Involvement of gene polymorphisms of thymidylate synthase in gene expression, protein activity and anticancer drug cytotoxicity using the NCI-60 panel

https://doi.org/10.1016/j.ejca.2006.12.012Get rights and content

Abstract

A significant association has been established, in clinical studies, between the expression or activity of thymidylate synthase (TYMS) and the efficiency of fluorouracil. TYMS expression is partly under the dependence of gene polymorphisms in the 5′ and 3′ untranslated regions (UTR), but conflicting results have been obtained about their roles on fluorouracil efficiency. In this study, we wanted to use the National Cancer Institute (NCI) panel of 60 human tumour cell lines to clarify this problem. Three relevant polymorphisms of the TYMS gene were studied: (i) the 5′UTR tandem repeat of 28-bp (2R/3R polymorphism); (ii) the single nucleotide polymorphism (SNP) within the second repeat (3C/3G polymorphism); (iii) the 3′UTR 6-bp deletion (+6/−6 polymorphism). Allele frequencies were close to those expected in a Caucasian population (2R/3C/3G: 53/29/18%; +6/−6: 68/32%), but the proportion of heterozygous genotypes was lower than expected from allele frequencies. The 2R and 3G alleles were significantly associated with the +6 and the −6 alleles, respectively. There was a significant association between the presence of the 3G allele and TYMS mRNA expression and catalytic activity, particularly in p53-mutated cell lines. However, no significant correlation existed between fluorouracil cytotoxicity, as extracted from the NCI databases, and TYMS expression, activity or polymorphisms.

Introduction

Thymidylate synthase (TYMS) is the target enzyme of fluorouracil, an antimetabolite widely used in the treatment of colorectal, head-and-neck and breast cancers. It is a key enzyme for the synthesis of the DNA-specific nucleotide thymidylate, by catalysing the reductive methylation of deoxyuridylic acid into thymidylic acid in the presence of N5,N10-methylene tetrahydrofolate as a cofactor.1 Fluorouracil and other fluoropyrimidines, as well as other folate analogues such as raltitrexed, exert their cytotoxic action by depleting the thymidylate pool, leading to the activation of apoptotic pathways and to cell death.2

The intracellular level of TYMS has been long ago recognised as a determinant of fluorouracil cytotoxicity in vitro3, 4 and in vivo.5, 6, 7, 8 Response of colorectal cancers to infusional fluorouracil has been associated to low expression levels of thymidylate synthase by several groups, using different methods addressing mRNA, protein or enzyme activity, and in palliative as well as in adjuvant settings.9 The correlation between low thymidylate synthase expression and fluorouracil response is also displayed in other malignancies which benefit from fluorouracil therapy, such as pancreatic cancer and hepatocellular carcinoma.10, 11

The transcription of the TYMS gene does not involve TATA or CAAT boxes, but an 80% proportion of CG bases is present in the 5′ untranslated region (UTR) of the gene. Kaneda et al.12 have identified in this region a tandem repeat of a 28-bp sequence harbouring a CGCCGCG motif able to form secondary stem-loop structures which regulates the transcription level of the TYMS gene. In addition, Chu et al.13 have identified a specific site for the fixation of TYMS at the level of the 5′ UTR of its own mRNA, involving the repeated sequence and the translation initiation codon, which would play a major role in TYMS translational regulation.

Several polymorphisms are present in the TYMS gene promoter and interfere with the mechanism of regulation of TYMS expression. A first variation lies in the number of repeats of the 28 bp sequence, which is either present in duplicate (2R) or in triplicate (3R).14 In addition, a C > G transversion can occur in the second repeat when three repeats are present, giving rise in that case to two possibilities for the 3R allele: 3C or 3G.15 A polymorphism is also present at the level of the 3′UTR of TYMS, consisting of the presence or absence of a 6 bp sequence close to the polyadenylation signal.16

These polymorphisms have been shown to be associated with the expression of TYMS and, consequently, with the sensitivity to fluorouracil. The presence of three repeats, at the heterozygous or homozygous states, was first shown to be associated with higher protein expression and lower fluorouracil efficiency.17, 18 However, opposite results were found later in terms of patients’ survival after adjuvant fluorouracil treatment,19, 20 but in that case the DNA used to determine the gene variation originated from normal tissue or peripheric blood cells instead of tumour tissue. It was later suggested that only the 3G allele was associated with these features. The 2R/2R, 2R/3C and 3C/3C diplotypes were thus characterised by lower gene expression while the 2R/3G, 3C/3G and 3G/3G diplotypes by higher gene expression.1, 15 Concerning the 3′ UTR polymorphism, it was shown that the deletion of the 6 bp sequence was associated with low gene expression,22 which would lead to a better efficiency of fluorouracil,21 but, there again, opposite results were found by another group.20

Looking for a cellular model able to resolve these discrepancies and provide clear indications on the relationships between TYMS gene polymorphisms and fluorouracil activity, we wanted to see whether the NCI-60 panel could serve as a surrogate tool able to provide answers to this problem. The NCI-60 panel consists of 60 human tumour cell lines in culture and was initially established for high-throughput screening of natural products and synthetic molecules, on the basis of their antiproliferative properties.23 In addition, a number of molecular markers and gene expression profiles have been determined in the panel, allowing to establish relationships between chemo-sensitivity or -resistance and the molecular features of the cells.24

We have recently studied several gene polymorphisms in the NCI-60 panel and have established relationships with the cytotoxicity of anticancer drug families.25 Since the cell lines are tumoural, they can present many somatic genetic alterations, distinct from an actual constitutive polymorphism present in the patient who hosted the tumour. We have shown, however, that the NCI panel could represent a valuable model to study the role of gene variations on anticancer drug activity.26 In the present study, we have determined the three known polymorphisms of TYMS and tentatively related these polymorphisms to the TYMS mRNA expression and catalytic activity as well as to fluorouracil in vitro cytotoxicity.

Section snippets

Materials and methods

Frozen cell pellets from 59 of the 60 NCI cell lines of the panel were kindly provided by Dr. S. Holbeck, Cancer Therapeutic Branch, NCI, Bethesda. One cell line, MDA-N, is no longer available in the NCI panel.

Genomic DNA was extracted from cell pellets using QIAamp® DNA minikit from Qiagen. It was quantified by spectrophotometry. Polymerase chain reactions (PCR) were performed on genomic DNA using appropriate primers (see below). Polymorphisms were detected using restriction fragment length

Results

Fig. 1 shows some representative electrophoretic profiles of PCR products before and after digestion with the appropriate restriction enzymes. Table 1 lists the 59 cell lines studied and their genotypic status for the polymorphisms considered. No special trend appeared concerning the presence or absence of a given variant as a function of the tissue of origin of the cells. There were 23 cell lines with the 2R/2R genotype, 17 with the 2R/3R genotype and 19 with the 3R/3R genotype. Among the 36

Discussion

The distribution of the TYMS genotypes of the NCI-60 panel was in agreement with the distribution found in Caucasian populations: the 2R/3C/3G allele proportions were 48, 29 and 23%, respectively, in the study of Krajinovic et al.29 working on leukaemic Canadian children of European descent, and 50, 28 and 22% in the study of Graziano et al.30 in an Italian population, while we found 53, 29 and 18% in the NCI-60 panel. This distribution is different from the one observed in Japanese subjects by

Conflict of interest statement

None declared.

Acknowledgements

This study was supported by grants from the Ligue Nationale Française contre le Cancer, comités de la Dordogne, de la Charente et de la Charente Maritime. It is part of the Master 2 research project of NN. We are grateful to Pr. Jacques Bonnet and Dr. Hélène Jacquemin-Sablon for expert advice and to Mrs. Patricia Formento for teaching the technique of evaluation of the catalytic activity of TYMS.

References (40)

  • T. Yamachika et al.

    A new prognostic factor for colorectal carcinoma, thymidylate synthase, and its therapeutic significance

    Cancer

    (1998)
  • C. Aschele et al.

    Immunohistochemical quantitation of thymidylate synthase expression in colorectal cancer metastases predicts for clinical outcome to fluorouracil-based chemotherapy

    J Clin Oncol

    (1999)
  • S. Popat et al.

    Thymidylate synthase expression and prognosis in colorectal cancer: a systematic review and meta-analysis

    J Clin Oncol

    (2004)
  • K.H. Link et al.

    Thymidylate synthase quantitation and in vitro chemosensitivity testing predicts responses and survival of patients with isolated nonresectable liver tumors receiving hepatic arterial infusion chemotherapy

    Cancer

    (2000)
  • Y.C. Hu et al.

    Thymidylate synthase expression predicts the response to 5-fluorouracil-based adjuvant therapy in pancreatic cancer

    Clin Cancer Res

    (2003)
  • E. Chu et al.

    Autoregulation of human thymidylate synthase messenger RNA translation by thymidylate synthase

    Proc Natl Acad Sci USA

    (1991)
  • N. Horie et al.

    Functional analysis and DNA polymorphism of the tandemly repeated sequences in the 5’-terminal regulatory region of the human gene for thymidylate synthase

    Cell Struct Funct

    (1995)
  • K. Kawakami et al.

    Identification and functional analysis of single nucleotide polymorphism in the tandem repeat sequence of thymidylate synthase gene

    Cancer Res

    (2003)
  • C.M. Ulrich et al.

    Searching expressed sequence tag databases: discovery and confirmation of a common polymorphism in the thymidylate synthase gene

    Cancer Epidemiol Biomarkers Prev

    (2000)
  • K. Kawakami et al.

    Different lengths of a polymorphic repeat sequence in the thymidylate synthase gene affect translational efficiency but not its gene expression

    Clin Cancer Res

    (2001)
  • Cited by (0)

    View full text