Discovery of diphenyl amine based sodium channel blockers, effective against hNav1.2

Article abstract of DOI:10.1016/j.bmc.2006.09.010

The development of new therapies for chronic pain is an area of unmet medical need. Central to pathways of chronic pain is the upregulation of voltage-gated sodium channels. The use of tricyclic antidepressants, which also have sodium channel activity, in chronic pain therapy prompted us to develop novel compounds from this scaffold. Herein, we show that the tricyclic moiety is not needed for effective inhibition of the [³H]-BTX binding site and sodium currents of hNa_v1.2. Our lead compound 6, containing a diphenyl amine motif, demonstrated a 53% inhibitory block of Na_v1.2 currents at 10 μM, which is greater than 50% increase in current block in comparison to the amitriptyline standard. Altogether our study establishes that the tricyclic motif is unnecessary for hNa_v1.2 activity and modification of the amine portion is detrimental to sodium channel block.

Full text of DOI:10.1016/j.bmc.2006.09.010

Bioorganic & Medicinal Chemistry 14 (2006) 8366–8378
Discovery of diphenyl amine based sodium channel blockers,
effective against hNav1.2
Debjani P. Hudgens,a Catherine Taylor,a Timothy W. Batts,b
Manoj K. Patelb and Milton L. Browna,c,
*
aDepartment of Chemistry, University of Virginia, Charlottesville, VA 22904, USA
bDepartment of Anesthesiology, University of Virginia, Charlottesville, VA 22908, USA
cDepartments of Oncology and Neuroscience, Drug Discovery Program, Georgetown University, Washington, DC 20057, USA
Received 21 June 2006; revised 31 August 2006; accepted 7 September 2006
Available online 10 October 2006
Abstract—The development of new therapies for chronic pain is an area of unmet medical need. Central to pathways of chronic pain
is the upregulation of voltage-gated sodium channels. The use of tricyclic antidepressants, which also have sodium channel activity,
in chronic pain therapy prompted us to develop novel compounds from this scaffold. Herein, we show that the tricyclic moiety is not
needed for effective inhibition of the [3H]-BTX binding site and sodium currents of hNav1.2. Our lead compound 6, containing a
diphenyl amine motif, demonstrated a 53% inhibitory block of Nav1.2 currents at 10 lM, which is greater than 50% increase in cur-
rent block in comparison to the amitriptyline standard. Altogether our study establishes that the tricyclic motif is unnecessary for
hNav1.2 activity and modification of the amine portion is detrimental to sodium channel block.
ꢀ 2006 Elsevier Ltd. All rights reserved.
O
1. Introduction
N
Tricyclic antidepressants (TCAs) have been clinically
utilized as mood stabilizing agents since their discovery
N
N
N
N
H
in the early 1960s.1 However, upon administration of
these drugs to patients with chronic pain, it was seren-
dipitously elucidated that TCAs possessed clinically rel-
evant analgesic properties. Amitriptyline is a potent
member of this class and has some clinical effectiveness
against certain pain syndromes (Fig. 1). Specifically, this
drug has been therapeutically utilized in the treatment of
migraines,2,3 diabetic neuropathy,4 postherpetic neural-
gia,5 and chronic lower back pain.6 The mechanistic
pathway by which amitriptyline functions in the treat-
ment of pain has yet to be fully determined. Although,
it is reasonable to propose that its remarkable capabili-
ties as a sodium channel blocker correlate to these phys-
iological functions.7
Amitriptyline
Doxepin
Nortriptyline
Imipramine
Figure 1. Class of tricyclic antidepressants (TCAs). Amitriptyline is a
potent member of the tricyclic antidepressant drug class, with effects
against numerous pain syndromes.
nels in the pain pathway. It is known that sodium
channels are associated with the sensitivity of neuronal
firing, especially during a state of injury when hyperex-
citability is observed. This hyperexcitability is typically
found at both sites of damage and within the dorsal root
ganglion neurons (DRGs).8,9 Current findings further
demonstrate that damaged neurons can display a mor-
phological change in sodium channel accumulation,
with upregulation of certain sodium channel isoforms
and downregulation of others.10 Numerous electrophys-
iological and pharmacological experiments have provid-
ed a direct relationship of sodium channels with pain
sydromes.11,12 Thus, we therefore view the sodium chan-
nel as an important target in regards to elucidating new
therapies for treating painful neuropathies.
Recent studies have provided us with a better under-
standing of the molecular mechanism of sodium chan-
Keywords: Diphenyl amine; Neuropathic pain; Sodium channel block-
ers; Amitriptyline.
*
Corresponding author. Tel.: +1 202 687 8603; fax: +1 202 687
0617; e-mail: mb544@georgetown.edu
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.09.010

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8367
Voltage-gated sodium channels, VGSCs, consist of a
family of nine known isoforms (Nav1.1–Nav1.9), each
containing a 260 kDa a-subunit and auxiliary 33–
36 kDa b-subunits.13 However, it has been found that
the a-subunits alone contains the key pharmacological
and physiological elements, with all known drug interac-
tions occurring within its pore.14 This a-subunit com-
prises of four homologous domains (DI–DIV), each
with six transmembrane segments (S1–S6) which orient
to form the ion pore.
initially screened in the [3H]-BTX displacement assay
and compounds showing substantial binding to the sodi-
um channel protein were further evaluated for inhibitory
effects on hNav1.2 using patch-clamp electrophysiology.
From our findings we have been able to establish key
elements for activity, as well as develop a more potent
structural class of sodium channel blockers.
2. Results
Within the channel there is a location termed site 2, in
which batrachotoxin (BTX) a known neurotoxin from
the poison dart frog (Phyllobates aurotaenia) preferen-
tially binds.15 This site is also shared by local anesthet-
ics, anticonvulsants, and antiarrhythmics.16 Therefore
use of a competitive inhibition assay with [3H]-BTX
has provided a facile assay17 for measuring the binding
affinity for compounds interacting with the sodium
channel protein.
Analogues 1–6 were initiated by Grignard addition of
the dimethylaminopropylmagnesium chloride (DMAP-
Cl) complex to the commercially available starting ke-
tones. Tertiary alcohols 2, 3, and 5 were subsequently
dehydrated to their corresponding alkenes by refluxing
in a 7 M HCl/EtOH solution (Scheme 1). Addition of
these organomagnesium complexes proved to be low
yielding, in accordance with previously reported stud-
ies.24 Herein, we have found that a higher yielding and
more direct route to these analogues were achieved by
formation of the corresponding phosphonium salt for
subsequent Wittig addition. Analogues 7, 8, and 13–25
were synthesized according to these routes in yields
ranging from 52 to 93% (Schemes 2 and 4). Higher yields
were obtained utilizing a freshly distilled THF solution
and accurately determining nBuLi concentration
through titration or NMR analysis,25 as previously
reported. Analogue 9 was constructed via imine conden-
sation providing the product in 96% yield (Scheme 3).
Doebner modified-Knoevenagel condensation yielded
acid 10, which was further coupled to dimethylamine
using DCC conditions to provide dimethylamide 11 in
56% yield. Analogue 12 was designed from the potent
sodium channel blocker, phenytoin, and subsequent
synthesis by Bucherer–Berg conditions provided the
hydantoin in good yield.26 All reactions resulting in a
mixture of alkene stereoisomers were used without fur-
ther separation, so that sufficient quantities for biologi-
cal testing could be obtained.
The incidence of neuropathic pain syndromes within the
general population is quite common, with 2.1–4.7 per-
sons per 100,000 being affected by trigeminal neuralgia,
11–16% of Type I and Type II diabetics displaying dia-
betic neuropathies, and 34 persons per 100,000 suffering
from postherpetic neuralgia.18 Interestingly tricyclic
antidepressants, specifically amitriptyline, have been
the most clinically utilized therapeutic for the treatment
of such ailments.19 Other commonly known drugs, such
as the long-acting local anesthetic bupivacaine, substan-
tially block nerves, but suffer from a high degree of
cardiotoxicity. Anticonvulsants, phenytoin and carbam-
azepine, also have adverse pharmacokinetic interactions
by production of liver enzymes.20 TCAs, themselves,
suffer from some adverse effects as well, and in one par-
ticular study 20% of patients withdrew from treatment
due to the severity of these side effects, which included
drowsiness, blurred vision, arrhythmias, and heart
block.21 However TCAs, such as amitiptyline, serve as
one of the best treatments for neuropathic pain to
date.22
Compounds were evaluated at 10 lM for the ability to
displace [3H]-BTX binding from the neuronal voltage-
gated sodium channel (NVSC) in prepared rat synapto-
somes. All evaluations were carried out in duplicate. In
our study, amitriptyline was used as a standard and
found to have 89% inhibition at 10 lM.
In this study, we propose to further develop analogues
of amitriptyline in order to create a more potent and
well-tolerated therapeutic for the potential treatment
of neuropathic pain syndromes.23 We have been able
to design, synthesize, and evaluate novel compounds de-
signed from a common amitriptyline scaffold. Modifica-
tion at three major sites (Fig. 2) was ensued in order to
elucidate the structural requirements necessary for en-
hanced sodium channel inhibition. All compounds were
Incorporation of a tertiary alcohol into the amitriptyline
scaffold, as in analogue 2, provided equivalent activity
with 82% block. Other analogues, 3–13 and 14 (Table
1), provided lower binding affinities than amitriptyline
ranging from 7% to 71%. Phenyl ring replacement by
a cyclohexyl provided analogue 7 with 85% inhibition.
However modification of the diphenyl amine structural
class, by various substitutions, led to numerous ana-
logues 15–18 and 20–22 (Table 2) with similar or much
higher displacement of [3H]-BTX than amitriptyline,
thereby demonstrating that this novel monocyclic class
are better binders of the sodium channel protein.
Site 1
Site 2
Site 3
N
Figure 2. Major sites of scaffold modification. Three major sites of the
amitriptyline scaffold have been modified in order to determine the
structural requirements for activity.
Amitriptyline compounds of this novel class demon-
strated inhibitory tonic effects against firing of hNav1.2

8368
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
Site 1
Amitriptyline
1
N
N
Cl
7M HCl, EtOH
63%
HO
HO
Mg, I , THF
2
O
O
O
68%
N
N
N
1
4
2
N
Cl
7M HCl, EtOH
56%
Mg, I , THF
2
39 %
N
3
N
Cl
7 M HCl, EtOH
78%
Mg, I , THF
2
HO
61%
N
N
6
5
Scheme 1. Modification at site 1, phenyl ring orientation. Modifications of the amitriptyline scaffold were conducted to determine phenyl ring
orientation, by both ring contraction and removal.
hNav1.2 current. Of the 18 analogues tested, only eight
were found to have attenuated block of the sodium
current compared with amitriptyline (Table 3). These
Site 2
findings were contrary to our initial [3H]-BTX
evaluation which yielded few compounds showing
N
substantial binding, but evaluation by patch-clamping
electrophysiology has revealed the majority of analogues
+
-
EtOH
15 hrs
77%
+
-
Br
PPh Br
3
N
H
N
PPh Br
3
+
to be better functional blockers of the sodium channel,
than amitriptyline. Our initial lead, diphenyl analogue
6, provided 53 3 block (n = 6) of hNav1.2 current,
which is a 50% increase of sodium current block
when compared to amitriptyline (Fig. 3). Similar or
enhanced activity was found for substituted diphenyl
analogues 16 and 21.
nBuLi
+
-
+
+
N
N
PPh Br
3
THF
66%
O
N
7
3. Discussion
nBuLi
THF
+
-
PPh Br
3
O
93%
The demand for a highly effective treatment of chronic
pain without the toxicities associated with most has
led us to pursue this area of unmet medical need. In this
investigation we have successfully made analogues of
amitriptyline, the most commonly utilized therapeutic
for neuropathic pain syndromes. Design and synthesis
of analogues ensued in order to develop more potent
therapeutics and to determine the essential structural
properties of amitriptyline and its analogues for binding
to the sodium channel protein. Modifications at three
major sites (Fig. 2) were conducted and consist of alter-
ation of the phenyl ring orientation, replacement of the
phenyl ring, and variation of the amine functional
group.
N
8
Scheme 2. Modification at site 2, phenyl replacement. Modifications
were conducted to determine the necessity of the phenyl ring and
replacement was carried out by both cyclohexyl and ethyl
replacements.
in a range of 10–54% at 10 lM. Currents were elicited
using depolarizing pulses from a holding potential of
À100 to +10 mV, at 15 s intervals. Our standard,
amitriptyline, at 10 lM concentration provided 34
2
block (n = 11, where n is the number of trials) of the

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8369
indicates a degree of steric bulk is required upon phenyl
ring replacement for activity.
Site 3
Analogues consisting of amine isosteres, through mod-
ification to a phenyl imine 9, replacement by a carbox-
ylic acid 10, conversion to the dimethyl amide 11, and
insertion of a hydantoin ring 12 were evaluated for
[3H]-BTX inhibition. Analogue 9 afforded very low
binding affinity with only 7% inhibition of [3H]-BTX,
but had slightly better block of sodium current than
amitriptyline. This may indicate that analogue 9 is
blocking the sodium channel by means of an alternate
site separate from the BTX site. Analogues 10–11 were
poor inhibitors of both [3H]-BTX and hNav1.2 sodium
current, indicating that modification of the amine func-
tionality is detrimental to sodium channel activity.
Analogue 12 was found to be a better binder than its
predecessor phenytoin providing 30% inhibition of
[3H]-BTX at 10 lM, in comparison to only 50% inhibi-
tion of [3H]-BTX at 40 lM with phenytoin.29 This indi-
N
A. Imine Formation
NH
2
Et O
2
96%
+
H
H
O
N
9
B. Amide Insertion
O
O
+
piperdine
75
HO
OH
O
H
HO
10
O
´
˚
N
cates that an increase of 2–3 A of steric volume can be
H
tolerated by the compound and provide an increase in
binding affinity. Overall analogue 12 did not prove to
be a significant inhibitor of either [3H]-BTX or hNav1.2
sodium current when compared to other derivatives of
this potent class.
EDCI, HOBt,
DIEA, DMF
58%
N
O
11
C. Hydantoin Isostere
Analogue 6 demonstrated to be the most potent diphen-
yl analogue, and we therefore were interested in con-
ducting an optimization of the phenyl ring portion of
this lead compound. Modifications incorporating ortho,
meta, and para substitutions of the diphenyl structure
by –Cl, –OMe, and –Me groups were designed and syn-
thesized. The 2,2-dichloro and 4,4-dichloro analogues
were also constructed by similar methods. Substituted
analogues 15–18 and 20–22 provided similar or higher
displacement of [3H]-BTX than amitriptyline, and sub-
stantial block of hNav1.2 current by analogues 16 and
21 was shown to have equal or enhanced activity when
compared to lead analogue 6 (Chart 1).
KCN, (NH ) CO
4 2
3
75%
H
O
NH
O
O
N
H
12
Scheme 3. Modification at site 3, amine isosteres. Modifications of the
amine functionality were conducted by imine formation, amide
insertion, and hydantoin isostere.
Alteration of the tricyclic portion by either ring contrac-
tion or removal of the internal ring resulted in better
blockers of hNav1.2 current. [3H]-BTX evaluation of
the intermediate tertiary alcohols 2 and 5 showed similar
binding affinities as amitriptyline, whereas alcohol 3
provided substantially less binding and was therefore
not pursued for inhibition of Nav1.2. Dehydration of
these analogues led to 4 and 6, which had lower binding
in the [3H]-BTX evaluation, but functional block re-
vealed that these analogues were better inhibitors of
hNav1.2 current than amitriptyline. Particularly com-
pound 6, which provided 53% inhibition at 10 lM in
comparison to amiptyline with 33% at 10 lM. From
these initial findings, we have been able to develop an
inhibitor with over 50% more activity than amitriptyline
and provide evidence against current design,27,28 by
revealing that the tricyclic portion is not essential for
inhibition of sodium channel current.
From these findings we have been able to determine that
the [3H]-BTX displacement assay does not sufficiently
explain the functional effects on sodium channel current.
This is consistent with a recent study which suggests that
the TCA binding site overlaps with only a portion of the
local anesthetic site in the channel.30 The dogma that the
tricyclic portion is essential for sodium activity has been
disproven, and we have thereby introduced a new class
of diphenyl amitriptyline analogues.
4. Conclusion
There is a great need for the discovery and development
of new analgesics. Herein, we have been able to success-
fully design and synthesize analogues derived from an
amitriptyline scaffold, thereby revealing a potent new
class of diphenyl amine sodium channel blockers. We
have also been able to prove that the tricyclic portion
is unnecessary for block of neuronal sodium channel
current as previously reported. Also alteration of the
amine portion was shown to be detrimental to both
Analogues were also constructed by replacement of one
phenyl ring, by either a cyclohexyl ring 7 or an ethyl
chain 8. Both 7 and 8 showed poor results in the [3H]-
BTX assay, and 7 provided similar functional block to
amitriptyline, whereas 8 was substantially lower. This

8370
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
A. Chain Length Modification
nBuLi
THF
25%
N
+
PPh +Br-
3
O
N
13
B. Heterocylce Insertion
nBuLi
THF
29%
+
N
PPh +Br-
3
N
N
O
N
14
C. Phenyl Ring Substitution
R
R
1
2
R
1
R
2
nBuLi
THF
+
N
PPh +Br-
3
O
N
15-25
Compound
R1
2-Cl
3-Cl
4-Cl
2-OMe
3-OMe
4-OMe
2-Me
3-Me
4-Me
2-Cl
R2
%
91
80
81
82
91
52
48
43
89
68
83
15
16
17
18
19
20
21
22
23
24
25
H
H
H
H
H
H
H
H
H
2-Cl
4-Cl
4-Cl
Scheme 4. Optimization of lead Compound 6. Optimizations of our lead analogue 6 were conducted by chain length modification, heterocyclic
isosteres, and phenyl ring substitution.
binding of the sodium channel protein and block of
sodium channel current.
the University of Illinois Urbana-Champaign School
of Chemical Sciences.
5.2. General procedure A.1
5. Experimental
5.1. Chemistry
A flame dried flask equipped with condenser, stir bar,
and drying tube was kept under N2. Magnesium shav-
ings (1.5 equiv), finely ground with a mortar and pestle
were placed in the flask along with one small crystal of
I2. Mild heat was applied until an atmosphere of iodine
could be visualized. THF, freshly distilled over CaH2,
was now added along with a small amount of MeI
(0.001 equiv). The solution was refluxed vigorously for
several minutes, after which a freshly distilled solution
of dimethylaminopropyl chloride (DMAP-Cl) (1.5
equiv) was added. The solution was refluxed approxi-
mately 2–3 h, or until disappearance of all magnesium
shavings. A separate flame dried flask containing the ke-
tone (1.0 equiv) in THF was obtained and kept at 0 ꢁC,
while under an atmosphere of N2. The Grignard solu-
tion was added dropwise to the flask, while trying to de-
cant any remaining magnesium solid. The reaction
All syntheses requiring anhydrous conditions were kept
under inert N2 atmosphere and conducted in flame-
dried glassware. Solvents were obtained from activated
alumina stills or were of commercial grade quality.
THF obtained directly from the alumina still was fur-
ther dried by distillation over CaH2. N-Butyllithium
concentrations were determined by NMR methods.
Reactions resulting in a mixture of alkene stereoisomers
were used without further separation. Melting points
were recorded from an Electrothermal Mel-TempTM
1
melting point apparatus and are uncorrected. H and
13C NMRs were conducted on a Varian 300 MHz
NMR in CDCl3 at ambient temperature. High resolu-
tion mass spectral (HRMS) data were determined at

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8371
Table 1. [3H]-BTX displacement of analogues
Table 1 (continued)
Compound
ID
Percent inhibition of
[3H]-BTX at 10 lM (%)
Compound
ID
Percent inhibition of
[3H]-BTX at 10 lM (%)
1
88
82
19
8
68
N
N
2
HO
9
7
N
N
3
HO
10
27
N
O
OH
4
48
11
25
N
O
N
5
71
HO
H
N
12
35
O
N
N
H
O
6
67
13
69
N
N
7
85
N
14
34
N
N

8372
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
Table 3. Percent block of hNav1.2 current at 10 lM
Table 2. Optimization of lead Compound 6
Compound
ID
Percent block of hNav1.2
current at 10 lM
R1
R2
N
1
33.78 1.91 (n = 11)
48.19 5.87 (n = 9)
18.83 7.95 (n = 2)
Compound
ID
Percent Inhibition
of [3H]-BTX at 10 lM (%)
N
R1 = 2-Cl, R2 = H
R1 = 3-Cl, R2 = H
R1 = 4-Cl, R2 = H
R1 = 2-OMe, R2 = H
R1 = 3-OMe, R2 = H
R1 = 4-OMe, R2 = H
R1 = 2-Me, R2 = H
R1 = 3-Me, R2 = H
R1 = 4-Me, R2 = H
R1 = 2-Cl, R2 = 2-Cl
R1 = 4-Cl, R2 = 4-Cl
15
16
17
18
19
20
21
22
23
24
25
83
79
88
99
70
85
85
76
41
63
56
4
N
mixture was allowed to slowly warm to room tempera-
ture overnight and was later quenched by addition of
a saturated NH4Cl solution. The crude product was
extracted into MeCl2 (3 · 50 mL), washed with brine
(1 · 20 mL), dried over MgSO4, filtered, and evaporated
to provide the crude product material. Flash chroma-
tography was carried out using a MeCl2/MeOH system
from (100:0) to (9:1).
HO
5
N
5.3. General procedure A.2
6
7
8
9
53.24 2.55 (n = 6)
35.32 3.22 (n = 5)
10.27 2.79 (n = 3)
37.49 8.98 (n = 3)
In a flask equipped with stir bar and condenser was add-
ed the tertiary alcohol (1.0 equiv) obtained from Proce-
dure A.1. Dehydration was then carried out by refluxing
in an ethanolic solution of 7 M HCl (>25 equiv) for 3-
4 h. The solution was evaporated, dissolved in water,
and basified to a pH 10 with K2CO3. The material was
taken up in MeCl2 and extracted (3 · 50 mL), dried over
MgSO4, filtered, and evaporated. Flash chromatogra-
phy of the crude material was carried out using a
MeCl2/MeOH system from (100:0) to (10:1).
N
N
N
5.4. General procedure B.1
(3-Bromopropyl) triphenylphosphonium bromide (1.0
equiv) was added to a flame dried flask under N2. The
salt was then stirred in absolute EtOH at 0 ꢁC and a
dimethylamine solution in THF (1.5 equiv) was added
dropwise. The solution was stirred with warming to
room temperature for over 15 h. The solvent was then
evaporated and the salt used directly without further
purification. Note: (4-Dimethylaminobutyl) triphenyl-
phosphomium bromide was made similarly for analogue
13.
N
5.5. General procedure B.2
(3-Dimethylaminopropyl) triphenylphosphonium bro-
mide (1.5–2.0 equiv) obtained from Procedure B.1 was
placed in a flame dried flask containing freshly distilled

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8373
Table 3 (continued)
Table 3 (continued)
Compound
ID
Percent block of hNav1.2
current at 10 lM
Compound
ID
Percent block of hNav1.2
current at 10 lM
O
10
23.80 2.66 (n = 4)
20
22.55 4.36 (n = 5)
O
OH
N
21
24
25
54.33 18.85 (n = 2)
28.43 4.98 (n = 2)
39.51 6.03 (n = 3)
11
12
15
16
17
18
15.35 7.98 (n = 3)
N
O
N
Cl
Cl
N
H
N
13.19 2.39 (n = 2)
36.36 8.43 (n = 7)
49.18 6.54 (n = 3)
35.78 6.97 (n = 3)
27.04 4.78 (n = 4)
O
N
H
O
Cl
Cl
Cl
N
N
THF and kept under N2. The mixture was cooled to
0 ꢁC and a solution of 2.5 M nBuLi in hexane (1.5–2.0
equiv) was added dropwise. Stirring was continued for
several minutes, after which the ketone (1.0 equiv) was
added slowly. The reaction mixture was allowed to con-
tinue stirring overnight with warming to room tempera-
ture and was later quenched by addition of a saturated
NH4Cl solution. The crude material was then extracted
into MeCl2 (3 · 50 mL), washed with brine (1 · 20 mL),
dried over MgSO4, filtered, and evaporated to provide
the crude product. Flash chromatography was carried
out using a MeCl2/MeOH system from (100:0) to (9:1).
Cl
Cl
N
5.5.1. [3-(10,11-Dihydro-dibenzo[a,d]cyclohepten-5-yli-
dene)- propyl]-dimethyl- amine hydrochloride (1). Proce-
dure A.2 was carried out using the tertiary alcohol
(0.5 g, 1.7 mmol) and 7 M HCl/EtOH (15 mL). The
product was obtained in 0.29 g as an off-white solid in
N
1
63% yield. TLC: MeCl2/MeOH (10:1), Rf = 0.431. H
NMR (CDCl3, 300 MHz) d 7.27–7.03 (m, 8H), 5.78–
5.73 (t, 3H, J = 7.5 Hz), 3.30–3.26 (m, 2H), 3.12–3.10
(m, 1H), 2.97–2.88 (m, 2H), 2.80–2.76 (m, 1H), 2.66
(br s, 8H); 13C NMR (CDCl3, 75.5 MHz) d 146.55,
139.39, 138.57, 138.34, 136.42, 129.58, 127.85, 127.57,
127.09, 126.95, 125.65, 125.56, 122.89, 56.27, 42.40,
41.65, 33.02, 31.35, 23.69. ESIMS: 278 (M+H). Lit mp
193–194 ꢁC,31 mp 177–181 ꢁC.
O
N

8374
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
80
70
60
50
40
30
20
10
0
1
4
5
6
7
8
9
10 11 12 15 16 17 18 20 21 24 25
Compound ID
Chart 1. Percent block of hNav1.2 upon compound administration. The tonic block of amitriptyline analogues on hNav1.2 is shown. Percent block
was determined at 10 lM, by holding cells at À100 mV and then stepping to +10 mV to elicit a current and repeating every 15 s.
(d, 2H, J = 9.0 Hz), 7.38-7.10 (m, 4H), 4.80 (br s, 1H),
2.05–1.91 (m, 4H), 1.85 (s, 6H), 1.21–1.09 (m, 2H); 13C
NMR (CDCl3, 75.5 MHz) d 149.62, 139.24, 128.29,
127.54, 123.53, 119.56, 81.31, 59.39, 44.51, 38.03,
21.62. ESIMS: 268 (M+H). Lit mp 101–103 ꢁC,33 mp
10 μM Amitriptyline
78–82 ꢁC.
Control
5.5.3. (3-Fluoren-9-ylidene-propyl)-dimethyl-amine (4).
Procedure A.2 was carried out using the tertiary alcohol
(0.1 g, 0.37 mmol) and 7 M HCl/EtOH (10 mL). The
product was obtained in 54 mg as an orange solid in
56% yield. TLC: MeCl2/MeOH (10:1), Rf = 0.30. 1H
10 μM Compound 6
NMR (CDCl3, 300 MHz) d 7.88–7.65 (m, 4H), 7.22–
7.45 (m, 4H), 6.78–6.73 (t, 1H, J = 7.5 Hz), 3.15–2.95
Control
(m, 2H), 2.68–2.63 (t, 2H, J = 7.2 Hz), 2.35 (s, 6H);
13C NMR (CDCl3, 75.5 MHz) d 148.76, 132.16,
131.83, 130.85, 128.94, 128.70, 127.93, 127.07, 119.97,
59.20, 44.85, 27.48. ESIMS: 250 (M+H). Mp 111–
115 ꢁC.
500 pA
1 ms
Figure 3. Comparing sample current traces of amitriptyline and
Compound 6. Example current traces demonstrate block of hNav1.2
by amitriptyline and our novel analogue 6.
5.5.4. (3-Fluoren-9-ylidene-propyl)-dimethyl-amine (5).
Procedure A.1 was carried out using DMAP-Cl
(0.77 g, 6.2 mmol), Mg0 (0.15 g, 6.2 mmol), benzophe-
none (0.73 g, 4.0 mmol), and THF (10 mL). The product
was obtained in 0.65 g as a white solid in 61% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.19. 1H NMR
(CDCl3, 300 MHz) d 7.55–7.49 (m, 4H), 7.34–7.27 (m,
5H), 7.23–7.18 (m, 1H), 2.57–2.51 (m, 4H), 2.24 (s,
6H), 1.68–1.61 (m, 2H). ESIMS: 270 (M+H). Lit mp
120–123 ꢁC,34 mp 118–120 ꢁC.
5-(3-Dimethylamino-propyl)-10,11-dihydro-5H-dibenzo-
[a,d]cyclohepten-5-ol (2). Procedure A.1 was carried out
using DMAP-Cl (1.25 g, 10.25 mmol), Mg0 (0.25 g,
10.25 mmol), dibenzosuberone (1.6 g, 7.6 mmol), and
THF (5 mL). The product was obtained in 1.5 g as a
white solid in 68% yield. TLC: MeCl2/MeOH (10:1),
Rf = 0.24. 1H NMR (CDCl3, 300 MHz) d 8.13–8.11
(appt d, 2H,J = 7.5 Hz), 7.28-7.11 (m, 6H), 3.56–3.48
(m, 2H), 3.07–3.02 (m, 2H), 2.57–2.53 (m, 2H), 2.24
(br s, 8H), 1.40 (br s, 2H); 13C NMR (CDCl3,
75.5 MHz) d 146.63, 137. 99, 130.91, 127.55, 127.44,
126.61, 77.01, 60.37, 45.89, 45.06, 34.47, 23.06. APC-
IMS: 296 (M+H). Lit mp 118–120 ꢁC,32 mp 119–120 ꢁC.
5.5.5. (4,4-Diphenyl-but-3-enyl)-dimethyl-amine (6). Pro-
cedure A.2 was carried out using the tertiary alcohol
(0.25 g, 0.93 mmol) and 7 M HCl/EtOH (10 mL). The
product was obtained in 0.18 g as a pale yellow oil in
78% yield. TLC: MeCl2/MeOH (10:1), Rf = 0.36. 1H
NMR (CDCl3, 300 MHz) d 7.37–7.18 (m, 10H), 6.13–
6.08 (t, 1H, J = 7.2 Hz), 2.45–2.41 (t, 2H, J = 6.6 Hz),
2.34–2.30 (t, 2H, J = 7.1 Hz), 2.22 (s, 6H); 13C NMR
(CDCl3, 75.5 MHz) d 142.67, 142.43, 139.98, 132.04,
131.96, 131.86, 129.71, 128.49, 128.40, 128.17, 128.00,
127.15, 126.95, 126.87, 59.48, 45.22, 28.09. ESIMS:
252 (M+H).
5.5.2. 9-(3-Dimethylamino-propyl)-9H-fluoren-9-ol (3).
Procedure A.1 was carried out using DMAP-Cl
(0.91 g, 7.5 mmol), Mg0 (0.18 g, 7.5 mmol), fluorenone
(0.9 g, 5.0 mmol), and THF (15 mL). The product was
obtained in 0.25 g as a yellow solid in 39% yield. TLC:
MeCl2/MeOH (10:1), Rf = 0.18. 1H NMR (CDCl3,
300 MHz) d 7.60–7.58 (d, 2H, J = 6.0 Hz), 7.45–7.42

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8375
5.5.6. (4-Cyclohexyl-4-phenyl-but-3-enyl)-dimethyl-amine
(7). Procedure B.2 was carried out using (3-dimethyla-
minopropyl) triphenylphosphomium bromide (1.14 g,
2.66 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(1 mL, 2.60 mmol), cyclohexyl phenyl ketone (0.25 g,
1.33 mmol), and THF (5 mL). The product was ob-
tained in 0.18 g as a pale yellow oil in 66% yield. TLC:
MeCl2/MeOH (10:1), Rf = 0.33. 1H NMR (CDCl3,
300 MHz) d 7.35–7.23 (m, 4H), 7.11–7.03 (m, 2 H),
5.42–5.34 (m, 0.80H), 5.21–5.19 (br m, 0.17H), 2.42–
2.33 (m, 4H), 2.20 (s, 6H) 2.08–2.05 (m, 2H), 1.74–
1.65 (br m, 5H), 1.29–1.07 (br m, 5H); 13C NMR
(CDCl3, 75.5 MHz) d 149.05, 141.38, 128.80, 128.00,
127.55, 126.49, 125.90, 121.54, 59.89, 59.68, 46.27,
45.44, 45.00, 32.43, 32.18, 26.83, 26.41. ESIMS: 258
(M+H). HRMS Calcd for C18H27N: 258.2216. Found:
258.2222.
TLC: Hex/EtOAc (1:1), Rf = 0.30. 1H NMR (CDCl3,
300 MHz) d 7.28–7.07 (m, 10H), 6.16–6.12 (t, 1H,
J = 7.1 Hz), 3.11–3.09 (d, 2H, J = 7.2 Hz); 13C NMR
(CDCl3, 75.5 MHz) d 178.29, 145.36, 141.95, 139.22,
129.88, 128.60, 128.33, 127.66, 119.74, 35.37. APCIMS:
239 (M+H). Lit mp 117–118 ꢁC,36 mp 105–108 ꢁC.
5.5.10. 4,4-Diphenyl-but-3-enoic acid dimethylamide (11).
Under anhydrous conditions, 4,4-diphenyl-but-3-enoic
acid (0.47 g, 1.96 mmol), EDCI (0.6 g, 3.06 mmol),
HOBt (0.34 g, 2.51 mmol), and 2.0 M dimethylamine
in THF (4.2 mL, 8.4 mmol) were added to a flask con-
taining dry DMF (30 mL). DIEA (1.83 mL, 10.5 mmol)
was added and the mixture was allowed to stir for 48 h
at room temperature. After this time the DMF was re-
moved by evaporation under reduced pressure, and the
resulting orange residue was taken up in MeCl2. The
material was extracted with water (6 · 50 mL) to remove
any remaining DMF solvent. The organic layer was then
washed with 1 M NaOH (3 · 50 mL), 10% citric acid
(3 · 50 mL), water (1 · 50 mL), and brine (1 · 50 mL).
The organic layer was dried over MgSO4, filtered, and
evaporated to provide 0.3 g of a brownish-red oil. Flash
chromatography was conducted using a MeCl2:MeOH
system from (100:1) to (10:1). The product was obtained
in 0.30 g as a pale orange solid in 58% yield. TLC:
MeCl2/MeOH (32:1), Rf = 0.27. 1H NMR (CDCl3,
300 MHz) d 7.39–7.21 (m, 10H), 6.35–6.30 (t, 1H,
J = 7.2 Hz), 3.19–3.17 (d, 2H, J = 7.2 Hz), 2.94 (s, 3H),
2.83 (s, 3H); 13C NMR (CDCl3, 75.5 MHz) d 170.61,
143.20, 141.45, 139.00, 131.53. 131.38, 129.20, 127.76,
127.47, 126.81, 126.74, 126.63, 121.57, 36.60, 34.90,
34.22. Mp 67–71 ꢁC.
5.5.7. Dimethyl-(4-phenyl-hex-3-enyl)-amine (8). Proce-
dure B.2 was carried out using (3-dimethylaminopropyl)
triphenylphosphomium bromide (1.6 g, 3.73 mmol)
from procedure B.1, 2.5 M nBuLi/Hex (1.5 mL,
3.73 mmol), propiophenone (0.25 g, 1.86 mmol), and
THF (5 mL). The product was obtained in 0.70 g as a
white solid in 93% yield. TLC: MeCl2/MeOH (10:1),
1
Rf = 0.6. H NMR (CDCl3, 300 MHz) d 7.32–7.18 (m,
4H), 7.18–7.08 (appt d, 1H, J = 8.4 Hz), 5.58–5.54 (t,
0.41H, J = 6.8 Hz), 5.41–5.36 (t, 0.58H, J = 7.2 Hz),
2.52–2.44 (m, 4H), 2.38 (s, 3H), 2.27 (s, 3H), 2.21–2.14
(appt q, 2H, J = 7.6 Hz), 0.98–0.90 (m, 3H); 13C NMR
(CDCl3, 75.5 MHz) d 146.01, 143.97, 142.52, 140.91,
128.28, 128.18, 126.86, 126.39, 123.63, 120.91, 59.02.
58.77, 44.73, 44.15, 32.26, 26.03, 25.68, 23.21, 13.67,
12.97. ESIMS: 204 (M+H). HRMS calc for C14H21N:
204.1747. Found: 204.1752.
5.5.11. 5-(2, 2-Diphenyl-vinyl)-imidazolidine-2,4-dione
(12). To a solution of b-phenyl cinnamaldehyde (1.0 g,
4.8 mmol) in 50% aq. EtOH (10 mL) were added KCN
(0.95 g, 14.4 mmol) and (NH4)2CO3 (3.0 g, 28.8 mmol).
The reaction mixture was heated to 65 ꢁC for 1 day,
after which the solution was cooled to room tempera-
ture and extracted with MeCl2 (3 · 50 mL). The organic
layers were combined, dried over MgSO4, filtered, and
evaporated to provide a crude brown solid. Purification
by flash chromatography was carried out in a system of
MeCl2/MeOH from (40:1) to (10:1). The product was
obtained as an off-brown solid in 75% yield. TLC:
Hex/EtOAc (1:1) Rf = 0.23. 1H NMR (CDCl3,
300 MHz) d 7.50–7.23 (m, 12H), 6.22–6.19 (m, 1H),
4.87 (br s, 1H); 13C NMR (CDCl3, 75.5 MHz) d
177.64, 169.23, 162.38, 159.43, 146.49, 144.84, 144.71,
143.05, 133.47, 132.91, 132.49, 132.27, 131.59, 131.48,
131.31, 130.38, 117.94, 61.47. APCIMS: 279 (M+H).
HRMS Calcd for C17H14N202: 279.1123. Found:
279.1134. Mp 135–138 ꢁC.
5.5.8. Dimethyl-(4-phenyl-hex-3-enyl)-amine (9). In a
flame dried flask under N2, b-phenyl cinnamaldehyde
(0.25 g, 1.2 mmol) was stirred in dry ether and kept at
room temperature. Aniline (0.11 mL, 1.2 mmol) was
now added dropwise and allowed to stir for an addition-
al 30 min. The solution was evaporated to provide the
product in 0.33 g as a yellow solid in 96% yield. TLC:
Hex: EtOAc (5:1), Rf = 0.32. 1H NMR (CDCl3,
300 MHz) d 8.17–8.14 (m, 1H), 7.46–7.07 (m, 15H),
6.80–6.69 (m, 1H); 13C NMR (CDCl3, 75.5 MHz) d
160.73, 141.04, 138.53, 130.77, 129.46, 129.38, 128.81,
128.75, 128.67, 128.44, 127.53, 126.31, 121.29, 118.82,
115.37. ESIMS 284 (M+H). Lit mp 99 ꢁC,35 mp 98–
100 ꢁC.
5.5.9. 4,4-Diphenyl-but-3-enoic acid (10). Diphenyl acet-
aldehyde (0.91 mL, 5.1 mmol), malonic acid (0.54 g,
5.1 mmol), and piperidine (0.5 mL, 5.1 mmol) were add-
ed to a flame dried flask and kept under inert conditions.
The mixture was heated to 85 ꢁC for 3 h. The solution
was then cooled to room temperature, evaporated, and
taken up in water to be acidified to pH = 2 with HCl.
Extraction was carried out by addition of EtOAc
(3 · 50 mL) and dried by washing with brine
(1 · 20 mL). The organic layers were combined, dried
over MgSO4, filtered, and evaporated to provide the
product in 0.9 g as an off-white solid in 75% yield.
5.5.12. (5, 5-Diphenyl-pent-4-enyl)-dimethyl-amine (13).
Procedure B.2 was carried out using (4-dimethylamin-
obutyl)
triphenylphosphomium
bromide
(1.0 g,
2.3 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(0.96 mL, 2.3 mmol), benzophenone (0.3 g, 1.50 mmol),
and THF (15 mL). The product was obtained in 0.44 g
as a pale yellow oil in 25% yield. TLC: MeCl2/MeOH
(10:1), Rf = 0.23. 1H NMR (CDCl3, 300 MHz) d

8376
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
7.75–7.63 (m, 2H), 7.31–7.05 (m, 8H), 5.99–5.94 (m,
1H), 2.65–2.60 (m, 2H), 2.48 (s, 6H), 2.16–2.05 (m,
2H), 1.82–1.74 (m, 2H); 13C NMR (CDCl3, 75.5 MHz)
d 142.78, 142.03, 139.56, 135.12, 133.53, 133.41,
130.50, 129.59, 128.16, 127.97, 127.41, 127.04, 126.97,
58.01, 43.72, 26.95, 25.82. ESIMS: 266 (M+H). HRMS
Calcd for C19H23N: 266.1913. Found: 266.1909.
in 0.54 g as an off-white semisolid in 81% yield. TLC:
MeCl2/MeOH (10:1), Rf = 0.22. 1H NMR (CDCl3,
300 MHz) d 7.70-7.10 (m, 9H), 6.12-6.07 (m, 1H),
2.42–2.37 (m, 2H), 2.32–2.26 (m, 2H), 2.19 (s, 6H); 13C
NMR (CDCl3, 75.5 MHz) d 142.18, 141.74, 139.65,
138.58, 132.25, 132.12, 132.05, 131.29, 129.80, 128.68,
128.58, 128.45, 128.26, 127.83, 127.73, 127.31, 127.29,
59.57, 45.44, 28.35, 28.27. APCIMS: 285 (M),
287(M+2).
5.5.13.
Dimethyl-(4-phenyl-4-pyridin-2-yl-but-3-enyl)-
amine (14). Procedure B.2 was carried out using (3-dim-
ethylaminopropyl) triphenylphosphomium bromide
(1.74 g, 4.0 mmol) from procedure B.1, 2.5 M nBuLi/
Hex (1.6 mL, 4.0 mmol), 2-pyridinyl phenyl ketone
(0.5 g, 2.7 mmol), and THF (5 mL). The product was
obtained in 0.68 g as a brown oil in 29% yield. TLC:
MeCl2/MeOH (10:1), Rf = 0.55. 1H NMR (CDCl3,
300 MHz) d 8.53–8.49 (m, 1H), 7.42–7.25 (m, 4 H),
7.16–7.13 (m, 2H), 7.01–6.98 (m, 1H), 6.88–6.77 (m,
2H), 2.41–2.36 (appt t, 2H, J = 6.9 Hz), 2.27–2.22 (appt
t, 2H, J = 7.7 Hz), 2.11 (s, 6H); 13C NMR (CDCl3,
75.5 MHz) d 158.45, 149.21, 141.82, 138.83, 136.36,
131.11, 129.94, 128.62, 127.41, 122.26, 121.81, 59.30,
45.32, 28.06. ESIMS: 253 (M+H). HRMS Calcd for
C17H20N2: 253.1701. Found: 253.1705.
5.5.17.
[4-(2-Methoxy-phenyl)-4-phenyl-but-3-enyl]-di-
methyl- amine (18). Procedure B.2 was carried out using
(3-dimethylaminopropyl) triphenylphosphomium bro-
mide (1.5 g, 3.6 mmol) from procedure B.1, 2.5 M nBu-
Li/Hex (1.4 mL, 3.6 mmol), 2-methoxybenzophenone
(0.5 g, 2.4 mmol), and THF (5 mL). The product was
obtained in 0.54 g as a white semisolid in 82% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.19. 1H NMR
(CDCl3, 300 MHz) d 7.32–6.95 (m, 9H), 6.24-6.19 (t,
1H, J = 7.4 Hz), 3.71 (s, 3H), 2.50-2.45 (m, 2H), 2.25-
2.18 (br s, 8H); 13C NMR (CDCl3, 75.5 MHz) d
157.19, 141.77, 139.01, 131.43, 128.76, 128.14, 127.62,
126.78, 126.34, 120.71, 111.26, 59.31, 55.64, 45.38,
28.40. ESIMS: 282 (M+H). HRMS calc for
C19H23NO: 282.1847. Found: 282.1858.
5.5.14. [4-(2-Chloro-phenyl)-4-phenyl-but-3-enyl]-dimeth-
yl-amine (15). Procedure B.2 was carried out using (3-
dimethylaminopropyl) triphenylphosphomium bromide
(1.5 g, 3.5 mmol) from procedure B.1, 2.5 M nBuLi/
Hex (1.4 mL, 3.5 mmol), 2-chlorobenzophenone (0.5 g,
2.3 mmol), and THF (5 mL). The product was obtained
in 0.6 g as a clear oil in 91% yield. TLC: MeCl2/MeOH
5.5.18.
[4-(3-Methoxy-phenyl)-4-phenyl-but-3-enyl]-di-
methyl- amine (19). Procedure B.2 was carried out using
(3-dimethylaminopropyl) triphenylphosphomium bro-
mide (2.4 g, 5.6 mmol) from procedure B.1, 2.5 M nBu-
Li/Hex (2.2 mL, 5.6 mmol), 3-methoxybenzophenone
(0.6 g, 2.8 mmol), and THF (15 mL). The product was
obtained in 0.75 g as a pale yellow oil in 91% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.19. 1H NMR
(CDCl3, 300 MHz) d 7.80–7.60 (m, 7H), 7.27–7.16 (m,
2H), 6.56–6.50 (m, 1H), 4.23 (s, 1H), 4.19 (s, 2H),
2.92–2.87 (m, 2H), 2.78–2.74 (m, 2H), 2.67 (s, 6H); 13C
NMR (CDCl3, 75.5 MHz) d 159.65, 159.54, 144.09,
142.75, 141.52, 140.00, 129.84, 129.11, 128.35, 127.23,
122.34, 120.00, 115.46, 113.34, 112.62, 112. 28, 59.60,
59.56, 55.24, 45.35, 28.23, 28.18. ESIMS: 282 (M+H).
HRMS Calcd for C19H23NO: 282.1851. Found:
282.1858.
1
(10:1), Rf = 0.22. H NMR (CDCl3, 300 MHz) d 7.71–
7.20 (m, 9H), 6.28–6.23 (t, 1H, J = 7.4 Hz), 2.65–2.63
(m, 2H), 2.37 (s, 6H), 2.33–2.25 (m, 2H); 13C NMR
(CDCl3, 75.5 MHz) d 141.24, 140.18, 138.45, 132.34,
132.21, 131.74, 130.04, 129.16, 128.74, 128.61, 127.64,
127.61, 127.23, 126.49, 126.25, 58.24, 44.43, 27.17,
27.08. APCIMS: 285 (M), 287(M+2). HRMS Calcd
for C18 H20ClN: 286.1355. Found: 286.1363.
5.5.15. [4-(3-Chloro-phenyl)-4-phenyl-but-3-enyl]-dimeth-
yl-amine (16). Procedure B.2 was carried out using (3-
dimethylaminopropyl) triphenylphosphomium bromide
(1.5 g, 3.5 mmol) from procedure B.1, 2.5 M nBuLi/
Hex (1.4 mL, 3.5 mmol), 3-chlorobenzophenone (0.5 g,
2.3 mmol), and THF (5 mL). The product was obtained
in 0.5 g as a pale yellow oil in 80% yield. TLC: MeCl2/
5.5.19.
[4-(4-Methoxy-phenyl)-4-phenyl-but-3-enyl]-di-
methyl- amine (20). Procedure B.2 was carried out using
(3-dimethylaminopropyl) triphenylphosphomium bro-
mide (1.5 g, 3.6 mmol) from procedure B.1, 2.5 M nBu-
Li/Hex (1.4 mL, 3.6 mmol), 4-methoxybenzophenone
(0.5 g, 2.4 mmol), and THF (10 mL). The product was
obtained in 0.34 g as a yellow-orange oil in 52% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.22. 1H NMR
(CDCl3, 300 MHz) d 7.52–7.15 (m, 7H), 6.97–6.95 (d,
1H, J = 8.1 Hz), 6.86–6.83 (d, 1H, J = 8.4 Hz), 6.11–
6.03 (m, 1H), 3.89 (s, 1.5H), 3.84 (s, 1.5H), 2.50–2.47
(m, 2H), 2.41–2.33 (m, 2H), 2.28 (s, 6H); 13C NMR
(CDCl3, 75.5 MHz) d 158.95, 158.79, 143.03, 142.57,
140.38, 135.31, 132.12, 131.08, 129.88, 128.40, 128.21,
127.48, 127.11, 124.96, 113.61, 59.64, 55.38, 45.29,
28.20, 28.03. ESIMS: 282 (M+H).
1
MeOH (10:1), Rf = 0.22. H NMR (CDCl3, 300 MHz)
d 7.39–7.09 (m, 9H), 6.12–6.07 (m, 1H), 2.51–2.46 (appt
t, 2H, J = 7.4 Hz), 2.37–2.33 (m, 2H), 2.27 (br s, 6H);
13C NMR (CDCl3, 75.5 MHz) d 144.58, 142.00,
139.48, 134.35, 134.31, 132.39, 132.25, 132..18, 129.90,
129.49, 128.81, 128.65, 128.43, 128.25, 127.98, 127.43,
127.16, 125.69, 59.58, 45.46, 45.42, 28.30, 28.25. APC-
IMS: 285 (M), 287(M+2).
5.5.16. [4-(4-Chloro-phenyl)-4-phenyl-but-3-enyl]-dimeth-
yl-amine (17). Procedure B.2 was carried out using (3-
dimethylaminopropyl) triphenylphosphomium bromide
(1.5 g, 3.5 mmol) from procedure B.1, 2.5 M nBuLi/
Hex (1.4 mL, 3.5 mmol), 4-chlorobenzophenone (0.5 g,
2.3 mmol), and THF (5 mL). The product was obtained
5.5.20. Dimethyl-(4-phenyl-4-o-tolyl-but-3-enyl)-amine
(21). Procedure B.2 was carried out using (3-dimethyla-

D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
8377
minopropyl) triphenylphosphomium bromide (1.9 g,
4.6 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(1.8 mL, 4.6 mmol), 2-methylbenzophenone (0.6 g,
3.1 mmol), and THF (10 mL). The product was ob-
tained in 0.4 g as a pale beige semisolid in 48% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.17. 1H NMR
(CDCl3, 300 MHz) d 7.25–7.16 (m, 8H), 7.12–7.10 (m,
1H), 6.25–6.20 (t, 0.82H, J = 7.4 Hz), 5.68–5.64 (t,
0.16H, J = 6.8 Hz), 2.55–2.49 (m, 0.38H), 2.45–2.40
(appt t, 1.65H, J = 7.8 Hz), 2.27 (s, 1H), 2.21 (s, 5H),
2.16 (br s, 1.6H), 2.13 (br s, 0.32H), 2.07 (s, 2.49H),
2.04 (s, 0.5H); 13C NMR (CDCl3, 75.5 MHz) d 141.95,
140.79, 139.02, 136.31, 130.10, 129.82, 129.08, 128.86,
128.15, 127.90, 127.23, 127.03, 126.82, 126.68, 126.20,
126.08, 125.65, 125.43, 59.49, 58.96, 45.13, 44.99,
27.71, 27.53, 20.36, 19.54. ESIMS: 266 (M+H). HRMS
Calcd for C19H23N: 266.1904. Found: 266.1909.
(M+H). HRMS Calcd for C18H19Cl2N: 320.0960.
Found: 320.0973.
5.5.24. [4,4-Bis-(4-chloro-phenyl)-but-3-enyl]-dimethyl-
amine (25). Procedure B.2 was carried out using (3-dim-
ethylaminopropyl) triphenylphosphomium bromide
(1.3 g, 3.0 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(1.2 mL, 3.0 mmol), 4,4-dichlorobenzophenone (0.5 g,
2.0 mmol), and THF (10 mL). The product was obtained
in 0.52 g as a yellow oil in 83% yield. TLC: MeCl2/MeOH
1
(10:1), Rf = 0.18. H NMR (CDCl3, 300 MHz) d 7.36–
7.08 (m, 9H), 6.10–6.06 (t, 1H, J = 7.2 Hz), 2.46–2.38
(m, 2H), 2.32–2.27 (m, 2H), 2.23 (s, 6H); 13C NMR
(CDCl3, 75.5 MHz) d 140.80, 140.63, 138.05, 133.34
133.13, 132.17, 131.22, 128.77, 128.59, 128.42, 128.10,
59.33, 45.30, 28.16. ESIMS: 320 (M+H). HRMS Calcd
for C18H19Cl2N: 320.0957. Found: 320.0973.
5.5.21. Dimethyl-(4-phenyl-4-m-tolyl-but-3-enyl)-amine
(22). Procedure B.2 was carried out using (3-dimethyla-
minopropyl) triphenylphosphomium bromide (1.9 g,
4.6 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(1.8 mL, 4.6 mmol), 3-methylbenzophenone (0.6 g,
3.1 mmol), and THF (10 mL). The product was ob-
tained in 0.35 g as a pale beige semisolid in 43% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.17. 1H NMR
(CDCl3, 300 MHz) d 7.37–6.99 (m, 9H), 6.08–6.03 (m,
1H), 2.52–2.47 (m, 2H), 2.35 (br s, 2H), 2.30 (br s,
2H), 2.27 (br s, 4H), 2.17 (s, 3H); 13C NMR (CDCl3,
75.5 MHz) d 143.22, 142.61, 140.23, 140.05, 137.95,
137.75, 130.50, 129.91, 128.37, 128.22, 127.94, 127.37,
127.15, 127.09, 126.98, 126.66, 124.64, 59.61, 45.30,
31.10, 28.08, 21.64. ESIMS: 266 (M+H).
5.6. Biology
5.6.1. [3H]-BTX experiment. In the sodium channel eval-
uation, the IC50, which represents the micromolar con-
centration of compound required to displace 50% of
specifically bound [3H]-BTX, was determined using rat
cerebral cortex synaptoneurosomes. In brief, rat fore-
brain membranes were incubated with [3H]-Batracho-
toxin (30–60 Ci/mmol). Reactions are carried out in
50 mM HEPES (pH 7.4) containing 130 mM choline
chloride at 37 ꢁC for 60 min. The reaction was terminat-
ed by rapid vacuum filtration of the reaction contents
onto glass fiber filters. Radioactivity trapped onto the
filters was determined and compared to control values
in order to ascertain any interactions of the test com-
pound with the sodium channel site 2 binding site.
Aconitine [1 lM] was used as a positive control.
5.5.22.
Dimethyl-(4-phenyl-4-p-tolyl-but-3-enyl)-amine
(23). Procedure B.2 was carried out using (3-dimethyla-
minopropyl) triphenylphosphomium bromide (2.6 g,
6.2 mmol) from procedure B.1, 2.5 M nBuLi/Hex
(2.5 mL, 6.2 mmol), 4-methylbenzophenone (0.6 g,
3.1 mmol), and THF (10 mL). The product was ob-
tained in 0.72 g as a pale white semisolid in 89% yield.
TLC: MeCl2/MeOH (10:1), Rf = 0.24. 1H NMR
(CDCl3, 300 MHz) d 7.42–7.10 (m, 9H), 6.13–6.09 (t,
1H, J = 7.1 Hz), 2.47–2.40 (m, 2H), 2.39–2.31 (m, 5H),
2.26 (br s, 6H); 13C NMR (CDCl3, 75.5 MHz) d
142.74, 136.77, 136.71, 129.89, 129.04, 128.90, 128.67,
128.31, 128.15, 127.39, 127.23, 126.98, 126.19, 59.74,
45.40, 28.22, 21.36, 21.18. ESIMS: 266 (M+H).
5.6.2. Cell culture and electrophysiology. Human embry-
onic kidney (HEK 293) cells stably expressing Nav1.2
were grown in DMEM/F12 media (Invitrogen Corp.,
CA, USA) supplemented with 10% fetal bovine serum,
penicillin (100 U/mL), streptomycin (100 lg/mL), and
geneticin (G418) (500 lg/mL; Sigma, MO, USA). Cells
were grown in a humidified atmosphere of 5% CO2
and 95% air at 37 ꢁC.
Sodium currents were recorded using the whole-cell con-
figuration of the patch-clamp technique with an Axo-
patch 200B amplifier (Axon Instruments, Foster City,
CA). All voltage protocols were applied using pCLAMP
8 software (Axon, USA) and a Digidata 1322 (Axon,
USA). Currents were amplified and low pass filtered
(2 kHz) and sampled at 33 kHz. Cells were plated on glass
coverslips and superfused with the following solution:
130 mM NaCl, 4 mM KCl, 1 mM CaCl, 5 mM MgCl2,
5 mM HEPES, and 5 mM glucose (pH adjusted to 7.4
with NaOH). Compounds were prepared as 100 mM
stock solutions in dimethylsulfoxide (DMSO) and diluted
to desired concentration in perfusion solution. The max-
imum DMSO concentration of .3% had no effect on cur-
rent amplitude. Borosilicate glass pipettes were pulled
using a Brown-Flaming puller (model P87, Sutter Instru-
ments Co, Novato, CA) and heat polished to produce
electrode resistances of 0.8–2.6 MX when filled with the
5.5.23. [4,4-Bis-(2-chloro-phenyl)-but-3-enyl]-dimethyl-
amine (24). Procedure B.2 was carried out using (3-dim-
ethylaminopropyl) triphenylphosphomium bromide
(1.3 g, 3.0 mmol) from procedure B.1, 2.5 M nBuLi/
Hex (1.2 mL, 3.0 mmol), 2,2-dichlorobenzophenone
(0.5 g, 2.0 mmol), and THF (10 mL). The product was
obtained in 0.43 g as a yellow oil in 68% yield. TLC:
MeCl2/MeOH (10:1), Rf = 0.26. 1H NMR (CDCl3,
300 MHz) d 7.40–7.14 9 (m, 8H), 6.04–5.99 (t, 1H,
J = 7.4 Hz), 2.48–2.43 (t, 2H, J = 7.8 Hz), 2.30–2.25 (t,
2H, J = 7.5 Hz), 2.22 (s, 6H); 13C NMR (CDCl3,
75.5 MHz) d 140.94, 138.48, 137.19, 134.64, 134.52,
133.88, 132.99, 132.22, 131.79, 130.27, 129.98, 128.79,
128.38, 126.59, 58.86, 45.44, 28.40. ESIMS: 320

8378
D. P. Hudgens et al. / Bioorg. Med. Chem. 14 (2006) 8366–8378
11. Wood, J. N.; Boorman, J. P.; Okuse, K.; Baker, M. D. J.
Neurobiol. 2004, 1, 55.
12. Hains, B. C.; Klein, J. P.; Saab, C. Y.; Craner, M. J.;
Black, J. A.; Waxman, S. G. J. Neurosci. 2003, 23, 8881.
13. Anger, T. A.; Madge, D. J.; Mulla, M.; Riddall, D. J.
Med. Chem. 2001, 44, 115.
following electrode solution: 130 mM CsCl, 1 mM
MgCl2, 5 mM MgATP, 10 mM HEPES, and 10 mM
BAPTA (pH adjusted to 7.4 with NaOH). Experiments
were performed at room temperature (20–22 ꢁC).
5.6.3. Data analysis. All data analysis was performed
using Clampfit 8 software (Axon Instruments, CA,
USA), Excel (Microsoft), and Origin 6.0 (Microcal Soft-
ware, MA, USA). Statistical analyses were performed
using a t test for normally distributed data, or the Wilco-
xon signed rank test for non-normalized data (Sigma Stat,
Jandel). Averaged data were presented as means stan-
dard error of the mean (SEM). Significance values of
p < 0.05 were considered.
14. Yu, F. H.; Catterall, W. A. Genome Biol. 1992, 8, 59.
15. Trainer, V. L.; Moreau, E.; Guedin, D.; Baden, D. G.;
Catterall, W. A. J. Biol. Chem. 1993, 268, 17114.
16. Worley, P. F.; Baraban, J. M. Proc. Natl. Acad. Sci. 1987,
84, 3051.
17. Denac, H.; Mevissen, M.; Scholtysik, G. Naunyn Schmi-
edebergs Arch. Pharmacol. 2000, 362, 453.
18. McQuay, H. J.; Smith, L. A.; Moore, R. A. Health Needs
Asses. (in press).
19. Sindrup, S. H.; Otto, M.; Finnerup, N. B.; Jensen, T. S.
Basic Clin. Pharmacol. Toxicol. 2005, 96, 399.
20. Kalso, E. Curr. Pharma. Des. 2005, 11, 3005.
21. Saarto, T.; Wiffen, P. J. Cochrane Collab. 2006, 1, 1.
22. Rice, A. S.; Hill, R. G. Annu. Rev. Med. 2006, 57, 535.
23. Amir, R.; Argoff, C. E.; Bennett, G. J.; Cummins, T. R.;
Durieux, M. E.; Gerner, P.; Gold, M. S.; Porreca, F.;
Strichartz, G. R. J. Pain 2006, 7, S1.
24. Miodownik, A.; Kreisberger, J.; Nussim, M.; Avnir, D.
Synthetic Commun. 1981, 11, 241.
25. Hoye, T. R.; Eklov, B. M.; Voloshin, M. Org. Lett. 2004,
6, 2567.
Acknowledgments
We thank NovaScreen, Inc. for technical support on the
[3H]-BTX studies, as well as Dr. H.A. Hartmann for the
kind gift of HEK 293 cells stably expressing hNav1.2.
We would also like to thank the National Institutes of
Health for financial support through NIH Grant
1RO1CA105435-01 and the Georgetown Drug Discovery
Program.
26. Zha, C.; Brown, G.; Brouillette, W. J. J. Med. Chem. 2004,
47, 6519.
27. Wang, G. K.; Gerner, P. Tricyclic antidepressants and
their analogues as long-acting local anesthetics and
analgesics. U.S. Patent 6,545,057 B2, 2003.
28. De Paulis, T. Acta. Chem. Scand. 1978, B32, 354.
29. Anderson, J. D.; Hansen, T. P.; Lenkowski, P. W.; Walls,
A. M.; Choudhury, I. M.; Schenck, H. A.; Friehling, M.;
Holl, G. M.; Patel, M. K.; Sikes, R. A.; Brown, M. L.
Mol. Cancer Ther. 2003, 2, 1149–1154.
References and notes
1. Barbui, C.; Hotopf, M. Br. J. Psychiatry 2001, 178, 129.
2. Cerbo, R.; Barbanti, P.; Fabbrini, G.; Pascali, M. P.;
Catarci, T. Headache 1998, 38, 453.
3. Ashina, S.; Bendsten, L.; Jensen, R. Pain 2004, 108, 108.
4. Collins, S. L.; Moore, R. A.; McQuay, H. J.; Wiffen, P. J.
Pain Symp. Manage. 2000, 6, 449.
30. Wang, K. G.; Russell, C.; Wang, S. Y. Pain 2004, 110,
166.
5. Bowsher, D. Eur. J. Pain 2003, 7, 1.
6. Atkinson, J. H.; Slater, M. A.; Williams, R. A.; Zisook, S.;
Patterson, T. L.; Grant, I.; Wahlgren, D. R.; Abramson,
I.; Garfin, S. R. Pain 1998, 76, 287.
7. McNeal, E. T.; Lewandowski, G. A.; Daly, J. W.;
Creveling, C. R. J. Med. Chem. 1985, 28, 381.
8. Song, J. H.; Ham, S. S.; Shin, Y. K.; Lee, C. S. Eur. J.
Pharmacol. 2000, 401, 297.
9. Wood, J. N.; Perl, E. R. Curr. Opin. Genet. Dev. 1999, 9, 328.
10. Waxman, S. G.; Dib-Hajj, S.; Cummins, T. R.; Black, J.
A. Proc. Natl. Acad. Sci. 1999, 96, 7635.
31. Zyczkowska, M. Pol. J. Chem. 1979, 53, 1311.
32. Winthrop, S. O.; Davis, M. A.; Myers, G. S.; Gavin, J. G.;
Thomas, R.; Barber, R. J. Org. Chem. 1962, 27, 230.
33. Bonvincino, G. E.; Arlt, H. G.; Pearson, K. M.; Hardy, R.
A. J. Org. Chem. 1961, 26, 2383.
34. Anderson, E. L.; Casey, J. E. J. Org. Chem. 1965, 30, 3959.
35. Cuvigny, T.; Normant, H. Organomet. Chem. Synth. 1971,
1, 237.
36. Staab, H. A.; Wendel, K. Justus Liebigs Ann. Chem. 1966,
694, 78.