| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Departments of1 Medicine & Therapeutics,2 Anatomical and Cellular Pathology, and3 Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong;
aaddress correspondence to this author at: Department of Medicine & Therapeutics, Prince of Wales Hospital, 30-32 Ngan Shing St., Shatin, Hong Kong; fax 852-2637-3852, e-mail wkleung{at}cuhk.edu.hk
Colorectal cancer is the fourth most common cancer worldwide (1). There has been intense interest in the search for potential tumor markers that can be used in the screening of colorectal cancer. Because of the continuous shedding of intestinal cells into the lumen, genetic alterations found in tumors can also been detected in stool, which offers a golden opportunity for the noninvasive screening of colorectal cancer. Previous studies have demonstrated the feasibility of detecting altered DNA, including BAT26, APC, K-ras, and p53 mutations, in the feces of colorectal cancer patients (2)(3)(4)(5).
Epigenetic gene silencing by promoter hypermethylation is increasingly recognized to play a crucial role in carcinogenesis (6). In colorectal cancer, several tumor-related genes have been found to have promoter hypermethylation in the CpG islands (7)(8)(9)(10). These epigenetic changes are detected in the early phase of colorectal cancer development before the development of K-ras mutations (11). We tested the feasibility of detecting promoter hypermethylation of multiple tumor-related genes in fecal DNA of patients with colorectal cancer.
We recruited 20 colorectal cancer patients (mean age, 69 years; range, 45–90 years; 7 males). Patients who had familial adenomatous polyposis or hereditary nonpolyposis colon cancer, inflammatory bowel diseases, or previous colon surgery were excluded. Colon tumor biopsies were obtained during colonoscopy. Stool samples were collected before initiation of bowel preparation. The stool specimens were stored in a household freezer and then transferred for long-term storage at –80 °C. Twenty age-matched controls were randomly selected from individuals who participated in a colonoscopy screening program during the same period and had normal colonoscopy results (12). Stool samples were collected before bowel preparation as described for cancer patients, and healthy colon tissues were collected during colonoscopy from these controls. All patients and controls gave informed consent for the collection of tissues and stool specimens for the study. The study protocol was approved by the Institutional Review Board of the Chinese University of Hong Kong.
Samples obtained from cancer patients and controls were randomly coded before processing. DNA was isolated from colonic tissues by use of the QIAamp DNA Mini Kit (Qiagen) and from stool samples (250 mg) by use of the QIAamp DNA Stool Mini Kit (Qiagen). The quality of human DNA recovered from stool was verified by PCR amplification of the human ß-globulin gene. We treated 2 µg of DNA with the EZ DNA Methylation Kit (Zymo Research) and resuspended it in Tris-EDTA buffer (10 mmol/L Tris, 1 mmol/L EDTA, pH 7.5). The presence of methylated DNA in colonic tissues and stool was detected by methylation-specific PCR (MSP; Fig. 1
) (13). We examined six tumor-related genes: APC, ATM, HLTF, MGMT, hMLH-1, and GSTP1 (14)(15)(16)(17)(18)(19). This selection was based on our previous study, which found that the former five genes were frequently methylated in colorectal cancer (11). GSTP1 was previously found to be unmethylated in colorectal cancer and was included as a negative control. CpGenomeTM Universal Methylated DNA (Chemicon International Inc.) was used as the positive control, and template-free distilled water was included as a negative control for amplification. Moreover, tumor DNAs with methylation status determined in previous experiments were included as positive and negative controls. All PCRs were duplicated to ensure consistency and reproducibility of the results.
|
Promoter hypermethylation was frequently detected in colorectal cancer [ATM (45%), APC (55%), MGMT (45%), hMLH1 (45%), and HLTF (50%)], but not in GSTP1. Moreover, all cancer samples had promoter hypermethylation detected in at least one of the five tumor-related genes. The mean number of genes methylated in colon cancer was 2.4 (range, 1–5). In contrast, we detected methylation in none of the normal colonic tissues from non-cancer controls. The possible association between promoter hypermethylation and the clinicopathologic characteristics of cancers is shown in Table 1
. Methylation in the APC gene was more commonly found in advanced-stage cancer (P = 0.009). Promoter hypermethylation in hMLH1 was more frequently detected in older patients, and there was a significant difference in the mean age of patients with methylated and unmethylated cancers (76.4 vs 62.5 years; P = 0.02). Similar age-related methylation of hMLH1 has been reported in autopsy samples as well as in gastric cancer (20)(21), which may explain the high frequency of hMLH1 methylation noted in this study. However, there was no association between methylation and location of tumor, which may be related to the small sample size.
|
We successfully amplified the ß-globulin gene in stool samples from all 20 cancer patients. In contrast, 3 (15%) non-cancer controls had no detectable human DNA in their stool, suggesting that the amount of human fecal DNA is lower in individuals without cancer (22). The corresponding frequency of detectable methylation in fecal DNA of cancer patients was as follows: ATM (25%), APC (20%), MGMT (20%), hMLH1 (20%), and HLTF (25%). Methylation in GSTP1 was again not detected in any stool samples. With the combination of all five markers, 14 cancer patients had methylated fecal DNA detected, giving a sensitivity of 70% (95% confidence interval, 46–88%). Methylation was not detected in the stool samples of colonoscopy-negative individuals (100% specificity). To ensure that the negative results for fecal DNA from non-cancer controls was not attributable to a low human DNA concentration, we successfully amplified unmethylated promoter sequences in all stool samples with detectable ß-globulin gene. The corresponding positive and negative predictive values were 100% and 73.9%, respectively.
The mean number of methylated genes in the fecal DNA of cancer patients was 1.1 (range, 0–3). Concordant methylation patterns were detected in the tumor and the paired fecal DNA for all patients. There was a significant association between the numbers of methylated markers in stool and in primary tumors (P = 0.043). There was no difference in the mean number of genes methylated in fecal DNA for proximal (1.1) and distal colon cancer (1.1; P = 0.96). The presence of methylation in fecal DNA was not associated with any clinicopathologic characteristics of the cancer.
To further test the feasibility of detecting methylated DNA in stool as a screening test for colorectal neoplasm, we examined stool samples from six patients with significant colonic adenoma. The human ß-globulin gene was successfully amplified in all samples, and methylated DNA was detected in four (67%) samples.
In this study, we explored the feasibility of detecting methylated fecal DNA as a possible screening test for colorectal cancer. The feasibility of detecting aberrant methylation in the serum and urine of patients with gastrointestinal and bladder cancers has been demonstrated (17)(18)(19)(23)(24). To overcome the low sensitivity of single methylation markers, we used a combination of five markers. The sensitivity (70%) was superior to that reported for detection of APC mutation in stool (57%), which involved 144 PCR amplifications for a single sample (4). It is estimated that the percentage of mutant APC genes present in stool could be as low as 0.4% of all APC genes (4). To detect this minute amount of mutated DNA within a larger pool of wild-type DNA, highly sophisticated techniques that are capable of distinguishing 1 mutant molecule in more than 250 wild-type molecules are necessary.
An inherent advantage of MSP is the detection of a positive signal rather than the identification of signal loss and the low detection limit of 0.1% methylated DNA in a background of wild-type DNA (13). For that reason, the requirement for stool processing and storage was also less stringent in our study than in previous reports (2)(3)(4)(5). Stool samples were collected by patients, and DNA isolation was carried out with commercially available reagents. We used only 250 mg of stool sample for DNA isolation, whereas up to 4 g of stool sample was used in previous studies focusing on genetic alterations (2)(3)(4)(5). Furthermore, unlike detection of microsatellite instability markers in fecal DNA, which is limited to proximal colon cancers (5), methylation was detected in the fecal DNA of patients with both proximal and distal colorectal cancer.
Recently, Müller et al. (25) reported the use of assays for methylation changes in fecal DNA. Using a very different panel of markers, they found that three genes (SFRP2, SFRP5, and PGR) were differentially methylated in cancer patients. With the use of a single methylation marker, SFRP2, they were able to detect 10 of 13 patients with colorectal cancer. On the other hand, three apparently healthy individuals were found to have methylated SFRP2 in their stool (77% specificity). The discrepancy in performance of our study and that of Müller et al. (25) is likely attributable to the selection of different methylation markers, but the use of a more sensitive methylation assay, MethyLight, may also account for the higher sensitivity and lower specificity. Moreover, it remains to be determined whether the methylation patterns of cancers differ in diverse ethnic groups.
The results from our preliminary work demonstrate the feasibility of using a panel of epigenetic markers in screening of colorectal cancer. With the selection and refinement of methylation markers, epigenetic testing of fecal DNA may be a simple and promising screening test for colorectal cancer.
Acknowledgments
This study was supported by the Innovation & Technology Support Program of the Hong Kong SAR Government (ITS007/02).
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  V. Melotte, M. H. F. M. Lentjes, S. M. van den Bosch, D. M. E. I. Hellebrekers, J. P. J. de Hoon, K. A. D. Wouters, K. L. J. Daenen, I. E. J. M. Partouns-Hendriks, F. Stessels, J. Louwagie, et al. N-Myc Downstream-Regulated Gene 4 (NDRG4): A Candidate Tumor Suppressor Gene and Potential Biomarker for Colorectal Cancer J Natl Cancer Inst, July 1, 2009; 101(13): 916 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Zou, J. J. Harrington, A. M. Shire, R. L. Rego, L. Wang, M. E. Campbell, A. L. Oberg, and D. A. Ahlquist Highly Methylated Genes in Colorectal Neoplasia: Implications for Screening Cancer Epidemiol. Biomarkers Prev., December 1, 2007; 16(12): 2686 - 2696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Zou, J. Harrington, R. L. Rego, and D. A. Ahlquist A Novel Method to Capture Methylated Human DNA from Stool: Implications for Colorectal Cancer Screening Clin. Chem., September 1, 2007; 53(9): 1646 - 1651. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
