Discovery of Small-Molecule Inhibitors of Ubiquitin Specific Protease 7 (USP7) Using Integrated NMR and in Silico Techniques

Article abstract of DOI:10.1021/acs.jmedchem.7b01293

USP7 is a deubiquitinase implicated in destabilizing the tumor suppressor p53, and for this reason it has gained increasing attention as a potential oncology target for small molecule inhibitors. Herein we describe the biophysical, biochemical, and comput

Full text of DOI:10.1021/acs.jmedchem.7b01293

Subscriber access provided by READING UNIV
Article
Discovery of Small-Molecule Inhibitors of Ubiquitin Specific
Protease 7 (USP7) Using Integrated NMR and In-Silico Techniques
Paola Di Lello, Richard Pastor, Jeremy M. Murray, Robert A. Blake, Frederick Cohen,
Terry D Crawford, Joy Drobnick, Jason Drummond, Lorna Kategaya, Tracy Kleinheinz,
Till Maurer, Lionel Rouge, Xianrui Zhao, Ingrid Wertz, Chudi Ndubaku, and Vickie Tsui
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01293 • Publication Date (Web): 22 Nov 2017
Downloaded from http://pubs.acs.org on November 23, 2017
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.
Journal of Medicinal Chemistry 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 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of SmallꢀMolecule Inhibitors of Ubiquitin
Specific Protease 7 (USP7) Using Integrated NMR
and InꢀSilico Techniques
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
†
‡*
†‡*
†
†
†
Paola Di Lello, Richard Pastor, Jeremy M. Murray, Robert A. Blake, Frederick Cohen,
†
†
†
†
†
Terry D. Crawford, Joy Drobnick, Jason Drummond, Lorna Kategaya, Tracy Kleinheinz,
†
†
†
†
†
†*
Till Maurer, Lionel Rougé, Xianrui Zhao, Ingrid Wertz, Chudi Ndubaku, Vickie Tsui
†
Genentech, Inc. 1 DNA Way, South San Francisco, California 94080, United States
ABSTRACT
USP7 is a deubiquitinase implicated in destabilizing the tumor suppressor p53 and for this reason
it has gained increasing attention as a potential oncology target for small molecule inhibitors.
Herein we describe the biophysical, biochemical and computational approaches that led to the
1
identification of 4ꢀ(2ꢀaminopyridinꢀ3ꢀyl)ꢀphenol compounds described by Kategaya et al. as
specific inhibitors of USP7. Fragment based lead discovery (FBLD) by NMR combined with
virtual screening and reꢀmining of biochemical highꢀthroughput screening (HTS) hits led to the
discovery of a series of ligands that bind in the “palm” region of the catalytic domain of USP7
and inhibit its catalytic activity. These ligands were then optimized by structureꢀbased design to
yield cellꢀactive molecules with reasonable physical properties. This discovery process not only
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 60
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
involved multiple techniques working in concert, but also illustrated a unique way in which hits
from orthogonal screening approaches complemented each other for lead identification.
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
Ubiquitination is a key form of postꢀtranslational modification. Ubiquitination largely targets
2
protein substrates for degradation by the 26S proteasome in a tightly regulated manner . As
such, ubiquitination involves substrateꢀspecific ubiquitin ligases (E3) that catalyze the formation
of an isopeptide bond between the carboxylꢀterminus of ubiquitin and the lysine sideꢀchain of the
3
target protein . A ubiquitin monomer can be attached via this mechanism to a protein substrate,
and multiple ubiquitins can be linked to each other resulting in polyubiquitination. Depending
on the type of linkage between the ubiquitin molecules in these chains, polyubiquitination can
4
ꢀ5
serve as a signal for degradation of the protein substrate
.
Opposing the function of these E3 are deubiquitinating enzymes, or DUBs. As the name
suggests, a DUB removes ubiquitin from a protein substrate, another ubiquitin within a
polyubiquitin chain, or a ubiquitin precursor. There are around 100 known human DUBs,
consisting of five major families: UbiquitinꢀSpecific Proteases (USPs), Ubiquitin Carboxylꢀ
terminal Hydrolases (UCHs), Ovarian Tumor proteases (OTUs), MachadoꢀJoseph Disease
protein domain proteases (MJDs) and JAMM/MPN domainꢀassociated metallopeptidases
6
(JAMMs) . In addition, two families of DUBs have been recently discovered, the Monocyte
7ꢀ8
Chemotactic ProteinꢀInduced Proteins (MCPIPs)
and a new class of cysteine proteases
9
belonging to the MINDY family (MIUꢀcontaining novel DUB family) . USPs are the largest
ACS Paragon Plus Environment
2

Page 3 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
family of DUBs and members of this family are cysteine proteases that rely on a “catalytic
triad” of cysteine, aspartic/glutamic acid and histidine to hydrolyze the isopeptide bond.
Several USPs have been reported to play important biological roles that make them attractive
1
0ꢀ11
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
as therapeutic targets
. An example is USP7 or herpes virus associated USP (HAUSP),
which is considered a potential oncology target due to its role in the destabilization of the tumor
1
2
suppressor p53 . As a transcription activator, p53 modulates the expression of genes like p21,
1
3
Bax, or Puma that control cell cycle arrest or apoptosis in response to various cellular stresses .
However, in the majority of human cancers p53 is inactivated by either mutations in p53 itself or
1
4ꢀ15
alterations in the pathways that regulate its levels
. Deubiquitination of MDM2, a substrate of
USP7 and itself a ligase for p53, is one of the mechanisms by which wildꢀtype p53 is inactivated.
Therefore, inhibition of USP7 can tip the balance towards ubiquitinated MDM2, thus triggering
its degradation and reactivating the p53 pathway in cancer cells that retain wildꢀtype p53.
We recently reported a new class of highly selective USP7 inhibitors, 28 (GNEꢀ6640) and 27
(
GNEꢀ6776), which modulated the enzymatic activity of USP7 by binding to a novel functional
1
site within the “palm” region of the enzyme and interfering with substrate binding . In this
manuscript we describe the structureꢀbased drug discovery strategies that led to the identification
of the aforementioned new functional site on USP7 and to the discovery of the “palm”
compounds. The discovery process that led to this new class of USP7 inhibitors involved the
interplay between biochemical, inꢀsilico and biophysical techniques, including Nuclear Magnetic
Resonance (NMR) and crystallography. In addition, hits from multiple small molecule libraries
(
fragment, HTS and sample management libraries) were combined to generate this novel series.
The detailed account of how the techniques and libraries complemented each other, as well as the
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 60
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
structureꢀbased optimization leading to molecules with improved biochemical potency, and
observable cellular activity will be presented.
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
RESULTS
Discovery, biophysical characterization and structure-based optimization of the
oxadiazole series.
We have applied different screening approaches (a fragment screen by NMR and a full HTS
campaign) to search for small molecule ligands that would specifically bind and inhibit the
1
catalytic domain of USP7 . In particular, a fragment screen by NMR led to the discovery of a
series of small molecules that bind in the active site near the catalytic cysteine (Cys223), and
were further developed into moderately active inhibitors of the enzymatic activity (unpublished
results). In a parallel approach we also used computational methods to look for additional
scaffolds that could bind in the enzyme active site and potentially block its activity. Compound 1
(Table 1), a fragmentꢀscreen hit that binds in the enzyme active site, was used as the query
molecule to carry out ligandꢀbased virtual screening against compounds in the Genentech
internal sample management library. The 500 topꢀscoring hits from the virtual screen were then
analyzed by visual inspection, leading to the selection of twentyꢀone compounds. To verify that
these twentyꢀone compounds bound to the catalytic domain of USP7 (USP7ꢀCD) we tested them
by NMR using Saturation Transfer Difference (STD) experiments. Although many of these
molecules did not confirm as ligands, a smaller subset of twelve compounds showed binding to
USP7ꢀCD. These twelve compounds were then evaluated for biochemical activity and, to further
characterize their binding site, they were also tested in NMR titration experiments with labeled
USP7ꢀCD. Addition of each of these hits to labeled USP7ꢀCD induced chemical shifts
ACS Paragon Plus Environment
4

Page 5 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
15
perturbations and/or line broadening of signals in the twoꢀdimensional [ H ꢀ N]ꢀTROSY
spectrum of labeled USP7ꢀCD, confirming an interaction with the protein. However, of the
twelve compounds tested with labeled USP7ꢀCD, three induced unspecific signals line
broadening suggesting possible protein aggregation, whereas five appeared to bind in more than
one site; in fact they caused changes in the resonances of residues located in the active site as
well as in other regions of the protein. The remaining four compounds (2ꢀ5) induced chemical
shifts perturbations only in a select subset of NMR signals, suggesting they interacted with
USP7ꢀCD at a specific site (Figure 1B). These four hits, all of which contained an oxadiazole
core with a 4ꢀphenol on one side and piperidine or pyrrolidine on the other side, were the only
ones of the twelve tested that had detectable biochemical activities (Table 1).
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
Surprisingly, the chemical shifts perturbations that 2 and its three analogs (3ꢀ5) induced on
labeled USP7ꢀCD (Figure 1B) also involved a different subset of peaks from that observed for
the active site ligand 1 (Figure 1A), thus pointing to a different binding site on the protein
surface. Nevertheless, the KD values of 2 and 3, estimated by NMR to be in the range 12ꢀ27 µM
for USP7ꢀCD, were within 5 fold of the biochemical IC ’s determined for full length USP7
5
0
(
Table 1).
In the case of the active site ligand 1 the amino acids showing the largest chemical shifts
perturbations upon addition of the ligand to labeled USP7ꢀCD were Met292, Gln293, His294,
Asp295 (located in the switching loop), Met410, Phe411, Asp412, Asn418, Ile419, Lys420,
Ile421 (located in the β0ꢀβ0’ sheet) and Gly458, Asp459, Asn460, Gly462, Gly463, His464
(
located in the loop connecting strands β10 and β11). When mapped onto the structure of apo
USP7ꢀCD, these amino acids clearly defined a ligand binding site (Figure 1C) that overlapped
with the catalytic site. On the contrary in the case of 2 the largest chemical shifts perturbations
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 60
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
and/or signals lineꢀbroadening were observed for a different set of amino acids, namely Ile320,
Lys322 (helix α6), Gly326, Met328 (strand β1), Ile350, Gln351, Leu352, Ser353 (strand β3),
Tyr347, Tyr348, Asp349 (located in the loop connecting strands β2 and β3), His403, Met407
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
(strand β8), Asp295, Gln297, Glu298, Cys300, Arg301, Leu304, Asp305, Val307, Glu308 (helix
α5), and Lys281, Gly284, Trp285. When mapped onto the structure of apo USP7ꢀCD, these
amino acids clustered within the “palm site” of USP7ꢀCD, in a region adjacent to the catalytic
site (Figure 1D), thus revealing a novel functional site.
The location of the binding site identified by the NMR chemical shift perturbation studies was
consistent with the crystal structure of the USP7ꢀCD/2 complex that was subsequently
determined. The structure also revealed several critical interactions between ligand and protein.
The phenol ring was buried in a hydrophobic pocket with an edgeꢀtoꢀface interaction with
Phe324 and the phenol hydroxyl engaged in a hydrogen bond interaction with His403 (Figure 2).
While the oxadiazole moiety was partially solvent exposed, it was involved in a faceꢀtoꢀface πꢀ
stacking interaction with Tyr348. The piperidine was, for the most part solvent exposed, and
based on the electron density, was not involved in any interactions with the protein. SAR on the
piperidine (Table 1) indicated that it made minimal contribution to activity.
To follow up on 3, analogues were designed and prioritized based on the crystal structure
(Table 2). Replacement of the oxadiazole core with a 2,4 substituted oxazole (6) resulted in no
change in activity. However, employing a 2,5ꢀsubstituted oxazole (7) led to a 3ꢀfold potency
increase. For 7 the oxygen (rather than the nitrogen in 6) in the oxazole is closer in space to the
methylene part of the Arg301 sidechain and this potentially leads to a better Van der Waals
interaction because of the smaller negative dipole of the oxygen relative to the nitrogen. An
ACS Paragon Plus Environment
6

Page 7 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
additional 2ꢀfold increase in activity was observed when phenol was replaced by 3ꢀfluorophenol,
likely due to increased phenol acidity or slight changes in conformational preference (Figure S1).
Discovery and characterization of the aminopyridine series.
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 discovered that the “palm” region of USP7ꢀCD contains an additional functional site
that could be exploited to modulate the enzymatic activity through small molecules binding, we
applied an NMR driven scaffoldꢀhopping strategy to search for structurally novel compounds
that could bind to this site and potentially be developed into selective USP7 inhibitors.
Instead of generating structurally novel compounds through a computational approach to
scaffold hopping, we reꢀmined the hits from our original NMR fragment screen, searching for
ligands that had a binding fingerprint on labeledꢀUSP7 similar to that observed for 2 and its
analogs. Using this approach we were able to identify several heterocyclicꢀsubstituted phenols
(i.e. aminopyridineꢀphenols and aminopyrazineꢀphenols, Figure 3A) that showed perturbation
maps remarkably similar to those observed for 2 (Figure 3BꢀC), suggesting that they bind USP7ꢀ
CD in the same site and possibly with the same binding pose as compounds of the oxazole series.
Interestingly, the aminopyridineꢀphenol motif found in some of the hits from the fragment
screen was also present in 14 and 15, two of the 76 hits from the highꢀthroughput screening
(HTS) (Table 3) of Genentech proprietary library of ~2 million compounds. However, these two
HTS hits had been previously deꢀprioritized for reꢀconfirmation in favor of hits that were
1
commercially available and significantly more potent .
Efforts that led to the reꢀsynthesis of compounds 14 and 15 also produced compound 16 as a
synthetic byꢀproduct and all three compounds demonstrated micromolar biochemical activities
(Table 3). A coꢀcrystal structure of 16 was obtained and it showed that the compound bound in a
similar mode to that of 3, as both occupied the same “palm” binding site, where significant
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 60
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
affinity was gained by engaging His403 and Phe324 with a phenol moiety (Figure 4A).
However, electron density for the other components of the ligand were not wellꢀdefined, leaving
its exact binding mode undetermined. Instead, four binding modes were possible based on the
observed electron density as shown in Figure 5, with each binding mode having a distinct
orientation of the solventꢀexposed phenol and the amino moiety. Modeling and occupancy
refinement of multiple possible binding orientations also revealed an unoccupied hydrophobic
pocket, proximal to the ligand, which was targeted for affinity optimization (Figure 4B). Initial
medicinal chemistry efforts focused on both affinity optimization and elucidation of the
inhibitory binding mode.
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
Attempts to optimize 16 began by substituting at the 2, 4 and 6 positions of the central pyridine
ring (Table 4). Deletion of the 2ꢀamino moiety (17) led to a complete loss in activity, while
addition of a 6ꢀamino group (18) showed comparable activity. Appending an ethyl at C4 (20)
led to a fiveꢀfold activity increase, whereas propyl was too long for the small pocket (21). C4ꢀ
isopropyl was fiveꢀfold less active than C4ꢀethyl (22 vs. 20) and C4ꢀcyano (23) led to complete
loss of activity. This preference for small, unbranched hydrophobic groups in the C4 position
was consistent with two of the possible ligand binding modes (Figure 5, C and D).
C3 and C5 analogue synthesis was undertaken, where one of the goals was to identify which
phenol moiety made the critical His403 interaction (Table 5). The resulting SAR is shown in
Table 5 and unexpectedly, the SAR at both positions was very similar. At both positions,
phenyl substitution led to a moderate reduction in activity, cyanophenyl substitution was well
tolerated, and indazole substitution improved activity by several fold, resulting in the most active
compounds from this series. In addition, the observation that compounds without a phenol in C3
or C5 (Table S1) were significantly less potent demonstrated that a phenol was required for
ACS Paragon Plus Environment
8

Page 9 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
detectable activity. Taken together, the results suggested the latter two conformations in Figure
5
C or D could be adopted to place phenol in the His403 pocket.
Cell-based assay results.
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 representative set of our most biochemically potent and chemically diverse compounds was
tested in a MDM2 MSD cellꢀbased assay with SJSA cells, which are osteosarcoma cells with
1
6
high level of MDM2 expression , to study their effects on levels of ubiquitinated MDM2 vs the
total amount of MDM2 in a doseꢀresponse fashion (Table 6). The rank order of activities
corresponded well with biochemical potency. Compound 28 demonstrated subꢀmicromolar
activity and was the most potent analog of this group, consistent with its relative biochemical
activity. Compound 27 demonstrated single digit micromolar activity, while 18 and 31 both
were an order of magnitude less active. The negative control 29 had EC >50 µM.
5
0
Compounds 18, 27 and 28 were further tested in EOLꢀ1 cells alongside 29 as a negative
control (NC). These compounds were selected based on their EC values in the MDM2 MSD
5
0
assay. The effects of these compounds on the levels of proteins downstream of the
USP7/MDM2 pathway (Figure 6A) were examined at 1.3 µM and 5 µM by Western Blots
(Figure 6B). Using tubulin as a loading control, the normalized levels of p53 and p21 were
quantitatively compared to DMSO and plotted in Figures 6C and 6D, respectively. As expected,
the negative control had minimal effect on the levels of p53 and p21, whereas compounds 27 and
2
8 showed a significant increase in p53 levels. Compounds 27, 28, and to a lesser extent 18, also
upꢀregulated p21.
Furthermore, the effects of USP7 inhibitors on the viability and caspase activity of EOLꢀ1 cells
were examined (Fig. 7AꢀB) using imaging techniques. The percent confluence measured the
area of each well occupied by cells, and doseꢀdependent decrease in cell viability was observed
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 60
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
over 48 hours when treated with each of the active inhibitors while treatment with the negative
control did not have significant effects. Caspase activity was increased and the effects were
doseꢀdependent following treatment with 18 and 28 over 48 hours, whereas all doses resulted in
minimal effects on caspase activation for the negative control arm. This doseꢀdependent effect
on caspase activity was lower for 27, leading to the hypothesis that 28 decreases cell viability via
both apoptosis and cell cycle arrest (Figure 6A), whereas 27 mainly causes cell cycle arrest. Cell
cycle arrest is generally triggered first in response to an insult, followed by induction of
apoptosis in case the cell damage cannot be repaired. Therefore, treatment with a more potent
compound like 28 can lead to apoptosis in addition to cell cycle arrest, whereas treatment with a
less potent compound like 27 would mainly trigger cell cycle arrest. For compound 18, which is
the least potent in inhibiting USP7 (Table 4 and Table 6) and in modulating the levels of p53 and
p21 (Figure 6CꢀD), the observed increase in caspase activity (Figure 7B) and the resulting
decrease in cell viability (Figure 7A) could be due to offꢀtarget effects.
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
This hypothesis can also explain the CellTiterꢀGlo results (Figure 7C) measuring ATP levels at
the 48ꢀhour end point. Unlike the negative control, all three inhibitors led to decreased ATP
levels as their concentrations were increased. Compound 28 was more potent, while the effects
of 27 were seen at higher inhibitor concentrations. As expected, the negative control had
minimal effects.
17ꢀ19
There are several reports of closely related analogs being kinase inhibitors
potential contributions of offꢀtarget kinase activity to the cellular phenotypes that were observed,
7 and 28 were profiled against a kinase panel consisting of 219 kinases. We observed only
modest inhibition, with IC values in the low micromolar to subꢀmicromolar range, against 6
and to assess
2
5
0
kinases (data not shown). Most importantly, 27 and 28, our most potent compounds, as well as
ACS Paragon Plus Environment
10

Page 11 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the negative control 29 were equipotent against all of these kinases, suggesting that kinaseꢀ
inhibitory effects did not contribute to effects on cellular viability. This conclusion is further
1
supported by published studies demonstrating that 28 and 27 are onꢀpathway, USP7ꢀselective
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
inhibitors. Both compounds reduce cell viability in wild type HCTꢀ116 cells but not in USP7ꢀ
null HCTꢀ116 cells indicating that USP7 expression is required for cellular activity. Furthermore
2
7, 28, and 29 were all profiled in 3ꢀ and 5ꢀday viability assays in large panels of cell lines.
While 27 and 28 significantly reduced viability of a number of cell lines, 29 was inactive in all
cell lines evaluated. Collectively, these data indicate that the modest in vitro kinase inhibition
observed for our USP7 inhibitors does not translate to cellular activity.”
Chemistry
The oxadiazoles and oxazoles shown in Tables 1 and 2 were prepared using a suite of
heterocycleꢀforming reactions in both a convergent and linear manner (Schemes 1 – 3).
Condensations between acid and hydroxyamidine components promoted by carbonyl
diimidazole (CDI) led to the described oxadiazoles. The acid components were commercially
sourced whereas the hydroxyamidine coupling partners were formed from nitrile precursors.
The 2,4 substituted oxazoles were generated from convergent couplings between
corresponding commercially available αꢀbromoꢀketones and amide monomers. On the other
hand, the 2,5 substituted oxazoles resulted from a linear sequence which began with addition of
nitromethane to NꢀBocꢀpiperidine aldehyde to afford the 2ꢀnitroꢀalcohol. Nitro reduction, amine
acylation, and alcohol oxidation affords the βꢀacylaminoꢀketone which was then cyclized under
Burgess reagent promoted conditions to form the oxazole ring.
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 60
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
Preparation of the bisꢀphenol substituted aminopyridines shown in Table 4 utilized a bisꢀ
bromination with NBS followed by a bisꢀSuzuki coupling with 4ꢀanisole boronic acid on the
corresponding pyridine precursors (Scheme 4). Methoxy demethylation with BBr then affords
3
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
1
7 – 23.
Synthesis of both the C3 and C5 phenol replacements in the aminopyridine core both involved
conducting Suzuki couplings on the corresponding bromoꢀaminopyridine lateꢀstage
intermediates 50 and 55 (Schemes 5 and 6).
Synthesis of 50 began with bisꢀbromination of commercially available 4ꢀethylꢀaminopyridine
followed by selective debromination at the more labile C5 position to afford 47. Suzuki coupling
to install a 4ꢀmethoxyꢀphenyl group at C3 followed by C5 bromination and methoxy
demthylation with HBr affords 50.
Synthesis of 55 began with selective C5 bromination of 4ꢀethylꢀaminopyridine with NBS.
Installation of the 4ꢀMeOꢀPhenyl group by Suzuki coupling followed by C3 bromination and
methoxydemthylation then affords 55.
DISCUSSION AND CONCLUSIONS
Herein, we describe the details of how the USP7 tool compounds 27 and 28 reported in
1
Kategaya et al. were discovered. These are the first small molecule allosteric inhibitors of any
USPs that have been fully characterized using biophysical and biochemical methods as well as
cellꢀbased assays. The lack of comprehensive data in past publications represents both the
20
novelty and challenge of this target class , and we could show that by combining experimental
and computational methods with a variety of chemical libraries we were able to discover these
“palm” site binding series.
ACS Paragon Plus Environment
12

Page 13 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NMR played a significant role in the discovery of lead matter, and it is no surprise that our
lead matter came from a biophysical screen, since a cysteine protease like USP7 could turn up
many reactive false positives in a biochemical HTS campaign. In fact, the HTS hits that were
followed up by reꢀmining efforts described in this publication progressed farther than other HTS
series. Most of the biochemical HTS hits failed to confirm upon reꢀsynthesis, did not show
specific activity to the catalytic domain, or were shown to be potential aggregators.
Furthermore, none of the hits from the biochemical HTS were successful in coꢀcrystallization
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
1
studies, hindering rational structureꢀbased design .
Our discovery process started with the application of a shapeꢀbased virtual screen using the
activeꢀsite binder as a query molecule against compounds in the Genentech internal sample
management library. Combining this virtual screen with a comprehensive biophysical and
biochemical characterization of the virtual hits was a critical step in the discovery process. It
allowed us to identify a structurally unique chemical series, the oxadiazole series, and most
importantly it unveiled the presence of a second functional site, the “palm” site, besides the
catalytic site. The subsequent approach of reꢀmining the NMR fragment hit set for additional
scaffolds with similar binding fingerprints to the oxadiazole compounds but with different
chemical structures was key in turning out several hits containing the aminopyridine and
aminopyrazine substructures. Even though these fragment hits were not potent enough to
observe biochemical activity, taking the chemical moiety back to reꢀexamine the deprioritized
HTS hits led to the successful discovery of our most potent series, the aminopyridine series.
Not only were crystal structures important in the lead discovery stage, they were also essential
for optimization of the aminopyridine series. Even though the electron densities provided
multiple possible binding modes, medicinal chemistry optimization helped to narrow down the
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 60
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
binding modes to two possibilities. During this process, the ethyl group was installed in a small
hydrophobic pocket in the 4ꢀposition of the aminopyridine to boost potency, and unsubstituted
indazoles at either the C3 or C5 position of the aminopyridine were found to be most active,
when coupled with a phenol on the other side. Compounds with measurable cellular EC were
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
5
0
1
obtained, and they had excellent DUB selectivity as described in Kategaya et al. . Furthermore,
compounds were shown to upregulate p21 and p53 in EOLꢀ1 cells in a doseꢀdependent manner,
in agreement with the pathway biology of inhibiting USP7. Increase in caspase activity and
decreased cell viability were also observed, consistent with the expected outcome of USP7
inhibition.
In order to progress this series as a clinical candidate, the key challenges will be to improve
cellular potency and plasma protein binding properties while maintaining good
pharmacokinetics. 27 demonstrated good oral bioavailability and halfꢀlife in mice, but only
transiently achieved target plasma exposures, due to its high plasma protein binding and lack of
1
subꢀmicromolar cellular EC . Therefore, improving both these parameters will enhance the
50
ability of this series to achieve efficacious concentrations in vivo. In addition, profiling of this
series in both in vitro and in vivo toxicity assays would be required to identify potential negative
side effects.
Our experience here provides a lesson for identifying lead matter for challenging targets such
as DUBs, in that an integrated approach is crucial and involves multiple biophysical,
biochemical and in silico techniques, as well as a variety of compound libraries, which normally
would be studied separately. The process we described to reꢀmine hits with various techniques
and from different hit sets, as opposed to more traditional drug discovery efforts, can improve
ACS Paragon Plus Environment
14

Page 15 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the chance of identifying potent inhibitors of new classes of enzymes and proteins otherwise
considered undruggable.
EXPERIMENTAL SECTION
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
Ligand-based virtual screening
Ligandꢀbased virtual screening was carried out using fastROCS (OpenEye, Inc.
www.eyesopen.com) as implemented at Genentech. The specifics of the integration of
2
1
fastROCS with a variety of databases have been described . Compound 1 was used as the
query molecule against a proprietary library of ~100,000 compounds in Genentech’s sample
management, and a maximum of 500 topꢀscoring compounds using default parameters were
returned without postꢀclustering. Visual inspection was carried out with consideration of leadꢀ
likeness as well as dry powder availability.
Protein expression and purification for NMR
The catalytic domain of USP7 (USP7ꢀCD), encompassing residues 208ꢀ554, was expressed as
a HisꢀTag fusion protein in E. coli Rosetta 2 (DE3) cells. Unlabeled USP7ꢀCD was produced
22
15
13
2
13
13
according to the protocol reported elsewhere. Uniformly ([ Nꢀ Cꢀ H], δ1[ CH ]ꢀIle, [ CH ]ꢀ
3
3
1
3
Leu/Val and [ CH ]ꢀMet)ꢀlabeled USP7ꢀCD was expressed and purified as previously
3
2
3
described.
Samples for NMR
1
The Genentech fragment library was screened as reported elsewhere .
Samples for binding validation by Saturation Transfer Difference (STD) experiments
contained 250 ꢀM of ligand and 5ꢀM of unlabeled USP7ꢀCD in NMR buffer [PBS (pH 7.4) in
1
00% D O, 50 ꢀM 4,4ꢀDimethylꢀ4ꢀSilapentaneꢀ1ꢀSulfonic acid (DSS).
2
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 60
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
Samples for binding site mapping consisted of 250ꢀ340 ꢀM of labeled USP7ꢀCD in PBS (pH
7
.4), 300 ꢀMꢀ2.5 mM of ligand, 1 mM TCEPꢀd , 10% D O and up to 2.5% DMSOꢀd .
16
2
6
For the determination of the binding constants, each sample consisted of 270 ꢀM of
isotopically labeled USP7ꢀCD in PBS (pH 7.4), 1 mM TCEPꢀd , 10% D O and up to 2.5%
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
16
2
DMSOꢀd . Individual ligands were added to the protein sample stepwise, over a range of
6
increasing concentrations [250ꢀM, 500 ꢀM, 1000 ꢀM, 2000 ꢀM and 4000 ꢀM].
NMR experiments and data analysis
NMR experiments were performed on 500 MHz and 600 MHz spectrometers equipped with
1
.7 mm and 5 mm cryoprobes, respectively.
The backbone resonances assignment for labeled USP7ꢀCD was obtained as previously
2
3
reported . Chemical shifts of the proton, carbon and nitrogen nuclei were referenced externally
to that of DSS at 0 ppm.
24
The fragment library was screened using Saturation Transfer Difference (STD) experiments
run at 280 K. Hits from the primary screen were further characterized by analyzing the ligandꢀ
1
15
25
induced chemical shifts changes of labeled USP7ꢀCD in [ Hꢀ N]ꢀTROSY experiments
STD experiments were processed and analyzed using TOPSPIN 3.0. Data collected for binding
2
6
27
site mapping were processed with NMRPipe/NMRDraw and analyzed with NMRView .
1
15
The dissociation constants (KD) were derived from the H and N chemical shifts changes
1
15
observed in the [ Hꢀ N]ꢀTROSY spectra of isotopically labeled USP7ꢀCD upon addition of
increasing amounts of ligand. Chemical shift changes, measured at various protein/ligand ratios,
were plotted as function of the ligand concentration, and then fitted according to the equation:
ꢄꢅ + ꢆꢇꢈ + ꢆꢉꢈ − ꢊ(ꢄꢋ + ꢆꢇꢈ + ꢆꢉꢈ)ꢌꢍ − 4ꢆꢇꢈꢆꢉꢈ
∆ꢀ = ∆ꢀꢁꢂꢃ
2
ꢆꢇꢈ
ACS Paragon Plus Environment
16

Page 17 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
where ꢁ
δ
is the chemical shift change at various protein/ligand ratios, ꢁδmax is the chemical
shift change at saturation, KD is the dissociation constant, and [L] and [P] are the ligand and
protein concentrations, respectively.
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
Biochemical assays
Biochemical USP7 assays used UbiquitinꢀRho110 as a substrate. The reaction buffer consisted
of 50 mM Tris (pH 7.5), 0.01%(v/v) Triton Xꢀ100, 2.5 mM Dithiothreitol, 0.1% (w/v) bovine
gamma globulin (Sigma cat # G5009ꢀ25G); USP7, fullꢀlength, native CꢀTerminus, 0.2 nM; the
substrate, UbiquitinꢀRho110 (Boston Biochem cat # Uꢀ555), 1 µM. Reactions were carried out
for 1 hour at room temperature, in black 20µL volume polystyrene ProxiPlate 384 F Plus
(
PerkinElmer cat # 6008260).
Test compounds, including a control USP7 inhibitor (Ubꢀaldehyde, Boston Biochem cat # Uꢀ
201) were serially diluted in DMSO, in 384ꢀwell clear Vꢀbottom polypropylene plates (Greiner
cat # 781280). Compounds in DMSO were diluted 10ꢀfold into Reaction Buffer, to achieve 3ꢀ
fold the final desired concentration. The substrate, UbiquitinꢀRho110 (Boston Biochem cat # Uꢀ
555), was prepared at 3 µM (3ꢀfold the final concentration) and 5 µl was dispensed into the
reaction plate. 5 µl of the compounds (diluted in reaction buffer at 3ꢀfold the final concentration)
were transferred to the reaction plate. 5 µl of 0.6 nM USP7 (diluted in reaction buffer at 3ꢀfold
the final concentration) was transferred to the reaction plate to initiate the reaction. After 1 hour
incubation at room temperature the reaction was quenched by the addition of 5 µl of 400 mM
acetic acid. The enzymatic product was measured by quantifying the fluorescence signal of
cleaved Rhodamineꢀ110 using excitation at 485 nm and emission at 535 nm. When preꢀ
incubation of USP7 with compounds was required, the order of addition of reagents was
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 60
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
modified to preꢀmix the compounds with USP7 (with a 1 hour incubation period), prior to the
addition of the substrate and the initiation of the reaction period. Percentage inhibition values
were calculated relative to a no enzyme control and an uninhibited enzyme control. Curve fitting
and IC calculations were carried out using Genedata Screener software.
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
5
0
Crystallization and data collection
Crystals were grown by the hangingꢀdrop method by mixing the USP7 catalytic domain
residues 208–554) at 15 mg/ml with an equal volume of reservoir solution containing 100 mM
(
Tris,ꢀHCl, pH 7.0, and 20% PEG1000 (v/v). Coꢀstructures with compounds were obtained by
soaking crystals with 1 mM of compound overnight. Crystals were cryoprotected with reservoir
solution supplemented with 20% glycerol (v/v) and flash frozen in liquid nitrogen. Data
collection and refinement statistics are detailed in Table S2.
MDM2 MSD cell-based assays
SJSA cells were maintained in RPMI media containing 10% FBS (Sigma Aldrich, cat #F6765)
and 1% GlutaMAX (ThermoFisher, cat #35050061). SJSA cells were seeded at 120,000 cells per
well in 90 µ l of low serum RPMI supplemented with 0.5% FBS and 1% GlutaMAX into a 96
well TCꢀtreated plate (Greiner cat #655090). Cells were allowed to attach to the plate for 2 hours
at 37 °C.
Test compounds were serially diluted in DMSO, in a 96 well clear Vꢀbottom polypropylene
plate (Greiner, cat #651261). 5 µl of compound diluted in DMSO was added to an intermediate
plate containing 95 µl/well of low serum media per well of a 96ꢀwell clear Vꢀbottom
polypropylene plate. 10 µl of intermediate dilution of compound in media were added to cell
plate yielding a 1:200 final dilution of test compound. Cell plates were incubated with compound
at 37 °C with 5% CO overnight for 16 hours.
2
ACS Paragon Plus Environment
18

Page 19 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
After compound incubation 20 µM final concentration of MG132 (Cayman Chemical,
Cat#10012628) was added to cell plate. Plates were incubated at 37 °C for 1 hour.
Lysis buffer was prepared per vendor instructions. 15 µl per well 5x cold lysis buffer was
added to cell plates and incubate on a shaker at 4 °C for 30 mins.
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
The 96ꢀwell MSD plates provided in Ubiquitinated, Total MDM2 MULTIꢀSPOT 96ꢀWell 4
spot Plate Kit (Cat#N45168Bꢀ1; Meso Scale Discovery, Gaithersburg, MD) were prepared by
incubation with 150ꢁL/well of 3% Blocker Buffer A (Meso Scale Discovery Cat# R93BAꢀ4) in
1
3
X MSD Tris Wash Buffer (Meso Scale Discovery Cat# R61TXꢀ1) for 1h. The plate was washed
times with Tris Wash buffer.
100 µl of cellular lysate was transferred to each of the MSD Assay plate.
Cell lysates were incubated in plate for 1 hour at room temperature while shaking (650 rpm) in
the dark.
The MSD plates were washed 3 times with 1×Tris wash buffer (Meso Scale Discovery) then
incubated with 25 ꢁL of 1X SULFOꢀTAG AntiꢀTotal MDM2 Antibody in antibody buffer
containing blockers specified in manufacturer protocol. Antibody was incubated in plates for 1
hour at room temperature while shaking (650 rpm) in the dark.
The MSD plates were washed 3 times with 1x Tris Wash buffer.
1
50 ꢀl microliters of 1×read buffer T with surfactant (Meso Scale Discovery) was added to the
plate and the relative light units (RLU) were recorded in a Sector Imager 6000 (Meso Scale
Discovery). Dose response of compound induced changes to Ratio of ubꢀMDM2/Total MDM2
was fit using Genedata Screener software.
Cell viability assays
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 60
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
EOLꢀ1 cells were maintained in RPMI media containing 10% FBS. 15,000 cells were seeded
in 1ꢀwell of a 96ꢀwell plate (Corning cat # 3904). Cells were treated immediately or the next day
and then monitored for 48h via live imaging using an Incucyte instrument. To monitor caspase
activity, 2 ꢁM CellEvent Caspase 3/7 reagent (Life Technologies cat # C10423) was added to
seeding media. Scans were collected every 3 h, using a 4X objective. Phase contrast was used to
measure cell confluence while green fluorescence was used to measure caspase activity. The
images were analyzed using IncuCyte software (Basic Analysis parameters). All treatments were
done in triplicate. 48 h after imaging, CellTiterꢀGlo reagent (Promega cat # G7570) was added to
each well as an independent measure of number of remaining cells at the end of the experiment.
CTG reagent was added following the vendor protocol.
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
Western blots
6
A total of 5x10 cells were seeded into 1 well of a 12ꢀwell plate and treated with DMSO, 1.3
ꢁM or 5 ꢁM cmpds, for 4 h prior to lysis for Western Blot analysis. Blots were scanned and
analyzed using a Licor technology. Antibodies used: USP7 (Abcam cat # ab84098), MDM2
(Santa Cruz cat # scꢀ965), tubulin (Licor cat # 926ꢀ42211), p53 (Thermo Scientific cat # MS738ꢀ
P1), p21 (Millipore cat # 05ꢀ655).
Synthesis
General Methods. All solvents and reagents were used as obtained. NMR analysis performed
in deuterated solvent on Bruker Avance 400ꢀ or 500ꢀMHz NMR spectrometers. The spectra were
referenced internally to tetramethylsilane (TMS). Chemical shifts are reported in ppm (δ) (in the
NMR description, s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, and br = broad
peak). All coupling constants (J) are reported in Hertz. Mass spectra were measured with a
Finnigan SSQ710C spectrometer using an ESI source coupled to a Waters 600MS high
ACS Paragon Plus Environment
20

Page 21 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
performance liquid chromatography (HPLC) system operating in reverseꢀphase mode with an Xꢀ
bridge Phenyl column of dimensions 150 mm by 2.6 mm, with 5 mm sized particles.
Preparatoryꢀscale silica gel chromatography was performed using mediumꢀpressure liquid
chromatography (MPLC) on a CombiFlash Companion (Teledyne ISCO) with RediSep normal
phase silica gel (35−60 ꢁm) columns and UV detection at 254 nm. Reverseꢀphase (HPLC) was
used to purify compounds as needed by elution from a Phenomenex GeminiꢀNX C18 column
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
(
20.2 × 50 mm, 5 micron) as stationary phase using mobile phase indicated, and operating at a 35
mL/min flow rate on a Waters 3100 massꢀdirected prep instrument. Chemical purities were
95% for all final compounds as assessed by LC/MS analysis. The following analytical method
was used to determine chemical purity of final compounds: HPLCꢀAgilent 1200, water with
.05% TFA, acetonitrile with 0.05% TFA (buffer B), Agilent SBꢀC18, 1.8 mM, 2.1 x 30 mm, 25
>
0
ºC, 3–95% buffer B in 8.5 min, 95% in 2.5 min, 400 mL/min, 220 nm and 254 nm, equipped
with Agilent quadrupole 6140, ESI positive, 90ꢀ1300 amu.
1
Example 1 was obtained from Sigma-Aldrich (catalog # CDS002148). H NMR (400
MHz, DMSOꢀd6) δ 8.87 (s, 2H), 7.31 (td, J = 8.4, 7.2 Hz, 1H), 6.87 – 6.71 (m, 3H), 3.88 (d, J =
6.3 Hz, 2H), 3.28 (dt, J = 12.7, 3.3 Hz, 2H), 2.88 (td, J = 12.7, 3.1 Hz, 2H), 2.05 (dddd, J = 14.2,
8.0, 6.4, 3.0 Hz, 1H), 1.90 (dd, J = 14.2, 3.5 Hz, 2H), 1.57 – 1.41 (m, 2H). LCMS (ESI m/z):
210.1 (M+H).
Synthesis of Compound 2 (Table 1)
4-(3-(1-methylpiperidin-4-yl)-1,2,4-oxadiazol-5-yl)phenol (2). To a solution of compound 3
(400 mg, 1.6 mmol) and 37% HCHO (0.26 mL) in water (5.0 mL) was added AcOH (0.3 mL)
followed by addition of NaBH CN (200 mg, 3.18 mmol). The mixture was stirred at room
3
temperature for 1 h. It was quenched with aq. NaHCO and concentrated to give the crude
3
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 60
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
1
product, which was purified by prepꢀHPLC (FA) to give the desired product (160 mg, 38%). H
NMR (400 MHz, CD OD) δ 8.52 (br, 1H), 7.98 (d, J = 6.8 Hz, 2H), 6.96 (d, J = 6.8 Hz, 2H),
3
3.51ꢀ3.48 (m, 2H), 3.20ꢀ3.18 (m, 3H), 2.85 (s, 3H), 2.37ꢀ2.32 (m, 2H), 2.17ꢀ2.14 (m, 2H). LCMS
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
(ESI m/z): 259.9 (M+H)
4
-(3-(piperidin-4-yl)-1,2,4-oxadiazol-5-yl)phenol (3). To a solution of 35 (345 mg, 1.0
mmol) in EtOAc (3.0 mL) was added HCl/EtOAc (5.0 mL). The formed mixtre was stirred for 1
1
h at room temperature. Solvent was removed to give the desired product (280 mg, 99%). H
NMR (400 MHz, CD OD) δ 7.96 (d, J = 6.8 Hz, 2H), 6.93 (d, J = 6.8 Hz, 2H), 3.49ꢀ3.46 (m,
3
2
H), 3.25ꢀ3.19 (m, 3H), 2.30 (m, 2H), 2.08ꢀ2.06 (m, 2H). LCMS (ESI m/z): 246.4 (M+H)
Compounds 4 and 5 were prepared in a manner similar to compound 2 (Scheme 1)
1
4
-(3-(1-methylpyrrolidin-3-yl)-1,2,4-oxadiazol-5-yl)phenol (4). H NMR (500 MHz, DMSOꢀ
d6) δ 7.93 – 7.89 (m, 2H), 6.97 – 6.92 (m, 2H), 3.55 – 3.47 (m, 1H), 2.94 – 2.89 (m, 1H), 2.69 –
2.61 (m, 2H), 2.58 – 2.53 (m, 1H), 2.30 (s, 3H), 2.26 – 2.17 (m, 1H), 2.12 – 2.03 (m, 1H).
LCMS (ESI m/z): 246.4 (M+H).
1
4
-(3-(pyrrolidin-3-yl)-1,2,4-oxadiazol-5-yl)phenol (5). H NMR (500 MHz, DMSOꢀd6) δ
1
0.47 (s, 1H), 9.22 (s, 1H), 7.99 – 7.89 (m, 2H), 7.03 – 6.93 (m, 2H), 3.78 (p, J = 7.6 Hz, 1H),
.64 (dd, J = 8.2, 11.7 Hz, 1H), 3.41 (dd, J = 7.4, 11.7 Hz, 1H), 3.39 – 3.25 (m, 2H), 2.45 – 2.36
3
(m, 1H), 2.26 – 2.14 (m, 1H). LCMS (ESI m/z): 232.4 (M+H).
Synthesis of Compound 6 (Scheme 2)
4-(4-(piperidin-4-yl)oxazol-2-yl)phenol (6).
A
mixture of tertꢀbutyl 4ꢀ(2ꢀ(4ꢀ
methoxyphenyl)oxazolꢀ4ꢀyl)piperidineꢀ1ꢀcarboxylate (37, 0.1 g, 0.3 mmol) and hydrobromic
acid (2 mL) was heated to 120º C for 16 h under a nitrogen atmosphere. After cooling to room
temperature, the mixture was concentrated in vacuo. The crude residue was purified by reverse
ACS Paragon Plus Environment
22

Page 23 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
phase chromatography (hydrochloric acid) to give the title compound (70 mg, 80%) as a white
1
solid. H NMR (400 MHz, DMSOꢀd ) δ 9.15 (s, 1H), 8.85 (s, 1H), 7.89 (s, 1H), 7.74 (d, J = 8.4
6
Hz, 2H), 6.87 (d, J = 8.4 Hz, 2H), 3.28 ꢀ 3.23 (m, 2H), 3.01 ꢀ 2.91 (m, 2H), 2.85 ꢀ 2.79 (m, 1H),
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
2
.07 ꢀ 2.04 (m, 2H), 1.82 ꢀ 1.72 (m, 2H). LCMS (ESI m/z) 245.4.
-(5-(piperidin-4-yl)oxazol-2-yl)phenol (7). mixture of tertꢀbutyl 4ꢀ(2ꢀ(4ꢀ
4
A
methoxyphenyl)oxazolꢀ5ꢀyl)piperidineꢀ1ꢀcarboxylate (42, 0.4 g, 1.06 mmol) and hydrobromic
acid (10 mL) was heated to 120º C for 6 h under a nitrogen atmosphere. After cooling to room
temperature, the mixture was concentrated in vacuo. The crude residue was purified by reverse
phase chromatography (formic acid) to give the title compound (66 mg, 24%) as a white solid.
1
H NMR (400 MHz, DMSOꢀd6) δ 10.09 (s, 1H), 9.18 (d, J = 9.5 Hz, 1H), 8.87 (d, J = 10.7 Hz,
1
H), 7.91 (d, J = 1.0 Hz, 1H), 7.82 – 7.73 (m, 2H), 6.94 – 6.86 (m, 2H), 3.29 (dt, J = 12.8, 3.1
Hz, 2H), 2.99 (dtd, J = 12.7, 9.9, 2.7 Hz, 2H), 2.86 (tdd, J = 11.4, 10.4, 4.0, 2.0 Hz, 1H), 2.09
(dd, J = 14.3, 3.6 Hz, 2H), 1.88 – 1.73 (m, 2H). LCMS (ESI m/z): 245.4 (M+H).
1
4
-(5-(piperidin-4-yl)oxazol-2-yl)phenol (8). Prepared in a manner similar to example 7. H
NMR (400 MHz, DMSOꢀd ) δ 8.36 (s, 1H), 7.75 ꢀ 7.68 (m, 1H), 7.00 (s, 1H), 6.73 ꢀ 6.65 (m,
6
2
1
H), 3.22 ꢀ 3.15 (m, 2H), 3.04 ꢀ 2.98 (m, 1H), 2.90 ꢀ 2.83 (m, 2H), 2.05 ꢀ 2.00 (m, 2H), 1.74 ꢀ
.63 (m, 2H). LCMS (ESI m/z): 263.3 (M+H).
1
4
-[5-(4-hydroxyphenyl)-3-pyridyl]phenol (17). Prepared in a manner similar to 18. H NMR
(400 MHz, DMSOꢀd6) δ 9.66 (s, 2H), 8.71 (d, J = 2.2 Hz, 2H), 8.09 (t, J = 2.2 Hz, 1H), 7.69 –
7.60 (m, 4H), 6.94 – 6.85 (m, 4H). LCMS (ESI m/z): 264.5 (M+H).
Synthesis of Compound 18 (Scheme 4)
4
-[2,6-diamino-5-(4-hydroxyphenyl)-3-pyridyl]phenol (18): Into a 500ꢀmL 3ꢀnecked
roundꢀbottom flask purged and maintained with an inert atmosphere of nitrogen was placed a
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 60
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
solution of 44 (12 g, 59.68 mmol, 1.00 equiv) in CH CN (100 mL), (4ꢀmethoxyphenyl)boronic
3
acid (11 g, 72.39 mmol, 1.20 equiv), Na CO (120 mL, sat.), and Pd(dppf) Cl (1.2 g, 1.64 mmol,
2
3
2
2
0.03 equiv). The resulting solution was stirred at 110º C for 1 h, diluted with of 500 mL of
EtOAc and then extracted 2 X 500 mL with EtOAc. The combined organic layers were washed 3
X 200 mL with brine, dried over anhydrous sodium sulfate and concentrated under vacuum. The
crude residue was carried forward without further purification.
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
The crude residue was dissolved in dichloromethane (100 mL) and charged with
tribromoborane (19.6 g, 78.24 mmol, 3.00 equiv) at 0º C. The resulting solution was stirred at
room temperature for 1 h and then quenched by the addition of 100 mL of NaHCO (1M) at 0º C.
3
The solids were collected by filtration and then washed with 1x100 mL of H O and 1 X 300 mL
2
1
of EtOAc/PE (1:1) to afford 6.3 g (83%) of the title compound as a white solid. H NMR (400
MHz, DMSOꢀd6) δ 9.34 (s, 2H), 7.24 – 7.15 (m, 4H), 6.89 (s, 1H), 6.83 – 6.74 (m, 4H), 5.06 (s,
1
3
4H).
C NMR (101 MHz, DMSOꢀd6) δ 156.23, 154.35, 139.83, 129.86, 115.98, 110.55.
LCMS (ESI m/z): 294.1 (M+H).
Compounds 19 -23 were prepared according to the synthesis of 18 (Scheme 4)
1
4
-[6-amino-5-(4-hydroxyphenyl)-4-methyl-3-pyridyl]phenol (19). H NMR (400 MHz,
DMSOꢀd6) δ 9.48 (s, 2H), 8.14 (s, 1H), 7.71 (s, 1H), 7.14 – 7.00 (m, 4H), 6.92 – 6.84 (m, 2H),
6
.84 – 6.75 (m, 2H), 4.95 (s, 2H), 1.81 (s, 3H). LCMS (ESI m/z): 293.0 (M+H).
1
4
-[6-amino-4-ethyl-5-(4-hydroxyphenyl)-3-pyridyl]phenol (20).
H NMR (400 MHz,
DMSOꢀd6) δ 9.48 (s, 2H), 8.14 (s, 1H), 7.65 (s, 1H), 7.13 – 7.00 (m, 4H), 6.92 – 6.83 (m, 2H),
6
.84 – 6.75 (m, 2H), 4.90 (s, 2H), 2.24 (q, J = 7.4 Hz, 2H). LCMS (ESI m/z): 306.1 (M+H).
ACS Paragon Plus Environment
24

Page 25 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
4
-[6-amino-5-(4-hydroxyphenyl)-4-propyl-3-pyridyl]phenol (21).
H NMR (400 MHz,
DMSOꢀd6) δ 13.80 (s, 1H), 9.82 (s, 1H), 9.70 (s, 1H), 7.77 (s, 1H), 7.22 – 7.04 (m, 6H), 7.00 –
6.88 (m, 2H), 6.91 – 6.79 (m, 2H), 1.04 (h, J = 7.3 Hz, 2H). LCMS (ESI m/z): 321.1 (M+H).
1
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
4
-[6-amino-5-(4-hydroxyphenyl)-4-isopropyl-3-pyridyl]phenol (22). H NMR (400 MHz,
DMSOꢀd6) δ 9.48 (s, 2H), 8.14 (s, 1H), 7.58 (s, 1H), 7.09 – 6.97 (m, 5H), 6.92 – 6.83 (m, 2H),
.81 – 6.73 (m, 2H), 2.87 (p, J = 7.2 Hz, 1H), 0.79 (d, J = 7.2 Hz, 7H). LCMS (ESI m/z): 321.1
M+H).
-amino-3,5-bis(4-hydroxyphenyl)pyridine-4-carbonitrile (23).
6
(
1
2
H NMR (400 MHz,
DMSOꢀd6) δ 9.75 (s, 2H), 8.15 (s, 1H), 8.09 (s, 1H), 7.37 – 7.29 (m, 2H), 7.30 – 7.21 (m, 2H),
6.95 – 6.81 (m, 4H), 5.86 (s, 2H). LCMS (ESI m/z): 304.1 (M+H).
Compounds 24 - 27 were prepared according to the synthesis of compound 28 (Scheme
5
)
1
4-(2-amino-4-ethyl-5-phenyl-3-pyridyl)phenol (24). HNMR (400 MHz, DMSOꢀd6) δ 9.52
s, 1H), 7.70 (s, 1H), 7.46 – 7.27 (m, 6H), 7.10 – 7.02 (m, 2H), 6.93 – 6.84 (m, 2H), 5.01 (s, 2H),
(
3
.27 (s, 2H), 2.26 (q, J = 7.4Hz, 2H), 0.61 (t, J = 7.4 Hz, 3H). LCMS (ESI m/z): 291.1 (M+H).
1
3
-[6-amino-4-ethyl-5-(4-hydroxyphenyl)-3-pyridyl]benzonitrile (25).
HNMR(400MHz,
DMSOꢀd6) δ 9.54 (s, 1H), 7.81 (dd, J = 7.3, 1.5Hz, 2H), 7.77 – 7.52 (m, 4H), 7.13 – 7.01 (m,
2
H), 6.93 – 6.85 (m, 2H), 5.10 (s, 2H), 3.28 (s, 2H), 2.24 (q, J = 7.4Hz, 2H), 0.60 (t, J = 7.5Hz,
3
H). LCMS (ESI m/z): 316.1 (M+H).
1
4-[2-amino-4-ethyl-5-(1H-indazol-6-yl)-3-pyridyl]phenol (26).
H NMR (400 MHz,
DMSOꢀd6) δ 13.07 – 13.02 (m, 1H), 9.55 (s, 1H), 8.09 (s, 1H), 7.81 – 7.74 (m, 2H), 7.42 (d, J =
.3 Hz, 1H), 7.07 (td, J = 8.3, 1.7 Hz, 3H), 6.94 – 6.84 (m, 2H), 5.12 (s, 2H), 2.29 (q, J = 7.4 Hz,
1
2H), 0.62 (t, J = 7.4 Hz, 3H). LCMS (ESI m/z): 331.1 (M+H).
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 60
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
5
-[6-amino-4-ethyl-5-(4-hydroxyphenyl)-3-pyridyl]-N-methyl-pyridine-2-carboxamide
1
(
27): H NMR (400 MHz, DMSOꢀd6) δ 9.56 (s, 1H), 8.80 (q, J = 4.7 Hz, 1H), 8.59 (dd, J = 2.3,
0.9 Hz, 1H), 8.07 (dd, J = 8.0, 0.9 Hz, 1H), 7.96 (dd, J = 8.0, 2.2 Hz, 1H), 7.79 (s, 1H), 7.11 –
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
7
.02 (m, 2H), 6.94 – 6.86 (m, 2H), 5.19 (s, 2H), 2.84 (d, J = 4.8 Hz, 3H), 2.26 (q, J = 7.4 Hz,
H), 0.62 (t, J = 7.4 Hz, 3H). LCMS (ESI m/z): 349.1 (M+H).
2
4
-[2-amino-4-ethyl-5-(1H-indazol-5-yl)-3-pyridyl]phenol (28): Into a 250ꢀmL 3ꢀnecked
roundꢀbottom flask purged and maintained with an inert atmosphere of nitrogen was placed 50
1.0 g, 3.41 mmol, 1.00 equiv), 6ꢀ(tetramethylꢀ1,3,2ꢀdioxaborolanꢀ2ꢀyl)ꢀ2Hꢀindazole (880 mg,
.41 mmol, 1.00 equiv), potassium carbonate (3.3 g, 23.88 mmol, 7.00 equiv), water (30 mL),
1,4ꢀdioxane (25 mL), and Pd(dppf) Cl (200 mg, 0.3 mmol, 0.1 equiv). The resulting solution
(
3
2
2
was stirred at 80º C for 16 h, diluted with 500 mL of H O and 500 mL of ethyl acetate. The
2
organic layer was washed twice with 250 mL of brine and concentrated under vacuum. The
residue was purified on a silica gel column eluting with DCM/CH OH (20:1ꢀ10:1) to afford the
3
1
title compound. H NMR (400 MHz, DMSOꢀd6) δ 13.07 (s, 1H), 9.52 (s, 1H), 8.07 (d, J = 1.0
Hz, 1H), 7.74 (s, 1H), 7.65 (dd, J = 1.6, 0.8 Hz, 1H), 7.56 (dt, J = 8.6, 0.9 Hz, 1H), 7.28 (dd, J =
8
0
4
.5, 1.6 Hz, 1H), 7.10 – 7.04 (m, 2H), 6.92 – 6.86 (m, 2H), 4.94 (s, 2H), 2.26 (q, J = 7.4 Hz, 2H),
.60 (t, J = 7.4 Hz, 3H). LCMS (ESI m/z): 331.1 (M+H).
-[2-amino-4-ethyl-5-(2-methylindazol-6-yl)-3-pyridyl]phenol (29): Prepared according to
1
the synthesis of 28. Refer to Kategaya et al. for spectral information
Synthesis of Compound 30 (Scheme 6)
4
-(6-amino-4-ethyl-5-phenyl-3-pyridyl)phenol (30): A solution of 4ꢀ(6ꢀaminoꢀ5ꢀbromoꢀ4ꢀ
ethylpyridinꢀ3ꢀyl)phenol (59 mg, 0.2 mmol) in 0.5 ml of 1,4ꢀdioxane was charged with
phenylboronic acid (36 mg, 0.3 mmol),bis(diphenylphosphino)ferrocene]dichloropalladium(II)
ACS Paragon Plus Environment
26

Page 27 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
14 mg, 0.02 mmol), and 0.5 ml of 1M K CO . The mixture was then heated at 120º C for 5
2
3
minutes. The mixture was then diluted with ethyl acetate and water. The layers were separated
and the organic layer was washed once with water, dried over Na SO , filtered, and concentrated
2
4
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
in vacuo. The residue was then purified by reverseꢀphase preparatory HPLC to afford the title
1
compound (29 mg, 50% yield). H NMR (400 MHz, DMSOꢀd6) δ 9.39 (s, 1H), 7.70 (s, 1H),
7
.56 – 7.46 (m, 2H), 7.46 – 7.35 (m,1H), 7.30 – 7.22 (m, 2H), 7.14 – 7.06 (m, 2H), 6.84 – 6.75
(
m, 2H), 4.90 (s, 2H), 2.22 (q, J = 7.5 Hz, 2H), 0.61 (t, J = 7.4 Hz, 3H). LCMS (ESI m/z): 291.0
M+H).
(
Compounds 31 and 32 were prepared according to the synthesis of compound 30
(Scheme 6)
-[2-amino-4-ethyl-5-(4-hydroxyphenyl)-3-pyridyl]benzonitrile (31). H NMR (400 MHz,
1
3
DMSOꢀd6) δ 9.43 (s, 1H), 7.87 (dt, J = 7.7, 1.5 Hz, 1H), 7.78 – 7.63 (m, 4H), 7.61 (dt, J = 7.8,
1.5 Hz, 1H), 7.15 – 7.02 (m, 3H), 6.86 – 6.76 (m, 3H), 5.19 (s, 3H). LCMS (ESI m/z): 316.0
(
M+H).
-[6-amino-4-ethyl-5-(1H-indazol-5-yl)-3-pyridyl]phenol (32).
DMSOꢀd6) δ 13.14 (s, 1H), 9.42 (s, 1H), 8.14 (s, 1H), 8.10 (d, J = 0.9 Hz, 1H), 7.70 (s, 1H), 7.70
1
4
H NMR (400 MHz,
–
7.62 (m, 2H), 7.19 (dd, J = 8.3, 1.7 Hz, 1H), 7.16 – 7.00 (m, 2H), 6.85 – 6.75 (m, 2H), 4.97 (s,
H). LCMS (ESI m/z): 331.1 (M+H).
2
tert-butyl 4-(5-(4-hydroxyphenyl)-1,2,4-oxadiazol-3-yl)piperidine-1-carboxylate (35). To
a solution of compound 33 (2.1 g, 10 mmol) in EtOH (5 mL) was added aqueous NH OH (1.5
4
mL, 50%). The mixture was heated at reflux for 70 min. The mixture was concentrated in vacuo
to give the crude product, which was used for the next step directly. A mixture of compound 34
(2.14 g, 10 mmol) and CDI (1.62 g, 10 mmol) in DMF (5.0 mL) was stirred for 0.5 hour at room
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 60
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
temperature. Then to this mixture was added the above crude product, and the mixture was
heated to 110º C for 2 h. Solvent was removed in vacuo and the crude product was purified by
silica gel column chromatography (DCM/CH OH=30/1) to afford the desired product as a white
3
1
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
solid (345 mg, 10%). H NMR (400 MHz, CD OD) δ 7.98 (d, J = 6.8 Hz, 2H), 6.95 (d, J = 6.8
3
Hz, 2H), 4.13ꢀ4.10 (m, 2H), 3.36ꢀ3.32 (m, 3H), 2.04 (m, 2H), 1.82ꢀ1.73 (m, 2H), 1.50 (s, 9H).
LCMS (ESI m/z): 245.5 [M+H].
tert-butyl 4-(2-(4-methoxyphenyl)oxazol-4-yl)piperidine-1-carboxylate (37). A mixture of
4
ꢀmethoxybenzamide (0.5 g, 3.31 mmol) and tertꢀbutyl 4ꢀ(2ꢀbromoacetyl)piperidineꢀ1ꢀ
carboxylate (36, 2.03 g, 6.62 mmol) in toluene (10 mL) was heated to 100º C for 16 h under a
nitrogen atmosphere. After cooling to room temperature, the mixture was concentrated in vacuo.
The crude residue was purified by silica gel chromatography to give the title compound (0.1 g,
1
8
%) as a white solid. H NMR (400 MHz, DMSOꢀd ) δ 9.15 (s, 1H), 8.85 (s, 1H), 7.89 (s, 1H),
6
7.74 (d, J = 8.4 Hz, 2H), 6.87 (d, J = 8.4 Hz, 2H), 3.28 ꢀ 3.23 (m, 2H), 3.01 ꢀ 2.91 (m, 2H), 2.85 ꢀ
2
.79 (m, 1H), 2.07 ꢀ 2.04 (m, 2H). LCMS (ESI m/z): 246.4 (M+H)
tert-butyl-4-(1-hydroxy-2-nitroethyl)piperidine-1-tert-butyl-4-(1-hydroxy-2-
nitroethyl)piperidine-1-carboxylatecarboxylate (39).
To a solution of tertꢀbutyl 4ꢀ
formylpiperidineꢀ1ꢀcarboxylate (38, 20 g, 93.8 mmol) and nitromethane (11.45 g, 187 mmol) in
THF (100 mL) and tꢀBuOH (100 mL) was added tꢀBuOK (10.52 g, 93.8 mmol). The mixture
was stirred at room temperature for 2 h. The mixture was acidified with AcOH to pH 6 and then
extracted with EtOAc (100 mL × 3). The combined organic layers were washed with brine (100
mL), dried over anhydrous Na SO and concentrated in vacuo. The crude residue was purified by
2
4
1
silica gel chromatography to give the title compound (16 g, 62%) as a faint yellow solid. H
NMR (400 MHz, DMSOꢀd ) δ 4.19 ꢀ 4.01 (m, 2H), 3.28 ꢀ 3.24 (m, 1H), 2.89 ꢀ 2.86 (m, 1H),
6
ACS Paragon Plus Environment
28

Page 29 of 60
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
1
.55 ꢀ 2.64 (m, 2H), 2.22 ꢀ 2.13 (m, 4H), 1.84 ꢀ 1.79 (m, 1H), 1.56 ꢀ 1.41 (m, 2H), 1.43 (s, 9H),
.28 ꢀ 1.17 (m, 2H). LCMS (ESI m/z): 275.4 (M+H).
tert-butyl 4-(2-amino-1-hydroxyethyl)piperidine-1-carboxylate (40). To a solution of tertꢀ
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
butyl 4ꢀ(1ꢀhydroxyꢀ2ꢀnitroethyl)piperidineꢀ1ꢀcarboxylate (39, 16 g, 58 mmol) in MeOH (25
mL) was added 10% Pd/C (1.6 g). The mixture was stirred at room temperature for 12 h under a
hydrogen atmosphere (50 psi). The mixture was filtered and the filtrate was concentrated in
vacuo to give the title compound (12 g, crude) as a yellow solid that required no further
1
purification. H NMR (400 MHz, DMSOꢀd ) δ 5.41 (d, J = 6.4 Hz, 1H), 4.76 ꢀ 4.72 (m, 1H),
6
4
.40 ꢀ 4.34 (m, 1H), 3.95 ꢀ 3.89 (m, 3H), 2.70 ꢀ 2.52 (m, 2H), 1.69 ꢀ 1.65 (m, 1H), 1.55 ꢀ 1.52 (m,
2H), 1.39 (s, 9H), 1.23 ꢀ 1.10 (m, 2H). LCMS (ESI m/z): 246.4 (M+H).
tert-butyl 4-(4-methoxybenzamido)acetyl)piperidine-1-carboxylate (41). To a solution
of 4ꢀmethoxybenzoic
acid
(1.5
g,
8.8
mmol) and
tertꢀbutyl
4ꢀ(2ꢀaminoꢀ1ꢀ
hydroxyethyl)piperidineꢀ1ꢀcarboxylate (40, 2.58 g, 10.58 mmol) in DMF (20 mL) was added
HATU (6.7 g, 17.63 mmol) and triethylamine (3.67 mL, 26.45 mmol). The mixture was stirred at
room temperature for 6 h. The mixture was concentrated in vacuo. The crude residue was
purified by silica gel chromatography (petroleum ether / EtOAc = 1 : 1) to give tertꢀbutyl 4ꢀ(2ꢀ
(2ꢀfluoroꢀ4ꢀmethoxybenzamido)ꢀ1ꢀhydroxyethyl)piperidineꢀ1ꢀcarboxylate(1.78 g, 51%) as a
faint yellow solid.
To a solution of tertꢀbutyl 4ꢀ(4ꢀmethoxybenzamido)ꢀ1ꢀhydroxyethyl)piperidineꢀ1ꢀcarboxylate
(1.2 g, 3.03 mmol) in DCM (20 mL) was added DessꢀMartin reagent (3.2 g, 7.55 mmol). The
mixture was stirred at room temperature for 4 h. The mixture was concentrated in vacuo. The
crude residue was purified by silica gel chromatography to give the title compound (1.1 g,
1
92%) as a white solid. H NMR (400 MHz, DMSOꢀd ) δ 7.84 ꢀ 7.79 (m, 1H), 7.66 ꢀ7.61 (m, 1H),
6
ACS Paragon Plus Environment
29