The Lantibiotic Mersacidin Is an Autoinducing Peptide

Bacterial strains, plasmids, and growth conditions.Bacterial strains were maintained as glycerol cultures at −70°C and grown on blood agar or in tryptone soy broth. All strains used are listed in Table 1. Bacillus sp. strain TTEX was constructed as described previously (33) using an altered sites in vitro mutagenesis system (Serva, Heidelberg, Germany), and a stop codon was introduced in position 4 of the leader peptide employing the primer E4stopmersacidin (Table 2). Appropriate antibiotics (chloramphenicol, 20 mg/liter; and erythromycin, 25 mg/liter) were added to the media used for culturing and selecting mutants. For the detection of mersacidin production, Bacillus sp. strain HIL Y-85,54728 and its respective mutants were grown in double-strength synthetic medium (1). Antibacterial activities in the supernatant were determined by agar diffusion assays using the indicator strain Micrococcus luteus ATCC 4698. When the producer strains had been cultivated in the presence of chloramphenicol, Staphylococcus carnosus TM 300(pTV0MCS), which harbors a chloramphenicol resistance gene, was employed as the indicator strain. Staphylococcus carnosus TM 300(pTV0MCS) was incubated overnight at 30°C, Micrococcus luteus ATCC 4698 was incubated at 37°C. For determination of bacteriocin production, Mueller-Hinton agar or Columbia blood agar was used.

The plasmid pOPAR1 contains the natural EcoRI-HindIII fragment that harbors the putative operator region upstream of mrsA, the promoter, the structural gene mrsA, and the gene mrsR1 (nucleotides 4843 to 6046 of GenBank accession number AJ250862) (Fig. 1B and C). The insert was cut out from the plasmid pMER2, which is a derivative of pMER1 (1), by digestion with EcoRI and HindIII and was ligated into pCU1 (2). The insert of the plasmid pPAR1 starts at the EcoRV site upstream of mrsA and was amplified by PCR using primers containing an EcoRI site (primer “5′mrsAR1”) and the natural HindIII restriction site (primer “3′mrsAR1”) (nucleotides 4998 to 6046 in accession number AJ250862). In contrast to pOPAR1, pPAR1 does not contain the putative operator region between the EcoRI and EcoRV restriction site (Fig. 1B and C). Both plasmids were introduced into Bacillus sp. strain HIL Y-85,54728 Rec1.

Isolation of DNA.DNA was prepared using QIAGEN (Hilden, Germany) genomic tips according to the manufacturer's recommendations. General protocols were followed for cloning strategies and enzymatic DNA modifications (27). Digested DNA fragments were eluted from agarose gels employing a MinElute gel extraction kit (QIAGEN).

Construction of a SigH knockout.Inactivation of the structural gene spo0H encoding SigH was performed employing the temperature-sensitive vector pTV0MCS, which carries a chloramphenicol resistance gene (11). Bacillus strains and Staphylococcus carnosus TM 300 were transformed by protoplast transformation (9, 10). All primers employed are listed in Table 2.

For inactivation of spo0H, an internal 339-bp fragment of spo0H was amplified by PCR using the primers SigmaH5′ and SigmaH3′. The protein that is encoded by this fragment carries a deletion in the N terminus, which is responsible for binding to the −10 region of the SigH promoter, and a deletion in the C terminus that is involved in binding to the −35 region of the SigH promoter. After digestion with KpnI and XbaI, the fragment was ligated to digested pTV0MCS, which replicates only in gram-positive organisms, and the ligation assay was transformed into Staphylococcus carnosus TM 300, yielding pΔSIGH1. The recombinant plasmid was then introduced into Bacillus sp. strain TT and integrated into the chromosome by a single crossover. Colonies that had integrated the plasmid into the chromosome were selected by a temperature shift to 42°C in the presence of chloramphenicol (20 mg/liter). The integration of the plasmid was confirmed by PCR using the primer combinations SigH1/pTV0mcsIns-1 and SigH2/pTV0mcsIns-2. The SigH1/2 primers anneal on the chromosome in regions that are not covered by the cloned fragment. The pTV0MCS primers anneal on the plasmid, and the above combinations amplify 596- and 404-bp products, only if the plasmid has been integrated into the chromosome. Loss of the circular form was checked by PCR using pTV0mcsIns-1 and pTV0mcsIns-2, which yields a 496-bp product if circular plasmid is present in the cell. Four knockout strains were chosen for further characterization and were designated Bacillus sp. strain TT ΔSIGH1.1, Bacillus sp. strain TT ΔSIGH1.2, etc. (Table 1).

Determination of sporulation frequency of the knockout mutant.Cultures of the knockout mutants and the wild-type strain were grown in LB medium for at least 48 h at 37°C and 180 rpm. In order to kill vegetative cells and induce spore germination, 1 ml of the cultures was incubated at 90°C for 30 min. The concentrated cell suspension and 10-fold dilutions were plated on nutrient agar and LB (20 mg/liter chloramphenicol). The SigH-deficient strain B. subtilis 168 1S20 served as an asporogenous control (25). After 16 h of incubation, the colonies were counted. Microscopic examinations were performed after Gram staining or by phase-contrast microscopy.

Assay of the producer self-protection of the sigH knockout mutants.For determination of producer self-protection (“immunity”) to mersacidin, the wild-type producer strain Bacillus sp. strain HIL Y-85,54728 and the sigH knockout mutants were grown to an optical density at 600 nm of 0.3 to 0.5 in 10 ml of half-concentrated Mueller-Hinton broth. Mersacidin was added at a concentration of 10 mg/liter, and growth was monitored at 600 nm.

Mapping of the mrsA mRNA using 5′ rapid amplification of cDNA ends (RACE).Isolation of total RNA from Bacillus sp. strain HIL Y-85,54728 was performed with a QIAGEN RNeasy midi kit. Before harvesting, the cells were treated with RNAprotect bacteria reagent (QIAGEN, Hilden, Germany) to ensure stability of total RNA. Cultures of the wild-type strain were grown in 15 ml synthetic medium (1) and harvested at an optical density at 600 nm of 2. The cells were lysed by incubation with lysozyme (100 mg/ml) in Tris-EDTA buffer for 20 min at room temperature. The further procedure was performed as described by the manufacturer of the 5′/3′ RACE kit (Roche, Penzberg, Germany). For the 5′ RACE experiment, 0.855 μg of RNA was introduced into the cDNA synthesis using the sequence-specific primer RT-5 in a concentration of 12.5 μM. After a purification step with a HighPure PCR product purification kit (Roche, Penzberg, Germany), the cDNA served as template for a first PCR, using the primers RT-5 and oligo-dT-anchor (Roche, Penzberg, Germany). Since the amount of the resulting PCR product was too small, a second PCR was performed, using the nested primers RT-12 and the PCR anchor primer according to the instructions of the manufacturer (Roche, Penzberg, Germany) (Table 2).

DNA sequence analysis.PCR products were sequenced by Sequiserve (Vaterstetten, Germany). Processing of the nucleotide sequence data was performed by using ClustalW 1.82 at the EMBL website (http://www.ebi.ac.uk/clustalw ).

Induction of mersacidin biosynthesis and real-time PCR of mrsA transcripts.For the induction experiments, the appropriate strains were cultivated overnight in tryptone soy broth. A portion of this culture (100 μl) served as an inoculum for 10 ml synthetic medium (1), and the cultures were incubated for about 5 h until an optical density at 600 nm of 0.5 to 0.7 had been reached. Mersacidin in a concentration of 2 mg/liter or 2 ml of a sterilized 16-h supernatant was added to the exponentially growing cultures, and the first sample was withdrawn immediately afterwards. Subsequently, growth and production of antibacterial activity were monitored by taking aliquots every hour. The culture supernatants employed for induction were obtained after cultivating the wild-type producer or Bacillus sp. strain TTEX for 16 h in synthetic medium, which allows mersacidin production (1). The cells were then harvested, and the culture supernatant was sterilized by filtration. The experiments were performed three times.

To confirm the data on a transcriptional level, a real-time PCR experiment was performed by using a LightcyclerFast start DNA Master^Plus SYBR green I kit from Roche (Penzberg, Germany). The appropriate strains were cultivated in 25 ml of synthetic medium (1) until an optical density at 600 nm of 0.5 had been reached. After induction with mersacidin at a concentration of 2 mg/liter, the cultures were further incubated for 15 min and 1 h before the total RNA was extracted. The isolated RNA (3 μg) was transcribed into cDNA using the SuperScript II reverse transcriptase from Invitrogen (Karlsruhe, Germany) following the manufacturer's instructions. As a housekeeping gene, the gene encoding the 16S rRNA was amplified by PCR using the primers 5′16S rRNA lang and 3′16S rRNA lang rev. For real-time PCR, the primers 5′16S rRNA and 3′16S rRNA were used for amplification. For the target gene mrsA, the primers RT-4 and RT-5 were employed and the plasmid pPAR1 (Fig. 1B) served as a control. PCR products were analyzed on a 2% agarose gel.

Nucleotide sequence accession numbers.The nucleotide sequence of the complete biosynthetic gene cluster of mersacidin is available under GenBank accession number AJ250862. The nucleotide sequence of the gene spo0H is available under accession number NC_000964.

Inactivation of spo0H and the influence of SigH on mersacidin production and immunity.The transcription factor SigH was inactivated by insertion of the plasmid pΔSIGH1 into its structural gene spo0H by a single crossover. For further analysis, four colonies were picked and designated Bacillus sp. strain TT ΔSIGH1.1, Bacillus sp. strain TT ΔSIGH1.2, etc. In all clones, the correct insertion of pΔSIGH1 into the chromosome of Bacillus sp. strain HIL Y-85,54728 was confirmed by two PCR experiments that employ primer combinations which simultaneously anneal on the chromosome and the plasmid and yield only amplification products when the plasmid is inserted in the chromosome. These experiments yielded amplification products of the expected size. In contrast, PCR experiments that rely on circular plasmid as a template did not yield a product, indicating that the plasmid was correctly integrated into the chromosome. After integration of the plasmid, two copies of spo0H were present in the chromosome; however, both copies encoded inactive proteins. The upstream copy carries a deletion of the C terminus (bp 489 to 657), and the downstream copy encodes a protein with a deletion in the N terminus (bp 1 to 164). After storage in the cold for 4 days, the colonies showed the typical translucent morphology that had been described by Rogolsky (25) (data not shown). Microscopic examination of the mutant cultured in the presence of chloramphenicol did not show spores, and the absence of spores was also confirmed by heat inactivation of vegetative cells. In contrast to the wild type (1.69 × 10⁸ colonies/ml), the knockout mutants did not form any colonies after 30 min of incubation of the culture at 90°C, indicating that no spores had been formed during growth.

Mersacidin is primarily produced in synthetic medium. Previous experiments with the mersacidin knockout mutant Bacillus sp. strain HIL Y-85,54728 Rec1 had shown that the strain is able to excrete at least two other antibacterial substances in this medium; however, these activities were degraded quickly and after 72 h of culture, mersacidin constituted the main antibacterial activity in the supernatant (1, 11). Therefore, the mersacidin production yield of the mutants was analyzed after 72 h of culture in synthetic medium by an agar diffusion assay using Staphylococcus carnosus TM 300(pTV0MCS) as an indicator organism. In Fig. 2A, the results of the diffusion assay are summarized. The inhibition zones clearly show that SigH had no influence on final mersacidin yields; in comparison to the wild-type strain, the yield of antibacterial activity by the mutants was not diminished.

• Open in new tab
• Download powerpoint

FIG. 2.

Agar diffusion assay. (A) Mersacidin production of the wild-type producer strain Bacillus sp. strain HIL Y-85,54728 (1), Bacillus sp. strain HIL Y-85,54728 Rec1 (2), Bacillus sp. strain TT ΔSIGH1.1 (3), Bacillus sp. strain TT ΔSIGH1.2 (4), Bacillus sp. strain TT ΔSIGH1.3 (5), and Bacillus sp. strain TT ΔSIGH1.4 (6) in synthetic medium was analyzed by agar diffusion assay using Staphylococcus carnosus TM 300(pTV0MCS) as the indicator strain. (B) Production of bacteriocins other than mersacidin by Bacillus sp. strain HIL Y-85,54728 (1), Bacillus subtilis 168 1S20 (2), Bacillus sp. strain TT ΔSIGH1.1 (3), Bacillus sp. strain TT ΔSIGH1.2 (4), Bacillus sp. strain TT ΔSIGH1.3 (5), and Bacillus sp. strain TT ΔSIGH1.4 (6) in LB using Micrococcus luteus ATCC 4698 as the indicator organism.

In order to analyze a possible influence of SigH on producer self-protection, the wild-type strain and the knockout mutants were cultured in half-concentrated Mueller-Hinton broth. After an optical density of 0.5 had been reached, mersacidin at a concentration of 10 mg/liter was added. After a lag phase of about 3 h, all strains resumed growth. The SigH knockout mutants were not impaired in growth, and no significant differences in susceptibility between the mutants and the wild type could be observed (data not shown).

As mentioned above, the wild-type producer strain is able to synthesize at least two other substances with antibiotic activity in addition to mersacidin. Antibacterial activity is also produced in complex media, where transcription of mrsA is inhibited (A. Hoffmann, unpublished data). To assess production of these other antibacterial substances by the knockout mutants, the strains were grown overnight in a complex medium (Luria-Bertani broth). Examination of the culture supernatants showed that production of these substances was impaired in the mutant strains. Similar results were obtained with a SigH mutant of B. subtilis 168 that was included in the test (Fig. 2B). In conclusion, the experiments indicated that the alternative sigma factor H had no influence on production of mersacidin and immunity against mersacidin under the chosen conditions, but it did affect the production of other antibacterial substances by the producer strain.

Characterization of the mrsA promoter and the influence of a putative operator region on mrsA transcription.In order to map the mrsA mRNA, a 5′ RACE experiment was performed. The 5′ end of the mrsA mRNA is indicated in Fig. 1C. A possible Pribnow box (−10 region) of a SigA-dependent promoter with four matches to the consensus sequence (TATAAT) is located 31 bp upstream of the start codon ATG and 5 bp upstream of the +1 position of the mrsA mRNA. A tentative −35 region with three matches to the consensus −35 sequence (TTGACA) is located 53 bp upstream of the ATG codon (Fig. 1C). In between there is a characteristic spacer of 17 bp. The consensus sequences indicate that mrsA is most probably transcribed by the housekeeping SigA-dependent RNA polymerase. The upstream region of the promoter contains an AT-rich sequence between positions −36 and −70 that is typical for the upstream regions of Bacillus subtilis SigA promoters and may contact the α-unit of the RNA polymerase (14).

In order to analyze whether an operator region is localized upstream of the mrsA promoter, Bacillus sp. strain HIL Y-85,54728 Rec1 (1), which harbors an erythromycin resistance cassette instead of the structural gene mrsA, was transformed with the plasmids pPAR1 (carrying mrsA, mrsR1, and the promoter of mrsA) and pOPAR1 (harboring an insert that starts 235 bp upstream of the transcriptional start site and includes the promoter and the putative operator sequence in addition to mrsA and mrsR1) (Fig. 1B). The transcription of mrsA was analyzed in the clones and the wild-type strain in real-time PCR experiments. The 16S rRNA served as a standard housekeeping gene. The transcription rates of the wild-type producer, Bacillus sp. strain HIL Y-85,54728 (5.1 × 10⁵ ± 3.1 × 10⁵ copies/10⁶ copies 16S rRNA), and Bacillus sp. strain HIL Y-85,54728 Rec1(pOPAR1) (4.33 × 10⁵ ± 2.9 × 10⁵ copies/10⁶ copies of 16S rRNA) were in a similar range, whereas the transcription of mrsA in Bacillus sp. strain HIL Y-85,54728 Rec1(pPAR1) (5.16 × 10³ ± 1.6 × 10³ copies/10⁶ copies of 16S rRNA) was 100-fold decreased. These data show that this region has a positive influence on transcription of mrsA and indicate that a putative operator region, which may serve as a binding site of at least one regulatory protein, or a second promoter exists upstream of the SigA-dependent promoter.

Mersacidin biosynthesis can be induced.To analyze whether mersacidin biosynthesis can be induced, induction experiments employing pure mersacidin as well as 16-h culture supernatant were performed with Bacillus sp. strain HIL Y-85,54728 as described in Materials and Methods. The 16-h supernatant of the producer strain used in these experiments contained 10.6 ± 1.9 mg/ml of mersacidin as well as any other molecule that might be involved in induction of mersacidin biosynthesis, since after 16 h of culture, mersacidin production is well under way (1). After the addition of 2 mg/ml mersacidin, as well as after the addition of 2 ml of sterilized supernatant to 10 ml of culture (introducing about 1.8 mg/ml mersacidin), a production of antibacterial activity that was earlier than that of the control culture could be detected (Fig. 3A).

• Open in new tab
• Download powerpoint

FIG. 3.

Induction of the production of antibacterial activity by mersacidin. Growth and production of antibacterial activity by Bacillus sp. strain HIL Y-85,54728 (A, C) or Bacillus sp. strain TT ΔSIGH1.1 (B). In panels A and B, the antibacterial activity produced by the cells after the addition of 2 ml of sterilized 16-h supernatant (▪), or pure mersacidin at a concentration of 2 mg/liter (▴), and the control (no addition) (⧫) are plotted against the growth of the cultures. In panel C, production of antibacterial activity by Bacillus sp. strain HIL Y-85,54728 after addition of vancomycin (0.5× MIC) (□), mersacidin (0.25 mg/liter) (×), mersacidin (0.5 mg/liter) (⋄), 16-h supernatant of Bacillus sp. strain TTEX (Δ), and the control (no addition) (⧫) are plotted against the optical density (OD₆₀₀) of the cultures. Micrococcus luteus ATCC 4698 was used as the indicator strain in an agar diffusion assay. The addition of mersacidin/vancomycin or supernatant is marked by an arrow.

However, the producer strain is able to excrete several substances with antibacterial action. To show that the antibacterial activity in the induced culture supernatants is elevated because of the production of mersacidin, the same experiment was performed with one of the SigH mutants, which—as shown above—is impaired in production of antibacterial substances other than mersacidin. After the induction with mersacidin and 16-h supernatant of Bacillus sp. strain HIL Y-85,54728, an earlier production of antibacterial activity could also be observed with the knockout mutant (Fig. 3 B). Since the effect was mediated not only by culture supernatant but also by pure mersacidin in a comparable concentration, the result indicated that mersacidin may be the active component in the 16-h supernatant and may be recognized by the cell.

The next question was whether this mechanism is based on specific recognition of the mersacidin molecule itself, i.e., possibly mediated by the specific two-component regulatory system MrsR2/MrsK2, which is encoded in the mersacidin biosynthetic gene cluster, or more unspecifically due to the action of mersacidin on the cell, i.e., inhibition of cell wall biosynthesis by complexing lipid II (4). Therefore, another inhibitor of cell wall biosynthesis, the glycopeptide antibiotic vancomycin, was added in a subinhibitory concentration (0.5× MIC) to an exponentially growing culture. The mode of action of vancomycin is very similar to that of mersacidin: both compounds complex lipid II, thereby inhibiting transglycosylation. After the addition of vancomycin, no earlier onset of production could be detected in comparison to that of the control (Fig. 3C), indicating that inhibition of cell wall biosynthesis by complexing lipid II is not sufficient to induce mersacidin biosynthesis.

To make sure that it was not a metabolic by-product or other antibacterial substance excreted by the producer strain that was involved in induction, 16-h supernatant of Bacillus sp. strain TTEX, a producer strain which carries a point mutation in the mersacidin structural gene introducing a stop codon after the first three amino acids of the leader peptide, was prepared and tested for induction. The TTEX supernatant was inactive in the experiment, which further underlines that the induction is directly dependent on mersacidin (Fig. 3C).

In order to define the minimum concentration of mersacidin needed for induction, lower mersacidin concentrations were tested. However, an induction of mersacidin production after the addition of 0.25 mg/liter and 0.5 mg/liter of mersacidin was not detectable (Fig. 3C). Experiments using 1 mg/ml sometimes showed a slight effect (data not shown).

To confirm the results observed above on the transcriptional level, quantitative real-time PCR experiments were performed. Figure 4 shows the transcription rates after the induction of exponentially growing cultures with mersacidin in a concentration of 2 mg/liter. After addition of mersacidin, the wild-type strain indeed showed an increased expression of the structural gene at 1 h postinduction. This result strongly supports the previous observations and suggests that an autoinducing mechanism is involved in mersacidin biosynthesis. Knockout mutants in the regulatory proteins of the mersacidin gene cluster were also included in this test. In Bacillus sp. strain TT ΔmrsK2/R2, the two-component regulatory system, MrsR2 and MrsK2, has been inactivated by insertion of an erythromycin cassette (11). In contrast to the wild-type producer, this strain did not increase transcription of the mersacidin mRNA, indicating that the two-component regulatory system is involved in the recognition of the signal and regulation of transcription of mrsA. In Bacillus sp. strain TT ΔmrsR1, the single regulatory protein MrsR1 has been inactivated (11). This strain showed a very slightly increased expression of mrsA after induction (Fig. 4).

• Open in new tab
• Download powerpoint

FIG. 4.

Influence of an autoinducing mechanism in mersacidin biosynthesis. Percentage of copies of mrsA per 10⁶ copies of 16S rRNA after addition of mersacidin (2 mg/liter) to the wild-type producer (Bacillus sp. strain HIL Y-85,54728) and knockout mutants (Bacillus sp. strain TT ΔmrsR2/K2 and Bacillus sp. strain TT ΔmrsR1). The strains were grown to an optical density at 600 nm of 0.5, mersacidin was added, and aliquots were withdrawn after 15 min (white bars) and after 60 min (black bars) for real-time PCR. Controls show the transcript levels of cultures grown in the absence of mersacidin. The transcript levels of the controls (no addition) measured 15 min after addition of mersacidin to the test cultures were set to 100% transcription.