Discovery and Lead Optimization of Atropisomer D1 Agonists with Reduced Desensitization

Article abstract of DOI:10.1021/acs.jmedchem.8b01622

The discovery of D1 subtype-selective agonists with drug-like properties has been an enduring challenge for the greater part of 40 years. All known D1-selective agonists are catecholamines that bring about receptor desensitization and undergo rapid metabolism, thus limiting their utility as a therapeutic for chronic illness such as schizophrenia and Parkinson's disease. Our high-throughput screening efforts on D1 yielded a single non-catecholamine hit PF-4211 (6) that was developed into a series of potent D1 receptor agonist leads with high oral bioavailability and CNS penetration. An important structural feature of this series is the locked biaryl ring system resulting in atropisomerism. Disclosed herein is a summary of our hit-to-lead efforts on this series of D1 activators culminating in the discovery of atropisomer 31 (PF-06256142), a potent and selective orthosteric agonist of the D1 receptor that has reduced receptor desensitization relative to dopamine and other catechol-containing agonists.

Full text of DOI:10.1021/acs.jmedchem.8b01622

Subscriber access provided by Kaohsiung Medical University
Article
Discovery and Lead Optimization of Atropisomer
D1 Agonists with Reduced Desensitization
Jennifer E. Davoren, Deane M Nason, Jotham Coe, Keith Dlugolenski, Christopher J. Helal, Anthony
R. Harris, Erik LaChapelle, Sidney Liang, Yue Liu, Rebecca E O'Connor, Christine C Orozco,
Brajesh Kumar Rai, Michelle Salafia, Brian M. Samas, Wenjian Xu, Rouba Kozak, and David Gray
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01622 • Publication Date (Web): 15 Nov 2018
Downloaded from http://pubs.acs.org on November 16, 2018
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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery and Lead Optimization of Atropisomer D1
Agonists with Reduced Desensitization
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
Jennifer E. Davoren,*† Deane Nason,‡ Jotham Coe,‡ Keith Dlugolenski,§ Christopher Helal,‡
Anthony R. Harris,‡ Erik LaChapelle,‡ Sidney Liang,‡ Yue Liu,§ Rebecca O’Connor,‡ Christine
C. Orozco,‡ Brajesh Rai,† Michelle Salafia,‡ Brian Samas,‡ Wenjian Xu,‡ Rouba Kozak,§ and
David Gray§
†Medicine Design & §Internal Medicine Research Unit, Pfizer Worldwide Research and
Development, 610 Main Street, Cambridge, Massachusetts 02139, United States; ‡Medicine
Design, Pfizer Worldwide Research and Development, Eastern Point Road, Groton, Connecticut
06340, United States
ADDRESS: Pfizer, 610 Main Street, Cambridge, Massachusetts 02139
KEYWORDS: Dopamine receptor; GPCR, partial agonist; biased signaling, cAMP, -arrestin,
desensitization, Parkinson’s disease, schizophrenia, atropisomer, D1; D1R; hit to lead, high-
throughput screen, ligand-based drug optimization.
Abstract
The discovery of D1 subtype-selective agonists with drug-like properties has been an enduring
challenge for the greater part of 40 years. All known D1-selective agonists are catecholamines
which bring about receptor desensitization and undergo rapid metabolism thus limiting their
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 45
1
2
3
4
5
6
7
8
9
utility as a therapeutic for chronic illness such as Schizophrenia and Parkinson’s disease. Our
high-throughput screening efforts on D1 yielded a single non-catecholamine hit, PF-4211 (6) that
was developed into a series of potent D1 receptor agonist leads with high oral bioavailability and
CNS penetration. An important structural feature of this series is the locked biaryl ring system
resulting in atropisomerism. Disclosed herein is a summary of our hit-to-lead efforts on this
series of D1 activators culminating in the discovery of atropisomer 31 (PF-06256142), a potent
and selective orthosteric agonist of the D1 receptor that has reduced receptor desensitization
relative to dopamine and other catechol-containing 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
Introduction:
The dopamine (DA) receptors are a family of five G protein-coupled receptors (GPCRs)
expressed throughout the nervous system.1 Based on their pharmacological profiles, localization,
and mechanism of action, they are roughly characterized as either D1-like (D1 and D5) or D2-
like (D2, D3 and D4).2 Activated D1-like receptors signal via the Gs subunit which increases
intracellular cAMP,3 while D2 receptors signal through Gi, which decreases cAMP levels in the
cell. D1 receptors play a central role in synaptic plasticity, basal ganglia-mediated motor
function, and specific domains of cognitive function including spatial learning and memory,
reversal learning, extinction learning, and incentive learning.4 D1 receptor signaling is deficient
in a variety of psychiatric, neurological, and endocrine disorders such as Parkinson’s disease
(PD)5 and schizophrenia.6
Development of a D1-selective (vs. D2) activator with properties suitable for oral
administration has been an enduring challenge for medicinal chemists for over 40 years.7
Benzodiazepine and benzoquinoline orthosteric agonist chemical series (Figure 1) have been at
the forefront of published D1-selective chemical matter for decades.8 Both of these chemical
ACS Paragon Plus Environment

Page 3 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
series, as well as other known D1-selective agonists,9 are based on a dopamine-containing core,
and retain the amine and catechol functionalities (blue) of the native ligand. A hallmark of these
chemical series is impaired brain penetration; additionally, rapid in vivo clearance following
bioconjugation, mediated by systems which specifically recognize and metabolize the catechol
moiety, further hampers the development of this chemotype.8c, 10
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
An impressive number of medicinal chemistry efforts have been specifically directed toward
identifying clinically viable compounds11 but, to date, the only D1-selective agent to reach the
market is Fenoldopam (5) (Figure 1),12 a rapid-acting, peripherally restricted13 vasodilator that is
dosed intravenously (IV) to treat acute hypertensive crisis.14 Despite these challenges, the D1
system remains an attractive therapeutic target.
Herein we describe the screening and medicinal chemistry strategies which led to the
identification and optimization of a novel series of non-catecholamine D1-orthosteric agonists
that are both orally bioavailable15 and CNS-penetrant. Lead examples from this series have an
asymmetric sterically hindered bi-aryl bond and exist as stable atropisomer enantiomers. These
ligands display robust activity in vivo and evidence of significantly reduced D1 receptor
desensitization and recruitment of -arrestin relative to dopamine and other catechol-containing
agonists. This pharmacology profile was unanticipated, but is well suited for the treatment of
chronic diseases, such as schizophrenia and Parkinson’s disease where the duration of
therapeutic effect is an important component of patient quality of life.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 45
1
2
3
4
5
6
HO
HO
NH2
Catechol
7
Dopamine (1)
8
9
R2
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
H
NH
H
HO
HO
Dihydrexidine (2)
HO
HO
NH
R1
R1, R2 =H, SKF 38393 (3)
R1=Cl, R2=H, SKF 81297 (4)
R1=Cl, R2=OH, Fenoldopam (5)
Figure 1: Representative early medicinal chemistry efforts, based on a modified dopamine scaffold, for achieving
D1 vs D2 selectivity. The embedded phenethylamine group (blue) is required for binding and agonism.
Results and Discussion
Screening: Available literature suggested that potent and even modestly selective activators of
D1 that lack the catecholamine motif are rare, although recent reports by Bristol-Myers Squibb16
and Astellas Pharma17 have disclosed the discovery of non-catecholamine D1-selective positive
allosteric modulators (PAMs). In our effort to identify novel activators of this recalcitrant target
we opted to run a high-throughput screen (HTS).18 The HTS was run using cAMP homogeneous
time-resolved fluorescence (HTRF) kits from Cisbio. We found that for this target, cell lines
overexpressing the hD1 receptor in an HEK cell background were more sensitive than those in a
CHO cell background; we thus elected to screen in an HEK293 cell line.
To increase the probability of detecting weak or low partial agonists in an HTS format, we
initially chose to analyze our functional response data in ratio mode relative to dopamine, using
its signal as the 100% effect, as opposed to the more conventional analysis method employing a
cAMP standard curve to convert the data to nM cAMP prior to normalizing to the dopamine
control. Analyzing the data in ratio mode has the effect of enhancing a partial agonist’s maximal
ACS Paragon Plus Environment

Page 5 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
effect relative to dopamine which we theorized would help us identify compounds that would
otherwise fail to be detected. For example, the known D1 partial agonist 319 appears as a
relatively full agonist when analyzed relative to dopamine, but as a partial agonist when analyzed
using the cAMP standard curve (Figure 2A). This analysis was only used to triage HTS hits.
Once SAR efforts commenced in earnest, we switched to using a conventional cAMP standard
curve analysis of the data (cAMP mode, Table 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
Using this approach, we queried approximately 3 million compounds in the Pfizer screening
file in a compressed format (multiple compounds per well) in the presence of an EC10 of
dopamine, so as to capture both PAMs and agonists. In this way, 1,200 compounds with ≥30%
maximal response relative to dopamine were identified (Figure 2B). Catechol- and phenol-
containing compounds, ergot alkaloid derivatives,20 and compounds with either D2 preference
(>10-fold D2/D1) or a strong response in a non-D1-overexpressing parental HEK293-cell line
were eliminated. A final round of confirmatory screening using samples prepared from solid
material with confirmed purity yielded only one hit, compound 6, which exhibited an EC50 of
approximately 1 µM and an 83% maximal response relative to dopamine. It is instructive to note
that had we used the cAMP standard curve to analyze the screening data, we might have missed
this hit, as it gave only borderline potency and efficacy values (~2.5 M and 34%, respectively;
Figure 2C).
In addition to elevation of cAMP in the HEK cell line, in an orthogonal readout of target
engagement, compound 6 displayed direct binding to the human D1 (hD1) receptor in a
radioligand displacement assay utilizing [3H]SCH-2339021 Ki = 2.2 µM). This data suggested
that compound 6 binds in or near the orthosteric site. Compound 6 does not share structural
similarity with known pan-assay interference compounds (PAINs)22 and was not promiscuous in
a focused pharmacology panel comprising GPCRs (9 targets, agonist and antagonist functional
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 45
1
2
3
4
5
6
endpoints, >10 µM), ion channels (4 targets, >10 µM) and phosphodiesterases (11 targets, >100
µM).
7
8
9
A
B
Ratio Mode
cAMP Mode
>30% cAMP Efficacy @ 20 uM
100
75
50
25
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
Remove Catechols/Phenols
and Ergot Alkaloids
D1 > D2
-11
-9
-7
-5
Log[M], Compound 3
Confirm From
Powder
C
Ratio Mode
cAMP Mode
N
N
100
75
50
25
0
O
N
O
-11
-9
-7
-5
PF-4211 (6)
Log[M], Compound 6
Figure 2: A) Composite cAMP functional dose-response curve of D1 partial agonist compound 3 (N=≥50) in an
HEK cell line overexpressing hD1R. Data plotted according to 2 different analysis methods: ratio mode (blue) and
cAMP mode (red); B) High-level D1 HTS screening funnel. Compound 6 was the sole confirmed hit from this
screening campaign; C) Composite cAMP functional dose-response curve in ratio mode (EC50: 1.0 M, 83%
maximal effect) and cAMP mode (EC50: ~2.5 M, 34% maximal effect) of D1 HTS hit compound 6 (N=≥40) in the
absence of exogenous dopamine in an HEK293 cell line overexpressing the hD1 receptor.
Synthesis: The modular structure of compound 6 was exploited to rapidly generate SAR
around the A, B and C ring systems (Figure 3) entirely by parallel medicinal chemistry (PMC).15
Early work targeting ether linker replacements was non-productive, leading us to theorize that
the ether function between rings B and C provides a desirable orientation for D1 binding, in
which the pyridine lone pair repels the oxygen lone pairs, creating an orientation biased as drawn
(Figure 3).23 The A-B linkage was established through coupling reactions of available
monomers, using standard palladium-catalyzed Suzuki cross-coupling, or via condensation
ACS Paragon Plus Environment

Page 7 of 45
Journal of Medicinal Chemistry
1
2
3
4
reactions to assemble the A ring heterocycles. SNAr coupling reactions delivered the B-C ether
5
6
bond through use of the commercial chlorofuropyridine 7.
7
8
9
Suzuki Coupling or
Condensation
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
R
SNAr or Ullman
A
X
Coupling
( )
n=1,2
B
C
O
N
O
X= C, N
Figure 3: Key synthetic disconnections for analogs of compound 6. Highlighted: heterocyclic A ring (blue), phenyl
core B ring (orange), furopyridine C ring (red).
Pyrazole-containing analogs with diversity introduced at the 1-nitrogen of the pyrazole
were prepared as shown in Route A (Scheme 1). Synthesis of these derivatives proceeded via
SNAr reaction between chlorofuropyridine 7 and the requisite phenolic acetophenone.
Subsequent treatment of 8a,b with DMF-DMA provided PMC templates 9a,b, which were then
condensed with substituted hydrazines in a parallel format to give substituted pyrazoles 18–20.
Pyrazole 6 was prepared in an analogous fashion, as shown in Route C, via an Ullman coupling
of 7 with intermediate 13.
Heterocyclic diversity on the A ring was primarily achieved using Route B. Aryl bromide
templates 10a and 10b were prepared by SNAr reaction of 7 with 4-bromophenol and 4-bromo-3-
methylphenol, respectively. Subsequent Suzuki cross-coupling of 10b with the appropriate
heterocyclic boronate provided compound 17. An orthogonal approach involved synthesis of
boronate templates 11a,b via treatment of bromides 10a,b with bis(pinacolato)diboron under
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 45
1
2
3
4
5
6
palladium catalysis. Heterocyclic halides24 were coupled with 11a,b under palladium-catalyzed
cross-coupling conditions to provide analogs 23–29.
7
8
9
N-Linked azabenzimidazoles at the A ring were constructed as shown using route D: 4-
chloro-3-nitropyridine (15) underwent SNAr displacement with commercial anilines 14a,b to
provide biaryl intermediates. Reduction of the nitro functional group via hydrogenation,
followed by condensation with ethyl orthoacetate, afforded the corresponding N-aryl
azabenzimidazoles. Demethylation of the methyl ether with BBr3 provided phenols 16a,b, which
were coupled under Ullmann conditions with chlorofuropyridine 7 to provide compounds 21 and
22.
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

Page 9 of 45
Journal of Medicinal Chemistry
1
2
3
4
O
5
6
BnO
7
12
8
9
h, i, j
Route 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
N
N
R1
O
R1
O
HO
N
13
a
b
c
O
k
O
Route A
R1
N
N
Het
O
O
Cl
9a:
9b:
R1 = H
R1 = CH3
8a:
8b:
R1 = H
R1 = CH3
O
N
N
O
O
7
R1
6, 18, 21, 23, 24:
R1 = H
17, 19, 20, 22, 25-29:
R2
R1 = CH3
O
d
f
g
or
p
N
Route B
O
N
R1
N
10a:
10b:
R1 = H, R2 = Br
R1 = CH3, R2 = Br
N
e
HO
11a:
11b:
R1 = H, R2 = pinacol boronate
R1 = CH3, R2 = pinacol boronate
16a:
16b:
R1 = H
R1 = CH3
Route D
l, m, n, o
NH2
R1
O2N
Cl
N
OMe
15
14a:
14b:
R1 = H
R1 = CH3
Scheme 1: (a) 1-(4-hydroxyphenyl)ethan-1-one or 1-(4-hydroxy-2-methylphenyl)ethan-1-one, Cs2CO3, CsF, NMP,
120 °C, 72 h; (b) DMF-DMA, DMF, Et2O-BF3, 110 °C, 6 h; (c) t-butylhydrazine HCl, MeOH, H2O, HCl, 30 °C, 16
h; (d) 4-bromophenol or 4-bromo-3-methylphenol, Cs2CO3, DMSO, 125 °C, 16 h; (e) bis(pinacolato)diboron,
Pd(dppf)Cl2, KOAc, 85 °C to 100 °C, 16 h; (f) compound 10a or 10b, heteroaryl boronic acid, Na2CO3, Pd(PPh3)4,
dioxane, H2O, 130 °C, 1 h, microwave; (g) compound 11a or 11b, heteroaryl bromide, Pd(OAc)2 or Pd(PPh3)4,
K2CO3, DMF/H2O, 70 °C to 80 °C, 0.5–18 h; (h) DMF-DMA, reflux, 16 h; (i) methylhydrazine, 75 °C, 3 h; (j) 1 atm
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 45
1
2
3
4
5
6
H2, Pd/C in EtOH, H2O, rt, 16 h; (k) compound 7, Cs2CO3, CuI, 2,2,6,6-tetramethyl-3,5-heptanedione, NMP, 130
°C, 16 h; (l) Et3N, EtOH, 16 h; (m) 45 psi H2, Pd/C, MeOH, 24 h; (n) Ac2O, CH3C(OEt)3, 145 °C, 1 h then 100 °C,
48 h; (o) BBr3, CH2Cl2, −78 °C to rt; (p) compound 7, Cs2CO3, DMSO, 140 °C, 16 h.
7
8
9
SAR Studies: In the initial rounds of PMC, monomer diversity was prioritized over the
physical chemical properties of the final analogues in order to maximize diversity and increase
the probability of identifying potency trends. Final compounds were limited to those with a
cLogP < 6.5 and molecular weight <450. Varying the A and C rings using established chemistry
furnished nearly 1,000 compounds in this initial design cycle. For brevity, only compounds
whose SAR guided the discovery of compound 2925 are showcased (Table 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
Preliminary SAR around the pyrazole tail group was generated using condensation chemistry
(Scheme 1, Route A). This library yielded compounds 17–20, which were more potent activators
of D1 than HTS hit 6 and also had increased affinity in the binding assay. The greatest impact of
this data was the observation that increasing steric bulk about the biaryl bond imparted
significant gains in both D1 affinity and functional activity, leading to the structural hypothesis
that orthogonal presentation of the A and B rings (Figure 4) results in improved receptor
interactions. These gains in activity were also realized with the matched molecular pair (MMP)26
imidazopyridines 21 and 22 which were derived using Route D. This pair, featuring di- and tri
ortho-substituted biaryl bonds, showed only a modest increase in functional potency and binding
affinity, 0.3 and 2-fold respectively relative to each other indicating limits to the activity gains
that can be acheived using this strategy.
Methoxyisoquinoline 23 and imidazopyridine 24 were notable because they were the first
compounds synthesized in this chemical series with a binding affinity Ki < 100 nM. The mono
ortho-substituted biaryl system of compound 24 provided an opportunity to use the knowledge
gained from the preliminary SAR generated on the pyrazoles to prospectively design for
ACS Paragon Plus Environment

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
increased potency, by increasing the energy required for the biaryl bond to rotate, as depicted in
Figure 4.
7
8
9
Replacing the phenyl core with an ortho-tolyl ring (25) gave a 3-fold boost in potency and
affinity over its MMP compound 24. Adding a methyl on the imidazopyridine tail group,
affording compound 26, provided impressive gains in potency and affinity, 7-fold and 12-fold
respectively, versus MMP compound 24. Combining the methyl imidazopyridine tail group with
the toluene core gave compound 27, which exhibited no gain in affinity relative to compound 26
but was functionally 2-fold more potent. Importantly, compound 27 demonstrated a functional
EC50 < 100 nM in the recombinant cell system; its cLogP (5.1), however, was outside of our
targeted range. Triazolopyridine 28, which offers lower lipophilicity, lost considerable potency
and affinity relative to its MMP triazolopyridine 25. This can be partially attributed to a
significantly reduced torsional barrier (7.1 vs 12.4 kcal/mol). The potency and affinity of
imidazopyridine 29, however, was similar to compound 27, while providing a 1.1 unit decrease
in cLogP.
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 LipE27 analysis (Figure 5) of the compounds in Table 1 summarizes the
improvement in chemical matter made from HTS hit 6 (binding LipE: 2.1) over the three
described design cycles that ultimately culminated in compound 29 (binding LipE 4.1). The first
design cycle, using PMC to target replacements of the A ring, provided 24, whose increased
activity was offset by increased lipophilicity resulting in a marginal improvement in LipE
(binding LipE : +0.2). The second design cycle sought to improve potency by increasing the
torsional barrier about the biaryl bond between the A and B rings, resulting in the design and
synthesis of compound 27. Despite the increase in lipophilicity of 27, the additional potency and
affinity netted nearly a full unit increase in LipE (binding LipE : +0.8). In the third design
cycle, efforts to decrease lipophilicity by incorporating additional heteroatoms into the A ring
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 45
1
2
3
4
5
6
were rewarded by discovery of imidazopyridine 29, which retained similar D1 potency and
affinity to 27, thus affording a full unit improvement in LipE.
7
8
9
Table 1: SAR
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
R2
R1
O
N
O
D1 EC50
D1 hKib
LipEc
Ed
(nM),
Cmpd
R1
R2
% Effecta
232, 100%
(nM)
Functional Binding (kcal/mol)
#
1
N/A
N/A
243
6.2
2.1
6.4
2.1
N/A
2.1
N
N
6
H
2,293, 37%
1,232, 34% e
2,242
254
N
N
N
N
N
17
Me
2.2
0.9
2.9
1.0
5.7
0.3
N
18
19
H
2,411, 11% e
1,108, 7%e
1,680 e
N
N
Me
563f
1.0
1.1
1.3
2.0
11.1
16.2
20
Me
895, 18%e
121f
ACS Paragon Plus Environment

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
N
21
22
H
1,141, 73%
416, 82%
376
244
1.9
1.5
2.3
2.1
6.7
N
N
N
N
Me
14.5
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
(Racemic)
N
23
H
1,333, 64%
56.3
1.3
2.6
9.9
OMe
N
N
N
N
24
25
H
821, 61%
262, 54%
98.9
38.1
1.4
1.7
2.3
2.6
4.8
N
N
Me
12.4
26
H
123, 62%
8.5
1.9
3.2
10.6
N
Me
N
27
28
29
64, 64%
5,485, 26%
106, 86%
6.8
2,346
8.9
2.1
2.6
3.0
3.1
2.8
4.1
17.2
7.1
(Racemic)
N
N
N
N
N
Me
N
N
Me
17.4
(Racemic)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 45
1
2
3
4
5
6
7
8
9
N
N
Me
(R-
N
N
30
31
141, 89%
33, 91%
21.6
11.9
2.9
3.4
3.7
4.0
17.4
17.4
N
N
enantiomer)
Me
(S-
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
enantiomer)
aFunctional potency (EC50) and intrinsic activity (eMax) as a % relative to the maximal effect of dopamine in
HEK293T cells overexpressing the human D1 receptor (geometric mean; N ≥ 3 unless otherwise specified); Data
was converted to nM cAMP prior to normalization to dopamine control; bRadioligand binding (geometric mean; N ≥
3 unless otherwise specified); cLipophilic efficiency (LipE) defined as D1 functional pEC50 or binding pKi – cLogP;
dTorsion scan calculations were performed by rotating around the bond connecting the two rings in the biaryl motif
over a range of −180 to +180 degree in 15 degree angle increments using DFT(B3LYP)/6-31G**//6-31G** from a
TeraChem software package; eN = 1; f N = 2; N/A = not applicable.
A
B
20
15
10
5
N
N
N
-200 -100
0
100 200
Angle
C
N
N
N
N
N
N
N
N
N
+100o
0o
-100o
Figure 4: (A) Highlighted atoms about the biaryl bond indicate the torsion motif used for the dihedral scan
measurement; (B) Rotational energy barrier and example QM torsion plot of compound 29 over a range of −180 to
+180 degrees in 15 degree angle increments using DFT(B3LYP)/6-31G**//6-31G** from a TeraChem software
package; (C) Defined torsional angles.
ACS Paragon Plus Environment

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8.4
8.0
7.6
7.2
6.8
6.4
6.0
5.6
5.2
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
2.8
3.2
3.6
4.0
4.4
4.8
5.2
cLogP
Figure 5: LipE plot of cLogP (X-axis) vs D1 activity (Y-axis). Open circles are the D1 agonist functional activity
pEC50; Solid circles are the D1 binding pKi. Red is compound 6; Yellow is compound 24; Blue is compound 27;
Purple is compound 29; All other compounds are grey. Black arrows indicate the progress made over three design
cycles, while the green arrow is the progress made from the initial D1 HTS hit compound 6 to compound 29.
Atropisomerism:
The physical consequence of increasing the barrier to rotation
between two rings is atropisomerism.28 In our experience, a biaryl bond with a calculated energy
barrier to rotation of >15 kcal/mol (Table 1, Figure 4) is likely to provide two stable chiral
atropisomers at physiologically relevant temperatures29. The individual atropo-enantiomers of
compound 29,
compounds 30 and 31, were separated by chiral supercritical fluid
chromatography (SFC)30 and tested in our primary pharmacology assays. Enantiomer 31 had a 4-
fold higher potency and 2-fold higher affinity towards D1 than its mirror image 30.
o
An analytical sample of 31 was heated to 37 C in 2-butanol for 72h and checked for
racemization. No racemization was detected by HPLC analysis indicating that the thermal barrier
to interconversion is high enough to have physiological relevance. The absolute stereochemistry
of atropo-enantiomer 31 was determined to be aS/P (depending on the nomenclature)31 by light
atom X-ray crystallography using the Hooft method.32 This assignment was further supported by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 45
1
2
3
4
5
6
independent assignment of the less active enantiomer as aR/M 30 using the orthogonal
vibrational circular dichroism (VCD) spectroscopy method.33
7
8
9
Understanding that the energetics of atropisomer interconversion would be relevant for
both synthetic chemistry and any therapeutic advancement for compounds in this hindered biaryl
bond series, we used a streamlined segmented flow method34 to experimentally determine
accurate ΔG‡ and T1/2 values for racemization of compound 30 in solvents of interest (Figure 6;
see SI for data). For example, using this approach, we determined these parameters for 31 and
chiral intermediates leading to 31 (Figure 6B). Route development efforts for compound 31
were facilitated by the discovery that these compounds can be resolved early in the synthesis and
then handled in the laboratory for short periods of time, even at temperatures up to 100 °C in
solution (extrapolated T1/2 = 18 days), without risk of significant erosion of enantiopurity.
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

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
A
5
6
Eyring Plot of Compound 30 Racemization
-12.0
-12.5
-13.0
-13.5
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
195 C
200 C
205 C
210 C
2.06
2.08
2.10
2.12
2.14
1/T (K)*1000
Measured Activation Parameters
B
C
‡
ΔS‡
units
kcal/mol & kcal/mol•K
Δ
H
26.8
-0.01621
Extrapolations from Y = -13489X + 15.604
T (°C)
Δ
G kcal/mol
T1/2
units
0
25
37
60
100
250
31.2
31.6
31.8
32.2
32.9
35.4
37620
548
91.6
4.2
years
years
years
years
days
18.0
35
seconds
Figure 6: (A) Eyring plot for the racemization of compound 30 in a 3-methylbutan-1-ol/toluene (97:3) solution; (B)
Calculated activation parameters for compound 30; (C) Extrapolated ΔG‡ and T1/2 at temperatures ranging from 0–
250 °C.
While our in vitro kinetics study predicted an acceptable half-life for potential
development of 31 as a single atropo-enantiomer, the presence of serum proteins in vivo can
affect the rate of interconversion;35 thus, an in vitro study was conducted to determine the rate of
interconversion under conditions simulating an in vivo system, by incubating compounds 29–31
in human blood, human plasma, and human serum albumin at 37 °C for 24 h followed by
quantitative analysis. In all matrices tested, there was no evidence of interconversion.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 45
1
2
3
4
Physical Chemical Properties and Pharmacokinetics:
With the exception of
5
6
7
8
moderately high lipophilicity (cLogP and measured ElogD36) (Table 2), compound 31 has well-
aligned physical chemical properties (EPSA,37 pKa, and lack of HBD) for CNS penetration (CNS
MPO38: 4.9). Low MDR and BCRP efflux ratios,39 high cell permeability, and low turnover by
human liver microsomes (HLM) contribute to its favorable profile. In rat, compound 31 had
excellent brain availability (Cb,u/Cp,u = 0.8, good IV clearance and half-life, and high oral-
bioavailability.
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 2: Summary of calculated and measured chemical properties, ADME, and rat in vivo PK
for compound 31
N
N
N
O
N
31
O
In vitro ADME
Chemical Properties
MW: 356.4
MDR: 1.7b
BCRP: 1.3c
ElogD/cLogP: 3.4/4.0
EPSA: 57
RRCK: 31 x 10-6 cm/secd
HLM: 22 mL/min/kge
Melting pt: 142-144 oC
Aq.sol., pH 7.4: 132 Ma
pKa: 4.1
Protein Bindingf
Rat In vivo PK
B/Pg = 2.2
human Fu,p: 0.04
rat Fu,p: 0.04
mouse Fu,p: 0.04
rat Fu,b: 0.02
Cb,u/Cp,uh = 0.8
T1/2 = 2.3 h
Cl = 23 mL/min/kgi
F% = 85%j
ACS Paragon Plus Environment

Page 19 of 45
Journal of Medicinal Chemistry
1
2
3
4
b
aMeasured thermodynamic solubility of crystalline solid. Ratio of permeability measured as a rate in 10-6 cm/s in
5
6
7
8
and out (BA/AB) of a Madin-Darby canine kidney (MDCK) epithelial cell line transfected with the MDR1 gene
encoding the P-gp efflux transporter. cRatio of permeability measured as a rate in 10-6 cm/s in and out (BA/AB) of a
BCRP-transfected MDCK cell line.40 dPermeability in the MDCKII-LE (RRCK)40 cell line, given as a rate in 10-6
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
e
f
cm/s. Intrinsic clearance in human liver microsomes (HLM). Fraction unbound41 in plasma (Fu,p) or brain (Fu,b).
gAUC ratio of the drug level in rat brain to the drug level in plasma. AUC ratio of the unbound drug level in rat
h
brain to the unbound drug level in plasma from a single subcutaneous dose of 5.6 mg/kg. iObserved blood clearance
in units of mL/min/kg following a single IV bolus of 5.0 mg/kg, dosed as a solution in saline containing 3 molar
equivalents of HCl. jOral bioavailability following 5 mg/kg dosed PO in a 0.5% methylcellulose vehicle.
Biology and Pharmacology:
Compound 29 and its atropo-enantiomer 31 were
invaluable tool compounds to strengthen our confidence that, despite the lack of structural
similarity to known D1 agonist ligands, this chemical series exerts its pharmacology at the D1
receptor both in-vitro and in-vivo. Compound 31 was submitted to a broad panel of targets to
assess its promiscuity. The panel consisted of agonist and antagonist functional readouts of 9
GPCR’s, 5 ion channels, 3 transporters, and 11 phosphodiesterases (PDEs). Compound 31
exhibited IC50’s <5 µM as an antagonist at the following 4 targets: M1: 4.9 M, CB1: 2.1 M,
H1: 4.6 M, and Nav 1.5: 1.1 M. It has a hERG IC50 of approximately 12 M. It is important to
reiterate that D1 shares considerable homology with D5 and that none of the currently available
agonists distinguish between D1 and D5; this includes 31 which has a D5 Ki of 4.8 nM. It is,
however, exquisitely selective vs D2 (Ki >10 µM).
Agonist activation of D1-like receptors (D1R), which include the D1 and D5 subtypes,
has been previously reported to increase locomotor activity in mice42 via a cAMP-dependent
mechanism. Therefore, the ability of D1 agonist compounds of interest to increase locomotor
activity in mice was used to confirm in vivo functional activity. Compound 29 was tested in the
habituated mouse locomotor assay at multiple test doses (Figure 7A). Similar to other D1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 45
1
2
3
4
5
6
7
8
9
agonists, treatment with compound 29, at the 10 and 32 mg/kg doses produced significant
increases in locomotor activity during the 0-120 minute post-dose time period as compared to
vehicle treatment (Figure 7A). The specificity of this activity was verified when pre-treatment
with D1 antagonist SCH-2339043 (32) attenuated the increase in a dose-dependent manner
(Figure 7B).
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
Cultured rat primary striatal neurons endogenously expressing D144 were used to assess
compound-driven D1 desensitization. Cells were incubated with 10 µM of D1 agonists 1–4,
A77636 (33)45, and SKF-83959 (34)46 for 90 min, thoroughly washed, and then challenged with
D1 catecholamine agonist 4 to measure cAMP response. Figure 7C shows that the D1
catecholamine agonist group (1–4, 33 and 34) induced significant levels of D1 desensitization
(~60-100%), whereas the D1 non-catecholamine agonist group (29–31) had a significantly
reduced degree of desensitization (~20%). The vehicle-pretreated group was normalized to 0%
desensitization. This result suggests that, compared to the catecholamines, the non-
catecholamine agonists are biased towards less desensitization in an endogenous D1 cell system.
Further, we have previously published that, in-vivo, a compound with less -arrestin recruitment
relative to 33 has a sustained in-vivo response, even after repeated dosing, in a cynomolgus
monkey eye blink rate study.15
ACS Paragon Plus Environment

Page 21 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A
B
Vehicle / Vehicle
Vehicle / Cmpd 29 (10 mg/kg)
Cmpd 32 (0.001 mg/kg) / Cmpd 29 (10 mg/kg)
Cmpd 32 (0.0032 mg/kg) / Cmpd 29 (10 mg/kg)
Cmpd 32 (0.01 mg/kg) / Cmpd 29 (10 mg/kg)
Cmpd 32 (0.032 mg/kg) / Cmpd 29 (10 mg/kg)
Vehicle
Cmpd 29 (0.32 mg/kg)
Cmpd 29 (1 mg/kg)
Cmpd 29 (3.2 mg/kg)
Cmpd 29 (10 mg/kg)
Cmpd 29 (32mg/kg)
20000
15000
10000
5000
0
20000
**
**
15000
10000
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
**
**
**
5000
0
C
100
80
60
40
20
0
e
1
2
3
4
33
34
29
30
31
Vehicl
Cmpd
Cmpd
Cmpd
Cmpd
Cmpd
Cmpd
Cmpd
Cmpd
Cmpd
Figure 7: Locomotor activity (total beam breaks during 2 hours) in response to administration of D1
agonists/antagonists. (A) Treatment with compound 29, at 10 and 32 mg/kg, SC, increases the number of beam
breaks. (B) Pretreatment with the D1 antagonist, compound 32, dose-dependently blocks the hyperactivity induced
by compound 29 at 10 mg/kg, SC. One-sided ANOVA with Dunnett’s post-test adjusted ** p < 0.05, ** p < 0.01 vs
vehicle or vs vehicle plus compound 29 (10 mg/kg). (C) Desensitization by catechol agonists (orange) 1-4, 33 and
34 versus non-catechol agonist compounds 29–31 (green) using a cAMP readout, towards subsequent challenge with
agonist SKF-81297.46
Conclusions
Herein we describe the discovery and optimization of the first D1-selective agonist chemical
series with suitable physicochemical properties for oral absorption, brain exposure, and low
clearance. These compounds represent a significant step forward for the exploration of D1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 45
1
2
3
4
5
6
7
8
9
activation. Extensive SAR generated on the A and B rings by PMC quickly provided compounds
with increased potency and binding affinity for the D1 receptor, as well as an early observation
that conformationally restraining these rings into an orthogonal relationship provides a consistent
potency and binding affinity benefit; this led to the discovery of compound 29. Separation of
compound 29 into its atropisomers provided the more active enantiomer 31. Compound 31 has
excellent D1 potency and is selective against D2 and other GPCRs, ion channels, and PDEs.
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 pharmacokinetic properties of 29/31 such as metabolic stability, high passive permeability
and low MDR and BCRP efflux ratios, make this compound suitable for oral dosing and in vivo
study. In vitro pharmacology experiments with 29–31 revealed an intriguing D1 receptor
desensitization profile relative to catechol-containing D1 agonists including dopamine. This
pharmacological profile merits clinical study because in chronic diseases, such as schizophrenia
and Parkinson’s disease, the duration of therapeutic effect is an important component of patient
quality of life.
Experimental:
All procedures performed on animals in this study were in accordance with established
guidelines and regulations, and were reviewed and approved by the Pfizer Institutional Animal
Care and Use Committee. Pfizer animal care facilities that supported this work are fully
accredited by AAALAC International.
Chemistry General Information. All solvents and reagents were obtained from commercial
sources and were used as received. All reactions were monitored by TLC (TLC plates F254,
Merck) or UPLC-MS analysis (Waters Acquity, ESCI +/−, APCI +/−). Melting points were
1
obtained with a Thomas–Hoover melting point apparatus and are uncorrected. H NMR spectra
ACS Paragon Plus Environment

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were obtained using deuterated solvent on a Varian or Bruker 400 MHz instrument. All 1H NMR
shifts are reported in δ units (ppm) relative to the signals for chloroform (7.27 ppm), DMSO
(2.50 ppm) and MeOH (3.31 ppm). All coupling constants (J values) are reported in hertz (Hz).
NMR abbreviations are as follows: br, broadened; s, singlet; d, doublet; t, triplet; q, quartet; p,
pentuplet; m, multiplet; dd, doublet of doublets; ddd, doublet of doublet of doublets. HPLC
purity analysis of the final test compounds was carried out using one of three methods. Method
A: UPLC/UV. WuXi AppTec, Shanghai, China. Column: Agilent Xtimate C18, 5 × 30 mm, 3
μm; UV purity detected at 220 nm; Mobile phase A = 0.1% TFA in H2O; Mobile phase B =
0.1% TFA in CH3CN. Gradient: 1% B to 100% B in 5.0 min. Flow rate: 1.2 mL/min. Method B:
UPLC/UV WuXi AppTec, Shanghai, China. Column: XBridge C18, 2.1 × 50 mm, 5 μm; UV
purity detected at 220 nm; Mobile phase A = 0.0375% TFA in H2O; Mobile phase B = 0.01875%
TFA in CH3CN. Gradient: 1% B to 5% B in 0.6 min, 5% B to 100% B in 4.4 min, 100% B to 1%
B for 0.3 min, hold at 1% B for 0.4 min. Flow rate: 0.8 mL/min. Method C: Column: Waters
Atlantis C18, 4.6 x 50 mm, 5 μm; UV purity detected at 215 nm; Mobile phase A: 0.05% TFA in
H2O (v/v); Mobile phase B: 0.05% TFA in CH3CN (v/v); Gradient: 5% B linear to 95% B in 4.0
min, hold at 95% B to 5.0 min. Flow rate: 2 mL/min. All final compounds were determined to
have a purity of >95% by one of the aforementioned methods.
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
Experimental
4-(4-(1-Methyl-1H-pyrazol-5-yl)phenoxy)furo[3,2-c]pyridine (6). To a stirred solution of 13
(60 mg, 0.340 mmol) and 4-chlorofuro[3,2-c]pyridine (7) (106 mg, 0.688 mmol) in NMP (2 mL)
were added cesium carbonate (283 mg, 0.860 mmol), copper(I)iodide (33 mg, 0.172 mmol), and
2,2,6,6-tetramethyl-3,5-heptanedione, and the reaction mixture was heated to 130 ºC for 16 h.
After addition of water (30 mL), the layers were separated and the aq layer was extracted with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 45
1
2
3
4
5
6
7
8
9
1:1 diethyl ether/hexanes (4 x 30 mL). The combined organic layers were dried over sodium
sulfate, filtered, and concd in vacuo. Purification via silica gel chromatography (Gradient: 30%
to 60% EtOAc in heptanes) afforded a white solid. The white solid was further purified by
HPLC. Xbridge column, (TFA as modifier, 5% to 100% acetonitrile in water) to afford 6 as 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
1
white solid in 49% yield. H NMR (400 MHz, CD3OD) δ 7.96 (d, J = 6.1 Hz, 1H), 7.89 (d, J =
2.1 Hz, 1H), 7.56 (d, J = 8.6 Hz, 2H), 7.51 (d, J = 2.0 Hz, 1H), 7.37 (dd, J = 6.0, 0.9 Hz, 1H),
7.32 (d, J = 8.8 Hz, 2H), 6.93 (dd, J = 2.2, 0.9 Hz, 1H), 6.39 (d, J = 2.0 Hz, 1H), 3.91 (s, 3H).
MS (ES+): m/z 292.3 [M + 1].
1-(4-(Furo[3,2-c]pyridin-4-yloxy)phenyl)ethan-1-one (8a). A stirring mixture of 1-(4-
hydroxyphenyl)ethan-1-one (4.9 g, 36 mmol), 7 (5.5 g, 36 mmol), cesium fluoride (5.5 g, 36
mmol), and cesium carbonate (11.7 g, 36 mmol) in NMP (50 mL) was heated at 120 ºC for 72 h.
The reaction mixture was partitioned between water (150 mL) and EtOAc (300 mL); the organic
layer was washed with 5% NaOH and water, dried over magnesium sulfate, and concd in vacuo.
Purification by silica gel chromatography (Eluent: hexane/EtOAc, 10:1) afforded 8a as an oil in
30% yield.
3-(Dimethylamino)-1-(4-(furo[3,2-c]pyridin-4-yloxy)phenyl)prop-2-en-1-one (9a). A stirring
mixture of 8a (5.8 g, 23 mmol) and DMF-DMA (7.74 g, 65 mmol) in DMF (20 mL) and several
drops of boron trifluoride etherate was heated at 110 ºC for 6 h. After removal of solvent under
reduced pressure, the residue was treated with EtOAc (200 mL) and washed with water (150
mL). The organic layer was separated, dried (sodium sulfate), and concd in vacuo. The residue
was treated with diethyl ether (50 mL) and the resulting precipitate was filtered off, washed with
hexane, and dried to provide 9a as a yellow solid in 73% yield.
4-(4-Bromophenoxy)furo[3,2-c]pyridine (10a). To a solution of 7 (10 g, 65 mmol) in NMP
(200 mL) were added cesium carbonate (65 g, 200 mmol) and 4-bromophenol (14.7 g, 91 mmol),
ACS Paragon Plus Environment

Page 25 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and the reaction was heated to 130 °C for 16 h. At this point, the reaction mixture was cooled to
rt, poured into water (400 mL), and extracted with EtOAc (3 x 200 mL). The combined organic
layer was dried over anhydrous sodium sulfate, filtered, and concd in vacuo. Purification by
chromatography on silica gel (Eluent: petroleum ether) afforded 10a as a pale yellow solid in
86% yield. 1H NMR (400 MHz, CDCl3) δ 8.01 (d, J = 5.9 Hz, 1H), 7.66 (d, J = 2.4 Hz, 1H), 7.54
(d, J = 8.6 Hz, 2H), 7.22 (d, J = 6.7 Hz, 1H), 7.13 (d, J = 8.6 Hz, 2H), 6.90 (d, J = 1.2 Hz, 1H).
MS (ES+): m/z 291.9 [M + 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
4-(4-Bromo-3-methylphenoxy)furo[3,2-c]pyridine (10b). To a stirred solution of 7 (120 g, 781
mmol) in DMSO (1.56 L) were added cesium carbonate (509 g, 1.56 mol) and 4-bromo-3-
methylphenol (161 g, 861 mmol), and the reaction was heated to 125 °C for 16 h. At this point,
the reaction mixture was cooled to rt, poured into water (5 L), and extracted with EtOAc (2 x 2.5
L). The combined organic extracts were washed with water (2.5 L), washed with saturated aq
NaCl solution (2.5 L), dried over anhydrous sodium sulfate, filtered, and concd in vacuo.
Purification by chromatography on silica gel (Eluent: 2% EtOAc in petroleum ether) afforded
10b as a pale yellow solid in 86% yield. 1H NMR (400 MHz, CDCl3)  8.00 (d, J = 6.2 Hz, 1H),
7.64 (d, J = 2.1 Hz, 1H), 7.55 (d, J = 8.3 Hz, 1H), 7.20 (dd, J = 5.8, 0.8 Hz, 1H), 7.12 (d, J = 2.9
Hz, 1H), 6.93 (dd, J = 8.5, 2.7 Hz, 1H), 6.88 (dd, J = 2.5, 0.8 Hz, 1H), 2.41 (s, 3H) . MS (ES+):
m/z 304.0, 306.0 [M + 1].
4-[4-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)phenoxy]furo[3,2-c]pyridine (11a). To a
stirred solution of 10a (14 g, 48.3 mmol) in 1,4-dioxane (150 ml) was added
bis(pinacolato)diboron (14.7 g, 58.0 mmol), potassium acetate (14.2 g, 145 mmol), and [1,1′-
bis(diphenylphosphino)ferrocene]dichloropalladium(II) (3.4 g, 4.83 mmol) and the reaction
mixture was heated at 100 °C for 16 h. After cooling to rt, the reaction mixture was poured into
water (400 mL) and EtOAc (150 mL). The aq layer was extracted with EtOAc (3 x 200 mL), and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
the combined organic layer was washed with brine, dried over anhydrous sodium sulfate,
filtered, and concd in vacuo. The residue was purified by silica gel chromatography (Eluent:
petroleum ether/EtOAc 50:1 to 10:1) to provide 11a as a white solid in 70% yield. 1H NMR (400
MHz, CDCl3)  8.02–8.01 (d, 1H), 7.89–7.87 (d, 2H), 7.62–7.61 (d 1H), 7.21–7.19 (m, 3H),
6.81–6.80 (m, 1H), 1.35 (s, 12H). MS (ES+): m/z 338.2 [M + 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
4-[3-Methyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenoxy]furo[3,2-c]pyridine
(11b). To a stirred solution of 10b (50.0 g, 164 mmol) in 1,4-dioxane (1.02 L) was added
bis(pinacolato)diboron (41.76 g, 164.4 mmol), potassium acetate (64.6 g, 658 mmol) and [1,1’-
bis(diphenylphosphino)ferrocene]dichloropalladium(II) (6.0 g, 8.2 mmol), and the reaction
mixture was heated at 85 °C for 16 h. After cooling to rt, it was filtered through a pad of Celite,
and the pad was washed with EtOAc. The combined filtrates were concd in vacuo and the
residue was purified by silica gel chromatography (Eluent: 2% EtOAc in petroleum ether) to
afford 11b as a white solid in 70% yield. 1H NMR (400 MHz, CDCl3)  8.02 (d, J = 5.8 Hz, 1H),
7.84 (d, J = 7.5 Hz, 1H), 7.61 (d, J = 2.1 Hz, 1H), 7.19 (d, J = 5.8 Hz, 1H), 7.00 (m, 2H), 6.80
(m, 1H), 2.56 (s, 3H), 1.34 (s, 12H). MS (ES+): m/z 352.2 [M + 1].
5-[4-(Benzyloxy)phenyl]-1-methyl-1H-pyrazole. To a stirred solution of compound 12 (15.32
g, 67.71 mmol) in DMF (30 mL) was added DMF-DMA (94%, 19.0 mL, 134 mmol) and the
reaction mixture was heated at reflux for 18 h, whereupon the reflux condenser was replaced
with a distillation head, and distillation was carried out until the temp of the distillate reached
140 ºC. The material in the reaction pot was cooled to rt, treated with methylhydrazine (98%, 7.4
mL, 136 mmol), and heated at 75 ºC for 3 h. The reaction mixture was cooled, diluted with
EtOAc, washed four times with aq 5% NaCl solution, dried over magnesium sulfate, filtered, and
concd in vacuo. Purification via silica gel chromatography (Gradient: 2% to 10% EtOAc in
DCM) afforded 5-[4-(benzyloxy)phenyl]-1-methyl-1H-pyrazole as a light yellow solid in 77%
ACS Paragon Plus Environment

Page 27 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
yield. 1H NMR (400 MHz, DMSO-d6) δ 7.49–7.32 (m, 8H), 7.12 (d, J = 8.6 Hz, 2H), 6.32 (d, J =
1.9 Hz, 1H), 5.17 (s, 2H), 3.81 (s, 3H). MS (ES+): m/z 265.1 [M + 1].
7
8
9
4-(1-Methyl-1H-pyrazol-5-yl)phenol (13). To a stirred solution of 5-[4-(benzyloxy)phenyl]-1-
methyl-1H-pyrazole (13.49 g, 51.04 mmol) in EtOH (125 mL) was added 10% palladium on
carbon (~50% in water, 1.46 g). The reaction mixture was hydrogenated at rt and 1 atm hydrogen
for 16 h, then filtered and concd in vacuo. The residue was triturated with heptane to afford 13 as
a colorless solid in 98% yield. 1H NMR (400 MHz, DMSO-d6)  9.73 (br s, 1H), 7.40 (d, J = 1.9
Hz, 1H), 7.31 (br d, J = 8.7 Hz, 2H), 6.86 (br d, J = 8.7 Hz, 2H), 6.26 (d, J = 1.9 Hz, 1H), 3.79
(s, 3H). MS (ES+): m/z 175.1 [M + 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
N-(4-Methoxy-2-methylphenyl)-3-nitropyridin-4-amine.
A
solution of 4-methoxy-2-
methylaniline (14b) (23.8 g, 173 mmol), 4-chloro-3-nitropyridine (15) (25 g, 160 mmol), and
triethylamine (33.0 mL, 237 mmol) in EtOH (250 mL) was stirred at rt for 16 h, then concd
under reduced pressure. The residue was dissolved in EtOAc (200 mL) and filtered through a
thick pad of silica gel (Eluent: EtOAc, 1 L). The filtrate was concd in vacuo to afford the title
compound as a purple oil, which solidified on standing in 100% yield. This material was used
without further purification. MS (ES+): m/z 260.1 [M + 1].
N4-(4-Methoxy-2-methylphenyl)pyridine-3,4-diamine. Palladium on carbon (10%, 3 x 2.12 g)
was added to each of three batches of N-(4-methoxy-2-methylphenyl)-3-nitropyridin-4-amine
(each approximately 10 g; total 31 g, 120 mmol) in MeOH (3 x 100 mL). The three suspensions
were independently hydrogenated under 45 psi hydrogen at rt on a Parr shaker for 24 h. The
three reaction mixtures were combined, filtered through a pad of Celite, and concd in vacuo.
Purification by silica gel chromatography [Gradient: 2% to 10% (1.7 M ammonia in MeOH) in
1
DCM] afforded the title compound as a light brown solid in 88% yield. H NMR (400 MHz,
CDCl3)  8.01 (s, 1H), 7.88 (d, J = 5.5 Hz, 1H), 7.08 (d, J = 8.6 Hz, 1H), 6.84 (br d, J = 2.8 Hz,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 45
1
2
3
4
5
6
1H), 6.78 (br dd, J = 8.6, 3.0 Hz, 1H), 6.34 (d, J = 5.5 Hz, 1H), 5.66 (br s, 1H), 3.82 (s, 3H), 2.20
(br s, 3H). MS (ES+): m/z 230.1 [M + 1].
7
8
9
1-(4-Methoxy-2-methylphenyl)-2-methyl-1H-imidazo[4,5-c]pyridine. A stirred mixture of N4-
(4-methoxy-2-methylphenyl)pyridine-3,4-diamine (3.95 g, 17.2 mmol), acetic anhydride (1.96
mL, 20.7 mmol), and triethyl orthoacetate (99%, 15.9 mL, 86.4 mmol) was heated at 145 ºC for
1 h, then at 100 ºC for 48 h. After being cooled to rt, the reaction mixture was diluted with
EtOAc (100 mL), washed with saturated aq sodium bicarbonate solution (30 mL), washed with
water, dried over sodium sulfate, filtered, and concd under reduced pressure. Purification by
silica gel chromatography (Gradient: 2% to 5% MeOH in DCM) afforded the title compound as
a light pink oil in 94% yield. 1H NMR (400 MHz, CDCl3)  9.07 (br d, J = 0.8 Hz, 1H), 8.36 (d,
J = 5.5 Hz, 1H), 7.15 (d, J = 8.6 Hz, 1H), 6.97–6.89 (m, 3H), 3.90 (s, 3H), 2.42 (s, 3H), 1.94 (br
s, 3H). MS (ES+): m/z 254.1 [M + 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
3-Methyl-4-(2-methyl-1H-imidazo[4,5-c]pyridin-1-yl)phenol (16b). Boron tribromide (1 M
solution in DCM, 44.1 mL, 44.1 mmol) was added drop-wise to a solution of 1-(4-methoxy-2-
methylphenyl)-2-methyl-1H-imidazo[4,5-c]pyridine (3.72 g, 14.7 mmol) in DCM (150 mL) at
−78 ºC. The reaction mixture was stirred at −78 ºC for 15 min, then the cooling bath was
removed and the reaction mixture was allowed to gradually warm to rt. After 20 h at rt, the
reaction mixture was re-cooled to −78 ºC and slowly quenched with MeOH (20 mL). At this
point, the cooling bath was removed; the mixture was allowed to reach rt and then stir for 15
min. Volatiles were removed in vacuo, MeOH (100 mL) was added, and the mixture was heated
at reflux for 30 min. After concentration under reduced pressure, the resulting solid 16b was
taken directly to the next step. MS (ES+): m/z 240.1 [M + 1].
4-(3-Methyl-4-(1-methyl-1H-pyrazol-5-yl)phenoxy)furo[3,2-c]pyridine (17).
A
stirred
mixture of 10b (200 mg, 0.658 mmol), 1-methyl-1H-pyrazol-5-ylboronic acid (100 mg, 0.794
ACS Paragon Plus Environment

Page 29 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
mmol), sodium carbonate (141 mg, 1.32 mmol), and tetrakis(triphenylphosphine)palladium(0)
(80 mg, 0.066 mmol) in dioxane/water (13:3, 4 mL) was heated in a microwave at 130 ºC for 1
h. The reaction mixture was cooled, filtered through a Celite pad, and concd in vacuo.
9
1
Purification via silica gel chromatography afforded 17 as an off-white solid in 44% yield. H
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
NMR (400 MHz, CD3OD) δ 7.93 (d, J = 5.9 Hz, 1H), 7.85 (d, J = 2.3 Hz, 1H), 7.53 (br s, 1H),
7.34 (d, J = 5.9 Hz, 1H), 7.27 (d, J = 8.2 Hz, 1H), 7.16 (d, J = 2.0 Hz, 1H), 7.08 (dd, J = 8.3, 2.2
Hz, 1H), 6.89 (s, 1H), 6.26 (d, J = 1.4 Hz, 1H), 3.67 (br s, 3H), 2.14 (s, 3H). MS (ES+): 306.1 [M
+ 1].
4-(4-(1-(tert-Butyl)-1H-pyrazol-5-yl)phenoxy)furo[3,2-c]pyridine (18). In library format, a
solution of 9a (75 µmol, 1 eq) and tert-butylhydrazine hydrochloride (90 µmol, 2 eq) in MeOH
(750 µL) and 2M aq HCl (75 µL, 2 eq) were shaken at 30 ºC for 16 h. Solvents were removed in
vacuo and the residue was purified by preparative HPLC to afford 18 as a white solid in 52%
yield. 1H NMR (400 MHz, CDCl3) δ 8.07 (d, J = 5.9 Hz, 1H), 7.68 (d, J = 2.4 Hz, 1H), 7.51 (d, J
= 1.6 Hz, 1H), 7.41 (d, J = 8.6 Hz, 2H), 7.32–7.22 (m, 3H), 6.92 (d, J = 1.6 Hz, 1H), 6.21 (d, J =
1.6 Hz, 1H), 1.53 (s, 9H). MS (ES+): m/z 334.4 [M + 1].
4-(4-(1-(tert-Butyl)-1H-pyrazol-5-yl)-3-methylphenoxy)furo[3,2-c]pyridine (19). The title
compound was prepared using methods closely related to those exemplified for 18. MS (ES+):
m/z 348.0 [M + 1].
4-(4-(1-(tert-Butyl)-4-methyl-1H-pyrazol-5-yl)-3-methylphenoxy)furo[3,2-c]pyridine (20).
The title compound was prepared using methods closely related to those exemplified for 18. MS
(ES+): m/z 362.0 [M + 1].
4-(4-(2-Methyl-1H-imidazo[4,5-c]pyridin-1-yl)phenoxy)furo[3,2-c]pyridine (21). The title
compound was prepared using methods closely related to those exemplified for 22 to afford 21
1
as a white solid in 90% yield. H NMR (400 MHz, CDCl3) δ 9.07 (s, 1H), 8.40 (d, J = 5.7 Hz,
ACS Paragon Plus Environment