Design, Synthesis, and Preclinical Evaluation of 3-Methyl-6-(5-thiophenyl)-1,3-dihydro-imidazo[4,5-b]pyridin-2-ones as Selective GluN2B Negative Allosteric Modulators for the Treatment of Mood Disorders

Article abstract of DOI:10.1021/acs.jmedchem.9b02113

Selective inhibitors of the GluN2B subunit of N-methyl-d-aspartate receptors in the ionotropic glutamate receptor superfamily have been targeted for the treatment of mood disorders. We sought to identify structurally novel, brain penetrant, GluN2B-selective inhibitors suitable for evaluation in a clinical setting in patients with major depressive disorder. We identified a new class of negative allosteric modulators of GluN2B that contain a 1,3-dihydro-imidazo[4,5-b]pyridin-2-one core. This series of compounds had poor solubility properties and poor permeability, which was addressed utilizing two approaches. First, a series of structural modifications was conducted which included replacing hydrogen bond donor groups. Second, enabling formulation development was undertaken in which a stable nanosuspension was identified for lead compound 12. Compound 12 was found to have robust target engagement in rat with an ED70 of 1.4 mg/kg. The nanosuspension enabled sufficient margins in preclinical toleration studies to nominate 12 for progression into advanced good laboratory practice studies.

Full text of DOI:10.1021/acs.jmedchem.9b02113

Subscriber access provided by The University of Melbourne Libraries
Article
Design, Synthesis and Preclinical Evaluation of 3-Methyl-6-(5-
thiophenyl)-1,3-dihydro-imidazo[4,5-b]pyridin-2-ones as Selective GluN2B
Negative Allosteric Modulators for the Treatment of Mood Disorders
Christa C Chrovian, Akinola Soyode-Johnson, Brice Stenne, Daniel J. Pippel, Jeffrey Schoellerman,
Brian Lord, Alexandra S Needham, Chunfang Xia, Kevin J Coe, Kia Sepassi, Freddy Schoetens, Brian
Scott, Leslie Nguyen, Xiaohui Jiang, Tatiana Koudriakova, Bartosz Balana, and Michael A Letavic
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b02113 • Publication Date (Web): 31 Jul 2020
Downloaded from pubs.acs.org on August 2, 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 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis and Preclinical Evaluation of 3-
Methyl-6-(5-thiophenyl)-1,3-dihydro-imidazo[4,5-
b]pyridin-2-ones as Selective GluN2B Negative
Allosteric Modulators for the Treatment of Mood
Disorders
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
AUTHOR NAMES. Christa C. Chrovian,* Akinola Soyode-Johnson, Brice Stenne,
Daniel J. Pippel, Jeffrey Schoellerman, Brian Lord, Alexandra S. Needham, Chungfang
Xia, Kevin J. Coe, Kia Sepassi, Freddy Schoetens, Brian Scott, Leslie Nguyen, Xiaohui
Jiang, Tatiana Koudriakova, Bartosz Balana, and Michael A. Letavic.
AUTHOR ADDRESS Janssen Research & Development, LLC, 3210 Merryfield Row,
San Diego, California 92121-1126, United States.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 88
1
2
3
4
5
6
7
8
KEYWORDS. NMDA, NR2B, antagonist, enabling formulation, receptor occupancy,
autoradiography, biodistribution, non-clinical studies, toxicokinetics, GluN2B NAM.
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
ABSTRACT. Selective inhibitors of the GluN2B subunit of N-methyl-D-aspartate
receptors in the ionotropic glutamate receptor superfamily have been targeted for the
treatment of mood disorders. We sought to identify structurally novel, brain penetrant,
GluN2B-selective inhibitors suitable for evaluation in a clinical setting in patients with
major depressive disorder. We identified a new class of negative allosteric modulators of
GluN2B that contain a 1,3-dihydro-imidazo[4,5-b]pyridin-2-one core. This series of
compounds had poor solubility properties and poor permeability, which was addressed
utilizing two approaches. First, a series of structural modifications was conducted which
included replacing hydrogen bond donor groups. Second, enabling formulation
development was undertaken in which a stable nanosuspension was identified for lead
compound 12. Compound 12 was found to have robust target engagement in rat with an
ED70 of 1.4 mg/kg. The nanosuspension enabled sufficient margins in pre-clinical
ACS Paragon Plus Environment
2

Page 3 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
toleration studies to nominate 12 for progression into advanced good laboratory practice
6
7
8
studies.
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
INTRODUCTION. The neurotransmitter glutamate is prevalent to such an extent in the
central nervous system (CNS) of complex organisms that it is involved in nearly all
aspects of normal CNS functioning and it is used by most excitatory synaptic connections
in the mammalian brain. Glutamate regulates the activity of two classes, or superfamilies,
of receptors: metabotropic glutamate receptors that are G protein-coupled receptors
(GPCRs) and ionotropic glutamate receptors (iGluRs) that are ion channels. Ionotropic
glutamate receptors function as tetramers and each contain two clamshell-shaped
extracellular N-terminal domains, a ligand binding domain, a transmembrane domain and
an intracellular C-terminal domain.1 The most well studied iGluRs are the α-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA)
receptors. A key distinction between the two classes is that AMPA receptors are active
during basal synaptic transmission whereas NMDA receptors are activated only after a
depolarization event of the surrounding neuronal membrane, because Mg2+ blocks the
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 88
1
2
3
4
5
6
7
8
9
pore under basal conditions. When the Mg2+ block is relieved after a depolarization event,
the pore becomes permeable to Ca2+. The influx of Ca2+ through NMDA receptors triggers
signaling cascades and regulates gene expression that is critical for different forms of
synaptic plasticity, including both long-term potentiation (LTP)2 and long-term depression
(LTD).3 Growing evidence suggests that disturbances in brain and synaptic plasticity
mediated by the glutamate system contribute to the pathophysiology of mood disorders.4-6
The NMDA receptor tetramers are comprised of two GluN1 and two GluN2 subunits.
Four unique GluN2 subunits have been identified and they have been assigned the
nomenclature GluN2A-GluN2D. The receptors require co-activation by both glycine (D-
serine) and glutamate, which bind to GluN1 and GluN2, respectively.1,7
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
The four GluN2 subunits of the NMDARs exhibit distinct distribution patterns across the
mammalian brain, depending on the species and individual’s development age. For adult
mammals, GluN2B subunits are expressed primarily in the cortex, striatum, and notably
the hippocampus, a region consistently demonstrating the highest receptor density
throughout the full life cycle.8-9
ACS Paragon Plus Environment
4

Page 5 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The NMDA receptor has been studied extensively since the early 1980s when it was
determined that Mg2+ confersa blockade of the ion channel under basal conditions, which
is the salient feature conferring its voltage dependence.10-11 Selectively blocking the
function of GluN2B-containing receptors, on the other hand, was targeted later, when it
was determined that ifenprodil (1) discriminates the subtypes of NMDAR.12 Compound 1
is an anti-hypertensive agent discovered in the 1960s that is a highly potent inhibitor of
the GluN2B subunit, in addition to being a significant modulator of alpha adrenergic and
sigma receptors.13
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
Compound 1 is a negative allosteric modulator (NAM) that was found to bind to the
NMDAR complex at a site positioned between the GluN1 and GluN2B subunits.14 The
site is not present on the other three GluN2 subunits and inhibitors binding at this interface
have excellent selectivity for GluN2B over GluN2A, GluN2C and GluN2D. This feature
has enabled the scientific community the opportunity to evaluate the safety and
pharmacology of several NMDAR small molecule modulators that are exquisitely
selective for GluN2B versus the other three GluN2 subunits.15-17 Several NAMs have
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 88
1
2
3
4
5
6
7
8
been evaluated in clinical settings, two of which were assessed in patients with
depressive disorders.
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
The first reported GluN2B-selective NMDAR inhibitor assessed in patients diagnosed
with major depressive disorder was CP-101,606 (2, traxoprodil).17 Compound 2 was
originally discovered by Pfizer through a medicinal chemistry program based on 1. It was
identified as an inhibitor of NMDA receptor function and was shown to be neuroprotective
in vitro against glutamate toxicity.18 As the tetrameric structure of the NMDA receptor was
unknown at the time, the selectivity of 2 for the GluN2B subunit was not revealed until
later.19 Compound 2 progressed into clinical trials in subjects with traumatic brain injury.20
Several years later it was assessed in patients diagnosed with treatment resistant
depression (TRD) in Phase II clinical trials as an add on therapy to paroxetine. It achieved
a positive response in 60% of subjects in a current depressive episode.17 Compound 2
has limited oral bioavailability and its route of administration was intravenous (i.v.). The
cardiovascular safety of 2 was also limited due to observed QT interval prolongation, and
trials were ceased.
ACS Paragon Plus Environment
6

Page 7 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CERC-301 (3, Rislenemdaz, MK-0657) is an orally available GluN2B-selective NAM
discovered by Merck, and was originally intended for Parkinson’s disease or pain and
related disorders.21-23 The results of a Phase II study were reported in 2009 in which 3 did
not result in motor function improvement in patients with Parkinson's disease.22 It was
later developed for the treatment of major depressive disorder (MDD) and evaluated in
two Phase II trials in patients with MDD.23-26 The drug was generally well tolerated and
there were no discontinuations due to adverse events (AE). Both trials failed to meet their
Primary Endpoints, although in the second of the two Phase II studies, a notable
improvement in symptoms was measured on Day 2 as determined by the secondary
rating scale called the Hamilton Depression Rating Scale (HDRS-17).25,26
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
The goal of medicinal chemistry efforts in our drug discovery program was to identify
novel GluN2B-selective NAMs that were well tolerated and suitable for once a day oral
dosing in the clinic, in order to assess their performance in patients with psychiatric
(mood) disorders.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 88
1
2
3
4
HO
HO
O
5
6
7
F
O
N
N
N
H
OH
N
N
OH
OH
8
9
N
CP-101,606
Traxoprodil
CERC-301
Rislenemdaz
Ifenprodil
1
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
2
3
Figure 1. GluN2B selective negative allosteric modulators.
RESULTS AND DISCUSSION:
A high-throughput screening (HTS) campaign identified compounds that inhibited
functional activity of human GluN1a/GluN2B channels expressed in CHO cells as
measured by Ca2+ flux (FLIPR) elicited by addition of co-agonists glutamate and glycine.
Confirmed hits were counter-screened in heterologous cells expressing human
GluN1a/GluN2A. From this effort, a selective series of GluN2B inhibitors containing a 1,3-
dihydro-imidazo[4,5-b]pyridin-2-one core was identified. Compound 4 was potent in the
FLIPR assay (hGluN2B IC50 = 22 nM) and was selective over the GluN2A, GluN2C,
GluN2D subunits (IC50s all > 10 μM). Two salient features of 4, however, were its poor
solubility (<4 μM at pHs 2 and 7) and high efflux ratio (B-A/A-B =18). Compound 4 has
two hydrogen bond donors and a characteristically insoluble urea moiety27 imbedded in
ACS Paragon Plus Environment
8

Page 9 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the core, likely contributing to its P-glycoprotein (Pgp) substrate potential, indicated by its
high efflux ratio in the MDCK cell line transfected with MDR1. Compounds with efflux
ratios >10 are expected substrates for Pgp and are less likely to cross the blood brain
barrier (BBB) and access the brain compartment. We were therefore encouraged to find
that eliminating the H-bond donors by replacing the cyclopropyl- of the secondary amide
with azetidine to provide instead a tertiary amide, and adding N-methyl substitution at the
urea N-H, resulted in compounds such as 5 that have comparable GluN2B potency to 4.
Compound 5 had a GluN2B IC50 of 112 nM and measurable solubility (23 μM at pH 7).
Compound 5 also had an improved efflux ratio (9.3) and a higher plasma free fraction
(5.2%) than 4 (0.4%). Neither 4 nor 5 displayed hERG channel inhibition. The majority of
subsequent 1,3-dihydro-imidazo[4,5-b]pyridin-2-one core compounds that were
synthesized and assessed for hERG inhibition, including all of those described in this
manuscript, also had hERG IC50 values >10 μM.
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
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 88
1
2
3
4
5
3
4
H
B
3a
7a
A
N
CH3
5
N
2
N
6
O
N
N
1
O
6
1
O
7
6
N
2
O
1
7
8
9
3
HN
CF3
N
CF3
5
4
First SAR
From HTS
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
hGluN2B IC50 = 112 nM
solubility at pH 2 and 7 = 19 / 23 M
Efflux ratio: 9.3
hGluN2B IC50 = 22 nM
solubility at pH 2 and 7 = <4 M
Efflux ratio: 18
hERG IC50 > 10 M
plasma protein binding:
94.8% bound (human)
hERG IC50 > 10 M
plasma protein binding:
99.6% bound (human)
Figure 2. GluN2B selective negative allosteric modulators. (A) Compound 4 was
identified from a high throughput screen. (B) Compound 5 shows improved aqueous
solubility, efflux ratio and human plasma protein binding compared to HTS hit 4.
The early SAR of the 1,3-dihydro-imidazo[4,5-b]pyridin-2-one series indicated that poor
solubility was one of the prominent features of the core, which was predicted to be an
obstacle. However, low metabolic turnover in vitro was also observed, a feature that was
preferred. One trend that we focused on for optimization was the high efflux ratios that
appeared to be the result of Pgp-mediated efflux. It was rationalized that if drug
candidates could be identified that have both high intrinsic permeability and relatively low
efflux liability, the need for high thermodynamic solubilities could be somewhat obviated.28
Our strategy for the 1,3-dihydro-imidazo[4,5-b]pyridin-2-ones was to negotiate solubility
ACS Paragon Plus Environment
10

Page 11 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
properties in such a way that suitable absorption, distribution, metabolism and excretion
(ADME) in this series could be achieved, as well as good GluN2B potencies and brain
uptake. Since much of the early SAR on the 1H-pyrrolo[3,2-b]pyridines has already been
disclosed,29 a deeper focus of this disclosure will be Late Lead Optimization SAR, in vivo
data and formulations that enabled clinical candidate selection.
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
The first 1,3-dihydro-imidazo[4,5-b]pyridin-2-one compounds prepared had a meta-
trifluoromethylphenyl group at the 6-position of the 1,3-dihydro-imidazo[4,5-b]pyridin-2-
one core (referred to as left hand side (LHS) throughout this discussion, Table 1: 5-7).
These analogs had moderate human GluN2B potencies (IC50 values of 47-118 nM) and
their efflux ratios were moderately high, ranging from 7-9 (Table 1). In the presence of a
Pgp inhibitor, their permeability was very good, ranging from Papp = 35 to 68 (x 10-6
cm/sec), suggesting that the measured cellular efflux potential is Pgp mediated. N-
Azetidine containing amides 5 and 7 had comparable solubilities in aqueous buffer,
although N,N-dimethylamide 6 had measurably less (6 μM at pH 2 and 7).
Table 1. GluN2B potency and affinity and in vitro ADME for 3-methyl-6-(3-
(trifluoromethyl)phenyl)-1,3-dihydro-imidazo[4,5-b]pyridin-2-ones.
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 88
1
2
3
4
5
6
4
CH3
3a
7a
3
N
5
N
2
O
N
1
6
1
O
7
7
8
9
2
3
NR2
CF3
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
Solub
GluN2B
GluN2B
Efflux
ratio e
f
cpd NR2
at pH
2/7c
ER (hu/r)d
Papp
IC50a (nM) Kib (nM)
N
5
97, n=1
47, n=1
19 ± 7
19 / 23 <0.3 / <0.2 9.0
39
35
N
6
19, n=1
6 / 6
0.38 / 0.31 7.0
N
7
118 ± 82
25 ± 21
17 / 16 <0.3 / <0.2 7.0
68
F
a IC50 values were determined by a calcium mobilization assay in inducible CHO T-Rex
cells heterologously expressing the hGluN1a/GluN2B receptor. Values reported are the
b
mean ± SD of at least three experiments unless otherwise stated. Ki values were
determined using a radioligand competitive binding assay in rat cortex membranes using
3-[3H]
1-(azetidin-1-yl)-2-[6-(4-fluoro-3-methyl-phenyl)pyrrolo[3,2-b]pyridin-1-
yl]ethanone.29 Values reported are the mean ± SD of at least three experiments unless
otherwise stated. c Kinetic solubility in aqueous buffer. d Stability in human and rat liver
e
microsomes. Data reported as microsomal extraction ratio. Efflux ratio was measured
ACS Paragon Plus Environment
12

Page 13 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
using MDCK-MDR1 cells transfected with the P-glycoprotein (MDR-1). The efflux ratio is
f
reported as the ratio of velocities (B-A/A-B). Papp is A to B permeability (10-6 cm/s)
7
8
measured in the presence of 2 μM of the PgP inhibitor elacridar.
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
A strategy to lower efflux ratios emerged at around this point in the SAR, when
concurrent work on a second series of GluN2B NAMs containing a 1H-pyrrolo[3,2-
b]pyridine core indicated that somewhat lower efflux ratios were measured for many of
the examples had thiophenes on the LHS as phenyl bioisosteres.29 An example from that
report in which 4-fluoro-3-methylphenyl was applied to the 1H-pyrrolo[3,2-b]pyridine core
LHS (8, Figure 3) had an efflux ratio of 5.4 and plasma Cmax of 1380 ng/mL in rat, whereas
the comparator molecule with a 5-(trifluoromethyl)thiophen-2-yl group in place of 4-fluoro-
3-methylphenyl (9), had an efflux ratio of 3.5 and higher Cmax of 2330 ng/mL in rat (Figure
3).29 We were thus prompted to apply these learnings and revisit thiophenes with the 1,3-
dihydro-imidazo[4,5-b]pyridine-one core, hopeful that the combination of high plasma
exposure and low efflux ratios in the pyrrolo[3,2-b]pyridine series would translate to the
1,3-dihydro-imidazo[4,5-b]pyridine-one core, and also lead to pharmacologically relevant
brain exposures.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 88
1
2
3
4
5
6
N
8
N
N
N
O
O
S
F
N
N
F3C
CH3
9
7
8
9
hGluN2B IC50 = 47 nM
Solubility at pH 2 and 7 = >400 / 100 M
Efflux ratio: 3.5
hGluN2B IC50 = 25 nM
Solubility at pH 2 and 7 = >400 / 264 M
Efflux ratio: 5.4
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
in vivo PK rat 5 mg/kg p.o.
AUC: 4728 h*ng/mL
Cmax: 2330 ng/mL
in vivo PK rat 5 mg/kg p.o.
AUC: 1587 h*ng/mL
Cmax: 1380 ng/mL
Figure 3. Summary of 1H-pyrrolo[3,2-b]pyridine series SAR.29
The thiophene LHS SAR for the 1,3-dihydro-imidazo[4,5-b]pyridin-2-one core indicated
that methyl- or fluoromethyl- groups incorporated at the 5-position (of thiophen-2-yl) were
generally effective in both human GluN2B and rat GluN2B binding assays (Table 2).
Analogs had poorer IC50 and Ki values when the ring substitution was at the 4-position of
the thiophene compared to the 5-position (10 vs 11). Trifluoromethyl thiophene analogs
with N,N-dimethylacetamid- or 2-(azetidin-1-yl)-2-oxoethyl groups were potent NAMs (12:
IC50 = 31 nM, 13: IC50 = 29 nM) with robust rat GluN2B Ki values (12: Ki = 16 nM, 13: Ki
= 12 nM). 5-Cyclopropyl substituted thiophenes were tolerated (17: IC50 = 81 nM, n=1)
but were not as potent as the corresponding methyl (11: IC50 = 25 nM) or fluoromethyl
(12: IC50 = 31 nM, 15: IC50 = 6 nM) analogs.
ACS Paragon Plus Environment
14

Page 15 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The efflux ratios for methyl and trifluoromethyl-substituted thiophenes were in the
desirable range of 1.4 to 3.7 (10-14) and lower than for compounds 5-7, as we had hoped.
Intrinsic permeability values (Papp) for 10-14 were excellent (37 to 68 x10-6 cm/sec). Given
their low efflux ratios (<4) and high Papp, compounds 10-14 were predicted to be BBB
penetrant. Difluoromethyl thiophenes 15 and 16 had higher efflux ratios (5.8 and 9.1,
respectively) but they were found to decompose in aqueous buffer after 24 h at pHs 2
and 7. So despite its excellent potency, compound 15 (hGluN2B IC50 = 6 nM) was
deprioritized due to its chemical instability.
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
The trifluoromethyl-substituted phenyl LHS (5-7) and thiophene LHS (12-14) showed
excellent stability in human and rat liver microsomes. Extraction ratios (ER) were
moderate for difluoromethyl-substituted thiophenes (15-16) and poor for 4- and 5-methyl-
substituted thiophenes (10-11).
The solubility of compounds 5-7 was moderately low at both pHs 2 and 7. Examples
that contained a trifluoromethylthiophene (12-14) displayed low solubility below the limit
of the assay (<4 μM).
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 88
1
2
3
4
5
6
7
8
Table 2. GluN2B potency and affinity and in vitro ADME for 3-methyl-6-(thiophenyl)-1,3-
dihydro-imidazo[4,5-b]pyridin-2-ones.
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
CH3
N
N
O
S
N
O
R
NR2
Solub
at pH
2/7c
Thio- GluN2B
GluN2B
Efflux
ratio e
f
cpd NR2
ER (hu/r)d
Papp
phene IC50a (nM) Kib (nM)
S
N
10
216 ± 85
25 ± 7
270 ± 59 31 / 25 0.59 / 0.82 3.5
38
37
S
N
11
44 ± 31
16 ± 7
48 / 39 0.56 / 0.78 3.2
S
N
12
31 ± 16
29 ± 9
<4 / <4 <0.3/<0.17 2.1
47
54
CF3
<0.3 /
S
N
13
12 ± 3
<4 / <4
1.4
CF3
<0.17
ACS Paragon Plus Environment
16

Page 17 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
S
N
14
15
16
71 ± 30
6 ± 5
21 ± 9
19 ± 12
59± 25
<4 / <4 <0.3/<0.17 3.7
68
41
51
F
CF3
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
S
N
decomp 0.46/0.58
5.8
CF2H
S
N
79 ± 63
decomp <0.3 / 0.52 9.1
FH2C
S
N
17
81, n=1
63, n=1
nd
nd
nd
nd
a-f Same as Table 1.
Ultimately, compounds 12-14 were prioritized over the other examples due to their low
microsomal turnover and GluN2B activity, both in the human recombinant cell line and in
rat cortex binding assay. It is of note, however, that 12-14 were the least soluble of the
series and possible challenges were anticipated due to their poor solubility properties.
Compounds 15-16 looked promising, but were not stable at pH 2 or 7 in aqueous buffer
after 24 h. Compound 11 was soluble in buffer up to 39 and 48 μM at pHs 2 and 7,
respectively, making it the most soluble of present examples. However, it had moderately
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 88
1
2
3
4
5
6
7
8
9
high turnover in liver microsomes, with a human ER of 0.56. If 11 were to exhibit a good
correlation between in vitro and in vivo PK (IVIVC) in rat, this predicted ER value would
not preclude a therapeutically relevant Cmax and oral bioavailability in human. Although it
was foreseeable that intestinal dissolution may not be a limiting factor in absorption of 12-
14, in which case plasma drug concentrations could still reach levels well in the range of
many drug substances that demonstrate robust bioavailabilities, due to the uncertainty
surrounding proper absorption of 12-14 into systemic circulation, 11 was also selected for
in vivo progression.
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
The synthesis of 5-7 and 10-11 was conducted using a sequence beginning with 2-
chloro-5-bromo-3-nitropyridine, which underwent an SNAr reaction with aqueous
methylamine at rt. The nitro group was reduced to the primary amine with zinc in the
presence of ammonium chloride and the 1,3-dihydro-imidazo[4,5-b]pyridine-one core was
formed in the next step using 1,1’-carbonyldiimidazole (CDI). Intermediate 18 was
deprotonated with NaH and underwent SN2 addition to ethyl 2-bromoacetate to give 19 in
88% yield. Lithium hydroxide was used to saponify ester 19 to carboxylic acid 20, which
was subjected to amide forming conditions using propanephosphonic acid anhydride
ACS Paragon Plus Environment
18

Page 19 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(T3P), followed by Suzuki coupling to provide final products 5-7 and 10-11. For the
formation of final products containing the 5-(trifluoromethyl)thiophen-2-yl moiety (12-14),
ethyl ester 19 was subjected to Suzuki coupling conditions to first append the 5-
(trifluoromethyl)thiophen-2-yl group to give 24, and then ester hydrolysis and amide
formation gave final products 12-14. The synthesis of final products 15-17 was conducted
through Suzuki couplings of 22 and thiophene boronic esters.
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
Scheme 1. The synthetic route for compounds 5-7 and 10-17.a
CH3
N
Cl
N
N
NHMe
NO2
N
NHMe
NH2
a
b
d
N
c
O
N
H
Br
Br
NO2
Br
Br
Br
18
CH3
CH3
CH3
N
CH3
N
N
N
N
N
N
f
N
g
e
O
O
O
O
N
N
Ar
N
N
Br
O
Br
O
O
O
20
19
5-7, 10-11
NR2
OH
OEt
NR2
21: NR =
e
N,N-dimethylamin
N-azetidine
2
22: NR =
2
h
i
23: NR =
N-3-fluoroazetidine
2
CH3
CH3
CH3
N
N
O
N
N
N
N
N
O
N
j,
f
O
O
N
O
O
S
S
S
N
X
15-17
OEt
NR2
24
12-14
F3C
F3C
a Reagents and conditions: (a) 40% NH2CH3 in water, THF, 0 °C, then rt 16 h (99%); (b)
zinc, ammonium chloride, acetone/water (10:1), rt, 72 h; (c) CDI, DMF, rt to 60 °C, 16 h
(88% over 2 steps); (d) sodium hydride, DMF, rt, 30 min; then 2-bromoethylacetate, DMF,
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 88
1
2
3
4
5
6
7
8
9
rt, 4 h, 88%; (e) lithium hydroxide, THF/water, rt, 16 h (83%) (f) secondary amine, n-
propylphosphonic anhydride (T3P®, 50% in ethyl acetate), Hunig’s base, CH2Cl2, rt, 16 h
(61-85%) (g) Boronic Acid, cesium carbonate, PdCl2(dppf), dioxane, 100 °C, 16 h (9-46%)
(h) Bromothiophene, bis(pinacolato)diboron, potassium acetate, PdCl2(dppf), dioxane, 90
°C, 2 h; then add 22, cesium carbonate, 90 °C, 3h addnl. (10-48%); (i) 4,4,5,5-tetramethyl-
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
2-(5-(trifluoromethyl)thiophen-2-yl)-1,3,2-dioxaborolane,
potassium
carbonate,
PdCl2(dppf), toluene/water (10:1), 105 °C, 3 h (91%); (j) lithium hydroxide, THF/water, rt,
16 h (98%).
Compounds 12-14 were evaluated in PK experiments in rat followed by receptor
occupancy studies in the rodent brain (Table 3). Compound 11 was evaluated in rat PK
experiments as well but showed very high i.v. CL, suggestive of extra-hepatic clearance
mechanisms, as well as low bioavailability (1%). As such, it was not further assessed in
receptor occupancy studies. Compounds 12, 13 and 14 all had low to moderate CL (14-
30 mL/mg/kg) and moderate volumes of distribution (Vss) (0.9-1.9 L/kg). All three
compounds had excellent bioavailabilies, indicative of complete absorption. It is of note
that 13 was found to be insoluble at 0.5 mg/mL in three standard excipients, 20%
hydroxypropyl-β-cyclodextrin (HP-β-CD), 30% sulfobutylether-β-cyclodextrin (SBE-β-CD)
ACS Paragon Plus Environment
20

Page 21 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 100% polyethylene glycol 400 (PEG-400). The 20% HP-β-CD formulation did allow
for a 0.1 mg/mL solution, which was enough to support a 0.5 mg/kg dose study.
Next, GluN2B receptor occupancy and brain concentrations were measured for 12, 13
and 14 over a six-hour time course after oral administration in rat (Table 3). Receptor
occupancy was assessed by autoradiography (ARG) by application of the GluN2B-
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
selective
3H
radiotracer
3-[3H]
1-(azetidin-1-yl)-2-[6-(4-fluoro-3-methyl-
phenyl)pyrrolo[3,2-b]pyridin-1-yl]ethanone29 to brain slices ex vivo. Thirty min after dosing
at 3 mg/kg, all three compounds showed >70% occupancy as measured against
background control. At 6 h, 13 had >70% GluN2B occupancy, although brain
concentrations decreased from 1257 ng/mL at 2 h (Table 2) to 598 ng/g at 6 h (full data
shown in SI). Due to the difficulty in formulating 13, however, it was decided to deprioritize
13 and progress 12.
Table 3. Pharmacokinetic data (0.5 mg/kg i.v / 2.5 mg/kg p.o., n=3 ± SD, vehicle: 20%
HP-β-CD) and GluN2B occupancy (3 mg/kg p.o., n=2 ± SEM) data for 11-14 in rat.
% Occupancy
Timecourse (3 mg/kg
Cl
Vss
t1/2
Cmax
AUCinf
t/max %F
(h)
Cmax Tmax
(mL/mi (L/kg) (h)
(ng/mL) (hr.ng/
(h)
-brain
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 88
1
2
3
4
5
n/kg)
mL)
g)
p.o.)a (vehicle: 20%
HP-β-CD)
6
7
0.5 h
2.0 h
6.0 h
8
9
11
114 ±24 1.1
0.2
±0.0
6±3
--
--
1±1
132
--
--
--
--
--
±0.3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12b 30±0.9
1.9
±0.1
0.8
±0.0
479 ±172 1829
±535
1.3
±0.6 ±39
87
±2.3
76
765
0.5
31,
±1.6
±91
n=1
13c 14±6
1.2
±0.1
1.3
±0.3
188 ±68
1001
912
±50
1.0
±0.0 ±38
121
79
±2.0
80
±0.9
77
±2.2
1257 2.0
±128
20±18
0.9 ±1
0.9
3614
±240
1.0
±0.0 ±7
109
74
±3.9
71
±0.7
24
±1.0
728
±128
2.0
d
14
±0.6 ±235
a
Ex vivo GluN2B labelling was expressed as the percentage of GluN2B labelling in
corresponding brain areas of vehicle-treated animals.b Vehicle is 1:1 PEG400/water for
PK study. c Doses were 0.1 mg/kg iv and 0.5 mg/kg p.o. d Vehicle is 30% SBE-β-CD for
PK study.
The brain biodistribution of 12 was evaluated in rodents in a microdosing experiment,
in order to obtain a preliminary assessment of brain uptake of 12 in brain regions known
to be enriched with the GluN2B subunit. Although NMDA-subunit expression has been
shown to change in the mammalian brain from infancy to adulthood, the hippocampal
region has the highest receptor density of GluN2B protein and mRNA throughout the
lifetime. GluN2B is also highly expressed in the cortex, and moderately in the striatum.
GluN2B is not expressed in the cerebellum.
ACS Paragon Plus Environment
22

Page 23 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Rats (n = 3 each group) were administered a single i.v. dose of 0.03 mg/kg of 12
formulated in 50% (w/v) PEG/water at pH 7.4. Animals were sacrificed at 5, 10, 30, 60-
and 120-min following dosing. The cerebellum, cortex, striatum and hippocampus were
collected. The sample weight was measured and then homogenized to analyze levels of
12 by LC-MS/MS.
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
Compound 12 was detected in all four brain regions at each timepoint (Figure 4). In the
cerebellum, concentrations were highest at the 5 min timepoint (19.3 ± 1.08 ng/mL) and
decreased over time to 6.9 ± 0.65 ng/mL at 120 min, consistent with null expression of
GluN2B in rat cerebellum. In the cortex, concentrations of 12 were higher at 10 min
compared to 5 min and decreased at each of the subsequent timepoints. The highest
levels of 12 were measured at 10, 30 and 60 min in the hippocampus, where GluN2B is
highly enriched. Levels in the hippocampus reached a significance versus the cerebellum
at 30 min (22.7 ± 2.49 vs. 13.8 ± 1.62 ng/ mL in cerebellum, P<0.05) and 60 min (20.9 ±
3.47 vs.11.9 ± 1.27 ng/ mL in cerebellum, P<0.05) time points. At 120 min after dosing,
both hippocampus and striatum showed higher levels of 12, reaching significance
(10.1±0.13 and 9.8±0.20 ng/ mL respectively) vs. cerebellum (6.9±0.65 ng/ mL) P<0.05.
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 88
1
2
3
4
5
6
7
8
30
20
10
0
*
*
cerebellum
cortex
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
striatum
hippocampus
* *
5
10
30
60
120
time(min)
Figure 4. Biodistribution of 12 in rat brain regions at different time points after a single
i.v. injection of 30 µg/kg. All animals were sacrificed at different time points (5, 10, 30,
60 and 120 min after the injection). The result is normalized by tissue weight (g).
*P<0.05 significant increased vs. cerebellum. N=3.
The synthesis of 12 on gram scale was conducted beginning from 6-bromo-1,3-dihydro-
imidazo[4,5-b]pyridin-2-one (Scheme 2). It should be noted that this route was chosen in
part due to the immediate availability of ethyl 6-bromo-2-oxo-2,3-dihydro-imidazo[4,5-
b]pyridine-1-carboxylate in house. In order to achieve the desired protecting group
positioning on the core beginning with 6-bromo-1,3-dihydro-imidazo[4,5-b]pyridin-2-one,
ACS Paragon Plus Environment
24

Page 25 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a carbamate protecting group was selectively installed using ethyl pyridine-2-yl
carbonate, which was followed by trityl protection of the 3-position N-H. The ethyl
carbamate was then deprotected with isopropylamine and the resulting N-H was
subjected to SN2 reaction conditions with 1-(azetidin-1-yl)-2-chloroethan-1-one. Finally, a
Suzuki coupling followed by trityl deprotection and N-methylation provided 12.
Scheme 2. Multi-gram route for compound 12.a
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
3
Tr
4
H
N
Tr
N
H
N
N
g
3a
7a
N
N
b
5
N
N
c
2
a
O
O
O
O
N
Br
N
N
H
1
Br
6
N
Br
Br
7
OEt
H
OEt
O
O
H
Tr
N
N
N
e,f
d
N
O
12
O
N
N
O
Br
O
S
N
F3C
N
des-methyl-12
a Reagents and conditions: (a) ethyl pyridin-2-yl carbonate, potassium carbonate, DMF,
75 °C (93%); (b) trityl chloride, triethylamine, rt (94%); (c) isopropylamine, THF, rt (100%);
(d) sodium hydride, DMF, rt, 20 min; then 1-(azetidin-1-yl)-2-chloroethan-1-one, DMF, rt,
2 h (100%); (e) Boronic Acid, cesium carbonate, PdCl2(dppf), dioxane, 100 °C, 16 h; (f)
TFA, CH2Cl2, rt, 2h (64% over 2 steps); (g) sodium hydride, DMF, rt, 30 min; then
iodomethane, DMF, rt, 2 h (80%).
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 88
1
2
3
4
5
6
7
8
9
The dose dependency of 12 was obtained from GluN2B rat hippocampal occupancy
measurements after oral administration at 8 different doses (Table 3 and Figure 5).
Receptor occupancy was determined at 60 min post dose (3 animals per group), by ex
vivo ARG.29 The calculations were made vs naïve control as described for the 6 hour time
course (vide supra). The 60 min time point measurement was chosen based on the tmax
from the rat PK experiment. Compound 12 occupied the receptor in rat hippocampus with
a plasma EC50 of 407 ng/mL and a corresponding ED50 of 0.86 mg/kg (Figure 5). The
receptor occupancy we targeted in order to both produce desired pharmacology and have
the potential to minimize unwanted CNS effects15 was 70% for 1-2 h. The total plasma
concentration (bound + unbound) required to achieve 70% occupancy was 716 ng/mL
(95% confidence interval 558 to 919 ng/mL).
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
ACS Paragon Plus Environment
26

Page 27 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
100
90
80
70
60
50
40
30
20
10
0
EC50= 407 ng/mL
EC70= 716 ng/mL
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
10
100
1000
10000
100000
Plasma Concentration (ng/mL)
A
B
100
ED50= 0.86 mg/kg
ED70= 1.44 mg/kg
80
60
40
20
0
0.001
0.01
0.1
1
10
100
Dose (mg/kg)
Figure 5. Ex Vivo GluN2B receptor occupancy with 12 in rat hippocampus: dose
dependency after p.o. administration shown as(A) Plasma concentration vs % GluN2B
occupancy, and (B) Dose v.s. % GluN2B occupancy. Results are expressed as average
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 88
1
2
3
4
5
6
percentage receptor occupancy vs. naïve control rats ± SEM (n=3) and vehicle is 20%
HP-β-CD.
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
The in vitro profile of compound 12 is shown (Table 4). The unbound fraction in human
and rat plasma was 4.6% and 5.3%, respectively, giving an unbound plasma EC70 for 12
of 40 ng/mL in rat (based on total bound + unbound plasma EC70 = 716 ng/mL).
Compound 12 displayed no significant inhibition of cytochrome P450s (CYP) 3A4, 1A2,
2C19, 2C8 or 2C9 up to 50 μM, and no time-dependent inhibition of CYP3A4. It was found
to be a moderate inhibitor of CYP2D6 with an IC50 of 15 μM. It showed no induction of
CYP3A4 or CYP1A2. Compound 12 did not inhibit the hERG channel at any concentration
assessed. It did not have any appreciable effect in a radioligand binding assay screen
against a panel of 50 targets (GPCRs, ion channels, and transporters) at a concentration
of 10 μM.
Table 4. In vitro ADME and selectivity data for compound 12.
GluN2B rat a / human 16 ± 7 /
ACS Paragon Plus Environment
28

Page 29 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
Ki b (nM)
15 ± 6
6
7
8
9
GluN2- A / C / D
>10 µM
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
c
IC50
h / rat / dog / mky
CLint microsomesd
5.4 / 26.3 /
14.7 / 16.6
μL/min/mg
CYP isoform IC50
(for IC50 < 50 µM)e
2D6
(15 µM)
3A4 time-dependent
inhibition
no f
CYP induction
no g
<4
Solubility pH 4
ACS Paragon Plus Environment
29