Effects of log P and phenyl ring conformation on the binding of 5- phenylhydantoins to the voltage-dependent sodium channel

Article abstract of DOI:10.1021/jm960692v

Binding to the neuronal voltage-dependent sodium channel (NVSC) was evaluated for 12 5-phenylhydanteins which systematically varied either log P and/or 5-phenyl ring orientation. The linear correlation of log P with in vitro sodium channel binding activit

Full text of DOI:10.1021/jm960692v

+
+
602
J . Med. Chem. 1997, 40, 602-607
Effects of log P a n d P h en yl Rin g Con for m a tion on th e Bin d in g of
5-P h en ylh yd a n toin s to th e Volta ge-Dep en d en t Sod iu m Ch a n n el
Milton L. Brown,† George B. Brown,‡ and Wayne J . Brouillette*,†
Department of Chemistry and Department of Psychiatry and Behavioral Neurobiology, University of Alabama at Birmingham,
Birmingham, Alabama 35294
Received October 4, 1996X
Binding to the neuronal voltage-dependent sodium channel (NVSC) was evaluated for 12
5-phenylhydantoins which systematically varied either log P and/or 5-phenyl ring orientation.
The linear correlation of log P with in vitro sodium channel binding activity (log IC50) for
hydantoins 1-12 and diphenylhydantoin (DPH) (r2 ) 0.638) suggested that simple partitioning
into the lipid phase is important but not sufficient to account for the effects of hydantoins on
the NVSC. Comparisons among different hydantoins with the same log P but different low-
energy phenyl ring orientations revealed that, in addition to log P, the correct 5-phenyl
orientation is important for efficient binding.
The neuronal voltage-sensitive sodium channel (NVSC)
is a transmembrane protein complex1 which, in concert
with other voltage-dependent ion channels, is essential
for the generation of action potentials in neurons and
other excitable cells.2-4 Action potentials nominally
consist of two phases: (1) depolarization caused by the
rapid entry of sodium into the cell through the NVSC
and/or calcium entry into the cell through voltage-
sensitive Ca2+ channels and (2) potassium movement
out of the cell through voltage-sensitive K+ channels,
repolarizing the cell and setting the resting membrane
potential (-60 mV).5
The NVSC can exist in at least three states, and
transitions between the states involve voltage-depend-
ent rate constants. These states are (1) closed and
activatable (resting); (2) open (activated); and (3) closed
and nonactivatable (inactivated). Activation controls
the rate and voltage dependence of the increase in
intracellular sodium concentration following depolar-
ization. Inactivation influences the rate and voltage
dependence of positive potential returning to negative
potential as the activated sodium channel returns to the
resting level during maintained depolarization.1,6
transmembrane segments, although additional isoforms
have more recently been described.10
The NVSC undergoes specific binding with neurotox-
ins, and at least seven neurotoxin binding sites have
been identified.11 Site 2 interacts with grayanotoxin
and the alkaloids veratridine, aconitine, and batra-
chotoxin, resulting in persistent activation of the sodium
channel. A batrachotoxin derivative, [3H]batrachotoxi-
nin A 20-R-benzoate ([3H]BTX-B),12-14 has been exten-
sively used in in vitro radioligand binding assays for
evaluating the interactions of small molecules with the
NVSC. For example, the allosteric inhibition of [3H]-
BTX-B binding by the antimaximal electroshock (anti-
MES) anticonvulsant diphenylhydantoin (phenytoin or
DPH)15 at therapeutically relevant concentrations (40
µM) first implicated the NVSC as a potential anticon-
vulsant receptor site.16 Other classes of drugs also
interact with the NVSC and inhibit [3H]BTX-B binding,
including the local anesthetics and antiarrhythmics.
Consistently, these three classes of therapeutic agents
often possess overlapping pharmacological activities,
illustrated by the antiarrhythmic properties exhibited
by both the anticonvulsant DPH and the local anesthetic
lidocaine. Proposals for one-site17 and two-site18 models
continue to provide controversy concerning whether
anticonvulsants, local anesthetics, and antiarrhythmics
act at either the same or different sites on the NVSC.
Interconversions between NVSC states may be im-
portant in considering the modes of action of these
drugs. The distribution of channels among the three
states, resting, opened, and inactivated, is a function
of membrane potential and time.19,20 Compounds that
bind to the NVSC may, because of changes in access or
interconversion rates, have different binding affinities
to the resting, activated, or inactivated state of the
sodium channel. This provides one possible explanation
for the differences in binding and therapeutic profiles
among anticonvulsants, local anesthetics, and antiar-
rythmics.21,22 DPH binds to the inactivated state of the
NVSC in a voltage- and frequency-dependent manner,23
providing an explanation for its selective effects on
hyperactive versus normal neurons.
The structural morphology of neuronal sodium chan-
nels is species and tissue dependent. For example,
NVSC from eel electroplax consists of only a single
R-subunit.7 However, sodium channels from rat brain
contain three polypeptides: (1) a 260 kDa R-subunit,
(2) a 36 kDa â1-subunit, and (3) a 33 kDa â2-subunit,1,5,8
with the R-subunit independently sufficient for func-
tional expression.5 Furthermore, sodium channels found
in neurons have pharmacological and physiological
properties different from those in skeletal muscle or
cardiac cells.
NVSC subtypes from many species have been cloned,
and amino acid sequences have been determined for
examples from rat and human brain and skeletal and
cardiac tissue.4 For example, three major NVSC R-sub-
unit isoforms from rat brain have been designated as
type I, II, and III,1 with each containing six R-helical
* Address correspondence to this author.
Examination of membrane effects for DPH using
fluorescent fatty acid probes24 implied that DPH may
exert an inhibitory effect on the NVSC by simple
† Department of Chemistry.
‡ Department of Psychiatry and Behavioral Neurobiology.
X Abstract published in Advance ACS Abstracts, J anuary 15, 1997.
S0022-2623(96)00692-9 CCC: $14.00 © 1997 American Chemical Society

+
+
Binding of 5-Phenylhydantoins
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 603
tions placed the methyl of the N-ethyl group either above (7a )
or below (7b) the plane of the hydantoin ring. In contrast, for
compound 4, higher rotational energy barriers were observed,
and two torsion angle ranges of phenyl ring conformers were
identified within 2.0 kcal/mol of the lowest energy minimum
(Table 2). The lowest energy minimum within each range
positioned the o-methyl group either above (4a ) or below (4b)
the plane of the phenyl ring. For DPH the conformation of
the X-ray crystal structure31 was used. Other than the torsion
angles of the phenyl groups (see footnote k in Table 2), the
calculated torsion angles, bond angles, and bond lengths for
DPH were in close agreement with those of the crystal
structure.
Conformational searches on rotationally restricted 5-phenyl
analogs 8-12 were performed using RANDOM SEARCH. In
this approach all bonds that are part of the aliphatic ring,
except the fusion bond shared with the benzene ring (and the
hydantoin ring for 12), were selected for searching. The
method sets up to three of the selected bonds to random
torsional angles, and the resulting conformer is submitted to
energy minimization to locate the nearest local minimum. The
minimization is performed in two stages so that molecules with
energy higher than a cutoff value (the default value of 70 kcal/
mol was used) are eliminated early. The second stage of
minimization uses gradient termination to ensure complete
minimization. The resulting conformer is then compared to
all previous structures based upon an RMS match of the
atoms; if different, it is retained. The other default values
utilized were 1000 maximum perturbation/minimization cycles,
an RMS threshold of 0.2 Å (the minimum difference between
two conformations to be designated as unique), a convergence
threshold of 0.005, and six maximum hits (the number of times
each unique conformer must be identified to terminate the
search). Each search was repeated, starting with a different
conformer, to verify results. Table 2 reports all minima whose
energies were within 2 kcal/mol of the global minimum.
partitioning into the lipid bilayer. However, DPH
exhibits saturable binding to the NVSC, which is
associated with properties of specific receptor-ligand
interactions.25 Studies18 also indicate that sodium
channel blockade by DPH is affected by changes in
intracellular, but not extracellular, pH.
The effects of log P26 and 5-phenyl ring orientation27
on whole animal anticonvulsant activity have been
summarized. Our preliminary investigations suggested
that log P28 and 5-phenyl ring orientation29 may also
be important for the efficient binding of hydantoins to
the NVSC. Here we systematically evaluate the effects
of log P and NVSC binding for a structurally diverse
group of hydantoins 1-12 which possess a range of log
P values (0.96-2.96). Within this group, we also make
comparisons between structural isomers that vary
5-phenyl ring orientation while maintaining a constant
log P.
To estimate phenyl torsion angle ranges within 2 kcal/mol
of the lowest energy conformer for 8-12, each local energy
minimum was evaluated in the following manner. The bond
connecting the alkyl ring fragment to the phenyl ring was
removed, the N1,C5,C6,C7 torsion angle was rotated by 5°,
and the bond was reconnected. Atoms N1, C5, C6, and C7
were defined as an aggregate (fixing the torsion angle), and
the resulting conformer was energy-minimized. This was
performed in each direction (+ and -) from the local energy
minimum until a conformer was generated whose energy was
2 kcal/mol greater than the global energy minimum. The
results are given in Table 2.
Meth od s
Biologica l Da ta . All compounds 1-12 were evaluated for
relative binding to the NVSC (Table 2) in an in vitro assay
using rat cerebral cortex synaptoneurosomes. Results are
expressed as log IC50, where IC50 represents the micromolar
concentration of compound required to inhibit the specific
binding of [3H]BTX-B by 50%.
Con for m a tion a l An a lysis. All compounds were energy
minimized with the Tripos force field30 (SYBYL software) using
default bond distances and angles and neglecting electrostat-
ics. While all compounds were synthesized and evaluated as
racemic mixtures, for internal consistency the geometries of
1-12 were modeled in SYBYL using only the arbitrarily
chosen R-configuration at C5. Conformational searching was
utilized to estimate the energetically reasonable range of
torsion angles (N1,C5,C6,C7) for rotation of the phenyl ring
relative to the hydantoin ring.
For 1-7 we performed, using GRIDSEARCH, conforma-
tional searches of torsion angle N1,C5,C6,C7 rotated 0-179°
(for symmetrical phenyl rings in 1-3 and 5-7) or 0-359° (for
the unsymmetrical phenyl ring in 4) in 1° increments. Briefly,
GRIDSEARCH utilizes a torsion angle driver followed by
energy minimization (Tripos force field). For 1-6 the starting
position of the R1 group was fully extended with an N1,C5,-
C1′,C2′ torsion angle of 50°. After viewing the results, selected
conformations near each apparent local energy minimum were
submitted to an additional minimization to confirm that the
local energy minimum had been identified. For compound 4,
selected conformations with energies roughly 2 kcal/mol above
the global minimum were also energy-minimized after first
defining the N1,C5,C6,C7 atoms as a rigid aggregate (to fix
the N1,C5,C6,C7 torsion angle); this procedure more accurately
defined the range of torsion angles within 2 kcal/mol of the
global minimum. The results revealed that, for analogs 1-3
and 5-7, all conformations from 0 to (180° are within 2 kcal/
mol of the lowest energy minimum. A single energy minimum
was found for all but 7, whose two minimum energy conforma-
Ch em istr y
The syntheses of analogs 1-3,28 7,29 8,32 9,29 10-11,33
and 1229 were previously described. In the preparation
of hydantoin 11,33 we synthesized benzocylooctenone34
by literature methods, and this was converted to 11 in
30% yield via synthetic method B (see the Experimental
Section). New hydantoins 4, 5, and 6 were also pre-
pared by the Bucherer-Berg35 procedure from the ap-
propriate ketones.36 The starting ketones were synthe-
sized using Grignard reactions of o-, m-, and p-tolyl
nitrile and butylmagnesium bromide. Table 1 contains
selected data for the compounds prepared in this study.
Resu lts a n d Discu ssion
The neuronal voltage-sensitive sodium channel is a
site of action for the anticonvulsant diphenylhydantoin,
one of the most widely used agents for the treatment of
generalized seizures. Unfortunately, the exact location
and nature of this site has not yet been determined. We
have been interested in investigating structure-activity
relationships for the interactions of hydantoins with the
NVSC in order to define structural features that give

+
+
604 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4
Ta ble 1. Selected Properties for Hydantoins 1-12
Brown et al.
In a preliminary study we reported the NVSC binding
of 1 vs 9 and 7 vs 12, and the results suggested that
the 5-phenyl orientation may be important for the in
vitro binding of hydantoins to the NVSC.29 To further
investigate the effects of 5-phenyl orientation, indepen-
dent from the effects of log P, we designed, synthesized,
and evaluated NVSC binding for new hydantoins 4-6
and we also prepared and evaluated known spirohy-
dantoins 8-11 and tricyclic hydantoin 12. Compounds
4 and 8-12 all contain phenyl rings with more re-
stricted energetically favorable rotameric conformations
as compared to 1-3, 5, and 6. Compounds 5 and 6 serve
as isomeric models of 4 that do not experience signifi-
cant changes in phenyl ring conformational accessibility
as compared to 1-3.
To estimate the 5-phenyl ring orientations that are
most highly populated in 1-12, the torsional angles N1,-
C5,C6,C7 for all minimum energy conformers within 2
kcal/mol of the lowest energy minimum were calculated.
Around each energy minimum, the range of N1,C5,C6,-
C7 torsional angles that provided conformations with
energies e2 kcal/mol greater than the global energy
minimum was also calculated. As illustrated in Table
2, all phenyl ring orientations of compounds 1-3 and
5-7 meet the above criterion, while compound 4 exhib-
its one large and one narrow torsional angle range.
Compounds 8-11 represent a homologous series of
spirohydantoins containing phenyl rings that are con-
formationally restricted about N1,C5,C6,C7 due to a
5-alkyl side chain that is covalently bound to the ortho
position of the 5-phenyl ring, and the range of accessible
torsion angles defining the 5-phenyl conformations
increases for larger ring sizes. Compound 12 is much
more highly constrained.
compd
% isolated yield (method)
64 (A)
mp, °C
1
169-171
(165-166)a
204-205
(204-205)
119-122
(125-127)
144-145
162-164
158-160
176-177
238-240
(238-240)
240-242
250-255
(252-253)
201-202
(201-202)
154-156
2
3
65 (C)
91 (A)
4
5
20 (A)
56 (A)
60 (A)
54
6
7b
8
31 (A)
9b
10
91
51 (C)
11
30 (B)
54
12b
a
Numbers in parentheses are literature melting ranges: 1-3,
b
11 (ref 35); 8 (ref 32); 10 (ref 33). Data previously reported in
ref 29.
rise to efficient binding.28,29 This approach may lead
to new anticonvulsants with enhanced activity.
Previous investigations revealed that log P may be
important for whole animal anticonvulsant activity,26
and our preliminary studies suggested that both log P
and phenyl ring orientation may be important for in
vitro binding to the NVSC.28,29 In the present study we
have selected hydantoin analogs which provide more
systematic evaluations for each of these properties.
We conveniently estimated log P values for new
compounds, as we did in an earlier study,28 by the
addition of π values to the experimentally determined
log P of the appropriate parent structure. We thus
estimated the log P values in Table 2 for compounds
4-12 (the others were taken from ref 8) by adding the
π values for substituents at C5 and N1 to the log P for
5-phenylhydantoin (0.46)26 to provide the final log P.
The π values used were 0.50 for each CH2 or CH3 in
the alkyl chain at C5 and 0.56 for an N-CH3 substitu-
ent.37 For example, the log P for hydantoin 7 was
One notes in Table 2 that the sodium channel binding
activity increases in going from 8-10, with no further
increase in activity for 11. One possible explanation for
this trend is that increases in log P may correlate with
increased binding activity. However, an exception is
evident in that compound 11, which exhibits a larger
log P than 10, does not exhibit a corresponding increase
in sodium channel binding activity. Another possible
explanation for this trend is the inability of the 5-phenyl
substituent of the smaller ring systems to adopt a
preferred conformation. A comparison between hydan-
toins with identical log P values from the spiro series
(8-11) and the least restricted series (1-3) reveals that
1 exhibits roughly the same IC50 as 10 and that 2
exhibits roughly the same IC50 as 11, suggesting that
10 and 11 may each be able to adopt an optimum phenyl
ring orientation.
calculated as (log Pparent + π5-methyl +πN1-methylene
+
π
methyl) ) (0.46 + 0.50 + 0.56 + 0.50) ) 2.02. Since the
ring closure of 7 provides the bicyclic hydantoin 12, the
log P of 12 was calculated as (log P7 + πring closure) ) 2.02
+ (-0.50) ) 1.52.
We previously observed that structurally similar
5-alkyl-5-phenylhydantoin homologs 1-3 demonstrated
enhanced binding to the NVSC with increases in log P.28
However, as shown in Table 2, log P is not the only
important variable. Comparisons of structurally dis-
similar hydantoins with the same log P (e.g., 1 vs 7, 2
vs 11, 9 vs 12) demonstrate that, in addition to log P,
structural features influence the potency of binding to
the NVSC. This is further supported by the observation
that compounds 2 and 11, which have log P values
equivalent to that for DPH, do not exhibit comparable
binding potencies. The linear correlation of log P versus
sodium channel binding (log IC50) for 1-12 and DPH
(Figure 1) is only a moderately good model (r2 ) 0.64).
This result indicates that simple nonspecific partitioning
of hydantoins into the membrane lipid phase, which has
been suggested by others as a possible mechanism for
the effect of DPH on the NVSC,24 is not sufficient for
potent inhibition.
To further pursue this question, we prepared and
evaluated the isomeric series 4-6, for which all ex-
amples have the same log P. These compounds were
designed to have the same log P value (2.96) as the most
potent compound in this study, hydantoin 3. In this
series the o-methyl group of 4 restricts phenyl ring
orientation as compared to isomers 3, 5, and 6. Note
that even though log P is constant, there is a 5.8-fold
range in the sodium channel IC50 values. Interestingly,
compounds 3, 5, and 6 all exhibit potent binding to the
NVSC as compared to compound 4 (Table 2). This
suggests that compound 4 may not be able to signifi-
cantly populate the phenyl ring conformation required
for optimum sodium channel binding activity.

+
+
Binding of 5-Phenylhydantoins
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 605
Ta ble 2. Conformational Analysis Results, Lipophilicities, and Sodium Channel Binding Activities for Hydantoins 1-12 and DPH
torsion angle range, deg
N1,C5,C6,C7b
low-energy conformer torsion angle,
deg N1,C5,C6,C7c
energy,
Na+ channel
IC50 (µM)
hydantoina
kcal/mold
log P
1
0 to (180
0 to (180
4
4
7.8
8
1.96
(2.10)f
2.46
(2.63)
2.96
(3.16)
2.96
(3.08)
162e
[136-193]g
2
103e
[85-124]
39e
[32-47]
225
3
0 to (180
4
8.1
11
4a
-108 to 1
-37
[218-233]
4b
5
147-169
0 to (180
159
4
12
8.2
2.96
(3.13)
2.96
58
[57-60]
95
[86-106]
720n
6
0 to (180
0 to (180
89-159
4
8.2
(3.13)
2.02
7a
7b
8a
-42
-57
144
7
7
15
(-)h
0.96
2112
[2021-2207]
(0.91)
8b
9a
104
138
15
8.1
95-158
1.46
(1.47)
851i
[761-953]
9b
10a
115
87
8.2
14.3
67-102
1.96
(2.03)
251
[225-280]
10b
10c
10d
10e
10f
11a
107 - (-177)
163
135
118
152
107
153
13.7
15
15
15
16
83-178
-4 to 26
20
2.46
(2.59)
251
[228-277]
11b
12a
118
11
20
9.3
1.52
250n
40m
(-)h
12b
DPH
32-42
0 to (180
37
2j
10.9
7.1k
2.46l
(2.08)
a
b
Letters in parentheses designate different low-energy conformers when more than one was identified. The range of torsion angles
calculated to be within 2 kcal/mol of the lowest energy minimum. c The lowest energy conformation(s) within the indicated torsion angle
range. Calculated using the Tripos Force Field (Maximin2) within SYBYL 6.1 and neglecting electrostatics. e The sodium channel binding
d
data was taken from ref 28. f For comparison, numbers in parentheses are log P values calculated by the ClogP for Windows 1.0.1 program
g
h
distributed by BioByte Corp. of Claremont, CA. Range describing -1 to +1 standard deviation. Could not be calculated by ClogP due
i
j
to a missing fragment value. The sodium channel value was determined in the present study. Taken from the X-ray crystal structure
in ref 31. k The energy calculated for the x-ray structure. The energy calculated for the minimized structure (Tripos force field) (N1,C5,C6,C7
l
m
n
torsion angle ) -11°) was similar (6.9 kcal/mol). The experimental value from ref 37. Reference 15. Reference 29.
structurally dissimilar compounds exhibiting the same
log P values. In comparing 9 and 12 (log P ) 1.5), IC50
changes 3.4-fold; for 1, 7, and 10 (log P ) 2), IC50
changes 4.4-fold; for 2, 11, and DPH (log P ) 2.5), IC50
changes 6.3-fold. Within each group the minimum
energy conformation for the most potent examples
contain phenyl ring torsion angles (N1,C5,C6,C7) of -11
to 11°. Energy minima for the least potent compounds
contain considerably different phenyl ring torsion angles.
These results suggest an optimum phenyl torsional
angle range.
In conclusion, log P is clearly an important parameter
for the potent binding of hydantoins to the NVSC. This
may result from the necessity of hydantoins to enter or
cross the cell membrane prior to interaction with the
channel protein. However, the above studies, which
compare structurally diverse hydantoins with identical
log P, reveal that different hydantoins with the same
log P value may exhibit significantly different binding
affinities, suggesting that hydantoins likely bind to a
specific site on the NVSC. Furthermore, these results
suggest a preferred phenyl ring orientation that gives
rise to the optimum binding of 5-phenylhydantoins to
this site.
F igu r e 1. A linear correlation of log P versus sodium channel
binding activity for hydantoins 1-12 and DPH.
Besides structurally similar hydantoins 3-6 (log P
) 3), the compounds in Table 1 can also be divided into
three distinct groups of hydantoins which contain

+
+
606 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4
Brown et al.
yl)pentanone (3.2 g, 18 mmol), KCN (2.4 g, 36 mmol) and
(NH4)2CO3 (8.3 g, 73 mmol). The solution was warmed to 50-
60 oC for 48 h. After cooling to room temperature, the
precipitate was filtered and the filtrate was acidified (pH 2)
by the addition of concentrated HCl. The filtrate was concen-
trated to half-volume, cooled, and filtered again. The combined
solids were recrystallized from hot ethanol to give pure 5 (2.5
Exp er im en ta l Section
Melting points were recorded on an Electrothermal melting
point apparatus and are uncorrected. IR spectra were re-
corded on a Beckman Acculab 6 and Nicolet IR/42 spectrom-
eters, and elemental analyses were performed by Atlantic
Microlabs of Norcross, GA. 1H NMR and 13C NMR spectra
were recorded on GE (NT series) and Bruker (ARX series)
NMR spectrometers operating at 300.1 MHz (for 1H). The
spectra were obtained in DMSO-d6 (for hydantoins) and CDCl3
(for all other compounds) at ambient temperature and refer-
enced internally to tetramethylsilane (TMS). Mass spectra
were obtained on a Hewlett-Packard 5885 GC/MS. Spectral
data for new hydantoins 4-6 are summarized below.
[3H]Batrachotoxinin A 20-R-benzoate ([3H]BTX-B) with a
specific activity of 30 Ci/mol was obtained from New England
Nuclear (Boston, MA).
Meth od A. To a stirring solution of 50% ethanol were
added ketone (0.66 mol/L), KCN (1.33 mol/L) and (NH4)2CO3
(2.66 mol/L). The solution was warmed to 50-65 °C for 12 h.
After the solution was cooled to room temperature, the
hydantoin precipitate was filtered, and the filtrate was acidi-
fied (pH 2) by the addition of concentrated HCl to give more
precipitate, which was filtered again. The filtrate was con-
centrated to half-volume and cooled, and the hydantion
product that precipitated was filtered. The solids were
combined and recrystallized from hot ethanol to give the final
product.
1
g, 56% yield): mp 162-164 °C; H NMR (DMSO-d6) δ 7.42-
7.36 (m, 2H, Ph), 7.21-7.15 (m, 2H, Ph), 6.44-6.38 (s, 1H,
NH), 2.35-2.33 (s, 3H, CH3), 2.24-1.98 (m, 2H, CH2), 1.41-
1.19 (m, 4H, CH2), 0.90-0.84 (m, 3H, CH3); 13C NMR (DMSO-
d6) δ 175.9, 157.6, 138.5, 137.6, 129.1, 128.7, 125.9, 122.3, 68.9,
38.5, 25.8, 22.5, 21.6, 13.8; IR (KBr) 1750, 1705 (CdO) cm-1
;
MS (EI) 246 (M+). Anal. (C14H18N2O2) C, H, N.
5-Bu tyl-5-(4-m eth ylp h en yl)h yd a n toin (6). To a stirring
solution of 50% ethanol (30 mL) were added 1-(4-methylphen-
yl)pentanone (2.6 g, 15 mmol), KCN (1.9 g, 30 mmol), and
(NH4)2CO3 (6.7 g, 59 mmol). The solution was warmed to 50-
60 oC for 48 h. After the solution was cooled to room
temperature, the precipitate was filtered, and the filtrate was
acidified (pH 2) by the addition of concentrated HCl. The
filtrate was concentrated to half-volume, cooled, and filtered
again. The combined solids were recrystallized from hot
o
ethanol to give pure 6 (2.1 g, 60% yield): mp 158-160 C; 1H
NMR (DMSO-d6) δ 7.42-7.36 (m, 2H, Ph), 7.21-7.15 (m, 2H,
Ph), 6.44-6.38 (s, 1H, NH), 2.35-2.33 (s, 3H, CH3), 2.24-1.98
(m, 2H, CH2), 1.41-1.19 (m, 4H, CH2), 0.90-0.84 (m, 3H, CH3);
13C NMR (DMSO-d6) δ 175.5, 156.7, 138.3, 134.6, 129.6, 125.1,
68.8, 38.3, 25.8, 22.5, 21.0, 13.8; IR (KBr) 1710, 1700 (CdO)
cm-1; MS (EI) 246 (M+). Anal. (C14H18N2O2) C, H, N.
Meth od B. To a solution of 50% ethanol contained in a
300 mL Parr pressure apparatus were added ketone (0.7 mol/
L), KCN (1.3 mol/L), and (NH4)2CO3 (2.7 mol/L). The solution
was heated at 125 °C for 24 h, the apparatus was cooled to
room temperature, and the hydantoin precipitate was collected
by filtration. The filtrate was adjusted to pH 2, concentrated
to half-volume, cooled, and the hydantoin precipitate was
filtered. The crude solids were combined and recrystallized
in hot ethanol to give the final hydantoin product.
Sod iu m Ch a n n el Bin d in g Assa y. We previously reported
the details of this procedure.39 Briefly, synaptoneurosomes (∼1
mg of protein) from rat cerebral cortex were incubated for 40
min at 25 °C with the test compound (seven different concen-
trations spanning the IC50) in a total volume of 320 µL
containing 10 nM [3H]BTX-B and 50 µg/mL of scorpion venom.
Incubations were terminated by dilution with ice-cold buffer
and filtration through a Whatman GF/C filter paper, and the
filters were washed four times with ice-cold buffer. Filters
were counted in a Beckmann scintillation counter. Specific
binding was determined by subtracting the nonspecific bind-
ing, which was measured in the presence of 300 µM veratri-
dine, from the total binding of [3H]BTX-B. All experiments
were performed in triplicate and included a control tube
containing 40 µM DPH. The IC50 values were determined from
a Probit analysis of the dose-response curve and excluded
doses producing less than 10% or greater than 90% inhibition.
Meth od C. Ketone and trimethylsilyl cyanide (TMSCN)
were combined without solvent in a 1:2 molar ratio under
anhydrous conditions, and ZnI2 (5-10 mg) was added as a
catalyst. This mixture was stirred at room temperature under
a nitrogen atmosphere for 12 h. The reaction was monitored
by the disappearance of the CdO stretching peak in the IR
spectrum of the reaction mixture. The TMS ether was not
purified but was directly hydrolyzed to the cyanohydrin by
adding equal amounts of ether and 15% HCl and stirring
vigorously at room temperature for 1 h.38 The ether layer was
separated, and the acidic layer was washed three times with
ether. The ether extracts were combined and evaporated to
give cyanohydrin in 100% yield. The cyanohydrin was con-
verted to hydantoin35 by dissolving cyanohydrin and (NH4)2-
CO3 in a 1:2 molar mixture in 50% ethanol. The mixture was
Ack n ow led gm en t. This work was taken in part
from the Ph.D. dissertation submitted in partial fulfill-
ment of the requirements for the Ph.D. degree in
chemistry by M.L.B. M.L.B gratefully acknowledges
financial support from the Patricia Robert Harris Fel-
lowship, the National Consortium for Educational Ac-
cess, the UAB Comprehensive Minority Faculty Devel-
opment Program, and the UAB Department of
Chemistry. We also thank Ms. Bereaval Webb, a 1993
Alabama Alliance for Minority Participation (AMP)
summer intern, for technical support.
o
then heated at 55-65 C for 12 h. The reaction mixture was
adjusted to pH 2 by the addition of HCl, concentrated to half-
volume, and cooled, and the hydantoin precipitate was filtered.
The crude solid was recrystallized from hot ethanol to give
the final hydantoin product.
5-Bu tyl-5-(2-m eth ylp h en yl)h yd a n toin (4). To a solution
of 50% ethanol (80 mL) contained in a 300 mL Parr pressure
apparatus were added 1-(2-methylphenyl)pentanone (1.5 g, 8.5
mmol), KCN (1.1 g, 17 mmol), and (NH4)2CO3 (3.9 g, 34 mmol).
o
The solution was heated at 125 C for 24 h, and the apparatus
was cooled to room temperature. The precipitate was filtered,
and the filtrate was acidified (pH 2) by the addition of
concentrated HCl. The filtrate was concentrated to half-
volume, cooled in an ice bath, and filtered again. The
combined solids were recrystallized from hot ethanol to give
Refer en ces
(1) Catterall, W. A. Structure and Function of Voltage-Gated Ion
Channels. Annu. Rev. Biochem. 1995, 64, 493-531.
(2) Kirsch, G. E. Na+ Channels: Structure, Function, and Clas-
sification. Drug Dev. Res. 1994, 33, 263-276.
1
pure 4 (0.50 g, 20% yield): mp 144-145 °C; H NMR (DMSO-
(3) Catterall, W. A. Molecular Mechanisms of Inactivation and
Modulation of Sodium Channels. Renal Physiol. Biochem. 1994,
17, 121-125.
d6) δ 7.42-7.36 (m, 2H, Ph), 7.21-7.15 (m, 2H, Ph), 6.44-
6.38 (s, 1H, NH), 2.35-2.33 (s, 3H, CH3), 2.24-1.98 (m, 2H,
CH2), 1.41-1.19 (m, 4H, CH2), 0.90-0.84 (m, 3H, CH3); 13C
NMR (DMSO-d6) δ 177.0, 157.0, 137.0, 132.6, 128.2, 127.4,
126.1, 67.5, 36.7, 25.2, 22.3, 20.4, 14.1; IR (KBr) 1760, 1700
(CdO) cm-1; MS (EI) 246 (M+). Anal. (C14H18N2O2) C, H, N.
5-Bu tyl-5-(3-m eth ylp h en yl)h yd a n toin (5). To a stirring
solution of 50% ethanol (30 mL) were added 1-(3-methylphen-
(4) Kallen, R. G.; Cohen, S. A.; Barchi, R. L. Structure, Function
and Expression of Voltage-Dependent Sodium Channels. Mol.
Neurobiol. 1993, 7, 383-428.
(5) Catterall, W. A. Structure and Function of Voltage-Sensitive Ion
Channels. Science 1988, 242, 50-61.
(6) Armstrong, C. M. Voltage-Dependent Ion Channels and Their
Gating. Physiol. Rev. 1992, 72, S5-S13.

+
+
Binding of 5-Phenylhydantoins
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 607
(7) Noda, M.; Shimizu, S.; Tanabe, T.; Takai, T.; Kayano, T.; Ikeda,
T.; Takahashi, H.; Nakayama, H.; Kanaoka, Y.; Minamino, N.;
Kangawa, K.; Matsu, H.; Raftery, M. A.; Hirose, T.; Inayama,
S.; Hayashida, H.; Miyata, T.; Numa, S. Primary Structure of
Electrophorus-Electricus Sodium Channel Deduced from Comple-
mentary DNA Sequence. Nature 1984, 312, 121-127.
(8) Scheuer, T.; Auld, V. J .; Boyd, S.; Offord, J .; Dunn, R.; Catterall,
W. A. Functional Properties of Rat Brain Sodium Channels
Expressed in a Somatic Cell Line. Science 1990, 247, 854-858.
(9) Black, J . A.; Westenbroek, R.; Ransom, B. R.; Catterall, W. A.;
Waxman, S. G. Type II Sodium Channels in Spinal Cord
Astrocytes In Situ: Immunocytochemical Observations. Glia
1994, 12, 219-227.
(10) George, A. L., J r.; Varkony, T. A.; Drabkin, H. A.; Han, J .; Knops,
J . F.; Finley, W. H.; Brown, G. B.; Ward, D. C.; Haas, M.
Assignment of Human Heart Tetrodotoxin-Resistant Voltage-
Gated Na+ Channel R-Subunit Gene (SCN5A) to Band 3p21.
Cytogenet. Cell Genet. 1995, 68, 67-70.
(11) Fainzilber, M.; Kofman, O.; Zlotkin, E.; Gordon, D. A New
Neurotoxin Receptor Site on Sodium Channels is Identified by
a Conotoxin that Affects Sodium Channel Inactivation in Mol-
luscs and Acts as an Antagonist in Rat Brain. J . Biol. Chem.
1994, 269, 2574-2580.
(23) Catterall, W. A. Voltage Clamp Analysis of the Inhibitory Actions
of Diphenylhydantoin and Carbamazepine on Voltage-Sensitive
Sodium Channels in Neuroblastoma Cells. Mol. Pharmacol.
1985, 27, 549-558.
(24) Harris, W. E.; Stahl, W. L. Interactions of Phenytoin with Rat
Brain Synaptosomes Examined by Flourescent Fatty Acid
Probes. Neurochem. Int. 1988, 13, 369-377.
(25) Francis, J .; Burnham, W. M. [3H]Phenytoin Identifies a Novel
Anticonvulsant-Binding Domain on Voltage-Dependent Sodium
Channels. Mol. Pharmacol. 1992, 42, 1097-1103.
(26) Lien, E. Structure-Activity Correlations for Anticonvulsant
Drugs. J . Med. Chem. 1970, 13 ,1189-1191.
(27) Wong, M. G.; Defina, J . A.; Andrews, P. R. Conformational
Analysis of Clinically Active Anticonvulsant Drugs. J . Med.
Chem. 1986, 29, 562-572.
(28) Brouillette, W. J .; J estkov, V. P; Brown, M. L.; Aktar, M. S.;
Delorey, T. M.; Brown, G. B. Bicyclic Hydantoins with Bridge-
head Nitrogen. Comparison of Anticonvulsant Activities with
Binding to the Neuronal Voltage-Dependent Sodium Channel.
J . Med. Chem. 1994, 37, 3289-3293.
(29) Brouillette, W. J ; Brown, G. B.; Delorey, T. M; Liang, G. Sodium
Channel Binding and Anticonvulsant Activities of Hydantoins
Containing Conformationally Constrained 5-Phenyl Substitu-
ents. J . Pharm. Sci. 1990, 79, 871-874.
(30) Vinter, J . G.; Davis, A.; Saunder, M. R. Strategic Approaches to
Drug Design. 1. An Integrated Software Framework for Molec-
ular Modeling. J . Comput.-Aided Mol. Des. 1987, 1, 31-55.
(31) Camerman, A.; Camerman, N. The Stereochemical Basis of
Anticonvulsant Drug Action. I. The Crystal and Molecular
Structure of Diphenylhydantoin, a Noncentrosymmetric Struc-
ture Solved by Centric Symbolic Addition. Acta Crystallogr.
1971, B27, 2205-2211.
(32) Sarges, R.; Schnur, R. C.; Belletire, J . L.; Peterson, M. J . Spiro
Hydantoin Aldose Reductase Inhibitors. J . Med. Chem. 1988,
31, 230-243.
(33) Huisgen, R.; Ugi, I. Polycyclische Systeme mit Heteroatomen.
Liebigs Ann. Chem. 1957, 610, 57-66.
(34) Huisgen, R; Rapp, W.; Mitteil, I. 1,2-Benzo-cycloocten-(1)-on-(3)*.
Chem. Ber. 1952, 85, 826-835.
(35) Novelli, A.; Lugones, Z. M.; Velasco, P. Hydantoins III. Chemical
Constitution and Hypnotic Action. An. Asoc. Quim. Argent. 1942,
30, 225-231.
(36) Cazes, B; J ulia, S. Re`arrangements Sigmatropiques [2.3] en Se`rie
Benzylique; Pre`paration de Cetones Ortho-Me`thyl-Aryle`es. ([2,3]
Sigmatropic Rearrangements of a Benzylic Series; Preparation
of Ortho-methyl Aryl Ketones.) Tetrahedron 1979, 35, 2655-
2660.
(37) Lipinski, C. A.; Giese, E. F.; Korst, R. J . pKa, Log P and
MedChem CLOGP Fragment Values of Acidic Heterocyclic
Potential Bioisosteres. Quant. Struct.-Act. Relat. 1991, 10, 109-
117.
(38) Grunewald, G. L.; Brouillette, W. J .; Finney, J . Synthesis of
R-Hydroxyamides via the Cyanosilylation of Aromatic Ketones.
Tetrahedron Lett. 1980, 21, 1219-1220.
(12) Catterall, W. A.; Morrow, C. S.; Daly, J . W.; Brown, G. B. Binding
of Batrachotoxinin A 20-R-Benzoate to a Receptor Site Associated
with Sodium Channels in Synaptic Nerve Ending Particles. J .
Biol. Chem. 1981, 10, 8922-8927.
(13) Brown, G. B.; Tieszen, S. C.; Daly, J . W.; Warnick, J . E.;
Albuquerque, E. X. Batrachotoxin-A 20-R-Benzoate: A New
Radioactive Ligand for Voltage-Sensitive Sodium Channels. Cell
Mol. Neurobiol. 1981, 1, 19-40.
(14) Brown, G. B. Batrachotoxin: A Window on the Allosteric Nature
of the Voltage-Sensitive Sodium Channel. Int. Rev. Neurobiol.
1988, 29, 7-116.
(15) Willow, M.; Catterall, W. A. Inhibition of Binding of [3H]-
Batrachotoxinin A 20-R-Benzoate to Sodium Channels by the
Anticonvulsant Drugs Diphenylhydantoin and Carbamazepine.
Mol. Pharmacol. 1982, 22, 627-635.
(16) Catterall, W. A. Common Modes of Drug Action on Na+ Chan-
nels: Local Anesthetics, Antiarrhythmics and Anticonvulsants.
Trends Pharmacol. Sci. 1987, 8, 57-65.
(17) Zima´nyi, I.; Weiss, S. R. B.; Lajtha, A.; Post, R. M.; Reith, M. E.
A. Evidence for a Common Site of Action of Lidocaine and
Carbamazepine in Voltage-Dependent Sodium Channels. Eur.
J . Pharmacol. 1989, 167, 419-422.
(18) Barber, M. J .; Starmer, C. F.; Grant, A. O. Blockade of Cardiac
Sodium Channels by Amitriptyline and Diphenylhydantoin,
Evidence for Two Use-Dependent Binding Sites. Circ. Res. 1991,
69, 677-696.
(19) Koumi, S.; Sato, R; Katori, R.; Hisatome, I.; Nagasawa, K.;
Hayakawa, H. Sodium Channel States Control Binding and
Unbinding Behavior of Antiarrhythmic Drugs in Cardiac Myo-
cytes for the Guinea Pig. Cardiovasc. Res. 1993, 26, 1199-1205.
(20) Courtney, K. R.; Etter, E. F. Modulated Anticonvulsant Block
of Sodium Channels in Nerve and Muscle. Eur. J . Pharmacol.
1983, 88, 1-9.
(21) Catterall, W. A. Inhibition of Voltage-Sensitive Sodium Channels
in Neuroblastoma Cells by Antiarryhthmics Drugs. Mol. Phar-
macol. 1981, 20, 356-362.
(39) Brouillette, W. J ; Brown, G. B.; Delorey, T. M; Shirali, S. S. ;
Grunewald, G. L. Anticonvulsant Activities of Phenyl-Substi-
tuted Bicyclic 2,4-Oxazolidinediones and Monocyclic Models.
Comparison with Binding to the Neuronal Voltage-Dependent
Sodium Channel. J . Med. Chem. 1988, 31, 2218-2221.
(22) Catterall, W. A. Inhibition of Voltage-Sensitive Sodium Channels
in Neuroblastoma Cells and Synaptosomes by the Anticonvul-
sant Drugs Diphenylhydantoin and Carbamazepine. Mol. Phar-
macol. 1984, 25, 228-234.
J M960692V