Insights into the Correlation and Immune Crosstalk Between COVID-19 and Sjögren’s Syndrome Keratoconjunctivitis Sicca via Weighted Gene Coexpression Network Analysis and Machine Learning

Yaqi Cheng, Liang Zhao, Huan Yu, Jiayi Lin, Meng Li, Huini Zhang, Haocheng Zhu, Huanhuan Cheng, Qunai Huang, Yingjie Liu, Tao Wang, Shiqi Ling

Background: Although autoimmune complications of COVID-19 have aroused concerns, there is no consensus on its ocular complications. Sjögren’s syndrome is an autoimmune disease accompanied by the ocular abnormality keratoconjunctivitis sicca (SS-KCS), which may be influenced by COVID-19. Thereby, we explored the possible interaction between COVID-19 and SS-KCS, and we aimed to elucidate the potential correlated mechanism. Methods: Differentially expressed genes (DEGs) in COVID-19 and SS-KCS transcriptome data obtained from the gene expression omnibus database were identified, and COVID-19-related genes were screened using weighted gene coexpression network analysis. Common genes were verified using four machine-learning diagnostic predictors. The clinical relationship between the two common hub genes of COVID-19 was analyzed. Finally, the immune cell types infiltrating the microenvironment in the COVID-19 dataset were analyzed using CIBERSORT, and the interrelation between key genes and differentially infiltrating immune cells was verified via Pearson correlation. Results: Ten potential primary hub mRNAs were screened by intersecting the COVID-19 DEGs, SS-KCS DEGs, and WGCNA genes. After a multifaceted evaluation using four mainstream machine-learning diagnostic predictors, the most accurate and sensitive random forest model identified CR1 and TAP2 as the common hub genes of COVID-19 and SS-KCS. Together with the clinical information on COVID-19, the expression of CR1 and TAP2 was significantly correlated with the status and severity of COVID-19. CR1 and TAP2 were significantly positively correlated with M0 and M2 macrophages, neutrophils, and CD4+ memory resting T cells and negatively correlated with activated NK cells, monocytes, and CD8+ T cells. Conclusions: We validated the hub genes associated with both COVID-19 and SS-KCS, and we investigated the immune mechanisms underlying their interaction, which may help in the early prediction, identification, diagnosis, and management of SARS-CoV-2 infection-related SS-KCS syndrome or many other immune-related complications in the long COVID period.