The human AGS (no. 89090402) and HGC-27 (no. 94042256) GC cell lines were purchased from the European Collection of Animal Cell Cultures (ECACC; Salisbury, UK). The cells were cultured in Dulbecco's modified Eagle's medium (DMEM)-F12 medium (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), supplemented with 10% fetal calf serum (FCS; Sigma-Aldrich; Merck KGaA) and 1% antibiotics (penicillin and streptomycin; Sigma-Aldrich; Merck KGaA) at 37°C in a humidified 5% CO₂ atmosphere.

Two cohorts were used in the present study, namely a generic collection and a collection with chemosensitive information. The first cohort had a total of 321 GC tissues and 173 adjacent normal tissues were collected from patients with GC that underwent surgical resection, and some were treated with preoperative radiotherapy or chemotherapy at the Peking University Cancer Hospital (Beijing, China). Tissue and data were obtained between September 2003 and December 2007. The mean age of the patients in the cohort was 59.8 years with male/female ratio of 229:92 (Table I). The other cohort comprised of 87 patients with GC that had received chemotherapy with oxaliplatin/5-fluorouracil/leucovorin prior to surgery at the Peking University Cancer Hospital between December 2001 and November 2006. All of the tissues were stored immediately following surgery at −80°C until use. Clinicopathological and follow-up information was obtained from patient data. GC staging was determined according to the 2010 tumor-node-metastasis (TNM) classification recommended by the American Joint Committee on Cancer (AJCC 7th edition) (21). This study was approved by Peking University Cancer Hospital Research Ethics Committee (ethics no. 2006021) and written consent was obtained from all patients involved.

Total RNA was extracted from cells or homogenized tissues using TRI reagent (Sigma-Aldrich; Merck KGaA) following the manufacturer's instructions, and subsequently reverse transcribed into cDNA using Promega Reverse Transcription kit (Promega Corporation, Madison, WI, USA). The reverse transcription conditions were annealing at 25°C for 5 min, extension at 42°C for 60 min, and inactivation at 70°C for 15. PCR was then performed using GoTaq Green MasterMix (Promega Corporation) under the cycling conditions as follows: 95°C for 2 min, followed by 40 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 30 sec, and a final extension of 5 min at 72°C. The PCR products were visualized on 1–2% agarose gels stained with SYBR Safe (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The primer sequences used for PCR in this study are listed in Table II.

qPCR was run on the StepOne Plus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.), using Precision FAST 2X qPCR MasterMix (Primerdesign Ltd., Chandler's Ford, UK) in a final volume of 10 µl. cDNA samples from GC cells and two cohorts of GC tissues were examined for psoriasin transcript expression with standard and negative controls. Reaction conditions were as follows: 94°C for 10 min, followed by 100 cycles of 94°C for 10 sec, 55°C for 30 sec and 72°C for 10 sec. GAPDH or actin was used as an internal control. The primer sequences used for qPCR are listed in Table II. Primers for qPCR were designed to the specific target using the Beacon Designer software (version 2; Premier Biosoft International, Palo Alto, CA, USA). An additional sequence, known as the Z sequence (5′-ACTGAACCTGACCGTACA-3′), which acts as a template for the amplifluor uniprimer probe, was added with the reverse primers. Expression of the target sequence was qualified in conjunction with a range of known dilution series of podoplanin gene transcript, which are used to generate a standard curve to enable the calculation of transcript copy.

Full-length human psoriasin coding sequence was amplified from human prostate tissues with appropriate primers (Table I). The synthesized product was cloned into the pEF6/V5-His-TOPO plasmid vector (K961020; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The overexpression constructs and control plasmids (25 µg pre transfection) were then transfected into 2.5×10⁶ AGS and HGC-27 cells using a Gene Pulser Xcell electroporation system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). After 2 weeks of selection using 6 µg/ml blasticidin (Melford Laboratories, Ltd., Ipswich, UK), stable transfectants were obtained and verified for psoriasin overexpression. Then cells were cultured in DMEM with a lower concentration of blasticidin (0.6 µg/ml) to maintain plasmid expression.

AGS and HGC-27 were cultured in serum-free medium for 4–6 h, and then treated with 1 µM ERK inhibitor (FR 180204; Tocris Bioscience, Bristol, UK) or 16.6 µM cisplatin (Tocris Bioscience) for 24 h, followed by PCR, western blotting or flow cytometry analysis.

AGS and HGC-27 were isolated by pre-chilled lysis buffer (50 mM Tris-Cl, pH 7.5, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.05% SDS, 1 mM EDTA, 150 mM NaCl) containing protease inhibitor cocktail (aprotinin and leupeptin) and phosphorylation inhibitor (sodium vanadate) and qualified with a DCTM protein assay kit (Bio-Rad Laboratories, Inc.). Protein extract of each sample (30 µg) was separated by 10% or 12% SDS-PAGE and transferred onto a 0.45 µm polyvinylidenedifluoride membrane (EMD Millipore, Billerica, MA, USA). The membranes were blocked with 5% non-fat dried milk in PBST solution (0.05% Tween-20 in PBS) for 1 h at room temperature. The blocked membranes were then incubated in a diluted primary antibodies (1:500) overnight as 4°C, washed in PBST and incubated with a diluted specific horseradish peroxidase (HRP)-conjugated secondary antibody (1:2,000) for 1 h at room temperature. The following primary antibodies were used in the study: Anti-S100A7/psoriasin (clone 47C1068; catalog no.; Novus Biologicals, Ltd., Cambridge, UK); phosphor (p)-ERK (Tyr 204; cat. no. sc-7976), ERK1/2 (clone C-9; cat. no. sc-514302), p21 (cat. no. sc-6246), p53 (clone DO-1; cat. no. sc-126), vimentin (cat. no. sc-66002), E-cadherin (cat. no. sc-1500), N-cadherin (cat. no. sc-7939), zinc finger E-box binding homeobox 1 (Zeb-1; cat. no. sc-81428) and GAPDH (cat. no. sc-32233; all Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The HRP secondary antibodies (cat. no. A5278, anti-mouse IgG; cat. no. A0545, anti-rabbit IgG) were obtained from Sigma-Aldrich (Merck, KGaA). Detection was performed using an EZ-ECL chemiluminescence kit (Biological Industries USA, Inc., Cromwell, CT, USA). Immunoreactive bands were visualized and quantified by densitometry using the Syngene G:BOX chemiluminescence imaging system and Gene Tools 4.03 (Syngene Europe, Cambridge, UK).

Briefly, AGS and HGC-27 cells were seeded into 96-well plate (2×10³ cells/well) in 6 replicates followed by incubation for 0, 1, 2, 3, 4 and 5 days. Following fixation in 4% formalin for 30 min at room temperature, the cells were stained with 0.5% (w/v) crystal violet (Sigma-Aldrich) for 15 min at room temperature. Following washing, the cells were then extracted with 10% (v/v) acetic acid and cell density was determined by measuring the absorbance at 540 nm using a spectrophotometer (Elx800; BioTek Instruments, Inc., Winooski, VT, USA).

To evaluate the cytotoxicity of cisplatin, AGS and HGC-27 cells were plated into 96-well plate (cells/well) in 6 replicates. After 24-h incubation, the cells were exposed to serial concentrations of cisplatin (0, 3.125, 6.25, 12.5, 25 and 50 µM, respectively) for 72 h. Cells viability was assessed by staining with 0.5% (w/v) crystal violet for 15 min at room temperature.

AGS and HGC-27 were plated into 6-well plate at the density of 400 cells/well and incubated at 37°C and at an atmosphere of 5% CO₂ for 2 weeks. The cell colonies were fixed in 4% formalin for 30 min and stained with 0.5% (w/v) crystal violet for 15 min. After images were taken with a camera, the colonies were dissolved in 10% (v/v) acetic acid, and the absorbance was measured at 540 nm using a spectrophotometer (Elx800; BioTek Instruments, Inc.).

A total of 2×10⁴ cells were plated in each well of a black 96-well plate previously coated with Matrigel (10 µg/well; Corning Incorporated, Corning, NY, USA) in 6 replicates. After incubation for 1 h, the cells were washed once with PBS and then incubated in cell dissociation solution (100 µl/well; Sigma-Aldrich; Merck KGaA) for 30 min containing 0.002 µg/µl calcein AM (eBioscience; Thermo Fisher Scientific, Inc.). Fluorescence of the cell suspension was read using a GloMax^®-Multi Detection System (Promega Corporation) at 495 nm for excitation and 519 nm for emission.

Cells (5×10⁴) were loaded into an 8 µm-pore ThinCertTM 24-well plate insert (Greiner Bio-One International GmbH, Kremsmünster, Austria) coated with 100 µg Matrigel (each sample in triplicate). The upper chamber was provided with serum-free medium, and the lower chamber was filled with DMEM with 10% FCS. Following incubation for 24 h, the underside of the insert was incubated in 350 µl dissociation solution containing 1 µM calcein AM for 1 h. The cell suspension was then aliquoted into a black 96-well plate (100 µl/well) and read using the GloMax^®-Multi Detection System at 495 nm for excitation and 519 nm for emission.

Electric cell-substrate impedance sensing (ECIS) instruments (Applied Biophysics, Inc., Troy, NY, USA) were used to monitor cell migration by detecting impedance variation based on the current flow change on the gold electrodes of a 96-well array (Applied Biophysics Inc.) (22). There are two parameters, resistance and capacitance, derived from current and voltage measurements. Among them, resistance, consisting of para- and trans-cellular current flow, is better to represent the quality and function of cell barrier (23). Briefly, 6×10⁴ cells diluted in 200 µl DMEM with or without ERK inhibitor (1 µM) were seeded into each ECIS 96-plate well, and the cells were wounded by applying electric current (3,000 µA, 60 kHz) when confluent monolayers were formed. The array was placed into a CO₂ incubator, which was connected to the ECIS Model 9600 Controller, and the impedance data, which indicate cellular migration, were collected continuously for 6 h. In this study, the normalized resistance measured at 4,000 Hz was chosen to evaluate cell function due to its high sensitivity to the current flow variation caused by the change of cell behavior as described in the literature (24).

For protein expression analysis, AGS and HGC-27 were fixed with IC Fixation Buffer (eBioscience; Thermo Fisher Scientific, Inc.) at room temperature for 30 min, and then permeabilized in ice-cold 100% methanol at 4°C for 30 min. Following blocking with PBS containing 0.1% Tween-20 (v/v) and 1% bovine serum albumin (w/v) (Sigma-Aldrich; Merck KGaA), the cells were incubated with a cleaved poly(ADP-ribose) polymerase 1 (PARP1) antibody (1:40; cat. no. 44-698G; eBioscience; Thermo Fisher Scientific, Inc.) overnight at 4°C. Mouse IgG2b K isotype antibody (eBioscience; Thermo Fisher Scientific, Inc.) was used as a control. The samples were then washed by PBS twice and resuspended in PBS containing 2 mM EDTA and 5% FCS for flow cytometry.

For cell cycle analysis, the cells were fixed with 70% ethanol for 15 min on ice, and stained with FxCycle™ PI/RNase staining solution (Thermo Fisher Scientific, Inc.) for 30 min. For apoptosis analysis, the cells were stained using an Annexin V Apoptosis Detection kit (eBioscience; Thermo Fisher Scientific, Inc.) following the manufacturer's instructions. Flow cytometry was performed using a Canto II flow cytometer (BD Bioscience, San Jose, CA, USA), and the data were analyzed using FCS Express version 4 Research Edition (De Novo Software, Glendale, CA, USA).

The results are presented as data from three independent experiments and expressed as the mean ± standard deviation. The cumulative survival curves were generated using Kaplan-Meier plots and analyzed using log-rank test. Kruskal-Wallis test was used for comparisons of three or more groups followed by Tukey Honest Significant Difference as a post hoc test, and Mann-Whitney U test was used for comparisons between two groups. Statistical analysis was performed using SPSS v18 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 6.0 (GraphPad Software, La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

To determine the clinical relevance of psoriasin in GC progression, the transcript levels of psoriasin in GC (n=321) and the adjacent normal specimens (n=173) were examined by RT-qPCR. As presented in in Table II, psoriasin was higher in GC tissues than in adjacent normal tissues (P<0.001), and a higher level of psoriasin was observed in advanced stages of GC (TNM III+IV) compared with early stages (TNM I+II; P<0.001; Table II). Notably, higher expression of psoriasin was detected in tumors with deeper infiltration (T3–4 vs. T1-2; P=0.011), lymph node involvement (P<0.001) and distance metastasis (P<0.001, Table II). Furthermore, psoriasin exhibited higher expression in patients with GC that had died prior to the end of follow-up than in those who remained alive at the end of the follow-up period (P<0.001; Table II). Kaplan-Meier analysis and log-rank test revealed that patients with GC with higher expression of psoriasin had worse overall survival (P<0.001; Fig. 1A) and disease-free survival (P<0.001; Fig. 1B), compared with those with lower expression.

AGS and HGC27 are two well-characterized gastric cancer cell lines (25,26). As demonstrated in Fig. 1, the gene expression of psoriasin is low in these two GC cell lines, as indicated by RT-qPCR (Fig. 1C and D) and conventional PCR (Fig. 1E). Western blotting also demonstrated that the protein levels of psoriasin in AGS and HGC27 were low (Fig. 1F). Attempts were made to subculture other GC cell lines, such as NUGC-1, MKN-45, MKN7 and MKN74. Unfortunately, they were either floating (NUGC-1, MKN-45) or was difficult expand during the establishment of the stable cells lines (MKN7 and MKN74). Therefore, AGS and HGC-27 were to establish stable psoriasin-overexpressed cell lines (AGS PSOexp and HGC-27 PSOexp) to determine the role of psoriasin in the tumorigenesis and progression of GC. The mRNA and protein levels of psoriasin was markedly increased in AGS PSOexp and HGC-27 PSOexp cells compared with the cells that were transfected with the vector alone (AGS pEF and HGC-27 pEF) as verified by RT-qPCR (Fig. 1C and D), PCR (Fig. 1E) and western blotting (Fig. 1F).

There was a significant increase in cell proliferation in psoriasin-overexpressed cells compared with the controls, which was assessed by crystal violet assay (P<0.001; Fig. 2A and B). Similarly, cell colony formation was also enhanced in the psoriasin-overexpressed cells (Fig. 2C and D). However, there was no effect on the cell cycle profile following psoriasin overexpression (data not shown).

As psoriasin expression was positively associated with lymph node involvement and distant metastasis (Table II), the migratory and invasive abilities of GC cells were evaluated in vitro. The Matrigel invasion assay and ECIS assay demonstrated that the overexpression of psoriasin dramatically enhanced migration (Fig. 2E and F) and invasion (Fig. 2G) in GC cell lines compared with the vector controls. Furthermore, upregulation of psoriasin significantly decreased cell-matrix adhesion in HGC-27 cells, but not in AGS cells (Fig. 2H). Together, these data suggest that psoriasin may function as an inducer of cell migration and invasion in GC.

Several EMT-associated molecules have been implicated in the development and progression of GC by altering the malignant properties of cancer cells (14,27). Therefore, the expression of epithelial markers, mesenchymal markers and EMT transcription factors were evaluated following overexpression of psoriasin by PCR and western blotting. The results demonstrated that the levels of N-cadherin and Snail expression were increased, while E-cadherin expression was decreased in AGS PSOexp and HGC-27 PSOexp cells compared with the empty vector controls. In addition, the upregulation of psoriasin promoted Slug and Zeb-1 expression levels in HGC-27 cells compared with the control cells, whereas the vimentin level was unchanged in both cell lines (Fig. 3A and B). Collectively, the activation of psoriasin may alter the patterns of EMT in GC cells, promoting cell invasion and migration.

The ERK signaling cascade is one of the key signaling pathways that regulates cell proliferation, invasion and migration by controlling multiple nuclear and cytoplasmic targets in cancer (28). In the present study, the effect of psoriasin on the ERK signaling pathway was investigated in GC. The upregulation of psoriasin significantly promoted the level of phosphorylated ERK (Tyr 204) protein, without altering the total ERK1/2 mRNA and protein levels in AGS and HGC-27 cells, compared with the control cells (Fig. 3C and D).

By contrast, there was no difference in the levels of cyclin D1 and cyclin D2 mRNA between the control cells and the cells that overexpressed psoriasin (Fig. 3C). Therefore, the effect of psoriasin on the viability of GC cells was unlikely due to the regulation of these regulatory proteins of cyclin-dependent kinases.

Treatment with an ATP-competitive ERK inhibitor suppressed the migratory ability of AGS cells (PSOexp cells and control cells), particularly in AGS PSOexp cells (Fig. 3E and F). Together, these results suggest that psoriasin may enhance cell migration at least partially via activation of ERK. Further study is required to determine whether ERK is a direct substrate of psoriasin.

To determine the effect of psoriasin on cisplatin resistance in GC, the gene expression of psoriasin was analyzed in specimens from patients with GC that received pre-operative chemotherapy (n=87). The results indicated that levels of psoriasin expression were higher in patients that exhibited no response to neoadjuvant chemotherapy (P=0.023; Table III). Flow cytometry analysis of cleaved PARP was performed to evaluate the levels of apoptosis in the stable cell lines that were treated with cisplatin. As presented in Fig. 4A and B, in the presence of cisplatin, the less AGS PSOexp cells expressed cleaved PARP protein (8.95%) than the AGS pEF cells (16.87%), suggesting an increase in resistance to cisplatin. However, such increase was not observed in HGC-27 cells, as the cleaved PARP expression was not markedly altered between the HGC-27 PSOexp cell line (63.03%) and the HGC-27 pEF cell line (59.38%; Fig. 4A and B). The cell cycle analysis indicated that in response to the cisplatin treatment, a higher portion of AGS cells was arrested in the G2/M phase compared with PBS treatment in AGS pEF cells (58.68% vs. 22.51%) and AGS PSOexp cells (61.04% vs. 23.55%). However, there was no difference in cell cycle profile between the AGS PSOexp and AGS pEF cell lines in the presence of cisplatin. By contrast, the S-phase population of the HGC-27 PSOexp cells was increased compared with the pEF control cells in the presence of cisplatin (44.68% in HGC-27 pEF cells and 60.84% in HGC-27 PSOexp cells; Fig. 4C–F). Furthermore, psoriasin overexpression in AGS cells inhibited the cisplatin-induced increase in the levels of p53 and p21 proteins compared with the cells transfected with the backbone pEF plasmid (Fig. 4G and H). However, the cell viability assay showed that although psoriasin overexpression improved survival rates in the presence of particular doses of cisplatin (e.g. 6.25 µM) in transfected AGS and HGC-27 cells compared with the pEF controls (P<0.05), the overall dose-dependent cytotoxic effect of cisplatin on the cell lines was similar to the control cell lines (data not shown). Therefore, the upregulation of psoriasin in GC cells may desensitize cells to cisplatin by promoting the proportion of cells in the S-phase or inhibiting apoptosis, instead of a direct effect via cellular cytotoxicity.