Skip Navigation

Human Molecular Genetics 2007 16(R1):R28-R49; doi:10.1093/hmg/ddm021
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
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 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 arrowRequest Permissions
Google Scholar
Right arrow Articles by Miremadi, A.
Right arrow Articles by Caldas, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miremadi, A.
Right arrow Articles by Caldas, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Cancer genetics of epigenetic genes

Ahmad Miremadi1, Mikkel Z. Oestergaard2, Paul D.P. Pharoah1,2 and Carlos Caldas1,*

1 Cancer Genomics Program, Department of Oncology, Hutchison/MRC Research Centre and 2 Department of Public Health, Strangeways Research Laboratories, University of Cambridge, Cambridge, UK

* To whom correspondence should be addressed at: Breast Cancer Functional Genomics Laboratory, Cancer Research UK, Cambridge Research Institute and Department of Oncology, University of Cambridge, Li Ka-Shing Centre, Robinson Way, Cambridge CB2 0RE, UK. Email: cc234{at}cam.ac.uk

Received January 9, 2007; Revised February 7, 2007;
   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
The cancer epigenome is characterised by specific DNA methylation and chromatin modification patterns. The proteins that mediate these changes are encoded by the epigenetics genes here defined as: DNA methyltransferases (DNMT), methyl-CpG-binding domain (MBD) proteins, histone acetyltransferases (HAT), histone deacetylases (HDAC), histone methyltransferases (HMT) and histone demethylases. We review the evidence that these genes can be targeted by mutations and expression changes in human cancers.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
Neoplasias have a distinct pattern of disrupted pathways, which are the result not only of genetic alterations but also of heritable patterns of disrupted gene expression (1). Epigenetics refers to these clonal changes in patterns of gene expression that are mediated by mechanisms that do not alter the primary DNA sequence.

Epigenetic changes in tumors mostly result in inappropriate gene silencing (2,3). The transcriptionally silenced state is the consequence of a concert of alterations in chromatin structure, including CpG island hypermethylation and histone modification (reviewed in 2,3). Epigenetic changes therefore are the result of two processes, DNA methylation and histone modifications. These processes are interlinked and the nature of this relationship is only now starting to be understood, in particular how these events differ between normal stem/precursor cells and neoplastic cells (reviewed in 4). A direct link between the two processes has been recently unravelled with the report that EZH2, a histone methylase, recruits DNA methyltransferases (DNMTs) to selected target genes (5).

DNA global hypomethylation and promoter-localized hypermethylation were, until recently, felt to be the key components of the cancer epigenome. Of these two, the better characterized epigenetic abnormality in cancer is the aberrant hypermethylation of promoter region CpG islands (2,3). Recent findings suggest that the picture is much more complex, with cancer-specific DNA hypermethylation (associated with histone methylation) affecting whole gene ‘neighborhoods’ up to an entire chromosome band (6). It is also becoming clear that cancers have also altered patterns of histone modification, with loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 as a common hallmark (7). These global histone modification patterns in cancer might have prognostic value (8). More recently, a novel mechanism of cancer-specific loss of expression of neighboring genes as a result of aberrant histone methylation without DNA methylation has been reported (9).

The proteins responsible for the alterations characteristic of the cancer epigenome are the enzymes that catalyze DNA methylation, the proteins that bind methylated DNA at promoters and contribute to silencing and the chromatin modifier enzymes that catalyze histone acetylation, deacetylation, methylation and demethylation. Here we review the evidence that genes encoding these proteins are themselves targeted during tumor initiation and progression. For the purposes of this review, we narrowed down the definition of epigenetic genes, leaving out, for example, members of the SWI/SNF complex, Polycomb group 1 proteins and small non-coding RNAs, all of which might also be involved directly or indirectly in epigenetic silencing (10).

Genetic abnormalities associated with cancer genes may occur as germline and/or somatic events. Germline variants can be classified as highly penetrant mutations that cause syndromes associated with cancer predisposition or as low to modest penetrance alleles that have a small effect on cancer risk. Somatic mutations include small sequence changes that activate/inactivate gene function, gross chromosomal rearrangements (resulting in translocations, deletions, duplications and amplifications) and expression changes. In this review, we have searched the literature for reports of both germline and somatic alterations in our list of epigenetic genes in human cancers.


   METHODOLOGY OF REVIEW
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
We defined as epigenetic genes those belonging to the following families: DNMTs, methyl-CpG-binding domain (MBD) proteins, histone acetyltransferases (HAT), histone deacetylases (HDAC), histone methyltransferases (HMT) and histone demethylases (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. List of epigenetic genes

 
GeneOntology (11) was searched (November 2006) for the following molecular function attributes: DNMT activity (GO:0009008), methyl-CpG binding (GO:0008327), HAT activity (GO:0004402), HDAC activity (GO:0000118), HMT activity (GO:0042054) and histone demethylase activity (GO:0051864). Only human genes were included. For protein complexes annotated with one of the functions, we only included component gene products if the component itself was annotated with the function. None of the evidence codes was Inferred from Electronic Annotation. Additional genes were added from previous literature reviews when the GO annotation was missing (see references in Table 1).

To investigate associations of human germline variants and risk of cancer, we carried out a systematic literature search of published studies using the Medline (National Library of Medicine, Washington, DC, USA) database. For studies of low-penetrance variants, our inclusion criteria were: the total number of subjects in a study should be greater than 400 for population-based association studies and greater than 200 for studies using familial cases (cases enriched for family history of cancer) (12). The included low-penetrance studies (n = 30) are summarized in Table 2. We regarded a study result as ‘significant’ if the main effect was significant (at {alpha} = 5%) and/or if a subgroup analysis with greater than 100 total subjects was significant (at {alpha} = 1%). To evaluate publication bias in low-penetrance association studies, we classified a study as ‘small’ or ‘large’ based on the number of total subjects. A study was classified as small if it had fewer than 800 total subjects in population-based association studies or less than 400 in association studies with familial cases. The remaining studies were classified as large studies. There was no indication of publication bias. For both small (n = 12) and large studies (n = 18) 50% showed significant associations.


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of low-penetrance association studies of epigenetic genes and cancer associated phenotypes

 
We identified three genes where germline mutations causing hereditary syndromes in which patients have an increased risk of cancer have been reported: Sotos syndrome (NSD1) and Rubinstein–Taybi syndrome (CREBBP and EP300).

Similarly, Medline was searched for publications reporting cancer-related somatic alterations in the epigenetic genes, including mutations, chromosomal rearrangements and expression changes. The details and references are listed in separate tables for each family of epigenetic genes (Tables 3–7).


View this table:
[in this window]
[in a new window]

  		Table 3. DNMT: somatic changes in cancer

 

   DNA METHYLTRANSFERASES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
Methylation of CpG islands in gene promoters is known to have an important regulatory effect on gene expression and aberrant patterns of DNA methylation have been implicated in carcinogenesis (2–4). The DNMTs are a group of enzymes responsible for transfer of methyl groups to cytosine (2–4). Four members of this family have been identified in mammals: DNMT1, TRDMT1, DNMT3A and DNMT3B.

Germline variants
Germline single nucleotide polymorphisms (SNPs) in DNMT3B have been suggested to be associated with risk of breast cancer (–149C > T) (29), lung adenocarcinoma (–283T > C; –579G > T) (30) and lung cancer (–149C > T) (31) (Table 2). The three investigated SNPs are all in the promoter region of DNMT3B. The SNPs –283T > C and –579G > T were in strong linkage disequilibrium in a Korean population (30). In vitro functional assays by Lee et al. (30) attributed the association with lung cancer to a significantly lower promoter activity of the –283T allele. The mechanism for the suggested association between the SNP –149C > T and lung cancer is unknown but the T allele is hypothesized to increase the promoter activity of DNMT3B (31).

The reported association between the SNP –149C > T in DNMT3B and risk of breast cancer by Montgomery et al. (29) was not confirmed in the UK-based study by Cebrian et al. (32), which has the largest sample size to date of epigenetic germline cancer studies. Furthermore, the SNP was not found to be associated with DNA methylation levels at CpG islands in six suggested cancer genes (RARß2, CDH1, ER, BRCA1, CCND2, p16 and TWIST) in breast cancer cells, which suggests that if an association between the SNP and breast cancer is real, the etiology is unlikely to include alterations in CpG island promoter methylation (33).

Somatic changes
Overexpression of DNMTs occurs in many human cancers and overexpression of DNMTs has been associated with hypermethylation of CpG islands in some cancers, although this has not been confirmed in recent studies (59–74) (Table 3). In one study, when the levels of DNMT1, DNMT3A, DNMT3B and MBDs in lung cancer cell lines were normalized against PCNA (proliferating cell nuclear antigen) no overexpression was observed (74), suggesting ‘overexpression’ of some of these genes may be a reflection of increased cell proliferation.

Mutations of DNMT1 have been reported in two out of 29 colorectal cancers, suggesting that inactivation of DNMT1 can be involved in carcinogenesis (75). Mice deficient in Dnmt1 show a tumor resistance phenotype on an ApcMin/+ background (76). This has been attributed to a decrease in aberrant methylation of tumor suppressor genes (77). On the other hand, reduced expression of Dnmt1 results in a reduced global DNA methylation level which is associated with genomic instability. In a mouse model for sarcomas, Dnmt1-deficient mice developed sarcomas at an earlier age (78). Another study showed that mice carrying a hypomorphic and a null allele of Dnmt1 developed aggressive T cell lymphomas (79). Taken together, these murine data suggest that DNMT could behave as a tumor suppressor gene.


   MBD PROTEINS
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
MBD proteins bind to methylated CpG islands and mediate transcriptional repression of affected genes. The MBD family of proteins consists of five members: MeCP2, MBD1, MBD2, MBD3 and MBD4 (reviewed in 81).

Germline variants
Three polymorphisms in MBD1 (634G > A, 501delT and Pro401Ala) have been investigated for their association with lung cancer risk (36) (Table 2). The study suggests that the G/G genotype for the SNP 634G > A increases the overall risk of lung cancer. Further subgroup analyses suggested that all three polymorphisms are associated with an increased risk of lung adenocarcinoma. The three polymorphisms are in strong linkage disequilibrium, and in vitro analysis of the promoter activity of the variants of 634G > A and 501delT attributed much of the functional effect to lower MBD1 expression from the G allele of 634G > A.

For MBD2 Zhu et al. (37) found a reduced risk of premenopausal breast cancer for the rare variants of two SNPs (rs1259938 and rs609791). No association was observed in postmenopausal women. The two SNPs, rs1259938 and rs609791, are non-coding and intronic, respectively. However, the number of premenopausal women in the study was very small (<90 cases and <150 controls), and the finding needs further investigation. Cebrian et al. (32) found no evidence of an association between SNPs or haplotypes in MBD2 and breast cancer risk in a similar population in a study that had more than 10 times the sample size of Zhu et al., but sub-group analyses by menopausal status were not reported. Previous findings of heterogeneity in breast cancer etiology between pre- and postmenopausal women suggest that further studies in these populations are required (82,83).

In MBD4, the SNP rs140693 (Glu346Lys) has been suggested to alter the risk of cancer. The polymorphism results in a neutral amino acid substitution that does not lie in any known functional domains of the protein (38). The Lys/Lys genotype was reported to increase the risk of esophageal squamos cell carcinoma by Hao et al. (40). Shin et al. (38) reported a reduced risk of lung adenocarcinoma associated with the Lys/Lys genotype in a Korean population, whereas Sakiyama et al. (39) reported a null association with lung adenocarcinoma and squamous cell carcinoma in a Japanese population.

Somatic changes
The role of somatic alterations in MBD proteins in cancer is unclear. It has been shown that MBD2 binds to the aberrantly methylated promoter of tumor suppressor genes (e.g. p14/ARF and p16/Ink4A in colon cancer cell lines) and suppresses their expression (84). Abnormalities in the expression level of MBD proteins in cancer cell lines, however, may be due to the increased cell proliferation in these cells (74). MECP2 overexpression has also been observed in breast cancer and appears to be associated with oestrogen receptor positivity (85).

The study of Mbd2-deficient mice crossed on to an ApcMin/+ background has shown that these mice are resistant to the development of intestinal tumors and the dosage of Mbd2 is important for the tumor resistant effect (86).


   HISTONE ACETYLTRANSFERASES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
The acetylation of lysine residues on the N-terminus of histones is generally associated with active gene transcription (15,16). The HATs can be grouped into three main families based on their sequence similarities: Gcn5/PCAF, p300/CBP and the MYST family of HAT proteins (reviewed in 15). Most HATs are present as part of large protein complexes and act as transcriptional coactivators (15,16). Many of them have also been shown to acetylate proteins other than histones (15,16).

Germline variants
NCOA3 (nuclear receptor coactivator-3), also called AIB-1, encodes a protein that interacts with nuclear hormone receptors and facilitates hormone-dependent transcription (87). A CAG/CAA repeat length polymorphism has been suggested to modify the risk of breast cancer for carriers of high-penetrance mutations in BRCA1 (42,47,48) and BRCA2 (43), to be associated with mammographic density (50), a phenotype that has been associated with the risk of breast cancer (88,89) (Table 2). The CAG/GAA repeat length polymorphism was reported to predispose to prostate cancer in a Chinese population (52) but to have no effect on prostate cancer susceptibility in a Caucasian population (53). The polymorphism results in variable number of glutamine residues within the C-terminal region of NCOA3 of which the functional significance is unknown.

The two largest studies examining the NCOA3 CAG/CAA repeat length polymorphism and breast cancer risk modification in BRCA1/2 mutation carriers have not found any evidence of risk modification for BRCA1 (41,43). Hughes et al. (43) reported a reduced breast cancer risk for BRCA2 mutation carriers [harzard ratio (95% CI): 0.67 (0.51–0.88)] for women carrying two alleles of NCOA3 with 28 or more repeats compared with women with at least one allele with 27 or fewer repeats. However, Hughes et al. did not find any risk modification for women carrying two alleles with 29 or more repeats, which provides little biological corroboration for their results. Furthermore, Spurdle et al. (41) did not find evidence of a modification of breast cancer risk in the largest BRCA2 mutation carrier cohort to date.

Burwinkel et al. (46) reported that the rare variants of two NCOA3 SNPs rs2230782 (Q586H) and rs2076546 (T960T) had protective roles for development of breast cancer in the general population. In silico analysis showed that the gln586 is highly conserved compared with orthologous and homologous genes and the authors hypothesized that the T960T polymorphism may influence the translation and/or transcriptional efficiency of NCOA3. In summary, the role of NCOA3 gene as a low penetrance cancer predisposition allele has not been established and the evidence remains conflicting.

Heterozygous germline mutations in CREBBP (CREB-binding protein) have, for more than a decade, been known to cause the Rubinstein–Taybi syndrome (90). The syndrome (MIM 180849 [OMIM] ) is a developmental disorder characterized by mental retardation and growth abnormalities such as facial abnormalities, broad thumbs and broad big toes. Patients have an increased risk of formation of various tumors, in particular tumors in the brain or nervous system (91). As many as ~61% of patients have mutations in CREBBP (92). Recently, Roelfsema et al. (93) identified three out of 92 Rubinstein–Taybi syndrome patients with causative mutations in EP300.

Germline mutations in EP300 have not been found in BRCA1 and BRCA2 negative families with breast cancer and gastric, pancreatic or colorectal cancer (54). Furthermore, a large association study failed to show any association between EP300/CREBBP SNPs and haplotypes and increased breast cancer risk (32).

Somatic changes
Mutations in a number of HATs have been observed in solid tumors (reviewed in 94) (Table 4). Biallelic mutations of EP300 have been identified in epithelial tumors (95,96). Furthermore, both EP300 and CREBBP, as well as other HATs (MYST3, MYST4), are commonly involved in chromosomal translocations in hematological malignancies, and less commonly in solid tumors (MYST4, NCOA1) (Table 4). It appears that these translocations are involved in leukemogenesis through aberrant acetylation caused by mistargeting of HATs. NCOA3 (AIB1) is frequently amplified and overexpressed in human breast cancer and behaves as a classical oncogene (87,97–101).


View this table:
[in this window]
[in a new window]

  		Table 4. HAT: somatic changes in cancer

 

   HISTONE DEACETYLASES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
HDACs promote gene repression through removal of acetyl groups from lysine residues in histone tails. At least 18 HDAC genes have been recognized in the human genome, grouped into three main classes based on sequence homology to the yeast counterparts Rpd3, Hda1 and Sir2/Hst. HDACs act mostly as part of large multiprotein complexes that function as transcriptional corepressors.

Germline variants
Germline variants in HDAC3, HDAC4 and HDAC5 were examined for lung cancer risk by Park et al. (55) and HDAC2 and HDAC5 for breast cancer risk by Cebrian et al. (32) (Table 2). Neither study found evidence of association. An insertion of a CAG triplet in the 5'-UTR of HDAC2 was recently identified in 18% of 181 cancer samples tested versus 10% of 192 normal DNA controls (P < 0.01, Fisher's exact test) (127), but this potential cancer susceptibility association needs to be confirmed in larger studies.

Somatic changes
HDACs are implicated in cancer partly through their aberrant recruitment and consequent silencing of tumor suppressor genes. For example, inactivation of p21WAF1 is associated with hypoacetylation of its promoter and can be reversed by HDAC inhibitors through hyperacetylation of the histones in the promoter (136).

Recently, two independent reports identified truncating mutations of HDAC2 in human epithelial cancers with microsatellite instability (MSI) (Table 5). One group screened HDAC1 and HDAC2 for mutations in 181 cancer samples (116 primary tumors of breast, ovarian and colorectal origin and 65 cancer cell lines of breast, ovarian, lung, pancreatic and colorectal origin) and identified a single nucleotide deletion, resulting in a frameshift in exon 12 of HDAC2 in the HCT15 colorectal cell line (127). The other group screened six colorectal and four endometrial cancer cell lines with MSI for all the exonic mononucleotide repeats in the coding sequences of HDAC1 and HDAC2 (in addition to pCAF, G9a/EHMT2, DNMT1, DNMT3b, MBD1, MBD2 and MeCP2) and found no mutations except for the A9 repeat of exon 1 of HDAC2 (137). This truncating mutation in HDAC2 was detected in colonic, gastric and endometrial primary tumors with MSI and absent in normal tissues and MSI–colorectal cancers. The mutation was shown in functional assays to confer resistance to the antiproliferative and proapoptotic effects of HDAC inhibitors (137).


View this table:
[in this window]
[in a new window]

  		Table 5. HDAC: somatic changes in cancer

 
HDAC4 mutations have been identified in breast cancer samples at significant frequency in the recent large-scale sequencing study of breast and colorectal cancers (138).

Overexpression of HDAC1 has been reported in prostate and gastric cancers (139,140). In a recent comprehensive expression study, HDAC1, HDAC2, HDAC4, HDAC5, HDAC7a and SIRT1 were profiled by RT–PCR in a large number of human epithelial malignancies (127). These studies are of particular clinical importance as HDAC inhibitors are considered as a new class of anticancer drugs.


   HISTONE METHYLTRANSFERASES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 
Methylation of arginine and lysine residues of histones is involved in the regulation of a wide range of processes including gene activity, chromatin structure and epigenetic memory (reviewed in 142). Arginine can be either mono- or dimethylated, with the latter in symmetric or asymmetric configurations. Lysine can be in mono-, di- or trimethylated forms. In general, lysine methylation at H3K9, H3K27 and H4K20 is associated with gene silencing, whereas methylation at H3K4, H3K36 and H3K79 is associated with gene activation. However, it is now clear that these correlations are far from absolute and the presence of other histone modifications as well as the mono-, di- or tri-state of methylation can alter the functional consequences of these modifications.

Germline variants
Tsuge et al. (58) reported a strong increase in the risk of breast cancer, colorectal cancer and hepatocellular carcinoma for individuals homozygous for a tree repeat unit (CCGCC) in the regulatory region of SMYD3 compared with the combined group of heterozygous + homozygous for two repeats units (Table 2). The odds ratios (95% CI) were: for breast cancer, 4.48 (2.71–7.41); colorectal cancer, 2.58 (1.68–3.94) and for hepatocellular carcinoma, 3.50 (2.22–5.51). Functional assays of plasmids containing the variable number tandem repeat (VNTR) polymorphism showed that the transcriptional factor E2F-1 binds to the tandem-repeat sequence and that plasmids containing the three repeat unit had higher transcriptional activation than alleles with the two repeat unit. Frank et al. (57) recently investigated the same VNTR polymorphism in a case–control study of familial breast cancer but did not find evidence of an association.

Intragenic mutations and microdeletions encompassing the gene NSD1 (nuclear receptor SET domain-containing protein 1) are causative and specific to the Sotos syndrome (MIM 117550 [OMIM] ) (143). The syndrome is characterized by facial dysmorphism, learning disabilities and childhood overgrowth and has been suggested as a cancer syndrome. Because of reporter bias, the estimate of an increased risk of cancer for patients of Sotos syndrome is uncertain but has been reported to be in the range of 2–3% (144,145). The type and sites of origin of tumors associated with the syndrome vary greatly, but include Wilms tumor, neuroblastoma and hepatocellular carcinoma (145–147). Tatton-Brown et al. (143) studied 530 patients with phenotypic characteristics similar to the Sotos syndrome. They identified 266 patients with NSD1 abnormalities and 99% of these were diagnosed with the Sotos syndrome (the penetrance estimate is likely to be lower in the general population as phenotype was selected for in the study).

Somatic changes
Lysine HMTs

  • MLL (SET1) family
  • The mixed lineage leukemia (MLL) gene is located on chromosome (11q23), recognized as a recurrent locus of chromosomal translocation in AML and ALL (156). More than 50 translocations of MLL have been reported with distinct fusion partners, and cases with MLL translocations are associated with an intermediate to poor prognosis (156) (Table 6). Partial tandem duplications (PTDs) of MLL have also been observed in acute leukemias with trisomy of chromosome 11 or normal karyotype (148). Clinical studies suggest that MLL-PTD confers a worse prognosis with shortened overall and event-free survival in childhood and adult AML (reviewed in 148). It is notable, however, that scrambled transcripts of MLL are seen in both normal and leukemic cells and RT–PCR based methods cannot be used for the detection of genomic MLL rearrangements (149). MLL amplification has also been recognized in cases of acute myeloid leukemia and myelodysplastic syndrome suggesting an etiologic role for MLL gain of function in myeloid malignancies (194–204). MLL encodes a large protein with several domains involved in transcriptional regulation. The N-terminal part of the protein contains three AT-hooks, two sub-nuclear localization domains (SNL1 and SNL2) and two repression domains (RD1 and RD2). The main region of gene fusion lies to the C-terminal side of the repression domain, known as the major break domain (MBR) (156). The region distal to the MBR, including the SET domain, is lost in the fusion protein. But it must be noted that several of the fusion partners may be HMTs themselves. A total of 87 different MLL rearrangements have been identified, of which 51 have been characterized at the molecular level. The four most frequently found partners are AF4, AF9, ENL and AF10 (156). These encode nuclear proteins that are involved in H3K79 methylation, suggesting a functional selection of MLL translocation partners. Knock-in and translocator mouse models have been used to study the function of MLL fusion partners (150–155). MLL appears to be a major regulator of class I homeobox (HOX) genes, a group of transcription factors involved in embryonic development and hematopoietic cell differentiation (156). Disturbances of HOX expression patterns have been observed in leukemias with MLL fusion proteins. The mechanism for dysregulation of HOX genes is not clear, but can be due to the acquisition of an activation domain in the fusion protein or dimerization of the MLL N-terminus in the fusion protein (reviewed in 156).
  • MLL3, a homolog of ALR/MLL2, maps to chromosome 7q36, an area commonly deleted in hematological malignancies (157,158) and recently was identified as a novel colorectal cancer gene, with six mutations, two of which truncating (138). MLL4 (formerly known as MLL2), the closest MLL homolog, is amplified in solid tumor cell lines of pancreatic and glial origin (159).
  • Enhancer of Zeste (EZH2) is part of the Polycomb repressive complex 2 (PRC2) which also includes EED, SUZ12, RbAp48, and is involved in silencing (reviewed in 10). EZH2 contains the highly conserved SET domain and exerts its silencing function through methylation of H3 Lys27. The study of genes silenced by the PcG shows a strong preference for genes involved in cell fate decisions, including genes in Hox, Notch, Hedgehog, Wnt and TGF signaling. EZH2 overexpression has been observed in prostate, breast, endometrium and bladder cancer. In breast cancer, its overexpression is associated with poor prognosis. Although the mechanism of action of EZH2 in cancer is not yet clear, it appears to play a role in the regulation of pRB–E2F growth control pathway, as well as genes involved in homologous recombination pathway of DNA repair (160). A recent study (161) has shown that EZH2 is also overexpressed in preneoplastic breast lesions, as well as morphologically normal breast epithelium adjacent to the pre-invasive and invasive lesions, and may therefore mark epithelium at higher risk for neoplastic transformation.
  • RIZ family.
  • Several RIZ family members have been implicated as cancer genes, mostly as a result of point mutations or chromosomal rearrangements.
  • PRDM1 (BLIMP1) is frequently mutated in human B-cell lymphomas, mostly truncating mutations associated with inactivation of the second allele, pointing to a role as a tumor suppressor gene (162,163).
  • PRDM2 (RIZ1) maps to chromosome 1p36, a region commonly deleted in cancer. Both frameshift and missense mutations affect PRDM2 in human cancers and cancer cell lines (164–167). Targeted inactivating mutations of PRDM2 in mice are associated with development of large B-cell lymphoma and a range of unusual tumors (168). PRDM2 shows underexpression in a number of tumors including breast, colon, liver and lung cancers, as well as neuroblastoma, melanoma and osteosarcomas (164,169,170).
  • PRDM3 (EVI1) maps to 3q26 and is targeted in myeloid leukemias by chromosomal rearrangement resulting in overexpression (171,172).
  • PRDM12 (PFM9) is the strongest candidate tumor suppressor gene that is deleted at 9q33–q34 in poor outcome chronic myeloid leukemia (173,174).
  • PRDM16 (MEL1, PFM13) maps to 1p36.23–p33 and is targeted by t(1:3) (p36;q21) in both myelodysplastic syndrome and acute myeloid leukemias and is also aberrantly expressed in T-cell leukemia (175–177).
  • SET2 family.
  • NSD1 (nuclear receptor binding SET domain protein 1) is involved in translocations with a nucleoporin gene (NUP98) to produce a fusion protein in childhood acute myeloid leukemia (Table 6).
  • NSD2 is the gene disrupted by the t(4:14) in multyple myeloma producing an IgH/NSD2 hybrid transcript resulting in overexpression of transcripts originating from the NSD2(MMSET/WHSC1) locus (178,179).
  • NSD3 was the third cloned family member, maps to 8p12, and is amplified in human breast cancer cell lines and primary tumors (180), and subsequently was identified at the breakpoint of t(8;11)(p11.2;p15), resulting in a fusion of the NUP98 and NSD genes (181).
  • Other lysine HMTs.
  • SMYD3 (Set and MYND containing 3) is overexpressed in colorectal, hepatocellular and breast cancer (Table 6).


View this table:
[in this window]
[in a new window]

  		Table 6. HMTs: somatic changes in cancer

 
Arginine HMTs
The evidence for the involvement of arginine HMTs in human cancers is not as solid but underexpression of PRMT1 in breast cancer (182) and overexpression of PRMT5 (SKB1) in gastric cancer (183) have been described.


   HISTONE DEMETHYLASES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
DISCUSSION
 REFERENCES
 
Three distinct classes of histone demethylases have now been recognized (reviewed in 24). The first class includes PADI4 (petidylarginine deiminase 4) which functions by converting a methyl-lysine to citrulline. The second class includes LSD1 (lysine-specific demethylase 1) which can reverse histone H3K4 and H3K9 modifications by an oxidative demethylation reaction. The third class of demethylases, and the largest so far, includes Jumonji C (JmjC)-domain containing histone demethylases (JHDMs) which, unlike LSD1, are capable of demethylating all three methylated states (mono-, di- and tri-methylated lysine). So far, JHDMs have been shown to demethylate H3K36 (JHDM1), H3K9 (JHDM2A) and H3K9/K27 (JHDM3 and JMJD2A-D).

Somatic changes
The link between this novel group of enzymes and cancer is still unclear (Table 7). However, a number of genes previously linked to cancer have recently been recognized as possible histone demethylases. JARID1B (PLU-1) is a putative histone demethylase, containing a JmjN domain and is shown to be overexpressed in breast cancers (292). JMJD2C (GASC1), a putative oncogene which is amplified in squamous carcinoma, belongs to the JMJD2 subfamily of the Jumonji family of histone demethylases (293).


View this table:
[in this window]
[in a new window]

  		Table 7. Histone demethylases: somatic changes in cancer

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
REFERENCES
 
The role of epigenetic alterations in cancer development has been the focus of increasing interest in recent years. These epigenetic abnormalities are present in almost all cancers and, alongside genetic changes, drive tumor progression. The mechanisms underlying the epigenetic alterations however are still largely unknown, namely whether epigenetic genes are directly involved in cancer.

Several lines of evidence are now converging to emphasize the role of genetic changes in DNMT, methyl-binding protein and histone modifier genes (‘epigenetic genes’) in cancer. These include studies on germline variation, somatic mutations and aberrant expression patterns of these genes. The upshot of these studies is that several epigenetic genes are clearly targeted in cancer initiation and progression, and therefore they should be added to a growing list of oncogenes and tumor suppressor genes.

The culmination of data on some of these epigenetic genes has started to uncover their role as tumor suppressor genes and provide insights into the mechanistic routes of their action. Germline mutations in CREBBP and EP300 are associated with increased predisposition to childhood cancers in Rubinstein–Taybi syndrome and biallelic mutations in both these genes have been identified in solid tumors (90–93,95,96,102–107,111,112). We believe therefore that both genes should now be classified as classical tumor suppressor genes. In the case of EP300, there is functional evidence that besides its epigenetic roles the protein might also be involved in the regulation of the TP53-dependent response to DNA damage and the G1-S transition of the cell cycle (295,296). The role of common germline genetic variation is less clear. Despite a substantial body of work and the reporting of associations in individual studies, no definitive low to moderate penetrance alleles, confirmed in several studies at stringent levels of statistical significance, have emerged.

Our catalog of epigenetic genes somatic alterations in cancer is improving as a result of large-scale studies of mutations in common cancers. For example, a recent study has identified mutations in two epigenetic genes, HDAC4 and MLL3, in breast and colorectal cancers, respectively (138). Further systematic studies of mutations in epigenetic genes may uncover other somatic alterations and help to clarify their role in tumorigenesis.

In addition to germline variation and somatic mutations, many of the epigenetic genes show aberrant patterns of expression in neoplastic cells, as reviewed here. A recent comprehensive analysis of the expression alterations of a number of these genes provides evidence for the existence of characteristic epigenetic gene expression signatures that can be used either to distinguish cancer and normal tissue or to classify different cancers (127).

Another important addition to our understanding of the role of epigenetic genes in carcinogenesis comes from recent studies examining global histone modifications in cancer. These post-translational histone modifications affect large regions of chromatin, including coding, promoter and non-promoter sequences, and have been shown to be associated with clinical outcome (7,8,297). An area of special interest for future research will be to correlate alterations in the epigenetic genes with patterns of global histone modifications and DNA methylation. Such correlations can provide an insight into the mechanisms by which these epigenetic genes contribute to the development and progression of cancer and might provide novel diagnostic and therapeutic insights.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODOLOGY OF REVIEW
 DNA METHYLTRANSFERASES
 MBD PROTEINS
 HISTONE ACETYLTRANSFERASES
 HISTONE DEACETYLASES
 HISTONE METHYLTRANSFERASES
 HISTONE DEMETHYLASES
 DISCUSSION
 REFERENCES
 

  1. Hanahan D., Weinberg R.A. The hallmarks of cancer. Cell (2000) 100:57–70.[CrossRef][ISI][Medline]

  2. Jones P.A., Baylin S.B. The fundamental role of epigenetic events in cancer. Nat. Rev. Genet. (2002) 3:415–428.[ISI][Medline]

  3. Laird P.W. Cancer epigenetics. Hum. Mol. Genet. (2005) 14(Spec. no. 1):R65–R76.[Abstract/Free Full Text]

  4. Ting A.H., McGarvey K.M., Baylin S.B. The cancer epigenome—components and functional correlates. Genes Dev. (2006) 20:3215–3231.[Abstract/Free Full Text]

  5. Vire E., Brenner C., Deplus R., Blanchon L., Fraga M., Didelot C., Morey L., Van Eynde A., Bernard D., Vanderwinden J.M., et al. The Polycomb group protein EZH2 directly controls DNA methylation. Nature (2006) 439:871–874.[CrossRef][Medline]

  6. Frigola J., Song J., Stirzaker C., Hinshelwood R.A., Peinado M.A., Clark S.J. Epigenetic remodeling in colorectal cancer results in coordinate gene suppression across an entire chromosome band. Nat. Genet. (2006) 38:540–549.[CrossRef][ISI][Medline]

  7. Fraga M.F., Ballestar E., Villar-Garea A., Boix-Chornet M., Espada J., Schotta G., Bonaldi T., Haydon C., Ropero S., Petrie K., et al. Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat. Genet. (2005) 37:391–400.[CrossRef][ISI][Medline]

  8. Seligson D.B., Horvath S., Shi T., Yu H., Tze S., Grunstein M., Kurdistani S.K. Global histone modification patterns predict risk of prostate cancer recurrence. Nature (2005) 435:1262–1266.[CrossRef][Medline]

  9. Stransky N., Vallot C., Reyal F., Bernard-Pierrot I., de Medina S.G., Segraves R., de Rycke Y., Elvin P., Cassidy A., Spraggon C., et al. Regional copy number-independent deregulation of transcription in cancer. Nat. Genet. (2006) 38:1386–1396.[CrossRef][ISI][Medline]

  10. Sparmann A., van Lohuizen M. Polycomb silencers control cell fate, development and cancer. Nat. Rev. Cancer (2006) 6:846–856.[CrossRef][ISI][Medline]

  11. Ashburner M., Ball C.A., Blake J.A., Botstein D., Butler H., Cherry J.M., Davis A.P., Dolinski K., Dwight S.S., Eppig J.T., et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. (2000) 25:25–29.[CrossRef][ISI][Medline]

  12. Antoniou A.C., Easton D.F. Polygenic inheritance of breast cancer: implications for design of association studies. Genet. Epidemiol. (2003) 25:190–202.[CrossRef][ISI][Medline]

  13. Esteller M. Epigenetics provides a new generation of oncogenes and tumor-suppressor genes. Br. J. Cancer (2006) 94:179–183.[CrossRef][ISI][Medline]

  14. Fraga M.F., Esteller M. Towards the human cancer epigenome: a first draft of histone modifications. Cell Cycle (2005) 4:1377–1381.[ISI][Medline]

  15. Yang X.J. The diverse superfamily of lysine acetyltransferases and their roles in leukemia and other diseases. Nucleic Acids Res. (2004) 32:959–976.[Abstract/Free Full Text]

  16. Santos-Rosa H., Caldas C. Chromatin modifier enzymes, the histone code and cancer. Eur. J. Cancer (2005) 41:2381–2402.[CrossRef][ISI][Medline]

  17. Gibbons R.J. Histone modifying and chromatin remodelling enzymes in cancer and dysplastic syndromes. Hum. Mol. Genet. (2005) 14(Spec no. 1):R85–R92.[Abstract/Free Full Text]

  18. Schultz D.C., Ayyanathan K., Negorev D., Maul G.G., Rauscher F.J. III. SETDB1: a novel KAP-1-associated histone H3, lysine 9-specific methyltransferase that contributes to HP1-mediated silencing of euchromatic genes by KRAB zinc-finger proteins. Genes Dev. (2002) 16:919–932.[Abstract/Free Full Text]

  19. Zhang Y., Leaves N.I., Anderson G.G., Ponting C.P., Broxholme J., Holt R., Edser P., Bhattacharyya S., Dunham A., Adcock I.M., et al. Positional cloning of a quantitative trait locus on chromosome 13q14 that influences immunoglobulin E levels and asthma. Nat. Genet. (2003) 34:181–186.[ISI][Medline]

  20. Ringrose L., Paro R. Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Annu. Rev. Genet. (2004) 38:413–443.[CrossRef][ISI][Medline]

  21. Schneider R., Bannister A.J., Kouzarides T. Unsafe SETs: histone lysine methyltransferases and cancer. Trends Biochem. Sci. (2002) 27:396–402.[CrossRef][ISI][Medline]

  22. Ruault M., Brun M.E., Ventura M., Roizes G., De Sario A. MLL3, a new human member of the TRX/MLL gene family, maps to 7q36, a chromosome region frequently deleted in myeloid leukemia. Gene (2002) 284:73–81.[CrossRef][ISI][Medline]

  23. Raaphorst F.M. Deregulated expression of Polycomb-group oncogenes in human malignant lymphomas and epithelial tumors. Hum. Mol. Genet. (2005) 14(Spec no. 1):R93–R100.[Abstract/Free Full Text]

  24. Klose R.J., Kallin E.M., Zhang Y. JmjC-domain-containing proteins and histone demethylation. Nat. Rev. (2006) 7:715–727.[CrossRef]

  25. Trojer P., Reinberg D. Histone lysine demethylases and their impact on epigenetics. Cell (2006) 125:213–217.[CrossRef][ISI][Medline]

  26. Klose R.J., Yamane K., Bae Y., Zhang D., Erdjument-Bromage H., Tempst P., Wong J., Zhang Y. The transcriptional repressor JHDM3A demethylates trimethyl histone H3 lysine 9 and lysine 36. Nature (2006) 442:312–316.[CrossRef][Medline]

  27. Cloos P.A., Christensen J., Agger K., Maiolica A., Rappsilber J., Antal T., Hansen K.H., Helin K. The putative oncogene GASC1 demethylates tri- and dimethylated lysine 9 on histone H3. Nature (2006) 442:307–311.[CrossRef][Medline]

  28. Yamane K., Toumazou C., Tsukada Y., Erdjument-Bromage H., Tempst P., Wong J., Zhang Y. JHDM2A, a JmjC-containing H3K9 demethylase, facilitates transcription activation by androgen receptor. Cell (2006) 125:483–495.[CrossRef][ISI][Medline]

  29. Montgomery K.G., Liu M.C., Eccles D.M., Campbell I.G. The DNMT3B C -> T promoter polymorphism and risk of breast cancer in a British population: a case–control study. Breast Cancer Res. (2004) 6:R390–R394.[CrossRef][ISI][Medline]

  30. Lee S.J., Jeon H.S., Jang J.S., Park S.H., Lee G.Y., Lee B.H., Kim C.H., Kang Y.M., Lee W.K., Kam S., et al. DNMT3B polymorphisms and risk of primary lung cancer. Carcinogenesis (2005) 26:403–409.[Abstract/Free Full Text]

  31. Shen H., Wang L., Spitz M.R., Hong W.K., Mao L., Wei Q. A novel polymorphism in human cytosine DNA-methyltransferase-3B promoter is associated with an increased risk of lung cancer. Cancer Res. (2002) 62:4992–4995.[Abstract/Free Full Text]

  32. Cebrian A., Pharoah P.D., Ahmed S., Ropero S., Fraga M.F., Smith P.L., Conroy D., Luben R., Perkins B., Easton D.F., et al. Genetic variants in epigenetic genes and breast cancer risk. Carcinogenesis (2006) 27:1661–1669.[Abstract/Free Full Text]

  33. Li S.Y., Rong M., Iacopetta B. Germ-line variants in methyl-group metabolism genes and susceptibility to DNA methylation in human breast cancer. Oncol. Rep. (2006) 15:221–225.[ISI][Medline]

  34. Chang K.P., Hao S.P., Liu C.T., Cheng M.H., Chang Y.L., Lee Y.S., Wang T.H., Tsai C.N. Promoter polymorphisms of DNMT3B and the risk of head and neck squamous cell carcinoma in Taiwan: a case–control study. Oral Oncol (2006) Published online on August 22, 2006, 10.1016/j.oraloncology.2006.04.006.

  35. Wang Y.M., Wang R., Wen D.G., Li Y., Guo W., Wang N., Wei L.Z., He Y.T., Chen Z.F., Zhang X.F., et al. Single nucleotide polymorphism in DNA methyltransferase 3B promoter and its association with gastric cardiac adenocarcinoma in North China. World J. Gastroenterol. (2005) 11:3623–3627.[Medline]

  36. Jang J.S., Lee S.J., Choi J.E., Cha S.I., Lee E.B., Park T.I., Kim C.H., Lee W.K., Kam S., Choi J.Y., et al. Methyl-CpG binding domain 1 gene polymorphisms and risk of primary lung cancer. Cancer Epidemiol. Biomarkers Prev. (2005) 14:2474–2480.[Abstract/Free Full Text]

  37. Zhu Y., Brown H.N., Zhang Y., Holford T.R., Zheng T. Genotypes and haplotypes of the methyl-CpG-binding domain 2 modify breast cancer risk dependent upon menopausal status. Breast Cancer Res. (2005) 7:R745–R752.[CrossRef][ISI][Medline]

  38. Shin M.C., Lee S.J., Choi J.E., Cha S.I., Kim C.H., Lee W.K., Kam S., Kang Y.M., Jung T.H., Park J.Y. Glu346Lys polymorphism in the methyl-CpG binding domain 4 gene and the risk of primary lung cancer. Jpn. J. Clin. Oncol. (2006) 36:483–488.[Abstract/Free Full Text]

  39. Sakiyama T., Kohno T., Mimaki S., Ohta T., Yanagitani N., Sobue T., Kunitoh H., Saito R., Shimizu K., Hirama C., et al. Association of amino acid substitution polymorphisms in DNA repair genes TP53, POLI, REV1 and LIG4 with lung cancer risk. Int. J. Cancer (2005) 114:730–737.[CrossRef][ISI][Medline]

  40. Hao B., Wang H., Zhou K., Li Y., Chen X., Zhou G., Zhu Y., Miao X., Tan W., Wei Q., et al. Identification of genetic variants in base excision repair pathway and their associations with risk of esophageal squamous cell carcinoma. Cancer Res. (2004) 64:4378–4384.[Abstract/Free Full Text]

  41. Spurdle A.B., Antoniou A.C., Kelemen L., Holland H., Peock S., Cook M.R., Smith P.L., Greene M.H., Simard J., Plourde M., et al. The AIB1 polyglutamine repeat does not modify breast cancer risk in BRCA1 and BRCA2 mutation carriers. Cancer Epidemiol. Biomarkers Prev. (2006) 15:76–79.[Abstract/Free Full Text]

  42. Colilla S., Kantoff P.W., Neuhausen S.L., Godwin A.K., Daly M.B., Narod S.A., Garber J.E., Lynch H.T., Brown M., Weber B.L., et al. The joint effect of smoking and AIB1 on breast cancer risk in BRCA1 mutation carriers. Carcinogenesis (2006) 27:599–605.[Abstract/Free Full Text]

  43. Hughes D.J., Ginolhac S.M., Coupier I., Barjhoux L., Gaborieau V., Bressac-de-Paillerets B., Chompret A., Bignon Y.J., Uhrhammer N., Lasset C., et al. Breast cancer risk in BRCA1 and BRCA2 mutation carriers and polyglutamine repeat length in the AIB1 gene. Int. J. Cancer (2005) 117:230–233.[CrossRef][ISI][Medline]

  44. Wilkening S., Burwinkel B., Grzybowska E., Klaes R., Pamula J., Pekala W., Zientek H., Hemminki K., Forsti A. Polyglutamine repeat length in the NCOA3 does not affect risk in familial breast cancer. Cancer Epidemiol. Biomarkers Prev. (2005) 14:291–292.[Free Full Text]

  45. Montgomery K.G., Chang J.H., Gertig D.M., Dite G.S., McCredie M.R., Giles G.G., Southey M.C., Hopper J.L., Campbell I.G. The AIB1 glutamine repeat polymorphism is not associated with risk of breast cancer before age 40 years in Australian women. Breast Cancer Res. (2005) 7:R353–R356.[CrossRef][ISI][Medline]

  46. Burwinkel B., Wirtenberger M., Klaes R., Schmutzler R.K., Grzybowska E., Forsti A., Frank B., Bermejo J.L., Bugert P., Wappenschmidt B., et al. Association of NCOA3 polymorphisms with breast cancer risk. Clin. Cancer Res. (2005) 11:2169–2174.[Abstract/Free Full Text]

  47. Kadouri L., Kote-Jarai Z., Easton D.F., Hubert A., Hamoudi R., Glaser B., Abeliovich D., Peretz T., Eeles R.A. Polyglutamine repeat length in the AIB1 gene modifies breast cancer susceptibility in BRCA1 carriers. Int. J. Cancer (2004) 108:399–403.[CrossRef][ISI][Medline]

  48. Rebbeck T.R., Wang Y., Kantoff P.W., Krithivas K., Neuhausen S.L., Godwin A.K., Daly M.B., Narod S.A., Brunet J.S., Vesprini D., et al. Modification of BRCA1- and BRCA2-associated breast cancer risk by AIB1 genotype and reproductive history. Cancer Res. (2001) 61:5420–5424.[Abstract/Free Full Text]

  49. Haiman C.A., Hankinson S.E., Spiegelman D., Colditz G.A., Willett W.C., Speizer F.E., Brown M., Hunter D.J. Polymorphic repeat in AIB1 does not alter breast cancer risk. Breast Cancer Res. (2000) 2:378–385.[CrossRef][ISI][Medline]

  50. Haiman C.A., Hankinson S.E., De Vivo I., Guillemette C., Ishibe N., Hunter D.J., Byrne C. Polymorphisms in steroid hormone pathway genes and mammographic density. Breast Cancer Res. Treat. (2003) 77:27–36.[CrossRef][ISI][Medline]

  51. Jernstrom H., Chu W., Vesprini D., Tao Y., Majeed N., Deal C., Pollak M., Narod S.A. Genetic factors related to racial variation in plasma levels of insulin-like growth factor-1: implications for premenopausal breast cancer risk. Mol. Genet. Metab. (2001) 72:144–154.[CrossRef][ISI][Medline]

  52. Hsing A.W., Chokkalingam A.P., Gao Y.T., Wu G., Wang X., Deng J., Cheng J., Sesterhenn I.A., Mostofi F.K., Chiang T., et al. Polymorphic CAG/CAA repeat length in the AIB1/SRC-3 gene and prostate cancer risk: a population-based case–control study. Cancer Epidemiol. Biomarkers Prev. (2002) 11:337–341.[Abstract/Free Full Text]

  53. Platz E.A., Giovannucci E., Brown M., Cieluch C., Shepard T.F., Stampfer M.J., Kantoff P.W. Amplified in breast cancer-1 glutamine repeat and prostate cancer risk. Prostate J. (2000) 2:27–32.[CrossRef]

  54. Campbell I.G., Choong D., Chenevix-Trench G. No germline mutations in the histone acetyltransferase gene EP300 in BRCA1 and BRCA2 negative families with breast cancer and gastric, pancreatic, or colorectal cancer. Breast Cancer Res. (2004) 6:R366–R371.[CrossRef][ISI][Medline]

  55. Park J.M., Lee G.Y., Choi J.E., Kang H.G., Jang J.S., Cha S.I., Lee E.B., Kim S.G., Kim C.H., Lee W.K., et al. No association between polymorphisms in the histone deacetylase genes and the risk of lung cancer. Cancer Epidemiol. Biomarkers Prev. (2005) 14:1841–1843.[Free Full Text]

  56. Yoon K.A., Hwangbo B., Kim I.J., Park S., Kim H.S., Kee H.J., Lee J.E., Jang Y.K., Park J.G., Lee J.S. Novel polymorphisms in the SUV39H2 histone methyltransferase and the risk of lung cancer. Carcinogenesis (2006) 27:2217–2222.[Abstract/Free Full Text]

  57. Frank B., Hemminki K., Wappenschmidt B., Klaes R., Meindl A., Schmutzler R.K., Bugert P., Untch M., Bartram C.R., Burwinkel B. Variable number of tandem repeats polymorphism in the SMYD3 promoter region and the risk of familial breast cancer. Int. J. Cancer (2006) 118:2917–2918.[CrossRef]