Regulation of Rot Expression in Staphylococcus aureus

Abstract

Repressor of toxins (Rot) is known to be a global regulator of virulence gene expression in Staphylococcus aureus. The function of Rot, but not the transcription of rot, is regulated by the staphylococcal accessory gene regulator (Agr) quorum-sensing system. In addition, the alternative sigma factor (σ^B) has a repressive effect on rot expression during the postexponential phase of growth. The transcriptional profiles of Rot in σ^B-positive and σ^B-negative strains in the postexponential and stationary phases of growth were compared. An upregulation of rot expression was observed during the stationary phase of growth, and this upregulation occurred in a σ^B-dependent manner. The effects of other staphylococcal transcriptional factors were also investigated. Electrophoretic mobility shift assays revealed that proteins present in staphylococcal lysates retarded the mobility of the rot promoter fragment and that the effect was reduced, but not eliminated, with lysates from strains lacking a functional SarS protein. A modest upregulation of rot expression was also observed in sarS-negative strains. Affinity purification of proteins binding to the rot promoter fragment, followed by N-terminal protein sequencing, identified the SarA and SarR proteins. Primer extension analysis of the rot promoter revealed a number of discreet products. However, these RNA species were not associated with identifiable promoter activity and likely represented RNA breakdown products. Loss of Rot function during the postexponential phase of growth likely involves degradation of the rot mRNA but not the inhibition of rot transcription.

Staphylococcus aureus is an important pathogen in food poisoning and nosocomial and community-acquired infections. Currently, the increasing case numbers of multiply antibiotic-resistant staphylococcal infections in the hospital environment have made this organism a public health concern. The virulence factors of this organism include a variety of exoproteins and cell wall-associated components (24). The secreted components include enterotoxins (A to E and G to R); leukotoxin; exofoliative toxins; alpha-, beta-, gamma-, and delta-hemolytic toxins; toxic shock syndrome toxin; coagulase; and secreted enzymes such as nuclease and proteases (24-27). Coordinated regulation of expression of these virulence genes is an important feature of the pathogenicity of S. aureus. In addition, the regulatory networks might provide sites of possible therapeutic intervention in the treatment of staphylococcal infections.

Several loci have been reported to be global regulators for expression of various virulence factors, including the accessory gene regulator (Agr), staphylococcal accessory regulator (SarA), and repressor of toxins (Rot) (12, 32, 43). Among these regulatory systems, the Agr system has been the best characterized. The Agr system consists of two transcripts RNAII and RNAIII, which are transcribed from P2 and P3 promoters (12). RNAII encodes the structural components of the quorum-sensing system, including AgrBDCA (30). AgrC is a transmembrane protein functioning as a histidine kinase, which is the sensory component of the two-component regulatory system (13). When a threshold concentration of the autoinduction peptide is detected in the environment, AgrC undergoes autophosphorylation (11). This AgrD-encoded autoinduction signal peptide is processed and transported by AgrB (45). The phosphorylated AgrC transduces the information to a response regulator, AgrA (21). Activated AgrA transcriptionally activates the P2 and P3 promoters increasing the RNAII and RNAIII levels. Although RNAIII encodes delta-toxin, it is the RNAIII molecule itself that is the regulatory molecule of the Agr system.

The S. aureus genome encodes a number of transcriptional factors, the Sar (staphylococcal accessory regulator) family of proteins, including SarA, SarR, SarS, SarT, SarV, and Rot (2). SarA is a transcriptional activator for the Agr system (3), as well as a transcriptional regulator that activates or represses a number of staphylococcal genes (4). SarA, for example, is a repressor of spa (staphylococcal protein A) and an upregulator for the fibronectin-binding protein A (6, 44). SarA is also required for biofilm formation (42). SarR binds to the sarA promoter region to downregulate transcription from its P1 promoter and thus reduces SarA protein expression (15). SarS is a positive regulator of spa transcription (6, 38). SarT has been shown to positively regulate sarS transcription and negatively regulates expression of hla (which encodes alpha-hemolysin) and sarU (16). SarU is proposed to be a positive regulator of agr expression (16). SarV is thought to be an important regulator in the autolytic pathway of S. aureus (18). SarX is a negative regulator of Agr (17). MgrA has been shown to be an activator of microcapsule synthesis, nuclease expression, and norA transcription but represses the expression of alpha-toxin, coagulase, protease, protein A, and certain genes involved in autolysis (11, 18, 39).

Rot was identified as a negative transcription regulator of alpha-hemolysin and protease expression (20). It was subsequently shown by gene array analysis to be a global regulator of gene expression in S. aureus, with 86 genes activated and 60 genes repressed by Rot (32). The regulation of Rot production is not well understood. Although Rot was first identified as a repressor of certain Agr system-regulated exoprotein genes, the transcription of Rot is not regulated by the Agr system in the postexponential phase of growth. It has been reported that Rot is an activator of its own transcription (40) and that expression of Rot is not growth phase dependent (32). However, the translation of Rot is negatively regulated by the Agr system through mechanisms involving cleavage of the rot mRNA (1, 7). In the present study, the regulatory relationships among Agr, SarA, SarS, SigB, and Rot in rot promoter activation were investigated.

MATERIALS AND METHODS

Media and culture conditions.

Escherichia coli cultures were grown in L broth (1% NaCl, 1% tryptone, and 0.5% yeast extract) with shaking at 200 rpm or on L agar (L broth with 1.5% agar) at 37°C overnight. S. aureus cultures were grown in tryptic soy broth (TSB; Difco) with shaking at 200 rpm or on tryptic soy agar (Difco) at 37°C overnight. The bacterial strains and plasmids used are listed in Table 1. The antibiotic concentrations utilized were as follows: ampicillin at 100 μg/ml, erythromycin at 400 μg/ml for E. coli and at 20 μg/ml for S. aureus, kanamycin at 25 μg/ml, tetracycline at 20 μg/ml for E. coli and at 10 μg/ml for S. aureus, and zeocin at 20 μg/ml for S. aureus.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Relevant phenotypes	Source or reference
Strains
Escherichia coli
DH5α	φ80dlacZΔM15 recA1 endA1 gyrA96 thi-1 hsdR17 (rK− mK−) supE44 relA1 deoR Δ(lacZYA-argF)U169	Gibco-BRL
MUS3566	AbleK pREP4, pDT112	41
TOP10F′	F′ [lacIq Tn10 (Tetr)] mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74 deoR recA1 araΔ139 Δ(ara-leu)7697galU galK rpsL endA1 nupG	Invitrogen
Staphylococcus aureus
GP269	rsbUVW sigB-tetL (SigB positive)	8
KSI2054	agr+	35
KSS5186	KSI2054 Δagr rot::tetM	40; this study
KSS5187	KSI2054 sarS::erm	38; this study
KSS5188	Δagr sarS::erm	This study
KSS5477	sarS::erm rot::tetM	This study
KSS5488	Δagr sarS::erm rot::tetM	This study
KSS5523	Δagr sarA::kan	This study
KSS5524	rsbUVW sigB -tetL sarA::kan	This study
KSS5525	sarA::kan rot::tetM	This study
KSS5526	Δagr sarA::kan rot::tetM	This study
KSS5527	sarA::kan sarS::erm	This study
KSS5528	Δagr sarA::kan sarS::erm	This study
KSS5529	sarA::kan sarS::erm rot::tetM	This study
KSS5530	Δagr sarA::kan sarS::erm rot::tetM	This study
KSS5595	rsbUVW sigB -tetL Δagr	This study
KSS5596	rsbUVW sigB -tetL Δagr sarA::kan	This study
PM466	Δagr	20
PM783	rot::tetM	20
RN4220	Accepts foreign DNA (r−)	12
UAMS979	sarA::kan	36
Plasmids
pDT41	rot promoter element in pMH109	41
pDT113	R1 rot promoter deletion in pMH109	This study
pDT114	R2 rot promoter deletion in pMH109	This study
pDT115	R3 rot promoter deletion in pMH109	This study
pDT117	R4 rot promoter deletion in pMH109	This study
pDT118	R5 rot promoter deletion in pMH109	This study
pDT119	R6 rot promoter deletion in pMH109	This study
pHH3735	C-terminal His-tagged rot in pMK4	This study
pMH109	Promoter cloning plasmid	9

DNA and genetic manipulations.

Chromosomal DNA was isolated from S. aureus as previously described (29). DNA fragments were generated by PCR, and the primer pairs utilized are listed in Table 2. PCR amplified promoter elements were cloned into pMH109 (9) after SacI and XbaI digestion. Subsequently, the cloned DNA fragments were subjected to DNA sequence analysis to confirm the identity of the inserted sequence. Plasmids were isolated from E. coli by using a Wizard Plus SV miniprep kit (Promega). Plasmids were introduced into S. aureus strain RN4220 by electroporation (6) and then transduced into other S. aureus strains with different genetic backgrounds using phage 80α (31). Plasmid DNA was isolated from S. aureus strains, and the inserts were resequenced to verify their identity.

TABLE 2.

Primers used for PCR amplification

Primer	Primer sequencea	Purpose
CAT3	GGTTATACTAAAAGTCGTTTGTTGGTTC	Primer extension
DTrev	GCGAGCTCAACTTGTATGTGGTAACTTATGC	R5+R6 deletions
lac5pX	GCTCTAGACTGCATTGTTAAAATATGTATC	EMSA
lac3pSc	GCGAGCTCGTTTAAATTATAAACATAAAC	EMSA
Protbio	5Bio-tctagaCAGTAGATGCTCATCTTTTTTTAG	Affinity purification
R13pSc	GCGAGCTCGAAAATATATCAATATTACCATTAAATTG	R1 deletion
R23pSc	GCGAGCTCATTTTATAATTTATATACAAAG	R2 deletion
R33pSc	GCGAGCTCAGAAATAAGATAATAGTAC	R3 deletion
R43pSc	GCGAGCTCGTTGAATAAAATTAAATAAG	R4 deletion
R55pX	GCTCTAGATATTCAACTTTGACAATTGAATAG	R5 deletion
R65pX	GCTCTAGACTTATTTCTAAATATTAACTC	R6 deletion
rot3pSc	GAGCTCGTGTAAATAAACTTGCTTTC	EMSA
rot5pX	CGGGATCCTTTGTTAATACTTGTATAG	EMSA
rot5p2X	TCTAGACAGTAGATGCTCATCTTTTTTTAG	R1-R4 deletions
RotHis3pS	GTCGACTTAGTGATGGTGATGGTGATGCACAGCAATAATTGCGTTTAAAC	Construction of pHH3735
RotHis5pX	TCTAGACAGTAGATGCTCATCTTTTTTTAGAACTTT	Construction of pHH3735
Rotprom3p	GCTAAACATCTCCCAATTAATAATC	Affinity purification

^aUnderlining signifies bases added to primers to create restriction endonuclease recognition sites.

Determination of the N terminus of the Rot protein.

A His-tagged Rot derivative was constructed by PCR amplification of a DNA fragment encompassing the intact rot promoter region, the rot open reading frame, and a C-terminal six-histidine tag using the primers RotHis3pS and RotHis5pX (Table 2). The PCR fragment was cloned into pSC-A (Stratagene Corp.) and subcloned into the shuttle vector pMK4 (37) to generate pHH3735. The insert was confirmed by DNA sequencing at the MU DNA core facility, and the plasmid was electroporated into S. aureus strain RN4220 (12). The C-terminal His-tagged Rot protein was expressed from its native transcription and translational signals. The S. aureus expressed protein was purified using the His-Spin purification system (Zymo Research), and the protein was detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using the SuperSignal West HisProbe kit (Pierce Chemical Co.). The purified protein was subjected to SDS-PAGE, transferred to a polyvinylidene difluoride membrane, and stained with Coomassie brilliant blue, and the N-terminal amino acid sequence was determined at the Protein Facility at the Iowa State University Office of Biotechnology.

CAT assays.

Chloramphenicol acetyltransferase (CAT) assays were performed by the spectrophotometric method of Shaw (33), modified to a microplate format (41). Bacterial cultures were grown in TSB with appropriate antibiotic selection overnight at 37°C. Then, 10 ml of prewarmed TSB (37°C) was inoculated with an overnight culture to an A₅₄₀ of 0.2 unless otherwise indicated. The cultures were then incubated at 37°C with shaking, until the A₅₄₀ reached 2.5, unless otherwise indicated. A total of 5 ml of each bacterial culture was collected, the cells were harvested by centrifugation (5,000 × g, 10 min), and the cell pellets were washed with TE buffer (50 mM Tris-HCl, 10 mM EDTA [pH 8.0]). The cell pellets were resuspended in 1 ml of TE buffer, and the cells were then lysed with 0.1-mm glass beads, thrice for 1 min at 4°C using an eight-sample bead beater (Biospec Products) with a 1-min cool-down interval between each cycle. The lysed bacterial samples were centrifuged (2,500 × g, 10 min) at 4°C, and the supernatant was saved and stored at −70°C. Then, 2 to 25 μl of cell lysate was added to 37.5 μl of 0.4% of 5,5′-dithiobis-2-nitrobenzoic acid (DTNB; Sigma) in 100 mM Tris-HCl (pH 8.0), and 7.5 μl of 5 mM acetyl coenzyme A (Amersham Pharmacia). Distilled water was added to give a final volume of 250 μl. The reaction mixtures were then incubated at 37°C for 10 min prior to the addition of 5 pmol of chloramphenicol (in 10 μl of 50% ethanol). The changes of absorbance in 412 nm were measured by using a microplate reader (Molecular Devices). The results were normalized with respect to the total protein concentration of the cell extracts as determined by using a Bio-Rad protein assay kit. CAT values were calculated as the change in absorbance per minute divided by 13.6 (the molar extinction value for DTNB) and by the amount of protein added. CAT values were expressed as nanomoles of chloramphenicol acetylated per minute per milligram of protein.

Primer extensions.

A total of 10 ml of TSB with antibiotics was inoculated with the bacterial culture at 37°C overnight with shacking at 200 rpm. The overnight culture was used to inoculate fresh prewarmed 20 ml of TSB to an A₅₄₀ of 0.1. The cultures were then incubated at 37°C with shaking at 200 rpm until the sample reached an A₅₄₀ of 2.5. A 10-ml portion of culture was transferred into a tube containing 10 ml of ice-cold ethanol-acetone (1:1) mixture. The samples were then collected by centrifugation at 5,000 × g for 10 min at 4°C. Cells were twice washed with 10 ml of TES buffer (30 mM Tris-HCl, 50 mM NaCl, 2.5 mM EDTA [pH 8.0]) and resuspended in 200 μl of TES buffer. Lysostaphin was added to a final concentration of 5 μg/ml. Then, 80 μl of the suspension was placed in a fresh 1.5-ml microfuge tube, and the samples were incubated at 37°C for 30 min. Total RNA was purified by using RNA-Bee (Tel-Test, Inc.) (34). The CAT3 primer was labeled with [γ-³²P]ATP using T4 polynucleotide kinase at 37°C for 1 h. Primer extension reactions were performed by using an avian myeloblastosis virus primer extension kit (Promega). A total of 30 μg of RNA was mixed with 100 fmol of labeled primer and 40 U of RNasin (Promega). The reaction mixtures were then incubated at 70°C for 5 min for denaturation, followed by 53°C for 20 min for primer annealing. The samples were then incubated at 42°C for 45 min. The resulting DNA products were analyzed by electrophoresis in 6% polyacrylamide-8 M urea gels, followed by radioautography.

Identification of DNA-binding proteins.

A 100-ml portion of TSB was inoculated with 1 ml of an overnight culture (initial optical density at 540 nm of 0.05), followed by incubation at 37°C until the optical density at 540 nm reached 2.0 (postexponential phase of growth). The cells were harvested by centrifugation, and the cell pellets were washed with 10 ml of TE buffer. The supernatant was removed, and the cell pellets were kept at −80°C overnight. The cell pellets were resuspended with 500 μl of DNase I reaction buffer (10 mM Tris-HCl, 2.5 mM MgCl₂, 0.5 mM CaCl₂ [pH 7.6]). The cells were lysed by bead beating with 0.1-mm beads consisting of three cycles of 1 min bead beating with a 1-min cooling interval at 4°C. The cell lysates were clarified by centrifugation, and the supernatants obtained were then treated with DNase I (2 U/μl) and RNase A (5 mg/ml) for 1 h at room temperature. The cell lysate was then dialyzed overnight against phosphate-buffered saline (PBS; pH 7.4) at 4°C and then divided into aliquots and stored at −80°C.

The rot promoter-containing DNA fragment was amplified by PCR using primers Protbio and Rotprom3p with the upstream primer biotinylated, and PCR fragments were purified with the QiaQuick PCR purification kit (Qiagen). The PCR fragments were bound to 0.5 ml streptavidin agarose (Sigma) and placed in a column. The staphylococcal protein extract (1 mg) in 1 ml of PBS containing 1 μg of poly(dI-dC) was added to the column, followed by incubation for 15 min at room temperature. The column was washed twice with 500 μl of PBS containing salmon sperm DNA (5 mg/ml) and five times with 500 μl of PBS and then eluted with 500 μl of PBS plus 500 mM NaCl. After dialysis overnight against distilled water, the samples were concentrated by vacuum centrifugation. Protein concentrations were determined by using a Nanodrop spectrophotometer, and proteins bound to the DNA were resolved by SDS-PAGE on 15% polyacrylamide gels and detected by silver staining. In parallel, the affinity-purified Rot promoter-binding proteins were subjected to SDS-PAGE, transferred to a polyvinylidene difluoride membrane, and stained with Coomassie brilliant blue, and the N-terminal amino acid sequence was determined at the protein facility at the Iowa State University Office of Biotechnology.

EMSAs.

The DNA fragments utilized in the electrophoretic mobility shift assay (EMSA) experiments were amplified from S. aureus chromosomal DNA utilizing the primers indicated in Table 2. The rot DNA fragment was a 400-bp fragment terminating 11 bases upstream of the first putative ATG codon shown in Fig. 1. The staphylococcal lac promoter fragment was a 250-bp DNA fragment upstream of the lacA determinant (29). DNA fragments were end labeled with [γ-³²P]ATP using T4 polynucleotide kinase. Cell lysates were prepared from postexponential-phase cultures (A₅₄₀ = 2.5) unless otherwise indicated. The cell pellets were washed and resuspended in TE buffer. Cell suspensions were then standardized to contain 5 mg of cells (dry weight)/ml in lysis buffer (50 mM Tris-HCl [pH 8], 10 mM EDTA, 10% glycerol). The cells were lysed by beating them twice for 1 min using 0.1-mm diameter glass beads in a bead beater. Cell debris was then removed by centrifugation (2,500 × g for 10 min. The binding reactions were performed according to a previously described protocol in a buffer system containing 10 mM Tris-HCl (pH 7.5), 1 mM EDTA, 50 mM NaCl, and 5% glycerol (14).

FIG. 1.

Sequence of the 5′ end of the rot open reading frame. The sequence was taken from GenBank accession number AF189239. Potential ATG start codons are in boldface. The putative ribosome-binding site sequence is underlined. The deduced amino acid sequence is shown below the nucleotide sequence, and the N-terminal amino acid sequence obtained is shown in boldface.

Statistical analysis.

Each result was represented as the mean ± the standard deviation from at least four independently collected samples. At least four sets of independent experiments were performed. The results were subjected to pair-sampled two-tailed Student t tests. Results with P values of <0.05 were considered significant.

RESULTS

Verification of the N terminus of the Rot protein.

Examination of the DNA sequence of the rot open reading frame reveals several in-frame ATG codons that could serve as potential start codons for Rot translation (Fig. 1). The fourth of these was identified by Oscarsson et al. as the correct site based upon electrophoretic mobility (28). This translational start site was supported by deletion analysis conducted by Geisinger et al. (7). This start site has a reasonable ribosome-binding site sequence upstream of the start site. However, the actual rot open reading frame had not been identified with certainty. To confirm the translational start site of the rot determinant, the open reading frame was fused at its carboxy terminus to six histidine residues and the construct, containing the native transcriptional and translational signals, was introduced into S. aureus. The Rot protein was expressed and purified using the His tag. The purified protein was subjected to SDS-PAGE and N-terminal amino acid sequence determination. The N-terminal sequence was determined to be M-K-K-V-N-N-d-T-V-F, indicating that the fourth ATG site was indeed the true translational start codon of the rot open reading frame. Only a single species was identified by Western blotting using the Pierce SuperSignal West His probe kit (data not shown), indicating that only one translational start site was demonstrable during the in vitro culture conditions utilized.

Expression profile of the rot promoter.

The intact rot promoter was positioned in front of a cat reporter gene (insert extending 554 bp upstream of the start codon; pDT41), and the rot promoter activity was assayed in the regulatory mutant strains of S. aureus. No significant differences in the rot promoter activity in both the exponential and the postexponential-phase samples between wild-type and Agr-null strains were observed (Fig. 2). This is consistent with previous findings (40). Expression of rot was found to be biphasic, with elevated expression seen in the exponential and stationary phases of growth and a reduced (2- to 2.5-fold) expression during the postexponential phase of growth. The decrease in postexponential-phase expression was found to be SigB dependent, since reporter expression remained elevated in the SigB-negative hosts. Furthermore, the recovery of expression during the stationary phase of growth was found to be Agr dependent. Thus, the Agr quorum-sensing system does have an impact on rot transcription, but only during the stationary phase of growth. Modest regulation of rot transcription occurs when the cells transition from the exponential phase to the postexponential phase of growth and again when the culture enters the stationary phase of growth.

FIG. 2.

rot promoter activities in SigB⁺ Agr⁺, SigB⁺ Agr⁻, SigB⁻ Agr⁺, and SigB⁻ Agr⁻ strains from exponential, postexponential, and stationary growth phases. Overnight culture was used to inoculate a fresh TSB to an A₅₄₀ of 0.1. Samples were collected when the A₅₄₀ reached 0.3 for exponential phase (□) and 2.5 for postexponential phase (▪), and overnight cultures were taken for the stationary phase of growth (▧). The promoter activity is expressed as nanomoles of chloramphenicol acetylated per minute per milligram of protein.

Promoter activities analyses of rot in different transcription regulator mutant background isogenic strains.

Transcription of rot has been reported to be regulated by Rot, SigB, and SarA (19, 40). To investigate the roles of the staphylococcal transcriptional regulators on rot expression, we transduced plasmids containing the rot promoter into strains with different genetic backgrounds. The resulting postexponential-phase promoter activities are presented in Fig. 3. The SigB-associated twofold effect on rot transcription was evident. Inactivation of SarS resulted in a twofold increase in postexponential-phase rot transcription, but only in an Agr⁺ background. In the Agr⁻ background, a twofold reduction in rot expression was observed. The SarS-associated effects on rot expression were dependent upon the presence of a functional SarA protein. In a SarS⁻ and SarA⁻ background, the fluctuations in rot expression were minimal. It is not known whether the effects of SarS on expression from the rot promoter were due to a direct binding of SarS to the rot promoter or if the effects were indirect. Rot has been reported to be a positive regulator of SarS expression (32), and thus the loss of Rot should partially mimic the SarS-negative phenotype. However, no significant effect of Rot on rot expression was observed (compare Rot⁺ Agr⁺ with Rot⁻ Agr⁺ and Rot⁺ Agr⁻ with Rot⁻ Agr⁻ in Fig. 3). We did not, however, observe an increase in rot promoter activity in the SarA-deficient strain as reported by Manna and Ray (19). With our promoter fusions, the SarA effect was only detected in the Agr⁺ SarS⁻ host. In this background, the effect was a reduction in rot promoter activity rather than an elevation.

FIG. 3.

Rot promoter activities from postexponential-phase cultures of different mutant strains. The results are represented as means ± the standard deviation from a set of three independently collected samples. The promoter activity is expressed as nanomoles of chloramphenicol acetylated per minute per milligram of protein.

EMSAs.

To determine whether proteins are present in the staphylococcal cytoplasm that are capable of binding to the rot promoter fragment, EMSAs were conducted. The mobility of the 400-bp rot promoter-containing DNA fragment was retarded when incubated with cell lysates collected from Agr⁻, Agr⁻ SarA⁻, Agr⁻ SarS⁻, Agr⁻ Rot⁻, Agr⁻ SarA⁻ SarS⁻, Agr⁻ SarA⁻ Rot⁻, and Agr⁻ SarA⁻ SarS⁻ Rot⁻ strains (results are shown for Agr⁻ SarS⁻ and Agr⁻ SarA⁻ SarS⁻ Rot⁻ lysates in Fig. 4). Proteins were present in the Agr⁻ strain lysate that bound to and retarded the mobility of the rot promoter fragment. The use of lysate prepared from an isogenic SarS-negative strain resulted in a substantially reduced electrophoretic mobility retardation, suggesting that SarS, or protein(s) whose presence requires a functional SarS protein, was responsible for much of the binding observed with the SarS-positive lysate. The lysate prepared from a mutant strain lacking the SarS, SarA, and Rot proteins gave essentially the same pattern as that of the SarS-deficient lysate, indicating that Rot and SarA were not the proteins responsible for the residual electrophoretic mobility retardation observed under these conditions. Control lysates prepared from E. coli DH5α and B. subtilis strain 168 did not exhibit mobility retardation of the rot promoter fragment (data not shown).

FIG. 4.

EMSA using cells lysates from S. aureus mutant strains. The rot promoter fragment was incubated with cell lysates collected from Agr⁻, Agr⁻ SarS⁻, and Agr⁻ SarA⁻ SarS⁻ Rot⁻ strains. The labeled DNA fragment was incubated with no lysate (−) or with 2, 4, or 6 μl of the Agr⁻ strain lysate and with 2, 4, 6, or 8 μl of the indicated cell lysates.

Identification of rot promoter-binding proteins.

To identify proteins that might be involved in the transcriptional regulation of rot, proteins with affinity for the promoter region of this determinant were isolated. Two species reproducibly bound to this DNA fragment, and a larger protein was occasionally observed (Fig. 5). The identity of the 65-kDa protein that was occasionally detected by this approach is under investigation. The two smaller species were identified as SarA and SarR by N-terminal sequence determination. SarA has recently been shown to bind to the rot promoter by EMSAs (19). SarR, a regulator of sarA expression (15), has not been previously identified as being involved in the regulation of rot expression. The promoter activity was not significantly altered in a sarR mutant strain (19). However, Manna and Cheung identified multiple putative SarR binding sites within the sarA promoter region by using a nuclease protection assay (15). The sarA P2-associated binding site consisted of the sequence TAAATTAA-(12 bp)-ATAATTTA. The other two sarA promoters contained only a single repeat (15). A sequence matching the P2 SarR putative binding site can be found in the rot promoter 402 bp upstream of the initiation codon. The sequence, TAAATTTA-(11 bp)-TTAATTTA matches in seven of the eight positions in each arm of the sarA P2 repeat.

FIG. 5.

SDS-PAGE analysis of proteins bound to promoter DNA fragments. Proteins from S. aureus GP269 (Agr⁺ and SigB⁺) bound to the rot promoter fragment. Lanes: 1, lysate proteins from the flow through; 2, PBS eluate; 3, high-salt eluate; 4, protein molecular weight markers (Invitrogen). Arrows indicate the positions of the SarA and SarR proteins.

Because SarS was not identified in this binding assay, it suggests that the effects on rot expression associated with SarS, both in the reporter expression assays and the lysate EMSA, may result from indirect effects of SarS on other transcriptional regulators.

Transcription profile of the rot promoter.

Results from the rot expression profile presented above indicated that SigB-associated differences in rot expression were evident, especially in the stationary phase of growth. Therefore, it was possible that the rot promoter contains both σ^A- and σ^B-dependent promoter elements. The nature and the number of promoters upstream of rot have not been definitively established, although multiple transcriptional start sites have recently been predicted (19). To determine whether multiple transcription start sites are present and whether they are influenced by the SigB status of the cells, a primer extension approach was utilized. The primer extension results from total RNA isolated from postexponential- and stationary-phase cultures of SigB⁺ and SigB⁻ strains were compared in Fig. 6A. A number of possible transcripts were identified from postexponential- and stationary-phase SigB⁺ and SigB⁻ cultures and, among them, the moderate and major species were designated “a” though “e” (Fig. 6A). The postexponential- and stationary-phase patterns showed distinct differences in the SigB-positive versus SigB-negative hosts. Both hosts, however, yielded a prominent species (labeled “a”) that was more strongly expressed in the postexponential growth phase and produced a stronger signal from the SigB-negative strain RNA. This species initiates 293 bp upstream of the rot initiation codon and at a site positioned downstream from a good match to a consensus σ^A promoter (TTGCAA and TATATT separated by 17 nucleotides; Fig. 6B). The smallest species (“e”) was more pronounced with the RNA sample from the SigB-positive strain in the stationary phase and only faintly present with the corresponding sample from the SigB-negative strain. However, there is not a good match to a SigB consensus sequence immediately upstream of this site (5). The lack of a SigB consensus sequence suggests that if species “e” represents a start site of transcription, the modulation by SigB may be an indirect effect, rather than involving transcription by the SigB-bearing RNA polymerase initiating at this site.

FIG. 6.

Primer extension of the rot promoter containing DNA fragment. (A) Total RNA was collected from SigB⁺ and SigB⁻ strains from postexponential-phase cultures (A₅₄₀ = 2.5) and stationary-phase cultures (A₅₄₀ = 5). Lanes: I, postexponential-phase SigB⁺; II, stationary-phase SigB⁺; III, postexponential-phase SigB⁻; IV, stationary-phase SigB⁻. Letters along the right-hand side refer to the prominent species identified in panel B. (B) DNA sequences from the corresponding region. Arrows: large curved arrow, major species; small curved arrow, moderate species; thin right-angled arrow, minor species. The underlined sequences denote the −35 position and Pribnow box sequences associated with the expression of the rot transcript “a”.

Do the species identified by primer extension represent transcription start sites for rot?

The primer extension approach identifies the 5′ terminus of stable RNA species present in a sample. These could represent start sites of transcription or stable breakdown RNA products of one or more transcripts. To determine whether the “a” through “e” species represented distinct rot transcripts, CAT reporter fusions were created that positioned isolated putative promoter elements in front of the reporter, and these constructs were introduced into S. aureus. CAT assays were conducted at the exponential, postexponential, and stationary growth phases. The results of these studies are presented in Fig. 7. The R2 deletion construct includes only the “a” species identified in the primer extension. This promoter is active in all three growth phases. The expression pattern differed from the intact promoter in that the activity was higher, and there was no decrease in postexponential-phase expression. Because the upstream sequences are present in this construct, the differences most likely result from the loss of the downstream sequences. The R1 deletion, which terminates 4 bp upstream of the “a” transcript start point, showed no reporter gene activity in this assay at any of the growth phases, indicating that the sequence context at the +1 site is critical for the production of this transcript.

FIG. 7.

Deletions of the rot promoter region. (A) Map of the deletion constructs. The lowercase letters refer to the primer extension species indicated in Fig. 5. Minor species are denoted by the small arrows. R1 through R6 contain the putative promoters indicated. (B) CAT assay results for R1, R2, R3, R4, R5, and R6 in strain GP269. Bars: □, exponential phase (A₅₄₀ = 0.4); ░⃞, postexponential phase (A₅₄₀ = 2.5); ▧, stationary phase (A₅₄₀ = 4.5).

Extension of the promoter-containing DNA fragment (deletion constructs R3 and R4) resulted in changes in the magnitude of the CAT assay values, but not the qualitative pattern of expression. The R5 and R6 deletion fragments, which encompass the putative “e” and “b + c + d + e” transcripts, respectively, displayed no promoter activity at any of the growth phases. Thus, these sequences do not appear to carry promoter elements, and the species identified by primer extension must represent relatively stable breakdown products of the rot mRNA. Hence, it can be concluded that there is only a single promoter active under the in vitro growth conditions used and transcription initiating from this promoter produces to the “a” fragment identified in Fig. 6A.

The effect of the SigB sigma factor on the biphasic expression of rot could arise from two effects. SigB may direct the expression of a transcriptional factor that downregulates the expression of rot during the postexponential growth phase. Alternatively, the differences may arise from differences in rot mRNA stability in the isogenic sigB strains. To address this, the R2 promoter region deletion plasmid was introduced into isogenic SigB-positive and SigB-negative strains, and the expression of the cat reporter was measured (Fig. 8). The R2 deletion lacks the “b” through “e” cleavage sites in the 5′ untranslated region of the rot mRNA and thus should not be susceptible to these endonucleolytic events. The overall pattern of rot expression with this construct mimics that of the intact rot promoter, indicating that the single identified rot promoter with its upstream sequences retains the transcriptional control found with the undeleted promoter (pDT41) and the expression profile resulted from transcriptional control rather than posttranscriptional control of the rot mRNA stability. Although the qualitative pattern of expression between the two plasmids was the same, the CAT assays revealed an eight- to ninefold increase in CAT activity with the R2 plasmid relative to pDT41 (compare Fig. 7 and 8). Deletion of the mRNA cleavage sites from the R2 construct may have led to an increased mRNA stability of the R2 transcript. The rot expression pattern was thus found to be transcriptionally regulated, whereas the cleavage of the mRNA was responsible for reduced expression of the reporter genes.

FIG. 8.

Expression of the R2 promoter in SigB-positive (GP269) and SigB-negative (KSI2054) host strains. Overnight cultures were used to inoculate a fresh TSB to an A₅₄₀ of 0.1. Samples were collected when the A₅₄₀ reached 0.5 for the exponential phase (□), 2.5 for the postexponential phase (▪), and 4.0 for the stationary phase (▧) of growth.

DISCUSSION

Rot is an important regulator of virulence gene expression in S. aureus. It functions as a repressor of transcription of a number of exotoxin genes in this pathogen. This Rot-mediated inhibition of exotoxin gene transcription occurs during the logarithmic phase of growth. Rot functions as a positive regulator of sarS (which encodes a positive regulator of staphylococcal protein A gene expression), clfB (clumping factor B), and sdrC (cell surface adhesion) (32). During the postexponential growth phase when the Agr quorum-sensing system is activated, Rot is inactivated, and transcription of the exotoxin genes ensues while the positive effects on adhesion gene expression is lost.

The promoter of the rot determinant is active at all growth phases of S. aureus. Therefore, a significant component of the Agr system's inactivation of Rot must occur posttranscriptionally. Geisinger et al. have demonstrated a mechanism that would explain one possible mechanism for this regulation (6). These researchers demonstrated that RNAIII, which is expressed at high levels in staphylococcal cells upon induction of the Agr system, could associate with the 5′ untranslated region on the rot mRNA through loop-loop interactions, and the association leads to an inhibition of translation of the rot message and to nucleolytic cleavage of the rot mRNA as well. Similar findings have recently been reported by Boisset et al. (1). The interaction of RNAIII with the ribosome-binding site regions of target mRNAs appears to be a common mechanism for inhibition and stimulation of translation with this mechanism having been reported for the hla (stimulation) and spa (inhibition) determinants (1, 10, 22). The spa translation inhibition also parallels the situation with rot in that it is also associated with cleavage of the mRNA (10). The cleavage of the RNAIII-mRNA pairs is thought to be due to the action of RNase III (1, 10).

In the present study we demonstrated that the rot determinant is expressed at all stages of the staphylococcal growth cycle from a single promoter element, and its expression is influenced to a modest degree by SigB and the Agr system. We have identified the single start site of transcription of this determinant. We also detected RNA species associated with the 5′ untranslated region of the rot mRNA that are sufficiently stable to be detected in a primer extension analysis. Secondary structure predictions of the rot message from the +1 site (species “a” in the primer extension) through the untranslated region containing the 5′ termini of the primer extension-identified fragments revealed a potential stem-loop structure shown in Fig. 9. Interestingly, the 5′ termini of the major species identified in Fig. 6 corresponded to sites centered around a predicted stem-loop structure, suggesting that the putative RNase responsible for processing the rot mRNA targets this domain in the rot message. Since the rot domain identified by Geisinger et al. is not contained within our full-length promoter construct (pDT41), this instability of the 5′ end of the mRNA is either independent of the interaction with RNAIII or additional RNAIII interactive sites exist further upstream of the rot ribosome binding site region. Because the presence of the RNA breakdown products is more pronounced in the stationary phase of growth, there may be RNA or protein factors interacting with the mRNA to influence cleavage either by preventing the hairpins from forming in the exponential and early postexponential phases or by promoting their formation in the stationary phase of growth. Given that the pDT41 CAT reporter is expressed during both postexponential and stationary growth phases, the cat reporter mRNA is translated. With pDT41, a gram-positive consensus ribosome-binding site is immediately distal to the sequence shown in Fig. 9, and it is then followed by the cat open reading frame. Expression of the reporter at the postexponential and stationary growth phases during which the rot mRNA breakdown products are evident indicates that either the cleavage at the extreme 5′ end of the mRNA does not prevent translation of the message (which retains an intact ribosome-binding site) or only a subset of the mRNAs are cleaved. Removal of the sequences downstream of the start site of transcription (the R2 deletion) resulted in higher levels of expression of the CAT reporter, suggesting that cleavage of this message did not occur and more efficient translation of the reporter mRNA ensued. Although interactions of RNAIII with the ribosome-binding site sequences are clearly the major mechanism for regulation of Rot expression in S. aureus, the results reported here indicate that additional effects in the mRNA sequences upstream of the ribosome-binding site element may also be important in the regulation of Rot expression.

FIG. 9.

Mfold-predicted secondary structure of the rot mRNA from +1 through the rot-cat junction of pDT41 (+153). The 5′ termini of the major species identified by primer extension (“a to e”) are denoted by arrows.

We demonstrated that SarA and SarR were able to bind to the rot promoter region. Manna and Ray found a pronounced increase in rot promoter activity in a sarA mutant strain but demonstrated no effect of sarR inactivation on rot expression (19). However, their promoter construct lacks the putative sarR binding site (their fragment begins 24 bp downstream of this putative SarR binding site) and thus the activities they detected may result from an incomplete rot promoter, one lacking the SarR-associated cis element.