Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Evaluation of Approaches to Monitor Staphylococcus aureus Virulence Factor Expression during Human Disease

Evaluation of Approaches to Monitor Staphylococcus aureus Virulence Factor Expression during Human Disease

  • Wouter Rozemeijer, 
  • Pamela Fink, 
  • Eduardo Rojas, 
  • C. Hal Jones, 
  • Danka Pavliakova, 
  • Peter Giardina, 
  • Ellen Murphy, 
  • Paul Liberator, 
  • Qin Jiang, 
  • Douglas Girgenti
PLOS
x

Abstract

Staphylococcus aureus is a versatile pathogen of medical significance, using multiple virulence factors to cause disease. A prophylactic S. aureus 4-antigen (SA4Ag) vaccine comprising capsular polysaccharide (types 5 and 8) conjugates, clumping factor A (ClfA) and manganese transporter C (MntC) is under development. This study was designed to characterize S. aureus isolates recovered from infected patients and also to investigate approaches for examining expression of S. aureus vaccine candidates and the host response during human infection. Confirmation of antigen expression in different disease states is important to support the inclusion of these antigens in a prophylactic vaccine. Hospitalized patients with diagnosed S. aureus wound (27) or bloodstream (24) infections were enrolled. Invasive and nasal carriage S. aureus isolates were recovered and characterized for genotypic diversity. S. aureus antigen expression was evaluated directly by real-time, quantitative, reverse-transcriptase PCR (qRT-PCR) analysis and indirectly by serology using a competitive Luminex immunoassay. Study isolates were genotypically diverse and all had the genes encoding the antigens present in the SA4Ag vaccine. S. aureus nasal carriage was detected in 55% of patients, and in those subjects 64% of the carriage isolates matched the invasive strain. In swab samples with detectable S. aureus triosephosphate isomerase housekeeping gene expression, RNA transcripts encoding the S. aureus virulence factors ClfA, MntC, and capsule polysaccharide were detected by qRT-PCR. Antigen expression was indirectly confirmed by increases in antibody titer during the course of infection from acute to convalescent phase. Demonstration of bacterial transcript expression together with immunological response to the SA4Ag antigens in a clinically relevant patient population provides support for inclusion of these antigens in a prophylactic vaccine.

Introduction

The Gram positive bacterium S. aureus is both a human commensal and opportunistic pathogen. The disease syndromes caused by S. aureus are diverse, ranging from relatively mild skin infections to more severe and invasive infections including endocarditis, necrotizing fasciitis, osteomyelitis, and pneumonia. While S. aureus is the primary cause of healthcare-associated infection, with nearly 1% of all US hospital inpatient admissions affected by S. aureus disease, community-associated S. aureus disease has become prominent both in the US and globally [1]. Clinical treatment options have been impacted by the emergence of drug-resistant organisms and the absence of chemotherapeutic agents with novel mechanism(s) of action. While there are no licensed S. aureus vaccines, there is a clear medical need for a vaccine to prevent the widespread disease manifestations of S. aureus and reduce the substantial burden that the pathogen imposes for healthcare systems.

S. aureus uses several virulence strategies to cause diverse pathologies. An efficacious prophylactic vaccine must generate immune responses that functionally challenge several different virulence factors. A prophylactic S. aureus 4-antigen (SA4Ag) vaccine comprising two capsular polysaccharide conjugates (CP5-CRM197 and CP8-CRM197), together with recombinant protein antigens clumping factor A (ClfA) and manganese transporter C (MntC) is currently in development [2]. Capsular polysaccharides (CP) prevent opsonophagocytosis of bacteria, a well-established immune evasion strategy; however, antibodies to CP facilitate phagocytosis by host immune cells and lead to functional antibacterial activity [3]. ClfA is a highly conserved (>88% sequence identity) cell surface adhesin protein detected in >99% of S. aureus clinical isolates [4]. ClfA-mediated binding of S. aureus to the C-terminus of the host plasma fibrinogen gamma chain promotes fibrin cross-linking, pathogen binding to platelets and thrombus formation [5]. The fourth component of the SA4Ag vaccine, MntC, is a highly conserved (>98% sequence identity) manganese binding surface lipoprotein [2]. Manganese is an essential cofactor of several S. aureus enzymes, including superoxide dismutase (SOD). Manganese-dependent bacterial SODs play a key role in immune evasion by detoxifying superoxide metabolites generated by engulfing neutrophils [6].

Understanding the expression of these candidate S. aureus vaccine antigens during clinical infection is essential. We therefore conducted a prospective observational study of S. aureus antigen expression and antibody response to these antigens in early stage bacteremia and wound infections in hospitalized patients.

Results

Clinical study population

The overall patient flow is illustrated in Fig. 1. Fifty-one patients were enrolled in the study including 27 with a S. aureus wound infection and 24 with culture-proven S. aureus bacteremia. Twenty-five patients completed all study procedures; 16 patients completed only two blood sample time points and ten patients completed only the initial blood sample time point, either because of discharge from the hospital (n = 24) or death (n = 2). Antibiotic therapy was initiated prior to study inclusion in most patients (82%). Clinical characteristics of the study population are summarized in Table 1. S. aureus nasal carriage was detected in both bacteremic (n = 15) and wound infection (n = 13) patients.

thumbnail
Fig 1. Study design.

Although not indicated, additional blood or wound swab samples were recovered from some of the patients over the course of the study.

https://doi.org/10.1371/journal.pone.0116945.g001

Characterization of S. aureus isolates associated with disease

Invasive S. aureus isolates were recovered from each patient at the time of enrollment, and serial isolates were obtained from seven wound infection patients and from three bacteremic patients. In addition to the primary invasive isolates, 24 isolates were collected from other sources (e.g., localized infection sites, catheters, pacemaker cords, urine). All isolates were sensitive to methicillin. Approximately half of the patients had infections subsequent to surgical procedures (23/51; seven bacteremia, 16 wound) and these wound infections all occurred at the surgical site. The remainder of patients (28) had infections that were non-surgical in origin and were mostly bacteremic. Descriptive detail for each strain is presented in S1 Table.

Spa typing, which categorizes bacterial relatedness based on the DNA sequence of the protein A gene (spa), was conducted to assess the diversity of disease-causing strains in this patient group. Forty different spa types were identified among the invasive isolates, illustrating a high degree of heterogeneity (Fig. 2). If patients had multiple isolates from different disease-associated specimens, the isolates had identical spa type 86% of the time (6/7) in the wound cohort, and 82% of the time (14/17) in the bacteremic cohort. In four patients the spa type of the primary isolate was distinct from the secondary isolate. In two instances the spa types of the isolates were slightly different while the multi-locus sequence type (MLST) profiles were the same, indicating that the isolates may be related. For the other two cases, both the MLST and spa genotypes were different, indicating that these two strains were not related.

thumbnail
Fig 2. Distribution of isolate spa types in clonal complexes (CC).

Major CCs are shown (CC45, 15, 30, 7, and 5) together with spa types of isolates associated with the respective CCs. Additional spa types for the remainder of the isolates are indicated as “Other”. PI (W), primary isolate from wound; PI (B), primary isolate from blood; NI (W), nasal isolate from wound patient; NI (B), nasal isolate from bacteremia patient.

https://doi.org/10.1371/journal.pone.0116945.g002

MLST analysis revealed that the S. aureus disease isolates belonged to the 5 major global clonal complexes (CC) common to The Netherlands (Fig. 2). Wound and bacteremia isolates were distributed evenly within the CCs, with the exception of CC7 (where 4/5 isolates were associated with bacteremia) and CC15 primary disease isolates (where 5/7 isolates were associated with wound infections).

The capsule polysaccharide (CP) genotype was determined for each isolate by PCR detection of either the cap5J or cap8J gene. As outlined in Fig. 3A, CP8 was the prevalent capsule genotype in each of the disease-related isolate sub-populations. All disease isolates contained clfA and mntC. Anderson et al [2] previously demonstrated that MntC is highly conserved (maximum sequence ID 99%), whereas ClfA protein variants are less well conserved [3]. ClfA variant sequences identified in this study were within the maximum sequence diversity known for ClfA (Fig. 3B). Of the over 20 ClfA variants detected, ten were associated with primary disease-causing strains from multiple patients (n = 39). Most of these isolates had different spa sequences indicating that the strains were not closely related. Comparison of ClfA variants detected in bacteremic versus wound patients revealed that 11 variants were unique to wound patients, six to bacteremic patients and eight were detected in both cohorts. For isolates other than the primary disease isolate, 14 ClfA variants were observed and in two cases these did not match the variant carried by the primary disease isolate. When comparing ClfA variant and capsule type, it was noted that different isolates carrying the same protein variant were always of the same capsule type.

thumbnail
Fig 3. Characterization of isolates by capsule type and clfA allele.

Panel A. CP5 (open bar) and CP8 (solid bar) distribution in wound, blood, and nasal isolates. Panel B. clfA alleles carried by S. aureus isolates displayed on phylogenetic tree. Alleles occurring only in invasive isolates are indicated by red circles, those detected only in nasal isolates are indicated by blue circles, and those alleles found in both types of isolates are indicated by the blue-red circles.

https://doi.org/10.1371/journal.pone.0116945.g003

The relationship between disease and carriage isolates

Nasal carriage isolates were identified in 28 (55%) of the patients (15 bacteremic and 13 wound patients). As with the disease isolates, the nasal carriage isolates were diverse. In addition to 16 spa types shared with the invasive isolates, there were also seven spa types unique to nasal isolates (Fig. 2). Clonal complex distribution was very similar to that of the invasive isolates except that there were no CC7 isolates in the carriage subset. Again, CP8 was the predominant capsule type (64%, Fig. 3A). All isolates were positive for clfA and mntC. There were 15 ClfA variants observed in the carriage isolates, including one (ClfA_019) that was not seen in the disease isolates (Fig. 3B).

Spa-matched disease/carriage isolates were detected for 18 (64%) of the 28 carriage-positive patients and these were evenly distributed between both disease cohorts. Also, similar numbers of surgical (11/16) and non-surgical (7/12) infections had matched nasal carriage isolates. When the spa types of carriage and disease isolates did not match (10/28 or 36%), there was no discernible difference between the numbers of bacteremic (6/10) versus wound (4/10) patients with heterologous carriage strains. Carriage isolates were identified that matched the spa types of primary disease isolates from other patients, highlighting the potential of strains to spread within populations (Fig. 2).

Gene expression analysis

The RNA expression profile of S. aureus genes associated with virulence was evaluated in 25 wound swabs (23 patients, two with serial swabs) and 53 nasal swabs (51 patients, two with duplicate swabs). Despite the administration of antibiotics to patients prior to swab sampling, S. aureus 16S rRNA, a highly abundant transcript, was detected in most of these samples (94% and 87% of the wound and nasal swabs, respectively). Most (27/28) patients with culture-positive nasal swabs also demonstrated 16S rRNA positivity. Some (19/23) patients with culture-negative nasal swabs were positive for 16S rRNA transcripts, illustrating the greater sensitivity of the molecular approach. Samples were tested for clfA, mntC, capH, isdB, and for the biofilm-associated icaB gene transcripts [7]. There were eight wound and four nasal swabs with detectable virulence gene transcripts, primarily mntC and clfA; icaB was not detected from any swab. The S. aureus triosephosphate isomerase (tpi) gene was included as a control for expression from a single-copy housekeeping gene. Of the twelve swabs, 16S rRNA and tpi transcripts were both detected in only five wound and three nasal swabs. The relative copy number of the 16S rRNA, tpi and the respective virulence gene transcripts are indicated in Table 2. The wound samples had higher tpi expression levels than the nasal swabs, indicating a higher bacterial load. In both wound and nasal swabs, relative transcript copy numbers demonstrate that mntC expression levels were typically higher than those of clfA. Expression of capH and isdB was detected in two wound samples but not in the nasal swabs. In both of these wound samples (patients 100106 and 100123), cap transcripts were specific for CP8 and consistent with the respective CP genotype of the invasive isolates recovered from these patients.

thumbnail
Table 2. In vivo expression of S. aureus genes in humans with confirmed S. aureus infection.

https://doi.org/10.1371/journal.pone.0116945.t002

Serological evaluation

Patients were enrolled in this study only after confirmation of a S. aureus infection. As a consequence, there were no pre-infection sera available for comparative analysis. The serum geometric mean antibody titer (GMT) values determined using the competitive Luminex immunoassay (cLIA) are summarized in Table 3. Antibody titer specific for each of the four candidate vaccine antigens was detected in this study population. Comparisons of GMT values for the four antigens between patients with wound and blood infections are outlined in Table 3. Unlike the CP antigens, there was a trend for higher serum antibody titer for both protein antigens in bacteremic than in wound patients.

thumbnail
Table 3. Serum Geometric Mean Titer (GMT) values from Competitive Luminex Immunoassay (cLIA)a.

https://doi.org/10.1371/journal.pone.0116945.t003

Most of the patients who were enrolled were infected with CP8 isolates (Fig. 3A). A comparison of serum antibody titer to the four antigens between patients with CP5 and CP8 infections is also captured in Table 3. In patients with CP8 invasive isolates, the CP8 GMT was greater than that of CP5, although the difference was not significant (p = 0.215). Antibody titers to CP5 were higher than to CP8 in patients with CP5 infections, which reached significance (p = 0.043). Serum titers to the SA4Ag vaccine antigens were monitored from the acute to the convalescent phase of the infection. The data in Fig. 4A illustrate that the antibody titers tended to rise from the acute phase of infection (day 1–2) to the convalescent phase (day 14). In addition, the percentage of patients with serum titers above the assay limit of quantitation for CP8, ClfA, and MntC increased over that 2-week interval (Fig. 4B).

thumbnail
Fig 4. Antibody titers to the SA4Ag vaccine antigens.

Panel A. Antibody titers to the four antigens at study baseline (day 1–2) to Day 7–14. Scatter plots of titers are shown for the 25 patients who had blood samples drawn at three different times over the two-week course of the study. If the titer was reported as “below limit of quantitation”, a value that is ½ LLOQ was used in the plots; for ClfA, MntC, CP5 and CP8, the values are 24.45, 21.35, 23.5 and 18.3 U/mL, respectively. Points above the diagonal line indicate a higher titer in the later samples. Panel B. Percentage of patients with titers above the limit of quantitation over time. Patient population is as described for Panel A (above).

https://doi.org/10.1371/journal.pone.0116945.g004

Discussion

In the current study a cohort of 51 patients ≥ 18 years of age with diagnosed S. aureus wound or bloodstream infections were enrolled at two hospitals in The Netherlands. Invasive and carriage isolates from these patients were genotypically characterized. Nasal carriage of S. aureus is considered a risk factor for infection in certain patient populations [8] [9] [10,11] [12]. In this study, a nasal carriage isolate was recovered in 55% of the patients. In those patients where a carriage isolate was obtained, the invasive isolate was genotypically indistinguishable in 69% (11/16) of surgical patients and 58% (7/12) of non-surgical patients. Overall, nasal carriage isolates were linked to disease strains in approximately 35% of patients. All isolates, regardless of origin, were sensitive to methicillin. This corresponds well with the observed low incidence of methicillin resistant S. aureus (MRSA) isolates in The Netherlands and points to the effectiveness of the “search and destroy” strategy to control MRSA [13]. Although the isolates in this study were methicillin sensitive, two bacteremic patients (100108 and 100126) died from sepsis, highlighting the fact that MSSA strains continue to be significant pathogens.

Despite the size of the study population and the geographic proximity of the two study sites, there was considerable genotypic heterogeneity in the S. aureus disease isolates. Comparative analysis of the nasal carriage (PFESA317) and primary disease isolate (PFESA304) from patient 100125 illustrates the discriminatory power of spa typing. The isolates have identical capsule, clfA and MLST genotypes, but their spa profiles differ from each other by a small insertion and sequence replacement in two regions of the locus (S1 Table). Examples of isolates with spa identity but distinct clfA and MLST alleles were also identified. Of the nine spa 21 type isolates characterized, four had clfA 015–1 and ST15 genotypes while five had clfA047–1 and ST582 genotypes (S1 Table). Bacterial genotyping using whole genome sequence data will provide the best discrimination between closely related isolates [14].

In other studies of invasive S. aureus disease in Europe, the major CCs observed are CC5, 8, 15, 30 and 45 [15] [16]. The major CCs were also well represented in the current study, but there were also several CC7 isolates (n = 14). Other studies of S. aureus bacteremic infections in The Netherlands have recovered CC7 strains, so the current findings correlate well with previous results. However, in contrast to the current study, others have reported that CC7 isolates are also recovered from nasal carriage [17]. Except for a small number of CC6, CC34, CC72, and CC121 isolates, each CC in this patient population is represented in the larger collection of S. aureus isolates reported earlier [4].

The second objective of this study was to determine whether SA4Ag vaccine antigens are expressed during infection or carriage. Two approaches were used to detect antigen expression. The first was qRT-PCR and the second was a serological assay. For a vaccine strategy to be effective, the vaccine must contain antigens expressed by S. aureus during the course of natural infection. Many studies use PCR to detect the presence of specific virulence genes in bacteria isolated from infections as a measure of their pathogenic potential [16] [18]. In recent studies with human subjects, S. aureus RNA transcript levels were quantitated using qRT-PCR in persistent, highly-colonized nasal carriers [19]. Similarly S. aureus transcript levels were monitored in subjects with cutaneous abscesses using microarray hybridization and qRT-PCR [20]. In the current study, gene expression of virulence factors was measured in different types of human infections and in nasal carriage. This is the first study to assess antigen expression by diverse S. aureus strains during infection. The nature of the study presented substantial technical challenges. For example, wound and nasal swabs were collected only following confirmation of a S. aureus infection in patients. Even prior to confirmation, the standard of care is to administer antibiotics. Consequently, patients in the study may have been on antibiotic therapy for up to 48 hours before the wound/nasal swabs were obtained. This could negatively impact recovery of bacteria on swabs for antigen expression analysis, and low numbers of bacteria in tissue samples has been identified as a limitation in this type of experiment [20]. The frequency of tpi negative samples points to a less than optimal recovery of bacteria and/or bacterial RNA from many of the swabs. Despite the experimental challenges, RNA transcripts corresponding to capsule, clfA and mntC genes were observed, with consistent detection of clfA and mntC transcripts where expression of tpi was also detected. Burian et al [19] also reported that clfA transcripts were seen in nasal samples from colonized subjects. In our study, mntC was always expressed at higher levels than other genes in nose and wound samples. Date et al. [20] also observed that mntC expression is slightly up-regulated in the cutaneous abscess study. Previous studies [21] [22] demonstrated a temporal expression of S. aureus virulence factors in preclinical in vivo studies. Given that gene expression was examined at only one time point in the current study, the lack of consistent expression of other virulence factor genes is not surprising. Also, transcript stability may differ between these genes, especially in the polycistronic capsule operon, making it more difficult to monitor expression with this approach.

An alternative approach to monitor antigen expression indirectly is to measure the presence of bacterial antigen-specific antibodies in serum [23] [24], [25], [26] [27]. The lack of serum samples from these patients prior to infection for comparative analysis is one limitation of the current study. However, serology results demonstrated that most of the patients responded to ClfA and MntC, and the percentage of responders increased over time. Antibodies are generated as a result of immune exposure to the organism and the detection of specific antibodies indicates that the subject has encountered the S. aureus antigen. The quality of native immune responses was illustrated by Hawkins et al [28] who mimicked the natural situation by immunizing non-human primates with killed S. aureus cells. These animals generated antibodies that bound to ClfA but did not inhibit the binding of ClfA to fibrinogen, its natural ligand. Importantly, the cLIA titers to ClfA observed following natural infection in the current study are several orders of magnitude less than titers observed in subjects receiving the investigational SA4Ag vaccine [29].

Patients in the current study with documented S. aureus infections also had measurable and specific antibody titer to CP5 and CP8. Specific antibody titers to the respective CP antigens are greater in patients infected with S. aureus isolates expressing the same CP type. Despite the challenges described earlier, expression of capH was detected in two wound swabs and in both cases cap transcripts were specific for CP8, consistent with the capsule genotype identified in the invasive isolate.

S. aureus employs a number of strategies to enable both colonization of the host and survival in vivo. The bacterium uses relevant virulence factors, including capsule polysaccharide, binding proteins and toxins, to adapt to a variety of host niches and cause a multitude of diverse infections. The broad spectrum of pathogenic mechanisms renders S. aureus a challenging vaccine target. An essential characteristic for a S. aureus vaccine antigen is that immunization must elicit a functional response capable of killing the diverse collection of disease-causing isolates. Thus, the vaccine antigen must be (i) present in the genome of geographically dispersed invasive isolates, (ii) highly conserved, (iii) expressed by the pathogen in host micro-environments, and (iv) accessible to antibodies. In this study we demonstrated that the four antigens present in the investigational SA4Ag vaccine were distributed across disease and carriage isolates. We obtained evidence, both direct (from wound) and indirect (from sera), that these antigens are expressed during disease, providing assurance that they will be accessible to antibodies generated by the SA4Ag vaccine in patients who are at risk of invasive S. aureus disease.

Materials and Methods

Patients and study design

Between March 2008 and March 2009 patients were recruited from two hospitals, a tertiary university hospital (VU University Medical Center in Amsterdam, The Netherlands) and a large teaching hospital (Amphia Hospital in Breda, The Netherlands). The medical ethical committee of each study site approved the study. Adult (≥18 years) hospitalized patients with culture proven S. aureus wound infections or bacteremia, were assessed for eligibility. Definitive microbiological diagnosis was required within 48 hours after cultures were obtained. Patients admitted to the intensive care unit, those with hematological diseases or being treated with immunosuppressive drugs were excluded. Patients were included after written informed consent had been given.

Sample collection

At inclusion into the study, blood was drawn and the anterior nares were swabbed with duplicate swabs. One swab was cultured on blood agar by standard microbiological methods and the second was frozen at -80°C for real-time, quantitative, reverse-transcriptase PCR (qRT-PCR) analysis. Culture isolates obtained from blood and/or wound sites were handled similarly, as were isolates obtained from any other accessible infection site(s) when present. S. aureus identification was confirmed by coagulase and/or PCR testing; methicillin susceptibility was assessed by standard antibiotic susceptibility testing. Blood samples were also drawn at one and two weeks after the initial assessment to evaluate immunological changes over time.

Molecular analysis of S. aureus isolates

S. aureus isolates collected at time of diagnosis (primary disease isolates), subsequently isolated disease isolates and nasal isolates were characterized by MLST (according to protocols at saureus.mlst.net/[30]), spa typing [31] [32], capsule serotype-specific PCR [4], and clfA and mntC DNA sequencing [2][4]. Genomic DNA was isolated from S. aureus cultures grown in tryptic soy broth (TSB, Difco) using the Qiagen DNeasy blood and tissue kit according to manufacturer’s instructions, except for the inclusion of lysostaphin (25–30U per 200 μL) in the first step. Spa typing was performed using the spa DNA sequence information and the eGenomics software (http://epigene.egenomics.com). ClfA protein variants were assigned as previously reported [4]. For bioinformatics analysis, multiple alignments were created with MAFFT [33]. Phylogenetic trees were constructed with MEGA 4[34].

Gene expression analysis

Wound and nasal carriage swabs were examined for gene expression by qRT-PCR. Swabs were considered of sufficient quality for qRT-PCR if they were positive using 16S rRNA and tpi primer/probe sets. In vivo antigen expression was monitored for clfA, capH, mntC, iron surface binding protein B (isdB), and intercellular adhesion protein B (icaB). RNA was extracted from bacteria on the swabs using the RiboPure kit (Ambion) and converted to cDNA using the Quantitect kit (Qiagen) as recommended by the manufacturers. qRT-PCR reactions were conducted on the 7500 Real Time PCR System (Applied Biosystems). The analysis performed by the 7500 System SDS Software v1.3.1 generated the threshold cycle (Ct) for each sample. The relative transcript copy number was determined from corresponding standard curves generated using 102 to 107 copies of S. aureus genomic DNA.

Competitive Luminex Immunoassay (cLIA)

A qualified multiplex cLIA was developed to indirectly measure the level of functional serum antibody titer to each of the four SA4Ag vaccine antigens. Each antigen was coupled to a distinct fluorescent Luminex microsphere. Antigen-coated microspheres were incubated overnight at 4°C with appropriately diluted serum samples, controls, or reference standard serum. Antigen-specific functional mouse monoclonal antibodies (mAbs) were then added to the microsphere/serum mixture and bound mAbs were detected with R-Phycoerythrin (PE) labeled rat anti-mouse IgG1 secondary antibody (Southern Biotech) using a BioPlex reader (BioRad) to measure fluorescence. As this is a competitive assay, the magnitude of the fluorescent PE signal is inversely proportional to the amount of antigen-specific antibody in the sample.

Statistical analysis

The cLIA serology data were descriptively summarized with GMT and corresponding 95% confidence intervals (CI). If a titer was below the limit of quantitation (LLOQ), ½ × lower limit of quantitation (½ LLOQ) was used to impute assay values when calculating GMT.

The CIs were constructed by back transformation of the confidence limits computed for the mean of the logarithmically transformed assay data based on Student t distribution. P-values based on 2-sample t-test or Wilcoxon test were presented to identify potential differences in immunoglobulin concentration.

Supporting Information

S1 Table. Characterization of S. aureus Isolates.

https://doi.org/10.1371/journal.pone.0116945.s001

(PDF)

S2 Table. qRT-PCR Results for Nasal and Wound Samples.

https://doi.org/10.1371/journal.pone.0116945.s002

(PDF)

Acknowledgments

The authors would like to thank Dr Emilio Emini for his thoughtful comments regarding this research. In addition, thanks to Lubomira Andrew and Lorna Nuñez for molecular genetics support, Li Hao for DNA sequence analysis, and the High Throughput Clinical Testing laboratory for performing the Luminex serology assays. All are employees of (Pfizer Vaccine Research and Development.

Author Contributions

Conceived and designed the experiments: WR PF ER CHJ EM QJ DG RP PS KJ ASA JK. Performed the experiments: DP PG ER. Analyzed the data: PL WR PF ER CHJ EM QJ DG RP PS KJ ASA JK DP PG. Wrote the paper: PL WR PF ER CHJ EM QJ DG RP PS KJ ASA JK DP PG.

References

  1. 1. Nathwani D (2009) Health economic issues in the treatment of drug-resistant serious Gram-positive infections. J Infect 59 Suppl 1: S40–50. pmid:19766889
  2. 2. Anderson AS, Scully IL, Timofeyeva Y, Murphy E, McNeil LK, et al. (2012) The Staphylococcus aureus transporter MntC is a highly conserved cell surface protein that elicits protective immunity against both S. aureus and Staphylococcus epidermidis. Journal of Infectious Diseases 205: 1688–1696. pmid:22474033
  3. 3. Nanra JS, Buitrago SM, Crawford S, Ng J, Fink PS, et al. (2013) Capsular polysaccharides are an important immune evasion mechanism for Staphylococcus aureus. Hum Vaccin Immunother 9: 480–487. pmid:23249887
  4. 4. Murphy E, Lin SL, Nunez L, Andrew L, Fink PS, et al. (2011) Challenges for the evaluation of Staphylococcus aureus protein based vaccines: monitoring antigenic diversity. Hum Vaccin 7 Suppl: 51–59. pmid:21245656
  5. 5. Siboo IR, Cheung AL, Bayer AS, Sullam PM (2001) Clumping factor A mediates binding of Staphylococcus aureus to human platelets. Infect Immun 69: 3120–3127. pmid:11292731
  6. 6. Handke LD, Hawkins JC, Miller AA, Jansen KU, Anderson AS (2013) Regulation of Staphylococcus aureus MntC Expression and Its Role in Response to Oxidative Stress. PLoS One 8: e77874. pmid:24205007
  7. 7. O’Gara JP (2007) ica and beyond: biofilm mechanisms and regulation in Staphylococcus epidermidis and Staphylococcus aureus. FEMS Microbiol Lett 270: 179–188. pmid:17419768
  8. 8. von Eiff C, Becker K, Machka K, Stammer H, Peters G (2001) Nasal Carriage as a Source of Staphylococcus aureus Bacteremia. New England Journal of Medicine 344: 11–16. pmid:11136954
  9. 9. Wertheim HFL, Melles DC, Vos MC, van Leeuwen W, van Belkum A, et al. (2005) The role of nasal carriage in Staphylococcus aureus infections. The Lancet Infectious Diseases 5: 751–762. pmid:16310147
  10. 10. Kluytmans JA, Mouton JW, Ijzerman EP, Vandenbroucke-Grauls CM, Maat AW, et al. (1995) Nasal carriage of Staphylococcus aureus as a major risk factor for wound infections after cardiac surgery. J Infect Dis 171: 216–219. pmid:7798667
  11. 11. Kluytmans J, A van Belkum A, Verbrugh H (1997) Nasal carriage of Staphylococcus aureus: epidemiology, underlying mechanisms, and associated risks. Clin Microbiol Rev 3: 505–520. pmid:9227864
  12. 12. Bode LG, Kluytmans JA, Wertheim HF, Bogaers D, Vandenbroucke-Grauls CM, et al. (2010) Preventing surgical-site infections in nasal carriers of Staphylococcus aureus. N Engl J Med 362: 9–17. pmid:20054045
  13. 13. van Cleef BA, Kluytmans JA, van Benthem BH, Haenen A, Monen J, et al. (2012) Cross border comparison of MRSA bacteraemia between The Netherlands and North Rhine-Westphalia (Germany): a cross-sectional study. PLoS One 7: e42787. pmid:22880109
  14. 14. Harris SR, Cartwright EJ, Torok ME, Holden MT, Brown NM, et al. (2013) Whole-genome sequencing for analysis of an outbreak of meticillin-resistant Staphylococcus aureus: a descriptive study. Lancet Infect Dis 13: 130–136. pmid:23158674
  15. 15. Blomfeldt A, Aamot HV, Eskesen AN, Muller F, Monecke S (2013) Molecular characterization of methicillin-sensitive Staphylococcus aureus isolates from bacteremic patients in a Norwegian University Hospital. J Clin Microbiol 51: 345–347. pmid:23135934
  16. 16. Rasmussen G, Monecke S, Ehricht R, Soderquist B (2013) Prevalence of Clonal Complexes and Virulence Genes among Commensal and Invasive Staphylococcus aureus Isolates in Sweden. PLoS One 8: e77477. pmid:24130888
  17. 17. Nulens E, Stobberingh EE, van Dessel H, Sebastian S, van Tiel FH, et al. (2008) Molecular characterization of Staphylococcus aureus bloodstream isolates collected in a Dutch University Hospital between 1999 and 2006. J Clin Microbiol 46: 2438–2441. pmid:18463215
  18. 18. Verkaik NJ, Boelens HA, de Vogel CP, Tavakol M, Bode LG, et al. (2010) Heterogeneity of the humoral immune response following Staphylococcus aureus bacteremia. Eur J Clin Microbiol Infect Dis 29: 509–518. pmid:20186449
  19. 19. Burian M, Wolz C, Goerke C. (2010) Regulatory adaptation of Staphylococcus aureus during nasal colonization of humans. PLoS One 5: e10040. pmid:20386721
  20. 20. Date SV, Modrusan Z, Lawrence M, Morisaki JH, Toy K et al. (2014) Global gene expression of methicillin-resistant Staphylococcus aureusUSA300 during human and mouse infection. J Infect Dis 209: 1542–1550. pmid:24286981
  21. 21. Nanra JS, Timofeyeva Y, Buitrago SM, Sellman BR, Dilts DA, et al. (2009) Heterogeneous in vivo expression of clumping factor A and capsular polysaccharide by Staphylococcus aureus: implications for vaccine design. Vaccine 27: 3276–3280. pmid:19200819
  22. 22. Kuklin NA, Clark DJ, Secore S, Cook J, Cope LD, et al. (2006) A novel Staphylococcus aureus vaccine: iron surface determinant B induces rapid antibody responses in rhesus macaques and specific increased survival in a murine S. aureus sepsis model. Infect Immun 74: 2215–2223. pmid:16552052
  23. 23. Verkaik NJ, Dauwalder O, Antri K, Boubekri I, de Vogel CP, et al. (2010) Immunogenicity of toxins during Staphylococcus aureus infection. Clin Infect Dis 50: 61–68. pmid:19947854
  24. 24. Verkaik NJ, Lebon A, de Vogel CP, Hooijkaas H, Verbrugh HA, et al. (2010) Induction of antibodies by Staphylococcus aureus nasal colonization in young children. Clin Microbiol Infect 16: 1312–1317. pmid:19832714
  25. 25. Kolata J, Bode LG, Holtfreter S, Steil L, Kusch H, et al. (2011) Distinctive patterns in the human antibody response to Staphylococcus aureus bacteremia in carriers and non-carriers. Proteomics 11: 3914–3927. pmid:21805632
  26. 26. den Reijer PM, Lemmens-den Toom N, Kant S, Snijders SV, Boelens H, et al. (2013) Characterization of the Humoral Immune Response during Staphylococcus aureus Bacteremia and Global Gene Expression by Staphylococcus aureus in Human Blood. PLoS One 8: e53391. pmid:23308212
  27. 27. Zorman JK, Esser M, Raedler M, Kreiswirth BN, Ala’aldeen DA, et al. (2013) Naturally occurring IgG antibody levels to the Staphylococcus aureus protein IsdB in humans. Hum Vaccin Immunother 9: 1857–1864. pmid:23778314
  28. 28. Hawkins J, Kodali S, Matsuka YV, McNeil LK, Mininni T, et al. (2012) A recombinant clumping factor A-containing vaccine induces functional antibodies to Staphylococcus aureus that are not observed after natural exposure. Clin Vaccine Immunol 19: 1641–1650. pmid:22896688
  29. 29. Richmond P, Nissen M, Marshall H, Shakib S, Hodsman P, et al. (2011) A randomised, placebo controlled Phase 1 first-in-human study of a novel 3-antigen Staphylococcus aureus vaccine in healthy adults. ECCMID. Milan, Italy.
  30. 30. Enright MC, Day NP, Davies CE, Peacock SJ, Spratt BG (2000) Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J Clin Microbiol 38: 1008–1015. pmid:10698988
  31. 31. Koreen L, Ramaswamy SV, Graviss EA, Naidich S, Musser JM, et al. (2004) spa typing method for discriminating among Staphylococcus aureus isolates: implications for use of a single marker to detect genetic micro- and macrovariation. J Clin Microbiol 42: 792–799. pmid:14766855
  32. 32. Mathema B, Mediavilla J, Kreiswirth BN (2008) Sequence analysis of the variable number tandem repeat in Staphylococcus aureus protein A gene: spa typing. Methods Mol Biol 431: 285–305. pmid:18287764
  33. 33. Katoh K, Toh H (2008) Recent developments in the MAFFT multiple sequence alignment program. Brief Bioinform 9: 286–298. pmid:18372315
  34. 34. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24: 1596–1599. pmid:17488738