The effect of particle wear debris on NFkappaB activation and pro-inflammatory cytokine release in differentiated THP-1 cells

Activation of NF-kappaB signalling and TNFalpha-expression in THP-1 macrophages by TiAlV- and polyethylene-wear particles

Wear particles are believed to induce periprosthetic inflammation which contributes to periprosthetic osteolysis. TNFalpha plays a pivotal role in the pathogenesis of this process. The molecular mechanisms leading to the development of periprosthetic inflammation with upregulated TNFalpha expression in monocytic cells in response to different wear particles have yet to be defined. In this study we evaluated the effects of polyethylene- and TiAlV-particles on activation of NF-kappaB signalling pathways and TNFalpha biosynthesis and release in monocytic cells with respect to periprosthetic osteoclastogenesis. THP-1 monocytic cells were differentiated to macrophage-like cells and exposed to LPS-detoxified polyethylene and prosthesis-derived TiAlV-particles. TNFalpha release was analyzed in culture supernatant by ELISA. NF-kappaB activation was examined by electrophoretic mobility shift assay (EMSA), and NF-kappaB target promoter activities including transactivation of the TNFalpha promoter were determined by luciferase reporter gene assays. Differentiated THP-1 macrophages were exposed to increasing numbers of particles for 0, 60, 180 and 360 min. Both, polyethylene- and TiAlV-particles induced a significant activation of both NF-kappaB and TNFalpha promoters at 180 min. A significant TNFalpha release was detected after 360 min exposure to polyethylene- and TiAlV-particles in a dose dependent manner. In comparison, LPS induced a much greater activation of NF-kappaB and TNFalpha promoters, and TNFalpha secretion into the supernatant was strongly induced. These results provide evidence that induction of the NF-kappaB signal transduction pathway in macrophages plays a major role in initiating and mediating the inflammatory response leading to periprosthetic osteolysis.

A New Cellular Model for In Vitro Biocompatibility Evaluation of Microcapsule Materials

Fibrosis caused by the host response to long-term transplanted microcapsules and the limitation of traditional L929 cell model for biocompatibility testing inspire the development of an assay of biocompatibility based on macrophage behavior. In this paper, the human monocytic cell line THP-1 was utilized for biocompatibility evaluation of microcapsule materials. The cell viability and secretion of nitric oxide (NO) and cytokines served as index of biocompatibility were assayed. It was found that the evaluated microcapsule materials had no effect on the stimulation of NO and cytokines secretion, which meant that these materials were biocompatible. Furthermore, it suggests the THP-1 cell a convenient in vitro experimental model that might be useful for long-term predictions of material biocompatibility.

In vivo cytokine secretion and NF-kappaB activation around titanium and copper implants

The early biological response at titanium (Ti), copper (Cu)-coated Ti and sham sites was evaluated in an in vivo rat model. Material surface chemical and topographical properties were characterized using Auger electron spectroscopy, energy dispersive X-ray spectroscopy and interferometry, respectively. The number of leukocytes, cell types and cell viability (release of lactate dehydrogenase) were determined in the implant-interface exudate. The contents of activated nuclear transcription factor NF-kappaB, interleukin-6 (IL-6) and interleukin-10 (IL-10) were determined by enzyme linked immunosorbent assay. An increase in the number of leukocytes, in particular, polymorphonuclear leukocytes, was observed between 12 and 48 h around Cu. A marked decrease of exudate cell viability was found around Cu after 48 h. The total amounts of activated NF-kappaB after 12 h was highest in Ti exudates whereas after 48 h the highest amount of NF-kappaB was detected around Cu. The levels of cytokine IL-6 were consistently high around Cu at both time periods. No differences in IL-10 contents were detected, irrespective of material/sham and time. The results show that materials with different toxicity grades (titanium with low and copper with high toxicity) exhibit early differences in the activation of NF-kappaB, extracellular expression and secretion of mediators, causing major differences in inflammatory cell accumulation and death in vivo.

Effect of titanium surface topography on macrophage activation and secretion of proinflammatory cytokines and chemokines

The macrophage has a major role in normal wound healing and the reparative process around implants. Murine macrophage-like cells RAW 264.7 were used to investigate the effect of titanium surfaces on macrophage activation and secretion of proinflammatory cytokines [interleukin (IL)-1 beta, IL-6, and tumor necrosis factor (TNF)-alpha] and chemokines (monocyte chemoattractant protein-1 and macrophage inflammatory protein-1 alpha). Four topographies were used: those produced by mechanically polishing, coarse sand blasting, acid etching, and sandblasting and acid etching (SLA). Macrophages were plated on the four titanium surfaces at a population density of 5 x 10(5) cells/mL/well. Tissue culture plastic and tissue culture plastic plus lipopolysaccharide (LPS) served as negative and positive control, respectively. In addition, all surfaces were tested for their effects on macrophages in the presence of LPS. Supernatants were collected for assays after 6, 24, and 48 h and the numbers of macrophages attached to the surfaces were quantified using the DAPI (4,6-di-amidino-2-phenylindole) assay. Cytokine and chemokine levels were measured with sandwich enzyme-linked immunosorbent assays. Statistical comparison between the surfaces and the controls was determined by using the two-way analysis of variance including interaction effect (two tailed and p < or = 0.05). Unstimulated macrophages increased their secretion of the proinflammatory cytokine (TNF-alpha) when attached to rough surfaces (acid etching and SLA, p < or = 0.05). In macrophages stimulated with LPS, the roughest surface SLA produced higher levels of IL-1 beta, IL-6, and TNF-alpha at 24 and 48 h than all other surfaces (p < or = 0.05). Surface topography also modulated the secretion of the chemokines monocyte chemoattractant protein-1 and macrophage inflammatory protein-1 alpha by macrophages. Unstimulated macrophages attached to the SLA surface down-regulated their production of chemokines (p < or = 0.05) whereas LPS-stimulated macrophages attached to the SLA surface up-regulated their production (p < or = 0.05). Moreover, the SLA surface was found to act synergistically with LPS as well as the combination of blasting and etching features of the SLA surface resulted in significant release of proinflammatory cytokines and chemokines by stimulated macrophages at 24 and 48 h (p < or = 0.05). This in vitro study has demonstrated that surface topography, in particular the SLA surface, modulated expression of proinflammatory cytokines and chemokines by macrophages in a time-dependent manner.

Particle bioreactivity and wear-mediated osteolysis

This review focuses on wear debris-mediated osteolysis, a major factor compromising the long-term success of total joint arthroplasty. Studies on retrieved implants and animal models, as well as in vitro studies on particle bioreactivity, suggest that wear-mediated periprosthetic osteolysis is unlikely to be caused solely by 1 particular cell type or particulate species, but is rather the cumulative consequence of a number of biological reactions. Our recent findings suggest 3 novel mechanisms of particle bioreactivity that may contribute to osteolysis: 1) exacerbated inflammation caused by elevated reactive oxygen species production by activated macrophages and osteoclasts, (2) impaired periprosthetic bone formation secondary to disrupted osteogenesis, and (3) compromised bone regeneration resulting from increased cytotoxic response of mesenchymal osteoprogenitor cells. Understanding the pathogenesis of wear-mediated osteolysis is needed to improve orthopedic implant biocompatibility and wear reduction, and to develop effective pharmacotherapies.

Effect of cyclic mechanical stretch and titanium particles on prostaglandin E2 production by human macrophages in vitro

Early implant instability has been proposed as a critical factor in the onset and progression of aseptic loosening and periprosthetic osteolysis in total joint arthroplasties. Previous in vitro studies have reported that macrophages stimulated with cyclic mechanical strain release inflammatory mediators. Little is known, however, about the response of these cells to mechanical strain with particles, which is often a component of the physical environment of the cell. We therefore studied the production of prostaglandin E(2) (PGE(2)), an important mediator in aseptic loosening and periprosthetic osteolysis in total joint arthroplasties, for human macrophages treated with mechanical stretch alone, titanium particles alone, and mechanical stretch and particles combined. A combination of mechanical stretch and titanium particles resulted in a statistically synergistic elevation of levels of PGE(2) compared with the levels found with either stretch or particles alone. Exposure of human macrophages to mechanical stretch with particles upregulated the expression of cyclooxygenase (COX)-2 mRNA but not COX-1 mRNA, this expression resulting in a 97-fold increase in PGE(2) production compared to the nonstimulated cells. The current study is the first to investigate the effects of mechanical stretch with particles on cultured macrophages and include an investigation of the time course of PGE(2) production and COX-2 mRNA expression. Our results suggest that, while mechanical strain may be one of the primary factors responsible for macrophage activation and periprosthetic osteolysis, mechanical strain with particles load may contribute significantly to the osteolytic potential of macrophages in vitro. The synergistic effect observed between mechanical stretch and particles could accelerate implant loosening and implies that reduction in either cyclic mechanical strain or wear debris load would lead to a reduction of osteolysis.

Detection of bacterial endotoxin in human tissues

Detection of bacterial lipopolysaccharide (LPS) in the absence of overt infection is a challenging problem in tissue homogenates and other complex samples. We found that conventional Limulus amebocyte lysate (LAL) assays are not suitable for this purpose due to interference from β-glucan-like molecules. In contrast, a modified LAL assay that is unaffected by β-glucan-like molecules was able to detect LPS in infected tissue and in a subset of clinically aseptic tissues. A two-step LAL assay was used to exclude the possibility of false positives due to nonspecific amidases. False positives due to sample color were also excluded, as were false negatives due to assay inhibition. This is the first report to successfully detect LPS in tissue in the absence of overt infection. This approach may be extremely useful in assessing recent hypotheses that subclinical levels of bacteria contribute to a wide range of chronic diseases.

Erythromycin inhibits wear debris-induced osteoclastogenesis by modulation of murine macrophage NF-kappaB activity

Activation of nuclear factor kappa B (NF-kappaB) signaling in response to cell stimulation by wear debris may be critical in the pathogenesis of aseptic loosening. Erythromycin (EM), a macrolide antibiotic, has been shown to effectively suppress some types of inflammatory reactions. In this study, we examined the effect of EM on wear debris-induced osteoclastic bone resorption in vitro. EM inhibited Ca+ release from neonatal calvaria co-cultured with conditioned medium from mouse RAW264.7 macrophages activated by wear debris. Inhibition of Ca+ release was associated with a decreased number of tartrate-resistant acid phosphatase (TRAP)-positive cells in cultured bones. To investigate the mechanism whereby EM inhibits bone-resorption, RAW cells were incubated with wear debris in the presence EM. Real time RT-CR analysis revealed that EM (5 microg/ml) significantly inhibited mRNA expression of NF-kappaB, cathepsin K (CPK), IL-1beta and TNFalpha, but not RANK in RAW cells stimulated with wear debris. Furthermore, electrophoretic mobility-shift assay showed that EM (0.2 microg-5 microg/ml) could reduce DNA-binding activity of NF-kappaB in RAW cells stimulated with wear debris. The inhibition of inflammatory osteoclastogenesis by EM treatment was further confirmed by an osteoclast (OC) formation assay using primary cultures of mouse bone marrow progenitor cells stimulated with macrophage colony-stimulating factor and RANK ligand (RANKL). EM treatment (5 microg/ml) resulted in more than 70% reduction in multinucleated OC formation and 50% reduction of TRAP+ cells by bone marrow progenitor cells. Our findings support that EM suppresses wear debris-induced osteoclastic bone resorption by, at least, down-regulation of NF-kappaB signaling pathway. It appears that EM represents a potential therapeutic candidate for the treatment and prevention of aseptic loosening.

Does endotoxin contribute to aseptic loosening of orthopedic implants?

Aseptic loosening of orthopedic implants caused by wear particles is a major clinical problem. This review examines the hypothesis that bacterial endotoxin contributes to aseptic loosening. Clinical findings support this hypothesis: bacterial biofilms exist on many implants from patients with aseptic loosening and antibiotics in bone cement reduce the rate of aseptic loosening. Three approaches were used to demonstrate that adherent endotoxin increases bioactivity of titanium particles. These experiments measured cytokine production and osteoclast differentiation in vitro and murine calvarial osteolysis in vivo. First, removal of >99.9% of the adherent endotoxin from titanium particles significantly ablates their biological activity. Second, adding lipopolysaccharide back to these "endotoxin-free" particles restores their biological activity. Third, cells or mice that are genetically hyporesponsive to endotoxin are significantly less responsive to titanium particles than are wild-type controls. Other investigators have confirmed and extended these results to include virtually all orthopedically relevant types of particles, including authentic titanium alloy particles retrieved from patients with loosening. Our recent studies suggest that adherent endotoxin on orthopedic implants may also inhibit initial osseointegration of the implants. Taken together, these studies suggest that bacterial endotoxin may have a significant role in induction of aseptic loosening.

Particles and periimplant bone resorption

There is compelling evidence that the most important factor in late periprosthetic bone resorption is an inflammatory reaction to debris. Based on results from several laboratories, it seems likely that opsonized particles activate the macrophage nuclear factor-kappa B signal transduction system via membrane receptors, leading to release of tumor necrosis factor-alpha and other cytokines and growth factors. Tumor necrosis factor stimulates osteoblasts to release cytokines that recruit inflammatory cells and osteoclast precursors to the site and promote the differentiation of early osteoclasts. Tumor necrosis factor influences fibroblasts to release tissue metalloproteinases, and induces c-src in osteoclast precursors, the expression of which is necessary for additional bone resorption. Phagocytosis of debris by osteoblasts may reduce collagen synthesis, whereas phagocytosis by fibroblasts may induce chemokines that amplify inflammation. Bone has been partially protected from particle-induced resorption in animals with defective or inhibited tumor necrosis factor or nuclear factor-kappa B signaling. Many aspects of this inflammatory reaction require clarification, including identifying the factors that influence variability among patients, and testing the importance of costimulatory molecules such as bacterial endotoxin, but the fundamental importance of particles in most cases of aseptic loosening seems certain.

Adherent endotoxin mediates biological responses of titanium particles without stimulating their phagocytosis

Aseptic loosening of orthopaedic implants is thought to be primarily due to stimulation of cytokine production by wear particles from the implants. The cytokines increase osteoclast differentiation, leading to osteolysis and implant loosening. Accumulating evidence indicates that adherent endotoxin mediates the biological responses induced by the wear particles. One mechanism by which adherent endotoxin may act is by increasing phagocytosis of the wear particles. To test this hypothesis, the effect of adherent endotoxin on phagocytosis of titanium particles was determined. First, we developed reliable confocal and fluorescence microscopy methods to examine both the attachment and internalization steps of phagocytosis. Use of these methods showed that adherent endotoxin does not detectably alter the rate or the extent of phagocytosis of titanium particles by RAW 264.7 cells. Despite this lack of an effect on phagocytosis, adherent endotoxin dramatically increases the ability of RAW 264.7 cells to produce TNF-alpha and induce osteoclast differentiation. Thus, adherent endotoxin mediates these biological responses by a mechanism that does not rely on increased phagocytosis. These results also demonstrate that phagocytosis is not sufficient to induce cytokine production and osteoclast differentiation but do not rule out the possibility that phagocytosis is required for induction of these responses by titanium particles with adherent endotoxin.