The expression of FLVCR1-AS1 and CRC clinical data were downloaded from the TCGA database (https://tcga-data.nci.nih.gov/tcga/). Kaplan-Meier analysis and the log-rank test were used to analyze survival curves. Cut-off values were determined using mean expression level of FLVCR1-AS1.

A total of 26 pairs of CRC tissues and adjacent non-tumor tissues were obtained from patients with CRC (14 males and 12 females) with a mean age of 52 years (range, 34–82 years) between April 2017 and September 2018 at Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine. Written informed consent was obtained from all subjects and the study was approved by the Ethics Committee of Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine. The specimens were immediately cryopreserved in liquid nitrogen after surgery.

Four human CRC cell lines, namely Caco-2, SW480, LoVo and SW1116, the NCM460 normal colonic epithelial cell line and 293T cells were purchased from the American Type Culture Collection. All cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (both Gibco; Thermo Fisher Scientific, Inc.) at 37°C in a humified atmosphere containing 5% CO₂.

Short hairpin RNA (shRNA) targeting FLVCR1-AS1 (shFLVCR1-AS1) and its corresponding control shRNA (shNC), miR-381 mimics, 5′-UACAGUACUGUGAUAACUGAA-3′; NC mimics, 5′UUUGUACUACACAAAAGUACUG-3′; miR-381 inhibitor, 5′-UUCAGUUAUCACAGUACUGUA-3′; NC inhibitor, 5′-UCACAACCUCCUAGAAAGAGUAGA-3′. The full-length of FLVCR1-AS1 and RAP2A were subcloned into pcDNA3.1 to overexpress FLVCR1-AS and RAP2A levels, with empty pcDNA3.1 serving as control. The overexpression plasmid and pcDNA3.1 were bought from Shanghai GenePharma Co., Ltd. shRNA (100 nm), miRNA (50 nM) or plasmid (1 µg/ml) was transfected into cells using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). Subsequent experiments were performed at 48 h post-transfection.

Total RNA was isolated from tissues and cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Subsequently, total RNA was reverse transcribed into cDNA using the PrimeScript RT reagent kit (Takara Bio, Inc.) at 37°C for 15 min. RT-qPCR was performed on an ABI 7500 Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.) with SYBR-Green PCR Master Mix kit (Takara Bio, Inc.). The PCR amplification reaction was carried out using cDNA as template with the following conditions: 95°C for 10 min, and 40 cycles of 95°C for 15 sec, 62°C for 30 sec, and 72°C for 30 sec. Primers were as follows: FLVCR1-AS1 forward, 5′-GAAAGCTGCAACATGCTCCC-3′ and reverse, 5′-TCCATGTCGTCCCAGTTGGT-3′; miR-381 forward, 5′-GCAGGGCTTCTGAGCTCCTTAA-3′ and reverse, 5′-CAAATTCGTGAAGCGTTCCATAT-3′; RAP2A forward, 5′-ACACGTCTGGAGGAGACAGC-3′ and reverse, 5′-GAGAGGTTGTGCCGGATAGA-3′; GAPDH forward, 5′-ACCACAGTCCATGCCATCAC-3′ and reverse, 5′-TCCACCACCCTGTTGCTGTA-3′; U6 forward, 5′-CTCGCTTCGGCAGCACA-3′ and reverse, 5′-AACGCTTCACGAATTTGCGT-3′. The relative expression of genes was calculated using the 2^−ΔΔCq method (23), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and U6 were used as internal references.

Cell viability was evaluated using a Cell Counting Kit 8 (CCK-8; Dojindo Molecular Technologies, Inc.) assay. Briefly, 2×10⁴ cells/well were seeded into 96-well plates and 48 h following transfection, each well was supplemented with 10 µl CCK-8 reagent. Subsequently, the optical density value at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, Inc.).

An in situ Cell Death Detection kit (cat. no. 11684817910; Roche Diagnostics GmbH) was used to analyze cell apoptosis. Caco-2 and SW480 cells were washed with PBS, and fixed in 4% paraformaldehyde solution for 1 h at 4°C. The cells were permeabilized in a solution containing 0.1% Triton X-100 for 2 min and cultured with TUNEL reaction mixture (Roche Diagnostics GmbH) for 1 h at 37°C in the dark. The TUNEL-stained coverslips were then washed with PBS and then counterstained with DAPI (1:20 dilution; Beyotime Institute of Biotechnology) for 15 min at room temperature. A fluorescence microscope (magnification, ×100; Olympus Corporation) was utilized to observe TUNEL-positive cells in at least 5 fields of view.

The starBase database (http://starbase.sysu.edu.cn/) were used for predicting the binding sites between miR-381 and FLVCR1-AS1 or RAP2A. The wild-type (wt) or mutant (mut) FLVCR1-AS1/RAP2A was inserted into a pGL3.0 vector (Shanghai GenePharma Co., Ltd.). Subsequently, 293T cells were transfected with one of the above vectors and either miR-381 mimics or NC mimics using Lipofectamine 2000. Following incubation for 48 h, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega Corporation). Firefly luciferase activity was normalized to Renilla luciferase gene activity.

Biotin-labeled wt miR-381 (bio-miR-381-wt), mut miR-381 (bio-miR-381-mut) and miR-NC were obtained from GenePharma Co., Ltd. The biotinylated RNAs were transfected into Caco-2 and SW480 cells using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), and the cell lysates were incubated with streptavidin magnetic beads (Invitrogen; Thermo Fisher Scientific, Inc.) overnight. Following RNA extraction, the abundance of FLVCR1-AS1 was detected by RT-qPCR.

Cell migration and invasion abilities were assessed using Transwell chambers (8.0 µm pore size; BD Biosciences). For cell migration, 1×10⁵ transfected cells and 200 µl serum-free DMEM were added to the upper chamber with an uncoated membrane. Then, complete DMEM was added to the lower chamber. Following seeding for 48 h, cells in the upper chamber were removed, and those in the lower chamber were stained with 0.1% crystal violet (Sigma-Aldrich; Merck KGaA). For cell invasion, the membranes were precoated with Matrigel at 37°C for 2 h, and the other steps were the same as those carried out in the migration assay. Cell numbers were counted under a light microscope. (magnification, ×200; Zeiss GmbH).

Total protein extracts were isolated from transfected cells using RIPA lysis buffer (Beyotime Institute of Biotechnology) and protein concentrations were determined using a bicinchoninic acid protein assay kit (Pierce; Thermo Fisher Scientific, Inc.). Subsequently, 10 µg protein/lane was separated by 10% SDS-PAGE (Bio-Rad Laboratories, Inc.). The proteins were then electrotransferred onto a PVDF membrane. Following blocking with non-fat milk for 2 h at room temperature, membranes were first incubated with primary antibodies against RAP2A (1:1,000; cat. no. ab101369; Abcam) and GAPDH (1:1,000; cat. no. ab9485; Abcam) at 4°C overnight and then with corresponding secondary antibodies (1:1,000; goat anti-rabbit IgG, cat. no. ab205718; Abcam) for 1 h at room temperature. Finally, blots were visualized with an enhanced chemiluminescent detection system (EMD Millipore). Protein expression was evaluated using Image-Pro^® Plus software (version 6.0; Media Cybernetics, Inc.).

SPSS 19.0 (IBM Corp.) software was used to perform the statistical analyses, and results are expressed as the mean ± standard deviation. Each experiment was performed at least three times. Clinicopathological characteristics were evaluated using the χ² test. Paired or unpaired Student's t-tests, or one-way ANOVA followed by Tukey's post hoc test were used to compare the groups. The survival curve was generated using the Kaplan-Meier method and log-rank test with the ‘survival’ R package (version 3.4.1; http://bioconductor.org/packages/survivalr/). The patients were divided into a high and a low expression group according to the mean gene expression. The correlation between gene expression levels was analyzed using Pearson's correlation coefficient. P<0.05 was considered to indicate a statistically significant difference.

Firstly, the expression of FLVCR1-AS1 in CRC tissues was examined, and the results indicated that the FLVCR1-AS1 expression levels were significantly elevated in CRC tissues compared with the adjacent normal tissue (Fig. 1A). Consistent with this, FLVCR1-AS1 was found to be upregulated in CRC cells (Caco-2, SW480, LoVo and SW111) compared with normal colonic epithelial cells (Fig. 1B). Analysis of the expression of FLVCR1-AS1 according to different clinicopathological characteristics of the patients with CRC indicated that high levels of FLVCR1-AS1 were positively associated with TNM stage, distant metastasis and lymph node metastasis (Table I). Notably, the expression of FLVCR1-AS1 was significantly higher in CRC tissues of advanced TNM stages compared with earlier ones (Fig. 1C). Furthermore, Kaplan-Meier analysis showed that CRC patients with high levels of FLVCR1-AS1 exhibited a shorter survival time than those with low FLVCR1-AS1 levels (Fig. 1D). These findings suggest that FLVCR1-AS1 is involved in CRC tumorigenesis.

The biological role of FLVCR1-AS1 in CRC was further explored. The expression of FLVCR1-AS1 was significantly decreased in Caco-2 and SW480 cells following transfection with shFLVCR1-AS1 compared with shNC (Fig. 2A). CCK-8 and Transwell assays demonstrated that the knockdown of FLVCR1-AS1 attenuated CRC cell viability, migration and invasion (Fig. 2B-D). Furthermore, TUNEL assay revealed that FLVCR1-AS1 silencing induced Caco-2 and SW480 cell apoptosis (Fig. 2E). Subsequently, RT-qPCR analysis showed that FLVCR1-AS1 was efficiently overexpressed following transfection with an FLVCR1-AS1 overexpression plasmid (Fig. 2F). CCK-8 and Transwell assays indicated that transfection with pcDNA3.1/FLVCR1-AS1 increased the viability, migration and invasion of Caco-2 and SW480 cells (Fig. 2G-I). In addition, TUNEL assay revealed that FLVCR1-AS1 overexpression markedly attenuated CRC cell apoptosis (Fig. 2J). The aforementioned data indicate that FLVCR1-AS1 promotes CRC progression.

To explore the molecular mechanism of FLVCR1-AS1 in CRC, its downstream targets were identified. Using the starBase online tool, the binding sequence between miR-381 and FLVCR1-AS1 was predicted (Fig. 3A). RT-qPCR analysis showed that miR-381 expression was successfully elevated in Caco-2 and SW480 cells following transfection with miR-381 mimics compared with NC mimics (Fig. 3B). A luciferase reporter assay indicated that the luciferase activity of FLVCR1-AS1-wt was significantly inhibited by miR-381 mimics. However, miR-381 mimics had no effect on the luciferase activity of FLVCR1-AS1-mut in 293T cells (Fig. 3C). Moreover, the RNA pull-down assay demonstrated that the FLVCR1-AS1 expression levels were elevated in Caco-2 and SW480 cells following transfection with bio-miR-381-wt (Fig. 3D). Furthermore, RT-qPCR analysis revealed that the expression of miR-381 was downregulated in CRC tissues compared with adjacent normal tissues (Fig. 3E), and the knockdown of FLVCR1-AS1 increased miR-381 expression in Caco-2 and SW480 cells (Fig. 3F). Following confirmation by RT-qPCR analysis that miR-381 was successfully knocked down in Caco-2 and SW480 cells following transfection with miR-381 inhibitor (Fig. 3G), whether FLVCR1-AS1 promotes CRC progression by modulating miR-381 was further investigated. CRC cells were transfected with shNC, shFLVCR1-AS1 and shFLVCR1-AS1 + miR-318 inhibitor. CCK-8 and Transwell assays of the transfected cells demonstrated that the miR-381 inhibitor attenuated the inhibitory effect of FLVCR1-AS1 knockdown on CRC cell viability, migration and invasion (Fig. 3H-J). The aforementioned findings suggest that FLVCR1-AS1 may act as a molecular sponge for miR-381 in CRC.

Bioinformatics analysis was conducted using starBase software to predict the downstream targets of miR-381. The analysis revealed that RAP2A includes a potential binding site for miR-381 (Fig. 4A). A luciferase reporter assay demonstrated that miR-381 mimics inhibit the luciferase activity of RAP2A-wt but have no effect on the luciferase activity of RAP2A-mut (Fig. 4B). In addition, RT-qPCR results showed that RAP2A was highly expressed in CRC tissues compared with adjacent normal tissue (Fig. 4C). In addition, RT-qPCR and western blot assays indicated that overexpression of miR-381 significantly reduced RAP2A expression in Caco-2 and SW480 cells (Fig. 4D). Subsequently, the association between RAP2A and FLVCR1-AS1 was evaluated. The results indicated that FLVCR1-AS1 knockdown downregulated RAP2A at the mRNA and protein levels (Fig. 4E). In addition, transfection with pcDNA3.1/FLVCR1-AS1 attenuated the miR-381-mediated inhibition of RAP2A expression (Fig. 4F). Additionally, Pearson's correlation analysis showed that the expression of miR-381 was negatively correlated with that of FLVCR1-AS1 and RAP2A in CRC tissues (Fig. 4G). The aforementioned data suggest that FLVCR1-AS1 increases RAP2A expression via the repression of miR-381 in CRC.

To investigate whether the molecular functions of FLVCR1-AS1 are mediated by the upregulation of RAP2A in CRC, the Caco-2 and SW480 cells were transfected with shNC, shFLVCR1-AS1 or shFLVCR1-AS1 + RAP2A, and RT-qPCR and western blot assays were then carried out. First, it was confirmed that RAP2A expression was successfully increased by transfection with RAP2A overexpression vector (Fig. 5A). The RT-qPCR and western blotting results demonstrated that RAP2A overexpression attenuated the FLVCR1-AS1 knockdown-mediated reduction of RAP2A expression (Fig. 5B). Similarly, CCK-8 and Transwell assays confirmed that RAP2A overexpression attenuated the effect of FLVCR1-AS1 downregulation on Caco-2 and SW480 cell proliferation, migration, and invasion (Fig. 5C-E). These findings indicate that FLVCR1-AS1 promotes the development of CRC by modulating the expression of RAP2A.