Search

OBJECTIVE In vitro, activated neutrophils create a microenvironment in which proteinase inhibitors are inactivated through the coordinate action of reactive oxygen species and released elastase. We investigated whether such a mechanism may contribute to the destruction of the joint tissues in arthritis. METHODS We analyzed the state of alpha 1-antitrypsin (alpha 1AT) and alpha 1-antichymotrypsin (alpha 1ACT), the two major inhibitors of the neutrophilic serine proteinases, in synovial fluid (SF) from patients with inflammatory arthropathies (n = 71) and osteoarthritis (OA) (n = 11), and related the results to neutrophil activation in SF. RESULTS The ratio of functional to antigenic levels of alpha 1AT in SF of patients with inflammatory joint diseases was similar to that of alpha 1AT in normal plasma, whereas that of alpha 1ACT was significantly decreased. Patients with inflammatory arthropathies had significantly higher levels of inactivated alpha 1AT (i alpha 1AT) and inactivated alpha 1ACT (i alpha 1ACT) in SF (as determined with monoclonal antibodies specific for the inactivated [i.e., proteolytically inactivated and/or complexed] forms of these inhibitors) than patients with OA (P < 0.005). Inactivated alpha 1AT and i alpha 1ACT levels corresponded to 0.3-11% and 3-99%, respectively, of the total amount of these inhibitors in SF. Most of the i alpha 1AT in SF had a lower M(r) than that of native alpha 1AT. Inactivated alpha 1ACT in SF had an M(r) identical to that of nonfunctional alpha 1ACT in plasma treated with chymotrypsin. Levels of both i alpha 1AT and i alpha 1ACT correlated significantly with lactoferrin and elastase levels. CONCLUSION These results suggest that alpha 1AT and alpha 1ACT in arthritic joints are inactivated in part by activated neutrophils, suggesting a role for these cells in impairment of the local balance between proteinases and their inhibitors in arthritis.

We studied the state of alpha 2-macroglobulin (alpha 2M), an important inhibitor of cartilage-degrading proteinases, in relation to activation of neutrophils in 82 patients with several types of arthritis, including 52 with rheumatoid arthritis and 11 with osteoarthritis. Levels of total inactive alpha 2M (i alpha 2M), which comprises alpha 2M complexed to proteinases and alpha 2M inactivated by oxidation or hydrolysis, were measured with a monoclonal antibody specific for i alpha 2M. In addition, levels of alpha 2M complexed to proteinases were quantitated with specific assays. Neutrophil activation was assessed by measuring elastase-alpha 1-antitrypsin complexes and lactoferrin. In 83% of the 82 patients tested, the synovial fluid (SF) to plasma ratio of i alpha 2M exceeded 1, indicating an intraarticular generation. Levels of i alpha 2M significantly correlated with neutrophil numbers (P less than 0.0005) and with levels of elastase-alpha 1-antitrypsin complexes and of lactoferrin (P less than 0.00001 for both). Moreover, part of i alpha 2M consisted of alpha 2M complexed to elastase-like and chymotrypsin-like proteinases, presumably, neutrophil elastase and cathepsin G, respectively. However, the amount of i alpha 2M was approximately 10-fold larger than the amount complexed to these proteinases. In vitro inactivation of alpha 2M by activated neutrophils was only partly inhibitable by eglin C, a specific inhibitor of both elastase and cathepsin G. Release of reactive oxygen species was presumably responsible for the additional inactivation of alpha 2M, because eglin C completely abolished the inactivation of alpha 2M by cell-free supernatant of activated neutrophils. Thus, our results suggest a predominant role of neutrophils in the inactivation of alpha 2M in the SF of patients with inflammatory joint diseases. However, this inactivation could be explained only in part by the release of neutrophilic proteinases. We propose that the inactivation of alpha 2M in SF was due to the concerted action of both reactive oxygen species and lysosomal proteinases.