Biofilm formation is an important aspect of the pathogenesis of staphylococcal infections. A β-1,6-linked N-acetyl glucosamine polysaccharide is critical to biofilm elaboration and is synthesized by proteins encoded by the intercellular adhesion (ica) locus. These studies were undertaken to characterize the mechanism by which transcription of the ica locus in S. aureus is regulated using isogenic S. aureus MN8 and MN8 mucoid (MN8m) strains, the latter of which constitutively overproduces biofilm. Transformation of the ica locus from MN8m to the ica knock-out mutants of two strains, MN8 and NCTC 10833, conferred a strong biofilm-producing phenotype. Sequence analysis revealed a 5-nucleotide deletion within the promoter region of the ica locus in MN8m compared with the sequence in the wild-type locus. Deletion or substitution of these 5 nucleotides within the wild-type ica locus augmented transcription of the ica locus and induced the strong biofilm-producing phenotype. Gel shift analysis demonstrated that a protein(s) within cell-free lysates from strain MN8 bind(s) specifically to oligonucleotides representative of the wild-type ica promoter sequence and that this binding is greatly diminished by the deletion or substitution of the 5 nucleotides. DNase I footprint analysis revealed that purified IcaR, thought to be a regulator of ica transcription, also binds to the ica promoter sequence just upstream of the ica start codon, but its affinity for the ica promoter is unaffected by deletion of the 5-nucleotide motif. These findings identify a 5-nucleotide motif within the ica promoter region that has a functional role in transcriptional regulation of the ica locus that is independent of IcaR, and also show that IcaR binds to the promoter region of the S. aureus ica locus.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Historically, prokaryotes have been deemed as elementary, autonomous, unicellular organisms. It is becoming increasingly apparent, however, that communication between bacterial cells allows them to sense and adapt to their surroundings. One such method of adaptation is the formation of a complex, structured, multicellular network called a biofilm. A major component required for biofilm formation is the production of an extracellular glycocalyx, which is principally composed of polysaccharides of bacterial origin. Staphylococcus aureus, which is particularly noteworthy for its genetic plasticity and ability to thrive under a variety of adverse conditions, regulates the production of this glycocalyx in response to its environment.
The major component of the S. aureus glycocalyx is polymeric N-acetyl glucosamine (PNAG). Originally described in Staphylococcus epidermidis and designated as the polysaccharide intercellular adhesin (PIA) (Mack et al., 1996), PNAG production and biofilm elaboration are thought to protect the bacteria from both the host immune system and antibiotics, and can complicate the treatment of S. aureus infections. Once an S. aureus biofilm has been established on an implanted medical device, the infection can be difficult to treat and may require removal of the infected device and repair of damaged tissue. Understanding why and how biofilm formation is induced in vivo is a necessary prerequisite for the development of more suitable therapies that might make the removal of infected devices unnecessary.
Currently, our knowledge about the regulation of PNAG synthesis is insufficient. Production of the polysaccharide is dependent upon the proteins encoded by the intercellular adhesion (ica) locus, originally detected in S. epidermidis and later found by McKenney et al. and Cramton et al. to be present in S. aureus (Heilmann et al., 1996; McKenney et al., 1998; Cramton et al., 1999). The ica locus is made up of five genes: icaR, icaA, icaD, icaB and icaC. The icaR gene is transcribed divergently from the other four genes, and the icaA, B, C and D genes appear to be translated from a single transcript. Several known stimuli induce PNAG production in vitro including elevated glucose, high osmolarity, low levels of ethanol, iron restriction and oxygen deprivation (Rachid et al., 2000a,b; Cramton et al., 2001a). These stimuli may work, in part, through the alternative sigma factor, σB, which couples with the core RNA polymerase complex to initiate transcription of its target operon in response to conditions of stress. Stressful conditions such as ethanol and high osmolarity also induce transcription of the σB operon. These findings, along with reports that mutations within the σB activator RsbU abrogate PNAG production, suggest that σB plays an important role in the transcriptional activity of the ica locus (Rachid et al., 2000c; Knobloch et al., 2001). A consensus binding site for σB is not present within the ica promoter, however, so its role may be an indirect one.
The IcaR protein appears to be another regulatory factor controlling the ica locus. The amino-terminal third of IcaR exhibits 72% homology and 40% identity with other regulatory proteins of the tetracycline-regulatable (TetR) family. There is evidence that IcaR represses transcription of the ica locus (Conlon et al., 2002; Götz, 2002), but whether repression of the ica locus by IcaR is direct or indirect is not known.
The goal of this investigation was to characterize the mechanism by which biofilm formation is controlled in S. aureus at the level of transcriptional regulation of the ica locus. S. aureus strain MN8 is a clinical isolate that fails to elaborate biofilm in vitro in the absence of a stimulus such as elevated glucose. A spontaneous mutant of MN8, which constitutively overproduces PNAG, was previously isolated in our laboratory and named MN8 mucoid (MN8m) (McKenney et al., 1999). In this study, we used strain MN8m to identify components affecting the molecular genetic regulation of ica transcription and PNAG synthesis in S. aureus and investigated the interaction of the IcaR protein with the ica promoter region.
The phenotype of a PNAG-hyperproducing strain results from changes within the ica locus
We initially hypothesized that PNAG overproduction by S. aureus strain MN8m resulted from a mutation within the ica locus itself and that the phenotype could be transferred to other strains by a plasmid-borne copy of the MN8m ica locus. To test this, we cloned the entire ica locus from S. aureus MN8 or S. aureus MN8m into the temperature-sensitive plasmid pBT9 to create pWT (from strain MN8) and pMUC (from strain MN8m). The plasmids were electroporated into S. aureus RN4220, then transduced using phage 80 into two S. aureus strains in which the chromosomal copy of icaADBC and part of icaR were replaced with a tetracycline resistance cassette (Cramton et al., 1999). One of the strains, NCTC 10833, is a clumping factor-positive strain, and the other, MN8, is a TSST-1-producing strain. The pWT construct restored the wild-type biofilm phenotype in MN8Δica::tet and 10833Δica::tet, in which a modest biofilm was produced but only in the presence of 1% glucose in the growth medium (Fig. 1). The pMUC construct conferred approximately sevenfold greater biofilm production relative to pWT, and strains 10833Δica::tet/pMUC and MN8Δica::tet/pMUC elaborated a thick biofilm with macroscopic aggregates of bacterial cells visible on the plastic wells (Fig. 1). These results supported our hypothesis that the mucoid phenotype of MN8m results from mutations within the ica locus itself.
A 5-nucleotide deletion in the ica promoter region augments transcription of the ica locus and induces constitutive hyperproduction of PNAG
We performed sequence analysis to search for mutations within the ica locus of S. aureus strain MN8m that could be responsible for the PNAG hyperproducing phenotype. A 5 bp deletion was found within the promoter region of the ica locus from strain MN8m compared with the same sequence from strain MN8 (Fig. 2). The derived sequence of the ica locus in strain MN8 was identical to that of the previously determined ica sequence from S. aureus strain 113 (Cramton et al., 1999). Site-directed mutagenesis was then used to delete the 5 bp motif in the plasmid pWT to create plasmid pDEL. As Fig. 3 shows, pDEL induced heavy biofilm and cell cluster formation in S. aureus strains 10833Δica::tet/pDEL and MN8Δica::tet/pDEL, indicating that the 5 bp deletion was sufficient to induce the constitutive PNAG-overproducing phenotype.
In order to determine whether the loss of the exact sequence or the change in length of DNA was responsible for the overproducing phenotype, site-directed mutagenesis was used to substitute the original sequence, TATTT, with five complementary bases, ATAAA, within the ica promoter region of plasmid pWT to create plasmid pSUB. The substitution mutant plasmid, pSUB, also augmented biofilm formation and cell cluster formation in 10833Δica::tet/pSUB and MN8Δica::tet/pSUB, although the increase in biofilm elaboration by pSUB was not as pronounced as that induced by pMUC and pDEL (Fig. 3).
To confirm that the increase in biofilm elaboration in the recombinant S. aureus strains resulted from an increase in transcription of icaADBC genes, we performed Northern slot-blot analysis. As would be expected, MN8Δica::tet and 10833Δica::tet, which are deleted in icaADBC and partially deleted in icaR, did not contain ica transcript. The pWT plasmid restored wild-type levels of ica transcription to S. aureus strains MN8Δica::tet and 10833Δica::tet, whereas the pMUC, pDEL and pSUB constructs induced levels of ica expression in the ica knock-out strains comparable with that detected in strain MN8m (Fig. 4). Owing to recent evidence implicating IcaR in the repression of PNAG production, we hypothesized that the deletion mutation would have either no effect or a negative effect on the transcriptional activity of icaR. Northern analysis using a probe specific for icaR demonstrated that icaR transcript levels were equivalent in MN8 and MN8m and were restored in the knock-out strains to wild-type levels by the plasmid constructs, but transcription was unaffected by the 5 bp deletion within the ica promoter of strain MN8m. Together, these results indicate that this particular TATTT motif within the ica promoter region plays a critical role in regulating icaADBC but not icaR transcription and subsequent biofilm production.
Cell-free lysate from S. aureus induces shifts in the electrophoretic mobility of an oligonucleotide containing the TATTT motif
The results from the genetic analysis led us to hypothesize that the 5 bp motif occurred within the binding site for a repressor of ica transcription. The promoter for the entire ica locus is believed to be located within the 164 bp region between the divergently transcribed start codons of icaR and icaA. A 198 bp stretch of DNA containing the promoter plus the first few nucleotides of the icaR and icaA genes was generated from S. aureus strains MN8 or MN8m total DNA by polymerase chain reaction (PCR), radiolabelled and combined with cell-free lysate from S. aureus strain MN8 for gel shift analysis. Reproducible mobility shifts occurred with the wild-type (WT) probe (Fig. 5), and four bands were consistently noted in the presence of MN8 cell-free lysates. A higher lysate concentration (10 µg) caused the bands to smear. Figure 5 also shows that an unlabelled specific competitor competes for binding to the protein causing the mobility shift (Fig. 5, lane 4), whereas a non-specific competitor has no effect on the mobility shift of the target DNA sequence (Fig. 5 lane 3). When analysed for the ability to bind to a probe containing the 5 bp deletion in the S. aureus strain MN8m locus, it was found that the lower bands were also detected by gel shift analysis with the MUC probe, whereas the upper band was greatly diminished, even in the presence of 10 µg of whole-cell lysate (Fig. 5).
A series of oligonucleotides was designed to identify the shortest sequence containing the 5 bp motif that was shifted by factors within the lysate of strain MN8. A 53 bp probe, WTshort, was the shortest oligonucleotide containing the 5 bp sequence to be shifted by the lysate from S. aureus strain MN8 (Fig. 6). The lysate caused two distinct shifts in the WTshort probe, which were competitively inhibited by unlabelled WTshort but not by an unlabelled non-specific probe.
The DNA-binding factor in S. aureus MN8 lysate has a higher affinity for the wild-type sequence
Short (53 bp) oligonucleotides were also generated that contained the 5 bp deletion (DEL) or the 5 bp substitution (SUB). Figure 7 indicates that the DEL and SUB probes were not shifted by the factor(s) in the MN8 lysate as was the WT probe. These results demonstrate that the 5 bp motif is required for binding of at least one staphylococcal DNA-binding protein. These findings thus support the hypothesis that this unknown DNA-binding protein is a repressor and that the constitutive PNAG overproduction by strain MN8m and S. aureus strains transduced with the pMUC, pDEL and pSUB constructs results from the inability of this repressor to bind to the altered ica promoter.
Recombinant IcaR induces a mobility shift in the wild-type and mutant probes
The protein encoded by the icaR gene exhibits homology to the TetR family of regulatory proteins. In addition, Götz (2002) recently stated in a review that deletion of icaR augments polysaccharide production in S. epidermidis, whereas overexpression of the gene abrogates polysaccharide synthesis in S. epidermidis. Similarly, Conlon et al. (2002) reported that deletion of the icaR gene in S. epidermidis augments transcription of icaA. We therefore hypothesized that IcaR was the DNA-binding protein detected in the mobility shift assays. Recombinant, Xpress epitope-tagged, histidine-tagged IcaR was expressed in Escherichia coli and purified by nickel affinity chromatography. The purified protein was subjected to gel shift analysis using the 198 bp WT probe. Recombinant IcaR produced a shift in the mobility of the 198 bp WT probe (Fig. 8). The specificity of the mobility shift is indicated by the loss of the shifted band when a 100-fold excess of specific unlabelled competitor DNA was used. A non-specific competitor, a 198 bp stretch of DNA from the icaA gene coding sequence, had no effect on the mobility shift (Fig. 8). The vector control (Xpress epitope plus histidine tag) induced a weak shift, but this shift was not reduced by a 100-fold excess of specific competitor, indicating the non-specific nature of the minor shift observed. There was no detectable difference in the binding of recombinant IcaR to the WT and MUC probes (Fig. 9). When we used the 53 bp probe, WTshort (see Fig. 6), in a gel shift assay with IcaR, no binding to the probe was seen (data not shown). Thus, it is not IcaR but another DNA-binding protein(s) involved in the regulation of ica transcription that is affected by the 5 bp deletion in the ica locus of strain MN8m.
IcaR binds to the DNA sequence proximal to icaA
The finding that binding of IcaR is unaffected by the 5 bp deletion and that it fails to shift the WTshort probe impelled us to investigate the precise DNA sequence recognized by the protein. We used a solid-phase DNase I protection technique to locate the IcaR-binding domain within the ica promoter. Recombinant IcaR protected a 42 bp region immediately upstream from the icaA gene but did not bind in the region containing the 5 bp TATTT motif (Fig. 10). This finding supports our conclusion that the role of the 5 bp motif in transcriptional control of the ica locus is independent of IcaR and suggests that IcaR may function by sterically hindering the binding of the σ-factor of the RNA polymerase complex to the icaA promoter.
Evidence for the importance of the ica locus as a staphylococcal virulence factor is abundant. It is now known that virtually all clinical isolates of S. epidermidis and S. aureus carry the ica genes (Muller et al., 1993; Ziebuhr et al., 1997). The ica locus was the only genetic marker found by Galdbart et al. (2000) to correlate with biomaterial adherence, a feature that differentiated commensal skin isolates of S. epidermidis from isolates recovered from prosthetic joint infections. Peacock et al. (2002) showed that ica was one of seven out of 33 studied virulence factors of S. aureus found to contribute independently to pathogenesis in blood isolates.
Despite its importance in pathogenesis, very little is known about the regulation of transcription of the ica locus. Several reports implicate a role for the alternative sigma factor σB, but the lack of a σB consensus binding site within the promoter sheds doubt on its direct involvement in the transcriptional activation of ica (Rachid et al., 2000a,c; Knobloch et al., 2001). The S. aureus strain MN8m was shown previously by immunochemical analysis to constitutively overproduce PNAG (McKenney et al., 1999). Using this strain to probe the regulation of ica transcription and biofilm formation, we demonstrate here that a specific 5 bp motif in the ica promoter region plays an important role in the control of transcription of this locus and that the spontaneous loss of this putative binding site in S. aureus strain MN8m is the cause of the PNAG-overproducing phenotype. Transfer of the strain MN8m ica allele to two S. aureus strains deleted for the chromosomal ica locus also resulted in hyperproduction of PNAG as determined in the biofilm assay. Gel mobility shift studies showed the presence of a cytoplasmic DNA-binding protein(s) with much greater affinity for the wild-type ica promoter than for the mutant ica promoter. Further analysis of the IcaR protein also showed that it bound specifically to the ica promoter immediately upstream from icaA and that there was no difference in the binding of IcaR to wild-type or mutant promoter sequences.
The IcaR protein, which exhibits homology to the TetR family of transcriptional regulators, appears to be a repressor of transcription of icaA in S. epidermidis, but its mechanism has not been described (Conlon et al., 2002; Götz, 2002). The data presented in this report support these findings and also provide evidence that IcaR is indeed a DNA-binding protein that attaches specifically to the ica promoter region. The finding that IcaR binds just 5′ of the icaA initiation codon suggests that it may obscure the −35 and −10 sites and interfere with binding of the σ-factor of the RNA polymerase complex to the ica promoter.
The finding that IcaR bound equally well to the full-length ica promoter regions from strains MN8 (WT) and MN8m (MUC) was somewhat unexpected given that the absence of the 5-nucleotide motif results in high levels of ica transcription. Although the DNase I footprint analysis demonstrated that IcaR does not bind near the 5 bp motif, the absence of this motif in the ica locus of strain MN8m probably affects the negative regulatory activity of IcaR in the presence of the intact, chromosomal copy of ica in this strain. Additionally, the transcription of icaR was unaffected by the 5 bp deletion in the ica locus of strain MN8m, indicating not only that this portion of the promoter does not affect icaR transcription but also that IcaR is produced in the overexpressing strains but is unable to manifest its negative regulatory activity. Concurrent expression of both icaR and icaA has also been noted by Conlon et al. (2002), who found that inclusion of 4% NaCl in the growth media induces biofilm elaboration and icaA transcription in S. epidermidis even though the level of icaR transcript is also substantial under these conditions.
The addition of a cellular lysate from strain MN8 resulted in several bandshifts using both full-length WT and MUC promoter sequences, suggesting the presence of regulatory proteins in addition to IcaR, the binding of which to ica is not affected by the 5 bp deletion. However, we also found that there were mobility shifts that occurred when the WT, but not the MUC, promoter sequence containing the 5 bp deletion of TATTT was used, indicating the presence of an ica-binding protein that require the 5 bp sequence for binding. In addition, a protein within the cell-free extracts was able to shift a 53 bp portion of the WT ica promoter sequence, which was not shifted by recombinant IcaR, but did not bind to the MUC ica promoter sequence, further substantiating the presence of DNA-binding, and potentially regulatory, proteins that require the TATTT motif within the ica promoter for binding. Together, these findings indicate that other, as yet uncharacterized, DNA-binding proteins are involved in interactions with the ica promoter and appear to have a predominant role in negatively regulating its transcription, and this protein(s) is also required for IcaR to have a negative regulatory role in the transcription of ica.
The role played by additional DNA-binding proteins in the manifestation of the negative regulatory activity of IcaR may be to expose the IcaR binding site in the chromosomal ica locus. For example, formation of the well-characterized Gal repressor (GalR) repressosome complex requires concurrent binding of the DNA-binding protein HU (Aki and Adhya, 1997; Orth et al., 2000). The histone-like protein HU binds to a site between two GalR binding sites and plays a role in bending of the DNA. The HU-induced bending allows the two DNA-bound GalR dimers to interact, forming a DNA loop (Kar and Adhya, 2001). Thus, one plausible explanation for the failure of IcaR to discriminate between the WT and MUC promoter sequences is that the TATTT motif that is deleted in the MN8m ica promoter is normally required for binding of an additional regulatory factor that exposes the IcaR binding site in the ica promoter but, in a 198 bp recreation of the promoter, the IcaR binding site is readily accessible to this protein.
Stretches of adenosine and thymidine residues cause intrinsic bends in DNA. If IcaR does indeed bind just upstream of the ica start codon in the chromosomal copy of the ica promoter, then an intrinsic bend might be required to form the proper binding site for a protein involved in making the IcaR-binding region available to the IcaR protein. Using the program model-it (Munteanu et al., 1998), we found that deletion of TATTT does in fact reduce an intrinsic bend within the ica promoter, and substitution of the motif with ATAAA causes the DNA to bend in the opposite direction (not shown).
In summary, the TATTT motif located in the centre of the ica promoter plays a critical but as yet, not fully understood role in the regulation of the ica locus. The IcaR protein is a DNA-binding protein that attaches specifically to the DNA sequence adjacent to the icaA gene. At least one other staphylococcal DNA-binding protein binds within the ica promoter, and this binding is dependent upon the presence of the TATTT motif. The protein(s) requiring the TATTT motif for binding must have a dominant regulatory role in ica transcription, as the absence of the TATTT motif in the ica locus of strain MN8m allows for overexpression of ica transcripts and biofilm formation in spite of the normal transcription of icaR and a lack of an effect on the binding site for IcaR. These findings provide additional insight into understanding the mechanism of transcriptional regulation of the staphylococcal ica locus.
Staphylococcus aureus strain MN8 is a clinical isolate originally obtained from Schlievert and Blomster (1983). Strain MN8m was a spontaneous mutant isolated from a chemostat culture of strain MN8 (McKenney et al., 1999). Strain NCTC 10833 (ATCC 25904) is a clumping factor-positive variant of a throat swab isolate. Partial deletion of the ica locus to produce strains 10833Δica::tet and MN8Δica::tet was performed as described previously (Cramton et al., 1999; 2001b).
Plasmids, primers and cloning and expression of genes in the ica locus
All plasmid purifications were performed with the QIAprep spin miniprep kit (Qiagen). All primers were custom synthesized by Qiagen Operon. Restriction enzymes and DNA-modifying enzymes were purchased from New England Biolabs. The expression plasmid pCRT7-NT (Invitrogen) was used to clone the icaR gene and express recombinant IcaR protein. The gene was amplified using the primer pair icaRF (TTTCTTCAAAAATATATTTAGTAGCGAATACAC) and icaRR (AAGGATAAGATTATTGATAACGCAATAAC). The icaR gene was cloned into the vector and expressed in E. coli BL21pLysS cells according to the manufacturer's instructions. The pCRT7-NT vector adds a tag of six histidine residues and an Xpress epitope to the amino-terminus of the protein. The IcaR protein and the vector expression control (the histidine tag and Xpress epitope alone) were purified using Probond nickel affinity chromatography resin (Invitrogen). Plasmid pBT9 is a temperature-sensitive plasmid derived from pBT2 that lacks the EcoRI restriction site (Bruckner, 1997). All plasmid constructs were initially transformed into the restriction-deficient S. aureus strain RN4220 according to the method of Lee (1993). Constructs were transferred to other strains of S. aureus by transduction using phage 80 (Kasatiya and Baldwin, 1967; Novick, 1967). The ica loci from different strains were amplified by PCR using previously described primers SA11 and SA12 (Cramton et al., 1999) and the Elongase kit (Invitrogen). PCR products were gel purified using Ultrafree DA spin columns (Millipore), digested with KpnI and ligated into the KpnI site of pBT9. PCR-based site-directed deletion mutagenesis was performed using the following primer pair: DelFwd (CCGTT TAATTATAACAA CAATCTATTGC) and DelRev (TTTGTAAT TGCAACTTAATTTTCCTGTAAC). PCR-based substitution mutagenesis was performed using the following primer pair: DelRev and SubFwd (ATAAACCGTTTAATTATAACAACAAT CTAATTGC). PCR was performed using the Elongase kit, template DNA was digested with DpnI for 30 min at 37°C, DNA ends were phosphorylated with T4 kinase, and the PCR products were blunt-end ligated. All mutations were confirmed by DNA sequencing, which was carried out by the Microbiology Core Facility at Harvard Medical School (Boston, MA, USA).
Microtitre plate assays for biofilm production were performed essentially as described by Christensen et al. (1985) with minor modifications. Cultures were grown overnight in 10 ml of tryptic soy broth (TSB) +1% glucose, diluted 1:200 in TSB + glucose and aliquoted into 96-well polystyrene flat-bottomed microtitre plates from Corning. After 24 h at 30°C (the permissive temperature for the pBT9 vector that was present in strains used in the biofilm assays), the wells were emptied and washed twice with phosphate-buffered saline (PBS). The plates were dried at ambient temperature, stained for 30 s with safranin, washed under gently running tap water and scanned using a digital scanner. The stained biofilms were resuspended in 100 µl of PBS by gentle sonication, transferred to new microtitre wells, diluted 1:4 in PBS and analysed by spectrophotometry at OD450 using an enzyme-linked immunosorbent assay (ELISA) reader.
RNA slot-blot analysis
Staphylococcus aureus cultures were grown in TSB + 1% glucose at 30°C for 16 h. RNA was extracted from 109 cells using the RNeasy miniprep kit (Qiagen) as described in the manufacturer's instructions except that 0.5 mg ml−1 lysostaphin was used in place of lysozyme to lyse the cells. RNA samples were treated with DNase, RNA concentrations were determined by absorbance at 260 nm, and 4 µg of each sample was immobilized on a nylon membrane. The single icaADBC transcript was detected by hybridization analysis essentially as described by Cramton et al. (1999) except that the DNA probe was labelled using the ECL direct nucleic acid labelling and detection system (Amersham). A DNA probe for the icaR transcript was generated by PCR using the following primer pair: icaRFwd (TTTCTTCAAAAATATGTTTAGTAGC GAATACAC) and icaRRev (AAGGATAAGATTATTGATAACG CAATAAC).
Preparation of S. aureus lysates for gel electrophoretic mobility shift assays (EMSAs)
TSB (500 ml) was inoculated with 5 ml of an overnight culture of S. aureus MN8. Cultures were well aerated and grown at 37°C for 5 h (OD600 = 0.9). Cells were lysed essentially as described by Fournier et al. (2000); they were first washed in 25 ml of buffer A [20 mM Tris-HCl, 50 mM MgCl2, 1 mM dithiothreitol (DTT), 0.1 mM EDTA, 5% glycerol], frozen for 2 h at −80°C and lysed in 5 ml of buffer A containing 100 µg ml−1 lysostaphin and EDTA-free Complete Protease Inhibitors (Boehringer Mannheim) for 3.5 h on ice. Buffer A and 1.3 M KCl (3 ml) were added, and the incubation was continued on ice for 30 min. Lysates were cleared by centrifugation, filter sterilized and dialysed through a 3500 MWCO membrane (Pierce) overnight against 4 l of 10 mM Hepes, 1 mM MgCl2, 0.5 mM DTT at 4°C.
Oligonucleotide probes for EMSAs
Double-stranded oligonucleotides were generated by combining a primer with its reverse complement (10 pM each) in 50 mM Tris-HCl + 50 mM NaCl, warming the reaction to 80°C and allowing it to cool at 1°C min−1 to 25°C. The following three primers were used: WTshort (CTATGTTACAGGAAAAT TAAGTTGCAATTACAAATATTTCCGTTTAATTATAA), SUB (CT A T GTT ACAGGAAAATTAAGTTGCAATTAAAAATAAACC GTTTAATTATAA) and MUC (CTATGTTACAGGAAAATTA AGTTGCAATTACAAACCGTTTAATTATAA). The 198 bp probe was generated by PCR using genomic DNA from strain MN8 as a template obtained from either S. aureus MN8 to yield the WT probe or S. aureus MN8m to yield the MUC probe using the following primer pair: icaFWD (ATTGCGT TATCAATAATCTTATCCTTC) and icaREV (TTGCAATTTCTT TACCTACCTTTC). The non-specific competitor used in gel shift assays was also a 198 bp PCR product representing a sequence from the icaA gene and was amplified from MN8 total DNA using the following primer pair: icaA-FW (CCTG TATTTATGTCTATTTACTGG) and icaA-Rev (CTTCTCG TATTTGAGTGCAAG). The PCR products were purified using the Qiagen PCR purification kit. For gel shift analysis, the double-stranded probes were labelled with [γ-32P]-ATP using T4 kinase.
Electrophoretic mobility shift
Gel shift assays were performed essentially as described by Fournier et al. (2000). A 20 µl binding reaction containing 0.25–1 µg of protein from the cell-free lysate, or 1 µg of purified recombinant IcaR, 1 µg of sonicated salmon sperm DNA and 1 µg of poly-(dI–dC) in binding buffer (10 mM Hepes, 60 mM KCl, 4 mM MgCl2, 0.1 mM EDTA, 0.1 mg ml−1 BSA and 0.25 mM DTT) was incubated at 21°C for 10 min before adding 1 µl of (20 000 c.p.m.) radiolabelled probe. The reaction was incubated for an additional 15 min, loaded onto a 5% non-denaturing polyacrylamide gel and electrophoresed in prechilled 1× Tris/borate/EDTA (TBE) at 360 V for 1 h. Gels were dried and exposed to radiographic film overnight at −80°C.
DNase I footprint analysis
Footprinting was carried out essentially as described by Sandaltzopoulos and Becker (1994). icaREV (10 pmol) was end-labelled with [γ-32P]-ATP using T4 kinase. The labelled oligonucleotide was EtOH precipitated and used in a 50 µl PCR with 10 pmol of 5′-biotinylated icaFWD and genomic DNA from MN8 as a template for the amplification of the 198 bp probe representing the ica promoter region. The biotinylated, radiolabelled PCR product was immobilized on 500 µg of streptavidin-coated M-280 Dynabeads (Dynal) according to the manufacturer's instructions. The magnetic beads were resuspended in 100 µl of binding buffer (BB; 10 mM Tris-Cl, 5 mM MgCl2, 2 mM DTT, 50 µg ml−1 BSA, 2 µg ml−1 poly-(dI–dC), 0.5 µg ml−1 sonicated salmon sperm DNA, 100 mM KCl). A sample of 10 µl of these oligonucleotide-coated beads, 2.5–10 µg of purified recombinant IcaR in Tris buffer and 25 µl of BB were combined and incubated for 10 min at 21°C. A sample of 5 µl of a DNase I solution (10 mM Tris-Cl, 5 mM MgCl2, 2 mM DTT, 10 mM CaCl2, 100 mM KCl, 2.5 U ml−1 DNase I) was added, and the reaction was carried out for 2 min at 21°C before the addition of 50 µl of 2× stop buffer (4 M NaCl, 100 mM EDTA). A similar solid-phase technique was used for the sequencing control, which was produced as a 198 bp probe by PCR using biotinylated icaFWD and cold icaREV primers that were then immobilized on streptavidin-coated Dynabeads. The beads were heated to 95°C and washed with Tris-EDTA buffer to remove the antisense strand. P32-labelled icaREV was annealed to the immobilized, single-stranded DNA, and sequencing of the antisense strand was carried out using the dideoxyadenosine 5-triphosphate nucleotide mix included in the Sequenase version 2.0 sequencing kit (USB). The beads from the DNase I protection and sequencing samples were resuspended in 95% formamide, 6 µg ml−1 bromophenol blue, heated to 76°C for 5 min and loaded on a 6% prerun sequencing gel at 55 W for 25 min. The gel was dried at 80°C and analysed by autoradiography.
We thank Drs Jean Lee and Christine Heilmann for their invaluable advice and assistance with cloning and genetics in S. aureus. This work was supported by NIH grants AI46707, 5T32AI07410 and F32AI51892. S.E.C. was supported by NRSA Postdoctoral Fellowship AI09626 from the National Institute of Allergy and Infectious Diseases and by the Deutsche Forschungsgemeinschaft (FOR:449/1).