New Insights into a Controversial Topic: The Methylation–Cancer Connection

Abstract

Suzuki K, Suzuki I, Leodolter A, Horiuchi S, Yamashita K, Perucho M (Burnham Institute for Medical Research, Cancer Genetics and Epigenetics Research Program, La Jolla, CA). Global DNA demethylation in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell 2006;9:199–207.Suzuki et al studied the relationship between genetic and epigenetic alterations in gastrointestinal cancer integrating the data from 2 DNA fingerprinting methods that detect tumor-specific changes throughout the genome. Because fingerprinting methods are unbiased, they constitute powerful tools to obtain a panoramic view of the genetic and epigenetic changes accompanying tumorigenesis.Genomic DNA was prepared from paired normal tumor tissues in a series of 86 gastric cancer and 67 colon cancer patients who had undergone surgery; tissues were collected as fresh-frozen samples. Genomic damage was analyzed by arbitrarily primed PCR (AP-PCR), which provides an estimation of DNA copy number alterations. This technique, first described by Welsh and McClelland in 1990, can be used to detect tumor-specific somatic gains and losses of sequences throughout the genome. Quantitation of genomic alterations in tumors was done comparing intensities of the bands between normal and tumor DNA. Either change, a gain (stronger band in tumor DNA) or a loss (fainter band in tumor DNA), was scored as 1 alteration. The sum of alterations for the 2 arbitrary primers divided by the total number of bands in each normal–tumor pair fingerprinting was the estimated genomic damage fraction. By contrast, epigenomic damage was analyzed by methylation-sensitive amplified fragment length polymorphism. This technique can detect simultaneously both hypermethylation and hypomethylation events in tumor samples compared with the corresponding normal tissue, thereby allowing the investigation of DNA methylation alterations of hundreds of NotI sites in a particular cell genome. Scoring of quantitative changes between normal and tumor DNA was made by visual inspection and gave rise to the epigenomic damage fraction.In this elegant study of Suzuki et al, the extent of genomic damage was gradual and correlated with survival of patients with gastrointestinal cancer. Tumors with low genomic damage fraction revealed a better prognosis, independently of microsatellite instability status and of the cutoff point used to segregate tumors with relative low and high levels of genomic damage. By contrast, the distribution of hypermethylation and hypomethylation alterations was also gradual in both colon and gastric cancers, with no hint of a bimodal distribution. Interestingly, tumors with high levels of hypermethylation also exhibited high levels of hypomethylation. More important, high levels of hypomethylation were indicators of poor prognosis in both gastric and colon cancers. In contrast with hypomethylation, the hypermethylation alterations did not show as strong an association with different survival. In gastric cancer, patients with tumors showing high levels of hypermethylation had poorer survival than those with low levels, but in colon cancer hypermethylation did not correlate with a different survival rate. Both hypermethylation and hypomethylation somatic alterations in gastric and colon cancer correlated with the relative extent of genomic damage. Finally, whereas epigenetic alterations were age dependent, genetic alterations were not.CommentDNA methylation occurs principally at cytosine residues located in dinucleotide CpG sites. CpG sites are underrepresented in the genome, but are found concentrated at the expected levels in C+G–rich regions termed “CpG islands” that frequently coincide with promoter or gene regulatory regions. However, the bulk of lone CpG dinucleotides are found within the intergenic and intronic regions of DNA, particularly within repeat sequences and transposable elements. About 70% of all CpG dinucleotides in the human genome are heavily methylated and the remaining are typically observed in CpG-rich regions of ≥200 bp that span the promoters and sometimes the first exons of genes. These genomic patterns of CpG methylation are reprogrammed in the early embryo and maintained thereafter with considerable fidelity, and are of great functional relevance in normal cells. Patterns of methylation cooperate in the differential expression of genes, such as silencing of genes on the inactive X chromosome, and the production of age-related and tissue-specific gene expression (Gut 2007;56:140–148). Given the role of epigenetic changes in changing gene expression, as well as their close relationship with development, it is not surprising that cancer cells show a noticeable change in the configuration of epigenetic marks in their genomes.Changes in human DNA methylation patterns are an important feature of cancer development and progression. The cancer genome is characterized frequently by an overall decrease in the level of 5-methyl-cytosine concurrently with hypermethylation of promoter regions of some specific genes. These epigenetic processes are frequently linked with altered chromatin structure, changes in DNA methyltransferase activity, and loss of imprinting. The resultant aberrant transcription and chromosomal instability is believed to contribute to disease onset or progression, and increased tumor frequency and malignancy. In particular, hypomethylation of the genome largely affects the intergenic and intronic regions of the DNA, thus resulting in chromosomal instability and increased mutation events, whereas hypermethylation is responsible for promoter silencing of several tumor suppressor genes (reviewed in Biochimica et Biophysica Acta 2007;1775:138–162).Gastrointestinal cancer, for many years a prototypic model for the genetic basis of cancer, is now increasingly cited as an exemplar of the role of epigenetic changes in tumorigenesis. Indeed, the wide range of accessible and well-characterized colorectal lesions, from aberrant crypt foci to carcinoma, provides an excellent opportunity to understand how epigenetics and genetics conspire to produce malignancy. In the present outstanding study (Cancer Cell 2006;9:199–207), both processes were simultaneously evaluated by unbiased DNA fingerprinting methods, thus favoring a panoramic view of the genetic and epigenetic changes accompanying tumorigenesis. Interestingly, screening colon cancers by AP-PCR, Ionov et al (Nature 1993;363:558–561) showed that some tumors accumulated hundreds or thousands of somatic mutations in microsatellite sequences, granting the first description of an alternative genetic pathway for colorectal and other cancers. More than 10 years later, a similar approach from the same group provides new insights into the epigenetics and genetics of malignancy.The main finding of the study is the dominant role of hypomethylation over DNA hypermethylation in linking epigenotype with cancer genotype (genomic damage), and with cancer phenotype (prognosis), as well as the age-dependent nature of global DNA hypomethylation in gastrointestinal cancer (Cancer Cell 2006;9:199–207). Considering that cancer is a disease of aging and that alterations in DNA copy number were independent of age, these authors suggest that epigenetic, but not genetic, alterations may occur in normal tissue, and they are exposed in tumors by clonal expansion after neoplastic transformation. Thus, epigenetic alterations seem to precede the genetic changes.It should be noted that involvement of focal DNA hypermethylation in carcinogenesis has been well established in the last decade. Toyota et al (Proc Natl Acad Sci U S A 1999;96:8681–8686) proposed for the first time a CpG island methylator phenotype (CIMP) in a subset of colorectal cancers to explain the somatic hypermethylation associated with silencing of MLH1 and several tumor suppressor genes. The existence of a methylator phenotype has been supported by other studies (Gastroenterology 2006;131:797–808; Gut 2006;55:1000–1006; Nat Genet 2006;38:787–793), with growing evidence that CIMP tumors are clinically, pathologically, and genetically different (Gastroenterology 2005;129:837–845). However, the controversy surrounding remains as to whether CIMP tumors represent a biologically distinct group of colorectal cancers or are an artificially selected group from a continuum of tumors showing different degrees of methylation at particular loci remains. In this context, the results of Suzuki et al (Cancer Cell 2006;9:199–207), demonstrating a gradual distribution of DNA hypermethylation with no evidence of bimodality or drastic discontinuity, argue against the existence of a methylator phenotype. More studies are needed to clarify this controversial issue. Suzuki K, Suzuki I, Leodolter A, Horiuchi S, Yamashita K, Perucho M (Burnham Institute for Medical Research, Cancer Genetics and Epigenetics Research Program, La Jolla, CA). Global DNA demethylation in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell 2006;9:199–207. Suzuki et al studied the relationship between genetic and epigenetic alterations in gastrointestinal cancer integrating the data from 2 DNA fingerprinting methods that detect tumor-specific changes throughout the genome. Because fingerprinting methods are unbiased, they constitute powerful tools to obtain a panoramic view of the genetic and epigenetic changes accompanying tumorigenesis. Genomic DNA was prepared from paired normal tumor tissues in a series of 86 gastric cancer and 67 colon cancer patients who had undergone surgery; tissues were collected as fresh-frozen samples. Genomic damage was analyzed by arbitrarily primed PCR (AP-PCR), which provides an estimation of DNA copy number alterations. This technique, first described by Welsh and McClelland in 1990, can be used to detect tumor-specific somatic gains and losses of sequences throughout the genome. Quantitation of genomic alterations in tumors was done comparing intensities of the bands between normal and tumor DNA. Either change, a gain (stronger band in tumor DNA) or a loss (fainter band in tumor DNA), was scored as 1 alteration. The sum of alterations for the 2 arbitrary primers divided by the total number of bands in each normal–tumor pair fingerprinting was the estimated genomic damage fraction. By contrast, epigenomic damage was analyzed by methylation-sensitive amplified fragment length polymorphism. This technique can detect simultaneously both hypermethylation and hypomethylation events in tumor samples compared with the corresponding normal tissue, thereby allowing the investigation of DNA methylation alterations of hundreds of NotI sites in a particular cell genome. Scoring of quantitative changes between normal and tumor DNA was made by visual inspection and gave rise to the epigenomic damage fraction. In this elegant study of Suzuki et al, the extent of genomic damage was gradual and correlated with survival of patients with gastrointestinal cancer. Tumors with low genomic damage fraction revealed a better prognosis, independently of microsatellite instability status and of the cutoff point used to segregate tumors with relative low and high levels of genomic damage. By contrast, the distribution of hypermethylation and hypomethylation alterations was also gradual in both colon and gastric cancers, with no hint of a bimodal distribution. Interestingly, tumors with high levels of hypermethylation also exhibited high levels of hypomethylation. More important, high levels of hypomethylation were indicators of poor prognosis in both gastric and colon cancers. In contrast with hypomethylation, the hypermethylation alterations did not show as strong an association with different survival. In gastric cancer, patients with tumors showing high levels of hypermethylation had poorer survival than those with low levels, but in colon cancer hypermethylation did not correlate with a different survival rate. Both hypermethylation and hypomethylation somatic alterations in gastric and colon cancer correlated with the relative extent of genomic damage. Finally, whereas epigenetic alterations were age dependent, genetic alterations were not. CommentDNA methylation occurs principally at cytosine residues located in dinucleotide CpG sites. CpG sites are underrepresented in the genome, but are found concentrated at the expected levels in C+G–rich regions termed “CpG islands” that frequently coincide with promoter or gene regulatory regions. However, the bulk of lone CpG dinucleotides are found within the intergenic and intronic regions of DNA, particularly within repeat sequences and transposable elements. About 70% of all CpG dinucleotides in the human genome are heavily methylated and the remaining are typically observed in CpG-rich regions of ≥200 bp that span the promoters and sometimes the first exons of genes. These genomic patterns of CpG methylation are reprogrammed in the early embryo and maintained thereafter with considerable fidelity, and are of great functional relevance in normal cells. Patterns of methylation cooperate in the differential expression of genes, such as silencing of genes on the inactive X chromosome, and the production of age-related and tissue-specific gene expression (Gut 2007;56:140–148). Given the role of epigenetic changes in changing gene expression, as well as their close relationship with development, it is not surprising that cancer cells show a noticeable change in the configuration of epigenetic marks in their genomes.Changes in human DNA methylation patterns are an important feature of cancer development and progression. The cancer genome is characterized frequently by an overall decrease in the level of 5-methyl-cytosine concurrently with hypermethylation of promoter regions of some specific genes. These epigenetic processes are frequently linked with altered chromatin structure, changes in DNA methyltransferase activity, and loss of imprinting. The resultant aberrant transcription and chromosomal instability is believed to contribute to disease onset or progression, and increased tumor frequency and malignancy. In particular, hypomethylation of the genome largely affects the intergenic and intronic regions of the DNA, thus resulting in chromosomal instability and increased mutation events, whereas hypermethylation is responsible for promoter silencing of several tumor suppressor genes (reviewed in Biochimica et Biophysica Acta 2007;1775:138–162).Gastrointestinal cancer, for many years a prototypic model for the genetic basis of cancer, is now increasingly cited as an exemplar of the role of epigenetic changes in tumorigenesis. Indeed, the wide range of accessible and well-characterized colorectal lesions, from aberrant crypt foci to carcinoma, provides an excellent opportunity to understand how epigenetics and genetics conspire to produce malignancy. In the present outstanding study (Cancer Cell 2006;9:199–207), both processes were simultaneously evaluated by unbiased DNA fingerprinting methods, thus favoring a panoramic view of the genetic and epigenetic changes accompanying tumorigenesis. Interestingly, screening colon cancers by AP-PCR, Ionov et al (Nature 1993;363:558–561) showed that some tumors accumulated hundreds or thousands of somatic mutations in microsatellite sequences, granting the first description of an alternative genetic pathway for colorectal and other cancers. More than 10 years later, a similar approach from the same group provides new insights into the epigenetics and genetics of malignancy.The main finding of the study is the dominant role of hypomethylation over DNA hypermethylation in linking epigenotype with cancer genotype (genomic damage), and with cancer phenotype (prognosis), as well as the age-dependent nature of global DNA hypomethylation in gastrointestinal cancer (Cancer Cell 2006;9:199–207). Considering that cancer is a disease of aging and that alterations in DNA copy number were independent of age, these authors suggest that epigenetic, but not genetic, alterations may occur in normal tissue, and they are exposed in tumors by clonal expansion after neoplastic transformation. Thus, epigenetic alterations seem to precede the genetic changes.It should be noted that involvement of focal DNA hypermethylation in carcinogenesis has been well established in the last decade. Toyota et al (Proc Natl Acad Sci U S A 1999;96:8681–8686) proposed for the first time a CpG island methylator phenotype (CIMP) in a subset of colorectal cancers to explain the somatic hypermethylation associated with silencing of MLH1 and several tumor suppressor genes. The existence of a methylator phenotype has been supported by other studies (Gastroenterology 2006;131:797–808; Gut 2006;55:1000–1006; Nat Genet 2006;38:787–793), with growing evidence that CIMP tumors are clinically, pathologically, and genetically different (Gastroenterology 2005;129:837–845). However, the controversy surrounding remains as to whether CIMP tumors represent a biologically distinct group of colorectal cancers or are an artificially selected group from a continuum of tumors showing different degrees of methylation at particular loci remains. In this context, the results of Suzuki et al (Cancer Cell 2006;9:199–207), demonstrating a gradual distribution of DNA hypermethylation with no evidence of bimodality or drastic discontinuity, argue against the existence of a methylator phenotype. More studies are needed to clarify this controversial issue. DNA methylation occurs principally at cytosine residues located in dinucleotide CpG sites. CpG sites are underrepresented in the genome, but are found concentrated at the expected levels in C+G–rich regions termed “CpG islands” that frequently coincide with promoter or gene regulatory regions. However, the bulk of lone CpG dinucleotides are found within the intergenic and intronic regions of DNA, particularly within repeat sequences and transposable elements. About 70% of all CpG dinucleotides in the human genome are heavily methylated and the remaining are typically observed in CpG-rich regions of ≥200 bp that span the promoters and sometimes the first exons of genes. These genomic patterns of CpG methylation are reprogrammed in the early embryo and maintained thereafter with considerable fidelity, and are of great functional relevance in normal cells. Patterns of methylation cooperate in the differential expression of genes, such as silencing of genes on the inactive X chromosome, and the production of age-related and tissue-specific gene expression (Gut 2007;56:140–148). Given the role of epigenetic changes in changing gene expression, as well as their close relationship with development, it is not surprising that cancer cells show a noticeable change in the configuration of epigenetic marks in their genomes. Changes in human DNA methylation patterns are an important feature of cancer development and progression. The cancer genome is characterized frequently by an overall decrease in the level of 5-methyl-cytosine concurrently with hypermethylation of promoter regions of some specific genes. These epigenetic processes are frequently linked with altered chromatin structure, changes in DNA methyltransferase activity, and loss of imprinting. The resultant aberrant transcription and chromosomal instability is believed to contribute to disease onset or progression, and increased tumor frequency and malignancy. In particular, hypomethylation of the genome largely affects the intergenic and intronic regions of the DNA, thus resulting in chromosomal instability and increased mutation events, whereas hypermethylation is responsible for promoter silencing of several tumor suppressor genes (reviewed in Biochimica et Biophysica Acta 2007;1775:138–162). Gastrointestinal cancer, for many years a prototypic model for the genetic basis of cancer, is now increasingly cited as an exemplar of the role of epigenetic changes in tumorigenesis. Indeed, the wide range of accessible and well-characterized colorectal lesions, from aberrant crypt foci to carcinoma, provides an excellent opportunity to understand how epigenetics and genetics conspire to produce malignancy. In the present outstanding study (Cancer Cell 2006;9:199–207), both processes were simultaneously evaluated by unbiased DNA fingerprinting methods, thus favoring a panoramic view of the genetic and epigenetic changes accompanying tumorigenesis. Interestingly, screening colon cancers by AP-PCR, Ionov et al (Nature 1993;363:558–561) showed that some tumors accumulated hundreds or thousands of somatic mutations in microsatellite sequences, granting the first description of an alternative genetic pathway for colorectal and other cancers. More than 10 years later, a similar approach from the same group provides new insights into the epigenetics and genetics of malignancy. The main finding of the study is the dominant role of hypomethylation over DNA hypermethylation in linking epigenotype with cancer genotype (genomic damage), and with cancer phenotype (prognosis), as well as the age-dependent nature of global DNA hypomethylation in gastrointestinal cancer (Cancer Cell 2006;9:199–207). Considering that cancer is a disease of aging and that alterations in DNA copy number were independent of age, these authors suggest that epigenetic, but not genetic, alterations may occur in normal tissue, and they are exposed in tumors by clonal expansion after neoplastic transformation. Thus, epigenetic alterations seem to precede the genetic changes. It should be noted that involvement of focal DNA hypermethylation in carcinogenesis has been well established in the last decade. Toyota et al (Proc Natl Acad Sci U S A 1999;96:8681–8686) proposed for the first time a CpG island methylator phenotype (CIMP) in a subset of colorectal cancers to explain the somatic hypermethylation associated with silencing of MLH1 and several tumor suppressor genes. The existence of a methylator phenotype has been supported by other studies (Gastroenterology 2006;131:797–808; Gut 2006;55:1000–1006; Nat Genet 2006;38:787–793), with growing evidence that CIMP tumors are clinically, pathologically, and genetically different (Gastroenterology 2005;129:837–845). However, the controversy surrounding remains as to whether CIMP tumors represent a biologically distinct group of colorectal cancers or are an artificially selected group from a continuum of tumors showing different degrees of methylation at particular loci remains. In this context, the results of Suzuki et al (Cancer Cell 2006;9:199–207), demonstrating a gradual distribution of DNA hypermethylation with no evidence of bimodality or drastic discontinuity, argue against the existence of a methylator phenotype. More studies are needed to clarify this controversial issue. ReplyGastroenterologyVol. 132Issue 5PreviewWe thank Drs Gonzalez-García and Castells for their commentary on our paper (Cancer cell, 2006;9:199–207). We have little to add, but we appreciate the opportunity we have been given to highlight the interesting difference between the genetic and epigenetic alterations observed in gastrointestinal cancer. Although somatic mutations must be present in normal tissues, they are not reproducibly found by the techniques used for mutational analyses. This means that mutations must be very rare. If they are nonfunctional, they remain undetectable, confined to a few normal cells in the intestinal epithelium. Full-Text PDF