Document 640GBJ4M1j0K6mk2mBLzrk96

Fu r t h e rANNUAL REVIEWS Click here for quick links to Annual Reviews content online, including: Other articles in this volume Top cited articles Top downloaded articles Our comprehensive search Epigenetic Changes in Cancer Christine A. Iacobuzio-Donahue Departments of Pathology and Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland 21231; email: ciacobu@jhmi.edu Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Annu. Rev. Pathol. Mech. Dis. 2009. 4:22949 First published online as a Review in Advance on October 7, 2008 The Annual Review of Pathology: Mechanisms of Disease is online at pathmechdis.annualreviews.org This article's doi: 10.1146/annurev.pathol.3.121806.151442 Copyright c 2009 by Annual Reviews. All rights reserved 1553-4006/09/0228-0229$20.00 Key Words methylation, chromatin, histone modification, imprinting, progenitor cell, genetics Abstract Cancer is as much an epigenetic disease as it is a genetic disease, and epigenetic alterations in cancer often serve as potent surrogates for genetic mutations. Normal epigenetic modifications of DNA encompass three types of changes: chromatin modifications, DNA methylation, and genomic imprinting, each of which is altered in cancer cells. This review addresses the various epigenetic modifications that are pervasive among human tumors and traces the history of cancer epigenetics from the first observations of altered global methylation content to the recently proposed epigenetic progenitor model, which provides a common unifying mechanism for cancer development. 229 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. A BRIEF HISTORY OF CANCER GENETICS Human cancer is, in essence, a genetic disease. The first suggestions of a genetic basis for cancer are credited to David Hansemann (1858 1920) and Theodor Boveri (18621915), both of whom observed abnormal numbers of chromosomes arising by multipolar mitoses and suggested that this abnormality is the cause of tumor formation (1, 2). However, evidence of a specific gene or genes associated with cancer was not recognized until the discovery of the Rous sarcoma virus in 1911, followed by the discovery of the virus's cellular homolog src, a tyrosine kinase with oncogenic activity, more than 60 years later (3). In 1987, the retinoblastoma gene was identified on 13q14; it was the first tumor-suppressor gene to be discovered (4). Since then, numerous genes that are abnormally activated or disrupted by mutation in carcinogenesis have been identified. These genes affect diverse pathways regulating cell growth, apoptosis, cell signaling, and DNA repair (5). The classical view of cancer is that it arises from a single cell that, through a series of both dominantly and recessively activating mutations, undergoes progressive waves of clonal expansion, which lead to a neoplasm that is clonally heterogeneous. Each mutation leads to the selective overgrowth of a population of tumor cells, and each population of cells contains one or more significant properties related to invasiveness, metastatic ability, and therapeutic resistance. Accumulation of genetic changes during tumor progression has been well documented for many tumors. Some mutations, such as in TP53, are found in almost all tumor types. Others, however, are specific to tumors from a specific tissue type, for example c-KIT mutations in gastrointestinal stromal tumors. In most cases, the genetic progression model has been successful in predicting so-called gatekeeper mutations, that is, mutations that seem to be necessary for the earliest stages of tumor growth, as in APC mutations in colorectal cancer (5). Thus, classical genetics has provided a firm foundation for our understanding of carcinogenesis. However, many properties of cancer cells that cannot be accounted for by genetic mutations include heterogeneity in tumor cell growth, invasion, metastasis, and resistance to therapy. As I discuss in this review, one factor increasingly recognized as contributing to all facets of cancer behavior is epigenetics. THE EMERGENCE OF CANCER EPIGENETICS The structure of DNA encodes all the information necessary to establish an organism. However, beyond this essential code, epigenetic phenomena are required for proper development and cellular differentiation within normal tissues. Epigenetics is defined as modifications of DNA or associated factors that have information content and are hereditable (aside from the DNA sequence itself) (6). Normal epigenetic modifications of DNA encompass three types of changes, all of which are interrelated: chromatin modifications, DNA methylation, and genomic imprinting. The first to use the term chromatin to describe nuclear proteins was Flemming, who in 1882 wrote: ". . . The word chromatin may stand until its chemical nature is known, and meanwhile stands for that substance in the cell nucleus which is readily stained" (reviewed in Reference 7). More than 80 years later, one of the earliest observations regarding epigenetic control was made by Allfrey et al. (8), who noticed a correlation between histone acetylation (a form of chromatin modification) and transcriptional activity. Since then, the study of epigenetic changes in normal and diseased tissues has become a major focus of medical research. The first molecular epigenetic abnormality in cancer was identified in 1983 by Feinberg & Vogelstein (9), who observed a global reduction of methylation content in colon cancer cells compared to levels in normal colonic tissues. This finding was soon independently verified by Gama-Sosa et al. (10), who also demonstrated 230 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. ab Figure 1 Morphologic appearance of chromatin in human tissues. (a) In this high-power view of normal colonic epithelium, the nuclei are evenly spaced and uniform in size and shape (i.e., monomorphic). The nuclear membrane has a smooth contour, and the chromatin within is evenly dispersed. (b) In colon cancer, the nuclei are enlarged and irregular (i.e., pleomorphic), and the nuclear content is irregularly distributed so that regions of dark-staining chromatin alternate with regions of pale-staining chromatin within an individual nucleus. global reduction of 5-methylcytosine content in every tumor type studied. Importantly, these changes were found in both preinvasive and invasive cancer tissues. Since these seminal observations were made 25 years ago, a virtual explosion in our understanding of epigenetic changes in cancer has transformed our view of carcinogenesis. Pathological epigenetic changes are now considered potent alternatives to mutations and chromosomal alterations in neoplasia, particularly as many epigenetic changes appear prior to the development of invasive cancer (6, 11, 12). Thus far, most known epigenetic modifications of DNA or associated factors in normal tissues are altered in cancer. Despite these advances, the field of pathology has long recognized that chromatin patterns in cells, which are morphologic markers of the epigenome, are important features of normal and disease processes. In fact, the appearance and quality of chromatin are often used for pathologic interpretation of normal or diseased cells in human specimens. For example, in a normal nucleus the chromatin pattern observed is of evenly dispersed euchromatin, the actively transcribed chromatin, and hete- rochromatin, which is condensed, inactive, and often associated with the nuclear membrane. By contrast, neoplastic cells are commonly recognized by their abnormally heterogeneous chromatin distribution within an irregular nuclear membrane (Figure 1). Chromatin Structure and Modifications DNA that is packaged into chromatin forms the basis for all processes that affect genetic activity. Nucleosomes, the smallest units of chromatin, contain 146 base pairs of DNA wrapped around a core of eight histone proteins. This octamer comprises two copies each of the H2A, H2B, H3, and H4 proteins, which are highly conserved throughout evolution. All four histones have an amino-terminal tail that is lysine rich and that contains about half of the positively charged residues of the polypeptide backbone. These lysine-rich tails protrude out of the nucleosome and are subject to posttranslational modifications. Each nucleosome is separated by approximately 50 base pairs of DNA, which is packaged by the linker histone protein H1 (13) (Figure 2). www.annualreviews.org Epigenetic Changes in Cancer 231 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. a H2A H3 H2B H4 A K K HDAC HAT AA A K K K NH2 M HMT HDAC HMT HDAC b HAT DNMT DNMT A A K K HMT A M K K M MBP M K CpG MBP M MBP CpG M Figure 2 Nucleosome organization and modifications of chromatin that underlie gene silencing in normal and malignant cells. (a) Nucleosomes comprise eight histone proteins, two each of H2A, H2B, H3, and H4. The amino-terminal tails of histones protrude from this protein core, and the lysine (K) residues are subject to acetylation by histone acetyltransferase (HAT) or methylation by histone methyltransferase (HMT), although other modifications are also possible. (b) Gene expression is associated with lysine acetylation that is mediated by HAT. Histone deacetylases (HDACs) deacetylate lysine residues as a prerequisite for their subsequent methylation by HMT. DNA can also be methylated at CpG dinucleotides. This process is mediated by DNA methyltransferases (DNMTs) that participate in multiprotein complexes, which contain HDACs and HMTs. Methyl-CpG-binding domain proteins (MBPs) can be loaded onto methylated DNA through their interactions with both HDACs and HMTs. Histone amino-terminal tails sustain an overall basic charge and only marginally contribute to nucleosome stability. Thus, posttranslational modification of the aminoterminal tails is critical to maintaining chromatin structure in addition to controlling gene expression without changes in the DNA sequence. These posttranslational modifications encompass the largest variation in epigenetic control, with more than 50 known sites of modification that include acetylation, methylation, citrullination, phosphorylation, SUMOylation, and ADP-ribosylation (14, 15). The major forms of histone modification are acetylation/deacetylation, methylation, incorporation of histone variants, and ATPdependent chromatin remodeling. These modifications are crucial in modulating how tightly or loosely the chromatin is compacted and therefore regulate gene expression (15). The incorporation of variant histone proteins provides another level of regulation to the capacity of chromatin-remodeling mechanisms to store cellular information. For example, the H3-like variant CENPA (centromeric protein A) replaces H3 in centromeric nucleosomes to maintain a unique structure that is essential for chromosome segregation during mitosis (16). It has been proposed that a regulatory "code" exists in the patterns of posttranslational modifications by phosphorylation, acetylation, methylation, and/or ADP-ribosylation, of which the histone amino-terminal tails are the targets (17, 18). If so, this code is read by nonhistone proteins and protein complexes that constitute the transcriptional molecular machinery. These modifications have been linked to many normal biological processes that depend upon the accessibility of chromatin, including gene expression, DNA repair, chromosome segregation during mitosis, X chromosome inactivation, and apoptosis (reviewed in Reference 19). Histone modifications in tumorigenesis are increasingly recognized as important epigenetic features of cancer, with histone lysine acetylation and methylation being the bestcharacterized changes in chromatin remodeling to date (18, 20). In particular, global reductions in monoacetylated H4-K16 and trimethylated H4-K20 have been demonstrated to be general features of cancer cells that occur in the early stages of carcinogenesis and accumulate with tumor progression (21). Histone Acetylation Acetylation of histone amino-terminal tails neutralizes the positive charge on the basic 232 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. lysine residues, thereby weakening electrostatic interactions between histones and the phosphate backbone of DNA and promoting gene expression (22). Acetylation of the lysine residues on histones H3 and H4 are correlated with active or open chromatin, which allows various transcription factors access to promoters of target genes. Histone H1 is also important in determining the level of DNA condensation, but it is not regulated by acetylation. Two classes of enzymes can affect the acetylation of histone proteins: histone acetyltransferases (HATs) and histone deacetylases (HDACs) (23, 24). Acetylation of lysine residues in the N-terminal tails of histones by HATs is a key process associated with active gene transcription (20). Members of the HAT family can be categorized based on their numbers of highly conserved structural motifs such as the GNAT family (Gcn5-related Nacetyl transferase), the MYST family (named after its members MOZ, YBF2/SAS3, SAS2, and TIP60), and the p300/CBP family. Members of the last group differ from other HATs in that they do not bind directly to DNA but are recruited to promoters by means of DNA-bound transcription factors (23). Acetylation of histone lysine tails is not random; HATs preferentially acetylate specific histone tails (17, 23). Interestingly, HAT enzymes also target nonhistone proteins such as the transcription factors TP53, E2F1, and GATA1 (25 27). By contrast, deacetylation by HDACs removes charge-neutralizing acetyl groups from histone lysine tails, leading to condensation of chromatin and to gene inactivation (28, 29). HDACs can be divided into three classes based on their homology to the yeast HDAC Rpd3 (Class I), the yeast HDAC Hda1 (Class II), and the yeast and murine protein Sir2 (Class III) (30). Unlike Class I and II HDACs, Class III HDACs depend on NAD+ for activity and are insensitive to the HDAC inhibitor trichostatin A (31, 32). Like HATs, HDACs target nonhistone proteins such as the transcription factors p53, E2F1, GATA1, TFIIE, TFIIF, and the glucocorticoid receptors (28, 29, 3335). Genetic alterations of proteins in the HAT/HDAC family provide a compelling link between aberrant histone lysine acetylation and cancer. In fact, specific disruption of HATor HDAC-related proteins is associated with carcinogenesis (36, 37). Individuals with the developmental disorder RubinsteinTaybi syndrome are characterized by a >300-fold increased cancer risk in association with germline mutation in CBP (37). The other allele is somatically mutated in these malignancies, leading to a loss of HAT activity. Somatic missense and nonsense mutations have also been found in p300 in gastric and colorectal cancers (38, 39). In most cases, the second allele was lost, resulting in biallelic inactivation of p300 as well. HATs and their related proteins-- CBP, p300, MOZ, and MORF--are also occasionally incorporated into fusion proteins that arise from chromosomal translocations associated with leukemia (40, 41). In these instances, the fusion proteins represent gain-of-function mutants that lead to mistargeting of HATs, resulting in perturbations in histone acetylation and gene activation. Truncating mutations in HDAC2 have also been described in colorectal cancers with microsatellite instability, where they lead to resistance to HDAC inhibitors (36). A more general mechanism by which HDACs promote carcinogenesis is their targeting to specific chromosomal regions by other transcription factors. Many HDACs exist as components in multiprotein complexes, which act as transcriptional corepressors. The best characterized of these interactions is the recruitment of HDACs to the hypermethylated CpG islands of tumor-suppressor genes via methyl-binding proteins (28). A potentially more important role for HDACs in cancer that occurs independently of DNA methylation involves the recruitment of HDACs to tumorsuppressor genes via oncogenic DNA-binding factors. For example, in contrast to wild-type AML1 protein, the translocation-generated fusion protein AMLETO actively suppresses transcription via the aberrant recruitment of HDAC-containing corepressors (42). A similar www.annualreviews.org Epigenetic Changes in Cancer 233 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. mechanism underlies the action of the fusion protein PMLRARa (43). Histone Methylation Unlike histone acetylation, which is transient in nature, histone methylation by histone methyltransferases (HMTs) is widely regarded as a mark of long-standing cellular memory (44). Histone tail methylation can result in either activation or repression of gene expression, depending on the residue methylated. Thus, the pattern of histone tail modification can be interpreted as a marker of either transcriptionally active or inactive chromatin. Most known sites of histone lysine methylation occur on H3 and H4. For example, methylation of histone H3lysine 4 (H3-K4), H3-K36, or H3-K79 is a marker of gene expression and active chromatin, and methylation of H3-K9, H3-K27, or H4-K20 corresponds to gene silencing and inactive chromatin (4547). Histone methylation affects gene expression in part through its association with DNA methylation, as many proteins involved in DNA methylation (such as DNA methyltransferases and methyl-binding proteins) also directly interact with histonemethylating enzymes. These processes indicate a convergence of DNA and histone methylation pathways, which may cooperate to silence tumor-suppressor genes in cancer cells (21). Histone methylation also differs from histone acetylation in that more than one methyl group per lysine residue can be present. Trimethylated H3-K9 is enriched in pericentromeric chromatin compared to mono- or dimethylated H3K9, and trimethylated H3-K27 is enriched at the inactive X chromosome, suggesting that histone modifications mark specific domains of heterochromatin in addition to regions of active or inactive chromatin (48). At least 17 HMTs are known, all of which are characterized by their evolutionary conserved SET [Su(var)3 9, Enhancer-of-zeste, Trithorax] domain. Like HATs, HMTs methylate proteins other than histones (49). The functional consequences of histone methylation are best understood in relation to the proteins that recognize these modifications. For example, repressive proteins such as heterochromatin protein 1 and the Drosophila Polycomb protein contain a chromodomain that allows them to recognize the appropriate repressive methylation matrix attachment regions H3-K9 and H3-K27, respectively. By contrast, the chromodomain helicase DNAbinding protein 1 activator protein from Saccharomyces cerevisiae uses its chromodomain to bind the activating methylated H3-K4 mark (50). Additional protein domains that bind methylated lysine residues include (a) the Tudor domain within the DNA-repair checkpoint protein p53-binding protein 1, which recognizes methylated H3-K79, and (b) the WD40 repeats of the vertebrate transcriptional activator WDR5, which bind to di- and trimethylated H3-K4 (51, 52). Until recently, methylation of histones was believed to be irreversible, partly because no histone demethylases were known. However, in 2004 investigators identified the enzyme LSD1 (lysine-specific demethylase 1, also known as BHC110 and p110b), which demethylates K4 within histone H3 (53). Demethylation by LSD1 is limited to mono- or dimethylated H3K4 but not trimethylated H3-K4, which is associated with active gene expression. Because gene transcription is a highly dynamic process, it is likely that additional histone demethylases are yet to be identified. Overexpression of key histone methyltransferases that catalyze the methylation of either H3-K4 or H3-K27 residues is a frequent event in neoplasia (6). As with HATs and HDACs, researchers have identified genetic alterations of HMTs or related proteins that further implicate histone methylation in carcinogenesis. For example, germline deletions or point mutations in NSD1, a histone methylase, give rise to Sotos syndrome, which is associated with a 170-fold-increased risk of developing cancer (54). A wide variety of neoplasms have been described in Sotos syndrome, including Wilms' tumor, neuroblastoma, acute lymphocytic or lymphoblastic leukemia, hepatocellular carcinoma, and small cell carcinoma of the lung. 234 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. NSD1 is also targeted somatically, as the recurrent translocation t(5:11)(q35;p15.5) in childhood acute myeloid leukemia often includes NSD1 (55). The mixed lineage leukemia (MLL) gene, which contains a SET domain, is also a common target for chromosomal translocations associated with human acute leukemias. The SET domain is typically absent from these oncogenic fusion proteins; thus, HMT activity does not seem to play a part in what appear to be gain-of-function mutations. Nevertheless, these MLL mutants support the notion that unregulated MLL function contributes to leukemogenesis (56). Additional HMTs implicated in cancer include EZH2, RIZ1, and SUV39H1. EZH2, a H3-K27 methyltransferase, is a member of the polycomb group complexes PRC2 and PRC3. These complexes also include SUZ12, a component required for the complexes' full activity (57, 58). The H3-K27 mark provides the docking signal for binding of the repressive polycomb complex PRC1 via the chromodomain of the polycomb. EZH2 is required for cell proliferation and acts in association with the retinoblastoma (Rb) family of proteins to inactivate the tumor suppressor p16/INK4 (59). In keeping with this role, EZH2 and its related protein SUZ12 are highly expressed, amplified, or genetically rearranged in a variety of tumor types, with the level of expression often correlating with the degree of aggressiveness (6064). Note, however, that in the absence of transgenic mouse models with increased EZH2 expression, it is not clear whether overexpression of EZH2 is a cause or consequence of malignancy. RIZ1 (PRDM2) is another HMT implicated in cancer. RIZ1 is a H3-K9 methyltransferase originally identified through its interaction with pRb (65). Biallelic inactivation of the RIZ1 gene through mutation or promoter methylation has been identified in a variety of tumor types (66, 67). Furthermore, geneknockout mice that lack RIZ1 are prone to developing B cell lymphomas (68). Of particular interest is the finding that cancer-specific mutations in RIZ1 reduce or abolish the histone methylase activity of the protein, which suggests an important role for this activity in tumor suppression (69). Finally, SUV39H1 is a H3-K9 methyltransferase that is associated predominantly with pericentromeric heterochromatin. This HMT is also associated with cancer development, possibly through its role in maintaining genomic stability or cell-cycle progression (70, 71). Global DNA and Promoter DNA Methylation In mammalian cells, methylation occurs by a covalent modification of DNA in which a methyl group is transferred from Sadenosylmethionine to the C-5 position of cytosine by a family of cytosine (DNA-5)methyltransferases (DNMTs). Only cytosine bases that are located 5 to a guanosine (CpG dinucleotides) are methylated. CpG dinucleotides are underrepresented in the genome, but there exist short regions rich in CpG content, known as CpG islands. Most CpG islands are found in repetitive elements including centromeres, microsatellite sequences, and proximal promoter regions of approximately half the genes in the genome of normal cells, where the islands are generally unmethylated (72, 73). Progressive DNA methylation is thought to contribute to normal aging and to various pathologic states, including cancer. Methylation at CpG regions affects transcription by recruiting methyl-CpG-binding domain proteins (MBPs) that function as adaptors between methylated DNA and chromatinmodifying proteins (7476). DNA methylation is regulated by a family of DNMTs that includes DNMT1, DNMT3A, and DNMT3B. All three DNMTs are embryonic lethal in knockout mice models. However, DNMT3A and 3B are important for patterning of DNA methylation during embryogenesis, after which DNMT1 acts as a maintenance methyltransferase (77, 78). However, the separable roles of the DNMT enzymes have been challenged in cancer models: Some investigators have shown that severe depletion www.annualreviews.org Epigenetic Changes in Cancer 235 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. of DNMT1 produces (a) negligible decreases in overall DNA methylation and promoter methylation and (b) undetectable changes in expression of silenced tumor-suppressor genes (7982). DNA methylation and histone modifications are interrelated phenomena that together determine gene expression (83). DNMTs recruit HDACs, leading to histone deacetylation and transcriptional repression. Methylated DNA is also recognized by a family of MBPs, which also recruit HDACs and ATP-dependent chromatin-remodeling proteins, resulting in chromatin condensation and gene inactivation. CpG island promoter methylation can additionally block transcription by interfering with binding of transcription factors to their target binding sites (84). The primary role of global hypermethylation in normal cells is hypothesized to maintain the integrity of the genome through the heritable repression of repetitive element transcription; many repetitive elements are retroviruses that have been trapped in the genome by DNA methylationdependent silencing or by Alus, which are short, inverted repeats (85, 86). Thus, global hypomethylation observed in cancer cells may cause expression of normally silent genes by two possible mechanisms. First, gene promoters that are normally methylated after embryogenesis and organismal development may become reactivated, resulting in aberrant expression. The cancer/testis antigen CAGE is an example of a gene reactivated by hypomethylation in cancer (87, 88). Other genes that mediate overexpression under hypomethylation include NAT1 in breast cancer (89) and CD30 in anaplastic large cell lymphoma (90); both of these genes have been suggested to have a causal relationship with cancer formation. Second, global demethylation may result in an overall increase in genomic instability, leading to cancer formation. For example, hypomethylation has been shown to increase the frequency of spontaneous tumor formation in mouse models (9193), and decreasing overall methyl-cytosine content has been correlated with greater degrees of aneuploidy (94, 95). A causal relationship between genomic instability and hypomethylation is supported by the developmental disorder ICF syndrome (immunodeficiency, chromosome instability, and facial abnormalities), which is caused by loss-of-function mutations in the DNA methyltransferase DNMT3B (96). Strong experimental evidence of a relationship between global hypomethylation and genomic instability also comes from experiments using a mouse model of neurofibromatosis type I, wherein the mice showed a more than twofold increase in loss of heterozygosity events when a hypomorphic Dnmt1 allele was introduced (92). Moreover, Lengauer et al. (97) showed that colon cancer cells with microsatellite instability retained the ability to methylate ectopic DNA sequences, whereas cells characterized by large-scale chromosome aberrations and aneuploidy did not, providing additional support for this relationship. However, global hypomethylation is not the only epigenetic abnormality that may contribute to genomic instability. For example, CENPA, the H3-like histone commonly found in centromeric nucleosomes, becomes overproduced in colorectal cancer and thus may contribute to aneuploidy (98). How might global hypomethylation lead to tumor formation? Hypomethylation in cancer cells is particularly prominent in pericentromeric regions, such as those found on chromosomes 1 and 16 (94, 96, 99); thus, hypomethylation of these regions may predispose them to recombination events. Recurrent unbalanced chromosomal translocations with breakpoints in these pericentromeric DNA sequences have been reported in breast cancers, ovarian cancers, and Wilms' tumors (94, 99, 100). Alternatively, hypomethylation of latent viral sequences may allow their reexpression and consequent tumor progression. For example, cervical cancer latency has been linked to hypermethylation of the HPV16 genome, and activation of the HPV genome has been linked to progressive hypomethylation that results in cervical dysplasia (101). Epstein-Barr virus (EBV) latency follows a similar pattern in EBV-associated lymphomas (102). 236 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Although global hypomethylation is a common feature of cancer and may be a causative factor in cancer development, promoter hypermethylation leading to transcriptional silencing is also a well-characterized epigenetic change in human tumors (77, 103). Aberrant promoter methylation has been associated with loss of gene function, which can provide a selective advantage to neoplastic cells that is similar to that seen for classical mutations in the clonal progression model. The first evidence of promoter hypermethylation and loss of gene expression was found by Baylin et al. (104) in their studies of calcitonin. However, the first evidence of tumor-suppressor gene hypermethylation was found for the RB gene (105, 106). Based on this observation and on the finding of allele-specific hypermethylation of RB, which indicates the specificity of this phenomenon, Sakai et al. (106) suggested the role of promoter hypermethylation as a more general mechanism of tumorsuppressor gene silencing. Shortly thereafter, Ohtani-Fujita et al. (107, 108) provided evidence for the first direct relationship of RB gene expression silencing in tumors with promoter hypermethylation of the RB promoter. Following reports of this relationship, studies of tumor-suppressor genes implicated in sporadic and familial cancer revealed that they were also targeted by methylation. Promoter hypermethylationmediated gene silencing is now a well-known feature of many tumor-suppressor genes, including the cyclindependent kinase inhibitor CDKN2A (p16), the mismatch repair enzyme MLH1, the von HippelLindau tumor-suppressor gene VHL, CDH1, and BRCA1 (109). In these examples, the extensive mutational data of each gene support the epigenetic inactivation of the gene as biologically relevant to carcinogenesis. In fact, almost 50% of genes that cause familial forms of cancer by mutation or deletion, such as BRCA1 or CDH1, are now recognized as undergoing promoter hypermethylation in the sporadic forms of those same tumors (109). However, whereas global hypomethylation has been linked to cancer formation, promoter methylation has not. More than 100 genes with tumor-specific promoter hypermethylation are currently recognized (109). Many genes with promoter hypermethylation do not otherwise undergo genetic mutation in human cancer; thus their tumor-suppressor function is based on their known function in normal cells. For example, the gene RASSF1a (Ras-association domain, family 1, isoform A), which is commonly methylated and silenced in several types of cancer, was identified in a region of frequent allelic loss on chromosome 3p (110). This gene is an effector of the Ras oncoproteins and modulates multiple apoptotic and cell-cycle checkpoint pathways (111). The functional significance of gene silencing by hypermethylation can be appreciated in the context of the two-hit hypothesis of tumor-suppressor gene inactivation. Several studies (109) have now shown that tumors can stably maintain mutations in one allele of a gene while the other allele is hypermethylated, thereby leading to functional inactivation. Although promoter hypermethylation is considered a surrogate manner of tumorsuppressor gene inactivation, additional classes of genes may be inactivated by promoter hypermethylation such as transcription factors, cell-cycle regulation, signal transduction pathways, and tissue remodeling (77, 103). For example, one of the most comprehensive ways to alter gene expression is to silence the expression of a transcription factor, and consequently, its downstream targets. Examples of transcription factors that are hypermethylated in cancer are OCT3/4 in testicular cancer (112), NFATC1 in lymphoma (113), and RUNX3 in esophageal cancer (114). In some instances, epigenetic silencing of a transcription factor and its downstream targets in a single tumor may be found, as has been shown for the GATA-4 and GATA-5 transcription factors and their downstream target genes in colorectal and gastric cancers (115). GATA-4 and GATA-5 are transcription factors that play a role in gastrointestinal development and are epigenetically silenced, together with their target genes Dab2, inhibin-, and TFF1 and 2. www.annualreviews.org Epigenetic Changes in Cancer 237 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Other sequelae of promoter hypermethylation in cancer may be to promote gene inactivation by classical mutational events. For example, hypermethylation of the DNA mismatch repair gene MLH1 is a frequent finding in sporadic colorectal tumors characterized by microsatellite instability. MLH1 hypermethylation may also be found in the normal mucosa of the same patients, consistent with hypermethylation preceding the development of microsatellite instability and colon cancer (116, 117). MGMT is another DNA-repair gene that undergoes promoter hypermethylation and epigenetic silencing in cancer (118). O6-MGMT normally acts to remove O6methylguanine adducts from DNA that arise from carcinogen exposure. Left unrepaired, these adducts result in GA transition mutations in key genes. Genomic Imprinting The reproductive process in mammalian species allows inheritance of two copies of every gene, one from the mother and one from the father. For most genes, both alleles are actively transcribed and functionally equivalent. Imprinted genes are an exception to this rule; they are caused by an epigenetic form of gene regulation that results in only one transcriptionally active allele that is determined by the parent of origin (119). To explain the evolution of genomic imprinting, Haig & Westoby (120) put forth the parent-offspring conflict hypothesis in 1989. Based on the tenets that a female is equally related to all her offspring and that any male that produces offspring with her is not necessarily related to her other offspring, the authors hypothesized that the paternal genome has evolved the ability to enhance maternal resources for its own offspring by paternal allele specific expression of growth-promoting genes and that the maternal genome counters this ability with maternal allelespecific expression of growth-inhibitory genes. Consistent with this hypothesis, the majority of imprinted genes identified to date can significantly affect fetal size in mouse models (reviewed in Reference 119). Imprinted genes tend to be clustered within the genome, and genes within an imprinting cluster share many regulatory elements and often have similar developmental and tissuespecific patterns of expression. Clusters of imprinted genes are regulated coordinately by an element known as an imprinting center (IC). The significance of ICs in regulating imprinted genes comes from studies in which (a) the IC was deleted or disrupted with resultant loss of imprinting or (b) naturally occurring mutations in ICs led to imprinting disorders such as PraderWilli syndrome (reviewed in Reference 121). One of the best understood ICs is on chromosome 11p15; it was recognized as a differentially methylated region upstream of H19. Deletion of this region in mice causes loss of imprinting at both H19 and Igf2 (insulin-like growth factor 2), and naturally occurring mutations in this region are present in individuals with BeckwithWiedemann syndrome (BWS) (122). In addition to differential DNA methylation, ICs also show allelic differences in chromatin structure, such as to DNAse I hypersensitivity and covalent modifications of histone tails. Allele-specific DNAse Ihypersensitive sites have been found at the Igf2/H19, Air, and SNURF ICs, among others. Repressive histone modifications such as methylation at H3-K9 or H3-K27 have been found at the ICs of the methylated allele, whereas activating histone modifications such as H3 and H4 acetylation have been found on the unmethylated allele (6). ICs can also act as chromatin insulators. An insulator is a sequence of DNA that blocks enhancers from interacting with gene promoters when positioned between the two. An insulator also acts as a barrier to the spread of transcriptionally repressive condensed chromatin. The CCCTC binding factor (CTCF) is known to bind to insulators and mediate their enhancer-blocking activity. A CTCFdependent insulator exists within the Igf2/H19 IC. On the unmethylated maternal allele, the insulator protein CTCF binds to target sites in 238 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. the IC, thus preventing the enhancers downstream of H19 from accessing the Igf2 promoters. In contrast, on the paternal allele, DNA methylation blocks CTCF binding, thereby abolishing insulator activity and allowing the enhancers to access Igf2 promoters, which results in Igf2 transcription. Putative binding sites for CTCF have also been found in the Gtl2 and Grb10 loci. The significance of CTCF in cancer was shown in Wilms' tumors, in which loss of imprinting (LOI) depends upon hypermethylation of CTCF sites located upstream of H19 (123). In colorectal cancers, both hypomethylation of Igf2 (which does not involve CTCF sites) and hypermethylation of CTCF have been described (124, 125). More recently, amplification of BORIS was identified in breast cancers. This gene is a paralog of CTCF and is believed to interfere with CTCF binding (126). A similar phenomenon, which has been described in advanced-stage prostate cancers, is mediated by overexpression of the gene EZH2, which is linked to generalized hypermethylation (60). The earliest indication that genomic imprinting is involved in cancer came from cytogenetic studies of two human neoplasms: hydatidiform moles and ovarian teratomas. A hydatidiform mole is a malignant trophoblastic tumor caused by the abnormal fusion of two complete sets of the paternal genome (127). Ovarian teratomas are benign tumors with many tissue types that arise from two complete sets of the maternal genome (128) (Figure 3). These tumors indicate that an imbalance between maternal and paternal genomes causes neoplasia, and an excess in the parental genome determines what type of tumor forms. Molecular support for a relationship between genomic imprinting and cancer came from studies showing a parent-of-origin bias in loss of heterozygosity for chromosome 11p15 alleles in Wilms' tumors and in embryonal rhabdomyosarcoma, with invariable loss of maternal alleles and duplication of paternal alleles (129). These observations suggest that one of the two hits affecting the responsible gene may represent epigenetic silencing of a specific parental al- lele by genomic imprinting. Studies of the pediatric disorder BWS have provided additional evidence of genomic imprinting. BWS is characterized by a variety of birth defects, prenatal overgrowth, and a predisposition to childhood tumors including Wilms' tumor, which maps to chromosome 11p15 (Figure 4). Perhaps the most convincing evidence has been provided by Mannens et al. (130), who showed the chromosome rearrangements at 11p15 in BWS were all of maternal origin. LOI refers to activation of the normally silenced allele, or silencing of the normally active allele, of an imprinted gene. LOI of IGF2 accounts for half of all Wilms' tumors in children (131). LOI of IGF2 is also a common epigenetic variant in adults and is associated with a fivefold-increased frequency of colorectal neoplasia (132, 133). LOI of IGF2 may cause cancer by increasing the progenitor cell population in the kidney in Wilms' tumor (131) and in the gastrointestinal tract in colorectal cancer (134, 135). Examples of other genes that show LOI in cancer include PEG1/MEST (paternally expressed gene 1/mesoderm-specific transcript homolog) in lung cancer (136), p57 KIP2 (also known as CDKN1C) in pancreatic cancer (137), and TP73 in gastric cancer (138). The Epigenetic Progenitor Model of Cancer The relatively new idea that epigenetic modifications may play a role in cancer predisposition was put forth by Andrew Feinberg and colleagues (11). Within normal tissues, stem cells have been suggested to be the target of initiating events upon which genetic events may occur; this suggestions provides a unifying view of cancer etiology (11). The stem cell is increasingly recognized as the target of initiating events, and epigenetic alterations underlie stem cell identity (139, 140); thus, it is conceivable that the earliest event in carcinogenesis is an epigenetic disruption of progenitor cells upon which gatekeeper mutations occur. This disruption may indicate (a) an increase in the number of progenitor cells upon which genetic events www.annualreviews.org Epigenetic Changes in Cancer 239 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. ab ST CT cd ef Figure 3 Histopathology of neoplasms that arise in association with an imbalance of parental chromosomes. Panels a and b illustrate the gross and microscopic features of a hydatidiform mole, a trophoblastic tumor caused by a pregnancy arising from two complete sets of the paternal genome. Grossly, these neoplasms are characterized by an abnormal placenta consisting of a mass of tissue with grapelike, swollen chorionic villi. Although a hydatidiform mole is usually a benign process, a significant proportion of cases lead to either invasive mole or choriocarcinoma, a highly aggressive malignancy with a propensity for early metastatic spread (shown in panels c and d ). Grossly, a choriocarcinoma is a soft, hemorrhagic, and nodular mass. Microscopically, choriocarcinomas contain malignant proliferations of cytotrophoblast (CT) that are separated by masses of syncytiotrophoblast (ST). Panels e and f show features of ovarian teratoma, a form of benign tumor comprising a variety of tissue types. Ovarian teratomas arise from cells containing two complete sets of the maternal genome. Photos provided by Dr. Bridgette Ronnett, Department of Pathology, Johns Hopkins Institutions. 240 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. occur, (b) an increase in the cells' ability to become neoplastic, or (c) an increase in the cells' "stemness" (11) (Figure 5). Sakatani et al. (134) provided evidence in support of this model. In their study, a mouse model of Igf2 LOI was created by crossing female H19+/- mice with male Apc+/Min mice. Mice with LOI developed twice as many intestinal tumors as did control littermates, and the intestinal epithelium in LOI mice showed a shift toward a less-differentiated epithelium, which was observed as an increase in crypt length and in staining with progenitor cell markers. Importantly, a similar shift in differentiation was observed in the normal colonic mucosa of humans with LOI. The evidence supporting epigenetic mechanisms as initiating events to cancer is twofold. First, epigenetic mechanisms are generally accepted to be potent surrogates for oncogene activation or tumor-suppressor gene silencing in cancer initiation and progression. Hypermethylation and associated silencing of the SFRP genes may be important initiating events in colorectal cancer (141), and hypomethylation with associated overexpression of Bcl-2 has been described in B cell lymphocytic leukemias (142). Second, epigenetic events such as global hypomethylation may be related to genomic instability, a pervasive feature of human cancers and among the earliest changes to be observed therein (92, 143). Perhaps the best evidence in favor of epigenetic changes as initiating events comes from studies of normal tissues from patients who develop cancer. These normal tissues represent a principal target of environmental, genetic, and age-dependent exposure that largely accounts for the long latency of cancer. Consistent with this notion, age-dependent methylation alterations have been observed in the estrogen receptor gene, RASSF1a, and cyclin D2 (144146), but none of these changes has been linked to familial risk or to increased cancer frequency. However, in 2003 Cui et al. (132) showed that unlike LOI of IGF2 that is an epigenetic variant present in 10% of the normal population, it is found in normal colonic mucosa of about 30% of colorectal cancer patients. LOI of IGF2 is associated with personal a b c Figure 4 Histopathology of Wilms' tumor. (a) Wilms' tumors are typically solitary, well-circumscribed masses with a soft consistency. On cut section, the tumor is solid and shows areas of hemorrhage and cystic change. (b) Microscopically, Wilms' tumors comprise undifferentiated blastema, mesenchymal (stromal) tissue, and epithelial tissue. (c) Perilobar nephrogenic rest comprising blastema and primitive tubules. Photos provided by Dr. Pedram Argani, Department of Pathology, Johns Hopkins Institutions. www.annualreviews.org Epigenetic Changes in Cancer 241 a bcd Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Stem cell Differentiated enterocyte Paneth cell Change in gene expression Genetic mutation Epigenetic plasticity Figure 5 The epigenetic progenitor model of cancer. Normal colonic epithelium (a) includes a proliferative zone that contains stem cells (orange), which give rise to differentiated enterocytes further up the crypt (light pink) or to Paneth cells at the crypt bases (dark pink). The epigenetic progenitor model suggests that (b) the stem cell compartment is altered epigenetically, which can involve an expansion of the progenitor compartment or changes in gene expression (blue), followed by (c) genetic mutation (red ). Subsequent evolution of the tumor involves (d ) genetic and epigenetic plasticity (turquoise and green). Epigenetic plasticity allows expression of phenotypic features related to invasion, metastasis, and drug resistance, which are inherent properties of the stem cell progenitor (11). or family history of colorectal adenomas or cancer, but not with environmental, dietary, or occupational exposures, suggesting that it may be a hereditary risk factor for colorectal cancer and a valuable predictive marker of an individual's risk of developing this disease (147). The epigenetic model of cancer has important implications for understanding cancer biology from a pathologist's standpoint. For example, the proposed existence of epigenetically disrupted progenitors of cancer implies that the earliest stages in neoplastic progression occur even before what a pathologist would recognize as a precursor lesion to infiltrating cancer. Thus, identification of morphologic correlates of epigenetic alterations in otherwise normal tissues, or immunohistochemical markers of such alterations, may become an important aspect of cancer risk assessment. Metastatic ability may also be explained by the epigenetic progenitor model. A commonly accepted model of metastasis hypothesizes that clonal evolution within a tumor leads to the emergence of highly metastatic clones within a heterogeneous mass (148). However, this model fails to explain the common occurrence of late metastatic disease that has markedly different properties from the primary tumor and that arises years after resection of a small primary malignancy. The ability to metastasize does not require subsequent mutation and clonal selection within a large tumor mass; this assertion is supported by studies that indicate a metastatic phenotype of primary carcinomas (149). Rather, metastatic ability may be an inherent property of the progenitor cell from which the tumor arises, even in earlystage disease, which requires common epigenetic changes rather than a rare mutational event. Third, the epigenetic progenitor model indicates that such alterations are inherently polyclonal. This is in contrast with the widely accepted model of cancer as a monoclonal 242 Iacobuzio-Donahue disorder that arises from an initiating mutation (150). These differences between the two models are particularly significant because the genetic progression model was proposed and accepted when little was known about epigenetic phenomena in cancer. SUMMARY Cancer is as much an epigenetic disease as a genetic disease. The global changes to the epigenome, including hypomethylation, hypermethylation, and chromatin alterations, will be made clearer by a systematic examination of the epigenome in cancer at the molecular level. The epigenetic progenitor model encompasses these many alterations in cancer and proposes that cancer formation involves epigenetic disruption of progenitor cells, followed by an initiating mutation and then genetic and epigenetic plasticity. The initial epigenetic disruption may perturb the normal balance-- either their number or their capacity for aberrant differentiation--between undifferentiated progenitor cells and differentiated committed cells within a given anatomical compartment, thereby providing a common mechanism that unifies neoplasia development. Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. DISCLOSURE STATEMENT The author is not aware of any biases that might be perceived as affecting the objectivity of this review. ACKNOWLEDGMENTS I am indebted to Dr. Andrew Feinberg for useful discussions of epigenetic changes in cancer and for his critical reading of this manuscript. LITERATURE CITED 1. Hansemann D. 1890. U ber asymmetrische Zelltheilung in Epithelkrebsen und deren biologische Bedeutung. Virchows Arch. Pathol. Anat. 119:299326 2. Boveri T. 1902. U ber mehrpolige Mitosen als Mittel zur Analyse des Zellkerns. Verh. D. Phys. Med. Ges. 35:6790 3. Martin GS. 2004. The road to Src. Oncogene 23:791017 4. Lee WH, Bookstein R, Hong F, Young LJ, Shew JY, et al. 1987. Human retinoblastoma susceptibility gene: cloning, identification, and sequence. Science 235:139499 5. Vogelstein B, Kinzler KW. 2004. Cancer genes and the pathways they control. Nat. Med. 10:78999 6. Feinberg AP, Tycko B. 2004. The history of cancer epigenetics. Nat. Rev. Cancer 4:14353 7. Olins DE, Olins AL. 2003. Chromatin history: our view from the bridge. Nat. Rev. Mol. Cell Biol. 4:80914 8. Allfrey VG, Faulkner R, Mirsky AE. 1964. Acetylation and methylation of histones and their possible role in the regulation of RNA synthesis. Proc. Natl. Acad. Sci. USA 51:78694 9. Feinberg AP, Vogelstein B. 1983. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature 301:8992 10. Gama-Sosa MA, Slagel VA, Trewyn RW, Oxenhandler R, Kuo KC, et al. 1983. The 5-methylcytosine content of DNA from human tumors. Nucleic Acids Res. 11:688394 11. Feinberg AP, Ohlsson R, Henikoff S. 2006. The epigenetic progenitor origin of human cancer. Nat. Rev. Genet. 7:2133 12. Baylin SB, Ohm JE. 2006. Epigenetic gene silencing in cancer--a mechanism for early oncogenic pathway addiction? Nat. Rev. Cancer 6:10716 13. Luger K. 2003. Structure and dynamic behavior of nucleosomes. Curr. Opin. Genet. Dev. 13:12735 14. Jenuwein T. 2006. The epigenetic magic of histone lysine methylation. FEBS J. 273:312135 15. Kouzarides T. 2007. Chromatin modifications and their function. Cell 128:693705 www.annualreviews.org Epigenetic Changes in Cancer 243 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 16. Regnier V, Vagnarelli P, Fukagawa T, Zerjal T, Burns E, et al. 2005. CENP-A is required for accurate chromosome segregation and sustained kinetochore association of BubR1. Mol. Cell Biol. 25:396781 17. Strahl BD, Allis CD. 2000. The language of covalent histone modifications. Nature 403:4145 18. Wang GG, Allis CD, Chi P. 2007. Chromatin remodeling and cancer. Part I: covalent histone modifi- cations. Trends Mol. Med. 13:36372 19. Jenuwein T, Allis CD. 2001. Translating the histone code. Science 293:107480 20. Gibbons RJ. 2005. Histone modifying and chromatin remodelling enzymes in cancer and dysplastic syndromes. Hum. Mol. Genet. 14:R8592 21. Fraga MF, Ballestar E, Villar-Garea A, Boix-Chornet M, Espada J, et al. 2005. Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nat. Genet. 37:391400 22. Margueron R, Trojer P, Reinberg D. 2005. The key to development: interpreting the histone code? Curr. Opin. Genet. Dev. 15:16376 23. Roth SY, Denu JM, Allis CD. 2001. Histone acetyltransferases. Annu. Rev. Biochem. 70:81120 24. Dokmanovic M, Clarke C, Marks PA. 2007. Histone deacetylase inhibitors: overview and perspectives. Mol. Cancer Res. 5:98189 25. Luo J, Li M, Tang Y, Laszkowska M, Roeder RG, et al. 2004. Acetylation of p53 augments its site-specific DNA binding both in vitro and in vivo. Proc. Natl. Acad. Sci. USA 101:225964 26. Boyes J, Byfield P, Nakatani Y, Ogryzko V. 1998. Regulation of activity of the transcription factor GATA-1 by acetylation. Nature 396:59498 27. Martinez-Balbas MA, Bauer UM, Nielsen SJ, Brehm A, Kouzarides T. 2000. Regulation of E2F1 activity by acetylation. EMBO J. 19:66271 28. Deckert J, Struhl K. 2001. Histone acetylation at promoters is differentially affected by specific activators and repressors. Mol. Cell Biol. 21:272635 29. Khochbin S, Verdel A, Lemercier C, Seigneurin-Berny D. 2001. Functional significance of histone deacetylase diversity. Curr. Opin. Genet. Dev. 11:16266 30. Zhou X, Marks PA, Rifkind RA, Richon VM. 2001. Cloning and characterization of a histone deacetylase, HDAC9. Proc. Natl. Acad. Sci. USA 98:1057277 31. Landry J, Sutton A, Tafrov ST, Heller RC, Stebbins J, et al. 2000. The silencing protein SIR2 and its homologs are NAD-dependent protein deacetylases. Proc. Natl. Acad. Sci. USA 97:580711 32. Frye RA. 1999. Characterization of five human cDNAs with homology to the yeast SIR2 gene: Sir2like proteins (sirtuins) metabolize NAD and may have protein ADP-ribosyltransferase activity. Biochem. Biophys. Res. Commun. 260:27379 33. Gray SG, Ekstrom TJ. 2001. The human histone deacetylase family. Exp. Cell Res. 262:7583 34. Robertson KD, Ait-Si-Ali S, Yokochi T, Wade PA, Jones PL, et al. 2000. DNMT1 forms a complex with Rb, E2F1 and HDAC1 and represses transcription from E2F-responsive promoters. Nat. Genet. 25:33842 35. Ito K, Barnes PJ, Adcock IM. 2000. Glucocorticoid receptor recruitment of histone deacetylase 2 inhibits interleukin-1-induced histone H4 acetylation on lysines 8 and 12. Mol. Cell Biol. 20:6891903 36. Ropero S, Fraga MF, Ballestar E, Hamelin R, Yamamoto H, et al. 2006. A truncating mutation of HDAC2 in human cancers confers resistance to histone deacetylase inhibition. Nat. Genet. 38:56669 37. Roelfsema JH, Peters DJ. 2007. Rubinstein-Taybi syndrome: clinical and molecular overview. Expert. Rev. Mol. Med. 9:116 38. Muraoka M, Konishi M, Kikuchi-Yanoshita R, Tanaka K, Shitara N, et al. 1996. P300 gene alterations in colorectal and gastric carcinomas. Oncogene 12:156569 39. Ionov Y, Nowak N, Perucho M, Markowitz S, Cowell JK. 2004. Manipulation of nonsense mediated decay identifies gene mutations in colon cancer cells with microsatellite instability. Oncogene 23:63945 40. Hess JL. 2004. Mechanisms of transformation by MLL. Crit. Rev. Eukaryot. Gene. Expr. 14:23554 41. Yang XJ. 2004. The diverse superfamily of lysine acetyltransferases and their roles in leukemia and other diseases. Nucleic Acids Res. 32:95976 42. Wang J, Hoshino T, Redner RL, Kajigaya S, Liu JM. 1998. ETO, fusion partner in t(8;21) acute myeloid leukemia, represses transcription by interaction with the human N-CoR/mSin3/HDAC1 complex. Proc. Natl. Acad. Sci. USA 95:1086065 244 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 43. Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, et al. 1998. Fusion proteins of the retinoic acid receptor recruit histone deacetylase in promyelocytic leukaemia. Nature 391:81518 44. Shi Y. 2007. Histone lysine demethylases: emerging roles in development, physiology and disease. Nat. Rev. Genet. 8:82933 45. Klose RJ, Zhang Y. 2007. Regulation of histone methylation by demethylimination and demethylation. Nat. Rev. Mol. Cell Biol. 8:30718 46. Sanders SL, Portoso M, Mata J, Bahler J, Allshire RC, et al. 2004. Methylation of histone h4 lysine 20 controls recruitment of Crb2 to sites of DNA damage. Cell 119:60314 47. Sims RJ 3rd, Millhouse S, Chen CF, Lewis BA, Erdjument-Bromage H, et al. 2007. Recognition of trimethylated histone H3 lysine 4 facilitates the recruitment of transcription postinitiation factors and premRNA splicing. Mol. Cell 28:66576 48. Plath K, Fang J, Mlynarczyk-Evans SK, Cao R, Worringer KA, et al. 2003. Role of histone H3 lysine 27 methylation in X inactivation. Science 300:13135 49. Ivanov GS, Ivanova T, Kurash J, Ivanov A, Chuikov S, et al. 2007. Methylation-acetylation interplay activates p53 in response to DNA damage. Mol. Cell Biol. 27:675669 50. Bannister AJ, Kouzarides T. 2005. Reversing histone methylation. Nature 436:11036 51. Huyen Y, Zgheib O, Ditullio RA Jr, Gorgoulis VG, Zacharatos P, et al. 2004. Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks. Nature 432:40611 52. Wysocka J, Swigut T, Milne TA, Dou Y, Zhang X, et al. 2005. Wdr5 associates with histone H3 methylated at K4 and is essential for H3 K4 methylation and vertebrate development. Cell 121:85972 53. Shi Y, Lan F, Matson C, Mulligan P, Whetstine JR, et al. 2004. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119:94153 54. Baujat G, Cormier-Daire V. 2007. Sotos syndrome. Orphanet. J. Rare Dis. 2:36 55. Jaju RJ, Fidler C, Haas OA, Strickson AJ, Watkins F, et al. 2001. A novel gene, NSD1, is fused to NUP98 in the t(5;11)(q35;p15.5) in de novo childhood acute myeloid leukemia. Blood 98:126467 56. Krivtsov AV, Armstrong SA. 2007. MLL translocations, histone modifications and leukaemia stem-cell development. Nat. Rev. Cancer 7:82333 57. Cao R, Zhang Y. 2004. The functions of E(Z)/EZH2-mediated methylation of lysine 27 in histone H3. Curr. Opin. Genet. Dev. 14:15564 58. Pasini D, Bracken AP, Jensen MR, Lazzerini Denchi E, Helin K. 2004. Suz12 is essential for mouse development and for EZH2 histone methyltransferase activity. EMBO J. 23:406171 59. Kotake Y, Cao R, Viatour P, Sage J, Zhang Y, et al. 2007. pRB family proteins are required for H3 K27 trimethylation and polycomb repression complexes binding to and silencing p16INK4 tumor suppressor gene. Genes Dev. 21:4954 60. Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, Kumar-Sinha C, et al. 2002. The polycomb group protein EZH2 is involved in progression of prostate cancer. Nature 419:62429 61. Visser HP, Gunster MJ, Kluin-Nelemans HC, Manders EM, Raaphorst FM, et al. 2001. The polycomb group protein EZH2 is upregulated in proliferating, cultured human mantle cell lymphoma. Br. J. Haematol. 112:95058 62. Kleer CG, Cao Q, Varambally S, Shen R, Ota I, et al. 2003. EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells. Proc. Natl. Acad. Sci. USA 100:1160611 63. Bracken AP, Pasini D, Capra M, Prosperini E, Colli E, et al. 2003. EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer. EMBO J. 22:532335 64. Kirmizis A, Bartley SM, Farnham PJ. 2003. Identification of the polycomb group protein SU(Z)12 as a potential molecular target for human cancer therapy. Mol. Cancer Ther. 2:11321 65. Buyse IM, Shao G, Huang S. 1995. The retinoblastoma protein binds to RIZ, a zinc-finger protein that shares an epitope with the adenovirus E1A protein. Proc. Natl. Acad. Sci. USA 92:446771 66. Chadwick RB, Jiang GL, Bennington GA, Yuan B, Johnson CK, et al. 2000. Candidate tumor suppressor RIZ is frequently involved in colorectal carcinogenesis. Proc. Natl. Acad. Sci. USA 97:266267 67. Tokumaru Y, Nomoto S, Jeronimo C, Henrique R, Harden S, et al. 2003. Biallelic inactivation of the RIZ1 gene in human gastric cancer. Oncogene 22:695458 www.annualreviews.org Epigenetic Changes in Cancer 245 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 68. Steele-Perkins G, Fang W, Yang XH, Van Gele M, Carling T, et al. 2001. Tumor formation and inactivation of RIZ1, an Rb-binding member of a nuclear protein-methyltransferase superfamily. Genes Dev. 15:225062 69. Kim KC, Geng L, Huang S. 2003. Inactivation of a histone methyltransferase by mutations in human cancers. Cancer Res. 63:761923 70. Garcia-Cao M, O'Sullivan R, Peters AH, Jenuwein T, Blasco MA. 2004. Epigenetic regulation of telomere length in mammalian cells by the Suv39h1 and Suv39h2 histone methyltransferases. Nat. Genet. 36:9499 71. Peters AH, O'Carroll D, Scherthan H, Mechtler K, Sauer S, et al. 2001. Loss of the Suv39h histone methyltransferases impairs mammalian heterochromatin and genome stability. Cell 107:32337 72. Jones PA, Baylin SB. 2002. The fundamental role of epigenetic events in cancer. Nat. Rev. Genet. 3:41528 73. Klose RJ, Bird AP. 2006. Genomic DNA methylation: the mark and its mediators. Trends Biochem. Sci. 31:8997 74. Jones PL, Veenstra GJ, Wade PA, Vermaak D, Kass SU, et al. 1998. Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat. Genet. 19:18791 75. Ng HH, Zhang Y, Hendrich B, Johnson CA, Turner BM, et al. 1999. MBD2 is a transcriptional repressor belonging to the MeCP1 histone deacetylase complex. Nat. Genet. 23:5861 76. Wade PA, Gegonne A, Jones PL, Ballestar E, Aubry F, et al. 1999. Mi-2 complex couples DNA methy- lation to chromatin remodelling and histone deacetylation. Nat. Genet. 23:6266 77. Ting AH, McGarvey KM, Baylin SB. 2006. The cancer epigenome--components and functional corre- lates. Genes Dev. 20:321531 78. Okano M, Bell DW, Haber DA, Li E. 1999. DNA methyltransferases DNMT3a and DNMT3b are essential for de novo methylation and mammalian development. Cell 99:24757 79. Ting AH, Jair KW, Suzuki H, Yen RW, Baylin SB, et al. 2004. CpG island hypermethylation is maintained in human colorectal cancer cells after RNAi-mediated depletion of DNMT1. Nat. Genet. 36:58284 80. Ting AH, Jair KW, Schuebel KE, Baylin SB. 2006. Differential requirement for DNA methyltransferase 1 in maintaining human cancer cell gene promoter hypermethylation. Cancer Res. 66:72935 81. Rhee I, Jair KW, Yen RW, Lengauer C, Herman JG, et al. 2000. CpG methylation is maintained in human cancer cells lacking DNMT1. Nature 404:10037 82. Rhee I, Bachman KE, Park BH, Jair KW, Yen RW, et al. 2002. DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. Nature 416:55256 83. D'Alessio AC, Szyf M. 2006. Epigenetic tete-a-tete: the bilateral relationship between chromatin mod- ifications and DNA methylation. Biochem. Cell Biol. 84:46376 84. Perini G, Diolaiti D, Porro A, Della Valle G. 2005. In vivo transcriptional regulation of N-Myc target genes is controlled by E-box methylation. Proc. Natl. Acad. Sci. USA 102:1211722 85. Walsh CP, Bestor TH. 1999. Cytosine methylation and mammalian development. Genes Dev. 13:2634 86. Babushok DV, Ostertag EM, Kazazian HH Jr. 2007. Current topics in genome evolution: molecular mechanisms of new gene formation. Cell Mol. Life Sci. 64:54254 87. Cho B, Lee H, Jeong S, Bang YJ, Lee HJ, et al. 2003. Promoter hypomethylation of a novel cancer/testis antigen gene CAGE is correlated with its aberrant expression and is seen in premalignant stage of gastric carcinoma. Biochem. Biophys. Res. Commun. 307:5263 88. Lee TS, Kim JW, Kang GH, Park NH, Song YS, et al. 2006. DNA hypomethylation of CAGE promotors in squamous cell carcinoma of uterine cervix. Ann. N. Y. Acad. Sci. 1091:21824 89. Kim SJ, Kang HS, Chang HL, Jung YC, Sim HB, et al. 2008. Promoter hypomethylation of the Nacetyltransferase 1 gene in breast cancer. Oncol. Rep. 19:66368 90. Watanabe M, Ogawa Y, Itoh K, Koiwa T, Kadin ME, et al. 2008. Hypomethylation of CD30 CpG islands with aberrant JunB expression drives CD30 induction in Hodgkin lymphoma and anaplastic large cell lymphoma. Lab. Invest. 88:4857 91. Gaudet F, Hodgson JG, Eden A, Jackson-Grusby L, Dausman J, et al. 2003. Induction of tumors in mice by genomic hypomethylation. Science 300:48992 92. Eden A, Gaudet F, Waghmare A, Jaenisch R. 2003. Chromosomal instability and tumors promoted by DNA hypomethylation. Science 300:455 246 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 93. Chen RZ, Pettersson U, Beard C, Jackson-Grusby L, Jaenisch R. 1998. DNA hypomethylation leads to elevated mutation rates. Nature 395:8993 94. Narayan A, Ji W, Zhang XY, Marrogi A, Graff JR, et al. 1998. Hypomethylation of pericentromeric DNA in breast adenocarcinomas. Int. J. Cancer 77:83338 95. Ehrlich M, Woods CB, Yu MC, Dubeau L, Yang F, et al. 2006. Quantitative analysis of associations between DNA hypermethylation, hypomethylation, and DNMT RNA levels in ovarian tumors. Oncogene 25:263645 96. Tuck-Muller CM, Narayan A, Tsien F, Smeets DF, Sawyer J, et al. 2000. DNA hypomethylation and unusual chromosome instability in cell lines from ICF syndrome patients. Cytogenet. Cell Genet. 89:12128 97. Lengauer C, Kinzler KW, Vogelstein B. 1997. DNA methylation and genetic instability in colorectal cancer cells. Proc. Natl. Acad. Sci. USA 94:254550 98. Tomonaga T, Matsushita K, Yamaguchi S, Oohashi T, Shimada H, et al. 2003. Overexpression and mistargeting of centromere proteinA in human primary colorectal cancer. Cancer Res. 63:351116 99. Qu GZ, Grundy PE, Narayan A, Ehrlich M. 1999. Frequent hypomethylation in Wilms tumors of pericentromeric DNA in chromosomes 1 and 16. Cancer Genet. Cytogenet. 109:3439 100. Qu G, Dubeau L, Narayan A, Yu MC, Ehrlich M. 1999. Satellite DNA hypomethylation versus overall genomic hypomethylation in ovarian epithelial tumors of different malignant potential. Mutat. Res. 423:91101 101. Badal V, Chuang LS, Tan EH, Badal S, Villa LL, et al. 2003. CpG methylation of human papillomavirus type 16 DNA in cervical cancer cell lines and in clinical specimens: Genomic hypomethylation correlates with carcinogenic progression. J. Virol. 77:622734 102. Hutchings IA, Tierney RJ, Kelly GL, Stylianou J, Rickinson AB, et al. 2006. Methylation status of the Epstein-Barr Virus (EBV) BamHI W latent cycle promoter and promoter activity: analysis with novel EBV-positive Burkitt and lymphoblastoid cell lines. J. Virol. 80:1070011 103. Esteller M. 2007. Epigenetic gene silencing in cancer: the DNA hypermethylome. Hum. Mol. Genet. 16:R5059 104. Baylin SB, Fearon ER, Vogelstein B, de Bustros A, Sharkis SJ, et al. 1987. Hypermethylation of the 5 region of the calcitonin gene is a property of human lymphoid and acute myeloid malignancies. Blood 70:41217 105. Greger V, Passarge E, Hopping W, Messmer E, Horsthemke B. 1989. Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma. Hum. Genet. 83:15558 106. Sakai T, Toguchida J, Ohtani N, Yandell DW, Rapaport JM, et al. 1991. Allele-specific hypermethylation of the retinoblastoma tumor-suppressor gene. Am. J. Hum. Genet. 48:88088 107. Ohtani-Fujita N, Fujita T, Aoike A, Osifchin NE, Robbins PD, et al. 1993. CpG methylation inactivates the promoter activity of the human retinoblastoma tumor-suppressor gene. Oncogene 8:106367 108. Ohtani-Fujita N, Dryja TP, Rapaport JM, Fujita T, Matsumura S, et al. 1997. Hypermethylation in the retinoblastoma gene is associated with unilateral, sporadic retinoblastoma. Cancer Genet. Cytogenet. 98:4349 109. Shames DS, Minna JD, Gazdar AF. 2007. DNA methylation in health, disease, and cancer. Curr. Mol. Med. 7:85102 110. Dammann R, Schagdarsurengin U, Seidel C, Strunnikova M, Rastetter M, et al. 2005. The tumor suppressor RASSF1A in human carcinogenesis: an update. Histol. Histopathol. 20:64563 111. Donninger H, Vos MD, Clark GJ. 2007. The RASSF1A tumor suppressor. J. Cell Sci. 120:316372 112. De Jong J, Weeda S, Gillis AJ, Oosterhuis JW, Looijenga LH. 2007. Differential methylation of the OCT3/4 upstream region in primary human testicular germ cell tumors. Oncol. Rep. 18:12732 113. Akimzhanov A, Krenacs L, Schlegel T, Klein-Hessling S, Bagdi E, et al. 2008. Epigenetic changes and suppression of the nuclear factor of activated t cell 1 (NFATC1) promoter in human lymphomas with defects in immunoreceptor signaling. Am. J. Pathol. 172:21524 114. Long C, Yin B, Lu Q, Zhou X, Hu J, et al. 2007. Promoter hypermethylation of the RUNX3 gene in esophageal squamous cell carcinoma. Cancer Invest. 25:68590 115. Akiyama Y, Watkins N, Suzuki H, Jair KW, van Engeland M, et al. 2003. GATA-4 and GATA-5 transcription factor genes and potential downstream antitumor target genes are epigenetically silenced in colorectal and gastric cancer. Mol. Cell Biol. 23:842939 www.annualreviews.org Epigenetic Changes in Cancer 247 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 116. Ricciardiello L, Goel A, Mantovani V, Fiorini T, Fossi S, et al. 2003. Frequent loss of hMLH1 by promoter hypermethylation leads to microsatellite instability in adenomatous polyps of patients with a single first-degree member affected by colon cancer. Cancer Res. 63:78792 117. Kawakami K, Ruszkiewicz A, Bennett G, Moore J, Grieu F, et al. 2006. DNA hypermethylation in the normal colonic mucosa of patients with colorectal cancer. Br. J. Cancer 94:59398 118. Jacinto FV, Esteller M. 2007. Mutator pathways unleashed by epigenetic silencing in human cancer. Mutagenesis 22:24753 119. Smith FM, Garfield AS, Ward A. 2006. Regulation of growth and metabolism by imprinted genes. Cytogenet. Genome Res. 113:27991 120. Haig D, Westoby M. 1989. Parent-specific gene expression and the triploid endosperm. Am. Nat. 134:147 121. Lewis A, Reik W. 2006. How imprinting centres work. Cytogenet. Genome Res. 113:8189 122. Thorvaldsen JL, Duran KL, Bartolomei MS. 1998. Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2. Genes Dev. 12:3693702 123. Cui H, Niemitz EL, Ravenel JD, Onyango P, Brandenburg SA, et al. 2001. Loss of imprinting of insulin- like growth factorII in Wilms' tumor commonly involves altered methylation but not mutations of CTCF or its binding site. Cancer Res. 61:494750 124. Cui H, Onyango P, Brandenburg S, Wu Y, Hsieh CL, et al. 2002. Loss of imprinting in colorectal cancer linked to hypomethylation of H19 and Igf2. Cancer Res. 62:644246 125. Nakagawa H, Chadwick RB, Peltomaki P, Plass C, Nakamura Y, et al. 2001. Loss of imprinting of the insulin-like growth factor II gene occurs by biallelic methylation in a core region of H19-associated CTCF-binding sites in colorectal cancer. Proc. Natl. Acad. Sci. USA 98:59196 126. Loukinov DI, Pugacheva E, Vatolin S, Pack SD, Moon H, et al. 2002. BORIS, a novel male germline-specific protein associated with epigenetic reprogramming events, shares the same 11-zinc-finger domain with CTCF, the insulator protein involved in reading imprinting marks in the soma. Proc. Natl. Acad. Sci. USA 99:680611 127. Kajii T, Ohama K. 1977. Androgenetic origin of hydatidiform mole. Nature 268:63334 128. Linder D. 1969. Gene loss in human teratomas. Proc. Natl. Acad. Sci. USA 63:699704 129. Scrable H, Cavenee W, Ghavimi F, Lovell M, Morgan K, et al. 1989. A model for embryonal rhabdomyosarcoma tumorigenesis that involves genome imprinting. Proc. Natl. Acad. Sci. USA 86:748084 130. Mannens M, Hoovers JM, Redeker E, Verjaal M, Feinberg AP, et al. 1994. Parental imprinting of human chromosome region 11p15.3-pter involved in the Beckwith-Wiedemann syndrome and various human neoplasia. Eur. J. Hum. Genet. 2:323 131. Ravenel JD, Broman KW, Perlman EJ, Niemitz EL, Jayawardena TM, et al. 2001. Loss of imprinting of insulin-like growth factorII (IGF2) gene in distinguishing specific biologic subtypes of Wilms' tumor. J. Natl. Cancer. Inst. 93:1698703 132. Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, et al. 2003. Loss of IGF2 imprinting: a potential marker of colorectal cancer risk. Science 299:175355 133. Woodson K, Flood A, Green L, Tangrea JA, Hanson J, et al. 2004. Loss of insulin-like growth factorII imprinting and the presence of screen-detected colorectal adenomas in women. J. Natl. Cancer. Inst. 96:40710 134. Sakatani T, Kaneda A, Iacobuzio-Donahue CA, Carter MG, de Boom Witzel S, et al. 2005. Loss of imprinting of IGF2 alters intestinal maturation and tumorigenesis in mice. Science 307:197678 135. Kaneda A, Wang CJ, Cheong R, Timp W, Onyango P, et al. 2007. Enhanced sensitivity to IGF-II signaling links loss of imprinting of IGF2 to increased cell proliferation and tumor risk. Proc. Natl. Acad. Sci. USA 104:2092631 136. Nakanishi H, Suda T, Katoh M, Watanabe A, Igishi T, et al. 2004. Loss of imprinting of PEG1/MEST in lung cancer cell lines. Oncol. Rep. 12:127378 137. Sato N, Matsubayashi H, Abe T, Fukushima N, Goggins M. 2005. Epigenetic down-regulation of CDKN1C/P57KIP2 in pancreatic ductal neoplasms identified by gene expression profiling. Clin. Cancer Res. 11:468188 138. Kang MJ, Park BJ, Byun DS, Park JI, Kim HJ, et al. 2000. Loss of imprinting and elevated expression of wild-type p73 in human gastric adenocarcinoma. Clin. Cancer. Res. 6:176771 248 Iacobuzio-Donahue Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. 139. Pardal R, Clarke MF, Morrison SJ. 2003. Applying the principles of stem-cell biology to cancer. Nat. Rev. Cancer 3:895902 140. Dai B, Rasmussen TP. 2007. Global epiproteomic signatures distinguish embryonic stem cells from differentiated cells. Stem Cells 25:256774 141. Suzuki H, Watkins DN, Jair KW, Schuebel KE, Markowitz SD, et al. 2004. Epigenetic inactivation of SFRP genes allows constitutive wnt signaling in colorectal cancer. Nat. Genet. 36:41722 142. Hanada M, Delia D, Aiello A, Stadtmauer E, Reed JC. 1993. Bcl-2 gene hypomethylation and high-level expression in B cell chronic lymphocytic leukemia. Blood 82:182028 143. Nakagawa T, Kanai Y, Ushijima S, Kitamura T, Kakizoe T, et al. 2005. DNA hypomethylation on pericentromeric satellite regions significantly correlates with loss of heterozygosity on chromosome 9 in urothelial carcinomas. J. Urol. 173:24346 144. Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE, et al. 1994. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat. Genet. 7:53640 145. Peters I, Vaske B, Albrecht K, Kuczyk MA, Jonas U, et al. 2007. Adiposity and age are statistically related to enhanced RASSF1A tumor suppressor gene promoter methylation in normal autopsy kidney tissue. Cancer Epidemiol. Biomarkers Prev. 16:252632 146. Matsubayashi H, Sato N, Fukushima N, Yeo CJ, Walter KM, et al. 2003. Methylation of cyclin D2 is observed frequently in pancreatic cancer but is also an age-related phenomenon in gastrointestinal tissues. Clin. Cancer Res. 9:144652 147. Cruz-Correa M, Cui H, Giardiello FM, Powe NR, Hylind L, et al. 2004. Loss of imprinting of insulin growth factor II gene: a potential heritable biomarker for colon neoplasia predisposition. Gastroenterology 126:96470 148. Fidler IJ. 2002. Critical determinants of metastasis. Semin. Cancer Biol. 12:8996 149. Ramaswamy S, Ross KN, Lander ES, Golub TR. 2003. A molecular signature of metastasis in primary solid tumors. Nat. Genet. 33:4954 150. Fearon ER, Vogelstein B. 1990. A genetic model for colorectal tumorigenesis. Cell 61:75967 www.annualreviews.org Epigenetic Changes in Cancer 249 Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Contents Annual Review of Pathology: Mechanisms of Disease Volume 4, 2009 The First Fifty Years in Research Peter A. Ward p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1 Graft Vascular Disease: Immune Response Meets the Vessel Wall Richard N. Mitchell p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p19 Molecular Pathology of Head and Neck Cancer: Implications for Diagnosis, Prognosis, and Treatment Sara I. Pai and William H. Westra p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p49 Mechanisms of Endothelial Dysfunction, Injury, and Death Jordan S. Pober, Wang Min, and John R. Bradley p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p71 The Pathogenesis of Pituitary Tumors Sylvia L. Asa and Shereen Ezzat p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p97 PTEN and the PI3-Kinase Pathway in Cancer Nader Chalhoub and Suzanne J. Baker p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 127 Pathogenesis of Classical and Lymphocyte-Predominant Hodgkin Lymphoma Roland Schmitz, Jens Stanelle, Martin-Leo Hansmann, and Ralf Kppers p p p p p p p p p p p p p 151 Molecular Genetics of Acute Lymphoblastic Leukemia Michael A. Teitell and Pier Paolo Pandolfi p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 175 MicroRNAs in Cancer Yong Sun Lee and Anindya Dutta p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 199 Epigenetic Changes in Cancer Christine A. Iacobuzio-Donahue p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 229 Molecular Pathogenesis and Diagnostics of Bladder Cancer Anirban P. Mitra and Richard J. Cote p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 251 Ovarian Cancer Kathleen R. Cho and Ie-Ming Shih p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 287 v Annu. Rev. Pathol. Mech. Dis. 2009.4:229-249. Downloaded from arjournals.annualreviews.org by Texas A&M University - College Station on 02/15/10. For personal use only. Drosophila Models of Neurodegenerative Diseases Bingwei Lu and Hannes Vogel p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 315 Serrated Polyps and Colorectal Cancer: New Pathway to Malignancy Amy E. Noffsinger p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 343 Nod-Like Receptors: Role in Innate Immunity and Inflammatory Disease Grace Chen, Michael H. Shaw, Yun-Gi Kim, and Gabriel Nun~ ez p p p p p p p p p p p p p p p p p p p p p p p 365 Tumor Suppressors, Chromosomal Instability, and Hepatitis C VirusAssociated Liver Cancer David R. McGivern and Stanley M. Lemon p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 399 The Immunopathogenesis of Rheumatoid Arthritis John B. Imboden p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 417 The Pathology of Chronic Obstructive Pulmonary Disease James C. Hogg and Wim Timens p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 435 Linking the Cellular Functions of BRCA Genes to Cancer Pathogenesis and Treatment Ashok R. Venkitaraman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 461 Regulation of Hepcidin and Iron-Overload Disease Pauline L. Lee and Ernest Beutler p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 489 The Brainstem and Serotonin in the Sudden Infant Death Syndrome Hannah C. Kinney, George B. Richerson, Susan M. Dymecki, Robert A. Darnall, and Eugene E. Nattie p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 517 Molecular Pathogenesis of Cutaneous Melanocytic Neoplasms Nageatte Ibrahim and Frank G. Haluska p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 551 Indexes Cumulative Index of Contributing Authors, Volumes 14 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 581 Cumulative Index of Chapter Titles, Volumes 14 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 583 Errata An online log of corrections to Annual Review of Pathology, Mechanisms of Disease articles may be found at http://pathol.annualreviews.org vi Contents