Network pharmacology and RNA-sequencing reveal the molecular mechanism of Xuebijing injection on COVID-19-induced cardiac dysfunction

Abstract

Background: Coronavirus disease 2019 (COVID-19) is an emerging infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Up to 20%-30% of patients hospitalized with COVID-19 have evidence of cardiac dysfunction. Xuebijing injection is a compound injection containing five traditional Chinese medicine ingredients, which can protect cells from SARS-CoV-2-induced cell death and improve cardiac function. However, the specific protective mechanism of Xuebijing injection on COVID-19-induced cardiac dysfunction remains unclear.

Methods: The therapeutic effect of Xuebijing injection on COVID-19 was validated by the TCM Anti COVID-19 (TCMATCOV) platform. RNA-sequencing (RNA-seq) data from GSE150392 was used to find differentially expressed genes (DEGs) from human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) infected with SARS-CoV-2. Data from GSE151879 was used to verify the expression of Angiotensin I Converting Enzyme 2 (ACE2) and central hub genes in both human embryonic-stem-cell-derived cardiomyocytes (hESC-CMs) and adult human CMs with SARS-CoV-2 infection.

Results: A total of 97 proteins were identified as the therapeutic targets of Xuebijing injection for COVID-19. There were 22 DEGs in SARS-CoV-2 infected hiPSC-CMs overlapped with the 97 therapeutic targets, which might be the therapeutic targets of Xuebijing injection on COVID-19-induced cardiac dysfunction. Based on the bioinformatics analysis, 7 genes (CCL2, CXCL8, FOS, IFNB1, IL-1A, IL-1B, SERPINE1) were identified as central hub genes and enriched in pathways including cytokines, inflammation, cell senescence and oxidative stress. ACE2, the receptor of SARS-CoV-2, and the 7 central hub genes were differentially expressed in at least two kinds of SARS-CoV-2 infected CMs. Besides, FOS and quercetin exhibited the tightest binding by molecular docking analysis.

Conclusion: Our study indicated the underlying protective effect of Xuebijing injection on COVID-19, especially on COVID19-induced cardiac dysfunction, which provided the theoretical basis for exploring the potential protective mechanism of Xuebijing injection on COVID19-induced cardiac dysfunction.

Keywords: COVID-19; Cardiac dysfunction; Molecular docking; Network pharmacology; RNA-sequencing; Xuebijing injection.

Fig. 1

Flow diagram of the analysis process.

Fig. 2

The network of effective compound-target. (A) Venn diagram of COVID-19 related therapeutic target genes from 5 databases. (B) Venn diagram for candidate targets in Xuebijing injection and COVID-19. (C) The effective compound-target network of Xuebijing injection on COVID-19. The two outer rings were composed of 97 therapeutic target proteins. In the inner circle, the effective ingredients in Chi Shao, Chuan Xiong, Dang Gui, Hong Hua, Dan Shen were shown in circle colored yellow, orange, green, blue, pink, respectively. Besides, the purple circle represented effective ingredients in multiple herbs.

Fig. 3

Data analysis of GSE150392. (A) Circos plot of DEGs in GSE150392. The outer track represented each chromosome, the second track was gene names of 22 overlapped genes between DEGs of SARS-CoV-2 infected hiPSC-CMs and the 97 therapeutic targets, the third track was a heatmap of 2029 DEGs (a sample of a control group), the fourth track was a heatmap of 2029 DEGs (a sample of a case group), the fifth track was the bar plot for log2 (foldchange) of 2029 DEGs, the sixth track was the scatter plot for log2 (adjust p value) of 2029 DEGs, the innermost track was the PPI connection of the 22 hub DEGs. (B) Expression heatmap of ACE2 and 22 hub DEGs in GSE150392.

Fig. 4

The GSEA enrichment results for 22 hub DEGs by WebGestalt based on the KEGG gene set. (A) The GSEA enrichment results shown in a bar chart. (B) The GSEA enrichment results exhibited in a volcano plot. (C) The GESA enrichment plots of typical pathways.

Fig. 5

The GSEA enrichment results for 22 hub DEGs by WebGestalt on the grounds of Reactome Pathway Knowledgebase. (A) The GSEA enrichment results shown in a bar chart. (B) The GSEA enrichment results exhibited in a volcano plot. (C) The GESA enrichment plots of typical pathways.

Fig. 6

Functional analysis of 22 hub DEGs. (A) GO functional enrichment analysis of 22 hub DEGs. BP: biological process, CC: cellular component, MF: molecular function. GeneRatio: The ratio of genes enriched in each term. (B) KEGG pathway enrichment analysis of 22 hub DEGs. GeneRatio: The ratio of genes enriched in each pathway.

Fig. 7

PPI network and the correlation among 7 central hub genes. (A) PPI network of the 22 hub DEGs. The up-regulated genes were shown in red, while the down-regulated ones were shown in blue. The deeper color of the genes indicated increased log2 (foldchange), and the larger diameter stood for a more significant statistical difference. (B) PPI network of the 7 central hub genes. (C) Correlation heatmap of the 7 central hub genes. The area of the colored pie and the depth of color were positively associated with the correlation coefficient. The specific correlation coefficients were also marked as numbers in the corresponding sites.

Fig. 8

Heatmap showed the detailed expression patterns of ACE2 and 7 central hub genes in three kinds of SARS-CoV-2 infected CMs (hiPSC-CMs, hESC-CMs, adult human CMs). The borders of DEGs with p value < 0.05 were marked in black. While the borders of DEGs with p value > 0.05 were marked in gray.

Fig. 9

Relationship between 7 central hub genes and their corresponding effective compounds. (A) Chord plot of the 7 central hub genes and their corresponding effective compounds from the components of Xuebijing injection. The outter semicircle showed the components source of corresponding compounds. The inner chord plot showed the interreaction among the 7 central hub genes and the effective compounds. (B) Molecular docking diagram of FOS and quercetin. The green helix represented the molecular structure of FOS. The predicted bond between FOS and quercetin were exhibited in the dashed box.

Fig. 10

Potential cell signal transduction pathways of 7 central hub genes in CMs infected with SARS-CoV-2. SARS-CoV-2 might bind to TLR4, then activated IRF3 and IRF7, leading to increased expression of IFNB1 and IFN-β. IFN-β could stimulate STAT1 and finally resulting in myocardial fibrosis. Besides, the TLR4-induced c-JUN/c-FOS activation could enhance the expression of ACE2 to allow more SARS-CoV-2 to entry into the host cells. At the same time, the activited c-JUN/c-FOS might eventually accelerate the apoptosis of CMs. CCL2 bound to its receptor CCR2 and activated the PKC signaling pathway, inducing the increase of AP-1 (c-JUN/c-FOS), NF-κB and ACE2, and ultimately leading to CM apoptosis and myocardial ischemia. NF-κB signaling pathway could also be activated by IL-1A/IL-1B and subsequently contributed to myocardial ischemia. The combination of CXCL8 and its receptor CXCR1/2 promoted cell senescence, and enhanced cell invasion in conjunction with SERPINE1 through MMP pathway.