In spite of recent advances in the development of both diagnostic and therapeutic tools, gastric cancer is still a major cause of death throughout the world. Moreover, because many patients are still diagnosed at only the late stages of this disease, recurrent tumors are often detected even after curative surgery. In such patients, chemotherapy is one of the main treatment strategies that is employed, as it has been shown to improve prognoses.1, 2, 3 However, a standard regimen of the treatment of metastatic gastric cancer has not been yet established in Japan.
5-Fluorouracil (5-FU) is widely used an anticancer agent and is currently considered to be a key drug in clinical chemotherapeutic treatments for gastrointestinal cancers, such as esophageal,4 gastric5 and colorectal.6 In addition, 5-FU has been also used to treat both breast7 and cervical8 cancers. 5-FU is catabolized to 2-fluoro-β-alanin by the first and rate-limiting enzyme of its metabolic pathway, dihydropyrimidine dehydrogenase (DPD). The functional effects of 5-FU are thought to occur through its active metabolite, 5-fluorodeoxyuridine diphosphate (FdUMP)9, 10 which, together with the coenzyme 5,10-methylenethtrahydrofolate (M-THF), forms a covalent ternary complex with the DNA de novo synthesizing enzyme thymidine synthetase (TS). This complex blocks the conversion of deoxyuridine monophosphate (dUMP) to thymidine monophosphate (dTMP) and thus inhibits DNA synthesis. Hence, the pharmacogenetic variability of 5-FU-related enzymes, such as DPD and TS, might be a major determinant of outcomes for gastrointestinal cancer patients treated with 5-FU.
S-1 is a novel oral fluorouracil antitumor drug that contains a combination of the three pharmacological agents tegafur (FT), which is a prodrug of 5-FU, 5-chloro-2,4-dihydroxypyrimidine (CDHP), which inhibits the activity of dihydropurimidine dehydrogenase (DPD) activity and potassium oxonate (Oxo), which reduces the gastrointestinal toxicity of 5-FU. Phase II studies have demonstrated that S-1 is active against gastric carcinomas.11, 12, 13, 14, 15, 16, 17 Moreover, S-1 has now been adopted in front-line chemotherapeutic regimens for the treatment of gastric cancer in Japan and its use in combination therapies with new agents has generated much recent interest. We have selected docetaxel (TXT) as the combination agent to test in our current study because of its overlapping antitumor spectra,18, 19, 20, 21 including against breast,22 esophageal,23 gastric,21 and head and neck cancers.24 In addition, TXT has a different mechanism of action than S-1 and has no cross resistance to other chemotherapeutic agents. Furthermore, we have recently demonstrated the tolerability and efficacy of TXT and S-1 combination chemotherapies.25, 26 However, little is known about the mechanisms underlying the synergistic effects of TXT upon either 5-FU or S-1. In the present study, we examined both the expression and activity profiles of enzymes that function in 5-FU metabolism, including TS, DPD and OPRT, so as to elucidate whether synergistic effects occur between TXT and S-1 against gastric cancer cells and to gain further insight into the mechanisms involved.
Five gastric carcinoma cell lines were routinely grown in RPMI 1640 (Nissui Co., Tokyo) supplemented with 5% fetal bovine serum (Whittaker M.A. Bioproducts Inc., MD) in a humidified atmosphere of 5% CO2 at 37°C. The cell lines used were as follows: TMK-1 is a poorly differentiated adenocarcinoma cell line established in our laboratory. The additional cell lines in this study were kindly provided by Dr. Suzuki (Fukushima Medical College, Japan) and included MKN-1, an adenosquamous-cell carcinoma; MKN-28, and -74, well-differentiated adenocarcinoma lines and MKN-45, a poorly differentiated adenocarcinoma.
5-Fluorouracil was purchased from Kyowa Hakko Co. (Tokyo, Japan) and docetaxel (Taxotere) was obtained from Rhone-Poulenc Rorer (Antony, France).
Cell growth was assessed by a standard 3-(4,5-dimethyl-2-tetrazolyl)-2,5-diphenyl-2H tetrazoluim bromide (MTT) assay (CellTiter 96 aqueous nonradioactive MTT Cell Proliferation Assay, Promega, Madison, WI), which detects viable dehydrogenase activity. Briefly, gastric carcinoma cells were seeded into 96-well culture plates (5×103 cells/well) in RPMI containing 7% FBS, 2.0 mM L-glutamine and 1% nonessential amino acids. Following a 24-hr incubation in RPMI 1640 medium with 5% FBS, the cells were incubated for 24, 48, 72, 96 and 120 hr at 37°C in a humidified atmosphere of 5% CO2 with varied concentrations of 0.01 to 10 nM TXT, in the absence or in the presence of 1 μM 5-FU. After incubation, 10 μl of MTT (Sigma) solution (5mg/ml) was added to each well, and the plates were then incubated for 3 hr at 37°C. The growth medium was then replaced with 150 μl of dimethyl sulfoxide (Wako) per well, and the absorbance at 540 nm was measured using a Titertek Multiscan.
Total RNA was isolated using the Rneasy mini kit (Qiagen Inc., Chtsworth, CA), according to the manufacturer's instructions. 1 μg of total RNA was converted to cDNA using the GeneAmp RNA PCR Core Kit (Applied Biosystems, Tokyo, Japan). PCR was performed with the QuantiTect SYBR Green PCR kit (Roche) for the OPRT and TS genes, and using the SYBR Premix Ex Taq (Takara, Shiga, Japan) for the DPD gene. Detection of the emission intensity in real time of SYBR Green bound to double-strand DNA was carried out using the Roche PCR Thermal Cycler MP.
The PCR conditions for OPRT and TS amplification consisted of an initial 15 sec degradation at 95°C, followed by 55 cycles of 94°C for 15 sec, 55°C for 20 sec and 72°C for 15 sec. This was followed by a final 15 sec extension at 65°C. The PCR conditions for the DPD gene consisted of initial 15 sec degradation at 95°C, followed by 50 cycles of 95°C for 5 sec and 60°C for 20 sec. The initial template concentrations were derived from the cycle number at which the fluorescent signal crossed a threshold in the exponential phase of the PCR reaction. The relative gene expression was determined by the threshold cycles (cross point) for both the OPRT and TS genes, and for the ACTB gene (initial control). OPRT primer sequences were 5′-TCC TGG GCA GAT CTA GTA AAT GC-3′ (OPRT-1107F) and 5′-TGC TCC TCA GCC ATT CTA ACC-3′ (OPRT-1282R). TS primer sequences were 5′-GAA TCA CAT CGA GCC ACT GAA A-3′ (TS-1) and 5′-CAG CCC AAC CCC TAA AGA CTG A-3′ (TS-r3). DPD primer sequences were 5′-AAT GAT TCG AAG AGC TTT TGA AGC-3′ (DPD-F11) and 5′-GTT CCC CGG ATG ATT CTG G-3′ (DPD-R11). ACTB primer sequences were 5′-CCA ACT GGG ACG ACA TGG AG-3′ (ACTB-L) and 5′-GCA CAG CCT GGA TAG CAA CG-3′ (ACTB-R).
Western blot analysis
To further examine the cooperative effects of 5-FU and TXT on the expression of OPRT, TS and DPD genes in TMK-1 cells, Western blot analysis was performed. Gastric carcinoma cells were seeded into 250 ml flasks, grown to confluence and incubated in the presence of 10 nM TXT either alone or in simultaneous combination with 10 μM 5-FU for different periods of incubation up to 24 hr. Cells were then harvested and lysed in a modified protein lysis buffer (10 mM Tris-HCl, pH 8.0, 135 mM NaCl, 1 mM EDTA, 10 mM CHAPS, 10 μg/ml aprotinin, 0.02 mM APMSF), and the protein concentration of the lysates was measured. Total cell protein extracts (15 μg/lane) were separated by SDS-PAGE using readygels J (Bio-Rad Laboratories, Milan, Italy) and were electrophoretically transferred onto PVDF membranes. The membranes were blocked with 5% nonfat dried milk in PBS buffer and 0.05% Tween 20. The filters were then incubated with primary antibodies against β-actin (Sigma), and OPRT, DPD and TS (Taiho pharma, Tokyo, Japan). Each of these antibodies was diluted as recommended by the manufacturer. Membranes were then washed with PBS-T buffer (PBS with 0.1% Tween20) and incubated with the appropriate secondary antibodies. Immunoreactive proteins were visualized by enhanced chemiluminescence (Amersham International, Buckinghamshire, UK).
Thymidylate synthase activity
TS activity was quantified using a tritiated FdUMP-binding assay according to the method of Spears et al.27 Briefly, tumors were thawed at 4°C and placed in 3 volumes of 0.2 M Tris-HCl buffer, pH 7.4, containing 20 mM 2-mercaptoethanol (2-ME), 15 mM cytidine 5′-monophosphate (5′-CMP) and 100 mM NaF. The tissue was then homogenized by sonication and centrifuged for 15 min at 3000 rpm (+4°C), and the supernatant was centrifuged for 1 hr at 105,000g (+4°C). This was followed by the addition of 50 μl of Buffer A (600 mM NH4HCO3 buffer (pH 8.0)/100 mM 2-ME, 100 mM NaF, 15 mM 5′-CMP) to 50 μl aliquots of the supernatant. TS activity levels were then assayed following the addition of 50 μl of [6-3H] FdUMP (7.8 pmol), plus 25 μl of cofactor solution containing 50 mM potassium phosphate buffer (pH 7.4), 20 mM 2-ME, 100 mM NaF, 15 mM 5′-CMP, 2% bovine serum albumin, 2 mM tetrahydrofolic acid, 16 mM sodium ascorbate and 9 mM formaldehyde. The mixture was then incubated at 30°C for 20 min followed by the addition of 100 μl of 2% BSA and 275 μl of 1 N HClO4 and centrifugation for 15 min at 3000 rpm (+4°C). The precipitate was applied to 2 ml of 0.5 N HClO4, followed by sonication and centrifugation for 15 min at 3000 rpm (+4°C) once more. The precipitate was solubilized with 0.5 ml of concentrated acetic acid, and after the addition of 10 ml ACS II, radioactivity levels were counted.
Digydropyrimidine dehydrogenase activity
DPD activity was determined by a catalytic assay according to the method described by Yamada et al.28 Briefly, approximately 100 mg of tissue sample was thawed and placed in 1 ml of homogenization buffer containing 10 mM Tris-HCl, 1 mM EDTA and 0.5 mM dithiothreitol (DTT), pH 7.4. The tissue was then homogenized using a polytron, and the homogenate was centrifuged at 105,000g for 60 min (+4°C). The supernatant fluid (cytosol) was collected as the enzyme source. Internal substrate, including uracil and thymine, which might inhibit DPD activity, was removed from the cytosol using a MicroSpin G-25 column (Amersham Pharmacia Biotech). The resulting cytosolic fraction was frozen and stored −80°C until use. The reaction mixture (25 μl) contained 20 mM [6-14C] 5-FU, 0.25 mM β-nicotinamide adenine dinucleotide phosphate (NADPH), 2.5 mM MgCl2 and 35 mM NaH2PO4, and was preincubated at 37°C for 1 min. The assay mixture consisted of incubating 25 μl of cytosol with reaction mixture (25 μl) for 10–30 min at 37°C. In cases of highly reactive cytosol, the duration of the incubation was shortened, and in instances of DPD deficiency, the incubation period was increased to 60 min. These reactions were terminated by heating in boiled water (∼90°C) for 3 min. After the addition of 0.36 M KOH (25 μl), the mixture was left for 30 min at room temperature. Subsequently, 25 μl of 0.36 M HClO4 was added and this was followed by centrifugation for 5 min at 15,000 rpm (+4°C). A 5 μl aliquot of the supernatant was then applied to a thin-layer chromatography (TLC) plate (Merck TLC plates which silica gel 60 F254 pre-coated). TLC was first carried out for ca. 5 cm in a mixture of 99% ethanol and 1 M ammonium acetate (5:1, v/v) and then performed for ca.15 cm in ethylether-acetone-chloroform-water (50:50:40:1, v/v/v/v). The metabolites on the TLC plate were analyzed by radioluminography (BSA2000, Fuji film). DPD activity was calculated by taking into account the sum of α-fluoro -β-alanin (FBAL),and 2-F-3-ureidopropionate (FUPA) peaks and expressed as picomole of 14C-FU catabolized per min per milligram protein. The protein concentrations in the cytosol were determined by the Bio-Rad protein dye using a standard curve of BSA, fraction V.
Five-week-old female BALB/c-nu/nu mice were obtained from Japan SLC, Inc. (Hamamatsu, Japan) and maintained for 1 week in our animal facility before tumor inoculation. Animals were housed on a 12-hr light/dark cycle with food and water provided ad libitum in a barrier facility fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. Animals from an international origin were examined before commencing our experiments, to ensure that they were in good health and had acclimatized to a laboratory environment. All animal experiments were conducted in accordance with the “Guidelines for the Care and Use of Laboratory Animals in Nippon Roche Research Center.”
Tumor xenograft studies of TXT and 5-FU
Suspensions of TMK-1 tumor cells (5–10 × 106 viable cells/mouse) were inoculated s.c. into the bilateral flank region of the BALB/c-nu/nu mice. Analyses of these human tumor xenografts were initiated when the tumor volumes reached 50 mm3. Six to seven mice each were randomly allocated to different groups to be treated with: (i) vehicles: sterile normal saline by intraperitoneal (i.p). injection every day; (ii) 10 mg/kg TXT dissolved in saline containing 2.5% ethanol and 2.5% polysorbate 80 by i.p. injection on day 1 and 22; (iii) 15 mg/kg 5-FU in sterile saline by i.p. injection on days 1–14 and 22–35; (iv) 10 mg/kg TXT dissolved in saline containing 2.5% ethanol and 2.5% polysorbate 80 by i.p. injection on day 1 and day 22 and 15 mg/kg 5-FU in sterile saline by i.p. injection on days 1–14 and 22–35. The tumor volumes were estimated by V = ab2/2, where a and b are tumor length and width, respectively. To evaluate the antitumor effects of the 5-FU and TXT, tumor sizes and body weights were measured 2–3 times per week. Carcass body weight was calculated by subtracting the tumor weight, which was estimated from the tumor volume, from the body weight.
Tumor xenograft studies of S-1 vs. 5-FU
Fragments of TMK-1 tumor tissues were implanted s.c. into the bilateral flank under anesthesia and allowed to establish growth for about 1 week after implantation. Analysis of these xenografts were commenced when tumor volume reached <100 mm3. Six mice each were randomly allocated to different groups that were treated with: (i) vehicle: sterile normal saline by i.p. injection daily for 2 cycles (1 cycle occurs for 2 weeks and the following cycle initiated after a 1 week absence); (ii) 20 mg/kg 5-FU in sterile saline by i.p. injection daily for 2 cycles; (iii) 10 mg/kg S-1 by daily oral gavage for 2 cycles; (iv) 10 mg/kg TXT dissolved in saline containing 2.5% ethanol and 2.5% polysorbate 80 by i.p. injection on the first day of each cycle and 20 mg/kg 5-FU in sterile saline by i.p. injection daily for 2 cycles; (v) 10 mg/kg TXT dissolved in saline containing 2.5% ethanol and 2.5% polysorbate 80 by i.p. injection on the beginning day of each cycle and 10 mg/kg S-1 by oral gavage daily for 2 cycles. Measurements were performed as described earlier.
Cell growth inhibition assay
We analyzed the synergistic growth inhibitory effects of 5-FU and TXT on the 5 gastric cancer cell lines TMK-1, MKN-1, MKN-28, MKN-45 and MKN-74. MTT assays were performed to examine the growth inhibitory effect of TXT alone (Fig. 1a) and in combination with 5-FU (Fig 1b). Each cell line was treated with 0.1, 1, 10 or 100 nM TXT, either alone or in combination with 1 μM 5-FU, and the time-course effects were observed at 24, 48, 72 and 96 hr posttreatment. MTT assays were performed to analyze the inhibitory effects of the drugs at each time point, compared to nontreated control cells. Growth inhibitory effects following TXT treatment were observed in both a time- and dose-dependent manner for each of the 5 gastric cancer cell lines under study. The TXT IC50 levels of each were found to be 4.72, 114.7, 2.25, 3.11 and 14.8 nM for TMK-1, MKN-1, MKN-28, MKN-45 and MKN-74, respectively. To determine the biochemical effects of TXT upon 5-FU activity, the growth inhibitory effects of 5-FU treatment alone or in combination with TXT in these cell lines were evaluated. The IC50 measurements of 5-FU for each cell line were compared with those from 5-FU treatment in combination with 1.0 nM TXT (Table I). These values were significantly reduced in each case, from 2.98- to 4.30-fold, suggesting that there were clear synergistic effects, which resulted from the combination of 5-FU and TXT, upon cell growth inhibition in gastric cancer cells.
Table I. The IC50 Values of Gastric Cancer Cell Lines by 5-FU Alone and in Combination with 1.0 nM TXT
5-FU alone (μM)
5-FU 1.0 nM TXT (μM)
Growth inhibitory effect of TXT alone and in combination with 5-FU in the 4 gastric cancer lines, TMK-1 and MKN-28, -45 and -74. MTT assays were performed. The lC50 values of each treatment were shown in 4 gastric cancer cell lines at 72 hr after treatments.
Changes in the expression levels of the OPRT, DPD and TS metabolic enzymes after treatment with TXT and 5-FU
To further elucidate the synergy of action between TXT and 5-FU, we examined possible changes in the expression levels of OPRT, DPD and TS, which are genes that function in the metabolic pathways of 5-FU. These expression profiles were examined by Western blot and RT-PCR analysis following treatment of gastric cancer cells with TXT alone or in combination with 5-FU.
TMK-1 cells were treated with 10 nM TXT alone (Fig. 2a) or in combination with 10 μM 5-FU (Fig. 2b) for up to 24 hr, and the expression levels of the OPRT, DPD and TS proteins were investigated by Western blot analysis. We observed a transient increase in the protein levels of OPRT (at 6 hr posttreatment, a 2.5-fold increase was measured compared with the control; at 24 hr, a 2.1-fold increase was detected) and DPD (at 12 hr posttreatment, a 2.6-fold increase was evident) following TXT exposure alone. Significantly, however, we detected a marked time-dependent increase in OPRT (3.9-fold after 24 hr treatment) and decrease in DPD protein expression (at 27% of control levels after 24 hr treatment) following the combined TXT/5-FU treatment. Moreover, the expression of TS protein was slightly decreased after TXT treatment alone (72% of control levels after 24 hr treatment), but showed a greater reduction (49% of the control levels after 24 hr treatment) in a time-dependent manner after treatment with both TXT and 5-FU. These findings strongly suggest that the stability of 5-FU can be increased, likely due to increased phosphorylation of the protein, and that the potency of this compound is enhanced in conjunction with a reduction in TS protein levels.
This synergism between TXT and 5-FU was also confirmed following the dosage experiments with TXT (Fig. 3). TMK-1 cells were grown until they reached confluence and then incubated in the presence of 1, 10 or 100 nM TXT in combination with 10 μM 5-FU for 24 hr. The protein expression levels of the OPRT, DPD, TS and β-actin genes were then examined by Western blot analysis. A significant dose-dependent increase in OPRT expression, and a dose-dependent decrease in both DPD and TS expression, could be observed after treatment with a combination of TXT and 5-FU (Fig. 3).
To examine whether the expression changes in the 5-FU metabolic pathway genes were due to regulation at the transcriptional level, their mRNA levels were analyzed by RT-PCR (Fig. 4). TMK-1 cells were treated with 10 nM TXT, either alone or in combination with 10 μM 5-FU, for 12 hr and mRNA was then isolated from the cells. As shown in Figure 4, treatment with TXT in combination with 5-FU significantly increased the mRNA expression levels of OPRT (1.78 ± 0.18-fold increase following 100 nM TXT in combination with 5-FU, compared with control cells) in a dose-dependent manner. In contrast, under the same conditions, the DPD transient levels were significantly reduced in a dose-dependent manner (decreased to 25% ± 0.5, 23% ± 5, 21% ± 5 of the control levels for 1, 10, 100 nM TXT, in combination with 5-FU, respectively). In the case of TS mRNA expression, although 5-FU treatment caused a significant increase in these transcript levels (10 μM 5-FU treatment induced a 2.88 ± 0.06-fold increase in TS mRNA, compared to the untreated control), a combination of TXT and 5-FU was resulted in a clear decrease in these levels (TS mRNA levels were measured at 46% ± 1 and 31% ± 0.3 for 10 and 100 nM TXT in combination with 5-FU, respectively, compared to cells treated with 5-FU alone).
Activity assays for DPD and TS in TMK-1 cells
The results from our analyses of the gene expression of enzymes that operate during 5-FU metabolism prompted us to examine the specific activities of both TS and DPD to further elucidate the biochemical modulation of 5-FU by TXT co-stimulation (Fig. 5 and Table II). Figures 5a and 5b show the results of our subsequent TS and DPD activity assays in TMK-1 cells treated by 5-FU alone, TXT alone or TXT and 5 FU in combination. Marked increases in the activities of DPD and TS were detectable 5-FU treatment alone in TMK-1 cells. However, these activities were reduced after TXT treatment. Moreover, as demonstrated also by our protein expression analyses, a marked inhibition of both DPD and TS activity was observed after treatment with a TXT and 5-FU combination, suggesting that these enzyme activities are modulated by synergistic actions of these 2 compounds.
Table II. TS and DPD Activity Levels in the Presence of TXT and 5-FU
TS activity (%)
TXT (nM) + 10 μM 5-FU
In vivo effects of a combination treatment with TXT and either 5-FU or S-1 in human tumor xenografts
As already shown in our current in vitro experiments, the synergy between TXT and 5-FU is most likely to be due to the biochemical modulation of 5-FU by TXT. To assess the potential synergistic effects of these drugs in vivo, we performed human tumor xenograft experiments. In these analyses, we further examined the growth inhibitory effects of TXT, in combination with either 5-FU or S-1, in TMK-1 cells that had been transplanted into nude mice (Figs. 6 and 7). We adopted treatment schedule from the method of Yoshida et al.26 Briefly, TXT was administrated by i.v. infusion on day 1 at 3 weeks intervals, whereas S-1 was administered orally each day for 2 weeks, followed by a 1 week interval. The other regimens utilized in these experiments are described in detail in the Materials and Methods. The concentration of TXT that was used, 10 mg/kg, is 10% of the highest nonlethal dose levels for this compound.10 In addition, the dosage levels of 5-FU that were administered, 15 mg/kg, are at 60% of the maximum tolerable in mice.29 The resulting growth curves of gastric cancer xenografts during the various treatment schedules are shown in Figure 6. Dosages of 15 mg/kg 5-FU, 10 mg/kg TXT and a combination of these two, induced a 5, 40 and 52% of inhibition of tumor growth at day 15 (at the end of the 1st cycle), respectively. Moreover, tumor growth inhibition of 18, 28 and 65% was observed at day 36 (at the end of the 2nd cycle), for these 3 treatments, respectively, further suggesting the synergistic effects of the TXT/5-FU combination. In addition, none of the treated animals in these experiments showed carcass weight loss due to drug toxicity.
We next investigated the growth-inhibitory effects of S-1 and TXT in our xenograft system, in comparison with 5-FU and TXT. S-1 is a novel anticancer agent, which contains a DPD inhibitor, and is currently in widespread use in Japan because of its superior response against advanced gastric cancers.13, 15 The growth curves of the gastric cancer xenografts following 5-FU and S-1 treatments are shown in Figure 7. Interestingly, 20 mg/kg 5-FU, 10 mg/kg S-1, 20 mg/kg 5-FU and 10 mg/kg TXT in combination, and 10 mg/kg S-1 and 10 mg/kg TXT in combination, induced a 36.5, 29.0, 63.0 and 71.0% inhibition of tumor growth, compared to the controls, on day 15 (at the end of the 1st cycle). This is compared with a 34.0, 37.1, 53.1 and 78.4% inhibition of tumor growth at day 36 (at the end of the 2nd cycle), respectively. Once again, none of the treatment groups in this experiment showed carcass weight loss due to drug toxicity. These results demonstrate that the combination of TXT and S-1/5-FU is effective against gastric cancer and has the potential to be a new therapeutic tool for future treatments of these tumors.
Although there is a general consensus that the prognosis for nonresectable or recurrent gastric cancer patients is improved with chemotherapy,1, 2, 3 a standard chemotherapeutic regimen for gastric cancer has not yet been established. S-1, an oral fluorouracil antitumor drug, has been identified as an effective agent for the treatment of gastric cancer, and this drug has gradually been adopted as part of the front-line regimens for treatments of these tumors in Japan. Moreover, combination therapies with S-1 and new agents, including the taxanes, CDDP30 and CPT-11,31 have generated great interest recently. As reported in our recent phase I/II study, we have examined S-1 and TXT (S-1/TXT) as a combination therapy and demonstrated both its efficacy and tolerability.26 In our present study, we attempted to further elucidate the apparently synergistic effects of TXT and either 5-FU or S-1, using both in vitro and in vivo studies in the TMK-1 gastric cancer cell line.
Our results show that the synergistic effects of TXT and either 5-FU or S-1 were evident both in vitro and in vivo and could be explained by our observations showing biochemical modulation of the expressions and activity of the TS, DPD and OPRT enzymes, which play key roles in the functional activities of 5-FU or S-1. The cotreatment of TXT and 5-FU increased the expression of OPRT and decreased TS and DPD activity, which can explain why the combination of TXT and 5-FU enhances antitumor activity of 5-FU and reduces resistant potential to 5-FU. 5-FU is catabolized to 2-fluoro-β-alanin by DPD which is the first and rate-limiting enzymes in this pathway. Moreover, the enhanced activity of DPD by 5-FU has been shown in a number of previous studies to confer tumor resistance to 5-FU, both in vitro and in clinical studies.32, 33, 34, 35 Low levels of DPD mRNA expression has also been associated with a low sensitivity to 5-FU and has also been correlated with a low sensitivity to fluoropyrimidine-based chemotherapy in colorectal cancer patients.36, 37, 38, 39
TS as well as DPD is an enzyme that plays key roles in 5-FU resistance in carcinoma cells. High intratumoral TS mRNA levels are believed to confer 5-FU resistance due to resulting inefficient TS inhibition.40 In primary gastric and colorectal cancers, TS mRNA expression has also been correlated with the response and survival of the primary and metastatic tumors when treated using a 5-FU containing regimens.35, 41, 42, 43 Our results show that TS expression and TS activity was decreased by combination treatment of TXT and 5-FU despite of induction of TS expression and TS activity by 5-FU treatment alone. Nishiyama et al.44 has demonstrated that it was not the basal TS gene expression level, but it was the induction of TS gene expression that was related to resistance in 5-FU resistant cells. Moreover, TS mRNA and protein induction by 5-FU seems to be due to the inhibition of a negative-feedback mechanism in which ligand-free TS proteins bind to TS mRNA, inhibiting their translation.45 Although the precise mechanism how TS gene expression is induced after 5-FU treatment is not fully understood, the resistant cells to 5-FU may have the capacity to overcome the expression of TS, which is consumed and reduced by formation of ternary complex with 5-FU. It has also been suggested recently that polymorphisms within the TS gene, rather than the number of gene copies, may influence the degree of protein expression.46, 47
Concerning the expression of DPD, the mRNA expression of DPD following 10 μM 5-FU exposure alone decreased as demonstrated in Figure 4b and DPD activity was increased following 5-FU treatment dose-dependently as demonstrated in Figure 5b. However, there is a general consensus that there can be some discrepancies among mRNA, protein expressions and the enzyme activity, because there can be several regulatory mechanisms including posttranscriptional and posttranslational regulation in the expression of a gene and also there can be a conditional regulation concerning the enzyme activity. Regulatory mechanism of DPD activity may be associated with negative-feedback system, regulation by transcriptional factor of DPD gene, such as AP-1,48 methylation of the promoter region49 or unknown DPD inhibitor. Further investigation should be required for precise explication of regulatory mechanism of DPD activity. Our data indicate that the cotreatment of TXT and 5-FU efficiently inhibits the growth of gastric tumor cells by reduction of TS and DPD activity, although the exposure of TMK-1 cells to 5-FU alone induces resistance to this agent. Hence, we speculate that the cotreatment of TXT and 5-FU may recover the sensitivity of the cells to 5-FU.
It is noteworthy that expression levels of OPRT were increased following treatment with TXT and 5-FU, suggesting the efficient uptake of 5-FU into the DNA of the exposed cells. OPRT directly metabolizes 5-FU to FUMP in the presence of 5-phosphoribosyl–1- pyrophosphate (PRPP), and has been shown to correlate with a higher sensitivity to 5-FU in cell lines and human xenograft models.50 In studies using human colorectal cancer tissues, a higher level of OPRT enzymatic activity was observed in vitro in 5-FU-sensitive tissues, than in nonsensitive ones, in chemosensitivity tests.50, 51 Moreover, Ichikawa et al.52 have reported that both the expression of the OPRT gene and the OPRT/DPD ratio might be useful as a predictive parameter for the efficacy of fluoropyrimidine-based chemotherapeutics in metastatic colorectal cancer. These reports supported our hypothesis that synergistic effects of TXT and 5-FU were relevant to induction of OPRT expression by the combined TXT/5-FU treatment. We suggest that not only DPD and TS but also OPRT expression closely correlates with the synergism between TXT and 5-FU.
There can still be arguments concerning that the decrease of TS and DPD may be due to the cell cycle arrest or apoptotic pathway, because the expressions of apoptotic related genes, including caspase 8, caspase 3 and PARP, were activated markedly (data not shown) following the cotreatment of TXT and 5-FU. However, we have examined the expression of β-actin as the control, which was not altered after the treatments. Moreover, the expression of OPRT was increased as we have discussed earlier. Although it should be elucidated whether the phenomenon is due to the cell cycle related or not, we suggest that the changes of expression of TS, DPD and OPRT can be regarded as a specific response to the combination treatment of TXT and 5-FU.
In our in vivo studies, we found a remarkable synergistic effect between TXT and 5-FU, and marked growth inhibitory effects of TXT and S-1 in combination, which was more potent than TXT and 5-FU in combination. A possible explanation for the higher potency of TXT and S-1 is that antitumor effects of S-1 in gastric cancer have been shown not to be influenced by intratumor DPD gene expression, although intratumor DPD expression inversely correlates with the sensitivity to 5-FU.29, 39, 53 Overall, however, predicting the efficacy of S-1 in a clinical setting might be of major interest in the future including assessments of the status of OPRT, DPD and TS. Finally, our data strongly indicate that the combination chemotherapy of TXT and S-1 is effective against gastric carcinomas and is therefore a good candidate as a standard chemotherapeutic strategy in treating these tumors.