The cervical cancer cell line, HeLa and the OSCC-derived cell line, Sa3 were used in the present study. CDDP-resistant cells established from these cell lines, HeLa-R and Sa3-R, were used (20). Cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum and 50 U/ml of penicillin and streptomycin.

The present study was approved by the Ethics Committee of the Graduate School of Medicine, Chiba University (approval number, 236) and performed according to the tenets of the Declaration of Helsinki. All the animal experiments were performed in accordance with the ethical standards of Canadian Council on Animal Care (CCAC) and institutional guidelines.

Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. The quality of the total RNA was determined using the Agilent Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). cDNA was synthesized from total RNA using Ready-to-Go You-Prime first-strand beads (GE Healthcare, Buckinghamshire, UK) and oligo (dT) primer (Sigma Genosys, Ishikari, Japan) according to the manufacturer's protocol.

Real-time qRT-PCR was performed using LightCycler^® 480 Probes Master kit (Roche Diagnostics GmbH, Mannheim, Germany) according to manufacturer's instructions. PCR reactions were performed in the Light Cycler (Roche Diagnostics GmbH) apparatus. Transcript amount was estimated using standard curves and normalized against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcripts determined in corresponding samples. Nucleotide sequences of the specific primers for AKR1C1, AKR1C2, AKR1C3, AKR1C4, and GAPDH are shown in Table I.

SMARTpool siRNA targeting AKR1Cs (siAKR1C1, siAKR1C2, siAKR1C3, and siAKR1C4) (Dharmacon, Lafayette, CO, USA) were used in gene silencing. Vehicle control and siCONTROL nontargeting siRNA pool (D-001206-13-20; siNT) were used as negative controls. cyclophilin B (siCONTROL cyclophilin B; siCyclo) was used as a positive silencing control to ascertain the transfection efficiency in each experiment. Cells were transfected with siRNAs using DharmaFECT 1 siRNA transfection reagent (Dharmacon).

Cells (2000 cells/well) were seeded in triplicate in a 96-well plate and cultured for 72 h. MTS reagent (Promega) was then added and incubated for 4 h, and relative cell viability was determined by recording the absorbance at 490 nm. Assays were performed with or without anticancer agent [CDDP, 5-fluorouracil (5-FU)] or non-steroidal anti-inflammatory drugs (NSAIDs). The experiment was repeated 3 times.

Expression of AKR1C family proteins was detected using western blot analysis. Cells were pelleted and resuspended to a concentration of 10⁸ cells/ml in lysis buffer (150 mM NaCl, 20 mM Tris-HCl, pH 7.4, containing 1% Triton X-100, 10% glycerol, 5 mM EDTA, 10 mM NaF, 1 mM sodium orthovanadate, 100 U/ml aprotinin, 10 mM iodoacetamide, and 25 µg/ml p-nitrophenyl-p'-guanidinobenzoate) with a cocktail of proteinase inhibitors (Roche Diagnostics GmbH). After centrifugation at 15,000 × g for 20 min, supernatants (cell lysates) were collected and subjected to SDS-PAGE (4–12%) and transferred to nitrocellulose membranes (Invitrogen). After blocking with Blocking One (Nacalai Tesque, Kyoto, Japan), the membrane was incubated with each antibody against AKR1C1, AKR1C2, AKR1C3, AKR1C4 (Sigma-Aldrich, St. Louis, MO, USA), respectively. Then, the membrane was incubated with the horseradish peroxidase-conjugated anti-mouse or anti-rabbit IgG antibody (Promega, Madison, WI, USA). The signals were detected using SuperSignal West Pico Chemiluminescent substrate (Thermo) and were visualized using ATTO Light-Capture II (Atto, Tokyo, Japan). Protein expression profiles of AKR1Cs were normalized with the internal control, β-actin.

HeLa and HeLa-R cells were used in the xenograft experiment. Cells (1×10⁷) were dissolved in 200 µl of phosphate buffered saline (PBS) and injected subcutaneously into the backs of 6-week-old female athymic nude mice; BALB/cAnNcrj-nu/nu (Charles River Japan Inc., Kanagawa, Japan). When the transplanted tumor volume reached 100 mm³, the mice were divided into 4 groups; control (n=5), anticancer agent (CDDP or 5-FU) alone (n=5), mefenamic acid alone (n=5), and anticancer agent plus mefenamic acid (n=5). CDDP (2 mg/kg) was administered intraperitoneally once weekly for 4 weeks. 5-FU (7.5 mg/kg) was administered intraperitoneally five times a week for 3 weeks. The freeze-dried diet containing 0.0125% of mefenamic acid was prepared and fed to the mice. In addition to tumor volume, body weight of mice was monitored throughout the experiment period. Six weeks after medication was initiated, tumor tissues were resected and analyzed.

The average values and standard deviation were analysed, and P-value was calculated using two-tailed, Student's t-test in MTS assay and in xenograft experiment. For all tests, α-level was 5% and the criterion for statistical significance was P<0.05.

The cervical cancer cell line, HeLa (Japanese Collection of Research Biosources Cell Bank, Osaka, Japan) and the OSCC-derived cell line, Sa3 (RIKEN BioResource Center, Ibaraki, Japan) were used in the present study. Moreover, CDDP-resistant cells established from these cell lines, HeLa-R and Sa3-R, were used (20). The sensitivity of these cells to various concentrations of CDDP was determined using MTS assay. Sa3-R and HeLa-R cells showed significantly higher viable cell rates than the parent clones with the same concentration of CDDP. The 50% inhibitory concentration (IC₅₀) values (µM) in the Sa3, Sa3-R, HeLa, and HeLa-R cells were 1.4, 18.4, 1.0, and 12.9, respectively. CDDP-resistant cell lines also showed drug resistance to 5-FU, suggesting that the cell lines potentially had cross-resistance to anticancer reagents. The IC₅₀ values (µg/ml) in the Sa3, Sa3-R, HeLa, and HeLa-R cells were 1.1, 38.3, 51.5, and 212.6, respectively. No morphologic difference was observed between CDDP-resistant cell lines and the parent cell lines. The CDDP-resistant cells as well as the parent cell lines had proliferative ability.

Real-time quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) analysis revealed that aldo-keto reductase (AKR) 1C1, AKR1C2, AKR1C3, and AKR1C4 were significantly upregulated in the CDDP-resistant cell lines (Fig. 1). Data are expressed as the average ± standard deviation (SD) of two independent experiments with samples in triplicate. The results indicated that AKR1C family members were associated with resistance to CDDP; thus, the AKR1C family was subjected to further investigation.

To determine whether AKR1C gene silencing contributes to CDDP sensitivity, cells were transfected with siRNAs. The AKR1C protein expression was examined by western blot analysis in Sa3-R and HeLa-R cells 120 h after transfection with siRNAs. AKR1C protein levels in non-targeting siRNA (siNT)-transfected cells were comparable to those of AKR1Cs in non-treated cells. AKR1C protein levels reduced significantly compared to the siNT-transfected cells (Fig. 2). These transfected cells were subjected to functional analysis to determine the effect of AKR1C gene silencing on CDDP sensitivity. The sensitivities of these cells to CDDP were determined by MTS assay. Increased sensitivity to CDDP and 5-FU was observed in Sa3-R cells transfected with siAKR1C1, siAKR1C2, siAKR1C3, and siAKR1C4, compared to that observed in the corresponding cells treated with siNT (Figs. 3 and 4).

Since mefenamic acid is an effective inhibitor of AKR1Cs (21), mefenamic acid was added, and the alteration of CDDP sensitivity in tumor cell lines was determined by MTS assay. CDDP-resistant cells treated with mefenamic acid (Sa3-R and HeLa-R) demonstrated restored sensitivity to CDDP and 5-FU in a dose-dependent manner (Fig. 5). As a result, mefenamic acid was used for further experiments in vivo.

Mouse tumor xenografts were established to investigate whether an AKR1C inhibitor enhances sensitivity to CDDP. HeLa and HeLa-R cells were subcutaneously implanted into female athymic nude mice, and the tumor volume was measured regularly until day 42. When CDDP was administered, tumor growth was suppressed in HeLa cell transplanted mice, however, tumor volume gradually increased in HeLa-R cell transplanted mice probably due to the CDDP-resistant characteristic (Fig. 6). The increased tumor volume significantly reduced in HeLa-R cell xenografts when mefenamic acid was added to CDDP (Fig. 7). Similarly, in the 5-FU administered mice, tumor volume in the mice treated with mefenamic acid was significantly smaller than without it (Fig. 8). Body weights of the mice were also measured to examine the systemic effect induced by these chemical reagents. The data revealed that the systemic effects observed after combined administration of mefenamic acid and anticancer agents were not different from that observed with individual administration (Figs. 9 and 10).