Substituted pyrimidines as cannabinoid CB1 receptor ligands

Article abstract of DOI:10.1016/j.bmcl.2009.06.069

Cannabinoid CB1 receptors have been the avenue of extensive studies since the first clinical results of rimonabant (SR141716) for the treatment of obesity and obesity-related metabolic disorders were reported in 2001. To further evaluate the properties of

Full text of DOI:10.1016/j.bmcl.2009.06.069

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4692–4697
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Substituted pyrimidines as cannabinoid CB1 receptor ligands
Min Ju Kim a, Jong Yup Kim a, Hee Jeong Seo a, Junwon Lee a, Sung-Han Lee a, Mi-Soon Kim a,
Jahyo Kang b, Jeongmin Kim a, Jinhwa Lee a,
*
a Central Research Laboratories, Green Cross Corporation, 303 Bojeong-Dong, Giheung-Gu, Yongin, Gyeonggi-Do 446-770, Republic of Korea
b Department of Chemistry, Sogang University, 1 Shinsu-Dong, Mapo-Gu, Seoul 121-742, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Cannabinoid CB1 receptors have been the avenue of extensive studies since the first clinical results of
rimonabant (SR141716) for the treatment of obesity and obesity-related metabolic disorders were
reported in 2001. To further evaluate the properties of CB receptors, we have designed and efficiently pre-
pared a series of substituted pyrimidines based on chemical structure of Merck’s taranabant, a cannabi-
noid CB1 receptor inverse agonist. Noticeably, N4-((2S,3S)-3-(3-bromophenyl)-4-(4-chlorophenyl)butan-
2-yl)-N6-butylpyrimidine-4,6-diamine (13b) demonstrated good binding affinity and decent selectivity
for CB1 receptor (IC50 = 16.3 nM, CB2/CB1 = 181.6).
Received 9 April 2009
Revised 27 May 2009
Accepted 18 June 2009
Available online 21 June 2009
Keywords:
Rimonabant
Taranabant
Antiobesity
Antagonist
Pyrimidine
Ó 2009 Elsevier Ltd. All rights reserved.
New development of obesity drugs reveals that it is possible to
control appetite and diminish weight by blocking cannabinoid
receptors in the brain, liver or muscle, via cannabinoid (CB1) recep-
tor antagonists or CB1 receptor inverse agonists.1,2 Cannabinoid
CB1 receptor antagonist is designed to hinder the effects of endog-
enous cannabinoids. This type of drug is particularly intriguing
since it not only causes weight loss but also reverses the metabolic
effects of obesity such as insulin resistance and hyperlipidemia.3
Cannabinoid CB2 receptor, the other cannabinoid receptor is clo-
sely related to immune regulation and neurodegeneration.4 There-
fore, the CB2/CB1 selectivity should be taken into consideration for
new drug development of anti-obesity agent.
The first specialized cannabinoid CB1 receptor antagonist, rimo-
nabant was discovered in a high-throughput screening program at
Sanofi-Synthélabo in 1994.5 Several CB1 receptor antagonists
including SR141716 (rimonabant), SLV319 (ibipinabant),6a CP-
945,598 (otenabant)6b,c and MK-0364 (taranabant)7 had been in
various phase of clinical trials.8,9 Unfortunately, some of cannabi-
noid-1 receptor ligands were recently discontinued from all ongo-
ing clinical developments.12
CB1 receptor has been reported.7c Similar to rimonabant, tarana-
bant interacted with a group of aromatic residues (Phe200,
Trp279, Trp356, and Tyr275) of CB1R through the two phenyl rings
and with Phe170 and Leu387 through the CF3-pyr ring. The strong
hydrogen bond formed between the NH of taranabant and the hy-
droxyl of Ser383 was reported to be essential to the superior CB1R
binding affinity of taranabant.7c,11b We envisioned that the key car-
bonyl group of taranabant might be superseded by the correspond-
ing imine-type moiety or ‘imine’-containing heterocycle. Among
many heterocycles involving ‘imine-type’ functionality, we were
particularly intrigued by pyrimidine although to the best of our
knowledge, pyrimidine has been unknown as a bioisostere for
amide.22 We speculated that replacement of the amide moiety into
pyrimidine ring could provide a favorable balance of biological
activity and pharmacokinetics to allow for further evaluation.
Herein, we report the design, chemical synthesis, and biological
evaluation of substituted pyrimidine analogues as our additional
research efforts toward discovery of a novel antiobesity agent
(see Fig. 1).7c
We started our synthesis with a classical Evans asymmetric
reaction route previously developed by our laboratory as shown
in Scheme 1.11a Our objective was to develop a user-friendly, prac-
tical, and efficient chemical synthesis for medicinal chemists to
utilize and to provide a reasonable amount of key intermediates.
We were able to modify and refine some of the previous procedure
to provide the requisite substituted amine 6 in ‘tens of grams’
amount. Thus, 3-bromophenyl acetic acid 1 was coupled with
(S)-4-benzyloxazolidin-2-one via pivaloyl mixed anhydride in
With our efforts to discover and develop a new medicine for the
treatment of obesity,10 we have recently reported a convenient to-
tal synthesis of taranabant,11a and subsequent evaluation of triaz-
olyl analogues of taranabant.11b Recently, a pharmacophore model
for the binding of a low energy conformation of taranabant in the
* Corresponding author. Tel.: +82 31 260 9892; fax: +82 31 260 9020.
E-mail address: jinhwalee@greencross.com (J. Lee).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.069

M. J. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4692–4697
Asp366-Lys192
4693
The synthesis of target compounds is illustrated in Scheme
2.18,19 Reaction of substituted amine 6 with 4,6-dichloropyrimidine
in the presence of DIPEA (N,N-diisopropylethylamine) under
microwave irradiation (160 °C, 30 min) provided chloropyrimidine
7 as a white solid in 91% yield. The key chloropyrimidine 7 thus ob-
tained was useful for preparing a variety of nitrogen, oxygen, and
carbon-linked congeners. In the present case, primary and second-
ary amines react with chloropyrimidine 7 under microwave irradi-
ation (180 °C, 1 h) to give the desired diaminopyrimidines 8
Trp279
O
A
B
O
N
C
Phe170
Leu387
NC
Cl
N
H
CF3
Phe200
Leu360
Ser383
uneventfully. Alternatively, treating chloropyrimidine
7 with
Trp356
in situ generated sodium alkoxide under mildly heated conditions
provided the desired alkoxypyrimidines 9 as shown in Scheme 2.
Meanwhile, as shown in Scheme 3, the iron-catalyzed cross-
coupling reaction14 of alkylmagnesium halides and dichloropyrim-
idine produced chloropyrimidine 11, which was reacted with
substituted amine 6 under microwave irradiation to provide the
carbon-linked target compounds 12.
Figure 1. Taranabant (MK-0364) and its potential receptor–ligand interaction.
O
O
a
b
Br
N
O
CO2H
Br
The target analogues were evaluated in vitro at a rat CB1 bind-
ing assay,15,17and the results are shown in Table 1.20 Unsubstituted
N-propylamine 13a had modest in vitro activity for rat CB1 recep-
tor (IC50 = 48.1 nM). As the size of the carbon chain of amine sub-
stituent on pyrimidine increases, increase in the binding affinity
for rat CB1 receptor is observed up to C-4 chain. Thus, n-butyl
13b improved rat binding affinity for CB1 receptor in threefold,
showing IC50 = 16.3 nM, while the binding affinities for CB1 were
decreased when the alkyl chain became more prolongated (For
example, N-pentylamine 13d, IC50 = 195 nM; N-hexylamine 13g,
IC50 = 342 nM). Branched aliphatic chains displayed moderate
binding affinity, showing IC50 = 93.6–155 nM for 13c, 13e, 13f,
and 13h.
Ph
O
1
2
O
O
OMe
Br
Cl
N
Br
N
c
d
O
Me
Ph
Cl
4
3
Me
NH2
Me
e
HCl
Br
Cl
Br
OH
Me
NH2
Me
N
N
a
HCl
Br
Cl
Br
Cl
N
Cl
Cl
H
6
5
Scheme 1. Reagents and conditions: (a) (i) pivaloyl chloride, Et3N, THF, 0 °C; (ii) n-
BuLi, (S)-4-benzyloxazolidin-2-one, THF, À78 °C, 72% (two steps); (b) 1-(bromo-
methyl)-4-chlorobenzene, NaHMDS, THF, À78 °C, 79%; (c) (i) LiOOH, THF, H2O, 0 °C;
(ii) (COCl)2, CH2Cl2; (iii) (MeO)NHMe, Et3N, CH2Cl2, 79% (three steps); (d) (i)
MeMgBr, THF; (ii) LiBH(sec-Bu)3, THF, À78 °C, 96% (two steps); (e) (i) MsCl, Et3N,
toluene; (ii) NaN3, DMF, 120 °C; (iii) PPh3, toluene–H2O; (iv) HCl–isopropanol, 66%
(four steps).
6
7
Me
N
N
N
N
Me
N
N
NR1R2
b
Br
Cl
N
Br
Cl
N
Cl
H
H
72% yield. Next, alkylation of 2 using NaHMDS with 1-(bromo-
methyl)-4-chlorobenzene provided the alkylated product 3 in
79% yields. Subsequently, the chiral auxiliary of acyloxazolidinone
3 was cleaved by standard conditions to produce the correspond-
ing acid, which was treated with oxalyl chloride provided the cor-
responding acyl chloride. The Weinreb amide 4 was then obtained
by treatment with N,O-dimethylhydroxylamine in the presence of
triethylamine (79%, three steps). Treatment of the Weinreb amide
7
7
8
Me
Br
Cl
N
OR
c
H
9
4 with MeMgBr, followed by hydride reduction with L-Select-
ride7a,11generated the desired alcohol 5 (>99% de, 96% yield, two
steps). Then the next three steps to prepare substituted amine 6
proceeded in a straightforward fashion. Thus, treatment of the
alcohol 5 with mesyl chloride provided the corresponding mesy-
late, which was directly reacted with sodium azide at high temper-
ature (ca. 120 °C) to smoothly produce the corresponding azide
(96%, two steps). Transformation of the azide thus obtained to free
amine was conducted using the Staudinger reaction conditions
(PPh3 in mildly heated toluene/water).7,11,13 Finally, key intermedi-
ate 6 was obtained by acidification of the free amine with HCl in
isopropanol, followed by filtration of the resulting salt form in
75% yield as shown in Scheme 1.
Scheme 2. Reagents and conditions: (a) 4,6-dichloropyrimidine, DIPEA, microwave,
160 °C, 30 min., 91%; (b) R1R2NH, microwave, ꢀ50%; (c) NaOR, ROH, 50 °C, ꢀ50%.
Me
N
N
a
b
N
N
N
N
Br
Cl
N
R
H
Cl
Cl
Cl
R
10
11
12
Scheme 3. Reagents and conditions: (a) RMgCl, Fe(acac)3, THF–NMP, 0 °C to rt, 20%;
(b) 6, THF, microwave, 180 °C, 20%.

4694
M. J. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4692–4697
Table 1
Table 1 (continued)
Binding affinity of substituted acyclic pyrimidines to rCB1 receptora
R1
R2
Compound
rCB1 IC50
69.3
b
R1
N
N
N
Me
13s
O
Br
Cl
N
H
R2
N
N
Me
Me
Me
Me
13t
116
unit: nM
Ref. Rimonabant ( 5.0 1.0c )
13
13u
13v
13w
21.0
94.4
76.5
R1
R2
Compound
rCB1 IC50
b
Me
13a
48.1
16.3
155
195
113
93.6
342
103
24.7
27.6
45.3
70.5
115
HN
N
N
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
13b
13c
13d
13e
13f
HN
HN
Me
Me
N
N
Me
13x
283
Me
HN
HN
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
13y
75.0
73.8
39.8
63.1
136
76.6
203
135
95.3
136
63.2
569
636
285
CF3
13z
HN
HN
13aa
13ab
13ac
13ad
13ae
13af
13ag
13ah
13ai
13aj
13ak
13al
13g
13h
13i
HN
Me
O
O
HN
HN
13j
O
O
13k
13l
HN
HN
O
O
O
13m
HN
HN
Me
Me
13n
13o
67.4
97.8
F
HN
HN
HN
HN
Et
Me
Me
Me
13p
13q
13r
195
264
106
Et
HN
N
N
HN
Et
a
b
c
CB1 receptor was collected from brain tissue of SD rat.
These data were obtained by single determinations.
This data was obtained by multiple determinations.
N
Me

M. J. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4692–4697
4695
Table 2
Table 2 (continued)
Binding affinity of substituted pyrimidines to rCB1 receptora
R1
R2
R3
H
Compound
rCB1 IC50
b
R1
H
14q
319
N
N
HN
HN
Br
Cl
N
H
R2
R3
H
H
H
H
H
H
14r
14s
14t
811
unit: nM
14
ref. Rimonabant (5.0 1.0c)
1540
71.1
R1
R2
R3
H
Compound
rCB1 IC50
b
HN
N
Me
14a
82.0
HN
a
CB1 receptor was collected from brain tissue of SD rat.
These data were obtained by single determinations.
This data was obtained by multiple determinations.
Me
Me
H
H
14b
14c
98.6
22.1
b
c
HN
HN
Preferable in vitro binding affinity displayed for the carbocycle
(N-cyclopentylamine, 13i, IC50 = 24.7 nM) than straight chain (N-
pentylamine, 13d, IC50 = 195 nM) or branched chain (N-isopentyl-
amine, 13e, IC50 = 113 nM or N-(pentan-3-yl)amine, 13f,
IC50 = 93.6 nM) if the same number of carbons are counted. As
the size of carbocycle increases, the binding affinity for CB1 recep-
tor slightly decreases. For example, N-cycloheptylamine, 13k
showed the weaker binding affinity for CB1 receptor
(IC50 = 45.3 nM), whereas N-cyclopentylamine, 13i showed the
good binding affinity for CB1 receptor (IC50 = 24.7 nM). Methy-
lene-linked carbocycle such as N-(cyclohexyl)methylamine, 13l
(IC50 = 70.5 nM) deteriorated the binding affinity for CB1 receptor
in approximately threefold, compared to N-cyclohexylamine, 13j
(IC50 = 27.6 nM). Methylene-linked heterocycle such as N-(tetrahy-
drofuranyl)methylamine, 13m, even further weakened the binding
affinity for CB1 receptor (IC50 = 115 nM). This tendency for CB1
receptor binding affinity is also observed in N-phenylamine series.
Thus, methylene-linked arylamine such as N-benzylamine, 13p
(IC50 = 195 nM) deteriorated the binding affinity for CB1 receptor
in approximately threefold, compared to N-cyclohexylamine, 13n
(IC50 = 67.4 nM). Methylene-linked heteroaryl such as pyridinylm-
ethylamine 13q once again even got worse in terms of CB1 recep-
tor binding affinity (IC50 = 264 nM), indicating the importance of
non-polar moiety in order to optimally bind to a hydrophobic area
of CB1 receptor. N,N-Disubstituted amines such N-ethyl-N-methyl
13r or morpholine 13s appeared to be tolerated for the replace-
ment to a degree. As the size of rings become bigger, the binding
affinities for CB1 receptor were fluctuated. Among the rings with
different size tested on pyrimidine, the best result was obtained
when pyrimidine has simple piperidine 13u. It showed reasonable
binding affinity for rat CB1R (IC50 = 21.0 nM). In order to evaluate
the binding affinity levels of piperiniylpyrimidine analogues for
the CB1 receptor, a structural derivatization of piperidine moiety
was undertaken. 4-Methylpiperidinylpyrimidine 13w decreased
binding affinity for CB1 receptor in 3–4-fold, showing
IC50 = 76.5 nM. Replacement of 4-methyl group with 4-trifluoro-
methyl group in 13y did not affect but simply maintain its CB1
binding affinity (IC50 = 75.0 nM). Disubstitution on piperidine ring
as in 13x even more deteriorated its binding affinity for CB1 recep-
tor (IC50 = 283 nM), suggesting that drug size or shape may be crit-
ical to the interaction of the substituted pyrimidine series of
compounds with the CB1 receptor binding site.
Me
Me
Me
H
H
H
14d
14e
14f
66.5
160
130
HN
HN
CF3
Me
N
Me
N
Me
H
H
H
14g
14h
14i
28.9
69.6
57.3
SMe
SMe
HN
HN
SMe
H
14j
33.0
HN
N
SMe
CF3
CF3
CF3
H
H
H
H
14k
14l
109
43.5
59.3
22.4
HN
HN
14m
14n
HN
CF3
CF3
H
H
14o
14p
38.9
196
HN
N
Simple aliphatic chain or carbocycles (13z, 13aa, 13ab) seemed
to be tolerated for the replacement to a degree (IC50 = 39.8–
73.8 nM), but none appeared more potent than the parent piperi-
dine group. A structurally related series of alkoxypyrimidine or

4696
M. J. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4692–4697
4. (a) Carrier, E. J.; Kearn, C. S.; Barkmeier, A. J.; Breese, N. M.; Yang, W.;
Nithipatikom, K.; Pfister, S. L.; Cambell, W. B.; Hillard, C. J. Mol. Pharmacol. 2004,
65, 999; (b) Klein, T. W.; Newton, C.; Larsen, K.; Lu, L.; Perkins, I.; Nong, L.;
Friedman, H. J. Leukocyte Biol. 2003, 74, 486.
5. (a) Rinaldi-Carmona, M.; Barth, F.; Héaulme, M.; Alonso, R.; Shire, D.; Congy, C.;
Soubrié, P.; Breliére, J.-C.; le Fur, G. Life Sci. 1995, 56, 1941; (b) Rinaldi-Carmona,
M.; Barth, F.; Héaulme, M.; Shire, D.; Calandra, B.; Congy, C.; Martinez, S.;
Maruani, J.; Néliat, G.; Caput, D.; Ferrara, P.; Soubrié, P.; Breliére, J.-C.; le Fur, G.
FEBS Lett. 1994, 350, 240; (c) Sanofi, US005462960, 1995.
6. (a) Lange, J. H.; Coolen, H. K.; van Stuivenberg, H. H.; Dijksman, J. A.;
Herremans, A. H.; Ronken, E.; Keizer, H. G.; Tipker, K.; McCreary, A. C.;
Veerman, W.; Wals, H. C.; Stork, B.; Verveer, P. C.; den Hartog, A. P.; de Jong, N.
M.; Adolfs, T. J.; Hoogendoorn, J.; Kruse, C. G. J. Med. Chem. 2004, 47, 627; (b)
Griffith, D. A.; Hadcock, J. R.; Black, S. C.; Iredale, P. A.; Carpino, P. A.; DaSilva-
Jardine, P.; Day, R.; DiBrino, J.; Dow, R. L.; Landis, M. S.; O’Connor, R. E.; Scott, D.
O. J. Med. Chem. 2009, 52, 234; (c) Pfizer, Inc. US 7,129,239, 2006.
7. (a) Lin, L. S.; Lanza, T. J.; Jewell, J. J. P.; Liu, P.; Shah, S. K.; Qi, H.; Tong, X.; Wnag,
J.; Xu, S. S.; Fong, T. M.; Shen, C.-P.; Lao, J.; Xiao, J. C.; Shearman, L. P.; Stribling,
D. S.; Rosko, K.; Strack, A.; Marsh, D. J.; Feng, Y.; Kumar, S.; Samuel, K.; Yin, W.;
der Ploeg, L. V.; Mills, S. G.; MacCoss, M.; Goulet, M. T.; Hagmann, W. K. J. Med.
Chem. 2006, 49, 7584; (b) Chen, C.-y.; Frey, L. F.; Shultz, S.; Wallace, D. J.;
Marcantonio, K.; Payack, J. F.; Vazquez, E.; Springfield, S. A.; Zhou, G.; Liu, P.;
Kieczykowski, G. R.; Che, A. M.; Phenix, B. D.; Singh, U.; Strine, J.; Izzo, B.; Krska,
S. W. Org. Process Res. Dev. 2007, 11, 616; (c) Lin, L. S.; Ha, S.; Ball, R. G.; Tsou, N.
N.; Castonguay, L. A.; Doss, G. A.; Fong, T. M.; Shen, C.-P.; Xiao, J. C.; Goulet, M.
T.; Hagmann, W. K. J. Med. Chem. 2008, 51, 2108.
8. (a) Barth, F.; Congy, C.; Martinez, S.; Rinaldi, M. WO97/19063, 2000; (b) Barth,
F.; Casellas, P.; Congy, C.; Martinez, S.; Rinaldi, M. C07D/23114, 1995.
9. Hutst, D. P.; Lynch, D. L.; Barnett-Norris, J.; Hyatt, S. M.; Seltzman, H. H.; Zhong,
M.; Song, Z.-H.; Nie, J.; Lewis, D.; Reggio, P. H. Mol. Pharmacol. 2002, 62, 1274.
10. (a) Kang, S. Y.; Lee, S.-H.; Seo, H. J.; Jung, M. E.; Ahn, K.; Kim, J.; Lee, J. Bioorg.
Med. Chem. Lett. 2008, 18, 2385; (b) Lee, S. H.; Seo, H. J.; Lee, S.-H.; Jung, M. E.;
Park, J.-H.; Park, H. J.; Yoo, J.; Yun, H.; Na, J.; Kang, S. Y.; Song, K.-S.; Kim, M.-a.;
Chang, C.-H.; Kim, J.; Lee, J. J. Med. Chem. 2008, 18, 2385; (c) Kim, J. Y.; Seo, H. J.;
Lee, S.-H.; Jung, M. E.; Ahn, K.; Kim, J.; Lee, J. Bioorg. Med. Chem. Lett. 2009, 19,
142; (d) Lee, S. H.; Seo, H. J.; Kim, M. J.; Kang, S. Y.; Song, K.-S.; Lee, S.-H.; Jung,
M. E.; Kim, J.; Lee, J. Bioorg. Med. Chem. Lett. 2009, 19, 1899; (e) Song, K.-S.; Kim,
M. J.; Seo, H. J.; Lee, S.-H.; Jung, M. E.; Kim, S.-U.; Kim, J.; Lee, J. Bioorg. Med.
Chem. 2009, 17, 3080.
11. (a) Kim, M.-a.; Kim, J. Y.; Song, K.-S.; Kim, J.; Lee, J. Tetrahedron 2007, 63,
12845; (b) Kim, M.-a.; Yun, H.; Kwak, H.; Kim, J.; Lee, J. Tetrahedron 2008, 64,
10802.
12. As of November 5, 2008, Sanofi-Aventis, Merck, and Pfizer announced that they
have decided to withdraw their ongoing clinical trials about rimonabant
(SR141,716), taranabant (MK-0364), and otenabant (CP-945,598), respectively
based upon changing regulatory prospects on the risk-benefit profile of the CB1
class and probable new regulatory requirements for approval.
cycloalkyloxypyrimidine derivatives was prepared in an analogous
fashion as shown in Scheme 2. The binding affinity data of these
oxygen-linked pyrimidines demonstrate that replacement of nitro-
gen-linked pyrimidine with oxygen-linked pyrimidine tends to de-
crease binding affinity in 2–5-fold (see Table 1: 13b, 13j, 13l vs
13ad, 13ag, 13ah), with the exception of n-hexyloxypyrimidine
13af. The structure–activity trends change dramatically in the case
of extension of methyl to ethyl on R1 substitution. Thus, ethyl-
substituted aminopyrimidines 13ai, 13aj, 13ak, and 13al provided
an approximately 1 order of magnitude decrease in activity. The
interesting compound such as 13b was further evaluated with
observation of the hCB2 receptor binding affinity. The IC50 value
was measured for the recombinant human CB2 receptor expressed
in CHO cells and employing [3H]WIN-55,212-2 as a radio-ligand.16
The binding affinity of 13b against hCB2 receptor was reasonably
good (IC50 = 2960 nM), thereby showing decent CB2/CB1 selectivity
(CB2/CB1 = 181.60) for the compound.
In order to investigate CB1 receptor binding affinity levels of C2-
or C5-substituted pyrimidine derivatives, a series of pyrimidine
derivatives was prepared in a comparable fashion previously de-
scribed in Scheme 2. The CB1 binding affinity data of these C2-or
C5-substituted pyrimidine derivatives are exhibited in Table 2. 2-
Methylpyrimidines (see 14a–14g) or 2-(methylthio)pyrimidines
(see 14h–14k) showed moderate binding affinities, while 2-(tri-
fluoromethyl)pyrimidines (see 14l–14o) slightly improved binding
affinities for CB1 receptor. The surprisingly poor result for 5-meth-
ylpyrimidines (see 14q–14s) suggested that there might be an
unfavorable steric interaction of the 5-substituted pyrimidine to
hinder good binding to the active site of CB1 receptor.
Up to date, the best results in the substituted pyrimidine series
were obtained when cyclohexylamine was introduced on 2-meth-
ylpyrimidine 14c (IC50 = 22.1 nM) or 2-(trifluoromethyl)pyrimi-
dine 14n (IC50 = 22.4 nM).
In conclusion, we investigated a series of substituted pyrimi-
dine derivatives for their binding affinity for cannabinoid CB1
receptor. Several compounds in this series exhibited reasonably
good CB1 receptor binding affinities, but in comparison with tar-
anabant, the binding affinity of substituted pyrimidines for CB1
receptor was significantly diminished. This reduced affinity might
be attributed to the rigidity of pyrimidine linker which may posi-
tion the C-4 substituent of pyrimidine at somewhat narrow and
deep pocket that can accommodate n-butyl or chair-like conforma-
tion of cyclohexane ring.21 Additional PK and in vivo efficacy stud-
ies in addition to further SAR studies of the substituted pyrimidine
derivatives will be the subject of future investigations.
13. Yokokawa, F.; Asano, T.; Shioiri, T. Org. Lett. 2000, 2, 4169.
14. Fürstner, A.; Leitner, A.; Méndez, M.; Krause, H. J. Am. Chem. Soc. 2002, 124,
13856.
15. Kuster, J. E.; Stevenson, J. I.; Ward, S. J.; D’Ambra, T. E.; Haycock, D. A. J.
Pharmacol. Exp. Ther. 1993, 264, 1352.
16. Murphy, J. W.; Kendall, D. A. Biochem. Pharmacol. 2003, 65, 1623.
17. CB1 and CB2 Receptor Binding Assay. For the CB1 receptor binding studies, rat
cerebellar membranes were prepared aspreviously described by the methods
of Kuster et al.15 MaleSprague-Dawley rats (200–300 g) were sacrificed by
decapitation and the cerebella rapidly removed. The tissue was homogenized
in 30 volumes of TME buffer (50 mM Tris–HCl, 1 mM EDTA, 3 mM MgCl2, pH
7.4) using a Dounce homogenizer. The crude homogenates were immediately
centrifuged (48,000g) for 30 min at 4 °C. The resultant pellets were
resuspended in 30 volumes of TME buffer, and protein concentration was
determined by the method of Bradford and stored at À70 °C until use. For the
CB2 receptor binding studies, CHO K-1 cells were transfected with the human
CB2 receptor as previously described, and cell membranes were prepared as
described above.16 Competitive binding assays were performed as described.
Acknowledgments
Financial support was provided by Green Cross Corporation
(GCC). We thank Drs. Chong-Hwan Chang and Eun Chul Huh for
their many helpful discussions throughout small molecule pro-
grams at GCC. Also we are grateful to Mrs. Jung Ho Kim and
Jung-Won Jeon at GCC Office of R&D planning and coordination
for their supports and services.
Briefly, approximately 10 lg of rat cerebella membranes (containing CB1
receptor) or cell membranes (containing CB2 receptor) were incubated in 96-
well plate with TME buffer containing 0.5% essentially fatty acid free bovine
serum albumin (BSA), 3 nM [3H]WIN55,212-2 (for CB2 receptor, NEN; specific
activity 50–80 Ci/mmol) or 3 nM ([3H]CP55,940, [3H]2-[(1S,2R,5S)-5-hydroxy-
2-(3-hydroxypropyl) cyclohexyl]-5-(2-methyloctan-2-yl)phenol, for CB1
receptor,
concentrations of the synthesized cannabinoid ligands in a final volume of
200 L. The assays were incubated for 1 h at 30 °C and then immediately
NEN;
specific
activity
120–190 Ci/mmol)
and
various
Supplementary data
l
filtered over GF/B glass fiber fiber filter (PerkinElmer Life and Analytical
Sciences, Boston, MA) that had been soaked in 0.1% PEI for 1 h by a cell
harvester (PerkinElmer Life and Analytical Sciences, Boston, MA). Filters were
washed five times with ice-cold TBE buffer containing 0.1% essentially fatty
acid free BSA, followed by oven-dried for 60 min and then placed in 5 mL of
scintillation fluid (Ultima Gold XR; PerkinElmer Life and Analytical Sciences,
Boston, MA), and radioactivity was quantitated by liquid scintillation
spectrometry. In CB1 and CB2 receptor competitive binding assay,
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.06.069.
References and notes
1. Lange, J. H. M.; Kruse, C. G. Drug Discovery Today 2005, 10, 693.
2. Howlett, A. C.; Breivogel, C. S.; Childers, S. R.; Deadwyler, S. A.; Hampson, R. E.;
Porrio, L. J. Neuropharmacology 2004, 47, 345.
3. Lafontan, M.; Piazza, P. V.; Girard, J. Diabetes Metab. 2007, 33, 85. and references
cited therein.
nonspecific binding was assessed using
WIN55,212-2, respectively. Specific binding was defined as the difference
between the binding that occurred in the presence and absence of 1
concentrations of rimonabant or WIN55,212-2 and was 70–80% of the total
1 lM rimonabant and 1 lM
l
M

M. J. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4692–4697
4697
binding. IC50 was determined by nonlinear regression analysis using Graph-
Pad PRISM. All data were collected in triplicate in our experiments.
18. Preparation of chloropyrimidine 7: (2S,3S)-3-(3-bromophenyl)-4-(4-chloro-
phenyl)butan-2-amine HCl (6) (3.00 g, 8.00 mmol) was added to a microwave
reactor tube containing 4,6-dichloropyrimidine (2.38 g, 16.0 mmol) and DIPEA
(4.18 mL, 24.0 mmol) in acetonitrile (24 mL). The capped reactor was placed
in a microwave reactor and the mixture was irradiated at 160 °C for 30 min.
The reaction mixture was evaporated in vacuo, diluted with EtOAc and
extracted three times. Collected organic layer was dried over MgSO4 and
purified by preparative HPLC to obtain 3.27 g (7.26 mmol, 91%) of 7 as a
white solid. 1H NMR (400 MHz, CDCl3) d 8.30 (s, 1H), 7.36 (d, J = 8.00 Hz,
1H), 7.25 (s, 2H), 7.18–7.12 (m, 3H), 6.99–6.94 (m, 3H), 6.18 (br, 1H), 3.11–
3.07 (m, 1H), 3.02–2.99 (m, 1H), 2.94–2.88 (m, 2H), 1.08 (d, J = 6.4 Hz, 3H).
MH+ 450.
and purified by preparative HPLC to obtain 76.4 mg (0.16 mmol, 47%) of 13b as
a yellow solid. 1H NMR (400 MHz, CDCl3) d 7.99 (s, 1H), 7.35 (d, J = 7.6 Hz, 1H),
7.25 (s, 3H), 7.12–7.03 (m, 3H), 7.00–6.96 (m, 2H), 4.94 (br, 1H), 3.9 (br, 1H),
3.26–2.89 (m, 5H), 1.66–1.59 (m, 2H), 1.51–1.39 (m, 2H), 1.10 (d, J = 6.8 Hz,
2H), 0.97 (t, J = 7.2 Hz, 3H). MH+ 487.
20. In this Letter, CB1R data were obtained by single determinations. Whenever we
use rimonabant as our reference in our in-house assay, the CB1R binding
affinity for rimonabant showed
a certain number in the close range
(IC50 = 5.0 1.0 nM) in each different assay (>1500 compounds tested).
Therefore, we believe that all SAR discussions in the manuscript are
scientifically meaningful.
21. As pointed out by a referee, substituted pyrimidines described in this report do
not look like derivatives of acyclic amide taranabant. Rather they might look
more like analogues of rimonabant, having pyrimidine heterocycle at the
center replacing pyrazole.
19. Preparation of the target compound 13b: chloropyrimidine
7 (150 mg,
0.33 mmol) was added to a microwave reactor tube containing n-butylamine
(1 mL). The capped reactor was placed in a microwave reactor and the mixture
was irradiated at 180 °C for 1 h. The reaction mixture was evaporated in vacuo
22. This Letter describes conformationally constrained analogues of taranabant.
Such conformationally constraining approaches have already been described
for the CB1 receptor antagonist rimonabant.