The human 293 cells, the human colon cancer cell lines (LoVo, HCT 116, DLD-1, SW480, HT-29 and RKO) and normal colon epithelial cell lines (FHC and CCD-18Co) were obtained from the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (both from Invitrogen, Carlsbad, CA, USA) containing penicillin/streptomycin. The cells were grown in a humidified 5% CO₂ atmosphere at 37°C in an incubator. The LoVo cell cultures were exposed to chemotherapy with oxaliplatin as previously described (24), at the Cancer Research Institute of Guangzhou Medical University (Guangzhou, China). The oxaliplatin-resistant cell line was established from LoVo parental cells by gradually treating the cells with increasing concentrations of oxaliplatin for 3 months for the selection of oxaliplatin-resistant cells. The cells were subcultured at 10% confluency by splitting at a ratio of 1:200 and exposed to increasing concentrations of oxaliplatin (0, 50, 100 and 200 µg/ml). The cultures that survived after 8 cycles of subculturing and oxaliplatin treatment (maximum treatment, 200 µg/ml) were considered chemoresistant and used in the subsequent experiments. It should be noted that the LoVo cells were selected as the target cells for use in the experiments, as they were more invasive than the other cells (data not shown). The IC₅₀ value was read from the drug concentration which had the half maximal inhibitory effect on cell viability. In addition, the accumulation of autophagosomes in the colon cancer cells was measured using an electron microscope (Philips CM-120 transmission electron microscope; Philips, Amsterdam, The Netherlands) as previously described (12). Autophagosomes appear in cells as single membrane vesicles containing different cytoplasmatic material.

A total of 30 human samples were used in this study, including 20 human colon tumor tissue samples and 10 human normal colon tissue (adjacent) samples. The samples were obtained from patients undergoing surgery for colon cancer between 2013 and 2014 at the Department of Abdominal Surgery, The Affiliated Cancer Hospital of Guangzhou Medical University (Guangzhou, China) and stored in liquid nitrogen until analysis. The tumor and normal samples were obtained following a protocol approved by the Ethics Committee of the Department of Guangzhou Medical University, The Affiliated Cancer Hospital (Guangzhou, China). All patients provided written informed consent for the use of their samples for experimental purposes.

Female BALB/c nude mice (n=20, 6 weeks old, weighing 25–30 g) were obtained from the Medical Experimental Animal Center of Guangdong Province (Guangzhou, China) and housed under pathogen-free conditions with free access to food and water. The animal experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee of Guangzhou Medical University.

The wild-type (WT) 3′-UTR of human Beclin-1 (5′-AGGTTGAG AAAGGCGAGACA-3′) and the 3′-UTR sequences carrying mutations (C5878T, A5999T or A6328G) were amplified and subcloned into the pGL3 expression vector downstream of the promoter and coding region of firefly luciferase (Promega Corp., Madison, WI, USA); subcloning was performed using the XbaI and NotI restriction sites. The resulting expression plasmids were transfected into the chemoresistant LoVo cells in the presence or absence of 20 nmol/ml miR-409-3p mimic (Gene Pharma, Shanghai, China) using Lipofectamine transfection reagent (Invitrogen) according to the manufacturer's instructions. The cells were harvested 48 h after transfection and the luciferase activity was measured using a Dual-Luciferase Reporter assay kit (Promega Corp.). Luciferase activity was normalized to that measured in the control cells transfected with the empty pGL3 expression vector in the presence or absence of miR-409-3p mimic. In addition TargetScan was used to confirm the target of miR-409-3p.

The chemoresistant LoVo cells were pre-treated with miR-409-3p mimic or scrambled miRNA (5′-ATTAATCATAGAGGAAATCCACG-3′; Shanghai Kangcheng Biological engineering Co. Ltd. Shanghai, China) for 1 h and then exposed to oxaliplatin at 50 µg/ml for 24 h. Cell growth and viability were evaluated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Sangon Biotech, Shanghai, China). Briefly, the cells were seeded in 96-well plates at a density of 5×10³ cells/200 µl. Following transfection and chemotherapy, fresh medium containing MTT solution [5 mg/ml diluted in phosphate-buffered saline (PBS), 20 µl/well] was added followed by incubation for an additional 4 h. The resulting formazan was resuspended in dimethyl sulfoxide (200 µl/well). Finally, the absorbance was determined at 490 nm using an ELISA reader (BioTek Instruments, Inc., Winooski, VT, USA).

The colony-forming ability was assessed at 15 days following transfection. The medium was discarded and the cell colonies were stained with crystal violet (Sigma-Aldrich (Shanghai, China) prior to counting using the Leica Zoom 2000 dissecting microscope (Leica Microsystems Inc., Buffalo Grove, IL, USA). Measurements were taken in quadruplicate from 3 independent experiments.

miRNA mimics and control miRNAs were purchased from Applied Biosystems Life Technologies (Foster City, CA, USA). Lipofectamine 2000 (Invitrogen) was used for the transfection of miRNA mimics or plasmid DNA according to the manufacturer's instructions. At 48 h after transfection, the expression of miR-409-3p was detected by reverse transcription quantitative-polymerase chain reaction (RT-qPCR).

A total of 1×10⁴ cells per well were seeded in 48-well plates overnight. The medium was replaced with fresh medium with or without oxaliplatin at the indicated concentrations and incubated for 48 h. Cell viability was measured by MTT assay. The absorbance was measured spectrophotometrically at 570 nm using the Universal Microplate Reader EL800 (BioTek Instruments, Inc.).

Total RNA was isolated from the cells using TRIzol reagent (Invitrogen) and small RNA was extracted using the mirVana kit (Ambion, Austin, TX, USA) according to the manufacturer's instructions. The corresponding cDNA was generated using M-MLV reverse transcriptase (Clontech Laboratories, Palo Alto, CA, USA) and the TaqMan microRNA reverse transcription kit (Applied Biosystems, Foster City, CA, USA). The RT-qPCR mixtures, which contained cDNA templates, primers and SYBR-Green qPCR Master Mix, were subjected to RT-qPCR quantification, and Beclin-1 and miR-409-3p expression levels were quantified using the 2^−ΔΔCt method. The expression levels were expressed relative to the levels of the following internal control genes: glyceraldehyde- 3-phosphate dehydrogenase (GAPDH) for Beclin-1 or U6 small nuclear (sn)RNA for miR-409-3p.

Total cell lysates were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes (Amersham; GE Healthcare, Little Chalfont, UK). Non-fat dry milk (2.5%) was used to block the membranes, which were then incubated with primary antibody against LC3 (ab63817) and Beclin-1 (ab62557) (both from Abcam, Cambridge, UK) or cleaved caspase-3 (PC679-50UG; Millipore Corp., Boston, MA, USA) and anti-GAPDH antibody (bs-2188R; BIOSS, Beijing, China). The membranes were then incubated with horseradish peroxide (HRP)-conjugated antibody and antibody binding was visualized using enhanced chemiluminescence (ECL; Amersham; GE Healthcare). Band intensities were quantitated using Image Pro Plus 6.0 (Olympus, Osaka, Japan) and normalized to the levels of GAPDH in the same sample.

Chemoresistant LoVo cells (2×10⁶) diluted in 200 µl PBS pre-transfected with lentiviral vector expressing miR-409-3p or scrambled miRNA control were injected subcutaneously into the right groin of the BALB/c nude mice. Tumor length and width were measured daily and the volume was calculated using the following formula: length × width² × π/6. When the tumor volume reached approximately 500 mm³, the tumor was injected once daily with oxaliplatin (50 µg/ml, 1 ml) or PBS as a control and the volume was measured daily for at least 36 more days. On day 41, the animals were euthanized by cervical dislocation following exposure to CO₂. The tumors were removed and either frozen in liquid nitrogen or fixed in 10% formalin for further analysis.

Data are expressed as the means ± standard deviation (SD) of 3 independent experiments. Differences between or within groups were assessed for significance using, respectively, Student's two-tailed t-test or one-way ANOVA. The threshold for significance was defined as P<0.05.

Previous studies have indicated that miR-409-3p acts as a tumor suppressor in various types of cancer (21–23). To examine the role of miR-409-3p in human colon cancer, we measured the expression of miR-409-3p in colon cancer and normal colon cell lines. As expected, the miR-409-3p levels were significantly downregulated in all the colon cancer cell lines compared with the 2 normal cell lines, FHC and CCD-18Co (Fig. 1A). To support these findings, we compared the miR-409-3p levels between the human colon cancer samples and the adjacent benign tissue samples. Consistent with these results, miR-409-3p expression was downregulated in the colon cancer samples, compared with the normal colon tissue samples (Fig. 1B), thus suggesting that miR-409-3p acts as a tumor suppressor in colon cancer. We then transfected miR-409-3p mimics or control miRNA into the LoVo colon cancer cells. Our results revealed that cell proliferation was suppressed by the overexpression of miR-409-3p (Fig. 1C). Cell proliferation at 48, 72 and 96 h was significantly suppressed following transfection with miR-409-3p mimic. Moreover, we observed that miR-409-3p overexpression in the LoVo cells significantly inhibited the colony-forming ability of th cells (Fig. 1D). Taken together, these results revealed a tumor suppressor role for of miR-409-3p in colon cancer.

The evasion of chemotherapeutic agents has been recognized as one of the hallmarks of cancer (3). In this study, to investigate the putative roles of miR-409-3p in enhancing the chemosensitivity of colon cancer cells, we established the oxaliplatin-resistant cell line from LoVo parental cells by gradually treating the parental cells with increasing concentrations of oxaliplatin for 3 months to select oxaliplatin-resistant cells. Fig. 2A illustrates that the IC₅₀ value of the oxaliplatin-resistant cells was approximately 5-fold greater than the IC₅₀ value of the LoVo parental cells, indicating that the resistant cells may be tolerant to higher concentrations of oxaliplatin. Of note, miR-409-3p expression was significantly lower in the oxaliplatin-resistant cultures than in the parental cells (Fig. 2B), indicating that miR-409-3p may be a therapeutic target for the treatment of chemoresistant cancer cells.

The above-mentioned findings demonstrated the negative correlation between the expression of miR-409-3p and sensitivity to oxaliplatin in colon cancer cells. To elucidate the mechanisms responsible for the miR-409-3p-mediated chemo-sensitivity, we compared the expression levels of Beclin-1 and its downstream factors, LC3-I and LC3-II between the parental and oxaliplatin-resistant LoVo cells. As expected, western blot analysis revealed that Beclin-1 and LC3-II protein expression was present at much higher levels and that LC3-I protein expression was downregulated in the chemoresistant cells compared with the parental cells, indicating enhanced autophagic activity in the oxaliplatin-resistant cells (Fig. 3A). Consistent with these findings, the mRNA expression levels of LC3-I were downregulated and those of LC3-II were upregulated in the chemoresistant cells (Fig. 3B). However, we did not observe a significant difference in Beclin-1 mRNA expression levels between the resistant and sensitive cells, indicating that the post-transcriptional regulation of Beclin-1 may occur in oxaliplatin-resistant cells (data not shown). To further confirm our findings, we measured autophagosome formation in the chemoresistant (Fig. 3C, panel a) and parental (Fig. 3C, panel b) LoVo cancer cells. The LC3-II dots (representing the autophagic activity; black arrowheads represent autophagosomes) in the oxaliplatin-resistant cells were more evident than in the sensitive cells (Fig. 3C). The above-mentioned data (Figs. 2 and 3) indicate that miR-409-3p expression is downregulated in chemotherapy-resistant colon cancer cells and that autophagy is activated in these resistant cells, thus suggesting a link between the downregulation of miR-409-3p expression and the activation of autophagy.

To obtain more direct evidence that miR-409-3p regulates autophagy, we performed a screening of the possible binding sites of miR-409-3p in the 3′-UTRs of gene transcripts related to autophagy using TargetScan. This analysis highlighted Beclin-1 as a putative target of miR-409-3p. To verify that miR-409-3p targets Beclin-1, we measured the protein expression levels of Beclin-1 and LC3-I/II in the LoVo cells in response to miR-409-3p overexpression induced by transfection with miR-409-3p mimics. In the cells transfected with miR-409-3p mimics, we observed decreased Beclin-1 expression levels, as well as increased LC3-I levels and decreased LC3-II levels (Fig. 4A). To verify that miR-409-3p binds to the 3′-UTR of Beclin-1 directly, we transfected 293 cells with a dual-luciferase reporter system in which the luciferase transcript was fused to the 3′-UTR of Beclin-1. In this system, luciferase activity is lower in the presence of elements that bind to the UTR. The cells were transfected with either the reporter system alone or with the reporter system together with the miR-409-3p mimic. Other cells were co-transfected with a reporter system containing a mutated version of the Beclin-1 3′-UTR lacking the predicted miR-409-3p binding site. The relative luciferase activity was significantly lower in the cultures co-transfected with miR-409-3p mimic and pGL3 containing the WT 3′-UTR of Beclin-1 than in the cells co-transfected with the mimic and pGL3 containing the mutant 3′-UTR (Fig. 4B). These results confirmed that Beclin-1 is a target gene of miR-409-3p. Consistent with the results from western blot analysis, the mRNA levels of LC3-I and LC3-II were significantly altered by transfection with miR-409-3p (Fig. 4C). Taken together, the above-mentioned results demonstrate that miR-409-3p suppresses autophagy through the direct inhibition of Beclin-1.

To determine the effects of miR-409-3p on the drug sensitivity of colon cancer cells, we pre-treated the LoVo cells with miR-409-3p mimic or scrambled miRNA control and measured the cell viabilitys in response to treatment with oxaliplatin. Exogenous miR-409-3p overexpression led to significantly higher apoptotic levels (Fig. 5A, shown by a decrease in the number of surviving cells). These results suggest that miR-409-3p enhances the sensitivity of LoVo colon cance4r cells by triggering apoptosis. As expected, the overexpression of miR-409-3p in LoVo oxaliplatin-resistant LoVo cells sensitized the cells to oxaliplatin. The oxaliplatin-resistant cells transfected with control miRNA showed an IC₅₀ value of 250 µg/ml, which was significantly higher than that of the cells transfected with miR-409-3p mimic (Fig. 5B), indicating miR-409-3p may be a therapeutic agent which may be used to overcome resistance to oxaliplatin. To obtain additional evidence that miR-409-3p sensitizes chemoresistant LoVo cells to chemotherapy by regulating Beclin-1 expression, we co-transfected the cells with miR-409-3p mimic and an expression plasmid carrying the Beclin-1 coding sequence upstream of a 3′-UTR mutated to eliminate the miR-409-3p binding site. The overexpression of Beclin-1 (Fig. 6A) almost completely blocked the ability of miR-409-3p to inhibit the growth of chemoresistant LoVo cells, as well as the ability of miR-409-3p to trigger apoptosis (Fig. 6B). These results support the hypothesis that miR-409-3p sensitizes chemoresistant LoVo colon cancer cells to chemotherapy by targeting Beclin-1-mediated autophagy.

To verify our in vitro experiments using the LoVo cell cultures, we determine whether miR-409-3p was capable of strengthening the ability of chemotherapeutic drugs to suppress the growth of tumor implanted tumors in nude mice. Twenty animals were equally divided into 2 groups, and subcutaneously injected with LoVo cells pre-transfected with lentiviral vector expressing miR-409-3p or scrambled miRNA control. When the tumors attained a volume of approximately 500 mm³ (around day 16), the tumors were injected once daily with oxaliplatin (50 µg/ml, 1 ml). The tumors overexpressing miR-409-3p were significantly more sensitive to chemotherapy than the tumors expressing the scrambled control (Fig. 7A and B). In addition, the analysis of the xenograft tumor samples at the end of chemotherapy indicated that the tumors overexpressing miR-409-3p expressed lower levels of Beclin-1 and exhibited less autophagic activity (Fig. 7C–E). These results are consistent with those from our in vitro experiments, suggesting that miR-409-3p sensitizes LoVo cancer cells to chemotherapy by blocking Beclin-1-mediated autophagy.