Gene silencing in phenomena related to DNA repair

Promoter methylation status of MGMT, hMSH2, and hMLH1 and its relationship to corresponding protein expression and TP53 mutations in human esophageal squamous cell carcinoma

To determine the relevance of O-6-methylguanine-DNA methyltransferase (MGMT), human mutS homolog 2 (hMSH2), and human mutL homolog 1 (hMLH1) in TP53 mutations in esophageal squamous cell carcinoma, we employed methylation-sensitive high-resolution melting technology and methylation-specific polymerase chain reaction (PCR) to analyze promoter hypermethylation of MGMT, hMSH2, and hMLH1, respectively, in 51 paired tumors and their adjacent normal tissues. The protein expression of the three proteins was also evaluated by Western blot analysis, and the PCR products of TP53, from exon 5 to exon 8, were directly sequenced to measure the mutation spectrum. Esophageal tumor tissues embraced statistically higher MGMT and hMSH2 promoter methylation level than normal tissue. The promoter methylation status of MGMT and hMSH2 corresponds positively with the protein expression level of MGMT and hMSH2. However, such relevance was not found for hMLH1. Furthermore, TP53 mutation status was well associated with MGMT and hMSH2 promoter methylation status, indicating that silencing of the two genes could lead to TP53 mutation in ESCC.

Methylation variable position profiles of hMLH1 promoter CpG islands in human sporadic colorectal carcinoma

Aberrant hypermethylation of CpG islands (CGIs) in hMLH1 promoter regions has been well known to play an important role in the tumorigenesis of human sporadic colorectal carcinoma (SCRC). In this study, bisulfite sequencing was performed to analyze the methylation variable positions (MVPs) profiles of hMLH1 promoter CGIs in 30 clinical SCRC patients, and further analysis was carried out to evaluate the associations between the CGI methylation and the clinicopathological features in SCRC. Among the 2 CGIs in the hMLH1 promoter, that is, CGI-I and CGI-II, 20% (6/30) and 13% (4/30) of the patients had methylated CGI-I and CGI-II, respectively. Suppressed expression of hMLH1was significantly correlated with methylation of CGI-I but not CGI-II. Further analysis of the MVP profiles of CGI-I showed that most of the MVPs were hypermethylated and others were poorly methylated or unmethylated. The profiles could be classified into at least 4 groups based on the methylation status of 3 MVPs at positions 21 to 23 in CGI-I. All 6 patients with methylated CGI-I belonged to group I. This result suggests that the above 3 MVPs in CGI-I should be a targeted region to further analyze the epigenetic features of hMLH1 in human SCRC. Our results further suggest that MVP profiling is useful for identifying the aberrantly methylated CGIs associated with suppressed gene expression.

Sequence Variants and the Risk of Head and Neck Cancer: Pooled Analysis in the INHANCE Consortium

Previous molecular epidemiological studies on head and neck cancer have examined various single nucleotide polymorphisms (SNPs), but there are very few documented associations. In the International head and neck cancer epidemiology (INHANCE) consortium, we evaluated associations between SNPs in the metabolism, cell cycle, and DNA repair pathways and the risk of head and neck cancer. We analyzed individual-level pooled data from 14 European, North American, Central American, and Asia case-control studies (5,915 head and neck cancer cases and 10,644 controls) participating in the INHANCE consortium. Unconditional logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for SNP effects, adjusting for age, sex, race, and country. We observed an association between head and neck cancer risk and MGMT Leu84Phe heterozygotes (OR = 0.79, 95% CI = 0.68-0.93), XRCC1 Arg194Trp homozygotes Arg/Arg (OR = 2.3, 95% CI = 1.1-4.7), ADH1B Arg48His homozygotes Arg/Arg (OR = 2.7, 95% CI = 1.9-4.0), ADH1C Ile350Val homozygotes Ile/Ile (OR = 1.2, 95% CI = 1.1-1.4), and the GSTM1 null genotype (OR = 1.1, 95% CI = 1.0-1.2). Among these results, MGMT Leu84Phe, ADH1B Arg48His, ADH1C Ile350Arg, and the GSTM1 null genotype had fairly low false positive report probabilities (<20%). We observed associations between ADH1B Arg48His, ADH1C Ile350Arg, and GSTM1 null genotype and head and neck cancer risk. No functional study currently supports the observed association for MGMT Leu84Phe, and the association with XRCC1 Arg194Trp may be a chance finding.

Epigenetics, disease, and therapeutic interventions

Heritable changes in gene expression that do not involve coding sequence modifications are referred to as "epigenetic". Epigenetic mechanisms principally include DNA methylation and a variety of histone modifications, of which the best characterized is acetylation. DNA hypermethylation and histone hypoacetylation are hallmarks of gene silencing, while DNA hypomethylation and acetylated histones promote active transcription. Aberrant DNA methylation and histone acetylation have been linked to a number of age related disorders including cancer, autoimmune disorders and others. Since epigenetic alterations are reversible, modifying epigenetic marks contributing to disease development may provide an approach to designing new therapies. Herein we review the role of epigenetic changes in disease development, and recent advances in the therapeutic modification of epigenetic marks.

Mutually exclusive promoter hypermethylation patterns of hMLH1 and O6-methylguanine DNA methyltransferase in colorectal cancer

Hypermethylation of CpG islands in gene promoter regions is an important mechanism of gene inactivation in cancer. Many cellular pathways, including DNA repair, are inactivated by this type of epigenetic lesion, resulting in proposed mutator phenotypes. Promoter hypermethylation of hMLH1 has been implicated in a subset of colorectal cancers that show microsatellite instability (MSI). Transcriptional silencing of O6-methylguanine DNA methyltransferase (MGMT) has also been described in a variety of neoplasms and has been associated with a consequent mutational spectrum. We investigated the relationship between hMLH1 promoter hypermethylation and MGMT promoter hypermethylation in 110 colorectal cancers using methylation-specific polymerase chain reaction. Expression of hMLH1 and MGMT was assessed by immunohistochemistry. MSI testing was performed using the National Cancer Institute consensus panel of five microsatellite markers. Promoter hypermethylation of hMLH1 was detected in 12% of tumors. This was significantly associated with the MSI-high phenotype (P < 0.01) and loss of hMLH1 expression (P < 0.01). Methylation of the MGMT promoter was detected in 43% of tumors, which were mostly microsatellite stable or MSI-low (P = 0.041) and showed loss of MGMT expression (P < 0.01). We demonstrated an inverse relationship between hMLH1 promoter hypermethylation and MGMT promoter hypermethylation (P = 0.041), suggesting that a number of distinct hypermethylation-associated pathways may exist in colorectal cancer.

A human beta-cell line for transplantation therapy to control type 1 diabetes

A human pancreatic beta-cell line that is functionally equivalent to primary beta-cells has not been available. We established a reversibly immortalized human beta-cell clone (NAKT-15) by transfection of primary human beta-cells with a retroviral vector containing simian virus 40 large T-antigen (SV40T) and human telomerase reverse transcriptase (hTERT) cDNAs flanked by paired loxP recombination targets, which allow deletion of SV40T and TERT by Cre recombinase. Reverted NAKT-15 cells expressed beta-cell transcription factors (Isl-1, Pax 6, Nkx 6.1, Pdx-1), prohormone convertases 1/3 and 2, and secretory granule proteins, and secreted insulin in response to glucose, similar to normal human islets. Transplantation of NAKT-15 cells into streptozotocin-induced diabetic severe combined immunodeficiency mice resulted in perfect control of blood glucose within 2 weeks; mice remained normoglycemic for longer than 30 weeks. The establishment of this cell line is one step toward a potential cure of diabetes by transplantation.

Genomic stability and tumorigenesis

Cancer research needs to explain the observed incidence of cancer. Many factors determine this process, including: the number of susceptible cells in the tissue of origin; the number of normal cell divisions through which susceptible cells pass in normal development and turnover; the number of cell divisions during tumorigenesis; the selective advantage which tumour cells have acquired; the ease of detection of a cancer; the number of mutations which are required for malignancy; the possible stepwise nature of tumorigenesis; the value of the normal mutation rate; the presence of extrinsic factors such as mutagens or growth promoters; the presence and strength of genomic instability, a phenomenon which has received a great deal of attention. We know very little about most of the factors and their influence in humans. We cannot, therefore, readily answer the question as to whether or not the observed incidence of cancer is what we would expect. Specifically, it is not yet possible to assess whether or not genomic instability is a prerequisite for carcinogenesis. Mathematical models, which assess the importance of genomic instability in tumorigenesis can be helpful, but require interpretation in the context of our overall ignorance. Experimental data have shown genomic instability in some cancers, but we do not yet know whether or not hypermutation initiates and drives tumorigenesis.

Killing and mutagenic actions of dacarbazine, a chemotherapeutic alkylating agent, on human and mouse cells: effects of Mgmt and Mlh1 mutations

Among various types of drugs designed for use in cancer chemotherapy, some have the potential for alkylation. After metabolic activation, these chemicals attack DNA and alkylate their bases, thereby preventing multiplication of rapidly growing tumor cells. Some of alkylated bases cause mutations, leading to untoward induction of tumors. To search for the rationale to separate lethal and mutagenic effects of alkylation drugs, we investigated actions of dacarbazine, a monofunctional triazene, on mouse and human cell lines defective in the Mgmt and/or the Mlh1 gene, the former encoding a DNA repair methyltransferase and the latter a protein involved in mismatch repair and induction of apoptosis. Mgmt-deficient cells are hypersensitive to the killing action of dacarbazine. On the other hand, cells defective in both Mgmt and Mlh1 genes are as resistant to the drug as are wild-type cells, in terms of survival, but do have many mutations after dacarbazine treatment. Thus, the killing and mutagenic actions of dacabazine can be dissociated by manipulating actions of these gene products.

Mutational-reporter transgenes rescued from mice lacking either Mgmt, or both Mgmt and Msh6 suggest that O6-alkylguanine-induced miscoding does not contribute to the spontaneous mutational spectrum

O6-methylguanine methyltransferase, Mgmt, constitutes the first line of defense against O6-alkylguanine, which can result in G : C to A : T transitions upon DNA replication. Mgmt has been found in organisms as diverse as archaebacteria and mammals. This evolutionary conservation suggests that all organisms may be exposed to either endogenous or environmental alkylating agents. We thus hypothesized that tissues of Mgmt-/- mice would exhibit elevated mutant frequencies. Employing the Big Blue trade mark transgenic system, we evaluated lacI mutants rescued from liver and small intestinal DNA of young Mgmt-/- mice. Interestingly, while there was a small difference between Mgmt-/- mice and controls with respect to lacI mutant frequency, no differences attributable to Mgmt deficiency were apparent in the mutational spectra. Although mutations stemming from O6-guanine alkylations would be predicted to be cumulative, we found no evidence of an Mgmt-dependent alteration in mutation spectrum in DNA samples from 12 month-old mice. To optimize our ability to detect mutations resulting from O6-alkylguanine-induced G : T mismatches, mice with combined deficiencies of Mgmt and the DNA mismatch repair molecule, Msh6, were analysed. In spite of this strategy, we observed no significant differences between Mgmt-/- Msh6-/- and Msh6-/- mouse lacI mutations, except for a trend towards a greater percentage (of total transitions) of G : C to A : T changes in Mgmt-/-Msh6-/- livers. Therefore, despite the striking evolutionary conservation of Mgmt, deficiency of this gene did not significantly impact the spontaneous lacI mutational spectrum in vivo.

Transcription factor accessibility and histone acetylation of the progesterone receptor gene differs between parental MCF-7 cells and a subline that has lost progesterone receptor expression

The human progesterone receptor (PgR) gene has a complex promoter that produces alternate mRNAs encoding the PgRA (94 kDa) and PgRB (120 kDa) protein isoforms. Expression of PgR is induced by estradiol (E(2)) in the breast, reproductive tract and many cell lines despite the lack of a classical estrogen responsive element (ERE) in the promoter regions. We employed chromatin immunoprecipitation (ChIP) to analyze the sites of estrogen receptor alpha (ERalpha) and Sp1 occupancy of the PgR promoters in vivo. We also assessed the functional relevance of histone acetylation levels on the accessibility of transcription factors to the promoter and subsequent hormone-induced transcription. We utilized MCF-7 human breast cancer cells that express PgR in response to E(2) and the MCF-7 derived C4 cell strain that has lost PgR expression as a model system. We found that promoter-wide levels of histone acetylation were not decreased in C4 cells, but that access was partially blocked for Sp1 and completely blocked for ERalpha. The basal level of histone acetylation at six localized regions of the promoter did show some differences between cell lines, but it did not correlate with transcription factor binding. Furthermore, we found only a modest and highly localized change in histone acetylation levels in response to E(2) at only one of three sites of ERalpha binding in MCF-7 cells. This was at the B1 site at the distal 5' end of the promoter. This site also showed a significant decrease in basal histone acetylation in C4 compared to MCF-7 cells. We speculate that the histone acetylation level at this site may be a marker for chromatin structure that affects the access of transcription factors to the whole promoter.

The mouse Murr1 gene is imprinted in the adult brain, presumably due to transcriptional interference by the antisense-oriented U2af1-rs1 gene

The mouse Murr1 gene contains an imprinted gene, U2af1-rs1, in its first intron. U2af1-rs1 shows paternal allele-specific expression and is transcribed in the direction opposite to that of the Murr1 gene. In contrast to a previous report of biallelic expression of Murr1 in neonatal mice, we have found that the maternal allele is expressed predominantly in the adult brain and also preferentially in other adult tissues. This maternal-predominant expression is not observed in embryonic and neonatal brains. In situ hybridization experiments that used the adult brain indicated that Murr1 gene was maternally expressed in neuronal cells in all regions of the brain. We analyzed the developmental change in the expression levels of both Murr1 and U2af1-rs1 in the brain and liver, and we propose that the maternal-predominant expression of Murr1 results from transcriptional interference of the gene by U2af1-rs1 through the Murr1 promoter region.

Silencing effect of CpG island hypermethylation and histone modifications on O6-methylguanine-DNA methyltransferase (MGMT) gene expression in human cancer

O6-methylguanine-DNA methyltransferase (MGMT) repairs the cytotoxic and mutagenic O6-alkylguanine produced by alkylating agents such as chemotherapeutic agents and mutagens. Recent studies have shown that in a subset of tumors, MGMT expression is inversely linked to hypermethylation of the CpG island in the promoter region; however, how the epigenetic silencing mechanism works, as it relates to hypermethylation, was still unclear. To understand the mechanism, we examined the detailed methylation status of the whole island with bisulfite-sequencing in 19 MGMT non-expressed cancer cell lines. We found two highly methylated regions in the island. One was upstream of exon 1, including minimal promoter, and the other was downstream, including enhancer. Reporter gene assay showed that methylation of both the upstream and downstream regions suppressed luciferase activity drastically. Chromatin immunoprecipitation assay revealed that histone H3 lysine 9 was hypermethylated throughout the island in the MGMT negative line, whereas acetylation on H3 and H4 and methylation on H3 lysine 4 were at significantly high levels outside the minimal promoter in the MGMT-expressed line. Furthermore, MeCP2 preferentially bound to the CpG-methylated island in the MGMT negative line. Given these results, we propose a model for gene silencing of MGMT that is dependent on the epigenetic state in cancer.