Pyrido[3,4-d]pyrimidin-4(3H)-one metabolism mediated by aldehyde oxidase is blocked by C2-substitution

Article abstract of DOI:10.1080/00498254.2016.1230245

1.We have previously described C8-substituted pyrido[3,4-d]pyrimidin-4(3H)-one derivatives as cell permeable inhibitors of the KDM4 and KDM5 subfamilies of JmjC histone lysine demethylases. 2.Although exemplar compound 1 exhibited moderate clearance in mo

Full text of DOI:10.1080/00498254.2016.1230245

Xenobiotica
the fate of foreign compounds in biological systems
ISSN: 0049-8254 (Print) 1366-5928 (Online) Journal homepage: http://www.tandfonline.com/loi/ixen20
Pyrido[3,4-d]pyrimidin-4(3H)-one metabolism
mediated by aldehyde oxidase is blocked by C2-
substitution
Angela Hayes, N. Yi Mok, Manjuan Liu, Ching Thai, Alan T Henley, Butrus
Atrash, Rachel M Lanigan, Jimmy Sejberg, Yann-Vaï Le Bihan, Vassilios
Bavetsias, Julian Blagg & Florence I Raynaud
To cite this article: Angela Hayes, N. Yi Mok, Manjuan Liu, Ching Thai, Alan T Henley,
Butrus Atrash, Rachel M Lanigan, Jimmy Sejberg, Yann-Vaï Le Bihan, Vassilios Bavetsias,
Julian Blagg & Florence I Raynaud (2016): Pyrido[3,4-d]pyrimidin-4(3H)-one metabolism
mediated by aldehyde oxidase is blocked by C2-substitution, Xenobiotica, DOI:
10.1080/00498254.2016.1230245
To link to this article: http://dx.doi.org/10.1080/00498254.2016.1230245
Accepted author version posted online: 12
Sep 2016.
Published online: 12 Sep 2016.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ixen20
Download by: [La Trobe University]
Date: 13 September 2016, At: 00:03

Pyrido[3,4-d]pyrimidin-4(3H)-one metabolism mediated by aldehyde
oxidase is blocked by C2-substitution
Angela Hayes, N. Yi Mok, Manjuan Liu, Ching Thai, Alan T Henley, Butrus
Atrash, Rachel M Lanigan, Jimmy Sejberg, Yann-Vaï Le Bihan, Vassilios
Bavetsias, Julian Blagg, Florence I Raynaud.
1Cancer Research UK Cancer Therapeutics Unit, The Institute of Cancer
Research, London, UK.
Corresponding author: Dr FI Raynaud 1Cancer Research UK Cancer
Therapeutics Unit, The Institute of Cancer Research, 15 Cotswold Road,
Sutton, Surrey, SM2 5NG, UK. Florence.Raynaud@icr.ac.uk
Keywords: Aldehyde oxidase, metabolism, Pyrido[3,4-d]pyrimidin-4(3H)-one

1.We have previously described C8-substituted pyrido[3,4-d]pyrimidin-4(3H)-
one derivatives as cell permeable inhibitors of the KDM4 and KDM5 subfamilies
of JmjC histone lysine demethylases.
2.Although exemplar compound 1 exhibited moderate clearance in mouse liver
microsomes, it was highly cleared in vivo due to metabolism by aldehyde
oxidase (AO). Similar human and mouse AO-mediated metabolism was
observed with the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold and other C8-
substituted derivatives.
3.We identified the C2-position as the oxidation site by LC-MS and 1H-NMR and
showed that C2-substituted derivatives are no longer AO substrates.
4.In addition to the experimental data, these observations are supported by
molecular modelling studies in the human AO protein crystal structure.

Introduction
Methylation and demethylation of lysine residues on the histone H3 tail
constitute important epigenetic modifications that differentiate transcriptionally
active and inactive chromatin (Shi et al., 2007; Siau et al., 2013; Kooistra et al.,
2012). These processes are highly deregulated in cancer offering opportunities
for drug discovery and development (Helin et al., 2013; Lohse et al., 2011;
Hoffman et al., 2012). It has been demonstrated that the KDM4 and KDM5
subfamilies of Jumonji C (JmjC) histone lysine demethylases are involved in
oncogenesis and drug resistance and their overexpression is commonly
observed in various human tumour types (Young et al., 2013; Berry et al., 2013;
Liu et al. 2009; Kawasu et al. 2011; Li et al. 2011; Walters et al., 2014, Yang et
al., 2015; Blair et al. 2011; Sharma et al.; 2010; Roesch et al., 2013). Cell
penetrant inhibitors of the KDM4 and KDM5 subfamilies have been reported
(Westaway et al., 2016), and we are currently evaluating the potential of dual
KDM4 and KDM5 subfamily small-molecule pyrido[3,4-d]pyrimidin-4(3H)-one-
based inhibitors as anticancer agents (Bavetsias et al., 2016).
In addition to evaluating the potency of our inhibitors on various JmjC histone
lysine demethylases, our assay cascade includes the assessment of their
metabolic stability in liver microsomes. Our current assay evaluates clearance
by CYP450 and UGT enzymes, which metabolise the majority of xenobiotics.
Once compound stability in mouse liver microsomes has been established, we
evaluate the in vivo pharmacokinetics in mice following iv and po administration
to select compounds suitable for further pharmacokinetic-pharmacodynamic
testing.

We have previously described C8-substituted pyrido[3,4-d]pyrimidin-4(3H)-one-
based JmjC histone demethylase inhibitors as potent and cell permeable
KDM4/KDM5 subfamily inhibitors (Bavetsias et al., 2016). Exemplar compound
1 exhibited low clearance in mouse liver microsomes, but very high in vivo
clearance.
In the current study, we characterised the metabolic route of 1 in mouse and
human cytosolic fractions and established the nature and the position of
metabolism. These experimental observations led to the hypothesis that the
pyrido[3,4-d]pyrimidin-4-one scaffold is liable to C2-oxidation by aldehyde
oxidase (AO). The succeptibility of the scaffold to AO metabolism was
confirmed experimentally. In addition, we showed that C2-substituted
pyrido[3,4-d]pyrimidin-4-one derivatives are protected from oxidation by AO.
Further molecular modelling studies in the human AO protein crystal structure
suggested a ligand binding mode that is consistent with our experimental
observations regarding metabolic liability.
Material and Methods
Compound synthesis
The syntheses of 1 and close analogue 2 have been previously reported
(Bavetsias et al., 2016). Compound 3 is commercially available. Compounds 4,
5 and 6 were prepared by methodologies analogous to those reported for the
synthesis of C2-unsubstituted pyrido[3,4-d]pyrimidin-4-one derivatives
(Bavetsias et al. 2016). Detailed experimental data including compound
characterisation is provided in the Supporting Information.

Microsomal and Cytosolic Clearance Studies
Liver microsomes and cytosolic fractions derived from female CD1 mice (MLM),
female Sprague-Dawley rats (RLM), a pool of 50 mixed gender human donors
(HLM) and mouse and human liver cytosol were obtained from Tebu Bio (Le
Perrey-en-Yvelines, France). CYP450 and AO-mediated metabolism of
compounds (1 µM DMSO) were assessed by incubation with liver microsomes
and cytosol (0.5 mg/ml) respectively in 0.1 M PBS at 37ºC. Microsomal
incubations were initiated by addition of NADPH (1 mM), UDPGA (2.5 mM) and
MgCl2 (3 mM) for phase I and II reactions. Control incubations were generated
by the omission of NADPH and UDPGA from the incubation reaction. Cytosol
incubations were carried out at pH 7.4 in the presence of 0.1 mM EDTA. They
were initiated by addition of 1 μM compound in the presence and absence of
selective AO inhibitor raloxifene and stopped by addition of 3 volumes of
acetonitrile at 0, 5, 10, 15 and 30 minutes, then centrifugation at 3000 rpm for
30 minutes at 4ºC. Supernatant was analysed by LC-MS/MS to determine
clearance. We also tested the effect of the selective xanthine oxidase inhibitor
allopurinol on the cytosolic clearance of compound 1.
In Vivo mouse Pharmacokinetic Studies
In vivo experiments were conducted according to the UK guidelines for the
welfare and use of animals in cancer research UKCC (Workman et al., 2010).
Female Balb/c mice (n = 3 per route) received either a single intravenous
(bolus) injection or a single oral administration (by gavage) of compound 1
(Figure 4) (5 mg/kg) in 10% DMSO, 0.1% Tween20 in Saline. Samples of whole

blood at specific time points were spotted onto Whatman FTA cards (GE
Healthcare, United Kingdom), allowed to dry at ambient temperature for at least
2 hours. Compound 1 was quantified by 6mm punch outs and extracted using
acetonitrile as previously described (Roberts et al., 2016).
Mass Spectrometric Analysis
In vitro microsomal incubation and in vivo samples were analysed using a
Waters Xevo TQ-S mass spectrometer by multiple reaction monitoring.
Conditions were 0.2% formic acid (mobile phase A) and acetonitrile (mobile
phase B). Separation was achieved on a Phenomenex Kinetex C18 column (2.1
x 50 mm; 2.6 µm). The column was equilibrated at initial condition of 95% A and
5% B for 0.5 min, linear gradient over 3 min to 100% B, held over 1 min,
followed by linear gradient back to 5% B over 0.1 min, at 0.6 mL/min flow rate.
AO incubation samples were analysed for metabolites of 1 using an Agilent
6520 QTOF MS. Using the above mobile phases, separation was achieved on a
Phenomenex Kinetex C18 column (2.1 x 100 mm; 2.6 µm). The column was
equilibrated at initial condition of 95% A and 5% B (0.5 min), linear gradient (15
min) to 50% B, held for 1 min, followed by linear gradient back to 5% B (0.5
min), at 0.6 mL/min flow rate.
Liquid Chromatographic Separation
Standard injections (with needle wash) of the sample were made onto a
ZORBAX Eclipse XDB-C18 column (4.6 x 150 mm, 5 μm, Agilent, Santa Clara,
USA). Chromatographic separation was carried out at 30°C using a 1260 Series
HPLC (Agilent, Santa Clara, USA) over gradient elution at a flow rate of 1.5

mL/min. UV-Vis spectra were acquired at 254nm on a 1260 Series diode array
detector (Agilent, Santa Clara, USA). Fractions were collected using an Agilent
analytical scale fraction collector.
Structural Identification of the Metabolites by 1H NMR
1H NMR data was collected on a Bruker Avance 500 spectrometer equipped
with a 1.7 mm TXI probe. The 1H NMR spectrum was referenced to the internal
deuterated solvent. The operating frequency for 1H was 500 MHz. All NMR data
were acquired at the temperature of 295 K. All data were acquired and
processed using Bruker Topspin 2.1.
Aldehyde Oxidase Ligand Binding Predictions
The protein-ligand co-crystal structure of human aldehyde oxidase (hAOX1)
(PDB code: 4uhw) (Coelho et al., 2015) was prepared using Protein Preparation
Wizard (Maestro v9.3, Schrödinger, LLC: New York, NY). To predict proposed
binding modes of the ligands, Glide (Grid-based Ligand Docking with
Energetics) (Glide v5.8, Schrödinger, LLC: New York, NY) was used for the
docking experiments. The receptor grid was defined by a grid box of 30 × 30 ×
30 Å3 with a default inner box (10 × 10 × 10 Å3) centered on the molybdenum
cofactor (MoCo) in the hAOX1 catalytic domain in structure PDB 4uhw.
Ligands were prepared using LigPrep, applying the OPLS_2005 force-field with
possible tautomeric states of neutral species within pH range 5.0-9.0 generated
using Epik metal binding state settings (Ligprep v2.5, Schrödinger, LLC: New
York, NY). Using Extra Precision (XP) settings, flexible docking of compound 3

was performed unconstrained. The predicted binding pose of compound 3
within the hAOX1 catalytic domain was subsequently used as the core
coordinates constraint to evaluate the feasibility of other synthesised ligands
adopting the same predicted binding mode.
Results
Exemplar pyrido[3,4-d]pyrimidin-4-one derivative 1 displayed moderate
clearance in mouse and human liver microsomes (45.4 and 34.1 µL/min/mg,
respectively) and low clearance in rat liver microsomes (10.4 µL/min/mg), which
prompted us to investigate the pharmacokinetics of this compound in mice.
Following iv and po administration of 5 mg/kg of compound 1 to female Balb C
mice, the blood clearance was greater than hepatic blood flow with only 2% oral
bioavailability (Figure 1). The measured mouse cytosolic clearance of 1 was
462.5 µL/min/mg, consistent with the observed high in vivo blood clearance.
Addition of raloxifene, an inhibitor of aldehyde oxidase, to the cytosolic
incubation greatly reduced the cytosolic clearance of compound 1 to 19
µL/min/mg, indicating that compound 1 is a substrate of mouse aldehyde
oxidases. Compound 1 was also highly metabolised in human cytosol with a
clearance of 752 µL/min/mg, which was reduced to 5 µL/min/mg by the addition
of raloxifene. Addition of allopurinol to the cytosolic incubation did not affect the
clearance showing that xanthine oxidase does not contribute to the cytosolic
clearance of compound 1(data not shown).
In order to characterise the metabolite(s) of compound 1, we evaluated the
mass spectrometric profile of the cytosolic incubation in comparison with the
control cytosol by full scan mass spectrometric analysis. We identified a strong

signal at M+16 (m/z 395.17) consistent with an oxidation product.
Fragmentation of the parent (m/z 379.17) and of this metabolite (m/z 395.17) in
the mass spectrometer revealed oxidation on the pyrido[3,4-d]pyrimidin-4(3H)-
one core scaffold (m/z 162.03 corresponding to m/z 146.04+16). Further
fragmentation of the pyrido[3,4-d]pyrimidin-4(3H)-one was not possible and the
exact position of the hydroxylation on the core could not be established by this
method (Figure 2). We therefore separated and isolated the metabolite by liquid
chromatography to carry out 1H NMR analysis of both the parent compound 1
and its metabolites. The 1H NMR traces (Figure 3 and supplemental Figure)
show that protons C5 and C6 are still present in the metabolite suggesting that
that oxidation has occurred at C2 of the pyrido[3,4-d]pyrimidin-4(3H)-one
scaffold. We evaluated both the mouse and human AO clearance of the parent
pyrido[3,4-d]pyrimidin-4(3H)-one scaffold (compound 3) that lacks any
substituent, and its C2-substituted derivatives (compounds 4 and 5, Figure 4).
In addition, we tested an additional pair of C8-substituted compounds, 2 and its
C2-methyl counterpart 6. Consistent with our previous observations regarding
compound 1, the cytosolic clearance of C2-unsubstituted analogues 2 and 3
was high (>90 ml/min/mg protein ), but it was significantly reduced by the
addition of raloxifene (Figure 4). In contrast, all C2-substituted derivatives
(compounds 4, 5 and 6) were stable in cytosolic incubations (Figure 4).
These results show that the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold is a
substrate for aldehyde oxidase at the C2-position, and that the introduction of a
C2-substituent blocks this AO-mediated metabolism. Since the crystal structure
of human aldehyde oxidase 1 (hAOX1) was recently published (Coelho et al.,

2015), we attempted to predict the binding modes of our pyrido[3,4-d]pyrimidin-
4(3H)-one–based JmjC histone lysine demethylase inhibitors in the hAOX1
substrate binding site. Using the crystal structure of substrate-free hAOX1 (PDB
code 4uhw), molecular docking studies of the core scaffold 3 suggested a
binding mode that involved cofactor-ligand interactions mediated via the 3-
position NH of the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold with the 3 position
hydrogen atom at 1.7 Å from one of the metal-coordinating oxygen atoms
within the molybdenum cofactor (MoCo). In this binding mode, the carbon atom
at the C2-position of the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold was 3.0 Å
away from the other metal-coordinating oxygen atom of MoCo and would be
reasonably positioned for nucleophilic oxidation (Alfaro and Jones, 2008). On
the contrary, this proposed binding mode of scaffold 3 would not be compatible
for exemplars of C2-substituted analogues including compounds 4 and 5 due to
steric clash of the alkyl substituent with the AO protein (Figure 5). This
observation provided a structural hypothesis explaining why C2-substituted
pyrido[3,4-d]pyrimidin-4(3H)-ones are not susceptible to AO-mediated
metabolism.
To further evaluate the relevance of this proposed binding mode for the
pyrido[3,4-d]pyrimidin-4(3H)-one-based demethylase inhibitor series, compound
1 was modelled in the hAOX1 binding site. The proposed binding mode of
compound 1 via the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold was found to be
compatible with the accommodation of its bulky C8 substituent in hAOX1
binding site. (Figure 6). These modelling results suggest a binding mode of the
pyrido[3,4-d]pyrimidin-4(3H)-one scaffold that is consistent with a hAOX1
oxidation mechanism at the C2-position and provide a structural insight to

explain our observation that introduction of a C2-substituent on the pyrido[3,4-
d]pyrimidin-4(3H)-one scaffold can abrogate AO-mediated metabolism in this
chemical series.
Discussion
Aldehyde oxidase (AO) is a molybdenum-containing enzyme located in the
cytosolic compartment of many cells and tissue types and is present in all
vertebrates (Carpani et al., 1990; Garattini et al., 2008; Falcani et al., 1994).
The fact that compound 1 is a substrate of mouse AO explains why mouse
microsomal clearance failed to predict the in vivo clearance of this compound.
The metabolism of AO substrates has been shown to vary greatly among
species with high expression in monkey, low expression in rat and minimal in
dogs (Dalgaard, 2015). Various AO isoforms have been described: in humans,
only one AOX1 functional gene is present, although several AOX1 orthologs are
found in most mammals with four in rodents (Terao et al.; 2015). Mammalian
AOX gene products are expressed in a tissue-specific manner in different
organisms. For example in humans, AOX1 is expressed predominantly in the
liver, but detectable amounts are also found in lung and kidney. All these factors
impact negatively on the potential accuracy of predictions of human
pharmacokinetics of AO substrates (Pryde et al., 2010; Sanoh et al., 2015).
Consequently, we decided to understand the basis of the metabolism of
compound 1 by human aldehyde oxidase 1 to ascertain if it could be blocked.
Our studies established that AO-mediated oxidation occurs at the C2-position of
the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold, which is consistent with literature

reports showing AO-mediated nucleophilic attack of an oxygen atom on the
electron deficient carbon atoms adjacent to nitrogen of certain aromatic
azaheterocycles (Dalvie et al, 2015). The introduction of a C2-alkyl substituent
(compounds 4, 5 and 6) was sufficient to block AO-mediated metabolism.
Subsequent molecular modelling studies predicted ligand binding modes of core
scaffold 3, and other elaborated analogues including compound 1, that were
compatible with the observed oxidation at the C2 position of the pyrido[3,4-
d]pyrimidin-4(3H)-one scaffold. On the contrary, steric clashes with the protein
were predicted for C2-substituted pyrido[3,4-d]pyrimidin-4(3H)-one derivatives,
giving a structural explanation for the lack of AO-mediated metabolism on these
compounds.
Compounds based on the pyrido[2,3-d]pyrimidin-7-one scaffold such as the p38
kinase inhibitor 6-(2,4-difluorophenoxy)-2-[(1-hydroxypropan-2-yl)amino]-8-(2-
hydroxypropyl)-7H,8H-pyrido[2,3-d]pyrimidin-7-one have been shown to be
extensively metabolized by human aldehyde oxidase into a 4-hydroxylated
metabolite (Zhang et al., 2011). In this instance, the compound which was
predicted to have a 6h half-life in man failed in Phase 1 clinical trial with a half-
life of 0.7h. It is therefore highly desirable to obtain compounds that are not
substrates for the AO-mediated metabolism.
In conclusion, our data show that the C2 position in pyrido[3,4-d]pyrimidin-
4(3H)-one-based derivatives is vulnerable to metabolism mediated by AO in
both mice and human. Blocking this susceptibility to AO-mediated metabolism

at the C2 position should therefore be considered in developing pyrido[3,4-
d]pyrimidin-4(3H)-one-based tool compounds for in vivo studies, and also in
exploring the potential of such compounds as anticancer agents.
References.
Alfaro JF, Jones JP. (2008). Studies on the mechanism of aldehyde oxidase and
xanthine oxidase. The Journal of organic chemistry, 73, 9469-72.
Bavetsias V, Lanigan RM, Ruda GF, Atrash B, McLaughlin MG, Tumber A, Mok NY, Le
Bihan YV, Dempster S, Boxall KJ and others. (2016). 8-Substituted Pyrido[3,4-
d]pyrimidin-4(3H)-one Derivatives As Potent, Cell Permeable, KDM4 (JMJD2)
and KDM5 (JARID1) Histone Lysine Demethylase Inhibitors. Journal of
medicinal chemistry, 59, 1388-409.
Berry WL, Janknecht R. (2013). KDM4/JMJD2 histone demethylases: epigenetic
regulators in cancer cells. Cancer research, 73, 2936-42.
Blair LP, Cao J, Zou MR, Sayegh J, Yan Q. (2011). Epigenetic Regulation by Lysine
Demethylase 5 (KDM5) Enzymes in Cancer. Cancers, 3, 1383-404.
Carpani G, Racchi M, Ghezzi P, Terao M, Garattini E. (1990). Purification and
characterization of mouse liver xanthine oxidase. Archives of biochemistry and
biophysics, 279, 237-41.
Cerqueira NM, Coelho C, Bras NF, Fernandes PA, Garattini E, Terao M, Romao MJ,
Ramos MJ. (2015). Insights into the structural determinants of substrate
specificity and activity in mouse aldehyde oxidases. Journal of biological
inorganic chemistry : JBIC : a publication of the Society of Biological Inorganic
Chemistry, 20, 20
Coelho C, Foti A, Hartmann T, Santos-Silva T, Leimkuhler S, Romao MJ. (2015).
Structural insights into xenobiotic and inhibitor binding to human aldehyde
oxidase. Nature chemical biology, 11, 779-83.
Coelho C, Mahro M, Trincao J, Carvalho AT, Ramos MJ, Terao M, Garattini E,
Leimkuhler S, Romao MJ. (2012). The first mammalian aldehyde oxidase
crystal structure: insights into substrate specificity. The Journal of biological
chemistry, 287, 40690-702.
Dalgaard L. (2015). Comparison of minipig, dog, monkey and human drug metabolism
and disposition. Journal of pharmacological and toxicological methods, 74, 80-
92.
Dalvie D, Zientek M. (2015). Metabolism of xenobiotics by aldehyde oxidase. Current
protocols in toxicology / editorial board, Mahin D. Maines (editor-in-chief) ... [et
al.], 63, 4.41.1-13.
Falciani F, Terao M, Goldwurm S, Ronchi A, Gatti A, Minoia C, Li Calzi M, Salmona M,
Cazzaniga G, Garattini E. (1994). Molybdenum(VI) salts convert the xanthine
oxidoreductase apoprotein into the active enzyme in mouse L929 fibroblastic
cells. The Biochemical journal, 298 ( Pt 1), 69-77.
Foti A, Hartmann T, Coelho C, Santos-Silva T, Romao MJ, Leimkuhler S. (2016).
Optimization of the expression of Human Aldehyde Oxidase for Investigations
of Single Nucleotide Polymorphisms. Drug metabolism and disposition: the
biological fate of chemicals.
Garattini E, Fratelli M, Terao M. (2008). Mammalian aldehyde oxidases: genetics,
evolution and biochemistry. Cellular and molecular life sciences : CMLS, 65,
1019-48.

Helin K, Dhanak D. (2013). Chromatin proteins and modifications as drug targets.
Nature, 502, 480-8.
Hillringhaus L, Yue WW, Rose NR, Ng SS, Gileadi C, Loenarz C, Bello SH, Bray JE,
Schofield CJ, Oppermann U. (2011). Structural and evolutionary basis for the
dual substrate selectivity of human KDM4 histone demethylase family. The
Journal of biological chemistry, 286, 41616-25.
Hoffmann I, Roatsch M, Schmitt ML, Carlino L, Pippel M, Sippl W, Jung M. (2012). The
role of histone demethylases in cancer therapy. Molecular oncology, 6, 683-
703.
Kawazu M, Saso K, Tong KI, McQuire T, Goto K, Son DO, Wakeham A, Miyagishi M,
Mak TW, Okada H. (2011). Histone demethylase JMJD2B functions as a co-
factor of estrogen receptor in breast cancer proliferation and mammary gland
development. PloS one, 6, e17830.
Kooistra SM, Helin K. (2012). Molecular mechanisms and potential functions of histone
demethylases. Nature reviews. Molecular cell biology, 13, 297-311.
Li W, Zhao L, Zang W, Liu Z, Chen L, Liu T, Xu D, Jia J. (2011). Histone demethylase
JMJD2B is required for tumor cell proliferation and survival and is
overexpressed in gastric cancer. Biochemical and biophysical research
communications, 416, 372-8.
Liu G, Bollig-Fischer A, Kreike B, van de Vijver MJ, Abrams J, Ethier SP, Yang ZQ.
(2009). Genomic amplification and oncogenic properties of the GASC1 histone
demethylase gene in breast cancer. Oncogene, 28, 4491-500.
Lohse B, Kristensen JL, Kristensen LH, Agger K, Helin K, Gajhede M, Clausen RP.
(2011). Inhibitors of histone demethylases. Bioorganic & medicinal chemistry,
19, 3625-36.
Pryde DC, Dalvie D, Hu Q, Jones P, Obach RS, Tran TD. (2010). Aldehyde oxidase:
an enzyme of emerging importance in drug discovery. Journal of medicinal
chemistry, 53, 8441-60.
Roesch A, Vultur A, Bogeski I, Wang H, Zimmermann KM, Speicher D, Korbel C,
Laschke MW, Gimotty PA, Philipp SE and others. (2013). Overcoming intrinsic
multidrug resistance in melanoma by blocking the mitochondrial respiratory
chain of slow-cycling JARID1B(high) cells. Cancer cell, 23, 811-25.
Sanoh S, Tayama Y, Sugihara K, Kitamura S, Ohta S. (2015). Significance of aldehyde
oxidase during drug development: Effects on drug metabolism,
pharmacokinetics, toxicity, and efficacy. Drug metabolism and
pharmacokinetics, 30, 52-63.
Sharma SV, Lee DY, Li B, Quinlan MP, Takahashi F, Maheswaran S, McDermott U,
Azizian N, Zou L, Fischbach MA and others. (2010). A chromatin-mediated
reversible drug-tolerant state in cancer cell subpopulations. Cell, 141, 69-80.
Shiau C, Trnka MJ, Bozicevic A, Ortiz Torres I, Al-Sady B, Burlingame AL, Narlikar GJ,
Fujimori DG. (2013). Reconstitution of nucleosome demethylation and catalytic
properties of a Jumonji histone demethylase. Chemistry & biology, 20, 494-9.
Terao M, Romao MJ, Leimkuhler S, Bolis M, Fratelli M, Coelho C, Santos-Silva T,
Garattini E. (2016). Structure and function of mammalian aldehyde oxidases.
Archives of toxicology, 90, 753-80.
Trojer P, Zhang J, Yonezawa M, Schmidt A, Zheng H, Jenuwein T, Reinberg D.
(2009). Dynamic Histone H1 Isotype 4 Methylation and Demethylation by
Histone Lysine Methyltransferase G9a/KMT1C and the Jumonji Domain-
containing JMJD2/KDM4 Proteins. The Journal of biological chemistry, 284,
8395-405
Walters ZS, Villarejo-Balcells B, Olmos D, Buist TW, Missiaglia E, Allen R, Al-Lazikani
B, Garrett MD, Blagg J, Shipley J. (2014). JARID2 is a direct target of the
PAX3-FOXO1 fusion protein and inhibits myogenic differentiation of
rhabdomyosarcoma cells. Oncogene, 33, 1148-57.

Westaway SM, Preston AG, Barker MD, Brown F, Brown JA, Campbell M, Chung CW,
Drewes G, Eagle R, Garton N and others. (2016). Cell Penetrant Inhibitors of
the KDM4 and KDM5 Families of Histone Lysine Demethylases. 2. Pyrido[3,4-
d]pyrimidin-4(3H)-one Derivatives. Journal of medicinal chemistry, 59, 1370-87.
Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ, et al.
Guidelinefor the welfare and use of animals in cancer research. Br J Cancer.
2010;102:1555-77.
Yang J, AlTahan AM, Hu D, Wang Y, Cheng PH, Morton CL, Qu C, Nathwani AC,
Shohet JM, Fotsis T and others. (2015). The role of histone demethylase
KDM4B in Myc signaling in neuroblastoma. Journal of the National Cancer
Institute, 107, djv080.
Young LC, Hendzel MJ. (2013). The oncogenic potential of Jumonji D2 (JMJD2/KDM4)
histone demethylase overexpression. Biochemistry and cell biology = Biochimie
et biologie cellulaire, 91, 369-77.
Zhang X, Liu HH, Weller P, Zheng M, Tao W, Wang J, Liao G, Monshouwer M, Peltz
G. (2011). In silico and in vitro pharmacogenetics: aldehyde oxidase rapidly
metabolizes a p38 kinase inhibitor. The pharmacogenomics journal, 11, 15-24.

Figure legends:
Figure 1:
Concentration versus time profile of Compound 1 following 5mg/kg iv (red) and po
(blue) in 10% DMSO, 1% tween 80 in saline. Profiles are from individual mice (n=3).
Analysis was carried out by LC-MS/MS with selected reaction monitoring. Clearance
and bioavailability have been calculated with Phoenix non-compartmental analysis.
Figure 2:
Fragmentation pattern of Compound 1 (m/z 379.17) and its oxidation product (m/z
395.17). The core pyrido[3,4-d]pyrimidin-4(3H)-one fragment in 1 (m/z 146.04) is
oxidised in the metabolite (m/z 162.03). No further fragmentation was observed.
Figure 3:
1H NMR spectrum of compound 1 and its oxidation product referenced to internal
deuterated solvent. Protons C5 and C6 are still present in the metabolite suggesting
that the oxidation is at position C2 of the pyrido[3,4-d]pyrimidin-4(3H)-one scaffold.
Figure 4:
Structure of a number of pyrido[3,4-d]pyrimidin-4(3H)-one derivatives and clearance in
mouse and human cytosol following incubation of 1μM compound in the absence and
presence of the aldehyde oxidase inhibitor raloxifene. Values are means + SD of n=3
replicate analysis. Statistical analysis carried out with paired t test comparing clearance
with and without inhibitor. ***p<0.001.

Figure 5:
Proposed binding mode of compound 3 (orange sticks) in the human hAOX1 substrate
binding site (PDB code 4uhw, grey). Cartoon representation of the protein with the
proposed binding mode of 3 to the metal-coordinating oxygen atoms within the
molybdenum cofactor MoCo (left); Surface representation illustrating the proposed
binding mode (right).
Figure 6:
Proposed binding mode of compound 1 (orange sticks) in the human hAOX1 substrate
binding site (PDB code 4uhw, grey surface), showing that the bulky C8-substituent can
be accommodated in the AO protein. (left); close-up illustrating the proposed binding
mode interactions (right).

Cl=1650ml/min/kg
F=2%
Figure 1

Figure 2
+ O
+ O

Figure 3
Compound 1
Compound 1
C5’-
C3’-
C5-H
C6-H
C2-H
C6-H
Metabolite
C5’-
C5-H
C3’-

Figure 4
***
***
***
***
***
***
***
***

Figure 5
MoCo
1.7 Å
3.0 Å

Figure 6
1.7 Å
3.1 Å