• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Jun 2005; 187(12): 3980–3989.
PMCID: PMC1151730

Regulation of Bacteriocin Production in Streptococcus mutans by the Quorum-Sensing System Required for Development of Genetic Competence


In Streptococcus mutans, competence for genetic transformation and biofilm formation are dependent on the two-component signal transduction system ComDE together with the inducer peptide pheromone competence-stimulating peptide (CSP) (encoded by comC). Here, it is shown that the same system is also required for expression of the nlmAB genes, which encode a two-peptide nonlantibiotic bacteriocin. Expression from a transcriptional nlmAB′-lacZ fusion was highest at high cell density and was increased up to 60-fold following addition of CSP, but it was abolished when the comDE genes were interrupted. Two more genes, encoding another putative bacteriocin and a putative bacteriocin immunity protein, were also regulated by this system. The regions upstream of these genes and of two further putative bacteriocin-encoding genes and a gene encoding a putative bacteriocin immunity protein contained a conserved 9-bp repeat element just upstream of the transcription start, which suggests that expression of these genes is also dependent on the ComCDE regulatory system. Mutations in the repeat element of the nlmAB promoter region led to a decrease in CSP-dependent expression of nlmAB-lacZ. In agreement with these results, a comDE mutant and mutants unable to synthesize or export CSP did not produce bacteriocins. It is speculated that, at high cell density, bacteriocin production is induced to liberate DNA from competing streptococci.

The gram-positive bacterium Streptococcus mutans is a major cause of human dental caries due to its capacity to metabolize carbohydrates to generate lactic acid, even at low pH. Some of the virulence factors that determine colonization and survival of S. mutans in dental plaque, such as adherence, acid resistance, acidogenicity, and resistance to other stress conditions, have been well studied (4). Another determinant that could be important for colonization is the production of bacteriocins, ribosomally synthesized peptides or proteins with antibacterial activity. In a survey of 143 strains of S. mutans, 70% were shown to produce one or more bacteriocins in vitro (29). However, little information is available about the ecological significance of bacteriocins produced by S. mutans. It is not known whether bacteriocins are produced in dental plaque, although some studies have provided indirect evidence that this is the case (9).

Bacteriocins from gram-positive bacteria have been classified into three groups: class I bacteriocins contain the posttranslationally modified bacteriocins or lantibiotics, class II comprise small nonmodified, heat-stable peptide bacteriocins, and class III bacteriocins contain large, heat-labile proteins (21). Class II bacteriocins can be further divided into two groups: those that resemble the antilisterial bacteriocin pediocin from Pediococcus species (class IIA) and those that are composed of two peptides (class IIB). The class II bacteriocins are synthesized as prepeptides that contain a leader peptide with two glycine residues at the cleavage site. The mature peptide is cleaved off from the leader peptide after these two glycine residues and released in the medium by an ABC transporter. The class II bacteriocins act on other organisms by insertion in the membrane, causing formation of pores that perturb the membrane potential. Bacteriocin-producing organisms are resistant to their own bacteriocins through the action of immunity proteins.

Most of the bacteriocins from S. mutans characterized up to now belong to the group I lantibiotics. Production of two of these, mutacin I and mutacin 1140, appears to be dependent on the culture conditions, since they could be detected only after growth on agar or agarose-containing medium (10, 27). A nonlantibiotic class IIB bacteriocin, mutacin IV, was purified from S. mutans strain UA140, which also produces the lantibiotic mutacin I (27). Whereas mutacin I could be produced only when cells were grown on solid medium, mutacin IV was produced in liquid culture exclusively. Mutacin IV is encoded by the nlmA and nlmB genes, which are probably organized in an operon.

Expression of class II bacteriocins generally requires an inducer peptide pheromone and a two-component signal transduction system (TCSTS) (20). The TCSTS is composed of a histidine protein kinase that serves to sense the presence of the peptide outside of the cell and transmits the signal over the membrane to a response regulator that activates transcription of the target genes. In many cases, these systems are autoregulated (20). Related two-component regulatory systems are required for competence development in streptococci. In S. mutans, development of competence is under the control of the comCDE system, where comC encodes the precursor of the pheromone competence-stimulating peptide (CSP) and comD and comE encode the histidine kinase sensor and response regulator, respectively (15). The precursor of CSP is processed and exported through an ATP-binding cassette transporter and an accessory protein, which are encoded by comA and comB, respectively (37).

In this study, it is shown that expression of S. mutans nlmAB, encoding mutacin IV, and that of other class IIB bacteriocins and associated genes are induced upon addition of CSP and that inactivation of the two-component regulatory system required for competence development abolishes the production of bacteriocins.



Reagents for molecular biology were obtained from Fermentas (Vilnius, Lithuania) and New England Biolabs (Beverly, MA). CSP from S. mutans (NH2-SGSLSTFFRLFNRSFTQALGK-COOH) was synthesized by ESGS (Evry, France). The primers used for generation of mutants and lacZ fusion strains are listed in Table Table11 and were obtained from Microsynth (Balgach, Switzerland).

Primers used for construction of mutants and lacZ fusion strains

Strains and growth conditions.

The strains of S. mutans that were used in this study are listed in Table Table2.2. OMZ1001 is a derivative of strain UA159 (ATCC 700610), characterized by a high transformation efficiency. It was isolated in our laboratory. S. mutans OMZ1001 and OMZ67 (GS-5), Streptococcus oralis OMZ607 (12), and Streptococcus gordonii OMZ505 (ATCC 10558) were grown routinely in Todd-Hewitt broth supplemented with 0.3% yeast extract (THY) at 37°C under aerobic conditions or under anaerobic conditions (5% CO2, 10% H2, 85% N2). When required, antibiotics were used at the following concentrations: erythromycin, 15 μg/ml; kanamycin, 750 μg/ml.

S. mutans strains used in this study

Escherichia coli strain DH5α was used as a host for plasmids and grown in LB medium. When required, antibiotics were added at the following final concentrations: erythromycin, 200 μg/ml; kanamycin, 50 μg/ml.

DNA manipulations.

Plasmid DNA purification from E. coli, restriction enzyme digestion, agarose gel electrophoresis, DNA ligation, and transformation of E. coli were carried out using standard methods (3). Chromosomal DNA from S. mutans was isolated by using the Genelute bacterial genomic DNA kit (Sigma, Buchs, Switzerland).

Construction of lacZ fusion strains.

To generate a chromosomal nlmAB-lacZ fusion, a fragment internal to the nlmAB genes (bp 205 to bp 614 [27]) was amplified by PCR using the primers nlmAB1 and nlmAB2 (Table (Table1)1) and chromosomal DNA from OMZ1001. The PCR fragment was digested with BamHI and EcoRI and cloned into the vector pSF151 (31) to give plasmid pOMZ46. Subsequently, a BamHI fragment harboring a promoterless lacZ gene (11) was inserted downstream of nlmAB in the same direction of transcription, which resulted in plasmid pOMZ47. The plasmid was then introduced into S. mutans OMZ1001 by transformation (23) followed by selection for kanamycin resistance. Since pSF151 and its derivatives cannot replicate in S. mutans, kanamycin-resistant colonies must have arisen from single crossover and contain a lacZ fusion to nlmAB. The resulting strain was designated OMZ1008.

For the construction of the lacZ fusions with the bsmA, bsmE, bsmHI, bsmK, immB, and immA genes of S. mutans, a derivative of plasmid pSF151 was prepared by digestion with XbaI and PstI, followed by filling in with T4 DNA polymerase and self-ligation, resulting in pOMZ119. A 3.1-kb BamHI fragment from pALH122 harboring the lacZ gene was cloned in the BamHI site of pOMZ119 to give pOMZ125. Fragments harboring part of the gene of interest were generated by PCR amplification and cloned into the SphI/XbaI sites of plasmid pOMZ125. The resulting plasmids were introduced into S. mutans OMZ1001 by transformation, again followed by selection for kanamycin resistance. For each strain, correct integration was confirmed by PCR analysis.

Transcriptional fusions between mutated nlmAB promoter regions and lacZ were generated as follows. PCR of the region upstream of and including the prospective mutation was carried out with a primer that contained an XbaI restriction site (nlmAB10) and a primer that introduced a second restriction site and the desired mutation (Table (Table1).1). In a second PCR, the region downstream of the mutation was amplified with a primer that introduced an SphI site (nlmAB11) and a primer that introduced the same restriction site and mutation as those in the first PCR. The two PCR products were digested with the enzymes that recognized the newly created restriction sites and ligated together into pOMZ125, which had been digested with SphI and XbaI. The resulting plasmids were subsequently transformed into S. mutans. Restriction analysis of PCR products was then used to confirm that the mutation was present at the expected location.

Construction of mutants.

Mutants were constructed by first cloning fragments containing portions of the 5′ and the 3′ ends of the gene or genes of interest up- and downstream of the erythromycin resistance gene of the plasmid pFW15 (26). The resulting plasmids were then linearized and introduced into S. mutans by transformation and selection for erythromycin resistance. The mutants were verified by PCR analysis. As summarized in Table Table22 this procedure was applied to the S. mutans strains OMZ1001 and OMZ67, as well as to various derivatives of OMZ1001 containing lacZ fusions.

Specifically, a mutant of S. mutans in which part of both the comD and the comE gene was replaced by the erythromycin resistance gene was constructed as follows. A fragment was amplified by PCR using the primers comDE3 and comDE4 and chromosomal DNA from S. mutans OMZ1001. The PCR product was then digested with NcoI and PstI and cloned into the vector pFW15 to give plasmid pOMZ43. A second fragment was obtained by PCR with primers comDE1 and comDE2. This fragment was first cloned in pBluescript KS as a BamHI/XbaI fragment, recovered by digestion with BamHI and SacI, and cloned into pOMZ43 to give plasmid pOMZ61. pOMZ61 was then linearized with BamHI and introduced into S. mutans by transformation. Erythromycin-resistant colonies were selected, and one of the colonies was purified for further use. Analogously, mutants were generated, which contained a deletion/erythromycin resistance insertion in comC, ciaRH, comA, and comX. The vicK mutants were constructed by transformation of S. mutans with chromosomal DNA from strain SMUHK1, a vicK mutant obtained from D. Cvitkovitch.

β-Galactosidase and bacteriocin assays.

β-Galactosidase was measured according to the method of Miller (17) with o-nitrophenylgalactoside as substrate.

To measure bacteriocin production, producer strains were grown overnight in THY broth, stab inoculated into THY agar, and grown anaerobically for 24 or 48 h. The indicator strains S. oralis OMZ607 and S. gordonii OMZ505 were grown overnight in THY broth, and 0.1 ml of the indicator culture was mixed with 4 ml of molten THY top agar and poured over the plate. The plates were incubated for 24 h at 37°C in an anaerobic chamber, and the diameter of the zone of inhibition around the producing strains was measured.

Determination of the transcription initiation site of bsmA.

S. mutans OMZ1001 was grown to the early-exponential phase in THY at 37°C under aerobic conditions, CSP was added to a final concentration of 100 ng/ml, and growth was continued for 2 h. RNA was isolated by using the FastRNA Pro Blue kit (Qbiogene, Inc., Carlsbad, Calif.) and the FastPrep instrument (Qbiogene). For primer extension analysis, 20 pmol of primer bsmAPE (5′-GCTGTACCGCCTGCAGTAGCCATATAAC-3′), labeled at its 5′ end with 6-carboxyfluorescein, was annealed with 20 μg of RNA. cDNA was produced by using Superscript II (Invitrogen, Basel, Switzerland). The product was precipitated with ethanol and dissolved in 2 μl of water, and 1 μl was loaded on an ABI Prism sequence analyzer model 3100 (Applied Biosystems, Rotkreuz, Switzerland). Fragment sizes were calculated by applying Genescan DB-30 markers (Applied Biosystems) together with the sample. To verify the size of the reverse transcriptase reaction product, a PCR was carried out with the 6-carboxyfluorescein-labeled primer bsmAPE and a second primer, using chromosomal DNA from S. mutans OMZ1001 as template. The product, which should be 200 bp in size, was applied together with the Genescan DB-30 standard. The determined product size was 197 bp. Therefore, to correct for this difference, 3 bp was added to the size determined for the reverse transcription product.


Expression of β-galactosidase from a transcriptional nlmAB′-lacZ fusion.

Expression of β-galactosidase from the transcriptional nlmAB′-lacZ fusion present in strain OMZ1008 was measured during aerobic growth in THY broth (Fig. (Fig.1).1). An initial decrease upon entry in the exponential growth phase and an increase upon entry in the stationary growth phase were observed. When S. mutans OMZ1008 was grown anaerobically for 48 h in solid THY medium, expression of β-galactosidase from the nlmAB′-lacZ fusion was increased approximately sixfold compared to cells that were grown anaerobically in liquid THY medium (253 ± 125 Miller units and 41 ± 6 Miller units, respectively). There was a small difference in expression between aerobically and anaerobically grown cells after 48 h (58 ± 10 and 41 ± 6 Miller units, respectively).

FIG. 1.
Induction of nlmAB-lacZ expression by CSP. An overnight culture of S. mutans strain OMZ1008 was diluted 100-fold in fresh medium and grown aerobically at 37°C. After 4 h of growth, the culture was split and CSP was added to a final concentration ...

It was investigated whether addition of the competence-stimulating peptide had an effect on expression of nlmAB-lacZ. Addition of CSP at a concentration of 0.5 μg/ml to an aerobic liquid culture led to a strong increase of β-galactosidase activity (Fig. (Fig.1).1). After 6 h of growth, the activity of β-galactosidase was about 60-fold higher in cells that were grown with CSP than in cells that were grown in its absence.

The amount of CSP added to the growth medium was varied from 1 ng/ml to 500 ng/ml, and β-galactosidase activity was measured (Fig. (Fig.2).2). Above 5 ng/ml, an increase in β-galactosidase activity was observed. This stimulation was linear until a CSP concentration of about 100 ng/ml. Thus, the expression of nlmAB is dependent on the amount of CSP present in the growth medium and NlmAB bacteriocin synthesis is therefore regulated as part of the comCDE quorum-sensing regulon.

FIG. 2.
Influence of the amount of CSP on expression of nlmAB-lacZ. An overnight culture of S. mutans strain OMZ1008 was diluted 100-fold in fresh medium and grown aerobically at 37°C. After 4 h of growth, the culture was divided in several aliquots. ...

Expression of nlmAB′-lacZ in mutants.

To consolidate the above results and to investigate the effect of mutations in genes known to influence competence development, β-galactosidase expressed from the nlmAB′-lacZ fusion was measured in different mutants (Table (Table3).3). Activity in the comC and comA mutants was enhanced as in the wild-type strain, but the comDE mutant did not show an increase in activity of β-galactosidase after CSP addition. This confirms that the three-component comCDE system is necessary for expression of nlmAB.

β-Galactosidase activity expressed from a transcriptional nlmAB′-lacZ fusion in wild-type and mutant strains

In analogy to the situation in Streptococcus pneumoniae (13), the comX gene of S. mutans (SMU.1997) is thought to encode an alternative sigma factor required for transcription of the late competence genes, whose expression is induced about 5 min later than that of the genes directly regulated by ComE (25). The comX gene of S. mutans was previously found to be necessary for competence (15). But the comX mutation had no effect on expression of β-galactosidase from the nlmAB′-lacZ fusion, indicating that the ComX protein is not required as an alternative sigma factor for transcription of nlmAB.

The ciaRH genes (SMU.1128 and SMU.1129) of S. mutans encode a two-component signal transduction system that is involved in competence development. Inactivation of ciaH, encoding the histidine kinase, led to a decrease of competence (28), whereas interruption of the complete ciaRH gene cluster resulted in an increase of competence (J. R. van der Ploeg, unpublished results). Interestingly, a ciaH deletion, but not a ciaR deletion, was found to abolish production of the lantibiotic bacteriocin mutacin but had no influence on production of mutacin IV (encoded by nlmAB) (28). The ciaRH mutant investigated in this study is defective in both ciaR and ciaH, yet this mutation had no apparent effect on expression of nlmAB-lacZ (Table (Table33).

Another two-component signal transduction system, encoded by vicRK, has been shown to influence competence development (D. Cvitkovitch, personal communication). The vicK gene (SMU.1516) encodes a histidine kinase, whereas vicR (SMU.1517) encodes a response regulator. When a vicK mutation was introduced into OMZ1008, the level of nlmAB′-lacZ expression in uninduced cultures was higher than in uninduced cultures of the wild-type strain (Table (Table3),3), indicating that vicK acts negatively on nlmAB expression.

Several putative bacteriocins are regulated by CSP and the comDE system.

Analysis of the S. mutans genome sequence (1) revealed 10 small open reading frames with high similarity to the leader peptides of NlmA and NlmB that could encode class II bacteriocins (Fig. (Fig.33 and and4).4). The putative bacteriocins, designated Bsm (bacteriocin Streptococcus mutans) ranged in size from 47 to 87 amino acids, had leader peptides from 22 to 25 amino acids (Table (Table44 and Fig. Fig.3),3), and contained a double glycine motif that could be recognized by the ComAB processing and export system. Some of the genes encoding the putative bacteriocins were located in tandem, indicating that they might act cooperatively, as is typical for class IIB bacteriocins (Table (Table44 and Fig. Fig.4).4). Most of the putative bacteriocins were predicted to be of hydrophobic nature. Out of these putative bacteriocins, eight are encoded within a region of about 20 kb in size, which also harbors the comC, comD, and comE genes (Fig. (Fig.4)4) (15) Two pairs of bacteriocin-encoding genes (bsmFG and bsmHI) appear to be separated by an insertion element (Fig. (Fig.4).4). One of the bacteriocin-encoding genes (bsmA, SMU.1914c) is located immediately upstream of the comC gene (SMU.1915c), but in the opposite direction of transcription.

FIG. 3.
Clustal W sequence alignment of the leader peptide sequences of putative type II bacteriocins and CSP. Asterisk, identical residue; colon, conserved residue; period, semiconserved residue.
FIG. 4.
Genetic organization of loci harboring the bsm, imm, and competence-related genes. Genes shown by black arrows and arrowheads encode putative bacteriocins, whereas genes associated with competence and bacteriocin production are shown by shaded arrows. ...
Putative bacteriocins encoded by S. mutans UA159

DNA sequence analysis showed that the region upstream of the putative bacteriocin-encoding genes bsmA, bsmB, bsmC, and nlmAB was highly conserved. This region was also found to be present upstream of an open reading frame encoding a putative immunity protein (immB, SMU.925). The ImmB protein sequence was 26% identical to BlpL, a putative immunity protein from S. pneumoniae (6). The conserved region contains a 9-bp repeat, which is separated by 12 bp (Fig. (Fig.5).5). The sequence of the repeat element matched with the previously proposed consensus binding site sequence for response regulators from the AlgR/AgrA/LytR family (22). Putative bacteriocins encoded by tandemly organized bsmH and bsmI also contained a repeat element upstream of a possible extended −10 region, but the repeat did not completely match with the proposed consensus sequence (not shown).

FIG. 5.
Sequence alignment of upstream regions from four genes encoding putative bacteriocins and one gene encoding a putative bacteriocin immunity protein. The direct repeat elements and the extended −10 promoter regions are underlined. The consensus ...

The presence of this highly conserved region indicated that expression of these bacteriocin-encoding genes and associated genes is under coordinate control by the comDE two-component signal transduction system. In order to verify this hypothesis, chromosomally encoded transcriptional lacZ fusions to bsmA and immB were constructed and β-galactosidase activity was measured in a wild-type strain and in a comDE background with or without addition of CSP to the medium. Expression of the bsmA-lacZ and immB-lacZ fusions was found to be inducible by CSP and dependent on functional ComDE (Table (Table5),5), supporting the assumption that the five loci are coregulated. Expression of bsmA-lacZ was highest in the stationary phase of growth (data not shown). A gene encoding a second putative bacteriocin immunity protein, designated immA (SMU.1913c), whose deduced protein sequence has 80% identity to that of immB, was detected downstream of bsmA. β-Galactosidase expressed from an immA′-lacZ fusion was also stimulated by CSP and dependent on comDE (Table (Table5).5). Thus, both immA, which is probably transcribed in one unit with bsmA, and immB are coexpressed with bacteriocins.

β-Galactosidase activities expressed from transcriptional lacZ fusions in a wild-type background and in a comDE mutanta

The transcription start site of the bsmA gene was mapped 42 bp downstream from the repeat element closest to the start codon of bsmA (data not shown) (Fig. (Fig.5).5). Upstream of the transcription start, a well-conserved extended −10 region was present.

To further confirm that the direct repeat is required for CSP-dependent expression, transcriptional lacZ fusions to genes that lacked this repeat (bsmE, bsmHI, and bsmK) were generated and β-galactosidase was measured (Table (Table5).5). Expression of the bsmE-lacZ and the bsmHI-lacZ fusions was not increased by addition of CSP and not dependent on comDE. The bsmK-lacZ fusion exhibited a small increase in β-galactosidase activity upon addition of CSP and showed less-than-twofold-reduced levels of activity in a comDE background. It is possible that this dependence on comCDE results from expression from the bsmBL promoter, which is located about 2.3 kb upstream of bsmK and which contains a conserved direct repeat element (Fig. (Fig.44).

Mutational analysis of the nlmAB direct repeat.

To substantiate the hypothesis that the direct repeat could function as a binding site for a transcriptional regulator (most likely ComE) and thus be required for CSP-dependent expression, chromosomally located transcriptional fusions between the nlmAB promoter region and lacZ were constructed. Mutations were introduced in the direct repeat element, and expression of these nlmAB-lacZ fusions was measured in cultures that had received CSP (Fig. (Fig.6).6). Both mutations in repeat 2 led to approximately 10-fold-reduced levels of β-galactosidase activity, whereas the mutation in repeat 1 resulted in 40-fold-lower levels of β-galactosidase. Removal of both repeats and the region in between abolished expression nearly completely. These data clearly demonstrate that the direct repeat is important for full expression of nlmAB.

FIG. 6.
Mutations in the direct repeat of the nlmAB promoter region and their effect on activity of β-galactosidase expressed from a transcriptional nlmAB′-lacZ fusion. Direct repeats are underlined. Residues that were mutated are shown in lowercase ...

Competence mutants do not produce bacteriocins.

The results presented above have shown that expression of several genes related to bacteriocin synthesis is controlled by the comCDE quorum-sensing system. It was investigated whether this could be correlated with in vivo bacteriocin production by studying growth inhibition of selected indicator strains by S. mutans strains with mutations in the competence genes (Table (Table6).6). S. mutans strain OMZ1001 bacteriocins were assayed with S. gordonii OMZ505, whereas S. oralis OMZ607 served as an indicator for bacteriocins produced by S. mutans OMZ67. The zone of growth inhibition of the indicator strains triggered by the two wild-type strains was not affected by the presence of CSP. Apparently, the amount of CSP produced was sufficient to induce production of bacteriocins. While comC and comDE mutants did not produce bacteriocins in the absence of CSP, the comC mutant could be complemented by exogenous CSP. Both comA mutants failed to produce bacteriocins in the absence of CSP. Addition of CSP could complement the comA mutant of OMZ67 but not the comA mutant of OMZ1001. This suggests that in the latter strain the ComAB system is required for both processing and transport of both CSP and bacteriocins, whereas strain OMZ67 might have a second system for processing and export of bacteriocins. Other mutations tested had no effect on production of bacteriocins by either strain of S. mutans.

Growth inhibition by S. mutans wild-type and mutant strainsa


Two-component signal transduction systems that respond to the presence of inducer peptide pheromones are known to regulate a variety of processes: bacteriocin production, competence for transformation, biofilm formation, and virulence (30). The results presented here and those from previous studies (14, 15) show that S. mutans uses one and the same TCSTS to control at least three of these at first sight unrelated processes. Mutants in comA, comC, and comDE are competence deficient (14), form biofilms that are different from those formed by wild-type strains (15), and do not produce bacteriocins (this study; 32, 36).

Unlike competence, which exhibits an optimum during a short interval of the exponential phase (data not shown) (19, 23), expression of nlmAB-lacZ was lowest in this period but reached a maximum in the stationary phase. Since both activities are controlled by the same regulatory system, the question arises as to how this differential regulation is established. The key to the answer may lie in the fact that expression of genes required for competence development is governed by the accessory sigma factor ComX, whereas production of bacteriocins is not, as shown in the present study. In S. pneumoniae, the comX gene, under the regulatory control of the comCDE system, is transcribed only during the first 15 min after addition of CSP, but then transcription is shut off by an unknown mechanism (25). This short period of transcription of comX, together with the short half-life of the ComX protein, results in transient expression of the late competence genes. A similar situation probably applies to S. mutans, since expression of a comX-lacZ fusion was induced only during the first minutes after addition of CSP (2). In contrast, the nlmAB-lacZ fusion exhibited an increase of β-galactosidase activity more than 2 h after addition of CSP.

Many of the bacteriocin-encoding genes are located in the vicinity of the comCDE genes. It is therefore likely that ComE binds to the direct repeat upstream of these genes to activate transcription of these genes, but it cannot be ruled out that ComE regulates the activity of a second transcription factor that then acts on the promoter regions of bacteriocins. In S. pneumoniae, the ComE protein binds to the promoter region of the comCDE and comAB (33), and it was expected to bind as well upstream of comX (13). However, no such binding sites in front of the S. mutans comAB, comX, comC, and comDE genes could be detected.

As many as 12 putative bacteriocins are encoded on the chromosome of S. mutans. There could be even more, as pointed out by a recent survey (7). The finding of so many type II bacteriocins raises several questions: (i) towards which species are these bacteriocins active, (ii) under what conditions are they expressed, and (iii) is the production of bacteriocins relevant for colonization or survival of S. mutans in dental plaque?

Although there is no obvious sequence similarity among the mature peptides, all putative bacteriocins from S. mutans are hydrophobic, suggesting that they interact with the bacterial cytoplasmic membrane. Since purified NlmAB bacteriocin is active against a broad range of oral streptococci (27), it is likely that the others are also directed towards these organisms. Detailed analysis with purified peptides is necessary to obtain conclusive evidence about the spectrum of the microbes that are sensitive. Class IIB bacteriocins generally act much more efficiently than two-peptide systems. It is possible that different combinations of bacteriocins have different antimicrobial spectra. The 12 peptides encoded in the genome of S. mutans would thus allow for a large repertoire.

Consistent with this hypothesis is the observation that S. mutans OMZ1001 produces bacteriocins against S. gordonii OMZ505 but not against S. oralis OMZ607, whereas S. mutans strain OMZ67 inhibited growth of OMZ607 but not of OMZ505 (results not shown), assuming heterogeneity with respect to the bacteriocins produced by these strains.

The second question concerning conditions of bacteriocin expression has been answered to some extent in this study. At least the six bacteriocins that contain the conserved upstream direct repeat element are likely to be regulated by the comCDE signal transduction system. The level of expression of nlmAB was found to be dependent on the concentration of CSP. It is not known how expression of the comC gene is regulated in S. mutans, but it seems likely that the concentration of CSP reaches its maximum in the stationary phase. Growth in biofilms is also expected to lead to high concentrations of CSP. Indeed, when strains were grown on plates, conditions comparable to growth, expression of nlmAB′-lacZ was higher than when strains were grown in liquid cultures. This correlates with the observation that addition of CSP to plates did not lead to a large increase of bacteriocin production by the wild-type strain, whereas β-galactosidase activities were strongly increased upon addition of CSP in liquid cultures. My results appear to contrast with those of Qi et al., who found higher expression of nlmAB in liquid cultures than on plates (27). Recently, it was shown that S. mutans GS-5, which is the same as strain OMZ67 used in this study, produces a two-peptide lantibiotic bacteriocin named Smb. Expression of the smb operon that encodes the bacteriocin structural proteins, immunity, processing, and transport were found to be dependent on the comCDE system (36). The promoter region of this operon contained a very similar direct repeat element as described here. Thus, the comCDE system appears to regulate nonlantibiotic as well as lantibiotic bacteriocin synthesis.

Derivatives of strain OMZ1001 and OMZ67 deficient in comA, comC, or comDE lost the ability to inhibit growth of S. gordonii OMZ505 and S. oralis OMZ607, respectively. This can likely be ascribed to the loss in expression of the bacteriocin genes that contain the direct repeat region. But these mutants may have kept the ability to produce one or more of the remaining bacteriocins. The observation that the com mutants did not inhibit growth of indicator strains suggests either that these remaining bacteriocins are not expressed under the conditions investigated or that they are expressed but active against other organisms. The latter explanation appears most likely, since the bsmE, bsmK, and bsmHI genes were expressed when S. mutans was grown in liquid medium.

The loss of growth inhibition by the comA mutant of OMZ1001 could not be complemented by addition of CSP. This suggests that, at least in this strain, processing and export of CSP and bacteriocins are carried out by the same transport proteins. This is not completely unexpected, since the N-terminal sequences of the bacteriocins and that of CSP are similar in size and sequence (Fig. (Fig.3).3). The existence of a unique transporter for bacteriocins and for the inducing peptide has also been proposed for bacteriocin production by Lactobacillus sakei (5). The comA mutant of strain OMZ67 could be complemented by addition of CSP. It might well be that processing and export of the lantibiotic bacteriocin Smb produced by this strain require a specific protein. Indeed, a gene that could encode a bacteriocin transporter was present in the smb operon (36).

What is the link between competence, biofilm formation, and bacteriocin production? Several studies have shown that oral streptococci lacking one of the components of the comCDE system produce poorer or differently structured biofilms (15, 16). In Streptococcus intermedius, biofilm formation was stimulated by addition of CSP (24). This stimulation was detectable at CSP concentrations that were in the same range as the concentrations required for induction of expression of nlmAB-lacZ. In addition, competence of S. mutans was found to be increased when grown in biofilms (14). Other studies have also suggested a connection between competence and bacteriocin production. In Bacillus subtilis, production of the dipeptide bacilysin was regulated by the system required for genetic competence (35). In S. pneumoniae, genes encoding bacteriocins are also induced by CSP and regulated by comDE, but they belong to the late-induced genes. The late-induced genes do not contain a direct repeat sequence that could act as a binding site for ComE but instead require ComX and have a conserved element (TACGAATA) just upstream of the transcription start site (13). This element could not be detected upstream of the genes encoding bacteriocins in S. mutans.

It has been shown recently that S. pneumoniae releases chromosomal DNA after addition of CSP and that this DNA release is dependent on ComD and ComE (18). Interestingly, the highest amounts of DNA were released in the stationary phase of growth, which coincides with the maximum in expression of bacteriocins in S. mutans. Although a speculative hypothesis, it is possible that bacteriocins produced during the stationary phase aid in release of DNA either from S. mutans itself or from competing streptococci by permeabilization of the cytoplasmic membrane. The DNA that is liberated could serve several functions. First, it might function as a component of a biofilm structure. It has recently been shown that addition of DNase inhibited biofilm formation of S. intermedius and Pseudomonas aeruginosa (24, 34), indicating that DNA contributes to biofilm development. Second, the DNA, and possibly other cellular constituents, might fulfill a nutritional requirement for starving cells (8). The “classical” role of competence development, uptake of DNA and subsequent recombination to increase the fitness, remains as a third possibility.


I thank Bernhard Guggenheim for generous support, Verena Osterwalder for excellent technical assistance, and Christoph Wyss and Rudolf Gmür for critical reading of the manuscript. A. Honeyman, A. Podbielski, and D. Cvitkovitch are gratefully acknowledged for providing plasmids or strains.

This work was supported by a grant from the University of Zürich.


1. Ajdic, D., W. M. McShan, R. E. McLaughlin, G. Savic, J. Chang, M. B. Carson, C. Primeaux, R. Tian, S. Kenton, H. Jia, S. Lin, Y. Qian, S. Li, H. Zhu, F. Najar, H. Lai, J. White, B. A. Roe, and J. J. Ferretti. 2002. Genome sequence of Streptococcus mutans UA159, a cariogenic dental pathogen. Proc. Natl. Acad. Sci. USA 99:14434-14439. [PMC free article] [PubMed]
2. Aspiras, M. B., R. P. Ellen, and D. G. Cvitkovitch. 2004. ComX activity of Streptococcus mutans growing in biofilms. FEMS Microbiol. Lett. 238:167-174. [PubMed]
3. Ausubel, F. M., R. Brent, R. E. Kingston, D. E. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. Wiley, New York, N.Y.
4. Bowden, G. H., and I. R. Hamilton. 1998. Survival of oral bacteria. Crit. Rev. Oral Biol. Med. 9:54-85. [PubMed]
5. Brurberg, M. B., I. F. Nes, and V. G. Eijsink. 1997. Pheromone-induced production of antimicrobial peptides in Lactobacillus. Mol. Microbiol. 26:347-360. [PubMed]
6. de Saizieu, A., C. Gardes, N. Flint, C. Wagner, M. Kamber, T. J. Mitchell, W. Keck, K. E. Amrein, and R. Lange. 2000. Microarray-based identification of a novel Streptococcus pneumoniae regulon controlled by an autoinduced peptide. J. Bacteriol. 182:4696-4703. [PMC free article] [PubMed]
7. Dirix, G., P. Monsieurs, K. Marchal, J. Vanderleyden, and J. Michiels. 2004. Screening genomes of Gram-positive bacteria for double-glycine-motif-containing peptides. Microbiology 150:1121-1126. [PubMed]
8. Finkel, S. E., and R. Kolter. 2001. DNA as a nutrient: novel role for bacterial competence gene homologs. J. Bacteriol. 183:6288-6293. [PMC free article] [PubMed]
9. Hillman, J. D., A. L. Dzuback, and S. W. Andrews. 1987. Colonization of the human oral cavity by a Streptococcus mutans mutant producing increased bacteriocin. J. Dent. Res. 66:1092-1094. [PubMed]
10. Hillman, J. D., J. Novak, E. Sagura, J. A. Gutierrez, T. A. Brooks, P. J. Crowley, M. Hess, A. Azizi, K. Leung, D. Cvitkovitch, and A. S. Bleiweis. 1998. Genetic and biochemical analysis of mutacin 1140, a lantibiotic from Streptococcus mutans. Infect. Immun. 66:2743-2749. [PMC free article] [PubMed]
11. Honeyman, A. L., C. K. Cote, and R. Curtiss III. 2002. Construction of transcriptional and translational lacZ gene reporter plasmids for use in Streptococcus mutans. J. Microbiol. Methods 49:163-171. [PubMed]
12. Kilian, M., L. Mikkelsen, and J. Henrichsen. 1989. Taxonomic study of viridans streptococci: description of Streptococcus gordonii sp. nov. and emended descriptions of Streptococcus sanguis (White and Niven 1946), Streptococcus oralis (Bridge and Sneath 1982), and Streptococcus mitis (Andrewes and Horder 1906). Int. J. Syst. Bacteriol. 39:471-484.
13. Lee, M. S., and D. A. Morrison. 1999. Identification of a new regulator in Streptococcus pneumoniae linking quorum sensing to competence for genetic transformation. J. Bacteriol. 181:5004-5016. [PMC free article] [PubMed]
14. Li, Y. H., P. C. Lau, J. H. Lee, R. P. Ellen, and D. G. Cvitkovitch. 2001. Natural genetic transformation of Streptococcus mutans growing in biofilms. J. Bacteriol. 183:897-908. [PMC free article] [PubMed]
15. Li, Y. H., N. Tang, M. B. Aspiras, P. C. Lau, J. H. Lee, R. P. Ellen, and D. G. Cvitkovitch. 2002. A quorum-sensing signaling system essential for genetic competence in Streptococcus mutans is involved in biofilm formation. J. Bacteriol. 184:2699-2708. [PMC free article] [PubMed]
16. Loo, C. Y., D. A. Corliss, and N. Ganeshkumar. 2000. Streptococcus gordonii biofilm formation: identification of genes that code for biofilm phenotypes. J. Bacteriol. 182:1374-1382. [PMC free article] [PubMed]
17. Miller, J. H. 1992. A short course in bacterial genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
18. Moscoso, M., and J.-P. Claverys. 2004. Release of DNA into the medium by competent Streptococcus pneumoniae: kinetics, mechanism and stability of the liberated DNA. Mol. Microbiol. 54:783-794. [PubMed]
19. Murchison, H. H., J. F. Barrett, G. A. Cardineau, and R. Curtiss III. 1986. Transformation of Streptococcus mutans with chromosomal and shuttle plasmid (pYA629) DNAs. Infect. Immun. 54:273-282. [PMC free article] [PubMed]
20. Nes, I. F., and V. G. H. Eijsink. 1999. Regulation of group II peptide bacteriocin synthesis by quorum-sensing mechanisms, p. 175-192. In G. M. Dunny and S. C. Winans (ed.), Cell-cell signaling in bacteria. ASM Press, Washington, D.C.
21. Nes, I. F., and H. Holo. 2000. Class II antimicrobial peptides from lactic acid bacteria. Biopolymers 55:50-61. [PubMed]
22. Nikolskaya, A. N., and M. Y. Galperin. 2002. A novel type of conserved DNA-binding domain in the transcriptional regulators of the AlgR/AgrA/LytR family. Nucleic Acids Res. 30:2453-2459. [PMC free article] [PubMed]
23. Perry, D., and H. K. Kuramitsu. 1981. Genetic transformation of Streptococcus mutans. Infect. Immun. 32:1295-1297. [PMC free article] [PubMed]
24. Petersen, F. C., D. Pecharki, and A. A. Scheie. 2004. Biofilm mode of growth of Streptococcus intermedius favored by a competence-stimulating signaling peptide. J. Bacteriol. 186:6327-6331. [PMC free article] [PubMed]
25. Peterson, S., R. T. Cline, H. Tettelin, V. Sharov, and D. A. Morrison. 2000. Gene expression analysis of the Streptococcus pneumoniae competence regulons by use of DNA microarrays. J. Bacteriol. 182:6192-6202. [PMC free article] [PubMed]
26. Podbielski, A., B. Spellerberg, M. Woischnik, B. Pohl, and R. Lütticken. 1996. Novel series of plasmid vectors for gene inactivation and expression analysis in group A streptococci (GAS). Gene 177:137-147. [PubMed]
27. Qi, F., P. Chen, and P. W. Caufield. 2001. The group I strain of Streptococcus mutans, UA140, produces both the lantibiotic mutacin I and a nonlantibiotic bacteriocin, mutacin IV. Appl. Environ. Microbiol. 67:15-21. [PMC free article] [PubMed]
28. Qi, F., J. Merritt, R. Lux, and W. Shi. 2004. Inactivation of the ciaH gene in Streptococcus mutans diminishes mutacin production and competence development, alters sucrose-dependent biofilm formation, and reduces stress tolerance. Infect. Immun. 72:4895-4899. [PMC free article] [PubMed]
29. Rogers, A. H. 1976. Bacteriocinogeny and the properties of some bacteriocins of Streptococcus mutans. Arch. Oral Biol. 21:99-104. [PubMed]
30. Sturme, M. H., M. Kleerebezem, J. Nakayama, A. D. Akkermans, E. E. Vaughan, and W. M. de Vos. 2002. Cell to cell communication by autoinducing peptides in gram-positive bacteria. Antonie Leeuwenhoek 81:233-243. [PubMed]
31. Tao, L., D. J. LeBlanc, and J. J. Ferretti. 1992. Novel streptococcal-integration shuttle vectors for gene cloning and inactivation. Gene 120:105-110. [PubMed]
32. Wang, B.-Y., and H. K. Kuramitsu. 2005. Interactions between oral bacteria: inhibition of Streptococcus mutans bacteriocin production by Streptococcus gordonii. Appl. Environ. Microbiol. 71:354-362. [PMC free article] [PubMed]
33. Ween, O., P. Gaustad, and L. S. Håvarstein. 1999. Identification of DNA binding sites for ComE, a key regulator of natural competence in Streptococcus pneumoniae. Mol. Microbiol. 33:817-827. [PubMed]
34. Whitchurch, C. B., T. Tolker-Nielsen, P. C. Ragas, and J. S. Mattick. 2002. Extracellular DNA required for bacterial biofilm formation. Science 295:1487. [PubMed]
35. Yazgan, A., G. Özcengiz, and M. A. Marahiel. 2001. Tn10 insertional mutations of Bacillus subtilis that block the biosynthesis of bacilysin. Biochim. Biophys. Acta 1518:87-94. [PubMed]
36. Yonezawa, H., and H. K. Kuramitsu. 2005. Genetic analysis of a unique bacteriocin, Smb, produced by Streptococcus mutans GS5. Antimicrob. Agents Chemother. 49:541-548. [PMC free article] [PubMed]
37. Yoshida, A., and H. K. Kuramitsu. 2002. Multiple Streptococcus mutans genes are involved in biofilm formation. Appl. Environ. Microbiol. 68:6283-6291. [PMC free article] [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...