Background
Autoantibodies (AAbs), particularly those targeting type 1 interferon subtype alpha (aIFNα), nuclear antigens (ANAs), cardiolipin (aCL), and beta 2 glycoprotein 1 (aβ2GP1), have been associated with the severity of acute COVID-19. Whether SARS-CoV-2 induces these AAbs, however, and whether there are differences across settings, remains underexplored. We investigated AAbs prevalence and induction in individuals from the Johns Hopkins Hospital Emergency Department (JHHED) in Baltimore Maryland and the Rakai Community Cohort Study (RCCS) in Uganda.
Methods
Serum samples were derived from 224 JHHED and from 150 RCCS unique individuals before and after SARS-CoV-2 infection. Pre-infection samples were collected between July 2020 – February 2022, for JHHED and between June 2018 – October 2020 for the RCCS. Post-infection samples were collected between September 2020 – December 2022 for JHHED, and February 2021 – March 2023 for RCCS. aIFNα, aCL, and aβ2GP1 AAbs were assessed using commercially available ELISAs. ANAs were measured using a line immunoassay. McNemar’s tests assessed within-cohort changes, and Fisher’s exact tests compared prevalence across cohorts, with significance set at p<0.05 after Benjamini-Hochberg correction.
Results
For both cohorts, no differences in AAb prevalence were observed before and after SARS-CoV-2 infection. Additionally, individuals were as likely to see the appearance of an AAb as have one resolve. The prevalence of AAbs for the JHHED were: aIFNα (2% vs 2%); ANAs (25% vs 25%); aCL (3% vs 4%,); aβ2GP1 (6% vs 4%). For the RCCS: aIFNα (1% vs 2%); ANAs (45% vs 45%); aCL (1% vs 3%); aβ2GP1 (5% vs 10%). Pre-SARS-CoV-2 infection, there was a significantly higher prevalence of six ANAs in Ugandan individuals compared to their Baltimore counterparts: aPCNA (10% vs 0%, p<0.001); aSS-B/LA (13% vs 2%, p<0.001); aCENPB (4% vs 0%, p=0.01); aU1snRNP (10% vs 1%, p=0.001); aPM-Scl (12% vs 4%, p=0.01); and aKu (22% vs 3%, p<0.001). There was no effect of age or sex on AAb prevalence.
Conclusions
We found no evidence that SARS-CoV-2 induced the examined AAbs in either the US or Ugandan population. Compared to the JHHED, we observed a significantly higher prevalence of several ANAs in the RCCS. Given ANA’s association with various autoimmune disorders, this disparity warrants further investigation.
Table or Figure
