Exosomal hsa-miR-151a-3p and hsa-miR-877-5p are potential novel biomarkers for predicting bone metastasis in lung cancer

Abstract

Exosomal miRNAs (exo-miRNAs) have arisen as novel diagnostic biomarkers for various cancers. However, few reports on exo-miRNAs related to bone metastasis (BM) in lung cancer exist. This study aims to screen out key exo-miRNAs and estimate their prognostic values for predicting BM in lung cancer. The differentially expressed exo-miRNAs between the highly-metastatic (95D) and lowly-metastatic (A549) human lung cancer cell lines were comprehensively analyzed using high-throughput sequencing followed by bioinformatic analyses. 29 candidate exo-miRNAs were identified, and 101 BM-related target genes were predicted. Enrichment analysis revealed that these target genes were mainly involved in regulating transcription and pathways in cancer. An exosomal miRNA-mRNA regulatory network consisting of 7 key miRNAs and 10 hub genes was constructed. Further function analysis indicated that these 10 hub genes were mainly enriched in regulating cancer's apoptosis and central carbon metabolism. The survival analysis indicated that 7 of 10 hub genes were closely related to prognosis. Mutation analysis showed that lung cancer patients presented certain genetic alterations in the 7 real hub genes. GSEA for a single hub gene suggested that 6 of 7 real hub genes had close associations with lung cancer development. Finally, ROC analysis revealed that hsa-miR-151a-3p and hsa-miR-877-5p provided high diagnostic accuracy in discriminating patients with bone metastasis (BM+) from patients without bone metastasis (BM-). These findings provided a comprehensive analysis of exo-miRNAs and target genes in the regulatory network of BM in lung cancer. In particular, hsa-miR-151a-3p and hsa-miR-877-5p may be novel biomarkers for predicting BM in lung cancer.

Keywords: biomarker; bone metastasis; exosome; lung cancer; miRNAs.

Figure 1

Analysis of precipitated exosomes isolated from supernatants of cultured tumor cells. (A) The morphology of exosomes was observed by TEM. (B) The exosomes were measured by NTA. (C) CD81, TSG101, and HSP70 were determined by western blot. Abbreviations: TEM: transmission electron microscopy; NTA: nanoparticle tracking analysis.

Figure 2

Analysis of the DE-exo-miRNAs. (A) Histogram showing the upregulated and downregulated DE-exo-miRNAs between the A549 and 95D cells. Red represented upregulated DE-exo-miRNAs; blue represented downregulated DE-exo-miRNAs. (B) The volcano map shows the distribution of DE-exo-miRNAs between A549 and 95D cells according to their P values and fold changes. Red dots represented upregulated DE-exo-miRNAs; blue dots represented downregulated DE-exo-miRNAs; black dots represented non-differentially-expressed miRNAs. (C) Heatmap showing DE-exo-miRNAs from A549 and 95D cells. The colors in the heatmap represented normalized gene expression values, with high expression values being colored in red and low expression values being colored in green. (D) The Venn diagram representing the 29 overlapping exo-miRNAs. Abbreviations: DE-exo-miRNAs: differentially expressed exosomal miRNAs; BM+: patients with bone metastasis; BM-: patients without bone metastasis.

Figure 3

GO and KEGG pathway enrichment analysis of the target genes of 29 exo-miRNAs using the DAVID database. Top 10 significant terms of GO BP (A), CC (B), MF (C), and KEGG pathway (D) enrichment analysis of the target genes. Abbreviations: exo-miRNAs: exosomal miRNAs; BP: biological process; CC: cellular component; MF: molecular function.

Figure 4

PPI network, modular analysis, and miRNA-mRNA network construction. The PPI network of 101 target genes of the 29 exo-miRNAs was analyzed using the STRING database (A) and Cytoscape software (B). (C) The key module was identified from the PPI network using the MCODE plug-in of Cytoscape. (D) The hub genes (degree: top 10) identified by the Cytohubba plug-in. (E) miRNA-hub gene network. In the miRNA-hub mRNA network, the pink ellipses represented hub genes, and the green triangles represented key miRNAs.

Figure 5

GO and KEGG analysis of hub genes using the WebGestalt database. Top 10 significant terms of GO BP (A), CC (B), MF (C), and KEGG pathway (D) enrichment analysis. Abbreviations: BP: biological process; CC: cellular component; MF: molecular function.

Figure 6

Validation of the mRNA expression level of 10 hub genes in lung cancer using the Sangerbox website database. (A) ESR1. (B) HIF1A. (C) DDX3X. (D) MAPK10. (E) KRAS. (F) KIT. (G) FXR1. (H) BCL2L11. (I) FOXO1. (J) MCL1. ^*p < 0.05; ^**p < 0.01; ^***p < 0.0001.

Figure 7

Validation of the protein expression level of screened 10 hub genes in lung cancer samples according to the IHC images in The Human Protein Atlas database. (A) ESR1. (B) HIF1A. (C) DDX3X. (D) MAPK10. (E) KRAS. (F) KIT. (G) FXR1. (H) BCL2L11. (I) FOXO1. (J) MCL1. Abbreviation: IHC: Immunohistochemistry.

Figure 8

Results for the OS analysis of the hub mRNAs in lung cancer patients based on the Kaplan-Meier plotter online database. (A) LUAD. (B) LUSC. Abbreviations: OS: overall survival; LUAD: Lung adenocarcinoma; LUSC: Lung squamous cell carcinoma.

Figure 9

Genetic alterations linked to 7 real hub genes in lung cancer in cBioPortal online database. (A) Seven hub gene alterations in LUAD (TCGA, PanCancer Atlas) and LUSC (TCGA, PanCancer Atlas). (B) Alteration frequencies of seven hub genes based on the LUAD (TCGA, PanCancer Atlas) and LUSC (TCGA, PanCancer Atlas). Kaplan-Meier plots showing OS (C) and DFS (D) in cases with and without hub gene alterations. Abbreviations: LUAD: Lung adenocarcinoma; LUSC: Lung squamous cell carcinoma; OS: overall survival; DFS: disease-free survival.

Figure 10

Significant genes related to six real hub genes and hallmark pathways in lung cancers were obtained by GSEA based on the Sangerbox website database using GSE175601. Top five gene sets according to a GSEA enrichment score for FXR1 (A), ESR1 (B), KIT (C), MCL1 (D), BCL2L11 (E), and HIF1A (F). Abbreviation: GSEA: Gene set enrichment analysis.

Figure 11

ROC curve analysis of the six key exo-miRNAs. (A) hsa-miR-222-3p. (B) hsa-miR-221-3p. (C) hsa-miR-199ª-5p. (D) hsa-miR-223-3p. (E) hsa-miR-151ª-3p. (F) hsa-miR-877-5p. The ROC curves to discriminate the BM- group from healthy controls (H) in the validation set are marked by red lines; the ROC curves to differentiate the BM+ group from the BM- group in the validation set are marked by blue lines; the ROC curves to differentiate the BM+ group from H in the validation set are marked by green lines. Abbreviations: ROC: receiver operator characteristic curve; BM+: patients with bone metastasis; BM−: patients without bone metastasis.