We measured the levels of az-macroglobulin (azM)-proteinase complexes in the plasma of 18 patients with classic rheumatoid arthritis, 11 age-matched patients with noninflammatory back pain and osteoarthritis, and 8 healthy volunteers. In contrast with previous reports, we found no evidence of a2M-proteinase complexes in the plasma samples from individuals in any of the groups. In our assays, all activity that might have been the result of the presence of such complexes in the plasma samples proved instead to be an artifact attributable to contamination of the anti-a2M antibody immobilized on the AffiGel solid phase with a trypsin-like proteinase. When the contaminating activity was eliminated by pretreatment of the antibody with 1 mM diisopropyl fluorophosphate, no degradation of substrate was detected with any of the plasma samples. However, the ability of the solid-phase assay to detect and quantitate a2M-proteinase complexes when they are present was confirmed in control experiments with plasma samples to which preformed a2M-trypsin complexes had been added, or in which a2M-kallikrein complexes had been generated by activation of Hageman factor (coagulation factor XII). We therefore conclude that neither normal plasma nor that from rheumatoid arthritis patients contains measurable amounts of aZMproteinase complexes.It has been reported by Teodorescu and coworkers (1-6) that patients with classic seropositive rheumatoid arthritis (RA) consistently have abnormally high plasma concentrations of complexes of a2-macroglobulin ( a2M) with a trypsin-like proteinase. Such concentrations were not found in healthy individuals or patients with nonrheumatoid inflammatory arthritides. If confirmed, this observation could form the basis of a valuable laboratory test to distinguish patients with RA from those with other inflammatory arthritides. The report was surprising, however, in that complexes between proteinases and azM are known to be cleared very rapidly from the circulation, and so have not previously been found to accumulate to appreciable concentrations in the plasma (7,8).Teodorescu et a1 (6) used a modification of the solid-phase immunosorbent assay which was initially developed by Harpel and Hayes (9) for measuring a2M-proteinase complexes. In this assay, AffiGel agarose beads coated with monospecific antibodies against a2M are used to adsorb the complexes from the plasma. The enzymic activity of the adsorbed complexes is then measured as the hydrolysis of the low M , synthetic substrate, benzyloxycarbonyl-lvalylglycyl-L-arginine p-nitroanilide (Z-Val-Gly-ArgNEIPhNOZ), also known as Chromozym-Try . The activity detected in the arthritis plasma samples (5,6) was not fully characterized and was described as trypsinlike because of its action on Chromozym-Try, abroadspectrum trypsin substrate.In the present study, we attempted to repro-