The discovery of benzoxazine sulfonamide inhibitors of NaV1.7: Tools that bridge efficacy and target engagement

Article abstract of DOI:10.1016/j.bmcl.2017.05.070

The voltage-gated sodium channel Na_V1.7 has received much attention from the scientific community due to compelling human genetic data linking gain- and loss-of-function mutations to pain phenotypes. Despite this genetic validation of Na_V1.7 as a target for pain, high quality pharmacological tools facilitate further understanding of target biology, establishment of target coverage requirements and subsequent progression into the clinic. Within the sulfonamide class of inhibitors, reduced potency on rat Na_V1.7 versus human Na_V1.7 was observed, rendering in vivo rat pharmacology studies challenging. Herein, we report the discovery and optimization of novel benzoxazine sulfonamide inhibitors of human, rat and mouse Na_V1.7 which enabled pharmacological assessment in traditional behavioral rodent models of pain and in turn, established a connection between formalin-induced pain and histamine-induced pruritus in mice. The latter represents a simple and efficient means of measuring target engagement.

Full text of DOI:10.1016/j.bmcl.2017.05.070

Accepted Manuscript
The Discovery of Benzoxazine Sulfonamide Inhibitors of NaV1.7: Tools that
Bridge Efficacy and Target Engagement
Daniel S. La, Emily A. Peterson, Christiane Bode, Alessandro A. Boezio,
Howard Bregman, Margaret Y. Chu-Moyer, James Coats, Erin F. DiMauro,
Thomas A. Dineen, Bingfan Du, Hua Gao, Russell Graceffa, Hakan Gunaydin,
Angel Guzman-Perez, Robert Fremeau Jr., Xin Huang, Christopher Ilch,
Thomas J. Kornecook, Charles Kreiman, Joseph Ligutti, Min-Hwa Jasmine Lin,
Jeff S. McDermott, Isaac Marx, David J. Matson, Stefan I. McDonough, Bryan
D. Moyer, Hanh Nho Nguyen, Kristin Taborn, Violeta Yu, Matthew M. Weiss
PII:
S0960-894X(17)30568-1
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.05.070
BMCL 25015
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
19 April 2017
22 May 2017
23 May 2017
Please cite this article as: La, D.S., Peterson, E.A., Bode, C., Boezio, A.A., Bregman, H., Chu-Moyer, M.Y., Coats,
J., DiMauro, E.F., Dineen, T.A., Du, B., Gao, H., Graceffa, R., Gunaydin, H., Guzman-Perez, A., Fremeau, R. Jr.,
Huang, X., Ilch, C., Kornecook, T.J., Kreiman, C., Ligutti, J., Jasmine Lin, M-H., McDermott, J.S., Marx, I., Matson,
D.J., McDonough, S.I., Moyer, B.D., Nho Nguyen, H., Taborn, K., Yu, V., Weiss, M.M., The Discovery of
Benzoxazine Sulfonamide Inhibitors of NaV1.7: Tools that Bridge Efficacy and Target Engagement, Bioorganic &
Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.05.070
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Discovery of Benzoxazine Sulfonamide Inhibitors of
NaV1.7: Tools that Bridge Efficacy and Target
Engagement
Daniel S. La,*a Emily A. Peterson,*a Christiane Bode,a Alessandro A. Boezio,a Howard Bregman,a Margaret
Y. Chu-Moyer,a James Coats,a Erin F. DiMauro,a Thomas A. Dineen,a Bingfan Du,a Hua Gao,b Russell
Graceffa,a Hakan Gunaydin,b Angel Guzman-Perez,a Robert Fremeau Jr.,d Xin Huang,b Christopher Ilch, d
Thomas J. Kornecook, e Charles Kreiman, a Joseph Ligutti, ǁǁ Min-Hwa Jasmine Lin,c Jeff S. McDermott,d
Isaac Marx, a David J. Matson, d Stefan I. McDonough,d Bryan D. Moyer,e Hanh Nho Nguyen,a Kristin
Taborn,d Violeta Yu, d Matthew M. Weiss a
aDepartment of Medicinal Chemistry, bDepartment of Molecular Engineering, cDepartment of Pharmacokinetics
and Drug Metabolism, and dDepartment of Neuroscience, Amgen Inc., 360 Binney Street, Cambridge,
Massachusetts 02142, and eDepartment of Neuroscience, Amgen Inc., One Amgen Center Drive, Thousand
Oaks, California 91320, United States
Abstract
The voltage-gated sodium channel NaV1.7 has received much attention from the scientific community due to
compelling human genetic data linking gain- and loss-of-function mutations to pain phenotypes. Despite this
genetic validation of NaV1.7 as a target for pain, high quality pharmacological tools facilitate further
understanding of target biology, establishment of target coverage requirements and subsequent progression into
the clinic. Within the sulfonamide class of inhibitors, reduced potency on rat NaV1.7 versus human NaV1.7 was
observed, rendering in vivo rat pharmacology studies challenging. Herein, we report the discovery and
optimization of novel benzoxazine sulfonamide inhibitors of human, rat and mouse NaV1.7 which enabled
pharmacological assessment in traditional behavioral rodent models of pain and in turn, established a
connection between formalin-induced pain and histamine-induced pruritus in mice. The latter represents a
simple and efficient means of measuring target engagement.

______________________________________________________________________________________________
Chronic pain is a significant unmet medical need that remains unaddressed by current therapies which
often do not provide adequate pain relief and carry significant undesirable side effects. It has been shown that
humans with gain-of-function mutations in the gene that encodes for the voltage-gated sodium channel NaV1.7,
SCN9A, experience enhanced pain sensation whereas loss-of-function mutations lead to the inability to sense
pain.1 Accordingly, NaV1.7 has become a highly pursued target within the biopharmaceutical industry as
human genetics provide a compelling rationale that target engagement with an inhibitor would translate to pain
relief in the clinic.
NaV1.7 is a transmembrane (TM) voltage-gated sodium channel consisting of a 260 kDa pore-forming
subunit. NaV1.7 consists of 24 transmembrane helices over four domains (DI-DIV), in which each domain is
comprised of six TM segments (S1-S6). The S5-S6 segments from each domain create the pore module of the
channel and the S1-S4 segments form the voltage-sensing domain (VSD). A recent report from Genentech
revealing an X-ray crystal structure of a sulfonamide small molecule ligand bound to a human NaV1.7 bacterial
NaVAb channel chimera clearly pinpointed the binding site of this chemical class in the external cleft of
VSDIV.2 This structural elucidation is consistent with chimera studies from Pfizer and Amgen which also map
the binding site of sulfonamide inhibitors to the VSDIV.3 This class of sulfonamide inhibitors is of particular
importance because it is one of the few chemical series that has been reported to consistently demonstrate high
selectivity over NaV isoforms, particularly, NaV1.5, which is expressed in cardiac muscle and presents a
significant off-target liability. Despite this inherent selectivity advantage, optimization of pharmacokinetics to
afford adequate in vivo tools to evaluate target engagement and efficacy has been challenging. In this report, we
highlight a benzoxazine sulfonamide inhibitor as a tool compound that show efficacy in both rat and mouse
models of formalin-induced pain as well as a histamine-induced pruritus model in mice. With a single
molecule, we thereby demonstrate NaV1.7 target engagement in vivo and further support a link between
behavior models of pain and itch. Furthermore, this finding enables the field, with greater confidence, to take

advantage of the histamine model as a measure of target of engagement which was been the cornerstone of our
in vivo assessment of compounds in a practical and timely fashion.
We and others have previously reported NaV1.7 inhibitors from several different structural classes4 and
more recently we have focused on acylsulfonamides5 and bicyclic sulfonamides such as 1 (Scheme 1).6 Early
bicyclic sulfonamide inhibitors, whilst selective over NaV1.5, have suffered from poor pharmacokinetic
properties, demonstrating poor passive permeability and high rat in vivo clearance (Table 1).7 Attempts to
validate early leads from this class in rodent pain models, such as the rat formalin model, were unsuccessful due
to weak inhibition of rat NaV1.7 and the inability to achieve adequate target coverage.
Scheme 1. Progression of bicyclic sulfonamide scaffolds
In addition to the lack of in vivo efficacy in rat for compounds such as 1, the high lipophilicity and planar,
aromatic nature of these molecules were also concerning, as both of these parameters often correlate with higher
attrition.8 In an effort to enter more “drug-like” chemical space, benzoxazine 2 was synthesized (Table 1). This
new bicyclic scaffold had reduced lipophilicity and planarity, maintained modest NaV1.7 potency and while it
demonstrated reduced levels of selectivity over NaV1.5 it did exhibit slightly improved unbound clearance
(CLu) in rat. Similar to naphthalene 1, benzoxazine 2 demonstrated poor passive permeability, presumably due
to the basic amine present in the tetrahydropyridine and the resulting zwitterionic nature. A systematic
scanning with chemical libraries of 2,4-disubstituted aryl C-rings led to the removal of the basic amine and

identification of more efficient analogs such as 3 with decreased lipophilicity, logD and molecular weight with
respect to 1.9 Compound 3 had improved permeability and PK properties albeit with moderately decreased
potency on human Nav1.7.10 A hypothesis detailing differences in residues about the binding pocket will be
provided to account for the observed reduced potency on rat Nav1.5 (vide infra).
Table 1. Improved permeability and PK as a result of truncated benzoxazine core
Nav1.5 PX
hNav1.7
PX IC50
(µM)
rNav1.7
PX IC50
(µM)
Rat IV CL
[CLu] a
(L/hr/kg)
Papp
(µcm/s) b
IC50 (µM)
[fold over
Nav1.7]
>30
[>330x]
1.5 [11X]
4.6 [10x]
cLog D
(pH7.4)
LLE
2.0
MW
511
Compound
0.090
>30
1.7 [40]
1.3
0.69
1
0.14
0.46
>30
3.0
2.6 [14]
0.17 [7]
1.0
9.7
-1.4
-0.2
2.6
2.5
518
461
2
3
a Clearance unbound was calculated by dividing rat clearance by unbound fraction as determined by equilibrium
dialysis. 0.5 mg/kg as a solution in DMSO to male rats. CLu = clearance unbound. b Papp = passive
permeability, LLE = ligand lipophilicity efficiency. MW = molecular weight. MDCK vector control, (A→B +
B→A) /2.
In an effort to improve the potency of benzoxazine 3, different heterocyclic sulfonamide A-rings were
investigated within the context of the newly discovered benzoxazine core (Table 2). Although 1,2,4-
thiadiazole 4 was the most potent inhibitor of NaV1.7, permeability decreased with this heterocycle. 4-
Amino-thiazole 7 demonstrated the best passive permeability of the heterocycles examined, but the potency
was reduced in relation to 3 and many of the other heterocycles evaluated. With the exception of compound
7 all of the compounds demonstrated low turnover in both human and rat liver microsomes (HLM/RLM).
Other five-membered ring heterocycles, represented by 2-aminooxazole 6 and 5-aminoisoxazole 8, also
resulted in significantly diminished potency and were not investigated further. In addition, the pKa of the
sulfonamide nitrogen and therefore the population existing in the anionic state responsible for binding did
not correlate with NaV1.7 inhibition as the experimental pKa’s for 3, 4, and 5 were 5.8, 3.8, and 7.0
respectively.11 Despite it’s potential for the formation of reactive metabolites, the 2-aminothiazole 5
provided the best balance of permeability, unbound clearance and potency on human NaV1.7 and was
therefore, kept constant for subsequent optimization.12 The identification of a suitable replacement for the

2-aminothiazole, which lacked the inherent reactivity associated with this moiety, provided a significant
challenge that could not be overcome within the context of this particular class of molecules.
Table 2. Exploration of the heterocyclic A-ring
Rat IV CL
[CLu]
(L/hr/kg)b
hNav1.7 PX
IC50 (µM)
RLM / HLM
(µL/(min•mg))
Ave Papp
(µcm/s)a
Compound
A-ring
0.46
<14 / <14
9.7
0.17 [7]
3
0.09
0.19
2.6
<14 / 34
<14 / <14
15 / <14
55 / <14
1.6
22
0.64 [71]
0.33 [31]
ND
4
5
6
7
6.4
39
2.8
ND
9.2
32 / <14
<1
ND
8
a Papp = passive permeability, MDCK vector control, (A→B + B→A) /2. b Clearance unbound was calculated by
dividing rat clearance by unbound fraction as determined by equilibrium dialysis. 0.5 mg/kg as a solution in
DMSO to male rats. CLu = clearance unbound.
Efforts next shifted to the C-ring portion of the molecule (Table 3). Using insight from previously reported
SAR,6 the majority of the substitutions explored were centered upon the 2- and 4-positions of the C-ring. With

the aim of identifying compounds with suitable pharmacokinetics to enable oral dosing in appropriate
pharmacodynamics and/or efficacy models, attention focused on improving the potency while maintaining low
turnover in liver microsomes and good passive permeability. The most potent human Nav1.7 inhibitors
contained lipophilic groups at C4, such as CF3 (see 5 and 9). The 2-position of the C-ring was tolerant of a wide
variety of substitution and it was found that a methoxy group provided a good balance across the parameters
being evaluated. In an attempt to mitigate the higher log D, the ortho-OMe group was paired with either a
nitrile, a hydrogen or a fluorine at the para position to give 10 – 12. These compounds demonstrated the best
combination of potency on NaV1.7 and selectivity over NaV1.5. Unfortunately, 11 and 12 demonstrated high
turnover in both human and rat liver microsomes, thereby limiting their utility moving forward.
Table 3. Exploration of C-Ring SAR
HLM /
Nav1.5 PX Rat IV CL
[CLu]
hNav1.7 PX
IC50 (µM)
RLM
(µL/(min•
mg))
Ave Papp
(µcm/s)b
rNav1.7 PX
IC50 (µM)
Compound
R1
R2
IC50 (µM)
[fold]
(L/hr/kg) a
CN
OMe
OMe
CF3
CF3
CN
0.19
0.12
0.25
1.1
1.5 [8]
2.0 [17]
14 [56]
0.33 [31]
0.24 [40]
0.84 [53]
<14 / <14
30 / 53
36 / 22
22
32
20
5
9
0.39
0.85
10
OMe
OMe
H
F
0.38
0.28
1.6
>30 [>79]
ND
ND
68 / 297
28
36
11
12
0.88
>30 [>100]
110 / >399
a Clearance unbound was calculated by dividing rat clearance by unbound fraction as determined by equilibrium
dialysis. 0.5 mg/kg as a solution in DMSO to male rats. CLu = clearance unbound. b Papp = passive
permeability, MDCK vector control, (A→B + B→A) /2.

Keeping the thiazole constant as the A-ring, susbstitution of the B-ring was also explored (Table 4).
Despite the greater NaV1.5 selectivity of 10, the trifluormethyl C-ring substituent of 5 was chosen for further
exploration given the greater body of historical data that existed with the 2-cyano-4-trifluromethyl C-ring
which allowed for ease of comparison between analogs. Substitution of the aromatic portion of the
benzoxazine B-ring revealed that although small electron-withdrawing groups could be tolerated, they did
not dramatically alter potency, intrinsic stability (HLM / RLM) or the passive permeability. The position
most tolerant to fluoro-substitution was R6 (compare 13, 14 and 15). Within this series, the R6 fluoro and
chloro-analogs (15 and 16) led to diminished selectivity over NaV1.5 as compared to compound 5.
Furthermore, as selectivity over NaV1.5 eroded, the potency on rat NaV1.7 was improved. Substitution of the
sp3-hybridized carbons linking the oxygen and nitrogen atoms of the benzoxazine ring was also investigated
with groups such as methyl, ethyl, phenyl and carbonyl with no observed improvement in either potency or
selectivity (data not shown).
Table 4. Modification of the benzoxazine B-ring
Compound
R5 R6 R8
hNav1.7 PX
IC50 (µM)
0.19
rNav1.7 PX
IC50 (µM)
1.1
Nav1.5 PX
IC50 (µM)
1.5
Ave Papp
(µcm/s)a
HLM / RLM
(µL/(min•mg))
<14 / <14
<14 / <14
<14 / <14
<14 / <14
22 / 21
H
H
F
H
H
H
F
H
F
22
38
26
19
27
5
0.52
2.0
1.7
0.56
2.1
0.65
13
14
15
16
H
H
H
H
H
0.14
0.39
0.067
0.05
0.32
0.61
Cl
a Papp = passive permeability, MDCK vector control, (A→B + B→A) /2.
The recent advancement in NaV1.7 structural biology enables a closer examination of ligand/protein
interaction, providing insight into the origins of potency and varying selectivity over rNaV1.7.2 Genentech

obtained an X-ray structure of 17 (GX-936, Figure 1A)13 bound to a Nav1.7 bacterial NaVAb channel chimera,
clearly depicting the ligand/protein interactions within the VSDIV pocket. Binding of our sulfonamide
benzoxazine inhibitors to this pocket was confirmed by a hNaV1.7 radioligand binding assay employing a
tritiated ligand14 that is routinely displaced by compounds highly homologous to 17. Employing this assay,
bicyclic sulfonamides 1, 4, 15 and 16 were determined to have IC50 values equal to 0.030, 0.021, 0.037 and
0.020 µM respectively, suggesting that the binding site for these compounds and 17 is the same.5 This data, in
turn, enabled computational docking to understand the critical protein interactions with the benzoxazines.15
Compounds 1, 2 and 3 in their ionized states were docked into the VSDIV binding site of the reported structure.
The acidic sulfonamide moiety engages Arg1602, Arg1608 and Asn1540 in an H-bonding network while
additionally participating in a charge-charge interaction with Arg 1605 and Arg1608 as shown in the interaction
map of Figure 1C and 3-dimensionally in Figure 2A. All four residues are conserved between hNaV1.7,
rNaV1.7 and hNaV1.5 (Table 5 in blue). Critical to this series of key interactions is the two-point contact
between Arg1608 and both the deprotonated sulfonamide nitrogen and the adjacent heterocyclic A-ring nitrogen
(N at the 3-position on 2 see Figure 2B). Consistent with this sulfonamide binding mode, potency is lost with
compound 8 where the 2-pyridyl A-ring is replaced by a 5-aminoisoxazole ring (N vs. O H-bond acceptor).
Figure 1. (A) Compound 17, used by Genentech as a ligand for X-ray crystallography. (B) Docking with
compound 2 performed in the ionized state. (C) Compound 2-Protein interaction map. Pink circle: charged or
H-bond interaction. Green circle: hydrophobic interaction. Purple dotted line: charge-charge interaction. Green
dotted line: H-bond

Despite the high global homology between hNaV1.7 and rNaV1.7 (93%),16 the selectivity of certain
sulfonamides over rNaV1.7 can be rationalized by the differences in residues surrounding the substituent at the
ortho-position of the C-ring. For compounds 1 and 2, no rat activity is observed (rNaV1.7 IC50 >30 µM).
Glu1534 (Figures 1C and 2C), which is unique to hNaV1.7 (Table 5), engages the basic tetrahydropyridine
moiety. In contrast, rNaV1.7 contains a tyrosine at the same position which cannot participate in this strong
enthalpic interaction and may be electrostatically disfavored, consistent with the lack of rat activity. Proximal
to Glu1534 on the same S2 helix is Tyr1537 which engages the ligand in a π-stacking interaction as shown in
Figure 2C. Residues 1531 and 1533, which also differ amongst the proteins, reside on the same helix but do not
directly engage the ligand, suggesting that the local environment could also playing a role in the observed levels
of selectivity. rNaV1.7 potency increases when the basic tetrahydropyridine is replaced by a smaller nitrile (3).
The shorter nitrile of 3 can interact with Trp1538 (conserved in both human and rat) through either an H-bond
with the indole NH or a π-π interaction.
Table 5. Selected residues from sequences alignments between hNav1.7 and rNav1.7. a
1531
His
1533
Thr
1534
Glu
1537
Tyr
1538
Trp
1540
Asn
1541
Val
1587
Leu
1602
Arg
1605
Arg
1608
Arg
hNav1.7
rNav1.7
Tyr
Asp
Tyr
His
Trp
Asn
Met
Met
Arg
Arg
Arg
a Residues are highlighted in blue (conserved), red (directly interact with ligand) and black (further
removed from ligand region)
The effect of halogenation of the benzoxazine core on enhancing rNaV1.7 activity highlighted in Table 4
can be explained by the residue differences at position 1541 (Figure 2D and Table 5). Within range of R6 (15
and 16, Table 4) Met1541 (in rat) extends beyond a shorter valine (human) and as a result, it could engage in
either an enhanced van der Waals (VDW) contact or a halogen-bonding interaction. The VDW contact is
consistent with the tolerance of the adjacent R5 fluorine on compound 14 which is also proximal to Met1541.

The majority of residue differences residing proximal to the ortho-substituent on the C-ring highly
implicates this region of the protein as the main driver of selectivity profile. These explanations for increased
rNaV1.7 potency, however, do not account for the changes in helix movement as these arguments are primarily
rooted in residue differences in the DIV pocket that are in direct contact with the ligands. Despite the high
sequence homology for human and rat, especially outside the immediate binding pocket, discrepancies at
remote sites may also govern selectivity profile. For instance, Leu1587 is located on the S6 helix and remote
form the binding site. This distal position is in stark contrast to the two other residues at 1531 and 1533 which
are directly adjacent to those that make direct contact with the ligand (outer sphere residues are highlighted in
black in Table 5.). While it is less clear how remote residue differences effect binding and protein movement,
this analysis of proximal residue differences provides a rationale for binding and selectivity which in turn could
enhance prospective design.

Figure 2. (A) Critical binding interaction between the sulfonamide moiety of compound 2 and surrounding
charged arginine residues. (B) Critical 2-point interaction between Arg1608 and the sulfonamide moiety of
compound 2 (C) D-Ring interactions with residues that govern rat and Nav1.5 selectivity. (D) Met1541 in rat
(Val in human) is poised to interact with the R5 and R6 halogen.
Compounds 15 and 16 proved unique within the sulfonamide class due to their cross-species inhibition
of Nav1.7 and their ability to interrogate in vivo pharmacology associated with inhibitors that act through
binding with the DIV pocket. As mentioned previously, inhibitors with aromatic bicyclic cores such as 1 did not
show efficacy in the rat formalin pain model. Because of the significant homology between the human and rat
NaV1.7 channels (93% homology), it was anticipated that the sulfonamides would demonstrate similar potency
for the inhibition of human and rat NaV1.7; it was subsequently found that this was not the case. While the
benzoxazines had diminished selectivity over NaV1.5, the observation that these compounds demonstrated
robust inhibition of rat Nav1.7 was an unexpected and useful discovery because of the scaffold’s potential to
deliver valuable in vivo tools.
Although several rodent models of nociceptive pain are commonly used to assess analgesic efficacy,17
our own prior experience,4 as well as other reports of NaV1.7 inhibitors blocking pain response in the rat
formalin model,18 prompted us to investigate a representative compound in the formalin assay.19 Compound 15
was chosen to assess the efficacy of a sulfonamide NaV1.7 inhibitor in a rat model of formalin-induced pain due
to its potency on rat NaV1.7 and its pharmacokinetic profile in rat (10 mg/kg PO dose: Cmax = 0.99 µM AUC =
20 µM•h, %F = 57, 0.5 mg/kg IV dose: CL [CLu] = 0.6 [39] L/h/kg, Fu = 0.015).20 Compound 15 showed dose-
and exposure-dependent efficacy in the formalin model, with statistically significant analgesic effects at 30 and
100 mg/kg (Figure 3A). The 30 mg/kg and 100 mg/kg doses produced 39 and 70% reductions in nociceptive
behavior, respectively with unbound plasma coverage of 1.3 and 3.6-fold over the rat NaV1.7 IC50 (See Table
6). We considered it very unlikely that formalin responses were due to inhibition of NaV1.8, a tetrodotoxin-
resistant (TTX-R) Nav isoform implicated in nociception and action potential electrogenesis,21 since the IC50 for
15 to block TTX-R currents in rat dorsal root ganglion (DRG) neurons was >30 µM, when evaluated on closed
or partially-inactivated channels, a concentration well above free drug concentrations achieved in our in vivo

studies. Notably, a previously reported NaV1.8 inhibitor was found to be efficacious in models of neuropathic
and inflammatory pain, but ineffective in a model of formalin-induced nociception. 22
The difference in sulfonamide potency between human and rat NaV1.7 prompted us to test potency
across a broader species panel. Similar to a report published from Pfizer demonstrating the species- and
isoform-selectivity of sulfonamide 18 (PF-04856264),13 compound 1, which was selective over human NaV1.5,
potently inhibited NaV1.7 channels from cynomolgus monkey and mouse, but not rat (Table 6).23 Compounds
15 and 16 demonstrated potent inhibition of both mouse and monkey NaV1.7 channels and this observation
enabled establishing a pharmacodynamic model in mice. In addition, compound 15 was further investigated in a
mouse TTX-sensitive dorsal root ganglion electrophysiology assay and was found to be potent (IC50 of 53 nM).
This potency correlates well with the mouse NaV1.7 potency determined in our in vitro HEK assay and further
reinforces the connectivity between mouse, rat and human activity. Since compound 15 demonstrated
statistically significant analgesia in the rat formalin model, it was postulated that this molecule could be used to
validate a similar pain model in mouse.
Table 6. Isoform and species selectivity profiles for selected compounds
human Nav1.7
IC50 (µM)
0.20 / 0.028a
0.09
cyno Nav1.7 IC50
mouse Nav1.7
IC50 (µM)
0.13a
rat Nav1.7 IC50
(µM)
human Nav1.5
IC50 (µM)
>30 / >10a
>30
Compound
(µM)
0.02a
0.10
0.37
0.30
4.2a
>30
18
4.3
0.06
1
0.14
0.39
0.07
0.05
0.33
0.61
15
16
0.10
aPublished data
When dosed orally in mice, 15 provided an acceptable pharmacokinetic profile (Cmax = 58 µM, AUC =
790 µM•h, t1/2=7 h and Fu = 0.006) and it was anticipated that sufficient unbound plasma concentrations (Cu)
could be achieved in order to enable informative in vivo pharmacology studies in mice (see Table 7.). To
determine its analgesic effect in mice, 15 was dosed at 30, 60, and 100 mg/kg in C57BL/6 male mice and
nociceptive (flinching and licking) behavior was measured after subcutaneous injection of formalin (2%) into
the dorsal hind paw (Figure 3B). It was found that 15 demonstrated robust inhibition of formalin-induced pain

behavior in a dose-dependent fashion, with statistically significant effects at the 60 mg/kg (45% MPE) and 100
mg/kg (77% MPE) doses. The free plasma concentration multiple over mouse NaV1.7 IC50 (Cu/m NaV1.7 IC50)
required to elicit an analgesic effect in mouse formalin (16x) was slightly higher than that required to elicit a
similar response it the rat formalin model (3.6x). The reasons for this modest difference in coverage
requirements is currently unclear. (Figure 3A, Table 8). These results provided an important proof-of-concept
for establishing a measure of efficacy in mouse for evaluating selective sulfonamide inhibitors.
Table 7. Rat and mouse PK of 15
Plasma
Species
Cmax
Tmax F (%)
(h)
Route
(mg/kg)
AUCinf
(µM•h)
CL
(L/h/kg)
Vd
(L/kg)
protein
binding
(%)
T1/2 (h)
(µM)
Rat
Rat
i.v. (0.5)
p.o. (10)
p.o. (30)
1.8
20
0.60
3.6
9.8
7.1
2.5
98.5
98.5
0.95
58
6.7
2.0
57
mouse
790
99.4
Despite success in demonstrating that the mouse formalin model is sensitive in detecting analgesic
activity of our NaV1.7 inhibitors, the labor-intensive nature of manually assessing formalin-induced pain
behavior in mice led us to pursue a more sensitive and higher throughput pharmacodynamic model. In addition
to its role in sensing pain stimuli, NaV1.7 has also been implicated in the perception of itch. Some individuals
with congenital insensitivity to pain (CIP) are also insensitive to itch, while others with gain-of-function
mutations suffer from neuropathic itch.24 Accordingly, we have found global NaV1.7 knockout mice
recapitulate the human CIP phenotype. In addition to the lack of response to chemical, mechanical and thermal
pain stimuli, these knockout mice also exhibit no scratching response to an intra-dermal injection of
histamine.25
Due to the underlying commonality between the neural anatomy responsible for conveying itch and pain
information, it was hypothesized that a rodent itch model could provide a reliable means of estimating NaV1.7
target engagement while also improving project process flow.26 Toward this end, C57BL/6 male mice were
dosed orally with 30, 60 or 100 mg/kg of 15 120 minutes prior to receiving a challenge dose of histamine, a

known pruritogen. Histamine was administered intradermally to the nape of the neck immediately prior to
testing and the number of resulting scratch bouts was measured over a 0-30 minute period. Gratifyingly, all
three doses of 15 provided a statistically significant reduction in itch response compared to vehicle (Figure
3C).27 The unbound concentration of the 30, 60 and 100 mpk doses were 0.37, 0.66 and 0.68 µM respectively,
which corresponds to multiples of 6 – 11-fold over the mouse NaV1.7 IC50 (Table 8). This data is consistent
with that obtained in the mouse formalin assay in which the efficacious concentration at the 60 mg/kg dose
(0.62 µM) was 10-fold over IC50. The similarity in target coverage between the two models helped confirm for
the first time in our program that a histamine-induced pruritis assay could be used as a model to assess target
engagement. In turn, the development of this assay significantly streamlined in vivo assessment of future
compounds that will be disclosed shortly.

Figure 3. (A) Total phase II formalin-induced nociceptive behavior (flinches) in Sprague Dawley rats
administered vehicle, morphine sulfate (subcutaneous dosing, 2 mg/kg), or 10, 30, or 100 mg/kg of 15 (PO,
formulation: 30% (w/v) hydroxypropyl β-cyclodextrin in water pH adjusted to 10.0 with KOH). Unbound
concentrations are expressed above bars. (B) Total phase II formalin-induced nociceptive behavior (paw licking
and lifting) in C57Bl/6 mice administered vehicle, morphine sulfate (subcutaneous dosing, 3 mg/kg), or 30, 60,
or 100 mg/kg of 15 (PO, formulation: 30% (w/v) hydroxypropyl β-cyclodextrin in water pH adjusted to 10.0
with KOH). Unbound concentrations are expressed above bars. (C) Effect of 15 on histamine-induced pruritus
in C57BL/6 male mice administered vehicle, diphenhydramine (30 mg/kg; 90 min prior to test, PO,
formulation: water), 30, 60, or 100 mg/kg of 15 (120 min prior to test, PO, formulation: 30% (w/v)
hydroxypropyl β-cyclodextrin in water pH adjusted to 10.0 with KOH); y-axis: total number of scratch bouts
over 30-min time period following histamine administration (8.15mM i.d.). Unbound concentrations are
expressed above bars.
Table 8. Plasma exposure and unbound concentration target coverage
Mean Plasma
Conc. (µM)
Cu/ rodent
NaV1.7 IC50
Effect
(%)
17
In Vivo Assay
Rat Formalin
Dose (mpk)
10
30
1.7
4.2
11
0.5
1.3
3.6
5
39
70
100
30
51
5
45
Mouse
Formalin
60
100
150
61
10
16
6
100
30
77
37
Mouse
Histamine
Pruritis
60
100
109
113
10
11
71
80

The synthesis of key compounds 15 and 10 is illustrated below in Scheme 2. The syntheses of these
benzoxazine inhibitors can be achieved by two complementary routes that allow late-stage structural
modification of either the A- or C-rings. Route 1, which is exemplified by the synthesis of 15, starts with SNAr
arylation of benzoxazine 17 with an appropriately functionalized fluoroarene such as 18 followed by
chlorosulfonylation to provide sulfonyl chloride 20. Sulfonamide formation proceeds upon mixing the
heterocyclic amine 21 with the sulfonyl chloride and treating the mixture with LHMDS at 0 °C. Aqueous
workup and treatment with trifluoroacetic acid provides inhibitors such as 15.
Alternatively, for SAR exploration of the C-ring (Table 4), the synthesis using Route 2 started from
commercially available sulfonyl chloride 22, which after installation of the protected A-ring heterocyclic amine,
underwent hydrolysis and subsequent reduction to give phenolic aniline 24. Cyclization of 24 with
chloroacetylchloride (25) followed by borane reduction of the lactam provided benzoxazine 26, a useful late-
stage intermediate for C-ring diversification. Palladium-catalyzed amination of bromide 27 followed by
deprotection of the sulfonamide with trifluoroacetic acid afforded 10.

Scheme 2. Synthetic routes for the synthesis of benzoxazine Nav1.7 inhibitors. Route 1: (a) NaH, THF, 20,
0→60 °C, 63% (b) SO2Cl2, DCM, 0 °C, 75%; (c) i. LHMDS, THF, 23, 0 °C→rt; ii. TFA, DCM, rt, 56% for 2
steps. Route 2: (d) LHMDS, THF, 23, -78 °C, 73%; (e) i. NaOTMS, THF, 0 °C; ii. Fe, HOAc, THF, 70 °C,
92% yield over 2 steps; (f) i. 27, K2CO3, DMF, 0-50 °C, 84%; ii. BH3•THF, THF, 0 °C, 78%; (g) 29, XantPhos,
Pd2(dba)3, NaO-tBu, 130 °C, 56%.
In an attempt to improve upon the physical properties of the naphthalene–based Nav1.7 inhibitors
represented by compound 1, the benzoxazine class of inhibitors was synthesized to decrease planarity and
reduce lipohilicity. Optimization of this scaffold led to the identification of an appropriate A-ring for improved
permeability and in vivo clearance followed by recognition of a C-7 fluoride on the benzoxazine ring for
improved rat potency. This work ultimately led to the discovery of 15. The fortuitous rat potency and general
disparity between rat and human activity are rationalized by docking studies employing the reported chimera X-
ray structure. This in silico study highly implicates the region around the C-ring as critical to the selectivity
profile. Compound 15 had sufficient exposure and rodent potency to progress to in vivo studies and
demonstrate target coverage. 15 became a critical in vivo tool, for the first on our program, linking efficacy
between a traditional model of pain sensitivity (formalin) and the histamine-induced pruritus assay in mouse.
Although selectivity could be improved, the potential for cardiac toxicity due to NaV1.5 inhibition precluded
further investigation of this specific series for the development of a pain therapeutic. Nevertheless, the
sulfonamide series was further optimized to related selective structures which will be the subject of future
publications.
Supporting Information Available: In vitro and in vivo pharmacology assay protocols, PKDM assay protocols,
synthetic experimental procedures, compound characterization.
Corresponding Author Information: Phone: 617-949-4131. Fax: 617-299-8379. E-mail: Daniel.la@sagerx.com
Author Information: Emily Peterson, Biogen Inc. Howard Bregman, Agios Pharmaceuticals. Erin DiMauro,
Merck Inc. Thomas Dineen, Novartis AG. Bingfan Du, Mersana Therapeutics. Hakan Gunaydin, Merck Inc.

Robert Fremeau, NeuroRx Consulting. Jeff McDermott, Eli Lilly and Co. Isaac Marx, Biogen Inc. Stefan
McDonough, Pfizer Inc. Hanh Nguyen, BlinkBio Inc.
Abbreviations Used: NaV1.7, voltage gated sodium channel 1.7; SCN9A, sodium valtage-gated channel alpha
subunit 9; TM, transmembrane; VSD, voltage sensing domain; CLu, unbound clearance; HLM, human liver
microsome; RLM, rat liver microsome; PX, patch express; LLE, ligand lipophilicity efficiency. MW,
molecular weight. MDCK, Madin-Darby canine kidney epithelial; SAR, structure activity relationship; VDW,
van der Waals; MPE, maximal pharmacological effect; CIP, congenital insensitivity to pain
References and Notes
(1) (a) Faber, C. G.; Hoeijmakers, J. G. J.; Ahn, H.-S.; Checng, X.; Han, C.; Choi, J.-S.; Estacion, M.; Lauria,
G.; Vanhoutte, E. K. Gerrits, M. M.; Dib-Hajj, S.; Drenth, J. P. H.; Waxman, S. G.; Merkies, I. S. J. Gain of
function Nav1.7 mutations in idiopathic small fiber neuropathy. Ann. Neurol. 2012, 71, 26–39; (b) Dib-Hajj, S.
D.; Binshtok, A. M.; Cummins, T. R.; Jarvis, M. F.; Samad, T.; Zimmermann, K. Voltage-gated soldium
channels in pain states: Role in pathophysiology and targets for treatment. Brain Res. Rev. 2009, 60, 65–83. (c)
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. An SCN9A channelopathy causes congenital inability to experience
pain. Nature 2006, 444, 894–898. (d) 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. SCN9A Mutations in
paroxysmal clinical study extreme pain disorder: allelic variants underlie distinct channel defects and
phenotypes. Neuron 2006, 52, 767–774. (e) Yang, Y.; Wang, Y.; Li, S.; Xu, Z.; Li, H.; Ma, L.; Fan, J.; Bu, D.;
Liu, B.; Fan, Z.; Wu, G.; Jin, J.; Ding, B.; Zhu, X.; Shen, Y. Mutations in SCN9A, encoding a sodium channel
a subunit, in patients with primary erythermalgia. J. Med. Genet. 2004, 41, 171–174. (f) Fab.; Vanhoutte, E. K.;
Gerrits, M. M.; Dib-Haij, S.; Drenth, J. P.; Waxman, S G.; Merkies, I. S. Gain of function NaV1.7 mutations in
idiopathic small fiber neuropathy. Ann. Neurol. 2012, 71, 26–39.
(2) Ahuja, S.; Mukund, S.; Deng, L.; Khakh, K.; Chang, E.; Ho, H.; Shriver, S.; Young, C.; Lin, S.; Johnson, J.
P., Jr.; Wu, P.; Li, J.; Coons, M.; Tam, C.; Brillantes, B.; Sampang, H.; Mortara, K.; Bowman, K. K.; Clark, K.
R.; Estevez, A.; Xie, Z.; Verschoof, H.; Grimwood, M.; Dehnhardt, C.; Andrez, J. C.; Focken, T.; Sutherlin, D.

P.; Safina, B. S.; Starovasnik, M. A.; Ortwine, D. F.; Franke, Y.; Cohen, C. J.; Hackos, D. H.; Koth, C. M.;
Payandeh, J. Structural basis of Nav1.7 inhibition by an isoform-selective small-molecule antagonist. Science
2015, 350 , aac5464.
(3) McCormick, K.; Santos, S.; Chapman, M.L.; Krafte, D.S. Marron, B.E.; West, C.W.; Krambis, M.J.;
Antonio, B.M.; Zellmer, S.G.; Printzenhoff, D.; Padilla, K.M.; Lin, Z.; Wagoner, P.K.; Swain, N.A.; Stupple,
P.A.; de Groot, M.; Butt, R.P.; Castle, N.A. Voltage sensor interaction site for selective small molecule
inhibitors of voltage-gated sodium channels. PNAS, 2013, 110, 1-9.
(4) a) Bregman, H.; Berry, L.; Buchanan, J. L.; Chen, A.; Bu, B.; Feric, E.; Hierl, M.; Huang, L.; Immke, D.;
Janosky, B.; Johnson, D.; Li, X.; Liguitti, J.; Liu, D.; Malmberg, A.; Matson, D.; McDermott, J.; Miu, P.;
Nguyen, H. N.; Patel, V. F.; Waldon, D.; Wilenkin, B.; Zheng, X. M.; Zou, A.; McDonough, S. I.; DiMauro, E.
F. Identification of a potent, state-dependent inhibitor of Nav1.7 with oral efficacy in the formalin model of
persistent pain. J. Med. Chem. 2011, 54, 4427–4445. b) Chakka, N.; Bregman, H.; Du, B.; Nguyen, H. N.;
Buchanan, J. L.; Feric, E.; Ligutti, J.; Liu, D.; McDermott, J. S.; Zou, A.; McDonough, S. I.; Dimauro, E. F.
Discovery and hit-to-lead optimization of pyrrolopyrimidines as potent, state-dependent Nav1.7 antagonists.
Bioorg. Med. Chem. Lett. 2012, 22, 2052-2062. c) Nguyen, H. N.; Bregman, H.; Buchanan, J. L.; Du, B.; Feric,
E.; Huang, L.; Li, X.; Ligutti, J.; Liu, D.; Malmberg, A. B.; Matson, D. J.; McDermott, J. S.; Patel, V. F.;
Wilenkin, B.; Zou, A.; McDonough, S. I.; DiMauro, E. F. Discovery and optimization of aminopyrimidinones
as potent and state-dependent Nav1.7 antagonists. Bioorg. Med. Chem. Lett. 2012, 22, 1055-1060. d) Matson,
D. J.; Hamamoto, D. T.; Bregman, H.; Cooke, M.; DiMauro, E. F.; Huang, L.; Johnson, D.; Li, X.; McDermott,
J.; Morgan, C.; Wilenkin, B.; Malmberg, A. B.; McDonough, S. I.; Simone, D. A. Inhibition of inactive states of
tetrodotoxin-sensitive sodium channels reduces spontaneous firing of C-fiber nociceptors and produces
analgesia in formalin and complete freund's adjuvant models of pain. PloS one 2015, 10, e0138140. d) Focken,
T.; Liu, S.; Chahal, N.; Dauphinais, M.; Grimwood, M.E.; Chowdhury, S.; Hemeon, I.; Bichler, P.; Bogucki, D.;
Waldbrook, M.; Bankar, G.; Sojo, L.E.; Young, C.; Lin, S.; Shuart, N.; Kwan, R.; Pang, J.; Chang, J.H.; Safina,
B.S.; Sutherlin, D.P.; Johnson Jr. J.P.; Dehnhardt, C.M.; Mansour, T.S.; Oballa R.M.; Cohen, C.J.; Robinette.

C.L. Discovery of aryl sulfonamides as isoform-selective inhibitors of NaV1.7 with efficacy in rodent pain
models. ACS Med. Chem. Lett. 2016, 7, 277-282. e) Wu, Y.; Guernon, J.; Shi, J.; Ditta, J.; Robbins, K.J.;
Rajamani, R.; Easton, A.; Newton, A.; Bourin, C.; Mosure, K.; Soars, M.G.; Knox, R.J.; Matchett, M.; Pieschl,
R.L.; Post-Munson, D.J.; Wang, S.; Herrington, J.; Graef, J.; Newberry, K.; Bristow, L.J.; Meanwell, N.A.;
Olson, R.; Thompson, L.A.; Dzierba, C. Development of new benzenesulfonamides as potent and selective
Nav1.7 inhibitors for the treatment of pain. J. Med. Chem. 2017, 60, 2513-2525. f) Roecker, A.J.; Egbertson,
M.; Jones, K.L.G.; Gomez, R.; Kraus, R.L.; Li, Y.; Koser, A.J.; Urban, M.O.; Klein, R.; Clements, Daley, M.C.;
Wang, J.; Finger, E.N.; Majercak, J.; Santarelli, V.; Gregan, I.; Cato, M.; Filzen, T.; Jovanovska, A.; Wang,
Y.H.; Wang, D.; Joyce, L.A.; Sherer, E.C.; Peng, X.; Wang, X.; Sun, H.; Coleman, P.J.; Houghton, A.K.;
Layton, M.E. Discovery of selective, orally bioavailable, N-linked arylsulfonamide Nav1.7 inhibitors with pain
efficacy in mice. Bioorg. Med. Chem. Lett. 2017, Ahead of print. g) For recent work in the area of peptide
inhibition, please see: Murray, J.K.; Ligutti, J.; Liu, D.; Zou, A.; Poppe, L.; Li, H.; Andrews, K.L.; Moyer,
B.D.; McDonough, S.I.; Favreau, P.; Stocklin, R.; Miranda, L.P. Engineering Potent and Selective Analogues of
GpTx-1, a Tarantula Venom Peptide Antagonist of the NaV1.7 Sodium Channel J. Med. Chem. 2015, 58, 2299-
2314.
(5) DiMauro, E. F.; Altmann, S.; Berry, L. M.; Bregman, H.; Chakka, N.; Chu-Moyer, M.; Feric Bojic, E.; Foti,
R. S.; Fremeau, R. T., Jr.; Gao, H.; Gunaydin, H.; Guzman-Perez, A.; Hall, B. E.; Huang, H.; Jarosh, M.;
Kornecook, T.; Lee, J.; Ligutti, J.; Liu, D.; Moyer, B. D.; Ortuno, D.; Rose, P. E.; Schenkel, L. B.; Taborn, K.;
Wang, J.; Wang, Y.; Yu, V. L.; Weiss, M. M. Application of a parallel synthetic strategy in the discovery of
biaryl acyl sulfonamides as efficient and selective NaV1.7 inhibitors. J. Med. Chem. 2016, 59, 7818-7839.
(6) a) Marx, I.E.; Dineen, T.A.; Able, J.; Bode, C.; Bregman, H.; Chu-Moyer, M.; DiMauro, E.F.; Du, B.; Foti,
R.S.; Fremeau Jr., R.T.; Gao, H.; Gunaydin, H.; Hall, B.E.; Huang, L.; Kornecook, T.; Kreiman, C.R.; La,
D.S.; Ligutti, J.; Lin, M.J.; Liu, D.; McDermott, J.; Moyer, B.D.; Peterson, E.A.; Roberts, J.T.; Rose, P.; Wang,
J.; Youngblood, B.D.; Yu, V.; Weiss, M.M. Sulfonamides as selective NaV1.7 inhibitors: optimizing potency
and pharmacokinetics to enable in vivo target engagement. ACS Med. Chem. Lett. 2016, 7, 1062-1067. b)

Graceffa, R.F.; Boezio, A.A.; Able, J.; Altmann, S.; Berry, L.M.; Boezio, C.M.; Butler, J.R.; Chu-Moyer, M.Y.;
Cooke, M.; Di Mauro, E.F.; Dineen, T.A.; Feric Bojic, E.; Foti, R.S.; Fremeau, Jr., R.T.; Guzman-Perez, A.;
Gao, H.; Gunaydin, H.; Huang, H.; Huang, L.; Ilch, C.; Michael Jarosh, Thomas Kornecook, Kreiman, C.R.; La,
D.S.; Ligutti, J.; Milgram, B.C.; Lin, M.J.; Marx, I.E.; Nguyen, H.N.; Peterson, E.A.; Rescourio, G.; Roberts, J.;
Schenkel, L.B.; Shimanovich, R.; Sparling, B.A.; Stellwagen, J.; Taborn, K.; Vaida, K.R.; Wang, J.; Yeoman,
J.T.S.; Yu, V.L.; Zhu, D.; Moyer, B.D.; Weiss, M.M. Sulfonamides as Selective NaV1.7 Inhibitors: Optimizing
Potency, Pharmacokinetics, and Metabolic Properties to Obtain Atropisomeric Quinolinone (AM-0466) that
Affords Robust In Vivo Activity. J. Med. Chem. 2017, Ahead of print. c) Weiss, M.M.; Dineen, T.A.; Marx,
I.E.; Altmann, S.; Boezio, A.A.; Bregman, H.; Chu-Moyer, M.Y.; Di Mauro, E.F.; Feric Bojic, E.; Foti, R.S.;
Gao, H.; Graceffa, R.F.; Gunaydin, H.; Guzman-Perez, A.; Huang, H.; Huang, L.; Jarosh, M.; Kornecook, T.;
Kreiman, C.R.; Ligutti, J.; La, D.S.; Lin, M.J.; Liu, D.; Moyer, B.D.; Nguyen, H.N.; Peterson, E.A.; Rose, P.E.;
Taborn, K.; Youngblood, B.D.; Yu, V.L.; Fremeau, Jr., R.T. Sulfonamides as Selective NaV1.7 Inhibitors:
Optimizing Potency and Pharmacokinetics While Mitigating Metabolic Liabilities. J. Med. Chem. 2017, Ahead
of print.
(7) The vast majority of zwitterionic inhibitors were non-progressible due to poor pharmacokinetics.
Compound 1 was a unique exception to this general trend. See reference 6.
(8) a) Gleeson, M.P. Generation of a Set of Simple, Interpretable ADMET Rules of Thumb. J. Med. Chem.
2008, 51, 817-834. b) Lovering, F.; Bikker, J.; Humblet, C. Escape from flatland: increasing saturation as an
approach to improving clinical success. J. Med. Chem. 2009, 52 , 6752-6756. c) Ritchie, T. J.; Macdonald, S. J.
The impact of aromatic ring count on compound developability--are too many aromatic rings a liability in drug
design? Drug Discovery Today 2009, 14, 1011-1020. d) Ritchie, T. J.; Macdonald, S. J.; Young, R. J.; Pickett,
S. D., The impact of aromatic ring count on compound developability: further insights by examining carbo- and
hetero-aromatic and -aliphatic ring types. Drug Discovery Today 2011, 16, 164-171.
(9) The LipE calculation is based upon clogP.

(10) (a) NaV1.7 potency was measured using a Patch-Xpress (PX) electrophysiology assay conducted with a
protocol that established ~20% inactivation and hence enabled the identification of state dependent inhibitors
(b) PX IC50 determinations were made using at least four different concentrations of test compound at half log
units applied individually, with washout, recovery of current, and resetting of holding voltage between each
individual concentration. Percent inhibition as a function of compound concentration was pooled from at least n
= 10 different cells, with two to three data points per concentration, and fitting the resulting data set with a Hill
(4-parameter logistic) fit in DataXpress 2.0 software to produce a single IC50 curve. See Experimental for
complete details of protocols.
(11) pKa values were calculated from capillary electrophoresis data obtained with pKa PRO Analyzer (AATI,
Ames, IA) in the pH range of 1.7 to 11.2
(12) (a) Evans, D. C.; Watt, A. P.; Nicoll-Griffith, D. A.; Baillie, T. A. Drug-protein adducts: an industry
perspective on minimizing the potential for drug bioactivation in drug discovery and development. Chem. Res.
Toxicol. 2004, 17, 3–16. (b) Kalgutkar, A. S.; Gardner, I.; Obach, R. S.; Shaffer, C. L.; Callegari, E.; Henne, K.
R.; Mutlib, A. E.; Dalvie, D. K.; Lee, J. S.; Nakai, Y.; O’Donnell, J. P.; Boer, J.; and Harriman, S. P. A
comprehensive listing of bioactivation pathways of organic functional groups. Curr. Drug Metab. 2005, 6, 161–
225. (c) Kalgutkar, A. S.; Soglia, J. R. Minimising the potential for metabolic activation in drug discovery.
Expert Opin. Drug Metab. Toxicol. 2005, 1, 91–142.
(13) Both GX-936 and PF-04856264 first appeared in the following patent. Beaudoin, S.; Laufersweiler, M.C.;
Markworth, C.J.; Marron, B.E.; Millan, D.S.; Rawson, D.J.; Reister, S.M.; Sasaki, K.; Storer, R.I.; Stupple,
P.A.; Swain, N.A.; West, C.W.; Shulan, Z. Preparation of heteroarylsulfonamide derivatives for use as
analgesics. WO 2010079443 A1, July 10, 2010.
(14) The tritiated ligand employed for the displacement assay: (P)-1-(3’-chloro-2-fluoro-5,5’-dimethoxy-
[1,1’biphenyl]-4-yl)-N-(isoxazol-3-yl)-2-oxo-1,2-dihydroquinoline-6-sulfonamide[5’-methoxy-C3H3]; based
upon compound 501 from the following patent: Boezio, C.; Boezio, A.; Bregman, H.; Chakka, N.; Coats, J. R.;
Copeland, K. W.; DiMauro, E. F.; Dineen, T.; Gao, H.; La, D.; Marx, I. E.; Nguyen, H-N.; Peterson, E. A.;

Weiss, M. Dihydrobenzoxazine and tetrahydroquinoxaline sodium channel inhibitors. PCT Int. Appl.
WO2013122897, Aug. 22. 2013.
(15) The X-ray ID for the protein structure employed is 5EK0 and the docking protocol was Schrodinger Glide
with SP (standard precision).
(16) Percent homology was determined by comparison of the linear sequences and based upon identity of
residue. This calculation was perform employing the Molecular Operating Environment (MOE) software.
(17) a) Le Bars, D.; Gozariu, M.; Cadden, S. W. Animal Models of Nociception. Pharmacol. Rev.2001, 53,
597–652. b) Bannon, A.W.; Malmberg, A.B. Unit 8.9 models of nociception: hot plate, tail flick, and formalin
tests in rodents. Curr. Protoc. Neurosci. 41, 16.
(18) Liang, J., Brochu, R. M., Cohen, C.J.; Dick, I.E.; Felix, J.P.; Fisher, M.H.; Garcia, M.L.; Kaczorowski,
G.J.; Lyons, K.A.; Meinke, P.T. Priest, B.T.; Schmalhofer, W.A.; Smith, M.M.; Tarpley, J.W.; Williams, B.S.;
Martin, W.J.; Parsons, W.H. Discovery of potent and use-dependent sodium channel blockers for treatment of
chronic pain. Bioorg. Med. Chem. Lett. 2005, 15, 2943–2947.
(19) The automated nature of this experiment makes this approach appealing as it removes operator bias in the
quantitation of pain: Yaksh, T. L.; Ozaki, G.; McCumber, D.; Rathbun, M.; Svensson, C.; Malkmus, S.; Yaksh,
M. C. An automated flinch detecting system for use in the formalin nociceptive bioassay. J. Appl. Physiol.
2001, 90, 2386–2402.
(20) Rat IV, 0.50 mg/kg, formulation: DMSO; Rat PO, 10 mg/kg, formulation: 30% (w/v) hydroxypropyl β-
cyclodextrix in water pH adjusted to 10.0 with KOH
(21) Blair, N.T.; Bean, B.P. Roles of tetrodotoxin (TTX)-sensitive Na+ current, TTX-resistant Na+ current, and
Ca2+ current in the action potentials of nociceptive sensory neurons. J. Neurosci. 2002, 22, 10277-10290
(22) Jarvis, M.F.; Honore, P.; Shieh, C.; Chapman, M.; Joshi, S.; Zhang, X.; 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.K.; Sullivan, J.P.; Faltynek, C.R.;

Krafte, D.S. A-803467, a potent and selective Nav1.8 sodium channel blocker, attenuates neuropathic and
inflammatory pain in the rat. PNAS 2007, 104, 8520-8525.
(23) Chimera work performed using chimeras made from human Nav1.7 and human Nav1.5 with 1 comparable
to that reported by Pfizer revealed similarly that selective sulfonamide hNav1.7 inhibitors such as 1 were
binding in the DIV voltage sensing domain. Further scrutiny of the protein sequence for the DIV domain and
comparison of human and rat Nav1.7 and human Nav1.5 led to the hypothesis that our inhibitors were
interacting across the S2 and S4 helices, in a manner likely similar to that reported for PF-04856264, which
shows a similar selectivity profile to 1.
(24) (a) Davidson, S.; Giesler, G.J. The Multiple Pathways for Itch and their interactions with Pain. Cell 2010
33, 550-558. (b) Devigili, G.; Eleopra, R.; Pierro, T.; Lombardi, R.; Rinaldo, S.; Lettieri, C.; Faber, C.G.;
Ingemar, M.S.J.; Waxman, S.G.; Lauria, G. Paroxysmal itch caused by gain-of-function NaV1.7 mutation.
Pain 2014, 155, 1702-1707.
(25) Gingras, J.; Smith, S.; Matson, D. J.; Johnson, D.; Nye, K.; Couture, L.; Feric, E.; Yin, R.; Moyer, B. D.;
Peterson, M. L.; Rottman, J. B.; Beiler, R. J.; Malmberg, A. B.; McDonough, S. I. Global Nav1.7 knockout
mice recapitulate the phenotype of human congenital indifference to pain. PloS One 2014, 9, e105895.
(26) Rats do not respond to a histamine challenge and therefore the focus of this investigation was on mouse.
(27) 15 didn’t demonstrate any significant binding to H1, 2 or 3 in a Cerep panel (single point displacement @
10 µM).

Graphical Abstract