SpoIIIL is required for the oblong shape of the forespore and the timing of σ^G activity

Previous work proposed SpoIIIL as a component of the A-Q complex [10]. This role for SpoIIIL was based on microscopy data showing that the ΔspoIIIL mutant exhibits hallmark phenotypes of A-Q complex mutants (reduced σG activity and rounder spores) [10]. To better analyze the spoIIIL phenotypes, we developed a strain to quantitively monitor spore shape and σG activity in the same cell (Fig 1A). This strain harbored two fluorescent reporters in the forespore compartment: a CFP reporter produced under the control of σF that allows examination of spore shape from early during developmental stages (P_spoIIQ-cfp) and a YFP reporter produced under the control of σG that allows measurement of σG activity (P_sspB-yfp). Using these reporters, we examined forespore shape and σG activity in WT and the ΔspoIIIL mutant over a sporulation time-course (Fig 1A). Consistent with previous work [23], WT forespores became more oblong overtime (Fig 1C). Furthermore, and consistent with previous results [15,18,24], σG activity in WT forespores peaked after engulfment completion at T3.5 (Fig 1B). As expected, the ΔspoIIIL mutant forespores were rounder than WT (Figs 1C and S1A). On the other hand, the ΔspoIIIL mutant forespores exhibited a slight delay in σG activity, but reached WT σG activity levels at T4.5 (Fig 1B), suggesting that SpoIIIL’s role is not related to the A-Q complex.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Forespore shape and timing of σ^G activity in WT and ∆spoIIIL mutant.

(A) Representative images of WT and ∆spoIIIL mutant expressing CFP (under the control of a σF-dependent promoter) and YFP (under the control of a control of σG-dependent promoter) at T3.5 and T5. Merged images show membrane in red and YFP in cyan. Scale bar is 2 μm. (B) Mean intensity (errors bars are ± STDEV of 3 biological replicates; n > 500, for each biological replicate) of σG-dependent YFP signal in WT and ∆spoIIIL mutant forespores, during a sporulation time-course. (C) Mean forespore aspect ratio (length/width) (errors bars are ± STDEV of 3 biological replicates; n > 500, for each biological replicate) of WT and ∆spoIIIL mutant forespores, during a sporulation time-course.

https://doi.org/10.1371/journal.pgen.1011768.g001

SpoIIIL localizes uniformly in the forespore membrane and undergoes a change in localization during spore maturation

Our data above suggest that SpoIIIL’s function is unrelated to the A-Q complex. If this is the case, then SpoIIIL may exhibit a different localization than SpoIIQ, a well-defined A-Q complex protein produced in the forespore that becomes localized in the septal membrane facing the mother cell during engulfment [18,25,26]. We generated a functional SpoIIIL-GFP fluorescent-fusion (S1B and S1C Fig) and examined its localization during a sporulation time-course (Fig 2A). At early stages of development (T2.5), SpoIIIL-GFP localized uniformly in the forespore inner membrane, with no specific enrichment in the forespore engulfing septal membrane facing the mother cell (Fig 2A and 2B). The uniform localization of SpoIIIL-GFP in the forespore inner membrane was maintained after engulfment completion (Fig 2A) and was comparable to the localization of a GFP-fusion to two transmembrane segments of MalF (integral membrane protein involved in maltose import in Escherichia coli; MalF-GFP) (S1F Fig). Furthermore, although protein topology prediction tools suggest that SpoIIIL is a cytoplasmic protein (Phobius) [27], SpoIIIL contains a series of hydrophobic residues towards the N-terminus and C-terminus, (S1D Fig) which may contribute to its membrane localization. Consistent with this idea, computational simulations of SpoIIIL on a 2D planar membrane suggest that SpoIIIL remains membrane associated over a nanoscale timeframe (S1D Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. SpoIIIL localization during early-to-intermediate stages of sporulation.

(A) Representative images of SpoIIIL-GFP during a sporulation time-course. Membrane in merged image is shown in red and SpoIIIL-GFP in cyan. Scale bar is 2 μm. (B) Representative image showing that SpoIIIL-GFP does not become enriched in the engulfing membrane (pointed out by yellow arrowheads). Scale bar is 2 μm. (C) Bar graph (left) showing the mean proportion of uniform (in blue) and non-uniform (in red) SpoIIIL-GFP localization at T3.5 and T5.5 (error bars are ± STDEV of 3 biological replicates; n > 100, for each biological replicate). Representative examples (right) of sporangia with uniform (outlined in blue) and non-uniform (outlined in red) localization of SpoIIIL-GFP. (D) Immunoblot analysis of cell lysates containing SpoIIIL-GFP, collected during a sporulation time-course. SpoIIIL-GFP was immunodetected using anti-GFP antibodies. The positions of SpoIIIL-GFP and GFP are indicated to the left and position of molecular weight markers to the right.

https://doi.org/10.1371/journal.pgen.1011768.g002

Interestingly, as forespores progressed through development, SpoIIIL-GFP changed localisation. The SpoIIIL-GFP signal became less uniform and could be observed as blotchy patches near the membrane or in the forespore cytoplasm (Fig 2A). For simplicity, we designated this less uniform signal as non-uniform. A similar change in localization was observed for an almost fully functional fluorescent fusion with GFP at the N-terminus of SpoIIIL (GFP-SpoIIIL) (S1B and S1E Fig). Quantification of the number of forespores with uniform and non-uniform signal at two time-points, T3.5 and T5.5, confirmed the change in SpoIIIL-GFP localization: at T3.5, 89.8% of spores displayed uniform SpoIIIL-GFP signal, while at T5.5 the fraction of spores with uniform signal decreased to 16.8% (Fig 2C). Immunoblot analysis suggests that the change in SpoIIIL-GFP localization between T3.5 and T5.5 is not due to changes in protein levels, as SpoIIIL-GFP levels were maintained across these time-points (Fig 2D). These results indicate that SpoIIIL undergoes a change in localization during spore maturation.

A synthetic sporulation screen identifies a relationship between spoIIIL and genes involved in the assembly of the spore envelope

The data above suggest that SpoIIIL does not function in the A-Q complex. To identify genes that function with SpoIIIL during sporulation, we used transposon-sequencing (Tn-seq) combined with sporulation heat-kills to screen for genes that become critical for sporulation in its absence. Saturated transposon libraries were constructed in the WT and the ΔspoIIIL mutant (see Materials & Methods). Analysis of the transposon insertion profiles revealed a set of genes in which insertions were underrepresented in the ΔspoIIIL mutant when compared to the WT (Fig 3A and 3C) (S1 Table). Interestingly, many of the genes identified encode proteins with roles in the assembly of various spore envelope layers, suggesting a genetic relationship between SpoIIIL and the spore envelope (S1 Table). These include genes involved in the spore coat (safA, spoVID and cotE), germ cell wall (pbpF), spore shape and cortex (ssdC, Fig 3C) and PG precursor synthesis (murAB, Fig 3C) [5,23,28–30]. Deleting most of these genes in combination with a ΔspoIIIL mutation only caused a moderate decrease in sporulation efficiency relative to the ΔspoIIIL mutant (S1 Table). By contrast, combining ΔssdC or ΔmurAB with ΔspoIIIL resulted in pronounced sporulation defects, with sporulation efficiencies of 0.01% and 0.38%, respectively (Fig 3B). Prompted by the severity of the ΔspoIIIL ΔmurAB and ΔspoIIIL ΔssdC double mutant sporulation defect, we investigated these mutants further.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Genetic relationship between spoIIIL and spore envelope genes revealed by Tn-seq and cytological analysis.

(A) Scatterplot showing fold-reduction of transposon insertions in ∆spoIIIL relative to WT, with corresponding P values. Genes involved in PG synthesis (murAB, pbpF), spore coat (safA, spoIVD and cotE), cortex (ssdC) and σ^K activation (ctpB) with high fold-reduction and low P value are labelled and colored red. (B) Sporulation efficiency of ∆spoIIIL, ∆murAB and ∆ssdC single mutants and ∆spoIIIL ∆murAB and ∆spoIIIL ∆ssdC double mutants (n = 3, mean ± STDEV; **P < 0.01, ***P < 0.001, ****P < 0.0001, Tukey’s multiple comparisons test). (C) Tn-seq profiles at the murAB and ssdC locus of WT and ∆spoIIIL cells following 30 h of growth and sporulation in exhaustion medium. The height of the vertical lines represents the number of Tn-seq reads at each position (scale is 200 sequencing reads). Shaded regions highlight the significant reduction in sequencing reads at the murAB and ssdC locus. (D) Representative phase contrast images showing spore refractivity at 9 (T9) and 30 (T30) hours after the onset of sporulation in WT and in ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutant sporangia. Scale bar is 2 μm. (E) Mean frequency of phase bright spores in the WT and ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutants (error bars are ± STDEV of 3 biological replicates; n > 100, for each biological replicate). (F) Representative electron micrographs of sporangia sections in the WT and ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutants. Scale bar is 200 nm and 100 nm in standard and zoomed-in images, respectively. White arrowheads point to a distinct light grey cortex. Yellow arrowheads point to amorphous grey material, reflecting a cortex synthesis defect.

https://doi.org/10.1371/journal.pgen.1011768.g003

Sporangia lacking SpoIIIL and MurAB exhibit a severe cortex synthesis defect

Our previous work indicated that SsdC is required for forespore shape and for proper cortex development, with the ΔssdC mutant resulting in phase bright spores with denser and less uniform cortex [23]. MurAB on the other hand has been shown to be required for efficient PG precursor synthesis which promotes efficient engulfment and may also promote efficient cortex synthesis [28]. Based on these observations, we reasoned that the sporulation defects of the ΔspoIIIL ΔssdC and ΔspoIIIL ΔmurAB double mutants may result from more pronounced defects in forespore shape, engulfment and/or cortex assembly.

To test this, we first examined whether the ΔspoIIIL ΔssdC and ΔspoIIIL ΔmurAB double mutants exhibit a more pronounced defect in spore shape, engulfment and σG activity compared to ΔspoIIIL, by examining the mutants in the CFP/YFP reporter background during a time-course covering early-to-intermediate stages of sporulation (S2 Fig). The ΔspoIIIL ΔssdC double mutant produced rounder forespores compared to either of the single mutants suggesting that the absence of SsdC exacerbates the shape defect of ΔspoIIIL (Figs S2D and 2E). Furthermore, σG activity reached similar levels in the ΔspoIIIL ΔssdC double mutant as the ΔssdC mutant, although following a similar delay as observed for the ΔspoIIIL mutant (S2C and S2E Fig). Engulfment completion (S2H Fig) and spore shape (S2G Fig) was comparable between ΔspoIIIL ΔmurAB and ΔmurAB. Interestingly, the ΔmurAB mutant also exhibited a slight delay in σG activation and did not reach WT levels (S2F Fig). The ΔspoIIIL ΔmurAB double mutant on the other hand, although more delayed than either of the single mutants, eventually reached WT levels of σG activity (S2F Fig). Thus, while the ΔspoIIIL ΔssdC and ΔspoIIIL ΔmurAB double mutants exhibit mild differences in spore shape and σG activity relative to their single mutant counterparts, these differences are unlikely to be the major cause of their pronounced sporulation defect.

Next, we examined whether the ΔspoIIIL ΔssdC and ΔspoIIIL ΔmurAB exhibit a more pronounced defect in cortex synthesis, which results in a change in forespore refractivity, from phase dark to phase bright, as they mature [23]. Interestingly, while the ΔspoIIIL ΔssdC double mutant produced phase-bright spores (S2A Fig), the ΔspoIIIL ΔmurAB double mutant produced virtually none (Fig 3D). Quantification showed that the WT and ΔspoIIIL mutant produced similar amounts of phase bright spores (T30; 88% and 80%, respectively) (Fig 3E). The ΔmurAB mutant on the other hand produced fewer phase-bright spores compared to WT (70.6%) (Fig 3E), suggesting that PG precursor synthesis by MurAB also contributes to efficient cortex synthesis. Recapitulating the microscopy data, the ΔspoIIIL ΔmurAB double mutant only produced 0.8% phase-bright spores, corresponding to a 100-fold reduction in phase bright spores compared to WT, similar to its sporulation efficiency (Figs 3E and S2B). Transmission electron microscopy (TEM) confirmed the absence of a distinct light grey cortex in the ΔspoIIIL ΔmurAB double mutant and instead an amorphous grey material was observed (Fig 3F). Together, these data suggest that SpoIIIL is required for cortex synthesis in cells lacking MurAB.

SpoIIIL is required for efficient accumulation of PG precursors at intermediate stages of development

The data above indicate that sporulating cells lacking SpoIIIL are sensitive to mild defects in PG synthesis, with the ΔspoIIIL ΔmurAB double mutant producing very few phase-bright spores with a mature cortex. How could the combined absence of SpoIIIL and MurAB impact cortex synthesis? Although the cortex PG synthases SpoVD and SpoVE are localized in the forespore outer membrane throughout engulfment, cortex synthesis only occurs post-engulfment with σK activation in the mother-cell and subsequent transcriptional upregulation of a subset of genes required for production of PG precursors (murAA, murB, murC and murF; Fig 4A) [9]. Since MurAB plays a direct role in maintaining efficient PG precursor levels during sporulation [28], but most sporangia lacking MurAB still assemble a cortex (Fig 3E), we hypothesised that SpoIIIL indirectly influences the upregulation of PG precursors such that the absence of both proteins would affect PG precursor levels and thereby compromise cortex assembly. To test this, we examined PG precursor levels in WT and ΔspoIIIL mutant cells by liquid chromatography-mass spectrometry (LC-MS) analysis. We were able to confidently quantify four PG precursors: UDP-MurNAc, UDP-MurNAc-L-Ala-D-Glu (UDP-M2), UDP-MurNAc-L-Ala-D-Glu-mDAP-D-Ala (UDP-M4) and MurNAc-L-Ala-D-Glu-mDAP-D-Ala-D-Ala (UDP-M5) (Fig 4B). In all cases, PG precursor levels were significantly reduced in the ΔspoIIIL mutant compared to WT (Fig 4B). Furthermore, although variation between biological replicates compromised statistical significance, there was a reduction in the mean values for three of the four PG precursors examined in the ΔspoIIIL ΔmurAB double mutant, relative to either of the single mutants (Fig 4B). These data suggest that SpoIIIL is required for efficient accumulation of PG precursors and that the ΔspoIIIL ΔmurAB cortex phenotype results from defects in separate pathways that contribute to PG precursor synthesis.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. PG precursor levels in the absence of SpoIIIL and MurAB.

(A) Steps in PG precursor biosynthesis leading up to the production of UDP-MurNAc-L-Ala-D-Glu-mDAP-D-Ala-D-Ala (UDP-M5) by MurF. The proteins labelled in red have been shown to be upregulated by σK (see text). (B) Peptidoglycan precursor levels in WT, ∆murAB, ∆spoIIIL and ∆murAB ∆spoIIIL at 4.5 hours after the onset of sporulation (T4.5). Four PG precursors were identified by LC-MS: UDP-MurNAc (UDP-M), UDP-MurNAc-L-Ala-D-Glu (UDP-M2), UDP-MurNAc-L-Ala-D-Glu-mDAP-D-Ala (UDP-M4) and UDP-MurNAc-L-Ala-D-Glu-mDAP-D-Ala-D-Ala (UDP-M5). UDP-M4 is not synthesized but is likely a by-product of the cleavage of UDP-M5 by an unknown enzyme. (n = 3, mean ± STDEV; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Tukey’s multiple comparisons test).

https://doi.org/10.1371/journal.pgen.1011768.g004

SpoIIIL is required for efficient SpoIVB activity and pro-σK processing

As SpoIIIL is produced in the forespore, one way in which it could contribute to maintaining sufficient PG precursor levels for cortex synthesis is by influencing the function of the forespore-produced, secreted protease, SpoIVB which initiates the RIP signalling pathway that leads to σK activation (Fig 5A) [13]. Thus, we first tested if SpoIIIL influences spoIVB transcription and SpoIVB levels.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. SpoIVB activity and σ^K activation in the absence of SpoIIIL.

(A) Activation of σK is controlled by a signalling pathway across the forespore membranes that involves Regulated Intramembrane Proteolysis (RIP) of an inactive form of σK (pro-σK) to its active form (σK). This RIP signalling pathways involves the forespore produced Site-1 protease, SpoIVB, and the mother cell produced Site-2 protease, SpoIVFB, which cleaves pro-σK to σK. SpoIVFB activity is inhibited by SpoIVFA (and BofA) until the SpoIVB protease is produced and secreted into the intermembrane space, where it then cleaves SpoIVFA, releasing the inhibition on SpoIVFB and promoting processing of pro-σK to σK. Binding of σK to the promoter of genes required for PG precursor synthesis, leads to increased PG precursor synthesis required for cortex PG synthesis. (B) Immunoblots comparing SpoIVB activity in WT and ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutants. Full-length (FL) and autoproteolyzed (AP) forms of SpoIVB are pointed out on the right and the position of molecular weight markers are shown to the left. For uncropped blots showing lower molecular weight autoproteolytic forms of SpoIVB refer to S3B Fig. (C) Representative images of pro-σK-CFP localization at 3.5 (T3.5), 4.5 (T4.5) and 5.5 (T5.5) hours after the onset of sporulation in WT and ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutant backgrounds. Scale bar is 2 μm. For the same image panels showing the DNA and merge refer to S3E Fig. (D) Representative example of processed (cleaved) and unprocessed (uncleaved) pro-σK-CFP, used as the basis of processed pro-σK-CFP quantification. Yellow arrowheads point to the mother cell nucleoid. White arrowheads point to the location of the pro-σK-CFP signal. (E) Mean frequency of processed pro-σK-CFP at 3.5, 4.5 and 5.5 hours after the onset of sporulation in WT and ∆spoIIIL, ∆murAB and ∆spoIIIL ∆murAB mutant backgrounds (errors bars are ± STDEV of 3 biological replicates, > 100 cells for each biological replicate; **P < 0.01, ****P < 0.0001, Tukey’s multiple comparisons test).

https://doi.org/10.1371/journal.pgen.1011768.g005

To examine spoIVB transcription, we fused the spoIVB promoter to cfp (P_spoIVB-cfp) and monitored CFP signal at 4.5 and 5.5 hours after the onset of sporulation (T4.5 & T5.5), which coincides with the window of SpoIVB activity [31]. Quantitative analysis of CFP signal suggests that spoIVB transcription is similar between the WT and ΔspoIIIL mutant, indicating that SpoIIIL does not influence spoIVB transcription (S3C and S3D Fig). To examine SpoIVB levels, we used immunoblotting. SpoIVB levels were examined at four time-points after the onset of sporulation (T2.5, T3.5, T4.5 and T5.5) that coincide with the activation of σG which is required for efficient transcription of SpoIVB, and SpoIVB levels, that signal pro-σK processing [8,32,33]. SpoIVB is an autoproteolytic serine peptidase and its activity leads to its autoproteolysis overtime, resulting in the formation of proteolyzed forms that by immunoblot appear at a lower molecular weight than the full-length protein [32–34]. Consistent with previous reports [32–34], in WT cells we predominantly observed 2 bands, a fainter band that corresponds to full-length SpoIVB (FL) and another darker band that corresponds to autoproteolyzed SpoIVB (AP) (Fig 5B). Importantly, in WT cells SpoIVB was almost completed proteolyzed by T4.5 (Fig 5B). In the ΔspoIIIL mutant however, we observed a band corresponding to full-length SpoIVB across all time points examined (Fig 5B), suggesting that SpoIVB autoproteolysis is incomplete and that SpoIIIL is required for efficient SpoIVB activity. Similar observations were made in ΔspoIIIL ΔmurAB double mutant (Fig 5B). In the ΔmurAB, although a faint band of full-length SpoIVB was observed at all time-points, there was a higher degree of SpoIVB autoproteolysis than in the ΔspoIIIL mutant and this was particularly evident at T3.5 and T4.5 (Fig 5B).

To test if the decreased SpoIVB activity observed in the ΔspoIIIL and ΔspoIIIL ΔmurAB mutants affects σK activation, we examined and quantified the frequency of pro-σK processing during a sporulation time-course that coincides with the window of SpoIVB activity and σK activation, using a previously characterized pro-σK-CFP fluorescent fusion [35]. Pro-σK-CFP initially appears as a diffuse cytoplasmic signal in the mother cell (Fig 5D; unprocessed). As a result of SpoIVB-dependent signalling, pro-σK-CFP is cleaved by SpoIVFB and then co-localizes with the mother cell nucleoid forming a bright CFP patch that is consistent with its proteolytic processing and association with core RNA polymerase (Fig 5D; processed) (Fig 5C) [36]. Importantly, in the ΔspoIVB mutant pro-σK-CFP fails to produce bright CFP patches on the nucleoid and instead appears as a patchy cytoplasmic signal in the mother cell [36] (S3A Fig). In WT and ΔmurAB cells, bright CFP patches could be observed in many cells at T3.5 and increased in similar frequency by T5.5 (Fig 5C and 5E). In the ΔspoIIIL and ΔspoIIIL ΔmurAB double mutant however, we observed fewer cells with bright CFP patches across all time-points (Fig 5C and 5E). By T5.5, only 59.4% and 64.6% of the cells had bright CFP patches in ΔspoIIIL and ΔspoIIIL ΔmurAB, respectively. For comparison, bright CFP patches were observed in 92.3% and 90.4% in the WT and ΔmurAB mutant, respectively. These data suggest that pro-σK-CFP processing is delayed and less efficient in the absence of SpoIIIL and that ΔspoIIIL ΔmurAB double mutant cortex phenotype results from defects in separate pathways that contribute to PG precursor synthesis: a σK-dependent pathway involving SpoIIIL and σK-independent pathway involving MurAB.

Sporangia lacking SpoIIIL and CtpB exhibit a subtle delay in the timing of pro-σK processing

One gene in which transposon insertions were underrepresented in the ΔspoIIIL mutant was ctpB (S4B Fig) (S1 Table). CtpB is serine protease required for efficient cleavage of SpoIVFA, thereby fine-tuning efficient SpoIVFB activation and pro-σK processing (S4A Fig) [35,37,38]. Confirming the Tn-seq data, while the ΔctpB mutant exhibited a mild sporulation defect (79.8%), the ΔctpB ΔspoIIIL double mutant only produced 6.3% spores (S4D Fig). Since the above data suggest that SpoIIIL is required for efficient pro-σK processing, it seemed likely that ΔspoIIIL ΔctpB would exhibit a more pronounced defect in pro-σK processing compared to either of the single mutants. To test this, we compared pro-σK processing using the pro-σK-CFP fluorescent fusion mentioned above, in the ΔctpB, ΔspoIIIL and ΔspoIIIL ΔctpB mutants (S4C Fig). Consistent with previous reports showing that CtpB contributes to the timing of pro-σK processing [35], in the ΔctpB mutant we observed fewer cells with bright CFP patches at T3.5 compared to WT: 15.7% versus 36.2%, respectively (S4E Fig). At T4.5 and T5.5 pro-σK processing was similar between the WT and ΔctpB mutant (S4E Fig). Similarly, compared to the ΔspoIIIL mutant, the ΔspoIIIL ΔctpB double mutant had fewer cells with bright CFP patches at T3.5 and T4.5 (S4E Fig): 4.7% vs 1.8% and 24.3% vs 14.2%, in the ΔspoIIIL mutant vs ΔspoIIIL ΔctpB double mutant, respectively. However, by T5.5 pro-σK processing was similar between the ΔspoIIIL mutant and ΔspoIIIL ΔctpB double mutant (S4E Fig). Thus, the absence of CtpB results in a subtle enhancement of the ΔspoIIIL mutant pro-σK processing defect. However, since we observed the same number of cells with processed pro-σK-CFP at T5.5 in the ΔspoIIIL and ΔspoIIIL ΔctpB mutants but that ΔspoIIIL ΔctpB yields fewer heat-resistant spores than the ΔspoIIIL (S4D Fig), we conclude that CtpB likely plays an additional role in maintaining sporulation in the absence of SpoIIIL. Nonetheless, these data provide additional evidence that SpoIIIL functions in the cell-cell signalling pathway that leads to σK activation.

The change in SpoIIIL localization at intermediate stages of spore development depends on cortex PG synthesis

Our data suggest that the absence of SpoIIIL leads to a decrease in PG precursor synthesis, thus affecting cortex synthesis. Since the change in SpoIIIL-GFP localization at intermediate stages of development coincides with the timing of cortex synthesis, we wanted to know if the change in SpoIIIL’s localisation depends on the cortex. To this end, we examined SpoIIIL-GFP localization in cells lacking the cortex synthases SpoVD and SpoVE at 3.5 and 5.5 hours after the onset of sporulation (T3.5 and T5.5) (Fig 6). These time-points were chosen because they show clear changes in SpoIIIL-GFP localization from uniform to non-uniform in WT cells. (Fig 2A and 2C). In the absence of SpoVD and SpoVE, SpoIIIL-GFP did not undergo a change in localization from uniform to non-uniform to the same extent as WT (Fig 6A). Like WT, most forespores lacking SpoVD and SpoVE exhibited uniform SpoIIIL-GFP at T3.5 (Fig 6A and 6B). At T5.5 however, and unlike the WT, most forespores in the ΔspoVD ΔspoVE double mutant displayed uniform SpoIIIL-GFP localization: 67.8% vs 16.8% in ΔspoVD ΔspoVE and WT, respectively (Fig 6B). Immunoblot analysis suggests that SpoIIIL-GFP levels are comparable between the WT and ΔspoVD ΔspoVE double mutant (Fig 6C). These data suggest that cortex synthesis by SpoVD and SpoVE is required for the change in SpoIIIL-GFP localization.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. SpoIIIL-GFP localization in sporangia lacking the cortex synthases, SpoVD and SpoVE.

(A) Representative images of SpoIIIL-GFP in otherwise WT and in the ∆spoVD ∆spoVE double mutant. Membrane in merged image is shown in red and SpoIIIL-GFP in cyan. Scale bar is 2 μm. (B) Bar graph showing the mean frequency of uniform (in blue) and non-uniform (in red) SpoIIIL-GFP localization at T3.5 and T5.5 in WT and in the ∆spoVD ∆spoVE double mutant (errors bars are ± STDEV of 3 biological replicates; n > 100, for each biological replicate). (D) Immunoblot analysis of cell lysates containing SpoIIIL-GFP in WT and in the ∆spoVD ∆spoVE double mutant, collected at T3.5 and T5.5. SpoIIIL-GFP was immunodetected using anti-GFP antibodies. The positions of SpoIIIL-GFP and GFP are indicated to the left and position of molecular weight markers to the right.

https://doi.org/10.1371/journal.pgen.1011768.g006

spoIIIL is exclusively found within the Bacillales and its co-occurrence with the ComG operon is species-specific

Previous work suggests that spoIIIL may be restricted to a subset of spore-forming Firmicutes [17,39,40]. To comprehensively gauge the phylogenetic distribution of spoIIIL, we conducted an analysis on 497 Firmicutes genomes (Fig 7A). This analysis revealed that spoIIIL is exclusively found in a subset of species within the Bacillales (Fig 7A), indicating that SpoIIIL function during sporulation reflects a species-specific specialization. We also examined the co-occurrence of spoIIIL with the comG operon in species of Bacillales where spoIIIL is present (Fig 7B). As mentioned in the introduction, in B. subtilis spoIIIL is the last gene in the comG operon and is transcribed with the comG genes by ComK. Interestingly, this does not appear to be the case in other species of Bacillales. Only a few species display the same genetic organisation for comG and spoIIIL as that observed in B. subtilis. Other species exhibit different genetic organisation, for example: comG appears to be transcribed in the opposite orientation to spoIIIL or comG is not found within the same genetic neighbourhood as spoIIIL.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. SpoIIIL phylogenetic distribution within the Firmicutes and co-occurrence with ComG operon.

(A) Phylogenetic mapping of SpoIIIL (in red) and Spo0A_C as a marker of sporulation (in green) on a maximum likelihood reference phylogeny of the Firmicutes. The mapping is based on a supermatrix containing 497 taxa with 3,776 amino acid positions and inferred with IQ-Tree using the evolutionary model LG + I + G4. Black circles correspond to ultrafast bootstrap supports higher than 80%, and the scale bar corresponds to the average number of substitutions per site. (B) Co-occurrence of SpoIIIL (in red) and ComG operon (in green). Co-occurrence of SpoIIIL and ComG in B. subtilis in highlighted in pale orange. Refer to high-resolution figure for finer details (https://doi.org/10.6084/m9.figshare.28937915.v1).

https://doi.org/10.1371/journal.pgen.1011768.g007

Based on the above we tested if SpoIIIL is required for DNA uptake during competence. We compared DNA uptake (i.e., transformation efficiency) by transforming WT and the ΔspoIIIL mutant with genomic DNA conferring spectinomycin resistance and counting the number of spectinomycin resistant colonies. The ΔspoIIIL mutant exhibited a 1.7-fold reduction in transformation efficiency compared to WT (S5C Fig), indicating that SpoIIIL is required for efficient DNA uptake. If so, the ΔspoIIIL mutant would likely exhibit a synergistic defect with other mutants defective in DNA uptake. To test this, we combined ΔspoIIIL with a mutant lacking comFB and comFC (ΔspoIIIL ΔcomFBC), which encode two proteins that contribute to efficient DNA uptake [41]. While the ΔcomFBC double mutant resulted in 7.9-fold reduction in transformation efficiency, the ΔspoIIIL ΔcomFBC triple mutant resulted in 47.2-fold reduction (S5C Fig). Thus, SpoIIIL is required for efficient DNA uptake and likely functions with other ComG proteins in genetic competence in B. subtilis. Collectively, this data suggest that SpoIIIL contributes to genetic competence and sporulation in some Bacillales species.

General methods

Bacillus subtilis strains are listed in S2 Table, plasmids (and their construction details) are listed in S3 Table and oligonucleotide primers are listed in S4 Table. Sporulation induction was achieved through resuspension at 37°C following the Sterlini-Mandelstam method [48] or by nutrient depletion in a supplemented Difco sporulation medium (DSM) [49] consisting of 8 g/L Bacto nutrient broth (Difco), 0.1% (wt/vol) KCl, 1 mM MgSO₄, 0.5 mM NaOH, 1 mM Ca(NO₃)₂, 0.01 mM MnCl₂, and 0.001 mM FeSO₄. Sporulation efficiency was determined through the heat-kill assay conducted on cultures grown for 30 hours at 37°C in the DSM medium. The total number of heat- resistant (80°C for 20 minutes) CFUs (colony-forming units) is compared with wild-type heat-resistant CFUs. Transformation efficiency was determined through a transformation assay conducted on cultures grown for 4 h at 37°C in 1XMC medium, followed by growth for 2 h at 37°C in the same spent medium but with added purified genomic DNA conferring spectinomycin resistance (0.01 mg; ycgO::spec).

Transposon insertion sequencing

Transposon insertion sequencing (Tn-seq) was performed on wild-type (bDR2413) and ΔspoIIIL libraries as described previously [10]. Approximately 750,000 transformants were pooled, aliquoted, and frozen. An aliquot was thawed, washed in DSM, and diluted into 50 mL DSM at an optical density at 600 nm (OD₆₀₀) of 0.05. Samples were harvested 24 h later (T24). The T24 samples were incubated at 80°C for 20 min and then plated on LB agar. Approximately 750,000 colonies from germinated spores from each sample were pooled. Genomic DNA was extracted from both samples and digested with MmeI, followed by adapter ligation. Transposon-chromosome junctions were amplified in 16 PCR cycles. PCR products were gel purified and sequenced on the Illumina HiSeq platform using TruSeq reagents (Tufts University TUCF Genomics Facility). Reads were mapped to the B. subtilis 168 genome (NCBI NC_000964.3) and tallied at each TA site, and genes in which reads were statistically underrepresented or overrepresented were identified using the Mann-Whitney U test (full list can be found here: https://doi.org/10.6084/m9.figshare.28633853.v4). Visual inspection of transposon insertion profiles was performed with the Sanger Artemis Genome Browser and Annotation tool.

Fluorescence microscopy

Live-cell fluorescence imaging was performed by placing cells on a 2% (w/v) agarose pad prepared with resuspension medium and set using a Gene Frame (Thermofisher). For sporulating cell cultures prepared using the resuspension method, at specific time points, 250 μL of culture was pelleted via centrifugation and resuspended in 10 μL of resuspension medium containing 0.05 mM TMA-DPH [1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene p-toluenesulfonate] or DAPI (10 µg/mL). Next, 2 μL of cell suspension was spread onto the agarose pad, and a coverslip was placed on top of the adhesive Gene Frame. Cells were imaged using a Zeiss Axio Observer 7 microscope equipped with a Plan-APOCHROMAT 100x/1.4 Oil Ph3 objective and a Colibri 7 Type R[G/Y]CBV-UV fluorescent light source. Images were captured with an Axiocam 712 mono camera. The TMA-DPH membrane dye was excited with a Zeiss 02 DAPI filter, using an exposure time of 100 ms. GFP was excited with a Zeiss 38 HE filter, with an exposure time of 150 ms. YFP and CFP was excited with a Zeiss 108HE filter, with an exposure time 150 and 300 ms, respectively. The mCherry fluorophore was excited with a Zeiss 90 HE LED filter, with an exposure time of 1000 ms. DAPI was excited with a Zeiss 02 DAPI filter, with an exposure time of 150 ms.

Image analysis & Statistics

Microscopy images were processed by adjusting the brightness and contrast using the Fiji software. The MicrobeJ plugin [50], designed for the Fiji software [51], was used to analyse forespore shape (P_spoIIQ-cfp or P_spoIIQ-mCherry), σG activity (P_sspB-yfp) and spoIVB transcription (P_spoIVB-cfp). The proportion of cells containing phase-bright cortex (Fig 3E) or bright patches of pro-σK-CFP signal (Figs 5E and S4E) was determined manually, using the Cell Counter plug-in for Fiji.

To perform quantitative analysis of forespore shape and σG activity using P_spoIIQ-cfp and P_sspB-yfp signal, image background was first subtracted (Process > Subtract Background) to avoid false-positive detection of the fluorescent signal. Next, the “Bacteria” tab on MicrobeJ was set to “Smoothed” to detect the outline of CFP signal from the forespores in Channel 1 and the YFP signal from the same forespores in Channel 2. Three parameters: “Exclude on Edges”, “Shape descriptors”, “Segmentation” and “Intensity” were checked. The generated CFP outlines were further refined by setting the shape descriptors (area, length, width) for each time-point corresponding to the outlines of individual forespores. The manual editing tool was also used to resolve unprocessed forespores. Results of the analyses of Channel 1 and Channel 2, including aspect ratio and mean intensity were exported to Microsoft Excel to process raw data into line charts. A similar approach was used to analyse forespore shape in the SpoIIIL-GFP strain, in this case using P_spoIIQ-mCherry set to Channel 1, to detect forespore outlines (no Channel 2 was used).

To perform quantitative analysis of spoIVB transcription using P_spoIIQ-mCherry and P_spoIVB-cfp signal, image background was first subtracted (Process > Subtract Background) to avoid false-positive detection of the fluorescent signal. Next, the “Bacteria” tab on MicrobeJ was set to “Smoothed” to detect the outline of mCherry signal from the forespores in Channel 1 and the CFP signal from the same forespores in Channel 2. Three parameters: “Exclude on Edges”, “Shape descriptors”, “Segmentation” and “Intensity” were checked. The generated mCherry outlines were further refined by setting the shape descriptors (area, length, width) for each time-point corresponding to the outlines of individual forespores. The manual editing tool was also used to resolve unprocessed forespores. Results of the analyses of Channel 2 (CFP), including mean intensity were exported to Microsoft Excel.

Violin plots were generated through inputting MicrobeJ data into Superplots (https://huygens.science.uva.nl/SuperPlotsOfData/) [52]. The non-parametric Kolmogorov-Smirnov test was used to compare the aspect ratio distribution, and P_spoIVB-cfp or pro-σK-CFP signal distributions between populations of wild-type and mutant sporulating cells. ANOVA, followed by the Tukey’s multiple comparisons test were used to compare means between WT and mutant strains as indicated in the legend. Numerical values used in graphs can be found here: https://doi.org/10.6084/m9.figshare.28633853.v4.

Immunoblot analysis

Whole-cell lysates from sporulating cells were prepared as previously described [23]. Samples were heated for 15 min at 50 ̊C prior to loading. Equivalent loading was based on the OD₆₀₀ at T2.5, since after T3.5 optical density readings do not accurately correlate with cell density, due to changes in spore refractivity as a result of spore maturation. Samples were separated on a 10% polyacrylamide gel and transferred to a PVDF membrane. Membranes were blocked in 5% non-fat milk with 0.5% Tween-20 for 1 h. Blocked membranes were probed with anti-GFP (1:2500) (Abcam) or anti-SpoIVB (1:10000) primary antibodies diluted into PBS with 0.05% Tween-20 containing 3% Bovine Serum Albumin (w/v) (Sigma) at 4 ̊C overnight. Primary antibodies were detected with horseradish-peroxidase anti-rabbit antibodies (BioRad) and detected with ECL Prime Western Blotting Detection reagent (Amersham) as described by the manufacturer.

PG precursor analysis

Sporulating cells (10 mL) were harvested at T4.5, normalized to an OD₆₀₀ equal to 1 (measured at T3.5) and pelleted by centrifugation. Cell pellets were washed three times with ice-cold 0.9% NaCl, resuspended in 100 μL 0.9% NaCl, and boiled for 5 min to lyse the cells and extract the soluble peptidoglycan (PG) precursors. Lysates were centrifuged at 21,000 × g for 5 min to pellet insoluble material, and the supernatant was filtered (0.22 μm pore size) for LC-MS analysis. Extracted precursors were analysed using a UPLC system (Waters) with an Acquity UPLC BEH C18 column (130-Å, 1.7-μm, 2.1 × 150 mm) coupled to a Xevo G2-XS QTOF mass spectrometer (Waters). Separation was achieved using a gradient from 0.1% formic acid in water to 0.1% formic acid in acetonitrile over 18 min at 45°C. The QTOF operated in positive-ion mode with the MSe mode used for data acquisition. MS parameters were set as follows: capillary voltage, 3 kV; source temp, 120°C; desolvation temp, 350°C; sample cone voltage, 40 V; cone gas flow, 100 L/h; desolvation gas flow, 500 L/h. Data were acquired and processed using Masslynx software. Quantification was done by extracting ion chromatograms for the expected m/z ratios of each ion and integrating the resulting peaks. The extracted m/z values were: UDP-MurNAc: 702.09; UDP-M2:880.18; UDP-M4:1123.30 and UDP-M5: 1194.33.

Muropeptide analysis of mature spores

Spore pellets of the WT and ΔspoIIIL mutant, normalized to an equal OD₆₀₀, were resuspended in 2 mL decoating solution (50 mM Tris-HCl pH 8.0, 8 M urea, 1% SDS and 50 mM dithiothreitol) and incubated at 37 °C for 1 h with shaking. The spores were pelleted by centrifugation at 14,000 × g for 8 min and washed 5 times in 2 mL water to remove the SDS. The final pellets were resuspended in 0.9 mL trypsin digestion solution (20 mM Tris-HCl pH 8.0, 10 mM CaCl₂, 0.1 mg/mL trypsin) and incubated overnight at 37 °C with shaking. The next day, 100 μL 10% SDS was added to each reaction, which were then boiled for 15 min. Insoluble PG sacculi were pelleted by ultracentrifugation at 150,000 × g for 13 min, washed 3 times with 3 mL water and resuspended in 100 μL water. Next, 10 μL of 2 mg/mL muramidase stock solution was added and the samples were incubated overnight at 37 °C with shaking to digest the PG into soluble muropeptides. The next day, the muramidase was inactivated by boiling for 5 min, and the samples were centrifuged at 24,000 × g for 15 min to remove the insoluble material. The supernatants were transferred to new tubes, dried in a SpeedVac system, and resuspended in 50 μL water. Next, 5 μL 0.5 M sodium borate buffer (pH 9) was added to adjust the pH to 8.5-9, before the muropeptides were reduced by adding 25 μL of freshly made 76 mg/mL NaBH₄. The samples were incubated at RT for 20 min, and then adjusted to pH to 3.5-4 using 25% orthophosphoric acid. The final samples were filtered through a 0.22 μm pore-size filter for analysis by LC.

Liquid chromatography was carried out using an Acquity H-Class UPLC system (Waters) equipped with an Acquity UPLC BEH C18 column (2.1 mm × 150 mm, 130 Å pore size and 1.7 μm particle size, Waters) maintained at 45 °C. Chromatographic separation was achieved using a linear gradient from 0.1% formic acid in water to 0.1% formic acid in acetonitrile. UV detection was done at 204 nm. Peaks were identified using in-line LC-MS with a Xevo G2-XS mass spectrometer (Waters) operated in positive ion mode, with data collection performed in the data-independent MSe mode. MS parameters were set as follows: capillary voltage 3.0 kV, source temperature 120 °C, desolvation temperature 350 °C, sample cone voltage 40 V, cone gas flow 100 L h⁻¹ and desolvation gas flow 500 L h⁻¹. Data acquisition and processing was performed using the UNIFI software (Waters). Compounds were built and their expected masses calculated using the ChemSketch software (ACD labs).

UV chromatograms were integrated to quantify peaks using the Empower software (Waters). The relative molar abundance of each muropeptide was calculated by dividing peak areas by the calculated average molar mass of each compound relative to GlcNAc-MurNAc-tetrapeptide, then dividing the adjusted peak area by the total peak area to give percent molar abundance. Crosslinking was calculated as the percentage of peptide stems in crosslinks, and muramic δ-lactam was calculated as the percentage of total muramic acid residues.

Chemical fixation, thin sectioning and electron microscopy of B. subtilis sporangia and spores

For TEM of mature spores, spore pellets were fixed in 3% (w/v) glutaraldehyde at 4°C overnight. Samples were washed in 0.1 M sodium cacodylate buffer and further incubated for 2 h at room temperature in 1% (w/v) osmium tetroxide for secondary fixation. Spore pellets were washed twice with 0.1 M sodium cacodylate buffer and dehydrated by incubating with increasing concentrations of ethanol [50% (v/v), 75% (v/v), 95% (v/v), 100% (v/v) dried ethanol], followed by two pure propylene oxide 15 min incubations for complete dehydration. Samples were incubated overnight at room temperature in a 1:1 mix of propylene oxide and Epon resin to allow for infiltration. Resin was removed and excess propylene oxide evaporated at room temperature. Samples were incubated for two consecutive 4h periods in pure Epon resin before being embedded into the final fresh resin. Resin was polymerised by incubation at 60°C for 48 h. Thin sections (80 nm) were produced using an Ultracut E Ultramicrotome (Reichert-Jung) and floated onto 300-square mesh nickel TEM grids. Sections were stained in 3% (w/v) uranyl acetate for 30 min, washed with dH₂O, stained with Reynold’s lead citrate for 5 min and further washed with dH₂O. Sections were imaged on a Tecnai G2 spirit BioTwin (FEI) 120 kV microscope, at a magnifications between 16,000–32,000 x (pixel size of 0.56 and 0.28 nm), equipped with an Orius SC1000B CCD camera (Gatan).

For TEM of sporangia, sporulating cells were harvested at T9 by centrifugation at 3,220 × g for 10 min. The resulting cell pellet (~ 1.5 μL) was placed on the 200-μm side of a type A 3-mm gold platelet (Leica Microsystems), covered with the flat side of a type B 3-mm aluminum platelet (Leica Microsystems), and vitrified by high-pressure freezing using an HPM100 system (Leica Microsystems). Freeze-substitution was carried out at −90°C for 80 h in acetone supplemented with 1% OsO₄. The samples were then gradually warmed at a rate of 2°C/h to −60°C using an AFS2 system (Leica Microsystems). After incubation at −60°C for 8–12 h, the temperature was raised to −30°C (2°C/h), and then to 0 °C within 1 h to enhance the osmium action and improve the membrane contrast. The sample was then cooled down to −30 °C within 30 min and subsequently rinsed 4 times in pure acetone. For resin embedding, the samples were progressively infiltrated with increasing concentrations of Embed812 resin (EMS) in acetone (1:2, 1:1, 2:1 [v/v]) over 2 h for each ratio, while gradually raising the temperature to 20°C. Pure resin was then added at room temperature, and polymerization was carried out at 60°C for 48 h. Thin sections (80 nm) were prepared using an ultramicrotome UC7 (Leica Microsystems) and collected on formvar-carbon-coated 100-mesh copper grids (Agar Scientific). The sections were post-stained with 2% aqueous uranyl acetate for 5 min, rinsed in water, further stained with lead citrate for 2 min and rinsed again. The samples were imaged using a Tecnai G2 spirit BioTwin (FEI) 120 kV microscope, at a magnifications between 16,000–32,000 x (pixel size of 0.56 and 0.28 nm), equipped with an Orius SC1000B CCD camera (Gatan).

Phylogenetic and co-conservation analysis

We assembled a local database of Firmicutes containing 497 genomes and inferred the corresponding reference phylogenetic tree as described in Chan et al., 2022 [28]. To infer sporulating taxa in the firmicutes, we searched for homologues of Spo0A against our Firmicutes database using the pfam domain PF08769 (Spo0A_C) and the HMMsearch tool from the HMMER package [53], with the option cut_ga. To infer the presence of SpoIIIL, we used the pfam domain PF14038 (YqzE) and HMMsearch tool. All SpoIIIL hits were retrieved and curated using alignment, phylogeny and domains to discard false positives. The presence of Spo0A and SpoIIIL was then mapped on the reference phylogeny of the Firmicutes using IToL [54].

To explore the co-conservation of SpoIIIL with the ComG operon in the Firmicutes, we retrieved the pfam domains PF00437, PF11773, PF15980, and PF14173 corresponding to the proteins ComGA, ComGE, ComGF, and ComGG respectively. As the remaining proteins do not have a corresponding pfam domain, we extracted the alignments corresponding to NF041012 (T4P_ComGB) and NF040982 (ComGD) NCBI families, and the COG family COG4537 (ComGC) from the Conserved Domain Database [55]. HMM profiles were first generated using the downloaded alignments with HMMbuild tool and then searched against the Firmicutes database using HMMsearch. All the hits retrieved were used to highlight the occurrence of one or more proteins from the ComG operon next to SpoIIIL using GeneSpy [56].

Computational simulations

An AlphaFold2 [57] prediction of the full length SpoIIIL protein, was inserted along its hydrophobic facing residues onto a 20:70:10 POPG:POPE:CDL1 membrane using the MemProtMD pipeline [58]. After automatic partial submersion in the lipids as predicted by memembed [59] and stabilisation by coarse Grained simulation under default MemProtMD parameters, CG2AT [60] was used to produce a full atom simulation box for production simulation of the SpoIIIL-membrane assembly. The full atom box was then simulated using GROMACs [61] charmm36 forcefields for 500 ns at 310K in 150mM NaCl at pH 7 (S1 Movie).