Structural manipulation on the catecholic fragment of dopamine D 1 receptor agonist 1-phenyl-N-methyl-benzazepines

Article abstract of DOI:10.1016/j.ejmech.2014.07.059

A series of new benzazepines with modification on the catecholic fragment were designed. The 8-hydroxyl group, other than the 7-hydroxyl was confirmed crucial to the interaction with the dopamine D₁ receptor. Subsequent replacement of the 7-hyd

Full text of DOI:10.1016/j.ejmech.2014.07.059

European Journal of Medicinal Chemistry 85 (2014) 16e26
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Structural manipulation on the catecholic fragment of dopamine D1
receptor agonist 1-phenyl-N-methyl-benzazepines
,
,
Jing Zhang a 1, Jiye Huang b 1, Zilan Song a, Lin Guo c, Wenxian Cai b, Yun Wang c,
Xuechu Zhen b **, Ao Zhang a
,
c
,
, *
a CAS Key Laboratory of Receptor Research, and Synthetic Organic & Medicinal Chemistry Laboratory (SOMCL),
Shanghai Institute of Materia Medica (SIMM), Chinese Academy of Sciences (CAS), Shanghai 201203, China
b Department of Pharmacology, Shanghai Institute of Materia Medica, CAS, China
c Department of Pharmacology, College of Pharmaceutical Sciences, Soochow University, Suzhou 215123, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of new benzazepines with modification on the catecholic fragment were designed. The 8-
hydroxyl group, other than the 7-hydroxyl was confirmed crucial to the interaction with the dopa-
mine D1 receptor. Subsequent replacement of the 7-hydroxyl with benzylamino groups was found
tolerable. 7-(m-Chlorophenyl)methylamino- and 7-(m- or o-tolyl)methylamino-substituted benzaze-
pines 13bed displayed Ki values of 270e370 nM at the D1 receptor, which were slightly more potent than
that of parent compound 1. In addition, 7-(arylmethyl)amino-benzazepines 13aec were found pos-
sessing high binding affinities less than 10 nM at the 5-HT2A receptor. Among them, the non-substituted
7-benzylamino analogue 13a was the most potent showing a Ki values of 4.5 nM at the 5-HT2A receptor
and a 5-HT2A/D1 selectivity of 147.
Received 7 February 2014
Received in revised form
11 July 2014
Accepted 18 July 2014
Available online 24 July 2014
Keywords:
Arylbenzazepine
Dopamine receptor
Serotonin 5-HT2A receptor
Structural modification
Catecholic fragment
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
series [6e8]. Unfortunately, most of these compounds eventually
failed as drug candidates due to their limited in vivo efficacy, poor
Dopamine (DA) is one of the major cerebral neurotransmitters
and plays an essential role in the pathophysiology of many neu-
robehavioral and neuropsychiatric disorders. DA exerts its agonistic
actions primarily through its five major DA receptors (D1eD5),
among which, D1,eD3 receptors are the most studied DA receptors
[1e3] and are the primary targets of current clinically prescribed
dopaminergic drugs [4,5]. Although the D1 receptor was discovered
very early with high abundance in the mammalian brains, clinically
useful D1 receptor agonists and antagonists are very limited [4e6].
Among the reported D1 receptor-targeting agents, the skeleton of
1-aryl-N3-benzazepines remains the most reliable structural scaf-
fold in terms of the affinity and selectivity against the D1 receptor.
Many widely used D1 receptor tool drugs (e.g. D1 agonist SKF-
38393, D1 antagonist SCH-23390, Fig. 1) were born from this
pharmacokinnetics (PK) and several other unwanted side effects
[6,9,10].
Since the catechol fragment in the 1-aryl-N3-benzazepine
framework is an essential feature for effective binding to the
amino acid residues of the D1 receptor, most of the reported
structural modification is focused on other sites, especially the 1-
aryl, azepine ring, and C6 [11e13]. A few early reports also dis-
cussed the possibility of replacing the catecholic component, but
only lower alkyls and halogens (especially 7-Cl) were investigated
as the replacement of 7-OH [14e19]. As a continuation of our
structureeactivity relationship (SAR) study [12,13,20e23] on the 1-
aryl-N3-benzazepine skeleton, here we report our structural
manipulation on the catecholic fragment and the binding affinity
and selectivity of the new compounds at the DA (D1eD3) receptors.
2. Chemistry
* Corresponding author.
** Corresponding author.
Although the R-enantiomers of benzazepines 1e4 are generally
more active than corresponding S-enantiomers, there is no signif-
icant difference between the racemates and their R-enantiomers at
the DA receptor binding level [6,12,13]. Therefore, to quickly
E-mail addresses: zhenxuechu@suda.edu.cn (X. Zhen), aozhang@mail.shcnc.ac.
cn (A. Zhang).
1
These two authors contributed equally to this work.
http://dx.doi.org/10.1016/j.ejmech.2014.07.059
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
17
Fig. 1. Established SAR and representative 1-aryl-3-benzazepines.
identify new compounds for further study, all compounds in cur-
rent report were prepared and evaluated as racemates. As shown in
Scheme 1, by following a literature procedure [12,13,22,23], 7- and
8-methoxy-N-methyl-1-benzazepines 7 and 8 were prepared from
corresponding 3- or 4-methoxyphenyl ethanamine and 2-
phenyloxirane in three steps [18,19]. Removal [23] of the O-
methyl group by refluxing in 48% HBr aqueous solution led to
monohydroxyl benzazepines 9 and 10 in 90% yield. Nitration of 8-
hydroxy-N-methyl-1-phenylbenzazepine 10 with fuming HNO3
and HOAc provided compounds 11 and 110 in 96% overall yield, with
nearly no regioselectivity (1.1/1). Reduction of nitrobenzene 11
with Pd/C gave 7-amino-8-hydroxybenzazepine 12 in 96% yield.
Reductive amination [13] of 12 by treating with aryl aldehydes
followed by NaBH4 yielded corresponding benzylamines 13aee in
80e91% yields. Acylation of amine 12 led to benzazepine 14 in 41%
yield, and diethylamino-substituted benzazepine 15 was prepared
in 81% yield by treating 12 with acetaldehyde and NaBH(OAc)3.
Meanwhile, treatment [24] of 12 with 2,5-dimethoxy-
Scheme 1. Synthesis of compounds 7e16.

18
J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
Scheme 2. Synthesis of compounds 17e19.
tetrahydrofuran in HOAc/H2O at 110 ꢀC afforded 7-(1H-pyrrol-1-
yl)-8-hydroxy-benzazepine 16 in 81% yield.
Similarly, nitration [18,19] of 7-hydroxybenzazepine 9 with
fuming HNO3 and HOAc provided 7-hydroxy-8-nitrobenzazepine
26 and 6-nitro-7-hydroxybenzazepinf 27 in 43% and 24% yields,
respectively (Scheme 4). Reduction of 26 with Pd/C yielded 7-
hydroxy-8-aminobenzazepine 28 in 90% yield. Treating 28 with
CDI in refluxing THF provided benzoxazolone 29 in 68% yield.
Acylation of 28 with 2-chloroacetyl chloride afforded benzo[b][1,4]
oxazin-3-one 30 in 75% yield. Benzo[d]oxazole 31 was prepared in
93% yield by refluxing amine 28 with orthoethyl formate in EtOH.
Meanwhile, reaction [25] of amine 28 with CS2 and KOH in EtOH
offered benzo[d]oxazole-2-thiol 32 in 89% yield, which was then
treated with N-methylpiperazine under microwave irradiation at
100 ꢀC to provide benzazepine 33 in 72% yield [26].
Aryl methanol 18 was prepared [23] by treating phenol 9 with
formaldehyde and anhydrous MgCl2 in refluxing THF, followed by
reduction with NaBH4 in 45% overall yield (Scheme 2). Meanwhile,
the benzaldehyde intermediate 17 was reacted with hydroxylamine
hydrochloride followed by reduction with LiAlH4 to afford phe-
nylmethyl amine 19 in 58% overall yield [24].
Synthesis of 7-aryl substituted benzazepine analogues 21aee
was described in Scheme 3. Bromination of 8-methoxy-3-H-ben-
zazepine 80 followed by N-methylation provided bromide 20 in 40%
yield, which was then subjected to palladium-catalyzed Suzuki
coupling with arylboronic acids followed by O-demethylation
[22] with BBr3 at ꢁ78 ꢀC to deliver target compounds 21aee in
27e73% overall yields. Meanwhile, triflation of 7- and 8-
hydroxylbenzazepines 9 and 10 with Tf2O and Et3N, followed by
Pd(OAc)2-catalyzed CeN bond formation [23] and subsequent hy-
drolysis provided 7- and 8-aminobenzazepines 22 and 23 in 82%
overall yields. N-Methylation was completed by treating amine 23
with ethyl formate followed by LiAlH4 giving benzazepine 24 in 48%
overall yield. Additionally, reacting amine 23 with formalin and
formic acid at 90 ꢀC led to dimethylamino analogue 25 (Scheme 3).
3. Results and discussion
3.1. Binding assay at the human dopamine D1eD3 receptors
All new compounds, together with the prototypic catecholic
benzazepine 1 were bioassayed for their binding at the dopamine
D1 receptor using membrane preparation obtained from stable
transfected HEK293 or CHO cells. Meanwhile, to evaluate the
Scheme 3. Synthesis of compounds 20e25.

J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
19
Scheme 4. Synthesis of compounds 26e33.
selectivity of these compounds among the DA receptor subtypes,
binding affinities at the D2 and D3 receptors were also tested. The
procedure is similar to those reported by us previously
[12,13,22,23]. As shown in Table 1, significant discrepancy in D1
receptor binding was observed between the two mono-hydroxyl
benzazepines 9 and 10. Compared to catechol 1, 8-hydroxyl
analogue 10 is slightly more potent at the D1 receptor showing a
Ki value of 207 nM, whereas the 7-hydroxyl congener 9 is inactive
D3 receptors, confirming the preference of 1-phenyl-N-methyl-
benzezapine skeleton as a privileged scaffold for the D1 receptor.
From the results above, the 7-hydroxyl group in the prototypical
benzazepine 1 was confirmed replaceable by an appropriately
substituted benzylamino-(e.g. 13b) or substituted aryl moiety (e.g.
21b, 21e). To further explore the importance of the nearly ‘ignored’
8-hydroxyl in 1, 8-hydroxymethyl and 8-aminomethyl analogues 18
and 19 were designed first to retain the H-bonding donor property
and to improve the metabolic stability. Unfortunately, both com-
pounds did not show appreciable binding affinity at the D1 receptor
(Table 2). As to the 7- or 8-amino substituted benzazepines (22 and
23), only the 8-amino substituted analogue 23 retained good af-
finity at this receptor with a Ki value of 583 nM, which is only 1.5-
fold less potent than that of 1. Quite disappointingly, further sub-
stitutions on the 8-amino moiety, including alkylamino-(24, 25),
cyclic amido-(29, 30), and aromatic N-heterocycle (31, 33) led to
complete loss of binding affinity at the D1 receptor.
(Ki > 1
mM). The result further confirms the necessity of the 8-
hydroxyl group to the interactions with the D1 receptor, as that in
benzazepines 1e6. Interestingly, replacing either the 7- or 8-
hydroxyl of benzazepine 1 with an amino group did not impact
the D1 receptor binding at all, and both compounds 12 and 28 dis-
played Ki values (330e400 nM) nearly identical to that of compound
1. It was found that variously substituted benzylamino groups were
relatively tolerant as the 7-substituent, and compounds 13aee
displayed Ki values in the range of 277e756 nM at the D1 receptor.
Electron-withdrawing group substituted 7-benzylamino-benzaze-
pine 13b showed a Ki value of 277 nM at the D1 receptor, which is
higher than that of the prototypic 1. Compound 13b is also slightly
more potent than both non-substituted (13a) and electron-
donating group substituted benzylamino analogues (13cee).
Among the three methyl substituted benzylamines 13cee, the
ortho-isomer is most potent having a Ki value of 292 nM, whereas
the para-methyl analogue is the least potent. Surprisingly, 7-
acylamino-8-hydroxybenzazepine 14 lost binding affinity at the
D1 receptor, indicating that a weak H-bond donor or an H-bond
acceptor at the 7-position is not beneficial to the interaction with
the D1 receptor. This was further confirmed by the inactivity of 7-
dimethylamino- and 7-pyrrolinyl-substituted benzazepines 15
and 16. Interestingly, among the 7-aryl substituted benzazepines
21aee, m-tolyl analog 21b and p-hydroxyphenyl analog 21e
showed good binding affinities at the D1 receptor with Ki values of
216 and 349 nM, respectively, which were even more potent than
that of compound 1. All the compounds were inactive at the D2 and
3.2. Binding assay of the D1 active compounds at the rat 5-HT1A and
human 5-HT2A receptors
In our previous studies [12,13], we found that benzazepines with
substitutions at the N3- or C6 showed substantial binding affinities
at the serotonin (5-HT) receptors, especially 5-HT1A and 5-HT2A
subtypes. To investigate the selectivity of our newly synthesized
benzazepine analogues over the 5-HT receptors, binding assays of
the D1-active compounds were further evaluated at the 5-HT1A and
5-HT2A receptors by using [3H]8-OH-DPAT and [3H]ketanserin as
the respective standard radioligands [12,13,22,23]. As summarized
in Table 3, most of these compounds displayed moderate to high
affinities at the 5-HT2A receptor, yet low or no affinities at the 5-
HT1A receptor. Compared to the D1 receptor affinity, much higher
affinity was observed for 8-hydroxybenzazepine 10 at the 5-HT2A
receptor showing a Ki value of 133 nM. Interestingly, introducing a
7-amino group to compound 10 abolished the 5-HT2A binding,

20
J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
Table 1 (continued)
Table 1
Binding affinities of phenolic benzazepines at human DA D1eD3 receptors.a
Cpd
Structure
Ki (nM)
Cpd
Structure
Ki (nM)
hD1
hD2
hD3
hD1
hD2
NA
hD3
16
>1000
>1000
e
1
393
5
e
9
>1000
>1000
>1000
>1000
>1000
21a
21b
21c
21d
21e
40-Me-Ph
30-Me-Ph
20-Me-Ph
20-Napthyl
40-OH-Ph
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
e
e
e
216 28
912 97
>1000
10
207 25
349 41
a
Values are means of three to five experiments, and all compounds were tested in
racemic form; dash line indicates that the data is not available.
12
28
334 37
>1000
>1000
>1000
>1000
whereas the D1 receptor affinity was retained. Similar result was
observed in 7-hydroxy-8-aminobenzazepine 28, indicating that a
potential interaction site of the 5-HT2A receptor might exist at the
C7-position of the benzazepine analogues. Significant discrepancy
of the 5-HT2A binding affinities was observed among the 7-(aryl-
methyl)amino substituted benzazepines 13aee. Compounds 13aec
displayed high affinities at this receptor and showed Ki values of
less than 10 nM. Non-substituted 7-benzylamino benzazepine 13a
is the most potent with a Ki value of 4.5 nM at the 5-HT2A receptor
and a high 5-HT2A/D1 selectivity of 147. The lower affinities of
compounds 13d and 13e suggested that the substitution pattern
other than electronic property of the arylmethylamino moiety play
an important role to the interactions with the receptor. Moderate
affinities at the 5-HT2A receptor were observed for the 7-aryl
benzazepines 21b and 21e showing Ki values of 163 and 826 nM,
respectively. Compound 21b with a lipophilic substituent (methyl)
is 5-fold more potent than the phenol analogue 21e.
351 67
13a
13b
665 66
>1000
>1000
>1000
277 32
e
4. Conclusion
13c
13d
370 40
>1000
>1000
e
e
As a part of our long-term objective of developing novel benza-
zepine analogues useful for the treatment of CNS diseases, we pre-
pared a series of new benzazepines with modifications on the
catecholic fragment that potentially lead to better metabolic sta-
bility. First, the importance of the 8-hydroxyl group to the interac-
tion with the D1 receptor was confirmed, and then several series of
new analogues bearing a larger substituent at the C7 position were
subsequently investigated. In comparison with the binding affinity
of the prototypical 1, non-/substituted benzylamino groups were
well tolerated. 7-(m-Chlorophenyl)methylamino- and 7-(m- or o-
tolyl)methylamino-substituted benzazepines (13b,c,d) displayed Ki
values of 270e370 nM at the D1 receptor, which were slightly more
potent than that of 1. Slightly higher affinities were also observed for
7-(m-methylphenyl)- and 7-(p-hydroxyphenyl)-substituted ben-
zazepines 21b and 21e. Meanwhile, these D1-active compounds
were further evaluated at the 5-HT1A and 5-HT2A receptors. Most
compounds displayed good binding affinities at the 5-HT2A receptor,
whereas negligible affinity at the 5-HT1A receptor. 7-(Arylmethyl)
amino-benzazepines 13aec displayed Ki values of less than 10 nM,
with the non-substituted 7-benzylamino analogue 13a as the most
potent showing a Ki values of 4.5 nM at the 5-HT2A receptor and a
high 5-HT2A/D1 selectivity of 147. Moderate affinities at the 5-HT2A
receptor were observed for 7-aryl-benzqzepines 21b and 21e. They
292 16
13e
756 162
>1000
e
14
15
>1000
>1000
>1000
>1000
e
>1000

J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
21
Table 2
Table 3
Binding affinities of nonphenolic benzazepines at human DA D1eD3 receptors.a
In vitro binding assays of new benzazepines at rat 5-HT1A and human 5-HT2A re-
ceptors. a
Cpd
Structure
Ki (nM)
Compound
Ki (nM)
hD1
hD2
hD3
r5-HT1A
h5-HT2A
hD1
393
1
10
12
28
13a
13b
13c
13d
13e
21b
21e
e
e
5
18
>1000
>1000
>1000
>1000
>1000
>1000
>1000
479 85
>1000
>1000
>1000
>1000
>1000
133 17
>1000
>1000
4.52 0.07
10.3 1.8
9.95 1.10
192 37
862 17
163 9.4
826 84
207 25
334 37
351 67
665 66
277 32
370 40
292 16
756 162
216 28
349 41
19
22
>1000
>1000
>1000
>1000
>1000
>1000
a
Values are means of three to five experiments, and all compounds were tested in
racemic form; dash line indicates that the data is not available.
within 0.4% of theoretical values. Analytical thin-layer chroma-
tography (TLC) was carried out on 0.2-mm Kieselgel 60F 254 silica
gel plastic sheets (EM Science, Newark). Flash column chromatog-
raphy was conducted on silica gel. The column output was moni-
tored with TLC. Yields of all the reactions were not optimized.
Compounds 9 and 10 were prepared by following a literature
procedure [17,18].
23
24
25
29
583 25
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
e
e
e
5.2. 7-Nitro and 9-nitro-substituted 8-hydroxy-1-phenyl-N-methyl
benzazepines 11 and 110
To a solution of phenol 10 (990 mg, 3.91 mmol) in acetic acid
(10 mL) was added foaming nitric acid (320 mg, 5.08 mmol) slowly.
The reaction was allowed to stir at room temperature for 2 h before
ice-water was added. The resulting mixture was basified with
NH3$H2O, and extracted with CHCl3. The organic phase was washed
with water and brine, dried over anhydrous Na2SO4 and evapo-
rated. The residue was purified by flash column chromatograph
(CHCl3:MeOH:NH3$H2O ¼ 100:1:1) to give 11 (577 mg, 50%) and 110
(531 mg, 46%). For 11: 1H NMR (300 MHz, CDCl3)
d 7.84 (s, 1H), 7.33
30
31
>1000
>1000
>1000
>1000
>1000
>1000
e
e
e
(m, 3H), 7.16 (m, 2H), 6.43 (s, 1H), 4.35 (d, J ¼ 8.4 Hz, 1H), 3.15 (m,
2H), 2.88 (m, 3H), 2.35 (m, 4H); for 110: 1H NMR (300 MHz, CDCl3)
d
7.28 (m, 2H), 7.18 (m, 2H), 7.07 (m, 2H), 6.87 (d, J ¼ 8.4 Hz, 1H),
4.68 (m, 1H), 3.56 (dd, J ¼ 6.0, 13.2 Hz, 1H), 2.80 (m, 3H), 2.52 (m,
1H), 2.34 (m, 4H).
5.3. 7-Amino-8-hydroxy-1-phenyl-N-methyl benzazepine 12
A solution of compound 11 (1.5 g, 5.03 mmol) and 10% Pd/C
(450 mg) in MeOH/THF (1:1, 40 mL) was stirred overnight under H2
at room temperature. After filtration, the filtrate was concentrated
to give a yellow solid 12 (1.3 g, 96%). 1H NMR (300 MHz, CDCl3)
33
a
d
7.73 (m, 3H), 7.38 (m, 2H), 6.97 (s, 1H), 6.23 (s, 1H), 4.55 (d,
J ¼ 8.7 Hz, 1H), 3.60 (m, 1H), 3.35 (m, 2H), 3.03 (m, 2H), 2.66 (m,
4H); 13C NMR (100 MHz, CDCl3)
143.6, 142.6, 135.0, 132.7, 132.1,
Values are means of three to five experiments, and all compounds were tested in
racemic form; dash line indicates that the data is not available.
d
were still slightly more potency at this receptor than at the D1 re-
ceptor. All the new synthetic benzazepines were either inactive or
significantly less active at the D2, D3 and 5-HT1A receptors.
128.6, 128.4, 126.2, 117.5, 115.4, 63.1, 57.0, 47.4, 46.5, 34.5; EI-MS m/
z: 268 (Mþ); HR-MS calcd for C17H20N2O (Mþ) 268.1576. Found:
268.1576.
5. Experimental
5.4. 7-(Arylmethyl)amino-8-hydroxy-1-phenyl-N-methyl
benzazepines 13aee
5.1. General methods
A solution of amine 12 (1.0 mmol) and an appropriate aryl
aldehyde (1.3 mmol) in dry ethanol (6 mL) was stirred at room
temperature for 3 h. The mixture was then cooled to 0 ꢀC, and
NaBH4 (1.5 mmol) was added slowly. After stirring at room
1H and 13C NMR spectra were recorded on a Brucker AC300
spectrometer using tetramethylsilane as an internal reference.
Element analyses, performed by the Analytic Lab, SIMM, were

22
J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
temperature for 1 h, the reaction was quenched with water, and
extracted with CHCl3. The combined organic phase was washed
with brine, dried over anhydrous Na2SO4 and then evaporated. The
yellowish residue was subjected to column chromatograph
(CHCl3:MeOH ¼ 30:1) to give corresponding benzylamines 13aee.
residue was subjected to column chromatograph (EtOAc:MeOH ¼
15:1) to give a yellow liquid 14 (43 mg, 41%). 1H NMR (300 MHz,
CDCl3)
1H), 4.08 (d, J ¼ 9.0 Hz,1H), 2.84 (m, 3H), 2.61 (m, 2H), 2.20 (m, 7H);
13C NMR (100 MHz, CDCl3)
169.6, 145.3, 142.9, 141.6, 132.0, 128.6,
d 7.95 (s, 1H), 7.40 (s, 1H), 7.22 (m, 3H), 6.91 (m, 2H), 6.00 (s,
d
128.4, 126.6, 124.0, 121.9, 117.2, 62.9, 57.0, 48.2, 46.8, 34.7, 24.1; EI-
MS m/z: 310 (Mþ); HR-MS calcd for C19H22N2O2 (Mþ) 310.1681.
Found: 310.1684.
5.4.1. 7-(Benzylamino)-3-methyl-8-hydroxy-1-phenyl-benzazepine
13a
White solid (89%). 1H NMR (300 MHz, CDCl3)
d 7.27 (m, 8H), 6.84
(m, 2H), 6.40 (s, 1H), 5.74 (s, 1H), 4.34 (s, 2H), 4.05 (d, J ¼ 9.3 Hz,1H),
5.6. 7-(Diethylamino)-8-hydroxy-3-methyl-1-phenyl-benzazepine
15
3.13 (m, 1H), 2.85 (m, 2H), 2.53 (m, 2H), 2.10 (m, 4H); 13C NMR
(100 MHz, CDCl3)
d 143.8, 142.0, 139.9, 135.5, 132.8, 132.0, 128.6,
128.4, 128.3, 127.4, 126.9, 126.1, 114.3, 112.6, 63.1, 56.8, 48.4, 47.1,
46.3, 34.8; EI-MS m/z: 358 (Mþ); HR-MS calcd for C24H26N2O (Mþ)
358.2045. Found: 358.2046.
A solution of compound 12 (50 mg, 0.19 mmol), acetaldehyde
(25 mg, 0.57 mmol) and NaBH(OAc)3 (79 mg, 0.38 mmol) in 1,2-
dichloroethane (7 mL) was stirred at room temperature over-
night. After filtration, the filtrate was concentrated and chroma-
tographied (CHCl3:MeOH ¼ 30:1) on neutral aluminum oxide to
yield diethylamine 15 as white solid (49 mg, 81%). 1H NMR
5.4.2. 7-(30-Chlorobenzylamino)-3-methyl-8-hydroxy-1-phenyl-
benzazepine 13b
Colorless liquid (82%). 1H NMR (300 MHz, CDCl3)
d
7.35 (s, 1H),
(300 MHz, CDCl3) d 7.27 (m, 5H), 6.86 (s, 1H), 6.24 (s, 1H), 4.28 (d,
7.20 (m, 6H), 6.87 (d, J ¼ 7.2 Hz, 2H), 6.27 (s, 1H), 5.83 (s, 1H), 4.28 (s,
J ¼ 8.4 Hz, 1H), 3.10 (m, 2H), 2.81 (m, 7H), 2.33 (m, 4H), 0.95 (t,
2H), 4.17 (d, J ¼ 9.0 Hz, 1H), 3.09 (m, 3H), 2.61 (m, 2H), 2.29 (m, 4H);
J ¼ 7.5 Hz, 6H); 13C NMR (100 MHz, CDCl3)
d 151.8, 143.3, 142.8,
13C NMR (100 MHz, CDCl3)
d
142.7, 142.4, 142.1, 135.4, 134.3, 132.1,
133.5, 132.5, 128.5, 128.4, 126.3, 123.8, 113.5, 57.7, 49.9, 49.6, 47.6,
35.9, 13.0; EI-MS m/z: 324 (Mþ); HR-MS calcd for C21H28N2O (Mþ)
324.2202. Found: 324.2203.
131.1, 129.8, 128.5, 127.3, 127.1, 126.5, 125.3, 114.4, 112.4, 62.3, 56.4,
47.7, 46.1, 45.8, 33.6; EI-MS m/z: 392 (Mþ); HR-MS calcd for
C
24H25ClN2O (Mþ) 392.1655. Found: 392.1647.
5.7. 7-(1H-pyrrol-1-yl)-8-hydroxy-3-methyl-1-phenyl-benzazepine
5.4.3. 7-((3-Methylbenzyl)amino)-3-methyl-8-hydroxy-1-phenyl-
16
benzazepine 13c
white solid (91%). 1H NMR (300 MHz, CDCl3)
d
7.15 (m, 7H), 6.91
(m, 2H), 6.35 (s, 1H), 5.91 (s, 1H), 4.38 (m, 1H), 4.25 (s, 2H), 3.26 (m,
3H), 2.85 (m, 1H), 2.49 (m, 8H); 13C NMR (100 MHz, CDCl3)
143.4,
A mixture of amine 12 (105 mg, 0.39 mmol) and 2,5-dimethoxy-
tetrahydrofuran (52 mg, 0.39 mmol) in HOAc/H2O (6:1, 7 mL) was
stirred at 110 ꢀC for 1.5 h. The reaction mixture was then cooled and
evaporated in vacuum. The obtained residue was dissolved in
CHCl3, washed with NH3$H2O and brine, then evaporated. The
crude product was further purified by column chromatograph
(CHCl3:MeOH ¼ 30:1) to afford compound 16 (101 mg, 81%) as a
d
142.0, 139.7, 138.0, 135.7, 132.5, 131.9, 128.6, 128.4, 128.3, 128.2,
127.7,126.3,124.5,114.4,112.6, 62.9, 56.8, 48.5, 46.9, 46.3, 34.5, 21.4;
EI-MS m/z: 372 (Mþ); HR-MS calcd for C25H28N2O (Mþ) 372.2202.
Found: 372.2211.
yellow liquid. 1H NMR (300 MHz, CDCl3)
d 7.26 (m, 3H), 7.00 (m,
5.4.4. 7-((2-Methylbenzyl)amino)-3-methyl-8-hydroxy-1-phenyl-
3H), 6.90 (s, 2H), 6.28 (s, 2H), 6.10 (s, 1H), 5.53 (brs, 1H), 4.22 (d,
benzazepine 13d
J ¼ 9.0 Hz, 1H), 3.07 (m, 2H), 2.75 (m, 3H), 2.21 (m, 4H); 13C NMR
white solid (80%). 1H NMR (300 MHz, CDCl3 þ CD3OD)
d
7.15 (m,
(100 MHz, CDCl3) d 148.2, 144.3, 142.7, 132.5, 128.7, 128.5, 126.7,
9H), 6.30 (s, 1H), 5.89 (s, 1H), 4.38 (d, J ¼ 9.3 Hz, 1H), 4.15 (s, 2H),
126.5, 126.1, 121.6, 117.0, 109.3, 62.8, 56.9, 48.2, 46.7, 34.5; EI-MS m/
z: 318 (Mþ); HR-MS calcd for C21H22N2O (Mþ) 318.1732. Found:
318.1729.
3.25 (m, 3H), 3.00 (m, 1H), 2.56 (m, 5H), 2.26 (s, 3H); 13C NMR
(100 MHz, CDCl3)
d 143.8, 142.0, 137.5, 136.2, 135.7, 132.7, 132.0,
130.1, 128.6, 128.4, 128.1, 127.0, 126.2, 126.0, 114.3, 112.4, 63.1, 56.9,
47.1, 46.5, 46.4, 34.8, 18.9; EI-MS m/z: 372 (Mþ); HR-MS calcd for
C25H28N2O (Mþ) 372.2202. Found: 372.2202.
5.8. 7-Hydroxy-3-methyl-1-phenyl-benzazepine-8-carbaldehyde
17
5.4.5. 7-((4-Methylbenzyl)amino)-3-methyl-8-hydroxy-1-phenyl-
A mixture of phenol 9 (550 mg, 2.17 mmol), paraformaldehyde
(649 mg, 21.62 mmol), anhydrous MgCl2 (616 mg, 6.47 mmol) and
Et3N (660 mg, 6.52 mmol) in dry THF was refluxed under N2
overnight. The reaction was quenched by ice-water, and extracted
with CHCl3. The combined organic phase was washed with brine,
dried over anhydrous Na2SO4, and then evaporated. The residue
was further purified by chromatography (CHCl3:MeOH ¼ 50:1) to
give aldehyde 17 (277 mg, 45%) as a light yellow liquid. 1H NMR
benzazepine 13e
white solid (85%). 1H NMR (300 MHz, CDCl3)
d 8.19 (brs, 1H), 7.17
(m, 7H), 6.88 (m, 2H), 6.35 (s, 1H), 5.89 (s, 1H), 4.24 (m, 3H), 3.15 (m,
3H), 2.67 (m, 2H), 2.38 (m, 7H); 13C NMR (100 MHz,
CDCl3 þ CD3OD)
d 143.3, 141.9, 136.3, 135.0, 132.6, 131.9, 128.8,
128.2, 127.2, 126.0, 114.0, 112.8, 63.1, 57.0, 48.0, 47.8, 46.6, 34.8, 20.6;
EI-MS m/z: 372 (Mþ); HR-MS calcd for C25H28N2O (Mþ) 372.2202.
Found: 372.2200.
(300 MHz, CDCl3)
d 10.94 (brs, 1H), 9.56 (s, 1H), 7.28 (m, 5H), 6.76
(m, 2H), 4.32 (d, J ¼ 8.7 Hz, 1H), 2.98 (m, 5H), 2.38 (m, 4H); 13C NMR
5.5. 7-(Acetylamino)-8-hydroxy-3-methyl-1-phenyl-benzazepine
(100 MHz, CDCl3) d 196.0, 160.1, 151.8, 142.5, 136.4, 132.5, 128.7,
14
128.2, 126.6, 118.4, 118.2, 63.1, 56.4, 48.6, 47.4, 36.8; EI-MS m/z: 281
(Mþ); HR-MS calcd for C18H19NO2 (Mþ) 281.1416. Found: 281.1421.
Acetyl chloride (39 mg, 0.50 mmol) was added to a solution of
compound 12 (90 mg, 0.34 mmol) and NaOAc (55 mg, 0.67 mmol)
in dry THF in ice-bath. The mixture was stirred at room tempera-
ture for 1 h, and then concentrated. The residue was taken in CHCl3,
washed with saturated NH4Cl solution and brine, and then dried
over anhydrous Na2SO4. After evaporation of the solvents, the
5.9. 8-(Hydroxymethyl)-3-methyl-7-hydroxy-1-phenyl-benzazepine
18
A solution of aldehyde 17 (35 mg, 0.12 mmol) and NaBH4
(9.4 mg, 0.24 mmol) in MeOH (5 mL) was stirred at room

J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
23
temperature for 1 h. Water was added, and the mixture was
extracted with CHCl3. The combined organic phase was washed
with brine, dried over anhydrous Na2SO4 and evaporated. The
crude product was purified by column chromatography
(CHCl3:MeOH ¼ 15:1) to give alcohol 18 (35 mg, 99%) as a white
Tol)3 (0.8 mmol) and K3PO4 (2.0 mmol) were added under N2. The
reaction was irradiated in microwave reactor at 110 ꢀC for 1 h. After
cooled to room temperature, the mixture was poured into water,
and extracted with Et2O. The combined organic phase was washed
with H2O and brine, dried over anhydrous Na2SO4, and then
evaporated. The residue was then subjected to demethylation by
using a procedure previously reported to yield the desired products
21aee.
solid. 1H NMR (300 MHz, CDCl3 þ CD3OD)
d 7.26 (m, 5H), 6.64 (s,
1H), 6.31 (s, 1H), 4.53 (s, 2H), 4.27 (d, J ¼ 8.4 Hz, 1H), 2.93 (m, 5H),
2.37 (m, 4H); 13C NMR (100 MHz, CDCl3 þ CD3OD)
d 153.9, 142.8,
141.4, 134.9, 128.5, 128.2, 128.0, 126.3, 123.0, 117.0, 63.1, 62.6, 56.8,
49.2, 47.0, 35.1; EI-MS m/z: 283 (Mþ); HR-MS calcd for C18H21NO2
(Mþ) 283.1572. Found: 283.1564.
5.12.1. 7-(p-Tolyl)-3-methyl-1-phenyl-8-hydroxy-benzazepine 21a
Light yellow solid (73% yield for two steps). 1H NMR (300 MHz,
CDCl3)
J ¼ 8.7 Hz, 1H), 3.13 (m, 2H), 2.90 (m, 1H), 2.75 (m, 2H), 2.40 (s, 3H),
2.28 (m, 4H); 13C NMR (100 MHz, CDCl3)
151.2, 145.1, 143.1, 136.9,
d 7.32 (m, 7H), 7.10 (m, 2H), 7.04 (s, 1H), 6.13 (s, 1H), 4.30 (d,
5.10. 8-(Aminomethyl)-3-methyl-7-hydroxy-1-phenyl-benzazepine
19
d
134.5, 132.7, 131.2, 129.4, 128.9, 128.6, 126.5, 125.6, 115.9, 63.1, 57.2,
48.7, 47.0, 35.0, 21.2; EI-MS m/z: 343 (Mþ); HR-MS calcd for
A mixture of aldehyde 17 (66 mg, 0.23 mmol), hydroxylamine
hydrochloride (33 mg, 0.47 mmol) and NaOAc (38 mg, 0.47 mmol)
in MeOH was heated to reflux for 2 h. The mixture was cooled and
concentrated. The residue was basified with NH3$H2O, and
extracted with CHCl3. The combined organic phase was washed
with brine, dried over anhydrous Na2SO4, and then evaporated to
give the oxime intermediate (61 mg, 88%) as a white solid. The
intermediate (32 mg, 0.11 mmol) was then dissolved in dry THF
(10 mL), and lithium aluminum hydride (12 mg, 0.32 mmol) was
added. The mixture was heated to reflux for 3 h, then cooled and
quenched with 10% NaOH aqueous solution slowly. The white
suspension was filtered, and the filtrate was dried over anhydrous
Na2SO4. After evaporation, the crude material was chromatogra-
phied (CHCl3:MeOH ¼ 5:1) to give amine 19 (20 mg, 66%) as a
C
24H25NO (Mþ) 343.1936. Found: 343.1933.
5.12.2. 7-(m-Tolyl)-3-methyl-1-phenyl-8-hydroxy-benzazepine 21b
Yellow solid (60% yield for two steps). 1H NMR (300 MHz, CDCl3)
d
7.30 (m, 6H), 7.15 (m, 3H), 7.02 (s, 1H), 6.18 (s, 1H), 4.36 (d,
J ¼ 8.7 Hz, 1H), 3.17 (m, 2H), 2.86 (m, 3H), 2.36 (m, 7H); 13C NMR
(100 MHz, CDCl3)
d 151.0, 145.2, 142.8, 138.6, 137.1, 132.7, 131.2,
129.7, 128.8, 128.7, 128.5, 128.2, 126.6, 126.0, 125.8, 115.9, 63.0, 57.3,
48.7, 47.1, 34.8, 21.5; EI-MS m/z: 343 (Mþ); HR-MS calcd for
C
24H25NO (Mþ) 343.1936. Found: 343.1936.
5.12.3. 7-(o-Tolyl)-3-methyl-1-phenyl-8-hydroxy-benzazepine 21c
colorless liquid. 1H NMR (300 MHz, CDCl3)
1H), 6.23 (s, 1H), 4.25 (d, J ¼ 8.1 Hz, 1H), 3.91 (brs, 2H), 2.94 (m, 5H),
2.40 (m, 4H); 13C NMR (100 MHz, CDCl3)
156.4, 143.6, 141.7, 134.7,
d
7.26 (m, 5H), 6.66 (s,
White solid (70% yield for two steps). 1H NMR (300 MHz, CDCl3)
d
7.26 (m, 9H), 6.88 (s, 1H), 6.16 (s, 1H), 4.33 (d, J ¼ 9.0 Hz, 1H), 3.14
(m, 2H), 2.81 (m, 3H), 2.26 (m, 7H); 13C NMR (100 MHz, CDCl3)
d 151.1, 145.3, 143.1, 137.2, 136.5, 132.4, 131.3, 130.4, 130.3, 128.6,
d
128.5, 128.4, 127.9, 126.2, 117.9, 63.4, 57.2, 49.0, 47.6, 45.0, 36.0; EI-
MS m/z: 282 (Mþ); HR-MS calcd for C18H22N2O (Mþ) 282.1732.
Found: 282.1727.
128.5, 127.9, 126.5, 126.0, 125.5, 115.3, 63.1, 57.4, 48.9, 47.1, 35.0,
19.9; EI-MS m/z: 343 (Mþ); HR-MS calcd for C24H25NO (Mþ)
343.1936. Found: 343.1935.
5.11. 7-Bromo-3-methyl-8-methoxy-1-phenyl-benzazepine 20
5.12.4. 7-(2-Naphthyl)-3-methyl-1-phenyl-8-hydroxy-benzazepine
21d
Br2 (0.98 mL, 19.20 mmol) was added to a mixture of compound
8-methoxy-1-phenyl-2,3,4,5-tetrahydro-1H-benzo[d]azepine
80
White solid (41% yield for two steps). 1H NMR (300 MHz, CDCl3)
(4.3 g, 16.08 mmol) in glacial acetic acid (30 mL), then the mixture
was stirred at 100 ꢀC for 1 h. The reaction mixture was cooled,
poured into water, and then extracted with CH2Cl2. The combined
organic phase was washed with 2 N NaOH solution, followed by
water and brine. After the solvents were evaporated, the residue
was purified by column chromatography (CHCl3:MeOH ¼ 12:1) to
give bromide intermediate (2.6 g, 49%) as a yellow liquid. 1H NMR
d
7.97 (s, 1H), 7.87 (m, 3H), 7.68 (m, 1H), 7.49 (m, 2H), 7.29 (m, 3H),
7.16 (s, 1H), 7.09 (m, 2H), 6.16 (s,1H), 4.33 (d, J ¼ 9.3 Hz,1H), 3.14 (m,
2H), 2.84 (m, 3H), 2.27 (m, 4H); 13C NMR (100 MHz, CDCl3)
151.6,
d
145.4, 143.0, 135.2, 133.5, 132.7, 132.4, 131.5, 128.6, 128.1, 128.0,
127.7, 127.6, 127.5, 126.5, 126.1, 125.9, 125.7, 116.1, 63.0, 57.2, 48.6,
46.9, 34.8; EI-MS m/z: 379 (Mþ); HR-MS calcd for C27H25NO (Mþ)
379.1936. Found: 379.1933.
(300 MHz, CDCl3)
d 7.31 (m, 4H), 7.12 (m, 2H), 6.44 (s, 1H), 4.24 (d,
J ¼ 6.3 Hz, 1H), 3.70 (s, 3H), 3.57 (m, 1H), 3.35 (m, 1H), 2.95 (m, 2H),
2.78 (m, 2H), 2.64 (brs, 1H). A mixture of the bromide intermediate
(610 mg, 1.84 mmol) and 37% formaldehyde aqueous solution
(7 mL) in formic acid (20 mL) was stirred under 100 ꢀC overnight.
After cooled down and concentrated, the obtained residue was
taken up in H2O and basified with 2 N NaOH aqueous solution. The
suspension was extracted with EtOAc, and the organic phase was
washed with H2O and brine. After evalopated, the obtained crude
product was purified by chromatography (EtOAc:MeOH ¼ 8:1) to
give the yellow liquid 20 (520 mg, 81.8%). 1H NMR (300 MHz, CDCl3)
5.12.5. 7-(p-Hydroxyphenyl)-3-methyl-1-phenyl-8-hydroxy-
benzazepine 21e
White solid (27% yield for two steps). 1H NMR (300 MHz,
CD3OD) d 7.31 (m, 7H), 7.02 (s, 1H), 6.79 (m, 2H), 6.17 (s, 1H), 4.35 (d,
J ¼ 8.7 Hz, 1H), 3.03 (m, 5H), 2.43 (m, 4H); 13C NMR (100 MHz,
CD3OD)
d 157.7, 153.9, 145.2, 144.6, 133.3, 133.0, 131.8, 131.5, 130.4,
130.1,128.3,127.9,117.6,116.3, 64.7, 59.2, 50.1, 48.0, 35.8; EI-MS m/z:
345 (Mþ); HR-MS calcd for C23H23NO2 (Mþ) 345.1729. Found:
345.1736.
d
7.31 (m, 4H), 7.18 (m, 2H), 6.24 (s, 1H), 4.30 (d, J ¼ 8.4 Hz, 1H), 3.60
(s, 3H), 2.90 (m, 5H), 2.39 (m, 4H).
5.13. 7-Amino or 8-amino-3-methyl-1-phenyl-benzazepine 22 or
23
5.12. 7-Aryl-3-methyl-8-hydroxy-1-phenyl-benzazepines 21aee
Amines 22 and 23 were prepared by triflation of corresponding
phenols 9 and 10 followed by a CeN coupling using a set of pro-
cedures similar to that reported by us early [22,23].
To a mixture of bromide 20 (1.0 mmol) in dry DMF (4 mL), an
appropriate arylboronic acid (2.0 mmol), Pd(OAc)2 (0.2 mmol), P(o-

24
J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
5.13.1. 7-Amino-3-methyl-1-phenyl-benzazepine 22
5.16.2. 6-Nitro-7-hydroxy-3-methyl-1-phenyl-benzazepines 27
Yellow liquid (24%). 1H NMR (300 MHz, CDCl3)
11.01 (brs, 1H),
1
Yellow liquid (82%). H NMR (300 MHz, CDCl3)
d
7.28 (m, 5H),
d
6.51 (s, 1H), 6.40 (m, 2H), 4.27 (d, J ¼ 8.1 Hz, 1H), 3.52 (brs, 2H), 3.06
7.30 (m, 3H), 7.06 (d, J ¼ 7.2 Hz, 2H), 6.44 (m, 2H), 4.34 (d, J ¼ 9.0 Hz,
(m, 2H), 2.88 (m, 2H), 2.74 (m, 1H), 2.40 (m, 4H); 13C NMR
1H), 3.01 (m, 5H), 2.51 (m, 4H).
(100 MHz, CDCl3)
d 144.6, 143.6, 142.2, 134.4, 129.2, 128.4, 128.3,
126.1, 116.6, 112.4, 63.3, 57.1, 48.9, 47.7, 36.4; EI-MS m/z: 252 (Mþ);
5.17. 8-Amino-7-hydroxy-3-methyl-1-phenyl-benzazepines 28
HR-MS calcd for C17H20N2 (Mþ) 252.1626. Found: 252.1632.
This compound was prepared as white solid (90%) from 26 by
using a procedure similar to that for preparation of amine 12. 1H
5.13.2. 8-Amino-3-methyl-1-phenyl-benzazepine 23
NMR (300 MHz, CDCl3)
(brs, 2H), 4.25 (d, J ¼ 8.4 Hz,1H), 2.96 (m, 4H), 2.44 (m, 4H), 2.27 (m,
1H); 13C NMR (100 MHz, CDCl3)
143.4, 142.7, 136.1, 132.4, 131.6,
d 7.25 (m, 5H), 6.40 (s, 1H), 5.94 (s, 1H), 5.10
Light yellow liquid (83%). 1H NMR (300 MHz, CDCl3)
d 7.28 (m,
5H), 6.93 (d, J ¼ 8.1 Hz, 1H), 6.45 (dd, J ¼ 2.1, 7.5 Hz, 1H), 5.99 (s, 1H),
d
4.31 (d, J ¼ 8.7 Hz, 1H), 3.51 (brs, 2H), 3.10 (m, 2H), 2.90 (m, 2H),
128.5, 128.4, 126.3, 117.0, 63.3, 57.4, 48.2, 47.2, 34.8; EI-MS m/z: 268
2.75 (m,1H), 2.38 (m, 4H); 13C NMR (100 MHz, CDCl3)
d 145.2,144.5,
(Mþ); HR-MS calcd for C17H20N2O (Mþ) 268.1576. Found: 268.1578.
143.1, 131.1, 130.3, 128.5, 126.3, 115.4, 112.5, 63.0, 57.6, 49.4, 47.5,
35.2; EI-MS m/z: 252 (Mþ); HR-MS calcd for C17H20N2 (Mþ)
252.1626. Found: 252.1630.
5.18. 7-Methyl-3,5,6,7,8,9-hexahydro-2-oxazolo[40,50:4,5]
benzazepin-2-one 29
A mixture of compound 28 (43 mg, 0.16 mmol) and carbon-
yldiimidazole (CDI, 78 mg, 0.48 mmol) in dry THF (6 mL) was
refluxed for 1 h. After cooled, the mixture was concentrated, and
the residue was dissolved in CHCl3. The organic phase was washed
with saturated NH4Cl solution and then dried over anhydrous
Na2SO4. After evaporation, the crude material was subjected to
chromatograph on silica gel (CHCl3:MeOH ¼ 30:1) to afford product
5.14. 8-(N-methylamino)-3-methyl-1-phenyl-benzazepine 24
A solution of amine 23 (36 mg, 0.14 mmol) in ethyl formate
(7 mL) was heated to reflux for 3 h under N2. After cooled, the
mixture was concentrated. The residue was subjected to flash
chromatography to yield N-formyl product (32 mg, 80%) as a yellow
liquid, which was then subjected to reduction with LiAlH4 (15 mg,
0.39 mmol). After work-up, the crude product was purified by
preparative TLC (CHCl3:MeOH ¼ 12:1) to give product 24 (17 mg,
29 (32 mg, 68%) as a yellow liquid. 1H NMR (300 MHz, CDCl3)
d
7.22
(m, 5H), 6.95 (s, 1H), 6.25 (s, 1H), 4.37 (d, J ¼ 8.4 Hz, 1H), 2.97 (m,
5H), 2.34 (m, 4H); 13C NMR (100 MHz, CDCl3)
156.1, 142.7, 142.1,
60%) as a colorless liquid. 1H NMR (300 MHz, CDCl3)
d 7.28 (m, 5H),
d
140.4, 135.4, 128.7, 128.3, 127.8, 126.7, 110.8, 109.9, 62.7, 56.8, 49.0,
47.1, 35.8; EI-MS m/z: 294 (Mþ); HR-MS calcd for C18H18N2O2 (Mþ)
294.1368. Found: 294.1364.
6.98 (d, J ¼ 7.8 Hz, 1H), 6.38 (dd, J ¼ 2.1, 7.8 Hz, 1H), 5.89 (s, 1H), 4.50
(d, J ¼ 9.3 Hz, 1H), 3.15 (m, 4H), 2.79 (m, 1H), 2.66 (s, 3H), 2.51 (m,
4H); 13C NMR (100 MHz, CDCl3)
d 147.9, 144.4, 142.2, 130.4, 128.8,
128.6, 128.5, 126.6, 113.6, 109.2, 62.6, 57.6, 48.3, 47.0, 34.1, 30.6; EI-
MS m/z: 266 (Mþ); HR-MS calcd for C18H22N2 (Mþ) 266.1783.
Found: 266.1786.
5.19. 6-Phenyl-4,6,7,8,9,10-hexahydro-[1,4]oxazino[20,30:4,5]
benzazepin-3-one 30
To a suspension of 28 (35 mg, 0.13 mmol) and Cs2CO3 (127 mg,
0.39 mmol) in CH3CN (6 mL) in ice-bath, was added slowly a so-
lution of chloroacetic chloride (18 mg, 0.16 mmol) in CH3CN. The
mixture was allowed to stir at room temperature overnight and
concentrated. The obtained residue was dissolved in CHCl3, washed
with brine, dried over anhydrous Na2SO4 and then evaporated.
The residue was further purified by column chromatography
(CHCl3:MeOH ¼ 30:1) to give compound 30 (30 mg, 75%) as a
5.15. 8-(N,N-dimethylamino)-3-methyl-1-phenyl-benzazepine 25
A mixture of amine 23 (40 mg, 0.16 mmol) in formalin (2 mL)
and formic acid (6 mL) was stirred at 90 ꢀC under N2 overnight. The
reaction was cooled and then concentrated. The obtained residue
was basified with 2 N NaOH solution and extracted with CHCl3. The
combined organic phase was washed with brine, and dried over
anhydrous Na2SO4. The crude product was then subjected to pre-
parative TLC (CHCl3:MeOH ¼ 12:1) to give compound 25 (8 mg,
colorless liquid. 1H NMR (300 MHz, CDCl3)
d
8.34 (s, 1H), 7.26 (m,
5H), 6.75 (s, 1H), 6.04 (s, 1H), 4.49 (s, 2H), 4.25 (d, J ¼ 9.0 Hz, 1H),
2.89 (m, 5H), 2.34 (m, 4H); 13C NMR (100 MHz, CDCl3)
165.3,142.8,
18%) as a colorless liquid. 1H NMR (300 MHz, CDCl3)
d 7.27 (m, 5H),
d
7.01 (d, J ¼ 8.4 Hz, 1H), 6.51 (dd, J ¼ 2.4, 8.1 Hz, 1H), 6.08 (s, 1H), 4.46
141.5, 139.0, 137.1, 128.7, 128.3, 126.6, 123.7, 117.7, 116.0, 67.3, 62.9,
(d, J ¼ 8.4 Hz, 1H), 3.11 (m, 4H), 2.80 (m, 1H), 2.74 (s, 6H), 2.49 (m,
57.0, 48.9, 47.4, 35.6; EI-MS m/z: 308 (Mþ); HR-MS calcd for
4H); 13C NMR (100 MHz, CDCl3)
d 149.2, 144.3, 142.5, 130.2, 128.5,
C
19H20N2O2 (Mþ) 308.1525. Found: 308.1522.
128.4, 126.5, 113.6, 110.3, 62.6, 57.7, 49.2, 47.3, 40.5, 34.4; EI-MS m/z:
280 (Mþ); HR-MS calcd for C19H24N2 (Mþ) 280.1939. Found:
280.1932.
5.20. 7-Methyl-5-phenyl-6,7,8,9-tetrahydro-5H-oxazolo[40,50:4,5]
benzazepine 31
A mixture of compound 28 (50 mg, 0.19 mmol) and orthoethyl
formate (200 mg, 1.35 mmol) in EtOH (6 mL) was heated to reflux
overnight. After cooled to room temperature, the reaction mixture
was concentrated. The residue was taken-up in CH2Cl2, washed
with 2 N NaOH solution and brine, and then evaporated. The ob-
tained residue was subjected to column chromatograph
(CHCl3:MeOH ¼ 50:1) to give compound 31 (48 mg, 93%) as a
5.16. 6-And 8-nitro-7-hydroxy-3-methyl-1-phenyl-benzazepines
26 and 27
These two compounds were prepared by nitration of phenol 8
following a similar procedure as that for preparation of 11 and 110.
5.16.1. 8-Nitro-7-hydroxy-3-methyl-1-phenyl-benzazepines 26
colorless liquid. 1H NMR (300 MHz, CDCl3)
d
7.97 (s, 1H), 7.24 (m,
7H), 4.46 (d, J ¼ 8.4 Hz,1H), 3.24 (m, 2H), 2.92 (m, 3H), 2.39 (m, 4H);
13C NMR (100 MHz, CDCl3)
152.2, 148.3, 142.9, 141.8, 139.7, 138.1,
128.6, 128.3, 126.5, 119.8, 111.1, 63.0, 57.2, 49.5, 47.4, 36.6; EI-MS m/
Yellow liquid (43%). 1H NMR (300 MHz, CDCl3)
d 7.30 (m, 6H),
6.93 (s, 1H), 6.87 (d, J ¼ 8.7 Hz, 1H), 3.11 (m, 2H), 2.88 (m, 3H), 2.41
d
(m, 4H).

J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
25
z: 278 (Mþ); HR-MS calcd for C18H18N2O (Mþ) 278.1419. Found:
278.1410.
previously [12,13,22,23]. Binding buffer contains 50 mM Tris, 4 mM
MgCl2, pH7.4 in the presence of tested compound and. 0.7 nM
respective radioligand ([3H]8-OH-DPAT for 5-HT1A receptor, [3H]
Ketanserin for 5-HT2A receptor, [3H]SCH23390 for D1 dopamine
receptor, [3H]Spiperone for D2 and D3 dopamine receptor). Dupli-
cated tubes were incubated at 30 ꢀC for 50 min with increasing
concentrations of respective compounds and different radioligands
5.21. 7-Methyl-5-phenyl-tetrahydro-5H-oxazolo[40,50:4,5]
benzazepine-2-thiol 32
A mixture of compound 28 (60 mg, 0.22 mmol), KOH (31 mg,
0.56 mmol) in CS2/EtOH (CS2:EtOH ¼ 1:2, 9 mL) was heated to
reflux for 3 h. After cooled, the mixture was concentrated, and the
residue was dissolved in CHCl3. The organic phase was washed with
saturated NH4Cl and brine, and then dried over anhydrous Na2SO4.
After concentrated, the crude product was purified by column
chromatograph (CHCl3:MeOH ¼ 30:1) to afford product 32 (62 mg,
in a final volume of 200
mL binding buffer. Nonspecific binding was
determined by parallel incubations with 10
mM WAY100635 for 5-
HT1A receptor, Ketanserin for 5-HT2A receptor, SCH23390 for D1
dopamine receptor, Spiperone for D2 and D3 dopamine receptor
respectively. The competition binding reaction was started by
adding membrane homogenates (15 ng/tube) and stopped by rapid
filtration through Whatman GF/B glass fiber filter, and subse-
quently washed with cold washing buffer using a Brandel 24-well
cell harvest. Scintillation cocktail was added and the radioactivity
was determined in a Microbeta liquid scintillation counter. The IC50
and Ki values were calculated by nonlinear regression (PRISM,
Graphpad, San Diego, CA) using a sigmodial function.
89%) as a white solid. 1H NMR (300 MHz, CDCl3 þ CD3OD)
d 7.24 (m,
5H), 6.99 (s, 1H), 6.25 (s, 1H), 4.56 (d, J ¼ 8.4 Hz, 1H), 3.08 (m, 5H),
2.56 (m, 4H); 13C NMR (100 MHz, CDCl3 þ CD3OD)
d 181.2, 147.8,
141.9, 140.0, 135.2, 132.0, 128.7, 128.2, 126.9, 110.9, 110.3, 61.6, 56.2,
47.8, 46.4, 34.5; EI-MS m/z: 310 (Mþ); HR-MS calcd for C18H18N2OS
(Mþ) 310.1140. Found: 310.1140.
5.22. 7-Methyl-2-(4-methylpiperazinyl)-5-phenyl-oxazolo[40,50:4,5]
benzazepine 33
Acknowledgments
This work was supported by grants from the Chinese National
Sciences Foundation (81125021, 81373277, 81130023), Chinese
National Science & Technology Major Project on ‘Key New Drug
Creation and Manufacturing Program’ (2012ZX09103-101-035,
2012ZX09301001-001), and National Basic Research Plan (973) of
the Ministry of Science and Technology of China (2009CB522000,
2011CB5C4403). Supporting grants from the “Interdisciplinary
Cooperation Team” Program for Science and Technology Innovation
(CAS), and grant from the State Key Laboratory of Drug Research
(SIMM1302KF-08), Shanghai Institute of Materia Medica were also
appreciated.
To a solution of thiol 32 (40 mg, 0.13 mmol) in dry toluene
(4 mL), was added N-methylpiperazine (0.5 mL). The mixture was
irradiated in microwave reactor at 100 ꢀC for 50 min. After cooled,
the mixture was concentrated and dissolved in CHCl3. The organic
phase was washed with brine, dried over anhydrous Na2SO4, and
then concentrated. The obtained residue was further purified by
chromatograph (CHCl3:MeOH ¼ 30:1) to yield product 33 (35 mg,
72%) as a colorless liquid. 1H NMR (300 MHz, CDCl3)
d 7.25 (m, 5H),
7.03 (s, 1H), 6.67 (s, 1H), 4.37 (d, J ¼ 8.7 Hz, 1H), 3.64 (t, J ¼ 5.1 Hz,
4H), 3.13 (m, 2H), 2.88 (m, 3H), 2.46 (m, 4H), 2.35 (m, 7H); 13C NMR
(100 MHz, CDCl3)
d 162.2, 147.0, 143.4, 140.9, 140.7, 134.3, 128.5,
128.4, 126.3, 116.2, 109.4, 63.1, 57.5, 54.1, 49.5, 47.6, 46.2, 45.5, 36.4;
EI-MS m/z: 376 (Mþ); HR-MS calcd for C23H28N4O (Mþ) 376.2263.
Found: 376.2272.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.07.059.
5.23. Established stable expression of cell lines
References
The rat 5-HT1A receptor gene, human 5-HT2A receptor gene,
human D1, D2 and D3 receptor genes were individually cloned into
pcDNA 3.0 vector. The cell line that stable expressed respective
receptor was prepared as described before [12,13,22,23]. Briefly, the
5-HT1A and 5-HT2A plasmid was transfected into CHO cells. The 5-
HT2A, D1, D2, D3 receptor were transfected into HEK293 cells
respectively. After 24 h, the cells were selected with antibiotic G418
at a concentration of 800 mg/L. After several rounds of selection,
monoclonal transfected cells were isolated and cultured in medium
containing Ham's F12 nutrient mixture for CHO cells and DMEM for
HEK293 cells (Gibco), respectively, with 10% fetal bovine serum
(Gibco), 100 U/mL penicillin (Sigma), 100 U/mL streptomycin
(Sigma), and 200 mg/L G418 at 37 ꢀC and 5% CO2.
[1] P.F. Spano, S. Govoni, M. Trabucchi, Studies on the pharmacological properties
of dopamine receptors in various areas of the central nervous system, Adv.
Biochem. Psychopharmacol. 19 (1978) 155e165.
[2] C.A. Tamminga, Partial dopamine agonists in the treatment of psychosis,
J. Neural. Transm. 109 (2002) 411e420.
[3] P. Maharajan, V. Maharajan, G. Ravagnan, G. Paino, The weaver mutant
mouse: a model to study the ontogeny of dopamine transmission systems and
their role in drug addiction, Prog. Neurobiol. 64 (2001) 269e276.
[4] A. Zhang, J.L. Neumeyer, R.J. Baldessarini, Recent progress in the development
of dopamine receptor subtype compounds: potential therapeutic agents for
neurological and neuropsychiatric disorders, Chem. Rev. 107 (2007) 274e303.
[5] N. Ye, J.L. Neumeyer, R.J. Baldessarini, X. Zhen, A. Zhang, Recent progress in the
development of dopamine receptor subtype compounds: potential thera-
peutic agents for neurological and neuropsychiatric disorders, Chem. Rev. 113
(2013) PR123ePR178.
[6] J. Zhang, B. Xiong, X. Zhen, A. Zhang, Dopamine D1 receptor ligands: where are
we now and where are we going, Med. Res. Rev. 29 (2009) 272e294.
[7] P.E. Setler, H.M. Sarau, C.L. Zirkle, H.L. Saunders, The central effects of a novel
dopamine agonist, Eur. J. Pharmacol. 50 (1978) 419e530.
[8] C. Kaiser, P.A. Dandridge, E. Garvey, R.A. Hahn, H.M. Sarau, P.E. Setler, L.S. Bass,
J. Clardy, Absolute stereochemistry and dopaminergic activity of enantiomers
of 2,3,4,5-tetrahydro-7,8-dihydroxy-1-phenyl-1H-3-benzazepine, J. Med.
Chem. 25 (1982) 697e703.
[9] J.A. Bourne, SCH 23390: the first selective dopamine D1-like receptor antag-
onist, CNS Drug Rev. 7 (2001) 399e414.
[10] M.M. Lewis, X. Huang, D.E. Nichols, R.B. Mailman, D1 and functionally selec-
tive dopamine agonists as neuroprotective agents in Parkinson's disease, CNS
Neurol. Disord. Drug Targets 5 (2006) 345e353.
5.24. Binding assay of new compounds at the D1eD3 and 5-HT1A, 5-
HT2A receptors
The affinity of compounds binding to respective dopamine
(D1eD3) or 5-HT (5-HT1A, 5-HT2A) receptors was measured by
competition binding assay according our published method pre-
viously [12,13,22,23]. All radioligands were purchased from Perki-
nElmer, USA. All compounds were dissolved in DMSO to prepare a
stock solution at a concentration of 10-1 M. Membrane homoge-
nates of 5-HT1A-CHO cells, 5-HT2A-HEK293 cells, D1-HEK293 cells,
D2-HEK293 cells and D3-HEK293 cells were prepared as described
[11] J.L. Neumeyer, N.S. Kula, J. Bergman, R.J. Baldessarini, Receptor affinities of
dopamine D1 receptor-selective novel phenylbenzazepines, Eur. J. Pharm. 474
(2003) 137e140.

26
J. Zhang et al. / European Journal of Medicinal Chemistry 85 (2014) 16e26
[12] J. Zhang, X. Chen, L. Yu, X. Zhen, A. Zhang, Synthesis of 6-substituted 1-
phenylbenzazepines and their dopamine D1 receptor activities, Bioorg. Med.
Chem. 16 (2008) 9425e9431.
[13] J. Zhang, H. Zhang, W. Cai, L. Yu, X. Zhen, A. Zhang, “Click” D1 receptor agonists
with a 5-HT1A receptor pharmacophore producing D2 receptor activity, Bio-
org. Med. Chem. 17 (2009) 4873e4880.
[14] J.H. Shah, R.H. Kline, B.G. Douglass, S. Izenwasser, J.M. Witkin, A.H. Newman,
3-[40-(N,N-dimethylamino)cinnamyl]benzazepine analogs: novel dopamine
D1 receptor antagonists, J. Med. Chem. 39 (1996) 3423e3428.
[15] E.S. Tan, M. Miyakawa, J.R. Bunzow, D.K. Grandy, T.S. Scanlan, Exploring the
structure-activity relationship of the ethylamine portion of 3-
iodothyronamine for rat and mouse trace amine-associated receptor 1,
J. Med. Chem. 50 (2007) 2787e2798.
[16] N. Baindur, M. Tran, H.B. Niznik, H.C. Guan, P. Seeman, J.L. Neumeyer, 3-Allyl-
7-halo-8-hydroxy-1-phenyl-2,3,4,5-tetrahedro-1H-3-benzazepines as selec-
tive high affinity D1 dopamine receptor antagonists: synthesis and structure-
activity relationship, J. Med. Chem. 35 (1992) 67e72.
[17] K.G. Holden, C. Kaiser, J. Weinstock, 7,8-amino, hydroxyl-1-phenyl-2,3,4,5-
tetrahedro-1H-3-benzazepines, process for their preparation and pharmaco-
logical compositions containing them. EP0033632 (Smithkline Corp.).
[18] W.E. Bondinell, 1-Phenyl-3-benzazepine compounds and medicaments con-
taining these compounds for treating gastrointestinal motility disorders,
EP0244088 (Smithkline Beckman Corp.).
[20] J. Zhang, A. Zhang, Unprecedented FeCl3$6H2O-promoted skeleton-
rearrangement of 1-aryl-2,3,4,5-tetrahydro-1H-3-benzazepines: a new strat-
egy for the synthesis of C-1 quaternary tetrahydroisoquinolines, Chem. Eur. J.
15 (2009) 11119e11122.
[21] X. Chen, J. Zhang, J.L. Neumeyer, G.-Z. Jin, G.-Y. Hu, A. Zhang, X. Zhen, Arylben-
zazepines are potent modulators for the delayed rectifier Kþ channel: a po-
tential mechanism for their neuroprotective effects, PLoS One 4 (2009) e5811.
[22] H. Zhang, N. Ye, S. Zhou, L. Guo, L. Zheng, Z. Liu, B. Gao, X. Zhen, A. Zhang,
Identification of N-propylnoraporphin-11-yl 5-(1,2-dithiolan-3-yl)pentanoate
as a new anti-Parkinson's agent possessing a dopamine D2 and serotonin 5-
HT1A dual-agonist profile, J. Med. Chem. 54 (2011) 4324e4338.
[23] Z. Liu, H. Zhang, N. Ye, J. Zhang, Q.Q. Wu, P. Sun, L. Li, X. Zhen, A. Zhang,
Synthesis of dihydrofuroaporphine derivatives: identification of a potent and
selective serotonin 5-HT1A receptor agonist, J. Med. Chem. 53 (2010)
1319e1328.
[24] K. Liu, H. Lu, L. Hou, Z. Qi, C. Teixeira, F. Barbault, B.T. Fan, S. Liu, S. Jiang, L. Xie,
Design, synthesis, and biological evaluation of N-carboxyphenylpyrrole de-
rivatives as potent HIV fusion inhibitors targeting gp41, J. Med. Chem. 51
(2008) 7843e7854.
[25] C. Fortenberry, B. Nammalwar, R.A. Bunce, Ammonium chloride-catalyzed
synthesis of benzo-fused heterocycles from o-substituted anilines and
orthoesters, Org. Prep. Proced. Int. 45 (2013) 57e65.
[26] R.G. Wilde, J.T. Billheimer, S.J. Germain, P.J. Gillies, C.A. Higley, H.S. Kezar,
T.P. Maduskuie, E.S. Shimshick, R.R. Wexler, Acyl CoA:cholesterol acyl-
transferase (ACAT) inhibitors: ureas bearing heterocyclic groups bioisosteric
for an imidazole, Bioorg. Med. Chem. Lett. 5 (1995) 167e172.
[19] K.J. Young, Y.B. Hoon, H.S. Kyung, O.C. Min, P.M. Seon, S.K. Ok, Urea derivative
useful as an anticancer agent and process for preparing same, WO 0195856
(Chaconne NSI Co., LTD).