Human immortal gastric epithelial cell line GES-1 and human gastric cancer cell lines SGC7901, MKN28, MKN45 and MGC803 were obtained from the Cell Resource Center, Shanghai Institute of Biochemistry and Cell Biology at the Chinese Academy of Sciences. Cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) (both from Gibco, Carlsbad, CA, USA) at 37°C in a humidified incubator containing 5% carbon dioxide.

The siRNA oligonucleotides targeting HOTTIP, HOXA13 and the negative control were obtained from GenePharma Co., Ltd. (Shanghai, China). Transfection of the oligonucleotides was conducted with X-tremeGENE siRNA transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN, USA) according to the manufacturer's instructions. The sequences of siRNAs used in the present study are listed in Table I.

Total RNA was extracted from cells or tissues using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The cDNA was synthesized using the RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Inc., Rockford, IL, USA). Quantitative real-time PCR was performed using a SYBR Premix Ex Taq™ II (Takara Biotechnology Co., Ltd., Dalian, China) on a Bio-Rad CFX-96 Real-Time PCR system. GAPDH was used as an internal control. The sequences of the primers are listed in Table II. All qRT-PCR reactions were performed in triplicate.

Cell proliferation was measured by the Cell Counting Kit-8 (CCK-8) assay (7Sea Biotech Co., Ltd., Shanghai, China). Cells transfected with siRNA were seeded and cultured into 96-well plates (3×10³ cells/well) in 100 µl medium. At different time points indicated in the figures, 10 µl CCK-8 solution was added into the medium and further incubated with the cells for 3 h. The optical density (OD) was measured using a microplate reader at 450 nm. The CCK-8 assays were performed in triplicate.

For colony formation assay, cells transfected with different siRNAs were seeded into 6-well plates at 300 cells/well. After 14 days of incubation, cells were fixed with methyl alcohol and stained with 0.5% crystal violet. The number of colonies (≥50 cells/colony) was counted. Each experiment was performed in triplicate.

For migration assays, 5×10⁴ cells were plated in the top chamber with a non-coated membrane (24-well insert; pore size, 8-µm; Corning, Corning, NY, USA). For invasion assays, 1.5×10⁵ cells were plated in the top chamber with a Matrigel-coated membrane (24-well insert; pore size, 8-µm; Corning). Medium without serum was used in the top chamber in both assays. Medium with 10% FBS was added to the lower chamber. After incubation for 24 h (migration assay) or 48 h (invasion assay), respectively, the cells that did not migrate or invade through the pores were removed using a cotton swab. Cells on the lower surface of the membrane were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. The number of migrated or invaded cells was counted. Each experiment was performed in triplicate.

Cells were lysed with RIPA buffer containing complete protease inhibitor mixture (Roche Molecular Biochemicals). Proteins were separated by dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes (Pall Life Sciences, Ann Arbor, MI, USA). The membranes were blocked in 5% non-fat milk and blotted with antibodies against GAPDH (1:2,000) and HOXA13 (1:200) (both from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), respectively. The membranes were then incubated with horseradish peroxidase-conjugated secondary antibodies and visualized with an enhanced chemiluminescence reagent.

A total of 50 paired gastric tissue samples (cancer lesions and adjacent non-tumor mucosae) of gastric cancer patients were obtained from the Department of General Surgery, The First Affiliated Hospital of Xi'an Jiaotong University between June 2013 and February 2014. All patients did not receive chemotherapy or radiotherapy prior to surgery. All samples were collected in the same manner. The samples were immediately frozen in liquid nitrogen and stored at −80°C until they were used. Informed consent was obtained from each patient before the surgery. The present study was approved by the Research Ethics Committee of Xi'an Jiaotong University.

Statistical analysis was performed using IBM SPSS Statistics software (IBM Corp., Armonk, NY, USA). Student's t-test for parametric variables was used. Spearman test was used to establish the correlation between HOTTIP and HOXA13. Data are presented as mean ± SEM unless otherwise indicated. All P-values were determined from two-sided tests, and statistical significance was determined based on a P-value of 0.05.

To determine the role of HOTTIP in gastric cancer, we first investigated the expression of HOTTIP in the GES-1, MKN28, MGC803, SGC7901 and MKN45 cell lines. In addition, we found that HOTTIP was upregulated in gastric cancer cell lines compared with that noted in the GES-1 cells (Fig. 1A). Then, we investigated the effect of HOTTIP on cell growth by downregulating HOTTIP expression in the SGC7901 and MKN45 cells. Efficiency of HOTTIP knockdown in the SGC7901 and MKN45 cells by three specific siRNAs was confirmed by qRT-PCR and siHOTTIP #1 was used in the following experiments (Fig. 1B). Knockdown of HOTTIP inhibited cell proliferation in the SGC7901 and MKN45 cells (Fig. 1C). The inhibition of cell growth by HOTTIP knockdown was further confirmed by colony formation assay. Downregulation of HOTTIP decreased colony numbers in the SGC7901 and MKN45 cells (Fig. 1D1 and D2). These results suggest that HOTTIP plays a growth-promoting role in gastric cancer cells.

We next investigated the effect of HOTTIP on the migration and invasion of SGC7901 and MKN45 cells. Downregulation of HOTTIP led to a 2- to 3-fold reduction in the migratory and invasive capabilities of the SGC7901 cells (Fig. 2A1 and A2). Similar results were observed in the MKN45 cells with decreased expression of HOTTIP (Fig. 2B1 and B2). These results suggest that HOTTIP promotes both migration and invasion of gastric cancer cells.

HOTTIP knockdown was previously found to lead to a reduction in HOXA gene expression in primary human fibroblasts (28), hepatocellular carcinoma (29) and pancreatic cancer cells (30,31). To ascertain whether HOTTIP exhibits a similar function in gastric cancer cells, we measured the expression of several HOXA genes (HOXA13, HOXA11, HOXA10 and HOXA9) in the SGC7901 cells treated with siHOTTIP #1. Downregulation of HOTTIP led to different degrees of decrease in the expression levels of these genes, among which HOXA13 expression was decreased the most (Fig. 3A). Downregulation of HOXA13 expression was further confirmed in MKN45 cells by qRT-PCR (Fig. 3A). Knockdown of HOTTIP inhibited the HOXA13 protein level in the SGC7901 and MKN45 cells (Fig. 3B). These results suggest that HOTTIP regulates HOXA13 expression in gastric cancer cells.

We investigated the expression of HOXA13 in the GES-1 and gastric cancer cell lines. HOXA13 was upregulated in the gastric cancer cell lines compared with GES-1, which was similar to the HOTTIP expression pattern (Fig. 4A). To investigate the role of HOXA13 in gastric cancer, three specific siRNAs against HOXA13 were used to inhibit HOXA13 mRNA expression in the SGC7901 and MKN45 cells. siHOXA13 #2 showed most significant knockdown efficiency and was used in the following experiments (Fig. 4B). siHOXA13 #2 led to a clear reduction in the protein level of HOXA13 (Fig. 4C).

Knockdown of HOXA13 also inhibited cell growth (Fig. 5A and B), migration and invasion (Fig. 5C) in the SGC7901 and MKN45 cells, which resembled the inhibitory effects of HOTTIP knockdown. These results indicate that HOXA13 was involved in HOTTIP-induced malignant phenotypes of gastric cancer cells.

To further understand the relationship between HOTTIP and HOXA13 in gastric cancer, we investigated HOTTIP and HOXA13 expression levels in 50 pairs of primary gastric cancer tissues and their counterpart non-tumorous tissues by qRT-PCR. The results showed that HOTTIP and HOXA13 were both markedly upregulated in the gastric cancer tissues when compared with these levels in the non-tumorous tissues (Fig. 6A and B), which were consistent with the expression patterns of HOTTIP and HOXA13 in the gastric cancer cells. Correlations between the HOTTIP or HOXA13 expression levels and clinicopathologic characteristics of gastric cancer are summarized in Tables III or IV, respectively. The data revealed that expression levels of HOTTIP and HOXA13 were both higher in gastric cancer which was poorly differentiated (P<0.05), at advanced TNM stages (P<0.05) and showed lymph node metastasis (P<0.01). Spearman analyses indicated that HOTTIP and HOXA13 had a positive correlation both in non-tumor mucosae (Fig. 6C) and cancer lesions (Fig. 6D). These results suggest that HOTTIP and HOXA13 are highly correlated and associated with gastric cancer progression.