OBJECTIVE: The BXD2 mouse strain spontaneously develops glomerulonephritis and erosive arthritis. The goal of this study was to identify the antigenic target proteins and epitopes and to unravel the mechanisms by which the related conditions arise in BXD2 mice. METHODS: Individual hybridomas isolated from the spleen of a 10-month-old BXD2 mouse were injected intraperitoneally into nonautoimmune mice for evaluation of pathogenicity of each autoantibody. Autoantigens were immunoprecipitated with the pathogenic autoantibody L3A4. Autoantigens were identified using enzyme-linked immunosorbent assay, Western blotting, 2-dimensional gel electrophoresis, and matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MS) and tandem MS. Antigenic epitopes were determined using a high-throughput epitope mapping method. RESULTS: The production of autoantibodies in BXD2 mice occurred in an orderly progression, with peak levels of autoantibodies to nitrotyrosine (NT)-modified enolase, Ro, alpha-actin, and heat-shock proteins (HSPs) preceding peak levels of antihistone, anti-DNA, and rheumatoid factor. Two monoclonal autoantibodies, L3A4 and T56G10, were identified that could induce immune complexes, renal disease, and/or arthritis. Both L3A4 and T56G10 were polyreactive, and each reacted with separate sets of autoantigens. The antigenic targets of L3A4 consisted of NT-modified enolase, ATP5b, alpha-actin, and Hsp70 family proteins including Hspa5 and Hsp74. The antigenic epitopes of NT-modified enolase and Hspa5 exhibited sequence homology and cross-reactivity, suggesting that epitope spreading may occur through a molecular mimicry mechanism. CONCLUSION: The polyreactivity of autoantibodies that target a novel class of autoantigens may enable these autoantibodies to induce erosive arthritis or glomerulonephritis either by direct pathogenic mechanisms or indirectly via Fc or immune complex deposition.