The relationship between autoantibodies targeting GPCRs and the renin-angiotensin system associates with COVID-19 severity

Marques, O. C.; Halpert, G.; Schimke, L. F.; Ostrinski, Y.; Zyskind, I.; Lattin, M. T.; Tran, F.; Schreiber, S.; Marques, A. H. C.; Filgueiras, I. S.; Placa, D. R.; Baiocchi, G. C.; Freire, P. P.; Fonseca, D. L. M.; Humrich, J. Y.; Lange, T.; Muller, A.; Giil, L. M.; Grasshoff, H.; Schumann, A.; Hackel, A. M.; Junker, J.; Meyer, C.; Ochs, H. D.; Lavi, Y. B.; Schulze-Forster, K.; Silvergerg, J. I.; Amital, H.; Zimmerman, J.; Heidecke, H.; Rosenberg, A. Z.; Riemekasten, G.; Shoenfeld, Y.

The coronavirus disease 2019 (COVID-19) can evolve to clinical manifestations resembling systemic autoimmune diseases, with the presence of autoantibodies that are still poorly characterized. To address this issue, we performed a cross-sectional study of 246 individuals to determine whether autoantibodies targeting G protein-coupled receptors (GPCRs) and renin-angiotensin system (RAS)-related molecules were associated with COVID-19-related clinical outcomes. Moderate and severe patients exhibited the highest autoantibody levels, relative to both healthy controls and patients with mild COVID-19 symptoms. Random Forest, a machine learning model, ranked anti-GPCR autoantibodies targeting downstream molecules in the RAS signaling pathway such as the angiotensin II type 1 and Mas receptor, and the chemokine receptor CXCR3 as the three strongest predictors of severe disease. Moreover, while the autoantibody network signatures were relatively conserved in patients with mild COVID-19 compared to healthy controls, they were disrupted in moderate and most perturbed in severe patients. Our data indicate that the relationship between autoantibodies targeting GPCRs and RAS-related molecules associates with the clinical severity of COVID-19, suggesting novel molecular pathways for therapeutic interventions.