Novel and potent inhibitors of stearoyl-CoA desaturase-1. Part II: Identification of 4-ethylamino-3-(2-hydroxyethoxy)-N-[5-(3-trifluoromethylbenzyl)thiazol-2-yl]benzamide and its biological evaluation

Article abstract of DOI:10.1016/j.bmcl.2009.05.123

The continuing investigation of SAR studies of 3-(2-hydroxyethoxy)-N-(5-benzylthiazol-2-yl)-benzamides as stearoyl-CoA desaturase-1 (SCD-1) inhibitors is reported. Our prior hit-to-lead effort resulted in the identification of 1a as a potent and orally efficacious SCD-1 inhibitor. Further optimization of the structural motif resulted in the identification of 4-ethylamino-3-(2-hydroxyethoxy)-N-[5-(3-trifluoromethylbenzyl)thiazol-2-yl]benzamide (37c) with sub nano molar IC₅₀ in both murine and human SCD-1 inhibitory assays. This compound demonstrated a dose-dependent decrease in the plasma desaturation index in C57BL/6J mice on a non-fat diet after 7 days of oral administration.

Full text of DOI:10.1016/j.bmcl.2009.05.123

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4159–4166
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel and potent inhibitors of stearoyl-CoA desaturase-1. Part II: Identification
of 4-ethylamino-3-(2-hydroxyethoxy)-N-[5-(3-trifluoromethylbenzyl)thiazol-2-
yl]benzamide and its biological evaluation
b
a
a
a
a
Yoshikazu Uto a, , Tsuneaki Ogata , Yohei Kiyotsuka , Yuriko Miyazawa , Yuko Ueno , Hitoshi Kurata ,
Tsuneo Deguchi c, Makiko Yamada c, Nobuaki Watanabe c, Toshiyuki Takagi d, Satoko Wakimoto e, Ryo
Okuyama e, Masahiro Konishi e, Nobuya Kurikawa e, Keita Kono e, Jun Osumi e
*
a Medicinal Chemistry Research Laboratories I, Daiichi Sankyo Co., Ltd, Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
b Global Project Management Department, Daiichi Sankyo Co., Ltd, Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
c Drug Metabolism & Pharmacokinetics Research Laboratories, Daiichi Sankyo Co., Ltd, Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
d Clinical Development Department I, Daiichi Sankyo Co., Ltd, Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
e Biological Research Laboratories II, Daiichi Sankyo Co., Ltd, Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The continuing investigation of SAR studies of 3-(2-hydroxyethoxy)-N-(5-benzylthiazol-2-yl)-benzam-
ides as stearoyl-CoA desaturase-1 (SCD-1) inhibitors is reported. Our prior hit-to-lead effort resulted in
the identification of 1a as a potent and orally efficacious SCD-1 inhibitor. Further optimization of the
structural motif resulted in the identification of 4-ethylamino-3-(2-hydroxyethoxy)-N-[5-(3-trifluorom-
ethylbenzyl)thiazol-2-yl]benzamide (37c) with sub nano molar IC50 in both murine and human SCD-1
inhibitory assays. This compound demonstrated a dose-dependent decrease in the plasma desaturation
index in C57BL/6J mice on a non-fat diet after 7 days of oral administration.
Received 17 March 2009
Revised 27 May 2009
Accepted 29 May 2009
Available online 6 June 2009
Keywords:
SCD-1 inhibitors
Thiazole
Ó 2009 Elsevier Ltd. All rights reserved.
Desaturation index
Stearoyl-CoA desaturase-1 (SCD-1), a microsomal enzyme, is a
rate-limiting enzyme in the synthesis of monounsaturated fatty
acids from their saturated fatty acid precursors.1,2 In adult mice,
SCD-1 isoform is expressed in lipogenic tissues including the liver
and adipose tissue. Deficiency of SCD-1 has been shown to cause
defective hepatic cholesterol ester and triglyceride synthesis,3
resistance against obesity,3 and reduced liver steatosis in rodents.4
In humans, a higher desaturation index (the ratio of oleate to stea-
rate or 18:1/18:0) is strongly correlated with higher plasma tri-
glyceride levels.5 Even though the detailed mechanism by which
SCD-1 deficiency affects body weight and adiposity is not com-
pletely understood, inhibition of SCD-1 may represent a novel ap-
proach for the treatment of metabolic syndromes. In the preceding
article,6 we reported optimization of the HTS hit compound to
identify potent and orally bioavailable SCD-1 inhibitors, such as 1
(Fig. 1). The improvement in oral bioavailability that was accom-
plished in the course of the optimization was significant, however,
more improvement in bioavailability is desirable to achieve more
potency in pharmacological studies in vivo. In this article, we
would like to report further exploration in SAR of the lead com-
pound to improve bioavailability and in vivo potency. Our plans
to modify 1 are summarized in Figure 1. Structurally, the hydroxy-
ethoxy functional group in the 3-position of the right-hand phenyl
is considered to be essential for both strong SCD-1 inhibitory
R1
S
N
1
1a
: R = H
NH
O
1b: R1 = CF3
O
F3C
OH
1
O
Essential 3-hydroxyethoxy
R1
V
S
W
R2
X
Y
OH
F3C
O
R2, V, W, X, Y are investigated.
* Corresponding author. Tel.: +81 3 3492 3131; fax: +81 3 5436 8563.
E-mail address: uto.yoshikazu.pw@daiichisankyo.co.jp (Y. Uto).
Figure 1. Plans for SAR studies of 1.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.05.123

4160
Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
activity and good oral exposure,6 whereas the methoxy on the 4-
position is presumed to be modifiable (R2 in Fig. 1). In regard to
the central pharmacophore, we tried to replace the thiazole with
other S-containing heteroaryls (X and Y in Fig. 1). The linkers be-
tween the thiazole core and the terminal phenyls on both ends
(V and W in Fig. 1) were investigated as well. As for the substitu-
ents on the left-hand phenyl, we reported in the preceding manu-
script6 that halogens and haloalkyls were preferred on the 3-, 3,4-
or 3,5-positions. We assumed it was not necessary to further inves-
tigate substitutions on the left hand phenyl at this point.
addition of 2-lithiothiophene to 3-trifluoromethylbenzaldehyde
and reduction of the resulting alcohol in the presence of an excess
amount of TMS–Cl and NaI9 provided 2-(3-trifluoromethylben-
zyl)thiophene (6) in 85% yield over two steps. Regioselective nitra-
tion by Cu(NO3)2 in acetic anhydride gave the desired
nitrothiophene (7) in 83% yield. Reduction, condensation with 10,
and THP deprotection gave 12. The thiadiazole analog (13)10 was
prepared from 5-(3-trifluoromethylbenzyl)-[1,3,4]thiadiazol-2-
ylamine (9) and 10 in the analogous procedures utilized for the
preparation of 11.
The synthetic routes for the compounds in Tables 1–3 are out-
lined in Schemes 1–3.7 Condensation of the commercially available
2-(3,5-bis-trifluoromethylphenyl)thioacetamide (2) and 2-chloro-
3-oxo-propionic acid ethyl ester (3)8 efficiently provided the de-
sired 2-benzylthiazole (4) in 97% yield. Saponification of the ethyl
ester, Curtius rearrangement in t-BuOH, and deprotection of the
N-Boc with TFA provided the 5-aminothaizole (5) in 38% yield over
three steps. Coupling between 5 and the benzoic acid (10) in the
presence of HATU was extremely sluggish and the desired product
11 was obtained in only 6% yield after THP deprotection
(Scheme 1). For the synthesis of thiophene analogs, nucleophilic
For optimization of the V linkers (Fig. 1), the synthetic routes
are outlined in Scheme 2. Lithium–halogen exchange on (5-bro-
mothiazol-2-yl)-carbamic acid tert-butyl ester (14)11 with n-BuLi,
addition of 3-trifluoromethylbenzaldehyde, Dess–Martin oxida-
tion, and acidic deprotection of the N-Boc group gave 15 in 42%
yield over three steps. The aminothiazole (15) was coupled with
10 and subsequent THP deprotection provided the 5-benzoylthiaz-
ole (16) in 54% yield over two steps. The ketone in 16 was reduced
with NaBH4 to produce the thiazole–phenyl–methanol analog (17).
Table 3
Evaluation of linkers to the thiazole core (V and W)
V
S
Table 1
Evaluation of heteroaryl cores
W
O
N
OH
OH
F3C
O
O
R
a
a
A
NH
O
No.
V
W
IC50 (nM)
IC50 (nM)
Inhibition %a at
10 M human D6
O
mouse
D
9
human
D
9
l
H
N
No.
R
IC50 (nM)
IC50 (nM)
Inhibition %a at
10
a
a
1a
2
3
22
A
O
mouse
D
9
human
D
9
lM
human
D6
H
N
O
16
17
20
24
29
>1000
276
NTb
884
>1000
828
32
NTb
8
S
N
O
O
O
O
1a
3-CF3
2
3
15
H
N
OH
S
1b
11
12
13
3,5-Di-CF3
3,5-Di-CF3
3-CF3
0.6
44
0.3
32
NTb
22
H
N
NTb
16
<5
N
S
O
>1000
836
N
N
S
76
145
8
22
H
S
N
N
49
3-CF3
10
<5
N
O
a
a
Values are the geometric means of at least two experiments.
NT—not tested.
Values are the geometric means of at least two experiments.
NT—not tested.
b
b
Table 2
Summaries of PK profilesa in C57BL/6J mice
S
N
S
NH
O
NH
N
O
O
N
O
F3C
OH
F3C
O
OH
13
1a
O
No.
PK profilesa (po, 20 mg/kg)
b
b
b
b
Cmax
(
lg/mL)
t1/2 (h)
Tmax (h)
AUC(0–8
(lg h/mL)
F (%)
h)
1a
13
1.7
0.12
3.5
2.5
0.7
1.0
8.2
0.4
12
2
a
A dose of each compound was either intravenously (5 mg/kg, DMA/Tween80/saline 10/10/80) injected into the tail vein of C57BL/6Jmice (n = 2) or orally (20 mg/kg, 0.5%
MC, n = 3) administered using an intubation tube. Plasma samples (20 lL) were collected up to 8 h after intravenous or oral administration. The plasma concentrations of the
compounds were determined by LC/MS.
b
Values are the geometric means of at least two experiments.

Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
4161
O
S
F3C
NH2
a
S
b-d
O
O
Ar
Ar
NH2
N
S
N
OEt
+
H
OEt
Cl
CF3
4
5
2
3
F3C
Ar =
CF3
S
S
e, f
g
NO2
S
CF3
CF3
NH2
6
7
OH
S
h, i
N
O
N
8
9
CF3
CF3
R
S
k, l (for 11 and 13)
j, k, l (for 12)
NH
O
X
5 7
,
9
, or
Y
O
O
CF3
OH
HO
O
No. R
X
Y
yield
10 =
O
O
11 CF3
N
CH
6% (2 steps)
8% (3 steps)
17% (2 steps)
OTHP
12
13
H
H
CH CH
N
N
Scheme 1. Reagents and conditions: (a) toluene, reflux, 97%; (b) 1 N NaOH, 1,4-dioxane, 60 °C, 94%; (c) DPPA, Et3N, t-BuOH, rt to reflux, 40%; (d) TFA, CH2Cl2, rt, quantitative
yield; (e) n-BuLi/THF, then 3-trifluoromethylbenzaldehyde, 97%; (f) TMS–Cl, NaI, MeCN, 87%; (g) Cu(NO3)2, Ac2O, 83%; (h) thiosemicarbazide, BOP reagent, Et3N, THF, (i)
CH3SO3H, toluene, reflux, 48% (two steps); (j) Zn Powder, 1 N HCl/2-PrOH; (k) HATU, 10, Et3N, DMA, rt to 70–80 °C; (l) 1 N HCl, MeOH, rt.
Condensation of 2-amino-5-bromothiazole (18) and 3-trifluoro-
methylphenol provided 19. Amide bond formation and deprotec-
tion gave 20. As for the amide linker (W in Fig. 1), synthetic
routes for the methylated amide analogs and the reverse amide
analogs are shown in Scheme 3. Since it was presumed that alkyl-
ation of acylated aminothiazoles would provide undesired alkyl-
ation products (i.e., alkylation at the 3-position of thiazole),12 we
tried to aminate the 2-bromothiazole (22) with (4-methoxyben-
zyl)methylamine. The synthesis was initiated with diazonium for-
mation and subsequent bromination on the aminothiazole (21)
under Sandmeyer reaction conditions gave the 2-bromothiazole
(22). Palladium catalyzed amination with (4-methoxyben-
zyl)methylamine produced 23. Deprotection of the PMB group in
TFA, HATU mediated amide bond formation and subsequent THP
deprotection gave the N-methylated thiazole analog (24). Synthe-
sis of the reverse amide analog (29) was initiated with alkylation
of 2-methoxy-5-nitrophenol (25) with 2-bromoethanol, followed
by TBS protection, and reduction of the nitro group to provide
the aniline (26) in 70% yield over three steps. The isocyanate
(27), prepared from 26 and triphosgene, was immediately reacted
O
O
Br
S
S
S
a-c
d-e
NH
O
NH
O
NH2
N
O
N
O
16
N
CF3
O
CF3
14
15
OH
OH
S
N
f
NH
O
O
CF3
O
17
OH
O
S
O
S
NH
O
Br
g
d-e
NH2
S
N
O
N
NH2
N
O
CF3
19
CF3
18
20
OH
Scheme 2. Reagents and conditions: (a) n-BuLi/THF, then 3-trifluoromethylbenzaldehyde, À78 to À5 °C, 68%; (b) Dess–Martin periodinane, CH2Cl2, 71%; (c) TFA, CH2Cl2, 87%;
(d) HATU, 10, Et3N, DMA, rt to 80 °C; (e) 1 N HCl, MeOH, rt, (16; 54% over two steps), (20, 38% over two steps); (f) NaBH4, MeOH/THF, 0 °C to rt, 48%; (g) NaH, 3-
trifluoromethylphenol, THF/DMF (1:9), 39%.

4162
Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
S
S
S
a
b
N
Br
NH2
PMB
N
N
N
CF3
23
CF3
21
22
CF3
S
c-e
N
O
O
N
OH
CF3
24
O
H2N
O
O
i
OCN
O
O
O2N
OH
O
f-h
OTBS
OTBS
25
26
27
O
S
S
N
Li
k-l
j
22
HN
O
N
OH
CF3
CF3
O
29
28
Scheme 3. Reagents and conditions: (a) tert-butyl nitrite, CuBr2, CH3CN, rt, 23%; (b) (4-methoxybenzyl)methylamine, K3PO4, Pd(OCOCF3)2, P(t-Bu)3, toluene, 80 °C, 85%; (c)
TFA, rt; (d) HATU, 10, Et3N, DMA, rt to 80 °C; (e) 1 N HCl, MeOH, rt, 18% (three steps); (f) 2-bromoethanol, K2CO3, CH3CN, reflux, 79%; (g) TBS–Cl, imidazole, DMF, rt; (h) H2, Pd/
C, THF, rt, 88% (two steps); (i) triphosgene, Et3N, CH2Cl2, rt; (j) n-BuLi, THF, À78 °C; (k) THF, À78 °C to rt; (l) 1 N HCl, MeOH, rt, 22% (four steps).
with lithiated thiazole (28), which was freshly prepared from 22.
After acidic deprotection, 29 was obtained.
The SCD-1 inhibitory activity13 of the compounds prepared in
Scheme 1 is summarized in Table 1. The data suggest that the
Table 5
nitrogen atom at the 3-position of the thiazole core should play a
pivotal role in causing powerful SCD-1 inhibition because the
2-benzylthiazole core (11) and the thiophene core (12) exhibited
significant decrease in SCD-1 inhibition. The sulfur atom in the
core also turned out to be very important because the 5-benzylox-
azole analog was >100 times weaker in SCD-1 inhibition than the
Evaluation of 4-amino(–NR2R3) analogs
S
NH
O
R2
R3
N
N
F3C
37
O
OH
a
a
a
R2
R3
No.
IC50 (nM) mouse
IC50 (nM) human
IC50 (nM) human
cell
D
9
D
9
Table 4
N
Modification of 4-methoxy group (R)
S
NH
N
R
F3C
O
OH
37a
37b
37c
37d
37e
37f
7
15
4
23
5
N
N
O
a
a
a
No.
R
IC50 (nM) mouse IC50 (nM) human IC50 (nM) human
D
9
D
9
cell
4
1a OMe
2
3
15
36a
2
3
2
22
6
NH
N
0.4
6
0.04
2
O
O
O
N
36b
36c
36d
0.6
12
11
44
O
NTb
22
>9999
16
NTb
NTb
NH
O
8
119
NTb\
O
O
OH
NH
S
>9999
NTb
O
O
a
a
Values are the geometric means of at least two experiments.
NT—not tested.
Values are the geometric means of at least two experiments.
NT—not tested.
b
b

Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
4163
O
O
OTHP
R1
Et
yield (4 steps)
43%
33%
No.
OH a-d
F
O
HO
HO
31a
OR1
31b -(CH2)2-OCH3
31
31c -(CH2)2-N(CH3)2 33%
30
O
O
b, d
OH
OH
O
EtO
HO
OTHP
58%
OTHP
O
32
31d
O
O
O
O
b, e
f, d
OH
O
HO
OTHP
MeO
O
OTHP
R2
quantitative
N
NO2
NH2
35
33
34
R3
No. R2
35a
35b Et
35c
R3
Me
Et
yield (2 steps)
O
Me
86%
23%
50%
g, d
97%
O
N
HO
OTHP
OTHP
34
34
H
Et
O
35d
O
No. R2
R3
H
H
35e
yield (2 steps)
h, d (for
i, d (for
)
O
HO
35e
35f
Ac
Ms
48%
66%
R2
35f
)
N
R3
Scheme 4. Reagents and conditions: (a) H2SO4, ROH (R = Et for 31a, R = Me for 31b and 31c), reflux; (b)2-(2-bromoethoxy)tetrahydro-2H-pyran, K2CO3, CH3CN, reflux; (c)
NaOEt, EtOH, reflux (for 31a); or R1OH, t-BuOK, 100 °C (for 31b and 31c); (d) 1–2 N NaOH; (e) H2, Pd/C, EtOAc, rt; (f) HCHO (for 35a) or CH3CHO (for 35b and 35c), NaBH3CN,
AcOH, THF; (g) bis(2-bromoethyl)ether, K2CO3, NaI, DMF; (h) acetic anhydride, Et3N, THF; (i) MsCl, pyridine.
S
a, b
S
NH
O
OR1
NH2
31a~31d
21 =
N
36
N
O
CF3
CF3
OH
yield (2 steps)
No.
36a
R1
Et
40%
47%
23%
39%
36b -(CH2)2OCH3
36c
-(CH2)2N(CH3)2
36d -(CH2)2OH
R2
N
S
a, b
NH
35a~35f
R3
N
O
O
CF3
37
OH
No. R2
37a Me
37b Et
R3
Me
Et
yield (2 steps)
58%
47%
36%
14%
23%
76%
37c
H
Et
37d -(CH2)2O(CH2)2-
37e Ac
37f
H
H
Ms
Scheme 5. Reagents and conditions: (a) 21, HATU, Et3N, DMA, rt to 70 °C; (b) 1 N HCl, MeOH, rt to 50 °C.

4164
Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
Table 6
PK parameters of 37b and 37c in C57BL/6J micea
S
NH
O
R
N
N
F3C
O
37
OH
No.
R
PK profilesa (iv, 5 mg/kg)
PK profilesa (po, 20 mg/kg)
CIb (mL/min/kg)
Vdb (L/kg)
AUC(0–8
(l
g h/mL)
Cmax
t1/2
Tmax
AUC(0–8
(lg h/mL)
h)
F (%)
b
b
b
b
b
t1/2 (h)
h)
37b
37c
Et
H
0.7
1.1
83
21
2.7
1.2
1.0
4.0
0.8
1.8
1.5
1.4
0.8
1.0
2.0
4.3
51
27
a
A dose of each compound was either intravenously (5 mg/kg, DMA/Tween80/saline 10/10/80) injected into the tail vein of C57BL/6J mice (n = 2) or orally (20 mg/kg, 0.5%
MC, n = 3) administered using an intubation tube. Plasma samples (20 lL) were collected up to 8 h after intravenous or oral administration. The plasma concentrations of the
compounds were determined by LC/MS.
b
Values are the geometric means of at least two experiments.
corresponding 5-benzylthiazole analog (data not shown). The thia-
diazole core (13) demonstrated comparable enzymatic SCD-1
inhibitory activity (IC50 (human) = 8 nM) and selectivity toward
those in 37a–37f (Table 5). As indicated in the SAR of the 4-alkoxy
analogs, a variety of alkyl amines were tolerated, dimethylamino
(37a), diethylamino (37b), and morpholino (37d) retained strong
inhibitory activity against SCD-1. Remarkably, ethylamino (37c)15
was found to be the most active SCD-1 inhibitor in this series, with
IC50 (human) = 0.04 nM. In the case of acylated and sulfonylated
analogs, acetlylamino (37e) retained potency for both murine
and human SCD-1, while methylsulfonylamino analog (37f)
showed only marginal inhibitory activity.
D
6 isozyme of the desaturase (<5% inhibition at 10 lM) as the 5-
benzylthiazole core (1a). The pharmacokinetic profiles of 1a and
13 are compared in Table 2. 13 exhibited only marginal plasma
exposure (AUC = 0.4 lg h/mL) and bioavailability (F = 2%) after oral
administration, indicating that the thiazole core is important not
only for potency but also for desirable pharmacokinetics.
The results of the linker modification are summarized in
Table 3. As for linker V, modifications of methylene in 1a to car-
bonyl (16), carbinol (17), and ether (20), resulted in a significant
decrease of activity, indicating that the methylene linker in 1a is
crucial for robust SCD-1 inhibition. In the case of amide linker W,
methylation of the amide proton (24) caused a >100-fold loss of
activity, whereas the reverse amide (29) retained weaker SCD-1
inhibitory activity (IC50 (human) = 32 nM). At this point, we were
quite convinced that the N-(5-benzylthiazol-2-yl) benzamide sys-
tem is a critical structural requirement for the development of po-
tent SCD-1 inhibitors. In the next phase of the optimization, we
turned to the right-hand portion of the lead compound.
The synthetic procedures for the compounds in Tables 4 and 5
are outlined in Schemes 4 and 5.7 Synthesis of the 4-alkoxy analog
was initiated by esterification of 4-fluoro-3-hydroxybenzoic acid
(30), followed by alkylation with 2-(2-bromoethoxy)tetrahydro-
2H-pyran, an SN2Ar reaction with a corresponding alcohol and sub-
sequent saponification to provide 31a–31c. Compound 31d was
prepared by dialkylation of 32 and saponification. To prepare 4-
amino derivatives, the commercially available 3-hydroxy-4-nitro-
benzoic acid methyl ester (33) was alkylated with 2-(2-bromoeth-
For the analysis of the in vivo efficacy of SCD-1 inhibitors, we
took note that Attie and co-workers reported that the hepatic tri-
glyceride levels of mice on a very low-fat diet increased by
240%.16 We assumed that the SCD-1 activity in the liver of these
mice was very high and were interested in the inhibitory effect
of the most potent SCD-1 inhibitor (37c) and its structurally re-
lated analog (37b) against the liver SCD-1 in C57BL/6J mice on a
non-fat diet. The inhibitory activity was determined by measuring
the ratio of [14C] stearate and [14C] oleate in the liver.17 The dose at
which 50% of the conversion is inhibited is described as ID50. As
shown in Tables 6 and 7, both compounds showed similar PK pro-
files and in vivo activity. A combination of strong enzymatic inhib-
itory activity (IC50 (mouse) = 0.4 nM) and good oral exposure right
after oral administration (Cmax = 1.8 lg/mL and Tmax = 1 h) of 37c
contributed to strong potency in liver SCD-1 inhibition (ID50 (2–
3 h) = 0.8 mg/kg). The decrease in potency at 6–7 h (ID50 = 2 mg/
kg) could be attributed to the relatively short plasma half-life
(t1/2 = 1.4 h) and fast clearance (Cl = 21 mL/min/kg). While showing
about 10-fold weaker enzymatic SCD-1 inhibition and relatively
lower plasma exposure, 37b demonstrated potency equal to that
of 37c in liver SCD-1 inhibition in mice. It is assumed that the con-
centration of 37b in the liver was higher than that of 37c, though
no data on their hepatic concentrations were available at this
oxy)tetrahydro-2H-pyran and reduced under
a
hydrogen
atmosphere to generate aniline (34). The aniline was alkylated
with a corresponding aldehyde in the presence of NaBH3CN and
saponified to provide 35a–35c. Preparation of morpholino (35d),
acetylamino (35e), and methylsulfonylamino (35f) are also de-
picted in Scheme 4. As shown in Scheme 5, the 4-alkoxy or 4-ami-
no benzoic acids thus prepared were condensed with
aminothiazole 21 in the presence of HATU. Subsequent THP depro-
tection gave 4-alkoxy analogs (36a–36d) and 4-amino series (37a–
37f).
Table 7
The liver SCD-1 inhibition by 37b and 37c in C57BL/6J mice on a non-fat dieta
S
NH
O
R
N
N
F3C
As shown in Table 4, elongation of the methoxy to ethoxy in 36a
retained strong potency (IC50 (human) = 2 nM).13,14 The methoxy-
ethoxy analog (36b) also displayed potency equal to that of lead
compound 1a, while the terminal dimethylamino (36c) completely
lost activity and the hydroxyethoxy group at the 4-position (36d)
resulted in a slight decrease in activity. It is fair to say there is a ste-
ric tolerance at this position and that lipophilic groups are pre-
ferred. The alkoxy groups were replaced with amines such as
O
37
OH
ID50 (mg/kg)a
No.
R
At 2–3 h
At 6–7 h
37b
37c
Et
H
1.0
0.8
2.0
2.0
a
Values are the geometric means of at least two experiments.

Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
4165
5. Attie, A. D.; Krauss, R. M.; Gray-Keller, M. P.; Brownlie, A.; Miyazaki, M.;
Kastelein, J. J.; Lusis, A. J.; Stalenhoef, A. F. H.; Stoehr, J. P.; Hayden, M. R.;
Ntambi, J. M. J. Lipid Res. 2002, 43, 1899. 3.
Plasma lipid desaturation index
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
6. Uto, Y.; Ogata, T.; Harada, J.; Kiyotsuka, Y.; Ueno, Y.; Miyazawa, Y.; Kurata, H.;
Deguchi, T.; Watanabe, N.; Takagi, T.; Wakimoto, S.; Okuyama, R.; Abe, M.;
Kurikawa, N.; Kawamura, S.; Yamato, M.;Osumi, J. Bioorg. Med. Chem. Lett. 2009,
19, in press.
7. All the final compounds were characterized by 1H NMR and mass spectroscopy.
8. Plouvier, B.; Bailly, C.; Housin, R.; Henichart, J.-P. Heterocycles 1991, 32, 693.
9. Stoner, E. J.; Cothron, D. A.; Balmer, M. K.; Roden, B. A. Tetrahedron 1995, 41,
11043.
10. (a) Compound 13 was prepared as follows: 5-(3-Trifluoromethylbenzyl)-
[1,3,4]thiadiazol-2-ylamine (9):
A mixture of (a,a,a-trifluoro-m-tolyl)acetic
acid (8, 1.15 g, 6.12 mmol), thiosemicarbazide (1.15 g, 12.6 mmol), BOP
reagent (3.24 g, 7.33 mmol), and Et3N (1.7 mL, 12 mmol) in anhydrous THF
(20 mL) was stirred at room temperature for 20 h. The reaction mixture was
concentrated, diluted with H2O and extracted with EtOAc (twice). The
combined organic layers were washed with saturated aqueous NaHCO3 and
brine, dried (Na2SO4), and concentrated. The off-white solid thus obtained
(1.72 g) was mixed with methanesulfonic acid (0.48 mL, 7.4 mmol) and
toluene (20 mL). The suspension was heated to reflux for 4 h and cooled to
room temperature. The reaction mixture was diluted with EtOAc, washed with
saturated aqueous NaHCO3 (Â2) and brine, dried (Na2SO4), and concentrated.
Chromatography of the residue on SiO2 (CH2Cl2/MeOH 30:1–10:1) gave
Vehicle
0.3 mg/kg
1 mg/kg
37c
3 mg/kg
Figure 2. Plasma desaturation index lowering effect of the treatment with SCD-1
inhibitor 37c for 7 days (q.d.) in C57BL/6J mice fed with a non-fat diet.
760 mg (48%) of
(2-Hydroxyethoxy)-4-methoxy-N-[5-(3-trifluoromethylbenzyl)-[1,3,4]thiadiazol-
2-yl]benzamide (13). solution of (100 mg, 0.386 mmol), 3-(2-
hydroxyethoxy)-4-methoxybenzoic acid (83 mg, 0.39 mmol), HATU (178 mg,
0.468 mmol), and Et3N (90 L, 0.65 mmol) in DMA (2 mL) was stirred at room
9 as a
white solid. MS (ESI) m/z: 260 (M+H)+.; (b) 3-
point. For multiple dosing studies of SCD-1 inhibitors, 37c was
tested in a 7-day efficacy study using C57BL/6J mice on a non-fat
diet.18 The desaturation index, calculated as the ratio of
C18:1N9(cis)/C18:0, was used as an in vivo biomarker. After
once-daily administration for 7 days, 37c dose-dependently
reduced the desaturation index, with a 65% reduction at 3 mg/kg
(Fig. 2). In the preliminary analysis, we did not observe any abnor-
malities in the skin or eyes of the C57BL/6J mice at 3 mg/kg (Cuta-
neous abnormalities and narrow eye fissure have been reported in
studies on SCD-1 deficient mice19). We assume that the balanced
combination of the strong potency and short plasma half life of
37c resulted in pharmacological efficacy in vivo and may be bene-
ficial in ameliorating adverse events. The liver can be a key tissue
for metabolizing xenobiotics, and orally administered drugs can
obtain significant concentrations in this tissue. Since SCD-1 is ex-
pressed in the liver, significant systemic exposure of SCD-1 inhib-
A
9
l
temperature for four days. After work-up, the crude product was purified by
chromatography on SiO2 (CH2Cl2/MeOH 20:1) and subsequent recrystallization
(2-PrOH) to give 29 mg (17%) of 13 as a white solid:1H NMR(400 MHz, DMSO-
d6): d 12.8 (1H, br s), 7.76 (3H, d, J = 9.8 Hz), 7.68 (2H, dd, J = 8.1 and 8.1 Hz),
7.61 (1H, dd, J = 7.8 and 7.8 Hz), 7.62 (1H, d, J = 7.9 Hz), 4.88 (1H, t, J = 5.2 Hz),
4.52 (2H, s), 4.07 (2H, t, J = 4.9 Hz), 3.85 (3H, s), 3.76 (2H, dt, J = 4.8 and 4.9 Hz);
MS (ESI) m/z: 454 (M+H)+.
11. Kuo, G.-H.; Wang, A.; Emanuel, S.; DeAngelis, A.; Zhang, R.; Connolly, P. J.;
Murray, W. V.; Gruninger, R. H.; Sechler, J.; Fuentes-Pesquera, A.; Johnson, D.;
Middleton, S. A.; Jolliffe, L.; Chen, X. J. Med. Chem. 2005, 48, 1886.
12. Manaka, A.; Ishii, T.; Takahashi, K.; Sato, M. Tetrahedron Lett. 2005, 46, 419.
13. Desaturase enzymatic assay: The SCD-1 activity was determined by measuring
the conversion of stearate to oleate. In each reaction tube, test compounds
were preincubated with 10
lL microsomes for 10 min at room temperature.
The SCD-1 reaction was started by the addition of 40
lL
of mixture
a
containing 250 mM sucrose, 150 mM KCl, 40 mM NaF, 5 mM MgCl2, 100 mM
sodium phosphate, pH7.4, 1 mM ATP, 1.5 mM reduced glutathione, 0.06 mM
itors may not be necessary to realize
a pharmacodynamic
reduced coenzyme A, 0.33 mM nicotinamide, 1.25 mM NADH and 0.01
14C] stearate. After 60 min incubation at 37 °C, the reaction was stopped by
adding 50 methanol containing 10% KOH and then the mixture was
saponified at 80 °C for 30 min. The free fatty acids in the reaction were
protonated by the addition of 5 N HCl (15 L) and extracted with 100 L ethyl
acetate. 30 L of the ethyl acetate extracts of each reaction was charged to an
AgNO3–TLC plate (20 Â 20 cm LK5D plates, 150 Å pore diameter, 250 m thick)
and differentiated in a solvent consisting of chloroform/methanol/acetate/
lCi
response. While this is a preliminary speculation, the relatively
short plasma half-life of 37c may help to accomplish favorable tis-
sue selectivity (liver over eyes or skin). Histopathological analysis
of the key tissues (eyes, skin, and liver) of the C57BL/6J mice after
a 7-day treatment with the SCD-1 inhibitor 37c is currently in pro-
gress and will be reported elsewhere along with more details about
the pharmacological studies of the 3-(2-hydroxyethoxy)-N-(5-ben-
zylthiazol-2-yl)benzamide-based SCD-1 inhibitors.
[
lL
l
l
l
l
water (90:8:1:0.8).
[ [
14C] stearate and 14C] oleate were quantified with
BAS2500 (Fujifilm) and SCD-1 activity was determined as the ratio of [14C]
oleate to [14C] stearate. The IC50 values were calculated by linear regression
using the straight line portions of the concentration–response curve. To
measure the delta-6 desaturase activity, [14C] linolenic acid was used as the
substrate and the delta-6 desaturase activity was determined as the ratio of
In summary, we prepared and assayed compounds derived from
1, which was identified as a potent SCD-1 inhibitor in the preced-
ing article. SAR studies of this lead compound proved that the N-(5-
[
14C] C18:3n À 3 to [14C] C18:4n À 3.
benzylthiazol-2-yl)benzamide system is
a critical structural
14. A 293A cell-based desaturase assay was performed in a 96-well plate. Human
SCD-1 gene was cloned into the expression vector pCMV-script (Stratagene).
293A cells stably expressing human SCD-1 were obtained by transfecting the
requirement for the development of potent SCD-1 inhibitors. Fur-
ther delineation of SAR studies of this lead compound, especially
in the right-hand portion, resulted in the identification of 4-ethyl-
amino-3-(2-hydroxyethoxy)-N-[5-(3-trifluoromethylbenzyl)thia-
zol-2-yl] benzamide (37c) with sub nano molar IC50 in both murine
and human SCD-1 inhibitory assays. Compound 37c demonstrated
a dose-dependent decrease in the plasma desaturation index in
C57BL/6J mice on a non-fat diet after once-daily 7-day oral admin-
istration. Further optimization and pharmacological and toxicolog-
ical evaluation of this series of compounds will be reported in due
course.
expression vector to 293 cells and selected with G418. 293A cells in 100 lL
media (DMEM + 10% FBS) were seeded to each well of the 96-well plate and
grown overnight to be confluent. The cells were preincubated with test
compound in fresh media for 30 min, after which 10
0.1
Ci [14C] stearate was added to each well and incubated for another 4 h.
Then the cells in each well were washed with cold PBS and the cellular lipids
were saponified directly by adding 100 L of 5% KOH in methanol/H2O (1:1).
lL media containing
l
l
The samples were processed as described for the SCD-1 enzymatic assay to
determine the SCD-1 activity by quantifying the ratio of [14C] oleate to [14C]
stearate.
15. (a) Compound 37c was prepared as follows: 4-Amino-3-[2-(tetrahydropyran-
2-yloxy)ethoxy]benzoic acid methyl ester (34). A suspension of 3-hydroxy-
4-nitro-benzoic acid ethyl ester (33, 5.02 g, 25.5 mmol), 2-(2-bromo-
ethoxy)tetrahydro-2H-pyran (5.8 mL, 38 mmol), K2CO3 (7.04 g, 50.9 mmol) in
DMA (70 mL) was heated at 80 °C for 5 h, cooled to room temperature, and
extracted with EtOAc. The organic layer was washed with H2O (Â4) and brine,
dried (Na2SO4), and concentrated. Chromatography of the residue on SiO2
(hexanes/EtOAc 17:3–1:1) to give 8.50 g of the alkylated compound as a yellow
oil: 1H NMR(400MHz,CDCl3): d 7.84 (1H, s), 7.83 (1H, d, J = 9.8 Hz), 7.71 (1H, d,
J = 9.7 Hz), 4.73 (1H, t, J = 3.3 Hz), 4.42–4.34 (2H, m), 4.16–4.08 (1H, m), 3.97
(3H, s), 3.92–3.83 (2H, m), 3.57–3.52 (1H, m), 1.86–1.71 (2H, m), 1.65–1.51
(4H, m). A suspension of the yellow oil (8.50 g, 25.5 mmol) and Pd/C (10 wt %,
References and notes
1. Dobrzyn, A.; Ntambi, J. M. Obesity Rev. 2005, 6, 169.
2. Ntambi, J. M.; Miyazaki, M. Curr. Opin. Lipidol. 2003, 14, 255.
3. (a) Miyazaki, M.; Kim, Y.-C.; Gray-Keller, M. P.; Attie, A. D.; Ntambi, J. M. J. Biol.
Chem. 2000, 275, 30132; (b) Flowers, M. T.; Ntambi, J. M. Curr. Opin. Lipidol.
2008, 19, 248.
4. Dobrzyn, P.; Dobrzyn, A. Drug Development Res. 2006, 67, 643.

4166
Y. Uto et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4159–4166
1.28 g) in EtOAc (100 mL) was treated with H2 gas for 3 h. The reaction mixture
(1H, d, J = 8.6 Hz), 5.78 (1H, t, J = 5.6 Hz), 4.98 (1H, t, J = 6.3 Hz), 4.23 (2H, s),
4.03 (2H, t, J = 4.7 Hz), 3.79–3.75 (2H, m), 3.24–3.17 (2H, m), 1.19 (3H, t,
J = 7.2 Hz). MS (ESI) m/z: 466 (M+H)+.
was filtered through a plug of Celite, concentrated, and dried in vacuo to give
7.78 g (quantitative yield) of 34 as a colorless oil: MS(ESI) m/z: 296 (M+H)+.; (b)
4-Ethylamino-3-[2-(tetrahydropyran-2-yloxy)ethoxy]benzoic acid (35c). To
16. Flowers, M. T.; Groen, A. K.; Oler, A. T.; Keller, M. P.; Choi, Y. J.; Schueler, K. L.;
Richards, O. C.; Lan, H.; Miyazaki, M.; Kuipers, F.; Kendziorski, C. M.; Ntambi, J.
M.; Attie, A. E. J. Lipid Res. 2006, 47, 2668.
a
solution of 34 (4.45 g, 15.3 mmol) in THF/MeOH (2:1, 90 mL) were
successively added acetic acid (0.88 mL, 15 mmol), acetaldehyde (4.3 mL,
77 mmol) and sodium cyanoborohydride (2.89 g, 46.0 mmol) at 0 °C. To
complete the reaction, additional amounts of the reagents were added and
the reaction was heated to 60 °C. The reaction mixture was concentrated,
diluted with EtOAc and saturated aqueous NaHCO3. The separated organic
layer was washed with water and brine, dried (Na2SO4), and concentrated.
Chromatography of the residue on SiO2 (CH2Cl2/EtOAc, and then CH2Cl2/
MeOH) gave 3.19 g (64%) of the monoalkylated compound as a colorless oil and
1.63 g (30%) of the dialkylated compound as a colorless oil. To a solution of the
monoalkylated compound (514 mg, 1.59 mmol) in 1,4-dioxane (10 mL) was
added 1 N NaOH (2.4 mL) at room temperature. The reaction mixture was
heated to reflux for 16 h. To complete the conversion, an additional amount of
1 N NaOH was added during the reaction. The reaction mixture was
concentrated, and diluted with EtOAc and citric acid (aq). The separated
organic layer was washed with H2O and brine, dried (Na2SO4), and
concentrated. The residue was triturated in hexanes/iPr2O, collected by
filtration, and dried in vacuo to give 385 mg (78%) of 35c as an off-white
solid: MS (ESI) m/z: 310 (M+H)+.; (c) 4-Ethylamino-3-(2-hydroxyethoxy)-N-[5-
(3-trifluoromethylbenzyl)thiazol-2-yl]benzamide (37c). A solution of 21 (95
mg, 0.37 mmol), 35c (114 mg, 0.369 mmol), HATU (154 mg, 0.405 mmol), and
Et3N (0.10 mL, 0.74 mmol) in DMA (6 mL) was heated at 70 °C for 3 days. The
reaction mixture was diluted with EtOAc and citric acid (aq). The separated
organic layer was washed with H2O (Â3) and brine, dried (Na2SO4), and
17. 9-Week-old C57BL/6J mice grown with a normal chow diet were fed a non-fat
diet (Research Diets, Inc., D05052506) for 7 days. SCD-1 inhibitors were
administered to the mice (n = 2) 2 h or 6 h prior to the administration of [14C]
stearate. Then the mice were injected ip with 5 mL/kg of 20
stearate solution in saline containing 2% BSA, resulting in a bolus amount of
100
Ci/kg. One hour after the injection of [14C] stearate, the mice were
sacrificed and their livers were removed and quickly frozen in liquid nitrogen.
of
l
Ci/mL [14C]
l
The livers were homogenized in 9Â volume of cold PBS, and 250
lL
homogenate was mixed with an equal volume of methanol containing 10%
KOH. Then the samples were processed as described for the SCD-1 enzymatic
assay to determine the SCD-1 activity by quantifying the ratio of [14C] oleate to
[
14C] stearate. The dose at which 50% of SCD-1 activity is inhibited is described
as ID50
.
18. 9-Week-old male C57BL6J mice grown with a normal chow diet were fed with
a non-fat diet (Research Diets, Inc., D05052506) for 7 days. Compound 37c was
administered daily at doses of 0.3, 1, and 3 mg/kg to male C57BL6J mice on a
non-fat diet for 7 days in the evening by oral gavage in propylene glycol/Tween
80 (4/1) formulation (dosing vehicle). The animals were allowed free access to
the non-fat diet and water throughout the study. After the seventh dose, the
animals were sacrificed in the morning and the blood serum was assayed for
the desaturation index as follows. 30 lL serum was mixed with 4 mL 0.5 N
KOH in methanol and saponified at 100 °C for 30 min. The free fatty acids in the
reaction were protonated by the addition of 2 mL of 1 N HCl and extracted with
3 mL of n-hexane. Two milliliters of the hexane extracts were dried up and
dissolved with 1 mL of BF3–methanol and esterified at 100 °C for 15 min. The
samples were mixed with 1 mL of H2O and extracted with 1 mL n-hexane. One
microliter of the final hexane extracts was loaded to a gas chromatogram to
determine the stearate and oleate content. The desaturation index was
calculated as the ratio of oleate to stearate.
concentrated. Chromatography of the residue on SiO2 (KP-NH, 40–75
100A, Biotage, hexanes/CH2Cl2 3:7 to 0:1) gave 122 mg (60%) of the amide
compound as pale yellow oil. To solution of the amide (122 mg,
lm,
a
a
0.221 mmol) in MeOH (10 mL) was added 1 N HCl (0.55 mL). The reaction
mixture was heated at 50 °C for 1.5 h and concentrated. The residue was
diluted with CH2Cl2, iPr2O, hexanes and saturated aqueous NaHCO3. The
resulting precipitate was collected by filtration, washed with iPr2O and H2O,
and dried in vacuo to give 62 mg (61%) of 37c as
a
white solid: 1H
19. Miyazaki, M.; Man, W. C.; Ntambi, J. M. J. Nutr. 2001, 131, 2260.
NMR(400MHz, DMSO-d6): d 12.1 (1H, s), 7.67–7.56 (6H, m), 7.32 (1H, s), 6.58