|Home | About | Journals | Submit | Contact Us | Français|
Bacillus subtilis strain 168 produces the extremely stable and broad-spectrum lantibiotic sublancin 168. Known sublancin 168-susceptible organisms include important pathogens, such as Staphylococcus aureus. Nevertheless, since its discovery, the mode of action of sublancin 168 has remained elusive. The present studies were, therefore, aimed at the identification of cellular determinants for bacterial susceptibility toward sublancin 168. Growth inhibition and competition assays on plates and in liquid cultures revealed that sublancin 168-mediated growth inhibition of susceptible B. subtilis and S. aureus cells is affected by the NaCl concentration in the growth medium. Added NaCl did not influence the production, activity, or stability of sublancin 168 but, instead, lowered the susceptibility of sensitive cells toward this lantibiotic. Importantly, the susceptibility of B. subtilis and S. aureus cells toward sublancin 168 was shown to depend on the presence of the large mechanosensitive channel of conductance MscL. In contrast, MscL was not involved in susceptibility toward the bacteriocin nisin or Pep5. Taken together, our unprecedented results demonstrate that MscL is a critical and specific determinant in bacterial sublancin 168 susceptibility that may serve either as a direct target for this lantibiotic or as a gate of entry to the cytoplasm.
Lantibiotics are small posttranslationally modified peptides with antimicrobial activity that are produced by gram-positive bacteria (9, 27, 36). The Bacillus subtilis strain 168 is known to produce an extremely stable lantibiotic named sublancin 168. This lantibiotic exhibits bactericidal activity against other gram-positive bacteria, including important pathogens such as Bacillus cereus, Streptococcus pyogenes, and Staphylococcus aureus (30, 40). Sublancin 168 is encoded by the sunA gene, which is located within the SPβ prophage region of the B. subtilis 168 chromosome (16, 25, 30). Newly synthesized sublancin 168 is exported from the cytoplasm by the ABC transporter SunT, the gene for which is located immediately downstream of sunA (12, 38). SunT also contains a proteolytic domain (27) and is, therefore, thought to be required both for sublancin 168 export and for concomitant removal of the leader peptide (12, 30). It is noteworthy that sublancin 168 displays the extraordinary characteristic for lantibiotics of having two disulfide bonds in addition to a β-methyllanthionine bridge (30). The thiol-disulfide oxidoreductase BdbB, which is encoded by a gene downstream of sunA and sunT, appears to be of major importance for the formation of the disulfide bonds (12, 20).
The cellular target(s) of sublancin 168 and the determinant(s) for producer immunity against this lantibiotic have remained elusive for a long time. Very recently, however, we have identified the SunI protein (also known as YolF) as the immunity protein that protects producer cells against sublancin 168 (13). SunI was found to be both required and sufficient for sublancin 168 immunity, even when produced in a heterologous sublancin-sensitive host organism, such as S. aureus. Interestingly, localization studies showed that the SunI protein is anchored to the membrane through a single N-terminal membrane-spanning domain with the bulk of the protein facing the cytoplasm. This is a topology that has not been reported before for other known bacteriocin immunity proteins (13).
The present studies were aimed at identifying bacterial determinants that confer susceptibility to sublancin 168. To date, two major mechanisms for bactericidal lantibiotic activity have been reported. Type A lantibiotics, such as nisin (22, 39), epidermin (32), and Pep5 (28), usually act by forming pores in the cytoplasmic membrane of sensitive target organisms in processes that may involve specific molecules, such as the cell wall precursor lipid II (6, 43). In contrast, type B lantibiotics, such as cinnamycin (14) and mersacidin (10), inhibit particular enzyme functions. On the basis of its leader peptide sequence, sublancin 168 was previously classified as a type A lantibiotic (30). Nevertheless, sublancin 168 does not display the usual flexible, elongated, and amphipathic molecular shape that is so characteristic of other type A lantibiotics (29), suggesting that sublancin 168 might have a specific mode of action. Consistent with this idea, our present results show that the sublancin 168 susceptibility of B. subtilis and S. aureus is determined by the presence of large-conductance mechanosensitive channels, which is an unprecedented finding.
The bacterial strains and plasmids used in this study are listed in Table Table1.1. The standard LB medium consisted of 1% tryptone, 0.5% yeast extract, and 1% NaCl, pH 7.4. Where appropriate, the NaCl content of the LB medium was adjusted to concentrations ranging from 0 to 5%. S. aureus strains were grown in brain heart infusion broth (BHI) or tryptone soy broth. Where necessary, media were supplemented with antibiotics at the concentrations indicated: ampicillin, 100 μg/ml (Escherichia coli); kanamycin, 20 μg/ml (E. coli, B. subtilis, and S. aureus); chloramphenicol, 5 μg/ml (E. coli and B. subtilis); tetracycline, 10 μg/ml (E. coli and B. subtilis); erythromycin, 100 μg/ml (E. coli), 2 μg/ml (B. subtilis), or 5 μg/ml (S. aureus); spectinomycin, 100 μg/ml (B. subtilis). To visualize α-amylase activity (specified by the amyE gene), LB plates were supplemented with 1% starch. To visualize β-galactosidase activity, plates contained 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal) at a final concentration of 80 μg/ml.
Procedures for DNA amplification, restriction, ligation, and transformation of E. coli DH5α were carried out according to standard laboratory procedures (37). Chromosomal DNA of B. subtilis was isolated according to the method of Bron and Venema (7). B. subtilis was transformed as described by Kunst and Rapoport (24). S. aureus was transformed by electroporation as described by Kreiswirth et al. (21). All primers used for PCR are listed in Table Table2.2. PCR products were purified using the High Pure PCR purification kit (Roche Applied Science).
A B. subtilis 168 strain lacking the mscL gene (originally named ywpC) was obtained from the B. subtilis Functional Analysis Program. This ΔmscL strain was constructed in the laboratory of G. Rapoport according to a protocol described in detail on the Micado website (http://genome.jouy.inra.fr/cgi-bin/micado/index.cgi). Briefly, a PCR fragment containing the 500-bp region upstream of the mscL gene was fused to a fragment containing the 500-bp region downstream of the mscL gene and cloned into a pMUTIN4mcs vector using the BamHI and HindIII sites. This plasmid was subsequently used to transform B. subtilis 168, thereby replacing the mscL gene with the pMUTIN vector in a double-crossover recombination event, which yielded the ΔmscL strain. The ΔSPβ ΔmscL strain was constructed by transformation of the ΔSPβ strain with genomic DNA of the ΔmscL strain and selection for erythromycin-resistant transformants. Correct deletion of mscL from the genomes of the ΔmscL and ΔSPβ ΔmscL strains was verified by PCR.
Mutants of S. aureus were constructed using the chromosomal integration-excision approach described by Arnaud et al. (1). Primers for the downstream (mscL-F1/mscL-R1) and upstream (mscL-F2/mscL-R2) regions of mscL (Table (Table2)2) were designed to obtain PCR products of 524 and 522 bp, respectively, including a 24-bp linker. These PCR products were linked in 10 PCR cycles. The resulting 1,025-bp product was directly ligated into the TOPO vector (Invitrogen) according to the manufacturer's protocol. This construct was digested with BamHI, and the 1,030-bp product was ligated into the chromosomal integration-excision vector pMAD, resulting in pMAD-mscL. S. aureus RN4220 cells were transformed with the pMAD-mscL plasmid by electroporation and grown on tryptic soy agar (TSA) plates containing erythromycin and X-Gal for 48 h at 30°C. To obtain cells with a chromosomally integrated copy of pMAD-mscL, blue colonies were used to inoculate overnight cultures in BHI medium. Next, 10 ml BHI was inoculated with 100 μl overnight culture, grown for 1 h at 30°C, and then transferred to 42°C for 6 h. To select cells with a chromosomally integrated copy of pMAD-mscL, dilutions (1,000×) of the culture were plated on TSA plates with erythromycin and X-Gal and incubated for 48 h at 42°C. To subsequently obtain cells that had excised pMAD-mscL from the chromosome, blue colonies with integrated pMAD-mscL were used to inoculate overnight cultures in BHI medium at 42°C. Next, 10 ml BHI was inoculated with 10 μl of the overnight culture and growth was continued for 6 h at 30°C. Dilutions (1,000×) of the cultures were plated on TSA plates with X-Gal and incubated at 42°C for 48 h. White colonies were tested for erythromycin sensitivity and checked for the presence or absence of mscL by colony PCR. The correct deletion of mscL was confirmed by PCR on isolated genomic DNA using the bacterial genomic DNA isolation kit (Sigma).
A sublancin 168-induced B. subtilis growth inhibition assay was performed on plates essentially as described by Dorenbos et al. (12). Briefly, indicator strains and strains to be tested for sublancin 168 production were grown overnight in LB broth containing the appropriate antibiotic(s). Overnight cultures of the indicator strains were then diluted 100-fold in LB, and 100-μl aliquots of the diluted cultures were plated on LB agar. After the plates were dried, 2-μl aliquots of undiluted overnight cultures of strains to be tested for sublancin 168 production were spotted. Alternatively, aliquots of nisin, Pep5, or concentrated spent medium with sublancin 168 were spotted. The plates were then incubated overnight at 37°C, and growth inhibition of the indicator strain was analyzed on the next day. Nisin was obtained from Sigma, and Pep5 was kindly provided by Hans-Georg Sahl.
The concentration of sublancin 168 in spent medium of B. subtilis 168 cells was increased by lyophilization. For this purpose, B. subtilis 168 or the negative-control ΔsunA strain was grown in 25 ml of LB broth containing 0% NaCl. At an optical density at 600 nm (OD600) of 2.5, cells were separated from the growth medium by centrifugation and medium fractions were frozen in liquid nitrogen. The frozen growth media were lyophilized for 4 days under vacuum using a freeze-dryer. After this period, the lyophilized medium was resuspended in 500 μl demineralized water and filtered (0.22 μm) before use. Two microliters of the concentrate was spotted on agar plates.
B. subtilis 168 and S. aureus strains were grown separately overnight in LB medium. In the morning, cultures were diluted to an OD600 of 0.05 in fresh LB medium. Next, specific strains to be tested together were mixed in a 1:1 ratio, resulting in cocultures consisting of 50% sublancin-producing cells (B. subtilis 168 Cm) and 50% nonproducing cells (either the B. subtilis ΔSPβ Tc strain, the B. subtilis ΔSPβ ΔmscL strain, S. aureus RN4220 Em, or S. aureus RN4220 ΔmscL). Upon mixing, growth was continued for 8 or 9 hours. Samples for plating were taken at hourly intervals during growth. The samples thus obtained were diluted 104- or 106-fold and plated on LB agar containing specific antibiotics that permit growth of only one of the two cocultured strains. After overnight incubation at 37°C, resistant colonies were counted, and numbers of CFU per ml of culture of each strain at the time of sampling were calculated. In the case of the S. aureus RN4220 ΔmscL strain, which does not contain an antibiotic resistance marker, no antibiotics were present in the plates used to determine the CFU numbers of this strain. Specifically, the CFU number of the ΔmscL strain was calculated by subtraction of the CFU number of the cocultivated antibiotic-resistant strain from the total CFU number of the two cocultivated strains. The presence of B. subtilis and/or S. aureus in samples used for plating was detected by light microscopy.
The promoter regions of mscL and sunA were amplified by PCR from genomic DNA prepared from B. subtilis 168 using the primers described in Table Table2.2. The resulting PCR fragments were prepared for ligation-independent cloning using T4 DNA polymerase (Novagen) and dTTP (2). The BaSysBioII cloning vector was digested with SmaI, purified from an agarose gel, and prepared for ligation-independent cloning using T4 DNA polymerase (Novagen) and dATP. One microliter of the treated vector and 3 μl of the treated insert were mixed and left to anneal at room temperature before transformation of E. coli. The resulting constructs PmscL-GFP (green fluorescent protein) and PsunA-GFP were used to transform B. subtilis 168, and transformants were selected on LB plates containing spectinomycin. Strains with mscL-GFP or sunA-GFP fusions were precultured in LB containing 1% or 5% NaCl and diluted 1:100 in the same medium 3 hours before the start of the growth experiments. Next, the cells were diluted in the respective medium to an OD600 of 0.01. Growth was continued in triplicate wells of a 96-well black optical-bottom microtiter plate (Nunc) that was placed in a Biotek Synergy 2 plate reader (37°C, variable shaking). The OD600 and fluorescence (excitation, 485/20; emission, 528/20) of the strains were measured for 14 h. Fluorescence measurements were processed essentially as described by Ronen et al. (35). Before the experiment was started, the OD977 and OD900 of the wells were measured to allow light path correction to 1 cm. For each of the recorded fluorescence data points, the blank of neat LB (average of three wells) was removed and each point was corrected for a light path of 1 cm using the following equation: [0.18/(OD977 − OD900)]. The average of the three samples was then calculated. The fluorescence data were further processed by calculating the average of the three samples and subtracting the background fluorescence of the parental strain 168 (without GFP) for each data point. The promoter activity was calculated using the following equation: [GFP(t) − GFP(t − 1)]/OD(t). We acknowledge the collaborative effort with the teams of Stéphane Aymerich, Kevin Devine, Vincent Fromion, and Tony Wilkinson in standardizing the fluorescence measurements. Each experiment was repeated at least three times.
To study the activity of sublancin 168 under different growth conditions, we made use of a previously developed sublancin growth inhibition plate assay. The essence of this assay, in which strains potentially producing sublancin 168 were spotted onto a lawn of sensitive or immune indicator cells, is demonstrated in Fig. Fig.1A.1A. This figure shows that the ΔSPβ strain, which lacks all genes of the SPβ prophage (including those for sublancin production and immunity), is not able to grow in the vicinity of the sublancin 168-producing parental strain 168. This growth inhibition is strictly dependent on the presence of an intact copy of the sunA gene for sublancin 168 since no zone of growth inhibition is formed around ΔsunA spotted cells on a plated ΔSPβ cell layer (Fig. (Fig.1A).1A). As previously demonstrated (13), the sublancin 168 susceptibility of the ΔSPβ cells is solely due to the absence of the sunI (yolF) immunity gene, since full sublancin 168 immunity can be restored by ectopic expression of sunI (Fig. (Fig.1A).1A). To screen for possible sublancin 168 activity determinants, we deployed this assay to monitor the effects of growth medium composition on the activity of sublancin 168. It was thus noticed that the sublancin 168 activity was dependent on the type of growth medium used in the plate assays (data not shown). Upon inspection of the composition of the tested media, it was found that, in particular, the NaCl contents seemed to differ. To investigate whether the NaCl concentration might influence the outcome of our sublancin growth inhibition assay, we performed a series of sublancin 168 activity assays in which the LB broth and agar media used for growth of the 168 and ΔSPβ cells contained increasing concentrations of NaCl, ranging from 0 to 5%. The results of these experiments showed that the size of the growth inhibition zone of the ΔSPβ cells is inversely correlated with the NaCl content of the growth medium (Fig. (Fig.1B).1B). Next, we investigated whether this effect of NaCl was related to the plate assay or whether this also occurred in liquid media. For this purpose, we performed coculturing and competition experiments in liquid medium by inoculation of the sublancin 168-producing B. subtilis strain 168 amyE::pX (chloramphenicol resistant) in growth medium in a 1:1 ratio with the nonproducing B. subtilis strain ΔSPβ amyE::pXTC (tetracycline resistant). This was done both in LB medium containing the standard concentration of 1% NaCl and in LB medium containing 5% NaCl. The results of cocultivation and subsequent transfer of samples to plates containing either chloramphenicol or tetracycline showed that the ΔSPβ strain was able to survive for only a few hours in the presence of the sublancin 168-producing strain when these strains were grown in standard LB medium (Fig. (Fig.2A).2A). In contrast, when the cocultures were grown in LB containing 5% NaCl, the ΔSPβ strain was not inhibited by the presence of B. subtilis 168 (Fig. (Fig.2B).2B). As expected, the deleterious effect of the 168 strain on the ΔSPβ strain was not observed when the sunA gene was deleted from the 168 strain (Fig. (Fig.2C).2C). These findings were fully consistent with those obtained by the sublancin 168 growth inhibition assays on LB agar (Fig. (Fig.1B).1B). Taken together, the results show that the NaCl concentration in the growth medium influences the outcome of sublancin 168 growth inhibition assays. This suggests that either the production of sublancin 168, the activity of sublancin 168, or the susceptibility of the target cells is influenced by the concentration of NaCl in the growth medium.
To explore the nature of the effect of NaCl, as observed in the sublancin 168 activity assays, we first investigated a possible influence of NaCl on the production of sublancin 168. As an indication for this, we measured the activity of the sunA promoter in the B. subtilis 168 strain grown in LB medium with either 1% or 5% NaCl, using a transcriptional fusion between the sunA promoter and the GFP gene (PsunA-GFP). The results of this analysis show that the sunA promoter displayed similar average promoter activities in the two media, although the 168 cells grew somewhat slower in LB medium with 5% NaCl than in LB medium with 1% NaCl (Fig. (Fig.3A).3A). This suggests that the different outcomes in the sublancin 168 growth inhibition assays in the presence of 1% or 5% NaCl are not due to differences in the levels of sublancin 168 production by B. subtilis 168. Interestingly, expression of sunA is rather suppressed until late exponential phase, which seems to coincide with the onset of growth inhibition as observed in Fig. Fig.2A2A.
As an alternative, we investigated the possible direct effects of NaCl on the stability or activity of sublancin 168. For this purpose, we concentrated sublancin 168 by lyophilization of spent medium (without NaCl) from the sublancin producer B. subtilis 168. By doing so, we increased the sublancin 168 concentration 50-fold. Figure Figure4A4A shows that a spotted concentrate of this spent medium inhibits the growth of plated ΔSPβ cells, whereas a control concentrate from the spent medium of a ΔsunA strain does not inhibit the growth of plated ΔSPβ cells. Next, we used the concentrate from the 168 spent medium to establish whether NaCl might directly affect the stability or activity of sublancin 168. For this purpose, the concentrated spent media were incubated overnight with or without 5% NaCl, prior to spotting. The very similar sizes of the resulting growth inhibition zones of ΔSPβ cells on plates without NaCl (Fig. (Fig.4B)4B) revealed that the activity of the concentrated sublancin 168 was not affected by overnight incubation with 5% NaCl. This indicated that the absence of sublancin 168-directed growth inhibition during growth on LB with 5% NaCl is not due to an irreversible inactivating effect of NaCl on sublancin 168.
To investigate the possible effect of NaCl on the ΔSPβ indicator cells, we also spotted the concentrated 168 medium on ΔSPβ cells that were plated on LB agar plates containing 5% NaCl. The results show that ΔSPβ cells plated on LB agar containing 5% NaCl were no longer sensitive to the spotted sublancin 168, in contrast to ΔSPβ cells plated on LB agar containing 0% NaCl (Fig. (Fig.4B).4B). Taken together, these findings indicate that NaCl does not influence the production, activity, or stability of sublancin 168. Instead, NaCl seems to influence the sublancin 168 susceptibility of B. subtilis.
In a search for possible cellular mechanisms that could be involved in the NaCl-dependent susceptibility of target cells toward sublancin 168, we explored the available literature for known effects of NaCl on bacterial growth and survival. This drew our attention to a possible role of the mechanosensitive channels of ion conductance. These channels are located in the cytoplasmic membrane and catalyze the efflux of ions and osmolytes from the cytoplasm when cells encounter a downshift in the osmolarity of their environment (5, 26, 34, 41). The opening or closing of these channels is dependent on the osmolarity (salt content) of the medium in which cells are growing. At high osmolarity of the medium, the channels will mostly be closed, while at low osmolarity they may be open more frequently due to a constantly increasing turgor pressure in the cell. We therefore investigated a possible involvement of these channels in the observed NaCl-dependent sublancin 168 susceptibility of B. subtilis. For this purpose, the gene encoding the largest mechanosensitive channel, named MscL, was deleted from the genome of the sublancin 168-sensitive ΔSPβ strain. Then, sublancin 168 growth inhibition assays were performed with this ΔSPβ ΔmscL strain to assess its susceptibility to sublancin 168 in the presence of different concentrations of NaCl. The results, shown in Fig. Fig.1C,1C, were striking, since the growth of ΔSPβ ΔmscL B. subtilis was not at all inhibited by the sublancin 168 producer strain, irrespective of the absence or presence of NaCl (Fig. (Fig.1C).1C). These results were confirmed by coculturing experiments with the 168 and ΔSPβ ΔmscL strains in liquid media, which showed that growth of the ΔSPβ ΔmscL strain was not inhibited by the parental strain producing sublancin 168, irrespective of the NaCl concentration (Fig. 2D and E). This shows that the susceptibility of the ΔSPβ cells to sublancin 168 depends on the presence of the mscL gene. To investigate a possible effect of NaCl on the production level of MscL, we monitored the mscL promoter activity in B. subtilis cells during growth in LB medium containing 1% or 5% NaCl. This was done with a transcriptional PmscL-GFP fusion construct. The results of this analysis show that the mscL promoter activity profile of cells grown in LB medium with 5% NaCl is comparable to that of cells grown in LB with 1% NaCl (Fig. (Fig.3B).3B). Growth in LB medium with 5% NaCl therefore does not seem to prevent MscL production, which is consistent with our previously published results on the effects of hypo-osmotic shock on MscL-proficient and -deficient cells (19). Taken together, these observations imply that the susceptibility of sensitive B. subtilis cells toward sublancin 168 is dependent on the production of MscL.
To investigate whether MscL is also involved in the susceptibility of B. subtilis to lantibiotics other than sublancin 168, we tested the sensitivity of ΔSPβ ΔmscL cells for the type A lantibiotics nisin and Pep5 using a plate assay (Fig. (Fig.1D).1D). The results showed that ΔSPβ ΔmscL cells have no altered sensitivity toward 2.5 mg/ml nisin or 10 mg/ml Pep5 compared to the ΔSPβ strain. These results therefore indicate that MscL is a specific determinant for sublancin 168 susceptibility in B. subtilis.
Sublancin 168 has antimicrobial activity against a broad range of gram-positive bacteria. The observed MscL-dependent sublancin 168 susceptibility of B. subtilis prompted us to investigate whether this phenomenon is specific for B. subtilis 168 or whether MscL is also involved in the sublancin 168 susceptibility of other bacteria. Therefore, the sublancin 168 activity against the pathogenic gram-positive bacterium S. aureus was also investigated. As a first approach, the sublancin 168-producing B. subtilis strain 168 amyE::pX (chloramphenicol resistant) was used to inoculate LB growth medium in a 1:1 ratio with the S. aureus strain RN4220 Em (erythromycin resistant). NaCl was present at concentrations of either 1% or 5%. The results of this cocultivation and subsequent transfer of samples to plates (Fig. 5A and B) were comparable to those obtained with the B. subtilis ΔSPβ strain (Fig. 2A and B). When grown in normal LB containing 1% NaCl, the S. aureus strain was able to survive for only a few hours in the presence of the sublancin 168-producing B. subtilis strain (Fig. (Fig.5A),5A), whereas the S. aureus strain was hardly inhibited by the sublancin 168-producing B. subtilis strain when grown in LB medium containing 5% NaCl (Fig. (Fig.5B).5B). To confirm that this inhibitory effect was indeed due to the sublancin 168 produced by the B. subtilis 168 strain, we also cocultured the ΔsunA strain with the S. aureus strain RN4220 Em. Indeed, the results show that the S. aureus strain was hardly inhibited by the ΔsunA strain, although a slight inhibitory effect of B. subtilis on the growth of S. aureus was still detectable (Fig. (Fig.5,5, compare panels A and C). Notably, when we performed sublancin 168 growth inhibition assays on agar plates during which the ΔsunA strain was spotted on a lawn of plated RN4220 Em cells, a growth inhibition zone around the ΔsunA strain was still observed (data not shown). This indicated that B. subtilis also produces other antimicrobial factors to which S. aureus is sensitive. However, sublancin 168 seems to represent the most effective antistaphylococcal activity of B. subtilis, as can be concluded from the strong inhibitory effect on the growth of S. aureus shown in Fig. Fig.5.5. Finally, we tested whether the sublancin 168 susceptibility of S. aureus was also dependent on MscL. For this purpose, we constructed an S. aureus RN4220 strain lacking the mscL gene and performed subsequent coculturing experiments of this strain together with the sublancin-producing strain B. subtilis 168. The results of this coculturing showed that growth of the RN4220 ΔmscL strain was no longer inhibited by the deleterious effects of B. subtilis 168 producing sublancin (Fig. (Fig.5D).5D). Also in this case, the S. aureus RN4220 ΔmscL strain was still slightly inhibited by the B. subtilis 168 strain, but this effect was comparable to the effect observed when the S. aureus RN4220 strain was cocultured with the B. subtilis ΔsunA strain. This clearly shows that the S. aureus RN4220 ΔmscL strain is not sensitive to sublancin 168. Taken together, these observations demonstrate that the susceptibility of S. aureus to sublancin 168 is dependent on the presence of an MscL channel and that this feature is conserved in B. subtilis and S. aureus.
In the present paper, we report on bacterial and environmental factors that determine bacterial susceptibility to the lantibiotic sublancin 168. Since its discovery in 1980, the antimicrobial mechanism of this broad-range and highly stable lantibiotic has never been so much as hinted at. We now show that the sublancin 168 susceptibility of cells lacking the immunity protein SunI is dependent on the NaCl content of the growth medium. In contrast, the production, stability, and activity of the sublancin 168 seem to remain largely unaffected by NaCl. Furthermore, we show that MscL plays a critical and unprecedented role in the mode of action of sublancin 168, since this channel is indispensable for sublancin 168 susceptibility. As MscL determines sublancin 168 susceptibility in both B. subtilis and S. aureus, it is well conceivable that the respective mechanism is conserved in many sublancin 168-susceptible bacteria.
Mechanosensitive ion channels are membrane-embedded channels that are present in all three domains of life but are especially widespread among bacteria (5, 26, 34, 41). So far, three families of mechanosensitive channels have been distinguished, named MscM (mini), MscS (small), and MscL (large). These channels are activated at different levels of applied pressure (3). The MscL channel opens at the highest applied pressure and also has the greatest pore diameter (30 Å). Mechanosensitive channels catalyze the efflux of osmolytes or osmoprotectants upon hypo-osmotic shock (4). When cells encounter a sudden downshift in osmolarity of their environment, they respond by excreting ions and osmolytes from the cytoplasm in order to maintain an appropriate turgor pressure. The cells thereby protect themselves from death by lysis due to overpressure. Possession of effective osmoregulatory protection mechanisms seems, therefore, of vital importance, in particular for soil-dwelling organisms that are frequently exposed to hypo-osmotic shocks, such as rain. In fact, it has been reported that B. subtilis cells lacking functional mechanosensitive channels, especially MscL, are severely compromised in their ability to survive a hypo-osmotic shock (17, 42). Notably, in the present studies we did not expose the cells to hypo-osmotic shock but merely grew them in LB media with various concentrations of NaCl. Furthermore, we also did not observe any difference in viability between the B. subtilis or S. aureus parental strains and their ΔmscL derivative strains under the applied conditions as long as sublancin 168 was absent. It seems therefore unlikely that sublancin 168 exerts its bactericidal activity by blocking the MscL channels of sensitive organisms. Yet, the sublancin 168 susceptibility was clearly shown to depend on the presence of MscL. Although MscL channels are vital for lowering the turgor pressure during osmotic shock, it is not unlikely that these channels can also occasionally open during a constant environmental osmolarity, especially if cells are grown in low-osmolarity media. Since an open state of the MscL channel is potentially hazardous for cells due to ion loss, and since a closed state of the MscL channel is more or less equivalent to the absence of MscL, it is very well conceivable that sublancin 168 susceptibility relates to an open state of the MscL pore. This would fit with the observed protective effect of salt against the detrimental effects of sublancin 168. In fact, this protective effect of salt was the prime reason for us to investigate a possible involvement of MscL in bacterial susceptibility to sublancin 168. The MscL channels open at a certain level of pressure on the membrane, which is usually the result of cell swelling due to a difference in the osmolarity (or salt content) between the cytoplasm and the environment. The higher the salt content of the medium, the less likely it is that mechanosensitive channels are open. Therefore, the observation that addition of salt to the medium lowers the susceptibility to sublancin 168, apparently without affecting sublancin 168 production or activity, seems to indicate that the MscL channel needs to be in an open state to allow for bactericidal activity of sublancin 168. If so, sublancin 168 might prevent the MscL pore from closing, which would result in cell death by leakage of ions from the cytoplasm. This putative mechanism would in fact be consistent with the classification of sublancin 168 as a group A lantibiotic, since lantibiotics of this class are known to create pores in the membrane. In this case, however, sublancin 168 would keep an already existing pore in an open state. Clearly, several alternative hypotheses can still be entertained. For example, MscL might function as a gate for sublancin 168 import into the cell, allowing it to perform bactericidal activity by interacting with an unidentified essential cytoplasmic target. Furthermore, MscL might be involved in an indirect process that is required to promote the bactericidal activity of sublancin 168.
Interestingly, even though MscL is conserved in all gram-positive organisms, not all of these organisms are inhibited by sublancin 168. This might be due to the existence of systems for lantibiotic producer immunity or natural lantibiotic resistance. Recently, we have identified the B. subtilis sublancin 168 producer immunity protein. This protein, SunI (YolF), was also found to give S. aureus immunity against sublancin 168 when heterologously expressed in this organism. SunI was shown to be a membrane protein with an Nout-Cin topology, the bulk of the protein facing the cytoplasm (13). This topology was unprecedented for known lantibiotic producer immunity proteins, but notably, it seems compatible with all above-mentioned mechanisms by which MscL might confer susceptibility to sublancin 168. For example, SunI might close an MscL pore that is kept in an open state by sublancin 168, it might prevent interactions between sublancin 168 and MscL (or other compounds) in the membrane, or it might block the entry of sublancin 168 into the cytoplasm. If producer immunity proteins analogous to SunI are active in other gram-positive bacteria, they may perhaps also be able to counteract the bactericidal effects of sublancin 168. A second and perhaps more likely possibility is the existence of effective resistance mechanisms. For example, it was shown that certain genes of the σW regulon confer sublancin 168 resistance on B. subtilis (8). Furthermore, an altered membrane or cell wall with an increased net positive charge might protect against the bactericidal effects of cationic lantibiotics, like sublancin 168 (31, 33). Such a resistance mechanism has been shown to exist in Staphylococcus epidermidis, which, indeed, is resistant to sublancin 168 (30).
In conclusion, the present studies have focused a general interest on mechanosensitive channels as potential determinants for bacterial susceptibility toward bacteriocins. Specifically, we have identified a critical role of the MscL channel in the susceptibility of B. subtilis and S. aureus toward the lantibiotic sublancin 168. Our findings suggest that this may be a conserved phenomenon in sublancin 168-sensitive organisms. Therefore, ongoing and future studies, including electrophysiology and interaction studies, are aimed at identifying the precise mechanisms by which (i) sublancin 168 exerts its bactericidal activity and (ii) MscL confers susceptibility to sublancin 168. Such studies will also further increase our knowledge of mechanosensitive channels in general, which is important as these channels serve important functions in all domains of life, including humans.
We thank Magda van der Kooi-Pol for technical assistance; Hans-Georg Sahl for the gift of purified Pep5; Leslie Aïchaoui-Denève, Stéphane Aymerich, Eric Botella, Kevin Devine, Mark Fogg, Vincent Fromion, Annette Hansen, Matthieu Jules, Pascal Neveu, Sjouke Piersma, Patrick Veiga, and Tony Wilkinson for their collaboration in developing tools and protocols for pBaSysBioII-based fluorescence measurements; and other colleagues from the BaSysBio program for helpful discussions.
Funding for this project was provided by the CEU projects LSHG-CT-2004-503468, LSHG-CT-2004-005257, LSHM-CT-2006-019064, and LSHG-CT-2006-037469; the transnational SysMO initiative through project BACELL SysMO; the European Science Foundation under the EUROCORES Programme EuroSCOPE; and grant 04-EScope 01-011 from the Research Council for Earth and Life Sciences of The Netherlands Organization for Scientific Research.
Published ahead of print on 8 September 2009.