|Home | About | Journals | Submit | Contact Us | Français|
This investigation examines the role of the SaeR/S two-component system in USA300, a prominent circulating clone of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA). Using a saeR/S isogenic deletion mutant of USA300 (USA300ΔsaeR/S) in murine models of sepsis and soft-tissue infection revealed this sensory system is critical to USA300 pathogenesis during both superficial and invasive infection. Oligonucleotide microarray and real time RT-PCR identified numerous extracellular virulence genes down-regulated in USA300Δ saeR/S. Unexpectedly, an up-regulation of mecA and mecR1 corresponded to increased methicillin-resistance in USA300Δ saeR/S. 5′-RACE analysis defined transcript start sites for sbi, efb, mecA, lukS-PV, hlb, SAUSA300_1975, and hla, to underscore a conserved consensus sequence within promoter regions of genes under strong SaeR/S transcriptional regulation. EMSA experiments illustrated direct binding of SaeRHis to promoter regions containing the conserved consensus sequence. Collectively, this investigation demonstrates SaeR/S directly interacts with virulence gene promoters to significantly influence USA300 pathogenesis.
Staphylococcus aureus is a prominent bacterial pathogen responsible for a wide range of disease in humans (1). Within the last decade there has been an emergence of virulent methicillin-resistant S. aureus (MRSA) distinct from previously characterized hospital-associated MRSA (HA-MRSA) (2). Termed community-associated MRSA (CA-MRSA) due to apparent origins from outside the healthcare setting, these strains are alarming in their ability to cause life-threatening infections in healthy individuals as well as produce diseases not generally associated with S. aureus pathogenesis (3–6). Infection caused by MRSA is a serious health concern even in developed countries; invasive MRSA disease caused over 18,000 deaths during 2005 in the United States alone (7). The recent identification of at least two separate S. aureus lineages resistant to vancomycin (8), the antibiotic of choice to treat MRSA infection, emphasizes the necessity to further understand fundamental mechanisms of S. aureus pathogenesis to promote novel treatment strategies against this bacteria.
The S. aureus genome encodes 16 putative two-component gene-regulatory systems (TCSs) identified by sequence homology that collectively monitor environment signals and adjust gene transcription in response (9–11). The SaeR/S TCS is thought to be critical for S. aureus pathogenesis by transcriptionally up-regulating numerous virulence genes in response to host specific signals (12–14), yet how this sensory system exerts influence on S. aureus transcription has remained uncertain. We have generated an isogenic saeR/S deletion mutant in USA300 (USA300Δ saeR/S) to examine the role of this TCS during disease caused by a major circulating strain of clinically relevant CA-MRSA. Collectively this investigation demonstrates SaeR/S is critical to USA300 pathogenesis by up-regulating virulence gene transcription through direct interactions with a conserved consensus sequence within gene promoters, a major step towards clearly defining the SaeR/S mediated transcriptome in S. aureus.
Staphylococcus aureus strains were cultured and harvested as described previously (14–17) at mid-exponential (ME) and early-stationary (ES) growth defined by OD600 = 0.75 or OD600 = 2.0, respectively. Escherichia coli BL21 and Top10 cells (Invitrogen) were grown in Luria Bertani broth supplemented with 100 mg/L ampicillin (LBA) (EMD chemicals). S. aureus pulse-field gel electrophoresis type USA300 strain LAC (18) was used to generate USA300Δ saeR/S as described previously (14;16;19) (Fig 1). USA400 strain MW2 and USA400Δ saeR/S were generated in prior investigations (14).
Murine models of infection were performed as previously described (14–16). All animal studies conformed to National Institute of Health guidelines and were approved by the Animal Care and Use Committee at Montana State University-Bozeman. Female CD1 Swiss and Crl;SKH1-hrBR hairless mice were purchased from Charles River Laboratories (Wilmington, MA). Survival statistics were performed using a Logrank test (GraphPad Prism version 4.0).
Oligonucleotide microarray analysis and TaqMan real-time RT-PCR was performed as previously described (14;15;17). For oligonucleotide microarrays, protocol for Affymetrix prokaryotic target preparation was followed (www.affymetrix.com/support/downloads) using the Standard format hybridized to GeneChip S. aureus Genome arrays (Affymetrix). Each experiment was repeated in triplicate. Real-time RT-PCR was performed using primers in Table 1 on at least two separate experiments with triplicate analysis of each sample.
Rapid-amplification of 5′-complementary DNA ends (5′-RACE) analysis was performed using previously defined protocols (20) with SuperScript III Reverse Transcriptase (Invitrogen Life Technologies), terminal transferase (New England Biolabs), and gene specific primers (Table 1). 5′-RACE products were visualized using 1.5% DNA agarose gel electrophoresis, excised, purified using a QIAquick PCR purification kit (Qiagen), and cloned into pCR2.1 Topo (Invitrogen). Subsequent plasmids were purified using a QIAprep spin miniprep kit (Qiagen) and sequenced with a BigDye terminator v3.1 kit (Applied Biosystems). Sequence alignment of 5′-RACE products against the USA300 genome (NCBI accession # NC_007793) was performed using nucleotide BLAST (http://blast.ncbi.nlm.nih.gov). Multiple sequence alignment was performed using CLUSTALW (http://align.genome.jp/) and a consensus sequence logo derived using weblogo (http://weblogo.berkeley.edu/). Virtual Footprint Bacterial Regulon Analyzer (http://prodoric.tu-bs.de/vfp/) and nucleotide BLAST were used to examine intergenic regions within the USA300 genome.
Expression of recombinant SaeR (SaeRHis) was achieved using saeR specific primers (Table 1) and a Champion pET Directional TOPO Expression Kit (Invitrogen). E. coli BL21 cells transformed with pET100D vector containing saeR were grown at 37°C with agitation in 4 L LBA to an OD600 = 0.8, induced with 0.5 mM IPTG, and grown for 6 more hours. Cells were harvested by centrifugation (4,000 × g and 4°C for 10 minutes), resuspended in 20 mM Tris, pH 8, (Tris-HCl) then lysed via sonication on ice for 15 minutes. Lysed cells were centrifuged (15,000 × g and 4°C for 10 minutes), samples resuspended in 60 mL 8 M Urea, 0.5 M NaCl, Tris-HCl, sonicated on ice for 15 min, and purified using NI-NTA affinity (Qiagen) and DEAE anion exchange chromatography (GE Healthcare) with a BioLogic LP Chromatography System (Bio-Rad). Purified SaeRHis was dialyzed against 3 L Tris-HCl and concentrated using a Centricon Plus-20 centrifugal filter (Millipore). Concentration of SaeRHis was determined using a micro BSA protein assay kit (Pierce). Electrophoretic mobility shift assays (EMSAs) were performed as previously described (21) using primers and probes in Table 1. SaeRHis was combined with indicated quantity of promoter sequence and 500 ng salmon sperm DNA when indicated in 20 μL binding buffer (100 mM KCl, 1 mM EDTA, 0.1 mM DTT, 5% v/v glycerol, 10 ng/mL BSA) and incubated at room temperature for 15 minutes. Samples were run on a 10% native gel and visualized using a Kodak Image Station 2000mm with excitation at 465 nm and a 535 nm filter or stained with SYBR green (Sigma-Aldrich) and viewed using a UVP BioDoc-It imaging system
As USA300 is a major cause of skin infections, we examined the role of the SaeR/S TCS in this strain using a murine model of soft-tissue infection (Fig. 2). In contrast to mice inoculated with USA300, mice inoculated with USA300ΔsaeR/S displayed a significant reduction in the incidence and severity of infection. The majority of soft-tissue infections caused by USA300 exhibited dermonecrotic lesions at the dose administered (Fig. 2A and 2B), similar to previous findings (16;22;23). Surprisingly, nearly half the mice inoculated with USA300ΔsaeR/S did not display any signs of infection and no dermonecrosis was observed (Fig. 2A and B). Moreover, the size of skin infections generated by USA300ΔsaeR/S were decreased relative to USA300 (Fig. 2C). This experiment was repeated to show bacterial loads of mice infected with USA300ΔsaeR/S were the same as USA300 at day 2 post infection but lower after day 4 (Fig. 2D).
A murine sepsis model was used to examine the influence of SaeR/S during invasive infection caused by USA300 (Fig 2E). In accordance with prior research (16), USA300 invasive infection was typically lethal within 24 hours at the dose administered. Conversely, mortality was delayed and considerably reduced in mice infected with USA300ΔsaeR/S. These findings demonstrate SaeR/S significantly influences the virulence of USA300 during both soft-tissue and invasive infection.
Gene transcription profiles of USA300ΔsaeR/S relative to USA300 were defined during in vitro growth using oligonucleotide microarray analysis (Fig 3). Approximately 1.9% of genes in USA300ΔsaeR/S exhibited a greater than two-fold change in transcription relative to USA300, including numerous hemolysins, proteases, host immunomodulatory proteins, and adhesion factors. A transcriptional up-regulation of the mecA (penicillin-binding protein 2′) and mecR1 genes that impart methicillin resistance was observed in USA300ΔsaeR/S, findings not reported for isogenic saeR/S deletion mutants in other S. aureus strains (12;14;24). However, USA300ΔsaeR/S is primarily characterized by a down-regulation of transcripts encoding a variety of secreted virulence genes relative to USA300. These include different hemolysins and leukocidins, such as α-toxin (hla), β-hemolysin (hlb), and two-component leukotoxins (hlgB-hlgC, lukE-lukD, and SAUSA300_1974-SAUSA300_1975). A down-regulation was also observed for known immunomodulatory genes, namely sbi, efb, and chs (encoding CHIPS), secreted into the extracellular environment to subvert the host immune response (25–27). The transcriptional regulator tcaR, a member of the tcaR-tcaA-tcaB operon, was down-regulated in USA300ΔsaeR/S. In addition to associations with teicoplanin resistance and cell wall biosynthesis pathways (28), tcaR is thought to be a repressor of the ica locus (29) and directly regulate sasF and sarS transcription (30). Other transcripts down-regulated in USA300ΔsaeR/S included various host-binding proteins (fnbA, fnbB, empbp, coa, and SAUSA300_1052), several lipase precursors, and the majority of the spl operon containing 6 distinct serine proteases.
TaqMan real-time RT-PCR was performed to support oligonucleotide microarray analysis and to determine the influence of SaeR/S on transcription of two genes not included on the oligonucleotide microarray (Fig 4), Panton-Valentine Leukocidin (PVL) and novel arginine catabolic mobile element (ACME) of USA300 (18). TaqMan real time RT-PCR confirmed the isogenic deletion of saeR and saeS (Fig 4A), down-regulation of sbi, efb, hla, hlb, SAUSA300_1975 (Fig 4B), and up-regulation of mecA (Fig. 4C) in USA300ΔsaeR/S. Complementation with a plasmid encoding saeR and saeS (USA300ΔsaeR/S + Comp) partially restored gene transcription in USA300ΔsaeR/S, reinforcing real-time RT-PCR and microarray results. Transcriptional regulation of lukS-PV (PVL component) was down-regulated in USA300ΔsaeR/S, though USA300ΔsaeR/S + Comp did not exhibit restored transcription of lukS-PV (Fig 4B). Transcription of arcA (ACME component) was similar between USA300 and USA300ΔsaeR/S, indicating SaeR/S is not involved with regulating ACME (Fig 4C). The up-regulation of mecA and mecR1 in USA300ΔsaeR/S relative to USA300 was coupled with significantly improved growth in sub-inhibitory concentrations of oxacillin (Fig 4D and E). Improved growth (Fig 4F and G) and a subtle increase of mecA transcription during growth in oxacillin (Fig 4H) was also observed for USA400ΔsaeR/S relative to USA400, further supporting results demonstrating a transcriptional up-regulation of methicillin-resistance genes in the absence of SaeR/S.
Sequence homology of SaeR/S with other bacterial TCSs suggests the DNA-binding response regulator SaeR controls S. aureus virulence through direct interactions with virulence gene promoters (13). To define a consensus SaeR recognition sequence specific to virulence gene promoters influenced by SaeR/S, 5′-RACE was used to first identify various transcript start sites (TSSs) in USA300 and USA300ΔsaeR/S (Fig 5). Genes shown to be positively regulated by SaeR/S produced more robust 5′-RACE products in USA300 relative to USA300ΔsaeR/S (Fig 5B). Alternatively, transcription of agrB was not influenced by SaeR/S (Fig 4C) and 5′-RACE products for agrB were indistinguishable between USA300 wt and the USA300ΔsaeR/S (Fig 5A). 5′-RACE products for mecA, a transcript up-regulated in USA300ΔsaeR/S relative to USA300, were also identical between USA300 and USA300ΔsaeR/S (Fig 5A). These results indicate the transcriptional down-regulation of sbi, efb, SAUSA300_1975, lukS-PV, hla, and hlb in USA300ΔsaeR/S likely accounts for the diminished 5′-RACE products for these genes in this strain.
5′-RACE products were excised, cloned, sequenced and aligned against the USA300 genome to define the TSS for these genes (Fig 5C and D). For hla and agrB, the TSS was found to be 330 and 18 bp upstream, respectively, from the start codon in accordance with prior research (31;32). Previously undefined TSSs of sbi, efb, SAUSA300_1975, hlb, lukS-PV, and mecA were also determined. Sequences of 5′-RACE products for lukS-PV, agrB, and mecA isolated from USA300ΔsaeR/S corresponded to those obtained from USA300. The 5′-RACE product for sbi from USA300ΔsaeR/S mapped to the intergenic region immediately upstream from 50S ribosomal protein L13 and is presumed to be a product of nonspecific primer binding.
Multiple sequence alignment performed on intergenic sequences immediately upstream from the defined TSS of genes positively regulated by SaeR/S revealed two conserved domains corresponding to the −10 and −35 promoter boxes (Fig 5C). In contrast, intergenic sequences immediately upstream from the TSS of agrB and mecA did not share the extensive homology observed for genes up-regulated by SaeR/S (Fig 5D). A consensus sequence 17 bp in length was derived from the conserved domain immediately upstream from the −35 promoter box (Fig 5E). This putative SaeR recognition sequence (SRS) has a relatively low overall G/C content and is partially palindromic. In silico examination of intergenic regions within the USA300 genome identified the SRS within putative promoter regions of numerous genes under SaeR/S regulation including coa, saeP, splA, SAUSA300_1052, SAUSA300_0215, and SAUSA300_0407 (Fig 3, Fig 5F). Investigators elsewhere have demonstrated a sae deletion mutant of strain COL exhibits a reduced expression of Ear and increased expression of Plc (24). Interestingly, the SRS was found within the putative promoter region of ear while the reverse compliment of the SRS was located within the putative promoters of plc and the phosphate starvation-inducible protein phoH (Fig 5H). Subsequent real time RT-PCR confirmed a down-regulation of ear in USA300ΔsaeR/S (Fig 4B). The SRS was also found immediately upstream from −35 promoter region of rot defined by primer extension analysis (33) and between the −35 and −10 promoter regions of the arlR/S TCS also defined by primer extension analysis (34) (Fig 5G).
EMSAs were used to demonstrate recombinant SaeRHis binds numerous virulence gene promoter regions in a concentration dependent manner (Fig. 6). Purified SaeRHis (Fig 6A) specifically bound the −35 region of the hla and sbi promoters defined by 5′-RACE in a concentration dependent manner while no binding was detected for the −35 region of the agrB promoter (Fig 6B and C). The hla promoter probe sequence contains two SRSs, the sbi promoter probe contains one SRS, while the agrB promoter probe has no SRS, paralleling the relative affinity of these probes for SaeRHis. Dissection of the hla promoter using overlapping probe sequences suggests binding of SaeRHis is most robust to the sequence containing both SRSs immediately upstream from the −35 promoter region (Fig. 6D). EMSAs indicate strong binding of SaeRHis to the arlR, saeP, and splA promoter regions (Fig 6E). SaeRHis binding was also observed for the efb, hlb, rot, lukS-PV, and SAUSA300_1975 promoter regions while no binding was detected for the mecA promoter region (Fig 6E). These findings establish direct SaeR-DNA interaction specific to virulence gene promoters containing the SRS.
Cumulative evidence indicates S. aureus virulence is determined by the coordinated expression of specific virulence gene ensembles that further bacterial survival and dissemination (2;15;16;35–38). S. aureus sensory mechanisms are thought to be primarily composed of 16 TCSs that dictate gene expression in response to specific environmental signals (9–11). TCSs offer intriguing drug targets; inhibition of these sensory systems may effectively “blind” S. aureus and result in the reduced expression of numerous virulence genes (39). Indeed, preliminary investigations by others demonstrates targeted inhibition of the agr quorum-sensing TCS attenuates soft-tissue infections caused by S. aureus (40;41). The SaeR/S TCS is thought to be critical during S. aureus pathogenesis by up-regulating expression of numerous secreted virulence genes during infection. As approximately 22% of the S. aureus genome can differ between isolates (42) and virulence appears to be largely dictated by differences in expression of common genes between strains (36), we verified the importance of SaeR/S in clinically relevant USA300.
Results demonstrating SaeR/S significantly influences virulence during soft-tissue infections caused by USA300 were unexpected given previous findings showing this TCS is dispensable during soft-tissue infections caused by CA-MRSA strain USA400 (14). Compared to USA400, the SaeR/S mediated transcriptome in USA300 is almost exclusively composed of extracellular virulence genes with SaeR/S exerting a stronger regulatory influence on many of these genes. For example, hla transcription was reduced by over 33-fold in USA300ΔsaeR/S compared to a 4-fold reduction of hla transcription in USA400ΔsaeR/S (14). Others have shown in vitro transcription of α-toxin (hla), agr, saeR/S, and sarA is higher in USA300 relative to USA400 (37). Thus it is likely differences in genomic content and extent of USA300 SaeR/S transcriptional regulation on common virulence gene expression that accounts for the significant influence of this TCS on virulence and pathogen survival during USA300 soft-tissue infections.
The USA300 SaeR/S mediated transcriptome is primarily composed of secreted virulence genes that interact directly with host components including numerous hemolysins and leukocidins, immunomodulatory genes, cell surface fibronectin binding proteins, lipases, nucleases and proteases similar to findings in other S. aureus strains (14;24;43). We also report an up-regulation of mecA and mecR1 in USA300ΔsaeR/S that corresponds to improved in vitro growth in the presence of subinhibitory concentrations of oxacillin, findings supported by the improved growth of USA400ΔsaeR/S relative to USA400 in the presence of oxacillin. Direct binding could not be detected between SaeRHis and the mecA promoter region defined by 5′-RACE analysis, suggesting an indirect up-regulation of mecA in USA300ΔsaeR/S by SaeR/S. The down-regulation of the transcriptional regulator tcaR in USA300ΔsaeR/S, involved with cell wall biosynthesis and implicated in methicillin resistance (28), may account for the differential regulation of mecA and mecR1.
We identified a conserved palindromic consensus sequence 17 bp in length specific to promoters of genes regulated by SaeR/S. EMSAs revealed SaeRHis directly bound various virulence gene promoters containing the conserved consensus sequence while no binding was detected for promoter regions lacking this sequence motif. Corresponding to published findings in other SaeR/S deficient S. aureus strains (12;14;24), a dramatic down-regulation of saeP was observed in USA300ΔsaeR/S. Not surprisingly, the SRS was identified within the previously defined P1 primary start site for the SaeR/S transcript (44). EMSAs demonstrated direct binding of SaeRHis to this promoter region, supporting a positive feedback loop of SaeR/S transcription suggested by others (12). The promoter region of lukS-PV was defined using 5′-RACE and weak binding of SaeRHis to this promoter sequence reinforces recent work indicating SaeR/S positively regulates luk-PV transcription (45). The SRS was also located between the −35 and −10 promoter regions of the arlR/S operon (34) and strong binding of SaeRHis to this promoter region was observed. The ArlR/S TCS is thought to negatively regulate transcription of numerous secreted virulence genes including hla, hlb, and coa, implicating a negative-feedback loop on secreted virulence gene expression (34). Additionally, the SRS specifically mapped to the region immediately upstream from the −35 promoter box of the repressor of secreted virulence genes, rot (46), defined by others using primer extension analysis (33). Recent work on strain COL has demonstrated the repression of hla expression by rot requires saeR/S and that rot downregulates the expression of SaeR/S (47). Weak binding of SaeRHis to the rot promoter supports the hypothesis made by others (47) that SaeR/S mediated transcription of rot acts as a negative feedback loop to curtail over expression of SaeR/S and secreted virulence genes positively regulated by SaeR/S. We were unable to identify a differential regulation of the rot or arlR transcription in USA300ΔsaeR/S during in vitro growth. However, because SaeR/S is thought to be activated by host tissue-specific signals it follows that changes in transcription of virulence gene regulators controlled by SaeR/S may only be resolved during growth in vivo. Taken together, these findings indicate a complex regulatory network of multiple feedback loops influences the SaeR/S mediated transcriptional regulation of secreted virulence genes in USA300.
This investigation is the first to examine the role of SaeR/S in USA300 pathogenesis and demonstrates this TCS is important during both invasive and superficial disease caused by this clinically relevant strain. Our results define the TSSs for sbi, efb, SAUSA300_1975, hlb, lukS-PV, and mecA and are congruent with previously established TSSs for hla and agrB (31;32). A conserved partially palindromic SRS was identified within virulence gene promoters transcriptionally regulated by SaeR/S and EMSAs established direct binding of SaeRHis to promoter regions containing the conserved consensus sequence. Collectively these findings demonstrate SaeR/S plays a major role during USA300 pathogenesis of both superficial and invasive infections by directly binding the SRS within virulence gene promoters to up-regulate transcription of a precise exoprotein array that furthers S. aureus survival and dissemination.
We would like to thank Kate McInnerney and Dr. Mensur Dlakic, Department of Microbiology Montana State University – Bozeman, for advice and technical support.
Financial support: This work was supported by NIH-PAR98-072 and NIH-NCRR grants P20RR020185, and P20RR16455-07 (T. K. N., J.M.V.), an equipment grant from the M.J. Murdock Charitable Trust and the Montana State University Agricultural Experimental Station.
Potential Conflicts of Interest: The authors have declared that there are no competing financial interests.
Presented in Part: The Wind River Conference on Prokaryotic Biology, Annual Conference June 3-7, Estes Park, CO