Small-molecules that bind to the ubiquitin-binding motif of REV1 inhibit REV1 interaction with K164-monoubiquitinated PCNA and suppress DNA damage tolerance

Article abstract of DOI:10.1016/j.bmc.2018.03.028

REV1 protein is a mutagenic DNA damage tolerance (DDT) mediator and encodes two ubiquitin-binding motifs (i.e., UBM1 and UBM2) that are essential for the DDT function. REV1 interacts with K164-monoubiquitinated PCNA (UbPCNA) in cells upon DNA-damaging stress. By using AlphaScreen assays to detect inhibition of REV1 and UbPCNA protein interactions along with an NMR-based strategy, we identified small-molecule compounds that inhibit the REV1/UbPCNA interaction and that directly bind to REV1 UBM2. In cells, one of the compound prevented recruitment of REV1 to PCNA foci on chromatin upon cisplatin treatment, delayed removal of UV-induced cyclopyrimidine dimers from nuclei, prevented UV-induced mutation of HPRT gene, and diminished clonogenic survival of cells that were challenged by cyclophosphamide or cisplatin. This study demonstrates the potential utility of a small-molecule REV1 UBM2 inhibitor for preventing DDT.

Full text of DOI:10.1016/j.bmc.2018.03.028

Accepted Manuscript
Small-molecules that bind to the ubiquitin-binding motif of REV1 inhibit REV1
interaction with K164-monoubiquitinated PCNA and suppress DNA damage
tolerance
Murugendra Vanarotti, Benjamin J. Evison, Marcelo L. Actis, Akira Inoue,
Ezelle T. McDonald, Youming Shao, Richard J. Heath, Naoaki Fujii
PII:
S0968-0896(18)30056-7
https://doi.org/10.1016/j.bmc.2018.03.028
BMC 14266
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
11 January 2018
9 March 2018
17 March 2018
Please cite this article as: Vanarotti, M., Evison, B.J., Actis, M.L., Inoue, A., McDonald, E.T., Shao, Y., Heath, R.J.,
Fujii, N., Small-molecules that bind to the ubiquitin-binding motif of REV1 inhibit REV1 interaction with K164-
monoubiquitinated PCNA and suppress DNA damage tolerance, Bioorganic & Medicinal Chemistry (2018), doi:
https://doi.org/10.1016/j.bmc.2018.03.028
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Small-molecules that bind to the ubiquitin-binding motif of REV1 inhibit REV1 interaction with K164-
monoubiquitinated PCNA and suppress DNA damage tolerance
Murugendra Vanarotti1, Benjamin J. Evison1, Marcelo L. Actis1, Akira Inoue1, Ezelle T. McDonald1, Youming
Shao2, Richard J. Heath2, and Naoaki Fujii*1
1Department of Chemical Biology and Therapeutics and 2Protein Production Facility, St. Jude Children’s
Research Hospital, Memphis, TN, USA; *corresponding author
Keywords: REV1, ubiquitin-binding motif, PCNA, small-molecule, protein-protein interaction
ABSTRACT
REV1 protein is a mutagenic DNA damage tolerance (DDT) mediator and encodes two ubiquitin-binding motifs
(i.e., UBM1 and UBM2) that are essential for the DDT function. REV1 interacts with K164-monoubiquitinated
PCNA (UbPCNA) in cells upon DNA-damaging stress. By using AlphaScreen assays to detect inhibition of REV1
and UbPCNA protein interactions along with an NMR-based strategy, we identified small-molecule compounds
that inhibit the REV1/UbPCNA interaction and that directly bind to REV1 UBM2. In cells, one of the compound
prevented recruitment of REV1 to PCNA foci on chromatin upon cisplatin treatment, delayed removal of UV-
induced cyclopyrimidine dimers from nuclei, prevented UV-induced mutation of HPRT gene, and diminished
clonogenic survival of cells that were challenged by cyclophosphamide or cisplatin. This study demonstrates
the potential utility of a small-molecule REV1 UBM2 inhibitor for preventing DDT.
INTRODUCTION
Chemotherapy remains indispensable for frontline cancer therapy, but its efficacy is often partially limited by
1-2
rapid and erroneous repair of DNA damage promoted by DNA replication irrespective of the damage
leading to drug resistance and mutations. REV1 protein is indispensable for replicating over a variety of
chemotherapeutic DNA damage types, including DNA crosslinks 3-6. The mouse Rev1 is responsible for
acquisition of resistance after repeated engraftment and cyclophosphamide treatment of lymphoma;
,
depletion of Rev1 gene in the lymphoma dramatically improved the drug efficacy, suppressed the resistance,
and inhibited the drug-induced mutations 7. Carcinogen-induced adenomas developed in Rev1 transgenic mice
approximately twice as often as in normal mice, with elevated DNA damage tolerance and mutations 8, further
indicating role of Rev1 in tumor generation. REV1 is also responsible for the resistance and mutations in
ovarian cancer cells 9 and for lung carcinogenesis 10. Importantly, Rev1-knockout mice are healthy except for

an approximate10% reduction in body growth; Rev1−/− B cells undergo normal proliferation and apoptosis 11.
These findings suggest that REV1 inhibition is not likely to be mechanistically toxic unless a chemotherapy drug
is co-administrated.
REV1 encodes the ubiquitin-binding motif (UBM) of REV1. When a DNA replication fork is stalled by DNA
damage, PCNA on the fork is monoubiquitinated on K164 (UbPCNA) and recruits REV1, likely via its UBM 12-13
.
REV1 coordinates Polζ, which extends over a variety of DNA lesions 14, mediates most mutagenic DNA damage
tolerance (DDT) 15-16, and is composed of the regulatory subunit REV7 and the catalytic subunit REV3L. These
observations suggest that the DDT processes may be inhibited by blocking one of the protein-protein
interactions in the UbPCNA-REV1-REV7-REV3L complex (Figure 1). In accordance, we have reported small-
molecules that inhibit UbPCNA-REV1 interaction by binding to PCNA 17 and those that inhibit REV7 interaction
with a partial structure of REV3L by binding to REV7 18. Another group has reported small-molecules that
inhibit REV1 interactions with DNA polymerases 19. Here, we focus on REV1 UBM as another potential target
for inhibiting DDT.
Figure 1. The REV1 machinery for DNA damage tolerance and mutation. PCNA (homotrimer, green) is K164-
monoubiquitinated (yellow) (UbPCNA) when a large DNA lesion such as a DNA crosslink is detected on the
replication fork. REV1 (orange) has UBMs and binds to the UbPCNA. REV1 also binds to REV7 (wine red),
recruiting REV3L (blue), which engages DNA replications that induce mutations (red arrowhead).
RESULTS
Small-molecule compounds inhibit REV1/UbPCNA interaction and directly bind to REV1 UBM
Our plan to identify inhibitor compounds was to screen a chemical library for UbPCNA-REV1 interaction
inhibition and for direct REV1 protein binding. Previously, we used an immunoprecipitation assay using crude
nuclear lysate that was harvested from cells challenged with DNA damage (i.e., ex vivo protocol using
endogenous UbPCNA) to identify a UbPCNA inhibitor 17, but this method is too tedious and not scalable for

chemical library screening purposes. Therefore, we intended to use a scalable in vitro protocol. In our hands,
full-length REV1 protein was very difficult to produce in large enough amounts for in vitro screening. Because
studies have shown that BRCT domain 12,20 and UBM 6 may be responsible for REV1 interaction with UbPCNA,
we alternatively generated REV1Δ, a truncated REV1 protein containing these domains (Figure 2A). Linear
chimeras of PCNA-Ub fusion have been used as surrogates of UbPCNA for in vitro studies 6,21. But in cells,
PCNA is specifically monoubiquitinated at K164 upon DNA damage 22. Therefore, to find compounds that are
potentially active in cells, an assay should work with bona fide UbPCNA but not with the chimera. We have
acquired an enzymatic protocol to produce UbPCNA in vitro 23-24. This protocol ligated either non-tagged or N-
terminus–tagged (hexahistidine, biotin-AviTag) ubiquitin to PCNA (Figure 2B).
Figure 2. Proteins used to screen compounds for REV1/UbPCNA interaction inhibition. A: REV1 (1251 AA) and
its truncated variant (REV1Δ, 695 AA), which was used to screen chemical compounds for inhibiting
REV1/UbPCNA interaction. Dotted lines indicate the truncation to generate REV1Δ. B: In vitro production of
K164-UbPCNA tagged with a biotin on the ubiquitin N-terminus. The reaction mixtures were analyzed by
Coomassie-staining (left) or immunoblotting for pan-PCNA (right, PC-10 antibody) to confirm the UbPCNA
bands.
To develop a scalable method to screen chemical compounds using these proteins, we used AlphaScreen that
has been successfully applied to assay fairly weak protein-protein interactions 25. This property is desirable for
assaying ubiquitin interactions because affinities of ubiquitin-binding domains are weak or modest 26. In
addition, high sensitivity of the proximity signal generation by AlphaScreen is desirable for assaying
compounds on proteins whose supply is limited such as UbPCNA by using relatively low protein concentrations.
AlphaScreen cannot be used to determine the EC50 (surrogate of KD) of the REV1Δ/UbPCNA interaction

because it is a sandwich assay that enforces the hooking effect in the direct binding curve. Despite, significant
signals were generated by treating mixture of His-tagged REV1Δ and UbPCNA protein in which a biotin-AviTag
is incorporated at N-terminus of the ubiquitin moiety by adding AlphaScreen beads (Figure 3A). We have
screened ~50 chemical compounds that we have previously produced for another project27 to select ones that
inhibit the signal generation in dose-dependent manner. Two structurally similar compounds, 1 and 2,
appeared to be inhibitory for the REV1/UbPCNA interaction in this assay (Figure 3B). These compounds did not
inhibit the signal generation from biotin-tagged polyhistidine peptide, indicating that they are not assay
artifacts.
Figure 3. Identification of small-molecule compounds that inhibit interaction of UbPCNA with REV1. A: The
AlphaScreen assay for interaction of His-tagged REV1Δ and biotin-tagged UbPCNA that identified compounds
that inhibit UbPCNA-REV1 interaction. B: Dose-response of the compounds for inhibiting the REV1Δ/UbPCNA
interaction. The IC50 values of 1 and 2 were fitted as 3.4 μM and 9.7 μM, respectively.
We next used a protein NMR approach to show whether the compound binds to REV1Δ. Because REV1Δ and
UbPCNA are still difficult to produce in large enough amounts for NMR experiments, we produced a small
protein encoding only REV1 UBM1-UBM2 because they are potential site if the compounds inhibit the
REV1/UbPCNA interaction by binding to REV1. In the saturation transfer difference (STD) NMR experiment 28
of the compounds, they showed positive NOE peaks with the UBM1-UBM2 protein but no or negative NOE
peaks without the protein (Figure 4A), indicative of direct binding to UBM1-UBM2. We next characterized the
structural outcome of the REV1 interaction of compounds by 2D [1H, 15N]-HSQC NMR. Because we have found
that HSQC peaks of UBM2 but not UBM1 of REV1 were perturbed by adding ubiquitin (manuscript under
review), we anticipated that REV1/UbPCNA interaction is mediated by UBM2 and thus the compound should
bind to UBM2. Compound 1 is not soluble enough to acquire clear HSQC spectra but 2 is soluble. In the 2D [1H,
15N]-HSQC spectra, peaks of several residues of REV1 UBM2 protein were shifted by adding 2 in dose-

dependent manner (Figure 4B-C). By mapping the chemical shift perturbation (CSP) values (Figure 4D) on the
NMR structure of REV1 UBM2, we found that all residues perturbed by 2 are on the α-helix structures (Figure
4E), which is similar to what we observed in the spectra of REV1 UBM2 with ubiquitin. This finding may explain
the potential structural mechanism of Compound 2’s inhibition of UbPCNA- REV1Δ interaction (Figure 3).

Figure 4. Compounds 1 and 2 bind to REV1 UBM2. A: One-dimensional proton spectra of 1 (upper) and 2
(lower) with REV1 UBM2 protein. Red trace indicates the 1D reference spectrum of each compound. Blue
trace indicates the STD spectrum of each compound. B: Superposition of the 2D [1H, 15N]-HSQC spectra of
REV1 UBM2 with 2 at different molar ratios: 1:0 (black), 1:2 (green), 1:4 (cyan), 1:6 (magenta), 1:8 (blue), and
1:10 (red). Arrows show the direction of perturbation. C: Selected peaks from the HSQC spectra are enlarged.
D: Chemical shift perturbations (CSPs) are plotted as a function of the residue number for REV1 UBM2 with 2
at molar ratio 1:10. The secondary structure elements in the REV1 UBM2 are indicated at the top: α1-helices 1
and α2-helices 2. The CSP, with the average value of 0.006 displayed as a red dotted line. E: Assignment of CSP
on the NMR structure of free REV1 UBM2 (6AXD.pdb) in a cartoon-embedded surface model. The secondary
structure elements are indicated. Residues are colored on a blue (largest) to white (zero) gradient based on
the CSP value.
We have attempted to accurately measure KD of the Compound 2 binding to REV1 UBM2. However, we could
not observe saturation of the binding in either isothermal titration calorimetry (ITC) or surface plasmon

resonance (SPR) (not shown), because compound 2 precipitated when the REV1 UBM2: 2 ratio exceeds 1:10.
This suggests that the KD is significantly higher than the IC50 value observed in the AlphaScreen (Figure 3B).
Therefore, we relied on NMR data to estimate the dose for the subsequent cellular experiments.
The small-molecule prevented cisplatin-mediated recruitment of REV1 that colocalized with PCNA on
chromatin
To verify whether the inhibitors identified above in vitro can inhibit the full-length REV1/UbPCNA interaction
in cells, we used a chromatin imaging approach that we have used to show that a compound that binds to and
inhibits interactions of UbPCNA prevents chromatin localization of REV1 protein with PCNA foci in cells
challenged with cisplatin (i.e., in situ generation of UbPCNA by DNA damage) 17, indicating inhibition of
REV1/UbPCNA interaction. The cells were treated with 1 in the concentration in which the NMR experiments
(Figure 4A-D) indicated the UBM2 binding. Compound 2 was too toxic in this concentration. EGFP-tagged full-
length REV1 protein was expressed in U2OS cells, and the cells were pulse-treated with cisplatin followed by
Compound 1. The cells were pre-extracted to remove any non-chromatin proteins and the chromatin was
immunostained for PCNA foci to image them with the EGFP-REV1. The recruitment of the EGFP-REV1 to
chromatin was significantly suppressed by 1 (Figure 5).
Figure 5. Compound 1 inhibits chromatin localization of REV1 following cisplatin treatment. A: Chromatin co-
foci of full-length EGFP REV1/PCNA in U2OS cells that were pulse-treated with cisplatin (33 M) followed by
incubation with DMSO or 1 (150 μM). Pre-extracted cell nuclei were immunostained for PCNA. Confocal
images of representative cells are shown. B: Colocalization of EGFP-REV1 and PCNA was determined by
calculating Pearson’s correlation coefficients between red and green signals of all the pixels within a DAPI-
positive area, n=32 (DMSO) and 37 (1). The dots in the graph indicate the correlation coefficient of individual
cells. Data were analyzed by performing Student’s t-test (two-tailed). The bars in the graph indicate averages.

The small-molecule delayed removal of UV-induced DNA crosslinks from nuclei and prevented UV-induced
mutation of the HPRT gene
REV1 supports DNA replication over various DNA lesions including crosslinks to prevent replication arrest and
support the ultimate removal of the lesions 5. Therefore, we assayed Compound 1 for inhibition of removal of
DNA crosslinks. The cyclopyrimidine dimer (CPD) is a DNA crosslink that is generated by UVC and that can be
reliably quantified on cell nuclei by using a specific antibody. The cells were UVC-irradiated to induce CPDs on
their DNA and allowed to remove the CPDs while being incubated in Compound 1, which significantly
suppressed removal of the CPD foci (Figure 6A-B). The incomplete inhibition regardless of the presence of 1 is
presumably because the CPDs have been removed by Polη 29 in addition to REV1.
REV1 supports DNA replication over a DNA crosslink by Polζ that is highly mutagenic (Figure 1). Therefore, we
assayed 1 for inhibition of HPRT gene mutagenesis that is mediated by DNA crosslinking. WTK-1 human
lymphoblastoid cells were given UVC with cisplatin followed by DMSO or 1 and cultured to induce mutation
and express HPRT protein. Mutation of the UVC-crosslinked HPRT locus deactivates it; thus, the mutated
clones are resistant against 6-thioguanine. Compound 1 significantly reduced the number of resistance clones
(Figure 6C), indicating suppression of cisplatin-mediated mutagenesis. Few of the cells that received the
highest UVC dose in this experiment (128 J/m2) survived, presumably because the amount of CPD generated
by this UVC dose was too excessive to allow recovery under the presence of 1 and induced cell death. Given
that CPD removal was suppressed incompletely by 1 (Figure 6A-B), we anticipated that the mutagenesis
suppression by 1 would be more significant for DNA crosslinks other than CPD. However, the HPRT assays
using cisplatin instead of UVC were unsuccessful because of enhanced cisplatin toxicity by the Compound 1
treatment (next section).

Figure 6. A: Compound 1 suppresses CPD removal. HeLa cells were irradiated by UVC (2.5 J/m2), and treated
as indicated; chromatin was immunostained with a CPD antibody. (a) untreated cells; (b) cells <1 h after UVC
irradiation; (c) cells 24 h after UVC irradiation and recovery under DMSO; and (d) ibid under 1 (150 µM). B: Dot
blots quantify the CPD foci normalized with the DAPI staining for each panel in (A) as indicated. Data were
analyzed by performing Student’s t-test (two-tailed). The bars in the graph indicate averages. C: Compound 1
(150 µM) suppresses HPRT gene mutations induced by UV. WTK-1 cells were UVC-challenged, treated as
indicated, and cultured to allow HPRT protein expression. Mutation frequency was determined by culturing
480,000 cells with 6-TG for 10 days and calculated from the number of surviving colonies in each dish
normalized to plating efficiency before adding 6-TG. Error bars represent standard deviation (n=2, 3, or 4).
The small-molecule diminished clonogenic survival of cells that were challenged by chemotherapy drugs
We performed clonogenic cell survival assays to validate whether a small-molecule REV1 UBM2 inhibitor
enhances the efficacy of chemotherapy drugs. U2OS cells received pulse treatment of 4-
hydroxycyclophosphamide (active metabolite of cyclophosphamide) or cisplatin in doses not lethal to the cells,
followed by Compound 1 or DMSO. Clonogenic survival of the cells that received only 1 was approximately 2-
times lower than that of control cells (Figure 7), but survival of cells that received 1 in addition to non-lethal
dose of either drug was approximately 10-times lower than that of control cells, indicating that 1 enhanced
the drugs’ efficacy.

Figure 7. Compound 1 enhances the efficacy of chemotherapy drugs. Clonogenic survival of U2OS cells after 4-
hydroxycyclophosphamide (left) or cisplatin (right) treatment with or without 1 (75 μM). Error bars represent
standard deviation (n=3).
DISCUSSION
Previously, we have shown that a small-molecule compound that binds to PCNA adjacent to K164 (to be
monoubiquitinated upon DTT) inhibits UbPCNA protein interactions and suppresses REV1 recruitment to
chromatin 17. Here, we identified another compound that binds to REV1 UBM2 and inhibits the REV1/UbPCNA
interaction. Because we have found that REV1 UBM2 can interact with ubiquitin (manuscript under review), it
is likely that REV1/UbPCNA interaction is mediated by the UBM2. Altogether, the REV1/UbPCNA interaction
can be inhibited by targeting either UbPCNA or REV1 UBM2. A potential limitation for targeting REV1 UBM2 is
that intrinsic affinity of the UBM2 to UbPCNA is likely to be modest given that affinities of ubiquitin binding
domains are weak or modest 26. At a glance, inhibiting a protein-protein interaction with a modest affinity
sounds easy, but it may not be because the modest affinity is due to a native property of the protein; that is,
REV1 UBM2 itself may modestly interact with either ubiquitin or small-molecules intrinsically, which may
partially explain the modest UBM2-inhibition potencies of Compounds 1 and 2. To generate potent analogs of
these compounds, a fragment-based strategy could be applied by using protein NMR to identify fragment
compounds that bind to REV1 at a site other than UBM2 and then conjugating the fragment compounds to 1
or 2.
A concern for targeting REV1 UBM is non-selectivity among other UBMs such as those of DNA Polι 30, although
there are few studies indicating potential toxicities by inhibiting UBMs other than those of REV1. However, we
have observed several structural differences between REV1 UBM2 (6AXD.pdb) and Polι UBM (2KHU.pdb),
indicating that a structure-based strategy is possible to rationally modify Compounds 1/2 or their fragment-
conjugates to generate REV1 UBM2–specific small-molecules.

Once potent and selective REV1 UBM2 inhibitors are identified, such compounds will be useful chemical tools
with which to study specific REV1 functions that UBM2 modulates because such UBM2 inhibitors will not
affect the pathways that other protein interaction domains of REV1 (e.g., BRCT 12) mediate. A study has shown
6
that cells expressing REV1 point mutants in UBM are hypersensitive to UV and cisplatin , genetic evidence
that disrupting REV1 UBM interactions will deactivate REV1 functions for DDT. Given that Rev1 knock-out mice
are viable 11, small-molecules that potently and specifically inhibit REV1 would not be mechanistically toxic.
Such inhibitors would reserve potential as future therapeutic agents to improve existing chemotherapies that
generate interstrand DNA crosslinking (ICL), such as cisplatin and cyclophosphamide (Figure 7), because REV1
is required to remove ICL 4 and REV1 has been validated for resistance against ICL drugs in mice 7. A recent
study demonstrating utility of small-molecules that inhibit DDT by targeting REV1 interactions to DNA
polymerases 19 also supports this therapeutic rationale.
AUTHOR CONTRIBUTIONS
MV produced proteins and performed protein NMR and SPR experiments. BJE performed the CPD removal
assay, REV1 chromatin imaging, and HPRT mutagenesis assay. MLA performed chemical syntheses and
AlphaScreen assays. AI generated protein expression constructs and performed AlphaScreen, HPRT
mutagenesis, and cell survival assays. YS, RJH, and ETM produced proteins. NF designed the research and
wrote the paper. All authors have approved the final version of the manuscript.
ACKNOWLEDGEMENTS
We thank the Sanger DNA Sequencing Core for plasmid sequencing, Weixing Zhang for supervising the NMR,
Jennifer Peters for assistance with microscopic foci quantification, Sivaraja Vaithiyalingam and Brett Waddell
for attempting affinity measurements, Christine Canman for providing plasmids encoding human REV1, Hans
Haecker for providing pBirA plasmid, Titia Sixma for providing pET3a- human Uba1 (E1) plasmid through
Addgene (#63571, Cambridge, MA), Rachel Klevit for providing pET28a-Ubc5Hc S22R (E2) plasmid through
Addgene (#12644), and Cherise Guess for editing the manuscript. The study was funded by ALSAC. The Cell &
Tissue Imaging Center is supported by NCI P30 CA021765-34.
Conflict of interest statement
BJE is currently an employee of Noxopharm Ltd. The authors declare no other conflict of interest.
MATERIALS AND METHODS

Reagents
All chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise stated. All oligonucleotides
for plasmid constructions were synthesized by Integrated DNA Technologies (Coralville, IA). The In-Fusion
cloning kit (Takara Bio USA, Mountain View, CA) was used according to the manufacturers’ recommendations.
The AlphaScreen assay kit was purchased from PerkinElmer (Waltham, MA). Anti-PCNA PC10 mouse IgG was
purchased from Cell Signaling Technology (Danvers, MA). All cells were obtained from American Type Culture
Collection (Manassas, VA) and cultured in Dulbecco's Modified Eagle Medium containing 10% FBS in an
incubator at 37°C in a humidified atmosphere of 5% carbon dioxide.
Chemical synthesis
See Supporting Information.
Production of His-tagged REV1Δ protein
To obtain a plasmid expressing minimal components of human REV1 with a histidine tag (REV1Δ) in bacteria,
two coding regions were retrieved by PCR from pLLEV- human REV1 plasmid and introduced into pET15b
vector (Millipore, Billerica, MA) by an in vitro recombination system using In-Fusion HD Cloning (Takara Bio,
Mountain View, CA). The first fragment encoding codons 31 through 90 was retrieved by primers 5’-AGG AGA
TAT ACC ATG GAA CAG TTT CGA TCA GAT GC-3’ and 5’-ACT CAG AAG AAA AGC ATA TGT GTT GTT TTA GAT CTG
G-3’. The second fragment encoding codons 730 through 1251 was retrieved by 2 consecutive rounds of PCR,
which added a thrombin cleavage site and a stretch of 8 histidines at the C-terminus of the truncated protein.
The primers for the first round of PCR were 5’-GCT TTT CTT CTG AGT CTT TCA G-3’ and 5’-GTG GCT GCT GCC
GCT GCC GCG CGG CAC CAG GCC GCT GCT TGT AAC TTT TAA TGT GCT TCC-3’, the latter of which contained the
thrombin site and a histidine. The product was served for the second round of PCR, with primers 5’-GCT TTT
CTT CTG AGT CTT TCA G-3’ and 5’-GGA TCC TCG AGC ATA TCA ATG ATG GTG ATG ATG ATG ATG ATG GCT GCT
GCC-3’. The latter primer was designed to extend the histidine stretch. The 2 fragments of 210 bp and 1640 bp
were introduced into NcoI and NdeI sites of pET15b vector by In-Fusion. The resultant sequence encodes the
REV1Δ protein (695 AA) consisting of BRCT, UBM1, and UBM2 domains, of which the C-terminus is tagged with
the thrombin site and the histidines but not with most of the catalytic core domain. The underscored
nucleotides in the primer sequences denote the sequence used for recombination.
The resultant plasmid was used to transform E. coli strain BL21(DE3) (Millipore). A 10-L culture was grown at
37°C in LB media plus ampicillin to an OD600nm of 0.6. The culture was then shifted to 16°C, and REV1Δ protein
expression was induced with 0.1 mM IPTG. Growth was continued for a further 16 h, at which point the cell

pellet was collected by centrifugation. The pellet was resuspended in 20 mM HEPES, pH 7.4, 500 mM NaCl,
10% glycerol and 2 mM tris (2-carboxyethyl) phosphine hydrochloride (TCEP) and lysed by passage through a
microfluidizer. The lysate was cleared by centrifugation and applied to a 5-mL bed volume Ni-NTA column (GE
Healthcare). The REV1Δ protein was eluted by adding a linear gradient of imidazole in the same buffer.
Fractions containing REV1Δ were pooled and dialyzed against 20 mM HEPES pH 7.4, 500 mM NaCl, 10%
glycerol, and 2 mM TCEP overnight at 4°C.
Production of site-specifically biotin-tagged UbPCNA protein
Human Uba1 and UbcH5c S22R proteins were expressed in E. coli BL21(DE3) cells (Novagen). Cells were
initially grown at 37°C and induced with 0.5 mM IPTG at 18°C overnight. Cells were harvested in a lysis buffer
of 25 mM Tris, pH 8.0, 500 mM NaCl, and 10% glycerol. Affinity purification was initially done through a Ni-
NTA column followed by a second purification using a 5-mL HiTrap Q HP column (GE Healthcare). Final
purification was performed by size-exclusion chromatography using a HiPrep 16/60 Sephacryl S-200
HR prepacked column in 20 mM sodium phosphate, pH 7.0, 100 mM NaCl, and 2 mM TCEP. This expression
and purification procedure is applicable to both Uba1 and UbcH5c S22R. Next, ubiquitination of PCNA was
performed —initially in small-scale reactions that contained 80 nM Uba1 (E1 ubiquitin ligase), 16 µM UbcH5c
S22R (E2 ubiquitin ligase), 32 µM biotin-AviTag-ubiquitin (Supporting Information), and 16 µM PCNA in a
buffer containing 50 mM Tris or malic acid-MES-Tris (MMT) buffer, pH 8.0, 100 mM NaCl, 3 mM MgCl2, 0.5
mM TCEP, and 3 mM ATP. Experiments were performed for 16 h at 37°C in a 50 µL volume. Finally, the
reaction was optimized for a 500 µL volume consisting of 100 µM Uba1, 128 µM UbcH5c S22R, 256 µM biotin-
AviTag-ubiquitin, and 64 µM PCNA in a buffer containing 50 mM MMT pH 9.0, 25 mM NaCl, 10 mM MgCl2, 0.5
mM TCEP, and 3 mM ATP. Ubiquitinated PCNA was further purified by using a 1-mL HiTrap Q XL column (GE
Healthcare) with 0 to 500 mM NaCl gradient in 20 mM HEPES, pH 8.0, and 0.5 mM TCEP.
AlphaScreen assays
The AlphaScreen assay was conducted in a solution consisting of 200 nM biotin-AviTag-ubiquitin-PCNA protein
and 40 nM N-terminal His-tagged REV1Δ in the AlphaScreen buffer (1x PBS, pH 7.4, 0.1% BSA, and 0.05%
Tween-20). Briefly, 5 μL of the assay solution containing the indicated serial dilution of indicated compound
was transferred into each well of a white OptiPlate-384 (PerkinElmer, Waltham, MA). AlphaScreen nickel
chelate acceptor beads (PerkinElmer) diluted 1:100 in 10 μL of the AlphaScreen buffer were added. After 1 h,
streptavidin AlphaScreen donor beads (PerkinElmer) diluted 1:100 in 10 μL of the AlphaScreen buffer were
added. The AlphaScreen signal was read an hour later by using an EnVision plate reader (PerkinElmer). The

dose-response curve was fitted by using Prism 7 (GraphPad Software, La Jolla, CA) for final concentration of
each serial dilution.
Production of REV1 UBM1-UBM2 and UBM2 proteins
The coding region of UBM1 and UBM2 domains (324 bp, 108 aa) was retrieved by PCR of pLLEV- human REV1
plasmid by using the primers 5’-AAG AGT TCC GGT CTG CCG TCA CCT TCC CAG CTG GAT C-3’ and 5’-TTA GCA
GCC GGA TCC TAT TGT CTT TGA TCA TAC GCT GC -3’. The linear vector was generated by PCR of pET15b by
using the primers 5’-CAG ACC GGA ACT CTT AAT GC-3’ and 5’-GGA TCC GGC TGC TAA CAA AGC C-3’. These
fragments were recombined by In-Fusion. The resultant sequence encodes a protein (137 aa) with UBM1 and
UBM2 domains, with the histidines and thrombin site at its N-terminus. The REV1 UBM1-UBM2 protein was
produced using a method similar to that described for the REV1Δ protein followed by removal of the histidine-
tag by trypsin digestion. Preparation of the REV1 UBM2 protein is reported elsewhere.
STD NMR experiments
STD experiments were performed at 280 K at a concentration of 10 μM protein and 200 μM of compound. For
saturation transfer difference (STD) experiments, pre-saturation was achieved using a pulse train of 30 ms
Gaussian shaped pulses for a 4-s duration and an offset of 0 ppm (on resonance) or −40 ppm (off resonance).
To obtain STD spectra, 256 on and off resonance of free induction decays were recorded. Subtraction of the
on-resonance from the off-resonance spectra yielded the STD signal.
NMR titrations for interaction of compound with REV1 UBM2
A series of two-dimensional [1H, 15N]-HSQC spectra were collected on a Bruker 600 spectrometer at 298K. The
REV1 UBM2 concentration for the entire titration was 50 µM, and the procedure was performed in a 20 mM
sodium phosphate buffer (pH 7.0) and 100 mM NaCl. Chemical shifts (Δ) of individual resonances from 15N-
labeled REV1 UBM2 were monitored as a function of increasing compound 2 concentration (REV1 UBM2:
Compound 2 ratios: 1:2, 1:4, 1:6, 1:8, and 1:10). Compound 2 precipitated when the ratio exceeds 1:10 (not
shown). 15N and 1H chemical shift values for the displaced peaks were determined for each of the successive
titration points by using CARA 1.8.4 31. To determine the per-residue chemical shift perturbation upon binding
and account for differences in spectral widths between 15N and 1H resonances 32, weighted average chemical-
shift differences [Δav(HN)] were calculated for the backbone amide 1H and 15N resonances by using the
equation ∆av (NH) = [(∆H2+ (∆N/5)2)/2]1/2, where ΔH and ΔN are chemical-shift differences for 1H and 15N,
respectively 33-34
.

Chromatin co-localization studies
U2OS cells (4 × 105/well) were initially seeded onto coverslips in six-well cluster plates and allowed to attach
overnight. Each sample was transfected with pEGFP-REV1 (2.5 µg) by using 7.5 µL of Lipofectamine 2000
(Thermo Fisher Scientific) according to the manufacturer’s instructions. Two days following transfection, cells
were treated with 33 µM cisplatin for 3.5 h prior to release into media containing either DMSO vehicle or 150
µM of Compound 1 for 6 h. Samples were then pre-extracted with ice-cold CSK buffer (100 mM NaCl, 300 mM
sucrose, 10 mM HEPES pH 7.4, 3 mM MgCl2, 1 × Halt Protease Inhibitor Cocktail [Thermo Fisher Scientific],
0.5% Triton X-100) for two minutes to remove non-chromatin bound protein and then fixed with 4%
paraformaldehyde for 10 min at ambient temperature. Cells were exposed to 100% methanol for 15 min at
−20°C prior to blocking in 3% FBS in PBS overnight at 4°C. The cells were then immunostained by probing with
1:2000 mouse α PCNA monoclonal antibody (PC10, Cell Signaling Technology) for 60 min followed by
incubation with 1:200 donkey α mouse IgG Alexa 555 (Thermo Fisher Scientific) for 30 min. Cells were then
washed extensively with PBS and mounted onto glass slides by using VectaShield containing 1 μg/mL DAPI.
Immunostained cells were subsequently examined, and images were captured on a TE2000 microscope
equipped with a C1Si confocal (Nikon, Tokyo, Japan). Nikon NIS-elements software (Nikon, Tokyo, Japan) was
used to calculate the Pearson’s correlation coefficient of red (PCNA) and green (REV1) fluorescence signals for
each nucleus, using DAPI staining to define the perimeter of each nuclei.
CPD immunofluorescence assay
HeLa cells (1.5 × 105 cells/dish) were seeded onto coverslips in 35-mm culture dishes and allowed to settle
overnight. Cells were irradiated with 0 or 2.5 J/m2 UVC at 254 nm by using a Stratalinker UV Crosslinker 1800
(Stratagene, La Jolla, CA). Following irradiation, cells were released into media containing DMSO or 150 µM
Compound 1. At designated time points, samples were rinsed with PBS, fixed with 3.7% formaldehyde in PBS
for 15 min at room temperature, permeabilized with 0.5% triton X-100 for 5 min at room temperature, and
stored at 4°C in PBS until all samples had been collected. Samples were subsequently denatured in 2 M HCl for
30 min and blocked in 3% FBS in PBS for 10 min at room temperature. Cells were probed with 1:400 mouse α
CPD monoclonal antibody (clone TDM-2, Cosmo Bio USA) in 3% FBS in PBS for 70 min, thoroughly rinsed with
PBS, and then exposed to 1:200 donkey anti-mouse IgG Alexa 488 (Thermo Fisher) in 3% FBS in PBS for 30 min
at room temperature. Cells were then washed extensively by using 3 × PBS rinses and then mounted onto
glass slides by using VectaShield containing 1 μg/mL DAPI. Samples were visualized by using an Olympus IX81
epi-fluorescence microscope (Olympus Corp., Tokyo, Japan), and Nikon NIS-elements software (Nikon, Tokyo,

Japan) was used to calculate the mean green fluorescence intensity in each nucleus, using the DAPI staining
pattern to define the perimeter of the nucleus.
HPRT gene mutagenesis assay
WTK-1 human lymphoblastoid cells were initially cleansed to minimize background levels of mutation in the
HPRT gene by exposing cells to CHAT in RPMI media supplemented with 10% FBS for 2 days. Cells were
allowed to recover in media containing THC for a further two days prior to conditioning in RPMI media
containing 10% heat-inactivated horse serum. Conditioned WTK-1 cells (2 × 106) were subsequently exposed
to UVC (0 – 128 J/m2) by using a UVP model UVGL-58 Handheld UV Lamp (Upland, CA) in the short-wavelength
(254 nm) mode. Immediately following UVC treatment, cells were treated with either DMSO or 150 µM
Compound 1 for 21 h. Following treatment, cells were resuspended in drug-free RPMI media supplemented
with 10% horse serum, 1% penicillin/streptomycin, and 200 µg/mL sodium pyruvate. A portion of each
suspension (0.5 mL) was used to evaluate survival (PE0) by immediate re-distribution into 96-well cluster
plates at an average cell density of 1.6 cells per well. PE0 survival plates were incubated for 11 days and then
scored for wells exhibiting colony growth. The remaining bulk of each cell suspension was cultured (4.8 × 106
cells per treatment) for a further five days to enable fixation of drug-induced mutations in the HPRT gene.
Following fixation, cells were re-plated into 96-well cluster plates at a density of 5,000 cells per well in
complete media containing 0.5 µg/mL 6-thioguanine (6TG) as a selective agent. Cells were also re-distributed
in parallel into 96-well cluster plates without selective agent at an average density of 1.6 cells per well to
evaluate survival (PE6). All 6TG and PE6 plates were incubated for a11 – 12 more days and then scored for
colony formation.
Clonogenic cell survival assay
After U2OS cells (300 cells/well) were settled in 6-well plates overnight, each indicated final concentration of
cisplatin or 4-hydroxycyclophosphamide was added. After 24 h, Compound 1 or DMSO was added as indicated.
After another 24 h, the culture medium was replaced with fresh medium, and cells were cultured for another
7 days. Each treatment was done in triplicate. The cells were gently rinsed twice with PBS, fixed with 3.7%
formaldehyde, and stained by using 0.5% crystal violet. Colonies containing more than approximately 40 cells
were counted. The survival rate was reported as 1 for the average of colony numbers in wells that received
vehicle control (0.9% sodium chloride for cisplatin and DMSO for 4-hydroxycyclophosphamide).

ToC graphic

REFERENCES
(1)
Helleday, T.; Petermann, E.; Lundin, C.; Hodgson, B.; Sharma, R. A. DNA repair pathways as targets for cancer
therapy. Nature reviews. Cancer 2008, 8, 193-204.
(2)
(3)
Deans, A. J.; West, S. C. DNA interstrand crosslink repair and cancer. Nature reviews. Cancer 2011, 11, 467-480.
Ho, T. V.; Scharer, O. D. Translesion DNA synthesis polymerases in DNA interstrand crosslink repair.
Environmental and molecular mutagenesis 2010, 51, 552-566.
(4) Sharma, S.; Canman, C. E. REV1 and DNA polymerase zeta in DNA interstrand crosslink repair. Environmental
and molecular mutagenesis 2012, 53, 725-740.
(5) Hicks, J. K.; Chute, C. L.; Paulsen, M. T.; Ragland, R. L.; Howlett, N. G.; Gueranger, Q.; Glover, T. W.; Canman, C. E.
Differential roles for DNA polymerases eta, zeta, and REV1 in lesion bypass of intrastrand versus interstrand DNA cross-
links. Molecular and cellular biology 2010, 30, 1217-1230.
(6)
Guo, C.; Tang, T. S.; Bienko, M.; Parker, J. L.; Bielen, A. B.; Sonoda, E.; Takeda, S.; Ulrich, H. D.; Dikic, I.; Friedberg,
E. C. Ubiquitin-binding motifs in REV1 protein are required for its role in the tolerance of DNA damage. Molecular and
cellular biology 2006, 26, 8892-8900.
(7)
Xie, K.; Doles, J.; Hemann, M. T.; Walker, G. C. Error-prone translesion synthesis mediates acquired
chemoresistance. Proceedings of the National Academy of Sciences of the United States of America 2010, 107, 20792-
20797.
(8)
Sasatani, M.; Xi, Y.; Kajimura, J.; Kawamura, T.; Piao, J.; Masuda, Y.; Honda, H.; Kubo, K.; Mikamoto, T.;
Watanabe, H.; Xu, Y.; Kawai, H.; Shimura, T.; Noda, A.; Hamasaki, K.; Kusunoki, Y.; Zaharieva, E. K.; Kamiya, K.
Overexpression of Rev1 promotes the development of carcinogen-induced intestinal adenomas via accumulation of
point mutation and suppression of apoptosis proportionally to the Rev1 expression level. Carcinogenesis 2017, 38, 570-
578.
(9)
human ovarian carcinoma cells. Molecular pharmacology 2006, 69, 1748-1754.
(10) Dumstorf, C. A.; Mukhopadhyay, S.; Krishnan, E.; Haribabu, B.; McGregor, W. G. REV1 is implicated in the
development of carcinogen-induced lung cancer. Molecular cancer research : MCR 2009, 7, 247-254.
(11) Jansen, J. G.; Langerak, P.; Tsaalbi-Shtylik, A.; van den Berk, P.; Jacobs, H.; de Wind, N. Strand-biased defect in
Lin, X.; Okuda, T.; Trang, J.; Howell, S. B. Human REV1 modulates the cytotoxicity and mutagenicity of cisplatin in
C/G transversions in hypermutating immunoglobulin genes in Rev1-deficient mice. The Journal of experimental
medicine 2006, 203, 319-323.
(12)
interacts with PCNA: significance of the REV1 BRCT domain in vitro and in vivo. Mol Cell 2006, 23, 265-271.
(13) Sharma, N. M.; Kochenova, O. V.; Shcherbakova, P. V. The non-canonical protein binding site at the monomer-
Guo, C.; Sonoda, E.; Tang, T. S.; Parker, J. L.; Bielen, A. B.; Takeda, S.; Ulrich, H. D.; Friedberg, E. C. REV1 protein
monomer interface of yeast proliferating cell nuclear antigen (PCNA) regulates the Rev1-PCNA interaction and
Polzeta/Rev1-dependent translesion DNA synthesis. J Biol Chem 2011, 286, 33557-33566.
(14)
203.
(15)
Lawrence, C. W. Cellular functions of DNA polymerase zeta and Rev1 protein. Adv Protein Chem 2004, 69, 167-
Sharma, S.; Shah, N. A.; Joiner, A. M.; Roberts, K. H.; Canman, C. E. DNA Polymerase Zeta is a Major Determinant
of Resistance to Platinum-based Chemotherapeutic Agents. Molecular pharmacology 2012.
(16) Wu, F.; Lin, X.; Okuda, T.; Howell, S. B. DNA polymerase zeta regulates cisplatin cytotoxicity, mutagenicity, and
the rate of development of cisplatin resistance. Cancer Res 2004, 64, 8029-8035.
(17) Inoue, A.; Kikuchi, S.; Hishiki, A.; Shao, Y.; Heath, R.; Evison, B. J.; Actis, M.; Canman, C. E.; Hashimoto, H.; Fujii, N.
A small molecule inhibitor of monoubiquitinated Proliferating Cell Nuclear Antigen (PCNA) inhibits repair of interstrand
DNA cross-link, enhances DNA double strand break, and sensitizes cancer cells to cisplatin. J Biol Chem 2014, 289, 7109-
7120.
(18)
Actis, M. L.; Ambaye, N. D.; Evison, B. J.; Shao, Y.; Vanarotti, M.; Inoue, A.; McDonald, E. T.; Kikuchi, S.; Heath, R.;
Hara, K.; Hashimoto, H.; Fujii, N. Identification of the first small-molecule inhibitor of the REV7 DNA repair protein
interaction. Bioorganic & medicinal chemistry 2016, 24, 4339-4346.

(19)
Sail, V.; Rizzo, A. A.; Chatterjee, N.; Dash, R. C.; Ozen, Z.; Walker, G. C.; Korzhnev, D. M.; Hadden, M. K.
Identification of Small Molecule Translesion Synthesis Inhibitors That Target the Rev1-CT/RIR Protein-Protein Interaction.
ACS chemical biology 2017, 12, 1903-1912.
(20)
synthesis DNA polymerase Rev1 mediated by Rev1-BRCT domain. Journal of molecular biology 2013, 425, 3091-3105.
(21) Chen, J.; Ai, Y.; Wang, J.; Haracska, L.; Zhuang, Z. Chemically ubiquitylated PCNA as a probe for eukaryotic
translesion DNA synthesis. Nature chemical biology 2010, 6, 270-272.
(22) Hoege, C.; Pfander, B.; Moldovan, G. L.; Pyrowolakis, G.; Jentsch, S. RAD6-dependent DNA repair is linked to
modification of PCNA by ubiquitin and SUMO. Nature 2002, 419, 135-141.
(23) Lau, W. C.; Li, Y.; Zhang, Q.; Huen, M. S. Molecular architecture of the Ub-PCNA/Pol eta complex bound to DNA.
Scientific reports 2015, 5, 15759.
(24) Hibbert, R. G.; Sixma, T. K. Intrinsic flexibility of ubiquitin on proliferating cell nuclear antigen (PCNA) in
translesion synthesis. J Biol Chem 2012, 287, 39216-39223.
(25) Wigle, T. J.; Herold, J. M.; Senisterra, G. A.; Vedadi, M.; Kireev, D. B.; Arrowsmith, C. H.; Frye, S. V.; Janzen, W. P.
Pustovalova, Y.; Maciejewski, M. W.; Korzhnev, D. M. NMR mapping of PCNA interaction with translesion
Screening for inhibitors of low-affinity epigenetic peptide-protein interactions: an AlphaScreen-based assay for
antagonists of methyl-lysine binding proteins. Journal of biomolecular screening 2010, 15, 62-71.
(26)
(27)
Hurley, J. H.; Lee, S.; Prag, G. Ubiquitin-binding domains. The Biochemical journal 2006, 399, 361-372.
Actis, M.; Inoue, A.; Evison, B.; Perry, S.; Punchihewa, C.; Fujii, N. Small molecule inhibitors of PCNA/PIP-box
interaction suppress translesion DNA synthesis. Bioorganic & medicinal chemistry 2013, 21, 1972-1977.
(28) Mayer, M.; Meyer, B. Group epitope mapping by saturation transfer difference NMR to identify segments of a
ligand in direct contact with a protein receptor. Journal of the American Chemical Society 2001, 123, 6108-6117.
(29) McCulloch, S. D.; Kokoska, R. J.; Masutani, C.; Iwai, S.; Hanaoka, F.; Kunkel, T. A. Preferential cis-syn thymine
dimer bypass by DNA polymerase eta occurs with biased fidelity. Nature 2004, 428, 97-100.
(30) Burschowsky, D.; Rudolf, F.; Rabut, G.; Herrmann, T.; Peter, M.; Wider, G. Structural analysis of the conserved
ubiquitin-binding motifs (UBMs) of the translesion polymerase iota in complex with ubiquitin. J Biol Chem 2011, 286,
1364-1373.
(31)
Keller, R. The Computer Aided Resonance Assignment Tutorial: Verlag, Cantina, Switzerland. 2004.
(32)
Farmer, B. T., 2nd; Constantine, K. L.; Goldfarb, V.; Friedrichs, M. S.; Wittekind, M.; Yanchunas, J., Jr.; Robertson,
J. G.; Mueller, L. Localizing the NADP+ binding site on the MurB enzyme by NMR. Nature structural biology 1996, 3, 995-
997.
(33)
Grzesiek, S.; Bax, A.; Clore, G. M.; Gronenborn, A. M.; Hu, J. S.; Kaufman, J.; Palmer, I.; Stahl, S. J.; Wingfield, P. T.
The solution structure of HIV-1 Nef reveals an unexpected fold and permits delineation of the binding surface for the
SH3 domain of Hck tyrosine protein kinase. Nat.Struct.Biol. 1996, 3, 340-345.
(34)
Garrett, D. S.; Seok, Y. J.; Peterkofsky, A.; Clore, G. M.; Gronenborn, A. M. Identification by NMR of the binding
surface for the histidine-containing phosphocarrier protein HPr on the N-terminal domain of enzyme I of the Escherichia
coli phosphotransferase system. Biochemistry 1997, 36, 4393-4398.