The present study was approved by the Ethics Committee of the Fujian Medical University Union Hospital (Fuzhou, China; reference no. 2014KY031) and written informed consent was obtained from each patient. A total of 105 paraffin-embedded GC tissue samples and 43 adjacent normal gastric mucosa tissue samples were randomly selected from the Department of Pathology, Fujian Medical University Union Hospital (Fuzhou, China), for retrospective study. The GC cases consisted of 80 men and 25 women, aged 34–81 years (mean age, 57 years), between October 2011 and December 2014. Tumors were staged according to the 7th edition of the AJCC Cancer Staging Manual: Stomach (26). The clinicopathological characteristics of the patients are summarized as follows (some data missing): Tumor size (<5 cm, 53 cases; ≥5 cm, 52 cases); tumor invasion (T1-T2, 28 cases; T3-T4, 76 cases); differentiation (poor, 52 cases; middle-well, 42 cases); clinical stage (I+II, 44 cases; III+IV, 61 cases); lymph nodes metastasis (negative, 26 cases; positive, 78 cases). All patients were treated by radical surgical resection and had not received chemotherapy or radiotherapy prior to surgery.

Formalin-fixed and paraffin-embedded specimens from the Department of Pathology were sectioned at a thickness of 4-µm, then deparaffinized twice in 100% dimethylbenzene for 30 min, then rehydrated in a graded ethanol series (100, 95, 70 and 50%). The slides were placed in antigen retrieval buffer (sodium citrate 10 mM, pH 6.0) and microwaved at high power for 15 min, followed by blocking endogenous peroxidase activity in 0.3% hydrogen peroxide for 20 min at room temperature. Protein expression was detected using the following primary antibodies: 12-LOX (dilution 1:100; cat. no. GTX80966), E-cadherin (dilution, 1:500; cat. no. GTX100443) and N-cadherin (dilution 1:150; cat. no. GTX12221; all GeneTex, Inc., Irvine, CA, USA), which were incubated with the sections overnight at 4°C. The slides were then incubated with secondary antibodies for 20 min in a humidified chamber at 37°C. All slides were stained with 10% 3′-diaminobenzidine (OriGene Technologies, Inc., Beijing, China) for 2 min at room temperature, washed with PBS and then stained with 0.1% hematoxylin for 3 min at room temperature. The sections were washed again with PBS and then washed with running tap water for 10 min. All the slides were observed under a light microscope (magnification, ×400) in 10 randomly selected fields of view, and qualitatively scored by 2 pathologists. The final score was based on the percentage of positively stained cells as follows: 0, 1%; 1, 1–25%; 2, 25–50%; 3, 50–75%; and 4, >75%. This was multiplied by the intensity of the staining, which was scored as follows: 0, none; 1, weak; 2, moderate, and 3, strong. Scores of ≤3 were considered to indicate negative expression, whereas scores ≥4 were considered to indicate positive expression.

The human GC cell line, SGC-7901 (The Department of Gastroenterology, Fujian Medical University Union Hospital), was cultured in DMEM (HyClone; GE Healthcare, Chicago, IL, USA) supplemented with 10% fetal bovine serum (HyClone; GE Healthcare) at 37°C in 5% CO₂.

The lentivirus vector for 12-LOX gene overexpression was constructed by Shanghai GeneChem Co., Ltd. (Shanghai, China). An empty green fluorescent protein (GFP) vector was used as a negative control (Shanghai GeneChem Co., Ltd.). A total of 6×10⁴ SGC-7901 cells were seeded in each well of a 6-well plate. When confluency reached 30–40%, the 12-LOX overexpression vector (2×10⁷ TU/ml) or the empty GFP vector (2×10⁷ TU/ml) was transfected into the SGC-7901 cells. The status of cells was observed by fluorescence microscope (magnification, ×200) for 8–12 h, according to the lentivirus vector manufacturer's instructions, and the medium was replaced. When confluency reached 90%, the cells was transferred into a culture flask. Following transfection, cells were used in subsequent experiments after 2 passages. Three groups of cells were used in the subsequent experiments: Untransfected cells (control group), cells transfected with the empty GFP vector (LV-vector group) and cells transfected with the 12-LOX expression vector (LV-12-LOX group).

A total of 4×10⁵ cells were seeded into each well of a 6-well plate and allowed to grow in DMEM supplemented with 10% fetal bovine serum (both HyClone; GE Healthcare, Chicago, IL, USA) until ~100% confluent. A 200-µl pipette tip was used to scratch across the cell monolayer. Cellular debris was removed by washing with PBS three times, and the plate was cultured for another 24 h. Images were captured at 0 and 24 h after wounding using an inverted light microscope. The extent of wound-healing was quantified using the following formula: (W_{0 h}-W_{24 h})/W_{0 h}, where W_{0 h} and W_{24 h} represent the width of the wound at 0 and 24 h, respectively.

A 24-well plate containing Transwell inserts with a pore size of 8 µm (Merck KGaA, Darmstadt, Germany) was coated with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) for the invasion assay and left uncoated for the migration assay. The following steps were then performed for the two assays. A total of 1.5×10⁴ cells suspended in serum-free DMEM were placed in the upper chamber. DMEM containing 10% fetal bovine serum was placed in the lower chamber as a chemoattractant. The cells were incubated for 24 h at 37°C prior to removal of cells remaining on the upper side of the chamber with a cotton swab. Cells on the lower side of the membrane were fixed in 4% paraformaldehyde for 10 min at room temperature and dyed with 0.1% crystal violet for 15 min at room temperature. The number of cells was then counted under an inverted light microscope (magnification, ×400) by a technician blinded to the experimental settings in 5 randomly selected fields of view in each plate. The experiments were repeated 3 times.

RT-qPCR was used to detect the mRNA expression of 12-LOX and N-cadherin in the transfected and untransfected cells. Total RNA was extracted using TRIzol (Thermo Fisher Scientific, Inc., Waltham, MA, USA). cDNA was synthesized using 2 µg total RNA using the PrimeScript^™ RT reagent kit (Takara Biotechnology Co., Ltd. Dalian, China). RT-qPCR analysis was performed using SYBR^® Premix Ex Taq^™ II (Tli RNaseH Plus; Takara) with 7500 Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The PCR amplification was performed under the following conditions: 95°C for 2 min, then 45 cycles of denaturation at 95°C for 15 sec, annealing at 63°C for 15 sec and elongation at 72°C for 20 sec. The primer sequences for 12-LOX, N-cadherin and GAPDH are listed in Table I (Shanghai Yaolin Bio-Technology Co., Ltd., Shanghai, China). The gene expression levels were normalized to GAPDH and are presented as relative fold change compared to the control (27). The data were analyzed using the 2^−ΔΔCq method. A total of 3 independent repeat experiments were performed.

Total protein was extracted using cell lysis buffer (Beyotime Institute of Biotechnology, Haimen, China) containing 1 mmol/l phenylmethylsulfonyl fluoride. Protein concentration was determined using a bicinchoninic acid assay kit (Beyotime Institute of Biotechnology). A total of 40 µg protein per lane was subjected to 10% SDS-PAGE and transferred to a nitrocellulose membrane (GE Healthcare, IL, Chicago). The membrane was blocked in 5% skimmed milk dissolved by PBS for 2 h at room temperature and subsequently incubated with the following rabbit polyclonal primary antibodies overnight at 4°C: 12-LOX (dilution 1:80; cat. no. GTX80966), N-cadherin (dilution 1:80; cat. no. GTX12221; both GeneTex, Inc.) or GAPDH (dilution 1:1,500; cat. no. TA309157; OriGene Technologies, Inc.). The membrane was then incubated with the appropriate HRP-conjugated Goat anti-Rabbit IgG (H+L) (dilution 1:2,000; cat. no. ZB2301; OriGene Technologies, Inc.) for 1 h at room temperature. The proteins were visualized using an ECL Advance Detection system (Origene Technologies, Inc.).

Data are expressed as the mean ± standard deviation. The differentiation between expression levels and clinicopathological parameters were analyzed using the χ² test. The associations between groups were compared using analysis of variance followed by the Least-Significant-Difference test. Statistical analyses were performed using SPSS 19.0 software (IBM Corp., Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

Immunohistochemical analysis was used to evaluate the protein expression levels of 12-LOX, N-cadherin and E-cadherin in GC tissue and adjacent normal mucosa tissue. It was demonstrated that 12-LOX and N-cadherin were mainly expressed in the cytoplasm of tumor cells, whereas E-cadherin was localized in the cell membrane (Fig. 1). The protein expression level of 12-LOX and N-cadherin was significantly increased in GC tissue compared with that in adjacent normal gastric mucosa tissue (P<0.05; Table II). By contrast, the expression of E-cadherin was significantly decreased in GC tissues compared with that in adjacent normal gastric mucosa tissue (P<0.05). The associations between patient clinicopathological characteristics and the protein expression levels of 12-LOX, N-cadherin and E-cadherin were examined. As demonstrated in Table III, there was a close association between the levels of E-cadherin protein expression and two factors, patient age and tumor differentiation (P<0.05), whereas the level of N-cadherin protein expression was associated with tumor size (P<0.05). 12-LOX expression was higher in tumors ≥5 cm, with a depth of invasion of T3-T4 and clinical stage of III+IV, however, these differences were not statistically significant (P>0.05). The expression of 12-LOX was positively associated with that of N-cadherin in GC tissues (r=0.263; P=0.007; Table IV).

A stably transfected 12-LOX-overexpressing GC SGC-7901 cell line was established, and the green fluorescent cells were >95% confluent prior to fluorescence microscopy (magnification, ×200). The results of RT-qPCR and western blotting revealed that the expression of 12-LOX in the LV-12-LOX group was significantly increased compared with that in the LV-vector and control groups (Fig. 2A).

The results of the scratch wound-healing (Fig. 2B) and Transwell (Fig. 2C) assays demonstrated that 12-LOX promotes the migration and invasion of GC cells in vitro. Significantly more cells migrated in 24 h in the LV-12-LOX group compared with those in the LV-vector and control groups (P<0.05). The number of cancer cells that invaded through the Matrigel was significantly higher in the LV-12-LOX group compared with that in the LV-vector and control groups (P<0.05).

The results of RT-qPCR revealed that the mRNA level of N-cadherin was significantly increased in the LV-12-LOX group compared with that in the LV-vector and control groups (Fig. 3A; P<0.05). Western blotting revealed that the protein expression level of N-cadherin was markedly increased in the LV-12-LOX group compared with that in the LV-vector and control groups (Fig. 3B).