Design and Synthesis of a New Series of 4-Heteroarylamino-1′-azaspiro[oxazole-5,3′-bicyclo[2.2.2]octanes as α7 Nicotinic Receptor Agonists. 1. Development of Pharmacophore and Early Structure-Activity Relationship

Article abstract of DOI:10.1021/acs.jmedchem.6b01506

The design and synthesis of a series of quinuclidine-containing spirooxazolidines ("spiroimidates") and their utility as α7 nicotinic acetylcholine receptor partial agonists are described. Selected members of the series demonstrated excellent selectivity

Full text of DOI:10.1021/acs.jmedchem.6b01506

Subscriber access provided by UNIV TORONTO
Article
Design and Synthesis of a New Series of 4-Heteroarylamino-1'-
azaspiro[oxazole-5, 3’-bicyclo[2.2.2]octanes as #7
Nicotinic Receptor Agonists. 1. Development of
Pharmacophore and Early Structure Activity Relationship.
James H. Cook, F. Christopher Zusi, Ivar M. McDonald, Dalton King, Matthew D Hill, Christiana
Iwuagwu, Robert A. Mate, Haiquan Fang, Rulin Zhao, Bei Wang, Jingfang Cutrone, Baoqing Ma, Qi
Gao, Ronald J. Knox, Michele Matchett, Lizbeth Gallagher, Meredith Ferrante, Debra Post-Munson,
Thaddeus Molski, Amy Easton, Regina Miller, Kelli Jones, Siva Digavalli, Francine Healy, Kimberley
A. Lentz, Yulia Benitex, Wendy J Clarke, Joanne Natale, Judith A. Siuciak, Nicholas J. Lodge,
Robert Zaczek, Rex Denton, Daniel Morgan, Linda J. Bristow, John E. Macor, and Richard E. Olson
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01506 • Publication Date (Web): 18 Nov 2016
Downloaded from http://pubs.acs.org on November 20, 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 47
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
4
5
6
7
8
9
Design and Synthesis of a New Series of 4-Heteroarylamino-1'-azaspiro[oxazole-5,
3’-bicyclo[2.2.2]octanes as α7 Nicotinic Receptor Agonists. 1. Development of
Pharmacophore and Early Structure Activity Relationship.
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
James Cook*, F. Christopher Zusi, Ivar M. McDonald, Dalton King, Matthew D. Hill,
Christiana Iwuagwu, Robert A. Mate, Haiquan Fang, Rulin Zhao, Bei Wang, Jingfang
Cutrone, Baoqing Ma, Qi Gao, Ronald J. Knox, Michele Matchett, Lizbeth Gallagher,
Meredith Ferrante, Debra Post-Munson, Thaddeus Molski, Amy Easton, Regina Miller,
Kelli Jones, Siva Digavalli, Francine Healy, Kimberley Lentz, Yulia Benitex, Wendy
Clarke, Joanne Natale, Judith A. Siuciak, Nicholas Lodge, Robert Zaczek, Rex Denton,
Daniel Morgan, Linda J. Bristow, John E. Macor, Richard E. Olson
Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT
06492, USA.
Abstract: The design and synthesis of a series of quinuclidine-containing
spirooxazolidines (“spiroimidates”) and their utility as α7 nicotinic acetylcholine receptor
partial agonists are described. Selected members of the series demonstrated excellent
selectivity for α7 over the highly homologous 5-HT3A receptor. Modification of the N-
spiroimidate heterocycle substituent led to (1S,2R,4S)-N-isoquinolin-3-yl)-4’H-4-
azaspiro[bicyclo[2.2.2]octane-2,5’oxazol]-2’-amine (BMS-902483), a potent α7 partial
agonist, which improved cognition in preclinical rodent models.
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
9
Schizophrenia is a chronic and highly debilitating mental disorder, affecting
approximately 1% of the general population, with an annual health care cost in the U.S.
estimated in the tens of billions of dollars. Patients with schizophrenia show an increase
of about 2.5-fold in all-cause mortality and have an elevated lifetime risk of suicide,
resulting in a life expectancy reduction of about 10 years.1 Both genetic and
environmental factors are thought to be involved in the etiology of the disease.2 The
clinical features of the disorder include disturbances in perceptual, emotional, cognitive,
and motor processes that are clustered into positive, negative, and cognitive symptom
domains. Current antipsychotic treatments can alleviate positive symptoms (e.g.,
delusions and hallucinations) however they lack consistent and robust efficacy against
negative (apathy, anhedonia, social withdrawal) or cognitive (deficits in attention,
working memory, executive function) symptoms. Additionally, patient compliance with
therapy is poor due to adverse extrapyramidal and metabolic side effects. This represents
a significant unmet medical need. Preclinical and early clinical data suggest that α7
nicotinic acetylcholine receptor (nAChR) agonists have potential for the treatment of
cognitive and negative symptoms in schizophrenia patients.3,4
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
Nicotinic acetylcholine receptors are pentameric acetylcholine gated ion channels that are
expressed in brain regions including the hippocampus, cortex, and subcortical limbic
regions. There are genes encoding for 16 individual nicotinic subunits, allowing for a
large number of potential pentameric configurations, but the most abundant in the brain
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
6
7
8
9
are the α4β2 and α7 nAChRs.5 Given the localization of α7 nAChRs in brain regions
associated with cognition and genetic association studies linking α7 nAChRs to sensory
processing deficits,6 preclinical efforts have focused on the testing of α7 nAChR agonists
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 models of learning, memory, and sensory gating. Thus several α7 nAChR agonists
have been shown to improve learning and memory in normal and aged animals, reverse
scopolamine-induced memory deficits, reverse pharmacologically-induced gating
deficits, and possess some anxiolytic properties.7 The effects of α7 nAChR agonists on
executive cognitive function have led to the hypothesis that they could have therapeutic
benefit in the treatment of cognitive deficits in schizophrenia patients. Positive phase II
clinical results in schizophrenia with encenicline (EVP-6124, 1) were reported,3 though
this compound failed to meet its co-primary endpoints in phase III trials (Figure 1).3d In
mild-to-moderate Alzheimer’s disease another α7 nAChR agonist, nelonicline (ABT-
126, 2), showed trends toward improvements in cognition in phase II trials but also failed
to meet its primary endpoints (Figure 1).4 This area has been recently reviewed.8, 9
The investigation of ligands for the α7 receptor has focused on structures which share
important pharmacological elements with natural ligand acetylcholine. Many of these
structures contain three pharmacophoric elements: a rigid basic amine, a central polar
group incorporating a hydrogen bond acceptor, and a lipophilic aryl or heteroaryl group
extending from the amine.10 An azabicyclic group, frequently quinuclidine, has been a
common motif for the constrained amine because of its consistently protonated form at
physiological pH, mimicking the quaternary ammonium feature in acetylcholine. The
amine fragment and the lipophilic aryl group have been demonstrated by Mullen et al.10
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to influence the affinity and the selectivity of ligands for the α7 receptor. A variety of
functional groups (carbamates,11-16 azaheterocyclic ethers,17-20 and amides21-28) have been
employed as the central hydrogen bond acceptor (Figure 1). The lipophilic pocket in the
α7 receptor seems to accommodate significant structural changes.
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
Nicotinic α7 and serotonergic 5-HT3A receptors have high sequence homology.29 Many
of the published α7 nicotinic full or partial agonists demonstrate antagonist activity at the
5-HT3A receptor with varying selectivity.21,30,31 Because of the potential for adverse side
effects related to gastrointestinal motility (constipation) associated with inhibition of the
5-HT3A receptor,32,33 we included selectivity against this activity as part of our criteria for
compound progression.
Investigation of the nicotinic agonist literature helped identify several SAR trends that
could lead to a selective α7 nAChR agonist. Spirooxazolidinone-quinuclidine α7
nAChR agonists utilizing a carbamate as the hydrogen-bond acceptor have been
described. These feature both a heteroatom substituent on the carbon beta to the
quinuclidine nitrogen, which may modulate quinuclidine basicity,26 and a rigid
framework, which is believed to project a group into a hydrophobic pocket.9,13-15
Compound 3 (AR-R17779) exhibits submicromolar binding potency for α7 nAChR (Ki =
92 nM) with very good selectivity versus the 5-HT3A receptor (inhibition = 2.5% at 10
ꢀM) and acts as a highly efficacious agonist displaying 96% maximal efficacy at rat α7
receptors.10 Additionally, spirooxazolidinone 4 was shown to be a partial agonist at the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
4
α7 receptor (Ki = 3 nM) with similar potency as an antagonist at the 5-HT3A receptor (Ki
5
6
= 10 nM).15
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 1: Examples of Known α7 nAChR Agonists.
Upon ligand (agonist) binding to the receptor’s extracellular site, an ion current is
produced34,35 and the receptor quickly reverts to a closed, desensitized state in which the
response to further agonist is blunted.36,37 Receptor desensitization has the potential to
limit efficacy in vivo from the perspective of maintained efficacy after chronic dosing but
also from the perspective of the efficacy/dose relationship after acute administration. 30,38
The latter can result in steep U shaped dose/effect relationships leading to challenges in
ACS Paragon Plus Environment

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
identification of active doses in preclinical studies and more importantly clinical trials.
We chose to focus on the identification of partial agonists since agents with this profile
retain robust efficacy in preclinical models over a more extensive dose/exposure range
than agents with high levels of intrinsic efficacy. Additionally, recent literature suggests
that a7 nAChR signalling may in part involve G-protein interactions which are
independent of channel activation.39 In summary, our goal was to identify novel selective
partial α7 nAChR agonists with favorable pharmacokinetic and off-target selectivity
properties. Our focus was on spiroazabicyclic structures, as these have shown potential
for both potency and 5-HT3A receptor selectivity.
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 prototypical spiroazabicyclic α7 nAChR agonist 3, with zero rotatable bonds, can be
considered a fully conformationally restricted analog of acetylcholine. It has been shown
that N-arylation of this core, as in spirooxazolidinone 4, can lead to a significant
improvement in potency. We hypothesized that similar boosts in potency could arise
from aryl substitution at the exocyclic heteroatom of this system as in spirocyclic imidate
11, which has the potential to exist in two tautomeric forms (Figure 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
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 2. Conception of Novel Amino-Substituted Spirooxazolo-Quinuclidine.
SYNTHETIC CHEMISTRY
Amino-substituted spirooxazolo-quinuclidines (“spiroimidates”) 11 (Scheme 1) were
initially synthesized as racemates by combining the known amino alcohol 1240, 41 (borane
adduct) with an isothiocyanate to obtain the thiourea intermediate 13. Treatment with
diisopropylcarbodiimide (DIC) followed by deprotection of the quinuclidine borane
complex with HCl afforded the spiroimidates 11.
Scheme 1a
aReagents: (a) RNCS, DMF; (b) DIC, DMF, 80 °C; (c) 1 N HCl, acetone.
ACS Paragon Plus Environment

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Isothiocyanates 15 were prepared from aromatic or heteroaromatic amines 14 via
treatment with either thiocarbonyldiimidazole or thiocarbonyldi-2(1H)pyridone (Scheme
2). In some cases, isothiocyanates did not form directly; however, the intermediate
mixed imidazole thioureas (16) were employed with equivalent results. Certain weakly
basic heteroaryl amines did not react efficiently under these conditions and were instead
treated with excess strong base, carbon disulfide (CS2), and methyl iodide (MeI) to give
imidodithioates 17, which could be condensed directly with 12 to furnish products 11.
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 2a
aReagents: (a) 1,1’-thiocarbonyldiimidazole or 1,1’thiocarbonyldipyridin-2(1H)-
one, DMF or MeCN; (b) 10 N NaOH, CS2, 10 N NaOH, 2 equiv. MeI.
Alternatively, some unsubstituted pyridine derivatives 20 (derived from highly basic
aminopyridines) were not obtained from the hydrolytic conditions employed for the
quinuclidine-borane deprotection and were accessed via their bromo-derivatives 18, with
the bromo substituent removed by catalytic hydrogenation (Scheme 3, see compound 33).
Scheme 3a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
6
7
8
Br
Br
HN
N
N
HN
N
HN
N
N
N
O
O
a
O
b
9
N+
N
N
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
-
BH3
18
19
20
aReagents: (a) 1 N HCl, acetone; (b) H2, Pd/C
Homochiral spiroimidates were obtained either by chiral chromatographic separation of
the racemic final products or by synthesis from chiral aminomethyl alcohol 22. Alcohol
22 was generated by treating borane-protected quinuclidine 12 with benzyl chloroformate
(CBZ-Cl) to give the corresponding racemic protected amino alcohol 21 (Scheme 4).
Racemic 21 was then separated into its individual enantiomers by chiral chromatography.
The CBZ- and borane-protecting groups were removed in a one-pot procedure by
treatment with aqueous HCl followed by hydrogenolysis in the presence of catalytic
palladium to afford amino alcohol 22 as the dihydrochloride salt. Final conversion to the
enantiomerically pure spiroimidates 24 was achieved by condensing amino alcohol 22
with the appropriate isothiocyanate followed by ring closure with
diisopropylcarbodiimide.
Scheme 4a
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
HN R
N
OH
O
OH
OH
H
N
a, b
c, d
e, f
NH2
NH2
Cbz
N+
N+
N
N
2 HCl
-
-
BH3
BH3
9
11
21
22
24
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
aReagents: (a) Cbz-Cl, Na2CO3, CH2Cl2, H2O; (b) chiral HPLC separation; (c) 1:1
acetone: 3 N HCl; (d) 10% Pd/C, H2, MeOH; (e) Cs2CO3, RNCS, DMF; (f) DIC, DMF,
80 °C
The quinuclidine analogs 25-41 and 53-58, and the [2.2.1] spiroimidates 45 and 47 (Table
4) were synthesized according to Scheme 1 starting from the appropriate amino alcohols,
which were prepared according to Swain et. al.40-42 The [3.2.1] analogs 47 and 48 were
prepared from the corresponding ketone43 by a route similar to that employed to prepare
the [2.2.1] analog (see Supplemental Information). The isomers in the [2.2.1] and [3.2.1]
heterocyclic systems were selectively prepared via temperature control of the
cyanohydrin formation. The “reversed spiroimidate” 42, N-substituted analog 43, and the
6-membered ring spiroimidate 44, (Table 3) were prepared utilizing the method of
Scheme 1 from the appropriate amino alcohols or diamines using the methods of Mullen
et. al.10
BIOLOGY
Our screening tier involved an in vitro primary assay using a stable HEK293 cell line co-
expressing the rat α7 nACh receptor with human RIC-3, an accessory protein required
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
4
5
6
7
8
9
for the proper folding and/or assembly of the α7 nACh receptors at the cell surface.44
Successful generation of this cell line enabled the development of a high throughput
functional FLIPR assay that was able to detect agonist activity without the need to co-
apply α7 nACh receptor positive allosteric modulators to produce a robust and
measureable Ca2+ flux response. We also developed functional, FLIPR assays to counter
screen agents at other nicotinic receptor subtypes namely rat α1β1δε, rat α3β4 and rat
α4β2 (all three in agonist mode) and at the human 5HT3A receptor (antagonist mode).
Potency and intrinsic efficacy were determined for selected compounds using a whole
cell, patch clamp electrophysiology (EP) assay in HEK-293 cells expressing the rat α7
nAChR (+ RIC3). We recorded peak intensity and the area under the current curve with
the current area EC50 serving as a measure of potency and the peak and current area Ymax
(%) as measures of intrinsic activity relative to acetylcholine. Competition binding
studies were conducted to determine the in vitro binding potency of selected compounds
by displacing [125I]-BTX binding from recombinant rat α7 receptors.
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
RESULTS AND DISCUSSION
Quinuclidine N’-phenyl spiroimidates were explored first. The phenyl spiroimidate 25
displayed weak α7 nAChR potency with selectivity favoring the 5-HT3A receptor (Table
1). We therefore explored a number of heterocyclic ring systems including thiazoles,
benzothiazoles, and pyridines. The α7 nAChR potency was improved 27-fold by
exchanging the phenyl (25) for a thiazole ring (26). In a patch clamp electrophysiology
(EP) assay utilizing HEK-293 cells expressing rat α7 nAChR, compound 26 was found to
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
be a partial agonist (peak Ymax = 47%; area Ymax = 63%) with an EC50 = 1.3 ꢀM.
However, the selectivity still favored the 5-HT3A receptor. Further investigation into
substituted thiazole analogs showed the C5 phenyl analog 27 had slightly diminished α7
nAChR potency but was 750-fold selective for the α7 nAChR versus the 5-HT3A
receptor. The opposite effect was seen with the 4-phenyl isomer 28, which demonstrated
that the substituent position on the heterocyclic ring had a profound influence on α7
receptor potency and 5-HT3A receptor selectivity.
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
Table 1: Functional Potency at α7 nAChR and 5-HT3A Receptors for Initial
Spiroimidate Compounds.a
HN R
O
N
N
R
5-HT3A
(IC50, nM
+SEM (n)b
FLIPR
FLIPR α7
EC50, nM
+SEM (n)b
Selectivity Ratio
vs 5-HT3A
5400
+1700 (3)
200
230
+4 (3)
25
0.04
0.1
25
26
+60 (6)
+6 (6)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
4
5
6
400
20000
51
27
7
+420 (6)
+4900 (6)
S
8
9
N
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
450
+100 (4)
20
11
+3 (4)
36
S
0.02
1.8
13
28
29
30
31
32
33
N
S
+12 (7)
3100
+12 (7)
41000
+32000 (2)
16
N
S
N
+1400 (2)
180
0.1
0.1
5
N
+50 (7)
21000
+3 (7)
2300
N
N
+13000 (7)
11000
+1000 (7)
58000
+10000 (4)
+6000 (4)
aAll compounds are racemic. bNumber of determinations in parentheses.
A substantial increase in α7 nAChR potency was observed with the 2-
aminobenzothiazole (29) and 2-aminopyridine (31) heterocycles. However, 3- and 4-
substituted aminopyridines 32 and 33, along with reverse benzothiazole 30, had reduced
α7 nAChR potency similar to the phenyl analog 25. One possible explanation for these
results was that the improved potency was associated with conformational restriction
from an intramolecular hydrogen bond involving the spiroimidate and heterocyclic
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nitrogen atoms. This would favor a conformation where the exocyclic nitrogen can act as
a hydrogen-bond acceptor and resulted in key binding interaction with the α7 receptor
(Figure 3). Interestingly, such an interaction was observed in an X-ray structure of a
related analog (see below).
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. Hypothesized Intramolecular Hydrogen Bond Interaction.
Compound 29 maintained partial agonist activity in a patch-clamp electophysiology assay
with peak Ymax = 5%; area Ymax = 34% and an EC50 = 0.5 ꢀM. Even though α7 nAChR
potency for benzothiazole 29 was improved 270-fold over phenyl analog 25, selectivity
was still suboptimal. As we demonstrated with phenyl thiazoles 27 and 28, the aryl
substitution could significantly alter α7 nAChR/5-HT3A selectivity. We examined
whether selectivity could be improved with 2-aminobenzothiazoles by substitution on the
benzothiazole ring. Introduction of a methoxy group at the C4 or C5 positions, (Table 2,
entries 34 and 35, respectively), resulted in a loss of α7 nAChR potency while potent 5-
HT3A receptor activity was maintained. However, we were pleased to discover that
substitution at C6 (compound 36) resulted in good selectivity wherein α7 nAChR
potency decreased slightly while 5-HT3A receptor potency was reduced 500 fold
compared to the unsubstituted benzothiazole 29 . Compound 36 also demonstrated
partial agonism with peak Ymax = 29%; area Ymax = 51% and an EC50 = 0.6 ꢀM. The C6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
7
8
9
substituted chloro compound 39 and the C6 methyl compound 40 also retained α7
nAChR potency and good selectivity vs 5-HT3A, suggesting that this selectivity is driven
by steric rather than electronic factors. Compounds with other substituents at C6 (38, 39,
and 41) also showed enhanced selectivity but had weaker α7 nAChR potency. Not
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
surprisingly, disubstitution at C5 and C6 (37) resulted in a substantial loss of α7 nAChR
potency.
Table 2: Benzothiazole Spiroimidate Substituent Effects.a
7
S
N
6
R
5
HN
N
O
4
N
FLIPR α7
EC50, nM
+SEM (n)b
Structure
5-HT3A
IC50, nM
+SEM (n)b
FLIPR
Selectivity Ratio
vs 5-HT3A
830
+330 (4)
1100
13
+3 (4)
19
4-OMe
5-OMe
6-OMe
5,6-OMe
0.02
0.02
153
3
34
35
36
37
+800 (4)
120
+4 (4)
19000
+4000 (6)
27000
+90 (6)
11000
+2900 (4) +11000 (4)
ACS Paragon Plus Environment

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
320
14000
+6900 (9)
8100
6-OEt
6-Cl
42
53
32
11
38
39
40
41
+390, (9)
150
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
+90 (4)
120
+3900 (4)
3800
6-Me
6-OH
+30 (6)
980
+140 (6)
11000
+530 (4)
+400 (4)
aAll compounds are racemic. bNumber of determinations in parentheses.
In parallel to our efforts to investigate the heterocyclic SAR, we examined several
variants of the quinuclidine and spiroimidate (Table 3). Reversing the oxazolidine
heteroatoms as in analog 42 or N-alkylation of the spiroimidate nitrogen (43) resulted in
significantly diminished α7 nAChR activity. Interestingly, insertion of a methylene into
the oxazolidine ring (44) diminished the potency only slightly. These results are
consistent with the internal hydrogen-bond hypothesis, as neither 42 nor 43 can achieve
this interaction. On the other hand, 44 can achieve a cyclic hydrogen-bond arrangement,
and the altered geometry resulted in significantly reduced α7 nAChR and 5-HT3A
activity.
Table 3. Spiroimidate Variants of Benzothiazole 36.a
5-HT3A
IC50, nM
+SEM (n)b
FLIPR
Selectivity
Ratio vs 5-
FLIPR α7
EC50, nM
+SEM (n)b
Structure
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
HT3A
5
6
7
8
120
19000
153
36
42
43
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
+90 (6)
+4000 (6)
OMe
OMe
S
N
7800
47000
HN
O
6
N
+3900 (5)
+18000 (5)
N
N
S
N
N
N
>100000
>100000
N/A
O
Me
OMe
S
3700
64000
HN
17
44
N
+3300 (7)
+29000 (7)
O
N
N
aAll compounds are racemic. bNumber of determinations in parentheses.
We next investigated replacements for the quinuclidine [2.2.2] bicycle (Table 4).
Interestingly, the [2.2.1] (compounds 45 and 46) and [3.2.1] (compounds 47 and 48)
bicyclics showed good potency at the α7 receptor, with one diastereomer in both cases
having better selectivity versus the 5-HT3A receptor. Electrophysiology (EP) results
showed 45 to be a partial agonist with peak Ymax = 76%; area Ymax = 24%) and an EC50 =
0.28 ꢀM.
Table 4: Quinuclidine Variations.
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
N
HN
S
O
N
R
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
FLIPR
Selectivity
Ratio vs 5-
HT3A
FLIPR α7
EC50, nM
+SEM (n)a
5-HT3A
IC50, nM
+SEM (n)a
R
Stereochemistry
Racemic
21
+22 (7)
22
21
+5 (7)
480
1
29
45
46
47
48
N
H
Homochiral
Racemic
22
N
+26 (4)
260
+280 (4)
35
0.06
3
N
+180 (4)
290
+5 (4)
950
H
Racemic
N
+110 (4)
700
+290 (4)
7
Racemic
< 0.01
N
+360 (5)
+0 (5)
aNumber of determinations in parentheses.
Benzothiazole 29 and 6-methoxy benzothiazole 36 were separated using chiral
chromatography into enantiomers 49/50, and 51/52, respectively. An X-ray of 49
determined that the stereogenic quaternary carbon has the stereochemistry as shown in
Figure 4.45
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
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
S
N
S
N
HN
N
N
N
O
O
H
N
N
Figure 4. X-ray crystal structure of compound 49
In each case, most of the α7 nAChR potency resided in one of the enantiomers (49 and
51, respectively) as shown in Table 5.
Table 5: Enantiomers of the Spiroimidates 29 and 36.
R
S
HN
N
O
N
N
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5-HT3A
IC50, nM
FLIPR α7
EC50, nM
FLIPR
Selectivity
Ratio vs 5-HT3A
R
H
Enantiomer
+SEM (n)a
+SEM (n)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
21
21
+5 (7)
7300
1
6
49
+22 (7)
1300
50
51
52
+400 (9)
140
+1500 (9)
9100
66
85
+70 (8)
1900
+1800 (8)
>97000
(5)
OMe
+200 (5)
aNumber of determinations in parentheses.
Benzothiazole 51 was further profiled in vitro and in vivo (data collected in Tables 6a and
6b). In a patch clamp electrophysiology (EP) assay, compound 51 was found to be a
partial agonist (peak efficacy = 23%; area efficacy = 47%) with an EC50 = 0.4 ꢀM. As
noted above, its selectivity against the 5-HT3A receptor was acceptable (66-fold), and
selectivity versus other nicotinic receptors was greater than 700-fold. The microsomal
stability was relatively good (> 60% parent remaining in human, mouse, and rat
microsomes after 10 min.). Compound 51 displayed high clearance (320 mL/min/Kg in
rat), moderate oral bioavailability (24%), and good brain accumulation (5300 nM after a
10 mg/kg dose s.c.; brain/plasma ratio = 7). The pharmacokinetic properties of 51 were
deemed sufficient to proceed to in vivo testing in rodent behavioral models.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
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
Table 6a: In Vitro Data for Spiroimidate 51.
Assay
51
FLIPR rat α7 (EC50, nM)
140 + 70 (8)
35, 38 (2)
400
Rat α7 binding ([125I]Tyr54-α-Bungarotoxin, Ki, nM)
Area EC50 (nM)
rat α7 nAChR
23, 47
>100
Electrophysiology
Peak Ymax, Area Ymax (%)
Nicotinic receptors:
Rat α1β1δε, α3β4, α4β2 (EC50, ꢀM)
Table 6b: ADME, Pharmacokinetic Data for Spiroimidate 51.
Assay
51
320
24%
760
5300
7
Rat Cl (mL/min/kg)
Rat F% (PO 10 mg/kg)
Rat plasma level 10 mg/kg s.c. dose (nM)
Rat brain level 10 mg/kg s.c. dose (nM)
Rat brain/plasma ratio
The activity of compound 51 was assessed in the mouse novel object recognition (NOR)
model which has been extensively used to examine visual recognition memory after
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
treatment with α7 nACh receptor agonists (see Supplementary section for details). The
model relies on the innate tendency of mice to explore novelty and is conducted over 2
sessions; in the first session (training) the subject is allowed to explore 2 identical objects
and in the second session (testing), conducted 24 h later, the subject is placed in the same
test chamber which now contains 1 previously encountered object and 1 novel object.
The time spent exploring each object is used as a measure of memory retention with
improved retention reflected by an increase in novel object exploration time. Treatment
with spiroimidate 51, 30 min prior to training, significantly increased novel object
exploration at doses of 3 mg/kg and 10 mg/kg. These data supported the hypothesis that
α7 partial agonists can improve cognitive performance.
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
40
Novel object
*
*
Familiar object
30
20
10
0
Veh
3
10
Compound 51 (mg/kg s.c.)
Figure 5. Compound 51 improves object recognition memory in mice. Mice were
treated with compound 51 (3 or 10 mg/kg, s.c.) 30 min prior to training and tested 24 h
later. Results are presented as the mean SEM (n=12-14/group) time spent exploring the
novel or familiar object and were analyzed by paired t test at each dose; *p<0.05 vs
familiar object.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
9
Even though 51 was efficacious in the NOR model, we anticipated an insufficient safety
margin versus hERG activity (IC50 = 1.3 ꢀM) at active doses based on the rat
pharmacokinetic data. We therefore focused on the pyridine series 31, which was
equipotent at α7 nAChR to the original thiazole 26. Although 31 did not show the
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
potency at α7 nAChR or selectivity against the 5-HT3A receptor that we desired, we
sought to apply the SAR learnings from the thiazole series to improve both of these
parameters.
Table 7: Functional Potency at α7 nAChR and 5-HT3A Receptors for 2-
Aminopyridines and Fused Heterocyclesa
5-HT3A
IC50, nM
+SEM (n)b
FLIPR
Selectivity Ratio
vs 5-HT3A
FLIPR α7
EC50, nM
+SEM (n)b
R
110
15
0.1
9.6
53
54
N
+40 (4)
+2 (4)
8.9
85
+6.7 (4)
+26 (4)
750
22000
30
55
56
+350 (4)
+4000 (4)
26
8
0.3
+15 (10)
+4 (11)
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
9.3
480
52
57
58
5
6
7
8
+5.3 (29)
210
+160 (29)
1600
9
7.5
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
+150 (4)
+700 (4)
aAll compounds are enantiopure. bNumber of determinations in parentheses.
Introduction of a phenyl substituent to the pyridine ring (54 and 55), as observed for
thiazole 27, improved the 5-HT3A receptor selectivity by 9 and 30 fold, respectively
(Table 7). Surprisingly, 54 showed excellent α7 nAChR potency (EC50 = 8.9 nM), while
55 was much weaker than 53. Structure activity relationship trends in the thiazole series
pointed to the more active benzothiazole (29). Accordingly, we hypothesized a potential
boost in potency with an isoquinoline or quinoline ring system. Heterocycles 56 and 57
(BMS-902483) improved α7 nAChR potency to 26 nM and 9.3 nM, respectively, with
analog 57 also providing a significant selectivity advantage against the 5-HT3A receptor.46
Compound 58 resulted in a loss of α7 nAChR potency. The SAR trends observed with
heterocycles 56-58 further supported the observation made in the benzothiazole series
that steric factors drive α7 nAChR potency and 5-HT3A receptor independently.
Isoquinoline 57 projects into a “sweet spot” where α7 nAChR potency is boosted while
5-HT3A receptor potency is diminished (Figure 6).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
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. Steric model describing α7 nAChR and 5-HT3A receptor SAR for
quinuclidine spiroimidates.
Based on the improvement in potency and selectivity for 57 compared to 51, isoquinoline
57 was further profiled. Compound 57 was examined in whole cell patch clamp
electrophysiology and shown to exhibit a partial agonist profile at the rat α7 nAChR; the
peak Ymax and area Ymax responses were 36% and 60% respectively and the EC50 was
0.14 µM. In a separate experiment, the cells were treated with the α7 nAChR antagonist
methyllycaconitine (MLA, 10 nM) prior to treatment with compound 57 (10 µM) in the
continued presence of 10 nM MLA, which caused a nearly complete inhibition of the
signal and indicated that the currents were produced by interaction with α7 nACh
receptors (Figure 7).
B Current Area
A Peak Current
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-2
-1
0
1
2
-2
-1
0
1
2
Log[Compound] (ꢀM)
Log[Compound] ( ꢀM)
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
C
D
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
3mM ACh
10ꢀM compound 57
10ꢀM compound 57
& 10nM MLA
1ꢀM compound 57
3ꢀM compound 57
10ꢀM compound 57
Figure 7. Compound 57 exhibits a partial agonist profile in whole cell patch clamp
assays conducted on HEK293/rat α7 nAChR cells. Concentration response curves for
A) peak current amplitude normalized to the peak current response induced by 3 mM
acetylcholine and B) area under the current response curve normalized to the current area
curve for 300 µM acetylcholine. All results are presented as the mean + SEM (n=9).
C) Representative current traces for compound 57. D) Inhibition of current produced by
compound 57 upon treatment with the α7 nAChR antagonist methyllycaconitine (MLA);
black trace indicates 10 µM compound 57, red trace indicates application of 10 nM MLA
followed by10 µM compound 57 in the continued presence of 10 nM MLA.
Isoquinoline 57 had good stability in human and mouse microsomes and showed low
potential for significant drug-drug interactions based on CYP inhibition data (Table 8a).
Plasma free fraction levels for many quinuclidine spiroimidates in our program were
generally high with 57 following that trend. The potential of 57 to undergo P-gp
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
4
5
6
7
8
9
mediated efflux, as evaluated in a bidirectional Caco-2 assay, was found to be low (efflux
ratio = 1.1). Additionally, the hERG potency was modestly improved (IC50 = 3.22 ꢀM)
Table 8a: Selected in vitro Screening Data of 57
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
FLIPR rat α7 (EC50, nM)
57
9.3 5.3
4.4 1.1
140
Rat α7 binding ([125I]Tyr54-α-Bungarotoxin, Ki, nM)
Area EC50 (nM)
Rat α7
Peak Ymax, Area Ymax (%)
36, 60
Electrophysiology
Nicotinic receptors (Agonism):
>100
Rat α1β1δε, α3β4, α4β2 (EC50, ꢀM)
Human 5-HT3A Antagonism (IC50, nM)
Metabolic stability, % remaining (human, rat, mouse)
CYP inhibition, IC50 (ꢀM) a
480 160
96, 1, 89
>40.0
Patch Clamp hERG (IC50, ꢀM)
Plasma free fraction, % free (human, rat, mouse)
Caco-2, efflux ratio
3.22
24.7, 26.9, 23.0
1.1
aPanel of human CYP isozymes: 3A4-BFC, 3A4-BZR, 1A2, 2B6, 2C8, 2C9, 2C19, 2D6
Isoquinoline 57 was evaluated for in vivo activity in the mouse NOR model (Table 8b
and Figure 8). Treatment with compound 57, 30 min prior to training, significantly
improved 24 h object recognition memory at doses of 0.1, 0.3, and 1 mg/kg, s.c. (Figure
8A). The associated plasma and brain exposures for 57, determined 30 min after dosing
in separate groups of mice, were 16 nM and 57 nM, respectively, at the minimum
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
efficacious dose (MED) of 0.1 mg/kg. In a separate study we also examined the effect of
0.3 mg/kg (s.c.) 57 administered 30 min, 2 h or 6 h prior to the training session. In all
cases a significant improvement in memory retention was observed (Fig 8B).
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
Familiar object
Novel object
A
B
Familiar object
Novel object
30
20
10
0
30
*
**
*
**
**
20
10
0
**
Veh
0.03
0.1
0.3
1.0
Veh
30 min
0.3 mg/kg s.c. Compound 57
Pretreatment Time Prior to Training
2 h
6 h
Compound 57 (mg/kg s.c.)
Figure 8. Compound 57 improves object recognition memory in mice. A) NOR dose
response for compound 57 administered 30 min prior to training (n=13-15/group) and B)
NOR time course for 0.3 mg/kg compound 57 administered 30 min, 2 h or 6 h prior to
training (n=10-12/group). Results are presented as the mean SEM object exploration
time determined 24 h after training and were analyzed for each treatment group by 2
tailed paired t test (*p<0.05, **p<0.01 versus familiar object).
Table 8b. Drug Concentrations of 57 30 min after dosing in mice
Mean SD
Time
(h)
Dose
Drug Concentration (nM)
NOR Efficacy
(mg/kg s.c.)
Brain
Plasma
4 0.7
0.5
0.03
19 6
no
ACS Paragon Plus Environment