Common DPYD Mutation Associated with 5-Fluorouracil Toxicity Detected by PCR-mediated Site-directed Mutagenesis

The human dihydropyrimidine dehydrogenase gene (DPYD) encodes dihydropyrimidine dehydrogenase (DPD; EC 1.3.1.2), the first and rate-limiting enzyme in the three-step pathway of uracil and thymine catabolism. DPD is also the principal enzyme involved in detoxification of pyrimidine-based antimetabolic analogs, such as 5-fluorouracil (5-FU), a drug that is commonly used in the treatment of solid tumors (colon, breast, head, neck, ovary, and skin). Because >80% of the administered 5-FU is degraded by DPD (1), the DPD catalytic activity in cancer patients could affect the efficacy of 5-FU treatment. In cancer patients with very low DPD activity, toxic reactions (e.g., diarrhea, stomatitis, mucositis, myelosuppression, and neurotoxicity) were reported that in some cases were life-threatening and sometimes fatal (2). A frequency as high as 3% of putative heterozygotes for DPD deficiency was also estimated based on catalytic activities in population studies (3)(4). The identification and characterization of the human DPD cDNA (5) made possible the identification and molecular analysis of mutations that affect DPD expression and catalytic activity. The most common mutation (6) associated with severe toxicity is a G→A transition at the 5′-splicing donor consensus sequence in intron 14 that leads to exon 14 skipping (7)(8)(9)(10): c.1905+1G→A [according to mutation nomenclature (11)]. By itself, the G-to-A nucleotide change destroys a unique restriction site only for the expensive MaeII endonuclease (isoschizomers, TaiI and TscI). We used PCR-mediated site-directed mutagenesis (PSM) utilizing a PCR primer with a single-base mismatch near the mutation site to introduce into the amplified wild-type product an allele-specific SnaBI restriction site. In this way, the amplification product encompassing this polymorphic site can be restriction-digested and electrophoresed to resolve alleles easily. Our strategy was such that the enzyme we used cut the amplified wild type once, but not the amplified homozygous mutant type. This choice provides in most cases a positive control for the SnaBI digestion. Using the Hardy-Weinberg equilibrium, the frequency of heterozygotes allows the estimation of up to 1 in 1000 homozygotes for DPYD mutations. We designed in the 3′ end of exon 14 of the DPYD gene the forward primer DPD-PSM1 (5′-CTAAAGGCTGACTTTCCAGACTAC-3′) to contain a single-base mismatch (A→T), creating a novel SnaBI restriction site (TAC↓GTA) in the amplified wild-type allele. DPD-PSM1 was designed based on the sequence from GenBank accession no. U20938. The reverse primer DPD-del-R (5′-CAGCAAAGCAACTGGCAGATTC-3′) was located in intron 14 (10). The amplification product length before digestion was 155 bp. Digestion by SnaBI restriction endonuclease generated two fragments (131 and 24 bp; the 24-bp fragment migrates quickly and is not seen on the gel) in the wild-type allele (Fig. 1 ) and does not cut the product from the homozygous mutant c.1905+1G→A allele (not shown). After electrophoresis in an agarose gel, a heterozygote for the mutation theoretically shows three bands of 155, 131, and 24 bp, which correspond to the two alleles (Fig. 1 ). Because 5-FU is one of the most commonly prescribed chemotherapeutic drugs in cancer treatment (in monotherapy or polytherapy) and the c.1905+1G→A mutation is frequently linked to severe toxicity, molecular screening of cancer patients could be done routinely, coupled with analysis of DPD activity in peripheral blood mononuclear cells, before the start of treatment to avoid the toxic effects of 5-FU. Because economic problems are very important in health-screening strategies, screening tests must be the least expensive. The use of SnaBI in a PSM method produces an 18-fold decrease in the enzyme cost ($0.70 vs $12.87 US per reaction) compared with the previous PCR-restriction method using MaeII (7)(8). Detection of c.1905+1G→A mutation by PSM. Reactions were carried out with 500 ng of genomic DNA in a total volume of 50 μL containing (final concentration) 10 mmol/L Tris-HCl (pH 8.3), 50 mmol/L KCl, 2.5 mmol/L MgCl2, 0.25 mmol/L of the four deoxynucleotide triphosphates, 1 U of Taq polymerase (Perkin-Elmer Cetus), and 0.25 μmol/L of each of the primers. PCR conditions were as follows: preliminary denaturation at 95 °C for 5 min, followed by 30 cycles of 30 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. The reaction was ended with 5 min at 72 °C. The digestion reactions contained 10 μL of PCR product, 2 μL of SnaBI (8 U; New England Biolabs), 2 μL of 1× NEBuffer 4 (supplied with the enzyme), and 100 mg/L bovine serum albumin, in a final volume of 20 μL. These components were incubated for 1 h at 37 °C. After the reaction ended, 10 μL of the PCR mixture was mixed with a loading buffer and then electrophoresed in a 4% agarose 1000® (Life Technologies) gel. Bands were made visible by ethidium bromide staining of the gel. Lane 1, HaeIII-digested pBR322 size marker; lane 2, amplification control; lane 3, heterozygote carrier for c.1905+1G→A, SnaBI-digested; lane 4, amplification control; lane 5, wild-type control, SnaBI-digested. This work was supported in part by a generous gift from Caisse des Professions Libérales, Province: CAMPLP. We thank Dr. Roch for his support. We are indebted to Paul Bennett for revision of our manuscript.