SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The oral pathogen Streptococcus mutans employs a variety of mechanisms to maintain a competitive advantage over many other oral bacteria which occupy the same ecological niche. Production of the bacteriocin, mutacin I, is one such mechanism. However, little is known about the regulatory mechanisms associated with mutacin I production. Previous work has demonstrated that the production of mutacin I greatly increased with cell density. In this study, we found that high cell density also triggered high level mutacin I gene transcription. However, this response was abolished upon deletion of luxS. Further analysis using real-time reverse transcription polymerase chain reaction (RT-PCR) demonstrated that in the luxS mutant transcription of both the mutacin I structural gene mutA and the mutacin I transcriptional activator mutR was impaired. Through microarray analysis, a putative transcription repressor annotated as Smu1274 in the Los Alamos National Laboratory Oral Pathogens Sequence Database was identified, which was strongly induced in the luxS mutant. Characterization of Smu1274, which we referred to as irvA, suggested that it may act as an inducible repressor to suppress mutacin I gene expression. A luxS and irvA double mutant regained the ability to produce mutacin I; whereas a constitutive irvA-producing strain was impaired in mutacin I production. These findings reveal a novel regulatory pathway for mutacin I gene expression, which may provide clues to the regulatory mechanisms of other cellular functions regulated by luxS in S. mutans.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Streptococcus mutans is a Gram-positive, facultative anaerobic oral bacterium, which lives in an extremely diverse and high cell density biofilm on the tooth surface (Hamilton, 2000; Kolenbrander, 2000; Brinig et al., 2003; Munson et al., 2004). The dental biofilm community experiences constant cycles of feast and famine dictated by the food intake and oral hygiene of the host. The high density and great biodiversity of the oral biofilm community coupled with a limited food supply should create an environment conducive to fierce competition among community members. Not surprisingly, many species in the oral biofilm have devised strategies to maximize their ability to compete for these limited resources (Bowden and Hamilton, 1989; Tang-Larsen et al., 1995; Basson, 2000; Quivey et al., 2001; Paramaesvaran et al., 2003; Shah and Russell, 2004).

Streptococcus mutans possesses several abilities that enable it to persist in the oral biofilm community, such as the production of bacteriocins called mutacins (Hillman, 2002). Most clinical isolates of S. mutans produce at least one kind of mutacin (Fabio et al., 1987; Gronroos et al., 1998). Currently, two classes of mutacins have been characterized, the lantibiotics and the non-lantibiotics (Mota-Meira et al., 1997; Chen et al., 1998; 1999; Hillman et al., 1998; Qi et al., 1999a,b; 2000; 2001; Balakrishnan et al., 2000; Yonezawa and Kuramitsu, 2005). Lantibiotics are ribosomally synthesized and undergo extensive post-translational modifications that create the unusual amino acid lanthionine (Lan) or β-methyllanthionine (MeLan) (McAuliffe et al., 2001; Twomey et al., 2002). In contrast, the non-lantibiotics consist of either one or two small unmodified peptides (Balakrishnan et al., 2000; Qi et al., 2001).

In previous studies, we have characterized four mutacins named mutacin I–IV (Novak et al., 1994; Qi et al., 1999a; 2000; 2001). Mutacin I, produced by the strain UA140, belongs to the lantibiotic class with a wide spectrum of activity against large numbers of Gram-positive bacteria including numerous Streptococcus and Staphylococcus species (Qi et al., 1999a; 2000). Mutacin I, like many lantibiotics, is genetically encoded by a large operon of at least 11 genes required for the biosynthesis of the mature peptide (Qi et al., 2000). Very little is known about the regulation of the mutacin I operon. However, it was previously shown that both the mutacin I-specific transcriptional activator mutR (Qi et al., 1999b; Kreth et al., 2004) and the two-component sensor ciaH (Qi et al., 2004) were required for expression. Additionally, previous mutacin I isolation experiments demonstrated that the production of mutacin I was virtually undetectable in planktonic culture, but easily isolated from cells grown on solid surfaces (Qi et al., 2000).

Based on this observation, it was hypothesized that mutacin I production was likely controlled by cell density and therefore a quorum sensing or ‘quorum sensing-like’ system may be involved in regulating the expression of mutacin I.

Currently, there are two recognized density-dependent signalling systems in S. mutans. The first is the intraspecies quorum sensing system encoded by comCDE, which controls natural competence, biofilm formation (Li et al., 2002), and was also recently found to regulate the production of a lantibiotic mutacin called Smb (Yonezawa and Kuramitsu, 2005). The other is the interspecies signalling system mediated by the gene luxS (Merritt et al., 2003). This system involves the production of autoinducer-2 (AI-2) signal molecules in response to cell density and has so far been detected in a large number of Gram-positive and Gram-negative bacteria (McNab and Lamont, 2003; Xavier and Bassler, 2003). It has been demonstrated in numerous studies that the AI-2 molecules produced by luxS-positive bacteria all have the capacity to induce the AI-2 reporter species Vibrio harveyi (Xavier and Bassler, 2003). This raises the possibility that AI-2 is a universally recognized signal, which may be used to mediate interspecies interactions in a multispecies biofilm such as the dental plaque (Kolenbrander et al., 2002). Previous studies of the S. mutans luxS gene from our laboratory and others have demonstrated that luxS is required for a variety of cellular processes ranging from biofilm formation to acid tolerance (Merritt et al., 2003; Wen and Burne, 2004).

In this study, we tested the involvement of the comCDE and luxS systems in mutacin I production. While comCDE had no effect on mutacin I production, the luxS null mutation abolished mutacin I production. Further characterization revealed that an inducible repressor may serve as a mediator between the luxS signalling system and mutacin I gene expression.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Mutacin I gene expression is induced at high cell density

As it was previously observed that mutacin I production was stimulated by growth on solid surfaces (Qi et al., 2001), we examined whether mutacin I gene expression would be induced by growth at a high cell density. A green fluorescent protein (gfp) reporter fusion to the mutA promoter was constructed and introduced into the mutacin I-producing strain UA140. The mutA–gfp reporter strain was first grown to early log phase (OD600∼0.1) under planktonic conditions and subsequently half of the culture was briefly centrifuged. After 2 h of incubation, cells in both cultures were analysed for fluorescence. As a control, we included an isogenic ldh–gfp transcriptional fusion strain grown under the same conditions. Previous work in our laboratory had demonstrated the ldh (lactate dehydrogenase) gene to be a highly expressed constitutive gene (Merritt et al., 2005) and therefore, the ldh–gfp reporter was not expected to be influenced by the growth conditions of our assay. As shown in Fig. 1A and B, there is a tremendous discrepancy in the level of GFP fluorescence in the mutA–gfp reporter strain grown under planktonic and pelleted conditions. Cells from the planktonic culture exhibited virtually no fluorescence (Fig. 1A), while cells incubated in the pelleted form displayed bright fluorescence (Fig. 1B). These results are in sharp contrast with that of the ldh–gfp reporter strain which showed similar strong fluorescence under both conditions (Fig. 1C and D). We also tested planktonic cells for mutA transcription over the growth curve using a luciferase reporter and found the highest expression during the log/stationary-phase transition. However, total reporter activity was much lower than cells incubated after centrifugation (data not shown). This suggested that the mutA promoter was likely activated as a result of growth at a high cell density.

image

Figure 1. The mutA–gfp reporter responds to high cell density. The mutA–gfp reporter strain (A and B) and the ldh–gfp reporter strain (C and D) were grown to OD600 0.05 and then split into two samples: one sample remained in planktonic phase (A and C), while the other was pelleted (B and D) by centrifugation. Samples were incubated for 2 h before fluorescence microscopy. Image magnification: 640×.

Download figure to PowerPoint

LuxS is required for density-dependent mutacin I expression

Based on the apparent induction of the mutacin I operon expression at a high cell density, we hypothesized that mutacin I gene expression could be controlled by a density-dependent cell signalling system. To test this, we transferred the comD and luxS null mutations to the mutacin I-producing strain UA140 and assayed for mutacin I production by the deferred antagonism assay (plate assay) as described in Experimental procedures. While the comD null mutation did not inhibit mutacin I production (data not shown), the luxS null mutation completely abolished mutacin I production as shown by the absence of a zone of inhibition around the colonies (Fig. 2). This result suggested that the luxS gene was required for cell density-dependent mutacin I production. As UA140 is known to produce the non-lantibiotic mutacin IV as well (Qi et al., 2001), we also tested whether the luxS mutation affected mutacin IV production. The same plate assay was performed using Streptococcus sanguinis NY101 as the indicator, which was previously demonstrated to be sensitive to both mutacin I and mutacin IV (Qi et al., 2001). We found that mutacin IV production was not affected by the luxS mutation (data not shown). Taken together, these results suggested that luxS was involved in regulating the production of mutacin I, but was not required for the production of mutacin IV.

image

Figure 2. A mutation in luxS abolishes mutacin I production. Wild type UA140 and the luxS deletion strain were spotted onto a TH agar plate from an overnight culture. After incubation for 2 days, the plate was overlaid with the indicator strain, OMZ176, and incubated for an additional 16 h. The three spots on the top of the plate are of the wild type UA140 and the three lower spots are the luxS mutant. Clearing zone indicates mutacin I production.

Download figure to PowerPoint

LuxS affects the transcription of mutA and mutR

To determine whether luxS affects mutacin I production at the transcriptional level, we used real-time reverse transcription polymerase chain reaction (RT-PCR) to compare the transcript levels of mutA in the wild type and the luxS mutant strains. Total RNA was extracted from cells grown on agar plates for 2 days. Under these conditions, mutacin I was readily detected in the wild type but not in the luxS mutant strain by the overlay assay (Fig. 2). As shown in Fig. 3A, mutA gene expression was downregulated more than 400-fold in the luxS mutant compared with the wild type. In contrast, transcription of the mutacin IV structural gene nlmA was not affected by the luxS mutation (data not shown). These results indicated that the effect of a luxS mutation on mutacin I production was largely at the transcriptional level.

image

Figure 3. Effects of luxS mutation on mutA (A) and mutR (B) gene transcription. Wild-type UA140 and the luxS mutant were grown on agar plates for 2 days as described in Experimental procedures. The RNA was extracted and quantified by real-time RT-PCR for mutA and mutR gene expression. The quantity of mutA and mutR cDNA measured by real-time PCR was normalized to the 16S cDNA abundance within each unique reaction. Each experiment was repeated three times with triplicate samples. Shown here are data from a representative experiment. Error bars indicate the variance between triplicate samples within the real-time PCR reaction. The relative cDNA abundance of the wild-type samples was arbitrarily assigned as 1.

Download figure to PowerPoint

Previous studies have indicated that mutA gene expression is dependent on a specific positive regulator, mutR, which is located upstream of the mutA operon (Qi et al., 2000). To further investigate the regulatory mechanism of luxS on mutacin I gene expression, we tested whether the luxS effect on mutacin I gene expression was mediated through MutR. Real-time RT-PCR was used to quantify mutR RNA extracted from the wild type and luxS mutant cells grown under the same conditions as described above. As shown in Fig. 3B, mutR expression was downregulated 10-fold in the luxS mutant as compared with the wild-type strain. This suggested that the mutacin-negative phenotype of the luxS mutant was manifested at least partially through reduced transcription of mutR.

Screening for intermediate regulators in the luxS and mutacin I pathway

Results presented above suggested the involvement of the luxS system in the regulation of mutacin I gene transcription. To search for genes that might potentially mediate the luxS regulation of mutA or mutR gene expression, we employed microarray analysis. Cells were collected from both the wild type and the luxS mutant grown on plates and their RNA was extracted. Microarray was performed three times with three independently prepared RNA samples. Genes whose expression was altered consistently in all three experiments were selected. Overall, nine genes were found to be differentially expressed twofold or greater in the luxS mutant (Table 1). As luxS regulation on mutacin production was at the transcriptional level, the conserved hypothetical gene Smu1274 was chosen for further analysis because it was annotated as a transcription repressor in the Los Alamos National Laboratory Oral Pathogens Sequence Database.

Table 1. Microarray analysis of genes affected by the luxS mutation.
Gene name/IDFold change in mutantFunction/class
Smu843 2×Membrane protein; DUF1304
Smu1131 2×Putative dihydrooratase family protein
Smu127412×Putative repressor
gbpC  2.5×Exopolysaccharide binding
pstS  3.5×Phosphate-binding protein
punA  5.5×Purine nucleoside phosphorylase
rgpBc  0.5×Rhamnosyltransferase
spaP  0.33×Adhesin
sodA  2.3×Oxidative stress protection

Smu1274 is a small protein (79 amino acids) with no characterized function. However, it has putative assignment as a transcriptional repressor based on its homology to several phage repressors in various Gram-positive bacterial species. The gene is immediately upstream of the glucan-binding protein C gene (gbpC) (Sato et al., 2002) and also shares an intergenic region with an upstream hypothetical protein (Smu1275) (Fig. 4A). We confirmed the induction of this gene by real-time RT-PCR on the same RNA samples used for microarray studies. As shown in Fig. 4B, the expression of Smu1274 was increased by greater than 300-fold in the luxS mutant, which validated the upregulation detected during microarray screening.

image

Figure 4. Analysis of Smu1274. A. The genetic organization of the Smu1274 locus. Gene names or gene IDs are indicated as well as the direction of transcription. The shaded areas represent intergenic regions. B. Smu1274 gene expression in the wild type and luxS mutant as analysed by real-time RT-PCR. Relative cDNA abundance and error bars are defined as described in the legend of Fig. 3.

Download figure to PowerPoint

Deletion of Smu1274 altered mutacin I production in the wild type and alleviated the mutacin defect in the luxS mutant

It was apparent from the RT-PCR data that luxS was able to affect the transcription of Smu1274 and that increased expression of Smu1274 in the luxS mutant correlated with decreased mutacin I transcription. However, it was as yet unclear whether this association was coincidental or Smu1274 was epistatic to luxS in the regulatory pathway leading to mutacin transcription. To address this issue, we performed mutational analysis of Smu1274 to determine whether deletion of Smu1274 would affect mutacin I gene expression. An allelic replacement mutant of Smu1274 was constructed (see Experimental procedures) and found to have no obvious growth defects (data not shown). This strain was then tested for mutacin I production with a plate assay. As shown in Fig. 5A, the ΔSmu1274 strain was able to produce more mutacin I than the wild type after 16 h of growth on brain–heart infusion (BHI) agar plates. These results suggested that the ΔSmu1274 mutation could affect the production of mutacin I.

image

Figure 5. Effects of Smu1274 (irvA) mutations on mutacin I production. A. The wild type (shown on the left) and the Smu1274 (irvA) deletion mutant (shown on the right) were inoculated onto a BHI plate and incubated for 24 h before overlay with the indicator strain OMZ176. B. The wild type (shown on the left) and the Smu1274 (irvA) and luxS double mutant (shown on the right) were inoculated onto a TH plate and incubated for 48 h before overlay with OMZ176.

Download figure to PowerPoint

To further explore the role of ΔSmu1274 in mutacin I production, we constructed a luxS and Smu1274 double deletion strain. If the luxS null mutation exerted its effect on mutacin I transcription through induction of Smu1274, then deletion of Smu1274 in the luxS mutant background should alleviate the negative effects of the luxS mutation on mutacin I gene expression. This double mutant strain was tested for mutacin I production by a plate assay. As shown in Fig. 5B, this strain indeed regained the ability to produce mutacin I. These results suggested that Smu1274 was negatively controlled by luxS and possibly served as a repressor for mutacin I gene expression in the pathway downstream of luxS. Based on its proposed role as a repressor of mutacin I transcription, we named this gene ‘inducible repressor of virulence’ or irvA.

Constitutive expression of irvA suppressed mutacin I gene expression

Analysis of irvA expression demonstrated very low expression in the wild type (Fig. 4B). This suggested that this gene was poorly transcribed under normal growth conditions. Based on our previous data, it was predicted that artificially increasing the level of irvA expression would result in suppressed mutacin I production. To test this hypothesis, we created a constitutive irvA-expressing mutant by putting this gene under the control of the ldh promoter. The ldh promoter has been shown to be highly expressed in S. mutans (Merritt et al., 2005) therefore the ldh–irvA construct was expected to yield high levels of irvA transcript. As show in Fig. 6A, real-time RT-PCR quantification demonstrated an over 400-fold increase in irvA transcription in the constitutive mutant strain compared with the wild-type strain. To test whether the increase in irvA gene expression would result in suppression of mutacin I production, we performed a plate assay. As shown in Fig. 6B, the constitutive irvA mutant strain was impaired in mutacin I production. Interestingly, this effect was only noticeable during the first 24 h of growth. After 2 days, the irvA constitutive strain was able to regain the ability to produce mutacin I and had only a slightly smaller halo size than the wild type (data not shown). To rule out the possibility of a growth defect accounting for the difference in mutacin I production, we measured the growth rates of the wild type and the irvA constitutive strain and found no evidence for slower growth in the mutant (data not shown). Taken together, these results suggested that high levels of irvA transcript were associated with decreased mutacin I production, even in the presence of an intact luxS gene, and that additional regulatory elements may be involved in alleviating the suppressive effect of irvA at later growth stages.

image

Figure 6. Effects of constitutively expressed irvA (ldh-1274) on mutacin production and mutacin gene expression. A. Quantification of the irvA transcript in the wild type and the ldh-1274 gene fusion strain. B. Mutacin I production in the wild type (shown on the left) and the irvA constitutive expression (ldh-1274) mutant (shown on the right). Cells were inoculated onto a BHI plate and incubated for 16 h before overlay with the indicator strain OMZ176. C. The expression of mutA in the wild type and the irvA constitutive expression strain as analysed by real-time RT-PCR. Relative cDNA abundance and error bars are defined as described in the legend of Fig. 3. D. The expression of mutR in the wild type and the irvA constitutive expression strain as analysed by real-time RT-PCR.

Download figure to PowerPoint

To further determine whether irvA repression on mutacin production was at the transcriptional level, we used real-time RT-PCR to quantify the mutA and mutR transcripts in cells grown on agar plates for 24 h. As shown in Fig. 6C and D, mutA transcript level was reduced > 35-fold in the irvA constitutive expression strain and mutR gene expression was reduced approximately 10-fold. These results were consistent with our proposal that irvA acts as a transcriptional repressor and that its function is likely epistatic to luxS in the pathway leading to mutA and mutR transcription.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

In this article we describe the effects of a luxS deletion upon the production of mutacin I. Mutacin I transcription was shown to be responsive to high cell density in the wild type, but this response was completely abolished in the luxS mutant. The luxS mutation caused a reduction in transcription of both the mutacin I-specific regulator, mutR, and the mutacin operon itself. In addition, we provided evidence to suggest that the phenotype of the luxS mutant was relatively specific for mutacin I rather than a general repression of all bacteriocin production. Further analysis using microarray identified an uncharacterized putative repressor, irvA (Smu1274), as strongly induced in the luxS mutant strain. This result was verified by real-time RT-PCR. Deletion of irvA resulted in a conditional increase in mutacin production during early stages of growth, whereas a double mutation of both luxS and irvA was able to rescue mutacin I production under all growth conditions. Additionally, a constitutive irvA expressing mutant strain was impaired in mutacin I production at the transcriptional level.

Based on these data, we propose the following model for the possible mechanism of luxS regulation on mutacin I gene expression (Fig. 7). LuxS normally synthesizes AI-2 from its substrate, S-ribosylhomocysteine (Schauder et al., 2001), which is then secreted to the outside of the cell. In the presence of a functional LuxS, transcription of irvA is repressed. However, when luxS is deleted or the cell encounters environmental stress, irvA gene expression is dramatically increased. High levels of irvA lead to a repression of mutA and mutR transcription, which results in impaired mutacin I production. It should also be noted that while this model explains the experimental observations made in this study, the exact mechanisms of how luxS regulates irvA and how irvA represses mutA and mutR transcription await further experimentation.

image

Figure 7. Proposed model for luxS control over mutacin I expression. The pathways printed in bold have been tested experimentally and those in normal print are hypothetical.

Download figure to PowerPoint

Since 1999, there have been numerous studies with a wide range of bacterial species that suggest AI-2 as being an universal signal mediating interspecies interactions (Xavier and Bassler, 2003). Such a signalling system may be of particular importance in the oral biofilm community, as interspecies interactions are necessary for different species to develop into a structured biofilm community (Kolenbrander, 1995; 2000; Kolenbrander et al., 2002). For example, the periodontal pathogen Porphyromonas gingivalis generally does not form robust biofilms as a monospecies culture, but forms confluent biofilms in mixed cultures with Streptococcus gordonii (Cook et al., 1998). This ability was further demonstrated to be dependent on the luxS gene (McNab et al., 2003). Similarly, interspecies signalling could also play a pivotal role in interspecies competitive behaviours as well. The lactic acid bacteria, for example, are known to produce a myriad of different bacteriocins as a means to eliminate competition from other Gram-positive species (Eijsink et al., 2002). In an environment such as the dental plaque, energy expenditures may be better utilized by producing bacteriocins when the total population density is high and competitors are likely to be present. This may be particularly true for mutacin I production, which requires at least 11 gene products to make a mature peptide. It is also interesting to note that of the two mutacins produced in strain UA140, the wide spectrum bacteriocin, mutacin I, is regulated by the interspecies signalling system luxS, while the narrow spectrum bacteriocin, mutacin IV, is regulated by the intraspecies signalling system comCDE (Kreth et al., 2005).

It is also worth noting that gbpC, which is adjacent to irvA on the chromosome, was also induced in the luxS mutant in our microarray assay (Table 1). This induction was further confirmed by real-time RT-PCR (data not shown). GbpC is suggested to be a stress-induced glucan-binding protein (Sato et al., 2000; 2002). This could lead to an interesting alternative explanation for the effect of the luxS mutation on mutacin I production, i.e. the luxS mutation could have created a stress signal, which caused repression of mutacin I production. This would also seem to be consistent with a study by Wen and Burne (2004) in which they proposed a function for luxS in regulating the S. mutans acid and oxygen stress responses.

While the mechanism of luxS control over irvA and subsequent mutacin I gene expression awaits further characterization, genomic analysis of irvA may give some clues as to its function. A blastp search revealed that residues 19–77 showed 67% identity, 78% similarity to a phage-associated repressor in Streptococcus pyogenes (Beres et al., 2002), and residues 17–77 were 57% identical and 73% similar to the Cro protein of bacteriophage EJ-1 (Diaz et al., 1992). IrvA also shows significant homology to the DNA-binding domains of several other Streptococcus and Lactococcus phage repressors. A pfam search placed residues 22–76 with a family of transcriptional regulators referred to as the HTH-3 family, which all share a similar helix–turn–helix domain (PF01381). In contrast, the first 20 amino acids of IrvA, which are outside of the HTH-3 domain, have no detectable homology in any other organisms. Interestingly, the genomic organization and sequences of both irvA and the upstream putative repressor, Smu1275, bear a striking similarity to the CI/Cro system of many bacteriophages such as λ (Johnson et al., 1981). Bacteriophages use this system of repression and antirepression to regulate the switch between lysogeny and lytic replication in response to stress (Ptashne et al., 1980). A similar system was also found in Bacillus subtilis, which regulates the cell's entry into sporulation via the SinR repressor (Lewis et al., 1998). Therefore, it is tempting to speculate that irvA (and perhaps Smu1275) may also be part of a larger adaptive response, which can affect mutacin I gene expression. Future experiments will be aimed at deciphering the regulatory pathways between luxS, irvA and mutacin gene expression.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains and culture conditions

Bacterial strains and their relevant characteristics are listed in Table 2. S. mutans strains were grown in either BHI or Todd–Hewitt (TH) medium (Difco). For the selection of antibiotic-resistant colonies, TH plates were supplemented with either 15 µg ml−1 erythromycin (Sigma) or 800 µg ml−1 spectinomycin (Sigma). All S. mutans strains were grown anaerobically (80% N2, 10% CO2 and 10% H2) at 37°C. Escherichia coli cells were grown in Luria–Bertani (LB; Fisher) medium with aeration at 37°C. E. coli strains carrying plasmids were grown in LB medium containing 250 µg ml−1 erythromycin or 250 µg ml−1 spectinomycin.

Table 2. Bacterial strains used in this study.
StrainRelevant characteristicsReference
E. coli DH5 α supE44 lacU169 (80 lacZ M15)hsdR17recA1 endA1 gyrA96 thi-1 relA1 luxS Hanahan (1983); Surette et al. (1999)
S. mutans UA140Wild-type S. mutans Kans Erms Qi et al. (2001)
S. sobrinus OMZ176Wild-type S. sobrinus mutacin Is Caufield et al. (1990)
140-luxUA140 ΔluxS ErmrThis work
140-1274UA140 ΔSmu1274 SpcrThis work
ldh-1274UA140::Φ(ldhp-Smu1274) SpcrThis work
JK1gfpUA140::Φ(mutAp–gfp) Spcr Kreth et al. (2004)
JK2gfpUA140::Φ(ldhp–gfp) SpcrThis Work

Visualization of mutA–gfp gene expression

To visualize mutA expression under high cell density, we performed a cell pelleting assay using the mutA reporter strain JK1gfp and the control strain JK2gfp (Table 2). Cells of an overnight culture were freshly inoculated into BHI medium to an OD600∼0.05. After two doubling times in anaerobic conditions at 37°C, the culture was divided into 1 ml aliquots in microfuge tubes. Cells either were pelleted by centrifugation at 13 200 r.p.m. for 1.5 min in a tabletop centrifuge (Eppendorf) or were left as dispersed (planktonic cells). The two sets of samples were further incubated in the medium for 2 h anaerobically. After the incubation period, both planktonic and pelleted cells were twice washed in PBS buffer and then left for 10 min at room temperature to aerate the cells for proper folding of the GFP fluorophore. GFP fluorescence was visualized using fluorescent microscopy (Nikon ECLIPSE E400 microscope) with appropriate filters.

Plate assay for mutacin production

To assay for mutacin I production, UA140 wild type and mutant cells were first grown overnight in liquid cultures under standard anaerobic conditions. Five microlitres of the overnight culture were spotted onto TH or BHI plates and incubated anaerobically at 37°C for various periods of time depending on the assay. For measuring the luxS effect, TH plate was used and the cells were incubated for 48 h. For detecting the effect of irvA deletion, BHI plate and 24 h incubation were found to give consistent results. For testing the effect of a constitutive irvA expression mutant, BHI plate and 16 h incubation showed the most dramatic effect. After the incubation period, an overnight culture of the mutacin I indicator strain S. sobrinus strain OMZ176 was used to overlay the test plate. After an overnight incubation at 37°C anaerobically, the zone of inhibition (halo) was assessed.

Construction of reporters and mutants

The luxS mutant strain was generated previously (Merritt et al., 2003) and this mutation was transferred to strain UA140 via transformation of gDNA. The deletion of Smu1274 was created via a double cross-over homologous recombination. To generate the construct, two fragments corresponding to approximately 1 kb of upstream and downstream sequence of the target gene were generated by polymerase chain reaction (PCR) using Pfu polymerase and the primer pairs 1274upF (5′-TTGTCCGTGATGGAT AGCTTT-3′)/1274upR (5′-CATCCCCTTTCAACTTACCA-3′), and 1274downF (5′-TGAAATTCCCATTCAACTGGC-3′)/1274downR (5′-GATCATATGGATTTGCTTGGG-3′). These fragments were subsequently treated with Taq polymerase to generate 5′ overhangs and cloned into pCR2.1 (Invitrogen). All fragments were digested with appropriate restriction enzymes, gel purified, and ligated with the spectinomycin resistance cassette and the cloning plasmid pBlueScript (Stratagene) in a four-piece ligation reaction. The resulting plasmid was confirmed via restriction digestion and PCR and linearized for transformation into S. mutans. The Smu1274 constitutively expressing strain was created by removing the luciferase (luc) gene from the plasmid pldh-luc (Merritt et al., 2005) and replacing it with the coding sequence of Smu1274. The coding sequence for Smu1274 was generated by PCR using the primers 1274 orfF (5′-CGGGATCCATGAATAATACTGCTATTTTTAGG-3′) and 1274 orfR (5′-CGAAGCTTAACCAAAGTGCTTTCTCT TATC-3′) and Pfu/Taq polymerase and then cloned into pCR2.1. This fragment was then digested with appropriate restriction enzymes, gel purified and ligated into compatible sites on pldh-luc. The resulting plasmid (pldh-1274) was transformed into UA140 and integrated into the chromosome via single cross-over homologous recombination. Construction of the mutA–gfp (JK1gfp) strain has been described previously (Kreth et al., 2004). Construction of the ldh–gfp reporter strain was created by removing the 1 kb ldh promoter fragment from the vector pldh-luc (Merritt et al., 2005) with SpeI and XhoI and subsequently ligating to compatible sites on the vector pFW5::gfp (Kreth et al., 2004) to generate the suicide vector pFW5::Φ(ldh–gfp). This plasmid was then transformed into UA140 and integrated into the chromosome via homologous recombination.

RNA extraction and real-time RT-PCR

UA140 and derivative strains were grown on TH plate under the same conditions as for the plate assay. Cells were harvested on ice by scraping the plates using a sterile dacron swab and several millilitres of PBS, pelleted and stored at −80°C. Multiple plates were used for each sample to generate sufficient amounts of cells. One plate was saved for each sample and overlaid with the indicator strain OMZ176 to confirm the phenotype. For RNA extraction, frozen cell pellets were resuspended in 700 µl of tris-EDTA buffer + 200 µl of 10% SDS + 1.5 ml of hot acidic phenol (65°C, pH 4.3) (Fisher). The solution was then incubated at 65°C for 15 min with frequent mixing. After incubation, the solution was centrifuged and the supernatant was saved for extraction. The supernatant was twice extracted with 2–3 volumes of TRIzol® (Invitrogen) + 0.2 volume of chloroform. RNA was precipitated with 1/10 volume of 3 M sodium acetate + 2.5 volumes of 100% ethanol. Total RNA (3 µg) was used for cDNA synthesis using Stratascript RT (Stratagene) according to the manufacturer's protocol. For real-time RT-PCR, primers were designed according to sequence data provided by the Los Alamos National Laboratory Oral Pathogens Sequence Database (http://www.oralgen.lanl.gov/oralgen/bacteria/smut/) and SYBR green (Bio-Rad) was used for fluorescence detection with the iCycler™ (Bio-Rad) real-time PCR system according to manufacturer's protocol. Real-time RT-PCR primer sequences are as follows: mutR rtF (5′-AAACCATGT TGGCAGCTGAT-3′), mutR rtR (5′-CCTCATAATGGCT CAAAGCA-3′), mutA rtF (5′-CTATTGTGGCAAGCAACG AC-3′), mutA rtR (5′-AAACTAGGATTTTTCACCCCTGT-3′), 1274 rtF (5′-TGGTAAGTTGAAAGGGGATG-3′), 1274 rtR (5′-ACAATCAGCCACCTCTTTGG-3′), 16s F (5′-GATAATTGAT TGAAAGATGCAAGC-3′) and 16s R (5′-ATTCCCTACTGCT GCCTCCC-3′). Total cDNA abundance between test samples was normalized using the 16s rRNA gene as a housekeeping control.

Microarray

Microarray slides were prepared by The Institute for Genomic Research (TIGR) and consisted of 70-mer oligonucleotides representing 1960 open reading frames. RNA was prepared as described above and 3 µg of RNA was used for cDNA synthesis. Cy dye incorporation and hybridization procedures were performed similarly as described on the TIGR website (http://pfgrc.tigr.org/protocols/M007.pdf and http://pfgrc.tigr.org/protocols/M008.pdf). Image analysis was performed using GenePix® Pro 6.0 software.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We greatly appreciate the Los Alamos National Laboratories Oral Pathogens Sequence Database website for useful annotation of the S. mutans genome. We also would like to acknowledge TIGR for production of the S. mutans microarray slides. This work was supported in part by NIH Grant R01 DE 014757 to F. Qi, an NIDCR T32-DE007296 and T32-DE007296-08 Training grant to J. Merritt, an U01-DE15018, a BioStar/C3 Scientific Corporation grant and a Washington Dental Service grant to W. Shi.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References