High cathepsin B activity in arthroplasty interface membranes. A histochemical study of 9 loose cemented total hip prostheses

Polyethylene wear particles do not induce inflammation or gelatinase (MMP-2 and MMP-9) activity in fibrous tissue interfaces of loosening total hip arthroplasties

In vitro and in vivo studies have suggested that polyethylene wear particles are the main cause for osteolysis in prosthetic loosening. Elevated amounts of proteases including gelatinases (or matrix metalloproteinases MMP-2 and MMP-9) have been found in fibrous tissue interfaces of loosened total hip arthroplasties suggesting that proteolysis plays a role in osteolysis. The presence of proteases does not mean that they are active, because activity of proteases is highly regulated at the post-translational level. We investigated whether the activity of two major proteases that are active extracellularly and have been associated with loosening, MMP-2 and MMP-9, is involved in loosening of non-cemented hip implants with polyethylene acetabular components. Eight interface tissues retrieved during revision were studied with light and electron microscopy and by in situ zymography to localize MMP-2 and MMP-9 activity in combination with immunohistochemistry to localize MMP-2 and MMP-9 proteins. All interface tissues contained large amounts of polyethylene wear particles, either in large accumulations or dispersed in the extracellular matrix or intracellularly in fibroblasts. Particles were not encountered in association with MMP-2 or MMP-9 activity or leukocytes. Inflammation was never found. MMP-9 activity was restricted to macrophages and MMP-2 activity was restricted to microvascular endothelial cells mainly outside areas where particles were present. Our data indicate that wear particles do not induce activation of leukocytes or MMP-2 or MMP-9 activity. Therefore, aseptic loosening may not be particle induced but initiated by other mechanisms such as mechanical stress.

Expression of matrix-degrading cysteine proteinase cathepsin K in cholesteatoma

Cholesteatoma is a nonneoplastic lesion of the middle ear space or mastoid that is histologically characterized by a progressive bone erosion of the ossicles and surrounding bone. Several matrix-degrading enzymes have been implicated as mediators of this bone erosion. Because the novel cysteine proteinase cathepsin K has been shown to play a central role in bone resorption, we examined the expression of this enzyme in tissue specimens of cholesteatoma. Tissue specimens of 9 patients with cholesteatoma were obtained during middle-ear surgery. Expression of cathepsin K mRNA was determined by RT-PCR using specific primers. Immunohistochemical analysis of cathepsin K protein expression in tissue sections was performed by using the streptavidin-alkaline phosphatase technique. Expression of both cathepsin K mRNA and protein was detected in areas affected by cholesteatoma, whereas specimens of nonaffected ear cartilage and surrounding tissue were not positive. In addition, cathepsin K was detected in numerous multinucleated giant cells, particularly osteoclasts at the site of bone degradation. In contrast, keratinized squamous epithelium was negative for cathepsin K. These data demonstrate that the matrix-degrading cysteine proteinase cathepsin K may be involved in bone erosion in cholesteatoma. Strong expression of this collagenolytic enzyme in osteoclasts suggests that these cells are mainly involved in cathepsin K-mediated bone destruction.

Immunohistochemical evaluation of interface membranes from failed cemented and uncemented acetabular components

Aseptic loosening of acetabular components is a primary factor compromising the long-term outcome of cemented and cementless total hip replacement. It is unknown whether the pathogenesis of the loosening process is identical for both types of fixation. The specific aim of this study was to determine whether there is a difference in the cellular and cytokine profiles of interface membranes removed from between the implant and the host bone from failed cemented (n = 5) and failed cementless acetabulae (n = 5). Routine histology and immunohistochemical evaluations were completed on each tissue specimen. The monoclonal antibodies used included those specific for cell types (macrophages, fibroblasts, T lymphocytes) and for cytokines (IL-1beta, IL-6, TNF-alpha). The patients were all revised for loosening. The time to revision was significantly longer for the cemented group (16.6 yr; 13-21 range) than for the cementless group (8.9 yr; 4-13 range). In all cases, slides from each group stained positively for each of the cell types and cytokines evaluated. Immunohistochemical analyses indicated a predominance of macrophages and ubiquitous staining for the cytokines IL-1beta and TNF-alpha within the membranes of both patient groups. The intensities of cytokine staining were similar for both patient groups. More regions of fibroblastic connective tissue were observed surrounding failed cementless components as compared to those of the cemented group. The clinical ramification of our findings is that, despite differences in the cellular composition of the periprosthetic membranes, the membranes from failed cemented and cementless implants contain cytokines, which have been shown to be capable of modulating the inflammatory response. These inflammatory mediators are likely to play a significant role in the development of osteolysis and prosthetic loosening.

The plasminogen activation system is upregulated in loosening of total hip prostheses

Interface tissues and pseudocapsules from loose total hip replacements were removed during revision of 11 cases and were investigated for the plasminogen activation system and IL-1beta. Control samples of synovium were taken during knee arthroscopy (n 8), and from the hip joint during primary total hip replacement (n 5). The concentrations of urokinase plasminogen activator (uPA), tissue type plasminogen activator (tPA), plasminogen activator inhibitor-1 (PAI-1) and interleukin 1beta were all found to be significantly different in interfaces and in pseudocapsules, compared to controls. Immunohistochemistry disclosed localization in periprosthetic tissues of uPA, uPA-receptor and tPA in macrophages with phagocytosed metal, polyethylene, cement particles or accompanying pieces of necrotic bone. PAI-1 staining was present in the neighboring areas that stained for uPA or tPA, but PAI-1 staining was also found overlapping and outside these areas. These findings suggest a role for the uPA/uPA- receptor and PAI-1 in activation and focalization of extracellular matrix degradation in periprosthetic tissues. The expression of the plasminogen activation system by macrophages containing phagocytosed material suggests undegradable microdebris as a possible initiating and perpetuating stimulus for a proteolytic activation cascade, which may contribute to loosening of the prosthesis.