The role of mesangial complement in the hematuria of experimental IgA nephropathy.

We sought to determine if codeposits of IgG and IgM and glomerular complement, observed in most cases of human IgA nephropathy, might be important for inducing hematuria. All combinations of three binary variables, the protein immunogen, the duration of oral immunization, and the protein used for intravenous challenge, were accommodated by eight groups of BALB/c mice in an active model of IgA nephropathy. Mice drank 0.1% solutions of either of two proteins for either 6 or 14 weeks, and then were challenged intravenously with either the same protein or the alternate protein. After 6 weeks, all mice had significant increases of serum IgA, IgG, and IgM antibody to the oral immunogen. At 14 weeks, IgG and IgM antibodies were reduced, presumably due to the onset of oral tolerance, but IgA titers persisted. Nearly all mice had mesangial deposits of IgA and oral immunogen. However, only mice immunized for 6 weeks and challenged with the same protein had significant IgG and IgM deposits (100%), C3 deposits (76%), and significant microhematuria. To distinguish between the role of IgG/IgM codeposits and C3 in the pathogenesis of the hematuria, we induced passive IgA nephropathy with immune complexes of monoclonal IgA anti-dinitrophenyl antibody, dinitrophenyl-bovine albumin as antigen, and one of two monoclonal IgG antibodies specific for dinitrophenyl; one of the IgGs fixes complement, the other does not. Despite comparable mesangial deposits of IgA, IgG, and antigen, only mice given immune complexes containing the complement-fixing IgG had glomerular C3 and hematuria. Furthermore, when mice depleted of serum complement via cobra venom factor were given immune complexes containing the complement-fixing IgG, no glomerular complement was observed and no hematuria ensued. We conclude that IgG/IgM codeposits in murine IgA nephropathy do not directly cause hematuria but do induce the deposition of complement, which is in turn required for glomerular injury.