Excessive CpG island hypermethylation in cancer cell lines versus primary human malignancies

CpG island methylation and expression of tumour-associated genes in lung carcinoma

In this study, microarray analysis was used to identify tumour-related genes that were down regulated in lung carcinoma. The promoter sequences of the identified genes were analysed for methylation patterns. In lung cancer cell lines, CpG island methylation was frequently detected for TIMP4 (64%), SOX18 (73%), EGF-like domain 7 (56%), CD105 (71%), SEMA2 (55%), RASSF1A (71%), p16 (56%) SLIT2 (100%) and TIMP3 (29%). Methylation was however rarely observed in cell lines for SLIT3 (18%) and DLC1 (18%). In primary lung tumours, methylation of TIMP4 (94%), SOX18 (100%), EGF-like domain 7 (100%), CD105 (69%), SEMA2 (93%), DLC1 (61%), RASSF1A (44%), p16 (47%), SLIT2 (100%) and TIMP3 (13%) was also detected. Methylation of several CpG islands was frequently found in normal lung tissue of cancer patients and this may have been attributed to epigenetic field defect and/or infiltrating tumour cells. Interestingly, inactivation of RASSF1A and p16 correlated well with an extended smoking habit (P=0.02), and exposure to asbestos (P=0.017) or squamous cell carcinoma (P=0.011), respectively. These results have identified genes whose aberrant promoter methylation could play a crucial role in the malignancy of lung carcinoma.

Analysis of somatic NF1 promoter methylation in plexiform neurofibromas and Schwann cells

Neurofibromatosis 1 (NF1) is an autosomal dominant disorder with the characteristic feature being the neurofibroma. It is believed that both NF1 alleles must be inactivated as the first step in tumorigenesis. However, often the somatic mutations are not identified, suggesting that epigenetic changes such as methylation could account for the "second hit" in some tumors. The literature reports that the region of the NF1 promoter surrounding the transcription start site is completely unmethylated in several normal tissues and some NF1-related dermal and plexiform neurofibromas. We analyzed the methylation state of the NF1 promoter in normal Schwann cells (the cell type clonally expanded in neurofibromas) and in NF1-related plexiform tumor samples with unidentified somatic mutations. In a region of 451 bp surrounding the transcription start site, a low level of methylation was found at several specific cytosines in 12 of 18 tumor samples. Overall, epigenetic silencing through methylation does not appear to be a major mechanism for the second hit. However, this study, which analyzed the largest number of NF1-related plexiform tumors and is the first to include Schwann cell-enriched tumor cultures, detected greater methylation than in any previous reports. This suggests that methylation, especially at potential transcription factor binding sites, is moderately perturbed in some plexiform neurofibromas and should be investigated further.

The clinical application of targeting cancer through histone acetylation and hypomethylation

Methods of gene inactivation include genetic events such as mutations or deletions. Epigenetic changes, heritable traits that are mediated by changes in DNA other than nucleotide sequences, play an important role in gene expression. Two epigenetic events that have been associated with transcriptional silencing include methylation of CpG islands located in gene promoter regions of cancer cells and changes in chromatin conformation involving histone acetylation. Recent evidence demonstrates that these processes form layers of epigenetic silencing. Reversal of these epigenetic processes and up-regulation of genes important to prevent or reverse the malignant phenotype has therefore become a new therapeutic target in cancer treatment.

DNA methylation changes in multiple myeloma

Using a candidate gene approach, we analyzed the methylation status of the promoter-associated CpG islands of 11 well-characterized tumor suppressor genes by methylation-specific polymerase chain reaction in five multiple myeloma (MM) cell lines and 56 patients with malignant plasma cell disorders. The frequency of aberrant methylation among the patient samples was 46.4% for SOCS-1, 35.7% for p16, 21.4% for E-cadherin, 12.5% for DAP kinase and p73, 1.8% for p15, MGMT as well as RARbeta, and 0% for TIMP-3, RASSF1A and hMLH1. We found at least one hypermethylated gene in 80.4% of the primary patient samples, while 33.9% harbored two or more hypermethylated genes. For the first time, we show that p73 may be hypermethylated in MM and thus be involved in the pathogenesis of plasma cell disorders. Hypermethylation of p16 at diagnosis was associated with a poorer prognosis. In patients with plasma cell leukemia, we found frequent simultaneous hypermethylation of p16, E-cadherin and DAP kinase. We conclude that aberrant methylation of tumor suppressor genes is a common event in malignant plasma cell disorders and that there is a correlation between methylation patterns and clinical characteristics in MM patients.

A CpG island hypermethylation profile of primary colorectal carcinomas and colon cancer cell lines

Background

Tumor cell lines are commonly used as experimental tools in cancer research, but their relevance for the in vivo situation is debated. In a series of 11 microsatellite stable (MSS) and 9 microsatellite unstable (MSI) colon cancer cell lines and primary colon carcinomas (25 MSS and 28 MSI) with known ploidy stem line and APC, KRAS, and TP53 mutation status, we analyzed the promoter methylation of the following genes: hMLH1, MGMT, p16^INK4a (CDKN2A α-transcript), p14^ARF (CDKN2A β-transcript), APC, and E-cadherin (CDH1). We compared the DNA methylation profiles of the cell lines with those of the primary tumors. Finally, we examined if the epigenetic changes were associated with known genetic markers and/or clinicopathological variables.

Results

The cell lines and primary tumors generally showed similar overall distribution and frequencies of gene methylation. Among the cell lines, 15%, 50%, 75%, 65%, 20% and 15% showed promoter methylation for hMLH1, MGMT, p16^INK4a, p14^ARF, APC, and E-cadherin, respectively, whereas 21%, 40%, 32%, 38%, 32%, and 40% of the primary tumors were methylated for the same genes. hMLH1 and p14^ARF were significantly more often methylated in MSI than in MSS primary tumors, whereas the remaining four genes showed similar methylation frequencies in the two groups. Methylation of p14^ARF, which indirectly inactivates TP53, was seen more frequently in tumors with normal TP53 than in mutated samples, but the difference was not statistically significant. Methylation of p14^ARF and p16^INK4a was often present in the same primary tumors, but association to diploidy, MSI, right-sided location and female gender was only significant for p14^ARF. E-cadherin was methylated in 14/34 tumors with altered APC further stimulating WNT signaling.

Conclusions

The present study shows that colon cancer cell lines are in general relevant in vitro models, comparable with the in vivo situation, as the cell lines display many of the same molecular alterations as do the primary carcinomas. The combined pattern of epigenetic and genetic aberrations in the primary carcinomas reveals associations between them as well as to clinicopathological variables, and may aid in the future molecular assisted classification of clinically distinct stages.

Apoptotic Speck Protein-Like, a Highly Homologous Protein to Apoptotic Speck Protein in the Pyrin Domain, Is Silenced by DNA Methylation and Induces Apoptosis in Human Hepatocellular Carcinoma

We have identified a novel gene encoding a pyrin domain protein of 89 amino acids that is expressed in various tissues including liver, brain, and spleen. The protein is highly homologous to the pyrin domain of apoptosis-associated speck-like protein (ASC). Therefore, we termed it ASC-like (ASCL). We found that ASCL gene was densely and frequently (80%) methylated in hepatocellular carcinoma (HCC) cell lines. In contrast, normal liver samples did not show any significant methylation. This aberrant methylation correlated well with the suppression of RNA expression. Furthermore, a demethylating agent, 5-aza-2'-deoxycytidine, reactivated the ASCL expression in the methylation-silenced cells, indicating that ASCL is silenced by the associated DNA methylation. ASCL methylation was also found in primary HCC (4 of 17 samples), although the frequency was less than that in cell lines. In addition, we found that ASC was also methylated in primary samples (6 of the 17). Interestingly, either ASCL or ASC methylation was observed in 53% (9 of the 17) of primary HCC samples. Significantly, the restoration of ASCL in the methylation-silenced cells demonstrated growth suppression in colony formation assay. This growth suppression effect of ASCL was supported by apoptotic changes observed in ASCL-transfected cells in which annexin-V binding was positive and caspase-3 was activated. Based on the methylation-silencing and the growth suppression activity, we propose that ASCL plays a significant role in the development of HCC.

Progesterone receptor gene inactivation and CpG island hypermethylation in human leukemia cancer cells

Previous studies showed that progesterone receptor (PR), one of the hormone receptor superfamily, was only connected with the sex-correlated cancers such as breast cancer, endometrial cancer, prostate cancer, etc. This article deals with the PR gene in leukemia. We investigated the methylation status and the expression of the two different PR isoforms, PRA and PRB, in three leukemia cancer cell lines using methylation-specific polymerase chain reaction (MSP-PCR) and reverse transcription-PCR. The correlation of PR methylation and expression together with DNA methyltransferase (DNMT1) was further studied. We found that DNMT1 is required to maintain CpG methylation and aberrant gene silencing of PR gene in human leukemia cancer cells. The activity of 5-aza-2'-deoxycytidine in demethylation and gene reactivation may be through depleting cellular DNMT1 levels. In addition, extensive methylation of PRA and PRB was also observed in leukemia samples. Our results suggest that PR CpG island aberrant hypermethylation could be one molecular and genetic alteration in leukemia.

Implication of the INK4a/ARF locus in gastroenteropancreatic neuroendocrine tumorigenesis

The INK4a/ARF locus on chromosome 9p21 is one of the important defenses against tumor development and engages both the Rb and the p53 tumor suppressor pathways through its capacity to encode two distinct proteins, p16(INK4a) and p14(ARF). Despite controversial reports, the body of present data suggests that tumor suppressors p16(INK4a) and p14(ARF) are targets of in-activation in GEP-NETs. Moreover, tumor type-specific aberrant p16(INK4a) silencing appears to be associated with advanced tumor stage and may function as a predictor of patients' outcome after surgical resection. Since conventional histological and biochemical assessment are limited with respect to predicting GEP-NET behavior or outcome, methylation profiles including INK4a/ARF might offer a tool to refine future diagnosis and therapeutic management of GEP-NET patients.

Profiling epigenetic inactivation of tumor suppressor genes in tumors and plasma from cutaneous melanoma patients

Aberrant methylation of CpG islands in promoter regions of tumor suppressor genes (TSG) has been demonstrated in epithelial origin tumors. However, the methylation profiling of tumor-related gene promoter regions in cutaneous melanoma tumors has not been reported. Seven known or candidate TSGs that are frequently hypermethylated in carcinomas were assessed by methylation-specific polymerase chain reaction (MSP) in 15 melanoma cell lines and 130 cutaneous melanoma tumors. Four TSGs were frequently hypermethylated in 86 metastatic tumor specimens: retinoic acid receptor-beta2 (RAR-beta2) (70%), RAS association domain family protein 1A (RASSF1A) (57%), and O6-methylguanine DNA methylatransferase (MGMT) (34%), and death-associated protein kinase (DAPK) (19%). Hypermethylation of MGMT, RASSF1A, and DAPK was significantly lower in primary melanomas (n=20) compared to metastatic melanomas. However, hypermethylation of RAR-beta2 was 70% in both primary and metastatic melanomas. Cell lines had hypermethylation profiles similar to those of metastatic melanomas. The analysis of these four markers of metastatic tumors demonstrated that 97% had > or =1 gene(s) and 59% had > or =2 genes hypermethylated. The methylation of genes was verified by bisulfite sequencing. The mRNA transcripts could be re-expressed in melanoma cell lines having hypermethylated genes following treatment with 5'-aza 2'-deoxycytidine (5Aza-dC). Analysis of melanoma patients' plasma (preoperative blood; n=31) demonstrated circulating hypermethylated MGMT, RAR-beta2, and RASSF1A DNA for at least one of the markers in 29% of the patients. Our findings indicate that the incidence of TSG hypermethylation increases during tumor progression. Methylation of TSG may play a significant role in cutaneous melanoma progression.

DNA methylation may restrict but does not determine differential gene expression at the Sgy/Tead2 locus during mouse development

Soggy (Sgy) and Tead2, two closely linked genes with CpG islands, were coordinately expressed in mouse preimplantation embryos and embryonic stem (ES) cells but were differentially expressed in differentiated cells. Analysis of established cell lines revealed that Sgy gene expression could be fully repressed by methylation of the Sgy promoter and that DNA methylation acted synergistically with chromatin deacetylation. Differential gene expression correlated with differential DNA methylation, resulting in sharp transitions from methylated to unmethylated DNA at the open promoter in both normal cells and tissues, as well as in established cell lines. However, neither promoter was methylated in normal cells and tissues even when its transcripts were undetectable. Moreover, the Sgy promoter remained unmethylated as Sgy expression was repressed during ES cell differentiation. Therefore, DNA methylation was not the primary determinant of Sgy/Tead2 expression. Nevertheless, Sgy expression was consistently restricted to basal levels whenever downstream regulatory sequences were methylated, suggesting that DNA methylation restricts but does not regulate differential gene expression during mouse development.

Identification of maspin and S100P as novel hypomethylation targets in pancreatic cancer using global gene expression profiling

DNA hypomethylation is one of the major epigenetic alterations in human cancers. We have previously shown that genes identified as hypomethylated in pancreatic cancer are expressed in pancreatic cancer cell lines, but not in normal pancreatic ductal epithelium and can be reexpressed in nonexpressing cells using 'epigenetic modifying agents' such as DNA methyltransferase inhibitors. To identify additional targets for aberrant hypomethylation in pancreatic cancer, we used oligonucleotide microarrays to screen for genes that displayed expression patterns associated with hypomethylation. This analysis identified a substantial number of candidates including previously reported hypomethylated genes. A subset of eight genes were selected for further methylation analysis, and two cancer-related genes, maspin and S100P, were found to be aberrantly hypomethylated in a large fraction of pancreatic cancer cell lines and primary pancreatic carcinomas. Combined treatment with 5-aza-2'-deoxycytidie and trichostatin A resulted in synergistic induction of maspin and S100P mRNA in MiaPaCa2 cells where both genes were methylated. Furthermore, there was an inverse correlation between methylation and mRNA expression level for maspin and S100P in a large panel of pancreatic cancer cell lines. We also found a significant difference in the methylation patterns of maspin and two previously identified hypomethylated genes (trefoil factor 2 and lipocalin 2) between pancreatic and breast cancer cell lines, suggesting cancer-type specificity for some hypomethylation patterns. Thus, our present results confirm that DNA hypomethylation is a frequent epigenetic event in pancreatic cancer, and suggest that gene expression profiling may help to identify potential targets affected by this epigenetic alteration.

DNA methylation controls the responsiveness of hepatoma cells to leukemia inhibitory factor

The related members of the interleukin 6 (IL-6) family of cytokines, IL-6, leukemia inhibitory factor (LIF), and oncostatin M, act as major inflammatory mediators and induce the hepatic acute phase reaction. Normal parenchymal liver cells express the receptors for these cytokines, and these receptors activate, to a comparable level, the intracellular signaling through signal transducer and activator of transcription (STAT) proteins and extracellular-regulated kinase (ERK). In contrast, hepatoma cell lines show attenuated responsiveness to some of these cytokines that is correlated with lower expression of the corresponding ligand-binding receptor subunits. This study tests the hypothesis that the reduced expression of LIF receptor (LIFR) observed in hepatoma cells is mediated by altered DNA methylation. H-35 rat hepatoma cells that have a greatly reduced LIF responsiveness were treated with 5-aza-2'-deoxycytidine, an inhibitor of DNA methyltransferase. Surviving and proliferating cells showed reestablished expression of LIFR protein and function. Restriction landmark genomic scanning (RLGS) demonstrated genome-wide drug-induced alterations in DNA methylation status, with striking similarities in the demethylation pattern among independently derived clonal lines. Upon extended growth in the absence of 5-aza-2'-deoxycytidine, the cells exhibit partial reversion to pretreatment patterns. Demethylation and remethylation of the CpG island within the LIFR promoter that is active in normal liver cells correlate with increased and decreased usage of this promoter in H-35 cells. In conclusion, these results indicate that transformed liver cells frequently undergo epigenetic alterations that suppress LIFR gene expression and modify the responsiveness to this IL-6 type cytokine.

Epigenetically mediated loss of UDP-GlcNAc 2-epimerase/ManNAc kinase expression in hyposialylated cell lines

The bifunctional enzyme UDP-GlcNAc 2-epimerase/ManNAc kinase is the key enzyme in sialic acid biosynthesis. Loss of UDP-GlcNAc 2-epimerase activity results in a hyposialylated phenotype as shown for two human hematopoietic cell lines that lack the specific mRNA. We found that treatment with the DNA methylation inhibitor 5'-aza-2'-deoxycytidine (5-aza-dC) restored the UDP-GlcNAc 2-epimerase/ManNAc kinase mRNA, as well as enzyme activity and cell surface sialylation. Increase of UDP-GlcNAc 2-epimerase activity by 5-aza-dC treatment was also found for a rat Morris hepatoma cell line. These results indicate a regulation of UDP-GlcNAc 2-epimerase/ManNAc kinase expression on the transcriptional level by DNA methylation.

The multiple promoter methylation profile of PR gene and ERalpha gene in tumor cell lines

The changes of methylation status of various gene promoters are a common feature of malignant cells and these changes can occur early in the progression process. Therefore, abnormal methylation can be used as cancer marker. Such studies will first require the development of a panel of methylated markers that are methylated in cancer tissues but unmethylated in normal tissues or methylated status is different between cancer tissues and normal tissues. By using methylation-specific PCR (MSP) assay method, we observed alterations in DNA methylation at the double promoter regions of the progesterone receptor (PR) gene and estrogen receptor (ERalpha) gene in various tumor cell lines. Compared with normal white blood cell, the methylation status of PRA promoter in various cancer cell lines changed from unmethylation pattern to methylation pattern. That of PRB promoter changed from both unmethylated and methylated alleles to only methylated allele. The methylation status of ERalpha-A and ERalpha-B promoter in various cancer cell lines are cell -specific. This study indicates that PR promoter methylation may be a molecular marker in various cancer detections. And the methylation status of ERalpha-A and ERalpha-B is cell-specific.

Hypermethylation of the spleen tyrosine kinase promoter in T-lineage acute lymphoblastic leukemia

Sequence analysis of the noncoding first exon (exon 1) of the Syk gene demonstrated the presence of a previously cloned CpG island (GenBank #Z 65706). Transient transfection analysis in Daudi cells demonstrated promoter activity (18-fold increase over parental luciferase plasmid) for a 348 bp BstXI-BsrBI fragment containing this island. This region exhibits a high GC content (approximately 75%), contains several SP1 binding sites and a potential initiator sequence, but lacks a strong TATA consensus. Bisulfite sequencing and methylation-specific PCR (MSP) of this region demonstrated that the Syk promoter CpG island was largely unmethylated in B-lineage leukemia cell lines, control peripheral blood cells, human thymocytes and CD3(+) T lymphocytes. However, dense methylation was seen in four T-lineage leukemia cell lines, Jurkat, H9, Molt 3 and HUT 78. MSP screening of leukemia cells from six T-lineage acute lymphoblastic leukemia (ALL) patients demonstrated methylation of the Syk promoter CpG island in one T-lineage ALL patient. Promoter methylation was correlated with reduced to absent expression of Syk mRNA and SYK protein in the T-lineage leukemia cell lines. Treatment of the leukemia lines Ha and Molt 3, with the methylation inhibitor, 5-aza-2'-deoxycytidine (5-aza-CdR) resulted in increased Syk mRNA expression. The presence of a methylated promoter sequence in these T-lineage leukemia cell lines and in one T-lineage patient suggests a potential role for SYK as a tumor suppressor in T-ALL.

Unanswered questions about the role of promoter methylation in carcinogenesis

It has been proposed that tumor suppressor genes can be silenced by ectopic de novo methylation during tumor progression and that this epigenetic silencing is an alternative to mutation in tumor suppressor inactivation during oncogenic transformation. However, methylation may follow inactivation and may not directly participate in tumor progression. There are no genetic data that implicate ectopic de novo methylation in cancer, and no DNA methyltransferase gene has been shown to be mutated in any cancer. Promoter methylation at tumor suppressor loci may be a consequence of transcriptional inactivity imposed by mutations in upstream components of the transcriptional machinery or signal transduction pathways. Current estimates of the importance of epigenetic changes in the etiology of cancer may be inflated, and consequences may have been mistaken for causes in some cases.

CpG island methylation is a common finding in colorectal cancer cell lines

Tumour cell lines are commonly used in colorectal cancer (CRC) research, including studies designed to assess methylation defects. Although many of the known genetic aberrations in CRC cell lines have been comprehensively described, no studies have been performed on their methylation status. In this study, 30 commonly used CRC cell lines as well as seven primary tumours from individuals with hereditary nonpolyposis colorectal cancer (HNPCC) were assessed for methylation at six CpG islands known to be hypermethylated in colorectal cancer: hMLH1, p16, methylated in tumour (MINT-)-1, -2, -12 and -31. The cell lines were also assessed for microsatellite instability (MSI), ploidy status, hMLH1 expression, and mutations in APC and Ki-ras. Methylation was frequently observed at all examined loci in most cell lines, and no differences were observed between germline-derived and sporadic cell lines. Methylation was found at MINT 1 in 63%, MINT 2 in 57%, MINT 12 in 71%, MINT 31 in 53%, p16 in 71%, and hMLH1 in 30% of cell lines. Overall only one cell line, SW1417, did not show methylation at any locus. Methylation was found with equal frequency in MSI and chromosomally unstable lines. MSI was over-represented in the cell lines relative to sporadic CRC, being detected in 47% of cell lines. The rate of codon 13 Ki-ras mutations was also over three times that expected from in vivo studies. We conclude that CpG island hypermethylation, whether acquired in vivo or in culture, is a ubiquitous phenomenon in CRC cell lines. We suggest that CRC cell lines may be only representative of a small subset of real tumours, and this should be taken into account in the use of CRC cell lines for epigenetic studies.

Oligonucleotide-based microarray for DNA methylation analysis: principles and applications

Gene silencing via promoter CpG island hypermethylation offers tumor cells growth advantages. This epigenetic event is pharmacologically reversible, and uncovering a unique set of methylation-silenced genes in tumor cells can bring a new avenue to cancer treatment. However, high-throughput tools capable of surveying the methylation status of multiple gene promoters are needed for this discovery process. Herein we describe an oligonucleotide-based microarray technique that is both versatile and sensitive in revealing hypermethylation in defined regions of the genome. DNA samples are bisulfite-treated and PCR-amplified to distinguish CpG dinucleotides that are methylated from those that are not. Fluorescently labeled PCR products are hybridized to arrayed oligonucleotides that can discriminate between methylated and unmethylated alleles in regions of interest. Using this technique, two clinical subtypes of non-Hodgkin's lymphomas, mantle cell lymphoma, and grades I/II follicular lymphoma, were further separated based on the differential methylation profiles of several gene promoters. Work is underway in our laboratory to extend the interrogation power of this microarray system in multiple candidate genes. This novel tool, therefore, holds promise to monitor the outcome of various epigenetic therapies on cancer patients.

Profile of methylated CpG sites of hMLH1 promoter in primary gastric carcinoma with microsatellite instability

A recent study using colorectal cancer cell lines has identified methylation on a small region of hMLH1 promoter (-248 to -178 relative to the transcription start site) to be critical for gene silencing, but shown that methylation on a more upstream region is frequent in cell lines with hMLH1 expression. Because cultured cell lines have a higher degree of CpG methylation than primary tumors, we attempted to examine methylation profiles of CpG sites of hMLH1 promoter in primary gastric carcinomas with or without microsatellite instability (MSI). Seven cases with MSI and six cases without MSI were assessed for the methylation status of hMLH1 promoter by bisulfite-sequencing. All of the MSI-positive cases previously showed loss of hMLH1 expression and six cases displayed methylated alleles in methylation-specific PCR (MSP) for hMLH1. Sequencing analysis revealed that: (i) CpG sites were overall methylated in MSI-positive tumors with positive MSP results; (ii) a small region (-248-178) was almost invariably methylated in MSI-positive tumors; and (iii) the vast majority of CpG sites were unmethylated in MSI-negative tumors, including a more upstream region (proximal to -248). Our study suggests that methylation of a more upstream region observed in colon cancer cell lines may be an acquired change during cell line establishment and it was not identified in primary gastric carcinomas without MSI.

Integrated genomic and epigenomic analyses pinpoint biallelic gene inactivation in tumors

Aberrant methylation of CpG islands and genomic deletion are two predominant mechanisms of gene inactivation in tumorigenesis, but the extent to which they interact is largely unknown. The lack of an integrated approach to study these mechanisms has limited the understanding of tumor genomes and cancer genes. Restriction landmark genomic scanning (RLGS; ref. 1) is useful for global analysis of aberrant methylation of CpG islands, but has not been amenable to alignment with deletion maps because the identity of most RLGS fragments is unknown. Here, we determined the nucleotide sequence and exact chromosomal position of RLGS fragments throughout the genome using the whole chromosome of origin of the fragments and in silico restriction digestion of the human genome sequence. To study the interaction of these gene-inactivation mechanisms in primary brain tumors, we integrated RLGS-based methylation analysis with high-resolution deletion maps from microarray-based comparative genomic hybridization (array CGH; ref. 3). Certain subsets of gene-associated CpG islands were preferentially affected by convergent methylation and deletion, including genes that exhibit tumor-suppressor activity, such as CISH1 (encoding SOCS1; ref. 4), as well as genes such as COE3 that have been missed by traditional non-integrated approaches. Our results show that most aberrant methylation events are focal and independent of deletions, and the rare convergence of these mechanisms can pinpoint biallelic gene inactivation without the use of positional cloning.

An AscI boundary library for the studies of genetic and epigenetic alterations in CpG islands

Knudson's two-hit hypothesis postulates that genetic alterations in both alleles are required for the inactivation of tumor-suppressor genes. Genetic alterations include small or large deletions and mutations. Over the past years, it has become clear that epigenetic alterations such as DNA methylation are additional mechanisms for gene silencing. Restriction Landmark Genomic Scanning (RLGS) is a two-dimensional gel electrophoresis that assesses the methylation status of thousands of CpG islands. RLGS has been applied successfully to scan cancer genomes for aberrant DNA methylation patterns. So far, the majority of this work was done using NotI as the restriction landmark site. Here, we describe the development of RLGS using AscI as the restriction landmark site for genome-wide scans of cancer genomes. The availability of AscI as a restriction landmark for RLGS allows for scanning almost twice as many CpG islands in the human genome compared with using NotI only. We describe the development of an AscI-EcoRV boundary library that supports the cloning of novel methylated genes. Feasibility of this system is shown in three tumor types, medulloblastomas, lung cancers, and head and neck cancers. We report the cloning of 178 AscI RLGS fragments via two methods by use of this library.

Restriction landmark genomic scanning for DNA methylation in cancer: past, present, and future applications

The field of molecular biology was revolutionized by the advent of gel electrophoresis. Restriction landmark genomic scanning (RLGS) is a type of two-dimensional electrophoresis employed in the genome-wide assessment of genomic alterations. RLGS has been used to study genetic and epigenetic changes in normal tissues, primary tumors, cancer cell lines, and various organisms such as mice, rats, hamsters, bacteria, and plants. An RLGS profile displays over 2000 radiolabeled restriction landmark sites in a single assay. When conducted with methylation-sensitive restriction enzymes whose sites are preferentially located in CpG island regulatory regions, RLGS becomes a very versatile tool for the investigation of both normal and aberrant methylation patterns. Early studies performed on tumor DNA were mainly descriptive in nature, essentially a catalogue of loci that were changed to varying degrees in different tumor types. Over time, as investigators have become more proficient with RLGS and have undertaken high-throughput studies, the need for efficient cloning, imaging, and analysis systems has become paramount. Current studies focus on identifying specific genes and pathways involved in deregulated methylation in cancer. As such, RLGS analysis of tumor samples has made tremendous contributions to our understanding of the role of DNA methylation in cancer. Future directions will take advantage of the abundant genomic sequence data available to link all of the RLGS loci to genes and create biologically relevant methylation profiles of cancer. This review discusses practical considerations of using RLGS as a genome scanning tool and the past, present, and future applications in cancer biology.

Aberrant methylation of the CDKN2a/p16INK4a gene promoter region in preinvasive bronchial lesions: a prospective study in high-risk patients without invasive cancer

Among the identified factors involved in malignant transformation, abnormal methylation of the CDKN2A/p16(INK4a) gene promoter has been described as an early event, particularly in bronchial cell cancerization. Precancerous bronchial lesions (n = 70) prospectively sampled during fluorescence endoscopy in a series of 37 patients at high risk for lung cancer were studied with respect to the methylation status of the CDKN2A gene. Methylation-specific polymerase chain reaction was performed on DNA extracted from pure bronchial cell populations derived from biopsies and detection of p16 protein was studied by immunohistochemistry on contiguous parallel biopsies. Aberrant methylation of the CDKN2A gene promoter was found in 19% of preinvasive lesions and its frequency increased with the histologic grade of the lesions. Methylation in at least 1 bronchial site was significantly more frequent in patients with cancer history, although there was no difference in the outcome of patients with or without methylation in bronchial epithelium. The other risk factors studied (tobacco and asbestos exposure) did not influence the methylation status. There was no relationship between CDKN2A methylation and the evolutionary character of the lesions. Our results confirm that abnormal methylation of the CDKN2A gene promoter is an early event in bronchial cell cancerization, which can persist for several years after carcinogen exposure cessation, and show that this epigenetic alteration cannot predict the evolution of precancerous lesions within a 2-year follow-up.

Methylation of the RASSF1A gene in human cancers

Loss of genetic material from chromosome 3p21.3 is one of the most common and earliest events in the pathogenesis of lung cancer and many other solid tumors. The chromosomal area 3p21.3 is thought to harbor at least one important tumor suppressor gene, which, despite many years of investigation, has remained elusive. In our previous studies, we have identified and cloned a gene from the common homozygous deletion area at 3p21.3. The gene, named RASSF1A (Ras ASSociation domain Family 1A), has homology to a mammalian Ras effector. The RASSF1A gene is epigenetically inactivated in a large percentage of human lung cancers, in particular small cell carcinomas. A high frequency of methylation of RASSF1A is found also in breast cancers, renal cell carcinomas, ovarian, gastric and bladder cancers, and in neuroblastomas. The RASSF1A gene is a candidate for a tumor suppressor gene in 3p21.3.

A genomic screen for genes upregulated by demethylation and histone deacetylase inhibition in human colorectal cancer

Aberrant hypermethylation of gene promoters is a major mechanism associated with inactivation of tumor-suppressor genes in cancer. We previously showed this transcriptional silencing to be mediated by both methylation and histone deacetylase activity, with methylation being dominant. Here, we have used cDNA microarray analysis to screen for genes that are epigenetically silenced in human colorectal cancer. By screening over 10,000 genes, we show that our approach can identify a substantial number of genes with promoter hypermethylation in a given cancer; these are distinct from genes with unmethylated promoters, for which increased expression is produced by histone deacetylase inhibition alone. Many of the hypermethylated genes we identified have high potential for roles in tumorigenesis by virtue of their predicted function and chromosome position. We also identified a group of genes that are preferentially hypermethylated in colorectal cancer and gastric cancer. One of these genes, SFRP1, belongs to a gene family; we show that hypermethylation of four genes in this family occurs very frequently in colorectal cancer, providing for (i) a unique potential mechanism for loss of tumor-suppressor gene function and (ii) construction of a molecular marker panel that could detect virtually all colorectal cancer.

Altered methylation patterns in cancer cell genomes: cause or consequence?

CpG islands are associated with at least half of all cellular genes and are normally methylation-free. Dense methylation of cytosine residues within islands causes strong and heritable transcriptional silencing. Such silencing normally occurs almost solely at genes subject to genomic imprinting or to X chromosome inactivation. Aberrant methylation of CpG islands associated with tumor suppressor genes has been proposed to contribute to carcinogenesis. However, questions of mechanisms underlying the cancer changes and the precise consequences for tumorigenesis exist in the field, and must continue to be addressed before the importance of abnormalities in genomic methylation patterns in carcinogenesis can be fully understood. In this article, two workers in DNA methylation, one concentrating on cancer biology and the other on developmental biology, address recurrent questions about cancer epigenetics from different perspectives. The goal is to highlight important controversies in the field which can be productive targets of ongoing and future research.

Variable promoter region CpG island methylation of the putative tumor suppressor gene Connexin 26 in breast cancer

Intercellular communication via gap junctions is a mechanism for tumor suppression. Connexin 26 (Cx26) is a structural component of gap junctions expressed by breast epithelial cells. Expression levels of Cx26 are reduced in many breast tumors. Methylation-sensitive single-stranded conformation analysis showed variable methylation in the promoter region CpG island in 11 out of 20 (55%) breast cancer patients. Heterogeneity in methylation patterns was observed both between and within tumors. The degree of methylation ranged from a few CpG dinucleotides to almost all the CpG dinucleotides in the analyzed region. The most frequently methylated CpG was in an Sp1 site known to be important for Cx26 gene expression. One of eight breast cancer cell lines (MD-MBA-453) was methylated in the promoter region and did not express Cx26. Treatment of MDA-MB-453 with 5-aza-2'-deoxycytidine resulted in the re-expression of Cx26 mRNA. Methylation of the promoter region is likely to be an important mechanism in modulating the expression of Cx26 in breast cancer.

Global and gene-specific methylation patterns in cancer: aspects of tumor biology and clinical potential

Heritable alterations of DNA that do not affect the base pair sequence itself but nevertheless regulate the predetermined activity of genes are referred to as epigenetic. Epigenetic mechanisms comprise diverse phenomena including stable feedback loops, nuclear compartmentalization, differential replication timing, heritable chromatin structures, and, foremost, DNA cytosine methylation (1-3). DNA cytosine methylation has recently gained major attention in the field of basic molecular biology as well as in studies of human diseases including cancer. Changes in DNA methylation patterns in human malignancies have been shown to contribute to carcinogenesis in multiple ways. Both hypo- and hypermethylation events have been described in various neoplasias leading to chromosomal instability and transcriptional gene silencing. DNA methylation research has entered the clinical arena and methylation patterns have become a major focus of clinicians seeking novel prognostic factors and therapeutic targets. The following minireview covers aspects of the basic molecular biology of DNA methylation and summarizes its importance in human cancers.

DNA methylation, imprinting and cancer

It is well known that a variety of genetic changes influence the development and progression of cancer. These changes may result from inherited or spontaneous mutations that are not corrected by repair mechanisms prior to DNA replication. It is increasingly clear that so called epigenetic effects that do not affect the primary sequence of the genome also play an important role in tumorigenesis. This was supported initially by observations that cancer genomes undergo changes in their methylation state and that control of parental allele-specific methylation and expression of imprinted loci is lost in several cancers. Many loci acquiring aberrant methylation in cancers have since been identified and shown to be silenced by DNA methylation. In many cases, this mechanism of silencing inactivates tumour suppressors as effectively as frank mutation and is one of the cancer-predisposing hits described in Knudson's two hit hypothesis. In contrast to mutations which are essentially irreversible, methylation changes are reversible, raising the possibility of developing therapeutics based on restoring the normal methylation state to cancer-associated genes. Development of such therapeutics will require identifying loci undergoing methylation changes in cancer, understanding how their methylation influences tumorigenesis and identifying the mechanisms regulating the methylation state of the genome. The purpose of this review is to summarise what is known about these issues.