SAR study of bicyclo[4.1.0]heptanes as melanin-concentrating hormone receptor R1 antagonists: Taming hERG

Article abstract of DOI:10.1016/j.bmc.2007.05.068

To improve the ex vivo potency of MCH inhibitor 1a and to address its hERG liability, a structure-activity study was carried out, focusing on three regions of the lead structure. Introduction of new side chains with basic nitrogen improved in vitro and ex

Full text of DOI:10.1016/j.bmc.2007.05.068

Bioorganic & Medicinal Chemistry 15 (2007) 5369–5385
SAR study of bicyclo[4.1.0]heptanes as melanin-concentrating
hormone receptor R1 antagonists: Taming hERG
Jing Su,a,* Brian A. McKittrick,a Haiqun Tang,a Duane A. Burnett,a John W. Clader,a
William J. Greenlee,a Brian E. Hawes,b Kim O’Neill,b Brian Spar,b Blair Weig,b
Timothy Kowalski,b Steve Sorota,c Cheng Lid and Tongtong Liud
aDepartment of Chemical Research, Schering-Plough Research Institute K15 2545, 2015 Galloping Hill Road,
Kenilworth, NJ 07033, USA
bDepartment of Cardiovascular/Metabolic Diseases, Schering-Plough Research Institute K15 3600,
2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
cDepartment of Neurobiology, Schering-Plough Research Institute K15 2600, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
dDepartment of Exploratory Drug Metabolism, Schering-Plough Research Institute K15 3800, 2015 Galloping Hill Road,
Kenilworth, NJ 07033, USA
Received 10 April 2007; revised 23 May 2007; accepted 29 May 2007
Available online 2 June 2007
Abstract—To improve the ex vivo potency of MCH inhibitor 1a and to address its hERG liability, a structure–activity study was
carried out, focusing on three regions of the lead structure. Introduction of new side chains with basic nitrogen improved in vitro
and ex vivo bindings. Many potent compounds with Ki < 10 nM were discovered (compounds 6a–j) and several compounds (14–17)
had excellent ex vivo binding at 6 h and 24 h. Attenuating the basicity of nitrogen on the side chain, and in particular, introduction
of a polar group such as aminomethyl on the distal phenyl ring significantly lowered the hERG activity. Further replacement of the
distal phenyl group with heteroaryl groups in the cyclohexene series provided compounds such as 28l with excellent ex vivo activity
with much reduced hERG liability.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
species examined so far and it appears to be involved
in the regulation of food intake and energy balance
based on the following observations: direct icv adminis-
tration of MCH in rats results in increases in food intake
in a dose dependent manner;3 MCH mRNA is overex-
pressed in ob/ob mice and in fasted mice;4a MCH over-
expressing mice are hyperphagic, mildly obese,
hyperglycemic, and insulin resistant;4b MCH knockout
mice are leaner than wild-type mice.5 It is generally be-
lieved that the MCH receptor R1 mediates the orexi-
genic effects of MCH.6,7 Many reports on MCH
receptor R1 antagonists for the potential treatment of
obesity have appeared in recent years,8,9 including at
least two compounds in clinical trials.10,11 In a recent
publication from Schering-Plough, compound 1a was
identified as a potent orally active MCH R1 antagonist
that did not have the Ames liability of the biphenylani-
line series from which it was derived.12 This compound
had good ex vivo binding but was also a potent blocker
of the hERG channel, as shown by the Rb efflux assay.13
In this paper, we describe our SAR exploration in this
The last decade witnessed a steady rise of the obese pop-
ulation worldwide. This has happened in developed
countries as well as developing countries and the poten-
tial economic impact can be significant. As maintaining
weight loss solely by lifestyle changes remains difficult
and the use of currently available drugs is limited by
their side effects, the search for new drugs for the
treatment of obesity continues to be an area of intense
interest around the globe. In recent years, the melanin-
concentrating hormone (MCH) receptor has emerged
as an important target for obesity.1 MCH, a cyclic 19-
amino acid neuropeptide responsible for color changes
in fish skin,2 is present in the brains of all vertebrate
Keywords: Melanin-concentrating hormone receptor R1 antagonist; Ex
vivo activity; hERG; Parallel synthesis.
*
Corresponding author. Tel.: +1 908 740 7489; fax: +1 908 740
7164; e-mail: jing.su@spcorp.com
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.05.068

5370
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 1. MCH R1 binding data for compounds 1a–p
series and our discovery of several compounds with
improved ex vivo binding and significantly diminished
hERG liability.
Compound
R
MCH R1 Ki (nM)
1b
1c
1d
1e
1f
1g
1h
1i
Ph
2-FPh
132 17
1884 232
3-FPh
3-OMePh
4-FPh
20
2
2. Chemistry and results
373 55
54 14
3,5-F2Ph
3,4-F2Ph
3-F, 4-MePh
3,4-OCH2OPh
3,5-Me2Ph
Cyclohexyl
15
20
48
2
5
8
Our SAR investigation on 1a (Fig. 1) covered three re-
gions of the molecule: the urea moiety, the side chain
with basic nitrogen, and the distal phenyl ring. As
shown in Scheme 1, a parallel synthesis approach was
used to first modify the urea moiety. Compound 3 was
synthesized from 2b12 via reductive amination. Subse-
quent treatment with an aryl isocyanate followed by
resin cleanup afforded compounds 1a–p.14 A urea li-
brary of 1 · 70 dimensions was synthesized using this
protocol and the SAR results are represented in Tables
1 and 3.15 These data clearly suggested the following
trends: (a) N-aryl ureas were more active than N-alkyl
ones (compound 1b vs 1l);15 (b) for mono-substitution
on N-phenyl ring, 3- substitution was more active than
2- and 4-substitution (compound 1d vs 1c, 1f), and elec-
tron withdrawing groups were more active than electron
donating ones (compound 1d vs 1e); (c) disubstitution,
particularly 3,4- and 3,5-disubstitution with electron
withdrawing groups, was as active as the 3-substituted
ones (compounds 1g and 1h). Other modifications in this
region included replacing the urea moiety by amide, ter-
tiary amine, and sulfonamide groups. All these modifi-
cations led to the loss of MCH R1 activity.16
1j
134 20
1k
1l
539
2213
6
6
Scheme 2 outlines the synthesis of compounds for SAR
exploration of the basic nitrogen side chain. Ketone 2b
was reacted with 1-hydroxyethylamine followed by
Boc protection of the nitrogen atom and subsequent
Dess–Martin oxidation to afford aldehyde 4. N-Boc
deprotection and urea formation provided the desired
compounds represented in Table 2 (1 · 24 dimensions).
These data indicate that a variety of amines were well
tolerated for MCH R1 binding, including cyclic amines
and non-cyclic amines. Among the cyclic amines, pyrrol-
idine, piperidine, and homopiperidine with substituents
at various positions were essentially equally potent.
The acyclic amine 6a showed Ki 5 nM in the human
receptor binding assay. In an effort to quickly determine
the ability of a compound to reach the mouse brain and
bind to the MCH R1 receptor at levels high enough to
cause an effect, a mouse ex vivo assay was developed
in house to directly measure MCH R1 receptor occu-
pancy in brain sections of treated mice.17 This assay
obviates the need to radiolabel each individual com-
pound and allows one to evaluate the correlation be-
tween the extent and duration of receptor occupancy
with in vivo efficacy. Compound 6a was found to be
more potent in the mouse ex vivo assay (83% inhibition
at 6 h, 30 mg/kg, po, Table 5) than compound 1a (49%
inhibition at 6 h). Compound 5d also had strong binding
in the ex vivo study (78% inhibition at 6 h).
Cl
F
NH
OH
N
O
N
The encouraging results of 6a and 5d prompted us to fur-
ther explore the development of SAR around these leads.
Based on their structural similarity, for simplicity of
synthesis, we fixed the side chain as the one in 6a and
reinvestigated SAR in the urea region. A 1 · 65 urea
library was assembled according to Scheme 3 and Table 3
CN
1a
Figure 1.
R
NH
OH
OH
N
O
N
HN
O
N
OH
1. RNCO
2. PS-NH2
N
H2N
3. PS-TsOH
4. NH3/MeOH
70-80%
Ti(OiPr)4, NaBH4
65%
CN
CN
CN
1a-p
2b
3
Scheme 1. Parallel synthesis of 1 with various ureas.

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5371
O
O
BocN
Ti(OiPr)4
NaBH4
1. (Boc)2O
2. Dess-Martin
40%
HNRR′
NaBH(OAc)3
40-50%
OH
H2N
48%
CN
CN
2b
4
Cl
F
O
NRR'
N
N
H
TFA
Ar-NCO
50-60%
CN
5
Scheme 2. Synthesis of compound 5.
Table 2. MCH R1 binding data for compounds 5a–j and 6a
shows some results with comparison to analog 1 which
had (R)-3-hydroxyprrolidine on the side chain. These
data confirmed our previous SAR observation that elec-
tron withdrawing groups at 3- and 4-positions were fa-
vored for MCH R1 activity. Many potent compounds
with Ki < 10 nM were discovered in this library. In addi-
tion, it was found that the series of compounds 6a–j
were uniformly more active than their corresponding
analogs from series 1. These observations demonstrated
the significance of N,N-diisopropyl group on the side
chain and it was therefore decided to keep this basic
nitrogen moiety unchanged while exploring SAR in
the distal phenyl region.
Compound
NRR0
MCH R1 Ki (nM)
N
5a
16
23
15
2
6
3
5b
5c
NHAc
N
N
OH
N
N
5d
5e
5f
7
15
15
22
0.9
3
OH
OH
The synthetic route to compounds with different substit-
uents on the distal phenyl ring is shown in Scheme 4.
Based on the previous observation that a 3-CN group
was critical for MCH R1 binding,12 our investigation
was concentrated on substitution at the 3-position.
Indeed, Table 4 shows that a simple hydrogen at 3-posi-
tion (compound 8a) led to a 30-fold drop in MCH R1
binding. Analogs with methyl ester 8e or alcohol 8f were
also much less potent. However, replacement of 3-CN
with 3-Br (8c) or oxime (8g) provided compounds with
good affinity for the MCH R1 receptor (Ki < 10 nM).
In the mouse ex vivo study, the oxime 8g showed 62%
inhibition at 6 h, demonstrating for the first time in this
series that a 3-CN group could be replaced while main-
taining MCH R1 ex vivo activity.
2
N
N
OH
5g
2
O
5h
5i
53 10
N
N
15
11
1
2
Some ex vivo data of compounds with further modifica-
tion on the side chain are shown in Table 5. As discussed
before, compound 6a showed 83% binding at 6 h.
Replacing one of the N-isopropyl groups with various
hydroxyalkyl groups did not affect their Ki’s (com-
pounds 9–11). The use of N-fluoroethyl on the side chain
resulted in similar binding (compound 12). It was
observed, however, that the use of 2,6-dichloropyridyl
as an aryl group in the urea region consistently
OH
5j
N
N
6a
5
1.6

5372
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Ar
NH
O
Ti(OiPr)4
NaBH4
N
O
N
ArNCO
45-50%
N
H2N
CN
CN
2b
6
Scheme 3. Parallel synthesis of compound 6.
Table 3. Comparison of MCH binding between 1a–p and 6a–j
Ar
Table 4. MCH R1 binding data for compounds 8a–h
Compound
R
Ar
MCH Ki (nM)
92
Ar
NH
NH
8a
6h
8c
8d
8e
8f
H
CN
3,4-F2Ph
3,4-F2Ph
5
N
N
OH
O
N
O
N
2.8 1.2
7.8 3.1
142 12
Br
3,4-F2Ph
4-F, 3-ClPh
3,4-F2Ph
CONH2
CO2Me
CH2OH
CH@NOMe
CH@NOEt
34
48
2
7
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
8g
8h
4.1 0.6
29
CN
CN
5
Ar
Compound MCH R1 Compound MCH R1
Ki (nM) Ki (nM)
Compounds 17 and 18, two of the possible metabolites
of 16, were also synthesized and both showed strong
binding at 6 h. Additionally, the use of 2,5-disubstituted
pyrrolidine such as compound 19 led to a drop in the
ex vivo binding compared with prolinol 5d. Once again,
the preferred 2,6-dichloropyridyl urea moiety restored
the ex vivo binding back to 82% at 6 h (compound 20).
4-F, 3-ClPh
3-NO2Ph
3-BrPh
6a
6b
6c
6d
6e
6f
5
1.6 1a
9
60
2
2
2.7 0.3 1m
5.7 1.9 1n
2.1 0.7 1d
34 0.5
3-FPh
3-ClPh
20
27
2
3
6
2.2 1o
3-CNPh
3,5-F2Ph
3,4-F2Ph
4-FPh
7.6 1.6 1p
4.5 1.1 1g
2.8 1.2 1h
10.4 3.8 1f
92 10
6g
6h
6i
15
20
2
5
Although we were able to improve the ex vivo activities
of these compounds, further studies and progression of
them were not pursued due to their strong inhibition
of hERG K+ channel since blocking of the hERG K+
channel could lead to prolonged QTc and potentially
sudden cardiac death.19 In the above SAR study it
showed that in bicyclo[4.1.0]heptane series, the urea,
the side chain, and the bicyclo[4.1.0]heptane moieties
were essential for the MCH R1 antagonist activity.
While deleting any one of these three moieties of this
series could get rid of the hERG liability, it also depleted
54 14
2,6-Cl2pyridyl 6j
2
0.3
improved the ex vivo binding significantly. In one case
(compound 15 vs 11), the binding at 6 h was improved
from 75% to 95%. Similar results were observed for
compound 13 versus 6a, 14 versus 9, and 16 versus 12.
Among all the compounds tested, 16 showed almost full
receptor occupancy at 24 h (99% at 6 h, 97% at 24 h).18
Ar
NH
N
O
O
N
O
O
see Scheme 3
1. KOH for 2d-e
2. MeI for 2e
3. TFA
X
X
Y
7a X = H
7b X = CN
7c X = Br
8
2a X = H
2b X = CN
2c X = Br
2d X = CONH2
2e X = CO2Me
Scheme 4. Synthesis of 8 with various substituents on the phenyl ring.

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 5. MCH in vitro, ex vivo and hERG data for 6a analogs
5373
R1
NH
R2
O
N
R3
Compound
R1
R2
R3
MCH Ki (nM)
Ex vivo % inh. at
6 h (30 mpk, po)
Ex vivo %
inh. at 24 h
Rb % inh.
at 5 lg/mL
1a
4-F, 3-ClPh
CN
9.2 2.8
49
2
NA
NA
80
N
OH
6a
4-F, 3-ClPh
CN
CN
5.0 1.6
83
4
78
N
N
5d
8g
4-F, 3-ClPh
4-F, 3-ClPh
7.0 0.9
4.1 0.6
78
62
7
4
NA
NA
75
76
OH
CHNOMe
N
N
OH
9
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
CN
8.7 0.6
9.4 1.3
5.2 1.4
73
73
75
81
99
99
2
3
2
5
4
3
38
25
2
2
2
5
4
88
NA
82
R
OH
OH
F
10
11
12
13
14
CN
N
S
CN
35
N
CN
12
1
66
50
N
N
N
Cl
CN
2.4 0.5
1.6 0.4
58
59
N
Cl
Cl
OH
CN
NA
NA
N
Cl
Cl
S
OH
F
15
16
CN
CN
1.6 0.2
99
99
3
1
81
97
3
1
60
50
N
N
Cl
Cl
10
3
N
N
Cl
(continued on next page)

5374
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 5 (continued)
Compound
R1
R2
R3
MCH Ki (nM)
Ex vivo % inh. at
6 h (30 mpk, po)
Ex vivo %
inh. at 24 h
Rb % inh.
at 5 lg/mL
Cl
Cl
H
N
17
18
19
20
CN
1.9 0.2
93
89
31
2
2
2
70
32
2
2
2
2
NA
84
N
N
Cl
Cl
F
CN
CN
CN
4.4 0.7
NH
N
4-F, 3-ClPh
11
2
78
OH
Cl
1.2 0.1
82 11
41 11
60
N
N
OH
Cl
Ar
O
and 22k. Elongation of the side chain (compound 22i)
or replacing the piperidine ring with pyrrolidine (com-
pound 22l) resulted in 10-fold loss of MCH R1 binding.
NH
O
Ti(OiPr)4
N
m
NaBH4
ArNCO
NBoc
n
90-95%
Another strategy we pursued was to increase the polar-
ity of our lead.20 We had shown in Table 4 that replac-
ing the 3-CN group on the distal phenyl maintained
MCH R1 activity. However, all compounds in that table
showed strong inhibition in the Rb efflux assay. The use
of polar groups such as 3-hydroxymethyl (compound 8f,
Table 7) did slightly lower the hERG activity but its
MCH R1 binding was also weaker. When primary
amine 8i or dimethylamine 8o were introduced, they re-
tained their MCH R1 activity. Importantly, they consis-
tently showed much weaker hERG activity.
50%
CN
CN
2b
21
Ar
NH
N
O
R
TFA
The combination of the above two strategies created
compounds shown in Table 8. Indeed the hERG activity
was further reduced to an acceptable level. To boost the
ex vivo binding, 2,6-dichloropyridyl was introduced as
the urea moiety. However, even though their MCH R1
bindings were excellent, their ex vivo bindings were dis-
appointing. Selected compounds were tested in the
CaCO-2 bidirectional assay and the results indicated
that they were pgp substrates.21
CN
22
Scheme 5. Synthesis of compound 22.
their MCH R1 activity (data not shown). Therefore, we
decided to keep the skeleton of the series unchanged
while exploring other peripheral changes to reduce the
inhibition of hERG channel. One of our strategies was
to decrease the basicity of the side-chain nitrogen. As
shown in Scheme 5, a piperidine unit was installed in
the side chain and several analogs were synthesized.
The data (Table 6) suggested that introduction of urea
and amide moieties at the piperidine nitrogen site could
lead to decreased inhibition in the Rb efflux assay as well
as decreased MCH R1 activity (compounds 22b and
22c). An exception was observed for compounds 22d
and 22e, which had a methanesulfonamide moiety and
maintained high MCH R1 affinity. Elongation to ethane-
sulfonamide 22f led to a drop in MCH R1 activity (Ki
21 nM) and an increase in hERG activity. Other
methanesulfonamides (compounds 22g and 22h) were
also synthesized and demonstrated the similar trend.
The position of the nitrogen on the piperidine was critical
for MCH R1 binding as demonstrated by compounds 22j
Based on the observation that introduction of a polar
group on the distal phenyl ring could significantly
reduce hERG inhibition (compound 8i versus 6a), we
decided to explore this region of the lead by replacing
the distal phenyl group with more polar heterocycles,
hoping that such modification could optimize the overall
profile of the lead structure. To quickly get some SAR
information, we decided to explore cyclohexene analogs
28 since Xu et al. demonstrated that aryl cyclohexenes
closely mimicked bicyclo[4.1.0]heptanes in terms of
MCH R1 activity.12 Scheme 6 outlines the synthesis of
these analogs. After reductive amination of 24 and
Boc protection, the ketal group was cleaved using cer-
ium chloride and a mixture of 25a/25b was obtained.22
After conversion of this mixture to the enol triflate
26a/26b, compound 26b was isolated and further

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 6. MCH R1 in vitro, hERG data for compound 22
5375
Compound
Ar
R
MCH Ki (nM)
Rb % inh. at 5 lg/mL
Rb % inh. at 1.5 lg/mL
N
22a
4-F, 3-ClPh
8.7 0.8
91
75
14
32
6
O
H
N
N
22b
22c
22d
22e
22f
22g
22h
22i
4-F, 3-CF3Ph
4-F, 3-ClPh
4-F, 3-ClPh
3,4-F2Ph
181
2
18
37
24
34
59
ꢀ1
47
20
9
O
N
312 29
O
S
7.2
1
N
N
N
N
N
O2
5.4 1.1
15
25
ꢀ1
25
18
3
S
O2
3,4-F2Ph
21
26
2
4
S
O2
4-F, 3-CF3Ph
3-FPh
S
O2
9.7 1.5
S
O2
4-F, 3-CF3Ph
4-F, 3-ClPh
80
7
4
N
S
O2
O2
S
22j
211
N
O2S
N
22k
22l
4-F, 3-ClPh
4-F, 3-ClPh
295 49
559 20
3
31
32
O2S
N
13
converted to urea 27. The Suzuki coupling reaction pro-
vided the desired products 28 in good to excellent yields.
Among 48 analogs that were synthesized, the furans 28d
and 28e and thiophenes 28g–i were very promising, espe-
cially 28i, which showed excellent Ki and ex vivo bind-
ing, yet strong hERG activity. The reduction of the
cyano group led to the primary amine 28j with excellent
Ki, a bit lower yet still significant ex vivo binding but
much lower hERG activity. Based on these encouraging
results, we further synthesized the thiazole derivatives
28k–n to increase the polarity of this region. We were
gratified to find that while 28n had poor ex vivo binding
at 6 h, its regioisomer 28l did exhibit excellent ex vivo
binding and much lower hERG activity. The strong
ex vivo binding (75% at 24 h) came as a surprise since
these compounds possess two basic nitrogen atoms.23
This result indicated that it is possible for a molecule
with two basic nitrogen atoms to cross the blood–brain
barrier (BBB) at physiological pH. Overall, these results
in Table 9 confirmed that replacing the distal phenyl
ring with heterocycles did maintain MCH R1 activity
and that increasing the polarity of the molecule in
specific regions could decrease hERG activity
significantly.
3. Summary
In summary, a systematic SAR study on three regions of
lead 1 was carried out. Once the N,N-diisopropylethyel-
endiamine was discovered as an important side chain for
MCH R1 activity, a re-examination of SAR in the urea

5376
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 7. MCH R1 in vitro, hERG data for compound 8
Ar
NH
N
O
N
R
Compound
Ar
R
MCH Ki (nM)
Rb % inh. at 5 lg/mL
Rb % inh. at 1.5 lg/mL
6a
8e
8f
8d
8i
4-F, 3-ClPh
3,4-F2Ph
CN
CO2Me
5.0 1.6
78
85
67
44
32
46
78
44
57
43
21
22
17
62
34
48
2
7
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
CH2OH
CONH2
142 12
2.3 0.5
CH2NH2
CH2NHSO2Me
CH2NHCOMe
8j
8k
18
31 0.3
3
N
8l
4-F, 3-ClPh
4-F, 3-ClPh
21
54
2
1
ꢀ1
ꢀ4
OH
N
8m
13
10
N
8n
8o
4-F, 3-ClPh
4-F, 3-ClPh
CH2NEt2
CH2NMe2
46
13
2
3
30
5
15
ꢀ1
Table 8. MCH R1 in vitro, ex vivo and hERG data for compounds 23
Ar
NH
O
N
N
SO2Me
R
Compound
Ar
R
MCH Ki (nM)
Rb % inh. at 5 lg/mL
Rb % inh. at 1.5 lg/mL
Ex vivo % at 6 h
23a
23b
4-F, 3-ClPh
4-F, 3-ClPh
CH2NH2
CH2NMe2
9.9 1.1
15
14
8
8
7
0
26
5
5
1
Cl
N
23c
23d
CH2NH2
2.4 0.2
3.3 0.6
3
7
ꢀ1
0
0
4
5
Cl
Cl
CH2NMe2
8
N
Cl
Cl
N
23e
22d
8.9 0.4
0
ꢀ7
1
0
3
8
N
OH
Cl
4-F, 3-ClPh
CN
7.2
1
24
6

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5377
N
O
RN
1. amine, Ti(OiPr)4
Tf2NPh
CeCl3
NaBH4
2. Boc2O
O
O
O
24
25a R = Boc
25b R = H
O
N
O
N
N
RN
N
Ar
Ar
N
H
N
H
N
ARNCO
R'B(OH)2
OTf
OTf
R'
26a R = Boc
26b R = H
27
28
Scheme 6. Synthesis of compound 28.
region was done and the results not only confirmed pre-
vious SAR but also led to many new compounds with
Ki < 10 nM. A SAR study on the distal phenyl ring
revealed that the 3-CN group could be replaced by other
groups such as bromo or oxime. Further modification
on the side chain resulted in compounds such as 16 with
excellent ex vivo binding. Several different strategies
were utilized to tackle the hERG issue of 1a and other
analogs. Our efforts to adjust the basicity of nitrogen
on the side chain and to increase molecular polarity
on the distal phenyl ring led to decreased hERG activity
in many cases, with both compounds 8i and 8o showing
excellent Ki’s with a much improved hERG profile.
Finally, these results prompted us to explore replace-
ment of the distal phenyl ring with heterocycles and
ultimately led to the discovery of new derivatives in
the cyclohexene series such as 28j and 28l with reduced
hERG liability.
(25 mM HEPES, 10 mM MgCl2, 5 mM MnCl2, and
0.1% BSA, pH 7.4) for 5 min on ice forming a bead/
membrane mixture. The bead/membrane mixture was
centrifuged (4 min at 300g) and resuspended in binding
buffer. The bead/membrane mixture was then pelleted
again (4 min at 300g), resuspended in binding buffer,
and set aside. Binding buffer (50 lL/well) containing
vehicle alone (2% DMSO), various compound concen-
trations, or 4 lM MCH (for non-specific binding) was
added to a 96-well plate. Subsequently, 50 lL of binding
buffer containing 0.5 nM 125I-MCH was added to each
well of the 96-well plate. Finally, 100 lL of the bead/
membrane mixture was added to each well of the 96-well
plate. The binding reaction mixture were incubated for
2–4 h at room temperature. Binding of 125I-MCH to
the bead/membrane mixture was detected using a TOP-
COUNT (Packard). Ki values were determined using
non-linear regression analysis. They are as mean
values the range from two determinations, each
performed in quadruplicate.
4. Experimental
4.2.2. MCH R1 ex vivo assay. Animals were adminis-
tered compounds via oral gavage and the brains were
harvested and frozen at the indicated time point after
dosing. Frontal sections from the caudate were placed
on slides and incubated with radiolabeled [125I]-S36057
(NEN) for 30 min. The sections were rinsed with bind-
ing buffer and dried, and radioactivity bound to the sec-
tion was quantified using a Storm 860 phosphorimager.
4.1. General methods
All reagents were used as received. H and 13C NMR
1
spectra were obtained on a Varian XL-400 (400 MHz)
instrument and are reported as ppm downfield from
Me4Si. LCMS analysis was performed on an Applied
Biosystems API-100 mass spectrometer and Shimadzu
SCL-10A LC column: Altech platinum C18, 3 micron,
33 mm · 7 mm ID; gradient flow: 0 min—10% CH3CN,
5 min—95% CH3CN, 7 min—95% CH3CN, 7.5 min—
10% CH3CN, 9 min—stop. Chromatography was per-
formed with Selecto Scientific flash silica gel, 32–63 lM.
[
125I]-S36057 binding to the brain sections was specific to
MCH binding sites, as addition of non-labeled MCH to
the reaction completely abolished radiolabel binding.24
Three mice were used for each time point for each
compound.
4.2. Biological assays
4.2.3. Rubidium efflux assay. hERG-CHO cells were pla-
ted into 96-well flat-bottomed dishes and returned to
the incubator for 24 h. On the day of study, culture med-
ium was removed and replaced by 150 lL of a HEPES-
buffered saline solution containing 5.4 mM RbCl
(rubidium loading buffer), and the cells were then re-
turned to the tissue culture incubator for 3 h to permit
4.2.1. MCH R1 receptor binding assay. The Ki values of
MCH receptor R1 antagonists were determined using a
SPA-based radioligand binding assay. Membranes from
CHO cells expressing MCH R1 (0.1 mg/mL) were incu-
bated with SPA beads (1 mg/mL) in binding buffer

5378
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
Table 9. MCH R1 in vitro, ex vivo and hERG data for compound 28
Compound
Ar
R0
MCH Ki
(nM)
Rb % inh.
at 5 lg/mL
Rb % inh.
at 1.5 lg/mL
Ex vivo % inh.
at 6 h
Ex vivo % inh.
at 24 h
28a
4-F, 3-ClPh
147 16
85
N
N
28b
28c
28d
28e
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
30
1
16
3
N
1969 128
117 10
51 15
NH
O
O
47
CHO
O
28f
4-F, 3-ClPh
316 0.6
S
28g
28h
4-F, 3-ClPh
4-F, 3-ClPh
85 0.2
12
43
S
31
7
59
1
3
1
O
S
28i
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4-F, 3-ClPh
4.2 0.1
8.2 0.4
57 0.5
74
26
84
77
4
2
97
57
4
2
CN
S
S
28j
NH2
28k
28l
N
N
CN
S
13
2
5
10
88
1
75
1
NH2
N
S
28m
28n
648
N
N
S
17 0.5
14
37
3
NH2
rubidium-potassium exchange. Test compounds were
prepared at a 4-fold final concentration in rubidium
loading buffer containing 10% DMSO. Individual com-
pounds were dispensed (50 lL) into wells on the cell
plate, returned to the incubator, and allowed to equili-
brate for 30 min at 37 ꢁC. Cell plates were then washed
three times with HEPES-buffered saline containing
5.4 mM KCl but no rubidium. After the final wash, cells
were depolarized by the addition of 200 lL of HEPES-
buffered saline containing 45.4 mM KCl. A 5-min depo-
larization was found to be adequate for efficient efflux of
rubidium. Supernatants were collected and analyzed for

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5379
rubidium content using automated flame atomic absor-
bance spectroscopy (ICR-8000 spectrometer, Aurora
Biosciences, Vancouver, BC, Canada). Percent inhibi-
tion was quantified based on a single window defined
by vehicle (no block) and dofetilide (10 lM, full block)
reference wells. Liquid handling for the rubidium efflux
screen was implemented using a pipetting robot capable
of making simultaneous additions and removals from 96
wells (Quadra 96, Tomtec, Hamden, CT).
MeOH for 1 h twice and dried. For characterization of
1a, see Ref. 12.
Compound 1b: MS for (MH+) C27H33N4O2: 445.
Compound 1c: MS for (MH+) C27H32FN4O2: 463.
Compound 1d: MS for (MH+) C27H32FN4O2: 463.
Compound 1e: MS for (MH+) C28H35N4O3: 475.
Compound 1f: MS for (MH+) C27H32FN4O2: 463.
Compound 1g: MS for (MH+) C27H31F2N4O2: 481.
Compound 1h: MS for (MH+) C27H31F2N4O2: 481.
Compound 1i: MS for (MH+) C28H34FN4O2: 477.
Compound 1j: MS for (MH+) C28H33N4O4: 489.
Compound 1k: MS for (MH+) C29H37N4O2: 473.
Compound 1l: MS for (MH+) C27H39N4O2: 451.
Compound 1m: MS for (MH+) C27H32N5O4: 490.
Compound 1n: MS for (MH+) C27H32BrN4O2: 523, 525.
Compound 1o: MS for (MH+) C27H32ClN4O2: 479.
Compound 1p: MS for (MH+) C29H32N5O2: 470.
4.3. General procedures for chemical synthesis
4.3.1. Synthesis of 3. 5.76 g of (R)-3-hydroxypyrrolidine
(66.2 mmol, 1 equiv) was treated with iodoacetonitrile
(11.50 g, 1.04 equiv), K2CO3 (27.60 g, 3 equiv) in
160 mL DCM and the mixture was stirred at rt for
overnight. The crude was filtered through Celite and
after removal of solvent, 9.7 g of brown oil was ob-
tained. Part of this oil (8.5 g) was treated with LiAlH4
(71 mL, 1 N in ethyl ether) in 100 mL THF at ꢀ30 ꢁC
under nitrogen. After gradually warming up to rt, the
reaction was refluxed at 75 ꢁC for 2.5 h. After cooling
to rt, 2.7 mL of water, 8.1 mL of 15% NaOH, and
2.7 mL of water were added in sequence and the mixture
was stirred for 30 min. The crude was filtered through
Celite and washed with DCM, ether, and THF. After re-
moval of solvent, 6.9 g of 2-(3R-hydroxypyrrolidin-1-
yl)-ethylamine as yellow oil was obtained in 91% yield
in two steps and used in the next reaction without fur-
ther purification.
The ketone 2b12 (1.48 g, 7.01 mmol, 1 equiv), the above
amine (1.37 g, 1.5 equiv), and Ti(OiPr)4 (1.99 g,
1.0 equiv) in 20 mL DCM were stirred at rt for 3 h.
The mixture was cooled to 0 ꢁC and NaBH4 (0.53 g,
2 equiv) was added. The reaction mixture was kept at
this temperature for 4 h and MeOH was added to
quench the reaction. The solvent was removed and
200 mL EtOAc was added. After washing with 10 mL
of 1 N NaOH and removal of solvent, flash chromato-
graphy (5% MeOH in DCM + 1% NH3 (2 N in MeOH))
provided the desired product 3 (1.38 g, 61% yield) along
with 0.35 g of a mixture of ketone 2b and the corre-
sponding cis-alcohol. The use of other reducing agent
such as Na(OAc)3BH gave a mixture of cis- and trans-
isomers.
4.3.3. Synthesis of 4. The ketone 2b (0.97 g, 4.60 mmol,
1 equiv), hydroxyethylamine (0.308 g, 1.1 equiv), and
Ti(OiPr)4 (1.44 g, 1.1 equiv) in 20 mL DCM were stirred
under nitrogen at rt for overnight. The mixture was
cooled to 0 ꢁC and NaBH4 (0.16 g, 1 equiv) was added.
The reaction mixture was kept at this temperature for
30 min and then warmed up to rt for 4 h. Five millime-
ters of MeOH was added to quench the reaction. The
solvent was removed and 200 mL EtOAc was added.
After washing with 10 mL 1 N of NaOH and removal
of solvent, flash chromatography (5% MeOH in
DCM + 1% NH3 (2 N in MeOH)) provided the desired
product (0.56 g, 48% yield). The amine was treated with
Boc2O (0.63 g, 1.3 equiv), Na2CO3 (0.35 g, 1.5 equiv) in
a mixture of 17 mL THF and 17 mL of water. After stir-
ring at rt for 1.5 h, the solvent was removed and the
crude was extracted with EtOAc. Purification by flash
chromatography (5% MeOH in DCM + 1% NH3) pro-
vided the desired product (0.79 g, 100% yield). 1H
NMR (CDCl3): d 0.88 (m, 1H) 0.99 (m, 1H) 1.30 (m,
2H) 1.40 (s, 9H) 1.60–1.70 (m, 3H) 2.00 (m, 1H) 2.30
(m, 2H) 3.30 (br, 2H) 3.70 (m, 2H) 7.30–7.50 (m, 4H).
MS for (MH+) C21H29N2O3: 357.
1
Compound 3: H NMR (CDCl3): d 0.79 (m, 1H) 0.98
(m, 1H) 1.00 (m, 1H) 1.20–1.35 (m, 2H) 1.60 (m, 2H)
1.90 (m, 1H) 2.10–2.35 (m, 4H) 2.40–2.50 (m, 2H)
2.50–2.78 (m, 5H) 2.82 (m, 1H) 4.30 (m, 1H) 7.30–7.58
(m, 4H).
MS for (MH+) C20H28N3O: 326.
4.3.2. Synthesis of 1. Compound 3 (10 mg) in 1 mL
DCM was treated with 1.1 equiv of isocyanate and stir-
red at rt. Once the starting material was consumed (15–
30 min), 2 equiv of resin-bound trisamine was added
and stirred for 2 h. After filtration, the solution was
treated with 10–20 equiv of resin-bound p-toluenesulfo-
nic acid in MeOH for overnight. The final product was
released from this resin by stirring with 2 N NH3 in
Dess–Martin oxidation of this Boc-protected amino
alcohol was done by treating the above alcohol (2.0 g,
5.62 mmol) with 2.86 g of Dess–Martin reagent
(1.2 equiv) in 30 mL DCM at rt for 3 h. The solvent
was removed and EtOAc was added. After washing with
saturated Na2CO3 solution and filtration through Celite,

5380
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
the solvent was removed followed by purification by
flash chromatography (25% EtOAc/hexane). The de-
sired product 4 was isolated in 70% yield (1.40 g). H
NMR (CDCl3): d 0.90 (m, 1H) 0.99 (m, 1H) 1.30 (m,
1H) 1.40 (s, 9H) 1.40–1.60 (m, 3H) 2.10 (m, 1H) 2.30
(m, 2H) 3.80 (m, 2H) 4.20 (m, 1H) 7.30–7.60 (m, 4H)
9.70 (s, 1H). MS for (MH+) C21H27N2O3: 355.
Compound 6f: MS for (MH+) C30H38N5O: 485.
Compound 6g: MS for (MH+) C29H37F2N4O: 495.
Compound 6h: MS for (MH+) C29H37F2N4O: 495.
Compound 6i: MS for (MH+) C29H38FN4O: 477.
Compound 6j: MS for (MH+) C28H36Cl2N5O: 529.
1
4.3.4. Synthesis of 5. A general procedure as follows: the
aldehyde 4 was treated with 2 equiv of amine and
2 equiv of Na(OAc)3BH in DCE at rt for overnight.
After quenching the reaction with MeOH, the product
was purified by flash chromatography. The desired
product was treated with TFA (50% in DCM) for
12 h. After removal of solvent, EtOAc was added and
the solution was washed with 1 N NaOH. The free base
was then treated with 1.1 equiv of 3-chloro-4-fluoro-
benzonitrile in DCM at rt for 2 h. The workup proce-
dure follows that for the synthesis of 1.
4.3.5. Synthesis of 7a. Compound 7a was synthesized
according to Ref. 12.
1H NMR (CDCl3): d 0.78 (m, 1H) 1.00 (m, 1H) 1.20 (m,
1H) 1.42 (m, 1H) 1.63 (m, 1H) 1.80 (db, 1H,
J = 13.5 Hz) 2.10–2.28 (m, 3H) 3.90–4.00 (m, 4H) 7.10
(m, 1H) 7.20 (m, 4H).
MS for (MH+) C15H19O2: 231.
Compound 5a: MS for (MH+) C28H33ClFN4O: 496.
Compound 5b: MS for (MH+) C29H34ClFN5O2: 539.
Compound 5c: MS for (MH+) C28H33ClFN4O2: 512.
Compound 5d: MS for (MH+) C28H33ClFN4O2: 512.
Compound 5e: MS for (MH+) C29H35ClFN4O: 510.
Compound 5f: MS for (MH+) C29H35ClFN4O2: 526.
Compound 5g: MS for (MH+) C28H33ClFN4O2: 512.
Compound 5h: MS for (MH+) C27H31ClFN4O2: 498.
Compound 5i: MS for (MH+) C29H35ClFN4O: 510.
Compound 5j: MS for (MH+) C27H33ClFN4O2: 500.
Compound 7b was synthesized according to Ref. 12.
Compound 7c was synthesized according to Ref. 12.
1H NMR (CDCl3): d 0.78 (m, 1H) 1.00 (m, 1H) 1.20 (m,
1H) 1.42 (m, 1H) 1.60 (m, 1H) 1.80 (db, 1H,
J = 13.5 Hz) 2.10–2.30 (m, 3H) 3.90–4.00 (m, 4H) 7.10
(m, 1H) 7.20 (m, 2H) 7.41 (s, 1H).
MS for (MH+) C15H18BrO2: 309, 311.
4.3.6. Synthesis of 2e. Compound 7b (19 g, 74.5 mmol)
was treated with KOH (60 g), EtOH (180 mL), and
water (120 mL), and the mixture was heated to 100 ꢁC
for 36 h. The crude was cooled to rt and most of the sol-
vent removed. EtOAc was added and 2 N HCl was
added to adjust the pH to 1. After extraction with
EtOAc 3· 200 mL and removal of the solvent, the crude
was treated with 60 g of K2CO3, 50 mL MeI, and
200 mL acetone at 66 ꢁC for 6 h. The solution was fil-
tered and solvent was removed. This crude was treated
with 15 mL of concentrated HCl, 150 mL of acetone,
and 30 mL of water at rt for overnight. Removal of sol-
vent, extraction with EtOAc followed by flash chroma-
tography (25% EtOAc/hexane) gave 12 g of the desired
product 2e in 66% yield.
1
Compound 6a: H NMR (CDCl3): d 0.79 (m, 1H) 1.00
(m, 1H) 1.10 (d, 12H, J = 7.5 Hz) 1.30 (m, 1H) 1.40
(m, 1H) 2.10 (m, 1H) 2.25 (m, 1H) 2.40 (m, 1H) 2.70
(br, 2H) 3.10 (m, 2H) 3.30 (m, 2H) 6.99 (t, 1H,
J = 8.8 Hz) 7.10 (m, 1H) 7.30–7.80 (m, 5H) 10.80 (s,
1H).
13C NMR (CDCl3): d 18.11, 18.59, 20.35, 24.78, 25.52,
27.78, 31.54, 45.80, 48.62, 50.66, 52.32, 112.20, 116.21,
118.79, 119.06, 120.66, 121.48, 129.03, 129.43, 131.26,
132.23, 137.01, 149.07, 154.44, 157.58.
1H NMR (CDCl3): d 1.05 (m, 1H) 1.13 (m, 1H) 1.50 (m,
1H) 2.20 (m, 2H) 2.40 (m, 2H) 2.70 (m, 1H) 2.85 (m, 1H)
3.90 (s, 3H) 7.39 (m, 1H) 7.54 (d, 1H, J = 7.8 Hz) 7.88
(d, 1H, J = 7.9 Hz) 7.98 (s, 1H).
LCMS: rt = 5.29 min. Purity: 98.0%. MS for (MH+)
C29H37ClFN4O: 511.3. HRMS: Calcd 511.2640. Found:
511.2656.
MS for (MH+) C15H17O3: 245.
1
Compound 2a: H NMR (CDCl3): d 0.95 (m, 1H) 1.10
(m, 1H) 1.50 (m, 1H) 2.25 (m, 2H) 2.44 (m, 2H) 2.70
(m, 1H) 2.88 (m, 1H) 7.20 (m, 1H) 7.22 (m, 4H).
Compound 6b: MS for (MH+) C29H38N5O3: 505.
Compound 6c: MS for (MH+) C29H38BrN4O: 539, 537.
Compound 6d: MS for (MH+) C29H38FN4O: 477.
Compound 6e: MS for (MH+) C29H38ClN4O: 493.
MS for (MH+) C13H15O: 187.2.
1
Compound 2c: H NMR (CDCl3): d 1.00 (m, 2H) 1.10
(m, 1H) 1.42 (m, 1H) 2.20 (m, 2H) 2.40 (m, 2H) 2.61
(m, 1H) 2.80 (m, 1H) 7.20 (m, 3H) 7.30 (m, 1H).

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5381
MS for (MH+) C13H14BrO: 265, 267.
3.18 (m, 2H) 3.25 (m, 2H) 4.20 (m, 1H) 4.64 (s, 2H)
7.00 (m, 1H) 7.10–7.30 (m, 5H) 7.55 (m, 1H) 10.80
(s, 1H).
Compound 2d: Compound 2b (8.09 g, 38.34 mmol) was
treated with KOH (4.3 g, 2 equiv) in 150 mL EtOH and
the mixture was heated to 80 ꢁC for 14 h. The crude was
cooled to rt and most of the solvent removed. Water
(30 mL), EtOAc were added and HCl (5 N) was added
to adjust pH to 2. After extraction with EtOAc 3·
200 mL, removal of the solvent, the organic layer was
dried with Na2SO4. Removal of solvent gave 6.4 g of
the desired product 2d in 66% yield as brownish syrup
(73% yield).
MS for (MH+) C29H40ClFN3O2: 516.
Compound 8g: following the Dess–Martin oxidation
condition for 2b, 8f (40 mg) was converted to the alde-
hyde intermediate which, after washing with saturated
Na2CO3 and without purification, was treated with
N-methylhydroxyamine hydrochloride (10 mg) in
0.2 mL pyridine at rt over weekend. Preparative TLC
purification provided the desired product (16 mg, 38%
yield).
1H NMR (CDCl3): d 1.00–1.10 (m, 2H) 1.50 (m, 1H)
2.20 (m, 2H) 2.40 (m, 2H) 2.64 (m, 1H) 2.80 (m, 1H)
6.20 (br s, 2H) 7.38 (t, 1H, J = 8.1 Hz) 7.42 (m, 1H)
7.60 (m, 1H) 7.80 (s, 1H).
1H NMR (CDCl3): d 0.66 (m, 1H) 1.06 (m, 1H) 1.05 (d,
12H, J = 6.7 Hz) 1.20–1.40 (m, 2H) 1.60 (m, 2H) 2.18
(m, 1H) 2.22–2.40 (m, 2H) 2.65 (m, 2H) 3.18 (m, 2H)
3.25 (m, 2H) 4.00 (s, 3H) 4.20 (m, 1H) 7.00 (m, 1H)
7.10 (m, 1H) 7.22–7.30 (m, 2H) 7.70 (m, 2H) 8.00 (s,
1H) 10.80 (s, 1H).
MS for (MH+) C14H16NO2: 230.
Compound 8a: MS for (MH+) C28H38F2N3O: 470.
1
Compound 8c: H NMR (CDCl3): d 0.70 (m, 1H) 1.00
LCMS: rt = 5.78 min. Purity: 89.7%. MS for (MH+)
C30H41ClFN4O2: 543.1.
(m, 1H) 1.05 (d, 12H, J = 6.7 Hz) 1.22 (m, 1H) 1.38
(m, 1H) 1.60 (m, 2H) 2.10 (m, 1H) 2.20 (m, 1H) 2.38
(m, 1H) 2.65 (m, 2H) 3.10–3.25 (m, 4H) 4.20 (m, 1H)
6.90 (m, 1H) 6.95 (m, 1H) 7.10–7.20 (m, 2H) 7.40 (m,
2H) 10.80 (s, 1H).
Compound 8h: MS for (MH+) C29H40ClFN4O: 515.
Compound 8i: compound 8d (0.50 g, 0.945 mmol) in
9 mL THF under nitrogen at ꢀ78 ꢁC was treated with
1.4 mL of 1 M LiAlH4 in THF. After slowly warming
up to rt in 2.5 h, the mixture was heated to 35 ꢁC for
6 h and then rt for overnight. After quenching the reac-
tion with 0.07 mL water, 0.22 mL of 15% NaOH, and
0.07 mL water, the crude was filtered through Celite
and the solvent was removed. The product 8i was
purified by flash chromatography (33 mg, 7% yield).
MS for (MH+) C28H37BrF2N3O: 548, 550.
1
Compound 8d: H NMR (CDCl3): d 0.72 (m, 1H) 1.00
(m, 1H) 1.05 (d, 12H, J = 6.7 Hz) 1.22 (m, 1H) 1.38
(m, 1H) 1.60 (m, 2H) 2.16 (m, 1H) 2.28 (m, 1H) 2.40
(m, 1H) 2.65 (m, 2H) 3.15 (m, 2H) 3.25 (m, 2H) 4.19
(m, 1H) 5.80 (br s, 1H) 6.30 (br s, 1H) 7.00 (m, 1H)
7.10 (m, 1H) 7.35 (m, 1H) 7.42 (m, 1H) 7.55 (m, 1H)
7.60 (m, 1H) 7.77 (s, 1H) 10.80 (s, 1H).
1H NMR (CDCl3): d 0.60 (m, 1H) 0.99 (m, 1H) 1.03 (d,
12H, J = 6.7 Hz) 1.20–1.40 (m, 2H) 1.60 (m, 2H) 2.10–
2.40 (m, 3H) 2.62 (m, 2H) 3.10 (m, 2H) 3.20 (m, 2H)
3.80 (s, 2H) 4.16 (m, 1H) 6.95 (m, 1H) 7.00–7.20 (m,
5H) 7.55 (m, 1H) 10.77 (s, 1H).
MS for (MH+) C29H39ClFN4O2: 529.
1
Compound 8e: H NMR (CDCl3): d 0.70 (m, 1H) 1.05
(m, 1H) 1.10 (d, 12H, J = 6.7 Hz) 1.30 (m, 1H) 1.40
(m, 1H) 1.60 (m, 2H) 2.18 (m, 1H) 2.28 (m, 1H) 2.40
(m, 1H) 2.69 (br s, 2H) 3.18 (m, 2H) 3.30 (m, 2H) 3.95
(s, 3H) 4.19 (m, 1H) 6.95 (m, 1H) 7.00 (m, 1H) 7.30–
7.50 (m, 3H) 7.82 (m, 1H) 7.95 (s, 1H) 10.80 (s, 1H).
MS for (MH+) C29H40ClFN4O: 515.
1
Compound 8j: H NMR (CDCl3): d 0.60 (m, 1H) 0.99
(m, 1H) 1.00 (d, 12H, J = 6.7 Hz) 1.20–1.40 (m, 2H)
1.60 (m, 2H) 2.00–2.40 (m, 3H) 2.62 (m, 2H) 2.80 (s,
3H) 3.10 (m, 2H) 3.20 (m, 2H) 3.95 (m, 1H) 4.20 (s,
2H) 4.80 (br s, 1H) 6.99–7.20 (m, 6H) 7.50 (m, 1H)
10.77 (s, 1H).
MS for (MH+) C30H40F2N3O3: 528.
Compound 8f: The methyl ester analog of 8f (0.31 g,
0.57 mmol) in 2 mL THF under nitrogen at ꢀ78 ꢁC
was treated with 0.6 mL of 1 M LiAlH4 in THF for
0.5 h. After quenching the reaction with 2 mL MeOH,
the solvent was removed and after extraction with
EtOAc and removal of solvent, the crude NMR indi-
cated a mixture of starting material and the product.
The product 8f was purified by flash chromatography
(16 mg, 5%).
MS for (MH+) C30H43ClFN4O3S: 593.
Compound 8k: MS for (MH+) C31H43ClFN4O2: 557.
Compound 8l: MS for (MH+) C33H47ClFN4O2: 585.
Compound 8m: MS for (MH+) C34H50ClFN5O: 598.
Compound 8n: MS for (MH+) C33H49ClFN4O: 571.
Compound 8o: MS for (MH+) C31H45ClFN4O: 543.
1H NMR (CDCl3): d 0.66 (m, 1H) 1.06 (m, 1H) 1.05
(d, 12H, J = 6.7 Hz) 1.22–1.40 (m, 2H) 1.60 (m, 2H)
2.16 (m, 1H) 2.28 (m, 1H) 2.40 (m, 1H) 2.65 (m, 2H)

5382
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
4.3.7. Synthesis of 9. The synthesis followed the proce-
dure for 5 with key procedure as: (a) reductive amina-
tion of 4 with isopropylamine; (b) alkylation with 2-
bromoethanol (1.5 equiv), NaHCO3 (2 equiv), EtOH,
78 ꢁC, 14 h; (c) 50% TFA/DCM, 2 h; (d) ArNCO
(0.9 equiv).
(1 equiv), K2CO3 (1 equiv), 45 ꢁC, 14 h, 90% yield; (c)
LiAlH4 (1.1 equiv) reduction, THF, 70 ꢁC, 12 h, 60%
yield; (d) reductive alkylation with 2b; (e) ArNCO
(0.9 equiv).
MS for (MH+) C29H35ClFN4O2: 525.
(MH+) C28H35ClFN4O2: 513.
Compound 20: MS for (MH+) C28H34Cl2N5O2: 542.
4.3.8. Synthesis of 10. The synthesis followed the proce-
dure for 5 with key procedure as: (a) (R)-alaninol
(1 equiv), 2-iodopropane (1 equiv), CH3CN, 40 ꢁC,
14 h; (b) reductive amination with 4; (c) 50% TFA/
DCM, 2 h; (d) ArNCO (0.9 equiv).
4.3.9. Synthesis of 22a. The synthesis was accomplished
following the procedure for 1 with key procedure as:
(a) reductive alkylation of 2b with N-Boc 4-(amino-
methyl)piperidine; (b) ArNCO; (c) 50% TFA/DCM; (d)
reductive alkylation with tetrahydro-4H-pyran-4-one.
MS for (MH+) C29H37ClFN4O2: 527.
MS for (MH+) C32H39ClFN4O2: 565.
Compound 11: synthesis followed the procedure for 10.
Compound 22b: MS for (MH+) C30H34F4N5O2: 572.
Compound 22c: MS for (MH+) C29H33ClFN4O2: 523.
MS for (MH+) C29H37ClFN4O2: 527.
Compound 12: synthesis followed the procedure for 5
with key procedure as: (a) reductive amination of 4 with
2-fluoroethylamine; (b) reductive alkylation with ace-
tone; (c) 50% TFA/DCM, 2 h; (d) ArNCO (1 equiv).
Compound 22d: 1H NMR (CDCl3): d 0.82 (m, 1H) 1.00
(m, 1H) 1.40 (m, 3H) 1.50–1.70 (m, 2H) 1.80 (m, 3H)
2.10 (m, 1H) 2.36 (m, 2H) 2.60 (m, 2H) 2.77 (s, 3H)
3.10 (m, 2H) 3.70 (m, 1H) 3.80 (m, 2H) 6.56 (br s, 1H)
7.02 (m, 1H) 7.19 (m, 1H) 7.20–7.60 (m, 5H).
MS for (MH+) C28H34ClF2N4O: 515.
Compound 13: MS for (MH+) C28H36Cl2N5O: 528.
MS for (MH+) C28H33ClFN4O3S: 559.
1
Compound 14: H NMR (CDCl3): d 0.77 (m, 1H) 0.99
Compound 22e: MS for (MH+) C28H33F2N4O3S: 543.
Compound 22f: MS for (MH+) C29H35F2N4O3S: 557.
Compound 22g: MS for (MH+) C29H33F4N4O3S: 593.
Compound 22h: MS for (MH+) C28H34FN4O3S: 525.
Compound 22i: MS for (MH+) C30H35F4N4O3S: 607.
(m, 1H) 1.03 (d, 6H, J = 6.7 Hz) 1.20–1.40 (m, 2H)
1.60 (m, 2H) 2.10 (m, 1H) 2.30–2.40 (m, 2H) 2.60–2.80
(m, 4H) 3.10 (m, 1H) 3.30 (m, 2H) 3.80 (m, 2H) 4.16
(m, 1H) 7.35–7.60 (m, 6H) 10.77 (s, 1H).
MS for (MH+) C27H34Cl2N5O2: 530.
Compound 15: MS for (MH+) C28H36Cl2N5O2: 544.
1
Compound 16: H NMR (CDCl3): d 0.77 (m, 1H) 1.01
Compound 22j: 1H NMR (CDCl3): d 0.82 (m, 1H)
0.99 (m, 1H) 1.44 (m, 3H) 1.60–1.80 (m, 5H) 1.80–
2.00 (m, 2H) 2.30 (m, 2H) 2.74 (s, 3H) 2.90 (m, 1H)
2.82–3.10 (m, 2H) 3.30–3.70 (m, 4H) 6.42 (d, 1H,
J = 9.3 Hz) 7.00 (t, 1H, J = 8.6 Hz) 7.10 (m, 1H)
7.30–7.50 (m, 5H).
(m, 1H) 1.04 (d, 6H, J = 6.7 Hz) 1.30–1.40 (m, 2H)
1.60 (m, 2H) 2.10 (m, 1H) 2.30–2.40 (m, 2H) 2.60 (m,
2H) 2.80–2.98 (m, 2H) 3.10 (m, 1H) 3.30 (m, 2H) 3.80
(m, 2H) 4.16 (m, 1H) 4.56 (m, 1H) 4.63 (m, 1H) 7.35–
7.60 (m, 6H) 10.89 (s, 1H).
LCMS: rt = 4.88 min. Purity: 98.5%. MS for (MH+)
C27H33Cl2FN5O: 532.1.
MS for (MH+) C28H33ClFN4O3S: 559.
Compound 22k: MS for (MH+) C28H33ClFN4O3S: 559.
Compound 22l: MS for (MH+) C27H31ClFN4O3S: 545.
Compound 17: synthesis followed the procedure for 5
with key procedure as: (a) 4, acetone, NaBH(OAc)3,
DCM; (b) TFAA (2 equiv), Et3N (3 equiv), DCM, rt,
12 h; (c) 50% TFA in DCM, 24 h; (d) ArNCO; (e)
NaOH (1 N in MeOH, 2 equiv), rt, 14 h.
Compound 23a: 1H NMR (CDCl3): d 0.70 (m, 1H) 1.00
(m, 1H) 1.30 (m, 3H) 1.45 (m, 1H) 1.60–1.80 (m, 7H)
2.10 (m, 1H) 2.30 (m, 2H) 2.58 (m, 2H) 2.76 (s, 3H)
3.04 (m, 2H) 3.60–3.80 (m, 5H) 6.50 (br s, 1H) 7.00–
7.20 (m, 5H) 7.44 (m, 1H).
MS for (MH+) C25H30Cl2N5O: 486.
Compound 18: MS for (MH+) C24H27Cl2FN5O: 490.
MS for (MH+) C28H37ClFN4O3S: 563.
Compound 19: synthesis followed the procedure for 5
with key procedure as: (a) LiAlH4 (1 equiv) reduction
of 5-methyl-2-(S)-proline methyl ester to the alcohol,
THF, 70 ꢁC, 12 h, 68% yield; (b) iodoacetonitrile
Compound 23b: MS for (MH+) C30H41ClFN4O3S: 591.
Compound 23c: MS for (MH+) C27H36Cl2N5O3S: 580.

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5383
Compound 23d: MS for (MH+) C29H40Cl2N5O3S: 608.
Compound 23e: MS for (MH+) C31H42Cl2N5O4S: 650.
for 3 h. Saturated NaCl solution was added followed
by extraction with EtOAc. Flash chromatography
(40:100 EtOAc/hexane) gave 27 as white solid (1.48 g,
96% yield).
4.3.10. Synthesis of 27. Ketone 24 (9.9 g, 63.5 mmol,
1 equiv)/200 mL DCM/N-(2-aminoethyl)pyrrolidine (8.7 g,
1.2 equiv)/Ti(OiPr)4 (21.6 g, 1.2 equiv) was stirred at
room temperature for overnight. The mixture was
cooled to 0 ꢁC followed by addition of NaBH4 (3.4 g,
1.4 equiv). After another overnight stirring, 5 mL of
MeOH was added. After 2 h, saturated Na2CO3 was
added followed by extraction with EtOAc. Flash chro-
matography (10:1:90 MeOH/NH3/DCM) provided
9.8 g of the desired product (60% yield).
1H NMR (CDCl3): d 1.86 (m, 6 H) 2.19 (m, 1H) 2.32–
2.36 (m, 2H) 2.56 (m, 1H) 2.67–2.72 (m, 6H) 3.23 (m,
2H) 4.36 (m, 1H) 5.69 (m, 1H) 6.96 (t, 1H, J = 8.8 Hz)
7.08 (m, 1H) 7.40 (m, 2H) 11.07 (m, 1H).
Suzuki coupling reaction: Starting material 27 (25 mg,
0.049 mmol), boronic acid (2 equiv), LiCl (1.5 equiv),
Na2CO3 (3 equiv), PdCl2 (dppf) (0.15 equiv), toluene/
EtOH/H2O (3:1:1) under nitrogen, at 50 ꢁC for 10 min,
then at 80 ꢁC for 1 h. Prep TLC (5% MeOH in DCM
with 1% NH3) gave the desired product 28 in 50–80%
yield.
1H NMR (CDCl3): d 1.30–1.40 (m, 2H) 1.41–1.55 (m,
2H) 1.65 (br s, 6H) 1.80 (m, 2H) 2.40 (m, 5H) 2.52 (t,
3H, J = 7.1 Hz) 2.62 (t, 3H, J = 7.1 Hz) 3.80 (s, 4H).
Compound 28a: MS for (MH+) C23H28ClFN5O: 444.
Compound 28b: MS for (MH+) C24H29ClFN4O: 443.
Compound 28c: MS for (MH+) C23H29ClFN4O: 431.
Compound 28d: MS for (MH+) C23H28ClFN3O2: 432.
Compound 28e: MS for (MH+) C24H28ClFN3O3: 460.
Compound 28f: MS for (MH+) C27H30ClFN3O2: 482.
Compound 28g: MS for (MH+) C24H30ClFN3OS: 462.
Compound 28h: MS for (MH+) C25H30ClFN3O2S: 490.
MS: C14H27N2O2: 255 (MH)+.
This material (3.1 g, 12.2 mmol, 1 equiv) was treated
with Boc2O (3.2 g, 1.2 equiv), NaHCO3 (2 g, 1.5 equiv)
in 40 mL THF and 40 mL water. The mixture was
heated to 55 ꢁC for 48 h. After cooling, extraction with
EtOAc, and drying the organic layer with Na2CO3,
evaporation of the solvent afforded colorless syrup
(4.4 g). This syrup was treated with CeCl3 ꢁ 7H2O (9 g,
2 equiv), NaI (0.54 g, 0.3 equiv) in 80 mL of CH3CN
at 80 ꢁC under nitrogen for 2 h. Another 4.5 g of
CeCl3ꢁ7H2O (1 equiv), NaI (0.54 g, 0.3 equiv), and
60 mL of CH3CN were added and the mixture was
heated for 16 h. After cooling to room temperature,
extraction with EtOAc, and drying with Na2SO4, evap-
oration of solvent provided light yellow syrup 25a and
25b (3.1 g total, 25a/25b = 1.4:1).
1
Compound 28i: H NMR (CDCl3): d 1.50 (m, 2H) 1.70
(m, 1H) 1.80 (m, 4H) 2.19 (m, 1H) 2.40 (m, 1H) 2.55 (m,
2H) 2.70 (m, 5H) 4.20 (m, 1H) 6.20 (br s, 1H) 6.83 (m,
1H) 6.94 (m, 1H) 7.10 (m, 1H) 7.40 (m, 2H) 11.00 (s, 1H).
The above material (3.1 g, 10 mmol) was dissolved in 1:1
toluene/THF 50 mL, cooled to ꢀ78 ꢁC followed by
addition of 5.36 g of Tf2NPh (15 mmol, 1.5 equiv) and
slow addition of KHMDS solution 30 mL (0.5 N in
toluene, 1.5 equiv). After 4 h of stirring, 40 mL of water
was added into reaction mixture. This reaction mixture
was slowly warmed up to room temperature followed
by extraction of EtOAc. Flash chromatography
(3:1:100 MeOH/NH3/DCM) gave the first fraction as
colorless syrup 26a (1.5 g).
MS for (MH+) C24H27ClFN4OS: 473.
Compound 28j: 1H NMR (CDCl3): d 1.60–1.80 (m, 4H)
1.82 (m, 4H) 2.19 (m, 1H) 2.36 (m, 1H) 2.55 (m, 2H)
2.60–2.80 (m, 4H) 3.22 (m, 2H) 3.90 (s, 2H) 4.22 (m,
1H) 6.00 (br s, 1H) 6.70 (s, 2H) 6.94 (m, 1H) 7.10 (m,
1H) 7.40 (m, 2H) 11.00 (s, 1H).
MS for (MH+) C24H31ClFN4OS: 477.
1H NMR (CDCl3): d 1.38 (s, 9H) 1.69 (m, 4H) 1.80
(m, 1H) 1.88 (m, 1H) 2.20 (m, 1H) 2.26–2.30 (m,
2H) 2.47 (m, 7H) 3.13 (br s, 2H) 4.03 (br s, 1H)
5.62 (m, 1H).
Compound 28k: MS for (MH+) C23H26ClFN5OS: 474.
Compound 28l: 1H NMR (CDCl3): d 1.60–1.90 (m, 8H)
2.10 (m, 1H) 2.30 (m, 1H) 2.56 (m, 2H) 2.70 (m, 4H)
3.22 (br s, 2H) 4.04 (br s, 2H) 4.20 (m, 1H) 5.98 (br s,
1H) 6.95 (m, 1H) 7.10 (m, 1H) 7.40 (m, 2H).
Chromatography also gave the second fraction as color-
less syrup 26b (1.1 g).
1H NMR (CDCl3): d 1.53 (m, 1H) 1.67 (m, 4H) 1.89–
1.92 (m, 3H) 2.31 (m, 2H) 2.42 (m, 5H) 2.52 (t, 2H,
J = 6 Hz) 2.67 (m, 3H) 5.59 (m, 1H).
LCMS: rt = 3.77 min. Purity: 97.5%. MS for (MH+)
C23H30ClFN5OS: 478.1.
Compound 28m: MS for (MH+) C25H34ClFN5OS: 506.
This material 26b (1 g, 2.9 mmol, 1 equiv)/ 15 mL DCM/
4-fluoro-3-chlorophenyl isocyanate (0.6 g, 3.5 mmol,
1.2 equiv) was stirred at room temperature under N2
Compound 28n: the procedure is similar to the above: 2-
tributylstannylthiazole (1.35 g, 3.62 mmol) was mixed

5384
J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5. Shimada, M.; Tritos, N.; Lowell, B.; Flier, J.; Maratos-
Flier, E. Nature 1998, 396, 670.
with 26a (0.8 g, 1.81 mmol), Pd(PPh3)4 (63 mg,
0.03 equiv), and LiCl (0.23 g, 3 equiv) in 10 mL THF.
The mixture was heated to 75 ꢁC under nitrogen for
2 h. EtOAc was added and after washing with water,
the solvent was removed and chromatography (5%
MeOH in DCM with 1% NH3) provided 0.51 g coupling
product in 75% yield. This compound was dissolved in
5 mL THF and treated with LDA (1.25 equiv) at
ꢀ78 ꢁC MS for 20 min. DMF (3 equiv) was added to
quench the reaction. After 1 h, water was added and
after extraction with EtOAc, organic layer was dried
and the crude syrup was directly dissolved in 50%
DCM in MeOH and NaBH4 (1.2 equiv) was added at
ꢀ30 ꢁC. Warming up to 0 ꢁC in 30 min and extraction
with EtOAc, drying the solvent, and chromatography
provided the alcohol (0.135 g). This alcohol was treated
with 4 equiv of dppa, 1.5 equiv of DBU in 3 mL of THF
and the mixture was stirred at 70 ꢁC for 2 h. After work-
up, chromatography (5% MeOH in DCM with 1% NH3)
provided the desired azide which was immediately trea-
ted with 10% TFA in DCM at rt for 1.5 h. After removal
of TFA, EtOAc was added and the organic layer was
washed with Na2CO3. Drying the solvent provided
76 mg of the amine. This amine was then treated with
aryl isocyanate followed by reduction of azide with
resin-bound triphenylphosphine (2 equiv) in 2 mL
THF/0.1 mL water at rt for 24 h. Chromatography pro-
6. (a) Chambers, J.; Amers, R. S.; Bergsma, D.; Muir, A.;
Fitzgerald, L. R.; Hervieu, G.; Dytko, G. M.; Foley, J. J.;
Martin, J.; Liu, W.-S.; Park, J.; Ellis, C.; Ganguly, S.;
Konchar, S.; Cluderay, J.; Leslie, R.; Wilson, S.; Sarau, H.
M. Nature 1999, 400, 261; (b) Saito, Y.; Nothacker, H. P.;
Wang, Z.; Lin, H. S.; Leslie, F.; Civelli, O. Nature 1999,
400, 265; (c) Sailer, A. W.; Sano, H.; Zeng, Z.; McDonald,
T.; Pan, J.; Pong, S.-S.; Feighner, S. D.; Tan, C. P.;
Fukami, T.; Iwaasa, H.; Hreniuk, D. L.; Morin, N. R.;
Sadowski, S. J.; Ito, M.; Bansal, A.; Ky, B.; Figueroa, D.
J.; Jiang, Q.; Austin, C. P.; MacNeil, D. J.; Ishihara, A.;
Ihara, M.; Kanatani, A.; Van der Ploeg, L. H. T.;
Howard, A. D.; Liu, Q. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 7564–7569.
7. Goodfellow, V.; Rowbottom, M.; Dyck, B. P.; Tamiya, J.;
Zhang, M.; Grey, T. D. WO 04081005 A1, 2004.
8. For a recent review, see: Handlon, A. L.; Zhou, H. J. Med.
Chem. 2006, 49, 4017.
9. (a) Dyck, B.; Markison, S.; Zhao, L.; Tamiya, J.; Grey, J.;
Rowbottom, M. W.; Zhang, M.; Vickers, T.; Sorensen, K.;
Norton, C.; Wen, J.; Heise, C. E.; Saunders, J.; Conlon,
P.; Madan, A.; Schwarz, D.; Goodfellow, V. S. J. Med.
Chem. 2006, 49, 3753; (b) Kym, P. R.; Souers, A. J.;
Campbell, T. J.; Lynch, J. K.; Judd, A. S.; Iyengar, R.,
et al. J. Med. Chem. 2006, 49, 2339; (c) Kowalski, T. J.;
Spar, B. D.; Weig, B.; Farley, C.; Cook, J.; Ghibaudi, L.;
Fried, S.; O’Neil, K.; Del Vecchio, R. A.; McBriar, M.;
Guzik, H.; Clader, J.; Hawes, B. E.; Hwa, J. Eur. J.
Pharmacol. 2006, 535, 182–191; (d) McBriar, M. D.;
Guzik, H.; Shapiro, S.; Paruchova, J.; Xu, R.; Palani, A.;
Clader, J. W.; Cox, K.; Greenlee, W. J.; Hawes, B. E.;
Kowalski, T. J.; ONeil, K.; Spar, B. D.; Weig, B.; Weston,
D. J.; Farley, C.; Cook, J. J. Med. Chem. 2006, 49, 2294;
(e) Palani, A.; Shapiro, S.; McBriar, M. D.; Clader, J. W.;
Greenlee, W. J.; Spar, B.; Kowalski, T. J.; Farley, C.;
Cook, J.; van Heek, M.; Weig, B.; O’Neil, K.; Graziano,
M.; Hawes, B. J. Med. Chem. 2005, 48, 4746; (f)
Vasudevan, A.; LaMarche, M. J.; Blackburn, C.; Che, J.
L.; Luchaco-Cullis, C. A.; Lai, S., et al. Bioorg. Med.
Chem. Lett. 2005, 15, 4174; (g) Kym, P. R.; Iyengar, R.;
Souers, A. J.; Lynch, J. K.; Judd, A. S.; Gao, J., et al.
J. Med. Chem. 2005, 48, 5888; (h) McBriar, M. D.; Guzik,
H.; Xu, R.; Paruchova, J.; Li, S.; Palani, A.; Clader, J. W.;
Greenlee, W. J.; Hawes, B.; Kowalski, T. J.; O’Neil, K.;
Spar, B.; Weig, B. J. Med. Chem. 2005, 48, 2274; (i) Su, J.;
McKittrick, B. A.; Tang, H.; Czarniecki, M.; Greenlee, W.
J.; Hawes, B. E.; O’Neill, K. Bioorg. Med. Chem. 2005, 13,
1829; (j) Vasudevan, A.; Wodka, D.; Verzal, M. K.;
Andrew, J.; Gao, J.; Brodjian, S.; Fry, D.; Dayton, B.;
Marsh, K. C.; Hernandez, L. E.; Ogiela, C. A.; Collins, C.
A.; Kym, P. R. Bioorg. Med. Chem. Lett. 2004, 14, 4879;
(k) Arienzo, R.; Clark, D. E.; Cramp, S.; Daly, S.; Dyke,
H. J.; Lockery, P.; Norman, D.; Roach, A. G.; Stuttle, K.;
Tomlinson, M.; Wong, M.; Wren, S. P. Bioorg. Med.
Chem. Lett. 2004, 14, 4099; (l) Carpenter, A. J.; Hertzog,
D. L. Exp. Opin. Therap. Patents 2002, 12, 1639; (m)
Borowsky, B.; Durkin, M. M.; Ogozalek, K.; Marzabadi,
M. R.; DeLeon, J.; Lagu, B.; Heurich, R.; Lichtblau, H.;
Shaposhnik, Z.; Daniewska, I.; Blackburn, T. P.; Bran-
chek, T. A.; Gerald, C.; Vaysse, P. J.; Forray, C. Nat.
Med. 2002, 8, 1039.
1
vided the final desired product 28n (68 mg). H NMR
(CDCl3): d 1.10 (m, 1H) 1.60–1.80 (m, 3H) 1.90 (m,
5H) 2.20 (m, 1H) 2.40 (m, 1H) 2.60–2.80 (m, 8H) 3.30
(m, 2H) 4.00 (s, 2H) 4.50 (m, 1H) 6.50 (m, 1H) 6.99
(m, 1H) 7.15 (m, 1H) 7.40 (m, 2H).
MS for (MH+) C23H30ClFN5OS: 478.
Acknowledgments
We thank Drs. T.M. Chan, Andy Evans, Jianshe Kong,
Charles McNemar, Hong Mei, Cymbelene Nardo, Re-
becca Osterman, Daniel Weston, Jesse Wong, and
Xue-Song Zhang for technical supports in this program.
We also thank Drs. Ruo Xu, Mark McBriar, Wen-Lian
Wu, T.K. Sasikumar, Michael Czarniecki, and Michael
Graziano for their helpful suggestions, discussions, and
support.
References and notes
1. For a review of current approaches for the treatment of
´
´
obesity, see: (a) Fernandez-Lopez, J.; Remesar, X.; Foz,
M.; Alemany, M. Drugs 2002, 62, 915–944; (b) Bray, G. A.
J. Med. Chem. 2006, 49, 4001.
2. Kawauchi, H.; Kawazoe, I.; Tsubokawa, M.; Kishida, M.;
Baker, B. L. Nature 1983, 305, 321.
3. Rossi, M.; Choi, S.; O’Shea, D.; Miyoshi, T.; Ghatei, M.
A.; Bloom, S. R. Endocrinology 1997, 138, 351.
4. (a) Qu, D.; Ludwig, D. S.; Gammeltoft, S.; Piper, M.;
Pelleymounter, M.; Cullen, M. J.; Mathes, W. F.; Przypek,
J.; Kanarek, R.; Maratos-Flier, E. Nature 1996, 380, 243;
(b) Ludwig, D. S.; Tritos, N. A.; Mastaitis, J. W.;
Kulkarni, R.; Kokkotou, E.; Elmquist, J.; Lowell, B.;
Flier, J. S.; Maratos-Flier, E. J. Clin. Invest. 2001, 107,
379.
10. Amgen: AMG 076, 2005.
11. Neurogen: NGD-4715, November 14, 2006.
12. Xu, R.; Li, S.; Paruchova, J.; McBriar, M. D.; Guzik, H.;
Palani, A.; Clader, J. W.; Cox, K.; Greenlee, W. J.; Hawes,
B. E.; Kowalski, T. J.; O’Neil, K.; Spar, B. D.; Weig, B.;
Weston, D. J. Bioorg. Med. Chem. 2006, 14, 3285.

J. Su et al. / Bioorg. Med. Chem. 15 (2007) 5369–5385
5385
13. Sorota, S.; Zhang, X.-S.; Margulis, M.; Tucker, K.;
Priestly, T. Assay Drug Dev. Technol. 2005, 3, 47–57.
14. Fivush, A. M.; Wilson, T. M. Tetrahedron Lett. 1997, 38,
7151.
for 21 h. LC/MS spectrum showed only the starting
material.
19. Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G.
J. Med. Chem. 2006, 49, 5029.
15. (a) Among 70 isocyanates, 51 were aryl ones and 19 were
alkyl ones. All alkyl products were inactive; (b) All Ki values
were determined in the human MCH R1 receptor binding
assay. See details in experimental part. All the compounds
reported in this article had been tested in the MCH mediated
Ca2+ influx functionalFLIPR assay and allof them were full
MCH R1 antagonists. For example, 6a: Kb 7.0 nM; 8g: Kb
4.6 nM; 16: Kb 11.4 nM; 28l: Kb 14.5 nM.
16. For example, the amide analog of 1h (with 3,4-difluor-
ophenylacetyl) has MCH R1 Ki of 7115 nM.
17. All compounds were dosed at 30 mg/kg. See also
Ref. 9c.
20. Roy, M.-L.; Dumaine, R.; Brown, A. M. Circulation 1996,
94, 817–823.
21. (a) For example, compound 22d had apical (AP) ! basal
(BL) permeability of 4.1 nm/s and BL ! AP permeability
of 21 nm/s, efflux ratio of 5; compound 23b had AP ! BL
permeability of 2.2 nm/s and BL ! AP permeability of
194 nm/s, efflux ratio of 88. A compound with efflux ratio
over 2 is considered as a pgp substrate; (b) Teodori, E.; Dei,
S.; Scapecchi, S.; Gualtieri, F. IL Farmaco 2002, 57, 385.
22. Marcantoni, E.; Nobili, F.; Bartoli, G.; Bosco, M.;
Sambri, L. J. Org. Chem. 1997, 62, 4183.
23. Calculated values for pKa: 9.1, 8.3, logP 4.4, logD 2.0
(using ACD/Labs software).
24. Audinot, V.; Lahaye, C.; Suply, T.; Beauverger, P.;
Rodriguez, M.; Galizzi, J.-P.; Fauchere, J.-L.; Boutin, J.
A. Br. J. Pharma. 2001, 133, 3931.
18. The stability of this 2,6-dichloropyridine moiety was
examined. For example, an analog of 23a with R as
hydroxymethyl (in place of aminomethyl) was mixed with
10 equiv of N-acyl cysteine in 50% water/DMSO at 37 ꢁC