ABSTRACT
Extracytoplasmic function (ECF) σ factors are a diverse family of alternative σ factors that allow bacteria to sense and respond to changes in the environment. σV is an ECF σ factor found primarily in low-GC Gram-positive bacteria and is required for lysozyme resistance in several opportunistic pathogens. In the absence of lysozyme, σV is inhibited by the anti-σ factor RsiV. In response to lysozyme, RsiV is degraded via the process of regulated intramembrane proteolysis (RIP). RIP is initiated by cleavage of RsiV at site 1, which allows the intramembrane protease RasP to cleave RsiV within the transmembrane domain at site 2 and leads to activation of σV. Previous work suggested that RsiV is cleaved by signal peptidase at site 1. Here we demonstrate in vitro that signal peptidase is sufficient for cleavage of RsiV only in the presence of lysozyme and provide evidence that multiple Bacillus subtilis signal peptidases can cleave RsiV in vitro. This cleavage is dependent upon the concentration of lysozyme, consistent with previous work that showed that binding to RsiV was required for σV activation. We also show that signal peptidase activity is required for site 1 cleavage of RsiV in vivo. Thus, we demonstrate that signal peptidase is the site 1 protease for RsiV.
IMPORTANCE Extracytoplasmic function (ECF) σ factors are a diverse family of alternative σ factors that respond to extracellular signals. The ECF σ factor σV is present in many low-GC Gram-positive bacteria and induces resistance to lysozyme, a component of the innate immune system. The anti-σ factor RsiV inhibits σV activity in the absence of lysozyme. Lysozyme binds RsiV, which initiates a proteolytic cascade leading to destruction of RsiV and activation of σV. This proteolytic cascade is initiated by signal peptidase, a component of the general secretory system. We show that signal peptidase is necessary and sufficient for cleavage of RsiV at site 1 in the presence of lysozyme. This report describes a role for signal peptidase in controlling gene expression.
INTRODUCTION
The ability of bacteria to respond to stress is important for survival under a wide variety of environmental conditions. Bacteria utilize transmembrane signal transduction mechanisms to transmit information from outside the cell across the cell membrane, to alter transcriptional responses and adapt to a variety of environmental stresses (1, 2). One type of signal transduction system bacteria use to respond to stress is alternative σ factors, which can initiate large changes in gene expression in response to environmental conditions (1, 3). The extracytoplasmic function (ECF) σ factors represent a diverse family of alternative σ factors that can specifically respond to extracellular signals (4, 5). The activities of most ECF σ factors are inhibited by an anti-σ factor (1, 4, 5). ECF σ factor activity is induced by modifying the activity of the anti-σ factor. This can occur by one of several mechanisms: (i) degradation of the anti-σ factor in response to cell envelope stress (1, 4, 5), (ii) sensing of a signal by the anti-σ factor, which results in a conformational change that releases the σ factor (1, 4, 5), and (iii) a partner switching mechanism in which the phosphorylation of an anti-anti-σ factor allows it to bind the anti-σ factor, freeing the σ factor (1, 4, 5).
Regulated intramembrane proteolysis (RIP) is one mechanism by which the activities of several ECF σ factors are controlled (6, 7). RIP involves the degradation of a membrane-spanning anti-σ factor by a series of proteases that act in a sequential manner. The initial cleavage occurs at site 1, followed by cleavage at site 2 within the transmembrane segment of the anti-σ factor by a membrane-embedded protease. The remaining portion of the anti-σ is further degraded by cytosolic proteases (6–8). In nearly all cases, site 1 cleavage is the rate-limiting step for σ factor activation and the remainder of the process is constitutive (9–12). Thus, the main regulatory step in ECF σ factor activation is site 1 cleavage of the anti-σ factor. The site 1 protease is thought to be responsible for sensing the stress signal and initiating RIP of the anti-σ factor in several ECF σ factor systems. For example, the binding of unfolded outer membrane proteins to the site 1 protease DegS initiates site 1 cleavage of RseA the σE anti-σ factor in Escherichia coli (8, 11–15). Similarly, it is thought that PrsW, the site 1 protease controlling σW activation in Bacillus subtilis, is the sensor for cell envelope stress. This is supported by the isolation of gain-of-function prsW mutants that promote cleavage of the anti-σ factor RsiW and thus σW activation even in the absence of inducing signal (16, 17).
Bacillus subtilis is a Gram-positive, soil-dwelling bacterium that inhabits a variable and competitive environment. B. subtilis contains seven known ECF σ factors (σM, σW, σV, σX, σY, σZ, and σYlaC). The best understood of these ECF σ factors are σM, σW, σX, and σV. σM activity is required for transcription of genes that are essential for cell envelope synthesis and division, and it is induced by antibiotics, acid, and ethanol (18–26). The activity of σW is induced by a variety of cell envelope stresses, including antimicrobial peptides, detergents, and alkali stress (16, 19, 27–29). Antibiotics that inhibit cell wall synthesis induce σX activity, which regulates genes that serve to alter the surface properties, providing protection against antimicrobial peptides (30, 31).
The B. subtilis ECF σ factor σV belongs to the ECF30 subfamily, primarily found in Firmicutes or low-GC Gram-positive bacteria (32, 33). σV is activated by lysozyme and induces resistance to lysozyme (33, 34). Lysozyme is a component of the innate immune system and can cleave the β-(1,4)-linked N-acetylglucosamine and N-acteylmuramic acid residues in the peptidoglycan of bacterial cells (35–37). In B. subtilis, σV is required for lysozyme-inducible expression of oatA, which encodes a peptidoglycan O-acetylase, and dltABCDE, which changes the cell surface charge by d-alanylating cell wall teichoic acids (33, 34, 38). σV homologs from related bacteria, Clostridium difficile and Enterococcus faecalis, are also induced by lysozyme (39–41). In addition, σV homologs from C. difficile and E. faecalis control expression of genes whose products are involved in peptidoglycan modification and d-alanylation of cell wall teichoic acids (41–44).
In the absence of lysozyme, the activity of σV is inhibited by the transmembrane anti-σ factor RsiV (45, 46). Previous work from our laboratory has shown that activation of σV requires RIP-mediated degradation of RsiV (9). We also demonstrated that RsiV acts as a receptor for lysozyme and binding to lysozyme is required for σV activation (47, 48). Upon binding to lysozyme, the extracellular domain of RsiV is apparently cleaved at site 1 by signal peptidase (47). This leads to cleavage of RsiV at site 2 by the membrane protease RasP (9). The cytosolic portion of RsiV is presumably degraded by cytosolic proteases, resulting in the release of σV and allowing σV to bind to RNA polymerase.
Signal peptidases are a critical component of the cellular secretion system in bacterial cells (49). Membrane-bound type 1 signal peptidases are responsible for the cleavage of the signal peptides from preproteins during or shortly after translocation across the membrane (49–52). Preproteins contain a signal peptide sequence which is recognized by signal peptidase with high fidelity (53). Once removed, the mature protein is released from the membrane. In Gram-positive bacteria, signal peptidases contain a short cytoplasmic domain of a few amino acids followed by a single transmembrane domain that anchors the enzyme to the membrane. The C-terminal domain contains the active site and is involved in the recognition of the correct cleavage site (54). B. subtilis contains four chromosomally encoded type I prokaryotic signal peptidases, SipS, SipT, SipU, and SipV, and one eukaryotic type I signal peptidase, SipW (49). The two major signal peptidases are SipS and SipT, while SipU, SipV, and SipW are minor contributors to the processing of secretory preproteins (53). SipS and SipT are of major importance for the secretion of degradative enzymes and cell viability. In the absence of SipT, a functional SipS is required for growth, as cells lacking both SipS and SipT are not viable (55). This shows that the presence of a single major signal peptidase is sufficient for growth of B. subtilis, suggesting that the secretory precursor processing machinery is functionally redundant.
In the present study, we developed an in vitro assay to monitor site 1 cleavage of RsiV. We found, using purified proteins, that SipS and SipT, but not SipU or SipV, can efficiently cleave RsiV in vitro only in the presence of lysozyme. We determined that site 1 cleavage of RsiV by signal peptidase is dependent upon the lysozyme concentration. Finally, we show that either SipS or SipT is required for site 1 cleavage of RsiV in response to lysozyme in vivo. Thus, either of the major B. subtilis signal peptidases, SipS or SipT, is necessary and sufficient for site 1 cleavage of RsiV in the presence of lysozyme.
RESULTS
Full-length SipS is sufficient for cleavage of RsiV in vitro.We previously demonstrated that SipS induces cleavage of RsiV at site 1 in vitro and that this cleavage occurs only in the presence of lysozyme (47). However, both RsiV and SipS were produced by in vitro transcription/translation using wheat germ extract; thus, these proteins had not been purified (47). To determine if SipS was sufficient to cleave RsiV at site 1, we purified full-length RsiV, full-length SipS, and a soluble form of SipS, which lacks the transmembrane domain SipS35-184 (see Fig. S1 in the supplemental material). We combined RsiV with the extracellular domain of SipS35-184 in the presence or absence of lysozyme and incubated the reaction mixtures at 37°C for 10 min or 2 h. We found that in the presence of lysozyme, SipS35-184 cleaves RsiV at site 1; however, this cleavage was not efficient, as only ∼8% of RsiV was cleaved after 2 h (Fig. 1B). We confirmed that RsiV was cleaved at site 1 between A66 and M67 by performing N-terminal sequencing (Fig. 1A). This is the same position as we previously observed in vivo (47). Interestingly, we found that in the absence of lysozyme, SipS35-184 cleaved full-length RsiV but the size of the cleaved product was smaller than the expected (Fig. 1B; see also Fig. S2 in the supplemental material). We hypothesized that the extracellular domain of SipS35-184 cleaves RsiV at a different site when incubated in the absence of lysozyme. N-terminal sequencing confirmed that the extracellular domain of SipS35-184 cleaves RsiV between A79 and I80, 13 amino acids from the in vivo cleavage site, in the absence of lysozyme (Fig. 1A).
Full-length signal peptidase is required for efficient site 1 cleavage RsiV in vitro. (A) Model showing location of site 1 cleavage of RsiV. The ECF σ factor σV binds in the N-terminal cytoplasmic portion of RsiV. Lysozyme binds in the C-terminal region. The arrow indicates that RsiV is cleaved between A66 and M67. (B) Cleavage assays were performed using purified 6×His-RsiV, 6×His-SipS, the extracellular domain of SipS (SipS35-184), and lysozyme. SipS (4 μM) or the extracellular domain of SipS35-184 (20 μM) was added to RsiV (20 μM) in the presence (+) or absence (−) of lysozyme (50 μM) and incubated at 37°C for 10 min or 2 h. Reactions were stopped by the addition of an equal volume of 2× Laemmli sample buffer. Samples were analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue.
We then combined full-length SipS and RsiV with or without lysozyme and incubated the reaction mixtures at 37°C for 10 min or 2 h (Fig. 1B). After 10 min, ∼18% of RsiV was cleaved by SipS in the absence of lysozyme, and after 2 h ∼20% of the protein was cleaved (Fig. 1B). In the presence of lysozyme, ∼85% of RsiV was cleaved after 10 min of incubation with SipS (Fig. 1B). We again confirmed that full-length SipS cleaved RsiV between A66 and M67 by N-terminal sequencing (Fig. 1A). This suggests that full-length SipS recognizes and cleaves RsiV in vitro efficiently only in the presence of lysozyme. Our data also show that full-length SipS is more efficient than soluble SipS at cleaving RsiV, similar to previous work which showed soluble SipS is 100-fold less active than full-length SipS (54). Thus, we chose to use full-length signal peptidase for the remainder of our studies.
Cleavage of RsiV is lysozyme dependent.Our data suggest that in the presence of lysozyme, full-length SipS cleaves RsiV at the expected signal peptide cleavage site in vitro. We decided to investigate the requirements for cleavage of RsiV by SipS. We performed a time course analysis with RsiV and full-length SipS in both the presence and absence of lysozyme. We found that in the absence of lysozyme, full-length SipS cleaves RsiV, although with limited efficiency (Fig. 2A). Note that after 1 min ∼8% of RsiV was cleaved and after 2 h ∼32% of the protein was cleaved at site 1 (Fig. 2A). In contrast, when RsiV was incubated with full-length SipS in the presence of lysozyme, SipS rapidly cleaves RsiV, with ∼96% of RsiV being cleaved at site 1 within 10 min (Fig. 2A). This suggests that in detergent, in the absence of all other B. subtilis proteins, SipS can cleave RsiV efficiently only when lysozyme is present.
Signal peptidase is sufficient for site 1 cleavage of RsiV in vitro. (A) Site 1 cleavage of full-length RsiV is lysozyme dependent. A cleavage assay was performed with RsiV, 6×His-SipS, and lysozyme. SipS (4 μM) was added to RsiV (20 μM) in the presence or absence of lysozyme (50 μM) and allowed to incubate at 37°C. (B) Degradation of RsiV is blocked in the presence of lysozyme and SipSS43A. RsiV is not cleaved by SipSS43A in the absence or presence of lysozyme. Cleavage assays were performed in which purified SipSS43A (4 μM) was added to RsiV (20 μM) in the presence or absence of lysozyme (50 μM) and incubated at 37°C for 10, 30, 60, or 120 min. (C) Degradation of RsiVA66W is blocked in the presence of lysozyme and SipS. RsiVA66W is not cleaved by SipS in the presence or absence of lysozyme. Cleavage assays were performed using purified RsiVA66W, SipS, and lysozyme. SipS (4 μM) was added to RsiVA66W (14 μM) in the presence or absence of lysozyme (50 μM) and incubated at 37°C for 10, 30, 60, or 120 min. Reactions were stopped by the addition of an equal volume of 2× Laemmli sample buffer. Samples were analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue.
Disruption of SipS active site blocks RsiV degradation.We sought to confirm that SipS activity was required to cleave RsiV at site 1 in vitro by constructing an active-site mutant of full-length SipS (56). The serine at position 43 was mutated to an alanine in SipS and tested for its ability to cleave RsiV. Purified RsiV and SipSS43A were incubated in the presence and absence of lysozyme for various times at 37°C. We found that SipSS43A was unable to cleave RsiV in the absence of lysozyme (Fig. 2B). After 2 h, we observed a small amount of cleaved RsiV in the presence of lysozyme (Fig. 2B). This putative cleavage product was confirmed to be RsiV by Western blotting using anti-RsiV59-285 antibodies (Fig. S3). This suggests that SipSS43A is almost completely blocked for site 1 cleavage of RsiV in vitro.
Disruption of the RsiV cleavage site blocks degradation by SipS in the presence of lysozyme.Previous data demonstrated that site 1 cleavage of RsiV in vivo can be blocked by mutating the alanine at position 66 of RsiV to tryptophan, yielding RsiVA66W (47). Thus, we sought to determine if the RsiVA66W mutant protein blocked cleavage at site 1 by SipS in vitro. We purified RsiVA66W and combined it with purified SipS with or without lysozyme at 37°C. We found that RsiVA66W was not cleaved by SipS in the absence or presence of lysozyme (Fig. 2C). This suggests that disruption of site 1 in full-length RsiV prevents cleavage by SipS in vitro.
RsiV cleavage occurs in a SipS concentration-dependent manner.Our data indicate that SipS is sufficient for cleavage of RsiV at site 1 in vitro. We sought to determine the effect of SipS concentration on the cleavage of RsiV at site 1 in the presence of lysozyme. Increasing concentrations of full-length SipS were added to purified RsiV and lysozyme at 37°C for 10 min. We observed that as the concentration of SipS increased, RsiV cleavage at site 1 increased in response (Fig. 3A). At the highest concentrations of SipS (4.3 μM), ∼93% of RsiV was cleaved at site 1 (Fig. 3A). We repeated the assay in the absence of lysozyme. We found that we were able to detect cleaved RsiV only at the highest concentrations of SipS (Fig. 3B). This provides further evidence that SipS efficiently cleaves RsiV at site 1 only in the presence of lysozyme.
Site 1 cleavage of RsiV increases with SipS concentration. (A) A SipS concentration dependence assay was performed using purified 6×His-RsiV, 6×His-SipS, and lysozyme. SipS was added (final concentrations, 0, 0.14, 0.28, 0.57, 1.14, 2.3, and 4.2 μM) to 6×His-RsiV (20 μM) and lysozyme (50 μM). Samples were incubated at 37°C for 10 min, and the reactions were stopped by the addition of an equal volume of 2× Laemmli sample buffer. (B) A SipS concentration dependence assay was performed using purified 6×His-RsiV and 6×His-SipS in the absence of lysozyme. SipS was added (final concentrations, 0, 0.14, 0.28, 0.57, 1.14, 2.3, and 4.2 μM) to 6×His-RsiV (20 μM). Samples were incubated at 37°C for 10 min, and reactions were stopped by the addition of an equal volume 2× Laemmli sample buffer. Samples were analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue.
Signal peptidase cleavage of RsiV is dependent upon lysozyme concentration.Based on our previous work, we developed a model that RsiV binding to lysozyme is required for site 1 cleavage of RsiV by signal peptidase. Thus, we sought to determine the effect of lysozyme concentration on the ability of SipS to induce RsiV degradation. Increasing concentrations of lysozyme were added to purified full-length RsiV and full-length SipS and incubated for 10 min at 37°C. At higher lysozyme concentrations, ∼92% of RsiV was cleaved by SipS (Fig. 4). We observed that as the concentration of lysozyme (14 μM) fell below a 1:1 molar ratio with RsiV, the amount of RsiV cleaved also decreased (Fig. 4). This is consistent with the previous observation that RsiV binds lysozyme at a 1:1 molar ratio and provides further evidence that cleavage of RsiV is dependent upon binding lysozyme (47, 48).
Site 1 cleavage of RsiV increases with lysozyme concentration. A lysozyme concentration dependence assay was performed using purified 6×His-RsiV, 6×His-SipS, and lysozyme. Lysozyme was added (final concentrations, 0, 0.56, 1.4, 3.5, 5.6, 7.3, 14, 28, 56, 91, and 118 μM) to RsiV (20 μM) and SipS (3 μM) and allowed to incubate at 37°C for 10 min. The reactions were stopped by the addition of an equal volume of 2× Laemmli sample buffer. Samples were analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue.
SipT can cleave RsiV at site 1 in the presence of lysozyme.B. subtilis contains four bacterial type 1 signal peptidases, with SipS and SipT as the two major type 1 signal peptidases (53). We sought to determine if other signal peptidases are able to cleave RsiV at site 1. We purified full-length SipT, SipU, and SipV and combined them with RsiV in the presence and absence of lysozyme at 37°C. We found that in the presence of SipT and in the absence of lysozyme, ∼2% of RsiV was cleaved after 10 min and ∼18% of RsiV was cleaved after 2 h at site 1 (Fig. 5A). In the presence of lysozyme, SipT cleaved ∼87% of RsiV after 10 min, and after 2 h, 95% of RsiV had been cleaved. This suggests that similar to the case with SipS, the major signal peptidase SipT is able to efficiently cleave RsiV in the presence of lysozyme.
Multiple signal peptidases cleave RsiV in vitro in the presence of lysozyme. The ability of signal peptidases SipT, SipU, and SipV to cleave RsiV in vitro was tested. (A) RsiV is cleaved at site 1 by SipT in the absence and presence of lysozyme. Purified SipT (4 μM) was added to RsiV (14 μM) in the presence and absence of lysozyme and incubated at 37°C for various times. (B) RsiV is not efficiently cleaved by SipU. Full-length SipU (3.7 μM) was added to RsiV (14 μM) in the presence and absence of lysozyme (50 μM) and allowed to incubate at 37°C. (C) RsiV is not efficiently cleaved at site 1 by SipV in the presence of lysozyme. Purified full-length SipV (4.4 μM) was incubated with RsiV (14 μM) in the presence and absence of lysozyme (50 μM) at 37°C. Samples were analyzed by 10% SDS-PAGE and stained with Coomassie brilliant blue.
We performed a time course analysis with RsiV and purified SipU in the presence and absence of lysozyme. In the absence of lysozyme, we were not able to detect a cleaved product for RsiV (Fig. 5B). After 2 h of incubation in the presence of lysozyme, ∼7% of RsiV was cleaved at site 1 by SipU (Fig. 5B). Western blot analysis confirmed that this putative cleaved product was RsiV (Fig. S3). This suggests that SipU can cleave RsiV at site 1 in vitro but does not do so as efficiently as the major signal peptidases SipS and SipT.
RsiV was incubated with purified SipV with or without lysozyme for various times. Similar to SipU, a cleaved product was not detected for RsiV in the absence of lysozyme (Fig. 5C). In the presence of lysozyme, SipV is not able to efficiently cleave RsiV (Fig. 5C). Although we detected a band after 2 h in the presence of lysozyme, Western blot analysis using anti-RsiV59-285 antibodies showed that it does not cross-react, suggesting that it is not RsiV (Fig. S3). Taken together, our data demonstrate that in vitro the major signal peptidases, SipS and SipT, efficiently cleave RsiV at site 1 in the presence of lysozyme, while the minor signal peptidases SipU and SipV do not.
Signal peptidase activity is required for site 1 cleavage of RsiV in vivo.We previously demonstrated that mutant strains lacking either SipS or SipT did not block degradation of RsiV in vivo (47). Since SipS and SipT are sufficient for site 1 cleavage of RsiV in vitro and others have shown that SipS and SipT are the major signal peptidases of B. subtilis (53, 57), we chose to determine if SipS and SipT were required for site 1 cleavage of RsiV in vivo. We constructed strains containing either ΔsipS or ΔsipT null mutations and created a temperature-sensitive allele of sipS, sipSR84H (55). We chose to construct the sipSR84H mutant because it was previously shown that SipSR84H was rapidly degraded upon shifting of cells to 48°C (55, 57).
We measured cleavage of RsiV in strains in the presence and absence of lysozyme at permissive and nonpermissive temperatures. We found that RsiV was cleaved at site 1 in the presence of lysozyme in all strains at the permissive temperature of 25°C (Fig. 6; see also Fig. S4). However, when cells were shifted to 48°C, RsiV was not cleaved in the presence of lysozyme at site 1 in the strain which lacked SipT and contained temperature-sensitive SipSR84H (ΔsipT sipS(Ts) [Fig. 6; see also Fig. S4]). In contrast, at both 25°C and 48°C, RsiV was cleaved at site 1 in both the sipS and sipT single-mutant strains (Fig. 6; see also Fig. S4). This demonstrates that either SipS or SipT is required for site 1 cleavage of RsiV and that without a functional copy of sipS or sipT, efficient cleavage of RsiV does not occur. Thus, signal peptidase is the site 1 protease for RsiV.
Signal peptidase is required for the cleavage of RsiV in the presence of lysozyme. B. subtilis wild-type (LTL380), ΔsipT (LTL381), ΔsipS (LTL413), sipS(Ts) (LTL412), and ΔsipT sipS(Ts) (LTL414) strains were subcultured 1:100 from an overnight culture and grown to an OD600 of 0.8 to 1 at 30°C. The cultures were then split in half and incubated at either 25°C (A) or 48°C (B) for 30 min. Cells were pelleted then resuspended in 100 μl of LB with or without 100 μg/ml of lysozyme and incubated 25°C or 48°C for 10 min. One hundred microliters of 2× Laemmli sample buffer was added to stop the reaction, and cells were lysed by sonication. The samples were electrophoresed on a 15% SDS-PAGE gel, and proteins were detected by Western blotting using anti-RsiV59-285 rabbit serum and goat anti-rabbit IgG–IRDye 800CW and streptavidin-IRDye 680LT as a loading control. The gel is shown in color in Fig. S4.
DISCUSSION
Signal peptidase is responsible for site 1 cleavage of RsiV.Here we provide evidence that signal peptidase is the protease required for site 1 cleavage of RsiV. This is supported by the following observations. (i) RsiV is not cleaved at site 1 in a B. subtilis ΔsipT sipS(Ts) double mutant. (ii) Multiple purified signal peptidases are sufficient to cleave RsiV at site 1 in vitro. (iii) Signal peptidase cleavage of RsiV in vitro is lysozyme dependent. (iv) Disruption of the signal peptide cleavage site of RsiV blocks cleavage. (v) Mutation of the active site of signal peptidase blocks cleavage of RsiV. Taken together, these data demonstrate that signal peptidase is responsible for site 1 cleavage of RsiV both in vitro and in vivo. This provides a novel role for signal peptidase in controlling activation of a bacterial stress response system.
SipS and SipT are necessary and sufficient for site 1 cleavage of RsiV.In species which encode multiple type I signal peptidases, like B. subtilis, some of the signal peptidases have been reported to be more important for efficient preprotein processing than others (53). In B. subtilis, SipS and SipT the major signal peptidases, while SipU and SipV have a minor role in protein secretion and are involved in nonessential processes (50, 53, 55). In strains lacking genes for multiple signal peptidases, i.e., ΔsipSUVW or ΔsipTUVW strains, there is only a mild defect in preprotein processing and secretion (49). Cells lacking both SipS and SipT completely block signal peptide cleavage and are not viable, likely due to the jamming of the secretion machinery with secretory preproteins (49, 53, 55).
Our in vivo and in vitro experiments suggest that in the absence of SipS and SipT, SipU and SipV are not sufficient for cleavage of RsiV. Consistent with these observations, we found that SipS and SipT cleave RsiV at site 1 efficiently only in the presence of lysozyme. In contrast, we found that SipU and SipV are not able to efficiently cleave RsiV in vitro even in the presence of lysozyme. While we cannot rule out the possibility that SipU and SipV are not active under our in vitro conditions, it is clear that they are not sufficient for site 1 cleavage of RsiV in vivo. Taken together, these data suggest that SipS and SipT are the major signal peptidases required for cleavage of RsiV at site 1. Given the inability of SipU to cleave RsiV and the inefficiency by which SipV cleaves RsiV in vitro, it is tempting to speculate that it may be due to an inability of these signal peptidases to recognize RsiV. This is supported by the observation that the replacement of the amino-terminal residues of the minor signal peptidase SipV with the corresponding residues of the major signal peptidase SipS was sufficient to convert a minor signal peptidase into a major signal peptidase (49, 50).
Benefits of using signal peptidase to control σV activation.Upon binding to lysozyme, RsiV is cleaved first by a signal peptidase and then by a membrane-embedded site 2 protease, leading to σV activation. σV homologs in C. difficile and E. faecalis are activated by lysozyme (39, 40, 42, 43). Interestingly, RsiV homologs and other ECF σ factors are often horizontally transferred between strains (32). One advantage of utilizing the signal peptidase to cleave RsiV is that signal peptidase and site 2 protease are present in all cells. Since the binding of RsiV to lysozyme is the signal for initiating site 1 cleavage by signal peptidase, if an ECF σ factor is horizontally transferred it does not require an additional protease to also be horizontally transferred. Instead, the horizontally transferred ECF σ factors “plug” into a system present in all cells to control σ factor activation. However, at this time it is not known if signal peptidase is responsible for site 1 cleavage of E. faecalis or C. difficile RsiV.
Our work demonstrates a novel role for signal peptidase in controlling gene expression and raises the following question: are there other examples of inducible signal peptidase playing a role in regulation? One potential example is the β-lactam sensor/signal transducer BlaR1 in Staphylococcus aureus (58). BlaR1 binds to antibiotics and transduces the information to the cytoplasm, leading to the removal of the sensor domain by signal peptidase (59). However, it is not yet known if these cleavage events play a critical role in signal transduction (59, 60).
RsiV is the sensor for lysozyme.We demonstrate that SipS and SipT are sufficient for site 1 cleavage of RsiV in vitro and that efficient cleavage is dependent upon the presence of lysozyme. We confirmed that SipS cleaves RsiV at the same position in vitro as in vivo by performing N-terminal sequencing on the cleaved product. We also show that a mutant of the signal peptide cleavage site, RsiVA66W, blocks signal peptidase cleavage in vitro.
We show that cleavage is dependent upon the concentration of lysozyme. As the lysozyme concentration falls below an ∼1:1 ratio, the amount of RsiV cleaved decreases in a linear manner. These results are consistent with the previous observation that RsiV binds lysozyme at a 1:1 ratio and that cleavage in vivo is dependent upon the concentration of lysozyme (47, 48). Taken together, these data demonstrate that site 1 cleavage of RsiV is signal peptidase and lysozyme dependent. Thus, these studies provide further evidence that RsiV is the sensor for lysozyme and support our previous studies showing that binding of RsiV to lysozyme is required for degradation of RsiV and activation of σV (48). However, several questions remain. How does RsiV avoid cleavage by signal peptidase in the absence of lysozyme? What are the structural changes that occur upon lysozyme binding that allow RsiV to be efficiently recognized by signal peptidase and cleaved at site 1?
MATERIALS AND METHODS
Bacterial growth conditions and medium supplements.All E. coli and B. subtilis strains were grown in LB broth or on LB agar. Antibiotics were used at the following concentrations: ampicillin, 100 μg/ml; spectinomycin, 100 μg/ml; erythromycin plus lincomycin, 1 μg/ml and 25 μg/ml; and kanamycin, 10 μg/ml. Isopropyl-β-d-1-thiogalactopyranoside (IPTG) was used at a final concentration of 1 mM unless otherwise noted.
Strain construction.All B. subtilis strains used are isogenic derivatives of PY79, a prototrophic derivative of B. subtilis strain 168 (61). All strains used in this study are listed in Table 1. All plasmid constructs are listed in Table 2 and were confirmed by DNA sequencing (Iowa State DNA Sequencing Facility). For purification purposes, plasmids were introduced into E. coli BL21λDE3 Rosetta for overexpression of proteins.
Strains used in this study
Plasmids used in this study
The ΔsipS::erm strain was obtained from the Bacillus Genetic Stock Center (62). The markerless deletion ΔsipS was created with pDR244 looping out the BKE cassette via lox sites flanking the cassette (62).
The temperature-sensitive sipS mutant (sipSR84H [57]) was introduced onto the chromosome of PY79 by homologous recombination using the temperature-sensitive plasmid pMAD (63). PCR products for sipSR84H were amplified using primer pairs CDEP3169/CDEP1722 and CDEP3170/CDEP1721. The resulting PCR products were cloned into SmaI-digested pMAD using isothermal assembly (64). The sipSR84H mutation was introduced into a ΔsipS mutant and confirmed by PCR and DNA sequencing.
To construct the ΔsipT:spec mutation, the regions flanking sipT were amplified by PCR with primer pairs CDEP1701/CDEP1711 and CDEP1712/CDEP1704. The spectinomycin antibiotic resistance cassette was amplified using primer pair CDEP1954/CDEP1955 and pAH54 as a template. The resulting PCR products were assembled by isothermal assembly (64). The assembled products were transformed into B. subtilis, resulting in JLH858, and the deletion was confirmed by PCR.
To construct the ΔsigV::kn Phs-rsiV+ mutant, the regions flanking sigV were amplified by PCR with primer pairs CDEP1031/CDEP1690 and CDEP1691/CDEP1012. The kanamycin antibiotic resistance cassette and Phs (an IPTG-dependent promoter) were amplified from pDP111 using primer pair CDEP1692/CDEP1693. The resulting PCR products were assembled by isothermal assembly (64). The assembled products were transformed into B. subtilis, resulting in CDE2373, and the deletion was confirmed by PCR.
To clone rsiV, we conducted PCR amplification with rsiV using primer pair CDEP3267/CDEP3165. The resulting PCR product was cloned into pET21b digested with NcoI and BamHI using isothermal assembly, resulting in pAC108. To clone rsiVA66W, we conducted PCR amplification with rsiV using primer pairs CDEP1561/CDEP3165 and CDEP1562/CDEP3267. The resulting PCR products were cloned into pET21b digested with NcoI and BamHI using isothermal assembly, resulting in pAC115 (64).
All plasmids were constructed initially in the E. coli strain Omnimax-2 (Invitrogen). All oligonucleotide sequences are listed in Table 3. To generate constructs for purification, the target genes were amplified with primers listed and cloned into the expression vector pET21 6′His-rtev digested with the enzymes NcoI and BamHI (New England BioLabs): SipS35-184 (CDEP3054/CDEP3055; pAC112), full-length SipS (CDEP3385/CDEP3054; pAC110), SipT (CDEP3363/CDEP3364; pCE601), SipU (CDEP3367/CDEP3368; pCE602), and SipV (CDEP3365/CDEP3366; pCE603).
Oligonucleotides used in this study
To clone full-length SipSS43A (CDEP3386/CDEP3387), we conducted PCR amplification with sipS using primer pairs CDEP3054/CDEP3387 and CDEP3388/CDEP3055. The resulting PCR products were cloned into NcoI- and BamHI-HF digested pET21 6′His-rtev using isothermal assembly, resulting in pAC113.
Expression of recombinant proteins in E. coli.Overnight cultures of E. coli BL21λDE3 containing either RsiV (ANC108), RsiVA66W (ANC128), SipS38-184 (ANC114), SipS (ANC220), SipSS43A (ANC232), SipT (ANC259), SipU (ANC258), or SipV (ANC260) were subcultured 1:100 into 100 ml of LB plus ampicillin. Cultures were incubated at 37°C to an optical density at 600 nm (OD600) of 0.5 to 0.6. IPTG was added to a final concentration of 1 mM to induce protein production. Cultures were incubated for an additional 4 h at 30°C. Cells were collected by centrifugation at 4,000 × g for 15 min. Cell pellets were stored at −80°C until time for purification.
Purification of 6×His-tagged proteins from E. coli.Cell pellets were thawed on ice and resuspended in 2 ml of lysis buffer (50 mM Tris, 250 mM NaCl, 10 mM imidazole, 3 mM Triton X-100 [pH 8]) per 50 ml of initial culture volume. Cells were lysed by sonication twice, and the lysate was centrifuged at 47,000 × g for 1 h at 25°C to pellet cellular debris. Cleared lysate was applied to a nickel affinity column to bind the 6×His-tagged protein (Thermo Scientific). The column was washed with 5 ml of wash buffer (50 mM Tris, 250 mM NaCl, 20 mM imidazole, 0.3 mM Triton X-100 [pH 8]). Protein was eluted with elution buffer (50 mM Tris, 250 mM NaCl, 250 mM imidazole, 0.03 mM Triton X-100 [pH 8]) and collected in 0.5-ml fractions. Samples from each fraction were analyzed by SDS-PAGE and stained with Coomassie brilliant blue. Protein concentration was determined using the bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific). The partially purified signal peptidases are shown in Fig. S1.
Standard RsiV in vitro cleavage assay conditions.For in vitro assays, 25 μl of purified full-length RsiV (14 μM) was incubated with 10 μl of full-length SipS (3.9 μM) in the presence or absence of 5 μl of 50 mM hen egg white (HEW) lysozyme (Sigma; >98% pure). Reaction mixtures were incubated for various times at 37°C, and reactions were stopped by the addition of 40 μl of 2× Laemmli sample buffer. Samples were analyzed by SDS-PAGE and stained with Coomassie brilliant blue. All cleavage assays were performed at a minimum in triplicate, and a representative gel is shown. The percent cleaved RsiV was determined by quantifying the intensities of full-length RsiV and cleaved RsiV using Image Studio Lite.
SipS concentration dependence assay.Increasing concentrations of full-length SipS were mixed with purified full-length RsiV (13 μM) in the presence and absence of HEW lysozyme (50 μM; Sigma; >98% pure). The mixture (45 μl) was incubated at 37°C for 10 min. Reactions were analyzed by SDS-PAGE; the gel was stained with Coomassie brilliant blue and imaged using the Gel Doc XR+ system (Bio-Rad).
Lysozyme concentration dependence assay.Increasing concentrations of HEW lysozyme (Sigma; >98% pure) were mixed with purified full-length RsiV (21 μM) and full-length SipS (3.2 μM). The RsiV-SipS-lysozyme (45 μl) mixture was incubated at 37°C for 10 min. Samples from each reaction were analyzed by SDS-PAGE; the gels were stained with Coomassie brilliant blue. Gels were imaged using the Gel Doc XR+ system (Bio-Rad).
Edman degradation.For N-terminal protein sequence analysis, cleavage reactions were separated by SDS-PAGE and then transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore). Transferred proteins were visualized on the PVDF blots by Coomassie brilliant blue staining, and the desired bands were excised for sequence analysis by Edman degradation (Iowa State University Protein Facility).
In vivo RsiV degradation assay.Overnight cultures were subcultured 1:100 in LB plus 0.01 mM IPTG and grown to an OD600 of 0.8 to 1 at 30°C. Cells were split into 1-ml samples and placed at 25°C or 48°C for 30 min. Cells were pelleted by centrifugation and the supernatant was removed. Pellets were resuspended in 100 μl of LB with or without 100 μg/ml of HEW lysozyme (Sigma; >98% pure) and incubated at 25°C or 48°C for 10 min. The reaction was stopped by addition of 100 μl of 2× Laemmli sample buffer. Cells were lysed with a Branson model 450 sonifier, and RsiV degradation was measured by Western blot analysis using anti-RsiV59-285 sera.
Western blotting.Samples were electrophoresed on a 15% SDS-polyacrylamide gel (Bio-Rad) and blotted onto nitrocellulose (GE Healthcare, Amersham). Nitrocellulose was blocked with 5% bovine serum albumin (BSA), and proteins were detected with a 1:10,000 dilution of anti-RsiV59–285 (9) and a 1:2,500 dilution of streptavidin-IRDye 680LT to detect PycA and AccB to serve as loading controls (65). Nitrocellulose was washed, then incubated with a 1:10,000 dilution of goat anti-rabbit IgG–IRDye 800CW (Li-Cor), and imaged on an Odessey CLx (Li-Cor).
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grant R01AI087834 from the National Institute of Allergy and Infectious Diseases to C.D.E. and training grant T32GM008629 to A.N.C.
We thank members of the Ellermeier lab for helpful comments.
FOOTNOTES
- Received 7 November 2017.
- Accepted 18 January 2018.
- Accepted manuscript posted online 22 January 2018.
- Address correspondence to Craig D. Ellermeier, craig-ellermeier{at}uiowa.edu.
↵* Present address: Jessica L. Hastie, FDA, Silver Spring, Maryland, USA.
Citation Castro AN, Lewerke LT, Hastie JL, Ellermeier CD. 2018. Signal peptidase is necessary and sufficient for site 1 cleavage of RsiV in Bacillus subtilis in response to lysozyme. J Bacteriol 200:e00663-17. https://doi.org/10.1128/JB.00663-17.
Supplemental material for this article may be found at https://doi.org/10.1128/JB.00663-17.
REFERENCES
- Copyright © 2018 American Society for Microbiology.
