LncRNA TRERNA1 facilitates hepatocellular carcinoma metastasis by dimethylating H3K9 in the CDH1 promoter region via the recruitment of the EHMT2/SNAI1 complex

Abstract

Objectives: Long non-coding RNAs (LncRNAs) play an important role in hepatocellular carcinoma development, however, as a crucial driver of hepatocellular carcinoma (HCC) metastasis, their functions in tumour metastasis remain largely unknown.

Materials and methods: The lncRNA TRERNA1 expression levels were detected in HCC by quantitative real-time PCR (qPCR). The function of TRERNA1 was examined by wound-healing assays, transwell assays and tail vein injection experiments. The potential regulatory mechanisms of TRERNA1 on its target genes were explored by ChIP, RIP, IP and WB assays.

Results: Elevated TRERNA1 levels promoted HCC cell migration and invasion in vitro and in vivo. TRERNA1 recruited EHMT2 to dimethylate H3K9 in the CDH1 promoter region. Furthermore, EHMT2 bound to SNAI1 to suppress CDH1 expression in HCC cells. After inhibiting TRERNA1, the expression level of CDH1 was restored and was involved in the regulation of the EHMT2/SNAI1 complex. The level of TRERNA1 was positively correlated with tumour metastasis and was negatively correlated with the expression of CDH1 in HCC tissues.

Conclusions: For the first time, the current study reveals that TRERNA1 promotes cell metastasis and the invasion of HCC via the recruitment of EHMT2 and/or the EHMT2/SNAI1 complex to suppress CDH1. These data identify a novel mechanism that regulates TRERNA1 in metastatic HCC and provides a potential targeted therapy for HCC patients.

Keywords: EHMT2; TRERNA1; hepatocellular carcinoma; metastasis.

Figure 1

TRERNA1 promotes the cell migration and invasion of HCC in vitro. A, Cell migration assays were performed using transwell assays in HepG2 and HepG2.215 cells. The average number of cells exhibiting migration from three random microscopic fields is presented in the histogram. B, Cell invasion assays were performed using Matrigel‐coated transwell membranes. The average number of cells exhibiting invasion from three random microscopic fields is presented in the histogram. C and D, Wound‐healing assays in TRERNA1‐overexpressing HepG2 cells and in TRERNA1‐depleted HepG2.215 cells. The scratch was measured at 0, 24 and 48 h. Data are presented as the mean ± SD; n = 3. *P < 0.05, **P < 0.01

Figure 2

TRERNA1 promotes the metastasis of HCC cells in vivo. A and B, Representative livers derived from severe combined immunodeficient (SCID) mice after tail vein injection with HepG2‐TRERNA1 and HepG2.215‐shTRERNA1 were shown. Haematoxylin and eosin‐stained (H&E) images of liver tissues isolated from the mice were shown. Scale bars represent 500 µm (left) and 100 µm (right). Arrows indicate metastasis nodules. C and D, Ectopic TRERNA1 expression promoted the metastasis of HepG2 cells in vivo, and the number of liver metastasis nodules in the mice was determined after 8 weeks (n = 6 per group). E and F, The knockdown of endogenous TRERNA1 by shRNA inhibited the liver metastasis nodules of HepG2.215 cells in nude mice (n = 6 per group). Data are presented as the mean ± SD; **P < 0.01 (Student's t test)

Figure 3

TRERNA1 functions as a scaffold to suppress CDH1 expression by recruiting EHMT2. A, The relative mRNA expression levels of epithelial and mesenchymal markers were measured in TRERNA1‐overexpressing HepG2 cells. B, The relative mRNA expression levels of epithelial and mesenchymal markers were measured in TRERNA1‐depleted HepG2.215 cells. C, Western blot analysis of CDH1 expression in HepG2 cells transfected with TRERNA1 or pcDNA3.1. Western blot analysis of CDH1 expression in HepG2.215 cells transfected with shTRERNA1 or the control. D, The gel electrophoresis results of the PCR products from the RIP assay of the enrichment of EHMT2 on TRERNA1 in HepG2.215 cells was shown. SNRNP70 and IgG were used as positive and negative controls, respectively. E, TRERNA1, U1 snRNA and β‐actin from the RIP assay were also analysed by qRT‐PCR. C, The relative mRNA level of CDH1 was detected by qRT‐PCR after knocking down EHMT2 in HepG2.215 cells. F, The relative protein level of CDH1 was detected by a Western blot after knocking down EHMT2 in HepG2.215 cells. G, Schematic diagram showed the primer position of 3 detection sites (A, B and C) on the CDH1 promoter by ChIP. H, Anti‐EHMT2 ChIP assay and qRT‐PCR were employed to detect the binding of EHMT2 on the CDH1 promoter region after the inhibition of TRERNA1 in HepG2.215 cells. I, Anti‐H3K9me2 ChIP assay and qRT‐PCR were performed to detect the enrichment of H3K9me2 on the CDH1 promoter region after the knockdown of TRERNA1 in HepG2.215 cells. RNA polymerase II and IgG were used as positive and negative controls in (H, I), respectively. Data are shown as the mean ± SD; n = 3. *P < 0.05, **P < 0.01 (Student's t test)

Figure 4

TRERNA1 recruits EHMT2 binding with SNAI1 to repress CDH1 expression. A, Immunoprecipitation of endogenous EHMT2 and its associated endogenous SNAI1 was analysed by Western blot. B, The CDH1 expression level was analysed by qRT‐PCR in HepG2/HepG2.215 cells after transfection with SNAI1 or siSNAI1. C, After the overexpression or knockdown of SNAI1, the CDH1 expression was detected by a Western blot in HepG2/HepG2.215 cells. D, After the overexpression or knockdown of TRERNA1, the SNAI1 expression levels were analysed by a Western blot analysis. E, The mRNA levels of CDH1 were analysed by qRT‐PCR in HepG2 cells transfected with ConsiRNA, pcDNA3.1, TRERNA1, siTRERNA1, SNAI1, siSNAI1 and siEHMT2. Data are shown as the mean ± SD; n = 3. *P < 0.05, **P < 0.01 (Student's t test)

Figure 5

TRERNA1 promotes the ability of SNAI1 to treat HCC metastasis. A, Cell migration assays were performed using transwell membranes. The average number of cells exhibiting migration from three random microscopic fields is presented in the histogram. B, Cell invasion assays were performed using Matrigel‐coated transwell membranes. The average number of cells exhibiting invasion from three random microscopic fields is presented in the histogram. C and D, A wound‐healing assay in SNAI1‐overexpressing HepG2 cells and in SNAI1‐depleted HepG2.215 cells. The scratch was measured 0, 24 and 48 h later. E, Cell migration assays using transwells in HepG2 cells treated with the ctrl, SNAI, SNAI + TRERNA1, or SNAI + siTRERNA1. The average number of cells exhibiting migration from three random microscopic fields was presented in the histogram. F, Cell migration assays using transwells in HepG2.215 cells treated with the ctrl, siSNAI1, siSNAI + siTRERNA1, or siSNAI1 + TRERNA1. The average number of cells exhibiting migration from three random microscopic fields is presented in the histogram. The results are presented as the mean ± SD; n = 3. *P < 0.05, **P < 0.01 (Student's t test)

Figure 6

Overexpressed TRERNA1 levels are correlated with metastasis and are negatively correlated with tumour metastasis repressor gene CDH1 expression in HCC patients. A, The expression levels of TRERNA1 in HCC tissues and in paired, adjacent non‐tumour tissues were analysed by qRT‐PCR. The horizontal lines in the box plots represent the medians, the boxes represent the interquartile range, and the whiskers represent the 2.5th and 97.5th percentiles. The significant differences were analysed by the Wilcoxon signed‐rank test. B, The TRERNA1 expression level in HCC metastasis tissues (n = 37) and in non‐metastatic tissues (n = 32). β‐actin was used as an internal control. C, The expression level of CDH1 in HCC tissues and in paired adjacent non‐tumour tissues by qRT‐PCR (n = 69). The horizontal lines in the box plots represent the medians, the boxes represent the interquartile range, and the whiskers represent the 2.5th and 97.5th percentiles. The significant differences were analysed by the Wilcoxon signed‐rank test. D, CDH1 expression levels were examined in HCC metastasis tissues (n = 37) and in non‐metastatic tissues (n = 32). E, The correlation between the CDH1 mRNA level and the TRERNA1 transcript level was measured by qRT‐PCR in metastatic tumour tissues. The 2^‐ΔΔCt values (normalized to β‐actin) were subjected to Pearson's correlation analysis (P = 4.42178E‐4, R ² = 0.28069). F, The correlation between the CDH1 mRNA level and the TRERNA1 transcript level was measured by qRT‐PCR in non‐metastatic tumour tissues. The 2^‐ΔΔCt values (normalized to β‐actin) were subjected to Pearson's correlation analysis (P = 0.21022, R ² = 0.02021). Data are presented as the mean ± SD; n = 3. *P < 0.05, **P < 0.01

Figure 7

A schematic diagram of the lncRNA TRERNA1 functions in HCC metastasis. LncRNA TRERNA1, which could be upregulated in HCC, recruits EHMT2 as a scaffold and forms an EHMT2/SNAI1 complex to suppress the expression of CDH1 via the dimethylation of H3K9 in promoting HCC metastasis