An assay for DNA polymerase β lyase inhibitors that engage the catalytic nucleophile for binding

Article abstract of DOI:10.1016/j.bmc.2020.115642

DNA polymerase β (Pol β) repairs cellular DNA damage. When such damage is inflicted upon the DNA in tumor cells treated with DNA targeted antitumor agents, Pol β thus diminishes their efficacy. Accordingly, this enzyme has long been a target for antitumor therapy. Although numerous inhibitors of the lyase activity of the enzyme have been reported, none has yet proven adequate for development as a therapeutic agent. In the present study, we developed a new strategy to identify lyase inhibitors that critically engage the lyase active site primary nucleophile Lys72 as part of the binding interface. This involves a parallel evaluation of the effect of the inhibitors on the wild-type DNA polymerase β (Pol β) and Pol β modified with a lysine analogue at position 72. A model panel of five structurally diverse lyase inhibitors identified in our previous studies (only one of which has been published) with unknown modes of binding were used for testing, and one compound, cis-9,10-epoxyoctadecanoic acid, was found to have the desired characteristics. This finding was further corroborated by in silico docking, demonstrating that the predominant mode of binding of the inhibitor involves an important electrostatic interaction between the oxygen atom of the epoxy group and N^ε of the main catalytic nucleophile, Lys72. The strategy, which is designed to identify compounds that engage certain structural elements of the target enzyme, could find broader application for identification of ligands with predetermined sites of binding.

Full text of DOI:10.1016/j.bmc.2020.115642

Bioorganic&MedicinalChemistry28(2020)115642
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
An assay for DNA polymerase β lyase inhibitors that engage the catalytic
nucleophile for binding
Sasha M. Daskalovaa, Brian M. Eisenhauerb, Mingxuan Gaoa, Xizhi Fengb, Xun Jia, Qi Chenga,
⁎
NourEddine Fahmib, Omar M. Khdoura, Shengxi Chena, Sidney M. Hechta,b,
a Biodesign Center for BioEnergetics and School of Molecular Sciences, Arizona State University, Tempe, AZ 85287, United States
b Departments of Chemistry and Biology, University of Virginia, Charlottesville, VA 22904, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
DNA polymerase β (Pol β) repairs cellular DNA damage. When such damage is inflicted upon the DNA in tumor
cells treated with DNA targeted antitumor agents, Pol β thus diminishes their efficacy. Accordingly, this enzyme
has long been a target for antitumor therapy. Although numerous inhibitors of the lyase activity of the enzyme
have been reported, none has yet proven adequate for development as a therapeutic agent. In the present study,
we developed a new strategy to identify lyase inhibitors that critically engage the lyase active site primary
nucleophile Lys72 as part of the binding interface. This involves a parallel evaluation of the effect of the in-
hibitors on the wild-type DNA polymerase β (Pol β) and Pol β modified with a lysine analogue at position 72. A
model panel of five structurally diverse lyase inhibitors identified in our previous studies (only one of which has
been published) with unknown modes of binding were used for testing, and one compound, cis-9,10-epox-
yoctadecanoic acid, was found to have the desired characteristics. This finding was further corroborated by in
silico docking, demonstrating that the predominant mode of binding of the inhibitor involves an important
electrostatic interaction between the oxygen atom of the epoxy group and Nε of the main catalytic nucleophile,
Lys72. The strategy, which is designed to identify compounds that engage certain structural elements of the
target enzyme, could find broader application for identification of ligands with predetermined sites of binding.
Lyase inhibitor
Non-proteinogenic amino acid
Active site interaction
Enzyme modification
Targeted inhibitor selection
1. Introduction
Epstein-Barr virus.17,18 Overexpressed Pol β is redirected to the re-
plication forks, causing interference with chromosomal DNA replica-
DNA polymerase beta (Pol β) is the smallest known DNA poly-
merase; with an error rate of 1/1500 nucleotides it has the lowest fi-
delity of the DNA polymerases.1 Structurally, Pol β folds into two in-
dependent domains connected by a short linker. The smaller, 8 kDa N-
terminal domain, known as the lyase domain, recognizes gapped abasic
DNA substrates and participates in the removal of the 5′-deoxyribose
phosphate (dRP) group.2–4 The C-terminal 31 kDa domain has weak
binding affinity for double-stranded DNA and, using its nucleotidyl
transferase activity, fills the gap generated by the lyase activity with the
nucleotides required for DNA duplex repair.2,5 The combination of lyase
and polymerase activities within a single protein moiety is well suited
to the primary role of mammalian Pol β in both short-6–8 and long-
patch9–11 base-excision repair (BER) pathways. In light of this house-
keeping function, it is unsurprising that defects in Pol β are associated
with a number of pathologies, including cancer12 and neurodegenera-
tive diseases.13 Pol β is frequently overexpressed in uterine, ovarian,
prostate and stomach tumors,14 as well as colon and esophageal can-
cers,15,16 and in patients infected with human papillomavirus 16 and
tion.19 This ultimately leads to genetic instability20,21 and plays a major
role in cancer predisposition and tumor progression. Pol β-mediated
non-canonical pathway of DNA synthesis also appears to contribute to
neuronal death of β-amyloid peptide treated neurons.22 Cells en-
gineered to overexpress Pol β acquire a spontaneous mutator phenotype
and display decreased sensitivity to cancer chemotherapeutic DNA-
damaging agents such as cisplatin, melphalan, and mechlorethamine,23
presumably due in part to the repair of DNA lesions induced by the DNA
damaging agents.24 In fact, Pol β overexpression appears to be a pri-
mary cause for the limited efficacy of many antitumor drugs.24–31
In light of the foregoing findings, much research has been focused
on the identification and development of potent Pol β inhibitors.32
Numerous compounds that inhibit the function of Pol β have been
isolated from natural sources or prepared by synthesis. However, given
that there are 16 different human DNA polymerases that participate in
DNA replication, repair and recombination,33 it has been challenging to
identify specific and potent Pol β inhibitors. Theoretically, there could
be several approaches to inhibiting Pol β activity. These include
⁎ Corresponding author.
https://doi.org/10.1016/j.bmc.2020.115642
Received 12 May 2020; Received in revised form 3 July 2020; Accepted 5 July 2020
Availableonline09July2020
0968-0896/©2020ElsevierLtd.Allrightsreserved.

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
targeting the DNA binding affinity of the enzyme, disrupting Pol β-
dependent DNA repair complexes, affecting post-translational mod-
ifications critical for enzyme function or directly inhibiting Pol β lyase
or polymerase activities. Lyase inhibitors might well be thought to hold
the greatest promise for specificity, given both the uniqueness of this
domain, and the obligatory intermediacy of a normally transient lyase-
DNA covalent binary complex.
of distinguishing the nature of binding interactions at the lyase active
site, we prepared and characterized three analogues of Lys72, namely
homolysine (III), p-aminophenylalanine (IV) and p-aminomethylphe-
nylalanine (V), that would reposition the side chain amine by about
0.1–1.3 Å yet retain similar nucleophilicity (Fig. 3A, Table 1). The ly-
sine analogues were chosen to cause minimal perturbation of the lyase
catalytic pocket other than repositioning of the lysine side chain amine,
providing modified enzymes that retained significant lyase activity.
Additionally, we included N-methyl-lysine (II) as a positive control in
addition to wild type. The structural similarity of II to lysine (I), with
identical Cβ-Nε distance and slightly altered nucleophilicity and steric
bulk of the ε-amino group, was anticipated to have little impact on the
lyase activity of the modified enzyme. In comparison, substitution of
Lys72 with valine (VI) should afford an enzyme with no lyase activity,
thereby fulfilling the role of a negative control.
While numerous inhibitors of the lyase activity have been reported,
only limited work has been done to characterize the nature of the
binding mechanism(s) which conduce to effective and selective in-
hibition. Perhaps more surprisingly, little has been done to identify
inhibitors that engage the catalytic nucleophile in the lyase active site,
i.e. Lys72. Such inhibitors might be expected to affect the formation or
stability of the normally transient covalent protein-DNA binary com-
plex. Accordingly, a key goal of this study was to identify Pol β lyase
inhibitors that engage the primary catalytic nucleophile of Lys72 as part
of the inhibitor Pol β complex. To achieve that outcome, we developed
a strategy predicated on comparing the effect of Pol β inhibitors on the
wild-type enzyme vs modified enzymes containing lysine analogues at
position 72. An inhibitor having the desired characteristic of interacting
with the Lys72 side chain amine to form contacts critical for Pol β
binding would be expected to decrease the lyase activity of the wild-
type enzyme, but have a diminished (or no) effect on the activity of the
modified enzyme. This is illustrated in Fig. 1 in which inhibitor I1
blocks substrate conversion to product by the wild-type and modified
Pol β enzymes, while inhibitor I2 affects only the wild-type enzyme,
putatively by interaction with the side chain amino group of Lys72.
Presently, we describe our study of the interaction of five Pol β lyase
inhibitors (Fig. 2) with wild-type and modified enzymes having Nε-
methyl-lysine (II) or homolysine (III) at position 72 (Fig. 3). Com-
pounds 2–5 are newly reported inhibitors of Pol β. One of these in-
hibitors (cis-9,10-epoxyoctadecanoic acid (4)) exhibited binding to the
enzyme critically dependent on the presence of the primary catalytic
nucleophile of Lys72. A less pronounced, but qualitatively analogous
observation was also made for 3,4-dihydroxybenzoic acid (3).
The actual preparation of each of the polymerases was carried out as
described in Fig. 4. As illustrated in the Fig. for homolysine, N-pente-
noylhomolysyl-pdCpA was ligated to a suppressor tRNACUA in vitro
transcript lacking the 3′-terminal pCpAOH moiety by the use of T4 RNA
ligase. The N-pentenoyl-protected aminoacyl-tRNAs were deprotected
by brief treatment with aqueous iodine.34,35 Each of the misacylated
tRNAs was used in a coupled in vitro transcription/translation of rat Pol
β from a DNA template having an amber stop codon corresponding to
position 72 of Pol β. The enzymes so derived were purified by Ni-NTA
affinity chromatography, then analyzed by sodium dodecyl-poly-
acrylamide gel electrophoresis (SDS-PAGE). The detailed procedures
have been described previously.36
The six amino acid analogues shown in Fig. 3A were incorporated
successfully and the corresponding modified enzymes were purified
(Fig. 3B). As expected, the N-methyllysine-modified enzyme demon-
strated lyase activity similar to the wild-type enzyme, about 86%.
Among the other enzymes prepared, only the homolysine-substituted
Pol β (III) showed moderate lyase activity, ~35% of the wild-type Pol β
levels. The enzymes having IV and V at position 72 had only minimal
lyase activity at best (Fig. 3C), and were not investigated further. Based
on these results, we employed wild-type enzyme expressed in vivo as a
control and, in addition, in vitro synthesized enzymes having II or III at
position 72 to study the effects of compounds 1–5. The enzyme mod-
ified with II allowed the steric effects of a small substituent to be as-
sessed, and also verified that the method of enzyme production did not
affect the results. Homolysine-substituted Pol β was the only suitable
choice for a modified enzyme capable of distinguishing the nature of
binding interactions at the lyase active site, as a consequence of having
the NH2 group of its side chain repositioned, relative to that in wild
type.
The predominant mode of binding of the cis-epoxide of oleic acid
(4), was studied further by in silico docking, and the results obtained
provided additional support for the experimental findings. While none
of the compounds studied here is known to be selective for inhibition of
Pol β, nor is any potent enough to be employed for the inhibition of the
enzyme under physiological conditions, the compounds do provide
proof of principle for a new strategy that can be used to identify
structural motifs that may be developed further as useful inhibitors.
2. Results and discussion
To measure the activity of wild type and modified Pol β enzymes,
we employed an assay described previously in which a synthetic DNA
In order to identify a Pol β analogue modified at position 72 capable
Fig. 1. Inhibition of wild-type Pol β and modified Pol β containing homolysine at position 72 in lieu of lysine by two hypothetical inhibitors. Inhibitor I1 affects the
conversion of substrate to product by both enzymes; inhibitor I2 affects only wild type.
2

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Fig. 2. Structures of compounds investigated for their Pol β lyase inhibitory activities, including β-stigmasterol (1), 3-β-(benzoyloxy)stigmastane (2), 3,4-dihy-
droxybenzoic acid (3), cis-9,10-epoxyoctadecanoic acid (4) and 9,10-dihydroxyoctadecanoic acid (5). Compounds 2–5 have not been reported previously to inhibit
Pol β. Compounds 4 and 5 are racemic.
duplex having a single site on one strand bearing an apurinic acid was
cleaved with AP endonuclease and then used as a substrate for 5′-dRP
excision by each of the chosen Pol β enzymes. The course of the excision
reaction was monitored by 20% denaturing polyacrylamide gel elec-
trophoresis and autoradiography.
lysine (I) and N-methyl-lysine (II) to the extent of roughly 50% under
the reaction conditions, but produced only 25–30% inhibition of Pol β
containing homolysine under the same conditions (Fig. 6). This ob-
servation implied that the side chain N residue of Lys72 participates in
binding this inhibitor, although not crucially.
Potential inhibitors of 5′-dRP excision were tested initially using rat
DNA Pol β over a range of concentrations (typically 10–50 μM) and
representative data are presented in Figs. 5–7.
A more dramatic difference was observed when using cis-9,10-
epoxyoctadecanoic acid (4) as the inhibitor (Fig. 7). Essentially com-
plete inhibition of Pol β s containing I and II was achieved using 50 μM
compound 4. In comparison, Pol β III was unaffected by inhibitor 4,
even though this enzyme had only about one-third the activity of Pol β s
containing I and II in the absence of any inhibitor. Ring opening of the
epoxide in 4 with perchloric acid afforded diol 5.
As shown in Fig. 5 for β-stigmasterol (1) and 3-β-(benzoyloxy)
stigmastane (2), both of these compounds were strongly inhibitory to
5′-dRP excision of Pol β enzymes containing I, II and III when employed
at 50 μM concentrations. In comparison, when used at the same con-
centration, 3,4-dihydroxybenzoic acid (3) inhibited Pol β containing
As shown in Fig. 7, this compound was a more potent inhibitor of all
Fig. 3. Synthesis and lyase activity of modified rat Pol β enzymes. (A) Structures of lysine and lysine analogues incorporated into Pol β at position 72. (B) SDS-PAGE
analysis of Ni2+ NTA-purified modified Pol β enzymes visualized by autoradiography. (C) Relative 5′-dRP excision activities of the modified enzymes. The activity of
in vitro synthesized wild-type Pol β was determined to be 8.64
0.78 pmol/min/mg under the conditions of the assay.
3

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Table 1
to Schiff base formation.
Amino Acid Analogues Incorporated into Pol β Position 72 and their Calculated
The binding interface of inhibitor 5 was different. The key binding
residues were found to be Leu11, Ile15, Met18, Leu19, Tyr39, Lys72,
Phe76 and Leu82 (Fig. 9C, Table S3). Lys72 Nε is in proximity to the
carboxylate oxygen moiety of 9R,10R-5 (4.7 Å), but quite distant from
the vicinal diol moiety, and was not engaged in any interaction assisting
binding of Pol β to compound 5 (Table S3). Analogous findings were
made for 9S,10R-5 and 9R,10S-5 (Fig. S1, Table S4). However, the
binding of 9S,10S-5 to Pol β did position the vicinal diol moiety of 5
close to Lys72 (Fig. S1, Table S4). The distance between the O atom of
the 9-OH group of 9S,10S-5 and the Lys72 Nε group was 3.63 Å, while
the analogous distance for the 10-OH O atom was 4.8 Å. While the
proximity of these OH groups to the Lys72 Nε group could in principle
allow stabilization of the binding of 9S,10S-5 to Pol β, the vicinol diol is
likely more mobile conformationally than the epoxide moiety of 4. This
seems consistent with the lack of selectivity of compound 5 toward wild
type and modified enzymes (Fig. 7B). Also, it may be noted that for
9S,10S-5 there are polar bonds involving Lys35 and Tyr39 that may
stabilize Pol β binding.
Side Chain Lengths and pKa Values.a
Pol β
Amino acid 72
distance from Cβ
to side chain N (Å)
pKa of side chain
nucleophilesb
I
Lysine
6.3
6.3
7.6
6.4
7.3
–
10.3
10.6
10.3
4.6
9.3
–
II
N-methyl-lysine
homolysine
III
IV
V
p-aminophenylalanine
p-aminomethylphenylalanine
valine
VI
a
Energy minimization of the side chain was performed using Chem3D Ultra
Calculated values (https://chemicalize.com/welcome).
7.0.
b
three DNA polymerases but failed to show selectivity at any tested
concentration.
The experiments supporting the key conclusions of this work were
repeated using human DNA Pol β with very similar results (Fig. 8). As
for rat Pol β, 25 μM 4 was inhibitory to Pol β enzymes I and II, but not
to Pol β III. In comparison, diol 5 inhibited all three forms of Pol β.
The foregoing findings are consistent with the interpretation that
the selectivity observed for compound 4 was due to the actual parti-
cipation of the catalytic amino groups of lysine and N-methyl-lysine at
position 72 in inhibitor binding, using a spatial orientation plausibly
not accessible to the analogous amino group in homolysine. (cf Fig. 3A
and Table 1). To explore this possibility further, the predominant mode
of binding of compound 4 and, for comparison, structurally related 5
were also studied by molecular docking. Eight residues, namely Met18,
Leu19, Leu22, Glu26, Lys35, Ala38, Tyr39 and Lys72 were identified as
being in close proximity to (9S,10R)-4 (Fig. 9A, Table S1). Because
synthetic 4 is a mixture of two isomers, analogous docking with
(9R,10S)-4 was carried out and the two highest ranking models in-
dicated a structurally related docking mode but with the epoxide
moiety of 4 slightly further from Lys72 Nε (Fig. 9B, Table S2). In
agreement with the biochemical results (Fig. 7A), for (9S,10R)-4 the
side chain amino group of nucleophile Lys72 was identified as part of
the binding network and predicted to form a polar bond engaging Lys72
Nε and the epoxide O of the potential inhibitor. This was the only polar
bond stabilizing (9S,10R)-4 interaction with Pol β. For (9R,10S)-4, the
lysine Nε – inhibitor epoxide O distance was somewhat greater than for
(9S,10R)-4 (3.9 – 5.1 Å, Fig. 9B). The estimated free energy of binding
and inhibitory constant Ki of (9S,10R)-4 (-3.88 kcal/mol and 1.44 mM,
respectively) suggested the low potency of the compound as an in-
hibitor; therefore, further kinetics study was not performed. Finally, it
should be mentioned that while we observed no effect of preincubation
of Pol β + 4 on possible inhibition of the lyase activity by covalent
reaction of Lys72 Nε with the electrophilic epoxide in 4, and Lys72 Nε is
not available for reaction with the epoxide once Schiff base formation
has occurred, we cannot exclude the possibility of covalent reaction
between Pol β and 4 subsequent to protein–DNA interaction, but prior
We also performed blind docking of β-stigmasterol (1) to the Pol β
lyase domain. The biochemical data (Fig. 5A) suggest that the position
of the catalytic nucleophile in amino acid 72 is not important for in-
hibitor binding. Our in silico prediction was entirely consistent, as the
inhibitor was predicted to bind to a region of Pol β quite distant from
Lys72 (Fig. 10, Table S5).
As a consequence of the key role of Pol β in base excision repair
(BER) and its importance as a therapeutic target, substantial effort has
been directed toward identifying potent, selective inhibitors. While Pol
β has two active sites that can be targeted,24 compounds capable of
affecting the lyase activity of the enzyme appear to be the most pro-
mising approach to achieve specificity, as only four of the 16 human
DNA polymerases other than Pol β employ 5′-dRP excision and thus
could potentially be affected by Pol β lyase inhibitors. These are Pol λ37
which is involved in DNA double-strand break repair38 and possibly
BER,39 Pol θ40 that plays multiple roles in DNA repair and regulation of
genome integrity,41 Pol ι,42 which is capable of facilitating limited
translesion replication of a thymine–thymine cyclobutane pyrimidine
dimer,43 and mitochondrial Pol γ44 which is involved in replication,
recombination and repair of mitochondrial DNA.45 However, Pol λ is
predominantly expressed in testis and has only basal levels in other
tissues,46 Pol γ is localized in the mitochondria, while a possible Pol ι
inhibition should not be as detrimental considering that Pol η plays the
central role in translesion DNA synthesis.47
The obligatory covalent Schiff base formed transiently between the
5′-dRP intermediate and Lys72 of Pol β is also formally analogous to the
covalent intermediates formed reversibly between the DNA duplex and
the DNA duplex unwinding enzymes DNA topoisomerases I and II. The
formed binary complexes between these enzymes and DNA are the
targets of several topoisomerase poisons currently used clinically as
antitumor agents.48,49
The identification of naturally occurring Pol β lyase inhibitors
Fig. 4. Preparation of a suppressor tRNACUA transcript activated with homolysine, and incorporation of homolysine into position 72 of Pol β in an in vitro protein
biosynthesizing system.
4

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Fig. 5. Evaluation of β-stigmasterol (1) and (B) 3-β-(benzoyloxystigmastane) (2) for their ability to engage Lys72-Nε as part of the rat Pol β-inhibitor binding
interface. Lyase 5′-dRP excision assays and the corresponding histograms showing the relative activities of Pol β enzymes containing lysine (I), N-methyl-lysine (II) or
homolysine (III) at position 72 in the absence (gray bar) or presence (black bar) of (A) β-stigmasterol (1) or (B) 3-β-(benzoyloxystigmastane) (2). WT, wild-type Pol β.
generally involves use of the 5′-dRP excision assay, although com-
pounds that stabilize the Schiff base intermediate have also been of
interest.50 Numerous natural and synthetic products with the potential
to suppress 5′-dRP excision have been identified in this fashion.51–60
However, as currently practiced the lyase assay itself does not provide
information about the binding site of the inhibitor. Therefore, addi-
tional techniques such as NMR chemical shift mapping61 or molecular
docking simulations62 have been employed to determine the exact or
predominant mode of interaction between the inhibitors and the Pol β
lyase domain. The currently used lyase assay also fails to focus on
compounds that bind the active site cleft of the lyase, which might be
thought to afford the best lead compounds for further optimization.
The strategy employed in the present study provides a means to
identify inhibitors that engage catalytic residue Lys72 to achieve
binding. This should conduce to the identification of compounds ex-
hibiting improved selectivity for binding of Pol β lyase in comparison
with other DNA polymerases, but was not studied with the present
compounds as it was not a goal of this investigation. We employed
natural products, as well as a synthetic library whose construction was
guided by structural elements common to a number of reported Pol β
inhibitors,32 to identify structurally diverse compounds for detailed
characterization. Four of the compounds studied (2–5) are new Pol β
inhibitors. Compounds which afforded responses characteristic of those
studied included the plant-derived sterol β-stigmasterol (1) and its O-
benzoylated, fully reduced analogue 3-β-(benzoyloxy)stigmastane (2).
Because many of the characterized Pol β inhibitors involve (un-
saturated) fatty acids, and aromatic derivatives thereof, we also
investigated several fatty acids, as well as simple, functionalized aro-
matic compounds. The results obtained with compounds of this type are
exemplified by those obtained for 3,4-dihydroxybenozic acid (3) and
oxygenated fatty acids 4 and 5. When tested in a modified version of the
5′-dRP excision assay using an enzyme in which the putative catalytic
side chain nucleophile of Pol β Lys72 was repositioned by replacing
lysine with homolysine, these five compounds displayed a range of
responses which provided insights into their modes of Pol β lyase
binding.
As shown in Fig. 5, both stigmasterol (1) and 3-β-(benzoyloxy)
stigmastane (2) inhibited 5′-dRP excision by Pol β lyase. The fact that
the Pol β containing homolysine at position 72 was inhibited to an
extent comparable to the enzymes containing lysine and N-methyl-ly-
sine is consistent with the interpretation that the side chain amino
group of residue 72 was not involved in inhibitor binding. In fact,
computational docking of β-stigmasterol (1) to the Pol β lyase domain
(Fig. 10, Table S5) indicated that binding did not involve any structural
element close to amino acid 72.
The binding of a Pol β inhibitor to a region distant from the active
site is not without precedent. Pamoic acid, a naphthoic acid derivative,
activates ERK by binding to G-protein coupled receptor GPR35.63 Pa-
moic acid also inhibits DNA Pol β, and the binding interaction has been
studied in some detail. The computational docking of pamoic acid to
Pol β illustrated in Fig. S2 and the interactions summarized in Table S6
are consistent with previous findings that its binding site mapped to the
groove between helices 2 and 4 and not including Lys72.62,64 Conse-
quently, pamoic acid exerts its inhibitory effect via interference with
5

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
inhibitors that interact with Lys72 may disfavor formation of the ob-
ligatory Schiff base normally formed between the Pol β lyase domain
and the abasic DNA intermediate formed during DNA repair, thus more
effectively blocking the repair of DNA by this enzyme.
It may be noted that the purpose of this study was to demonstrate
the feasibility of a novel strategy to identify compounds that engage
active site residue Lys72 of Pol β in binding to the lyase active site.
None of the compounds studied is a potent inhibitor of the lyase, which
is a common characteristic of presently known lyase inhibitors.32 The
selectivity of the five compounds studied as Pol β inhibitors, as com-
pared with other cellular DNA polymerases, is likewise unknown at
present. While additional experimentation will be required to identify
potent and selective inhibitors of Pol β lyase, the novel strategy em-
ployed in this study to identify inhibitors that engage active site residue
Lys72 in binding to Pol β should be extensible to other active site re-
sidues in Pol β and other DNA polymerases.
Finally, while the current study establishes proof of principle for
identifying compounds that utilize a specific active site residue for
target site binding, the results do not demonstrate that the inhibitors so
identified will be superior to those that function by binding elsewhere
to the target enzyme, e.g. allosteric inhibitors. Given the unique
structural characteristics of many enzyme active sites, it seems logical
to suggest that those inhibitors that do bind to enzyme active sites may
more readily exhibit selectivity of target enzyme binding, but this will
require additional study and could well vary from one enzyme to an-
other.
3. Conclusions
DNA polymerase β is one of the enzymes employed to repair DNA
damage. While this is generally beneficial, the enzyme has been im-
plicated in the repair of DNA damage inflicted upon cancer cells treated
with DNA targeted antitumor agents, and is thought to diminish the
therapeutic efficacy of such agents. While inhibitors of DNA polymerase
β have been proposed as adjuvants for antitumor therapy by DNA da-
maging agents, to date the inhibitors identified have lacked the potency
and selectivity required of potential therapeutic agents. The present
study provides a straightforward strategy for identifying Pol β in-
hibitors that interact with the catalytic nucleophile of Lys72 in the lyase
domain of the enzyme. One compound that employs this critical
binding mode has been identified, shown to mediate such interactions
in both rat and human DNA polymerases, and characterized in greater
detail. The ability to express a broad range of enzymes appropriately
modified by coupled in vitro transcription/translation in the presence of
chemically misacylated tRNAs, suggests that the new strategy ex-
emplified in this study could realize broader application for the iden-
tification of inhibitors having specific binding characteristics.
Fig. 6. Evaluation of 3,4-dihydroxybenzoic acid (3) for its ability to engage
Lys72-Nε as part of the rat Pol β-inhibitor binding interface. Lyase 5′-dRP ex-
cision assays and the corresponding histograms showing the relative activities
of Pol β enzymes containing lysine (I), N-methyl-lysine (II) or homolysine (III)
at position 72 in the absence (gray bar) or presence (black bar) of 50 μM 3,4-
dihydroxybenzoic acid (3). WT, wild-type enzyme.
single-stranded DNA recognition rather than by directly affecting the
catalytic event.64
While the binding of compounds 1 and 2 do not seem to involve
interaction with Lys72, two of the compounds studied do seem likely to
interact with the side chain amino group of Lys72. These include 3,4-
dihydroxybenzoic acid (3) and the cis-epoxide of oleic acid (4). Binding
of the latter was further studied by parallel molecular docking (Fig. 9).
The binding pocket of cis-9,10-epoxyoctadecanoic acid (4) is located
between helix 1 and helix 2 of the 8 kDa domain. The compound is
anchored to the protein mainly via hydrophobic interactions and hy-
drogen bonds in a position that favors electrostatic interaction between
the oxygen atom of the epoxy group and Nε of the main catalytic nu-
cleophile Lys72 (helix 4). Two other key residues for the lyase reaction,
Tyr39, known to stabilize the deprotonated form of Lys7265 and Lys35,
which assists in phosphate elimination, also form contacts with the
inhibitor involving C4 and C10 atoms and C16, C18, O2 and O3 atoms,
respectively. In the context of the foregoing analysis, it is worth noting
that anacardic acids containing double bonds have been found to be
more potent Pol β inhibitors than those with fully saturated side
chains66 and sulfolipids having 1–3 cis-double bonds were quite potent
competitive Pol β inhibitors.67
4. Experimental section
4.1. General experimental procedures
A quick change site-directed mutagenesis kit was purchased from
Agilent Technologies. Uracil DNA glycosylase, apurinic endonuclease
and restriction endonucleases SalI, NdeI and BamH1 were obtained
from New England BioLabs. Vector pET28b(+) was obtained from
Novagen. T7 RNA polymerase transcription kits were purchased from
Epicentre Biotechnologies. β-Stigmasterol (1) and 3,4-dihydrox-
ybenzoic acid (3) were purchased from Sigma-Aldrich Chemicals. 3-β-
(Benzoyloxy)stigmastane (2) was prepared from β-stigmasterol (1)
(Scheme S1), while Pol β lyase test compounds 4 and 5 were prepared
from commercially available oleic acid (Scheme S2).
The foregoing analysis involving Pol β analogues modified at re-
sidue 72 provides a facile method for the identification of Pol β lyase
inhibitors that critically bind the primary catalytic residue Lys72 as part
of their protein binding interface; this could potentially afford lead
inhibitors with greater affinity and specificity for Pol β. Plausibly, Pol β
6

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Fig. 7. Evaluation of cis-9,10-epoxyoctadecanoic acid (4) and 9,10-dihydroxyoctadecanoic acid (5) for their ability to engage Lys72-Nε as part of the rat Pol β-
inhibitor binding interface. Lyase 5′-dRP excision assays and the corresponding histograms showing the relative activities of Pol β enzymes containing lysine (I), N-
methyl-lysine (II) or homolysine (III) at position 72 in the absence (gray bar) or presence (black bar) of cis-9,10-epoxyoctadecanoic acid (4) (panel A) or 9,10-
dihydroxyoctadecanoic acid (5) (panel B). WT, wild-type Pol β.
4.2. Biochemical procedures
reaction mixture (1 mL total volume) containing 7.5 mM ATP, CTP, and
UTP, 5 mM GTP, 20 mM GMP, 10 mM dithiothreitol, 20 nM template
DNA, and 100 μL of the T7 RNA polymerase preparation at 42 °C for
12 h. The transcribed tRNA was precipitated from the reaction mixture
by the addition of 3 vol of cold EtOH, then collected by centrifugation.
The sample was diluted 3-fold in 0.1 M NaOAc, pH 5.2, and purified on
a 1 mL DEAE-Sepharose CL-6B column. The column was washed with
1.5 mL of 0.1 M NaOAc, pH 5.2, and then with 1.5-mL portions of the
same buffer containing 0.1 M, 0.2 M, 0.3 M, 0.4 M, 0.5 M, 0.6 M, 0.7 M,
0.8 M, and then 0.9 M NaCl. Each eluate was precipitated with 2-pro-
panol at 0 °C, washed with 70% EtOH, dissolved in 100 μL of RNase-free
water, and analyzed by 8% denaturing PAGE; the RNA was visualized
4.2.1. Construction of E. coli S-30 expression plasmids containing the gene
for rat DNA polymerase β
Expression plasmids denoted pETPOLB (wild-type rat DNA poly-
merase β) and pETPOLB72 (TAG codon at position 72), were con-
structed using vector pET28b(+). The rat Polb gene was excised from
the original ampicillin resistant vector using restriction endonucleases
NdeI and SalI, and inserted into a pre-cut kanamycin resistant [pET28b
(+)] vector, yielding pETPOLB. The hexahistidine fusion peptide was
included in the vector, and coded for a 2-kDa peptide (MGSSHHHHH-
HSSGLVPRGSH) at the N-terminus of the protein. A TAG stop codon
was inserted at position 72 using a mutagenesis kit, yielding the
plasmid pETPOLB72.
by UV shadowing.70 The fractions (0.6–0.8
tRNACUA-COH were pooled and stored at −80 °C.
M NaCl) containing
The analogous human gene was modified in the same fashion.
4.2.3. Preparation of misacylated suppressor tRNACUA-COH
4.2.2. In vitro transcription of suppressor tRNA
Transfer RNA misacylation reactions were carried out in 100 μL
(total volume) of 50 mM Na Hepes, pH 7.5, containing 0.5 mM ATP,
15 mM MgCl2, 50 μg of suppressor tRNACUA-COH, 1.0 A260 unit of
protected aminoacyl-pdCpA (5 to 10-fold molar excess), 20% di-
methylsulfoxide, and 200 units of T4 RNA ligase. After incubation at
37 °C for 60 min, the reaction was quenched by the addition of 0.1 vol
of 3 M NaOAc, pH 4.5, and the tRNA was precipitated with 2.5 vol of
EtOH. The efficiency of ligation was estimated by acidic gel electro-
phoresis (pH 5.2).71 Pentenoyl protected aminoacyl-tRNAs were de-
protected at room temperature for 20 min in 5 mM I2 from a stock
Phe 68,69
Plasmid pYRNA8, encoding the yeast suppressor tRNA
,
was
CUA
digested with restriction endonuclease FokI and then transcribed using
an AmpliScribe T7 RNA polymerase transcription kit. Preparative FokI
digests typically contained 100 μg of plasmid DNA, 20 mM Tris-acetate,
pH 7.9, 10 mM magnesium acetate, 50 mM potassium acetate, 1 mM
dithiothreitol (DTT), and 40 units of FokI in a total volume of 100 μL.
Reaction mixtures were incubated at 37 °C for 4 h. The solution con-
taining the digested plasmid DNA was extracted with phenol–chloro-
form, precipitated with ethanol, and then transcribed using an Ampli-
Scribe T7 transcription kit. Transcription was carried out in a buffered
solution
containing
25
mM
I2
dissolved
in
1:1
7

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Fig. 8. Evaluation of cis-9,10-epoxyoctadecanoic acid (4) and 9,10-dihydroxyoctadecanoic acid (5) for their ability to engage Lys72-Nε as part of the human Pol β-
inhibitor binding interface. Lyase 5′-dRP excision assays and the corresponding histograms showing the relative activities of Pol β enzymes containing lysine (I), N-
methyl-lysine (II) or homolysine (III) at position 72 in the absence (gray bar) or presence (black bar) of cis-9,10-epoxyoctadecanoic acid (4) (panel A) or 9,10-
dihydroxyoctadecanoic acid (5) (panel B). WT, wild-type Pol β.
tetrahydrofuran − H2O.34,35 Following deprotection, the solution was
centrifuged, and the clear supernatant was adjusted to 0.3 M NaOAc,
then treated with 3 vol of EtOH to precipitate the aminoacylated tRNA
and remove trace amounts of I2. The tRNA pellet was washed with 70%
EtOH and dried. The deprotected aminoacyl-tRNA was used im-
mediately in in vitro suppression experiments. The pdCpA derivatives of
N-methyl-lysine and homolysine were prepared as described.36
thiogalactopyranoside, 20 mM ATP and GTP, 5 mM CTP and UTP, and
10 mM cAMP), 100 μM amino acids minus methionine, 50 μM me-
thionine, 40 μCi of [35S]-L-methionine, and 30 μL of S-30 extract.
Suppression reactions (100 μL total volume) contained 25 μg of de-
blocked misacylated suppressor tRNACUA dissolved in 10 μL of diethyl
pyrocarbonate-treated water, and were incubated at 37 °C for 45 min.
For the purpose of analysis by SDS-PAGE, 1 μL of the crude protein
biosythesizing system was mixed with 9 μL of SDS-PAGE loading buffer
(250 mM Tris-HCl, pH 6.8, 1% SDS, 10% glycerol, 0.01% bromophenol
blue, 20 mM β-mercaptoethanol), and loaded on a gel. The gel was run
for 2 h at 5 W, and fixed in a solution containing 50% water, 40%
ethanol, and 10% acetic acid for 10 min. Dried gels were visualized and
analyzed using an Amersham Storm 820 phosphorimager and Mole-
cular Dynamics ImageQuant software.
4.2.4. E. coli S-30 extract preparation
The preparation of the S-30 extract from E. coli BL21-DE3 was
carried out according to a published procedure.72 E. coli cells were
grown in an incubator (37 °C) in eight 2-L flasks until the optical
density at 450 nm was 1.7. IPTG (isopropyl β-D-1-thiogalactopyrano-
side) was then added to induce the synthesis of T7 RNA polymerase.
The cells were collected by centrifugation, washed three times, and
lysed in a chilled French cell press at 8,400 p.s.i. The S-30 extract was
centrifuged, and then stored at –80 °C until use (the total yield was
20 mL).
4.2.6. Purification of enzymes by Ni-NTA affinity chromatography
The crude S-30 reaction mixtures containing the radiolabeled pro-
teins of interest were applied to a 200 μL Ni-NTA agarose column that
had been equilibrated with 50 mM Na phosphate, pH 8.0, containing
300 mM NaCl, 20 mM imidazole, and 100 μg/mL of bovine serum al-
bumin. The column was washed with five 100-μL portions of the same
buffer; the protein was then eluted from the column by washing with
five 100-μL portions of 50 mM Na phosphate, pH 8.0, containing
300 mM NaCl, 250 mM imidazole, and 10 μg/mL of bovine serum al-
bumin. The samples were diluted and stored in polymerase β storage
buffer (50 mM Tris-HCl, pH 7.5, containing 0.1 mM EDTA, 1 mM DTT,
and 50% glycerol) at –20 °C.
4.2.5. Translation in S-30 E. coli extract coupled transcription/translation
protein biosynthesizing system
Translation reactions carried out in the presence of S-30 extract
were performed as described by Lesley et al.73 except with the use of
plasmid DNA instead of mRNA transcripts. In a typical experiment,
proteins were synthesized in a reaction mixture (100 μL total volume)
that contained 10 μg of plasmid DNA dissolved in diethyl pyr-
ocarbonate-treated water, 40 μL of premix (35 mM Tris acetate, pH 7.0,
containing 190 mM potassium glutamate, 30 mM ammonium acetate,
2 mM dithiothreitol, 11 mM magnesium acetate, 20 mM phospho(enol)
pyruvate, 0.8 mg/mL of E. coli tRNA, 0.8 mM isopropyl β-D-1-
4.2.7. 3′-[32P]-End labeling of an oligonucleotide substrate
A 34-nucleotide (nt) DNA (5′-CTGCAGCTGATGCGCUGTACGGATC
8

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
Fig. 10. Computational docking of β-stigmasterol (1) to the structure of the
8 kDa domain of rat Pol β (pdb 1dk3). Residues constituting the binding in-
terface are identified by name and colored light brown. The primary catalytic
residue Lys72 (not part of the binding interface) is shown in coral blue.
preparative 20% polyacrylamide gel and then annealed to the com-
plementary DNA oligonucleotide (5′-GTACCCGGGGATCCGTACGGCG
CATCAGCTGCAG-3′) to form a duplex. This was accomplished by
heating the solution at 70 °C for 3 min, followed by slow cooling to
25 °C.
4.2.8. Apurinic site preparation
The reaction mixture (1 mL) contained 43 nM 3′-[32P]-end labeled
DNA duplex having a single deoxyuridine in 10 mM K Hepes buffer, pH
7.4, containing 50 mM KCl, 5 mM MgCl2, 10 mg/mL of bovine serum
albumin, 3 units of AP endonuclease, and 2.4 units of uracil-DNA gly-
cosylase. The reaction mixture was incubated at 25 °C for 10 min, and
the product was used directly in the dRP excision assay.
4.2.9. dRP excision assay
The reaction mixture (48 μL total volume) contained 40 nM 32P-end
labeled duplex DNA containing an apurinic site, 25 nM rat DNA poly-
merase β, and pure test compounds dissolved in DMSO. The final DMSO
concentration did not exceed 2%. The reaction mixture was incubated
at 25 °C for up to 5 min, treated with 50 mM NaBH4, and incubated at
25 °C for 10 min. The samples were then heated to 75 °C for 10 min to
destroy excess NaBH4. The reaction products were separated on a se-
Fig. 9. Computational docking of (A) cis-9S,10R-epoxyoctadecanoic acid (4)
and (C) 9R,10R-dihydroxyoctadecanoic acid (5) to the structure of the 8 kDa
domain of rat Pol β (pdb 1dk3). Residues constituting the binding interface are
identified by name and colored in blue in panel A and yellow in panel C, re-
spectively. In panel A, the primary catalytic residue Lys72 is shown in coral
blue and the lysine Nε – inhibitor epoxide O distance (3.7 Å) is shown. An
expanded version of part of panel A showing the bound inhibitor 4 and the
nearest active site residues is shown in the top portion of panel B. The lower
portion of panel B illustrates the binding of (9R,10S)-4; the lysine Nε – inhibitor
epoxide O distance was 5.1 Å. In panel C, inhibitor 5 binding clearly does not
involve Lys72.
quencing
20%
denaturing
polyacrylamide
gel
(40 cm × 40 cm × 0.4 mm) at 50 W for 5 h. After overnight exposure,
the storage phosphor screen was scanned on an Amersham Storm 820
phosphorimager. The data was analyzed using Molecular Dynamics
ImageQuant software.
4.2.10. Inhibition of dRP excision
The fraction of the product formed was calculated by dividing the
band intensity of the product band by the band intensities of the pro-
duct and substrate bands. Enzyme activity of Pol β (the amount of
product generated for 3 min under the conditions of the assay) was used
as reference (100%) to calculate the effect of the tested compounds on
the lyase activity.
CCCGGGTAC-3′) containing an internal deoxyuridine at a single posi-
tion on one strand was 3′-[32P]-end labeled using terminal deoxy-
transferase and [α-32P]ddATP. The product was purified on
a
9

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
4.3. Computational methods
flap endonuclease 1 enzymatic specificities sustain DNA synthesis for long patch base
excision repair. J Biol Chem. 2005;280:3665–3674.
11. Sung JS, DeMott MS, Demple B. Long-patch base excision DNA repair of 2-deoxyr-
ibonolactone prevents the formation of DNA-protein cross-links with DNA poly-
merase β. J Biol Chem. 2005;280:39095–39103.
4.3.1. Docking calculations
Docking calculations were carried out using Virtua Drug Ltd
DockingServer.74 The MMFF94 force field75 was used for energy
minimization of ligand molecules uploaded from PubChem. Gasteiger
partial charges were added to the ligand atoms. Non-polar hydrogen
atoms were merged, and rotatable bonds were defined. Docking cal-
culations were carried out on the 1dk3 protein model of the lyase do-
main of rat DNA polymerase β.76 Essential hydrogen atoms, Kollman
united atom type charges and solvation parameters were added with
the aid of AutoDock tools.77 Affinity (grid) maps of 20 × 20 × 20 Å
grid points and 0.375 Å spacing were generated using the Autogrid
program.76 AutoDock parameter set- and distance-dependent dielectric
functions were used in the calculation of the van der Waals and the
electrostatic terms, respectively. Docking simulations were performed
using the Lamarckian genetic algorithm and the Solis & Wets local
search method.78 Initial position, orientation, and torsions of the ligand
molecules were set randomly. Each docking experiment was derived
from 100 different runs that were set to terminate after a maximum of
2,500,000 energy evaluations. The population size was set to 150.
During the search, a translational step of 0.2 Å, and quaternion and
torsion steps of 5 were applied.
12. Starcevic D, Dalal S, Sweasy JB. Is there a link between DNA polymerase β and
cancer? Cell Cycle. 2004;3:998–1001.
13. Copani A, Hoozemans JJ, Caraci F, et al. DNA Polymerase-β is expressed early in
neurons of Alzheimer’s disease brain and is loaded into DNA replication forks in
neurons challenged with β-amyloid. J Neurosci. 2006;26:10949–10957.
14. Albertella MR, Lau A, O’Connor MJ. The overexpression of specialized DNA poly-
merases in cancer. DNA Repair. 2005;4:583–593.
15. Dong ZM, Zheng NG, Wu JL, Li SK, Wang YL. Difference in expression level and
localization of DNA polymerase beta among human esophageal cancer focus, ad-
jacent and corresponding normal tissues. Dis Esophagus. 2006;19:172–176.
16. Yu J, Mallon MA, Zhang W, et al. DNA Repair pathway profiling and microsatellite
instability in colorectal cancer. Clin Cancer Res. 2006;12:5104–5111.
17. Liu SN, Bai WY, Frye RM, Hou L, Zhang B. Specific up-regulation of DNA polymerase
by human papillomavirus 16. Clin Med Sci J. 2008;23:108–112.
18. Faumont N, Le Clorennec C, Teira P, et al. Regulation of DNA polymerase β by the
LMP1 oncoprotein of EBV through the nuclear factor-κB pathway. Cancer Res.
2009;69:5177–5185.
19. Pillaire MJ, Betous R, Conti C, et al. Upregulation of error-prone DNA polymerases β
and κ slows down fork progression without activating the replication checkpoint. Cell
Cycle. 2007;6:471–477.
20. Bergoglio V, Pillaire MJ, Lacroix-Triki M, et al. Deregulated DNA polymerase β in-
duces chromosome instability and tumorigenesis. Cancer Res. 2002;62:3511–3514.
21. Canitrot Y, Capp JP, Puget N, et al. DNA polymerase β overexpression stimulates the
Rad51-dependent homologous recombination in mammalian cells. Nucleic Acids Res.
2004;32:5104–5112.
The server performance was tested by docking pamoic acid, a
known DNA polymerase β inhibitor whose site of binding to the 8 kDa
lyase domain has already been identified. Our results (Supporting
Information, vide infra) are in accordance with the literature data.64
22. Copani A, Sortino MA, Caricasole A, et al. Erratic expression of DNA polymerases by
β-amyloid causes neuronal death. FASEB J. 2002;16:2006–2008.
23. Canitrot Y, Cazaux C, Fréche M, et al. Overexpression of DNA polymerase β in cell
results in a mutator phenotype and a decreased sensitivity to anticancer drugs. Proc
Natl Acad Sci USA. 1998;95:12586–12590.
24. Gao Z, Maloney DJ, Dedkova LM, Hecht SM. Inhibitors of DNA polymerase β:
Activity and mechanism. Bioorg Med Chem. 2008;16:4331–4340.
25. Vaisman A, Masutani C, Hanaoka F, Chaney SG. Efficient translesion replication past
oxaliplatin and cisplatin GpG adducts by human DNA polymerase η. Biochemistry.
2000;39:4575–4580.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
26. Vaisman A, Warren MW, Chaney SG. The effect of DNA structure on the catalytic
efficiency and fidelity of human DNA polymerase β on templates with platinum-DNA
adducts. J Biol Chem. 2001;276:18999–19005.
27. Raaphorst GP, Cybulski SE, Sobol R, Ng CE. The response of human breast tumor cell
lines with altered polymerase-β levels to cisplatin and radiation. Anticancer Res.
2001;21:2079–2083.
Declaration of Competing Interest
28. Raaphorst GP, Ng CE, Yang DP. Comparison of response to radiation, hyperthermia
and cisplatin in parental and polymerase β knockout cells. Int J Hyperthermia.
2002;18:33–39.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
29. Boudsocq F, Benaim P, Canitrot Y, et al. Modulation of cellular response to cisplatin
by a novel inhibitor of DNA polymerase β. Mol Pharmacol. 2005;67:1485–1492.
30. Trivedi RN, Wang XH, Jelezcova E, Goellner EM, Tanh J, Sobol RW. Human methyl
purine DNA glycosylase and DNA polymerase β expression collectively predict sen-
sitivity to temozolomide. Mol Pharmacol. 2008;72:505–516.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmc.2020.115642.
31. Yang J, Parsons J, Nicolay NH, et al. Cells deficient in the base excision repair
protein, DNA polymerase beta, are hypersensitive to oxaliplatin chemotherapy.
Oncogene. 2010;29:463–468.
32. Barakat KH, Gajewski MM, Tuszynski JA. DNA polymerase beta (pol β) inhibitors: A
comprehensive overview. Drug Discov Today. 2012;17:913–920.
References
33. Jain R, Aggarwal AK, Rechkoblit O. Eukaryotic DNA polymerases. Curr Opin Struct
Biol. 2018;53:77–87.
1. Beard WA, Wilson SH. Purification and domain-mapping of mammalian DNA poly-
merase β. Methods Enzymol. 1995;262:98–107.
34. Lodder M, Golovine S, Hecht SM. A chemical deprotection strategy for the elabora-
tion of misacylated transfer RNA’s. J Org Chem. 1997;62:778–779.
35. Lodder M, Golovine S, Laikhter AL, Karginov VA, Hecht SM. Misacylated transfer
RNAs having a chemically removable protecting group. J Org Chem.
1998;63:794–803.
2. Kumar A, Widen SG, Williams KR, Kedar P, Karpel R. Wilson SH. Studies of the
domain structure of mammalian DNA polymerase β. Identification of a discrete
template binding domain. J Biol Chem. 1990;265:2124–2131.
3. Prasad R, Beard WA, Wilson SH. Studies of gapped DNA substrate binding by
mammalian DNA polymerase β. Dependence on 5'-phosphate group. J Biol Chem.
1994;269:18096–18101.
36. Daskalova SM, Bhattacharya C, Dedkova LM, Hecht SM. Probing the flexibility of the
catalytic nucleophile in the lyase catalytic pocket of human DNA polymerase β with
unnatural lysine analogues. Biochemistry. 2017;56:500–513.
4. Matsumoto Y, Kim K. Excision of deoxyribose phosphate residues by DNA polymerase
β during DNA repair. DNA Repair. 1995;269:699–702.
37. García-Díaz M, Bebenek K, Kunkel TA, Blanco L. Identification of an intrinsic 5′-
deoxyribose-5-phosphate lyase activity in human DNA polymerase λ. A possible role
in base excision repair. J Biol Chem. 2001;276:34659–34663.
5. Pelletier H, Sawaya MR, Wolfe W, Wilson SH, Kraut J. Crystal structures of human
DNA polymerase β complexed with DNA: implications for catalytic mechanism,
processivity, and fidelity. Biochemistry. 1996;35:12742–12761.
6. Chagovetz AM, Sweasy JB, Preston BD. Increased activity and fidelity of DNA
polymerase β on single-nucleotide gapped DNA. J Biol Chem.
38. Lee JW, Blanco L, Zhou T, et al. Implication of DNA Polymerase λ in alignment-based
gap filling for nonhomologous DNA end joining in human nuclear extracts. J Biol
Chem. 2004;279:805–811.
39. Braithwaite EK, Prasad R, Shock DD, Hou EW, Beard WA, Wilson SH. DNA
Polymerase λ mediates a back-up base excision repair activity in extracts of mouse
embryonic fibroblasts. J Biol Chem. 2005;280:18469–18475.
1997;272:27501–27504.
7. Wilson SH. Mammalian base excision repair and DNA polymerase beta. Mut Res.
1998;407:203–215.
40. Prasad R, Longley MJ, Sharief FS, Hou EW, Copeland WC, Wilson SH. Human DNA
polymerase θ possesses 5′-dRP lyase activity and functions in single-nucleotide base
excision repair in vitro. Nucleic Acids Res. 2009;37:1868–1877.
8. Podlutsky AJ, Dianova II, Wilson SH, Bohr VA, Dianov GL. DNA synthesis and dRPase
activities of polymerase β are both essential for single-nucleotide patch base excision
repair in mammalian cell extracts. Biochemistry. 2001;40:809–813.
9. Podlutsky AJ, Dianova II, Podust VN, Bohr VA, Dianov GL. Human DNA polymerase β
initiates DNA synthesis during long-patch repair of reduced AP sites in DNA. EMBO J.
2001;20:1477–1482.
41. Beagan K, McVey M. Linking DNA polymerase theta structure and function in health
and disease. Cell Mol Life Sci. 2016;73:603–615.
42. Prasad R, Bebenek K, Ho E, et al. Localization of the deoxyribose phosphate lyase
active site in human DNA Polymerase ι by controlled proteolysis. J Biol Chem.
2003;278:29649–29654.
10. Liu Y, Beard WA, Shock DD, Prasad R, Hou EW, Wilson SH. DNA polymerase β and
10

S.M. Daskalova, et al.
Bioorganic&MedicinalChemistry28(2020)115642
43. Tissier A, Frank EG, McDonald JP, Iwai S, Hanaoka F, Woodgate R. Misinsertion and
bypass of thymine-thymine dimers by human DNA polymerase ɩ. EMBO J.
2000;19:5259–5266.
inhibitors of DNA polymerase β by NMR chemical shift mapping. J Biol Chem.
2004;279:39736–39744.
62. Barakat K, Tuszynski J. Relaxed complex scheme suggests novel inhibitors for the
lyase activity of DNA polymerase beta. J Mol Graph Model. 2011;29:702–716.
63. Zhao P, Sharir H, Kapur A, et al. Targeting of the orphan receptor GPR35 by pamoic
acid: A potent activator of extracellular signal-regulated kinase and β-arrestin2 with
antinociceptive activity. Mol Pharmacol. 2010;78:560–568.
44. Pinz KG, Bogenhagen DF. Characterization of a catalytically slow AP lyase activity in
DNA polymerase γ and other family A DNA polymerases. J Biol Chem.
2000;275:12509–12514.
45. Graziewicz MA, Longley MJ, Copeland WC. DNA polymerase γ in mitochondrial DNA
replication and repair. Chem Rev. 2006;106:383–405.
64. Hazan C, Boudsocq F, Gervais V, Saurel O, et al. Structural insights on the pamoic
acid and the 8 kDa domain of DNA polymerase beta complex: Towards the design of
higher-affinity inhibitors. BMC Struct Biol. 2008;8(1):22.
46. García-Díaz M, Domínguez O, López-Fernández LA, et al. DNA polymerase lambda
(Pol λ), a novel eukaryotic DNA polymerase with a potential role in meiosis. J Mol
Biol. 2000;301:851–867.
65. Prasad R, Batra VK, Yang XP, et al. Structural insight into the DNA polymerase β
deoxyribose phosphate lyase mechanism. DNA Repair. 2005;4:1347–1357.
66. Chen J, Zhang Y-H, Wang L-K, Sucheck SJ, Snow AM, Hecht SM. Inhibitors of DNA
polymerase β from Schoepfia californica. Chem Commun. 1998:2769–2770.
67. Mizushina Y, Tanaka N, Yagi H, et al. Fatty acids selectively inhibit eukaryotic DNA
polymerase activities in vitro. Biochim Biophys Acta. 1996;1308:256–262.
68. RajBhandary UL, Chang SH, Stuart A, Faulkner RD, Hoskinson RM, Khorana HG.
Studies on polynucleotides, LXVIII. The primary structure of yeast phenylalanine
transfer RNA. Proc Natl Acad Sci USA. 1967;57:751–758.
47. Cruet-Hennequart S, Gallagher K, Sokòl AM, Villalan S, Prendergast AM, Carty MP.
DNA polymerase η, a key protein in translesion synthesis in human cells. Subcell
Biochem. 2010;50:189–209.
48. Liu L. DNA topoisomerase poisons as antitumor drugs. Annu Rev Biochem.
1989;58:351–375.
49. Froelich-Ammon SJ, Osheroff N. Topoisomerase poisons: Harnessing the dark side of
enzyme mechanism. J Biol Chem. 1995;270:21429–21432.
50. Hecht SM. Inhibitors of the lyase activity of DNA polymerase β. Pharmaceut Biol.
2003;41:68–77.
69. Noren CJ, Anthony-Cahill SJ, Suich DJ, Noren KA, Griffith MC, Schultz PG. In vitro
suppression of an amber mutation by a chemically aminoacylated transfer RNA
prepared by runoff transcription. Nucleic Acids Res. 1990;18:83–88.
70. Hassur SM, Whitlock Jr HW. UV shadowing–a new and convenient method for the
location of ultraviolet-absorbing species in polyacrylamide gels. Anal Biochem.
1974;59:162–164.
51. Cao S, Gao Z, Thomas SJ, Hecht SM, Lazo JS, Kingston DG. Marine sesquiterpenoids
that inhibit the lyase activity of DNA polymerase β. J Nat Prod. 2004;67:1716–1718.
52. Chaturvedula VS, Zhou BN, Gao Z, Thomas SJ, Hecht SM, Kingston DG. New lupane
triterpenoids from Solidago canadensis that inhibit the lyase activity of DNA poly-
merase β. Bioorg Med Chem. 2004;12:6271–6275.
53. Chaturvedula VS, Gao Z, Jones SH, Feng X, Hecht SM, Kingston DG. A new ursane
triterpene from Monochaetum vulcanicum that inhibits DNA polymerase β lyase. J Nat
Prod. 2004;67:899–901.
71. Varshney U, Lee CP, RajBhandary UL. Direct analysis of aminoacylation levels of
tRNAs in vivo. Application to studying recognition of Escherichia coli initiator tRNA
mutants by glutaminyl-tRNA synthetase. J Biol Chem. 1991;266:24712–24718.
72. Pratt M. Transcription and translation: a practical approach. Hames BD, Higgins SJ
(eds.). Oxford: 1984; 179-209.
54. Chaturvedula VS, Schilling JK, Miller JS, Andriantsiferana R, Rasamison VE,
Kingston DG. New cytotoxic terpenoids from the wood of Vepris punctata from the
Madagascar rainforest. J Nat Prod. 2004;67:895–898.
73. Lesley SA, Brow MA, Burgess RR. Use of in vitro protein synthesis from polymerase
chain reaction-generated templates to study interaction of Escherichia coli tran-
scription factors with core RNA polymerase and for epitope mapping of monoclonal
antibodies. J Biol Chem. 1991;266:2632–2638.
55. Feng X, Gao Z, Li S, Jones SH, Hecht SM. (2004) New neolignans that inhibit DNA
polymerase β lyase. J Nat Prod. 2004;67:964–967.
56. Li SS, Gao Z, Feng X, Jones SH, Hecht SM. (2004) Plant sterols as selective DNA
polymerase β lyase inhibitors and potentiators of bleomycin cytotoxicity. Bioorg Med
Chem. 2004;12:4253–4258.
74. Bikadi Z, Hazai E. Application of the PM6 semi-empirical method to modeling pro-
teins enhances docking accuracy of AutoDock. J Cheminfo. 2009;1:15.
75. Halgren TA. Merck molecular force field. I. Basis, form, scope, parameterization, and
performance of MMFF94. J Comp Chem. 1996;17:490–519.
57. Li SS, Gao Z, Feng X, Hecht SM. Biscoumarin derivatives from Edgeworthia gardneri
that inhibit the lyase activity of DNA polymerase β. J Nat Prod. 2004;67:1608–1610.
58. Arian D, Hedayati M, Zhou H, et al. Irreversible inhibition of DNA polymerase β by
small molecule mimics of a DNA lesion. J Am Chem Soc. 2014;136:3176–3183.
59. Strittmatter T, Brockmann A, Pott M, Hantusch A, Brunner T, Marx A. Expanding the
scope of human DNA polymerase λ and β inhibitors. ACS Chem Biol.
2014;9:282–290.
76. Maciejewski MW, Liu D, Prasad R, Wilson SH, Mullen GP. Backbone dynamics and
refined solution structure of the N-terminal domain of DNA polymerase β.
Correlation with DNA binding and dRP lyase activity. J Mol Biol. 2000;296:229–253.
77. Morris GM, Goodsell DS, Halliday RS, et al. Automated docking using a Lamarckian
genetic algorithm and an empirical binding free energy function. J Comp Chem.
1998;19:1639–1662.
60. Gowda AS, Suo Z, Spratt TE. Honokiol inhibits DNA polyerases β and λ and increases
bleomycin sensitivity of human cancer cells. Chem Res Toxicol. 2017;30:715–725.
61. Hu HY, Horton JK, Gryk MR, et al. Identification of small molecule synthetic
78. Solis FJ, Wets RJB. Minimization by random search techniques. Math Oper Res.
1981;6:19–30.
11