Discovery of Potent and Selective MTH1 Inhibitors for Oncology: Enabling Rapid Target (In)Validation

Article abstract of DOI:10.1021/acsmedchemlett.9b00420

We describe the discovery of three structurally differentiated potent and selective MTH1 inhibitors and their subsequent use to investigate MTH1 as an oncology target, culminating in target (in)validation. Tetrahydronaphthyridine 5 was rapidly identified as a highly potent MTH1 inhibitor (IC₅₀ = 0.043 nM). Cocrystallization of 5 with MTH1 revealed the ligand in a φ-cis-N-(pyridin-2-yl)acetamide conformation enabling a key intramolecular hydrogen bond and polar interactions with residues Gly34 and Asp120. Modification of literature compound TH287 with O- and N-linked aryl and alkyl aryl substituents led to the discovery of potent pyrimidine-2,4,6-triamine 25 (IC₅₀ = 0.49 nM). Triazolopyridine 32 emerged as a highly selective lead compound with a suitable in vitro profile and desirable pharmacokinetic properties in rat. Elucidation of the DNA damage response, cell viability, and intracellular concentrations of oxo-NTPs (oxidized nucleoside triphosphates) as a function of MTH1 knockdown and/or small molecule inhibition was studied. Based on our findings, we were unable to provide evidence to further pursue MTH1 as an oncology target.

Full text of DOI:10.1021/acsmedchemlett.9b00420

Letter
pubs.acs.org/acsmedchemlett
Cite This: ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX
Discovery of Potent and Selective MTH1 Inhibitors for Oncology:
Enabling Rapid Target (In)Validation
Julie Farand,*,† Jeffrey E. Kropf,‡ Peter Blomgren,‡ Jianjun Xu,‡ Aaron C. Schmitt,‡
Zachary E. Newby,† Ting Wang,† Eisuke Murakami,† Ona Barauskas,† Jawahar Sudhamsu,† Joy Y. Feng,†
Anita Niedziela-Majka,† Brian E. Schultz,† Karen Schwartz,† Serge Viatchenko-Karpinski,†
Dmytro Kornyeyev,† Adam Kashishian,‡ Peidong Fan,† Xiaowu Chen,† Eric B. Lansdon,†
Michael O. Ports,‡ Kevin S. Currie,‡ William J. Watkins,† and Gregory T. Notte†
†Gilead Sciences, Inc. 333 Lakeside Drive, Foster City, California 94404, United States
‡Gilead Sciences, Inc. 199 East Blaine Street, Seattle, Washington 98102, United States
S
* Supporting Information
ABSTRACT: We describe the discovery of three structurally
differentiated potent and selective MTH1 inhibitors and their
subsequent use to investigate MTH1 as an oncology target,
culminating in target (in)validation. Tetrahydronaphthyridine 5
was rapidly identified as a highly potent MTH1 inhibitor (IC50
=
0.043 nM). Cocrystallization of 5 with MTH1 revealed the
ligand in a Φ-cis-N-(pyridin-2-yl)acetamide conformation
enabling a key intramolecular hydrogen bond and polar
interactions with residues Gly34 and Asp120. Modification of
literature compound TH287 with O- and N-linked aryl and alkyl
aryl substituents led to the discovery of potent pyrimidine-2,4,6-
triamine 25 (IC50 = 0.49 nM). Triazolopyridine 32 emerged as a
highly selective lead compound with a suitable in vitro profile and
desirable pharmacokinetic properties in rat. Elucidation of the
DNA damage response, cell viability, and intracellular concentrations of oxo-NTPs (oxidized nucleoside triphosphates) as a
function of MTH1 knockdown and/or small molecule inhibition was studied. Based on our findings, we were unable to provide
evidence to further pursue MTH1 as an oncology target.
KEYWORDS: MutT homologue 1(MTH1) inhibition, 8-oxo-dGTP, tetrahydronaphthyridine, triazolopyridine, DNA damage
n 2018, 59 new molecular entities were approved by the
IFDA’s Center for Drug Evaluation and Research (CDER),
27% of which were developed as oncology therapies (12 new
chemical entities and 4 biologics).1,2 Between 2014 and 2018,
cancer drugs consistently achieved the highest number of FDA
approvals by therapeutic area, accounting for nearly 25% of all
approvals. Addressing the need for safe and effective cancer
therapies has been challenging, and prosecuting drug discovery
programs for oncology targets requires rigorous preclinical
target validation. In 2014, two back-to-back Nature publica-
tions described MTH1 as cancer phenotypic lethal with great
promise as a small molecule targeted therapy for oncology.3,4
MutT homologue 1 (MTH1 or NUDT1) is a member of the
Nudix phosphohydrolase superfamily of enzymes and is a
pyrophosphatase that selectively recognizes oxidized purine
nucleoside triphosphates, thus converting 8-oxo-dGTP or 2-
OH-dATP to 8-oxo-dGMP or 2-OH-dAMP. In normal cells,
MTH1 has been described as a sanitizing enzyme capable of
preventing the incorporation of oxo-dNTPs into DNA. Cancer
cells, however, have an altered redox state with higher levels of
reactive oxygen species (ROS) leading to elevated levels of
oxidative stress. Disruption of nucleotide pool homeostasis and
dysregulation of DNA repair by MTH1 inhibition in tumor
cells could increase the incorporation of mismatched oxidized
bases into DNA, thus leading to transversion mutations, DNA
damage, and cancer cell death.5 Gad et al.3 demonstrated that
MTH1 inhibition with small molecules TH287 and TH588, or
via genetic knockdown, resulted in DNA damage and reduced
clonogenic survival and cell viability. Based on these
compelling results, we initiated a preclinical research program
to evaluate MTH1 as a potential target for oncology. At the
onset of our research, only the small molecules shown in
Figure 1 were known in the literature to demonstrate
biochemical inhibition of MTH1.3,4,6 When TH287, TH588,
SCH51344, and (S)-crizotinib were profiled in our three-day
cell viability assay, we observed discrepancies between their
Special Issue: Women in Medicinal Chemistry
Received: September 12, 2019
Accepted: November 12, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Figure 1. Structures of literature MTH1 inhibitors. Internal biochemical potencies (8-oxo-dGTP as substrate) and three-day cell viability
(CellTiter-Glo) in U2OS and SW480 cells are reported (*literature values).
Figure 2. Representative structures of a selection of reported tool compounds used for target validation studies and their MTH1 biochemical
potencies (8-oxo-dGTP as substrate). Cocrystal structures of AZ compound 15,7 AZ compound 19,7 and BAY-70710 bound to MTH1 (PDB code
5ANU, 5ANV, 5NHY, respectively) revealed hydrogen bond donors and acceptors (blue) engaging Asp119, Asp120, Gly34, and/or Asn33 in the
MTH1 active site.
Figure 3. Our initial chemistry efforts yielded highly potent MTH1 inhibitors. For number of replicates and standard error of the mean, see the
Supporting Information.
biochemical potencies (nM) and their phenotypic cellular
effects (μM) in U2OS and SW480 cell lines. Additionally,
Huber et al.4 had reported that MTH1 overexpression rescued
SW480 cells from the cell killing effects of SCH51344 but not
(S)-crizotinib. Thus, our initial efforts aimed to address these
discrepancies by concurrently (1) expressing the full length
human MTH1 protein for structural studies, (2) identifying
structurally differentiated, selective, and potent small molecule
tool compounds, and (3) performing validation studies in
shRNA-mediated MTH1 knockdown cell lines.
Several other research groups also initiated MTH1 programs
based upon the same literature reports, systematically executed
elegant target validation studies, and have since shared their
findings.7−10 Common means to elucidate the cellular effect of
MTH1 inhibition included, but are not limited to, thermal
stabilization assays to assess target engagement, kinase profiling
to evaluate off-target liabilities, cell viability and growth
inhibition assays, immunofluorescence/immunoblots of DNA
damage response signaling markers, proteomics, cell cycle
analysis, and genetic knockdown of MTH1. Importantly, the
identification of structurally differentiated MTH1 tool
compounds proved to be essential for all studies performed
(Figure 2). Herein, we share a similar approach and present
the discovery of three potent and selective MTH1 chemotypes,
one of which achieved unprecedented levels of biochemical
potency (tetrahydronaphthyridine 5, IC50 = 0.043 nM).
Furthermore, we share methodology that was developed to
measure intracellular concentrations of oxo-NTPs as an in vitro
biomarker of MTH1 target engagement.
Examination of the published MTH1 cocrystal structures
with 8-oxo-dGMP and inhibitors TH287 and TH588 revealed
common key hydrogen bonding interactions with active site
residues Asp119, Asp120, and Asn33.11,3 After establishing that
the 2,3-dichlorophenyl substituent in TH287 (MTH1 IC50
=
4.1 nM) could be replaced with 2,3-dimethylphenyl without
significant loss in potency (MTH1 IC50 of 1 = 7.2 nM), we set
out to further understand the relationship between the
nitrogen-mediated hydrogen bonds in the TH-series and
MTH1 biochemical potency by conducting a series of nitrogen
deletions, acetylations, and ring annulations (Figure 3). The
>1000-fold loss in potency upon removal of N3 (2 vs 1) was
suggestive of disruption of a key hydrogen bond. However, one
B
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
order of magnitude in potency was restored upon incorpo-
ration of a bicycle in the form of 3, synthesized as illustrated in
Scheme 1. These results prompted us to prepare naphthyridine
N6 of naphthyridine 4 and N6 of tetrahydronaphthyridine 5
had significantly different pKa values, 3.5 and 7.5, respectively,
yet both compounds showed similar permeability rates at
physiological pH with no efflux (AB/BA = 37/45 × 10−6 and
35/32 × 10−6 cm/s, respectively). The excellent permeability
of 5, together with an identical logD7.4 (2.5) to its nonbasic
congener 4 suggested the presence of an intramolecular
hydrogen bond between protonated N6 and the acetamide
carbonyl in solution (although this was not evident in a crystal
structure of the hydrochloride salt (see Supporting Informa-
tion)). As in the transposition of 3 to 2, removal of the bicycle
in 5 to give the aminopyridine 6 resulted in an order of
magnitude loss in potency (IC50 = 0.33 nM). This residual
potency was abolished when the pyridine core was replaced
with phenyl (data not shown), demonstrating the critical role
of the DMAP-like basic nitrogen in 5 and 6 and its ability to
facilitate the adoption of nearly coplanar acetamide con-
formations via intramolecular hydrogen bonds.
In light of the unprecedented potency of tetrahydronaph-
thyridine 5, we confirmed that it is a tight reversible binder
(Ki(app) = 5.2 0.9 pM, Ki = 1.7 pM), and no acetylation of
MTH1 was observed. The compound is chemically stable at
pH 6, 7.5, and 9 at 40 °C, is unreactive with glutathione, and
displays desirable kinetic solubility at pH 2 and 7 (97 and 70
μM, respectively). A preliminary screen revealed no significant
affinity for kinases (S(35) = 0 at 10 μM; 97 kinase panel).12
However, 5 was not fully stable in human and rat plasma (t1/2
= 252 and 446 min, respectively) and predicted hepatic
clearance from microsomal stability assays approached liver
blood flow in human (0.89 L/h/kg) and rat (3.5 L/h/kg).
Metabolite identification studies in human liver microsomes
indicated deacetylation as the major metabolic pathway (72%),
followed by oxidative metabolism of the resulting 7-amino-
tetrahydronapthyridine.
a
Scheme 1. General Synthesis of Tetrahydronaphthyridines
a
Reagents and conditions: (a) ArB(OH)2, Pd(PPh3)4, 1,4-dioxane/
water (2:1), 80 °C, 1 h, 96%; (b) RC(O)NH2, Pd2(dba)3, t-BuXPhos,
Cs2CO3, 1,4-dioxane, 100 °C, 18 h, 20−82%; (c) PtO2, H2 (40 psi), 4
M HCl in 1,4-dioxane, 15 min, 20−76%; (d) when R = OtBu, TFA/
DCM (3:1), rt, 5 h, 12%.
and tetrahydronaphthyridine acetamides 4 and 5 to interrogate
the electronic preferences of the ligand’s nitrogen atoms with
Asp residues 119 and 120. Naphthyridine 4 was moderately
potent (MTH1 IC50 = 81 nM), while, remarkably, tetrahy-
dronaphthyridine 5 demonstrated a 1800-fold improvement in
potency (IC50 = 43 pM, LLE = 7.9) relative to 4.
Cocrystallization of 4 with MTH1 revealed key hydrogen
bonding interactions with both aspartates, as observed with
TH287, a π−π stacking interaction with Trp 117; in making a
hydrogen bond to a water molecule, the acetamide of 4 is 23°
out of the plane of the aryl system (Figure 4A). Interestingly,
Thus, we attempted to replace the acetamide moiety to
improve metabolic stability while maintaining the Gly34
interaction (Table 1). Ureas 13 and 14 were among the
most potent analogues, yet solely lowering the logD was
insufficient to achieve full stability. Compounds 16, 17, and 3
in which the amide was replaced or absent demonstrated that
reasonable stability could be achieved, albeit at the expense of
potency. We also investigated O- and N-linked aryls and
variously substituted alkyls in the pyrimidine-2,4-diamine
series (Table 2). Compounds 19 and 21 validated that
heteroatoms were tolerated in the linker, albeit with a slight
loss in potency. Further optimization toward (S)-ethoxyphenyl
Figure 4. Cocrystal structures of MTH1 with inhibitors. MTH1
(gray) is drawn from the complex with compound 5. (A) Overlay of 4
(yellow, 1.5 Å, PDB code 6US3) and 5 (cyan) illustrating Φ-cis
dihedral angles of 23° and 6°, respectively, and the resulting shifts of
the 2,3-dimethyphenyl moieties. Unlike tetrahydronaphthyridine 5,
naphthyridine 4 engages Asn33 via a water (yellow sphere)-mediated
interaction. (B) Cocrystal structure of 5 (cyan) bound to MTH1 (1.8
Å, PDB code 6US2) showing the shifted register of hydrogen bond
interactions relative to TH287 (orange, PDB code 4N1T). No
significant changes to the protein were observed between MTH1
complexed to 4 and 5.
analogue 23, (S)-N-methyl-1-phenylethan-1-amine 25 (IC50
=
0.49 nM, LLE = 7.3), and piperidine derivatives 26, 29, and 30
successfully restored MTH1 IC50 to ≤1 nM. Although
excellent biochemical potency, kinetic solubility, and perme-
ability were achievable, this lipophilic series suffered from poor
metabolic stability and was intolerant to polar and logD
lowering substituents. The 2-amino and 4-methylamino groups
of 1 showed little tolerance to change, emphasizing the
importance of preserving optimal interactions with Asp120 and
Asp119.
Our third series of small molecule MTH1 inhibitors was
designed to explore interactions with Asn33 in the active site.
This residue has been reported to be critical for 8-oxo-dGMP
binding and specificity, interacting with both the 2-amino
group and N3 of the ligand. Since the 2,4-diaminopyrimidine
of TH287 shared similar interactions, we proposed 2-
aminotriazolopyridine and 2-aminobenzimidazole cores to
tetrahydronaphthyridine 5 when cocrystallized with MTH1
displayed the same key interactions as 4, with the added
benefit of adopting a nearly coplanar acetamide conformation
(6° dihedral angle) and permitting a unique interaction with
the backbone NH of Gly34 in the absence of a water molecule
(Figure 4B).
C
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Table 1. Acetamide Moiety Replacements in the
Table 2. MTH1 Biochemical Potencies with O- and N-
a
Tetrahydronaphthyridine Series in an Attempt To Improve
Metabolic Stability while Maintaining MTH1 Biochemical
Linked R2 Groups
a
Potency
MTH1 IC50
(nM)
Human MS Cl
(L/h/kg)
Compound
R1
5
8
9
-NHC(O)Me
-NHC(O)Et
0.043
0.17
0.06
0.61
<0.05
0.11
<0.05
<0.05
0.15
62
0.89
0.88
0.38
0.73
0.45
0.81
0.33
0.28
0.39
0.18
0.11
0.23
-NHC(O)cPr
-NHC(O)iPr
-NHC(O)cBu
-NHC(O)CF2H
-NHC(O)NH2
-NHC(O)NHMe
-NHC(O)OMe
-NHC(=NH)Me
-NHSO2Me
10
11
12
13
14
15
16
17
3
773
952
-NH2
a
For number of replicates and standard error of the mean, see the
Supporting Information.
interrogate their potential to interact with Asn33 and yield
potent inhibitors (Table 3). We discovered that 2-amino-
triazolopyridine 32 maintained good potency (MTH1 IC50
=
13 nM, LLE = 6.0), while the R5 phenyl derivative 31 lost 9-
fold in potency. Cocrystallization of 32 with MTH1 confirmed
key hydrogen bonding interactions with Asn33, Asp119,
Asp120, and an additional π−π stacking interaction with
Trp117 (Figure 5). Although the triazolopyridine 34 was
designed to engage Asp119, the additional amino group at R3
failed to yield a more potent compound. Acetylation of 32
resulted in a 35-fold loss in potency (data not shown), and the
triazolopyridine regioisomer 40 was similarly less potent.
Methylation at the R4 position yielded the most potent
compound in this series (MTH1 IC50 of 36 = 4.1 nM).
Triazolopyridine 32 was found to be stable in human liver
microsomes (pred. hep. Cl = 0.11 L/h/kg) and moderately
stable in rat liver microsomes (pred. hep. Cl = 0.86 L/h/kg).
When administered to Sprague−Dawley rats at 5 mg/kg PO
and 1 mg/kg IV, compound 32 showed excellent oral
bioavailability (F = 86%), good in vitro/in vivo correlation
(Cl = 0.73 L/h/kg), moderate volume (2.7 L/kg), and a half-
life of 3.2 h. It lacks inhibition of hERG and five CYP isoforms
(IC50 > 25 μM), while demonstrating high permeability (Caco-
2 AB/BA = 41/34 × 10−6 cm/s) and high solubility at pH 2
and 7 (93/74 μM). In addition, it had low affinity for kinases
(S(35) = 0.01 at 10 μM; 97 kinase panel) and did not activate
human PXR at 15 μM. The 2-aminobenzimidazole 37 also
emerged as a useful lead, with an in vitro MTH1 IC50 of 15 nM
and good metabolic stability (pred. hep. Cl = 0.11 and 0.22 L/
h/kg in human and rat, respectively). With three highly potent
scaffolds capable of inhibiting MTH1, we then selected tool
molecules 5, 25, 26, 32, and 37 to further evaluate their
antiproliferative effects and ability to induce DNA damage in
cancer cell lines.
a
For number of replicates and standard error of the mean, see the
Supporting Information.
viability (CTG) assay in U2OS cells. We were surprised by the
observation that our potent inhibitors, particularly tetrahy-
dronaphthyridine 5 (IC50 = 43 pM, CTG EC50 = 8.0 μM,
Figure 6A and Table 4), did not elicit a similar or greater
antiproliferative effect than TH287 (CTG EC50 = 0.7 μM).
Several subsequent 3-day CTG experiments containing
additional positive controls (TH287, TH588, and/or (S)-
crizotinib) showed our compounds were consistently less
effective than the tool compounds listed above. An in vitro
assessment of cytotoxicity was conducted in various cell lines,
and we found that TH287 exhibited cytotoxic effects in MT4
cells (CC50 = 726 nM), whereas 5 and 32 showed greatly
reduced cytotoxicity (CC50 = 19 and >50 μM, respectively).13
Since all compounds tested had relatively similar binding in
cell culture media, we hypothesized that the differential
cytotoxicity could indicate that MTH1 inhibition alone was
insufficient for antiproliferative activity, and we proceeded to
further characterize the effect of this structurally diverse set of
MTH1 inhibitors on the purported downstream markers of
MTH1 inhibition.
To characterize the effect of these novel MTH1 inhibitors
on the growth of cancer cells, we employed a 3-day cell
D
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Table 3. Structural Modifications to Improve Potency in the
a
Triazolopyridine and Benzimidazole Series
Compound Core
R3
R4
R5
MTH1 IC50 (nM)
31
32
33
34
35
36
37
38
39
40
X
X
X
X
X
X
Y
Y
Y
Z
H
H
H
NH2
H
H
H
H
H
H
H
H
H
Me
Me
H
Me
OMe
H
Ph
936
13
40
207
946
4.1
15
20
13
467
2,3-diClPh
2,3-diMePh
2,3-diClPh
piperidinyl
2,3-diClPh
2,3-diClPh
2,3-diClPh
2,3-diClPh
2,3-diClPh
H
a
For number of replicates and standard error of the mean, see the
Supporting Information.
Figure 5. (A) Cocrystal structure of compound 32 (yellow) with
MTH1 (1.95 Å, PDB code 6US4) and TH287 (orange) overlaid,
revealing common interactions with Asn33, Asp119, and Asp120.
Compound 32 also forms a π−π stacking interaction with Trp117
(not shown in this figure). (B) Overlay of 5 (cyan) and 32 (yellow)
showing Gly34 has moved away to allow Asn33 to interact with 32.
Figure 6. (A) Three-day cell viability assay in U2OS cells using
CellTiter-Glo readout. (B) Expression of p-p53 and β-actin in U2OS
cells treated with TH287, TH588, and compound 5 at 5 μM and
mitoxantrone at 2.25 μM, measured using Peggy Sue Western blot.
(C) Induction of DNA damage foci in U2OS cells with MTH1
inhibitors. Cells were incubated for 72 h with DMSO, MTH1
inhibitors (20 μM), and H2O2 (150 μM). Ninety to 120 nuclei were
counted per experiment (*p < 0.001).
Induction of DNA damage and activation of the DNA
damage response pathway have been reported to be associated
with inhibition of MTH1, and we interrogated both end points
in cellular assays.3 Activation of p53 in response to DNA
damage has been well characterized and recently reviewed.14
Using Simple Western, we evaluated the effect of TH287 and
several structurally diverse MTH1 inhibitors on p53
phosphorylation in U2OS cells treated for 4 and 24 h. We
found that treatment with TH287, TH588, and the positive
control (mitoxantrone) resulted in activation of p53, while the
newly synthesized MTH1 inhibitors 5 (Figure 6B) and 32
(Supporting Information) failed to do so. As a final step to
profile the cellular phenotypes of MTH1 inhibition, we used
confocal microscopy to determine the extent of γH2AX foci
formation in U2OS cells treated with MTH1 inhibitors. A
normal cellular response to initiate DNA double-stranded
break repair involves the recruitment and phosphorylation of
multiple H2AX histones in close proximity to the site of
repair.15 As expected, treatment with both TH287 and
hydrogen peroxide resulted in a larger number of γH2AX
foci observed in each cell nucleus (Figure 6C). However, when
5, 32, and 26 were tested at a concentration of 20 μM,
induction of foci was not observed. Notably, these compounds
Table 4. MTH1 Biochemical Potencies (8-oxo-dGTP as
Substrate) and Cellular Potencies (Three-day Cell Viability
Assay in U2OS Cells) of Representative Tool Compounds
a
from Our Various Chemotypes
MTH1 IC50
(nM)
Cell viability
EC50 (nM)
Chemotype
Compound
2,4-Diaminopyrimidine
Tetrahydronaphthyridine
2,4,6-Triaminopyrimidine
2-Aminotriazolopyridine
2-Aminobenzimidazole
TH287
5
25
32
37
4.1
699
0.043
0.49
13
8048
24916
20323
14416
15
a
For number of replicates and standard error of the mean, see the
Supporting Information.
E
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
all demonstrated good cellular permeability and were tested in
the foci formation assay at concentrations >1000-fold above
their respective IC50 values.
Synthetic procedures and characterization of MTH1
inhibitors, small molecule X-ray structures of 4 and 5,
MTH1 crystallography, data collection and refinement
statistics, kinase selectivity profile, biochemical and
cellular assay protocols, and intracellular measurements
of oxo-NTPs (PDF)
Although the collective results from the cellular assays
described thus far clearly show the compounds described
herein failed to recapitulate the cell killing effect of previously
described MTH1 inhibitors, the end points assessed are all
several steps downstream, and the lack of a direct cellular
readout for MTH1 inhibition hampered our ability to make
definitive conclusions about the cellular activity of our series.
To address this unanswered question, we developed method-
ology to measure the intracellular concentration of oxidized
nucleoside triphosphates (oxo-NTPs), as inhibition of MTH1
was hypothesized to increase oxo-NTP levels in cells (Table 5,
Accession Codes
Cocrystal structures of MTH1 with 4, 5 and 32 have been
deposited with the Protein Data Bank under Accession ID
codes 6US3, 6US2 and 6US4.
AUTHOR INFORMATION
Corresponding Author
*E-mail: julie.farand@gilead.com. Phone: +1-650-372-7668.
■
Table 5. Intracellular Concentrations of Oxo-NTPs in the
Presence of Control shRNA, MTH1 shRNA, and Small
Molecule MTH1 Inhibitor 37
ORCID
Julie Farand: 0000-0001-9991-4328
Brian E. Schultz: 0000-0001-9757-2547
Eric B. Lansdon: 0000-0001-9461-1475
a
Intracellular Concentration (pmol/million cells)
Author Contributions
Cell Type
8-oxo-dGTP 8-oxo-rGTP dGTP
rGTP
b
d
The manuscript was written through the contributions of all
authors. All authors have given approval to the final version of
the manuscript.
SW480
0.004
0.008
0.006
0.010
-
6.31
17.0
7.43
5.58
12.0
2790
4370
ND
ND
ND
c
U2OS
0.058
0.034
0.022
0.040
e
U2OS, NT shRNA
f
U2OS, MTH1 shRNA
U2OS + 20 μM 37
Notes
0.007
The authors declare no competing financial interest.
a
b
c
d
n = 1. SW480:12 × 106 cells. U2OS: 30 × 106 cells. LOQ = 0.004
e
f
pmol/million cells. Nontargeting (NT). About 50% knockdown
when harvested.
ACKNOWLEDGMENTS
■
The authors would like to thank the staff at the Advanced
Light Source for help in data collection. The Advanced Light
Source is a Department of Energy Office of Science User
Facility under Contract No. DE-AC02-05CH11231. The
authors also wish to thank Professor Arnold L. Rheingold for
providing small molecule X-ray crystallography data.
see the Supporting Information for methods). A large number
of cells were required for this analysis as 8-oxo-dGTP and 8-
oxo-rGTP levels were consistently low and near the limit of
quantification. However, intracellular dGTP and rGTP
concentrations remained within ranges reported in the
literature.16 U2OS cells treated with either MTH1 shRNA or
representative small molecule inhibitor 37 showed no evidence
of increased 8-oxo-dGTP or 8-oxo-rGTP levels in vitro,
suggesting that MTH1 inhibition may not disrupt homeostasis
of the oxo-NTP pool.
In this study, the validity of MTH1 as an oncology target
was assessed by the discovery and characterization of
structurally diverse chemotypes in conjunction with specific
assessment of cellular effects. Despite nano- and subnanomolar
potencies, the tool compounds did not elicit the desired
antiproliferative effects, failed to induce p53 activation, and did
not increase foci formation or 8-oxo-dGTP levels in U2OS
cells, leading us to conclude that MTH1 alone is insufficient
for cancer cell death.17,18 During the months following the
conclusion of our MTH1 efforts, several articles, perspectives,
and comments were published corroborating our findings. The
use of dissimilar chemotypes for phenotypic confirmation was
evident in many of these programs and can be a valuable tactic
at an early stage of a project. Publication of efforts such as
these furthers our collective understanding of methods for in
vitro target validation.
ABBREVIATIONS
■
FDA, Food and Drug Administration; 8-oxo-dGMP, 8-oxo-2′-
deoxyguanosine-5′-monophosphate; dNTP, deoxynucleoside
triphosphate; shRNA, short hairpin ribonucleic acid; Asp,
aspartic acid; Asn, asparagine; Trp, tryptophan; Gly, glycine;
LLE, lipophilic ligand efficiency; DMAP, 4-dimethylaminopyr-
idine; PXR, pregnane X receptor
REFERENCES
■
(1) Novel Drug Approvals for 2018. https://www.fda.gov/drugs/
new-drugs-fda-cders-new-molecular-entities-and-new-therapeutic-
biological-products/novel-drug-approvals-2018 (accessed September
12, 2019).
(2) Mullard, A. 2018 FDA Drug Approvals. Nat. Rev. Drug Discovery
2019, 18, 85−89.
(3) Gad, H.; Koolmeister, T.; Jemth, A. S.; Eshtad, S.; Jacques, S. A.;
̈
̈
Strom, C. E.; Svensson, L. M.; Schultz, N.; Lundback, T.; Einarsdottir,
̈
̈
B. O.; Saleh, A.; Gokturk, C.; Baranczewski, P.; Svensson, R.;
̈
Berntsson, R. P.; Gustafsson, R.; Stromberg, K.; Sanjiv, K.; Jacques-
Cordonnier, M. C.; Desroses, M.; Gustavsson, A. L.; Olofsson, R.;
Johansson, F.; Homan, E. J.; Loseva, O.; Brautigam, L.; Johansson, L.;
̈
̈
Hoglund, A.; Hagenkort, A.; Pham, T.; Altun, M.; Gaugaz, F. Z.;
Vikingsson, S.; Evers, B.; Henriksson, M.; Vallin, K. S.; Wallner, O. A.;
̈
̈
́
Hammarstrom, L. G.; Wiita, E.; Almlof, I.; Kalderen, C.; Axelsson, H.;
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.9b00420.
■
̈
Djureinovic, T.; Puigvert, J. C.; Haggblad, M.; Jeppsson, F.; Martens,
S
U.; Lundin, C.; Lundgren, B.; Grenelli, I.; Jensen, A. J.; Artursson, P.;
Nilsson, J. A.; Stenmark, P.; Scobie, M.; Berglund, U. W.; Helleday, T.
MTH1 Inhibition Eradicates Cancer by Preventing Sanitation of the
dNTP pool. Nature 2014, 508, 215−221.
F
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(4) Huber, K. V.; Salah, E.; Radic, B.; Gridling, M.; Elkins, J. M.;
̈
̈
̈
Stukalov, A.; Jemth, A. S.; Gokturk, C.; Sanjiv, K.; Stromberg, K.;
Pham, T.; Berglund, U. W.; Colinge, J.; Bennett, K. L.; Loizou, J. I.;
Helleday, T.; Knapp, S.; Superti-Furga, G. Stereospecific Targeting of
MTH1 by (S)-Crizotinib as an Anticancer Strategy. Nature 2014, 508,
222−227.
(5) Curtin, N. J. DNA Repair Dysregulation from Cancer Driver to
Therapeutic Target. Nat. Rev. Cancer 2012, 12, 801−817.
(6) Streib, M.; Kraling, K.; Richter, K.; Xie, X.; Steuber, H.; Meggers,
̈
E. An Organometallic Inhibitor for the Human Repair Enzyme 7,8-
Dihydro-8-oxoguanosine Triphosphatase. Angew. Chem., Int. Ed. 2014,
53, 305−309.
(7) Kettle, J.; Alwan, H.; Bista, M.; Breed, J.; Davies, N. L.;
Eckersley, K.; Fillery, S.; Foote, K. M.; Goodwin, L.; Jones, D. R.;
Kack, H.; Lau, A.; Nissink, J. W. M.; Read, J.; Scott, J. S.; Taylor, B.;
Walker, G.; Wissler, L.; Wylot, M. Potent and Selective Inhibitors of
MTH1 Probe Its Role in Cancer Cell Survival. J. Med. Chem. 2016,
59, 2346−2361.
(8) Petrocchi, A.; Leo, E.; Reyna, N.; Hamilton, M. M.; Shi, X.;
Parker, C. A.; Mseeh, F.; Bardenhagen, J. P.; Leonard, P.; Cross, J. B.;
Huang, S.; Jiang, Y.; Cardozo, M.; Draetta, G.; Marszalek, J. R.;
Toniatti, C.; Jones, P.; Lewis, R. T. Identification of Potent and
Selective MTH1 Inhibitors. Bioorg. Med. Chem. Lett. 2016, 26, 1503−
1507.
(9) Kawamura, T.; Kawatani, M.; Muroi, M.; Kondoh, Y.; Futamura,
Y.; Aono, H.; Tanaka, M.; Honda, K.; Osada, H. Proteomic Profiling
of Small-Molecule Inhibitors Reveals Dispensability of MTH1 for
Cancer Cell Survival. Sci. Rep. 2016, 6, 26521.
(10) Ellermann, M.; Eheim, A.; Rahm, F.; Viklund, J.; Guenther, J.;
̈
Andersson, M.; Ericsson, U.; Forsblom, R.; Ginman, T.; Lindstrom, J.;
́
Silvander, C.; Tresaugues, L.; Giese, A.; Bunse, S.; Neuhaus, R.;
Weiske, J.; Quanz, M.; Glasauer, A.; Nowak-Reppel, K.; Bader, B.;
Irlbacher, H.; Meyer, H.; Queisser, N.; Bauser, M.; Haegebarth, A.;
́
́
Gorjanacz, M. Novel Class of Potent and Cellularly Active Inhibitors
Devalidates MTH1 as Broad-Spectrum Cancer Target. ACS Chem.
Biol. 2017, 12, 1986−1992.
(11) Svensson, L. M.; Jemth, A.-S.; Desroses, M.; Loseva, O.;
̈
Helleday, T.; Hogbom, M.; Stenmark, P. Crystal Structure of Human
MTH1 and the 8-Oxo-dGMP Product Complex. FEBS Lett. 2011,
585, 2617−2621.
(12) DiscoverX kinase selectivity score S(35) = (number of
nonmutant kinases with % control < 35)/(number of nonmutant
kinases tested).
(13) Feng, J. Y.; Xu, Y.; Barauskas, O.; Perry, J. K.; Ahmadyar, S.;
Stepan, G.; Yu, H.; Babusis, D.; Park, Y.; McCutcheon, K.; Perron,
M.; Schultz, B. E.; Sakowicz, R.; Ray, A. S. Role of Mitochondrial
RNA Polymerase in the Toxicity of Nucleotide Inhibitors of Hepatitis
C Virus. Antimicrob. Agents Chemother. 2016, 60, 806−817.
(14) Williams, A. B.; Schumacher, B. p53 in the DNA-Damage-
Repair Process. Cold Spring Harbor Perspect. Med. 2019, 6, a026070.
(15) Nakamura, A. J.; Rao, V. A.; Pommier, Y.; Bonner, W. M. The
Complexity of Phosphorylated H2AX Foci Formation and DNA
Repair Assembly at DNA Double-Strand Breaks. Cell Cycle 2010, 9,
389−397.
(16) Traut, T. W. Physiological Concentrations of Purines and
Pyrimidines. Mol. Cell. Biochem. 1994, 140, 1−22.
(17) Papeo, G. MutT Homolog 1 (MTH1): The Silencing of a
Target. J. Med. Chem. 2016, 59, 2343−2345.
̈
(18) Einarsdottir, B. O.; Karlsson, J.; Soderberg, E. M. V.; Lindberg,
M. F.; Funck-Brentano, E.; Jespersen, H.; Brynjolfsson, S. F.; Olofsson
Bagge, R.; Carstam, L.; Scobie, M.; Koolmeister, T.; Wallner, O.;
Stierner, U.; Warpman Berglund, U.; Ny, L.; Nilsson, L. M.; Larsson,
E.; Helleday, T.; Nilsson, J. A. A Patient-Derived Xenograft Pre-
Clinical Trial Reveals Treatment Responses and a Resistance
Mechanism to Karonudib in Metastatic Melanoma. Cell Death Dis.
2018, 810.
G
DOI: 10.1021/acsmedchemlett.9b00420
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX