LINC02418 promotes colon cancer progression by suppressing apoptosis via interaction with miR-34b-5p/BCL2 axis

Human CRC cell lines including SW460, HCT116, HT-29, LoVo, Colo205, SW480 and normal colon epithelial cell line NCM460 were obtained from American Type Culture Collection (ATCC, Manassas, VA). Cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (Gibco, Carlsbad, CA), 100U/ml penicillin and 100 mg/ml streptomycin. Cells were normally maintained at 37 °C in an incubator with 5% CO² until use.

The sequence of BCL2 was inserted into vector pcDNA3.1 (Santa Cruz, Dallas, TX) for its ectopic expression in HCT116 and LoVo cell lines. Constructions, miR-34b-5p mimic, miR-34b-5p inhibitor and miR-NC mimic were delivered into CRC cells by Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) according to manufactures instructions.

Three shRNAs sequences targeting LINC02418 and negative control sequence were inserted into HuSH shRNA GFP Lenti Cloning Vector (Origene, Rockville, MD) following commercial guidance. The lentivirus was packaged using 293T cells following common protocols.

For CCK-8 experiment, the cells were seeded at density of 5 × 10⁴ cells per well on 48-well plate. Cells were harvested at 12 h, 24 h, 48 h and 72 h, and cell proliferation was assessed using cell counting kit-8 (Beyotime, Shanghai, CHA) according to the manufacturer’s instructions. The optical density (OD) at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, Waltham, MA).

For colony formation assay. Total number of 3000 cells were seeded in 6-well plates and maintained in RPMI 1640 medium containing 10% FBS. After culture of 14 days, cells were fixed with 4% paraformaldehyde for 15 min and then stained with 0.1% crystal violet (Sigma-Aldrich, St. Louis, MO) for another 5 min till manually counting of visible colonies.

Male BALB/c nude mice (5–6 weeks old) were maintained under specific pathogen-free facility and were manipulated according to protocols approved by the China-Japan Union Hospital of Jilin University. All the mice were randomly divided into four groups, and each group contained 5 mice.

HCT116-sh-LINC02418, HCT116-sh-NC, LoVo-sh-LINC02418 and LoVo-sh-NC cells (2 × 10⁶ cells) resuspended in 200 μl of medium were subcutaneously inoculated into nude mice. After 7 days post-injection, tumor size was measured every 3 days, and the tumor volumes were recorded. After 21 days post-injection, mice were sacrificed by cervical dislocation. Tumors were separated from mice and the weight of tumor was measured. The survival curve analysis was not involved in this experiment.

Cells migration assays were performed in a 24-well transwell chamber (CoStar, Badhoevedorp, Netherlands). Cells were plated and allowed to migrate through 8 μm-pore sized polycarbonate membrane. The chamber for invasion assay was pre-coated with 1 mg/ml Matrigel (Sigma-Aldrich, St. Louis, MO). A number of 5 × 10⁴ cells were added to the upper chamber of the transwells with FBS-free medium and the lower chamber was filled with 500 μl RPMI 1640 medium containing 10% FBS. After 24 h incubation, the cells were fixed by 4% formaldehyde for 10 min, stained by 0.1% crystal violet for 20 min. Images were captured under microscope.

Total RNA was extracted from clinical tissue and CRC cells by TRIzol reagent (Invitrogen, CA, USA). RNA reverse transcription for mRNA and miRNA was performed using Prime ScriptTM RT Master Mix (Takara, Otsu, Japan) and TaqMan MicroRNA Reverse Transcription system (Thermo Fisher Scientific, MA, USA), respectively. The quantitative PCR was carried out by SYBR Premix Ex Taq II (TaKaRa Biotechnology, Dalian, China). Commercial miRNA qRT-PCR primers for miR-34b-5p and U6 were purchased by RiboBio (Guangzhou, China). The available primers sequences were as follows: LINC02418-F: 5′-ATTTCCATGGCGTTTCTCAC, LINC02418-R: 5′-AGGCAGGAGAATTGCTTGAA; BCL2-F: 5′-GGCATCTTCTCCTTC CAG-3′, BCL2-R: 5′-CATCCCAGCCTCCGTTAT-3′; GAPDH-F:5′-ACAACTTTGGTATCGTGGAAGG-3′, GAPDH-R:5′GCCATCACGCCACAGTTTC-3′. Applied Biosystems 7900 Real-Time PCR System (Applied Biosystems, Foster City, CA) was used for RNA quantification assay. U6 and GAPDH were used as internal control of miRNAs and mRNAs, respectively. Fold change was calculated by the 2^−△△Ct method.

The fragments containing the binding sites or the mutated sites were synthesized and inserted into a pGL3-basic vector for dual luciferase assay. HCT116 and LoVo Cells were seeded in a 12-well plate and co-transfected with reporter plasmids and miR-NC/miR-34b-5p mimic. After 48 h, cells were harvested, dual-luciferase reporter assays were performed according to the protocol using a Dual-Luciferase Reporter Assay System (Promega, Madison, WI) on a GloMax 20/20 luminometer (Promega, Madison, WI).

HCT116 and LoVo cells reaching 70% confluency were fixed in 4% paraformaldehyde for 20 min at room temperature, followed by permeabilized treatment in 70% ethanol at 4 °C overnight. For cellular miR-34b-5p detection, FISH assay was conducted following previous procedures [26]. Specific Digoxigenin (DIG)–labeled locked nucleic acid (LNA) probe against miR-34b-5p was purchased from QIAGEN (Hilden, Germany). The 4′,6-diamidino-2-phenylindole (DAPI) (Invitrogen, CA, USA) was adopted to stain the cell nucleic.

The tumor samples from CRC patients were fixed in 10% formaldehyde, embedded in paraffin and then sectioned into slices. For IHC assay, tumor slices were firstly deparaffinized and rehydrated. After washing, slices were treated with H₂O₂ to reduce the endogenous peroxidase activities. Slices were then incubated with primary antibody against BCL2 (1:50, Abcam, Cambridge, UK) overnight at 4  °C. With three times washing in PBS, the slides were incubated with secondary streptavidin–horseradish peroxidase-conjugated antibody (1:3000, Abcam, Cambridge, UK) for 1 h, and reacted with 3,3-diaminobenzidine tetrahydrochloride (DAB) solution (Yeasen Biotech, Shanghai, China) for 5 min. Finally, slides were counterstained with hematoxylin solution (Beyotime Biotechnology, Shanghai, CHN) for 1 min, dehydrated and mounted with neutral gum.

Protein samples were separated on 10% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene fluoride membranes (PVDF) (Millipore, Billerica, MA) for protein bands detection. The membranes were blocked in 5% nonfat milk. Primary antibodies including anti-GAPDH (1:2000, Abmart, Shanghai, CHN), anti-BCL2 (1:500, Abcam, Cambridge, UK), anti-Caspase 9 (1:1000, Cell Signaling Techonolgy, MA, USA), anti-cleaved-Caspase 9 (1:500, Cell Signaling Techonolgy, MA, USA), anti-Caspase 3 (1:1000, Abcam, Cambridge, UK), and anti-cleaved-Caspase 3 (1:1000, Abcam, Cambridge, UK) were used for incubation overnight at 4 °C. The membranes were incubated with HRP-conjugated secondary antibody (1:3000, Jackson Immuno Research, PA, USA) at room temperature for 1 h and the protein bands were detected by Pierce Fast Western Blot Kit (Thermo Fisher Scientific, MA, USA).

Apoptosis assays of HCT116 and LoVo cells were performed using Annexin V-FITC/propidium iodide (AV/PI) Apoptosis Detection Kit (Abcam, Cambridge, UK) according to the commercial instruction. Cells were incubated with ice-cold 75% ethanol at 4 °C overnight, followed by resuspending in 500 µl of 1 × Annexin V Binding Buffer. Then, cells were stained with 5 μl Annexin V/FITC and 5 μl Propidium Iodide (PI) in the dark for 15 min at the room temperature. Apoptosis rates were examined and analyzed by flow cytometry using FACS Calibur (BD Biosciences, CA, USA).

To identify the lncRNAs expression profile in CRC patients, we examined the expressional level of lncRNAs in human CRC samples and normal intestinal tissue. By using the Cancer Genome Atlas (TCGA) database, we found LINC02418 abundance was significantly up-regulated in CRC samples (n = 478) when compared with normal tissues (n = 41) (Fig. 1a).To validate the results, LINC02418 level in 20 pairs of CRC samples and adjacent tissues were examined by RT-qPCR. Similarly, quantification test showed LINC02418 was highly expressed in 19 out of the 20 CRC patients (Fig. 1b). These results implied that expression of LINC02418 in CRC tissues was markedly higher than that in normal tissues. Moreover, the expression of LINC02418 is increased by 2–6 folds in SW620, HCT116, LoVo, Colo205, HT-29 and SW480 cell line when compared with normal colon epithelial cell line NCM460 (Fig. 1c). HCT116 and LoVo cell lines were selected for further experiments, because of the highest expression level of LINC02418.

Next, to study the clinical significance of LINC02418 in CRC, the correlation between LINC02418 expression and prognosis was assessed by Kaplan–Meier survival assays using 437 patients information collected from TCGA database. As shown in Fig. 1d and e, in both early (Stage 1–2) and late stages (Stage 3–4), CRC patients with high expressional level of LINC02418 had poorer overall survival (OS) rates, indicating that LINC02418 was one potential indicator for prognosis prediction, and LINC02418 may promote CRC progress.

In order to dissect the effect of LINC02418 on the biologic activity of CRC cell lines, HCT116 and LoVo cells with reduced LINC02418 levels were established by lentivirus infection. Based on the endogenous level of LINC02418, HCT116-sh-LINC02418#1 and LoVo-sh-LINC02418#1 were selected for further analysis (Fig. 2a). The CCK-8 assay revealed that compared with sh-NC transduced CRC cells group and blank control (without shRNA transducing) group, stable knockdown of LINC02418 statistically inhibited the proliferation of HCT116 and LoVo cells (Fig. 2b). The colony formation assay was carried out to further analyze the effect of LINC02418 on tumor cell growth. As revealed in Fig. 2c, stable down-regulation of LINC02418 significantly reduced the proliferation of HCT116 and LoVo cells. To evaluate whether LINC02418 affected CRC growth in vivo, subcutaneous tumor formation experiments were set up in nude mice. HCT116 and LoVo cells transduced with sh-LINC02418 or sh-NC were subcutaneously injected into the nude mice for 21 days. The tumor size and tumor weight were recorded at indicated times. As shown in Fig. 2d, e and Additional file 1: Fig. S1, tumor size and tumor weight of xenograft tumors from sh-LINC02418-transduced HCT116 and LoVo cells were obviously less than those from CRC cells transduced with sh-NC, which was consistent with the results of in vitro experiments. The transwell experiments showed that inhibition of LINC02418 significantly repressed mobility and invasiveness of HCT116 and LoVo cells (Fig. 2f). Taken together, these data suggested that LINC02418 contributed to CRC cell growth and metastasis both in vivo and in vitro.

It is documented that lncRNAs can exert their function trough acting as ceRNA of miRNAs [27]. The lncRNA-miRNA-mRNA regulatory network has been shown to be involved in the treatment of malignant tumors. To assess whether LINC02418 interacted with other miRNAs in CRC, the potential binding partner for LINC02418 was analyzed by online software StarBase v2.0. As shown in Fig. 3a, LINC02418 contained putative binding sequence for miR-34b-5p.

To confirm the binding between LINC02418 and miR-34b-5p, dual luciferase assays were performed in HCT116 and LoVo cell lines. The luciferase activity was markedly inhibited when LINC02418-WT and miR-34b-5p mimic were co-transfected into HCT116 and LoVo cells. Co-transfection of LINC02418-WT and miR-NC or co-transfection of LINC02418-MUT and miR-34b-5p mimic had no significant impact on luciferase activity (Fig. 3b). In addition, knockdown of LINC02418 in HCT116 and LoVo cells greatly enhanced the expression level of miR-34b-5p (Fig. 3c). Taken together, it could be found that LINC02418 negatively regulated miR-34b-5p expression in CRC cells.

Subsequently, we detected the relationship between LINC02418 and miR-34b-5p in clinical samples. In 20 pairs of CRC samples, LINC02418 expressional level negatively correlated with miR-34b-5p level (Fig. 3d, e). Moreover, fluorescence in situ hybridization (FISH) experiment showed that compared with CRC cells transduced with sh-NC, CRC cells transduced with sh-LINC02418 expressed higher level of miR-34b-5p, which provided evidence for the negative correlation between LINC02418 and miR-34-5p (Fig. 3f).

BCL2, the gatekeeper for cell apoptosis, was identified as one possible target for miR-34b-5p by using software TargetScan (Fig. 4a). Interestingly, previous articles reported that miR-34b-5p regulates multiple cellular processes including cell apoptosis and cell proliferation through participating in a several critical signal pathway like VEGF-A, BCL2 and Notch [28]. To determine the interaction between BCL2 and miR-34b-5p, wild type 3′UTR (containing miR-34b–5p recognition site) and the mutant 3′UTR of BCL2 were sub-cloned into luciferase reporter plasmids. Dual-luciferase assays showed that miR-34b-5p mimic transfection reduced luciferase activity in wild type construction but not in mutant type construction (Fig. 4b, c). In parallel, miR-34b-5p transfection reduced endogenous protein level of BCL2, while the addition of miR-34b-5p inhibitor significantly restored BCL protein expression in both HCT116 and LoVo cells (Fig. 4d). These results proved that BCL2 was one of the target genes of miR-34b-5p. Quantification of BCL2 in 20 pairs of CRC samples revealed that BCL2 expressional level was positively correlated with the amount LINC02418 (Fig. 4e, f). We randomly selected 3 pairs of CRC samples for further confirmation. As shown in immunohistochemical (IHC) and western blot assays, the protein level of BCL2 was greatly increased in CRC samples from patients (Fig. 4g, h).

To elucidate the effect of LINC02418 on miR-34b-5p/BCL2 axis, protein level of several apoptotic markers were detected in HCT116 and LoVo cells. Western blot analysis showed that knockdown of LINC02418 greatly enhanced the amount of cleaved-Caspase 9 and cleaved-Caspase 3, and strongly reduced BCL2 expression. Transfection of miR-34b-5p inhibitor into cells with defective expression of LINC02418 not only reduced the level of cleaved forms of Caspase 9 and Caspase 3, but also promoted protein expression of BCL2. Overexpression of BCL2 also reduced protein level of cleaved-Caspase 9 and cleaved-Caspase 3 (Fig. 5a). Subsequently, CCK-8 and colony formation assays showed inhibition of LINC02418 repressed cell growth in HCT116 and LoVo cells. However, transfection of miR-34b-5p inhibitor or overexpression of BCL2 abolished the effect of down-regulated LINC02418 on cell proliferation (Fig. 5b, c). Inversely, knockdown of LINC02418 dramatically elevated the apoptosis rate of CRC cells while addition of miR-34b-5p inhibitor and BCL2 reduced insufficient LINC02418 expression-promoted cell apoptosis (Fig. 5d). These findings demonstrated that LINC02418 regulated colon cancer cell proliferation through upregulating BCL2 expression via sponging miR-34b-5p.