Autoimmune diseases such as ankylosing spondylitis (AS) can be driven by emerging neoantigens that disrupt immune tolerance. Here, we developed a workflow to profile posttranslational modifications involved in neoantigen formation. Using mass spectrometry, we identified a panel of cysteine residues differentially modified by carboxyethylation that required 3-hydroxypropionic acid to generate neoantigens in patients with AS. The lysosomal degradation of integrin αIIb [ITGA2B (CD41)] carboxyethylated at Cys96 (ITGA2B-ceC96) generated carboxyethylated peptides that were presented by HLA-DRB1*04 to stimulate CD4 + T cell responses and induce autoantibody production. Immunization of HLA-DR4 transgenic mice with the ITGA2B-ceC96 peptide promoted colitis and vertebral bone erosion. Thus, metabolite-induced cysteine carboxyethylation can give rise to pathogenic neoantigens that lead to autoreactive CD4 + T cell responses and autoantibody production in autoimmune diseases.
Background: Systemic inflammatory diseases including rheumatoid arthritis (RA), ankylosing spondylitis (AS), IBD and long covid share convergent multi-organ phenotypes. Long covid provides a tractable model for dissecting gut-driven mechanisms of systemic inflammation, given its defined temporal onset and treatment-naïve postinfectious context. Objective: To characterise a gut-driven mechanism of systemic inflammation in long covid and assess its cross-disease correlates in RA, AS and IBD. Design: Comparative metagenomic analyses across RA, AS, IBD and long covid cohorts. Long covid was established as a paradigm for postdysbiotic inflammatory diseases, single-cell RNA sequencing and functional studies in longitudinal human cohorts and co-infection mouse models (SARS-CoV-2 and Pseudomonas aeruginosa ) were employed to dissect cellular and molecular mechanisms. Genetic and pharmacological interventions targeting the interleukin (IL)-15-arachidonic acid (ARA) axis were validated for therapeutic efficacy. Results: Flagellated bacterial expansion defined a shared intestinal signature across all four diseases. Mechanistic studies in long covid demonstrated that flagellated bacteria activated toll-like receptor 5 (TLR5) on neutrophils, triggering the formation of neutrophil extracellular trap (NET) and IL-15 release. IL-15 subsequently stimulated macrophage ARA production. The co-infection murine model recapitulated multi-organ pathophysiology of long Covid, including pulmonary fibrosis and intestinal lymphoid aggregates. Genetic ablation of macrophage ARA synthesis or neutrophil IL-15 attenuated lung pathology, whereas gut microbiome clearance with gentamicin uniquely suppressed systemic inflammation. Conclusions: We delineate a flagellin-TLR5-IL-15-ARA axis as a candidate mechanism driving systemic inflammation in long covid. These findings position intestinal flagellin as a candidate therapeutic target and ARA as a potential biomarker for long covid, warranting prospective validation across inflammatory disease boundaries.