Generation of three different fragments of bound C3 with purified factor I or serum. I. Requirements for factor H vs CR1 cofactor activity.

Abstract

Human C3b that was either disulfide bonded to activated Thiol-Sepharose (ATS-C3b) or fixed to sheep erythrocytes (EC3b) was treated with purified factors H and I (β1H and C3b inactivator), heat-inactivated serum, or heat-inactivated serum supplemented with purified erythrocyte complement receptor type one (CR1, C3b-C4b receptor), and then the resulting C3 gragments generated were analyzed. The C3 fragments produced with heat-inactivated serum appeared to be identical to those generated with purified H and I. Treatment of ATS-C3b in low ionic strength buffer with H and limiting I generated iC3b1 and iC3b2 in which the α-chain of C3b was cleaved at either one or two sites, respectively. iC3b1 consisted of three chains of 75 kilodaltons (Kd), 68 Kd, and 46 Kd, whereas iC3b2 consisted of three chains of 75 Kd, 68 Kd, and 43 Kd, with the 43 Kd chain apparently derived from the 46 Kd chain of iC3b1. The iC3b2 fragment had the same structure as has usually been described for iC3b (or C3bi). As the ATS-C3b was treated with H and increasing I in low ionic strength buffer, 10% to 90% of the bound iC3b2 was cleaved at a site in the 68 Kd α-chain fragment, producing a bound 41 Kd α-chain fragment, and releasing the 145 Kd C3c fragment that consisted of the intact β-chain (75 Kd) disulfide bonded to two α-chain fragments of 43 Kd and 27 Kd. This cleavage of bound iC3b by I required H and low ionic strength buffer, and was not observed with H in isotonic salt buffer. The bound 41 Kd fragement generated with H at low ionic strength was not degraded further by I. Treatment of the 41 Kd fragment with trypsin or plasmin cleaved it into C3d (31 Kd) and a fragment(s) of estimated 10 Kd (herein designated C3g) that was not detected by gel electrophoresis. Because the bound 41 Kd fragment consisted of C3d and C3g, it was designated C3d-g. Treatment of fluid-phase C3b with H and I in low ionic strength buffer resulted in generation of iC3b and also some further breakdown of iC3b into C3c and C3d-g. By contrast, when EC3b were incubated in isotonic heat-inactivated serum for 4 hr at 37°C, little or none of the bound iC3b generated was converted to C3d-g. On the other hand, when EC3b were treated with this same isotonic serum supplemented with 10 μg/ml of purified CR1 for 1 hr at 37°C, 50% of the bound C3b was cleaved into bound C3d-g and fluid-phase C3c. A low ionic strength (4 mS) serum reagent was also examined for its ability to convert EC3b into EC3d-g. Low ionic strength serum cleaved 80% of the bound C3b into C3d-g after 30 min at 37°C, and cleaved more than 95% of the bound C3b into C3d-g after 2 hr at 37°C. Thus, purified H or serum H served as a cofactor for iC3b cleavage by factor I only at low ionic strength, whereas purified CR1 had this same cofactor activity at isotonic salt concentrations. These findings suggested that intact erythrocyte CR1 rather than H, might be the normal cofactor for factor I proteolysis of bound iC3b in whole blood. Because the conversion of EC3d-g to EC3d was not detected with either CR1-supplemented serum or hypotonic-serum, bound C3d-g and not C3d, is probably the final breakdown product of bound C3b in vivo.