RagB stimulates the activity of the peptidoglycan polymerase RodA in Bacillus subtilis

Abstract

The bacterial cell wall is primarily composed of peptidoglycan (PG), a polymer essential for its protective envelope function, and any defect in its synthesis or repair can potentially result in bacterial lysis. Class A Penicillin-Binding Proteins (aPBPs) and Shape, Elongation, Division, and Sporulation (SEDS) proteins are PG polymerases acting in concert to ensure bacterial cell wall growth. Here, we identify the first regulator of the SEDS protein RodA in the Gram-positive model bacterium Bacillus subtilis. In the presence of the antibiotic moenomycin, which specifically inhibits glycosyltransferase activity of aPBPs, or in a strain deleted for all four aPBPs, bacterial survival depends on the presence of the YrrS protein (renamed RagB) and can be rescued by overexpression of RodA. No effect of RagB is observed on the rodA gene expression level or on the speed of circumferentially moving RodA associated with PG elongation by the Rod complex. However, we demonstrate that RagB interacts with RodA. We propose that RagB stimulates RodA activity and becomes essential in the absence of aPBPs and in particular of the major aPBP, PBP1.

Keywords: Bacillus subtilis; Peptidoglycan Synthesis; Regulation; SEDS; aPBP.

Figure 1. Growth and morphology of strains deleted for ponA, gpsB and or ragB genes.

(A) Growth of the different strains was tested on a 96-well plate in triplicate and incubated on a microplate reader at 37 °C for 7 h with stirring and OD measurements every 30 min. The representative growth curves of three biological replicates are shown. Growth of strains wild-type 168 (WT) in green, PS2062 (ponA::spec) in blue, JR46 (gpsB::kan) in pink, SG314 (ponA::spec gpsB::kan) in gray, BKE27300 (ragB::erm) in red and SG667 (ponA::spec ragB::erm) in black. (B) Microscopy images of strains wild-type 168 (WT) in green, PS2062 (ponA::spec) in blue, BKE27300 (ragB::erm) in red, SG667 (ponA::spec ragB::erm) in black, JR46 (gpsB::kan) in pink and SG314 (ponA::spec gpsB::kan) in gray cultured in LB at 37 °C up to OD_600nm = 0.3 were taken to observe their cell morphology. Scale bar is 5 µm. (C) Cell diameters of mutant strains were measured from microscopy images of strains wild-type 168, BKE27300 (ragB::erm), PS2062 (ponA::spec), SG667 (ponA::spec ragB::erm), JR46 (gpsB::kan) and SG334 (ponA::spec gpsB::kan) cultured in LB at 37 °C up to OD_600nm = 0.4. Violin plots of cell diameter (µm) whose average value is shown at top right, with N indicates the number of cells analyzed for each strain from three biological replicates. Blue boxes are box plots and dark lines indicate the median. Diameters are measured with the MicrobeJ plug-in. Statistically significant differences between the wild-type and mutant diameters are determined with ANOVA using a Tukey’s test and exact P values are indicated. Data information: In (C), data are presented as Violin and box plots. Each box represents the interquartile range (IQR), with the lower and upper bounds corresponding to the 25th and 75th percentiles, respectively. The center line indicates the median (50th percentile). Whiskers extend to the minimum and maximum values within 1.5 × IQR; data points beyond this range are considered outliers. ns not significant, ***P < 0.001 (Tukey’s test). Source data are available online for this figure.

Figure 2. Complementation of SG667 (ponA::spec ragB::erm) phenotypes by RodA overproduction.

(A) Growth of the different strains was tested in triplicate on a 96-well plate at 37 °C for 7 h. Growth in LB (top graph, empty symbols) and LB + 1% xylose (bottom graph, solid symbols) of the strains wild-type 168 (WT) in green, PS2062 (ponA::spec) in blue, SG667 (ponA::spec ragB::erm) in black, SG712 (ponA::spec ragB::erm amyE::P_xylragB) in purple, SG713 (ponA::spec ragB::erm amyE::P_xylrodA) in yellow and SG933 (ponA::spec ragB::erm amyE::P_xylrodA(D280A)) in red are presented on the graph. The representative growth curves of three biological replicates are shown. (B) Microscopy images of strains WT in green, PS2062 (ponA::spec) in blue, SG667 (ponA::spec ragB::erm) in black, SG712 (ponA::spec ragB::erm amyE::P_xylragB) in purple, SG713 (ponA::spec ragB::erm amyE::P_xylrodA) in yellow and SG933 (ponA::spec ragB::erm amyE::P_xylrodA(D280A)) in red, cultured in LB at 37 °C up to OD_600nm = 0.3, were taken to analyze their cell morphology. Scale bar is 5 µm. (C) Strains were grown at 37 °C on LB medium supplemented with 15 mM MgSO₄ to an OD_600nm of 0.4. A 800 μl aliquot of each culture was centrifuged for 3 min at 7000 rpm. Cell pellets were resuspended in 300 µl of LB, and 200 µl were spread on a LB-Agar plate containing 1% xylose. In total, 40 µg of moenomycin were deposited on filter paper disks placed in the center of the plates that were incubated at 30 °C overnight. Pictures were taken to compare the size of the inhibition zone for each strain: wild-type 168 (WT), BKE27300 (ragB::erm), PS2062 (ponA::spec), SG667 (ponA::spec ragB::erm), SG712 (ponA::spec ragB::erm amyE::P_xylragB), SG713 (ponA::spec ragB::erm amyE::P_xylrodA) and SG933 (ponA::spec ragB::erm amyE::P_xylrodA(D280A)). All the experiments were realized in triplicate and representative experiments are shown here. (D) Bar graph of the zone of growth inhibition (in mm) by moenomycin for the strains shown in (C). Each obtained value from the three biological replicates is represented by a plot. Error bars of the graph are the SD. Statistically significant differences between the wild-type and mutant inhibition diameters were determined with ANOVA using a Tukey’s test and exact P values are indicated. Data information: In (D), data are presented as mean ± SD. ns not significant, **P < 0.01, ***P < 0.001 (Tukey’s test). Source data are available online for this figure.

Figure 3. rodA expression is not affected by RagB.

(A) Bar graph of rodA expression levels in the strains wild-type 168 (WT), PS2062 (ponA::spec), BKE27300 (ragB::erm) and SG667 (ponA::spec ragB::erm) grown in LB medium until OD_600nm = 0.5 at 37 °C. The expression level was determined by quantitative RT-PCR and represented (as colored bars) for the three strains relative to that of the WT strain (value = 1). Four biological replicates were realized and each obtained value is indicated by a plot; error bars represent the SD. The statistically significant differences between wild-type and mutant strains were determined with ANOVA using a Tukey’s test and the P values are indicated. (B) Bar graph of ragB expression levels quantified by quantitative RT-PCR for the strains wild-type 168 (WT) and PS2062 (ponA::spec) grown in LB medium until OD_600nm = 0.5 at 37 °C. Each obtained value from the three biological replicates is represented by a plot. Error bars of the graph are the SD. The statistically significant differences between wild-type and mutant strains were determined with a Mann–Whitney U test and the P value is indicated. (C) Comparative western blot showing RagB in wild-type (WT), PS2062 (ponA::spec) and BKE27300 as negative control (ragB::erm) strains. Strains were grown in LB medium until OD_600nm = 0.5 at 37 °C, crude extracts were loaded (2, 4, or 8 µl) and separated on 12.5% SDS-PAGE, transferred to a nitrocellulose membrane, and RagB was detected using anti-RagB antibodies. Data information: In (A, B), data are presented as the mean ± SD. ns not significant, *P < 0.05, **P < 0.01 (Tukey’s test and Mann–Whitney U test, respectively). Source data are available online for this figure.

Figure 4. RagB and RodA interact in vitro and in vivo.

(A) Detection of protein interactions by bacterial two-hybrid assay. The T18 and T25 fragments of the adenyl cyclase protein were fused to the N-termini of RagB, GpsB, and RodA. Co-transformed strains of E. coli BTH101 were spotted onto LB medium supplemented with Xgal and IPTG and incubated at 30 °C overnight. Blue colonies indicate a positive interaction. Some were detected between GpsB and RagB, RodA and RagB, but no interaction was detected between GpsB and RodA. (B) Detection of protein interactions by pull-down assay. An extract of membrane proteins from the SG825 (ragB::erm amyE::P_xylrodA-gfp) strain containing RodA-GFP (lane 1) was incubated for 1 h at 4 °C with either buffer (lanes 2, 3, 4) or 100 µg of 6His-RagB protein (lanes 5, 6, 7) and then purified on a Ni-NTA column. The purification fractions were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The presence of RagB (bottom gel) and RodA-GFP (top gel) proteins in the different fractions was detected using specific antibodies. In the protein extract (lane 1) RodA-GFP is detected confirming its production and extraction (top gel). In the non-retained fraction (lanes 2 and 5), some unbound RodA-GFP is detected, but not in wash fractions (lanes 3 and 6). During purification of 6His-RagB (bottom gel), an excess of protein is found in the non-retained (lane 5) and wash (lane 6) fractions, but the majority is eluted (lane 7). RodA-GFP is co-eluted (lane 7) with 6His-RagB. (C) Detection of protein interactions by Co-immunoprecipitation assay. Extracts of membrane proteins from the SG187 (amyE::P_xylgfp), SG818 (amyE::P_xylrodA-gfp) and SG825 (ragB::erm amyE::P_xylrodA-gfp) strains were purified on GFP-affinity resin and eluted proteins RagB (bottom gel) and GFP or RodA-GFP (top gel) were detected using anti-RagB or anti-GFP antibodies, respectively. No trace of non-specifically co-purified RagB is detected (lane 1, bottom gel) with GFP alone (lane 1, top gel) showing that RagB does not bind non-specifically to either GFP or the resin. RodA-GFP produced in SG825 and SG818 strains is well purified (lanes 2 and 3, top gel), and RagB from SG818 is specifically co-purified with RodA (lane 3, bottom gel) but not from SG825 (lane 2, bottom gel). Source data are available online for this figure.

Figure 5. RodA overproduction rescues the growth of Δ4 strain only in the presence of RagB.

(A) Serial dilutions of strains wild-type 168 (WT), AG157 (pbpG::kan ΔpbpD ΔpbpF ΔponA named Δ4), SG1150 (Δ4 amyE::P_xylrodA) and SG1158 (Δ4 ragB::erm amyE::P_xylrodA) grown in LB supplemented with 15 mM MgSO₄ until OD_600nm = 0.4 were spotted (10 µl) on LB-Agar + 15 mM MgSO₄, LB-Agar and LB-Agar + 0.5% xylose plates and incubated overnight at 37 °C. Differences in growth between strains for both conditions are monitored by the appearance of colonies for each dilution. (B, C) Statistical analysis from microscopic morphology monitoring of ∆4 strain and derivatives. (B) Visible light microscopy images were taken for the 4 strains, wild-type 168 (WT), AG157 (pbpG::kan ΔpbpD ΔpbpF ΔponA named Δ4), SG1150 (Δ4 amyE::P_xylrodA) and SG1158 (Δ4 ragB::erm amyE::P_xylrodA) grown in LB supplemented with 2.5 mM MgSO₄ and 0.5% xylose until OD_600nm of 0.4 at 37 °C. A representative image of the morphology of the cells of each strain is shown here. Ten images per strain were taken in order to be able to statistically process the morphology parameters in a significant number of cells. Data were collected from three biological replicates. Scale bar is 5 µm. (C) Statistical analysis of the values for four morphological parameters of cell shape representative of curvature measurement: lateral and longitudinal asymmetries (defined as the overlap ratio between the two halves of the cell contour separated by the median axis or the axis perpendicular to the middle of the cell, respectively, where 1 is the 100% of overlap), width variation (defined as the standard deviation of the cell width measured perpendicular to the median axis at a distance of one pixel apart) and solidity (defined as the ratio of cell surface to convex cell surface (convex hull) with 1 means that the cell shape is convex). Graphs represent the solidity values (0 to 1 ratio), the lateral and longitudinal asymmetry values (0 to 1 ratio) and the width variation along the main axis of the cell for each strain (µm). The statistical analysis of the whole data was obtained by a linear mixed-effects model (LMM) or a generalized linear mixed-effects model (GLMM). The contrast between strains was obtained from a global nonlinear model and shows significant differences; the exact P values of the two-by-two comparisons are indicated. Data information: In (C), data are presented as box plots. Each box represents the interquartile range (IQR), with the lower and upper bounds corresponding to the 25th and 75th percentiles, respectively. The center line indicates the median (50th percentile). Whiskers extend to the minimum and maximum values within 1.5 × IQR; data points beyond this range are considered outliers. ns = not significant, *P < 0.05, **P < 0.01, ****P < 0.0001 (global nonlinear model test). Source data are available online for this figure.

Figure 6. Cellular localization of RagB during cell cycle.

(A) The wild-type (WT, green) PS2062 (ponA::spec, blue), SG667 (ponA::spec ragB::erm, black) and SG1014 (ponA::spec ragB::erm amyE::P_xylgfp-ragB, yellow) strains were inoculated from an exponentially grown culture at an OD_600nm 0.1 in LB medium supplemented with 1% xylose in a 96-well plate in triplicate and grown until the stationary phase. The representative growth curves of three biological replicates are shown. (B, C) Strain SG661 (ragB::erm, amyE::P_xyl gfp-ragB) observed by phase contrast (left) or under epifluorescence illumination at 488 nm (middle in green) or 561 nm (middle in red) or a merged image (right). SG661 was grown in LB medium supplemented with 1% xylose at 37 °C, and GFP-RagB was localized by fluorescence microscopy (middle panels in green). For membrane labeling (middle panels in red), 1 ml of bacterial culture at OD_600nm = 0.4 (in exponential phase, (B) or at OD_600nm = 3.7 (in stationary phase, (C) was incubated for 1 min with 1 µl of 1 mg/ml FM4-64 reagent before preparing the slides. GFP-RagB is mainly observed at the cell membrane and enriched at the septum during the exponential phase and in the sidewalls during the stationary phase. (D, E) GFP-RagB (SG661; ragB::erm, amyE::P_xylgfp-ragB) observed by TIRF microscopy epifluorescence (E) and under bright-field illumination (D). Scale bars are 5 µm. (F) Typical kymographs monitored on tracks drawn perpendicularly to the cell long axis, on a time-lapse acquisition by TIRFM of SG661. No specific trajectory is observed. Scale bars, 1 µm (horizontal) and 10 s (vertical). Source data are available online for this figure.

Figure 7. Dynamic localization of RodA.

(A) The wild-type 168 (blue) and RCL1445 (rodA::rodA-halo; red) strains were inoculated at an OD_600nm of 0.01 in a 96-well plate in triplicate, from an exponentially grown culture, and were grown in LB medium until the stationary phase. Error bars are the standard deviation to the mean of technical replicas. (B, C) RCL1445 (rodA::rodA-halo) grown exponentially in LB medium and spotted on a 2% agarose-LB pad, observed in bright-field illumination (B) or in TIRFM (C). Cells display no shape defect (B) and fluorescent RodA shows a discrete distribution of localization in the membrane (C). Scale bars, 5 µm. (D) Typical kymographs observed on tracks drawn perpendicularly to the cell long axis, on a time-lapse acquisition by TIRFM of RCL1445. Scale bars, 1 µm (horizontal) and 10 s (vertical). Arrows indicate typical traces of moving foci. (E, F) Mean speed (E) and density (F) of directionally moving RodA-Halo subpopulation in wild-type (WT) or three mutant backgrounds: ∆ragB (RCL1690), ∆ponA (RCL1693) and ∆ragB ∆ponA (RCL1694). Plots are the sum of all values from at least 115 particles per strain, from three or four biological replicates, distinguished by colors. Large dots represent the mean for each replicate, black lines the mean of the pooled replicates. Nested t test on the non-pooled replicates were performed to analyze the differences between the means (speed or density) of the wild-type and each mutant. Only the mean density between the WT and the ∆ragB ∆ponA (RCL1694) mutant was significantly different (* = P < 0.05, P value = 0.0482). The speed differences observed in E are not significant (ns). (G, H) SPT and CDF analysis of diffusing RodA-Halo particles in wild-type (WT) and three mutant backgrounds: ∆ragB (RCL1690), ∆ponA (RCL1693) and ∆ragB ∆ponA (RCL1694), grown on s-EZRDM, revealing two subpopulations. Solid bars are median values of four biological replicates (dots). (G) Ratio of slow diffusing particles over total trajectories. (H). Diffusion coefficient (DC) of the slow (gray) and fast (blue) diffusing subpopulations. Data information: In (E, F), data are presented as Means (for each replicate and for the pool of replicates). ns not significant, *P < 0.05 (Nested t test). Source data are available online for this figure.

Figure 8. Model of the molecular role of RagB on the stimulation of RodA.

In the presence of aPBPs, GT activities of RodA and PBP1 are sufficient for PG synthesis and repair. A stimulatory effect of RagB on RodA activity is dispensable. In the absence of aPBPs (or of PBP1), RodA assumes a fraction of the repairing activity while the elongasome components are upregulated (Patel et al, 2020) and RagB stimulatory effect on RodA activity becomes essential. Indeed, enough active Rod complex should take on these two functions (synthesis and repair). In this context, RagB may detect PG damages by its IDR and recruits freely diffusible molecules of RodA which, associated to its bPBP partners, will be able to repair them. RagB could also maintain an active conformation of RodA/bPBP complex to stimulate its activity outside or within the elongasome.