Development of a novel 3D in vitro model to measure cellular response to antioxidant doped highly cross-linked ultra high molecular weight polyethylene wear debris

Osteolysis, and subsequent aseptic loosening, is the leading cause of revision in total hip replacements. The biological response to ultra-high molecular weight polyethylene (UHMWPE) wear debris is known to trigger the release of several inflammatory mediators such as tumour necrosis factor-alpha (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6), interleukin-8 (IL-8), leukotriene B4 (LTB4) and prostaglandin E2 (PGE2); all of which are involved in the signalling cascade that triggers the differentiation of osteoclasts and begins osteolysis. UHMWPE has undergone modifications such as cross-linking (to reduce the volume of wear debris produced) and antioxidant doping (to neutralise free radicals produced by irradiative crosslinking and prevent oxidative embrittlement). Previous studies at the University of Leeds, using peripheral blood mononuclear cells (PBMNCs) cultured on wear particle containing agarose gel, have shown that antioxidant doped, highly cross-linked UHMWPE may reduce the production of osteolytic cytokines in comparison to highly cross-linked UHMWPE. This model has limitations; namely donor heterogeneity associated with the use of PBMNC’s in addition to the use of a 2D model for what is a 3D in vivo environment. The aims of this study were to develop a novel 3D model in which mononuclear phagocytes and wear particles were encapsulated within a gel and to use the model to compare the cellular production of osteolytic cytokines when treated with highly cross-linked UHMWPE and antioxidant doped, highly cross-linked UHMWPE wear debris. Three different cell types (PBMNC’s, U937 human histiocytes and RAW 264.7 murine macrophages) and two gel types (agarose and type I rat tail collagen) were investigated during the development of the 3D model. It was found that the use of RAW 264.7 murine macrophages and UHMWPE wear debris simultaneously trapped inside a collagen gel matrix was the most promising model. A large volume of sterile highly cross-linked UHMWPE wear debris (with and without antioxidant doping) was generated using pin-on-plate (POP) multidirectional wear simulator rigs for use in the 3D model. RAW 264.7 murine macrophages were cultured with both unfractionated UHMWPE wear debris, or a filtered, critical size range (0.1 – 1 μm) of UHMWPE wear debris. RAW 264.7 murine macrophages cultured in the 3D model with unfractionated UHMWPE wear debris showed no difference in oxidative stress levels or TNF-α, IL-1β, IL-6, KC (CXCL-1 – a murine homologue of IL-8), LTB4 and PGE2 levels produced in response to highly cross-linked UHMWPE wear debris with and without antioxidant doping compared to the cells only controls. Similarly, RAW 264.7 murine macrophages cultured in the 3D model with a critical size range (0.1 – 1 μm) of UHMWPE wear debris showed no difference in oxidative stress levels or TNF-α, IL-1β, IL-6, KC, LTB4 and PGE2 levels when treated with highly cross-linked UHMWPE wear debris with and without antioxidant doping compared to the cells only controls. RAW 264.7 murine macrophages were also stimulated with LPS in the 3D model and then treated with Protein Kinase C (PKC) inhibitors (Calphostin C and Bisindolylmaleimide I), 5-lipoxygenase (5-LO) inhibitors (MK 886 and REV 5901), Cycloxygenase (COX) inhibitor (Indomethacin) and antioxidants (α –tocopherol and pentaerythritol tetrakis(3,5-di-tert-butyl-4-hydroxyhydrocinnamate)) after three hours. RAW 264.7 murine macrophages treated with both PKC inhibitors and both antioxidants showed reduced levels of TNF-α and IL-6 when compared to the LPS only control. All inhibitors and antioxidants reduced the production of IL-1β when compared to the LPS only control. Cells treated with Calphostin C exhibited a reduced production of LTB4 and PGE2 when compared to the LPS only control. Microscopy images of RAW 264.7 murine macrophages cultured in the novel 3D model incubated with 0.1 – 1 μm antioxidant doped, highly cross-linked UHMWPE wear debris and FluoSpheres® showed that the RAW 264.7 murine macrophages sequestered the UHMWPE wear debris in the perinuclear lysosomes, indicating attempted phagocytosis, which is consistent with the previous studies. Overall, the RAW 264.7 murine macrophages cultured in the novel 3D model largely failed to produce inflammatory and osteolytic cytokines, chemokines and mediators in response to the UHMWPE wear debris generated in this study. However the novel model showed some promise by demonstrating the link between vitamin E and PKC in reducing the production of inflammatory and osteolytic mediators as shown in previous studies. The novel culture model requires further work to produce a more accurate in vitro representation of the 3D periprosthetic environment for use in future studies.