Autoantigen-specific T Cell and Antibody Reactivity to a Human Gut Commensal in Antiphospholipid Syndrome

William Ruff¹, Carina Dehner¹, Alex Roth¹, Silvio M. Vieira¹, Cassyanne L. Aguiar², Andrew Goodman³, Doruk Erkan², Martin A. Kriegel^1,4

¹Department of Immunobiology, Yale School of Medicine, New Haven, CT

²Department of Medicine, Hospital for Special Surgery, New York, NY

 ³Microbial Sciences Institute, Department of Microbial Pathogenesis, Yale School of Medicine, New Haven, CT

⁴Department of Medicine, Section of Rheumatology, Yale School of Medicine, New Haven, CT

Background/Purpose: Antiphospholipid syndrome (APS) is a serious autoimmune clotting disorder of unknown etiology but with a well-defined major autoantigen, β₂-glycoprotein I (β₂GPI). Infectious triggers have been implicated in transient autoantibody production, but the persistent stimuli for anti-β₂GPI antibodies remain unknown. Given the vast antigenic potential of the gut microbiota, we hypothesize that human gut commensal bacteria induce and sustain autoreactivity via cross-reactivity. To this end, we characterized APS β₂GPI-specific T cell and autoantibody reactivity to in silico candidates.

Methods: NCBI BLAST was used to identify microbial protein sequences with high homology to major β₂GPI epitopes. Sorting of IgA-coated fecal microbiota followed by 16S rDNA sequencing (IgA-Seq) was performed on APS and control microbiomes. Human-derived candidate and control strains were cultured anaerobically. Blood and stool samples were collected from APS patients and controls. A novel strain-specific real-time PCR strategy was developed and validated using isolated strains and defined fecal microbiomes. Memory CD4⁺ T cells specific to β₂GPI were cloned using a T cell library assay and a monoclonal antibody specific to the RGGMR domain I epitope of β₂GPI was expressed for cross-reactivity studies.

Results: Systematic in silico searches revealed Roseburia intestinalis as a major candidate with high homology to the main B and T cell epitopes of β₂GPI. R. intestinalis colonization load was abundant and persistent throughout the study population. IgA-Seq showed IgA coating of the genus Roseburia. APS PBMC proliferated significantly more to protein extracts from R. intestinalis versus control subjects (p=0.0002), and also compared to the phylogenetically related, but mimic-deficient gut commensal Eubacterium rectale (p=0.02). APS memory CD4+ T cells specific for a T cell epitope in domain V cross-react with R. intestinalis mimic peptide. Further, an APS patient-derived anti-domain I β₂GPI monoclonal antibody bound significantly to R. intestinalis lysates whereas a control antibody against an unrelated β₂GPI epitope did not.

Conclusion: We have identified a common gut commensal, R. intestinalis, as a potential chronic cross-reactive trigger in APS. Roseburia is IgA-coated in APS microbiomes suggesting adaptive immune responses to the candidate in vivo. In vitro studies support cross-reactivity of β₂GPI-specific CD4 memory T cells and a pathogenic autoantibody with R. intestinalis. This study provides a concept for common human gut commensals as chronic cross-reactive triggers in genetically susceptible autoimmune hosts such as APS patients.