Freezing the Butterfly Motion of Carbamazepine Derivatives

Article abstract of DOI:10.1055/s-0035-1560201

Atropisomeric properties were found in carbamazepine derivatives. The atropisomers of N-acyl and N-thiocarbamoyl derivatives of carbamazepine were isolated, with high stereochemical stability. It has been elucidated that the rotation about the N-C1′ axis around the outer amide N-(C=O) does not coordinate with the rotation of the butterfly-like motion.

Full text of DOI:10.1055/s-0035-1560201

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
© Georg Thieme Verlag Stuttgart · New York
2015, 47, 3907–3913
paper
3907
Y. Kanase et al.
Paper
Synthesis
Freezing the Butterfly Motion of Carbamazepine Derivatives
atropisomer A
(E/Z)
Yuki Kanase
R
R
O
O
Mai Kuniyoshi
E
Z
Hidetsugu Tabata
Yuka Takahashi
Susumu Kayama
Shintaro Wakamatsu
Tetsuta Oshitari
Hideaki Natsugari
Hideyo Takahashi*
fast
OMe
OMe
N
N
N
N
slow
slow
atropisomer B
(E/Z)
fast
OMe
OMe
E
Z
Faculty of Pharma Sciences, Teikyo University,
2-11-1 Kaga, Itabashi-ku, Tokyo 173-8605,
Japan
O
O
R
R
hide-tak@pharm.teikyo-u.ac.jp
Received: 23.07.2015
Accepted after revision: 04.08.2015
Published online: 09.09.2015
10
11
DOI: 10.1055/s-0035-1560201; Art ID: ss-2015-f0451-op
N
N
Abstract Atropisomeric properties were found in carbamazepine de-
rivatives. The atropisomers of N-acyl and N-thiocarbamoyl derivatives
of carbamazepine were isolated, with high stereochemical stability. It
has been elucidated that the rotation about the N–C1′ axis around the
outer amide N–(C=O) does not coordinate with the rotation of the but-
terfly-like motion.
H
O
NH2
carbamazepine (1)
iminostilbene (2)
O
O
O
Me
HN
HN
HN
S
Key words amides, atropisomerism, chirality, heterocycles, acylation
N
N
N
N
N
N
Me
O
O
Carbamazepine (5H-dibenz[b,f]azepine-5-carboxamide,
1) (Figure 1) is a widely used agent for the treatment of epi-
lepsy and trigeminal neuralgia.1 In order to study the struc-
ture–activity relationships of carbamazepine, many carba-
mazepine derivatives have been synthesized.2 A structural
study using X-ray analysis elucidated that the seven-mem-
bered ring of 1 is not planar, being described as a ‘butterfly-
like shape’.3 More specifically, the planar benzene rings in 1
are bent upwards or downwards from the puckered central
seven-membered ring (boat conformation). Such a tricyclic
ring system has been investigated energetically in structur-
ally related compounds (e.g., pirenzepine, telenzepine, and
nevirapine) (Figure 1).4 In the course of our stereochemical
and physicochemical analyses of biologically active mole-
cules,5 we became interested in the butterfly motion in 1.
Since nonplanarity of the central ring appears to be a re-
quirement for high activity, knowledge of the active confor-
mation of this seven-membered-ring system should be
helpful in developing the next-generation carbamazepine.
Although there has been little discussion of the conforma-
tion of carbamazepine (1) from the viewpoint of atrop-
isomerism,6 we anticipated that an atropisomeric property
should exist in association with the butterfly-like shape.
N
N
nevirapine
N
N
Me
Me
pirenzepine
telenzepine
Figure 1 Carbamazepine and structurally related compounds
While symmetric iminostilbene (2) (Figure 1) has no
chance of chirality (i.e., it is achiral), carbamazepine (1),
which has another planarity with the N-carbamoyl group,
theoretically has the origin of atropisomerism. In essence,
carbamazepine (1) might adopt the chiral conformations 1-
A and 1-B,7 which are closely related to the E/Z-amide rota-
tion about the N–C1′ axis (Scheme 1). Furthermore, we be-
came interested in the relationship between the inner sp2–
sp2 axes (ax1–ax4) and the outer sp2–sp2 amide axis (ax5).
In silico conformational analysis of pirenzepine carried out
by Trummlitz and co-workers4a elucidated that the butter-
fly motion should be accompanied by outer amide rotation.
This led us to the question of whether a similar coordinated
rotation of five axes occurred in carbamazepine. In this re-
spect, we considered that the interconversion of the tricy-
clic system (butterfly motion) in 1 is a result of the coordi-
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3908
Y. Kanase et al.
Paper
Synthesis
nated rotation (geared rotation) of four sp2–sp2 axes (ax1–
ax4). The focus is the effect of the outer E/Z-amide rotations
on the geared rotation of the inner tricyclic system. There-
fore, we examined the conformations of carbamazepine de-
rivatives by changing the N-carbamoyl group to an N-acyl
or N-thiocarbamoyl group, and investigating their butterfly
motion. Here, we report that atropisomers of N-acyl and N-
thiocarbamoyl derivatives of 1 were isolated, with high ste-
reochemical stability. The manner of the rotation of the in-
ner and the outer axes is discussed in detail.
Table 1 Synthesis of N-Acyl-5H-dibenz[b,f]azepinesa
R
R
NaH
AcCl
DMF
r.t., 1 h
N
N
H
2: R = H
3: R = OMe
Me
O
4: R = H, quant.
5: R = OMe, 96%
COCl
R
X1
X1
E/Z rotamers
ax3
ax4
(atropisomers)
KHMDS
ax1
ax2
THF
r.t., 1 h
N
N
N
X2
X1
ax5
H2N
O
6a–c
7a–c
O
1' NH2
O
X1
X2
1-A
1-B
R
X1
X2
Cl
Yield (%)
6a
6b
6c
7a
7b
7c
H
H
96
E/Z rotamers
(atropisomers)
H
Cl
Cl
98
N
N
H
Me
H
Me
Cl
99
H2N
O
O
NH2
OMe
OMe
OMe
quant
96
Cl
Cl
Scheme 1 Stereoisomers of carbamazepine (1)
Me
Me
95
a Reaction conditions for 4 and 5: 5H-dibenz[b,f]azepine (5.0 mmol), NaH
(6.0–7.5 mmol), AcCl (6.0–7.5 mmol), DMF (4 mL), 23 °C; for 6 and 7: 5H-
dibenz[b,f]azepine (1.0 mmol), KHMDS (1.5 mmol), acyl chloride (1.5
mmol), THF (4 mL), 23 °C.
1
First, the H NMR spectrum (CDCl3) of carbamazepine
(1) was examined. In the spectrum, the two alkene protons
(H10 and H11) were observed as a singlet peak, which im-
plies that 1 exists as an achiral compound at room tempera-
ture. Since the carbamoyl group is insufficient to induce
chirality in carbamazepine, we planned to synthesize vari-
ous N-acyl derivatives in anticipation of hindrance to the
rotation of the axes. As shown in Table 1, iminostilbene (2)
and 10-methoxy-5H-dibenz[b,f]azepine (3) were acylated
successfully following the conventional synthetic route.
The conformation of compounds 4 and 6a–c was exam-
ined based on the 1H NMR spectra (CDCl3). Interestingly,
each of the alkene protons (H10 and H11) of N-acetylated
compound 4 was observed as a separated peak, at δ 6.90
and 6.96 ppm (each 1 H, doublet), which indicates that the
protons are differentiated from each other. In contrast, in
the spectrum of N-4-chlorobenzoyl compound 6a, the two
alkene protons (7.05 ppm) are broad singlets and the aro-
matic protons (6.70–7.80 ppm) are also broad multiplets. In
compound 6b, which bears an N-2,4,6-trichlorobenzoyl
group, each alkene proton was observed as a separated
peak, at δ 6.94 and 6.98 ppm (each 1 H, doublet), and the
aromatic protons (7.01–7.58 ppm) were also sharp. Similar
results were observed in the spectrum of compound 6c
bearing the N-2,4,6-trimethylbenzoyl group. We succeeded
in obtaining 6b as a single crystal, and its X-ray analysis8
yielded important information (Figure 2).
Figure 2 Conformation of compound 6b
In compound 6b, the dihedral angle (φ C2–N–C7′–O) of
178.9° confirmed the planarity of the outer amide moiety.
The dihedral angle (φ C2′–C1′–C7′–O) of 79.0° also con-
firmed that the benzene ring is nearly orthogonal to the
carbonyl group. It is obvious that the substituents at the 2′-
and 6′-positions of the benzoyl affect the C7′–C1′ axis, to
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3909
Y. Kanase et al.
Paper
Synthesis
form a twisted conformation. These results imply the possi-
bility of the atropisomeric property arising from the steri-
cally encumbered benzoyl moiety.
and co-workers have reported a similar energy barrier for
the racemization of telenzepine based on unique biological
experiments.4b
We therefore attempted to separate the atropisomers of
4, 6b, and 6c using chiral columns; however, incomplete
separation was observed, and we failed to isolate each en-
antiomer at room temperature. Considering that the E/Z-
amide rotation around the outer [N–(C=O)] bond of com-
pounds 4 and 6a–c is essentially related to the axial chirali-
ty (Scheme 1), it is impossible to evaluate the rotational
barrier of the coordinated rotation of the four axes (ax1–
ax4) independently. The same is true of carbamazepine (1),
in which the rotational barrier of the four sp2–sp2 axes
(ax1–ax4) could not be estimated.
slow
MeO
OMe
10
R
R
1'
10
1'
N
N
E
E
O
O
fast
fast
slow
O
O
MeO
OMe
10
10
In order to distinguish the rotational barrier of the inner
sp2–sp2 axes (ax1–ax4) of the tricyclic system from that of
the outer amide bond (N–C1′ axis: ax5), we decided to in-
troduce a 10-methoxy group to the tricyclic system
(Scheme 2). The 10-methoxy substituent may also encum-
ber the coordinated rotation of the sp2–sp2 axes (ax1–ax4)
to provide more stable atropisomers. For N-acetylated com-
pound 5, two sets of resonances corresponding to E/Z-iso-
1'
1'
N
N
Z
Z
R
R
atropisomer A (E/Z)
5: R = Me
atropisomer B (E/Z)
7a: R = 4-chlorophenyl
7b: R = 2,4,6-trichlorophenyl
7c: R = 2,4,6-trimethylphenyl
1
mers (E/Z = 1:1) were observed in the H NMR spectrum
Scheme 2 Four stereoisomers in compounds 5 and 7a–c
(CDCl3). The E/Z-isomers, however, could not be separated
at room temperature using nonchiral HPLC, which implies
that the energy barrier between the E/Z-isomers of 5 is
low.9 On the other hand, the atropisomers of 5 (5-A and 5-
B) were separated at room temperature on a chiral column
(Chiralpak IA). It is obvious that the geared rotation of the
sp2–sp2 axes (ax1–ax4) is not accompanied by the rotation
of the N–C1′ axis (ax5) which provides E/Z-amide rotamers
(Scheme 2). The atropisomers of compound 5 (5-A and 5-B)
were successfully isolated using preparative chiral HPLC,
and stereochemical stability with a ΔG‡ value of 107.7
kJ/mol was observed.10 It is interesting that the coordinated
rotation of the sp2–sp2 axes (ax1–ax4) is completely re-
stricted by the less-hindered acetyl group. Meanwhile, the
1H NMR spectrum of N-4-chlorobenzoylated compound 7a
showed broad signals at room temperature, so that the E/Z
conformers were not distinguished; however, chiral HPLC
(Chiralpak IC) results suggest that compound 7a exists as
racemates of separable atropisomers at room temperature.
The atropisomers (7a-A and 7a-B) were isolated by prepar-
ative chiral HPLC, with a ΔG‡ value of 95.3 kJ/mol.10 It must
Interestingly, 10-methoxy-5-(2,4,6-trimethylbenzoyl)-
5H-dibenz[b,f]azepine (7c) provided two sets of resonanc-
1
es, corresponding to the E and Z conformations, in the H
NMR spectrum (E/Z = 1.6:1). Although 7c was observed as
one peak on nonchiral HPLC, the separation of the stereo-
isomers including the E/Z-isomers partially succeeded at
2 °C on a chiral column: three peaks [tR = 32.3 min (50%),
48.9 min (32%), and 60.0 min (18%)] were separated on a
Chiralpak ID column (Figure 3). Based on the ratio of each
peak, we determined that the first peak (tR = 32.3 min) con-
sisted of the enantiomer of 7c-E (major diastereomer) and
the enantiomer of 7c-Z (minor diastereomer), and thus, the
second peak (tR = 48.9 min) contained another enantiomer
of 7c-E (major diastereomer) and the third peak (tR = 60.0
min) contained another enantiomer of 7c-Z (minor diaste-
reomer). We succeeded in isolating these three peaks by
preparative chiral HPLC. Since the absolute stereochemistry
has not yet been determined, we tentatively called the iso-
lated second peak (tR = 48.9 min) 7c-E-B, and the third peak
(tR = 60.0 min) 7c-Z-A. The pure enantiomer (7c-E-B) was
converted into a mixture of diastereomers (7c-E-B/7c-Z-B =
ca. 1.5:1) after standing at room temperature for two hours.
Similarly, the isolated pure enantiomer (7c-Z-A) was con-
verted into a mixture of diastereomers (7c-E-A/7c-Z-A = ca.
1.8:1) after standing at room temperature for 2 hours. Thus,
it is apparent that the interconversion between 7c-E-B and
7c-Z-B, and between 7c-Z-A and 7c-E-A, at room tempera-
ture should originate from the E/Z rotation about the N–C1′
axis (ax5) and not from the coordinated rotation of the in-
ner axes (ax1–ax4). All of these results make it clear that
1
be noted that the broad signals of 7a in the H NMR spec-
trum do not reflect the stability of the stereoisomers aris-
ing from the four inner axes (ax1–ax4). For compound 7b
bearing the N-2,4,6-trichlorobenzoyl group, two sets of res-
onances corresponding to the E and Z conformations were
1
observed in the H NMR spectrum (E/Z = 1.8:1). Although
compound 7b was observed as one peak on nonchiral HPLC,
each atropisomer (7b-A and 7b-B) was isolated at room
temperature on a preparative chiral column (Chiralpak IB),
with a ΔG‡ value of 135.3 kJ/mol.10 Such high stability of the
enantiomers of 7b might be surprising, although Eveleigh
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3910
Y. Kanase et al.
Paper
Synthesis
the coordinated rotation of the sp2–sp2 axes (ax1–ax4) at
Ar–N(C=O) and Ar–(C(OMe)=CH) has no relation to the N–
C1′ axis that provides E/Z-amide rotamers.
ers were not differentiated and suggests a low energy barri-
er of the rotation of the N–C1′ axis (ax5). Although com-
pound 8 provided one peak on nonchiral HPLC, the atrop-
isomers were partially separated at room temperature on a
chiral column (Chiralpak IA). Compared with the N-acetyl-
ated 10-methoxy-5H-dibenz[b,f]azepine 5, in which the at-
ropisomers were isolated, the energy barrier between the
atropisomers of N-carbamoyl compound 8 is low. Therefore,
it is clear that the N-carbamoyl group is not sufficient to
hinder the coordinated rotation of the sp2–sp2 axes (ax1–
ax4). The electronic or steric property of the thiocarbamoyl
group should be more effective than that of the carbamoyl
group. In the 1H NMR spectrum of N-thiocarbamoylated
5H-dibenz[b,f]azepine 9, the aromatic protons (7.38–7.59
ppm) and the two alkene protons (H10 and H11) (6.92–7.04
ppm) are partially sharp. We attempted to separate the ste-
reoisomers of 9 using chiral columns; however, no separa-
tion was achieved. In contrast, the 1H NMR spectrum of N-
thiocarbamoylated 10-methoxy-5H-dibenz[b,f]azepine 10
provided two sets of resonances corresponding to the E and
Z conformations (E/Z = 1:1).9 Although thiocarbamoylated
compound 10 was observed as one peak on nonchiral HPLC,
the atropisomers were separated at room temperature on a
chiral column (Chiralpak IB), which means that the N–C1′
axis (ax5) can allow rotation even though the rotation of
the inner axes (ax1–ax4) is hindered. The enantiomers of
compound 10 (10-A and 10-B) were successfully isolated
using preparative chiral HPLC, and stereochemical stability
with a ΔG‡ value of 100.7 kJ/mol was observed.10
Figure 3 Separation of 7c
We next investigated the thiocarbamoyl group (deriva-
tives 9 and 10) compared with the carbamoyl group (com-
pound 8) (Scheme 3).
In conclusion, the stereochemistry of carbamazepine
and its derivatives (N-carbamoyl-, N-thiocarbamoyl-, and
N-acyl-substituted 5H-dibenz[b,f]azepines) was investigat-
ed from the viewpoint of atropisomerism. Although the at-
ropisomers of compounds 4 and 6 were not isolated, those
of 10-methoxy-substituted compounds 5, 7a–7c, and 10
were isolated. The 10-methoxy substituent enables the dis-
crimination of the rotational barrier of the sp2–sp2 axes
(ax1–ax4) of the tricyclic system from that of the outer am-
ide bond (N–C1′ axis: ax5). It should be noted that the ener-
gy barrier of the rotation about the N–C1′ axis (ax5) is low-
er than that of the coordinated rotation of the sp2–sp2 axes
(ax1–ax4). Hence, it was revealed that the E/Z-amide or
E/Z-urea rotamers exist with little relation to the coordinat-
ed rotation of the inner sp2–sp2 axes. In contrast to the sta-
bility of the atropisomeric conformers of the N-carbamoyl
derivative, the N-acyl and N-thiocarbamoyl derivatives of
10-methoxy-5H-dibenz[b,f]azepine provided sufficient sta-
bility to isolate each atropisomer. We hope that this under-
standing of the unique stereochemical properties originat-
ing in the stereogenic elements at the four sp2–sp2 axes at
Ar–N(C=O) and Ar–(CR=CH) in carbamazepine derivatives
will contribute to next-generation carbamazepine drug de-
sign.
O
R
10
R
11
O
XCN
CH2Cl2, r.t., 16 h
for 8
DCE, 80 °C, 24 h
for 9, 10
N
1'
N
H
2: R = H
3: R = OMe
H2N
X
8: X = O, R = OMe, 80%
9: X = S, R = H, 78%
10: X = S, R = OMe, 57%
Scheme 3 Synthesis of N-carbamoyl and N-thiocarbamoyl compounds
So far, little attention has been paid to the properties of
the thiocarbamoyl group from the viewpoint of atropisom-
erism. We anticipated that the planar thiourea moiety
would be a source of atropisomerism and examined its ste-
ric or electronic influence on the conformation of the tricy-
clic system. The synthesis is shown in Scheme 3, and the
conformation of compounds 8, 9, and 10 was examined
based on the 1H NMR spectra (CDCl3) at room temperature.
In the spectrum of N-carbamoylated 10-methoxy-5H-
dibenz[b,f]azepine 8, only a single conformer with sharp
peaks was observed, which indicates that the E/Z conform-
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3911
Y. Kanase et al.
Paper
Synthesis
mixture was stirred at 23 °C for 1 h. The reaction was quenched with
ice–water and the mixture was extracted with EtOAc. The organic lay-
er was washed with sat. aq NaHCO3, dried over MgSO4, and concen-
trated in vacuo. The residue was purified by silica gel column chroma-
tography (EtOAc–hexane, 1:2) to give 6a as a white solid; yield: 318
mg (0.960 mmol, 96%); mp 154–157 °C.
Materials were obtained from commercial suppliers. NMR spectra
were recorded on a JEOL-ECA600 spectrometer at 600 MHz for 1H
NMR and at 150 MHz for 13C NMR acquisitions. Chemical shifts are
given in parts per million (ppm) downfield from TMS as internal stan-
dard, and coupling constants (J) are reported in hertz (Hz). Splitting
patterns are abbreviated as follows: singlet (s), doublet (d), triplet (t),
quartet (q), multiplet (m), and broad (br). IR spectra were recorded on
a JASCO FT/IR-4200 FT-IR spectrophotometer equipped with ATR
(Diamond). High-resolution mass spectra (HRMS) were obtained on a
Shimadzu LCMS-IT-TOF mass spectrometer operated in the ESI mode.
Melting points were recorded on a Yanaco micro melting point appa-
ratus and are uncorrected. Optical rotations were determined with a
JASCO P-2200 digital polarimeter. Analytical thin-layer chromatogra-
phy was performed on precoated, glass-backed silica gel plates. Col-
umn chromatography was performed using silica gel (45–60 μm).
HPLC was performed with a Shimadzu Prominence system. Extract
solutions were dried over anhydrous MgSO4 or Na2SO4. Solvents were
evaporated under reduced pressure.
IR (ATR): 1650 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.80–6.70 (br, 8 H), 7.15 (m, 2 H), 7.11
(m, 2 H), 7.05 (br, 2 H).
13C NMR (150 MHz, CDCl3): δ = 168.8, 135.7, 133.8, 129.7, 129.1,
128.4, 128.0, 127.5; several signals overlap.
HRMS (ESI): m/z [M + Na]+ calcd for C21H14ClNO: 354.0656; found:
354.0642.
5-(2,4,6-Trichlorobenzoyl)-5H-dibenz[b,f]azepine (6b)
White solid; yield: 589.7 mg (98%); mp 219–220 °C.
IR (KBr): 1667 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.58 (d, J = 8.4 Hz, 1 H), 7.49 (dt, J = 8.4,
1.2 Hz, 1 H), 7.41 (dd, J = 7.8, 1.2 Hz, 1 H), 7.37 (m, 2 H), 7.31 (d, J = 1.8
Hz, 1 H), 7.28 (dd, J = 8.4, 1.2 Hz, 1 H), 7.19 (t, J = 7.8 Hz, 1 H), 7.12 (dt,
J = 7.8, 1.2 Hz, 1 H), 7.01 (d, J = 1.8 Hz, 1 H), 6.98 (d, J = 11.4 Hz, 1 H),
6.94 (d, J = 11.4 Hz, 1 H).
13C NMR (150 MHz, CDCl3): δ = 164.3, 138.6, 138.0, 135.0, 134.9,
134.1, 133.7, 133.6, 132.3, 130.7, 130.3, 129.7, 129.5, 129.2, 128.8,
128.3, 128.2, 127.9, 127.8, 127.7, 126.1.
5-Acetyl-5H-dibenz[b,f]azepine (4); Typical Procedure
To a solution of iminostilbene (2; 965 mg, 5.0 mmol) in DMF (4 mL) at
0 °C under argon were added successively NaH (60% in oil; 240 mg,
6.0 mmol) and AcCl (0.47 mL, 6.0 mmol), and the mixture was stirred
at 23 °C for 1 h. The reaction was quenched with ice–water and the
mixture was extracted with EtOAc. The organic layer was washed
with sat. aq NaHCO3, dried over MgSO4, and concentrated in vacuo.
The residue was purified by silica gel column chromatography
(EtOAc–hexane, 1:2) to give 4 as a white solid; yield: 1.27 g (quant);
mp 106 °C.
HRMS (ESI): m/z [M + H]+ calcd for C21H12Cl3NO: 400.0057; found:
400.0056.
IR (ATR): 1677 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.45–7.29 (m, 8 H), 6.96 (d, J = 12 Hz, 1
H), 6.90 (d, J = 12 Hz, 1 H), 1.85 (s, 3 H).
5-(2,4,6-Trimethylbenzoyl)-5H-dibenz[b,f]azepine (6c)
White solid; yield: 337.6 mg (99%); mp 172–175 °C.
13C NMR (150 MHz, CDCl3): δ = 170.8, 140.5, 140.3, 134.7, 133.5,
131.5, 129.3, 129.3, 129.3, 129.2, 129.0, 128.5, 127. 9, 127.9, 127.4,
22.1.
HRMS (ESI): m/z [M + Na]+ calcd for C16H13NO: 258.0889; found:
258.0886.
IR (ATR): 1661 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.53 (d, J = 7.2 Hz, 1 H), 7.49 (dt, J = 7.2,
1.2 Hz, 1 H), 7.40 (dd, J = 7.2, 1.2 Hz, 1 H), 7.34 (dt, J = 7.2, 1.2 Hz, 1 H),
7.21 (dd, J = 7.2, 1.2 Hz, 1 H), 7.10 (dt, J = 7.2, 1.2 Hz, 1 H), 7.00 (m, 3
H), 6.93 (d, J = 12 Hz, 1 H), 6.80 (s, 1 H), 6.48 (s, 1 H), 2.66 (s, 3 H), 2.14
(s, 3 H), 1.92 (s, 3 H).
13C NMR (150 MHz, CDCl3): δ = 171.2, 139.7, 139.6, 137.7, 134.5,
134.0, 133.8, 133.3, 133.3, 130.9, 130.4, 129.5, 129.4, 128.9, 128.7,
128.6, 127.9, 127.8, 127.5, 127.3, 127.1, 21.0, 20.0, 19.5.
HRMS (ESI): m/z [M + H]+ calcd for C24H21NO: 340.1696; found:
340.1707.
5-Acetyl-10-methoxy-5H-dibenz[b,f]azepine (5)
White solid; yield: 2.544 g (96%); Z/E = 1:1; mp 118–120 °C.
IR (ATR): 1674 cm–1
.
1H NMR (600 MHz, CDCl3): δ (Z-isomer) = 7.76 (d, J = 8.4 Hz, 1 H), 7.46
(m, 1 H), 7.41–7.23 (m, 6 H), 6.11 (s, 1 H), 3.87 (s, 3 H), 1.90 (s, 3 H); δ
(E-isomer) = 7.73 (d, J = 8.4 Hz, 1 H), 7.46 (m, 1 H), 7.41–7.23 (m, 6 H),
6.18 (s, 1 H), 3.90 (s, 3 H), 1.90 (s, 3 H).
5-(4-Chlorobenzoyl)-10-methoxy-5H-dibenz[b,f]azepine (7a)
White solid; yield: 372.7 mg (quant); mp 184 °C.
13C NMR (150 MHz, CDCl3): δ = 170.5, 170.2, 156.9, 154.9, 140.5,
139.2, 138.8, 134.5, 132.9, 132.7, 131.6, 130.3, 130.2, 129.1, 128.9,
127.9, 127.9, 127.7, 127.4, 127.3, 127.3, 127.2, 127.1, 102.8, 100.9,
55.4, 22.1; E/Z-isomers were not distinguished.
HRMS (ESI): m/z [M + H]+ calcd for C17H15NO2: 266.1176; found:
266.1182.
IR (ATR): 1652 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.80–6.80 (br, 8 H), 7.19 (d, J = 8.4 Hz, 2
H), 7.12 (d, J = 8.4 Hz, 2 H), 6.26 (br, 1 H), 3.96 (s, 3 H).
13C NMR (150 MHz, CDCl3): δ = 168.2, 135.8, 133.7, 130.1, 129.9,
128.9, 128.0, 127.8, 127.4, 127.3, 127.2, 127.1, 55.6; several signals
overlap.
Chiral HPLC separation: shorter tR: [α]D20 +256.6 (c 0.24, CHCl3); 100%
ee; longer tR: [α]D20 –250.3 (c 0.22, CHCl3); 98.6% ee.
HRMS (ESI): m/z [M + Na]+ calcd for C22H16ClNO2: 384.0762; found:
384.0755.
5-(4-Chlorobenzoyl)-5H-dibenz[b,f]azepine (6a); Typical Proce-
dure
Chiral HPLC separation: shorter tR: [α]D20 +107.1 (c 0.15, CHCl3); 96.8%
ee; longer tR: [α]D20 –104.3 (c 0.15, CHCl3); 96.5% ee.
To a solution of iminostilbene (2; 193 mg, 1.0 mmol) in THF (4 mL) at
0 °C under argon were added successively 0.5 M KHMDS (3.0 mL, 1.5
mmol) and 4-chlorobenzoyl chloride (0.20 mL, 1.5 mmol), and the
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3912
Y. Kanase et al.
Paper
Synthesis
10-Methoxy-5-(2,4,6-trichlorobenzoyl)-5H-dibenz[b,f]azepine
HRMS (ESI): m/z [M + H]+ calcd for C16H14N2O2: 267.1128; found:
(7b)
267.1123.
White solid; yield: 413.3 mg (96%); Z/E = 1:1.8; mp 205–206 °C.
5H-Dibenz[b,f]azepine-5-carbothioamide (9)
IR (ATR): 1671 cm–1
.
To a solution of iminostilbene (2; 96.6 mg, 0.5 mmol) in DCE (3 mL) at
r.t. under argon was added FmocNCS (198.9 mg, 0.71 mmol), and the
mixture was stirred at 80 °C for 24 h. When the starting material was
diminished, piperidine (2.5 mL, 25.2 mmol) and MeOH (13 mL) were
added successively. After the removal of MeOH, the residue was ex-
tracted with EtOAc. The organic layer was dried over MgSO4 and con-
centrated in vacuo, and the residue was purified by silica gel column
chromatography (EtOAc–hexane, 1:2) to give 9 as a white solid; yield:
98.9 mg (0.392 mmol, 78%); mp 199 °C.
1H NMR (600 MHz, CDCl3): δ (Z-isomer) = 7.78 (dd, J = 8.4, 1.2 Hz, 1
H), 7.54 (m, 2 H), 7.38 (m, 1 H), 7.31 (m, 1 H), 7.24 (m, 2 H), 7.13 (dt,
J = 7.8, 1.2 Hz, 1 H), 7.02 (d, J = 1.8 Hz, 1 H), 6.98 (dt, J = 7.8, 1.2 Hz, 1
H), 6.15 (s, 1 H), 3.88 (s, 3 H); δ (E-isomer) = 7.66 (dd, J = 7.8, 1.2 Hz, 1
H), 7.54 (m, 1 H), 7.38 (m, 3 H), 7.30 (m, 2 H), 7.22 (m, 1 H), 7.18 (dt,
J = 7.8, 1.8 Hz, 1 H), 7.05 (d, J = 1.8 Hz, 1 H), 6.18 (s, 1 H), 3.90 (s, 3 H).
13C NMR (150 MHz, CDCl3): δ (Z-isomer) = 163.8, 155.8, 139.0, 136.9,
135.0, 134.6, 134.1, 133.8, 132.5, 131.8, 130.4, 129.1, 128.0, 127.5,
126.8, 125.7, 102.9, 55.5; δ (E-isomer) = 163.8, 155.9, 138.2, 137.4,
135.1, 134.0, 133.6, 133.2, 133.0, 132.1, 129.9, 129.4, 128.3, 127.9,
127.8, 127.7, 127.4, 127.3, 125.7, 102.1, 55.4.
IR (ATR): 3393, 3180, 1619, 1504, 1468, 1381, 1185 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.59 (br, 1 H), 7.51–7.38 (m, 7 H), 7.04
(d, J = 10.8 Hz, 1 H), 6.92 (d, J = 10.8 Hz, 1 H), 6.50–5.10 (br, 2 H).
13C NMR (150 MHz, CDCl3): δ = 183.6, 142.2, 138.6, 134.3, 133.5,
HRMS (ESI): m/z [M + H]+ calcd for C22H14Cl3NO2: 430.0163; found:
430.0180.
131.8, 130.1, 129.7, 129.4, 129.1, 129.1, 128.8, 128.5, 127.8, 126.7.
HRMS (ESI): m/z [M + H]+ calcd for C15H12N2S: 253.0794; found:
Chiral HPLC separation: shorter tR: [α]D23 +156.4 (c 0.17, CHCl3); 99.9%
ee; longer tR: [α]D23 –153.3 (c 0.10, CHCl3); 97.2% ee.
253.0793.
10-Methoxy-5-(2,4,6-trimethylbenzoyl)-5H-dibenz[b,f]azepine
(7c)
10-Methoxy-5H-dibenz[b,f]azepine-5-carbothioamide (10)
FmocNCS (210 mg, 0.75 mmol) was subjected to reaction with 10-
methoxy-5H-dibenz[b,f]azepine (3; 112 mg, 0.5 mmol), following the
same procedure as that used for the synthesis of 9 described above, to
afford 10 as a white solid; yield: 81.1 mg (0.287 mmol, 57%); E/Z =
1:1; mp 218 °C.
White solid; yield: 351.8 mg (95%); Z/E = 1:1.6; mp 178–179 °C.
IR (ATR): 1654 cm–1
.
1H NMR (600 MHz, CDCl3): δ (Z-isomer) = 7.78 (d, J = 7.2 Hz, 1 H), 7.52
(m, 2 H), 7.36 (m, 1 H), 7.16 (d, J = 7.2 Hz, 1 H), 7.04 (dt, J = 7.2, 1.8 Hz,
1 H), 6.91 (m, 2 H), 6.81 (s, 1 H), 6.50 (s, 1 H), 6.13 (s, 1 H), 3.90 (s, 3
H), 2.66 (s, 3 H), 2.15 (s, 3 H), 1.96 (s, 3 H); δ (E-isomer) = 7.60 (d, J =
7.8 Hz, 1 H), 7.52 (m, 1 H), 7.36 (m, 2 H), 7.28 (t, J = 7.2 Hz, 1 H), 7.13
(dt, J = 7.8, 1.2 Hz, 1 H), 7.07 (dt, J = 7.2, 1.2 Hz, 1 H), 6.99 (d, J = 7.8 Hz,
1 H), 6.79 (s, 1 H), 6.49 (s, 1 H), 6.22 (s, 1 H), 3.90 (s, 3 H), 2.66 (s, 3 H),
2.13 (s, 3 H), 1.94 (s, 3 H).
13C NMR (150 MHz, CDCl3): δ (Z-isomer) = 170.9, 156.4, 140.2, 138.5,
137.8, 134.8, 133.8, 133.6, 132.0, 130.5, 128.8, 128.3, 128.0, 127.9,
127.4, 127.4, 127.2, 126.8, 126.7, 102.5, 55.6, 21.1, 20.1, 19.6; δ (E-
isomer) = 170.8, 155.9, 140.0, 138.7, 137.9, 134.5, 133.5, 133.4, 133.3,
132.4, 129.7, 129.4, 128.2, 128.0, 127.9, 127.5, 127.4, 127.4, 127.1,
126.7, 102.5, 55.5, 21.1, 20.1, 18.4.
IR (ATR): 3429, 3141, 1612, 1433 cm–1
.
1H NMR (600 MHz, CDCl3): δ = 7.83 (d, J = 7.8 Hz, 1 H), 7.76 (d, J = 7.8
Hz, 1 H), 7.58–7.28 (m, 14 H), 6.23 (s, 1 H), 6.09 (s, 1 H), 6.40–4.85 (br,
4 H), 3.89 (s, 3 H), 3.87 (s, 3 H); two diastereomers, not distinguished.
13C NMR (150 MHz, CDCl3): δ = 183.2, 183.0, 157.3, 154.5, 142.3,
140.8, 138.6, 137.3, 134.5, 133.0, 132.30, 131.6, 130.8, 130.1, 129.9,
129.0, 128.9, 128.8, 128.6, 128.4, 127.8, 127.7, 127.6, 127.3, 127.1,
126.2, 126.2, 103.0, 100.5, 55.6, 55.5; two diastereomers, not distin-
guished.
HRMS (ESI): m/z [M + H]+ calcd for C16H14N2OS: 283.0900; found:
283.0898.
Chiral HPLC separation: shorter tR: [α]D20 +228.6 (c 0.21, CHCl3); 97.7%
ee; longer tR: [α]D20 –234.6 (c 0.21, CHCl3); 96.6% ee.
HRMS (ESI): m/z [M + Na]+ calcd for C25H23NO2: 392.1621; found:
392.1616.
10-Methoxy-5H-dibenz[b,f]azepine-5-carboxamide (8)
To a solution of 10-methoxy-5H-dibenz[b,f]azepine (3; 223.3 mg, 1.0
mmol) in CH2Cl2 (5 mL) at r.t. under argon was added FmocNCO
(397.9 mg, 1.5 mmol). The mixture was stirred for 16 h, then piperi-
dine (2.5 mL, 25.2 mmol) and MeOH (13 mL) were added successively.
After the removal of MeOH, the residue was extracted with EtOAc.
The organic layer was dried over MgSO4 and concentrated in vacuo,
and the residue was purified by silica gel column chromatography
(EtOAc–hexane, 1:2) to give 8 as a white solid; yield: 213 mg (0.800
mmol, 80%); mp 193–194 °C.
Acknowledgment
This work was supported in part by a Grant-in-Aid for Scientific Re-
search (C) (24590020) from the Japan Society for the Promotion of
Science. H. Takahashi is grateful for financial support from the
Astellas Foundation for Research on Metabolic Disorders (2013) and
MEXT-Supported 53.6 Program for the Strategic Research Foundation
at Private Universities (2013–2017).
IR (ATR): 3476, 1666, 1600, 1402, 1133 cm–1
1H NMR (600 MHz, CDCl3): δ = 7.73 (d, J = 7.3 Hz, 1 H), 7.45 (m, 3 H),
.
Supporting Information
7.35 (m, 1 H), 7.27 (m, 3 H), 6.14 (s, 1 H), 5.20–4.10 (br, 2 H), 3.88 (s, 3
H).
Supporting information for this article is available online at
http://dx.doi.org/10.1055/s-0035-1560201.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
o
nrtogI
f
rmoaitn
13C NMR (150 MHz, CDCl3): δ = 156.8, 156.0, 140.1, 138.5, 135.0,
134.6, 133.0, 130.4, 129.2, 128.2, 128.1, 127.7, 127.6, 127.5, 127.3,
55.4.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913

3913
Y. Kanase et al.
Paper
Synthesis
References
Oshitari, T.; Natsugari, H.; Takahashi, H. Org. Lett. 2014, 16,
1514. (c) Wakamatsu, S.; Takahashi, Y.; Tabata, H.; Oshitari, T.;
Tani, N.; Azumaya, I.; Katsumoto, Y.; Tanaka, T.; Hosoi, S.;
Natsugari, H.; Takahashi, H. Chem. Eur. J. 2013, 19, 7056.
(d) Takahashi, H.; Wakamatsu, S.; Tabata, H.; Oshitari, T.;
Harada, A.; Inoue, K.; Natsugari, H. Org. Lett. 2011, 13, 760.
(e) Tabata, H.; Nakagomi, J.; Morizono, D.; Oshitari, T.;
Takahashi, H.; Natsugari, H. Angew. Chem. Int. Ed. 2011, 50,
3075. (f) Takahashi, Y.; Wakamatsu, S.; Tabata, H.; Oshitari, T.;
Natsugari, H.; Takahashi, H. Synthesis 2015, 47, 2125.
(1) Loiseau, P.; Duché, P. In Antiepileptic Drugs; Levy, R. H.;
Mattson, R. H.; Meldrum, M. S., Eds.; Raven Press: New York,
1995, 555–566.
(2) (a) Carril, M.; SanMartin, R.; Domínguez, E.; Tellitu, I. Tetrahe-
dron 2007, 63, 690. (b) Benes, J.; Parada, A.; Figueiredo, A. A.;
Alves, P. C.; Freitas, A. P.; Learmonth, D. A.; Cunha, R. A.; Garrett,
J.; Soares-da-Silva, P. J. Med. Chem. 1999, 42, 2582. (c) Tian, M.;
Abdelrhaman, A.; Weinhausen, S.; Hinz, S.; Weyer, S.; Dosa, S.;
Tayeb, A. E.; Müller, C. E. Bioorg. Med. Chem. 2014, 22, 1077.
(d) Ravinder, B.; Reddy, S. R.; Sridhar, M.; Mohan, M. M.;
Srinivas, K.; Reddy, A. P.; Bandichhor, R. Tetrahedron Lett. 2013,
54, 2841. (e) Martínez, R.; Espinosa, R. C.; Toscano, R. A.;
Cogordán, J. A.; Arellano, M. D. R.; Angeles, E.; Posada, M. E.;
Maya, B.; Martínez, L. J. Heterocycl. Chem. 1996, 33, 715.
(3) (a) Lowes, M. M. J.; Caria, M. R.; Lötter, A. P.; Van Der Watt, J. G.
J. Pharm. Sci. 1987, 76, 744. (b) Grzesiak, A. L.; Lang, M.; Kim, K.;
Matzger, A. J. J. Pharm. Sci. 2003, 92, 2260. (c) Lang, M.; Kampf, J.
W.; Matzger, A. J. J. Pharm. Sci. 2002, 91, 1186. (d) Harris, R. K.;
Ghi, P. Y.; Puschmann, H.; Apperley, D. C.; Griesser, U. J.;
Hammond, R. B.; Ma, C.; Roberts, K. J.; Pearce, G. J.; Yates, J. R.;
Pickard, C. J. Org. Process Res. Dev. 2005, 9, 902.
(6) The existence of stereoisomers of pirenzepine, telenzepine, and
nevirapine has been discussed; see refs. 4a, 4b, and 4c, respec-
tively.
(7) Conformation 1-A is described as (ax1: aS, ax2: aR, ax3: aS, ax4:
aR). Correspondingly, 1-B is described as (ax1: aR, ax2: aS, ax3:
aR, ax4: aS). The terms aS and aR (chiral axis nomenclature) cor-
respond to P and M (helix nomenclature). See the Supporting
Information (CH-COSY and HH-COSY NMR spectra) for the
determination of E/Z conformations.
(8) CCDC 1406904 contains the supplementary crystallographic
data for this paper. The data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/getstructures.
(4) (a) Trummlitz, G.; Schmidt, G.; Wagner, H.-U.; Luger, P.
Arzneim.-Forsch./Drug. Res. 1984, 34, 849. (b) Eveleigh, P.;
Hulme, E. C.; Schudt, C.; Birdsall, N. J. M. Mol. Pharmacol. 1989,
35, 477. (c) Burke, E. W. D.; Morris, G. A.; Vincent, M. A.; Hillier,
I. H.; Clayden, J. Org. Biomol. Chem. 2012, 10, 716. (d) Mui, P. W.;
Jacober, S. P.; Hargrave, K. D.; Adams, J. J. Med. Chem. 1992, 35,
201.
(9) DG‡ value of 75.3 kJ/mol for compound 5, and DG‡ value of 66.9
kJ/mol for compound 10 (see the Supporting Information); for
the determination of ΔG‡ values by variable temperature NMR
experiments, see: Boiadjiev, S. E.; Lightner, D. A. Tetrahedron
2002, 58, 7411.
(10) For the determination of ΔG‡ values, see: Petit, M.; Lapierre, J.
B.; Curran, D. P. J. Am. Chem. Soc. 2005, 127, 14994.
(5) (a) Yoneda, T.; Tabata, H.; Tasaka, T.; Oshitari, T.; Takahashi, H.;
Natsugari, H. J. Med. Chem. 2015, 58, 3268. (b) Tabata, H.;
Kayama, S.; Takahashi, Y.; Tani, N.; Wakamatsu, S.; Tasaka, T.;
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 3907–3913