Clinical Cancer Research Meeting Calendar Metabolism
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gagnon, J.-F.
Right arrow Articles by Guillemette, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gagnon, J.-F.
Right arrow Articles by Guillemette, C.
Clinical Cancer Research Vol. 12, 1850-1858, March 2006
© 2006 American Association for Cancer Research

Cancer Therapy: Preclinical

Irinotecan Inactivation Is Modulated by Epigenetic Silencing of UGT1A1 in Colon Cancer

Jean-François Gagnon2, Olivier Bernard2, Lyne Villeneuve2, Bernard Têtu3 and Chantal Guillemette1,2

Authors' Affiliations: 1 Canada Research Chair in Pharmacogenomics and 2 Oncology and Molecular Endocrinology Research Center, Centre Hospitalier de l'Université Laval Research Center, Faculty of Pharmacy and 3 Department of Pathology, Hôtel-Dieu de Québec, Laval University, Quebec, Canada

Requests for reprints: Chantal Guillemette, Canada Research Chair in Pharmacogenomics, Pharmacogenomics Laboratory, Centre de Recherche du Centre Hospitalier de l'Université Laval Research Center, T3-67, 2705 Boul. Laurier, T3-48 Quebec, Canada G1V 4G2. Phone: 418-654-2296; Fax: 418-654-2761; E-mail: chantal.guillemette{at}crchul.ulaval.ca.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: Irinotecan is used in the first-line treatment of metastatic colorectal cancer. The UGT1A1-metabolizing enzyme, expressed in liver and colon, is primarily involved in the inactivation of its active metabolite 7-ethyl-10-hydroxycamptothecin (SN-38). Herein, we explored the role of DNA methylation in the silencing of UGT1A1 gene expression in colon cancer and its influence on cellular SN-38 detoxification.

Experimental Design and Results: UGT1A1 mRNA was repressed in most primary tumors (41 of 50; 82%) and in three colon cancer cell lines (HCT-116, HCT-15, and COLO-320DM). Bisulfite sequencing of the UGT1A1 gene revealed the aberrant methylation of specific CpG islands in UGT1A1-negative cells. Conversely, hypomethylation was observed in HT-29, HT-115, and LOVO cells that overexpress UGT1A1. Direct methylation of the UGT1A1 promoter resulted in the complete repression of transcriptional activity. Treatment with demethylating and histone deacetylase inhibitor agents had the capacity to reverse aberrant hypermethylation and to restore UGT1A1 expression in hypermethylated UGT1A1-negative cells but not in hypomethylated cells. Loss of UGT1A1 methylation was further associated with an increase in UGT1A1 protein content and with an enhanced inactivation of SN-38 by 300% in HCT-116 cells.

Conclusions: We conclude that DNA methylation represses UGT1A1 expression in colon cancer and that this process may contribute to the level of tumoral inactivation of the anticancer agent SN-38 and potentially influence clinical response.

Colorectal cancer is the third leading cause of cancer-related deaths worldwide. Of patients suffering from colorectal cancer, up to 60% already have overt metastases at the time of diagnosis (1). Irinotecan is the anticancer drug used in first-line treatment of metastatic colorectal cancer (2–5). However, there is marked variability in response to therapy as well as frequency and severity of toxicities. Its active metabolite, 7-ethyl-10-hydroxycamptothecin (SN-38), is primarily inactivated through glucuronidation, mediated mainly by the UDP-glucuronosyltransferase UGT1A1 enzyme (6–8). The UGT1A1 gene encodes an enzyme that acts as a drug detoxifier by catalyzing conjugation reactions with glucuronic acid. Interindividual differences in the inactivating pathway of SN-38 caused by common genetic polymorphisms in the glucuronidation pathway have been shown as major determinants for irinotecan toxicity and clinical outcome (9–13). In particular, the UGT1A1*28 promoter allele, leading to a lower expression of the gene, was identified as clinically valuable for predicting severe irinotecan toxicity in cancer patients (9, 10, 12).

Another factor that needs to be considered when regarding a patient's response to an anticancer drug is the intrinsic susceptibility of cancer cells defined by the expression of an array of genes that determine local tumor drug concentration. This includes activities of metabolic enzymes, such as UGT1A1, which likely establish local tissue drug concentration. It has been shown that the metabolic capacities of tumor cells influence response outcome, such as the local formation of SN-38 from CPT-11 by carboxylesterases (14). Recently, glucuronidation has been identified as a mechanism of intrinsic drug resistance to SN-38 in human colon cancer cells (15–17). In addition, clinical studies support the hypothesis that the individual glucuronidation capacity contributes to tumoral response, likely through the modulation of both plasma and tumoral concentrations of SN-38 (11, 13). UGT1A1 is expressed with a large variability in primary colon tumors and during colon tumorigenesis, suggesting that the level of UGT1A1 expression may contribute to the differential chemosensitivity of colon tumors (18–20). Although the mechanisms involved in the variability of UGT1A1 expression in colon tumors remain unknown, recent observations support that mechanisms other than genetic polymorphisms may contribute to the differential expression of the UGT1A1 metabolic pathway in colon tumors.

It is well documented that colon cancer arises from a series of genetic alterations that includes a variety of mechanisms of gene silencing, such as point mutations, loss of heterozygosity, and homozygous deletions. In addition to these genetic abnormalities, promoter DNA hypermethylation is an additional mechanism that plays a critical role in the progression of colorectal cancer. It has been shown that sporadic colorectal cancers are often associated with an abnormal methylation of CpG-rich sites (CpG islands) in promoter regions of multiple loci, including genes involved in the cell cycle, growth regulation, DNA repair, metabolism, and apoptosis, whose silencing is a common and early event in human colon neoplasia mediated by epigenetic occurrences (21–24). Cytosine methylation results in transcriptional repression either by interfering with transcription factor binding or by inducing a repressive chromatin structure (25). While exploring the mechanisms involved in the variable expression of UGT1A1 in colon tumors, we hypothesized that, in addition to genetic polymorphisms, such as the common UGT1A1*28 allele, silencing of the glucuronidation pathway by epigenetic mechanisms could have the potential to induce various effects at the systemic and tissue levels, including an enhanced intratumoral exposure to SN-38. Previous studies support that therapeutic efficacy of anticancer drugs can be predicted by epigenetic modifications (26–29). A notable example is the MGMT gene, responsible for repairing DNA damage caused by alkylating chemotherapeutic drugs (30). The extent of promoter methylation of the MGMT gene predicts a favorable response to antineoplastic therapy in cancer patients (31).

The mechanisms underlying the inactivation of UGT1A1 expression in a large proportion of colon tumors and the extent of its role in determining tumoral response to anticancer drugs remain unknown (18, 19, 32–34). Our study was thus designed to assess if promoter hypermethylation arises as a mechanism that modulates UGT1A1 expression in colon cancer cells and, consequently, that could contribute to determine SN-38 inactivation in colon tumors. Results indicate that epigenetic modifications in the UGT1A1 gene occur in colon tumor cell lines and that the consequence of UGT1A1 repression is a modulation of anticancer drug exposure in cancer cells.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Colon cell lines and primary colon tumor samples. Colon cancer cells HT-29, HCT-116, COLO-320DM, HCT-15, and LOVO were obtained from American Type Culture Collection (Manassas, VA). HT-115 cells were obtained from European Collection of Cell Culture (Salisbury, United Kingdom). All cell lines were growth in the medium recommended by American Type Culture Collection and European Collection of Cell Culture. HCT-116 and HT-29 cells were kept in McCoy's 5A medium with 1.0 mmol/L sodium pyruvate (Sigma, Oakville, Ontario, Canada). HT-115 cells were kept in DMEM with 2 mmol/L L-glutamine (all from Wisent, St-Bruno, Quebec, Canada). LOVO cells were kept in F-12K medium with 2 mmol/L L-glutamine. COLO-320DM and HCT-15 were kept in a modified RPMI 1640 to obtain a final concentration of 2 mmol/L L-glutamine, 1.5 g/L sodium bicarbonate, 4.5 g/L glucose, 10 mmol/L HEPES, and 1.0 mmol/L sodium pyruvate. All medium were added with 100 IU/mL penicillin, 50 µg/mL streptomycin, and 10% fetal bovine serum, except for the HT-115 cell line, which required 15% fetal bovine serum (Wisent). All cells were incubated at 37°C in a humidified incubator with 5% CO2 as specified by the supplier.

Primary tumor specimens were obtained from 50 patients who underwent surgery at Hôtel-Dieu de Québec. All patients provided an informed consent and the project was approved by the institutional human research review board. All specimens were quick frozen in liquid nitrogen and stored at –80°C until processing.

Reexpression of UGT1A1 by 2'-deoxy-5-azacytidine and trichostatin A treatment of colon carcinoma cell lines. 2'-Deoxy-5-azacytidine (5-Aza-dC; Sigma) and trichostatin A (New England Biolabs, Beverly, MA) were resuspended in sterile 1x PBS and 100% DMSO, respectively. Treatment with both reagents resulted in final concentrations of 0.1% DMSO or PBS in the culture medium. The treatment groups consisted of trichostatin A alone, 5-Aza-dC alone, or combined trichostatin A and 5-Aza-dC. The control group consisted of medium with 0.1% PBS for the groups treated with 5-Aza-dC alone and medium with 0.1% DMSO for groups treated with trichostatin A alone or in combination with 5-Aza-dC. Cells were seeded 24 hours before treatment in 100-mm2 dishes to obtain 95% confluence of the control cells after 4 days of culture. For the 5-Aza-dC-treated groups, cells were cultured with 5 µmol/L 5-Aza-dC for 72 hours, whereas trichostatin A–treated groups were incubated for 72 hours in culture medium before adding 300 nmol/L trichostatin A for 16 hours. The groups treated with both agents were incubated with 5 µmol/L 5-Aza-dC for 72 hours followed by trichostatin A treatment (300 nmol/L) for 16 hours. All treatments with 5-Aza-dC and trichostatin A were done as described previously (35).

Expression analysis of UGT1A1 by quantitative real-time PCR. RNA from primary tumor samples and cancer cell lines was extracted with Tri-Reagent (Molecular Research Center, Inc., Cincinnati, OH) as described in the manufacturer's protocol. RNA was resuspended in RNAsecure water (Ambion, Austin, TX) and concentrations were determined by spectrophotometric reading. RNA (1 µg) was converted to cDNA with SuperScript II RNase H– (Invitrogen, Burlington, Ontario, Canada) using the manufacturer's protocol. Relative expression of UGT1A1 was measured by quantitative real-time PCR (QRT-PCR) using the ABI Prism 7000 Sequence Detector. Analyses were done in 25 µL reaction volumes using 1 µL of cDNA template, 12.5 µL of 2x Taqman Universal PCR Master Mix (Applied Biosystems, Foster City, CA), 850 nmol/L of each primer, and 200 nmol/L UGT1A1 probe. The primers and probes for UGT1A1 were designed with the Primer Express Software (Applied Biosystems). All primer sequences are shown in Table 1 and were designed to be exon-exon so that none amplified genomic DNA (data not shown). PCR amplification began with a 95°C, 10-minute step to activate the AmpliTaq Gold DNA polymerase followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute. The housekeeping genes used were eukaryotic endogenous 18S rRNA and human endogenous ß2-microglobulin control (Applied Biosystems) quantified according to the manufacturer's protocols. Relative gene expression data were calculated using the {Delta}({Delta})Ct method described previously by K. Livak (PE-ABI; Sequence Detector User Bulletin 2). This method uses a single sample, termed calibrator sample, for comparison with samples of unknown gene expression level. The calibrator sample is analyzed on every assay plate with the unknown samples of interest. Results are obtained with the following formula: fold induction = 2–[{Delta}({Delta})Ct]; {Delta}({Delta})Ct is defined as [Ct gi (unknown sample) – Ct ß2-microglobulin (unknown sample)] – [Ct gi (calibrator sample) – Ct ß2-microglobulin (calibrator sample)], where gi is the gene of interest. The calibrator sample can be any sample chosen to represent 1-fold expression of the gene of interest (in this study, untreated cells).


View this table:
[in this window]
[in a new window]
  		Table 1. Primers sequences and probes

 
Transient transfection of UGT1A1 promoter constructs in colon cell lines. Variable-length sequences of the UGT1A1 promoter region (269, 540, 1,316, and 2,218 bp) were amplified from genomic DNA using the primers listed in Table 1 and were directionally cloned into the MluI/XhoI sites for 540-, 1,316-, and 2,218-bp constructs and the SacI/XhoI sites for 269-bp constructs of the pGL3 basic vector (Promega, Madison, WI). All constructs were sequenced before functional analysis. Each plasmid was methylated and mock methylated (without S-adenosylmethionine) with SssI methylase (New England Biolabs) using the manufacturer's protocol. Confirmation of complete methylation was assessed by digestion with the HpaII restriction enzyme that cuts only the unmethylated CCGG restriction sites in the plasmid. Migration on agarose gel confirmed that methylated plasmids remain undigested after HpaII digestion in contrast to unmethylated plasmids that were digested in multiple fragments depending on the length of the promoter regions. This method has been successfully done in previous studies (36–38). DNA was purified with the Qiaex II kit (Qiagen, Hilden, Germany) and resuspended in Tris-HCl (pH 7.9) for subsequent transfection. Cells were plated in 24-well microplates at 70% confluence in growth medium. For each well, 0.7 µg reporter construct was cotransfected with 0.1 µg Renilla luciferase plasmid and 3 µL Exgen-500 (Promega) into HCT-116, HCT-15, COLO-320DM, LOVO, and HT-115 cells using Exgen-500 transfection reagent (MBI Fermentas, Burlington, Ontario, Canada) according to the manufacturer's protocol. For the cell line HT-29, transfections were done with Lipofectamine 2000 (Invitrogen) as described by the manufacturer. Briefly, cells were plated in 24-well microplates at 90% confluence in growth medium. Reporter construct (0.6 µg) was cotransfected with 0.2 µg Renilla luciferase plasmid (Promega) using 3 µL Lipofectamine 2000 per well. Cells were harvested 24 hours after transfection and assayed for promoter activity by use of the Dual-Luciferase Reporter Assay System according to the manufacturer's recommendations (Promega). Luciferase activity was measured by using 40 µL of cell lysate in a 96-well plate on an LB96V microplate luminometer (EG&G Berthold, Bad Wildbad, Germany).

Sodium bisulfite genomic DNA modification and DNA sequencing. To analyze CpG methylation of the UGT1A1 5' region, genomic DNA was extracted from colon cell lines with Tri-Reagent as described in the manufacturer's protocol. The bisulfite reaction, converting all unmethylated but not methylated cytosines to uracil, was done as follows. DNA (1 µg) was modified by sodium bisulfite using the CpGenome DNA Modification kit (Intergen, Purchase, NY) according to the manufacturer's instructions. Amplification primers used for bisulfite sequencing are described in Table 1. All primers used were designed with the demo version of the Primer Premiere software (Premier Biosoft, Palo Alto, CA), which predicts primer hairpins, dimers, and false priming and consequently allows very high primer quality. Nested primers were used for sequencing whenever amplification primers did not allow good readings. Sequences obtained using such primers had excellent quality and low background. Modified DNA (100 ng) was amplified with the following conditions for each primer pairs: 95°C, 10-minute step to activate the AmpliTaq Gold DNA polymerase followed by 40 cycles at 95°C for 30 seconds, 58°C for 30 seconds, and 72°C for 1 minute. PCR products were purified with the QIAquick PCR Purification kit (Qiagen) and eluted in 30 µL nuclease-free water. DNA cycle sequencing was carried out in 20 µL reactions on a Perkin-Elmer GenAmp PCR System 9700 (Perkin-Elmer, Foster City, CA) using 6 µL purified PCR product, 3.2 pmol primer, and the ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction kit (Perkin-Elmer) as described by the manufacturer. The methylation percentage corresponds to the fluorescence peak area values of cytosine divided by the addition of the fluorescence peak area values of cytosine and thymidine. Methylation status was calculated as the mean of the methylation percentages of CpG for each CpG region and includes all CpG in that region. The same procedure was used in previous studies (39, 40).

Western blot analyses. Western blot analyses of UGT1A1 protein were done on microsomal preparations isolated from the HCT-116 cell line after treatment with 5-Aza-dC alone or in combination with trichostatin A as described previously (41). Commercial microsomal preparations of liver and ileum (n = 5 subjects pools; Tissue Transformation Technologies, Edison, NJ), in addition to a preparation of HEK293-UGT1A1 (BD Gentest, Woburn, MA), were used as comparative models. Briefly, 90 µg of microsomal proteins were separated by 10% SDS-PAGE. The separated proteins were transferred onto nitrocellulose membranes and probed with a specific anti-human UGT1A1 antiserum (1:2,500 dilution) as described previously (42).

SN-38 glucuronidation assays. SN-38 was generated from irinotecan HCl (McKesson, ON, CA) as follows: 1.25 mL NaOH (6 mol/L) was added to 2.5 mL irinotecan HCl (20 mg/mL) and incubated overnight at 55°C in the dark. The solution was neutralized with 1.25 mL HCl (6 mol/L), and 150 µL HCl (6 mol/L) was added to obtain pH ≤3. The solution was incubated for 3 hours at room temperature in the dark. The pH was readjusted to 3 with 1 mol/L NaOH and the solution was centrifuged at 3,000 rpm for 10 minutes. The supernatant was removed and the pellet was washed twice with 2.5 mL ultrapure water. The product was frozen in a bath of dry ice/ethanol and placed in a lyophilizator. The product was certified by high-performance liquid chromatography with mass spectrometry and UV detection analysis for purity ≥97%. SN-38 glucuronide was produced from enzymatic assays with human liver microsomes, isolated, and quantified on a SN-38 calibration curve. The enzymatic assays with human colon cancer cell lines were done as described previously (41). Briefly, microsomal fractions (120 µg) were added to the reaction mixture containing alamethicin (60 µg/mg protein) and 10 µmol/L SN-38 in a final volume of 100 µL. The assays were incubated at 37°C in a shaking water bath for 4 hours. The reaction was stopped in 200 µL methanol plus 1% HCl (2 N) followed by centrifugation at 14,000 x g for 10 minutes. The supernatant was evaporated under nitrogen at 35°C and resuspended in a solution of methanol and water (1:1) and 0.5% HCl (2.0 mol/L). Detection of SN-38 and SN-38 glucuronide was done by high-performance liquid chromatography coupled with fluorescence detection as described previously (41).

Statistical analyses. All statistical analyses were done using the JMP V4.0.2 software (SAS Institute, Cary, NC). All P values shown were obtained with rank sums of the Wilcoxon/Krustal-Wallis test.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
UGT1A1 expression in primary colon tumors and colon cancer cell lines. To confirm previous reports of variable expression of UGT1A1 in colon tumor samples, its expression was analyzed by QRT-PCR in 50 primary colon tumor samples. Results showed a coefficient of variability of 350% with two groups of baseline levels of UGT1A1 mRNA; tumors with high expression levels (18%; up to 12,112 arbitrary units) and those with undetectable or very low UGT1A1 mRNA content [82%; expression level <400 arbitrary units (1-382); data not shown]. With the objective to study cellular models that represent this array of UGT1A1 expression, we analyzed six human colon carcinoma cell lines. Of the cell lines investigated, half did not express UGT1A1. At baseline, expression of the UGT1A1 mRNA transcript was undetectable in HCT-116 and COLO-320DM (1 relative unit) and very low in HCT-15 (108 relative units). In contrast, HT-29, LOVO, and HT-115 cell lines were associated with high baseline expression of UGT1A1 (Fig. 1 ). The role of methylation in the silencing of UGT1A1 gene expression in cell lines was then investigated.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Relative expression of UGT1A1 mRNA in colon cancer cell lines. Columns, mean of relative expression of UGT1A1 (arbitrary units) determined by QRT-PCR. The housekeeping gene was used as an endogenous control and showed a coefficient of variation of 15%.

 
Effects of 5-Aza-dC and trichostatin A treatment on UGT1A1 expression in colon cancer cell lines. To study the potential role of methylation in the silencing of UGT1A1 gene expression, cell lines were treated with 5-Aza-dC and/or trichostatin A, demethylating and histone deacetylase inhibitor agents, respectively. QRT-PCR analysis revealed, on treatment with 5-Aza-dC alone, a significant increase in UGT1A1 mRNA (up to 19-fold) for cells with a very low expression of the gene (HCT-116, COLO-320DM, and HCT-15; Fig. 2 , Table 2 ). The combination of 5-Aza-dC and trichostatin A treatment increased the expression of UGT1A1 by 13-fold for the HCT-15 cell line and up to 75- to 128-fold for the COLO-320DM and HCT-116 cell lines, respectively (Fig. 2), compared with a modest increase (up to 3-fold) for cells with high basal UGT1A1 expression (data not shown). For all cell lines, trichostatin A alone failed to increase significantly UGT1A1 expression. These results suggest that demethylation by 5-Aza-dC restored the expression of the UGT1A1 gene, supporting that methylation is a mechanism affecting the transcription of UGT1A1 mRNA in various colon cancer cell lines with very low basal expression of this gene.


Figure 2
View larger version (10K):
[in this window]
[in a new window]
  		Fig. 2. UGT1A1 expression following 5-Aza-dC (5 µmol/L) and/or trichostatin A (TSA; 300 nmol/L) treatment in colon cancer cell lines. Results for all three low UGT1A1 mRNA expression cell lines: (A) HCT-116, (B) COLO-320DM, and (C) HCT-15. All expression results were assessed by QRT-PCR using ß2-microglobulin as an endogenous control. DMSO and PBS did not modify UGT1A1 mRNA expression. Columns, mean of two separated experiments done in triplicate; bars, SD. *, P < 0.05.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Relative UGT1A1 expression compared with the control group for each cell line

 
In vitro methylation of the UGT1A1 promoter represses transcriptional activity. To assess if the methylation of CpG in the promoter region of UGT1A1 affects directly the transcriptional activity of the gene, we studied various lengths of methylated and unmethylated UGT1A1 promoters. Four promoter constructions were analyzed for luciferase activity after in vitro methylation (Fig. 3A ). For methylated promoter constructions, a plasmid digestion was done with the HpaII restriction enzyme to ensure that methylation was complete. HpaII cuts only unmethylated CCGG sites (Fig. 3B). Luciferase expression of unmethylated plasmids (Fig. 3C-E) varied from 5- to 25-fold over the pGL3 basic vector and UGT1A1 expression was almost completely abolished for the methylated plasmids. Significant promoter activity in cell lines expressing undetectable or very low UGT1A1 mRNA levels indicated the presence of transcription factors required for UGT1A1 expression. These results show that methylation of the UGT1A1 promoter directly affects the transcriptional level of the gene.


Figure 3
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Repression of UGT1A1 promoter activity by DNA methylation in luciferase constructs. A, UGT1A1 promoter constructs of variable lengths and their associated CpG. B, digestion of plasmid DNA with the methylation-sensitive restriction enzyme HpaII, which cuts CCGG but not CmCGG sites. Luciferase activity in fully methylated (black) and unmethylated (white) promoter constructs for all three low UGT1A1 mRNA expression cell lines: (C) HCT-116, (D) COLO-320DM, and (E) HCT-15. Columns, mean of two separate experiments done in triplicate; bars, SD. LUC, luciferase; M, methylated; U, unmethylated. **, P < 0.01.

 
Methylation profiles of colon cancer cell lines correlate with UGT1A1 expression. The promoter of UGT1A1 encompasses a region of at least 3.3 kb upstream of the start codon involved in the constitutive and inducible expression of the gene (43). We therefore studied at least 5 kb of the 5' untranslated region for the presence of CpG islands and included sequences of the first exon.

The methylation profiles of all six colon cancer cell lines were determined for CpG +15 to –21 in the proximal promoter and first exon and for CpG –65 to –49 in the distal promoter (Fig. 4A ). Correlation analyses were then done between gene expression and methylation levels. Five CpG regions covering the promoter and the first exon of the UGT1A1 gene showed significant differences between high and low UGT1A1 expression cell lines (Fig. 4B). Regions, including CpG –16 to –13 (region D), –5 to –1 (region B), and +3 to +6 (region A), exhibited the most significant correlations with UGT1A1 mRNA expression (P < 0.01). Two other regions, including CpG –21 to –18 (region E) and –9 to –7 (region C), were also found in the promoter (P < 0.05). All other CpG islands analyzed (CpG +8 to +15 and CpG –65 to –49) were found to be fully methylated in all cell lines (data not shown). Three of these regions were predicted as CpG islands by the CpGPlot program (http://www.ebi.ac.uk/emboss/cpgplot/) and include CpG regions A, D, and F. Regions A and D exhibited the highest differences in methylation profiles between high and low UGT1A1 expression cell lines. In contrast, the third region (F) predicted by CpGPlot (from –47 to –56) was fully methylated for all CpG sites. These results suggest that the extent of methylation of specific CpG regions in the UGT1A1 gene predicts its expression in colon cancer cell lines.


Figure 4
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Schematic representation of the UGT1A1 gene methylation profile. A, schematic map of the CpG-rich regions (regions A-F) in UGT1A1. Vertical ticks, CpG sites. Amplicons 1 to 8 produced by PCR for bisulfite sequencing are also indicated. Arrow, direction of the bisulfite sequencing (for each amplicon). Nucleotide position of the C moiety of CpG regions relative to the ATG are as follows: F (–5,000 to –4,885), E (–1,157 to –892), D (–704 to –642), C (–513 to –432), B (–330 to –4), and A (+21 to +233) based on the UGT1A1 reference sequence AF297093. B, methylation profiles of colon cancer cell lines in regions A to E. Methylation status is presented as the mean of the methylation percentages of CpG for the indicated region and includes all CpG in that region. Comparison was done between percentages for a specific region in UGT1A1-negative and UGT1A1-positive cell lines. For significantly different methylation percentages between low-expressing and high-expressing cell lines, regions were reamplified and resequenced multiple times (at least thrice) for confirmation. *, P < 0.05; **, P < 0.01.

 
Demethylating treatment increases the inactivation of the anticancer agent SN-38 in HCT-116 cells. Compared with the SN-38-sensitive HCT-116 cell line, HT-29 cells were reported previously to exhibit intrinsic resistance to SN-38, owing to enhanced drug clearance via glucuronidation (15, 17). UGT1A1 is undetectable in HCT-116 cells but is abundant in HT-29 cells. These data suggest that the extent of promoter methylation could determine UGT1A1 expression and, ultimately, local SN-38 tumor concentrations. To ascertain this hypothesis, we assessed the capacity of 5-Aza-dC and trichostatin A treatment to modulate UGT1A1 protein levels and SN-38 glucuronidation in HCT-116 cells. Western blotting and SN-38 glucuronidation activities were done with microsomal subcellular fractions, where UGTs are localized. In treated cells, UGT1A1 protein levels were significantly higher compared with control cells (Fig. 5A ). In parallel, SN-38 glucuronide formation was increased by 300% in treated HCT-116 cells compared with untreated cells (Fig. 5B). The analysis of the UGT1A1 genomic regions by bisulfite treatment followed by sequencing confirmed the absence of methylated CpG in treated cells. These results show that the absence of a high percentage of methylation in the UGT1A1 promoter is associated with a significantly higher SN-38 glucuronidation capacity of tumor cells.


Figure 5
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Demethylation treatment increases protein levels and inactivation of SN-38 through glucuronidation. A, Western blot analysis of UGT1A1 protein after treatment with 5-Aza-dC alone or in combination with trichostatin A for the HCT-116 cell line. B, enzymatic assays for SN-38 glucuronidation with the isolated microsomal preparations from cells treated with both 5-Aza-dC and trichostatin A. Percentage of SN-38 glucuronide formation.

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
Transcriptional silencing by CpG island hypermethylation is a crucial event in colon carcinogenesis. This epigenetic mechanism has also been identified as a predictor of therapeutic efficiency for anticancer drugs used to treat various neoplasia (26, 27, 29, 44). In this study, we showed that DNA methylation governs the expression of the UGT1A1-metabolizing gene in colon cancer and that this mechanism may contribute to the differential chemosensitivity of colon tumors to irinotecan by influencing cellular inactivation of its active metabolite SN-38.

In this study, of the 50 primary colon tumors investigated for UGT1A1 expression, >82% showed undetectable or extremely low levels of UGT1A1 mRNA content. Previous studies showed that UGT1A1 is expressed in normal colon, but our results clearly indicate that its expression is lost in most colon tumors (32–34). On the other hand, >18% of tumors overexpressed UGT1A1, a characteristic recently associated with a resistance phenotype to SN-38 in colon cancer cells (15).

Transcriptional regulation of drug-metabolizing genes could occur through a variety of mechanisms, and DNA methylation constitutes a well-documented process for regulating gene transcription, especially in cancer (21–24). The role of epigenetics has never been reported previously for UGT genes. To investigate on the role of methylation in silencing UGT1A1 gene expression, UGT1A1-negative adenocarcinoma colon cell lines were selected as in vitro models representing a large proportion of primary colon tumor samples that do not express UGT1A1. In these colon cancer cells lacking basal UGT1A1 expression, hypermethylation of CpG islands was observed. Treatment of UGT1A1-negative colon cell lines with the DNA methyltransferase inhibitor 5-Aza-dC restored UGT1A1 expression, supporting that this gene is regulated by methylation. Besides, the drastic repression of reporter gene expression in the presence of methylated UGT1A1 promoters compared with unmethylated constructs in colon cell lines is consistent with a direct effect of DNA methylation on UGT1A1 gene expression. Additionally, the significant luciferase activity induced by unmethylated UGT1A1 promoter in colon cell lines that do not express basal UGT1A1 mRNA indicates that the loss of UGT1A1 expression in these cells is not caused by a lack of transcription factors required for UGT1A1 expression. Furthermore, recent studies have established a link between two important epigenetic modifications, DNA methylation and histone acetylation, in the regulation of gene expression and silencing. In addition to DNA methylation, histone acetylation mediates changes in the nucleosomal and chromatin structure of promoters, presumably affecting the accessibility of transcription factors to their cis-regulatory elements (45). In our experiments, a remarkable synergic increase of UGT1A1 expression was shown for the combined treatment of 5-Aza-dC and trichostatin A. Results support that this combined treatment is more active on methylation status and chromatin conformation to increase UGT1A1 mRNA expression as observed for other hypermethylated genes (46, 47). Altogether, results designate DNA hypermethylation as one of the mechanisms responsible for UGT1A1 repression in colon cancer cell lines.

Analyses of UGT1A1 promoter methylation status in cell lines revealed that the extent of methylation in specific regions of the promoter predicts UGT1A1 gene expression. Several dinucleotide CpG sequences are located within the core motif of putative sites for transcription factors. In vitro experiments with variable lengths of UGT1A1 promoters were also valuable in defining the potential CpG sites involved in transcriptional silencing. Data suggest that UGT1A1 is silenced as a result of CpG hypermethylation within a minimal region encompassing 260 bp upstream of the start codon. A previous study reported that UGT1A1 is regulated by HNF1{alpha} and HNF1ß through a putative HNF1 site in the proximal promoter (48). The localization of CpG islands in this study reveals that the HNF1 site is located in the methylated region B, between CpG –1 and –4. Additional analyses are needed to determine if methylation prevents the binding of HNF1{alpha} and HNF1ß or other transcriptional factors on these putative binding sites in colon tissues.

The exact role of UGT1A1 in colon carcinogenesis remains unknown, but there are potential consequences to the repression of this metabolic pathway in relation to cancer therapy. In our study, treatment with 5-Aza-dC and trichostatin A had the capacity to restore UGT1A1 protein expression in HCT-116 cells, lacking basal UGT1A1 expression, and resulted in cells that efficiently inactivate SN-38 through glucuronidation. These results support that exposure to SN-38 and the intrinsic susceptibility of cancer cells may be partially defined by UGT1A1 promoter methylation status and expression. The large family of UGT enzymes, broadly expressed in human tissues, belongs to the phase II drug-metabolizing enzymes involved in ~35% of phase II drug-metabolizing reactions for therapeutic drugs, including several anticancer drugs (49). Similar to other cytotoxic drugs, response to irinotecan in colon cancer patients may be determined by alterations in the metabolism of active metabolites. SN-38 is largely metabolized in human by glucuronidation and mostly by UGT1A1 (6–8). An important consequence associated with positive UGT1A1 methylation, and subsequent repression of the UGT1A1-mediated metabolic pathway, is the prospect that colon tumors associated with lower rates of SN-38 glucuronidation would retain higher levels of the compound. This could lead to a higher sensitivity to irinotecan. Conversely, tumor cells that overexpress UGT1A1 have the ability to inactivate a large proportion of the active drug that reaches the tumor. The presence of high levels of UGT activity and expression was identified as a characteristic recently associated with a resistance phenotype to SN-38 in colon cancer cells (15).

Several studies recently exposed the clinical importance of variable UGT1A1 activity governed by genetic polymorphisms in response to irinotecan-based chemotherapy (9–13). These investigations were focused on the role of constitutive variations of the UGT1A1 gene and showed that UGT1A1-deficient patients are at higher risk for severe hematologic and gastrointestinal toxicities. However, another important finding is the link between glucuronidation genotypes and variable response of cancer patients to irinotecan (11, 13). These observations further support that impaired glucuronidation may determine chemotherapeutic response by influencing drug concentration in tumors. Consequently, a contribution of the epigenetic silencing of UGT1A1 to irinotecan sensitivity of colon tumors is expected. In addition, silencing of UGT1A1 activity would affect other molecules metabolized primarily by this enzyme.

In conclusion, the present findings support the notion that methylation of UGT1A1, in addition to common genetic polymorphisms, may contribute to define tumors likely to respond to irinotecan in opposition to tumors that overexpress UGT1A1. The effect on irinotecan therapy outcome of aberrant UGT1A1 hypermethylation in colon cancer deserves further attention.


   Acknowledgments
 
We thank Patrick Caron for technical assistance and Dr. Olivier Barbier for helpful discussion.


   Footnotes
 
Grant support: Canadian Institutes of Health Research grant MOP-42392 and Canada Research Chair Program (C. Guillemette); Fonds de l'enseignement et de la recherche, Université Laval (J-F. Gagnon) and Fonds de la Recherche en Santé du Québec (O. Bernard) graduate studentship awards.

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 9/29/05; revised 12/ 5/05; accepted 1/ 9/06.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. American Cancer Society. Colorectal cancer facts & figures. Special edition 2005. Atlanta: American Cancer Society; 2005.
  2. Saltz L. Irinotecan-based combinations for the adjuvant treatment of stage III colon cancer. Oncology (Huntingt) 2000;14:47–50.
  3. Saltz LB, Douillard JY, Pirotta N, et al. Irinotecan plus fluorouracil/leucovorin for metastatic colorectal cancer: a new survival standard. Oncologist 2001;6:81–91.[Abstract/Free Full Text]
  4. Cunningham D, Maroun J, Vanhoefer U, Van Cutsem E. Optimizing the use of irinotecan in colorectal cancer. Oncologist 2001;6:17–23.[Abstract/Free Full Text]
  5. Rothenberg ML. Efficacy and toxicity of irinotecan in patients with colorectal cancer. Semin Oncol 1998;25:39–46.[Medline]
  6. Iyer L, Hall D, Das S, et al. Phenotype-genotype correlation of in vitro SN-38 (active metabolite of irinotecan) and bilirubin glucuronidation in human liver tissue with UGT1A1 promoter polymorphism. Clin Pharmacol Ther 1999;65:576–82.[CrossRef][Medline]
  7. Gagne JF, Montminy V, Belanger P, Journault K, Gaucher G, Guillemette C. Common human UGT1A polymorphisms and the altered metabolism of irinotecan active metabolite 7-ethyl-10-hydroxycamptothecin (SN-38). Mol Pharmacol 2002;62:608–17.[Abstract/Free Full Text]
  8. Hanioka N, Ozawa S, Jinno H, Ando M, Saito Y, Sawada J. Human liver UDP-glucuronosyltransferase isoforms involved in the glucuronidation of 7-ethyl-10-hydroxycamptothecin. Xenobiotica 2001;31:687–99.[CrossRef][Medline]
  9. Ando Y, Saka H, Ando M, et al. Polymorphisms of UDP-glucuronosyltransferase gene and irinotecan toxicity: a pharmacogenetic analysis. Cancer Res 2000;60:6921–6.[Abstract/Free Full Text]
  10. Iyer L, Das S, Janisch L, et al. UGT1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity. Pharmacogenomics J 2002;2:43–7.[CrossRef][Medline]
  11. Carlini LE, Meropol NJ, Bever J, et al. UGT1A7 and UGT1A9 polymorphisms predict response and toxicity in colorectal cancer patients treated with capecitabine/irinotecan. Clin Cancer Res 2005;11:1226–36.[Abstract/Free Full Text]
  12. Rouits E, Boisdron-Celle M, Dumont A, Guerin O, Morel A, Gamelin E. Relevance of different UGT1A1 polymorphisms in irinotecan-induced toxicity: a molecular and clinical study of 75 patients. Clin Cancer Res 2004;10:5151–9.[Abstract/Free Full Text]
  13. Font A, Sanchez JM, Taron M, et al. Weekly regimen of irinotecan/docetaxel in previously treated non-small cell lung cancer patients and correlation with uridine diphosphate glucuronosyltransferase 1A1 (UGT1A1) polymorphism. Invest New Drugs 2003;21:435–43.[CrossRef][Medline]
  14. Xu G, Zhang W, Ma MK, McLeod HL. Human carboxylesterase 2 is commonly expressed in tumor tissue and is correlated with activation of irinotecan. Clin Cancer Res 2002;8:2605–11.[Abstract/Free Full Text]
  15. Cummings J, Ethell BT, Jardine L, et al. Glucuronidation as a mechanism of intrinsic drug resistance in human colon cancer: reversal of resistance by food additives. Cancer Res 2003;63:8443–50.[Abstract/Free Full Text]
  16. Cummings J, Zelcer N, Allen JD, et al. Glucuronidation as a mechanism of intrinsic drug resistance in colon cancer cells: contribution of drug transport proteins. Biochem Pharmacol 2004;67:31–9.[CrossRef][Medline]
  17. Cummings J, Boyd G, Ethell BT, et al. Enhanced clearance of topoisomerase I inhibitors from human colon cancer cells by glucuronidation. Biochem Pharmacol 2002;63:607–13.[CrossRef][Medline]
  18. Lin YM, Furukawa Y, Tsunoda T, Yue CT, Yang KC, Nakamura Y. Molecular diagnosis of colorectal tumors by expression profiles of 50 genes expressed differentially in adenomas and carcinomas. Oncogene 2002;21:4120–8.[CrossRef][Medline]
  19. Yu J, Shannon WD, Watson MA, McLeod HL. Gene expression profiling of the irinotecan pathway in colorectal cancer. Clin Cancer Res 2005;11:2053–62.[Abstract/Free Full Text]
  20. Strassburg CP, Nguyen N, Manns MP, Tukey RH. Polymorphic expression of the UDP-glucuronosyltransferase UGT1A gene locus in human gastric epithelium. Mol Pharmacol 1998;54:647–54.[Abstract/Free Full Text]
  21. Jubb AM, Quirke P, Oates AJ. DNA methylation, a biomarker for colorectal cancer: implications for screening and pathological utility. Ann N Y Acad Sci 2003;983:251–67.[Medline]
  22. Lofton-Day C, Lesche R. DNA methylation markers in patients with gastrointestinal cancers. Current understanding, potential applications for disease management and development of diagnostic tools. Dig Dis 2003;21:299–308.[CrossRef][Medline]
  23. Toyota M, Itoh F, Imai K. DNA methylation and gastrointestinal malignancies: functional consequences and clinical implications. J Gastroenterol 2000;35:727–34.[CrossRef][Medline]
  24. Jass JR, Whitehall VL, Young J, Leggett BA. Emerging concepts in colorectal neoplasia. Gastroenterology 2002;123:862–76.[CrossRef][Medline]
  25. Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003;33:245–54.[CrossRef][Medline]
  26. Plumb JA, Strathdee G, Sludden J, Kaye SB, Brown R. Reversal of drug resistance in human tumor xenografts by 2'-deoxy-5-azacytidine-induced demethylation of the hMLH1 gene promoter. Cancer Res 2000;60:6039–44.[Abstract/Free Full Text]
  27. Ferreri AJ, Dell'Oro S, Capello D, et al. Aberrant methylation in the promoter region of the reduced folate carrier gene is a potential mechanism of resistance to methotrexate in primary central nervous system lymphomas. Br J Haematol 2004;126:657–64.[CrossRef][Medline]
  28. Esteller M. DNA methylation and cancer therapy: new developments and expectations. Curr Opin Oncol 2005;17:55–60.[CrossRef][Medline]
  29. Baker EK, Johnstone RW, Zalcberg JR, El-Osta A. Epigenetic changes to the MDR1 locus in response to chemotherapeutic drugs. Oncogene 2005;8:8.
  30. Ludlum DB. DNA alkylation by the haloethylnitrosoureas: nature of modifications produced and their enzymatic repair or removal. Mutat Res 1990;233:117–26.[CrossRef][Medline]
  31. Esteller M, Garcia-Foncillas J, Andion E, et al. Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 2000;343:1350–4.[Abstract/Free Full Text]
  32. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 2000;40:581–616.[CrossRef][Medline]
  33. Tukey RH, Strassburg CP. Genetic multiplicity of the human UDP-glucuronosyltransferases and regulation in the gastrointestinal tract. Mol Pharmacol 2001;59:405–14.[Abstract/Free Full Text]
  34. Strassburg CP, Manns MP, Tukey RH. Expression of the UDP-glucuronosyltransferase 1A locus in human colon. Identification and characterization of the novel extrahepatic UGT1A8. J Biol Chem 1998;273:8719–26.[Abstract/Free Full Text]
  35. Kondo Y, Shen L, Issa JP. Critical role of histone methylation in tumor suppressor gene silencing in colorectal cancer. Mol Cell Biol 2003;23:206–15.[Abstract/Free Full Text]
  36. Shmelkov SV, Jun L, St Clair R, et al. Alternative promoters regulate transcription of the gene that encodes stem cell surface protein AC133. Blood 2004;103:2055–61. Epub 2003 Nov 20.[Abstract/Free Full Text]
  37. Singal R, van Wert J, Bashambu M. Cytosine methylation represses glutathione S-transferase P1 (GSTP1) gene expression in human prostate cancer cells. Cancer Res 2001;61:4820–6.[Abstract/Free Full Text]
  38. Bearzatto A, Szadkowski M, Macpherson P, Jiricny J, Karran P. Epigenetic regulation of the MGMT and hMSH6 DNA repair genes in cells resistant to methylating agents. Cancer Res 2000;60:3262–70.[Abstract/Free Full Text]
  39. Sasaki M, Kaneuchi M, Sakuragi N, Dahiya R. Multiple promoters of catechol-O-methyltransferase gene are selectively inactivated by CpG hypermethylation in endometrial cancer. Cancer Res 2003;63:3101–6.[Abstract/Free Full Text]
  40. Liu H, Liu W, Wu Y, et al. Loss of epigenetic control of synuclein-{gamma} gene as a molecular indicator of metastasis in a wide range of human cancers. Cancer Res 2005;65:7635–43.[Abstract/Free Full Text]
  41. Villeneuve L, Girard H, Fortier LC, Gagne JF, Guillemette C. Novel functional polymorphisms in the UGT1A7 and UGT1A9 glucuronidating enzymes in Caucasian and African-American subjects and their impact on the metabolism of 7-ethyl-10-hydroxycamptothecin and flavopiridol anticancer drugs. J Pharmacol Exp Ther 2003;307:117–28.[Abstract/Free Full Text]
  42. Duguay Y, McGrath M, Lepine J, et al. The functional UGT1A1 promoter polymorphism decreases endometrial cancer risk. Cancer Res 2004;64:1202–7.[Abstract/Free Full Text]
  43. Sugatani J, Kojima H, Ueda A, et al. The phenobarbital response enhancer module in the human bilirubin UDP-glucuronosyltransferase UGT1A1 gene and regulation by the nuclear receptor CAR. Hepatology 2001;33:1232–8.[CrossRef][Medline]
  44. Esteller M. Dormant hypermethylated tumour suppressor genes: questions and answers. J Pathol 2005;205:172–80.[CrossRef][Medline]
  45. Loo S, Rine J. Silencing and heritable domains of gene expression. Annu Rev Cell Dev Biol 1995;11:519–48.[CrossRef][Medline]
  46. Yang X, Phillips DL, Ferguson AT, Nelson WG, Herman JG, Davidson NE. Synergistic activation of functional estrogen receptor (ER)-{alpha} by DNA methyltransferase and histone deacetylase inhibition in human ER-{alpha}-negative breast cancer cells. Cancer Res 2001;61:7025–9.[Abstract/Free Full Text]
  47. Nakajima M, Iwanari M, Yokoi T. Effects of histone deacetylation and DNA methylation on the constitutive and TCDD-inducible expressions of the human CYP1 family in MCF-7 and HeLa cells. Toxicol Lett 2003;144:247–56.[CrossRef][Medline]
  48. Bernard P, Goudonnet H, Artur Y, Desvergne B, Wahli W. Activation of the mouse TATA-less and human TATA-containing UDP-glucuronosyltransferase 1A1 promoters by hepatocyte nuclear factor 1. Mol Pharmacol 1999;56:526–36.[Abstract/Free Full Text]
  49. Evans WE, Relling MV. Pharmacogenomics: translating functional genomics into rational therapeutics. Science 1999;286:487–91.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Ann OncolHome page
C. Postma, M. Koopman, T. E. Buffart, P. P. Eijk, B. Carvalho, G. J. Peters, B. Ylstra, J. H. van Krieken, C. J. A. Punt, and G. A. Meijer
DNA copy number profiles of primary tumors as predictors of response to chemotherapy in advanced colorectal cancer
Ann. Onc., June 1, 2009; 20(6): 1048 - 1056.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
E. Cecchin, F. Innocenti, M. D'Andrea, G. Corona, E. De Mattia, P. Biason, A. Buonadonna, and G. Toffoli
Predictive Role of the UGT1A1, UGT1A7, and UGT1A9 Genetic Variants and Their Haplotypes on the Outcome of Metastatic Colorectal Cancer Patients Treated With Fluorouracil, Leucovorin, and Irinotecan
J. Clin. Oncol., May 20, 2009; 27(15): 2457 - 2465.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gagnon, J.-F.
Right arrow Articles by Guillemette, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gagnon, J.-F.
Right arrow Articles by Guillemette, C.

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