The human cholangiocarcinoma cell lines SK-ChA-1 and TFK-1 and the human umbilical vein endothelial cells (HUVECs) were purchased from Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Medical Sciences. The SK-ChA-1 cells were cultured in Dulbecco's modified Eagle medium (DMEM) and TFK-1 cells were cultured in RPMI-1640 (both from Gibco; Thermo Fisher Scientific, Inc.), supplemented with 10% foetal bovine serum (FBS; HyClone; GE Healthcare Life Sciences), 100 U/ml penicillin and 100 mg/ml streptomycin. HUVECs were cultured in endothelial cell growth medium (PromoCell GmbH). All cell cultures were maintained at 37°C in a humidified atmosphere with 5% CO₂.

For leptin treatment, cells in the log phase were treated with leptin (R&D Systems, Inc.) at 0, 50, 100 or 200 ng/ml, for 0, 24 or 48 h. The morphology of cells was imaged under a light microscope (magnification, ×400).

Negative control siRNA (si-NC; 5′-GGCCAGACTGGGAAGAAAA-3′), siRNA targeting PKM2 (si-PKM2; 5′-CUUGUCCGAUGUUACCCAATT-3′), miRNA mimics negative control (miR-NC; sense 5′-UUCUCCGAACGUGUCACGUTT-3′ and antisense 5′-ACGUGACACGUUCGGAGAATT-3′), miR-122 mimics (sense 5′-UGGAGUGUGACAAUGGUGUUUG-3′ and antisense 5′-AACACCAUUGUCACACUCCAUU-3′), miRNA inhibitor negative control (inhibitor-NC; 5′-CAGUACUUUUGUGUAGUACAA-3′) and inhibitor miR-122 (5′-CAAACACCAUUGUCACACUCCA-3′) were designed and synthesized by Shanghai GenePharma Co., Ltd.

Transfection of siRNA, miRNA mimics or inhibitor was performed using Lipofectamine 2000 reagents (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. In brief, SK-ChA-1 and TFK-1 cells were seeded on 6-well plates (5×10⁵ cells/well) until ~75% confluency was achieved. si-PKM2 (100 nM), miR-122 mimics (50 nM), miR-122 inhibitor (100 nM) or the corresponding scrambled sequence (100 nM; si-NC, miR-NC and inhibitor NC, respectively) that served as negative control were transfected into the cells in serum-free medium using Lipofectamine 2000. At 48 h post-transfection, the levels of PKM2 and miR-122 were quantified using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) to evaluate the effect of transfection.

To assess the effects of leptin on cytotoxicity and cell proliferation, TFK-1 cells were seeded into 96-well plates (Corning Incorporated) in triplicate at 5×10⁵ cells/100 μl/well. Upon treating the cells with leptin at 0, 100 or 200 ng/ml for 24 h or 48 h, 20 μl of MTT reagent (5 mg/ml) was added to each well and incubated at 37°C for a further 4 h. Following centrifugation at 250 × g for 5 min at room temperature, the supernatant was discarded from each well. DMSO (150 μl/well) was added to each well to dissolve the formazan crystals. The absorbance was measured using a microplate reader at 490 nm.

Transwell chambers coated with Matrigel matrix (BD Biosciences) and uncoated Transwell chambers (8-μm pore; Corning Incorporated) were used for the invasion and migration assays, respectively. Briefly, 1×10⁵ SK-ChA-1 or TFK-1 cells were seeded into the top chamber, and 600 μl of medium containing leptin was added to the lower chamber. After 24 h, cells that did not invade through the membrane were removed using a cotton swab. Cells that invaded the membrane and attached to the bottom side of the inserted membrane were stained with 1% crystal violet at room temperature for 10 min, imaged and counted under a light microscope (magnification, ×100).

Total RNA was extracted from SK-ChA-1 or TFK-1 cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. cDNA was then synthesized with M-MLV reverse transcriptase kit (Invitrogen; Thermo Fisher Scientific, Inc.) according to manufacturer's protocol. Briefly, 1 μg total RNA and 1 μl dNTP mix (10 mM each dATP, dGTP, dCTP and dTTP) were mixed and incubated at 65°C for 5 min. Then, the reaction mixtures, including 4 μl 5X first-strand buffer, 2 μl 0.1 mol DTT and 40 U of ribonuclease inhibitor were incubated at 37°C for 2 min, followed by incubation with 1 μl M-MLV RT at 37°C for 2 min, 50 min at 37°C and heat inactivation at 70°C for 15 min. qPCR was performed using SuperScript® III Platinum® SYBR Green One-Step qRT-PCR kit (Invitrogen; Thermo Fisher Scientific, Inc.) on the ABI PRISM 7300 Fast Real-Time PCR system (Ambion; Thermo Fisher Scientific, Inc.) according to manufacturer's protocol. The thermocycling conditions were as follows: Initial 3 min denaturation step at 95°C; followed by 40 cycles at 95°C 3 sec for denaturation; and at 60°C for 30 sec for annealing and extension. Primer sequences are listed in Table I. Each reaction was performed in triplicate, and the relative expression levels of each target gene were calculated as a ratio to that of GAPDH, the internal control, using the 2^−∆∆Cq method (20).

SK-ChA-1 or TFK-1 cells were collected and lysed using cell lysis buffer (Beyotime Institute of Biotechnology). Protein concentration was determined using a Pierce BCA Protein Assay kit (Thermo Fisher Scientific, Inc.). Subsequently, 30 μg total protein/lane were boiled at 95-100°C for 5 min and separated on a 12% SDS-PAGE gel, and then proteins were transferred onto a polyvinylidene difluoride membrane. Upon blocking the membrane in TBST buffer (10 mM Tris, pH 8.0, 150 mM NaCl, 0.5% Tween-20) containing 5% non-fat milk at room temperature for 1 h, the target protein was probed with one of the following primary antibodies at 4°C overnight: Rabbit anti-human E-cadherin, (1:1,000; cat. no. A3044), rabbit anti-human β-catenin (1:1,000; cat. no. A10834), rabbit anti-human N-cadherin (1:1,000; cat. no. A0433), rabbit anti-human vimentin (1:1,000; cat. no. A11952), rabbit anti-human VEGFA (1:1,000; cat. no. A0280), rabbit anti-human angiotensin 1 (Ang1; 1:1,000; cat. no. A945; all from ABclonal Biotech Co., Ltd.), anti-PKM2 (1:1,000; cat. no. 4053; Cell Signaling Technology, Inc.), or rabbit anti-human GAPDH (internal control; 1:5,000, AC001, ABclonal Biotech Co., Ltd.). The membrane was then incubated with horseradish peroxidase-conjugated anti-rabbit IgG (1:5,000; cat. no. 7074; Cell Signaling Technology, Inc.) at room temperature for 2 h. The signal was developed using the Pierce™ Enhanced Chemiluminescence system (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The signal density was analysed using ImageJ software (version 1.51q; National Institutes of Health), and the relative protein level was calculated as the density ratio of the target protein to GAPDH (internal control).

To assess in vitro angiogenesis, conditioned medium was collected from cholangiocarcinoma cells. Briefly, 2×10⁵ cells were seeded into 10-cm tissue culture plates. Following an overnight culture, cells were treated with 0 or 200 ng/ml leptin for 24 h. Following three washes with DMEM, cells were cultured in serum-free DMEM for a further 24 h. The conditioned medium was collected and centrifuged at 2,000 × g for 10 min at 4°C to remove any cell debris. A 24-well plate was coated with Matrigel. Upon gel solidification, HUVECs (1×10⁵ cells/well) were seeded on top of the Matrigel in triplicate, treated with a mixture of conditioned medium:EGM2 medium (volume ratio 2:1), and incubated at 37°C with 5% CO₂ for 6 h. Each well was imaged under an Olympus DP71 microscope (Olympus Corporation) at ×100 magnification, and the quantification of branching points was performed using ImageJ software (version 1.51q).

starBase (http://starbase.sysu.edu.cn/) and miRDB (http://mirdb.org) software tools were used to identify potential miRNAs that may bind to the 3′-untranslated region (3′-UTR) of the human PKM2 gene (21,22). For the luciferase reporter assay, the 3′-UTR sequence of the PKM2 gene containing the potential miR-122-binding site was cloned into a psiCHECK-2 luciferase reporter plasmid (Promega Corporation) and transfected into TFK-1 cells using Lipofectamine 2000 according to the manufacturer's protocol. Upon treating the cells with vehicle control (PBS only), or transfecting miR-122 mimics or miR-122 inhibitor for 36 h, luciferase activity was detected using the Dual Luciferase Reporter assay system (Promega Corporation) according to the manufacturer's protocol. Renilla luciferase activity was normalized to firefly luciferase activity in cells.

The present study was approved by the Institutional Review Board of Hunan Provincial People's Hospital, and written informed consent was obtained from all participants. A total of 15 patients with cholangiocarcinoma admitted to the Hunan Provincial People's Hospital between January 2016 and December 2016 were recruited for the present study. The diagnosis for cholangiocarcinoma was established by combining clinical symptoms and examination with ultrasound, computed tomography, and magnetic resonance imaging. Fifteen sex- and age-matched healthy individuals were also recruited as controls. Venous blood was collected from each participant (total 15 patients), and the serum concentration of leptin (cat. no. RAB0333; Sigma-Aldrich; Merck KGaA) and plasma concentration of PKM2 (cat. no. LS-F29732-1; LifeSpan BioSciences, Inc.) were measured by ELISA kit, following the manufacturer's protocols.

All data were analysed using SPSS 13.0 software (SPSS, Inc.) and presented as the mean ± standard deviation. The difference between two groups was analysed using an unpaired two-tailed Student's t-test. One-way analysis of variance was used for comparison among multiple groups and multiple comparisons were further performed using the post hoc Tukey's test. P≤0.05 was considered to indicate a statistically significant difference.

Accumulating evidence suggests that leptin is a pleiotropic cytokine involved in cancer development (23,24). However, in cholangiocarcinoma, few studies have investigated the functions or molecular mechanisms of leptin (19). To examine the biological significance of leptin, its biosafety and effect on the proliferation of the human TFK-1 cholangiocarcinoma cell line was first assessed using an MTT assay. As shown in Fig. 1A, no significant decreases in cell viability were observed when the cells were treated with 100 or 200 ng/ml leptin for 24 h, and leptin at 200 ng/ml promoted TFK-1 cell proliferation when detected at 48 h, suggesting that leptin is not toxic to TFK-1 cells. Next, the effect of leptin on the migratory and invasive behaviours of TFK-1 cells was examined. In response to increasing concentrations of leptin (0, 100 and 200 ng/ml), leptin significantly stimulated the migration and the invasion of cells in a dose-dependent manner, as indicated by Transwell assay (Fig. 1B and C). Monitoring the changes in cell morphology revealed that leptin transformed the cholangiocarcinoma cells from cuboidal shaped cells to elongated spindle fibroblast-like cells (Fig. 1D), resembling the changes associated with EMT. To follow up, the levels of several EMT promoters, including E-cadherin, β-catenin, N-cadherin and vimentin were measured in cell lines following leptin treatment. As shown in Fig. 1E and F, leptin treatment significantly reduced the protein expression of epithelial markers E-cadherin and β-catenin, while enhancing that of mesenchymal markers N-cadherin and vimentin in a dose-dependent manner. Collectively, these data support that leptin promotes EMT in cholangiocarcinoma cells.

To examine the activity of leptin in angiogenesis, TFK-1 cells were treated with 0 or 100 ng/ml leptin and the conditioned medium was collected from these cells. By applying the conditioned medium to HUVECs, monitoring their angiogenesis (Fig. 2A, left), and measuring the number of branching points (Fig. 2A, right), conditioned medium from leptin-treated cholangiocarcinoma cells was demonstrated to induce tubular formation of HUVECs. Concomitantly, leptin significantly increased the protein expression of VEGFA and Ang1 (Fig. 2B), which are known pro-angiogenic factors in cholangiocarcinoma cells, suggesting that leptin stimulates the pro-angiogenic capability of cholangiocarcinoma cells.

To explore the molecular mechanisms underlying the EMT- and angiogenesis-stimulating activities of leptin, PKM2, a pyruvate kinase essential for leptin-induced EMT in breast cancer, was examined (15). By measuring PKM2 on the steady-state mRNA and the protein level in TFK-1 and sk-chA-1 cells, leptin was revealed to significantly upregulate the expression of PKM2 in a dose-dependent manner (Fig. 3A). The time-course study revealed that PKM2 mRNA and protein (Fig. 3B) peaked at 24 h after leptin treatment, and gradually reduced thereafter in TFK-1 and sk-chA-1 cells. Taken together, these results suggest that leptin increased the expression of PKM2 in cholangiocarcinoma cells.

To assess the significance of upregulated PKM2 in leptin-mediated EMT or pro-angiogenesis of cholangiocarcinoma cells, the expression of endogenous PKM2 was targeted with siRNA (si-PKM2). RT-qPCR and western blot analysis demonstrated that si-PKM2 significantly and specifically knocked down the level of PKM2 in TFK-1 cells (Fig. 4A). When compared with vehicle control (PBS only) treatment, leptin alone or leptin together with si-NC, si-PKM2 completely abolished the effects of leptin on cell migration (Fig. 4B), invasion (Fig. 4C), and mesenchymal morphology (Fig. 4D). Then, the expression of EMT markers E-cadherin, β-catenin, N-cadherin, and vimentin were detected by western blot analysis, which demonstrated that si-PKM2 abolished the effects of leptin (Fig. 4E). Furthermore, conditioned medium from si-PKM2-treated cholangiocarcinoma cells abolished tubular formation of HUVECs (Fig. 4F) and inhibited the expression of VEGFA or Ang1 (Fig. 4G) in cholangiocarcinoma cells. Taken together, these data suggest that PKM2 is an essential signalling molecule for leptin-induced EMT and pro-angiogenesis of cholangiocarcinoma cells.

To identify the molecules controlling leptin-mediated regulation on PKM2, a bioinformatics search was performed, and miR-122 was predicted as a potential miRNA that interacts with the 3′-UTR of PKM2 (Fig. 5A). The luciferase reporter assay demonstrated that miR-122 mimics significantly reduced the luciferase activity of the reporter carrying the 3′-UTR of PKM2. In contrast, a mutated luciferase reporter was developed by site-directed mutagenesis, and the results suggested that mutations on the binding sites successfully abolished the suppressive effect of miR-122 (Fig. 5B). Next, miR-122 mimics and miR-122 inhibitors were used for gain-of-function and loss-of-function experiments, respectively. RT-qPCR analysis revealed that miR-122 mimics elevated the miR-122 level to ~23-fold of that in cells transfected with miR-NC, while miR-122 inhibitor significantly reduced the level of miR-122 when compared with the expression in cells transfected with inhibitor NC (Fig. 5C). Consistently, miR-122 mimics significantly reduced, while miR-122 inhibitors potently increased the endogenous mRNA (Fig. 5D) as well as the protein level of PKM2 (Fig. 5E) in cholangiocarcinoma cells. The addition of leptin reduced the inhibitory effect of miR-122 mimics on PKM2 expression (Fig. 5D and E). Furthermore, leptin significantly reduced the miR-122 level in a dose-dependent manner (Fig. 5F). Collectively, these data suggest that leptin may target miR-122, thus releasing miR-122-mediated inhibition on PKM2.

To analyse the functional significance of miR-122 as a negative regulator for leptin-mediated expression of PKM2, cholangiocarcinoma cells were treated with leptin together with miR-122 mimics. As shown in Fig. 6A and B, miR-122 mimics were sufficient to reduce leptin-induced migration and invasion of cholangiocarcinoma cells to the basal level (when cells were treated with PBS) and a level significantly lower compared with that in cells treated with leptin alone or leptin with vehicle control (leptin+NC). In addition, miR-122 mimics significantly suppressed the changes in cell morphology (Fig. 6C) and gene expression (Fig. 6D) associated with leptin-induced EMT. The conditioned medium collected from cholangiocarcinoma cells treated with leptin+miR-122 mimics presented similar activities as the medium from cells treated with leptin+si-PKM2 (Fig. 5E), disrupting tubular formation of HUVECs (Fig. 6E), which was associated with the reduced expression of VEGFA and Ang1 (Fig. 6F). Taken together, these results suggested that increasing miR-122 activity is sufficient to reverse leptin-induced EMT as well as the pro-angiogenic capability of cholangiocarcinoma cells.

The data presented above support the essential role of the leptin/miR-122/PKM2 axis in regulating EMT and angiogenesis of cholangiocarcinoma. To analyze the significance of this axis in the clinical setting, the serum levels of leptin and miR-122, and plasma level of PKM2 from patients with cholangiocarcinoma and age- and sex-matched healthy individuals were measured. As shown in Fig. 7A-C, serum leptin and PKM2 levels were significantly higher, while that of miR-122 was significantly reduced in patients when compared with the corresponding levels in healthy controls. Further correlation analysis revealed that in the patients, the serum level of leptin positively correlated with that of PKM2, but negatively with that of miR-122 (Fig. 7D).