Linc00659, a long noncoding RNA, acts as novel oncogene in regulating cancer cell growth in colorectal cancer

Eight human CRC cell lines, namely DLD-1, colo205, colo320DM, LS174T, HCT116, SW620, LOVO, and HT29, were obtained from the American Type Culture Collection. Cells were cultured in high-glucose Dulbecco’s modified Eagle’s medium (Invitrogen, Grand Island, NY, USA) with 10% fetal bovine serum (Hyclone Laboratories, Inc., South Logan, UT, USA) and 1% penicillin/streptomycin (Gibco, Thermo Fisher Scientific Inc., Waltham, MA, USA) in plastic tissue culture plates in a humidified atmosphere containing 5% CO₂ at 37 °C.

Colon cancer tissue and the corresponding adjacent normal mucosa samples were obtained from 91 patients with CRC who underwent surgery at the Department of Surgery, Taipei Veterans General Hospital, Taiwan. Informed consent was obtained from all patients.

Total RNA from tissue was extracted using TRIzol reagent (Invitrogen, Grand Island, NY, USA), according to instructions in the user’s manual. In brief, tissue samples were homogenized in 1 mL TRIzol reagent and mixed with 0.2 mL chloroform to extract protein, before RNA was precipitated using 0.5 mL isopropanol. The concentration, purity, and amount of total RNA were determined using a Nanodrop 1000 spectrophotometer (Nanodrop Technologies Inc., Wilmington, DE, USA).

The transcriptome expression profiles of colon cancer were downloaded from The Cancer Genome Atlas (TCGA) data portal (https://​cancergenome.​nih.​gov). The expression profiles of 616 colon cancer tissues and 51 adjacent normal tissues were obtained from TCGA data portal. In this study, the transcriptome profiles of 29 N-T pairs were used for coexpression analysis and 616 cases were included in the survival analysis.

In this reaction, 2 μg of total RNA was reverse transcribed with random primers (Invitrogen; Thermo Fisher Scientific Inc., Waltham, MA, USA) and SuperScript IV Reverse Transcriptase according to the user’s manual (Invitrogen; Thermo Fisher Scientific Inc., Waltham, MA, USA). The reaction was performed with incubation at 42 °C for 1 h, and the enzyme was subsequently inactivated by incubation at 85 °C for 5 min. cDNA was used for real-time polymerase chain reaction (PCR) analysis with gene-specific primers, and gene expression was detected using a Fast SYBR Green Master Mix (Applied Biosystems; Thermo Fisher Scientific Inc., Waltham, MA, USA). The expression of lncRNA was normalized to that of glyceraldehyde 3-phosphate dehydrogenase (GAPDH; △Ct = target lncRNA Ct – GAPDH Ct). The individual primers used in this study were as follows:

Small interfering RNA (siRNA) oligonucleotides targeting Linc00659 (si-Linc00659, sense: 3′-TTGGGAGGAACACGAAGUCUU-5′ and antisense: 5′-UUCUGAAGCACAAGGAGGGTT-3′) and a scrambled oligo as a negative control were designed and synthesized by GenDiscovery Biotechnology (Taipei, Taiwan). Cells were transfected with a final concentration (10 mM) of individual siRNA or control using Lipofectamine RNAiMAX (Invitrogen; Thermo Fisher Scientific Inc., Waltham, MA, USA). After transfection for 24 h, the RNA was extracted and knockdown efficiency was evaluated by real-time PCR.

Stable HCT116 cells with Linc00659 knockdown were generated by infecting colon cancer HCT116 cells with lentiviruses expressing sh-Linc00659 in the presence of 8 μg/mL polybrene for 24 h followed by puromycin (4 μg/mL) selection for 3–5 days. The shLuc vector targeting the luciferase gene provided puromycin resistance and was used as the control. Linc00659 expression was confirmed by a real-time PCR assay. In this study, we designed two shRNA sequence targeted Linc00659, and individual sequence of shRNA used for constructs in this study were as follows:

For cell proliferation analysis, 2500 living cells were transfected with either siLinc00659 or a scrambled control and plated onto 96-well plates. Cell growth was determined at 0, 1, 2, 3, and 4 days. Cell viability was determined using a CellTiter-Glo One Solution cell proliferation assay (Promega Corporation, Madison, WI, USA), which was performed according to the manufacturer’s instructions.

In brief, this assay was performed as follows: The base agar layer was prepared with 1% agar (Laboratorios CONDA, Torrejón de Ardoz, Madrid, Spain) dissolved in 1.5 mL phosphate-buffered saline on 6-well plates; subsequently, 1.5 mL of 0.7% agarose (Invitrogen, Grand Island, NY, USA) solution containing the cells (10,000 cells per well) was inoculated on top of the base agar layer. After allowing the solution to harden, 0.5 mL of fresh medium was added to the top of the hard agar layer. After 3–4 weeks, agar plates were stained with iodonitrotetrazolium chloride (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and colonies were counted.

A total of 4000 cells were seeded on each well of a 6-well plate, and the cells were transfected with individual siRNA oligonucleotides by using Lipofectamine RNAiMAX (Invitrogen; Thermo Fisher Scientific Inc., Waltham, MA, USA) in the 6-well plate. After incubation at 37 °C for 3 days, the cultured medium was replaced with a new medium. The cells were allowed to incubate at 37 °C for 10 days. Cell culture plates containing colonies were fixed with 4% formaldehyde for 2 min and colonies were stained with crystal violet solution (containing 0.5% crystal violet, 5% formaldehyde, 50% ethanol, and 0.85% sodium chloride) for 2 h. Wells were rinsed with H₂O after air drying. The crystal violet staining of cells from each well was solubilized using 1 mL per well of 10% acetic acid, and the absorbance (optical density) of the solution was measured on a spectrophotometer at a wavelength of 595 nm.

A total of 1 × 10⁶ cells were collected and mixed with 70% ethanol in a fixative at − 20 °C overnight. Cells were then stained with 4′,6-diamidino-2-phenylindole (ChemoMetec, Gydevang, Lillerød, Denmark) and detected using NucleoView NC-3000 software (ChemoMetec, Gydevang, Lillerød, Denmark).

For the image flow cytometry assay, CF488A Annexin V and PI Apoptosis kit (Biotium, Fremont, CA, USA) and Hoechst 33,342 (ChemoMetec, Gydevang, Lillerød, Denmark) were used to detect cells in early apoptotic and late apoptotic stages. After staining, the cells were analyzed in NucleoCounter NC-3000 (ChemoMetec, Gydevang, Lillerød, Denmark) and NucleoView NC-3000 for automated image flow cytometry analysis.

The cells were harvested 24 h after transient transfection, washed with phosphate-buffered saline, and subjected to lysis in radioimmunoprecipitation assay buffer (50 mM Tris–HCl pH 8.0, 150 mM NaCl, 1% NP40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate) at 4 °C for 30 min. Lysed cells were collected and centrifuged to remove cell debris. Protein assays were performed with the Bio-Rad Protein Assay kit based on the Bradford dye-binding procedure (Bio-Rad). Protein samples (40 μg) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis in 10% or 12% resolving gel using a Mighty Small II Deluxe Mini Vertical Electrophoresis Unit (Hoefer, Inc., Holliston, MA, USA). Proteins were then electrotransferred to polyvinylidene difluoride membranes (PerkinElmer, Inc., Waltham, MA, USA) by Mighty Small Transfer Tank (Hoefer, Inc., Holliston, MA, USA). Subsequently, the membranes were blocked with a blocking buffer (50 mM Tris–HCl pH 7.6, 150 mM NaCl, 0.1% Tween 20, 5% nonfat dry milk, 0.05% sodium azide) for 1 h at room temperature and incubated with the following primary antibodies overnight at 4 °C: CCNA2 (1:1000; 18,202–1-AP, Proteintech Group, Inc., Rosemont, IL, USA), CCNB1(1:1000; 55,004–1-AP, Proteintech Group, Inc., Rosemont, IL, USA), CCND1 (1:1000; RM-9104-S, Thermo Fisher Scientific Inc., Waltham, MA, USA), CDK4 (1:1000; MS-299-P, Thermo Fisher Scientific Inc., Waltham, MA, USA), CDKN1B (1:1000; #3686, Cell Signaling Technology, Inc., Beverly, MA, USA), CDKN1A (1:1000; #2947, Cell Signaling Technology, Inc., Beverly, MA, USA), PARP (1:1000; sc-8007, Santa Cruz Biotechnology, Inc. Santa Cruz, CA, USA), CASP3 (1:1000; #9662, Cell Signaling Technology, Inc., Beverly, MA, USA), E2F1 (1:200, sc-251, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), PI3K p85 (1:1000; #4292, Cell Signaling Technology, Inc., MA, USA), AKT (1:1000; #4691, Cell Signaling Technology, Inc., MA, USA), Phospho-AKT (Thr308) (1:1000; #4056, Cell Signaling Technology, Inc., MA, USA), Phospho-AKT (Ser473) (1:1000; #4060, Cell Signaling Technology, Inc., MA, USA), GSK3β (1:1000; #9315, Cell Signaling Technology, Inc., MA, USA), Bcl-2 (1:1000; #4223, Cell Signaling Technology, Inc., MA, USA), BAX (1:1000; #5023, Cell Signaling Technology, Inc., MA, USA), Bad (1:1000; #9239, Cell Signaling Technology, Inc., MA, USA), Phospho-Bad(Ser136) (1:1000; #4366, Cell Signaling Technology, Inc., MA, USA), β-catenin (1:4000; C2206, Sigma-Aldrich, Inc., Beverly, MO, USA) and ACTB (1:2000, MAB1501, EMD Millipore, Billerica, MA, USA). The membranes were then incubated with anti-rabbit (sc-2004) or anti-mouse (sc-2005) immunoglobulin G horseradish peroxidase-conjugated secondary antibodies (1:10000, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) for 1 h at room temperature. After three washes with Tris-buffered saline with Tween-20 buffer (50 mM Tris–HCl pH 7.6, 150 mM NaCl, 0.1% Tween-20), immunoreactive bands were detected using a WesternBright ECL (Advansta, Menlo Park, CA, USA).

The expression levels of Linc00659 in colon cancer from TCGA database were analyzed using Student t tests. The correlation of Linc00659 with the protein-coding genes in colon cancer was determined through Pearson coefficient analysis, with r and p values as indicated. Cumulative survival curves were estimated using the Kaplan–Meier method, and comparison between survival curves was performed using the log-rank test. The difference was considered significant when p < 0.05. The data on the expression levels of lncRNAs in paired colon tissues were analyzed using a paired t test. Cell proliferation, colony formation, and soft agar assay experiments were performed in triplicate. Histograms present the mean values, and the error bars indicate the standard deviation. These data were analyzed using Student t tests.

We determined the gene expression profiles of the primary tumor and the corresponding adjacent normal tissues in two patients with CRC by using a microarray approach, in which protein-coding genes and lncRNAs were included (Agilent SurePrint G3 Human V2 GE; Fig. 1a). Most genes were consistently expressed between the CRC and the corresponding adjacent normal tissues in the two patients (R² > 0.86; Fig. 1b and c). In total, 5790 protein-coding genes were detected as differentially expressed (fold change ≤ 2 or ≥ 0.5) between CRC and the corresponding adjacent normal mucosa, comprising 4866 upregulated genes and 924 downregulated genes, respectively (Fig. 1d, upper panels). We further examined the putative biological function of these differentially expressed genes by using pathway enrichment analysis. Our data revealed that these differentially expressed genes were significantly enriched in cell cycle, DNA replication, RNA transport, and pyrimidine metabolism signaling pathways (Fig. 1d, bottom panel). Abnormal lncRNAs in CRC were also identified by analyzing the same microarray data. Consistent with protein data, most lncRNAs were consistently expressed between the CRC and the corresponding adjacent normal tissues in the two patients with CRC (R² > 0.83; Fig. 2a and b). Only a small subset of lncRNAs was differentially expressed between CRC and the corresponding adjacent normal tissues. A total of 201 lncRNAs were either upregulated (n = 165; fold change ≥ 4) or downregulated (n = 36; fold change ≤ 0.25) in the two CRC cases.

The top 10 of these abnormal lncRNAs upregulated and downregulated are presented in Table 1. To validate the microarray analysis results, 4 of the top 10 differentially expressed lncRNAs were selected for validation in 89 matched CRC samples by using reverse transcription-quantitative PCR. Our results indicated that the expression levels of Linc00659 and MNX1-AS1 were significantly increased, whereas those of Loc339524 and Linc00657 were significantly decreased in colon cancer (Fig. 2d–i), which were consistent with our microarray results.

The biological and clinical effects of Linc00659 expression were further examined by analyzing TCGA database. A total of 667 transcriptome data items were downloaded from TCGA database, comprising 51 normal and 616 CRC tissues. As shown in Fig. 3a, the expression levels of Linc00659 significantly increased in colon cancer compared with normal mucosa. Furthermore, high expression levels of Linc00659 were significantly correlated with a short survival curve of patients with CRC (Fig. 3b), implying that Linc00659 might play an oncogenic role in colon cancer progression. We further explored the biological role of Linc00659 in colon cancer tumorigenesis by identifying coexpression gene networks. A total of 1218 protein-coding genes were coexpressed with Linc00659 in colon cancer (r > 0.7 or <− 0.7; Fig. 4a). By determining these signaling pathways of the protein-coding genes coexpressed with Linc00659, we found that these highly interconnected genes were enriched in several cell proliferation-related signaling pathways (Fig. 4b). Notably, we also found that Linc00659 expression had a highly positive correlation with cell cycle-related gene expression in colon cancer, implying that Linc00659 might have an oncogenic role in colon cancer cell growth by accelerating cell cycle progression (Fig. 4c).

We examined the expression levels of Linc00659 in eight colon cancer cell lines, namely colo320dm, colo205, DLD1, Lovo, SW620, HCT116, LS174T, and HT29, by using real-time PCR. Compared with adjacent normal mucosa, the expression levels of Linc00659 were clearly overexpressed in eight colon cancer cell lines by more than 20-fold (Fig. 5a). Based on this result, we selected a loss-of-function approach to study the biological function of Linc00659 in colon cancer cells. As demonstrated in Fig. 5b, Linc00659 expression significantly decreased by 70% in HCT116 cells after siRNA transfection.. Knockdown of Linc00659 expression significantly suppressed cancer cell proliferation and colony formation in HCT116 (Fig. 5c–e). Using a soft agar assay, we indicated that Linc00659 knockdown could significantly suppress anchorage-independent growth in HCT116 cells (Fig. 5f and g). We also observed that the cell proliferation of Lovo, LS174T, DLD-1 and SW620 could be suppressed after Linc00659 knockdown for 4 days (Additional file 1: Figure S1). We further established stable knockdown of Linc00659 expression through sh-Linc00659 transfection and puromycin selection for 2 weeks. As illustrated in Fig. 6a and Additional file 2: Figure S2A, the expression levels of Linc00659 were significantly decreased by 60% in sh-Linc00659 compared with the control. The stable knockdown of Linc00659 expression exhibited similar effects on cell growth in accordance with si-Linc00659 treatment. The cell proliferation, colony formation, and anchorage-independent growth abilities were significantly decreased in HCT116 cells with stable Linc00659 knockdown (Fig. 6b–f and Additional file 2: Figure S2B-F).

Our results revealed that Linc00659 could regulate colon cancer growth. Coexpression data indicated that Linc00659 had a high correlation with cell cycle-related genes; therefore, we explored the mechanism underlying the regulation of cell growth by examining the cell cycle distribution and apoptosis after Linc00659 knockdown. Image flow assay results indicated that the cell cycle distribution was disrupted after Linc00659 knockdown. Transient knockdown of Linc00659 cells also indicated that the G₁ phase decreased in HCT116 cells (Fig. 7a and b). Furthermore, an examination of the cell cycle-related protein expression levels revealed that the expression levels of cyclin A2, cyclin B1, cyclin D1, CDK4, and E2F1 decreased in HCT116 cells with Linc00659 knockdown (Fig. 7c). Notably, the sub-G₁ population cells were obviously increased in Linc00659 cells subjected to transient knockdown, including HCT116 (Fig. 7a and b). Apoptosis assay results indicated that the apoptotic cells significantly increased and that PARP and caspase-3 cleavage was induced in HCT116 cells with si-linc00659 transfection (Fig. 7d–f). These data indicate that Linc00659 knockdown could impair cell cycle progression and induce cell apoptosis.

To explore whether Linc00659 expression is involved in the regulation of drug responsiveness, we assessed the effects of Linc00659 knockdown on cell apoptosis in HCT116 with drug treatment, including oxaliplatin, 5-fluorouracil (5-FU), and irinotecan treatment. Linc00659 was stably knocked down and the respective control cells were exposed to various concentration of oxaliplatin, 5-FU, or irinotecan for 48 h, and cell viability was examined. As presented in Fig. 8a, Linc00659 knockdown could obviously accelerate the reduction of cell viability in HCT116 following oxaliplatin or 5-FU, and slightly reduced cell viability with irinotecan treatment. An examination of the cell cycle progression revealed that Linc00659 knockdown resulted in a significant increase of the sub-G₁ phase of the cell cycle distribution after oxaliplatin and 5-FU treatment (Fig. 8b and c). In addition, Linc00659 knockdown increased apoptotic cells after oxaliplatin treatment compared to control group (Fig. 8d and e). Overall, the expression levels of Linc00659 conferred drug sensitivity in colon cancer by inducing apoptosis.

Studies have indicated that AKT activation plays a crucial role in the cell-cycle progression and drug resistance of colon cancer [22‐24]. Therefore, we sought to determine whether the P13K-AKT signaling pathway is involved in suppressing growth in HCT116 with Linc00659 knockdown. As shown in Fig. 9a, the protein levels of PI3K and p-AKT were decreased in HCT116 after Linc00659 knockdown. In addition, studies have indicated that AKT signaling acts has a crucial role in colon cancer cell survival by modulating BCl2 and Bad [25, 26]. Therefore, we examined these apoptosis-related proteins, revealing that the levels of anti-apoptosis proteins BCl2 and p-Bad both decreased after Linc00659 knockdown (Fig. 9b); we also observed that GSK3β expression was increased in Linc00659 knockdown compared with the control group. We further examined the expression levels of B-catenin, revealing that total protein and nuclear-B-catenin were decreased in HCT116 with Linc00659 knockdown (Fig. 9c). Taken together, our results demonstrate a putative mechanism of Linc00659 knockdown that impairs cell cycles and induces apoptosis; this may be a result of suppressing the PI3K-AKT-GSK3B axis in colon cancer.