We recently described the isolation from human serum of a serum protein (C4 binding protein) that functions as an essential cofactor for C3b inactivator in the proteolysis of fluid-phase C4b and, to a much lesser extent, C3b.e show here the role of C4 binding protein in the foruntion and function of the classical pathway C3 convertase (C42). C4 binding protein interferes with the assembly of the membrane-bound C3 convertase of the classical pathway and accelerates the decay of C42 in a dose-ependent fashion. Its The classical complement pathway is initiated by the interaction of antigen-antibody complexes with the first component of complement, C1 (1, 2). The bound C1 develops proteolytic activity (CY) (3) that is responsible for the activation of C4 (4) and C2 (5) and the formation of an enzymatic active complex (C4b2a or C42), the C3 convertase (6Lwhich mediates the cleavage of CS into C3a and C3b. The C42 enzyme is labile due to the decay of C2 (C2a) (7), but it can be regenerated on the residual C4 (C4b) by the uptake of additional C2 and its cleavage by CY (7).The orderly progression of the classical pathway is modulated, in part, by this instability of the C42 enzyme (7). In this paper we describe a previously unrecognized control of the classical C3 convertase by a normal serum protein, C4 binding protein (C4-bp) (8). This protein serves as a cofactor for the enzyme C3b inactivator (C3bINA) in the further degradation of fluid-phase C4b (9, 10) and, to a lesser extent, C3b (11, 12).The experiments described here demonstrate that C4-bp acts as a modulator of the C42 enzyme, accelerating the decay of C2a. In addition, they show that C4-bp participates, in conjunction with C3bINA, in the inactivation of the hemolytically active C4b bound to the erythrocyte membrane, thereby preventing the regeneration of the classical pathway convertase. MATERIALS AND METHODSBuffers. Veronal-buffered saline (pH 7.5), 0.15 M, containing 0.1% gelatin (GVB); GVB containing 0.15 mM CaC12 and 0.5 mM MgCl2 (GVB2+); 75 mM dextrose/veronal-buffered saline (DGVB2+); and 40 mM EDTA buffer in GVB (EDTA GVB) were prepared as described (13) prepared from sheep erythrocytes as described and converted to EAC14 by using highly purified C4 at a concentration of 6 ,ug/l X 108 cells (22). The specific activity of C4 was 1.1 X 1010 units/,ug of protein. The C4 deposited on the cellular intermediate represented 10% of the C4 input, as assessed by uptake of 125I-labeled C4. C4, C2, and C3 were measured by hemolytic titrations (23). Rat serum diluted 1:20 in EDTA GVB was used as a source of C3-C9. C5, C6, C7, C8, and C9 used in C3 titrations were purchased from Cordis Laboratories (Miami, FL).Depletion The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

No abstract

Human C4 and mouse Ss proteins show extensive structural homologies. They are antigenically related (1) and are composed of three polypeptide chains of similar molecular weights, linked by disulfide bonds (2, 3). In addition, it is very likely that the major histocompatibility complex of these species contains structural genes for beth proteins (3-5). However, some functional differences between the Ss protein and the C4 hemolytic activity of mouse serum have been reported (6).We recently described a new protein in mouse serum which forms complexes with the Ss protein, and also with C4 of human or guinea pig origin (7, 8). Because of the remarkable specificity of its interaction with Ss (C4) we named this protein Ss-or C4-binding protein, and suggested it ~might be a new complement component. This question could not be appropriately studied in mouse serum, whose complement system is still poorly characterized. Therefore we approached the problem by searching for a C4-binding protein (C4-bp)' analogue in human serum.In this paper we report the isolation and characterization of a human serum protein that differs from all known complement components, which has properties very similar to those of mouse C4-bp. Materials and MethodsMaterials. Dimethyl suberimidate, Aldrich Chemical Co., Inc., Milwaukee, Wis.; agarose and human transferrin, Behring Diagnostics, American Hoechst Corp., Sommerville, N. J.; Biorex 70, N, acrylamide, N,N,N',, ammonium persulfate, sodium dodecyl sulfate (SDS), Bio-Rad Laboratories, Richmond, Calif.; diisopropylfluorophosphate (DFP), Calbiochem, San Diego, Calif.; sucrose, Fisher Scientific Co.,

The activated form of C4 (C4b) 1 is an essential component of two key enzymes of the classical pathway, C3 convertase (C4b2a) and C5 convertase (C4b2a3b). The function of these enzymes is limited by the decay of C2a, which is very labile at 37°C. There is also evidence that another serum enzyme, C3b-inactivator (C3bINA) can inhibit the activity of cell-bound C4b2a, probably by cleaving C4b (1). Some uncertainty exists regarding the activity of C3bINA on C4b. According to Shiraishi and Stroud (2) Mouse (4) and human serum (5) contain a high molecular weight protein with several combining sites for C4b. This protein, named C4-hinding protein (C4-bp), differs from all known complement components and inactivators. Because in mouse and human serum most or all C4b generated by activation of the complement cascade is found in combination with C4-bp, we considered the possibility that C4-bp might be the cofactor which modulates the function and/or assembly of C3 and C5-convertase in the fluid phase.

The first component of complement, C1, was isolated unactivated from human serum by repeated additions of di-isopropyl phosphorofluoridate during isolation. The unactivated subcomponents were also isolated, and evidence is given that the three subcomponents C1q, C1r and C1s account wholly for the activity of component C1 in serum. No evidence could be found for a fourth subcomponent, C1t. The approximate molar proportions of the subcomponents in serum are C1q/C1r/C1s = 1:2:2. Optimum activity by haemolytic assay was found at approximate molar proportions C1q/C1r/C1s of 1:4:4. No activity was found when subcomponents were assayed singly or in pairs, except for subcomponents C1q and C1s, which in molar ratio 1:4 gave 15-20% of the activity of the mixture C1q + C1r + C1s. The proteolytic activity of the isolated subcomponent C1s varied according to the method of activation used. Subcomponents C1q + C1r + C1s and C1q + C1s in the presence of antibody-antigen aggregates were activated and inactivated simultaneously, showing a peak of activity and subsequent loss of activity. Both reactions are probably due to proteolysis, and analysis of the peptide bonds split will be necessary to distinguish these two phenomena.