Upregulation of LINC02154 promotes esophageal cancer progression by enhancing cell cycling and epithelial-mesenchymal transition

Abstract

Long noncoding RNAs (lncRNAs) play crucial roles in the progression of human malignancies; however, their involvement in esophageal cancer (ESCA) remains incompletely understood. In this study, we screened for lncRNAs upregulated in ESCA and identified 12 lncRNAs significantly upregulated in primary ESCA tumors. Among those, elevated LINC02154 expression correlated positively with advanced T stages. LINC02154 knockdown in ESCA cell lines suppressed cell proliferation and migration, while ectopic expression of LINC02154 enhanced colony formation. Depletion of LINC02154 suppressed genes involved in various oncogenic processes, including cell cycling, epithelial-mesenchymal transition (EMT), and metabolism. We also found that LINC02154 promotes EMT and enhances chemoresistance, at least in part, through suppression of miR-200b. Finally, RNA-pulldown and mass spectrometry analysis revealed that LINC02154 interacts with proteins involved in the cornified envelope or desmosome. These findings suggest that LINC02154 exerts oncogenic effects through modulation of multiple oncogenic signaling pathways in ESCA and that LINC02154 is a potential therapeutic target.

Keywords: Cell cycle; ESCA; Esophageal cancer; VIM; lncRNA; miR-200b.

Graphical abstract

Fig. 1

Detection of LINC02154 upregulation in primary ESCA tumors. (A) Venn diagram showing lncRNAs upregulated in primary ESCA tumors in the TCGA-ESCA and GSE130078 datasets. (B) Levels of LINC02154 expression in normal esophageal tissues (n = 11) and primary tumors (n = 160) in TCGA-ESCA dataset. (C) Levels of LINC02154 expression in normal tissues (n = 23) and primary tumors (n = 23) in the GSE130078 dataset. (D) Correlations between levels of LINC02154 expression and those of T-factors (left), N-factors (middle) and M-factors (right) in TCGA-ESCA dataset. (E) Results of GSEA for the indicated gene sets based on genes upregulated in ESCA with high LINC02154 expression. NES, normalized enrichment score; FDR, false discovery rate. (E) Results of GSEA for the indicated ontology gene sets based on genes upregulated in ESCA with high LINC02154 expression. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, NS, not significant.

Fig. 2

Functional analysis of LINC02154 in ESCA cells. (A) qRT-PCR analysis of LINC02154 expression in normal esophageal tissue and ESCA cell lines. (n = 3). (B) qRT-PCR analysis of LINC02154 expression in ESCA cells transfected with a control siRNA or siRNAs targeting LINC02154. (n = 3). (C) Cell viability assays in ESCA cells transfected with the indicated siRNAs. (n = 6). (D) qRT-PCR analysis of LINC02154 expression in ESCA cells infected with a control vector or a LINC02154 expression vector. (n = 3). (E) Colony formation assays using ESCA cells with or without ectopic LINC02154 expression. Summarized results are shown on the right. (n = 3). (F, G) Cell migration (F) and invasion (G) assays with TE-9 cells transfected with the indicated siRNA. Representative results are shown on the left, summarized results are shown on the right. (n = 3). Error bars represent SDs. ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 3

The effects of LINC02154 knockdown on gene expression profiles in ESCA cells. (A) Results of RNA-seq in TE-9 cells transfected with a control siRNA (siCtrl) or a siRNA targeting LINC02154 (siLINC). Shown is a heatmap of genes altered (>2-fold) by LINC02154 knockdown. (B, C) Pathway (B) and gene ontology (GO) analyses (C) of genes downregulated by LINC02154 knockdown. (D) Results of GSEA for indicated gene sets based on the RNA-seq data in (A). (E) qRT-PCR analysis of indicated genes in ESCA cells transfected with the indicated siRNAs. (n = 3). (F) Western blot analysis of cyclin B1 in ESCA cells transfected with the indicated siRNAs. (G) Correlations between expression levels of LINC02154 and those of CCNB1 or VIM in TCGA-ESCA dataset. Error bars represent SDs. ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 4

LINC02154 promotes EMT through suppression of miR-200b. (A) Correlation between levels of miR-200b expression and those of LINC02154 expression in TCGA-ESCA dataset. (B) Putative miR-200b-3p binding sites in the LINC02154 sequence. (C) qRT-PCR analysis of miR-200b in TE-5 cells transfected with a control siRNA or siRNAs targeting LINC02154. (n = 3). (D) Western blot analysis of ZEB2 in TE-5 cells transfected with the indicated siRNAs. (E) Correlation between levels of miR-200b expression and those of VIM in TCGA-ESCA dataset. (F) Correlation between VIM expression and T-factors (left) and clinical stages (right) in TCGA-ESCA dataset. (G) qRT-PCR analysis of VIM in ESCA cells transfected with a miR-mimic control or a miR-200b mimic. (n = 3). (H) qRT-PCR analysis of VIM in ESCA cells transfected with a miR-inhibitor control or a miR-200b inhibitor. (n = 3). (I, J) Results of cell viability assays in ESCA cells infected with the indicated vectors. Cells were treated with the indicated concentrations of cisplatin (H) or 5-FU (I). (n = 6). Error bars represent SDs. ∗∗P < 0.01, ∗∗∗P < 0.001, NS, not significant.

Fig. 5

Identification of proteins that interact with LINC02154 in ESCA cells. (A) qRT-PCR analysis of LINC02154 in the cytoplasmic and nuclear fractions extracted from TE-9 cells. (n = 3). (B) Gene ontology (GO) analysis of proteins identified by mass spectrometry that potentially interact with LINC02154. (C) Interaction network among the proteins that potentially interact with LINC02154. Error bars represent SDs. ∗P < 0.05.