Subtype-selective Nav1.8 sodium channel blockers: Identification of potent, orally active nicotinamide derivatives

Article abstract of DOI:10.1016/j.bmcl.2010.08.121

A series of aryl-substituted nicotinamide derivatives with selective inhibitory activity against the Na_v1.8 sodium channel is reported. Replacement of the furan nucleus and homologation of the anilide linker in subtype-selective blocker A-803467 (1) provided potent, selective derivatives with improved aqueous solubility and oral bioavailability. Representative compounds from this series displayed efficacy in rat models of inflammatory and neuropathic pain.

Full text of DOI:10.1016/j.bmcl.2010.08.121

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6812–6815
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Subtype-selective Nav1.8 sodium channel blockers: Identification of potent,
orally active nicotinamide derivatives
Michael E. Kort a, , Robert N. Atkinson b, James B. Thomas b, Irene Drizin a, Matthew S. Johnson b,
⇑
Matthew A. Secrest b, Robert J. Gregg a, Marc J. C. Scanio a, Lei Shi a, Ahmed H. Hakeem a,
Mark A. Matulenko a, Mark L. Chapman b, Michael J. Krambis b, Dong Liu b, Char-Chang Shieh a,
XuFeng Zhang a, Gricelda Simler a, Joseph P. Mikusa a, Chengmin Zhong a, Shailen Joshi a, Prisca Honore a,
Rosemarie Roeloffs b, Stephen Werness b, Brett Antonio b, Kennan C. Marsh a, Connie R. Faltynek a,
Douglas S. Krafte b, Michael F. Jarvis a, Brian E. Marron b,
⇑
a Global Pharmaceutical Research and Development, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064-6100, USA
b Icagen, Inc., 4222 Emperor Blvd., Durham, NC 27703, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of aryl-substituted nicotinamide derivatives with selective inhibitory activity against the Nav1.8
sodium channel is reported. Replacement of the furan nucleus and homologation of the anilide linker in
subtype-selective blocker A-803467 (1) provided potent, selective derivatives with improved aqueous
solubility and oral bioavailability. Representative compounds from this series displayed efficacy in rat
models of inflammatory and neuropathic pain.
Received 31 May 2010
Accepted 24 August 2010
Available online 18 September 2010
Keywords:
Sodium channel blocker
Ó 2010 Elsevier Ltd. All rights reserved.
Nav1.8
PN3
Pain
Voltage-gated sodium (Nav1.x) channels are heteromeric trans-
membrane protein complexes which open in response to changes
in membrane potential to enable selective permeability for sodium
ions.1 The Nav1.x channels are critical for the initiation and propa-
gation of action potentials in primary sensory neurons2; in both
the peripheral and central nervous systems, they play a central role
in pain signaling.3 The nine known sodium channel isoforms can be
distinguished pharmacologically on the basis of their sensitivity to
blockade by natural toxins, particularly tetrodotoxin (TTx).4 Noci-
ceptive neurons express a number of voltage-gated sodium chan-
nel subtypes (Nav1.3, Nav1.7, Nav1.8, Nav1.9) that may contribute
to the ectopic hyperexcitability characteristic of chronic pain
states.5 Consistent with this hypothesis, nonselective blockade of
sodium channels contributes to the analgesic activity of a number
of clinically used agents, including mexiletine, lamotrigine, and
carbamazepine. Unfortunately, the utility of these drugs is limited
by their relatively narrow therapeutic index versus CNS and car-
diovascular adverse events.6 Subtype-selective small molecule
Nav1.x channel blockers have the potential to deliver the therapeu-
tic efficacy of their nonselective counterparts with a concomitant
amelioration of side effects, and are therefore an attractive class
of therapeutic agents for the treatment of pain.7
We recently have described the discovery of A-803467 (1,
Fig. 1), an isoform-selective inhibitor of the TTx-resistant (TTx-r)
sodium channel Nav1.8, that is efficacious in a variety of pre-
clinical pain models.8,9 This 5-aryl furfuranilide derivative potently
blocks recombinant human Nav1.8 (IC50 8 nM) under conditions
(À40 mV) where the channels reside at half-maximal inactiva-
tion.10 Comparable electrophysiology protocols run against other
subtypes (Nav1.2, Nav1.3 and Nav1.5) demonstrated a selectivity
ratio of 100- to 1000-fold for 1. In dissociated dorsal root ganglion
(DRG) neurons from rats, 1 displays a modest right-shift (IC50
125 nM) in its inhibitory potency against native TTx-resistant cur-
rents. Although dose-dependent attenuation of mechanical allo-
dynia was observed in neuropathic pain models with 1 upon
intraperitoneal (ip) administration, the poor oral bioavailability
OMe
Cl
O
O
O
R1
R2
Ar
N
N
H
H
OMe
A-803467 (1)
2
⇑
Corresponding authors. Tel.: +1 847 935 4945 (M.E.K.).
E-mail addresses: michael.e.kort@abbott.com (M.E. Kort), bmarron@icagen.com
(Ar = 5-, 6-ring heterocycle)
(B.E. Marron).
Figure 1. Expansion of the furfuramide SAR for Nav1.8.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.121

M. E. Kort et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6812–6815
6813
Table 1
(ꢀ10%) of this compound, a result of limited aqueous solubility
(<0.1 g/mL) and metabolic susceptibility (CLint 140 L/min/mg,
In vitro structure–activity relationships for 3,5-dimethylanilide derivatives
l
l
rat microsomes), precluded its further development. Therefore,
attention was directed toward the development of agents with en-
hanced physicochemical and pharmacokinetic properties. Herein,
we disclose the in vitro and in vivo profiles of selective, orally-effi-
cacious nicotinamide-derived Nav1.8 blockers.
It was anticipated that replacement of the furan nucleus in 1
with less oxidatively labile nitrogen-containing heterocyclic moie-
ties (e.g., 2) would offer an incremental improvement in aqueous
solubility, and potentially, in systemic exposure (Fig. 1). To this
end, a collection of five- and six-membered ring analogs could be
readily fashioned from known or commercially available bromo-
ester building blocks 3 (Scheme 1).11 Cross-coupling of these acti-
vated precursors with commercially available arylboronic acids
under standard Suzuki–Miyaura conditions12 afforded aryl esters
4, which were saponified subsequently to give the corresponding
carboxylic acids 5. Activated ester coupling using EDCI and 3,5-
R
O
Ar
N
H
Compd
R
Ar
mNav1.8
IC50
hNav1.8
TTx-r
IC50 (lM)
a
a
(
l
M)
IC50
(l
M)
O
O
6
7
OCF3
OCF3
0.29
0.002
0.28
0.13
0.34
1.4
0.95
2.0*
N
N
8
9
OCF3
OCF3
OCF3
OCF3
0.18
O
N
0.87
0.40
0.28
0.47
0.82
0.38
0.52
N O
N
10
11
0.051
0.008
dimethylaniline,
a fragment which produced optimal Nav1.8
S
N
in vitro potency in the furfuramide series,10 generated the desired
anilides 6–18.
S N
N
Relative to 2,5-furfuramide 6 (Table 1), the oxazole 7 possessed
essentially equipotent activity in recombinant mouse Nav1.8
(isotopic flux assay), but was considerably less active when
characterized electrophysiologically in recombinant human and na-
tive rat Nav1.8.10,13 Five-membered oxygen-containing heterocycles
wherein the aryl and carboxamide residues are spaced by a nitrogen
atom, as in regioisomeric oxazole 8 and 1,2,4-oxadiazole 9, also were
less potent across species. Interestingly, structurally complemen-
tary sulfur analogs such as 2-aryl thiazole 10 and 5-aryl-1,2,4-thia-
diazole 11 restored significant potency at mouse and human Nav1.8,
although both were characterized by unacceptably low microsomal
12
13
14
15
16
OCF3
OCF3
OCF3
OCF3
CN
0.057
1.5
0.021
0.064
0.026
0.78
0.034
0.51
N
N
0.53
11
0.097
ND
N
N
stability (CLint >300 lL/min/mg, rat and human). In accord with the
0.19
0.023
0.091
observations with 10 and 11, the isosteric six-membered ring het-
erocyclic core variants, pyrazine (12) and 2,6-pyridine (13), dis-
played favorable hNav1.8 inhibitory profiles (IC50 <100 nM), as did
the 3,5-pyridine isomer 14. Interestingly, the corresponding 4,6-
pyridine 15 was markedly less active at both mouse and human
Nav1.8 than its regioisomeric cohorts. The greater basicity of pyri-
dine relative to pyrazine, and the qualitatively improved kinetic
aqueous solubility of 14 versus 13 prompted the evaluation of
additional related 3,5-pyridine derivatives (16–18). Among these,
the 4-cyano (16) and 4-chloro (18) congeners offered in vitro po-
N
N
N
17
18
F
1.9
0.10
0.21
Cl
0.61
0.009
0.087
ND = not determined.
a
Estimated IC50 values from electrophysiology data generated at multiple testing
concentrations (see note 13; a description of * values is included).
tency comparable to 6. All of the anilide derivatives in Table 1 pos-
sessed selectivity versus Nav1.2 and the hERG channel in the range
of 30- to 300-fold (data not shown).
R
R
O
O
a
O
c
Br
Ar
N
Ar OEt
Ar OR1
4 (R1 = Et)
5 (R1 = H)
With the 3,5-pyridine scaffold established as a more tractable
starting point than the 2,5-furan motif, improvements in solubil-
ity-limited oral absorption were required to enable the broad
in vivo characterization necessary for clinical development. To re-
duce the crystallinity imparted by the highly planar biaryl anilide
system and refine the physicochemical properties of the chemo-
type, an extensive survey of one-carbon homologated amide sub-
stituents was undertaken, employing 5-(4-chlorophenyl)nicotinic
acid (19) as a starting material (Scheme 2).
H
3
6-18
b
R
O
F3CO
O
X
Y
X
N
H
N
Z
Y Z
H
Cmpd
Cmpd
R
X
Y
Z
X
Y
Z
6
7
8
9
10
11
O
O
N
N
N
N
CH
N
O
N
S
CH
CH
CH
O
CH
N
12 OCF3
13 OCF3
14 OCF3 CH
15 OCF3 CH
N
N
N
CH
N
CH
N
N
CH
CH
CH
N
CH
CH
CH
For elaboration of the amide, a diverse array of substituted ben-
zylic amines (20) and picolinamines (21) were chosen from com-
mercial sources. Because the presence of a basic nitrogen in 21
was a particularly attractive design element, a number of these
heteroatom-substituted (R = N- and O-containing functionality)
picolinamine building blocks were fashioned from substituted 2-
chloro-3-cyano pyridines via a known nucleophilic addition, nitrile
reduction sequence.14 Coupling of the two fragments was accom-
plished either with BOP reagent15 or via the intermediacy of the
16
17
18
CN
F
Cl
CH
CH
CH
S
N
Scheme 1. Reagents and conditions: (a) arylboronic acid, PdCl2(PPh3)2, Na2CO3, i-
PrOH, reflux; (b) LiOH, aq dioxane, 23 °C; (c) 3,5-dimethylaniline, EDCI, Et3N, DMF,
23 °C.

6814
M. E. Kort et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6812–6815
32). Unfortunately, Nav1.2 selectivity was eroded in 32 and 33.
Morpholine functionality provided still greater aqueous solubility
(>50 g/mL for 33), but reduced hNav1.8 inhibitory potency
approximately 6-fold compared with 32. We also examined the ef-
fects of halogenation of the central nicotinamide core and embel-
lishment of the flanking picolinamide residue at various locations
with additional sterically unencumbered (one-carbon append-
ages); a detailed discussion of these structure–activity relation-
ships will be the subject of a subsequent publication.
The potency, selectivity, pharmacokinetic profiles, and analgesic
effects of two representative 3-nicotinamide derived Nav1.8 block-
ers (16 and 31) are summarized in Table 3. Consistent with their
improved metabolic stability in rat microsomes and improved
aqueous solubility relative to the furfuramides, 16 and 31 were
more orally bioavailable and provided plasma concentrations that
were multiple-fold above their respective in vitro IC50 values
(3 mg/kg dose). These derivatives were further characterized by
uniformly high levels of serum protein binding (>97%) and appre-
ciable CNS penetration. Both displayed dose-dependent attenua-
tion of mechanical allodynia in the L5/L6 spinal nerve ligation
Cl
R
CO2H
H2N
X
l
+
N
19
20 (X = CH)
21 (X = N)
a or b,c
Cl
O
R
X
N
H
N
22-33
Scheme 2. Reagents and conditions: (a) BOP reagent, Et3N, CH2Cl2, 23 °C; (b) 19,
(COCl)2, cat. DMF, CH2Cl2, 23 °C; (c) 20 or 21, Et3N, CH2Cl2, 23 °C.
acid chloride, with the latter proving more practical for gram-scale
syntheses.
As shown in Table 2, several substituted benzyl (22–26) and 3-
picolinyl (27–33) carboxamides were potent and subtype-selective
Nav1.8 blockers. This finding stands in contrast to the SAR gener-
ated for the fufuramides, wherein aniline derivatives delivered
optimal activity.10 The potent benzyl amide 22 possessed an
attractive profile with 22 nM hNav1.8 potency and 86-fold selectiv-
ity versus Nav1.2 (electrophysiology characterization in recombi-
nant human cell line). Introduction of an ortho-methyl group on
the benzylic residue (22?23) provided a 10-fold boost in Nav1.8
activity across species. Comparable increases in potency at mouse
and human Nav1.8 were also observed for the corresponding
chloro analog 24, which showed a Nav1.2 selectivity profile supe-
rior to that of 23. Target potency was also enhanced for piperidine
and morpholine derivatives 25 and 26, respectively, albeit with a
concomitant erosion in selectivity at Nav1.2. The unsubstituted 3-
picoline analog 27 had greatly enhanced aqueous solubility, but
was a markedly less efficient Nav1.8 blocker; the regioisomeric 2-
Table 3
In vitro, pharmacokinetic, and in vivo profilesa of 16 and 31
16
31
hNav1.8 IC50
Rat TTx-r IC50
hNav1.2 IC50
hNav1.5 IC50
(
l
M)
M)
0.023
0.091
3.4
>30
16
270
1.4
17
0.003
0.028
2.8
4.7
19
68
3.1
95
0.35
0.6
(
l
(
(
l
l
M)
M)b
hERG IC50
CLint, (
L/min/mg)d
Aq solubility ( g/mL)
F, po, 3 mg/kg (%)e
Cmax g/mL)
(
l
M)c
l
l
(l
0.74
1.0
CLp (L/(h kg))
Vss (L/kg)
0.5
2.0
Brain/plasma
1.3
1.0
Plasma protein binding, rat (%)
Chung ED50, (mg/kg) [% effect, max]
99.5
72 [60%]
BZD
98.7
65 [66%]
Cl channel
and 4-picolines were essentially inactive (Nav1.8 IC50 >30 lM, data
not shown). With 27, introduction of a 2-methyl group (28) had
negligible impact on in vitro activity, though potency was restored
to some degree with ortho-chloro substitution (29). Encouraged by
the results with 29, it was discovered that heteroatom appendages
containing oxygen (30 and 31) or nitrogen (32 and 33) restored po-
tency and selectivity levels observed with 23, while affording a sig-
Cerep (>70 @ 10 l
M)e
a
Values shown for n = 6 per dose group.
Estimated IC50 values from electrophysiology data generated at multiple testing
concentrations (see note 13).
b
c
hERG flux assay determination (see Ref. 10).
Incubation with rat microsomes (1 h, 37 °C).
Determined in rats (n = 6).
d
e
nificant enhancement in solubility (e.g., 28 lg/mL for piperidine
Table 2
In vitro structure–activity relationships for amide substitutiona
Cl
O
R
N
X
H
N
a
a
a
Compd
X
R
mNav1.8 IC50
(lM)
hNav1.8 IC50
(lM)
TTx-r IC50
(l
M)
hNav1.2 IC50 (lM)
22
23
24
25
26
27
28
29
30
31
32
33
CH
CH
CH
CH
CH
N
N
N
N
N
H
Me
Cl
0.62
0.022
0.002
0.004
0.008*
0.010
ND
ND
0.14
0.007
0.003
0.014
0.087
0.085
0.009
0.066
0.045
0.130
ND
ND
0.56
0.032
0.028
0.008
0.098
1.9
0.11
1.2
0.050
0.12*
ND
<0.030b
0.081
<0.030b
0.032
7.9
N-Piperidinyl
N-Morpholino
H
Me
Cl
2.6
0.68
0.088
0.087
0.026
0.34
ND
18*
OEt
0.67*
2.8
OCH2CF3
N-Piperidinyl
N-Morpholino
N
N
0.070
1.6
ND = not determined.
a
Estimated IC50 values from electrophysiology data generated at multiple testing concentrations (see note 13; a description of * values is included).
An IC50 value could not be calculated due to high potency.
b

M. E. Kort et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6812–6815
6815
(Chung) model16 of neuropathic pain in the absence of discernable
sedative effects. Compounds 16 and 31 also were tested against
other channels and receptors expressed in peripheral sensory neu-
5. (a) Lai, J.; Hunter, J. C.; Porreca, F. Curr. Opin. Neurobiol. 2003, 13, 291; (b)
Waxman, S. G.; Cummins, T. R.; Dib-Hajj, S.; Fjell, J.; Black, J. A. Muscle Nerve
1999, 22, 1177.
6. (a) Kinloch, R. A.; Cox, P. J. Expert Opin. Ther. Targets 2005, 9, 685; (b) Priest, B. T.
Top. Med. Chem. 2008, 3, 121.
7. On number of reviews on the subject have appeared recently: (a) Matulenko,
M. A.; Scanio, M. J. C.; Kort, M. E. Curr. Top. Med. Chem. 2009, 9, 362; (b) Marron,
B. Ann. Rep. Med. Chem. 2006, 59; (c) Kyle, D. J.; Ilyin, V. I. J. Med. Chem. 2007, 50,
2583; (d) Priest, B. T. Curr. Opin. Drug Discov. Dev. 2009, 12, 682; (e) Bear, B.;
Asgian, J.; Termin, A.; Zimmerman, N. Curr. Opin. Drug Discov. Dev. 2009, 12,
543; (f) Dib-Hajj, S. D.; Binshtok, A. M.; Cummins, T. R.; Jarvis, M. F.; Samad, T.;
Zimmerman, K. Brain Res. Rev. 2009, 60, 65.
rons, including TRPV1, P2X2/3
KCNQ2/3 potassium channels; neither had significant activity at
these channels (IC50 >10 M). In addition, both showed minimal
cross-reactivity upon evaluation (10 M) in a broad screening pa-
, Cav2.2 calcium channels, and
l
l
nel (n = 70) of cell-surface receptors, ion channels, and enzymes
(CEREP, Poitiers, France).
In summary, we have identified a novel series of potent, selec-
tive, orally active pyridine-based blockers of the Nav1.8 sodium
channel by expansion of the central core of our previously dis-
closed furan, A-803467 (1).8,10 The present SAR studies indicate
that potency, confirmed at recombinant human Nav1.8 channels
and in isolated rat DRG neurons using electrophysiological record-
ings, and selectivity (against hNav1.2 and hERG) can be modulated
with core permutations and judicious choice of amine monomers
for amide elaboration. The tolerability of a one-carbon homologa-
tion of the amide linkage set the stage for production of com-
pounds with significantly enhanced physicochemical properties
relative to their furfuramide predecessors. Oral administration of
16 and 31 produced dose-dependant reversal of allodynia in the
spinal nerve ligation (Chung) model of neuropathic pain in rat.
The results presented herein reinforce the hypothesis that selective
pharmacological blockade of Nav1.8 can produce significant anti-
nociceptive effects in animal models of pain while minimizing ob-
served side-effects.
8. Jarvis, M. F.; Honore, P.; Shieh, C. C.; Chapman, M.; Joshi, S. K.; Zhang, X. F.; Kort,
M. E.; Carroll, W. A.; Marron, B. E.; Atkinson, R. N.; Thomas, J. B.; Liu, D.;
Krambis, M. J.; Liu, Y.; McGaraughty, S.; Chu, K.; Roeloffs, C. R.; Zhong, C.;
Mikusa, J. P.; Hernandez, G.; Gauvin, D.; Wade, C.; Zhu, C.; Pai, M.; Marc Scanio,
M. J. C.; Shi, L.; Drizin, I.; Gregg, R. J.; Matulenko, M. A.; Ahmed Hakeem, A. H.;
Gross, M. F.; Johnson, M. S.; Marsh, K. C.; Wagoner, K.; Sullivan, J. P.; Faltynek, C.
R.; Krafte, D. S. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 8520.
9. McGaraughty, S.; Chu, K. L.; Scanio, M. J. C.; Kort, M. E.; Faltynek, C. R.; Jarvis, M.
F. J. Pharmacol. Exp. Ther. 2008, 324, 1204.
10. 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.
11. Schnurch, M.; Flasik, R.; Khan, A. F.; Spina, M.; Mihovilovic, M. D.; Stanetty, P.
Eur. J. Org. Chem. 2006, 15, 3283.
12. For an excellent recent review, see: Nicolaou, K. C.; Bulger, P. G.; Sarlah, D.
Angew. Chem., Int. Ed. 2005, 44, 4442.
13. For ease of comparison, the indicated IC50 values are estimates calculated from
electrophysiology data (À40 mV for Nav1.8 and TTx-r, À60 mV for Nav1.2,
À90 mV for Nav1.5) obtained at multiple (n = 2–5) testing concentrations using
the following equation: (tested concentration,
lM) Â [(100 À % inhibition)/(%
inhibition)]. Inhibition values <20% and >80% were excluded from the
*
calculation unless indicated otherwise ( in the Table).
14. (a) Saari, W. S.; Halczenko, W.; King, S. W.; Huff, J. R.; Guare, J. P., Jr.; Hunt, C. A.;
Randall, W. C.; Anderson, P. S.; Lotti, V. J.; Taylor, D. A.; Clineschmidt, B. V. J.
Med. Chem. 1983, 26, 1696; (b) Romero, D. L.; Morge, R. A.; Biles, C.; Berrios-
Pena, N.; May, P. D.; Palmer, J. R.; Johnson, P. D.; Smith, H. W.; Busso, M.; Tan, C.
K.; Voorman, R. L.; Reusser, F.; Altaus, I. W.; Downey, K. M.; So, A. G.; Resnick,
L.; Tarpley, W. G.; Aristoff, P. A. J. Med. Chem. 1994, 37, 999.
References and notes
1. Catterall, W. A. Neuron 2000, 26, 13.
2. Lai, J.; Porreca, F.; Hunter, J. C.; Gold, M. S. Annu. Rev. Pharmacol. Toxicol. 2004,
44, 371.
3. Wood, J. N.; Boorman, J. Curr. Top. Med. Chem. 2005, 5, 529.
4. Catterall, W. A.; Goldin, A. L.; Waxman, S. G. Pharmacol. Rev. 2005, 57, 397.
15. Le Nguyen, D.; Castro, B. Peptide Chem. 1988, 231.
16. Kim, S. H.; Chung, J. M. Pain 1992, 50, 355.