Identification and Optimization of Anthranilic Acid Based Inhibitors of Replication Protein A

Article abstract of DOI:10.1002/cmdc.201500479

Replication protein A (RPA) is an essential single-stranded DNA (ssDNA)-binding protein that initiates the DNA damage response pathway through protein-protein interactions (PPIs) mediated by its 70N domain. The identification and use of chemical probes that can specifically disrupt these interactions is important for validating RPA as a cancer target. A high-throughput screen (HTS) to identify new chemical entities was conducted, and 90 hit compounds were identified. From these initial hits, an anthranilic acid based series was optimized by using a structure-guided iterative medicinal chemistry approach to yield a cell-penetrant compound that binds to RPA70N with an affinity of 812 nm. This compound, 2-(3- (N-(3,4-dichlorophenyl)sulfamoyl)-4-methylbenzamido)benzoic acid (20 c), is capable of inhibiting PPIs mediated by this domain.

Full text of DOI:10.1002/cmdc.201500479

DOI: 10.1002/cmdc.201500479
Full Papers
Very Important Paper
Identification and Optimization of Anthranilic Acid Based
Inhibitors of Replication Protein A
James D. Patrone,[a, d] Nicholas F. Pelz,[a] Brittney S. Bates,[a] Elaine M. Souza-Fagundes,[a]
Bhavatarini Vangamudi,[a] Demarco V. Camper,[a] Alexey G. Kuznetsov,[a] Carrie F. Browning,[a]
Michael D. Feldkamp,[a] Andreas O. Frank,[a] Benjamin A. Gilston,[a] Edward T. Olejniczak,[a]
Olivia W. Rossanese,[a] Alex G. Waterson,[b, c] Walter J. Chazin,[a, c] and Stephen W. Fesik*[a, b, c]
Replication protein A (RPA) is an essential single-stranded DNA
(ssDNA)-binding protein that initiates the DNA damage re-
sponse pathway through protein–protein interactions (PPIs)
mediated by its 70N domain. The identification and use of
chemical probes that can specifically disrupt these interactions
is important for validating RPA as a cancer target. A high-
throughput screen (HTS) to identify new chemical entities was
conducted, and 90 hit compounds were identified. From these
initial hits, an anthranilic acid based series was optimized
by using a structure-guided iterative medicinal chemistry
approach to yield a cell-penetrant compound that binds
to RPA70N with an affinity of 812 nm. This compound, 2-(3-
(N-(3,4-dichlorophenyl)sulfamoyl)-4-methylbenzamido)benzoic
acid (20c), is capable of inhibiting PPIs mediated by this
domain.
Introduction
Replication protein A (RPA), the primary single-strand DNA
(ssDNA)-binding protein in eukaryotes, is essential for DNA rep-
lication, damage response, and repair. In addition to binding
to and protecting ssDNA from degradation, RPA recruits part-
ner proteins involved in these processes. RPA is composed of
three subunits, each bearing oligonucleotide/oligosaccharide-
binding (OB)-fold domains.[1,2] The N-terminal domain of the
70 kDa subunit (RPA70N) is one of two key sites that mediates
the recruitment of partner proteins.[3] This domain is particular-
ly important for the recruitment of DNA damage response pro-
teins to sites of DNA damage via interaction with the RPA70N
central basic cleft.[3–6]
possible that specific inhibition of this RPA function may repre-
sent an attractive pathway for therapeutic intervention in
cancer. We and others are pursuing inhibitors of the RPA70N-
mediated PPIs that do not interfere with the ability of RPA to
bind to and protect ssDNA, as these would allow for further
exploration of the role of RPA in checkpoint signaling, enable
studies to confirm the therapeutic potential of RPA inhibition,
and serve as a potential starting point for new cancer drugs.
Based on this unique opportunity for small-molecule inhibi-
tors of RPA as potential cancer therapeutics, research on RPA
inhibitors has intensified over the last several years. Turchi and
colleagues have identified dihydropyrazole 1 (Figure 1), which
binds to a DNA-binding domain of RPA and disrupts its interac-
tion with ssDNA.[7,8] Oakley and colleagues identified fumaropi-
maric acid (2, Figure 1), which was shown to disrupt both
Based on the key role of RPA70N-mediated protein–protein
interactions (PPIs) in initiating the DNA damage response, it is
[a] Dr. J. D. Patrone, N. F. Pelz, B. S. Bates, Prof. E. M. Souza-Fagundes,
B. Vangamudi, D. V. Camper, Dr. A. G. Kuznetsov, C. F. Browning,
Dr. M. D. Feldkamp, Dr. A. O. Frank, B. A. Gilston, Prof. E. T. Olejniczak,
Prof. O. W. Rossanese, Prof. W. J. Chazin, Prof. S. W. Fesik
Department of Biochemistry
Vanderbilt University, Nashville, TN 37232 (USA)
E-mail: stephen.fesik@vanderbilt.edu
[b] Prof. A. G. Waterson, Prof. S. W. Fesik
Department of Pharmacology
Vanderbilt University, Nashville, TN 37232 (USA)
[c] Prof. A. G. Waterson, Prof. W. J. Chazin, Prof. S. W. Fesik
Department of Chemistry, Vanderbilt University, Nashville, TN 37232 (USA)
[d] Dr. J. D. Patrone
Current address: Department of Chemistry, Rollins College, 1000 Holt
Avenue, Winter Park, FL 32789 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500479.
This article is part of a Special Issue on Protein–Protein Interactions. To
view the complete issue, visit: http://onlinelibrary.wiley.com/doi/10.1002/
cmdc.v11.8/issuetoc.
Figure 1. Previously reported RPA PPI inhibitors.
ChemMedChem 2016, 11, 893 – 899
893
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
RPA70N–Rad9 and RPA70N–p53 interactions.[9,10] The Oakley re-
search group has also reported on HAMNO (3, Figure 1), which
was shown to inhibit RPA binding to RAD9 and cause in-
creased replicative stress and cytotoxicity in cancer cells and
was shown to slow the progression of squamous cell carcino-
ma in a xenograft model.[11]
We previously reported the results of an NMR-based frag-
ment screen to identify novel molecules that bind to RPA70N.
This screen revealed several distinct chemotypes of fragments
that bind to the domain. Remarkably, this single screen identi-
fied two distinct binding locations in the basic cleft of RPA70N
(Site-1 and Site-2) which can be independently and simultane-
ously occupied by two different compounds.[12,13] From these
results, we also described the results of two optimization cam-
paigns. Initially a fragment-merging strategy was used, result-
ing in triazole 4 (Figure 1), which bound to only one site in the
basic cleft.[12] We also described the results of a fragment-link-
ing strategy to generate compounds that span the entire cleft
and incorporate features of two distinct fragment hits (5,
Figure 1).[13] Herein we describe a different class of molecules
that was identified using a high-throughput screen (HTS) and
further optimized using iterative medicinal chemistry and
structure-based design.
Results and Discussion
Figure 2. Selected HTS hit compounds with Kd and ligand efficiency (LE)
values indicated.
Using a previously reported fluorescence polarization anisotro-
py (FPA) screening assay, 90000 compounds from the Vander-
bilt collection were screened at a single concentration of
30 mm for their ability to disrupt the binding of a fluorescently
labeled ATRIP-derived probe to RPA70N.[14] This screen identi-
fied 674 compounds that displaced >10% of the probe from
RPA70N at this concentration. These initial hits were further fil-
tered to remove compounds that exhibited fluorescence inter-
ference and were prioritized for follow-up on the basis of the
lack of potentially reactive chemical functionality and concord-
ance with commonly accepted measures of drug-likeness.[15,16]
After this analysis, concentration–response curves were collect-
ed for 90 compounds to determine IC50 values, from which Kd
values were calculated. Of these 90 compounds, 52 were iden-
tified with a Kd value <100 mm. Several of the most potent hits
are depicted in Figure 2.[3–6]
To guide the optimization of 11, a co-crystal structure of
compound 11 in complex with RPA70N was obtained
(Figure 3). The binding mode of compound 11 shares several
important contacts with the binding mode of the p53 peptide
and our previously reported molecules. The 4-bromophenyl
portion of the molecule occupies the hydrophobic Site-
1 pocket (Figure 3A), but lies flat against the surface of
RPA70N and sits in a much more shallow position than com-
pounds 4 and 5 (Figure 3C). The sulfonamide group of the
molecule appears to establish the proper geometry necessary
to orient the 4-bromophenyl into Site-1. The middle phenyl
ring occupies the center of the RPA70N cleft and overlays well
with the indole moiety of the tryptophan of the p53 peptide[18]
(Figure 3B) or the phenylalanine of a previously reported
ATRIP-derived peptide.[19] The carboxylic acid of the anthranilic
acid portion of the molecule engages in a charge–charge inter-
action with Arg41 of RPA70N, a common interaction amongst
our fragments and linked small-molecule inhibitors of RPA70N.
In addition, compound 11 also makes a unique hydrogen
bond interaction to Asn85 of RPA70N using the carbonyl
oxygen atom of the amide bond. We hypothesized the amide
in this molecule to be important owing to both this interaction
with RPA and its ability to form an internal hydrogen bond
with the anthranilic acid of the molecule, thus maintaining the
planarity of the molecule in its binding pose.
Compound 6, with the highest ligand efficiency (LE)
amongst the hit set, was briefly investigated. The results from
this work were reported previously.[17] Because of the high lipo-
philicity of the series and generally flat SAR, further work on
the series was halted. Notably, Turchi and co-workers previous-
ly described a series of inhibitors of the interaction between
RPA and ssDNA with a chemical structure similar to that of
compound 9.[17] Compounds 7 and 8 were of relatively low in-
terest. Compounds 10 and 11 are similar and together form an
anthranilic acid based series. SAR evident in the HTS hit set in-
dicated the nitro group to be essential for binding of ether-
based exemplars such as 10. Because of this, as well as the rea-
sonably favorable combination of potency, LE, and the pros-
pect of a modular synthetic route, we focused follow-up efforts
on sulfonamide variants such as compound 11.
Based on the binding mode of hit molecule 11 and our pre-
vious knowledge of small molecules binding to RPA70N, we
devised a strategy to improve potency by optimizing the hy-
drophobic interactions of each of the phenyl rings while main-
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
894
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 3. A) Compound 11 in complex with RPA70N. B) Compound 11 in complex with RPA70N with p53 peptide[18] superimposed. C) Compound 11 in com-
plex with RPA70N with compounds 4 and 5 superimposed. D) SAR strategy for compound 11.
taining the hydrophilic interactions of the amide and carboxyl-
ic acid of the molecule. The first goal was to optimize the
phenyl sulfonamide portion of the molecule for binding to the
hydrophobic pocket of Site-1. An initial compound library con-
taining various phenyl substituents and phenyl replacements
was constructed by using a combination of chemical synthesis
and analogue purchases. Despite the majority of the analogues
being less potent than the original hit, this library provided im-
portant SAR insight (Table 1).
To explore the SAR around the phenyl ring of the anthranilic
acid portion of the molecule, a library of analogues was syn-
thesized with varying R2 substituents, while R1 was fixed as
either 3,4-dichloro, 3-chloro, or 4-bromo. From this library, sev-
eral clear SAR tends emerged (Table 2). Halogen R2 options
were more beneficial at the 4-position than at the 5-position,
leading to an improvement of two- to tenfold. This observa-
tion can be rationalized from the co-crystal structure, in which
one can envision the 5-position substituent clashing with the
lip of the cleft, whereas the 4-position substituent is oriented
toward a hydrophobic gap. A 5-chloro substitution at R2 con-
sistently led to poor physicochemical properties, such as limit-
ed solubility, as evidenced by precipitation under the assay
conditions.
The most effective option at R2 for all three different R1 sub-
stituents was replacement of the anthranilic acid phenyl ring
with a naphthyl moiety (11 l, 18l, and 20l). These analogues
displayed binding affinities of 1–4 mm. Based on the co-crystal
structure of 11, the naphthyl substitution most likely occupies
the hydrophobic space adjacent to both the 4- and 5-posi-
tions. The analogue with the best binding affinity (20c), how-
ever, contained a 3,4-dichloro R1 substitution and a 4-bromo R2
substitution. This analogue was slightly superior to the R2 =
naphthyl analogue 20l and had a more attractive LE (0.27
compared to 0.24 for 20l). Furthermore, compound 20c repre-
sents the best binding affinity yet observed for a molecule
with only one acidic moiety.
Analogues bearing 3- or 4-chloro (18, 19) were found to be
equipotent with 11, whereas non-halogen substituents such as
3-methyl, 4-methyl, or 4-methoxy (13, 14, or 15) were five- to
eightfold less potent. In concordance with previously de-
scribed SAR, analogue 20 (3,4-dichloro) displayed the best
binding affinity of the initial set, showing a fourfold improve-
ment over 11. Both of the chlorine atoms are necessary, as re-
placing either or both with a methyl group (21–23) decreases
binding affinity by six- to 20-fold relative to compound 20.
Methylation of the sulfonamide is not tolerated, as all methy-
lated analogues were two- to fivefold less potent than the des-
methyl analogues (data not shown). Replacement of the
phenyl ring with saturated ring systems (compounds 27–32,
34, and 35) in attempts to increase the hydrophobic interac-
tions and increase the sp3 character of the molecule, were un-
successful, as all analogues were two- to tenfold lower in bind-
ing affinity to RPA70N. Surprisingly, the 3,4-dichloro-substituted
biphenyl compound 25 was found to have an affinity (Kd) of
4 mm, despite its increased size within Site-1. However, deriva-
tives of this molecule were not pursued further due to poor LE
(0.20) and cLogP (6.08), as well as potential solubility limita-
tions.
The final strategy to optimize compound 11 was exploration
of several substituents at R3 on the middle phenyl ring. Howev-
er, several planned analogues (R3 =chloro, bromo, or methoxy,
for example) were synthetically intractable, as intermediates re-
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
895
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Chemistry
Table 1. Structure–activity relationships in Site-1.
The synthesis of the anthranilic acid based inhibitors 11–36
used a modular route, allowing for the introduction of diversity
at each step and only one chromatographic purification.[20] The
synthesis begins with an aromatic sulfonylation, upon treating
a para-substituted benzoic acid with chlorosulfonic acid. The
carboxylic acid (compounds 40–42) is converted into an acid
chloride, and the methyl ester of the appropriate anthranilic
acid is added to afford sulfonyl chlorides in >90% yield
(Scheme 1). After a water workup, the final substituted phenyl
ring is added to the sulfonyl chloride by the addition of the
appropriate substituted aniline. This is followed by saponifica-
tion of the methyl ester to yield the desired analogue.
Compd
R1
Kd [mm][a]
LE[c]
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4-bromo
H
3-methyl
4-methyl
4-methoxy
4-ethyl
4-isopropyl
3-chloro
4-chloro
3,4-dichloro
3,4-dimethyl
30Æ6
156Æ11
165Æ34
196
0.21
0.18
0.18
0.17
0.16
0.18
0.17
0.21
0.20
0.23
0.17
0.19
0.18
0.17
0.20
0.17
0.18
0.18
n.c.[d]
0.19
0.20
0.19
n.c.[d]
0.18
0.17
n.c.[d]
234Æ3
95Æ6.5
76Æ5.5
29Æ2
44Æ0
7Æ3
150Æ11
43Æ3.5
58Æ11
83Æ8
3-chloro, 4-methyl
3-methyl, 4-chloro
2-naphthyl[b]
3’,4’-dichloro-[1,1’-biphenyl]-3-amine[b]
indane[b]
4Æ0.5
106Æ5.5
>250
cyclopentyl[b]
cyclohexyl[b]
193
>250
110 Æ4
81Æ10
126Æ1
>250
208
222Æ28
>250
4-aminotetrahydropyran[b]
cycloheptyl[b]
trans-4-methylcyclohexyl[b]
cyclohexylmethyl[b]
benzyl[b]
azepane[b]
octahydrocyclopenta[c]pyrrole[b]
isoindoline[b]
Scheme 1. General synthesis of anthranilic acid based RPA inhibitors. Re-
agents and conditions: a) chlorosulfonic acid, reflux, 16 h; b) thionyl chloride,
758C, 4 h; c) methyl 2-aminobenzoate–R2, THF, 12 h; d) aniline–R1, toluene,
708C, 12 h; e) 2m LiOH, 558C, 2 h.
[a] Average Kd values (n=2) calculated using the Cheng–Prusoff equation
from IC50 values measured by FPA competition assay. [b] The entire ring
system replaces the phenyl. [c] Ligand efficiency values calculated using
LE=1.4”pKd/HAC (HAC=number of non-hydrogen atoms) using FPA
data. [d] Not calculated.
Conclusions
quired for the synthesis of these molecules were unstable
under the conditions necessary for sulfonamide formation or
saponification. Despite these challenges, several alkyl ana-
logues were obtained (Table 3). The desmethyl analogue 20m
was twofold less potent than compound 20. Further extension
of the methyl to an ethyl (20n) or isopropyl group (20o)
showed marginal improvements in affinity (Kd: 4 and 5 mm, re-
spectively). However, this slight gain in potency for these ana-
logues was offset by a decrease in solubility, with both 20n
and 20o showing some evidence of precipitation at the high-
est concentrations under the assay conditions.
We conducted a high-throughput screen and initial compound
optimization toward the discovery of new and selective chemi-
cal probes to validate inhibition of the protein–protein interac-
tions mediated by RPA70N. Inhibitor 11 was initially identified
as an attractive starting point for structure-based optimization.
Subsequent optimization using an iterative medicinal chemis-
try process and structure-based design principles led to the
discovery of 20c, which binds to RPA70N with an affinity of
812 nm and displays adequate permeability and solubility char-
acteristics for use in cellular studies.
Using a standard fluorescence-based DNA binding assay, we
established that compound 20c does not affect ssDNA binding
to RPA; the Kd value for ssDNA binding to RPA70AB in the
RPA70NAB construct was the same in the absence and pres-
ence of the compound. Thus, 20c appears to bind selectively
to the RPA70N domain. Furthermore, compound 20c was
taken forward for characterization in cellular studies. The mole-
cule was found to possess very high protein binding (99.8%),
but also exhibits high permeability (Papp A!B value of 29.2”
10À6 cmsÀ1 in the Caco-2 line) relative to our previously report-
ed compounds. Studies to define the cellular activity of this
compound are underway and will be reported in due course.
Experimental Section
Chemistry
General methods: All chemicals, reagents, and solvents were used
as purchased from commercial sources, without further purifica-
tion. All NMR spectra were recorded at room temperature on
a
400 MHz Bruker spectrometer with a DRX-400 console,
a 500 MHz Bruker spectrometer with a DRX-500 console, or
a 600 MHz Bruker spectrometer with an AV-II console. 1H NMR
chemical shifts (d) are reported in ppm downfield with the deuter-
ated solvent as the internal standard. Data are reported as follows:
chemical shift, multiplicity (s=singlet, d=doublet, t=triplet, q=
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
896
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. SAR of anthranilic acid ring substituents.
Compd
R1
R2
Kd [mm][a]
LE[c]
Compd
R1
3-Cl
3-Cl
3-Cl
3-Cl
3-Cl
3-Cl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
R2
Kd [mm][a]
LE[c]
20
3,4-diCl
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
4-Br
3-Cl
3-Cl
3-Cl
3-Cl
3-Cl
3-Cl
H
7Æ3
0.23
0.21
n.c.[e]
0.22
0.21
0.20
0.20
0.20
0.19
n.c.[e]
0.22
0.21
0.22
0.23
n.c.[e]
0.24
0.22
0.21
0.20
18g
18h
18i
18j
18k
18l
20a
20b
20c
20d
20e
20 f
20g
20h
20i
5-Me
6-Me
30Æ1
134Æ1
>250
0.20
0.17
n.c.[e]
0.23
0.22
0.22
0.24
n.c.[e]
0.27
0.22
0.22
0.21
0.23
0.19
0.18
0.23
0.23
0.24
11 a
11 b
11 c
11 d
11 e
11 f
11 g
11 h
11 i
11 j
11 k
11 l
18a
18b
18c
18d
18e
18 f
4-Cl
5-Cl
4-Br
5-Br
4-F
5-F
5-Me
6-Me
25Æ2
ppt[d]
4,5-diMe
5-Et
18Æ1.5
20Æ3.5
34Æ3.5
48Æ8
6Æ0.37
5-iPr
7Æ0.25
4,5-naphthyl[b]
4-Cl
4Æ0.1
4Æ1
33Æ2
5-Cl
4-Br
5-Br
4-F
5-F
5-Me
6-Me
ppt[d]
73Æ0.5
0.81Æ0.3
8Æ1.3
8Æ2
4,5-diMe
5-Et
>250
9Æ0.83
9Æ0.54
3Æ0.14
8Æ0.45
ppt[d]
5-iPr
15Æ1
4,5-naphthyl[b]
4-Cl
7Æ0.5
36Æ1.5
62Æ3.5
4Æ0.19
3Æ0.012
1Æ0.04
5-Cl
4-Br
5-Br
4-F
4,5-diMe
5-Et
6Æ0.2
14Æ0.5
22Æ3
20j
20k
20l
5-iPr
4,5-naphthyl[b]
5-F
35Æ0
[a] Average Kd values (n=2) calculated using the Cheng–Prusoff equation from IC50 values measured by FPA competition assay. [b] The entire ring system
replaces the phenyl. [c] Ligand efficiency values calculated using LE=1.4”pKd/HAC (HAC=number of non-hydrogen atoms) using FPA data. [d] Visible pre-
cipitation in assay wells. [e] Not calculated.
General procedure for anthranilic acid based inhibitors: The an-
thranilic acid based inhibitors 11 a–l, 12–17, 18a–l, 19, 20a–p, and
Table 3. SAR of substituents on middle phenyl ring.
21–36 were prepared by similar procedures. This procedure is ex-
emplified for compound 11.[18]
3-(Chlorosulfonyl)-4-methylbenzoic acid 40: 4-Methylbenzoic acid
37 (1.0 g, 7.35 mmol, 1 equiv) was dissolved in chlorosulfonic acid
(10 mL). The reaction was heated at reflux and stirred overnight.
Compd
R3
R4
Kd [mm][a]
LE[b]
The next day, the reaction was cooled to room temperature and
then poured onto ice. The solid was filtered, dissolved in CH2Cl2
(150 mL), and washed with 1m HCl (50 mL). The CH2Cl2 layer was
then dried (Na2SO4) and evaporated in vacuo to give the desired
product 37 as a white solid (1.22 g, 71%). 1H NMR (600 MHz,
[D6]DMSO): d=8.32 (d, J=1.9 Hz, 1H), 7.77 (dd, J=2.0, 7.7 Hz, 1H),
7.26 (d, J=7.9 Hz, 1H), 2.58 ppm (s, 3H); 13C NMR (125 MHz,
[D6]DMSO): d=167.2, 146.4, 141.2, 131.2, 129.6, 127.7, 127.5,
20.3 ppm; MS (ESI) [M+H]+ m/z=234.9.
20
Me
H
Et
iPr
Me
H
H
H
H
Me
7Æ3
0.23
0.22
0.23
0.22
0.21
20m
20n
20o
20p
17Æ1
4Æ0.25[c]
5Æ0.19[c]
14Æ1.5
[a] Average Kd values (n=2) calculated using the Cheng–Prusoff equation
from IC50 values measured by FPA competition assay. [b] Ligand efficiency
values calculated using LE=1.4”pKd/HAC (HAC=number of non-hydro-
gen atoms) using FPA data. [c] Some evidence of precipitation noted at
highest concentrations.
Methyl 2-(3-(chlorosulfonyl)-4-methylbenzamido)benzoate 43a:
The intermediate 40 (235 mg, 1 mmol, 1 equiv) was dissolved in
thionyl chloride (4 mL). The reaction was heated at 758C and
stirred for 4 h. Solvents were removed in vacuo. The resulting
syrup was dissolved in toluene (3”5 mL) and evaporated. The
product was taken forward without further purification. The appro-
priate methyl-2-aminobenzoate (151 mg, 1 mmol, 1 equiv) was dis-
solved in THF (4 mL), and NaH (40 mg, 1 mmol, 1 equiv) was added
and stirred for 20 min. The acyl chloride (1 mmol, 1 equiv) was
added, and the reaction was stirred at RT for 2 h. The reaction was
diluted with CH2Cl2 (100 mL) and washed with water (25 mL). The
CH2Cl2 layer was dried (Na2SO4) and then evaporated in vacuo. The
white solid residue was taken forward without further purification
quartet, br=broad, m=multiplet), coupling constant (Hz), and in-
tegration. Low-resolution mass spectra were obtained on an Agi-
lent 1200 series 6140 mass spectrometer with electrospray ioniza-
tion. All samples were of ꢀ90% purity as analyzed by LC–UV/Vis–
MS. Analytical HPLC was performed on an Agilent 1200 series with
UV detection at l 214 and 254 nm along with ELSD detection. LC–
MS parameters were as follows: Phenomenex-C18 Kinetex column,
50 mm”2.1 mm, 2 min gradient, 5–100% (H2O/MeCN with 0.1%
TFA). Preparative purification was performed on a Gilson HPLC
(Phenomenex-C18, 100 mm”30 mm, 10 min gradient, (H2O/MeCN
with 0.1% TFA) or by automated flash column chromatography
(Teledyne Isco Inc., Combiflash Rf).
1
(367 mg, quant). H NMR (600 MHz, [D6]DMSO): d=8.57 (dd, J=0.8,
8.5 Hz, 1H), 8.39 (d, J=2.1 Hz, 1H), 8.00 (dt, J=1.9, 8.0 Hz, 1H),
7.80 (dd, J=2.1, 7.8 Hz, 1H), 7.67 (m, 1H), 7.37 (d, J=7.9 Hz, 1H),
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
897
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
7.22 (m, 1H), 3.90 (s, 3H), 2.62 ppm (s, 3H); 13C NMR (150 MHz,
[D6]DMSO): d=168.0, 164.7, 146.9, 140.5, 140.3, 134.3, 131.4, 131.1,
130.8, 127.0, 125.7, 123.3, 120.9, 117.1, 52.7, 20.2 ppm; MS (ESI)
[M+H]+ m/z=368.0.
to disrupt the binding of an ATRIP-based probe to RPA70N. The
protocol is described in full detail in the report by Souza-Fagundes
et al.[14] FPA competition assays were conducted as previously de-
scribed with minor modifications.[12,14] Compounds were diluted in
a ten-point, threefold serial dilution scheme in DMSO for a final
concentration range of 500–0.025 mm. Compounds were added to
assay buffer (50 mm HEPES, 75 mm NaCl, 5 mm DTT, pH 7.5) con-
taining FITC-labeled probe and appropriate RPA70 protein in
a final reaction volume of 50 mL containing 5% DMSO. All assays
were conducted using a protein concentration equal to 1”Kd for
the protein–probe interaction. Therefore, competition for binding
to RPA70N was measured using either the FITC-ATRIP peptide
(FITC-Ahx-DFTADDLEELDTLAS-NH2; 50 nm with 6 mm RPA70N) or
the FITC-ATRIP2 peptide (FITC-Ahx-DFTADDLEEWFAL-NH2; 25 nm
with 350 nm RPA70N). Binding to RPA70NAB was measured using
200 nm RPA70NAB and 25 nm FITC-ATRIP2. Following incubation
for 1 h, emission anisotropy was measured using an EnVision plate
reader (PerkinElmer). IC50 values were generated using a four-pa-
rameter dose–response (variable slope) equation in XLfit and were
converted into Kd values. Reported Kd values are the average of
two independent experiments, run in duplicate.
2-(3-(N-(4-Bromophenyl)sulfamoyl)-4-methylbenzamido)benzoic
acid 11: The sulfonyl chloride 43a (62 mg, 0.17 mmol, 1 equiv) was
dissolved in toluene (2 mL). The 4-bromoaniline (86 mg, 0.5 mmol,
3 equiv) was added, and the reaction was stirred at 708C over-
night. The solvents were removed in vacuo. The resulting residue
was dissolved in CH2Cl2 (50 mL) and washed with water (20 mL).
The CH2Cl2 layer was evaporated in vacuo, and the residue was dis-
solved in THF (2 mL), and 2m LiOH (0.5 mL) was added. The reac-
tion was stirred at 558C for 2 h. The reaction was neutralized with
2m HCl (0.5 mL), and the solvents were removed in vacuo. The res-
idue was purified by preparative HPLC to give the desired product
1
as a white solid (23 mg, 28%). H NMR (600 MHz, [D6]DMSO): d=
10.78 (s, 1H), 8.67 (dd, J=0.8, 8.4 Hz, 1H), 8.52 (d, J=1.9 Hz, 1H),
8.08–8.05 (m, 2H), 7.68 (m, 1H), 7.62 (d, J=8.1 Hz, 1H), 7.43–7.41
(m, 2H), 7.24 (m, 1H), 7.09–7.06 (m, 2H), 3.39 (brs, 1H), 2.66 ppm
(s, 3H); 13C NMR (150 MHz, [D6]DMSO): d=170.6, 163.5, 141.6,
141.3, 138.3, 137.1, 134.9, 134.1, 133.0, 132.7, 131.9, 131.8, 128.7,
123.8, 121.5, 120.5, 117.2, 116.2, 20.2 ppm; MS (ESI) [M+H]+ m/z=
489.1.
X-ray crystal structures of complexes with RPA70N
2-(3-(N-(3-Chlorophenyl)sulfamoyl)-4-methylbenzamido)benzoic
acid 18: Synthesized as a white solid according to procedure for
11 in 42% yield (29 mg). 1H NMR (600 MHz, [D6]DMSO): d=10.88
(s, 1H), 8.67 (dd, J=0.9, 8.4 Hz, 1H), 8.55 (d, J=1.9 Hz, 1H), 8.09–
8.06 (m, 2H), 7.68 (m, 1H), 7.63 (d, J=8.1 Hz, 1H), 7.27–7.23 (m,
2H), 7.12–7.09 (m, 2H), 7.04 (m, 1H), 3.40 (brs, 1H), 2.67 ppm (s,
3H); 13C NMR (150 MHz, [D6]DMSO): d=170.1, 162.9, 141.2, 140.7,
138.8, 137.8, 134.4, 133.6, 133.5, 132.5, 132.4, 131.4, 131.3, 131.1,
128.2, 123.4, 123.3, 120.0, 118.1, 116.9, 19.7 ppm; MS (ESI) [M+H]+
m/z=445.2.
Crystals of the E7R mutant of RPA70N were grown as described
previously.[21] X-ray diffraction data were collected at sector 21 (Life
Sciences Collaborative Access Team, LS-CAT) of the Advanced
Photon Source (Argonne, IL, USA). All data were processed by HKL-
2000.[22] E7R crystallized in space group P212121 and contained one
molecule in the asymmetric unit. Initial phases were obtained by
molecular replacement with PHASER[23] using the structure of the
free protein (PDB ID: 4IPC) as a search model. Iterative cycles of
model building and refinement were performed using COOT[24]
and PHENIX.[25] The structure of compound 20c bound to E7R are
deposited at the RCSB Protein Data Bank under accession code
5E7N. The program PyMOL (Schrçdinger) was used to visualize and
analyze the structures.
4-Bromo-2-(3-(N-(3-chlorophenyl)sulfamoyl)-4-methyl benzami-
do)benzoic acid 18c: Synthesized as a white solid according to
procedure for 11 in 34% yield (28 mg). 1H NMR (600 MHz,
[D6]DMSO): d=10.91 (s, 1H), 8.93 (d, J=2.0 Hz, 1H), 8.55 (d, J=
2.0 Hz, 1H), 8.06 (dd, J=1.8, 7.4 Hz, 1H), 7.99 (d, J=8.4 Hz, 1H),
7.63 (d, J=8.0 Hz, 1H), 7.44 (dd, J=2.0, 8.6 Hz, 1H), 7.27 (t, J=
8.1 Hz, 1H), 7.13–7.10 (m, 2H), 7.05 (m, 1H), 3.42 (brs, 1H),
2.68 ppm (s, 3H); 13C NMR (150 MHz, [D6]DMSO): d=168.0, 161.5,
140.1, 139.8, 137.1, 136.2, 132.1, 131.9, 131.3, 130.3, 129.8, 129.5,
126.6, 126.1, 124.5, 121.8, 120.6, 116.5, 115.2, 114.2, 18.1 ppm; MS
(ESI) [M+H]+ m/z=568.9.
Protein binding and cellular permeability studies
The studies on 20c were performed by Absorption Systems, a pre-
clinical contract research organization. Brief details of the studies
can be found in the Supporting Information.
4-Bromo-2-(3-(N-(3,4-dichlorophenyl)sulfamoyl)-4-methyl benza-
mido)benzoic acid 20c: Synthesized as a white solid according to
procedure for 11 in 42% yield (34 mg). 1H NMR (600 MHz,
[D6]DMSO): d=11.05 (s, 1H), 8.92 (d, J=2.1 Hz, 1H), 8.53 (d, J=
1.9 Hz, 1H), 8.06 (dd, J=1.9, 7.9 Hz, 1H), 7.98 (d, J=8.5 Hz, 1H),
7.64 (d, J=8.2 Hz, 1H), 7.50 (d, J=8.8 Hz, 1H), 7.43 (dd, J=2.1,
8.5 Hz, 1H), 7.28 (d, J=2.6 Hz, 1H), 7.12 (dd, J=2.6, 8.9 Hz, 1H),
3.39 (brs, 1H), 2.67 ppm (s, 3H); 13C NMR (150 MHz, [D6]DMSO): d=
169.6, 163.1, 141.8, 141.5, 137.6, 137.4, 133.8, 133.0, 132.1, 131.6,
131.6, 131.4, 128.2, 127.7, 126.1, 125.6, 122.2, 119.8, 118.3, 115.8,
19.7 ppm; MS (ESI) [M+H]+ m/z=556.9.
Acknowledgements
We thank Dr. David Cortez (Vanderbilt University Department of
Biochemistry) for his intellectual contributions in the conception
of this project. We also acknowledge the Vanderbilt High-
Throughput Screening Core, an institutionally supported facility
in which some of these experiments were performed and whose
staff provided invaluable assistance. Funding of this research was
provided in part by US National Institutes of Health (NIH) grants
5DP1OD006933/8DP1A174419 (NIH Director’s Pioneer Award) to
S.W.F., R01M065484 and P01CA092584 to W.J.C., and an Ameri-
can Recovery and Reinvestment Act (ARRA) stimulus grant
(5RC2A148375) to Lawrence J. Marnett, as well as funding from
the Brazilian National Council for Scientific and Technological De-
velopment (CNPq) and the Federal University of Minas Gerais
(Brazil) to E.M.S.-F. A.O.F. was supported by a postdoctoral fellow-
Fluorescence polarization anisotropy (FPA) assays
90000 compounds from the Vanderbilt Institute of Chemical Biol-
ogy compound collection were screened at the High-Throughput
Screening core at a single concentration of 30 mm for their ability
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
898
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[14] E. M. Souza-Fagundes, A. O. Frank, M. D. Feldkamp, D. C. Dorset, W. J.
Chazin, O. W. Rossanese, E. T. Olejniczak, S. W. Fesik, Anal. Biochem.
2012, 421, 742.
[15] J. B. Baell, G. A. Holloway, J. Med. Chem. 2010, 53, 2719.
[16] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv. Drug Delivery
Rev. 1997, 23, 3.
ship from the Deutscher Akademischer Austauschdienst (DAAD),
J.D.P. by an NIH National Research Service Award (NRSA) post-
doctoral fellowship (F32CA174315-02), and M.D.F. by an NIH
NRSA postdoctoral fellowship (F32ES021690-01).
[17] V. J. Anciano Granadillo, J. N. Earley, S. C. Shuck, M. M. Georgiasdis, R. W.
Fitch, J. J. Turchi, J. Nucleic Acids 2010, 1.
[18] E. Bochkareva, L. Kasutov, A. Ayed, G.-S. Yi, A. Pineda-Lucena, J. C. C.
Liao, A. L. Okorokov, J. Milner, C. H. Arrowsmith, A. Bochkarev, Proc. Natl.
Acad. Sci. USA 2005, 102, 15412.
Keywords: high-throughput screening · hit optimization ·
medicinal chemistry · protein–protein interactions · replication
protein A
[19] A. O. Frank, B. Vangamudi, M. D. Feldkamp, E. M. Souza-Fagundes, J. W.
Luzwick, D. Cortez, E. T. Olejniczak, A. G. Waterson, O. W. Rossanese,
W. J. Chazin, S. W. Fesik, J. Med. Chem. 2014, 57, 2455.
[20] A. Thorarensen, C. J. Ruble, D. L. Romero (Pfizer Inc.), Int. PCT Pub No.
WO 2004018428 A3, 2004.
[1] M. S. Wold, Annu. Rev. Biochem. 1997, 66, 61.
[2] M. S. Wold, T. Kelly, Proc. Natl. Acad. Sci. USA 1988, 85, 2523.
[3] X. Xu, S. Vaithiyalingam, G. G. Glick, D. A. Mordes, W. J. Chazin, D.
Cortez, Mol. Cell. Biol. 2008, 28, 7345.
[21] M. D. Feldkamp, A. O. Frank, J. P. Kennedy, J. D. Patrone, B. Vangamudi,
A. G. Waterson, S. W. Fesik, W. J. Chazin, Biochemistry 2013, 52, 6515.
[22] Z. M. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307.
[23] M. D. Winn, C. C. Ballard, K. D. Cowtan, E. J. Dodson, P. Emsley, P. R.
Evans, R. M. Keegan, E. B. Krissinel, A. G. Leslie, A. McCoy, S. J. McNicho-
las, G. N. Murshudov, N. S. Pannu, E. A. Potterton, H. R. Powell, R. J.
Read, A. Vagin, K. S. Wilson, Acta Crystallogr. Sect. D 2011, 67, 235.
[24] P. Emsley, K. Cowtan, Acta Crystallogr. Sect. D 2004, 60, 2126.
[25] P. D. Adams, P. V. Afonine, G. Bunkoczi, V. B. Chen, I. W. Davis, N. Echols,
J. J. Headd, L. W. Hung, G. J. Kapral, R. W. Grosse-Kunstleve, A. J. McCoy,
N. W. Moriarty, R. Oeffner, R. J. Read, D. C. Richardson, J. S. Richardson,
T. C. Terwilliger, P. H. Zwart, Acta Crystallogr. Sect. D 2010, 66, 213.
[4] E. Fanning, V. Klimovich, A. R. Nager, Nucleic Acids Res. 2006, 34, 4126.
[5] D. Cortez, S. Guntuku, J. Qin, S. J. Elledge, Science 2001, 294, 1713.
[6] K. A. Cimprich, D. Cortez, Nat. Rev. Mol. Cell Biol. 2008, 9, 616.
[7] B. J. Andrews, J. J. Turchi, Mol. Cancer Ther. 2004, 3, 385.
[8] S. C. Shuck, J. J. Turchi, Cancer Res. 2010, 70, 3189.
[9] J. G. Glanzer, S. Q. Liu, G. G. Oakley, Bioorg. Med. Chem. 2011, 19, 2589.
[10] J. G. Glanzer, K. A. Carnes, P. Soto, S. Liu, L. J. Parkhurst, G. G. Oakley, Nu-
cleic Acids Res. 2013, 41, 2047.
[11] J. G. Glanzer, S. Liu, L. Wang, A. Mosel, A. Peng, G. G. Oakley, Cancer Res.
2014, 74, 5165.
[12] J. D. Patrone, J. P. Kennedy, A. O. Frank, M. D. Feldkamp, B. Vangamudi,
N. F. Pelz, O. W. Rossanese, A. G. Waterson, W. J. Chazin, S. W. Fesik, ACS
Med. Chem. Lett. 2013, 4, 601.
[13] A. O. Frank, M. D. Feldkamp, J. P. Kennedy, A. G. Waterson, N. F. Pelz,
J. D. Patrone, B. Vangamudi, D. V. Camper, O. W. Rossanese, W. J. Chazin,
S. W. Fesik, J. Med. Chem. 2013, 56, 9242.
Received: October 14, 2015
Revised: December 8, 2015
Published online on January 8, 2016
ChemMedChem 2016, 11, 893 – 899
www.chemmedchem.org
899
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim