Application of a Parallel Synthetic Strategy in the Discovery of Biaryl Acyl Sulfonamides as Efficient and Selective NaV1.7 Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.6b00425

The majority of potent and selective hNa_V1.7 inhibitors possess common pharmacophoric features that include a heteroaryl sulfonamide headgroup and a lipophilic aromatic tail group. Recently, reports of similar aromatic tail groups in combination with an acyl sulfonamide headgroup have emerged, with the acyl sulfonamide bestowing levels of selectivity over hNa_V1.5 comparable to the heteroaryl sulfonamide. Beginning with commercially available carboxylic acids that met selected pharmacophoric requirements in the lipophilic tail, a parallel synthetic approach was applied to rapidly generate the derived acyl sulfonamides. A biaryl acyl sulfonamide hit from this library was elaborated, optimizing for potency and selectivity with attention to physicochemical properties. The resulting novel leads are potent, ligand and lipophilic efficient, and selective over hNa_V1.5. Representative lead 36 demonstrates selectivity over other human Na_V isoforms and good pharmacokinetics in rodents. The biaryl acyl sulfonamides reported herein may also offer ADME advantages over known heteroaryl sulfonamide inhibitors.

Full text of DOI:10.1021/acs.jmedchem.6b00425

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Application of a Parallel Synthetic Strategy in the Discovery of
V
Biaryl Acyl Sulfonamides as Efficient and Selective Na1.7 Inhibitors
Erin F. DiMauro, Stephen Altmann, Loren M. Berry, Howard Bregman, Nagasree Chakka,
Margaret Chu-Moyer, Elma Feric Bojic, Robert S. Foti, Robert T. Fremeau, Jr., Hua Gao, Hakan
Gunaydin, Angel Guzman-Perez, Brian E. Hall, Hongbing Huang, Michael Jarosh, Thomas
Kornecook, Josie Lee, Joseph Ligutti, Dong Liu, Bryan D. Moyer, Daniel Ortuno, Paul E. Rose,
Laurie B Schenkel, Kristin Taborn, Jean Wang, Yan Wang, Violeta L. Yu, and Matthew M. Weiss
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00425 • Publication Date (Web): 21 Jul 2016
Downloaded from http://pubs.acs.org on July 21, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Application of a Parallel Synthetic Strategy in the Discovery of
Biaryl Acyl Sulfonamides as Efficient and Selective Na 1.7
V
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
Inhibitors
,
†
§,˧
‡
†
Erin F. DiMauro* , Stephen Altmann , Loren M. Berry , Howard Bregman , Nagasree
†
†
§,˧
‡
§,˧
Chakka , Margaret ChuꢀMoyer , Elma Feric Bojic , Robert S. Foti , Robert Fremeau , Hua
ǁ
ǁ,˧
†
±
†,˧
Gao , Hakan Gunaydin , Angel GuzmanꢀPerez , Brian E. Hall , Hongbing Huang , Michael
§
§
§
§
§
§
Jarosh , Thomas Kornecook , Josie Lee , Joseph Ligutti , Dong Liu , Bryan D. Moyer , Daniel
§
,˧
±
†
§
§
±
Ortuno , Paul E. Rose , Laurie B. Schenkel , Kristin Taborn , Jean Wang , Yan Wang , Violeta
§
†
Yu , and Matthew M. Weiss
†
ǁ
‡
Department of Medicinal Chemistry, Department of Molecular Engineering, Department of Pharmacokinetics and
§
±
Drug Metabolism, Department of Neuroscience, and Department of Biologics, Amgen Inc., 360 Binney Street,
Cambridge, Massachusetts 02142, and One Amgen Center Drive, Thousand Oaks, CA 91320, United States.
ABSTRACT
The majority of potent and selective hNa 1.7 inhibitors possess common pharmacophoric
V
features that include a heteroaryl sulfonamide headgroup and a lipophilic aromatic tail group.
Recently, reports of similar aromatic tail groups in combination with an acyl sulfonamide
headgroup have emerged, with the acyl sulfonamide bestowing levels of selectivity over hNa 1.5
V
comparable to the heteroaryl sulfonamide. Beginning with commercially available carboxylic
acids that met selected pharmacophoric requirements in the lipophilic tail, a parallel synthetic
approach was applied to rapidly generate the derived acyl sulfonamides. A biaryl acyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
sulfonamide hit from this library was elaborated, optimizing for potency and selectivity with
attention to physicochemical properties. The resulting novel leads are potent, ligand and
lipophilic efficient, and selective over hNa 1.5. Representative lead 36 demonstrates selectivity
V
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
over other human Na isoforms and good pharmacokinetics in rodents. The biaryl acyl
V
sulfonamides reported herein may also offer ADME advantages over known heteroaryl
sulfonamide inhibitors.
INTRODUCTION
Aberrant action potential firing of nociceptive neurons can lead to pain. The fast
upstroke of the action potential is produced by entry of sodium ions through voltageꢀgated
1
sodium channels (Na ). There are nine different isoforms of voltageꢀgated sodium channels
V
(
Na 1.1 – Na 1.9), which have distinct expression patterns in neurons, cardiac muscle and
V V
2
skeletal muscle. Nonꢀselective Na inhibitors such as lidocaine, mexiletine, and carbamazepine
V
show clinical efficacy in chronic pain, but are limited in dose and in use, likely due to effects on
3
other Na isoforms outside of the pain pathway. The tetrodotoxinꢀsensitive (TTXꢀS) voltageꢀ
V
gated sodium channel 1.7 (Na 1.7) encoded by the SCN9A gene, is required to sense pain in
V
4
humans and mice. Rare genetic forms of severe chronic pain, primary erythromelalgia (PE) and
paroxysmal extreme pain disorder (PEPD), result from mutations that increase the activity of
5
hNa 1.7. Conversely, the root cause of the genetic disorder congenital insensitivity to pain
V
6
(
CIP) is a loss of function mutation of hNa 1.7. These studies indicate that hNa 1.7 is critical
V
V
for human pain perception. Additionally, human and mouse genetics highlight a critical role of
Nav1.7 in itch. A human Na 1.7 gain of function mutation leads to both paroxysmal itch and
v
7
pain, whereas Na 1.7 knockout mice show no withdrawal responses to painful stimuli or
v
4
c
scratching responses to histamine. Accordingly, a small molecule therapeutic agent that
ACS Paragon Plus Environment

Page 3 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selectively inhibits hNa 1.7 should effectively treat pain or itch in humans. Several
V
contemporary reviews summarize progress in the quest for potent and selective small molecule
8
inhibitors of hNa 1.7. In recent years, we and others have focused on identifying novel and
V
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
advantageous leads from within the heteroaryl sulfonamide class of inhibitors that was originally
9
10
pioneered by Pfizer/Icagen, and later advanced by Genentech/Xenon, as these routinely
demonstrate high levels of potency and selectivity over the human cardiac sodium channel 1.5
(
hNa 1.5). In a complementary fashion, we embarked on a secondary pursuit to discover a novel
V
and structurally distinct class of inhibitors that could demonstrate similar potency and selectivity,
coupled with improved efficiency and favorable ADME properties.
RESULTS AND DISCUSSION
11
In 2012, the first acyl sulfonamide hNa 1.7 inhibitors were disclosed by Pfizer.
V
Representative acyl sulfonamides 1 and 2 were determined to inhibit hNav1.7 with selectivity
1
2
over hNa 1.5 in PatchꢀXpress (PX) electrophysiological assays and human etherꢀaꢀgoꢀgo
V
3
13
related gene (hERG) in a [ H]dofetilide displacement assay (Table 1). A comparison to 6,6ꢀ
1
4
fused heteroaryl sulfonamide leads developed at Amgen was made on the basis of potency,
1
5,16
16
selectivity, ligand efficiency (LE), lipophilic efficiency (LipE),
and pKa,
using
14
representative sulfonamide 3. Initial data for our heteroaryl sulfonamide series suggested the
potential for transporterꢀmediated clearance, specifically in regard to the organic anionꢀ
1
4a, 17
transporting polypeptide (OATP) transporters.
Under the hypothesis that the heteroaryl
sulfonamide headgroup was a recognition element for hepatic drug transporters such as OATP, it
was envisioned that the acyl sulfonamides, possessing an alternative pharmacophore and slightly
different pK , may be cleared by a different mechanism, and thereby offer a potentially
a
advantageous ADME profile. Furthermore, the accumulation of internal data on sulfonamides
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
had alerted us to a tendency towards undesirable increases in molecular weight and lipophilicity
in the quest for potency, leading to issues with selectivity over inhibition of CYP isoforms
(particularly 2C9) or induction of CYPs via PXR activation. With respect to efficiency, the acyl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sulfonamides 1 and 2 generally compared favorably to our early heteroaryl sulfonamide leads,
and a strategy that specifically focused on lipophilic efficiency from the outset of lead discovery
was envisioned to be advantageous for delivering metabolically stable leads with minimal safety
risks.
Table 1. Representative Pfizer acyl sulfonamide and Amgen heteroaryl sulfonamide hNa 1.7
V
a
inhibitors
Cmpd
hNa 1.7 IC (ꢀM)
1
2
3
b
c
0.74
>30
>30
0.076
11
0.59
27
V
50
hNa 1.5 IC (ꢀM)
V
50
hERG K (ꢀM)
>30
2.0
i
d
0
.35 / 4.83
0.39 / 5.84 0.28 / 4.57
LE / LipE
e
pKa
3.37
3.41
3.62
a
b
Reported data was collected at Amgen. hNa 1.7 data were collected using a PatchXpress automated electrophysiology
v
c
platform using a protocol where cells were held at a voltage yielding 20ꢀ50% channel inactivation. hNa 1.5 data were collected
v
d
on the same platform using a protocol where cells were held at ꢀ50 mV. LipE was calculated using hNa 1.7 PX IC and
V
50
e
Amgen Na 1.7 project cLogD(7.4), which was calculated using daylight cLogP and an Amgen pK calculator. Amgen pKa
V
a
ACS Paragon Plus Environment

Page 5 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
calculator was based on an empirical statistical in silico model derived using the Cubist statistical method and a pool of 820
molecular descriptors.
A conformational analysis and molecular overlay of inhibitors 1, 2 and 3 was performed
(Scheme 1). On the basis of this overlay, it was anticipated that the acquired wealth of 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
from our internal efforts on heteroaryl sulfonamides possessing 6,6ꢀfused aryl cores related to
14
3
, might be applied towards the rapid generation of novel acyl sulfonamide leads with high
efficiency. Thus, the acyl sulfonamide headgroup was grafted onto our established 6,6ꢀfused aryl
cores. While the resulting 6,6ꢀfused aryl analogs, such as isoquinoline 4, were only weakly
potent, the incorporation of an Oꢀlinker between the isoquinoline and phenyl rings afforded
significantly enhanced potency, as exemplified by 5. Further elaboration of the bottom aryl ring
with a focus on driving potency and LipE delivered moderately improved pyridyl leads, such as
6
, which exhibited reasonably good potency, selectivity over hNa 1.5, low intrinsic clearance in
V
liver microsomes (HLM, RLM CL : 17, < 14 ꢀL/min/mg), and high solubility (PBS pH 7.4: 486
int
ꢀM, FaSSIF pH 6.8: 462 ꢀM). Since heteroaryl sulfonamides have suffered from CYP
10
inhibition issues, we evaluated acyl sulfonamide lead 6 for inhibitory activity on hCYP3A4,
1
8
2
D6 and 2C9 isoforms in both standard and timeꢀdependent assays (Table 2). Compound 6
displayed significant inhibitory activity on CYP3A4 and 2C9, and was therefore not ideally
suited for further optimization. Furthermore, while leads 5 and 6 afforded LipE >5 and >40ꢀfold
selectivity over hNav1.5, the observation that smaller, more flexible ethers such as 1 had LipE =
5.8 and selectivity of 144ꢀfold, led us to speculate that the rigid 6,6ꢀcore might not be a
competitive starting point for lead optimization. Finally, we sought to identify a novel, but more
modular core system, wherein rapid synthetic chemistry would allow for SAR exploration in
divergent directions.
a
Scheme 1. Evolution of isoquinoline acyl sulfonamides
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cmpd
4
5
6
b
c
hNa 1.7 IC (ꢀM)
1.55
0.28
14.4
>30
0.20
8.7
>30
V
50
hNa 1.5 IC (ꢀM)
>30
V
50
hERG K (ꢀM)
>30
i
d
0.28 / 4.25
0.31 / 5.01 0.31 / 5.23
LE / LipE
a
Deprotonated species ꢀ 1 orange; 2 blue; 3 gray. Full conformational analyses were conducted with MMFF94, followed by a
b
rigid body alignment of the conformers with lowest energy in MOE.
hNav1.7 data were collected using a PatchXpress
automated electrophysiology platform using a protocol where cells were held at a voltage yielding 20ꢀ50% channel inactivation.
c
d
hNav1.5 data were collected on the same platform using a protocol where cells were held at ꢀ50 mV. LipE was calculated
using hNa 1.7 PX IC and Amgen Na 1.7 project cLogD(7.4), which was calculated using daylight cLogP and an Amgen pK
a
V
50
V
calculator.
Table 2. hCYP Inhibitory activity of 6
hCYP
3A4
30.7
2D6
>50
>50
2C9
9.4
a
IC (ꢀM)
50
b
1
8.4
11.1
Shifted IC (ꢀM)
5
0
a
b
Compound 6 was coꢀincubated with probe substrate of enzyme activity. Compound 6 was preꢀincubated with enzyme and
coꢀfactor for 30 minutes prior to addition of probe substrate to determine potential for timeꢀdependent inhibition.
ACS Paragon Plus Environment

Page 7 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Accordingly, an unbiased parallel synthetic strategy was adopted to identify leads with novel
and efficient cores, beginning with commercially available carboxylic acids possessing a
preferred 3,4ꢀdichlorophenyl ring connected via any type of linker to the carboxylate group. A
diverse library of acyl sulfonamides was prepared in parallel, using a single coupling reaction
followed by high throughput parallel purification and characterization (HTꢀLC/MS, HTꢀNMR).
Using this strategy, weakly potent biaryl lead 7a was identified. The majority of other products
from this library did not significantly inhibit hNav1.7 (< 5% at 5 ꢀM; See SI Table S1 for a
comprehensive table of products 7b-t). Lead 7a was rapidly elaborated to 8 and 9, both of
which showed improved potency and LipE, while maintaining high selectivity. The marked
substituent effects on the top aryl ring suggested that a further increase in potency could be
realized via exploration of this region of the scaffold. Furthermore, lead 9 was promising from
the standpoint of minimal observed CYP inhibition (Table 3) and the 2ꢀalkoxy pyridyl moiety
offered a convenient retrosynthetic disconnect for exploration of tail group SAR using parallel
SNAr 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
Scheme 2. Evolution of biaryl acyl sulfonamides
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cmpd
7a
8
9
a
hNa 1.7 IC (ꢀM)
10.7
2.54
>42
>30
0.53
>30
>30
V
50
b
hNa 1.5 IC (ꢀM)
>42
V
50
hERG Ki (ꢀM)
>30
c
LE / LipE
0.32 / 3.65
0.33 / 4.30 0.32 / 4.98
a
hNav1.7 data were collected using a PatchXpress automated electrophysiology platform using a protocol where cells were
b
held at a voltage yielding 20ꢀ50% channel inactivation. hNav1.5 data were collected on the same platform using a protocol
c
where cells were held at ꢀ50 mV. LipE was calculated using hNa 1.7 PX IC and Amgen Na 1.7 project cLogD(7.4), which
V
50
V
was calculated using daylight cLogP and an Amgen pK calculator.
a
Table 3. hCYP Inhibitory activity of 9
hCYP
3A4
>50
2D6
>50
>50
2C9
34.1
34.7
a
IC (ꢀM)
50
b
4
8.6
Shifted IC (ꢀM)
5
0
a
b
Compound 9 was coꢀincubated with probe substrate of enzyme activity. Compound 9 was preꢀincubated with enzyme and
coꢀfactor for 30 minutes prior to addition of probe substrate to determine potential for timeꢀdependent inhibition.
ACS Paragon Plus Environment

Page 9 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Having identified 9 as a promising lead with a novel and modular biaryl core, with a favorable
CYP inhibition profile relative to 6, we set out to optimize the substitution pattern on the top aryl
ring. Table 4 describes selected SAR in this portion of the molecule. Cꢀ2 was found to be
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
generally intolerant to substitution (R ), as substantial potency loss was observed with groups
1
1
larger than H and F (e.g., R =Me, 13 and R =Cl, 14). Cꢀ3 (A) was similarly intolerant to
substitution. A 2ꢀfold loss in potency and a significant loss in selectivity were observed with A
=
CF (15 vs. 11), and a 16ꢀfold potency loss was observed when CH was replaced with N (16 vs.
2
11). In contrast, Cꢀ5 was broadly tolerant to substitution (R ), with a variety of small lipophilic
groups or rings affording compounds of comparable potency (17ꢀ20). However, increased
lipophilicity or size at this position generally corresponded with a substantial decrease in human
microsomal stability. Introduction of some polarity at Cꢀ5 was well tolerated with ethers, such as
dihydrofuran 19 and methoxy 21, while also realizing high selectivity. However, some potency
loss was observed upon increasing polarity (e.g., nitrile 22), and larger, highly polar substituents
such as dimethylamide were not well tolerated (23). Ultimately, we opted to proceed with 5ꢀOMe
phenyl (21) as the near optimal top ring on the basis of its modest advantages with respect to
potency, lipophilic efficiency, and microsomal stability.
Table 4. Selected SAR of the top aryl ring
1
2
a
b
c
Cmpd R , R , A
hNa 1.7
hNa 1.5
HLM / RLM
LipE
V
V
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
IC50
ꢀM)
IC50
CLint
(
(ꢀM)
(ꢀL/min/mg)
9
F, F, CH
F, H, CH
H, F, CH
H, H, CH
Me, H, CH
Cl, H, CH
H, F, CF
H, F, N
0.53
3.04
0.66
1.19
> 42
> 42
1.24
10.6
0.29
0.27
0.26
> 30
18
47 / 22
4.98
4.36
4.66
4.57
ꢀꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
< 14 / 28
27 / 22
1
1
1
1
2
3
23
32
14 / < 14
22 / < 14
26 / < 14
22 / <14
18 / 18
N.D.
N.D.
5.2
14
ꢀꢀ
1
1
1
5
6
7
4.24
4.25
4.74
4.35
5.63
N.D.
9.5
H, Cl, CH
30 / 16.5
> 399 / < 14
> 399 / < 14
18
H, CF
H,
3
, CH
3.1
1
9
> 42
, CH
, CH
2
2
0
1
H, OCF
3
0.14
0.35
3.7
> 399 / < 14
44 / 32
4.84
5.39
4.81
ꢀꢀ
H, OMe, CH
H, CN, CH
> 30
24
22
1.62
20 / < 14
19 / < 14
2
3
H, CONMe
2
, CH > 42
N.D.
a
hNav1.7 data were collected using a PatchXpress automated electrophysiology platform using a protocol where cells were
b
held at a voltage yielding 20ꢀ50% channel inactivation. hNav1.5 data were collected on the same platform using a protocol
d
c
where cells were held at ꢀ50 mV. N.D. = not determined. LipE was calculated using hNa 1.7 PX IC and Amgen Na 1.7
V
50
V
project cLogD(7.4), which was calculated using daylight cLogP and an Amgen pK calculator.
a
With the top aryl ring fixed as 5ꢀmethoxy phenyl, a second parallel synthetic strategy was
applied to explore SAR at the bottom ring and linker. Taking advantage of the convenient
synthetic handle afforded by the chloroꢀpyridyl central ring, intermediate 24 (Table 5) was
rapidly generated on scale, using readily available building blocks and two synthetic steps.
ACS Paragon Plus Environment

Page 11 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Broad SAR explorations of both the linker (L = O, S, NR’) and bottom substituents (R) were
then conducted using the appropriate S Ar reactions, followed by high throughput parallel
N
purification and characterization with HTꢀLC/MS and HTꢀNMR. These libraries were designed
using virtual enumeration, with ranking and scoring of virtual targets based on calculated cLogD,
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
PSA, and predicted CL in LM. In this manner, a vast assortment of diverse compounds was
int
quickly generated and tested for biological activities. Selected SAR from this effort is shown in
Table 5.
A variety of bottom substituents (R) and linkers (L) were tolerated, with an ether linker in
combination with a ringꢀcontaining substituent affording the best balance of lipophilic efficiency,
selectivity and microsomal stability. Ethers containing 2ꢀ3 small lipophilic substituents (Cl, Me,
CF , OCF ) arranged in a variety of substitution patterns around a phenyl ring proved to be the
3
3
most potent compounds (e.g., 25). Bisꢀfluorinated phenyl rings such as 26 were moderately
potent and efficient. Substituted pyridyl methoxy ethers lent a boost in lipophilic efficiency with
a modest loss of potency (e.g., 27). Substituted benzylic ether leads with good potency,
selectivity, and efficiency were also identified (e.g., 28). Weakly potent leads containing small,
branched alkyl and cycloalkyl ethers were considered advantageous due to their relatively high
ligand and lipophilic efficiency (e.g., 29 and 30), though modest potency and/or high intrinsic
clearance in HLM (30) prevented these leads from advancing. Moderately potent compounds
containing phenyl thioethers (e.g., 31) and secondary amines (e.g., 32) were discovered, but
these generally suffered a loss in human microsomal stability. Leads emerging from this library
effort that met established criteria for potency (Nav1.7 IC < 0.075 ꢀM), selectivity (>100 fold
5
0
over hNav1.5), and RLM CL (< 30 ꢀL/min/mg) were advanced to rat IV PK studies. As such,
int
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compound 25 (CL = 0.56 L/h/kg) was deemed favorable and was thus used as a point of
reference for exploration of middle ring Cl replacement SAR.
Table 5. Selected SAR of the bottom substituent (R) and linker (L)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
d
Cmpd ꢀLR
hNa 1.7
hNa 1.5
Pred. HLM /
RLM CLint
HLM / RLM
CLint
LipE
5.26
V
V
IC50
IC50
c
(
ꢀM)
(ꢀM)
(ꢀL/min/mg)
(ꢀL/min/mg)
2
5
0.036
6.0
13 / 14
< 14 / < 14
2
2
2
6
7
8
0.100
0.150
0.109
23.7
11.6
21
15 / 27
15 / 16
14 / 14
< 14 / 33
5.99
6.12
5.36
< 14 / < 14
< 14 / < 14
2
3
9
0
0.20
0.55
9.3
18 / 15
95 / 14
< 14 / < 14
102 / 17
5.61
5.68
17.8
ACS Paragon Plus Environment

Page 13 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
3
1
2
0.16
8.9
2.5
205 / 15
393 / 33
199 / < 14
385 / 40
4.5
0.088
5.77
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
hNav1.7 data were collected using a PatchXpress automated electrophysiology platform using a protocol where cells were
b
held at a voltage yielding 20ꢀ50% channel inactivation. hNav1.5 data were collected on the same platform using a protocol
c
where cells were held at ꢀ50 mV. Empirical statistical in silico model derived using the Cubist statistical method and a pool of
d
820 molecular descriptors. LipE was calculated using hNa 1.7 PX IC and Amgen Na 1.7 project cLogD(7.4), which was
V
50
V
calculated using daylight cLogP and an Amgen pK calculator.
a
With the potent and relatively low clearance 3,5ꢀdichlorophenyl ether tail group in place,
replacement of the pyridyl ring Cl with CF was found to afford approximately equivalent
3
potency with a slight loss in selectivity and efficiency (33, Table 6). More polar substituents
such as CHF (34) and CN (35) were fairly well tolerated and afforded selective, efficient leads.
2
While compound 35 proved most efficient in this series, the PSA (118) and rat clearance (CL =
1
.8 L/h/kg) were high relative to the more potent lead 25 (PSA = 95; rat IV CL = 0.734 L/h/kg).
While both compounds were highly bound in rat plasma (rat PPB > 99.9 % bound), the clearance
was lower for the less polar lead 25. Since the chloroꢀpyridyl middle ring afforded a good
balance of favorable properties, it was retained for further optimization.
Table 6. Selected middle ring Cl replacement SAR
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
b
d
Cmpd ꢀR
hNa 1.7
hNa 1.5
Pred. HLM /
RLM CLint
HLM / RLM
CLint
LipE
V
V
IC50
IC50
c
(
ꢀM)
(ꢀM)
(ꢀL/min/mg) (ꢀL/min/mg)
2
3
5
3
Cl
0.036
0.028
6.0
13 / 14
14 / 14
< 14 / < 14
< 14 / < 14
5.26
4.97
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CF
2.35
3
34
CHF
2
0.054
0.077
18.8
16.1
18 / 15
28 / 15
< 14 / < 14
39 / 15
5.41
5.75
3
5
CN
a
hNav1.7 data were collected using a PatchXpress automated electrophysiology platform using a protocol where cells were
b
held at a voltage yielding 20ꢀ50% channel inactivation. hNav1.5 data were collected on the same platform using a protocol
c
where cells were held at ꢀ50 mV. Empirical statistical in silico model derived using the Cubist statistical method and a pool of
d
8
20 molecular descriptors. LipE was calculated using hNa 1.7 PX IC and Amgen Na 1.7 project cLogD(7.4), which was
V
50
V
calculated using daylight cLogP and an Amgen pK calculator.
a
We then returned to weaker (Na 1.7 IC > 0.075 ꢀM), efficient leads identified in the library
v
50
exploration described in Table 5 that incorporated relatively small bottom rings, as we imagined
that a boost in potency and/or intrinsic stability could be gained via reꢀintroduction of fluorine at
Cꢀ2 in the top ring. Indeed, this afforded improved potency, selectivity and LipE (36, AMꢀ
1
9
0358 vs. 26; 37 vs. 29; 38 vs. 30) as well as improved RLM CL for 36. Selected lipophilic
int
efficient leads, incorporating F at Cꢀ2, are represented in Table 7.
Table 7. Selected leads incorporating F at Cꢀ2
ACS Paragon Plus Environment

Page 15 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
d
Cmpd
ꢀOR
hNa 1.7
hNa 1.5
Pred. HLM /
RLM CLint
HLM / RLM LipE
CLint
LE
V
V
IC50
IC50
c
(
ꢀM)
(ꢀM)
(ꢀL/min/mg)
(ꢀL/min/mg)
3
3
6
7
0.069
6.9
20 / < 14
27 / < 14
6.37
0.31
0.34
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
0.062
11
93 / < 14
110 / < 14
6.35
6.60
3
8
0.110
18
124 / < 14
161 / < 14
0.35
a
hNav1.7 data were collected using a PatchXpress automated electrophysiology platform using a protocol where cells were
b
held at a voltage yielding 20ꢀ50% channel inactivation. hNav1.5 data were collected on the same platform using a protocol
c
where cells were held at ꢀ50 mV. Empirical statistical in silico model derived using the Cubist statistical method and
d
a pool of 820 molecular descriptors. Amgen Na 1.7 project cLogD(7.4) was calculated using daylight cLogP and an Amgen
V
pK calculator
.
a
As compounds 37 and 38 suffered from high CLint in HLM, they were not advanced. In
accordance with its low CLint in RLM, arylꢀether acyl sulfonamide 36 demonstrated lowꢀ
moderate clearance (CL = 0.39 L/h/kg), as well as high oral bioavailability (F > 100%) and high
plasma protein binding (99.6% bound) in rats (Table 8). Based on experimental data suggesting
that the reported acyl sulfonamides are substrates for breast cancer resistance protein (BCRP;
ABCG2), the high oral bioavailability observed in rats is most likely attributable to a
combination of low clearance, high passive permeability (Papp Avg. MDCK: 22.4 ꢁcm/s), high
solubility (PBS, FaSSIF: 398, 500 ꢁM), and presumed saturation of rBCRP (hPgp ER 1.4;
hBCRP ER 23.7; rBCRP ER 25.7) and/or OATP in the intestine or liver. Further supporting the
potential for transporter saturation is the expected intestinal concentration of approximately 454
ꢁ
M based on a molar dose of 5.1 ꢁmol (10 mg/kg, 0.25 kg rat, MW = 486.9) and an intestinal
20
volume of 11.3 mL. In general, in vitro ADME assays underꢀpredicted the clearance for these
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
21
compounds, as is often the case when transporterꢀmediated clearance pathways are suspected.
It is also possible that the observed volume of distribution for 36 is partially a function of
2
2
hepatobiliary uptake into hepatocyte, as is often the case for acidic compounds. In addition to
its favorable rat PK profile, 36 demonstrated no appreciable in vitro timeꢀdependent inhibition of
hCYP3A4, 2D6 and 2C9 at concentrations up to 45 ꢀM (Table 9).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 8. Rat PK of 36
Route
AUCinf
CL
T1/2
(h)
Vd
F (%)
Plasma
Protein
Binding
(%)
(ꢀMꢀh) (L/h/kg)
(L/kg)
(
mg/kg)
b
i.v. (1)
5.4 ±
.2
0.39 ±
1.84 ±
0.93 ±
ꢀꢀ
1
0.08
0.21
0.19
9
0.01
9.6 ±
c
p.o. (10)
377 ±
44.2
ꢀꢀ
ꢀꢀ
ꢀꢀ
>100
a
b
c
Male, Fed. Vehicle: 100% DMSO. Vehicle: 1%Tween 80, 2%HPMC, 97% water/NaOH(pH 8.5).
Table 9. hCYP Inhibitory activity of 36
hCYP
3A4
>50
2D6
>50
>50
2C9
45.3
49.6
a
IC (ꢀM)
50
b
>
50
Shifted IC (ꢀM)
50
a
b
Compound 36 was coꢀincubated with probe substrate of enzyme activity. Compound 36 was preꢀincubated with enzyme
and coꢀfactor for 30 minutes prior to addition of probe substrate to determine potential for timeꢀdependent inhibition.
While 36 was not a potent inhibitor of rat Na 1.7 heterologously expressed in HEK cells (IC
V
50
=
3.69 ꢀM), its potency on mouse Na 1.7 (IC = 0.115 ꢀM) was comparable to that observed on
V
50
the human channel, and it potently inhibited native TTXꢀS currents in mouse DRG neurons, with
IC = 0.27 ꢀM.
5
0
Consequently, the pharmacokineticꢀpharmacodynamic (PKꢀPD) behavior of 36 was evaluated in
a mouse histamineꢀinduced scratching model (Figure 1, A), which we have shown to be Na 1.7
V
ACS Paragon Plus Environment

Page 17 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4c
7
dependent. The importance of Na 1.7 in itch pathways has also been established in humans,
V
and we anticipate a potential utility of itch as a biomarker in human clinical trials.
dose of 300 mg/kg, 36 demonstrated robust reduction of histamineꢀinduced scratching bouts
75%, which was comparable to the antiꢀhistamine control, diphenhydramine, at 30 mg/kg) with
With an oral
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
(
unbound plasma concentrations (mPPB = 98.6%, Cu plasma = 4.15 ꢀM) ~15ꢀfold greater than
the IC as measured on native TTXꢀS currents in mouse DRG neurons. Notably, the same oral
5
0
dose and similar plasma exposure was not associated with any significant reductions in activity
in a separate openꢀfield study in naïve mice (Figure 1, B). At lower doses (30, 100 mg/kg), the
effect on histamineꢀinduced scratching bouts was not statistically significant. The steep doseꢀ
response and the requirement for relatively high target coverage (~15ꢀfold vs. mouse DRG IC )
5
0
in this model are in line with the PKꢀPD relationship spanning a substantial internal data set of
related compounds. Furthermore, this is consistent with published coverage multiples in rodent
1
0a
PD assays from an aryl sulfonamide compound which ranged from 10x to 330x.
Also
consistent with these findings, inhibition of a large fraction of Nav current is required to block
2
3
action potential firing in neurons.
Figure 1. Dose response of compound 36 in mice: (A) HistamineꢀInduced Scratching; (B) Open
a
Field Assessment (PK/PD)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Dose (mg/kg)
mean Plasma
Cu (ꢀM)
30
100
300
A. HIS
0.294 ± 0.099 1.27 ± 0.379
0.640 ± 0.250 2.39 ± 0.650
4.16 ± 1.58
3.72 ± 0.990
B. Open Field
ACS Paragon Plus Environment

Page 19 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
A: Mouse Histamine-Induced Scratching – Total number of histamineꢀinduced scratch bouts in animal cohorts orally
administered either vehicle (veh, n =12), diphenhydramine (DPH, 30 mg/kg, n = 8), or 30 (n = 9), 100 (n = 10), or 300 (n = 10)
mg/kg of 36. All treatment groups received histamine i.d. ***, p < 0.001 versus vehicle group (oneꢀway ANOVA followed by
Dunnett’s test). B: Mouse Open Field Assessment – Total basic movement in an openꢀfield arena recorded following oral
administration of either vehicle (veh, n = 10) or 36 at 30, 100, or 300 mg/kg (n = 10 for each dose).
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 36 was evaluated, via manual patch clamp electrophysiology, for inhibitory activity
against a panel of other human sodium channel isoforms, where it demonstrated marked
selectivity (Table 10). Furthermore, 36 was highly selective over other important antiꢀtargets
24
such as hERG (K > 30 ꢀM), hCYPs (Table 9), and hBSEP (IC50 = 13.3 ꢀM).
Notably,
i
selectivity over hCYP2C9 is substantially enhanced relative to reported heteroaryl sulfonamide
10
hNa 1.7 inhibitors.
Furthermore, no significant induction of hPXR was observed at
V
concentrations 200ꢀfold in excess of the hNa 1.7 IC (3.2 % at 2 ꢀM, 3.2 % at 10 ꢀM).
V
50
a
Table 10. Manual Eꢀphys – Inhibitory activity of 36 across the human Na family panel
V
hNa 1.x
1.7
1.1
8.5 ± 9.8 ± 12.5 1.1 ± 11.9 9.4 ± 15.6 ±
2.26 3.7 ± 3.9 0.32 ± 2.2 0.46 0.49
Values represent the mean of two experiments. Data were collected by manual patch clamp electrophysiology using a
1.2
1.3
1.4
1.5
1.6
1.8
V
IC (ꢀM)
0.051 ±
5
0
0
.012
a
protocol where cells were held at a voltage yielding 20% channel inactivation.
In order to measure useꢀdependence of channel inhibition, a representative set of compounds
was evaluated for inhibitory activity on hNav1.7 and hNav1.5 using an IonWorks Quattro (IWQ)
automated electrophysiology system. The protocol allowed for IC measurements at pulse 1
5
0
(
tonic block) and pulse 26 (useꢀdependent block) (See SI, Table S4). In general, the biaryl acyl
sulfonamides are not useꢀdependent inhibitors of these channels. For example, compound 36:
Nav1.7 IWQ IC tonic = 0.044 ꢀM, IC use = 0.042 ꢀM; Nav1.5 IWQ IC tonic = 13.6 ꢀM,
5
0
50
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
IC use = 11.4 ꢀM). In order to measure stateꢀdependence, compound 36 was evaluated on
5
0
hNav1.7 using manual patch clamp electrophysiology. In this experiment, we measured
inhibitory activity on the channels in the resting/closed state (Vꢀhold = ꢀ140 mV) and a partially
inactivated state (initial Vꢀhold = ꢀ140 mV, then switch to a voltage that yielded ~ 20% channel
inactivation). The data set indicates that the inhibition of hNav1.7 by 36 is stateꢀdependent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
mean IC on resting/closed channels >30 ꢀM (n = 3 cells); mean IC on ~20% inactivated
5
0
50
channels = 0.051 ꢀM (n = 2 cells)).
In a hNav1.7 radioligand binding assay using a tritiated ligand that was derived from our
2
5
internal optimization of 6,6ꢀfused aryl heteroaryl sulfonamide leads such as 3, acyl sulfonamide
3
6 potently displaced the labeled heteroaryl sulfonamide with IC = 0.017 ꢀM. This data, in
5
0
combination with parallel SAR from the two series, including markedly decreased potency on rat
Na 1.7, suggests that our heteroaryl sulfonamides and acyl sulfonamides occupy the same
V
binding pocket in Na 1.7, as initially assumed. 36 was docked into the known binding pocket of
V
GXꢀ936 in the Na 1.7 VSD4 region, using the recently disclosed coꢀcrystal structure (Figure
V
2
6
2). The presumed binding mode of 36 is in agreement with the observed SAR and biological
activity. With this docked model, we hypothesize that the SO group of 36 aligns with the
2
thiadiazole group of GXꢀ936, to interact with R1608, while the CO group of 36 aligns with the
SO group of GXꢀ936 to interact with R1602 and R1605. The model also suggests that the top
2
ring Cꢀ5 methoxy substituent of 36 aligns with the CN substituent of GXꢀ936, projecting towards
polar D1586, thus explaining tolerance for small polar substituents at this position (e.g., OMe,
CN). In contrast, the top ring Cꢀ3 of 36 faces a lipophilic wall, consistent with the observed
intolerance for substitution larger than fluorine or replacement with pyridyl (N). The docked
model aligns the pyridyl ring Cꢀ3 Cl with the positively charged azetidine ring of GXꢀ936,
ACS Paragon Plus Environment

Page 21 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
projecting towards polar residue E1534, supporting the observed tolerance for small polar
substituents (e.g., CHF , CN) at this position. As was observed with GXꢀ936, the model predicts
2
that the top ring of 36 forms a πꢀstacking interaction with Y1537, while the terminal 2,5ꢀ
difluorophenyl ring forms a unique πꢀstacking interaction with W1538. Based on sequence
alignment between Na 1.7 and Na 1.5, the residues in the positions of Y1537 and W1538 in
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
V
V
hNav1.5 are alanine and lysine, respectively. According to this model, the inability of biaryl
acyl sulfonamides to form πꢀstacking interactions with these residues in Na 1.5 likely
V
contributes to the observed selectivity.
a
Figure 2. 36 docked into VSD4 of chimeric Na 1.7
V
A.
B.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
All displayed AA residues are within 4.5 Å of compound 36 in the binding surface model. A. Docked conformation of 36
(
blue) superimposed onto the experimental coordinates of the GXꢀ936 (orange)ꢀVSD4ꢀNa Ab crystal structure as viewed from
V
two angles. B. 36ꢀprotein interaction map. Orange: πꢀstacking or hydrophobic interaction; Green: polar or chargeꢀcharge
interaction.
CHEMISTRY
Scheme 3 describes the synthesis of isoquinoline thiadiazole sulfonamide 3 and isoquinoline
acyl sulfonamides 4-6. Palladium catalyzed coupling of 6ꢀbromoꢀ1ꢀchloroisoquinoline 39 and
phenylmethanethiol afforded benzylmercaptan 40. 40 was converted to 41 via oxidation with
1,3ꢀdichloroꢀ5,5ꢀdimethylimidazolidineꢀ2,4ꢀdione followed by condensation with 2,3,4,5,6ꢀ
pentafluorophenol. Reaction with Nꢀ(2,4ꢀdimethoxybenzyl)ꢀ1,2,4ꢀthiadiazolꢀ5ꢀamine in the
presence of LiHMDS afforded 42, which was converted to 3 via Suzuki coupling to (3ꢀchloroꢀ4ꢀ
(trifluoromethyl)phenyl)boronic acid. 1ꢀChloroisoquinolineꢀ6ꢀcarboxylic acid 43 was subjected
to Suzuki coupling with (3ꢀchloroꢀ4ꢀ(trifluoromethyl)phenyl)boronic acid to afford 44, which
was then condensed with methanesulfonamide in the presence of EDC to afford 4. Alternatively,
EDCꢀpromoted coupling of methanesulfonamide and 43 afforded 45. S Ar of 45 with 4ꢀchloroꢀ
N
3
ꢀ(trifluoromethyl)phenol or 48 afforded 5 and 6, respectively. Phenol 48 was derived from
ACS Paragon Plus Environment

Page 23 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
bromide 46 via S Ar with 2ꢀmethylpropanꢀ1ꢀol to give intermediate 47, followed by borylation
N
and oxidation.
a
Scheme 3. Preparation of Compounds 3, 4, 5, and 6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: (a) Xantphos, Pd (dba) , DIPEA, phenylmethanethiol, 1,4ꢀdioxane (63%); (b) i. 1,3ꢀdichloroꢀ5,5ꢀ
2
3
dimethylimidazolidineꢀ2,4ꢀdione, CH CN, water, AcOH, 0 °C; ii. 2,3,4,5,6ꢀpentafluorophenol, CH CN (74%); (c) Nꢀ(2,4ꢀ
3
3
dimethoxybenzyl)ꢀ1,2,4ꢀthiadiazolꢀ5ꢀamine, LiHMDS, THF (83%); (d) (3ꢀchloroꢀ4ꢀ(trifluoromethyl)phenyl)boronic acid,
Pd(PPh ) , K CO , 1,4ꢀdioxane/water, 100 °C (14ꢀ66%); (e) NH SO Me, EDCꢀHCl, DMAP, DCM (47ꢀ60%); (f) 4ꢀchloroꢀ3ꢀ
3
4
2
3
2
2
(trifluoromethyl)phenol (for 5) or 48 (for 6) or 2ꢀmethylpropanꢀ1ꢀol (for 47), Cs CO , DMSO, 90 °C (37ꢀ86%); (g)
2
3
.
bis(pinacolato)diboron, PdCl (dppf) DCM, KOAc, 1,4ꢀdioxane, 90 °C (47%); (h) oxone, acetone/water, 0 °C (65%).
2
Scheme 4 demonstrates synthetic routes to biaryl acyl sulfonamides 8-9, 11-18, 20-33 and 36-
38. 4ꢀBromoꢀ2ꢀfluoroꢀ5ꢀmethoxybenzoic acid 51a was prepared from 49, following a sequence
of saponification and bromination to provide intermediate 50, followed by perꢀmethylation and
saponification. 51b-p were obtained from commercial suppliers. Aryl bromides 52a-p were
prepared from EDCꢀpromoted coupling of carboxylic acids 51a-p. Suzuki coupling of 52a-p
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with either commercial boronates or 47 afforded final compounds 8-9 and 11-18, and
intermediates 24 and 53-54.
Fluoride 24, prepared from 4ꢀbromoꢀ3ꢀmethoxyꢀNꢀ
(
methylsulfonyl)benzamide (52p), was reacted with alcohols under parallel S Ar conditions to
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
afford etherꢀlinked final compounds 25-30. In a similar fashion, 33 was prepared from fluoride
53 and 36ꢀ38 were prepared from fluoride 54. Similarly, baseꢀpromoted S Ar reactions of thiols
N
or amines with fluoride 24 afforded compounds such as 31 and 32.
a
Scheme 4. Preparation of Compounds 8-9, 11-18, 20-33 and 36-38
a
Reagents and conditions: (a) LiOH, water/THF, RT (91ꢀ94%); (b) Br , AcOH, DCM, RT (47%); (c) Me SO , K CO ,
2
2
4
2
3
CH CN, 80 °C (38%); (d) NH SO Me, EDCꢀHCl, DMAP, DCM, RT (81ꢀ92%); (e) aryl boronate, Pd(PPh ) , K CO (or
3
2
2
3 4
2
3
Na CO ), 1,4ꢀdioxane/water, 60 °C (or reflux) (37ꢀ91%); (f) alcohol, Cs CO , DMSO, 90 ꢀ 120 °C (18ꢀ79%); (g) 3ꢀ
2
3
2
3
ACS Paragon Plus Environment

Page 25 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
methylbenzenethiol, Cs CO , DMSO, 100 °C (26%); (h) i. 2ꢀisopropylpyrrolidine, NEt , DMSO, 90 °C (68%); ii. SFC with
2
3
3
ChiralpakꢀIC.
The synthesis of compounds 10 and 19 is described in Scheme 5. Carboxylic acids 55 and 57
were converted into the corresponding acyl sulfonamides, which were then coupled under Suzuki
conditions to 5ꢀbromoꢀ3ꢀchloroꢀ2ꢀisobutoxypyridine or 3ꢀchloroꢀ2ꢀisobutoxypyridineꢀ5ꢀboronic
acid, respectively, affording compound 10 and intermediate 59. Suzuki coupling of bromide 59
to 2ꢀ(2,5ꢀdihydrofuranꢀ3ꢀyl)ꢀ4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaborolane afforded compound 19.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 5. Preparation of Compounds 10 and 19
a
Reagents and conditions:
(b) NH SO Me, EDCꢀHCl, DMAP, DCM, RT (34ꢀ43%); (b) 5ꢀbromoꢀ3ꢀchloroꢀ2ꢀ
2 2
isobutoxypyridine, Pd(PPh ) , K CO , 1,4ꢀdioxane/water, 60 °C (44%); (c) 3ꢀchloroꢀ2ꢀisobutoxypyridineꢀ5ꢀboronic acid,
3
4
2
3
.
PdCl (dppf) DCM, Pd (dba) K CO , 1,4ꢀdioxane/water, tꢀBuOH, 50 °C (49%); (d) 2ꢀ(2,5ꢀdihydrofuranꢀ3ꢀyl)ꢀ4,4,5,5ꢀ
2
2
3,
2
3
.
tetramethylꢀ1,3,2ꢀdioxaborolane, PdCl (dppf) DCM, Pd (dba) K CO , 1,4ꢀdioxane/water, tꢀBuOH, 95 °C (12%).
2
2
3,
2
3
The synthesis of compounds 34 and 35 is described in Scheme 6. S Ar of fluoroꢀpyridyl
N
boronic ester 60 with 3,5ꢀdichlorophenol afforded boronate 61, which was coupled to 52p to
generate aldehyde 62. Reaction of 62 with DAST afforded difluoromethyl compound 34.
Boronate 64 was prepared from bromide 63 and coupled to 52p to generate aniline 65.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Sandmeyer reaction with CuBr afforded bromide 66, which was converted to nitrile 67 by
microwaveꢀpromoted palladiumꢀcatalyzed coupling to Zn(CN) . S Ar reaction with 3,5ꢀ
2
N
dichlorophenol afforded final compound 35.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 6. Preparation of Compounds 34 and 35
a
Reagents and conditions: (a) 3,5ꢀdichlorophenol, Cs CO , DMF, 50 °C (26%); (b) 52p, Pd(PPh ) , K CO (or Na CO ), 1,4ꢀ
2
3
3 4
2
3
2
3
.
dioxane/water 90 °C (42ꢀ46%); (c) DAST, DCM 50 °C (8%); (d) bis(pinacolato)diboron, PdCl (dppf) DCM, KOAc, 1,4ꢀdioxane,
2
9
0 °C (84%); (e) isoamylnitrite, CuBr, CH CN (40%); (f) Zn(CN) , Pd (dba) XantPhos, DMF, 100 °C, ꢀW (63%); (g) 3,5ꢀ
3 2 2 3,
dichlorophenol, Cs CO , DMSO, 90 °C (80%).
2
3
CONCLUSION
Parallel synthetic strategies, with focus on molecular weight and lipophilicity, were applied in
the rapid generation of potent, selective, and efficient leads in a novel class of biaryl acyl
sulfonamide Na 1.7 inhibitors. Representative lead 36 demonstrated marked selectivity against
V
seven other hNaV isoforms, hERG, hCYPs, hPXR, and hBSEP.
Its good rodent
pharmacokinetics and robust pharmacodynamic response in a mouse histamineꢀinduced
ACS Paragon Plus Environment

Page 27 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scratching model (without associated locomotor side effects) are typical of this series of
compounds. As such, leads from this series are anticipated to be quality tools for probing the
target coverage requirements for efficacy in various preclinical rodent models of acute and
chronic pain. Furthermore, compound 36 and related leads described herein compare favorably
to reported heteroaryl sulfonamides with respect to their lack of CYP inhibition. While the
current leads in this biaryl acyl sulfonamides series would benefit from further gains in potency
without compromising PK, thus affording lower efficacious doses, the low probability of drugꢀ
drug interactions with future derived compounds is noteworthy. Moreover, the modular nature of
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 scaffold and modeling in the Na 1.7 VSD4 selectivity pocket should enable rapid lead
V
optimization. Related efforts will be reported in due course.
EXPERIMENTAL SECTION
Unless otherwise noted, all materials were obtained from commercial suppliers and used without
further purification. Anhydrous solvents were obtained from Aldrich and used directly. All
reactions involving airꢀ or moisture–sensitive reagents were performed under a nitrogen or argon
atmosphere. Microwaveꢀassisted reactions were conducted with a Biotage Initiator. Purifications
were performed using standard column chromatography in glass columns or medium pressure
liquid chromatography (MPLC) on a CombiFlash Companion (Teledyne Isco) or a Biotage
Isolera with prepacked RediSep or Biotage normalꢀphase silica gel (35−60 ꢁm) columns and UV
detection at 254 nm. Preparative reversedꢀphase highꢀperformance liquid chromatography
(HPLC) and high throughput parallel purification were performed with reversed phase
preparative LC/MS: Waters autoꢀpurification system; liquid transfer system: Tecan; drying
system: Genevac; prep. Column: Xbridge (19 x 100 mm, C18, 10 ꢀm); flow rate: 40 mL/min;
general gradient: 5 – 95% B, 0.1% additive in both A and B; 10 min gradient; Mobile Phase A:
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 79
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
H O, Mobile Phase B, CH CN; Additive: TFA or NH OH. Chiral method development was
2
3
4
performed on an analytical Thar SFC/MS. Preparative chiral separations were performed on Thar
SFC Prep 80 or SFC Prep 350 instruments. All final compounds were purified to ≥ 95% purity as
determined by HPLC. Purity (3 min methods only) and reaction analyses were measured using
Agilent 1100 Series HPLC systems with UV detection at 254 nm and 215 nm (System A:
Agilent Zorbax SBꢀC18 3.0 x 50 mm, 3.5 micron, 5 to 95% CH CN in H O with 0.1% TFA for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
2
3
.6 min at 1.5 mL/min or Halo PhenylꢀHexyl, 3 x 50 mm, 2.7 micron, 5 to 95% CH CN in H O
3 2
with 0.1% TFA for 1.01 min at 2.0 mL/min; System B: Waters Xbridge C18, 3 x 50 mm, 3.5
1
micron, 5 to 95% CH CN in H O with 0.1% formic acid for 3.6 min at 1.5 mL/min. H NMR
3
2
spectra were recorded at ambient temperature on a Bruker AVꢀ400 (400 MHz) spectrometer or a
Varian 400 MHz spectrometer or a Bruker Advance III spectrometer, operating at a proton
frequency of 500.34 MHz using a Protasis CapNMR flowꢀprobe, equipped with Discovery
Tower™ Sample management system and a Waters Liquid Handler, made by CTC, Switzerland
(Model 2777). All observed protons are reported as parts per million (ppm) downfield from
tetramethylsilane (TMS) or other internal reference in the appropriate solvent indicated. Data are
reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet,
br = broad, m = multiplet), coupling constants, and number of protons. Lowꢀresolution mass
spectral (MS) data were determined on an Agilent 1100 Series LCMS with UV detection at 254
nm and a low resonance electrospray mode (ESI). Exact mass confirmation was performed on an
Agilent 1200 series high performance liquid chromatography (HPLC) system (Santa Clara, S2
CA, U.S.) by flow injection analysis, eluting with a binary solvent system A and B (A, water
with 0.1% formic acid; B, CH CN with 0.1% formic acid) under gradient conditions (5ꢀ95% B
3
ACS Paragon Plus Environment

Page 29 of 79
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
over 3 min) at 0.3 mL/min with MS detection by an Agilent 6510ꢀQꢀTOF mass spectrometer
(
Santa Clara, CA, U.S.).
-(3-Chloro-4-(trifluoromethyl)phenyl)-N-1,2,4-thiadiazol-5-yl-6-isoquinolinesulfonamide
3). Step 1A. To a suspension of 1,2,4ꢀthiadiazolꢀ5ꢀamine (3.35 g, 33.1 mmol) and 2,4ꢀ
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
(
dimethoxybenzaldehyde (5 g, 30.1 mmol) in DCM (150 mL) was added chlorotitanium
triisopropoxide (19.60 g, 75 mmol) in DCM (10 mL) portion wise over 5 min. After stirring for
15 min, sodium triacetoxyborohydride (31.9 g, 150 mmol) was added portionwise and the
reaction mixture was stirred for 1 h and quenched with saturated aq. NaHCO to pH~7, added 50
3
mL of water to aid layer separation. The mixture was then extracted with DCM (3 x 100 mL).
The combined organics were dried over MgSO , filtered, and concentrated to afford a yellow
4
residue, which was purified by column chromatography using isocratic Hex:EtOAc (50:50) to
obtain Nꢀ(2,4ꢀdimethoxybenzyl)ꢀ1,2,4ꢀthiadiazolꢀ5ꢀamine (3 g, 11.94 mmol, 39.7 % yield) as an
offꢀwhite solid. Step 1B. A 3 neck 2L flask equipped with overhead stirrer, thermocouple and
nitrogen inlet was charged with 6ꢀbromoꢀ1ꢀchloroisoquinoline 39 (60 g, 247 mmol), Xantphos
(7.16 g, 12.37 mmol) and Pd (dba) (5.66 g, 6.19 mmol). To reaction flask was charged with 1,4ꢀ
2
3
dioxane (540 mL) and DIPEA (64.8 mL, 371 mmol). The flask was purged with N and heated
2
to 63 °C upon which a solution of phenylmethanethiol (30.5 mL, 260 mmol) in 180 mL of
sparged 1,4ꢀdioxane was charged to the reaction mixture dropwise over 1 h. The reaction
mixture was filtered and concentrated to obtain an oil. Upon charging 500 mL isopropanol to the
oil, a crystalline slurry was formed. This was stirred at RT for 2 h, cooled to 0 °C and filtered.
The solids were washed with 50% IPA/heptane, then dried under vacuum with a nitrogen sweep
to obtain 6ꢀ(benzylthio)ꢀ1ꢀchloroisoquinoline 40 (44.3 g, 63% yield) as a yellow solid. MS (ESI,
+
positive ion) m/z: 286.0 [M+1] . Step 2B: A 3 neck 2000 mL flask equipped with overhead
ACS Paragon Plus Environment