Identification of the first small-molecule inhibitor of the REV7 DNA repair protein interaction

Article abstract of DOI:10.1016/j.bmc.2016.07.026

DNA interstrand crosslink (ICL) repair (ICLR) has been implicated in the resistance of cancer cells to ICL-inducing chemotherapeutic agents. Despite the clinical significance of ICL-inducing chemotherapy, few studies have focused on developing small-molec

Full text of DOI:10.1016/j.bmc.2016.07.026

Bioorganic & Medicinal Chemistry xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Identification of the first small-molecule inhibitor of the REV7 DNA
repair protein interaction
Marcelo L. Actis a, Nigus D. Ambaye a, Benjamin J. Evison a, Youming Shao b, Murugendra Vanarotti a,
Akira Inoue a, Ezelle T. McDonald a, Sotaro Kikuchi c, Richard Heath b, Kodai Hara c, Hiroshi Hashimoto c,
Naoaki Fujii a,
⇑
a Department of Chemical Biology and Therapeutics, St. Jude Children’s Research Hospital, Memphis, TN, USA
b Protein Production Facility, St. Jude Children’s Research Hospital, Memphis, TN, USA
c School of Pharmaceutical Sciences, University of Shizuoka, Shizuoka, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
DNA interstrand crosslink (ICL) repair (ICLR) has been implicated in the resistance of cancer cells to ICL-
inducing chemotherapeutic agents. Despite the clinical significance of ICL-inducing chemotherapy, few
studies have focused on developing small-molecule inhibitors for ICLR. The mammalian DNA polymerase
f, which comprises the catalytic subunit REV3L and the non-catalytic subunit REV7, is essential for ICLR.
To identify small-molecule compounds that are mechanistically capable of inhibiting ICLR by targeting
REV7, high-throughput screening and structure–activity relationship (SAR) analysis were performed.
Compound 1 was identified as an inhibitor of the interaction of REV7 with the REV7-binding sequence
of REV3L. Compound 7 (an optimized analog of compound 1) bound directly to REV7 in nuclear magnetic
resonance analyses, and inhibited the reactivation of a reporter plasmid containing an ICL in between the
promoter and reporter regions. The normalized clonogenic survival of HeLa cells treated with cisplatin
and compound 7 was lower than that for cells treated with cisplatin only. These findings indicate that
a small-molecule inhibitor of the REV7/REV3L interaction can chemosensitize cells by inhibiting ICLR.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 17 June 2016
Revised 11 July 2016
Accepted 14 July 2016
Available online xxxx
Keywords:
Interstrand crosslink repair
DNA damage response
REV7
REV3L
Polf
Chemotherapy
High-throughput screening
1. Introduction
for therapeutic intervention. However, few chemical inhibitors
that target the DDR molecules required for ICLR are currently avail-
DNA interstrand cross links (ICLs) are a principal mechanism by
which bifunctional chemotherapeutic drugs (e.g., cisplatin) elimi-
nate cancer cells. Mammalian cells have extensive mechanisms
for ICL repair (ICLR), which comprise complex networks of DNA
damage response (DDR) processes such as nucleotide excision
repair (NER), translesion synthesis (TLS) of DNA, and homologous
recombination (HR).1,2 Because cancer cells co-opt these mecha-
nisms to promote their survival when challenged by ICL-inducing
agents, each pathway involved in ICLR provides an attractive target
able, which poses a challenge in the optimal sensitization of ICL-
inducing chemotherapies.
Several DDR molecules are essential for ICLR.1 In mammalian
cells, DNA polymerase zeta (Polf) catalyzes the TLS step of
ICLR.2–6 Polf is composed of a large 350-kDa catalytic subunit
REV3L and a small 24-kDa non-catalytic subunit REV7/MAD2L2/
MAD2B (hereafter designated as REV7).7–9 Studies have identified
that genetic elimination of REVL3 can chemosensitize cancer
cells.10–12 However, the large size of mammalian REV3L hinders
the identification of Polf inhibitors, because it is difficult to pro-
duce REV3L for enzymatic assays.13 Therefore, we focused on
REV7 as a target for inhibiting Polf-mediated ICLR. REV7 is essen-
tial to recruit REV3L to the DNA replication fork.14 The depletion of
REV7 from a cell-free system derived from Xenopus egg extracts by
immunogenic deprivation significantly delayed ICL removal from a
plasmid DNA in the extracts.15 Moreover, the genetic elimination
of REV7 from cells markedly suppressed cell survival after cisplatin
treatment.4,16 These findings suggest that although REV7 has func-
Abbreviations: AlphaScreen, amplified luminescent proximity homogeneous
assay screen; BSA, bovine serum albumin; DMSO, dimethyl sulfoxide; FBS, fetal
bovine serum; FP, fluorescence polarization; GB1, protein G B1 domain; HEPES, 2-
[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid; HR, homologous recombi-
nation; HTS, high-throughput screening; ICL, interstrand crosslink; MBP, maltose
binding protein; NMR, nuclear magnetic resonance; PBS, phosphate buffered saline;
PCR, polymerase chain reaction; TAMRA, tetramethyl-6-carboxyrhodamine; TLS,
translesion DNA synthesis; WaterLOGSY, water-ligand observed via gradient
spectroscopy.
tions independent of Polf,17–22 it is
a promising target for
⇑
Corresponding author.
http://dx.doi.org/10.1016/j.bmc.2016.07.026
0968-0896/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

2
M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
chemotherapeutic sensitization.23,24 Because REV7 by itself does
not have an enzymatic activity, we focused to inhibit the REV7/
REV3L interaction.
and 10% glycerol. The overall yield was approximately 4 mg pro-
tein per liter of culture. A 100- L sample was desalted and eluted
l
by using 50% acetonitrile (Honeywell Burdick & Jackson, Morris-
town, NJ) plus 2% formic acid (Sigma–Aldrich) and then loaded
onto a nanospray emitter tip (New Objective, Woburn, MA) for sta-
tic ionization, followed by time-of-flight analysis (Waters LCT Pre-
mier XE Mass Spectrometer). The Waters MaxEnt1 software was
used for deconvolution to determine average mass of the sample,
which matched that of the His-REV7(R124A)/biotin-AviTag-
REV3L(1846–1898).
2. Materials and methods
2.1. Materials
Chemical libraries for HTS were acquired from various sources.
An additional batch of compound 1 was purchased from Chem-
Bridge Corp. (San Diego, CA) for revalidation. All other chemicals
were purchased from Sigma–Aldrich (St. Louis, MO) unless stated
otherwise. Restriction DNA endonucleases and DNA ligases were
purchased from New England Biolabs (Beverly, MA). SYBR green I
stain was purchased from Life Technologies (Carlsbad, CA). The
AlphaScreen assay kit was purchased from PerkinElmer (Waltham,
MA). Taq DNA polymerases for PCR, the PCR purification kit, and
the Plasmid Miniprep kit were purchased from Qiagen (Valencia,
CA). COS7 and HeLa cells were obtained from American Type Cul-
ture Collection (Manassas, VA) and cultured in Dulbecco’s Modified
Eagle Medium containing 10% FBS. All cells were maintained in an
incubator at 37 °C in a humidified atmosphere of 5% carbon
dioxide.
2.4. Preparation of MBP-REV7(R124A)
cDNA encoding full-length wild-type REV7 was amplified by
Taq PCR, using the primers 50-ATACCATGGATCATCATCATCATC
ATCATCATCATACCACGCTCACACGACAAG-30 and 50-ATAGTCGA
CTTATTTATCATCATCATCTTTGTAATCGCTGCCTTTATGAGCGCGC-30.
The amplicon containing a 50-octahistidine tag and a 30-FLAG tag
(for possible use for the AlphaScreen and pull-down assays) was
digested with SalI-HF and NcoI-HF (New England Biolabs), ligated
into the SalI/NcoI site of a pMAL-c5X vector (New England Biolabs)
by using the Quick Ligation kit (New England Biolabs), and
transformed into the E. coli DH5a strain (Invitrogen) according
to the manufacturer’s instructions. Positive clones of the
MBP-His-REV7-FLAG expression vector were used as templates to
generate the MBP-His-REV7(R124A)-FLAG mutant vector by
site-directed mutagenesis, using the GeneArt Site-Directed
Mutagenesis System (Invitrogen) and the primers 50-CATGTGGAG-
CAGCTGCTCGCAGCCTTCATCCTGAAGATC-30 and 50-GATCTTCAG-
GATGAAGGCTGCGAGCAGCTGCTCCACATG-30. The sequence of the
mutated vector was subsequently confirmed by Sanger sequenc-
ing, using an MBP sequencing primer (50-GATGAAGCCCTGAAA-
GACGCGCAG-30).
2.2. Chemical synthesis
For synthetic procedure and characterization data of com-
pounds 2–17, see Supporting information.
2.3. Production of His-REV7(R124A)/biotin-AviTag-REV3L(1846–
1898) pre-complex
The pETDuet1-REV7(R124A)/REV3L(1847–1898) plasmid25 was
digested with restriction enzymes NdeI and XhoI and purified to
excise the region encoding REV3L(1847–1898). Concurrently, the
region encoding the REV3L(1847–1898) was amplified by Taq
PCR, using this parent plasmid as the template and 50-ATACATATG
AGTTCTGGACTAAACGACATATTCGAGGCACAGAAGATAGAGTGGCA
CGAGGATATGTTGACACCAACTCCT-30 and 50-AATCTCGAGTTACT
AGTCATGATCCAACAAAGTTGCC-30 as the primers to generate a
cDNA fragment encoding AviTag-REV3L(1846–1898), which con-
tains an extra 3 nucleotides encoding REV3L(1846) inserted as a
linker between the AviTag and REV3L(1847–1898). The cDNA
was digested with NdeI and XhoI, ligated to the cut plasmid
prepared above by using the Quick Ligation Kit (New England
The plasmid pMAL-His8-REV7(R124A)-FLAG was transformed
into BL21(DE3) (Novagen). Cells were grown in LB media at 37 °C
to a cell density of OD0.6 at 600 nm and induced with 1 mM IPTG.
Cells were further grown for another 3 h at 37 °C. Cells were har-
vested by centrifugation, suspended in lysis buffer (20 mM Tris–
HCl, pH 8.0, 500 mM NaCl, and 10% glycerol), and disrupted by
using a microfluidizer. After clearing the lysate by centrifugation,
amylose resin (New England Biolabs) was added and the mixture
was stirred for 1 h at 4 °C. The mixture was added into an empty
chromatography column, and the resin was allowed to settle by
gravity. After washing with the lysis buffer, the bound protein
was eluted with 10 mM maltose (Sigma–Aldrich) in lysis buffer.
Biolabs), and cloned by transforming the E scherichia coli DH5
a
2.5. Hit identification by AlphaScreen HTS
strain to generate the pETDuet1-REV7(R124A)/AviTag-REV3L
(1846–1898) plasmid. The perfect match of sequence was con-
firmed by Sanger sequencing, using the Duet UP2 primer
(TTGTACACGGCCGCATAATC).
HTS was conducted by using the AlphaScreen assay to analyze
the bioactive (5760 compounds) and kinase inhibitor-like scaffold
(2650) compounds chemical library in a solution consisting of
The plasmids described above and pBirA+ were co-transformed
100 nM
His-REV7(R124A)/REV3L(1846–1898)
protein
and
into BL21(DE3) (Novagen, Madison, WI) with ampicillin (100
mL) and chloramphenicol (34 g/mL) selection. Cells were grown
in LB media at 37 °C to a density of 0.6 at 600 nm, induced with
1 mM IPTG, and 50 M biotin was added. Cells were grown for
lg/
10 nM C-terminal biotin-tagged REV3L(1875–1895) peptide
(ANILKPLMSPPSREEIMATLL) in the AlphaScreen buffer (25 mM
HEPES, 100 mM NaCl, pH 7.4, 0.1% BSA, and 0.05% Tween-20).
l
l
Briefly, 5 lL of the assay solution was transferred into each well
another 6 h at 25 °C. Then, cells were harvested by centrifugation
and suspended in lysis buffer (50 mM HEPES-NaOH, pH 7.8,
100 mM NaCl, and 20 mM imidazole) and disrupted using a
microfluidizer. The clarified supernatant was applied to a Nickel
column (GE Healthcare Life Sciences, Marlborough, MA) and
of a white OptiPlate-384 (PerkinElmer) by using a Matrix WellMate
liquid dispenser (Thermo Scientific, Waltham, MA). Then, 20 nL of
the test compounds in DMSO were transferred in duplicate with a
pin tool (V&P Scientific, Inc., San Diego, CA), by using a Biomek
(Beckman Coulter, Indianapolis, IN), into the solution to give a final
washed with
a buffer containing 50 mM Tris–HCl, pH 8.0,
drug concentration of 40 lM in each well. The negative control in
500 mM NaCl, and 40 mM imidazole. The bound protein was
eluted by a linear gradient of imidazole in the same buffer. Frac-
tions containing the protein of interest were pooled and dialyzed
in a buffer containing 25 mM HEPES-NaOH, pH 7.5, 100 mM NaCl,
each plate was DMSO, and the positive control was unlabeled
REV3L(1875–1895) peptide as a self-competitor. After incubating
the plates overnight at room temperature, AlphaScreen nickel che-
late (Ni-NTA) acceptor beads (PerkinElmer) diluted 1:100 in 10 lL
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
3
of the AlphaScreen buffer were added by using the Wellmate liquid
dispenser. After 1 h, streptavidin AlphaScreen donor beads (Perki-
nElmer) diluted 1:100 in 10 lL of the AlphaScreen buffer were
was measured by using an EnVision plate reader (PerkinElmer)
with 555-nm excitation (38-nm bandpass) and 632-nm emission
(45-nm bandpass) filters and a D595fp/D635 dichroic mirror. FP
values were calculated in millipolarization (mP) units. Absence of
the autofluorescence background was concurrently confirmed for
each compound in the same assay plate.
added. One hour later, the AlphaScreen signal was read by using
an EnVision plate reader (PerkinElmer). Data were processed by
using in-house programs in Pipeline Pilot (Accelrys, v.7.0.1). Hit
compounds showing more than a 50% decrease in the AlphaScreen
signal were further validated by establishing a dose response in
triplicate, using another batch of the compound. Nonspecific inhi-
bitors of the AlphaScreen signal generation were identified and
eliminated by assaying the hit compounds on AlphaScreen for bio-
tin-tagged hexahistidine. Compounds that nonspecifically dena-
tured the His-REV7(R124A)/REV3L(1846–1898) protein were
identified and eliminated by the thermal shift assay. After this
selection process, only compound 1 was left for further evaluation.
2.9. WaterLOGSY NMR
All NMR experiments were performed at 298 K (25 °C), using a
Bruker Avance 600-MHz spectrometer equipped with a TCI cryo-
genic gradient probe. Compound 7 was dissolved in DMSO-d6 at
a concentration of 10 mM. For NMR experiments, 10 lL of each
compound was dissolved in a solution containing 20mM sodium
phosphate, pH 7.4, 100 mM NaCl, 10% D2O, or buffer containing
the MBP-His8-REV7(R124A)-FLAG protein (10
lM), or MBP protein
2.6. Thermal shift assay
to yield a final concentration of 200 lM for each compound. One-
dimensional (1D) 1H-WaterLOGSY NMR spectra (mixing time of
1.2 s)26 were recorded in the presence and absence of the protein
described above.
For the thermal shift assay, a mixture of a 5 lM solution of His-
REV7(R124A)/REV3L(1846–1898) protein in buffer (25 mM HEPES,
100 mM NaCl, pH 7.3) and a 1:2000 dilution of SYPRO orange dye
(Sigma) was prepared. To a 96-well plate, 47.5
solution was added with either 2.5 L DMSO or 2.5
1 (from a 10 mM DMSO stock) to yield a final concentration of
500 M. Then, 20 L of the mixture was added in duplicate to a
l
L of the protein
2.10. Cell toxicity assay
l
lL compound
HeLa cells (1000 cells/well) were seeded into 96-well plates and
allowed to attach overnight. Next day, the culture medium was
carefully removed and fresh medium containing compound 7 or
DMSO vehicle control was added in triplicate. After 3 days, the ala-
marBlue reagent (Thermo Fisher Scientific) was added as per man-
ufacturer’s instructions. After 3 h, the florescence signal was read
by using an EnVision plate reader (PerkinElmer) with 510-nm exci-
tation and 590-nm emission filters. Cell viability was calculated by
normalizing the signal value with that of cells treated with the
DMSO vehicle control.
l
l
384-well MicroAmp optical reaction plate (Applied Biosystems,
Foster City, CA). The plate was covered with a clear film and cen-
trifuged for 1 min at 1500 rpm. The plate was incubated for 4 h
at room temperature and then analyzed on an ABI 7900HT Fast
Real-Time PCR System (Applied Biosystems). The temperature
was increased from 25 °C to 95 °C at the rate of 1 °C/min, and flu-
orescence was monitored for the SYPRO orange dye (kex 492 nm
and kem 610 nm). The Prism software (GraphPad, La Jolla, CA)
was used to generate a Boltzmann sigmoid fit on a plot of fluores-
cence versus temperature to calculate the melting temperature at
the inflection point of the curve.
2.11. ICL repair assay
COS7 cells (80,000 cells/well) were initially seeded into 24-well
cluster plates and allowed to settle. The medium was removed
2.7. Pull-down assay
after 5 h and replaced with 200
the final concentration of each compound or DMSO vehicle control.
Transfection mixtures (100
L) containing pGL-ICL27 (10 ng) or
pGL4.50 (2.5 ng; Promega, Madison, WI), pRL-TK (1 g; Promega),
L of FuGENE 6 (Promega), and 95 L of OptiMEM (Invitrogen)
l
L of a medium containing 1.5ꢀ
A mixture of REV7(R124A)/AviTag-REV3L(1846–1898) (1
and DMSO (0.5%), compound 1 (50 M), or REV3L(1875–1895)
peptide (5 M) in 50 mM HEPES, pH 7.4, 150 mM NaCl, 0.1% Tween
20 (total 100 L) was incubated by vibrating at 1500 rpm over
lM)
l
l
l
l
l
4
l
l
WellMate (Eppendorf, Hamburg, Germany) at 4 °C for 20 h. Each
were immediately prepared according to the manufacturer’s
treatment was performed in duplicate. Washed streptavidin-con-
instructions and added to each well. Cells were detached by adding
jugated agarose beads (10
Danvers, MA) were added to the mixture and incubated for 4 h.
l
L, settled, Cell Signaling Technologies,
100
resuspended in 200
centration of compound 7 or DMSO vehicle control. All samples
were then re-dispensed (70 L/well) in quadruplicate into a white
l
L of trypsin 5 h after the transfection, and then cells were
lL of medium containing 1.5ꢀ the final con-
Beads were then washed with the same buffer as used above (5ꢀ
with 500
lL of buffer), and bound proteins were eluted by heating
l
at 85 °C for 10 min with the SDS-loading buffer, analyzed by the
NuPAGE gel system (10% Bis-Tris, MES running buffer; Thermo
Fisher Scientific, Waltham, MA), and imaged by using the Oriole
Gel Stain (Bio-Rad, Hercules, CA).
opaque 96-well cluster plates. After the cells were incubated for
another 13 h, the medium was removed, and the signals of both
firefly and Renilla luciferase reporters were assayed by using the
Dual-Luciferase Reporter Assay System (Promega) according to
the manufacturer’s instructions. The ICL repair activity was calcu-
lated as previously described.27
2.8. Fluorescence polarization assay
The assay solution consisted of 200 nM MBP-His8-REV7
(R124A)-FLAG protein and 10 nM C-terminal TAMRA-tagged
REV3L(1875–1895) peptide in a buffer consisting of 35 mM HEPES,
10% glycerol, 0.01% Triton X-100, pH 7.3. Each compound was seri-
ally diluted from a 10 mM stock solution in DMSO in the assay
solution in a 96-well plate. In each plate, the negative control
was DMSO and the positive control was the unlabeled REV3L
(1875–1895) and REV3L(1875–1887) peptide as a self-competitor.
2.12. Cisplatin sensitization by the clonogenic survival assay
HeLa cells (450 cells/well) were seeded into 6-well plates and
allowed to attach overnight. Each indicated final concentration of
cisplatin was added. After 24 h, compound 7 at a final concentra-
tion of 75 lM or DMSO vehicle control was added to the wells,
and plates were incubated for another 24 h. The culture medium
was replaced with fresh medium, and cells were cultured for
another 6 days. Cells were gently rinsed twice with PBS, fixed with
3.7% formaldehyde (w/v), and stained with 0.5% crystal violet
After 4 h, 20
lL of the assay mixture was transferred in triplicate
into each well of a black 384-well plate. The florescence signal
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

4
M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
(w/v). Colonies containing more than approximately 40 cells were
counted. Each treatment was performed in triplicate. The survival
rate was reported as 1 for the average of colony numbers in wells
that received the vehicle control of cisplatin [0.9% sodium chloride
(w/v)].
Two additional assays were performed to confirm that com-
pound 1 is a true inhibitor of the REV7-REV3L(1875–1895) interac-
tion. In a thermal shift assay, Compound 1 showed a clear shift in
the denaturation point of the His-REV7(R124A)/REV3L(1847–
1898) complex (Fig. 1D), suggesting that compound 1 exchanged
with REV3L(1847–1898) in the complex but was not nonspecifi-
cally bound on the complex. To verify this, a His-REV7(R124A)/bio-
tin-AviTag-REV3L(1846–1898) complex was generated. The
complex was incubated with compound 1 or non-tagged REV3L
(1875–1895) and a pull-down assay using streptavidin-conjugated
beads was performed. The isolation of His-REV7(R124A) was sig-
nificantly inhibited (Fig. 1E), suggesting that both compound 1
and REV3L(1875–1895) can exchange with the biotin-AviTag-
REV3L(1846–1898) in the complex, resulting in the release of
His-REV7(R124A) from the beads.
3. Results
3.1. Structure-based design was unsuccessful for designing a
REV7 inhibitor
REV7 binds to short motifs of REV3L,28,29 which are essential for
resistance to cisplatin.30,31 Crystal structures of the complex of
REV7 with the REV3L(1847–1898) fragment have been deter-
mined. The most notable feature of this interaction is that a portion
of the REV3L(1847–1898) containing a PxxxxP motif is inserted
into the bridge-like ‘safety-belt’ structure of REV7.25,32,33 In con-
trast, no structures of uncomplexed REV7 have been solved, likely
due to the intrinsically disordered state of this protein. 15N-HSQC
NMR spectra of the GB1-fused34 REV7 protein revealed peaks in
only the GB1 region, and there were only faint peaks clustered in
the region of the unstructured portion (data not shown), indicating
that the uncomplexed REV7 portion is almost unstructured. This
property of REV7 precludes the use of a structure-based approach.
In fact, none of compounds selected by virtual screening of approx-
imately 450,000 compounds using Glide docking to the REV7
structure of the complex were active in the binding inhibition
assay (next section). Therefore, we used a non-rational screening
approach to identify a hit compound that inhibits the REV7
interaction.
3.3. Structure–activity relationship (SAR) of the hit scaffold
Next, we studied the SAR of the chemical scaffold of compound
1. Because the HTS method that uses a protein complex is atypical
to evaluate the inhibitors of protein–protein interactions, we
wished to use an uncomplexed REV7(R124A) protein to study
SAR, which does not require large-scale protein production. In a
recent study, the MBP-fused REV7 protein was produced and used
for functional studies.13 The MBP conjugation improves protein
stability and increases fluorescence anisotropy because of the
increase in overall protein size (from 24 kDa to 67 kDa). Thus, we
produced the MBP-REV7(R124A) protein and titrated it with a
TAMRA-tagged REV3L(1875–1895) peptide in a fluorescence polar-
ization (FP) assay. The EC50 of the dose–response curve was
approximately 75 nM (Fig. 2A). When a mixture of MBP-REV7
(R124A) and REV3L(1875–1895)-TAMRA was titrated by the non-
tagged REV3L(1875–1895), a dose-responsive competition was
observed, which indicated the self-competition (Fig. 2B). This find-
ing confirms that the FP competition assay can be used to evaluate
inhibitors of the interaction between REV7(R124A) and REV3L
(1875–1895)-TAMRA.
Analogs of compound 1 were synthesized and assayed by the FP
competition assay to study SAR (Fig. 2C and D). None of com-
pounds in which the furan ring was replaced with a different
hydrophobic group (compounds 2–5) were active. Changing the
orientation of the ring to a 3-furan configuration neither yielded
an active compound (compound 6), highlighting the importance
of the 2-furan moiety. Among examined, the only tolerated substi-
tution on the furan ring of compound 1 was a 5-methyl group
3.2. Hit identification by HTS
Despite the small size of REV7, we were unable to produce the
native REV7 protein in a standard bacterial expression protocol.
Although the R124A mutation of REV7 can stabilize the protein
without losing the interaction to REV3L(1847–1898),35 expression
of the REV7(R124A) protein was not still productive enough to
yield large amounts. Thus, to enable running HTS, we had to take
an approach using REV7(R124A)/REV3L(1847–1898) complex that
can be produced in large amounts. A hexahistidine (His)-REV7
(R124A) was expressed and purified concurrently with REV3L
(1847–1898).25 Titration of the His-REV7(R124A)/REV3L(1847–
1898) complex with a biotin-tagged REV3L(1875–1895), but not
with another biotin-tagged unrelated peptide of similar length
[Frizzled-7(544–566)36], generated a strong signal in the amplified
luminescent proximity homogenous assay screen (AlphaScreen),
which detects the proximity of a His-tag to biotin (Fig. 1A). The fact
that the structure of REV7(R124A)/REV3L(1847–1898) complex25
does not harbor an additional site to adopt another REV3L peptide
suggested that part of the REV3L(1847–1898) fragment could be
exchanged with the exogenous REV3L(1875–1895) peptide in this
(compound 7), which yielded higher activity (IC50 = 78 lM) than
compound 1 in the FP assay. A series of amide substitutions on
the furan ring yielded compounds 8–10 that were inactive,
whereas the morpholino analog (compound 11) showed some
activity. Therefore, a 5-methyl substitution was made on the furan
ring to explore the acetylpiperidinemethyl moiety of compound 7.
Removing the entire moiety (compound 12) eliminated activity.
Comparable activity was observed on substituting the piperidine
ring by an azetidine (compound 13) or a pyrrolidine ring (com-
pound 14) or shifting the position of the acetyl group (compound
15). However, replacing the acetylpiperidine moiety with a phenyl
(compound 16) or pyridine (compound 17) ring yielded inactive
compounds, which supports the notion that the N-acyl-heterocycle
moiety plays a role in activity.
assay condition. When
a mixture of His-REV7(R124A)/REV3L
(1847–1898) complex and REV3L(1875–1895)-biotin was titrated
with the non-tagged REV3L(1875–1895), the signal in the AlphaSc-
reen assay was reduced in a dose-dependent manner by non-
tagged REV3L(1875–1895) but not by non-tagged Frizzled-7
(544–566) (Fig. 1B). These findings indicate that the interaction
of biotin-tagged REV3L(1875–1895) to the His-REV7(R124A) was
inhibited specifically by the self-competition. Therefore, this assay
should identify a compound that inhibits the interaction of His-
REV7 with the REV3L(1875–1895)-biotin. We performed HTS of
approximately 8400 compounds by using this protocol. After
removing false-positive hits, only compound 1 (Fig. 1C) was left.
HTS of approximately 6500 more compounds did not yield
additional true hits.
3.4. Direct REV7 binding of the lead compound
Compound 7 was chosen for further evaluation because of its
potency and a symmetric structure that makes the NMR spectra
simple, and thereby feasible for the water-ligand observed via
gradient spectroscopy (WaterLOGSY)26 study. To verify whether
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
5
Figure 1. HTS identification of a REV7 inhibitor by AlphaScreen assays. (A) Dose response of indicated biotin-tagged peptide titrated over His-REV7(R124A)/REV3L(1847–
1898) in AlphaScreen. A hooking peak characteristic to AlphaScreen assay is observed only for REV3L(1875–1895)-biotin peptide. (B) Dose response of competition by the
indicated peptide titrated over a mixture of His-REV7(R124A)/REV3L(1847–1898) and REV3L(1875–1895)-biotin peptide. IC50 for REV3L(1875–1895) is read as 80 nM. An
unrelated peptide of similar length [Fz7(544–566)] served as a negative control. Error bars represent standard deviation (n = 2). Hit identification and validation by
independent assays. (C) Structure of the hit compound 1. (D) Destabilization, but not nonspecific denaturation, of His-REV7(R124A)/REV3L(1847–1898) protein by compound
1 measured by thermal shift assay. (E) Dissociation of His-REV7(R124A)/biotin-AviTag-REV3L(1847–1898) protein complex (1
lM) by compound 1 (50 lM) or REV3L(1875–
1895) peptide (5 M), on streptavidin-conjugated agarose beads. DMSO was used as a vehicle control for compound 1. Each treatment was duplicated, as shown by the bars.
l
compound 7 directly binds to REV7(R124A), WaterLOGSY NMR
spectra of compound 7 were recorded in the absence and presence
of MBP-REV7(R124A) or MBP. Except for the negative resonance
peaks of buffer salts, almost no other resonances appeared in the
absence of MBP-REV7(R124A). However, positive peaks that
matched those from the 1H NMR of compound 7 appeared in the
presence of MBP-REV7(R124A) (Fig. 3) but not MBP (data not
shown), indicating that compound 7 directly binds to REV7
(R124A) but not MBP. The relative size of several peaks in Water-
LOGSY spectra were smaller than those in the 1H NMR (Fig. 3,
arrowheads), which suggests that those hydrogens are exposed
out from the interface of the REV7(R124A)/compound 7 complex.
However, the attempt to determine structural detail of interaction
of compound 7 to REV7 was not successful, due to failure of crys-
tallization of REV7/compound 7 complex.
COS7, HEK293T), resulting in removal of the ICL during replication
and the restoration of luciferase activity. COS7 cells were trans-
fected with this plasmid. Compound 7 suppressed luciferase activ-
ity in cells transfected with the ICL plasmid, after normalizing with
those in cells transfected with the corresponding non-ICL plasmid,
indicating that compound 7 inhibits ICLR (Fig. 4C). Finally, a cell-
survival assay was performed to validate the chemotherapeutic
potential of compound 7. HeLa cells were first treated with cis-
platin and then compound 7 or vehicle control, and the number
of cell colonies was counted. The normalized number of colonies
was lesser for cells treated with cisplatin and compound 7 than
for cells treated with cisplatin only (Fig. 4D), which suggests a
chemosensitization effect.
4. Discussion
3.5. Functional activity of the lead compound
We did not observe a sharp SAR of the compound 1 scaffold for
REV7 inhibition, which suggests that the scaffold might not be
enough capable for antagonizing entire REV3L(1875–1895). A pre-
vious study identified that the PxxxxP motif of REV3L(1880–1885)
is the main region that interacts with REV7.28 In this interaction, a
helix structure of REV7 binds with the PxxxxP motif, and the bridge
structure of REV7 stabilizes the PxxxxP motif interaction.25 How-
ever, the PxxxxP motif might not be sufficient for a tight REV7
interaction. A small helical motif in the REV3L(1887–1893) frag-
ment is not the main region binding to REV7, but might increase
At a concentration approximately at IC50 in the FP assay, com-
pound 7 decreased HeLa cell growth but did not completely elim-
inate the cells (Fig. 4A). Thus, this concentration was used to
validate compound 7 for inhibiting the DDR in cells. A luciferase
plasmid containing an ICL between the promoter and luciferase-
coding regions was used (Fig. 4B) to test ICLR activity.27 This plas-
mid also contains the SV40 origin sequence and thus is replicated
in mammalian cells expressing the SV40 large-T antigen (e.g.,
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

6
M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
Figure 2. SAR study by fluorescence polarization assays. (A) Dose response of MBP-REV7(R124A) titrated over REV3L(1875–1895)-TAMRA peptide in fluorescence
polarization. EC50 is read as 75 nM. (B) Dose response of self-competition by indicated peptide titrated over a mixture of MBP-REV7(R124A) (200 nM) and REV3L(1875–
1895)-TAMRA peptide (10 nM). IC50 is read as 0.3
(C) Dose response of competition by indicated compound titrated over a mixture of MBP-REV7(R124A) and REV3L(1875–1895)-TAMRA peptide. The 17 compounds, REV3L
(1875–1895) (⁄), REV3L(1875–1886) (#), and DMSO (^) were assayed all together for direct side-by-side comparison at 300, 100, 30, 10, and 3
M (from left to right for each
compound). DMSO was used as a vehicle control for the compounds. The mP value of DMSO treatment represents 0% inhibition and that of REV3L(1875–1995) peptide does
100% inhibition. Error bars represent standard deviation (n = 3). IC50 of compound 7 is read as 78 M. (D) Structure of the compounds 1–17.
lM [REV3L(1875–1895): square] and 30 lM [REV3L(1875–1886): triangle]. Error bars represent standard deviation (n = 3).
l
l
REV7 association by promoting a structural transition of REV7 from
a disordered to an ordered state. Indeed, the competition by the
REV3L(1875–1886) peptide (i.e., the PxxxxP motif only) was
Numerous studies have reported the Polf-independent func-
tions of REV7, such as cell cycle regulation by inhibiting APC37,38
and resection of double-strand DNA breaks.19,21 It is reasonable
to assume that an inhibitor of the REV7/REV3L interaction might
be inert to REV3L-independent (i.e., Polf-independent) functions.
However, given the intrinsically disordered nature of uncomplexed
REV7, a compound that induces a global change in the structure of
REV7 may also allosterically inhibit REV7 interactions with other
REV7 partners for Polf-independent functions. Given this possibil-
ity, the suppressive effect of compound 7 on cell growth (Fig. 4A)
might be in part through mechanistic inhibition of multiple func-
tions of REV7. From a chemotherapeutic standpoint, it will be
merely 30 lM (Fig. 2B), which was 100 times weaker than that of
the REV3L(1875–1895) peptide (Fig. 2B), regardless of the fact that
the REV3L(1885–1895) peptide by itself (i.e., the helical motif only)
showed no competition (not shown). It is possible that compound
1 scaffold antagonizes the PxxxxP motif but not the entire REV3L
(1875–1895) sequence, given that the IC50 of compound
7
(78 M, Fig. 2C) is only 2–3 times weaker than that of the REV3L
l
(1875–1886) PxxxxP peptide. Thus, it is likely that the structure
transition of REV7 to adopt compound 7 is not successful.
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
7
intriguing to compare inhibition of Polf-independent REV7 func-
tions and that of Polf-specific REV7 function for ICLR and the
chemotherapeutic sensitization. The functional validation of
REV7 by genetic elimination of REV7 may not address this objec-
tive, because this approach is not capable to distinguish particular
REV7 interactions and functions. An approach in which the specific
interaction site of REV7 is mutated would be the most plausible
approach to validate specific REV7 functions, but interaction sites
for Polf-independent REV7 functions have not been identified to
date. Hence, another approach is required to determine whether
Polf-independent REV7 functions are druggable target.
Author contributions
MLA performed all chemical syntheses, FP assays, thermal sta-
bilization assays, and additional HTS for ꢁ6500 compounds; and
analyzed the data. NDA performed virtual screenings, initial HTS
for ꢁ8400 compounds to identify compound 1, and thermal stabi-
lization assays. BJE generated the expression plasmid for MBP-
REV7(R124A), optimized the FP assay, and performed the ICLR
assay. YH and RH generated the His-REV7(R124A)/biotin-AviTag-
REV3L(1847–1898) and MBP-REV7(R124A) proteins. MV per-
formed the WaterLOGSY analysis. AI performed cell growth and
clonogenic assays. ETM attempted HSQC NMR analysis of the
REV7/compound 7 complex. SK, KH, and HH generated the plasmid
for expressing His-REV7(R124A)/REV3L(1847–1898), produced
this protein, validated the interaction of REV3L(1875–1895) to
uncomplexed REV7, and attempted to crystallize the REV7/
compound 7 complex. NF generated the plasmid for express-
ing His-REV7(R124A)/AviTag-REV3L(1846–1898), developed the
AlphaScreen and FP assays, performed the pull-down assay, con-
ducted the project, analyzed the data, and wrote the manuscript.
All authors have approved the final version of the manuscript.
Figure 3. Direct binding of the compound to REV7(R124A). Spectra 1H NMR, and
WaterLOGSY NMR with or without MBP-REV7(R124A), for compound 7. See text for
detail. Arrowheads: peaks smaller in the WaterLOGSY NMR with MBP-REV7(R124A)
than in the 1H NMR, and their assignments (left below).
Funding
The study was funded by the American Cancer Society Research
Scholar Grant #RSG CDD-120969 (to NF) and by the American
Lebanese and Syrian Associated Charities (ALSAC).
Acknowledgements
We thank Jaeki Min for HTS robotics, the Compound Manage-
ment office for providing the compounds, the Macromolecular Syn-
thesis Lab for peptide synthesis, the Sanger DNA Sequencing Core
for plasmid sequencing, the Proteomics & Mass Spectrometry Facil-
ity for protein characterization, Cynthia Jeffries for compound
characterization, Weixing Zhang for supervising the NMR, Aman
Singh for assisting with the thermal shift assay, M. Brett Waddell
for performing the molecular interaction analysis, Sourav Das for
cheminformatics support, Christine E. Canman for providing the
REV7 plasmid and for technical advice, Hans Haecker for providing
the pBirA plasmid, and Vani Shanker for editing the manuscript.
Supplementary data
Figure 4. Functional activity of the compound. (A) Viability of HeLa cells after
compound 7 treatment (75 lM) measured by alamarBlue reagent, normalized by
DMSO treatment. (B) Reporter plasmid used in the ICLR assay. (C) ICLR activity
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2016.07.026.
measured in COS7 cells transfected with the plasmid (B) and treated with
References and notes
compound 7 (75
unpaired t-test. (D) Clonogenic survival of HeLa cells that were treated with
indicated concentration of cisplatin followed by compound 7 (75 M) or DMSO.
lM) or DMSO. Statistical analysis was performed by two-tailed
1. Deans, A. J.; West, S. C. Nat. Rev. Cancer 2011, 11, 467.
2. Clauson, C.; Scharer, O. D.; Niedernhofer, L. Cold Spring Harb. Perspect. Biol.ogy
2013, 5, a012732.
3. Shen, X.; Jun, S.; O’Neal, L. E.; Sonoda, E.; Bemark, M.; Sale, J. E.; Li, L. J. Biol.
Chem. 2006, 281, 13869.
l
Data is indicated as survival fractions normalized by that of no cisplatin group for
each treatment. DMSO was used as a vehicle control, same volume of the compound
solution. Error bars represent standard deviation (n = 3 (B, D), 4 (C)).
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026

8
M. L. Actis et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
4. Hicks, J. K.; Chute, C. L.; Paulsen, M. T.; Ragland, R. L.; Howlett, N. G.; Gueranger,
22. Zhang, L.; Yang, S. H.; Sharrocks, A. D. Mol. Cell. Biol. 2007, 27, 2861.
23. Cheung, H. W.; Chun, A. C.; Wang, Q.; Deng, W.; Hu, L.; Guan, X. Y.; Nicholls, J.
M.; Ling, M. T.; Chuan Wong, Y.; Tsao, S. W.; Jin, D. Y.; Wang, X. Cancer Res.
2006, 66, 4357.
24. Niimi, K.; Murakumo, Y.; Watanabe, N.; Kato, T.; Mii, S.; Enomoto, A.; Asai, M.;
Asai, N.; Yamamoto, E.; Kajiyama, H.; Shibata, K.; Kikkawa, F.; Takahashi, M.
Cancer Sci. 2014, 105, 545.
25. Hara, K.; Hashimoto, H.; Murakumo, Y.; Kobayashi, S.; Kogame, T.; Unzai, S.;
Akashi, S.; Takeda, S.; Shimizu, T.; Sato, M. J. Biol. Chem. 2010, 285, 12299.
26. Dalvit, C.; Fogliatto, G.; Stewart, A.; Veronesi, M.; Stockman, B. J. Biomol. NMR
2001, 21, 349.
27. Inoue, A.; Kikuchi, S.; Hishiki, A.; Shao, Y.; Heath, R.; Evison, B. J.; Actis, M.;
Canman, C. E.; Hashimoto, H.; Fujii, N. J. Biol. Chem. 2014, 289, 7109.
28. Hanafusa, T.; Habu, T.; Tomida, J.; Ohashi, E.; Murakumo, Y.; Ohmori, H. Genes
Cells 2010, 15, 281.
Q.; Glover, T. W.; Canman, C. E. Mol. Cell. Biol. 2010, 30, 1217.
5. Sharma, S.; Canman, C. E. Environ. Mol. Mutagen. 2012, 53, 725.
6. Ho, T. V.; Scharer, O. D. Environ. Mol. Mutagen. 2010, 51, 552.
7. Murakumo, Y. Mutat. Res. 2002, 510, 37.
8. Lawrence, C. W. Adv. Protein Chem. 2004, 69, 167.
9. Gan, G. N.; Wittschieben, J. P.; Wittschieben, B. O.; Wood, R. D. Cell Res. 2008,
18, 174.
10. Doles, J.; Oliver, T. G.; Cameron, E. R.; Hsu, G.; Jacks, T.; Walker, G. C.; Hemann,
M. T. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20786.
11. Xu, X.; Xie, K.; Zhang, X. Q.; Pridgen, E. M.; Park, G. Y.; Cui, D. S.; Shi, J.; Wu, J.;
Kantoff, P. W.; Lippard, S. J.; Langer, R.; Walker, G. C.; Farokhzad, O. C. Proc. Natl.
Acad. Sci. U.S.A. 2013, 110, 18638.
12. Wu, F.; Lin, X.; Okuda, T.; Howell, S. B. Cancer Res. 2004, 64, 8029.
13. Lee, Y. S.; Gregory, M. T.; Yang, W. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 2954.
14. Hashimoto, K.; Cho, Y.; Yang, I. Y.; Akagi, J.; Ohashi, E.; Tateishi, S.; de Wind, N.;
Hanaoka, F.; Ohmori, H.; Moriya, M. J. Biol. Chem. 2012, 287, 9613.
15. Raschle, M.; Knipscheer, P.; Enoiu, M.; Angelov, T.; Sun, J.; Griffith, J. D.;
Ellenberger, T. E.; Scharer, O. D.; Walter, J. C. Cell 2008, 134, 969.
16. Fattah, F. J.; Hara, K.; Fattah, K. R.; Yang, C.; Wu, N.; Warrington, R.; Chen, D. J.;
Zhou, P.; Boothman, D. A.; Yu, H. PLoS Genet. 2014, 10, e1004419.
17. Sale, J. E. EMBO J. 2015, 34, 1609.
29. Murakumo, Y.; Roth, T.; Ishii, H.; Rasio, D.; Numata, S.; Croce, C. M.; Fishel, R. J.
Biol. Chem. 2000, 275, 4391.
30. Tomida, J.; Takata, K.; Lange, S. S.; Schibler, A. C.; Yousefzadeh, M. J.; Bhetawal,
S.; Dent, S. Y.; Wood, R. D. Nucleic Acids Res. 2015, 43, 1000.
31. Khalaj, M.; Abbasi, A.; Yamanishi, H.; Akiyama, K.; Wakitani, S.; Kikuchi, S.;
Hirose, M.; Yuzuriha, M.; Magari, M.; Degheidy, H. A.; Abe, K.; Ogura, A.;
Hashimoto, H.; Kunieda, T. J. Biol. Chem. 2014, 289, 3811.
32. Kikuchi, S.; Hara, K.; Shimizu, T.; Sato, M.; Hashimoto, H. J. Biol. Chem. 2012,
287, 33847.
18. Bhat, A.; Wu, Z.; Maher, V. M.; McCormick, J. J.; Xiao, W. Cell Cycle 2015, 14,
3929.
19. Boersma, V.; Moatti, N.; Segura-Bayona, S.; Peuscher, M. H.; van der Torre, J.;
Wevers, B. A.; Orthwein, A.; Durocher, D.; Jacobs, J. J. Nature 2015, 521, 537.
20. Pirouz, M.; Rahjouei, A.; Shamsi, F.; Eckermann, K. N.; Salinas-Riester, G.;
Pommerenke, C.; Kessel, M. Cell Cycle 2015, 14, 1596.
21. Xu, G.; Chapman, J. R.; Brandsma, I.; Yuan, J.; Mistrik, M.; Bouwman, P.;
Bartkova, J.; Gogola, E.; Warmerdam, D.; Barazas, M.; Jaspers, J. E.; Watanabe,
K.; Pieterse, M.; Kersbergen, A.; Sol, W.; Celie, P. H.; Schouten, P. C.; van den
Broek, B.; Salman, A.; Nieuwland, M.; de Rink, I.; de Ronde, J.; Jalink, K.;
Boulton, S. J.; Chen, J.; van Gent, D. C.; Bartek, J.; Jonkers, J.; Borst, P.;
Rottenberg, S. Nature 2015, 521, 541.
33. Wojtaszek, J.; Lee, C. J.; D’Souza, S.; Minesinger, B.; Kim, H.; D’Andrea, A. D.;
Walker, G. C.; Zhou, P. J. Biol. Chem. 2012, 287, 33836.
34. Zhou, P.; Wagner, G. J. Biomol. NMR 2010, 46, 23.
35. Hara, K.; Shimizu, T.; Unzai, S.; Akashi, S.; Sato, M.; Hashimoto, H. Acta
Crystallogr., Sect. F: Struct. Biol. Cryst. Commun. 2009, 65, 1302.
36. Punchihewa, C.; Ferreira, A. M.; Cassell, R.; Rodrigues, P.; Fujii, N. Protein Sci.
2009, 18, 994.
37. Chen, J.; Fang, G. Genes Dev. 2001, 15, 1765.
38. Pfleger, C. M.; Salic, A.; Lee, E.; Kirschner, M. W. Genes Dev. 2001, 15, 1759.
Please cite this article in press as: Actis, M. L.; et al. Bioorg. Med. Chem. (2016), http://dx.doi.org/10.1016/j.bmc.2016.07.026