The human gastric carcinoma cell lines, AGS, TMC-1, SNU-1, TMK-1, SCM-1, MKN-45, and KATO III, were cultured in RPMI-1640 (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS), sodium bicarbonate (2%, w/v), L-glutamine (0.29 mg/ml), penicillin (100 U/ml), and streptomycin (100 μg/ml) (Invitrogen) at 37°C in a humidified 5% CO₂ incubator.

Specific antibodies against S100A4, ribophorin II (RPN2) and β-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-PARP and anti-caspase-3 were obtained from Cell Signaling Technology (Beverly, MA, USA). Anti-Myc was purchased from Millipore (Millipore Corporation, Bedford, MA, USA). Cisplatin was purchased from Sigma (St. Louis, MO, USA).

Cells (5×10³) were seeded in 96-well culture plates, incubated overnight at 37°C in medium containing 10% FBS, and then treated with the indicated concentrations of anticancer drugs for 48 h. Cell viability was determined using an MTS colorimetric assay (CellTiter 96^® cell proliferation assay kit; Promega, Madison, WI, USA) as described by the manufacturer. All experiments were performed at least in triplicate, on three separate occasions. A dose-response curve was plotted, and the drug concentration that decreased color development by 50% (i.e., the IC₅₀ value) was calculated for each drug. The data are presented as means ± SDs.

For small-interfering RNA (siRNA) knockdown of S100A4, ON-TARGET plus SMART pool siRNAs against S100A4 were purchased from Dharmacon Research (Lafayette, CO, USA). Non-targeting siRNA duplexes were used as negative controls (Dharmacon Research). Cells were transfected with siRNA using Lipofectamine RNAiMAX (Invitrogen) and incubated in glucose-free Opti-MEM (Invitrogen) according to the manufacturer’s recommendations.

Cell extracts were prepared in lysis buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 50 mM NaF, 1 mM Na₃VO₄, 10% glycerol, and a protease inhibitor cocktail). Equal amounts of proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were blocked, washed, probed with the indicated primary antibodies, washed again, and incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h. Finally, the blots were washed, and then developed using enhanced chemiluminescence (ECL) reagents (Millipore) according to the manufacturer’s protocol.

RNA was isolated from cultured cells using the TRIzol reagent (Invitrogen) according to the manufacturer’s instructions, and cDNA was synthesized from 2 μg of total RNA by reverse transcription using the ImProm-II Reverse Transcriptase kit (Promega) with oligo(dT)_12–18 primers. The resulting cDNA was used for subsequent PCR using specific PCR primers for S100A4 (forward, 5′-ATGGCGTGCCCTCTGGAG-3′ and reverse, 5′-TTTCTTCCTGGGCTGC-3′).

The full-length human S100A4 cDNA was inserted into pcDNA3.1 (Invitrogen) to express the S100A4-Myc recombinant protein. Asp-10 was substituted with a valine residue using a QuikChange II Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA) using specific DNA oligonucleotides (D10V forward, 5′-GAGAAGGCCCTGGTTGTGATGGTGTCC-3′ and D10V reverse, 5′-GGACACCATCACAACCAGGGCCTTCTC-3′).

For continuous monitoring of cell migration, cells (1×10⁴ cells/well) suspended in serum-free medium were seeded into the upper compartment of CIM-plates 16 (Roche, Mannheim, Germany). The lower compartment was then filled with medium containing 10% FBS and incubated in the RTCA station (xCELLigence System, Roche). Cell migration was monitored for 24 h with impedance measured every 15 min. Cell impedance was represented as cell index (CI) = (Z_i-Z₀) [Ohm]/15[Ohm], where Z₀ was the background resistance and Z_i was the resistance at a given time-point. A normalized cell index was determined as the cell index at a given time-point (CI_ti) divided by the cell index at the normalization time-point (CI_{nml_time}).

The in vitro cell migration assay was performed in Transwell chambers (Millipore), using 8.0-μm pore-size filters according to the manufacturer’s recommendations. Briefly, 1×10⁵ cells were suspended in serum-free DMEM and seeded to the upper compartment of a Transwell insert, while the lower compartment was loaded with a 24-well dish containing medium supplemented with 10% FBS. After incubation at 37°C in 5% CO₂ for 18 h, a cotton swab was used to remove the non-migrated cells from the upper surface of the membrane. The cells that had migrated through the membrane and adhered to the lower surface were fixed with methanol and then stained with crystal violet (1% crystal violet in 75% ethanol). The cells were examined under a microscope and counted. Experiments were performed in triplicate, and the results were calculated by averaging the total number of cells from three membranes.

All experiments were performed in triplicate. The significances of between-group differences were determined using the Student’s t-test. A P-value <0.05 was considered to indicate a statistically significant difference.

First, we examined the expression levels of S100A4 in seven gastric cancer cell lines. S100A4 was found to be highly expressed in MKN-45, SNU-1 and TMK-1 cells at both the protein and mRNA levels, whereas AGS and SCM-1 cells exhibited much lower expression of S100A4 at the protein and mRNA levels (Fig. 1). We therefore used AGS, MKN-45, TMK-1, and SCM-1 cells in our subsequent studies. To examine the functional significance of S100A4 in the cell migration of these gastric cancer cell lines, we used the RTCA system, which is a label-free, real-time automated continuous-monitoring platform that assesses cell migration by measuring changes in the electrical impedance at the electrode/cell interface. As shown in Fig. 2, higher expression levels of endogenous S100A4 were significantly associated with the higher migration ability of MKN-45 cells compared to AGS cells. Moreover, the migration ability of AGS cells was markedly enhanced by ectopic expression of S100A-Myc, whereas siRNA-mediated knockdown of endogenous S100A4 expression attenuated the cell migration of MKN-45 cells (Fig. 2A–C). Consistent with the results from previous reports, our data confirmed that S100A4 plays a critical role in the cell migration of gastric cancer cell lines.

To investigate the role of S100A4 in anticancer drug resistance, AGS cells expressing exogenous S100A4-Myc were exposed to the half-maximal inhibitory concentrations (IC₅₀) of six conventional anticancer drugs for 48 h. Immunoblot analysis of apoptosis-related protein markers revealed that overexpression of S100A4 slightly decreased the cisplatin-induced cleavage of caspase-3 and poly(ADP-ribose) polymerase (PARP) in AGS cells (Fig. 3A). MTS-based cell viability analyses showed that overexpression of S100A4 slightly (but not significantly) increased cell viability relative to empty-vector controls in AGS cells treated with any of the tested drugs (Fig. 3B). Next, MKN-45 cells were subjected to siRNA-mediated knockdown of S100A4 (siS100A4) for 24 h and then challenged with anticancer drugs for 48 h. S100A4 knockdown slightly increased the cisplatin-induced cleavage of caspase-3 and PARP, and decreased BCL2 protein levels (Fig. 4A). However, our MTS-based analysis of cell viability indicated that knockdown of S100A4 did not have any discernible effect on cell survival (Fig. 4B). A similar pattern of drug responsiveness was obtained in TMK-1 cells, wherein significant knockdown of S100A4 (Fig. 5A) did not alter the resistance to various anticancer drugs (Fig. 5B).

Numerous proteins have been identified as potentially interacting with S100 proteins, including Annexin II. (18,31) We selected nonsynonymous single-nucleotide polymorphisms (SNPs) in S100A4 from dbSNP, and examined whether they alter the biological function of the protein. We identified an SNP of S100A4 (NM_002961.2: c.29A>T, rs1803245) that resulted in the substitution of an Asp residue with a Val residue (NP002952.1:p.Asp10Val), and further found that it localizes within a sequence that is conserved among the members of the S100A protein family (Fig. 6A). Computer modeling of the S100A4 protein structure showed that the D10 residue is localized within the binding surface for Annexin II (Fig. 6B). To examine the potential significance of this SNP in cancer cell migration and anticancer drug resistance, we next examined whether cancer cells expressing this S100A4 protein variant showed changes in their metastasis-related properties. AGS cells were transfected with wild-type S100A4^D10 (A allele) or S100A4^D10V (T allele), and cell motility was assessed by an in vitro cell migration assay. Compared to cells expressing S100A4^D10, those expressing S100A4^D10V showed significantly fewer instances of traversing the membrane (P<0.01) (Fig. 6C and D). This suggests that S100A4^D10V is negatively correlated with cancer cell motility compared to the wild-type.

The drug responsiveness of cells expressing S100A4^D10V was further investigated. AGS cells overexpressing S100A4^D10V were individually treated with the six tested anticancer drugs for 48 h, and cell viability was determined. As shown in Fig. 7, the D10V substitution was not found to correlate with drug responsiveness. Similar results were obtained in SCM-1 cells expressing S100A4^D10V (Fig. 8). Overexpression of both S100A4 versions did not have any discernible effect on cell growth in the absence of drugs (Figs. 7 and 8), with the exception of a slight enhancement of viability among AGS cells (Fig. 7). Taken together, our results suggest that S100A4 contributes to cancer cell migration but not the resistance to anticancer drugs.

Ribophorin II (RPN2) is a prognostic marker and has been shown to contribute to resistance against chemotherapeutic agents in human breast tumors and animal models of breast cancer (32–36). Therefore, we tested whether RPN2 depletion could have a synergistic effect in S100A4-knockdown gastric cancer cells. MKN-45 cells were subjected to simultaneous siRNA-mediated depletion of RPN2 and S100A4 for 24 h, and individually challenged with the six tested anticancer drugs for another 48 h. MTS-based assays of cell survival revealed that there was no change in cell viability among cultures subjected to RPN2 knockdown, S100A4 knockdown, or dual knockdown of RPN2 plus S100A4, regardless of drug treatment (Fig. 9).