Divergence of V-region sequences from the human proteome as a factor in autoimmune diseases and the immunogenicity of therapeutic antibodies. - Project Summary: Millions of antibody sequences from both healthy individuals and patients with a variety of immune-mediated disorders are now freely available to the research community. This has been made possible by support primarily provided by NIAID for the research projects themselves and for the databases that curate the data. To maximize the benefits of this financial support, the Institute now funds efforts to use the curated data to explore new initiatives which are beyond the scope of the original research projects. We propose to use approximately one million of these antibody sequences to test a hypothesis concerning rare antibody sequences that was first proposed over 35 years ago but which now can be examined on a large scale by means of these NIAID supported databases. The expansion of the antibody repertoire by V(D)J gene rearrangement, N-nucleotide addition, and somatic hypermutation (SHM) introduces non-germline encoded sequence changes throughout the antibody variable domain. Many of these sequences are either totally absent from, or very rare in, the human proteome. It is already clear that these “foreign” sequences are common in V regions of antibodies even from healthy human controls, so it is unlikely that the majority are immuno-pathogenic. However, there is reason to test the hypothesis that a small subset of these forbidden/rare pentapeptides will be clinically immunogenic, i.e., will be able to induce immune responses that should be absent from normal antibodies. If this can be confirmed, it will help us address at least two major challenges now facing medical immunology. First is the persistent problem of immunogenicity in new therapeutic antibodies which substantially raises the cost of development of these agents and limits their efficacy in patients. If the proposed work is successful and the immunogenic pentapeptides are identified, it could lead to the development of new algorithms that more accurately predict the immunogenicity of therapeutic antibodies prior to expensive clinical trials. Second is the need for more specific and effective therapies for autoimmune diseases such as SLE, MS, and type 1 diabetes. A large body of evidence indicates that many of the self-reactive, pathogenic autoantibodies contain idiotopes in their structures which are recognized by anti-idiotypic antibodies. Further, it appears that the presence or absence of these anti- idiotypic antibodies may in part be responsible for periods of remission or active disease. Few idiotope structures are known, but it seems reasonable that these immunogenic segments in antibody variable regions may also contain the foreign pentapeptides; this could serve as a tool for identifying idiotopes in the sequences of pathogenic self-reactive antibodies. A major outcome of the proposed studies is a bioinformatics research tool, the Penta-Scope, which will take as input the sequences of large numbers of antibodies and return the frequencies of their component pentapeptides in human and murine proteomes. This tool will be made freely available online for other investigators to use in studies of antibody structure and/or immunogenicity.
