To clarify the pathogenesis of lupus nephritis, we developed an assay that defines a glomerular binding activity (GBA) in both murine and human lupus sera highly correlated with nephritis. In the current study, we used a cross-adsorption strategy to establish that the GBA in MRL lpr serum binds to the glomerular basement membrane (GBM). We subsequently observed that this binding to the GBM was competitively inhibited by either exogenous DNA or histone, abrogated by pretreatment of the GBM with DNAse, and restored after DNase treatment with DNA/histone in a synergistic fashion. GBM binding was also completely inhibited by pretreatment of GBM with collagenase but not heparatinase. The effect of collagenase was not reversed by the subsequent addition of DNA, but was restored by the sequential re-addition of type IV collagen and DNA. By using purified basement membrane components, we found that MRL lpr serum bound avidly to DNA coated on type I collagen but less well (or not at all) to DNA coated on type IV collagen, laminin, or fibronectin. Histone pretreatment of type IV collagen before DNA addition, however, synergistically enhanced binding in a fashion similar to that seen with native GBM. Thus, the GBA in MRL lpr serum seems to be comprised of autoantibodies that bind to histones and/or DNA that adhere to type IV collagen within the GBM. These data support the planted Ag hypothesis as the principal pathogenic mechanism in lupus nephritis and suggest that multiple autoantibodies may contribute to this disorder.