A novel benzazepinone sodium channel blocker with oral efficacy in a rat model of neuropathic pain

Article abstract of DOI:10.1016/j.bmcl.2013.03.121

A series of benzazepinones were synthesized and evaluated for block of Na_v1.7 sodium channels. Compound 30 from this series displayed potent channel block, good selectivity versus other targets, and dose-dependent oral efficacy in a rat model of neuropathic pain.

Full text of DOI:10.1016/j.bmcl.2013.03.121

Bioorganic & Medicinal Chemistry Letters 23 (2013) 3640–3645
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A novel benzazepinone sodium channel blocker with oral efficacy
in a rat model of neuropathic pain
a
b
d
c
b
Scott B. Hoyt a, , Clare London , Catherine Abbadie , John P. Felix , Maria L. Garcia , Nina Jochnowitz ,
Bindhu V. Karanam d, Xiaohua Li a, Kathryn A. Lyons a, Erin McGowan b, Birgit T. Priest c,
McHardy M. Smith c, Vivien A. Warren c, Brande S. Thomas-Fowlkes c, Gregory J. Kaczorowski c,
Joseph L. Duffy a
⇑
a Department of Medicinal Chemistry, Merck Research Laboratories, Rahway, NJ 07065, United States
b Department of Pharmacology, Merck Research Laboratories, Rahway, NJ 07065, United States
c Department of Ion Channels, Merck Research Laboratories, Rahway, NJ 07065, United States
d Department of Drug Metabolism, Merck Research Laboratories, Rahway, NJ 07065, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of benzazepinones were synthesized and evaluated for block of Nav1.7 sodium channels. Com-
pound 30 from this series displayed potent channel block, good selectivity versus other targets, and
dose-dependent oral efficacy in a rat model of neuropathic pain.
Received 27 November 2012
Revised 25 March 2013
Accepted 27 March 2013
Available online 5 April 2013
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Benzazepinone
Sodium channel
Nav1.7
Neuropathic pain is a chronic pain state resulting from injury to
the peripheral or central nervous system.1 Its onset is typically
associated with underlying conditions such as diabetes, shingles,
or chemotherapy use that cause nerve damage through a variety
of mechanisms. Because few drugs have been developed specifi-
cally to treat neuropathic pain, patients are often prescribed anti-
convulsants or antidepressants as therapies. Optimized for other
indications, these brain-penetrant agents provide inadequate relief
for some patients, and elicit dose-limiting CNS side effects such as
sedation and coordination impairment.
Neuropathic pain signaling can be initiated by the firing of ecto-
pic, high-frequency action potential bursts in or near damaged
neurons. These action potentials are generated and propagated
along neuronal axons via the opening of voltage-gated sodium
channels (Nav1.x). Of the nine voltage-gated sodium channel iso-
forms that have been cloned and functionally expressed to date
(Nav1.1–Nav1.9), three (Nav1.3, Nav1.7 and Nav1.8) appear to play
roles in pain signaling.2
efficacy in the treatment of neuropathic pain, thereby providing
validation for this approach (Fig. 1).3,4 Blockers with improved po-
tency and/or selectivity have been more recently reported (1, ralfi-
namide and A-803467, Fig. 1), and have demonstrated efficacy in
preclinical pain models.5–9
Data from human genetic studies support a key role for Nav1.7
in pain signaling. Gain of function mutations in SCN9A, the gene
that encodes Nav1.7, have been identified as causal for two inher-
ited pain syndromes, primary erythermalgia (PE) and paroxysmal
extreme pain disorder (PEPD).10,11 In PE, mutations in SCN9A alter
the kinetics of Nav1.7 activation and deactivation, lower the
threshold for action potential firing, and produce hyperexcitability
in nociceptive neurons. Affected individuals experience burning
pain in their extremities in response to stimuli such as mild
warmth or exercise —inflammatory- like symptoms that also reca-
pitulate aspects of neuropathic allodynia. In PEPD, a different set of
mutations produce channels with a reduced capacity for fast inac-
tivation. Once activated, the mutant channels remain open much
longer than wild-type channels, producing long-duration action
potentials and persistent signaling in affected neurons. These
mutations yield a phenotype characterized by spontaneous parox-
ysms of rectal, ocular, or submandibular pain.
Nav1 sodium channel blockers have been investigated as treat-
ments for neuropathic pain. Weak, non-subtype selective blockers
such as lidocaine and carbamazepine have demonstrated clinical
Loss of function mutations in SCN9A have also been reported,
and they lead to a rare and remarkable condition: the complete
⇑
Corresponding author.
E-mail address: scott_hoyt@merck.com (S.B. Hoyt).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.121

S. B. Hoyt et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3640–3645
3641
O
NH2
improvement of pharmacokinetic properties and efficacy in this
series. These efforts have led to the discovery of 30, a state-depen-
dent Nav1.7 blocker that is moderately selective against other Nav
isoforms (Nav 1.5, Nav 1.6, and Nav 1.8), and that displays dose-
dependent oral efficacy in a preclinical model of neuropathic pain.
Our primary objectives were to improve pharmacokinetics (PK)
and efficacy in this series. An initial benchmark compound had dis-
played a moderate PK profile in rat and dog (compound 1, Table 1).
Incubation of 1 in the presence of rat liver microsomes had re-
vealed metabolic oxidation at the N-Boc tert-butyl group, and at
one or more unspecified sites on the benzazepinone phenyl ring
(C6–C9, Table 1). Prior efforts to block this metabolism via chlori-
nation of the benzazepinone 7-position had led to reduced clear-
ance and improved oral exposure.15 Intrigued by this initial
result, we wanted to systematically examine the effect of blocking
positions 6–9 on the benzazepinone core.
H
N
O
N
N
O
OCF3
N
O
N
H
NH
O
O
Carbamazepine
Lidocaine
1
H2N
O
H
N
H3CO
O
Cl
N
H
O
F
O
CH3O
A-803467
Figure 1. Nav1 sodium channel blockers.
Ralfinamide
inability to sense pain.12,13 Individuals with nonsense mutations in
SCN9A are healthy and normal in terms of most non-nociceptive
sensory functions, but are congenitally unable to sense pain in
any form.14 Given that several Nav1 isoforms have been implicated
in pain signaling, it is striking that loss of function in just one is suf-
ficient to block all pain perception. Together with the gain-of-func-
tion findings, these data provide compelling human genetic
validation for Nav1.7 as an important pain target.
To be used safely in the clinic, sodium channel blockers must in-
hibit chronic pain signaling while leaving acute pain perception,
motor coordination and other nerve functions intact. We believe
that this can be achieved through a combination of subtype selec-
tivity and/or state-dependent channel block. Voltage-gated sodium
channels exist in three main conformational states: resting, open
and inactivated. In healthy nerve and cardiac tissue, these channels
reside predominantly in the resting state. In contrast, the repeti-
tive, high-frequency firing patterns characteristic of some chronic
pain syndromes (e.g., trigeminal neuralgia) cause sodium channels
to accumulate in the inactivated state. Compounds that selectively
bind and stabilize the inactivated state should block chronic pain
signaling preferentially, thus minimizing the potential for mecha-
nism-based adverse effects.
Substituted benzazepinones were synthesized as shown in
Scheme 1. Known 7-fluorobenzazepinone 2 was prepared in four
steps from commercially available 2-bromo-4-fluoroaniline using
our recently reported method.16 Sequential treatment of 2 with so-
dium hydride and 2-iodopropane effected lactam alkylation to yield
N-isopropyl derivative 3. That compound could then be elaborated
to amine 4 using a modified version of the process developed by
Armstrong et al.17 Thus, exposure of 3 to tetramethylethylenedi-
amine (TMEDA), iodotrimethylsilane (TMSI) and iodine effected
iodination alpha to the lactam carbonyl. Nucleophilic displacement
of iodide with azide and subsequent hydrogenation then furnished
racemic amine 4. That amine was coupled with N-Boc-D-2-F-Phe
under standard conditions (BOP, i-Pr2NEt) to give two diastereo-
meric products that were separable via HPLC.18 Exposure of the
fast-eluting (R,R) diastereomer 5 to standard conditions for N-Boc
deprotection (trifluoroacetic acid in dichloromethane) provided a
crude TFA salt that could be coupled with 2-trifluoromethyl-4-flu-
orobenzoic acid to afford target compound 6.
N-Cyclopropyl benzazepinones were prepared as outlined in
Scheme 2. Known trifluorobenzazepinone 7 was synthesized from
2-bromo-3,4,6-trifluoroaniline using our recently reported meth-
od.16 Upon heating in the presence of 10 mol % copper iodide and
N,N0-dimethylethylenediamine,
7
underwent Cu(I)-catalyzed
The goal of the present study is to develop state-dependent
Nav1.7 blockers as treatments for chronic pain. Toward that end,
we recently reported the discovery of a structurally novel class of
benzazepinone Nav1.7 blockers. An early example from this class
(compound 1, Fig. 1) displayed potent, state-dependent block of
Nav1.7, a moderate pharmacokinetic profile, and oral efficacy in a
rat model of neuropathic pain. Subsequent work has targeted the
Buchwald coupling with vinyl bromide or 2-bromopropene to pro-
vide 8a or 8b, respectively.19 Treatment of 8a or 8b with diethyl-
zinc, trifluoroacetic acid and diiodomethane then effected
cyclopropanation to yield 9a or 9b.20 Those compounds were sub-
jected to the sequence shown in Scheme 1 for the conversion of 3
to 6 to provide target compounds 28–31.
Table 1
Effect of benzazepinone substitution on Nav1.7 potency and rat pharmacokinetics in the 2-OCF3 Phe series
O
9
6
O
OCF3
N
8
7
N
H
R
NH
O
O
Compound
R
Nav1.7 (IC50, nM)
MK-0499 (% inh @ 10
lM)
F (%)
AUCN (po,
lM h/mpk)
Cmax
Clp (mL/min/kg)
t1/2 (h)
1
H
9-F
8-F
7-F
35
70
18
34
108
181
494
57
17
24
26
24
32
7
19
21
24
24
27
20
35
31
59
21
30
n.d.
0.31
0.39
0.20
0.66
0.71
0.41
0.22
0.54
n.d.
0.30
0.38
0.24
0.47
0.53
0.09
0.01
0.32
n.d.
24
21
29
16
13
39
26
17
n.d.
2.3
2.7
2.5
2.3
3.0
1.9
1.3
2.7
n.d.
10
11
12
13
14
15
16
17
7-Cl
7-CF3
8-CF3
7,9-di-F
6,7,9-tri-F
67

3642
S. B. Hoyt et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3640–3645
O
H
N
O
O
N
4
N
3
e, f
a
b - d
NH2
F
F
F
F
2
O
O
O
F
O
F
N
N
g, h
N
H
N
H
NH
O
F
NH
O
O
5
6
F3C
F
Scheme 1. Reagents and conditions: (a) NaH (60%/oil), 2-iodopropane, DMF, 0–60 °C (78%); (b) TMEDA, TMSI, I2, CH2Cl2, ꢀ10–0 °C (70%); (c) NaN3, DMF, 40 °C (81%); (d) H2
(1 atm), Pd/C, EtOH (86%); (e) BOP, i-Pr2NEt, N-Boc-D-2-F-Phe, CH2Cl2 (79%); (f) HPLC (Chiralcel AD column, 30% i-PrOH/heptane); (g) TFA, CH2Cl2; (h) BOP, i-Pr2NEt, 2-CF3-4-
F-benzoic acid, CH2Cl2 (76% over 2 steps).
Specific details of the experimental protocols employed have re-
cently been described.23 Selected compounds were also examined
in whole cell electrophysiology experiments to determine their
affinities for the resting and inactivated states (Kr and Ki, respec-
tively) of Nav1.7.24
R
R
F
F
F
F
O
O
O
H
N
N
N
c
a or b
F
F
F
F
F
Potent Nav1.7 blockers were counterscreened against a variety
of other targets. Because block of hERG channels has been associ-
ated with potentially lethal ventricular arrhythmias, compounds
were tested in a binding assay that measures displacement of
35S-labelled MK-0499, a known hERG blocker.25
A benchmark compound was further characterized in manual
whole cell electrophysiology assays designed to measure the func-
tional block of the Nav1.5, Nav1.6, and Nav1.8 resting and inacti-
vated states. This compound was also tested for selectivity
against a broad set of >170 ion channels, receptors, and enzymes
(Ricerca Pharma Services, Taipei, Taiwan).
Finally, selected blockers were evaluated in a rat pharmacoki-
netic assay, and in rat pharmacology assays that measured analge-
sic efficacy and propensity for CNS side effects. These assays
included a modified version of the Chung spinal nerve ligation
(SNL) model of neuropathic pain and the rota-rod test of motor
coordination.26–28
Using the method outlined in Scheme 1, we synthesized a series
of compounds wherein the effects of benzazepinone substitution
on Nav1.7 block and rat pharmacokinetics were examined. As
shown in Table 1, fluorination of the 9- or 8-positions (compounds
10 and 11) had little effect on potency or PK.29 Fluorination of the
9a (R = H)
9b (R = CH3)
7
8a (R = H)
8b (R = CH3)
Scheme 2. Reagents and conditions: (a) conversion of 7–8a: 10 mol % CuI,
N,N0dimethylethylenediamine, K2CO3, vinyl bromide, toluene, 100 °C (90%); (b)
conversion of 7–8b: 10 mol % CuI, N,N0dimethylethylenediamine, K2CO3, 2-bromo-
propene, toluene, 100 °C (83%); (c) Et2Zn, TFA, CH2I2, CH2Cl2, 0 °C (78% for
conversion of 8a–9a, 80% for conversion of 8b–9b).
Compounds listed in Tables 1–4 were synthesized according to
these procedures using the appropriate commercially available
starting materials. In several instances (compounds 10–17), com-
pounds were prepared via the coupling of amines such as 4 with
N-Boc-D-2-trifluoromethoxy phenylalanine, a non-commercially
available amino acid that was synthesized via the method of Schol-
kopf.21 Representative experimental procedures for the prepara-
tion of these compounds have been published previously.22
Compounds were assayed for their ability to block human
Nav1.7 channels that were stably expressed in HEK293 cells. The
extent of channel block was determined in a functional, membrane
potential-based assay that measures the fluorescence resonance
energy transfer (FRET) between two membrane-associated dyes.
Table 2
Effect of benzazepinone substitution on Nav1.7 potency and rat pharmacokinetics in the 2-F Phe series
O
9
6
O
F
N
8
7
N
H
R
NH
O
F3C
F
Compound
R
Nav1.7 (IC50, nM)
MK-0499 (% inh @ 10
lM)
F (%)
AUCN (po,
lM h/mpk)
Cmax
Clp (mL/min/kg)
t1/2 (h)
6
7-F
7-Cl
7-CF3
7,9-di-F
7-Cl-9-F
6,7,9-tri-F
254
30
359
264
260
103
36
32
10
34
17
20
56
90
75
31
74
68
0.45
0.58
0.60
0.42
0.93
1.00
0.28
0.33
0.36
0.37
0.64
0.53
36
44
33
21
21
19
1.7
2.8
2.5
1.9
1.9
5.6
18
19
20
21
22

S. B. Hoyt et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3640–3645
3643
Table 3
Effect of sidechain amide on Nav1.7 potency and rat pharmacokinetics in the 2-F Phe series
F
O
O
F
N
N
H
F
NH
R
O
F
Compound
R
Nav1.7 (IC50, nM)
MK-0499 (% inh @ 10
lM)
F (%)
AUCN (po,
lM h/mpk)
Cmax
Clp (mL/min/kg)
t1/2 (h)
23
24
25
26
27
29
2-Pyridyl
2-CF3-Ph
2-CF3-4-F-Ph
(c-Pr)CH3
(c-Pr)CF3
(c-Bu)CF3
443
229
103
103
137
107
22
10
20
0
2
8
n.d.
n.d.
68
47
62
n.d.
n.d.
1.00
0.44
0.79
0.20
n.d.
n.d.
0.53
0.57
0.61
0.14
n.d.
n.d.
19
35
23
n.d.
n.d.
5.6
0.8
2.2
2.4
28
44
Table 4
Effect of benzazepinone N-substituent on Nav1.7 potency and rat pharmacokinetics
R2
F
O
O
F
N
N
H
F
NH
R1
M)
O
F
Compound
R1
R2
NAv1.7 (IC50, nM)
MK-0499 (% inh @ 10
l
F (%)
AUCN (po,
lM h/mpk)
Cmax
Clp (mL/min/kg)
t1/2 (h)
22
28
29
26
30
31
2-CF3-4-F-Ph
2-CF3-4-F-Ph
2-CF3-4-F-Ph
(c-Pr)CF3
(c-Pr)CF3
(c-Pr)CF3
i-Pr
c-Pr
(c-Pr)CH3
(c-Pr)CH3
c-Pr
103
85
131
137
117
101
20
31
44
2
26
54
68
1.00
n.d.
n.d.
0.79
0.72
n.d.
0.53
n.d.
n.d.
0.61
0.53
n.d.
19
9.6
n.d.
n.d.
2.2
1.9
n.d.
n.d.
n.d.
62
100
n.d.
n.d.
n.d.
23
43
n.d.
(c-Pr)CH3
A
B
hNav1.7
hNav1.5
0
-1
-2
-3
-4
0.0
-0.5
-1.0
-1.5
-2.0
Vehicle
3 µM Compound 30
Vehicle
3 µM Compound 30
-120
-110
-100
-90
-80
-70
-60
-100
-80
-60
-40
-20
Voltage (mV)
Voltage (mV)
Figure 2. Whole-cell electrophysiology experiments measuring block of Nav1.7 and Nav1.5 by compound 30. At a concentration of 3
dependence of inactivation of Nav1.7 channels (A) and Nav1.5 channels (B) relative to vehicle, consistent with a mechanism of state-dependent block. Fitting the data to the
Hill equation yields a Ki = 0.34 M and Kr = 24 M for Nav1.7, and a Ki = 2.0 M and Kr >10 M for Nav1.5.
lM, compound 30 shifts the voltage-
l
l
l
l
7-position, however (compound 12), led to a doubling of oral expo-
sure (AUCN) and a reduction in clearance (Clp) relative to unsubsti-
tuted parent compound 1, presumably due to block of metabolic
oxidation at the activated C7 position. The 7-chloro analog 13,
though somewhat less potent, displayed a similar PK profile, while
the 7- and 8-CF3 derivatives 14 and 15 displayed both reduced
Nav1.7 block and weaker pharmacokinetic properties. Finally, the
7,9-difluoro analog 16 maintained, but did not improve upon, the
rat PK profile of 7-F analog 12.
The effects of benzazepinone substitution were further exam-
ined in a series of compounds containing an N-amide-D-2-F-Phe
sidechain that had proved optimal in earlier work. In this series,
we focused on benzazepinones that were substituted at the 7-po-
sition. The 7-F, 7-Cl and 7-CF3 derivatives 6, 18 and 19 featured
comparable rat PK profiles, with 18 proving most potent (Table 2).
Two compounds that contained additional fluorination at the 9-po-
sition (compounds 20 and 21) showed reduced clearance relative
to their monosubstituted counterparts (compounds 6 and 18).
Consistent with results from the 2-OCF3 Phe series, the 7,9-difluoro
derivative 20 offered no substantive improvements in potency or
other PK parameters relative to the 7-F analog 6. The 6,7,9-trifluoro
derivative 22, on the other hand, displayed improved Nav1.7 block,
reduced clearance, and increased oral exposure and half-life rela-
tive to 6, and on balance provided the best combination of potency
and PK in this series.
In the 6,7,9-trifluoro series, we next examined the impact of the
sidechain amide on Nav1.7 block and rat PK. While the 2-pyridyl
and 2-CF3-Ph amides 23 and 24 proved somewhat less potent,

3644
S. B. Hoyt et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3640–3645
Table 5
the steady-state inactivation curve was shifted by 16 mV in the
presence of 3 M compound 30. Assuming a simple two-state
Block of the inactivated state of Nav1 subtypes by compound 30
l
model where compound 30 can bind with different affinities to
either the resting (Kr) or inactivated state (Ki) of Nav1.7, the shift
in the voltage-dependence of inactivation can be used to calculate
Kr and Ki, yielding Kr = 24 lM and Ki = 0.34 lM for the experiment
shown. Average values are presented in Table 5.
Subtype
Ki (
l
M)
n
Nav1.5
Nav1.6
Nav1.7
Nav1.8
2.38 0.55
3.88 1.12
0.44 0.13
3.02 0.08
5
4
3
2
By comparison, the effects of compound 30 on Nav1.5 were less
pronounced (Fig. 2B). In the presence of 3
mid-point of the steady-state inactivation curve was shifted by
5 mV, yielding Kr >10 M and Ki = 2.0 M for the experiment shown.
Similar experiments were carried out for Nav1.6 and Nav1.8, and re-
sults are summarized in Table 5. Compound 30 is thus moderately
selective for block of Nav1.7 versus 1.5, 1.6 and 1.8.
lM compound 30, the
several bulky alkyl amides (compounds 25–27) displayed Nav1.7
block equal to that of 22 as well as lower activity in the MK-
0499 counterscreen (Table 3). From that set, the trifluoro-
methyl(cyclopropyl) derivative 26 displayed the lowest clearance,
highest oral exposure and best overall profile.
l
l
Compound 30 was further examined for activity against a broad
set of >170 ion channels, receptors, and enzymes. This set included
pain targets such as cyclooxgenase (COX-1), prostanoid receptor
(EP1), bradykinin receptor (BK1), cannabinoid receptor (CB2), and
Finally, in the 2-CF3-4-F-Ph and trifluoromethyl(cyclopropyl)
amide series, we performed focused optimizations of the benzaz-
epinone N-substituent. Prior work had shown that medium-sized
alkyl groups provided the best balance of potency and reduced
clearance (Clp). With that in mind, we prepared N-isopropyl, N-
cyclopropyl and N-(cyclopropyl)methyl analogs in each amide ser-
ies (Table 4, compounds 22, 26 and 28–31). Within each series, the
three analogs displayed comparable Nav1.7 block. In the trifluoro-
methyl(cyclopropyl) amide series, the N-isopropyl and N-cyclopro-
pyl derivatives 26 and 30 displayed broadly similar rat PK profiles,
with 30 offering higher oral bioavailability and 26 lower clearance.
Compound 30 was profiled in a whole-cell electrophysiology
(EP) assay that measures affinities for the Nav1.7 inactivated and
resting states to determine the state-dependence of Nav1.7 channel
opiod receptors (d, K,
l). At a concentration of 25 lM, compound
30 exhibited <50% activity against every target in this screen, and
is thus highly selective for block of sodium Nav1 channels versus
a broad range of other targets.
Compound 30 displayed dose-dependent oral efficacy in a rat
spinal nerve ligation (SNL) model of neuropathic pain. When dosed
orally at 10 and 30 mg/kg, compound 30 produced significant
reversal of allodynia at 2 and 4 h post dose (Fig. 3A), with an
ED50 = 15 mg/kg. At an oral dose of 30 mg/kg, 30 produced 62%
and 55% reversal of allodynia at 2 and 4 h post-dose. In terms of
both maximal reversal and ED50, these results compare favorably
with those we obtained using the clinical standards gabapentin
and mexiletine, which displayed ED50s of 50 and 45 mg/kg, respec-
tively (Fig. 3C; gabapentin 30 mg/kg p.o.: % reversal @ 2 h = 36%;
mexiletine 30 mg/kg p.o.: % reversal @ 2 h = 42%).
block.23 At a concentration of 3
lM, compound 30 blocked 48% of
the current evoked by 6 ms depolarizations to 0 mV from a holding
potential of ꢀ70 mV (Fig. 2A). Block was dependent on holding po-
tential, consistent with preferential block of channels in the inacti-
vated state. This mechanism is further supported by the observed
drug-dependent shift in the voltage-dependence of inactivation.
The maximum current available on depolarization from negative
holding potentials was minimally affected, but the mid-point of
Finally, compound 30 did not induce any significant impair-
ment of motor coordination as measured in a rat rotarod assay.
In this assay, rats were initially trained on a rotarod apparatus
100%
15
62%
***
55%
39% ***
3 mg/kg (n=7)
10 mg/kg (n=15)
30 mg/kg (n=8)
4*3*%
10
5
***
28%
*
*
20%
18%
15%
*
**
10%
7%
4%
1%
0
pre-SNL
post-SNL
2 h
4 h
8 h
24 h
Time post-dose
A
80
30 mg/kg P.O.
100 mg/kg P.O.
60
40
20
0
300
200
100
0
Compound 30
Mexiletine
Gabapentin
-20
-40
-60
Baseline 2 h
4 h
1
3
10
30 100
Dose (mg/kg P.O.)
B
C
Figure 3. Effects of compound 30 in rat models of neuropathic pain and motor coordination. In the rat SNL model (A and C, upper), 30 dose-dependently reversed SNL-
induced allodynia with a duration of action greater than 4 h post-dose (A), and its antiallodynic effects were greater than mexiletine or gabapentin at 2 h post-dose (C, upper).
Antiallodynic effects of 30 and standard analgesics at 2 h post-dose, expressed as percentage of reversal of allodynia calculated as indicated in reference 26 (C, upper). In the
rat rotarod model (B and C, lower), 30 elicited no impairment of motor coordination at doses of 30 and 100 mg/kg PO (B). In comparison, mexiletine at doses of 10 mg/kg PO
and higher, and gabapentin at doses of 30 mg/kg PO and higher elicited significant impairment of motor coordination (>20% reduction in latency time) at 2 h post-dose (C,
lower).

S. B. Hoyt et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3640–3645
3645
12. Cox, J. J.; Reimann, F.; Nicholas, A. K.; Thornton, G.; Roberts, E.; Springell, K.;
Karbani, G.; Jafri, H.; Mannan, J.; Raashid, Y.; Al-Gazali, L.; Hamamy, H.;
Valente, E. M.; Gorman, S.; Williams, R.; McHale, D. P.; Wood, J. N.; Gribble, F.
M.; Woods, C. G. Nature 2006, 444, 894.
13. Goldberg, Y. P.; MacFarlane, J.; MacDonald, M. L.; Thompson, J.; Dube, M.-P.;
Mattice, M.; Fraser, R.; Young, C.; Hossain, S.; Pape, T.; Payne, B.; Radomski, C.;
Donaldson, G.; Ives, E.; Cox, J.; Younghusband, H. B.; Green, R.; Duff, A.;
Boltshauser, E.; Grinspan, G. A.; Dimon, J. H.; Sibley, B. G.; Andria, G.; Toscano,
E.; Kerdraon, J.; Bowsher, D.; Pimstone, S. N.; Samuels, M. E.; Sherrington, R.;
Hayden, M. R. Clin. Genet. 2007, 71, 311.
for 3 min at a speed of 10 rpm. A single dose of test compound or
vehicle was then administered, and rats were tested at 2 and 4 h
post-dose. During testing, the speed was set at 10 rpm for 60 s,
then accelerated to 60 rpm, and the time required for rats to fall
after the beginning of the acceleration was recorded. When dosed
orally at 30 and 100 mg/kg, compound 30 did not affect time to fall
(Fig. 3B). Thus, at oral doses up to 10-fold higher than those re-
quired for analgesic efficacy, compound 30 does not impair motor
coordination. In contrast, gabapentin and mexiletine induce signif-
icant impairment at oral doses at or below those required for SNL
efficacy (Fig. 3C). Compound 30 thus affords an improved thera-
peutic window of preclinical analgesic efficacy versus impairment
when compared to these standards. On the basis of a broad biolog-
ical screen that included numerous pain targets, the only identified
mechanism of analgesic efficacy for 30 is block of Nav1 sodium
channels. With regard to the three Nav1 isoforms that are thought
to be involved in pain signaling (Nav1.3, 1.7 and 1.8), manual EP
14. As reported in Ref. 12,13, individuals with loss of function mutations in SCN9A
display an impaired sense of smell (anosmia).
15. Hoyt, S. B.; London, C.; Ok, H.; Gonzalez, E.; Duffy, J. L.; Abbadie, C.;
Dean, B.; Felix, J. P.; Garcia, M. L.; Jochnowitz, N.; Karanam, B. V.; Li,
X.; Lyons, K. A.; McGowan, E.; MacIntyre, D. E.; Martin, W. J.; Priest, B.
T.; Smith, M. M.; Tschirret-Guth, R.; Warren, V. A.; Williams, B. S.;
Kaczorowski, G. J.; Parsons, W. H. Bioorg. Med. Chem. Lett. 2007, 17,
6172.
16. Hoyt, S. B.; London, C.; Park, M. Tetrahedron Lett. 1911, 2009, 50.
17. Armstrong, J. D.; Eng, K. K.; Keller, J. L.; Purick, R. M.; Hartner, F. W.; Choi, W.-
B.; Askin, D.; Volante, R. P. Tetrahedron Lett. 1994, 35, 3239.
18. The two diastereomeric products were separated via HPLC on a Chiralcel AD
column eluting with 30% i-PrOH/heptane.
19. Jiang, L.; Job, G. E.; Klapars, A.; Buchwald, S. L. Org. Lett. 2003, 5, 3667.
20. Yang, Z.; Lorenz, J. C.; Shi, Y. Tetrahedron Lett. 1998, 39, 8621.
21. Schollkopf, U. Tetrahedron 1983, 39, 2085.
22. Hoyt, S. B.; London, C.; Ok, D.; Parsons, W. H. WO 145922, 2007.
23. Felix, J. P.; Williams, B. S.; Priest, B. T.; Brochu, R. M.; Dick, I. E.; Warren, V. A.;
Yan, L.; Slaughter, R. S.; Kaczorowski, G. J.; Smith, M. M.; Garcia, M. L. Assay
Drug Dev. Technol. 2004, 2, 260.
24. Hamill, O. P.; Marty, A.; Neher, E.; Sakmann, B.; Sigworth, F. J. Pflugers Arch.
1981, 391, 85. Procedure: Sodium currents were examined by whole cell
voltage clamp using an EPC-9 amplifier and Pulse software (HEKA Electronics,
Lamprecht, Germany). Experiments were performed at room temperature.
experiments indicate that 30 blocks Nav1.7 (Ki = 0.44
lM) in mod-
est preference to Nav1.8 (Ki = 3.02
lM). Block of Nav1.3 has not yet
been determined.
We have reported studies aimed at improving pharmacokinetics
and efficacy in a series of benzazepinone Nav1.7 blockers. A bench-
mark compound from this series, compound 30, displays potent
block of Nav1.7, high selectivity versus a broad range of other bio-
logical targets, and a good rat PK profile. In a rat model of neuro-
pathic pain, compound 30 displays dose-dependent oral efficacy
comparable to that of the clinical standard gabapentin. Finally, at
oral doses 10-fold higher than those required for efficacy, 30 does
not impair motor coordination, a potential advantage over existing
therapies that elicit impairment in both preclinical and clinical
settings.
Electrodes were fire-polished to resistances of 1.5–4 MX. Voltage errors were
minimized by series resistance compensation (75–85%), and the capacitance
artifact was canceled using the amplifier’s built-in circuitry. Data were
acquired at 50 kHz and filtered at 10 kHz. The bath solution consisted of
40 mM NaCl, 120 mM NMDG-Cl, 1 mM KCl, 2.7 mM CaCl2, 0.5 mM MgCl2,
10 mM NMDG-HEPES, pH 7.4, and the internal (pipet) solution contained
110 mM Cs-methanesulfonate, 5 mM NaCl, 20 mM CsCl, 10 mM CsF, 10 mM
BAPTA (tetra Cs salt), 10 mM Cs-HEPES, pH 7.4. Liquid junction potentials were
less than 4 mV and were not corrected for.
References and notes
25. Wang, J.; Della Penna, K.; Wang, H.; Karczewski, J.; Connolly, T. M.; Koblan, K.
S.; Bennett, P. B.; Salata, J. J. Am. J. Physiol. Heart Circ. Physiol. 2002, 284, H256.
26. Chaplan, S. R.; Bach, F. W.; Pogrel, J. W.; Chung, J. M.; Yaksh, T. L. J. Neurosci.
Methods 1994, 53, 55.
1. For a recent review, see: Jensen, T. S.; Madsen, C. S.; Finnerup, N. B. Curr. Opin.
Neurol. 2009, 22, 467.
2. Priest, B. T. Curr. Opin. Drug Discov. Devel. 2009, 12, 682.
3. Kalso, E.; Tramer, M. R.; McQuay, H. J.; Moore, R. A. Eur. J. Pain 1998, 2, 3.
4. (a) Campbell, F. G.; Graham, J. G.; Zilkha, K. J. J. Neurol. Neurosurg. Psychiatry
1966, 29, 265; (b) Rull, J. A.; Quibrera, R.; Gonzalez-Millan, H.; Lozano, C. O.
Diabetologia 1969, 5, 215.
5. For reviews of the medicinal chemistry of sodium channel blockers, see: (a)
Matulenko, M. A.; Scanio, M. J.; Kort, M. E. Curr. Top. Med. Chem. 2009, 9, 362;
(b) Zuliani, V.; Patel, M. K.; Fantini, M.; Rivara, M. Curr. Top. Med. Chem. 2009, 9,
396.
6. Hoyt, S. B.; London, C.; Gorin, D.; Wyvratt, M. J.; Fisher, M. H.; Abbadie, C.; Felix,
J. P.; Garcia, M. L.; Li, X.; Lyons, K. A.; McGowan, E.; MacIntyre, D. E.; Martin, W.
J.; Priest, B. T.; Ritter, A.; Smith, M. M.; Warren, V. A.; Williams, B. S.;
Kaczorowski, G. J.; Parsons, W. H. Bioorg. Med. Chem. Lett. 2007, 17, 4630.
7. Veneroni, O.; Maj, R.; Calabresi, M.; Favarelli, R. G.; Salvati, P. Pain 2003, 102,
17.
8. Jarvis, M. F.; Honore, P.; Shieh, C.-C.; Chapman, M.; Joshi, S.; Zhang, X.-F.; Kort,
M.; Carroll, W.; Marron, B.; Atkinson, R.; Thomas, J.; Liu, D.; Krambis, M.; Liu, Y.;
McGaraughty, S.; Chu, K.; Roeloffs, R.; Zhong, C.; Mikusa, J. P.; Hernandez, G.;
Gauvin, D.; Wade, C.; Zhu, C.; Pai, M.; Scanio, M.; Shi, L.; Drizin, I.; Gregg, R.;
Matulenko, M.; Hakeem, A.; Gross, M.; Johnson, M.; Marsh, K.; Wagoner, P.;
Sullivan, J. P.; Faltynek, C. R.; Krafte, D. S. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
8520.
27. A modified version of the model described in Ref. 26 was employed. Rats were
anesthetized with isoflurane and placed on
a heating pad. Using aseptic
techniques, the L5 spinal nerve was exposed, ligated, and transected. Muscle
and skin were then closed with 4–0 polydiaxone and wound clips, respectively.
Allodynia was assessed 7 days post surgery, and only rats that developed
allodynia as defined by a significant decrease in their mechanical threshold
using von Frey filaments were used. Mechanical threshold refers to the force
required to elicit a 50% withdrawal rate. Seven days post surgery, rats were
given a single oral dose (3, 10, or 30 mg/kg) of test compound or vehicle. Tactile
allodynia was then assessed at 2, 4, 8 and 24 h post-dose with calibrated von
Frey filaments (Stoelting Co.) using an up-down paradigm. Percent of Maximal
Possible Effect (%MPE) was calculated as: (post-compound—post-SNL) / (pre-
SNL—post-SNL) ꢁ 100, where 100% is equivalent to complete reversal of
allodynia.
28. In the rota-rod test of motor coordination, rats were initially trained on a rota-
rod apparatus (Stoelting Co.) for 3 min at a speed of 10 rpm. For testing, the
speed was set at 10 rpm for 60 s and then accelerated to 60 rpm. A single oral
dose (30 or 100 mg/kg) of test compound or vehicle was administered, and the
time required for rats to fall after the beginning of the acceleration was
recorded.
29. Rat PK experiments were conducted as follows: Test compounds were typically
formulated as 1.5 mg/mL solutions in mixtures of PEG300/water or DMSO/
PEG300/water. Fasted male Sprague–Dawley rats were given either a 1.0 mg/
kg iv dose of test compound solution via a cannula implanted in the femoral
vein (n = 3) or a 3.0 mg/kg p.o. dose by gavage (n = 3). Serial blood samples
were collected at 5 (i.v. only), 15, and 30 min, and at 1, 2, 4, 6, and 8 h post
dose. Plasma was collected by centrifugation, and plasma concentrations of
test compound were determined by LC–MS/MS following protein precipitation
with acetonitrile.
9. Kort, M. E.; Drizin, I.; Gregg, R. J.; Scanio, M. J. C.; Shi, L.; Gross, M. F.; Atkinson,
R. N.; Johnson, M. S.; Pacofsky, G. J.; Thomas, J. B.; Carroll, W. A.; Krambis, M. J.;
Liu, D.; Shieh, C.-C.; Zhang, X.; Hernandez, G.; Mikusa, J. P.; Zhong, C.; Joshi, S.;
Honore, P.; Roeloffs, R.; Marsh, K. C.; Murray, B. P.; Liu, J.; Werness, S.; Faltynek,
C. R.; Krafte, D. S.; Jarvis, M. F.; Chapman, M. L.; Marron, B. E. J. Med. Chem. 2008,
51, 407.
10. Dib-Hajj, S. D.; Rush, A. M.; Cummins, T. R.; Hisama, F. M.; Novella, S.; Tyrrell,
L.; Marshall, L.; Waxman, S. G. Brain 1847, 2005, 128.
11. Fertleman, C. R.; Baker, M. D.; Parker, K. A.; Moffatt, S.; Elmslie, F. V.;
Abrahamsen, B.; Ostman, J.; Klugbauer, N.; Wood, J. N.; Gardiner, R. M.; Rees,
M. Neuron 2006, 52, 767.