Identification of the feature genes involved in cytokine release syndrome in COVID-19

Abstract

Objective: Screening of feature genes involved in cytokine release syndrome (CRS) from the coronavirus disease 19 (COVID-19).

Methods: The data sets related to COVID-19 were retrieved using Gene Expression Omnibus (GEO) database, the differentially expressed genes (DEGs) related to CRS were analyzed with R software and Venn diagram, and the biological processes and signaling pathways involved in DEGs were analyzed with GO and KEGG enrichment. Core genes were screened using Betweenness and MCC algorithms. GSE164805 and GSE171110 dataset were used to verify the expression level of core genes. Immunoinfiltration analysis was performed by ssGSEA algorithm in the GSVA package. The DrugBank database was used to analyze the feature genes for potential therapeutic drugs.

Results: This study obtained 6950 DEGs, of which 971 corresponded with CRS disease genes (common genes). GO and KEGG enrichment showed that multiple biological processes and signaling pathways associated with common genes were closely related to the inflammatory response. Furthermore, the analysis revealed that transcription factors that regulate these common genes are also involved in inflammatory response. Betweenness and MCC algorithms were used for common gene screening, yielding seven key genes. GSE164805 and GSE171110 dataset validation revealed significant differences between the COVID-19 and normal controls in four core genes (feature genes), namely IL6R, TLR4, TLR2, and IFNG. The upregulated IL6R, TLR4, and TLR2 genes were mainly involved in the Toll-like receptor signaling pathway of the inflammatory pathway, while the downregulated IFNG genes primarily participated in the necroptosis and JAK-STAT signaling pathways. Moreover, immune infiltration analysis indicated that higher expression of these genes was associated with immune cell infiltration that mediates inflammatory response. In addition, potential therapeutic drugs for these four feature genes were identified via the DrugBank database.

Conclusion: IL6R, TLR4, TLR2, and IFNG may be potential pathogenic genes and therapeutic targets for the CRS associated with COVID-19.

Fig 1. Analysis of the COVID-19 DEGs.

(A) The Principal Component Analysis (PCA) of the COVID-19 (n = 10) and HC samples (n = 5). When PCA1 = 34.8% and PCA2 = 20.5%, COVID-19 and HC samples are clearly distinguished. (B) The 6950 DEGs volcano plots. The red dots indicate upregulated genes, the blue dots denote downregulated genes, and the gray dots represent non-DEGs, with FC≥1.0 and P-value<0.05. (C) The heatmaps showing the results of the clustering analysis based on 6950 DEGs.

Fig 2. The common genes and their functional analysis.

(A) The Venn diagram of the 6950 DEGs and 971 CRS genes. A total of 34 DEGs, namely common genes, were obtained. (B) The KEGG analysis of 34 common genes. (C) The GO analysis of 34 common genes.

Fig 3. The network map of the common genes regulated by transcription factors.

The 34 common genes are mainly regulated by 5 transcription factors RELA, YY1, IRF1, ETS2, and EGR1. Red represents the transcription factors, while blue denotes the common genes.

Fig 4. Key gene screening.

(A) PPI network of common gene. There are 26 nodes with 92 edges, and the average degree value is 7.08. The size of the nodes in the figure and the depth of their colors are proportional to their degree value. (B)Venn diagram shows the intersection of the results of two algorithms.

Fig 5. A comparison and validation between the key genes of the COVID-19 and healthy groups.

(A) A histogram of the feature genes in the GSE164805 dataset. * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001. (B) A histogram of the feature genes in the GSE171110 dataset. *indicates p<0.05, ***indicates p<0.001,and ****indicates p<0.0001.

Fig 6. The PPI network and KEGG analysis of the feature genes.

(A) The PPI network diagram of the upregulated genes. (B) The PPI network diagram of the downregulated genes. (C) The KEGG analysis of the upregulated and associated genes. (D) The KEGG analysis of the downregulated and associated genes.

Fig 7. Immune infiltration analysis.

(A) A comparison between the immune cell counts of the COVID-19 and healthy groups. * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001. (B) The effect of higher feature gene expression on the immune cells. * indicates p<0.05, ** indicates p<0.01, and *** indicates p< 0.001.