Anti-IFN-Gamma Autoantibodies Disease---From Etiology to Therapy

Background: Anti-IFN- autoantibodies (autoAbs) is a new emerge adult onset immunodeficiency and patients suffered severe, life-threatened nontuberculous mycobacteria (NTM) infection. Recently, large number of patients had been found in Southeast Asia, mainly from Taiwan and Thailand and the prevalence and severity of this disease might highly under-estimated. We have observed high prevalence of anti-IFN- autoAbs in Taiwanese/Han patients. Patients present very unique and narrow manifestations: severe NTM and salmonella infections, normal immunological profile and absence of clinical signs or laboratory evidences of autoimmune diseases. Two-third patients suffered from chronic and severe NTM infection even with anti-mycobacterial antibiotic treatment; development of new treatment of NTM infection is urgent for these patients. In previous studying, we identified HLA-DRB1*16:02/DQB1*05:02 is strong associated with this disease, and this finding showed the genetic factor is involved and also explained the ethnic/population specificity in this disease. Aims: In our preliminary data, we had identified one shared epitope of anti-IFN-γ autoAbs in our patients and this epitope had high similarity with a foreign protein diguanylate cyclase from Clostridium. To compete our preliminary observation and have a compressive conclusion, in this proposal, we plan to achieve four main goals: (1) characterize the epitope of anti-IFN- autoAbs and clone the autoantibodies production B cells; (2) establish a proof of principle therapeutic strategy by using modified recombinant anti-IFN- without neutralizing by autoantibodies; (3) validate the potential mimicry antigen diguanylate cyclase; (4) establish anti-IFN- autoAbs mice model to perform in vivo studying. Expected Results and Significance: We expect to demonstrate epitope of anti-IFN-γ is limited in a small region on the C-terminal of IFN-γ; Anti-IFN-γ autoAbs could cross-react with diguanylate cyclase and vice versa plasma antibodies from mice immunized with similar region on diguanylate cyclase; generate a mutant protein to active the IFN-γ receptor by evading the recognition the anti-IFN-γ autoantibodies and test its function in mice model. These data will support our hypothesis that the autoantibodies to IFN-γ in these patients with particular HLA class II alleles, are generated from the antibodies to diguanylate cyclase which could cross-react with IFN-γ through the molecular mimicry mechanism. Moreover, we can develop new therapeutic strategy based on our finding of epitope specificity and hope to apply on clinic in the feature. This project could help us to prevent, predict, diagnose and even to treat anti-IFN-γ autoAbs disease by knowing the property of autoAbs and the mechanism of anti-IFN-γ autoAbs disease.

Project ID:PC10308-1314
External Project ID:MOST103-2320-B182-012