Diphenylpyrazoles as replication protein A inhibitors

Article abstract of DOI:10.1021/ml5003629

Replication Protein A is the primary eukaryotic ssDNA binding protein that has a central role in initiating the cellular response to DNA damage. RPA recruits multiple proteins to sites of DNA damage via the N-terminal domain of the 70 kDa subunit (RPA70N). Here we describe the optimization of a diphenylpyrazole carboxylic acid series of inhibitors of these RPA-protein interactions. We evaluated substituents on the aromatic rings as well as the type and geometry of the linkers used to combine fragments, ultimately leading to submicromolar inhibitors of RPA70N protein-protein interactions.

Full text of DOI:10.1021/ml5003629

Subscriber access provided by LSU Libraries
Letter
Diphenylpyrazoles as Replication Protein A inhibitors
Alex G. Waterson, J. Phillip Kennedy, James D Patrone, Nicholas F. Pelz,
Michael D Feldkamp, Andreas O. Frank, Bhavatarini Vangamudi, Elaine Maria
Souza-Fagundes, Olivia W. Rossanese, Walter J Chazin, and Stephen W. Fesik
ACS Med. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/ml5003629 • Publication Date (Web): 11 Nov 2014
Downloaded from http://pubs.acs.org on November 14, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Medicinal Chemistry Letters is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 6
ACS Medicinal Chemistry Letters
1
2
3
4
5
6
7
8
9
Diphenylpyrazoles as Replication Protein A inhibitors
Alex G. Waterson2,3, J. Phillip Kennedy1†, James D. Patrone1, Nicholas F. Pelz1, Michael D. Feldkamp1,
Andreas O. Frank1†, Bhavatarini Vangamudi1†, Elaine M. Souza-Fagundes1†, Olivia W. Rossanese1,
Walter J. Chazin1,3, Stephen W. Fesik1,2,3*
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1Department of Biochemistry, 2Department of Pharmacology Vanderbilt University School of Medicine, 3Department of
Chemistry, Vanderbilt University, Nashville, TN 37232 (USA)
KEYWORDS: Replication protein A, fragment-based discovery, medicinal chemistry
ABSTRACT: Replication Protein A is the primary eukaryotic ssDNA binding protein that has a central role in initiating the cellu-
lar response to DNA damage. RPA recruits multiple proteins to sites of DNA damage via the N-terminal domain of the 70 kDa
subunit (RPA70N). Here we describe the optimization of a diphenylpyrazole carboxylic acid series of inhibitors of these RPA-
protein interactions. We evaluated substituents on the aromatic rings as well as the type and geometry of the linkers used to com-
bine fragments, ultimately leading to sub-micromolar inhibitors of RPA70N protein-protein interactions.
Replication Protein A (RPA), the eukaryotic ssDNA binding
protein, is a widely used scaffold for DNA transactions that
functions by protecting and organizing ssDNA, as well as re-
cruiting other proteins of the DNA processing machinery.1,2
RPA is a heterotrimer; the 70-kDa subunit contains an OB-
fold at its N-terminus, RPA70N, that serves as a protein re-
cruitment module.1-3 In particular, this domain plays a crucial
role in recruiting DNA damage response and repair proteins to
sites of DNA damage, using a conserved motif that interacts
with a basic cleft at the surface of RPA70N.4-6
volves a cyclodehydration reaction to generate functionalized
diphenylpyrazoles that are elaborated in various ways to con-
struct the inhibitor compounds. Full details of the compound
syntheses are reported in the supplementary information.
Cl
Cl
N
N
HS
N
CO2H
CO2H
S
N
S
N
S
N
Cl
Cl
Cl
Cl
1
11 µM
2
18 µM
Based on the critical role of the RPA70N interactions with
target proteins in initiating DNA damage response, inhibition
of these interactions has the potential to suppress damage re-
sponse pathways and may represent a promising point of in-
tervention for cancer therapy. A selective inhibitor of the
RPA70N-protein interactions that does not abrogate ssDNA
binding or important scaffolding activities of the entire protein
would provide a powerful tool for focused exploration of the
role of RPA in checkpoint signaling and enable studies to con-
firm the therapeutic potential of RPA70N inhibition, as well as
a potential starting point for new cancer drugs. Indeed, a re-
cent report describes a small molecule that binds to RPA70N
and demonstrates activity in a xenograft tumor model.7
CO2H
N
N
Cl
HO2C
H
N
O
Cl
S
Cl
3
0.19 µM
Figure 1. Our previously reported RPA70N inhibitors.8,9 Values
represent binding affinities determined in an FPA assay.
We explored the SAR of the 1,5-diphenyl pyrazoles via an
organized system of chemical synthesis and targeted purchas-
ing (Table 1). The initial analogs were evaluated for binding to
RPA70N using HSQC NMR-based titrations. Of note, the
carboxylic acid was critical for binding, as all analogs devoid
of this acid, or with ester or amide derivatives, displayed
sharply reduced binding affinity to the protein (not shown).
Toward the discovery of new inhibitors, we have reported the
use of a fragment-based screen to identify small molecules
that bind to the basic cleft of RPA70N, the elaboration of these
fragments to produce two triazole series (1, 2), and prelimi-
nary linking efforts to obtain molecules (e.g. 3) that span the
entire cleft (Figure 1).8,9 Here, we describe the optimization of
the linked compounds, including the details of the structure-
activity relationships (SAR) for binding to RPA70N and the
physicochemical properties of selected compounds.
We quickly established a preference for 3- and 4- position
substituents on this ring (compare 4a versus 4 c,d). The most
robust trend evident in this initial group was a very strong
preference for halogens in the 3- and 4-positions of the 1-
phenyl ring (e.g. 4f). Indeed, we found a distinct binding affin-
ity advantage for a 3,4-dichlorophenyl substitution, mirroring
the preference found in a prior series of RPA70N binding
The synthetic route used to produce linked compounds such as
38 was adapted to explore the SAR of the diphenylpyrazoles
and linked compounds described herein. The basic route in-
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 2 of 6
Table 1. Pyrazole SAR in Site-1.
thyl at the 3-position. Larger substitutions in either the 3- or 4-
1
2
3
4
5
6
7
8
position, for example, the use of bicyclic ring systems and
larger alkyl groups such as isopropyl, were completely inef-
fectual (data not shown). However, as observed with 4f, com-
bining the 3- and 4-chloro substitutions produces a large gain
in binding affinity (4m). Finally, we established an advantage
for a phenyl substituent on the 1- position of the pyrazole
(compare 4n versus 4a).
O
OH
R2
N
N
R1
NMR K
Cpd
R1
R2
Kd (µM)b % inhc LEd
(µM)a
Holding the 3,4-dichlorophenyl substitution pattern constant,
we initiated an investigation of 5-phenyl ring modifications. In
general, a relatively wide range of substitutions was tolerated
with minimal changes in binding affinity or ligand efficiency
(LE) (Table 2). We observed a slight preference for 4-position
substitutions over 3-position analogs (5b vs. 5c, 5d vs. 5e). A
4-cyano substituent was 4-fold less potent than the methyl
analog (5b v. 5f) and the corresponding amine-containing 5g
was completely inactive. While some polar groups were toler-
ated in this region (e.g. 5d, 5e, 5h, and 5i), heterocyclic ring
systems were generally disfavored (e.g. 5j). In addition, a
methylene could not be introduced between the pyrazole and
the phenyl ring without a substantial loss of binding affinity
(e.g. 5k).
9
4a
4b
4c
4d
4e
-H
-H
1150
895
580
260
410
0.21
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4-OMe
3-Me
4-Cl
-H
>500
>500
>500
>500
18
0.19
0.22
0.23
0.21
-H
4-Cl
4-Cl
24
29
2,4-diCl
4f
3,4-diCl
3-Me
4-Cl
54
0.25
N/A
67±9
4g
4-OMe
>500
20
15
4h
4i
4-Me
3-Br
4-OMe
4-OMe
4-OMe
4-OMe
4-OMe
4-OMe
>500
N/A
0.23
N/A
N/A
N/A
0.24
N/A
Many SAR trends in this series can be rationalized by the co-
crystal structures of the molecules bound to the basic cleft of
RPA70N. As previously described,8 1,5-diarylpyrazoles bind
140±2.8
>500
4j
4-Br
24
35
19
Table 2. Pyrazole 5-position SAR.
4k
4l
3-Cl
>500
O
4-Cl
>500
OH
R
4m
4n
3,4-diCl
79±1.4
>500
N
N
cyclohexyle 4-OMe
9
aKd values determined from 6 point NMR titrations. Average Kd
values (n=2) calculated using Cheng-Prusoff equation from IC50
values measured in FPA competition assay. cAmount of dis-
placement of labeled probe at the highest compound concentration
used, 500µM. dLE = ligand efficiency. Values calculated using LE
= 1.4*pKd/HAC, using FPA data if available, NMR data if the
b
Cl
Cl
CpdR
Kd (µM)a
LEc
4f 4-Cl
0.25
0.25
0.25
0.25
0.22
0.24
0.20
N/A
0.22
0.24
N/A
67±9
e
FPA data did not give a Kd value. Represents replacement of the
5a
H
123±4.2
90±1.4
72±4.9
129±17
59±3.5
310±21
entire phenyl ring.
5b 3-Me
compounds and in a potent stapled helix peptide inhibitor of
RPA70N.9,10 It is likely that the halogen substituents lead to
additional hydrophobic interactions and a possible halogen
bond in the lipophilic cavity near Ser55 in RPA70N (Site-1) in
which the N-phenyl ring of the pyrazole compounds bind.8
5c 4-Me
5d 3-CO2H
5e 4-CO2H
5f 4-CN
Compound 4f displayed sufficient NMR-derived binding af-
finity to also be evaluated in an FPA competition assay,11
which produced a comparable binding affinity of 67 µM. This
assay was used exclusively to profile later analogs. To identify
additional substituents that might effectively fill Site-1, we
initiated a more thorough investigation of the 1-phenyl ring
SAR, systematically exploring lipophilic and halogen substitu-
tions at the 3- and 4- positions, using a 4-methoxyphenyl ring
at the pyrazole 1-position as a standard substitution. Unfortu-
nately, most of these analogs displayed a loss of binding affin-
ity 10-fold or greater compared with 4f, with only the 3-Br
analog 4i producing a full concentration-response curve. Some
rank ordering of the compounds was determined by examining
the maximal displacement of the FPA probe at the highest
compound concentration. From this, it was generally evident
that 3-position substitutions were favored over 4-position sub-
stitutions and that a bromide was favored over chloro and me-
5g 4-CH2NH2 >500
5h 4-NH2
5i 4-OH
227±27
104±15
>500
5j
4-(2-
furanyl)b
HO
N/A
5k
>500
b
aAverage Kd values (n=2) calculated using Cheng-Prusoff equa-
tion from IC50 values measured in FPA competition assay. bEntire
c
ring system replaces the phenyl. LE = ligand efficiency. Values
values calculated using LE = 1.4*pKd/HAC, using the FPA data.
ACS Paragon Plus Environment

Page 3 of 6
ACS Medicinal Chemistry Letters
Table 3. Extension of the diphenylpyrazoles. parts, indicating the likelihood of a specific interaction. Inter-
1
2
3
4
estingly, several attempts to replace the appended phenyl rings
with heterocycles produced compounds with extremely weak
binding to the protein (data not shown), reinforcing the prefer-
ence for lipophilic substitutions in this region of the protein.
O
N
OH
R
N
5
6
7
8
Fragment linking to generate high affinity inhibitors is consid-
ered difficult to achieve, due to the specific geometrical con-
straints that must be precisely matched to successfully link
two molecules that bind to adjacent sites in a protein. Howev-
er, several successful examples demonstrate the power of this
approach.12 RPA70N appeared ideally suited for this strategy,
as a single fragment screen identified multiple series of frag-
ments that bind to two different, nearby sites in the protein.
Cl
Cl
9
Cpd
R
Kd (µM)a
LEb
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4f
4-Cl
0.25
0.18
67±9
6a
4-CONHCH2(phenyl)
83±18
An ether linker should provide flexibility and allow a linked
compound to achieve a binding pose similar to the original
fragments. However, connecting the optimized diphenylpyra-
zole and a furan-based fragment with either a two atom link8
or a longer, PEG-inspired 4 atom link (data not shown) proved
less successful than anticipated. We virtually docked
4-CONHCH2(4-
methoxyphenyl)
6b
6c
6d
45±11
90±15
72±7.8
0.18
0.18
0.17
4-CH2NHCO(phenyl)
4-CH2NHCO(4-
methoxyphenyl)
Table 4. Fragment linking and optimization.
6e
6f
4-CH2NHCO(4-bromophenyl) 35±2.1
0.19
0.21
O
OH
4-CSNHCH2(phenyl)
18±6.4
23±4.2
N
R
N
4-CSNHCH2(4-
methoxyphenyl)
linker
6g
0.19
Cl
O
Cl
6h
6i
4-CH2NHCS(4-bromophenyl) 15±1.4
0.20
0.17
O
OH
4-CH2NHSO2(phenyl)
70±9.2
56±2.8
Linker
a
Cpd
R
Linker
LEb
4-CH2NHSO2(4-
trifluoromethoxyphenyl)
Kd (µM)
position
6j
0.16
N
H
7a
7b
H
H
para
para
57±1.4
8.2±2.6
0.15
0.17
aAverage Kd values (n=2) calculated using Cheng-Prusoff equa-
tion from IC50 values measured in FPA. LE = ligand efficiency.
Values calculated using LE = 1.4*pKd/HAC, using the FPA data.
b
N
Ac
O
7c
7d
7e
H
meta
para
para
22±4.2
2.5±0.1
1.5±0.3
0.16
0.20
0.20
with good affinity to Site-1 and are located above Ser55 in the
protein. The 3,4-dichloro substitution more effectively fills the
Site-1 pocket compared with other substitution patterns. Inter-
estingly, the 4-position chloride is located 3.5-3.8 Å away
from the sidechain oxygen of Ser 55, at bond angles of 150-
160°, indicating a possible halogen bond interaction. In addi-
tion, the 5-position phenyl ring is located such that substitu-
tions at the 3- and 4-positions are aimed directly down the
basic cleft. Thus, it is reasonable to expect that a range of
modifications would be tolerated on this ring. The hydropho-
bic nature of the central portion of the cleft may also explain
the poor binding affinity of the basic amine-containing 5g.
N
H
S
N
H
c
S
H
N
H
S
7f
H
para
para
para
para
0.55±0.1 0.22
N
H
O
7g
7h
7i8
Cl
Cl
Cl
9.4±3.8
0.17
N
H
Because the 5-phenyl ring is ideally positioned in the cleft of
RPA70N to tolerate larger functionality, we elected to employ
a fragment growing strategy to seek affinity improvements.
Using the carboxylic acid of 5e and the amine of 5g as syn-
thetic handles, we synthesized an exploratory series of substi-
tuted amides, thioamides, and sulfonamides with phenyl moie-
ties predicted to extend down the cleft of RPA70N (Table 3).
Most of the amide and sulfonamide analogs failed to demon-
strate substantial improvement over 4f or 5e. In general, the
best analogs within this group (6e-h) employ highly lipophilic
thioamide-containing substitutions and display binding affini-
ties 2-3 fold better than 4f. As these substitutions did not add
significant affinity to the molecules to offset large increases in
molecular weight, the corresponding ligand efficiencies are
lower. However, it should be noted that the thioamides had
generally higher ligand efficiencies than their amide counter-
S
0.48±0.1 0.22
N
H
O
2.9±0.8
0.18
N
H
S
38
7j
Cl
-
para
para
0.19±0.03 0.23
0.21
N
H
N
S
0.54±0.2
aAverage Kd values (n=2) calculated using Cheng-Prusoff equa-
b
tion from IC50 values measured in FPA. LE = ligand efficiency.
Values calculated using LE = 1.4*pKd/HAC, using the FPA data.
cn=1 data. cThe linker is attached to the meta position of the furan
phenyl ring and the para position of the pyrazole phenyl ring.
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 4 of 6
prospective molecules into RPA70N using the Induced Fit the occupation of space in the binding site above Met57 (Fig-
Docking Protocol available from Schroedinger.13 This sug-
gested exploration of three atom linkers. Thus, we used the
amine of 5g as a synthetic handle (Table 4). The initial com-
pound, 7a, reinforced the lack of tolerance for a charged atom
in this region of the protein. Acetylating this amine produced a
10-fold gain in binding affinity (7b). However, the binding
mode of the compound in complex with RPA70N reveals a
linker conformation that is kinked to accommodate the re-
quirements of the original fragment molecules in Site-1 and
Site-2 (Figure 2A).
Further molecular modeling also suggested that an sp2 hybrid-
ized carbon in the linker would improve the binding geometry.
This idea was explored in the context of both meta- and para-
linked compounds, with a carbonyl or thiocarbonyl attached to
the 5-phenyl ring of the pyrazole. In accordance with the com-
putational predictions, linkers with an sp2 center were general-
ly superior to those with saturated linkers. We found a distinct
preference for a more lipophilic thioamide substitution over
the amide (7c vs. 7e, 7g vs. 7h). The thioamide appears to
occupy a small hydrophobic space under Leu87 (Figure 2B),
potentially explaining the enhanced binding affinity granted
by this substitution. Placing the linker on the 3-position of the
furan phenyl ring offered no significant advantage over the 4-
position (7d vs. 7e). In addition, placement of the thioamide
carbonyl on the furan phenyl ring was found to be superior to
placement on the 5-phenyl pyrazole (7f vs. 7e). The structure
of 7f in complex with RPA70N reveals that the sulfur is able
to access the Leu87 pocket with less contortion of the linker
portion of the molecule (blue circle, Figure 2C), leading to a
compound with submicromolar binding affinity to RPA70N.
ure 2D). The application of all of these strategies produced
compounds with the highest binding affinity to RPA70N and
with ligand efficiencies similar to the original pyrazole-
containing compounds (3).8 A cyclic thioamide linker (7j) also
places non-polar atoms under Leu87, occupies the space above
Met57 (Figure 2E), and binds with submicromolar affinity.
1
2
3
4
5
6
7
8
We explored removal or replacement of one or both of the two
carboxylic acids from the high affinity linked compounds to
guard against possible issues with poor cellular permeability.
Indeed, we assessed the permeability of compound 3 in the
Caco-2 cell line and found that it possesses poor intrinsic per-
meability of <0.21 x 10-6 cm/s and is subject to significant
pgp-mediated efflux (efflux ratio >32). Applying the optimal
thioamide positioning, we explored isosteres of the 2-furan
carboxylic acid designed to maintain the key binding interac-
tions with Arg31 (8-10, Figure 3). The best of these com-
pounds (9) displays low micromolar binding affinity. We also
explored the removal of the pyrazole carboxylic acid. Com-
pound 11, devoid of this carboxylic acid, has a 1.7 µM binding
affinity to RPA70N in our FPA assay and an intrinsic permea-
bility of 14.5 x 10-6 cm/s, with no pgp efflux noted. The X-ray
co-crystal structure of this molecule, overlaid with the previ-
ously reported structure of 7j (Figure 2F), demonstrates that
the lack of interaction with Arg31 leads to a slight relaxation
of the molecule in the pocket and a slight shift toward Site-2.
This shift, and the missing interaction with Arg31, may ex-
plain the 10-fold loss in binding affinity compared with 3.
9
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Discussion/Conclusions: Interactions mediated by the
RPA70N domain of RPA are essential for recruiting a number
of key proteins to initiate DNA damage response and repair
(e.g. ATRIP, MRE11, RAD9). Each of these interactions is
Introduction of a 3-position chloride (inspired by the fragment
SAR)8 on the phenylfuran produced affinity gains over the
des-chloro variants (e.g. 7h vs. 7e, 3 vs. 7f), probably due to
Figure 2. X-ray co-crystal structures of compounds in complex with RPA70N. A) Compound 7b (PDB 4R4Q), B) Compound 7e (4R4O),
C) Compound 7f (4R4T), D) Compound 7i (4R4C), E) Compound 7j (4R4I), F) Compound 11 (green, 4LWC) overlaid with 3 (orange).
ACS Paragon Plus Environment

Page 5 of 6
ACS Medicinal Chemistry Letters
Houston, TX, 77054; (E.M.S.-F.) Federal University of Minas
Gerais, Belo Horizonte, Brazil.
Figure 3. Linked compounds with acid replacements.a
1
2
3
4
HN
HN
Funding Sources
O
O
S
S
O
N
Funding of this research was provided in part by NIH grants
5DP1OD006933/8DP1CA174419 (NIH Director’s Pioneer
Award) and R01CA174887 to S.W.F., R01GM065484 and
S
N
N
OH
N
OH
N
N
O
H2N
NH
N
5
6
7
8
Cl
Cl
Cl
Cl
P01CA092584
to
W.J.C.,
ARRA
stimulus
grant
8
9
34 + 12.7 µM
(0.17)
2.2 + 1.0 µM
(0.21)
(5RC2CA148375) to Lawrence J. Marnett, F32ES021690 to
M.D.F., F32CA174315 to J.D.P. and 5T21CA9582-24 to Scott
Hiebert (Trainee: J.P.K.). A.O.F. was supported by the Deutscher
Akademischer Austausch Dienst (DAAD) postdoctoral fellowship
and E.M.S.-F. was supported by fellowship from the National
Council for Scientific and Technological Development – CNPq
and Federal University of Minas Gerais/Brazil. The NMR instru-
mentation used in this work was supported by NIH grant S10
RR025677-01 and by NSF grant DBI-0922862.
HN
HN
S
9
O
S
N
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
N
N
OH
N
N
Cl
NH
O
O
HO
Cl
Cl
O
Cl
Cl
10
11
9.3 + 3.8 µM
(0.19)
1.7 + 0.6 µM
(0.21)
aAverage Kd values (n=2) calculated using Cheng-Prusoff equa-
tion from IC50 values measured in FPA. Ligand efficiency (LE)
values in parentheses.
ACKNOWLEDGMENT
We would like to thank Dr. David Cortez for his intellectual con-
tributions in the conception of this project.
mediated by the binding of RPA70N to an amphiphilic helix
from the target protein. The amphiphilic nature of the
RPA70N binding site seems to produce strict requirements for
a tightly binding compound, as evidenced by our optimization
efforts. A significant amount of hydrophobic contact area must
be engaged, and the binding must be anchored by interaction
with charged residues at the outer edges of the cleft.
REFERENCES
(1) Wold, M. S. Replication protein A: A heterotrimeric, single-
stranded DNA-binding protein required for eukaryotic DNA metabo-
lism. Annu. Rev. Biochem. 1997, 66, 61-92.
(2) Wold, M. S.; Kelly, T. Purification and Characterization of
Replication Protein-a, a Cellular Protein Required for Invitro Replica-
tion of Simian Virus-40 DNA. Proc. Natl. Acad. Sci. U.S.A. 1988, 85,
2523-2527.
(3) Xu, X.; Vaithiyalingam, S.; Glick, G. G.; Mordes, D. A.;
Chazin, W. J.; Cortez, D. The Basic Cleft of RPA70N Binds Multiple
Checkpoint Proteins, Including RAD9, To Regulate ATR Signaling.
Mol. Cell Biol. 2008, 28, 7345-7353.
(4) Fanning, E.; Klimovich, V.; Nager, A. R. A dynamic model for
replication protein A (RPA) function in DNA processing pathways.
Nucleic Acids Res. 2006, 34, 4126-4137.
(5) Cortez, D.; Guntuku, S.; Qin, J.; Elledge, S. J. ATR and
ATRIP: Partners in checkpoint signaling. Science 2001, 294, 1713-
1716.
(6) Cimprich, K. A.; Cortez, D. ATR: an essential regulator of ge-
nome integrity. Nature Rev. Mol. Cell Biol. 2008, 9, 616-627.
(7) Glanzer J. G.; Liu, S.; Wang, L.; Mosel, A.; Peng, A.; Oakley,
G. G. Cancer Res. 2014, 74, 5165-5172.
(8) Frank, A. O.; Feldkamp, M. D.; Kennedy, J. P.; Waterson, A.
G.; Pelz, N. F.; Patrone, J. D.; Vangamudi, B.; Camper, D. V.; Ros-
sanese, O. W.; Chazin, W. J.; Fesik, S. W. Discovery of a potent in-
hibitor of replication protein a protein-protein interactions using a
fragment-linking approach. J. Med. Chem. 2013, 56, 9242-9250.
(9) Patrone, J. D.; Kennedy, J. P.; Frank, A. O.; Feldkamp, M. D.;
Vangamudi, B.; Pelz, N. F.; Rossanese, O. W.; Waterson, A. G.;
Chazin, W. J.; Fesik, S. W. Discovery of Protein-Protein Interaction
Inhibitors of Replication Protein A. ACS Med. Chem. Lett. 2013, 4,
601-605.
(10) Frank, A. O.; Vangamudi, B.; Feldkamp, M. D.; Souza-
Fagundes, E. M.; Luzwick, J. W.; Cortez, D.; Olejniczak, E. T.; Wat-
erson, A. G.; Rossanese, O. W.; Chazin, W. J.; Fesik, S. W. Discov-
ery of a potent stapled helix peptide that binds to the 70N domain of
replication protein A. J. Med. Chem. 2014, 57, 2455-2461.
(11) Souza-Fagundes, E. M.; Frank, A. O.; Feldkamp, M. D.; Dor-
set, D. C.; Chazin, W. J.; Rossanese, O. W.; Olejniczak, E. T.; Fesik,
S. W. A high-throughput fluorescence polarization anisotropy assay
for the 70N domain of replication protein A. Anal. Biochem. 2012,
421, 742-749.
We have generated molecules with sub-micromolar binding
affinity to this cleft using fragment linking and subsequent
optimization. The requirements to successfully bind to the
basic cleft of RPA70N result in small molecules with relative-
ly poor pharmaceutical properties. Careful molecular design to
remove one of the negatively charged groups, yet maintain the
other important contributors to binding affinity, has produced
a compound with enhanced permeability. However, this trade-
off results in loss of binding affinity to the protein, pointing
toward opportunities for future optimization. The molecules
described here represent a useful starting point for obtaining
potent and cell permeable RPA inhibitors that could be used as
tools to validate that inhibition of the RPA70N-protein interac-
tions is a therapeutically relevant avenue for suppression of
the DNA damage response as an adjuvant cancer treatment.
ASSOCIATED CONTENT
Supporting Information
Synthetic procedures, compound characterization, and assay pro-
tocols. This material is available free of charge via the Internet at
http://pubs.acs.org. The atomic coordinates and structure factors
for the X-ray crystal structures of RPA70N in complex with com-
pounds 7b (4R4Q), 7e (4R4O), 7f (4R4T), 7i (4R4C), 7j (4R4I),
and 11 (4LWC) have been deposited in the Protein Data Bank
RCSB PDB.
AUTHOR INFORMATION
Corresponding Author
* Tel: 615-322-6303. E-mail: stephen.fesik@vanderbilt.edu.
Present Addresses
†(J.P.K.) Revance Therapeutics, Newark, CA 94560, United
States; (AO.F.) Novartis Institutes for BioMedical Research
(NIBR), Global Discovery Chemistry, Emeryville, California
94608, United States; (B.V.) UT MD Anderson Cancer Center,
(12) Ichihara, O.; Barker, J.; Law, R. J.; Whittaker, M. Compound
Design by Fragment-Linking. Mol. Inform. 2011, 30, 298-306.
(13) Schrödinger Suite 2009 Induced Fit Docking protocol; Glide
version 5.5, Schrödinger, LLC, New York, NY, 2009; Prime version
2.1, Schrödinger, LLC, New York, NY, 2009.
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 6 of 6
SYNOPSIS TOC
1
2
3
4
5
6
7
8
9
Diphenylpyrazoles as Replication Protein A inhibitors
Alex G. Waterson2,3, J. Phillip Kennedy1†, James D. Patrone1, Nicholas F. Pelz1, Michael D. Feldkamp1,
Andreas O. Frank1†, Bhavatarini Vangamudi1†, Elaine M. Souza-Fagundes1†, Olivia W. Rossanese1,
Walter J. Chazin1,3, Stephen W. Fesik1,2,3*
10
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
H
O
N
OH
Cl
OH
H
N
N
N
N
N
N
O
O
N
HO
S
HO
S
O
O
Cl
Cl
Cl
Cl
1150 µM
0.54 µM
1.7 µM
ACS Paragon Plus Environment