Discovery of Bispecific Antagonists of Retinol Binding Protein 4 That Stabilize Transthyretin Tetramers: Scaffolding Hopping, Optimization, and Preclinical Pharmacological Evaluation as a Potential Therapy for Two Common Age-Related Comorbidities

Article abstract of DOI:10.1021/acs.jmedchem.0c00996

Accumulation of cytotoxic lipofuscin bisretinoids may contribute to atrophic age-related macular degeneration (AMD) pathogenesis. Retinal bisretinoid synthesis depends on the influx of serum all-trans-retinol (1) delivered via a tertiary retinol binding protein 4 (RBP4)-transthyretin (TTR)-retinol complex. We previously identified selective RBP4 antagonists that dissociate circulating RBP4-TTR-retinol complexes, reduce serum RBP4 levels, and inhibit bisretinoid synthesis in models of enhanced retinal lipofuscinogenesis. However, the release of TTR by selective RBP4 antagonists may be associated with TTR tetramer destabilization and, potentially, TTR amyloid formation. We describe herein the identification of bispecific RBP4 antagonist-TTR tetramer kinetic stabilizers. Standout analogue (±)-44 possesses suitable potency for both targets, significantly lowers mouse plasma RBP4 levels, and prevents TTR aggregation in a gel-based assay. This new class of bispecific compounds may be especially important as a therapy for dry AMD patients who have another common age-related comorbidity, senile systemic amyloidosis, a nongenetic disease associated with wild-type TTR misfolding.

Full text of DOI:10.1021/acs.jmedchem.0c00996

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Discovery of Bispecific Antagonists of Retinol Binding Protein
4
That Stabilize Transthyretin Tetramers: Scaffolding Hopping,
Optimization, and Preclinical Pharmacological Evaluation as a
Potential Therapy for Two Common Age-Related Comorbidities
Christopher L. Cioffi, Parthasarathy Muthuraman, Arun Raja,
Andras Varadi, Boglarka Racz, and Konstantin Petrukhin
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00996 • Publication Date (Web): 03 Sep 2020
Downloaded from pubs.acs.org on September 3, 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 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Bispecific Antagonists of Retinol Binding Protein 4
That Stabilize Transthyretin Tetramers: Scaffolding Hopping,
Optimization, and Preclinical Pharmacological Evaluation as a
Potential Therapy for Two Common Age-Related Comorbidities
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
Christopher L. Cioffi,*† Parthasarathy Muthuraman,† Arun Raja,† Andras Varadi,§
§
§
Boglarka Racz, and Konstantin Petrukhin*
†Albany College of Pharmacy and Health Sciences, Departments of Basic and Clinical
Sciences and Pharmaceutical Sciences, 106 New Scotland Ave, Albany, NY 12208
§Department of Ophthalmology, Columbia University Medical Center, New York, NY
1
0032
Abstract: Accumulation of cytotoxic lipofuscin bisretinoids may contribute to atrophic
age-related macular degeneration (AMD) pathogenesis. Retinal bisretinoid synthesis
depends on influx of serum all-trans-retinol (1) delivered via a tertiary retinol-binding
protein 4 (RBP4)–transthyretin (TTR)–retinol complex. We previously identified selective
RBP4 antagonists that dissociate circulating RBP4-TTR-retinol complexes, reduce serum
RBP4 levels, and inhibit bisretinoid synthesis in models of enhanced retinal
lipofuscinogenesis. However, release of TTR by selective RBP4 antagonists may be
associated with TTR tetramer destabilization and, potentially, TTR amyloid formation. We
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 114
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
describe herein identification of bispecific RBP4 antagonists – TTR tetramer kinetic
stabilizers. Standout analogue (±)-44 possesses suitable potency for both targets,
significantly lowers mouse plasma RBP4 levels, and prevents TTR aggregation in a gel-
based assay. This new class of bispecific compounds may be especially important as a
therapy for dry AMD patients who have another common age-related comorbidity, senile
systemic amyloidosis, a non-genetic disease associated with wild-type TTR misfolding.
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
All-trans-retinol (vitamin A, 1) (Figure 1) is an essential vitamin that serves as a precursor
1
2
for the biosynthesis of retinoic acid (2), 11-cis-retinal (3), and many other key retinoids
involved in multiple cellular processes and numerous critical biological functions
throughout the body. Extracellular delivery of fat-soluble 1 from the liver to vitamin A-
dependent tissues is accomplished via a protein transport complex involving the lipocalin
protein retinol binding protein 4 (RBP4) and transthyretin (TTR, thyroxine binding
3
prealbumin). The majority of circulating RBP4 is synthesized in the liver (~60%), where it
requires the binding of 1 (holo-RBP4) prior to secretion. The tertiary complex between
holo-RBP4 and TTR is required as the small size of RBP4 (21 kDa) makes it susceptible to
4
rapid glomerular filtration. Formation of the complex depends on binding of 1 to RBP4
as apo-RBP4 associates with TTR poorly.
2
ACS Paragon Plus Environment

Page 3 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CO H
2
OH
1
2
all-trans-retinol (vitamin A)
retinoic acid
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
CHO
3
1
1-cis-retinal
Figure 1. All-trans-retinol (1), retinoic acid (2) (a vital retinoid involved in morphogenesis), and 11-
cis-retinal (3) (a key retinoid required for phototransduction).
TTR is a 55-kDa homotetrameric protein largely synthesized in the liver for systemic
5
circulation and by the choroid plexus for secretion into the cerebral spinal fluid (CSF). In
plasma, TTR mainly functions as a transporter of holo-RBP4 while also serving as a
secondary transporter for the thyroid hormone thyroxine (T4, 4) (Figure 2). Thyroxine-
binding globulin (TBG) is the major T4 transport carrier in plasma with approximately 10-
1
5% bound to TTR.
I
O
I
I
OH
I
HO C
NH2
2
4
thyroxine (T4)
Figure 2. The thyroid hormone thyroxine (T4) (4).
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 114
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
The quaternary structure of TTR is composed of a dimer of dimer arrangement that
features two identical C2 symmetric binding sites for T4 at the central channel of the
5
tetramer formed where the dimers interface (Figure 3, A). Each dimer subunit is
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
constructed from two 127-residue -sheet-rich polypeptide monomers that associate via
their edge -strands. The dimer-dimer interface at the T4 binding sites is weak, and the
breakage of it constitutes the first step in the TTR tetramer dissociation process.6
Complexation between RBP4 and TTR requires that 1 be initially bound to RBP4. The
structure of RBP4 contains a cylindrical binding cavity for 1 and analysis of Protein Data
7
Bank (PDB) X-ray crystallographic data for holo-RBP4 (PDB 1RBP) shows the isoprene tail
of 1 occupying the tunnel with its pendant trimethyl cyclohexene ring projecting within
the inner hydrophobic -ionone cavity. The hydroxyl group of 1 resides near the binding
cavity opening and is exposed to solvent. Binding of 1 to RBP4 induces conformational
8
changes to exterior loops that provide favorable protein-protein interaction surfaces for
facile TTR engagement. These conformational changes permit two-fold axis of symmetry
docking to TTR (Figure 3, B). TTR features two equivalent binding sites for holo-RBP4,
providing a complex stoichiometry of one TTR tetramer to two holo-RBP4 molecules
5
(
Figure 3, C). However, a 1:1 molar complex is typically observed due limiting
concentrations of plasma RBP4. Lastly, an H-bond between the hydroxyl group of 1 and
9
TTR further stabilizes the protein transport complex and allows it to fully-conceal the
hydrophobic vitamin as it is transported in serum.
4
ACS Paragon Plus Environment

Page 5 of 114
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 3. Three-dimensional structures of TTR, holo-RBP4, and the holo-RBP4-TTR tertiary
complex. (A) Ribbon diagram of the quaternary homotetrameric structure of TTR with 4 bound
5
c
(PDB 2ROX). The dimers are shown as dimer subunit A and dimer subunit B and each monomer
subunit of the tetramer (labeled 1, 2, 3 and 4) is shown with its secondary structural elements and
colored differently. The monomer components of dimer A are shown in orange (monomer 1) and
cyan (monomer 2). The monomer components of dimer B are shown in violet (monomer 3) and
blue (monomer 4). The dimers are associated back-to-back creating a large channel through the
center of the tetramer that presents two C2 symmetrical T4-binding sites. The binding of 4 in both
T4 binding pockets of TTR is shown in ball and stick format (grey). (B) Ribbon diagram of holo-
7
RBP4 (PDB 1RBP). RBP4 is shown as yellow and 1 is depicted in ball and stick format (grey) bound
within the RBP4 hydrophobic -barrel binding cavity. (C) The holo-RBP4-TTR tertiary complex
10
(
PDB 1QAB) colored by chain and viewed from the front. The TTR tetramer is located at the
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 114
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
center of the complex with two holo-RBP4 molecules docked at a twofold axis of symmetry that
is orthogonal to the T4 binding sites. Compound 1 is depicted in ball and stick format (black). (D)
1
0
The holo-RBP4-TTR tertiary complex (PDB 1QAB) colored by chain and viewed from the side.
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
Compound 1 is depicted in ball and stick format (black).
It was hypothesized that reducing systemic levels of circulating RBP4 and 1 via
selective antagonists could impede the biosynthesis of retinal bisretinoids and prevent
geographic lesion formation and growth associated with late-stage dry (atrophic) age-
related macular degeneration (AMD) and Stargardt disease.11, 12 All-trans-retinol-
competitive antagonists of RBP4 prohibit holo-RBP4-TTR complexation, thereby inducing
reductions with circulating RBP4 and 1 levels via rapid renal clearance. A diminished influx
of 1 to the RPE results in a reduction of cytotoxic bisretinoid accumulation in the retina,
which is believed to underlie parts of the pathophysiology of dry AMD and Stargardt
disease.11a, 13-16 The approach is supported by proof-of-concept data obtained for
-/-
fenretinide (5) (Figure 4), which was studied preclinically with Abca4 transgenic mice
1
7
and in a Phase II proof-of-concept study with dry AMD patients. Thorough analysis of
the human data with 5 indicated that a reduction in serum RBP4 levels below the 1 M
threshold was required to significantly hinder expansion of demarcated atrophic lesions.
8
The non-retinoid RBP4 antagonist A1120 (6) was reported disrupt holo-RBP4-TTR
complexation in vitro and reduce rodent RBP4 plasma levels by >70% and reduce retinal
-/-
13
bisretinoid accumulation in Abca4 mice. Our selective and orally bioavailable non-
1
8
19
retinoid RBP4 antagonists 7 and BPN-14136 (8) displayed favorable pharmacokinetic
(PK) profiles and induced dose-dependent circulating RBP4 reductions in rodents.
6
ACS Paragon Plus Environment

Page 7 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 8 also robustly lowered serum RBP4 levels and exhibited excellent
pharmacokinetic-pharmacodynamic (PK-PD) correlations in non-human primates upon
oral administration.20 Lastly, 8 inhibited production of fluorophore bisretinoid N-
retinylidene-N-retinylethanolamine (A2E) while restoring homeostatic complement
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
-/-
system protein expression in the Abca4 mouse retina without altering visual cycle
kinetics.14
In addition to transporting 1 to targeted tissues, RBP4 has also been identified as an
adipokine potentially involved in metabolic disorders including type 2 diabetes,21
2
2
23
24
25
obesity, insulin resistance, cardiovascular disease, and hepatic steatosis. Thus, the
pharmacological reduction of circulating RBP4 serum levels may also hold promise for the
treatment of a myriad of metabolic diseases. Indeed, we have recently reported that our
antagonist 10 significantly lowered serum RBP4 levels in rodents (>80%), reduced the
concentration of circulating RBP4 produced in the adipose tissue, and ameliorated
hepatic steatosis in transgenic adi-hRBP4 mice. These data provide evidence that RBP4
antagonists may hold therapeutic promise for treating non-alcoholic fatty liver disease
(NAFLD).26
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 114
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
OH
O
N
H
5
fenretinide
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
CF3
F
CF3
CF3
CF3
CF3
H
H
N
H
H
H
H
N
CN
O
NH
N
N
N
O
NH
CO H
2
N
N
CO H
N
N
N
N
2
O
N
H C
CO H
CO H
2
3
2
6
7
8
9
10
A1120
BPN-14136
Figure 4. Examples of previously reported RBP4 antagonists that include fenretinide (5), A1120
(6), and previously identified RBP4 antagonists 7, BPN-14136 (8), 9, and 10.
The formation of amyloid aggregates derived from either mutant or wild-type
underlies TTR amyloidosis (ATTR) diseases such as senile systemic amyloidosis (SSA),
peripheral polyneuropathy (ATTR-PN), and cardiomyopathy (ATTR-CM).6, 27 As was noted
above, the breakage of the dimer-dimer interface in TTR tetramers constitutes the first
step in the TTR tetramer dissociation process that leads to TTR misfolding. Approximately
50% of serum TTR is associated with RBP4 and it is suggested that tertiary holo-RBP4-TTR
complexation serves to stabilize this fraction of serum TTR tetramers and prevent them
from dissociation and misfolding.28, 29 Based on the in vitro observation that RBP4-TTR
interaction is capable of conferring an additional stabilization to tetrameric TTR,28, 29 it
seems plausible that the release of TTR tetramers from RBP4-TTR-retinol complexes
8
ACS Paragon Plus Environment

Page 9 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
induced by selective RBP4 antagonists may lead to tetramer destabilization and its
enhanced dissociation to dimer subunits. The resulting dimers may then further
dissociate into monomers that can misfold, aggregate, oligomerize, and eventually form
insoluble TTR amyloid fibrils.28, 29 While selective RBP4 antagonists can be a safe and
effective therapy for the majority of dry AMD patients, this class of compounds may
potentially be counter-indicated for a fraction of AMD patients who may be prone to
developing ATTR. In addition to individuals with rare genetic forms of ATTR caused by pro-
amyloidogenic TTR mutations, the use of selective RBP4 antagonists may not be optimal
in patients with SSA, a late-onset non-genetic disease associated with misfolding and
aggregation of wild-type TTR. SSA affects approximately 25% of patients over the age of
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
0
8
0 and based on the high population frequency of this disease and dry AMD, significant
comorbidity between the two conditions is expected. In addition, the use of selective
RBP4 antagonists may not be optimal in older African-American patients with dry AMD
who have the increased chance of carrying a relatively high frequency pro-amyloidogenic
3
1
V122I mutation in the TTR gene. It is undesirable for an effective chronic treatment for
one of the two conditions to be counter-indicated for the use in patients with another
one, and developing an optimal therapy for dry AMD that can be safely used in patients
with ATTR comorbidities is an important objective.
ATTR pathophysiology begins with the sequential dissociation of TTR tetramers into
pro-amyloidogenic monomers with tetramer dissociation into dimer subunits as the rate-
6
limiting step in the process. While thyroxine binding was reported to stabilize TTR
3
2
tetramers, the majority of TTR in circulation (up to 90%), including TTR in a complex with
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 114
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
28
holo-RBP4, is not bound to its natural ligand. Current therapeutic approaches to treat
ATTR-CM include small molecule kinetic stabilizers of TTR tetramers that bind at the T4
binding sites and increase the energy barrier of tetramer dissociation.33 Two orally
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
bioavailable kinetic stabilizers clinically investigated to date include FDA-approved
3
4
35
tafamidis (11) (Figure 5), and AG10 (12). TTR stabilizer 11 is currently approved to treat
familial amyloid polyneuropathy and ATTR-CM patients, and 12 has demonstrated near-
complete stabilization of TTR in ATTR-CM patients with symptomatic, chronic heart failure
3
6
in Phase II clinical trials (a Phase III trial is ongoing). In addition, the repurposed FDA-
3
7
approved non-steroidal anti-inflammatory drug diflunisal (13) and catechol-O-methyl
38
transferase inhibitor tolcapone (14) are examples of additional small molecules that also
exhibit TTR tetramer kinetic stabilization activity and have been investigated for clinical
efficacy against ATTR-PN. Lastly, recently reported saturation transfer difference (STD)
NMR data by Gemino and coworkers show that -amyloid (A) peptides bind at the RBP4
binding pockets of TTR and that stabilization of the TTR tetramer by iododiflunisal
3
9
(
structure not shown) may be facilitating the TTR−Aβ interaction. These STD-NMR
binding experiments may shed some light on the mechanism by which iododiflunisal
produces its ameliorating effects in animal models for Alzheimer’s disease.40
10
ACS Paragon Plus Environment

Page 11 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HN
N
Cl
Cl
H C
CH3
F
F
H3C
3
O
N
O
O
F
CO H
HO
NO2
2
OH
OH
CO H
CO H
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
2
11
12
AG10
13
14
tafamidis
diflunisal
tolcapone
Figure 5. Examples of reported small molecule TTR tetramer kinetic stabilizers that include
tafamidis (11), AG10 (12), diflunisal (13), and tolcapone (14).
Our goal was to develop a treatment for dry AMD and ATTR comorbidities based on
bispecific ligands that can serve as both an RBP4 antagonist and TTR tetramer kinetic
stabilizer. Such drugs are predicted to provide therapeutic benefits associated with
reducing circulating RBP4 levels while simultaneously stabilizing unliganded TTR
tetramers released from the holo-RBP4-TTR complex, thus circumventing potential risks
of amyloid fibril formation as schematized in Figure 6. Furthermore, a
polypharmacological approach consisting of single bispecific molecule capable of
exhibiting dual activity for both targets may present advantages over the co-
administration of single agent for each target. Such advantages include improving patient
compliance, minimizing complex PK, and avoiding potential drug-drug interactions that
could arise from multiple drug intake.41
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 114
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 6. Bispecific RBP4 antagonists and TTR tetramer kinetic stabilizers for treating RBP4
indications with potential ATTR comorbidities. (A) Schematic illustrating the use of selective RBP4
antagonists for disrupting the holo-RBP4-TTR protein-protein interaction and inducing serum
reduction of 1 and RBP4. The concomitant release of unliganded TTR may induce its aggregation,
potentially contributing to ATTR in predisposed patients. (B) Bispecific ligands with dual RBP4
antagonist and TTR tetramer kinetic stabilization activity may induce reductions in circulating
levels of RBP4 and 1 while also preventing potential TTR aggregation and insoluble amyloid fibril
formation.
In our work reported herein, we initiated a structure-based drug design effort to
identify a novel class of non-retinoid bispecific compounds capable of exhibiting dual
RBP4 antagonist and TTR tetramer kinetic stabilization activity. The primary desirable
attributes for the newly designed analogues were: 1) to contain a non-retinoid framework
so as to avoid potential off-target activity at retinoic acid receptors, 2) to effectively
compete with 1 and antagonize the all-trans-retinol-dependent RBP4-TTR interaction, 3)
12
ACS Paragon Plus Environment

Page 13 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to induce robust in vivo serum RBP4 reduction by >70%, and 4) to bind to unliganded TTR
tetramers and effectively prevent TTR dissociation and aggregation as judged in a gel-
based wild-type TTR aggregation assay. It is for these reasons in our program that
compounds were measured for binding potency to non-TTR associated RBP4 via
competition with 1 (scintillation proximity assay (SPA) to generate dose-response IC50
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
values in the presence of fixed concentrations of [ H]-all-trans-retinol), compound
binding potency to unliganded TTR tetramers (fluorescence polarization assay (FP) to
generate dose-response IC50 values in the presences of a fixed concentration of a
fluorescence probe that binds at TTR tetramer T4 sites), and compound functional
antagonist potency for disruption of holo-RBP4-TTR complex formation (homogenous
time-resolved fluorescence assay (HTRF) to generate dose-response IC values in the
50
presence of fixed concentrations of all-trans-retinol).
DESIGN OF BISPECIFIC ANALOGUES
We discovered that our previously disclosed series of RBP4 antagonists could serve as
suitable benchmark scaffolds for our novel bispecific analogue design efforts. RBP4
1
8
19
antagonists 7 (RBP4 SPA IC = 72.7 nM, RBP4-TTR HTRF IC = 0.294 M) and 9 (RBP4
5
0
50
SPA IC = 20.8 ± 0.5 nM, RBP4-TTR HTRF IC = 79.7 nM) were found to exhibit significant
5
0
50
TTR tetramer binding potency (7 TTR FP IC = 2.4 μM; 9 TTR FP IC = 2.5 μM). These data,
50
50
in addition to the exemplary PK-PD properties19, 20 and in vivo Abca4 transgenic mouse
-/-
1
4
model efficacy exhibited by standout analogue 8 (RBP4 SPA IC = 12.8 ± 0.4 nM, RBP4-
5
0
TTR HTRF IC = 43.6 ± 10.5 nM), led us to focus our initial bispecific drug design efforts
5
0
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 114
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
starting from their common [3.3.0]-bicyclic core scaffold. Thus, our optimization strategy
was to identify novel compounds that presented the following attribute: 1) retaining or
improving in vitro activity observed for 7, 8, and 9, 2) exhibiting TTR binding potency that
could provide adequate TTR tetramer stabilization and prevent aggregation at
therapeutic concentrations required for >70% serum RBP4 reduction, 3) maintaining the
largely favorable ADME profile observed for 7, 8, and 9, and 4) maintaining the favorable
PK-PD properties of 7 and 8. A potential drawback for 8 was its ancillary agonist activity
at the nuclear peroxisome proliferator-activated receptor-gamma (PPARγ) (PPARγ IC =
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
3.6 M in the agonist-induced corepressor NCoR release assay). PPARγ plays a critical role
in regulating energy homeostasis, metabolism, and inflammation and we sought to avoid
such off-target activity due to potential side effects often associated with PPARγ agonists,
which include risk for fluid retention, weight gain, bone loss, and congestive heart
failure.42
Our structure-based optimization approach sought to exploit potential similarities
presented within the binding pockets of both proteins. X-ray crystallographic data of 4
5
c
bound to TTR (PDB 2ROX) shows that the T4 binding sites are segregated by an inner
and outer binding cavity. The two-fold axis symmetry of the T4 binding cavity presents six
halogen binding pockets designated as HBP1, HBP1’, HBP2, HBP2’, HBP3 and HBP3’
5c
(
Figure 7). The HBPs are defined according to the relative positions of the iodine atoms
of 4 that occupy them. The inner T4 binding pockets contain HBP3 and HBP3’ and are
located between the Ser117, Leu110, Thr119, and Ala108 residues of both subunits. The
innermost Ser117 hydroxyl groups may either participate in bridging hydrogen bond
14
ACS Paragon Plus Environment

Page 15 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
interactions with a resident structural water molecule or they may mediate hydrogen
3
5
bond interactions with bound TTR tetramer kinetic stabilizers, such as 12. The HBP2 and
HBP2’ pockets are located between the inner and outer binding cavities and are highly
lipophilic as they contain Leu17, Ala108, Ala109 and Leu110 residues. Lastly, the HBP1
and HBP1’ pockets reside near the outer binding cavity and feature residues Lys15, Glu54,
Leu17, Thr106, and Val121 of both subunits. The binding of 4 typically involves a forward
binding mode, which projects the bis-iodo phenol “head group” deep within the inner
cavity where the iodine atoms occupy HBP2 and HPB3 pockets. The linker phenyl ring
resides within the hydrophobic channel bridging the inner and outer cavities, with one of
its iodine atoms occupying a HBP1/HBP1’ binding pocket. Lastly, the amino acid “bottom
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
group” motif of 4 resides near the periphery of the binding site opening and forms salt-
bridge interactions with ionizable residues Lys15/Lys15’and Glu54/Glu54’.5c
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 114
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 7. The halogen binding pockets (HBPs) of the T4 binding site with 4 bound (PDB 2ROX).5c
Compound 4 is shown in ball and stick format and both symmetry-related binding modes related
by a two-fold axis are presented (4 is shown in violet and magenta). The iodine atoms of 4 occupy
the HBP1, HBP2, and HBP3 pockets while the amino acid appendage engages in ionic salt-bridge
interactions Lys15 and Glu54 near the opening of the binding site. Contacting amino acids are
designated and depicted in ball and stick form. Dashed blue lines represent H-bonds and dashed
cyan lines represent halogen bonds.
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 common pharmacophore among small molecule RBP4 antagonists and TTR
tetramer kinetic stabilizers is a “bottom group” carboxylic acid appendage connected
through a core linker scaffold to an aromatic “head group” moiety. The X-ray
8
crystallographic data PDB 3FMZ provides insight as to how 6 and related analogues 7 and
8
may bind to RBP4 (Figure 8, A). The binding pose of 6 involves the piperidine core linker
residing within the central cavity while the lipophilic phenyl ring bearing a trifluoromethyl
group extends into the inner lipophilic cavity in an orthogonal manner. The ionizable
anthranilic acid forms electrostatic binding interactions with Arg121 and a pair of H-bonds
with neighboring residues Tyr90 and Gln98. The structural data for 4 bound to the T4
5
c
binding site (PDB 2ROX) reveals how both molecular targets present binding pockets of
comparable geometries and van der Waals surfaces areas of hydrophobicity and polarity.
Ligand 4 binds in the aforementioned forward binding mode with its lipophilic phenolic
head group projecting into the inner binding cavity and oriented orthogonally (Figure 8,
C). The linker phenyl ring of 4 resides within the hydrophobic channel bridging the inner
and outer cavities, and similar to RBP4 antagonist 6, the ionizable amino acid appendage
16
ACS Paragon Plus Environment

Page 17 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
resides near the opening of its respective binding site where it also engages in salt-bridge
interactions.
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 8. X-ray crystallographic data showing similar binding poses for 6 bound to RBP4 (PDB
8
5c
3
FMZ) and 4 bound to the T4 binding site (PDB 2ROX). (A) RBP4 antagonist 6 is shown in ball
and stick format (dark blue) with polar binding interactions (e.g., H-bonds and salt bridges) shown
as blue dotted lines. (B) Ligand interaction diagram of 6 with RBP4 indicating key H-bonds and
salt-bridge interactions. (C) TTR ligand 4 is shown in ball and stick format (light blue) with the
phenolic aryl appendage projecting into the inner cavity in an orthogonal manner. The iodine
atoms of the head group occupy the HBP2 and HBP3 pockets while the iodine atoms of the central
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 114
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
core phenyl ring occupy the HBP1 pockets. The amino acid appendage engages in ionic salt-bridge
interactions Lys15 and Glu54 near the opening of the binding site. Polar binding interactions (e.g.,
H-bonds and salt bridges) shown as dark blue dotted lines, halogen bonds shown as light-blue
dotted lines. (D) Ligand interaction diagram of 4 with TTR indicating the key salt-bridge
interaction.
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
We planned to develop novel bispecific ligands that retain the key pharmacophoric
elements described for ligands of both targets. Importantly, critical salt-bridge
interactions with the basic Arg121 and Lys15 residues of RBP4 and TTR, respectively, are
mediated by an aromatic carboxylic acid for many known RBP4 antagonists and TTR
ligands. Hence, this motif was not altered in our novel bispecific analogue design. From
8
5c
an examination of PDB 3FMZ and PDB 2ROX) , we speculated that varying the length
and conformational flexibility in the [3.3.0]-bicyclic core linker of 8 might maintain RBP4
antagonist potency while potentially enhancing TTR activity. Figure 9 highlights how we
executed our sequential structure-activity relationship (SAR) plan by examining three
critical regions of our novel scaffolds.
We initially executed a scaffold hopping campaign that explored how altering the
shape and flexibility of the linker core of 8 (A-region) would affect potency at both targets.
We hypothesized that subtle changes to this region of the molecule may better position
an appropriately substituted aromatic head group to fully occupy both the RBP4 -ionone
pocket and inner T4 binding cavity. We synthesized initial illustrative examples of 8
containing alternative linker scaffolds that retained the ortho-trifluoromethylphenyl head
group and the 6-methylpyrimidine-4-carboxylic acid bottom group moieties. We chose to
18
ACS Paragon Plus Environment

Page 19 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
maintain the ortho-trifluoromethylphenyl head group as it is an established head group
motif for optimal RBP4 binding affinity and it has served as a suitable head group for
43
44
previously disclosed oxime and N-phenyl phenoxazine TTR amyloid fibril inhibitors
structures not shown). We postulated that the 6-methylpyrimidine-4-carboxylic acid
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
(
bottom group provides functionality for the aforementioned key ionic interactions while
also providing additional structural components that could also engage in favorable
binding interactions. For example, we hypothesized that the pyrimidine nitrogen para to
the carboxylic acid may be in a similar position as the urea carbonyl of 6 (PDB 3FMZ),
8
which hydrogen bonds with backbone Leu37. Furthermore, the nitrogen ortho to the
carboxylic acid group could also be within a suitable proximity to H-bond with
8
5c
Glu98/Tyr80 of RBP4 (PDB 3FMZ) and/or Glu54 of TTR (PDB 2ROX). Analysis of PDB
ROX suggests that the pyrimidine ring 6-position, which previously provided informative
2
1
9
RBP4 SAR, could also potentially provide a handle to further probe the HBP1/HBP1’
pocket within the T4 binding site.5c
The various scaffold hopping core linkers explored presented similar distances
between the aromatic head group and carboxylic acid bottom group of 4 and bicyclic
analogues 6, 7, and 8. Once a novel A-region core linker with suitable activity at both
targets was identified, it was then used to explore the SAR of alternatively substituted
aromatic head groups (B-region). Lastly, we subsequently studied carboxylic acid bottom
group (C-region) SAR with an identified optimal A- and B-region scaffold. This C-region
SAR sample set included analogues featuring picolinic and nicotinic acids, positional
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 114
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
variation of the carboxylic acid, carboxylic acid isosteres, and carboxamides that were
prepared and analyzed for potency effects at both targets.
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
Sequential SAR Approach Toward Novel Bispecific RBP4 Antagonists - TTR Tetramer Kinetic Stabilizers
Step 2.
Step 1.
Aromatic Head Group (B-Region):
Explore substituted aromatic head
groups that can bind within the RBP4
interior -ionone pocket and the
innermost TTR halogen binding pockets
HBP2/HBP2' and HBP3/HBP3'.
Core Linker Scaffold (A-Region):
Initially examine linker modifications
via scaffold hopping that contain the
ortho-trifluoromethyphenyl
aromatic
head group and 6-methylpyrimidine-4-
carboxylic bottom grop of 8 that lead to
novel compounds exhibiting balanced
activity for both RBP4 and TTR.
B-Region

-ionone pocket
HBP3/HBP3'
I
R1
HBP2/HBP2'
OH
I
CF3
Y
N
RBP4 Leu37
O
O
A-Region
H
H
R
I
I
N
N
R
H
X
X
Z
N
N
HBP1/HBP1'
NH2
R2
H3C
CO2H
C-Region
2
HO C
2
CO H
H
N
Glu54
TTR
8
4
H
H
N
Step 3.
H
H
H
Aromatic Carboxylic Acid Bottom Group (C-Region):
Explore substituted pyrimidines, pyridines, and other
aromatic carboxylic acid moieties capable of favorably
interacting with the ionizable and polar amino acid residues
residing at the opening of the RBP4 and T4 binding sites.
Lys15
N
H
N
Arg121
H
RBP4
X = N, CH, CF
Y = CH2, O, S, NH
Z = CO2H, carboxylic acid isostere, carboxamide
R2 = CH3, H
Figure 9. Medicinal chemistry strategy to identify bispecific RBP4 antagonists – TTR tetramer
kinetic stabilizers. Novel analogues were rationally designed from benchmark RBP4 antagonist 8
and endogenous TTR ligand 4. The initial SAR campaign involved scaffold hopping of the core
linker (A-region) while maintaining the aromatic head group and aromatic carboxylic acid bottom
group appendages of 8. A subsequent SAR campaign of the RBP4 -ionone and T4 inner cavities
involved exploration of the aromatic head group (B-region) with a newly identified A-region core
linker. Lastly, with an optimal A-region and B-region scaffold identified, aromatic carboxylic acid
analogues (C-region) were then investigated. These analogues were designed to potentially
engage in critical RBP4 H-bonding interactions with Arg121, Gln98 (not shown), and backbone
20
ACS Paragon Plus Environment

Page 21 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Leu37 (all RBP4 interactions shown in blue) while also engaging in key TTR binding interactions
with Lys15, Glu54 and within the HBP1/HBP1’ (all TTR interactions shown in magenta).
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
CHEMISTRY
The preparation of alkoxypiperidine analogue 21 shown in Scheme 1 begins with a
sodium borohydride (NaBH ) reduction of tert-butyl 4-oxopiperidine-1-carboxylate (15)
4
to give alcohol 16. Sulfonylation of 16 followed by bimolecular nucleophilic (S 2)
N
displacement of the corresponding sulfonate ester 17 with 2-(trifluoromethyl)phenol led
to formation of phenoxyether 18. Treatment of 18 with trifluoroacetic acid (TFA) provided
Boc-deprotected amine 19, which when reacted with methyl 2-chloro-6-
methylpyrimidine-4-carboxylate, afforded pyrimidine methyl ester 20. Hydrolysis of 20
with lithium hydroxide (LiOH) followed by subsequent acidification of the lithium
carboxylate salt with 2 N aqueous hydrochloric acid (HCl) provided desired carboxylic acid
2
1.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 114
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
Scheme 1a
F3C
O
O
OH
OTs
a
b
c
N
N
N
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
N
Boc
Boc
Boc
Boc
1
5
16
17
18
F3C
O
F3C
F3C
O
O
N
d
e
f
N
N
H
N
N
N
N
1
9
H C
CO CH3
H C
3
CO H
2
3
2
20
21
a
Reagents and conditions: (a) NaBH , CH OH, 0 °C to rt, 8 h; (b) TsCl, DMAP, Et N, CH Cl , 0 °C to
4
3
3
2
2
rt, 16 h; (c) 2-(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (d) TFA, CH Cl , 0 °C to rt, 8 h; (e)
2
3
2
2
methyl 2-chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (f) (i) LiOH, CH OH,
2
3
THF, H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
Scheme 2 highlights the preparation of racemic alkoxypyrrolidine analogue (±)-27.
Tert-butyl 3-(tosyloxy)pyrrolidine-1-carboxylate ((±)-23) was generated via tosylation of
commercially available racemic tert-butyl 3-hydroxypyrrolidine-1-carboxylate ((±)-22).
Treatment of sulfonate ester (±)-23 with 2-(trifluoromethyl)phenol provided ether (±)-24,
which was deprotected with TFA affording amine (±)-25. Conversion of (±)-25 to the
desired carboxylic acid (±)-27 followed the same two-step process outlined in Scheme 1.
22
ACS Paragon Plus Environment

Page 23 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 2a
CF3
CF3
OH
OTs
O
O
a
b
c
N
N
N
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
N
Boc
Boc
Boc
H
(
±)-22
(±)-23
(±)-24
(±)-25
CF3
CF3
O
O
d
e
N
N
N
N
N
N
H C
CO CH
H3C
CO2H
3
2
3
(
±)-26
(±)-27
a
Reagents and conditions: (a) TsCl, DMAP, Et N, CH Cl , 0 °C to rt, 16 h; (b) 2-
3
2
2
(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (c) TFA, CH Cl , 0 °C to rt, 8 h; (d) methyl 2-
2
3
2
2
chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (e) (i) LiOH, CH OH, THF,
2
3
H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
O-alkylation of tosylated azetidine 28 with 2-(trifluoromethyl)phenol afforded phenyl
ether 29 (Scheme 3). TFA-mediated Boc-deprotection of 29 followed by treatment with
methyl 2-chloro-6-methylpyrimidine-4-carboxylate yielded pyrimidine 31, which was
hydrolyzed with LiOH to provide acid 32.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 114
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
Scheme 3a
F3C
F3C
O
F3C
O
F3C
O
OTs
a
O
b
c
d
N
N
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
N
N
N
N
N
N
N
Boc
Boc
29
H
H C
CO CH3
H C
3
CO H
2
3
2
28
30
31
32
a
Reagents and conditions: (a) 2-(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (b) TFA, CH Cl ,
2
3
2
2
0
°C to rt, 8 h; (c) methyl 2-chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h;
2
(d) (i) LiOH, CH OH, THF, H O, rt, 16 h; (ii) 2 N aqueous HCl.
3
2
Construction of (hydroxymethyl)azetidine analogue 38 followed the route depicted in
Scheme 4. Tosylation of alcohol 33 gave sulfonate ester 34, which was subsequently
treated with 2-(trifluoromethyl)phenol to provide phenyl ether 35. TFA-mediated
deprotection of 35 followed by reaction with methyl 2-chloro-6-methylpyrimidine-4-
carboxylate provided analogue 37. LiOH promoted hydrolysis of methyl ester 37 and
subsequent acidification with 2 N aqueous HCl gave desired carboxylic acid 38.
24
ACS Paragon Plus Environment

Page 25 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 4a
CF3
OH
OTs
CF3
a
b
O
c
O
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
N
N
N
Boc
Boc
N
Boc
H
33
34
3
5
36
CF3
CF3
O
O
N
d
e
N
N
N
N
N
H C
CO CH
H C
CO H
2
3
2
3
3
3
7
38
a
Reagents and conditions: (a) TsCl, DMAP, Et N, CH Cl , 0 °C to rt, 16 h; (b) 2-
3
2
2
(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (c) TFA, CH Cl , 0 °C to rt, 8 h; (d) methyl 2-
2
3
2
2
chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (e) (i) LiOH, CH OH, THF,
2
3
H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
Racemic 3-(hydroxymethyl)pyrrolidine (±)-44 was achieved via the route shown in
Scheme 5. Sulfonate ester formation of racemic alcohol (±)-39 followed by SN2
displacement with 2-(trifluoromethyl)phenol in the presence of cesium carbonate (Cs CO ) gave
2
3
O-alkylated product (±)-41. Treatment of (±)-41 with TFA followed by reaction with methyl
2-chloro-5-methylpyrimidine-4-carboxylate yielded (±)-43, which was saponification with
LiOH to provide desired racemic carboxylic acid (±)-44.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 114
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
Scheme 5a
HO
TsO
O
CF3
O
CF3
a
b
c
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
N
N
N
N
Boc
Boc
Boc
H
(
±)-39
(±)-40
(±)-41
(±)-42
O
O
CF3
CF3
d
e
N
N
N
N
N
N
H C
CO CH3
2
H C
CO2H
3
3
(
±)-43
(±)-44
a
Reagents and conditions: (a) TsCl, DMAP, Et N, CH Cl , 0 °C to rt, 16 h; (b) 2-
3
2
2
(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (c) TFA, CH Cl , 0 °C to rt, 8 h; (d) methyl 2-
2
3
2
2
chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (e) (i) LiOH, CH OH, THF,
2
3
H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
Preparation of enantiomers (R)-50 and (S)-56 is highlighted in Scheme 6. The
synthesis of the enantiopure analogues was achieved using the same route outlined in
Scheme 5 but starting with either tert-butyl (R)-3-(hydroxymethyl)pyrrolidine-1-
carboxylate ((R)-45) or tert-butyl (S)-3-(hydroxymethyl)pyrrolidine-1-carboxylate ((S)-51).
26
ACS Paragon Plus Environment

Page 27 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 6a
HO
TsO
O
O
CF3
O
CF3
CF3
O
CF3
a
b
c
d
e
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
N
N
N
N
N
Boc
Boc
N
Boc
(R)-47
N
N
H
N
N
(R)-45
(R)-46
(R)-48
CO2H
H3C
CO2CH3
H3C
(
R)-49
(R)-50
HO
TsO
O
O
CF3
O
CF3
CF3
O
CF3
a
b
c
d
e
N
N
N
N
N
N
Boc
Boc
N
Boc
(S)-53
H
N
N
N
(S)-51
(S)-52
(S)-54
CO2H
H3C
CO2CH3
H3C
(
S)-55
(S)-56
a
Reagents and conditions: (a) TsCl, DMAP, Et N, CH Cl , 0 °C to rt, 16 h; (b) 2-
3
2
2
(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (c) TFA, CH Cl , 0 °C to rt, 8 h; (d) methyl 2-
2
3
2
2
chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (e) (i) LiOH, CH OH, THF,
2
3
H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
The preparation of the racemic piperidine carboxylic acid analogue (±)-62 begins with
SN2 displacement of tosylate (±)-58 with 2-(trifluoromethyl)phenol followed by treatment
with TFA gave secondary amine (±)-60 (Scheme 7). Reaction of amine (±)-60 with methyl
2
-chloro-5-methylpyrimidine-4-carboxylate yielded (±)-61, which upon saponification
provided desired carboxylic acid (±)-62.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 114
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
Scheme 7a
CF3
CF3
O
OH
OTs
O
a
b
c
N
N
Boc
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
N
Boc
(±)-59
N
H
Boc
(
±)-57
(±)-58
(±)-60
CF3
O
CF3
O
d
e
N
N
N
N
N
N
H C
CO CH3
H C
3
CO H
2
3
2
(
±)-61
(±)-62
a
Reagents and conditions: (a) TsCl, DMAP, Et N, CH Cl , 0 °C to rt, 16 h; (b) 2-
3
2
2
(trifluoromethyl)phenol, Cs CO , DMF, 80 °C, 16 h; (c) TFA, CH Cl , 0 °C to rt, 8 h; (d) methyl 2-
2
3
2
2
chloro-6-methylpyrimidine-4-carboxylate, i-Pr NEt, THF, reflux, 16 h; (e) (i) LiOH, CH OH, THF,
2
3
H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
Racemic 2,7-diazaspiro[4.4]nonane analogue (±)-67 was designed to explore the SAR
effects of introducing a ring-constrained core (Scheme 8). Aniline (±)-64 was generated
via
a
palladium-catalyzed amination between (±)-63 and 1-bromo-2-
(trifluoromethyl)benzene using tris(dibenzylideneacetone)dipalladium (0) (Pd (dba) )
2
3
and 2-dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl (XPhos). Boc-deprotection of
±)-64 with TFA gave amine (±)-65, which was subsequently reacted with methyl 2-chloro-
(
5-methylpyrimidine-4-carboxylate to give pyrimidine (±)-66. Saponification of (±)-66 with
28
ACS Paragon Plus Environment

Page 29 of 114
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
LiOH followed by acidification of the lithium carboxylate salt with 2 N HCl provided
desired spirocycle (±)-67.
Scheme 8a
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
NH
N
a
b
N
F3C
F3C
N
N
N
Boc
Boc
H
(
±)-63
(±)-64
(±)-65
N
N
c
F3C
d
F3C
N
N
N
N
N
N
H3C
CO2CH3
3
H C
2
CO H
(
±)-66
(±)-67
a
Reagents and conditions: (a) 1-bromo-2-(trifluoromethyl)benzene, X-Phos, Pd (dba) , Cs CO ,
2
3
2
3
1
4
,4-dioxane, 110 °C, 16 h; (b) TFA, CH Cl , 0 °C to rt, 8 h; (c) methyl 2-chloro-6-methylpyrimidine-
2
2
-carboxylate, i-Pr NEt, THF, reflux, 16 h; (d) (i) LiOH, CH OH, H O, rt, 16 h; (ii) 2 N aqueous HCl.
2
3
2
Racemic thioether analogue (±)-71 construction of is depicted in Scheme 9. The route
use for the synthesis of (±)-71 follows the aforementioned route described for (±)-44
except that the S 2 displacement of sulfonate ester (±)-39 was conducted with 2-
N
(trifluoromethyl)benzenethiol.
29
ACS Paragon Plus Environment