Discovery of 2-phenoxyacetamides as inhibitors of the Wnt-depalmitoleating enzyme NOTUM from an X-ray fragment screen

Article abstract of DOI:10.1039/c9md00096h

NOTUM is a carboxylesterase that has been shown to act by mediating the O-depalmitoleoylation of Wnt proteins resulting in suppression of Wnt signaling. Here, we describe the development of NOTUM inhibitors that restore Wnt signaling for use in in vitro disease models where NOTUM over activity is an underlying cause. A crystallographic fragment screen with NOTUM identified 2-phenoxyacetamide 3 as binding in the palmitoleate pocket with modest inhibition activity (IC₅₀ 33 μM). Optimization of hit 3 by SAR studies guided by SBDD identified indazole 38 (IC₅₀ 0.032 μM) and isoquinoline 45 (IC₅₀ 0.085 μM) as potent inhibitors of NOTUM. The binding of 45 to NOTUM was rationalized through an X-ray co-crystal structure determination which showed a flipped binding orientation compared to 3. However, it was not possible to combine NOTUM inhibition activity with metabolic stability as the majority of the compounds tested were rapidly metabolized in an NADPH-independent manner.

Full text of DOI:10.1039/c9md00096h

View Article Online
View Journal
MedChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: B. N. Atkinson, D.
Steadman, Y. Zhao, J. Sipthorp, L. Vecchia, R. R. Ruza, F. Jeganathan, G. Lines, S. Frew, A. Monaghan, S.
Kjaer, M. Bictash, Y. Jones and P. V. Fish, Med. Chem. Commun., 2019, DOI: 10.1039/C9MD00096H.
This is an Accepted Manuscript, which has been through the
Volume 7 Number 1 January 2016 Pages 1–204
Royal Society of Chemistry peer review process and has been
accepted for publication.
MedChemComm
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Broadening the field of opportunity for medicinal chemists
www.rsc.org/medchemcomm
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
Themed issue: Antibiotic Resistance
ISSN 2040-2503
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/medchemcomm

Page 1 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
Discovery of 2-phenoxyacetamides as Inhibitors of the Wnt-
depalmitoleating enzyme NOTUM from an X-ray Fragment Screen†
Benjamin N. Atkinson,a, ‡ David Steadman,a, ‡ Yuguang Zhao,b, ‡ James Sipthorp,a, c Luca Vecchia,b Reinis
R. Ruza,b Fiona Jeganathana, Georgie Lines,a Sarah Frew,a Amy Monaghan,a Svend Kjaer,c Magda
Bictash,a E. Yvonne Jones,b, * Paul V. Fish.a, c,
*
a Alzheimer’s Research UK UCL Drug Discovery Institute, University College London, Cruciform
Building, Gower Street, London, WC1E 6BT, U.K.
b Division of Structural Biology, Wellcome Centre for Human Genetics, University of Oxford, The
Henry Wellcome Building for Genomic Medicine, Roosevelt Drive, Oxford, OX3 7BN, U.K.
c The Francis Crick Institute, 1 Midland Road, London, NW1 1AT, U.K.
KEYWORDS: NOTUM inhibitor, Wnt signaling, Alzheimer’s disease, Crystallographic fragment
screen, 2-Phenoxyacetamide, NADPH independent metabolism.
ABSTRACT: NOTUM is a carboxylesterase that has been shown to act by mediating the O-
depalmitoleoylation of Wnt proteins resulting in suppression of Wnt signaling. Here, we describe the
development of NOTUM inhibitors that restore Wnt signaling for use in in vitro disease models where
NOTUM over activity is an underlying cause. A crystallographic fragment screen with NOTUM
identified 2-phenoxyacetamide 3 as binding in the palmitoleate pocket with modest inhibition activity
(IC50 33 M). Optimization of hit 3 by SAR studies guided by SBDD identified indazole 38 (IC50 0.032
μM) and isoquinoline 45 (IC50 0.085 μM) as potent inhibitors of NOTUM. The binding of 45 to NOTUM
was rationalized through an X-ray co-crystal structure determination which showed a flipped binding
orientation compared to 3. However, it was not possible to combine NOTUM inhibition activity with
metabolic stability as the majority of the compounds tested were rapidly metabolized in an NADPH-
independent manner.
† This work was first presented at the following international conferences:
E.Y.J. Biochemical Society Meeting. Acylation of intracellular and secreted proteins: mechanisms and
functional outcomes; 10-12th September 2018; Brighton, UK.
P.V.F. EFMC: XXVth International Symposium on Medicinal Chemistry; 2-6th September 2018;
Ljubljana, Slovenia.

MedChemComm
Page 2 of 21
View Article Online
DOI: 10.1039/C9MD00096H
Members of the Wnt family are secreted signaling proteins that play key roles in adult stem cell biology
as well as in embryonic development.1 Wnts initiate signaling by binding to cell surface receptors and
two main pathways have been identified downstream of Wnts: the so-called ‘canonical’ and ‘planar cell
polarity’ pathways. These pathways are triggered by the binding of Wnt to a member of the Frizzled
family of cell surface receptors and, for the canonical pathway, a member of the LDL-receptor-related
protein (LRP) family of cell surface receptors (typically LRP5 or LRP6). This binding elicits an
intracellular signaling cascade that results in both biochemical and transcriptional changes within the cell,
with the canonical pathway involving the accumulation and translocation of ß-catenin. Both pathways are
tightly regulated by a sophisticated network of modulators and feedbacks including secreted inhibitory
Dickkopf (DKK) proteins2 and post translational modifications (PTM).3, 4
Conversely, dysregulation of Wnt signaling is frequently associated with growth-related pathologies and
cancers,5 particularly those of tissues for which Wnts normally stimulate self-renewal and repair. Wnt
signaling is also implicated to have a role in neurodegenerative diseases such as Alzheimer’s disease
(AD). Cognitive impairments, characteristic of AD, correlate closely with the loss of synapses and current
knowledge suggests that excess amyloid-ß (A) causes synapse dysfunction by impairing synapse
maintenance, at least in part, through causing dysfunction of Wnt signaling.6, 7 Compromised Wnt
signaling may also be associated with AD through loss of blood-brain barrier (BBB) integrity8 and A
generation through -secretase (BACE1) expression.9
O-Palmitoleoylation of a conserved serine residue in Wnt proteins is a key PTM required for efficient
binding of Wnt proteins to Frizzled receptors, a requirement for signal transduction.10 A carboxylesterase
NOTUM was previously shown to antagonize Wnt signaling pathways.11 More recently, NOTUM has
been shown to act by mediating the depalmitoleoylation of Wnt proteins resulting in suppression of Wnt
signaling.12, 13 It follows that inhibition of NOTUM could restore Wnt signaling with potential benefit in
disease where Wnt deficiency is an underlying cause.
There are only a few reports of NOTUM inhibitors with compounds falling into two main categories.
Thienopyrimidine LP-922056 (1) is a potent, orally active inhibitor of NOTUM that has shown NOTUM
to be a potential drug target for stimulating bone formation and treating osteoporosis.14 However, although
1 demonstrates low clearance and good bioavailability, the structure contains an essential carboxylic acid
and acids tend to have low passive brain penetration;15-18 in fact, it is an accepted strategy to add a
carboxylic acid group to minimize brain penetration of drugs for peripheral targets.19 N-
Hydroxyhydantoin carbamates are potent and selective irreversible inhibitors of NOTUM discovered by
activity-based protein profiling (ABPP).20 An optimized inhibitor ABC99 (2) preserves Wnt-mediated

Page 3 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
cell signaling in the presence of NOTUM and was converted to a clickable ABPP probe for visualizing
NOTUM in biological systems. Due to their reactive mechanism of action forming covalent adducts with
NOTUM Ser232, it is unlikely that these compounds will be suitable for in vivo studies. Hence, our
objective was to discover potent small molecule inhibitors of NOTUM suitable for exploring the
regulation of Wnt signaling in the central nervous system (CNS) and modulation of AD phenotypes.
N
S
N
CO2H
Me
O
S
O
O
O
N
N
H
Cl
N
O
N
N O
N
Cl
O
1: LP-922056
2: ABC99
3
O
O
HNR1R2
O
CO2H +
O
OH
NR1R2
NR1R2
R
R
R
+
Cl
Scheme 1: Chemical structures of LP-922056 (1), ABC99 (2) and initial fragment hit 3. General scheme for the synthesis of
2-phenoxyacetamides reported in Tables 1 & 2.
In order to identify new small molecule inhibitors of NOTUM, a crystallographic fragment screen was
performed using the XChem platform at Diamond Light source. Crystals of C-terminal his-tagged
NOTUM(Ser81-Thr451 Cys330Ser) were soaked with the DSI-Poised library (XChem, 768 fragments).21
Crystal structures of NOTUM show a distinctive pocket that accommodates the palmitoleate group
(Figure 1A).12 Fragments observed to bind in the palmitoleate pocket were all re-synthesized as solid
samples to establish structure and purity. Inhibition of NOTUM carboxylesterase activity of these hits
was measured in a cell-free biochemical assay. In brief, test compounds (dispensed to give 10 point
concentration-response-curves) were incubated with NOTUM(81-451 Cys330Ser) and trisodium 8-
octanoyloxypyrene-1,3,6-trisulfonate (OPTS) as the substrate for 1 h, and fluorescence recorded; an
inhibitor of NOTUM would suppress fluorescence by binding to NOTUM and preventing hydrolysis of
OPTS (Figure S1).14, 22
One of the preferred hits from this set was 2-(2-methylphenoxy)-N-(pyridine-3-yl)acetamide (3)(IC50 33
± 4.7 M) which was selected for further investigation for several reasons: (1) excellent “hit-like”
properties (mw 242; cLogP 1.7; TPSA 51; LE = 0.35; LLE = 2.8) with no obvious reactive groups; (2)
chemically enabled to selectively functionalize each position to explore structure activity relationships

MedChemComm
Page 4 of 21
View Article Online
DOI: 10.1039/C9MD00096H
(SAR); (3) successful co-crystal structure with NOTUM to support a structure based drug design (SBDD)
program; and (4) structural features and physicochemical properties consistent with CNS drug-like
space23 including a favorable ‘CNS multiparameter optimization’ (CNS MPO) score (CNS MPO =
5.6/6.0).24, 25 Analysis of the crystal structure (Figure 1B) showed key pi-pi stacking interactions between
the pyridine ring and Trp128 at the outer pocket, and further pi-pi interactions between the tolyl ring and
Phe268/Phe320 within the deeper, lipophilic pocket. No hydrogen bonding interactions between the
ligand and the protein were observed, so it was considered that these pi-pi interactions were crucial for
binding, in addition to general lipophilic interactions, as observed for the palmitoleate ligand (Figure 1A).
Notably, the orientation of Trp128 is significantly shifted in our crystal structures compared to previous
structures like PDB 4UZQ. This shift shows a more open form of the binding site. Also in contrast to the
O-palmitoleoyl serine structure, 3 does not interact with the oxyanion hole of the active site.
Two general synthetic methods were used to prepare new analogues: either an activated amide coupling
reaction between the aminoheterocycle and the 2-phenoxyacetic acid; or a nucleophilic substitution
reaction between the 2-chloroacetamide and the relevant phenoxide (Scheme 1; see, Electronic
Supplementary Information). Inhibition of NOTUM carboxylesterase activity was routinely measured in
an OPTS biochemical assay as described above. Selected compounds were then screened for NOTUM
occupancy in a FP-biotin competition assay26 and/or inhibition of NOTUM activity in a Wnt/-catenin
signaling pathway TCF/LEF Reporter (Luciferase) HEK293 cell line with exogenous Wnt3a and
NOTUM.14, 22 Compounds were also screened for aqueous solubility, transit performance in MDCK-mdr1
cell lines for permeability, and metabolic stability in human and mouse liver microsomes (HLM and MLM
resp.) as a measure of clearance.

Page 5 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
Figure 1: X-ray crystal structures of ligands bound to NOTUM. (A) O-Palmitoleoyl serine (orange) (S232A mutant NOTUM),
PDB ID: 4UZQ; (B) Fragment hit 3 (green) with key pi-pi stacking interactions between the ligand and Trp128 and Phe268
indicated, PDB ID: 6R8P; (C) Benzotriazole 39 (blue) showing a similar binding mode to 3, but with flipping of Trp128, PDB
ID: 6R8Q; (D) Isoquinoline 45 (yellow) demonstrates a reorientation of the phenoxy and amide groups but with key pi-pi
stacking interactions maintained, PDB ID: 6R8R. Ligands are shown as sticks, and protein represented as ribbons with key
amino acids shown as lines. Key pi-pi stacking interactions are shown between the ligands and the protein as black dots. Phe320
lies above the plane of the Figure and is not shown for clarity.
The SARs were directed at exploring four principle areas of the original hit 3: the 2-heteroatom linker (4-
7), substituents on the phenoxy ring (8-18), the acetamide backbone (19, 20)(Table 1), and the amide
group (21-53)(Table 2). Furthermore, target compounds were designed to have molecular and
physicochemical properties consistent with CNS drug-like space. It is generally accepted that CNS drugs
tend to occupy a restricted property space compared to general orally available drugs,23 and extensive
efforts have been made to define attributes that are desirable for molecules to have good brain exposure.
One such model is the CNS MPO score24, 25 which we routinely used in our compound design. A second
useful design metric was lipophilic ligand efficiency (LLE)27, 28 used to track improvements in NOTUM
activity against lipophilicity (cLogP) of the inhibitors. The palmitoleate pocket of NOTUM is a very
lipophilic environment and we were keen to avoid gains in activity simply through increased compound
lipophilicity.
Initial studies suggested that replacing the O-atom linker was not an effective strategy for increasing
activity, with potency decreasing from O > NH > S > CH2 (3-6) (Table 1). The preference for an O-linker
was believed to be due to a favorable internal H-bond with the amide NH holding the molecule in a low
energy conformation that closely mimicked the NOTUM binding conformation. The NMe linker 7 was

MedChemComm
Page 6 of 21
View Article Online
DOI: 10.1039/C9MD00096H
similar in potency for this matched pair (3 v 7) although this did not prove to be a general trend, and so
the O-linker was retained for the next phase of SAR. N-Methylation of the amide group (19) or the 2-
phenoxy propanamide backbone (20) was tolerated but offered no advantage (Table 1).
Guided by the NOTUM-3 crystal structure, a concise investigation of SAR of the phenoxy ring indicated
the original 2-methyl group in 3 was preferred with a beneficial 2-fold increase in potency only achieved
by direct replacement with a 2-chlorine atom 13 (Table 1). Moving the Cl atom or Me group to the 3-, 4-
positions gave a significant decrease in potency (14, 15, 16) and was not pursued further at this point.
These results were consistent with the structural information showing minimal space to accommodate any
group larger than H- or F-atom (17) in these positions. The symmetrical 2,6-diMe phenoxy ether 18 was
tolerated but offered no advantage over a single 2-Me substituent. From this limited set, most groups were
detrimental, or offered little advantage, compared to having no substituent (8; R = H) with a preference
for small lipophilic groups at the 2-position of the ring such as 2-Me, 2-Cl and 2-CF3.

Page 7 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
Table 1: Inhibition of NOTUM activity. SAR of the heteroatom linker, the phenoxy ring and the
acetamide.
N
N
O
Me
O
X
O
N
N
R
H
R2 R1
3 - 18
19: R1 = Me; R2 = H
20: R1 = H; R2 = Me
#
3
4
5
6
7
8
9
X
R
IC50 (μM)a
O
NH
S
2-Me
2-Me
2-Me
2-Me
2-Me
H
33
60
110
CH2
NMe
O
140
23
150
O
2-F
290
10
11
12
13
14
15
16
17
18
19
20
O
2-OMe
2-CN
2-CF3
2-Cl
170
O
120
O
37
O
14
O
3-Cl
310
O
3-Me
4-Me
2-Me;4-F
2-Me;6-Me
-
37% I @ 100 M
O
20% I @ 100 M
O
20
44
O
-
180
100
-
-
a
Values are geometric means of n = 2-8 experiments quoted to 2 s.f. Screening data only passed the
quality control criteria if the screening plates demonstrated a Z’ > 0.5 and the NOTUM inhibition activity
of the positive control was within acceptable limits (IC50 0.6 – 1.1 nM). Differences of <2-fold should
not be considered significant. See also ESI Table S1 for s.e.m..

MedChemComm
Page 8 of 21
View Article Online
DOI: 10.1039/C9MD00096H
The next phase of SAR was to explore the N-acetamide substituent with a variety of heterocycles (Table
2). Although swapping the 3-Py group for the Ph (21) gave a boost in activity, this was at a significant
penalty in added lipophilicity (cLogP 2.9, LLE = 2.8), and pyrazine 22 proved to be a more efficient
inhibitor (cLogP 0.8, LLE = 4.2). Interestingly, the 3-(aminomethyl)pyridine amide 25 showed a decrease
in potency from 3 suggesting that the introduction of the methylene linker disrupted the positioning of the
ligand. The most significant increase in activity was achieved with benzo-fused heterocycles such as 26
and 27. These results prompted a more systematic investigation of both 6,5- and 6,6-heterocycles.
Fused 6,5-ring systems were well tolerated, with a range of indoles (28, 29), benzimidazoles (30, 31),
indazoles (32-38) and benzotriazoles (39-41) having NOTUM inhibition < 1 μM (Table 2). With good
activity now achieved within this series, indazole 34 (IC50 0.27 ± 0.06 M) (cLogP 3.1; LLE = 3.6; CNS
MPO 5.6) was selected as a representative early lead for in vitro ADME profiling. Indazole 34 had low
aqueous thermodynamic solubility (2 g/mL) and moderate cell permeability as measured by transit
performance in the MDCK-mdr1 cell line with no evidence of P-gp mediated efflux (AB/BA Papp 5.1/6.6
x106 cm.s-1). However, 34 had very poor metabolic stability in both HLM and MLM and was found to be
rapidly degraded in a NADPH-independent manner (Table 3). Hydrolytic stability of 34 was tested in
three buffer systems (pH 4, 7.4, 10) at multiple time points up to 18 h and only parent was detected by
UPLC-MS confirming degradation of 34 was not caused by simple hydrolysis. These results highlighted
the need to further improve NOTUM inhibition activity whilst simultaneously developing compounds
with good metabolic stability. Hence the microsomal stability assay became a priority secondary screen
(vide infra).
Further investigation of SAR within the indazole series showed activity was increased by addition of a
larger alkyl group (Me, Et, iPr) on the N1 or N2 position (35 v 37 v 38). Particularly noteworthy is the 2-
isopropylindazole 38 (IC50 0.032 ± 0.004 μM) (cLogP 4.0; LLE = 3.5; CNS MPO 5.1) which was the
most potent NOTUM inhibitor from this 6,5-series. Benzotriazoles (39-41) were well tolerated with the
SAR of the N1-Me (40) and N2-Me (41) derivatives tracking their counterparts in the indazole series (e.g.
32 v 35 v 41). The unsubstituted benzotriazole 39 (IC50 0.12 ± 0.04 μM) (cLogP 2.8; LLE = 4.2; CNS
MPO 5.0) offered a slight advantage and again showed that a free N1-H could be accommodated as seen
previously with indole 26. However, both 38 and 39 were also rapidly degraded in HLM and MLM in a
NADPH-independent manner.
Fused 6,6-ring systems showed more variation in their NOTUM inhibition activities with the position of
the N atom(s) critical for achieving good activity (Table 2). A systematic migration of a single N atom
around each available position of the heterocycle (42-48) showed a preference for the N atom in the distal

Page 9 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
ring with a clear advantage for the 6-isoquinoline isomer 45 (IC50 0.085 ± 0.011 μM) (cLogP 3.0; LLE =
4.1; CNS MPO 5.6). Multiple N atoms were accommodated in the distal ring of the 6,6-rings as shown
with quinazoline 49 but was still inferior to 45.
With good NOTUM inhibition activity achieved across a range of amides, preferred compounds (IC50
<0.50 M) were submitted for NOTUM X-ray co-crystallography studies to elucidate their binding
modes. Structures of benzotriazole 39 (Figure 1C) and isoquinoline 45 (Figure 1D) were solved and
showed that, while having similar potencies, they differed in their adopted binding conformations. Both
compounds formed pi-pi stacking interactions with Trp128 through their heterocyclic rings but there was
a distinct difference in the orientation of their phenoxyacetamide backbones. Benzotriazole 39 adopted a
similar orientation as the pyridinyl fragment 3 (Figure 1B). In contrast, isoquinoline 45 showed a 180o
rotation around the amide bond with the carbonyl oxygen now pointing away from the oxyanion hole and
the methyl group of the tolyl group occupying the opposite side of the lipophilic pocket.
An overlay of the crystal structures of 3 and 45 showed that the phenoxy ring of 45 was shifted slightly
(ca. 1.28 Å) towards the front of the binding pocket, suggesting the possibility of additional space for
further modification to this ring. Therefore, substitution on the phenoxy ring was revisited in the 45 series
in an attempt to achieve optimal occupancy of the lipophilic pocket and increase potency. However, once
again, only a 2-Cl group 50 (IC50 0.069 ± 0.011 μM) (cLogP 3.3; LLE = 3.9; CNS MPO 5.5) was
accommodated with the 3-Me 51 and 3-Cl 52 giving a significant decrease in potency, although still
superior to the direct analogues in the 3-pyridinyl series (14 and 15). Isoquinoline amides 45 and 50 were
screened in liver microsomes and also showed rapid degradation in a NADPH-independent manner.

MedChemComm
Page 10 of 21
View Article Online
DOI: 10.1039/C9MD00096H
Representative inhibitors were tested in a NOTUM occupancy assay using FP-biotin, a serine hydrolase
activity-based probe, whereby labelling of NOTUM with FP-biotin can be blocked by an inhibitor
occupying the active site of NOTUM (Figure 2A). All inhibitors showed some ability to prevent labelling
by FP-biotin, confirming they competitively bind to NOTUM, albeit with modest potency compared to 1,
and the rank order of binding tracks with NOTUM IC50 activity in the OPTS assay.
A
B
120
100
80
60
40
20
0
-8
-7
-6
-5
-4
-20
Log10 [Compound] M
Figure 2: (A) NOTUM activity-based occupancy assay was performed with FP-biotin (2 M) and test compounds (10 M)
for 10 minutes in conditioned media from HEK293S cells stably transfected with a NOTUM lentiviral construct. Relative
occupancy was calculated by optical density of fluorescent band detecting the level of biotinylation of NOTUM using Image
Studio Lite 5.2, compared to the control-treated sample which was set to 1. N = 3 with S.D. Statistical significance is calculated
using an unpaired t-test with welch’s adjustment (***p = <0.0005, **p = <0.005). UT, untreated. (B) Percentage activation of
the Wnt pathway via inhibition of NOTUM by compounds in the stable TCF-Lef SuperTOPFLASH luciferase reporter
HEK293T cell line. Cells were treated in 8 point concentration response curves from 3 nM to 10 µM for 18 hours. % Activation
of the TCF/Lef reporter gene was calculated by normalizing data to DMSO (minimum) and to compound 1 at 10 µM
(maximum) control wells. EC50s were calculated using a 4PL fit in Graphpad Prism and Dotmatics Studies. Compound key:
1 (●), 37 (■), 38 (○), 45 (▲), 49 (∆).
Preferred inhibitors were then screened in the cell-based TCF/LEF reporter gene (Luciferase) assay to
assess their ability to restore Wnt/-catenin signaling when activated by exogenous Wnt3a (100 ng/mL)
in the presence of NOTUM (500 ng/mL) (Figure 2B). Assay validation studies with a training set of
NOTUM inhibitors showed that nanomolar activity in the OPTS assay (IC50 < 1 M) was required to
show activity in the cell assay; these findings are consistent with previous reports.14, 22 Compound 1
showed an effective activation of Wnt signaling (EC50 0.138 ± 0.053 M; n = 4) in this model system
through inhibition of NOTUM (ESI, Table S2). Compound 37 also showed activation of Wnt signaling
via inhibition of NOTUM (EC50 0.386 ± 0.199 M; n = 4). Compound 38 demonstrated 60 % inhibition

Page 11 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
of NOTUM at 3 µM (estimated EC50 0.934 ± 0.657 µM; n = 4), whilst 45 and 49 showed 20 % and 38 %
inhibition at 10 µM respectively. More stringent EC50 determination was precluded by bell shaped dose
response curves for the compounds when tested at concentrations > 10 µM. In contrast, 27 and 35
demonstrated a supramaximal response consistent with dual activity in this assay, preventing accurate
EC50 determination. These compounds not only inhibit NOTUM as shown by the Wnt3a-mediated
activation of the reporter system but also acts as stabilizers of the luciferase enzyme giving the
supramaximal response.29 This luciferase activity is consistent with similar chemical structures to 3 that
have been reported to be inhibitors of luciferase in the PubChem database.30
11

MedChemComm
Page 12 of 21
View Article Online
DOI: 10.1039/C9MD00096H
Table 2: Inhibition of NOTUM activity. SAR of the amide heterocycles.
Me
O
O
N
O
R
O
N
N
H
R
H
21 - 49
50: R = 2-Cl
51: R = 3-Me
52: R = 3-Cl
Me
O
Me
O
Me
N
O
N
N
N
N
H
Me
53
54
#
R
N
IC50 (μM) a
#
R
IC50 (μM) a
3
33
21
1.6
Me
N
N
N
22
24
26
9.4
72
23
25
27
3.6
100
0.68
N
Me
N
N
N
N
N
N
H
0.21
Me
N
H
N
28
30
32
0.33
0.52
0.36
29
31
33
0.24
0.98
0.27
N
Me
N
N
N
Me
Me
N
N
N
N Me
12

Page 13 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
N
N Me
N
Me
N
34
36
38
0.27
35
37
39
0.28
0.068
0.12
N
N Et
N
Et
N
0.20
N
N
N
iPr
N
H
N
0.032
N
N
N
N
N
Me
N
40
0.44
41
0.27
Me
N
N
42
44
46
48
7.4
1.2
0.97
15
43
45
47
49
2.5
0.085
0.43
0.67
N
N
N
N
N
N
N
-
-
50
52
54
0.069
5.2
51
53
19
-
-
-
5.7
7.9
a
Values are geometric means of n = 2-8 experiments quoted to 2 s.f.. Screening data only passed the
quality control criteria if the screening plates demonstrated a Z’ > 0.5 and the NOTUM inhibition activity
of the positive control was within acceptable limits (IC50 0.6 – 1.1 nM). Differences of <2-fold should
not be considered significant. See also ESI Table S1 for s.e.m..
13

MedChemComm
Page 14 of 21
View Article Online
DOI: 10.1039/C9MD00096H
Table 3: In vitro microsomal stability data for selected compounds.
#
NOTUM IC50
HLM
MLM
cLogP
(t1/2 min)
(t1/2 min)
(μM)
3
33
9.4
-
1.1*
Not detected*
Not detected*
5.7*
1.7
0.8
3.1
4.0
2.8
3.0
3.3
3.1
3.5
22
34
38
39
45
50
53
54
-
0.27
0.032
0.12
0.085
0.069
5.7
11.6*
5.1*
-
1.8*
12.1*
11.0*
-
-
-
3.3*
5.2*
7.9
Not detected*
*NADPH independent metabolism.
Throughout the course of analogue preparation, preferred examples were assessed for microsomal
stability (Table 3). However, the majority of the compounds tested were rapidly metabolized in an
NADPH-independent manner with t1/2 <12.5 min. One approach to mitigate this issue was to introduce a
NMe linker as a direct replacement for the O linker with indazole 53 (cf 34) but this led to a 25-fold
decrease in activity and had no effect on alleviating the NADPH-independent metabolism. When an
isolipophilic non-aromatic amide was profiled in MLM, tetrahydroisoquinoline 54 was also found to have
rapid degradation in a NADPH-independent manner. Even revisiting some of the more polar analogues
such as 3 and 22 in MLM suffered a similar outcome. Rapid NADPH-independent metabolism of
structurally related 2-phenoxypropanamides has recently been reported suggesting a common metabolic
liability associated with this type of scaffold.31 Hence, in light of this work, we recommend assessing the
stability of N-aryl 2-phenoxyacetamide scaffolds to NADPH-independent metabolism in liver
microsomes early in the drug discovery process.
14

Page 15 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
In summary, a crystallographic fragment screen with carboxylesterase NOTUM identified 2-
phenoxyacetamide 3 as binding in the palmitoleate pocket with modest inhibition activity. Optimization
of hit 3 by SAR studies guided by SBDD identified indazole 38 and isoquinoline 45 as potent inhibitors
of NOTUM with properties entirely consistent with CNS drug space. Through direct inhibition of
NOTUM, 38 and 45 may prove to be valuable chemical tools for use in in vitro disease models where
over activity of NOTUM suppressing Wnt signaling is an underlying cause. However, ultimately, it was
not possible to combine NOTUM inhibition activity with metabolic stability in this series as the majority
of the compounds tested in liver microsomes were rapidly metabolized in an NADPH-independent
manner. These compounds represent a step forward in identifying new chemical matter to study the role
of NOTUM in Wnt signaling, and work is ongoing to identify compounds suitable for in vivo studies to
validate NOTUM as a target in Wnt signaling related CNS disorders.
15

MedChemComm
Page 16 of 21
View Article Online
DOI: 10.1039/C9MD00096H
ASSOCIATED CONTENT
Electronic Supporting Information
Synthetic procedures and characterization data; enzyme production and purification methods; OPTS assay
procedure; FP-biotin competition assay; TCF/LEF reporter assay; ADME screens; protein crystallization,
data collection and structure determination.
Accession Codes
Atomic coordinates have been deposited in the Protein Data Bank (PDB) and will be released upon
publication. PDB ID codes: 3: 6R8P; 39: 6R8Q; 45: 6R8R.
AUTHOR INFORMATION
Corresponding Authors
E.Y. J.:
Phone: +44 (0)1865 287 546; E-mail: yvonne@strubi.ox.ac.uk;
ORCiD 0000-0002-3834-1893.
P. V. F.:
Phone: +44 (0)20 7679 6971; E-mail: p.fish@ucl.ac.uk;
ORCiD 0000-0002-2117-2173.
Author Contributions
‡ B.N.A., D.S. and Y.Z. contributed equally. The manuscript was written through contributions from all
authors. B.N.A., D.S., J.S., A.M., E.Y.J. and P.V.F. wrote the manuscript. All authors have given approval
to the final version of the manuscript.
Funding Sources
E.Y.J.: Structural analysis was performed by Y.Z., L.V., R.R.R. and E.Y.J. supported by Cancer Research
UK (Programme Grant C375/A17721) and the Wellcome Trust (PhD Training Programme
102164/B/13/Z). The Wellcome Trust funds the Wellcome Centre for Human Genetics, University of
Oxford (Centre Grant 203141/Z/16/Z).
P.V.F.: This work was supported by the Alzheimer’s Research UK (ARUK) and the Francis Crick
Institute. The ARUK UCL Drug Discovery Institute is core funded by Alzheimer’s Research UK (ARUK-
16

Page 17 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
2018DDI-UCL). The Francis Crick Institute receives its core funding from Cancer Research UK
(FC001002), the UK Medical Research Council (FC001002), and the Wellcome Trust (FC001002).
Conflicts of Interest
There are no conflicts of interest to declare.
ACKNOWLEDGMENT
We thank our colleagues Jamie Bilsland, Will Mahy, Patricia Salinas, Fredrik Svensson, Ed Tate, J. P.
Vincent, Paul Whiting, Nicky Willis and Matthias Zebisch of our NOTUM Consortium for their support
and advice. The Cell Services and Structural Biology Science Technology Platforms (STPs) at the Francis
Crick Institute are gratefully acknowledged for their provision and purification of recombinant NOTUM.
We thank Artur Costa, Stefan Constantinou, and Kathryn Costelloe for assistance with the biochemical
screening, and Abil Aliev and Kersti Karu at the UCL Department of Chemistry for spectroscopic and
analytical services. The authors thank Frank von Delft and the XChem team for enabling the
crystallographic fragment screen (beamtime proposal IB16814) and the staff of Diamond Light Source
beamlines I04-1 and I24 for assistance with data collection (BAG application MX14744). ADME studies
reported in this work were independently performed by GVK Biosciences (Hyderabad, India) and/or
Cyprotex (Macclesfield, UK).
ABBREVIATIONS
A, amyloid-beta; ABPP, activity-based protein profiling; AD, Alzheimer’s disease; ADME, absorption
distribution metabolism elimination; CNS, central nervous system; FP, fluorophosphonate; HLM, human
liver microsomes; LE, ligand efficiency; LLE, lipophilic ligand efficiency; MLM, mouse liver
microsomes; MPO, multiparameter optimization; NADPH, nicotinamide adenine dinucleotide phosphate;
OPTS, trisodium 8-octanoyloxypyrene-1,3,6-trisulfonate; P-gp, P-glycoprotein; PTM, post translational
modifications; SAR, structure activity relationship; SBDD, structure based drug design; TPSA,
topological polar surface area; UPLC-MS, ultra performance liquid chromatography – mass spectrometer.
17

MedChemComm
Page 18 of 21
View Article Online
DOI: 10.1039/C9MD00096H
REFERENCES
1.
2.
3.
4.
Nusse, R., Clevers, H. Wnt/-Catenin Signaling, Disease, and Emerging Therapeutic Modalities.
Cell 2017, 169, 985-999.
Niehrs, C. Function and biological roles of the Dickkopf family of Wnt modulators. Oncogene
2006, 25, 7469–7481.
Niehrs C. The complex world of WNT receptor signaling. Nat. Rev. Mol. Cell. Biol. 2012, 13,
767-779.
Malinauskas T., Jones, E. Y. Extracellular modulators of Wnt signaling. Curr. Opin. Struct. Biol.
2014, 29, 77-84.
5.
6.
Polakis, P. Wnt signaling in cancer. Cold Spring Harb. Perspect. Biol. 2012, 4.
Liu, C.-C. et al. Deficiency in LRP6-Mediated Wnt Signaling Contributes to Synaptic
Abnormalities and Amyloid Pathology in Alzheimer’s Disease. Neuron 84, 63–77 (2014).
Marzo, A. et al. Reversal of synapse degeneration by restoring Wnt signaling in the adult
hippocampus. Curr. Biol.2016, 26, 2551-2561.
7.
8.
(a) Zhou, Y. et al. Canonical WNT signaling components in vascular development and barrier
formation. J. Clin. Invest. 124, 3825–3846 (2014).
(b) Liu, L, Wan, W, Xia, S., Kalionis, B., Li, Y. Dysfunctional Wnt/-catenin signaling contributes
to blood- brain barrier breakdown in Alzheimer's disease. Neurochem. Int. 75, 19-25 (2014).
Parr, C., Mirzaei, N., Christian, M. & Sastre, M. Activation of the Wnt/ β -catenin pathway
represses the transcription of the β -amyloid precursor protein cleaving enzyme (BACE1) via
binding of T-cell factor-4 to BACE1 promoter. FASEB J. 29, 623–635 (2015).
9.
10. Janda, C. Y., Garcia, K. C. Wnt acylation and its functional implication in Wnt signaling regulation.
Biochem. Soc. Trans. 43, 211–216 (2015).
11. Gerlitz, O., Basler, K. Wingful, an extracellular feedback inhibitor of Wingless. Genes Dev. 16,
1055–1059 (2002).
12. Kakugawa, S. et al. NOTUM deacylates Wnt proteins to suppress signaling activity. Nature 519,
187–192 (2015).
13. Zhang, X. et al. NOTUM is required for neural and head induction via Wnt deacylation, oxidation,
and inactivation. Dev. Cell 32, 719–30 (2015).
14. (a) Tarver, J. E. et al. Stimulation of cortical bone formation with thienopyrimidine based inhibitors
of NOTUM Pectinacetylesterase. Bioorg. Med. Chem. Lett. 26, 1525–1528 (2016).
(b) Brommage, R. et al. NOTUM inhibition increases endocortical bone formation and bone
18

Page 19 of 21
MedChemComm
View Article Online
DOI: 10.1039/C9MD00096H
strength. Bone Research 7, 2 (2019).
15. Manallack, D. T., The pKa Distribution of Drugs: Application to Drug Discovery. Perspect.
Medicin. Chem. 2007, 1, 25–38.
16.
Wager, T. T., Chandrasekaran, R. Y., Hou, X., Troutman, M. D., Verhoest, P. R., Villalobos,
A., Will, Y. Defining desirable central nervous system drug space through the alignment of
molecular properties, in vitro ADME, and safety attributes. ACS Chem Neurosci. 2010, 1, 420-34.
17.
18.
19.
Manallack, D. T., Prankerd, R. J., Yuriev, E., Oprea, T. I., Chalmers, D. K. The significance of
acid/base properties in drug discovery. Chem. Soc. Rev. 2013, 42, 485-496.
Charifson, P. S., Walters W. P. Acidic and basic drugs in medicinal chemistry: a perspective. J.
Med. Chem. 2014, 57, 9701-9717.
Di, L., Rong, H., Feng, B. Demystifying brain penetration in central nervous system drug
discovery. Miniperspective. J. Med. Chem. 2013, 56, 2-12.
20. Suciu, R. M., Cognetta, A. B., Potter, Z. E., Cravatt, B. F. Selective Irreversible Inhibitors of the
Wnt-Deacylating Enzyme NOTUM Developed by Activity-Based Protein Profiling. ACS Med.
Chem. Lett. 9, 563–568 (2018).
21. https://www.diamond.ac.uk/Instruments/Mx/Fragment-Screening/Fragment-Libraries.html.
Accessed 18th October 2018.
22.
Barbosa, J., Carson, K. G., Gardyan, M. W., He, W., Lombardo, V., Pabba, P., Tarver Jr, J.
Inhibitors of notum pectinacetylesterase and methods of their use. US20120065200.
23. Rankovic, Z. CNS Drug Design: Balancing Physicochemical Properties for Optimal Brain
Exposure. J. Med. Chem. 2015, 58 (6), 2584-2608.
24. Wager, T. T. et al. Moving beyond Rules: The Development of a Central Nervous System
Multiparameter Optimization (CNS MPO) Approach To Enable Alignment of Druglike Properties.
ACS Chem. Neurosci. 2010, 1 (6), 435-449.
25. Rankovic, Z., CNS Physicochemical Property Space Shaped by a Diverse Set of Molecules with
Experimentally Determined Exposure in the Mouse Brain. J. Med. Chem. 2017, 60 (14), 5943-
5954.
26. Liu, Y., Patricelli, M. P., Cravatt, B. F. Activity-based protein profiling: The serine hydrolases.
Proc. Natl. Acad. Sci. USA 1999, 96, 14694-14699.
27. Leeson, P. D., Springthorpe, B. The influence of drug-like concepts on decision-making in
19

MedChemComm
Page 20 of 21
View Article Online
DOI: 10.1039/C9MD00096H
medicinal chemistry. Nat. Rev. Drug. Discov. 2007, 6, 881-890.
28.
29.
30.
Johnson, T. W., Gallego, R. A., Edwards, M. P. Lipophilic Efficiency as an Important Metric in
Drug Design. J. Med. Chem. 2018, 61, 6401-6420.
Auld, D. S., Thorne, N., Nguyen, D.-T., Inglese, J. A Specific Mechanism for Non-Specific
Activation in Reporter-Gene Assays ACS Chem. Biol. 2008, 3, 463-470.
Thorne, N., Shen, M., Lea, W. A., Simeonov, A., Lovell, S., Auld, D. S., Inglese, J. Firefly
luciferase in chemical biology: A compendium of inhibitors, mechanistic evaluation of
chemotypes, and suggested use as a reporter. Chem Biol. 2012, 19, 1060–1072. See also:
https://pubchem.ncbi.nlm.nih.gov/substance/7972821#section=Top Accessed 10th December
2018.
31.
Chacko, S. et al. Expanding Benzoxazole-Based Inosine 5′-Monophosphate Dehydrogenase
(IMPDH) Inhibitor Structure–Activity As Potential Antituberculosis Agents. J. Med. Chem. 61,
4739–4756 (2018).
20

Page 21 of 21
MedChemComm
TOC graphic:
[Editable HD figure].
Discovery of 2-phenoxyacetamides as Inhibitors of the Wnt-depalmitoleating enzyme NOTUM from an X-ray Fragment
Screen
Benjamin N. Atkinson, David Steadman, Yuguang Zhao, James Sipthorp, Luca Vecchia, Reinis R. Ruza, Fiona Jeganathan, Georgie Lines, Sarah Frew,
Amy Monaghan, Svend Kjaer, Magda Bictash, E. Yvonne Jones, Paul V. Fish.
Optimization of fragment hit 3 identified isoquinoline 45 as a potent inhibitor of NOTUM with an unexpected flipped binding mode.
SAR evolution
SBDD
N
Me
O
N
Me
O
O
O
N
N
H
H
physicochemical
properties for CNS
Fragment hit 3
45
NOTUM IC50 33,000 nM
LE 0.35; LLE 2.8
NOTUM IC50 85 nM
LE 0.45; LLE 4.1
3 (green) and 45 (blue)
21