Targeting histone lysine demethylases — Progress, challenges, and the future Academic research paper on "Chemical sciences"

ARTICLE IN PRESS

BBAGRM-00731 ; No. of pages: 17; 4C: 2,4, 7, 9,10,11

Biochimica et Biophysica Acta xxx (2014) xxx-xxx

Review

Targeting histone lysine demethylases — Progress, challenges, and the future

Cyrille C. Thinnes1, Katherine S. England 1, Akane Kawamura, Rasheduzzaman Chowdhury, Christopher J. Schofield *, Richard J. Hopkinson *

The Chemistry Research Laboratory, Mansfield Road, Oxford, OX1 3TA, UK

ARTICLE INFO

ABSTRACT

Article history: Received 11 January 2014 Received in revised form 6 May 2014 Accepted 13 May 2014 Available online xxxx

Keywords:

Demethylase

Epigenetics

Methylation

Inhibition

Histone

Lysine

N-Methylation of lysine and arginine residues has emerged as a major mechanism of transcriptional regulation in eukaryotes. In humans, N^methyllysine residue demethylation is catalysed by two distinct subfamilies of demethylases (KDMs), the flavin-dependent KDM1 subfamily and the 2-oxoglutarate- (2OG) dependent JmjC KDM2-7 subfamily, which both employ oxidative mechanisms. Modulation of histone methylation status is proposed to be important in epigenetic regulation and has substantial medicinal potential for the treatment of diseases including cancer and genetic disorders. This article provides an introduction to the enzymology of the KDMs and the therapeutic possibilities and challenges associated with targeting them, followed by a review of reported KDM inhibitors and their mechanisms of action from kinetic and structural perspectives. This article is part of a Special Issue entitled: Methylation Multifaceted Modification — looking at transcription and beyond.

© 2014 Published by Elsevier B.V.

1. Introduction

Post-oligomerisation modifications to the nucleic acid and protein components of chromatin are of central importance in eukaryotic transcriptional regulation. Established modifications to chromatin include methylation of DNA and histones, and further modifications to histones include acetylation, phosphorylation, and ubiquitination, with the N-terminal tail of histone H3 being subjected to a particularly complex set of modifications. These modifications, together with other factors, are proposed to substantially contribute to the regulation of gene expression in a context-dependent manner in part by affecting the accessibility of DNA sequences for transcription [1-3]. The post-translational modifications affect chromatin structure and dynamics, the lifetimes of chromatin components, and mediate chromatin binding interactions. Aberrant modification patterns to chromatin have been associated with the onset and progression of both germline and somatic diseases, ranging from mental disorders to many cancers [4-6]. Presently, it seems quite possible that most, if not all, major human diseases (including at least some infectious diseases) will be linked to changes in post-oligomerisation modifications to chromatin. Consequently, many of the enzymes that catalyse these

☆ This article is part of a Special Issue entitled: Methylation Multifaceted Modification — looking at transcription and beyond.

* Corresponding authors. Tel.: +441865 275625. E-mail addresses: christopher.schofield@chem.ox.ac.uk (C.J. Schofield), richard.hopkinson@chem.ox.ac.uk (RJ. Hopkinson). 1 Authors contributed equally to the work.

modifications and their removal (sometimes referred to as 'writers' and 'erasers', respectively), as well as the multiple binding domains that interact with them, are being pursued as small-molecule targets for therapeutic benefit. Small molecules targeting chromatin-modifying enzymes for cancer treatment are already in clinical use (e.g. DNA methyltransferase inhibitors (Azacitidine, Decitabine) and histone deacetylase inhibitors (Vorinostat, Romidepsin)), and others are in trials [7-9].

N-Methylations of both DNA and histones are chromatin modifications of central importance in (epi)genetics. Due to the strength of the C-N bond, N-methylation was once thought to be irreversible, at least via direct enzymatic catalysis; however, it is now clear that N-methylation of DNA (and RNA) is both common and can be directly reversed by the action of demethylases (note that indirect methods of repair of methylated DNA, for example base excision repair, are well-established, but beyond the scope of this article).

The available evidence is that protein N-methylation occurs principally, but not exclusively, on the histone lysyl and arginyl side chains of histones H3 and H4. There is also evidence, both from 'global' prote-omic as well as focused studies, that many other proteins are N-methylated (and likely demethylated), including some associated with transcriptional regulation (e.g. methylation of p53 [10] and NF-kB [11,12]). In contrast to histone lysine acetylation, which results in transcriptional activation, histone methylation can either enable context-dependent activation or repression of transcription. In general, methylation of histone H3 at Lys9 or Lys27 (H3K9 or H3K27) or histone H4 at Lys20 (H4K20) correlates with transcriptional repression, whereas methylation of H3K4, H3K36 and H3K79 correlates with enhanced

http://dx.doi.org/10.1016/j.bbagrm.2014.05.009 1874-9399/© 2014 Published by Elsevier B.V.

ARTICLE IN PRESS

C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

transcription [13]. The outcomes of methylation depend on its site and extent, the presence of other chromatin modifications, and many other regulatory factors.

The available evidence is that lysine is one of, if not the, most diversely and extensively modified component monomer of any biological polymer [14,15]. Lysine residues can be altered by modifications including acetylation, crotonylation, ubiquitination (multiple forms), hydroxylation (at least several types), and by Ne-methylation. Lysine and arginine residues can each exist in four biologically identified methylation states (unmodified, mono-, di- and tri-Ne-methylation for lysine and unmodified, mono-, di-(symmetric)- and di-(asymmetric)-methylation for arginine) (Fig. 1A). The possibility of combinations of modifications both on the same residue and on other histone residues enables an enormity of potential complexity that is proposed to be a central factor in the regulation of gene expression. Defining how the combinations of post-oligomerisation modifications to chromatin, and associated other variables (e.g. association of binding partner proteins) regulate context-dependent transcription of individual open reading frames is a major current biological challenge. Substantial difficulties arise in epigenetic regulation, including with inhibition, due to the context-dependent nature of effects, linked, at least in higher organisms, to redundancy and adaptation. The precise mechanisms of how

N-methylation and other modifications regulate the transcription of in- 102 dividual open reading frames are complex and may take a long time, 103 possibly decades, to be unravelled in detailed chemical terms. Whilst 104 such an understanding is desirable, one would hope that it is not essen- 105 tial for therapeutic benefit to occur based on existing, or soon to be ac- 106 quired, knowledge. Many enzymes catalysing chromatin modifications 107 and their reversal, e.g. methylation/demethylation, have been identi- 108 fied; along with non-catalytic binding proteins, some are being pursued 109 as therapeutic targets. Currently, the main therapeutic focus regarding 110 methyltransferases and demethylases is cancer, but in the longer term 111 the complexity of transcriptional regulation, and in particular methyla- 112 tion, suggests that the possibilities for selective treatment of many 113 diseases involving transcriptional regulation are very substantial. 114 Histone methylation is catalysed by S-adenosylmethionine- (SAM) 115 dependent histone lysine methyltransferases (KMTs) and protein 116 arginine methyltransferases (PRMTs). Over-expression as well as both 117 activating and inactivating mutations of many KMTs correlate with can- 118 cer [16]. In several acute leukaemias, the mixed lineage leukaemia 119 (MLL) gene undergoes chromosomal translocation resulting in an 120 oncogenic MLL fusion protein, which recruits DOT1L, an H3K79 KMT, 121 causing aberrant methylation and transcription of leukaemeogenic 122 genes [17,18]. EZH2, an H3K27 KMT which is a catalytic component of 123

h2n^nh2*

HN^NH2* HN

H o Rme1

H o Rme2s

Kme1 HCHO

H2CVlell.HisO

Asp/Glu

Asp/Glu

Fig. 1. Histone lysyl demethylation is catalysed by histone demethylases (KDMs). (A) Methylation states of lysine and arginine residues in histones. Lysine residues may be mono-, di- or trimethylated on their Ne-amino groups (Kme1, Kme2 and Kme3 respectively), whereas arginine residues maybe either mono- or dimethylated on their N^-guanidino nitrogens (Rme1 and Rme2 respectively). Dimethylated arginine exists in either symmetric (Rme2s) or asymmetric (Rme2a) forms, depending upon the positions of methylation. Arginine may also be deaminated to form citrulline (Rcit), as catalysed by peptidylarginine deiminase activity. (B) Mechanisms of lysyl demethylation catalysed by the KDM1 and JmjC (KDM2-7) subfamilies. The KDM1 subfamily only accepts mono- and dimethylated lysines as substrates.

ARTICLE IN PRESS

C.C. Thinnes et al./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx 3

the PRC2 complex, is an oncogene overexpressed in many cancers and associated with poor prognosis in breast and prostate cancers [19-21]. Other KMTs, such as the Nuclear Binding SET Domain protein (NSD1), SUV39H1 and SUV39H2, are also linked to cancers (see Ref. [16] for review). A DOT1L KMT inhibitor is in clinical development for MLL-rearranged leukaemia [22]. A KMT EZH2 inhibitor [23] is being targeted at non-Hodgkin lymphoma [24]. Together, with the clinical use of DNA methyltransferase inhibitors [25], these studies validate the modulation of chromatin methylation status for cancer therapy.

Over 20 human histone Ne-methylated lysine residue demethylases (KDMs) have been identified, with many exhibiting dysregulated expression patterns in disease [26,27]. As with the KMTs, emerging evidence suggests that some KDMs are required for tumour growth. Consequently, KDMs have emerged as interesting new targets for therapeutic intervention. Demethylation of mono-N-methylated arginine residues on histones is proposed to be catalysed by the peptidyl arginine deiminases (PADIs) which catalyse hydrolysis to give citrulline and methylamine [28,29]. Although there are reports of a 'direct' N-methylated arginine residue demethylase [30,31 ], i.e. JMJD6, these are controversial. Subsequent work has revealed that JMJD6 is a lysine C-5 hydroxylase acting on arginine-rich regions of splicing regulatory proteins [32,33]. Thus, at present there is no definitive evidence for direct reversal of arginine N-methylation.

KDMs are classified into two subfamilies depending on their sequence homologies and catalytic mechanisms. Members of the first identified subfamily, the KDM1 subfamily (or lysine-specific demethylases, LSDs), are related to the well-characterised monoamine oxidases (MAOs) and utilise the cosubstrate flavin adenine dinucleotide (FAD) during catalysis of demethylation; methyl group oxidation is proposed to proceed via hydride transfer from the NE-methyl group onto FAD, forming an imine which is unstable to hydrolysis (Fig. 1B) [26]. For chemical reasons this mechanism cannot apply to quaternary amines, hence the KDM1 subfamily likely does not act on N-trimethylated lysine residues. KDM1A, the most studied KDM1 enzyme, catalyses removal of methyl groups from mono- and dimethylated lysine residues at H3K4 and, potentially, H3K9, while KDM1B catalyses demethylation specifically at H3K4 [26]. KDM1A is reported to be oncogenic and/or over-expressed in cancers including leukaemias and solid tumours [34-37]. Notably, KDM1A is specific for H3K4me1/me2 when a component of the CoREST co-repressor complex [38], but in vivo work has suggested that the presence of the androgen receptor (AR) may change its specificity from H3K4 to H3K9 [39]. Thus, KDM1A might potentially act either as a transcriptional corepressor or coactivator depending on its binding partners and substrates. This example nicely illustrates the importance of why a detailed biological understanding of context-dependent roles will be useful in interpreting the pharmacological effects of KDM inhibitors.

The JmjC domain-containing KDMs (KDM2-7 subfamilies; JmjC KDMs) represent the larger KDM class, comprising about 20 human enzymes which are grouped into five subfamilies (KDM2/7, KDM3, KDM4, KDM5, and KDM6) [40]. They can catalyse the demethylation of mono-, di- and trimethylated lysines at multiple sites, using 2OG and dioxygen as cosubstrates and Fe(II) as a cofactor [26]. The JmjC KDMs are over-expressed in multiple types of cancer cells [34,37]. Some JmjC KDMs are implicated in neural development and/or function and are associated with conditions including X-linked mental retardation, autism and midline defects [41-44]. The JmjC KDMs are related to N-methyl DNA and RNA demethylases (AlkB homologues and the fat mass and obesity-related protein, FTO) and other nucleic acid oxygenases including the TET (ten eleven translocation) hydroxylases which catalyse oxidation of the methyl group of 5-methylcytosine bases to give C-5 hydroxymethyl-, formyl, and carboxycytosine [45]. There is evidence that JmjC KDM-catalysed demethylation proceeds via the consensus mechanism for 2OG oxygenase catalysis: following sequential active site binding of 2OG, substrate, then dioxygen, oxidative decarboxylation occurs to give succinate, carbon dioxide and an Fe(IV)-oxo intermediate,

which reacts with the NE-methyl group, either via a radical rebound or direct insertion process to form an unstable hemiaminal [26]. Subsequent fragmentation produces the demethylated lysine and formaldehyde (Fig. 1B) [46,47]. Thus, the overall reactions of the two KDM subfamilies are closely related; however, the JmjC KDMs are capable of catalysing demethylation of trimethylated lysines because their mechanism does not require imine formation.

The catalytic domain of the KDM1 subfamily, the amine oxidase-like (AOL) domain, is related to the large superfamily of monoamine oxidases [48], with MAO-A and MAO-B being the closest homologues [49]. MAOs catalyse the oxidation of monoamine neurotransmitters, such as serotonin and dopamine. Similarly, the catalytic domains of JmjC KDMs, the JmjC domains, are also part of the large superfamily of 2OG-dependent oxygenases, which has approximately 60 human members [50]. In addition to the roles of the JmjC KDMs in epigenetic regulation, other 2OG oxygenases play key roles in human biology, including in fatty acid metabolism, collagen biosynthesis, protein biosynthesis, nucleic acid repair/modification and hypoxia response [50].

Pioneering inhibitor development strategies for both KDM subfamilies initially focused on the testing of generic mechanism-related inhibitors for each family of enzymes. For example, the first generation KDM1 inhibitor, tranylcypromine (2, Fig. 2), is an FDA-approved MAO inhibitor that is used to treat mood and anxiety disorders [51]. Broad-spectrum 2OG oxygenase inhibitors, such as the 2OG mimetics N-oxalylglycine (NOG, 32, Fig. 4) and pyridine-2,4-dicarboxylate (2,4-PDCA, 47, Fig. 4), are also JmjC KDM inhibitors [52,53]. Given the structural similarity of KDMs and the conservation of active site features with multiple homologous enzymes, one of the major challenges for future KDM inhibitor discovery is to develop inhibitors selective for particular (sets of) KDMs. Although beyond the scope of this review, it is also important to note that the activities of most, if not all KDMs, are likely directed by binding domains contiguous with the 'catalytic' JmjC domains (e.g. the KDM4 enzymes contain Tudor and PHD domains in addition to the JmjC domain) or by non-covalent interactions with targeting proteins (see above for KDM1A); these non-catalytic domains are also of therapeutic interest in isolation and in combination with KDM inhibitors. KDM inhibitors are required for target validation and for probing biological function. Below, we describe recent advances in the development of inhibitors for the KDM1 and JmjC KDM subfamilies, and discuss future strategies for KDM inhibitor discovery.

2. KDM1 inhibitors

2.1. Tranylcypromine and analogues

The known amine oxidase inhibitors phenelzine (1, Fig. 2A) [54], tranylcypromine (PCPA) (2, Fig. 2A) [54,55], and pargyline (3, Fig. 2A) [39] were amongst the first compounds reported to inhibit KDM1 catalysis. Although subsequent studies with pargyline did not reveal KDM1 inhibition [54,56], both the hydrazine derivative phenelzine 1 and cyclopropane derivative tranylcypromine 2 were subsequently found to inhibit KDM1A both in vitro and in cell-based experiments, with inhibitory potencies similar to those observed for these compounds against amine oxidases [54,57]. Structural and mechanistic studies with MAOs have shown that phenelzine 1 and tranylcypromine 2 react with the co-substrate FAD via their hydrazine and cyclopropyl groups respectively, forming covalent adducts within the active site [58-60]. Analogous 'suicide' inactivation has also been reported for KDM1A inhibition by tranylcypromine 2 (Fig. 2) [55,61-63].

Subsequent studies, which were partly informed by crystallographic analyses, have utilised tranylcypromine 2 as a scaffold for the development of more selective cyclopropane-based KDM1 inhibitors (for representative structures of analogues, see Fig. 2A) [61,64-71]. To date, a number of tranylcypromine analogues, modified on both the aromatic ring and amino groups, have been tested for KDM1 activity, resulting

ARTICLE IN PRESS

C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

Phenelzine, 1 Tranylcypromine, 2

Pargyline, 3

'1 H.3LV

h i' i

qtarkstggkaprkqla

GREEN = FAD-Tranylcypromine Adduct = FAD-Peptide 15 Adduct

4gut 2z3y 2xaj 2xag 2xah 2xas 2xaq 2z5u 3abt 3abu 4guu 2uxn

f y f f

ARTICLE IN PRESS

Namolin, 26

f ^—NH Cl 1 N

h2nAo A OVI

O 25 NH

F,C.. „0.

N .0 HN

Fig. 3. Structures of representative non-tranylcypromine-based KDM1 subfamily inhibitors. Bisguanidines, bisbiguanides, polyamines and (thio)urea-containing compounds (for examples see 21-24) have all been shown to inhibit KDM1 activity in vitro, with some showing effects in cells. A detailed mechanistic/structural understanding of KDM1 inhibition by these compounds is presently not available. Recently, a series of other inhibitors of the KDM1 subfamily have been identified, including Namolin (26), aminothiazoles (27-29), benzohydrazides (30) and a tricyclic pyridine (31).

in the generation of inhibitors which display both improved inhibitory potency against KDM1A and, in some cases, improved selectivity.

Improvements in both potency and selectivity were observed for compounds substituted on the aromatic ring of tranylcypromine [61,65,66,70, 71]; crystallographic analyses reveal that these groups likely protrude into the methylated lysyl binding pocket, forming favourable hydrogen-bonding and hydrophobic interactions (Fig. 2) [61,66]. Some of the tranylcypromine derivatives inhibit cell growth in cancer cell lines including murine acute promyelocytic leukaemia blasts [61,65,72]. In addition, a number of compounds substituted at the primary amino group display improved activity relative to tranylcypromine 2 and are selective for the KDM1 family over related amine oxidases [67-69]. Some of these compounds are brain-penetrating [67]. Promising preclinical data for treatment of acute myeloid leukaemia (AML) by targeting KDM1A with tranylcypromine 2 have provided target validation evidence. A KDM1A inhibitor has been recently described which, it is reported, shows both unprecedented in vitro potency (IC50 at 20 nM) and selectivity over all analysed amine oxidases including KDM1B. This compound (ORY-1001) is projected to enter clinical trials in 2014 [73].

2.2. Substrate mimetics

The preference of the KDM1 subfamily for protein substrates, as well as their susceptibility to mechanism-based inhibitors (e.g. tranylcypromine 2), has led to the development of a series of peptides containing reactive groups selective for KDM1 over MAOs. Peptide fragments of H3 were synthesised with K4 lysine analogues [57,74] and modified to contain reactive groups (e.g. cyclopropyl, propargyl) intended to react with the FAD cofactor of the KDM1s. A number of the peptides inhibited KDM1A activity in vitro, with a hydrazine-derivatised peptide showing the most potent activity ( 13, Fig. 2C, Ki = 4.4 nM). Mechanistic and crystallographic studies with the

propargyl group-containing peptides (14 and 15, Fig. 2) revealed that inhibition by these peptides is indeed induced, at least in part, through reaction with FAD [62,75] (a similar mechanism for inhibition by analogous peptides has also been proposed) [57].

2.3. Polyamines

Polyamines are known MAO inhibitors which also inhibit KDM1 [76, 77]. An early screen of bisguanidines and bisbiguanides identified noncompetitive inhibitors of KDM1A with low-micromolar IC50 values (for representative structures, see Fig. 3) [78]. Two of the compounds were tested in cell models, resulting in increased expression of tumour suppressor genes as well as fluctuations in levels of H3K4me2 (increase), H3K9ac (increase), H3K4me1 (decrease) and H3K9me2 (decrease) marks. Later studies with a series of polyamines (octamines and decamines) also revealed inhibition of recombinant KDM1Aand similar effects in cells (Fig. 3) [79]. Further derivatives incorporating bisurea and bisthiourea groups also exhibited KDM1A activity, with two compounds inhibiting growth of non-small cell lung carcinoma cells (Fig. 3) [80]. Two recent patents also reveal KDM1A inhibition by amidine- and dithiocarbamate-containing compounds [81,82]. However, more detailed mechanistic, selectivity and cell-based studies are required for the 'polyamine' inhibitor class. In particular, it is of interest to determine to what extent the cell-based observations result from selective KDM1 subfamily inhibition.

2.4. Other KDM1 inhibitors

Several types of reversible inhibitors of KDM1 have recently been reported. Namolin (26, Fig. 3), a cell-active, reversible inhibitor of KDM1A, was identified by screening against the closely related monoamine oxidases MAO-A and MAO-B [49]. Aminothiazoles have been identified as

Fig. 2. Mechanism-based inhibitors of the KDM1 subfamily. (A) Structures of representative mechanism-based KDM1 inhibitors. The MAO inhibitors phenelzine 1, tranylcypromine 2 and pargyline 3 were among the first reported inhibitors of KDM1, which led to the development of KDM1 -selective analogues. (B) Views from X-ray crystal structures of tranylcypromine (2) and peptidic inhibitor (14) cross-linked to FAD in the active site of KDM1A (PDB IDs: 2Z3Y [63] and 2UXN [155] respectively). Views from both structures are overlaid. (C) Views from crystal structures of tranylcypromine analogues (FAD adducts) in the active site of KDM1A (PDB IDs: 2XAJ, 2XAG, 2XAH, 2XAS and 2XAQ) [61 ]. The analogues protrude into the substrate binding pocket. (D) Proposed structures of adducts formed by reaction of tranylcypromine with FAD in the active site of KDM1s. There is evidence for three of the structures from crystallographic analyses (see section E). (E) Structures of FAD adducts of mechanism-based inhibitors bound in the active sites of KDM1s (views of structures are from PDB IDs: 2Z3Y [63], 2XAJ [61], 2XAG [61 ], 2XAH [61 ], 2XAS [61], 2XAQ[61], 2Z5U [63], 3ABT [66], 3ABU, 4GUU [156] and 2UXN [155]).

ARTICLE IN PRESS

6 C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

310 weak reversible inhibitors (compounds 27 and 28, IC50 at 249 and as a marker for KDM1A inhibition in THP1 cells [84] indicated that these 314

311 437 |aM respectively, Fig. 3) of KDM1A following a fragment-based compounds are not cell-active (>50 |jM). 315

312 screen [83]. The initial hits were developed to increase potency (29,low- A structure-based virtual screen, followed by computational 316

313 estIC50at7 |jM, Fig. 3); however cellular assays using CD86 expression analyses and biochemical screening, led to the identification of an 317

A TCA Cycle Intermediates and 2OG Mimetics

NOG, 32

Succinate, 33

Fumarate, 34

но' : • "

(SJ-2HG, 35

(RJ-2HG, 36

B Hydroxamic Acids and Daminozide

SAHA, 40

o^ n OH

O^ N OH

Methylstat, 44

Daminozide, 45

C Pyridine Derivatives

2,4-PDCA, 47

ARTICLE IN PRESS

C.C. Thinnes et al./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx

KDM4 Subfamily

KDM4A + NOG

E190 S196

KDM4A + Daminozide

KDM4A + 38

KDM4A + (R)-2HG

KDM4A + (S)-2HG

E190 »

KDM4A + 2,4-PDCA

Y177 N198

KDM4A + 52

KDM4A + 4C8HQ

KDM4A + IOX1

Fig. 5. Crystal structures of inhibitors bound to KDM4A. KDM4 inhibitor structures reported to date are for bidentate iron chelators. One chelating group from the inhibitor coordinates opposite to the iron-binding glutamate residue (Glu190 in KDM4A); however, the second chelating group may coordinate either opposite the first iron-binding histidine (His188 in KDM4A), as observed for 2OG chelation (and for NOG, see top left), or opposite the second histidine (His276 in KDM4A), as observed for 2,4-PDCA (see middle right). The preference of the two binding modes is likely determined by steric factors. Binding of IOX1 in the active site of KDM4A induces translocation of the active site metal (red arrow, bottom right). In all structures, the active site iron is substituted for nickel. PDB IDs: 2OQ6 [121], 4AI9 [124], 2WWJ [52], 2YBK [105], 2YBS [105], 2VD7 [104], 3PDQ[134], 4BIS [53] and 3NJY [135].

Fig. 4. Structures of representative iron-chelating inhibitors of the JmjC KDMs. (A) Structures of tricarboxylic acid (TCA) cycle intermediates and 2OG mimetics. 2OG is a co-substrate of the 2OG oxygenases. Some 2OG oxygenases are inhibited by succinate (33, a co-product of catalysis), fumarate (34), and 2-hydroxyglutarate (35 and 36). Levels of these small molecules can be substantially increased in some tumour cells. N-Oxalylglycine (NOG, 32) is a close isostere of 2OG. C-a derivatisation of NOG can confer selectivity for different 2OG oxygenase subfamilies. (B) Structures of hydroxamic acid-containing inhibitors and Daminozide (45). Hydroxamic acids are established metal chelators and with appropriate functionalisation can be potent and selective 2OG oxygenase inhibitors. Examples include the JmjC KDM inhibitor Methylstat (44), and SAHA (40), which is a clinically used inhibitor of histone deacetylases that also inhibits JmjC KDMs in vitro. Daminozide (45), a small achiral hydrazide, inhibits the KDM2/7 subfamily of JmjC KDMs. (C) Examples of pyridine-based KDM inhibitors. Pyridine-2,4-di-carboxylic acid (2,4-PDCA, 47) is a broad-spectrum 2OG oxygenase inhibitor that chelates active site-bound iron via its pyridyl nitrogen and 2-carboxylate. KDM4-selective derivatives of 2,4-PDCA have been prepared via substitution at the 3-position (e.g. 48). Recently, pyridine-containing fragments without a 2-carboxylate group (e.g. 49) have also been reported to inhibit JmjC KDMs via monodentate iron chelation. Other pyridine-containing inhibitors include bipyridyl compounds (e.g. 51 and 52), which chelate iron via both pyridyl nitrogens, and the pyridylhydrazone 50, which was recently identified from a high-throughput screen. (D) Structures of 8-hydroxyquinoline derivatives. 8-Hydroxyquinolines chelate iron in a bidentate manner via their pyridyl nitrogen and phenolic oxygen atoms. 5-Carboxy-8-hydroxyquinoline (IOX1, 54), which was identified from a high-throughput screen against KDM4E, is a broad-spectrum 2OG oxygenase inhibitor that exhibits moderate selectivity for some JmjC KDM subfamilies (KDM2/7, KDM3, KDM4, KDM6). Substitution at the 2-, 4-, 5-and 7- positions of the 8-hydroxyquinoline ring has resulted in improved selectivity for KDMs (e.g. 56-59). Crystallographic studies with KDM4A, KDM6B and the HIF asparaginyl hydroxylase FIH reveal that IOX1 can induce translocation of the active-site iron (see Figs. 5 and 7). Such metal movement is not observed for4-carboxy-8-hydroxyquinoline (4C8HQ 55) binding to KDM4A (Fig. 5).

ARTICLE IN PRESS

8 C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

N'-(1-phenylethylidene)-benzohydrazide series as KDM1A inhibitors. The optimised compound 30 was found to be highly potent (Kj = 31 nM) and selective over monoamine oxidases (MAO-Aand MAO-B >300 |jM), to inhibit cell proliferation of several cancer cell lines, and to increase H3K4me2 levels in VCaP cells in a dose-dependent manner [85]. Recently, GSK354, a novel potent (IC50 < 100 nM), highly selective (MAO IC50 > 200 |jM) and cell-active inhibitor (31, IC50 = 1.4 ^M, Fig. 3) of KDM1A has been declared as a chemical probe suitable for use in biological function studies [86].

3. JmjC KDM inhibitors

3.1. Overview

The 2OG oxygenases were originally identified as playing roles in collagen biosynthesis comprising prolyl and lysyl hydroxylations. Prolyl-4-hydroxylation stabilises the collagen triple helix and consequently, in pioneering inhibition work, collagen prolyl-4-hydroxylases were targeted for fibrotic diseases [87,88]. These studies led to the identification of various compounds that compete with 2OG and coordinate to the active site ferrous iron [89]. Prior to the identification of the JmjC KDMs, 2OG oxygenases were found to play key roles in the physiological response to limiting oxygen availability via the post-translational prolyl and asparaginyl hydroxylation of the hypoxia-inducible transcription factor (HIF) [90]. Whereas the HIF prolyl hydroxylases (PHD or EGLN enzymes) are related to the collagen prolyl hydroxylase subfamily of 2OG oxygenases, HIF asparaginyl hydroxylation is catalysed by a JmjC-type enzyme (factor inhibiting HIF, FIH) but one which catalyses a stable hydroxylation modification rather than demethylation via hydroxyl-ation [90]. Some of the inhibitors developed/applied to the collagen and HIF hydroxylases, such as N-oxalylglycine (NOG, 32, Fig. 4), have been found to be useful both as broad-spectrum 2OG oxygenase inhibitors, including for the KDMs, and as templates for development by functionalisation into more specific inhibitors [89].

There is now a considerable amount of structural information available for 2OG oxygenases [91,92]. The crystal structures have revealed substantially conserved Fe(II) and 2OG binding sites; however, differences in Fe(II) and 2OG binding sites are subfamily-characteristic and can be exploited for the development of selective inhibitors [92,93]. As may be anticipated from the diversity of substrate types for 2OG oxygenases - proteins, nucleic acids, small molecules and lipids -there are greater variations in the substrate binding elements, which separately, or in conjunction with Fe(II) binding, can be exploited for the development of selective inhibitors. However, approaches involving solely mimicking the binding of histone substrates to the JmjC KDMs raise the possibility of selectivity issues relating to non-2OG oxygenase proteins that bind the same/similar histone sequences. Further, it is important to note that 2OG oxygenase catalysis involves an ordered sequential mechanism involving conformational changes that are not yet fully defined and probably not (completely) accessible by crystallo-graphic studies. For example, recent studies with inhibitors have demonstrated that some compounds can induce movement (> 1.0 A) of the metal within the active site [53,94]. Thus, structural analyses by solution phase techniques (e.g. NMR) are of importance in the field [95,96].

It is also worth mentioning that inhibition of2OG oxygenases by endogenous small molecules is of pathophysiological relevance. Mutations to tricarboxylic acid (TCA) cycle enzymes are common in tumours and can result in very substantial increases in the concentrations of succi-nate (33, Fig. 4), fumarate (34, Fig. 4), or 2-hydroxyglutarate ((S/R)-2HG, 35 and 36, Figs. 4/5) [97-99]. The formation of 2HG is of particular biochemical interest — 'gain of function' mutations cause isocitrate dehydrogenase (IDH) 1 and 2 to convert 2OG into 2HG, as well as produce 2OG from isocitrate [100]. The increases in TCA intermediate/2HG levels are proposed to be pro-oncogenic, at least in part due to inhibition of 2OG oxygenases involved in transcriptional regulation. Succinate [101, 102], which is a co-product of 2OG oxygenase catalysis, fumarate [103,

<£1^0

R = OH, GSK-J2, 63 R = OEt, GSK-J5, 64

Fig. 6. Structures of the KDM6 subfamily inhibitor GSK-J1 and derivatives. GSK-J1 (61) is an iron chelator, which induces metal movement within the active site of KDM6B (for a view of a crystal structure of GSK-J1 bound in the active site of KDM6B, see Fig. 11). The ethyl ester prodrug (GSK-J4,62) is active in cells, whereas the prodrug form of the inactive analogue GSK-J2 (63) shows no cellular effects (GSK-J5,64).

104] and 2HG all inhibit 2OG oxygenases, though rather weakly (in 381

the |jM to mM range for KDM2A, KDM4A, KDM4C and KDM5B, as 382

shown with isolated proteins and in cells [105,106]), via competition 383

with 2OG [103,107], though more complex mechanisms are possible. 384

It is also notable that succinate release is rate-limiting for some 2OG 385

oxygenases [108] and that high levels of the cosubstrate 2OG can be in- 386

hibitory [109]. Further, in assessing the potential relevance of 2OG oxy- 387

genase inhibition by endogenous 2OG competitors, consideration 388

should be given to the 2OG/inhibitor concentrations in cells, enzyme 389

concentration, and 2OG KM/KD values; the latter can vary substantially 390

(> 100-fold) [110]. Ester derivatives of TCA intermediates, 2HG and 391

2OG have been used as prodrugs in cellular functional studies because 392

the parent di-acids have poor cell penetration [111]. 393

In the following sections we describe studies on various types of 394

JmjC KDM inhibitors — our objective is not to be comprehensive (espe- 395

cially with respect to the patent literature) but to illustrate different 396

types of inhibitors developed to date and make general points. 397

Reflecting current reports, our focus is on active site-directed inhibitors, 398

but (evidently) there is also potential for the inhibition of auxiliary do- 399

mains, and by targeting the non-active site regions in the core JmjC do- 400

main. The latter approach is exemplified by compounds that act on the 401

zinc-binding domain of KDM4 [112]. It is also notable that recent work 402

has focused on inhibition of the JmjC KDMs, with fewer reports to date 403

on the nucleic acid oxygenases such as FTO and the TET enzymes 404

[113-115]. 405

3.2. N-Oxalyl amino acid-based JmjC KDM inhibitors 406

N-Oxalylglycine (NOG, 32, Fig. 4) is a closely related analogue of 2OG 407

wherein the C-3 methylene group is replaced with an NH group to give 408

an N-oxalyl amide derivative that likely stalls the catalytic reaction by 409

hindering oxygen binding to the active site iron. Although first de- 410

scribed as a prolyl hydroxylase (PHD) inhibitor, NOG is a broad- 411

spectrum inhibitor of2OG oxygenases [116], which, following initial 412

work with KDM4E [104], has been shown to inhibit all tested JmjC 413

KDMs. [53] Multiple crystal structures for 2OG oxygenases complexed 414

with NOG have been reported (e.g. for KDM4A, KDM6A, PHF8; see 415

Figs. 5 and 11); the structures reveal that NOG binds to the active site 416

metal (iron or an iron surrogate) via its C-2 oxo and C-1 carboxylate 417

groups, with its glycine carboxylate occupying the same position as 418

PBIT, 65 Ebselen, 66

Fig. 7. Structures of 2-4(4-methylphenyl)-1,2-benzisothiazol-3(2H)-one (PBIT) and Ebselen. PBIT (65) and ebselen (66) inhibit both the KDM4 and KDM5 subfamilies of KDMs, probably via removal of the enzyme-bound zinc.

ARTICLE IN PRESS

ARTKQTAR-^-VSTGGK<aC>-NH2 ; H

C.C Thinnes etal./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx О

ARK{me3)S

,GGK-NH2

O Y он -A->nh

о HRY-NH2 STGGV(Kme3)KK V-NH

HO^O .s

BIX-01294, 70

E67, 71

E67-2, 72

Fig. 8. Structures of histone substrate-competitive JmjC KDM inhibitors. Compounds 67-69 comprise both iron-chelating groups and histone fragments to enable potency and selectivity for JmjC KDMs. The lysine methyltransferase inhibitor BIX-01294 (70) also inhibits JmjC KDMs; development of BIX-01294 analogues has resulted in JmjC KDM-specific inhibitors (71 and 72). Crystallographic studies of 71 with KDM7A suggest that the compound may compete with the histone substrate for binding (Fig. 11).

ОН О

Myricetin, 73

ОН о Baicalein, 74

Epigallocatechin gallate, 75

но" ^ v "NH2 Dopamine, 76

Haematoxylin, 77

Caffeic acid, 7s

Tripartin, 79

X = OH, OEt, nh2

Pyrido[1,2-a]indoles

Fig. 9. Structures of flavonoid, catechol, tripartin, and pyrido[1,2-a]indole inhibitors of JmjC KDMs. Numerous flavonoid and catechol compounds have been identified as JmjC KDM inhibitors from high-throughput screening, including myricetin (73) and dopamine (76). Precise mode(s) of inhibition, however, are unclear. Tripartin (79) is an indanone-based natural product that was shown to inhibit KDM4. Pyrido[1,2-a]indoles have been recently reported to inhibit KDM4C via unknown mechanism(s).

Fig. 10. View from a crystal structure of KDM4A with bound peptide substrate analogue (sequence ARK(me3)SCGGK, yellow) and N-oxalyl-D-cysteine (DNOC, pink). The peptide and DNOC form a disulfide linkage within the active site (dashed line, circled). Replacing the disulfide bond with a sulfide led to the development of a stable analogue exhibiting potent and selective inhibition for the KDM4 subfamily (68, Fig. 8). PDB ID: 3U4S [141].

ARTICLE IN PRESS

C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

A KDM6 Subfamily

KDM6A + NOG

KDM6B + IOX1

H1226 H1146

B KDM2/7 Subfamily PHF8 + NOG

$ S238

rff £

mKDM6B + GSK-J1

C1408 C1408

W1407 |

W1410 H1470

H13872OGih.

E1389 > ^ . 'OX1

N1400 ,N™V/ A1482

N189 I191

PHF8 + Daminozide

K264 L236 N189

KDM7A + E67-2

99 L271 N224

F266 V255

T244 H247

F266 V321 I1

V255 V255

I226 T279

T244 H247

H354 H282

V290 V290

Fig. 11. Views from crystal structures of inhibitors bound to the KDM6 and KDM2/7 subfamilies. As observed in inhibitor structures with KDM4A, bidentate iron chelators may adopt one of two coordination modes around the iron, with one iron-binding group positioned opposite to the iron-binding carboxylate residue (Glu1148 in KDM6A, Glu1389 in KDM6B, Asp249 in PHF8 and Asp284 in KDM7A) and the other opposite to one of the iron-binding histidines. Both IOX1 (54, Fig. 4) and GSK-J1 (61, Fig. 7) induce metal translocation in KDM6B (note: the crystal structure of GSK-J1 binding was solved using mouse KDM6B). E67 (71, an analogue of the lysine methyltransferase inhibitor BIX-01294, 70, Fig. 8) binds away from the iron binding site in KDM7A, suggesting no metal chelation. PDB IDs: KDM6A + NOG, 3AVS (metal = nickel) [157]; KDM6B + IOX1, 2XXZ (metal = nickel) [53]; mKDM6B + GSK-J1, 4ASK (metal = cobalt) [94]; PHF8 + NOG, 3KV4 (metal = iron) [158]; PHF8 + Daminozide,4DO0 (metal = zinc) [124]; KDM7A + E67,3U78 (metal = nickel) [142].

the C-5 carboxylate of 2OG. Although NOG is a relatively broad-spectrum inhibitor, it should be noted that the potency of 2OG oxygenase inhibition by NOG does vary substantially [53]. The cell-penetrating dimethyl ester form of NOG, dimethyl N-oxalylglycine (DMOG), has found widespread applications as a hypoxia mimetic [117,118] because it inhibits both HIF prolyl and asparaginyl hydroxylases resulting in upregulation and activation respectively of the HIF system [119]. It is possible that some of the observed cellular effects of DMOG result fromJmjC KDM, or other 2OG oxygenase, inhibition [53].

The potential for selective inhibition of the JmjC subfamily was demonstrated by pioneering studies on inhibition of the HIF hydroxylases. Comparison of PHD2 and FIH crystal structures led to the proposal that C-alpha derivatisation of NOG may lead to FIH-selective inhibitors

[120]. Indeed the NOG derivative N-oxalyl-D-phenylalanine (37, Fig. 4) is selective for FIH over the PHDs; the dimethyl ester prodrug form is an active FIH inhibitor in cells [119]. The benzyl group of 37 occupies a hydrophobic binding pocket in FIH close to the Fe(II) binding site which is absent in the PHDs.

As with FIH, the KDM4 subfamily also possess a hydrophobic pocket close to the Fe(II) centre, which may be exploited for selective inhibition

[121]. A combined approach involving structure-guided design coupled with disulfide-based dynamic combinatorial chemistry employing non-denaturing mass spectrometry (DC-MS) for screening was employed to identify disulfides that occupied this pocket, which is larger in the

KDM4s compared to FIH. Subsequently, stable analogues of disulfides 443 that were observed to bind relatively tightly by MS were prepared. It 444 was found that substitution at the para position of the phenyl ring of 445 N-oxalyl-D-tyrosine resulted in molecules that inhibit KDM4E in the 446 micromolar range and which are selective over both FIH and the PHDs 447 [52]. A crystal structure of KDM4A in complex with one of the potent 448 inhibitors (38) was obtained (Fig. 5). 449

Analysis of the X-ray crystal structure of KDM4A in complex with 450 NOG (occupying the 2OG binding site) and a trimethylated peptide 451 (occupying the substrate pocket, PDB ID: 2OQ6) [121] led to the design 452 and synthesis of NOG derivatives substituted with an alkyl-linked 453 dimethylaniline group in order to mimic the interactions of the 454 trimethylated peptide with the protein (Fig. 4) [122]. These derivatives 455 maintain the inhibitory action of NOG against KDM4A, and exemplify 456 the strategy of linking the 2OG and peptide substrate binding sites for 457 JmjC KDM inhibition (for representative structures, see 39, Fig. 4) [122]. 458

3.3. Hydroxamic acid-based JmjC KDM inhibitors 459

In early screenings for templates that inhibit the JmjC KDMs, 460 work with KDM4E identified hydroxamic acid inhibitors, including 461 conformationally constrained 2OG analogues. Interestingly, amongst 462 the hydroxamic acids found to inhibit KDM4E was the histone 463 deacetylase (HDAC) inhibitor SAHA (Vorinostat, 40, Fig. 4) [104]. This 464

ARTICLE IN PRESS

C.C. Thinnes et al./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx 11

observation raises the possibility that some of the pharmacological effects of SAHA may result from inhibition of targets other than HDACs, including theJmjC KDMs.

The inhibitory activity of SAHA and the knowledge that hydroxamic acids are good metal chelators led to the design of inhibitors where the 2-oxoacid group of2OG is replaced with a hydroxamic acid moiety [123, 124]. The resulting compounds containing the hydroxamic acid scaffold show good selectivity for the KDM4 family over the related HIF PHDs. Molecules where the hydroxamic acid scaffold is connected through an alkyl linker to a dimethylamino group were designed based on the KDM4A:NOG crystal structure (PDB ID: 2OQ6 [121]) [125]. The protein interaction of the protonated tertiary amine likely mimics those of the positively charged NE-methylated lysine residue. A linker of eight carbon atoms was found to give optimum potency for KDM4A and KDM4C inhibition (42, Fig. 4). Replacing the dimethylamino group of 42 with a cyclopropane ring (43, Fig. 4) shifted the selectivity from the KDM4 subfamily to the KDM2/7 subfamily. It is proposed that the cyclopropane group occupies a hydrophobic pocket present in the KDM2/KDM7 subfamily.

The strategy of designing inhibitors capable of binding to both the 2OG and substrate binding pockets was applied in the discovery of Methylstat (44, Fig. 4), which contains a carboxyhydroxamic acid designed to occupy the 2OG binding pocket attached to a protonated lysine mimic through a hydrocarbon linker. Whilst the free acid inhibits KDM4C in the micromolar range, the methyl ester was found to be necessary for cell activity [126].

3.4. Daminozide

Daminozide (45, Fig. 4) is a (now largely discontinued) agrochemical, which inhibits 2OG oxygenases involved in gibberellin biosynthesis in plants [127]. Following screening of a set of potential 2OG analogues, Daminozide was found to also inhibit the KDM2/7 family enzymes (KDM2A (1C50 = 1.5 |jM) and PHF8 (1C50 = 0.55 |jM)) - somewhat unexpectedly given its small size. Some analogues of Daminozide, including a hydroxamic acid variant (46, Fig. 4), also revealed inhibition, albeit

with poorer selectivity (and in some cases, poor compound stability). Daminozide is selective for the KDM2/7 subfamily over other members of the human JmjC KDMs. Crystallographic studies reveal Daminozide chelates the active site iron via its dimethylamino nitrogen lone pair and C-4 carbonyl group, with its C-1 carbonyl occupying the same site as the 2OG C-5 carboxylate (Figs. 5/11). This selectivity may be engendered by the more lipophilic region created by the Tyr257, Val255 and Ile191 residues adjacent to Fe(ll) in the KDM2/7 subfamily compared to the more hydrophobic residues in the corresponding regions in other JmjC demethylases, as indicated by crystallographic studies [124].

3.5. Pyridine dicarboxylates and related JmjC KDM inhibitors

Pyridine dicarboxylates are more rigid 2OG analogues than NOG and the acyclic hydroxamic acids. Of the various regioisomers, pyridine-2,4-dicarboxylic acid (2,4-PDCA, 47, Figs. 4/5/12) is the broadest-spectrum 2OG oxygenase inhibitor, and is an inhibitor of most JmjC KDMs. [104] It is used as dimethyl or diethylester prodrugs for cellular studies where it has been shown to inhibit JmjC KDM activity by antibody-and MS-based assays [128].

The observation that the 2OG binding pocket is smaller in the HIF prolyl hydroxylases than the KDM4 demethylase subfamily led to the design of 2,4-PDCA derivatives substituted at the C-3 position to block binding in PHD2 whilst retaining potency against KDM4E. Notably, one of the most potent inhibitors was the 3-amino derivative 48 (Fig. 4) [129]. Although the 2-carboxy group in 2,4-PDCA and its derivatives is involved in binding to Fe(ll) in the active site, 3-amino-4-carboxy pyridines lacking the 2-carboxy group (e.g. 49, Fig. 4) have also been recently described as JmjC inhibitors [130]. The 4-carboxy group, which mimics the 2OG carboxylate, is also not essential for JmjC inhibition; screening of the National Cancer Institute's 'Diversity Set' library resulted in the identification of a pyridylhydrazone derivative as a micromolar to sub-micromolar inhibitor of KDM4A-E, KDM6B and KDM5A (50, Fig. 4) [131]. This compound was also found to have an anti-proliferative effect in cancer cell lines and to inhibit tumour growth in mice. The Z isomer was less efficacious in causing

Fig. 12. Analysis of heterocyclic iron chelators binding to JmjC KDMs. Overlays of reported structures of heterocyclic inhibitors bound to JmjC KDMs reveal preferential positioning of 2aromatic groups within the active sites. In structures of KDM4A with bound 2,4-PDCA (section A), 1OX1 (section B),4C8HQ (section C) and the bipyridine compound 52 (section D), an aromatic group is positioned between the metal site and the 2OG C-5 carboxylate binding site defined by residues Tyr132 and Lys206 (ring position A, section E). In KDM4A, the position of this aromatic group is likely stabilised by n-n stacking interactions with Phe185. For bicyclic chelators (1OX1,4C8HQand 52), the other aromatic groups are positioned in one of two discrete orientations (ring positions B or C, sections E and F). Despite 1OX1 inducing translocation of the active site metal, the position of quinolyl bicyclic system is identical to that adopted by 4C8HQ, suggesting metal movement may be preferred to altered ring-positioning, at least in KDM4A.

ARTICLE IN PRESS

C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

cell death, possibly because of reduced demethylase activity as a result of being unable to chelate Fe(11) in the active site in a bidentate manner.

3.6. Bipyridine and related JmjC KDM inhibitors

2,2'-Bipyridine and its derivatives are precedented 2OG oxygenase inhibitors including of the H1F PHDs [132,133]. Screening efforts have shown that the carboxy derivative 4'-(methoxycarbonyl)-2,2'-bipyridine-4-carboxylic acid (51, Fig. 4) inhibits KDM4E in the micromolar range [104]. More potent analogues were discovered by replacing the methyl ester with a range of substituted amides. Optimisation of the amide substituent led to compound 52 (Fig. 4) which has 40-fold improved KDM4E activity (180 nM) compared to 51 [134]. Crystallograph-ic studies demonstrated that 52 binds to KDM4A in the active site, chelating Fe(11) through the two pyridyl nitrogens (Figs. 5/12).

3.7. 8-Hydroxyquinoline-based JmjC KDM inhibitors

A high-throughput screen led to the identification of a variety of 4-, 5-, and 7-substituted 8-hydroxyquinolines that are active against KDM4E [135]. The 5-substituted hit compounds were most active and were validated in a secondary assay against KDM4A and KDM4E. These compounds also showed activity against F1H and PHD2 although with somewhat weaker potency. The most potent compound, 5-carboxy-8-hydroxyquinoline (1OX1, 54, Fig. 4) displayed an 1C50 of 0.2 |jM in vitro and was active in cells with an EC50 of 87 |aM. 1OX1 is a broad-spectrum 2OG oxygenase inhibitor, displaying sub-micromolar in vitro potency against KDM4A/C/D/E and KDM3A, micromolar 1C50s against KDM6A, KDM2A, PHF8 and KDM5C, and high-micromolar activity against BBOX1 [53]. 1OX1 chelates Fe(11) in the enzyme active site in a bidentate manner via its quinoline nitrogen and phenol oxygen atoms. The carboxylic acid at the 5-position interacts with Tyr132 and Lys206 in a way similar to the acyclic 2OG mimetic NOG (Figs. 5/11/12). The regioisomer of 1OX1, 4-carboxy-8-hydroxyquinoline (4C8HQ 55, Fig. 4) binds in an analogous way to 1OX1 by chelating Fe(11) and interacting with Tyr132 and Lys206 (Figs. 5/12). However, in the case of 1OX1 54, the position of the active-site metal is shifted relative to its position in the absence of inhibitor (as observed for a selective KDM6 inhibitor, see below) [53]. This iron translocation may be related to the difference in potency between 1OX1 54 and 4C8HQ 55.

The activity of the 8HQ-based KDM4 inhibitor 59 (Fig. 4) has been reported to be related to suppression of initiation of infection by herpes simplex virus, suppression of propagation of infection to adjacent cells and suppression of reaction in latently infected sensory neurons via inhibition of the initiation stage [136]. These results suggest that chromatin remodelling via small-molecule modulation may be a viable strategy for antiviral therapy at an early stage of infection. More generally they suggest potential of 'epigenetic' therapies for treatment of infectious diseases.

Photoaffinity probes show promising utility for identification and quantification of proteins and mapping molecular interactions in cells [137]. Such a photoaffinity probe (60, Fig. 4) for 2OG oxygenases was developed by coupling 1OX1 to a photoactivatable crosslinking group and biotin as an affinity-purification tag [138]. The resulting probe allowed the identification of KDM2A in HEK293T cells via photo-crosslinking and affinity purification coupled to MS analysis. In an effort to develop this methodology for profiling various 2OG oxygenases from different cell types, the Ugi four-component reaction was exploited to enable empirically guided optimisation of probes for specific tasks by enabling the modular assembly of probe scaffolds with a diverse set of photoreactive groups, detection handles and inhibitor attachment points [139].

3.8. A small-molecule probe for KDM6

The first small-molecule selective probe for the KDM6 subfamily (KDM6A and KDM6B) is GSK-J1 (61, Fig. 6), with its corresponding

cell-active ethyl ester prodrug GSK-J4 (62, Fig. 6) [94]. This active site-binding inhibitor reduced the KDM6A- and KDM6B-mediated lipopolysaccharide-induced proinflammatory cytokine production by human primary macrophages. GSK-J1 chelates the active site Fe(11) and contains a propanoic acid side-chain that mimics 2OG side chain binding: like 1OX1 (see above), GSK-J1 induces movement of the active site metal (Fig. 11). The negative control compound GSK-J2 (63, Fig. 6), is a regioisomer of the pyridine moiety of GSK-J1 which cannot engage in bidentate metal chelation and renders the molecule inactive (for the corresponding prodrug GSK-J5, see 64, Fig. 6). However, GSK-J4 has also been reported to inhibit the KDM5 subfamily in vitro (-20fold less active relative to KDM6 inhibition) further highlighting the difficulties in achieving selectivity for this class of enzymes.

The benzisothiazol-3-one ring system is currently being pursued for KDM5 inhibitor discovery. 2-4(4-Methylphenyl)-1,2-benzisothiazol-3(2H)-one (PB1T, 65, Fig. 7) was recently identified as an inhibitor of KDM5B (1C50 = 3 |jM) through a high-throughput screen programme. Its analogue, Ebselen (66, Fig. 7), also exhibits similar inhibitory activity. PB1T inhibits KDM5A/B/C but does not inhibit KDM6A/B. While benzisothiazol-3-ones have also been shown to inhibit the KDM4 subfamily [112,135], it is possible that selectivity maybe achieved with further medicinal chemistry efforts. PB1T is cell-active and can inhibit transiently overexpressed KDM5B, thereby increasing global levels of H3K4me3. Although the mechanism of action of these compounds is unknown, it is possible that ejection of enzyme-bound zinc may be important for their activity (as observed for inhibition of KDM4A by Ebselen) [112].

3.9. JmjC KDM inhibitors targeting the substrate binding pocket

The JmjC KDMs have very distinct, sometimes overlapping, histone H3 substrate preferences. Emerging work is focusing on the use of histone peptide substrate mimetics as a strategy to achieve inhibitor selectivity for specific (sets of) JmjC KDMs. In one study, a systematic truncation analysis of histone H3K9me3/2 peptides and PHF8/KDM4A/ C turnover demonstrated differential activities amongst the JmjC KDMs for the same H3 sequence [140]. The peptides were then converted to inhibitors by conjugating a promiscuous iron chelator motif, uracil, onto the K9 position of the truncated peptides, and the degree of pep-tide length imparted selectivity for JmjC KDMs recognising the same H3K9 methylation (for representative structure, see 67, Fig. 8).

In another study, N-oxalyl-D-cysteine (DNOC) and a set of histone H3 peptides systematically substituted with cysteine residues were screened by techniques including non-denaturing mass spectrometry for binding to KDM4A and other KDM4 subfamily enzymes. Following identification of mixed disulfides that bound (for a view of a crystal structure on one cross-linked peptide-DNOC conjugate in the active site of KDM4A, see Fig. 10), stable analogues were synthesised (via radical ene reaction) resulting in an inhibitor selective for the KDM4 subfamily over KDM3A, KDM6B and KDM2A (68, Fig. 8) [141]. Moreover, by cross-linking to an H3K36 sequence (which binds to KDM4A-C but not D/E, 69, Fig. 8), it was possible to demonstrate intra-subfamily selectivity (i.e. selectivity for KDM4A-C over KDM4D-E). Overall, these results provide proof of principle that selective inhibition of not only JmjC subfamilies, but also isoform subgroups, is achievable by utilising binding to both 2OG and substrate binding sites.

The development of small molecules that mimic histone substrate binding for the JmjC KDMs is at an early stage. The G9a and G9a-like H3K9 methyltransferase inhibitors B1X-01294 (70, Fig. 8) and E67 (71, Fig. 8) were shown to inhibit the H3K9me2 and H3K27me2 demethylase KDM7A with 1C50 values in the low-micromolar range [142]. The E67-2 analogue (72, Fig. 8) retained potency against KDM7A with >30-fold selectivity over G9a-like methyltransferases. All three compounds demonstrated selectivity for KDM7A over KDM5C. Both E67 and E67-2 displayed reduced cytotoxicity in mouse and human primary fibroblasts compared with B1X-01294. It is proposed

ARTICLE IN PRESS

C.C. Thinnes etal./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx

that these compounds act as H3 substrate analogues as supported by crystallographic analyses of E67 complexed with KDM7A (Fig. 11). These studies imply further work on substrate-competing inhibitors of theJmjC KDMs will be worthwhile.

3.10. inhibition by flavonoids, catechols and other compounds

Flavonoids and catechols are natural products that have been demonstrated to inhibit a number of 2OG oxygenases, including the JmjC KDMs (and many other enzymes) [109,143-145]. A high-throughput screen of the LOPAC bioactive library against KDM4E identified many flavonoids (e.g. myricetin 73, baicalein 74 and epigallocatechin gallate 75, Fig. 9) and catechols (e.g. dopamine 76 and its derivatives, and haematoxylin 77, Fig. 9) as low-micromolar inhibitors [146]. In another study, a catechol series (caffeic acid (78, Fig. 9) and derivatives) from a diversity-optimised natural product library was found to inhibit KDM4C and KDM6A but not PHF8 [147]. Caffeic acid 78 was a hit from a KDM6B HTS [148]. While the metal-chelating properties of flavonoids and catechols is, in part, thought to contribute to KDM inhibition, direct competition with 2OG is not observed, and it is possible that multiple factors contribute to their inhibitory effect. Although these compounds are unlikely to be useful from a pharmaceutical perspective, these results do raise the possibility that natural products obtained from the diet may affect chromatin modification status.

Recently, the natural product Tripartin (79, Fig. 9), which is a dichlorinated indanone isolated from the culture broth of the Strepto-myces sp. associated with a larva of a dung beetle, has been reported as a JmjC KDM inhibitor; cell-based evidence for KDM4 inhibition was presented, though selectivity studies on isolated enzymes have not yet been reported [149].

The majority of JmjC KDM inhibitors identified to date incorporate carboxylic acids/carboxylic acid analogues, leading to use of pro-drug ester forms for sufficient cellular activity. Interestingly, a series of pyrido[1,2-a]indoles are reported to inhibit KDM4C in the sub-micromolar range as the acid, ester, and primary amide forms; the mode of action of these compounds is not yet established (Fig. 9) [150,151].

4. 'Pan' KDM1 and JmjC KDM inhibitors

'Pan'-histone demethylase inhibitors 80 and 81 (Fig. 13) targeting both KDM1 and JmjC KDMs were synthesised in an effort to address the observation that both KDM1 and KDM4 are coexpressed and colocalise with the androgen receptor in prostate cancer [152]. These inhibitors were created by covalently linking the known KDM1 inhibitor tranylcypromine (2, Fig. 2) with the JmjC KDM inhibitor templates 4-carboxy-2,2'-bipyridine (51, Fig. 4), or 1OX1 (54, Fig. 4) [153]. Both 80 and 81 increased H3K4 and H3K9 methylation levels in cells and led to growth arrest and apoptosis in LNCaP prostate and HCT116 colon cancer cell lines; this activity was not observed with single 2, 51 or 54, or a combination of 2 and 51. Further, the inhibitors were observed to cause little or no apoptosis in non-cancerous mesenchymal progenitor (MePR) cells. Thus, although there is undoubtedly

Fig. 13. Structures of pan-KDM inhibitors. Compounds 80 and 81 (as racemates) combine KDM1 inhibitor tranylcypromine 2 with the JmjC KDM inhibitors 4-carboxy-4'-carboxymethoxy-2,2'-pipyridine 51 and IOX1 54 respectively. These dual inhibitors induce cell growth inhibition and apoptosis in prostate and colon cancer cell lines, but not in noncancer mesenchymal progenitor cells.

considerable scope for optimisation, inhibitors targeting sets of both KDM1 and JmjC KDMs, may have potential for cancer-selective applications.

5. Future directions

In this review we hope to have summarised the pioneering progress on the inhibition of the human KDMs and given some sense of the excitement in the field from both basic science and therapeutic perspectives. Over the past decade or so our perceptions of the roles and extent of protein and nucleic acid methylation have changed considerably. A major change has come with the realisation that the methylation of protein, DNA and RNA molecules can be directly and frequently reversed by methyl group oxidation catalysed by enzymes from the amine oxidase and 2OG oxygenase superfamilies. A third mechanism of direct demethylation, i.e. by thiolate-mediated nucleophilic displacement, is involved in nucleic acid repair in both prokaryotes and eukary-otes, but has not been shown to be involved in transcriptional regulation [154].

Inhibition of histone demethylases and methyltransferases has substantial potential for the regulation of gene expression by treatment with small molecules. As outlined in the preceding sections, studies on the selective inhibition of the catalytic domain of both the human KDM1/LSD and theJmjC families of KDMs are progressing rapidly. Although these studies are at a relatively early stage, the signs are that with sufficient medicinal chemistry efforts, it will be possible to make highly potent and selective inhibitors of the catalytic domains of both families of human KDMs. To date most KDM1 and JmjC KDM inhibition efforts have been focused on the extension of known types of inhibitors for other family members to the KDMs, i.e. mechanism-based inhibition of the KDM1s and active site iron chelators for theJmjC KDMs. It is likely that the extension of those methods, including by competing with his-tone substrate binding interactions, will lead to highly selective inhibitors of the catalytic domains. As time progresses it is likely that new types of inhibitors will emerge — such as those involving substrate competition and allosteric interactions. Additionally, although more challenging, the identification of small molecules that promote KDM activity is of potential therapeutic interest and this might be the focus of some future academic efforts.

There is already strong evidence that the biological roles of the KDMs cannot be simply explained as 'erasure' of methylation. All human KDMs are predicted to contain non-catalytic 'binding' domains, and their interaction with nucleosomes/chromatin perhaps might be best viewed as protein-protein/nucleic acid interactions in which the covalent methylation/demethylation processes are a component. Perturbation ofthe interaction ofthe KDM binding domains offers potential for therapeutic manipulation. However, such an approach requires the issue of how best to assay KDM activity in cells.

To date, cell-based assays have mostly focused on antibody-based methods for measuring methylation status, sometimes coupled to 'reporter gene' assays, though not always under physiologically relevant conditions. Data relating to the methylation status associated with particular DNA sequences can also be acquired via chromatin immunopre-cipitation studies. As with studies on other chromatin-modifying enzymes, in few cases the effects of KDM inhibition are directly connected with clear phenotypes at the physiological level. Mutations to the genes encoding for some KDMs are linked to pathophysiological pheno-types — and one interesting avenue will be to attempt to replicate these phenotypes using selective small-molecule inhibitors. Accumulating evidence, however, indicates that multiple post-translational modifications contribute to the emergence of phenotypes including pathophys-iological ones.

Thus, small-molecule 'epigenetic therapies' may be required to inhibit multiple targets - at least for optimal therapeutic benefit - perhaps aimed at modulating sets ofpost-oligomerisation modifications associated with specific regions ofchromatin. Achieving this in a rational manner

ARTICLE IN PRESS

C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

770 will require substantially more knowledge of the chemical details of ge-

771 netic processes than we presently possess and acquiring such detailed

772 knowledge is likely to take some time — in the meantime it will be inter-

773 esting to study the effects of combinations of inhibitors targeting differ-

774 ent regulatory components of the (epi)genetic machinery.

775 Acknowledgements

Q3 We thank the Wellcome Trust, the Biotechnology and Biological

777 Sciences Research Council (BBSRC), European Union, the British Heart

778 Foundation (BHF), Cancer Research UK (CRUK), the BHF Centre of

779 Research Excellence, Oxford, the Structural Genomics Consortium,

780 Pfizer Ltd. and the Royal Society for funding our work on KDMs. RJH ac-

781 knowledges a William R. Miller Junior Research Fellowship (St. Edmund

782 Hall, University of Oxford). We thank all who have contributed to the

783 field of histone KDM inhibition and apologise for incomplete citations

784 due to space limitations.

M. Berdasco, M. Esteller, Aberrant epigenetic landscape in cancer: how cellular identity goes awry, Dev. Cell 19 (2010) 698-711.

T. Suganuma, J.L. Workman, Signals and combinatorial functions of histone modifications, Annu. Rev. Biochem. 80 (2011) 473-499.

G.E. Zentner, S. Henikoff, Regulation of nucleosome dynamics by histone modifications, Nat. Struct. Mol. Biol. 20 (2013) 259-266.

E.L. Greer, Y. Shi, Histone methylation: a dynamic mark in health, disease and inheritance, Nat. Rev. Genet. 13 (2012) 343-357.

A. Portela, M. Esteller, Epigenetic modifications and human disease, Nat. Biotechnol. 28 (2010) 1057-1068.

T. Waldmann, R Schneider, Targeting histone modifications—epigenetics in cancer, Curr. Opin. Cell Biol. 25 (2013) 184-189.

W.S. Xu, R.B. Parmigiani, P.A. Marks, Histone deacetylase inhibitors: molecular mechanisms of action, Oncogene 26 (2007) 5541-5552.

D. Rotili, A. Mai, Targeting histone demethylases: a new avenue for the fight against cancer, Gene Cancer 2 (2011 ) 663-679.

G.L. Gravina, C. Festuccia, F. Marampon, V.M. Popov, R.G. Pestell, B.M. Zani, V. Tombolini, Biological rationale for the use of DNA methyltransferase inhibitors as new strategy for modulation of tumor response to chemotherapy and radiation, Mol. Cancer 9 (2010).

S. Chuikov, J.K Kurash, J.R. Wilson, B. Xiao, N. Justin, G.S. Ivanov, K. McKinney, P. Tempst, C. Prives, S.J. Gamblin, N.A. Barlev, D. Reinberg, Regulation of p53 activity through lysine methylation, Nature 432 (2004) 353-360. D. Levy, AJ. Kuo, Y. Chang, U. Schaefer, C. Kitson, P. Cheung, A. Espejo, B.M. Zee, C.L. Liu, S. Tangsombatvisit, R.I. Tennen, AY. Kuo, S. Tanjing, R Cheung, K.F. Chua, P.J. Utz, X. Shi, R.K. Prinjha, K. Lee, B.A. Garcia, M.T. Bedford, A. Tarakhovsky, X. Cheng, O. Gozani, Lysine methylation of the NF-[kappa ] B subunit RelA by SETD6 couples activity of the histone methyltransferase GLP at chromatin to tonic repression of NF-[kappa] B signaling, Nat. Immunol. 12 (2011) 29-36. C.-K. Ea, D. Baltimore, Regulation of NF-kB activity through lysine monomethylation of p65, Proc. Natl. Acad. Sci. 106 (2009) 18972-18977. A. Barski, S. Cuddapah, K. Cui, T.Y. Roh, D.E. Schones, Z. Wang, G. Wei, I. Chepelev, K. Zhao, High-resolution profiling ofhistone methylations in the human genome, Cell 129 (2007) 823-837.

G.A. Khoury, R.C. Baliban, C.A. Floudas, Proteome-wide post-translational modification statistics: frequency analysis and curation of the Swiss-Prot database, Sci. Rep. 1 (2011).

C.T. Walsh, S. Garneau-Tsodikova, G.J. Gatto Jr., Protein posttranslational modifications: the chemistry of proteome diversifications, Angew. Chem. 44 (2005) 7342-7372.

M. Albert, K. Helin, Histone methyltransferases in cancer, Semin. Cell Dev. Biol. 21 (2010) 209-220.

K.M. Bernt, N. Zhu, A.U. Sinha, S. Vempati, J. Faber, A.V. Krivtsov, Z. Feng, N. Punt, A. Daigle, L. Bullinger, R.M. Pollock, V.M. Richon, A.L. Kung, S.A. Armstrong, MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1L, Cancer Cell 20 (2011) 66-78.

Y. Okada, Q. Feng, Y. Lin, Q.Jiang, Y. Li, V.M. Coffield, L. Su, G. Xu, Y. Zhang, hDOT1L links histone methylation to leukemogenesis, Cell 121 (2005) 167-178. R. Margueron, D. Reinberg, The Polycomb complex PRC2 and its mark in life, Nature 469 (2011) 343-349.

A.P. Bracken, D. Pasini, M. Capra, E. Prosperini, E. Colli, K. Helin, EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer, EMBO J. 22 (2003) 5323-5335.

S. Varambally, S.M. Dhanasekaran, M. Zhou, T.R. Barrette, C. Kumar-Sinha, M.G. Sanda, D. Ghosh, K.J. Pienta, R.G. Sewalt, A.P. Otte, M.A. Rubin, A.M. Chinnaiyan, The polycomb group protein EZH2 is involved in progression of prostate cancer, Nature 419 (2002) 624-629.

S.R. Daigle, E.J. Olhava, C.A. Therkelsen, A. Basavapathruni, L. Jin, P.A. Boriack-Sjodin, C.J. Allain, C.R. Klaus, A. Raimondi, M.P. Scott, N.J. Waters, R. Chesworth,

785 Reft

786 [1]

788 [2]

790 [3]

792 [4]

794 [5]

796 [6]

798 [7]

800 [8]

802 [9]

806 [10]

809 [11]

815 [12]

817 [13]

820 [14]

823 [15]

826 [16]

828 [17]

832 [18]

834 [19]

836 [20]

839 [21]

843 [22]

[23 [24]

[25 [26 [27 [28

[30 [31

[34 [35

[40 [41

[42 [43

[44] [45 [46 [47 [48 [49

M.P. Moyer, R.A. Copeland, V.M. Richon, R.M. Pollock Potent inhibition of DOT1L 845 as treatment of MLL-fusion leukemia, Blood 122 (2013) 1017-1025. 846

www.epizyme.com. 847

M.T. McCabe, H.M. Ott, G. Ganji, S. Korenchuk C. Thompson, G.S. Van Aller, Y. Liu, A. 848 P. Graves, A.D.P. Iii, E. Diaz, L.V. LaFrance, M. Mellinger, C. Duquenne, X. Tian, R.G. 849 Kruger, C.F. McHugh, M. Brandt, W.H. Miller, D. Dhanalk, S.K. Verma, P.J. Tummino, 850 C.L. Creasy, EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2- 851 activating mutations, Nature 492 (2012) 108-112. 852

M.A. Dawson, T. Kouzarides, Cancer epigenetics: from mechanism to therapy, Cell 853 150 (2012) 12-27. 854

L.J. Walport, R.J. Hopkinson, C.J. Schofield, Mechanisms of human histone and 855 nucleic acid demethylases, Curr. Opin. Chem. Biol. 16 (2012) 525-534. 856

S.M. Kooistra, K. Helin, Molecular mechanisms and potential functions of histone 857 demethylases, Nat. Rev. Mol. Cell Biol. 13 (2012) 297-311. 858

Y. Wang, J. Wysocka, J. Sayegh, Y.H. Lee, J.R Perlin, L. Leonelli, L.S. Sonbuchner, C.H. 859 McDonald, R.G. Cook, Y. Dou, R.G. Roeder, S. Clarke, M.R Stallcup, C.D. Allis, S.A. 860 Coonrod, Human PAD4 regulates histone arginine methylation levels via 861 demethylimination, Science 306 (2004) 279-283. 862

G.L. Cuthbert, S. Daujat, A.W. Snowden, H. Erdjument-Bromage, T. Hagiwara, M. 863 Yamada, R. Schneider, P.D. Gregory, P. Tempst, A.J. Bannister, T. Kouzarides, His- 864 tone deimination antagonizes arginine methylation, Cell 118 (2004) 545-553. 865

B. Chang, Y. Chen, Y. Zhao, R.K. Bruick, JMJD6 is a histone arginine demethylase, 866 Science 318 (2007) 444-447. 867 W. Liu, Q.Ma, K. Wong, W. Li, K. Ohgi, J. Zhang, A.K. Aggarwal, M.G. Rosenfeld, Brd4 868 and JMJD6-associated anti-pause enhancers in regulation of transcriptional pause 869 release, Cell 155 (2013) 1581-1595. 870

C.J. Webby, A. Wolf, N. Gromak, M. Dreger, H. Kramer, B. Kessler, M.L. Nielsen, C. 871 Schmitz, D.S. Butler, J.R. Yates III, C.M. Delahunty, P. Hahn, A. Lengeling, M. Mann, 872 N.J. Proudfoot, C.J. Schofield, A. Bottger, Jmjd6 catalyses lysyl-hydroxylation of 873 U2AF65, a protein associated with RNA splicing, Science 325 (2009) 90-93. 874 M. Mantri, N.D. Loik, R.B. Hamed, T.D. Claridge, J.S. McCullagh, C.J. Schofield, The 2- 875 oxoglutarate-dependent oxygenase JMJD6 catalyses oxidation of lysine residues to 876 give 5S-hydroxylysine residues, Chembiochem 12 (2011) 531-534. 877 Y. Shi, Histone lysine demethylases: emerging roles in development, physiology 878 and disease, Nature reviews, Genetics 8 (2007) 829-833. 879 S. Hayami, J.D. Kelly, H.-S. Cho, M. Yoshimatsu, M. Unoki, T. Tsunoda, H.I. Field, D.E. 880 Neal, H. Yamaue, B.A.J. Ponder, Y. Nakamura, R. Hamamoto, Overexpression of 881 LSD1 contributes to human carcinogenesis through chromatin regulation in vari- 882 ous cancers, Int. J. Cancer 128 (2011) 574-586. 883 William J. Harris, X. Huang, James T. Lynch, Gary J. Spencer, James R Hitchin, Y. Li, F. Q4 Ciceri, Julian G. Blaser, Brigit F. Greystoke, Allan M. Jordan, Crispin J. Miller, Donald 885 J. Ogilvie, Tim C.P. Somervaille, The histone demethylase KDM1A sustains the on- 886 cogenic potential of MLL-AF9 leukemia stem cells, Cancer Cell 21 (2012) 473-487. 887 I. Hoffmann, M. Roatsch, M.L. Schmitt, L. Carlino, M. Pippel, W. Sippl, M.Jung, The 888 role of histone demethylases in cancer therapy, Mol. Oncol. 6 (2012) 683-703. 889 F. Forneris, C. Binda, A. Adamo, E. Battaglioli, A. Mattevi, Structural basis of 890 LSD1 -CoREST selectivity in histone H3 recognition, J. Biol. Chem. 282 (2007) 891 20070-20074. 892

E. Metzger, M. Wissmann, N. Yin, J.M. Muller, R. Schneider, A.H. Peters, T. Gunther, 893 R. Buettner, R. Schule, LSD1 demethylates repressive histone marks to promote 894 androgen-receptor-dependent transcription, Nature 437 (2005) 436-439. 895 R.J. Klose, E.M. Kallin, Y. Zhang, JmjC-domain-containing proteins and histone 896 demethylation, Nat. Rev. Genet. 7 (2006) 715-727. 897 M. Tahiliani, P. Mei, R. Fang, T. Leonor, M. Rutenberg, F. Shimizu, J. Li, A. Rao, Y. Shi, 898 The histone H3K4 demethylase SMCX links REST target genes to X-linked mental 899 retardation, Nature 447 (2007) 601-605. 900

D. Castermans, J.R. Vermeesch, J.-P. Fryns, J.G. Steyaert, W.J.M. Van de Ven, J.W.M. 901 Creemers, K. Devriendt, Identification and characterization of the TRIP8 and 902 REEP3 genes on chromosome 10q21.3 as novel candidate genes for autism, Eur. 903 J. Hum. Genet. 15 (2007) 422-431. 904

F. Laumonnier, S. Holbert, N. Ronce, F. Faravelli, S. Lenzner, C.E. Schwartz, J. 905 Lespinasse, H. Van Esch, D. Lacombe, C. Goizet, F. Phan-Dinh Tuy, H. van Bokhoven, 906 J.-P. Fryns, J. Chelly, H.-H. Ropers, C. Moraine, B.C.J. Hamel, S. Briault, Mutations in 907 PHF8 are associated with X linked mental retardation and cleft lip/cleft palate, J. 908 Med. Genet. 42 (2005) 780-786. 909 J. Qiu, G. Shi, Y. Jia, J. Li, M. Wu, J. Li, S. Dong, J. Wong, The X-linked mental retarda- 910 tion gene PHF8 is a histone demethylase involved in neuronal differentiation, Cell 911 Res. 20 (2010) 908-918. 912 A. Thalhammer, W. Aik, E.A.L. Bagg, C.J. Schofield, The potential of 2-oxoglutarate 913 oxygenases acting on nucleic acids as therapeutic targets, Drug Discov. Today 9 914 (2012) e91-e100. 915 Y. Tsukada, J. Fang, H. Erdjument-Bromage, M.E. Warren, C.H. Borchers, P. Tempst, 916 Y. Zhang, Histone demethylation by a family ofJmjC domain-containing proteins, 917 Nature 439 (2006) 811-816. 918 R.J. Hopkinson, R.B. Hamed, N.R. Rose, T.D.W. Claridge, C.J. Schofield, Monitoring 919 the activity of 2-oxoglutarate dependent histone demethylases by NMR spectros- 920 copy: direct observation of formaldehyde, Chembiochem 11 (2010) 506-510. 921 Y. Shi, F. Lan, C. Matson, P. Mulligan, J.R. Whetstine, P.A. Cole, R.A. Casero, Y. Shi, 922 Histone demethylation mediated by the nuclear amine oxidase homolog LSD1, 923 Cell 119 (2004) 941 -953. 924 D. Willmann, S. Lim, S. Wetzel, E. Metzger, A. Jandausch, W. Wilk, M. Jung, I. Forne, 925 A. Imhof, A. Janzer, J. Kirfel, H. Waldmann, R. Schule, R. Buettner, Impairment of 926 prostate cancer cell growth by a selective and reversible lysine-specific 927 demethylase 1 inhibitor, Int. J. Cancer 131 (2012) 2704-2709. 928 C. Loenarz, C.J. Schofield, Expanding chemical biology of 2-oxoglutarate 929 oxygenases, Nat. Chem. Biol. 4 (2008) 152-156. 930

ARTICLE IN PRESS

C.C. Thinnes etal./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx 15

931 [51] J.C. Shih, K. Chen, M.J. Ridd, Monoamine oxidase: from genes to behavior, Annu. [73] T. Maes, I. Tirapu, C. Mascara, A. Ortega, A. Estiarte, N. Valls, J. Castro-Palomino, C. 1017

932 Rev. Neurosci. 22 (1999) 197-217. Buesa Arjol, G. Kurz, Preclinical characterization of a potent and selective inhibitor 1018

933 [52] N.R. Rose, E.C.Y. Woon, G.L Kingham, O.N.F. King, J. Mecinovic, I.J. Clifton, S.S. Ng, J. of the histone demethylase KDM1A for MLL leukemia, ASCO Meeting Abstracts, 31, 1019

934 Talib-Hardy, U. Oppermann, MA. McDonough, C.J. Schofield, Selective inhibitors of 2013, p. e13543. 1020

935 the JMJD2 histone demethylases: combined nondenaturing mass spectrometric [74] J.C. Culhane, L.M Szewczuk X. Liu, G. Da, R. Marmorstein, PA. Cole, A mechanism- 1021

936 screening and crystallographic approaches, J. Med. Chem. 53 (2010) 1810-1818. based inactivator for histone demethylase LSD1, J. Am. Chem. Soc. 128 (2006) 1022

937 [53] R.J. Hopkinson, A. Tumber, C. Yapp, R. Chowdhury, W. Aik K.H. Che, X.S. Li, J.B.L. 4536-4537. 1023

938 Kristensen, O.N.F. King, M.C. Chan, K.K. Yeoh, H. Choi, L.J. Walport, C.C. Thinnes, [75] L.M. Szewczuk, J.C. Culhane, M. Yang, A. Majumdar, H. Yu, P.A. Cole, Mechanistic 1024

939 J.T. Bush, C. Lejeune, A.M. Rydzik, N.R. Rose, E.A. Bagg, M.A. McDonough, T.J. analysis of a suicide inactivator of histone demethylase LSD1, Biochemistry 46 1025

940 Krojer, W.W. Yue, S.S. Ng, L. Olsen, P.E. Brennan, U. Oppermann, S. Muller, R.J. (2007) 6892-6902. 1026

941 Klose, P.J. Ratcliffe, C.J. Schofield, A. Kawamura, 5-Carboxy-8-hydroxyquinoline is [76] R. Federico, L. Leone, M. Botta, C. Binda, R. Angelini, G. Venturini, P. Ascenzi, 1027

942 a broad spectrum 2-oxoglutarate oxygenase inhibitor which causes iron transloca- Inhibition of pig liver and Zea mays L. polyamine oxidase: a comparative study, J. 1028

943 tion, Chem. Sci. 4 (2013) 3110-3117. Enzym Inhib. 16 (2001) 147-155. 1029

944 [54] M.G. Lee, C. Wynder, D.M. Schmidt, D.G. McCafferty, R. Shiekhattar, Histone H3 [77] M. Bianchi, F. Polticelli, P. Ascenzi, M. Botta, R. Federico, P. Mariottini, A. Cona, 1030

945 lysine 4 demethylation is a target of nonselective antidepressive medications, Inhibition of polyamine and spermine oxidases by polyamine analogues, FEBS J. 1031

946 Chem. Biol. 13 (2006) 563-567. 273 (2006) 1115-1123. 1032

947 [55] D.M. Schmidt, D.G. McCafferty, trans-2-Phenylcyclopropylamine is a mechanism- [78] Y. Huang, E. Greene, T. Murray Stewart, AC. Goodwin, S.B. Baylin, P.M. Woster, R.A 1033

948 based inactivator of the histone demethylase LSD1, Biochemistry 46 (2007) Casero, Inhibition of lysine-specific demethylase 1 by polyamine analogues results 1034

949 4408-4416. in reexpression of aberrantly silenced genes, Proc. Natl. Acad. Sci. 104 (2007) 1035

950 [56] F. Forneris, C. Binda, M.A. Vanoni, E. Battaglioli, A. Mattevi, Human histone 8023-8028. 1036

951 demethylase LSD1 reads the histone code, J.Biol. Chem. 280 (2005) 41360-41365. [79] Y. Huang, T.M. Stewart, Y. Wu, S.B. Baylin, L.J. Marton, B. Perkins, R.J. Jones, P.M. 1037

952 [57] J.C. Culhane, D. Wang, P.M. Yen, P.A. Cole, Comparative analysis of small molecules Woster, R.A. Casero Jr., Novel oligoamine analogues inhibit lysine-specific 1038

953 and histone substrate analogues as LSD1 lysine demethylase inhibitors, J. Am. demethylase 1 and induce reexpression of epigenetically silenced genes, Clin. 1039

954 Chem. Soc. 132 (2010)3164-3176. Cancer Res. 15 (2009) 7217-7228. 1040

955 [58] L. Hellerman, V.G. Erwin, Mitochondrial monoamine oxidase. II. Action of various [80] S.K. Sharma, Y. Wu, N. Steinbergs, M.L. Crowley, A.S. Hanson, R.A. Casero, P.M. 1041

956 inhibitors for the bovine kidney enzyme. Catalytic mechanism, J. Biol. Chem. 243 Woster, (Bis)urea and (bis)thiourea inhibitors of lysine-specific demethylase 1 as 1042

957 (1968) 5234-5243. epigenetic modulators, J. Med. Chem. 53 (2010) 5197-5212. 1043

958 [59] C. Binda, M. Li, F. Hubalek, N. Restelli, D.E. Edmondson, A. Mattevi, Insights into the [81 ] H.O.B.W.L.V.N.V. Zhang, T.S.U.TXN.D.S.G. Ye,J.S.U.TXN.D.S.G. Quan,J.S.U.T.X.N.D.S.G. 1044

959 mode of inhibition of human mitochondrial monoamine oxidase B from high- Wang, HISTONE DEMETHYLASE INHIBITORS AND USES THEREOF FOR TREATMENT 1045

960 resolution crystal structures, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 9750-9755. O F CANCER, in: O.B.W.L.V.N.V. Nevada Cancer Institute, S.U.T.X.N.D.S.G. Shenzhen 1046

961 [60] D.R. Patek L Hellerman, Mitochondrial monoamine oxidase. Mechanism ofinhibi- Graduate School Of Peking University (Eds.), vol. US2011/061954, WO, 2012. 1047

962 tion by phenylhydrazine and by aralkylhydrazines. Role of enzymatic oxidation, J. [82] H. Liu, Y. Zheng, Y. Duan, W. Lu, J. Ma, A. Yang, K. Yang, Application of amino dithio 1048

963 Biol. Chem. 249 (1974) 2373-2380. formic ester compound with triazolyl in preparing medicine taking LSD1 (Lysine 1049

964 [61 ] C. Binda, S. Valente, M. Romanenghi, S. Pilotto, R. Cirilli, A. Karytinos, G. Ciossani, O. Specificity Demethylase 1) as target, in, CN, 2013. 1050

965 A. Botrugno, F. Forneris, M. Tardugno, D.E. Edmondson, S. Minucci, A. Mattevi, A. [83] J.R. Hitchin, J. Blagg, R. Burke, S. Burns, M.J. Cockerill, E.E. Fairweather, C. Hutton, 1051

966 Mai, Biochemical, structural, and biological evaluation of tranylcypromine deriva- A.M. Jordan, C. McAndrew, A. Mirza, D. Mould, G.J. Thomson, I. Waddell, D.J. 1052

967 tives as inhibitors of histone demethylases LSD1 and LSD2, J. Am. Chem. Soc. 132 Ogilvie, Development and evaluation of selective, reversible LSD1 inhibitors 1053

968 (2010) 6827-6833. derived from fragments, Med. Chem. Commun. 4 (2013) 1513-1522. 1054

969 [62] M. Yang,J.C. Culhane, L.M. Szewczuk P. Jalili, H.L. Ball, M. Machius, P.A. Cole, H. Yu, [84] J.T. Lynch, M.J. Cockerill,J.R. Hitchin, D.H. Wiseman, T.C. Somervaille, CD86 expres- 1055

970 Structural basis for the inhibition of the LSD1 histone demethylase by the antide- sion as a surrogate cellular biomarker for pharmacological inhibition of the histone 1056

971 pressant trans-2-phenylcyclopropylamine, Biochemistry 46 (2007) 8058-8065. demethylase lysine-specific demethylase 1, Anal. Biochem. 442 (2013) 104-106. 1057

972 [63] S. Mimasu, T. Sengoku, S. Fukuzawa, T. Umehara, S. Yokoyama, Crystal structure of [85] V.Sorna,E.R.Theisen,B. Stephens, S.L. Warner, D.J. Bearss, H. Vankayalapati, S.Sharma, 1058

973 histone demethylase LSD1 and tranylcypromine at 2.25 A, Biochem. Biophys. Res. High-throughput virtual screening identifies novel N'-(1-phenylethylidene)- 1059

974 Commun. 366 (2008) 15-22. benzohydrazides as potent, specific, and reversible LSD1 inhibitors, J. Med. Chem. 1060

975 [64] D.M. Gooden, D.M. Schmidt, JA Pollock A.M. Kabadi, D.G. McCafferty, Facile syn- (2013). Q5

976 thesis of substituted trans-2-arylcyclopropylamine inhibitors of the human histone [86] D. Dhanak, Drugging the cancer genome, Proceedings of the 104th Annual Meeting 1062

977 demethylase LSD1 and monoamine oxidases A and B, Bioorg. Med. Chem. Lett. 18 of the American Association for Cancer Research, 2013. 1063

978 (2008) 3047-3051. [87] K. Majamaa, T. Sasaki, J. Uitto, Inhibition of prolyl hydroxylation during collagen 1064

979 [65] R. Ueda, T. Suzuki, K. Mino, H. Tsumoto, H. Nakagawa, M. Hasegawa, R. Sasaki, T. biosynthesis in human skin fibroblast cultures by ethyl 3,4-dihydroxybenzoate, 1065

980 Mizukami, N. Miyata, Identification of cell-active lysine specific demethylase J. Investig. Dermatol. 89 (1987) 405-409. 1066

981 1-selectiveinhibitors,J.Am. Chem. Soc. 131 (2009) 17536-17537. [88] T.J. Franklin, M. Hitchen, Inhibition of collagen hydroxylation by 2,7,8- 1067

982 [66] S. Mimasu, N. Umezawa, S. Sato, T. Higuchi, T. Umehara, S. Yokoyama, Structurally trihydroxyanthraquinone in embryonic-chick tendon cells, Biochem. J. 261 1068

983 designed trans-2-phenylcyclopropylamine derivatives potently inhibit histone (1989) 127-130. 1069

984 demethylase LSD1/KDM1, Biochemistry 49 (2010) 6494-6503. [89] N.R. Rose, M.A. McDonough, O.N.F. King, A. Kawamura, C.J. Schofield, Inhibition of 1070

985 [67] R. Neelamegam, E.L. Ricq, M. Malvaez, D. Patnaik, S. Norton, S.M. Carlin, I.T. Hill, 2-oxoglutarate dependent oxygenases, Chem. Soc. Rev. 40 (2011) 4364-4397. 1071

986 MA. Wood, S.J. Haggarty, J.M. Hooker, Brain-penetrant LSD1 inhibitors can block [90] C.J. Schofield, P.J. Ratcliffe, Oxygen sensing by HIF hydroxylases, Nat. Rev. Mol. Cell 1072

987 memory consolidation, ACS Chem. Neurosci. 3 (2012) 120-128. Biol. 5 (2004) 343-354. 1073

988 [68] S.A. Oryzen Genomics, Phenylcyclopropylamine Derivatives and their Medical Use, [91] I.J. Clifton, M.A. McDonough, D. Ehrismann, N.J. Kershaw, N. Granatino, C.J. 1074

989 2010. Schofield, Structural studies on 2-oxoglutarate oxygenases and related double- 1075

990 [69] N. JOHNSON, W. (1250 South Collegeville Road, Collegeville, PA, 19426, US), stranded ß-helix fold proteins, J. Inorg. Biochem. 100 (2006) 644-669. 1076

991 KASPAREC, Jiri (1250 South Collegeville Road, Collegeville, PA, 19426, US), MILLER, [92] M.A. McDonough, C. Loenarz, R. Chowdhury, I.J. Clifton, C.J. Schofield, Structural 1077

992 William, Henry (1250 South Collegeville Road, Collegeville, PA, 19426, US), ROUSE, studies on human 2-oxoglutarate dependent oxygenases, Curr. Opin. Struct. Biol. 1078

993 Meagan, B. (1250 South Collegeville Road, Collegeville, PA, 19426, US), SUAREZ, Dom- 20 (2010) 659-672. 1079

994 inic (1250 South Collegeville Road, Collegeville, PA, 19426, US), TIAN, Xinrong (1250 [93] W. Aik, M.A. McDonough, A. Thalhammer, R. Chowdhury, C.J. Schofield, Role of the 1080

995 South Collegeville Road, Collegeville, PA, 19426, US), CYCLOPROPYLAMINES AS LSD1 jelly-roll fold in substrate binding by 2-oxoglutarate oxygenases, Curr. Opin. Struct. 1081

996 INHIBITORS, in, GLAXOSMITHKLINE LLC (One Franklin Plaza, 200 North 16th Street, Biol. 22 (2012) 691-700. 1082

997 Philadelphia, PA, 19102, US),JOHNSON, Neil, W. (1250 South Collegeville Road, [94] L. Kruidenier, C.-w. Chung, Z. Cheng, J. Liddle, K. Che, G. Joberty, M. Bantscheff, C. 1083

998 Collegeville, PA, 19426, US),KASPAREC,Jiri (1250 South Collegeville Road, Collegeville, Bountra, A. Bridges, H. Diallo, D. Eberhard, S. Hutchinson, E. Jones, R. Katso, M. 1084

999 PA, 19426, US),MILLER, William, Henry (1250 South Collegeville Road, Collegeville, Leveridge, P.K. Mander, J. Mosley, C. Ramirez-Molina, P. Rowland, C.J. Schofield, 1085

1000 PA, 19426, US),ROUSE, Meagan, B. (1250 South Collegeville Road, Collegeville, PA, R.J. Sheppard, J.E. Smith, C. Swales, R. Tanner, P. Thomas, A. Tumber, G. Drewes, 1086

1001 19426, US),SUAREZ, Dominic (1250 South Collegeville Road, Collegeville, PA, 19426, U. Oppermann, D.J. Patel, K. Lee, D.M. Wilson, A selective Jumonji H3K27 1087

1002 US),TIAN, Xinrong (1250 South Collegeville Road, Collegeville, PA, 19426, US), 2012. demethylase inhibitor modulates the proinflammatory macrophage response, 1088

1003 [70] T.C.S.F.E.C.d.L. Maes, M.C.S.F.E.C.d.L. Martinell Pedemonte, J.C.C.S.F.E.C.d.L. Nature 488 (2012) 404-408. 1089

1004 Castro-Palomino Laria, LYSINE DEMETHYLASE INHIBITORS SUCH AS [95] L. Poppe, C.M. Tegley, V. Li, J. Lewis, J. Zondlo, E. Yang, R.J.M. Kurzeja, R. Syed, 1090

1005 CYCLYLCYLOPROPAMINE DERIVATIVES FOR USE IN THE TREATMENT OF INFLAM- Different modes of inhibitor binding to prolyl hydroxylase by combined use of 1091

1006 MATORY DISEASES OR CONDITIONS, in: S.A.C.S.F.E.C.d.L. Oryzon Genomics (Ed.), X-ray crystallography and NMR spectroscopy of paramagnetic complexes, J. Am. 1092

1007 vol. EP2012/059377, WO, 2012. Chem. Soc. 131 (2009) 16654-16655. 1093

1008 [71] S.V.S.M.I.O. Minucci, A.V.J.F.I.R. Mai, A.V.M.I.P. Mattevi, TRANYLCYPROMINE [96] B. Bleijlevens, T. Shivarattan, E. Flashman, Y. Yang, P.J. Simpson, P. Koivisto, B. 1094

1009 DERIVATIVES AS INHIBITORS OF HISTONE DEMETHYLASE LSD1 AND/OR LSD2, Sedgwick C.J. Schofield, S.J. Matthews, Dynamic states of the DNA repair enzyme 1095

1010 in: P.A.M.I.R. UniversitÄ Degli Studi Di Roma "La Sapienza, V.F.I.M. Fondazione AlkB regulate product release, EMBO Rep. 9 (2008) 872-877. 1096

1011 Ieo, C.S.N.I.P. UniversitÄ Degli Studi Di Pavia, V.F.d.P.I.M. UniversitÄ Degli Studi Di [97] G. Opocher, F. Schiavi, Functional consequences of succinate dehydrogenase 1097

1012 Milano (Eds.), vol. EP2011/055990, WO, 2011. mutations, Endocr. Pract. 17 (2011) 64-71. 1098

1013 [72] W.J. Harris, X. Huang, J.T. Lynch, G.J. Spencer, J.R. Hitchin, Y. Li, F. Ciceri, J.G. Blaser, [98] L Dang, D.W. White, S. Gross, B.D. Bennett, MA Bittinger, E.M. Driggers, V.R Fantin, H. 1099

1014 B.F. Greystoke, AM. Jordan, C.J. Miller, D.J. Ogilvie, T.C. Somervaille, The histone G.Jang, S.Jin, M.C. Keenan, K.M. Marks, R.M. Prins, P.S. Ward, K.E. Yen, L.M. Liau,J.D. 1100

1015 demethylase KDM1A sustains the oncogenic potential of MLL-AF9 leukemia stem Rabinowitz, L.C. Cantley, C.B. Thompson, M.G. Vander Heiden, S.M. Su, Cancer- 1101

1016 cells, Cancer Cell 21 (2012)473-487. associated IDH1 mutations produce 2-hydroxyglutarate, Nature 462 (2009) 739-744. 1102

ARTICLE IN PRESS

16 C.C. Thinnes et al. / Biochimica et Biophysica Acta xxx (2014) xxx-xxx

1103 [99] J.-A. Losman, R.E. Looper, P. Koivunen, S. Lee, R.K. Schneider, C. McMahon, G.S. Sasaki, T. Mizukami, N. Miyata, Design, synthesis, enzyme-inhibitory activity, and 1189

1104 Cowley, D.E. Root, B.L. Ebert, W.G. Kaelin, (R)-2-Hydroxyglutarate is sufficient to effect on human cancer cells of a novel series of Jumonji domain-containing 1190

1105 promote leukemogenesis and its effects are reversible, Science 339 (2013) protein 2 histone demethylase inhibitors, J. Med. Chem. 53 (2010) 5629-5638. 1191

1106 1621-1625. [124] N.R Rose, E.C.Y. Woon, A. Tumber, LJ. Walport, R Chowdhury, X.S. Li, O.N.F. King, C. Q7

1107 [100] S.Zhao,Y.Lin,W.Xu,W.Jiang,Z.Zha,P. Wang, W.Yu,Z.Li,L.Gong, Y.Peng,J.Ding, Lejeune, S.S. Ng, T. Krojer, M.C. Chan, A.M. Rydzik, RJ. Hopkinson, K.H. Che, M. 1193

1108 Q.Lei, K.L. Guan, Y. Xiong, Glioma-derived mutations in IDH1 dominantly inhibit Daniel, C. Strain-Damerell, C. Gileadi, G. Kochan, I.K.H. Leung, J. Dunford, K.K. 1194

1109 IDH1 catalytic activity and induce HIF-1alpha, Science 324 (2009) 261-265. Yeoh, P.J. Ratcliffe, N. Burgess-Brown, D.F. von, S. Muller, B. Marsden, P.E. 1195

1110 [101] U. Puistola, T.M. Turpeenniemihujanen, R. Myllyla, K.I. Kivirikko, Studies on the Brennan, M.A. McDonough, U. Oppermann, RJ. Klose, C.J. Schofield, A. 1196

1111 lysyl hydroxylase reaction. 1. Initial velocity kinetics and related aspects, Biochim Kawamura, Plant growth regulator daminozide is a selective inhibitor of human 1197

1112 Biophys. Acta 611 (1980)40-50. KDM2/7 histone demethylases, J. Med. Chem. 55 (2012) 6639-6643. 1198

1113 [102] R. Myllyla, L. Tuderman, K.I. Kivirikko, Mechanism ofprolyl hydroxylase reaction. 2. [125] T. Suzuki, H. Ozasa, Y. Itoh, P. Zhan, H. Sawada, K. Mino, L. Walport, R. Ohkubo, A. 1199

1114 Kinetic-analysis of reaction sequence, Eur. J. Biochem. 80 (1977) 349-357. Kawamura, M. Yonezawa, Y. Tsukada, A. Tumber, H. Nakagawa, M. Hasegawa, R 1200

1115 [103] M. Xiao, H. Yang, W. Xu, S. Ma, H. Lin, H. Zhu, L Liu, Y. Liu, C. Yang, Y. Xu, S. Zhao, D. Sasaki, T. Mizukami, C.J. Schofield, N. Miyata, Identification of the KDM2/7 histone 1201

1116 Ye, Y. Xiong, K.L. Guan, Inhibition of alpha-KG-dependent histone and DNA lysine demethylase subfamily inhibitor and its antiproliferative activity, J. Med. 1202

1117 demethylases by fumarate and succinate that are accumulated in mutations of Chem. 56 (2013) 7222-7231. 1203

1118 FH and SDH tumor suppressors, Genes Dev. 26 (2012) 1326-1338. [126] X. Luo, Y. Liu, S. Kubicek,J. Myllyharju, A. Tumber, S. Ng, K.H. Che, J. Podoll, T.D. 1204

1119 [104] N.R. Rose, S.S. Ng, J. Mecinovic, B.M.R. Lienard, S.H. Bello, Z. Sun, M.A. McDonough, Heightman, U. Oppermann, S.L. Schreiber, X. Wang, A selective inhibitor and 1205

1120 U. Oppermann, C.J. Schofield, Inhibitor scaffolds for 2-oxoglutarate-dependent probe of the cellular functions of Jumonji C domain-containing histone 1206

1121 histone lysine demethylases, J. Med. Chem. 51 (2008) 7053-7056. demethylases, J. Am. Chem. Soc. 133 (2011) 9451-9456. 1207

1122 [105] R Chowdhury, K.K. Yeoh, Y.M. Tian, L Hillringhaus, E.A. Bagg, N.R. Rose, I.K. Leung, [127] R.G.S. Brown, H. Kawaide, Y.-Y. Yang, W. Rademacher, Y. Kamiya, Daminozide and 1208

1123 X.S. Li, E.C. Woon, M. Yang, MA. McDonough, O.N. King, I.J. Clifton, RJ. Klose, T.D. prohexadione have similar modes of action as inhibitors of the late stages of 1209

1124 Claridge, P.J. Ratcliffe, C.J. Schofield, A. Kawamura, The oncometabolite gibberellin metabolism, Physiol. Plant. 101 (1997) 309-313. 1210

1125 2-hydroxyglutarate inhibits histone lysine demethylases, EMBO Rep. 12 (2011) [128] M.M. Mackeen, H.B. Kramer, K.-H. Chang, M.L Coleman, RJ. Hopkinson, C.J. Schofield, 1211

1126 463-469. B.M. Kessler, Small-molecule-based inhibition of histone demethylation in cells 1212

1127 [106] W. Xu, H. Yang, Y. Liu, Y. Yang, P. Wang, S.H. Kim, S. Ito, C. Yang, M.T. Xiao, L.X. Liu, assessed by quantitative mass spectrometry, J. Proteome Res. 9 (2010)4082-4092. 1213

1128 W.Q.Jiang,J.Liu,J.Y.Zhang,B.Wang, S.Frye,Y. Zhang,Y.H.Xu, Q.Y.Lei,K.L.Guan, S. [129] A.Thalhammer,J. Mecinovic, C. Loenarz, A. Tumber, N.R. Rose, T.D. Heightman, C.J. 1214

1129 M. Zhao, Y. Xiong, Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of Schofield, Inhibition of the histone demethylase JMJD2E by 3-substituted pyridine 1215

1130 alpha-ketoglutarate-dependent dioxygenases, Cancer Cell 19 (2011) 17-30. 2,4-dicarboxylates, Org. Biomol. Chem. 9 (2011) 127-135. 1216

1131 [107] J.-F. Couture, E. Collazo, P.A. Ortiz-Tello, J.S. Brunzelle, R.C. Trievel, Specificity and [130] M.D. Barker, M. Campbell, H. Diallo, C. Douault, P. Humphreys, J. Liddle, RJ. 1217

1132 mechanism ofJMJD2A, a trimethyllysine-specific histone demethylase, Nat. Struct. Sheppard, P.J. Thomas, D.M. Wilson, Pyridine-4-carboxylic acids as demethylase 1218

1133 Mol. Biol. 14 (2007) 689-695. enzymes inhibitors and their preparation, in, Glaxo Group Limited, UK, 2013. 1219

1134 [108] J.C. Price, E.W. Barr, L.M. Hoffart, C. Krebs, J.M. Bollinger, Kinetic dissection of (124pp). 1220

1135 the catalytic mechanism of taurine:a-ketoglutarate dioxygenase (TauD) from [131] L. Wang, J. Chang, D. Varghese, M. Dellinger, S. Kumar, A.M. Best, J. Ruiz, R Bruick, S. 1221

1136 Escherichia colif, Biochemistry 44 (2005) 8138-8147. Pena-Llopis,J.Xu, D.J. Babinski, D.E. Frantz, R.A. Brekken, A.M. Quinn, A. Simeonov, J. 1222

1137 [109] R.W. Welford, I. Schlemminger, L.A. McNeill, K.S. Hewitson, C.J. Schofield, The Easmon, E.D. Martinez, A small molecule modulates Jumonji histone demethylase 1223

1138 selectivity and inhibition of AlkB, J. Biol. Chem. 278 (2003) 10157-10161. activity and selectively inhibits cancer growth, Nat. Commun. 4 (2013) 2035. 1224

1139 [110] A.M. Rydzik I.K. Leung, G.T. Kochan, A. Thalhammer, U. Oppermann, T.D. Claridge, [132] J. Hurych, M. Chvapil, Influence of chelating agents on the biosynthesis of collagen, 1225

1140 C.J. Schofield, Development and application of a fluoride-detection-based fluores- Biochim. Biophys. Acta Gen. Subj. 97 (1965) 361-363. 1226

1141 cence assay for gamma-butyrobetaine hydroxylase, Chembiochem 13 (2012) [133] N.J. Hales, J.F. Beattie, Novel inhibitors of prolyl 4-hydroxylase. 5. The intriguing 1227

1142 1559-1563. structure-activity relationships seen with 2,2'-bipyridine and its 5,5'-dicarboxylic 1228

1143 [111] E.D. MacKenzie, M.A. Selak, D.A. Tennant, L.J. Payne, S. Crosby, C.M. Frederiksen, acid derivatives, J. Med. Chem. 36 (1993) 3853-3858. 1229

1144 D.G. Watson, E. Gottlieb, Cell-permeating alpha-ketoglutarate derivatives alleviate [134] K.-H. Chang, O.N.F. King, A. Tumber, E.C.Y. Woon, T.D. Heightman, M.A. 1230

1145 pseudohypoxia in succinate dehydrogenase-deficient cells, Mol. Cell. Biol. 27 McDonough, C.J. Schofield, N.R. Rose, Inhibition of histone demethylases by 1231

1146 ( 2007) 3282-3289. 4-carboxy-2,2'-bipyridyl compounds, ChemMedChem 6 (2011) 759-764. 1232

1147 [112] R Sekirnik N.R. Rose, A. Thalhammer, P.T. Seden, J. Mecinovic, C.J. Schofield, [135] O.N.F. King, X.S. Li, M. Sakurai, A. Kawamura, N.R. Rose, S.S. Ng, A.M. Quinn, G. Rai, 1233

1148 Inhibition of the histone lysine demethylase JMJd2a by ejection of structural B.T. Mott, P. Beswick, RJ. Klose, U. Oppermann, A. Jadhav, T.D. Heightman, D.J. 1234

1149 Zn(ii), Chem. Commun. (2009) 6376-6378. Maloney, C.J. Schofield, A. Simeonov, Quantitative high-throughput screening 1235

1150 [113] E.C. Woon, M. Demetriades, E.A. Bagg, W. Aik S.M. Krylova, J.H. Ma, M. Chan, L.J. identifies 8-hydroxyquinolines as cell-active histone demethylase inhibitors, 1236

1151 Walport, D.W. Wegman, K.N. Dack, M.A. McDonough, S.N. Krylov, C.J. Schofield, PLoS ONE 5 (2010) e15535. 1237

1152 Dynamic combinatorial mass spectrometry leads to inhibitors of a 2-oxoglutarate- [136] Y. Liang, J.L. Vogel, J.H. Arbuckle, G. Rai, A. Jadhav, A. Simeonov, D.J. Maloney, T.M. 1238

1153 dependent nucleic acid demethylase, J. Med. Chem. 55 (2012) 2173-2184. Kristie, Targeting the JMJD2 histone demethylases to epigenetically control 1239

1154 [114] B.Chen,F.Ye,L.Yu,G.Jia,X.Huang,X.Zhang,S.Peng,KChen,M. Wang,S.Gong,R. herpesvirus infection and reactivation from latency, Sci. Transl. Med. 5 (2013) 1240

1155 Zhang,J. Yin, H. Li, Y. Yang, H. Liu, J. Zhang, H. Zhang, A. Zhang, H. Jiang, C. Luo, C.G. (167ra165,111 pp). 1241

1156 Yang, Development of cell-active N6-methyladenosine RNA demethylase FTO in- [137] P.P. Geurink L.M. Prely, G.A. van der Marel, R. Bischoff, H.S. Overkleeft, 1242

1157 hibitor, J. Am. Chem. Soc. 134 (2012) 17963-17971. Photoaffinity labeling in activity-based protein profiling, in: S.A. Sieber (Ed.), 1243

1158 [115] W. Aik, M. Demetriades, M.K. Hamdan, E.A. Bagg, K.K. Yeoh, C. Lejeune, Z. Zhang, Activity-Based Protein Profiling, vol. 324, 2012, pp. 85-113. 1244

1159 MA. McDonough, C.J. Schofield, Structural basis for inhibition of the fat mass and [138] D. Rotili, M. Altun, A. Kawamura, A. Wolf, R. Fischer, Ivanhoe K.H. Leung, Mukram 1245

1160 obesity associated protein (FTO), J. Med. Chem. 56 (2013) 3680-3688. M. Mackeen, Y.-m. Tian, Peter J. Ratcliffe, A. Mai, Benedikt M. Kessler, Christopher J. 1246

1161 [116] C.J. Cunliffe, T.J. Franklin, N.J. Hales, G.B. Hill, Novel inhibitors of prolyl 4- Schofield, A photoreactive small-molecule probe for 2-oxoglutarate oxygenases, 1247

1162 hydroxylase. 3. Inhibition by the substrate analog N-oxaloglycine and its deriva- Chem. Biol. 18 (2011) 642-654. 1248

1163 tives,J. Med. Chem. 35 (1992) 2652-2658. [139] J.T. Bush, L.J. Walport, J.F. McGouran, I.K.H. Leung, G. Berridge, S.S. van Berkel, A 1249

1164 [117] D.R Mole, I. Schlemminger, L.A McNeill, K.S. Hewitson, C.W. Pugh, P.J. Ratcliffe, C.J. Basak, B.M. Kessler, C.J. Schofield, The Ugi four-component reaction enables 1250

1165 Schofield, 2-Oxoglutarate analogue inhibitors of hif prolyl hydroxylase, Bioorg. expedient synthesis and comparison of photoaffinity probes, Chem. Sci. (2013). Q8

1166 Med. Chem. Lett. 13 (2003) 2677-2680. [140] B. Lohse, A.L. Nielsen, J.B.L. Kristensen, C. Helgstrand, P.A.C. Cloos, L. Olsen, M. 1252 Q6 [118] Erinn B. Rankin, C. Wu, R. Khatri, Tremika L.S. Wilson, R Andersen, E. Araldi, Gajhede, R.P. Clausen, J.L. Kristensen, Targeting histone lysine demethylases by 1253

1168 Andrew L. Rankin, J. Yuan, Calvin J. Kuo, E. Schipani, Amato J. Giaccia, The HIF sig- truncating the histone 3 tail to obtain selective substrate-based inhibitors, 1254

1169 naling pathway in osteoblasts directly modulates erythropoiesis through the pro- Angew. Chem. Int. Ed. 50 (2011) 9100-9103. 1255

1170 duction of EPO, Cell 149 (2012) 63-74. [141] E.C.Y. Woon, A. Tumber, A. Kawamura, L. Hillringhaus, W. Ge, N.R. Rose, J.H.Y. Ma, 1256

1171 [119] Y.M. Tian, K.K. Yeoh, M.K. Lee, T. Eriksson, B.M. Kessler, H.B. Kramer, M.J. Edelmann, M.C. Chan, L.J. Walport, K.H. Che, S.S. Ng, B.D. Marsden, U. Oppermann, M.A. 1257

1172 C. Willam, C.W. Pugh, C.J. Schofield, P.J. Ratcliffe, Differential sensitivity of hypoxia McDonough, C.J. Schofield, Linking of 2-oxoglutarate and substrate binding sites 1258

1173 inducible factor hydroxylation sites to hypoxia and hydroxylase inhibitors, J. Biol. enables potent and highly selective inhibition of JmjC histone demethylases, 1259

1174 Chem. 286 (2011) 13041-13051. Angew. Chem. Int. Ed. 51 (2012) 1631-1634. 1260

1175 [120] M.A McDonough, L.A. McNeill, M. Tilliet, CA. Papamicael, Q.-Y. Chen, B. Banerji, K. [142] A.K. Upadhyay, D. Rotili,J.W. Han, R Hu, Y. Chang, D. Labella, X. Zhang, Y.S. Yoon, A. 1261

1176 S. Hewitson, C.J. Schofield, Selective inhibition of factor inhibiting hypoxia- Mai, X. Cheng, An analog of BIX-01294 selectively inhibits a family of histone H3 1262

1177 inducible factor, J. Am. Chem. Soc. 127 (2005) 7680-7681. lysine 9 Jumonji demethylases, J. Mol. Biol. 416 (2012) 319-327. 1263

1178 [121] S.S. Ng, K.L. Kavanagh, M.A. McDonough, D. Butler, E.S. Pilka, B.M.R Lienard, J.E. [143] J.C. Murray, R.H. Cassell, S.R Pinnell, Inhibition of lysyl hydroxylase by catechol 1264

1179 Bray, P. Savitsky, O. Gileadi, F. von Delft, N.R. Rose, J. Offer, J.C. Scheinost, T. analogs, Biochim. Biophys. Acta 481 (1977) 63-70. 1265

1180 Borowski, M. Sundstrom, C.J. Schofield, U. Oppermann, Crystal structures of his- [144] W.J. Wilson, L. Poellinger, The dietary flavonoid quercetin modulates HIF-1 alpha 1266

1181 tone demethylase JMJD2A reveal basis for substrate specificity, Nature 448 activity in endothelial cells, Biochem. Biophys. Res. Commun. 293 (2002)446-450. 1267

1182 (2007) 87-91. [145] Y.D. Zhou, Y.P. Kim, X.C. Li, S.R. Baerson, A.K. Agarwal, T.W. Hodges, D. Ferreira, D.G. 1268

1183 [122] S. Hamada, T.-D. Kim, T. Suzuki, Y. Itoh, H. Tsumoto, H. Nakagawa, R Janknecht, N. Nagle, Hypoxia-inducible factor-1 activation by (-)-epicatechin gallate: potential 1269

1184 Miyata, Synthesis and activity of N-oxalylglycine and its derivatives as Jumonji adverse effects of cancer chemoprevention with high-dose green tea extracts, J. 1270

1185 C-domain-containing histone lysine demethylase inhibitors, Bioorg. Med. Chem. Nat. Prod. 67 (2004) 2063-2069. 1271

1186 Lett. 19 (2009) 2852-2855. [146] M. Sakurai, N.R. Rose, L. Schultz, A.M. Quinn, A. Jadhav, S.S. Ng, U. Oppermann, C.J. 1272

1187 [123] S. Hamada, T. Suzuki, K. Mino, K. Koseki, F. Oehme, I. Flamme, H. Ozasa, Y. Itoh, D. Schofield, A. Simeonov, A miniaturized screen for inhibitors of Jumonji histone 1273

1188 Ogasawara, H. Komaarashi, A. Kato, H. Tsumoto, H. Nakagawa, M. Hasegawa, R demethylases, Mol. BioSyst 6 (2010) 357-364. 1274

ARTICLE IN PRESS

C.C. Thinnes et al./ Biochimica et Biophysica Acta xxx (2014) xxx-xxx

1275 [147] A.L. Nielsen, L.H. Kristensen, K.B. Stephansen, J.B. Kristensen, C. Helgstrand, M. Lees,

1276 P. Cloos, K. Helin, M. Gajhede, L. Olsen, Identification of catechols as histone-lysine

1277 demethylase inhibitors, FEBS Lett. 586 (2012) 1190-1194.

1278 [148] J. Sayegh,J. Cao, M.R. Zou, A. Morales, L.P. Blair, M. Norcia, D. Hoyer, AJ. Tackett,J.S.

1279 Merkel, Q. Yan, Identification of small molecule inhibitors of Jumonji AT-rich inter-

1280 active domain 1B (JARID1B) histone demethylase by a sensitive high throughput

1281 screen, J. Biol. Chem. 288 (2013) 9408-9417.

1282 [149] S.-H. Kim, S.H. Kwon, S.-H. Park, J.K. Lee, H.-S. Bang, S.-J. Nam, H.C. Kwon, J. Shin,

1283 D.-C. Oh, Tripartin, a histone demethylase inhibitor from a bacterium associated

1284 with a dung beetle larva, Org. Lett. 15 (2013) 1834-1837.

1285 [150] M.M.W.B.R.N.J. Labelle, C.A.G.N.R.R.W.O.O.X.D.S. Montalbetti, R.J.S.H.C.D.O.O.X.U.U.

1286 Mears, J.L.G.W.O.O.X.N.Y. Vile, D.M.F.A.M.E.V. Gelman, X.B.D.W.O.O.X.B.H. Gai,

1287 O.R.C.C.L.W.O.O.X.N.S. Barker, H.J.C.O.C.D.O.O.X.A.X. Deboves, S.P.T.C.W.O.OXB.B.

1288 East, T.L.t.D.K.C.0. Boesen, INHIBITORS OF HDME, in: O.M.V.D.K.C.N. Epitherapeutics

1289 Aps (Ed.), vol. DK2011/050280, WO, 2012.

1290 [151 ] M.M.W.B.R.N.J. Labelle, C.A.G.N.R.R.W.O.O.X.D.S. Montalbetti, R.J.S.H.C.D.O.O.X.U.U.

1291 Mears, INHIBITORS OF HDME, in: O.M.V.D.K.C.N. Epitherapeutics Aps (Ed.), vol.

1292 DK2011/050281, WO, 2012.

Q9 [152] M. Wissmann, N. Yin, J.M. Mueller, H. Greschik B.D. Fodor, T. Jenuwein, C. Vogler, R.

1294 Schneider, T. Guenther, R. Buettner, E. Metzger, R. Schuele, Cooperative demethylation

1295 by JMJD2C and LSD1 promotes androgen receptor-dependent gene expression, Nat.

1296 Cell Biol. 9 (2007) 347-U176. 1317

[153] D. Rotili, S. Tomassi, M. Conte, R. Benedetti, M. Tortorici, G. Ciossani, S. Valente, B. 1297 Marrocco, D. Labella, E. Novellino, A. Mattevi, L. Altucci, A. Tumber, C. Yapp, 1298 O.N.F. King, R.J. Hopkinson, A. Kawamura, C.J. Schofield, A. Mai, Pan-histone 1299 demethylase inhibitors simultaneously targeting Jumonji C and lysine-specific 1300 demethylases display high anticancer activities, J. Med. Chem. 57 (2014) 42-55. 1301

[154] B. Sedgwick, P.A. Bates, J. Paik, S.C. Jacobs, T. Lindahl, Repair of alkylated DNA: 1302 recent advances, DNA Repair 6 (2007) 429-442. 1303

[155] M. Yang, J.C. Culhane, L.M. Szewczuk C.B. Gocke, CA. Brautigam, D.R. Tomchick, M. 1304 Machius, P.A. Cole, H. Yu, Structural basis of histone demethylation by LSD1 1305 revealed by suicide inactivation, Nat. Struct. Mol. Biol. 14 (2007) 535-539. 1306

[156] R. Fang, F. Chen, Z. Dong, D. Hu, AJ. Barbera, E.A. Clark J. Fang, Y. Yang, P. Mei, M. 1307 Rutenberg, Z. Li, Y. Zhang, Y. Xu, H. Yang, P. Wang, M.D. Simon, Q. Zhou, J. Li, 1308 M.P. Marynick X. Li, H. Lu, U.B. Kaiser, R.E. Kingston, Y. Xu, Y.G. Shi, LSD2/KDM1B 1309 and its cofactor NPAC/GLYR1 endow a structural and molecular model for 1310 regulation of H3K4 demethylation, Mol. Cell 49 (2013) 558-570. 1311

[157] T. Sengoku, S. Yokoyama, Structural basis for histone H3 Lys 27 demethylation by 1312 UTX/KDM6A, Genes Dev. 25 (2011) 2266-2277. 1313

[158] J.R. Horton, A.K. Upadhyay, H.H. Qi, X. Zhang, Y. Shi, X.-D. Cheng, Enzymatic and 1314 structural insights for substrate specificity of a family of Jumonji histone lysine 1315 demethylases, Nat. Struct. Mol. Biol. 17 (2010) 38-43. 1316