According to previous studies (21,22), human CRC HCT-116 and HT-29 cell lines were chosen to evaluate the effect of miR-9-5p in CRC chemoresistance. HCT-116 and HT-29 cells were obtained from Procell Life Science & Technology Co., Ltd., and maintained in McCoy's 5A medium (Procell Life Science & Technology Co., Ltd.) containing 10% fetal bovine serum (FBS). 293T cells (Shanghai Zhongqiao Xinzhou Biotechnology Co., Ltd.) were incubated in Dulbecco's modified Eagle's medium (DMEM; Sigma-Aldrich; Merck KGaA) containing 10% FBS. Cells were cultured at 37°C with 5% CO₂.

Human miR-9-5p agomir (sense, 5′-UCUUUGGUUAUCUAGCUGUAUGA-3′ and antisense, 5′-AUACAGCUAGAUAACCAAAGAUU-3′) and its scrambled negative control (NC; sense, 5′-UUCUCCGAACGUGUCACGUTT-3′ and antisense, 5′-ACGUGACACGUUCGGAGAATT-3′), miR-9-5p antagomir (5′-UCAUACAGCUAGAUAACCAAAGA-3′) and its scrambled NC (5′-CAGUACUUUUGUGUAGUACAA-3′) were bought from Shanghai GenePharma Co., Ltd. HCT-116 and HT-29 cells were seeded in a 12-well plate (4×10⁴ cells/well) and incubated overnight at 37°C with 5% CO₂. At 70% confluency, HT-29 cells were transfected with 100 pmol miR-9-5p agomir or agomir NC at room temperature for 48 h using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer's protocol. Similarly, HCT-116 cells were transfected with 100 pmol miR-9-5p antagomir or antagomir NC. An HMGA2-overexpression pcDNA3.1 plasmid (over-HMGA2) was obtained from GenScript, whereas an empty pcDNA3.1 vector (GenScript) was applied as a NC (vector). A small interference (si)RNA sequence of HMGA2 (si-HMGA2; sense, 5′-AGAGGCAGACCUAGGAAAUTT-3′ and antisense, 5′-ATTTCCTAGGTCTGCCTCTTT-3′) was bought from JTS Scientific, Ltd. A scrambled siRNA (JTS Scientific, Ltd.) was used as a NC (siRNA NC; sense, 5′-UUCUCCGAACGUGUCACGUTT-3′ and antisense, 5′-ACGUGACACGUUCGGAGAATT-3′). A total of 1 µg of over-HMGA2 was transfected alone or co-transfected with 50 pmol miR-9-5p agomir into HT-29 cells at room temperature for 48 h using Lipofectamine 2000 following the manufacturer's protocol. HCT-116 cells were transfected with 50 pmol si-HMGA2 or co-transfected with 50 pmol miR-9-5p antagomir and si-HMGA2. Transfected cells were collected 48 h after incubation for subsequent experiments.

After fixation with 4% paraformaldehyde for 20 min at room temperature, two CRC cell lines (HCT-116 and HT-29 cells) were used to observe nuclear changes and apoptotic body formation using the Hoechst staining kit (Nanjing KeyGen Biotech Co., Ltd.) according to the manufacturer's protocol. Cells were visualized under a fluorescence microscope (IX53; Olympus Corporation; magnification, ×400) and counted by a professional researcher, who was blind to the grouping. The percentage of apoptotic cells was calculated as follows: Apoptotic cells (%)=(the number of apoptotic cells/the number of total cells) ×100%.

Total RNA from HCT-116 and HT-29 cells was isolated using the TRIpure lysis buffer (BioTeke Corporation). Subsequently, RT to cDNA was performed using M-MLV Reverse Transcriptase kit (cat. no. PR6502; BioTeke Corporation) according to the manufacturer's protocol. qPCR reactions were performed using the SYBR Green PCR kit (Sigma-Aldrich; Merck KGaA) according to the manufacturer's protocol. The PCR reaction was run for an initial denaturation for 5 min at 94°C, followed by 40 cycles of denaturation at 94°C for 15 sec, annealing at 60°C for 25 sec and extension at 72°C for 30 sec. Subsequently, the 2^−ΔΔCq comparative method was applied to determine the relative expression levels of mRNA and miRNA (23). U6 was used as the internal reference for miR-9-5p, while β-actin acted as the internal reference for HMGA2. The primer sequences used are listed in Table I.

TargetScan (http://www.targetscan.org) was used to predict the potential target genes of miR-9-5p. For the miR-9-5p-target analysis, a fragment of the HMGA2 3′-UTR containing either the wild-type (WT) or mutated (Mut) binding sites for miR-9-5p was cloned in a pmirGLO luciferase reporter vector (Promega Corporation) following the manufacturer's instructions. 293T cells were plated into 12-well plates and co-transfected using Lipofectamine 2000 with luciferase reporter plasmid and miR-9-5p agomir and agomir NC. After 4 h, cells were incubated in new DMEM for 48 h at 37°C. Subsequently, luciferase activities were detected using the Dual-Luciferase Reporter Assay kit (Promega Corporation) following the manufacturer's protocol. Luciferase activity was normalized to the Renilla luciferase activity.

Cells were lysed using the RIPA lysis buffer (Beyotime Institute of Biotechnology) and the supernatants were collected via centrifugation (10,000 × g for 5 min at 4°C). Protein concentration was measured using the BCA Protein Concentration Determination kit (Beyotime Institute of Biotechnology) following the manufacturer's protocol. Protein separation (40 µg/lane) was performed via 12 or 15% SDS-PAGE (Beyotime Institute of Biotechnology). Proteins were transferred onto polyvinylidene fluoride membranes (Thermo Fisher Scientific, Inc.) and blocked with 5% (m/v) skimmed milk, followed by incubation with the following primary antibodies at 4°C overnight: Polyclonal rabbit anti-HMGA2 (1:1,000; cat. no. 20795-1-AP; ProteinTech Group, Inc.) and monoclonal mouse anti-β-actin (1:1,000; cat. no. sc-47778; Santa Cruz Biotechnology, Inc.). After washing with TBS-Tween (0.1% Tween-20), proteins were incubated with HRP-conjugated goat anti-rabbit IgG (1:5,000; cat. no. A0208; Beyotime Institute of Biotechnology) and goat anti-mouse IgG (1:5,000; cat. no. A0216; Beyotime Institute of Biotechnology) secondary antibodies for 40 min at 37°C. The protein bands were visualized using an enhanced chemiluminescence kit (Beyotime Institute of Biotechnology) according to the manufacturer's protocol, and quantified using the Gel-Pro-Analyzer 4.0 software (Media Cybernetics, Inc.).

Briefly, 4×10³ cells/well were plated in a 96-well plate. Subsequently, CRC cells were subjected to distilled water (vehicle control group) or different concentrations of 5-FU (0.625, 1.25, 2.5, 5, 10, 20, 40 and 80 µM; Shanghai Aladdin Biochemical Technology Co., Ltd.). After 48 h at 37°C, the CCK-8 assay kit (Beyotime Institute of Biotechnology) was used to measure CRC cell viability at 37°C for 1 h following the manufacturer's protocol. Optical density values at 450 nm were detected using an ELX-800 microplate reader (BioTek Instruments, Inc.; Agilent Technologies, Inc.).

The Annexin V-FITC Apoptosis Detection kit (Nanjing KeyGen Biotech Co., Ltd.) was used to measure apoptosis. After washing two times with PBS, CRC cells were double-stained with Annexin V-FITC and PI in the dark for 10 min at room temperature. Transfected cells were subjected to 2.5 or 10 µM 5-FU. After 48 h, apoptosis was measured via flow cytometry (NovoCyte; ACEA Biosciences, Inc.; Agilent Technologies, Inc.) and analyzed using NovoExpress 1.2.5 (ACEA Biosciences, Inc.; Agilent Technologies, Inc.).

Cells were lysed in cold lysis buffer and maintained on ice for 15 min. After centrifugation at 16,000 × g for 15 min at 4°C, protein concentration in the cell supernatant was detected using the Bradford Protein Concentration Determination kit (Beyotime Institute of Biotechnology). Subsequently, caspase-3 and activities of the CRC cells were measured using the Caspase 3 Activity Assay kit (cat. no. C1116; Beyotime Institute of Biotechnology) and caspase-9 activities were detected using the Caspase 9 Activity Assay kit (cat. no. BC3890; Beijing Solarbio Science & Technology Co., Ltd.), following the manufacturers' protocol.

Statistical analysis was performed using GraphPad Prism (v8.0; GraphPad Software, Inc.) and data were presented as the mean ± SD (n=3). IC₅₀ values were measured using GraphPad Prism. The IC₅₀ value represents the drug concentration at which tumor cells are inhibited by half. Unpaired two-tailed Student's t-test was used for comparisons between two groups, while one-way or two-way ANOVA with Tukey's post-hoc test was used for multiple comparisons to analyze statistically significant differences. P<0.05 was considered to indicate a statistically significant difference.

The effect of 5-FU on two CRC cell lines, including HT-29 and HCT-116 cells, was assessed via CCK-8 assay. The cell viability of both CRC cell lines was significantly decreased in a concentration-dependent (0.65–80 µM) manner compared with the vehicle group (0 µM 5-FU) (Fig. 1A; P<0.05). HCT-116 cells were more sensitive to 5-FU than HT-29 cells (IC₅₀ 6.343±0.767 and 30.510±3.440 µM, respectively). A previous study demonstrated that miR-9-5p expression was markedly lower in CRC cells compared with that in normal colorectal epithelial cells (13); thus, the present study further assessed miR-9-5p expression in 5-FU-treated CRC cells. The RT-qPCR results revealed that treatment with 5-FU significantly upregulated miR-9-5p expression in both CRC cells in a dose-dependent manner compared with the vehicle group (Fig. 1B; P<0.05).

To explore the effect of miR-9-5p in overcoming 5-FU resistance, HT-29 cells with a high IC₅₀ value were transfected with miR-9-5p agomir or agomir NC, and HCT-116 cells with a low IC₅₀ value were transfected with miR-9-5p antagomir or antagomir NC. The results revealed that miR-9-5p overexpression significantly increased miR-9-5p expression in HT-29 cells, while miR-9-5p expression was significantly downregulated in HCT-116 transfected with antagomiR-9-5p (Fig. 2A; P<0.05). The CCK-8 analysis indicated that after transfection with miR-9-5p agomirs and antagomirs, treatment with 5-FU significantly decreased CRC cell viability in a dose-dependent manner (Fig. 2B; P<0.05). Upregulation of miR-9-5p significantly enhanced HT-29 cell sensitivity to 5-FU, with the IC₅₀ value decreasing from 28.488±4.069 µM in the agomir NC group to 15.680±2.001 µM in the miR-9-5p agomir group, whereas miR-9-5p-knockdown decreased HCT-116 cell sensitivity to 5-FU, with the IC₅₀ value increasing from 6.464±0.686 to 17.586±2.501 µM in the antagomir NC and miR-9-5p antagomir groups, respectively (Fig. 2B; P<0.05). Additionally, based on the IC₅₀ value of the two CRC cell lines (HT-29 and HCT-116), one-third of the IC₅₀ value (2.5 µM 5-FU for HCT-116 cells and 10 µM 5-FU for HT-29 cells) was chosen to explore the impact of miR-9-5p on the sensitivity of CRC cells to 5-FU. The results revealed that miR-9-5p overexpression significantly induced HT-29 apoptosis, whereas knockdown of miR-9-5p significantly repressed the apoptosis of HCT-116 cells (Fig. 2C; P<0.05). Hoechst staining revealed that the nuclei of apoptotic cells were densely stained after transfection with the miR-9-5p agomir compared with the agomir NC, while treatment with miR-9-5p antagomir decreased the staining number of apoptotic cells compared with the antagomir NC (Fig. 2D). Corresponding quantitative analysis revealed that the percentage of apoptotic cells was significantly increased by miR-9-5p agomir and significantly decreased by miR-9-5p antagomir compared with the corresponding NCs (Fig. 2D; P<0.05). Furthermore, the expression levels of apoptosis-associated markers, including caspase-3 and caspase-9, were significantly upregulated after miR-9-5p overexpression and downregulated after miR-9-5p-knockdown (Fig. 2E; P<0.05).

To clarify the functional effect of miR-9-5p in CRC cell sensitivity to 5-FU, bioinformatics analysis was applied to predict its target genes. The results revealed that HMGA2 was one of target genes of miR-9-5p. Therefore, HMGA2 expression was determined by RT-qPCR and western blot assays. The results indicated that compared with the vehicle group (0 µM 5-FU), treatment with 5-FU decreased HMGA2 expression in the two CRC cell lines (Fig. 3A and B). Additionally, miR-9-5p overexpression significantly downregulated HMGA2 expression in HT-29 cells, whereas miR-9-5p silencing significantly upregulated HMGA2 expression in HCT-116 cells compared with their respective controls (Fig. 3A and B; P<0.05). TargetScan predicted that miR-9-5p potentially bound to HMGA2 mRNA. There was a targeted binding association between miR-9-5p and HMGA2 (Fig. 3C). Results of the luciferase reporter assay verified that miR-9-5p overexpression significantly repressed the luciferase activity of the reporter via binding to the WT but not the MUT 3′-UTR of HMGA2 (Fig. 3C; P<0.05).

To confirm whether HMGA2 was involved in miR-9-5p-mediated CRC drug resistance response, HMGA2-overexpression plasmid was co-transfected with miR-9-5p agomir into HT-29 cells. Additionally, a HMGA2 siRNA was co-transfected with miR-9-5p antagomir into HCT-116 cells. Fig. 4A shows that HMGA2 expression was significantly upregulated in HT-29 cells after transfection with over-HMGA2 and significantly downregulated in si-HMGA2-transfected HCT-116 cells (P<0.05). After co-transfection with miR-9-5p agomir + vector/over-HMGA2 into HT-29 cells, and miR-9-5p antagomir + siRNA NC/si-HMGA2 into HCT-116 cells, CCK-8 assays revealed that overexpression of HMGA2 significantly increased the viability of miR-9-5p agomir-transfected HT-29 cells, whereas the viability of miR-9-5p antagomir-transfected HCT-116 cells was significantly inhibited by HMGA2-knockdown (Fig. 4B; P<0.05). Results of flow cytometry analysis exhibited that the apoptotic rates in miR-9-5p agomir-transfected HT-29 cells were significantly decreased after HMGA2 overexpression, while increased apoptosis was observed in miR-9-5p antagomir-transfected HCT-116 cells following downregulation of HMGA2 expression (Fig. 4C; P<0.05). A similar trend was observed in the caspase activity assays. As shown in Fig. 4D, HMGA2 overexpression significantly decreased the levels of caspase-3 and caspase-9 in miR-9-5p agomir-transfected HT-29 cells, while HMGA2-knockdown significantly increased the levels of caspase-3 and caspase-9 in miR-9-5p antagomir-transfected HCT-116 cells (P<0.05). The current findings indicated that HMGA2 overexpression reversed miR-9-5p-induced apoptosis of HT-29 cells.