HCT-8, HT-29 and HCT 116 cells were cultured in Roswell Park Memorial Institute medium (Welgene, Inc.), supplemented with 10% fetal bovine serum (Corning, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin (Welgene, Inc.). These cells (5×10⁵) were plated in 100-mm culture dishes and were cultured at 37°C in a 5% CO₂ humidified incubator. When the cells achieved 70–80% confluence, subculture was performed. For 2-Gy ionizing radiation (IR) treatment, cells were γ-irradiated for 34 sec from a 137 Cs γ-ray source (GC-3000 Elan; Atomic Energy of Canada, Ltd.) administered at a dose of 3.5 Gy/min.

Gene expression levels (n=478) were obtained from the TCGA data portal (http://tcga-data.nci.nih.gov/tcga/) and cBioPortal (http://www.cbioportal.org/public-portal/), to analyze the expression profile and prognostic significance of the NMU gene in patients with colorectal cancer. To calculate the effect of NMU expression on patient survival, UALCAN (ualcan.path.uab.edu) was used to create a Kaplan-Meier plot with TCGA data. The fragments per kilobase of transcripts per million (FPKM) for each gene were used for quantification of expression and patients were divided based on the expression level into one of the two groups: NMU low (<3.0 FPKM) or NMU high (>3.0 FPKM).

HCT-8, HT-29 and HCT 116 cells were detached by treatment with trypsin-EDTA and stained with 0.1% trypan blue for 5 min at room temperature. Cell numbers were measured using a LUNA-FX7 Automated Cell Counter (Logos Biosystems; Aligned Genetics).

Cells were seeded into 6-well cell culture plates at 100, 200 or 500 cells/well. After seeding, cells were irradiated with 0, 2 or 3 Gy IR for 0, 34 or 51 sec using a GC-3000 Elan (Atomic Energy of Canada, Ltd.). After irradiation, cells were incubated in a 5% CO₂ humidified incubator at 37°C for 2 weeks. When a single cell grew into a colony (>50 cells), the culture medium was removed and cells were fixed in 4% paraformaldehyde at 4°C for 20 min. After removing the fixation buffer, colonies were stained with 0.5% crystal violet solution at room temperature for 30 min. The crystal violet solution was gently removed and the number of colonies in the entire well was counted, and relative survival fraction was calculated by plating efficiency (PE) and surviving fraction (SF). PE=number of colonies formed/number of cells seeded ×100. SF=number of colonies formed after IR treatment/number of cells seeded × PE.

Cells were seeded at 3×10⁵ cells/well in 24-well plates. After cells became 100% confluent, scratch wounds were made using a 1-mm pipette tip, and the cells were incubated at 37°C in RPMI media containing 2% serum for 24 h. Wound healing was observed using an optical microscope (Eclipse Ts2; Nikon Corporation) and images were obtained using an MShot Digital Imaging System V1.1.6 (Guangzhou Micro-shot Technology Co., Ltd.) and analyzed using ImageJ software V1.53 (National Institutes of Health).

Cells were seeded at 1×10⁴ cells/well in 8-well slides and incubated at 37°C for 24 h. Each slide was exposed to 2 Gy radiation at room temperature. After 24 h, the culture medium was removed, and the cells were fixed using 4% paraformaldehyde for 25 min at 4°C, followed by permeabilization with 0.2% Triton X-100 at room temperature for 5 min. TUNEL staining was performed using a DeadEND™ Fluorometric TUNEL system kit (Promega Corporation). The cells were stained in a Coplin jar by immersing the slides in 40 ml propidium iodide solution freshly diluted to 1 µg/ml in PBS for 10 min at room temperature in the dark and mounted by 50 µl VECTASHIELD with DAPI (cat. no. H-1200; Vector Laboratories, Inc.) at room temperature. The fluorescence intensity was measured using a confocal microscope (LSM 900; Carl Zeiss AG).

RNA was extracted from cells using Tri-RNA Reagent (Favorgen), according to the manufacturer's protocol. Subsequently, 1 µg extracted RNA was subjected to complementary cDNA synthesis using the All-In-One cDNA Master Mix (CellScript™, CellSafe) at 42°C for 1 h. qPCR amplification was performed using the qPCR SYBR 2X Master Kit (cat. no. 18102; Mbiotech) and gene-specific oligonucleotide primer pairs (Table I) from Bioneer Corporation. Amplification was conducted with the CFX96 Touch Real-Time PCR detection system (Bio-Rad Laboratories, Inc.). The thermocycling conditions consisted of initial denaturation at 95°C for 30 sec, followed by 40 cycles of denaturation at 95°C for 5 sec and annealing/extension at 60°C for 30 sec. GAPDH expression levels were used to normalize the mRNA expression levels in each sample. The 2^−ΔΔCq method was used to calculate the relative gene expression (20).

HCT-8, HT-29 and HCT 116 cells were cultured to ~70% confluence in 60-mm plates. For NMU and NMUR1 knockdown, 20 nM small interfering RNAs (siRNAs) targeting NMU or NMUR1, or control siRNA (Table II) were transfected using G-fectin transfection reagent (Genolution, Inc.) overnight at 37°C. After changing the media, the cells were incubated at 37°C for 24 h and subsequent experiments for proliferation, migration and TUNEL assays were performed. To assess the exogenous effect of NMU on NMUR1 knockdown, the cells were treated with 1 µM NMU recombinant protein (NMU-25; cat. no. ab141007; Abcam) at 37°C for 24 h. For NMU overexpression, 1 µg NMU-expression pCMV3 vector (Sino Biological, Inc.) was transfected using Lipofectamine^® 2000 transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. An empty vector was used as the control.

Cells were washed twice with phosphate-buffered saline and harvested with protein extraction solution (Elpis Biotech, Inc.) containing protease-phosphatase inhibitor (Cell Signaling Technology, Inc.). The nuclear fraction was prepared using NE-PER™ Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. Protein concentrations were quantified with the bicinchoninic acid method (Pierce Biotechnology, Inc.) using 1 µg/ml bovine serum albumin (Pierce; Thermo Fisher Scientific, Inc.) as a standard.

Proteins (25 µg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on 10% gels and were transferred to a nitrocellulose membrane. The membranes were blocked using 5% skimmed milk (MilliporeSigma) for 1 h at room temperature, and soaked in primary antibody solutions (diluted to 1:1,000) against GAPDH (ca. no. sc47724), β-actin (cat. no. sc81178; Santa Cruz Biotechnology, Inc.), lamin A/C (cat. no. 4777), YAP (cat. no. 4912), TAZ (cat. no. 70148), LATS1 (cat. no. 9153), phosphorylated (p)-YAP (cat. no. 4911), p-TAZ (cat. no. 70148) and p-LATS1 (cat. no. 8654) (all Cell Signaling Technology, Inc.) overnight at 4°C. Membranes were then washed five times (10 min/wash) with Tris-buffer containing 1% Tween 20 and were incubated with horseradish peroxidase (HRP)-conjugated IgG goat anti-rabbit (cat. no. 7074) or HRP-conjugated IgG goat anti-mouse antibodies (cat. no. 7076) (diluted to 1:10,000; Cell Signaling Technology, Inc.) for 1 h at room temperature. Protein expression was detected using Western Lighting Plus-ECL (PerkinElmer, Inc.).

A total of 6 patients with pathologically proven locally advanced rectal cancer, tested between April 2018 and May 2019, whose neoadjuvant chemo-radiation therapy outcomes were categorized as complete response (n=3) or poor response (n=3) based on total mesorectal excision specimens, were selected for inclusion. Clinical data from patients with colorectal cancer were provided by the Korea Institute of Radiological and Medical Sciences (Seoul, South Korea; Table III). For the genetic information of patients with colorectal cancer, microarray data obtained from our previous study were used (5). Heatmaps were constructed using MeV software V4.7 (http;//mev.tm4.org/) based on the provided clinical data.

Data are presented as the mean ± standard deviation from three independent experiments and were analyzed using GraphPad Prism 9.0 software (GraphPad Software, Inc.). The statistical significance of differences between groups was determined using one-way ANOVA and subsequent Tukey's multiple comparison post hoc test, or using Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Microarray analysis was performed in our previous study (5) to confirm the gene transcriptional characteristics of radiation-resistant (RR) colorectal cancer. Six patients with colorectal cancer were classified as RR or radiation-sensitive (RS) according to tumor regression grade following radiation therapy. The patients were assessed by pelvis magnetic resonance imaging and 18-fluoro-2-deoxy-glucose positron emission tomography scanning before and after radiotherapy (5). The clinical information of these patients is summarized in Table III. Normal tissue was not used as a control for patients with cancer, but cancer tissues were compared according to the radiosensitivity of patients with cancer. The microarray analysis showed that 325 genes were expressed at higher levels in RR patients than in RS patients (cutoff of fold-change >2). The differential expression of the top 50 genes is presented in the heatmap in Fig. 1A. RT-qPCR was performed on the top 50 differentially expressed genes to verify the microarray results. Our previous study (5) showed that 40 genes were expressed at higher levels in RR colorectal cancer than in RS colorectal cancer (cutoff of fold-change >2). The expression levels of 20 genes are presented in Fig. 1B. By analyzing these data, NMU was selected as a candidate biomarker for RR colorectal cancer. The relationship between NMU expression and prognosis among patients with colorectal cancer was examined using data from The Cancer Genome Atlas, which revealed a significant association between NMU expression (Fig. 1C) and prognosis (Fig. 1D). Based on these data, NMU may affect the development of radiation resistance and the progression of colorectal cancer.

The present study first confirmed the expression levels of NMU and its receptors, NMUR1 and NMUR2, in HCT-8, HT-29 and HCT 116 colorectal cancer cells. NMU expression was high in HCT-8 cells, and relatively low in HT-29 and HCT 116 cells. Conversely, the expression levels of NMUR1 and NMUR2 were highest in HCT 116 cells (Fig. S1). Given the observed relationship between NMU upregulation and poor patient survival, the present study investigated whether NMU knockdown affected colorectal cancer cell proliferation in the presence and absence of radiation. siRNA targeting NMU was used to transfect HCT-8 and HCT 116 colorectal cancer cells, which resulted in an NMU depletion efficiency of ~40% (Fig. 2A). Colorectal cancer cells transfected with NMU siRNA exhibited reduced cell proliferation and this inhibition was further enhanced when these cells were irradiated (Fig. 2B). HCT-8 and HCT 116 cells transfected with NMU siRNA showed 17 and 34% reductions in colony-forming ability following 3 Gy irradiation in the clonogenic assay compared with that in cells transfected with control siRNA (Fig. 2C). Subsequently, the effect of NMU knockdown on the migration of colorectal cancer cells was assessed using the monolayer wound-healing assay. NMU siRNA reduced the migration of HCT-8 and HCT 116 cells, and migration was further reduced when these cells were exposed to radiation (Fig. 2D). TUNEL staining confirmed that NMU depletion increased the irradiation-induced apoptosis of colorectal cancer cells (Fig. 2E). These data indicated that NMU knockdown may affect cell survival, cell migration and radiation resistance in colorectal cancer cells.

To evaluate the effect of NMU overexpression on colorectal cancer cells, HT-29 and HCT 116 cells were transfected with a DNA vector containing the NMU open reading frame, resulting in the successful overexpression of NMU (Fig. 3A). Following exposure to 3 Gy irradiation, NMU-overexpressing HT-29 and HCT 116 cells showed 22 and 200% increases in colony formation compared with that in control vector-transfected cells (Fig. 3C). The proliferation and migration of colorectal cancer cells were increased following NMU overexpression. In addition, NMU overexpression restored or offset reductions in proliferation and migration induced by irradiation (Fig. 3B and D). Irradiation-induced apoptosis was also reduced in cells with NMU overexpression (Fig. 3E). These data indicated that NMU overexpression may promote malignant phenotypes and protect against the irradiation-induced apoptosis of colorectal cancer cells.

YAP and TAZ are associated with radiation resistance in various types of cancer (21,22). To investigate whether NMU is involved in the Hippo pathway, which regulates YAP and TAZ activity, changes in YAP and TAZ expression following NMU knockdown or overexpression were determined by western blotting. In HT-29 cells, NMU depletion increased the expression levels of p-LATS1, p-YAP and p-TAZ, and similar results were observed under irradiated conditions (Fig. 4A). To confirm that NMU enhances the nuclear translocation of YAP and TAZ, which are known to induce cancer cell proliferation and tumorigenesis, nuclear fractionation and western blotting were performed. siRNA-mediated NMU depletion decreased the nuclear translocation of YAP and TAZ (Fig. 4B). By contrast, NMU overexpression reduced the expression levels of p-LATS1, p-YAP and p-TAZ (Fig. 4C), and increased the nuclear translocation of YAP and TAZ (Fig. 4D). Similar outcomes were observed under irradiated conditions. Taken together, these results suggested that NMU may be involved in regulating YAP and TAZ activity in colorectal cancer cells.

The present study investigated whether agonist activation of NMUR1 by NMU affects colorectal cancer cell proliferation and migration. In HCT 116 cells, NMUR1 was depleted using siRNA (Fig. 5A), followed by treatment with NMU-25, which is the functional form of NMU in humans. Treatment with NMU-25 induced colorectal cancer cell proliferation and migration, similar to the effects observed with NMU overexpression; however, NMU-25 had no effect on cells transfected with NMUR1 siRNA (Fig. 5B and C). Nuclear fractionation and western blotting revealed increased nuclear translocation of YAP and TAZ in colorectal cancer cells following treatment with NMU-25, but the effects of NMU-25 treatment were attenuated by NMUR1 siRNA (Fig. 5D). As well as the increased nuclear translocation of YAP and TAZ in cells treated with NMU-25, the expression levels of the YAP and TAZ transcriptional target genes AXL, CTGF and CYR61 were increased, and these effects were abolished in NMUR1-depleted cells (Fig. 5E). These results suggested that the NMU-indued migratory phenotypes observed in colorectal cancer cells involve YAP and TAZ activity and are mediated through NMUR1 signaling.