Molecular dissection of the Goodpasture epitope

Sammanfattning: Goodpasture disease is a prototype autoimmune disease that is characterized by a rapidly progressive glomerulonephritis, with or without lung haemorrhage, associated with autoantibodies against the glomerular basement membrane. The major antigen has previously been shown to be the non-collagenous domain of the a3 chain of type IV collagen, one of the six known a chains of type IV collagen. On average, one percent of the total IgG fraction from the patients is comprised of anti-type IV collagen autoantibodies. Ninety percent of these antibodies are a3(IV)-specific and the remaining ten percent are low-affinity antibodies, showing cross-reactivity with the other a(IV) chains. Furthermore, the epitope specificity seems to be limited, as shown by monoclonal antibody inhibition. To test the hypothesis that only antibodies against certain epitopes are nephrotoxic, clinical and serological data from 77 anti-GBM positive patients were retrieved. The results showed that the anti-a3(IV) titre, and especially the antibodies that can be blocked using a monoclonal antibody, correlated with the outcome, in terms of kidney survival. Thus, the study strongly indicates the pathogenic role of the circulating autoantibodies. Epitope mapping of the anti-GBM antibodies has been done using overlapping synthetic peptides. The epitope specificity of the autoantibodies from one patient with anti-a1(IV) antibodies was revealed, and reactivity in ELISA was successfully blocked using a four-amino-acid-long peptide from the a1(IV) sequence. The same approach did not show any reactivity with the anti-a3(IV) autoantibodies. By using recombinant antigens expressed in a human cell line as chimeric proteins, where the a3(IV) sequence was exchanged for the corresponding sequence from the a1(IV) chain, we could show that only autoantibodies against the N-terminal third of the a3(IV) chain correlated with disease. The reactive sequence was further narrowed down to nine discontinuous amino acid residues. A chimeric protein comprised of a1(IV), but with these nine positions expressed as a3(IV), bound the toxic autoantibodies with approximately the same affinity as native a3(IV). These studies provide evidence that only autoantibodies against a very limited region of the a3(IV) chain carry a toxic potential. Furthermore, epitope spreading is relatively limited, thus indicating a possibility for new forms of therapy, including epitope immunomodulatory treatment. In the future, therapy might be adjusted for each individual patient, on the basis of diagnostic tests, e.g., regarding the fine specificity and titre of the autoantibodies, thereby minimising unnecessary discomfort caused by the side effects of immunosuppressive treatment.