In total, 196 patients with gastric cancer, including 132 males and 64 females (mean age, 57 years; age range, 26–80 years) were included in the present study and were diagnosed and surgically treated at Peking University Cancer Hospital (Beijing, China) between 1999 and 2007. Primary gastric carcinoma tissues and matched non-cancerous mucosal tissues were obtained from the patients and were fixed with 10% formaldehyde in PBS for immunohistochemistry. The investigations were performed following approval by the Ethics Committee of Peking University. General informed consent was obtained from each participant involved in the study.

The AGS and KATO 3 gastric cancer cell lines were obtained from American Type Culture Collection (Manassas, VA, USA). The BGC823 gastric cancer cell line was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The cell lines were routinely grown in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% (v/v) fetal calf serum (FCS; Gibco; Thermo Fisher Scientific, Inc.) and antibiotics at 37°C in a humidified 5% CO₂ atmosphere.

Sections (4 µm) of the formalin-fixed, paraffin-embedded tissues were mounted on poly-L-lysine-coated slides, deparaffinized in xylene, rehydrated with alcohol and rinsed with distilled water. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 15 min at room temperature. Following heating the slides under pressure (120°C and 103 kPa/15 psi) in 10 mmol/l EDTA (pH 8.0) for 3 min, the sections were incubated overnight at 4°C with mouse anti-S100A6 monoclonal antibody (1:500; cat. no. H00006277-M16; Abnova, Taipai, Taiwan), or mouse Ki-67 monoclonal antibody (1:100; cat. no. MS-1794-S0; LabVision, Fremont, CA, USA). Primary antibodies were detected using a two-step EnVision system (Dako, Glostrup, Denmark). As a negative control, the primary antibody was replaced with nonimmune mouse serum (Dako) to confirm its specificity.

Evaluation of the S100A6 staining was performed using previously described scoring criteria (20). The recorded information included the subcellular location of the S100A6 staining (nuclear and/or cytoplasmic), the intensity of staining (negative, weak, moderate and strong) and the percentage of cells demonstrating positive immunoreactivity.

The reactivity of Ki-67 was evaluated by counting the number of positive and negative tumor cell nuclei in at least four randomly selected fields in each section under a light microscope. The slides were visualized by two pathologists with no knowledge of the clinical data. Reproducibility of the scoring method between the two observers was >90%. In the remaining cases, if discrepancies were noted, the differences were settled through consensus following a review of the corresponding slides.

For double staining, mouse anti-human S100A6 antibody (1:500; Abnova) and rabbit anti-human Ki-67 antibody (1:100; cat. no. RM-9106-S0; LabVision) were used as the primary antibodies. The secondary antibodies used for double staining were fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse antibody and rhodamine (TRITC)-conjugated goat anti-rabbit antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA). Confocal images were captured using a Leica TCS SP5 confocal microscope (Leica Microsystems GmbH, Mannheim, Germany). The excitation wavelength for TRITC was 586 and for FITC was 488 nm. In addition, the nuclei of specimens were simultaneously stained with DAPI, with excitation at 358 nm.

The post-embedding immunogold method was used to detect the S100A6 proteins. The anti-S100A6 antibody, obtained from Abnova, was used at a dilution of 1:40. The gastric cancer tissues (1 mm³ in size) were immersed for 6 h in fresh glutaraldehyde fixative solution, following which the tissues were washed in 0.2 M sucrose and fixed in 1% osmium tetroxide for 4 h at 4°C. Subsequently, the tissues were dehydrated using a graded acetone series (−4°C), embedded in epoxy resin, and polymerized at 37°C for 24 h and 60°C for 48 h. Ultrathin sections (60 nm) were obtained on an ultramicrotome and collected on nickel grids. All sections were incubated in 10% H₂O₂, rinsed with phosphate-buffered saline enriched with 1% bovine serum albumin (Gibco; Thermo Fisher Scientific, Inc.) and 0.5% powdered skim milk three times (5 min each), and then incubated with the primary antibodies at 4°C overnight. Following incubation, the sections were washed three times in TBS (pH 7.6), following which the sections were incubated with goat anti-mouse IgG antibody conjugated to 10 nm gold particles (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) at a dilution of 1:100 in TBS (pH 7.6). The sections were rinsed three times (10 min each) in enriched TBS, rinsed three times (10 min each) in distilled water, and then stained with uranyl acetate and lead citrate. The sections were visualized and images were captured using a Zeiss 900 transmission electron microscope (Zeiss GmbH, Oberkochen, Germany).

The cDNAs encoding human S100A6 were generated from human tissues using polymerase chain reaction (PCR) with the following forward and reverse primers containing BamHI and XhoI sites: Forward 5′-AAGGATCCTACCGCTCCAAGCCCAGC-3′ and reverse 5′-CTCGAGCGCCCTTGAGGGCTTCATTGTAGAT-3′. The PCR products were digested with BamHI and XhoI, and then directly subcloned into the BamHI and XhoI sites of pcDNA.3.1/myc-His C vectors (Invitrogen; Thermo Fisher Scientific, Inc.), which produce fusion proteins.

The AGS and BGC823 human gastric cancer cell lines were maintained in high-glucose Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc.) containing 10% FCS. At the time of transfection, the cells were 90% confluent. The cells in 6-well plates (8×10⁵) were transfected with a total of 4.0 µg of plasmid DNAs using Lipofectamine™ 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The medium was replaced 6 h later. The cells were incubated at 37°C in a CO₂ incubator for 48 h prior to assessment of transgene expression. G418 (300 ng/ml; Gibco; Thermo Fisher Scientific, Inc.) was used to screen and isolate the resistant colonies. The efficiency of stable transfection was confirmed using PCR and western blot analyses.

RT-qPCR was performed to confirm successful transfection of the cell lines with S100A6 and to determine the expression levels of interleukin (IL)-8, IL-2, IL-13, IL-6R, cyclin-dependent kinase (CDK) 4, CDK5, minichromosome maintenance complex component 7 (MCM7), B-cell lymphoma 2 (Bcl2) and MYC. Total RNA was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Total RNA (1 µg) was reverse transcribed using the GoScript™ Reverse Transcription System (Promega Corporation, Madison, WI, USA). qPCR was performed using the SYBR Green PCR Master Mix kit (Applied Biosystems; Thermo Fisher Scientific, Inc.) with the following cycling conditions: 95°C for 10 min, followed by 40 cycles at 95°C for 15 sec, 60°C for 10 sec and 72°C for 60 sec. Each sample was run thrice. The primer sequences were as follows: S100A6 forward, 5′-AAGCTGCAGGATGCTGAAAT-3′ and reverse, 5′-CCCTTGAGGGCTTCATTGTA-3′; IL-8 forward, 5′-ACTGAGAGTGATTGAGAGTGGAC-3′ and reverse, 5′-AACCCTCTGCACCCAGTTTTC-3′; IL-2 forward, 5′-AACTCCTGTCTTGCATTGCAC-3′ and reverse, 5′-GCTCCAGTTGTAGCTGTGTTT-3′; IL-13 forward, 5′-CCTCATGGCGCTTTTGTTGAC-3′ and reverse, 5′-TCTGGTTCTGGGTGATGTTGA-3′; IL-6R forward, 5′-CCCCTCAGCAATGTTGTTTGT-3′ and reverse, 5′-CTCCGGGACTGCTAACTGG-3′; CDK4 forward, 5′-ATGGCTACCTCTCGATATGAGC-3′ and reverse, 5′-CATTGGGGACTCTCACACTCT-3′; CDK5 forward, 5′-GGAAGGCACCTACGGAACTG-3′ and reverse, 5′-GGCACACCCTCATCATCGT-3′; MCM7 forward, 5′-CCTACCAGCCGATCCAGTCT-3′ and reverse, 5′-CCTCCTGAGCGGTTGGTTT-3′; Bcl2 forward, 5′-GGTGGGGTCATGTGTGTGG-3′ and reverse 5′-CGGTTCAGGTACTCAGTCATCC-3′; MYC forward, 5′-GGCTCCTGGCAAAAGGTCA-3′ and reverse, 5′-CTGCGTAGTTGTGCTGATGT-3′; and β-actin forward, 5′-AAATCTGGCACCACACCTTC-3′ and reverse, 5′-GGGGTGTTGAAGGTCTCAAA-3′. The expression of each gene was determined as the ratio of the mRNA expression of each target gene to that of β-actin, using the 2^−ΔΔCq method (21).

Total proteins were extracted using TRIzol reagent, according to the manufacturer's protocol. Protein concentration was determined using a Bicinchoninic Acid Protein Assay kit (Thermo Fisher Scientific, Inc.). Samples were denatured in SDS sample loading buffer by heating at 100°C for 3 min and resolved by 12% SDS-PAGE, followed by transfer to polyvinylidene difluoride membranes. Blots were blocked with 5% nonfat milk in PBS for 30 min and then incubated with the mouse anti-S100A6 monoclonal antibody (1:2,000) and mouse anti-tubulin monoclonal antibody (1:5,000; cat. no. T5168; Sigma-Aldrich; Merck Millipore) for 2 h at room temperature. After washing, the membranes were incubated with the horseradish peroxidase-conjugated anti-mouse antibody (1:1,000; cat. no. sc-2005; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Protein-antibody complexes were visualized using the SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific, Inc.).

The BGC823, AGS, BGC823 cells transfected with pcDNA.3.1/myc-His C plasmid (BGC823/control), AGS cells transfected with pcDNA.3.1/myc-His C plasmid (AGS/control), BGC823 cells transfected with S100A6/pcDNA.3.1/myc-His C plasmid (BGC823/S100A6) and AGS cells transfected with S100A6/pcDNA.3.1/myc-His C plasmid (AGS/S100A6) were digested with trypsin-EDTA buffer (Gibco; Thermo Fisher Scientific, Inc.). A suspension of 2×10³ cells/90 µl medium was added to each well of 96-well plates and allowed to grow; each concentration had three duplicate wells. The plates were incubated for 1, 2, 3, 4, 5 and 6 days following which 10 µl of Cell Counting Kit-8 (CCK8) solution (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was added to each well and the plates were incubated for another 2 h at 37°C. Subsequently, the absorbance was read at 450 nm using a Model 680 microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) (22).

ChIP-Chip assay. KATO3 cells (10,000,000) were prepared, crosslinked with 1% formaldehyde and lysed with lysate buffer. Sonication was used to break genomic DNA into small DNA fragments. Anti-S100A6 (Abnova) was added and used to immunoprecipitate DNA fragments corresponding to the promoter regions. IgG was used as a control. The DNA fragments bound by proteins were then isolated, and DNA was hybridized to the GLAS-H20K-Chip (Aviva Systems Biology, Corp., San Diego, CA, USA), with the transcription starting point approximately in the −800 to +200 regions. The microarray was then scanned, and two images corresponding to anti-S100A6 and IgG (control) were captured. Subsequently, the ratio between anti-S100A6 and IgG was calculated. The factors for which the ratio was >2 were selected. Gene Ontology (GO) analysis was then performed to clarify the gene function (http://geneontology.org).

Spearman's rank correlation was performed to analyze the correlation between the mRNA expression of S100A6 and Ki67. Two-way analysis of variance was performed to analyze the results of the CCK8 assay. P<0.05 was considered to indicate a statistically significant difference and all tests were two-tailed. The statistical analysis was performed using SPSS V16.0 software (SPSS, Inc., Chicago, IL, USA).

A comparison between primary gastric carcinoma tissues and matched non-cancerous mucosal tissues revealed significant differences in the protein levels of S100A6 in these tissues. In the primary gastric carcinoma tissues, the intensity of S100A6 staining was markedly higher in the cytoplasm and nucleus of primary gastric carcinoma tissues, compared with the matched non-cancerous mucosa, which had either no or weak cytoplasmic staining. The S100A6 nuclear staining was significantly higher in the invading fronts with structural atypia, compared with the central portions with glandular structures (Fig. 1A and B). The ultrastructural images captured using an immunoelectron microscope showed that S100A6 was deposited within the cytoplasm and nucleus, whereas it was distributed in the nucleoplasmic and nucleolar structures in gastric cancer cells (Fig. 2A).

To investigate whether the expression of S100A6 was associated with in vivo cell proliferation, 108 samples underwent Ki-67 and S100A6 immunohistochemical staining, respectively. Statistical analysis revealed a positive correlation between high expression of S100A6 and a high Ki-67 labeling index (P=0.043; Fig. 2B and C). In total, 10 representative slides were selected for immunofluorescence staining and consecutive laser confocal scanning, which showed that the Ki-67 antigen was found localized to the nuclear region (Fig. 2D).

The correlation between S100A6 and cell proliferation was also investigated in vitro. Successful transfection of AGS and BGC823 cells with S100A6 was confirmed by western blotting (Fig. 3A and B). The present study examined whether S100A6 transfection promoted cell proliferation. The AGS, AGS/control, AGS/S100A6, BGC823, BGC823/control and BGC823/S100A6 cells were incubated for 1, 2, 3, 4, 5 and 6 days, and the growth rates were calculated using a CCK8 assay. The results suggested that the cell proliferation rate increased gradually and reached its highest level following 6 days of incubation. The proliferative abilities of the AGS/S100A6 and BGC823/S100A6 cells were markedly increased compared with those of the controls (Fig. 3C and D).

Based on the above results, it was concluded that S100A6 was located at nucleolar structures and that it regulated cell proliferation. Subsequently, a ChIP-Chip assay was performed, with the promoter Chip including approximately −800 to +200 regions around the transcription starting point. The results showed the involvement of 1,328 genes, which regulate metabolic process, cell proliferation, cell cycle and cell apoptosis. GO category analysis suggested that 57 factors were associated with cell proliferation (Table I). In addition, the common sites of the promoter regions of the 1,328 involved genes were analyzed using the human and mouse promoter database (−100 to +1). The results showed that the promoter regions of the majority of genes had SP1 protein binding sites, for example the cartilage oligomeric matrix protein gene (Fig. 4A and B).

To confirm the effect of S100A6 on the expression of factors associated with the malignant cell phenotype, the present study compared the expression levels of IL-8, IL-2, IL-13, IL-6R, CDK4, CDK5, MCM7, Bcl2 and MYC prior to and following S100A6 transfection of the BGC823 and AGS cell lines. RT-qPCR analysis was performed, which revealed that, following S100A6 transfection in the AGS cell lines, the mRNA expression levels of IL-8, CDK5, CDK4, MCM7 and Bcl2 were increased (Fig. 5A-E, respectively).