No abstract

A B S T R A C T A dynamic estimation of the involvement of the complement system in various diseases was obtained by the direct quantitation of breakdown products of C3 and of properdin factor B. The methods used were based, first, on the separation of native and fragmented molecules according to their molecular size through a precipitation with polyethylene glycol and, secondly, on an immunochemical quantitation, using specific antisera for the major antigens of C3 and factor B. The sensitivity and the specificity of these methods were demonstrated by activation of complement in vitro with generation of C3 and factor B fragments.A clinical investigation was carried out in 41 patients with systemic lupus erythematosus (SLE), 31 with membranoproliferative glomerulonephritis (MPGN), 26 with other types of glomerulonephritis, and 6 with severe alcoholic cirrhosis of the liver. The following observations were made: (a) an elevated plasma level of C3d fragment of C3 was found in 68% of SLE patients, in 87% of MPGN patients, in 62% of patients with other hypocomplementemic nephritis, and in 15% of those with normocomplementemic nephritis, but in only 33% of patients with liver cirrhosis and very low levels of C3; (b) a significant difference was observed between the levels of C3 obtained with either anti-"native" C3 or anti-C3c sera for immunochemical quantitation, in patients with SLE or MPGN, indicating the presence of "altered" or fragmented C3 in plasma; (c) an elevated plasma level of Ba fragment of properdin factor B was found in 46% of SLE patients, in 67% of MPGN patients, in 50% of patients with otherThis work was presented in part at the European Complement Workshop, Heidelberg, May 1974.Dr. Lambert's address is: Centre de Transfusion, HopitalCantonal, 1211 Geneva 4, Switzerland. Received for publication 9 December 1974 and in revised form 18 March 1975. hypocomplementemic nephritis, and in 9% of patients with normocomplementemic nephritis, while the level of properdin factor B was only slightly decreased in these diseases; (d) in SLE and MPGN there was an inverse correlation between the levels of C3d and Ba and the level of C3 in plasma. The level of these fragments was directly correlated with the clinical manifestations of SLE; (e) some patients with a normal C3 level exhibited an elevated plasma concentration of C3 and factor B fragments, suggesting the coexistence of an increased synthesis with a hypercatabolism of complement components.

In vitro, collagen and collagen-like material in GBM, were demonstrated to have a particular high affinity for any DNA tested (mammalian, bacterial, viral, and plant). GBM fixed DNA 40-80 times more than HGG and BSA and 10-40 times more than bacterial LPS. GBM has a higher affinity for SSDNA than for DSDNA. This binding was inhibited at low pH, low ionic strength, and in the presence of anionic detergents, indicating that the highly negatively charged DNA may interact with the basic site on collagen or GBM by electrostatic forces. This interaction was competitively interfered with by DNA-binding proteins such as Clq. Complexes formed of DNA and anti-DNA antibodies did not exhibit the same binding property as free DNA. However, DNA which was already bound to GBM or to collagen could very efficiently bind anti-DNA antibodies and form immune complexes which would remain on these structures. The biological significance of the binding of DNA to GBM or to collagen should be particularly considered in relation to the pathogenesis of SLE. It is possible that DNA released from disrupted or degenerating cells would bind to surrounding collagen fibers or to basement membranes and then act as an immunoabsorbant for circulating anti-DNA antibodies. Some evidence for an in vivo binding of SSDNA to renal structures was obtained in mice treated with bacterial LPS 2 days before the injection of SSDNA

After injection of lipopolysaccharides (LPS) in mice, there is first a release of DNA into plasma and secondly an induction of anti-DNA antibodies. The circulating DNA was purified from plasma and physico-immunochemically characterized. This DNA has a similar density to mammalian cellular DNA,is 4--6S insize, and probably represents a mixture of single-stranded DNA (SSDNA) and double-stranded DNA (DSDNA) or DSDNA with some single-stranded regions. This purified DNA was shown to react with anti-DNA antibodies which appeared as early as 3 days after a single injection of LPS in mice. In serum, DNA-anti-DNA complexes were not detected, although unidentified circulating immune complex-like material was demonstrated 5-8 days after the injection of LPS. In tissues, particularly in renal glomeruli, fine granular immune complex-type immunoglobulin deposits appeared along the glomerular capillary walls and in the mesangium 3 days after the injection of LPS. There is a direct correlation between the level of anti-DNA antibodies and the intensity of glomerular deposits and about 40% of immunoglobulins eluted from kidneys are anti-DNA antibodies, indicating that some of the immune complexes localized in kidneys are DNA-anti-DNA complexes. Based on these observations, the following hypothetical mechanism for the glomerular localization of DNA-anti-DNA complexes after the injection of LPS in mice is proposed. First, DNA, which has been released in circulating blood after injection of LPS, might bind to renal glomeruli, probably on glomerular basement membranes (GBM) through a high affinity of GBM for DNA; secondly, circulating anti-DNA antibodies, which appear later, might react with the glomerular-bound DNA and form immune complexes independently of circulating immune complexes. However, the possibility of direct deposition of immune complexes is not ruled out.

No abstract