RNA sequencing gene expression (Level 3) data were downloaded from The Cancer Genome Atlas-Colon Cancer (TCGA-COAD) database at the Genomic Data Commons data portal (https://portal.gdc.cancer.gov). Patients lacking RPL27 mRNA expression data were excluded. The dataset included 327 samples (286 colorectal tumors and 41 normal mucosa). The correlation between RPL27 and PLK1 mRNA expression was assessed by Pearson's correlation coefficient analysis.

Overall survival (OS) data (death event and survival time) of 228 patients with CRC were downloaded from the UCSD Xena Browser (xena.ucsc.edu). OS was calculated from date of diagnosis and mortality. For survival analysis, the median expression of RPL27 gene was used to divide the patients into high- and low-expression groups. Survival curves were determined by univariate Kaplan-Meier estimators and analyzed by log-rank test.

Human CRC cell lines HCT116 and HT29 were purchased from the Korean Cell Line Bank (KCLB; Seoul, South Korea). KCLB performed cell line authentication using STR profiling. Both cell lines were cultured in RPMI-1640 medium (Welgene, Inc.) supplemented with 1% penicillin/streptomycin and 10% fetal bovine serum (FBS) (Welgene, Inc.), in a 37°C humidified incubator with a mixture of 95% air and 5% CO₂.

Before siRNA transfection, cells were plated at a density of 30%. After 24 h, cells were transfected with 15 nM negative control (NC) or RPL27 siRNA for 5 h using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) and Opti-MEM (Thermo Fisher Scientific, Inc.) at room temperature. The sequences of NC siRNA (Bioneer) were as follows: 5′-ACG UGA CAC GUU CGG AGA A(UU)-3′ (sense) and 5′-UUC UCC GAA CGU GUC ACG U-3′ (antisense). The bracket indicates overhang. Three variants of RPL27-specific siRNA (RPL27-1 siRNA, cat. no. s12205; RPL27-2 siRNA, cat. no. 9273 and RPL27-3 siRNA, cat. no. 9361) were purchased from Ambion (Thermo Fisher Scientific, Inc.). The sequences of RPL27 siRNA were as follows: RPL27-1 sense, 5′-ACA AAA CUG UCG UCA AUA A-3′ and antisense, 5′-UUA UUG ACG ACA GUU UUG U-3′; RPL27-2 sense, 5′-GGA UGU CUU CAG AGA UCC U-3′ and antisense, 5′-AGG AUC UCU GAA GAC AUC C-3′ and RPL27-3 sense, 5′-GGU CAA GUU UGA AGA GAG A-3′ and antisense, 5′-UCU CUC UUC AAA CUU GAC C-3′.

Cell proliferation was assessed using MTT reagent according to the manufacturer's instructions (Duchefa Biochemie). HCT116 and HT29 cells (1×10³) were transfected with siRNAs, as aforementioned. After 96 h, cells were treated with MTT reagent for 1 h. DMSO was added to dissolve the purple formazan. The optical density was evaluated at a wavelength of 540 nm using an Asys IVM 340 microplate reader (Biochrom, Ltd.).

The mRNA expression was quantified using RT-qPCR. Total RNA in HCT116 or HT29 cells was purified and reverse-transcribed to cDNA using the RNeasy Plus Mini (Qiagen GmbH) and PrimeScript II 1st strand cDNA Synthesis kits (Takara Bio, Inc.), respectively, according to the manufacturer's protocol. Amplification of each target gene was performed using TB Green (Takara Bio) and corresponding pair of primers (RPL27 forward, 5′-TGG CTG GAA TTG ACC GCT AC-3′ and reverse, 5′-CCT TGT GGG CAT TAG GTG ATT G-3′; GAPDH forward, 5′-ACA TCA AGA AGG TGG TGA AG-3′ and reverse, 5′-GGT GTC GCT GTT GAA GTC-3′; polo-like kinase 1 (PLK1) forward, 5′-CTC AAC ACG CCT CAT CCT C-3′ and reverse, 5′-GTG CTC GCT CAT GTA ATT GC-3′; CD133 forward, 5′-AGT CGG AAA CTG GCA GAT AGC-3′ and reverse, 5′-GGT AGT GTT GTA CTG GGC CAA T-3′; NANOG forward, 5′-CGA TCT CCT GAC CTT GT-3′ and reverse, 5′-CAC GCC TGT AAA TCC CA-3′; CD44 forward, 5′-CTG CCG CTT TGC AGG TGT A-3′ and reverse, 5′-CAT TGT GGG CAA GGT GCT ATT-3′; OCT4 forward, 5′-CTT GAA TCC CGA ATG GAA AGG G-3′ and reverse, 5′-GTG TAT ATC CCA GGG TGA TCC TC-3′ and c-MYC forward, 5′-AAT GAA AAG GCC CCC AAG GTA GTT ATC C-3′ and reverse, 5′-GTC GTT TCC GCA ACA AGT CCT CTT C-3′) synthesized by Genotech. Thermocycling conditions were as follows: 95°C for 60 sec, followed by 45 cycles at 95°C for 10 sec, one cycle at 60°C for 10 sec and one cycle at 72°C for 10 sec. Relative mRNA expression was assessed using a LightCycler 96 and quantified using the 2^−ΔΔCq method (28) in LightCycler 96 software version 1.1 (Roche Diagnostics). In addition, mRNA levels were normalized to GAPDH as a control.

A total of 1×10³ HCT116 or HT29 cells transfected with NC or RPL27 siRNA were seeded onto 6-well plates and incubated in RPMI-1640 medium at 37°C for 7-10 days until viable cells propagated to colonies for quantification. The colonies were fixed with 100% methanol for 15 min at room temperature and stained with 0.5% crystal violet for 30 min at room temperature. The number of colonies including >50 cells was counted manually under a light microscope (1× magnification).

CRC cells were plated in 100-mm dishes and harvested following 72 h transfection with control or target siRNA. To monitor cell cycle progression and apoptosis induction, the collected cells were stained with propidium iodide solution containing RNase A and FITC Annexin V Apoptosis Detection kit I (both BD Biosciences). The proportion of cells in each cell cycle phase and proportion of apoptotic cells were analyzed using a FACS Canto II flow cytometer and quantified using FACSDiva software 7.0 (both BD Bioscience). The total percentage of apoptotic cells were estimated by counting early + late apoptotic cells.

To assess the therapeutic effect of RPL27 targeting in vivo, xenograft assay using 5 male BALB/c nude mice (age, 4 weeks; weight, 17.5-19.5 g; Orientbio) was performed. The housing conditions were as follows: 24°C, 50-60% humidity, 12:12 light/dark cycle and free access to food and water. HCT116 cells were transfected with 15 nM NC or RPL27-1 siRNA, as aforementioned. After 24 h, 1×10⁶ cells were resuspended in RPMI-1640 medium (Welgene) containing Matrigel (Corning, Inc.) and injected subcutaneously into the left or right flank of mice. Tumor size was measured for 21 days at 2-3 day intervals using a Vernier Caliper and the tumor volume was calculated as width² × length × 1/2 (n=5/treatment group). The mice were sacrificed at day 21 by CO₂ euthanasia with flow rate of 30% volume/min. Death was confirmed by loss of heartbeat, breathing and pupil response. The tumor weight was then evaluated. All animal experiments were approved (approval no. KM-2021-03R1) by Keimyung University Institutional Animal Care and Use Committee. Humane experimental endpoints were the occurrence of tumor >10% of animal body weight, tumor size >2,000 mm³ and weight loss >20% of animal body weight.

To compare the pattern of global gene expression between NC and RPL27-1 siRNA-treated CRC cells, RNA sequencing was performed as previously described (22). All sequencing data were deposited in the Gene Expression Omnibus (GEO) database (accession no. GSE78195). Differentially expressed genes were selected based on at least 2-fold in CRC cells with RPL27 depletion were using Bootstrap ANOVA method (29). P≤0.001 were considered statistically significant. The significantly overlapping functional pathways of differentially expressed genes were analyzed using ingenuity pathway analysis (IPA, www.ingenuity.com).

HCT116 and HT29 cells were suspended in RIPA buffer containing 0.01% protease and phosphatase inhibitor cocktail (both Thermo Scientific Fisher, Inc.) 48 h after siRNA transfection. The amount of protein was quantified by Pierce BCA protein assay kit. Equal amounts (50 µg) of total protein were separated by 10-15% SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Roche Diagnostics). The membranes were blocked at room temperature for 30 min with 5% milk/TBS + 0.05% Tween-20 and incubated overnight at 4°C with primary antibodies against RPL27 (cat. no. #PA5-88938; Invitrogen; Thermo Fisher Scientific, Inc.), PLK1 (cat. no. sc-17783; Santa Cruz Biotechnology, Inc.), phosphorylated (p-)CDC25C (cat. no. #9529; Cell Signaling Technology, Inc.), CDC25C (cat. no. sc-327; Santa Cruz Biotechnology, Inc.), CDK1 (cat. no. sc-54; Santa Cruz Biotechnology, Inc.), cyclin B1 (cat. no. sc-245; Santa Cruz Biotechnology, Inc.), β-actin (cat. no. sc-47778; Santa Cruz Biotechnology, Inc.), Flag (cat. no. F1804; Sigma Aldrich; Merck KGaA) and CD133 (cat. no. #64326; Cell Signaling Technology, Inc.). Horseradish peroxidase-conjugated anti-mouse (cat. no. 115-035-062; Jackson ImmunoResearch Laboratories, Inc.) and goat anti-rabbit IgG (cat. no. sc-2301; Santa Cruz Biotechnology, Inc.) were used as the secondary antibodies and incubated for 1 h at room temperature. Immuno-reactive bands were visualized using FUSIOM SOLO S (Vilber Lourmat) and quantified with ImageJ software (version 1.8.0, National Institutes of Health).

Migration and invasion assays were performed using Transwell migration assay (cat. no. #3422) and 24-well Matrigel invasion chambers (cat. no. #354480; both Corning, Inc.), respectively. siRNA-transfected cells were plated into the upper chamber with serum-free RPMI-1640 medium (Welgene, Inc.; 5×10⁴ cells/well) and incubated for 24 h at 37°C. The lower chamber were filled with 10% FBS-containing RPMI-1640 medium (Welgene, Inc.). The cells were fixed and stained with 0.5% crystal violet for 30 min at room temperature. Images were captured at 100× magnification under a light microscope.

To investigate the biological effects of RPL27 overexpression, HCT116 cells were cultured in a 24-well plate and transiently transfected with 0.8 µg/µl RPL27-expressing or pCMV6-entry vector (empty vector; both OriGene Technologies, Inc.) for 5 h using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) and Opti-MEM (Thermo Fisher Scientific, Inc.) at 37°C. After 24 h, RPL27 expression was compared by RT-qPCR and western blotting, as aforementioned.

CRC cells were suspended in serum-free DMEM/F-12 (containing 2% B27, 4 ng/ml insulin, 10 ng/ml basic FGF and 20 ng/ml EGF; all Invitrogen; Thermo Fisher Scientific, Inc.) and 5×10³ cells/well were plated on 24-well ultralow attachment plates (Corning, Inc.). Following 24 h culture at 37°C, 15 nM NC or RPL27-1 siRNA complexed with Lipofectamine RNAiMAX (Invitrogen; Thermo Fisher Scientific, Inc.) was added to CRC cells. Sphere size and number were evaluated by iSolution Lite software (version 26.1, IMT i-Solution, Inc.) under a light microscope 8 days after siRNA transfection.

CD133 MicroBead kit (cat. no. 130-100-857) and AutoMACS Pro (both Miltenyi Biotec, Inc.) were used to isolate CD133⁺ HT29 CSC populations from HT29 parental cell culture according to the manufacturer's instructions. Magnetically labeled CD133⁺ and unlabeled CD133⁻ cells were separately incubated with isotype control (mouse IgG2b APC conjugate; cat. no. 130-122-932; Miltenyi Biotec, Inc.) or CD133/2 antibody (anti-human APC conjugate; cat. no. 130-113-746; Miltenyi Biotec, Inc.). Separation efficiency was evaluated using a FACSCanto II flow cytometer and quantified using FACSDiva software 7.0 (both BD Biosciences).

All statistical analyses were repeated at least three times, except analysis of TCGA data and gene expression data. Unpaired Student's t test and one-way ANOVA with Dunnett's post hoc test were used for statistical comparison of two and multiple groups, respectively. Data are presented as the mean ± SEM. P<0.05 was considered to indicate a statistically significant difference.

RPL27 is upregulated in CRC tissue (27); the present study verified its overexpression by accessing TCGA-Colon adenocarcinoma big data (286 colorectal tumors vs. 41 normal mucosa). Expression of RPL27 in CRC tumors was significantly higher than in the normal mucosa (Fig. 1A). It was analyzed whether increased expression affected survival of patients with CRC using TCGA data. RPL27 overexpression resulted in poorer overall survival in the stage III and IV than in stage I and II groups (Fig. 1B and C), implying that the biological and clinical impact of RPL27 expression increased as CRC progressed.

To investigate the effect of RPL27 silencing on cell proliferation, HCT116 and HT29 cells were transfected with NC or RPL27 siRNA. To select the siRNA with the highest treatment efficacy and target gene silencing ability, three sequences of RPL27-specific siRNAs (RPL27-1, -2 and -3) were tested. After 96 h, all three RPL27 siRNAs caused significant proliferation suppression in both HCT116 and HT29 cells, as determined by microscopic observation and MTT assay (Fig. 2A and B). Proliferation of cells treated with RPL27 siRNAs was inhibited by >50% compared with the control group. RT-qPCR was used to measure mRNA levels of RPL27 after transfection. All three siRNAs significantly inhibited the expression of RPL27 and this was greatest following RPL27-1 siRNA transfection (Fig. 2C). Similarly, treatment with RPL27-1 siRNA efficiently decreased colony formation by ~90 and 86% in HCT116 and HT29 cells, respectively, after 7 and 9 days of target siRNA treatment (Fig. 2D). The ability of RPL27-3 siRNA to inhibit colony formation in HCT116 and HT29 cells was similar to that of RPL27-1 siRNA (Fig. 2E). RPL27-1 siRNA was selected for all subsequent experiments. These results suggested that RPL27 was functionally involved in CRC tumor cell proliferation.

It was investigated whether the inhibition of CRC cell proliferation via RPL27 silencing was elicited by decreased cell cycle progression and/or the induction of apoptotic cell death. Flow cytometry analysis confirmed that RPL27-silenced cells displayed an increased proportion of sub-G1 phase in both HCT116 and HT29 cells (Fig. 3A). As the sub-G1 population contains apoptotic cells (30), apoptosis assay was performed by staining CRC cells with Annexin V. This showed that targeting RPL27 increased the number of apoptotic cells at 72 h after transfection (Fig. 3B). These results indicated that RPL27 silencing-mediated inhibition of CRC cell proliferation occurred due to decreased cell cycle progression and apoptosis induction.

To investigate whether RPL27 targeting affects in vivo tumor growth, a xenograft mouse model was constructed. HCT116 cells were transfected with NC or RPL27-1 siRNA for 24 h and then injected into BALB/c nude mice. The size of the tumor was measured at 2-3 day intervals. After 14 days, tumors in the NC group were significantly larger than those in the RPL27 siRNA-group (Fig. 4A). After 21 days, a significant difference in weight was observed between the control and RPL27-silencing groups (Fig. 4A and B). These results showed that targeting RPL27 significantly decreased the growth of human CRC xenografts in vivo.

To determine molecular mechanisms by which RPL27 silencing induces phenotypical changes, RNA sequencing was performed to compare patterns of global gene expression in RPL27-silenced HCT116 and HT29 cells with those in control cells transfected with NC siRNA. When at least a 2-fold change was defined, global gene expression analysis revealed that RPL27 silencing resulted in up- and downregulation of 697 RNA transcripts in HCT116 and 3,023 RNA transcripts in HT29 cells (Fig. 5A). Overlapping these two gene sets generated 228 commonly dysregulated genes (69 up- and 159 downregulated genes), which were considered to be common RPL27-silencing signatures (Fig. 5A and B). Subsequent IPA showed that 228 mRNA transcripts were functionally enriched in the top five networks (Fig. S1). Expression of PLK1, which is known to play an important role in mitotic cell cycle progression and migration (31), was downregulated by RPL27 silencing (Fig. 5C). The decrease in PLK1 mRNA expression was validated by RT-qPCR in both HCT116 and HT29 cells (Fig. 5D). This association was confirmed in TCGA data: RPL27 mRNA expression was positively correlated with PLK1 expression (R=0.255) in CRC tissue (Fig. 5E). These findings suggested that the function of RPL27 was associated with PLK1 expression.

Consistent with transcriptional profiling of RPL27-silenced cells, the protein expression of PLK1 was also decreased (Fig. 6A). PLK1 phosphorylates CDC25C at serine 198 residue for subsequent cyclin B1/CDK1 activation and mitotic entry (32). Here, silencing of RPL27 decreased the levels of p-CDC25C (ser198) and CDC25C (Fig. 6A). When normalized to CDC25C expression, silencing of RPL27 decreased phosphorylation of CDC25C by ~20 and 65% in HCT116 and HT29 cell lines, respectively, and decreased levels CDK1 and cyclin B1 protein, which are downstream effectors of PLK1 signaling that induce transition of G2/M (33). PLK1 silencing leads to decreased migration and invasion of various types of cancer, including CRC (31,34-36). Therefore, it was investigated whether targeting RPL27 inhibited metastatic potential of CRC cells using migration and invasion assay. Silencing of RPL27 effectively decreased migration (Fig. 6B) and invasion (Fig. 6C) abilities of both HCT116 and HT29 cells. Next, it was determined whether RPL27 overexpression could reverse the effect of RPL27 silencing on CRC migration and if RPL27 silencing could inhibit CRC stemness. Overexpression assay was performed in HCT116 cells as the basal expression of RPL27 protein in HCT116 cells was lower than that in HT29 cells (Fig. S2A). As expected, HCT116 cells transfected with RPL27-expressing vector (Fig. S2B and C) exhibited improved migration ability compared with control vector-transfected cells (Fig. 6D). The levels of CSC marker CD133 and the efficiency of CSC sphere formation are associated (37). To verify the functional involvement of RPL27 in CRC stemness, HT29 cells were used as expression of CD133 in HT29 cells is significantly higher than in HCT116 cells (38). RPL27-1 siRNA suppressed sphere formation in HT29 parental cells compared with NC siRNA (Fig. 6E). PLK1 inhibitor decreased CD133⁺ CSC proliferation (39). Given the functional significance of PLK1 signaling in CSC (40) and suppressed PLK1 expression following RPL27 silencing, it was investigated whether RPL27 targeting affected CSC proliferation and stemness. CD133⁺ CSC population of HT29 cells were sorted from the CD133⁻ population by FACS using APC-conjugated CD133 antibody (Fig. 7A); successful separation was confirmed by demonstrating that expression of CD133 mRNA in the CD133⁺ population was higher than that in the CD133⁻ population (Fig. 7B). RPL27 silencing decreased levels of CD133 and PLK1 in CD133⁺ CSCs (Fig. 7C and D) and significantly decreased the proliferation rate and sphere-forming capacity of CD133⁺ cells (Fig. 7E and F). Taken together, these data indicated that RPL27 was functionally associated with CRC stemness via PLK1 signaling.