Identification of Selective Acyl Sulfonamide-Cycloalkylether Inhibitors of the Voltage-Gated Sodium Channel (NaV) 1.7 with Potent Analgesic Activity

Article abstract of DOI:10.1021/acs.jmedchem.8b01621

Herein, we report the discovery and optimization of a series of orally bioavailable acyl sulfonamide Na_V1.7 inhibitors that are selective for Na_V1.7 over Na_V1.5 and highly efficacious in in vivo models of pain and hNa_V1.7 target engagement. An analysis of the physicochemical properties of literature Na_V1.7 inhibitors suggested that acyl sulfonamides with high f_sp3 could overcome some of the pharmacokinetic (PK) and efficacy challenges seen with existing series. Parallel library syntheses lead to the identification of analogue 7, which exhibited moderate potency against Na_V1.7 and an acceptable PK profile in rodents, but relatively poor stability in human liver microsomes. Further, design strategy then focused on the optimization of potency against hNa_V1.7 and improvement of human metabolic stability, utilizing induced fit docking in our previously disclosed X-ray cocrystal of the Na_V1.7 voltage sensing domain. These investigations culminated in the discovery of tool compound 33, one of the most potent and efficacious Na_V1.7 inhibitors reported to date.

Full text of DOI:10.1021/acs.jmedchem.8b01621

Subscriber access provided by University of Winnipeg Library
Article
Identification of selective acyl sulfonamide-cycloalkylether inhibitors of
the voltage gated sodium channel (NaV) 1.7 with potent analgesic activity
Shaoyi Sun, Qi Jia, Alla Y Zenova, Michael S Wilson, Sultan Chowdhury, Thilo Focken, Jun Li, Shannon
Decker, Michael E Grimwood, Jean-Christophe Andrez, Ivan Hemeon, Tao Sheng, Chien-An Chen, Andy
White, David H. Hackos, Lunbin Deng, Girish Bankar, Kuldip Khakh, Elaine Chang, Rainbow Kwan, Sophia
Lin, Karen Nelkenbrecher, Benjamin D. Sellers, Antonio G DiPasquale, Jae Chang, Jodie Pang, Luis Sojo,
Andrea Lindgren, Matthew Waldbrook, Zhiwei Xie, Clint Young, James P Johnson, C. Lee Robinette, Charles
J Cohen, Brian Salvatore Safina, Daniel P. Sutherlin, Daniel Fred Ortwine, and Christoph M Dehnhardt
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01621 • Publication Date (Web): 30 Nov 2018
Downloaded from http://pubs.acs.org on December 1, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 74
Journal of Medicinal Chemistry
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 74
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
Identification of selective acyl sulfonamide-
cycloalkylether inhibitors of the voltage gated
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
sodium channel (Na ) 1.7 with potent analgesic
V
activity
┼
┼
┼
┼
┼, 
Shaoyi Sun, Qi Jia, Alla Y. Zenova, Michael S. Wilson, Sultan Chowdhury,
Thilo
┼
╪
┼
┼
┼
Focken, Jun Li, Shannon Decker, Michael E. Grimwood, Jean-Christophe Andrez, Ivan
Hemeon, Tao Sheng, Chien-An Chen, #, Andy White, #,○ David H. Hackos, Lunbin Deng,
┼
┼,
╪
╪
┼
┼
┼
┼
┼
Girish Bankar, Kuldip Khakh, Elaine Chang, Rainbow Kwan, Sophia Lin, Karen
┼
╪
╪
╪
╪
Nelkenbrecher, Benjamin D. Sellers, Antonio G. DiPasquale, Jae Chang, Jodie Pang, Luis
┼
┼
┼
┼
┼
Sojo, Andrea Lindgren, Matthew Waldbrook, Zhiwei Xie, Clint Young, James P. Johnson,
┼
┼
┼
╪
╪
C. Lee Robinette, Charles J. Cohen, Brian S. Safina, Daniel P. Sutherlin, Daniel F.
Ortwine *,╪ and Christoph M. Dehnhardt *,┼
┼Xenon Pharmaceuticals Inc, 200-3650 Gilmore Way, Burnaby, BC, V5G 4W8 Canada;
╪
#
Genentech Inc, 1 DNA Way, South San Francisco, California 94080-4990, USA; Chempartner,
Building No. 5, 998 Halei Rd., Zhangjiang Hi-Tech Park, Pudong New Area, Shanghai, P. R.
China.
ACS Paragon Plus Environment
1

Page 3 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
KEYWORDS
Voltage-gated sodium channel, SCN9A, Na 1.7, Na 1.5, pain, inherited erythromelalgia,
V
V
formalin test.
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
ABSTRACT
Herein, we report the discovery and optimization of a series of orally bioavailable acyl
sulfonamide Na 1.7 inhibitors that are selective for Na 1.7 over Na 1.5 and highly efficacious
V
V
V
in in vivo models of pain and hNa 1.7 target engagement. An analysis of the physicochemical
V
properties of literature Na 1.7 inhibitors suggested acyl sulfonamides with high f could
V
sp3
overcome some of the pharmacokinetic (PK) and efficacy challenges seen with existing series.
Parallel library syntheses lead to the identification of analogue 7, which exhibited moderate
potency against Na 1.7 and an acceptable PK profile in rodents, but relatively poor stability in
V
human liver microsomes (HLM). Further design strategy then focused on the optimization of
potency against hNa 1.7 and improvement of human metabolic stability utilizing induced fit
V
docking in our previously disclosed X-ray co-crystal of the Na 1.7 voltage sensing domain
V
(VSD). These investigations culminated in the discovery of tool compound 33, one of the most
potent and efficacious Na 1.7 inhibitors reported to date.
V
INTRODUCTION
Drugs that block voltage-gated sodium channels (Na s) are widely used therapeutically as anti-
V
arrhythmics, anticonvulsants, anesthetics, and analgesics. Non-subtype selective sodium channel
blockers such as mexiletine and carbamazepine have been used for treatment of neuropathic pain
ACS Paragon Plus Environment
2

Journal of Medicinal Chemistry
Page 4 of 74
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
conditions. Unfortunately, this class of drugs is known to exhibit dose-limiting side effects in
patients, such as somnolence, dizziness, and nausea, that limit their utility as analgesic agents.1, 2
These adverse effects are thought to be derived in large part from a lack of selectivity on the NaV
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
family of channels. Nine Na isoforms are known (Na 1.1–Na 1.9), and there are indications
V
V
V
that isoform selective blockers may have an improved therapeutic index compared to the non-
selective compounds presently in use. In particular, Na 1.7 is a highly validated target for the
V
4
treatment of pain conditions based upon human genetic studies. Congenital insensitivity to pain
(
CIP), caused by a genetic deficiency of the SCN9A gene encoding Na 1.7, is a rare autosomal
V
recessive disorder characterized by the complete absence of pain sensation typically associated
with noxious stimuli.5–10 Individuals with CIP show no deficits in motor or cognitive function.11
Conversely, gain of function mutations in SCN9A result in painful conditions, including
inherited erythromelalgia (IEM),12–15 paroxysmal extreme pain disorder (PEPD) and small fiber
16
neuropathies.17 These findings indicate that Na 1.7 has an essential role in mediating pain
V
signaling in humans and suggests that selective Na 1.7 antagonism may be a highly effective
V
strategy to treat painful conditions.4
Na 1.5 is the predominant sodium channel in cardiac myocytes and has been identified as an
V
important off-target to avoid in Na 1.7 drug discovery. Inhibition of Na 1.5 has been
V
V
1
8
demonstrated to lead to cardiac arrhythmias. As a consequence, our optimization campaign
focused on establishing selectivity for Na 1.7 over Na 1.5. Due to a high sequence homology
V
V
among Na isoforms, compounds that are highly specific for Na 1.7 have been difficult to
V
V
identify despite considerable efforts by the pharmaceutical industry. Recently, subtype selective
1
9–25
small molecule Na 1.7 inhibitors have been reported.
An example is aryl sulfonamide 1 (PF-
V
05089771) (Figure 1), which selectively (>600 fold) blocks Na 1.7 over Na 1.5, and was
V
V
ACS Paragon Plus Environment
3

Page 5 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
advanced into clinical trials.20a,b,d Our earlier work described the discovery of two structurally
distinct series of aryl sulfonamide Na 1.7 inhibitors exemplified by 2 and 3,21b as well as the
2
1a
V
2
6
X-ray structure of the binding site on hNa 1.7. We also recently disclosed a novel class of
V
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
21c
triazolyl sulfonamide Na 1.7 inhibitors, exemplified by 4.
V
H
F
O
N
N
F
O
H
Cl
F
O
H
S
N
N
N
N
Cl
O
S
N
Cl
S
S
O
O
N
S
O
F
O
O
F
Cl
F
H N
2
H N
HN
N
2
N
H
O
N
1
(PF-05089771)
2
3
O
O
S
F
O
O
N
N
O
N
N
H
S
O
N
O
N
N
N
O
H
S
F
N
N
H
O
Cl
OMe
O
O
F
Cl
4
5 (PF-05241328)
6
Figure 1. Structures of selected subtype-selective Na 1.7 inhibitors.
V
2
1b
Although potent on Na 1.7 and highly selective against Na 1.5, aryl sulfonamides such as 3
V
V
displayed poor ADME properties. Therefore, we sought to identify alternative chemical series
to mitigate these liabilities. A literature survey on selective Na 1.7 inhibitors revealed a primary
V
focus on aryl sulfonamides, with only a single paper on a companion series of acyl
sulfonamides.22a This report did not contain a characterization of the analogues in in vivo models
of pain. We and others21b,22a have found that for many series, large multiples of the Na 1.7 IC
V
50
values are needed to achieve acceptable efficacy in in vivo models. We recently reported in vivo
3
7
studies on a class of acyl sulfonamides that showed efficacy at low Na 1.7 IC multiples.
V
50
However, the design and optimization of compounds that are highly active in in vivo assays has
so far been elusive, which prompted the current study on acyl sulfonamide optimization that
focused on improving in vivo efficacy.
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 74
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
An analysis of the patent landscape through 2013 for aryl and acyl sulfonamide Na 1.7
V
2
7
inhibitors guided our start of the program. We noticed that acyl sulfonamides occupy a more
favorable physicochemical property space such as lower molecular weight (MW) and clogP
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
(
Figure 2). For example, aryl sulfonamides displayed an average clogP of 4.1 and average MW
of 490, while acyl sulfonamides showed an average clogP of 3.0 and an average MW of 400.
Given that optimization very often involves increases in molecular weight and clogP, acyl
sulfonamides represented an attractive series for progression within Lipinski property space
which was an important criterion for us at the start of the program.28 Additionally, acyl
sulfonamides represented a less explored chemical series compared to aryl sulfonamides,20a,22a
providing an opportunity to deepen our understanding of this class of compounds.
Known acyl sulfonamides such as 5 (PF-05241328)20a or 622a have a highly planar architecture
due to the presence of multiple aromatic rings. The number of aromatic rings in a molecule has
been shown to negatively affect aqueous solubility, and has been associated with an increased
29
attrition risk in drug development, suggesting a need for the identification of acyl sulfonamides
with improved drug-like properties. Our strategy to increase favorable properties focused partly
3
30
on increasing fsp3 (fraction of sp carbons relative to the total carbon count). These efforts
culminated in 33, a potent, molecularly selective and metabolically stable Na 1.7 inhibitor that
V
demonstrated efficacy in various models of pain.
ACS Paragon Plus Environment
5

Page 7 of 74
Journal of Medicinal Chemistry
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
Figure 2. Acyl sulfonamides generally occupy a more favorable physiochemical property space
(average cLogP of 3.0 versus 4.1 and MW of 400 versus 490 for acyl versus aryl sulfonamides)
for achieving oral bioavailability.
RESULTS AND DISCUSSION
Compounds were evaluated in vitro in a radioligand binding assay (LBA) that measured the
3
displacement of aryl sulfonamide [ H]GX–545 from HEK membranes that expressed hNa 1.7
V
including β1.26 Effects in cells were assessed by whole cell patch voltage clamp
electrophysiology measurements (EP) of sodium currents mediated by human hNa 1.7 and
V
2
6
hNa 1.5 heterologously expressed in HEK293 cells. The LBA assay was used as a high
V
throughput assay to guide design and to establish binding to a preferred site in voltage sensing
2
6
domain 4 (VSD4) of Na 1.7. The functional potency data obtained by EP is biologically more
V
relevant as it tracks the movement of the channel in a cell but is associated with a larger
measurement error. 31 Given these assay behaviors, we were most interested in compounds that
showed high potencies in both assays.
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 74
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
Acyl sulfonamide 7,21c,32 possessing a pendant adamantyl group, was discovered in early
efforts to explore ether linked saturated rings. This compound inhibited hNa 1.7 with an IC of
V
50
0.07 M in the LBA (Table 1). A voltage–clamp assay confirmed the ligand binding result (IC50
0.006 M) and indicated that 7 was highly selective against hNa 1.5 (IC50 = 1.9 M,
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
=
V
Na 1.5/Na 1.7 = 340-fold). Further profiling showed that 7 had good permeability, moderate
V
V
solubility (31 M), no significant CYP inhibition for isoforms tested (> 10 M for CYPs 3A4,
A2, 2C19, 2C9 and 2D6), and no hERG liability (IC > 10 M). However, 7 showed moderate
1
5
0
and poor stabilities in mouse and human liver microsomes (MLM = 32ml/min/Kg, HLM =
9ml/min/kg), respectively. Initial in vivo pharmacokinetic (PK) evaluation in mouse showed
1
good oral bioavailability and exposure. The observed exposure, despite moderate mouse liver
microsomal stability, was most likely a result of high plasma protein binding (> 99.9% bound)
that restricted the intrinsic clearance of the molecule. We were cognizant of the fact that high
plasma protein binding would require exquisite potency to achieve efficacy in vivo without
requiring very high exposure. In addition, optimization of metabolic stability is particularly
important for these highly bound analogues as their low volume of distribution requires good
metabolic stability to achieve a half-life suitable for BID dosing. Hence, our optimization
focused particularly on increasing potency and HLM stability. Given its potency, selectivity, and
biopharmaceutical properties, 7 represented a good starting point for a lead optimization
campaign.
ACS Paragon Plus Environment
7

Page 9 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. In vitro activity and pharmacokinetic properties of compound 7.
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
0
.07
[3
H]GX-545 LBA IC (M)
50
0.006
hNa 1.7 EP IC (M)
V
50
hNa 1.5 EP IC (M)/ sel. h1.5/1.7
1.9/340
V
50
LM CLhep (ml/min/kg)
human/mouse
1
9 / 28
MDCKIPapp (A to B)
-6
1
3 x 10 cm/s
PPB (%)h / m
99.9 / 99.9
Mouse IV PK (1 mg/kg)
CL: 9.2 ml/min/kg,
V : 1.2 l/kg, t : 1.7 h
ss
½
Mouse PO PK (30 mg/kg)
AUC: 240 µMh, t : 3.5 h,
½
Cmax: 45 µM, F: 100%,
To better understand how the present acyl sulfonamides might interact with VSD4 of the
Na 1.7 receptor, we modeled representative analogs into the crystal structure of VSD4 via
V
induced fit docking. A proposed fit of 7 in VSD4 is shown in Figure 3, supported in part by
mutagenesis and small molecule X-ray data. For the aryl sulfonamide GX-936 a large loss of
potency was seen for the R1608A and R1608K mutations (3000x and 400x, respectively) while
2
6
the R1602A mutation caused little change in affinity. In contrast, mutagenesis studies with acyl
sulfonamide 7 showed that both R1608 and R1602 were required for binding, with 300x and 50x
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 74
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
decreases in potency observed for the R1608A and R1602A mutations, respectively. We
therefore postulated that the acyl sulfonamide warhead anion is delocalized between the carbonyl
and sulfonyl moieties, making hydrogen bonding and ionic interactions with both R1608 and
R1602. Sulfonyl oxygens are known to form hydrogen bonds with arginines, although the
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
interaction is often weak. The combination of the SO -to-R1608 and the amide carbonyl oxygen
2
to R1602 interactions proposed for acyl sulfonamides rationalizes the roughly equipotent affinity
between the acyl and aryl class of inhibitors. A small molecule X-ray structure of the sodium
salt of 7 grown in aqueous media (Figure 4) shows a sodium ion complexed to an ionized amide
carbonyl oxygen, which may partially exist as its imino tautomer, consistent with our hypothesis
and binding to R1602 in VSD4. This proposed fit results in the linker phenyl ring being offset
26
from the analogous aryl sulfonamide phenyl ring in the X-ray crystal structure of GX-936, and
is consistent with the lack of corresponding SAR between aryl and acyl sulfonamides.
Specifically, the linker phenyl in 7 is predicted to be translated approximately 1.5-1.7 angstroms
out of the active site, providing room for the adamantyl group and substituted cyclohexyl rings of
the present series to fit into a mostly hydrophobic region flanked by W1538, V1541, F1583,
M1582, and the sidechain carbon atoms of D1586. This fit would drive the chlorine of 7 into a
different region than that occupied by the cyano in GX-936, essentially occupying the center of
the channel.
The X-ray structure of Na 1.7 VSD4 represents one conformation of a flexible system
V
designed to enable outward movement of S4 during a depolarization. This flexibility and
induced fit of ligands introduces uncertainty in the modeling calculations. Other states of this
protein can no doubt exist. Nonetheless, the modeling experiments using induced fit docking
ACS Paragon Plus Environment
9

Page 11 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
have served to rationalize parts of the SAR such as phenyl ring and sulfonamide substitution and
provided design guidance throughout the project.
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
Figure 3. Induced fit docking of 7 into the VSD4 of the hNa 1.7 receptor, shown in cyan. The
V
active site cavity and helices are from the induced fit version. Key residues and distances are
2
6
shown. The co-crystal structure of GX-936 and VSD4 of Na 1.7 (PDB code 5EK0) is
V
superimposed and shown in orange.
ACS Paragon Plus Environment
1
0

Journal of Medicinal Chemistry
Page 12 of 74
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
Figure 4. Small molecule crystal structure of the sodium salt of 7 (crystals grown in 50% MeOH
solution), with sodium ions complexed to the charged oxygen atom shown as orange spheres.
Two molecules of methanol are also shown. For details, see the Supporting Information.
With this data in hand, we focused our attention on improving the potency and metabolic
stability of compound 7. We were cognizant of the fact that large lipophilic groups like
adamantane are main contributors to high metabolic clearance, so our initial SAR exploration
focused on a diverse set of replacements aimed at understanding the impact on potency and
metabolic stability (Table 2). Since metabolic stability is very often linked to lipophilicity,33 we
also tracked measured LogD (mLogD) and cLogP as well as ligand lipophilic efficiency (LLE)
(
calculated using hNa 1.7 EP IC and cLogP) in the initial set of compounds to understand if
V 50
there is a correlation. In addition, lowering lipophilicity is an established medicinal chemistry
3
4
strategy to decrease the risk for drug promiscuity and unwanted pharmacology. The differences
ACS Paragon Plus Environment
1
1

Page 13 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
between cLogP and mLogD reflects the compound ability to exist as deprotonated species at
pH=7.4.
Cyclohexane 8 demonstrated a 7-fold loss in hNa 1.7 potency, but a slight increase in
V
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
HLM stability compared to 7. Incorporation of an oxygen atom into the cyclohexyl ring (9) was
not tolerated. Extending the linker by one carbon (10) only slightly increased the potency while
negatively effecting metabolic stability. Shortening the linker (11) or reducing the ring size by
one carbon (12) resulted in a further decrease in potency but enhanced metabolic stability. As
expected, metabolic stability and lipophilicity of the left-hand portion of the molecule were
generally correlated, with cyclopentyl ether 12 and ethylcyclopropyl ether 13 being the most
stable in this subset. However, both displayed unacceptably low potency in the LBA.
In this initial library, we found a relatively good correlation between mLogD and HLM
stability. For example, 13 showed one of the lowest mLogD values (0.4) and was also the most
stable in HLMs (HLM Cl hep = 7ml/min/kg). On the other hand, the homologated cyclohexane
1
0, which showed one of the highest LogD’s compared to other analogues in this subset, had low
stability in HLMs (HLM Clhep = 19ml/min/kg). However, in a larger data set this tendency
could not be confirmed. We suspect structural features are overriding the HLM-lipophilicity
trend.
Considering that 7 displayed the highest potency in a functional assay and the highest LLE, we
used this building block for further optimization to explore the SAR of substituents on the phenyl
core (Table 3). A 2-fluoro substitution was essential for hNa 1.7 potency, as removal (14) or
V
replacement with chlorine (15) resulted in significant drops in potency. Replacement of the 5-
chlorine in 7 with a fluorine (16) also significantly reduced potency. Likewise, replacement of
the 5-chlorine with methyl (17), ethyl (18) or isopropyl (19) were not beneficial to potency.
ACS Paragon Plus Environment
1
2

Journal of Medicinal Chemistry
Page 14 of 74
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
However, introducing a cyclopropyl group as a chloro replacement (20) turned out to be highly
favorable, providing a greater than 10–fold potency enhancement with minimal increases in
molecular weight or lipophilicity (LogP = 3.5 and 3.6 for 7 and 20, respectively) resulting in a
significant gain in LLE. Unfortunately, modification of the phenyl ring had no impact on HLM
stability. The marked increase in potency for 20 over 7 may be a result of the cyclopropyl
making additional hydrophobic contacts with the protein, or simply being more compatible with
the adjacent hydrophobic membrane environment.
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
Table 2. Effects of adamantane replacements on Na 1.7 potency and HLM stability.
V
F
O
O
S
N
O
R1
H
O
Cl
hNa 1.7 IC (M)
V
50
HLM CL
hep
c
mLogDd cLogP LLEe
Compd
R1
(ml/min/kg)
LBAa
EPb

7
0.07
0.006
19
2.4
3.5
4.8
8
9
0.18
>10
0.043
7.4
15
nd
1.4
3.2
1.7
4.2
3.4

O
0.21


1
1
1
0
1
2
0.50
1.2
0.017
0.26
19
10
7
1.5
1.3
1.1
3.6
3
4.2
3.6
4.5


0.52
0.068
2.7
ACS Paragon Plus Environment
1
3

Page 15 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
3
2.7
0.44
7
0.4
2.4
4

a
3
Ligand binding assay using [ H]GX-545. IC s are an average of at least two independent
5
0
determinations.
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
b
IC s are an average of two independent determinations.
5
0
c
Predicted hepatic Clearance CLhep extrapolated from in vitro experiments in human liver
3
5
microsomes (CLHep = 19 ml/min/kg correlates to a 100% extraction ratio); nd: not determined.
d
Measured logD@pH7.4
e
LLE was calculated using hNa 1.7 EP IC and cLogP
V
50
Table 3. Effect of substitutions on the phenyl core on Na 1.7 potency and HLM stability.
V
Y
O
O
S
N
O
H
O
X
hNa 1.7 IC (M)
HLM CLhep
V
50
Compd
X
Y
LLEd
LBAa
EPb
(ml/min/kg)c
1
1
1
1
1
1
4
5
6
7
8
9
Cl
Cl
F
H
Cl
F
0.33
2.3
0.059
0.57
19
nd
19
19
19
19
3.8
2.2
2.9
4.9
4.7
3.8
0.64
0.037
0.015
0.028
0.041
0.006
0.006
0.010
Me
Et
i-Pr

F
F
F
2
0
F
0.006
0.0003
19
5.9
a
3
Ligand binding assay using [ H]GX-545 IC s are an average of at least two independent
5
0
determinations.
ACS Paragon Plus Environment
1
4

Journal of Medicinal Chemistry
Page 16 of 74
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
b
c
IC s are an average of two independent determinations.
5
0
Predicted hepatic Clearance CLhep extrapolated from in vitro experiments in human liver
3
5
microsomes (CLHep =19 mL/min/kg correlates to a 100% extraction ratio); nd: not
determined.
d
LLE was calculated using hNa 1.7 EP IC and cLogP
V
50
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
Given the high potency and LLE for 20, we now explored ways of improving the metabolic
stability of the adamantane itself. Metabolite identification (MetID) studies on 7 indicated that
the adamantal skeleton was subject to oxidative metabolism. We therefore decided to investigate
blocking potential sites of metabolism and reducing lipophilicity by incorporating polar
substituents on this ring system. As shown in Table 4, introduction of a polar substituent (21)
significantly enhanced HLM stability, but compromised potency on hNa 1.7 Other attempts to
V
block the potential sites of oxidative metabolism via introduction of fluorine (22 and 23) resulted
in negligible effect on HLM stability for 22 and markedly decreased HLM for 23. Both 22 and
23 displayed significant loss of hNa 1.7 potency compared to 7. Given 23 had relatively good
V
potency and stability, we wanted to probe if replacing Cl with cyclopropyl would increase
potency and retain HLM stability as seen for 20. While cyclopropyl analogue 24 did indeed show
potency increases in ligand binding and EP, its HLM stability decreased significantly. Finally,
substitution of each tertiary carbon on the adamantane with fluorine (25) substantially improved
stability but again at a marked cost in hNa 1.7 potency and LLE relative to 7 or 20.
V
ACS Paragon Plus Environment
1
5

Page 17 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. Effects of adamantane modification on Na 1.7 potency and HLM stability.
V
F
O
O
S
N
O
R1
H
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
O
X
hNa 1.7 IC (M)
V
50
HLM CL
(ml/min/kg)
hep
c
LLEd
Compd
R1
X
LBAa
EPb

2
2
1
2
HO
Cl
Cl
>10
2.2
9
3.1
4.3


F
0.357
0.031
18
F
F
23
Cl
0.290
0.008
0.011
12
4.4
4.8



F
F
2
4
0.0031 16
F
25
F
Cl
2.398
0.752
6.7
3.5
F
a
3
Ligand binding assay using [ H]GX-545 IC s are an average of at least two independent
5
0
determinations.
b
IC s are an average of two independent determinations.
5
0
c
Predicted hepatic Clearance CLhep extrapolated from in vitro experiments in human liver
35
microsomes (CLHep =19 ml/min/kg correlates to a 100% extraction ratio); nd: not determined.
d
LLE was calculated using hNa 1.7 EP IC and cLogP
V
50
ACS Paragon Plus Environment
1
6

Journal of Medicinal Chemistry
Page 18 of 74
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
F
O
F
O
O
O
S
S
N
N
O
O
H
H
O
O
O
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
2
6
M1 (62%), M2 (15%), M3 (19%)
m/z 368
m/z 384
Figure 5. Summary of the major metabolic pathway of 26.
Lessons taken from the SAR developed in Tables 2-4 influenced the design of the next round
of analogs (Table 5). Specifically, our goal was to incorporate strategies used successfully to
block metabolism of the adamantyl group on the synthetically less complex cyclohexyl ring
while also taking advantage of potency improvements realized in the core. We also wanted to
maintain the relatively high fsp3 count of the overall molecule. Replacing the 5-chloro in 8 with a
cyclopropane (26) led to a significant boost in Na 1.7 potency, particularly in the functional
V
assay, but unfortunately retained the HLM instability seen with previous analogs. MetID studies
revealed three major metabolites resulting from oxidation of the cyclohexyl ring (Figure 5).
Further improvements in stability were obtained through gem-difluorination at the 4-position of
the cyclohexyl ring (27). This compound also retained some of the LLE from 26. However, it
showed only moderate LBA potency compared to others in the series. We next turned to
examining the effect of varying the acyl sulfonamide substitution on potency and stability.
Interestingly, replacement of the N-methylsulfonyl substituent in 27 with N-cyclopropysulfonyl
(
28) or N-((2-methoxyethyl)sulfonyl (29) enhanced stability. Potency was retained in 28, while
2
9 was much less potent, presumably due to size limitations in this region of the protein.
ACS Paragon Plus Environment
1
7

Page 19 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
From this small SAR study of the warhead, we learned that our ability to modulate potency
and stability via changes in acyl sulfonamide substitution was limited. We therefore turned our
attention back to the cyclohexyl-ether appendage. We examined the effect of adding methyl
groups36 by incorporation of a substituent on the tertiary carbon at the 1-position of the
cyclohexyl ring. Compound 30 provided a moderate improvement in both potency and stability
over 29. Increasing the size of the substituent at the tertiary position had a detrimental effect on
stability (31). Therefore, the methyl group was incorporated as a preferred substituent for further
cyclohexane explorations. Replacing the 4-difluoro group in 30 with a trifluoromethyl (32)
retained potency relative to 30, but decreased stability. As an alternative to blocking
metabolically labile sites by fluorine atoms, we also explored the incorporation of strained
systems. An example is 33, which showed the highest functional activity against hNa 1.7 (IC
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
V
50
=
0.6 nM) and retained moderate stability in human microsomes and hepatocytes (Table 6.)
Compounds 30 and 33 were selected for further characterization against a panel of human and
rodent Na isoforms as they showed distinct advantages over other analogs. Compound 33
V
showed the highest potency in EP, high LLE, and moderate stability, while 30 showed
acceptable potency, HLM stability, and displayed the highest LLE in the cyclohexane series.
Although 33 showed good selectivity over hNa 1.5, it was only about 10- and 2-fold selective
V
against hNa 1.1 and hNa 1.2. Compounds 30 and 33 also did not inhibit (IC >10uM) mNa 1.8
V
V
50
V
in TTX-resistant channels in mouse dorsal root ganglion neurons. Compound 30 showed high
selectivity over hNa 1.5 and greater selectivity against hNa 1.1 than acyl sulfonamide 33.
V
V
ACS Paragon Plus Environment
1
8

Journal of Medicinal Chemistry
Page 20 of 74
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
Table 5. Effects of cyclohexane modification on Na 1.7 potency and HLM stability.
V
F
O
O
R2
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
S
N
O
H
R1
O
hNa 1.7 IC (M)
HLM
CLhep
V
50
Compd
R1
R2
LLEd
(
ml/min/
LBAa
EPb
c
kg)
2
6
Me
Me
0.022
0.093
0.033
0.304
0.015
0.0007
0.015
0.008
0.075
0.003
19
5.9
4.5
3.9
3.7
4.0

F
F
27
4.9
3.7
2.7
2.2



F
F
2
2
3
8
9
0


O
F
F
F
F



Et
F
F
3
1
0.017
0.005
9.5
3.3

F3C
3
3
2
3


0.010
0.015
0.006
13
12
3.2
4.6

0.0006

a
3
Ligand binding assay using [ H]GX-545. IC s are an average of at least two independent
5
0
determinations.
ACS Paragon Plus Environment
1
9

Page 21 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
b
c
IC s are an average of two independent determinations.
5
0
Predicted hepatic Clearance CLhep extrapolated from in vitro experiments in human liver
3
5
microsomes (CLHep =19 ml/min/kg correlates to a 100% extraction ratio); nd: not determined.
d
LLE was calculated using hNa 1.7 EP IC and cLogP
V
50
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
Table 6. In vitro profiles of compounds 30 and 33
3
0
33
Compound
a
Na 1.7 IC (nM)
V
50
3 / 2.5 / 9
0.6 / 2.2 / 3.7
human / mouse / rat
hNa 1.5, 1.1, 1.2, 1.6
V
260 / 48 / 6 / 49
26
50 / 6 / 2 / 38
38
a
IC (nM)
5
0
% inh. TTX-R @ (10µM)
Liver Microsomes
2
/ 38 / 10 / 7
12 / 65 / 22 / nd
CLhep (ml/min/kg)b
human / mouse / rat / dog
Hepatocytes
2
.8 / 17 / 4.3 / 3.5
13 / 6.0 / 5.2 / 7.6
CLhep (mL/min/kg)c
human / mouse / rat / dog
CYP 3A4 / 2C9 / 2C19 /
2
D6 / 1A2
>
10 / > 10 / > 10 / > 10 / > 10
> 10 / 6.1 / > 10 / > 10 / > 10
IC (M)
5
0
MDCK1
-6
Papp (A to B) x 10 cm/s
Ratio (BA / AB)
2
2
9
ACS Paragon Plus Environment
2
0

Journal of Medicinal Chemistry
Page 22 of 74
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
.55
1.0
MDCK1-MDR1
-6
Papp (A to B) x 10 cm/s
11
3.7
1.2
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
Ratio (BA / AB)
1.4
MDCK2-mBCRP
-6
Papp (A to B) x 10 cm/s
Ratio (BA / AB)
PPB (%)
0.4
56
5.1
2.1
human / mouse / rat
Kinetic Solubility (M )
mLogD/LLE(EP/mLogD)d,e
99.6 / 99.8 / 99.8
99.9 / 99.9 / 99.9
32
10
1.4/6.8
2.1/7.1
aIC50 values are an average of two independent determinations using PatchXpress
electrophysiology technique.
b
Predicted hepatic clearance (CL ) extrapolated from in vitro human, mouse and rat
hep
3
5
microsome experiments.
c
Predicted hepatic clearance (CL ) extrapolated from in vitro human, mouse, rat and dog
hep
3
5
hepatocyte experiments.
d
Measured logD@pH7.4
e
LLE was calculated using hNa 1.7 EP IC and mLogD
V
50
Compounds 30 and 33 were also tested against mouse and rat Na 1.7 where they showed
V
potencies similar to that of the human isoform. The invariant human, rat, and mouse Na 1.7
V
potencies are a distinct characteristic the present acyl sulfonamide series relative to other
ACS Paragon Plus Environment
2
1

Page 23 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
reported acyl and aryl sulfonamides (e.g., 4) that bind to the VSD4 site. We believe that the
increased conformational flexibility between cycloalkyether and phenyl core allow our
compounds to adopt a binding mode in rat Na 1.7 that is similar the one shown in for compound
V
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
7
in Figure 3, and adapt that binding mode to residue changes in the VSD between human,
mouse, and rat. The less flexible, more deeply binding aryl sulfonamides such as 4 cannot readily
adjust to the residue differences in the rNa 1.7 binding pocket, and are less potent as a result.
V
Gratifyingly, both 30 and 33 were selective for hNa 1.7 over hNa 1.5. No significant DDI risk
V
V
was observed, as 30 and 33 did not show notable CYP inhibition (Table 6). We tracked
permeability in a mouse MDCK2-BCRP cell line, which contains the main transporter found at
the blood brain barrier, to better understand the brain exposure of our compounds. Compound
3
3 had good permeability and low efflux, while 30 proved to be a mBCRP substrate. Both
compounds showed high protein binding in human, mouse and rat (Table 6). We also evaluated
33 in a CEREP panel of 33 receptors at 10uM concentration, where we saw >75% inhibition of
only two targets: Na ion channels (80%), and PPAR (91%). The measured LogD’s at pH 7.4
V
were acceptable for both 30 (mlogD=1.4) and 33 (mLogD=2.1). Noticeably, LLE when
calculated based on mLogD was high for both 30 and 33, and in line with low off target activity
described for 33.
Table 7. Pharmacokinetic profiles for compounds 30 and 33
IV
CL
(ml/min/kg)
PO
Compd
Species
t½
h)
Vss
(l/kg)
Cmax
AUC
t (h)
F(%)
49
½
(
(µM ) (µMh)
3
0
Mousea
Ratb
2
9
.1
.0
1.6
0.28
0.29
15.6
16.9
69
25
350
160
30
0.4
66
ACS Paragon Plus Environment
2
2

Journal of Medicinal Chemistry
Page 24 of 74
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
Mousea
3
3
3
3
1.5
3.2
1.2
0.3
0.41
0.88
0.14
9.2
nc
22
13
9
113
78
31
49
68
Ratb
8
5
.1
.4
33
Dogc
6.0
83
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
aAverage of 3 male FVB mice, dosed IV with 1 mg/kg of compound 30 or 33 in 5%
DMSO/35% PEG400/60% PBS or dosed PO with 30 mg/kg of compound 30 or 33 as MCT
suspension.
bAverage of 3 male Sprague-Dawley rats, dosed IV with 1 mg/kg of compound 30 or 33 in
0% DMSO/50% PEG400/40% PBS or dosed PO with 5 mg/kg of compound 30 or 33 as
MCT suspension. nc: not calculated.
1
cAverage of 3 male beagle dogs, dosed IV with 33 in 0.5 mg/kg in 15% EtOH, 60% PEG400
in water, and orally dosed with 31 at 1 mg/kg as MCT suspension.
Based on their overall in vitro potencies and DMPK profiles, 30 and 33 were selected for
further in vivo evaluation. Their PK profiles were determined in mouse and rat, 33 was also
evaluated in dog. Results are summarized in Table 7. Consistent with the predicted CLhep and
high plasma protein binding, both compounds exhibited very low plasma clearance (CL). Both
30 and 33 displayed moderate half-lives (t ) in mice. Both molecules showed increased half-
½
lives in the rat IV experiments, with 33 showing acceptable half lives in dog IV and PO studies.
They also displayed low volume of distributions (Vss) values in these animal species, presumably
because of high plasma protein binding. In agreement with their high permeability and
solubility, both compounds displayed good to moderate oral plasma exposures (AUC) and
bioavailabilities (F). The oral bioavailabilities of 33 in mice, rats, and dogs were 31%, 49% and
6
8%, respectively.
ACS Paragon Plus Environment
2
3

Page 25 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To determine if the PK and potencies of our compounds were sufficient to demonstrate
analgesia in vivo, 30 and 33 were evaluated in transgenic mice21c,37 that express a human variant
3
8
of Na 1.7 that results in IEM (mutation I848T). Aconitine-induced pain in these mice provides
V
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 novel way to quantify engagement of hNa 1.7 in vivo and facilitates the characterization of
V
molecularly selective inhibitors that, most often, show species specificity for rodent orthologs.
This assay reports the number of nociceptor responses (shaking and licking) in response to the
sodium channel activator aconitine. As illustrated in Figures 6 and 7, 30 and 33 demonstrated
dose-related inhibition of aconitine-induced nociceptive behaviors, matched by dose-dependent
increases in plasma concentrations. Consistent with its significantly higher functional potency
against Na 1.7, 33 was much more potent in this study demonstrating >75% analgesic efficacy
V
starting at a plasma concentration of approximately 3.1 M (which corresponds to a unbound
plasma concentration of Cpu = 3nM, approximately 5-fold the hNa 1.7 IC ) (0.3 mg/kg, oral
V
50
dose). At these very low efficacious free plasma concentrations, we did not expect to see
significant inhibition of other Na isoforms. Plasma concentrations of 33 for the 3 and 10 mg/kg
V
groups were measured as 25 and 79 M, respectively. The brain concentrations were 1.9 and 6.2

M, respectively, which provided a low brain to plasma ratio of approximately 0.08. We were
gratified to see this remarkable efficacy for 30 and 33 despite their high plasma protein binding.
The significantly higher efficacy of 33 over 30 was rationalized by its 5-fold higher potency in
the functional assay for inhibition of hNa 1.7. In vivo effects of 33 were further assessed in a
V
39
formalin-induced pain model in mice, an assay that has been used to evaluate analgesic effects
for many sodium channel blockers.23, 24, 25b,40 This pain model utilizes formalin as a pain-inducing
agent and generates a biphasic pain response. The initial response (phase I), occurring within 5–
1
0 minutes post-injection of formalin, indicates an acute localized irritation (partially due to
ACS Paragon Plus Environment
2
4

Journal of Medicinal Chemistry
Page 26 of 74
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
41
activation of TRPA1 channels ), while a delayed response (Phase II) that begins 10–15 minutes
post-injection indicates neurogenic inflammation.42 As illustrated in Figure 8, 33 produced
significant response reduction in phase II of the formalin assay at a PO dose of 10 mg/kg. The
plasma concentration of 33 was measured at the end of the study (approximately 3 hours after
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
oral dosing) to be 50 µM. Despite the modest selectivity for the inhibition of hNa 1.7 over
V
hNa 1.1 and hNa 1.2, 33 was well tolerated in mice, as there were no clinical observations over
V
V
the course of this study. We also evaluated 30 and 33 for inhibition of mNa 1.8 in TTX-
V
resistant channels in mouse dorsal root ganglion neurons, and found minimal inhibition (IC50
>
10 µM), indicating that their efficacy was not caused by inhibition of this isoform.
Figure 6. Effects of 30 on aconitine-induced pain in humanized IEM transgenic mice. Each bar
represents the mean of at least 4 animals. Error bars represent standard errors of the mean. ****:
p < 0.0001 versus vehicle group. Statistical significance of results was calculated by two-way
ANOVA using Prism version 7 (GraphPad software)
ACS Paragon Plus Environment
2
5

Page 27 of 74
Journal of Medicinal Chemistry
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
Figure 7. Effects of 33 on aconitine-induced pain in humanized IEM transgenic mice. Each bar
represents the mean of at least 4 animals, except the 0.3 mg/kg and 1mg/kg groups (3 animals for
each group at these doses) Error bars represent standard errors of the mean. ****: p < 0.0001 and
*
**: p < 0.005 versus vehicle group for nociceptive events. Statistical significance of results was
calculated by two-way ANOVA using Prism version 7 (GraphPad software)
ACS Paragon Plus Environment
2
6

Journal of Medicinal Chemistry
Page 28 of 74
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
Figure 8. Dose response effect of 33 in the mouse formalin model. (A) Time course of inhibition
of formalin-evoked pain behavior. Each point represents the mean number of times that a mouse
(n = 4) shook or licked its injected paw over a 5 min period. (B) Quantification of the response
to formalin binned into two different phases: phase I (0–10 min after formalin injection) and
phase II (11–40 min after formalin injection). Each bar represents the mean of 4 animals. Error
bars represent standard errors of the mean. ***: p = 0.0021 versus vehicle group. Statistical
ACS Paragon Plus Environment
2
7

Page 29 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
significance of results was calculated by two-way ANOVA using prism version 7 software
(Graphpad software)
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
CHEMISTRY
The general synthetic routes to prepare acyl sulfonamides 7–33 are outlined in Schemes 1–8.
For syntheses of analogues 7–14, 16, 21–23 and 25 (Scheme 1), a simple and efficient
nucleophilic aromatic substitution of intermediates 37, 38 and 39 with an appropriate alcohol
was applied. This method was general, and a wide range of alcohols could be used in a library
fashion to prepare the desired molecules 7–14, 16, 21–23 and 25.
Scheme 1. Syntheses of compounds 7–14, 16, 21–23 and 25 a
Y
O
Y
O
Y
O
O
O
a
b
S
S
Z
N
N
O
O
H
H
R1
F
F
O
X
X
X
34: X = Cl, Y = F, Z = OH
35: X = Cl, Y = H, Z = OH
36: X, Y = F, Z = Cl
37: X = Cl, Y = F
38: X = Cl, Y = H
39: X, Y = F
7-13, 21-23, 25: X = Cl, Y = F
14: X = Cl, Y = H
16: X, Y = F
a
Reagents and conditions: (a) for 37 and 38: MeSO NH , EDCI, DMAP, CH Cl , rt, 58–96%;
2
2
2
2
1
for 39: MeSO NH , Et N, CH CN, rt, 54%; (b) R OH, t-BuOK, DMSO, rt, 9–58%.
2
2
3
3
Scheme 2. Synthesis of compound 15a
Cl
Cl
O
O
Cl
Cl
Cl
S
Br
Br
N
b
O
a
H
O
HO
O
Cl
4
0
41
15
ACS Paragon Plus Environment
2
8