Discovery of Novel Indazole Derivatives as Orally Available β3-Adrenergic Receptor Agonists Lacking Off-Target-Based Cardiovascular Side Effects

Article abstract of DOI:10.1021/acs.jmedchem.6b01197

We previously discovered that indazole derivative 8 was a highly selective β₃-adrenergic receptor (β₃-AR) agonist, but it appeared to be metabolically unstable. To improve metabolic stability, further optimization of this scaffold was carried out. We focused on the sulfonamide moiety of this scaffold, which resulted in the discovery of compound 15 as a highly potent β₃-AR agonist (EC₅₀ = 18 nM) being inactive to β₁-, β₂-, and α_1A-AR (β₁/β₃, β₂/β₃, and α_1A/β₃ > 556-fold). Compound 15 showed dose-dependent β₃-AR-mediated responses in marmoset urinary bladder smooth muscle, had a desirable metabolic stability and pharmacokinetic profile (C_max and AUC), and did not obviously affect heart rate or mean blood pressure when administered intravenously (3 mg/kg) to anesthetized rats. Thus, compound 15 is a highly potent, selective, and orally available β₃-AR agonist, which may serve as a candidate drug for the treatment of overactive bladder without off-target-based cardiovascular side effects.

Full text of DOI:10.1021/acs.jmedchem.6b01197

Subscriber access provided by University of Newcastle, Australia
Article
3
Discovery of Novel Indazole Derivatives as Orally Available #-Adrenergic
Receptor Agonists Lacking Off-Target-Based Cardiovascular Side Effects
Yasuhiro Wada, Seiji Nakano, Akifumi Morimoto, Ken-ichi Kasahara, Takahiko Hayashi, Yoshio Takada,
Hiroko Suzuki, Michiko Niwa-Sakai, Shigeki Ohashi, Mutsuhiro Mori, Takatsugu Hirokawa, and Satoshi Shuto
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01197 • Publication Date (Web): 29 Mar 2017
Downloaded from http://pubs.acs.org on March 30, 2017
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
Discovery of Novel Indazole Derivatives as Orally Available β3ꢀAdrenergic
6
7
Receptor Agonists Lacking OffꢀTargetꢀBased Cardiovascular Side Effects
8
9
Yasuhiro Wada*,a,b Seiji Nakano,a Akifumi Morimoto,a Kenꢀichi Kasahara,a Takahiko Hayashi,a Yoshio Takada,a
Hiroko Suzuki,a Michiko NiwaꢀSakai,a Shigeki Ohashi,a Mutsuhiro Mori,a Takatsugu Hirokawa,b,d,e and Satoshi
Shuto*b,c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aPharmaceutical Research Center, Asahi Kasei Pharma Corporation, 632ꢀ1, Mifuku, Izunokuni, Shizuoka 410ꢀ2321,
Japan
bFaculty of Pharmaceutical Sciences and cCenter for Research and Education on Drug Discovery, Hokkaido University,
Kitaꢀ12, Nishiꢀ6, Kitaꢀku, Sapporo 060ꢀ0812, Japan
dMolecular Profiling Research Center for Drug Discovery (molprof), National Institute of Advanced Industrial Science
and Technology (AIST), 2ꢀ4ꢀ7 Aomi, Kotoꢀku, Tokyo 135ꢀ0064, Japan
eDivision of Biomedical Science, Faculty of Medicine, University of Tsukuba, 1ꢀ1ꢀ1 Tennodai, Tsukuba, Ibaraki
305ꢀ8575, Japan
Running Title: Structurally Novel Human β3ꢀAdrenergic Receptor Agonists
Corresponding Author:
Yasuhiro Wada
Phone: +81ꢀ558ꢀ76ꢀ8494. Fax: +81ꢀ558ꢀ76ꢀ5755. Eꢀmail: wada.yp@om.asahiꢀkasei.co.jp
Satoshi Shuto
Phone & Fax: +81ꢀ11ꢀ706ꢀ3769. Eꢀmail: shu@pharm.hokudai.ac.jp
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
Abstract
4
5
6
7
8
We previously discovered that indazole derivative 8 was a highly selective β3ꢀadrenergic receptor (β3ꢀAR) agonist, but
it appeared to be metabolically unstable. To improve metabolic stability, further optimization of this scaffold was
carried out. We focused on the sulfonamide moiety of this scaffold, which resulted in the discovery of compound 15 as
a highly potent β3ꢀAR agonist (EC50 = 18 nM) being inactive to β1ꢀ, β2ꢀ, and α1AꢀAR (β1/β3, β2/β3, and α1A/β3
>556ꢀfold). Compound 15 showed doseꢀdependent β3ꢀARꢀmediated responses in marmoset urinary bladder smooth
muscle, had a desirable metabolic stability and pharmacokinetic profile (Cmax and AUC), and did not obviously affect
heart rate or mean blood pressure when administered intravenously (3 mg/kg) to anesthetized rats. Thus, compound 15
is a highly potent, selective, and orally available β3ꢀAR agonist, which may serve as a candidate drug for the treatment
of overactive bladder without offꢀtargetꢀbased cardiovascular side effects.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Introduction
The International Union of Pharmacology recognizes three adrenergic receptor (AR) subfamilies, the α1ꢀ, α2ꢀ, and
βꢀAR subfamilies, with β1ꢀ, β2ꢀ, and β3ꢀARs as members of the latter.1 The β1ꢀAR is the dominant receptor in heart and
adipose tissue, whereas the β2ꢀAR is responsible for relaxation of vascular, uterine, and airway smooth muscle.2 The
β3ꢀAR mediates metabolic effects such as lipolysis and thermogenesis in adipose tissues.3 Given that β3ꢀAR mRNA is
present in human white fat, gall bladder, stomach, small intestine, prostate, and urinary bladder detrusor,4 this particular
receptor represents an attractive target for drug discovery efforts that may culminate in the development of potent and
selective β3ꢀAR agonists.5 Because many β3ꢀAR agonists induce weight loss in obese animals and exert antiꢀdiabetic
effects in rodent models of type 2 diabetes, these agonists were investigated in clinical trials for the treatment of obesity
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and/or diabetes.6 However, these agonists failed during early clinical investigations, likely because the expression
pattern of β3ꢀAR in most animal species differs considerably from that in humans, particularly in adipose tissues.7
Moreover, many of the agonists also showed either poor pharmacokinetic (PK) profiles and/or cardiovascular adverse
effects due to a lack of selectivity for the β3ꢀAR over β1ꢀ and β2ꢀARs.8 However, these earlier β3ꢀAR agonists were
observed to relax isolated rat and human detrusor muscle strips, and to increase bladder capacity in rats.9 Therefore,
pharmaceutical companies have recently been endeavoring to discover potent, selective, and orally bioavailable β3ꢀAR
agonists for the treatment of overactive bladder (OAB).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
OAB is defined as urgency with or without urinary incontinence, usually associated with frequency and nocturia.11
Populationꢀbased estimations demonstrated that OAB affects 11.8% of adults, and that its rates are similar between men
and women and increase with age.12 OAB can be extremely distressing to the patient and has been shown to negatively
affect quality of life.13 OAB is associated with enhanced voiding frequency, increased risk of depression, sleep
disorders, skin infections, and falls with fractures.13 Detrusor overactivity is a urodynamic observation characterized by
involuntary detrusor contractions during the filling phase that may be spontaneous or provoked.14 Muscarinic
antagonists have been used as firstꢀline treatment drugs for OAB for many years.15 These drugs demonstrate
improvement in continence, number of micturitions, and urgency.16 However, they are associated with mechanistic side
effects such as dry mouth, constipation, and blurred vision, and in rare instances with urinary retention.16 A recent study
found that the persistence rates with pharmacotherapy for OAB were ≤35% for twelve months after the initial
prescription of muscarinic antagonists.17 The side effects can be related to the limitation of their longꢀterm adherence in
the pharmacotherapy of OAB, at least to some extent.18 Therefore, alternative drugs are sought to both avoid adverse
effects and provide more potent therapeutic options. The β3ꢀAR agonists have emerged as a promising class of such
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
drugs, because this class may be less prone towards inducing adverse effects in comparison with muscarinic
antagonists.19
6
7
8
9
Clinical studies have examined several β3ꢀAR agonists such as 1 (mirabegron, YMꢀ178),20 2 (solabegron,
GW427353),21 3 (ritobegron, KUCꢀ7483),22 and 4 (vibegron, MKꢀ4618),10a which are shown in Figure 1. In 2012
Mirabegron was approved for the treatment of OAB,23 and its treatment showed significant efficacy for OAB and
resulted in a lower dry mouth incidence compared to muscarinic antagonists.19 However, treatment with mirabegron
caused hypertension and increased heart rate (HR) in both animals and healthy individuals.19, 24 We hypothesize that,
the increased HR resulting from treatment with mirabegron is at least partially induced via activation of the β1ꢀAR,24a
and accordingly, highly selective novel β3ꢀAR agonists would be effective drugs for the treatment of OAB without the
risk of such cardiovascular side effects as increased HR. In contrast to mirabegron, which inhibits CYP2D6 and
generally should not be coadministered with CYP2D6 substrates,25 highly selective novel β3ꢀAR agonists should be
inactive to CYP enzymes.
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 our previous publication, we reported on novel indazole derivatives, represented by compound 8, which exhibited a
highly selective β3ꢀAR agonistic activity without cardiovascular side effects.10c However, PK studies of 8 in rats
indicated that it was metabolically unstable and had poor PK properties (Cmax and AUC). Based on these findings, we
performed a structure–activity relationship (SAR) study of the sulfonamide moiety of the indazole derivatives to
improve the metabolic stability whilst maintaining the excellent selectivity over α1AꢀAR, and obtained the optimized
compound 15, which had more desirable PK properties than 8 in rats. We also conducted a docking study at the β3ꢀAR
binding site, which identified the structural features required for high affinity for the β3ꢀAR as well as high selectivity
for the β3ꢀAR over α1Aꢀ, β1ꢀ, and β2ꢀARs. Thus, we have successfully identified compound 15 as a highly potent,
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
selective, and orally available β3ꢀAR agonist lacking cardiovascular side effects.
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 1. Chemical structures of 1, 2, 3, and 4.
Results and Discussion
Characteristics of compound 8. We evaluated the in vitro ADME and PK profiles of four 3ꢀsubstituted indazole
β3ꢀAR agonists 5ꢀ8 previously reported by us.10c As shown in Table 1, the human unbound microsomal intrinsic
clearance (hCLint,u) and the rat unbound microsomal intrinsic clearance (rCLint,u) for 7 and 8 were determined to be
unacceptably higher (thereby indicating greater metabolic instability) than 5, even though 7 and 8 possessed excellent
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
4
5
6
7
8
9
β3ꢀAR selectivity while 5 was poorly selective for the β3ꢀAR over α1AꢀAR. Furthermore, 8 showed an undesirable PK
profile (Cmax and AUC) in rats compared with 5 (Table 2). On the other hand, 8 had moderate MDCK cell monolayer
permeability and good solubility in aqueous solutions at a pH 1.2 and pH 6.8 (Table 2). The low Cmax and AUC of 8
may be due to its metabolic instability.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These previous results indicated that the selectivity of our indazole compounds for the β3ꢀAR and α1AꢀAR changed
depending on the substituent at the 3ꢀpositon. Therefore, when ordered in terms of their selectivity for the β3ꢀAR over
α1AꢀAR, the indazole compounds ranked as follows: isoꢀpropyl (8) > cyclobutyl (7) > ethyl (6) > methyl (5). In contrast,
the rank order of these compounds in terms of their metabolic stability (hCLint,u) was as follows: methyl (5) > ethyl (6)
> isoꢀpropyl (8) > cyclobutyl (7). Thus, highly selective compounds (7 and 8) showed poor metabolic stability.
Our aim was to improve the metabolic stability of these compounds while maintaining highly selective β3ꢀAR agonist
activity. In sulfonamideꢀcontaining AR agonists, the α1Aꢀ or β3ꢀAR agonistic activity can be manipulated by changing
the sulfonamide structure.26 Therefore, we decided to optimize the metabolically stable 3ꢀmethyl indazole derivative 5
by modifying its sulfonamide moiety in order to improve its β3ꢀAR agonistic selectivity over α1AꢀAR.
Table 1. AR Agonist Activity and In Vitro Metabolic Stability
a
a
Selectivity
Metabolic Stabilitya
hβ3
hα1A
Compound
R
EC50
IAb
EC50
IAc
hCLint,u
rCLint,u
α1A / β3
ACS Paragon Plus Environment

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(nM)
(%)
82
(nM)
(%)
71
(mL/min/kg) (mL/min/kg)
ꢀMe
ꢀEt
21 ± 0.67
14 ± 0.58
219 ± 6.9
857 ± 164
10
61
<10
42
103
113
5
6
78
31
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
66
68
69
>10000
>10000
4
>151
>769
463
193
641
211
7
8
ꢀiꢀPr
13 ± 1.5
9.1
aThe results are shown as the mean ± SEM (n = 3) or are presented as the average of two experiments. IA (intrinsic
b
c
activity): maximum response induced by isoproterenol was defined as 100%. IA: maximum response induced by
norepinephrine was defined as 100%.
Table 2. In vitro ADME and In Vivo Pharmacokinetics in Rats
Compound
R
5
8
ꢀMe
ꢀiꢀPr
Physicochemical Properties and in vitro ADME Profiles
Metabolic Stabilitya
hCLint,u (mL/min/kg)
rCLint,u (mL/min/kg)
<10
103
193
211
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
Permeabilitya
4
5
6
7
8
9
MDCK (x10ꢀ6 cm/s)
Solubilitya
2.4
4.1
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
pH 1.2 solution (µM)
pH 6.8 solution (µM)
285
273
295
267
Pharmacokinetics Parameters in Ratsa
5 mg/kgb
Cmax (µg/mL)
AUC (µg·hr/mL)
0.107 ± 0.034
0.542 ± 0.065
0.0452
0.200
aThe results are shown as the mean ± SD (n = 3) or are presented as the average of two experiments. bCompounds were
administered orally to rats in solution with saline or water.
Altenbach et al. reported a SAR study on potent sulfonamideꢀcontaining α1AꢀAR agonist 9 (Figure 2),26a for which
increases in the size of the aliphatic group (R1) of the sulfonamide resulted in decreased α1AꢀAR potency, while
aromatic substitution of the sulfonamide made the compounds inactive to α1AꢀAR.26a On the other hand, when Sawa et
al. compared aliphatic sulfonamide 10a with aromatic sulfonamide 10b, the β3ꢀAR agonistic potency of 10b was higher
than that of 10a.26b Taking these findings into account, we hypothesized that changing the size of the substitution on the
sulfonamide of compound 5 would improve selectivity for β3ꢀAR agonistic activity over α1AꢀAR without sacrificing its
desirable human microsome, rat microsome, and rat PK profile (Figure 2).
ACS Paragon Plus Environment

Page 9 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
Figure 2. Design strategy of sulfonamide moiety.
SAR of the sulfonamide moiety. Agonistic activities for the β3ꢀAR of analogues of 5, which were modified at the
sulfonamide moiety, are shown in Table 3. Replacement of the methyl substituent of 5 with an nꢀPr (12) slightly
increased β3ꢀAR agonistic potency (5, EC50 = 21 nM vs. 12, EC50 = 6.5 nM). However, by replacing the methyl
substituent with iꢀPr (13), the β3ꢀAR agonistic activity was decreased (5, EC50 = 21 nM vs. 13, EC50 = 70 nM). We next
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
6
7
8
9
synthesized and evaluated compounds with a cycloalkyl substitution; replacement with a cyclopropyl or cyclobutyl
group increased β3ꢀAR agonistic potency (13, EC50 = 70 nM vs. 14, EC50 = 43 nM; 15, EC50 = 18 nM). However,
replacement with a cyclopentyl group (16) decreased potency (13, EC50 = 70 nM vs. 16, EC50 = 309 nM). On the other
hand, the phenyl sulfonamide derivative 17 (EC50 = 3.9 nM) showed higher potency than the alkyl congeners. Thus, in
order of their β3ꢀAR agonist activity, the compounds were ranked as follows: 17 > 12 > 15, 5 > 14 > 13 > 16.
The agonistic potency of these compounds for the α1AꢀAR is also summarized in Table 3. Although replacement of the
methyl substituent with an nꢀPr did not change activity (5, EC50 = 219 nM vs. 12, EC50 = 163 nM), replacement with
iꢀPr (13) and cyclopropyl (14) showed remarkably weakened agonistic activity for the α1AꢀAR (13, EC50 = 1757 nM;
14, EC50 = 1250 nM). Finally, the change to a cyclobutyl (15), cyclopentyl (16), or phenyl (17) group made these
compounds inactive to α1AꢀAR (all EC50 >10000 nM). Thus, in order of α1AꢀAR agonist activity, the compounds were
ranked as 5, 12 > 13, 14 >> 15, 16, 17.
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
Based on their agonistic potencies for the β3ꢀ and α1AꢀARs, these compounds were ranked in terms of their selectivity
for β3ꢀAR over α1AꢀAR as follows: 17 > 15 >> 12, 13, 14, 16 > 5. The most selective, 15 and 17, were next evaluated
for β1ꢀAR and β2ꢀAR agonistic activities, and the results indicated that both compounds were inactive to these two AR
subtypes (Table 4). Thus, 15 and 17 were identified as highly selective β3ꢀAR agonists, and were selected for the next
pharmacological and ADME evaluation.
Table 3. Human β3ꢀ and α1AꢀAR Agonistic Activitiesa
ACS Paragon Plus Environment

Page 11 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
b
b
Compound
R1
α1A
Selectivity
β3
EC50 (nM)
21 ± 0.67
6.5 ± 0.94
70 ± 6.4
43 ± 4.3
18 ± 2.6
309 ± 69
3.9 ± 0.067
86 ± 3.7
NT
IA (%)c
82
EC50 (nM)
IA (%)d
α1A / β3
ꢀMe
ꢀnꢀPr
ꢀiꢀPr
219 ± 6.9
163
71
84
27
26
0
10
5
79
25
12
90
1757
25
13
72
1250
29
14
105
58
>10000
>10000
>10000
NT
>556
>32
15
16
0
ꢀPh
85
0
>2564
17
isoproterenol
norepinephrine
100
9.1 ± 0.52
100
aThe results are shown as the mean ± SEM (n = 3) or are presented as the average of two experiments. bHuman β3ꢀ or
α1AꢀAR agonist assay, reference 10c. cIA (intrinsic activity): maximum response induced by isoproterenol was defined
as 100%. dIA: maximum response induced by norepinephrine was defined as 100%. NT: not tested.
Table 4. Summary of Human βꢀ and α1ꢀARs Agonistic Activitiesa
b
b
b
b
Compound
β3
β1
β2
α1A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
4
5
6
7
8
9
EC50
(nM)
IA
(%)c
69
EC50
(nM)
IA
(%)c
11.9
4
EC50
(nM)
IA
(%)c
5.6
4
EC50
(nM)
IA
(%)d
9.1
0
13 ± 1.5
18 ± 2.6
3.9 ± 0.067
86 ± 3.7
NT
>10000
>10000
>10000
3.2 ± 0.29
NT
>10000
>10000
>10000
13 ± 3.0
NT
>10000
>10000
>10000
NT
8
15e
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
105
85
10
0
0
17
isoproterenol
norepinephrine
100
100
100
9.1 ± 0.52
100
aThe results are shown as the mean ± SEM (n = 3) or are presented as the average of two experiments. bHuman β1ꢀ, β2ꢀ,
c
β3ꢀ or α1AꢀAR agonist assay, reference 10c. IA (intrinsic activity): maximum response induced by isoproterenol was
d
e
defined as 100%. IA: maximum response induced by norepinephrine was defined as 100%. Compound 15 exhibited
insignificant agonistic activity for α1BꢀAR and α1DꢀAR (EC50 >10000 nM). NT: not tested.
ADME profiles. ADME profiles of 15 and 17 as well as 8 in vitro are summarized in Table 5. Although human
microsomal clearance of 15 was excellent (hCLint,u = 57 mL/min/kg), that of 17 was not acceptable (hCLint,u = 464
mL/min/kg, eightꢀfold worse than 15). Rat microsomal clearance of 15 was also acceptable (rCLint,u = 52 mL/min/kg,
fourꢀfold better than 8). Both 15 and 17 had moderate MDCK cell monolayer permeability and good solubility in pH
1.2 and pH 6.8 solutions. Compound 15 also had improved Cmax, AUC and halfꢀlife values in comparison with 8.
Compound 15 was next evaluated with a panel of cytochrome P450 enzymes, where it showed only insignificant
inhibition of the CYPs (< 34% inhibition at 10 ꢀM against CYP1A2, 2B6, 2C19, 2C8, 2C9, 2D6, and 3A4). In the
same evaluation, 1 showed a significant inhibition of 87% at 10 ꢀM against CYP2D6. The low inhibitory capacity of
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
15 against CYPs suggests that 15 has a much lower risk of drug–drug interactions than 1 when coadministered with
4
5
6
CYPꢀmetabolized drugs.
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 5. In Vitro ADME and In Vivo Pharmacokinetics of 8, 15, and 17 in Rats
Compound
8
15
17
R1
R2
ꢀMe
ꢀiꢀPr
ꢀPh
ꢀMe
ꢀMe
Physicochemical and in vitro ADMET profiles
Metabolic Stabilitya
hCLint,u (mL/min/kg)
rCLint,u (mL/min/kg)
Permeabilitya
193
211
57
52
464
NT
MDCK (x10ꢀ6 cm/s)
Solubilitya
4.1
5.0
5.1
pH 1.2 solution (µM)
pH 6.8 solution (µM)
295
267
299
278
300
229
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
Pharmacokinetics parameters in rata
4
5
6
7
8
9
5 mg/kgb
Cmax (µg/mL)
AUC (µg·hr/mL)
Halfꢀlife (hr)
0.0452
0.200
1.62
0.119 ± 0.077
NT
NT
NT
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
0.312 ± 0.100
3.01 ± 1.51
aThe results are shown as the mean ± SD (n = 3) or are presented as the average of two experiments. bCompounds were
administered orally to rats in solution with saline or water. NT: not tested.
Cardiovascular safety profiles and pharmacology. We compared the cardiovascular safety profile of 15, which has
highly selective β3ꢀAR agonistic activity and an excellent ADME profile, with the clinical drug 1. As shown in Table 6,
1 (i.v., 3 mg/kg) clearly increased HR and decreased mean blood pressure (MBP) in anesthetized rats. In the same
evaluation, compound 15 (i.v., 3 mg/kg) tended to increase HR and decrease MBP less than 1 as a preliminary
experiment.27 These data indicate that 15 possesses a desirable cardiovascular safety profile likely due to its minimal
offꢀtarget activities; it is highly active to the β3ꢀAR, but completely inactive to all other α1ꢀ and βꢀARs.
Next, we investigated the effect of 15 on the relaxation of urinary bladder smooth muscle strips from marmosets.10c
As shown in Figure 3, 15 fully relaxed isolated marmoset urinary bladder smooth muscle that was preꢀcontracted with
KCl, and its efficacy was similar to that of the nonꢀselective βꢀAR agonist isoproterenol.
Based on its excellent β3ꢀAR selectivity, oral availability, low risk of drug–drug interactions, and favorable relaxation
effect on marmoset urinary bladder smooth muscle, the optimized compound 15 was identified as a candidate drug for
the treatment of OAB without cardiovascular side effects.
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
Table 6. Effects of Intravenous Administration of 15 on HR and MBP in Anaesthetized Ratsa
6
7
8
9
HR
MBP
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
N
5
Increase (%)
1.8 ± 0.9
Decrease (%)
4.5 ± 0.8
ND
Increase (%)
1.0 ± 0.5
Decrease (%)
5.4 ± 1.9
Saline
1
3
10.0 ± 2.5
5.8 ± 0.4
3.5 ± 3.5
38.3 ± 3.4
12.9 ± 0.9
3
0.9 ± 0.4
2.0 ± 0.9
15
aCompounds were administered intravenously to rats (3 mg/kg) and HR and MBP recorded for 30 min. Maximum
increase or decrease of HR or MBP is shown as the percentage change relative to baseline. The results are shown as the
mean ± SEM. See reference 10c for further details of the experimental method. ND: Not detected.
Figure 3. Relaxation of marmoset urinary bladder smooth muscle by 15. Each point represents the mean of two
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
experiments. See reference 10c for further details of the experimental method.
4
5
6
7
8
9
Chemistry
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The synthesis of compounds 12ꢀ17 is shown in Scheme 1. Aminoalcohol 20 was prepared by coupling
(R)ꢀ2ꢀ(3ꢀnitrophenyl)oxirane 1828 with benzylamine 19 in 2ꢀPrOH. Protection of the hydroxyl of 20 with
triethylsilylchloride (TESCl) and imidazole afforded 21. Hydrogenolysis of 21 reduced the nitro group and deprotected
NꢀBn, and the alkyl amine was protected by using Boc2O, and then the OꢀBn group was deprotected under
hydrogenolysis conditions to afford alcohol 22. Mitsunobu reaction of 22 and 2310c using DMEAD29 and
triphenylphosphine proceeded to give the aniline 24. Coupling of a sulfonyl chloride with aniline 24 and subsequent
deprotection of Boc and TES groups afforded the corresponding target compounds 12ꢀ17.
Scheme 1. Synthesis of Target Compoundsa
ACS Paragon Plus Environment

Page 17 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
aReagents and conditions: (a) 2ꢀPrOH, reflux; (b) TESCl, imidazole, DMF, rt; (c) H2, 10% Pd/C, EtOH, 50 °C; (d)
Boc2O, THF, rt; (e) H2, 20% Pd(OH)2/C, THF, MeOH, 50 °C; (f) DMEAD, PPh3, toluene, rt; (g) R1ꢀSO2Cl (R1 = nꢀPr,
cyclopropyl, cyclobutyl, or phenyl), pyridine, CH2Cl2 or R1ꢀSO2Cl (R1 = iꢀPr or cyclopentyl), DBU, CH2Cl2, rt; (h) for
12ꢀ14, 16, and 17: 4 M HCl in dioxane, rt, for 1530: 1 M TBAF in THF, THF, rt then 4 M HCl in dioxane, rt.
Docking Study to Characterize β3ꢀAR Agonistic Activity and Selectivity over α1Aꢀ, β1ꢀ, and β2ꢀARs.
A docking study of the indazole series of the agonists was undertaken in order to understand the SAR underlying the
β3ꢀ and α1AꢀAR agonistic activities. No crystal structure of the β3ꢀAR is known; therefore, we constructed a homology
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
5
6
7
8
9
model of the β3ꢀAR. Previous studies showed that β3ꢀAR homology models, which were constructed using the crystal
structure of the β2ꢀAR, were useful for providing structural insight into ligand binding conformation, performing QSAR
analyses, and conducting in silico drug discovery.10a, 10d, 31 Recent resolutions of the β2ꢀAR active conformation32
allowed us to generate more accurate models of the β3ꢀAR with β3ꢀAR agonist using homology modeling and docking
simulation. We constructed an in silico homology model of the β3ꢀAR based on the β2ꢀAR active conformation (PDB
entry 3P0G). The β3ꢀAR sequence was aligned with the β1ꢀ, β2ꢀ, and α1AꢀARs using MOE 2015 software (Chemical
Computing Group, Montreal, Canada), and the alignment was refined manually to better match the position of amino
acids with the corresponding β2ꢀAR structure (Figure S1, Supporting Information). The sequence identity between β3ꢀ
and β2ꢀARs is 55.7% based on sequence alignment.31b
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
First, we constructed a homology model of the β3ꢀAR with compound 17, which is the most potent β3ꢀAR agonist in
the indazole series. Using this model, we performed a docking simulation with a structurallyꢀdiverse indazole series of
eight compounds, which included the previously reported compounds,10c in addition to another series that consisted of
KUCꢀ7322 (active form of compound 3),22 two analogues of 10,26b and isoproterenol. Previously, our group used a
similar strategy for a docking simulation to characterize the histamine H3 receptor binding mode.33 The correlation
between the calculated docking score and the ꢀpEC50 was determined. As shown in Figure 4, a strong correlation was
observed between docking score and ꢀpEC50 (R = 0.67 for all 12 compounds; R = 0.90 for the eight indazole
compounds). These results show that the homology model and docking pose of the ligand are in agreement with the
SAR. To further validate the docking pose of compound 15, which offers the best profile for potency, selectivity, and
ADME within our indazole series, a molecular dynamics (MD) simulation was performed. An enormous difference
between the pose suggested by docking and final ligand pose after a MD simulation has been reported in spite of a high
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
docking score.34 To examine temporal changes in compound 15 with the β3ꢀAR during 20 ns of the MD simulation, the
rootꢀmeanꢀsquare deviation (RMSD) was calculated for compound 15 and the β3ꢀAR throughout the simulation time
(Figure S2 in Supporting Information). The time series of the RMSD of compound 15 show that the binding mode is
overall stable. Accordingly, compound 15 with the β3ꢀAR after the MD refinement was further evaluated for its
interaction with and selectivity for the β3ꢀAR over α1Aꢀ, β1ꢀ, and β2ꢀARs.
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 proposed binding mode of compound 15 with the β3ꢀAR after the MD refinement is shown in Figure 5A. The
obtained mode suggested that compound 15 interacts with the three key conserved residues Asp117 (transmembrane
[TM] 3), Ser208 (TM5), and Asn332 (TM7), as observed in the crystal structure of the BIꢀ167107ꢀβ2ꢀAR complex
(PDB entry 3P0G). An additional hydrogen bonding was predicted between the indazole NH of compound 15 and
backbone of Cys196 (extracellular loop [ECL] 2); the indazole moiety participates in hydrophobic interactions with
residues Leu97 (TM2), Leu329 (TM7), and Trp333 (TM7) (Figure 5A and B). Moreover, hydrogen bonding also
occurred between the sulfonamide carbonyl oxygen and Asn312 (TM6); a Tꢀshaped πꢀπ interaction was predicted
between the phenyl ring and Phe309 (TM6); and the cyclobutane moiety occupied a hydrophobic pocket formed by
residues Val118 (TM3), Ile173 (TM4), and Tyr204 (TM5) (Figure 5A and B).
We next examined selectivity for the β3ꢀAR over α1AꢀAR based on the proposed β3ꢀAR binding model of compound
15 and the sequence alignment with the α1AꢀAR. Previously, we reported that the selectivity of different indazole
3ꢀmoieties for the β3ꢀAR over α1AꢀAR was ranked as follows: iꢀPr (8) > cyclobutyl (7) > ethyl (6) > ꢀMe (5) >
ꢀCH2OH.10c These results suggest that a hydrophobic substituent effectively induces improved selectivity for the β3ꢀAR
over α1AꢀAR. The indazole moiety participates in hydrophobic interactions with residues Leu97 (TM2), Leu329 (TM7),
and Trp333 (TM7) of the β3ꢀAR in the binding model. The residue Leu329 is not conserved in the α1AꢀAR; in α1AꢀAR,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
9
these residues are mutated to the hydrophilic residue Lys309. Therefore, the environment of this pocket in the α1AꢀAR
might be more hydrophilic than that in the β3ꢀAR. Thus, the SAR results on selectivity for the β3ꢀAR over α1AꢀAR are
supported by the results obtained from our docking model of the β3ꢀAR. On the other hand, when compared by their
selectivity for the β3ꢀAR over α1AꢀAR, the substituents at the sulfonamide moiety were ordered as follows: phenyl (17)
> cyclobutyl (15) >> nꢀpropyl (12) > methyl (5) (Table 2). This ordering suggests that a more bulky substituent
improves selectivity for the β3ꢀAR over α1AꢀAR. The cyclobutyl substituent displayed hydrophobic interactions with
Val118 (TM3), Ile173 (TM4), and Tyr204 (TM5) in β3ꢀAR homology modeling. The residue Ile173 is not conserved,
and is replaced with Leu162 in the α1AꢀAR. Moreover, Phe198, Ala199, and Ser200 of ECL2, which is located
proximal to the sulfone group of the sulfonamide in the β3ꢀAR, are not conserved in the α1AꢀAR; in the α1AꢀAR, these
residues correspond to Ile178, Asn179, and Glu180, respectively. Interactions of α1AꢀAR’s residues with the cyclobutyl
substituent and sulfone group may thus differ significantly from those of β3ꢀAR’s residues, based on differences in their
overall size, protonation potential, charge density, and overall hydrogen bondꢀforming ability with neighboring residues
or ligands. Consequently, these differences may account for the selectivity of β3ꢀAR over α1AꢀAR agonistic activity in
our indazole analogues.
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
We also considered the selectivity for the β3ꢀAR over β1ꢀ and β2ꢀARs based on the proposed β3ꢀAR binding model of
compound 15. As previously observed by Edmondon et al.,10a the Ala197 (ECL2) residue of β3ꢀAR is a critical
component for selectivity for activation of β3ꢀAR over binding to β1ꢀ and β2ꢀARs. The indazole moiety of compound 15
formed a hydrogenꢀbonding interaction with the backbone of Cys196, and the Ala197 residue is located proximal to the
indazole NH group (Figure 5A). The Ala197 replaces an Asp residue in β1ꢀ and β2ꢀARs (Asp217 in β1ꢀAR, Asp192 in
β2ꢀARs). The Asp residue in β1ꢀ and β2ꢀARs may lead to unfavorable interactions with this hydrogenꢀbonding with
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
respect to both electrostatic interactions and size. Therefore compound 15 is likely to show high selectivity for the
β3ꢀAR over β1ꢀ and β2ꢀARs.
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 4. Plot of docking score calculated by Glide extra precision (XP) based on the compound 17 β3ꢀAR model
versus experimental agonistic activity (ꢀpEC50) for eight indazole analogues (diamonds) and four other compounds
(circles). The coefficient of determination, R2, was 0.45 for the docking scores and ꢀpEC50 values of all 12 compounds.
See Supporting Information (Table S1) for individual ꢀpEC50 values and docking scores.
Figure 5. (A) Proposed model of compound 15 binding to the homology model of β3ꢀAR after MD refinement.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
5
6
7
8
9
Receptor residues within 4.0 Å of compound 15 are represented by lines. Compound 15 is shown as a ball and stick
model. All nonpolar hydrogen atoms of the receptor residues are omitted for clarity. Hydrogen bonding and salt bridges
to side chains of Asn117, Ser208, Asn312, and Asn332 are depicted by blue dots. Hydrogen bonding to the backbone of
Cys196 is also depicted by blue dots. (B) Schematic representation of the interactions between compound 15 and
β3ꢀAR residues using the Maestro Ligand Interaction Diagram module (Schrödinger, LLC, Portland, OR, USA).
Hydrogen bonding is indicated with dashed arrows, salt bridge with purple lines, cationꢀπ interactions with red lines,
and πꢀπ interactions with green lines.
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
Conclusion
We successfully developed the highly potent and selective β3ꢀAR agonist 15, which possesses a desirable metabolic
stability and PK profile. Compound 15 showed doseꢀdependent β3ꢀARꢀmediated responses in marmoset urinary bladder
smooth muscle and exhibited an excellent cardiovascular safety profile. These promising characteristics are likely due
to its minimal offꢀtarget effects: 15 is highly active towards only the β3ꢀAR, and completely inactive towards all other
αꢀ and βꢀARs. Furthermore, 15 is likely to have a lower risk of drug–drug interactions due to its low inhibition of CYPs.
Thus, compound 15 could be a drug candidate for the treatment of OAB without cardiovascular side effects. Detailed
pharmacology of 15 will be reported in a future pharmacology paper. We also developed a docking model of the β3ꢀAR
with 15 and related β3ꢀAR agonists, which identified the structural features required for highly selective binding to the
β3ꢀAR over α1Aꢀ, β1ꢀ, and β2ꢀARs, and which could therefore be useful for the design of other potent β3ꢀAR agonists.
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
Experimental Section
4
5
6
7
8
General Methods. All reagents and solvents were purchased from commercial sources and were used as received.
Anhydrous solvents were obtained from commercial sources. Thin layer chromatography (TLC) was carried out using
Merck GmbH precoated silica gel 60 F254 (Darmstadt, Germany). Chromatography on silica gel was carried out using
prepacked silica gel cartridges (Yamazen HiꢀFlash Column Silica gel, PurifpackꢀSi series or Biotage KPꢀSil series) and
the indicated solvent system (Yamazen Multi Prep YFLC, a Moritex Purifꢀα2 (50F) or Quad1 preparative isolation
system). 1H NMR spectra were recorded on either a JEOLALꢀ300 (300 MHz) or a Bruker BIOPIN AVANCE III HD
(400 MHz), with chemical shifts reported in δ values (ppm) relative to trimethylsilane. Electrospray ionization (ESI)
highꢀresolution mass spectra (HRMS) were recorded using a Thermo Scientific Orbitrap Velos Pro (Waltham, MA,
USA). Liquid chromatographyꢀmass spectrometry (LC/MS) data were recorded using a Waters single quadrupole type
mass spectrometer UPLC/SQD system (Milford, MA, USA) with an ESI method. All the final compounds were
purified to ≥ 95% purity, as determined by highꢀperformance liquid chromatography (HPLC). The HPLC method
conditions were as follows: YMC Meteoric Core C18 column (Kyoto, Japan), 2.7 ꢀm, 3.0 × 50 mm, 40 °C; solvent A
consisting of water (10 mM ammonium acetate); solvent B consisting of acetonitrile; 1.0 mL/min flow rate; linear
mobile phase gradient of 10ꢀ95% B (2 min), 95% B (2 min), 10% B (2 min); photodiode array detection (254 nm).
None of the final compounds were recognized as PAINS compounds.35 All animal experiments were approved by the
Committee on Ethics in Animal Experiments of Asahi Kasei Pharma Corporation.
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
(R)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)propaneꢀ1ꢀsulfonamide (12).
To a solution of 24 (95 mg, 0.15 mmol) and pyridine (75 ꢀL, 0.9 mmol) in anhydrous CH2Cl2 (1 mL) was added
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
9
propaneꢀ1ꢀsulfonyl chloride (86 mg, 0.6 mmol) at room temperature, and the solution was shaken (600 minꢀ1) overnight.
The mixture was purified by flash column chromatography (4:3 nꢀhexane/ethyl acetate). The sulfonamide product was
diluted with MTBE (100 ꢀL) and 4 M HCl in 1,4ꢀdioxane (1.5 mL, 6.0 mmol) was added at room temperature. The
mixture was shaken (600 minꢀ1) overnight and nitrogen gas was blown into the reaction solution to evaporate the
solvent. The residue was dried to give compound 12 (40.9 mg, 0.095 mmol, 63% yield) as dihydrochloride salt.
1HꢀNMR (400 MHz, DMSOꢀd6) δ 0.93 (3H, t, J = 7.6), 1.68 (2H, quin, J = 7.6), 2.44 (3H, s), 3.03ꢀ3.09 (1H, m), 3.07
(2H, t, J = 7.6), 3.23ꢀ3.27 (1H, m), 3.46 (2H, t, J = 5.1), 4.33ꢀ4.37 (2H, m), 4.99 (1H, dd, J = 2.2, 10.2), 6.21 (1H, brs),
6.77 (1H, dd, J = 1.8, 8.7), 6.90 (1H, d, J = 1.8), 7.11 (1H, d, J = 7.8), 7.15 (1H, dd, J = 1.3, 7.8), 7.30 (1H, s), 7.33 (1H,
t, J = 7.8), 7.60 (1H, d, J = 8.7), 8.97 (1H, brs), 9.22 (1H, brs), 9.88 (1H, s); 13CꢀNMR (100 MHz, DMSOꢀd6) δ 11.4,
12.5, 16.7, 45.9, 52.2, 53.6, 63.4, 67.9, 92.1, 111.6, 116.6, 117.1, 118.6, 120.9, 121.0, 129.3, 138.5, 140.7, 141.7, 143.0,
157.2; HRMS calculated for C21H28N4O4S + H+, 433.1904, found ESI: [M+H]+, 443.1897; LC/MSꢀESI (m/z): [M+H]+,
433; HPLC: purity 100%, RT 2.0 min.
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)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)propaneꢀ2ꢀsulfonamide (13).
Step 1. To a solution of 24 (96 mg, 0.15 mmol) and pyridine (18 ꢀL, 0.23 mmol) in anhydrous CH2Cl2 (2 mL) was
added propaneꢀ2ꢀsulfonyl chloride (26 mg, 0.18 mmol) at room temperature, and the solution was stirred overnight. To
the mixture was added DBU (134 ꢀL, 0.9 mmol) and propaneꢀ2ꢀsulfonyl chloride (20 ꢀL, 0.18 mmol) at room
temperature. The mixture was stirred overnight and DBU (33.5 ꢀL, 0.23 mmol) and propaneꢀ2ꢀsulfonyl chloride (20 ꢀL,
0.18 mmol) were added at room temperature. The mixture was stirred for 4 days and DBU (33.5 ꢀL, 0.23 mmol) and
propaneꢀ2ꢀsulfonyl chloride (30 ꢀL, 0.26 mmol) were added at room temperature. The mixture was stirred overnight
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and purified by flash column chromatography (81:19 to 60:40 nꢀhexane/ethyl acetate) to give a mixture of 24 and
sulfonamide product (41.5 mg). The mixture was diluted with CH2Cl2 (1.5 mL) and 1.7 mmol/g MPꢀIsocyanate (118
mg, 0.2 mmol) was added at room temperature. The mixture was stirred overnight, filtered, and concentrated to give
tertꢀbutyl
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)ꢀ6ꢀ(2ꢀ((tertꢀbutoxycarbonyl)(2ꢀ(3ꢀ((1ꢀmethylethyl)sulfonamido)phenyl)ꢀ2ꢀ((triethylsilyl)oxy)ethyl)amino)ethoxy)ꢀ3ꢀ
methylꢀ1Hꢀindazoleꢀ1ꢀcarboxylate (40 mg). LC/MSꢀESI (m/z): [M+H]+, 747.
Step 2. To tertꢀbutyl
(R)ꢀ6ꢀ(2ꢀ((tertꢀbutoxycarbonyl)(2ꢀ(3ꢀ((1ꢀmethylethyl)sulfonamido)phenyl)ꢀ2ꢀ((triethylsilyl)oxy)ethyl)amino)ethoxy)ꢀ3ꢀ
methylꢀ1Hꢀindazoleꢀ1ꢀcarboxylate (40 mg) was added 4 M HCl in 1,4ꢀdioxane (1.5 mL, 6.0 mmol) at room
temperature. The mixture was shaken (600 minꢀ1) overnight and nitrogen gas was blown into the reaction solution to
evaporate the solvent. The residue was dried to give compound 13 (28 mg, 0.055 mmol, 36% yield) as dihydrochloride
salt. 1HꢀNMR (400 MHz, DMSOꢀd6) δ 1.24 (6H, d, J = 6.8), 2.44 (3H, s), 3.02ꢀ3.09 (1H, m), 3.23 (1H, septet, J = 6.8),
3.24ꢀ3.26 (1H, m), 3.44ꢀ3.47 (2H, m), 4.31ꢀ4.40 (2H, m), 4.98 (1H, dd, J = 2.2, 10.3), 6.22 (1H, brs), 6.77 (1H, dd, J =
2.1, 8.7), 6.90 (1H, d, J = 2.1), 7.09 (1H, d, J = 7.7), 7.17 (1H, dd, J = 1.3, 8.1), 7.30ꢀ7.34 (2H, m), 7.60 (1H, d, J = 8.7),
8.95 (1H, brs), 9.20 (1H, brs), 9.85 (1H, s); 13CꢀNMR (100 MHz, DMSOꢀd6) δ 11.4, 16.0, 45.9, 51.2, 53.6, 63.4, 68.0,
92.1, 111.6, 116.5, 117.1, 118.5, 120.8, 120.9, 129.3, 138.8, 140.8, 141.6, 143.0, 157.0; HRMS calculated for
C21H28N4O4S + H+, 433.1904, found ESI: [M+H]+, 433.1895; LC/MSꢀESI (m/z): [M+H]+, 433; HPLC: purity 100%, RT
1.9 min.
(R)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)cyclopropanesulfonamide
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
5
6
7
8
9
(14). To a solution of 24 (96 mg, 0.15 mmol) and pyridine (18 ꢀL, 0.23 mmol) in anhydrous CH2Cl2 (2 mL) was added
cyclopropanesulfonyl chloride (25 mg, 0.18 mmol) at room temperature, and the solution was stirred overnight. To the
mixture was added pyridine (24 ꢀL, 0.3 mmol) and cyclopropanesulfonyl chloride (42 mg, 0.3 mmol) at room
temperature. The mixture was stirred overnight and purified by flash column chromatography (4/3 nꢀhexane/ethyl
acetate). The sulfonamide product was diluted with 1,4ꢀdioxane (0.2 mL) to which was added 4 M HCl in 1,4ꢀdioxane
(1.6 mL, 6.4 mmol) at room temperature. The mixture was shaken (600 minꢀ1) overnight and nitrogen gas was blown
into the reaction solution to evaporate the solvent. The residue was dried to give compound 14 (63 mg, 0.12 mmol, 83%
yield) as dihydrochloride salt. 1HꢀNMR (400 MHz, DMSOꢀd6) δ 0.90ꢀ0.97 (4H, m), 2.44 (3H, s), 2.58ꢀ2.64 (1H, m),
3.03ꢀ3.10 (1H, m), 3.23ꢀ3.28 (1H, m), 3.46 (2H, t, J = 5.0), 4.32ꢀ4.44 (2H, m), 5.01 (1H, dd, J = 2.2, 10.3), 6.25 (1H,
brs), 6.77 (1H, dd, J = 2.1, 8.8), 6.91 (1H, d, J = 2.1), 7.12 (1H, d, J = 7.8), 7.18 (1H, dd, J = 1.3, 7.8), 7.32ꢀ7.36 (2H,
m), 7.60 (1H, d, J = 8.8), 8.99 (1H, brs), 9.26 (1H, brs), 9.82 (1H, s); 13CꢀNMR (100 MHz, DMSOꢀd6) δ 4.9, 11.5, 29.5,
45.9, 53.6, 63.4, 67.9, 92.1, 111.5, 117.2, 117.4, 119.4, 120.8, 121.1, 129.1, 138.5, 140.8, 141.7, 142.9, 157.1; HRMS
calculated for C21H26N4O4S + H+, 431.1748, found ESI: [M+H]+, 431.1754; LC/MSꢀESI (m/z): [M+H]+, 431; HPLC:
purity 100%, RT 1.9 min.
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)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)cyclobutanesulfonamide
(15). Step 1. To a solution of 24 (14.95 g, 23.4 mmol) and pyridine (13.2 mL, 164 mmol) in anhydrous CH2Cl2 (100
mL) was added cyclobutanesulfonyl chloride (12.73 g, 81.9 mmol) at room temperature, and the solution was stirred
overnight. The mixture was diluted with ethyl acetate, washed twice with 0.5 M HCl, washed with brine, dried over
Mg2SO4, and concentrated in vacuo. The residue (28.72 g) was purified by flash column chromatography (73:27 to
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
52:48 nꢀhexane/ethyl acetate) to afford the sulfonamide product (11.14 g, 14.7 mmol, 63% yield). 1HꢀNMR (300 MHz,
CDCl3, 1:1 rotamers) δ 0.54 (6H, q, J = 7.8), 0.88 (9H, t, J = 7.8), 1.23ꢀ1.31 (2H, m), 1.47 (9H, s), 1.71 (9H, s),
1.88ꢀ2.02 (2H, m), 2.19ꢀ2.23 (2H, m), 2.47ꢀ2.60 (5H, m), 3.21ꢀ4.16 (5H, m), 4.88ꢀ4.93 and 5.02ꢀ5.06 (1H, each m),
6.03 and 6.44 (1H, each s), 6.86 (1H, d, J = 8.7), 7.09ꢀ7.12 (1H, m), 7.15ꢀ7.30 (2H, m), 7.47 (1H, dd, J = 1.6, 8.7), 7.54
(1H, s); LC/MSꢀESI (m/z): [M+H]+, 759.
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
Step 2. To a solution of sulfonamide (11.14 g, 14.7 mmol) in anhydrous THF (100 mL) was added 1 M TBAF in THF
(30 mL, 30 mmol) at room temperature, and the mixture was stirred for 2 h. The mixture was diluted with ethyl acetate
and the organic layer was washed with brine, water, and brine. The organic layer was dried over Na2SO4 and
concentrated in vacuo. The residue was purified by flash column chromatography (39:61 to 18:82 nꢀhexane/ethyl
acetate) to give the alcohol product (8.99 g, 13.9 mmol, 95% yield). LC/MSꢀESI (m/z): [M+H]+, 645.
Step 3. To a solution of the alcohol (8.99 g, 13.9 mmol) in anhydrous 1,4ꢀdioxane (13.9 mL) was added 4 M HCl in
1,4ꢀdioxane (139 mL, 556 mmol) at room temperature, and the mixture was stirred for 2 h. The sticky precipitate was
allowed to settle to the bottom of the flask, and the supernatant solution was decanted. The residue was diluted with
ethanol (150 mL) and water (15 mL). The solution was concentrated under reduced pressure. To the residue was added
ethanol, and the precipitate was collected by suction filtration to give compound 15 (5.15 g, 10 mmol, 72% yield) as
dihydrochloride salt. 1HꢀNMR (400 MHz, DMSOꢀd6) δ 1.81ꢀ1.93 (2H, m), 2.11ꢀ2.19 (2H, m), 2.27ꢀ2.36 (2H, m), 2.48
(3H, s), 3.02ꢀ3.09 (1H, m), 3.22ꢀ3.25 (1H, m), 3.46 (2H, t, J = 5.0), 3.86ꢀ3.94 (1H, quintet, J = 8.2), 4.34ꢀ4.43 (2H, m),
5.81 (1H, brs), 5.02 (1H, dd, J = 1.9, 10.2), 6.82 (1H, dd, J = 2.0, 8.8), 6.93 (1H, dd, J = 2.0), 7.11 (1H, d, J = 7.8), 7.14
(1H, d, J = 7.8), 7.29 (1H, s), 7.31 (1H, t, J = 7.8), 7.65 (1H, d, J = 8.8), 9.09 (1H, brs), 9.47 (1H, brs), 9.80 (1H, s);
13CꢀNMR (100 MHz, DMSOꢀd6) δ 11.2, 16.2, 23.2, 45.8, 53.1, 53.7, 63.4, 67.9, 92.1, 112.2, 116.8, 117.2, 119.1, 121.1,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
121.2, 129.1, 138.5, 140.7, 141.5, 142.9, 157.6; HRMS calculated for C22H28N4O4S + H+ 445.1904, found ESI: [M+H]+,
4
5
6
445.1895; LC/MSꢀESI (m/z): [M+H]+, 445; HPLC: purity 100%, RT 2.0 min.
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
(R)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)cyclopentanesulfonamide
(16). Step 1. To a solution of 24 (95 mg, 0.15 mmol) and DBU (134 µL, 0.9 mmol) in anhydrous CH2Cl2 (1 mL) was
added cyclopentanesulfonyl chloride (111 mg, 0.6 mmol) in CH2Cl2 (1 mL) at room temperature, and the mixture was
stirred for 2 days. Nitrogen gas was blown into the reaction solution to evaporate the solvent. The residue was diluted
with methanol (2 mL) and 5 M NaOH (200 µL, 1 mmol) was added at room temperature. The mixture was stirred for 3
days and poured into saturated aq NH4Cl. The aqueous layer was extracted twice with ethyl acetate. The organic layers
were washed twice with 2 M HCl, washed with brine, dried over Na2SO4, and concentrated in vacuo. The residue was
purified by flash column chromatography (47:53 to 20:80 nꢀhexane/ethyl acetate) and preparative TLC (10:1
CH3Cl/MeOH)
to
give
tertꢀbutyl
(R)ꢀ(2ꢀ(3ꢀ(cyclopentanesulfonamido)phenyl)ꢀ2ꢀhydroxyethyl)(2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)carbamate (35
mg, 0.063 mmol, 42% yield). LC/MSꢀESI (m/z): [M+H]+, 559.
Step 2. To the tertꢀbutyl
(R)ꢀ(2ꢀ(3ꢀ(cyclopentanesulfonamido)phenyl)ꢀ2ꢀhydroxyethyl)(2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)carbamate (35
mg, 0.063 mmol) was added 4 M HCl in 1,4ꢀdioxane (1.5 mL) at room temperature. The mixture was shaken (600
minꢀ1) overnight and nitrogen gas was blown into the reaction solution to evaporate the solvent. The residue was dried
to provide the product (34 mg, 0.064 mmol, quant.) as dihydrochloride salt. 1HꢀNMR (400 MHz, DMSOꢀd6) δ
1.48ꢀ1.57 (2H, m), 1.61ꢀ1.70 (2H, m), 1.79ꢀ1.95 (4H, m), 2.43 (3H, s), 3.02ꢀ3.09 (1H, m), 3.22ꢀ3.28 (1H, m), 3.46 (2H,
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
t, J = 4.7), 3.50ꢀ3.56 (1H, m), 4.32ꢀ4.40 (2H, m), 4.98 (1H, dd, J = 2.0, 10.1), 6.23 (1H, brs), 6.76 (1H, dd, J = 2.1, 8.7),
6.90 (1H, d, J = 2.1), 7.10 (1H, d, J = 7.8), 7.17 (1H, dd, J = 1.4, 7.8), 7.31 (1H, s), 7.33 (1H, t, J = 7.8), 7.59 (1H, d, J
= 8.7), 8.94 (1H, brs), 9.16 (1H, brs), 9.82 (1H, s); 13CꢀNMR (100 MHz, DMSOꢀd6) δ 11.5, 25.3, 27.2, 45.9, 53.6, 59.6,
63.4, 67.9, 92.1, 111.5, 116.9, 117.2, 118.9, 120.9, 121.0, 129.2, 138.7, 140.1, 141.6, 143.0, 157.1; HRMS calculated
for C23H30N4O4S + H+, 459.2061, found ESI: [M+H]+, 459.2068; LC/MSꢀESI (m/z): [M+H]+, 459; HPLC: purity 100%,
RT 2.0 min.
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)ꢀNꢀ(3ꢀ(1ꢀHydroxyꢀ2ꢀ((2ꢀ((3ꢀmethylꢀ1Hꢀindazolꢀ6ꢀyl)oxy)ethyl)amino)ethyl)phenyl)benzenesulfonamide (17).
To a solution of 24 (96 mg, 0.15 mmol) and pyridine (18 ꢀL, 0.23 mmol) in anhydrous CH2Cl2 (2 mL) was added
benzenesulfonyl chloride (32 mg, 0.18 mmol) at room temperature, and the solution was stirred overnight. The mixture
was purified by flash column chromatography (4:3 nꢀhexane/ethyl acetate). The sulfonamide product was diluted with
1,4ꢀdioxane (0.2 mL) and 4 M HCl in 1,4ꢀdioxane (1.6 mL, 6.4 mmol) was added at room temperature. The mixture
was shaken (600 minꢀ1) overnight and nitrogen gas was blown into the reaction solution to evaporate the solvent. The
residue was dried to provide the product (67 mg, 0.12 mmol, 83% yield) as dihydrochloride salt. 1HꢀNMR (400 MHz,
DMSOꢀd6) δ 2.44 (3H, s), 2.88ꢀ2.99 (1H, m), 3.14ꢀ3.17 (1H, m), 3.43 (2H, t, J = 4.9), 4.29ꢀ4.38 (2H, m), 4.92 (1H, dd,
J = 2.0, 10.4), 6.20 (1H, brs), 6.76 (1H, dd, J = 2.1, 8.8), 6.90 (1H, d, J = 2.1), 7.02 (1H, d, J = 8.0), 7.05 (1H, d, J =
8.0), 7.22 (1H, s), 7.23 (1H, t, J = 8.0), 7.52ꢀ7.56 (2H, m), 7.58ꢀ7.62 (2H, m), 7.77ꢀ7.79 (2H, m), 8.93 (1H, brs), 9.17
(1H, brs), 10.41 (1H, s); 13CꢀNMR (100 MHz, DMSOꢀd6) δ 11.4, 45.9, 53.5, 63.4, 67.8, 92.1, 111.6, 117.1, 117.2, 119.0,
120.9, 121.4, 126.6, 129.1, 132.8, 137.8, 139.4, 140.8, 141.6, 142.9, 157.2; HRMS calculated for C24H26N4O4S + H+,
467.1748, found ESI: [M+H]+, 467.1735; LC/MSꢀESI (m/z): [M+H]+, 467; HPLC: purity 95% RT 2.0 min.
ACS Paragon Plus Environment