Research Highlights

Abstract

EpigenomicsVol. 4, No. 1 News & ViewsFree AccessResearch HighlightsMoshe SzyfMoshe SzyfDepartment of Pharmacology & Therapeutics, McGill University, 3655 Sir William Osler Promenade, Montreal, Quebec, Canada. Search for more papers by this authorEmail the corresponding author at moshe.szyf@mcgill.caPublished Online:14 Feb 2012https://doi.org/10.2217/epi.11.115AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinkedInRedditEmail Evaluation of: Azhikina T, Kozlova A, Skvortsov T, Sverdlov E. Heterogeneity and degree of TIMP4, GATA4, SOX18, and EGFL7 gene promoter methylation in non-small cell lung cancer and surrounding tissues. Cancer Genet. 204(9), 492–500 (2011).DNA methylation alterations are hypothesized to play a role in the onset of cancer. DNA methylation patterns in otherwise normal tissue starting its course toward cancer would be expected to be heterogeneous, a mixture of normal and early tumor cells, and it might, therefore, be possible to detect these early changes in a tissue, prior to presentation of histopathological and clinical changes. The paper by Azhikina et al. reveals heterogeneous methylation in non-small-cell lung cancer and surrounding ‘normal’ tissues using high-resolution melting of CpG islands within promoters of the TIMP4, GATA4, SOX18 and EGFL7 genes [1]. Heterogeneity of DNA methylation has been shown before in several studies [2,3]; however, these studies used labor-intensive bisulfite sequencing. The high-resolution melting method measures both the overall level of DNA methylation, as well as the extent of methylation heterogeneity in the tissue by analyzing the melting curves of bisulfite converted amplified fragments using a standard light cycler. The authors show that this measure of heterogeneity per se clearly differentiates tumors from normal unaffected individuals. Surprisingly however, in contrast to normal tissue from unaffected controls, the state of DNA methylation of these genes is heterogeneous in ‘normal’ tissue from affected individuals. Heterogeneity therefore differentiates ‘normal’ samples in the cancer group from the normal samples in the unaffected individuals. This DNA methylation heterogeneity might emerge early in the progression of tumors even before other histopathological changes appear; as suggested by the fact that it appears in otherwise normal tissue. Therefore, it is plausible that DNA methylation heterogeneity could serve as a diagnostic tool to detect the earliest stages in tumorigenesis, even before any other changes appear.Although this might represent an exciting opportunity for developing early diagnostic biomarkers for cancer, there are a number of issues that need to be addressed first. DNA methylation is binary at the cellular level; a CpG site in a given allele in a single cell is either methylated or unmethylated. Therefore, heterogeneity in DNA methylation in a tissue sample implies that the sample is a mixture of some cells that are methylated at a given site and others are not. If DNA methylation provides a selective growth advantage, then one would predict that heterogeneity would eventually drift into homogeneity of the methylated methylotype; the degree of heterogeneity should decrease with advanced stages. However, the observations presented in this paper suggest that heterogeneity is maintained in advanced tumors. This points to the attractive hypothesis that heterogeneity itself is required for tumor growth and survival and that the presence of multi ‘methylotypes’ is required for the multicellular complex interactions occurring within a tumor. A tumor might not be just a clonal expansion of a growth selected single cellular phenotype’ but a synergistic ‘community’ combined with many subtly different cellular phenotypes. If this is the case, it is also possible that the heterogeneous DNA methylation pattern of methylated genes in the ‘normal’ adjacent tissue does not represent early stages of the tumor, but is rather triggered by the advanced tumor through the bilateral interactions of an advanced tumor and its cellular environment. If this is the case, the heterogeneous DNA methylation studied here in adjacent tumor tissue might not provide an early diagnosis. Evidence is required to demonstrate that this heterogeneity occurs in individuals at the preclinical stages of cancer. As this assay requires a lung biopsy, which is a highly invasive procedure, this is extremely difficult to acquire in this type of cancer.References1 Azhikina T, Kozlova A, Skvortsov T, Sverdlov E. Heterogeneity and degree of TIMP4, GATA4, SOX18, and EGFL7 gene promoter methylation in non-small cell lung cancer and surrounding tissues. Cancer Genet.204(9),492–500 (2011).Crossref, Medline, CAS, Google Scholar2 Heng HH, Bremer SW, Stevens JB, Ye KJ, Liu G, Ye CJ. Genetic and epigenetic heterogeneity in cancer: a genome-centric perspective. J. Cell Physiol.220(3),538–547 (2009).Crossref, Medline, CAS, Google Scholar3 Aggerholm A, Guldberg P, Hokland M, Hokland P. Extensive intra- and interindividual heterogeneity of p15INK4B methylation in acute myeloid leukemia. Cancer Res.59(2),436–441 (1999).Medline, CAS, Google ScholarEvaluation of: Cassinotti E, Melson J, Liggett T et al. DNA methylation patterns in blood of patients with colorectal cancer and adenomatous colorectal polyps. Int. J. Cancer doi:10.1002/ijc.26484 (2011) (Epub ahead of print).A critical challenge in cancer diagnostics is developing noninvasive diagnostic biomarkers that could detect the earliest stages of cancer. DNA methylation differences between tumors and normal cells have a high diagnostic potential and a large number of studies have derived cancer-specific, as well as cancer stage specific DNA methylation signatures [1]. However, since it is believed that changes in DNA methylation are molecular signatures of the tumor itself, highly invasive procedures such as biopsies are required for utilizing DNA methylation states as biomarkers. This makes them irrelevant for early screening in most instances. Since tumor cells circulate, die and shed their DNA in the circulation system, several groups have attempted to identify blood-based tumor DNA methylation markers. Such DNA methylation biomarkers could be used for early screening as well as follow-up for subjects with a high risk for certain cancers such as colorectal cancers (CRC). These biomarkers could perhaps even substitute invasive methods such as colonoscopy or other insensitive imaging methods. Cassinotti et al., have recently examined whether there are DNA methylation markers in cell-free circulating plasma DNA that could differentiate patients with early CRC, (stages I and II) and patients with adenomatous polyps from normal controls using DNA methylation profiling of 56 genes hypermethylated in cancer [2]. A group of six promoters (i.e., CYCD2, HIC1, PAX5, RASSF1A, RB1 and SRBC) were found to differentiate CRC patients from those without polyps with a sensitivity of 83.7% and a specificity of 67.9%. Interestingly, three informative genes (i.e., HIC1, MDGI and RASSF1A) were able to differentiate subjects with adenomatous polyps from subjects without any polyps with a sensitivity of 54.6% and a specificity of 64.5%. Two of these genes (i.e., HIC1 and RASSF1A) were informative for differentiation of both adenomatous polyps and CRC patients from controls. This suggests that methylation of these genes emerges early, prior to progression of the polyps to carcinoma. These DNA methylation biomarkers could potentially serve as diagnostic markers for early screening.This study is promising by revealing that DNA with distinct DNA methylation profiles is circulating in blood as early as the ademonatous polyps stage, implying that it might be possible to identify DNA methylation biomarkers of the earliest stages preceding CRC in blood. However, several questions remain. The data suggests that this approach is more effective in detecting higher grades of dysplasia, which might be a consequence of increasing concentrations of tumor-derived DNA in blood with an increase in the state of dysplasia. This might exclude the application of blood-derived DNA methylation markers for early diagnosis where such an assay is needed the most. Other remaining questions are whether it would be possible to differentiate ademonatous polyps from CRC. Would the accuracy and specificity of these DNA methylation biomarkers increase or decrease in larger clinical studies? Are there other DNA methylation biomarkers in the blood that would exhibit higher sensitivity and specificity that could be identified by a more detailed genome-wide profiling? Although it is unclear whether these blood-based DNA methylation biomarkers would be sufficiently specific and accurate to replace current screening methods in CRC, this appears to be a promising approach that deserves increased attention.References1 Yan PS, Perry MR, Laux DE, Asare AL, Caldwell CW, Huang TH. CpG island arrays: an application toward deciphering epigenetic signatures of breast cancer. Clin. Cancer Res.6(4),1432–1438 (2000).Medline, CAS, Google Scholar2 Cassinotti E, Melson J, Liggett T et al. DNA methylation patterns in blood of patients with colorectal cancer and adenomatous colorectal polyps. Int. J. Cancer doi:10.1002/ijc.26484 (2011) (Epub ahead of print).Medline, Google ScholarEvaluation of: Caretti A, Sirchia SM, Tabano S, Zulueta A, Dall’olio F, Trinchera M. DNA methylation and histone modifications modulate the β1,3 galactosyltransferase β 3Gal-T5 native promoter in cancer cells. Int. J. Biochem. Cell Biol. 44(1), 84–90 (2011).DNA demethylating agents are highly effective in activating genes silenced by methylation of their promoters [1]. DNA methylation instruct histone hypoacetylation [2]. DNA methylation inhibitor Vidaza® (5-azacytidine) and histone deacetylase inhibitors have been tested independently and in combination in studies in cancer cells [3]. However, there is a subset of methylated sequences that are not activated even with a combination of both trichostatin A (TSA) and 5-azacytidine [4]. A plausible hypothesis is that additional epigenetic modifications are silencing these genes in cancer cells rendering them resistant to hypomethylating agents and histone deacetylase inhibitors. In a recent study by Carettithe, response of β1,3 galactosyltransferase β 3Gal-T5, which encodes an enzyme that catalyzes type 1 chain oligosaccharide synthesis and is methylated in several cancers, to a combination of TSA and 5-azacytidine was examined [5]. This gene is methylated and silenced in several cell lines and is partially active in others. Whereas 5-azacytidine was able to weakly induce the gene in a cell line where it was partially expressed, it was unable to induce the gene expression in HCT-15 and MDA-MB-231 cells where this gene was silenced, although the gene was partially demethylated in all cases in response to the treatment. The authors examined the possibility that other chromatin modifications were involved in silencing of the gene, rendering it resistant to DNA hypomethylating agents. High expression of the transcript in HuCC-T1 cells was found together with high levels of H3K4me3, H3K79me2, H3K9Ac and H3K9–14Ac, and low levels of H3K27me2 and H4K20me3. Moderate-to-low expression of the transcript was associated with a similar pattern with quantitative differences. Absence of the transcript (HCT-15 and MDA-MB-231 cells) was related to the opposite histone modifications. Although Caretti et al. suggest that resistance of methylated β1,3 galactosyltransferase β 3Gal-T5 to the action of DNA demethylating agents is caused by several chromatin modifications that repress the gene, this is not examined directly and there is no causal evidence presented in the paper for the role of these histone modifications in the resistance to demethylation using either inhibition or depletion strategies. Nor is it known whether these chromatin modifications are indeed resistant to the effect of DNA demethylating agents since this was not measured. An alternative possibility that needs to be considered is that DNA methylation of another regulatory region that was not demethylated in response to 5-azacytidine and that was not measured in this study was responsible for silencing of the gene in the resistant cell lines. It is well established that 5-azacytidine does not cause complete DNA demethylation and that certain sites are more resistant than others to demethylation.Since DNA demethylation agents are emerging as therapeutics in the clinic, it is important to understand and classify DNA demethylation resistant genes and their role in normal physiology in cancer progression and in the response to anticancer therapy.References1 Jones PA. Altering gene expression with 5-azacytidine. Cell40(3),485–486 (1985).Crossref, Medline, CAS, Google Scholar2 Eden S, Hashimshony T, Keshet I, Cedar H, Thorne AW. DNA methylation models histone acetylation [letter]. Nature394(6696),842 (1998).Crossref, Medline, CAS, Google Scholar3 Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB. Synergy of demethylation and histone deacetylase inhibition in the re- expression of genes silenced in cancer. Nat. Genet.21(1),103–107 (1999).Crossref, Medline, CAS, Google Scholar4 McInerney JM, Nawrocki JR, Lowrey CH. Long-term silencing of retroviral vectors is resistant to reversal by trichostatin A and 5-azacytidine. Gene Ther.7(8),653–663 (2000).Crossref, Medline, CAS, Google Scholar5 Caretti A, Sirchia SM, Tabano S, Zulueta A, Dall’olio F, Trinchera M. DNA methylation and histone modifications modulate the β1,3 galactosyltransferase β3Gal-T5 native promoter in cancer cells. Int. J. Biochem. Cell Biol.44(1),84–90 (2011).Crossref, Medline, Google ScholarEvaluation of: Gowers IR, Walters K, Kiss-Toth E, Read RC, Duff GW, Wilson AG. Age-related loss of CpG methylation in the tumour necrosis factor promoter. Cytokine 56(3), 792–797 (2011).It is well established that as we age there are changes to our immune and inflammatory functions that are implicated in age-related diseases such a rheumatoid arthritis. What is the mechanism responsible for these age-related changes in immune functionality? DNA methylation has been shown to drift during aging with a global loss of methylation [1] and an increased hypermethylation of CpG islands [2]. A plausible hypothesis is that a drift in DNA methylation lies behind the changes in immune function during aging. Gowers et al. examines in a recent study [3] the changes with age in the state of methylation of a critical proinflammatory mediator TNF in peripheral blood leucocytes and macrophages. Using pyrosequencing assays, the authors detected small but statistically significant age-related demethylation of CpG motifs (304, 245 and 239) in the TNF promoter in human peripheral blood cells from 312 healthy controls (0.8% per decade) and primary monocyte-derived macrophages from a separate population of 78 healthy controls (1.4% per decade). There were differences in age-related DNA demethylation between the 5´ and proximal promoter regions suggesting selectivity and targeting in the process of age-related demethylation even within the same promoter. Interestingly, genomic DNA methylation assessed using a LINE-1 methylation assay did not change with age in either peripheral blood leucocytes or monocyte-derived macrophages in contrast to previous reports of global demethylation with aging. Although there is a gender difference in the emergence of inflammatory diseases with aging, no gender difference is observed in the age-related decline in methylation in the TNF promoter.The study tests the plausibility that a gene encoding a critical inflammatory mediator changes its methylation pattern with aging. However, the study does not examine an association between the degree of methylation of the TNF promoter, aging and inflammatory disease such as rheumatoid arthritis. This needs to be performed to test the hypothesis that inflammatory disease is indeed associated with an age-dependent reduction in methylation of TNF promoter.One of the main problems that the data present is that although the decline in DNA methylation of certain CpG sites with age is significant, the extent of the change in DNA methylation is minute. Since DNA methylation is a binary signal, the change in DNA methylation per cell is either 100% if both alleles are affected or 50% if one chromosome is affected. Therefore, the percentage change in DNA methylation that is measured by pyrosequencing, represents the fraction of cells that become fully methylated or unmethylated at a given site. It is possible that age-related demethylation is not random across the blood cell populations that were analyzed, but that it is targeted to a specific subpopulation of cells that plays a critical role in the inflammatory response. This putative discrete subpopulation of cells would exhibit an ‘all-or-none’ change in DNA methylation. If this is true, identifying the specific cell subfraction that exhibits age-dependent demethylation of TNF will help us map the critical cells that are involved in mediating age-related changes in inflammatory functions that lead to age-related inflammatory disease.Financial & competing interests disclosureThe author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript.References1 Bollati V, Schwartz J, Wright R et al. Decline in genomic DNA methylation through aging in a cohort of elderly subjects. Mech. Ageing Dev.130(4),234–239 (2009).Crossref, Medline, CAS, Google Scholar2 Ahuja N, Li Q, Mohan AL, Baylin SB, Issa JP. Aging and DNA methylation in colorectal mucosa and cancer. Cancer Res.58(23),5489–5494 (1998).Medline, CAS, Google Scholar3 Gowers IR, Walters K, Kiss-Toth E, Read RC, Duff GW, Wilson AG. Age-related loss of CpG methylation in the tumour necrosis factor promoter. Cytokine56(3),792–797 (2011).Crossref, Medline, CAS, Google ScholarFiguresReferencesRelatedDetailsCited ByDNA methylation and transcriptome comparative analysis for Lvliang Black goats in distinct feeding pattern reveals epigenetic basis for environment adaptation4 June 2021 | Biotechnology & Biotechnological Equipment, Vol. 35, No. 1DNA Ligase-Based Strategy for Quantifying Heterogeneous DNA Methylation without Sequencing1 January 2015 | Clinical Chemistry, Vol. 61, No. 1microRNAs are important players in head and neck carcinoma: A reviewCritical Reviews in Oncology/Hematology, Vol. 88, No. 3 Vol. 4, No. 1 Follow us on social media for the latest updates Metrics History Published online 14 February 2012 Published in print February 2012 Information© Future Medicine LtdPDF download