Expression profiling reveals alternative macrophage activation and impaired osteogenesis in periprosthetic osteolysis (original) (raw)

2008, Journal of Orthopaedic Research

Aseptic loosening is the most critical problem in total hip arthroplasty. It can occur in even a technically well-inserted prosthesis. Although many reports have been published on the pathogenesis of loosening, the precise mechanism of loosening has not been elucidated 1,2. In animal models proinflammatory cytokines have been proposed to have an important role 3. However, this has not been proved in human tissue. The current study was designed to determine the cellular the cytokine and chemokine network from periprosthetic tissues of loose hips joints and from synovium in hips with osteoarthritis. Materials and Methods. Interface tissues between bone and prosthesis were collected from 15 cases of aseptic loose artificial hip joint at revision. Synovial tissue samples were collected from 14 patients with hip osteoarthritis at primary surgery Table I. The samples were retrieved fresh, using a sterile technique, and immediately were placed in stabilizing solution (RNA later, QIAGEN) and stored at-20 0 C. Total RNA was isolated (Trizol) and the quality was confirmed by aragose gel electrophoresis. Real-time RT-PCR was carried out using iQ TM SYBR ® green supermix reagent (Bio-RAD, CA), and mRNA amounts were normalized relative to control gene HPRT. Generation of only the correct amplification products confirmed by melting curve analysis of the products. The Mann-Whitney test was applied to detect significant differences in the expression levels for each mRNA between the two groups. Results. Relative mRNA expression levels, fold increase or decrease in osteolysis patients compared to controls, and p values are shown in Table II. Osteoclastogenesis. Osteoclast markers (Cath K, TRAP, MMP9, DC-STAMP) were significantly higher in osteolysis patients. Osteoprotegerin was low in osteolysis while RANKL was not significantly different in two groups. Macrophage activation. There was no difference in levels of the pro-inflammatory cytokines TNFa and IL-6. However, the alternative macrophage activation markers chitinase and CCL18 were 66 and 3 fold more in osteolysis, respectively. BCL2A1, an antiapoptotic macrophage protein was 16 fold more abundant in osteolysis patients. Chemokines. IL-8 and MIP1A were significant higher in osteolysis Osteogenesis. BMP4 and FGF18 were significantly lower in osteolysis patients than controls. Discussion The data from this in vivo study show that proinflammatory cytokines are not expressed at significantly elevated levels in end stage osteolysis patients. Proinflammatory cytokines signaling is prominently involved in animal models of osteolysis, and may be involved in the early stages of osteolysis. However, our data showed a different pathogenesis during the chronic stage of osteolysis. Rather, an alternative activation of macrophages characterized by increased levels of markers as Chit-1, CCL18 and BCL2A1 is evident. Elevated expression of chemokines and decreased OPG (but unchanged RANKL expression) contributes to the increased osteoclastogenesis associated with this disease. In addition, this study shows an impairment of osteoblastic bone formation indicated by decreased BMP4 and FGF18