Deciphering the role of non-coding RNAs involved in sorafenib resistance

Abstract

Sorafenib is an important treatment strategy for advanced hepatocellular carcinoma (HCC). Unfortunately, drug resistance has become a major obstacle in sorafenib application. In this study, whole transcriptome sequencing (WTS) was conducted to compare the paired differences between non-coding RNAs (ncRNAs), including long non-coding RNAs (lncRNAs), circular RNAs (circRNAs), microRNAs (miRNAs), and mRNAs, in sorafenib-resistant and parental cells. The overlap of differentially expressed ncRNAs (DENs) between the SMMC7721/S and Huh7/S cells and their parental cells was determined. 2 upregulated and 3 downregulated lncRNAs, 2 upregulated and 1 downregulated circRNAs, as well as 10 upregulated and 2 downregulated miRNAs, in both SMMC7721/S and Huh7/S cells, attracted more attention. The target genes of these DENs were then identified as the overlaps between the differentially expressed mRNAs achieved using the WTS analysis and the predicted genes of DENs obtained using the "co-localization" or "co-expression," miRanda, and RNAhybrid analysis. Consequently, the potential regulatory network between overlapping DENs and their target genes in both SMMC7721/S and Huh7/S cells was explored. The "lncRNA-miRNA-mRNA" and "circRNA-miRNA-mRNA" networks were constructed based on the competitive endogenous RNA (ceRNA) theory using the Cytoscape software. In particular, lncRNA MED17-203-miRNA (miR-193a-5p, miR-197-3p, miR-27a-5p, miR-320b, miR-767-3p, miR-767-5p, miR-92a-3p, let-7c-5p)-mRNA," "circ_0002874-miR-27a-5p-mRNA" and "circ_0078607-miR-320b-mRNA" networks were first introduced in sorafenib-resistant HCC. Furthermore, these networks were most probably connected to the process of metabolic reprogramming, where the activation of the PPAR, HIF-1, Hippo, and TGF-β signaling pathways is governed. Alternatively, the network "circ_0002874-miR-27a-5p-mRNA" was also involved in the regulation of the activation of TGF-β signaling pathways, thus advancing Epithelial-mesenchymal transition (EMT). These findings provide a theoretical basis for exploring the mechanisms underlying sorafenib resistance mediated by metabolic reprogramming and EMT in HCC.

Keywords: Circular RNAs; Hepatocellular carcinoma; Long non-coding RNAs; MicroRNAs; Non-coding RNAs; Sorafenib.

Fig. 1

The determination for DENs in sorafenib-resistant cells, compared to sorafenib-sensitive HCC cells. (A and B) The volcano map for differentially expressed lncRNAs in SMMC7721/S and Huh7/S cells, compared to SMMC7721 and Huh7 cells, respectively. (C and D) The volcano plot image for differentially expressed circRNAs in SMMC7721/S compared to SMMC7721 cells (C), and in Huh7/S compared to Huh7 cells (D). (E and F) The volcano map for differentially expressed miRNAs in SMMC7721/S, compared to SMMC7721 cells (E), and in Huh7/S, compared to Huh7 cells (F). For the volcano plot image, the horizontal axis represents the change in the expression fold (log2FoldChange) of DENs in different experimental groups/samples, while the vertical axis represents the statistical significance of the change in the expression of DENs. The “UP” and “DOWN” represents the upregulated and downregulated DENs, respectively. Padj is the value obtained using Benjamini and Hochberg's method to perform multiple tests. Padj<0.05. Fold Change≥1.0.

Fig. 2

The identification for DENs, which partially play a vital role in the development of sorafenib resistance. (A) The Venn diagram showing the differentially expressed lncRNAs in both SMMC7721/S and Huh7/S cells, compared to their parental cells. (B) Details for these differentially expressed lncRNAs in both SMMC7721/S and Huh7/S cells, compared to SMMC7721 and Huh7 cells, respectively. (C and D) The Venn map (C) and the detail information (D) for the differentially expressed circRNAs in both SMMC7721/S and Huh7/S cells, compared to SMMC7721 and Huh7 cells respectively. (E and F) The Venn map (E) and the details (F) for the differentially expressed miRNAs in both SMMC7721/S and Huh7/S cells. The values in the intersection in the Venn map represent the number of DENs in both SMMC7721/S and Huh7/S cells.

Fig. 3

The construction of an interactive network in sorafenib-resistant HCC cells based on ceRNA theory using the Cytoscape software. (A) A network “lncRNA MED17-203-miRNA-mRNA” based on lncRNA MED17-203, miR-193a-5p, miR-197-3p, miR-27a-5p, miR-320b, miR-767-3p, miR-767-5p, miR-92a-3p, let-7c-5p and indicated mRNAs.(B) An interactive network “circ_0002874-miR-27a-5p-mRNA” depicts that circ_0002874 targeted miR-27a-5p and downstream mRNAs. (C) “Circ_0078607-miR-320b-mRNA” network showing the interactive relationship between circ_0078607, miR-320b, and mRNAs.

Fig. 4

GO functional and KEGG pathway enrichment analysis for the target genes in these networks. (A) and (B) The enrichment map of GO terms (A) and KEGG pathways (B) for indicated mRNAs in the “lncRNA MED17-203-miRNA-mRNA” network. (C) and (D) GO and KEGG bar for these mRNAs of functional (C) and KEGG pathway enrichment (D) in the “circ_0002874-miR-27a-5p-mRNA” network. (E) and (F) The identification of GO function (E) and KEGG pathway enrichment (F) for these mRNAs in the “circ_0078607-miR-miR-320b-mRNA” network. The corrected P-value (Padj) < 0.05 was considered significantly enriched by differentially expressed genes. The 20 most significant GO and KEGG enrichment pathways were drawn as a bar chart.

Fig. 4

GO functional and KEGG pathway enrichment analysis for the target genes in these networks. (A) and (B) The enrichment map of GO terms (A) and KEGG pathways (B) for indicated mRNAs in the “lncRNA MED17-203-miRNA-mRNA” network. (C) and (D) GO and KEGG bar for these mRNAs of functional (C) and KEGG pathway enrichment (D) in the “circ_0002874-miR-27a-5p-mRNA” network. (E) and (F) The identification of GO function (E) and KEGG pathway enrichment (F) for these mRNAs in the “circ_0078607-miR-miR-320b-mRNA” network. The corrected P-value (Padj) < 0.05 was considered significantly enriched by differentially expressed genes. The 20 most significant GO and KEGG enrichment pathways were drawn as a bar chart.

Fig. 4

GO functional and KEGG pathway enrichment analysis for the target genes in these networks. (A) and (B) The enrichment map of GO terms (A) and KEGG pathways (B) for indicated mRNAs in the “lncRNA MED17-203-miRNA-mRNA” network. (C) and (D) GO and KEGG bar for these mRNAs of functional (C) and KEGG pathway enrichment (D) in the “circ_0002874-miR-27a-5p-mRNA” network. (E) and (F) The identification of GO function (E) and KEGG pathway enrichment (F) for these mRNAs in the “circ_0078607-miR-miR-320b-mRNA” network. The corrected P-value (Padj) < 0.05 was considered significantly enriched by differentially expressed genes. The 20 most significant GO and KEGG enrichment pathways were drawn as a bar chart.

figs1

figs2

figs3

figs4

figs5

figs6

figs7

figs8

figs9

figs10

figs11

figs12

figs13

figs14

figs15

figs16

figs17

figs18

figs19

figs20

figs21

figs22

figs23

figs24