Exploring the mechanism of rosmarinic acid in the treatment of lung adenocarcinoma based on bioinformatics methods and experimental validation

Abstract

Objective: Rosmarinic acid (RosA) is a natural polyphenol compound that has been shown to be effective in the treatment of inflammatory disease and a variety of malignant tumors. However, its specific mechanism for the treatment of lung adenocarcinoma (LUAD) has not been fully elucidated. Therefore, this study aims to clarify the mechanism of RosA in the treatment of LUAD by integrating bioinformatics, network pharmacology and in vivo experiments, and to explore the potential of the active ingredients of traditional Chinese medicine in treating LUAD.

Methods: Firstly, the network pharmacology was used to screen the RosA targets, and LUAD-related differential expressed genes (DEGs) were acquired from the GEO database. The intersection of LUAD regulated by RosA (RDEGs) was obtained through the Venn diagram. Secondly, GO and KEGG enrichment analysis of RDEGs were performed, and protein-protein interaction networks (PPIs) were constructed to identify and visualize hub RDEGs. Then, molecular docking between hub RDEGs and RosA was performed, and further evaluation was carried out by using bioinformatics for the predictive value of the hub RDEGs. Finally, the mechanism of RosA in the treatment of LUAD was verified by establishing a xenograft model of NSCLC in nude mouse.

Results: Bioinformatics and other analysis showed that, compared with the control group, the expressions of MMP-1, MMP-9, IGFBP3 and PLAU in LUAD tissues were significantly up-regulated, and the expressions of PPARG and FABP4 were significantly down-regulated, and these hub RDEGs had potential predictive value for LUAD. In vivo experimental results showed that RosA could inhibit the growth of transplanted tumors in nude mice bearing tumors of lung cancer cells, reduce the positive expression of Ki67 in lung tumor tissue, and hinder the proliferation of lung tumor cells. Upregulated expression of PPARG and FABP4 by activating the PPAR signaling pathway increases the level of ROS in lung tumor tissues and promotes apoptosis of lung tumor cells. In addition, RosA can also reduce the expression of MMP-9 and IGFBP3, inhibit the migration and invasion of lung tumor tissue cells.

Conclusions: This study demonstrated that RosA could induce apoptosis by regulating the PPAR signaling pathway and the expression of MMP-9, inhibit the proliferation, migration and invasion of lung cancer cells, thereby exerting anti-LUAD effects. This study provides new insight into the potential mechanism of RosA in treating LUAD and provides a new therapeutic avenue for treatment of LUAD.

Keywords: Apoptosis; Bioinformatics; Lung adenocarcinoma; MMP-9; PPAR signaling pathway; Rosmarinic acid.

Fig. 1

Network pharmacology and bioinformatics analysis A, B Volcano plots of DEGs in GSE140797 and GSE116959 (LUAD group vs. adjacent non-neoplastic group), where red represents up-regulated genes and blue represents down-regulated genes. Heatmap of DEGs in GSE140797 and GSE116959 (blue area in the first row corresponds to adjacent non-neoplastic group, and red area corresponds to LUAD group), where red represents up-regulated genes and blue represents down-regulated genes. C Venn diagram show the overlap between DEGs and RosA-related genes in GSE140797 and GSE116959, with a total of 26 genes RDEGs that intersect with LUAD. D GO and KEGG enrichment analysis of RDEGs. E PPI network of RDEGs based on the STRING database and the top 6 hub RDEGs identified by the MMC method were MMP-9, MMP-1, PPARG, IGFBP3, PLAU, FABP4 (The darker the color, the higher the MCC score). F Box plots show the mRNA expression of 6 hub RDEGs in GSE140797 and GSE116959. The p-value was calculated using the Wilcoxon rank-sum test, and the differences between groups were statistically significant

Fig. 2

Molecular docking map of RosA and hub RDEGs A rosmarinic acid docked with MMP-9. B rosmarinic acid docked with MMP-1. C rosmarinic acid docked with PPARG. D rosmarinic acid docked with IGFBP3. E rosmarinic acid docked with PLAU. E rosmarinic acid docked with PLAU. F rosmarinic acid docked with FABP4

Fig. 3

Clinical significance of hub RDEGs. A Expression of 6 hub RDEGs were validated in the GSE27262 dataset. The p-value was calculated using the Wilcoxo rank sum test, and the difference between groups was statistically significant. B Expression of 6 hub RDEGs based on the TCGA database. The p-value was calculated using the paired-samples T-test, and the difference between groups was statistically significant. C ROC analysis curve for hub RDEGs. The closer the AUC is to 1, the better the diagnostic effect of this variable in predicting outcomes

Fig. 4

RosA inhibits tumor growth in nude mice bearing A549 lung cancer. A Images of tumor specimens from different treatment groups. B Changes in tumor volume in nude mice during RosA intervention. C Changes in body weight of nude mice during RosA intervention. D Pathological examination of HE staining of lung tumor tissues (200 ×). E Immunohistochemical analysis of Ki67 expression in lung tumor tissues (200 ×). F Statistical data on the percentage of Ki67-positive cells. All experiments were analyzed using 3 independent analyses, and data were expressed as mean ± SD. (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 when compared with control group)

Fig. 5

(A) ROS levels in lung tumor tissues were detected by fluorescent staining (200 ×). (B) Statistical data of the mean fluorescence intensity of ROS. (C) Apoptosis in lung tumor tissues were detected by TUNEL assay (200 ×). (D) Statistical data of the mean fluorescence intensity of TUNEL. All experiments were analyzed using 3 independent analyses, and data were expressed as mean ± SD. (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 when compared with control group)

Fig. 6

mRNA and protein expression levels of hub RDEGs after different doses of RosA intervention. A RT-qPCR to detect the mRNA level of hub RDEGs. B Western blotting to detect protein levels of hub RDEGs. C Relative protein expression. All experiments were analyzed using 3 independent analyses, and data were expressed as mean ± SD. (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 when compared with control group)

Fig. 7

Molecular mechanism of RosA inhibiting LUAD in nude mice bearing A549 lung cancer