Myositis-specific autoantibodies recognizing Mi2 also target the autoimmune regulator (AIRE) protein at a shared PHD-zinc finger.

OBJECTIVES: In dermatomyositis patients with anti-Mi2 autoantibodies, autoantibodies can enter muscle cells, leading to the aberrant expression of genes normally repressed by the Mi2/nucleosome remodeling and deacetylation (NuRD) complex. However, the mechanism by which autoantibodies interfere with Mi2/NuRD function remains unclear. This study aimed to identify additional autoantibodies in anti-Mi2-positive patients as well as the specific epitopes recognized by anti-Mi2 and any novel autoantibodies. METHODS: Phage ImmunoPrecipitation Sequencing (PhIP-Seq) was used to screen serum samples from anti-Mi2-positive myositis patients for autoantibodies. Enzyme-linked immunosorbent assays (ELISA) and luciferase immunoprecipitation system (LIPS) immunoassays were used to detect autoantibodies in serum samples from myositis patients and healthy controls. RESULTS: PhIP-Seq identified autoantibodies recognizing the autoimmune regulator (AIRE) in sera from anti-Mi2 autoantibody-positive patients. Both anti-AIRE and anti-Mi2 autoantibodies predominantly recognized a homologous region of the plant homeodomain zinc finger type I (PHD1), which is critical for AIRE and Mi2/NuRD function. ELISA and LIPS testing showed that anti-Mi2 autoantibody-positive patients were positive for anti-AIRE autoantibodies, while AIRE reactivity was largely absent in healthy comparators, anti-Mi2 autoantibody-negative-myositis, and other autoimmune diseases. Affinity-purified anti-Mi2 autoantibodies recognized both Mi2 and AIRE by ELISA, whereas anti-Mi2-depleted immunoglobulin fractions did not recognize either protein. CONCLUSIONS: Autoantibodies recognizing Mi2 also recognize AIRE at a homologous PHD1 finger. This region is required by the Mi2/NuRD complex to anchor the nucleosome and consequently repress gene expression. Our findings suggest that anti-Mi2 autoantibodies disrupt NuRD complex function by binding to the PHD1 domain. Further studies are needed to determine if anti-Mi2 autoantibodies bind other PHD1-containing proteins and their functional implications.