The association between COVID-19 and autoimmune diseases has gained increasing recognition, yet the specific targets of SARS-CoV-2-induced IgG are currently in focus for several studies. This study aims to explore the proteomic targets of these antibodies and their potential role in autoimmunity. We utilised a human proteome microarray encompassing 23 736 unique proteins, including isoform variants and fragments, as catalogued by the Human Protein Atlas. Serum samples were analysed from four groups: healthy controls (N-exp HC), individuals vaccinated with protein-based vaccines (N-Cov Vac) and patients with moderate or severe COVID-19 (COVID-Mod and COVID-Sev). The evaluation of SARS-CoV-2-induced IgG antibodies revealed their potential to recognise multiple human proteins. Key targets included interferon alpha (IFN-α), tumour growth factor beta (TGF-β), interleukin 1 (IL-1), CXCL16, TGF-β receptors, CD34, CD47 and BCL2. The antibodies also targeted proteins from genes overexpressed in various immune cells, such as CD4+ and CD8+ T cells, γδ T cells, B cells, dendritic cells and NK cells. Reactivity was also observed with proteins specifically expressed in multiple organs, including the brain, liver, lungs and heart. Targeting patterns differed between COVID-19 patients and controls, with some proteins showing differential recognition in moderate versus severe cases. Furthermore, we evaluated the protein-protein interaction network (PPIN) of all targeted proteins and observed minimal structural homology and co-expression among the evaluated proteins, with almost no relation to the SARS-CoV-2 immune system reactome. The results suggest that the profile of SARS-CoV-2-induced IgG autoantibodies is associated with disease severity. In contrast, protein-vaccinated individuals exhibited a profile similar to non-exposed controls, suggesting that autoreactive IgG is specifically linked to active SARS-CoV-2 infection. These findings reveal a complex network of SARS-CoV-2-induced IgG idiotypes capable of targeting human proteins, not merely through simple cross-recognition of homologous proteins. This highlights the need for further investigations to determine whether they may influence COVID-19 pathophysiology and its clinical outcomes.
Read full abstract