Discovery of benzhydrylpiperazine derivatives as CB1 receptor inverse agonists via privileged structure-based approach

Article abstract of DOI:10.1016/j.ejmech.2009.12.018

The present study describes the identification via privileged structure-based approach of the benzhydrylpiperazine moiety as a potential scaffold to develop novel CB₁ receptor modulators. Efficient structural optimization of the initial four hit compounds led to a high quality lead series, represented by compound 6c. Compound 6c is a highly potent and selective CB₁ receptor inverse agonist that is able to reduce body weight in diet-induced obese Sprague-Dawley rats. The preparation of privileged structure-based library, the progression from hit to lead, the structure-activity relationships in the lead series and in vitro and in vivo activity of compound 6c are discussed.

Full text of DOI:10.1016/j.ejmech.2009.12.018

European Journal of Medicinal Chemistry 45 (2010) 1133–1139
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Discovery of benzhydrylpiperazine derivatives as CB1 receptor inverse agonists via
privileged structure-based approach
,
a
a
a
b,
*
Tao Meng a, Jue Wang b, Hongli Peng a , Guanghua Fang , Min Li , Bing Xiong , Xin Xie
,
*
Yongliang Zhang a, Xin Wang a, Jingkang Shen a
,
*
a State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203, PR China
b The National Center for Drug Screening, 189 Guoshoujing Road, Shanghai 201203, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The present study describes the identification via privileged structure-based approach of the benzhy-
drylpiperazine moiety as a potential scaffold to develop novel CB1 receptor modulators. Efficient struc-
tural optimization of the initial four hit compounds led to a high quality lead series, represented by
compound 6c. Compound 6c is a highly potent and selective CB1 receptor inverse agonist that is able to
reduce body weight in diet-induced obese Sprague–Dawley rats. The preparation of privileged structure-
based library, the progression from hit to lead, the structure–activity relationships in the lead series and
in vitro and in vivo activity of compound 6c are discussed.
Received 23 October 2009
Received in revised form
2 December 2009
Accepted 4 December 2009
Available online 16 December 2009
Keywords:
Ó 2009 Elsevier Masson SAS. All rights reserved.
Benzhydrylpiperazine
Privileged structure
Cannabinoid receptor 1(CB1)
Anti-obesity
1. Introduction
other CB1 receptor inverse agonists including taranabant (MK-
0364) and otenabant (CP-945598) were also suspended in late-
The excess consumption of fat-enriched food and sedentary
lifestyle result in obesity, which is currently a worldwide health
problem [1,2]. In order to discover novel anti-obesity therapeutics,
numerous potential drug targets have been studied. Among those,
the endocannabinoid system is of particular interest because it
plays an important role in the regulation of body energy balance
[3]. Cannabinoids and endocannabinoids function via two G-
protein-coupled receptors (GPCRs) known as CB1 and CB2 recep-
tors. CB1 receptor is mainly expressed in the central nervous system
where it controls motivation for appetitive stimuli, including food
and drugs. Currently, CB1 antagonists and inverse agonists are
evaluated for obesity, metabolic disorders, smoking cessation, and
alcohol abuse [4–6].
Rimonabant (SR141716A), a potent and selective CB1 receptor
inverse agonist, shows promising weight-reducing effect in both
animal models and patients. It was launched by Sanofi-Aventis in
Europe for the treatment of obesity and associated metabolic
disorders [7], but recently withdrawn from the market due to its
psychiatric adverse effects (depression and anxiety) [8]. Several
stage clinical development [9,10]. So far, the mechanisms through
which CB1 receptor blockade increase depression and anxiety
remain unclear. It has been noted that CB1 antagonists are potential
usefulness in the treatment of obesity and associated metabolic
diseases as well as in other therapeutical areas [9]. Hence, devel-
oping novel CB1 receptor antagonists with unique structure and
pharmacological properties may provide an option to overcome the
safety problems associated with current CB1 antagonists and opti-
mize the therapeutic benefits of CB1 receptor antagonists.
Scaffold hopping and bioisosteric approaches have been previ-
ously applied in discovery of CB1 receptor antagonists [11–13].
However, the majority of the reported CB1 receptor antagonists and
inverse agonists can be described in terms of a general pharma-
cophore model [12]. The privileged structure concept has also been
used extensively to design libraries that possess a high probability
of producing lead compounds against a variety of targets, especially
for GPCRs [14–16]. A privileged structure is a molecular scaffold
that is able to provide potent and selective ligands for a range of
different biological targets through modification of functional
groups, and usually exhibits good drug-like properties that lead to
more drug-like compound libraries and leads [14,16].
Given the success of privileged structures in medicinal chem-
istry, we envisioned that a similar application of this concept could
rapidly identify novel CB1 antagonists. Herein, we report the
* Corresponding authors. Tel.: þ86 21 50806600 5407; fax: þ86 21 50807088.
E-mail addresses: penghl@mail.shcnc.ac.cn (H. Peng), xxie@mail.shcnc.ac.cn
(X. Xie), jkshen@mail.shcnc.ac.cn (J. Shen).
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.12.018

1134
T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
identification of the benzhydrylpiperazine moiety as a potential
scaffold to develop novel CB1 receptor modulators utilizing privi-
leged structure-based approach.
selective, exhibiting high binding affinity for hCB1 (Ki values
ranging from 50 to 250 nM). In addition, array 3 was selective to
CCR5 (Ki < 6.5
mM), while array 1 showed activity at both of MC4R
(Ki < 10 M) and NMUR1/2 (Ki < 10
m
mM). Remarkably, high hit rates
of w6% can be observed with this privileged structure library of 230
compounds, indicating the successful use of the GPCR-based priv-
ileged structure approach for hit finding.
2. Results and discussion
2.1. Privileged structure-based library construction and CB1 hits
identification
The preparation of privileged structure-based library and
general screening data of this library are outlined in Fig. 1. Six
molecular frameworks, including diphenylmethane, 2-amino-
thiazole, benzimidazole, piperazine, 4-aminopiperidine, and 1,2-di-
phenyl-substituted five-membered heterocyclic fragments were
selected as privileged structure subunits as these structures are
widely present in many GPCR ligands [14–16]. Furthermore,
phenyl-substituted bicyclic framework have been utilized
frequently in medicinal chemistry, and is commonly observed as
core elements of privileged substructures [17]. We then designed
and synthesized a privileged structure-based library of bicyclic
compounds, including substituted benzhydryloxyalkyl guanidines
(array 1), 1-phenylbenzimidazoles (array 2), 4,5-diphenyl-2-ami-
nothiazoles (array 3), N-benzhydrylpiperidin-4-amines (array 4),
diphenylpiperazines (array 5), and 1-phenyl-1,2,3,4-tetrahy-
droisoquinolines (array 6), by using the selected privileged
subunits.
2.2. Structure–activity relationships (SAR) of benzhydrylpiperazine
derivatives at hCB1 receptor
As shown in Table 1, four CB1 hits identified via privileged
structure-based approach share the same benzhydrylpiperazine
core structure that occurs in well-known class of antihistamine
drugs [18], and is a novel scaffold for CB1 modulators. Therefore,
this class of compounds was explored for further SAR studies.
Synthesis of the benzhydryl-1,4-diazepanes 1(a–d) is illustrated
in Scheme 1. The intermediate 1-benzhydryl-1,4-diazepane was
synthesized by condensation of homopiperazine with diphe-
nylmethyl chloride, followed by reaction with the corresponding
acyl chlorides, isocyanates and thioisocyanates to afford the desired
amide 1a, ureas 1(b–c), and thiourea 1d.
The benzhydrylpiperazine derivatives 4(a–g), 5(a–i), and 6(a–h)
were prepared according to Scheme 2. The key intermediates 2(a–
n) were synthesized by Grignard reaction of substituted phenyl-
magnesium bromide, followed by addition of the substituted
benzaldehyde in THF. Treatment of 2(a–n) with thionyl chloride
followed by refluxing with piperazine in acetonitrile afforded the
intermediates 3(a–n). The target compounds were obtained by the
coupling reaction of intermediate 3(a–n) with corresponding acyl
chlorides, isocyanates and thioisocyanates.
This library was evaluated by screening against several GPCRs,
including the human CB1 (hCB1) and CB2 (hCB2) receptors, mela-
nocortin receptor 4 (MC4R), chemokine receptor 5 (CCR5), and
neuromedin U receptor 1/2 (NMUR1/2), utilizing radioligand
binding or functional Ca2þ flux assays. The in vitro activities are
summarized in Fig. 1 and Table 1. Of the six arrays, only array 5 was
Fig. 1. Design of privileged structure-based library targeting GPCRs and pharmacological screening results.

T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
1135
Table 1
Activity of array 5 from the privileged structure-based library.a
Fragment C
hCB1 Ca2þ
IC50 (nM)
Fragment B
X
R
N
Cpd.
R
hCB1 Ki
(nM)
hCB2 Ki
(nM)
∗
N
R
N
Fragment A
N
Y
Fragment D
Fig. 2. Four fragments of the benzhydrylpiperazine core structure.
Rimonabant
hit-1
hit-2
hit-3
hit-4
–
0.74 ꢀ 0.10 126 ꢀ 12 3.2 ꢀ 0.3
164 ꢀ 16 3050 ꢀ 824 ND
52 ꢀ 13 2407 ꢀ 538 ND
239 ꢀ 95 2833 ꢀ 361 ND
Cyclohexylamide
Cyclohexylurea
n-Butylurea
Cyclohexylthiourea 106 ꢀ 26 1866 ꢀ 146 ND
at different positions of the phenyl ring in fragment
B
were examined. As shown in Table 2, the 2,4-dichlorophenyl(4a),
2,5-dichlorophenyl(4c), 2,6-dichlorophenyl(4d), 2-fluoro-4-bro-
mophenyl(4e), 2-chloro-4-fluorophenyl(4f), and 2-fluoro-4-chlor-
ophenyl(4g) significantly improved the hCB1 affinity as well as
selectivity over hCB2, except the 3,4-dichloro substitution (4b),
indicating that ortho-substitution in fragment B is necessary for
high affinity for hCB1. Functional Ca2þ assay revealed that this series
of compounds had inhibitory effects on CP-55940-induced Ca2þ
increase. Among these hCB1 antagonists, the 2,4-dichlorophenyl
moiety (4a) showed the best hCB1 binding affinity as well as
antagonistic activity.
a
Ki and IC50 (mean ꢀ SEM) (n ꢁ 3 independent experiments) were calculated
from dose-response curves. hCB1: human CB1 receptor. hCB2: human CB2 receptor.
ND: not detected. Rimonabant: CB1 reference compound.
All the target compounds were evaluated in vitro at the hCB1 and
hCB2 receptors, stably expressed into Chinese Hamster Ovary (CHO)
cells, utilizing radioligand binding studies. CB1 receptor antagonism
was measured using a CP-55940 (an agonist of the CBRs) induced
Ca2þ increase functional assay in CHO cells co-expressing hCB1
receptor and Ga15/16.
Having identified 2,4-dichloro as the optimal substitution of the
phenyl ring in fragment B, various acyl (amide, urea, thiourea)
wherein substituent was a 4–10 atom aliphatic ring or chain were
examined for the SAR of fragment C. As shown in Table 3, trans-
formation of urea of 4a to amide to give 5c or replacement of
cyclohexyl ring of 4a with piperidinyl to give 5h had a small or no
effect on binding affinity, but resulted in an increase in selectivity
for hCB1. On the other hand, a great loss in hCB1 activity was
observed when the cyclohexyl ring of 4a was replaced by the n-
butyl (5g), indicating the importance of the aliphatic ring structure.
Similarly, cyclohexylthiourea (5i) enhanced binding affinity and
selectivity for the hCB1 receptor.
The general structure of identified hit compounds was divided
into four fragments (A–D shown in Fig. 2), and each fragment was
modified for the SAR study separately. We first examined whether
enlargement of the piperazine ring (fragment A) would improve
hCB1 activity. Replacement of piperazine with homopiperazine led to
a complete loss of binding affinity (Ki > 10,000 nM)for both hCB1 and
hCB2, indicating that piperazine is essential to the biological activity.
The SAR from the hit compounds demonstrated that the cyclo-
hexylurea in fragment C (Table 1, hit 2) correlated with relative
higher binding affinity and selectivity for hCB1. Keeping fragment C
as the cyclohexylurea and fragment D unaltered, various halogens
R
N
NH
OH
N
N
i
ii
1a R = cyclohexylcarboxamide;
1b
R = cyclohexylurea;
1c R = n-butylurea;
1d R = cyclohexylthiourea;
Scheme 1. Reagents and conditions: (i) SOCl2, CH2Cl2, rt; homopiperazine, CH3CN, reflux; (iii) for compound 1a: cyclohexanecarbonyl chloride, Et3N, CH2Cl2, rt; for compound 1b–c:
isocyanates, CH2Cl2, rt; for compound 1d: cyclohexyl isothiocyanate, CH2Cl2, rt.
NH
X
X
OH
Br
N
CHO
ii
i
X
+
Y
Y
Y
2a - 2n
3a - 3n
W
N
R
X
N
iii
4a - 4g
;
(Table 2)
5a - 5i (Table 3);
6a - 6h (Table 4);
Y
Scheme 2. Reagents and conditions: (i) Mg/THF, rt then reflux; (ii) SOCl2, CH2Cl2, rt; piperazine, CH3CN, reflux; (iii) isocyanates, CH2Cl2, rt, or acyl chlorides, Et3N, CH2Cl2, rt, or
thioisocyanates, CH2Cl2, rt.

1136
T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
Table 2
Table 4
SAR of Fragment B Modification at hCB1 receptor.a
SAR of Fragment D Modification at hCB1 receptor.a
hCB2 Ki (nM) hCB1 Ca2þ IC50 (nM)
Cpd
Y
R
hCB1 Ki
(nM)
hCB2 Ki
(nM)
hCB1 Ca2þ
IC50 (nM)
Cpd.
Fragment B
X
hCB1 Ki (nM)
O
Cl
Cl
R
N
N
N
X
H
N
N
Y
Fragment D
4a
2,4-diCl
3,4-diCl
2,5-diCl
2,6-diCl
2-F-4-Br
2-Cl-4-F
2-F-4-Cl
1.6 ꢀ 0.2
60 ꢀ 17
432 ꢀ 118
356 ꢀ 174
899 ꢀ 292
831 ꢀ 169
322 ꢀ 89
9.2 ꢀ 1.6
462 ꢀ 97
43 ꢀ 18
55 ꢀ 37
87 ꢀ 11
68 ꢀ 26
124 ꢀ 52
6a
6b
6c
6d
6e
6f
3-Cl
4-Cl
4-CH3
4-CH3
Cyclohexylurea
Cyclohexylurea
Cyclohexylurea
1-(Piperidin-1-yl)urea
2.3 ꢀ 0.4
407 ꢀ 271 161 ꢀ 75
4b
4c
4d
4e
4f
0.23 ꢀ 0.06
26 ꢀ 2
19 ꢀ 6
2.7 ꢀ 0.5
15 ꢀ 2
5.9 ꢀ 1.7
2.2 ꢀ 0.3
1.8 ꢀ 0.5
8.1 ꢀ 1.5
3.1 ꢀ 1.4
0.15 ꢀ 0.04 329 ꢀ 71
3.0 ꢀ 1.2
2.6 ꢀ 1.4
1.1 ꢀ 0.4
1.2 ꢀ 0.3
2.5 ꢀ 1.1
317 ꢀ 67
290 ꢀ 197
52 ꢀ 10
4-OCH3 Cyclohexylurea
82 ꢀ 13
34 ꢀ 3
1300 ꢀ 309
526 ꢀ 432
4-OPh
4-CF3
4-F
Cyclohexylurea
Cyclohexylurea
Cyclohexylurea
4g
6g
6h
274 ꢀ 34
159 ꢀ 34
57 ꢀ 41
8.1 ꢀ 1.4
a
Ki and IC50 (mean ꢀ SEM) (n ꢁ 3 independent experiments) were calculated
a
from dose-response curves. hCB1: human CB1 receptor. hCB2: human CB2 receptor.
Ki and IC50 (mean ꢀ SEM) (n ꢁ 3 independent experiments) were calculated
from dose-response curves. hCB1: human CB1 receptor. hCB2: human CB2 receptor.
Following the above results, we decided to conduct the SAR
studies of fragment D on structures with 2,4-dichloro substitution
as fragment B, and cyclohexylurea and 1-(piperidin-1-yl)urea as
fragment C, respectively. The cyclohexylthiourea (5i) was excluded
due to the toxicity of thiourea [19]. As shown in Table 4, different
substituents at the meta-/para-position of the phenyl ring in frag-
ment D did not significantly influence binding affinity, but caused
a decrease in selectivity for hCB1. Only cyclohexylurea derivatives
with para-methyl substitution (6c) showed high potency
(Ki ¼ 0.15 nM) and selectivity (>2000) for hCB1.
2.4 h and the plasma-to-brain concentration ratios were 0.5 and
about 1.0, respectively at 3 and 12 h postdose (Fig. 3B). The plasma-
to-brain concentration ratios of rimonabant (10 mg/kg, p.o.) were
1.6 and near 4.4, respectively at 3 and 12 h postdose (Fig. 3B).
Compared with rimonabant, compound 6c was slowly and lowly
brain-penetrant. In diet-induced obesity (DIO) rats, a single dose of
The inverse agonist property of compound 6c was studied using
a functional cAMP assay in CHO cell expressing the hCB1 receptor.
Because CB1 receptor is a Gi/o-coupled receptor, activation of hCB1 by
CP55940 led to a dose-dependent inhibition of forskolin-stimulated
cAMP increase, whereas the inverse agonist, rimonabant, further
facilitated cAMP production. As shown in Fig. 3A, compound 6c
(EC50 ¼ 0.87 nM) displayed a similar cAMP stimulation pattern as
rimonabant (EC50 ¼ 0.11 nM) under the same assay conditions. Thus,
compound 6c appears to be a potent CB1 inverse agonist.
2.3. In vivo studies
Following a single oral administration of 10 mg/kg in Sprague–
Dawley rats, the plasma half-life of compound 6c is approximately
Table 3
SAR of Fragment C Modification at hCB1 receptor.a
Cpd.
R
hCB1 Ki (nM) hCB2 Ki (nM) hCB1 Ca2þ IC50 (nM)
Fragment C
Cl
Cl
R
N
N
5a
5b
5c
5d
5e
5f
5g
5h
5i
Cyclobutylamide
Cyclopentylamide
Cyclohexylamide
Adamantylamide
Piperidylamide
4-Morpholylamide
n-Butylurea
291 ꢀ 35
732 ꢀ 186
5.8 ꢀ 1.5
5.9 ꢀ 1.7
56 ꢀ 23
2116 ꢀ 273 > 10,000
4885 ꢀ 1911 403 ꢀ 95
2928 ꢀ 1104 9.3 ꢀ 8.5
25 ꢀ 16
288 ꢀ 256
987 ꢀ 206 >10,000
618 ꢀ 251 >10,000
918 ꢀ 353 173 ꢀ 30
1043 ꢀ 495 4.7 ꢀ 2.5
111 ꢀ 59
103 ꢀ 50
1-(Piperidin-1-yl)urea 1.2 ꢀ 0.2
Fig. 3. (A) Whole-cell cAMP assay using CHO cells expressing human CB1 receptor. (B)
Plasma (ng/mL) and brain (ng/g) concentration–time courses of compound 6c and
rimonabant following a single oral administration of 10 mg/kg in Sprague–Dawley rats.
Data represent mean ꢀ SE (n ¼ 5 rats/group).
Cyclohexylthiourea
0.20 ꢀ 0.08
673 ꢀ 94
18 ꢀ 6
a
Ki and IC50 (mean ꢀ SEM) (n ꢁ 3 independent experiments) were calculated
from dose-response curves. hCB1: human CB1 receptor. hCB2: human CB2 receptor.

T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
1137
4.1.1. General method for synthesis of 1-[(2,4-dichloro-
phenyl)(phenyl)methyl] piperazine (3a)
To a solution of 2,4-dichlorobenzaldehyde (10.0 g, 0.57 mol) in
anhydrous THF (100 mL) phenylmagnesium bromide (80 mL, freshly
prepared 0.9 M solution in THF) was added dropwise at room
temperature. After 1 h of stirring the reaction mixture was treated
with saturated NH4Cl aqueous solution (10 mL) and diethyl ether
(300 mL) and organic phase was washed successively with brine and
dried over anhydrous Na2SO4. After evaporation of the solvent under
vacuum the crude product (2,4-dichlorophenyl)(phenyl)methanol
(2a, 12.7 g, 88%) was obtained and used in following reactions
without further purification. 1H NMR (400 MHz, CDCl3)
d
7.59(d, J ¼ 8.4 Hz, 1H), 7.26–7.38(m, 9H), 6.16(d, J ¼ 3.7 Hz, 1H).
Benzhydrol (2a) was dissolve in CH2Cl2 (70 mL). After addition
of SOCl2 (5 mL, 0.69 mol) the reaction mixture was stirred at 40 ꢂC
for 3 h. The solvent was evaporated under vacuum and the residue
was dissolved in MeCN (80 mL). Then, piperazine (24.6 g, 0.29 mol)
was added and the mixture was refluxed for 12 h. The solvent was
removed under vacuum and the residue was dissolved in ethyl
acetate (250 mL) and washed with water (100 mL) followed by 1 N
HCl (100 mL). The acid phase was washed with ethyl acetate
(3 ꢃ 60 mL). The ethyl acetate layer was discarded and the
remaining water layer was neutralized with 3 N NaOH aqueous
solution (40 mL) to pH > 10. The aqueous solution was extracted
with CH2Cl2 (3 ꢃ 80 mL). The combined CH2Cl2 was washed
successively with brine, dried over Na2SO4 and evaporated under
vacuum to provide the title compound 3a as a yellow oil (12.5 g,
Fig. 4. Chronic effects of compound 6c in DIO rats. A significant reduction in body
weight gain was observed on days 12 at 10 mg/kg of compound 6c. *P < 0.05 vs vehicle.
Data represent mean ꢀ SE (n ¼ 11–12 rats/group).
compound 6c at 10 mg/kg (p.o.) suppressed 3 h and overnight
(18 h) food intake by 39% and 22%, respectively. Furthermore, the in
vivo efficacy of compound 6c in body weight reduction was
assessed with the DIO rat model. Chronic application of compound
6c (10 mg/kg, p.o.) once daily for 12 days led to a slower progression
of body weight and a significant reduction in the final body weight
gain on day 12 from starting weight (p < 0.05, Fig. 4).
78%). 1H NMR (400 MHz, CDCl3)
d
7.73(dd, J ¼ 12.5, 8.4 Hz, 1H),
7.35–7.39(m, 2H), 7.17–7.31(m, 5H), 4.72(s, 1H), 3.09–3.12(m, 2H),
2.59(br s, 2H), 2.48(br s, 2H), 2.37(br s, 2H); ESI: m/z (relative
intensity) 320.9(M þ 1, 100%), 235.0.
3. Conclusion
In this study, we described the identification via privileged
structure-based approach of benzhydrylpiperazine moiety as
a potential scaffold to develop novel CB1 receptor modulators.
Efficient structural optimization of the initial hit compounds led to
the discovery of high quality lead series, represented by compound
6c. Compound 6c is a highly potent and selective CB1 inverse
agonist that is able to reduce body weight gain in a DIO rat model.
A study by Song et al. [20] has depicted similar derivatives that
revealed very low CB1 binding activities, however, there were no
cellular function and in vivo information of such structures reported
in this study.
4.1.2. 1-[(4-Chlorophenyl)(2,4-dichlorophenyl)methyl]piperazine
(3i)
The procedure described for the synthesis of 3a was applied to
synthesize the title compound 3i as a yellow oil in 62% total yield.
1H NMR (400 MHz, CDCl3)
6H), 4.71(s, 1H), 2.85–2.88(m, 4H), 2.32–2.47(m, 4H). ESI: m/z
(relative intensity) 354.9(M þ 1, 100%), 269.0.
d
7.71(d, J ¼ 8.4 Hz, 1H), 7.23–7.36(m,
4.1.3. 1-[(2,4-Dichlorophenyl)(p-tolyl)methyl]piperazine (3j)
The procedure described for the synthesis of 3a was applied to
synthesize the title compound 3j as a yellow oil in 64% total yield.
4. Experiment procedure
1H NMR (400 MHz, CDCl3)
d
7.73(dd, J ¼ 13.5, 8.4 Hz, 1H), 7.20–
7.30(m, 4H), 7.05–7.10(m, 2H), 4.69(s, 1H), 3.14–3.17(m, 2H), 2.65(br
s, 2H), 2.52(br s, 2H), 2.38(br s, 2H), 2.28(s, 3H). ESI: m/z (relative
intensity) 334.9(M þ 1, 100%), 249.0.
4.1. Synthesis
All non-aqueous reactions were performed in dried glassware
under an atmosphere of Ar, unless otherwise specified. THF was
freshly distilled from sodium/benzophenone under Ar. All solvents
are of the highest available purity unless otherwise indicated.
Melting points were determined in open capillary tubes on an
electrothermal melting point apparatus. 1H and 13C NMR spectra
were recorded on Varian Mercury-400 spectrometer. Chemical
shifts are reported in ppm from TMS with the solvent resonance as
the internal standard. The following abbreviations were used to
explain the multiplicities: s ¼ singlet, d ¼ doublet, t ¼ triplet,
q ¼ quartet, dd ¼ doublet of doublets, m ¼ multiplet, br ¼ broad.
The LC-MS was carried out on Thermo Finnigan LCQDECAXP. Low-
resolution mass spectra were performed on Finnigan MAT 95 and
Finnigan LCQ Deca spectrometers and high resolution mass spectra
were measured on Finnigan MAT 95 and MicroMass Q-Tof ultimaÔ
mass spectrometers. General synthetic procedures used for the
preparation of the selected target compounds 4a, 5c, 5g–i and 6b–
d are described as follows.
4.1.4. N-Cyclohexyl-4-[(2,4-dichlorophenyl)(phenyl)methyl]-
piperazine-1-carboxamide (4a)
To a solution of compound 3a (0.20 g, 0.62 mmol) in CH2Cl2
(5 mL) cyclohexyl isocyanate (0.78 g, 0.62 mmol) was added. The
reaction mixture was subsequently stirred at room temperature
overnight, and evaporated under vacuum to get the crude product.
The residue was further purified by flash column chromatography,
petroleum ether:EtOAc (10:1 / 2:1) as an eluent to afford the title
compound 4a as a white solid (0.22 g, 79%), mp 165–168 ꢂC. 1H
NMR (400 MHz, CDCl3)
d
7.75(d, J ¼ 8.5 Hz, 1H), 7.38–7.41(m, 2H),
7.18–7.30(m, 5H), 4.75(s, 1H), 4.21(d, J ¼ 7.6 Hz, 1H, NH), 3.63(m,
1H), 3.31(m, 4H), 2.37(m, 4H), 1.93(dd, J ¼ 12.5, 3.1 Hz, 2H), 1.57–
1.72(m, 3H), 1.32–1.39(m, 2H), 1.04–1.16(m, 3H). 13C NMR (CDCl3)
d
157.04, 140.13, 138.49, 134.35, 132.96, 129.68, 129.45, 128.59 ꢃ 2,
128.31 ꢃ 2, 127.54 ꢃ 2, 69.83, 51.39(2 ꢃ CH2), 49.32, 43.80(2 ꢃ CH2),
33.93(2 ꢃ CH2), 25.62(CH2), 25.01(2 ꢃ CH2). HRMS(EI) m/z calcd for
C24H29Cl2N3O, 445.1688; found, 445.1683.

1138
T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
4.1.5. 1-(Cyclohexylcarbonyl)-4-[(2,4-dichloro-
phenyl)(phenyl)methyl]piperazine (5c)
(relative intensity) 462.0(M þ 1, 100%), 235.0; HRMS(EI) m/z calcd
for C24H29Cl2N3S, 461.1459; found, 416.1456.
To a solution of compound 3a (0.20 g, 0.62 mmol) and triethyl-
amine (0.17 mL, 1.3 mmol) in 5 mL dichloromethane, cyclo-
hexanecarbonyl chloride (84 mL, 0.62 mmol) was added with stirring
4.1.9. 4-[(4-Chlorophenyl)(2,4-dichlorophenyl)methyl]-N-
cyclohexylpiperazine-1-carboxamide (6b)
at 0 ꢂC. The reaction mixture was then stirred at room temperature
overnight and poured into water. The mixture was extracted with
CH2Cl2 (2 ꢃ 20 mL). The combined CH2Cl2 was washed with water
and brine, dried over Na2SO4, and evaporated under vacuum. The
residue was further purified by flash column chromatography,
petroleum ether:EtOAc (10:1 / 3:1) as an eluent to afford the title
compound 5c as a white solid (0.23 g, 87%), mp 125–128 ꢂC. 1H NMR
The procedure described for the synthesis of 4a was applied to
3i and cyclohexyl isocyanate providing the title compound 6b as
a white solid in 89% yield, mp 172–175 ꢂC, 1H NMR (400 MHz,
CDCl3)
d
7.70(d, J ¼ 8.2 Hz,1H), 7.27–7.36(m, 6H), 4.73(s,1H), 4.21(d,
J ¼ 7.3 Hz, 1H, NH), 3.63(m, 1H), 3.32(t, J ¼ 5.0 Hz, 4H), 2.36(ddd,
J ¼ 20.2, 11.2, 4.4 Hz, 4H), 1.93(dd, J ¼ 12.5, 3.4 Hz, 2H), 1.57–1.72(m,
3H), 1.26–1.43(m, 2H), 1.01–1.18(m, 3H). 13C NMR (CDCl3)
d 156.99,
(400 MHz, CDCl3)
1H), 3.59(m, 2H), 3.48(m, 2H), 2.38(m, 5H),1.79(m, 2H),1.67–1.71(m,
4H), 1.48–1.52(m, 2H), 1.22(m, 2H). 13C NMR (CDCl3)
174.44, 140.01,
138.37, 134.35, 133.02, 129.63, 129.48, 128.63 ꢃ 2, 128.29 ꢃ 2,
127.57 ꢃ 2, 69.85, 52.23(CH2), 51.65(CH2), 45.40(CH2), 41.53(CH2),
40.36, 29.34(CH2), 29.29(CH2), 25.81(3 ꢃ CH2). HRMS(EI) m/z calcd
for C24H28Cl2N2O, 430.1579; found, 430.1581.
d
7.77(d, J ¼ 8.1 Hz, 1H), 7.22–7.41(m, 7H), 4.75(s,
138.70,137.96,134.28,133.26 ꢃ 2,129.58 ꢃ 2,129.51 ꢃ 2,128.82 ꢃ 2,
127.68, 69.14, 51.35(2 ꢃ CH2), 49.35, 43.76(2 ꢃ CH2), 33.93(2 ꢃ CH2),
25.62(CH2), 25.00(2 ꢃ CH2). HRMS(EI) m/z calcd for C24H28Cl3N3O,
479.1298; found, 479.1296.
d
4.1.10. N-Cyclohexyl-4-[(2,4-dichlorophenyl)(p-tolyl)
methyl]piperazine-1-carboxamide (6c)
The procedure described for the synthesis of 4a was applied to
3j and cyclohexyl isocyanate providing the title compound 6c as
a white solid in 90% yield, mp 143–146 ꢂC. 1H NMR (400 MHz,
4.1.6. N-Butyl-4-[(2,4-dichlorophenyl)(phenyl)methyl]piperazine-
1-carboxamide (5g)
The procedure described for the synthesis of 4a was applied to 3a
and n-butylisocyanate providing the title compound 5g as a white
solid in 86% yield, mp 125–127 ꢂC. 1H NMR (400 MHz, CDCl3)
CDCl3)
d
7.76(d, J ¼ 8.4 Hz, 1H), 7.23–7.29(m, 4H), 7.07–7.09(m, 2H),
4.70(s, 1H), 4.21(d, J ¼ 7.7 Hz, 1H, NH), 3.63(m, 1H), 3.29–3.32(m,
4H), 2.35–2.37(m, 4H), 2.29(s, 3H), 1.91–1.95(m, 2H), 1.66–1.71(m,
2H), 1.59–1.62(m, 2H), 1.31–1.38(m, 2H), 1.03–1.14(m, 2H). 13C NMR
d
7.75(d, J ¼ 9.2 Hz, 1H), 7.24–7.41(m, 7H), 4.75(s, 1H), 3.33(m, 4H),
3.18–3.25(m, 2H), 2.37(m, 4H), 1.43–1.52(m, 2H), 1.29–1.37(m, 2H),
(CDCl3) d 157.04, 138.75, 137.22, 137.07, 134.27, 132.83, 129.57,
0.92(t, J ¼ 7.3 Hz, 3H). 13C NMR (CDCl3)
d
157.78, 140.10, 138.46,
129.43, 129.27 ꢃ 2, 128.22 ꢃ 2, 127.50, 69.58, 51.38(2 ꢃ CH2), 49.31,
43.80(2 ꢃ CH2), 33.93(2 ꢃ CH2), 25.63(CH2), 25.01(2 ꢃ CH2), 21.05.
HRMS(EI) m/z calcd for C25H31Cl2N3O, 459.1844; found, 459.1837.
134.33, 132.95, 129.67, 129.44, 128.59 ꢃ 2, 128.29 ꢃ 2, 127.54 ꢃ 2,
69.83, 51.40(2 ꢃ CH2), 43.80(2 ꢃ CH2), 40.58(CH2), 32.30(CH2),
20.05(CH2), 13.80. HRMS(EI) m/z calcd for C22H27Cl2N3O, 419.1531;
found, 419.1533.
4.1.11. 4-[(2,4-Dichlorophenyl)(p-tolyl)methyl]-N-piperidin-1-
ylpiperazine-1-carboxamide (6d)
4.1.7. 4-[(2,4-Dichlorophenyl)(phenyl)methyl]-N-piperidin-1-
ylpiperazine-1-carboxamide (5h)
The procedure described for the synthesis of 5h was applied to
3j and 1-aminopiperidine providing the title compound 6d as
a white solid in 89% yield, mp 167–168 ꢂC. 1H NMR (400 MHz,
To a solution of 1,10-carbonyldiimidazole(CDI, 0.18 g, 1.1 mmol)
in THF (10 mL) was added a solution of 1-aminopiperidine(0.11 g,
1.1 mmol) in THF (10 mL). The reaction mixture was stirred at
ambient temperature for 1.5 h. After addition of a solution of 3a
(0.30 g, 0.93 mmol) in THF (10 mL) the reaction mixture was heated
to reflux for 16 h and concentrated. The residue was dissolved in
CH2Cl2 (100 mL), washed with water and brine, and dried over
Na2SO4. After evaporation of the solvent under vacuum the residue
was further purified by flash column chromatography, petroleum
ether:EtOAc (2:1 / 1:2) as an eluent to afford the title compound
5h as a white solid (0.38 g, 91%), mp 138–140 ꢂC. 1H NMR (400 MHz,
CDCl3)
d
7.78(d, J ¼ 8.4 Hz, 1H), 7.23–7.30(m, 4H), 7.08–7.10(m, 2H),
4.91(s, 1H, NH), 4.70(s, 1H), 3.43–3.47(m, 4H), 2.64(m, 4H), 2.34–
2.37(m, 4H), 2.29(s, 3H), 1.58–1.66(m, 4H), 1.36(m, 2H). 13C NMR
(CDCl3)
d 158.29, 138.80, 137.18, 137.14, 134.27, 132.81, 129.62,
129.40, 129.26 ꢃ 2, 128.22 ꢃ 2, 127.49, 69.69, 57.60(2 ꢃ CH2),
51.66(2 ꢃ CH2), 44.71(2 ꢃ CH2), 25.71(2 ꢃ CH2), 23.22(CH2), 21.02.
HRMS(EI) m/z calcd for C24H30Cl2N4O, 460.1797; found, 460.1791.
4.2. Biology
CDCl3)
d
7.77(d, J ¼ 8.4 Hz, 1H), 7.38–7.42(m, 2H), 7.19–7.31(m, 5H),
4.2.1. Cell culture and preparation of stably transfected CHO cells
Chinese hamster ovarian (CHO) cells were stably transfected with
the plasmid encoding hCB1 or hCB2 receptors cloned by our labora-
tory. The recombinant CHO-hCB1/hCB2 cells were grown in F12
nutrient medium with 10% FBS, 100 mg/L penicillin, 100 mg/L
streptomycin, and 1 mg/mL G418. After confluence, cells were tryp-
sinized and collected by centrifugation at 800 rpm for 5 min, planted
onto 96-well plates at the density of 20,000 cells per well and
incubated at 37 ꢂC in a humidified atmosphere of 5% CO2 overnight.
4.97(s, 1H, NH), 4.74(s, 1H), 3.44–3.47(m, 4H), 2.65(br s, 4H), 2.35–
2.38(m, 4H), 1.58–1.66(m, 4H), 1.36(m, 2H). 13C NMR (CDCl3)
d
158.28, 140.20, 138.54, 134.34, 132.93, 129.73, 129.43, 128.56 ꢃ 2,
128.30 ꢃ 2, 127.52, 127.47, 69.93, 57.62(2 ꢃ CH2), 51.66(2 ꢃ CH2),
44.71(2 ꢃ CH2), 25.70(2 ꢃ CH2), 23.21(CH2); HRMS(EI) m/z calcd for
C23H28Cl2N4O, 446.1640; found, 446.1644.
4.1.8. N-Cyclohexyl-4-[(2,4-dichlorophenyl)(phenyl)methyl]-
piperazine-1-carbothioamide (5i)
The procedure described for the synthesis of 4a was applied to
3a and cyclohexyl isothiocyanate providing the title compound 5i
as a white solid in 92% yield, mp 166–168 ꢂC. 1H NMR (400 MHz,
CHO cells co-transfected with hCB1 receptors and G
hCB1-G 15/16) were plated onto 96-well plates at a density of
30,000 cells/100 L/well and cultured in same conditions.
a15/16 (CHO-
a
m
CDCl3)
d
7.76(d, J ¼ 8.5 Hz, 1H), 7.38–7.41(m, 2H), 7.19–7.32(m, 5H),
4.2.2. Whole-cell binding assay
4.77(s, 1H), 4.33(m, 1H), 3.69–3.83(m, 4H), 2.42–2.45(m, 4H), 2.05–
The assay was performed using recombinant CHO-hCB1/hCB2
cells prepared as previously described. In briefly, cells were starved
in serum-free F12 medium for 3 h and pretreated with different
concentrations of rimonabant or test compounds for 10 min before
the addition of [3H]-CP-55940. After 3 h incubation at 37 ꢂC, cells
2.11(m, 2H), 1.61–1.73(m, 4H), 1.35–1.46(m, 2H), 1.14(m, 2H). 13C
NMR (CDCl3) d 180.83, 139.84, 138.16, 134.35, 133.10, 129.57, 129.50,
128.64, 128.25 ꢃ 2, 127.63 ꢃ 2, 127.58, 69.62, 54.14, 51.12(2 ꢃ CH2),
47.15(2 ꢃ CH2), 33.07(2 ꢃ CH2), 25.53(CH2), 24.84(2 ꢃ CH2). ESI: m/z

T. Meng et al. / European Journal of Medicinal Chemistry 45 (2010) 1133–1139
1139
were washed 5 times with prewarmed phosphate-buffered saline
(PBS) and lysed by PBS with 1% SDS. The lysed pellets were trans-
ferred into 96-well isoplate (PerkinElmer, Waltham, MA, USA) and
bound radioligand was measured with 1450 microbeta liquid scin-
tillation luminescence counter (PerkinElmer, Waltham, MA, USA).
Assays were performed in triplicate. IC50 values were analyzed by
nonlinear regression analysis performed using the GraphPad Prism
4.0 software (GraphPad Software, San Diego). The Ki values were
calculated by the Cheng–Prusoff equation: Ki ¼ IC50/(1 þ L/Kd).
dosing and caged individually in plastic cages. Compound was
administered orally to DIO rats 1 h before the start of the dark cycle
(4:00 PM) at 10 mg/kg. Vehicle was 0.5% carboxymethyl cellulose in
water, and dosing volume was 2 mL/kg. Preweighed pelleted food
was provided in food cups that were reweighed at 3 and overnight
(18 h) postdose for cumulative food consumed.
For 12-day BW assay, DIO rats (n ¼ 11–12 rats/group) were caged
individually in plastic cages and dosed orally with vehicle or
compound (10 mg/kg) for 12 days. Rats were dosed 1 h before the
dark cycle. FI and BW were measured daily. Statistical analysis was
performed for the FI and BW data using Student’s t-test.
4.2.3. Calcium current assay
CHO-hCB1-Ga15/16 cells were loaded with 2 mM fluo-4 AM in
Hanks balanced salt solution (HBSS, containing 5.4 mM KCl,
0.3 mM Na2HPO4, 0.4 mM KH2PO4, 4.2 mM NaHCO3, 1.3 mM CaCl2,
4.2.6. Pharmacokinetic assays
Male Sprague–Dawley rats were dosed orally at 10 mg/kg after
overnight fasting for pharmacokinetic (PK) evaluations. The blood
samples were collected at various time points into lithium heparin
tubes and centrifuged. The plasma samples were kept at ꢄ20 ꢂC
until analysis. For brain penetration study, rats were administered
orally at 10 mg/kg and sacrificed at 3, 5, 9 and 12 h postdose,
respectively. The whole brains and the blood samples were
collected and stored at ꢄ20 ꢂC. The plasma samples were extracted
by protein precipitation and analyzed by LC/MS/MS. The whole rat
brain was homogenized, and was extracted by protein precipitation
prior to LC/MS/MS analysis.
0.5 mM MgCl2, 0.6 mM MgSO4, 137 mM NaCl, 5.6 mM
250
M sulfinpyrazone, pH 7.4) at 37 ꢂC for 50 min. After removing
the loading buffer, cells were washed with HBSS and incubated with
50 L HBSS containing various concentrations of test compounds or
DMSO (negative control, final concentration 1%). After 10 min incu-
bation at room temperature, 25 L CP55940 were dispensed into the
D-glucose and
m
m
m
well by FlexStation II micro-plate reader (Molecular Devices, Sun-
nyvale, CA, USA), and intracellular calcium change was recorded with
an excitation wavelength of 485 nm and emission wavelength of
525 nm. Assays were performed in triplicate.
4.2.4. Intracellular cAMP assay
Acknowledgment
CHO-hCB1 cells were harvested with 0.04% EDTA (trypsin free)
and diluted to the density of 1 ꢃ106 cells/mL with Stimulation Buf-
fer(1ꢃ HBSS, 5 mM HEPES, 0.1% BSA, 0.5 mM IBMX, pH ¼ 7.4). Cells
were incubated with the Alexa Fluor 647-anti cAMP antibody
(LANCEÔ cAMP 384 kit, PerkinElmer, Waltham, MA, USA), forskolin,
and various concentration of test compounds or agonist CP55940.
After 60 min incubation at room temperature, a detection mix (Eu-
This work was supported by 863 Program (the National High
Technology Research and Development Program of China, No.
2007AA02Z308), by National Science & Technology Major Project
‘‘Key New Drug Creation and Manufacturing Program’’, China (No.
2009ZX09501-010), and grants from the National Natural Science
Foundation of China (30672467, 30873162 and 30623008) and
Shanghai Commission of Science and Technology (08JC1407701).
W8044 labeled streptavidin 22.2 mM, Biotin-cAMP 4.4 mM, LANCEÔ
cAMP 384 kit) was added and maintained for 60 min. The mixture
was transferred into 384 well assay plate (Corning) and intracellular
cAMP levels were measured with Wallac Envision 2101 Multilabel
Reader. When adenylyl cyclase is activated by forskolin, activation of
Gi by CB1R in the presence of agonist (such as CP55940) will lead to
an inhibition of the forskolin-stimulated cAMP increase, and inverse
agonist will lead to a further increase. The maximal CP55940-
mediated inhibition of forskolin-stimulated cAMP increase is defined
as 100% negative efficacy, and the activity of all other compounds is
relative to the efficacy of CP55940. Positive efficacy indicates inverse
agonist property of test compounds.
Appendix. Supplementary information
Supplementary data associated with this article can be found in
the on-line version, at doi:10.1016/j.ejmech.2009.12.018.
References
[1] C. Li, E.S. Ford, A.H. Mokdad, S. Cook, Pediatrics 118 (2006) 1390–1398.
[2] A.H. Mokdad, E.S. Ford, B.A. Bowman, W.H. Dietz, F. Vinicor, V.S. Bales,
J.S. Marks, J. Am. Med. Assoc. 289 (2003) 76–79.
[3] S.C. Woods, Am. J. Med. 120 (2007) 19–25.
[4] G. Colombo, A. Orru, P. Lai, C. Cabras, P. Maccioni, M. Rubio, G.L. Gessa,
M.A. Carai, Mol. Neurobiol. 36 (2007) 102–112.
[5] J.H. Lange, C.G. Kruse, Chem. Rec. 8 (2008) 156–168.
4.2.5. In vivo studies
In vivo pharmacological studies in diet-induced obese (DIO) rats
were carried out using male Sprague–Dawley rats (Shanghai SLAC
Laboratory Animals Co., Shanghai, China). All animal care and
experimental procedures carried out in accordance with guidelines
of the Laboratory Animal Science Center at Shanghai Institute of
Materia Medica. Rats were maintained in a 12 h/12 h light–dark cycle
with free access to food and water in group housing conditions in
a temperature controlled environment(25 ꢂC). To induce obesity, rats
were fed a high fat diet (HFD) containing 1% cholesterol, 0.4% sodium
cholate, 32% saturated fat (lard), and18%casein, with equal quantities
of fiber and minerals as in the rat regular diet (Shanghai SLAC Labo-
ratoryAnimals Co., Shanghai, China) fromweaning. TheHFD supplied
60% of calories as fat and 20% ofcalories as carbohydrate comprised of
cornstarch and sucrose. Food intake (FI) and body weight (BW) were
measured weekly. DIO rats were used at 12–16 weeks of age and
conditioned to oral gavage of water for 5 days before experiments.
For inhibition of feeding experiments, DIO rats were random-
ized into groups (n ¼ 14 rats/group) for compounds and vehicle
[6] I. Svizenska, P. Dubovy, A. Sulcova, Pharmacol. Biochem. Behav. 90 (2008)
501–511.
[7] S. Xie, M.A. Furjanic, J.J. Ferrara, N.R. McAndrew, E.L. Ardino, A. Ngondara,
Y. Bernstein, K.J. Thomas, E. Kim, J.M. Walker, S. Nagar, S.J. Ward, R.B. Raffa, J.
Clin. Pharm. Ther. 32 (2007) 209–231.
[8] R. Christensen, P.K. Kristensen, E.M. Bartels, H. Bliddal, A. Astrup, Lancet 370
(2007) 1706–1713.
[9] D.R. Janero, A. Makriyannis, Expert. Opin. Emerg. Drugs 14 (2009) 43–65.
[10] D. Jones, Nat. Rev. Drug. Discov. 7 (2008) 961–962.
[11] G.G. Muccioli, D.M. Lambert, Curr. Med. Chem. 12 (2005) 1361–1394.
[12] J.H. Lange, C.G. Kruse, Drug. Discov. Today 10 (2005) 693–702.
[13] J. Adam, P.M. Cowley, T. Kiyoi, A.J. Morrison, C.J. Mort, Prog. Med. Chem. 44
(2006) 207–329.
[14] B.E. Evans, K.E. Rittle, M.G. Bock, R.M. DiPardo, R.M. Freidinger, W.L. Whitter,
G.F. Lundell, D.F. Veber, P.S. Anderson, R.S. Chang, et al., J. Med. Chem. 31
(1988) 2235–2246.
[15] A.A. Patchett, R.P. Nargund, Annu. Rep. Med. Chem. 35 (2000) 289–298.
[16] G. Muller, Drug. Discov. Today 8 (2003) 681–691.
[17] D.A. Horton, G.T. Bourne, M.L. Smythe, Chem. Rev. 103 (2003) 893–930.
[18] C.T. King, S.A. Weaver, J.E. Derr, Am. J. Obstet. Gynecol. 93 (1965) 563–565.
[19] R.A. Neal, J. Halpert, Annu. Rev. Pharmacol. Toxicol. 22 (1982) 321–339.
[20] K.S. Song, S.H. Lee, H.J. Chun, J.Y. Kim, M.E. Jung, K. Ahn, S.U. Kim, J. Kim, J. Lee,
Bioorg. Med. Chem. 16 (2008) 4035–4051.