Abstract
BACKGROUND: Orthopaedic wear particles activate the NLRP3 inflammasome to produce active interleukin 1β (IL1β). However, the NLRP3 inflammasome must be primed before it can be activated, and it is unknown whether wear particles induce priming. Toll-like receptors (TLRs) are thought to mediate particle bioactivity. It remains controversial whether pathogen-associated molecular patterns (PAMPs) and/or alarmins are responsible for TLR activation by wear particles. QUESTIONS/PURPOSES: (1) Does priming of the NLRP3 inflammasome by wear particles depend on adherent PAMPs? (2) Does priming of the NLRP3 inflammasome by wear particles depend on TLRs and TIRAP/Mal? (3) Does priming of the NLRP3 inflammasome by wear particles depend on cognate TLRs? (4) Does activation of the NLRP3 inflammasome by wear particles depend on adherent PAMPs? METHODS: Immortalized murine macrophages were stimulated by as-received titanium particles with adherent bacterial debris, endotoxin-free titanium particles, or titanium particles with adherent ultrapure lipopolysaccharide. To study priming, NLRP3 and IL1β mRNA and IL1β protein levels were assessed in wild-type, TLR4, TLR2, and TIRAP/Mal macrophages. To study activation, IL1β protein secretion was assessed in wild-type macrophages preprimed with ultrapure lipopolysaccharide. RESULTS: Compared with titanium particles with adherent bacterial debris, endotoxin-free titanium particles induced 86% less NLRP3 mRNA (0.05 ± 0.03 versus 0.35 ± 0.01 NLRP3/GAPDH, p < 0.001) and 91% less IL1β mRNA (0.02 ± 0.01 versus 0.22 ± 0.03 IL1β/GAPDH, p < 0.001). ProIL1β protein level was robustly increased in wild-type macrophages stimulated by particles with adherent PAMPs but was not detectably produced in macrophages stimulated by endotoxin-free particles. Adherence of ultrapure lipopolysaccharide to endotoxin-free particles reconstituted stimulation of NLRP3 and IL1β mRNA. Particles with adherent bacterial debris induced 79% less NLRP3 mRNA (0.09 ± 0.004 versus 0.43 ± 0.13 NLRP3/GAPDH, p < 0.001) and 40% less IL1β mRNA (0.09 ± 0.04 versus 0.15 ± 0.03 IL1β/GAPDH, p = 0.005) in TLR4 macrophages than in wild-type. Similarly, those particles induced 49% less NLRP3 mRNA (0.22 ± 0.10 versus 0.43 ± 0.13 NLRP3/GAPDH, p = 0.004) and 47% less IL1β mRNA (0.08 ± 0.02 versus 0.15 ± 0.03 IL1β/GAPDH, p = 0.012) in TIRAP/Mal macrophages than in wild-type. Particles with adherent ultrapure lipopolysaccharide induced 96% less NLRP3 mRNA (0.012 ± 0.001 versus 0.27 ± 0.05 NLRP3/GAPDH, p = 0.003) and 91% less IL1β mRNA (0.03 ± 0.01 versus 0.34 ± 0.07 IL1β/GAPDH, p < 0.001) expression in TLR4 macrophages than in wild-type. In contrast, those particles did not induce less NLRP3 and IL1β mRNA in TLR2 macrophages. IL1β protein secretion was equivalently induced by particles with adherent bacterial debris or by endotoxin-free particles in a time-dependent manner in wild-type macrophages. For example, particles with adherent bacterial debris induced 99% ± 2% of maximal IL1β secretion after 12 hours, whereas endotoxin-free particles induced 92% ± 11% (p > 0.5). CONCLUSIONS: This cell culture study showed that adherent PAMPs are required for priming of the NLRP3 inflammasome by wear particles and this process is dependent on their cognate TLRs and TIRAP/Mal. In contrast, activation of the NLRP3 inflammasome by titanium particles is not dependent on adherent PAMPs. Animal and implant retrieval studies are needed to determine whether wear particles have similar effects on the NLRP3 inflammasome in vivo. CLINICAL RELEVANCE: Our findings, together with recent findings that aseptic loosening associates with polymorphisms in the TIRAP/Mal locus, support that adherent PAMPs may contribute to aseptic loosening in patients undergoing arthroplasty.