The synovial production of collagenase and chondrocyte activating factors in response to cobalt

Concentration-Dependent Effects of Cobalt and Chromium Ions on Osteoarthritic Chondrocytes

OBJECTIVE

Cobalt and chromium (CoCr) ions from metal implants are released into the joint due to biotribocorrosion, inducing apoptosis and altering gene expression in various cell types. Here, we asked whether CoCr ions concentration-dependently changed viability, transcriptional activity, and inflammatory response in human articular chondrocytes.

DESIGN

Human articular chondrocytes were exposed to Co (1.02-16.33 ppm) and Cr (0.42-6.66 ppm) ions and cell viability and early/late apoptosis (annexin V and 7-AAD) were assessed in 2-dimensional cell cultures using the XTT assay and flow cytometry, respectively. Changes in chondrocyte morphology were assessed using transmitted light microscopy. The effects of CoCr ions on transcriptional activity of chondrocytes were evaluated by quantitative polymerase chain reaction (qPCR). The inflammatory responses were determined by measuring the levels of released pro-inflammatory cytokines (interleukin-1β [IL-1β], IL-6, IL-8, and tumor necrosis factor-α [TNF-α]).

RESULTS

CoCr ions concentration-dependently reduced metabolic activity and induced early and late apoptosis after 24 hours in culture. After 72 hours, the majority of chondrocytes (>90%) were apoptotic at the highest concentrations of CoCr ions (16.33/6/66 ppm). SOX9 expression was concentration-dependently enhanced, whereas expression of COL2A1 linearly decreased after 24 hours. IL-8 release was enhanced proportionally to CoCr ions levels, whereas IL-1β, IL-6, and TNF-α levels were not affected by the treatments.

CONCLUSIONS

CoCr ions showed concentration- and time-dependent effects on articular chondrocytes. Fractions of apoptotic articular chondrocytes were proportional to CoCr ion concentrations. In addition, metabolic activity and expression of chondrocyte-specific genes were decreased by CoCr ions. Furthermore, exposure to CoCr ions caused a release of pro-inflammatory cytokines.

Effects of Loading Conditions on Articular Cartilage in a Metal-on-Cartilage Pairing

The aim of this in vitro study was to investigate the response of articular cartilage to frictional load when sliding against a metal implant, and identify potential mechanisms of damage to articular cartilage in a metal-on-cartilage pairing. Bovine osteochondral cylinders were reciprocally slid against metal cylinders (cobalt-chromium-molybdenum alloy) with several variations of load and sliding velocity using a microtribometer. The effects of different loads and velocities, and the resulting friction coefficients on articular cartilage, were evaluated by measuring histological and metabolic outcomes. Moreover, the biotribocorrosion of the metal was determined. Chondrocytes stimulated with high load and velocity showed increased metabolic activity and cartilage-specific gene expression. In addition, higher load and velocity resulted in biotribocorrosion of the metal implant and damage to the surface of the articular cartilage, whereas low velocity and a high coefficient of friction increased the expression of catabolic genes. Articular cartilage showed particular responses to load and velocity when sliding against a metal implant. Moreover, metal implants showed tribocorrosion. Therefore, corrosion particles may play a role in the mechano-biochemical wear of articular cartilage after implantation of a metal implant. These findings may be useful to surgeons performing resurfacing procedures and total knee arthroplasty. © 2019 The Authors. Journal of Orthopaedic Research® published by Wiley Periodicals, Inc. on behalf of Orthopaedic Research Society J Orthop Res 37:2531-2539, 2019.

Activation of osteoclast-mediated bone resorption by the supernatant from a rabbit synovial cell line in response to polyethylene particles

It is unknown whether the soluble factors produced from cells activated by wear particles in the fibrous tissue around failed joint prostheses really activate osteoclastic bone resorption. In this study, the activation of osteoclast-mediated bone resorption by the products from a rabbit synovial cell line (HIG-82) stimulated by various particles was investigated using rabbit unfractionated bone cells cultured on a dentin slice. The HIG cells were challenged with the following laboratory-made particles: high-density polyethylene (PE), cobalt alloy (Co-Cr), titanium alloy (Ti6Al4V), pure titanium (Ti), and sintered hydroxyapatite (HA). The size of each particle was < 2 microns. The supernatants from HIG cells cultured with the appropriate concentration of wear particles were added to unfractionated bone cells on a dentin slice, and then resorbed areas were determined for each particle. Interestingly, resorbed areas significantly increased only when the culture medium from HIG cells with PE particles was added to unfractionated bone cells. This study demonstrates that PE particles stimulate the rabbit synovial cells to produce soluble factors that induce osteoclast-mediated bone resorption. Moreover, this experimental model is a useful method sensitively to evaluate the effects of soluble factors from the cells stimulated by particulate biomaterials from joint prostheses on osteoclastic bone resorption.

Biologic effects of cobalt chrome in cell and animal models

The literature on animal and cellular models used to study the response to cobalt chrome alloy implants and wear and corrosion products is reviewed. Animal studies show that in solid form cobalt chrome alloy is relatively well tolerated. Injections of large numbers of particles in a single bolus lead to acute inflammation and necrosis, followed by a chronic inflammatory response. Macrophages are the predominant cell type and may persist in the tissues for years. Long term studies have failed to confirm the induction of tumors. In vitro studies confirm the toxic effects of cobalt chrome alloy corrosion products and wear particles, especially cobalt, and show that intracellular corrosion is an important mechanism for early release of cobalt ions. In vitro studies show that cobalt chrome alloy particles induce the release of inflammatory mediators from macrophages before causing cell death. These mediators have significant effects on osteoblastlike cells, as well as inducing bone resorption. Variations in the cell types, implantation site, and characteristics of the particles used in experimental models make interpretation of the results difficult. Standardized methods to control for size, shape, and number of particles for testing are proposed. It is important that in vitro and in vivo findings not be taken in isolation, but be compared with the results of human studies.

The potential role of fibroblasts in periprosthetic osteolysis: fibroblast response to titanium particles

Periprosthetic osteolysis with or without aseptic loosening is a major clinical problem in total hip arthroplasty. While the macrophage response to prosthetic wear debris and its role in periprosthetic osteolysis has been extensively studied, information regarding other cell types (fibroblasts, osteoblasts) is limited. This study explored the response of fibroblasts to particulate wear debris. Fibroblasts isolated from interfacial membranes of patients with failed total hip replacements and normal synovial tissue, when challenged with small-sized ( < 3 microns) titanium (Ti) particles, responded with significantly enhanced expressions of collagenase, stromelysin and, to a much lesser extent, their tissue inhibitor of metalloproteinases (TIMP). These "regulated" expressions at both mRNA and protein levels were correlated with the size and composition of particles. De novo protein synthesis was required for the regulation of these mRNAs. A similar effect could be induced by the treatment of the cells with particle-free conditioned medium from Ti particle-stimulated fibroblasts. Furthermore, this conditioned medium significantly suppressed the mRNA levels of procollagen alpha 1 (I) and alpha 1 (III) in osteoblast-like MG-63 cells. It is concluded that fibroblasts stimulated with certain particle debris may play an important role in periprosthetic osteolysis by releasing bone-resorbing metalloproteinases and mediator(s) which resulted in suppressed collagen synthesis in osteoblasts.

Isolation and characterization of wear particles generated in patients who have had failure of a hip arthroplasty without cement

Wear particles from thirty-five membranes obtained during revision hip-replacement operations were studied after digestion of the soft tissue with papain. The particles were isolated and were characterized with use of light and scanning electron microscopic techniques, x-ray microanalysis, and an automated particle analyzer. The mean size of the polyethylene particles was 0.5 micrometer, and the metal particles were a mean of 0.7 micrometer, as determined with scanning electron microscopy. The automated particle analyzer revealed a mean particle diameter of 0.63 micrometer (more than 90 per cent of all particles were less than 0.95 micrometer) and a mean of 1.7 billion particles per gram of tissue, compared with only 143 million per gram of tissue for the control samples. X-ray microanalysis revealed metal debris in sixteen (46 per cent) of the thirty-five membranes after digestion. Thirteen (50 per cent) of the twenty-six membranes surrounding a titanium-alloy stem contained metal particles, compared with three of the nine membranes surrounding a chromium-cobalt stem. Metal debris was present in only one of the twelve membranes surrounding a titanium-alloy stem without a porous coating, compared with twelve of the fourteen membranes surrounding a titanium-alloy stem with a porous coating. This tenfold difference in prevalence was significant (p < 0.005). On the average, the total number of particles (expressed in millions per gram of tissue) associated with the bipolar acetabular components was twice that associated with the fixed acetabular components. In addition, there was a trend toward a larger mean size of the polyethylene particles in association with the bipolar cups. Our data indicate that particulate prosthetic debris in the tissues around failed femoral components that have been inserted without cement constitutes a class of particles that are predominantly less than one micrometer in size and are present in amounts of more than one billion particles per gram of tissue. Routine histological methods did not detect this class of wear debris and led to a gross underestimation of the amount of debris in these membranes.

Particle-induced synthesis of collagenase by synovial fibroblasts: an immunocytochemical study

The response of synoviocytes to wear particles has been implicated in several orthopaedic pathologies, including the synovitis associated with the failure of synthetic anterior cruciate ligament (ACL) replacements. To study the interactions of particles with synovial fibroblasts at the level of the individual cell, we employed immunocytochemistry, with use of antiserum, to lapine interstitial collagenase. Cultures of the HIG-82 lapine synovial cell line showed only weak immunofluorescence under resting conditions. Incubation with phorbol myristate acetate or autocrine factors known as cell-activating factors (CAFs) induced marked changes in morphology and intense immunofluorescence. This technique then was used to study the effects of standard particles of latex and of particles generated from two prosthetic ACL materials, Dacron and carbon. Internalized particles of latex, Dacron, and carbon all induced collagenase. Particles of latex that were too large for endocytosis failed to induce collagenase, whereas particles of carbon and, in particular, Dacron that remained extracellular, still provoked considerable synthesis of collagenase. Thus, both the size and the physical properties of these materials influence their ability to activate synoviocytes. Certain cells that appeared by visual inspection to contain no particles nevertheless produced collagenase when in co-culture with cells that did contain particles. This is consistent with earlier biochemical data showing that phagocytosis, in addition to inducing collagenase, also provokes the release of CAFs, which then activate additional cells in the culture. More rarely, cells were identified which, although containing particles, did not stain positively for collagenase.(ABSTRACT TRUNCATED AT 250 WORDS)

Adsorption of collagenase to particulate titanium: a possible mechanism for collagenase localization in periprosthetic tissue

Osteolysis is a central feature of aseptic loosening of orthopaedic joint prostheses. This destructive process is believed to result from phagocytosis of implant wear debris by periprosthetic and synovial macrophages and the subsequent release of proinflammatory mediators, including collagenase. Isolated murine macrophages were cultured in vitro with particulate titanium in order to explore the mechanism of macrophage activation by particulate wear debris. The results, in which the amount of secreted, soluble collagenase in culture supernatants was inversely proportional to titanium concentration, suggested that titanium strongly adsorbed secreted collagenase. This inference was confirmed by direct binding assays in which particulate titanium coated with adsorbed collagenase bound an alkaline phosphatase conjugated anti-collagenase antibody, but not a conjugated anti-IgG antibody. Adsorption of collagenase was not influenced by preincubation of titanium particles with albumin. The adsorbed collagenase remained enzymatically active as indicated by its ability to hydrolyze a synthetic peptide substrate. These results demonstrate that particulate titanium stimulates collagenase production by macrophages and then strongly adsorbs the secreted proinflammatory enzyme. The process of macrophage stimulation, collagenase secretion, and adsorption may represent an important mechanism for localization and concentration of collagenase in periprosthetic and synovial tissue, a mechanism that ultimately triggers bone resorption through osteoclast activation.

Fibroblast response to metallic debris in vitro. Enzyme induction cell proliferation, and toxicity

Bovine synovial fibroblasts in primary monolayer culture were exposed to particulate metallic debris. The effects of the metallic particles on the synthesis and secretion of proteolytic enzymes and on cell proliferation and viability were examined. Uniform suspensions of titanium, titanium-aluminum, cobalt, and chromium particles, ranging in size from approximately 0.1 to ten micrometers (average, one to three micrometers), were prepared; the particle concentrations (the volume of particles divided by the total volume of the suspension) ranged from 0.0005 to 5 per cent. Aliquots of the particle suspensions were added to the synovial fibroblast cultures. The final particle concentrations in the media ranged from 0.0000083 to 0.83 per cent. After seventy-two hours of exposure, each medium was harvested and was assayed for proteolytic and collagenolytic activity and for hexosaminidase levels. Neutral metalloproteases, quantified by collagenolytic and caseinolytic (proteolytic) activity, represent enzymes, secreted by cells, that are capable of degrading extracellular matrix. Hexosaminidase is a marker for lysosomal enzyme activity that can include more than thirty enzymes, such as proteases, lipases, nucleases, and phosphatases. Cell proliferation was quantified by uptake of 3H-thymidine. Cell morphology was examined by scanning electron microscopy. Titanium, titanium-aluminum, and chromium significantly stimulated 3H-thymidine uptake at low particle concentrations (p < 0.01, p < 0.002, and p < 0.002, respectively). Exposure to cobalt, even at the lowest particle concentration, resulted in a significant decrease in thymidine uptake (p = 0.027). At the highest particle concentrations, all particles were toxic, as evidenced by the absence of thymidine uptake. At high particle concentrations, all of the metals caused a decrease in caseinolytic (proteolytic) and collagenolytic activity in the culture media. Titanium elevated the lysosomal enzyme marker, hexosaminidase, except at high concentrations. Chromium and titanium-aluminum had no significant effect on hexosaminidase at any particle concentration, while cobalt decreased all enzyme markers at mid-particle to high-particle concentrations. Scanning electron microscopy demonstrated that the morphological response of fibroblasts to titanium included membrane-ruffling and extension of filopodia, typical of active fibroblasts. In contrast, exposure to cobalt at the same concentration resulted in cell crenation, indicative of cell death.

The biologic effects of implant materials

The interaction between implant materials and the surrounding biological environment continues to be an area of intense research and clinical interest. This article presents the information presented in a symposium, held during the 36th Annual Meeting of the Orthopaedic Research Society, in which several important issues concerning the biologic effects of implant materials were discussed. These issues included the mechanisms by which implant materials are released to the surrounding tissues and the ways in which these tissues respond to implant materials. The problem of bone loss around cementless implants was discussed as a specific example of a biologic effect resulting in both bone remodelling and endosteal erosion.

Studies on the autocrine activation of a synovial cell line

The lapine synovial cell line HIG-82 secretes factors that activate cultures of articular chondrocytes. We showed that these "chondrocyte-activating factors" (CAF) also activate quiescent cultures of HIG-82 cells in an autocrine fashion. After exposure to partially purified preparations of CAF, HIG-82 cells increased their synthesis of prostaglandin E2 (PGE2) and the neutral proteinases collagenase, gelatinase, and stromelysin. CAF also induced their own synthesis. Both PGE2 synthesis and endogenous production of CAF started to increase between 1 and 3 h after treatment of cells with exogenous CAF, but the neutral proteolytic activity of the conditioned medium took approximately 12 h to increase. Induction of neutral proteinases by CAF was inversely related to the degree of cell confluency, whereas their induction by phorbol myristate acetate (PMA) was independent of this parameter. Both CAF and PMA provoked morphologic changes in subconfluent cultures of HIG-82 cells. Although the intracellular concentration of free Ca2+ increased rapidly in response to CAF, the results of experiments with calcium channel blockers and ionophores failed to support a role for Ca2+ fluxes in induction of neutral proteinases. In similar types of experiments, no evidence could be found to implicate fluxes in cyclic AMP or cyclic GMP in the induction of collagenase, gelatinase, or stromelysin. Because PMA is such a strong inducer of these enzymes, protein kinase C may be involved in signal transduction, but further work is needed to determine whether this is so.

Cellular events in the mechanisms of prosthesis loosening

The functional restoration of a joint damaged by trauma or disease is obtained by prosthetic surgery. In particular the implantation of hip prostheses is regarded as routine in orthopedic surgery and thorough research has been developed in this field. The prosthetic replacement of the knee and even more so the ankle and elbow occurs less frequently in clinical practice and has been studied less intensively. The results of artifical hip replacement are generally good, both in terms of pain relief and the restoration of satisfactory joint function. Nevertheless, as time passes, a high rate of failures have been recorded due to prosthesis infections, fracture and wearing of the prosthetic components and prosthesis loosening by various causes. The use of ultra-filtered air and laminar flow in operating theatres and antibiotic prophylaxis have dramatically reduced the incidence of infections in total hip arthroplasty. Thanks to the setting up of new stem configurations and the use of superalloys that are highly resistant to fatigue failure, the fracture of the femoral component has been virtually eliminated as a complication of total hip arthroplasty replacements. Loosening is thus the most frequent complication in total hip replacement.