Analyzing the molecular mechanism of Scutellaria Radix in the treatment of sepsis using RNA sequencing

Abstract

Background: Sepsis is a life-threatening organ dysfunction, which seriously threatens human health. The clinical and experimental results have confirmed that Traditional Chinese medicine (TCM), such as Scutellariae Radix, has anti-inflammatory effects. This provides a new idea for the treatment of sepsis. This study systematically analyzed the mechanism of Scutellariae Radix treatment in sepsis based on network pharmacology, RNA sequencing and molecular docking.

Methods: Gene expression analysis was performed using Bulk RNA sequencing on sepsis patients and healthy volunteers. After quality control of the results, the differentially expressed genes (DEGs) were analyzed. The active ingredients and targets of Scutellariae Radix were identified using The Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP). Gene Ontology (GO) and Protein-Protein Interaction (PPI) analysis were performed for disease-drug intersection targets. With the help of GEO database, Survival analysis and Meta-analysis was performed on the cross-targets to evaluate the prognostic value and screen the core targets. Subsequently, single-cell RNA sequencing was used to determine where the core targets are located within the cell. Finally, in this study, molecular docking experiments were performed to further clarify the interrelationship between the active components of Scutellariae Radix and the corresponding targets.

Results: There were 72 active ingredients of Scutellariae Radix, and 50 common targets of drug and disease. GO and PPI analysis showed that the intersection targets were mainly involved in response to chemical stress, response to oxygen levels, response to drug, regulation of immune system process. Survival analysis showed that PRKCD, EGLN1 and CFLAR were positively correlated with sepsis prognosis. Meta-analysis found that the three genes were highly expressed in sepsis survivor, while lowly in non-survivor. PRKCD was mostly found in Macrophages, while EGLN1 and CFLAR were widely expressed in immune cells. The active ingredient Apigenin regulates CFLAR expression, Baicalein regulates EGLN1 expression, and Wogonin regulates PRKCD expression. Molecular docking studies confrmed that the three active components of astragalus have good binding activities with their corresponding targets.

Conclusions: Apigenin, Baicalein and Wogonin, important active components of Scutellaria Radix, produce anti-sepsis effects by regulating the expression of their targets CFLAR, EGLN1 and PRKCD.

Keywords: Molecular docking; Network pharmacology; RNA sequencing; Scutellariae Radix; Sepsis; Traditional Chinese medicine.

Fig. 1

Flow chart. The peripheral blood of 22 septic patients and 10 healthy volunteers was collected for Bulk RNA sequencing and differential expression analysis. At the same time, the TCMSP database was used to obtain the active ingredients and targets of Scutellariae Radix. Disease-drug intersection targets are used for GO and PPI analysis. Core targets were screened using survival analysis and meta-analysis, then single-cell RNA sequencing was performed on core targets to clarify cell line localization. Subsequently, the potential active ingredients and targets of Scutellariae Radix in the treatment of sepsis were screened. Finally, molecular docking experiments were performed to further verify the interrelationship between the drug active ingredient and the target

Fig. 2

Data quality control and differential RNA screening. (A), PCA analysis showed that the two groups of samples could be clearly distinguished, and there were no outlier samples. (B-C) Density distribution and box plot showed that the data of each sample were homogeneous and comparable. (D), Volcano map shows 2462 up-regulated differentially expressed genes in red, 2046 down-regulated differentially expressed genes in blue, the abscissa is the average expression of genes in sepsis samples, and the ordinate is the average expression of genes in normal samples

Fig. 3

Target screening of scutellariae radix in the treatment of sepsis. (A), Venn diagram shows 4508 differentially expressed genes for sepsis in green, 151 active targets of Scutellariae Radix in pink, and 50 intersecting targets in the middle are defined as potential targets for the treatment of sepsis by Scutellariae Radix. (B), Heat maps based on Bulk RNA sequencing show that PRKCD, CFLAR, EGLN1, IL4R, etc. are highly expressed in sepsis, and AR, MS4A2, ADRB1, FXYD2, etc. are low in sepsis. (C), The ingredient-target network shows that the pink on the right represents 50 intersecting targets, the green ○ on the left represents the 42 active ingredients acting on the target

Fig. 4

Intersection targets analysis. (A), BP results showed that intersection targets were mainly involved in cellular response to chemical stress, response to reactive oxygen species, response to oxygen levels, response to ketone, response to hypoxia, response to oxidative stress, response to metal ion, response to decreased oxygen levels, response to drug and other biological processes. (B), PPI results showed that 50 intersecting targets were closely connected, mainly involved in regulation of immune system process, regulation of cell death, regulation of cell communication, organic substance metabolic process, response to Stimulus and other biological processes

Fig. 5

Survival curve and meta-analysis. (A-C), Based on the public dataset GSE65682, a total of 478 patients with sepsis were included in the analysis. In the survival curve, the abscissa represents the survival days, and the ordinate represents the survival rate. The red curve indicates the high expression group of the related genes, and the green curve indicates the correlation based on the low expression group. Statistical tests were performed using the logrank test. Survival curve showed the 28-day survival rate of septic patients, and it can be seen from the figure that the survival rates of patients with high expression of PRKCD, EGLN1 and CFLAR are higher than those in patients with low expression (P < 0.05). It means that high expression of PRKCD, EGLN1 and CFLAR is a protective factor for sepsis patients. (D-F), Based on the datasets GSE54514, GSE63042 and GSE95233, we performed a meta-analysis of the above three core genes in the normal and sepsis group. The data was tested for heterogeneity (I²), using a random effects model at I²>50%, and a fixed effects model at I² ≤ 50%. For PRKCD, I²=69%, using a random-effects model, standardized mean difference (SMD) = 0.87, 95% confidence interval (CI): 0.39–1.36. For EGLN1, I²=15%, using a fixed effects model, SMD = 0.52, CI: 0.26–0.79. For CFLAR, I² = 27%, using a fixed effects model, SMD = 0.15, CI: -0.11-0.41. It showed that PRKCD, EGLN1 and CFLAR were high expression in the sepsis survival group and low expression in non-survivor group

Fig. 6

Single-cell RNA sequencing. (A), Mixed sample sequencing plot. The different colors represent cell lines containing different potential marker genes that are upregulated relative to the other cell lines. Where 1, 2, 6 and 8 represent T cells, 3 and 5 represent Macrophages, 4 represent NK cells, 7 represent B cells, and 9 represent Platelets. (B), Further visualization of the sequencing results of mixed samples. Different colors represent different cells. Orange represents Macrophages, purple represents T cells, green represents NK cells, blue represents B cells and red represents Platelets. (C). Each number in the ordinate represents a different cell line, with consistency with Fig. 6A. Different colors represent the average expression of the different cells. The diameter of the circle represents the percentage of the cells expressed. (D-F), Cell localization map showed that PRKCD were mainly localized in Macrophage cell lines, EGLN1 and CFLAR were widely localized in immune cells lines

Fig. 7

Molecular docking. (A-C), Binding mode of CFLAR and Apigenin based on molecular docking. CFLAR forms conventional hydrogen bond interactions with LEU412, SER416 of Apigenin, carbon-hydrogen bond interactions with SER419, and Pi-Alkyl interactions with LEU422. Molecular docking results showed that the Affinity values of CFLAR and Apigenin was − 6.7 kcal/mol. (D-F), Binding mode of EGLN1 and Baicalein based on molecular docking. EGLN1 forms conventional hydrogen bond interactions with HIS313, HIS374, ARG383 of Baicalein, Pi-Donor Hydrogen Bond interaction with TYR303, Pi-Pi Stacked interaction with TYR310, and Pi-Alkyl interactions with ALA301, VAL376, ALA385. Molecular docking results showed that the Affinity values of EGLN1 and EGLN1 was − 8.1 kcal/mol. (G-I), Binding mode of PRKCD and Wogonin based on molecular docking. PRKCD forms conventional hydrogen bond interactions with SER240, LYS260, and ASN267 of Wogonin, a Pi-Sigma interaction with THR242, and a Pi-Alkyl interaction with MET239. Molecular docking results showed that the Affinity values of PRKCD and Wogonin was − 6.0 kcal/mol