Oncogene (2012) 31, 3827–3844 & 2012 Macmillan Publishers Limited All rights reserved 0950-9232/12 www.nature.com/onc REVIEW Targeting the epigenome for treatment of

E-J Geutjes, PK Bajpe and R Bernards

Division of Molecular , Centre for Biomedical Genetics and Cancer Genomics Centre, The Netherlands Cancer Institute, Amsterdam, The Netherlands

Cancer genome analyses have revealed that the enzymes required for normal cellular physiology. Moreover, involved in epigenetic gene regulation are frequently epimutations can outnumber genetic abnormalities and deregulated in cancer. Here we describe the enzymes that often occur early in cancer development (Esteller et al., control the epigenetic state of the cell, how they are 1999; Schuebel et al., 2007; Chen et al., 2009). Such affected in cancer and how this knowledge can be epimutated genes are often tumour-suppressor genes exploited to treat cancer with a new arsenal of selective (TSGs), including the DNA-repair proteins MutL therapies. homologue-1 (MLH1)andBRCA1, the INK4A-ARF Oncogene (2012) 31, 3827–3844; doi:10.1038/onc.2011.552; locus-encoded cell-cycle inhibitors (CDKN2A and published online 5 December 2011 CDKN2B) and von Hippel–Lindau (VHL), which controls cell survival and angiogenesis. Many epigenetic Keywords: cancer epigenetics; epigenetic therapy; DNA regulators, including those responsible for silencing of methylation; histone modifications; nucleosome the above mentioned TSGs, are deregulated in cancer. remodelling Recent cancer genome analyses have identified an impressive and still increasing number of epigenetic enzymes that are deregulated in many types of , establishing a direct link between cancer genetics and Introduction cancer epigenetics. Recurrent loss-of-function and gain- of-function mutations affecting nucleosome-remodelling Epigenetics is the study of heritable changes in gene complexes, histone modifiers and DNA-modifying en- expression that do not involve alterations of the DNA zymes occur in many cancers (Table 1 and discussed sequence (Kelly et al., 2010). Many of these changes below), clearly demonstrating that at least three major involve alterations of the chromatin: the state in which epigenetic mechanisms are deregulated in cancer. DNA is stored inside the cell. Eukaryotic DNA is Epigenetic enzymes are attractive drug targets because packaged in nucleosomes, which comprise octamers of of the reversible nature of epigenetic modifications. The four core histone proteins (H2A, H2B, H3 and H4) recognition that epigenetic enzymes are frequently around which 147 bp of DNA is wrapped. Accessibility deregulated in cancer has fuelled the development of of DNA is determined by repositioning and restructur- inhibitors of these, referred to here as ‘epigenetic drugs’. ing of the nucleosomes by ATP-dependent chromatin Epigenetic drugs can be divided into two classes. The remodellers, through covalent modifications of the first class consists of non-selective agents, including the histones by histone-modifying enzymes and through clinically used inhibitors of DNA methyltransferases modifications on the DNA itself by DNA-modifying (DNMTi) and histone deacetylases (HDACi), and the enzymes. There is substantial interplay between the more recently discovered inhibitors of EZH2 and LSD1. many epigenetic regulators, which together control Of this class, both DNMTis and HDACis are approved transcription. Globally, chromatin can be in two states, for treatment of cancer. The second class consists of the open state (euchromatin) in which DNA is available more selective inhibitors, including those targeting for transcription and a closed state (heterochromatin) JAK2, G9a/GLP and DOT1L. In this review, we will in which the DNA is tightly packed, precluding its discuss those epigenetic enzymes deregulated in cancer transcription. and the feasibility to target these for cancer therapy. In addition to alterations in the primary DNA sequence, cancers often harbour multiple epigenetic DNA-modifying enzymes alterations, here referred to as epimutations, which also DNMTs convert the cytosine residues of CpG dinucleo- affect key regulatory signalling pathways that are tides into 5-methylcytosine, which can be further converted into 5-hydroxymethyl-20-deoxycytidine by Correspondence: Dr R Bernards, Division of Molecular Carcinogen- the Ten-Eleven-Translocation (TET) family (Tahiliani esis, Centre for Biomedical Genetics and Cancer Genomics Centre, et al., 2009; Ito et al., 2010). CpG dinucleotides are The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, highly enriched in small genomic regions, known as Amsterdam, The Netherlands. E-mail: [email protected] CpG islands, which frequently coincide with promoter Received 1 October 2011; accepted 27 October 2011; published online 5 regions. Most CpG dinucleotides are methylated except December 2011 those found in the promoter-located CpG islands, Targeting the epigenome for cancer treatment E-J Geutjes et al 3828 Table 1 Selection of epigenetic enzymes deregulated in cancer Epigenetic Function Cancer type Alteration Reference regulator

DNA-modifying enzymes DNMT3A Converts cytosine residues AML, MDS Mutations (Ley et al., 2010; Yan et al., 2011) to 5-methylcytosine TET2 Converts methylcytosine AML, MDS, MF Deletions, missense (Delhommeau et al., 2009; to hydroxy methylcytosine and nonsense Langemeijer et al., 2009; mutations Ko et al., 2010)

Histone-modifying enzymes Lysine acetyltransferases p300/CBP Acetylates histones H3, MDS, B-cell non- Translocations, (Lavau et al., 2000; Gui et al., 2011; H4 and non-histone proteins Hodgkin’s mutations Mullighan et al., 2011; and bladder cancer Pasqualucci et al., 2011) TIP60 Histones H2A and H4 Breast carcinoma LOH Gorrini et al. (2007) MYST3 Histones H2A and H4 Leukaemia, Translocations, (Kitabayashi et al., 2001b; medulloblastoma amplification Northcott et al., 2009) AIB-1 Co-activator , Amplification (Anzick et al., 1997; Sakakura et al., oesophageal cancer, 2000; Miller et al., 2003) gastric cancer TRRAP Adaptor protein for HAT Mutations Wei et al. (2008b) (2011) complexes

Lysine deacetylases HDAC1 Deacetylates histones and Prostate, gastric, colon Overexpression (Choi et al., 2001; Halkidou et al., non-histone proteins and breast cancers 2004; Zhang et al., 2005; Wilson et al., 2006; Weichert et al., 2008a) HDAC2 Deacetylates histones and Colorectal, cervical and Overexpression (Song et al., 2005; Weichert et al., non-histone proteins gastric cancers 2008a, 2008b) HDAC3 Deacetylates non-histone Overexpression Jin et al. (2008) proteins HDAC6 Deacetylates non-histone Breast cancer Overexpression Saji et al. (2005) proteins HDAC8 Deacetylates histones Neuroblastoma Overexpression Oehme et al. (2009)

Lysine methyltransferases G9a/KMT1C Histone H3K9 mono- and Lung and hepatocellular Overexpression (Kondo et al., 2008; Chen et al., 2010) dimethylase carcinoma SETDB1 Histone H3K9 trimethylase Melanoma Amplification Ceol et al. (2011) MLL1 Histone H3H4 trimethylase ALL, AML Translocations (Liu et al., 2009b; Marschalek, 2011) MLL2 and Histone H3H4 trimethylase Non-Hodgkin’s lym- Inactivating (Dalgliesh et al., 2010; Morin et al., MLL3 phoma, renal carcinoma mutations 2011; Parsons et al., 2011) and medulloblastoma EZH2 Histone H3K27 trimethylase Prostate, breast, endo- Amplification, (Varambally et al., 2002; Kleer et al., metrium, B-cell non- overexpression, 2003; Arisan et al., 2005; Raman et al., Hodgkin’s lymphoma gain-of-function 2005; Sudo et al., 2005; Weikert et al., and bladder carcinoma mutation 2005; Matsukawa et al., 2006) SETD2 Histone H3K36 methylase Clear cell renal Truncating mutations, (Dalgliesh et al., 2010; Varela et al., carcinoma, AML translocations 2011) NSD1 Histone H3K36 methylase Sotos syndrome, AML, Translocations, (Jaju et al., 2001; Kurotaki et al., 2002; neuroblastoma, glioma mutations, promoter Berdasco et al., 2009) methylation KMT8/RIZ1 Histone H3K9 trimethylase Colorectal cancer, Missense, frame shift (He et al., 1998; Chadwick et al., 2000; DLBL mutations, promoter Steele-Perkins et al., 2001) methylation Lysine demethylases KDM1A Histone H3K4me1/2 and Prostate and breast Overexpression (Metzger et al., 2005; Kahl et al., 2006; H3K9me1/2 demethylase cancer and neuroblastoma Schulte et al., 2009; Lim et al., 2010) KDM2B Histone H3K4 and K36 and Overexpression Frescas et al. (2008) demethylase adenocarcinomas KDM4C H3K9 and H3K36 Prostate cancer, breast Amplification, (Yang et al., 2000; Ehrbrecht et al., demethylase cancer, lymphoma, overexpression 2006; Liu et al., 2009a; Rui et al., 2010) medulloblastoma, oesophageal cancer KDM5A Histone H3K4 demethylase AML Translocations (Wang et al., 2009a; Zeng et al., 2010) KDM5B Histone H3K4 demethylase Breast and prostate Overexpression, (Lu et al., 1999; Roesch et al., 2005; cancer, melanoma downregulation Xiang et al., 2007) KDM5C Histone H3K4 demethylase Renal cell carcinoma Inactivating mutations (Dalgliesh et al., 2010; Varela et al., 2011) KDM6A Histone H3K27 demethylase Lymphoma, renal cell Inactivating mutations (van Haaften et al., 2009; Dalgliesh carcinoma and bladder et al., 2010; Gui et al., 2011; Varela carcinoma et al., 2011; Wartman et al., 2011)

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3829 Table 1 Continued Epigenetic Function Cancer type Alteration Reference regulator

Ubiquitin ligases BMI1 Cooperates with Ring1B B-cell lymphoma Amplification and (Bea et al., 2001; Rubio-Moscardo for H2A ubiquitination overexpression et al., 2005) BRCA1 Ubiquitinates H2A at Hereditary breast and satellite DNA ovarian cancers

Kinases JAK2 Phosphorylates H3Y41 MDS Gain-of-function Kilpivaara and Levine (2008) mutation

Chromatin remodellers ARID1A Component of the Ovarian clear cell Truncating mutations, (Jones et al., 2010; Wiegand et al., SWI/SNF complex carcinoma, endometroid homozygous deletions 2010; Parsons et al., 2011; carcinoma, Varela et al., 2011) medulloblastoma ARID2 Component of the Hepatocellular Mutations Li et al. (2011) SWI/SNF complex carcinoma SNF5 Component of the Rhabdoid tumours, Homozygous deletions, (Versteege et al., 1998; Grand et al., SWI/SNF complex medulloblastoma nonsense, missense and 1999; Rousseau-Merck et al., 1999; frame-shift mutations, Sevenet et al., 1999; Biegel et al., promoter methylation 2000a, 2000b; Yuge et al., 2000; Jackson et al., 2009) BRG1 ATPase subunit of the Non-small-cell-lung Truncating mutations, (Reisman et al., 2003; Fukuoka SWI/SNF complex cancer, , missense mutations and et al., 2004; Medina et al., 2004; medulloblastoma nonsense mutations, Rodriguez-Nieto et al., 2011; promoter methylation Parsons et al., 2011) PB1 Component of the Renal cell carcinoma Truncating, nonsense, Varela et al. (2011) SWI/SNF complex missense and frame-shift mutations BRD7 Component of the Breast cancer Genomic loss, reduced Drost et al. (2010) SWI/SNF complex expression CHD5 Gliomas, Homozygous deletion, (Bagchi et al., 2007; Fujita et al., 2008; neuroblastomas, promoter methylation Mulero-Navarro and Esteller, 2008) colorectal and breast carcinomas

Abbreviations: AML, acute myeloid leukaemia; CML, chronic myeloid leukaemia; CTCL, cutaneous T-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; LOH, loss-of-heterozygosity; MDS, myelodysplastic syndrome; MF, myelofibrosis; MM, multiple myeloma. excluding imprinted loci and the inactive X-chromo- of TSGs (Ting et al., 2006; Steine et al., 2011). DNMTs some. While the function of 5-hydroxylmethylation is can also be mislocalized in cancer, resulting in aberrant unclear, hyper-methylation of DNA at CpG islands hyper-methylation. For example, the leukaemia- inhibits transcription by preventing recognition by promoting PML-RARa fusion protein and several transcriptional activators and by serving as a recogni- Polycomb Group (PcG) proteins induce aberrant tion signal for specific chromatin interactors, which in promoter hyper-methylation by recruiting DNMTs to turn recruit co-repressors and nucleosome remodellers. target promoters of TSGs (Di Croce et al., 2002; Vire Of the five DNMTs only DNMT1, 3A and 3B have et al., 2006). Highly recurrent heterozygous mutations in catalytic activity. DNMT3A and DNMT3B can methy- DNMT3A, yielding a partially defective enzyme, have late previously unmethylated sites, which are then recently been discovered in acute myeloid leukaemia copied to daughter cells by DNMT1 during replication. (AML) and myelodysplastic syndrome, and are asso- The cancer epigenome is marked by genome-wide ciated with poor clinical outcome (Ley et al., 2010; DNA hypo-methylation and site-specific DNA hyper- Yan et al., 2011). These mutations were proposed to methylation. DNA hypo-methylation promotes genomic compromise the binding of DNMT3A to chromatin instability, activates tumour-promoting genes and leads or to proteins associating with DNMT3A. Bi- and to loss of imprinting, whereas DNA hyper-methylation monoallelic loss-of-function mutations in TET2 were blocks transcription factor-binding sites and silences identified in many haematological malignancies, in TSGs. DNA-modifying enzymes are overexpressed, association with global 5hmC depletion (Delhommeau mislocalized or mutated in cancer. The level and activity et al., 2009; Langemeijer et al., 2009; Ko et al., 2010). of DNMTs appear to be modestly elevated in various Biallelic Tet2 loss in the mouse impairs hematopoietic cancer types, which could lead to aberrant promoter differentiation and induces myeloid transformation. hyper-methylation, as DNMT3B overexpression in Tet2 haploinsufficiency confers increased self-renewal normal mouse colon induces de novo DNA methylation and myeloproliferation, suggesting that TET2 functions

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3830 as a haploinsufficient TSG (Moran-Crusio et al., 2011; are frequently mutated or dysregulated in cancer. Quivoron et al., 2011). Monoallelic mutation of CBP causes Rubinstein–Taybi DNMTis such as the azanucleosides azacitidine syndrome, a severe developmental disorder that predis- (5-azacytidine) and decitabine (5-aza02-deoxycytidine) poses to cancer development (Petrij et al., 1995). induce differentiation and/or . At low doses, Recently, frequent genomic deletions and mutations of azanucleosides sequester and degrade DNMTs after CBP, all concentrated in the catalytic domain, were incorporation into DNA, leading to global DNA detected in B-cell non-Hodgkin’s lymphoma, bladder demethylation and consequent reactivation of promoter cancer and tumours of patients with relapsed lympho- hyper-methylated genes, including aberrantly silenced blastic leukaemia (Gui et al., 2011; Mullighan et al., TSGs. However, azanucleosides also induce gene 2011; Pasqualucci et al., 2011). In B-cell non-Hodgkin’s expression changes independent of active DNA lymphomas, the CBP mutants were unable to acetylate demethylation, for example by inducing degradation of p53, thereby compromising its transcriptional activity, the histone methyltransferase G9a, which may also and BCL6, blocking its transcriptional repressor func- contribute to their clinical effectiveness (Gius et al., tion. Mice with monoallelic ablation of Cbp develop 2004; Wozniak et al., 2007). Azacitidine and decitabine highly penetrant leukaemia’s, demonstrating that CBP are clinically used in high-risk myelodysplastic syndrome has tumour-suppressive activity (Kung et al., 2000). and have demonstrated clinical efficacy in treatment of CBP and p300 are also dysregulated by viral oncopro- AML (reviewed by Kelly et al., 2010) (Table 2). The teins and translocations. The viral oncoproteins E1A many clinical trials testing DNMTis have failed to find and Large T misregulate p300 and CBP to activate other DNMTi-responsive cancers, in particular solid cell-cycle progression-promoting genes by histone acet- tumours (reviewed by Quintas-Cardama et al., 2010). ylation, leading thereby contributing to cellular trans- Only some therapeutic benefit of DNMTi treatment was formation (Eckner et al., 1994; Ferrari et al., 2008). observed in chronic myeloid leukaemia (Aribi et al., In , p300 and CBP are commonly fused to the 2007; Wijermans et al., 2008). Besides azacitidine and histone methyltransferase Mixed Lineage Leukemia decitabine, there are many new DNMTis that are all (MLL) (see also lysine methyltransferases (KMTs)) capable of reverting aberrant promoter hyper-methyla- and two other KATs, MYST3 or MYST4/MORF/ tion (Figure 1). Azanucleosides S-110 and zebularine, KAT6B. The MLL-CBP fusion protein induces and the quinoline-based DNMTi SGI-1027, are more myelodysplastic syndrome in mice (Lavau et al., 2000). stable than azacitidine and decitabine, and show activity In contrast to CBP and p300, MYST3 appears to in cancer cell lines and mouse models of cancer (Cheng have oncogenic properties. Recurrent amplifications of et al., 2003; Yoo et al., 2007; Datta et al., 2009). Isoform- MYST3 have recently been identified in medulloblasto- specific inhibitors of DNMT1 (RG108) and DNMT3B ma. (Northcott et al., 2009). Unlike MLL-CBP and (nanaomycin-A) have been developed, but only nanao- MLL-MYST4, the MYST3-CBP fusion protein func- mycin-A appears to have potent antiproliferative effects tions as a transcriptional repressor by blocking hema- (Brueckner et al., 2005; Kuck et al., 2010). topoietic stem cell differentiation by inhibition of RUNX1-dependent transcription (Kitabayashi et al., 2001a). TIP60 may function as a haploinsufficient TSG Histone-modifying enzymes as monoallelic losses of TIP60 have been detected in Histones can be modified dynamically by multiple lymphomas, breast carcinomas, and head and neck chemical groups, including acetyl, methyl, phosphoryl cancer (Gorrini et al., 2007). Monoallelic loss of Tip60 and ubiquityl groups. These modifications serve to open impairs the Myc-induced DNA-damage response and is or close the chromatin structure, and promote or sufficient to counteract Myc-induced lymphomagenesis occlude proteins from binding to the chromatin. in mouse experimental models (Gorrini et al., 2007). In Broadly, we can discriminate three classes of histone- addition, recurrent mutations in the transformation/ interacting proteins. The writers place histone modifica- transcription domain-associated protein (TRRAP), tions, which can subsequently be removed by erasers. encoding a cofactor of TIP60-containing complexes, readers Finally, bind to chromatin through specific were recently described in melanoma, suggesting that domains that read the histone modifications and can inhibition of TIP60 complexes contributes to carcino- deliver nucleosome, histone or DNA-modifying enzymes. genesis (Wei et al., 2011). Apart from the classical KATs, amplified in breast cancer-1 (AIB1) also has Lysine acetylases. Lysine acetyltransferases (KATs) intrinsic KAT activity and associates with CBP and acetylate the lysine residues of histone proteins and p300 (Chen et al., 1997). AIB1 is involved in many non-histone proteins, including p53 and E2F1. They can signalling pathways important for oncogenesis of breast be divided into three classes, namely GCN5/PCAF, cancer, for example by functioning as a co-activator for p300/CBP and MYST/TIP60. The GCN5/PCAF and oestrogen and progesterone signalling (Lahusen et al., p300/CBP groups mainly function as coactivators for 2009). Amplification of AIB1 frequently occurs in breast various transcription factors, but the MYST family also cancer and ablation of Aib1 in the mouse leads to B-cell has roles in other nuclear processes. lymphoma (Anzick et al., 1997; Coste et al., 2006). The structural and functional homologues p300/ Inhibitors of p300 and CBP, including garcinol, KAT3A and CBP/KAT3B, and the MYST family curcumin and anacardic acid, were reported to kill members TIP60/KAT5 and MYST3/MOZ/KAT6A, cancer cells but not non-malignant cells (Balasubrama-

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3831 Table 2 Selected clinical trials of epigenetic therapies Epigenetic target Agent (INN) Marketed by Study/indication Phase of study and Reference number of trials

DNMT inhibitors Azacitidine Celgene MDS FDA-approved Kaminskas et al. (2005) Azacitidine Celgene AML P3 (2) (Fenaux et al., 2009; Fenaux et al., 2010) Azacitidine Celgene MF, MM, prostate P2 (4) (Sonpavde et al., 2007; cancer, lung cancer Quintas-Cardama et al., 2008) Decitabine Supergen/Eisai MDS FDA-approved Kantarjian et al. (2006) Decitabine Supergen/Eisai AML P3 Decitabine Supergen/ (Gleevec-refractory) P2 (7) (Issa et al., 2005; Eisai CML Oki et al., 2008; Wijermans et al., 2008) HDAC inhibitors Vorinostat Merck CTCL FDA-approved Mann et al. (2007) Vorinostat Merck Advanced mesothelioma P3 Vorinostat Merck Many cancers P2 (23) (Crump et al., 2008; Luu et al., 2008; Modesitt et al., 2008; Vansteenkiste et al., 2008; Bradley et al., 2009; Galanis et al., 2009; Schaefer et al., 2009; Traynor et al., 2009; Woyach et al., 2009; Kirschbaum et al., 2011) Romidepsin Celgene CTCL FDA-approved (Piekarz et al., 2009; Whittaker et al., 2010) Romidepsin Celgene Many cancers P2 (25) (Stadler et al., 2006; Schrump et al., 2008; Whitehead et al., 2009; Molife et al., 2010; Otterson et al., 2010; Niesvizky et al., 2011) Panobinostat Hodgkin’s with CR P3 Dickinson et al. (2009) after Panobinostat Novartis CTCL P2/3 (2) Panobinostat Novartis Refractory CML P2/3 (2) Panobinostat Novartis Many cancers P2 (14) Entinostat Syndax Hodgkin’s, ER þ P2 Hauschild et al. (2008) breast cancer, melano- ma Mocetinostat Methylgene CLL, Hodgkin’s, P2 (3) Blum et al. (2009) AML, MDS SB939 S*Bio Sarcoma, MF, P2 (3) prostate cancer Givinostat Italfarmaco MF P2 (3) Rambaldi et al. (2010) Topotarget/ Many cancers P2(9) (Ramalingam et al., 2009; Spectrum Giaccone et al., 2011) JAK2 inhibitors Ruxolitinib Incyte MF P3 (3) (Verstovsek et al., 2010; Mesa et al., 2011) Ruxolitinib Incyte Prostate cancer, P2 (4) advanced leukaemia AZD1480 AstraZeneca MF P1/2 (1) SB1518 S*Bio MF, advanced P1/2 (3) leukaemia Lestaurtinib Cephalon MF, advanced P1/2 (3) (Knapper et al., 2006; leukaemia Santos et al., 2010) DNMT þ HDAC Azacitidine þ AML, MDS, DLBCL P2 (3) inhibitors Vorinostat Azacitidine þ Lung, breast and colon P2 (7) Entinostat cancer, AML, MDS, CML Decitabine þ AML, MDS P2 (2) (Garcia-Manero et al., 2006; Valproic acid Voso et al., 2009) Decitabine þ AML, MDS P1/2 (1) Panobinostat

Abbreviations: AML, acute myeloid leukaemia; CML, chronic myeloid leukaemia; CR, complete remission; Gleevec, BCR-ABL inhibitor; CTCL, cutaneous T-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; ER þ , oestrogen receptor-positive; MDS, myelodysplastic syndrome; MF, myelofibrosis; MM, multiple myeloma. Shown are all clinical trials in which epigenetic drugs are being tested for efficacy (phase-2 and 3) or have been approved (FDA) in cancer therapy as monotherapy or as combination therapy. P (phase) and number of trials are depicted in parentheses.

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3832 RG108 (bromo domains) of BRD2-4 and BRDT (Filippako- poulos et al., 2010) (Figure 2). Translocations of BDR4 with nuclear protein in testis (NUT) frequently occur in azacitidine an incurable subtype of squamous carcinoma. JQ1 was decitabine DNMT1 found to displace the BDR4 fusion oncoprotein from SGI-1027 M the chromatin in squamous carcinoma cell lines, M S-110 resulting in tumour regression and prolonged survival M in xenograft experiments with patient-derived tumour zebularine M DNMT3B cells expressing BRD4-NUT. Recently, JQ1 was re- disulfiram M M ported to block leukemogenesis in an AML mouse model in which BRD4 was identified as being critically Nanaomycin A required for tumourigenesis (Zuber et al., 2011). JQ1 Figure 1 Inhibitors of DNMTs. Shown are all inhibitors of also induced senescence in mouse models of multiple DNMTs. writers are depicted in yellow, erasers in blue and readers myeloma by downregulation of Myc, which was in purple. M, methylated. regulated by bromo domain-containing proteins (Delmore et al., 2011).

HDACs Lysine deacetylases. HDACs remove acetyl groups Ac Ac from lysine residues on histones and non-histone proteins, and are considered to be transcriptional vorinostat repressors. Eighteen HDACs have been identified that romidepsin are grouped into four classes. Class-I, II and IV HDACs panobinostat are Zn2 þ -dependent enzymes, whereas class-III HDACs, entinostat referred to as sirtuins, use NAD þ as cofactor. The mocetinostat HDAC isoforms function in many cellular processes and givinostat Ac Ac SB939 have distinct gene expression patterns, cellular localiza- tion and function (Haberland et al., 2009b). HDACs KAT3 BRD2-4 commonly interact with DNA-binding proteins, such as BRDT transcription factors, which in turn target HDACs to specific DNA sequences. Some HDACs reside in multi- garcinol protein complexes, which associate with specific geno- JQ1 curcumin anacardic mic loci. For example, HDAC1 and 2 are constituents of acid the NuRD, SIN3A and Co-REST repressor complexes, CTK7A and HDAC3 is found within the N-COR and SMRT Figure 2 Inhibitors of factors involved in lysine acetylation. repressor complexes. Shown are all inhibitors of KATs, HDACs and BRD proteins. The cancer epigenome is hallmarked by a global writers are depicted in yellow, erasers in blue and readers in purple. reduction of lysine hypo-acetylation and many HDACs Marks associated with gene activation are in green and marks are altered in expression or mislocalized in cancer. Class-I associated with gene repression are in red. Ac, acetylated. HDACs (1, 2, 3 and 8) are overexpressed in many tumours and their increased expression correlates with poor outcome in some tumours, suggesting that these HDACs nyam et al., 2003; Eliseeva et al., 2007) (Figure 2). In are oncogenes (Weichert, 2009). The tumour growth of addition, a novel small-molecule inhibitor of p300 transformed fibroblasts can be completely blocked by substantially reduced growth of patient-derived oral genetic deletion of both Hdac1 and 2 (Haberland et al., squamous cell carcinoma xenografts in mice, in which, 2009a). Transgenic expression of a catalytically inactive unlike most cancers, histones are highly hyper-acety- Hdac2 in tumour-prone Apc þ /À mice reduced intestinal lated (Arif et al., 2010). Given the tumour-suppressive tumour incidence (Zimmermann et al., 2007). Inactivation role of CBP, p300 is likely the relevant target of these of HDAC8 induces differentiation of neuroblastoma cells inhibitors. Inhibitors of p300 may be particularly (Oehme et al., 2009). The role of the other Zn2 þ -dependent effective in cancers in which p300 is translocated or HDACs in carcinogenesis has not been explored thor- dysregyulated. A recent study found that inhibition of oughly, with the exception of HDAC6. HDAC6 is p300 impaired leukemogenesis in mouse models expres- overexpressed at the protein level in breast cancer and sing the oncogenic translocation product AML1-ETO loss of Hdac6 cooperates with oncogenic Ras in mouse by blocking the acetylation of AML1-ETO by p300 fibroblasts (Saji et al., 2005; Lee et al., 2008). Similar to (Wang et al., 2011). Of the other KATs, MYST3 and DNMTs, HDACs can be recruited by oncogenic tran- AIB1 could represent novel drug targets in cancer scription factors. BCL6, PML-RARa, PLZF-RARa and therapy, given their pro-oncogenic role. Selective AML-ETO can recruit HDAC-containing repressor com- inhibitors of readers of histone lysine acetylation have plexes to repress specific target genes (Bolden et al., 2006). recently been developed. The small-molecule JQ1 binds HDACis can restore global lysine acetylation levels competitively to the acetyl-lysine recognition motifs in cancer cells and induce the de-acetylation of many

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3833 non-histone proteins (Choudhary et al., 2009). However, SUV39H1/KMT1A, SUV39H2/KMT1B, G9a/KMT1C the activities of HDACis may also be independent of and the closely related protein GLP/KMT1D, SETDB1/ HDAC inhibition as they have non-HDAC targets KMT1E and CLL8/KMT1F, of which G9a and (Bantscheff et al., 2011). HDACis induce apoptosis, SETDB1 may function as oncogenes. The KMT1 family differentiation and cell-cycle arrest in cancer cells, and of proteins methylate H3K9, which induces gene have anti-angiogenic, anti-invasive and immunomodula- silencing and heterochromatin formation. G9a is over- tory properties in vivo (Lane and Chabner, 2009). Current expressed in some tumours, including leukaemia and HDACis for cancer therapy are structurally very diverse lung cancer, and may suppress TSGs or regulate p53 agents that target Zn2 þ -dependent HDACs and have activity by methylation (Chen et al., 2010; Huang et al., marked differences in specificity and potency (Bradner 2010). SETDB1 is recurrently amplified in melanoma et al., 2010; Bantscheff et al., 2011). The HDACis and cooperates with oncogenic BRAF in accelerating vorinostat (SAHA) and romidepsin (FK228) are clinically oncogenesis, possibly by deregulation of HOX genes used in cutaneous T-cell lymphomas (Mann et al., 2007; (Ceol et al., 2011). In contrast to the KMT1 family, Whittaker et al., 2010) (Table 2 and Figure 2). Many other other H3K9 methyltransferases such as RIZ1/KMT8, HDACis are in clinical development, with the following SMYD4 and EHMT1 are inactivated by mutations, being tested in phase-2 clinical trials: belinostat (PXD- suggesting that these function as TSGs. KMT8/RIZ1 is 101), panobinostat (LBH589), SB939, givinostat commonly inactivated in breast, liver and colon cancers. (ITF2357), entinostat (SNDX-275/MS-275) and moceti- KMT8 can be silenced by promoter hyper-methylation; nostat (MGCD0103) (Wagner et al., 2010) (Table 2 and is located on the frequently deleted 1p36 chromosomal Figure 2). Most clinical trials testing HDACis found locus; and is subject to PR domain-targeting loss-of- variable therapeutic benefit of HDACis in cancers other function mutations (He et al., 1998; Chadwick et al., than T-cell lymphomas (reviewed by Wagner et al., 2010). 2000; Steele-Perkins et al., 2001). Kmt8-knockout mice Ongoing phase-3 clinical trials are testing the clinical develop tumours of a broad spectrum at a high efficacy of vorinostat in advanced mesothelioma, and incidence (Steele-Perkins et al., 2001). Finally, homo- panobinostat in Hodgkin’s lymphoma and refractory zygous deletions in poorly characterized SMYD4 and chronic myeloid leukaemia. Isoform-specific HDACis EHMT1 were found in medulloblastoma (Northcott targeting oncogenic HDACs have been developed (Fig- et al., 2009). ure 2). Selective inhibitors of HDAC6 or HDAC8 induce The MLL proteins (MLL1-5/KMT2A-E) are H3K4 tumour cell death by targeting pathways regulated by methyltransferases and are considered to be transcrip- these HDACs (Hideshima et al., 2005; Balasubramanian tional activators. Translocations of MLL occur in 10% et al., 2008; Namdar et al., 2010). HDACis can also be of AML and 70% of infant leukaemias, and are combined with DNMTis, given their ability to more associated with poor clinical outcome (Marschalek, effectively reactivate those TSGs, including CDKN2A and 2011). Seventy-one MLL fusion partners have been MLH1, that are silenced in association with histone hypo- identified that behave as dominantly acting oncogenes acetylation and promoter hyper-methylation (Cameron that promote leukaemogenesis. Many of these require et al., 1999). Moreover, this combination synergizes in the H3K79 methyltransferase DOT1L/KMT4 for leu- induction of cancer cell death in vitro and appears to have kemic cell growth by enhancing the expression of the enhanced clinical efficacy in early clinical studies (Zhu HOX genes HOXA9 and MEIS1, which are important et al., 2001; Garcia-Manero et al., 2006; Gore et al., 2006; for MLL-dependent leukaemogenesis (Ayton and Blum et al., 2007; Braiteh et al.,2008;Vosoet al., 2009; Cleary, 2003; Bernt et al., 2011; Daigle et al., 2011; Stathis et al., 2011). Marschalek, 2011; Nguyen et al., 2011). Recently, recurrent inactivating mutations in MLL2 and MLL3 Lysine methylases. Histone lysine methylation can were detected in non-Hodgkin’s lymphoma, renal induce gene activation or repression, depending on the carcinoma and medulloblastoma, suggesting that lysine involved. In general, methylation of H3K4, MLL2 and 3 can also function as tumour suppressors H3K36 and H3K79 is associated with gene activation, (Dalgliesh et al., 2010; Morin et al., 2011; Parsons et al., whereas methylation of H3K9, H3K27 and H4K20 is 2011). Given that MLL2 and 3 reside in complexes that characteristic of repressed chromatin. KMTs methylate are essential for normal development and differentia- the lysine residues of histone and non-histone proteins, tion, it was proposed that inactivation of MLL2 and 3 and all have a SET domain or a structurally related PR might lead to aberrant cellular proliferation (Parsons (PRDI-BF1 and RIZ homology) domain, which confers et al., 2011). the catalytic activity, with the exception of DOT1L/ The KMT3-family members function as transcrip- KMT4. Some KMTs are part of major transcription- tional activators by methylating H3K36 and consist of regulatory protein complexes. The MLL/KMT2-family SETD2/KMT3A, NSD1/KMT3B and SMYD2/ members reside in protein complexes that regulate the KMT3C, of which SETD2 and NSD1 are deregulated initiation or elongation of transcription, whereas EZH2/ in cancer. Truncating mutations in SETD2 were recently KMT6 is a core subunit of the Polycomb protein detected in clear cell renal carcinoma (Dalgliesh et al., complex-2 (PRC2), which controls the cell lineage by 2010; Varela et al., 2011). Deficiency of NSD1 leads to repressing HOX genes. Sotos syndrome, an overgrowth condition that leads Many KMTs are mutated, translocated or over- to developmental defects and predisposes to cancer expressed in cancer. The KMT1 family consists of development (Rahman, 2005). NSD1 is also frequently

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3834 fused to NUP98 in AML and induces leukaemia by H3K27 enforcing the expression of the proto-oncogenes Me HOXA7, HOXA9, HOXA10 and MEIS1 (Jaju et al., 2001). Finally, NSD1 promoter hyper-methylation was detected in many neuroblastoma and gliomas, and this KMT6 DZNep could predict poor outcome in high-risk neuroblastoma (Berdasco et al., 2009). BIX-01294 EZH2/KMT6 is an H3K27 methyltransferase, and UNC0638 KMT1D MeM MeMe together with SUZ12 and EED forms the core compo- E72 H3K9 H3K79 nents of the PRC2 repressor complex, which frequently BIX-01294 KMT1C UNC0638 induces aberrant gene repression in cancer. First, PRC2- KMT4 enriched genes frequently acquire aberrant promoter methylation, possibly by recruitment of DNMTs or HDACs by EZH2 and EED (van der Vlag and Otte, EPZ004777 1999; Vire et al., 2006; McGarvey et al., 2008). Second, Figure 3 Inhibitors of KMTs. Shown are all inhibitors of KMTs. EZH2 can associate with the melanoma antigen writers are depicted in yellow, erasers in blue and readers in purple. PRAME, which together appear to have a role in Marks associated with gene activation are in green and marks inhibiting retinoic acid signalling, which is known to associated with gene repression are in red. Me, methylated. suppress tumourigenesis (Epping et al., 2005). EZH2 is overexpressed in many cancers (Chase and Cross, 2011). EZH2 overexpression is associated with aggressive and Miranda et al., 2009). Given that PRC2 and HDACs can metastatic disease in both prostate and breast cancer, cooperate in transcriptional repression, 3-deazaneplano- and is associated with poor clinical outcome in breast cin and HDACis synergized in inducing apoptosis in carcinoma (Kleer et al., 2003). Recently, a gain-of- AML cells in association with enhanced reactivation of function mutation in the SET domain of EZH2, PRC2 target genes and in colorectal cancer cells by resulting in enhanced H3K27 trimethylation, was identi- reactivating a WNT inhibitor, repressed by hypo- fied in non-Hodgkin’s lymphoma (Morin et al., 2010; acetylation and H3K27 trimethylation (Jiang et al., Yap et al., 2011). Besides EZH2, the other PRC2 2008; Fiskus et al., 2009). component, SUZ12, can be translocated in endometrial cancer (Koontz et al., 2001). Inhibitors against some of the pro-oncogenic KMTs Lysine demethylases. Lysine demethylases (KDMs) re- have been developed (Figure 3). BIX-01294 and move methyl groups from (non-) histone proteins and are UNC638 are highly selective inhibitors of G9a and divided into two classes: lysine residues on the KDM1/ GLP, leading to demethylation of H3K9me2 in mouse LSD family and the Jumonji-C (JmjC) domain family embryonic stem cells (Kubicek et al., 2007; Chang et al., (KDM2–KDM8). KDM1 family enzymes are FAD- 2009; Vedadi et al., 2011). UNC638 inhibited cell dependent amine oxidases that only remove mono- and viability and growth of breast carcinoma cell lines in dimethylated lysines, whereas JmjC family enzymes are association with reduced H3K9 methylation of G9a- Fe(II) 2-oxogluterate-dependent enzymes that are also regulated genes (Vedadi et al., 2011). Of all the other able to remove trimethyl groups (Pedersen and Helin, H3K9 methyltransferases, only SETDB1 could represent 2010). a potential target for therapy of melanoma, perhaps LSD1/KDM1A demethylates H3K4me1/2 and most effectively in combination with a BRAF inhibitor, H3K9me1/2, and is required for development and given the functional cooperation between SETDB1 and differentiation (Wang et al., 2009b). LSD1 resides in BRAF in carcinogenesis. Targeting MLL-translocated many protein complexes that have either tumour-promot- leukaemias with DOT1L inhibitors is a promising novel ing or tumour-suppressive activities. Consequently, both therapeutic strategy. A potent and selective DOT1L high and low LSD1 expression have been linked to inhibitor (EPZ004777) blocked H3K79 methylation and carcinogenesis. High LSD1 expression is associated with consequent activation of leukaemogenic genes, including poor clinical outcome of prostate cancer, aggressive HOXA9 and MEIS1, in leukemic cells bearing the MLL biology of ER-negative breast cancer and poorly differ- translocations (Daigle et al., 2011). Importantly, this led entiated neuroblastoma (Metzger et al., 2005; Kahl et al., to selective killing of MLL-rearranged cells and induced 2006; Schulte et al., 2009; Lim et al., 2010). However, in survival in a mouse MLL xenograft model. NSD1 and certain breast cancers LSD1 suppresses when SETD2 can be excluded as drug targets as they appear to overexpressed (Wang et al., 2009c). function as tumour suppressors. EZH2, SUZ12 and The KDM4 family consists of JMJD2A/KDM4A, EED can be inhibited by 3-deazaneplanocin, leading to JMJD2B/KDM4B, JMJD2C/KDM4C and JMJD2D/ reactivation of PRC2-repressed genes and induction KMM4D, and catalyses the demethylation of of apoptotic cell death in cancer cell lines H3K9me2/3 and H3K36me2/3. A recent study found (Tan et al., 2007). However, 3-deazaneplanocin non- that the histone kinase JAK2 (see histone kinases) and specifically inhibits S-adenosyl-L-methionine-dependent GASC1, which are co-localized on the 9p24 chromosome activity of histone methyltransferases and was reported band that is frequently amplified in cancer, cooperate to to globally inhibit histone methylation (Tan et al., 2007; activate MYC (Rui et al., 2010). Enforced GASC1

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3835 expression in immortalized mammary epithelial cells leads H3K4 to transformation, and depletion of GASC1 inhibits the Me proliferation of oesophageal squamous carcinoma cells harbouring the GASC1 amplification (Cloos et al.,2006; Wissmann et al., 2007; Liu et al., 2009a). Collectively, polyamine these findings suggest a role for GASC1 as an oncogene. analogs KDM1A The KDM5 family consists of the H3K4me2/3 demethylases JARID1A/KDM5A, JARID1B/KDM5B, JARID1C/KDM5C and JARID1D/KDM5D, and func- Me Me tion as transcriptional repressors. JARID1A/KDM5A is H3K9 H3K36 commonly fused to NUP98 in AML and prevents the differentiation-associated removal of H3K4me3 at many KDM4 loci encoding leukaemogenic transcription factors, leading to their activation and consequent induction of leukaemia (Wang et al., 2009a). JARID1A is also overexpressed in hydroxymate gastric cancer and inactivates many senescence-associated analogs cell-cycle inhibitors, thereby preventing senescence induc- Figure 4 Inhibitors of KDMs. Shown are all inhibitors of KDMs, tion (Zeng et al., 2010). Inactivating mutations in HDACs and BRD proteins. writers are depicted in yellow, erasers KDM5C/JARID1C have been detected in renal cell in blue and readers in purple. Marks associated with gene carcinoma, often in association with loss of VHL,which activation are in green and marks associated with gene repression is commonly mutated in these tumours and is causal in are in red. Me, methylated. tumourigenesis (Dalgliesh et al., 2010; Varela et al., 2011). JARID1C was proposed to cooperate with VHL in promoting the activation of the hypoxia-inducible factor treatment of lymphomas that have amplified 9p24 as their pathway that promotes cell survival and angiogenesis. combined inhibition by RNA interference cooperated in Knockdown of JARID1C in VHL-deleted renal cell killing these lymphomas (Rui et al., 2010). carcinoma cells lead to reactivation of hypoxia-inducible factor-responsive genes and enhanced tumour growth in a Arginine methyltransferases. Arginine methyltrans- xenograft model (Niu et al., 2011). ferases add methyl groups to histone arginine residues KDM6A/UTX and KDM6B/JMJD3 antagonize PcG- and, similar to lysine methylation, the outcomes of these mediated gene repression by removing H3K27me3. modifications depend on the arginine residue involved. Mutations of UTX have been found in many cancers, Currently, there is only little evidence that arginine most notably lymphoma, renal cell and bladder carcino- ma (van Haaften et al., 2009; Dalgliesh et al., 2010; Gui methyltransferases are deregulated in cancer, although they interact with pro-oncogenic proteins (Bedford and et al., 2011; Varela et al., 2011; Wartman et al., 2011). Clarke, 2009). Inhibition of Jmjd3 in MEFs results in repression of the aforementioned cell-cycle-inhibitory Ink4a-Arf locus, a known target of the PRC1 complex (see histone Histone kinases. Many histone residues are phos- ubiquitinases), inducing their immortalization (Agger phorylated by kinases, including AKT, JAK2 and et al., 2009; Barradas et al., 2009; Wang et al., 2010). PIM1, which have well-established roles in cancer Of the KDMs, only GASC1 and LSD1 are potential development (Baek, 2011). Given that these kinases drug targets and inhibitors of these have been found regulate many other key signalling pathways, the (Figure 4). Specific polyamine analogues inhibit LSD1, contribution of their histone kinase activity to tumour- leading to increased H3K4me2 levels and reactivation of igenesis remains to be determined. We will therefore silenced TSGs, in conjunction with inhibition of cancer restrict our discussion to JAK2 for it has a clear role in cell growth in cell lines and xenograft models (Huang chromatin signalling and can be targeted in cancer et al., 2007, 2009). Given that silenced TSGs are often therapy. JAK2 initiates signalling cascades involved in depleted in H3K4me3 and histone acetylation, and the cell cycle and apoptosis. However, JAK2 also acquire aberrant CpG island hyper-methylation, com- phosphorylates H3Y41, leading to exclusion of hetero- bined use of LSD1 inhibitors and DNMTis or HDACis chromatin protein-1a from the chromatin, thereby synergized in TSG reactivation, and inhibited colorectal activating the haematopoietic oncogene LMO2 (Dawson and breast cancer growth (Huang et al., 2009, 2011). et al., 2009). Using a similar mechanism, JAK2 also However, although LSD1 inactivation inhibited the cooperates with GASC1 in the activation of MYC in xenograft growth of LSD1-overexpressing neuroblastoma leukaemias (see also histone demethylases) (Rui et al., cells, it induced tumourigenesis in Drosophila,raising 2010). JAK2-mediated chromatin signalling appears to concerns about the effects of LSD1 inhibition on non- be critical for embryonic stem cell self-renewal, a cancer cells (Schulte et al., 2009; Eliazer et al.,2011) process, which could contribute to the oncogenesis of (Wang et al., 2009c). Inhibitors of the KDM4 family have myeloproliferative diseases, given that these are clonal recently been developed (Hamada et al., 2010). Dual blood stem cell disorders (Griffiths et al., 2011). JAK2 inactivation of GASC1 by these compounds and JAK2 by gain-of-function mutations (JAK2V617F) are present in JAK2 inhibitors (see histone kinases) may be useful in the 50% of human myeloproliferative diseases and are

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3836 whereas the Drosophila H2B DUB USP36 is important JAK2JAK2 for stem cell maintenance (Henry et al., 2003; Buszczak H3Y41 et al., 2009). Drosophila USP7-GMPS specifically targets H2B for deubiquitination in polycomb mediated P P gene regulation and ecdysone pathway (van der Knaap et al., 2005; van der Knaap et al., 2010). USP3 is an ruxolitinib AZD1480 H2A and H2B de-ubiquitylase that is involved in the SB1518 DNA-damage response (Nicassio et al., 2007).

Nucleosome-remodelling enzymes Nucleosomes are dynamically repositioned to ensure DNA replication, DNA repair and transcription P P (Clapier and Cairns, 2009). Nucleosome occupancy induces chromatin condensation, which not only Figure 5 Inhibitors of JAK2. Shown are all inhibitors of JAK2. occludes the transcription factor-binding sites, but also writers are depicted in yellow, erasers in blue and readers in purple. impedes the progress of RNA and DNA polymerases. Marks associated with gene activation are in green and marks Specific complexes that contain ATPase activity associated with gene repression are in red. Y, phosphorylated. organize nucleosomes by ejecting, reconfiguring or moving nucleosomes to alternative positions along the DNA. ATP-dependent chromatin remodellers can be causal in the pathogenesis (Kilpivaara and Levine, subdivided into four classes: SWI/SNF, ISWI, CHD 2008). Several JAK2 inhibitors are in clinical develop- and INO80. Each family shares similar ATPase subunits ment (Figure 5 and Table 2). As much as half of all and core members, but has a different composition patients with myeloproliferative diseases, unselected for of unique subunits. The non-ATPase subunits are JAK2 mutations, responded to treatment with small required for recognition of DNA and histone modifica- molecules targeting JAK1/2 in phase-1–2 clinical trials tions, regulation of ATPase activity, and interaction (Verstovsek et al., 2010; Pardanani et al., 2011). with other chromatin-binding proteins or transcription factors. Histone-ubiquitinating and -de-ubiquitinating enzymes. SWI/SNF stands out as the most frequently deregu- Mono-ubiquitination of histones H2A and H2B has lated nucleosome-remodelling complex in cancer important transcriptional consequences. Whereas H2B (Wilson and Roberts, 2011). There are two SWI/SNF ubiquitination is mostly associated with transcriptional complexes, named as BAF (BRG1 or Brahma- activation and elongation, H2A ubiquitination is Associated Factors) and PBAF (Polybromo-associated associated with gene repression. H2B ubiquitination is BAF complex), which contain one of the two mutually catalysed by BRE1/RNF20 and H2A ubiquitination is exclusive ATPases BRM or BRG1, and the core catalysed by another PcG complex called PRC1 subunits SNF5/BAF47, BAF155 and BAF170 (Reisman consisting of the ubiquitin ligase RING1B/RNF2 and et al., 2009). The variant subunits may direct the SWI/ BMI1, which enhances the catalytic activity of SNF complexes to specific loci and these include RING1B. BMI1 may function as an oncogene as it is ARID1A/BAF250, which is only found in the BAF amplified in B-cell lymphoma (Bea et al., 2001; Rubio- complexes, and PB1/BAF180, ARID2/BAF200 and Moscardo et al., 2005), and Bmi1 cooperates with cMyc BRD7, which specifically reside in the PBAF complexes. in enhancing tumour aggressiveness and counteracts The SWI/SNF complexes usually promote transcrip- cMyc-induced apoptosis by repressing the Ink4a-Arf tional activation and regulate many processes such as locus (van Lohuizen et al., 1991; Jacobs et al., 1999). cell-cycle progression, differentiation and DNA repair Recently, it was shown that, besides BRE1, the tumour (Reisman et al., 2009). SNF5, ARID1A, ARID2, PB1, suppressor BRCA1 also ubiquitinates H2A, but speci- BRG1 and BRM are deleted, mutated or silenced in fically at DNA satellite repeats. Loss of BRCA1 induced many different tumours at a high frequency, showing H2A de-ubiquitination accompanied by de-repression of that SWI/SNF is an important tumour-suppressor satellite DNA, leading to genomic instability (Zhu et al., complex (Table 1). SWI/SNF can induce tumourigenesis 2011). by many mechanisms and these are likely to be cell type- Ubiquitin ligases are antagonized by de-ubiquitynat- dependent, given that many subunits are tissue specifi- ing enzymes (DUBs), which remove ubiquitin moieties cally expressed (Wilson and Roberts, 2011). This could from protein substrates. There are multiple DUBs also explain the distinct range of cancers associated with targeting histones in a gene-specific manner. the inactivation of each subunit. SNF5 is frequently De-ubiquitination of H2A by USP16 is important for homozygously deleted or mutated in rhabdoid tumours, normal cell-cycle progression and regulation of HOX which are very aggressive childhood cancers (Versteege genes, whereas 2A-DUB-mediated H2A de-ubiquitina- et al., 1998). In addition, one allele of SNF5 can be tion controls androgen receptor-dependent gene deleted whereas the other allele is silenced by promoter activation (Joo et al., 2007; Zhu et al., 2007). De- methylation (Versteege et al., 1998; Rousseau-Merck ubiquitination of H2B by USP22/UBP8 is required for et al., 1999; Biegel et al., 2000b, 2002; Biegel and SAGA complex-mediated transcriptional activation, Pollack, 2004). SNF5 is also heterozygously deleted in

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3837 the chronic and acute phases of chronic myeloid ways, inhibitors of JAK2, DOT1L and BRDs function leukaemia (Grand et al., 1999), and is mutated in by blocking those that are aberrantly hyper-activated in Hodgkin’s lymphoma (Yuge et al., 2000). Genetic cancer. The latter are highly selective inhibitors that ablation of Snf5 in the mouse recapitulates the incidence only appear to kill cancer cells with the deregulated and phenotype of tumours seen in humans (Klochendler- epigenetic target, possibly because the resultant epige- Yeivin et al., 2000; Roberts et al., 2000, 2002). netic changes may have addicted these cells to altered Both BRG1 and BRM show high degree of loss-of- signalling pathways (Filippakopoulos et al., 2010; heterozygosity in non-small-lung carcinoma and BRM is Verstovsek et al., 2010; Daigle et al., 2011). Thus, epigenetically silenced in various cell lines. BRG1 and tumours should be classified according to the deregula- BRM expression is simultaneously lost in 30–40% of tion of their epigenetic targets to allow effective lung cancer cell lines and 15–20% of non-small-cell lung epigenetic therapy. This becomes even more crucial carcinoma (Reisman et al., 2003; Fukuoka et al., 2004; with the recognition that some epigenetic enzymes Rodriguez-Nieto et al., 2011). Knockout of either Brg1 such as MLL2/3 and LSD1 can function both as or Brm leads to cancer development in mice with Brg1 tumour suppressors and oncoproteins. Molecular clas- functioning as a haploinsufficient TSG (Bultman et al., sification of tumours could also guide treatment with 2000, 2008; Glaros et al., 2008). Recently, high- broad-acting epigenetic drugs. DNMTis could be throughput sequencing identified highly recurrent muta- particularly useful in treating the subset of colorectal, tions in ARID1A, ARID2 and PB1 in many tumours. brain and breast tumours that have a high degree of Loss-of-function mutations in ARID1A were detected in CpG island promoter hyper-methylation (Toyota et al., approximately 50% of ovarian clear cell carcinomas, 1999; Noushmehr et al., 2010; Fang et al., 2011), which 30% of endometrioid carcinomas and 12% of bladder is now being clinically tested in colorectal cancer cancers (Jones et al., 2010; Wiegand et al., 2010; Gui patients. et al., 2011). PB1 is mutated in 41% of clear cell renal Although DNMTis and HDACis are clinically carcinomas and ARID2 is mutated in 18% of hepato- effective in some haematological malignancies, clinical cellular carcinomas (Li et al., 2011; Varela et al., 2011). proof of concept for solid tumours remains to be Besides SWI/SNF, the CHD-family member CHD5 is established. The clinical failures of current DNMTis and also frequently inactivated in cancer. Inactivation of HDACis are ascribed to dosing-related issues, drug CHD5 by homozygous deletions and promoter hyper- delivery problems, intrinsic resistance and the fact that methylation has been detected in gliomas, neuroblasto- these agents have pleiotropic effects. The incorporation mas, colorectal and breast cancers, and CHD5 ablation of DNMTis is DNA replication-dependent, meaning in the mouse enhances proliferation and blocks senes- that slowly proliferating cancer cells could survive cence through repression of the Ink4a-Arf locus (Bagchi DNMTi treatment (Issa and Kantarjian, 2009). Respon- et al., 2007; Fujita et al., 2008; Leary et al., 2008; siveness to DNMTis and HDACis is critically deter- Mulero-Navarro and Esteller, 2008). mined by pharmacogenetic factors that affect the uptake Strategies that restore the function of SWI/SNF and and metabolism of these agents (Voso et al., 2009; CHD5 are likely to be of therapeutic benefit, given the Geutjes et al., 2011). Such issues could potentially be importance of these complexes in cancer. BRG1, BRM avoided with the new DNMTis as they do not require and CHD5 can be inactivated by epimutations, raising metabolic activation or DNA replication for their the possibility that epigenetic therapy could also be used incorporation. Finally, global hypo-methylation and to revert these epimutations. Indeed, HDACis can non-specific HDAC inhibition activate both tumour- restore the expression of epimutated BRM and DNMTi promoting and tumour-inhibiting genes in a cell context- can reactivate silenced CHD5 in cancer cells (Yamamichi dependent manner (Gius et al., 2004; Piekarz and Bates, et al., 2005; Glaros et al., 2007; Mulero-Navarro and 2009). For example, oncogenes such as MYC, EGFR Esteller, 2008). and E2F1 were induced in liver cancer cell lines resistant to the DNMTi zebularine, but not in those sensitive to zebularine in which TSGs were reactivated (Andersen Future directions for epigenetic therapy et al., 2010). Although HDAC3 is overexpressed in We have provided an overview of the epigenetic enzymes many tumours, liver-specific inactivation of Hdac3 in that are deregulated in cancer and the feasibility to mice culminates in hepatocellular carcinoma (Bhaskara target these for cancer therapy. A bewildering amount et al., 2010). It has been proposed that the clinical of epigenetic enzymes, involved in multiple layers of inefficacy of DNMTis and HDACis in solid tumours epigenetic regulation, including DNA (hydroxyl)-methy- is caused by these agents promoting survival rather lation, histone modification and nucleosome remodel- than apoptosis in solid tumours as they have more ling, are deregulated at high frequency in many cancers (epi-) mutations and perhaps also more dysfunctional (Table 1). The clinical successes of DNMTis and apoptotic pathways than haematological malignancies HDACis clearly validate the usefulness of epigenetic (Nowell, 2002; Piekarz and Bates, 2009). In support therapy in cancer treatment, and other epigenetic of this, some solid tumours are sensitive to the enzymes such as MYST3, AIB1, SETDB1, GASC1 combination of DNMTis or HDACis with apoptosis- and BMI1 are potential drug targets in cancer therapy. inducing therapy, which together more effectively While inhibitors of DNMTs, HDACs, G9a/GLP, LSD1 trigger pro-apoptotic signalling pathways (reviewed by and EZH2 restore aberrantly silenced signalling path- Bolden et al., 2006; Boumber and Issa, 2011). Thus,

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3838 particularly in solid tumours, clinical success with The many epimutated TSGs and deregulated epige- DNMTis and HDACis, and possibly other broadly netic enzymes in cancer clearly demonstrate that acting agents, may require combinations with apoptosis- aberrant epigenetic regulation is a significant contribu- inducing therapy. tor to carcinogenesis. The reversibility of epigenetic Another approach would be to develop novel agents abnormalities makes the epigenetic enzymes responsible that selectively reactivate epimutated TSGs while for establishing these modifications good targets for avoiding global gene activation and off-target effects. therapeutic intervention. Molecular classification of However, for most silenced TSGs we have not identified tumours should guide treatment with epigenetic therapy the enzymes involved and consequently do not know the and can be combined with other anticancer agents to relevant drug targets. Loss-of-function genetic screens enhance clinical efficacy and to overcome or prevent with libraries that target chromatin modifiers could be drug resistance. The recent findings indicating that many useful in identifying such enzymes. RNA interference epigenetic enzymes are mutated in cancer also suggest screens have identified the mediators of epigenetic that cancer genome resequencing may become helpful to silencing of RASSF1A and the death receptor FAS guide the choice of epigenetic therapy. (Gazin et al., 2007; Palakurthy et al., 2009). Some of these were constituents of repressor complexes such as PcG. Targeting the subunits that are specific to these complexes might lead to reactivation of silenced TSGs Conflict of interest while avoiding global effects, perhaps translating into The authors declare no conflict of interest. improved clinical efficacy.

References

Agger K, Cloos PA, Rudkjaer L, Williams K, Andersen G, contributes to epigenetic control of INK4a/ARF by oncogenic Christensen J et al. (2009). The H3K27me3 demethylase RAS. Genes Dev 23: 1177–1182. JMJD3 contributes to the activation of the INK4A-ARF locus in Bea S, Tort F, Pinyol M, Puig X, Hernandez L, Hernandez S et al. response to oncogene- and stress-induced senescence. Genes Dev 23: (2001). BMI-1 gene amplification and overexpression in hematolo- 1171–1176. gical malignancies occur mainly in mantle cell lymphomas. Cancer Andersen JB, Factor VM, Marquardt JU, Raggi C, Lee YH, Seo D Res 61: 2409–2412. et al. (2010). An integrated genomic and epigenomic approach Bedford MT, Clarke SG. (2009). Protein arginine methylation in predicts therapeutic response to zebularine in human liver cancer. mammals: who, what, and why. Mol Cell 33: 1–13. Sci Transl Med 2: 54ra77. Berdasco M, Ropero S, Setien F, Fraga MF, Lapunzina P, Losson R Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, Guan et al. (2009). Epigenetic inactivation of the Sotos overgrowth XY et al. (1997). AIB1, a steroid receptor coactivator amplified in syndrome gene histone methyltransferase NSD1 in human neuro- breast and ovarian cancer. Science 277: 965–968. blastoma and glioma. Proc Natl Acad Sci USA 106: 21830–21835. Aribi A, Borthakur G, Ravandi F, Shan J, Davisson J, Cortes J et al. Bernt KM, Zhu N, Sinha AU, Vempati S, Faber J, Krivtsov AV et al. (2007). Activity of decitabine, a hypomethylating agent, in chronic (2011). MLL-rearranged leukemia is dependent on Aberrant H3K79 myelomonocytic leukemia. Cancer 109: 713–717. Methylation by DOT1L. Cancer Cell 20: 66–78. Arif M, Vedamurthy BM, Choudhari R, Ostwal YB, Mantelingu K, Bhaskara S, Knutson SK, Jiang G, Chandrasekharan MB, Wilson AJ, Kodaganur GS et al. (2010). Nitric oxide-mediated histone Zheng S et al. (2010). Hdac3 is essential for the maintenance of hyperacetylation in oral cancer: target for a water-soluble HAT chromatin structure and genome stability. Cancer Cell 18: 436–447. inhibitor, CTK7A. Chem Biol 17: 903–913. Biegel JA, Fogelgren B, Wainwright LM, Zhou JY, Bevan H, Rorke Arisan S, Buyuktuncer ED, Palavan-Unsal N, Caskurlu T, Cakir OO, LB. (2000a). Germline INI1 mutation in a patient with a central Ergenekon E. (2005). Increased expression of EZH2, a polycomb nervous system atypical teratoid tumor and renal rhabdoid tumor. group protein, in bladder carcinoma. Urol Int 75: 252–257. Genes Chromosomes Cancer 28: 31–37. Ayton PM, Cleary ML. (2003). Transformation of myeloid progeni- Biegel JA, Fogelgren B, Zhou JY, James CD, Janss AJ, Allen JC et al. tors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9. (2000b). Mutations of the INI1 rhabdoid tumor suppressor gene in Genes Dev 17: 2298–2307. medulloblastomas and primitive neuroectodermal tumors of the Baek SH. (2011). When signaling kinases meet histones and histone central nervous system. Clin Cancer Res 6: 2759–2763. modifiers in the nucleus. Mol Cell 42: 274–284. Biegel JA, Kalpana G, Knudsen ES, Packer RJ, Roberts CW, Thiele Bagchi A, Papazoglu C, Wu Y, Capurso D, Brodt M, Francis D et al. CJ et al. (2002). The role of INI1 and the SWI/SNF complex in the (2007). CHD5 is a tumor suppressor at human 1p36. Cell 128: 459–475. development of rhabdoid tumors: meeting summary from the Balasubramanian S, Ramos J, Luo W, Sirisawad M, Verner E, Buggy workshop on childhood atypical teratoid/rhabdoid tumors. Cancer JJ. (2008). A novel 8 (HDAC8)-specific inhibitor Res 62: 323–328. PCI-34051 induces apoptosis in T-cell lymphomas. Leukemia 22: Biegel JA, Pollack IF. (2004). Molecular analysis of pediatric brain 1026–1034. tumors. Curr Oncol Rep 6: 445–452. Balasubramanyam K, Swaminathan V, Ranganathan A, Kundu TK. Blum KA, Advani A, Fernandez L, Van Der Jagt R, Brandwein J, (2003). Small molecule modulators of histone acetyltransferase Kambhampati S et al. (2009). Phase II study of the histone p300. J Biol Chem 278: 19134–19140. deacetylase inhibitor MGCD0103 in patients with previously treated Bantscheff M, Hopf C, Savitski MM, Dittmann A, Grandi P, Michon chronic lymphocytic leukaemia. Br J Haematol 147: 507–514. AM et al. (2011). Chemoproteomics profiling of HDAC inhibitors Blum W, Klisovic RB, Hackanson B, Liu Z, Liu S, Devine H et al. reveals selective targeting of HDAC complexes. Nat Biotechnol 29: (2007). Phase I study of decitabine alone or in combination with 255–265. valproic acid in acute myeloid leukemia. J Clin Oncol 25: 3884–3891. Barradas M, Anderton E, Acosta JC, Li S, Banito A, Rodriguez- Bolden JE, Peart MJ, Johnstone RW. (2006). Anticancer activities of Niedenfuhr M et al. (2009). Histone demethylase JMJD3 histone deacetylase inhibitors. Nat Rev Drug Discov 5: 769–784.

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3839 Boumber Y, Issa JP. (2011). Epigenetics in cancer: what’s the future? Cloos PA, Christensen J, Agger K, Maiolica A, Rappsilber J, Antal T Oncology (Williston Park) 25: 220–226, 228. et al. (2006). The putative oncogene GASC1 demethylates tri- and Bradley D, Rathkopf D, Dunn R, Stadler WM, Liu G, Smith DC et al. dimethylated lysine 9 on histone H3. Nature 442: 307–311. (2009). Vorinostat in advanced prostate cancer patients progressing Coste A, Antal MC, Chan S, Kastner P, Mark M, O’Malley BW et al. on prior chemotherapy (National Cancer Institute Trial 6862): trial (2006). Absence of the steroid receptor coactivator-3 induces B-cell results and interleukin-6 analysis: a study by the Department of lymphoma. EMBO J 25: 2453–2464. Defense Prostate Cancer Consortium and University Crump M, Coiffier B, Jacobsen ED, Sun L, Ricker JL, Xie H et al. of Chicago Phase 2 Consortium. Cancer 115: 5541–5549. (2008). Phase II trial of oral vorinostat (suberoylanilide Bradner JE, West N, Grachan ML, Greenberg EF, Haggarty SJ, hydroxamic acid) in relapsed diffuse large-B-cell lymphoma. Ann Warnow T et al. (2010). Chemical phylogenetics of histone Oncol 19: 964–969. deacetylases. Nat Chem Biol 6: 238–243. Daigle SR, Olhava EJ, Therkelsen CA, Majer CR, Sneeringer CJ, Braiteh F, Soriano AO, Garcia-Manero G, Hong D, Johnson MM, Song J et al. (2011). Selective killing of mixed lineage leukemia Silva Lde P et al. (2008). Phase I study of epigenetic modulation cells by a potent small-molecule DOT1L inhibitor. Cancer Cell 20: with 5-azacytidine and valproic acid in patients with advanced 53–65. cancers. Clin Cancer Res 14: 6296–6301. Dalgliesh GL, Furge K, Greenman C, Chen L, Bignell G, Butler A Brueckner B, Garcia Boy R, Siedlecki P, Musch T, Kliem HC, et al. (2010). Systematic sequencing of renal carcinoma reveals Zielenkiewicz P et al. (2005). Epigenetic reactivation of tumor inactivation of histone modifying genes. Nature 463: 360–363. suppressor genes by a novel small-molecule inhibitor of human Datta J, Ghoshal K, Denny WA, Gamage SA, Brooke DG, DNA methyltransferases. Cancer Res 65: 6305–6311. Phiasivongsa P et al. (2009). A new class of quinoline-based DNA Bultman S, Gebuhr T, Yee D, La Mantia C, Nicholson J, Gilliam A hypomethylating agents reactivates tumor suppressor genes by et al. (2000). A Brg1 null mutation in the mouse reveals functional blocking DNA methyltransferase 1 activity and inducing its differences among mammalian SWI/SNF complexes. Mol Cell 6: degradation. Cancer Res 69: 4277–4285. 1287–1295. Dawson MA, Bannister AJ, Gottgens B, Foster SD, Bartke T, Bultman SJ, Herschkowitz JI, Godfrey V, Gebuhr TC, Yaniv M, Green AR et al. (2009). JAK2 phosphorylates histone H3Y41 and Perou CM et al. (2008). Characterization of mammary tumors from excludes HP1alpha from chromatin. Nature 461: 819–822. Brg1 heterozygous mice. Oncogene 27: 460–468. Delhommeau F, Dupont S, Della Valle V, James C, Trannoy S, Buszczak M, Paterno S, Spradling AC. (2009). Drosophila stem cells Masse A et al. (2009). Mutation in TET2 in myeloid cancers. N Engl share a common requirement for the histone H2B ubiquitin protease JMed360: 2289–2301. scrawny. Science 323: 248–251. Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM et al. Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB. (2011). BET bromodomain inhibition as a therapeutic strategy to (1999). Synergy of demethylation and histone deacetylase inhibition target c-Myc. Cell 146: 904–917. in the re-expression of genes silenced in cancer. Nat Genet 21: Dickinson M, Ritchie D, DeAngelo DJ, Spencer A, Ottmann OG, 103–107. Fischer T et al. (2009). Preliminary evidence of disease response to Ceol CJ, Houvras Y, Jane-Valbuena J, Bilodeau S, Orlando DA, the pan deacetylase inhibitor panobinostat (LBH589) in refractory Battisti V et al. (2011). The histone methyltransferase SETDB1 is Hodgkin lymphoma. Br J Haematol 147: 97–101. recurrently amplified in melanoma and accelerates its onset. Nature Di Croce L, Raker VA, Corsaro M, Fazi F, Fanelli M, Faretta M et al. 471: 513–517. (2002). Methyltransferase recruitment and DNA hypermethylation Chadwick RB, Jiang GL, Bennington GA, Yuan B, Johnson CK, of target promoters by an oncogenic transcription factor. Science Stevens MW et al. (2000). Candidate tumor suppressor RIZ is 295: 1079–1082. frequently involved in colorectal carcinogenesis. Proc Natl Acad Sci Drost J, Mantovani F, Tocco F, Elkon R, Comel A, Holstege H et al. USA 97: 2662–2667. (2010). BRD7 is a candidate tumour suppressor gene required for Chang Y, Zhang X, Horton JR, Upadhyay AK, Spannhoff A, Liu J p53 function. Nat Cell Biol 12: 380–389. et al. (2009). Structural basis for G9a-like protein lysine Eckner R, Ewen ME, Newsome D, Gerdes M, DeCaprio JA, methyltransferase inhibition by BIX-01294. Nat Struct Mol Biol Lawrence JB et al. (1994). Molecular cloning and functional 16: 312–317. analysis of the adenovirus E1A-associated 300-kD protein (p300) Chase A, Cross NC. (2011). Aberrations of EZH2 in cancer. Clin reveals a protein with properties of a transcriptional adaptor. Genes Cancer Res 17: 2613–2618. Dev 8: 869–884. Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash A, Nagy L et al. Ehrbrecht A, Muller U, Wolter M, Hoischen A, Koch A, Radlwimmer (1997). Nuclear receptor coactivator ACTR is a novel histone B et al. (2006). Comprehensive genomic analysis of desmoplastic acetyltransferase and forms a multimeric activation complex with medulloblastomas: identification of novel amplified genes and P/CAF and CBP/p300. Cell 90: 569–580. separate evaluation of the different histological components. Chen MW, Hua KT, Kao HJ, Chi CC, Wei LH, Johansson G et al. J Pathol 208: 554–563. (2010). H3K9 histone methyltransferase G9a promotes lung cancer Eliazer S, Shalaby NA, Buszczak M. (2011). Loss of lysine-specific invasion and metastasis by silencing the cell adhesion molecule demethylase 1 nonautonomously causes stem cell tumors in the Ep-CAM. Cancer Res 70: 7830–7840. Drosophila ovary. Proc Natl Acad Sci USA 108: 7064–7069. Chen SS, Raval A, Johnson AJ, Hertlein E, Liu TH, Jin VX et al. Eliseeva ED, Valkov V, Jung M, Jung MO. (2007). Characterization of (2009). Epigenetic changes during disease progression in a murine novel inhibitors of histone acetyltransferases. Mol Cancer Ther 6: model of human chronic lymphocytic leukemia. Proc Natl Acad Sci 2391–2398. USA 106: 13433–13438. Epping MT, Wang L, Edel MJ, Carlee L, Hernandez M, Bernards R. Cheng JC, Matsen CB, Gonzales FA, Ye W, Greer S, Marquez VE (2005). The human tumor antigen PRAME is a dominant repressor et al. (2003). Inhibition of DNA methylation and reactivation of of retinoic acid receptor signaling. Cell 122: 835–847. silenced genes by zebularine. J Natl Cancer Inst 95: 399–409. Esteller M, Catasus L, Matias-Guiu X, Mutter GL, Prat J, Baylin SB Choi JH, Kwon HJ, Yoon BI, Kim JH, Han SU, Joo HJ et al. (2001). et al. (1999). hMLH1 promoter hypermethylation is an early event Expression profile of histone deacetylase 1 in gastric cancer tissues. in human endometrial tumorigenesis. Am J Pathol 155: 1767–1772. Jpn J Cancer Res 92: 1300–1304. Fang F, Turcan S, Rimner A, Kaufman A, Giri D, Morris LG et al. Choudhary C, Kumar C, Gnad F, Nielsen ML, Rehman M, Walther (2011). Breast cancer methylomes establish an epigenomic founda- TC et al. (2009). Lysine acetylation targets protein complexes and tion for metastasis. Sci Transl Med 3: 75ra25. co-regulates major cellular functions. Science 325: 834–840. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, Santini V, Finelli C, Clapier CR, Cairns BR. (2009). The biology of chromatin remodeling Giagounidis A et al. (2009). Efficacy of azacitidine compared with complexes. Annu Rev Biochem 78: 273–304. that of conventional care regimens in the treatment of higher-risk

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3840 myelodysplastic syndromes: a randomised, open-label, phase III Griffiths DS, Li J, Dawson MA, Trotter MW, Cheng YH, Smith AM study. Lancet Oncol 10: 223–232. et al. (2011). LIF-independent JAK signalling to chromatin in Fenaux P, Mufti GJ, Hellstrom-Lindberg E, Santini V, Gattermann N, embryonic stem cells uncovered from an adult stem cell disease. Nat Germing U et al. (2010). Azacitidine prolongs overall survival Cell Biol 13: 13–21. compared with conventional care regimens in elderly patients with Gui Y, Guo G, Huang Y, Hu X, Tang A, Gao S et al. (2011). Frequent low bone marrow blast count acute myeloid leukemia. J Clin Oncol mutations of chromatin remodeling genes in transitional cell 28: 562–569. carcinoma of the bladder. Nat Genet 43: 875–878. Ferrari R, Pellegrini M, Horwitz GA, Xie W, Berk AJ, Kurdistani SK. Haberland M, Johnson A, Mokalled MH, Montgomery RL, (2008). Epigenetic reprogramming by adenovirus e1a. Science 321: Olson EN. (2009a). Genetic dissection of histone deacetylase require- 1086–1088. ment in tumor cells. Proc Natl Acad Sci USA 106: 7751–7755. Filippakopoulos P, Qi J, Picaud S, Shen Y, Smith WB, Fedorov O Haberland M, Montgomery RL, Olson EN. (2009b). The many roles et al. (2010). Selective inhibition of BET bromodomains. Nature of histone deacetylases in development and physiology: implications 468: 1067–1073. for disease and therapy. Nat Rev Genet 10: 32–42. Fiskus W, Wang Y, Sreekumar A, Buckley KM, Shi H, Jillella A et al. Halkidou K, Gaughan L, Cook S, Leung HY, Neal DE, Robson CN. (2009). Combined epigenetic therapy with the histone methyltrans- (2004). Upregulation and nuclear recruitment of HDAC1 in ferase EZH2 inhibitor 3-deazaneplanocin A and the histone hormone refractory prostate cancer. Prostate 59: 177–189. deacetylase inhibitor panobinostat against human AML cells. Blood Hamada S, Suzuki T, Mino K, Koseki K, Oehme F, Flamme I et al. 114: 2733–2743. (2010). Design, synthesis, enzyme-inhibitory activity, and effect Frescas D, Guardavaccaro D, Kuchay SM, Kato H, Poleshko A, on human cancer cells of a novel series of jumonji domain- Basrur V et al. (2008). KDM2A represses transcription of containing protein 2 histone demethylase inhibitors. J Med Chem centromeric satellite repeats and maintains the heterochromatic 53: 5629–5638. state. Cell Cycle 7: 3539–3547. Hauschild A, Trefzer U, Garbe C, Kaehler KC, Ugurel S, Kiecker F Fujita T, Igarashi J, Okawa ER, Gotoh T, Manne J, Kolla V et al. et al. (2008). Multicenter phase II trial of the histone deacetylase (2008). CHD5, a tumor suppressor gene deleted from 1p36.31 in inhibitor pyridylmethyl-N-{4-[(2-aminophenyl)-carbamoyl]-benzyl}- neuroblastomas. J Natl Cancer Inst 100: 940–949. carbamate in pretreated metastatic melanoma. Melanoma Res 18: Fukuoka J, Fujii T, Shih JH, Dracheva T, Meerzaman D, Player A 274–278. et al. (2004). Chromatin remodeling factors and BRM/BRG1 He L, Yu JX, Liu L, Buyse IM, Wang MS, Yang QC et al. (1998). expression as prognostic indicators in non-small cell lung cancer. RIZ1, but not the alternative RIZ2 product of the same gene, is Clin Cancer Res 10: 4314–4324. underexpressed in breast cancer, and forced RIZ1 expression causes Galanis E, Jaeckle KA, Maurer MJ, Reid JM, Ames MM, Hardwick G2–M cell cycle arrest and/or apoptosis. Cancer Res 58: 4238–4244. JS et al. (2009). Phase II trial of vorinostat in recurrent glioblastoma Henry KW, Wyce A, Lo WS, Duggan LJ, Emre NC, Kao CF et al. multiforme: a north central cancer treatment group study. J Clin (2003). Transcriptional activation via sequential histone H2B Oncol 27: 2052–2058. ubiquitylation and deubiquitylation, mediated by SAGA-associated Garcia-Manero G, Kantarjian HM, Sanchez-Gonzalez B, Yang H, Ubp8. Genes Dev 17: 2648–2663. Rosner G, Verstovsek S et al. (2006). Phase 1/2 study of the Hideshima T, Bradner JE, Wong J, Chauhan D, Richardson P, combination of 5-aza-20-deoxycytidine with valproic acid in patients Schreiber SL et al (2005). Small-molecule inhibition of proteasome with leukemia. Blood 108: 3271–3279. and aggresome function induces synergistic antitumor activity in Gazin C, Wajapeyee N, Gobeil S, Virbasius CM, Green MR. (2007). multiple myeloma. Proc Natl Acad Sci USA 102: 8567–8572. An elaborate pathway required for Ras-mediated epigenetic Huang J, Dorsey J, Chuikov S, Perez-Burgos L, Zhang X, Jenuwein T silencing. Nature 449: 1073–1077. et al. (2010). G9a and Glp methylate lysine 373 in the tumor Geutjes EJ, Tian S, Roepman P, Bernards R. (2011). Deoxycytidine suppressor p53. J Biol Chem 285: 9636–9641. kinase is overexpressed in poor outcome breast cancer and Huang Y, Greene E, Murray Stewart T, Goodwin AC, Baylin SB, determines responsiveness to nucleoside analogs. Breast Cancer Woster PM et al. (2007). Inhibition of lysine-specific demethylase 1 Res Treat (http://www.ncbi.nlm.nih.gov/pubmed/21465168). by polyamine analogues results in reexpression of aberrantly Giaccone G, Rajan A, Berman A, Kelly RJ, Szabo E, Lopez-Chavez A silenced genes. Proc Natl Acad Sci USA 104: 8023–8028. et al. (2011). Phase II study of belinostat in patients with recurrent Huang Y, Stewart TM, Wu Y, Baylin SB, Marton LJ, Perkins B et al. or refractory advanced thymic epithelial tumors. J Clin Oncol 29: (2009). Novel oligoamine analogues inhibit lysine-specific demethy- 2052–2059. lase 1 and induce reexpression of epigenetically silenced genes. Clin Gius D, Cui H, Bradbury CM, Cook J, Smart DK, Zhao S et al. Cancer Res 15: 7217–7228. (2004). Distinct effects on gene expression of chemical and genetic Huang Y, Vasilatos SN, Boric L, Shaw PG, Davidson NE. (2011). manipulation of the cancer epigenome revealed by a multimodality Inhibitors of histone demethylation and histone deacetylation approach. Cancer Cell 6: 361–371. cooperate in regulating gene expression and inhibiting growth in Glaros S, Cirrincione GM, Muchardt C, Kleer CG, Michael CW, human breast cancer cells. Breast Cancer Res Treat (http:// Reisman D. (2007). The reversible epigenetic silencing of BRM: www.ncbi.nlm.nih.gov/pubmed/21452019). implications for clinical targeted therapy. Oncogene 26: 7058–7066. Issa JP, Gharibyan V, Cortes J, Jelinek J, Morris G, Verstovsek S et al. Glaros S, Cirrincione GM, Palanca A, Metzger D, Reisman D. (2008). (2005). Phase II study of low-dose decitabine in patients with Targeted knockout of BRG1 potentiates lung cancer development. chronic myelogenous leukemia resistant to imatinib mesylate. J Clin Cancer Res 68: 3689–3696. Oncol 23: 3948–3956. Gore SD, Baylin S, Sugar E, Carraway H, Miller CB, Carducci M Issa JP, Kantarjian HM. (2009). Targeting DNA methylation. Clin et al. (2006). Combined DNA methyltransferase and histone Cancer Res 15: 3938–3946. deacetylase inhibition in the treatment of myeloid . Ito S, D’Alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y. Cancer Res 66: 6361–6369. (2010). Role of Tet proteins in 5mC to 5hmC conversion, Gorrini C, Squatrito M, Luise C, Syed N, Perna D, Wark L et al. ES-cell self-renewal and inner cell mass specification. Nature 466: (2007). Tip60 is a haplo-insufficient tumour suppressor required 1129–1133. for an oncogene-induced DNA damage response. Nature 448: 1063– Jackson EM, Sievert AJ, Gai X, Hakonarson H, Judkins AR, Tooke L 1067. et al. (2009). Genomic analysis using high-density single nucleotide Grand F, Kulkarni S, Chase A, Goldman JM, Gordon M, Cross NC. polymorphism-based oligonucleotide arrays and multiplex ligation- (1999). Frequent deletion of hSNF5/INI1, a component of the dependent probe amplification provides a comprehensive analysis of SWI/SNF complex, in chronic myeloid leukemia. Cancer Res 59: INI1/SMARCB1 in malignant rhabdoid tumors. Clin Cancer Res 3870–3874. 15: 1923–1930.

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3841 Jacobs JJ, Scheijen B, Voncken JW, Kieboom K, Berns A, van genes in endometrial stromal tumors. Proc Natl Acad Sci USA 98: Lohuizen M. (1999). Bmi-1 collaborates with c-Myc in tumorigenesis 6348–6353. by inhibiting c-Myc-induced apoptosis via INK4a/ARF. Genes Dev Kubicek S, O’Sullivan RJ, August EM, Hickey ER, Zhang Q, 13: 2678–2690. Teodoro ML et al. (2007). Reversal of H3K9me2 by a small- Jaju RJ, Fidler C, Haas OA, Strickson AJ, Watkins F, Clark K et al. molecule inhibitor for the G9a histone methyltransferase. Mol Cell (2001). A novel gene, NSD1, is fused to NUP98 in the 25: 473–481. t(5;11)(q35;p15.5) in de novo childhood acute myeloid leukemia. Kuck D, Caulfield T, Lyko F, Medina-Franco JL. (2010). Blood 98: 1264–1267. Nanaomycin A selectively inhibits DNMT3B and reactivates Jiang X, Tan J, Li J, Kivimae S, Yang X, Zhuang L et al. (2008). silenced tumor suppressor genes in human cancer cells. Mol Cancer DACT3 is an epigenetic regulator of Wnt/beta-catenin signaling in Ther 9: 3015–3023. colorectal cancer and is a therapeutic target of histone modifica- Kung AL, Rebel VI, Bronson RT, Ch’ng LE, Sieff CA, Livingston tions. Cancer Cell 13: 529–541. DM et al. (2000). Gene dose-dependent control of hematopoiesis Jin KL, Pak JH, Park JY, Choi WH, Lee JY, Kim JH et al. (2008). and hematologic tumor suppression by CBP. Genes Dev 14: Expression profile of histone deacetylases 1, 2 and 3 in ovarian 272–277. cancer tissues. J Gynecol Oncol 19: 185–190. Kurotaki N, Imaizumi K, Harada N, Masuno M, Kondoh T, Nagai T Jones S, Wang TL, Shih Ie M, Mao TL, Nakayama K, Roden R et al. et al. (2002). Haploinsufficiency of NSD1 causes Sotos syndrome. (2010). Frequent mutations of chromatin remodeling gene ARID1A Nat Genet 30: 365–366. in ovarian clear cell carcinoma. Science 330: 228–231. Lahusen T, Henke RT, Kagan BL, Wellstein A, Riegel AT. (2009). Joo HY, Zhai L, Yang C, Nie S, Erdjument-Bromage H, Tempst P The role and regulation of the nuclear receptor co-activator AIB1 in et al. (2007). Regulation of cell cycle progression and gene breast cancer. Breast Cancer Res Treat 116: 225–237. expression by H2A deubiquitination. Nature 449: 1068–1072. Lane AA, Chabner BA. (2009). Histone deacetylase inhibitors in Kahl P, Gullotti L, Heukamp LC, Wolf S, Friedrichs N, Vorreuther R cancer therapy. J Clin Oncol 27: 5459–5468. et al. (2006). Androgen receptor coactivators lysine-specific histone Langemeijer SM, Kuiper RP, Berends M, Knops R, Aslanyan MG, demethylase 1 and four and a half LIM domain protein 2 predict Massop M et al. (2009). Acquired mutations in TET2 are common risk of prostate cancer recurrence. Cancer Res 66: 11341–11347. in myelodysplastic syndromes. Nat Genet 41: 838–842. Kaminskas E, Farrell A, Abraham S, Baird A, Hsieh LS, Lee SL et al. Lavau C, Du C, Thirman M, Zeleznik-Le N. (2000). Chromatin- (2005). Approval summary: azacitidine for treatment of myelodys- related properties of CBP fused to MLL generate a myelodysplastic- plastic syndrome subtypes. Clin Cancer Res 11: 3604–3608. like syndrome that evolves into myeloid leukemia. EMBO J 19: Kantarjian H, Issa JP, Rosenfeld CS, Bennett JM, Albitar M, 4655–4664. DiPersio J et al. (2006). Decitabine improves patient outcomes in Leary RJ, Lin JC, Cummins J, Boca S, Wood LD, Parsons DW et al. myelodysplastic syndromes: results of a phase III randomized study. (2008). Integrated analysis of homozygous deletions, focal amplifi- Cancer 106: 1794–1803. cations, and sequence alterations in breast and colorectal cancers. Kelly TK, De Carvalho DD, Jones PA. (2010). Epigenetic modifica- Proc Natl Acad Sci USA 105: 16224–16229. tions as therapeutic targets. Nat Biotechnol 28: 1069–1078. Lee YS, Lim KH, Guo X, Kawaguchi Y, Gao Y, Barrientos T et al. Kilpivaara O, Levine RL. (2008). JAK2 and MPL mutations in (2008). The cytoplasmic deacetylase HDAC6 is required for efficient myeloproliferative neoplasms: discovery and science. Leukemia 22: oncogenic tumorigenesis. Cancer Res 68: 7561–7569. 1813–1817. Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE Kirschbaum M, Frankel P, Popplewell L, Zain J, Delioukina M, et al. (2010). DNMT3A mutations in acute myeloid leukemia. N Pullarkat V et al. (2011). Phase II study of vorinostat for treatment Engl J Med 363: 2424–2433. of relapsed or refractory indolent non-Hodgkin’s lymphoma and Li M, Zhao H, Zhang X, Wood LD, Anders RA, Choti MA et al. mantle cell lymphoma. J Clin Oncol 29: 1198–1203. (2011). Inactivating mutations of the chromatin remodeling gene Kitabayashi I, Aikawa Y, Nguyen LA, Yokoyama A, Ohki M. ARID2 in hepatocellular carcinoma. Nat Genet 43: 828–829. (2001a). Activation of AML1-mediated transcription by MOZ and Lim S, Janzer A, Becker A, Zimmer A, Schule R, Buettner R et al. inhibition by the MOZ-CBP fusion protein. EMBO J 20: 7184–7196. (2010). Lysine-specific demethylase 1 (LSD1) is highly expressed in Kitabayashi I, Aikawa Y, Yokoyama A, Hosoda F, Nagai M, Kakazu ER-negative breast cancers and a biomarker predicting aggressive N et al. (2001b). Fusion of MOZ and p300 histone acetyltrans- biology. Carcinogenesis 31: 512–520. ferases in acute monocytic leukemia with a t(8;22)(p11;q13) Liu G, Bollig-Fischer A, Kreike B, van de Vijver MJ, Abrams J, Ethier chromosome translocation. Leukemia 15: 89–94. SP et al. (2009a). Genomic amplification and oncogenic properties Kleer CG, Cao Q, Varambally S, Shen R, Ota I, Tomlins SA et al. of the GASC1 histone demethylase gene in breast cancer. Oncogene (2003). EZH2 is a marker of aggressive breast cancer and promotes 28: 4491–4500. neoplastic transformation of breast epithelial cells. Proc Natl Acad Liu H, Cheng EH, Hsieh JJ. (2009b). MLL fusions: pathways to Sci USA 100: 11606–11611. leukemia. Cancer Biol Ther 8: 1204–1211. Klochendler-Yeivin A, Fiette L, Barra J, Muchardt C, Babinet C, Lu PJ, Sundquist K, Baeckstrom D, Poulsom R, Hanby A, Meier- Yaniv M. (2000). The murine SNF5/INI1 chromatin remodeling Ewert S et al. (1999). A novel gene (PLU-1) containing highly factor is essential for embryonic development and tumor suppres- conserved putative DNA/chromatin binding motifs is specifically sion. EMBO Rep 1: 500–506. upregulated in breast cancer. J Biol Chem 274: 15633–15645. Knapper S, Burnett AK, Littlewood T, Kell WJ, Agrawal S, Chopra R Luu TH, Morgan RJ, Leong L, Lim D, McNamara M, Portnow J et al. (2006). A phase 2 trial of the FLT3 inhibitor lestaurtinib et al. (2008). A phase II trial of vorinostat (suberoylanilide (CEP701) as first-line treatment for older patients with acute hydroxamic acid) in metastatic breast cancer: a California Cancer myeloid leukemia not considered fit for intensive chemotherapy. Consortium study. Clin Cancer Res 14: 7138–7142. Blood 108: 3262–3270. Mann BS, Johnson JR, Cohen MH, Justice R, Pazdur R. (2007). FDA Ko M, Huang Y, Jankowska AM, Pape UJ, Tahiliani M, Bandukwala approval summary: vorinostat for treatment of advanced primary HS et al. (2010). Impaired hydroxylation of 5-methylcytosine in cutaneous T-cell lymphoma. Oncologist 12: 1247–1252. myeloid cancers with mutant TET2. Nature 468: 839–843. Marschalek R. (2011). Mechanisms of leukemogenesis by MLL fusion Kondo Y, Shen L, Ahmed S, Boumber Y, Sekido Y, Haddad BR et al. proteins. Br J Haematol 152: 141–154. (2008). Downregulation of histone H3 lysine 9 methyltransferase Matsukawa Y, Semba S, Kato H, Ito A, Yanagihara K, Yokozaki H. G9a induces centrosome disruption and chromosome instability in (2006). Expression of the enhancer of zeste homolog 2 is correlated cancer cells. PLoS One 3: e2037. with poor prognosis in human gastric cancer. Cancer Sci 97: 484–491. Koontz JI, Soreng AL, Nucci M, Kuo FC, Pauwels P, van Den McGarvey KM, Van Neste L, Cope L, Ohm JE, Herman JG, Van Berghe H et al. (2001). Frequent fusion of the JAZF1 and JJAZ1 Criekinge W et al. (2008). Defining a chromatin pattern that

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3842 characterizes DNA-hypermethylated genes in colon cancer cells. Noushmehr H, Weisenberger DJ, Diefes K, Phillips HS, Pujara K, Cancer Res 68: 5753–5759. Berman BP et al. (2010). Identification of a CpG island methylator Medina PP, Carretero J, Fraga MF, Esteller M, Sidransky D, Sanchez- phenotype that defines a distinct subgroup of glioma. Cancer Cell Cespedes M. (2004). Genetic and epigenetic screening for gene 17: 510–522. alterations of the chromatin-remodeling factor, SMARCA4/BRG1, Nowell PC. (2002). Tumor progression: a brief historical perspective. in lung tumors. Genes Chromosomes Cancer 41: 170–177. Semin Cancer Biol 12: 261–266. Mesa RA, Kantarjian H, Tefferi A, Dueck A, Levy R, Vaddi K et al. Oehme I, Deubzer HE, Wegener D, Pickert D, Linke JP, Hero B et al. (2011). Evaluating the serial use of the myelofibrosis symptom (2009). Histone deacetylase 8 in neuroblastoma tumorigenesis. Clin assessment form for measuring symptomatic improvement: perfor- Cancer Res 15: 91–99. mance in 87 myelofibrosis patients on a JAK1 and JAK2 inhibitor Oki Y, Jelinek J, Shen L, Kantarjian HM, Issa JP. (2008). Induction of (INCB018424) clinical trial. Cancer 117: 4869–4877. hypomethylation and molecular response after decitabine therapy in Metzger E, Wissmann M, Yin N, Muller JM, Schneider R, Peters AH patients with chronic myelomonocytic leukemia. Blood 111: et al. (2005). LSD1 demethylates repressive histone marks to 2382–2384. promote androgen-receptor-dependent transcription. Nature 437: Otterson GA, Hodgson L, Pang H, Vokes EE. (2010). Phase II study 436–439. of the histone deacetylase inhibitor Romidepsin in relapsed small Miller CT, Moy JR, Lin L, Schipper M, Normolle D, Brenner DE et al. cell lung cancer (Cancer and Leukemia Group B 30304). J Thorac (2003). Gene amplification in esophageal adenocarcinomas and Oncol 5: 1644–1648. Barrett’s with high-grade dysplasia. Clin Cancer Res 9: 4819–4825. Palakurthy RK, Wajapeyee N, Santra MK, Gazin C, Lin L, Gobeil S Miranda TB, Cortez CC, Yoo CB, Liang G, Abe M, Kelly TK et al. et al. (2009). Epigenetic silencing of the RASSF1A tumor (2009). DZNep is a global histone methylation inhibitor that suppressor gene through HOXB3-mediated induction of DNMT3B reactivates developmental genes not silenced by DNA methylation. expression. Mol Cell 36: 219–230. Mol Cancer Ther 8: 1579–1588. Pardanani A, Gotlib JR, Jamieson C, Cortes JE, Talpaz M, Stone RM Modesitt SC, Sill M, Hoffman JS, Bender DP. (2008). A phase II study et al. (2011). Safety and efficacy of TG101348, a selective JAK2 of vorinostat in the treatment of persistent or recurrent epithelial inhibitor, in myelofibrosis. J Clin Oncol 29: 789–796. ovarian or primary peritoneal carcinoma: a Gynecologic Oncology Parsons DW, Li M, Zhang X, Jones S, Leary RJ, Lin JC et al. (2011). Group study. Gynecol Oncol 109: 182–186. The genetic landscape of the childhood cancer medulloblastoma. Molife LR, Attard G, Fong PC, Karavasilis V, Reid AH, Patterson S Science 331: 435–439. et al. (2010). Phase II, two-stage, single-arm trial of the histone Pasqualucci L, Dominguez-Sola D, Chiarenza A, Fabbri G, Grunn A, deacetylase inhibitor (HDACi) romidepsin in metastatic castration- Trifonov V et al. (2011). Inactivating mutations of acetyltransferase resistant prostate cancer (CRPC). Ann Oncol 21: 109–113. genes in B-cell lymphoma. Nature 471: 189–195. Moran-Crusio K, Reavie L, Shih A, Abdel-Wahab O, Ndiaye-Lobry Pedersen MT, Helin K. (2010). Histone demethylases in development D, Lobry C et al. (2011). Tet2 loss leads to increased hematopoietic and disease. Trends Cell Biol 20: 662–671. stem cell self-renewal and myeloid transformation. Cancer Cell 20: Petrij F, Giles RH, Dauwerse HG, Saris JJ, Hennekam RC, Masuno 11–24. M et al. (1995). Rubinstein–Taybi syndrome caused by mutations in Morin RD, Johnson NA, Severson TM, Mungall AJ, An J, Goya R the transcriptional co-activator CBP. Nature 376: 348–351. et al. (2010). Somatic mutations altering EZH2 (Tyr641) in follicular Piekarz RL, Bates SE. (2009). Epigenetic modifiers: basic under- and diffuse large B-cell lymphomas of germinal-center origin. standing and clinical development. Clin Cancer Res 15: 3918–3926. Nat Genet 42: 181–185. Piekarz RL, Frye R, Turner M, Wright JJ, Allen SL, Kirschbaum MH Morin RD, Mendez-Lago M, Mungall AJ, Goya R, Mungall KL, et al. (2009). Phase II multi-institutional trial of the histone Corbett RD et al. (2011). Frequent mutation of histone-modifying deacetylase inhibitor romidepsin as monotherapy for patients with genes in non-Hodgkin lymphoma. Nature 476: 298–303. cutaneous T-cell lymphoma. J Clin Oncol 27: 5410–5417. Mulero-Navarro S, Esteller M. (2008). Chromatin remodeling factor Quintas-Cardama A, Santos FP, Garcia-Manero G. (2010). Therapy CHD5 is silenced by promoter CpG island hypermethylation in with azanucleosides for myelodysplastic syndromes. Nat Rev Clin human cancer. Epigenetics 3: 210–215. Oncol 7: 433–444. Mullighan CG, Zhang J, Kasper LH, Lerach S, Payne-Turner D, Quintas-Cardama A, Tong W, Kantarjian H, Thomas D, Ravandi F, Phillips LA et al. (2011). CREBBP mutations in relapsed acute Kornblau S et al. (2008). A phase II study of 5-azacitidine for lymphoblastic leukaemia. Nature 471: 235–239. patients with primary and post-essential thrombocythemia/ Namdar M, Perez G, Ngo L, Marks PA. (2010). Selective inhibition of polycythemia vera myelofibrosis. Leukemia 22: 965–970. histone deacetylase 6 (HDAC6) induces DNA damage and Quivoron C, Couronne L, Della Valle V, Lopez CK, Plo I, sensitizes transformed cells to anticancer agents. Proc Natl Acad Wagner-Ballon O et al. (2011). TET2 inactivation results in Sci USA 107: 20003–20008. pleiotropic hematopoietic abnormalities in mouse and is a recurrent Nguyen AT, Taranova O, He J, Zhang Y. (2011). DOT1L, the H3K79 event during human lymphomagenesis. Cancer Cell 20: 25–38. methyltransferase, is required for MLL-AF9-mediated leukemogen- Rahman N. (2005). Mechanisms predisposing to childhood esis. Blood 117: 6912–6922. overgrowth and cancer. Curr Opin Genet Dev 15: 227–233. Nicassio F, Corrado N, Vissers JH, Areces LB, Bergink S, Marteijn JA Ramalingam SS, Belani CP, Ruel C, Frankel P, Gitlitz B, Koczywas M et al. (2007). Human USP3 is a chromatin modifier required et al. (2009). Phase II study of belinostat (PXD101), a histone for S phase progression and genome stability. Curr Biol 17: deacetylase inhibitor, for second line therapy of advanced malignant 1972–1977. pleural mesothelioma. J Thorac Oncol 4: 97–101. Niesvizky R, Ely S, Mark T, Aggarwal S, Gabrilove JL, Wright JJ Raman JD, Mongan NP, Tickoo SK, Boorjian SA, Scherr DS, Gudas et al. (2011). Phase 2 trial of the histone deacetylase inhibitor LJ. (2005). Increased expression of the polycomb group gene, romidepsin for the treatment of refractory multiple myeloma. EZH2, in transitional cell carcinoma of the bladder. Clin Cancer Res Cancer 117: 336–342. 11: 8570–8576. Niu X, Zhang T, Liao L, Zhou L, Lindner DJ, Zhou M et al. (2011). Rambaldi A, Dellacasa CM, Finazzi G, Carobbio A, Ferrari ML, The von Hippel–Lindau tumor suppressor protein regulates gene Guglielmelli P et al. (2010). A pilot study of the histone-deacetylase expression and tumor growth through histone demethylase JAR- inhibitor Givinostat in patients with JAK2V617F positive chronic ID1C. Oncogene 31: 776–786. myeloproliferative neoplasms. Br J Haematol 150: 446–455. Northcott PA, Nakahara Y, Wu X, Feuk L, Ellison DW, Croul S et al. Reisman D, Glaros S, Thompson EA. (2009). The SWI/SNF complex (2009). Multiple recurrent genetic events converge on control and cancer. Oncogene 28: 1653–1668. of histone lysine methylation in medulloblastoma. Nat Genet 41: Reisman DN, Sciarrotta J, Wang W, Funkhouser WK, Weissman BE. 465–472. (2003). Loss of BRG1/BRM in human lung cancer cell lines and

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3843 primary lung cancers: correlation with poor prognosis. Cancer Res in patients with advanced solid tumors and non-Hodgkin’s 63: 560–566. lymphomas. Clin Cancer Res 17: 1582–1590. Roberts CW, Galusha SA, McMenamin ME, Fletcher CD, Orkin SH. Steele-Perkins G, Fang W, Yang XH, Van Gele M, Carling T, Gu J (2000). Haploinsufficiency of Snf5 (integrase interactor 1) et al. (2001). Tumor formation and inactivation of RIZ1, an Rb- predisposes to malignant rhabdoid tumors in mice. Proc Natl Acad binding member of a nuclear protein-methyltransferase superfamily. Sci USA 97: 13796–13800. Genes Dev 15: 2250–2262. Roberts CW, Leroux MM, Fleming MD, Orkin SH. (2002). Highly Steine EJ, Ehrich M, Bell GW, Raj A, Reddy S, van Oudenaarden A penetrant, rapid tumorigenesis through conditional inversion of the et al. (2011). Genes methylated by DNA methyltransferase 3b are tumor suppressor gene Snf5. Cancer Cell 2: 415–425. similar in mouse intestine and human colon cancer. J Clin Invest Rodriguez-Nieto S, Canada A, Pros E, Pinto AI, Torres-Lanzas J, 121: 1748–1752. Lopez-Rios F et al. (2010). Massive parallel DNA pyrosequencing Sudo T, Utsunomiya T, Mimori K, Nagahara H, Ogawa K, Inoue H analysis of the tumor suppressor BRG1/SMARCA4 in lung primary et al. (2005). Clinicopathological significance of EZH2 mRNA tumors. Hum Mutat 32: E1999–E2017. expression in patients with hepatocellular carcinoma. Br J Cancer Roesch A, Becker B, Meyer S, Wild P, Hafner C, Landthaler M et al. 92: 1754–1758. (2005). Retinoblastoma-binding protein 2-homolog 1: a retinoblas- Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y toma-binding protein downregulated in malignant . Mod et al. (2009). Conversion of 5-methylcytosine to 5-hydroxymethyl- Pathol 18: 1249–1257. cytosine in mammalian DNA by MLL partner TET1. Science 324: Rousseau-Merck MF, Versteege I, Legrand I, Couturier J, Mairal A, 930–935. Delattre O et al. (1999). hSNF5/INI1 inactivation is mainly Tan J, Yang X, Zhuang L, Jiang X, Chen W, Lee PL et al. (2007). associated with homozygous deletions and mitotic recombinations Pharmacologic disruption of Polycomb-repressive complex 2- in rhabdoid tumors. Cancer Res 59: 3152–3156. mediated gene repression selectively induces apoptosis in cancer Rubio-Moscardo F, Climent J, Siebert R, Piris MA, Martin-Subero JI, cells. Genes Dev 21: 1050–1063. Nielander I et al. (2005). Mantle-cell lymphoma genotypes identified Ting AH, McGarvey KM, Baylin SB. (2006). The cancer epigenome— with CGH to BAC microarrays define a leukemic subgroup of components and functional correlates. Genes Dev 20: 3215–3231. disease and predict patient outcome. Blood 105: 4445–4454. Toyota M, Ahuja N, Ohe-Toyota M, Herman JG, Baylin SB, Issa JP. Rui L, Emre NC, Kruhlak MJ, Chung HJ, Steidl C, Slack G et al. (1999). CpG island methylator phenotype in colorectal cancer. Proc (2010). Cooperative epigenetic modulation by cancer amplicon Natl Acad Sci USA 96: 8681–8686. genes. Cancer Cell 18: 590–605. Traynor AM, Dubey S, Eickhoff JC, Kolesar JM, Schell K, Huie MS Saji S, Kawakami M, Hayashi S, Yoshida N, Hirose M, Horiguchi S et al. (2009). Vorinostat (NSC# 701852) in patients with relapsed et al. (2005). Significance of HDAC6 regulation via estrogen non-small cell lung cancer: a Wisconsin Oncology Network phase II signaling for cell motility and prognosis in estrogen receptor- study. J Thorac Oncol 4: 522–526. positive breast cancer. Oncogene 24: 4531–4539. van der Knaap JA, Kozhevnikova E, Langenberg K, Moshkin YM, Sakakura C, Hagiwara A, Yasuoka R, Fujita Y, Nakanishi M, Verrijzer CP. (2010). Biosynthetic enzyme GMP synthetase co- Masuda K et al. (2000). Amplification and overexpression of the operates with ubiquitin-specific protease 7 in transcriptional AIB1 nuclear receptor co-activator gene in primary gastric cancers. regulation of ecdysteroid target genes. Mol Cell Biol 30: Int J Cancer 89: 217–223. 736–744. Santos FP, Kantarjian HM, Jain N, Manshouri T, Thomas DA, van der Knaap JA, Kumar BR, Moshkin YM, Langenberg K, Garcia-Manero G et al. (2010). Phase 2 study of CEP-701, an orally Krijgsveld J, Heck AJ et al. (2005). GMP synthetase stimulates available JAK2 inhibitor, in patients with primary or post- histone H2B deubiquitylation by the epigenetic silencer USP7. Mol polycythemia vera/essential thrombocythemia myelofibrosis. Blood Cell 17: 695–707. 115: 1131–1136. van der Vlag J, Otte AP. (1999). Transcriptional repression mediated Schaefer EW, Loaiza-Bonilla A, Juckett M, DiPersio JF, Roy V, Slack by the human polycomb-group protein EED involves histone J et al. (2009). A phase 2 study of vorinostat in acute myeloid deacetylation. Nat Genet 23: 474–478. leukemia. Haematologica 94: 1375–1382. van Haaften G, Dalgliesh GL, Davies H, Chen L, Bignell G, Schrump DS, Fischette MR, Nguyen DM, Zhao M, Li X, Kunst TF Greenman C et al. (2009). Somatic mutations of the histone et al. (2008). Clinical and molecular responses in lung cancer H3K27 demethylase gene UTX in human cancer. Nat Genet 41: patients receiving Romidepsin. Clin Cancer Res 14: 188–198. 521–523. Schuebel KE, Chen W, Cope L, Glockner SC, Suzuki H, Yi JM et al. van Lohuizen M, Verbeek S, Scheijen B, Wientjens E, van der (2007). Comparing the DNA hypermethylome with gene mutations Gulden H, Berns A. (1991). Identification of cooperating in human colorectal cancer. PLoS Genet 3: 1709–1723. oncogenes in E mu-myc transgenic mice by provirus tagging. Cell Schulte JH, Lim S, Schramm A, Friedrichs N, Koster J, Versteeg R 65: 737–752. et al. (2009). Lysine-specific demethylase 1 is strongly expressed in Vansteenkiste J, Van Cutsem E, Dumez H, Chen C, Ricker JL, poorly differentiated neuroblastoma: implications for therapy. Randolph SS et al. (2008). Early phase II trial of oral vorinostat in Cancer Res 69: 2065–2071. relapsed or refractory breast, colorectal, or non-small cell lung Sevenet N, Sheridan E, Amram D, Schneider P, Handgretinger R, cancer. Invest New Drugs 26: 483–488. Delattre O. (1999). Constitutional mutations of the hSNF5/INI1 Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, gene predispose to a variety of cancers. Am J Hum Genet 65: Kumar-Sinha C, Sanda MG et al. (2002). The polycomb group 1342–1348. protein EZH2 is involved in progression of prostate cancer. Nature Song J, Noh JH, Lee JH, Eun JW, Ahn YM, Kim SY et al. (2005). 419: 624–629. Increased expression of histone deacetylase 2 is found in human Varela I, Tarpey P, Raine K, Huang D, Ong CK, Stephens P et al. gastric cancer. Apmis 113: 264–268. (2011). Exome sequencing identifies frequent mutation of the Sonpavde G, Aparicio A, Guttierez I, Boehm KA, Hutson TE, Berry SWI/SNF complex gene PBRM1 in renal carcinoma. Nature 469: WR et al. (2007). Phase II study of azacitidine to restore 539–542. responsiveness of prostate cancer to hormonal therapy. Clin Vedadi M, Barsyte-Lovejoy D, Liu F, Rival-Gervier S, Genitourin Cancer 5: 457–459. Allali-Hassani A, Labrie V et al. (2011). A chemical probe Stadler WM, Margolin K, Ferber S, McCulloch W, Thompson JA. selectively inhibits G9a and GLP methyltransferase activity in cells. (2006). A phase II study of depsipeptide in refractory metastatic Nat Chem Biol 7: 566–574. renal cell cancer. Clin Genitourin Cancer 5: 57–60. Versteege I, Sevenet N, Lange J, Rousseau-Merck MF, Ambros P, Stathis A, Hotte SJ, Chen EX, Hirte HW, Oza AM, Moretto P et al. Handgretinger R et al. (1998). Truncating mutations of hSNF5/ (2011). Phase I study of decitabine in combination with vorinostat INI1 in aggressive paediatric cancer. Nature 394: 203–206.

Oncogene Targeting the epigenome for cancer treatment E-J Geutjes et al 3844 Verstovsek S, Kantarjian H, Mesa RA, Pardanani AD, Cortes-Franco Wilson AJ, Byun DS, Popova N, Murray LB, L’Italien K, Sowa Y J, Thomas DA et al. (2010). Safety and efficacy of INCB018424, a et al. (2006). Histone deacetylase 3 (HDAC3) and other class I JAK1 and JAK2 inhibitor, in myelofibrosis. N Engl J Med 363: HDACs regulate colon cell maturation and p21 expression and are 1117–1127. deregulated in human colon cancer. J Biol Chem 281: 13548–13558. Vire E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C et al. Wilson BG, Roberts CW. (2011). SWI/SNF nucleosome remodellers (2006). The Polycomb group protein EZH2 directly controls DNA and cancer. Nat Rev Cancer 11: 481–492. methylation. Nature 439: 871–874. Wissmann M, Yin N, Muller JM, Greschik H, Fodor BD, Jenuwein T Voso MT, Santini V, Finelli C, Musto P, Pogliani E, Angelucci E et al. et al. (2007). Cooperative demethylation by JMJD2C and LSD1 (2009). Valproic acid at therapeutic plasma levels may increase promotes androgen receptor-dependent gene expression. Nat Cell 5-azacytidine efficacy in higher risk myelodysplastic syndromes. Clin Biol 9: 347–353. Cancer Res 15: 5002–5007. Woyach JA, Kloos RT, Ringel MD, Arbogast D, Collamore M, Wagner JM, Hackanson B, Lubbert M, Jung M. (2010). Histone Zwiebel JA et al. (2009). Lack of therapeutic effect of the histone deacetylase (HDAC) inhibitors in recent clinical trials for cancer deacetylase inhibitor vorinostat in patients with metastatic radio- therapy. Clin Epigenet 1: 117–136. iodine-refractory thyroid carcinoma. J Clin Endocrinol Metab 94: Wang GG, Song J, Wang Z, Dormann HL, Casadio F, Li H et al. 164–170. (2009a). Haematopoietic malignancies caused by dysregulation of a Wozniak RJ, Klimecki WT, Lau SS, Feinstein Y, Futscher BW. chromatin-binding PHD finger. Nature 459: 847–851. (2007). 5-Aza-20-deoxycytidine-mediated reductions in G9A histone Wang J, Hevi S, Kurash JK, Lei H, Gay F, Bajko J et al. (2009b). The methyltransferase and histone H3 K9 di-methylation levels are lysine demethylase LSD1 (KDM1) is required for maintenance of linked to tumor suppressor gene reactivation. Oncogene 26: 77–90. global DNA methylation. Nat Genet 41: 125–129. Xiang Y, Zhu Z, Han G, Ye X, Xu B, Peng Z et al. (2007). JARID1B Wang JK, Tsai MC, Poulin G, Adler AS, Chen S, Liu H et al. The is a histone H3 lysine 4 demethylase upregulated in prostate cancer. histone demethylase UTX enables RB-dependent cell fate control. Proc Natl Acad Sci USA 104: 19226–19231. Genes Dev, (2010) 24: 327–332. Yamamichi N, Yamamichi-Nishina M, Mizutani T, Watanabe H, Wang L, Gural A, Sun XJ, Zhao X, Perna F, Huang G et al. (2011). Minoguchi S, Kobayashi N et al. (2005). The Brm gene suppressed The leukemogenicity of AML1-ETO is dependent on site-specific at the post-transcriptional level in various human cell lines is lysine acetylation. Science 333: 765–769. inducible by transient HDAC inhibitor treatment, which exhibits Wang Y, Zhang H, Chen Y, Sun Y, Yang F, Yu W et al. (2009c). antioncogenic potential. Oncogene 24: 5471–5481. LSD1 is a subunit of the NuRD complex and targets the metastasis Yan XJ, Xu J, Gu ZH, Pan CM, Lu G, Shen Y et al. (2011). programs in breast cancer. Cell 138: 660–672. Exome sequencing identifies somatic mutations of DNA methyl- Wartman LD, Larson DE, Xiang Z, Ding L, Chen K, Lin L et al. transferase gene DNMT3A in acute monocytic leukemia. Nat Genet (2011). Sequencing a mouse acute promyelocytic leukemia genome 43: 309–315. reveals genetic events relevant for disease progression. J Clin Invest Yang ZQ, Imoto I, Fukuda Y, Pimkhaokham A, Shimada Y, 121: 1445–1455. Imamura M et al. (2000). Identification of a novel gene, GASC1, Wei X, Walia V, Lin JC, Teer JK, Prickett TD, Gartner J et al. (2011). within an amplicon at 9p23-24 frequently detected in esophageal Exome sequencing identifies GRIN2A as frequently mutated in cancer cell lines. Cancer Res 60: 4735–4739. melanoma. Nat Genet 43: 442–446. Yap DB, Chu J, Berg T, Schapira M, Cheng SW, Moradian A et al. Weichert W. (2009). HDAC expression and clinical prognosis in (2011). Somatic mutations at EZH2 Y641 act dominantly through a human malignancies. Cancer Lett 280: 168–176. mechanism of selectively altered PRC2 catalytic activity, to increase Weichert W, Roske A, Gekeler V, Beckers T, Ebert MP, Pross M et al. H3K27 trimethylation. Blood 117: 2451–2459. (2008a). Association of patterns of class I histone deacetylase Yoo CB, Jeong S, Egger G, Liang G, Phiasivongsa P, Tang C et al. expression with patient prognosis in gastric cancer: a retrospective (2007). Delivery of 5-aza-20-deoxycytidine to cells using oligodeox- analysis. Lancet Oncol 9: 139–148. ynucleotides. Cancer Res 67: 6400–6408. Weichert W, Roske A, Niesporek S, Noske A, Buckendahl AC, Yuge M, Nagai H, Uchida T, Murate T, Hayashi Y, Hotta T et al. Dietel M et al. (2008b). Class I histone deacetylase expression has (2000). HSNF5/INI1 gene mutations in lymphoid malignancy. independent prognostic impact in human colorectal cancer: specific Cancer Genet Cytogenet 122: 37–42. role of class I histone deacetylases in vitro and in vivo. Clin Cancer Zeng J, Ge Z, Wang L, Li Q, Wang N, Bjorkholm M et al. (2010). The Res 14: 1669–1677. histone demethylase RBP2 Is overexpressed in gastric cancer and its Weikert S, Christoph F, Kollermann J, Muller M, Schrader M, inhibition triggers senescence of cancer cells. Gastroenterology 138: Miller K et al. (2005). Expression levels of the EZH2 polycomb 981–992. transcriptional repressor correlate with aggressiveness and Zhang Z, Yamashita H, Toyama T, Sugiura H, Ando Y, Mita K et al. invasive potential of bladder carcinomas. Int J Mol Med 16: (2005). Quantitation of HDAC1 mRNA expression in invasive 349–353. carcinoma of the breast*. Breast Cancer Res Treat 94: 11–16. Whitehead RP, Rankin C, Hoff PM, Gold PJ, Billingsley KG, Zhu P, Zhou W, Wang J, Puc J, Ohgi KA, Erdjument-Bromage H Chapman RA et al. (2009). Phase II trial of romidepsin (NSC- et al. (2007). A histone H2A deubiquitinase complex coordinating 630176) in previously treated colorectal cancer patients with histone acetylation and H1 dissociation in transcriptional advanced disease: a Southwest Oncology Group study (S0336). regulation. Mol Cell 27: 609–621. Invest New Drugs 27: 469–475. Zhu Q, Pao GM, Huynh AM, Suh H, Tonnu N, Nederlof PM et al. Whittaker SJ, Demierre MF, Kim EJ, Rook AH, Lerner A, Duvic M (2011). BRCA1 tumour suppression occurs via heterochromatin- et al. (2010). Final results from a multicenter, international, pivotal mediated silencing. Nature 477: 179–184. study of romidepsin in refractory cutaneous T-cell lymphoma. Zhu WG, Lakshmanan RR, Beal MD, Otterson GA. (2001). DNA J Clin Oncol 28: 4485–4491. methyltransferase inhibition enhances apoptosis induced by histone Wiegand KC, Shah SP, Al-Agha OM, Zhao Y, Tse K, Zeng T et al. deacetylase inhibitors. Cancer Res 61: 1327–1333. (2010). ARID1A mutations in endometriosis-associated ovarian Zimmermann S, Kiefer F, Prudenziati M, Spiller C, Hansen J, Floss T carcinomas. N Engl J Med 363: 1532–1543. et al. (2007). Reduced body size and decreased intestinal tumor rates Wijermans PW, Ruter B, Baer MR, Slack JL, Saba HI, Lubbert M. in HDAC2-mutant mice. Cancer Res 67: 9047–9054. (2008). Efficacy of decitabine in the treatment of patients Zuber J, Shi J, Wang E, Rappaport AR, Herrmann H, Sison EA et al. with chronic myelomonocytic leukemia (CMML). Leukemia Res (2011). RNAi screen identifies Brd4 as a therapeutic target in acute 32: 587–591. myeloid leukaemia. Nature 478: 524–528.

Oncogene