This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Woodson, K. Articles by Lanza, E. Search for Related Content PubMed PubMed Citation Articles by Woodson, K. Articles by Lanza, E. Related Collections Clinical Intervention Clinical Intervention: Biomarkers

Gene-Specific Methylation and Subsequent Risk of Colorectal Adenomas among Participants of the Polyp Prevention Trial

¹ Cancer Prevention Studies Branch, ² Basic Research Laboratory, Center for Cancer Research, ³ Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, Maryland and ⁴ Department of Surgery and of Biochemistry and Molecular Biology, Keck School of Medicine, University of Southern California, Los Angeles, California

Requests for reprints: Karen Woodson, Cancer Prevention Studies Branch, Center for Cancer Research, National Cancer Institute, Bethesda, MD 20892. Phone: 301-496-0651; Fax: 301-435-8645. E-mail: kw114v{at}nih.gov

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Hypermethylation of tumor suppressor and other regulatory genes is thought to play an important role in colorectal neoplasia and tumorigenesis. This study examined the association between gene methylation status in baseline adenomas and subsequent adenoma recurrence in a randomized dietary intervention study, the Polyp Prevention Trial. The methylation status of four genes [CDKN2A (p16), PTGS2 (COX2), ESR1 (ER-), and PGR(PR)] was determined by MethyLight in 284 baseline adenomas from 196 trial participants. The association of gene methylation with recurrence was determined using logistic regression models. Gene methylation was evaluated as percent of methylated reference, a measure of methylation of each gene relative to control DNA. ESR1methylation status was inversely associated with adenoma recurrence, odds ratio = 0.36 (95% confidence interval, 0.15-0.88; P = 0.02) for the highest compared with the lowest quartile of the ESR1methylation. Further, ESR1 methylation status was inversely associated with the recurrence of multiple adenomas, advanced adenomas, and the recurrence of adenomas in the proximal but not distal bowel. No association between CDKN2A, PTGS2, or PGR methylation and adenoma recurrence was observed. These data suggest that ESR1 methylation may play a role in subsequent adenoma recurrence.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Colorectal cancer is the third leading cause of cancer-related deaths worldwide. A majority of colorectal cancers are thought to arise through a series of steps, which include initial hyperproliferation of normal epithelial cells, formation of aberrant crypt foci and adenomas, and finally the transition to invasive carcinomas (1, 2).

In addition to genetic abnormalities, another potential mechanism that may be important in the progression of colorectal cancer is promoter DNA hypermethylation. Aberrant methylation of the CpG islands in the promoters of genes that are normally unmethylated has been linked to transcriptional silencing of that gene. A growing number of tumor suppressor genes and other regulatory genes have been shown to be hypermethylated in colorectal cancer. Prior studies have shown that colorectal cancers and adenomas (about 15-50%) have what is termed to be a "CpG island methylator phenotype," a condition characterized by the concurrent hypermethylation of a number of cancer-specific genes (3-7). In addition to promoter region hypermethylation, colorectal tumors have also been shown to possess widespread hypomethylation of CpG dinucleotides in noncoding parts of the genome, including repetitive elements (8). Thus, a global phenomenon of DNA methylation dysregulation may be occurring and contributing to the genesis and progression of colorectal cancers.

The causes of gene methylation aberrancies and their prognostic significance have not been established. Evidence suggests that individuals may be prone to methylation aberrancies as a consequence of genetic, dietary, or environmental exposures. Based on their study findings, Ricciardiello et al. (9) postulate that hypermethylation of the MLH1 gene (resulting in the loss of MLH1 expression) accounts for about 25% of adenomas that developed in subjects with a family history of colorectal cancer. The prevalence of MLH1 methylation was significantly higher in adenomas of those patients with a family history of colorectal cancer versus those without. Further, CDKN2A methylation in colorectal tumors has been associated with age, gender, and anatomic location (10, 11). For example, CDKN2A methylation was six times more likely to occur in colorectal tumors in women compared in men, and nine times more likely to occur in tumors located in the proximal colon compared with those in the distal bowel.

In this study, we evaluated the association of the methylation status of four genes with subsequent adenoma recurrence among participants of a randomized, dietary intervention trial, the Polyp Prevention Trial. The genes were selected, in part, based on data derived from studies suggesting that there are two distinct types of promoter gene methylation in colorectal cancer, one involving gene methylation in normal colon as a function of age (type A, i.e., ESR1) and the other involving genes exclusively methylated in neoplasia (type C, i.e., CDKN2A; ref. 3, 12). We included the PGR gene because it is part of the estrogen signaling pathway and shown to be methylated in other cancers (13, 14). We were interested in whether the presence of gene methylation in adenomas removed at the baseline colonoscopy visit was associated with adenoma recurrence detected at subsequent visits.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects and Specimens
The participants of the study are a subgroup of enrollees in the Polyp Prevention Trial, a randomized, dietary intervention trial designed to test whether a low-fat, high-fiber, fruit and vegetable diet can inhibit the recurrence of colorectal adenomas. The Polyp Prevention Trial was a collaboration between eight study centers and the National Cancer Institute and was approved by the Institutional Review Boards of all participating centers. The overall design, rationale, and trial results have been previously reported (15, 16). The trial findings showed that the rate of recurrent adenomas was not changed by the dietary intervention (15). This study consisted of 196 patients selected from those members of the study population (both intervention and control groups) who had completed the trial, had adequate polyp tissue slides available to provide sufficient DNA for methylation analysis, and included enough women participants to examine the influence of hormone replacement therapy on ESR1 methylation status. This sampling scheme resulted in a study population consisting of (a) women taking hormone replacement therapy at the baseline visit (n = 58); (b) women never taking hormones (at any visit; n = 59); and (c) men (n = 79). The three groups were frequency matched by age (±3 years). The polyp recurrence rate among the 196 subjects was not significantly different from that of the overall trial population (P = 0.20).

Methylation Analyses of Polyp Tissue
Polyp tissue was dissected from two 10-µm tissue sections by scraping the cells with a 25-gauge needle. Cells were lysed in a solution containing 100 mmol/L Tris-HCl (pH 8.0), 10 mmol/L EDTA (pH 8.0) 1 mg/mL proteinase K, and 0.05 mg/mL tRNA at 50°C overnight. The DNA from the cell lysate was bisulfite converted as previously described (17) and then purified using the a QIAamp 96 Blood Kit (Qiagen, Valencia, CA) according to the specifications of the suppliers. After purification, the bisulfite-converted DNA was desulfonated in 0.076 mol/L NaOH for 15 minutes at room temperature and then neutralized with HCl. Then a second purification step was done using the same QIAamp 96 Blood Kit to obtain bisulfite-converted DNA of high quality suitable for MethyLight analysis.

DNA methylation analysis was done by MethyLight as previously described (17, 18). All reactions used probes containing a 6-FAM fluorophore at the 5' end and a black hole quencher (BHQ) at the 3' end. Methylation reactions were done for CDKN2A-M2, PTGS2-M1, and ESR1-M1 using previously described primer and probe sequences (18). The PGR-M1 primer and probe sequences (in the 5' to 3' direction) are as follows: forward primer, GGCGGTGACGGTCGTATTC; reverse primer, ACAAACCGTCCCGCGAA; and probe, 6-FAM-AACAACCGCTCGCGCCCGA-BHQ. Control reactions for ACTB and COL2A1 were used to normalize for the amount of input DNA in each reaction. SssI-treated human peripheral blood leukocyte DNA was used as a fully methylated reference (18). The percentage of methylated reference (PMR) was calculated as previously described (18). Briefly, the PMR of specific genes was calculated by dividing the gene of interest/ACTB ratio of each sample by the ratio of a control sample (SssI-treated leukocyte DNA) and multiplying by 100. This was repeated for the second control gene, COL2A1, and the two values were averaged to obtain the PMR.

Statistical Analyses
All statistical analyses were conducted using STATA 8.0 (College Station, TX). The main analyses examined the association between gene methylation in baseline adenomas and adenoma recurrence over the course of the trial (between the T1 and T4 visits). Gene methylation was evaluated as percent methylation relative to a methylated reference DNA (PMR). PMR values were ranked and categorized into quartiles or tertiles. In the case where there was methylation data for more than one polyp per person, a single measure for methylation status per person was generated by averaging the PMR values across all polyps for each of the genes. The average PMR values for each of the genes were then ranked and categorized into quartiles or tertiles for subsequent analyses. To evaluate methylation status across all of the genes, we created a methylation index by adding PMRs for each gene and categorizing this composite into quartiles based on rank. The Wilcoxon rank-sum test was used to test for differences in recurrence status for continuous variables including PMR, age, and body mass index. The ² test was used for categorical variables. The association between gene methylation and recurrence was computed using logistic regression. Potential confounders were evaluated by assessing their association both with adenoma recurrence and gene methylation status. The association between gene methylation and adenoma multiplicity was evaluated using logistic regression with the outcome variable being those with more than one recurrent adenoma compared with those with a single adenoma or no adenoma recurrence as the reference group. The association between gene methylation and recurrence of advanced adenomas was evaluated using logistical regression with the outcome variable being those with recurrence of an advanced adenoma compared with those with recurrence of nonadvanced adenomas or no recurrence as the reference group. An adenoma(s) was considered to be "advanced" if it had a maximal diameter of at least 1 cm, had at least 25% villous elements, or displayed evidence of high-grade dysplasia. Logistic regression was also used to estimate the odds ratio and 95% confidence interval (OR; 95% CI) for recurrence of adenomas for each anatomic subsite (i.e., proximal, distal, or both, a combination of distal and proximal) relative to the no adenoma reference group (those with recurrent adenomas at locations other than the subsite being evaluated were not included in the analysis).

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
This study included an analysis of the methylation status of four genes (CDKN2A, PTGS2, ESR1, and PGR) in baseline adenomatous polyps from a selected group of participants of the Polyp Prevention Trial. The methylation status of the genes was evaluated by MethyLight analysis in 284 polyps from 196 patients (multiple polyps were evaluated in 60 of the patients). A description of selected baseline characteristics related to colorectal cancer risk by recurrence group is presented in Table 1. A higher proportion of men had an adenoma(s) recurrence; however, there were no significant differences for any of the other study factors, including age, smoking, alcohol consumption, hormone replacement therapy or nonsteriodal anti-inflammatory drug use, and family history of colorectal cancer, between participants who had or did not have an adenoma recurrence.

The association between ESR1methylation and recurrent adenoma characteristics, such as advanced polyp, multiple polyp, and polyp location in the large bowel, was evaluated (Table 4). ESR1methylation showed a strong inverse association with both recurrence of advanced adenoma(s) and recurrence of multiple adenomas (i.e., two or more adenomas). Interestingly, ESR1methylation status was also associated with the recurrence of proximally located adenomas but not with recurrence of adenomas in the distal bowel. Because CDKN2A methylation status has been reported to occur at a higher frequency in proximal tumors and in tumors from women, we also evaluated CDKN2A methylation across gender and location and found no differences (data not shown).

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

Our hypothesis was based, in part, on the postulated existence of a "methylator phenotype" in a subset of colorectal cancers, particularly those in the proximal bowel. Toyota et al. (3) hypothesized that CpG methylation of genes falls into two categories: type A, age related, where the gene is methylated in normal colon mucosa and the degree of methylation increases with age; and type C, cancer related, where methylation is specific to neoplastic or preneoplastic tissue. The genes selected in our analysis included representatives of both categories: ESR1, a type A gene, versus CDKN2A, a type C gene.

We did not find evidence for the correlation of methylation of any of the four genes with age, anatomic location of the adenomas, or gender. In contrast, Wiencke et al. (10) observed a 6-fold increase in CDKN2A methylation in colorectal tumors among women compared with men independent of age, degree of differentiation, and anatomic location. In addition, proximal tumors were nine times more likely to have CDKN2A methylation than tumors located distally. In another study, Hawkins et al. (19) found associations between CDKN2A methylation and age, sex, and anatomic location, with gene methylation occurring more frequently in elderly women and among proximal tumors. In contrast, Xiong et al. (11) evaluated the methylation status of four genes (including CDKN2A and ESR1) in microsatellite instability (+) and microsatellite instability (–) colorectal tumors and found no association between the methylation status of any of the genes and anatomic subsite, gender, or microsatellite instability status. Only ESR1 methylation was associated with age, with a higher extent of methylation observed in tumors from older women.

CDKN2A is a tumor suppressor gene and its expression is reduced in colorectal tumors. The presence of CDKN2A methylation in both adenomas and colorectal cancer has been well documented (3, 4, 10, 11). For example, Toyota et al. (4) evaluated CDKN2A methylation in 45 adenomas and 88 colorectal cancers and found CDKN2A methylation in 50% of adenomas and in 53% of colorectal cancers.

Although it is generally thought that up-regulation of PTGS2 expression is involved in colorectal carcinogenesis, a subset of tumors have shown a loss of PTGS2 expression, coincident with hypermethylation. In one study, PTGS2 methylation was observed in 13% of sporadic colorectal cancers and in 14% of colorectal adenomas (20).

Our finding of a high prevalence of extensive ESR1 methylation in adenomas is supported by prior studies (11, 12). Although ESR1 methylation is present in normal colonic tissue, colorectal tumors and adenomas show more extensive ESR1 methylation. Although we observed some methylation of ESR1 in 88% of all the adenomas, the extent of methylation across the adenomas was highly variable (the PMR ranged from 1% to 700%). We found an association between ESR1 methylation and the recurrence of proximally located adenomas but no association with the recurrence of adenomas in the distal bowel. In a prior study, ESR1 methylation was found to be highest in the "normal" colonic tissue from the proximal bowel, suggesting a potential field effect for ESR1 methylation in the proximal bowel (12).

Our observation of reduced risk of adenoma recurrence with ESR1 methylation seems to be in conflict with current theories of promoter-region gene methylation and tumorigenesis. According to this theory, extensive methylation of ESR1 would result in adenomas with reduced ER protein expression. There are two major forms of ER, ESR1 (ER-) and ESR2 (ER-ß). ESR1 has been shown to be present in low levels in the colon with no differences in mRNA or protein expression between normal colon, adenomas, and colon cancer (21, 22). ESR2 is the predominant form in normal colon and its expression is shown to be reduced in colorectal cancer. Weyant et al. (23) showed that relative levels of ESR1 and ESR2 expression were associated with inhibition of tumor formation in the Min/+ mouse model. Administration of 17ß-estradiol inhibited the number of tumors formed and this inhibition was concomitant with a decrease in ESR1 and an increase in ESR2 expression in the colon. Thus, a possible explanation of our findings is that a more extensive ESR1 methylation, corresponding to reduced protein expression, is beneficial for inhibition of tumor formation in the colon.

It is possible that the reduced risk for recurrence associated with extensive ESR1 methylation may not be a result of alterations in ER itself. ESR1 methylation may be a marker for some other molecular or cellular process related to recurrence. For example, methylation of ESR1, a type A gene, may reflect cell proliferation, senescence, or the rate of development of the polyp.

This study is strengthened by its conduct within a prospective clinical trial cohort that allowed for a detailed, controlled follow-up designed for the complete detection of adenoma recurrence. In addition, the evaluation of gene methylation in multiple adenomas per patient potentially allows for the improved classification of methylation status. Finally, the MethyLight technique used in this study allowed for quantitation of gene methylation, and the extent of gene methylation might be more biologically relevant than a categorical assessment of gene methylation in the study of adenoma recurrence.

In conclusion, this prospective study conducted within a clinical recurrence trial suggests that ESR1 gene methylation status in adenomas may have prognostic significance. This may have clinical relevance to colorectal cancer screening practice because individuals with recurrent adenomas are thought to be at increased risk for the development of colorectal cancer. The presence of low ESR1 methylation may indicate a need for increased surveillance or more frequent colonoscopy. These findings are intriguing but need to be confirmed in other study populations.

	Acknowledgments

 
We thank the investigators and personnel involved in the conduct of the Polyp Prevention Trial.

	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 10/ 4/04; revised 1/ 5/05; accepted 2/11/05.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell 1996;87:159–70.[CrossRef][Medline]
2. Lipkin M, Higgins P. Biological markers of cell proliferation and differentiation in human gastrointestinal diseases. Adv Cancer Res 1988;50:1–24.[Medline]
3. Toyota M, Ahuja N, Ohe-Toyota M, Herman JG, Baylin SB, Issa JP. CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A1999;96:8681–6.
4. Toyota M, Ohe-Toyota M, Ahuja N, Issa JP. Distinct genetic profiles in colorectal tumors with or without the CpG island methylator phenotype. Proc Natl Acad Sci U S A 2000;97:710–5.[Abstract/Free Full Text]
5. Rashid A, Shen L, Morris JS, Issa JP, Hamilton SR. CpG island methylation in colorectal adenomas. Am J Pathol 2001;159:1129–35.[Abstract/Free Full Text]
6. Bariol C, Suter C, Cheong K, et al. The relationship between hypomethylation and CpG island methylation in colorectal neoplasia. Am J Pathol 2003;162:1361–71.[Abstract/Free Full Text]
7. Lee S, Hwang KS, Lee HJ, Kim JS, Kang GH. Aberrant CpG island hypermethylation of multiple genes in colorectal neoplasia. Lab Invest 2004;84:884–93.[CrossRef][Medline]
8. Suter CM, Martin DI, Ward RL. Hypomethylation of L1 retrotransposons in colorectal cancer and adjacent normal tissue. Int J Colorectal Dis 2004;2:95–101.
9. Ricciardiello L, Goel A, Mantovani V, et al. Frequent loss of hMLH1 by promoter hypermethylation leads to microsatellite instability in adenomatous polyps of patients with a single first-degree member affected by colon cancer. Cancer Res 2003;63:787–92.[Abstract/Free Full Text]
10. Wiencke JK, Zheng S, Lafuente A, et al. Aberrant methylation of p16INK4a in anatomic and gender-specific subtypes of sporadic colorectal cancer. Cancer Epidemiol Biomarkers Prev 1999;8:501–6.[Abstract/Free Full Text]
11. Xiong Z, Wu AH, Bender CM, et al. Mismatch repair deficiency and CpG island hypermethylation in sporadic colon adenocarcinomas. Cancer Epidemiol Biomarkers Prev 2001;10:799–803.[Abstract/Free Full Text]
12. Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE, Baylin SB. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat Genet 1994;7:536–40.[CrossRef][Medline]
13. Widsschwendter A, Muller HM, Fiegl H, et al. DNA methylation in serum and tumors of cervical cancer patients. Clin Cancer Res 2004;15:565–71.
14. Liu ZJ, Maekawa M, Horii T, Morita M. The multiple promoter methylation profile of PR gene and ER gene in tumor cell lines. Life Sci 2003;73:1963–72.[Medline]
15. Schatzkin A, Lanza E, Corle D, et al. Lack of effect of a low-fat, high-fiber diet on the recurrence of colorectal adenomas. Polyp Prevention Trial Study Group. N Engl J Med 2000;342:1149–55.[Abstract/Free Full Text]
16. Schatzkin A, Lanza E, Freedman LS, et al. The polyp prevention trial I: rationale, design, recruitment, and baseline participant characteristics. Cancer Epidemiol Biomarkers Prev 1996;5:375–83.[Abstract]
17. Eads CA, Danenberg KD, Kawakami K, et al. MethyLight: a high-throughput assay to measure DNA methylation. Nucleic Acids Res 2000;28:E32.
18. Widschwendter M, Siegmund KD, Muller HM, et al. Association of breast cancer DNA methylation profiles with hormone receptor status and response to tamoxifen. Cancer Res 2004;64:3807–13.[Abstract/Free Full Text]
19. Hawkins N, Norrie M, Cheong K, et al. CpG island methylation in sporadic colorectal cancers and its relationship to microsatellite instability. Gastroenterology 2002;122:1376–87.[CrossRef][Medline]
20. Toyota M, Shen L, Ohe-Toyota M, Hamilton SR, Sinicrope FA, Issa JP. Aberrant methylation of the cyclooxygenase 2 CpG island in colorectal tumors. Cancer Res 2000;60:4044–8.[Abstract/Free Full Text]
21. Foley EF, Jazaeri AA, Shupnik MA, Jazaeri O, Rice LW. Selective loss of estrogen receptor ß in malignant human colon. Cancer Res 2000;60:245–8.[Abstract/Free Full Text]
22. Singh S, Sheppard MC, Langman MJ. Sex differences in the incidence of colorectal cancer: an exploration of oestrogen and progesterone receptors. Gut 1993;34:611–5.[Abstract/Free Full Text]
23. Weyant MJ, Carothers AM, Mahmoud NN, et al. Reciprocal expression of ER and ER ß is associated with estrogen-mediated modulation of intestinal tumorigenesis. Cancer Res 2001;61:2547–51.[Abstract/Free Full Text]

This article has been cited by other articles:

				Y. Suehiro, C. W. Wong, L. R. Chirieac, Y. Kondo, L. Shen, C. R. Webb, Y. W. Chan, A. S.Y. Chan, T. L. Chan, T.-T. Wu, et al. Epigenetic-Genetic Interactions in the APC/WNT, RAS/RAF, and P53 Pathways in Colorectal Carcinoma Clin. Cancer Res., May 1, 2008; 14(9): 2560 - 2569. [Abstract] [Full Text] [PDF]

				A. Y. Jung, E. M. Poole, J. Bigler, J. Whitton, J. D. Potter, and C. M. Ulrich DNA Methyltransferase and Alcohol Dehydrogenase: Gene-Nutrient Interactions in Relation to Risk of Colorectal Polyps Cancer Epidemiol. Biomarkers Prev., February 1, 2008; 17(2): 330 - 338. [Abstract] [Full Text] [PDF]

				S. Ogino, T. Kawasaki, M. Brahmandam, M. Cantor, G. J. Kirkner, D. Spiegelman, G. M. Makrigiorgos, D. J. Weisenberger, P. W. Laird, M. Loda, et al. Precision and Performance Characteristics of Bisulfite Conversion and Real-Time PCR (MethyLight) for Quantitative DNA Methylation Analysis J. Mol. Diagn., May 1, 2006; 8(2): 209 - 217. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Woodson, K. Articles by Lanza, E. Search for Related Content PubMed PubMed Citation Articles by Woodson, K. Articles by Lanza, E. Related Collections Clinical Intervention Clinical Intervention: Biomarkers