Comparative cytotoxicity of folate-based inhibitors of thymidylate synthase and 5-fluorouracil ± leucovorin in MGH-U1 cells

Thymidylate synthase (TS) is a critical enzyme in the synthesis of DNA and an important target for cancer chemotherapy. 5-Fluorouracil (5FU) combined with leucovorin (LV) has been used to inhibit TS, and inhibition is dependent on the formation of a ternary complex between a folate cofactor, TS, and 5-fluorodeoxyuridine monophosphate (FdUMP), a metabolite of FU. The folate-based TS inhibitors CB3717, its analogs, and BW1843U89 have been synthesized as specific inhibitors of TS that do not require activation or the presence of a cofactor. We have compared the cytotoxicity of 5FU ± LV with that of these folate-based TS inhibitors in human bladder cancer MGH-U1 cells using a colony-forming assay. After a 6-h exposure, FU+LV, CB3717, dCB3717, or C2 methyl dideazafolate analogs demonstrated similar cytotoxic potency that was 0.96 to 2.9 times that of 5FU alone. A 24-h exposure did not increase the potency of 5FU+LV relative to 5FU alone, but there was a marked increase in the cytotoxicity of the dideazafolates as compared with 5FU+LV. Similarly, BW1843U89 was more cytotoxic than 5FU+LV. This was reflected in a 3.2- to 1333-fold decrease in the 50% inhibitory concentration (IC₅₀). Simultaneous exposure to LV and thymidine (TdR) protected MGH-U1 cells from the cytotoxicity of CB3717, its analogs, and BW1843U89. We conclude that (a) the folate-based TS inhibitors are more potent than 5FU+LV after a 24-h exposure, (b) protection by LV and TdR indicates that TS inhibition is the primary site of action, and (c) BW1843U89 is more potent than D1694 in MGH-U1 cells.This study was supported by a grant from the National Cancer Institute of Canada     

5-FU sensitivity and thymidylate synthase in gastric cancer

The relationship of 5-FU-sensitivity to thymidylate synthase (TS) was investigated in a total of 82 gastric cancers of stage III or IV. The sensitivity test was done by Histoculture Drug Response Assay and TS expression was stained immunohistochemically using anti-TS antibody. Sensitivity to 5-FU of more than 50% of the inhibitory index (high sensitivity) was observed in 28.1% of the materials, less than 50% (low sensitivity) in 71.9% and less than 20% in 50.0%. The expression of TS was found in 25.6% of the patients. The incidence of positive TS expression was 34.5% in the high sensitivity group, against 22.0% in the low sensitivity group, with no statistical difference between them. No particular relationship was found between 5-FU-sensitivity group and TS expression rate when the tumor-histology was divided into high and low differentiation. In the present study, the ratio of patients negative for TS among the high 5-FU-sensitivity group both was 65.2%, both of which have been reported as producing high survival rates. In contrast, the ratio of patients positive for TS in the low 5-FU-sensitivity group was 22.0%. From these results, TS is considered to be responsible for approximately 40-50% of either high or low 5-FU-sensitivity, respectively, when the lower TS expression in the present study was taken into consideration and corrected.     

Schedule-dependent inhibition of thymidylate synthase by 5-fluorouracil in gastric cancer.

BACKGROUND. An optimal treatment schedule of 5-fluorouracil (5-FU) remains to be clarified. METHODS. A randomized study was conducted to investigate schedule-dependent thymidylate synthase (TS) inhibition by 5-FU in 16 patients with gastric cancer who underwent surgical resection. Surgical specimens of tumor, normal gastric mucosa, and regional lymph nodes were obtained 12 hours after administration of 5-FU either as a continuous infusion (1000 mg/m2 for 48 hours) or as a bolus injection (500 mg/m2 x 2 in 48 hours). RESULTS. The total TS activity (567.8 +/- 294 fmol/mg protein) and the rate of TS inhibition (74.7 +/- 23.1%) in cancer tissues were significantly higher in the continuous-infusion group than in the bolus-injection group (228.5 +/- 104.6 fmol/mg protein and 48.8 +/- 12%, respectively). Likewise, the total TS activity (807.4 +/- 440.3 fmol/mg protein) and the rates of TS inhibition in lymph nodes (72.3 +/- 17.1%) and in normal gastric mucosa (85.1 +/- 12.2%) were significantly higher in the continuous-infusion group than those in the bolus-injection group (232.4 +/- 142.3 fmol/mg protein and 53.6 +/- 17.0% in lymph nodes and 46.5 +/- 14.3% in normal gastric mucosa, respectively). There was a significant correlation between the total TS activity and TS inhibition. CONCLUSIONS. Continuous infusion of 5-FU provides a superior antimetabolic effect in the treatment of gastric cancer, which may lead to a superior antitumor effect.     

Apoptosis and thymidylate synthase inductions by 5-fluorouracil in gastric cancer cells with or without p53 mutation.

We studied apoptosis and thymidylate synthase (TS) inductions by 5-fluorouracil (5-FU) in gastric cancer cells with wild-type p53 (MKN-45 and MKN-74) and with mutated p53 (MKN-28 and KATO-III). Apoptotic inductions in MKN-45 and MKN-74 were stronger than those in MKN-28 and KATO-III, suggesting that wild-type p53 may contribute to the induction of apoptosis. After continuous exposure to 0.1 microg/ml of 5-FU for 96 h, no TS induction was obtained in KATO-III, while approximately twice the amount of TS was observed compared to non-treatment cells in MKN-45, MKN-74, and MKN-28. The results of immunohistochemical staining for TS and p53 showed no relation between these two protein expressions in endoscopic biopsy specimens of 25 patients with advanced gastric cancer. These results indicated that p53 status may not play a pivotal role in regulating TS expression. We found no significantly different effects of 5-FU between intermittent (repeat of 24-h continuous infusion and 24-h drug-free) and continuous treatments in either MKN-28 or stem cells (CD34+ hematopoietic progenitor cells) when the same area under the time-concentration curve of 5-FU was administered. The TS induction in MKN-28 cells by intermittent treatment was significantly higher than that by continuous treatment; however, most TSs in both types of 5-FU treatment cells were of the inactive form, i.e., TS bound to FdUMP, a 5-FU metabolite. Therefore, neither intermittent nor continuous treatment appears to induce 5-FU resistance related to the level of increment free TS. In conclusion, our observations suggested that p53 mutation may be associated with apoptotic induction by 5-FU; however, p53 status may not strongly affect TS induction by 5-FU. Intermittent treatment can be replaced with continuous treatment without causing 5-FU resistance.     

Inhibition of thymidylate synthase by the diastereoisomers of leucovorin

Summary The clinical formulation of leucovorin (5-CHO-FH₄) is a mixture of diastereoisomers with markedly different pharmacologic properties. Comparatively little information is available concerning the cellular pharmacology of reduced folate stereoisomers, due largely to the difficulty in preparing sufficient quantities of these compounds for in vitro use. Recent improvements in HPLC technology have now facilitated this process, enabling studies of folate stereochemistry on a larger scale. Using purified (6R) and (6S) leucovorin, we examined the effects of these compounds on the enzymatic activity ofLactobacillus casei thymidylate synthase (TS) in a cell-free system. The natural (6S), unnatural (6R), and racemic (6R,S) leucovorin preparations inhibited TS activity by 50% at concentrations of 0.11, 2.1, and 0.52 mM, respectively. Dixon plots demonstrated the inhibition to be competitive, with K_i values of 85μM, 1.59 mM, and 385μM for (6S), (6R), and (6R,S) leucovorin, respectively. In view of the high doses of leucovorin given clinically and the slow clearance of the unnatural isomer, our observations suggest that leucovorin may have important direct inhibitory effects on folate-requiring enzymes.     

Association of thymidylate synthase polymorphisms with gastric cancer susceptibility

We investigated in a case-control study a possible role of thymidylate synthase gene (TS) polymorphisms for gastric cancer susceptibility. Lymphocyte genomic DNA from 134 Italian gastric cancer patients and 139 controls was used for genotyping two polymorphisms in the TS 5'-untranslated region (5'-UTR); a double (2R) or triple (3R) 28-bp repeat and a G/C polymorphism within the triple repeats allele (3G allele). Samples were also genotyped at a 6-bp deletion/insertion (del6 or ins6) polymorphism at position 1494 in the TS 3'-untranslated region (3'-UTR). Unconditional regression with odd ratios (OR) and 95% confidence intervals (CI), haplotype and linkage disequilibrium analyses were used to investigate the association of the polymorphisms with the disease. The global allelic distribution was in Hardy-Weinberg equilibrium. Genotypes with the 3G allele (2R/3G, 3C/3G, 3G/3G) were significantly more frequent in patients than controls and were associated with gastric cancer risk (OR = 2.06; 95% CI = 1.26-3.35). A significant risk was also observed for carriers of the del6 allele in the 3'-UTR. Odds ratios for combined 3G-del6/ins6 and 3G-del6/del6 genotypes were 2.59 (95% CI = 1.36-4.94) and 2.81 (95% CI = 1.22-6.64), respectively. The 3G-del6 haplotype showed a significant association with the disease (p = 0.01). Polymorphisms in the TS gene may contribute to gastric cancer susceptibility and this finding deserve further investigation in the context of novel strategies for gastric cancer prevention. In vitro, 3G genotypes have been related to high TS mRNA expression, which may underlie one of the possible etiologic mechanisms.     

RAD001 shows activity against gastric cancer cells and overcomes 5-FU resistance by downregulating thymidylate synthase

We evaluated RAD001, an inhibitor of the mammalian target of rapamycin (mTOR) in human gastric cancer cell lines and determined the molecular mechanisms. RAD001 has marked growth inhibitory activity against the SNU-1 and SNU-216 cells. It inhibited phosphorylation of mTOR and S6 K, and induced G1 cell cycle arrest. Synergistic growth-inhibitory effects in combination with 5-fluorouracil (5-FU) was identified. Furthermore, RAD001 conferred sensitivity to 5-FU-resistant cell lines by downregulating thymidylate synthase (TS). In conclusion, RAD001 showed growth inhibitory activity against gastric cancer cells and acted synergistically with cytotoxic agents such as 5-FU by downregulating TS.     

Lack of effectiveness of combined 5-fluorouracil and leucovorin in patients with 5-fluorouracil-resistant advanced colorectal cancer

Favorable results have been reported for the treatment of advanced colorectal cancer with the combination of 5-fluorouracil (5-FU) and leucovorin (LV). In these investigations the highest response rates were obtained in non-pretreated patients. In the present study, 22 patients with primary or acquired resistance to single-agent 5-FU and documented progressive disease on 5-FU were given a bolus injection of LV at a dose of 200 mg/m², 1 h prior to a 2 h infusion of 5-FU at a dose of 370–770 mg/m² on 5 consecutive days, and this was repeated every 3 weeks. Whenever possible the dose of 5-FU was escalated to find the maximum tolerable dose. No objective response was observed. Five patients had short-lasting stable disease. Despite 5-FU dose escalation, toxicity was acceptable. One patient developed 5-FU-related angina pectoris with EKG changes. It is concluded that in the schedule used, combined LV/5-FU treatment is ineffective for patients with 5-FU-resistant advanced colorectal cancer.     

Dihydropyrimidine dehydrogenase and thymidylate synthase polymorphisms and their association with 5-fluorouracil/leucovorin chemotherapy in colorectal cancer

The causes of interpatient variation in severe toxicity resulting from treatment with weekly 5-fluorouracil (5-FU)/ leucovorin (LV) are poorly understood. This study was undertaken to examine the contribution of commonly occurring polymorphisms in the dihydropyrimidine dehydrogenase (DPYD) gene to interpatient variability in 5-FU pharmacokinetics and toxicity. Patients with stage III/IV colorectal cancer were treated by bolus intravenous (I.V.) injection with 500 mg/m2 doses of 5-FU and LV once every week. The pharmacokinetics of 5-FU was determined on weeks 1 and 4. Genotyping assays were developed for 8 polymorphisms in the DPYD gene. A well-characterized functional polymorphism in the 5' untranslated region of the thymidylate synthase (TS) gene was also analyzed. A cohort of 22 patients (15 male, 7 female) with a median age of 61 years was evaluated. Although there was no relationship between the area under the plasma concentration time curve (AUC) for the first dose of 5-FU and worst-grade toxicity during the first cycle of therapy, 3 of the 4 patients in whom the AUC on week 4 was more than equal to 5 microgram/h/mL greater than the value for the first dose experienced grade 3/4 toxicity during subsequent treatment. Among the 8 polymorphisms in the DPYD gene, 7 were found to vary in the study population but none were significantly associated with the AUC of 5-FU. There was no relationship between the DPYD and TS genotypes examined and 5-FU toxicity. Extensive polymorphism in the DPYD gene was observed; however, no conclusive correlations existed between the DPYD and TS genotype and 5-FU pharmacokinetics or toxicity. Decreases in 5-FU clearance in certain patients may provide insight into the increased toxicity following repetitive cycles of treatment with weekly I.V. bolus 5-FU. The present study offers useful themes for undertaking larger prospective pharmacogenetic studies in the future.     

Predictive value of thymidylate synthase expression in gastric cancer: a systematic review with meta-analysis

Purpose: The relationship between thymidylate synthase (TS) expression and outcomes in gastric cancer(GC) patients remains controversial, although most studies reported poor survival and reduced response tofluoropyrimidine were related to high TS in tumors. We carried out a systematic review of the literature withmeta-analysis to estimate the predictive value of TS expression from published studies. Methods: We indentified24 studies analysing the outcome data in gastric cancer stratified by TS expression. Effect measures of outcomewere hazard ratios (HRs) for overall survival (OS) and event-free survival (EFS), or the odds ratio (OR) foroverall response rate (ORR). HRs and ORs from these eligible studies were pooled using random-effects metaanalysis.Results: Fifteen studies investigated outcomes in a total of 844 patients with advanced GC, and ninestudies investigated outcomes in a total of 1,235 patients with localized GC undergoing adjuvant therapy. Metaanalysisof estimates showed high TS expression was significantly associated with poor OS in the advanced setting(HR: 1.43, 95%CI: 1.08 - 1.90), and poor EFS in the adjuvant setting (HR: 1.53, 95%CI: 1.01 - 2.32). Subgroupanalysis demonstrated TS expression to haves even greater value in predicting OS, EFS and ORR in advancedGC patients treated with fluoropyrimidine monotherapy (HR for OS: 2.32, 95%CI: 1.53 - 3.50; HR for EFS:1.76, 95%CI: 1.19 - 2.60; OR for ORR: 0.32, 95%CI: 0.11 - 0.95). Conclusion: High levels of TS expression wereasssociated with a poorer OS for advanced GC patients compared with low levels. In the adjuvant setting, highTS expression was also associated with a worse EFS. Additional studies with consistent methodology are neededto define the precise predictive value of TS.     

Enhancement of fluorinated pyrimidine-induced cytotoxicity by leucovorin in human lung cancer cell lines

Reduced folates have been shown to increase the cytotoxicity of 5-fluorouracil (FUra) by stabilizing the fluorodeoxyuridine monophosphate:thymidylate synthase complex, thus increasing the block in the DNA synthetic pathway. Using an in vitro tetrazolium colorimetric (MTT) cytotoxic assay, we tested the effects of FUra and 5-fluorodeoxyuridine (FUdR) with and without leucovorin (LV) on a panel of 7 human lung cancer cell lines. LV at a concentration of 20 microM enhanced the cytotoxicity of FUra and of FUdR in all of the cell lines. Quantitatively, LV had a higher degree of enhancement on FUdR than on FUra cytotoxicity in 6 cell lines. There was equivalent enhancement in the only remaining line. The differential effects of LV on the cytotoxicity of FUra vs. FUdR in these lung carcinoma lines contrasts with a quantitatively similar enhancement of cytotoxicity between FUra and FUdR in colon cancer lines previously reported from our laboratory. This suggests that the metabolism of FUra may be different in these lung cancer cell lines.     

The association of expression and gene polymorphism of the thymidylate synthase gene with 5-fluoro-5'-deoxyuridine sensitivity in gastric cancer cell lines

Thymidylate synthase (TS) is a target enzyme of 5-fluorouracil (5-FU), and the TS expression level of cancer tissues is a potential predictor of the response to 5-FU-based chemotherapy. The TS gene has a polymorphic tandem-repeat sequence, which is associated with its protein expression. Therefore, the TS polymorphism may also be a predictor of the response to 5-FU-based chemotherapy. In this study, we analyzed the TS genotype, TS protein level, and sensitivity to 5-fluoro-5'-deoxyuridine (5'-dFUrd) in 10 human gastric cancer cell lines. TS genotype was classified into 2R-homozygote (2R/2R, n=3), 3R-homozygote (3R/3R, n=5), and 2R/3R-heterozygote (2R/3R, n=2). The cell lines with 3R/3R showed a significantly higher IC50 value compared to those with 2R/2R or 2R/3R genotype. There was no relationship between TS protein level and 5'-dFUrd sensitivity. However, a statistically significant relationship was revealed between them when the subgroup with the genotypes of 2R/2R or 2R/3R was considered (r=0.815, p<0.05). In this subgroup, the cell lines with higher TS protein showed higher IC50 value for 5'-dFUrd, indicating less sensitivity to 5'-dFUrd. An identical relationship between the TS protein level and IC50 was also observed in the subgroup with 3R/3R genotype, although it did not reach statistical significance (r=0.745, p=0.09). These results suggest that the TS gene polymorphism and TS protein level may be independent predictors for 5-FU-based chemotherapy.     

The expression of thymidylate synthase and thymidine phosphorylase in the early-stage of gastric cancer

The immunohistochemical expression of thymidylate synthase (TS) and thymidine phosphorylase (TP) was investigated in 116 of early gastric cancer, in order to know whether or not these reflect malignity in an early stage. The materials conditioned on early gastric cancer with submucosal invasion and over 1 cm2 in size, were 57 with and 59 without lymph node metastasis. They were divided into two by the depth of invasion. The expressions of TS and TP in these group were compared with corresponding histopathological findings. Overall expressions of TS and TP were 54.3% and 34.5%, respectively. The TS-expression was not related with the depth of invasion and lymph node metastasis. The TP-expression, however, showed significant difference between with and without lymph node metastasis, and was so on the depth of submucosal invasion in the group without the nodal metastasis. Multivariate analysis showed that mucosal spread bordering 4 cm2 in size (p = 0.024) and lymphatic permeation (p = 0.099) in TS-expression, and lymph node metastasis (p = 0.041), submucosal invasion (p = 0.076) and venous permeation (p = 0.111) in TP-expression were the noticeable factors regarding to their high expression rates. Although these results were considered not to exceed gastric resection on the prognosis, they might be applicable as one of the indicators in postoperative follow-up on the minor resection of early gastric cancer such as EMR or local resection.     

Correlation between augmentative effects of leucovorin on 5-fluorouracil and thymidylate synthase inhibition

Background We studied the relationship between the augmentative effects of leucovorin (LV) on 5-fluorouracil (5-FU) and the inhibition rate of thymidylate synthase (TS) activity in Colon 26 murine colon carcinoma cells and L1210 murine leukemia cells. Methods In cytotoxic and TS inhibition studies, cells were exposed for 24 hours to varied concentrations of 5-FU alone or in combination with 1 or 10 μmol/L LV. Cytotoxicity was determined by colony forming efficiency or trypan blue dye exclusion, and TS inhibition rate was determined by [6-³H]-5-FdUMP binding assay. A growth inhibition curve was constructed for longer exposures. Results For tumor cell growth inhibition and cytotoxicity, the augmentative effect was observed when the lower 5-FU concentrations (0.1 μg/mL) were used. There was no difference in the effects between the low (1 μmol/L) and high (10 μmol/L) LV doses. Normally, TS enzyme levels rise acutely when cells are exposed to 5-FU. Both cell lines displayed an increase in TS after exposure to 5-FU. This exposure to 5-FU resulted in the maintenance of free enzyme. LV was able to increase the ternary complex and TS inhibition rate with little induction of TS, however, the inhibition rate was not dependent on doses of LV. Conclusions It was concluded that the augmentative effect of LV at a concentration of 0.1 μg/mL 5-FU occurred at the clinically achievable levels of 1 μmol/L of LV, and there was no difference in the effect between the low and high doses. TS was inhibited effectively by 5-FU and the addition of LV, without a marked induction of TS.     

胃癌组织中胸苷酸合成酶mRNA表达与预后的关系

目的探讨胃癌组织中胸苷酸合成酶（TS）mRNA表达水平与预后的关系。方法以G6PDH作内参照,采用实时定量RT-PCR技术检测41例胃癌组织TS mRNA表达水平。结果胃癌组织TS mRNA表达水平的中位数为0．93。TS高表达组（〉0．93）和低表达组（≤0．93）之间无瘤生存期和总生存期差异有显著性（P〈0．05）;而TS mRNA表达与年龄、性别、淋巴结转移、组织学分级及临床分期均无相关性（P〉0．05）。结论检测TS mRNA表达水平对判断胃癌病人预后有很好的预测价值。     

Effect of leucovorin on the antitumor efficacy of the 5-FU prodrug, tegafur-uracil, in human colorectal cancer xenografts with various expression levels of thymidylate synthase

The combination of oral tegafur-uracil (UFT) with leucovorin (LV) is used to treat patients with stage II to III colon cancer based on the results of postoperative randomized studies in which UFT/LV treatment showed an equivalent efficacy to intravenous 5-FU plus LV therapy. However, whether the addition of LV to UFT can elevate the antitumor activity of UFT in colorectal tumors with high expression levels of thymidylate synthase (TS), which affects 5-FU efficacy, remains to be clarified. This study investigated the effect of LV on the antitumor activity of UFT and/or 5-FU prodrugs in low folate diet-fed nude mice using human colorectal cancer xenografts with various expression levels of TS. The addition of LV to UFT resulted in a 55-79% inhibition of tumor growth among 11 types of colorectal tumor xenograft, whereas UFT alone showed 23-67% antitumor activity. Although there was an inverse relationship between the antitumor effect of UFT alone and UFT plus LV and tumoral TS activity, UFT plus LV appeared to have a more potent antitumor effect than UFT alone on colorectal tumors such as Co-3 and KM12C/5-FU with high expression levels of TS. This finding was confirmed by the significant positive correlation between the relative inhibition ratio of UFT/LV to UFT alone and TS levels in tumors. To investigate the reason for the higher efficacy of UFT/LV on colorectal cancer xenografts with high TS activity, intratumoral levels of reduced folates and a ternary complex of TS after oral UFT with or without LV were measured using Co-3 xenografts. Elevated levels of reduced folates and an increased ternary complex of TS in LV-treated tumors were noted. Our results indicate that a combined therapy of UFT with LV may contribute to the treatment of colorectal cancer patients with low and high expression levels of tumoral TS by increased formation of the ternary complex of TS leading to potentiated antitumor efficacy of UFT.     

The effect of dose and interval between 5-fluorouracil and leucovorin on the formation of thymidylate synthase ternary complex in human cancer cells.

We examined the importance of dosing interval between leucovorin (LCV) and 5-fluorouracil (5-FU) on intracellular thymidylate synthase (TS) ternary complex, free TS and total TS protein levels in human MCF-7 breast and NCI H630 colon cancer cell lines. A 2- to 3-fold increase in total TS was noted when either cell line was exposed to 5-FU 10 microM plus LCV (0.01-10 microM) compared with a 1.4- to 1.6-fold increase in total TS due to 5-FU 10 microM alone. The amount of TS ternary complex formed was 2- to 3-fold higher in both cell lines treated with the combination of 5-FU and LCV compared with 5-FU alone. TS complex formation and total TS protein increased with LCV dose (0.1-10 microM). In MCF-7 cells, the maximal increase in total TS protein and TS ternary complex formation was observed when 5-FU was delayed for 4 h after the start of LCV exposure. In NCI H630 cells, maximal total TS protein and ternary complex formation occurred when 5-FU was delayed for 18 h after the start of LCV exposure. The amount of free TS did not change in either cell line whether 5-FU was given concurrently with LCV or delayed for up to 24 h. The accumulation rate of intracellular folates in the form of higher glutamates Glu3-Glu5 was rapid in MCF-7 cells (maximal formation after 4 h), whereas in H630 cells accumulation of higher polyglutamates continued to increase up to 18 h. The time of peak folate polyglutamate (Glu3-Glu5) formation coincided with the time of peak TS complex formation and total TS protein in each cell line. In these human carcinoma cell lines, the LCV dose and interval between 5-FU and LCV play a role in increased TS total protein and TS ternary complex; however, the amount of free TS is independent of the interval between 5-FU and LCV. The time-and dose-dependent increases in TS ternary complex and TS total protein are associated with differences in the accumulation of folate polyglutamates in these cell lines.