The present clinical investigation (no. HMCH2010072508) was performed in strict accordance with the recommendations in the Guide for Haidian Maternal and Child Healthcare Center (Beijing, China) between May 2010 and October 2015. A total of 12 female patients with ovarian fibroma were required to review trial protocols and amendments, and to provide written informed consent. The present study was approved by the ethics committee of Haidian Maternal and Child Healthcare Center. The demographic and clinical pathological characteristics of the patients are summarized in Table I.

Ovarian cancer cells (1×10⁷) were isolated from patients with ovarian cancer using tumor cell separation methods (19). Ovarian cancer tissues and adjacent normal tissues were washed with PBS. Tumor tissue was cut into pieces and separated into individual cells using minimum essential medium (MEM) containing 5% pancreatin and penicillin/streptomycin (2 mM) (both from Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Cells were maintained for 12 h at 37°C in the presence of 5% CO₂. Subsequently, cells were filtered, collected and identified by microscopic investigation (20). Ovarian cancer cells were cultured in MEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.). Cells were cultured in a humidified atmosphere containing 5% CO₂ at 37°C in a cell culture incubator (Gibco; Thermo Fisher Scientific, Inc.).

Ovarian cancer cells transfected with siRNA or the eukaryotic expression vector for LKB1 were homogenized in lysate buffer containing protease-inhibitor (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and were centrifuged at 8,000 × g at 4°C for 10 min. The supernatant was used for analysis of the total protein using a bicinchoninic protein assay kit (Gibco; Thermo Fisher Scientific, Inc.). Protein samples (20 µg) were separated on 12% sodium dodecyl sulfate polyacrylamide gels and transferred onto polyvinylidene fluoride membranes (EMD Millipore, Billerica, MA, USA) as previously described (21). Protein was blocked with 5% bovine serum albumin reagent (Roche Diagnostics, Basel, Switzerland) for 1 h at 37°C. For western blotting, primary goat anti-human antibodies against TGF-β (cat. no. ab31013), zinc-finger protein SNAI1 (cat. no. ab53519), Snai2 (cat. no. ab187109), Twist-related protein 1 (cat. no. ab50887), zinc finger E-box-binding homeobox 1 (ZEB1; cat. no. ab71286), E-cadherin (cat. no. ab76319), vimentin (VIM; cat. no. ab137321), VEGF (cat. no. ab27278), angiotensin-1 (Ang-1; cat. no. ab53951) and β-actin (cat. no. ab8226) (all 1:500 dilution; Abcam, Cambridge, UK), were incubated overnight at 4°C, followed by incubation with horseradish peroxidase-conjugated polyclonal anti-rabbit immunoglobulin G antibody (1:10,000, cat. no. HAF008; R&D Systems, Inc., Minneapolis, MN, USA) for 1 h at room temperature. A Ventana Benchmark automated staining system was used for analyzing protein expression (Olympus BX51; Olympus, Tokyo, Japan).

Ovarian cancer cells (1×10⁶) were transfected with HiPerFect reagent (Qiagen GmbH, Hilden, Germany), according to the manufacturer's instructions (22). siRNA-TGF-β (100 pmol), siRNA-LKB1 (100 pmol) and siRNA-vector (100 pmol) were transfected into ovarian cancer cells for 24 h at 37°C. The sequences of siRNA-TGF-β, siRNA-LKB1 and siRNA-vector were designed and are listed in Table II. siRNA oligonucleotide pools containing three sequences targeting TGF-β or LKB1 were purchased from Eurogentec, Ltd. (Liège, Belgium).

In order to establish stable ovarian cancer cells with endogenous LKB1 expression, the eukaryotic expression vector pCMVp-NEO-BAN (Takara Biotechnology Co., Ltd., Dalian, China) was used to construct recombinant plasmids. LKB1 was cloned, sequenced and recombined into pCMVp-NEO-BAN to construct pCMVp-NEO-LKB1 (pLKB1). The recombinant plasmid pCMVp-NEO-LKB1 was subsequently transfected into ovarian cancer cells. pLKB1 (1.0 µg) or pvector (1.0 µg) was transfected into cultured ovarian cancer cells (5×10⁶) using Lipofectamine 2000 (Sigma-Aldrich; Merck KGaA), according to the manufacturer's instructions. Stable LKB1-overexpressing ovarian cancer cells were selected by G418 screening (23). After 48 h transfection, LKB1-overexpressing ovarian cancer cells were used to subsequent experimentation.

Apoptosis analysis of ovarian tumor cells was performed using flow cytometry. Human ovarian tumor cells (5×10⁶) were cultured in 6-well plates with paclitaxel (2.0 mg/ml; Sigma-Aldrich; Merck KGaA) for 48 h to achieve the maximal apoptosis rate. Ovarian tumor cells were harvested at 48 h post-treatment by trypsinization. Ovarian tumor cells were subsequently washed in cold PBS and adjusted to 1×10⁶ cells/ml with PBS. Following double staining with fluorescein isothiocyanate (FITC)-Annexin V and propidium iodide using the FITC Annexin V Apoptosis Detection kit I (BestBio Biotechnology, Shanghai, China), cells were analyzed using a FACScan^® flow cytometer equipped with Cell Quest software (version 3.3; BD Biosciences, San Jose, CA, USA), according to manufacturer's instructions, to detect apoptosis in ovarian tumor cells. All experiments were performed in triplicate.

Ovarian cancer cells were incubated (1×10³) with paclitaxel or PBS in 96-well plates for 48 h in triplicate. Subsequently, 20 µl MTT (5 mg/ml) in PBS solution was added to each well, the plate was further incubated for 4 h. Most of the medium was removed and 100 µl dimethyl sulfoxide was added into the wells to solubilize the crystals. The OD was measured using an ELISA reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at wavelength of 450 nm.

Ovarian tumor tissues were fixed in situ overnight in 10% buffered formalin. The fixed tissues were cut mid-sagittal and being embedded in paraffin (4 µm thickness) using standard protocols. Hematoxylin and eosin staining was used to visualize the area of myocardial infarction after treatment with matrine. Immunohistochemical staining was performed using an avidin-biotin-peroxidase technique. Tumor sections (4 µm) were deparaffinized in xylene, dehydrated through graded ethanol and treated with 0.3% hydrogen peroxide in methanol for 30 min at 37°C. Paraffin-embedded ovarian normal tissue and tumor sections were prepared and epitope retrieval was performed at 95°C for 15 min for further analysis. The paraffin sections were treated with hydrogen peroxide (3%) for 10–15 min, which subsequently was blocked with a blocking solution (5% skim milk powder) for 10–15 min at 37°C. Subsequently, the sections were incubated in goat anti-human anti-Snai2 (1:1,000, cat. no. ab187109), Twist (1:1,000, cat. no. ab50887), ZEB1 (1:1,000, cat. no. ab71286), LKB1 (1:1,000, cat. no. ab15095), SIK1 (1:1,000, cat. no. ab64428), phosphorylated (p-)LKB1 (1:1,000, cat. no. ab63473), p-SIK1 (1:1,000, cat. no. ab217809), TGF-β (1:1,000, cat. no. ab31013) and VIM (1:1,000, cat. no. ab137321) at 4°C for 12 h. All sections were washed three times and incubated with secondary rabbit anti-goat antibodies (1:2,000, cat. no. ab150117; Abcam) for 1 h at 37°C. For immunofluorescence, ovarian tumor cells were stained with goat anti-human vascular endothelial growth factor and angiopoietin-1 antibodies. Additionally, a terminal deoxynucleotidyl transferase dUTP nick end labeling assay was performed using a Peroxidase Apoptosis Detection kit (Chemicon; EMD Millipore). All sections were observed in six random fields in the confocal microscope at magnification, ×40 (Nikon E400, Nikon Corporation, Tokyo, Japan).

To analyze the effects of LKB1 overexpression on the cell cycle stage of ovarian cancer cells, flow cytometry was performed. Exponentially, culturing ovarian cancer cells (1×10⁶) or LKB1 overexpression were cultured for 24 h at 37°C. Cells were washed and trypsinized and rinsed with phosphate-buffered saline (PBS). All cells were fixed in 75% ice-cold ethanol for 5 min and then washed with PBS three times. The fixed cells were washed with RNase A (20 µg ml/l, Fermentas; Thermo Fisher Scientific, Inc.) and stained with propidium iodide (20 µg ml/l, Sigma-Aldrich; Merck KGaA) for 10 min at 37°C. The percentages of cells in G1 phase were analyzed using BD FACSCalibur (Becton Dickinson; BD Biosciences, San Jose, CA, USA).

Ovarian tumor cells subjected to different treatments (LKB1 overexpression or TGF-β inhibition) were used to analyze invasion and migration. Migration and invasion assays in ovarian tumor cells were conducted in a 24-well MEM culture plate with chamber inserts (BD Biosciences) with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) for 12 h at 37°C. For migration assays, 1×10³ cells/well ovarian tumor cells were placed into the upper chamber with a non-coated membrane. For the invasion assays, cells (1×10³ cells/well) were placed into the upper chamber with a Matrigel-coated membrane. Matrigel were fixed with 4% formaldehyde and stained with 4′,6-diamidino-2-phenylindole as well as counted in 6 random fields under a microscope. Invasion and migration were calculated in at least three randomly stained fields under a light microscope (Nikon E400; Nikon Corporation).

All data are presented as the mean ± standard error of triplicate samples, and analyses were performed using Prism 6.0 software (GraphPad Software, Inc., La Jolla, CA, USA). Statistical differences between experimental groups were analyzed using Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

The expression levels of EMT molecules, LKB1 and SIK1 were measured in ovarian tumor tissues. As presented in Fig. 1A, expression of the EMT pathway components Snai2, Twist and ZEB1 was upregulated in ovarian tumor tissues compared with normal ovarian tissues. The results in Fig. 1B and C demonstrated that the expression and phosphorylation levels of LKB1 and SIK1 were downregulated in ovarian tumor tissues compared with normal ovarian tissues. It was additionally observed that the expression levels of TGF-β and VIM were markedly increased in ovarian tumor tissues compared with normal ovarian tissues (Fig. 1D). The results of the present study demonstrated that the knockdown of LKB1 expression promoted TGF-β and Snai1 expression in ovarian tumor cells (Fig. 1E). Additionally, it was observed that the knockdown of LKB1 promoted the apoptosis resistance of ovarian tumor cells treated with paclitaxel (Fig. 1F). The present results suggested that the expression levels of EMT, LKB1 and SIK1 were increased in ovarian tumor tissues, which may be associated with the aberrant growth and aggressiveness of ovarian tumor cells.

As presented in Fig. 2A, the growth rate of ovarian tumor cells isolated from clinical patients was increased compared with normal ovarian cells. Proliferation and migration assays demonstrated that growth and aggressiveness was inhibited by the inhibition of TGF-β expression in ovarian tumor cells (Fig. 2B and C). Apoptosis experiments demonstrated that the inhibition TGF-β expression by siRNA promoted apoptotic sensitivity in cells treated with paclitaxel for 48 h (Fig. 2D). It was observed that the inhibition of TGF-β expression by siRNA inhibited the proliferation of ovarian tumor cells compared to si-vector-transfected cells (Fig. 2E). Additionally, it was observed that the inhibition of TGF-β expression suppressed the expression levels of Snai2, Twist and ZEB1, as determined by western blotting (Fig. 2F). The results of the present study suggested that TGF-β expression may be associated with the growth and apoptotic resistance of ovarian tumor cells.

The present study further analyzed the influences of LKB1 on TGF-β expression and the EMT signaling pathway in ovarian tumor cells. The results in Fig. 3A demonstrated that SIK1 expression was promoted by LKB1 upregulation in ovarian tumor cells, as determined by immunofluorescence analysis. Western blotting demonstrated that TGF-β expression was decreased by LKB1 (Fig. 3B). The apoptosis assay indicated that the apoptosis sensitivity of ovarian tumor cells was increased following upregulation of LKB1 (Fig. 3C). In addition, the results demonstrated that the expression of important regulatory factors in the EMT pathway, E-cad, VIM, Snail2 and γH2AX, was downregulated by LKB1 upregulation in ovarian tumor cells (Fig. 3D). Growth, migration and invasion assays demonstrated that growth and aggressiveness was inhibited following overexpression of LKB1 expression in ovarian tumor cells (Fig. 3E and F). The present findings suggested that activation of the LKB1 signaling pathway may inhibit the TGF-β-mediated EMT pathway and decrease growth, aggressiveness and apoptosis resistance in ovarian carcinoma cells.

The present study investigated the effects of LKB1 on cells survival- and apoptosis resistance-associated protein expression in ovarian tumor cells. The results demonstrated that LKB1 overexpression inhibited the survival and cell cycle progression of tumor cells treated with paclitaxel, compared with control cells (Fig. 4A and B). Western blotting demonstrated that the expression of Bcl-2, Bax and p53 was inhibited by LKB1 overexpression in ovarian tumor cells compared with control cells (Fig. 4C). However, caspase-3, caspase-8 and Apaf-1 expression was increased by LKB1 overexpression in ovarian tumor cells compared with control cells (Fig. 4D). The expression of aggressiveness-associated proteins FN and RNF-8 was decreased by LKB1 overexpression in ovarian tumor cells (Fig. 4E). Additionally, the results demonstrated that the expression levels of vascular endothelial growth factor (VEGF) and angiotensin-1 (Ang-1) were downregulated by LKB1 overexpression, as determined by immunofluorescence analysis (Fig. 4F). The present results suggested that LKB1 overexpression may inhibit aggressiveness- and apoptosis resistance-associated protein expression in ovarian tumor cells.