Discovery of Potent, Selective, and Orally Bioavailable Inhibitors of USP7 with in Vivo Antitumor Activity

Article abstract of DOI:10.1021/acs.jmedchem.0c00245

USP7 is a promising target for cancer therapy as its inhibition is expected to decrease function of oncogenes, increase tumor suppressor function, and enhance immune function. Using a structure-based drug design strategy, a new class of reversible USP7 inhibitors has been identified that is highly potent in biochemical and cellular assays and extremely selective for USP7 over other deubiquitinases. The succinimide was identified as a key potency-driving motif, forming two strong hydrogen bonds to the allosteric pocket of USP7. Redesign of an initial benzofuran-amide scaffold yielded a simplified ether series of inhibitors, utilizing acyclic conformational control to achieve proper amine placement. Further improvements were realized upon replacing the ether-linked amines with carbon-linked morpholines, a modification motivated by free energy perturbation (FEP+) calculations. This led to the discovery of compound 41, a highly potent, selective, and orally bioavailable USP7 inhibitor. In xenograft studies, compound 41 demonstrated tumor growth inhibition in both p53 wildtype and p53 mutant cancer cell lines, demonstrating that USP7 inhibitors can suppress tumor growth through multiple different pathways.

Full text of DOI:10.1021/acs.jmedchem.0c00245

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
Discovery of Potent, Selective, and Orally Bioavailable
Inhibitors of USP7 with In Vivo Anti-Tumor Activity
Paul Robert Leger, Dennis X Hu, Berenger Biannic, Minna Bui, Xinping Han, Emily Karbarz,
Jack Maung, Akinori Okano, Maksim Osipov, Grant M. Shibuya, Kyle Young, Christopher
Higgs, Betty Abraham, Delia Bradford, Cynthia Cho, Christophe Colas, Scott Jacobson,
Yamini M. Ohol, Deepa Pookot, Payal Rana, Jerick Sanchez, Niket Shah, Michael Sun,
Steve Wong, Dirk G. Brockstedt, Paul D. Kassner, Jacob B. Schwarz, and David J. Wustrow
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00245 • Publication Date (Web): 17 Apr 2020
Downloaded from pubs.acs.org on April 17, 2020
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 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 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.
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 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Potent, Selective, and Orally
Bioavailable Inhibitors of USP7 with In Vivo Anti-
Tumor Activity
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
‡
†‡
‡
‡
‡
Paul R. Leger , Dennis X. Hu , Berenger Biannic , Minna Bui , Xinping Han , Emily
‡
‡
‡
‡
‡
Karbarz , Jack Maung , Akinori Okano , Maksim Osipov , Grant M. Shibuya , Kyle
‡
§
‡
‡
‡
Young , Christopher Higgs , Betty Abraham , Delia Bradford , Cynthia Cho , Christophe
‡
‡
‡
‡
‡
Colas , Scott Jacobson , Yamini M. Ohol , Deepa Pookot , Payal Rana , Jerick
‡
‡
‡
‡
‡
Sanchez , Niket Shah , Michael Sun , Steve Wong , Dirk G. Brockstedt , Paul D.
‡
‡
‡
Kassner , Jacob B. Schwarz , David J. Wustrow*
‡RAPT Therapeutics, Inc., 561 Eccles Avenue, South San Francisco, CA 94080
§
th
Schrödinger, 120 West 45 Street, New York, NY 10036
ABSTRACT. USP7 is a promising target for cancer therapy as its inhibition is expected
to decrease function of oncogenes, increase tumor suppressor function, and enhance
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
immune function. Using a structure-based drug design strategy, a new class of
reversible USP7 inhibitors has been identified that is highly potent in biochemical and
cellular assays and extremely selective for USP7 over other deubiquitinases (DUBs).
The succinimide was identified as a key potency-driving motif, forming two strong
hydrogen-bonds to the allosteric pocket of USP7. Re-design of an initial benzofuran-
amide scaffold yielded a simplified ether series of inhibitors, utilizing acyclic
conformational control to achieve proper amine placement. Further improvements were
realized upon replacing the ether-linked amines with carbon-linked morpholines, a
modification motivated by Free Energy Perturbation (FEP+) calculations. This led to the
discovery of compound 41, a highly potent, selective, and orally bioavailable USP7
inhibitor. In xenograft studies, compound 41 demonstrated tumor growth inhibition in
both p53 wildtype and p53 mutant cancer cell lines, demonstrating that USP7 inhibitors
can suppress tumor growth through multiple different pathways.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
ACS Paragon Plus Environment
2

Page 3 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ubiquitin-specific protease 7 (USP7), also known as HAUSP, is an enzyme
1
which cleaves ubiquitin groups from specific ubiquitinated client proteins. This de-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ubiquitination activity can rescue the client proteins from proteasome-mediated
degradation and result in a net increase in concentration of these proteins. Through this
mechanism, USP7 is proposed to stabilize the oncogenic E3 ubiquitin ligase MDM2, a
2
negative regulator of the tumor suppressor p53. USP7 has also been reported to play a
role in stabilizing FOXP3, an essential transcription factor for the development and
function of regulatory T cells, and hence may play a role in the suppression of immune
3
responses to tumors. Additionally, USP7 may stabilize several proteins involved in
4
DNA damage repair. Pharmacological inhibition of USP7 may therefore be beneficial
for cancer therapy through multiple mechanisms.
A number of USP7-selective reversible inhibitors have recently been reported.5
These include 4-hydroxy piperidine inhibitors 1-4 (Figure 1) disclosed by Hybrigenics,6
7
8
9
10
FORMA Therapeutics, Almac Discovery, Servier, and others, as well as tri-aryl
inhibitors such as 5 disclosed by Genentech.11,12 4-hydroxy piperidine inhibitors such as
2-4 have been reported to bind to an allosteric pocket in the “palm” domain of USP7
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
Figure 2). Small molecule binding to this allosteric site is believed to stabilize an
7
inactive conformation of USP7, preventing USP7 catalytic activity. The tri-aryl inhibitors
such as 5 are reported to bind in a different allosteric pocket in the “palm” domain of
USP7.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
O
N
OH
OH
O
N
N
OH
N
F
N
N
Me N
N
Me
N
N
Me
N
F
N
O
N
O
N
O
F
1
F
2
3
H N
2
Hybrigenics
FORMA
Almac
Me
O N
2
N
N
S
O
O
N
N
NH2
OH
S
S
N
Me
F
N
N
N
N
Me
OH
H
O
F
O
5
P22077, (6)
4
Servier
Genentech
Progenra
Figure 1. Examples of previously described USP7 inhibitors.
Many studies interrogating the biological functions of USP7 have employed
irreversible active-site dual inhibitors of USP7 and USP47, such as P22077 (6, Figure
1).13,14 Potent, DUB-selective, and reversible inhibitors of USP7 of diverse chemotypes
would be useful tools in clarifying and assessing the many potential biological roles of
ACS Paragon Plus Environment
4

Page 5 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
USP7 and its relevance as a target for human cancer therapy. Here we report an
optimization campaign resulting in a new class of highly potent, selective, reversible,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and bioavailable inhibitors of USP7. These inhibitors have recently aided the
clarification of the biological activities of USP7.15
RESULTS AND DISCUSSION
Pyridylbenzofurans as Starting Chemical Matter
Our drug discovery campaign for USP7 inhibitors began with an evaluation of
compounds derived from a focused library screen of known DUB inhibitors as well as
other structurally related compounds. Anticipating DUB-selectivity to be a major
challenge for a USP7 program based on the conserved nature of DUB active sites, early
chemical matter was tested for biochemical activity against USP7 as well as against
USP1, UCH-L3, and USP47. Among the USP7-active compounds evaluated, the 7-
pyridylbenzofuran compound 7 (Figure 2) stood out as an attractive starting point for
optimization due to its relatively high selectivity for USP7 over related de-ubiquitinating
enzymes. Interestingly, co-crystallizations of 4-hydroxy piperidine compound 1 and
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
benzofuran 7 with USP7 revealed that both inhibitors occupy the same allosteric pocket
of the palm region, even though they belong to different chemotypes (Figure 2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. (Left) Schematic of the catalytic domain of USP7 (orange) bound to ubiquitin
(blue), with the palm region (below purple dashed line), tunnel area (red box), and
catalytic site (green circle) highlighted. The yellow arrow indicates the compound
binding region highlighted in the structures on the right. (Right) Comparison of co-crystal
structures of 4-hydroxy piperidine inhibitor 1 and pyridylbenzofuran compound 7. Both
inhibitors bind to the same allosteric site of USP7 and are selective inhibitors of USP7.
The PDB code for 1 bound to USP7 is 6VN4. The PDB code for 7 bound to USP7 is
6VN5.
ACS Paragon Plus Environment
6

Page 7 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Benzofuran compound 7 and 4-hydroxy piperidine compound 1 form many
similar interactions with the USP7 protein despite their distinct structures. Both
compounds form hydrogen-bonds with the backbone nitrogen of Val-296, the phenol
moiety of Tyr-465, and the carboxylate of Asp-295. Importantly, despite the two
compounds similar potencies, there appeared to be interactions made by compound 1
with USP7 that were not made by compound 7 (Figure 3), suggesting that it should be
possible to enhance compound 7’s affinity towards USP7 by introducing novel elements
that would take advantage of additional interactions in the binding pocket. It was
hypothesized that the potency of a compound such as benzofuran 7 could be improved
by introducing functionality which could target the backbone hydrogen bond donor of
Phe-409 with an appropriately linked hydrogen-bond acceptor, or which could better
engage the glutamine side chain of Gln-297.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Comparison of USP7 binding interactions of 4-hydroxy piperidine inhibitor 1
and pyridylbenzofuran inhibitor 7. Boxed residues indicate potential hydrogen-bonding
interactions that are not utilized by compound 7. The PDB code for 1 bound to USP7 is
6VN4. The PDB code for 7 bound to USP7 is 6VN5.
WaterMap16 analysis of the holo structure (ligand removed) revealed several
unfavorable hydration sites in the binding site of 7 (Figure 4). Unfavorable hydration
sites can be characterized as sites that have a predicted free energy G of > 0 kcal/mol
upon adding water while favorable hydration sites have a predicted free energy G of <
0 kcal/mol. Binding of compound 7 likely displaces several of the water molecules as
ACS Paragon Plus Environment
8

Page 9 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
shown by the magnified hydration sites of Figure 4A, including a high-energy site filled
by the aryl chloride moiety of 7. However, sites adjacent to the aminopyridine are not
displaced by the ligand, corroborating the proposed hypothesis that adding appropriate
functionality in this region could lead to improved potency. This was further confirmed
by a WaterMap analysis with compound 7 present in the complex, showing that the
ligand indeed does not displace water sites adjacent to the aminopyridine (Figure 4B).
In fact, the free energy G of these sites is even more unfavorable than predicted from
the structure with ligand removed, likely due to a loss of entropy as the waters are
somewhat trapped between the ligand and protein. The combined structural and
computational analysis strongly suggests that efforts to improve the potency of
compound 7 should be focused on changes to the upper pyridine section of the
molecule.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A)
B)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. WaterMap analysis of USP7 binding pocket. Hydration sites are shown as
spheres colored by their predicted free energies (G). Green spheres indicate a
favorable free energy G while red spheres indicate an unfavorable G. (A) Hydration
sites calculated with the ligand (compound 7) removed. Overlay of compound 7
indicates that the magnified (enlarged) spheres are predicted to be displaced by
compound 7 while the other (smaller) spheres are not. (B) Hydration sites with
compound 7 present. Three unfavorable water sites (red/orange) near the 2-
aminopyridyl headgroup of 7 suggest opportunities to further improve the potency of this
inhibitor. WaterMap analysis based on co-crystal structure of compound 7 and USP7,
PDB code 6VN5.
Addition to the Pyridyl Moiety Improves Potency
ACS Paragon Plus Environment
10

Page 11 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Several bicyclic analogs of 7 were first prepared as scaffolds for a potential
hydrogen-bond acceptor (Table 1). Azaindole 8 was isoefficient to 2-aminopyridine 7,
having increased potency of comparable magnitude to its increased lipophilicity. In
agreement with the selective co-crystallization of the R-isomer from racemic 2-
aminopyridine 7 with USP7, asymmetric synthesis of the enantiomers indicated that the
R-isomer 9 was significantly more active against USP7, while the S-isomer ent-9 was
found to be inactive. Alternative bicyclic scaffolds to the azaindole motif were surveyed.
Thienopyridine 10 was found to be more potent and isoefficient to the corresponding
aza-indole, but with one less hydrogen-bond donor. Substitution of the thienopyridine
with a methyl group (compound 11) did not lead to a gain in potency, but substitution
with a hydroxymethylene group (compound 12) did lead to an improvement in potency
while simultaneously reducing lipophilicity. Consistent with the binding mode of these
inhibitors to an allosteric pocket uniquely accessible in USP7, these inhibitors generally
did not display significant activity against USP47 or USP1.8
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 1. Evaluation of heterocyclic scaffoldsa
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NH
R
O
N
O
Cl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compoun
d
R Group
USP7
USP47
USP1
cLogP LipEf
c
d
e
IC50
nM)b
IC (µM)
IC (µM)
50
50
(
H2N
N
7g
8g
5600
930
>80
>80
>80
>80
2.2
3.1
2.9
2.9
H
N
N
H
N
N
N
9
450
>80
>80
3.1
3.2
1
0
1
90
>40
>80
>40
>80
3.9
4.4
3.1
2.5
S
N
N
1
130
Me
HO
S
12
43
>40
>40
2.9
4.5
S
a
c
b
Data represent the mean of n = 2 or more determinations. USP7 biochemical assay.
d
e
USP47 biochemical assay. USP1 biochemical assay. Calculated using ChemDraw.
f
g
LipE = p(IC ) – cLogP. Racemic compound.
50
ACS Paragon Plus Environment
12

Page 13 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To further improve potency toward USP7, compounds with an additional
hydrogen bond donor or acceptor near the thienopyridine were explored. It was
expected that additional binding affinity could be achieved in this region by
simultaneously forming interactions with the backbone N–H motifs of either Arg-408 or
Phe-409 and the amide side chain of Gln-297. Interactions with these two residues are
evident in co-crystal structures of USP7 with 4-hydroxy piperidine inhibitor 1 (see Figure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3) but are not fully utilized by thienopyridine alcohol 12. Introduction of an oxetane with
the intention of engaging the side chain of Gln-297 resulted in compound 13 which was
not more potent than compound 12 (Table 2). Serendipitously, the chlorohydrin 14
resulting from HCl-mediated chlorinolysis of 13 was found to be significantly more active
against USP7 while decreasing overall lipophilicity. Co-crystallization of 14 with USP7
confirmed that this compound was forming the originally desired contacts with Arg-408
and Gln-297 while binding in a reversible manner (Figure 5).
Table 2. Evaluation of heteroatom-containing headgroupsa
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NH
R
O
N
O
Cl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compoun
d
R Group
USP7
USP47 IC50
(µM)c
cLogPd
LipEe
IC50
nM)b
(
N
12
43
73
>40
>40
2.9
2.8
4.5
4.3
HO
S
S
O
N
13
HO
Cl
HO
HO
N
1
4
5
9
>40
>5
2.6
2.3
5.4
5.3
S
S
N
1
O
27
N
N
N
N
1
6
7
O
O
25
26
>80
>20
2.7
2.8
4.9
4.8
NH
N
S
S
1
N
18
0.18
>2
2.7
7.0
O
N
S
O
a
c
b
Data represent the mean of n = 2 or more determinations. USP7 biochemical assay.
d
e
USP47 biochemical assay. Calculated using ChemDraw. LipE = p(IC ) – cLogP.
50
ACS Paragon Plus Environment
14

Page 15 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Compound 14 co-crystal structure with USP7, highlighting key interactions
with the binding pocket. The PDB code for 14 bound to USP7 is 6VN6.
Alternative functional groups to the chloro-diol present in compound 14 were
evaluated (Table 2). The pyrimidinone side chain, whose derivatives are present in the
hydroxy-piperidine compounds 1-4, resulted in compound 15 which was found to have
slightly better potency than alcohol 12. Amide or lactam-containing molecules 16 and 17
were not found to have improved potency. However, the addition of a second hydrogen-
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bond acceptor resulted in succinimide 18, a compound with sub-nanomolar biochemical
activity against USP7 and excellent lipophilic efficiency. X-ray co-crystallization
confirmed that this succinimidyl functional group was making both interactions with Phe-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
409 and Gln-297 (Figure 6). No effect was observed on the potency of USP7 inhibition
with pre-incubation of the succinimide-based inhibitor, indicating that the compound
behaves as a reversible inhibitor as suggested by the crystal structure.17
ACS Paragon Plus Environment
16

Page 17 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 6. X-ray co-crystal structure of compound 18 with USP7, highlighting key
interactions with the binding pocket. The PDB code for 18 bound to USP7 is 6VN2.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A Simplified Ether Series
Compound 18 was evaluated in rodent PK and found to have very high clearance
and poor oral exposure, despite reasonable stability in an in vitro hepatocyte stability
assay (Table 3). The low oral exposure was suspected to be in part due to transporter-
mediated efflux resulting from the basic amine moiety of the molecule; however,
compounds without the basic amine motif, which makes multiple key interactions with
Asp-101 and a backbone carbonyl, were orders of magnitude less active against USP7.
Additionally, stability to incubation with hepatocytes was still found to be low for these
compounds. Several isosteric analogs of 18 were prepared where the amide linker was
replaced with a heterocyclic linker in an attempt to improve permeability and stability, a
subset of which are shown in Table 3. Oxazole 19 was found to be less potent than
amide 18 and be less stable to hepatocytes. Isoxazole 20 was found to have good
potency but poor metabolic stability, clearance, and oral exposure. Pyrazole 21 was
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
found to be more stable to hepatocytes, but unfortunately, this did not extend to a
significantly reduced clearance in vivo.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Evaluation of piperazine-amide replacements
N
O
N
S
O
O
R
Cl
Compound
R Group
USP7
IC50
Hepatocyte stability:
IV
PO AUC
(hr*ng/mL)d
% remaining (R / H)c Clearance
(L/hr/kg)d
(
nM)a,b
NH
18
0.18
14% / 37%
11.5
0
N
O
19
13
2% / 14%
N.D.
N.D.
N
N
O
H
N
2
0
1
0.38
1.1
4% / 26%
5.8
4.7
0
0
N
O
H
N
2
42% / 63%
N
N
Me
ACS Paragon Plus Environment
18

Page 19 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
Data represent the mean of n = 2 or more determinations. USP7 biochemical assay.
c% remaining after 60-minute incubation with rat (R) or human (H) hepatocytes.
dSprague-Dawley rats, dose of 0.5 mg/kg IV, 2 mg/kg PO. N.D. = not determined.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Having been unsuccessful at improving PK properties by simple amide
replacement, we were motivated to make more significant modifications to the
dihydrofuran-amide scaffold that had been omnipresent in our inhibitors. Additionally,
these dihydrofuran-containing compounds required relatively lengthy syntheses,
resulting in a slower discovery effort (see Scheme 2). A simplified and metabolically
stable replacement for this bottom portion was desired, but one that maintained the key
contacts with the binding pocket (as highlighted in Figure 6).
It was believed that a simplified structure such as ether-linked pyrrolidine 22
might maintain the desired binding interactions while providing better metabolic stability,
a simpler synthesis, and reduced molecular weight (Figure 7). However, this
modification resulted in a 1000-fold potency loss. It became quite clear that while many
aspects of the dihydrofuran-amide did not interact with the binding pocket, it served an
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
important scaffolding role to help place the basic nitrogen in a productive spatial
orientation. Computational modeling indicated that the lowest energy conformation of
dihydrofuran-amide 18 is one in which the piperazine ring is oriented upward, syn- to
the pyridine ring system – the same conformation that is adopted upon binding to USP7
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(see Figure 6). Conversely, the ground state of simplified system 22 has the basic
nitrogen directed anti- to the pyridine ring system. This is a non-productive conformation
and a significant energy penalty is required to achieve proper amine placement in the
binding pocket. To help overcome this unproductive ground state conformation, A-(1,3)
strain was leveraged by installing a methyl group ortho- to the ether linkage (as in
pyrrolidine 23). This raises the energy of the global minimum conformation of 23 relative
to its bound conformation, increasing the overall free energy of binding with USP7. We
were pleased to see that this simple modification results in a greater than 100-fold
potency gain. A USP7 co-crystal structure of 23 was obtained and showed the
pyrrolidine nitrogen making the same contacts with Asp-295 and Met-292 as the
previous piperazine-amide (Figure 8). Gratifyingly, compound 23 also demonstrated
moderate clearance and non-zero oral bioavailability.
ACS Paragon Plus Environment
20

Page 21 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
N
N
H
N
O
O
O
NH
N
S
N
S
N
S
F
O
O
O
H
O
O
O
N
O
F
NH
Cl
Cl
Cl
Me
18
22
USP7 IC50: 207 nM
23
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
USP7 IC50: 0.17 nM
USP7 IC50: 1.2 nM
Figure 7. A scaffold switch from the benzofuran-amide to a simplified ether is productive
with the inclusion of an ortho-methyl group.
Figure 8. X-ray co-crystal structure of ether 23 with USP7, highlighting key interactions
with the binding pocket. The PDB code for 23 bound to USP7 is 6VN3.
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
This structural simplification enabled extensive SAR exploration of the basic
amine portion of these molecules. Analogues were assessed based on their
biochemical USP7 inhibition as well as in a cellular assay to measure elevated p53
levels.18 This was accomplished by utilizing a p53 response element-driven luciferase
reporter assay in engineered RKO cells. In addition to pyrrolidines such as 23, many
other ring systems were well tolerated with respect to USP7 potency (Table 4). 3- and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4-piperidinyl compounds such as 24 and 25 were both tolerated, as were azetidines
such as 26. Even exocyclic amines such as cyclobutane 27 showed potent USP7
inhibition. However, most of these compounds still demonstrated only moderate
clearance and bioavailability. After screening many substituents on these ring systems,
it was found that a single methyl group afforded a drastic change in the PK profile of the
4-substituted-piperidine set of compounds. Relative to piperidine 25, methyl-piperidine
28 had 5-fold lower rat IV clearance and 20-fold greater PO AUC. The methylated
compound 28 was also more potent in line with its greater lipophilicity relative to the
parent compound 25.
ACS Paragon Plus Environment
22

Page 23 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. Highlights of compounds in the simplified ether series
N
O
N
S
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
R
Cl
Me
Compoun
d
R Group
USP7
IC50
p53 EC50
(nM)a,c
cLogPd
LipEe
IV
PO AUC
(hr*ng/mL)f
clearance
(L/hr/kg)f
nM)a,b
(
2
3
4
1.2
560
180
3.5
3.7
5.4
4.6
2.1
4.9
130
170
NH
F
2
4.6
NH
25
26
27
8.0
1.5
3.4
400
320
150
3.1
3.4
3.2
5.0
5.4
5.3
0.33
N.D.
N.D.
44
NH
NH
N.D.
N.D.
F
NH2
Me
2
8
2.1
150
3.6
5.1
0.07
860
NH
a
c
b
Data represent the mean of n = 2 or more determinations. USP7 biochemical assay.
d
e
p53-reporter gene assay in RKO cells. Calculated using ChemDraw. LipE = p(IC ) –
5
0
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
f
cLogP. Sprague-Dawley rats, dose of 0.5 mg/kg IV, 2 mg/kg PO. N.D. = not
determined.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Re-examination of the Succinimide Side Chain
Having established this new piperidine ether as providing an improved PK profile
relative to the benzofuran series, the succinimide side chain was re-evaluated. Although
the succinimide was a key potency-driver, this moiety was also chemically labile and
slowly hydrolyzed under aqueous conditions. Fortunately, substitution on the
succinimide backbone with a cyclopropyl ring mitigated hydrolysis while maintaining
potency (29, Table 5). Further succinimide substitution with a gem-dimethylcyclopropyl
ring resulted in a compound with greater biochemical potency in line with an increase in
lipophilicity (30), as well as with greatly enhanced chemical stability. Replacement of the
succinimide for a substituted uracil as in 31 also resulted in a potent compound;
however, the PK properties for uracil-containing compounds were sub-optimal, in part
due to greater efflux.
Table 5. Cyclic imide SAR
ACS Paragon Plus Environment
24

Page 25 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
R
S
Me
O
NH
Cl
Me
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compoun
d
R Group
USP7
IC50
(nM)a,b
p53 EC50
(nM)a,c
cLogPd
LipEe
IV
PO AUC
(hr*ng/mL)f
clearance
(L/hr/kg)f
2
8
9
2.1
1.1
150
170
3.6
3.8
5.1
5.1
0.07
860
O
N
N
O
O
2
N.D.
N.D.
O
30
O
0.18
0.26
65
42
5.1
4.2
4.6
5.4
0.11
1.0
4700
52
N
O
Me
O
Me
31
N
N
O
CF3
a
c
b
Data represent the mean of n = 2 or more determinations. USP7 biochemical assay.
d
e
p53-reporter gene assay in RKO cells. Calculated using ChemDraw. LipE = p(IC ) –
50
f
cLogP. Sprague-Dawley rats, dose of 0.5 mg/kg IV, 2 mg/kg PO. N.D. = not
determined.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Out of the above compounds, gem-dimethylcyclopropylsuccinimide 30 had the
best combination of potency and PK properties. Additionally, 30 was found to be potent
against the MM.1S and H526 cell lines. It is well-established that the MM.1S cell line, a
p53-wildtype multiple myeloma cell line, is sensitive to USP7 inhibition and other
mechanisms of p53 activation.7,9 Additionally, we have recently discovered that certain
p53-mutant cancer cell lines are also sensitive to USP7 inhibitors, including H526, a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
15
small-cell lung cancer cell line. This activity is presumably mediated through a different
mechanism other than p53 elevation since the p53 pathway is expected to be non-
functional in H526 cells.19 Furthermore, target engagement was demonstrated for
compound 30 in MM.1S cells, highlighting that our inhibitors are capable of preventing
USP7 from binding and conjugating to an active-site ubiquitin probe.20
Compound 30 possessed significant cytotoxic activity against both the MM.1S
and H526 cell lines, demonstrating a CC50 of 140 nM and 440 nM, respectively.
Additionally, 30 was tested against a wide variety of DUBs and was found to have
exquisite selectivity for USP7. Compound 30 showed an IC50 >10 µM for all the >40
ACS Paragon Plus Environment
26

Page 27 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DUBs that were assayed, corresponding to a >50,000-fold selectivity for USP7 over the
other DUBs.21 Unfortunately, 30 also exhibited some general cytotoxicity, observed by
sharply reduced ATP levels at 3.4 µM in RKO cells, suggestive of physicochemical
toxicity. Additionally, substantial hERG inhibition was observed with this compound in a
simple hERG binding assay (0.4 µM) as well as mild CYP3A4 inhibition (6.2 µM). These
sub-optimal properties along with the above-mentioned cytotoxicity encouraged further
SAR exploration into a somewhat different chemical space from ether-linked compound
30.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Carbon-Linked Compounds Replace the Ether Series
The ether oxygen atom had been omnipresent in our USP7 inhibitors up to this
point (first in the benzofurans and later in the simplified ethers) because it was assumed
that the buried hydrogen bond to Tyr-465 (as observed in co-crystal structures with
USP7) was critical for binding activity. The importance of this interaction, however, had
never been experimentally established. Free Energy Perturbation (FEP+) calculations22
had been successfully used throughout this program to assess different modifications in
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 85
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
order to help prioritize compounds, especially in occasions involving challenging or
lengthy syntheses. Using FEP+, the relative binding free energy was calculated for
compounds where the ether oxygen had been replaced with a nitrogen or carbon atom.
Much to our surprise, carbon-linked analogues were predicted to be equipotent to the
standard ether-linked compounds; nitrogen-linked analogues were predicted to be even
more potent. These calculations were validated when aniline 33 and methylene-
analogue 34 were synthesized and assayed (Figure 9). Compared to ether-linked
compound 32, carbon-linked compound 34 was slightly more potent. Nitrogen-linked
compound 33 displayed marked improvements in potency, especially in the cellular p53
assay. This phenomenon was rationalized by re-analyzing the USP7 crystal structure,
which revealed a small hole on the side of the protein opposite to the ligand binding site
which enables Tyr-465 to be solvent exposed.23 This could help make up for loss of
hydrogen bond when moving to the carbon-linker and enable additional hydrogen
bonding when Tyr-465 switches from an H-bond donor to an H-bond acceptor when
bound to aniline 33.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
28

Page 29 of 85
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
N
N
O
O
O
N
S
N
S
N
S
H
N
O
O
O
O
NH
NH
NH
Me
Me
Me
Me
Cl
Me
Me
Cl
Me
Me
Cl
Me
32
33
34
USP7: 2.4 nM
p53: 250 nM
USP7: 0.78 nM
p53: 33 nM
USP7: 1.3 nM
p53: 130 nM
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 9. Biochemical and cellular potencies with oxygen-, nitrogen-, and carbon-linked
compounds.
SAR of the new nitrogen-linked series showed that while these compounds were
quite potent, they had poor PK properties, likely in part due to an additional hydrogen-
bond donor. Additionally, this series was undesirable because it introduced an aniline
toxicophore. The carbon-linked series was found to be more promising – by removing a
heteroatom from the linker position, it enabled the introduction of a heteroatom into the
piperidine ring to forge either piperazines or morpholines. These saturated heterocycles
would result in less basic compounds and help mitigate the hERG inhibition seen in
ether-linked compound 30.
Initially, simple piperazine 35 was not a very potent compound (Table 6).
However, the addition of methyl groups to the 2- and 6-positions of the piperazine ring
ACS Paragon Plus Environment
29