Skip Navigation


Carcinogenesis Advance Access originally published online on February 25, 2006
Carcinogenesis 2006 27(8):1661-1669; doi:10.1093/carcin/bgi375
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Material
Right arrow All Versions of this Article:
27/8/1661    most recent
bgi375v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Cebrian, A.
Right arrow Articles by Ponder, B. A.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cebrian, A.
Right arrow Articles by Ponder, B. A.J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Genetic variants in epigenetic genes and breast cancer risk

Arancha Cebrian*, Paul D. Pharoah, Shahana Ahmed, Santiago Ropero1, Mario F. Fraga1, Paula L. Smith2, Don Conroy, Robert Luben3, Barbara Perkins, Douglas F. Easton2, Alison M. Dunning, Manel Esteller1 and Bruce A.J. Ponder

Cancer Research UK Human Cancer Genetics Research Group, Department of Oncology, University of Cambridge, Strangeways Research Laboratories Cambridge CB1 8RN, UK
1 Cancer Epigenetics Laboratory, Spanish National Cancer Centre (CNIO) Madrid, Spain
2 Cancer Research UK Genetic Epidemiology Unit, Strangeways Research Laboratories Cambridge CB1 8RN, UK
3 Department of Public Health and Primary Care, Strangeways Research Laboratories Cambridge CB1 8RN, UK

*To whom correspondence should be addressed. Tel: +44 1223 740 684; Fax: +44 1223 740 147; Email: acebrian{at}iib.uam.es


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 References
 
Epigenetic events, resulting changes in gene expression capacity, are important in tumour progression, and variation in genes involved in epigenetic mechanisms might therefore be important in cancer susceptibility. To evaluate this hypothesis, we examined common variants in 12 genes coding for DNA methyltransferases (DNMT), histone acetyltransferases, histone deacetyltransferases, histone methyltrasferases and methyl-CpG binding domain proteins, for association with breast cancer in a large case–control study (N cases = 4474 and N controls = 4580). We identified 63 single nucleotide polymorphisms (SNPs) that efficiently tag all the known common variants in these genes, and are also expected to tag any unknown SNP in each gene. We found some evidence for association for six SNPs: DNMT3b-c31721t [P (2 df) = 0.007], PRDM2-c99243 t [P (2 df) = 0.03] and t105413c [P-recessive = 0.05], EHMT1-g-9441a [P (2df) = 0.05] and g41451t (P-trend = 0.04), and EHMT2-S237S [P (2df) = 0.04]. The most significant result was for DNMT3b-c31721t (P-trend = 0.124 after adjusting for multiple testing). However, there were three other results with P < 0.05. The permutation-based probability of this occurring by chance was 0.335. These significant SNPs were genotyped in 75 human cancer cell lines from different tumour types to assess if there was an association between them and six epigenetic measures. No statistically significant association was found. However, a trend was observed: homozygotes for the rare alleles of the EHMT1, EHMT2 and PRDM2 had a mean value for both trimethylation of K9 and K27 of histone H3 remarkably different to the homozygotes for the common alleles. Thus, these preliminary observations suggest the possible existence of a functional consequence of harbouring these genetic variants in histone methyltransferases, and warrant the design of larger epidemiological and biochemical studies to establish the true meaning of these findings.

Abbreviations: df, degrees of freedom; DNMTs, DNA methyltransferases; HATs, histone acetyltransferases; HDACs, histone deacetylases; HMTs, histone methyltransferases; LD, linkage disequilibrium; MAF, minor allele frequency; MBDs, methyl-CpG binding domain proteins; ORs, odds ratios; SNPs, single nucleotide polymorphisms.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Supplementary material
 References
 
Breast cancer is the most common cancer among women in industrialized countries (1). A family history is well established as a risk factor for breast cancer (2), and twin studies suggest that most of the excess familial risk is due to inherited factors (3). However, germline mutations in so-called high-penetrance cancer susceptibility genes, such as BRCA1 and BRCA2, account for <25% of the excess risk (4). A range of genetic models may explain the remainder, but it seems likely that variants that are common in the population are important (5). The present study examines the hypothesis that genes involved in epigenetic modification may harbour such common variants.

The molecular mechanisms underlying the development of breast cancer are not well understood. It is generally believed that the initiation of breast cancer, like other cancers, is a progressive appearance of malignant cell behaviour that is triggered by the evolution of altered gene function. However, it is becoming clear that epigenetic events, or heritable changes in gene expression capacity without DNA alterations, are also central to tumour progression. The epigenetic control of gene function involves the formation of chromatin that modulates gene transcription. Two epigenetic modifications have emerged as critical components of transcriptional regulation. The first, histone acetylation, appears to be used by all eukaryotes as one layer of transcriptional control (6). The acetylation of the N-terminal tails of histones H3 and H4 by histone acetyltransferases (HATs) creates an accessible chromatin configuration that facilitates transcriptional activity. Removal of these acetyl groups by histone deacetylases (HDACs) facilitates chromatin compaction that is detrimental to transcription. The cell uses these HATs and HDACs as co-activators and co-repressors, respectively, to modulate promoter activity. The second epigenetic modification, DNA methylation, has been implicated as a critical layer of control by enhancing transcriptional silencing (7,8). In mammals, genomic methylation patterns are established during embryogenesis through the interplay of at least three DNA methyltransferases (DNMT1, DNMT3a, and DNMT3b). DNMT1 is thought to maintain these methylation patterns during DNA replication, whereas DNMT3a and DNMT3b act primarily as de novo methyltransferases to establish methylation patterns during embryogenesis (9). However, different experiments have demonstrated that DNA methylation itself does not interfere with transcription, but rather marks the DNA for establishment of a transcriptionally incompetent chromatin state. The mechanisms by which methylation mediates this process have recently emerged (10–13). DNA methylation depends on prior methylation of histone H3 at lysine 9 by the histone methyltransferases (HMTs) (14) and also depends on the methyl-CpG binding domain proteins (MBDs), which preferentially bind to methylated CpGs (15,16) and repress transcription partially through the action of HDACs. The association of virtually all of the methylation machinery, HMTs, DNMTs and MBDs, with HDACs provides a cooperative linkage in transcriptional silencing between DNA methylation and histone deacetylation (17,18).

Genetic variation and/or environmental exposures could lead to systemic epigenetic control defects that either directly increase the risk of developing cancer or represent surrogate markers for an increased risk (19). Nevertheless, few studies have examined genetic variation (single nucleotide polymorphisms, or SNPs) in genes coding for the proteins that establish alternate states of chromatin structure and histone modification (HATs, HDACs and HMTs), and DNA methylation (DNMTs and MBDs). Only one gene (DNMT3b) has been analysed in genetic association studies and these have evaluated only one or two SNPs (20,21) (Figure 1).


Figure 1
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 1 Epigenetic control. Schematic of some of the molecular events that occur at CpG-rich promoters undergoing epigenetic control of gene function in the cell. The open chromatin structure of a transcriptionally active gene with loosely spaced nucleosomes (black circles) is shown at the top and the transcriptionally silenced state with more tightly packed nucleosomes is shown at the bottom. The acetylation of the N-terminal tails of histones H3 and H4 by HATs creates an accessible chromatin configuration that facilitates transcriptional activity. DNA methyltransferases (DNMTs) may act only on chromatin that is methlylated at lysine 9 and is associated with HP1. MBDs recognize the methylated DNA and recruit HDACs and, possibly, histone methylases (HMTs) to perpetuate and maintain the repressed state. DNA is indicated by a think grey line wrapped around nucleosomes. Proteins are indicated by ovals, and grey small circles on the DNA line denote a methylated state.

 
The aim of this study was to evaluate the association between common variants in 14 genes coding for DNMTs (DNMT1, DNMT3a and DNMT3b), HATs (EP300 and CREBBP), HDACs (HDAC1, HDAC2 and HDAC5), HMTs (PRDM2, SUV39H1, SETDB1, EHMT1 and EHMT2) and MBDs (MBD2) and susceptibility to breast cancer. We have used a case–control study design and genotyped SNPs that tag all common variants present in each gene in our population.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Supplementary material
 References
 
Patients, controls and cell lines
Cases were drawn from SEARCH, an ongoing population-based study in which cases are ascertained through the East Anglian Cancer Registry. All patients diagnosed with invasive breast cancer below age 55 years since 1991 and still alive in 1996 (prevalent cases; median age, 48 years), together with all those diagnosed <70 years between 1996 and the present (incident cases; median age, 52 years), were eligible to take part. All study participants completed an epidemiological questionnaire and provided a blood sample for DNA analysis. Sixty-seven per cent of eligible breast cancer patients returned a questionnaire and 64% provided a blood sample. Controls were randomly selected from the Norfolk component of EPIC (European Prospective Investigation of Cancer). EPIC is a prospective study of diet and cancer being carried out in nine European countries. The EPIC-Norfolk cohort comprises 25 000 individuals resident in Norfolk, East Anglia, the same region from which the cases have been recruited. Controls are not individually matched to cases, but are broadly similar in age, being aged 42–81 years. The ethnic background of both cases and controls as reported on the questionnaires is similar, with >98% being white. All participants have given written consent, and the study is approved by the Eastern Region Multicentre Research Ethics Committee.

A total of 4474 cases, of whom 27% were prevalent cases, and 4560 controls were available for analysis. The samples have been split into two sets in order to conserve DNA and reduce genotyping costs. The first set (2271 cases, 2280 controls) was genotyped for all SNPs. Any SNP that showed association in Set 1 that was significant at the P < 0.1 level, using either of the test used was then tested in the second set (2203 cases, 2280 controls). If P < 0.1 for a comparison of haplotype frequencies in Set 1, all SNPs used in the haplotype analysis were genotyped in Set 2. This staged approach substantially reduces genotyping costs without significantly affecting statistical power. Using this design, for a causative SNP that is tagged with r2 > 0.8, a type I error rate of 0.0001 and genotyping success rate of 0.95, the study has 86% power to detect a dominant allele with MAF of 0.05 that confers a relative risk of 1.5 and 87% power to detect a dominant allele with MAF of 0.25 that confers a relative risk of 1.3. Power to detect recessive alleles is less: 53% for an allele with MAF of 0.25 and risk 1.5 and 71% for an allele with MAF 0.5 and risk 1.3. Cases with high yields of genomic DNA were selected for Set 1 from the first 3500 recruited, with Set 2 comprising the remainder of these plus the next 974 incident cases recruited. As the prevalent cases were the first recruited, the proportion of prevalent cases was somewhat higher in Set 1 than Set 2 (33 versus 20%). Median age at diagnosis was similar in both sets (51 and 52 years old, respectively). Median time from diagnosis to blood draw was slightly longer for Set 2 (15 months) than for Set 1 (9 months). There were no significant differences in the morphology, histopathological grade or clinical stage of the cases by set or by prevalent/incident status.

A panel of 75 human cancer cell lines were obtained from the American Type Culture Collection (Rockland, MD), the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany), and the Spanish National Cancer Centre (Madrid, Spain). These were used to assess measures of epigenetic status. Twelve different tumour types were represented in the set: lymphoma (n = 11), melanoma (n = 11), lung (n = 11), leukaemia (n = 7), breast (n = 7), prostate (n = 4), sarcoma (n = 3), renal (n = 2), cervical (n = 2), bladder (n = 2), glioma (n = 2), testis (n = 1), endometrium (n = 1), liver (n = 1), skin (n = 1) and thyroid (n = 1). DNA and histones were extracted as described previously (22,23).

Selection of tagging SNPs
The principal hypothesis underlying this experiment was that there are one or more common SNPs in the genes of interest that are associated with an altered risk of breast cancer. Thus, the aim of the SNP tagging was to identify a set of SNPs (stSNPs) that efficiently tags all the known SNPs. We postulate that such SNPs are also likely to tag any hitherto unidentified SNPs in the gene. We used data from the International HapMap Project (21-06-2005: HapMap last public release used in this study), which has genotyped a large number of SNPs in 30 parent–offspring trios. These samples were collected in 1980 from US residents with northern and western European ancestry by the Centre d'Etude du Polymorphisme Humain (CEPH).

The best measure of the extent to which one SNP tags another SNP is the pairwise correlation coefficient Formula, since the loss in power incurred by using a marker SNP in place of a true causal SNP is directly related to this measure. However, some SNPs are poorly correlated with other single SNPs but may be efficiently tagged by a haplotype defined by multiple SNPs, thus reducing the number of tagging SNPs needed. In this case, the relevant measure is the correlation coefficient Formula between the causative SNP and a haplotype. We therefore aimed to tag each SNP with an Formula or Formula of >0.8 with tag SNP or haplotype. The program tagSNPs (24) was used for selection of tagging SNPs.

Since tagging SNP selection is problematic when there is extensive haplotype diversity, where necessary we divided a gene into haplotype blocks and selected the stSNPs for each block separately. It is possible to use a variety of formal definitions of haplotype blocks, but we simply used the graphical representations of the pattern of linkage disequilibrium (LD) based on D' and selected blocks such that the common haplotypes in each block accounted for at least 80% of all haplotypes observed using the Haploview program (25). If assay design failed for a selected stSNP the tagging selection was repeated with forced exclusion of the failed SNP in order to select an alternative.

Assuming a minimum r2 of 0.8, this study had greater than 85% power to detect, at a significance level of P < 0.0001, any dominant susceptibility allele with a frequency of 5% or greater conferring a relative risk of at least 1.4 or a recessive allele with frequency 10% or greater conferring a relative risk of at least 2.

Genotyping
We genotyped all samples for the selected tag SNPs using the ABI PRISM 7900 sequence detection system or ‘Taqman’ (Applied Biosystems, Foster City, CA). We carried out PCR on DNA (10 ng) using TaqMan universal PCR master mix (Applied Biosystems), forward and reverse primers and FAM and VIC labelled probes designed by Applied Biosystems (ABI Assay-by-Designs) in a 5 µl reaction. Sequences of primers and probes are available on request. Amplification conditions on MJ Tetrad thermal cyclers (Genetic Research Instrumentation, MJ Research, Cambridge, MA) were as follows: 1 cycle of 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. We read the completed PCRs on an ABI PRISM 7900 Sequence Detector in end-point mode using the Allelic Discrimination Sequence Detector Software (Applied Biosystems). For the software to recognize the genotypes, we included two non-template controls in each 384-well plate. Cases and controls were arrayed together in twelve 384-well plates and a thirteenth plate contained eight duplicate samples from each of the twelve plates to ensure a good quality of genotyping (the concordance was >99% for all SNPs). Failed genotypes were not repeated (the rate for failed genotypes did not exceed 8.3% for any of the SNPs under study).

Epigenetic assays
Two epigenetic mechanisms of gene regulation were assessed: DNA methylation and histone modifications. For DNA methylation, we analysed the total genomic content of 5-methylcytosine DNA by High-Performance Capillary Electrophoresis (HPCE) and the number of genes with promoter CpG island hypermethylation by methylation-specific PCR (22). For the latter, the following 15 genes were analysed: p16INK4a, p14ARF, p15INK4b, THBS1, CDH1, p73, DAPK, MGMT, BRCA1, LKB1, TIMP3, hMLH1, RARB2, GSTP1 and RASSF1A (22). For histone modifications, the overall acetylation levels of histone H3 and histone H4 were quantified by HPCE (23), whilst assessment of the methylation levels at specific histone residues was accomplished by western blotting. The antibodies used were trimethyl-K9-H3 (1 : 1000) and trimethyl-K27-H3 (1 : 1000; both from Upstate). Antibody to total histone H4 (1 : 2000; Upstate) was used as a loading control.

Statistical methods
For each polymorphism, deviation of the genotype frequencies in the controls from those expected under Hardy–Weinberg equilibrium was assessed by a {chi}2-test. Genotype frequencies in cases and controls were compared using a {chi}2-test with 2 degrees of freedom (df) (P-heterogeneity), and the Armitage trend test ({chi}2 on 1 df) for the trend in breast cancer risk with number of rare alleles (P-trend). As a secondary analysis, we also considered the difference in the rare homozygote frequency. The relative risks of breast cancer for heterozygotes and for rare homozygotes, relative to common homozygotes, were estimated as odds ratios (ORs) with associated 95% confidence intervals (CI).

Haplotype frequencies were estimated and compared in cases and controls using the estimation maximization (EM) algorithm implemented in the Haploscore program (26). Haplotypes with a frequency <0.05 were pooled. The Haploscore program computes score statistics (and hence significance levels) to test for associations between individual haplotypes and disease status, along with the global test of association.

The potential functional effect of specific SNPs was examined using several tools. TESS (Transcription Element Search Software; http://www.cbil.upenn.edu/tess) is a web-based software tool for locating possible transcription factor binding sites in DNA sequence and for browsing the TRANSFAC database. It provides functionality beyond that of the TRANSFAC flat files and web site. PROMO (http://alggen.lsi.upc.es/home) is a program to predict potential transcription factor binding sites in DNA sequences among those already experimentally identified. The TRANSFAC database (27) is used as the source of known binding sites and transcription factors. Weight matrices representing the binding sites are constructed in a dynamic fashion from factor-specific collections of sites (28,29). PupaSNP (Putative Phenotypic Alterations caused by SNPs; http://pupasnp.bioinfo.cnio.esis) is a web-based search tool for SNPs with potential phenotypic effect at transcriptional level. PupaSNP inputs lists of genes (or generates them from chromosomal coordinates) and retrieves SNPs that could affect conserved regions that the cellular machinery uses for the correct processing of genes (intron/exon boundaries or exonic splicing enhancers), predicted transcription factor binding sites, or which cause changes in amino acids in proteins. The program uses the mapping of SNPs in the genome provided by Ensembl.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Supplementary material
 References
 
Genotyping results
We identified a total of 222 SNPs from the HapMap database in the genes of interest (Table I). On the basis of the LD among these SNPs, 63 tagging SNPs were identified. All SNPs were tagged with Formula > 0.8, and 197 were tagged with pairwise Formula > 0.8. For two of the genes, HDAC1 and SUV39H1, no SNPs polymorphic in the CEPH samples were identified. We therefore attempted to find additional SNPs by screening the coding regions of both genes in 48 breast cancer cases, using denaturing high-performance liquid chromatography (dHPLC) and direct sequencing. No variants were found and these two genes were therefore not considered further.


View this table:
[in this window]
[in a new window]
  		Table I Number of common SNPs and SNP density for the HapMap data for each gene, number of tag SNPs selected and their tagging performance

 
Genotype distributions in the controls did not differ significantly from that expected under Hardy–Weinberg equilibrium (HWE) for any of the SNPs, apart from SNP CREBBP-g154748a (rs9392) in Set 1 (genotype frequencies for cases and controls are shown in Supplementary Table S1). However, as the cases did not deviate from HWE and we observed a good quality of genotyping in all the plates, this observation may be attributable to chance. Table II shows the minor allele frequency (MAF) for all SNPs in controls, together with the genotype-specific risks and results of the tests for association for each SNP. There was no evidence for association of genotype with age in controls and, as expected, age-adjusted risks were similar to unadjusted risks (data not shown). Table III shows the estimated haplotype frequencies for each gene in cases and controls, together with the results of the global test for association and the haplotype-specific risks.


View this table:
[in this window]
[in a new window]
  		Table II Breast cancer risks associated with tag SNPs in 12 genes included in this study

 

View this table:
[in this window]
[in a new window]
  		Table III Estimated haplotype frequencies for each gene, global test for association and haplotype-specific risks

 
We studied the association of each individual tag SNP, and in situations where the combination of SNPs was required to tag another we examined that combination. Fifty-seven of the 63 tag SNPs showed no association with breast cancer. In addition, there was no evidence of association with the g–g haplotype of EHMT1 rs4634736 and rs6559218, which specifically tags rs10780185 (P = 0.49). Positive associations were seen for DNMT3b rs406193, PRDM2 rs2235515 and rs2235512, EHMT1 rs4634736 and rs3125795, and EHMT2 rs535586.

The SNP DNMT3b-c31721t (rs406193) showed a significant difference in genotype distribution between cases and controls, with the t-allele being associated with a reduced risk [OR ct versus cc = 0.85 (0.77–0.94); OR tt versus cc 0.89 (0.64–1.23); P-het = 0.007; P-Trend = 0.003].

PRDM2-c99243t (rs2235515) and t105413c (rs2235512) both showed some evidence of association. The first SNP, located in LD block 1, was associated with an increased risk of breast cancer in an apparently co-dominant manner [OR ct versus cc = 1.05 (0.96–1.15); OR tt versus cc 1.27 (1.05–1.52); P-trend = 0.02]. The heterogeneity and trend tests of association were not significant for t105413c (P = 0.12 and 0.08, respectively), but there was some evidence for a recessive effect of the rare allele [OR cc versus tt = 1.21 (1.01–1.46); P = 0.047].

Two of the 6 SNPs studied in EHMT1 showed some evidence of association with breast cancer (rs4634736 and rs3125795). The a-allele of g-9441a (rs4634736) was associated with a decreased risk of breast cancer [OR ga versus gg = 0.88 (0.79–0.97); OR aa versus gg 0.89 (0.59–1.34); P-Trend = 0.02]. Neither of the main tests were significant for rs3125795, but there was evidence of an increased risk in tt homozygotes [OR tt versus gg = 2.18 (1.12–4.23); P = 0.02]. These two SNPs are not correlated with each other Formulaand both SNPs were significant when included in a joint analysis using multiple logistic regression. There was also a marginal difference in haplotype frequencies between cases and controls for this (global P = 0.05) (Table III). This was primarily due to the difference in frequency of the haplotype carrying the rare allele of SNP g-9441a [OR 0.82 (0.70–0.98) P = 0.027].

Both EHMT2 SNPs were genotyped in the complete sample set, but only S234S (rs535586) was significant (P-trend = 0.01).

Associations between epigenetic assays and genotype
The six SNPs, significant at the 5% level in at least one analysis, were genotyped in 75 human cancer cell lines from different tumour types to assess whether there was an association between genotype and measures of epigenetic status. Six measures were used: 5-methylcytosine DNA content, number of hypermethylated CpG islands, acetylation content of histones H3 and H4 and the trimethylation levels of the lysine 9 (trimethyl-K9-H3) and lysine 27 of histone H3 (trimethyl-K27-H3). Illustrative examples of the different epigenetic analyses are shown in Supplementary Figure 1. For DNA methylation measures, hypermethylation of the 15 promoter CpG islands, described in Materials and methods, ranged from 1 to 11, with a median value of 4; 5-methylcytosine DNA content ranged from 1.4 to 5.6%, with a median value of 3.1. No statistically significant associations were found for any SNP in any assay, but those samples that were rare allele carriers of the three HMTs included in the study (EHMT1, EHMT2 and PRDM2) had a mean value, for both trimethylation of K9 and K27 of histone H3, substantially different to those observed in the homozygotes for the common allele (Table IV). In this regard, EHMT1, EHMT2 and RIZ1 have a preferential affinity to trimethylate K9 and K27 of histone H3 (30). These preliminary observations suggest a functional consequence for people who harbour these genetic variants in HMTs. Larger epidemiological and biochemical studies to unravel the true meaning of these findings are clearly warranted.


View this table:
[in this window]
[in a new window]
  		Table IV Trimethylation levels of the lysine 9 (trimethyl-K9-H3) and lysine 27 of histone H3 (trimethyl-K27-H3)

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Supplementary material
 References
 
We have assessed the evidence for association between common variants in 12 genes involved in epigenetics and the risk of breast cancer in a large population-based case–control study. We typed 63 SNPs that tag all the known common polymorphisms in these genes. Two different SNPs in DNMT3b gene (rs2424913 and a –283t > c variant in DNMT3b promoter) have been associated previously with lung cancer susceptibility: carriers of the rare allele of rs2424913 were found to have double the risk (20), whereas the –283 cc homozygotes were found to have a 50% reduction in risk in a different study (21). There are no published data on these variants, other variants in this gene, or variants in other chromatin or histone modification genes or methylation genes in breast cancer. Rs2424913 is in perfect LD Formulawith the variant a20506g (rs2424928), which was included in our study. We found no association of this variant with breast cancer risk, and so we have not directly analysed this SNP. However, we have virtually 100% power at a type I error rate of 0.0001 to detect an effect if this variant were associated with the same magnitude of risk as found for lung cancer.

Seven out of 63 SNPs showed evidence of association in the initial case–control set, significant at the 0.1 level using one of the two statistical tests, and these were genotyped in a further series of cases and controls. Four of these were significant at the 5% level in the complete data—DNMT3b-c31721t and EHMT1-g-9441a were associated with a reduced risk and PRDM2-c99243t and EHMT2-S237S were associated with an increased risk. In addition, two other SNPs, PRDM2-t105413c and EHMT1-g41451t, showed borderline association under a recessive genetic model, although both tests based on the overall genotype distribution and the trend test were not significant.

Of the six SNPs with possible associations, three are intronic and not strongly correlated with other known SNPs that are more likely to have a functional role. However, various in silico tools suggest that the other three may have direct functional effects. The sequence motif MATWAAT (where M is A or C), recognized by the transcription factor N-Oct-3, is missing when the t-allele of DNMT3b-c31721t is present (TESS program); and the presence of N-Oct-3 factor has been associated with the tumourigenic phenotype of melanoma (31,32). The rare a-allele of EHMT1-g-9441a, which is associated with a decreased risk of developing breast cancer, modifies the sequence motif recognized by two different transcription factors. The first, ATF3 (activating transcription factor 3), is associated with invasive potential in mammary epithelial cells (33) and colorectal cancer (34). The second, c-Jun, operates as mediator of cell proliferation and differentiation (35,36) and also cooperates with oncogenic alleles of ras in malignant transformation (37). The presence of the rare allele also creates a new sequence motif for the YY1 transcription factor involved in repressing and activating a diverse number of promoters (38–40). The third significant SNP that is likely to be functional is EHMT2-S237S. Although this is a silent polymorphism, an effect of synonymous codon usage on gene expression has been supported by the detection of epistatic interactions between nucleotides that are important in maintaining pre-mRNA/RNA secondary structures (41). The presence of the rare allele of SNP S237S might modify the mRNA folding, stability and translation of EHMT2 protein and consequently lead to altered protein activity. Our functional assays suggest that trimethylation levels of lysine 9 and 27 of histone H3, both epigenetic marks of transcriptional repression (30), may be partly dependent on the EHMT1-g-9441a and EHMT2-S237S alleles. If confirmed, this would provide an additional link between these genotypes and the putative regulation of genes involved in cellular transformation. Further functional studies will be necessary to test the biological effect of these SNPs.

Regardless of the functional consequences of these SNPs, observed associations with breast cancer risk may be due to chance or bias. Despite the large sample size, none of the associations were highly significant (none more significant than P = 0.003), and it is possible that the findings are simply type I statistical errors. Adjustments for multiple hypotheses testing using the Bonferroni correction are too conservative, since it assumes that the tests are independent. We therefore derived an empirical global test based on the most significant P-value (i.e. P-trend = 0.003 for DNMT3b-c31721t), using simulations in which case–control studies were permuted randomly. In 1000 random permutations, a P-value at least as significant as this was obtained on 124 occasions, giving a P-trend adjusted for multiple testing of 0.124. Using the same simulation, the probability of observing 4 results out of 63 with P < 0.05 was estimated to be 0.335.

An alternative explanation for a (false) positive association is bias due to hidden population stratification. This occurs when allele frequencies differ between population subgroups and cases and controls are drawn differentially from those subgroups. However, it seems unlikely that population stratification is relevant here because the cases and controls were drawn from the same north-western European group (>98% white). Furthermore, we have found no evidence for association between 23 unlinked markers (209 tests) in the controls, which suggests that there is unlikely to be significant substructure in our population. (42).

In summary, we found no evidence that common variations in DNMT1, DNMT3a, EP300, CREBBP, HDAC2, HDAC5, SETDB1 or MBD2 are associated with breast cancer. The SNPs under study were not selected because of their predicted effects on structure and function, but because they tag all known common variants in these genes. As these genes have not been resequenced in large numbers of individuals, it remains possible that important, unidentified variants in the genes were not efficiently tagged. Nevertheless, de Bakker et al. (43) have shown that sets of tags selected from the phase I HapMap dataset are expected to be almost as powerful as equal-sized sets chosen from complete resequencing reference panels. We found some evidence for association with breast cancer for variants in DNMT3b, PRDM2, EHMT1 and EHMT2, three of which may have functional effects. However, we believe that further studies to replicate these associations in both breast and other cancers are needed before a large commitment to carrying our detailed and expensive biological assays on the exact functional effects of these common gene variants is made.


   Supplementary material
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
References
 
Supplementary material can be found at: http://www.carcin.oxfordjournals.org/


   Acknowledgments
 
We thank the EPIC management team (K.-T. Khaw, S. Oakes, S. Bingham and J. Russell) for access to control DNA. B.A.J.P. is a Gibb Fellow of Cancer Research, UK.

Conflict of interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 References
 

  1. McPherson K., Steel C.M., Dixon J.M. (2000) ABC of breast diseases, breast cancer-epidemiology, risk factors, and genetics. BMJ 321:624–628.[Free Full Text]
  2. Collaborative group on hormonal factors in breast cancer. (2001) Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet 358:1389–1399.[CrossRef][Web of Science][Medline]
  3. Lichtenstein P., Holm N.V., Verkasalo P.K., Iliadou A., Kaprio J., Koskenvuo M., Pukkala E., Skytthe A., Hemminki K. (2000) Environmental and heritable factors in the causation of cancer-analyses of cohorts of twins from Sweden, Denmark, and Finland. N. Engl. J. Med. 343:78–85.[Abstract/Free Full Text]
  4. Easton D.F. (1999) How many more breast cancer predisposition genes are there. Breast Cancer Res. 1:14–17.[Medline]
  5. Houlston R.S. and Peto J. (2004) The search for low-penetrance cancer susceptibility alleles. Oncogene 23:6471–6476.[CrossRef][Web of Science][Medline]
  6. Grunstein M. (1997) Histone acetylation in chromatin structure and transcription. Nature 389:349–352.[CrossRef][Medline]
  7. Bird A. (1992) The essentials of DNA methylation. Cell 70:5–8.[CrossRef][Web of Science][Medline]
  8. Eden S. and Cedar H. (1994) Role of DNA methylation in the regulation of transcription. Curr. Opin. Genet. Dev. 4:255–259.[CrossRef][Medline]
  9. Okano M., Bell D.W., Haber D.A., Li E. (1999) DNA methyl-transferases Dnmt3a and Dnmt3b are essential for de novo methyl-ation and mammalian development. Cell 99:247–257.[CrossRef][Web of Science][Medline]
  10. Ordway J.M. and Curran T. (2002) Methylation matters: modelling a manageable genome. Cell Growth Differ. 13:149–162.[Free Full Text]
  11. Felsenfeld G. and Groudine M. (2003) Controlling the double helix. Nature 421:448–453.[CrossRef][Medline]
  12. Freiman R.N. and Tjian R. (2003) Regulating the regulators: lysine modifications make their mark. Cell 112:11–17.[CrossRef][Web of Science][Medline]
  13. Jaenisch R. and Bird A. (2003) Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat. Genet. 33:Suppl, 245–254.[CrossRef][Web of Science][Medline]
  14. Tamaru H. and Selker E.U. (2001) A histone H3 methyltransferase controls DNA methylation in Neurospora crassa. Nature 414:277–283.[CrossRef][Medline]
  15. Cross S.H., Meehan R.R., Nan X., Bird A. (1997) A component of the transcriptional repressor MeCP1 shares a motif with DNA methyl-transferase and HRX proteins. Nat. Genet. 16:256–259.[CrossRef][Web of Science][Medline]
  16. Hendrich B. and Bird A. (1998) Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mol. Cell Biol. 18:6538–6547.[Abstract/Free Full Text]
  17. Rountree M.R., Bachman K.E., Herman J.G., Baylin S.B. (2001) DNA methylation, chromatin inheritance, and cancer. Oncogene 20:3156–3165.[CrossRef][Web of Science][Medline]
  18. Kouzarides T. (2002) Histone methylation in transcriptional control. Curr. Opin. Genet. Dev. 12:198–209.[CrossRef][Web of Science][Medline]
  19. Laird P.W. (2005) Cancer epigenetics. Hum. Mol. Genet. 1:R65–R76.
  20. Shen H., Wang L., Spitz M.R., Hong W.K., Mao L., Wei Q. (2002) A novel polymorphism in human cytosine DNA-methyltransferase-3B promoter is associated with an increased risk of lung cancer. Cancer Res. 62:4992–4995.[Abstract/Free Full Text]
  21. Lee S.J., Jeon H.S., Jang J.S., et al. (2005) DNMT3B polymorphisms and risk of primary lung cancer. Carcinogenesis 26:403–409.[Abstract/Free Full Text]
  22. Paz M.F., Fraga M.F., Avila S., Guo M., Pollan M., Herman J.G., Esteller M. (2003) A systematic profile of DNA methylation in human cancer cell lines. Cancer Res. 63:1114–1121.[Abstract/Free Full Text]
  23. Fraga M.F., Ballestar E., Paz M.F., et al. (2005) Epigenetic differences arise during the lifetime of monozygotic twins. Proc. Natl Acad. Sci. USA 102:10604–10609.[Abstract/Free Full Text]
  24. Stram D.O., Haiman C.A., Hirschhorn J.N., Altshuler D., Kolonel L.N., Henderson B.E., Pike M.C. (2003) Choosing haplotype-tagging SNPs based on unphased genotype data using a preliminary sample of unrelated subjects with an example from the Multiethnic Cohort Study. Hum. Hered. 55:27–36.[CrossRef][Web of Science][Medline]
  25. Barrett J.C., Fry B., Maller J., Daly M.J. (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21:263–265.[Abstract/Free Full Text]
  26. Schaid D.J., Rowland C.M., Tines D.E., Jacobson R.M., Poland G.A. (2002) Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am. J. Hum. Genet. 70:425–434.[CrossRef][Web of Science][Medline]
  27. Wingender E., Chen X., Fricke E., et al. (2001) The TRANSFAC system on gene expression regulation. Nucleic Acids Res. 29:281–283.[Abstract/Free Full Text]
  28. Messeguer X., Escudero R., Farré D., Nuñez O., Martínez J., Albà M.M. (2002) PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics 2:333–334.
  29. Farré D., Roset R., Huerta M., Adsuara J.E., Roselló L., Albà M.M., Messeguer X. (2003) Identification of patterns in biological sequences at the ALGGEN server: PROMO and MALGEN. Nucleic Acids Res. 13:3651–3653.
  30. Sims R.J. III, Nishioka K., Reinberg D. (2003) Histone lysine methylation: a signature for chromatin function. Trends Genet. 19:629–639.[CrossRef][Web of Science][Medline]
  31. Thomson J.A., Murphy K., Baker E., Sutherland G.R., Parsons P.G., Sturm R.A., Thomson F. (1995) The brn-2 gene regulates the melanocytic phenotype and tumorigenic potential of human melanoma cells. Oncogene 11:691–700.[Web of Science][Medline]
  32. Cook A.L., Donatien P.D., Smith A.G., Murphy M., Jones M.K., Herlyn M., Bennett D.C., Leonard J.H., Sturm R.A. (2003) Human melanoblasts in culture: expression of BRN2 and synergistic regulation by fibroblast growth factor-2, stem cell factor, and endothelin-3. J. Invest. Dermatol. 121:1150–1159.[CrossRef][Web of Science][Medline]
  33. Iyengar P., Combs T.P., Shah S.J., et al. (2003) Adipocyte-secreted factors synergistically promote mammary tumorigenesis through induction of anti-apoptotic transcriptional programs and proto-oncogene stabilization. Oncogene 22:6408–6423.[CrossRef][Web of Science][Medline]
  34. Bottone F.G. Jr, Moon Y., Kim J.S., Alston-Mills B., Ishibashi M., Eling T.E. (2005) The anti-invasive activity of cyclooxygenase inhibitors is regulated by the transcription factor ATF3 (activating transcription factor 3). Mol. Cancer Ther. 4:693–703.[Abstract/Free Full Text]
  35. Bohmann D., Bos T.J., Admon A., Nishimura T., Vogt P.K., Tjian R. (1987) Human proto-oncogene c-jun encodes a DNA binding protein with structural and functional properties of transcription factor AP-1. Science 238:1386–1392.[Abstract/Free Full Text]
  36. Hong I.S., Kim S.H., Koong M.K., Jun J.H., Kim S.H., Lee Y.S., Kang K.S. (2004) Roles of p38 and c-jun in the differentiation, proliferation and immortalization of normal human endometrial cells. Hum. Reprod. 19:2192–2199.[Abstract/Free Full Text]
  37. Weiss C. and Bohmann D. (2004) Deregulated repression of c-Jun provides a potential link to its role in tumorigenesis. Cell Cycle 3:111–113.[Web of Science][Medline]
  38. Begon D.Y., Delacroix L., Vernimmen D., Jackers P., Winkler R. (2005) Yin Yang 1 cooperates with activator protein 2 to stimulate ERBB2 gene expression in mammary cancer cells. J. Biol. Chem. 280:24428–24434.[Abstract/Free Full Text]
  39. Seligson D., Horvath S., Huerta-Yepez S., et al. (2005) Expression of transcription factor Yin Yang 1 in prostate cancer. Int. J. Oncol. 27:131–141.[Web of Science][Medline]
  40. Wang C.C., Tsai M.F., Hong T.M., Chang G.C., Chen C.Y., Yang W.M., Chen J.J., Yang P.C. (2005) The transcriptional factor YY1 upregulates the novel invasion suppressor HLJ1 expression and inhibits cancer cell invasion. Oncogene 24:4081–4093.[Web of Science][Medline]
  41. Duan J., Wainwright M.S., Comeron J.M., Saitou N., Sanders A.R., Gelernter J., Gejman P.V. (2003) Synonymous mutations in the human dopamine receptor D2 (DRD2) affect mRNA stability and synthesis of the receptor. Hum. Mol. Gen. 12:205–216.[Abstract/Free Full Text]
  42. Goode E.L., Pharoah P.D.P., Wareham N., Easton D.F. (2005) No evidence for population substructure within the EPIC-Norfolk cohort. In Proceedings of 96th annual meeting of the American Association of Cancer Research..
  43. de Bakker P.I., Yelensky R., Pe'er I., Gabriel S.B., Daly M.J., Altshuler D. (2005) Efficiency and power in genetic association studies. Nat. Genet. 37:1217–1223.[CrossRef][Web of Science][Medline]
Received December 16, 2005; revised January 3, 2006; accepted February 6, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
D. L. Foley, J. M. Craig, R. Morley, C. J. Olsson, T. Dwyer, K. Smith, and R. Saffery
Prospects for Epigenetic Epidemiology
Am. J. Epidemiol., February 15, 2009; 169(4): 389 - 400.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
L. S. Kristensen and A. Dobrovic
Direct Genotyping of Single Nucleotide Polymorphisms in Methyl Metabolism Genes Using Probe-Free High-Resolution Melting Analysis
Cancer Epidemiol. Biomarkers Prev., May 1, 2008; 17(5): 1240 - 1247.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. E. Kelemen, T. A. Sellers, J. M. Schildkraut, J. M. Cunningham, R. A. Vierkant, V. S. Pankratz, Z. S. Fredericksen, M. K. Gadre, D. N. Rider, M. Liebow, et al.
Genetic Variation in the One-Carbon Transfer Pathway and Ovarian Cancer Risk
Cancer Res., April 1, 2008; 68(7): 2498 - 2506.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
A. Y. Jung, E. M. Poole, J. Bigler, J. Whitton, J. D. Potter, and C. M. Ulrich
DNA Methyltransferase and Alcohol Dehydrogenase: Gene-Nutrient Interactions in Relation to Risk of Colorectal Polyps
Cancer Epidemiol. Biomarkers Prev., February 1, 2008; 17(2): 330 - 338.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
M. K. Schmidt, S. Reincke, A. Broeks, L. M. Braaf, F. B.L. Hogervorst, R. A.E.M. Tollenaar, N. Johnson, O. Fletcher, J. Peto, J. Tommiska, et al.
Do MDM2 SNP309 and TP53 R72P Interact in Breast Cancer Susceptibility? A Large Pooled Series from the Breast Cancer Association Consortium
Cancer Res., October 1, 2007; 67(19): 9584 - 9590.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
K.-A. Yoon, S. Park, B. Hwangbo, H. D. Shin, H. S. Cheong, H.-R. Shin, and J. S. Lee
Genetic polymorphisms in the Rb-binding zinc finger gene RIZ and the risk of lung cancer
Carcinogenesis, September 1, 2007; 28(9): 1971 - 1977.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
A. Miremadi, M. Z. Oestergaard, P. D.P. Pharoah, and C. Caldas
Cancer genetics of epigenetic genes
Hum. Mol. Genet., April 15, 2007; 16(R1): R28 - R49.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Material
Right arrow All Versions of this Article:
27/8/1661    most recent
bgi375v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Cebrian, A.
Right arrow Articles by Ponder, B. A.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cebrian, A.
Right arrow Articles by Ponder, B. A.J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?