Human colorectal cancer cell lines, HCT116 and HT29, were purchased from the Korean Cell Line Bank (KCLB; Seoul, Republic of Korea). Both HCT116 and HT29 cells were cultured in RPMI-1640 medium (Corning Incorporated, Corning, NY, USA) supplemented with 10% of fetal bovine serum (FBS; HyClone Laboratories, Logan, UT, USA) and 1% of penicillin/streptomycin solution (HyClone Laboratories), in a 37°C humidified incubator with a mixture of 95% air and 5% CO₂. The phenotypes of these cell lines have been authenticated by the KCLB. Cells were plated at 30% density 24 h before transfection. A total of 15 nM siRNA was mixed with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) in Opti-MEM (Invitrogen) and the medium was replaced 6 h after transfection. RPL9 siRNA duplexes (si-RPL9) were chemically synthesized by Ambion (Austin, TX, USA; siRNA ID# s226955). The negative control siRNA (si-NC) that does not target any endogenous transcript was used for control experiments. The sequences of si-NC (Bioneer, Daejeon, South Korea) are as follows: 5′-ACGUGA CACGUUCGGAGAA(UU)-3′ (sense) and 5′-UUCUCCGAAC GUGUCACGU-3′ (antisense).

Cell growth was measured using the Cell Counting kit-8 (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's instructions. Log phase cells were trypsinized into single cell suspension and HCT116 (2×10³ cells) and HT29 (1×10³ cells) were seeded into 96-well plates. After 24 h, cells were transfected as described above and after day 1, day 2, day 3 and day 4 of cell culture, the CCK-8 solution was added into each well. After 2 h, OD value was measured at a wavelength of 450 nm using a VersaMax microplate reader (Molecular Devices, Sunnyvale, CA, USA).

The changes in target gene expression were detected using RT-qPCR. Total RNA was isolated using TRIzol (Ambion) and synthesized into cDNA by using first strand cDNA synthesis kit (Takara Bio, Shiga, Japan) according to the manufacturer's instruction. cDNAs were then amplified using corresponding pair of primers (RPL9 forward, 5′-GCACAG TTATCGTGAAGG GC-3′ and RPL9 reverse, 5′-TTACCC CACCATTTGTCAA CC-3′; GAPDH forward, 5′-GGGAGCCAAAAGGGTCATCA TCTC-3′ and GAPDH reverse, 5′-CCATGCCAGTGAGCTT CCCGTTC-3′) synthesized by Macrogen (Seoul, Republic of Korea). The relative quantification of mRNA was measured by LightCycler 96 (Roche, Basel, Switzerland) according to the manufacturer′s instructions and quantified using LightCycler 96 software version 1.1, comparing with the Ct (threshold cycle) values of each target gene. The mRNA levels of GAPDH were used for normalization.

siRNA transfected cells were seeded into three independent wells of 6-well culture plates (1×10³ cells/well) and cultured at 37°C in 5% CO₂. Cells were maintained without medium change to let the viable cells propagate to sizable colonies for quantification. The colonies were fixed with methanol and then stained with 0.5% crystal violet for 30 min at room temperature. The number of colonies formed in each well was counted under the microscope and statistically analyzed.

Cells were cultured in 60-mm culture dishes and harvested at 72 h after siRNA transfection. Cells were washed with cold PBS, and then fixed 24 h with 70% cold ethanol at −20°C. Cells were washed with cold PBS again and incubated in a dark condition with propidium iodide (PI) staining solution containing RNase A (BD Biosciences, San Diego, CA, USA) for 30 min at room temperature. The cell cycle was measured by FACSVerse flow cytometry (BD Biosciences) according to the manufacturer's instructions and quantified using FlowJo software program.

Cells were cultured in 60-mm culture dishes and harvested at 48 and 72 h after siRNA transfection. Apoptotic cells were stained with Annexin V and PI using FITC Annexin V apoptosis detection kit I (BD Biosciences) following the manufacturer's instruction. The cell death was measured by FACSVerse flow cytometry (BD Biosciences) and quantified using FlowJo software program.

Total RNA was extracted 48 h after siRNA transfection using the RNeasy mini kit (Qiagen, Valencia, CA, USA). The quantity of the total RNA was evaluated using RNA electropherograms (Bio-Rad Experion; Bio-Rad Laboratories, Hercules, CA, USA); RNA quality was assessed based on the RNA quality indicator (RQI). The total RNA from each sample with a RQI value of 8.0 or higher was used. The resulting mRNA samples were processed for sequencing libraries using the Illumina TruSeq Stranded mRNA sample preparation kit (Illumina, San Diego, CA, USA) according to the manufacturer's protocols. RNA sequencing was performed using the Illumina HiSeq 2500 to generate non-directional, paired-end 100-base-pair reads. Quality-filtered reads were mapped to the human reference genome sequence hg19 (UCSC Genome Bioinformatics, https://genome.ucsc.edu) using TopHat2 (http://ccb.jhu.edu/software/tophat). The relative transcript abundance was estimated by counting the fragments per kilo-base of the exon model per million mapped sequence reads (FPKM), and differentially expressed genes were evaluated using the cufflinks package (http://cole-trapnell-lab.github.io/cufflinks). The significantly overlapping pathways and Gene Ontology categories with differentially expressed genes were analyzed using DAVID (http://david.abcc.ncifcrf.gov) and IPA (ingenuity pathway analysis, www.ingenuity.com).

Cells were suspended in RIPA buffer (Thermo Fisher Scientific, Inc., Rockford, IL, USA) containing 0.01% of a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific) at 48 h after siRNA transfection. The amount of protein was quantified by using the Pierce BCA protein assay kit (Thermo Fisher Scientific). Nuclear-cytoplasmic fractionation was conducted using the NE-PER Nuclear and Cytoplasmic Extraction Reagents kit (Thermo Fisher Scientific) according to the manufacturer's protocol. Equal amounts of total proteins were fractionated by SDS-PAGE on a 10% gel and transferred to PVDF membranes (Roche, Basel, Switzerland). 0.1% naphthol blue black (NBB) (Sigma-Aldrich, Seoul, Republic of Korea) was used to stain PVDF membrane to confirm equal sample loading. The membrane was blocked with 5% milk/Tris-buffered saline plus Tween-20 (TBST) and incubated with primary antibodies against human Id-1 (sc-133104), PARP-1 (sc-8007), NF-κB p65 (sc-372), Lamin B1 (sc-30264), β-actin (sc-47778) (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA), caspase-3 (#9662; Cell Signaling Technology, Danvers, MA, USA), Phospho-IκBα (Ser32/36) (#9246; Cell Signaling Technology), GAPDH (NB600-502; Novus Biologicals LLC, Littleton, CO, USA), RPL9 (ab182556; Abcam, Cambridge, MA, USA). HRP goat anti-mouse IgG, HRP goat anti-rabbit IgG and HRP rabbit anti-goat IgG (Santa Cruz Biotechnology) were used as the secondary antibodies. Immunoreactive bands were visualized with an LAS-3000 Imager (Fujifilm Corp., Tokyo, Japan).

Tumor formation was assessed to define the effects of RPL9 silencing on tumorigenicity in vivo. HCT116 cells were transfected with si-NC or si-RPL9 using Lipofectamine 2000. Twenty-four hours after transfection, cells were harvested by trypsinization and then they were washed and resuspended in RPMI mixed with Matrigel (Corning Incorporated). Cells (1×10⁶) transfected with si-NC and si-RPL9 were injected subcutaneously into the left and right flank, respectively, of 4-week old male BALB/c nude mice (Orient Bio, Inc., Seongnam, Korea). Tumor formation was monitored at 2-day intervals for 14 days after tumor injection. The tumor size was measured using a vernier caliper and calculated as (width² × length × 1/2). The mice were then sacrificed and tumor weight of each mouse was evaluated. All the animal experiments were performed in accordance with the guidelines of IACUC (Institutional Animal Care and Use Committee) in Yonsei University Health System with the approval number 2015–0066.

All the statistical analysis was performed using Student's t-test, except for in vivo experiment data which was applied with Bootstrap t-test with 10,000 random repetitions. All data are shown as means ± SEM. P-values ≤0.05, ≤0.01 and ≤0.001 were considered as statistically significant.

To test the effect of RPL9 knockdown on colorectal cancer cell survival, we transfected HCT116 and HT29 cells with si-NC or si-RPL9. After 96 h, we found from microscopic observation that the cell viability of RPL9 knockdown cells had decreased notably (Fig. 1A). In addition, we observed the maximum growth suppression of ~60–80% in both HCT116 and HT29 cells after 4 days of treatment (Fig. 1B). Concordance with the phenotypic assay result, si-RPL9 effectively silenced target gene expression in both the examined tumor cell lines (Fig. 1C). Consistent with the short-term results, RPL9 knockdown effectively suppressed the long-term colony formation of HCT116 and HT29 cells at day 7 and day 10, respectively, by ~60% inhibition in both cell lines (Fig. 2). These data indicate that RPL9 is functionally involved in colorectal cell growth.

To further investigate the role of RPL9 in colorectal cancer growth, the effect of RPL9 silencing on cell cycle progression of HCT116 and HT29 cells were analyzed with flow cytometry (Fig. 3A). It revealed that, when compared to the control treatment, RPL9 knockdown cells showed increased percentage of sub-G1 population and a concomitant decrease in the G1-phase cells in both HCT116 and HT29 cells (Fig. 3B). Since sub-G1 population represents apoptotic cells, we then confirmed the phenomenon by staining CRC cells with the apoptotic marker Annexin V using FACS. Consequently, we affirmed that siRNA knockdown of RPL9 significantly induced the apoptotic cell death at 72 h after transfection (Fig. 4A). Moreover, the percentages of early-apoptotic cell populations (Q3 region) plus late-apoptotic cell populations (Q2 region) in HCT116 and HT29 cells have increased in both 48 and 72 h after transfection (Fig. 4B). Early apoptotic cells are considered as cells that have intact plasma membrane which expose phosphatidylserine (PS) on the surface and late apoptotic cells have permeabilized plasma membrane (16). These results suggest that RPL9 knockdown induces apoptosis in colorectal cancer cells.

To explore the molecular basis of the growth inhibition caused by RPL9 silencing, we compared the global gene expression profiles of RPL9 knockdown HCT116 and HT29 cells to those of control cells transfected with si-NC using next-generation RNA sequencing. The global gene expression analysis, defined as at least a 2.0-fold change, revealed that RPL9-specific knockdown resulted in the upregulation and downregulation of 918 RNA transcripts in HCT116 and 3178 transcripts in HT29 cells (Fig. 5A). Comparing these two gene sets generated a statistically significant overlap of 296 genes (128 upregulated and 168 downregulated genes) which are considered as a common RPL9 knockdown signature (Fig. 5B). Subsequent IPA showed that the 269 mRNA transcripts were functionally enriched in the top five networks (Table I). In particular, as shown in Fig. 5C, we found that the expression level of Id-1, which is involved in carcinogenesis (17) and NF-κB activation (18), was downregulated in mRNA level by RPL9 silencing. Consistent with our observation, it was previously demonstrated that Id-1 silenced colorectal cancer cells resulted in decreased proliferation and increased apoptotic rate (15). Subsequent western blot analysis confirmed that the protein level of Id-1 was decreased in RPL9 knockdown HCT116 and HT29 cells when compared to the control (Fig. 6A). In addition, a concomitant decrease in pro-caspase 3, which activate apoptotic signaling, and PARP-1, a nuclear enzyme essential for genomic stability, were observed (Fig. 6A). Meanwhile, IκBα is an inhibitory protein that binds with NF-κB in the cytosol making NF-κB into an inactivated state. However, when IκBα becomes phosphorylated it dissociates from NF-κB. Then the activated NF-κB translocates into the nucleus where it binds to the specific sequences of DNA causing changes in cell function such as promoting cell survival (19). We observed that the protein level of p-IκBα is decreased by RPL9 silencing (Fig. 6A), and NF-κB protein level also decreased in nucleic fraction of the RPL9 knockdown colorectal cancer cells (Fig. 6B). Since the expression level of housekeeping proteins may vary depending on the experimental condition (20,21), equal sample loading was confirmed by naphthol blue black staining of the western blot membrane (Fig. 6B). These findings indicate that the function of RPL9 is associated with Id-1/NF-κB signaling pathway in colorectal cancer cells.

To further validate the effects of RPL9 knockdown on tumorigenesis of colorectal cancer, tumor formation assay was performed in BALB/c nude mice. si-NC and si-RPL9 transfected HCT116 cells were subcutaneously inoculated into the left and right flank, respectively. Tumor sizes were measured in 2-day interval for 14 days. Eight days after the inoculation, tumor sizes between the control and RPL9 knock-down showed definite difference and that the control tumor sizes were larger (P<0.05) (Fig. 7A). After 14 days, a significant difference (P<0.001) was apparent in size and weight between the control and RPL9 knock-down (Fig. 7). The results show that silencing of RPL9 significantly suppresses the growth of human CRC xenografts in vivo.