Primary CRC tissues were obtained from 140 consecutive patients aged 27-91 years (49 females and 91 males who underwent surgical resection between January 2012 and December 2013 at Chiba University Hospital (Chiba, Japan). Corresponding liver metastasis tissue was obtained from 11 matched patients who underwent surgical resection at Chiba University Hospital. These patients ranged in age from 48 to 76 years, and consisted of 5 females and 6 males. Cases in which preoperative chemotherapy resulted in the total disappearance of tumour cells were excluded. Patients who underwent two-stage hepatectomy were excluded. The clinicopathological characteristics of patients are shown in Table SI. The Ethics Committees of the Department of General Surgery, Chiba University Hospital (Chiba, Japan) approved the study protocol (approval no. M10101) and written informed consent to participate was obtained from each patient before surgery. Union for International Cancer Control TNM classification of 8th edition (16) was used to assess the clinical outcome of the patients with CRC.

Paraffin-embedded tissue samples after 10% formalin fixed for 24 h at room temperature were cut to obtain 4-μm-thick slices and de-paraffinized with xylene and rehydrated with descending ethanol series. Antigen retrieval was performed by microwaving (500 W) slides in citric acid buffer (0.01 M; pH, 6.0) for 25 min at 100°C. Subsequently, endogenous peroxidase activity was blocked with hydrogen peroxide (3% in methanol) for 15 min at room temperature. Non-specific proteins were blocked with 5% bovine serum albumin (BSA; cat. no. 01860-36; Nacalai Tesque, Inc.) for 10 min at room temperature. Slides were incubated overnight at 4°C with the following primary antibodies: Anti-RYBP polyclonal (1:200; cat. no. HPA053357; Sigma-Aldrich; Merck KGaA) and anti-p53 Ab (1:100; cat. no. M7001; Dako; Agilent Technologies, Inc.). Secondary antibodies (undiluted; EnVision™ kits; cat. nos. K4001 and K4003; Dako; Agilent Technologies, Inc.) were applied for 30 min at room temperature, followed by staining with peroxidase DAB kit (cat. no. 25985-50; Nacalai Tesque, Inc.). Counterstaining was performed with haematoxylin before dehydration for 1 min at room temperature, penetration and mounting.

Using an inverted light microscope (cat. no. BX40; Olympus Corporation), the expression levels of RYBP and p53 were independently evaluated by two investigators and a pathologist, all of whom were blinded to clinical information. A 1% BSA/phosphate-buffered saline (PBS) solution without primary antibody served as a negative control. As a positive control, staining of normal liver tissue that originated from the tissue at least 1cm away from the tumour in a resected specimen from a patient with colorectal liver metastasis was confirmed because RYBP is expressed in the cytoplasm of normal hepatocytes (12). The evaluations were performed after establishing an inter-observer consensus using samples from preliminary experiments. In primary CRC, RYBP expression was evaluated by the percentage of positively stained nuclei in tumour cells relative to the total number of malignant cells in three positive high-power fields which were randomly selected (magnification, ×400); low and high expression were defined as <20 and ≥20%, respectively. The expression of p53 was evaluated by considering the distribution pattern of positive cells and the intensity of nuclear staining and categorised as wild-type (wt; absent and weakly and sporadically positive staining) and mutant p53 (strongly mosaic and diffuse staining).

The human colon cancer cell lines HCT116 (TP53^wt/wt) and HCT116 (TP53^−/−) were kindly provided by Dr Mamoru Takada (Chiba University, Chiba, Japan). SW48, DLD-1, HT29, COLO201 and SW620 were purchased from American Type Culture Collection. HCT116 cells were cultured in McCoy's 5A medium (cat. no. 16600082) with 10% foetal bovine serum (FBS; both Gibco; Thermo Fisher Scientific, Inc.) and incubated at 37°C with 5% CO₂. SW48 and SW620 cells were cultured in Leibovitz's L-15 medium (cat. no. 11415064; Gibco; Thermo Fisher Scientific, Inc.) with 10% FBS and incubated at 37°C without CO₂. DLD-1 and COLO201 cells were cultured in RPMI 1640 medium (cat. no. 22400089; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS and incubated at 37°C with 5% CO₂.

HCT116 (TP53^wt/wt), HCT116 (TP53^−/−), SW48 and DLD-1 cells (2-5×10⁵/well) were transfected with siRNA-1 and siRNA-2 [ON-TARGETplus Human RYBP (23429) siRNA, cat.no. J-015936-06 (target; 5′-GAA AGA UCC UCC UAG UGA A-3′) and J-015936-07 (target; 5′-CGA CAU GUC AGC AGU CAA U-3′); both GE Healthcare Dharmacon, Inc.] at a final concentration of 10 nmol/l or an equimolar concentration of control siRNA (AllStars Negative Control siRNA; sequence not available; cat. no. 1027280; Qiagen GmbH.) using Lipofectamine™ RNAiMAX transfection reagent (cat. no. 13778075; Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 24 h according to the manufacturer's recommendations. The expression (pcDNA3.1-RYBP) and empty vector (pcDNA3.1) were purchased from GenScript. Plasmid transfection was performed using 1 μg pcDNA and Lipofectamine^® 3000 transfection reagent (cat. no. L3000015; Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 24 h incubation, according to the manufacturer's instructions. Changes in RYBP levels were monitored using western blotting 48 h after transfection. The cells were used for subsequent assays 24 h after transfection.

Proteins were extracted from cultured cells using RIPA buffer (cat. no. 16488-34; Nacalai Tesque, Inc.). Each protein sample was lysed in Laemmli Sample Buffer (cat. no. 1610737; Bio-Rad Laboratories, Inc.) containing 5% 2-mercaptoethanol and incubated at 97°C for 10 min. After measuring the protein concentration of each sample using the Pierce™ BCA Protein Assay kit (cat. no. 23225; Thermo Scientific; Thermo Fisher Scientific, Inc.), 20 μg/lane protein was separated by electrophoresis on 10% XV PANTERA Gels (cat. no. NXV-224P; DRC) and transferred onto a polyvinylidene difluoride membrane. The membranes were blocked in 5% skimmed milk diluted in 0.1% Tris-buffered saline with Tween-20 or Blocking One-P (cat. no. 05999-84; Nacalai Tesque, Inc.) in the case of phosphorylated protein at room temperature (15-25°C) for 60 min. The membranes were incubated at 4°C overnight with primary antibodies against RYBP (1:1,000; cat. no. HPA053357; Sigma-Aldrich; Merck KGaA), p53 (1:1,000; cat. no. M7001; Dako; Agilent Technologies, Inc.), p21 (1:1,000; cat. no. 2947; Cell Signaling Technology, Inc.), cyclin D1 (1:1,000; cat. MA5-14512; Thermo Fisher Scientific, Inc.), Bax (1:1,000; cat. no. 5023; Cell Signaling Technology, Inc.), p-cyclin D1 (1:1,000; cat. no. 3300; Cell Signaling Technology, Inc.), YY-1 (1:1,000; cat. no. ab109228; Abcam) and β-actin (1:2,000; cat. no. 5125; Cell Signaling Technology, Inc.). Subsequently, the membranes were incubated with secondary antibodies at room temperature (15-25°C) for 60 min as follows: Anti-mouse (1:2,000; cat. no. SC516102; Santa Cruz Biotechnology, Inc.) and anti-rabbit HRP conjugated secondary antibodies (1:2,000; cat. no. 7074; Cell Signaling Technology, Inc.). The membranes were incubated with an enhanced chemiluminescence detection reagent (Chemi-Lumi One Ultra; Nacalai Tesque, Inc.) and developed using a LAS-4000UV mini luminescent image analyser (FUJIFILM Wako Pure Chemical Corporation). Band intensities were quantified by densitometric analysis using the ImageJ software version 1.53 (National Institutes of Health) to calculate the relative protein levels normalised to β-actin.

HCT116 (TP53^wt/wt), HCT116 (TP53^−/−), SW48 and DLD-1 cells were seeded in 96-well microplates at a density of 3,000 cells/well. At 0, 24, 48, 72, or 96 h, 10 μl Cell Count Reagent SF (cat. no. 07553044; Nacalai Tesque, Inc.) was added to 100 μl culture medium. After 2 h incubation at 37°C, the absorbance at 450 nm was measured using a microplate reader.

HCT116 (TP53^wt/wt), HCT116 (TP53^−/−) cells were collected 48 h after transfection, with or without 24 h pre-treatment with 4 or 10 μM oxaliplatin (Nippon Kayaku Co., Ltd.) at 37°C, washed with ice-cold PBS and fixed in 70% ethanol at -30°C overnight. Prior to staining, cells were washed with PBS. Pellets of the cells were mixed with 500 μl FxCycle™ PI/RNAse Solution (cat. no. 10797; Invitrogen; Thermo Fisher Scientific, Inc.) and incubated for 30 min at room temperature. Samples were analysed by fluorescence-activated cell sorting using FACS Canto II (BD Biosciences). All data were analysed using FlowJo v10.7.1 software (BD Biosciences).

Cell apoptosis was assessed using the FITC Annexin V/PI Apoptosis Detection kit (cat. no. 15342-54; Nacalai Tesque, Inc.), according to the manufacturer's instructions. Briefly, HCT116 (TP53^wt/wt), HCT116 (TP53^−/−) cells were collected, washed with Binding Buffer at 4°C and resuspended in 100 μl Binding Buffer. A total of 1×10⁵ cells were incubated with 5 μl Annexin V-FITC and PI at room temperature in the dark for 15 min. In total, 10,000-20,000 cells were analysed on a FACS Canto II (BD Biosciences) using FlowJo v10.7.1 (BD Biosciences). Oxaliplatin-treated cells were used as a positive control. The percentage of apoptotic cells was calculated based on the number of early apoptotic cells which were Annexin V-positive and PI negative.

HCT116 (TP53^wt/wt), HCT116 (TP53^−/−) cells were seeded in 96-well microplates at a density of 5,000 cells/well and 100 μl of cell-free McCoy's 5A medium was prepared as a blank. After 24 h pre-incubation at 37°C, oxaliplatin was added at concentrations of 0.001, 0.010, 0.1, 0.5, 1, 2, 4 and 10 μM and cells were incubated for 2 days at 37°C. A total of 10 μl Cell Count Reagent SF (cat. no. 07553044; Nacalai Tesque, Inc.) was added to each well. After 2 h incubation, absorbance at 450 nm was measured using a microplate reader. The cell viability was calculated as follows: Cell viability (%)=(absorbance of treated sample-absorbance of cell-free sample)/(absorbance of control sample-absorbance of cell-free sample) ×100.

The half-maximal inhibitory concentration (IC₅₀) was calculated with a non-linear regression fit to a sigmoidal dose-response curve and compared using the extra-sum-of-squares F test in Prism 9 (GraphPad Software, Inc.; Dotmatics). Experiments were performed six times independently.

Data are expressed as mean ± standard error of the mean or median ± standard deviation. The association between RYBP or p53 staining and patient characteristics were evaluated using χ² for categorical variables, unpaired Student's t for parametric continuous variables or Mann-Whitney U test for nonparametric continuous variables. Survival rates were calculated using Kaplan-Meier analysis and assessed using the log-rank test. Survival data were evaluated using univariate and multivariate Cox proportional hazards regression analysis. When analysing the association between RYBP expression in primary tumour and long-term outcomes, cancer-specific survival (CSS) and disease-free survival (DFS) were calculated from the date of primary tumour resection. When analysing the association between RYBP expression in liver metastases and long-term outcome, CSS and DFS were calculated from the date of initial hepatectomy. The in vitro experiments were performed at least three times independently and data were analysed using unpaired Student's t test or one-way or multivariate analysis of variance followed by Tukey's post hoc test. P≤0.05 was considered to indicate a statistically significant difference. All statistical analyses were performed using the JMP^® Pro15 software (SAS Institute Inc.).

To investigate RYBP expression and subcellular localisation in CRC, immunohistochemical staining of primary CRC and liver metastases was performed. In primary CRC, RYBP was predominantly expressed in the nuclei of both tumour and adjacent normal tissue (Fig. 1A). High RYBP expression was observed in 76 (54.3%) patients. The association between RYBP expression profiles and clinicopathological features is shown in Table I. The rate of distant metastases and recurrence after primary surgery were significantly higher in patients with low RYBP expression than in those with high RYBP expression (Table I). Kaplan-Meier analysis showed that patients with high RYBP expression had a significantly longer CSS and DFS after primary surgery than those with low RYBP expression (Fig. 1B). Univariate and multivariate analyses revealed that poor pathological differentiation, severe lymphatic invasion and low RYBP expression were independent prognostic factors associated with CSS (Table II).

Wt p53 expression was observed in 70 patients (50%), whereas mutant p53 expression was observed in 70 patients (50.0%; Fig. 1C). The status of p53 expression was not associated with clinicopathological features or prognosis (Table III). In patients with wt p53, CSS and DFS were significantly longer in those with higher RYBP expression (Fig. 1D). By contrast, in patients with mutant p53 expression, there were no significant differences in CSS and DFS (Fig. 1D). These data indicated that RYBP contributed to better prognosis in a TP53 status-dependent manner.

Nuclear expression of RYBP was not observed in 11 patients with patient-matched liver metastases. In all adjacent normal liver tissue samples, RYBP was moderately expressed in the cytoplasm (Fig. S1B). In nine of the 11 liver metastases, RYBP was also expressed in the cytoplasm, with a lower staining intensity than in adjacent normal tissue.

As the aforementioned data indicated that RYBP may have a tumour-suppressive role in CRC, in vitro experiments were performed to elucidate the molecular mechanisms by which RYBP regulates behaviour of CRC cells. Western blot analyses showed that the expression of both RYBP and p53 was high in HCT116 cells (TP53^wt/wt), whereas expression was low in SW48 cells. However, no association between RYBP and p53 expression was observed in cell lines with TP53 mutation, such as DLD-1, HT29, COLO201 and SW620 (Fig. 2A). To evaluate the effects of RYBP on TP53 wt colon cancer cells, HCT116 (TP53^wt/wt) and SW48 cells and TP53-mutant DLD-1 and TP53-null HCT116 (TP53^−/−) cells were used as the control groups for subsequent experiments.

To verify the function of RYBP, cell proliferation assay was performed following knockdown of RYBP using siRNAs and overexpression of RYBP using pcDNA3.1-RYBP plasmid. RYBP knockdown significantly increased proliferation of HCT116 (TP53^wt/wt) cells. However, knockdown did not alter the proliferation of SW48 cells, in which RYBP expression was low, and in TP53-mutant DLD-1 and TP53-null HCT116 (TP53^−/−) cells (Fig. 2B). Overexpression of RYBP significantly decreased proliferation of HCT116 (TP53^wt/wt) and SW48, but not TP53-mutant DLD-1 and TP53-null HCT116 (TP53^−/−) cells (Fig. 2C). These results suggested that RYBP may regulate tumour cell proliferation via a p53-associated pathway.

To elucidate the tumour-suppressive mechanisms associated with p53, cell cycle and apoptosis assays were performed. In HCT116 cells (TP53^wt/wt), RYBP knockdown significantly decreased the number of cells in G1/0 phase and increased the number of cells in S phase, whereas RYBP overexpression significantly increased the number of cells in G1/0 and decreased the number of cells in S phase (Fig. 3A). However, the number of HCT116 (TP53^−/−) cells in each cell cycle phase was not affected by modulation of RYBP expression (Fig. S2A). This also indicated that RYBP did not directly damage DNA. Western blot analysis was performed to evaluate expression of p21, a protein downstream of p53, and cyclin D1, a cell cycle G1-S checkpoint regulator (17). In HCT116 (TP53^wt/wt) cells, RYBP knockdown significantly decreased expression of p53 and p21 but induced cyclin D1 expression. Conversely, RYBP overexpression significantly induced expression of p53 and p21 but decreased cyclin D1 expression. However, neither knockdown nor overexpression had any effect on p53, p21 or cyclin D1 expression in TP53-mutant DLD-1 and TP53-null HCT116 (TP53^−/−) cells (Fig. 3C). We confirmed that RYBP bands were visible in cells transfected with the empty vector in preliminary experiments (Fig. S1C). Protein expression normalised to β-actin is shown in Fig. S3A-C. Corresponding β-actin from same membrane is shown in Fig. S4A and B.

In the apoptosis assay, knockdown of RYBP significantly decreased the number of HCT116 (TP53^wt/wt) cells in early apoptosis, whereas RYBP overexpression significantly increased the number of apoptotic cells (Fig. 3B). However, the number of apoptotic HCT116 (TP53^−/−) cells was not affected by modulation of RYBP expression (Fig. S2B). Western blot analysis revealed that knockdown of RYBP significantly reduced expression of Bax, a p53-related pro-apoptotic protein, and conversely, RYBP overexpression induced Bax expression (Figs. 3C and S3D).

To evaluate the effect of RYBP on oxaliplatin sensitivity, cell viability assay was used to calculate IC₅₀ value. In HCT116 (TP53^wt/wt) cells, RYBP knockdown significantly increased viability and the IC₅₀ value of oxaliplatin was significantly higher in cells treated with siRNA-1 and siRNA-2 than in the control (Fig. 4A). By contrast, RYBP overexpression decreased cell viability and the IC₅₀ was significantly lower in cells treated with pcDNA-RYBP than in the control (Fig. 4B). Low-dose (4.0 μM)oxaliplatin increased the number of cells in G1 phase and decreased the number of cells in the S phase. Higher doses (10 μM) of oxaliplatin further decreased the number of cells in S phase, indicating that disruption of DNA replication leads to anti-tumour effects (Fig. 4C). The results from the cell cycle assay following treatment of cells with 10 μM oxaliplatin showed that RYBP knockdown significantly increased the number of cells in the S phase, whereas RYBP overexpression further decreased the number of cells in S phase (Fig. 4D). These data suggested that RYBP enhanced the sensitivity of CRC cells to oxaliplatin.