Discovery of orally active carboxylic acid derivatives of 2-phenyl-5-trifluoromethyloxazole-4-carboxamide as potent diacylglycerol acyltransferase-1 inhibitors for the potential treatment of obesity and diabetes

Article abstract of DOI:10.1021/jm101580m

Diacylglycerol acyltransferase-1 (DGAT-1) is the enzyme that catalyzes the final and committed step of triglyceride formation, namely, the acylation of diacylglycerol with acyl coenzyme A. DGAT-1 deficient mice demonstrate resistance to weight gain on hig

Full text of DOI:10.1021/jm101580m

ARTICLE
pubs.acs.org/jmc
Discovery of Orally Active Carboxylic Acid Derivatives of
2-Phenyl-5-trifluoromethyloxazole-4-carboxamide as Potent
Diacylglycerol Acyltransferase-1 Inhibitors for the Potential
Treatment of Obesity and Diabetes
Yimin Qian,*,† Stanley J. Wertheimer,^ Mushtaq Ahmad,† Adrian Wai-Hing Cheung,† Fariborz Firooznia,†
Matthew M. Hamilton,† Stuart Hayden,† Shiming Li,† Nicholas Marcopulos,† Lee McDermott,† Jenny Tan,†
Weiya Yun,† Liang Guo,‡ Anjula Pamidimukkala,§ Yingsi Chen,|| Kuo-Sen Huang,|| Gwendolyn B. Ramsey,^
Toni Whittard,^ Karin Conde-Knape,^ Rebecca Taub,# Cristina M. Rondinone,^ Jefferson Tilley,† and
David Bolin†
†Department of Discovery Chemistry, ‡Non-Clinical Drug Safety, §Metabolism and Pharmacokinetics, Discovery Technologies, and
^Metabolic and Vascular Diseases, Hoffmann-La Roche Inc., 340 Kingsland Street, Nutley, New Jersey 07110, United States
#Via Pharmaceuticals, Inc., 101 College Road East, Princeton, New Jersey 08540, United States
S Supporting Information
b
ABSTRACT: Diacylglycerol acyltransferase-1 (DGAT-1) is the enzyme that
catalyzes the final and committed step of triglyceride formation, namely, the
acylation of diacylglycerol with acyl coenzyme A. DGAT-1 deficient mice demon-
strate resistance to weight gain on high fat diet, improved insulin sensitivity, and
reduced liver triglyceride content. Inhibition of DGAT-1 thus represents a potential
novel approach for the treatment of obesity, dyslipidemia, and metabolic syndrome.
In this communication, we report the identification of the lead structure 6 and our
lead optimization efforts culminating in the discovery of potent, selective, and orally efficacious carboxylic acid derivatives of
2-phenyl-5-trifluoromethyloxazole-4-carboxamides. In particular, compound 29 (DGAT-1 enzyme assay, IC50 = 57 nM; CHO-K1
cell triglyceride formation assay, EC50 = 0.5 μM) demonstrated dose dependent inhibition of weight gain in diet induced obese
(DIO) rats (0.3, 1, and 3 mg/kg, PO, qd) during a 21-day efficacy study. Furthermore, compound 29 demonstrated improved
glucose tolerance determined by an oral glucose tolerance test (OGTT).
’ INTRODUCTION
believe that inhibiting DGAT-1 might represent a novel ap-
proach for the treatment of obesity and the improvement of
insulin sensitivity.8,9
Triglycerides constitute the major form of energy storage in
eukaryotic organisms. In humans, excessive triglyceride accumu-
lation is often associated with diseases such as obesity, diabetes,
and steatohepatitis. The key enzyme catalyzing triglyceride
synthesis is diacylglycerol O-acyltransferase (DGAT). Two
DGAT isozymes, DGAT-1 and DGAT-2, each encoded by a
distinct gene family, play important roles in triglyceride
synthesis.1ꢀ3 The exact roles played by DGAT-1 and DGAT-2
in triglyceride synthesis are currently under investigation.
DGAT-1 deficient mice are resistant to diet induced obesity
and have decreased adiposity, in part through increased energy
expenditure.4 In addition, DGAT-1 null mice are reported with
improved insulin and leptin sensitivity.5 On the other hand, mice
lacking DGAT-2 are lipopenic and die soon after birth.6 More
recently it was reported that DGAT-1 deficient mice with
expression of DGAT-1 only in the intestine completely reversed
the resistance to diet induced obesity and hepatic steatosis, which
suggested that the beneficial effects of DGAT-1 inhibition might
be mainly due to inhibition of the enzyme in the intestine.7
Encouraged by the results from DGAT-1ꢀ/ꢀ mice, we and others
The rapid progress in the search for DGAT-1 inhibitors over
the past decade is reflected by the large number of potent
compounds disclosed in both the patent and peer reviewed
literature.10ꢀ15 Many of the reported compounds are car-
boxylic acids, and typical structures are listed in Figure 1.
Compound 1 was reported as a potent DGAT-1 inhibitor
(IC50 = 44 nM) and showed efficacy in reducing body weight
gain in diet induced obese C57BL/6 mice.10 Compounds 2 and
3 are examples of DGAT-1 inhibitors disclosed in patent
applications from Bayer and Japan Tobacco/Amgen, respec-
tively.11,12 Compound 3 was recently shown to be efficacious in
a mouse DIO model.12 Compound 4 was reported to be a
potent DGAT-1 inhibitor (IC50 = 7 nM) with oral efficacy by
an Abbott research team.13 On the basis of lead structure 3,
an AstraZeneca research team reported orally efficacious
Received: December 13, 2010
Published: March 17, 2011
r
2011 American Chemical Society
2433
dx.doi.org/10.1021/jm101580m J. Med. Chem. 2011, 54, 2433–2446
|

Journal of Medicinal Chemistry
ARTICLE
Figure 1. Selected examples of DGAT-1 inhibitors from the literature.
pig Langendorff heart model indicated that 6 induced dispersion
of repolarization and caused QT prolongation at low concentra-
tions consistent with the in vitro hERG data. In light of these
results, our lead optimization strategy focused on maintaining
DGAT-1 enzyme and cellular inhibition potency while reducing
the hERG channel interaction. In addition to resolving hERG
issues, we also sought to further improve the pharmacokinetic
properties of 6 to achieve efficacy at lower doses. Since carboxylic
acids are rarely associated with hERG liabilities,18 we investigated
the effects of incorporation of carboxylic acid moieties into our
lead structure. This strategy led to elimination of the hERG
inhibition while maintaining both in vitro and in vivo potency.
Herein we report the details of this work and the identification of
the orally active and potent DGAT-1 inhibitor 29, with improved
pharmacokinetic properties and significantly decreased hERG
liabilities.
Figure 2. DGAT-1 inhibitor hit molecule 6A, lead structure 6, and
optimized compound 29.
pyrimidinooxazinylbicyclooctaneacetic acid derivative 5.14 Re-
cently, scientists from Pfizer disclosed PF-04620110 as a
development candidate.15
’ CHEMISTRY
Through the screening of our compound library, we identified
a series of hydrazides as DGAT-1 inhibitors. The screening hit
6A (Figure 2) showed good potency in DGAT-1 enzyme assay
(IC50 = 230 nM). However, the high clearance of this class of
hydrazides in vivo prompted us to investigate further structural
modifications.16 During the lead identification process, we
replaced the hydrazide moiety with heterocycles and discovered
a series of 2-phenyl-5-trifluoromethyloxazole-4-carboxamides as
potent DGAT-1 inhibitors.16 In particular, when assayed for
human DGAT-1 enzyme inhibition (potency expressed as IC50)
and human DGAT-1 inhibition in CHOK1 cells (potency
expressed as EC50), compound 6 (IC50 = 38 nM, EC50 = 0.66
μM, >100-fold selectivity over DGAT-2 and ACAT) emerged as
a promising lead and showed desirable pharmacokinetic proper-
ties in rats (CL = 13.5 (mL/min)/kg, iv at 5 mg/kg) with good
oral bioavailability (F = 77%). When dosed orally in DIO rats for
3 weeks, 6 demonstrated good efficacy in reducing body weight
gain (4.3% of body weight gain reduction at 25 mg/kg, po, q.
d.).17 Unfortunately, compound 6 was also shown to block the
human ether-a-go-go-related gene (hERG) encoded potassium
channel (IC20 = 0.2 μM), and further assessment in the guinea
Compounds 6ꢀ14 were prepared as described in Scheme 1.
The commercially available 2-phenyl-5-trifluoromethyloxazole-
4-carboxylic acid or 2-methyl-5-trifluoromethyloxazole-4-car-
boxylic acid were coupled with the appropriate aromatic or
heteroaromatic amines in the presence of bromotripyrrolidino-
phosphonium hexafluorophosphate. The heteroaromatic amines
were in turn synthesized by the SNAr reaction of halogen-
substituted nitroaromatics with amines, followed by the reduc-
tion of the nitro group (Scheme 1).
To prepare amides and carbamates 15ꢀ24, the substituted
piperidine or piperazine hydrochlorides (31, 32, 33, 34, and 35)
were used as the key intermediates. The hydrochloride salt 31
was prepared through the amide coupling reaction of 2-phenyl-5-
trifluoromethyloxazole-4-carboxylic acid with the corresponding
amine followed by the deprotection of the tert-butoxycarbonyl
group (Scheme 2). Intermediates 32, 33, and 34 were prepared
via a similar method (see the Experimental Section). For the
synthesis of intermediate 35, 2-bromo-5-nitropyiridine was first
coupled with the pinacolatoborate of N-Boc-tetrahydropyridine
under Suzuki coupling conditions, followed by hydrogenation
2434
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Preparation of Compounds 6ꢀ14a
a Reaction conditions: (a) (i) triethylamine, THF, room temp or heat; (ii) H2 (50 psi), Pd/C (cat.), THF/MeOH; (b) PyBroP, triethylamine, DMF;
(c) methyl iodide, NaH, DMF; (d) azetidin-3-ol hydrochloride, K2CO3, NMP, heat; (e) (i) bromoacetic acid tert-butyl ester, NaH, THF/DMF; (ii) H2
(50 psi), Pd/C (cat.); (f) 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid, PyBroP, DMF; (g) CF3COOH, CH2Cl2.
and subsequent amide coupling reaction (Scheme 2). Treatment
of the intermediates 31ꢀ35 with the appropriate acyl chloride or
alkoxycarbonyl chlorides afforded compounds 15ꢀ24.
The synthesis of trans-1,4-cyclohexane carboxylic acid deriva-
tives is illustrated in Scheme 3. The intermediates required for
the preparation of compounds 25ꢀ30, namely, trans- and cis-1,4-
cyclohaxane dicarboxylic acid monobenzyl ester, were synthe-
sized through monoacyl chloride formation and quenching with
benzyl alcohol (Scheme 3).
solubility (mp 217 ꢀC, solubility of <1 μg/mL). On the other
hand, increasing the polarity significantly reduced hERG channel
interaction (compound 12), albeit accompanied by a 6-fold loss
of DGAT-1 inhibition potency relative to 6. Significant improve-
ments in hERG liability came from amide methylation or R1
substitution (compounds 7 and 8), which may be explained by a
conformational change or an increase in polarity. Unfortunately,
both compounds 7 and 8 lost DGAT-1 potency. Finally, the
incorporation of a carboxylic acid almost completely alleviated
hERG blockade despite a modest loss of DGAT-1 enzyme
inhibition (compound 14). This result is consistent with the
observation that the hERG channel tends to bind strongly with
positively charged molecules such as tertiary amines. The loss of
cellular potency in compound 14 was presumably due to its poor
cell permeability (cpKa = 2.9, log D = ꢀ0.20).
’ RESULTS AND DISCUSSION
Compounds prepared in the present work (6ꢀ30) were
assayed for human DGAT-1 enzyme inhibition (expressed as
IC50), human DGAT-1 inhibition in CHOK1 cells (expressed as
EC50), and hERG channel inhibition (expressed as IC20). As
shown in Table 1, 6 displayed significant hERG inhibition despite
the lack of a basic amine in the structure. Since hERG inhibition
can induce torsade de pointes (TdP) in human,19,20 prevention
of hERG potassium channel blocking has thus been adopted as a
routine drug toxicity derisking strategy. To overcome the hERG
blocking issue, we first looked at the structural fragments in
compound 6 with the goal of understanding their contributions
to hERG channel inhibition. The results are summarized in
Table 1 below.
The above results from compounds 6ꢀ10 indicated that
multiple factors contributed to hERG inhibition. Changing the
pyridine moiety in 6 to pyrimidine or benzene slightly decreased
hERG inhibition (compounds 9 and 10 compared to 6), while
cyclization of N-methyl-N-methoxyethylamine A to azetidine B
(compound 11) did not affect hERG inhibition significantly.
Interestingly, reducing the basicity of the nitrogen of the R3 by
placing fluorine atoms in the β position (compound 13) did not
have any significant impact on hERG inhibition either. To our
surprise, increased DGAT-1 inhibition in compound 13 was
accompanied by a loss in cellular activity, possibly due to the poor
Although we were able to decrease hERG inhibition by adding
a carboxylic acid moiety (compound 14), the same carboxylic
acid group also resulted in a loss of DGAT-1 cellular activity. We
next extended our investigation of the R3 substitution in order to
improve DGAT-1 potency. The results are summarized in
Table 2. Rigidification of the R3 substitution through the
incorporation of a piperazine or a piperidine moiety and capping
the distal nitrogens as amides or carbamates all resulted in potent
DGAT-1 inhibition in both enzyme and cellular assays (compo-
unds 15ꢀ24, Table 2). When compared to 6, compound 17
demonstrated 6-fold potency improvement in the DGAT-1
cellular assay and a 4-fold reduction in hERG channel activity.
With the exception of compound 15, all piperazine/piperidine
derivatives (16ꢀ24) demonstrated reduced hERG inhibition. Of
particular interest were compounds 23 and 24, which displayed
the most significant improvement in hERG inhibition. However,
when 23 and 24 were further evaluated in mice, they both
displayed very low oral exposure after dosing as an aqueous
suspension to DIO mice at 25 mg/kg, with compound 23
showing plasma Cmax of 76 ng/mL at Tmax of 1 h and compound
24 showing Cmax < 10 ng/mL.
2435
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
Scheme 2. Preparation of Piperazine/Piperidine Substituted Amides and Carbamatesa
a Reaction conditions: (a) N-Boc-piperazine, potassium carbonate, acetonitrile, 80 ꢀC; (b) (i) H2 (50 psi), Pd/C (cat.), THF/MeOH; (ii) 2-phenyl-5-
trifluoromethyloxazole-4-carboxylic acid, PyBroP, triethylamine, DMF; (iii) HCl in ether, CH2Cl2, MeOH; (c) cyclopropylcarbonyl chloride or
methoxycarbonyl chloride or ethoxycarbonyl chloride, triethylamine, CH2Cl2; (d) PdCl2dppf (cat.), potassium carbonate, toluene/ethanol, 100 ꢀC,
microwave.
Scheme 3. Synthesis of trans-1,4-Cyclohexane Carboxylic Acid Derivativesa
a Reaction conditions: (a) (i) oxalyl chloride, cat. DMF, THF; (ii) benzyl alcohol, THF; (b) (i) oxalyl chloride, CH2Cl2; (ii) Et3N, CH2Cl2; (iii) H2, Pd/
C, MeOH/THF
Therefore, to improve the pharmacokinetic properties of
compound 24 while maintaining potent DGAT-1 inhibition in
enzyme and cellular assays, we derivatized the piperazine/
piperidine groups with six-membered rings functionalized with
carboxylic acids. Results from these modifications are summar-
ized in Table 3. Only compounds with good potency in DGAT-1
enzyme and cellular assays (IC50 < 0.5 μM or EC50 < 1.0 μM)
were tested for hERG channel inhibition.
As illustrated in Table 3, incorporation of a carboxylic acid
significantly reduced hERG channel inhibition. When compared
to neutral compounds (Table 2), carboxylic acid derivatives
showed reduced cellular activity (compare 17 with 25).
2436
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
Table 1. Substitution Effects on DGAT-1 Activity and hERG Inhibitiona
compd
R1
R2
R3
X
Y
DGAT-1 IC50 (μM)
DGAT-1 EC50 (μM)
hERG IC20 (μM) or % inhibition at x (μM)
6
phenyl
methyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
phenyl
H
H
A
A
A
A
A
B
C
D
E
N
CH
CH
CH
N
0.038
15.1
0.661
>25
0.2
7
N
37% at 30
34% at 30
1.5
8
methyl
H
N
24.7
>25
9
N
0.048
0.325
0.068
0.237
0.016
0.202
0.534
1.803
0.228
5.66
10
11
12
13
14
H
CH
N
CH
CH
N
18% at 1
0.37
H
H
N
10.0
H
N
CH
CH
25.0
17% at 1
12% at 30
H
N
>25
a Mean values from at least three experiments.
Table 2. StructureꢀActivity Relationships of Substituted Piperazine and Piperidinea
compd
X
Y
M
R
DGAT-1 IC50 (μM)
DGAT-1 EC50 (μM)
hERG IC20 (μM) or % inhibition at x (μM)
15
16
17
18
19
20
21
22
23
24
N
CH
CH
CH
N
N
cyclopropyl
methoxy
0.059
0.028
0.041
0.071
0.128
0.084
0.136
0.165
0.067
0.084
0.138
0.216
0.093
0.41
0.18
N
N
1.94
N
N
ethoxy
0.80
N
N
methoxy
28% at 1.0
1.73
N
CH
CH
CH
CH
CH
CH
CH
CH
N
cyclopropyl
methoxy
0.378
0.219
0.165
0.190
0.362
0.371
N
16% at 1.0
35% at 3.0
16% @ 3.0
13% at 3.0
41% at 30.0
CH
CH
CH
CH
cyclopropyl
methoxy
N
CH
cyclopropyl
CH
methoxy
a Mean values from at least three experiments.
Incorporation of heteroaromatics such as pyridine into the
structures also led to reduced cellular potency (compare 26 with
29), possibly due to the lowered hydrophobicity (compound 29
clogP = 3.1, compound 26 clogP = 1.9). The literature carboxylic
acid 3 displayed similar difference between enzyme and cellular
potency in our assay despite its higher lipophilicity (clogP = 4.6).
Of these analogues, 29 exhibited the best combination of cellular
potency in the DGAT assay and minimal hERG liability.
Thus, compound 29 was selected for further investigation.
When assayed for its selectivity toward human DGAT-1 over
human DGAT-2 (IC50 > 100 μM) and ACAT (a mixture of
ACAT1 and ACAT2, IC50 = 68.4 μM), it showed over 4000-fold
and 2000-fold selectivity, respectively. When assayed for rat
DGAT-1 inhibition, it showed comparable potency to human
DGAT-1. Compound 29 also displayed desirable physicochem-
ical properties (solubility of 15.4 μg/mL; log D = 2.34; pKa =
5.78; caco-2 permeability of 245 ꢁ 10ꢀ7 cm/s; free fraction in
human serum protein binding of 0.7%; human hepatocyte
clearance of 3.14 (mL/min)/kg; rat hepatocyte clearance of
5.77 (mL/min)/kg; P450 inhibition for CYP2C9, 2C19, 2D6,
and 3A4, IC50 > 50 μM, no CYP3A4 time dependent inactivation
and no P-glycoprotein inhibition).
We next determined the pharmacokinetic (PK) profiles of 29
in rats and dogs. The PK results are summarized in Table 4. As
observed with other carboxylic acids in this class of molecules,
compound 29 has a low volume of distribution and low clearance.
2437
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
Table 3. DGAT-1 and hERG Inhibition Potency of Cyclohexane 1,4-Dicarboxylic Acid Derivativesa
compd
X
M
cis or trans
DGAT-1 IC50 (μM)
DGAT-1 EC50 (μM)
hERG IC20 (μM) or % inhibition at x (μM)
25
26
27
28
29
30
3
N
N
trans
trans
trans
cis
0.035
0.128
0.277
0.077
0.057
0.082
0.113
1.764
2.363
1.070
0.365
0.493
0.220
1.436
ND
N
CH
N
ND
CH
CH
CH
CH
5% at 1.0
>10
N
CH
CH
trans
cis
30.0
21% at 3.0
ND
a Mean value from at least three experiments. ND: not determined.
Table 4. Pharmacokinetic (PK) Properties of Compound 29
in Rat and Doga
PK parameter
rat
dog
iv dose (mg/kg)
Vd (L/kg)
5
5
0.27
0.38
8.6
0.38
1.3
5.3
10
CL ((mL/min)/kg)
T1/2 (h)
po dose (mg/kg)
10
AUCext (μM h)
566.8
34.8
12
77.9
4.7
2
3
Cmax (μM)
Tmax (h)
F (%)
73
33
a All numbers are the mean average (n = 3).
Figure 3. Effects of compound 29 on body weight change of DIO rats
(orally dosed at 0.3, 1, and 3 mg/kg) for 3 weeks.
The observed low clearance could be due to the high microsomal
stability and high protein binding. Interestingly, when the
corresponding cis-isomer compound 30 was orally dosed in
mice, both the cis-isomer and the trans-isomer compound 29
were detected in plasma in an almost 1:1 ratio resulting from
isomerization. On the other hand, when the trans-isomer was
dosed in mice, no in vivo isomerization was observed.
We also measured daily food intake on days 1ꢀ3, 8, 9, 15, and 16.
As shown in Figure 4, treatment with 29 for 3 weeks caused a dose
dependent decrease in food intake (17%, 19%, and 27% at doses of
0.3, 1, and 3 mg/kg, respectively, p < 0.01). These data suggested
that part of the effect of compound 29 on body weight may be due
to reduced food intake.
On the basis of its desirable pharmacokinetic properties,
compound 29 was selected for 1 week efficacy screening in male
DIO rats. When dosed at 3 mg/kg (po, q.d.), it caused a
significant reduction (4.0%) in body weight gain. At the end of
the study, rats were challenged with corn oil (4 h fast, 5 mL/kg),
and a reduction of plasma triglyceride levels by 85% was observed
at the 4 h time point compared to the vehicle-treated group (p =
0.015). Encouraged by these data, we initiated studies aimed at
determining the dose response of 29 on decreasing body weight
gain in DIO rats in a more chronic setting. Therefore, compound
29 was orally dosed at 0.3, 1.0, and 3.0 mg/kg once a day for 3
weeks. As shown in Figure 3, rats in the vehicle-treated group fed
a high fat diet for 3 weeks gained significant weight. However, in
rats treated with 29, we observed a dose dependent decrease in
body weight gain (by 5%, 6%, and 8% at doses of 0.3, 1.0, and 3.0
mg/kg, respectively, p < 0.01). This dose dependent efficacy
correlated well with the measured exposure levels (Cmax of 2.0,
4.8, and 15.1 μM at doses of 0.3, 1.0, and 3.0 mg/kg, respectively).
To further investigate the potential beneficial effects of
compound 29 on glucose homeostasis, an oral glucose tolerance
test (OGTT) was carried out on DIO rats on the last day of the
3-week efficacy studies. Our data indicated that 29 significantly
lowered the overall glucose excursion by 15% at a dose of 3 mg/
kg (Figure 5). Although lower doses (0.3 and 1 mg/kg) also
showed trends in lowering glucose excursion, data analysis
suggested that they did not reach significance. We also measured
insulin levels on day 21 prior to OGTT and compared them with
the insulin levels measured on day 1 prior to the treatment by
compound 29. Our data indicated a significant decrease in basal
insulin levels after DIO rats were treated with compound 29 (3
mg/kg, data in Supporting Information), while the basal glucose
level was comparable to that of the vehicle group (Figure 5).
When insulin levels were measured at the 30 min time point
during OGTT, we did not observe significant changes in insulin
levels between the vehicle group and the groups treated with 29
2438
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
Figure 4. Measured food intake in DIO rats dosed with compound 29 (0.3, 1, and 3 mg/kg). All measurements are daily food intake (10 rats per dose
group). Values are the average for each group.
was accompanied by a reduction in food intake. The exact
mechanism of how selective DGAT-1 inhibitors can affect food
intake is not clear. One possible explanation is that inhibiting
DGAT-1 can affect fatty acid composition, distribution, and
sensing mechanisms, which may have an impact on food intake.
In fact, it was recently reported that gastric emptying was
significantly delayed and incretin hormones GLP-1 and PYY
were elevated in DGAT-1 deficient mice compared to the wild
type after oral triglyceride loading.21 GLP-1 and PYY are intes-
tine incretin hormones that can affect food intake,22,23 which
could explain the observed anorectic effects of DGAT-1 inhibi-
tors. In our efficacy studies, plasma GLP-1 and PYY were not
measured. Further support for a local effect of DGAT in the gut
comes from the recent report that DGAT-1 deficient mice with
specific expression of DGAT-1 in intestine completely reversed
the resistance to diet induced obesity and hepatic steatosis
Figure 5. Effects of compound 29 on oral glucose tolerance test
normally associated with DGAT-1 knockout, suggesting that
(OGTT). Glucose was measured at 0, 30, 60, and 120 min after oral
the effects of DGAT-1 on body weight gain are largely con-
glucose treatment at 2 g/kg (10 rats each dose group). All numbers are
tributed by the inhibition of the enzyme in the intestine.7 The
the average: (/) p < 0.05 for glucose AUC, Anova contrast test.
effects of DGAT inhibition on food intake in rats treated with
compounds of completely different chemical structures have also
been reported in literature.24
at the three doses. However, the glucose level was significantly
To further assess selectivity of compound 29, a panel of 80
receptors and ion channels were selected for a radiolabeled
binding assay (data in Supporting Information). Compound
29 showed a clean profile (IC50 > 10 μM) for 77 out of 80
targets in the panel, including potential targets that can affect
food intake and body weight, such as cannabinoid (CB1),
cholecystokinin (CCK), dopamine (D2), ghrelin, histamine,
serotonin, NPY, nuclear receptor PPAR, and purinergic receptor
(PX). The only three targets from the panel that showed strong
interactions with compound 29 are the adenosine receptors
(A2A and A3) and the dopamine receptor (D3). It is unlikely
lower for the group treated with 29 (3 mg/kg, Figure 5). These
data suggested a potential insulin sensitizing effect of compound
29 in DIO rats. The encouraging OGTT result in terms of
reduced glucose excursion in addition to the effects on insulin
levels and on lowering body weight gain suggested that DGAT-1
inhibitor 29 could provide potential benefits for treating obesity
and diabetes.
Since the publication of mouse DGAT-1ꢀ/ꢀ data, significant
efforts have been carried out in the industry to identify potent
small molecule DGAT-1 inhibitors.8 In our studies, the reduction
of body weight gain in DIO rats by a selective DGAT-1 inhibitor
2439
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
that activities at the three receptors in the panel are related to the
food intake effects observed with compound 29, as another very
similar carboxylic acid analogue (structure not shown) with a
clean profile of binding to A2A, A3, and D3 demonstrated almost
identical effects on food intake and body weight in DIO rats.
Furthermore, when total brain concentration of compound 29
was measured, it showed brain to plasma ratio of only 3%. We
also did not observe any changes regarding locomotor activity,
clinical signs, or clinical chemistry in the 3-week efficacy study.
In summary, we have identified a novel class of 2-phenyl-5-
trifluoromethyloxazole-4-carboxamides, which demonstrated
good in vitro DGAT-1 inhibition and high selectivity for
DGAT-1, without affecting the hERG potassium channel. One
of the described compounds, 29, showed good PK properties and
significant efficacy in obesity models with beneficial effects on
glucose homeostasis. Compound 29 was evaluated in several in
vitro toxicity assays (cyp inhibition, protein covalent binding,
micronucleus test, and Ames test) and displayed no liabilities.
Our results provided further evidence that selective small mole-
cule DGAT-1 inhibitors could have beneficial effects on meta-
bolic syndromes.
bicarbonate solution and dried over sodium sulfate. Solvents were
evaporated, and the residue was purified through ISCO flash column
chromatography (linear gradient ethyl acetate in hexanes) to give
2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid {6-[(2-methox-
yethyl)methylamino]pyridin-3-yl}amide (828 mg, 73%). LCꢀMS calcd
for C20H19F3N4O3 (m/e) 420.14, obsd 421.0 (M þ H). 1H NMR (300
MHz, CDCl3) δ 8.73 (br s, 1H), 8.24 (br s, 1H), 8.02ꢀ8.19 (m, 3H),
7.48ꢀ7.65 (m, 3H), 6.54 (d, J = 8.15 Hz, 1H), 3.76 (br s, 2H), 3.60 (br s,
2H), 3.37 (s, 3H), 3.10 (s, 3H).
2-Methyl-5-trifluoromethyloxazole-4-carboxylic Acid
{6-[(2-Methoxyethyl)methylamino]pyridin-3-yl}amide (7).
N2-(2-Methoxy-ethyl)-N2-methylpyridine-2,5-diamine (56 mg, 0.309
mmol), 2-methyl-5-trifluoromethyloxazole-4-carboxylic acid (50 mg,
0.256 mmol), and Et3N (0.198 mL, 0.768 mmol) were treated with
BOP (119 mg. 0.268 mmol) in 2 mL of DMF. Following workup, the
crude material was purified by flash chromatography to yield 2-methyl-5-
trifluoromethyloxazole-4-carboxylic acid {6-[(2-methoxyethyl)methy-
lamino]pyridin-3-yl}amide as a lyophilized powder (28 mg, 30%).
LCꢀMS calcd for C15H17F3N4O3 (m/e) 358.318, obsd 359.2 (M þ
H). 1H NMR (300 MHz, DMSO-d6) δ 10.38 (s, 1H), 8.37 (d, J = 2.6 Hz,
1H), 7.85 (dd, J = 2.6, 9.2 Hz, 1H), 6.60 (d, J = 9.2 Hz, 1H), 3.60ꢀ3.69
(m, 2H), 3.42ꢀ3.50 (m, 2H), 3.22 (s, 3H), 2.98 (s, 3H), 2.58 (s, 3H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
{6-[(2-Methoxyethyl)methylamino]pyridin-3-yl}methylamide
(8). 2-Phenyl-5-trifluoromethyloxazole-4-carboxylic acid {6-[(2-methox-
yethyl)methylamino]pyridin-3-yl}amide (6) (50 mg, 0.119 mmol) was
treated with sodium hydride (60% in oil, 10.4 mg, 0.238 mmol) and
methyl iodide (22 μL, 0.357 mmol) in 5 mL THF at room temperature
for 16 h. Following workup, the crude material was purified by flash
chromatography to yield 2-phenyl-5-trifluoromethyloxazole-4-car-
boxylic acid {6-[(2-methoxyethyl)methylamino]pyridin-3-yl}methyl-
amide as a solid (29 mg, 56%). LCꢀMS calcd for C21H17F3N4O3 (m/
e) 434.416, obsd 435.2 (M þ H). 1H NMR (300 MHz, DMSO-d6) δ
7.82ꢀ7.92 (m, 3H), 7.49ꢀ7.67 (m, 3H), 7.40 (dd, J = 2.72, 9.06 Hz,
1H), 6.56 (d, J = 9.06 Hz, 1H), 3.53ꢀ3.62 (m, 2H), 3.33ꢀ3.37 (m,
2H), 3.31 (s, 3H), 3.08 (s, 3H), 2.90 (s, 3H).
’ EXPERIMENTAL SECTION
All reactions were carried out under an argon atmosphere. All solvents
were purchased from commercial sources without further drying.
Melting points were taken on a Buchi melting point B-545 apparatus
without correction. 1H NMR spectra were recorded with Mercury 300
and Unityplus 400 MHz spectrometers. The reference peaks are as
follows: CDCl3 δ ppm 7.27, CD3OD δ ppm 3.31, DMSO-d6
δ
ppm 2.50. All compounds were analyzed by LC/MS (liquid chroma-
tography/mass spectrometry) using Waters ZQ mass detector and
Waters LC system. Ionization was generally achieved via electron spray
(ES). The LC fraction detection consisted of both diode array detector
and evaporative light scattering detector, and all tested compounds had
purity greater than 98%. Combustion elemental analysis was applied to
compound 29. Thin layer chromatography was run on silica coated glass
plate and was visualized under 254 nm UV or iodine vapor. Preparative
high pressure liquid chromatography (HPLC) was carried out using a
Rainin HPLC employing a 41.4 mm ꢁ 300 mm, 8 μm, Dynamax C-18
column at a flow of 49 mL/min employing a gradient of acetonitrile/
water (each containing 0.75% TFA) typically from 5% to 95% acetoni-
trile over 35ꢀ40 min.
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
{2-[(2-Methoxyethyl)methylamino]pyrimidin-5-yl}amide (9).
This compound was prepared by the same method as described for the
preparation of compound (6) by using 2-chloro-5-nitropyrimidine.
LCꢀMS calcd for C19H18F3N5O3 (m/e) 421.14, obsd 422.0 (M þ H).
1H NMR (300 MHz, CDCl3) δ 8.57ꢀ8.65 (m, 3H), 8.09ꢀ8.17 (m, 2H),
7.50ꢀ7.65 (m, 3H), 3.80ꢀ3.88 (m, 2H), 3.57ꢀ3.65 (m, 2H), 3.36 (s,
3H), 3.23 (s, 3H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
{6-[(2-Methoxyethyl)methylamino]pyridin-3-yl}amide (6).
To a mixture of 2-chloro-5-nitropyridine (1.58 g, 9.96 mmol) and
2-methoxyethylmethylamine (1.78 g, 20 mmol) in THF (30 mL) was
added triethylamine (1.38 mL), and the solution was stirred at room
temperature overnight. The resulting mixture was filtered, and the
filtrate was evaporated. The residue was extracted with ethyl acetate
and water. The organic layer was washed with brine and dried over
sodium sulfate. Solvents were evaporated and the residue was dried
under vacuum overnight to give pure desired compound 2-meth-
oxyethylmethyl-(5-nitropyridin-2-yl)amine (2.13 g, 100%). This oily
material (570 mg, 2.70 mmol) was dissolved in methanol (30 mL), and
palladium on carbon (5%, 35 mg) was added. The mixture was
hydrogenated at 30 psi for 1 h and then filtered. Solvents were
evaporated, and the residue was dried under vacuum overnight. The
resulting material was combined with 2-phenyl-5-trifluoromethyloxa-
zole-4-carboxylic acid (694 mg, 2.70 mmol) and bromotripyrrolidino-
phosphonium hexafluorophosphate (1.26 g, 2.70 mmol) in DMF
(15 mL) containing triethylamine (0.8 mL). The solution was stirred
overnight, and solvents were evaporated. The residue was extracted with
dichloromethane and water. The organic layer was washed with sodium
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
{4-[(2-Methoxyethyl)methylamino]phenyl}amide (10). This
compound was prepared by the same method as described for the
preparation of compound (6) by using 1-fluoro-4-nitrobenzene. LCꢀ
MS calcd for C21H20F3N3O3 (m/e) 419.15, obsd 420.0 (M þ H). 1H
NMR (300 MHz, CDCl3) δ 8.79 (br s, 1H), 8.08ꢀ8.20 (m, 2H),
7.49ꢀ7.66 (m, 5H), 6.73 (d, J = 8.75 Hz, 2H), 3.46ꢀ3.64 (m, 4H), 3.37
(s, 3H), 2.99 (s, 3H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
[6-(3-Ethoxyazetidin-1-yl)pyridin-3-yl]amide (11). 2-(3-Ethox-
yazetidin-1-yl)-5-nitropyridine (59 mg, 0.264 mmol) was hydrogenated
and reacted with 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid
(74 mg, 0.29 mmol), Et3N (186 μL, 1.32 mmol), and BOP (122 mg,
0.277 mmol). Following workup, the crude material was purified by flash
chromatography to yield 2-phenyl-5-trifluoromethyloxazole-4-car-
boxylic acid [6-(3-ethoxyazetidin-1-yl)pyridin-3-yl]amide as a solid
(79 mg, 83%). LCꢀMS calcd for C21H19F3N4O3 (m/e) 432.41, obsd
1
433.1 (M þ H). H NMR (300 MHz, DMSO-d6) δ 10.46 (s, 1H),
8.39ꢀ8.44 (m, 1H), 8.13 (d, J = 7.55 Hz, 2H), 7.91 (dd, J = 2.26,
8.91 Hz, 1H), 7.56ꢀ7.72 (m, 3H), 6.43 (d, J = 9.06 Hz, 1H), 4.34ꢀ4.43
2440
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
(m, 1H), 4.08ꢀ4.18 (m, 2H), 3.71 (dd, J = 4.08, 8.60 Hz, 2H), 3.43 (q,
J = 6.90 Hz, 2H), 1.12 (t, J = 6.90 Hz, 3H).
was stirred at room temperature for 2 h. The reaction mixture was diluted
with 100 mL of ethyl acetate and extracted with 3 ꢁ 25 mL of saturated
sodium bicarbonate solution, 25 mL of brine, dried over magnesium
sulfate, filtered, and evaporated to give a dark oil. The crude material was
applied to a short sílica plug and eluted with hexanes and ether. The ether
fraction was evaporated to yield [1-(5-nitropyridin-2-yl)azetidin-3-ylox-
y]acetic acid tert-butyl ester as a bright yellow solid (210 mg, 76%).
LCꢀMS calcd for C14H19N3O5 (m/e) 309.32, obsd 310.1 (ES, M þ H).
[1-(5-Nitropyridin-2-yl)azetidin-3-yloxy]acetic acid tert-butyl ester
(210 mg, 0.679 mmol) was hydrogenated at 40 psi in methanol. The
mixture was filtered, and solvents were evaporated. The residue was
treated with 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (183
mg, 0.71 mmol), DIPEA (355 μL, 2.03 mmol), and BOP (315 mg, 0.74
mmol) in 3 mL of DMF. After overnight stirring, solvents were
evaporated and the residue was extracted with ethyl acetate and water.
The crude product was purified through flash column chromatography,
eluting with ethyl acetate in hexanes to give (1-{5-[(2-phenyl-5-trifluor-
omethyloxazole-4-carbonyl)amino]pyridin-2-yl}azetidin-3-yloxy)acetic
acid tert-butyl ester as a light brown solid (169 mg, 48%). LCꢀMS calcd
for C25H25F3N4O5 (m/e) 518.5, obsd 519.1 (M þ H).
(S)-2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid-
[2-(3-Hydroxypyrrolidin-1-yl)pyrimidin-5-yl]amide (12). To a
solution of 2-chloro-5-nitropyrimidine (638 mg, 4 mmol) in THF
(15 mL) cooled to 0 ꢀC was added (S)-3-hydroxypyrrolidine (392 mg,
4.5 mmol) and triethylamine (1.2 mL). The mixture was stirred at room
temperature overnight. Solids were filtered out, and the solution was
concentrated. The residue was extracted with ethyl acetate and water. The
organic layer was washed with dilute aqueous citric acid solution. After the
evaporation of solvents, the residue was treated with ether/petroleum
ether (1:1 ratio) and the white solid was filtered to give N-(5-nitropyr-
imidin-2-yl)pyrrolidin-3-ol (690 mg, 82.1%). LCꢀMS calcd for
C8H10N4O3 (m/e) 210.19, obsd 211.0 (ES, M þ H). This compound
(150 mg, 0.714 mmol) was dissolved in a mixture of THF and methanol,
and the solution was hydrogenated at 40 psi in the presence of a catalytic
amount of palladium on carbon. The mixture was filtered, and solvents
were removed. The residue was treated with 2-phenyl-5-trifluoromethy-
loxazole-4-carboxylic acid (183.6 mg, 0.714 mmol) and PyBroP (333 mg,
0.714 mmol) in DMF containing triethylamine (0.2 mL). After overnight
stirring, the mixture was worked up and the crude mixture was purified
through flash column chromatography, eluting with ethyl acetate in hexa-
nes to give a white solid as (S)-2-phenyl-5-trifluoromethyloxazole-4-
carboxylic acid [2-(3-hydroxypyrrolidin-1-yl)pyrimidin-5-yl]amide (178
mg, 59.4%). LCꢀMS calcd for C19H16F3N5O3 (m/e) 419.36, obsd 420.1
(ES, M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 2.00ꢀ2.27 (m, 2 H),
3.61ꢀ3.87 (m, 4 H), 4.56ꢀ4.74 (m, 1 H), 7.47ꢀ7.69 (m, 3 H), 8.06ꢀ8.20
(m, 2 H), 8.66 (s, 3 H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
[6-(3,3-Difluoroazetidin-1-yl)pyridin-3-yl]amide (13). 3,3-Di-
fluoroazetidine hydrochloride (259 mg, 2 mmol) was mixed with
2-chloro-5-nitropyridine (317 mg, 2 mmol) in THF (8 mL) containing
DIPEA (1.4 mL). The mixture was heated in a microwave at 120 ꢀC for
20 min. The resulting mixture was kept at room temperature overnight
and then filtered. The solution was evaporated to dryness, and the solid
material was triturated with methanol. Filtration of the mixture provided
2-(3,3-difluoroazetidin-1-yl)-5-nitropyridine (288 mg, 67%). 1H NMR
(300 MHz, CDCl3) δ ppm 4.53 (t, J = 11.6 Hz, 4 H), 6.34 (d, J = 9.1 Hz,
1 H), 8.29 (dd, J = 9.2, 2.6 Hz, 1 H), 9.07 (s, 1 H).
(1-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbo-
nyl)amino]pyridin-2-yl}azetidin-3-yloxy)acetic acid tert-butyl ester
(132 mg, 0.26 mmol) was treated with 8 mL of 97% TFA in water at
room temperature for 1 h. The reaction mixture was evaporated, re-
evaporated from 1 N HCl in CH3CN, and dried under vacuum to yield
(1-{5-[(2-phenyl-5-trifluoromethyloxazole-4-carbonyl)amino]pyridin-
2-yl}azetidin-3-yloxy)acetic acid hydrochloride as an off-white solid
(130 mg, 100%). LCꢀMS calcd for C21H17F3N4O5 (m/e) 462.39, obsd
463.0 (M þ H). 1H NMR (300 MHz, DMSO-d6) δ 10.81 (s, 1H), 8.48
(d, J = 2.11 Hz, 1H), 8.22 (d, J = 9.2 Hz, 1H), 8.13 (d, J = 6.34 Hz, 2H),
7.58ꢀ7.74 (m, 3H), 6.87 (d, J = 9.2 Hz, 1H), 4.56 (m, 1H), 4.33ꢀ4.46
(m, 2H), 4.13 (s, 2H), 4.04ꢀ4.11 (m, 2H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid (6-
Piperazin-1-ylpyridin-3-yl)amide Hydrochloride (31). 4-(5-
Aminopyridin-2-yl)piperazine-1-carboxylic acid tert-butyl ester (834
mg, 3.0 mmol) was mixed with 2-phenyl-5-trifluoromethyloxazole-4-
carboxylic acid (771 mg, 3.0 mmol) and bromotrispyrrolidinopho-
sphonium hexafluorophosphate (1.398 g, 3 mmol) in N,N-dimethylfor-
mamide (20 mL) and methylene chloride (5 mL) containing
triethylamine (0.85 mL). The mixture was stirred at room temperature
overnight, and the solvents were evaporated. The residue was parti-
tioned between ethyl acetate and water. The organic layer was dried over
sodium sulfate, and solvent was evaporated. The residue was triturated
with ethyl acetate and the solid was filtered to give 4-{5-[(2-phenyl-5-
trifluoromethyloxazole-4-carbonyl)amino]pyridin-2-yl}piperazine-1-
carboxylic acid tert-butyl ester (1.09 g, 70.3%). LCꢀMS calcd for
The above material (140 mg, 0.65 mmol) was hydrogenated and then
coupled with 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (163.9
mg, 0.64 mmol) to provide 2-phenyl-5-trifluoromethyloxazole-4-car-
boxylic acid [6-(3,3-difluoroazetidin-1-yl)pyridin-3-yl]amide (132.8 mg,
48.1%) as a white crystalline material. LCꢀMS calcd for C19H13F5N4O2
(m/e) 424.32, obsd 425.0 (ES, M þ H). 1H NMR (300 MHz, CDCl3) δ
ppm 4.39 (t, J = 12.1 Hz, 4 H), 6.45 (d, J = 9.1 Hz, 1 H), 7.50ꢀ7.67 (m, 3
H), 8.15 (d, J = 7.5 Hz, 2 H), 8.24 (dd, J = 8.6, 2.6 Hz, 1 H), 8.31 (s, 1 H),
8.82 (br s, 1 H).
1
C25H26F3N5O4 (m/e) 517.5, obsd 518.1 (M þ H). H NMR (300
MHz, DMSO-d6) δ ppm 1.42 (s, 9 H), 3.44 (d, J = 4.2 Hz, 8 H), 6.90 (d,
J = 9.4 Hz, 1 H), 7.56ꢀ7.74 (m, 3 H), 7.98 (dd, J = 9.2, 2.6 Hz, 1 H), 8.15
(dd, J = 8.0, 1.7 Hz, 2 H), 8.51 (d, J = 2.7 Hz, 1 H), 10.52 (s, 1 H).
4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbo-
(1-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]pyridin-2-yl}azetidin-3-yloxy)acetic Acid Hydro-
chloride (14). 2-Bromo-5-nitropyridine (1.82 g, 8.96 mmol),
azetidin-3-ol hydrochloride (0.98 g, 9.0 mol), and potassium carbo-
nate (3.71 g, 26.8 mmol) in 40 mL of NMP were stirred at 80 ꢀC for
24 h under argon. After cooling, the reaction mixture was diluted with
150 mL of EtOAc, washed with 4 ꢁ 25 mL of brine, dried over magne-
sium sulfate, filtered, and evaporated to furnish an oil. The crude
material was then dissolved in 100 mL of ethyl acetate and eluted
through a plug of sílica gel and evaporated to yield 1-(5-nitropyridin-
2-yl)azetidin-3-ol as a yellow solid (1.0 g, 57%). LCꢀMS calcd for
C8H9N3O3 (m/e) 195.18, obsd 196.0 (ES, M þ H).
nyl)amino]pyridin-2-yl}-piperazine-1-carboxylic acid tert-butyl ester
(300 mg, 0.58 mmol) prepared above was suspended in methylene
chloride (5 mL) and methanol (5 mL). To this mixture was added
hydrogen chloride in ether (4 N, 3 mL). The mixture was stirred at room
temperature overnight. Solvents were evaporated, and the residue was
dried in vacuum. The resulting solid was triturated with dry ether and
then filtered to give a hydrochloride salt of 2-phenyl-5-trifluoromethy-
loxazole-4-carboxylic acid (6-piperazin-1-ylpyridin-3-yl)amide (274 mg,
96.5%). LCꢀMS calcd for the free base C20H18F3N5O2 (m/e) 417.39,
obsd 418.0 (M þ H). 1H NMR (300 MHz, methanol-d4) δ ppm 3.49
(d, J = 4.8 Hz, 4 H), 4.00 (br s, 4 H), 7.53 (d, J = 10.0 Hz, 1 H),
7.57ꢀ7.70 (m, 3 H), 8.20 (d, J = 6.6 Hz, 2 H), 8.42 (d, J = 10.0 Hz, 1 H),
8.85 (s, 1 H).
A mixture of 1-(5-nitropyridin-2-yl)azetidin-3-ol (176 mg, 0.9 mmol)
and sodium hydride (60% in oil, 108 mg, 2.7 mmol) in 3 mL of THF and
3 mL of DMF was stirred in an ice bath under argon for 40 min. tert-Butyl
bromoacetate (199.7 μL, 1.35 mmol) was added, and the resulting mixture
2441
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid (2-
Piperazin-1-ylpyrimidin-5-yl)amide Hydrochloride (32). This
compound was prepared according to the same method as described for
the preparation of compound 31. LCꢀMS calcd for the free base
(d, J = 12.4 Hz, 2 H), 7.37 (d, J = 8.5 Hz, 1 H), 7.54ꢀ7.69 (m, 3 H),
8.11ꢀ8.30 (m, 3 H), 8.86 (d, J = 2.4 Hz, 1 H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
[6-(4-Cyclopropanecarbonylpiperazin-1-yl)pyridin-3-yl]-
amide (15). Compound 31 (40 mg) was suspended in dichloro-
methane (4 mL) containing triethylamine (0.07 mL) and cyclopropyl-
carbonyl chloride (0.01 mL) while the reaction mixture was cooled with
an ice bath. The mixture was stirred at room temperature overnight, and
solvents were evaporated. The residue was extracted with dichloro-
methane and brine. After evaporation of solvents, the residue was
triturated with methanol. The white solid was filtered to give 2-phen-
yl-5-trifluoromethyloxazole-4-carboxylic acid [6-(4-cyclopropanecarbo-
nylpiperazin-1-yl)pyridin-3-yl]amide (30 mg, 76%). LCꢀMS calcd for
1
C19H17F3N6O2 (m/e) 418.14, obsd 419.0 (M þ H). H NMR (300
MHz, DMSO-d6) δ ppm 3.16 (br s, 4 H), 3.78ꢀ4.09 (m, 4 H),
7.46ꢀ7.85 (m, 3 H), 8.14 (dd, J = 7.8, 1.5 Hz, 2 H), 8.80 (s, 2 H),
9.39 (br s, 2 H, NH, HCl), 10.71 (s, 1 H, amide).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid (4-
Piperazin-1-ylphenyl)amide (33). This compound was prepared
according to the same method as described for the preparation of
compound 31. LCꢀMS calcd for C21H19F3N4O2 (m/e) 416.14, obsd
417.1 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 3.05ꢀ3.10 (m, 4
H), 3.11ꢀ3.25 (m, 4 H), 6.90 ꢀ 6.98 (m, 2 H), 7.42ꢀ7.77 (m, 5 H),
8.14 (dd, J = 7.8, 1.5 Hz, 2 H), 8.84 (s, 1 H).
1
C24H22F3N5O3 (m/e) 485.17, obsd 486.1 (M þ H). H NMR (300
MHz, CDCl3) δ ppm 0.82 (dq, J = 7.2, 3.6 Hz, 2 H), 0.95ꢀ1.09 (m,
2 H), 1.69ꢀ1.87 (m, 1 H), 3.51 (d, J = 7.2 Hz, 2 H), 3.68 (br s, 2 H), 3.81
(br s, 4 H), 6.71 (d, J = 9.4 Hz, 1 H), 7.49ꢀ7.67 (m, 3 H), 8.08ꢀ8.27 (m,
3 H), 8.34 (d, J = 2.4 Hz, 1 H), 8.81 (s, 1 H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid (4-
Piperidin-4-ylphenyl)amide (34). This compound was prepared
from 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid and 4-(4-ami-
nophenyl)piperidine-1-carboxylic acid tert-butyl ester. 4-{4-[(2-Phenyl-
5-trifluoromethyloxazole-4-carbonyl)amino]phenyl}piperidine-1-car-
boxylic acid tert-butyl ester (245 mg, 0.475 mmol) was dissolved in
methylene chloride (2 mL) and trifluoroacetic acid (1 mL). The
mixture was stirred at room temperature, and the solvents were
evaporated. The residue was extracted with dichloromethane and
diluted sodium hydroxide solution. The organic layer was washed with
brine and dried over sodium sulfate. Evaporation of solvents gave
2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (4-piperidin-4-
yl-phenyl)amide (183 mg, 92.8%) as a white solid. LCꢀMS calcd
4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]pyridin-2-yl}piperazine-1-carboxylic Acid Methyl Es-
ter (16). Compound 16 was prepared from compound 31 and methyl
chloroformate using the same method as described for the preparation of
compound 15 (75%). LCꢀMS calcd for C22H20F3N5O4 (m/e) 475.15,
obsd 476.1 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 3.56 (d, J =
5.7 Hz, 4 H), 3.61 (br s, 4 H), 3.75 (s, 3 H), 6.71 (d, J = 9.1 Hz, 1 H),
7.50ꢀ7.66 (m, 3 H), 8.08ꢀ8.24 (m, 3 H), 8.32 (d, J = 2.4 Hz, 1 H), 8.81
(s, 1 H).
4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]pyridin-2-yl}piperazine-1-carboxylic Acid Ethyl Ester
(17). Compound 17 was prepared from compound 31 and ethyl
chloroformate using the same method as described for the preparation
of compound 15 (77%). LCꢀMS calcd for C23H22F3N5O4 (m/e) 489.16,
obsd 490.0 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 1.30 (t, J = 7.1
Hz, 3 H), 3.59 (d, J = 11.5 Hz, 8 H), 4.19 (q, J = 6.9 Hz, 2 H), 6.71 (d, J =
9.1 Hz, 1 H), 7.51ꢀ7.66 (m, 3 H), 8.15 (d, J = 8.2Hz, 2 H), 8.21 (d, J = 8.8
Hz, 1 H), 8.32 (d, J = 2.4 Hz, 1 H), 8.81 (s, 1 H).
1
for C22H20F3N3O2 (m/e) 415.15, obsd 416.0. H NMR (300 MHz,
CDCl3) δ ppm 1.55ꢀ1.76 (m, 3 H), 1.85 (d, J = 12.4 Hz, 2 H), 2.64 (t,
J = 12.1 Hz, 1 H), 2.76 (t, J = 11.8 Hz, 2 H), 3.21 (d, J = 11.2 Hz, 2 H),
7.24 (br s, 2 H), 7.49ꢀ7.62 (m, 3 H), 7.68 (d, J = 8.5 Hz, 2H), 8.15 (d,
J = 6.6 Hz, 2H), 8.90 (br s, 1H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
(10,20,30,40,50,60-Hexahydro[2,40]bipyridinyl-5-yl)amide (35).
To a mixture of 2-bromo-5-nitropyridine (0.5 g, 2.46 mmol) and N-
tert-butoxycarbonyl-1,2,3,6-tetrahydropyridine-4-boronic acid pinacol
ester (0.913 g, 2.95 mmol) in toluene (4 mL) and ethanol (1.0 mL)
were added potassium carbonate solution (2M, 2 mL) and PdCl2dppf
(180 mg, 0.246 mmol). The mixture was degassed with argon and heated
at 100 ꢀC in a microwave for 40 min with stirring. Solvents were
evaporated, and the residue was extracted with ethyl acetate. After
evaporation of the solvents, the residue was purified using flash chroma-
tography (eluting with ethyl acetate and hexanes) to give 5-nitro-30,60-
dihydro-20H-[2,40]bipyridinyl-10-carboxylic acid tert-butyl ester as a solid
(400 mg, 53.2%). This solid material (400 mg) was dissolved in methanol
(50 mL) and tetrahydrofuran (10 mL). To this mixture was added 10%
palladium on carbon (100 mg). The mixture was hydrogenated at 50 psi for
2 h. The mixture was filtered and the solvents were evaporated to give
5-amino-30,40,50,60-tetrahydro-20H-[2,40]bipyridinyl-10-carboxylic acid tert-
butyl ester as a white solid (363 mg, 100%). This amine (363 mg) was
coupled with 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid to pro-
vide 5-[(2-phenyl-5-trifluoromethyloxazole-4-carbonyl)amino]-30,40,50,60-
tetrahydro-20H-[2,40]bipyridinyl-10-carboxylic acid tert-butyl ester (535
mg, 79.1%). This amide (535 mg) was dissolved in a mixture of methylene
chloride (35 mL) and trifluoroacetic acid (9 mL). The mixture was stirred
at room temperature for 2 h. Solvents were evaporated, and the residues
were partitioned between methylene chloride and diluted sodium hydro-
xide solution. The organic layer was washed with brine and dried over
sodium sulfate. Evaporation of solvents gave 2-phenyl-5-trifluoromethylox-
azole-4-carboxylic acid (10,20,30,40,50,60-hexahydro[2,40]bipyridinyl-5-yl)-
amide (360 mg, 83.5%) as a solid. LCꢀMS calcd for C21H19F3N4O2
(m/e) 416.15, obsd 417.1 (M þ H). 1H NMR (300 MHz, CD3OD) δ
ppm 1.64ꢀ1.82 (m, 2 H), 1.85ꢀ2.00 (m, 2 H), 2.67ꢀ2.96 (m, 3 H), 3.18
4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)amino]-
pyrimidin-2-yl}piperazine-1-carboxylic Acid Methyl Ester
(18). Compound 18 was prepared from compound 32 and methyl
chloroformate (80.4%). LCꢀMS calcd for C21H19F3N6O4 (m/e)
476.14, obsd 477.0 (M þ H). 1H NMR (300 MHz, CDCl3) δ
ppm 3.52ꢀ3.63 (m, 4 H), 3.76 (s, 3 H), 3.80ꢀ3.93 (m, 4 H),
7.49ꢀ7.68 (m, 3 H), 8.14 (dd, J = 8.0, 1.4 Hz, 2 H), 8.59ꢀ8.78 (m, 3 H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid (10-
Cyclopropanecarbonyl-10,20,30,40,50,60-hexahydro[2,40]bipy-
ridinyl-5-yl)amide (19). To a mixture of compound 35 (50 mg, 0.12
mmol) and cyclopropane carboxylic acid (31 mg, 0.37 mmol) in
dichloromethane (5 mL) were added triethylamine (0.06 mL, 0.41
mmol) and BOP reagent (55.7 mg, 0.13 mmol). The mixture was stirred
at room temperature for 3 h, and solvents were evaporated. The residue
was extracted with ethyl acetate and dried. Solvents were evaporated and
the residue was triturated with ether and petroleum ether (1:1 ratio) to
give an off white solid as compound 19 (77.6%). LCꢀMS calcd for
1
C25H23F3N4O3 (m/e) 484.17, obsd 485.1 (M þ H). H NMR (300
MHz, CDCl3) δ ppm 0.77 (dd, J = 7.8, 3.0 Hz, 2 H), 1.01 (br s, 2 H),
1.80 (td, J = 8.1, 4.4 Hz, 3 H), 1.93ꢀ2.15 (m, 2 H), 2.73 (t,
J = 11.0 Hz, 1 H), 3.01 (t, J = 11.8 Hz, 1 H), 3.25 (t, J = 12.2 Hz,
1 H), 4.39 (d, J = 13.0 Hz, 1 H), 4.79 (d, J = 10.6 Hz, 1 H), 7.23 (d, J = 8.5
Hz, 1 H), 7.51ꢀ7.68 (m, 3 H), 8.16 (d, J = 7.8 Hz, 2 H), 8.39 (dd, J = 8.5,
2.4 Hz, 1 H), 8.65 (d, J = 2.4 Hz, 1 H), 8.98 (s, 1 H).
5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)amino]-
30,40,50,60-tetrahydro-20H-[2,40]bipyridinyl-10-carboxylic Acid
Methyl Ester (20). Compound 20 was prepared from compound 35
and methyl chloroformate (65%). LCꢀMS calcd for C23H21F3N4O4
2442
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
1
(m/e) 474.15, obsd 475.1 (M þ H). H NMR (300 MHz, CDCl3) δ
3.18 (t, J = 12.2 Hz, 1 H), 4.07 (d, J = 12.1 Hz, 1 H), 4.83 (d, J = 10.6 Hz,
1 H), 7.23 (d, J = 8.2 Hz, 1 H), 7.50ꢀ7.70 (m, 3 H), 8.17 (d, J = 7.2 Hz,
2 H), 8.52 (d, J = 8.2 Hz, 1 H), 8.72 (br s, 1 H), 9.04 (s, 1 H).
4-(4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperazine-1-carbonyl)-trans-cyclohexane-
carboxylic Acid (27). This compound was prepared from the amide
coupling of 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (4-pi-
perazin-1-yl-phenyl)amide (compound 33) and trans-1,4-cyclohexane
dicarboxylic acid using coupling reagent BOP in DMF (37 mg, 32%).
LCꢀMS calcd for C29H29F3N4O5 (m/e) 570.21, obsd 571.0 (M þ H).
1H NMR (300 MHz, DMSO-d6) δ 10.38 (s, 1H), 8.13 (d, J = 6.64 Hz,
2H), 7.55ꢀ7.72 (m, 5H), 6.97 (d, J = 8.75 Hz, 2H), 3.60 (br d, J = 13.30
Hz, 4H), 3.08 (br d, J = 15.40 Hz, 4H), 2.52ꢀ2.59 (m, 1H), 1.83ꢀ1.94
(m, 2H), 1.63ꢀ1.78 (m, 2H), 1.31ꢀ1.45 (m, 4H).
ppm 1.63ꢀ1.83 (m, 2 H), 1.97 (d, J = 12.7 Hz, 2 H), 2.93 (m, 3 H), 3.72
(s, 3 H), 4.30 (br s, 2 H), 7.22 (d, J = 6.9 Hz, 1 H), 7.49ꢀ7.69 (m, 3 H),
8.16 (d, J = 7.2 Hz, 2 H), 8.43 (d, J = 6.9 Hz, 1 H), 8.68 (br s, 1 H), 9.03
(br s, 1 H).
2-Phenyl-5-trifluoromethyloxazole-4-carboxylic Acid
[4-(4-Cyclopropanecarbonylpiperazin-1-yl)phenyl]amide
(21). Compound 21 was prepared from compound 33 and cyclopro-
pane carboxylic acid. LCꢀMS calcd for C25H23F3N4O3 (m/e) 484.17,
obsd 485.2 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 0.81 (d, J =
4.5 Hz, 2 H), 1.03 (br s, 2 H), 1.79 (br s, 1 H), 3.21 (d, J = 14.5 Hz, 4 H),
3.85 (br s, 4 H), 6.97 (d, J = 8.2 Hz, 2 H), 7.48ꢀ7.62 (m, 3 H), 7.68 (d,
J = 7.8 Hz, 2 H), 8.14 (d, J = 7.2 Hz, 2 H), 8.86 (br s, 1 H).
4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperazine-1-carboxylic Acid Methyl Ester
(22). Compound 22 was prepared from compound 33 and methyl
chloroformate. LCꢀMS calcd for C23H21F3N4O4 (m/e) 474.15, obsd
475.2 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 3.15 (br s, 4 H),
3.65 (br s, 4 H), 3.75 (s, 3 H), 6.96 (d, J = 8.2 Hz, 2 H), 7.49ꢀ7.62 (m,
3 H), 7.66 (d, J = 8.5 Hz, 2 H), 8.14 (d, J = 6.9 Hz, 2 H), 8.86 (br s, 1 H).
4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperidine-1-carboxylic Acid Methyl Ester
(24). Compound 24 was prepared from compound 34 and methyl
chloroformate (80.6%). LCꢀMS calcd for C24H22F3N3O4 (m/e) 473.16,
obsd 474.1 (M þ H). 1H NMR (300 MHz, CDCl3) δ ppm 1.64 (qd, J =
12.6, 4.4 Hz, 2 H), 1.86 (d, J = 12.1 Hz, 2 H), 2.61ꢀ2.74 (m, 1 H), 2.88 (t,
J = 11.9 Hz, 2 H), 3.73 (s, 3 H), 4.30 (d, J = 12.1 Hz, 2 H), 7.23 (d, J = 8.5
Hz, 2 H), 7.52ꢀ7.62 (m, 3 H), 7.69 (d, J = 8.5 Hz, 2 H), 8.15 (dd, J = 8.2,
1.5 Hz, 2 H), 8.91 (s, 1 H).
4-(4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperazine-1-carbonyl)-cis-cyclohexanecar-
boxylic Acid (28). This compound was prepared from the amide
coupling of 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (4-pi-
perazin-1-ylphenyl)amide (compound 33) and cis-1,4-cyclohexane di-
carboxylic acid using coupling reagent BOP in DMF (15 mg, 9%).
LCꢀMS calcd for C29H29F3N4O5 (m/e) 570.21, obsd 571.2 (M þ H).
1H NMR (300 MHz, DMSO-d6) δ 12.12 (br s, 1H), 10.39 (s, 1H), 8.13
(br d, J = 9.0 Hz, 2H), 7.54ꢀ7.75 (m, 5H), 6.96 (d, J = 9.0 Hz, 2H),
3.48ꢀ3.70 (m, 4H), 2.97ꢀ3.19 (m, 4H), 2.60ꢀ2.77 (m, 1H),
1.88ꢀ2.07 (m, 2H), 1.41ꢀ1.63 (m, 6H).
4-(4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperidine-1-carbonyl)-trans-cyclohexane-
carboxylic Acid (29). To a solution of compound 36 (2.62 g, 10.0
mmol) in dichloromethane (60 mL) was added oxalyl chloride (10 mL,
2 N in dichloromethane) and one drop of DMF. The resulting solution
was stirred at room temperature for 1 h until gas evolution ceased.
Solvents were evaporated, and the residue was treated with benzene
(20 mL). Solvents were evaporated, and the waxy material was dried
under vacuum. This material was dissolved in dichloromethane (30 mL),
and the solution was added dropwise to a dichloromethane solution
(100 mL) containing compound 34 (4.15 g, 10 mmol) and triethyla-
mine (2.8 mL) under ice bath. The solution was stirred for 30 min, and
the ice bath was removed. After 3 h of stirring at room temperature, the
solution was extracted with dichloromethane and 0.2 N hydrochloric
acid. The organic layer was washed with water and sodium bicarbonate
solution, dried over sodium sulfate, and concentrated. The residue was
passed through silica gel plug eluted with ethyl acetate in hexanes. After
evaporation of solvents, a pale yellow fluffy material was obtained (6.46 g,
98%). This material was dissolved in a solution containing THF (10 mL)
and ethanol (100 mL) in the presence of palladium on carbon (10%, 0.95
g). The mixture was hydrogenated at 45 psi on a Parr shaker until TLC
indicated complete consumption of the starting material. The mixture was
filtered through a layer of Celite and concentrated. The residue was
crystallized from ethyl acetate to provide compound 29 as a white solid
(4.55 g, 80%). Mp 208ꢀ209 ꢀC. HRMS calcd for C30H30F3N3O5
(M þ H) 570.2211, obsd 570.2210. 1H NMR (300 MHz, DMSO-d6) δ
ppm 1.32ꢀ1.61 (m, 6 H), 1.64ꢀ1.98 (m, 6 H), 2.17 (br s, 1 H), 2.59 (d,
J = 11.8 Hz, 2 H), 2.76 (t, J = 11.5 Hz, 1 H), 3.10 (t, J = 12.4 Hz, 1 H), 4.08
(d, J = 13.9 Hz, 1 H), 4.56 (d, J = 12.1 Hz, 1 H), 7.26 (d, J = 8.8 Hz, 2 H),
7.58ꢀ7.77 (m, 5 H), 8.15 (dd, J = 7.8, 1.5 Hz, 2 H), 10.53 (s, 1 H), 12.04
(brs, 1 H). Elementalanalysiscalcd, C 63.26%, H 5.31%, N 7.38%; obsd, C
63.19%, H 5.34%, N 7.31%.
trans-Cyclohexane-1,4-dicarboxylic Acid Monobenzyl Es-
ter (36). To a suspension of trans-cyclohexane-1,4-dicarboxylic acid (5
g, 29 mmol) in THF (60 mL) was added oxalyl chloride (2 N in
dichlrormethane, 15 mL) and two drops of DMF. The mixture was
stirred at room temperature for 3 h. The mixture was evaporated to
dryness, and the residue was treated with benzyl alcohol (5 mL). The
mixture was heated at 80 ꢀC for 20 min and then crystallized from
hexanes to give white needle crystals as trans-cyclohexane-1,4-dicar-
boxylic acid monobenzyl ester (3.2 g, 41%). LCꢀMS calcd for
1
C15H18O4 (m/e) 262.12, obsd 261.0 (M ꢀ H). H NMR (300 MHz,
DMSO-d6) δ ppm 1.22ꢀ1.47 (m, 4 H), 1.80ꢀ2.03 (m, 4 H), 2.10ꢀ2.24
(m, 1 H), 2.27ꢀ2.43 (m, 1 H), 5.08 (s, 2 H), 7.23ꢀ7.46 (m, 5 H), 12.09
(s, 1 H).
4-(4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]pyridin-2-yl}piperazine-1-carbonyl)-trans-cyclohex-
anecarboxylic Acid (25). This compound was prepared from the
amide coupling of 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid
(6-piperazin-1-ylpyridin-3-yl)amide hydrochloride (compound 31) and
trans-1,4-cyclohexane dicarboxylic acid using coupling reagent BOP in
DMF (25%). LCꢀMS calcd for C28H28F3N5O5 (m/e) 571.20, obsd
572.2 (M þ H). 1H NMR (300 MHz, DMSO-d6) δ ppm 1.40 (t, J = 10.4
Hz, 4 H), 1.72 (d, J = 6.9 Hz, 2 H), 1.84ꢀ1.99 (m, 2 H), 2.17 (br s, 1 H),
2.63 (br s, 1 H), 3.40ꢀ3.60 (m, 8 H), 6.90 (d, J = 9.1 Hz, 1 H),
7.57ꢀ7.73 (m, 3 H), 7.98 (dd, J = 9.2, 2.6 Hz, 1 H), 8.09ꢀ8.21 (m, 2 H),
8.51 (d, J = 2.4 Hz, 1 H), 10.52 (s, 1 H), 12.05 (s, 1 H).
4-{5-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)amino]-
30,40,50,60-tetrahydro-20H-[2,40]bipyridinyl-10-carbonyl}-trans-
cyclohexanecarboxylic Acid (26). This compound was prepared
from the amide coupling of 2-phenyl-5-trifluoromethyloxazole-4-car-
boxylic acid (10,20,30,40,50,60-hexahydro[2,40]bipyridinyl-5-yl)amide
(compound 35) and trans-1,4-cyclohexane dicarboxylic acid using
coupling reagent BOP in DMF (31%). LCꢀMS calcd for C29H29-
4-(4-{4-[(2-Phenyl-5-trifluoromethyloxazole-4-carbonyl)-
amino]phenyl}piperidine-1-carbonyl)-cis-cyclohexanecar-
boxylic Acid (30). This compound was prepared from the amide
coupling of 2-phenyl-5-trifluoromethyloxazole-4-carboxylic acid (4-pi-
peridin-4-ylphenyl)amide (compound 34) and cis-1,4-cyclohexane di-
carboxylic acid using coupling reagent BOP in DMF. HRMS calcd for
C30H30F3N3O5 (M þ H) 570.2211, obsd 570.2210. 1H NMR (300 MHz,
1
F3N4O5 (m/e) 570.21, obsd 571.2 (M þ H). H NMR (300 MHz,
CDCl3) δ ppm 1.39ꢀ1.79 (m, 5 H), 1.81ꢀ2.07 (m, 5 H), 2.16 (br s,
2 H), 2.39 (d, J = 11.8 Hz, 1 H), 2.49ꢀ2.78 (m, 2 H), 2.99 (br s, 1 H),
2443
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
DMSO-d6) δ ppm 12.13 (br s, 1 H), 10.53 (s, 1 H), 8.15 (dd, J = 7.8, 1.5
Hz, 2 H), 7.58ꢀ7.78 (m, 5 H), 7.26 (d, J = 8.5 Hz, 2 H), 4.55 (d, J = 12.1
Hz, 1 H), 4.05 (d, J = 14.2 Hz, 1 H), 3.10 (t, J = 12.4 Hz, 1 H), 2.61ꢀ2.84
(m, 2 H), 2.44ꢀ2.58 (m, 2 H), 2.00 (d, J = 7.8 Hz, 2 H), 1.70ꢀ1.88 (m, 2
H), 1.22ꢀ1.65 (m, 8 H).
plate on ice, scraping cells into medium, and then combining with
chloroform/methanol to extract lipids. Extracted lipids were resolved on
TLC plates in solvent containing petroleum ether, ether, acetic acid
(80:20:1, v:v:v). The resolved lipids were identified and quantified by
phosphoimager analysis.
Inhibition on Body Weight Gain. Male DIO rats, 19 weeks old,
on HFD for 12 weeks were treated with compound 29 at 0.3, 1, and 3
mg/kg, po, in Klucel (hydroxypropyl cellulose) suspension for 22 days.
Body weight and food intake were measured for the duration of the
experiment. On day 21, an OGTT was performed, and on day 22 animals
were euthanized and organ weights and fat pads were measured. Plasma
was collected, and lipids were analyzed.
Oral Glucose Tolerance Test. Male SpragueꢀDawley rats (n =
10/group), whose weight ranged from 550 to 650 g, were used for this
study. Animals were acclimated to a reverse light/dark cycle for 2 weeks
prior to the start of the experiment. Animals were placed on high fat diet
(D-12492, Research Diets, New Brunswick, NJ) at 7 weeks of age. The
study began at week 14 of high fat diet treatment and continued for an
additional 23 days. At day 23, an oral glucose tolerance test (OGTT) was
performed to study the effects of treatments on glucose tolerance. Prior
to the glucose challenge, animals were fasted for 2 h and then given an
oral glucose bolus (2 g/kg). Glucose measurements were taken via tail
vein blood collection at 0, 30, 60, and 120 min after glucose challenge.
DGAT-1 Enzyme Assays. Insect cells (High 5) were transfected
with recombinant baculovirus containing human DGAT-1 sequence and
harvested after 48 h. Cells (1 ꢁ 1010) were suspended in 500 mL of
buffer A (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, and
protease inhibitors (complete protease inhibitor cocktail from Roche
Diagnostics, catalog no. 11873580001)) at 4 ꢀC. Cells were lysed by
sonication (3 ꢁ 30 s) and membranes isolated by centrifugation at
100000g for 60 min. Pellets were resuspended in 500 mL of buffer A plus
0.5% Triton X-100 and stirred for 30 min, sonicated (2 ꢁ 30 s), and
centrifuged at 100000g for 60 min. DGAT-1 containing pellets were
then rinsed twice with small amounts of buffer A and homogenized in
300 mL of buffer A by sonication. Protein concentration was 9.85 mg/
mL as determined by Bio-Rad protein assay reagents (catalog no. 500-
00006). Aliquots were stored at ꢀ80 ꢀC.
PatchXpress hERG Inhibition Assay. The detailed method to
quantify hERG channel inhibition by the automated patch clamp system
PatchXpress 7000A (Molecular Devices, Sunnyvale, CA) has been
described elsewhere.25 In brief, Chinese hamster ovary (CHO) cells
transfected with the human ether-a-go-go-related gene (hERG) was
cultured in Ex-cell 302 medium supplemented with 10% fetal bovine
serum, 2 mM glutamine, and 0.25 mg/mL Geneticin and maintained in a
CO2 incubator at 37 ꢀC. For patch clamp electrophysiology, the external
buffer contained the following (in mM): 150 NaCl, 10 Hepes, 4 KCl, 1.2
CaCl2, 1 MgCl2. pH 7.4 was adjusted with HCl. The internal recording
solution contained the following (in mM): 140 KCl, 6 EGTA, 5 Hepes,
MgCl2, 5 ATP-Na2. pH 7.2 was adjusted with KOH. Once the cell was
loaded in the recording chamber and formed a gigaohm seal with the
planar glass electrodes (Sealchip), a whole-cell configuration was
achieved by rupturing the cell membrane. The membrane potential
was then clamped at ꢀ80 mV and the hERG channel activated by a 1 s
depolarizing pulse delivered at 0.1 Hz. The hERG current was measured
during the 500 ms repolarizing pulse to ꢀ40 mV. After an acceptable
hERG current recording was obtained, the cell was first exposed to 0.3%
DMSO as the vehicle control followed by the test article in three
ascending, full log interval concentrations and finally E-4031 at 1 μM (as
the positive control) to block the hERG current completely. Each test
article was tested on three or more cells and at concentrations up to
30 μM or the solubility limit, determined using the BD Gentest solubility
scanner (BD Biosciences, San Jose, CA). The inhibition of hERG current
at each concentration was normalized to that recorded in the vehicle
control and fitted with Hill equation to calculate IC20 and/or IC50.
Pharmacokinetic Analysis. Hanover Wistar rats (n = 3) were
cannulated in jugular and femoral vein. Rats were dosed either intrave-
nously via jugular vein or by oral gavage, and blood samples were
collected via femoral vein. Plasma was separated after centrifugation and
transferred into tubes containing EDTA. Beagle dogs (n = 3) were dosed
intravenously or via oral gavage. Compounds were formulated as the
following: iv 5 mg/kg solution in 2% DMA, 15% PEG400, 77% of 28%
HPβCD in 0.25 M, pH 7 phosphate buffered saline, po 2 mg/mL
suspension in 2% Klucel LF, and 0.1% Tween 80 in water. Quantitative
analysis of DGAT inhibitors in rat and dog plasma was carried out as
follows: Plasma levels of DGAT inhibitors were determined using LC/
MS/MS. Briefly, the compounds were extracted from plasma by protein
precipitation with acetonitrile. The extracts were diluted with an equal
volume of water and injected to a C18 column and analyzed by
LCꢀMS/MS to quantitate the compounds. The linear range of the
method was 5.0ꢀ5000 ng/mL. PK analysis was performed using a
noncompartmental model.
Rat DGAT-1 was also produced in High 5 insect cells and purified by
the same procedures as described above. Protein concentration was 35.9
mg/mL as determined by Bio-Rad assay. Aliquots were stored at ꢀ80 ꢀC.
(A) Radioactive DGAT-1 Phospholipid (PL) FlashPlate Assay. Hu-
man DGAT-1 activities were measured by radioactive PL-FlashPlate
assays. Typically, 10 mM 1,2-dioleoyl-sn-glycerol (DAG) suspended in
water containing 0.1% Triton X-100 was diluted to 0.5 mM with coating
buffer purchased from PerkinElmer (catalog no. SMP900A). The
diluted DAG solution was then added to 384-well PL-FlashPlates
(PerkinElmer, catalog no. SMP 108) at 60 μL per well and incubated
at room temperature for 2 days. Plates were washed twice with buffer B
(50 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 0.05% deoxycholic acid).
Test compounds at 0ꢀ2 mM in 100% DMSO were diluted 10-fold with
buffer C (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 0.01% BSA), and
20 μL of solutions were added into the PL-FlashPlates. To each well was
then added 15 μL of 8.3 μM palmitoyl-1-14C coenzyme A (PerkinElmer,
catalog no. NEC-555) in buffer C followed by 15 μL of human DGAT-1
enzyme solution (0.13 mg of total protein/mL) diluted in buffer A plus
0.2% Triton X-100. The reaction mixtures were incubated at 37 ꢀC for 1
h and washed 3 times with buffer B to stop the reaction. Plates were
sealed and read on a TopCount instrument from PerkinElmer.
Rat DGAT-1 was measured by the same procedure as described
above except that the final enzyme concentration in the reaction mixture
was 0.02 mg of total protein/mL.
DGAT Inhibition Cell Assay. Media, FBS, and supplements were
all from GIBCO (DMEM/F12 (Gibco no. 11330-032) þ 10% FBS þ
1% L-glutamine þ 20 μL G418/mL medium, DMEM high glucose). The
fatty acid free BSA, fraction V, was from Roche. TLC plates (20 cm ꢁ
20 cm silica gel 60 F 254, no. 5735/7 EM Science) were ordered from
VWR. 14C palmitic acid (no. NEC 075H, 0.25 mCi/2.5 mL, 55.0 mCi/
mmol) was obtained from PerkinElmer Life Sciences.
A CHO cell line that expresses DGAT1 (CHOK1/DGAT) was
plated at a density of 2.5 ꢁ 105 cells/well in six-well plates in DMEM/
F12 medium. The next day, medium was removed and the cells were
washed twice with phosphate buffered saline (PBS). PBS was removed,
and DMEM high glucose with 0.01% BSA (fatty acid free) was added
together with tested compounds at multiple dilutions starting at 25 μM.
Then 0.5 μCi 14C palmitic acid was added to each well. The cells were
incubated for 1 h at 37 ꢀC with the reaction terminated by placing the
(B) Radioactive DGAT-1 TLC Assay. DGAT-1 activity was also mea-
sured by a radioactive TLC method. This assay was used for comparing
compound potency against DGAT-2 and ACAT. Typically, DGAT-1
2444
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
(20 μL, 0.13 mg of total protein/mL) in buffer A plus 0.2% Triton X-100
was incubated with 10 μL of test compound solutions (0ꢀ600 μM) in
20% DMSO in buffer C in Eppendorf tubes for 15 min on ice. DAG
solution (50 μL of 0.5 mM in coating buffer) was then added followed by
20 μL of 8.3 μM palmitoyl-1-14C coenzyme A in buffer C and the
reaction mixture incubated at 37 ꢀC for 1 h. The reaction was stopped by
adding 0.3 mL of chloroform/methanol (2:1, v/v). The mixture was
shaken for 5 min and centrifuged. The aqueous phase was discarded, and
100 μL of the organic phase containing triglyceride product was spotted
on TLC plates. Lipids were separated with a solvent system of petroleum
ether/ether/acetic acid (80:20:1, v/v/v) for 35 min and plates analyzed
on a phosphoimager after overnight exposure.
’ ABBREVIATIONS USED
DGAT-1, diacylglycerol acyltransferase-1; hERG, human ether-
a-go-go-related gene; DIO, diet induced obesity; OGTT, oral
glucose tolerance test; GLP-1, glucagon-like peptide-1; PYY,
peptide YY; PyBroP, bromotripyrrolidinophosphonium hexa-
fluorophosphate; BOP, (benzotriazol-1-yloxy)tris(dimethylamino)-
phosphonium hexafluorophosphate; PdCl2(dppf), dichloro[1,10-
bis(diphenylphosphino)ferrocene]palladium(II)
’ REFERENCES
DGAT-2 Enzyme Assay. Sf9 insect cells were transfected with
recombinant baculovirus containing DGAT-2 sequence and harvested
after 48 h. Cells (4 ꢁ 109) were suspended in 230 mL of buffer A at 4 ꢀC.
Cells were lysed by sonication (3 ꢁ 30 s) and membranes isolated by
centrifugation at 100000g for 60 min. DGAT-2 containing pellets were
resuspended in 90 mL of buffer A and sonicated (1 ꢁ 30 s). Protein
concentration was 6.29 mg/mL as determined by Bio-Rad assay, and
aliquots were stored at ꢀ80 ꢀC.
(1) Cases, S.; Smith, S. J.; Zheng, Y-W; Myers, H. M.; Lear, S. R.;
Sande, E.; Novak, S.; Collins, C.; Welch, C. B.; Lusis, A. J.; Erickson,
S. K.; Farese, R. V., Jr. Identification of a gene encoding an acyl CoA:
diacylglycerol acyltransferase, a key enzyme in triacylglycerol synthesis.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13018–13023.
(2) Cases, S.; Stone, S. J.; Zhou, P.; Yen, E.; Tow, B.; Lardizabal,
K. D.; Voelker, T.; Farese, R. V., Jr. Cloning of DGAT2, a second
mammalian diacylglycerol acyltransferase, and related family members.
J. Biol. Chem. 2001, 276, 38870–38876.
DGAT2 enzymatic activity was measured by radioactive TLC assay.
DGAT-2 was diluted in 50 mM Tris-HCl, pH 7.5, 20 mM MgCl2 plus
0.25 M sucrose to 0.313 mg/mL, and 20 μL was incubated with 10 μL of
compound solutions (0ꢀ600 μM) in 20% DMSO in 100 mM Tris-HCl,
pH 7.5, 20 mM MgCl2, and 0.125% BSA in Eppendorf tubes for 15 min
on ice. DAG solution (50 μL of 0.5 mM in coating buffer) was then
added followed by 20 μL of 8.3 μM palmitoyl-1-14C coenzyme A in
buffer C and the reaction mixture incubated at 37 ꢀC for 5 min. The
reaction was stopped, and lipids were analyzed as described above.
ACAT Enzyme Assay. HepG2 cells (2 ꢁ 108) containing a mixture
of ACAT1 and ACAT2 were suspended in 10 mL of buffer A at 4 ꢀC.
Cells were lysed by sonication (3 ꢁ 30 s) and membranes isolated by
centrifugation at 100000g for 60 min. ACAT containing pellets were
resuspended in 5 mL of buffer A and sonicated (1 ꢁ 30 s). Protein
concentration was 2.6 mg/mL as determined by Bio-Rad assay, and
aliquots were stored at ꢀ80 ꢀC.
(3) Yen, C. E.; Stone, S. J.; Koliwad, D.; Harris, C; Farese, R. V., Jr.
DGAT enzymes and triacylglycerol biosynthesis. J. Lipid Res. 2008,
49, 2283–2301.
(4) Smith, S. J.; Cases, S.; Jensen, D. R.; Chen, H. C.; Sande, E.; Tow,
B.; Sanan, D. A.; Raber, J.; Eckel, R. H.; Farese, R. V., Jr. Obesity
resistance and multiple mechanisms of triglyceride synthesis in mice
lacking DGAT. Nat. Genet. 2000, 25, 87–90.
(5) Chen, H. C.; Smith, S. J.; Ladha, Z.; Jensen, D. R.; Ferreira, L. D.;
Pulawa, L. K.; McGuire, J. G.; Pitas, R. E.; Eckel, R. H.; Farese, R. V., Jr.
Increased insulin and leptin sensitivity in mice lacking acyl CoA:
diacylglycerol acyltransferase 1. J. Clin. Invest. 2002, 109, 1049–1055.
(6) Stone, S. J.; Myers, H. M.; Watkins, S. M.; Brown, B. E.; Feingold,
K. R.; Elias, P. M.; Farese, R. V., Jr. Lipopenia and skin barrier
abnormalities in DGAT2-deficient mice. J. Biol. Chem. 2004,
279, 11767–11776.
(7) Lee, B.; Fast, A. M.; Zhu, J.; Cheng, J.; Buhman, K. K. Intestine-
specific expression of acyl CoA:diacylglycerol acyltransferase 1 reverses
resistance to diet-induced hepatic steatosis and obesity in Dgat1ꢀ/ꢀ
mice. J. Lipid Res. 2010, 51, 1770–1780.
ACAT activity was measured by radioactive TLC assay. ACAT
solution was diluted in buffer A to 0.52 mg/mL, and 20 μL was incubated
with 10 μL of compound solutions (0ꢀ600 μM) in 20% DMSO in
buffer C for 15 min on ice followed by 70 μL of 2.37 μM palmitoyl-1-14C
coenzyme A in buffer C. The reaction mixture incubated at 37 ꢀC for 1 h.
For lipid analysis on TLC, a solvent system of petroleum ether/ether/
acetic acid (90:10:1, v/v/v) was used.
(8) King, A. J.; Judd, A. S.; Souers, A. J. Inhibitors of diacylglycerol
acyltransferase: a review of 2008 patents. Expert Opin. Ther. Patents
2010, 20, 19–29.
(9) (a) Birch, A. M.; Buckett, L. K.; Turnbull, A. V. DGAT1
inhibitors as anti-obesity and anti-diabetic agents. Curr. Opin. Drug
Discovery Dev. 2010, 13, 489–496. (b) Matsuda, D.; Tomoda, H.
Triazolo Compounds Useful as Diacylglycerol Acyltransferase1 Inhibi-
tor-WO2009126624. Expert Opin. Ther. Pat. 2010, 20, 1097–1102. (c)
Fox, B. M.; Iio, K.; Li, K.; Choi, R.; Inaba, T.; Jackson, S.; Sagawa, S.;
Shan, B.; Tanaka, M.; Yoshida, A.; Kayser, F. Discovery of pyrrolopyr-
idazines as novel DGAT1 inhibitors. Bioorg. Med. Chem. Lett. 2010,
20, 6030–6033.
’ ASSOCIATED CONTENT
S
Supporting Information. Results of radiolabeled binding
b
assay of compound 29 and effect of compound 29 on insulin
levels. This material is available free of charge via the Internet at
http://pubs.acs.org.
(10) (a) Nakada, Y.; Aicher, T. D.; Le Huerou, Y.; Turner, T.; Pratt,
S. A.; Gonzales, S. S.; Boyd, S. A.; Miki, H.; Yamamoto, T.; Yamaguchi,
H.; Kato, K.; Kitamura., S. Novel acyl coenzyme A (CoA):diacylglycerol
acyltransferase-1 inhibitors: synthesis and biological activities of diacy-
lethylenediamine derivatives. Bioorg. Med. Chem. 2010, 18, 2785–2795.
(b) Yamamoto, T.; Yamaguchi, H.; Miki, H.; Shimada, M.; Nakada, Y.;
Ogino, M.; Asano, K.; Aoki, K.; Tamura, N.; Masago, M.; Kato, K.
Coenzyme A:diacylglycerol acyltransferase 1 inhibitior ameliorates
obesity, liver steatosis, and lipid metabolism abnormality in KKAy mice
fed high-fat or high-carbohydrate diets. Eur. J. Pharmacol. 2010,
640, 243–249.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: (973) 235-3240. Fax: (973) 235-6263. E-mail: yimin.
qian@roche.com.
’ ACKNOWLEDGMENT
We thank Dr. Paul Gillespie, Dr. Robert Goodnow, Dr. Jianping
Cai, and Dr. Christophe Michoud for their DGAT-1 hit identifica-
tion work, Gino Sasso and Theresa Burchfield for 1H NMR help,
Vance Bell for mass spectrometry experiments, and Dr. Rama-
kanth Sarabufor helpful discussionsand readingof the manuscript.
(11) Smith, R.; Campbell, A.-M.; Coish, P.; Dai, M.; Jenkins, S.;
Lowe, D.; O’Connor, S.; Su, N.; Wang, G.; Zhang, M.; Zhu. L.
Preparation of Benzazolylaminobiphenyloxoalkanoates for the Treat-
ment of Obesity. Patent US 2004/0224997, 2004.
2445
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446

Journal of Medicinal Chemistry
ARTICLE
(12) (a) Fox, B. M.; Furukawa, N.; Hao, X.; Iio, K.; Inaba, T.;
Jackson, S. M.; Kayser, F.; Labelle, M.; Li, K.; Matsui, T.; McMinn, D. L.;
Ogawa, N.; Rubenstein, S. M.; Sagawa, S.; Sugimoto, K.; Suzuki, M.;
Tanaka, M.; Ye, G.; Yoshida, A.; Zhang, J. A Preparation of Fused
Bicyclic Nitrogen-Containing Heterocycles. Patent WO 2004/047755,
2004. (b) Fox, B. M.; Iio, K.; Li, K.; Inaba, T.; Labelle, M.; Rubenstein, S.;
Ye, G.; Zhang, J.; Jackson, S.; Choi, R.; Sagawa, S.; Shan, B.; Tanaka, M.;
Yoshida, A.; Kayser, F. Discovery of Pyrimidooxazines as Novel DGAT1
Inhibitors Efficacious in a Mouse Diet-Induced Obesity Model. Pre-
sented at the 32nd Annual National Medicinal Chemistry Symposium,
Minneapolis, MN, June 6ꢀ9, 2010.
(25) Guo, L.; Guthrie, H. Automated electrophysiology in the
preclinical evaluation of drugs for potential QT prolongation. J. Phar-
macol. Toxicol. Methods 2005, 52, 123–135.
(13) Zhao, G.; Souers, A. J.; Voorbach, M.; Falls, H. D.; Droz, B.;
Brodjian, S.; Lau, Y. Y.; Iyengar, R. R.; Gao, J.; Judd, A. S.; Wagaw, S. H.;
Ravn, M. M.; Engstrom, K. M.; Lynch, J. K.; Mulhern, M. M.; Freeman,
J.; Dayton, B. D.; Wang, X.; Grihalde, N.; Fry, D.; Beno, D. W. A.; Marsh,
K. C.; Su, Z.; Diaz, G. J.; Collins, C. A.; Sham, H.; Reilly, R. M.; Brune,
M. E.; Kym, P. R. Validation of diacyl glycerolacyltransferase I as a novel
target for the treatment of obesity and dyslipidemia using a potent and
selective small molecule inhibitor. J. Med. Chem. 2008, 51, 380–383.
(14) Birch, A. M.; Birtles, S.; Buckett, L. K.; Kemmitt, P. D.; Smith,
G. J.; Smith, T. J. D.; Turnbull, A. V.; Wang, S. J. Y. Discovery of a potent,
selective, and orally efficacious pyrimidinooxazinyl bicyclooctaneacetic
acid diacylglycerol acyltransferase-1 inhibitor. J. Med. Chem. 2009,
52, 1558–1568.
(15) Dow, R. L.; Li, J.-C.; Parel, L.; Perreault, C.; Munchhof, M. J.;
Piotrowski, D. W.; Gibbs, E. M.; Zavadoski, W. J.; Manion, T. B.;
Treadway, J. L.; La Perle, J. L. Discovery and Preclinical Pharmacology of
PF-04620110: A Selective Inhibitor of DGAT-1 for the Treatment of
Type-2 Diabetes. Presented at the 239th National Meeting of the
American Chemical Society, San Francisco, CA, March 21ꢀ25, 2010;
MEDI-315.
(16) (a) Wertheimer, S. J.; Bolin, D.; Erickson, S.; Conde-Knape, K.;
Belunis, C.; Konkar, A.; Taub, R.; Rondinone, C. M. Fatty acid
modulators for the treatment of diabesity. Drug Discovery Today: Ther.
Strategies 2007, 4 (2), 129–135.(b) Bolin, D. R.; Cheung, A. W.-H.;
Firooznia, F.; Hamilton, M. M.; Li, S.; McDermott, L. A.; Qian, Y.; Yun,
W. Diacylglycerol Acyltransferase Inhibitors. Patent US 7,714,126, May
11, 2010. (c) Bolin, D. R.; Michoud, C. Inhibitors of Diacylglycerol
Acyltransferase (DGAT). WO 2006/082010, 2006.
(17) Yun, W.; Bolin, D. R.; Li, S.; Ahmad, M.; Wertheimer, S. J.;
Conde-Knape, K.; Chen, Y.; Kazmer, S. Discovery and Optimization of
Oxazole Based Diacylglycerol Acyltransferase-1 Inhibitor for the Treat-
ment of Obesity. Presented at the 237th National Meeting of the
American Chemical Society, Salt Lake City, UT, March 22ꢀ26, 2009;
MEDI-256.
(18) Zhu, B.; Jia, Z.; Zhang, P.; Su, T.; Huang, W.; Goldman, E.;
Tumas, D.; Kadambi, V.; Eddy, P.; Sinha, U.; Scarborough, R. M.; Song,
Y. Inhibitory effects of carboxylic acid group on hERG binding. Bioorg.
Med. Chem. Lett. 2006, 16, 5507–5512.
(19) Redfern, W. S.; Carlsson, L.; Davis, A. S.; Lynch, W. G.;
MacKenzie, I.; Palethorpe, S.; Siegl, P. K. S.; Strang, I.; Sullivan, A. T.;
Wallis, R.; Camm, A. J.; Hammond, T. G. Relationships between
preclinical cardiac electrophysiology, clinical QT interval prolongation
and torsade de pointes for a broad range of drugs: evidence for a provisional
safety margin in drug development. Cardiovasc. Res. 2003, 58, 32–45.
(20) Lagrutta, A. A.; Trepakova, E. S.; Salata, J. J. The hERG channel
and risk of drug-acquired cardiac arrhythmia: an overview. Curr. Top.
Med. Chem. 2008, 8, 1102–1112.
(21) Okawa, M.; Fujii, K.; Ohbuchi, K.; Okumoto, M.; Aragane, K.;
Sato, H.; Tamai, Y.; Seo, T.; Itoh, Y.; Yoshimoto, R. Role of MGAT2 and
DGAT1 in the release of gut peptides after triglyceride ingestion.
Biochem. Biophys. Res. Commun. 2009, 390, 377–381.
(22) Murphy, K. G.; Bloom, S. R. Gut hormones and the regulation
of energy homeostasis. Nature 2006, 444, 854–859.
(23) Larsen, P. J. Mechanisms behind GLP-1 induced weight loss.
Br. J. Diabetes Vasc. Dis. 2008, 8, S34–S41.
(24) Ogawa, N.; Okuma, C.; Furukawa, N. Anorectic Compounds.
Patent WO2005/072740, 2005.
2446
dx.doi.org/10.1021/jm101580m |J. Med. Chem. 2011, 54, 2433–2446