|Home | About | Journals | Submit | Contact Us | Français|
Methicillin-resistant Staphylococcus aureus (MRSA) is problematic both in hospitals and the community. Currently, we have limited understanding of mechanisms of innate immune evasion used by S. aureus. To that end, we created an isogenic deletion mutant in strain MW2 (USA400) of the saeR/S two-component gene regulatory system and studied its role in mouse models of pathogenesis and during human neutrophil interaction. In this study, we demonstrate saeR/S plays a distinct role in S. aureus pathogenesis and is vital for virulence of MW2 in a mouse model of sepsis. Moreover, deletion of saeR/S significantly impaired survival of MW2 in human blood and after neutrophil phagocytosis. Microarray analysis of genes influenced by saeR/S demonstrated SaeR/S of MW2 influences a wide variety of genes with diverse biological functions. These data shed new insight into how virulence is regulated in S. aureus and associates a specific staphylococcal gene-regulatory system with invasive staphylococcal disease.
Staphylococcus aureus (S. aureus) is a leading cause of human infections worldwide. It is the causative agent of diverse diseases ranging in severity from mild to life-threatening (1). Recently, there has been an increase in the incidence of community-associated methicillin-resistant S. aureus (CA-MRSA) infections in otherwise healthy individuals. CA-MRSA primarily cause skin and soft-tissue infections, but the most prominent strains, classified as pulsed-field gel electrophoresis types USA300 and USA400, have also been associated with severe syndromes or pathologies, including septicemia, necrotizing pneumonia, and necrotizing fasciitis (2–6).
Human polymorphonuclear leukocytes (PMNs or neutrophils) are the first line of defense against bacterial infections. It follows that the ability of S. aureus to circumvent destruction by innate immunity includes survival after PMN phagocytosis (7). The ability of S. aureus to survive following PMN phagocytosis is dependent on the pathogen’s ability to moderate the hostile PMN environment. However, specific mechanisms used by S. aureus to survive following PMN phagocytosis are incompletely defined.
In previous studies, we used oligonucleotide and cDNA microarrays to identify complex transcriptional regulation used by S. aureus and group A Streptococcus (GAS) to survive after PMN phagocytosis (7;8). These studies revealed a role for two-component gene-regulatory systems during host cell-pathogen interaction. Several two-component systems were up-regulated by S. aureus during human PMN phagocytosis, including the saeR/S gene-regulatory system (7). The Sae regulatory system was designated Sae for “S. aureus exoprotein expression” due to altered exoprotein production in sae mutant strains (9–11). Several investigators have demonstrated a role for SaeR/S in regulation of virulence factors (9;12–18). However, the role of saeR/S in evasion of innate immunity has not been directly investigated. To that end, we generated a saeR/S isogenic deletion mutant in MW2 (USA400), the prototype CA-MRSA strain, and investigated the fundamental role of saeR/S during innate immune evasion.
To delete saeR/S in MW2, PCR amplified regions flanking the saeR/S locus were cloned into a spectinomycin cassette containing plasmid pBT2spec using a previously published protocol (24). DNA fragments upstream (PCR fragment 1) of saeR/S (5’- GCAACCCATGAGCTCAAACACTTCCTGTTCAC-3’ and 5’- CCGCTAGTTGTCGTTGTTACTTTGGATCCTTCATATTC-3’) and downstream (PCR fragment 2: 5’-GCAGTCGACTAGATGATGTAGGAACTACGATGACTGTAACATTAC-3’ and 5’- CTATTTGATAAAACAATACTCAGGTACAAGCTTAATCTTTTAAATAAAAAGGATG-3’ were amplified by PCR (restriction sites underlined). PCR product was transformed into TOPO TA PCR 2.1 cloning vector (TOPO) (Invitrogen Life Technologies) and sequence of purified product was verified by nucleotide sequencing. Left and right fragments were ligated to plasmid pBT2spec (24;25). The resulting plasmid pBTΔsaeR/S was transformed sequentially into E. coli, S. aureus strains RN42 and MW2. The MW2 construct was used for allelic replacement as described (24). Lack of saeR/S transcript in the saeR/S mutant strain (ΔsaeR/S) was verified by TaqMan real-time PCR (ABI 7500 Thermocycler Applied Biosystems, Foster City, CA) (figure 5A) and by PCR (figure 1). For the saeR/S complemented strain (compΔsaeR/S), saeR/S genes were cloned into TOPO using primers 5'- GGTATAAGTGGATCCTCGCAAATATAGTTGCACATACGAC-3' and 5'- GCCCTCATTAATGGGAGCAAGCTTTTAGTCTTTGC-3', cloned into plasmid pRB473 (24), and transformed into S. aureus strain ΔsaeR/S.
All studies conformed to NIH guidelines and were approved by the Animal Care and Use Committee at Rocky Mountain Laboratories, NIAID and at Montana State University - Bozeman. Female CD1 Swiss and Crl:SKH1-hrBR hairless mice were purchased from Charles River Laboratories (Wilmington, MA). MW2 wild-type and ΔsaeR/S S. aureus strains were cultured to mid-exponential phase of growth, washed twice with sterile DPBS, and resuspended in DPBS at 108 / 100 µL. Mice were inoculated with MW2 or ΔsaeR/S, and control animals received DPBS. Fifteen mice were used for each strain and/or control group in each model.
For the sepsis model, CD1 Swiss mice were inoculated via tail vein injection with 108 S. aureus in 100 µL as reported previously (7;7;23). Mice were monitored every 2 h for the first 48 h. If animals were unable to eat or drink or became immobile they were euthanized. Survival statistics were performed using a Logrank test (GraphPad Prism version 4.0 for Windows, GraphPad Software, Inc.).
For the abscess model, Crl:SKH1-hrBR mice were inoculated by subcutaneous injection in the right flank with 107 S. aureus or DPBS. Abscess size was calculated using the formula for a spherical ellipsoid (v + (π / 6) · l × w2) (26). To determine bacterial burden in abscesses, mice were inoculated subcutaneously with S. aureus as described above and three mice per treatment group were euthanized at designated days. Abscesses were excised, homogenized in 2 mL DPBS, and plated on tryptic soy agar for determination of cfus.
Heparinized venous blood of healthy donors was collected in accordance with a protocol approved by the Institutional Review Board for Human Subjects, NIAID and Montana State University, Bozeman, MT. All donors signed a written consent to participate in the study. MW2 and ΔsaeR/S were harvested at mid-exponential phase of growth. One mL of heparinized human blood was inoculated with ~ 105 cfu of S. aureus and incubated at 37° C for 1 and 3 h with shaking (250 rpm). Percent S. aureus survival in blood was determined by comparing cfus in each sample after 1 or 3 h to cfu at the start of the assay (T = 0 h). Statistical analysis was performed with repeated-measures ANOVA and Tukey’s posttest for multiple comparisons (GraphPad Prism version 4.0 for Windows; GraphPad Software, San Diego, CA).
PMNs were isolated under endotoxin-free conditions (< 25.0 pg/ml) as previously described (7). Cell viability and purity of preparations were assessed by flow cytometry (FACSCalibur; BD Biosciences). Cell preparations contained ~ 99% PMNs. Phagocytosis of human serum opsonized MW2 and ΔsaeR/S was synchronized by centrifugation using a published method (27) and percent ingested was determined with fluorescence microscopy as previously described (7). PMN bactericidal activity was determined as previously described (7;28). Colonies were enumerated the following day and percent S. aureus survival was determined. The assay measures viable ingested and uningested bacteria.
PMN lysis following phagocytosis of MW2 was determined with a fluorescence-based assay for release of lactate dehydrogenase (LDH) as described by the manufacturer (CytoTox-One Homogenous Membrane Integrity Assay, Promega; Madison, WI). Statistics were performed with repeated-measures ANOVA and Tukey’s posttest for multiple comparisons (GraphPad Prism version 4.0).
To compare transcript levels of MW2 and saeR/S, bacteria were grown to mid-exponential (OD600 = 0.75) and early stationary phases of growth (OD600 = 2.0), and processed for microarray analysis as described (7). S. aureus cDNA was hybridized to custom Affymetrix GeneChips (RMLChip 3) containing 99.3% coverage of genes from MW2 (2613 probe sets of 2632 ORFs remaining 0.7% represented by identical probe sets from other staphylococci). Samples were scanned according to standard GeneChip protocols (www.affymetrix.com/support/downloads/manulas/expression-s3-manual.pdf). Each experiment was repeated in triplicate. Microarray data were analyzed using GeneChip Operating Software verson 1.1 (Affymetrix) and GeneSpring version 7.0 (Silicon Genetics). Microarray data are posted on the Gene Expression Omnibus (GEO, www.ncbi.nlm.nih.gov/geo/, accession number GSE15067). To compare gene expression between MW2 and ΔsaeR/S strains, fold changes for each transcript were determined by comparing RMLChips hybridized with cDNA from MW2 to those with cDNA from ΔsaeR/S (comparisons matched by growth-phase).
RNA preparations for TaqMan analysis were performed with procedures and conditions identical to those used for microarray experiments. Relative quantification of S. aureus genes was determined by the change in expression of target transcripts relative to the housekeeping gene gyrB. Fold-change was determined by comparing transcript expression in strains MW2 and compΔsaeR/S to expression in ΔsaeR/S (according to Applied Biosystems Relative Quantification Manual). Primer probe sequences used for confirmation of saeR/S deletion and microarray results were as follows: gyrB forward primer, 5’-CAAATGATCACAGCATTTGGTACAG-3’, gyrB probe, 5’-AATCGGTGGCGACTTTGATCTAGCGAAAG-3’, gyrB reverse primer 5’- CGGCATCAGTCATAATGACGAT-3’; saeS forward primer, 5’- CGTACATTCAGAGTAGAAAACTCTCGTAATAC-3’, saeS probe, 5’- AGCCTAATCCAGAACCACCCGTT–3’, saeS reverse primer, 5’- GTTGCGCGAGTTCATTAG CTATATAT–3’, saeR forward primer, 5’- CTGCCAAAACACAAGAAC ATGATAC–3’, saeR probe, 5’- ATTTACGCCTTAACTTTAGGTGCAGAT-3’, saeR reverse primer, 5’- CTTGGACTAAATGGTTTTTTGACATAGT-3’; sbi forward primer, 5’- ATACATCAAAACATTACGCGAACAC-3’, sbi probe 5- CAGAACGTGCACAAGAAGTATTCTCTGAA-3’, sbi reverse primer 5’-CTGGGTTCTTGCTGTCTTTAAGTG-3’; orf4 forward primer 5’- TGGTGCTGTTGCCTCTGTATTAA-3’, orf4 probe, 5’-TTTAGGCGCTTGTGGTAATTCTAA-3’, orf4 reverse primer, 5’- TGTTCAGTTTTGTTACCTTGATCTTGT-3’; MW1037 forward primer, 5’-CAACGTTTGCCGGTGAATC-3’, MW1037 probe, 5’-CATGCACAAACTAAGGTTGA-3’, MW1037 reverse primer 5’-GCGTCCACAACTTTTTTATTTACTTG-3’; sec4 forward primer, 5’-ATACATCAAAACATTACGCGAACAC-3’, sec4 probe, 5’- CAGAACGTGCACAAGAAGTATTCTCTGAA-3’, sec4 reverse primer, 5’-CTGGGTTCTTGCTGTCTTTAAGTG-3’, and MW1040 forward primer 5’- CTACAATTGCGTCAACAGCAGAT-3’, MW1040 probe, 5’-CGAGCGAAGGATACGGTCCAAGAGAAA-3’, MW1040 reverse primer, 5’-ACCATCATTGTACTCTACGATATTGTGA-3’. TaqMan real-time PCR analysis was performed on two separate experiments each assayed in triplicate.
MW2 can cause a wide range of infections ranging from relatively mild skin infections to invasive and rapidly fatal disease (2;3;19). To investigate the role of saeR/S during invasive disease, we used a mouse model of staphylococcal sepsis (figure 2A). All mice infected intravenously with MW2 suffered morbidity and died within 36 h of infection. This result is consistent with previous studies (23). In contrast, only one mouse infected with ΔsaeR/S became sick and died (P < .0001). We next compared the ability of wild-type and ΔsaeR/S strains to cause abscesses and dermonecrosis (figure 2B–D). Abscess volumes in mice infected with wild-type and ΔsaeR/S strains were similar (figure 2B, P > .05 at all time points except day 5). Also, the number of mice with dermonecrosis was comparable between the strains (figure 2C). In a separate experiment, we determined the number of S. aureus cfus per abscess. There was no significant difference in cfus recovered between mice infected with MW2 and ΔsaeR/S (figure 2D). These results suggest saeR/S plays a distinct role in pathogenesis and is vital for virulence of MW2 following bloodstream infection.
To determine if the observed attenuated virulence of ΔsaeR/S was due to reduced survival in blood (i.e., killing by a component of blood), MW2 and ΔsaeR/S were cultured in whole human blood (figure 3). Although there was no significant difference in survival in whole blood between wild-type, complement, and mutant strains after 1 h of incubation, survival/growth of ΔsaeR/S was reduced significantly by 3 h (compare an average of ~ 130% survival ± 118% in the wild-type MW2 and ~ 118% ± 157% in the compΔsaeR/S strains to less than 30% survival ± 37% in the saeR/S mutant strain P < .01 versus wild-type and P < .05 versus complemented strain). The attenuated growth/survival of ΔsaeR/S in blood was not due to increased susceptibility to serum complement (data not shown). Importantly, complementation of the ΔsaeR/S strain with saeR/S (compΔsaeR/S) restored the wild-type phenotype. The reduced survival of ΔsaeR/S in human blood correlates with the dramatically reduced virulence in the sepsis model (figure 2A), and the data indicate the SaeR/S system is important for invasive S. aureus infection. Since PMNs are the major cellular component of host defense in blood, we hypothesized that the decreased survival of ΔsaeR/S in human blood is due to increased killing by PMNs.
To determine if decreased survival of the saeR/S mutant strain in blood was due to enhanced killing by human PMNs, we evaluated phagocytosis and killing of MW2 and ΔsaeR/S. There was no significant difference in the uptake of MW2 or ΔsaeR/S by human neutrophils (figure 4A). However, survival after uptake was significantly better for MW2 at all time points tested (figure 4B, e.g., survival at 1.5 h was 74.6% ± 22% for MW2 and 21.6% ± 11.2% for ΔsaeR/S, P < .001). Again, complementation of the ΔsaeR/S strain restored the wild-type phenotype (figure 4C, *P < .05 vs. MW2 and compΔsaeR/S).
Pathogenic strains of S. aureus produce numerous toxins including leukocidins (29). We have shown previously that MW2 destroys PMNs after phagocytosis (7). To determine if saeR/S contributes to the previously observed cytolytic capacity of MW2, we measured LDH release by human PMNs after phagocytosis of MW2 and ΔsaeR/S (figure 4D). Compared with the wild-type strain, lysis of human PMNs was significantly reduced during interaction with ΔsaeR/S (lysis of PMNs was 76 ± 5.7% after a 5-h incubation with MW2 versus 30% ± 12.5% for those incubated with ΔsaeR/S, P < .001).
As a first step toward identifying the saeR/S-regulated molecules that contribute to virulence (figure 2), evasion of PMN killing (figure 4B), and PMN lysis (figure 4D), we compared the transcriptomes of MW2 and ΔsaeR/S during growth in TSB (Table II, sections I–IV). Deletion of saeR/S significantly altered the MW2 transcriptome and influenced expression of ~ 212 genes (~ 8% of the MW2 genome using a ≥ 2.0 fold-change in transcript levels as a cutoff value). Transcripts encoded by ~ 3 % (80 genes) and ~ 5% (133 genes) of the MW2 genome were up- and down-regulated, respectively, in ΔsaeR/S (Table II sections I–IV). Genes encoding proteins with undefined functions comprised the largest category of genes influenced by saeR/S (~ 49% of the total genes influenced by saeR/S, Table II section III and IV). Consistent with observations made by others, deletion of saeR/S caused down-regulated expression of genes encoding virulence factors, including hla, hlgA, hlgB, hlgC, sbi, and adhesins such as fibrinogen-binding proteins (MW1040 and MW1037) (15;17). MW1037 (characterized as a hypothetical protein Table II section III) has homology to SA1000 a fibrinogen-binding protein of S. aureus that has been shown to be important in adherence and in internalization of S. aureus by endothelial cells and is regulated by saeR/S in S. aureus strain WCUH29 (15). Deletion of saeR/S in strain MW2 also altered expression of genes encoding staphylococcal exotoxin 26 (set26, down-regulated ~16.46 fold in ΔsaeR/S) and staphylococcal enterotoxin C4 (sec4, down-regulated 13.31- and 247.3-fold in ME and ES growth phases in ΔsaeR/S). Moreover, several S. aureus virulence factors previously shown to be induced during PMN phagocytosis of MW2 (7;23), were down-regulated in the saeR/S mutant strain (Table I). Collectively, over a dozen genes associated with virulence that were up-regulated after PMN phagocytosis of MW2 were down-regulated in the saeR/S mutant strain, suggesting the SaeR/S regulatory system plays an important role in S. aureus survival after PMN phagocytosis.
Transcripts involved in metabolism were also affected by deletion of saeR/S. For example, several transcripts involved in carbohydrate transport and metabolism (lacA–G were up-regulated), and amino acid transport and metabolism (urea–G were up-regulated, and arcA–D were down-regulated), were differentially regulated in ΔsaeR/S (Table II sections I and II). Several genes involved in nucleotide transport and metabolism were down-regulated in ΔsaeR/S (including purK, purC, purQ, purL, purF, purM, purN, purH, and purD) (Table II section I). Thus, our data suggest the SaeR/S regulatory system of MW2 has pleiotropic regulatory affects and alters expression of genes with diverse functions.
To confirm differences in gene expression in ΔsaeR/S and wild-type MW2 strains was due to the isogenic mutation in saeR/S we used TaqMan real-time RT-PCR to verify changes of selected transcripts (figure 5A and B). Using growth conditions identical to those used for the microarray experiments we measured transcript levels in seven genes by TaqMan analysis in strains MW2, ΔsaeR/S, and compΔsaeR/S. Complementation of the ΔsaeR/S mutant strain restored gene expression of saeR/S comparable to transcript levels measured in wild-type MW2 (figure 5A). TaqMan analysis also confirmed changes in gene expression identified by microarray analysis (figure 5B and Table II sections I and III).
In this study we created an isogenic saeR/S mutant in strain MW2 and investigated the role of this gene regulatory system in staphylococcal pathogenesis. We found that absence of saeR/S rendered MW2 essentially non-virulent in a mouse model of staphylococcal sepsis, but had virtually no effect on abscess formation (figure 2). Moreover, the mutant strain was significantly attenuated in its ability to survive after PMN phagocytosis (figure 4B), and microarray analysis demonstrated that deletion of saeR/S caused down-regulation of transcripts encoding several virulence factors in MW2 that were differentially-expressed after PMN phagocytosis (Table I) (7;23).
The observed dichotomous phenotype of ΔsaeR/S during two very different types of infection is intriguing and suggests two-component gene regulatory systems influence the type of staphylococcal infection based on the initial site of infection. This idea is supported by Wright et al., who concluded that agr activation and expression is essential for staphylococcal lesion development (30). The authors hypothesize that rapid agr activity precedes the innate immune response, allowing S. aureus to establish a critical quorum capable of producing large amounts of toxins to neutralize the bactericidal activity of the recruited PMNs. In our murine sepsis model, S. aureus encounter PMNs immediately, and therefore, quorum sensing activation of the staphylococcal virulon is unlikely under these conditions. Instead, we hypothesize that during bloodstream infection the staphylococcal virulon is at least in-part regulated by SaeR/S. Sepsis is the result of a complex cascade of events resulting in multi-organ failure that includes abnormal cytokine activation, neutropenia, and coagulation dysfunction (31). Inasmuch as cytolytic toxins have been shown to promote release of inflammatory mediators, and SaeR/S regulates release of cytolytic toxins, it is possible that these molecules are responsible at least in part for the observed mortality in the mouse sepsis model (figure 2). Alternatively, deletion of saeR/S resulted in the down-regulation of several genes that function or have putative functions that modulate complement including sbi, MW1040, MW1041, and MW1037. Sbi is a secreted protein that can inhibit all three complement pathways through consumption of C3 (32). MW1040 has homology to extracellular fibrinogen binding protein (Efb), and MW1037 has homology to extracellular complement binding protein (Ecb) (33). Efb and Ecb are complement evasion proteins that target C3b-containing convertases (33) and MW1041 has homology to chain A of the staphylococcal complement inhibitor (SCIN) (34) . These findings are important because complement promotes PMN recruitment, phagocytosis, and activation. In our abscess model, PMN recruitment did not appear to be affected by saeR/S (Figure 2C–D). However, the influence of saeR/S on PMN recruitment/activation following bloodstream infection remains to be determined.
Our microarray data supported observations made by others as to genes influenced by saeR/S, but also identified many additional genes under the influence of saeR/S in MW2 (Table II section I – IV). Differences in results are likely due to strains studied, growth phases investigated, type of microarray technology employed, and genetic constructs used to analyze the influence of saeR/S on S. aureus gene expression. For example, Liang et al. created a saeS mutant in strain WCUH29 and used oligonucleotide microarrays to analyze gene expression during mid-exponential (3 h) phase of growth (15). Rogasch et al. used cDNA microarrays to analyze gene expression in saeS deletion mutants constructed in S. aureus strains COL and Newman during late-exponential and stationary phases of growth (17). In comparison, our study analyzed a saeR/S deletion mutant in MW2 using oligonucleotide microarrays designed specifically with the genome of this strain. Moreover, we analyzed gene expression during mid-exponential and early-stationary phases of growth. These differences likely account for the some of the differences observed between the studies. For example, Rogasch et al. reported that an saeS deletion in strain COL did not up-regulate gene expression (17). Liang et al. showed that deletion of saeR/S in strain WCUH29 up-regulated the agr regulatory system (15). In contrast, we observed ~78 genes up-regulated in ΔsaeR/S (Table II sections II and IV), and down-regulation of the agr operon in ΔsaeR/S (agrA, agrC, agrD, agrB, and hld with fold-decreases of 2.0, 1.8, 2.0, 2.0 and 1.9, respectively). Collectively, microarray data suggest that the influence of saeR/S on transcriptional regulation is dependent on multiple factors including the strain of S. aureus studied (15;17).
The ability to rapidly recognize PMNs is likely a virulence strategy used by pathogenic strains of S. aureus. We infer from our published data and those presented herein that interaction with neutrophils activates saeR/S (7;35;36). GAS uses the Ihk-Irr two-component system to recognize PMN components including ROS and primary granule proteins (37). Salmonella typhimurium uses a similar system to recognize host cationic antimicrobial peptides, including C18G, LL-37, polymyxin B and protegrin via the PhoP/PhoQ two-component system (38;39). Li et al. described an antimicrobial peptide sensor (aps) and regulator in the Gram-positive pathogen S. epidermidis (40) that responds very specifically to cationic antimicrobial peptides and uses a mechanism distinct from the PhoP/PhoQ system (38–40). Thus, it appears that recognizing host factors via sensor/regulator systems is a strategy maintained and used by a broad range of bacterial pathogens to detect and respond to innate immunity. Palazzolo-Ballance et al. determined saeR/S was up-regulated in response to microbicides of human PMNs including hypochlorous acid, hydrogen peroxide, and azurophilic granules (35), and Geiger et al. identified promoter activities of saeS in response to sub-inhibitory concentrations of alpha-defensins (36).
Collectively, saeR/S appears to play an essential role during S. aureus evasion of innate immunity. Our study provides a foundation from which to pursue specific molecular mechanisms used by S. aureus at the host-pathogen interface and associates a two-component system to invasive S. aureus disease.
Financial support: This work was supported by NIH-PAR98-072 and NIH-NRRI grant P20RR020185, (J. M. V., M.D., T. K.N., J.J., S.G. and C. I.), P20RR16455-07 (T. K. N.) and the Intramural Program of the National Institutes of Allergy and Infectious Diseases, NIH (C.V., S.K., K.B., A.W., D.E.S., M. O. and F.R.D.).
Potential Conflicts of Interest: The authors have declared that there are no competing financial interests.
Presented in part: The Montana IDeA Network of Biomedical Research Excellence (INBRE) and Montana State University Center for Immunotherapies to Zoonotic Diseases (COBRE) Annual Fall Conference, September 27 -29, 2007, Big Sky, MT; and at the National IDeA Symposium of Biomedical Research Excellence, August 6–8, 2008, Washington, DC.