Titanium Wear Particles Exacerbate S. epidermidis-Induced Implant-Related Osteolysis and Decrease Efficacy of Antibiotic Therapy

Abstract

Total joint arthroplasty (TJA) surgeries are common orthopedic procedures, but bacterial infection remains a concern. The aim of this study was to assess interactions between wear particles (WPs) and immune cells in vitro and to investigate if WPs affect the severity, or response to antibiotic therapy, of a Staphylococcus epidermidis orthopedic device-related infection (ODRI) in a rodent model. Biofilms grown on WPs were challenged with rifampin and cefazolin (100 µg/mL) to determine antibiotic efficacy. Neutrophils or peripheral blood mononuclear cells (PBMCs) were incubated with or without S. epidermidis and WPs, and myeloperoxidase (MPO) and cytokine release were analyzed, respectively. In the ODRI rodent model, rats (n = 36) had a sterile or S. epidermidis-inoculated screw implanted in the presence or absence of WPs, and a subgroup was treated with antibiotics. Bone changes were monitored using microCT scanning. The presence of WPs decreased antibiotic efficacy against biofilm-resident bacteria and promoted MPO and pro-inflammatory cytokine production in vitro. WPs exacerbated osteolytic responses to S. epidermidis infection and markedly reduced antibiotic efficacy in vivo. Overall, this work shows that the presence of titanium WPs reduces antibiotic efficacy in vitro and in vivo, induces proinflammatory cytokine release, and exacerbates S. epidermidis-induced osteolysis.

Keywords: Staphylococcus epidermidis; antibiotics; microCT; osteomyelitis; wear particles.

Figure 1

In vitro antibiotic killing activity is reduced when bacteria are cultured on titanium WPs. (A) Antibiotic killing activity of rifampin/cefazolin (both 100 µg/mL) against S. epidermidis cultured for 24 h with 0.1 or 1 mg/mL WPs. Data represent the CFU/mL with and without antibiotic treatment. Statistical analysis using 2-way ANOVA with Sidak’s multiple comparison test was performed between the rifampin/cefazolin and no antibiotic group (p < 0.0001, not shown) and one-way ANOVA with Dunn’s multiple comparison test within the rifampin/cefazolin family, n = 3. (B) SEM images of bacteria growing for 24 h in the presence of 1 mg/mL WPs and after 24 h in the absence (left) or presence of rifampin/cefazolin (right). Arrows indicate bacteria; asterisks indicate the WPs.

Figure 2

WPs decrease the inhibitory effects of neutrophils on S. epidermidis growth and induce MPO release. (A) Primary neutrophils killing activity expressed as CFU/mL calculated after 3, 6 or 24 h of incubation with S. epidermidis in the absence or presence of 0.1 or 1 mg/mL WPs. Statistical analysis by 2-way ANOVA with Tukey’s multiple comparison test (n = 3 independent donors); (B) myeloperoxidase (MPO) release from primary neutrophils after 24 h culture in the absence or presence of 0.1 or 1 mg/mL WPs, and after activation with 100 nM N-formyl-Mel-Leu-Phe (FMLP) (n = 5 independent donors). Statistical analysis by ordinary one-way ANOVA with Dunnett’s multiple comparison test, n = 5.

Figure 3

WPs induce pro-inflammatory cytokine release from PBMCs in the presence of bacterial-derived factors. Peripheral blood mononuclear cells (PBMCs) were incubated for 24 h with 0.1 or 1 mg/mL WPs in the absence or presence of 10 µg/mL lipoteichoic acid (LTA). Cytokine levels in conditioned media were assessed using multiplexed ELISA: (A) IL-1β; (B) IL-6; (C) TNF-α; (D) IL-10. Left panels are unstimulated cultures; right panels are following LTA treatment. Statistical analysis by one-way ANOVA with Dunn’s multiple comparison test; n = 3 independent donors performed in duplicate.

Figure 4

Wear particles increase bacterial burden and diminish antibiotic efficacy in vivo. A sterile or S. epidermidis-inoculated screw was implanted into the proximal tibia of female Wistar rats in the presence of absence of 2 mg titanium WPs. In the antibiotic-treated groups (Infected + ABs), animals were administered rifampin plus cefazolin. At euthanasia (day 28) the number of bacteria (CFU) was determined by quantitative bacteriology. Results shown are from individual animals (n = 6 per group). The mean is indicated by the horizontal bar. Culture-negative samples were arbitrarily assigned a value of 1 for the purposes of displaying on a log₁₀ axis. Statistical analysis performed using the Kruskal–Wallis test with Dunn’s multiple comparison test. A Fisher’s exact test (indicated) was performed to compare proportions of infected animals between antibiotic-treated groups in the absence or presence of WPs.

Figure 5

MicroCT revealed increased S. epidermidis-induced osteolysis in the presence of WPs. A sterile or S. epidermidis-colonized screw was inserted in the unicortical hole created in the proximal tibia of female Wistar rats. Further subgroups included animals receiving 2 mg of sterilized titanium WPs into the screw hole and, in animals receiving the S. epidermidis-colonized screws, a combination antibiotic regimen (+ABs) was administered on days 7–21, followed by a 7-day washout period. The evolution of bone structure was assessed by in vivo microCT imaging at day 0, 6, 9, 14, 20 and 28. n = 6 per experimental group. Asterisks highlight osteolysis around the screw, arrows represent thickening of periosteal regions in proximity of the screw, and white arrows (diamond-headed arrow) represent WPs. Scale bar = 1 mm.

Figure 6

Quantitative morphometric analyses reveal changes in bone fraction, periosteal reaction bone formation and bone resorption in response to S. epidermidis infection and presence of wear particles over 28 days. S. epidermidis-inoculated screws induce marked changes in (A) bone fraction (BV/TV), (B) periosteal reaction, (C) bone formation (BF) and (D) bone resorption (BR) over time, which is affected by the presence of titanium WPs. Data shown are the mean ± SEM. Statistical analysis was by 2-way ANOVA with Tukey’s multiple comparison test: * indicates a significant difference between Infected and Infected+WPs (p < 0.05); # indicates a significant difference between Sterile and Sterile + WPs (# p < 0.05; ## p < 0.01; ### p < 0.001).