In Frontiers in immunology ; h5-index 100.0
BACKGROUND : A lot of studies have revealed that chronic urticaria (CU) is closely linked with COVID-19. However, there is a lack of further study at the gene level. This research is aimed to investigate the molecular mechanism of COVID-19-related CU via bioinformatic ways.
METHODS : The RNA expression profile datasets of CU (GSE72540) and COVID-19 (GSE164805) were used for the training data and GSE57178 for the verification data. After recognizing the shared differently expressed genes (DEGs) of COVID-19 and CU, genes enrichment, WGCNA, PPI network, and immune infiltration analyses were performed. In addition, machine learning LASSO regression was employed to identify key genes from hub genes. Finally, the networks, gene-TF-miRNA-lncRNA, and drug-gene, of key genes were constructed, and RNA expression analysis was utilized for verification.
RESULTS : We recognized 322 shared DEGs, and the functional analyses displayed that they mainly participated in immunomodulation of COVID-19-related CU. 9 hub genes (CD86, FCGR3A, AIF1, CD163, CCL4, TNF, CYBB, MMP9, and CCL3) were explored through the WGCNA and PPI network. Moreover, FCGR3A, TNF, and CCL3 were further identified as key genes via LASSO regression analysis, and the ROC curves confirmed the dependability of their diagnostic value. Furthermore, our results showed that the key genes were significantly associated with the primary infiltration cells of CU and COVID-19, such as mast cells and macrophages M0. In addition, the key gene-TF-miRNA-lncRNA network was constructed, which contained 46 regulation axes. And most lncRNAs of the network were proved to be a significant expression in CU. Finally, the key gene-drug interaction network, including 84 possible therapeutical medicines, was developed, and their protein-protein docking might make this prediction more feasible.
CONCLUSIONS : To sum up, FCGR3A, TNF, and CCL3 might be potential biomarkers for COVID-19-related CU, and the common pathways and related molecules we explored in this study might provide new ideas for further mechanistic research.
Zhang Teng, Feng Hao, Zou Xiaoyan, Peng Shixiong
COVID-19, bioinformatics, biomarker, chronic Urticaria (CU), immunology