Signal Peptidase-Mediated Cleavage of the Anti-σ Factor RsiP at Site 1 Controls σP Activation and β-Lactam Resistance in Bacillus thuringiensis

Abstract

In Bacillus thuringiensis, β-lactam antibiotic resistance is controlled by the extracytoplasmic function (ECF) σ factor σ^P. σ^P activity is inhibited by the anti-σ factor RsiP. In the presence of β-lactam antibiotics, RsiP is degraded and σ^P is activated. Previous work found that RsiP degradation requires cleavage of RsiP at site 1 by an unknown protease, followed by cleavage at site 2 by the site 2 protease RasP. The penicillin-binding protein PbpP acts as a sensor for β-lactams. PbpP initiates σ^P activation and is required for site 1 cleavage of RsiP but is not the site 1 protease. Here, we describe the identification of a signal peptidase, SipP, which cleaves RsiP at a site 1 signal peptidase cleavage site and is required for σ^P activation. Finally, many B. anthracis strains are sensitive to β-lactams yet encode the σ^P-RsiP signal transduction system. We identified a naturally occurring mutation in the signal peptidase cleavage site of B. anthracis RsiP that renders it resistant to SipP cleavage. We find that B. anthracis RsiP is not degraded in the presence of β-lactams. Altering the B. anthracis RsiP site 1 cleavage site by a single residue to resemble B. thuringiensis RsiP results in β-lactam-dependent degradation of RsiP. We show that mutation of the B. thuringiensis RsiP cleavage site to resemble the sequence of B. anthracis RsiP blocks degradation by SipP. The change in the cleavage site likely explains many reasons why B. anthracis strains are sensitive to β-lactams. IMPORTANCE β-Lactam antibiotics are important for the treatment of many bacterial infections. However, resistance mechanisms have become increasingly more prevalent. Understanding how β-lactam resistance is conferred and how bacteria control expression of β-lactam resistance is important for informing the future treatment of bacterial infections. σ^P is an alternative σ factor that controls the transcription of genes that confer β-lactam resistance in Bacillus thuringiensis, Bacillus cereus, and Bacillus anthracis. Here, we identify a signal peptidase as the protease required for initiating activation of σ^P by the degradation of the anti-σ factor RsiP. The discovery that the signal peptidase SipP is required for σ^P activation highlights an increasing role for signal peptidases in signal transduction, as well as in antibiotic resistance.

Keywords: cell envelope; gene expression; sigma factors; signal transduction; stress response; σ factors.

FIG 1

A predicted signal peptidase cleavage site is required for RsiP degradation. (A) Model. RsiP (red) is cleaved by RasP at site 2 (light blue), but the site 1 protease (dark blue) is unknown. PbpP was shown to play a role in β-lactam sensing and signaling for σ^P activation (44). The SignalP (50) predicted signal peptidase cleavage site (VQS) is shown along with the corresponding mutations. (B) Mutation of RsiP results in a loss of σ^P activation. All strains contain the reporter P_sigP-lacZ integrated into the thrC locus. The relevant genotypes are WT (THE2549), rsiP (EBT238), rsiP^S84W (EBT1136), rsiP^V82W (EBT1165), and rsiP^S84A (EBT1166). Cells were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) at 37°C. Cefoxitin (0.02 to 2 μg/mL) was added, and the cells were incubated for 1 h at 37°C. β-Galactosidase activities were calculated as described in Materials and Methods. Experiments were performed in technical and biological triplicates, and standard deviations are represented by error bars. a.u., arbitrary units. (C, D) Mutation of the predicted signal peptidase cleavage site inhibits RsiP degradation. A GFP-RsiP fusion was used with the following RsiP mutants in WT RsiP⁺ (EBT936), RsiP^S84W (EBT1207), RsiP^V82W (EBT1209), and RsiP^S84A (EBT1208) (C) or in ΔrasP mutant ΔrasP/RsiP⁺ (EBT939), ΔrasP/RsiP^S84W (EBT1210), ΔrasP/RsiP^V82W (EBT1212), and ΔrasP/RsiP^S84A (EBT1211) (D) strains. Cells were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) with 1 mM IPTG at 37°C, pelleted, and concentrated in LB or LB plus cefoxitin (5 μg/mL). Cells were incubated for 1 h at 37°C before sample buffer was added. Immunoblotting was performed using anti-GFP antisera. Streptavidin IR680LT was used to detect PycA (HD73_4231) and AccB (HD73_4487), which served as a loading control (68, 69). The color blot with both anti-GFP and streptavidin on a single gel is shown in Fig. S2A and B in the supplemental material. Numbers at the left indicate the molecular masses (in kilodaltons) of the ladder. The bands corresponding to the loading control and GFP-RsiP fragments are indicated on the right.

FIG 2

Identification of signal peptidases that are sufficient to activate σ^P. (A) Expression of B. thuringiensis signal peptidases in B. subtilis. All strains are B. subtilis and contain P_sigP-sigP⁺-rsiP⁺-lacZ and P_xyl-pbpP⁺. The strain-specific relevant genotypes are the empty vector (EV) (CDE3602), P_IPTG-bt2887 (CDE3603), P_IPTG-sipX (CDE3604), P_IPTG-bt1507 (CDE3605), P_IPTG-bt2973 (CDE3606), P_IPTG-bt3371 (CDE3608), P_IPTG-sipP (CDE3610), and P_IPTG-bt2898 (CDE3612). The strains were grown in the presence of xylose (1%) and IPTG (1 mM) at 37°C to an OD₆₀₀ of 1.6 to 1.8. β-Galactosidase activities were calculated as described in Materials and Methods. Experiments were performed in technical and biological triplicates, and standard deviations are represented by error bars. ns, not significant. (B) All strains are B. subtilis and contain P_sigP-sigP-rsiP-lacZ and the following relevant genotypes: P_IPTG (CDE3613), P_IPTG P_xyl-pbpP (CDE3602), P_IPTG-sipX (CDE3614), P_IPTG-sipX P_xyl-pbpP (CDE3604), P_IPTG-sipP (CDE3615), and P_IPTG-sipP P_xyl-pbpP (CDE3610). Cells were prepared and β-galactosidase activities were determined using the same methods as described for panel A. Strains grown in the absence of IPTG are shown in Fig. S3A and S3B. ****, P value of <0.0001.

FIG 3

SipP is required for σ^P activation in B. thuringiensis. (A) SipP is required for σ^P activation. All strains contain P_sigP-lacZ and the following relevant genotypes: WT (THE2549), ΔsipX (EBT1170), ΔsipP (EBT1202), and ΔsipX ΔsipP (EBT1213). (B) sipP complements ΔsipP. All strains contain P_sigP-lacZ and the following relevant genotypes: WT/EV (EBT728), WT/P_IPTG-sipP (EBT1269), ΔsipP/EV (EBT1244), ΔsipP/P_IPTG-sipP (EBT1218), ΔsipX ΔsipP/EV (EBT1246), and ΔsipX ΔsipP/P_IPTG-sipP (EBT1220). All strains were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) at 37°C. Cefoxitin (Cef, 0.02 to 2 μg/mL) was added, and the cells were incubated for another hour at 37°C. Cells were prepared and β-galactosidase activities were determined using the same methods as described for panel A. Experiments were performed in technical and biological triplicates, and standard deviations are represented by error bars.

FIG 4

sipP is required for degradation of RsiP. All strains contain P_IPTG-gfp-rsiP⁺ inserted in the ICEBs1 locus and the following relevant genotypes: WT (EBT936), ΔsipX (EBT1223), ΔsipP (EBT1222), and ΔsipX ΔsipP (EBT1224) (A) or WT (EBT936), ΔrasP (EBT939), ΔrasP ΔsipP (EBT1263), and ΔrasP ΔsipX ΔsipP (EBT1265) (B). Cells were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) with 1 mM IPTG at 37°C, pelleted, and concentrated in LB or LB plus cefoxitin (5 μg/mL). Cells were incubated for 1 h at 37°C before sample buffer was added. Immunoblotting was performed using anti-GFP antisera. Streptavidin IR680LT was used to detect PycA (HD73_4231) and AccB (HD73_4487), which served as loading controls (68, 69). The color blot with both anti-GFP and streptavidin on a single gel is shown in Fig. S4A and B. Numbers on the left indicate the molecular masses (in kilodaltons) of the ladder. The bands corresponding to the loading control and GFP-RsiP fragments are indicated on the right.

FIG 5

sipP is required for activation of σ^P by overexpression of pbpP. (A) pbpP overexpression does not activate σ^P in the absence of sipP. All strains contain P_sigP-lacZ inserted into the thrC locus and the following relevant genotypes: WT/EV (EBT728), WT/P_IPTG-pbpP⁺ (EBT1239), ΔsipX/EV (EBT1242), ΔsipX/P_IPTG-pbpP⁺ (EBT1241), ΔsipP/EV (EBT1244), ΔsipP/P_IPTG-pbpP⁺ (EBT1243), ΔsipX ΔsipP/EV (EBT1246), and ΔsipX ΔsipP/P_IPTG-pbpP⁺ (EBT1245). β-Galactosidase activities of cultures grown in the absence of IPTG are shown in Fig. S5A. (B) PbpP is produced in ΔsipP and ΔsipX ΔsipP mutants. The strains used are the same as described for panel A. After incubation to an OD of 1.6 to 1.8, 1 mL of each culture from panel A was concentrated, washed, and resuspended in Bocillin FL (50 μg/mL) for 30 min at room temperature. The color blot with both Bocillin FL and the ladder on a single gel is shown in Fig. S5B. Numbers on the left indicate the molecular masses (in kilodaltons) of the ladder. The band corresponding to PbpP is indicated on the right. (C) Overproduction of SipP can activate σ^P in the absence of PbpP. All strains contain P_sigP-lacZ inserted into the thrC locus and the following relevant genotypes: WT/EV (EBT728), WT/sipP⁺ (EBT1269), ΔpbpP/EV (EBT1270), and ΔpbpP/sipP⁺ (EBT1273). Cultures grown in the absence of IPTG as well as overexpression of bt0543 and pbpP are shown in Fig. S5C. (D) sipP overexpression is not sufficient for the cleavage of RsiP^S84W or RsiP^V82W. All strains contain P_sigP-lacZ and the following relevant genotypes: EV/EV (EBT1313), EV/P_IPTG-sipP (EBT1314), rsiP/EV (EBT1323), rsiP/P_IPTG-sipP (EBT1324), rsiP^V82W/EV (EBT1319), rsiP^V82W/P_IPTG-sipP (EBT1320), rsiP^S84W/EV (EBT1316), and rsiP^S84W/P_IPTG-sipP (EBT1317). Cultures grown in the absence of IPTG are shown in Fig. S5D. All strains were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) in the presence of IPTG (1 mM) at 37°C. β-Galactosidase activities were calculated as described in Materials and Methods. Experiments were performed in technical and biological triplicates, and standard deviations are represented by error bars.

FIG 6

pbpP and sipP are sufficient for cefoxitin-induced activation of σ^P in B. subtilis. All strains are B. subtilis and contain P_sigP-sigP⁺-rsiP⁺-lacZ and the following relevant genotypes: EV/EV (CDE3613), EV/P_xyl-pbpP⁺ (CDE3602), P_IPTG-sipP⁺/EV (CDE3615), and P_IPTG-sipP⁺/P_xyl-pbpP⁺ (CDE3610). All strains were grown to early mid-log phase (OD₆₀₀, 0.6 to 0.8) in the presence of IPTG (0.01 mM) and xylose (0.01%) at 37°C. Cefoxitin (0.05 to 0.2 μg/mL) was added, and the cells were incubated for another hour at 37°C. β-Galactosidase activities were calculated as described in Materials and Methods. Experiments were performed in technical and biological triplicates, and standard deviations are represented by error bars.

FIG 7

The signal peptidase cleavage site in RsiP from B. anthracis blocks RsiP degradation. All strains contain the following relevant genotypes: GFP-RsiP_Bt (EBT936), GFP-RsiP^S84I_Bt (EBT1328), GFP-RsiP_Ba (EBT1329), and GFP-RsiP^I84S_Ba (EBT1330). Cells were grown to mid-log phase (OD₆₀₀, 0.6 to 0.8) with 1 mM IPTG at 37°C, pelleted, and concentrated in LB or LB plus cefoxitin (5 μg/mL). Cells were incubated for 1 h at 37°C before sample buffer was added. Immunoblotting was performed using anti-GFP antisera. Streptavidin IR680LT was used to detect PycA (HD73_4231) and AccB (HD73_4487), which served as loading controls (68, 69). A color blot with both anti-GFP and streptavidin on a single gel is shown in Fig. S7. Numbers at the right indicate the molecular masses (in kilodaltons) of the ladder.

FIG 8

Model for σ^P activation. PbpP (gray) binds β-lactams (black) and conformationally changes. This allows a direct or indirect interaction between RsiP (red) or SipP (dark blue). This interaction results in site 1 cleavage of RsiP (red) by SipP (dark blue). Site 1 cleavage allows for site 2 cleavage by RasP (light blue). Degradation of RsiP (red) results in the release of σ^P (green).