Significance of thymidylate synthase expression for resistance to pemetrexed in pulmonary adenocarcinoma

Pemetrexed (PEM), a multi-targeted antifolate, has promising clinical activity in non-squamous non-small cell lung cancer. However, the majority of patients eventually acquire resistance to PEM. To evaluate the resistant mechanisms, the A549 lung adenocarcinoma cell line was exposed to stepwise increasing PEM concentrations of 1.6, 6.4 and 16 μM to establish three PEM-resistant lung cancer cell lines, A549/PEM-1.6, -6.4 and -16. Growth inhibition was determined by the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay. Expression of the genes encoding thymidylate synthase (TS), reduced folate carrier (RFC) and folypoly-γ-glutamate synthetase (FPGS) were analyzed by quantitative real-time reverse transcription polymerase chain reaction. The three A549 cell lines showed more resistance to PEM (3.7-, 17.3− and 38.0-fold, respectively) compared with that of the parental cell line, which also showed cross-resistance to cisplatin, but not to docetaxel, vinorelbine and 5-Fluorouracil (5-FU). TS gene expression was significantly increased in three PEM-resistant cells, relative to that of the parental cells, in a PEM dose-dependent manner. Knockdown of TS expression with siRNA enhanced the cytotoxicity of PEM in A549/PEM-16 cells. By contrast, the levels of RFC and FPGS gene expression in A549/PEM-1.6 and -6.4 cells were significantly decreased, whereas the levels of the two genes were restored in A549/PEM-16 cells. In summary, PEM-resistant A549 cells remained sensitive to docetaxel, vinorelbine and 5-FU. TS expression appeared to be associated with resistance to PEM, which may be a predictive marker for PEM sensitivity in lung adenocarcinoma.


Introduction
Lung cancer is the leading cause of cancer-related mortality worldwide, with ~226,160 new cases and ~160,340 mortalities predicted in 2012 in the United States (1). Non-small cell lung cancer (NSCLC) is a heterogeneous aggregate of histologies, including squamous cell carcinoma, adenocarcinoma and large cell carcinoma, and represents ~80-85% of all types of lung cancer (2). Despite the public awareness of NSCLC and increasing use of screening techniques, the majority of patients are likely to have advanced-stage non-operable disease at the time of diagnosis. Therefore, chemotherapy is often the first-line treatment for such patients.
Progress has been made in the treatment of advanced NSCLC during the past decade (3). The results of four previous multicenter randomized clinical trials evaluating the newer cytotoxic agents, alone or in combination with platinum-based chemotherapy, were shown to prolong survival, relieve symptoms in the majority of cases and improve patient quality of life (4)(5)(6)(7). It is clear from these studies that no single regimen demonstrated a significant superiority compared with any other combination. However, in the last three years, important advances have been achieved in the treatment of advanced NSCLC (8). Previous results arising from the availability of pemetrexed (PEM) show that histology represents an important variable in decision making (9).
PEM is a novel, multi-targeted antifolate and its primary mechanism of action is to inhibit at least three different enzymes in the folate pathway: thymidylate synthase (TS), dihydrofolate reductase and glycinamide ribonucleotide formyltransferase (10). These enzymes are involved in the synthesis of nucleotides and, therefore, inhibition ultimately hinders RNA and DNA synthesis. During the process, the primary vehicle for the uptake of PEM is reduced folate carrier (RFC), which is retained in cells as polyglutamates, a process catalyzed by folypoly-γ-glutamate synthetase (FPGS). Polyglutamation results in an increased intracellular drug concentration and cytotoxicity (11).
In chemotherapy-naive patients with advanced NSCLC, combination chemotherapy with PEM and cisplatin has an efficacy similar to that of gemcitabine and cisplatin, which has been the standard first-line treatment for patients with advanced NSCLC, with improved tolerability. The median overall survival time (MST) was 10.3 months in the two arms (12). However, a pre-planned analysis of this trial for the histological subtype of NSCLC reported that adenocarcinoma patients have a higher MST on cisplatin/PEM compared with cisplatin/gemcitabine (12.6, vs. 10.9 months, respectively; P=0.03) (9). PEM produced similar results and had an improved tolerance compared with that of docetaxel in advanced NSCLC patients following the failure of one prior chemotherapy regimen in a phase III trial (13), with an MST of 8.3 versus 7.9 months, respectively. No significant difference was identified in the outcome or toxicity between elderly and younger patients (14). Thus, the majority of patients acquired resistance to PEM between 2 and 5 months. Therefore, in the current study, PEM-resistant lung adenocarcinoma cell lines were established to further understand the resistance mechanisms.

Materials and methods
Cell lines and chemicals. A549 cells were purchased from the American Type Culture Collection (Manassas, VA, USA), which were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum, penicillin G (100 U/ml) and streptomycin (100 µg/ml) in a humidified chamber (37˚C, 5% CO 2 ). To observe the various mechanisms according to the degree of resistance, the A549 cell line was continuously exposed to stepwise increasing PEM concentrations of up to 1.6 µM for 5 months, 6.4 µM for 7 months and 16 µM for 10 months, which resulted in the following three PEM-resistant sublines: A549/PEM-1. Growth inhibition assay. Growth inhibition was evaluated using MTT assay, which measures the mitochondrial activity of viable cells. Cells were plated in flat bottom 96-well plates (Greiner Bio-One GmbH, Frickenhausen, Germany), with seeding densities of 2,000 cells per well for A549 and its sublines and allowed to attach for 24 h. Subsequently, cells were treated with RPMI-PBS containing serial dilutions of each chemotherapeutic agent for 96 h in a humidified chamber (37˚C, 5% CO 2 ). Following treatment, the medium was removed and cells were incubated for 4 h at 37˚C in 50 µl per well of MTT solution (final concentration, 0.42 mg ml-1). Formazan crystals that had formed were dissolved in 150 µl dimethyl sulfoxide per well and the absorbance was measured at 540 nm using a spectrophotometric microplate reader (iMARK; Bio-Rad, Hercules, CA, USA). Radiosensitivity was also determined using the MTT assay. Instead of adding chemotherapeutic drugs, cells in a 96-well microplate were irradiated at doses of 0 (control), 4 or 8 Gy (room temperature; linear accelerator; URA, Antwerp, Belgium). Following 96 h incubation at 37˚C in a humidified chamber with 5% CO 2 , the cells were assayed as abovementioned.
The drug concentration required to inhibit the growth of tumor cells by 50% (IC 50 ) was calculated by plotting the logarithm of the drug concentration versus the percentage of surviving cells. Each assay was performed in quadruplicate at least three times and the mean was calculated.
A growth rate of the cells was also determined by MTT assay and cells growing in exponential phase were plated in 48-well plates. The doubling time of each cell line was estimated from the duration of cell increment determined by measuring the mean absorbance of eight wells for seven successive days.
qPCR was performed using an ABI PRISM 7700 Sequence Detection system (Applied Biosystems, Inc., Foster City, CA, USA) . The protocol was as follows: 50˚C for 2 min and 95˚C for 10 min, followed by 50 cycles at 95˚C for 15 sec and 60˚C for 2 min. The mRNA levels were normalized using GAPDH expression. The reaction mix consisted of 12.5 µl SYBR Green master mix (Applied Biosystems, Inc.) and 2.5 µl of forward and reverse primers for target gene (Table I)   incubation, prior to measuring the absorbance at 490 nm using an ELISA plate reader. The siRNA oligonucleotides for TS (predesigned siRNA; ID 116928) and negative control siRNA (silence negative control 1 siRNA) were purchased from Ambion (Carlsbad, CA, USA).
Statistical analysis. Statistical significance was determined using one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference using two-sided analysis.     Cross-resistant patterns were also observed for these three cell lines (Table II). All three PEM-resistant sublines exhibited cross resistance to cisplatin, but not to docetaxel, vinorelbine and 5-FU, and also remained sensitive to MTX, a mother compound of PEM.
Effect of TS siRNA. Considering the importance of TS overexpression for acquired resistance to PEM, A549/PEM-16 cells were transfected with TS siRNA to investigate whether modification of TS gene expression may alter PEM cytotoxicity. At 24 h following transfection, total RNA was extracted and TS  gene expression was measured by real-time RT-PCR. Relative to A549/PEM-16 transfected with negative-control siRNA, the expression of the TS gene was significantly diminished by ~41% in cells treated with TS siRNA (P<0.05) and was not changed in non-transfected cells (Fig. 4A). At 72 h following transfection, cell viability was assessed using MTS assays, which showed that the cytotoxicity of PEM in A549/PEM-1.6 cells transfected with TS siRNA was greatly enhanced compared with cells transfected with negative-control siRNA (Fig. 4B). Therefore, decreased TS gene expression altered the sensitivity of PEM.

Discussion
Three PEM-resistant lung adenocarcinoma cell lines were established with three different PEM concentrations, which remained sensitive to 5-FU, docetaxel and vinorelbine, however, all cell lines showed resistance to cisplatin. TS has a central role in DNA biosynthesis and tumor biology and is the target of antifolate agents, such as 5-FU. The acute induction of TS has also been verified as one of several mechanisms of acquired resistance to 5-FU, since TS is stably bound to FdUMP and no longer has the ability to bind to its mRNA and suppress its own translation, which results in increased TS protein expression (15). Notably, incubation of TS with PEM also significantly impairs its ability to interact with TS mRNA (15). Therefore, acute induction of TS expression may be important for acquired resistance to PEM in the same manner as 5-FU. In the present study, the expression of TS mRNA was significantly increased in A549/PEM-1.6, -6.4 and -16 cells compared with A549 cells, which was consistent with previous studies (16)(17)(18)(19). In addition, following transfection with TS siRNA, the expression of the TS gene in A549/PEM-16 cells was diminished significantly and sensitivity to PEM was restored.
High TS expression in uterine cervical cancer cells has been reported to induce resistance to radiation, which has been explained by the suppression of p53 expression or promotion of DNA repair via TS increment (20). However, in the current study, no correlation between high TS expression and radiation was detected. By contrast, A549/PEM-1.6 and -6.4 cells showed increased sensitivity compared with parental cells to irradiation. Therefore, further investigation to clarify the correlation between TS levels and resistance to radiation is required. The results of the present study also indicated that PEM-resistant patients with locally advanced NSCLC in clinical settings may remain sensitive to irradiation, but receive thoracic radiotherapy. However, this is likely to be confirmed by future clinical trials.
RFC and FPGS activities may also be a determinant of PEM cytotoxicity (21,22). PEM utilizes RFC for entry into cells and then requires polyglutamation by FPGS to inhibit various target enzymes maximally. Therefore, decreased expression of RFC and/or FPGS may be associated with resistance to PEM, which has been determined in an L1210 murine leukemia cell line (23) and colon cell line (21). In the present study, RFC and FPGS gene expression in A549/PEM-1.6 and -6.4 cells was significantly decreased, while RFC and FPGS gene expression in A549/PEM-16 cells was restored to levels similar to those observed in parental A549 cells. Therefore, acquired resistance to PEM may result from the reduction of the intracellular concentration of PEM due to decreased levels of RFC gene expression and/or inhibition of polyglutamation due to decreased levels of FPGS gene expression in low (1.6 µM) or medium (6.4 µM) concentrations. However, in high (16 µM) concentrations, the determinant of acquired resistance to PEM is different, which may particularly depend on high TS gene expression. Although the determinants of acquired resistance to PEM may be altered by PEM concentration, TS overexpression may be one of the major determinants.
Three PEM-resistant lung adenocarcinoma cell lines were established, which remained sensitive to 5-FU, docetaxel and vinorelbine. It has been proposed that TS overexpression may be one of the major determinants of acquired resistance to PEM in lung adenocarcinoma, although, its interaction with other genes, such as RFC and FPGS, may also be important. In conclusion, we hypothesize that the level of TS gene expression may predict drug sensitivity to PEM. Therefore, examination of the correlation between TS gene expression and sensitivity to PEM in patients of lung adenocarcinoma is predicted.