Network module analysis and molecular docking-based study on the mechanism of astragali radix against non-small cell lung cancer

Abstract

Background: Most lung cancer patients worldwide (stage IV non-small cell lung cancer, NSCLC) have a poor survival: 25%-30% patients die < 3 months. Yet, of those surviving > 3 months, 10%-15% patients survive (very) long. Astragali radix (AR) is an effective traditional Chinese medicine widely used for non-small cell lung cancer (NSCLC). However, the pharmacological mechanisms of AR on NSCLC remain to be elucidated.

Methods: Ultra Performance Liquid Chromatography system coupled with Q-Orbitrap HRMS (UPLC-Q-Orbitrap HRMS) was performed for the qualitative analysis of AR components. Then, network module analysis and molecular docking-based approach was conducted to explore underlying mechanisms of AR on NSCLC. The target genes of AR were obtained from four databases including TCMSP (Traditional Chinese Medicine Systems Pharmacology) database, ETCM (The Encyclopedia of TCM) database, HERB (A high-throughput experiment- and reference-guided database of TCM) database and BATMAN-TCM (a Bioinformatics Analysis Tool for Molecular mechanism of TCM) database. NSCLC related genes were screened by GEO (Gene Expression Omnibus) database. The STRING database was used for protein interaction network construction (PIN) of AR-NSCLC shared target genes. The critical PIN were further constructed based on the topological properties of network nodes. Afterwards the hub genes and network modules were analyzed, and enrichment analysis were employed by the R package clusterProfiler. The Autodock Vina was utilized for molecular docking, and the Gromacs was utilized for molecular dynamics simulations Furthermore, the survival analysis was performed based on TCGA (The Cancer Genome Atlas) database.

Results: Seventy-seven AR components absorbed in blood were obtained. The critical network was constructed with 1447 nodes and 28,890 edges. Based on topological analysis, 6 hub target genes and 7 functional modules were gained. were obtained including TP53, SRC, UBC, CTNNB1, EP300, and RELA. After module analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis showed that AR may exert therapeutic effects on NSCLC by regulating JAK-STAT signaling pathway, PI3K-AKT signaling pathway, ErbB signaling pathway, as well as NFkB signaling pathway. After the intersection calculation of the hub targets and the proteins participated in the above pathways, TP53, SRC, EP300, and RELA were obtained. These proteins had good docking affinity with astragaloside IV. Furthermore, RELA was associated with poor prognosis of NSCLC patients.

Conclusions: This study could provide chemical component information references for further researches. The potential pharmacological mechanisms of AR on NSCLC were elucidated, promoting the clinical application of AR in treating NSCLC. RELA was selected as a promising candidate biomarker affecting the prognosis of NSCLC patients.

Keywords: Astragali radix; Molecular docking; Network pharmacology; Non-small cell lung cancer; TCGA.

Fig. 1

The flowchart of this study was based on UPLC-Q-Orbitrap HRMS, network pharmacology, molecular docking, and molecular dynamics mechanistic simulations to decipher the potential mechanism of astragali radix for the treatment of NSCLC

Fig. 2

Representative UPLC-Q-Orbitrap HRMS total ion chromatograph of AR in positive mode (A) and negative mode (B), and AR absorbed in blood in positive mode (C) and negative mode (D)

Fig. 3

Identification of the AR-NSCLC shared target genes. A The AR–related genes by taking a union of all the results from 4 databases. B The NSCLC–related genes by taking a union of all the results from 7 gene expression profiles. C The AR-NSCLC shared target genes by taking an intersection of AR target genes and NSCLC–related genes

Fig. 4

Protein–protein interaction network. A The component-targets interaction network. Triangles represent the components in AR. Circles represent the AR-NSCLC shared targets. The larger the degree of the node, the darker the color of the node. B The critical AR-NSCLC PIN and eight significant modules identified from the critical AR-NSCLC PIN via MCODE with a score of > 5. The circle represents module1, module 2, module 3, module 4, module 5, module 6 and module 7 respectively, starting from the red in a clockwise direction

Fig. 5

Topological properties of the network. A The degree distribution of AR-NSCLC PIN; B The degree distribution of critical AR-NSCLC PIN; C The shortest path length distribution of AR-NSCLC PIN; D The shortest path length distribution of critical AR-NSCLC PIN

Fig. 6

KEGG enrichment analysis of each module. Gene ratio refers to the ratio of enriched genes to all target genes. Counts refer to the number of the enriched genes

Fig. 7

Screening of core genes affecting the prognosis of NSCLC patients. A Forest plots based on the results of univariate Cox regression of RELA expression and other clinicopathological factors. B Forest plots are based on the results of multivariate Cox regression of RELA expression and other clinicopathological factors. HR and p-values were displayed. C RELA gene expression distribution in primary tissue and normal tissue of NSCLC tumor tissues. D RELA gene expression distribution in lymphoid tissue and normal tissue of NSCLC tumor tissues. E RELA gene expression distribution in metastatic grade and normal tissue of NSCLC tumor tissues. *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 by Kruskal–Wallis test. F Nomogram predicting the proportion of patients with OS. G Plots depict the calibration of model in terms of agreement between predicted and observed OS. Model performance is shown by the plot, relative to the 45-degree line, which represents perfect prediction

Fig. 8

The three-dimensional structure of the interaction between astragaloside IV and hub target genes

Fig. 9

RELA-Ligand interaction diagrams retrieved from Proteins Plus (Pose View). Directed bonds between protein and ligand are drawn as dashed lines and the interacting protein residues and the ligand are visualized as structure diagrams. Hydrophobic contacts are represented more indirectly through spline sections highlighting the hydrophobic parts of the ligand and the label of the contacting residue. A RELA-astragaloside IV; B RELA-Erlotinib; C RELA-Mobocertinib

Fig. 10

The RSMD simulation results of four complexes

Fig. 11

The signaling pathways that regulated by AR on NSCLC

Fig. 12

Simulation analysis of molecular dynamics of RELA-Astragaloside IV complexes. A The SASA between the RELA and astragaloside IV; B The HBond between astragaloside IV and RELA; C The Gyrate between astragaloside IV and RELA proteins

Fig. 13

Free energy morphology of RELA-Astragaloside IV complex