Accessory Gene Regulator Locus of Staphylococcus intermedius
http://www.100md.com
感染与免疫杂志 2006年第5期
Departments of Veterinary Clinical Sciences
Veterinary Basic Sciences, Royal Veterinary College, University of London, London, United Kingdom
ABSTRACT
The accessory gene regulator (agr) locus, a candidate system for the regulation of the production of virulence factors in Staphylococcus intermedius, has been characterized. Using PCR-based genome walking, we have obtained the first complete sequence (3,436 bp) of the accessory gene regulator (agr) gene in this organism. Sequence analysis of the agr gene has identified five open reading frames (ORFs), agrB, agrD, agrC, agrA, and hld. The translated ORF contained amino acid motifs characteristic of the response regulator and histidine protein kinase signal transducer of the classic two-component regulatory system. Sequencing of the agrD PCR products amplified from DNA from 20 different isolates has facilitated detection of genetic variation in the putative autoinducing peptide (AIP) within the agr gene of S. intermedius, revealing the presence of at least three agr specificity groups within this species. Classification of the agr gene from S. intermedius was supported by phylogenetic analysis. Real-time PCR also revealed that the effector molecule of the agr system, RNAIII, was regulated in an autocrine manner in S. intermedius and demonstrated positive correlation with the temporal gene expression patterns of luk and entC. Transcription of RNAIII was also dependent on self secreted cues. Cyclic self and nonself peptides were synthesized on the basis of the novel AIPs produced by S. intermedius, which lack the cysteine necessary to form the thiolactone ring in analogous peptides from Staphylococcus aureus and Staphylococcus epidermidis. Experiments with these synthetic cyclic peptides indicated that self peptides led to up-regulation of RNAIII—findings in support of the assumption that activation of the agr gene is initiated by growth- and species-specific factors generated during bacterial growth.
INTRODUCTION
Staphylococcus intermedius,first described in 1976 (10), is the most common cause of skin infection in dogs, with canine pyoderma being caused almost exclusively by S. intermedius (31). S. intermedius is also frequently isolated from dogs with otitis media and externa (3, 4, 28) and is commonly associated with seborrhoea, a cutaneous scaling disorder in dogs. Little is known, however, about the pathogenesis of S. intermedius. The results of studies investigating the production of virulence factors are unclear, and no distinct differences in the nature of the toxins produced by isolates from healthy dogs and dogs with pyoderma have been found (1). Thus far, there are differing opinions as to the underlying causes of canine pyoderma. Ihrke suggested factors that predispose dogs to staphylococcal pyoderma infections were due to a wide variety of underlying diseases, such as allergy, metabolic diseases, disruption of normal skin flora, immunodeficiency, and endoparasitic infestation (12), whereas according to Greene and Lammler, host-bacterium interactions and the immune status of the host appeared to be the factors most crucial in determining the outcome of infection (9). Many of the factors believed to be involved in S. intermedius virulence have also not been well characterized, and proof of their association with disease pathogenesis is often lacking. It is thus difficult for veterinary clinicians and microbiologists to determine the clinical significance of many S. intermedius isolates.
S. intermedius produces a diverse arsenal of virulence factors involving a large number of cell-bound and extracellular proteins (6, 11, 29, 32), which may be likely contributory factors that predispose dogs to staphylococcal pyoderma. The virulent components and products are thought to share overlapping roles, acting either in concert or alone, and may convert host tissue components into nutrients required for bacterial growth. On the basis of studies performed with corresponding virulence factors in Staphylococcus aureus, considerable knowledge is available about their contribution to the development of infection in humans (21). Less is known about the virulence factors of S. intermedius and their relative importance in canine pyoderma.
In the human pathogen S. aureus, the production of virulence factors is known to be controlled, in part, by the accessory gene regulator system. One of a number of two-component regulatory systems identified thus far, the agr system has received much attention owing to it being assigned, historically, a central role in the model of S. aureus pathogenesis. It is recognized as a quorum-sensing gene cluster that up-regulates production of secreted virulence factors and down-regulates production of cell-associated virulence factors in a cell density-dependent manner (15, 19, 25). When the staphylococci are in lag phase, it is thought that staphylococci initiate infection by synthesizing surface proteins. Once colonization is established, the bacteria multiply and enter exponential phase, activating a density-sensing mechanism that stimulates toxic exoprotein production, thereby enabling them to spread to new sites to prevent overcrowding (30). Synthesis of essential factors only as and when required is important from an economic perspective, as it avoids overwhelming the organism metabolically with unnecessary energy expenditure.
S. intermedius produces many toxins comparable to those of S. aureus, so we hypothesized that it is likely that the production of toxins in the former is under the regulation of a similar global regulator. To determine whether this is so, here we have ascertained the presence of the agr locus and analyzed its sequence. Our starting point was the agrD gene present in each strain of a collection of 20 S. intermedius strains isolated from cases of canine pyoderma. The elucidation of the agr locus facilitated assessment of its regulation during the growth of S. intermedius. In S. aureus, mutation and deletion analyses performed suggested that RNAIII itself, rather than any translation product, is the effector of exoprotein gene regulation (25). We have thereby quantified RNAIII, the effector molecule of agr, together with luk and entC, encoding leukotoxin and enterotoxin C, respectively, using real-time quantitative PCR. We have also synthesized two novel cyclic autoinducing peptides (AIPs), shown to be found in S. intermedius (17), but not in S. aureus or Staphylococcus epidermidis, and used these in functional assays monitoring RNAIII production.
MATERIALS AND METHODS
Bacterial strains and growth conditions. Twenty clinical isolates of S. intermedius were collected from the skin of dogs with pyoderma at the Royal Veterinary College. Long-term storage of the isolates was maintained in 20% glycerol in brain heart infusion broth at –70°C, and cultures for routine use were maintained on sheep blood agar plates. Isolates were identified as S. intermedius according to their colony morphology, biochemical reactions (production of coagulase and DNase), and on the basis of the analytical profile index (API) (bioMerieux, France).
PCR and sequencing. The collection of 20 S. intermedius isolates was screened by direct colony PCR. It was found that the initial hot start of the PCR was sufficient to break down bacterial cell walls from a pinprick colony of S. intermedius, allowing these cells to be used as the direct source of DNA. We used the following primers: 5'-TTCCAATCAATCGGTACT-3' and 5'-TTCCTCCTTCTAACAACTC-3'; their design was based on regions of the partial agr sequence derived from a pigeon isolate of S. intermedius (GenBank accession number AF346723). PCR was carried out in the GeneAmp PCR system 9700 thermocycler (Perkin-Elmer Biosystems, Foster City, California). Amplification was performed in a final volume of 50 μl containing dATP, dCTP, dGTP, and dTTP (each at a concentration of 200 μM), 10 μM of the corresponding primer, 5 μl of 10x Expand high-fidelity buffer (Roche, Nutley, NJ), and 2.6 U of Expand high-fidelity enzyme mix (Roche). The cycling parameters for PCR were as follows: (i) an initial denaturation step of 2 min at 94°C; (ii) 35 cycles, with 1 cycle consisting of 15 s at 94°C, 30 s at 45°C, and 60 s at 72°C; and (iii) a final extension step of 7 min at 72°C. The amplified fragments were cloned into the pGEM-T Easy vector (Promega, Madison, Wis.). Recombinant DNA molecules were transformed into electrocompetent Escherichia coli JM109 cells and sequenced (MWG-Biotech, Ebersberg, Germany).
Genome walking. Genome walking was initiated from the amplified and sequenced agrD DNA fragment using the Universal Genome-Walker kit (BD Biosciences Clontech, Palo Alto, CA). Total genomic DNA of S. intermedius strain 3 was isolated with Genomic-Tip 100/G (QIAGEN, Crawley, United Kingdom) and digested with DraI, EcoRV, PvuII, and StuI to obtain blunt ends. An aliquot of each DNA sample was run on a 0.7% agarose gel to ascertain quality. After purification, the fragments were ligated to a Genome-Walker adapter provided by the manufacturer (31). The Genome-Walker protocol consists of a series of PCR amplifications, starting initially with the adapter primers (AP1 or AP2) and two nested gene-specific primers designed on the basis of the agrD sequence of S. intermedius (see above), D1 (5'-CCAATCAATCGGTACTTTTGCTAGAATCCCTATTAGTACTGGA-3') and U1 (5'-AGATGAGCCAGAAATCCCTGCCGAGTTGTTAGAAGAGGAC-3'). Sequence extension occurred through further walks using novel gene-specific primers, and the primers were designed on the basis of initial sequence data obtained from previous steps. Care was taken to ensure that the sequences overlapped to some degree. Downstream chromosomal walking was continued with the following primers: D2 (5'-CGTGAAGATGATTTAGAAGGTTTGAAGCAA-3') and D3 (5'-GCTACCAATGACCCTTATGCTATCTTAGAAACTTCT-3'). The upstream region of agrD was analyzed in a similar way using primer U2 (5'-CAACAACTTACGCACTGGTTAACACAAGGAC-3'). Each amplification was performed in a final volume of 50 μl containing 5 μl 10x genomic PCR buffer, 1.1 mM magnesium acetate, 10 mM 50x deoxynucleoside triphosphate (dNTP) mix, 1 μl 50x Advantage genomic polymerase mix (BD Biosciences), and 10 μM of the corresponding primers. A typical PCR cycle consisted of 7 cycles of 94°C for 2 s and 72°C for 3 min, followed by 32 cycles of 94°C for 2 s and 67°C for 3 min, and a final elongation step at 67°C for 4 min. To facilitate PCR amplification of sequences that proved difficult to amplify, we used the Advantage GC genomic polymerase (BD Biosciences).
PCR products from genome walking were purified using the QIAquick gel extraction kit (QIAGEN) and directly sequenced with both primers (adapter and S. intermedius specific). For PCR amplicons that were longer than 0.6 kb, sequencing was performed via primer walking, involving repeated cycles of custom oligonucleotide synthesis and direct sequencing. Sequences obtained were confirmed by comparison of independent PCR experiments with different primers and genomic DNA. Alignments of amino acid sequences from the individual protein-coding genes were realized by using the ClustalW algorithm (http://www.ebi.ac.uk/clustalw) and default gap penalties (28). Regions of sequence that were difficult to align were removed from the data file and the sequences realigned. Ambiguous alignment regions, such as the N and C termini, were excluded from the alignments. Overlapping nucleotide sequences of DNA generated were manually aligned to form a contiguous agr operon and assessed for complete open reading frames (ORFs). DNA and protein sequence similarity searches were performed by the BLASTN and BLASTP (National Center for Biotechnology Information server). Conserved domains in each ORFs were analyzed using reverse position-specific BLAST (rpsBLAST) (v1.65).
Transmembrane topology prediction. The transmembrane topology of the various ORFs of the S. intermedius agr gene was analyzed using ConPredII (http://bioinfo.si.hirosaki-u.ac.jp/ConPred2/), a program based on a consensus approach that combines the results of several prediction methods, such as KKD, TMpred, TopPredII, DAS, TMAP, MEMSAT 1.8, SOSUI, TMHMM 2.0, and HMMTOP 2.0 (2).
Assessment of the regulation of leukotoxin, enterotoxin, and RNAIII transcription in S. intermedius. To examine growth phase-dependent expression, 2-ml cultures of S. intermedius (strain 3) were grown overnight with aeration in brain heart infusion broth at 37°C. Cells were then inoculated from the overnight culture to an initial optical density at 600 nm (OD600) of 0.05 and grown with shaking to the desired phase of growth, e.g., early exponential (OD600 of 0.5, 1 h), mid-exponential (OD600 of 4.5; 4 h), or post-exponential phase (OD600 of 8.0; 8 h). The various growth phases were determined in preliminary experiments (not shown). Growth was monitored turbidimetrically with a UV-160A spectrophotometer (Shimadzu) read at 600 nm. The results of growth phase effects were standardized for the number of bacteria, so that the increase in the total amount of protein is not simply a consequence of increasing number of bacterial cells. At indicated times, samples containing equal numbers of cells were immediately stabilized by the addition of 2 volumes of Bacterial Protect reagent (QIAGEN). The bacterial cells were subsequently harvested after 10 min by centrifugation at 13,000 x g for 10 min at 4°C, prior to RNA extraction.
Characterization of the activating factor for luk, entC, and RNAIII expression in S. intermedius. Cultures of S. intermedius strains 3 and 27 and S. aureus strain J3945 were inoculated from 1-day-old single colonies on sheep blood agar plates into 2 ml brain heart infusion broth. The cultures were grown overnight at 37°C with orbital shaking for 15 h. To synchronize cultures to an active mode of growth and to eliminate potential quorum signals that may have accumulated, 0.5 ml of overnight culture was centrifuged at 3,000 x g for 5 min at 4°C. After the supernatant was discarded and the bacteria were resuspended in fresh prewarmed medium (50 ml), they were grown at 37°C with shaking. Aliquots (10 ml) were harvested at intervals of 1, 4, and 8 h corresponding to the early, mid-, and post-exponential phases of bacterial growth. Bacterial cells were removed by centrifugation at 13,000 x g for 15 min at 4°C, and thereafter, the supernatants were filter sterilized through a 0.22-μm-pore-size filter (Millipore). All the supernatant filtrates obtained were lyophilized and stored at –80°C until use as a source of putative AIP.
Synthesis of the AgrD cyclic peptides. The cyclic peptides RIPISTGFF and KIPTSTGFF were custom synthesized by Peptide Protein Research Company Ltd. (Eastleigh, United Kingdom). Both peptides were cyclized by a lactone ring between the serine side chain and the C terminus and purified in acetonitrile and water containing 0.1% trifluoroacetic acid. The purified peptides were 95% homogenous as indicated by high-performance liquid chromatography and mass spectrometry.
In vitro biological activity assay. The spent culture supernatants or synthetic peptides (50 μg or 100 μg) were added separately to early exponentially growing S. intermedius strain 3 cells and incubated with shaking for 30 min, prior to stabilization with Bacterial Protect reagent (QIAGEN) and harvesting, in accordance with the manufacturer's recommendation. Following RNA isolation, the expression levels of luk, entC, RNAII,I and of the internal control 16S rRNA were analyzed by real-time reverse transcription-PCR.
RNA preparation and cDNA synthesis. Total bacterial RNA was extracted from S. intermedius and S. aureus using the RNeasy Mini kit (QIAGEN). DNase digestion of 7 μl of total RNA was performed with 2 U RNase-free DNase I (Promega) and 1 μl of the supplied buffer in a total volume of 10 μl for 30 min at 37°C. RNA concentrations were determined using the RiboGreen RNA quantification reagent (Molecular Probes, Leiden, The Netherlands). To assess RNA integrity, 2 μg of each RNA sample was electrophoresed in a 1.2% agarose-0.66 M formaldehyde gel. For cDNA synthesis, 5 μl of total RNA was mixed with 0.8 μl (400 ng) of random hexamer (Promega), 2 μl of 10x first-strand buffer, 2 μl of 5 mM dNTP, 10 U of RNasin (Promega), 1 μl of Sensiscript reverse transcriptase (QIAGEN), and distilled water to a total volume of 20 μl. Samples were incubated for 30 min at 37°C, followed by heat inactivation at 65°C for 10 min. Random priming was used to generate cDNA so as to reduce the number of reverse transcriptase reactions required.
Real-time quantitative PCR of RNAIII, leukotoxin, and enterotoxin C transcription in S. intermedius. Amplicon standards were generated for each primer pair by amplifying 100 to 500 ng of genomic S. intermedius DNA in a thermocycler. The following primers were used: 5'-GCCTAATACATGCAAGTC-3'and 5'-CATGTTATCCGGCATTAG-3' for 16S rRNA, 5'-CAGTTCACCATATGAAACAG-3' and 5'-TCCATTCTTTGTTGTAAGGTG-3' for enterotoxin C, 5'-GCAAGAGGTAACTTTAATCCTG-3' and 5'ATCAGAATGTGTTTTGATTAG-3' for leukotoxin, and 5'-GCAGCAGATATCATTAGC-3' and 5'-TGCTACAATGGCTTCA-3'for RNAIII. The design of the forward and reverse primers for RNAIII was based on the sequences obtained from genome walking. Amplicons were gel purified using QIAquick gel extraction kit (QIAGEN) before the copy number was calculated. The real-time PCR templates for each gene included fresh dilutions of the standards, cDNA samples, and distilled water for a negative control. A control for each primer pair and RNA sample also included sham cDNA synthesis reactions that lack reverse transcriptase, followed by PCR amplification to identify RNA preparations contaminated by residual genomic DNA. All reactions were performed in triplicate in a DNA Engine Opticon 2 system (MJ Research, Waltham, MA). Each reaction tube contained 10 μl of 2x Quantitect SYBR green PCR master mix (HotStarTaq DNA polymerase, Quantitect SYBR green PCR buffer, dNTP mix, SYBR green I, and 5 mM MgCl2), 0.3 μM of gene-specific forward and reverse primers, and 2 μl template, made up to a final volume of 20 μl with distilled water. Cycling parameters were as follows: initial activation step at 95°C for 15 min, denaturation at 94°C for 15 s, annealing at 55°C for 30s, extension at 72°C for 30 s, and fluorescence data acquisition at 72°C for 15 s. Melting curve analysis was performed at 50°C to 95°C with stepwise fluorescence acquisition at every 1°C s–1. Melting curves observed for each gene were confirmed to correspond to the correct amplicon size by agarose gel electrophoresis of PCR products (not shown). Sequence-specific standard curves were generated using 10-fold serial dilutions (103 to 107 copies μl–1) of the specific RNA standards. On the basis of the given copy number values for amplicon standards, the Opticon Monitor analysis software (version 2.01) generated a standard curve that was used to quantitate representative gene cDNA copy numbers in each sample. The mean cDNA copy number values obtained for each gene were divided by the corresponding mean 16S rRNA values to standardize for the number of bacterial cells present in each sample. Relative comparisons between corrected values were performed using the analysis of variance test in GraphPad Prism software package (version 4.0) applying the Bonferroni posthoc test to allow multiple comparisons for significance.
Nucleotide sequence accession number. The DNA sequence of the 3.436-kb fragment has been deposited in GenBank under the accession number AY965912.
RESULTS AND DISCUSSION
Cloning and sequencing of the agrD gene revealed that the 20 canine S. intermedius strains tested could be divided into three agr specificity groups based on the putative autoinducing peptide produced by each strain. Elucidation of the complete 3' agr (RNAIII) and the upstream agr flanking sequences was achieved using primers U1 and AP2 and primers U2 and AP2. For the downstream region of agrD, nested PCR gave a 5.5-kb amplified fragment. By repeated sequential PCR and primer walking, the entire agr determinant and further downstream flanking sequences were obtained (GenBank accession number AY965912) (Fig. 1).
Through genome walking, the agr locus in S. intermedius was found to span a total of 3,436 bp, with five ORFs, agrB, agrD, agrC, agrA, and hld, in two divergently transcribed transcripts, RNAII and RNAIII. The start codons of agrB, agrD, and agrC of S. intermedius are preceded by the sequence 5'-AGGAGGA-3', which is similar to the typical consensus ribosome-binding site 5'-AGGAGGT-3' in gram-negative bacteria. For agrA, the ribosome-binding site is 5'AGGATGT-3'.
The amino acid sequences of AgrB, AgrD, AgrC, and AgrA of S. intermedius were compared to those of known Agr proteins from other staphylococci, and analyzed by manual examination and publicly available computer programs. Only the agr loci from S. aureus group I (GenBank accession number X52543), S. aureus group II (GenBank accession number AP003135), S. aureus group III (GenBank accession number AP004829), S. epidermidis (GenBank accession number Z49220), Staphylococcus lugdunensis (GenBank accession number AF173933), and S. intermedius (GenBank accession number AY965912, this study) were incorporated in the multiple-alignment analysis, since sequence data for other staphylococci are incomplete. The identified agrB, agrD, and agrC regions of RNAII of S. intermedius share some similarity with the corresponding sequences of other staphylococci. However, regions of difference are found, and those regions of RNAII that show little similarity among the staphylococcal species are thought to be important for species-specific peptide processing and recognition involved in the agr response.
AgrA of S. intermedius. In the 3' region of RNAII, high sequence conservation throughout the agrA genes of S. intermedius and S. aureus is evident, with 83.6% identity. Evidence for the role of AgrA as a response regulator can also be inferred from analyses of its sequences using reverse position-specific BLAST (rpsBLAST) and the Superfamily 1.63 HMM library and genome assignments server. Analysis using ConPredII found no likely transmembrane helices in the sequence of AgrA of S. intermedius, in agreement with the expected cytoplasmic localization of response regulators (not shown). Being a putative response regulator, AgrA of S. intermedius is likely to have little direct involvement with the specificity of processing and of the ligand-receptor interactions that constitute the putative quorum-sensing circuits, which may account for the low sequence divergence among the AgrA proteins of various staphylococci.
AgrB of S. intermedius. Analysis of the amino acid sequence (188 residues) of the AgrB protein of S. intermedius reveals that it shares 41.5% identity to the AgrB proteins of S. aureus, S. lugdunensis, and S. epidermidis. Transmembrane topology analysis predicted the AgrB of S. intermedius to be a membrane-spanning protein with five potential helices, consistent with the proposed function of AgrB as seen in S. aureus, where it is responsible for processing of the AIP expressed by AgrD. Since AIP is derived from an internal fragment of the agrD-coding sequence and its synthesis involves the product of the agrB gene (15), it would be expected that polymorphism in the agrD gene would require agrB to vary accordingly, so as to retain the specificity of peptide processing. This may account for the high sequence variation of AgrB in the various staphylococcal species.
AgrC of S. intermedius. The S. intermedius AgrC amino acid sequence (433 residues) shows 15.7% identity with analogous products from S. aureus. Transmembrane topology analysis of AgrC of S. intermedius suggested the presence of a N-terminal hydrophobic domain. A search of conserved domains (rpsBLAST) revealed that the C terminus of AgrC is categorized as a member of the histidine protein kinase family. As agrC presumably encodes a receptor responsible for binding to the autoinducing peptide derived from agrD, divergence in the N terminus of the agrC of staphylococci would allow binding to its corresponding polymorphic AIP, thus retaining the specificity of the receptor-ligand interaction. This divergence may be responsible for speciation within staphylococci, similar to divergence within the structurally related comAP locus required for competence in Bacillus (33).
AgrD of S. intermedius. The short S. intermedius AgrD amino acid sequence (45 residues) shows only 24.4% similarity with the equivalent prepheromone of S. aureus. Characterization of the AIP of S. aureus by Ji and colleagues revealed that it is a modified peptide whose sequence is contained within an internal fragment of the agrD-coding sequence (15). On the basis of the primary sequence of the pheromone peptides determined from PCR analysis, four different agr subgroups have been classified in S. aureus (15, 26), and one group has been detected in 15 investigated strains of S. epidermidis (27). Our similar PCR sequence analyses of the agrD genes among 20 strains of S. intermedius isolated from clinical cases of canine pyoderma revealed the distribution of three AIP allelic variants. Among the 20 strains, 5% (1 strain) produce the AIP RIPTSTGFF, while 35% (7 strains) and 60% (12 strains) produce the RIPISTGFF (this study) and KIPTSTGFF peptides, respectively (Table 1). The AIP gene in the three different groups of S. intermedius was highly conserved and shows nucleotide polymorphisms that led to conservative amino acid changes only.
With S. aureus, the highly divergent AIP sequences in the different groups vary in length from seven to nine amino acyl residues. Through extensive structure-activity relationship studies, the AIP is found to be a thiolactone molecule containing a ring of five amino acids formed by a thioester linkage between the sulfhydryl group of a cysteine residue and the C-terminal carboxyl group, and a tail ranging from two to four amino acid residues, depending on the staphylococcal species and groups within the same species (13, 15, 22, 27). Prediction of the S. intermedius AIPs from the corresponding nucleotide sequences of agrD revealed that in contrast to the AIP sequences in the four groups of S. aureus, there are no conserved sequences of five cysteine residues from the COOH terminus in the putative AIPs of all three specificity groups in S. intermedius. This is consistent with the findings by Kalkum and colleagues, who reported that instead of a thiolactone structure, a lactone structure was discovered in a pigeon isolate of S. intermedius (17). Despite the high divergence of the agrD sequences among the different staphylococcal species, there is nonetheless a more or less conserved negatively charged C terminus. This negatively charged terminus has been proposed to be involved in the processing of AgrD (35), and it contributes to the translocation of the AIP (15).
RNAIII of S. intermedius. Upstream of the transcription start site, putative –10 and –35 promoter elements, similar to those in other staphylococci, were found in S. intermedius (Fig. 2). A direct repeat immediately upstream of the –35 promoter element was found in all species of staphylococci. Sequence conservation among these elements suggests their functionally important roles in regulating RNAIII gene expression (24), with similar control mechanisms likely to exist in all staphylococcal species. The 3' regions of the RNAIII sequences of the different staphylococcal species also showed a considerable degree of sequence conservation (Fig. 2). In S. aureus, the 3' region was found to be essential for the production of exoproteins, such as and toxin (18). Similarity in this region among the different staphylococci suggests that they may play roles to that of S. aureus.
The RNAIII gene also contains an open reading frame encoding the delta-lysin homologue. Delta-lysin is encoded by the hld gene, which gives rise to a transcript of 75 nucleotides in S. intermedius. The predicted delta-lysin genes of the various staphylococcal species are found at different distances from the transcription start point, with most being localized in the 5' half of RNAIII (Fig. 2). In S. intermedius, there is an additional insertion upstream of the structural gene for the delta-lysin open reading frame (Fig. 2). The open reading frame of the delta-lysin in S. intermedius possesses a reasonably high degree of sequence identity to those of S. aureus and other coagulase-negative staphylococci (Table 2). In S. intermedius, 16 of the 25 amino acids of delta-lysin are identical to those of S. aureus. From the alignment of the primary sequence of delta-lysin in Table 2, it can be seen that the distribution of the charged residues was conserved between all predicted molecules. Consequently, it is likely that the delta-lysin of S. intermedius may function as a pore-forming toxin, as has been reported for S. aureus delta-lysin (8).
Fitton et al. and Turner et al. determined the amino acid sequence of delta-lysin of S. aureus from human and canine isolates and found that the toxin obtained from S. aureus of the human strain differs from that made by the canine strain by nine amino acids (Table 2) (7, 34). Interestingly, the delta-lysin from a canine isolate of S. intermedius (this study) compared with that from a canine isolate of S. aureus (34) differs only by a single amino acid (Table 2). Given the higher similarity between S. aureus and S. intermedius of canine origin compared to that between S. aureus of human and canine origin, one might question the true identity of the suggested S. aureus of canine origin.
Temporal regulation of RNAIII, leukotoxin, and enterotoxin C expression. Despite differences in the primary nucleotide sequences of RNAIII from S. aureus and S. intermedius, the kinetics of transcription in both are comparable—notably, the expression of RNAIII reaches its maximum in late logarithmic phase of growth (Fig. 3A). The temporal regulation of RNAIII suggests that activation of the P3 promoter, which drives the expression of RNAIII, is growth phase dependent. Regulation of RNAIII in an autocrine manner is consistent with its hypothesized function as the effector of the agr system. Expression of enterotoxin C and leukotoxin were similarly temporally regulated, displaying a significant increase in the amount of transcripts in the post-exponential phase (Fig. 3B). Thus far, there is very little data available concerning the relationship between S. intermedius cutaneous infection and the production of leukotoxin and enterotoxin, respectively, because little is known about the local and general effects of the toxin products in vivo. The luk gene was chosen for this study, as all 20 clinical isolates of S. intermedius isolated from cases of canine pyoderma were found to harbor the two cotranscribed genes of leukotoxin, lukF and lukS (not shown), suggesting the likely role and importance of this toxin in the virulence of S. intermedius, at least in cutaneous infection. Staphylococcal enterotoxins, on the other hand, are known to have potent effects on cells of the immune system in general, functioning as superantigens to cause excessive nonspecific T-cell proliferation. In S. intermedius, the expression of both virulence factors followed a trend similar to that of RNAIII and was positively correlated (RNAIII versus entC, r = 0.8191 and P = 0.0898; RNAIII versus luk, r = 0.9001 and P = 0.0373), suggesting that changes in the regulation of both toxins studied may be attributed to the differential transcription of RNAIII.
Characterization of the activating factor for luk, entC, and RNAIII expression in S. intermedius. Given that the AgrC protein of S. intermedius resembles the classical sensory transducing proteins of bacteria, it seemed likely that agr, and hence RNAIII activation, would be initiated by an environmental signal generated during bacterial growth. To investigate whether quorum sensing via the agr gene was responsible for the exponential-phase induction of toxin transcription, the effects of cell-free spent culture supernatants obtained from different phases of growth from S. intermedius strains 3 and 27 and S. aureus strain J3945 on gene expression in S. intermedius strain 3 were assessed. The spent media collected were individually added to early exponential cultures of S. intermedius strain 3 at a time when expression of each gene studied was minimally activated, as determined in preliminary studies (not shown). The results obtained demonstrate that post-exponential-phase culture supernatants of S. intermedius contain a substance that, when added to early exponential self cultures, caused immediate activation of RNAIII, enterotoxin C, and leukotoxin transcription, with expression of RNAIII and both virulence factors being positively correlated (RNAIII versus entC, r2 = 0.9705 and P = 0.0295; RNAIII versus luk, r2 = 0.9821 and P = 0.0179) (Fig. 4). Exhibition of a concentration-dependent activation suggests that the production of virulence factors in S. intermedius is initiated by a growth-specific cue generated during bacterial growth. The activating substance appears to accumulate in the supernatant during the growth of S. intermedius, and upon reaching a threshold level, it regulates virulence factor gene expression. This type of kinetics represents a cell density-sensing mechanism similar to that observed with Pseudomonas, Vibrio, Agrobacterium, and other bacterial species that use homoserine lactones as autoinducers (5, 16, 36).
The abundance of mRNA from S. intermedius RNAIII, leukotoxin, and enterotoxin genes demonstrated dependency on secreted biological cues. With self biological cues, activation of the transcripts encoding three genes studied occurred when supernatants obtained from late-exponential growth of S. intermedius strain 3 (t = 8 h) were added to the same growing strain (Fig. 4). When post-exponential supernatant from a different strain, S. intermedius strain 27 (S27), was added to S. intermedius strain 3 cells, there was higher expression of RNAIII (P < 0.01) and luk (P < 0.05) than when the medium-only control was added (Fig. 4). To verify that activation of the above-mentioned transcripts following the addition of post-exponential-phase spent medium is due to specific signals elaborated by the bacterium itself, cyclic peptides were synthesized, according to the sequenced putative AIPs of strain 3 (self) and S27 (nonself) in order to test their efficacy to induce gene expression in S. intermedius strain 3. Up-regulation of RNAIII (P < 0.001), luk (P < 0.001), and entC (P < 0.001) expression was observed following the addition of "self" peptide (RIPISTGFF) to S. intermedius strain 3. With respect to the "nonself" peptide (KIPTSTGFF), there was no convincing activation of the agr response, in contrast to that observed when nonself spent culture medium (S27) was added to S. intermedius strain 3. The difference could be attributable to changes in environmental factors as S. intermedius enters the post-exponential phase, such as alterations in the levels of acidity, or to the presence of nonspecific endogenous substances in the spent culture medium.
With S. aureus and S. epidermidis, although the various peptide groups differ in their amino acid sequences, each has been shown to possess a cyclic thiolactone moiety (15, 22, 27). Previous studies have reported the essentiality of the high-energy thioester linkage between the central cysteine and the C-terminal carboxyl group in the thiolactone ring of these AIPs for full agr activation activity (20, 22, 23, 27). Despite being potent intergroup inhibitors, the corresponding lactone and lactam analogues of the group II AIP of S. aureus were also found to be inactive, as were linear version of this and other AIPs (14, 22). In our study, the addition of "self" synthetic peptide (RIPISTGFF) with a lactone cyclic structure to early exponentially growing S. intermedius activated transcription of RNAIII. The putative nonapeptide, RIPISTGFF, is thus the activating factor that turns on the agr response in S. intermedius strain 3. Results obtained from this study also suggest that the thiolactone ring is not an absolute requirement for agr activation in S. intermedius, as was proposed for other staphylococci.
Taken together, the response defined on the basis of increases in transcript levels of quorum-sensing-controlled genes, such as RNAIII of agr, indicates that S. intermedius is an active participant in activating its own virulence regulatory system. This type of regulation is a form of quorum sensing, whereby a population of bacteria responds in concert when a critical cell density is reached. The identification of the agr gene in S. intermedius is an important step towards understanding its pathogenesis. Given the similarities in the orientation, organization, and conservation of certain residues and domains of the agr gene of S. intermedius to those of S. aureus, it is likely that the agr system may play a comparable role in the pathology of S. intermedius canine pyoderma.
ACKNOWLEDGMENTS
This work was supported by a grant from Nestle Purina Petcare.
Present address: Department of Cellular and Molecular Medicine, St. George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, United Kingdom.
REFERENCES
1. Allaker, R. P., A. I. Lamport, D. H. Lloyd, and W. C. Noble. 1991. Production of virulence factors by Staphylococcus intermedius isolates from cases of canine pyoderma and healthy carriers. Microb. Ecol. Health Dis. 4:169-173.
2. Arai, M., H. Mitsuke, M. Ikeda, J. X. Xia, T. Kikuchi, M. Satake, and T. Shimizu. 2004. ConPred II: a consensus prediction method for obtaining transmembrane topology models with high reliability. Nucleic Acids Res. 32:W390-W393.
3. Colombini, S., S. R. Merchant, and G. Hosgood. 2000. Microbial flora and antimicrobial susceptibility patterns from dogs with otitis media. Vet. Dermatol. 11:235-239.
4. Dowling, P. M.1996 . Antimicrobial therapy of skin and ear infections.Can. Vet. J. 37:695-699.
5. Dunlap, P. V., and J. M. Ray. 1989. Requirement for autoinducer in transcriptional negative autoregulation of the Vibrio fischeri luxR gene in Escherichia coli. J. Bacteriol. 171:3549-3552.
6. Dziewanowska, K., V. M. Edwards, J. R. Deringer, G. A. Bohach, and D. J. Guerra. 1996. Comparison of the beta-toxins from Staphylococcus aureus and Staphylococcus intermedius. Arch. Biochem. Biophys. 335:102-108.
7. Fitton, J. E., A. Dell, and W. V. Shaw.1980 . The amino acid sequence of the delta haemolysin of Staphylococcus aureus. FEBS Lett. 115:209-212.
8. Freer, J. H., and T. H. Birkbeck. 1982. Possible conformation of delta-lysin, a membrane-active peptide of Staphylococcus aureus. J. Theor. Biol. 94:535-540.
9. Greene, R. T., and C. H. Lammler. 1993. Staphylococcus intermedius: current knowledge on a pathogen of veterinary importance. J. Vet. Med. B 40:206-214.
10. Hajek, V. 1976. Staphylococcus intermedius, a new species isolated from animals. Int. J. Syst. Bacteriol. 26:401-408.
11. Hendricks, A., H. J. Schuberth, K. Schueler, and D. H. Lloyd. 2002. Frequency of superantigen-producing Staphylococcus intermedius isolates from canine pyoderma and proliferation-inducing potential of superantigens in dogs. Res. Vet. Sci. 73:273-277.
12. Ihrke, P. J. 1987. An overview of bacterial skin disease in the dog. Br. Vet. J. 143:112-118.
13. Jarraud, S., G. J. Lyon, A. M. Figueiredo, G. Lina, F. Vandenesch, J. Etienne, T. W. Muir, and R. P. Novick. 2000. Exfoliatin-producing strains define a fourth agr specificity group in Staphylococcus aureus. J. Bacteriol. 182:6517-6522.
14. Ji, G., R. C. Beavis, and R. P. Novick.1995 . Cell density control of staphylococcal virulence mediated by an octapeptide pheromone. Proc. Natl. Acad. Sci. USA 92:12055-12059.
15. Ji, G., R. Beavis, and R. P. Novick. 1997. Bacterial interference caused by autoinducing peptide variants.Science 276:2027-2030.
16. Jones, S., B. Yu, N. J. Bainton, M. Birdsall, M. Byceoft, B. W. Chhabra, S. R. Cox, A. J. R. Golby, P. Reeves, P. J. Staphens, S. Winson, M. K. Salmond, G. P. C. Stewart, and P. Williams.1993 . The lux autoinducer regulates the production of exoenzyme virulence determinants in Erwinia carotovora and Pseudomonas aeruginosa. EMBO J. 12:2477-2482.
17. Kalkum, M., G. J. Lyon, and B. T. Chait.2003 . Detection of secreted peptides by using hypothesis-driven multistage mass spectrometry. Proc. Natl. Acad. Sci. USA 100:2795-2800.
18. Kornblum, J., B. N. Kreiswirth, S. J. Projan, H. Ross, and R. P. Novick. 1990. Agr: a polycistronic locus regulating exoprotein synthesis in Staphylococcus aureus, p. 373-402. In R. P. Novick (ed.), Molecular biology of staphylococci. VCH Publisher, New York, N.Y.
19. Lyon, G. J., P. Mayville, T. W. Muir, and R. P. Novick.2000 . Rational design of a global inhibitor of virulence response in Staphylococcus aureus, based in part on localization of the site of inhibition to the receptor-histidine kinase, AgrC. Proc. Natl. Acad. Sci. USA 97:13330-13335.
20. Lyon, G. J., J. S. Wright, A. Christopoulos, R. P. Novick, and T. W. Muir. 2002. Reversible and specific extracellular antagonism of receptor-histidine kinase signalling. J. Biol. Chem. 227:6247-6253.
21. Marrack, P., and J. Kappler. 1990. The staphylococcal enterotoxins and their relatives. Science 248:705-711.
22. Mayville, P., G. Ji, R. Beavis, H. Yang, M. Goger, R. P. Novick, and T. Muir. 1999. Structure-activity analysis of synthetic autoinducing thiolactone peptides from Staphylococcus aureus responsible for virulence. Proc. Natl. Acad. Sci. USA 96:1218-1223.
23. McDowell, P., Z. Affas, C. Reynolds, M. T. Holden, S. J. Wood, S. Saint, A. Cockayne, P. J. Hill, C. E. Dodd, B. W. Bycroft, W. C. Chan, and P. Williams. 2001. Structure, activity and evolution of the group I thiolactone peptide quorum sensing system of Staphylococcus aureus. Mol. Microbiol. 41:503-512.
24. Morfeldt, E., K. Tegmark, and S. Arvidson. 1996. Transcriptional control of the agr-dependent virulence gene regulator, RNAIII, in Staphylococcus aureus. Mol. Gen. Genet. 248:446-458.
25. Novick, R. P., H. F. Ross, S. J. Projan, J. Kornblum, B. Kreiswirth, and S. Moghazeh. 1993. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J. 12:3967-3975.
26. Otto, M., H. Echner, W. Voelter, and F. Gotz. 2001. Pheromone cross-inhibition between Staphylococcus aureus and Staphylococcus epidermidis. Infect. Immun. 69:1957-1960.
27. Otto, M., R. Sussmuth, C. Vuoung, G. Jung, and F. Gotz.1998 . Structure of the pheromone peptide of the Staphylococcus epidermidis agr system. FEBS Lett. 424:89-94.
28. Pearson, W. R., and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85:2444-2448.
29. Prevost, G., T. Bouakham, Y. Piemont, and H. Monteil. 1995. Characterisation of a synergohymenotropic toxin produced by Staphylococcus intermedius. FEBS Lett. 376:135-140. (Erratum, 381:272, 1996.)
30. Projan, S. J., and R. P. Novick. 1997. The molecular basis of pathogenicity, p.55 -82. In K. B. Crossley and G. L. Archer (ed.), The staphylococci in human diseases. Churchill Livingstone, New York, N.Y.
31. Scott, D. W., W. H. Miller, Jr., and C. E. Griffin. 2001. Bacterial skin diseases, p.275 -335. In D. W. Scott, W. H. Miller, Jr., and C. E. Griffin (ed.),Small animal dermatology , 6th ed. W. B. Saunders Company, Philadelphia, Pa.
32. Terauchi, J., H. Sato, T. Hasegawa, T. Yamaguchi, C. Aizawa, and N. Maehara. 2003. Isolation of exfoliative toxin from Staphylococcus intermedius and its local toxicity in dogs. Vet. Microbiol. 94:19-29.
33. Tortosa, P., and D. Dubnau. 1999. Competence for transformation: a matter of taste. Curr. Opin. Microbiol. 2:588-592.
34. Turner, W. H. 1978. Purification and characterization of an immunologically distinct delta-hemolysin from a canine strain of Staphylococcus aureus. Infect. Immun. 20:485-494.
35. van Wamel, W. J., G. van Rossum, J. Verhoef, C. M. Vanderbroucke-Grauls, and A. C. Fluit. 1998. Cloning and characterization of an accessory gene regulator (agr)-like locus from Staphylococcus epidermidis.FEMS Microbiol. Lett. 163:1-9.
36. Zhang, L., P. J. Murphy, A. Kerr, and M. E. Tae. 1993. Agrobacterium conjugation and gene regulation by N-acyl-L-homoserine lactones. Nature 362:446-448.(Julia M. L. Sung, Peter D)
Veterinary Basic Sciences, Royal Veterinary College, University of London, London, United Kingdom
ABSTRACT
The accessory gene regulator (agr) locus, a candidate system for the regulation of the production of virulence factors in Staphylococcus intermedius, has been characterized. Using PCR-based genome walking, we have obtained the first complete sequence (3,436 bp) of the accessory gene regulator (agr) gene in this organism. Sequence analysis of the agr gene has identified five open reading frames (ORFs), agrB, agrD, agrC, agrA, and hld. The translated ORF contained amino acid motifs characteristic of the response regulator and histidine protein kinase signal transducer of the classic two-component regulatory system. Sequencing of the agrD PCR products amplified from DNA from 20 different isolates has facilitated detection of genetic variation in the putative autoinducing peptide (AIP) within the agr gene of S. intermedius, revealing the presence of at least three agr specificity groups within this species. Classification of the agr gene from S. intermedius was supported by phylogenetic analysis. Real-time PCR also revealed that the effector molecule of the agr system, RNAIII, was regulated in an autocrine manner in S. intermedius and demonstrated positive correlation with the temporal gene expression patterns of luk and entC. Transcription of RNAIII was also dependent on self secreted cues. Cyclic self and nonself peptides were synthesized on the basis of the novel AIPs produced by S. intermedius, which lack the cysteine necessary to form the thiolactone ring in analogous peptides from Staphylococcus aureus and Staphylococcus epidermidis. Experiments with these synthetic cyclic peptides indicated that self peptides led to up-regulation of RNAIII—findings in support of the assumption that activation of the agr gene is initiated by growth- and species-specific factors generated during bacterial growth.
INTRODUCTION
Staphylococcus intermedius,first described in 1976 (10), is the most common cause of skin infection in dogs, with canine pyoderma being caused almost exclusively by S. intermedius (31). S. intermedius is also frequently isolated from dogs with otitis media and externa (3, 4, 28) and is commonly associated with seborrhoea, a cutaneous scaling disorder in dogs. Little is known, however, about the pathogenesis of S. intermedius. The results of studies investigating the production of virulence factors are unclear, and no distinct differences in the nature of the toxins produced by isolates from healthy dogs and dogs with pyoderma have been found (1). Thus far, there are differing opinions as to the underlying causes of canine pyoderma. Ihrke suggested factors that predispose dogs to staphylococcal pyoderma infections were due to a wide variety of underlying diseases, such as allergy, metabolic diseases, disruption of normal skin flora, immunodeficiency, and endoparasitic infestation (12), whereas according to Greene and Lammler, host-bacterium interactions and the immune status of the host appeared to be the factors most crucial in determining the outcome of infection (9). Many of the factors believed to be involved in S. intermedius virulence have also not been well characterized, and proof of their association with disease pathogenesis is often lacking. It is thus difficult for veterinary clinicians and microbiologists to determine the clinical significance of many S. intermedius isolates.
S. intermedius produces a diverse arsenal of virulence factors involving a large number of cell-bound and extracellular proteins (6, 11, 29, 32), which may be likely contributory factors that predispose dogs to staphylococcal pyoderma. The virulent components and products are thought to share overlapping roles, acting either in concert or alone, and may convert host tissue components into nutrients required for bacterial growth. On the basis of studies performed with corresponding virulence factors in Staphylococcus aureus, considerable knowledge is available about their contribution to the development of infection in humans (21). Less is known about the virulence factors of S. intermedius and their relative importance in canine pyoderma.
In the human pathogen S. aureus, the production of virulence factors is known to be controlled, in part, by the accessory gene regulator system. One of a number of two-component regulatory systems identified thus far, the agr system has received much attention owing to it being assigned, historically, a central role in the model of S. aureus pathogenesis. It is recognized as a quorum-sensing gene cluster that up-regulates production of secreted virulence factors and down-regulates production of cell-associated virulence factors in a cell density-dependent manner (15, 19, 25). When the staphylococci are in lag phase, it is thought that staphylococci initiate infection by synthesizing surface proteins. Once colonization is established, the bacteria multiply and enter exponential phase, activating a density-sensing mechanism that stimulates toxic exoprotein production, thereby enabling them to spread to new sites to prevent overcrowding (30). Synthesis of essential factors only as and when required is important from an economic perspective, as it avoids overwhelming the organism metabolically with unnecessary energy expenditure.
S. intermedius produces many toxins comparable to those of S. aureus, so we hypothesized that it is likely that the production of toxins in the former is under the regulation of a similar global regulator. To determine whether this is so, here we have ascertained the presence of the agr locus and analyzed its sequence. Our starting point was the agrD gene present in each strain of a collection of 20 S. intermedius strains isolated from cases of canine pyoderma. The elucidation of the agr locus facilitated assessment of its regulation during the growth of S. intermedius. In S. aureus, mutation and deletion analyses performed suggested that RNAIII itself, rather than any translation product, is the effector of exoprotein gene regulation (25). We have thereby quantified RNAIII, the effector molecule of agr, together with luk and entC, encoding leukotoxin and enterotoxin C, respectively, using real-time quantitative PCR. We have also synthesized two novel cyclic autoinducing peptides (AIPs), shown to be found in S. intermedius (17), but not in S. aureus or Staphylococcus epidermidis, and used these in functional assays monitoring RNAIII production.
MATERIALS AND METHODS
Bacterial strains and growth conditions. Twenty clinical isolates of S. intermedius were collected from the skin of dogs with pyoderma at the Royal Veterinary College. Long-term storage of the isolates was maintained in 20% glycerol in brain heart infusion broth at –70°C, and cultures for routine use were maintained on sheep blood agar plates. Isolates were identified as S. intermedius according to their colony morphology, biochemical reactions (production of coagulase and DNase), and on the basis of the analytical profile index (API) (bioMerieux, France).
PCR and sequencing. The collection of 20 S. intermedius isolates was screened by direct colony PCR. It was found that the initial hot start of the PCR was sufficient to break down bacterial cell walls from a pinprick colony of S. intermedius, allowing these cells to be used as the direct source of DNA. We used the following primers: 5'-TTCCAATCAATCGGTACT-3' and 5'-TTCCTCCTTCTAACAACTC-3'; their design was based on regions of the partial agr sequence derived from a pigeon isolate of S. intermedius (GenBank accession number AF346723). PCR was carried out in the GeneAmp PCR system 9700 thermocycler (Perkin-Elmer Biosystems, Foster City, California). Amplification was performed in a final volume of 50 μl containing dATP, dCTP, dGTP, and dTTP (each at a concentration of 200 μM), 10 μM of the corresponding primer, 5 μl of 10x Expand high-fidelity buffer (Roche, Nutley, NJ), and 2.6 U of Expand high-fidelity enzyme mix (Roche). The cycling parameters for PCR were as follows: (i) an initial denaturation step of 2 min at 94°C; (ii) 35 cycles, with 1 cycle consisting of 15 s at 94°C, 30 s at 45°C, and 60 s at 72°C; and (iii) a final extension step of 7 min at 72°C. The amplified fragments were cloned into the pGEM-T Easy vector (Promega, Madison, Wis.). Recombinant DNA molecules were transformed into electrocompetent Escherichia coli JM109 cells and sequenced (MWG-Biotech, Ebersberg, Germany).
Genome walking. Genome walking was initiated from the amplified and sequenced agrD DNA fragment using the Universal Genome-Walker kit (BD Biosciences Clontech, Palo Alto, CA). Total genomic DNA of S. intermedius strain 3 was isolated with Genomic-Tip 100/G (QIAGEN, Crawley, United Kingdom) and digested with DraI, EcoRV, PvuII, and StuI to obtain blunt ends. An aliquot of each DNA sample was run on a 0.7% agarose gel to ascertain quality. After purification, the fragments were ligated to a Genome-Walker adapter provided by the manufacturer (31). The Genome-Walker protocol consists of a series of PCR amplifications, starting initially with the adapter primers (AP1 or AP2) and two nested gene-specific primers designed on the basis of the agrD sequence of S. intermedius (see above), D1 (5'-CCAATCAATCGGTACTTTTGCTAGAATCCCTATTAGTACTGGA-3') and U1 (5'-AGATGAGCCAGAAATCCCTGCCGAGTTGTTAGAAGAGGAC-3'). Sequence extension occurred through further walks using novel gene-specific primers, and the primers were designed on the basis of initial sequence data obtained from previous steps. Care was taken to ensure that the sequences overlapped to some degree. Downstream chromosomal walking was continued with the following primers: D2 (5'-CGTGAAGATGATTTAGAAGGTTTGAAGCAA-3') and D3 (5'-GCTACCAATGACCCTTATGCTATCTTAGAAACTTCT-3'). The upstream region of agrD was analyzed in a similar way using primer U2 (5'-CAACAACTTACGCACTGGTTAACACAAGGAC-3'). Each amplification was performed in a final volume of 50 μl containing 5 μl 10x genomic PCR buffer, 1.1 mM magnesium acetate, 10 mM 50x deoxynucleoside triphosphate (dNTP) mix, 1 μl 50x Advantage genomic polymerase mix (BD Biosciences), and 10 μM of the corresponding primers. A typical PCR cycle consisted of 7 cycles of 94°C for 2 s and 72°C for 3 min, followed by 32 cycles of 94°C for 2 s and 67°C for 3 min, and a final elongation step at 67°C for 4 min. To facilitate PCR amplification of sequences that proved difficult to amplify, we used the Advantage GC genomic polymerase (BD Biosciences).
PCR products from genome walking were purified using the QIAquick gel extraction kit (QIAGEN) and directly sequenced with both primers (adapter and S. intermedius specific). For PCR amplicons that were longer than 0.6 kb, sequencing was performed via primer walking, involving repeated cycles of custom oligonucleotide synthesis and direct sequencing. Sequences obtained were confirmed by comparison of independent PCR experiments with different primers and genomic DNA. Alignments of amino acid sequences from the individual protein-coding genes were realized by using the ClustalW algorithm (http://www.ebi.ac.uk/clustalw) and default gap penalties (28). Regions of sequence that were difficult to align were removed from the data file and the sequences realigned. Ambiguous alignment regions, such as the N and C termini, were excluded from the alignments. Overlapping nucleotide sequences of DNA generated were manually aligned to form a contiguous agr operon and assessed for complete open reading frames (ORFs). DNA and protein sequence similarity searches were performed by the BLASTN and BLASTP (National Center for Biotechnology Information server). Conserved domains in each ORFs were analyzed using reverse position-specific BLAST (rpsBLAST) (v1.65).
Transmembrane topology prediction. The transmembrane topology of the various ORFs of the S. intermedius agr gene was analyzed using ConPredII (http://bioinfo.si.hirosaki-u.ac.jp/ConPred2/), a program based on a consensus approach that combines the results of several prediction methods, such as KKD, TMpred, TopPredII, DAS, TMAP, MEMSAT 1.8, SOSUI, TMHMM 2.0, and HMMTOP 2.0 (2).
Assessment of the regulation of leukotoxin, enterotoxin, and RNAIII transcription in S. intermedius. To examine growth phase-dependent expression, 2-ml cultures of S. intermedius (strain 3) were grown overnight with aeration in brain heart infusion broth at 37°C. Cells were then inoculated from the overnight culture to an initial optical density at 600 nm (OD600) of 0.05 and grown with shaking to the desired phase of growth, e.g., early exponential (OD600 of 0.5, 1 h), mid-exponential (OD600 of 4.5; 4 h), or post-exponential phase (OD600 of 8.0; 8 h). The various growth phases were determined in preliminary experiments (not shown). Growth was monitored turbidimetrically with a UV-160A spectrophotometer (Shimadzu) read at 600 nm. The results of growth phase effects were standardized for the number of bacteria, so that the increase in the total amount of protein is not simply a consequence of increasing number of bacterial cells. At indicated times, samples containing equal numbers of cells were immediately stabilized by the addition of 2 volumes of Bacterial Protect reagent (QIAGEN). The bacterial cells were subsequently harvested after 10 min by centrifugation at 13,000 x g for 10 min at 4°C, prior to RNA extraction.
Characterization of the activating factor for luk, entC, and RNAIII expression in S. intermedius. Cultures of S. intermedius strains 3 and 27 and S. aureus strain J3945 were inoculated from 1-day-old single colonies on sheep blood agar plates into 2 ml brain heart infusion broth. The cultures were grown overnight at 37°C with orbital shaking for 15 h. To synchronize cultures to an active mode of growth and to eliminate potential quorum signals that may have accumulated, 0.5 ml of overnight culture was centrifuged at 3,000 x g for 5 min at 4°C. After the supernatant was discarded and the bacteria were resuspended in fresh prewarmed medium (50 ml), they were grown at 37°C with shaking. Aliquots (10 ml) were harvested at intervals of 1, 4, and 8 h corresponding to the early, mid-, and post-exponential phases of bacterial growth. Bacterial cells were removed by centrifugation at 13,000 x g for 15 min at 4°C, and thereafter, the supernatants were filter sterilized through a 0.22-μm-pore-size filter (Millipore). All the supernatant filtrates obtained were lyophilized and stored at –80°C until use as a source of putative AIP.
Synthesis of the AgrD cyclic peptides. The cyclic peptides RIPISTGFF and KIPTSTGFF were custom synthesized by Peptide Protein Research Company Ltd. (Eastleigh, United Kingdom). Both peptides were cyclized by a lactone ring between the serine side chain and the C terminus and purified in acetonitrile and water containing 0.1% trifluoroacetic acid. The purified peptides were 95% homogenous as indicated by high-performance liquid chromatography and mass spectrometry.
In vitro biological activity assay. The spent culture supernatants or synthetic peptides (50 μg or 100 μg) were added separately to early exponentially growing S. intermedius strain 3 cells and incubated with shaking for 30 min, prior to stabilization with Bacterial Protect reagent (QIAGEN) and harvesting, in accordance with the manufacturer's recommendation. Following RNA isolation, the expression levels of luk, entC, RNAII,I and of the internal control 16S rRNA were analyzed by real-time reverse transcription-PCR.
RNA preparation and cDNA synthesis. Total bacterial RNA was extracted from S. intermedius and S. aureus using the RNeasy Mini kit (QIAGEN). DNase digestion of 7 μl of total RNA was performed with 2 U RNase-free DNase I (Promega) and 1 μl of the supplied buffer in a total volume of 10 μl for 30 min at 37°C. RNA concentrations were determined using the RiboGreen RNA quantification reagent (Molecular Probes, Leiden, The Netherlands). To assess RNA integrity, 2 μg of each RNA sample was electrophoresed in a 1.2% agarose-0.66 M formaldehyde gel. For cDNA synthesis, 5 μl of total RNA was mixed with 0.8 μl (400 ng) of random hexamer (Promega), 2 μl of 10x first-strand buffer, 2 μl of 5 mM dNTP, 10 U of RNasin (Promega), 1 μl of Sensiscript reverse transcriptase (QIAGEN), and distilled water to a total volume of 20 μl. Samples were incubated for 30 min at 37°C, followed by heat inactivation at 65°C for 10 min. Random priming was used to generate cDNA so as to reduce the number of reverse transcriptase reactions required.
Real-time quantitative PCR of RNAIII, leukotoxin, and enterotoxin C transcription in S. intermedius. Amplicon standards were generated for each primer pair by amplifying 100 to 500 ng of genomic S. intermedius DNA in a thermocycler. The following primers were used: 5'-GCCTAATACATGCAAGTC-3'and 5'-CATGTTATCCGGCATTAG-3' for 16S rRNA, 5'-CAGTTCACCATATGAAACAG-3' and 5'-TCCATTCTTTGTTGTAAGGTG-3' for enterotoxin C, 5'-GCAAGAGGTAACTTTAATCCTG-3' and 5'ATCAGAATGTGTTTTGATTAG-3' for leukotoxin, and 5'-GCAGCAGATATCATTAGC-3' and 5'-TGCTACAATGGCTTCA-3'for RNAIII. The design of the forward and reverse primers for RNAIII was based on the sequences obtained from genome walking. Amplicons were gel purified using QIAquick gel extraction kit (QIAGEN) before the copy number was calculated. The real-time PCR templates for each gene included fresh dilutions of the standards, cDNA samples, and distilled water for a negative control. A control for each primer pair and RNA sample also included sham cDNA synthesis reactions that lack reverse transcriptase, followed by PCR amplification to identify RNA preparations contaminated by residual genomic DNA. All reactions were performed in triplicate in a DNA Engine Opticon 2 system (MJ Research, Waltham, MA). Each reaction tube contained 10 μl of 2x Quantitect SYBR green PCR master mix (HotStarTaq DNA polymerase, Quantitect SYBR green PCR buffer, dNTP mix, SYBR green I, and 5 mM MgCl2), 0.3 μM of gene-specific forward and reverse primers, and 2 μl template, made up to a final volume of 20 μl with distilled water. Cycling parameters were as follows: initial activation step at 95°C for 15 min, denaturation at 94°C for 15 s, annealing at 55°C for 30s, extension at 72°C for 30 s, and fluorescence data acquisition at 72°C for 15 s. Melting curve analysis was performed at 50°C to 95°C with stepwise fluorescence acquisition at every 1°C s–1. Melting curves observed for each gene were confirmed to correspond to the correct amplicon size by agarose gel electrophoresis of PCR products (not shown). Sequence-specific standard curves were generated using 10-fold serial dilutions (103 to 107 copies μl–1) of the specific RNA standards. On the basis of the given copy number values for amplicon standards, the Opticon Monitor analysis software (version 2.01) generated a standard curve that was used to quantitate representative gene cDNA copy numbers in each sample. The mean cDNA copy number values obtained for each gene were divided by the corresponding mean 16S rRNA values to standardize for the number of bacterial cells present in each sample. Relative comparisons between corrected values were performed using the analysis of variance test in GraphPad Prism software package (version 4.0) applying the Bonferroni posthoc test to allow multiple comparisons for significance.
Nucleotide sequence accession number. The DNA sequence of the 3.436-kb fragment has been deposited in GenBank under the accession number AY965912.
RESULTS AND DISCUSSION
Cloning and sequencing of the agrD gene revealed that the 20 canine S. intermedius strains tested could be divided into three agr specificity groups based on the putative autoinducing peptide produced by each strain. Elucidation of the complete 3' agr (RNAIII) and the upstream agr flanking sequences was achieved using primers U1 and AP2 and primers U2 and AP2. For the downstream region of agrD, nested PCR gave a 5.5-kb amplified fragment. By repeated sequential PCR and primer walking, the entire agr determinant and further downstream flanking sequences were obtained (GenBank accession number AY965912) (Fig. 1).
Through genome walking, the agr locus in S. intermedius was found to span a total of 3,436 bp, with five ORFs, agrB, agrD, agrC, agrA, and hld, in two divergently transcribed transcripts, RNAII and RNAIII. The start codons of agrB, agrD, and agrC of S. intermedius are preceded by the sequence 5'-AGGAGGA-3', which is similar to the typical consensus ribosome-binding site 5'-AGGAGGT-3' in gram-negative bacteria. For agrA, the ribosome-binding site is 5'AGGATGT-3'.
The amino acid sequences of AgrB, AgrD, AgrC, and AgrA of S. intermedius were compared to those of known Agr proteins from other staphylococci, and analyzed by manual examination and publicly available computer programs. Only the agr loci from S. aureus group I (GenBank accession number X52543), S. aureus group II (GenBank accession number AP003135), S. aureus group III (GenBank accession number AP004829), S. epidermidis (GenBank accession number Z49220), Staphylococcus lugdunensis (GenBank accession number AF173933), and S. intermedius (GenBank accession number AY965912, this study) were incorporated in the multiple-alignment analysis, since sequence data for other staphylococci are incomplete. The identified agrB, agrD, and agrC regions of RNAII of S. intermedius share some similarity with the corresponding sequences of other staphylococci. However, regions of difference are found, and those regions of RNAII that show little similarity among the staphylococcal species are thought to be important for species-specific peptide processing and recognition involved in the agr response.
AgrA of S. intermedius. In the 3' region of RNAII, high sequence conservation throughout the agrA genes of S. intermedius and S. aureus is evident, with 83.6% identity. Evidence for the role of AgrA as a response regulator can also be inferred from analyses of its sequences using reverse position-specific BLAST (rpsBLAST) and the Superfamily 1.63 HMM library and genome assignments server. Analysis using ConPredII found no likely transmembrane helices in the sequence of AgrA of S. intermedius, in agreement with the expected cytoplasmic localization of response regulators (not shown). Being a putative response regulator, AgrA of S. intermedius is likely to have little direct involvement with the specificity of processing and of the ligand-receptor interactions that constitute the putative quorum-sensing circuits, which may account for the low sequence divergence among the AgrA proteins of various staphylococci.
AgrB of S. intermedius. Analysis of the amino acid sequence (188 residues) of the AgrB protein of S. intermedius reveals that it shares 41.5% identity to the AgrB proteins of S. aureus, S. lugdunensis, and S. epidermidis. Transmembrane topology analysis predicted the AgrB of S. intermedius to be a membrane-spanning protein with five potential helices, consistent with the proposed function of AgrB as seen in S. aureus, where it is responsible for processing of the AIP expressed by AgrD. Since AIP is derived from an internal fragment of the agrD-coding sequence and its synthesis involves the product of the agrB gene (15), it would be expected that polymorphism in the agrD gene would require agrB to vary accordingly, so as to retain the specificity of peptide processing. This may account for the high sequence variation of AgrB in the various staphylococcal species.
AgrC of S. intermedius. The S. intermedius AgrC amino acid sequence (433 residues) shows 15.7% identity with analogous products from S. aureus. Transmembrane topology analysis of AgrC of S. intermedius suggested the presence of a N-terminal hydrophobic domain. A search of conserved domains (rpsBLAST) revealed that the C terminus of AgrC is categorized as a member of the histidine protein kinase family. As agrC presumably encodes a receptor responsible for binding to the autoinducing peptide derived from agrD, divergence in the N terminus of the agrC of staphylococci would allow binding to its corresponding polymorphic AIP, thus retaining the specificity of the receptor-ligand interaction. This divergence may be responsible for speciation within staphylococci, similar to divergence within the structurally related comAP locus required for competence in Bacillus (33).
AgrD of S. intermedius. The short S. intermedius AgrD amino acid sequence (45 residues) shows only 24.4% similarity with the equivalent prepheromone of S. aureus. Characterization of the AIP of S. aureus by Ji and colleagues revealed that it is a modified peptide whose sequence is contained within an internal fragment of the agrD-coding sequence (15). On the basis of the primary sequence of the pheromone peptides determined from PCR analysis, four different agr subgroups have been classified in S. aureus (15, 26), and one group has been detected in 15 investigated strains of S. epidermidis (27). Our similar PCR sequence analyses of the agrD genes among 20 strains of S. intermedius isolated from clinical cases of canine pyoderma revealed the distribution of three AIP allelic variants. Among the 20 strains, 5% (1 strain) produce the AIP RIPTSTGFF, while 35% (7 strains) and 60% (12 strains) produce the RIPISTGFF (this study) and KIPTSTGFF peptides, respectively (Table 1). The AIP gene in the three different groups of S. intermedius was highly conserved and shows nucleotide polymorphisms that led to conservative amino acid changes only.
With S. aureus, the highly divergent AIP sequences in the different groups vary in length from seven to nine amino acyl residues. Through extensive structure-activity relationship studies, the AIP is found to be a thiolactone molecule containing a ring of five amino acids formed by a thioester linkage between the sulfhydryl group of a cysteine residue and the C-terminal carboxyl group, and a tail ranging from two to four amino acid residues, depending on the staphylococcal species and groups within the same species (13, 15, 22, 27). Prediction of the S. intermedius AIPs from the corresponding nucleotide sequences of agrD revealed that in contrast to the AIP sequences in the four groups of S. aureus, there are no conserved sequences of five cysteine residues from the COOH terminus in the putative AIPs of all three specificity groups in S. intermedius. This is consistent with the findings by Kalkum and colleagues, who reported that instead of a thiolactone structure, a lactone structure was discovered in a pigeon isolate of S. intermedius (17). Despite the high divergence of the agrD sequences among the different staphylococcal species, there is nonetheless a more or less conserved negatively charged C terminus. This negatively charged terminus has been proposed to be involved in the processing of AgrD (35), and it contributes to the translocation of the AIP (15).
RNAIII of S. intermedius. Upstream of the transcription start site, putative –10 and –35 promoter elements, similar to those in other staphylococci, were found in S. intermedius (Fig. 2). A direct repeat immediately upstream of the –35 promoter element was found in all species of staphylococci. Sequence conservation among these elements suggests their functionally important roles in regulating RNAIII gene expression (24), with similar control mechanisms likely to exist in all staphylococcal species. The 3' regions of the RNAIII sequences of the different staphylococcal species also showed a considerable degree of sequence conservation (Fig. 2). In S. aureus, the 3' region was found to be essential for the production of exoproteins, such as and toxin (18). Similarity in this region among the different staphylococci suggests that they may play roles to that of S. aureus.
The RNAIII gene also contains an open reading frame encoding the delta-lysin homologue. Delta-lysin is encoded by the hld gene, which gives rise to a transcript of 75 nucleotides in S. intermedius. The predicted delta-lysin genes of the various staphylococcal species are found at different distances from the transcription start point, with most being localized in the 5' half of RNAIII (Fig. 2). In S. intermedius, there is an additional insertion upstream of the structural gene for the delta-lysin open reading frame (Fig. 2). The open reading frame of the delta-lysin in S. intermedius possesses a reasonably high degree of sequence identity to those of S. aureus and other coagulase-negative staphylococci (Table 2). In S. intermedius, 16 of the 25 amino acids of delta-lysin are identical to those of S. aureus. From the alignment of the primary sequence of delta-lysin in Table 2, it can be seen that the distribution of the charged residues was conserved between all predicted molecules. Consequently, it is likely that the delta-lysin of S. intermedius may function as a pore-forming toxin, as has been reported for S. aureus delta-lysin (8).
Fitton et al. and Turner et al. determined the amino acid sequence of delta-lysin of S. aureus from human and canine isolates and found that the toxin obtained from S. aureus of the human strain differs from that made by the canine strain by nine amino acids (Table 2) (7, 34). Interestingly, the delta-lysin from a canine isolate of S. intermedius (this study) compared with that from a canine isolate of S. aureus (34) differs only by a single amino acid (Table 2). Given the higher similarity between S. aureus and S. intermedius of canine origin compared to that between S. aureus of human and canine origin, one might question the true identity of the suggested S. aureus of canine origin.
Temporal regulation of RNAIII, leukotoxin, and enterotoxin C expression. Despite differences in the primary nucleotide sequences of RNAIII from S. aureus and S. intermedius, the kinetics of transcription in both are comparable—notably, the expression of RNAIII reaches its maximum in late logarithmic phase of growth (Fig. 3A). The temporal regulation of RNAIII suggests that activation of the P3 promoter, which drives the expression of RNAIII, is growth phase dependent. Regulation of RNAIII in an autocrine manner is consistent with its hypothesized function as the effector of the agr system. Expression of enterotoxin C and leukotoxin were similarly temporally regulated, displaying a significant increase in the amount of transcripts in the post-exponential phase (Fig. 3B). Thus far, there is very little data available concerning the relationship between S. intermedius cutaneous infection and the production of leukotoxin and enterotoxin, respectively, because little is known about the local and general effects of the toxin products in vivo. The luk gene was chosen for this study, as all 20 clinical isolates of S. intermedius isolated from cases of canine pyoderma were found to harbor the two cotranscribed genes of leukotoxin, lukF and lukS (not shown), suggesting the likely role and importance of this toxin in the virulence of S. intermedius, at least in cutaneous infection. Staphylococcal enterotoxins, on the other hand, are known to have potent effects on cells of the immune system in general, functioning as superantigens to cause excessive nonspecific T-cell proliferation. In S. intermedius, the expression of both virulence factors followed a trend similar to that of RNAIII and was positively correlated (RNAIII versus entC, r = 0.8191 and P = 0.0898; RNAIII versus luk, r = 0.9001 and P = 0.0373), suggesting that changes in the regulation of both toxins studied may be attributed to the differential transcription of RNAIII.
Characterization of the activating factor for luk, entC, and RNAIII expression in S. intermedius. Given that the AgrC protein of S. intermedius resembles the classical sensory transducing proteins of bacteria, it seemed likely that agr, and hence RNAIII activation, would be initiated by an environmental signal generated during bacterial growth. To investigate whether quorum sensing via the agr gene was responsible for the exponential-phase induction of toxin transcription, the effects of cell-free spent culture supernatants obtained from different phases of growth from S. intermedius strains 3 and 27 and S. aureus strain J3945 on gene expression in S. intermedius strain 3 were assessed. The spent media collected were individually added to early exponential cultures of S. intermedius strain 3 at a time when expression of each gene studied was minimally activated, as determined in preliminary studies (not shown). The results obtained demonstrate that post-exponential-phase culture supernatants of S. intermedius contain a substance that, when added to early exponential self cultures, caused immediate activation of RNAIII, enterotoxin C, and leukotoxin transcription, with expression of RNAIII and both virulence factors being positively correlated (RNAIII versus entC, r2 = 0.9705 and P = 0.0295; RNAIII versus luk, r2 = 0.9821 and P = 0.0179) (Fig. 4). Exhibition of a concentration-dependent activation suggests that the production of virulence factors in S. intermedius is initiated by a growth-specific cue generated during bacterial growth. The activating substance appears to accumulate in the supernatant during the growth of S. intermedius, and upon reaching a threshold level, it regulates virulence factor gene expression. This type of kinetics represents a cell density-sensing mechanism similar to that observed with Pseudomonas, Vibrio, Agrobacterium, and other bacterial species that use homoserine lactones as autoinducers (5, 16, 36).
The abundance of mRNA from S. intermedius RNAIII, leukotoxin, and enterotoxin genes demonstrated dependency on secreted biological cues. With self biological cues, activation of the transcripts encoding three genes studied occurred when supernatants obtained from late-exponential growth of S. intermedius strain 3 (t = 8 h) were added to the same growing strain (Fig. 4). When post-exponential supernatant from a different strain, S. intermedius strain 27 (S27), was added to S. intermedius strain 3 cells, there was higher expression of RNAIII (P < 0.01) and luk (P < 0.05) than when the medium-only control was added (Fig. 4). To verify that activation of the above-mentioned transcripts following the addition of post-exponential-phase spent medium is due to specific signals elaborated by the bacterium itself, cyclic peptides were synthesized, according to the sequenced putative AIPs of strain 3 (self) and S27 (nonself) in order to test their efficacy to induce gene expression in S. intermedius strain 3. Up-regulation of RNAIII (P < 0.001), luk (P < 0.001), and entC (P < 0.001) expression was observed following the addition of "self" peptide (RIPISTGFF) to S. intermedius strain 3. With respect to the "nonself" peptide (KIPTSTGFF), there was no convincing activation of the agr response, in contrast to that observed when nonself spent culture medium (S27) was added to S. intermedius strain 3. The difference could be attributable to changes in environmental factors as S. intermedius enters the post-exponential phase, such as alterations in the levels of acidity, or to the presence of nonspecific endogenous substances in the spent culture medium.
With S. aureus and S. epidermidis, although the various peptide groups differ in their amino acid sequences, each has been shown to possess a cyclic thiolactone moiety (15, 22, 27). Previous studies have reported the essentiality of the high-energy thioester linkage between the central cysteine and the C-terminal carboxyl group in the thiolactone ring of these AIPs for full agr activation activity (20, 22, 23, 27). Despite being potent intergroup inhibitors, the corresponding lactone and lactam analogues of the group II AIP of S. aureus were also found to be inactive, as were linear version of this and other AIPs (14, 22). In our study, the addition of "self" synthetic peptide (RIPISTGFF) with a lactone cyclic structure to early exponentially growing S. intermedius activated transcription of RNAIII. The putative nonapeptide, RIPISTGFF, is thus the activating factor that turns on the agr response in S. intermedius strain 3. Results obtained from this study also suggest that the thiolactone ring is not an absolute requirement for agr activation in S. intermedius, as was proposed for other staphylococci.
Taken together, the response defined on the basis of increases in transcript levels of quorum-sensing-controlled genes, such as RNAIII of agr, indicates that S. intermedius is an active participant in activating its own virulence regulatory system. This type of regulation is a form of quorum sensing, whereby a population of bacteria responds in concert when a critical cell density is reached. The identification of the agr gene in S. intermedius is an important step towards understanding its pathogenesis. Given the similarities in the orientation, organization, and conservation of certain residues and domains of the agr gene of S. intermedius to those of S. aureus, it is likely that the agr system may play a comparable role in the pathology of S. intermedius canine pyoderma.
ACKNOWLEDGMENTS
This work was supported by a grant from Nestle Purina Petcare.
Present address: Department of Cellular and Molecular Medicine, St. George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, United Kingdom.
REFERENCES
1. Allaker, R. P., A. I. Lamport, D. H. Lloyd, and W. C. Noble. 1991. Production of virulence factors by Staphylococcus intermedius isolates from cases of canine pyoderma and healthy carriers. Microb. Ecol. Health Dis. 4:169-173.
2. Arai, M., H. Mitsuke, M. Ikeda, J. X. Xia, T. Kikuchi, M. Satake, and T. Shimizu. 2004. ConPred II: a consensus prediction method for obtaining transmembrane topology models with high reliability. Nucleic Acids Res. 32:W390-W393.
3. Colombini, S., S. R. Merchant, and G. Hosgood. 2000. Microbial flora and antimicrobial susceptibility patterns from dogs with otitis media. Vet. Dermatol. 11:235-239.
4. Dowling, P. M.1996 . Antimicrobial therapy of skin and ear infections.Can. Vet. J. 37:695-699.
5. Dunlap, P. V., and J. M. Ray. 1989. Requirement for autoinducer in transcriptional negative autoregulation of the Vibrio fischeri luxR gene in Escherichia coli. J. Bacteriol. 171:3549-3552.
6. Dziewanowska, K., V. M. Edwards, J. R. Deringer, G. A. Bohach, and D. J. Guerra. 1996. Comparison of the beta-toxins from Staphylococcus aureus and Staphylococcus intermedius. Arch. Biochem. Biophys. 335:102-108.
7. Fitton, J. E., A. Dell, and W. V. Shaw.1980 . The amino acid sequence of the delta haemolysin of Staphylococcus aureus. FEBS Lett. 115:209-212.
8. Freer, J. H., and T. H. Birkbeck. 1982. Possible conformation of delta-lysin, a membrane-active peptide of Staphylococcus aureus. J. Theor. Biol. 94:535-540.
9. Greene, R. T., and C. H. Lammler. 1993. Staphylococcus intermedius: current knowledge on a pathogen of veterinary importance. J. Vet. Med. B 40:206-214.
10. Hajek, V. 1976. Staphylococcus intermedius, a new species isolated from animals. Int. J. Syst. Bacteriol. 26:401-408.
11. Hendricks, A., H. J. Schuberth, K. Schueler, and D. H. Lloyd. 2002. Frequency of superantigen-producing Staphylococcus intermedius isolates from canine pyoderma and proliferation-inducing potential of superantigens in dogs. Res. Vet. Sci. 73:273-277.
12. Ihrke, P. J. 1987. An overview of bacterial skin disease in the dog. Br. Vet. J. 143:112-118.
13. Jarraud, S., G. J. Lyon, A. M. Figueiredo, G. Lina, F. Vandenesch, J. Etienne, T. W. Muir, and R. P. Novick. 2000. Exfoliatin-producing strains define a fourth agr specificity group in Staphylococcus aureus. J. Bacteriol. 182:6517-6522.
14. Ji, G., R. C. Beavis, and R. P. Novick.1995 . Cell density control of staphylococcal virulence mediated by an octapeptide pheromone. Proc. Natl. Acad. Sci. USA 92:12055-12059.
15. Ji, G., R. Beavis, and R. P. Novick. 1997. Bacterial interference caused by autoinducing peptide variants.Science 276:2027-2030.
16. Jones, S., B. Yu, N. J. Bainton, M. Birdsall, M. Byceoft, B. W. Chhabra, S. R. Cox, A. J. R. Golby, P. Reeves, P. J. Staphens, S. Winson, M. K. Salmond, G. P. C. Stewart, and P. Williams.1993 . The lux autoinducer regulates the production of exoenzyme virulence determinants in Erwinia carotovora and Pseudomonas aeruginosa. EMBO J. 12:2477-2482.
17. Kalkum, M., G. J. Lyon, and B. T. Chait.2003 . Detection of secreted peptides by using hypothesis-driven multistage mass spectrometry. Proc. Natl. Acad. Sci. USA 100:2795-2800.
18. Kornblum, J., B. N. Kreiswirth, S. J. Projan, H. Ross, and R. P. Novick. 1990. Agr: a polycistronic locus regulating exoprotein synthesis in Staphylococcus aureus, p. 373-402. In R. P. Novick (ed.), Molecular biology of staphylococci. VCH Publisher, New York, N.Y.
19. Lyon, G. J., P. Mayville, T. W. Muir, and R. P. Novick.2000 . Rational design of a global inhibitor of virulence response in Staphylococcus aureus, based in part on localization of the site of inhibition to the receptor-histidine kinase, AgrC. Proc. Natl. Acad. Sci. USA 97:13330-13335.
20. Lyon, G. J., J. S. Wright, A. Christopoulos, R. P. Novick, and T. W. Muir. 2002. Reversible and specific extracellular antagonism of receptor-histidine kinase signalling. J. Biol. Chem. 227:6247-6253.
21. Marrack, P., and J. Kappler. 1990. The staphylococcal enterotoxins and their relatives. Science 248:705-711.
22. Mayville, P., G. Ji, R. Beavis, H. Yang, M. Goger, R. P. Novick, and T. Muir. 1999. Structure-activity analysis of synthetic autoinducing thiolactone peptides from Staphylococcus aureus responsible for virulence. Proc. Natl. Acad. Sci. USA 96:1218-1223.
23. McDowell, P., Z. Affas, C. Reynolds, M. T. Holden, S. J. Wood, S. Saint, A. Cockayne, P. J. Hill, C. E. Dodd, B. W. Bycroft, W. C. Chan, and P. Williams. 2001. Structure, activity and evolution of the group I thiolactone peptide quorum sensing system of Staphylococcus aureus. Mol. Microbiol. 41:503-512.
24. Morfeldt, E., K. Tegmark, and S. Arvidson. 1996. Transcriptional control of the agr-dependent virulence gene regulator, RNAIII, in Staphylococcus aureus. Mol. Gen. Genet. 248:446-458.
25. Novick, R. P., H. F. Ross, S. J. Projan, J. Kornblum, B. Kreiswirth, and S. Moghazeh. 1993. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J. 12:3967-3975.
26. Otto, M., H. Echner, W. Voelter, and F. Gotz. 2001. Pheromone cross-inhibition between Staphylococcus aureus and Staphylococcus epidermidis. Infect. Immun. 69:1957-1960.
27. Otto, M., R. Sussmuth, C. Vuoung, G. Jung, and F. Gotz.1998 . Structure of the pheromone peptide of the Staphylococcus epidermidis agr system. FEBS Lett. 424:89-94.
28. Pearson, W. R., and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85:2444-2448.
29. Prevost, G., T. Bouakham, Y. Piemont, and H. Monteil. 1995. Characterisation of a synergohymenotropic toxin produced by Staphylococcus intermedius. FEBS Lett. 376:135-140. (Erratum, 381:272, 1996.)
30. Projan, S. J., and R. P. Novick. 1997. The molecular basis of pathogenicity, p.55 -82. In K. B. Crossley and G. L. Archer (ed.), The staphylococci in human diseases. Churchill Livingstone, New York, N.Y.
31. Scott, D. W., W. H. Miller, Jr., and C. E. Griffin. 2001. Bacterial skin diseases, p.275 -335. In D. W. Scott, W. H. Miller, Jr., and C. E. Griffin (ed.),Small animal dermatology , 6th ed. W. B. Saunders Company, Philadelphia, Pa.
32. Terauchi, J., H. Sato, T. Hasegawa, T. Yamaguchi, C. Aizawa, and N. Maehara. 2003. Isolation of exfoliative toxin from Staphylococcus intermedius and its local toxicity in dogs. Vet. Microbiol. 94:19-29.
33. Tortosa, P., and D. Dubnau. 1999. Competence for transformation: a matter of taste. Curr. Opin. Microbiol. 2:588-592.
34. Turner, W. H. 1978. Purification and characterization of an immunologically distinct delta-hemolysin from a canine strain of Staphylococcus aureus. Infect. Immun. 20:485-494.
35. van Wamel, W. J., G. van Rossum, J. Verhoef, C. M. Vanderbroucke-Grauls, and A. C. Fluit. 1998. Cloning and characterization of an accessory gene regulator (agr)-like locus from Staphylococcus epidermidis.FEMS Microbiol. Lett. 163:1-9.
36. Zhang, L., P. J. Murphy, A. Kerr, and M. E. Tae. 1993. Agrobacterium conjugation and gene regulation by N-acyl-L-homoserine lactones. Nature 362:446-448.(Julia M. L. Sung, Peter D)