A total of 68 matched CRC tissues and corresponding normal tissues were collected at The First Affiliated Hospital of the University of South China (Hengyang, China). Signed informed consent had been provided by all patients. None of the patients received chemotherapy or radiotherapy prior to surgical resection. The present study was approved by the Medical Ethics Committee of The First Affiliated Hospital of University of South China (approval no. LL20181102013).

Normal colorectal epithelial cells (FHC) and human CRC cells (HCT116, SW480, SW620, HT-29) were provided by The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. The cell lines were authenticated by STR profiling. The cells were grown in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin antibiotics (Invitrogen; Thermo Fisher Scientific, Inc.) in the presence of 5% CO₂ at 37°C.

Lentiviral knockdown vector specifically targeting SNHG20 [small interfering (si) RNA-SNHG20] and negative control (si-NC) were provided by Shanghai GenePharma Co., Ltd. miR-495 mimics, miR-495 inhibitor and their NCs were purchased from Shanghai GenePharma Co., Ltd. STAT3 overexpression plasmid [pcDNA3.1(+) STAT3] and its NC [pcDNA3.1(+)] were constructed by Shanghai GenePharma Co., Ltd. Before transfection, cells were subcultured in 6-well plates (Corning, Inc.) ~50% confluence (3×10⁵). RNAs (100 nM) or miR-495 mimics (50 nM) or miR-495 inhibitor (150 nM) or plasmids (1.5 µg per well) were transfected into cells. The lentivirus was prepared according to the User Manual of the Lenti-Pac™ HIV Expression Packaging kit (GeneCopoeia, Inc.). The viral packaging was performed in 293T cells after co-transfection of the Lv3-si-SNHG20, or empty lentiviral vector, and Lenti-Pac™ HIV packaging mix (GeneCopoeia, Inc.) using EndoFectin™ Lenti transfection reagent (GeneCopoeia, Inc.). The medium containing the retroviral supernatant was harvested 48–72 h after transfection. The retroviruses were added into HCT116 and SW480 cells for the infection. The cells were selected by culture in the presence of puromycin (5 µg/ml; Invitrogen; Thermo Fisher Scientific, Inc.) for up to 2 weeks. The sequences were as follows: si-SNHG20, 5′-GCCUAGGAUCAUCCAGGUUTT-3′; si-NC, 5′-UUCUCCGAACGUGUCACGUU-3′; miR-495 mimic, 5′-UGUGACGAAACAAACAUGGUGCACU-3′; miR-NC, 5′-CAGUACUUUUGUGUAGUACAA-3′; miR-495 inhibitor, 5′-GCCGAATTCTGGCTGCTATGATCTGAACT-3′; and inhibitor NC, 5′-CAGUACUUUUGUGUAGUACAA-3′. In order to establish the STAT3-overexpression cell model, the full-length cDNAs of human STAT3 were cloned into the pc-DNA3.1 vector. Transfection was performed when the cell density was 50%. All cells transfections were conducted using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). Transfection was performed at room temperature for 30 min. After incubation for 48 h, the cells were collected for the subsequent experiments.

Total RNA was isolated from tissue and cells with TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The mRNA expression levels of SNHG20 and STAT3 were detected following RNA extraction and RT-qPCR. Total RNA was reverse transcribed into cDNA using a Reverse Transcription kit (Takara Bio, Inc.). The PCR protocol consisted of cycling at 94°C for 3 min, followed by 30 cycles of 94°C for 30 sec, 55°C for 30 sec and 72°C for 60 sec and a final extension at 72°C for 5 min. RT-qPCR was carried out with SYBR Green (Takara Bio, Inc.). The expression levels of miR-495 were measured using the miRNA isolation kit (Guangzhou RiboBio Co., Ltd.). GAPDH and U6 were used as internal controls for RNAs and miRNAs, respectively. The relative expression levels of the genes were determined using the 2^−ΔΔCq method (13). The primers used are shown in Table I.

The cell proliferative rates were assessed by MTS and colony formation assays. The transfected cells were seeded in 96-well plates (2×10³ cells/well) with five replicate wells. At the time of harvest, 10 µl MTS was added to each well. Following incubation for 2 h, the absorbance was read at 490 nm on a microplate reader (Bio-Rad Laboratories, Inc.). The colony formation assay was performed with transfected cells seeded in 6-well plates (800/well) and incubated for 2 weeks. Subsequently, the cells were fixed with ethanol at room temperature for 10 min and stained with 0.1% crystal violet at room temperature for 10 min. when a single colony cell number >50 was achieved. Finally, colony counting was performed under an inverted light microscope (magnification, ×200; CKX41, Olympus).

The Transwell assay was carried out to assess the migratory (without Matrigel) and invasive (with Matrigel; BD Biosciences) ability of the cells. The BD Matrigel was placed in a 4°C refrigerator for 24 h. It was diluted to 1:9 ratios with RPMI-1640. The upper Transwell chamber was filled with 50 µl of the diluent. The Transwell insert was incubated for 3 h at 37°C until the Matrigel clotted. Briefly, the transfected cells (5×10⁴ cells/well) were seeded into the upper Transwell chamber with 250 µl serum-free RPMI-1640 medium, and the lower chamber was filled with 750 µl RPMI-1640 medium supplemented with 10% FBS. Following incubation for 48 h at 37°C, the invaded cells were fixed with ethanol at room temperature for 10 min and stained with 0.1% crystal violet at room temperature for 10 min. The stained cells were counted using an inverted light microscope (magnification, ×200; CKX41, Olympus Corporation).

The StarBase software V2.0 (http://starbase.sysu.edu.cn) was applied to analyze the potential miRNA target of SNHG20. The TargetScan software was performed to predict the potential mRNA target of miR-495. The wild-type (WT) or mutant (MUT) SNHG20 sequences containing the miR-495 binding sites were cloned into the pmir-GLO Dual-luciferase vector (Promega Corporation) and co-transfected with the miR-495 mimics or corresponding control sequences into the cells with Lipofectamine 2000. Following transfection, the cells were incubated for 48 h and the luciferase activity was measured with the Dual Luciferase Reporter Assay System (Promega Corporation). Similarly, pmirGLO-STAT3-WT or pmirGLO-STAT3-MUT were constructed and co-transfected with miR-495 mimics or the corresponding control into the cells. Following transfection, the cells were incubated for 48 h and the luciferase activity was measured. The Renilla luciferase was used as an internal control to homogenize the detection of the reporter gene.

Total protein from tissues or cells was extracted using pre-cooled RIPA lysis. Protein concentration was evaluated using the BCA protein assay regents (cat. no. 23225, Pierce; Thermo Fisher Scientific, Inc.). Equal amounts (50 µg/lane) of protein aliquots were loaded on to 10% SDS-PAGE gels, then, proteins transferred onto PVDF membranes. After blocking with 5% skimmed milk for 2 h at room temperature, the membranes were then incubated with primary antibodies at 4°C overnight. Next, the membranes were incubated with secondary antibody for 1 h at room temperature. The primary antibodies used in the present study were as follows: STAT3 (1:1,000; cat. no. 9139; Cell Signaling Technology, Inc.) and GAPDH (1:3,000; cat. no. 8245; Abcam). The secondary antibody used was HRP-conjugated goat anti-rabbit IgG (1:2,000; cat. no. 6728; Abcam). The immunoblots were detected using LI-COR Odyssey^® CLX Two-colour infrared laser imaging system (LI-COR Biosciences). ImageJ software V1.8.0 (National Institutes of Health) was implemented to perform densitometric analysis.

RIP assays were performed using the Magna RIP™ kit (EMD Millipore). Briefly, the cells were lysed using complete RIP lysis buffer (4°C; 1,000 × g; 10 min) and incubated with containing magnetic beads (2 µg) conjugated with human anti-Ago2 antibody (Abcam; cat. no. ab186733) or anti-IgG antibody (EMD Millipore; cat. no. MA5-27548) for overnight at 4°C. Samples were washed with ethanol and sodium acetate and incubated with proteinase K at 55°C for 30 min to isolate the RNA-protein complexes from beads (4°C; 1,000 × g; 15 min). Lastly, RT-qPCR assays were performed to determine SNHG20 and miR-495 expression.

Ten male nude mouse (age, 5–6 weeks; weight, 18–20 g) were housed in an isolated, clean, air-conditioned room at a temperature of 25–26°C and a relative humidity of ~50%, light 12 h/day. The mice were randomly divided into two groups (n=5/group). All rats were given free access to water and food. SW480 cells were stably transfected with si-SNHG20 lentivirus or si-NC (2×10⁶ cells/mouse; n=5 for each group). A total of 5×10⁶ cells suspended in 100 µl PBS were subcutaneously inoculated into the right flanks of each nude mouse (the mice were provided by the Experimental Animal Center of the Chinese Academy of Sciences). Tumor size was measured once every 4 days following the appearance of the tumors and calculated using the following formula: V (mm³)=width² (mm²) × length (mm)/2. Following 4 weeks of tumor growth, the mice mouse was euthanized by cervical dislocation under anesthesia with 1% pentobarbital (50 mg/kg) by intraperitoneal injection. The following signs were used to verify death of the mice: Eyes had turned pale, no heartbeat, and they did not respond to external stimuli. The tumors were collected for weight assessment and for RT-qPCR and western blot analyses. All the animal experiments were carried out in accordance with the Ethics Committee of The First Affiliated Hospital of University of South China.

Ten male nude mice were randomly divided into two groups (n=5/group). SW480 cells that exhibited stable knockdown of SNHG20 expression were transfected with firefly luciferase vector (LV16; 100 nM; Guangzhou RiboBio Co., Ltd.). Nude mouse (BALB/c-nu/nu; age, 5–6 weeks old; weight, 18–20 g) were anesthetized by intraperitoneal injection of 1% pentobarbital (50 mg/kg) and cells (2×10⁶ LV16-si-SNHG20 SW480 cells) were inoculated into the tail vein. Lung metastases were monitored by bioluminescent imaging for 6 consecutive weeks. All mice were sacrificed at 6 weeks following injection. Before collection of the lungs, the mice were euthanized by cervical dislocation under anesthesia with 1% pentobarbital (50 mg/kg) by intraperitoneal injection. The aforementioned indicators were used to verify death. The lung tissue were fixed using 4% paraformaldehyde at 4°C for 24 h and sections were cut into 4 µm pieces. The lung metastatic sites, the number of cells and the corresponding signal intensity were recorded and analyzed by using a small animal live imaging system (PerkinElmer, Inc.; Lumina X5).

All statistical analyses were carried out with GraphPad Prism 8.0 software (GraphPad Software, Inc.) or SPSS 20.0 (IBM Corp.). The data are presented as the mean ± SD. A paired Student's t-test was applied for the comparison of matched samples (CRC and normal adjacent tissues). An unpaired t-test was applied for the other comparisons between two groups. For the comparison of multiple groups, one-way ANOVA analysis was performed followed by Tukey's post hoc test. The statistical analysis between the clinical parameters of the patients with CRC and the SNHG20 expression groups was analyzed using a chi-squared test or the Fisher's exact test. Pearson correlation analysis was used to estimate the relationship between the expression level of different gene. All experiments were carried out at least three times. P<0.05 was considered to indicate a statistically significant difference.

The expression levels of SNHG20 were measured in 68 pairs of CRC and adjacent normal tissues by RT-qPCR. The results indicated that SNHG20 was highly expressed in CRC tissues (Fig. 1A and B). Similarly, SNHG20 expression was significantly increased in CRC cell lines (HCT116, SW480, SW620 and HT-29) compared with that of the FHC cell line (Fig. 1C). To further explore the clinical significance of SNHG20 expression in CRC, 68 patients with CRC were divided into a high (n=34) and a low expression group (n=34), according to the median expression levels of SNHG20 in all CRC samples. Notably, the clinicopathological data demonstrated that high expression of SNHG20 was significantly associated with tumor size (P=0.014), tumor invasion depth (P=0.019), positive lymph node status (P=0.022), distant metastasis (P=0.017) and TNM stage (P=0.038; Table II).

Loss of function assays were performed to further assess the biological function of SNHG20. SNHG20 expression was significantly reduced in HCT116 and SW480 cells that were stably transfected with si-SNHG20 (Fig. 1D). MTS and colony formation assays indicated that SNHG20 knockdown significantly suppressed the viability of HCT116 and SW480 cells (Fig. 1E and F). Moreover, Transwell assays revealed that SNHG20 knockdown resulted in a significant reduction of the migratory and invasive activities of HCT116 and SW480 cells (Fig. 1G).

It has been reported that specific lncRNAs serve as competing endogenous RNAs (ceRNAs) by sponging miRNAs (14). StarBase software was used to predict the target miRNAs of SNHG20. miR-495 was characterized as a potential target of SNHG20 due to its potential complementary sites (Fig. 2A) for SNHG20 and its role as a tumor suppressor in CRC (15). The present study indicated that miR-495 expression levels were significantly downregulated in CRC tissues and cell lines (HCT116, SW480, SW620 and HT-29) (Fig. 2B and C). The miR-495 mimic and inhibitor were transfected into HCT116 and SW480 cell lines. The expression of miR-495 was significantly increased in miR-495 mimic-transfected cells, whereas miR-495 expression was significantly decreased in miR-495 inhibitor-transfected cells (Fig. 2D). The dual luciferase reporter assays revealed that miR-495 mimics significantly reduced the relative luciferase activity in HCT116 and SW480 cells transfected with pmirGLO-SNHG20-WT (Fig. 2E). In addition, the expression of miR-495 were significantly increased following SNHG20 knockdown (Fig. 2F). Furthermore, Pearson's correlation analysis revealed that miR-495 expression was inversely associated with SNHG20 in CRC tissues (r=−0.53; P<0.001; Fig. 2G). RIP assays indicated that SNHG20 and miR-495 expression levels were significantly enriched in Ago2-containing microribonucleoprotein complexes (Fig. 2H). Taken together, the data clearly indicated that SNHG20 directly interacted with miR-495.

Subsequently, rescue experiments were performed by transfecting miR-495 inhibitor into SNHG20-knockdown cells. The results suggested that the miR-495 inhibitor reversed the suppressive effects of SNHG20 knockdown on CRC cell viability (Fig. 3A and B), migration and invasion (Fig. 3C). In summary, SNHG20 expression regulated the progression of CRC cells via sponging miR-495.

The present study demonstrated that STAT3 was a potential target of miR-495 (Fig. 4A). A Previous study reported that STAT3 serves as an oncogene in various cancer types, including CRC (16). The data demonstrated that STAT3 expression levels were significantly upregulated in CRC tissues (Fig. 4B). In addition, Pearson's correlation analysis confirmed that miR-495 expression was associated with STAT3 in CRC tissues (r=−0.59; P<0.001; Fig. 4C). The luciferase reporter assay revealed that miR-495 mimics significantly reduced the luciferase activity of STAT3-WT, whereas this effect was not noted for STAT3-MUT (Fig. 4D). In addition, overexpression of miR-495 resulted in a significant decrease in the mRNA and protein expression of STAT3 in both HCT116 and SW480 cells (Fig. 4E and F).

Accumulating evidence has revealed that lncRNAs regulate tumor progression by competing for target miRNAs and modulating mRNA expression (17). Therefore, the present study aimed to investigate whether SNHG20 expression contributed to CRC progression by modulating STAT3 expression. The data indicated that STAT3 mRNA and protein levels were significantly reduced following SNHG20 knockdown. This effect could be partially reversed by the addition of miR-495 inhibitor to the cells (Fig. 5A-C). Moreover, Pearson's correlation analysis confirmed that STAT3 expression was positively correlated with SNHG20 expression in CRC tissues (r=0.48; P<0.001; Fig. 5D). To further investigate the role of STAT3 and SNHG20 in CRC progression, pcDNA3.1(+) STAT3 was transfected into cells. The mRNA and protein levels of STAT3 were increased significantly following transfection of the cells with pcDNA3.1(+) STAT3 (Fig. 5E and F). In addition, the mRNA and protein levels of STAT3 were significantly reduced following SNHG20 knockdown, whereas co-transfection of pcDNA3.1(+) STAT3 and si-SNHG20 reversed this inhibitory effect (Fig. 5G and H). Functional assays indicated that STAT3 overexpression could rescue the inhibitory effect of SNHG20 knockdown on cell viability (Fig. 6A and B), migration and invasion (Fig. 6C). In summary, these data demonstrated that SNHG20 facilitated CRC progression by modulating the miR-495/STAT3 axis.

The tumor volume was significantly smaller in the SNHG20-knockdown group than that noted in the NC group (Fig. 7A) In addition, the average weight of the tumors in the SNHG20-knockdown group was significantly lower than that noted in the NC group (Fig. 7B). The tumor diameters of the tumors in the SNHG20-knockdown group were significantly smaller than that in the NC group (Fig. 7C). The expression levels of miR-495 in tumor tissues from the SNHG20-knockdown group were higher than those of the NC group as determined by RT-qPCR (Fig. 7D), which was consistent with the in vitro data. However, the expression levels of STAT3 in the SNHG20-knockdown group were significantly lower than those of the NC group (Fig. 7E). The results derived from the in vivo fluorescence imaging indicated that the lowest number of metastatic sites was present in the SNHG20-knockdown group, whereas the highest number was found in the NC group (Fig. 7F). In addition, SNHG20 knockdown significantly reduced the number of metastatic pulmonary nodules compared with that in the NC group (Fig. 7G). Taken together, the results indicated that SNHG20 knockdown inhibited CRC growth and metastasis in vivo.