INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Streptococcus salivarius is a primary colonizer of neonatal oral mucosal surfaces and a predominant component of the human adult oral microbiota and is not associated with disease in healthy individuals (39). Streptococcus pyogenes, on the other hand, persists in the pharynx in a carrier state in approximately 10% of the population, is a common cause of pharyngeal infections, especially in school-aged children, and is usually present in high numbers only during acute infection (1). All S. pyogenes strains tested have been found to be susceptible to growth inhibition by salivaricin A (SalA), a lantibiotic peptide produced by S. salivarius (34, 37), and it has been suggested that growth of S. pyogenes in vivo may be modulated by indigenous SalA-producing S. salivarius.

Lantibiotics are antimicrobial peptides that are produced by, and are active against, closely related gram-positive organisms. These peptides are ribosomally synthesized and then undergo posttranslational modifications, including amino acid dehydration (38) and thioether bridge formation (20). Lantibiotics form two families; type A lantibiotics are linear, and type B lantibiotics have more globular conformations. The peptides are synthesized as prepropeptides consisting of an inhibitor propeptide and a leader region, the features of which have been utilized to group the type A lantibiotics into three subclasses (36). In the members of one of these subclasses, subclass AII, the amino acid (aa) residues Gly-Gly, Gly-Ser, or Gly-Ala immediately precede the site of leader cleavage. These Gly-Gly type cleavage sites are more commonly found in nonmodified bacteriocins or pheromones, such as the ComC peptide responsible for competence induction in Streptococcus pneumoniae (15). The streptococcal pheromones are quorum-sensing molecules, analogous to the N-acyl homoserine lactones produced by gram-negative bacteria (35), that regulate microbial community responses.

SalA (22 aa residues) is a subclass AII lantibiotic produced by S. salivarius 20P3 via processing of a 48-aa prepropeptide encoded by the salA gene (34). By utilizing salA as a DNA hybridization probe it was shown that all SalA peptide-producing strains of S. salivarius contained salA sequences and that, somewhat surprisingly, 63 of 65 S. pyogenes strains of different M types contained a salA gene homolog designated salA1 (37). We hypothesized that S. pyogenes failed to produce SalA1 inhibitor and was sensitive to inhibition by SalA from S. salivarius because the genetic locus for lantibiotic production and immunity was incomplete or transcriptionally inactive. In the present study, the structure of the sal genetic locus in S. salivarius 20P3 was determined and the genes necessary for regulation of lantibiotic production were identified. In S. pyogenes SF370 (M type 1), the corresponding sal genetic locus had a deletion spanning three genes encoding peptide modification and export proteins, and the salA1 gene was transcriptionally inactive, thus accounting for the lack of SalA1 production. However, the salA gene was actively transcribed in S. pyogenes T11 (M type 4), which contained a complete sal genetic locus, and active SalA1 inhibitor peptide was produced. To the best of our knowledge, this is the first report of an autoregulatory lantibiotic produced by streptococci, and the evidence presented here suggests that SalA-like peptides form a family of signaling factors recognized by different species of streptococci.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and media. S. salivarius 20P3 (wild type, SalA⁺) (34), S. pyogenes SF340 (M type 1, SalA) (41), and S. pyogenes T11 (M type 4, SalA⁺ clinical isolate) were maintained on Columbia agar plates (Life Technologies Ltd., Paisley, United Kingdom) supplemented with human blood (5%, vol/vol) and CaCO₃ (0.1%, wt/vol) in a 5% CO₂-air atmosphere at 37°C. Streptococcal strains were routinely cultured in Todd-Hewitt broth (Difco Laboratories, Detroit, Mich.) or in M17 medium (Difco) supplemented with CaCO₃ (10 mM) and sucrose (10 mM) (M17CS medium) for maximal SalA peptide production. Escherichia coli DH5 (9) and the highly electrocompetent derivative strain DH10B (14) were maintained on Luria-Bertani (LB) agar (Difco) or were grown in LB broth with aeration at 37°C. The following antibiotic concentrations were used (where appropriate): ampicillin, 100 µg/ml (E. coli); and spectinomycin, 100 µg/ml (E. coli), 60 µg/ml (S. pyogenes), or 600 µg/ml (S. salivarius).

PCR and sequence analysis. Primers for PCR amplification of salA and the regions downstream of salA in S. salivarius 20P3 were designed on the basis of the corresponding sal locus sequence in S. pyogenes SF370 (University of Oklahoma Advanced Center for Genome Technology; http://www.genome.ou.edu/strep.html). The major primers utilized (and their corresponding target sites within the S. salivarius 20P3 sal locus [10,610 nucleotides; GenBank accession number AY005472]) were SalAF (positions 604 to 629; 5'GATATTTTGAACAATGCTATCGAAGA), DsintF (positions 856 to 876; 5'CAACATCAGTTTTACGAATAC), ORFIR (positions 2340 to 2320; 5'GTGAACTTCAATCTTTCATCG), SalYF (positions 6665 to 6684; 5'GGTCAAGCTCTAGTTATCGC), SalYR (positions 8413 to 8393; 5'GATTATGGATGTCACTAACCG), and SalRterm (positions 10610 to 10590; 5'TCAGAAATCCATAAAATACCC). PCRs were carried out with Taq polymerase (Roche Diagnostics Ltd., Lewes, England) for 30 cycles consisting of denaturation at 95°C for 30 s, annealing at between 55 and 64°C (as appropriate) for 30 s, and extension at 72°C for 1 min per kb of DNA to be amplified; this was followed by a final elongation step at 72°C for 5 min. PCR products were ligated into pGEM-T (Promega, Madison, Wis.) and sequenced with a Perkin-Elmer ABI 377A sequencer. Primary sequence data were collated with SeqEd sequencer software, and sequence alignment, translation, and general analyses were performed by using DNAMAN (Lynnon BioSoft, Vaudreuil, Canada). Deduced amino acid sequences for open reading frames were compared with sequences in protein sequence databases by using the BLAST facilities on the National Center for Biotechnology Information server (www.ncbi.nlm.nih.gov) and the University of Oklahoma server.

DNA manipulation and transformation. Streptococcal genomic DNA was isolated as described by Upton et al. (43). Briefly, confluent growth from one-half of an agar plate culture was collected in TE buffer (10 mM Tris-HCl [pH 8.0] containing 10 mM EDTA), the cells were harvested by centrifugation (8,000 × g for 10 min) and suspended in 0.3 ml of TE buffer, and 0.3 ml of a lysis mixture (50 mM Tris-HCl [pH 8.0] containing 10 mM EDTA and 2% [vol/vol] Triton X-100) was added. Lysis was achieved by adding 30 U of mutanolysin (Sigma) and then 0.3 mg of pronase (Sigma) and incubating the preparation at 37°C for 2 h. The suspension was extracted once with 0.6 ml of phenol and once with 0.6 ml of phenol-chloroform (0.3 ml of phenol plus 0.3 ml of chloroform-isoamyl alcohol [24:1]). Nucleic acids were precipitated from the aqueous phase with 2 volumes of 100% ethanol at 70°C and collected by centrifugation (12,000 × g for 20 min), and the pellet was washed with 70% (vol/vol) ethanol, air dried, and dissolved in TE buffer. In a number of experiments the solutions were then incubated with 50 µg of RNase A (Sigma)/ml at 37°C for 30 min and extracted with phenol-chloroform, and the DNA was precipitated, washed, and dissolved in TE buffer as described above.

Generation of mutants. To generate S. salivarius UB1309 salB::pMU1011 Sp^r, an internal 1.48-kb salB fragment (Fig. 1), PCR amplified from 20P3 DNA by using primers DsintF and ORF1R, was cloned into pFW5 to produce pMU1001, and this plasmid was then transformed onto the S. salivarius 20P3 chromosome. S. pyogenes UB1308 salB::pMU1016 Sp^r was generated in the same way by transforming pMU1016 (pFW5 carrying a 1.48-kb salB fragment from T11) onto the S. pyogenes T11 chromosome. S. salivarius UB1310 salR::pMU1002 Sp^r was generated by PCR amplifying a 1.16-kb fragment from the salKR gene region (Fig. 1) of 20P3 DNA with primers SalKF (positions 9249 to 9269; 5'GTTGGATTGTACTCATGAAGG) and SalRR (positions 10411 to 10391; 5'TCAACATAATCCTGAGATTCG), cloning this fragment into pFW5 to produce pMU1002, and transforming this plasmid into wild-type strain 20P3 as described above. Integration of pFW5 plasmid constructs onto the streptococcal chromosome was confirmed by PCR amplification with pFW5-specific primers pFW5F (5'GATCAGGAGTTGAGAGTGGAC) and pFW5R (5'TGGAGAAGATTCAGCCACTGC).

View larger version (14K): [in this window] [in a new window]   FIG. 1.   Genetic structures of the sal loci in S. salivarius 20P3 (A) and S. pyogenes SF370 (M type 1) (B). Open reading frames are depicted by arrows which show the numbers of aa residues in the deduced polypeptides. The sites of insertional inactivation with plasmid pFW5 constructs used to generate strains UB1309, UB1310, and UB1308 are shown. PCR-amplified segments of DNA used for hybridization probes or cloning in pFW5 are indicated as follows: I, 0.3-kb salA probe; II, 1.48-kb salB fragment; III, 1.16-kb salKR fragment. Symbols: , putative promoter; , inverted repeat.

RNA analysis. To prepare total RNA from streptococci, cells were harvested by centrifugation from M17CS medium, suspended in 0.2 ml of spheroplasting buffer (20 mM Tris-HCl [pH 6.8] containing 10 mM MgCl₂ and 26% [wt/vol] raffinose) containing 500 U of mutanolysin/ml and 0.1 mg of spectinomycin/ml, and incubated 37°C for 30 min. Spheroplasts were collected by centrifugation, and RNA was extracted by using a Qiagen RNAeasy kit and a Qiashredder column (Qiagen Ltd., Crawley, England) as recommended by the manufacturer. RNA samples (10 µg per lane) were subjected to electrophoresis through 1% (wt/vol) agarose in 1× MOPS (morpholinepropanesulfonic acid) buffer (0.2 M MOPS containing 50 mM sodium acetate and 10 mM EDTA; pH 7.0) supplemented with 1% (wt/vol) formaldehyde at 80 V for 3 h and were vacuum blotted onto a nylon membrane. For Northern analysis of sal transcripts, a PCR product generated with primers SalAF and SalAR (positions 901 to 883; 5'AGAAGTATCTAGTATGTCG) was radioactively labeled with [³²P]dATP by using Prime-a-Gene (Promega) as directed by the manufacturer. Hybridization reactions were carried out at 65°C for 18 h as described elsewhere (5). For reverse transcription (RT)-PCR analysis, DNase-treated RNA that had been extracted from exponentially growing cells was utilized for cDNA synthesis by using the SalRterm oligonuclotide primer and avian myeloblastosis virus reverse transcriptase (First Strand cDNA synthesis kit; Roche Diagnostics). PCR amplifications were then performed by utilizing the Expand Long Template PCR system (Roche Diagnostics) with primers SalAF and SalXR (positions 6510 to 6490; 5'CTCTCCTCTATCGGATAAAGC), primers SalY2S (positions 6571 to 6591; 5'CTATTTTCTAGCTACGATCGG) and SalRterm, or primers SalAF and SalRR.

SalA production and expression. SalA peptide was purified from S. salivarius 20P3 M17CS broth culture supernatant as previously described (34). To test for SalA induction of transcription, streptococcal strains were grown for 18 h in M17CS medium with or without SalA peptide. Cells were harvested by centrifugation (16,000 × g for 20 min at 4°C), and the cell-free supernatants were collected in sterile tubes. For medium swap experiments the bacterial pellets were suspended in the appropriate spent culture supernatant and incubated at 37°C for 4 h before RNA was extracted as described above. Levels of inhibitor production and of immunity to inhibition were determined by using a deferred antagonism assay performed on Columbia blood agar plates supplemented with 0.1% (wt/vol) CaCO₃ as described elsewhere (42). Levels of lantibiotic production by producer strains and growth inhibition of test strains were scored as 0 (no inhibition zone), 1 (narrow inhibition zone that was 1 to 2 mm wide), 2 (medium zone of inhibition that was 2 to 5 mm wide), 3 (larger zone of inhibition that was 5 to 8 mm wide), or 4 (inhibition zone that was more than 8 mm wide).

Nucleotide sequence accession number. The nucleotide sequence data for the S. salivarius 20P3 sal locus have been deposited in the GenBank database under accession number AY005472.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Structure of the sal locus in S. salivarius 20P3. To determine the sequence and genetic structure of the S. salivarius 20P3 sal gene locus, PCR primers were synthesized based on the nucleotide sequence of the approximately 7-kb region downstream of the salA1 gene in S. pyogenes SF370 (Fig. 1). These primers were utilized to PCR amplify segments that comprised the complete sal genetic locus in S. salivarius 20P3. The genetic locus comprised eight open reading frames designated salABCTXYKR (Fig. 1A). These open reading frame designations were assigned on the basis of database homology searches and were in accordance with conventional lantibiotic gene cluster nomenclature (7, 36).

Structure of the sal locus in S. pyogenes. S. pyogenes SF370 (M type 1) contains the salA-like gene designated salA1 (37) but does not produce detectable levels of lantibiotic inhibitor (Table 2). While the deduced aa sequence of the mature SalA1 peptide from strain SF370 has two conserved changes, residue R₂ to K and residue I₇ to F (Fig. 2), it is predicted that these changes do affect the hydrophobicity of the mature peptide or the maturation process. Indeed, the deduced aa sequence of SalA1 from the lantibiotic inhibitor-producing strain S. pyogenes T11 (M type 4) is identical to that of SF370 SalA1. In deferred antagonism tests the inhibitory profile of S. pyogenes T11 was similar to that of S. salivarius 20P3 (Table 2).

View larger version (10K): [in this window] [in a new window]   FIG. 2.   Comparison of the inferred aa sequences of the precursor SalA peptide in S. salivarius 20P3 and SalA1 peptide in S. pyogenes SF370. The arrow indicates the site of proteolytic cleavage (after residue 26) to generate the 22-aa SalA or SalA1 propeptides. The residues in boldface type are those likely to be involved in thioether bridge formation.

Transcriptional analysis of the sal locus in S. pyogenes and S. salivarius. For Northern analysis of salA mRNA transcript levels, nitrocellulose blots of streptococcal RNAs were incubated with a PCR-generated ³²P-labeled 0.3-kb DNA fragment (salA) (Fig. 1). The salA fragment probe reacted strongly with a single mRNA transcript approximately 200 bases long in S. salivarius 20P3 and less strongly with a band of similar size in S. pyogenes T11; no significant signal was obtained with mRNA extracts of S. pyogenes SF370 (Fig. 3 and 4). The relative salA mRNA levels corresponded well to the levels of inhibitor produced by these strains, as assayed by deferred antagonism on agar (Table 2). To confirm the identities of the other predicted transcripts of the locus, mRNA from S. salivarius 20P3 cells was subjected to RT-PCR analysis. A 5.9-kb product was generated with primers SalAF and SalXR (Fig. 5A), supporting the suggestion that the salABCTX genes were cotranscribed. In addition, amplification of a 4-kb fragment with primers SalYS and SalRterm demonstrated that there was transcriptional linkage of the salYKR genes (Fig. 5A). By utilizing primers SalAF and SalRR we were also able to demonstrate the presence of a 9.3-kb mRNA transcript corresponding to salABCTXYKR (Fig. 5B). In all these experiments no PCR products were obtained from RNA samples that had not been incubated with reverse transcriptase (Fig. 5).

View larger version (26K): [in this window] [in a new window]   FIG. 3.   Northern analysis of SalA1 gene transcription in S. pyogenes probed with a salA1 fragment corresponding to salA fragment I (Fig. 1B). Lane 1, S. pyogenes T11 grown in M17CS medium; lane 2, S. pyogenes T11 grown in S. pyogenes UB1308 salB culture medium; lane 3, S. pyogenes SF370 grown in M17CS medium; lane 4, S. pyogenes SF370 grown in strain T11 culture medium; lane 5, S. pyogenes UB1308 salB grown in M17CS medium; lane 6, S. pyogenes UB1308 salB grown in strain T11 culture medium. The position of the salA1 mRNA transcript (0.2 kb) is indicated by an arrow. The lanes contained equivalent amounts of total RNA (10 µg).

View larger version (50K): [in this window] [in a new window]   FIG. 4.   Autoinduction of SalA in S. salivarius 20P3: Northern analysis of salA mRNA transcripts from streptococci grown at 37°C for 30 min in M17CS medium (lane 1) and in M17CS medium containing purified SalA peptide at concentrations of 0.05 pmol/ml (lane 2), 0.5 pmol/ml (lane 3), and 5 pmol/ml (lane 4). The lanes contained equivalent amounts of total RNA (10 µg).

View larger version (53K): [in this window] [in a new window]   FIG. 5.   RT-PCR analysis of sal locus transcripts from S. salivarius 20P3. cDNA was generated from mRNA by using the oligonucleotide SalRterm. (A) Lanes 2 to 5 PCRs performed with primers SalAF and SalXR; lanes 7 to 10, PCRs performed with primers SalY2S and SalRterm; lanes 2 and 7, cDNA template generated by RT; lanes 3 and 8, RNA controls (no RT); lanes 4 and 9, chromosomal DNA template; lanes 5 and 10, no-template controls. (B) PCRs performed with primers SalAF and SalRR. Lane 2, cDNA template generated by RT; lane 3, RNA control; lane 4, chromosomal DNA; lane 5, no template. The molecular mass markers (panel A, lanes 1 and 6; panel B, lane 1) were  DNA HindIII fragments (23.1, 9.41, 6.56, 4.36, 2.32, and 2.02 kb).

Transcription of salA depends upon SalA modification and export. In order to test the hypothesis that the salBCT gene products were essential for production of SalA inhibitor, the salB gene in S. salivarius 20P3 was inactivated by inserting integrational plasmid pMU1011 in the salB coding region (Fig. 1A). A Northern blot analysis of mRNA extracted from UB1309 salB::pMU1011 in which the salA probe was used demonstrated that transcription of the salA gene was abrogated (Fig. 6, lane 1). S. salivarius UB1309 was phenotypically SalA as determined by an agar inhibition assay and did not significantly inhibit growth of S. pyogenes (Table 2). The analogous salB gene inactivation experiment was then carried out with S. pyogenes T11 by integrating plasmid pMU1016 in the salB chromosomal gene (Fig. 1B). The resulting isogenic mutant, UB1308, did not produce a 0.2-kb salA1 transcript (Fig. 3, lane 5) and was also inhibitor production negative (SalA1) as determined by the agar inhibition assay (Table 2). These results confirmed that the salBCT gene products were essential for production of SalA and SalA1 peptides and suggested that SalA and SalA1 lantibiotics modulated salA and salA1 gene transcription.

View larger version (44K): [in this window] [in a new window]   FIG. 6.   Northern analysis of salA gene transcription in S. salivarius UB1309 salB probed with salA fragment I (Fig. 1A). mRNAs were extracted from S. salivarius UB1309 cells grown at 37°C for 4 h. Lane 1, cells grown in their own cell-free spent culture medium; lane 2, cells grown in fresh M17CS medium containing 5 pmol of purified SalA/ml; lane 3, cells grown in cell-free S. salivarius 20P3 spent culture medium; lane 4, cells grown in cell-free S. pyogenes T11 spent culture medium.

Expression of SalA is autoregulated. To test the hypothesis that salA expression depended upon the presence of active SalA, S. salivarius 20P3 cells were removed from a late-exponential-phase culture in M17CS medium. The spent medium was inoculated with S. salivarius UB1309 salB cells; the control consisted of spent medium from a UB1309 culture. The cultures were incubated at 37°C for 4 h, and mRNAs were then extracted, used for Northern analysis, and probed with the salA gene fragment as described above. Following incubation in the S. salivarius 20P3 spent culture medium, the UB1309 salB cells produced a 0.2-kb salA mRNA transcript (Fig. 6, lane 3) that was absent from the control cells (Fig. 6, lane 1).

SalA and SalA1 are intra- and interspecies signaling molecules. To investigate the specificity of autoinduction by SalA and SalA1, medium swap experiments were carried out with the SalA1-producing organism S. pyogenes T11 and the non-SalA-producing organism S. salivarius UB1309 salB. In the latter strain, salA mRNA transcription was clearly induced by the S. pyogenes SalA1 peptide present in T11 culture medium (Fig. 6, lane 4), and the level was similar to that induced by the S. salivarius peptide SalA (Fig. 6, lane 3).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

It is now well established that gram-positive bacteria, especially lactic acid bacteria, produce an array of extracellular peptides with diverse functions. These molecules may act as highly specific intercellular signals for controlling competence development, mating responses, and virulence factor production in bacterial communities or may act as bacteriocin-like inhibitors of the growth of other neighboring species (10). Salivaricin A was one of the first bacteriocin-like inhibitors that were purified, sequenced, and characterized (34). This lantibiotic is produced by S. salivarius, and homologous SalA1 peptides are produced by M type 4 strains of S. pyogenes (37). In this study we extended our understanding of the production and function of this lantibiotic through comparative genetic analysis of the sal genetic loci in S. salivarius and S. pyogenes.

The genetic organization of the sal locus in strain 20P3 resembles that of other lantibiotic synthesis gene clusters (36). Transcriptional analyses suggested that there are at least two sal mRNA transcripts, a major (inducible) salA mRNA transcript approximately 200 bases long and a salABCTXYKR readthrough transcript. RT-PCR data are consistent with the proposal that the salBCTX genes are cotranscribed with salA, but the presence of an inverted repeat sequence downstream of the salA structural gene would be predicted to form a hairpin loop and attenuate transcriptional readthrough from salA into salBCTX. This would allow production of higher levels of SalA compared with the levels of SalBCTX, the modification and export machinery, and is a control feature found in a number of other lantibiotic synthesis gene operons (18, 23, 33). From the sequence analysis, we predicted that transcription of salYKR may also be initiated from a promoter immediately upstream of salY, but transcriptional start point mapping experiments would be necessary to confirm this. The significance of the different transcripts to lantibiotic production and control is not yet understood, but it is worth noting that among the transcripts from the L. lactis nisin lantibiotic gene locus, a transcript corresponding to the entire locus has been reported (31).

Comparison of the polypeptide sequences encoded by the sal gene loci in S. salivarius and S. pyogenes with the sequences in protein sequence databases allowed us to assign polypeptide functions. One unusual feature of the locus was the presence of two genes (salB and salC) encoding lantibiotic modification enzymes, since subclass AII lantibiotics with the characteristic Gly-Gly motif are usually modified by the product of a single gene (lanM) (36). The only gene for which a possible function of the product could not be inferred from a database match was salY. Since most lantibiotic gene clusters carry a self-immunity protein gene (36), it is possible that the 635-aa SalY protein is associated with self-protection against SalA. However, it is also possible that genetic determinants required for immunity are located elsewhere on the chromosome (2, 11).

Inactivation of the salB gene in S. salivarius 20P3 or S. pyogenes T11 led to a lantibiotic-negative (SalA or SalA1) phenotype, confirming that the salB gene product plays an essential role in lantibiotic production. However, inactivation of salB also resulted in abrogation of salA and salA1 transcription. This transcription was restored by exogenous addition of subinhibitory levels of SalA or SalA1 peptides, demonstrating that production of these lantibiotics is autoregulated, like production of nisin and subtilin (21, 23). Interestingly, salB mutants were also rendered sensitive to growth inhibition by SalA or SalA1 peptides, so lantibiotic production and immunity to the lantibiotic are coregulated. These results explain clearly why S. pyogenes strains which contain a deletion across the salBCT genes, such as SF370 (M type 1), are not producers of SalA1 lantibiotic and are also very sensitive to growth inhibition by SalA or SalA1 peptides.

A model for autoregulation of SalA production in S. salivarius is presented in Fig. 7; this model is based on experimental evidence presented here and comparative data for other lantibiotic synthesis systems (7, 8, 20, 21, 24). Transcription of the salABCTX genes is initiated at a promoter upstream of salA. The primary translation products include the 48-aa precursor pro-SalA and the SalB, SalC, SalT, and SalX protein modification and transport machinery. Modification of the precursor peptide occurs with a membrane-associated SalB-SalC enzyme complex, and the membrane-integral SalT-SalX polypeptides then mediate translocation of SalA across the cytoplasmic membrane (Fig. 7). Cleavage of the precursor SalA leader peptide results from cysteine protease activity located in the N-terminal region of SalT, which generates extracellular mature SalA peptide. The fate of the precursor SalA leader peptide is not known.

View larger version (19K): [in this window] [in a new window]   FIG. 7.   Model for autoregulation of SalA production in S. salivarius and S. pyogenes. The product of the salA gene, preproSalA peptide, is modified by the membrane-associated SalBC polypeptides and then exported with leader peptide cleavage by the membrane-integral SalTX polypeptides. Extracellular lantibiotic peptide from S. salivarius (SalA) or from S. pyogenes (SalA1) is sensed at the cell surface, possibly by the two-component SalKR system, and transcription of the salA promoter is upregulated via a phosphorylated (P) regulatory protein (R). Only the bacteria that respond to extracellular SalA peptide by producing active SalA are immune to the inhibitory effects of the lantibiotic peptide, but the mechanism of immunity is unknown.

The gene clusters encoding production of nisin (12), streptococcin SA-FF22 (26), and subtilin (22) each contain two genes for a two-component (sensor-regulator) system. It has been shown that in the nisin and subtilin systems the mature inhibitor peptide is recognized by the sensor kinase (21, 23, 30). We propose that mature SalA peptide is sensed by the sensor histidine kinase of a two-component system (putatively SalKR), which leads to activation of the response regulator (SalR) and induction of transcription from the salA promoter (Fig. 7). Although we were unable to generate a salK mutant and therefore were unable to formally test the notion that SalK is the cognate sensor for SalA, we generated a salR mutant that contained reduced salA mRNA transcript levels and exhibited reduced SalA inhibitor production. This is consistent with the proposed role for SalR in autoregulation of SalA, but since salA mRNA transcription was not shut down in this mutant, other regulatory pathways must also be operative.

An interesting finding is that SalA production by one streptococcal species may be induced by sensing of the homologous peptide from another streptococcal species. Interspecies signaling between taxonomically diverse streptococcal species has not been demonstrated previously. Most peptide signaling molecules are specific for their cognate sensor-response systems in streptococci (6, 10), but the SalA peptide sensing system apparently does not discriminate between SalA and SalA1 (Fig. 2). Minor modifications to other inhibitor peptides, such as epilancin K7 (44), streptococcin SA-FF22 (19), and subtilin (3), are sufficient to affect their biological activities, but the two conservative changes in SalA1 do not significantly affect the inhibitor activity or profile. Interspecies community sensing could regulate relative population levels of streptococci. The ability of SalA1-producing S. pyogenes strains to respond to SalA from S. salivarius (and vice versa) could provide a selective mechanism for cocolonization of the mucosal epithelium by pathogen and commensal cell populations. For example, SalA1 produced by rapidly multiplying S. pyogenes cells might stimulate production of SalA by S. salivarius strains, leading to modulation of the number of S. pyogenes cells. In this way, the number of S. pyogenes cells may be self-limiting, thus promoting maintenance of the population in a carrier state. An alternative scenario is that sensing of SalA by S. pyogenes may influence expression of other genes, including genes related to virulence. In this regard it has recently been demonstrated that a bacteriocin-like peptide (BlpC) in S. pneumoniae induces a set of 16 genes following activation of its cognate two-component sensor-regulator system (6). We are currently investigating the possibility that SalR, like CsrR (17) or Mga (25) proteins, may have a global regulatory function in S. pyogenes and modulate production of virulence factors, such as streptolysin S (28).

In summary, here we describe the genetic locus encoding the polypeptides necessary for autoregulated production of SalA and SalA1 lantibiotics by S. salivarius and S. pyogenes. The SalA and SalA1 peptides act as inhibitors of sensitive streptococcal strains which do not themselves produce these peptides, but they are also sensed by other SalA-producing or SalA1-producing strains and there is concomitant induction of lantibiotic synthesis. It could be envisaged, therefore, that SalA production and SalA1 production have a role in modulating the coexistence of streptococcal populations in vivo, where it is becoming apparent that signaling peptides are likely to have a major influence on oral microbial community structure.