In the last two decades an increasing number of local outbreaks of invasive group A streptococcus (GAS) infections including necrotizing fasciitis (NF) have been reported. We identified the streptococcal invasion locus (sil) which is essential for virulence of the M14 strain JS95 isolated from an NF patient. This locus contains six genes: silA/B and silD/E encoding two-component system (TCS) and ABC transporter, respectively, homologous to the corresponding entities in the regulon of Streptococcus pneumoniae involved in genetic competence. Situated between these two units are silC and silCR, which highly overlap and are transcribed from the complementing strand at opposite directions. SilCR is a putative competence stimulating peptide, but in the M14 strain it has a start codon mutation. Deletion of silC or addition of synthetic SilCR attenuates virulence of the M14 strain. Here we found that silC and silCR form a novel regulatory circuit that controls the sil locus transcription. Under non-inducing conditions silC represses the silCR promoter. Externally added SilCR peptide activates the TCS, which in turn stimulates silCR transcription. Ongoing silCR transcription mediates the repression of the converging and overlapping silC transcript. Transcription of bacteriocin-like peptide (blp) operon mirrors the inverse relationships between the silC and silCR transcripts. It is upregulated by either addition of SilCR or deletion of silC. Moreover, expression of silC from a plasmid in a silC deleted-mutant significantly represses blp transcription. Finally, we show that 18% of clinically relevant GAS isolates possess sil and produce SilCR. Based on these results we propose a working model for regulation gene expression and virulence in GAS by the SilCR signalling peptide.
Group A streptococcus (GAS) exclusively colonizes humans giving rise to a wide spectrum of diseases ranging from uncomplicated infections to life-threatening invasive diseases, such as necrotizing fasciitis (NF) (Bisno et al., 2003). A large number of GAS virulence factors are involved in host–pathogen interactions and facilitate GAS colonization, immune evasion and bacterial dissemination (Banks et al., 2003; Kreikemeyer et al., 2003; Musser and DeLeo, 2005). Two-component systems (TCSs) and stand-alone transcriptional regulators control the expression of these virulence factors in response to environmental cues. Complete genome sequences are now available for 12 GAS strains corresponding to nine serotypes. Analysis of 12 GAS genomes (Beres et al., 2006) revealed the presence of 13 conserved TCSs, seven of which have been implicated in host–pathogen interaction (Kreikemeyer et al., 2003; Nakata et al., 2005; Sitkiewicz and Musser, 2006). Except for CovR/S, which plays a central role in the switch between the pharyngeal and the invasive transcriptomes (Sumby et al., 2006), much less is known on the in vivo mechanism of action of the other TCSs.
In vivo screening of a transposon-tagged mutant library from an M14 GAS strain JS95 isolated from an NF patient identified the streptococcal invasion locus (sil) as essential for GAS rapid dissemination in a murine model of NF (Hidalgo-Grass et al., 2002). sil shares homology with the quorum-sensing competence (com) and bacteriocin-like peptide (blp) regulons of Streptococcus pneumoniae (Morrison and Lee, 2000; de saizieu et al., 2000). Quorum-sensing regulons enable S. pneumoniae and several other streptococcal species to interchange their genetic material, develop tolerance to acid and form biofilm (Cvitkovitch et al., 2003; Kreth et al., 2005). These complex processes are primarily initiated by a 17-residue unmodified peptide termed CSP –competence-stimulating peptide. CSP is matured and secreted by a dedicated ATP-binding cassette (ABC) transporter which removes a leader peptide after a Gly–Gly motif (double-glycine leader) of the pre-CSP (Pestova et al., 1996). In S. pneumoniae CSP activates the expression of both the ABC transporter and the TCS, forming an autocatalytic circuit (Morrison and Lee, 2000; Claverys and Havarstein, 2002). Once the response regulator is activated, it then activates the transcription of several genes, including comX. ComX acts as an alternative sigma factor recognizing a specific consensus sequence, termed com-box (or cin-box) in the promoter regions of competence-specific operons (Morrison and Lee, 2000; Opdyke et al., 2003).
sil is composed of the TCS SilA/B and an ABC transporter SilD/E (Fig. 1). Situated between these entities is a unique small open reading frame (ORF) silC. silCR overlaps silC on the complementary strand by more than 70%. It encodes a predicted CSP (Hidalgo-Grass et al., 2002; Smoot et al., 2002) consisting of a mature peptide of 17-residues (SilCR). In the wild-type (WT) M14 JS95 strain, a missense mutation changes the AUG start codon of silCR to AUA, suggesting that SilCR is not translated (Hidalgo-Grass et al., 2002).
The location of the transposon in the original sil attenuated mutant (JS95::pttm112) was in silC, disrupting both silC and silCR (Fig. 1) (Hidalgo-Grass et al., 2002). A deletion mutant lacking silC (ΔsilC) exhibited reduced virulence (Hidalgo-Grass et al., 2002). As SilCR is presumably not produced in the M14 type strain due to the start codon mutation, it implies that silC promotes virulence. Interestingly, deletion of silC in an M18 type strain reduced its ability to form biofilm (Lembke et al., 2006). While silC might promote virulence, synthetic SilCR obliterates virulence when coinjected into mice with the M14 JS95 strain (Hidalgo-Grass et al., 2004).
To start unravelling the mechanism by which silC and SilCR inversely affect virulence, we studied the transcriptional regulation of the sil locus. We found that silC and silCR form a novel regulatory circuit that is controlled by the SilCR signalling peptide. Furthermore, by polymerase chain reaction (PCR) analysis and immuno-dot-blot assay we showed that 18% of clinically relevant GAS strains possess sil and produce SilCR. Based on these results we propose a working model for gene and virulence regulation by the SilCR peptide.
The transcripts composing the sil locus
To identify the transcripts composing the sil locus, we employed reverse transcriptase PCR (RT-PCR) using primers pairs (Table 1) whose location on sil is shown in Fig. S1. The RT-PCR analysis shown in Fig. 1A and B demonstrates that silA and silB are situated on a single transcript which does not extend into silC. PCR products of similar size were obtained using primers amplifying a DNA segment stretching from silA to silB (Fig. S1 and Table 1, primers RT-silA-f and RT-silB-r) when cDNA or chromosomal DNA of the WT strain were used as templates. In contrast, chromosomal DNA, but not cDNA, produced a PCR product when primers amplifying a DNA segment stretching from silA to silC were used (Fig. S1 and Table 1, primers RT-silA-f and RT-silC-r). These results suggest that silA and silB are transcribed as an operon starting at the P1 putative promoter and terminating before the start of silC transcription (Fig. 1). Consistent with this notion is the presence of 69 bp DNA segment composed of two inverted repeats located between silB and silC, which could form a stable secondary structure [as predicted by the M-fold program of Zuker (Zuker, 2003)] thus acting as a transcriptional terminator (Fig. 1, TT and Fig. S3).
Table 1. Primers.
Construction (C) and verification (V)
RNase protection assay
The existence of an mRNA transcript encompassing silE, silD and silCR is evident from the results shown in Fig. 1C. When cDNA or chromosomal DNA of the WT were used as templates, PCR products of similar size were obtained using PCR primers amplifying a segment stretching from silE to silCR (Fig. S1 and Table 1, primers RT-silE-f and RT-silCR-r). In contrast, only chromosomal DNA but not cDNA, generated a PCR product, when primers amplifying a segment stretching from silE to downstream of the TT were used (Fig. S1, and Table 1, primers RT-silE-f and RT-TT-r). These results suggest that silE, silD and silCR are transcribed as an operon starting at the putative P3 promoter and ending before the TT (Fig. 1). The existence of a single transcript containing silE, silD and silCR was further supported by RT-RT-PCR analyses quantifying the amount of silE and silCR transcripts formed by the silD¯ mutant (disrupted by insertion inactivation), compared with that formed by the WT strain. These experiments were performed in the presence of SilCR, which considerably stimulates the transcription of silE/D/CR from the putative P3 promoter (see below). The amount of silE transcript formed by the silD¯ mutant was much higher than that of silCR (Fig. S2). In contrast, the relative amounts of these transcripts in the WT strain were similar, indicating that the interruption of silD by insertion inactivation is polar, impeding the formation of the silCR transcript.
To map the transcription start sites (TSS) of the transcripts identified by RT-PCR (Fig. 1), we performed primer-extension analyses. End-labelled primers (Table 1, PE-silA-r, PE-silC-r and PE-silE-r), were utilized by RT to synthesize cDNA molecule which were analysed on a denaturing polyacrylamide gels. The analysis of silA/B transcript showed that its TSS is 61 bp upstream to silA start codon whereas that of silE/D/CR is 20 bp upstream to the silE start codon, as illustrated in Fig. S3. The primer extension analysis of silC demonstrated that it is transcribed separately of silA/B. silC TSS is 53 bp upstream to the silC start codon and 24 bp downstream to the TT end (Fig. S3). silC TSS is located 24 bp upstream to the putative silC com-box (or cin-box) promoter (Hidalgo-Grass et al., 2002). The exact localization of silC TSS excludes the possibility that the putative com-box promoter drives silC transcription (Fig. S3).
SilCR induces the transcription of silE/D/CR via the TCS SilA/B
To determine whether sil locus is regulated like com (Morrison and Lee, 2000; Luo and Morrison, 2003), we measured silE/D/CR, silA/B and sigX[the homologue of comX in GAS (Opdyke et al., 2003)] transcripts in response to SilCR addition by RT-RT-PCR. Total RNA was isolated from cultures of the WT strain grown in the absence or presence of SilCR (10 μg ml−1) to either early or late logarithmic phases or to an early stationary phase as described in Experimental procedures. Upon addition of SilCR, the transcription of the silE/D/CR operon was highly induced (35- to 60-folds) as determined by quantifying silD (Fig. 2A) or silE (Fig. 2B) transcripts at either early or late logarithmic phases. No change in silE/D/CR transcription in response to SilCR was detected at early stationary phase. This suggests that like in S. pneumoniae com regulon (Morrison and Lee, 2000; Luo and Morrison, 2003; Moscoso and Claverys, 2004; Claverys et al., 2006) the SilCR-mediated upregulation in the transcription is autoregulated and growth-phase restricted (Fig. 2A and B).
To examine whether the upregulation in silE/D/CR transcript is dependent on SilCR concentration we grew the WT strain to a late logarithmic phase in the absence and presence of increasing concentrations of SilCR, ranging from 0.1 μg ml−1 (49 nM) to 100 μg ml−1, and quantified the amount of silE transcript. We found that silE/D/CR transcription responded to increasing concentration of SilCR in a dose-dependent fashion, reaching almost its peak level at 1 μg ml−1 (Fig. 2C). This concentration of SilCR CSP agrees well with the concentration required to induce either competence or biofilm formation in CSP-null mutants of Streptococcus mutans (Li et al., 2001; Yoshida and Kuramitsu, 2002). Because in previous studies we used SilCR at a concentration of 10 μg ml−1 (Hidalgo-Grass et al., 2002; Hidalgo-Grass et al., 2004), we elected to continue using the same concentration in this work as well.
The homology of the TCS SilA/B to ComD/E and BlpH/R (Hidalgo-Grass et al., 2002) coupled with its colocalization with the ABC transporter on the sil locus, suggested that SilA/B could mediate SilCR signalling. To test this assumption, we used the silA¯ mutant (for construction see Experimental procedures) and the silB¯ mutants (Hidalgo-Grass et al., 2002) and measured the ability of SilCR to induce silE transcription. Compared with the WT, none of the mutants responded to SilCR by increasing silE transcript (Fig. 2D). This suggested that upon sensing of SilCR, SilB activates the response regulator SilA, which in turn stimulates the transcription of silE/D/CR from the P3 promoter. In contrast to S. pneumoniae where the CSP peptide activates also the transcription of the TCS, we did not observe an increase in the transcription of silA. On the contrary, there was a small but significant decrease in silA transcription in the presence of SilCR (Fig. 3A). Because genes of the quorum-sensing circuit such as the comD/E and comX exhibit a strong rise-and-fall pattern of transcription peaking at 10 min after CSP addition (Luo and Morrison, 2003; Peterson et al., 2004), we tested whether such temporary rise might exists also for silA/B and sigX. We found that SilCR addition did not affect either silA/B or sigX transcription during the first 30 min after CSP addition (Fig. 3B).
SilCR suppresses silC transcription
silC and silCR ORFs overlap in a way that silC ORF starts 34 bp upstream from the end of silCR ORF and ends 28 bp downstream from the silCR putative translation start (Hidalgo-Grass et al., 2002). This configuration of silC and silCR makes it essentially impossible to distinguish between double stranded cDNA of silC and silCR by either RT-PCR or RT-RT-PCR. Because of that we employed RNase protection assay (RPA) for detecting silC transcript. While detectable at time zero, the silC transcript became undetectable 1 min after SilCR addition and remained so for the next 30 min (Fig. 4A, left panel). We obtained an opposite behaviour for silCR transcript; it was barely detectable at time zero but became strongly apparent 1 min after SilCR addition, and remained so for the next 30 min. The rapid increase in the rate of silE/D/CR transcription induced by SilCR addition was verified by quantifying silE transcript by RT-RT-PCR. Already 2.5 min after SilCR addition, silE transcript was increased by about 30-fold reaching almost its peak level (Fig. 4B). To demonstrate that the SilCR-mediated suppression of silC transcription requires the SilA/B TCS, we repeated the experiment using the silA¯ mutant instead of the WT. As expected from the results presented in Fig. 2C, lack of SilA prevented the up-shift in the silCR transcription. The transcript of silC remained low but clearly apparent for 30 min after SilCR addition (Fig. 4A, middle panel). This experiment demonstrated that the SilCR-mediated suppression of silC transcription is absolutely dependent on SilA and requires the ongoing induction of silCR transcription from the converging P3 promoter. Next, we examined whether silC suppression requires ongoing transcription of the overlapping silCR. We repeated the experiment described in Fig. 4A using silE¯ mutant instead of the WT. silE¯ mutant was constructed by insertion inactivation which disrupted silE in its central part. By RT-RT-PCR analysis we confirmed that addition of SilCR induces silE/D/CR transcription, upstream but not downstream, of the disruption site (not shown). The result presented in Fig. 4A (right panel) demonstrates that silC transcript is unaffected by SilCR addition in the silE¯ mutant. Therefore, ongoing transcription of silCR is required to suppress silC transcription.
The transcription of the blp operon mirrors the interplay between silC/silCR transcripts
We performed a preliminary transcriptosome analysis of the WT strain grown in the presence and absence of SilCR, using the microarray system described previously (Beyer-Sehlmeyer et al., 2005). The micoarray analysis needs to be repeated and validated. Nonetheless, it led us to examine the transcription of a cluster of genes, blpMH, blp0484 and blp0486, all encoding bacteriocin-like peptide and localized downstream to sil (Fig. 1). RT-PCR analysis with primers RT-blpM-f, RT-blp048 and RT-spy0488-r (Fig. S1 and Table 1) revealed that the three genes are a part of an operon, which is transcribed oppositely to silE/D/CR from the putative P4 promoter (Fig. 1E and F). Primer extension analysis with primer PE-blpMH-r (Table 1) revealed that the TSS of the blp operon is located 58 bp upstream to blpMH (Fig. S3).
Another gene whose expression was implicated to be affected by SilCR in the microarray analysis was a transposase, homologous to spyM3_1016. Here we demonstrate that SilCR addition increased the transcription of spyM3_1016 homologue by sevenfold compared with non-treated bacteria (Fig. 5A). The SilCR-mediated induction of both blp0484 and of spyM3_1016 transcription was dependent on the SilA/B TCS, because no induction was observed in silA¯ mutant (Fig. 5A). The transcript induction of these two genes was rapid, reaching a peak level within 2 min (Fig. 5B), and exhibiting a kinetic similar to that of silE induction (Fig. 4B). As SilCR suppresses silC transcription, we hypothesized that the transcription of silE/D/CR, blp and spyM3_1016 would be upregulated in the ΔsilC mutant. We discovered that the transcription of blp and the silE/D/CR operons was strongly induced (∼60 fold) in ΔsilC mutant compared with the WT strain, while the transcription of spyM3_1016 remained the same (Fig. 5C). In an attempt to reverse the ΔsilC effect on the upregulation of blp and silE/D/CR transcription, we transformed this mutant with a shuttle vector, pLZsilC (Table 2), harbouring a DNA segment stretching from the end of silB to the beginning of the silE (Fig. 1). There was sixfold reduction in the transcription of spy0484 compared with that of ΔsilC transformed with the pLZ12-Km vector alone (Fig. 5C). The reduction in the transcription of silE was less significant as some reduction in silE transcription was caused by transformation of the ΔsilC mutant with the pLZ12-Km vector alone. These observations, taken together, suggest that the silC acts as a repressor of the putative P3 and P4 promoters. The silC-mediated repression is alleviated by SilCR addition due to upregulation of the transcription from the P3 and P4 promoters via the SilA/B TCS. Furthermore, SilCR can also regulate transcription independently of silC, as demonstrated by the spyM3_1016 upregulation.
A pLZ12-Km derivate containing a 4 kb stretching from silB to silE containing silC
SilCR is produced by GAS in vitro
To test for production of the SilCR peptide by clinically isolated GAS strains we prepared polyclonal anti-SilCR antibodies in rabbits. The immuno-dot-blot analysis shown in Fig. 6 demonstrates that SilCR antiserum successfully detected ng quantities of the SilCR peptide and did not react with a non-related (NR) peptide of a similar size. We examined 94 strains isolated from cases of invasive diseases (Moses et al., 2003) and cases of pharyngitis in Israel for: (i) presence of the sil locus by PCR analyses and (ii) production of SilCR by the immuno-dot-blot. We found that among 94 strains, 26 were positive for the presence of sil using one set of PCR primers and 20 using two sets of primers (Table S1). Thirty-two strains were positive for SilCR production and 17 strains were positive for both sil presence and SilCR production (Table S1). NS88 strain possessing sil, as judged by two sets of PCR primers (Supporting information Table S1), generated a very strong signal on the immuno-dot-blot (Fig. 6). To ensure that this signal results from production of a large amount of SilCR, we subjected serially diluted NS88 and JS95 strains to the immuno-dot-blot analysis. While JS95 did not react with the anti-SilCR antibody even at the highest dose used, NS88 reacted in a dose-dependent fashion (Fig. 6). Further validation for the production of SilCR by NS88 was provided by the finding that NS88 and four other strains producing high amounts of SilCR (as judged by the dot-blot analysis) all possessed ATG and not ATA at the start silCR codon (not shown). Thus, a significant number of GAS strains produce SilCR during growth in-vitro.
In Gram-positive bacteria, quorum-sensing is accomplished by signalling systems that depend on the secretion and sensing of small peptides. These quorum-sensing systems regulate various cellular functions including: virulence, production of bacteriocins, development of genetic competence, resistance to acid, and biofilm formation. Despite the large plethora of cellular functions the systems share basic features. This includes binding of a peptide to the sensor of a TCS, which activates the cognate response regulator that in turn induces changes in gene expression. One of the first induced changes is upregulation of the expression of the peptide itself. This type of auto-induction enables a rapid burst of activity once the threshold of the quorum-sensing has been reached.
In this work we provide the first direct evidence that SilCR acts as a signalling peptide changing gene expression profile in GAS. We demonstrate that SilCR auto-induces its own transcription and this induction is dependent on SilCR dose, and absolutely requires the SilA/B TCS. In contrast to S. pneumoniae, the autoregulation in GAS is restricted to silCR and the genes composing the ABC transporter, and does not include the TCS genes. Furthermore, sigX does not seem to be involved in the silC/silCR circuit. Previously it was suggested that silC transcription is linked to sigX based on the existence of a typical com-box promoter located immediately up-stream to this gene (Hidalgo-Grass et al., 2002). In this work we demonstrated that silC transcript start is located upstream to the putative silC com-box promoter, excluding the possibility that the latter serves as P2 promoter. In addition, we did not observe an increased sigX transcription in the presence of SilCR. Nonetheless, we demonstrate that SilCR upregulates the transcription of bacteriocins and of a transposase. Because SilCR is produced under physiological conditions by about 18% of GAS clinical isolates containing sil, these findings support the notion that SilCR might act as a key signalling molecule in GAS.
In the M14 strain JS95 there is a missense mutation which changes the AUG start codon of silCR to AUA suggesting that SilCR is not produced by this strain. Indeed, we could not detect SilCR formation by the immuno-dot-blot analysis. Yet, addition of synthetically produced SilCR (Hidalgo-Grass et al., 2004) or deletion of the overlapping silC obliterated virulence (Hidalgo-Grass et al., 2002). In attempt to unravel the mechanism by which SilCR regulates GAS virulence in the M14 strain, we studied the relationships between silC and silCR transcripts in the absence and presence of SilCR. Here we show that addition of SilCR suppresses silC transcription. The suppression requires ongoing induction of the converging and overlapping silCR transcript by the SilCR peptide. Two mechanisms could account for these findings. The first, termed transcriptional interference is characterized by suppressive influence of one transcriptional process directly and in cis, on a second transcriptional process (Callen et al., 2004; Shearwin et al., 2005). The second, involves degradation of RNA by RNase E upon formation of a complex between the antisense RNA and its mRNA target (Carpousis, 2003). Further experiments are required to pinpoint which of the two above-mentioned mechanisms is implicated in silC/silCR circuit.
One of the possible reasons for the finding that silC product inhibits the transcription of silCR from the P3 promoter is to endow the sil locus with an ultra-sensitive switch. On one hand, silC-mediated inhibition of the P3 promoter ensures that the signalling system is not turned-on before the TCS senses a critical concentration of SilCR. On the other hand, it also ascertains that once the SilCR concentration drops below its threshold the signalling system is turned-off rapidly.
We were able to partially complement the repression of blp transcription from the P4 promoter by expressing silC from a plasmid in ΔsilC mutant. Reversal of silC-mediated suppression of P3 promoter needs to be further evaluated because the shuttle vector itself seems to induce some suppression. The precise way by which P3 and P4 are stimulated by SilA and inhibited by silC remains to be determined.
Bacteriocin transcription is switched on and off in a co-ordinated fashion with the silCR/silC circuit, suggesting that the bacteriocins may act in-concert with SilCR. Indeed, it has been reported that in some strains of S. mutans the same ABC transporter system is responsible for the processing and secretion of CSP and of the bacteriocins (van der Ploeg, 2005). Furthermore, it was shown that co-ordinated bacteriocin production and competence development enables S. mutans to acquire DNA from neighbouring species (Kreth et al., 2005).
Based on the results shown in this work we propose a general working model for gene and virulence regulation of GAS by sil. We hypothesize that the product of silC transcript acts as a transcriptional regulator of GAS virulence determinants (Fig. 7). SilCR, by interacting with the TCS SilA/B regulates gene expression also independently of the silC/SilCR circuit. Recently, we have shown that SilCR downregulates the activity of ScpC, a protease that inactivates CXC chemokines preventing polymorphonuclear neutrophil recruitment to the site of infection (Hidalgo-Grass et al., 2004; 2006), in a silC-independent fashion (not shown). Here we demonstrate that SilCR upregulates the transcription of transposase gene homologues to spyM3_1016, independently of silC. Further studies of sil regulation will add a new dimension to our understanding of GAS virulence regulation.
Bacterial strains, growth conditions and plasmids
Primers, bacterial strains and plasmids used in this study are described in Tables 1 and 2 respectively. Molecular cloning experiments utilized Escherichia coli JM109 strain, which was cultured in Luria–Bertani broth, Lennox (Becton, Dickinson, Sparks MD, USA). For culturing of GAS we employed Todd-Hewitt medium (Becton, Dickinson) supplemented with 0.2% yeast extract (Becton, Dickinson) (THY media) with incubation at 37°C in sealed tubes without agitation. To produce solid media, BactoTM Agar (Becton, Dickinson) was added to final concentration of 1.4%. Antibiotics were added at the following concentrations when necessary: for GAS: 250 μg ml−1, kanamycin (Km), 50 μg ml−1 spectinomycin (Spec) and 1 μg ml−1 erythromycin (Erm), for E. coli: 100 μg ml−1 ampicillin (Amp), 50 μg ml−1 Spec and 750 μg ml−1 Erm. All the antibiotics were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Plasmid DNA was isolated by commercial kits (High Pure Plasmid Isolation, Roche Applied Science, Basel, Switzerland or Wizard®Plus Midipreps, Promega, Madison WI, USA) according to manufacturer's instructions and used to transform E. coli by standard methods and to transform GAS by electroporation as described previously (Caparon and Scott, 1991). Restriction endonucleases, ligases and polymerases were used according to the recommendations of the manufacturers. Chromosomal DNA was purified from GAS as described previously (Caparon and Scott, 1991) or by using the Wizard® Genomic DNA Purification Kit (Promega). Linear DNA fragments were purified using Certified™ low-melt agarose (Bio-Rad, Hercules, CA, USA) prior to electroporation into GAS strains. PCR products were purified using commercial kit (QIAquick PCR Purification Kit, Qiagen, Hilden, Germany). DNA sequencing was performed using the Excel II cycle sequencing kit (Hy Laboratories, Rehovot, Israel). All other procedures were conducted according to standard protocols (Sambrook et al., 1989).
Construction of sil mutants
Mutant strains of the genotypes silA¯, silB¯, silD¯ and silE¯ were derived from strain JS95 by insertion inactivation. To construct the silA-disrupted mutant an internal 492 bp region of the gene was PCR amplified using the primers C-silA-f, C-silA-r (Table 1), followed by A-T cloning into pGEM-T-Easy (Promega). The corresponding plasmid pGsilA (Table 2) was digested with SacII and PstI and the silA internal fragment was cloned into the temperature-sensitive plasmid pJRS233 (Perez-Casal et al., 1993) digested with the same restriction enzymes to generate the plasmid pJsilA (Table 2). Transformation of JS95 with pJsil was conducted as previously described to identify plasmid integrants (Perez-Casal et al., 1993). The presence of the plasmid insertion was verified by PCR analysis of the mutant chromosomal DNA using two sets of primers: V-M13-f and V-silA-f V-M13-r and V-silB-r (Table 1). To construct the silE-disrupted mutant an internal 635 bp region of the gene was PCR amplified using the primers C-silE-f, C-silE-r (Table 1), followed by A-T cloning into pGEM-T-Easy (Promega). The corresponding plasmid pGsilE (Table 2) was digested with NcoI and PstI and the silA internal fragment was cloned into the temperature-sensitive plasmid pJRS233 (Perez-Casal et al., 1993) digested with the same restriction enzymes to generate the plasmid pJsilE (Table 2). Transformation of JS95 and integrants identification was done as described for silA¯. The insertion of the plasmid was verified by PCR analysis of the mutant chromosomal DNA using two sets of primers: V-M13-f and V-silE-f V-M13-r and V-silE-r (Table 1). silB¯ and silD¯ were constructed as described previously (Hidalgo-Grass et al., 2002). The mutation in each mutant was confirmed by sequencing the junction regions between the chromosomal and inserted plasmid DNA. To construct pLZsilC, a 4 kb PCR fragment, stretching from silB end to silE start codon was amplified from JS95, using primers C-silCcomp-f and C-silCcomp-R. These primers were designed to introduce PstI and BglII sites at the 5′- and 3′ ends of the amplified PCR fragment respectively. The amplified fragment was digested with PstI and BglII and subsequently ligated into PstI and BglII digested pLZ12-Km. The resulting pLZsilC was introduced into ΔsilC mutant by electroporation and transformants resistant to kanamycin and spectinomycin were isolated. The complementation of silC mRNA was verified by RT-PCR using primers V-silCcomp-f and V-silCcomp-r.
For all RNA preparations JS95 or derivative mutants were grown in THY. Cultures were harvested at different stages of growth, centrifuged at 5000 g for 10 min and pellets were frozen in liquid nitrogen. RNA was isolated by hot acidic phenol extraction as previously described (Ravins et al., 2000). Representative samples were assessed for RNA integrity by electrophoretic analysis. Measurement of the absorbance ratio at 260/280 nm were used to determine the RNA concentration and purity (accepted if > 1.9). Contaminating DNA was removed by DNase treatment according to the manufacturer's instructions (RQ1 RNase free DNase, Promega). Samples were rejected if PCR amplification performed with RNA templates (Table 1 SD-sra-f and SD-sra-r) indicated the presence of contaminating DNA.
SilCR was synthesized, purified to 96% purity by BioSight (Israel) and full size was confirmed by GC/MS or HPLC. It was dissolved in double-distilled water to a concentration of 5 mg ml−1 and appropriate volumes were added to THY culture medium to obtain the desired final SilCR concentrations. The cultures were grown until reaching the appropriate optical density at 600 nm (OD600) and RNA was isolated and purified as described above. To quantify transcript amount immediately after SilCR addition, 350 ml of JS95 or derived mutants were grown to an OD600 = 0.25. Successive 50 ml culture samples were withdrawn from the SilCR-treated culture 10 s before (zero time) and at specified intervals after SilCR addition. At these intervals the 50 ml cultures were transferred from 37°C to liquid nitrogen, cooled for 5 s and subjected to centrifugation at 5000 g for 10 min at 4°C. Pellets were frozen in liquid nitrogen and RNA was isolated as described above.
RT-PCR and RT-RT-PCR determinations
DNA-free total RNA (4 μg) was used for cDNA synthesis using MMLV reverse transcriptase (Promega), according to the manufacturer's protocol. Standard RT-PCR reactions were conducted using BIOTAQ™ DNA polymerase (Bioline) according to manufacturer's protocol. RT-RT-PCR, primers (Table 2) were designed using Primer Express™ software v2.0 (Applied Biosystems). SYBR-green mix (Absolute SYBR GREEN ROX MIX, ABgene) was used for fluorescence detection with Rotor-Gene 3000 A (Corbett) according to manufacturer's instructions. The cDNA amount of gyrase subunit A (gyrA) was used to normalize expression data for each target gene. Transcription of gyrA is considered constant under a variety of in vitro conditions (Graham et al., 2002). Each assay was performed in duplicates with at least two RNA templates prepared from bacteria from independent cultures grown on different days. The data were analysed according to the standard curve method (Rotor-gene analysis software 6.0) and are presented as abundance of transcript amount relative to that of gyrA.
Primer extension analysis
RNA samples (30 μg) were annealed to the corresponding end-labelled primers (70°C for 10 min, followed by incubation for 20 min at 4°C) and then subjected to primer extension (at 42°C for 45 min) with 1 unit of AMV-RT (CHIMERx Milwaukee WI, USA) and dNTPs (Promega) (0.5 mM each). The extension products were separated on 6% sequencing gels, alongside with sequencing reactions.
RNase-protection assay (RPA)
To produce DNA template for gyrA probe, a portion of the gene (513 bp) was PCR amplified using primers RPA-T7gyrA-r (introduces a T7 promoter at the antisense strand) and RPA-gyrA-f (Table 1). To produce DNA template for silCR probe, a 400 bp fragment including 150 bp of silCR ORF and 250 bp upstream was PCR amplified using primers RPA-T7silCR-f (introduces a T7 promoter at the antisense strand) and RPA-silD-r (Table 1). The gyrA and silCR DNA templates were purified using the Wizard® PCR Preps Kit (Promega). To produce template for silC probe, the plasmid pGsilC (Table 2) was constructed. A 351 bp fragment that includes BglII restriction site proceeded by the complete silC and part of silD 3′ was PCR-amplified from JS95 DNA with primers RPA-BglII-silC-f and RPA-silD-r. The PCR product was then cloned in pGEM-T-Easy to give pGsilC (Table 2). pGsilC contains a T7 RNA polymerase promoter 52 bp downstream to the multiple cloning region. It was linearized using BglII, which cuts the plasmid five bp upstream of silC start codon. The linear plasmid was purified using Wizard DNA clean-up system (Promega).
Anti-sense RNA probes were synthesized by in-vitro transcription with DIG RNA labelling mix (Roche) and T7 RNA polymerase (New England Biolabs) according to the recommendations of the manufacturers. The DIG-labelled probes were purified from acrylamide gel using probe elution buffer (RPAIII, Ambion), according to manufactures instructions. Determination of labelling efficiency was done in a spot assay according to manufactures instructions (DIG RNA labelling, Roche). The expression levels of specific mRNAs were determined by RPA using the RPAIII kit (Ambion) according to manufacturer's protocol. Briefly, 50 μg of RNA preparations generated at the specified intervals after SilCR addition, were coprecipitated with an excess (determined empirically) of a probe corresponding to the mRNA of either silC, or silCR or gyrA and the mixtures were allowed to hybridize overnight at 42°C. Hybridized samples were digested with a 1:50 dilution of RNase T1 provided within the RPAIII kit (Ambion). Protected fragments were resolved by denaturing PAGE (5% polyacrylamide/8M urea) and transferred to a positively charged nylon membrane (Roche) by semidry electro-blotter (Bio-Rad, Hercules, CA, USA). Membranes were cross-linked by UV and the protected RNA probes were detected by an immunoassay with anti-DIG-AP conjugate using the chemiluminescent substrate CDP-star (DIG luminescent detection kit, Roche) according to manufacturer's instruction.
Immuno-dot-blot assay for SilCR
Synthetic SilCR (500 μg ml−1) were diluted (1:1) with complete Freund's adjuvant and 2 ml were injected subcutaneously into a rabbit back. The first injection was followed by two booster injections at days 14 and 28 in incomplete Freund's adjuvant. Rabbits were bled at day 42 after the first injection. Immunological dot-blot assay was employed to confirm the production of SilCR by GAS. Two microlitres of bacterial suspension (or successive dilutions) in THY medium were spotted on nitrocellulose membrane (0.45 μM, Schleicher and Schuell Bioscience, GmbH, Dassel Germany). The membrane was allowed to dry and then it was blocked in 1% skim milk for 2–4 h. The membrane was washed three times with Tris-buffered PBS supplemented with 0.03% Tween 20 (TBS-T) and then incubated with the anti-SilCR antibody at dilution of 1:20 000 in TBS-T supplemented with 1% skim milk. After 2 h, the membrane was washed three times in TBS-T, followed by 1 h incubation with the secondary anti-rabbit IgG HRP-conjugated antibody (Promega), at a dilution of 1:20 000 in TBS-T. The membrane was washed three times in TBS-T and incubated for 3 min with chemiluminescent substrate Super Signal®WestPico (Pierce, Rockford, IL, USA) and dots were visualized on X-ray film after 1 min of exposure.
We are grateful to Dr Mary Dan-Goor for the preparation of the anti-SilCR antibody. We thank Miriam Ravins for careful reading of the manuscript and helpful advice. We also are grateful to Z. Korenman from Israel Ministry of Health Streptococcal Reference Laboratory, Jerusalem, Israel for providing the GAS strains isolated from cases of pharyngitis. This work was supported by grants from: The Israeli Science Foundation administered by the Israel Academy of Science and Humanities (to E.H.); The German-Israeli Foundation (to E.H. and A.B.). E.H. is an international research scholar from the Howard Hughes Medical Institute.