| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Clinical |
Authors' Affiliation: Division of Clinical Pharmacology and Toxicology, Comprehensive Cancer Center, University of Alabama at Birmingham, Birmingham, Alabama
Requests for reprints: Robert Diasio, Division of Clinical Pharmacology and Toxicology, University of Alabama at Birmingham, 1824 6th Avenue South, Wallace Tumor Institute, Room 620, Birmingham, AL 35294-3300. Phone: 205-975-9770; Fax: 205-975-5650; E-mail: robert.diasio{at}ccc.uab.edu.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: DPD deficiency was identified phenotypically by both enzyme assay and uracil breath test, and genotypically by denaturing high-performance liquid chromatography. The methylation status was evaluated in PCR products (209 bp) of bisulfite-modified DPYD promoter, using a novel denaturing high-performance liquid chromatography method that distinguishes between methylated and unmethylated alleles. Clinical samples included five volunteers with normal DPD enzyme activity, five DPD-deficient volunteers, and five DPD-deficient cancer patients with a history of 5-FU toxicity.
Results: No evidence of methylation was detected in samples from volunteers with normal DPD. Methylation was detected in five of five DPD-deficient volunteers and in three of five of the DPD-deficient cancer patient samples. Of note, one of the two samples from patients with DPD-deficient cancer with no evidence of methylation had the mutation DPYD*2A, whereas the other had DPYD*13.
Discussion: Methylation of the DPYD promoter region is associated with down-regulation of DPD activity in clinical samples and should be considered as a potentially important regulatory mechanism of DPD activity and basis for 5-FU toxicity in cancer patients.
85% of administered 5-FU. Genotypic studies have identified >32 sequence variants in the DPYD gene (1, 3). Expression analysis of these variants showed that many were polymorphisms with no obvious functional significance (4), with the exception of a few mutations. One example commonly associated with DPD deficiency and severe toxicity to 5-FU is an intronic sequence variation (IVS14 + 1 G > A, DPYD*2A), which results in a truncated protein that lacks 55 amino acids due to the skipping of exon 14 (5). A second example is a less common mutation (1679T > G, I560S, DPYD*13), which is associated with decreased DPD activity and 5-FU toxicity (6) due to a nonconservative amino acid change from isoleucine to serine at codon 560 (I560), which is 100% conserved among human, mouse, rat, bovine, and pig species, suggesting its importance in maintaining DPD enzyme activity (7). Of note is that this mutation has not been detected in individuals with normal DPD enzyme activity. However, in a significant number of patients, many of the detected DPYD sequence variants did not explain the molecular basis of the observed DPD deficiency and 5-FU toxicity (4). Recent advances in our understanding of molecular mechanisms involved in the activation and degradation of 5-FU have led to an increased awareness of the potential importance of epigenetic factors in deciding the sensitivity of patients to anticancer drugs. We hypothesize that in the absence of inactivating mutations, methylation of CpG islands located in the 5' regulatory region of the DPYD gene promoter, may inhibit the binding of transcriptional factors or stabilize the chromatin structure, thereby directly inhibiting transcription (8). The current study investigates the role of aberrant methylation of the DPYD promoter as a potential epigenetic regulatory mechanism of DPD enzyme activity that may clarify the unexplained molecular basis of DPD deficiency and 5-FU toxicity.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Clinical samples. A total of 15 individuals were selected from a healthy population study and a separate study of 5-FU toxic patients using approved Institutional Review Board protocols, following informed consent from all individuals. Demographic data of studied individuals is shown in Table 1.
|
13C-Uracil breath test. The 13C-uracil breath test measured as DOB50 (DOB50 defined as the 13CO2/12CO2 ratio of breath sample measured at 50 minutes as determined by IR spectroscopy using the UBiTIR 300 instrument) was done as previously described (10). Individuals having a DOB50 <128.9 were classified as DPD-deficient (D), and those with a DOB50 >128.9 were classified as normal (N). Three patients DP2, DP3, and DP5 with deficient DPD enzyme activity exhibited life-threatening 5-FU toxicity and were not available to perform the 13C-uracil breath test.
DNA preparation, bisulfite modification, and PCR amplification. Genomic DNA was extracted from RKO cell line and from peripheral blood mononuclear cells of studied individuals using Wizard SV genomic DNA purification system (Promega, Madison, WI). All samples were bisulfite-modified (BSM) as previously described (11). PCR primers were designed according to the CpG island of the sense strand of the DPYD gene (Genbank accession no. NM_000110). These primers amplified the same sequence areas from methylated and unmethylated BSM DNA. The designed primers were forward, 5'-TTTTTGTTTGTAGGTTGGG-3'; and reverse, 5'-CAACCAAAAAACCAAATAACAACAA-3', which generates a 209 bp fragment of the DPYD promoter (nucleotides +44 to –165 from tsp). A 50 µL volume PCR was done using buffers purchased from Epicenter Technology (Madison, WI) as follows: fail-safe buffer (G) 25 µL, forward and reverse primers (10 nmol/L) 1 µL each, BSM gDNA (20 ng/µL) 1 µL Platinum Taq enzyme (Invitrogen, Carlsbad, CA) 0.45 and 21.55 µL H2O. The reactions were done on a PTC200 DNA engine (MJ Research, Reno, NV) as follows: denaturation at 95°C for 10 minutes followed by 40 cycles of denaturing at 94°C for 50 seconds, annealing at 52°C for 50 seconds, elongation at 72°C for 1 minute, and a final extension step at 72°C for 10 minutes. Samples were then maintained at 4°C. Screening for sequence variations in the DPYD gene was done using denaturing high-performance liquid chromatography (DHPLC) analysis of amplified PCR products of reference controls and clinical samples as described previously (12). All DPYD sequence variants identified by DHPLC were confirmed by DNA sequence analysis using a dideoxynucleotide chain termination method (Big Dye Kit; Applied Biosystems, Foster City, CA) and capillary electrophoresis on an ABI 310 Automated DNA Sequencer (Applied Biosystems).
Denaturing high-performance liquid chromatography method for the detection of methylation status and single nucleotide polymorphisms within bisulfite-modified DNA fragments. A DHPLC method was developed to detect the methylation status and single nucleotide polymorphisms within the PCR-amplified BSM fragments on the Wave System (Transgenomic, Co., Omaha, NE). The detection of multiple variables in a single injection, using DHPLC was previously reported by our laboratory (13). In this study, the DHPLC method was optimized using a 6.6 minute gradient at 0.9 mL/min flow rate. The gradient starts at 0.5 minutes and 51.8% buffer B (0.1 mol/L TEAA, 25% acetonitrile) and stops at 5 minutes and 60.8% buffer B. DNA loading on DNASep column starts at 46.8% buffer B. Optimal screening temperatures were experimentally determined by repeatedly injecting 5 µL of the PCR product using the same gradient at temperatures from 50°C to 70°C as previously described (12). PCR products of BSM fragments were injected at the experimentally determined optimal screening temperatures (50°C, 56°C, and 57°C) to detect the methylation status and single nucleotide polymorphisms in studied samples. Reagents for DHPLC analysis were purchased from Transgenomic.
Sequencing of bisulfite-modified samples. PCR products of BSM samples were gel-purified using the QIAquick gel extraction kit (Qiagen, Valencia, CA). Purified fragments were then cloned into pGEM T-easy vectors (Promega) and transfected into Escherichia coli JM109 competent cells (Promega). Cells were cultured on Luria-Bertani agar medium containing B-gal-isopropyl-L-thio-ß-D-galactopyranoside. Plasmid DNA was isolated from five different clones per sample using Isopure DNA Purification Kits Spin Column Plasmid Mini-Preps (Denville Scientific, Inc., Metuchen, NJ). Isolated plasmid DNA was then sequenced on an ABI Prism 310 genetic analyzer using Big Dye terminator and M13 primers.
Statistical analysis. Statistical analysis was done using SPSS Software (version 10.0.5). Means of DPD enzyme activity in studied groups were compared using one-way ANOVA test. The association between DPD enzyme deficiency and methylation was examined using Pearson 
2 test.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Location of CpG sites in bisulfite-modified control samples. The DPYD promoter fragment examined in this study spanned 209 bp on the 5'-untranslated region starting at nucleotide +44 to –166 from the transcription start point. This region included 27 CpG sites, 11 of which lie within the sequence of two regulatory elements at nucleotides (–23 to –42) and (–51 to –72) which were previously reported by our laboratory to regulate DPD mRNA expression (14).
Methylation of the DPYD promoter. The DHPLC instrument automatically calculates the area under the curve (AUC) in the chromatogram, designating %AUC for each peak. In case of a methylated fragment, under the specified temperature and gradient conditions, the methylated allele (GC-rich) is retained longer on the column than the nonmethylated (A/T-rich) allele. In case of the BSM universal methylated control (all 27 CpG sites are methylated), the %AUC for the methylated allele (%MAUC) was 44.3 % (Fig. 1C). By calculation, it is possible to estimate the approximate number of methylated CpG sites in a fragment relative to the control sample, which have 27 of the 27 methylated CpG sites (Fig. 2; represented by 43% AUC), using a factor of 0.609 (27 of 44.3 or 0.609). This factor is obtained from the equation: [27 (known number of methylated CpG sites in control) x %AUC of test methylated fragment = number of CpG sites in test fragment x 44.3%].
|
|
14 CpG). The methylation status of an allele is defined by its %MAUC, retention time, elution pattern, and melting profile as well as its resolution under optimal screening temperatures using the optimized gradient.
Examining the association between enzyme deficiency and methylation, regardless of group assignment (using Pearson 2), illustrated that 80% (8 of 10) of the DPD-deficient individuals exhibited a positive methylation signal, whereas individuals with normal DPD enzyme activity (n = 5) tested negative for methylation (2 = 7.78; P < 0.05).
Denaturing high-performance liquid chromatography analysis of control reference samples. Universal methylated sperm DNA, steady state RKO cell line (hypermethylated), and the demethylated 5-azacytidine–treated RKO (negative control) showed a single peak pattern under nondenaturing temperature (50°C; Fig. 1A, D, and G). Under partially denaturing temperatures (56°C), chromatograms of both positive methylated controls (universal methylated sperm DNA and hypermethylated RKO cell line) slightly resolved into two peaks (Fig. 1B and E). These peaks further resolved at 57°C into distinct patterns characteristic of methylated samples. Samples with a positive methylation signal illustrated the presence of two alleles, a methylated allele with higher retention time due to high GC content, and an unmethylated allele, rich in A/T nucleotides, eluting earlier on the chromatogram (Fig. 1C and F). The negative control sample (demethylated RKO cell line) retained a single peak pattern at all temperatures (Fig. 1G, H, and I).
For confirmation of the DHPLC results, control samples were cloned and five colonies were sequenced for each sample (Fig. 2). Sequence analysis illustrated that two of five and one of five colonies of the universal methylated sperm DNA and RKO hypermethylated control samples, respectively, had 100% methylation (27 of 27 CpG sites). This finding could be attributed to different growth phases of the cells in culture. Bisulfite sequence analysis of five of the five colonies of the negative control sample illustrated the absence of methylation in all 27 CpG sites (Fig. 2).
Denaturing high-performance liquid chromatography analysis of volunteers with normal dihydropyrimidine dehydrogenase enzyme activity. Volunteers with normal DPD enzyme activity produced chromatogram patterns similar to those of the reference negative control sample, demonstrating a single peak eluting at 4.5 minutes in all samples (Fig. 3A-E). Bisulfite sequence analysis confirmed the absence of methylation in five of the five cloned samples (0 of 27 CpG sites; Fig. 4).
|
|
Denaturing high-performance liquid chromatography analysis of cancer patients with deficient dihydropyrimidine dehydrogenase enzyme activity. Chromatograms of cancer patients DP5 and DP17 with DPD deficiency and 5-FU toxicity showed a positive methylation signal of 18% MAUC (11 CpG methylated sites) and 19% MAUC (12 CpG methylated sites), respectively (Fig. 3M and N). In contrast, no methylation signal was detected in patients DP2 and DP20; however, the mutations DPYD*13 and DPYD*2A were detected in their DPYD gene, respectively (Fig. 3K and O). The DHPLC results were further confirmed by the absence of methylation in all CpG sites (0 of 27) in five of five clones sequenced for these two patients. In patient DP3, 13% MAUC was detected, which represents approximately eight methylated CpG sites. Additionally, the mutation DPYD*2A was detected in this patient's DPYD gene (Fig. 3L).
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, methylation has been recognized as a common epigenetic alteration in human cancer which leads to gene silencing (17). Studies in cancer cell lines (oral epidermoid carcinoma line KB, colon adenocarcinoma COLO20, oral squamous cell carcinoma lines HSC3, HSC4, and Ca9-22 and hepatoma HepG2 cell lines) showed the absence of genetic alterations in DPYD promoter region with full activity (8). Our results show the absence of DNA sequence variants in the DPYD promoter region in all studied individuals. Transcriptional studies examining steady state expression of DPD mRNA in cell lines showed variable levels of DPD expression (8). Experiments investigating endogenous promoter activity indicated that promoter methylation could down-regulate DPD expression (8, 18). Following treatment of these cell lines with 5-azacytidine, DPD mRNA levels increased significantly, implying that methylation could be a regulatory mechanism for DPD expression. Bisulfite sequence analysis of the DPYD promoter region revealed the presence of different methylation patterns in CpG islands of these cell lines (8). However, aberrant methylation of the DPYD promoter has not been investigated in cancer patients with unexplained DPD deficiency and 5-FU toxicity.
Previously, methylation detection has been hampered by technical limitations. These limitations include artifacts of bisulfite modification reactions (19); labor-intensive and time-consuming cloning and sequencing steps required for bisulfite genomic sequencing; as well as substantial amounts of high molecular weight DNA required in Southern blotting. Also, methods based on differential methylation states of alleles such as methylation-specific PCR (11), MethyLight (20), or methylation-sensitive single-stranded conformational polymorphisms (21, 22) are limited by the specific nucleotide variation located at the primer/probe binding sequence (23).
The DHPLC method developed in this study, unlike other techniques, circumvents the cloning and sequencing steps required for methylation detection. It also correctly differentiates between methylated and unmethylated alleles with high sensitivity and rapidity (6.6 minutes) and permits a semiquantitative assessment of the methylation status, allowing approximation of the number of methylated CpG sites relative to a 100% methylated control reference sample. Additionally, this method permits the detection of single nucleotide polymorphisms within fragments screened for methylation. These advantages allow the use of this DHPLC method in large population studies investigating the effect of aberrant methylation in cancer patients.
In the current study, individuals with normal DPD enzyme activity and normal 13C-uracil breath test showed single peak patterns (Fig. 3A-E) suggestive of unmethylated status (Figs. 1G and 4). Normal individuals and cancer patients with deficient DPD enzyme activity and altered 13C-uracil breath test results showed the presence of variable methylation pattern and %MAUC (5-24.7%; approximately 3 to 15 methylated CpG sites; Fig. 3F-J and Fig. 4). The lowest MAUC (5% MAUC) was detected in volunteer DVD2 (Fig. 3J) whose genotype also illustrated the presence of the mutation IVS14 + 1G > A, DPYD*2A. Two novel mutations 542A > C, K182T and 557A > G, Y186C were detected in two DPD-deficient volunteers DV75 and DV 130, respectively; their functional significance is currently being investigated in our laboratory. The DHPLC chromatograms of the two patients with colorectal cancer, DP5 and DP17, showed the presence of a strong methylation signal, 17.9% MAUC (11 CpG methylated sites) and 19.2% MAUC (12 CpG methylated sites), respectively (Fig. 3M and N). Interestingly, these two patients did not have inactivating mutations in their DPYD gene that could explain the molecular basis of DPD deficiency.
Taken collectively, a significant association between aberrant methylation of the DPYD promoter and DPD enzyme deficiency was detected in 80% (8 of 10) of DPD-deficient individuals, whereas all individuals with normal DPD enzyme activity (n = 5), tested negative for methylation (2 = 7.78; P < 0.05). Additionally, aberrant methylation of the DPYD promoter was observed in 100% of DPD-deficient individuals without inactivating mutations in their DPYD gene (n = 6; Table 1; Figs. 3 and 4). It should be noted that in some cases, the presence of genetic variations in the DPYD gene (Table 1), whether intronic (e.g., DPYD*2A) or in the coding region (e.g., DPYD*13) were accompanied by variable methylation patterns. Whereas methylation of the two regulatory elements detected in the DPYD promoter (ref. 14; Fig. 4) was observed in DPD-deficient individuals with wild-type DPYD (DPYD*1). The positive methylation signals detected by DHPLC were associated with a variable number of methylated CpG sites and variable %MAUC (Fig. 3F-J and L-N). Thus, genetic and/or epigenetic molecular mechanisms can act separately (Fig. 3F-I, K and M-O) or in concert (Fig. 3J and L) to down-regulate DPD enzyme activity.
However, it is important to emphasize that down-regulation of DPD enzyme activity by methylation is not an isolated cellular mechanism and that transcriptional silencing by methylation is a consequence of multiple mechanisms that requires the combined action of histone hypoacetylation, histone methylation, and methyl-binding proteins (24). Whether suppression of DPD expression by methylation is site-specific or pattern-specific (25) remains to be examined in a larger population of patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 7/13/05; revised 8/30/05; accepted 9/13/05.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Schwab, U. M. Zanger, C. Marx, E. Schaeffeler, K. Klein, J. Dippon, R. Kerb, J. Blievernicht, J. Fischer, U. Hofmann, et al. Role of Genetic and Nongenetic Factors for Fluorouracil Treatment-Related Severe Toxicity: A Prospective Clinical Trial by the German 5-FU Toxicity Study Group J. Clin. Oncol., May 1, 2008; 26(13): 2131 - 2138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. H. Ezzeldin and R. B. Diasio Predicting Fluorouracil Toxicity: Can We Finally Do It? J. Clin. Oncol., May 1, 2008; 26(13): 2080 - 2082. [Full Text] [PDF]
|
|
|
|
 |
|
 R. B Diasio Overview of Cancer Pharmacogenomics Am. Assoc. Cancer Res. Educ. Book, April 12, 2008; 2008(1): 461 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. H. Ezzeldin and R. B. Diasio Genetic testing in cancer therapeutics. Clin. Cancer Res., July 15, 2006; 12(14): 4137 - 4141. [Full Text] [PDF]
|
|
|
|
|
|
J. Yu, H. L. McLeod, H. H. Ezzeldin, and R. B. Diasio Methylation of the DPYD Promoter and Dihydropyrimidine Dehydrogenase Deficiency. Clin. Cancer Res., June 15, 2006; 12(12): 3864 - 3864. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
), whereas methylated cytosines were protected (
). Numbers located above the first row, nucleotide (nt) position of cytosine residue relative to the transcription start point. CpG sites located within regulatory elements are boxed in two rectangles.
