Discovery of a potent and selective DGAT1 inhibitor with a piperidinyl-oxy-cyclohexanecarboxylic acid moiety

Article abstract of DOI:10.1021/ml5003426

We report the discovery of a novel series of DGAT1 inhibitors in the benzimidazole class with a piperdinyl-oxy-cyclohexanecarboxylic acid moiety. This novel series possesses significantly improved selectivity against the A_2A receptor, no ACAT1 off-target activity at 10 μM, and higher aqueous solubility and free fraction in plasma as compared to the previously reported pyridyl-oxy-cyclohexanecarboxylic acid series. In particular, 5B was shown to possess an excellent selectivity profile by screening it against a panel of more than 100 biological targets. Compound 5B significantly reduces lipid excursion in LTT in mouse and rat, demonstrates DGAT1 mediated reduction of food intake and body weight in mice, is negative in a 3-strain Ames test, and appears to distribute preferentially in the liver and the intestine in mice. We believe this lead series possesses significant potential to identify optimized compounds for clinical development.

Full text of DOI:10.1021/ml5003426

Letter
pubs.acs.org/acsmedchemlett
Discovery of a Potent and Selective DGAT1 Inhibitor with a
Piperidinyl-oxy-cyclohexanecarboxylic Acid Moiety
Shuwen He,* Qingmei Hong, Zhong Lai, David X. Yang, Pauline C. Ting, Jeffrey T. Kuethe,
Timothy A. Cernak, Kevin D. Dykstra, Donald M. Sperbeck, Zhicai Wu, Yang Yu, Ginger X. Yang,
Tianying Jian, Jian Liu, Deodial Guiadeen, Arto D. Krikorian, Lisa M. Sonatore, Judyann Wiltsie, Jinqi Liu,
Judith N. Gorski, Christine C. Chung, Jack T. Gibson, JeanMarie Lisnock, Jianying Xiao, Michael Wolff,
Sharon X. Tong, Maria Madeira, Bindhu V. Karanam, Dong-Ming Shen, James M. Balkovec, Shirly Pinto,
Ravi P. Nargund, and Robert J. DeVita
Early Development and Discovery Sciences, Merck Research Laboratories, 2000 Galloping Hill Road, Kenilworth, New Jersey 07033,
United States
S
* Supporting Information
ABSTRACT: We report the discovery of a novel series of DGAT1 inhibitors in the
benzimidazole class with a piperdinyl-oxy-cyclohexanecarboxylic acid moiety. This novel
series possesses significantly improved selectivity against the A2A receptor, no ACAT1 off-
target activity at 10 μM, and higher aqueous solubility and free fraction in plasma as
compared to the previously reported pyridyl-oxy-cyclohexanecarboxylic acid series. In
particular, 5B was shown to possess an excellent selectivity profile by screening it against a
panel of more than 100 biological targets. Compound 5B significantly reduces lipid
excursion in LTT in mouse and rat, demonstrates DGAT1 mediated reduction of food
intake and body weight in mice, is negative in a 3-strain Ames test, and appears to distribute
preferentially in the liver and the intestine in mice. We believe this lead series possesses
significant potential to identify optimized compounds for clinical development.
KEYWORDS: DGAT1, inhibitor, benzimidazole, ACAT1, A2A receptor, cyclohexanecarboxylic acid, lipid tolerance test, epimerization,
metabolite, Ames test, skin
DGAT1 inhibitors have emerged as potential therapeutic agents
against diabetes and obesity.1 DGAT (acyl CoA:diacylglycerol
acyltransferase) catalyzes the final and committed step in the
synthesis of triglyceride: the formation of triacylglycerol from
diacylglycerol and acyl-CoA. DGAT1 is one of the isoforms and
shares only limited homology with DGAT2 in the terms of amino
acid sequence.2−4 DGAT1 has more sequence homology to acyl
CoA:cholesterol acyltransferase (ACAT1 and ACAT2), which
play a crucial role in cholesterol homeostasis.5 DGAT1 has
attracted much attention since the disclosure of the phenotype of
DGAT1 knockout mice, which were shown to be viable and
resistant to diet-induced obesity.6 DGAT1 knockout mice were
also reported to have enhanced insulin sensitivity compared to
wild-type mice.7 The data, taken together, has prompted
Figure 1. Structures of selected DGAT1 inhibitors.
significant research effort in identifying small molecule DGAT1
DGAT1 inhibitors as a novel treatment for diabetes and
obesity.18,19
Recently, we disclosed a series of novel DGAT1 inhibitors in
the benzimidazole class bearing a pyridyl-oxy-cyclohexanecar-
boxylic acid moiety.20 A representative of this series, 1A, is a
potent DGAT1 inhibitor with excellent selectivity against
inhibitors as a potential treatment for obesity and diabetes
(Figure 1)8−16 In a phase II clinical trial, 3-week dosing of LCQ-
908 (pradigastat) at 20 mg daily resulted in a 40% reduction in
fasting triglyceride levels in patients with familial chylomicrone-
mia syndrome, thus demonstrating the clinical proof of concept
that inhibition of DGAT1 in humans leads to reductions of
plasma triglycerides.17 Recent reports on the clinical results for
AZD-7687 demonstrated the ability of DGAT1 inhibitors to
attenuate postprandial triacylglyceride excursion. However, the
gastrointestinal side effects could hinder further development of
Received: February 9, 2014
Accepted: September 8, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
ACAT1 (Figure 2). Furthermore, 1A significantly reduces
triglyceride excursion in lipid tolerance (LTT) in both mice
In addition, the compounds with a piperidinyl linker series
(e.g., 3A) might have a lower risk to cause adverse effects in the
skin. DGAT1 is known to be expressed in the skin of mice and
human.23 DGAT1 KO mice were shown to display deficiency in
skin and fur, including sebaceous gland atrophy and hair loss.
These findings raised concerns that pharmacological inhibition
of DGAT1 in skin could lead to undesirable adverse effects in
humans. The calculated logD for 3A is about one unit lower than
that of 1A (0.35 vs 1.36, Figure 3). Given that the skin tissue is
largely lipophilic, we believed that an inhibitor with lower logD is
expected to distribute less in the skin and therefore has a lower
tendency to elicit undesirable effect.24
Figure 2. Representative DGAT1 inhibitors of the pyridyl-oxy-
cyclohexanecarboxylic acid series.
The synthesis for this series of compounds focused on the
preparation of the piperidine substructures 8A and its trans
isomer 8B (Scheme 1). The initial synthetic route involved
and dogs. However, 1A undergoes cis/trans isomerization in
vivo, which could complicate the further development.
During the profiling of 1A as the lead compound, other
potential issues for this structural series were uncovered. First, 1A
has low aqueous solubility. In the high throughput solubility test,
all the salt forms prepared, free form, formate, and ammonium
salt, have solubility less than 10 μM in PBS buffers at pH 2 and 7,
respectively. Second, 1A appears to be tightly bound to plasma
proteins. The unbound free fraction of 1A in human plasma is
only 0.36%. In mouse plasma, the unbound free fraction is
slightly higher (1.7%). Finally, this series of compounds were
shown to have an off-target interaction with the A2A receptor.21
Compound 2A, an earlier lead compound with a chloro
substitution on the benzimidazole ring, was screened against a
panel of known biological targets (Figure 2). In this test, 2A had
an IC50 of 247 nM in a radioligand based competition binding
assay with recombinant human A2A receptor. The activity was
later confirmed in a similar in-house human A2A receptor binding
assay. In this assay, 1A had IC50 of 334 nM, which is only ∼160-
fold selectivity against DGAT1. Consistent with this result, in an
a
Scheme 1. Synthesis of 3A and 3B
A
2A cAMP functional assay, compound 1A displayed antagonist
a
Reagents and conditions: (a) 1. PPh3, 4-hydroxypyridine, DIAD,
activity (IC50 = 269 nM with 101% inhibition at 30 μM). Given
the cis/trans isomerization issue and the potential issues
discussed above, further profiling of 1A was discontinued.
While profiling 1A and related compounds, we continued to
explore other structural variations to present both the
benzimidazole and the carboxylic acid pharmacophores in the
molecule. We became interested in a new design, which
incorporates a piperidinyl linker instead of a pyridyl linker
(e.g., structure 3A in Figure 3). We believed the piperidiny linker
THF, 55 °C, 2 days; 2. SFC (ChiralPak AD-H), 7A (14% for two
steps), 7B (23% for two steps); (b) 6-fluoronicotinaldehyde,
potassium peroxymonosulfate, DMF−water, 69%; (c) PtO2, TsOH,
H2 (45 psi), EtOH, RT, 5 days; (d) NaHCO3, DMSO, 110 °C, 43%
for two steps; (e) LiOH, THF−water, 83%; (f) PtO2, TsOH, H2 (45
psi), EtOH, 5 days; (g) NaHCO3, DMSO, 110 °C, 27% for two steps;
(h) potassium peroxymonosulfate, 4-fluorobenzene-1,2-diamine,
DMF−water, 18%; (i) LiOH, THF−water, 70%.
saturating a pyridyl ring to furnish the required piperidinyl
intermediates. The Mitsunobu reaction of commercially available
ethyl 4-hydroxycyclohexanecarboxylate 6 (a cis and trans
mixture) and 4-hydroxypyridine followed by SFC separation
gave the cis and trans isomers (7A and 7B). Saturation of the
pyridyl ring to a piperidinyl ring turned out to be challenging,
which we rationalized to be due to the electron donating effect of
the alkoxy substituent on 4-position of the pyridyl ring.
Eventually, we were able to hydrogenate 7A to piperidine 8A
with frustratingly long reaction time at 45 psi in a Parr shaker at
room temperature. The trans isomer 7B behaved similarly in the
hydrogenation step to give 8B. Coupling of 8A with 2-(6-
fluoropyridin-3-yl)-5-fluoro-1H-benzimidazole 9 gave ethyl ester
10A. The coupling reaction condition was identified from a
comprehensive screening of a variety of solvents and bases.
Eventually, we identified that the coupling worked best with a
weak inorganic base such as NaHCO3 in polar aprotic solvents
(such as DMSO and NMP).25 The coupling chemistry thus
developed proved to be pivotal for the rapid progression of this
structural series. Under the same conditions, 8B was coupled
Figure 3. Design of the piperidinyl-oxy-cyclohexanecarboxylic acid
series.
would impart more flexibility to the structure, as well as a more
basic piperidine nitrogen, which could improve aqueous
solubility with molecules of essentially the same molecular
weight. As an advantage to alternative designs, the symmetrical
nature of this piperidinyl linker would not impose additional
stereochemical complications to the synthesis. Furthermore,
given that many known A2A receptor modulators have multiple
aromatic rings in their structures, we expected that the saturation
of the pyridyl ring system could decrease the number of aromatic
rings and potentially reduce the affinity with the A2A receptor.22
B
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
with 6-fluoronicotinaldehyde to give aldehyde intermediate 11B.
Oxidative condensation of 11B with 4-fluorobenzene-1,2-
diamine afforded ester 10B. Finally, hydrolysis of ethyl esters
10A and 10B provided compounds 3A and 3B, respectively.
The difficulty in preparing the piperidine pieces (8A and 8B)
by the route shown above significantly hampered the synthesis
and profiling of this series of compounds. To overcome this
synthetic problem, we developed a new route starting with
benzyl 4-oxopiperidine-1-carboxylate (Scheme 2). Reductive
compounds in piperidinyl series maintain potent inhibition on
both human and mouse DGAT1 except for 3B, the least potent
analogue in this series.28 The compounds are slightly less potent
on mouse DGAT1 than on human DGAT1. The potency loss
appears to be more severe for the compounds in the trans series.
In contrast, in the pyridyl linked series reported earlier (e.g., 1A),
there is less differentiation of potencies (human vs mouse and cis
vs trans).16 However, the piperidinyl series shows much
improved off-target selectivity. All the compounds in the
piperdinyl series are now highly selective against ACAT1.29 In
contrast, in the pyridyl linker series, a small substituent on
benzimidazole (e.g., F or H) is required to offer good ACAT
selectivity, while maintaining potent DGAT1 inhibition.
Furthermore, all the piperidinyl linked compounds possess
improved selectivity against A2A as compared to 1A and 2A of the
pyridyl linker series. In particular, 3A, 4B, 5A, and 5B achieve
>1000-fold selectivity (the ratios of hDGAT1 IC50 and hA2A Ki).
Finally, the piperidinyl series displays improved aqueous
solubility and increased free fractions in plasma. For example,
the TFA salt of 5B has solubility of 130 and 151 μM in PBS buffer
at pH 2 and 7, respectively. The solubility of the free form is 167
μM and 153 μM at pH 2 and 7, respectively. The free fraction of
5B is 9.0% and 7.8% in human and mouse plasma, respectively.30
The compounds were subsequently tested in rodent lipid
tolerance test (LTT), an in vivo lipid excursion assay, to evaluate
their ability to inhibit plasma triglyceride accumulation. In mouse
LTT screening at 10 mpk oral dose, 3A and 4A show modest
efficacy reducing lipid excursion (∼50% at 20 h time point).31
Because of its poor potency on mouse DGAT1, 3B was excluded
from mouse LTT screening. However, 5A and 5B with a CF3
substituent have the best efficacy. Similarly, 4B with a chloro
substituent on benzimidazole ring gave excellent efficacy in
mouse LTT screening. However, it suffers from slightly increased
off-target activity on hERG ion channel in a competitive binding
assay with radiolabeled MK-499 (Ki = 4.0 μM).32 Because of its
excellent efficacy observed in mouse LTT screening, 5B was also
tested in rat LTT model (Figure 4).33 At 1 mpk and 3 mpk, 5B
reduces lipid excursion in rat by ∼80%. The corresponding
plasma concentrations of 5B were 20 and 58 nM.34 In this assay,
reduction of lipid excursion by the positive control, Cpd A (a
known DGAT1 inhibitor from the literature), at 10 mpk was
normalized to 100%.35
a
Scheme 2. Reductive Etherification Route
a
Reagents and conditions: (a) 1. ethyl 4-hydroxycyclohexanecarbox-
ylate (cis/trans mixture), triethylamine, TMSCl, 0 °C; 2. benzyl 4-
oxopiperidine-1-carboxylate, Et3SiH, TMSOTf, −78 to 0 °C, 91%; (b)
1. H2, Pd/C; 2. NaHCO3, NMP, 110 °C; 3. SFC (ChiralPak AD-H),
13A (37% for three steps), 13B (33% for three steps); (d) LiOH,
THF−water, RT, 73%; (e) LiOH, THF−water, RT, 81%.
etherification of this starting material with ethyl 4-hydroxycy-
clohexanecarboxylate (cis and trans) gave a mixture of 12A and
12B.26 Removal of Cbz-protecting group using standard
conditions furnished a mixture of 8A and 8B. Following the
chemistry described above, 5A and 5B were prepared. During the
reductive etherification reaction, trans/cis configuration in ethyl
4-hydroxycyclohexanecarboxylate was transferred to the product
with complete retention. Therefore, scale-up of this series of
compounds could begin with pure cis- or trans-ethyl 4-
hydroxycyclohexanecarboxylate to provide the desired inter-
mediate in cis or trans form, respectively.27
To establish the target engagement on DGAT1, 5B was
evaluated in wild type (WT) vs DGAT1 knockout (KO) mice for
the effect on body weight and food intake (Figure 5).36 In a six
day study with daily dosing at 10 mpk, 5B reduced cumulative
Applying the chemistry described above, we prepared the
compounds with several different substitutions on the
benzimidazole ring in both the cis and trans series. The
compounds and their profiles are listed in Table 1. Most of the
Table 1. Profiles of Compounds 3−5
human
mouse
hA2A
hERG
binding
(Ki, μM)
cis/
DGAT1 IC50 DGAT1 IC50 human ACAT1 IC50 binding Ki
compd trans
X
(nM)
(nM)
(% inh. at 10 μM)
(μM)
mouse LTT (10 mpk) triglyceride reduction @18 h
3A
3B
4A
4B
5A
5B
cis
F
5.8
15
26
>10 μM (−2%)
>10 μM (5%)
>10 μM (13%)
>10 μM (4%)
>10 μM (22%)
>10 μM (23%)
>10
9.7
ND
3.5
8.0
4.6
55%
ND
11
trans
cis
F
110
11
>60
15
Cl
2.2
2.9
2.5
3.9
52%
75%
75%
89%
trans
cis
Cl
29
4.0
7.0
10
CF3
CF3
8.0
23
trans
C
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
As 5B was extensively profiled as a potential development
candidate, 1A in the pyridyl linked series was shown to isomerize
to its trans isomer in vivo. Immediately, 5B was evaluated for this
possibility. In a hepatocyte incubation study, 5B does isomerize
but at different levels across the species (Table 2). In dog and
Table 2. Conversion of 5B to 5A and Vice Versa in Hepatocyte
a
Incubation
conversion of 5B (parent) to 5A (metabolite)
species
5A peak area as % of 5B
mouse
rat
3.3%
34.6%
0%
Figure 4. Compound 5B reduces lipid excursion in rat LTT.
dog
monkey
human
0%
5.7%
conversion of 5A (parent) to 5B (metabolite)
species
5B peak area as % of 5A
rat
67.0%
26.1%
human
a
Compound was added with final concentration of 10 μM into
hepatocytes diluted to 1 × 106 cell/mL in a buffer. The incubation was
performed at 37 °C for 120 min. The plates were centrifuged, and the
supernatant was analyzed by LC−MS/MS.
monkey, the isomerization was minimal, while in mouse and
human there was more isomerization. The isomerization was
most pronounced in rat hepatocytes. As a comparison, the
isomerization from 5A (the cis isomer) to trans-5B occurs more
extensively in rat and human.
In addition to the in vitro hepatocyte incubation study, 5B was
also tested for its in vivo pharmacokinetics in rat and dog (Table
3). Consistent with the in vitro data, about half of trans-5B was
a b
,
Table 3. Pharmacokinetic Data for 5B in Rat and Dog
PK parameters
rat
dog
F (%)
Cl (mL min−1 kg−1
8
126
88.9
4.54
1.16
4
Figure 5. Effect on body weight and food intake of 5B in DGAT1 KO vs
WT mice.
)
5.82
0.19
0.91
2
Vdss (L kg−1
)
t
1/2 (h)
body weight gain and food intake significantly in WT mice as
compared to vehicle treated groups starting from day 3, while the
body weight and food intake of 5B treated KO mice were
indistinguishable from the vehicle treated group. This result
strongly suggests that the reduction of body weight and food
intake was due to the inhibition of DGAT1.
During the profiling of 5B, there was some concern around the
possible mutagenicity due to the presence of a benzimidazole
moiety in the structure.37 Initial in silico prediction by
MultiCASE and DEREK indicated that the probability of
mutagenicity was low.38 Indeed, 5B was tested and shown to
be negative in a 3-strain microbial mutagenesis assay. Over a
concentration range of 30 to 5000 μg/plate, 5B did not produce
any 2-fold or greater increases in revertants relative to the solvent
control in any of the three strains tested (TA1535, TA98, and
TA100), either with or without metabolic activation. This result
offered us more confidence on the overall potential for this
structure class.
oral dose (mg kg−1
)
AUC (μM·h) 5B
AUC (μM·h) metabolite 5A
1.00
0.97
1.94
0.11
a
Compound 5B was dosed in Sprague−Dawley rats as a solution in
EtOH/PEG400/water (10:50:40) at 1 mg/kg, iv, and 2 mg/kg, po.
Compound 5B was dosed in beagle dogs as a solution in 30% captisol
(pH 8) at 1 mg/kg, iv, and in 0.5 methylcellulose at 4 mg/kg, po.
b
converted to the cis isomer 5A in rat in vivo as indicated by the
AUCs in plasma. Less isomerization occurred in dog (∼5%).
Finally, 5B was dosed orally in mice to determine the
distribution in various tissues of interest (Table 4). Considering
the isomerization of 5B to 5A, the drug levels of 5A were also
measured. Compound 5B showed much higher drug levels in
liver and small intestine compared with the concentrations in
plasma at both 3 and 24 h time points. The level of 5B in skin is
much lower compared to liver and intestine, but still high
compared with the IC50 on DGAT1 (17- and 7-fold over mouse
DGAT1 IC50 at 3 and 24 h, respectively). The tissue distribution
of 5A (the metabolite of 5B from isomerization) tracks that of 5B
albeit at lower concentrations. On the basis of this study, 5B
demonstrated favorable distributions in the target organs (liver
With the increasing interest in 5B as a potential development
candidate, 5B was screened against a panel of more than 100
biological targets (receptors and enzymes). In this set of assays,
5B showed an excellent selectivity profile with IC50 >10 μM
against all targets except A2A.
D
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
a
Table 4. Mouse Tissue Distribution after Oral Dosing of 5B
REFERENCES
■
(1) 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.
concentration of 5B
(μM)
concentration of 5A (μM)
tissue
plasma
3 h
24 h
3 h
24 h
(2) 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.
(3) 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.
(4) 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.
(5) Pramfalk, C.; Eriksson, M.; Parini, P. Cholesteryl esters and ACAT.
Eur. J. Lipid Sci. Technol. 2012, 114, 624−633.
(6) 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.
(7) Chen, H. C.; Rao, M.; Sajan, M. P.; Standaerdt, M.; Kanoh, Y.;
Miura, A.; Farese, R. V., Jr. Role of adipocyte-derived factors in
enhancing insulin signaling in skeletal muscle and white adipose tissue of
mice lacking Acyl CoA:diacylglycerol acyltransferase 1. Diabetes 2004,
53, 1445−1451.
(8) For recent reports of DGAT1 inhibitors, see: Serrano-Wu, M. H.;
Kwak, Y.; Coppola, G.; Foster, C.; Gilmore, T.; Gong, Y.; He, G.; Hou,
Y.; Kantor, A.; Li, J.; Mergo, W.; Nakajima, K.; Neubert, A.; Radetich, B.;
Stroup, B.; Sung, M.; Szklennik, P.; Tichkule, R.; Yang, L.; Yoon, T.;
Zhu, Y.; Wareing, J.; Hosagrahara, V.; Jain, M.; Chatelain, R.;
Commerford, R.; Dardik, B.; Meyers, D.; Hubbard, B. Discovery of a
DGAT1 inhibitor with robust suppression of postprandial triglyceride
levels in humans. Abstracts of Papers, 243rd ACS National Meeting &
Exposition, San Diego, CA, March 25−29, 2012.
(9) Serrano-Wu, M. H.; Coppola, G. M.; Gong, Y.; Neubert, A. D.;
Chatelain, R.; Clairmont, K. B.; Commerford, R.; Cosker, T.; Daniels,
T.; Hou, Y.; Jain, M.; Juedes, M.; Li, L.; Mullarkey, T.; Rocheford, E.;
Sung, M. J.; Tyler, A.; Yang, Q.; Yoon, T.; Hubbard, B. K. Intestinally
targeted diacylglycerol acyltransferase 1 (DGAT1) inhibitors robustly
suppress postprandial triglycerides. ACS Med. Chem. Lett. 2012, 3, 411−
415.
0.324
21.6
<0.01
4.02
0.054
6.91
<0.01
0.672
0.076
<0.010
0.018
liver
intestine (small)
fat (adipose)
skin
36.6
0.711
<0.01
0.167
5.026
0.021
0.046
0.237
0.404
a
Compound 5B was dosed in C57BL/6 mice as a solution in EtOH/
PEG400/water (10:50:40) at 10 mg/kg, po.
and small intestine). However, 5B is present in the skin at a
significant concentration. Furthermore, the isomerization of 5B
to an active metabolite (5A) could complicate the development
process.39 Therefore, although there were many desirable
attributes associated with 5B, the profiling of 5B as a potential
development candidate was discontinued.
In summary, we have described the discovery of a novel series
of DGAT1 inhibitors in the benzimidazole class with a
piperdinyl-oxy-cyclohexanecarboxylic acid moiety. This series
of compounds abolish the ACAT1 off-target activity associated
with the earlier pyridyl-oxy-cyclohexanecarboxylic acid series.
Meanwhile, the liability of A2A binding affinity is significantly
reduced by the introduction of the piperidinyl linker. The series
shows improved aqueous solubility and higher free fraction in
plasma. The most interesting compound in this series to date, 5B,
was shown to possess an excellent selectivity profile by screening
it against a panel of more than 100 biological targets. In addition
to its favorable in vitro profile, 5B significantly reduced lipid
excursion in LTT assays in mouse and rat, respectively. In a WT
vs DGAT1 KO mouse study, 5B demonstrated DGAT1
mediated reduction in food intake and body weight. Upon oral
dosing in mice, 5B seemed to distribute in liver and small
intestine preferentially. Furthermore, the risk of mutagenicity of
5B was low based on a 3-strain Ames test. All the exciting data
collected for 5B helped us to make a critical decision to
deprioritize the previous pyridyl linked series and instead focus
on this new series. Additional efforts focusing on addressing cis/
trans isomerization issue and further reducing the distribution of
the DGAT1 inhibitor in skin while maintaining other desirable
attributes will be disclosed in the future.
(10) Dow, R. L.; Li, J.-C.; Pence, M. P.; Gibbs, E. M.; LaPerle, J. L.;
Litchfield, J.; Piotrowski, D. W.; Munchhof, M. J.; Manion, T. B.;
Zavadoski, W. J.; Walker, G. S.; McPherson, R. K.; Tapley, S.; Sugarman,
E.; Guzman-Perez, A.; Dasilva-Jardine, P. Discovery of PF-04620110, a
potent, selective, and orally bioavailable inhibitor of DGAT-1. ACS Med.
Chem. Lett. 2011, 2, 407−412.
(11) Yeh, V. S.; Beno, D. W.; Brodjian, S.; Brune, M. E.; Cullen, S. C.;
Dayton, B. D.; Dhaon, M. K.; Falls, H. D.; Gao, J.; Grihalde, N.; Hajduk,
P.; Hansen, T. M.; Judd, A. S.; King, A. J.; Klix, R. C.; Larson, K. J.; Lau,
Y. Y.; Marsh, K. C.; Mittelstadt, S. W.; Plata, D.; Rozema, M. J.; Segreti, J.
A.; Stoner, E. J.; Voorbach, M. J.; Wang, X.; Xin, X.; Zhao, G.; Collins, C.
A.; Cox, B. F.; Reilly, R. M.; Kym, P. R.; Souers, A. J. Identification and
preliminary characterization of a potent, safe, and orally efficacious
inhibitor of acyl-CoA:diacylglycerol acyltransferase 1. J. Med. Chem.
2012, 55, 1751−1757.
(12) Plowright, A. T.; Barton, P.; Stuart, B.; Alan, M.; Birtles, S.;
Buckett, L. K.; Butlin, R. J.; Davies, R. D. M.; Ertan, A.; Gutierrez, P. M.;
Kemmitt, P. D.; Leach, A. G.; Svensson, P. H.; Turnbull, A. V.; Waring,
M. J. Design and synthesis of a novel series of cyclohexyloxy-pyridyl
derivatives as inhibitors of diacylglycerol acyl transferase 1. Med. Chem.
Commun. 2013, 4, 151−158.
ASSOCIATED CONTENT
■
S
* Supporting Information
Syntheses and characterization data for the new compounds.
This material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*(S.H.) Phone: 908-740-0881. E-mail: shuwen_he@merck.com.
Notes
The authors declare no competing financial interest.
(13) Barlind, J. G.; Bauer, U. A.; Birch, A. M.; Birtles, S.; Buckett, L. K.;
Butlin, R. J.; Davies, R. D. M.; Eriksson, J. W.; Hammond, C. D.;
Hovland, R.; Johannesson, P.; Johansson, M. J.; Kemmitt, P. D.;
Lindmark, B. T.; Morentin, P.; Noesk, T. A.; Nordin, A.; O’Donnell, C.
J.; Petersson, A. U.; Redzic, A.; Turnbull, A. V.; Vinblad, J. Design and
ACKNOWLEDGMENTS
■
We thank Mr. Thomas J. Novak and Dr. Li-Kang Zhang at Merck
Research Laboratories for measuring the high resolution mass.
E
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
optimization of pyrazinecarboxamide-based inhibitors of diacylglycerol
acyltransferase 1 (DGAT1) leading to a clinical candidate dimethylpyr-
azinecarboxamide phenylcyclohexylacetic acid (AZD7687). J. Med.
Chem. 2012, 55, 10610−10629.
(14) Waring, M. J.; Birch, A. M.; Birtles, S.; Buckett, L. K.; Butlin, R. J.;
Campbell, L.; Gutierrez, P. M.; Kemmitt, P. D.; Leach, A. G.; MacFaul,
P. A.; O’Donnell, C.; Turnbull, A. V. Optimisation of biphenyl acetic
acid inhibitors of diacylglycerol acetyl transferase 1: the discovery of
AZD2353 Med. Chem. Commun. 2013, 4, 159−164.
(15) Goldberg, F. W.; Birch, A. M.; Leach, A. G.; Groombridge, S. D.;
Snelson, W. L.; Gutierrez, P. M.; Hammond, C. D.; Birtles, S.; Buckett,
L. K. Discovery and optimization of efficacious neutral 4-amino-6-
biphenyl-7,8-dihydropyrimido[5,4-f][1,4]oxazepin-5-one diacylglycer-
ol acyl transferase-1 (DGAT1) inhibitors. Med. Chem. Commun. 2013, 4,
165−174.
(16) DeVita, R. J.; Pinto, S. Current status of the research and
development of diacylglycerol O-acyltransferase 1 (DGAT1) inhibitors.
J. Med. Chem. 2013, 56, 9820−9825.
(28) For a description of the protocols for enzyme inhibition assays,
see: Liu, J.; Gorski, J. N.; Gold, S. J.; Chen, D.; Chen, S.; Forrest, G.; Itoh,
Y.; Marsh, D. J.; McLaren, D. G.; Shen, Z.; Sonatore, L.; Carballo-Jane,
E.; Craw, S.; Guan, X.; Karanam, B.; Sakaki, J.; Szeto, D.; Tong, X.; Xiao,
J.; Yoshimoto, R.; Yu, H.; Roddy, T. P.; Balkovec, J.; Pinto, S.
Pharmacological inhibition of diacylglycerol acyltransferase 1 reduces
body weight and modulates gut peptide release: potential insight into
mechanism of action. Obesity 2013, 21, 1406−1415.
(29) Inhibition of ACAT has been shown to cause adverse changes in
preclinical species, see: Floettmann, J. E.; Buckett, L. K.; Turnbull, A. V.;
Smith, T.; Hallberg, C.; Birch, A.; Lees, D.; Jones, H. B. ACAT-selective
and nonselective DGAT1 inhibition: adrenocortical effects. A cross-
species comparison. Toxicol. Pathol. 2013, 41, 941−950.
(30) For protocols measuring aqueous solubility and free fractions in
plasma, see ref 24.
(31) For a description of mouse LTT, see ref 28.
(32) Radiolabeled MK-499 has been used in a hERG (potassium ion
channel Kv11.1) binding assay to provide an initial screen of a
compound’s potential to interact with hERG. Blocking hERG can result
in QTc prolongation, a serious adverse effect. Raab, C. E.; Butcher, J. W.;
Connolly, T. M.; Karczewski, J.; Yu, N. X.; Staskiewicz, S. J.; Liverton,
N.; Dean, D. C.; Melillo, D. G. Synthesis of the first sulfur-35-labeled
hERG radioligand. Bioorg. Med. Chem. Lett. 2006, 16, 1692−1695.
(33) For a description of rat LTT, see Supporting Information.
(34) The plasma concentrations reported here were the concentrations
of 5A and 5B combined at the 4 h time point postoral dosing in contrast
to mouse LTT where the plasma concentrations were measured at the
20 h time point postoral dosing.
(17) Meyers, C.; Gaudet, D.; Tremblay, K.; Amer, A.; Chen, J.; Aimin,
F. The DGAT1 inhibitor LCQ908 decreases triglyceride levels in
patients with the familial chylomicronemia syndrome. J. Clin Lipidol.
2012, 6, 266−267.
(18) Denison, H.; Nilsson, C.; Kujacic, M.; Lofgren, L.; Karlsson, C.;
̈
Knutsson, M.; Eriksson, J. W. Proof of mechanism for the DGAT1
inhibitor AZD7687: results from a first-time-in-human single-dose
study. Diabetes Obes. Metab. 2013, 15, 136−143.
(19) Denison, H.; Nilsson, C.; Lofgren, L.; Himmelmann, A.;
̈
Martensson, G.; Knutsson, M.; Al-Shurbaji, A.; Tornqvist, H.;
̊
(35) The positive control, Cpd A, was originally disclosed by scientists
at Japan Tobacco and Tularik: Fox, B. M.; Furukawa, N. H.; Hao, X.; Lio,
K.; Inaba, T.; Jackson, S. M.; Kayser, F.; Labelle, M.; Kexue, M.; 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, useful in the
treatment or prevention of metabolic and cell proliferative diseases.
WO 2004/ 047755A2. CAN 14138623.
Eriksson, J. W. Diacylglycerol acyltransferase 1 inhibition with
AZD7687 alters lipid handling and hormone secretion in the gut with
intolerable side effects: a randomized clinical trial. Diabetes Obes. Metab.
2014, 16, 334−343.
(20) He, S.; Hong, Q.; Lai, Z.; Wu, Z.; Yu, Y.; Kim, D. W.; Ting, P. C.;
Kuethe, J. T.; Yang, G. X.; Jian, T.; Liu, J.; Guiadeen, D.; Krikorian, A. D.;
Sperbeck, D. M.; Sonatore, L. M.; Wiltsie, J.; Chung, C. C.; Gibson, J. T.;
Lisnock, J.; Murphy, B. A.; Gorski, J. N.; Liu, J.; Chen, D.; Chen, X.;
Wolff, M.; Tong, S. X.; Madeira, M.; Karanam, B. V.; Shen, D.-M.;
Balkovec, J. M.; Pinto, S.; Nargund, R. P.; DeVita, R. J. ACS Med. Chem.
Lett. 2013, 4, 773−778.
(21) A2A receptor mediates many important physiological functions. de
Lera Ruiz, M.; Lim, Y.-H.; Zheng, J. Adenosine A2A receptor as a drug
discovery target. J. Med. Chem. 2014, 57, 3623−3650.
(22) For a review on small molecule adenosine receptor agonists and
(36) For a description of the food intake and body weight study, see ref
28.
antagonists, see: Muller, C. E.; Jacobson, K. A. Recent developments in
̈
adenosine receptor ligands and their potential as novel drugs. Biochim.
Biophys. Acta 2011, 1808, 1290−1308.
(37) Mercangoz, A.; Korkmaz, F. Mutagenicity of 1-ethyl-2,4,5-
̈
triphenyl-1H-imidazole and six derivatives in Salmonella typhimurium.
Gazi Univ. J. Sci. 2012, 25, 1−8.
(23) Chen, H. C.; Smith, S. J.; Tow, B.; Elias, P. M.; Farese, R. V. Leptin
modulates the effects of acyl CoA:diacylglycerol acyltransferase
deficiency on murine fur and sebaceous glands. J. Clin. Invest. 2002,
109, 175−181.
(38) For an excellent discussion on in silico prediction of mutagenicity,
see: Hillebrecht, A.; Muster, W.; Brigo, A.; Kansy, M.; Weiser, T.; Singer,
T. Comparative evaluation of in silico systems for Ames test
mutagenicity prediction: Scope and limitations. Chem. Res. Toxicol.
2011, 24, 843−854.
(24) Subsequently, we carried out a study to evaluate logD as a
predictor of skin exposure with subsequent induction of sebaceous gland
atrophy: Muise, E. S.; Zhu, Y.; Verras, A.; Karanam, B. V.; Gorski, J.;
Weingarth, D.; Lin, H. V.; Hwa, J.; Thompson, J. R.; Hu, G.; Liu, J.; He,
S.; DeVita, R. J.; Shen, D.-M.; Pinto, S. Identification and character-
ization of sebaceous gland atrophy-sparing DGAT1 inhibitors. PLoS One
2014, 9, e88908.
(25) For a report detailing the effort identifying the optimal coupling
condition, see: Cernak, T. A.; et al. Manuscript in preparation.
(26) For an initial disclosure of the reductive etherification as an
efficient strategy to prepare ether compounds, see: Hatakeyama, S.;
Mori, H.; Kitano, K.; Yamada, H.; Nishizawa, M. Efficient reductive
etherification of carbonyl compounds with alkoxytrimethylsilanes.
Tetrahedron Lett. 1994, 35, 4367−4370.
(39) Guidance for Industry: safety testing of drug metabolites. Center
for Drug Evaluation and Research, U.S. FDA, February, 2008. http://
w w w . f d a . g o v / d o w n l o a d s / D r u g s /
GuidanceComplianceRegulatoryInformation/Guidances/ucm079266.
pdf (downloaded January 2014).
(27) For a report on the application of the reductive etherification
strategy for the preparation of 12A, 12B, and related compounds, see:
Kuethe, J. T.; Janey, J. M.; Truppo, M.; Arradondo, J.; Li, T.; Yong, K.;
He, S. A biocatalytic/reductive etherification approach to substitued
piperidinyl ethers. Tetrahedron 2014, 70, 4563−4570.
F
dx.doi.org/10.1021/ml5003426 | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX