Discovery of the First α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic Acid (AMPA) Receptor Antagonist Dependent upon Transmembrane AMPA Receptor Regulatory Protein (TARP) γ-8

Article abstract of DOI:10.1021/acs.jmedchem.6b00125

Transmembrane AMPA receptor regulatory proteins (TARPs) are a family of scaffolding proteins that regulate AMPA receptor trafficking and function. TARP γ-8 is one member of this family and is highly expressed within the hippocampus relative to the cerebellum. A selective TARP γ-8-dependent AMPA receptor antagonist (TDAA) is an innovative approach to modulate AMPA receptors in specific brain regions to potentially increase the therapeutic index relative to known non-TARP-dependent AMPA antagonists. We describe here, for the first time, the discovery of a noncompetitive AMPA receptor antagonist that is dependent on the presence of TARP γ-8. Three major iteration cycles were employed to improve upon potency, CYP1A2-dependent challenges, and in vivo clearance. An optimized molecule, compound (-)-25 (LY3130481), was fully protective against pentylenetetrazole-induced convulsions in rats without the motor impairment associated with non-TARP-dependent AMPA receptor antagonists. Compound (-)-25 could be utilized to provide proof of concept for antiepileptic efficacy with reduced motor side effects in patients.

Full text of DOI:10.1021/acs.jmedchem.6b00125

Subscriber access provided by Weizmann Institute of Science
Article
The Discovery of The First #-Amino-3-Hydroxy-5-Methyl-4-
Isoxazolepropionic Acid (AMPA) Receptor Antagonist Dependent Upon
Transmembrane AMPA Receptor Regulatory Protein (TARP) Gamma-8
Kevin M. Gardinier, Douglas L. Gernert, Warren J. Porter, Jon K. Reel, Paul L. Ornstein, Patrick Spinazze,
F. Craig Stevens, Patric Hahn, Sean P. Hollinshead, Daniel Mayhugh, Jeff Schkeryantz, Albert Khilevich,
Oscar De Frutos, Scott D. Gleason, Akihiko S. Kato, Debra Luffer-Atlas, Prashant V. Desai, Steven
Swanson, Kevin D. Burris, Chunjin Ding, Beverly A. Heinz, Anne B. Need, Vanessa N Barth, Gregory
A Stephenson, Benjamin A Diseroad, Tim A. Woods, Hong Yu, David Bredt, and Jeffrey M. Witkin
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00125 • Publication Date (Web): 11 Apr 2016
Downloaded from http://pubs.acs.org on April 18, 2016
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 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The Discovery of The First αꢀAminoꢀ3ꢀHydroxyꢀ5ꢀ
Methylꢀ4ꢀIsoxazolepropionic Acid (AMPA)
Receptor Antagonist Dependent Upon
Transmembrane AMPA Receptor Regulatory
Protein (TARP) Gammaꢀ8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kevin M. Gardinier*, Douglas L. Gernert, Warren J. Porter, Jon K. Reel, Paul L. Ornstein,
Patrick Spinazze, F. Craig Stevens, Patric Hahn, Sean P. Hollinshead, Daniel Mayhugh, Jeff
Schkeryantz, Albert Khilevich, Oscar De Frutos, Scott D. Gleason, Akihiko S. Kato, Debra
Luffer-Atlas, Prashant V. Desai, Steven Swanson, Kevin D. Burris, Chunjin Ding, Beverly A.
Heinz, Anne B. Need, Vanessa N. Barth, Gregory A. Stephenson, Benjamin A. Diseroad, Tim A.
Woods, Hong Yu, David Bredt, and Jeffrey M. Witkin
†
Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285 USA
KEYWORDS: AMPA, TARP, Gammaꢀ8, TDAA, epilepsy, hippocampus, clinical candidate
ABSTRACT:
Transmembrane AMPA Receptor Regulatory Proteins (TARPs) are a family of scaffolding
proteins that regulate AMPA receptor trafficking and function. TARP gammaꢀ8 (γꢀ8) is one
member of this family and is highly expressed within the hippocampus relative to the
cerebellum. A selective TARP γꢀ8ꢀdependent AMPA receptor antagonist (TDAA) is an
innovative approach to modulate AMPA receptors in specific brain regions to potentially
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
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
increase the therapeutic index relative to known nonꢀTARPꢀdependent AMPA antagonists. We
describe here, for the first time, the discovery of a nonꢀcompetitive AMPA receptor antagonist
that is dependent on the presence of TARP γꢀ8. Three major iteration cycles were employed to
improve upon potency, CYP1A2ꢀdependent challenges, and in vivo clearance. An optimized
molecule, compound (ꢀ)ꢀ25 (LY3130481), was fully protective against pentylenetetrazoleꢀ
induced convulsions in rats without the motor impairment associated with nonꢀTARPꢀdependent
AMPA receptor antagonists. Compound (ꢀ)ꢀ25 could be utilized to provide proof of concept for
antiepileptic efficacy with reduced motor sideꢀeffects in patients.
0
1
2
3
4
5
6
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:
Despite decades of research, epilepsy remains a devastating disease with lifeꢀthreatening
1
implications, as well as, social and economic impact to patients and caregivers. While today’s
available antiepileptic drugs (AEDs) decrease seizure frequency and magnitude for many
patients, a significant population do not gain adequate therapeutic benefit. In most cases,
prescribers cycle through multiple AEDs and combinations as maintenance therapy in an attempt
to control the patient’s seizures. In addition, many AEDs are associated with unwanted side
effects such as dizziness, ataxia and sedation at therapeutic doses. Thus there is still significant
unmet medical need to identify new therapies for epilepsy that explore alternative mechanisms or
have a wider therapeutic window between efficacy and undesirable side effects.
Glutamate is the principal excitatory neurotransmitter in the brain and is the endogenous
ligand for metabotropic (Gꢀprotein coupled) and ionotropic (ionꢀchannel forming) glutamate
receptors. Ionotropic receptors include αꢀaminoꢀ3ꢀhydroxyꢀ5ꢀmethylꢀ4ꢀisoxazolepropionic acid
(AMPA) receptors which show predominant postsynaptic localization and mediate fast synaptic
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
transmission. Systemic or central administration of glutamate receptor agonists, such as
3
AMPA, kainate, or NꢀmethylꢀDꢀaspartate (NMDA) evoke seizures in rodents. Conversely,
antagonists of these receptors have been shown to be potent anticonvulsants. For example,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
Figure 1 shows several known AMPA receptor antagonists, including perampanel (1),
5
6
7
selurampanel (2) , talampanel (3) and GYKI 53784 (4).
Importantly, talampanel and
perampanel show antiepileptic efficacy in humans and the latter is approved to treat complex
8
partial seizures.
As these AMPA receptor antagonists block excitatory neurotransmission
across the brain, it is not surprising that motor impairment, dizziness, ataxia, and falling are
common side effects that occur at efficacious exposures.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
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
N
N
O
H
N
O
N
N
N
S
O
O
N
N
H
O
0
1
2
3
4
5
6
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
2
Perampanel
Selurampanel
Chiral
Chiral
H
N
O
O
O
O
N
N
N
N
O
O
H N
H N
2
2
3
4
Talampanel
GYKI53784
Figure 1. Structures of known nonꢀTARPꢀdependent AMPA antagonists.
9
AMPA receptors are associated with various auxiliary subunits, including TARPs.
Importantly, TARPs dramatically enhance AMPA receptor trafficking and gating. In addition
1
0
they modulate AMPA receptor pharmacology.
The identification of the TARP family of
proteins and the different regional expression of these family members across the central nervous
system (CNS) provides an opportunity to identify AMPA antagonists that are dependent upon
the coꢀexpression of AMPA receptors with various TARP proteins for potential therapeutic
ACS Paragon Plus Environment

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
11
advantage.
For example, TARP γꢀ2 is highly expressed within the cerebellum, which is
involved in motor coordination, while TARP γꢀ8 is highly expressed within the hippocampus, a
primary locus of epileptogenic activity. Of particular interest was the very low expression of
TARP γꢀ8 in the cerebellum. Thus, we sought to identify a TARP γꢀ8ꢀdependent AMPA
receptor antagonist (TDAA) and had hypothesized previously that such a molecule could provide
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
antiepileptic efficacy without motor impairment. We describe here the discovery of an
optimized TDAA, (ꢀ)ꢀ25, and test our hypothesis in a rodent seizure model.
RESULTS AND DISCUSSION
Highꢀthroughput assays designed to identify and profile TARPꢀsubtype selective AMPA
receptor antagonists were developed by monitoring glutamateꢀinduced calcium influx. Here we
used GluA1 subunit for the AMPA receptor principal subunit, since GluA1 is one of the major
hippocampal AMPA receptors. AMPA receptors have two different splice variants, flip and
flop, that differ in pharmacological properties, channel kinetics, and the amplitude of
1
3
desensitized currents.
The flop isoform desensitizes faster than flip, and this activity is
14
modulated by TARPs. The ability of compounds to block glutamateꢀstimulated increases in
2
+
intracellular Ca was examined in CHOꢀS cells expressing iGluA1flip alone (TARPꢀ
dependence) or iGluA1flop plus different TARP subtypes (TARP selectivity) using fluorescence
imaging plate reader (FLIPR)ꢀbased assays. With these in vitro assay systems in place, a high
throughput screen of a diverse subset of the inꢀhouse chemical collection identified compound 5
as an antagonist against AMPA receptors in cells coꢀexpressing TARP γꢀ8 (Figure 2) with an
IC = 2.97 µM (Table 1). This molecule showed no antagonism of AMPA receptors in the
50
absence of TARP γꢀ8 (IC50 >83 µM) or when coꢀexpressed with TARP γꢀ2 (IC >83 µM). In
5
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 58
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
contrast, perampanel (1), a nonꢀTARP dependent AMPA receptors antagonist, showed weak
functional blockade in each of these cell lines.
0
1
2
3
4
5
6
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
Figure 2. Compound 5 identified from an internal screen.
ACS Paragon Plus Environment

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
5
Table 1. Functional antagonist data in vitro for compounds 5 and perampanel (1).
Assay
Compound 5
2,970 ± 0.17 (6)
>83,300 (5)
Perampanel (1)
1,180 ± 0.41 (3)
485 ± 0.13 (4)
243 ± 0.16 (2)
623 ± 0.30 (2)
6,510 ± 0.74 (4)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
GluA1/TARP γꢀ8 IC (nM) ±SEM (n)
5
0
GluA1/TARP γꢀ2 IC (nM) ±SEM (n)
5
0
GluA1/TARP γꢀ3 IC (nM) ±SEM (n)
>83,300 (3)
5
0
GluA1/TARP γꢀ4 IC (nM) ±SEM (n)
>83,300 (3)
5
0
GluA1 IC (nM) ±SEM (n)
>83,300 (2)
5
0
Data represent the mean ± SEM of n≥2 from FLIPR calcium imaging methods using CHOꢀS
cells transfected with human clonal constructs.
The identification of this modestly potent, although highly TARP γꢀ8 selective molecule,
encouraged us to assess its in vitro metabolic profile to understand which other properties would
require further optimization. High clearance (CL) in rat and modest potency were among the
early issues identified. Liver microsomal incubations indicated 5 underwent extensive oxidative
metabolism in rats and humans (Table 2). To characterize the extent of secondary metabolism, 5
was incubated with microsomes in the presence of appropriate cofactors for uridine 5'ꢀ
diphosphoꢀglucuronosyltransferase (UGT). A significant amount of glucuronidation was
observed in both human and rat liver microsomal preparations. To better understand this high
susceptibility to metabolism and clearance, metabolite identification in both rat and human
hepatocytes was performed. As shown in the chromatograms in Figure 3, the primary
metabolites in rat hepatocytes were glucuronidation products of the parent with and without
primary oxidation. In human hepatocytes, sulfate conjugation was observed. Finally, in vivo
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 58
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
pharmacokinetic (PK) studies in rats exhibited a clearance rate that exceeded hepatic blood flow.
These collective results suggested the phenol group of 5 was a likely site for conjugating
metabolism and therefore became a focal point for early molecular optimization.
0
1
2
3
4
5
6
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
6
Table 2. In vitro metabolism data for initial screen active 5.
Result
1.0 ± 2.6
In Vitro Metabolism Assay
human microsomal metab
% rat microsomal metab
8
%
84.0 ± 1.3
33 (n=1)
99 (n=1)
%
%
human UGT metab
rat UGT metab
Human hepatocyte incubations
Rat hepatocyte incubations
Rat CL (mL/min/kg)
Primarily conjugate metabolism (sulfation)
Primarily conjugate metabolism (glucuronidation)
234 ± 47
Data represent the mean ± SEM of n > 2 experiments unless otherwise noted.
ACS Paragon Plus Environment

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. LC/MS/MS Ion chromatogram showing a) compound 5 with no hepatocyte incubation
and the in vitro metabolites of compound 5 upon incubation with cryopreserved b) rat and c)
1
7
human hepatocytes.
Following initial structure activity relationship (SAR) exploration which identified that a
proton donor was required within the phenolic region for activity, our first major SAR iteration
cycle explored NH containing heterocycles to replace the phenol with bioisosteres that could
1
8
lower susceptibility to similar metabolic conversion. Our initial attempts replaced the phenol
with bicyclic heterocycles like indazole and benzothiazolone as represented in compounds 6ꢀ8
(Figure 4). Scheme 1 shows the synthetic route to compound 6 and is representative of the routes
used for 7 and 8. The key step is a Buchwald coupling between aminothiazole 9 and 5ꢀ
bromoindazole 10 followed by methylation of the amine linker and THP deprotection to provide
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 58
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
compound 6. The full synthetic schemes and experimentals of 7 and 8 are included within the
supplemental section.
0
1
2
3
4
5
6
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
6
7
8
Figure 4. Indazole and benzothiazolone analogs 6ꢀ8 were prepared to address conjugation of the
phenol group in 5.
a
Scheme 1. Syntheses of indazole analog 6.
a
b
9
10
11
c
1
2
6
a
o
Reagents and conditions: (a) tꢀbutylꢀXꢀphos, Pd(OAc) , amylene hydrate/water, 100 C; 11
2
o
8
%; (b) MeI, NaH, DMF, 0 C; 12 82%; (c) HCl, MeOH; 6, 94%.
ACS Paragon Plus Environment

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Replacing the phenol with an indazole or benzothiazolone provided up to 70ꢀfold
improvement in potency while maintaining TARP γꢀ8 dependency and selectivity. Also, these
groups provided a marked reduction of the secondary metabolism profile (Table 3). Overall,
clearance in rat remained high for compounds 6ꢀ8 and appeared to correlate with the extent of
microsomal metabolism. While the rat liver microsomal stability was improved slightly,
compounds 6ꢀ8 continued to exhibit rat CL that was near the rate of hepatic blood flow.
Importantly, however, no conjugate metabolism was observed upon incubating these compounds
with rat microsomes in the presence of appropriate cofactors for UGT, demonstrating that the
phenol replacements successfully eliminated this mode of in vitro metabolism. Further in vivo
investigation of compound 6, however, showed significantly lower exposure following 4 days of
dosing in rats (30 mg/kg, po), suggesting this compound might be inducing metabolism (e.g.
cytochrome P450; CYP) under these conditions (Table 4). Further CYP profiling of these
analogs confirmed that they acted as both substrates and inhibitors for human CYP1A2, as well
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
9
as, inducers of CYP1A2, and presumably the rat CYP homolog (Table 5).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 58
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
5
Table 3. In vitro potency, selectivity and metabolism profiles of compounds 6ꢀ8.
Compound
GluA1 IC50
% human
% rat
microsomal
metab ± SEM
Rat CL
mL/min/kg ±
SEM
GluA1/TARPγ
IC50 (nM) ±
SEM (n)
ꢀ8
GluA1/TARPγ
IC50 (nM) ±
SEM (n)
ꢀ2
(
(
nM) ± SEM microsomal
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
metab ± SEM
5
6
7
8
2970 ± 0.17 (6)
147 ± 0.02 (17)
353 ± 0.07 (11)
70 ± 0.01 (5)
>83,300 (5)
>83,300 (15)
>83,300 (9)
>83,300 (4)
>83,300 (2)
>83,300 (16)
>83,300 (10)
>83,300 (6)
81.0 ± 2.6
52.7 ± 8.2
36.2 ± 12.1
54.5 ± 2.4
84.0 ± 1.3
65.6 ± 5.8
32.5 ± 4.9
82.2 ± 1.7
234 ± 47
71.5 ± 2.1
61.3 ± 4.8
96.6 ± 11.4
Data represent the mean ± SEM of n > 2 experiments from the FLIPR calcium mobilization
assay using CHOꢀS cells transfected with human clonal constructs.
Table 4. Plasma exposure (Cmax and AUC) of compound 6 following 4 days of dosing in rats at
1
5
30 mg/kg, po.
Compound Day 1 Cmax Day 4 Cmax AUC Day 1
µg/ml) (µg/ml) (µg*h/mL)
AUC Day 4
(µg*h/mL)
(
6
3.74
±
0.0.5
1.93
±
1.1
52.3
±
8.1
3.15 ± 1.7
Data represent the mean ± SEM of n =2
Our second major SAR iteration cycle therefore focused on addressing the CYP1A2
challenges. To identify structural features that might diminish CYP1A2 interactions, ligands
2
0
were docked in the crystal structure of this enzyme using Glide v5.8. Based on the docking
model, several structural modifications were explored to disrupt the binding of this chemical
series within the CYP1A2 structure. Among these, replacement of the nitrogen linker with an
sp3ꢀhybridized carbon atom was proposed to disrupt recognition by CYP1A2 while
simultaneously maintaining overall molecular shape and size needed for interactions with TARP
γꢀ8. To test this hypothesis, we synthesized a prototype molecule, compound 13 (racemic
ACS Paragon Plus Environment

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mixture), that replaced the nitrogen linker of compound 8 with a carbon atom (see Figure 5).
This minor modification successfully maintained potency and selectivity. Assessment of the
individual enantiomers revealed activity to reside primarily in one enantiomer, (+)ꢀ13 (see Table
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
). The proposed binding poses of compounds 8 and (+)ꢀ13 are shown in Figure 6. Compound 8
appeared to fit well into the planar binding pocket similar to the coꢀcrystallized ligand, αꢀ
2
0
naphthoflavone, in the reported crystal structure. The benzothiazolone group appeared to have
stacking interactions with Phe226 and Phe260. However, in the case of compound (+)ꢀ13, the
3
introduction of nonꢀplanar sp carbon linker resulted in less optimal stacking interactions
indicated by a relatively poor docking score and interaction energy between the ligand and
protein. More importantly, the methyl group in compound (+)ꢀ13 was oriented toward a
relatively shallow region of the binding pocket leading to potential steric clash with residues like
Gly316, Asp320 and Thr223, also contributing to significant loss in binding interactions with
CYP1A2.
Alternatively, another approach was explored to destabilize CYP1A2 recognition by
expanding the size of the group on the nitrogen linker, causing severe steric impedance. To test
this hypothesis, we alkylated the linker nitrogen with a methylpyrazole sidechain as represented
in compound 14 (Figure 5). Figure 6 (panels C, F) shows the attempted docking pose within the
CYP1A2 binding pocket and the marked steric clash that is expected from the inclusion of this
sidechain.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 58
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
1
3
(-)-13
(+)-13
1
4
Figure 5. SAR analogs to address CYP1A2 interactions include compound 13, which contains a
chiral carbon linker, and 14, which contains a larger methylꢀpyrazole sidechain, were prepared.
The synthesis of compound 13 is shown in Scheme 2. The critical step is lithiation of
commercially available 1ꢀtetrahydropyranꢀ2ꢀylpyrazole and addition into methylketone 16 to
give alcohol 17. Dehydration of the alcohol to the olefin followed by hydrogenation gave
racemate 19. Finally, addition of 19 to 2ꢀfluoropyridine was accomplished under basic
conditions to give the fully assembled target which was deprotected. The enantiomers were
obtained using chiral chromatography purification conditions and the absolute stereochemistry of
(
+)ꢀ13 was unambiguously determined through Xꢀray crystallography.
The synthesis of compound 14 was accomplished in a similar manner as compound 6 (see
Scheme 3). Alkylation of intermediate 11 was accomplished with commercially available 3ꢀ
chloromethyl)ꢀ1ꢀmethylꢀpyrazole. Deprotection of the MOMꢀprotecting group provided 14 in
(
good overall yield.
a
Scheme 2. Synthesis of carbonꢀlinked series and analog 13.
ACS Paragon Plus Environment

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
5
16
17
d
e
1
8
19
20
f
g
1
3
(-)-13
(+)-13
a
Reagents and conditions: (a) Cs CO , MOMꢀCl, DMF; 16 75%; (b) 1ꢀtetrahydropyranꢀ2ꢀ
ylpyrazole, t-BuLi, THF, ꢀ78 C; 17 84%; (c) TFA, DCM; 18 92%; (d) H , 5%Pd(C), EtOAc; 19
2%,(e) LiOtꢀBu, DMF, 140 C; 20 46%; (f) (i) TFA, reflux; (ii) NH OH, THF; 13 71%; (g)
chiral chromatography; (ꢀ)ꢀ13 45%, (+)ꢀ13 44%.
2
3
o
2
o
9
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 3. Synthesis of compound 14.
0
1
2
3
4
5
6
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
b
a
1
1
2
1
1
4
a
o
Reagents and conditions: (a) 3ꢀ(chloromethyl)ꢀ1ꢀmethylꢀpyrazole, MeI, NaH, DMF, 0 C; 21
2%; (b) HCl, MeOH; 14 95%.
6
ACS Paragon Plus Environment

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: Docking of compounds 8 (left: A, D), (+)ꢀ13 (middle: B, E) and 14 (right:C, F) in the
CYP1A2 binding pocket. Protein atoms are represented as ball and sticks with carbon in yellow.
Ligands are represented as tubes (top panel: AꢀC) and CoreyꢀPaulingꢀKoltun (CPK) space filling
model (bottom panel: DꢀF). The heme iron is depicted as a cyan sphere. The active site available
for ligand binding is represented as a semiꢀtransparent orange surface.
Consistent with the hypothesized mode of interaction of 8 with CYP1A2, compounds 13
and 14 showed significantly reduced CYP1A2 inhibition (Table 5). These compounds also had
relatively less dependence on CYP1A2 for oxidative metabolism (moderate dependence on
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 58
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
CYP1A2 for microsomal metabolism) compared to those with an Nꢀmethyl group, in accordance
with the proposed loss in binding interactions with CYP1A2. Interestingly, these compounds did
not appear to cause significant induction of CYP1A2 (Figure 7). Within the SAR, the carbon
linker provided a consistent mitigation of the CYP1A2 interaction concerns in an efficient and
atomꢀeconomic fashion. While this structural modification resolved the CYP inhibition and
induction issues associated with compounds 6ꢀ8, rat liver microsomal metabolism of 13 (rac and
individual enantiomers) remained high and translated to high in vivo clearance for (+)ꢀ13 (Table
6).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 5. CYP1A2 inhibition and substrate profiling for compounds 5ꢀ14
Compound
hCYP1A2 Inhibition IC (nM) Relative
dependence
on
5
0
CYP1A2 #for microsomal
metabolism
High
5
6
7
8
ND
222 (n=2)
230
High
High
360
High
1
3
8680
Moderate
Moderate
Moderate
Low
(+)ꢀ13
>10000
4670
(ꢀ)ꢀ13
14
>10000
#
Based on decrease in the peak area of the major microsomal metabolite in presence of αꢀ
naphthoflavone (a selective CYP1A2 inhibitor). (>70% High; 30ꢀ70% Moderate; <30% ꢀ Low).
ND = not determined. Data represent n = 1 unless otherwise noted.
ACS Paragon Plus Environment

Page 19 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9
0.0
80.0
70.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
5
4
3
0.0
0.0
0.0
0.0
0
.2 uM
2 uM
0 uM
2
20.0
10.0
0
.0
5
6
7
8
13
(+)-13 (-)-13
14
(-)-25
Compound
Figure 7. Percent fold induction of CYP1A2 for SAR compounds compared to that of
omeprazole in human hepatocytes (<15% ꢀ No; >30% ꢀ Yes) as a function of compound
concentration. Significance is achieved when fold induction is >30%.
Our final major SAR iteration cycle focused on the high in vivo clearance. To better
understand the high oxidative metabolism, we again performed rat and human hepatocyte
incubation studies. This revealed that oxidative metabolism followed by glucuronidation was
occurring within the leftꢀhand side of the molecule as shown in Figure 8 (panel A).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
B
7
2%
+
O and glucuronide
1
0%
5
1%
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
3
%
9
%
3
%
2
%
Figure 8. Region of compound 13 where oxidation followed by glucuronidation were observed
following incubation in rat and human hepatocytes (panel A) and regioselectivity map for
®
predicted sites of CYP2D6ꢀmediated oxidative metabolism using the StarDrop model (panel B).
Sites are labelled with a predicted percentage of products formed due to metabolism at that
position.
Also, CYPꢀisoformꢀdependent metabolism analysis demonstrated a strong influence of
human CYP2D6 on in vitro metabolic turnover. Given the multiple potential sites of oxidative
metabolism, we explored in silico prediction models to help pinpoint the most probable sites to
prioritize our SAR efforts. Figure 8 (panel B) shows the human CYP2D6 predicted sites of
2
1
metabolism using the StarDrop® model by Optibrium Ltd. UK and these sites were presumed
to be the same for the rat CYP homolog. Based on these predictions, the 5ꢀposition of the
pyridine ring seemed to be the most probable site for metabolism and was verified
experimentally through metabolite identification of similar compounds within the SAR. Thus,
we explored substitutions on the 5ꢀposition of the pyridine to block metabolism. For example,
we prepared a 5ꢀfluoro analog (+)ꢀ22 to block oxidation at the 5ꢀposition of the pyridine (see
ACS Paragon Plus Environment

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 9). As shown in Table 6, this substitution decreased microsomal metabolism modestly
with a corresponding decrease in rat in vivo CL relative to compound 13. However, more
dramatic improvements of in vitro and in vivo metabolic stability were achieved when more
polar sidechains were incorporated. For example, by incorporating hydroxyethyl, substituted
hydroxypropyl, or 2ꢀhydroxyethyloxy substituents as shown in compounds 23, 24, 25,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
respectively, microsomal metabolism was significantly decreased.
These analogs also
demonstrated low in vivo clearance in rats (Table 6). While compounds 23 & 24 were studied
extensively, the remainder of this discussion focuses on (ꢀ)ꢀ25 which was ultimately chosen as
22
our clinical candidate. Compound (ꢀ)ꢀ25 (or LY3130481) showed excellent bioavailability in
rats (Table 7) and, consistent with the previously described SAR, showed no measurable
CYP1A2 inhibition (IC >10 µM) or CYP1A2ꢀdependent metabolism.
5
0
(+)-22
(+)-23
(+)-24
(-)-25
LY3130481
Figure 9. SAR analogs designed to lower in vivo CL
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 58
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
Table 6. Potency, selectivity, and metabolism data for compounds 13ꢀ25.
Compound
% human
microsomal
metab
% rat
microsoma
l metab
Rat CL
GluA1/TARP γ
IC50 (nM) ±
SEM (n)
ꢀ8
GluA1/TARPγ
2 IC50 (nM) ±
SEM (n)
ꢀ
GluA1 IC50
(nM) ± SEM
0
1
2
3
4
5
6
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)
(mL/min/kg)
racꢀ13
122 ± 0.02 (3)
>83,300 (3)
>83,300 (3)
41.9 ± 2.2
84 ± 1.3
115 ± 2.4
(
+)ꢀ13
50.9 ± 0.02 (7)
615 ± 0.23 (10)
53.9 ± 0.01 (3)
53.6 ± 0.01 (5)
160 ± 0.04 (17)
61.6 ± 0.01 (12)
2820 ± 456 (9)
65.3 ± 0.01 (18)
1790 ± 306(12)
>83,300 (2)
>83,300 (3)
>83,300 (3)
>83,300 (2)
>9,260 (14)
>9,260 (8)
>9,260 (9)
>83,300 (3)
>83,300 (2)
>9,260 (9)
>83,300 (3)
>9,260 (10)
>83,300 (4)
>9,260 (11)
36.2 ± 0.9
44.2 ± 0.8
49.7 ± 0.1
19.6 ± 1.5
23.6 ± 2.6
15.2 ± 3.6
ND
95.3 ± 0.5
96.0 ± 0.7
98.5 ± 0.1
67.4 ± 1.5
43.7 ± 1.3
47.4 ± 3.5
ND
99 ± 1.4
144 ± 10.4
ND
(ꢀ)ꢀ13
1
4
(+)ꢀ22
42.2 ± 3.4
6.0 ± 0.0.4
16.3 ± 0.0.4
ND
(
+)ꢀ23
+)ꢀ24
(
25,700 ± 423 (3)
>9,260 (8)
(ꢀ)ꢀ24
ꢀ)ꢀ25
(+)ꢀ25
(
68,500 ± 566 (2)
>9,260 (11)
16.4 ± 6.9
ND
32.3 ± 8.1
ND
11.4 ± 0.8
ND
Data represent the mean + SEM of n >2 experiments from a FLIPR calcium mobilization assay
using CHOꢀS cells transfected with human clonal constructs. ND = not determined.
The synthesis of compound (ꢀ)ꢀ25 was accomplished as described in Scheme 4. As
shown previously, the advanced pyrazole intermediate was alkylated under SnAr conditions with
the appropriate substituted fluoropyridine (26). The THP and MOM protecting groups were then
removed in good yield to provide racemic 25 which was then separated under chiral
chromatography conditions. Consistent with the results from compound 13, one enantiomer (ꢀ)ꢀ
25 was significantly more active than the other enantiomer (+)ꢀ25 (see Table 6). The absolute
stereochemistry of (+)ꢀ24 and (ꢀ)ꢀ25 was unambiguously determined through XꢀRay
crystallography and determined to be in the (S)ꢀconfiguration (Figure 10, panels A & B,
respectively). The syntheses of compounds (+)ꢀ22, (+)ꢀ23, and (+)ꢀ24 were accomplished in a
ACS Paragon Plus Environment

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
similar fashion and the full schemes and experimentals are available in the supplemental
information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Molecular conformation and stereochemistry observed in the crystal structures of (+)ꢀ
24 (panel A) and (ꢀ)ꢀ25 (panel B).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 4. Synthesis of candidate compound (ꢀ)ꢀ25.
0
1
2
3
4
5
6
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
26
a
27
b
1
9
28
2
9
c
d
2
5
(-)-25
a
o
Reagents and conditions: (a) Cs CO , DMF; 28 91%; (b) LiOtꢀBu, DMF, 140 C; 29 74%; (c)
i) TFA, reflux; (ii) NH OH, THF; 25 76%; (d) chiral chromatography; (ꢀ)ꢀ25 45%.
2
3
(
4
ACS Paragon Plus Environment

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 7. Rat bioavailablity data for (ꢀ)ꢀ25.
Mean
Dose
Parameter
AUC
Vdss
Units
ng*Hours/mL
mL/kg
(n = 3)
1470
1280
11.4
4220
1.3
SEM
110
139
0.8
1
3
mg/kg
IV
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CL
mL/min/kg
ng*Hours/mL
Hours
AUC
Tmax
Cmax
%F
148
0.4
mg/kg
PO
ng/ml
628
36
97
6.9
Abbreviations: AUC=area under the plasma concentrationꢀtime curve, Vdss =volume of
distribution at steady state, CL= total clearance, T1/2 = plasma disappearance halfꢀlife, Tmax= time
of peak plasma concentration, Cmax= peak plasma concentration; %F= absolute oral
bioavailability.
Having now developed a TARPꢀdependent AMPA receptor antagonist (ꢀ)ꢀ25 with
23,24
appropriate PK characteristics and molecular properties consistent with CNS drugs,
we asked
whether it would demonstrate anticonvulsant efficacy at doses and exposures significantly lower
than those that impair motor coordination. Oral administration of (ꢀ)ꢀ25 doseꢀdependently
protected rats from clonic convulsions induced by subcutaneous administration of
pentylenetetrazole (PTZ) with an ED of 1.7 mg/kg (95% Confidence limits: 0.3 – 11.5, Table
5
0
8). In contrast, the nonꢀTARPꢀdependent AMPA receptor antagonist, perampanel, had an ED50
value approximately 10ꢀfold higher (10.6 mg/kg) than (ꢀ)ꢀ25. The anticonvulsant effects of (ꢀ)ꢀ25
were not due to any change in disposition of PTZ in rats, nor did PTZ change the disposition of
(ꢀ)ꢀ25 in either plasma or brain (data not shown).
The effects of (ꢀ)ꢀ25 on motor impairment were assessed in the same experiment using an
2
5
inverted screen (IS) test prior to administering the convulsant PTZ. Importantly, and fully
consistent with our hypothesis, (ꢀ)ꢀ25 up to a dose of 100 mg/kg did not significantly impair the
ability of the rats to climb to the top of screen when inverted (Table 9). In contrast, perampanel
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 58
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
full impaired behavior in this test at 30 mg/kg. The TD for impairment by perampanel in this
5
0
assay was determined to be 3.8 mg/kg. Based upon the efficacy in the PTZ assay and side
effects in the inverted screen assay, a protective index (PI = inverted screen TD50 / PTZ ED50)
can be calculated. This yielded a remarkably high protective index for (ꢀ)ꢀ25 (PI> 58) while the
PI for perampanel was equal to 0.4.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 8. Effects of orallyꢀadministered (ꢀ)ꢀ25 on blocking PTZ (35 mg/kg, sc)ꢀinduced
convulsions in rats (n=10 per group) and its corresponding receptor occupancy as a function of
dose (n=4 per group).
%
Receptor
Brain
Exposure
(ng/g) ±
Plasma
Exposure
ng/mL) ± SEM
Dose (mg/kg, po)
of (ꢀ)ꢀ25
% block of clonic
convulsions in rats*
Occupancy
(RO) ±
(
#
#
SEM
SEM
Vehicle
0
ꢀꢀ
ꢀꢀ
ꢀꢀ
1
3
10
40*
80*
100*
ꢀꢀ
377 ± 18
1269± 106
3671± 469
ꢀꢀ
45 ± 7
69 ± 7
68 ± 1
96 ± 10
133 ± 11
477± 69
1526 ± 188
2400 ± 701
30
Compound (ꢀ)ꢀ25 was administered po, 30 minutes prior to PTZ. PTZ was dissolved in saline;
ꢀ)ꢀ25 was suspended in 10% Acacia/0.05% DAF. *significantly different than PTZ alone
(
P<0.05 Fishers exact test.# data from satellite animals (n>4).
ACS Paragon Plus Environment

Page 27 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 9. Effects of (ꢀ)ꢀ25 and perampanel in the rat inverted screen assay of motor function.
Group
Inverted Screen Score
Plasma Exposure
(ng/mL)
Vehicle
0.4 ± 0.4
0.4 ± 0.4
0.4 ± 0.4
2.0 ± 0*
ꢀꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
(
ꢀ)ꢀ25, 10 mg/kg, po
ꢀ)ꢀ25, 100 mg/kg, po
3360 ± 265
8342 ± 216
1780 ± 201
Perampanel at 30
mg/kg, po
Data are means ± SEM in ≥4 rats/group. *Significantly different than vehicle control values.
We also sought to identify a molecule within the scaffold that possessed suitable
properties as a tracer to measure the level of preclinical target engagement (GLUA1+TARP γꢀ8)
2
6
and possibly generate a PET ligand for clinical development. To this end, we evaluated several
compounds within the structural series and identified compound 37 which was potent at TARP
γꢀ8 with an IC of 22 ± 0.01 nM and showed a desirable and rapid clearance profile. The
5
0
synthesis of the 37 is shown in Scheme 5. Starting from commercially available compound 30,
the thiazolone was protected with a trimethylsilylethoxymethyl (SEM) group and the nitro group
was reduced and functionalized as a thiourea (33). Removal of the benzoyl group followed by
condensation with bromoacetyl pyridine provided intermediate 35. Finally, alkylation of the
amino linker with 1ꢀbromoꢀ2ꢀethoxyꢀethane and SEM group deprotection provided the tracer
compound 37.
We have reported elsewhere a technique of using nonꢀradiolabelled molecules to measure
2
7,28
occupancy through the use of LCMS/MS methods.
To measure receptor occupancy (RO),
compound (ꢀ)ꢀ25 was preꢀdosed orally to rats 30 min prior to an intravenous bolus of the tracer,
compound 37. Forty minutes after tracer administration, the brain was collected and the amount
of tracer levels within the hippocampus (total binding tissue) and cerebellum (nonspecific
binding tissue or null region) were quantified. These tracer levels were used to calculate receptor
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 58
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
29
occupancy using the wellꢀestablished ratio method. Table 8 shows the relationship of the brain
exposure and receptor occupancy between doses of 1ꢀ30 mg/kg of (ꢀ)ꢀ25. Increasing oral doses
of (ꢀ)ꢀ25 produced increases in drug exposure of the brain, resulting in increases in target
occupancy for (ꢀ)ꢀ25. Within this experiment it is noted that the RO was similar at 3 and 10
mg/kg despite about a 3ꢀfold increase in brain exposure across the same dose range. This is
likely due to slightly more error associated with the RO data at 3 mg/kg dose group and would be
expected to average out across multiple experiments. Thus, at the drug exposures required to
have efficacy in the PTZ convulsion assay, the receptor occupancy of (ꢀ)ꢀ25 is estimated to be
between 45ꢀ68%.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 5. Synthesis of LC/MS preclinical receptor occupancy tracer 37.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
30
31
32
c
d
e
3
3
34
g
f
3
5
3
6
37
a
o
Reagents and conditions: (a) SEMꢀCl, NaH, DMF, 0 C; 31 81%; (b) H , PdꢀC, Ethanol; 32
82%;(c) benzoyl isothiocyanate, acetone; 33 93%; (d) NaOH, water/methanol; 34 99%; (e) 2ꢀ
bromoꢀ1ꢀ(2ꢀpyridyl)ethanone, DIEA, DMF, 100 C; 35 69%. (f) 1ꢀbromoꢀ2ꢀethoxyꢀethane, NaH,
DMF, 0 C; 36 66%; (g) (i) TFA, (ii) NH OH, THF; 37 91%.
2
o
o
4
CONCLUSION
We established a series of high throughput screening assays which allowed us to detect
novel AMPA receptor antagonists that are specific for AMPA receptors containing the auxiliary
ACS Paragon Plus Environment