Co-expression analyses of the interaction between miR-151a-5p and AGMAT were downloaded from the starbase database (http://starbase.sysu.edu.cn/), including COAD samples (n=450). The potential binding sites between miR-151a-5p and AGMAT were downloaded from the miRmap database (http://www.mirmap.ezlab.org). The expression levels of miR-151a-5p in tumor tissues (n=254) were directly compared with those in adjacent normal tissues (n=8) using the expression data from ~32 common types of cancer downloaded from the Gene Expression Display Server (http://bioinfo.life.hust.edu.cn/web/GEDS/). Additionally, the expression levels of miR-151a-5p in tumor tissues (n=445) were also directly compared with those in adjacent normal tissues (n=8) downloaded from the UCSC Xena database (http://xena.ucsc.edu/).

The human CRC cell lines HCT8, HCT116, HT29, SW480, SW620, LS174T and DLD1 were purchased from the American Type Culture Collection. In addition, the normal human colorectal cells NCM460 and 293T cells were conserved in the central laboratory of Shanghai Fengxian District Central Hospital. All cell lines were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (both from Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin solution in a humidified incubator at 37°C with 5% CO₂.

The full-length sequence of the human AGMAT gene was subcloned into the pCD513B vector (Geneppl Technology Co., Ltd.) and transfected to generate cells stably overexpressing AGMAT. To stably knock down AGMAT in CRC cells, small hairpin RNAs (shRNAs) targeting AGMAT were synthesized and subcloned into the lentiviral pPLK-GFP + Puro vector (Geneppl Technology Co., Ltd.). The sequences of the shRNA clones used were listed in Table I. The concentrations of shRNA clones used were as follows: NC, 964 ng/µl; shAGMAT#1, 1,142 ng/µl; and shAGMAT#3, 795 ng/µl. The mixture of 1 ml DMEM, 10 µg DNA and 10 µl (µg) GM easy™ Lentivirus mix (Genomeditech) was co-transfected into 293T cells using 60 µl Hg transgene™ reagent (Genomeditech), according to the manufacturer's instructions (cat. no. GMeasy 40; Genomeditech). The lentiviral particles were collected at 48 h following cell transduction.

When the cell density of ~80–90%, CRC cells seeded into 6-cm culture-plates were infected with filtered lentiviral particles (the 2nd generation system) plus 8 µg/ml Polybrene (Sigma-Aldrich; Merck KGaA) for 24 h. Following transduction, cells were treated with 0.5 µg/ml puromycin (Invitrogen; Thermo Fisher Scientific, Inc.) for >7 days to obtain stably transfected cells; afterwards, the stably transfected cells were maintained with 0.3 µg/ml puromycin. Subsequently, when reached ~60-70% confluency in the cell culture plate, HCT116 cells, SW480 cells and stably AGMAT expressing CRC cells were treated with a mixture containing 5 µl Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), 100 pmol miR-151a-5p and control oligonucleotides, according to the manufacturer's instructions. Following transfection for 48 h, the relative expression levels of miR-151a-5p were determined by reverse transcription-quantitative PCR (RT-qPCR). The sequences of miR-151a-5p mimics or inhibitor used were listed in Table I (Genomeditech).

The luciferase reporter constructs were generated by inserting the wild-type (WT) or mutant (MUT) 3′ untranslated region (3′ UTR) of AGMAT into the pGL3 luciferase vector (Geneppl Technology Co., Ltd.). Subsequently, the stable AGMAT overexpressing CRC cells were co-transfected with 2 µg WT or MUT AGMAT-3′ UTR, Renilla Luc, miR-151a-5p mimics or miR-NC using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions (Shanghai Yeasen Biotechnology Co., Ltd.). Following transfection for 48 h, the firefly and Renilla luciferase activities were determined by spectrophotometry using the Dual Luciferase Reporter Assay System (Glomax 96; Promega Corporation). The luciferase activity was then calculated. The results are expressed as the ratio of firefly luciferase activity to Renilla luciferase activity.

The cell proliferation ability was assessed using CCK-8 assay (Dojindo Molecular Technologies, Inc.), according to the manufacturer's protocol. Briefly, treated CRC cell lines were seeded into a 96-well plate at a density of 10³ cells/well in 100 µl DMEM supplemented with 10% FBS and were then cultured for 10, 24, 48 and 72 h at 37°C with 5% CO₂. Subsequently, cells were supplemented with 10 µl CCK-8 solution, followed by incubation for an additional 1 h in the dark. Finally, the optical density (OD) of each well was measured at a wavelength of 450 nm using a microplate reader (BioTek Instruments, Inc.).

For colony formation assays, treated CRC cells were seeded into a 12-well plate at a density of 3×10³ cells/well and cultured in DMEM supplemented with 10% FBS for 7 or 10 days. The formed colonies were fixed with 4% paraformaldehyde at room temperature for 30 min and were then stained with 0.1% crystal violet at room temperature for 1 h followed by washing with ddH₂O. Images of the formed colonies were captured under an inverted light microscope (Olympus Corporation). Finally, the number of colony cells (>50 cells per cell colony) were counted using ImageJ software (version 1.52a; National Institutes of Health).

The migration and invasion abilities of HCT116 and SW480 cells transfected with the aforementioned plasmids/miR mimics/inhibitors for 48 h were evaluated using Transwell chambers (0.8 µm; cat. no. 353097; Corning, Inc.). Briefly, treated HCT116 and SW480 cells at a density of 10⁵ cells/well were added to the upper chamber of the Transwell chamber coated or not with Matrigel (0.8 µm; cat. no. 354480; Corning, Inc.). The upper chambers of the Transwell chamber with Matrigel were precoated with 500 µl medium without serum for 1 h at 37°C with 5% CO₂. For migration assays, the lower chamber was supplemented with 600 µl medium with 10% FBS, cells were cultured for 48 h, while for the invasion assays, the lower chamber was supplemented with 600 µl medium with 20% FBS, cells were cultured for 24 h, the upper chamber was supplemented without serum. The migratory or invasive cells were fixed with 4% paraformaldehyde at room temperature for 30 min and were then stained with 0.1% crystal violet at room temperature for 1 h followed by ddH₂O washing. Finally, migratory or invasive cells were observed and images were captured under a fluorescence microscope (magnification, ×100).

Total RNA was extracted from treated cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The nuclear and cytoplasmic extracts were isolated using the cytoplasmic and nuclear RNA purification kit (cat. no. 21000; Norgen Biotek Corp.). Subsequently, 1 µg RNA was reverse transcribed into cDNA using the Evo M-MLV RT Mix kit with gDNA Clean for qPCR (cat. no. AG11728; Accurate Bio-Medical Technology Co., Ltd.) and the miRNA 1st strand cDNA synthesis kit (by stem-loop) (cat. no. MR101-02; Vazyme Biotech Co., Ltd.). qPCR was performed on the quantum Studio 6 real-time PCR system (Thermo Fisher Scientific, Inc.) using the SYBR Green premix Pro Taq HS qPCR kit (Rox plus) (cat. no. AG1718; Accurate Bio-Medical Technology Co., Ltd.). The thermocycling conditions used for reverse transcription PCR were as follows: For gDNA clean, 42°C 2 min, 4°C; for synthesis of first strand cDNA of miR-151a-5p, 25°C for 5 min, 50°C for 15 min, 85°C for 5 min; for synthesis of common primer cDNA, 37°C for 15 min, 85°C for 5 min, 4°C. The thermocycling conditions used for qPCR were as follows: For miR-151a-5p, 95°C for 5 min, 95°C 10 sec, 60°C for 5 min; for common primers, 95°C for 30 sec, 95°C for 5 sec, 60°C for 30 sec. The relative mRNA expression levels of the target genes were calculated using the 2^−ΔΔCq method (21). GAPDH served as an internal reference gene for AGMAT. The primers for U1 were purchased from Guangzhou RiboBio Co., Ltd. The primer sequences used are listed in Table II (Sangon Biotech Co., Ltd.).

The transfected cells were cultured in 48-well plates at a density of 6×10⁴ cells/well for 24 h. Following fixation with 4% paraformaldehyde at room temperature for 30 min, the cells were permeabilized with 1% Triton X-100 at room temperature for 15 min followed by blocking with 3% BSA (cat. no. 4240GR100; Biofroxx; neoFroxx) at 37°C for 1 h. After cell incubation with primary antibodies at 4°C overnight, the cells were treated with the corresponding secondary Goat anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 568 (Thermo Fisher Scientific, Inc.) at room temperature for 30 min in the dark. Nuclei were counterstained with DAPI (5 mg/ml; cat. no. C1006; Beyotime Institute of Biotechnology) at room temperature for 20 min in the dark. The stained cells were observed under a fluorescence microscope (magnification, ×200). The antibodies used for immunofluorescence are listed in Table III.

Total proteins were extracted from treated CRC cells using RIPA lysis buffer (New cell & Molecular Biotech Co., Ltd.) supplemented with phosphatase inhibitors (Beyotime Institute of Biotechnology) and protein concentration was measured with the BCA Protein Assay Kit (cat. no. P0012; Beyotime Institute of Biotechnology). Subsequently, the proteins (10–20 µl/lane, 25 mg/ml) were separated by 10% SDS-PAGE (New cell & Molecular Biotech Co., Ltd.) and were then transferred onto a nitrocellulose membrane (Pall Corporation). Following blocking with 5% skimmed milk (Beyotime Institute of Biotechnology) at room temperature for 1 h, the membrane was incubated with primary antibodies at 4°C overnight. After incubation with the corresponding secondary antibodies at room temperature for 1 h, the immunoreactive bands were visualized using an ECL Kit (Vazyme Biotech Co., Ltd.) and images were captured under the Tanon 4600 system (Tanon Science and Technology Co., Ltd.). The density of each band was measured using a computer-assisted imaging analysis system (Adobe Systems Inc.). The expression levels of the target proteins were normalized to those of β-tubulin. The following primary and secondary antibodies were used: Anti-AGMAT, anti-MMP2, anti-MMP9, anti-N-cadherin, anti-E-cadherin, anti-Vimentin, anti-β-catenin, HRP-conjugated Affinipure Goat anti-Mouse IgG (H+L) and HRP-conjugated Affinipure Goat anti-Ribbit IgG (H+L) (Table IV). The protein expression levels were assessed using ImageJ software (National Institutes of Health).

All experiments were repeated at least in triplicate. Data are expressed as the mean ± SD. All statistical analyses were carried out using GraphPad Prism 8 (GraphPad Software, Inc.). The differences between two groups were compared using unpaired Student's t-test. Comparisons among multiple groups were conducted using with one- or two-way ANOVA followed by Dunnett's multiple comparisons test used as post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To explore the pathophysiological relevance of miR-151a-5p in human CRC, the expression levels of miR-151a-5p were determined in CRC and adjacent normal tissues obtained from UCSC Xena database and the Gene Expression Display Server. Therefore, compared with normal colorectal tissues, miR-151a-5p was significantly upregulated in CRC tissues (Fig. 1A and B). These data suggested that miR-151a-5p was upregulated in CRC tissues and could therefore regulate the occurrence and development of CRC via acting as a tumor-promoting gene. In addition, the expression levels of miR-151a-5p were further detected in several CRC cell lines by RT-qPCR. The analysis showed that the expression levels of miR-151a-5p were higher in CRC cell lines compared with normal colorectal cells. Among all CRC cell lines examined, SW480 cells exhibited the highest miR-151a-5p expression levels, while HCT116 cells the lowest one (Fig. 1C). Therefore, HCT116 and SW480 cells were selected for the subsequent experiments. Furthermore, the transfection efficiency of miR-151a-5p in CRC cells was also verified (Fig. 1D and E).

To further uncover the potential role of miR-151a-5p in CRC, HCT116 or SW480 cells were transfected with miR-151a-5p mimics or inhibitor, respectively. Following cell transfection for 48 h, the cell proliferation ability was assessed by CCK-8 and colony formation assays. The results showed that compared with the NC group, the change in the overexpression levels of miR-151a-5p significantly promoted the proliferation of HCT116 cells in time-dependent manner (Fig. 2A), while as predicted, the proliferation capacity of SW480 cells transfected with miR-151a-5p inhibitor was significantly inhibited (Fig. 2B). In addition, the effect of miR-151a-5p on the proliferation of HCT116 and SW480 cells was evaluated by colony formation assay. The results demonstrated that the overexpression levels of miR-151a-5p promoted the proliferation of HCT116 cells (Fig. 2C). By contrast, the proliferation of SW480 cells was significantly inhibited after transfection with miR-151a-5p inhibitor (Fig. 2D). The enhanced migration and invasion abilities of CRC cells could also serve a key role in the growth of CRC. Therefore, to further evaluate the effect of miR-151a-5p on CRC cell invasion and migration, Transwell assays coated with Matrigel or not, respectively, were used. The data revealed that compared with the corresponding NC group, the migration and invasion abilities of HCT116 cells were significantly increased following cell transfection with miR-151a-5p mimics (Fig. 2E-G). However, the opposite effect was observed in miR-151a-5p-depleted SW480 cells (Fig. 2H-J). The aforementioned findings indicated that miR-151a-5p could significantly enhance the proliferation, migration and invasion abilities of CRC cells.

Epithelial cells are characterized by mesenchymal phenotype and changes in the expression of three significant biomarkers, including E-cadherin, vimentin and N-cadherin, eventually leading to reduced cell adhesion, loss of polarity and tight junction (22). Matrix metalloproteinases (MMPs) are proteins involved in the degradation of base membranes, thus affecting cell metastasis during the development of malignant tumors (23). Therefore, western blot analysis was performed and the result revealed that miR-151a-5p overexpression in HCT116 cells downregulated E-cadherin and upregulated N-cadherin, vimentin and β-catenin. Correspondingly, western blot analysis was also performed to evaluate the effect of miR-151a-5p on the expression levels of metastasis-related MMPs, such as MMP2 and MMP9. The results demonstrated that compared with the NC group, miR-151a-5p enhanced the protein expression levels of both MMP2 and MMP9 in HCT116 cells. By contrast, miR-151a-5p silencing in SW480 cells exhibited the opposite effect (Fig. 3A-C). To further determine the expression of EMT-related proteins in CRC cell lines, the expression of E-cadherin and vimentin was assessed by immunofluorescence analysis. The immunofluorescence analysis results demonstrated that miR-151a-5p overexpression in HCT116 cells downregulated E-cadherin and upregulated vimentin. However, miR-151a-5p silencing upregulated E-cadherin and downregulated vimentin (Fig. 3D). The aforementioned results verified that miR-151a-5p could promote EMT in CRC cells.

Polyamines, such as putrescine, spermidine and spermine, produced during the metabolic process, are low molecular weight, aliphatic and polycationic nitrogen compounds containing two or more amino groups. Currently, the application of new technologies has increased our understanding of the genetics and potential molecular biology of polyamine function in normal and tumor cells. Recognition of the effect of the increased polyamine levels on tumor transformation and progression has deepen our understanding of the significant role of polyamines in cancer, thus providing reasonable targets for intervention (24). It has been reported that miRNAs can regulate tumor-associated proteins in CRC (25,26). However, the association between genes and miRNAs is not a one-to-one correspondence. The results of the current study showed that following miR-151a-5p overexpression or silencing in CRC cell lines, the levels of the polyamine metabolism-related metabolites and key enzymes significantly altered, while the most differences were observed in the expression levels of AGMAT at the molecular level (Fig. 4A and B). These findings indicated that miR-151a-5p could promote polyamine metabolism via regulating AGMAT. Therefore, subsequently, the present study aimed to explore whether miR-151a-5p could regulate CRC progression via targeting AGMAT. Bioinformatic analysis using the starbase database predicted that AGMAT encompassed a binding site for miR-151a-5p and revealed that AGMAT was positively associated with miR-151a-5p expression in CRC (Fig. 4C). In addition, it was further predicted using the miRmap database that the potential miR-151a-5p binding site was encompassed in AGMAT-3′ UTR (Fig. 4D). Previous studies suggested that the majority of miRNAs in the cytoplasm could act as tumor suppressors, thus resulting in gene silencing via inhibiting gene translation (27). By contrast, other studies showed that several miRNAs, located at the nucleus, could positively regulate gene transcription via binding to and activating their target enhancers (28,29). In the present study, RNA isolation from nuclear and cytoplasmic extracts revealed that miR-151a-5p was mainly localized in the nucleus, as evidenced by RT-qPCR analysis (Fig. 4E and F). To further verify whether miR-151a-5p could regulate the expression of AGMAT via binding to AGMAT-3′ UTR, a dual luciferase reporter assay was performed. As revealed in Fig. 4G, the luciferase activity of AGMAT-3′ UTR was significantly increased in cells transfected with miR-151a-5p mimics. However, miR-151a-5p overexpression showed a modest effect on luciferase activity in cells transfected with mutated putative binding site. For the convenience of follow-up tests, the stable AGMAT expression cell lines were designed, namely, the overexpression of AGMAT in HCT116 cells and the knockdown of AGMAT in SW480 cells. The transfection efficiency of AGMAT in stable transmutation cells was determined by RT-qPCR and western blot analysis. ShAGMAT#1 was selected from the AGMAT-stably knocked down SW480 cells due to the best knockdown effect (Fig. S1). Furthermore, RT-qPCR and western blot analysis were carried out to reveal the relationship between miR-151a-5p and AGMAT expression. The results identified that the mRNA expression levels of AGMAT were significantly reduced in AGMAT-stably overexpressing HCT116 cells transfected with miR-151a-5p inhibitor (Fig. 5A). However, the opposite effect was observed in the AGMAT-stably knocked down SW480 cells transfected with miR-151a-5p mimics (Fig. 5B). Western blot analysis further verified the positive regulatory association between miR-151a-5p and AGMAT (Fig. 5C and D). This indicated that miR-151a-5p could promote the expression of AGMAT in CRC cells. The aforementioned data suggested that AGMAT could be a potential target of miR-151a-5p and could be positively associated with miR-151a-5p expression in CRC.

To further verify whether miR-151a-5p could regulate the potential functions of AGMAT in CRC, AGMAT-stably expressing HCT116 or SW480 cells were transfected with miR-151a-5p inhibitor or mimics. Following transfection for 48 h, the proliferation ability of AGMAT-stably expressing CRC cells was evaluated by CCK-8 and colony formation assays. Therefore, compared with the NC group, the proliferation of the AGMAT-stably overexpressing HCT116 cells was significantly decreased after transfection with miR-151a-5p inhibitor in a time-dependent manner (Fig. 6A). However, the significant opposite result of the proliferation was observed in a time-dependent manner in the AGMAT-stably knocked down SW480 cells transfected with miR-151a-5p mimics (Fig. 6B). Additionally, the effect of miR-151a-5p targeting AGMAT on the proliferation of the AGMAT-stably overexpressing HCT116 or SW480 cells was also evaluated by colony formation assays. As expected, the proliferation ability of the AGMAT-stably overexpressing HCT116 cells transfected with miR-151a-5p inhibitor was significantly decreased (Fig. 6C), but the significant opposite result of the proliferation was observed in a time-dependent manner in the AGMAT-stably knocked down SW480 cells transfected with miR-151a-5p mimics (Fig. 6D). To further explore whether miR-151a-5p could regulate CRC cell invasion and migration via targeting AGMAT, Transwell invasion and migration assays were performed. The results showed that compared with the NC group, the migratory and invasive abilities of the AGMAT-stably overexpressing HCT116 cells transfected with miR-151a-5p inhibitor were significantly decreased (Fig. 6E-G). The completely opposite effect was obtained in the AGMAT-stably knocked down SW480 cells transfected with miR-151a-5p mimics (Fig. 6H-J). The aforementioned results indicated that miR-151a-5p could enhance the biofunction of CRC cells via targeting AGMAT.