Infectious Dose Dictates the Host Response during Staphylococcus aureus Orthopedic-Implant Biofilm Infection

Abstract

Staphylococcus aureus is a leading cause of prosthetic joint infections (PJIs) that are typified by biofilm formation. Given the diversity of S. aureus strains and their propensity to cause community- or hospital-acquired infections, we investigated whether the immune response and biofilm growth during PJI were conserved among distinct S. aureus clinical isolates. Three S. aureus strains representing USA200 (UAMS-1), USA300 (LAC), and USA400 (MW2) lineages were equally effective at biofilm formation in a mouse model of PJI and elicited similar leukocyte infiltrates and cytokine/chemokine profiles. Another factor that may influence the course of PJI is infectious dose. In particular, higher bacterial inocula could accelerate biofilm formation and alter the immune response, making it difficult to discern underlying pathophysiological mechanisms. To address this issue, we compared the effects of two bacterial doses (10³ or 10⁵ CFU) on inflammatory responses in interleukin-12p40 (IL-12p40) knockout mice that were previously shown to have reduced myeloid-derived suppressor cell recruitment concomitant with bacterial clearance after low-dose challenge (10³ CFU). Increasing the infectious dose of LAC to 10⁵ CFU negated these differences in IL-12p40 knockout animals, demonstrating the importance of bacterial inoculum on infection outcome. Collectively, these observations highlight the importance of considering infectious dose when assessing immune responsiveness, whereas biofilm formation during PJI is conserved among clinical isolates commonly used in mouse S. aureus infection models.

INTRODUCTION

Staphylococcal species (Staphylococcus aureus and coagulase-negative staphylococcus) are a primary cause of prosthetic joint infections (PJIs) that are typified by biofilm formation (1–3). Many PJIs are thought to originate from bacterial colonization at the time of surgery, although clinical signs of infection may not appear until months to years following joint replacement (4, 5). The often protracted period between prosthetic joint replacement and infectious complications suggests that prostheses may also be colonized following systemic bacterial infection. Since S. aureus biofilms are adept at circumventing immune-mediated clearance (6–9) and are recalcitrant to antibiotic therapy, S. aureus PJIs have become increasingly difficult to treat (10–13). Understanding the series of inflammatory events that occur during S. aureus PJI is a critical step toward identifying potential therapeutic targets.

An important factor to consider when investigating mechanisms pertinent to human infection is whether animal models accurately reflect clinical features of human disease. Given its frequency in clinical infections (1–3), S. aureus is a logical choice of pathogen to demonstrate clinical relevance with regard to PJIs. However, it has been well documented that S. aureus strains can behave differently in terms of clinical presentation and pathogenicity (1, 14–18). The fact that ca. 30% of individuals are carriers for S. aureus and the recent emergence of community-acquired, methicillin-resistant S. aureus (CA-MRSA) clones as causative pathogens in hospital-associated infections (19, 20) highlight the importance of examining the immune response to different S. aureus strains and doses in the context of PJI. Therefore, we compared the immune responses to three distinct S. aureus clinical isolates from various epidemiological backgrounds during PJI. S. aureus LAC is a CA-MRSA USA300 strain that was isolated from a patient with a skin and soft tissue infection (16, 21). S. aureus MW2 is a CA-MRSA USA400 strain that was recovered from a pediatric patient with fatal septicemia and septic arthritis (19, 22, 23). S. aureus UAMS-1 is a community-acquired, methicillin-sensitive S. aureus (CA-MSSA) USA200 strain that was isolated from a patient with osteomyelitis (24). These strains were selected since they were originally responsible for human infections and are commonly used in mouse infection models and, as such, are well characterized. Since the purpose of this study was to demonstrate the influence of infectious dose on immune responses, we considered it important to focus on clinical isolates that are widely used in mouse models and that will best contribute to the current body of knowledge. Recent studies from our laboratory have focused on characterizing the immune response during S. aureus PJI using a mouse model of orthopedic-implant biofilm infection, which has provided a better understanding of the key leukocyte populations that contribute to chronic biofilm establishment. In particular, myeloid-derived suppressor cells (MDSCs) and anti-inflammatory polarized macrophages have been implicated in promoting S. aureus biofilm persistence (25–27). In this report, we determined whether biofilm formation or the resultant immune response was strain-dependent using our orthopedic-implant biofilm infection model with LAC, MW2, and UAMS-1. Differences in bacterial burdens between the various strains were negligible. Additionally, the immune response, as measured by immune cell influx and inflammatory mediator production in the soft tissue surrounding the orthopedic implant, revealed little variability between the three strains tested, an important finding considering the diverse origin of the bacterial strains examined.

The range of diseases caused by S. aureus is represented by the diversity of animal model systems currently being studied (28, 29). In terms of S. aureus biofilm infection, some models utilize a large infectious inoculum or introduce implants that are precoated with bacteria (30–36). A high-challenge dose, particularly in a confined space such as the joint/bone, would be predicted to elicit a vigorous proinflammatory cascade that could likely alter the course of the resultant immune response and bacterial survival. In contrast, human PJIs typically result from colonization with low numbers of bacteria, which may provide a survival advantage during acute infection by their inability to trigger a robust proinflammatory response. To examine the effect of infectious dose on the host immune response and biofilm formation, we compared disease progression following the administration of a standard low-dose S. aureus inoculum (10³ CFU) that we and others have used to establish chronic biofilm infection (25–27, 37–39) versus a 2-log-higher dose (10⁵ CFU). These studies utilized IL-12p40 knockout (KO) mice that were previously shown to have reduced myeloid-derived suppressor cell (MDSC) recruitment concomitant with improved bacterial clearance after low-dose challenge (10³ CFU) (26) to investigate whether a higher dose (10⁵ CFU) would negate this phenotype. Indeed, bacterial burdens and MDSC infiltrates were similar between IL-12p40 KO and wild-type (WT) mice after high-dose (10⁵ CFU) LAC challenge, whereas both readouts were significantly reduced in IL-12p40 KO animals in response to 10³ CFU of bacteria. Taken together, these observations highlight the importance of considering infectious dose when assessing immune responsiveness, whereas biofilm formation during PJI is conserved among clinical isolates commonly used in mouse S. aureus infection models.

MATERIALS AND METHODS

Mice.

For S. aureus strain comparison studies, male C57BL/6 mice (8 to 12 weeks old) were purchased from Charles River Laboratories (Frederick, MD). For S. aureus dose comparison studies, IL-12p40 KO mice (C57BL/6 background) were purchased from The Jackson Laboratory (Bar Harbor, ME) with breeding colonies maintained in an approved facility at the University of Nebraska Medical Center. Age- and sex-matched IL-12p40 KO and C57BL/6 WT mice were utilized at 8 to 12 weeks of age. These studies were conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animal use protocol was approved by the Institutional Animal Care and Use Committee of the University of Nebraska Medical Center (UNMC).

Mouse model of S. aureus prosthetic joint infection.

A mouse model of S. aureus PJI was utilized as previously described (25–27, 39, 40). Briefly, mice were anesthetized with a ketamine/xylazine cocktail (Hospira, Lake Forest, IL, and Akorn, Decatur, IL; 100 and 5 mg/kg of body weight, respectively), and the surgical site was shaved and disinfected with povidone-iodine. A medial parapatellar arthrotomy with lateral displacement of the quadriceps-patella was performed to access the distal femur. Next, a burr hole was created in the femoral intercondylar notch extending into the intramedullary canal using a 26-gauge needle, whereupon a precut 0.8-cm-length, orthopedic-grade Kirschner (K) wire (0.6 mm in diameter; Nitinol [nickel-titanium]; Custom Wire Technologies, Port Washington, WI) was inserted into the intramedullary canal, leaving ∼1 mm protruding into the joint space. A total of 10³ or 10⁵ CFU of S. aureus was inoculated at the implant tip depending on the experimental design. Analgesia (Buprenex, 0.1 mg/kg [administered subcutaneously]; Reckitt Benckiser, Hull, United Kingdom) was administered immediately after infection and again 24 h later for pain relief. After this interval, all mice exhibited normal ambulation and no discernible pain behaviors.

Bacterial strains and preparation of the bacterial inoculum.

Three bacterial strains, representing different epidemiological backgrounds, were utilized for these studies. S. aureus USA300 LAC, which was originally isolated from a patient with a skin and soft tissue infection (16, 21), was cured of two antibiotic resistance plasmids and designated USA300 LAC-13C (referred to as LAC in the present study) as previously described (41). S. aureus MW2 is a USA400 CA-MRSA clinical isolate recovered from a fatal pediatric infection and was kindly provided by Keer Sun at UNMC (22, 42). S. aureus UAMS-1, a USA200 MSSA strain, was isolated from a patient with osteomyelitis and was kindly provided by Mark Smeltzer at the University of Arkansas for Medical Sciences (Little Rock, AR) (24). All three isolates were utilized for strain comparison studies, whereas dose-response assessments were only performed with LAC and UAMS-1.

For all experiments, bacterial strains were prepared by inoculating 25 ml of brain heart infusion medium (BD Biosciences, Sparks, MD) with a single colony from a fresh streak plate made 24 h earlier. Cultures were grown at 37°C with constant aeration for 12 h, with a 1:10 volume-flask ratio, whereupon an aliquot was removed and diluted 1:10 with PBS, for optical density measurements. Absorbance readings (an optical density at 620 nm) were used to estimate culture densities, extrapolated from previous studies and growth curves. A total of 1 ml from the 12-h culture was removed and pelleted by centrifugation (20,000 × g, 5 min, 4°C), and the bacteria were washed three times with 1 ml of phosphate-buffered saline (PBS). The final bacterial pellet was resuspended in 1 ml of PBS, and infectious inocula were prepared by diluting organisms to 5 × 10⁵ or 5 × 10⁷ CFU/ml in PBS to achieve the desired infectious doses of 10³ or 10⁵ CFU/mouse. Actual bacterial inocula were calculated retrospectively by plating serial dilutions of each working stock onto blood agar plates, which were consistently found to accurately reflect the inoculum estimates (10³ or 10⁵ CFU/mouse).

Flow cytometry.

To characterize leukocyte infiltrates associated with S. aureus PJI, soft tissues surrounding the knee joint were dissociated in Hanks balanced salt solution (HBSS) containing 10% fetal bovine serum (FBS; HyClone, Logan, UT) using the flat end of a plunger from a 3-ml syringe and subsequently passed through a 70-μm-pore-size filter (BD Falcon, Bedford, MA). The resulting filtrate was washed with HBSS–10% FBS, and cells were collected by centrifugation (300 × g, 10 min), whereupon red blood cells were lysed using BD Pharm Lyse (BD Biosciences, San Diego, CA) according to the manufacturer's instructions. After lysis, the cells were washed and resuspended in PBS, followed by incubation in mouse Fc block (BD Biosciences) to minimize nonspecific antibody binding. The cells were then stained for 30 min with directly conjugated antibodies for multicolor flow cytometry analysis, which included two separate panels to identify innate immune populations (CD45-APC, Ly6G-PE, Ly6CPerCP-Cy5.5, and F4/80-PE-Cy7) or T cells (CD3ε-APC, CD4-Pacific Blue, CD8a-FITC, and TCR γδ-PE) with a viability dye to exclude dead cells (Live/Dead Fixable Blue dead cell stain kit; Life Technologies, Eugene, OR). All fluorochrome-conjugated antibodies were purchased from either BD Biosciences or eBioscience. Controls included cells stained with isotype control antibodies to assess the degree of nonspecific staining, as well as fluorescence minus one to identify gating thresholds (43). Analysis was performed using BD FACSDiva software with results presented as the percentage of CD45⁺ or CD3⁺ gated cells.

Multianalyte microbead arrays.

To compare inflammatory mediator profiles elicited by the various S. aureus strains during PJI, a custom-designed mouse microbead array was used (Milliplex; Millipore, Billerica, MA), which detects the following mediators: granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-1α (IL-1α), IL-1β, IL-6, IL-9, IL-10, IL-12p40, IL-13, IL-17, CCL2, CCL3, CCL5, CXCL2, CXCL9, CXCL10, tumor necrosis factor alpha (TNF-α), and vascular endothelial growth factor (VEGF). A Bio-Plex workstation (Bio-Rad, Hercules, CA) was used to analyze results with values normalized to the total amount of protein recovered from each sample to correct for differences in tissue sampling size.

Statistical analysis.

Significant differences between experimental groups were determined by using an unpaired two-tailed Student t test using GraphPad Prism version 4 (GraphPad Software, La Jolla, CA). For all analyses, P < 0.05 was considered statistically significant.

RESULTS

Distinct S. aureus clinical isolates elicit similar immunological responses and biofilm formation during PJI.

When examining immunological responses to S. aureus, determining the most appropriate isolate to best model clinical disease is an important consideration. To evaluate the importance of strain on infection outcome and inflammation, we compared three commonly used S. aureus isolates, namely, LAC, MW2, and UAMS-1, in a mouse model of orthopedic-implant biofilm infection that models PJI in humans. As previously reported in our laboratory with USA300 LAC (25–27), this is a well-established low dose infection model associated with rapid bacterial expansion, with titers most prominent in the soft tissue surrounding the infected knee joint followed by the tendons/ligaments comprising the joint, femur, and implant (Fig. 1). All three S. aureus strains displayed equivalent burdens in the soft tissue surrounding the knee (Fig. 1A) and, in general, the titers were similar in the other compartments, with the exception of MW2, which was significantly greater in the joint and femur at day 7 (Fig. 1B and C). However, these increases were only apparent during the first week after infection and dissipated thereafter, with all three strains displaying roughly equivalent infectious burdens at day 14 (Fig. 1).

FIG 1.

Establishment of S. aureus orthopedic biofilm infection is not strain dependent. C57BL/6 mice were infected with 10³ CFU of S. aureus LAC, MW2, or UAMS-1, whereupon the soft tissue surrounding the joint (A), knee joint (B), femur (C), and titanium implant (D) was collected 7 and 14 days after infection for quantitation of bacterial burdens. The results are expressed as CFU per milliliter for the implant and CFU per gram of tissue (for tissue, knee joint, and femur) to normalize for differences in sampling size and are presented from individual animals combined from two independent experiments (n = 8 to 12 mice per group). Significant differences between experimental groups were determined by an unpaired, two-tailed Student t test (*, P < 0.05; **, P < 0.01).

As previously described in our S. aureus PJI model (25, 26), the innate immune response is predominantly composed of MDSCs (50 to 80% of the CD45⁺ population), which persist throughout infection (Fig. 2A). Monocytes, macrophages, and T cells are also recruited to the site of infection but are much less abundant than the MDSC infiltrate (Fig. 2B to D). Overall, LAC, MW2, and UAMS-1 elicited similar inflammatory cell infiltrates, with the exception of UAMS-1 at day 7, where MDSCs were significantly, although modestly reduced (Fig. 2A). However, the biological significance of this finding is questionable, since it did not translate into enhanced monocyte or macrophage recruitment, which we have previously shown occurs when MDSC infiltrates are decreased (25–27). Likewise, levels of inflammatory cytokine/chemokine production in the soft tissue surrounding infected joints of mice challenged with LAC or UAMS-1 were similar (G-CSF, IL-10, CCL2, CCL5, CXCL2, and VEGF; Fig. 3 and data not shown). Although the production of some mediators was increased in UAMS-1-infected tissues compared to LAC at day 7 or 14 (IL-1α, IL-1β, IL-9, CCL3, CXCL9, and CXCL10; Fig. 3 and data not shown), this did not influence bacterial persistence or leukocyte infiltrates at the same time points (Fig. 1 and 2, respectively) and thus was not considered biologically significant. UAMS-1 and LAC were utilized for all subsequent dose-response studies, since UAMS-1 originated from an orthopedic infection and we have a great deal of knowledge about the behavior of LAC in the orthopedic-implant model (25–27, 40). This approach was further justified by the finding that MW2 behaved similarly to both UAMS-1 and LAC in terms of immune profiles and biofilm burdens.

FIG 2.

S. aureus clinical isolates recruit similar leukocyte infiltrates after PJI. C57BL/6 mice were infected with 10³ CFU of S. aureus LAC, MW2, or UAMS-1, whereupon MDSC (A), monocyte (B), macrophage (C), and T cell (D) infiltrates from the soft tissue surrounding the infected knee joint were analyzed by flow cytometry at days 7 and 14 postinfection. The results are presented as means ± the standard deviations (SD) from two independent experiments (n = 8 to 12 mice per group). Significant differences between experimental groups were determined by an unpaired, two-tailed Student t test (*, P < 0.05).

FIG 3.

S. aureus LAC and UAMS-1 elicit similar cytokine and chemokine profiles. C57BL/6 mice were infected with 10³ CFU of S. aureus LAC or UAMS-1, whereupon IL-1β, G-CSF, CXCL2, and CCL5 production was measured in the tissue surrounding the knee joint at days 7 and 14 postinfection by multianalyte bead arrays. The results were normalized to the amount of total protein to correct for differences in tissue sampling size and are presented as means ± the SD from one representative experiment (n = 5 mice per group). Significant differences between experimental groups were determined by an unpaired, two-tailed Student t test (*, P < 0.05).

Infectious dose influences inflammatory responses and bacterial persistence during S. aureus PJI.

Another important factor when translating mouse infection models to human disease is infectious dose. This is particularly relevant when considering the initial bacterial challenge to the immune system, especially since the number of organisms seeding a given site during human infection is usually lower than the high S. aureus inocula used in many mouse models (i.e., 10⁵ to 10⁷ CFU) (30, 44–46). For example, a high infectious dose is likely to elicit an immediate and robust proinflammatory response attributed to a greater bacterial biomass (mediated by lipoteichoic acids, peptidoglycan, etc.). In contrast, a low inoculum may afford S. aureus an early advantage to colonize and propagate by not eliciting an initial vigorous proinflammatory response, as suggested by our prior studies (25–27).

IL-12p40 KO mice were used to investigate how the immune response to S. aureus was influenced by infectious dose, since our prior study showed that these animals had significantly reduced MDSC infiltrates and improved bacterial clearance in response to low-dose LAC infection (10³ CFU), which was confirmed in the present study (Fig. 4 and 5, respectively) (26). However, when the bacterial inoculum was increased to 10⁵ these differences between WT and IL-12p40 KO mice were no longer apparent. Specifically, MDSCs were significantly reduced concomitant with increased monocytes and macrophages in IL-12p40 KO mice in response to 10³ CFU (Fig. 4, indicated by asterisks), which was negated with the 10⁵ CFU dose, where all three populations were equivalent between IL-12p40 KO and WT animals at both 7 and 14 days postinfection (Fig. 4).

FIG 4.

Leukocyte recruitment in response to S. aureus LAC is influenced by infectious dose. Wild-type (WT) or IL-12p40 knockout (KO) mice were infected with 10³ or 10⁵ CFU of S. aureus LAC, whereupon MDSC (A), monocyte (B), macrophage (C), and T cell (D) infiltrates from the soft tissue surrounding the infected knee joint were analyzed by flow cytometry at days 7 and 14 postinfection. The results are presented as means ± the SD from three independent experiments (n = 11 to 18 mice per group). Significance was determined using an unpaired, two-tailed Student t test, with differences between WT and IL-12p40 KO mice denoted by asterisks (*, P < 0.05; **, P < 0.01; ***, P < 0.005; ****, P < 0.001), whereas differences between WT or IL-12p40 KO animals receiving 10³ CFU versus 10⁵ CFU are indicated by hash signs (#, P < 0.05; ##, P < 0.01; ####, P < 0.001).

FIG 5.

Biofilm establishment by S. aureus LAC is dependent on infectious dose. WT or IL-12p40 KO mice were infected with 10³ or 10⁵ CFU of S. aureus LAC, whereupon the soft tissue surrounding the joint (A), knee joint (B), femur (C), and titanium implant (D) was collected 7 and 14 days after infection for quantitation of bacterial burdens. The results are expressed as CFU per milliliter for the implant and CFU per gram of tissue (for tissue, knee joint, and femur) to normalize for differences in sampling size and are presented from individual animals combined from three independent experiments (n = 15 to 18 mice per group). Significant differences between experimental groups were determined by an unpaired, two-tailed Student t test (*, P < 0.05; **, P < 0.01; ***, P < 0.005; ****, P < 0.001).

A dose-response relationship was also detected with bacterial burdens. As we previously reported, biofilm titers were significantly reduced in IL-12p40 KO animals at both 7 and 14 days postinfection with an inoculum of 10³ CFU. However, when the dose was increased to 10⁵ CFU, bacterial clearance was no longer evident, since the biofilm burdens in IL-12p40 KO animals were now nearly identical or increased compared to WT mice (Fig. 5).

Similar dose-dependent effects were observed with regard to inflammatory mediator production, where numerous cytokines/chemokines (G-CSF, IL-1α, IL-1β, CXCL2, CCL2, CCL3, CCL5; Fig. 6 and data not shown) were significantly reduced in IL-12p40 KO mice after a 10³ CFU challenge in agreement with our earlier report (26). In contrast, these same inflammatory mediators were similar between IL-12p40 KO and WT animals that received an inoculum of 10⁵ CFU (Fig. 6 and data not shown). The production of IL-9, CCL2, CXCL9, CXCL10, and VEGF in IL-12p40 KO and WT mice was very low and did not differ between the two infectious inocula. We previously reported that MDSCs support biofilm persistence, in part, through IL-10 induction (25). IL-10 levels were slightly reduced in IL-12p40 KO mice at days 7 and 14 postinfection with the 10³ CFU dose; however, these differences did not reach statistical significance. Nevertheless, reduced IL-10 production in IL-12p40 KO animals correlated with improved bacterial clearance (Fig. 5), which is in agreement with our prior report demonstrating a role for IL-10 in promoting biofilm persistence (25). Similar dosage-dependent changes in IL-12p40 KO mice were observed with UAMS-1 (data not shown).

FIG 6.

A high infectious dose of S. aureus LAC alters cytokine and chemokine responses. WT or IL-12p40 KO mice were infected with 10³ or 10⁵ CFU of S. aureus LAC, whereupon IL-1β, G-CSF, CXCL2, and CCL5 production was measured in the tissue surrounding the knee joint at days 7 and 14 postinfection by multianalyte bead arrays. The results are normalized to the amount of total protein to correct for differences in tissue sampling size and are presented as means ± the SD from a representative experiment (n = 5 mice per group). Significance was determined using an unpaired, two-tailed Student t test, with differences between WT and IL-12p40 KO mice denoted by asterisks (*, P < 0.05; **, P < 0.01; ***, P < 0.005; ****, P < 0.001), whereas differences between WT or IL-12p40 KO animals receiving 10³ CFU versus 10⁵ CFU are indicated by hash signs (#, P < 0.05; ##, P < 0.01; ###, P < 0.005).

Dose-response effects of S. aureus challenge were also observed when comparing differences within WT or IL-12p40 KO animals. For example, MDSC and macrophage recruitment was significantly increased in WT mice receiving 10⁵ CFU of S. aureus compared to 10³ CFU at day 7 postinfection (Fig. 4, indicated by hash signs). However, dose-dependent differences in MDSC infiltrates were more pronounced in IL-12p40 KO animals, where 10⁵ CFU resulted in major increases in MDSC recruitment at both days 7 and 14 postinfection (Fig. 4, indicated by hash signs). Differences in biofilm burdens were also observed in WT animals following low- versus high-dose bacterial challenge. Specifically, infection with 10³ CFU resulted in bacterial expansion from day 7 to 14 in WT mice; however, this increase did not reach statistical significance. In contrast, when the higher inoculum was used (10⁵ CFU), bacterial burdens were similar at both time points in WT animals. Our prior studies have demonstrated that biofilm burdens reach a threshold titer (normally by day 14 following a 10³ CFU challenge) that can no longer expand in vivo, even with increasing time postinfection. Our interpretation for the failure to observe bacterial expansion over time with the 10⁵ CFU dose in WT animals was that biofilm development was accelerated and reached its maximal titer by day 7, since less time is required to reach maximal device colonization, effectively masking bacterial expansion. This could have dramatic consequences on the stages of biofilm development, particularly when considering the prominent role of quorum sensing during biofilm formation. Finally, increasing the dose to 10⁵ CFU resulted in heightened inflammatory mediator production compared to 10³ CFU, although this was most evident in IL-12p40 KO mice at both 7 and 14 days after infection (Fig. 6, indicated by hash signs). Collectively, these findings have established that the infectious inoculum can dictate leukocyte recruitment, inflammatory mediator production, and biofilm growth/clearance and highlight the need to carefully consider infectious doses when examining inflammatory attributes of biofilm infection.

DISCUSSION

The number of patients undergoing primary total hip and knee arthroplasties has steadily increased over the past decade, with nearly 800,000 procedures being performed in the United States each year (2). Prosthetic joint infection (PJI) is a serious complication after arthroplasty, with S. aureus being a common inciting pathogen (2). A recent study using National Inpatient Sample data from 1990 to 2003 projected the infection incidence following total hip revision to increase from 3,400 in 2005 to 46,000 in 2030, and for total knee replacements to increase from 6,400 in 2003 to 175,500 in 2030 based on the increased volume of prosthetic joint replacement procedures with an aging population (47). In addition, S. aureus PJIs are challenging because of their treatment complexity and increased risk of recurrent infection (2, 5). It is currently unclear whether certain S. aureus isolates have a propensity to cause PJIs versus colonization of other tissue sites.

The three S. aureus clinical isolates examined in this study (LAC, MW2, and UAMS-1) displayed roughly equivalent abilities to establish infection and elicited comparable immune responses when WT animals were challenged with a low infectious inoculum (10³ CFU). Whereas strain-dependent preferences have been suggested for the colonization of native tissues (i.e., USA300 strains preferentially causing skin and soft tissue infections) (48, 49), it appears that the presence of an implant likely negates these differences and facilitates bacterial attachment/persistence regardless of the strain during PJI. In general, the immune responses elicited by all three S. aureus isolates were similar, which from a therapeutic standpoint is advantageous, since a conserved immune response would be more amenable to a singular treatment regimen targeted toward enhancing antimicrobial immunity.

S. aureus is a highly opportunistic pathogen that expresses numerous virulence determinants designed to circumvent host immune responses and promote bacterial persistence. As we previously demonstrated (26), IL-12 is a critical cytokine for shaping the immune response after PJI and determining infection outcome, since IL-12p40 KO mice were more resistant to S. aureus colonization and biofilm formation during PJI than WT animals following the administration of a low inoculum (10³ CFU). Importantly, this phenotype was negated when IL-12p40 KO animals were challenged with a 2-log-higher dose of LAC (10⁵ CFU). Since bacterial division is very rapid, we propose that the initial introduction of large numbers of organisms quickly transitions into a large bacterial biomass, which elicits a vigorous proinflammatory response that is not reminiscent of what occurs during native infection in humans where tissues are often colonized with few organisms. This is supported by our findings with IL-12p40 KO animals, showing that the large infectious dose elicits a different series of events that changes the inflammatory infiltrate and prevents S. aureus clearance. This emphasizes the critical importance of the infectious dose, particularly when evaluating the efficacy of potential therapeutics for PJI in preclinical studies.