Abstract
Background/Aim: MicroRNAs (miRNAs) are abnormally expressed and involved in the pathogenesis of various carcinomas. The present study aimed to identify novel miRNA genes associated with the pathogenesis and prognosis of oesophageal squamous cell carcinoma (ESCC). Materials and Methods: The miRNA profiling of 873 genes was performed using surgically resected oesophageal tissues from 35 patients with ESCC to identify candidate miRNAs. To examine the biological activities of candidate miRNAs, their proliferative, invasive, and migratory abilities were evaluated in ESCC cells subjected to miRNA mimic-mediated over-expression. The miRNA expression levels of the selected candidate miRNAs were analysed in the resected oesophageal tissues of 76 patients with ESCC from the two cohorts and correlated with the clinicopathological parameters. Results: Among the four candidate miRNAs identified by miRNA profiling, miR-877-3p was selected for subsequent analyses. In vitro analyses showed that the over-expression of miR-877-3p significantly suppressed the proliferation, invasion, and migration of ESCC cell lines compared with those of control cells. In the analyses of clinical specimens, the expression of miR-877-3p was down-regulated in ESCC tissues compared with that in adjacent normal oesophageal tissues. The down-regulation of miR-877-3p expression in ESCC tissues was significantly associated with advanced local progression and lymphatic involvement. The miR-877-3p down-regulation was also significantly associated with poor disease-free and disease-specific survival. Conclusion: miR-877-3p acts as a tumour suppressor gene in ESCC cells, and its down-regulation in ESCC tissues is associated with a poor prognosis. Thus, miR-877-3p may serve as a novel prognostic marker and promising therapeutic target.
Oesophageal cancer is the sixth most common cause of cancer-related deaths worldwide and one of the most common types of malignancies (1). Oesophageal squamous cell carcinoma (ESCC) and adenocarcinoma are two major subtypes of oesophageal cancer. ESCC accounts for 70% of all oesophageal cancer cases (2). Although advances in multidisciplinary treatment and diagnostic approaches have improved the patient’s outcomes, ESCC has poor prognosis (3). The identification of genes contributing to the prognosis of ESCC is an important research issue that will assist in the accurate prediction of prognosis and the development of new molecular targeted therapies.
MicroRNAs (miRNAs) are small non-coding RNAs approximately 22 nucleotides in length. They regulate the expression of target genes by binding to the 3′-untranslated regions (3′-UTRs) of target mRNAs in a sequence-complementary manner (4-6). Previous studies have shown that miRNAs are abnormally expressed in various cancer types (5-7). Furthermore, miRNAs have been associated with tumour occurrence and development, including cell proliferation, differentiation, apoptosis, invasion, and metastasis (8, 9). Hence, the identification of ESCC-associated miRNAs is one of the most important research topics (10). We also reported that several miRNAs contribute to oncogenesis and serve as prognostic markers of ESCC (11-13); however, the role of miRNAs in ESCC remains unclear.
In the present study, miRNA profiling of 873 genes was conducted using oesophageal tissues from patients with ESCC to identify candidate miRNA genes associated with ESCC pathogenesis and prognosis. Among the identified candidate miRNAs, we selected miR-877-3p as a promising candidate miRNA, and investigated its expression, function, and clinical significance in the development of ESCC. We evaluated the biological function of miR-877-3p in a panel of human ESCC cell lines over-expressing miRNA mimics of miR-877-3p. In addition, we measured the miR-877-3p expression levels in clinical samples and their relationship with the clinicopathological characteristics.
Materials and Methods
Ethics. The study was conducted in accordance with the ethical guidelines of the World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. This study was approved by the Institutional Review Board of the University of Toyama and Nagoya University, Japan. Written informed consent for the use of clinical data and samples was obtained from all patients as required by our institutional review board.
miRNA profiling with microarray assay. Surgically resected oesophageal tissues from 35 patients with ESCC were subjected to miRNA profiling. miRNA profiling was carried out using a 3D-Gene® miRNA oligo chip (Toray Industries, Tokyo, Japan) to compare the expression levels of 873 miRNAs in primary ESCC tissues with those in non-cancerous adjacent ooesophageal tissues. Details of the experiment have been described in previous studies (14). miRNAs that showed significant changes in expression levels in cancerous and non-cancerous tissues were selected as candidate miRNAs.
Bioinformatics analysis using The Cancer Genome Atlas (TCGA) dataset. For further selection of candidate miRNAs, we analysed the data of 96 ESCC patients from TCGA database (15). We analysed the association between the miRNA expression level and prognosis, and only those miRNAs that were significantly associated with prognosis were retained as candidates.
Cell lines. For the in vitro analyses, we used a panel of 10 human ESCC cell lines. KYSE30, KYSE140, KYSE150, KYSE410, KYSE510, KYSE590, and KYSE1440 were obtained from the Japanese Collection of Research Bio Resources Cell Bank (Osaka, Japan). TE2 and TTn cells were obtained from the American Type Culture Collection (Manassas, VA, USA). NUEC2 was established at Nagoya University (16). All cell lines were cultured in Roswell Park Memorial Institute 1640 medium (FUJIFILM Wako, Osaka, Japan) supplemented with 10% foetal bovine serum (Biosera, France) and 1% antibiotics (100 U/ml penicillin G and 100 μg/ml streptomycin; Gibco Thermo Fisher Scientific, Waltham, MA, USA). All cultures were maintained at 37°C in an incubator containing 5% CO2. The non-tumorigenic epithelial cell line (Het1A) was used as the control.
Patients and tissue specimens. For miRNA expression analysis using quantitative reverse-transcription polymerase chain reaction (qRT-PCR), ooesophageal tissues (primary ESCC and adjacent non-cancerous mucosa) were collected from 76 patients who underwent radical esophagectomy at two institutions. Of the 76 patients, 26 underwent surgical resection at Toyama University Hospital between 2018 and 2021 and the remaining at Nagoya University Hospital between 2001 and 2016. The tissue samples were directly frozen and stored at −80°C upon resection. The specimens were confirmed to be ESCC by histological classification based on the 8th edition of the Union for International Cancer Control (UICC) staging system for oesophageal cancer (17).
qRT-PCR analysis. miRNA expression analysis of hsa-miR-877-3p in cell lines and tissue samples was performed using the CFX Connect qRT-PCR systems (Bio-Rad, Hercules, CA, USA). The QIAzol® Lysis Reagent (Qiagen, Hilden, Germany) and miRNeasy® Mini Kit (Qiagen) were used to isolate the miRNAs from cultured cells or tissue samples, according to the manufacturer’s protocol. Complementary DNA was produced using the miRCURY LNA™ RT Kit (Qiagen), and the miRCURY LNA™ SYBR Green PCR Kit (Qiagen) and appropriate miRCURY LNA miRNA PCR assay primers (Qiagen) were used for performing qRT-PCR. The thermocycling conditions for qRT-PCR were as follows: 95°C for 2 min, 45 cycles of denaturation at 95°C for 10 s, and annealing/extension at 56°C for 60 s. We used the miR-103a-3p miRNA as an internal standard control to calculate the relative miR-877-3p miRNA levels in each sample. The 2−ΔΔCq method was used to calculate the relative miRNA expression levels.
miRNA mimic-mediated over-expression of miR-877-3p. To evaluate the biological function of miR-877-3p, a miRNA mimic-mediated over-expression experiment was performed. The miR-877-3p mimics and negative control (miR-NC) were purchased from Qiagen. The miR-877-3p mimic sequence was 5′-UCCUCUUCUCCCUCCUCCCAG-3′, while the miR-NC sequence was 5′-UCACCGGGUGUAAAUCAGCUUG-3′. Logarithmically grown cells were seeded in 12-well plates at a density of 1×105 cells/well. The cells were transfected with oligonucleotides (10 pmol) using Lipofectamine® RNAiMAX (Invitrogen; Thermo Fisher Scientific) according to the manufacturer’s protocol. Functional assays were performed 24 h after transfection.
Functional assays. Cell proliferation was determined using the Cell Counting Kit-8 (Dojindo Molecular Technologies, Kumamoto, Japan), migration was measured using a wound-healing assay (ibidi GmbH, Martinsried, Germany), and invasion was determined using BioCoat Matrigel Invasion Chambers (Corning, Glendale, AZ, USA), as described previously (18, 19). For the migration assay, the wound width in each well was measured 20 times at 100-mm intervals. The KYSE1440 and NUEC2 cells were used for functional assays owing to their transfection efficacy. Invasion was only assessed in the KYSE1440 cells, because NUEC2 cannot penetrate the Matrigel layer (20).
Clinical significance of the miR-877-3p miRNA expression level. The patients were divided into the miR-877-3p down-regulation and non-down-regulation groups using the cancerous to non-cancerous expression level ratio of 0.67 as the cut-off value, which was determined using the Youden index of the receiver operating characteristic (ROC) curve. The correlations between down-regulation or non-down-regulation of miR-877-3p expression, clinicopathological parameters, and long-term outcomes, including the disease-free survival (DFS) and disease-specific survival (DSS), were evaluated.
Statistical analysis. Data were compared between the two groups using the Mann–Whitney U-test or the χ2 test. The Kaplan–Meier method was used to calculate the DFS and DSS rates, and the differences in survival curves were analysed using the Cox proportional hazards model. All statistical analyses were performed using the JMP 16 software (SAS Institute Inc., Cary, NC, USA). A p-value of <0.05 was considered statistically significant.
Results
Identification of miR-877-3p as a candidate biomarker. First, we performed a comprehensive gene analysis of 873 miRNAs using 35 pairs of ESCC cells and adjacent non-tumour tissues. This finding revealed that 111 miRNAs were significantly down-regulated in cancerous areas compared with those in non-cancerous areas, while 12 miRNAs were up-regulated in cancerous areas. To narrow down the candidate miRNAs from the 123 miRNAs that showed significant down- or up-regulation of expression levels, TCGA public dataset was searched for miRNAs associated with ESCC prognosis. Of the 111 down-regulated miRNAs, the low expression of miR-491-5p, miR-877-3p, and miR-877-5p was significantly associated with poor prognosis. In addition, of the 12 up-regulated miRNAs, the high expression of miR-1291 was significantly associated with poor prognosis (Figure 1A). Next, we conducted a literature review on the functions of the candidate miRNAs and selected miR-877-3p for subsequent analyses owing to the following reasons (Table I): 1) among the four candidate miRNAs, miR-877-3p is among the genes with the lowest expression levels in primary ESCC tissues; 2) miR-877-3p acts as a suppressor gene in several cancer types (21, 22); and 3) no study related to the oncological roles of miR-877-3p in ESCC has been published. The results of the above analyses of miR-877-3p are shown in Figure 1B and C.
Outline of candidate gene selection and analyses of miR-877-3p. A) Flowchart of candidate gene selection by miRNA array analysis and bioinformatics analysis using TCGA dataset. B) The miR-877-3p expression levels in 35 pairs of oesophageal squamous cell carcinoma (ESCC) and adjacent non-tumour tissues. C) Kaplan–Meier analysis of the overall survival of 90 patients with ESCC from TCGA dataset.
Candidate miRNA genes identified from miRNA array and TCGA dataset.
Effect of miR-877-3p over-expression on the biological activities of ESCC cells. First, we analysed the expression of miR-877-3p in 10 human ESCC cell lines using qRT-PCR. The expression levels varied widely in ESCC cell lines, with no significant differences between ESCC cell lines with various degrees of differentiation (p=0.500) (Figure 2A).
Expression levels of miR-877-3p and effects of miR-877-3p over-expression in oesophageal squamous cell carcinoma (ESCC) cells. A) miRNA levels of miR-877-3p in 10 ESCC cell lines. B) Proliferation of NUEC2 and KYSE1440 cells with and without miRNA mimic-mediated miR-877-3p over-expression. C) Wound-healing assay with NUEC2 and KYSE1440 cells. The graph on the left shows the mean migration distance at the indicated times and the panels on the right show representative images of cells. D) Invasion assay with KYSE1440 cells. The graph on the left shows the mean number of invading cells in eight randomly selected fields. The panels on the right show representative images of stained cells (×200 magnification). *p<0.05.
The NUEC2 and KYSE1440 cells with lower miR-877-3p expression levels were selected for subsequent functional assays. These two cell lines were transfected with miR-877-3p-mimics or miR-NC. Cell proliferation was significantly decreased in NUEC2 and KYSE1440 cells by miR-877-3p over-expression from 72 h to 96 h compared with that in the control cells (Figure 2B). Next, the wound-healing assay revealed that the over-expression of miR-877-3p significantly decreased the migration of NUEC2 and KYSE1440 cells (Figure 2C). Moreover, miR-877-3p over-expression inhibited the invasion of KYSE1440 cells compared with that of the control cells (Figure 2D).
ESCC patients and miR-877-3p miRNA expression in ESCC tissues. Next, we examined the miR-877-3p expression levels in the primary ESCC tissues and adjacent normal tissues from 76 ESCC patients in two cohorts. Of the 76 patients, 58 were men and 18 were women, with a median age of 66 years (range=44–84 years). Based on the 8th edition of the UICC classification, 13, 25, 34, and 4 patients were classified as having pathological stages I, II, III, and IV, respectively. NAC was administered to 26 (34 %) patients. The median follow-up duration was 22.7 months, during which 34 patients (45%) experienced recurrence and 26 (34%) died from the disease.
The miR-877-3p levels were lower in the ESCC tissues compared with those in normal adjacent oesophageal tissues from 51 patients with ESCC (67%, p=0.028; Figure 3A).
Expression levels of miR-877-3p in oesophageal squamous cell carcinoma (ESCC) and normal oesophageal tissues and its prognostic value. A) Histogram of cancerous to non-cancerous miR-877-3p expression level ratio. B, C) Kaplan–Meier analyses of disease-free survival (B) and disease-specific survival (C) of 76 patients with ESCC who underwent radical surgery for ESCC.
Prognostic value of miR-877-3p miRNA level in ESCC tissues. The 76 ESCC patients were divided into the miR-877-3p down-regulation and non-down-regulation groups using the cancerous to non-cancerous expression level ratio of 0.67 as the cut-off value, and the relationships between expression and clinicopathological parameters were analysed (Table II). The down-regulation of miR-877-3p in ESCC tissues was significantly associated with advanced T-stage and lymphatic involvement.
Association between expression level of miR-877-3p and clinicopathological parameters in 76 patients with oesophageal squamous cell carcinoma.
Next, a survival analysis of the miR-877-3p miRNA expression was performed. DFS was significantly lower in the miR-877-3p down-regulation group compared with that in the non-down-regulation group (5-year DFS rates, 38% and 64%, respectively, p=0.045; Figure 3B). DSS was also significantly lower in the miR-877-3p down-regulation group compared with that in the non-down-regulation group (5-year DSS rates, 43% and 74%, respectively, p=0.039; Figure 3C).
Discussion
In the present study, to identify the novel miRNA genes associated with the pathogenesis and prognosis of ESCC, miRNA profiling was performed using a microarray assay and identified miR-877-3p as a candidate miRNA. In vitro analyses showed that the over-expression of miR-877-3p in ESCC cell lines significantly reduced the malignant potential, including cell proliferation, migration, and invasion. In the analyses of clinical specimens, the expression level of miR-877-3p in ESCC tissues was down-regulated compared with that in normal oesophageal tissues, which is significantly associated with poor prognosis. These results indicate the potential role of miR-877-3p as a tumour suppressor gene in ESCC, suggesting that miR-877-3p may serve as a novel prognostic marker as well as a promising therapeutic target for ESCC.
Previous basic studies on several types of malignancies evaluated the behaviour of miR-877-3p and reported the tumour-suppressive role of miR-877-3p in vitro. The over-expression of miR-877-3p inhibits the proliferation of bladder cancer cells (21) and decreases the migration and invasion of gastric and cervical cancer cells (22, 23). In the present study, the over-expression of miR-877-3p attenuated the malignant potential of ESCC cells. Recently, miRNA replacement therapy has attracted attention as a novel therapeutic strategy for cancer (24). This approach aims to restore the function of tumour-suppressor miRNAs in tumour cells using synthetic miRNA mimics or miRNA expression plasmids (25). Our results suggest that miR-877-3p acts as a tumour suppressor miRNA in ESCC and may be a promising therapeutic target for miRNA replacement therapy.
miR-877-3p has been associated with the prognosis of patients with several cancer types, such as cervical cancer (23, 26); however, the clinical significance of miR-877-3p in ESCC patients has not been reported. Therefore, we revealed the relationship between the miR-877-3p expression and clinicopathological parameters in ESCC patients. First, our analysis of the clinical specimens showed that the down-regulation of miR-877-3p expression in ESCC tissues is significantly associated with advanced local progression and lymphatic involvement. In basic research, the inhibition of miR-877-3p expression results in epithelial-mesenchymal transition and promotes the local progression of cancer and lymphatic involvement (27, 28). Indeed, the low expression of miR-877-3p in tumour tissues is reportedly associated with advanced T stage and lymphatic involvement in gastric cancer (23). Our results were consistent with the findings of these previous studies. Second, we showed that the down-regulation of miR-877-3p expression in ESCC tissues was associated with poor prognosis, such as shortened DFS and DSS, suggesting the possibility of miR-877-3p as a prognostic biomarker for ESCC. Although the prognostic value of miR-877-3p is not sufficient to fully predict the prognosis of patients with ESCC, we previously reported the applicability of miRNA profiles with several biomarkers in prognosis prediction (12). Therefore, further studies focusing on the use of miR-877-3p in combination with other biomarkers would be informative, as it can enhance the performance of miR-877-3p as a biomarker.
This study has several limitations. First, we did not investigate the target genes of miR-877-3p that display biological activity in ESCC cells. Previous studies reported that miR-877-3p acts as a tumour suppressor gene by inhibiting the production of oncogenes, such as p16, fibroblast growth factor 2, and vascular endothelial growth factor A (VEGFA) (21-23, 29). Therefore, further studies examining the network of miR-877-3p and these oncogenes in ESCC cells are desirable. Second, we retrospectively examined the clinical data. Third, we did not determine the optimal cut-off value applicable to ESCC patients aside from our cohort, and the cut-off value for actual clinical application should be redetermined in a larger-scale study.
In conclusion, our results showed that miR-877-3p acts as a tumour suppressor gene in ESCC cells, and its down-regulation in ESCC tissues is associated with a poor prognosis. Thus, miR-877-3p may serve as a novel prognostic marker and promising therapeutic target.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 22K08730.
Footnotes
Authors’ Contributions
TF, HB, and TO conceived the study’s concept and design, analysed the data. TF and HB wrote the manuscript. MK, TA, HT, TM, TW, KH, SS, IH, KS, SH, IY, KM, YK, and TF contributed to the data acquisition and interpretation. TO, MK, TA, HT, TM, TW, KH, SS, IH, KS, SH, IY, KM, YK, and TF revised the manuscript. All Authors have read and approved the final version of the manuscript.
Conflicts of Interest
The Authors declare that they have no conflicts of interest associated with this manuscript.
- Received November 3, 2022.
- Revision received November 10, 2022.
- Accepted November 11, 2022.
- Copyright © 2023 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
