The role of adsorbed endotoxin in particle-induced stimulation of cytokine release

Numerous in vitro models have demonstrated the capacity of wear particles to stimulate the release of soluble pro-inflammatory products with the ability to induce local bone resorption. Recent observations have demonstrated that binding of lipopolysaccharide (LPS) to particulate wear debris can significantly modulate the pattern of cell response in the in vitro models. These findings raise concerns over the possible role of LPS in the pathogenesis of aseptic loosening after total joint replacements, and also indicates the importance of controlling for possible confounding effects of LPS contamination in the in vitro models used to study the reactive nature of wear debris. Our studies were undertaken to rigorously analyze the effects of particle-associated LPS on cell responses and to assess the efficacy of different treatment protocols to inactivate LPS associated with different particulate materials. Particles of cobalt-chrome alloy, titanium-6-aluminum 4-vanadium, titanium nitride and silica were pretreated with LPS and exposed to multiple treatment protocols. When cells were treated with "as-received" particles prepared by washing in ethanol, small amounts of TNF-alpha, IL-1beta, and IL-1alpha were detected. In contrast, all particle species pretreated with LPS produced marked increases in TNF-alpha, IL-1alpha, and IL-1beta release, as well as upregulation of corresponding mRNA levels even after ethanol washing. Boiling the LPS-pretreated particles in 1% acetic acid or autoclaving and baking the particles also markedly reduced and in some instances abolished the effect of the LPS-pretreatment. This indicates that LPS binds to the surface of particles of diverse composition and that the bound LPS is biologically active. Treatment protocols to inactivate particle-associated LPS demonstrated significant differences in efficacy. When the most rigorous treatments were utilized, essentially all LPS activity could be eliminated. Particles treated with these methods retained some capacity to stimulate cytokine release, but activities were markedly reduced. These results provide further evidence indicating that LPS contamination of particulate materials can markedly enhance their biological activity. This potential confounding effect needs to be carefully monitored and controlled in the in vitro model systems used to evaluate wear particles. Furthermore, the presence of particle-associated endotoxin at the bone implant interface in vivo could markedly enhance the adverse biological activity of particulate wear debris. (C) 2002 Published by Elsevier Science Ltd. on behalf of Orthopaedic Research Society.