No abstract

An enzyme-linked immunosorbent assay for detection and quantification of the terminal complexes (SC5b-9 and membrane attack complex) of human complement is described. We separate the complex from the native complement components, to use antibodies against the native components in a 'double-antibody sandwich' technique. It is thereby possible to detect the terminal complement complex in solution without the requirement of specific antibodies against the neoantigens. The results show that the assay is both sensitive and specific. Evidence is presented that a terminal complement complex occurs in a normal plasma pool. The terminal complement complex may be valuable for evaluating both the physiology and pathophysiology of the complement system in vivo.

Synovial fluid (SF) and synovial tissue from 10 patients with juvenile rheumatoid arthritis were examined. The SFs were heterogeneous with respect to the degree of complement activation. Quantification of C3dg and the terminal complement complex revealed a positive correlation between activation of the early and the late parts of the cascade in all patients. The amount of C‐reactive protein and the number of white blood cells in the SF correlated significantly with the degree of complement activation. Weak deposits of C3, C3dg, or terminal complement complex were observed in a few vessels in the synovial tissue from 5 of the patients. There was no correlation between complement activity in SF and in the corresponding tissue. Furthermore, there was no correlation between clinical activity in the joints and the degree of complement activation. It is concluded that there is a discrepancy between synovial tissue and synovial fluid with respect to complement activation. C‐reactive protein may, to some extent, be responsible for activation in SF, and the accumulation of white blood cells may be due to complement activation products.

Complement activation products, C9 and C3-containing circulating immune complexes (CIC), were evaluated in plasma and synovial fluid (SF) from patients with rheumatoid arthritis (RA) and osteoarthritis. C3 activation products and the fluid phase terminal complement complex were considerably elevated in SF from RA patients reaching levels five- to eighttimes that in plasma, consistant with a local activation of the whole cascade in the joints. The results emphazise the importance of detecting C3 activation by neoepitope expression instead of single fragment determinations. The concentration of native C9 was lower in synovial fluid compared with plasma, consistant with the excessive local complement activation. Increased CIC levels which correlated with the degree of complement activation were also found in the SF from the RA patients.

We have investigated interactions between human blood platelets and the vitronectin-containing fluid-phase terminal complement complex, the SC5b-9, which is a non-cytolytic variant of the membrane attack complex C%9(m). SC5b-9 affinity for the platelet membrane glycoprotein IIb/IIIa (GPIIb/IIIa) complex was demonstrated by crossed radio-immunoelectrophoresis of solubilized, washed platelets followed by incubation of the immunoplates with (125)I-labelled SC5b-9 and exposure to X-ray films. Platelet adhesion to surfaces of polystyrene coated with the SC5b-9 complex was studied under static conditions in an enzyme immunoassay (EIA). Thrombin-stimulated platelets but not non-stimulated platelets adhered to the SCSb-9coated surface, and platelet adherence was inhibited in a dose-dependent manner by the tetrapeptide RGDS (Arg-Gly-Asp-Ser). This indicates an interaction between platelet GPIIb/IIIa and an RGD sequence in SC5b-9, possibly in its vitronectin moiety. The effect of the SC5b-9 complex on platelet aggregation was examined in a dual-channel aggregometer. Here the SC5b-9 complex inhibited both ADP-and thrombin-induced platelet aggregation in a dose-dependent manner. These results were confirmed by electron microscopical examination of the contents of the aggregometer cuvettes. Platelets which had been thrombin-stimulated in the presence of SC5b-9 appeared activated and had undergone secretion, but showed no aggregation. By contrast, platelets from the control experiments which had been thrombin-stimulated in the absence of SC5b-9 were aggregated. To the best of our knowledge, this is the 6rst report on a biological role of the SC5b-9 complex in platelet function.