Discovery and Optimization of Imidazopyridine-Based Inhibitors of Diacylglycerol Acyltransferase 2 (DGAT2)

Article abstract of DOI:10.1021/acs.jmedchem.5b01006

The medicinal chemistry and preclinical biology of imidazopyridine-based inhibitors of diacylglycerol acyltransferase 2 (DGAT2) is described. A screening hit 1 with low lipophilic efficiency (LipE) was optimized through two key structural modifications: (1) identification of the pyrrolidine amide group for a significant LipE improvement, and (2) insertion of a sp³-hybridized carbon center in the core of the molecule for simultaneous improvement of N-glucuronidation metabolic liability and off-target pharmacology. The preclinical candidate 9 (PF-06424439) demonstrated excellent ADMET properties and decreased circulating and hepatic lipids when orally administered to dyslipidemic rodent models.

Full text of DOI:10.1021/acs.jmedchem.5b01006

Article
pubs.acs.org/jmc
Discovery and Optimization of Imidazopyridine-Based Inhibitors of
Diacylglycerol Acyltransferase 2 (DGAT2)
,
†
,∥
∥
∥
†
Kentaro Futatsugi,* Daniel W. Kung,* Suvi T. M. Orr, Shawn Cabral, David Hepworth,
∥
∇
∥
†
†
∥
Gary Aspnes, Scott Bader, Jianwei Bian, Markus Boehm, Philip A. Carpino, Steven B. Coffey,
Matthew S. Dowling, Michael Herr, Wenhua Jiao, Sophie Y. Lavergne, Qifang Li, Ronald W. Clark,
∥
∥
∥
∥
∥
‡
‡
‡
⊥
‡
‡
‡
Derek M. Erion, Kou Kou, Kyuha Lee, Brandon A. Pabst, Sylvie M. Perez, Julie Purkal,
#
#
§
#
○
Csilla C. Jorgensen, Theunis C. Goosen, James R. Gosset, Mark Niosi, John C. Pettersen,
‡
‡
‡
Jeffrey A. Pfefferkorn, Kay Ahn, and Bryan Goodwin
†
‡
§
Worldwide Medicinal Chemistry, Cardiovascular, Metabolic and Endocrine Diseases Research Unit, and Pharmacokinetics,
Dynamics and Metabolism, Pfizer Worldwide Research & Development, Cambridge, Massachusetts 02139, United States
∥
⊥
#
Worldwide Medicinal Chemistry, Cardiovascular, Metabolic and Endocrine Diseases Research Unit, Pharmacokinetics, Dynamics
∇
○
and Metabolism, Pharmaceutical Sciences, and Drug Safety Research & Development, Pfizer Worldwide Research &
Development, Groton, Connecticut 06340, United States
*
S Supporting Information
ABSTRACT: The medicinal chemistry and preclinical biology
of imidazopyridine-based inhibitors of diacylglycerol acyltrans-
ferase 2 (DGAT2) is described. A screening hit 1 with low
lipophilic efficiency (LipE) was optimized through two key
structural modifications: (1) identification of the pyrrolidine
amide group for a significant LipE improvement, and (2)
3
insertion of a sp -hybridized carbon center in the core of the
molecule for simultaneous improvement of N-glucuronidation
metabolic liability and off-target pharmacology. The preclinical
candidate 9 (PF-06424439) demonstrated excellent ADMET
properties and decreased circulating and hepatic lipids when orally administered to dyslipidemic rodent models.
INTRODUCTION
esterification of diacylglycerol (DAG) with fatty acyl Coenzyme
A (CoA). In mammals, two structurally unrelated DGAT
■
6
Globally, 39% of adults are estimated to have hyper-
1
enzymes, DGAT1 and DGAT2, have been characterized.
DGAT1 is abundantly expressed in the intestine, where it
cholesterolemia. Reduction of low-density lipoprotein (LDL)
cholesterol via statin therapy has been demonstrated to
decrease adverse cardiovascular events in both the primary
and secondary prevention settings. As a class, statins have been
effective, but there remains significant residual risk for
cardiovascular events in patients unable to achieve LDL
7
plays a critical role in the absorption of dietary lipid. DGAT2 is
2
found in the liver and adipose tissue and, unlike DGAT1,
8
exhibits exquisite substrate specificity for DAG. Extensive
medicinal chemistry efforts have led to the development and
3
characterization of selective small-molecule inhibitors of
treatment target with current therapies. More recently, it has
9
DGAT1. Although these molecules exhibited an attractive
been proposed that plasma triglyceride (TG) levels causally
associate with coronary artery disease; however, the cardiovas-
cular benefit of pharmacological TG lowering, in isolation of
preclinical profile, clinical development has been hindered by
1
0,11
significant gastrointestinal side effects.
Nevertheless,
4
Novartis’s DGAT1 inhibitor LCQ908 (pradigastat) remains
in phase III trials for familial chylomicromia syndrome.
Targeted disruption of the murine Dgat2 gene results in
postnatal lethality and, as a result, much of our understanding
of the physiological function of DGAT2 is derived from
preclinical studies using antisense oligonucleotides (ASO) and
other lipid parameters, has not been established. Dietary TGs
are transported on chylomicrons, whereas TGs released and
secreted from the liver are carried on very-low density
lipoprotein (VLDL). In addition to the role of VLDL in
hepatic TG secretion, VLDL is the primary metabolic precursor
of LDL. One potential mechanism to reduce both circulating
TG and LDL would be the blockade of hepatic VLDL
secretion. However, therapies that directly target this pathway
have limited therapeutic utility due to concerns over the5
accumulation of excess liver fat and gastrointestinal toleration.
Diacylglycerol acyltransferases (DGAT) catalyze the terminal
step in the synthesis of triacylglycerol (TG) through the
12
overexpression. These studies consistently demonstrated that
ASO-mediated knockdown of hepatic DGAT2 resulted in
decreased VLDL secretion, lower plasma cholesterol and TG
Received: June 27, 2015
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1. General Synthetic Routes to Compounds 1−9
a
Conditions: (a) Et N, CH CN, or DMSO, 80−110 °C; (b) H , Pd/C, EtOH, (HCl); (c) R CHO, Na S O , EtOH−H O, 110 °C; (d) (i) H , Pd/
3
3
2
3
2
2
4
2
2
C, Boc O, (ii) R CO H, T3P, NMM, THF, reflux, (iii) MsOH, HOAc, then NaOAc; (e) R CHO, sulfur, DMF, toluene, 85 °C; (f) (i) BOP, NMM,
2
3
2
3
DMF, 50 °C, (ii) NaOMe, MeOH, iBuOH, 110 °C; (g) R C(NH)OEt, HOAc, Et N, EtOH, 100 °C.
3
3
1
3
levels, and reduced hepatic lipid burden. Improved hepatic
triaminopyridine D by hydrogenation. The free base of D was
prone to air oxidation, which was mitigated by immediate use
of the free base or by isolation of the dihydrochloride salt that
was stable upon storage. Imidazopyridine products E were
formed by reaction of triaminopyridine D with either
aldehyde, carboxylic acid, or imidate reagents, as shown in
Scheme 1. Imidazopyridine 1 was synthesized by cyclization of
13a
insulin resistance and glycemic control were also reported.
Although the precise molecular mechanisms underlying these
observations are not clear, it was hypothesized that inhibition of
DGAT2 led to suppression of de novo lipogenesis and
induction of oxidative pathways as a result of a transient
17
18
13a,c
increase in lipid intermediates.
Thus, the inhibition of
DGAT2 is an attractive therapeutic hypothesis which, if
successful, would lead to improved plasma lipid profile,
decreased hepatic fat accumulation, and improved glycemic
control in patients with metabolic disease.
triaminopyridine D with aldehyde R CHO. Imidazopyridine 6
3
was accessed by treatment of D with carboxylic acid R CO H in
3
2
a two-step amidation−cyclization sequence. Imidazopyridines
5, 8, and 9 were prepared by condensation of D with imidates
R3C(NH)OEt.
The cyclopropyl-substituted pyrazole carboxylic acid and
imidate intermediates used in the synthesis of compounds 7, 8,
and 9 were prepared as shown in Scheme 2. Bis-alkylation of
To date, several small-molecule DGAT2 inhibitors have been
reported, but most have suffered from weak to modest in vitro
9b,14,15
potencies, and no in vivo evaluations disclosed.
We
sought to identify an orally bioavailable, potent, and selective
small-molecule DGAT2 inhibitor that would (1) allow us to
evaluate the physiological function of DGAT2 in vivo in
preclinical models and (2) ultimately lead to a candidate with
the potential to evaluate the clinical utility of DGAT2 inhibition
in patients. Herein we describe the medicinal chemistry efforts
to achieve these objectives, culminating in the discovery of the
preclinical candidate 9 (PF-06424439).
a
Scheme 2. Synthesis of Pyrazole Intermediates
CHEMISTRY
■
The C2-aryl and C2-alkyl imidazopyridines described in this
work were synthesized by the general methods shown in
Scheme 1. The piperidine-3-carboxamides A, readily prepared
from N-protected piperidine-3-carboxylic acid, were reacted
a
Conditions: (a) for Y = CO tBu, LiHMDS, THF, 1,3,2-dioxathiolane-
2
with chloro-nitropyridine B under S Ar conditions to provide
N
2
,2-dioxide, −78 to 23 °C, or for Y = CN, NaH, 1,2-dibromoethane,
the diaminonitropyridine C. Several methods were used to
convert intermediate C to the fully functionalized imidazopyr-
idine core E. In one step, diaminonitropyridine C was
transformed directly to imidazopyridines (R)-1, (S)-1, 2, and
DMSO, 0−23 °C; (b) TFA, DCM; (c) sodium metal, EtOH, 70 °C.
the acetate ester or acetonitrile with 1,3,2-dioxathiolane-2,2-
3
by reductive cyclization with aldehydes R CHO using sodium
dioxide or 1,2-dibromoethane formed the cyclopropyl inter-
3
1
6
19
dithionite. Alternatively, in a multistep sequence, diaminoni-
tropyridine C was reduced and monoprotected in situ as a tert-
butyl carbamate. Amidation with carboxylic acid R CO H,
mediate G. Cleavage of the ester G (Y = CO tBu) afforded
2
acid H, while imidate I was derived from the corresponding
nitriles G (Y = CN) via sodium ethoxide addition.
3
2
followed by Boc deprotection and cyclization, afforded the
imidazopyridine product such as compound 7. In an alternative
two-step sequence, diaminonitropyridine C was first reduced to
Exploration of SAR of the piperidine carboxamide group via
parallel synthesis required a change in the order of steps to
allow for late-stage amidation of intermediate K as shown in
B
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 3. Synthesis of Carboxylic Acid K for Parallel Synthesis Variation of the Piperidine Amide
a
Conditions: (a) (i) iPr NEt, DMF, 80 °C, (ii) Zn, HCO NH , DMF, 45 °C; (b) HOAc, DMF, 60 °C.
2
2
4
Scheme 3. In analogous fashion to the routes described in
Scheme 1, piperidine-3-carboxylic acid was reacted with
and all other acyltransferase assays. When the Km value for
DAG (or MAG) was greater than its solubility limit, the highest
substrate concentration of the substrate achievable without
precipitation was chosen as the final assay condition. Each
preparation of the enzyme was titrated, and the reaction time
course was analyzed to ensure that the final reaction conditions
were chosen from the linear range of the reaction.
chloropyridine B under S Ar conditions, followed by reduction
N
of the nitro group to afford triaminopyridine J. Reaction of J
with the appropriate aldehyde afforded the parallel synthesis
substrate K. For example, compound 4 was synthesized in a
library of amidation products derived from acid K.
Mechanistic studies revealed that some but not all DGAT2
inhibitors from the imidazopyridine series described herein are
reversible, time-dependent inhibitors. These inhibitors bind to
DGAT2 through a rapid equilibrium followed by a much slower
isomerization of the initial enzyme−inhibitor complex to form a
much higher affinity complex. On the basis of this mechanism,
IC50 values were determined after a 120 min preincubation of
IN VITRO PHARMACOLOGY
■
DGAT2 activity was determined by measuring the incorpo-
14
ration of the [1- C]decanoyl moiety into triacylglycerol using
1
4
[
1- C]decanoyl-CoA and 1,2-didecanoyl-sn-glycerol as sub-
strates. After partitioning by the addition of a phase partition
1
4
scintillation fluid, the generated product [ C]tridecanoyl-
glycerol was quantified from the upper organic phase. We
have utilized a DGAT2 membrane fraction prepared from an
Sf9 insect cell expression system as a source of DGAT2
enzyme. In this system, we observed that the host cells
possessed an enzyme activity that efficiently hydrolyzed
21
DGAT2 with inhibitors.
RESULTS AND DISCUSSION
Initial hit compound 1 (Figure 1) was identified through
screening a subset of the Pfizer compound collection.
■
1
1
4
4
14
[
[
C]acyl-CoA to [ C]fatty acid and CoA. The generated
C]fatty acid products readily partitioned into the organic
1
4
layer along with the DGAT2 reaction product [ C]-
triacylglycerol, which precluded accurate quantitation of
DGAT2 activity without further separation steps. Furthermore,
1
4
under these conditions, [ C]acyl-CoA substrate concentration
decreases as the reaction proceeds. Others have previously
observed the hydrolyzing activity while running DGAT1−2 and
MGAT1−3 assays, where the quantitation of the TG/DAG
products was achieved after a separation step, typically by
2
0
TLC.
We have speculated that this acyl-CoA hydrolytic enzyme
activity is likely due to thioesterase(s) which belong(s) to the
serine hydrolase family of enzymes. We therefore assessed a
general serine hydrolase inhibitor methyl arachidonyl fluo-
rophosphonate (MAFP) and found that MAFP at 0.1−1 μM
indeed inhibited the thioesterase activity completely without
affecting DGAT2 activity. As expected, the IC50 values for
DGAT2 inhibitors were not affected by MAFP. Inhibition of
this endogenous thioesterase activity by MAFP allowed us to
carry out the DGAT2 assay by phase partitioning and
quantitation of TG in the upper organic layer without
separation steps in a 384-well plate format as described under
the Experimental Section. Similarly, we have observed that
MAFP inhibited endogenous thioesterase activity without
affecting DGAT1 and MGAT1−3 activity and thus utilized
MAFP in these assays that were used for selectivity assessment
of DGAT2 inhibitors. The membranes prepared from Sf9 cell
expression system were utilized as sources of these
Figure 1. Profile of screening hit 1 and its truncated analogue 2.
21
Compound 1 is a reversible inhibitor of DGAT2 and showed
modest inhibitory potency for DGAT2, with >10-fold
selectivity against MGAT2 and DGAT1. Compound 1
exhibited a less than ideal profile as a starting point, with low
22
lipophilic efficiency (LipE = pIC50 − cLogP) and high
intrinsic clearance in human liver microsomes (HLM), which
was consistent with its high lipophilicity (cLogP = 4.8).
However, we hypothesized that a compound design strategy
emphasizing physicochemical property improvement, coupled
to the large chemical space that could be rapidly accessed from
the parallel synthesis-enabled scaffold of 1, would provide an
opportunity for rapid hit optimization. Initial structure−activity
relationship (SAR) examinations around 1 revealed that (1)
removal of the diethyl amide group on the piperidine ring of 1
led to a significant decrease in potency (data not shown), (2)
2
1
acyltransferases.
Substrate concentrations for acyl-CoA and DAG (or
monoacylglycerol (MAG)) in DGAT1−2, and MGAT1−3
assays were established at (or near) the K values for DGAT2
the (R)-enantiomer was the eutomer (DGAT2 IC = 873 nM
m
50
C
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Optimization of Compound 2
a
b
Human DGAT2 potency reported as the geometric mean of at least three replicates. Numbers in parentheses are 95% confidence interval. Tested
c
as a formate salt. Data of single experiment.
and 19.9 μM for (R)-1 and (S)-1, respectively), and (3)
truncation of the diethyl amide to the corresponding dimethyl
amide retained potency (2, Figure 1). On the basis of this SAR,
the initial medicinal chemistry effort focused on decreasing the
lipophilicity of the most hydrophobic region of 2, the meta-
Table 2. Profile of Compound 5
OCF H-Ph group.
2
Table 1 summarizes key SAR from the optimization of
compound 2. With slightly increased lipophilicity and
comparable potency, the meta-cyclopropyl-phenyl analogue 3
possessed similar LipE to that of 2. Examination of heteroaryl
analogues of the phenyl ring showed that pyrimidine 4,
although somewhat less potent than the phenyl analogue 3,
provided a substantial reduction in lipophilicity; its significantly
improved LipE reflected a good balance of properties and,
importantly, the decreased lipophilicity also translated to
improved HLM stability. Compound 4, with its relatively low
cLogP, became the next starting point for LipE-guided potency
optimization. Considering the importance of the carboxamide
group for DGAT2 inhibition in this series (vide supra), it was
reasoned that small changes in the amide substituents might
significantly influence DGAT2 potency. Maintaining the
cyclopropyl-pyrimidine of 4 and limiting cLogP < 3 in order
to minimize possible increases in clearance, a small set of amide
groups was examined. Among those tested, tertiary amides
exhibited the highest potency; cyclic derivatives such as
pyrrolidine amide 5 demonstrated substantially improved
LipE and absolute potency. Compound 5 showed higher
metabolic turnover in HLM relative to 4, however, likely due to
increase in lipophilicity from 4 and the introduction of the site
of metabolism (pyrrolidine ring). On the basis of its good
physicochemical properties and attractive potency, compound 5
was selected for further characterization (Table 2).
2
3
MW 417, eLogD 2.4, TPSA 91
a
DGAT2 IC50 ± SEM (nM)
18 ± 2
>40/>50
50
a
MGAT2/DGAT1 IC50 (μM)
HLM CLint,app (μL/min/mg)
Human hepatocytes CLint,app (μL/min/million cells)
HLM-UGT CLint,app (μL/min/mg)
71
316
b
off-target activities >50% at 10 μM
8 hits
(CEREP panel/PDE selectivity panel)
a
Human DGAT2, MGAT2, or DGAT1 potency reported as the
geometric mean of at least three replicates. SEM = standard error of
the mean. Adenosine-A2A K = 3700 nM; muscarinic M1 K = 4500
nM; histamine H1 agonism, 50% effect at 10 μM; alpha 1a agonism,
2% effect at 10 μM. PDE IC s (nM): 3300 (1B), 1980 (5A), 9410
b
i
i
6
50
(6A), 1490 (11).
revealed several weak, but undesired off-target activities, notably
in a subset of PDEs (Table 2, footnote b). Metabolically, high
turnover in both HLM and human hepatocytes was also
observed. To determine the contribution of glucuronidation
pathway to overall metabolism of 5, follow-up studies using
alamethicin activated HLM with UDP-glucuronosyltransferase
(UGT) enzyme cofactors (HLM-UGT CLint assay) were
employed. It was revealed that 5 was rapidly metabolized in the
presence of the UGT cofactor uridine diphosphate-glucuronic
acid (UDPGA), which implicated an unanticipated phase II
24
25
Compound 5 exhibited an excellent selectivity profile against
other acyltransferases (MGAT2, DGAT1). Broader off-target
selectivity screening (CEREP panel, PDE selectivity panel)
metabolism via presumed N-glucuronidation on the imida-
zopyridine ring (HLM-UGT CLint,app = 316 μL/min/mg, Table
2).
D
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Design strategy for simultaneous optimization of off-target selectivity and N-glucuronidation.
Table 3. Optimization of Heteroaromatic Groups in Imidazopyridines with Cyclopropyl Spacers
a
b
Human DGAT2 potency reported as the geometric mean of at least three replicates. Numbers in parentheses are 95% confidence interval. Not
determined.
To address both the off-target pharmacology and the N-
glucuronidation liabilities, a single design strategy was
employed by increasing three-dimensional shape of the core
cyclopropyl was selected as the spacer between the
imidazopyridine and the terminal heteroaromatic ring (Figure
2, highlighted in blue). It was envisioned that variation of the
heteroaromatic ring would enable the fine-tuning of phys-
icochemical properties and potency. The SAR for selected
compounds within the scope of this design hypothesis is
summarized in Table 3.
Among initial heteroaromatic derivatives synthesized, the 2-
pyridyl analogue 6 showed no detectable turnover in the HLM-
UGT CLint assay, providing validation for the hypothesis of
mitigating N-glucuronidation by means of three-dimensionality
from the cyclopropyl ring. Pyridine 6 showed a 10-fold decrease
in DGAT2 potency but moderate clearance in HLM. The N-
3
of the molecule through introduction of an sp -hybridized
carbon center (Figure 2). The added three dimensionality was
26
expected to improve off-target selectivity. With regard to
metabolism, we hypothesized that increased steric bulk around
the likely sites of N-glucuronidation (imidazole N−H) might
27
diminish interaction with the UGT enzyme(s). It was deemed
3
important to limit the increase in lipophilicity from the sp -
center in order to mitigate increased CYP-mediated metabo-
lism. To balance the desired steric bulk with undesired
lipophilicity (and to avoid introducing another stereocenter),
E
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
linked pyrazole analogue 7 was examined as a structural mimic
of the 2-pyridyl analogue 6, with the expectation that its lower
lipophilicity would diminish metabolic turnover in HLM. While
the potency of pyrazole 7 did not increase as compared to that
of pyridine 6, its improved LipE, low metabolic turnover in
HLM, and relatively low lipophilicity prompted us to examine
further substitution on the pyrazole. By focusing on
substituents that were likely not to be metabolically labile, we
hoped to achieve a >10-fold increase in potency while
maintaining an acceptable clearance profile. The fluoro- and
chloro-pyrazole analogues 8 and 9 exhibited promising
combinations of potency and clearance properties. Chloro-
pyrazole 9 possessed the highest potency and LipE among
analogues tested while maintaining an acceptable clearance
profile in HLM and no measurable turnover in the HLM-UGT
of a DGAT1 inhibitor. The DGAT1 inhibitor (PF-04620110)
14
reduced [ C]glycerol incorporation into TG by ∼30%. The
addition of 9 on top of a DGAT1 inhibitor further lowered TG
synthesis by ∼40−50% (data not shown). In this assay,
compound 9 showed potent inhibitory activity with an IC of
5
0
1.3 nM.
A crystalline form of 9 was obtained as a methanesulfonic
acid salt (basic pK of 9 = 4.1), which exhibited excellent
a
thermodynamic solubility in FaSSIF (simulated fasted intestinal
fluid with bile salts) (0.83 mg/mL). The high solubility,
coupled with reasonable metabolic stability and high passive
permeability, prompted evaluation of the in vivo pharmacoki-
netic properties of 9 (Table 5). Compound 9 showed moderate
Table 5. in Vivo Pharmacokinetic Profile of 9 in Rat and
2
8
a
CLint assay. Furthermore, compound 9 did not show any
significant off-target activity (<50% effect at 10 μM) in the
CEREP and PDE selectivity panels, confirming the improve-
ment in promiscuity over the prototype 5. On the basis of
positive attributes mentioned above, compound 9 was further
profiled in a series of in vitro assays (Table 4).
Dog
dose
t1/2
(h)
CL,p
(mL/min/kg)
Vdss
(L/kg)
species (mg/kg) route
F (%)
b
,
f
h
rat
1
5
1
iv
po
iv
1.39
2.37
1.15
1.65
18.4
NA
1.16
NA
NA
c,g
rat
∼100
NA
e,
f
dog
dog
18.4
NA
2.01
NA
d,
g
5
po
∼100
Table 4. Summary of Pharmacology and ADME Data for
Compound 9
a
For rat PK studies, male Wistar-Han rats were utilized. For dog PK
studies, male beagle dogs were used. All data reported here are means
b
of two experiments. Vehicle was 10% DMSO/90% of 40% SBECD in
c
d
water. Vehicle was 0.5% methylcellulose/0.1% Tween 80. Vehicle
e
was 0.5% methylcellulose. Vehicle was 10% PEG200/90% of 12%
SBECD in water. Free base of 9 was dosed. Crystalline,
methanesulfonic acid salt of 9 was dosed. NA = not applicable.
f
g
h
clearance in rat and dog following intravenous administration,
consistent with the moderate metabolic stability observed in
liver microsomes of the respective species (CLint,app [μL/min/
mg] in rat and dog liver microsomes = 20 and 26, respectively).
2
3
MW 440, eLogD 2.2, TPSA 83, pK = 4.1 (basic)
a
DGAT2 IC50 ± SEM (nM) (human/rat/dog)
14 ± 1/38 ± 4/16 ± 1
>50
IC50 (MGAT1−3, DGAT1) (μM)
fresh human hepatocytes IC50 ± SEM (nM)
CEREP panel
Moderate steady-state volume of distribution (Vd ) resulted in
ss
1.3 ± 0.5
<50% at 10 μM
a short-to-moderate half-life. High solubility and high passive
permeability of 9 translated into high oral bioavailability (F ∼
HLM CLint,app (μL/min/mg)
HLM-UGT CLint,app (μL/min/mg)
human hepatocytes CLint,app
24
1
00%) in both species.
On the basis of its excellent in vitro pharmacological profile
<1.9
9.1
(
μL/min/million cells)
and favorable in vivo pharmacokinetic properties, compound 9
was selected to evaluate the acute effects of DGAT2 inhibition
on hepatic secretion of TGs in male Sprague−Dawley rats
6
RRCK Papp (cm/s)
33.9 × 10−
thermodynamic solubility
methanesulfonic acid salt)
0.83 mg/mL
(
(
Figure 3). Dose-dependent reductions in plasma TG were
FaSSIF (simulated fasted intestinal
observed after an oral administration of 9, with a maximal effect
of >50% reduction in plasma TG relative to vehicle control.
The magnitude of the effect on plasma TG reduction correlated
with the plasma free drug levels of 9 (Figure 3). In contrast to
isolated human hepatocytes (vide supra), the observed changes
in TGs were achieved in the absence of DGAT1 inhibitor in
this setting.
fluid with bile salts)
fu,p (human/rat/dog)
0.154/0.117/0.111
Compound 9 inhibited DGAT2 of different species (human,
rat, and dog) with similar potency. Among related acyltrans-
ferases, no significant inhibition was observed for compound 9
(
up to 50 μM) against human MGAT2 or MGAT3, DGAT1,
To evaluate the effects of DGAT2 inhibition in a
dyslipidemia rodent model, low-density lipoprotein receptor
knockout mice (Ldlr− ) were treated with 60 mg/kg/day
or mouse MGAT1, indicating a high selectivity (>2000-fold)
against these enzymes. To determine the effect of 9 on TG
synthesis in a cellular system, freshly isolated human
hepatocytes were cultured in the presence of [1,3- C]-glycerol
and the incorporation of radiolabel into TG was monitored
using TLC. To monitor DGAT2 activity, a selective DGAT1
inhibitor {trans-4-[4-(4-amino-5-oxo-7,8-dihydropyrimido[5,4-
f][1,4]oxazepin-6(5H)-yl)phenyl]cyclohexyl}acetic acid (PF-
/−
(administered as 30 mg/kg BID) 9 for 3 days. When
1
4
−/−
maintained on standard laboratory chow, Ldlr
mice exhibit
elevated plasma triglycerides and cholesterol, however, when
fed a high-fat, high-cholesterol diet (HFHC), these animals
develop pronounced hyperlipidemia with total plasma choles-
terol levels in excess of 1500 mg/dL (Figure 4). Treatment of
2
9
−/−
0
4620110, 3 μM) was added to completely inhibit
Ldlr
mice on a HFHC diet with 9 resulted in 61% (p <
endogenous DGAT1 activity. In this cell-based system,
DGAT1 appeared to play a redundant role and it was not
possible to consistently measure DGAT2 activity in the absence
0.001) and 34% (p < 0.001) reductions in plasma TG and
cholesterol levels, respectively (Figure 4A,B). The HFHC diet-
induced increase in circulating TG levels was completely
F
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
CONCLUSION
■
In summary, the first class of small-molecule DGAT2 inhibitors
suitable for in vivo testing has been identified by optimization
of a hit from high-throughput screening. The first round of
optimization afforded a potent DGAT2 inhibitor such as 5 with
significant improvement in LipE. The strategic introduction of
3
an sp -carbon spacer to simultaneously improve both N-
glucuronidation and off-target selectivity was a key design
element leading to the discovery of the preclinical candidate 9
(
PF-06424439). To our knowledge, PF-06424439 was the first
orally bioavailable small-molecule DGAT2 inhibitor evaluated
in vivo, demonstrating favorable effects on hepatic and
15
circulating lipid levels in rats. More detailed in vitro and in
vivo pharmacological characterization of PF-06424439 will be
described in future publications.
Figure 3. Acute effects of 9 on plasma triacylglycerol levels. Dose-
dependent effects of 9 on plasma TG levels in sucrose-fed rats.
Triglyceride (TG) levels were determined in blood drawn from the tail
vein 2 h after oral administration of the indicated dose of 9 or vehicle
EXPERIMENTAL SECTION
General Experimental Methods. All chemicals, reagents, and
solvents were purchased from commercial sources and used without
further purification unless otherwise noted. H NMR data are reported
■
1
(
0.5% methylcellulose, n = 8 per group). TG levels in untreated
relative to residual solvent signals and are reported as follows: chemical
shift (δ ppm), multiplicity, coupling constant (Hz), and integration.
The multiplicities are denoted as follows: s, singlet; d, doublet; t,
triplet; q, quartet; sept, septet; m, multiplet; br s, broad singlet; app,
apparent. Silica gel chromatography was performed using a medium
pressure Biotage or ISCO system and columns prepackaged by various
commercial vendors including Biotage and ISCO. Whatman precoated
silica gel plates (250 μm) were used for analytical thin-layer
chromatography (TLC). The terms “concentrated” and “evaporated”
refer to the removal of solvent at reduced pressure on a rotary
evaporator with a water bath temperature not exceeding 60 °C. Purity
of final compounds was assessed by reversed-phase HPLC with UV
detection at 215 nM; all tested compounds were >95% purity unless
otherwise noted. Safety note: Sodium hydride and DMSO can react
explosively, please take appropriate precautions in using this reagent/
solvent combination.
animals were 77.4 ± 6.9 mg/dL. Relationship between plasma free
drug levels and the pharmacodynamic effect (TG lowering). Data are
expressed as percent change from the vehicle-treated group and were
analyzed using one-way ANOVA followed by Dunnett’s multiple
comparison test. *p < 0.05, **p < 0.001, ***p < 0.0001 compared to
vehicle-treated animals. The values shown adjacent to the individual
data points indicate the dose group (mg/kg). Plasma concentrations of
9
that were below the lower limit of quantification using LC-MS/MS
were eliminated from the analysis.
inhibited by 9 (Figure 4A). Nonsignificant decreases in
circulating lipids were also observed following treatment of
chow-fed animals with 9 (Figure 4A,B). To characterize the
effects of 9 on lipoprotein distribution, plasma samples from
HFHC-fed mice were subjected to fast protein liquid
chromatography (FPLC) fractionation. As previously de-
(
R)-(1-(2-(2-Cyclopropylpyrimidin-4-yl)-3H-imidazo[4,5-b]pyridin-
5
-yl)piperidin-3-yl)(pyrrolidin-1-yl)methanone (5). Step 1: (R)-tert-
Butyl 3-(pyrrolidine-1-carbonyl)piperidine-1-carboxylate. To a solu-
tion of (R)-1-(tert-butoxycarbonyl)piperidine-3-carboxylic acid (50.6 g,
3
0
scribed, the vast majority (>90%) of circulating cholesterol
was carried in the VLDL and intermediate-density lipoprotein
2
21 mmol, 1 equiv) in anhydrous tetrahydrofuran (552 mL) at 2 °C
was added 1,1′-carbonyldiimidazole (77.6 g, 460 mmol, 2.1 equiv).
The mixture was stirred at room temperature for 2 h then cooled to 10
°C, whereupon pyrrolidine (74.0 mL, 890 mmol, 4.0 equiv) was added
slowly. The reaction mixture was stirred at room temperature for 18 h.
The solvent was removed under reduced pressure, and water (200
mL) was added to the residue. The mixture was extracted with ethyl
acetate (2×). The combined organics were washed sequentially with
(IDL)/LDL fractions. Treatment with 9 for 3 days resulted in a
dramatic decrease in VLDL-associated cholesterol. Decreases in
IDL/LDL cholesterol were also apparent (Figure 4C). Because
blockade of hepatic VLDL secretion has been associated with
31
hepatic lipid accumulation, hepatic lipid content was assayed.
−/−
Treatment with 9 in HFHC-fed Ldlr
mice resulted in a
1
N aqueous hydrochloric acid (4 × 200 mL) and with a saturated
significant decrease (32%, p < 0.05) in hepatic TG content.
Having demonstrated favorable in vivo efficacy in both acute
and subchronic settings, compound 9 was also profiled in a
panel of standard advanced in vitro safety assays. Compound 9
did not show significant hERG (human ether-a-go-go related
gene) inhibition (patch clamp IC50 = 95 μM; >300-fold
selectivity over DGAT2 IC ). Little or no reversible inhibition
aqueous solution of sodium bicarbonate (2 × 200 mL). The organics
were dried over magnesium sulfate, filtered, and concentrated under
reduced pressure to afford (R)-tert-butyl 3-(pyrrolidine-1-carbonyl)-
piperidine-1-carboxylate as a white solid (58.8 g, 94%). The compound
1
was used in the next step without further purification. H NMR (400
MHz, CD OD) δ 1.38−1.57 (m, 10H), 1.59−1.67 (m, 1H), 1.69−
3
1
2
.77 (m, 1H), 1.80−2.05 (m, 5H), 2.52−2.62 (m, 1H), 2.67−2.94 (m,
5
0
H), 3.33−3.46 (m, 2H), 3.51−3.66 (m, 2H), 3.98−4.12 (m, 2H). MS
of major cytochrome P450s (CYP2B6, 2C8, 2C19, 2D6, 3A4/
) was observed up to 30 μM (IC s > 30 μM), indicating a low
+
(M + H) = 283.3.
5
Step 2: (R)-Piperidin-3-yl(pyrrolidin-1-yl)methanone hydrochlor-
5
0
drug−drug interaction risk from reversible inhibition of these
CYPs. No genetic toxicology risks were identified, with negative
findings in an Ames mutagenicity assay and an in vitro
micronucleus assay in TK6 cells, both in the presence and
absence of the metabolic activation. On the basis of its favorable
overall profile, compound 9 (PF-06424439) was selected as a
preclinical candidate for evaluation in regulatory toxicology
studies.
ide. To a solution of (R)-tert-butyl 3-(pyrrolidine-1-carbonyl)-
piperidine-1-carboxylate (58.8 g, 208 mmol, 1 equiv) in anhydrous
dichloromethane (100 mL) was added hydrogen chloride in 1,4-
dioxane (4M, 260 mL, 1040 mmol, 5.0 equiv). The reaction mixture
was stirred vigorously at room temperature for 1 h. The solvent was
removed under reduced pressure, and the residue was left standing
overnight at room temperature. The residue was triturated with diethyl
ether (250 mL). The diethyl ether was decanted and dichloromethane
was added, followed by sonication in a warm water bath. The solvent
32
G
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Effects of 9 on plasma and hepatic lipids in LDLr knockout mice (Ldlr−/−). Male Ldlr−/− mice were maintained on regular chow or high-
fat, high-cholesterol diet (HFHC) for 2 weeks prior to treatment with either 60 mg/kg/day (30 mg/kg BID) 9 or vehicle (0.5% w/v
methylcellulose) for 3 days. (A) Plasma TG. (B) Total plasma cholesterol. (C) Plasma lipoprotein profiles were generated on plasma from HFHC
diet-fed mice as described in the Experimental Section. Lipid values were corrected for recoveries of total plasma lipid from combined fractions.
VLDL, IDL/LDL, and high-density lipoprotein (HDL) are defined as fractions 4−11, 12−30, and 31−52, respectively. (D) Hepatic TG content.
Data are mean ± SEM from 6−8 animals and were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test.
was removed under reduced pressure, and the resulting solid was
placed under high vacuum at 40 °C for 1 h to afford (R)-piperidin-3-
yl(pyrrolidin-1-yl)methanone hydrochloride (48.41 g, >100%). The
3.33−3.46 (m, 2H), 3.45−3.55 (m, 1H), 3.60−3.70 (m, 1H), 4.41 (br
s, 1H), 4.68 (br s, 1H), 6.23 (d, J = 9.6 Hz, 1H), 8.05 (d, J = 9.6 Hz,
1H). MS (M + H) = 320.4.
+
1
material was used in the next step without further purification. H
Step 4: (R)-(1-(5,6-Diaminopyridin-2-yl)piperidin-3-yl)(pyrrolidin-
1-yl)methanone dihydrochloride. Into a Parr bottle was added 10%
palladium-on-carbon (50% wet, 1.49 g) followed by ethanol (10 mL)
which was previously bubbled with nitrogen and cooled to 0 °C. (R)-
(1-(6-Amino-5-nitropyridin-2-yl)piperidin-3-yl)(pyrrolidin-1-yl)-
methanone (10.6 g, 33.2 mmol, 1 equiv) in ethanol (84 mL), and a
solution of concentrated hydrochloric acid (8.59 mL, 99.6 mmol, 3.0
equiv) in ethanol (10 mL) was added to the Parr bottle under a
nitrogen atmosphere. The reaction mixture was purged with nitrogen
and evacuated (3×). The mixture was submitted to a hydrogen
atmosphere (45 psi). A drop in pressure was observed, and hydrogen
pressure was increased to 46 psi. This process was repeated several
times over a period of 30 min. The mixture was shaken for a total of 1
h and filtered through Celite rinsing with ethanol (0.7 L) and
methanol (1.5 L). The filtrate was concentrated, and the resulting solid
was dried under high vacuum to afford the title compound (12.0 g,
NMR (400 MHz, CD OD) δ 1.75−2.05 (m, 8H), 3.04−3.17 (m, 2H),
3
+
3
.19−3.28 (m, 3H), 3.35−3.66 (m, 4H). MS (M + H) = 183.3.
Step 3: (R)-(1-(6-Amino-5-nitropyridin-2-yl)piperidin-3-yl) (pyrro-
lidin-1-yl)methanone. To a suspension of (R)-piperidin-3-yl-
pyrrolidin-1-yl)methanone hydrochloride (208 mmol, 1.1 equiv)
(
from step 2 in anhydrous acetonitrile (200 mL) at 10 °C was added
triethylamine (50.0 mL, 358 mmol, 1.9 equiv). The suspension was
poured into a 2 L three-neck flask equipped with a thermometer and a
magnetic stirrer. The original flask was rinsed with anhydrous
acetonitrile and triethylamine (80.0 mL, 573 mmol, 3.0 equiv), with
sonication. The suspensions were combined and cooled to 10 °C, and
6
-chloro-3-nitropyridin-2-amine (32.9 g, 190 mmol, 1 equiv) was
added. The bright-yellow solution was stirred under nitrogen, while
the temperature was increased gradually to 80 °C over 1 h. The
reaction mixture was then cooled to room temperature. The resulting
suspension was filtered, and the solids were rinsed with ethyl acetate.
The filtrate was concentrated under reduced pressure. A saturated
aqueous solution of ammonium chloride (500 mL) was added to the
residue, and the mixture was extracted into ethyl acetate (2×). The
combined organics were washed with brine, dried over magnesium
sulfate, filtered, and concentrated under reduced pressure, and the
resulting residue was dried for 18 h under high vacuum to afford (R)-
1
99%). The material was used without further purification. H NMR
(400 MHz, CD OD) δ 1.71−1.81 (m, 2H), 1.88−1.95 (m, 3H), 1.98−
3
2.12 (m, 3H), 2.89−2.95 (m, 1H), 3.38−3.55 (m, 5H), 3.66−3.72 (m,
1H), 3.97 (d, J = 12.7 Hz, 1H), 3.98 (dd, J = 13.7, 3.7 Hz, 1H), 6.39
+
(d, J = 9.3 Hz, 1H), 7.63 (d, J = 9.0 Hz, 1H). MS (M + H) = 290.2.
Step 5: 2-Cyclopropylpyrimidine-4-carbonitrile. To a solution of 2-
cyclopropylpyrimidine-4-carbaldehyde (29.1 g, 197 mmol, 1.0 equiv)
in N,N-dimethylformamide (200 mL) was added hydroxylamine
hydrochloride (13.9 g, 197 mmol, 1 equiv) and a solution of
triethylamine (35.0 mL, 250 mmol, 1.3 equiv) in N,N-dimethylforma-
mide (40 mL) at room temperature. A 50% solution of 1-
(
1-(6-amino-5-nitropyridin-2-yl)piperidin-3-yl)(pyrrolidin-1-yl)-
methanone as a yellow foam (63.5 g, > 100%), which was used without
1
further purification. H NMR (400 MHz, CD OD) δ 1.47−1.64 (m,
3
1
H), 1.71−2.07 (m, 7H), 2.50−2.70 (m, 1H), 2.98−3.04 (m, 2H),
H
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
propanephosphonic anhydride in N,N-dimethylformamide (140 mL,
keeping the internal temperature at or below 20 °C, followed by
stirring for an additional 1 h at room temperature. The mixture was
cooled to 10 °C, and then a saturated aqueous solution of ammonium
chloride (600 mL) was slowly added. The mixture was stirred at 10 °C
for 15 min and extracted with ethyl acetate (3 × 1 L). The combined
organics were washed with brine, dried over sodium sulfate, filtered,
and concentrated under reduced pressure. The crude material was
purified via flash chromatography (10−100% dichloromethane in
hexanes), and the resulting residue was crystallized from a mixture of
hexanes/diethyl ether to afford 1-(4-chloro-1H-pyrazol-1-yl)-
2
30 mmol, 1.2 equiv) was added, and the reaction mixture was stirred
at 110 °C for 18 h. The mixture was cooled to room temperature, and
a saturated aqueous solution of sodium bicarbonate and ethyl acetate
was added. The mixture was stirred vigorously, and solid sodium
bicarbonate was added. The mixture was filtered to remove
undissolved solids, and the layers were separated. The aqueous layer
was extracted with ethyl acetate (3×). The combined organics were
washed with water and brine, dried over sodium sulfate, filtered, and
concentrated under reduced pressure, and the resulting residue was
dried under high vacuum to afford 2-cyclopropylpyrimidine-4-
1
cyclopropanecarbonitrile (7.65 g, 65%). H NMR (400 MHz+,
carbonitrile (21.0 g, 73%). The material was used without further
CDCl ) δ 1.77−1.82 (m, 4H), 7.45 (s, 1H), 7.60 (s, 1H). MS (M)
3
1
purification. H NMR (400 MHz, CDCl ) δ 1.15−1.22 (m, 4H),
= 167.
3
2
.28−2.35 (m, 1H), 7.36 (d, J = 4.8 Hz, 1H), 8.76 (d, J = 4.8 Hz, 1H).
Step 6: Ethyl 2-cyclopropylpyrimidine-4-carbimidate. To a solution
Step 2: Ethyl 1-(4-chloro-1H-pyrazol-1-yl)cyclopropanecarbimidate.
To a solution of 1-(4-chloro-1H-pyrazol-1-yl)cyclopropanecarbonitrile
(17.3 g, 103 mmol, 1 equiv) in ethanol (50 mL) was added sodium
ethoxide (prepared from sodium metal (3.2 g, 139 mmol) in 150 mL
of ethanol, 1.3 equiv). The reaction mixture was stirred at 70 °C for 2
of 2-cyclopropylpyrimidine-4-carbonitrile (20.97 g, 144.5 mmol, 1
equiv) in ethanol (100 mL) was added a solution of sodium ethoxide
(
0
prepared from sodium metal (1.8 g, 78.9 mmol) in 30 mL of ethanol,
.55 equiv). The reaction mixture was stirred at room temperature for
h
t o a ff o r d e t h y l 1 - ( 4 - c h l o r o - 1 H - p y r a z o l - 1 - y l ) -
3
0 min. Diethyl ether (300 mL) and a saturated aqueous solution of
cyclopropanecarbimidate, which was used for the next step without
ammonium chloride (50 mL) were added sequentially. The mixture
was extracted with diethyl ether (3 × 100 mL). The combined
organics were washed with brine (2 × 50 mL), dried over magnesium
sulfate, filtered, and concentrated under reduced pressure, and the
further purification. An aliquot of the solution was concentrated for
1
analysis. H NMR (400 MHz, CDCl ): δ 1.24 (t, J = 7.24 Hz, 3H,
3
obscured by residual ethanol), 1.42−1.48 (m, 2H), 1.69−1.74 (m,
2H), 4.15 (q, J = 7.0 Hz, 2H), 7.50 (s, 1H), 7.51 (s, 1H).
resulting residue was dried under high vacuum. The material was used
Step 3: (R)-(1-(2-(1-(4-Chloro-1H-pyrazol-1-yl)cyclopropyl)-3H-
imidazo[4,5-b]pyridin-5-yl)piperidin-3-yl)(pyrrolidin-1-yl)methanone
(9). To a solution of (R)-(1-(5,6-diaminopyridin-2-yl)piperidin-3-
yl)(pyrrolidin-1-yl)methanone dihydrochloride (5.50 g, 15.2 mmol, 1
equiv) and acetic acid (17.4 mL, 305 mmol, 20 equiv) in ethanol (20
mL) was added a solution of ethyl 1-(4-chloro-1H-pyrazol-1-
yl)cyclopropanecarbimidate (3.25 g, 15.2 mmol, 1.0 equiv) in ethanol
(30 mL) followed by triethylamine (12.7 mL, 91.2 mmol, 6.0 equiv).
The resulting mixture was purged with nitrogen. The reaction mixture
was stirred at 100 °C for 18 h. The mixture was cooled to room
temperature, and the solvent was removed under reduced pressure.
The residue was partitioned between a saturated aqueous solution of
ammonium chloride (50 mL) and dichloromethane (50 mL). The
aqueous layer was extracted with dichloromethane (50 mL). The
combined organics were washed with a saturated aqueous solution of
sodium bicarbonate, and the resulting aqueous layer was extracted with
dichloromethane (50 mL). The organics were combined, washed with
brine, dried over sodium sulfate, filtered, and concentrated under
reduced pressure. The crude material was purified via flash
1
without further purification. H NMR (500 MHz, CDCl ) δ 1.06−1.13
3
(
1
m, 2H), 1.16−1.22 (m, 2H), 1.41 (t, J = 7.1 Hz, 3H), 2.27−2.35 (m,
H), 4.40 (q, J = 7.2 Hz, 2H), 7.48 (d, J = 5.1 Hz, 1H), 8.72 (d, J = 4.9
Hz, 1H) 9.26 (br s, 1H).
Step 7: (R)-(1-(2-(2-Cyclopropylpyrimidin-4-yl)-3H-imidazo[4,5-
b]pyridin-5-yl)piperidin-3-yl)(pyrrolidin-1-yl)methanone (5). To a
suspension of (R)-(1-(5,6-diaminopyridin-2-yl)piperidin-3-yl)-
(
pyrrolidin-1-yl)methanone dihydrochloride (1.0 g, 2.8 mmol, 1.0
equiv) and acetic acid (2.5 mL, 43 mmol, 15 equiv) in ethanol (7 mL)
was added triethylamine (1.5 mL, 10.8 mmol, 3.9 equiv) and a solution
of ethyl 2-cyclopropylpyrimidine-4-carbimidate (530 mg, 2.8 mmol, 1
equiv) in ethanol (3 mL). The reaction mixture was heated under
reflux for 1 h and cooled to room temperature. The solvent was
removed under reduced pressure, and ethyl acetate (100 mL) was
added to the residue followed by water (10 mL) and a saturated
aqueous solution of ammonium chloride (50 mL). The layers were
separated, and the organics were washed sequentially with a saturated
aqueous solution of sodium bicarbonate (75 mL) and brine (50 mL),
dried over magnesium sulfate, filtered, and concentrated under
reduced pressure. To the resulting residue was added dichloromethane
chromatography (0−5% methanol in dichloromethane) to afford 9
1
(6.33 g, 94%). H NMR (500 MHz, CDCl ) δ 1.55−1.70 (m, 1H),
3
(
4 mL) and methyl tert-butyl ether (35 mL). The mixture was stirred
1.72−2.05 (m, 11H), 2.60−2.70 (m, 1H), 2.90−2.95 (m, 1H), 3.03−
3.10 (m, 1H), 3.42−3.50 (m, 3H), 3.56−3.62 (m, 1H), 4.20−4.25 (m,
1H), 4.38−4.44 (m, 1H), 6.62−6.68 (m, 1H), 7.59 (s, 1H), 7.63 (s,
at 40 °C until it was homogeneous. The solution was cooled to room
temperature and stirred for 18 h. The resulting solids were filtered and
1
+
rinsed with ether to afford 5 (1.02 g, 62%). H NMR (400 MHz,
1H), 7.66−7.72 (m, 1H). MS (M + H) = 440.
CDCl ) δ 1.10−1.17 (m, 2H), 1.18−1.23 (m, 2H), 1.57−1.70 (m,
(R)-(1-(2-(1-(4-Chloro-1H-pyrazol-1-yl)cyclopropyl)-3H-imidazo-
[4,5-b]pyridin-5-yl)piperidin-3-yl)(pyrrolidin-1-yl)methanone Meth-
anesulfonate (9·MsOH). To a solution of 9 (56.5 g, 128 mmol, 1
equiv) in acetonitrile (130 mL) was added methanesulfonic acid (8.33
mL, 128 mmol, 1.0 equiv). The mixture was stirred at room
temperature for 18 h. The mixture was filtered, and the solids were
rinsed with acetonitrile. The solids were collected and dried under
3
1
2
3
6
1
H), 1.82−2.05 (m, 7H), 2.27−2.34 (m, 1H), 2.63−2.72 (m, 1H),
.97−3.07 (m, 1H), 3.14−3.22 (m, 1H), 3.43−3.54 (m, 3H), 3.58−
.67 (m, 1H), 4.37 (d, J = 13.7 Hz, 1H), 4.53 (d, J = 12.9 Hz, 1H),
.76 (d, J = 9.0 Hz, 1H), 7.82−7.90 (m, 2H), 8.67 (d, J = 5.3 Hz, 1H),
+
0.27 (br s, 1H). MS (M + H) = 418.5.
(
R)-(1-(2-(1-(4-Chloro-1H-pyrazol-1-yl)cyclopropyl)-3H-imidazo-
1
[
4,5-b]pyridin-5-yl)piperidin-3-yl)(pyrrolidin-1-yl)methanone (9).
high vacuum for 1 h to afford 9·MsOH (62.5 g, 90%). H NMR (400
Step 1: 1-(4-Chloro-1H-pyrazol-1-yl)cyclopropanecarbonitrile. Safety
note: Sodium hydride and DMSO can react explosively, please take
appropriate precautions in using this reagent/solvent combination.
Into a four-neck, 2 L round-bottom flask, previously dried with a heat
gun under high vacuum, was added DMSO (175 mL). The flask was
placed in a 10 °C bath, and sodium hydride (60% oil dispersion, 17.0
g, 300 mmol, 4.2 equiv) was added portionwise with stirring under
nitrogen. The resulting suspension was stirred for 10 min before a
solution of 2-(4-chloro-1H-pyrazol-1-yl)acetonitrile (10.0 g, 70.6
mmol, 1 equiv) and 1,2-dibromoethane (18.3 mL, 213 mmol, 3.0
equiv) in DMSO (175 mL) was added dropwise over a period of 30
min. The internal temperature was kept between 13 and 20 °C during
the addition process. The reaction mixture was stirred for 5.5 h while
MHz, CDCl ) δ 1.65−1.75 (m, 1H), 1.84−2.13 (m, 11H), 2.75−2.85
3
(m, 1H), 2.87 (s, 3H), 3.28−3.35 (m, 1H), 3.39−3.57 (m, 4H), 3.65−
3.72 (m, 1H), 4.02−4.09 (m, 1H), 4.11−4.17 (m, 1H), 6.73−6.82 (m,
1H), 7.55 (s, 1H), 7.76 (s, 1H), 8.13 (d, J = 9.0 Hz, 1H). MS (M +
+
H) = 440; mp = 212−214 °C. Anal. Calcd for C H ClN O·
22
26
7
CH O S: C, 51.53; H, 5.64; N, 18.29; Cl, 6.61; S, 5.98. Found: C,
4
3
51.28; H, 5.65; N, 18.19; Cl, 6.56; S, 5.96.
DGAT2 Assay and Determination of IC50 Values. DGAT2
activity was determined by measuring the incorporation of the
14
14
[1- C]decanoyl moiety into triacylglycerol using [1- C]decanoyl-
CoA and 1,2-didecanoyl-sn-glycerol as substrates. The final assay
1
4
mixture contained 50 mM Hepes−NaOH, pH 7.4, 6 μM [1- C]-
decanoyl-CoA (50 mCi/mmol, PerkinElmer, Waltham, MA), 25 μM
I
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1
,2-didecanoyl-sn-glycerol (Avanti Polar Lipids, Alabaster, AL), 10 mM
range, and data analysis for IC50 determinations were identical to those
described in the DGAT2 assay.
MgCl , 100 nM methyl arachidonyl fluorophosphonate (MAFP,
2
Cayman Chemical, Ann Arbor, MI), 0.01% BSA (fatty acid free,
Sigma-Aldrich, St. Louis, MO), 5% DMSO, 2.5% acetone, and 0.1 μg
of the detergent-solubilized DGAT2 membrane. The reactions were
carried out in 384-well white Polyplates (PerkinElmer) in a total
volume of 20 μL. To 1 μL of compounds dissolved in 100% DMSO
and spotted at the bottom of each well, 5 μL of 0.04% BSA was added
and the mixture was kept at room temperature for 20 min. To this
mixture, 10 μL of the detergent-solubilized DGAT2 membrane
fraction (0.01 mg/mL) diluted in 100 mM Hepes−NaOH, pH 7.4, 20
Determination of IC50 Values for TG Synthesis in Fresh
Human Hepatocytes. Fresh human hepatocytes (Triangle Research
Laboratories, Research Triangle Park, NC, Celsis IVT, Baltimore, MD,
or Life Technologies, Grand Island, NY) (50000 cells per well) were
seeded into collagen I coated 96-well plates (BD Biosciences, Woburn,
MA) and maintained overnight in Williams’ E media prior to
determination of DGAT2 activity. Media was aspirated and the cells
incubated with 400 μM sodium dodecanoate (Sigma-Aldrich, St.
Louis, MO) in Williams’ E media for 40 min prior to the addition of a
selective DGAT1 inhibitor {trans-4-[4-(4-amino-5-oxo-7,8-
dihydropyrimido[5,4-f ][1,4]oxazepin-6(5H)-yl)phenyl]cyclohexyl}-
acetic acid (PF-04620110, 3 μM) and eight concentrations of
compound 9 ranging from 0.0003 to 1 μM. Both compounds were
dissolved in dimethyl sulfoxide (DMSO), and the final concentration
of DMSO in each well was kept constant at 0.25% volume/volume (v/
mM MgCl containing 200 nM MAFP (dried from ethyl acetate stock
2
solution under argon gas and dissolved in DMSO as 5 mM stock) was
added. After this mixture was preincubated at room temperature for
1
20 min, DGAT2 reactions were initiated by the addition of 4 μL of
14
substrates containing 30 μM [1- C]decanoyl-CoA and 125 μM 1,2-
didecanoyl-sn-glycerol dissolved in 12.5% acetone. The reaction
mixtures were incubated at room temperature for 40 min, and the
reactions were stopped by addition of 5 μL of 1% H PO . After the
14
v). After a further 20 min incubation, 5 μL of [ C]glycerol (American
Radio Chemicals, St. Louis, MO) (0.2 μCi) was added to each well
and mixed by gentle pipetting. The plate was returned to the incubator
3
4
addition of 45 μL of MicroScint-E (PerkinElmer), plates were sealed
with Top Seal-A covers (PerkinElmer) and phase partitioning of
substrates and products was achieved using a HT-91100 microplate
orbital shaker (Big Bear Automation, Santa Clara, CA). Plates were
centrifuged at 2000g for 1 min and then were sealed again with fresh
covers before reading in a 1450 Microbeta Wallac Trilux scintillation
counter (PerkinElmer). DGAT2 activity was measured by quantifying
(37 °C, 5% carbon dioxide) for 3.5 h prior to aspiration of the media
and addition of 100 μL of isopropyl alcohol:THF (9:1). The plate was
shaken for 15 min, centrifuged at 3000 rpm for 5 min, and 50 μL of
supernatant from each well applied to TLC lane (Whatman LK6D
Silica Gel Plates). Radiolabeled lipids were resolved using a two-
solvent system. Solvent 1 contained a 100:100:100:40:36 mixture of
14
ethyl acetate:isopropyl alcohol:CHCl
3
:MeOH:0.25% KCl and solvent
the generated product [ C]tridecanoylglycerol in the upper organic
phase. Background activity obtained using 50 μM of (1R, 2R)-2-({3′-
fluoro-4′-[(6-fluoro-1, 3-benzothiazol-2-yl)amino]-1,1′-biphenyl-4-yl}-
carbonyl)cyclopentanecarboxylic acid (US 20040224997, example 26)
for complete inhibition of DGAT2 was subtracted from all reactions.
Inhibitors were tested at 11 different concentrations to generate IC50
values for each compound. The 11 inhibitor concentrations employed
typically included 50, 15.8, 5, 1.58, 0.50, 0.16, 0.05, 0.016, 0.005,
2
a 70:27:3 hexane:diethyl ether:acetic acid mix. The TLC plate was
14
dried under nitrogen for 30 min and [ C]-calibrators added to a
vacant lane. Bands were visualized and quantitated using a Molecular
Dynamics’ Storm 860 PhosphorImager system following 18−36 h
exposure to a PhosphorImager screen. The IC50 was determined using
GraphPad Prism (GraphPad Software, Inc., La Jolla, CA).
Acute Effects of Compound 9 on Plasma Triacylglycerol
Levels. Male Sprague−Dawley rats (∼200 g, Harlan Laboratories,
Inc., Madison, WA) were fed a high-sucrose, low-fat diet (TD.03045,
Harlan Laboratories, Inc.) 2 days prior to the study. On the day of the
study, animals were fasted for 4 h prior to treatment. Immediately
before dosing, blood was drawn from the lateral tail vein into a
dipotassium ethylenediaminetetraacetic acid (K2 EDTA)-containing
tube (Microtainer, Becton Dickinson, Franklin Lakes, NJ). Following
centrifugation, the plasma was transferred to a fresh tube for the
determination of TG levels using a Roche Hitachi Chemistry analyzer
0
.0016, and 0.0005 μM. The data were plotted as percentage of
inhibition versus inhibitor concentration and fit to the equation, y =
z
1
00/[1 + (x/IC ) ], where IC is the inhibitor concentration at 50%
50 50
inhibition and z is the Hill slope (the slope of the curve at its inflection
point).
Determination of IC50 Values for DGAT1 and MGAT1/2/3
Inhibition. Enzyme activity for DGAT1 and MGAT3 was determined
14
by measuring the incorporation of the [1- C]decanoyl moiety into
14
TG using [1- C]decanoyl-CoA and 1,2-didecanoyl-sn-glycerol as
(Roche Diagnostics Corporation, Indianapolis, IN). Animals were
substrates. Enzyme activity for MGAT1 was determined by measuring
dosed with 0.01, 0.03, 0.1, 0.3, 1, 3, 10 mg/kg po compound 9 as a
solution in 0.5% w/v methylcellulose (10 mL/kg dosing volume).
Vehicle-treated animals received 0.5% methylcellulose alone. After a
further 2 h, blood was drawn and TG determined as described above.
Data were expressed as percent change from vehicle-treated animals.
The level of 9 was determined in these plasma samples using LC-MS/
MS.
14
the incorporation of the [1- C]decanoyl moiety into DAG using
14
[
1- C]decanoyl-CoA and 2-oleoylglycerol as substrates. MGAT2
activity was determined by measuring the incorporation of the
14
14
[
1- C]decanoyl moiety into DAG using [1- C]decanoyl-CoA and 1-
decanoyl-rac-glycerol as substrates. All assays were performed in 384-
well Polyplates in a total volume of 20 μL. The final assay mixture
contained 50 mM Hepes−NaOH, pH 7.4, 10 mM MgCl , the
2
Effects of Compound 9 on Plasma and Hepatic Lipids in
−
/−
indicated concentrations of substrates (below), 100 nM MAFP, 0.01%
Triton X-100, 5% DMSO, 2.5% acetone, and the indicated amount of
each enzyme. The final substrate concentrations in the assay were 7.5
LDLr Knockout Mice (Ldlr ). Male low-density lipoprotein
tm1Her
receptor (LDLR) knockout mice (B6.129S7-Ldlr
/J) were
obtained from The Jackson Laboratory (Bar Harbor, Maine) at 7−8
weeks of age and maintained on either standard laboratory chow
(PicoLab Rodent Diet 20 no. 5053, LabDiet, St. Louis, MO) or a high-
fat, high-cholesterol diet (HFHC, D12108, Research Diets Inc., New
Brunswick, NJ) which contains ∼40 kcal% from fat and 1.25% w/w
cholesterol for 2 weeks prior to treatment. Animals were dosed with
either 60 mg/kg/day (30 mg/kg BID) of 9 or vehicle (0.5% w/v
methylcellulose) for 3 days (total of five doses). On the final day, food
was withdrawn at 06:00, and the animals were dosed at 10:00 and
sacrificed 2 h later. All animals were sacrificed by carbon dioxide
asphyxiation, and blood for lipoprotein analysis was collected by
cardiac puncture. Livers were excised, immediately frozen in liquid
nitrogen, and held at −80 °C until analysis. Plasma TG and cholesterol
were determined as described above. Plasma lipoproteins were
14
μM [1- C]decanoyl-CoA and 10 μM 1,2-didecanoyl-sn-glycerol for
14
DGAT1, 2 μM [1- C]decanoyl-CoA and 20 μM 2-oleoylglycerol
1
4
(
1
Sigma-Aldrich) for MGAT1, 4 μM [1- C]decanoyl-CoA and 25 μM
-decanoyl-rac-glycerol (Nu-Check Prep Inc., Elysian, MN) for
14
MGAT2, 2 μM [1- C]decanoyl-CoA and 35 μM 1,2-didecanoyl-sn-
glycerol for MGAT3. All compounds were preincubated with each
enzyme for 30 min before initiating reactions by addition of substrates.
Typically, 1 μg of DGAT1 microsomal fraction was required for each
reaction per well. MGAT1, MGAT2, and MGAT3 reactions required
0
.12, 0.006, and 0.05 μg of the detergent-solubilized membrane
fractions, respectively, for each reaction per well. Typical reaction
times were 30−70 min at room temperature. The radioactive DAG
and TG products generated were quantified by scintillation counting
following reaction quenching and phase partitioning as described in
the DGAT2 assay above. Background activity, typical dose response
3
3
separated by FPLC essentially as described elsewhere. Briefly, 200
μL aliquots of plasma pooled from eight mice were injected inline to
J
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
two Superose 6 columns (GE Healthcare Biosciences, Pittsburgh, PA)
connected in tandem at a flow rate of 0.2 mL/min and fractions were
collected. Cholesterol was determined by standard enzymatic assay.
Lipid values, nmol/fraction, were corrected for recoveries of total
plasma cholesterol from combined fractions. Total lipids were
REFERENCES
■
(
1) Raised cholesterol: Situation and Trends. Global Health
Observatory (GHO) Data; World Health Organization: Geneva,
015; http://www.who.int/gho/ncd/risk_factors/cholesterol_text/
en/.
2
34
extracted from the livers using the Folch method. All extraction
solvents contained 50 μM butylated hydroxytoluene as antioxidant.
Neutral lipids were isolated from the total lipid extracts using
(
2) Baigent, C.; Keech, A.; Kearney, P. M.; Blackwell, L.; Buck, G.;
Pollicino, C.; Kirby, A.; Sourjina, T.; Peto, R.; Collins, R.; Simes, R.;
Cholesterol Treatment Trialists' (CTT) Collaborators. Efficacy and
safety of cholesterol-lowering treatment: prospective meta-analysis of
data from 90,056 participants in 14 randomised trials of statins. Lancet
005, 366 (9493), 1267−1278.
3) (a) Ridker, P. M.; Genest, J.; Boekholdt, S. M.; Libby, P.; Gotto,
A. M.; Nordestgaard, B. G.; Mora, S.; MacFadyen, J. G.; Glynn, R. J.;
Kastelein, J. J.; JUPITER Trial Study Group. HDL cholesterol and
residual risk of first cardiovascular events after treatment with potent
statin therapy: an analysis from the JUPITER trial. Lancet 2010, 376
3
5
aminopropyl solid phase extraction columns as previously described.
The neutral lipid fraction was in turn dried down under nitrogen and
resuspended in isooctane:isopropyl alcohol (98:2). Neutral lipids were
separated and detected using normal phase cyanopropyl high-
performance liquid chromatography (HPLC) and a charged aerosol
2
(
36
detector. Data are expressed as mg/g wet weight of liver.
ASSOCIATED CONTENT
Supporting Information
■
*
S
(
9738), 333−339. (b) Sampson, U. K.; Fazio, S.; Linton, M. F.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.5b01006.
Residual cardiovascular risk despite optimal LDL cholesterol reduction
with statins: the evidence, etiology, and therapeutic challenges. Curr.
Atheroscler. Rep. 2012, 14 (1), 1−10.
(
4) (a) Do, R.; Willer, C. J.; Schmidt, E. M.; Sengupta, S.; Gao, C.;
Procedures for the synthesis of compounds 1−4 and 6−
Peloso, G. M.; Gustafsson, S.; Kanoni, S.; Ganna, A.; Chen, J.;
Buchkovich, M. L.; Mora, S.; Beckmann, J. S.; Bragg-Gresham, J. L.;
Chang, H. Y.; Demirkan, A.; Den Hertog, H. M.; Donnelly, L. A.;
Ehret, G. B.; Esko, T.; Feitosa, M. F.; Ferreira, T.; Fischer, K.;
Fontanillas, P.; Fraser, R. M.; Freitag, D. F.; Gurdasani, D.; Heikkila,
K.; Hypponen, E.; Isaacs, A.; Jackson, A. U.; Johansson, A.; Johnson,
T.; Kaakinen, M.; Kettunen, J.; Kleber, M. E.; Li, X.; Luan, J.;
Lyytikainen, L. P.; Magnusson, P. K.; Mangino, M.; Mihailov, E.;
Montasser, M. E.; Muller-Nurasyid, M.; Nolte, I. M.; O'Connell, J. R.;
Palmer, C. D.; Perola, M.; Petersen, A. K.; Sanna, S.; Saxena, R.;
Service, S. K.; Shah, S.; Shungin, D.; Sidore, C.; Song, C.; Strawbridge,
R. J.; Surakka, I.; Tanaka, T.; Teslovich, T. M.; Thorleifsson, G.; Van
den Herik, E. G.; Voight, B. F.; Volcik, K. A.; Waite, L. L.; Wong, A.;
Wu, Y.; Zhang, W.; Absher, D.; Asiki, G.; Barroso, I.; Been, L. F.;
Bolton, J. L.; Bonnycastle, L. L.; Brambilla, P.; Burnett, M. S.; Cesana,
G.; Dimitriou, M.; Doney, A. S.; Doring, A.; Elliott, P.; Epstein, S. E.;
Eyjolfsson, G. I.; Gigante, B.; Goodarzi, M. O.; Grallert, H.; Gravito,
M. L.; Groves, C. J.; Hallmans, G.; Hartikainen, A. L.; Hayward, C.;
Hernandez, D.; Hicks, A. A.; Holm, H.; Hung, Y. J.; Illig, T.; Jones, M.
R.; Kaleebu, P.; Kastelein, J. J.; Khaw, K. T.; Kim, E.; Klopp, N.;
Komulainen, P.; Kumari, M.; Langenberg, C.; Lehtimaki, T.; Lin, S. Y.;
Lindstrom, J.; Loos, R. J.; Mach, F.; McArdle, W. L.; Meisinger, C.;
Mitchell, B. D.; Muller, G.; Nagaraja, R.; Narisu, N.; Nieminen, T. V.;
Nsubuga, R. N.; Olafsson, I.; Ong, K. K.; Palotie, A.; Papamarkou, T.;
Pomilla, C.; Pouta, A.; Rader, D. J.; Reilly, M. P.; Ridker, P. M.;
Rivadeneira, F.; Rudan, I.; Ruokonen, A.; Samani, N.; Scharnagl, H.;
Seeley, J.; Silander, K.; Stancakova, A.; Stirrups, K.; Swift, A. J.; Tiret,
L.; Uitterlinden, A. G.; van Pelt, L. J.; Vedantam, S.; Wainwright, N.;
Wijmenga, C.; Wild, S. H.; Willemsen, G.; Wilsgaard, T.; Wilson, J. F.;
Young, E. H.; Zhao, J. H.; Adair, L. S.; Arveiler, D.; Assimes, T. L.;
Bandinelli, S.; Bennett, F.; Bochud, M.; Boehm, B. O.; Boomsma, D. I.;
Borecki, I. B.; Bornstein, S. R.; Bovet, P.; Burnier, M.; Campbell, H.;
Chakravarti, A.; Chambers, J. C.; Chen, Y. D.; Collins, F. S.; Cooper,
R. S.; Danesh, J.; Dedoussis, G.; de Faire, U.; Feranil, A. B.; Ferrieres,
J.; Ferrucci, L.; Freimer, N. B.; Gieger, C.; Groop, L. C.; Gudnason, V.;
Gyllensten, U.; Hamsten, A.; Harris, T. B.; Hingorani, A.; Hirschhorn,
J. N.; Hofman, A.; Hovingh, G. K.; Hsiung, C. A.; Humphries, S. E.;
Hunt, S. C.; Hveem, K.; Iribarren, C.; Jarvelin, M. R.; Jula, A.;
Kahonen, M.; Kaprio, J.; Kesaniemi, A.; Kivimaki, M.; Kooner, J. S.;
Koudstaal, P. J.; Krauss, R. M.; Kuh, D.; Kuusisto, J.; Kyvik, K. O.;
Laakso, M.; Lakka, T. A.; Lind, L.; Lindgren, C. M.; Martin, N. G.;
Marz, W.; McCarthy, M. I.; McKenzie, C. A.; Meneton, P.; Metspalu,
A.; Moilanen, L.; Morris, A. D.; Munroe, P. B.; Njolstad, I.; Pedersen,
N. L.; Power, C.; Pramstaller, P. P.; Price, J. F.; Psaty, B. M.;
Quertermous, T.; Rauramaa, R.; Saleheen, D.; Salomaa, V.; Sanghera,
D. K.; Saramies, J.; Schwarz, P. E.; Sheu, W. H.; Shuldiner, A. R.;
Siegbahn, A.; Spector, T. D.; Stefansson, K.; Strachan, D. P.; Tayo, B.
O.; Tremoli, E.; Tuomilehto, J.; Uusitupa, M.; van Duijn, C. M.;
8; generation of constructs for DGAT1/2 and MGAT1/
2
/3; expression and preparation of membrane and
microsomal fractions; selectivity data of compound 5
PDF)
(
Molecular formula strings (CSV)
AUTHOR INFORMATION
Corresponding Authors
For K.F.: E-mail, Kentaro.Futatsugi@pfizer.com.
For D.W.K.: E-mail, Daniel.W.Kung@pfizer.com.
■
*
*
Notes
The authors declare the following competing financial
interest(s): All authors were employed by Pfizer Inc at the
time this work was done.
ACKNOWLEDGMENTS
■
We thank David Bernhardson, Peter Dorff, and Matthew F.
Sammons for their contributions to synthesis, Jared Milbank
and Jian Wang for their contributions to data analytics and
computational chemistry, Steven Hawrylik, Tracy B. Phillips,
Tim Subashi, and George Tkalcevic for expression of DGAT1/
, MGAT1/2/3 enzymes, and Alfin Vaz for metabolite
identification works.
2
ABBREVIATIONS USED
■
DGAT2, diacylglycerol acyltransferase 2; LDL, low-density
lipoprotein; LDLr, low-density lipoprotein receptor; MGAT1,
monoacylglycerol acyltransferase 1; MGAT2, monoacylglycerol
acyltransferase 2; MGAT3, monoacylglycerol acyltransferase 3;
DGAT1, diacylglycerol acyltransferase 1; TG, triacylglycerol;
MAG, monoacylglycerol; DAG, diacylglycerol; TC, total
cholesterol; NASH, nonalcoholic steatohepatitis; NAFLD,
nonalcoholic fatty liver disease; MAFP, methyl arachidonyl
fluorophosphonate; LipE, lipophilic efficiency; HLM, human
liver microsome; CYP, cytochrome P450; Papp, apparent
passive permeability; SAR, structure−activity relationship; MW,
molecular weight; RRCK, Ralph Russ canine kidney cells;
UGT, UDP-glucuronosyltransferase; 1H-benzotriazol-1-
yloxytris(dimethylamino)phosphonium hexafluorophosphate;
T P, 1-propanephosphonic anhydride; NMM, N-methyl
morpholine; MsOH, methanesulfonic acid; fu,p, fraction
unbound in plasma
3
K
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Vollenweider, P.; Wallentin, L.; Wareham, N. J.; Whitfield, J. B.;
Wolffenbuttel, B. H.; Altshuler, D.; Ordovas, J. M.; Boerwinkle, E.;
Palmer, C. N.; Thorsteinsdottir, U.; Chasman, D. I.; Rotter, J. I.;
Franks, P. W.; Ripatti, S.; Cupples, L. A.; Sandhu, M. S.; Rich, S. S.;
Boehnke, M.; Deloukas, P.; Mohlke, K. L.; Ingelsson, E.; Abecasis, G.
R.; Daly, M. J.; Neale, B. M.; Kathiresan, S. Common variants
associated with plasma triglycerides and risk for coronary artery
disease. Nat. Genet. 2013, 45 (11), 1345−1352. (b) Goldberg, I. J.;
Eckel, R. H.; McPherson, R. Triglycerides and heart disease: still a
hypothesis? Arterioscler., Thromb., Vasc. Biol. 2011, 31 (8), 1716−1725.
(10) (a) Denison, H.; Nilsson, C.; Lofgren, L.; Himmelmann, A.;
Martensson, G.; Knutsson, M.; Al-Shurbaji, A.; Tornqvist, H.;
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 (4), 334−343. (b) Maciejewski, B. S.; LaPerle, J. L.;
Chen, D.; Ghosh, A.; Zavadoski, W. J.; McDonald, T. S.; Manion, T.
B.; Mather, D.; Patterson, T. A.; Hanna, M.; Watkins, S.; Gibbs, E. M.;
Calle, R. A.; Steppan, C. M. Pharmacological inhibition to examine the
role of DGAT1 in dietary lipid absorption in rodents and humans. Am.
J. Physiol Gastrointest Liver Physiol 2013, 304 (11), G958−G969.
(
c) ACCORD Study Group; Ginsberg, H. N.; Elam, M. B.; Lovato, L.
(11) A 12-week multi-center, randomized, double-blind, placebo-
C.; Crouse, J. R., III; Leiter, L. A.; Linz, P.; Friedewald, W. T.; Buse, J.
B.; Gerstein, H. C.; Probstfield, J.; Grimm, R. H.; Ismail-Beigi, F.;
Bigger, J. T.; Goff, D. C., Jr.; Cushman, W. C.; Simons-Morton, D. G.;
Byington, R. P. Effects of combination lipid therapy in type 2 diabetes
mellitus. N. Engl. J. Med. 2010, 362 (17), 1563−1574. (d) Jorgensen,
A. B.; Frikke-Schmidt, R.; Nordestgaard, B. G.; Tybjaerg-Hansen, A.
Loss-of-function mutations in APOC3 and risk of ischemic vascular
disease. N. Engl. J. Med. 2014, 371 (1), 32−41. (e) The TG and HDL
Working Group of the Exome Sequencing Project, National Heart,
Lung, and Blood Institute; Crosby, J.; Peloso, G. M.; Auer, P. L.;
Crosslin, D. R.; Stitziel, N. O.; Lange, L. A.; Lu, Y.; Tang, Z. Z.; Zhang,
H.; Hindy, G.; Masca, N.; Stirrups, K.; Kanoni, S.; Do, R.; Jun, G.; Hu,
Y.; Kang, H. M.; Xue, C.; Goel, A.; Farrall, M.; Duga, S.; Merlini, P. A.;
Asselta, R.; Girelli, D.; Olivieri, O.; Martinelli, N.; Yin, W.; Reilly, D.;
Speliotes, E.; Fox, C. S.; Hveem, K.; Holmen, O. L.; Nikpay, M.;
Farlow, D. N.; Assimes, T. L.; Franceschini, N.; Robinson, J.; North, K.
E.; Martin, L. W.; DePristo, M.; Gupta, N.; Escher, S. A.; Jansson, J.
H.; Van Zuydam, N.; Palmer, C. N.; Wareham, N.; Koch, W.;
Meitinger, T.; Peters, A.; Lieb, W.; Erbel, R.; Konig, I. R.; Kruppa, J.;
Degenhardt, F.; Gottesman, O.; Bottinger, E. P.; O’Donnell, C. J.;
Psaty, B. M.; Ballantyne, C. M.; Abecasis, G.; Ordovas, J. M.;
Melander, O.; Watkins, H.; Orho-Melander, M.; Ardissino, D.; Loos,
R. J.; McPherson, R.; Willer, C. J.; Erdmann, J.; Hall, A. S.; Samani, N.
J.; Deloukas, P.; Schunkert, H.; Wilson, J. G.; Kooperberg, C.; Rich, S.
S.; Tracy, R. P.; Lin, D. Y.; Altshuler, D.; Gabriel, S.; Nickerson, D. A.;
Jarvik, G. P.; Cupples, L. A.; Reiner, A. P.; Boerwinkle, E.; Kathiresan,
S. Loss-of-function mutations in APOC3, triglycerides, and coronary
disease. N. Engl. J. Med. 2014, 371 (1), 22−31.
controlled, parallel-group adaptive design study to evaluate the efficacy
on blood glucose control and safety of five doses of LCQ908 (2, 5, 10,
1
5 and 20 mg) or sitagliptin 100 mg on a background therapy of
metformin in obese patients with type 2 diabetes. Clinical Trial Results
Database; Novartis, 2010; http://www.novctrd.com/ctrdWebApp/
clinicaltrialrepository/displayFile.do?trialResult=4313.
(12) 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,
2
(
79 (12), 11767−11776.
13) (a) Choi, C. S.; Savage, D. B.; Kulkarni, A.; Yu, X. X.; Liu, Z. X.;
Morino, K.; Kim, S.; Distefano, A.; Samuel, V. T.; Neschen, S.; Zhang,
D.; Wang, A.; Zhang, X. M.; Kahn, M.; Cline, G. W.; Pandey, S. K.;
Geisler, J. G.; Bhanot, S.; Monia, B. P.; Shulman, G. I. Suppression of
diacylglycerol acyltransferase-2 (DGAT2), but not DGAT1, with
antisense oligonucleotides reverses diet-induced hepatic steatosis and
insulin resistance. J. Biol. Chem. 2007, 282 (31), 22678−22688.
(b) Liu, Y.; Millar, J. S.; Cromley, D. A.; Graham, M.; Crooke, R.;
Billheimer, J. T.; Rader, D. J. Knockdown of acyl-CoA:diacylglycerol
acyltransferase 2 with antisense oligonucleotide reduces VLDL TG
and ApoB secretion in mice. Biochim. Biophys. Acta, Mol. Cell Biol.
Lipids 2008, 1781 (3), 97−104. (c) Yu, X. X.; Murray, S. F.; Pandey, S.
K.; Booten, S. L.; Bao, D.; Song, X. Z.; Kelly, S.; Chen, S.; McKay, R.;
Monia, B. P.; Bhanot, S. Antisense oligonucleotide reduction of
DGAT2 expression improves hepatic steatosis and hyperlipidemia in
obese mice. Hepatology 2005, 42 (2), 362−371.
(14) (a) Kim, M. O.; Lee, S.; Choi, K.; Lee, S.; Kim, H.; Kang, H.;
Choi, M.; Kwon, E. B.; Kang, M. J.; Kim, S.; Lee, H.-J.; Lee, H. S.;
Kwak, Y.-S.; Cho, S. Discovery of a Novel Class of Diacylglycerol
Acyltransferase 2 Inhibitors with a 1H-Pyrrolo[2,3-b]Pyridine Core.
Biol. Pharm. Bull. 2014, 37 (10), 1655−1660. (b) Lee, K.; Kim, M.;
Lee, B.; Goo, J.; Kim, J.; Naik, R.; Seo, J. H.; Kim, M. O.; Byun, Y.;
Song, G. Y.; Lee, H. S.; Choi, Y. Discovery of indolyl acrylamide
derivatives as human diacylglycerol acyltransferase-2 selective inhib-
itors. Org. Biomol. Chem. 2013, 11 (5), 849−858. (c) Kim, M. O.; Lee,
S. U.; Lee, H. J.; Choi, K.; Kim, H.; Lee, S.; Oh, S. J.; Kim, S.; Kang, J.
S.; Lee, H. S.; Kwak, Y. S.; Cho, S. Identification and validation of a
selective small molecule inhibitor targeting the diacylglycerol
acyltransferase 2 activity. Biol. Pharm. Bull. 2013, 36 (7), 1167−
(
5) (a) Bell, D. A.; Hooper, A. J.; Burnett, J. R. Mipomersen, an
antisense apolipoprotein B synthesis inhibitor. Expert Opin. Invest.
Drugs 2011, 20 (2), 265−272. (b) Cuchel, M.; Rader, D. J.
Microsomal transfer protein inhibition in humans. Curr. Opin. Lipidol.
2
(
013, 24 (3), 246−250.
6) Yen, C. L.; Stone, S. J.; Koliwad, S.; Harris, C.; Farese, R. V., Jr.
Thematic review series: glycerolipids. DGAT enzymes and triacylgly-
cerol biosynthesis. J. Lipid Res. 2008, 49 (11), 2283−2301.
(
7) (a) Zammit, V. A. Hepatic triacylglycerol synthesis and secretion:
DGAT2 as the link between glycaemia and triglyceridaemia. Biochem. J.
2
013, 451 (1), 1−12. (b) Turchetto-Zolet, A. C.; Maraschin, F. S.; de
Morais, G. L.; Cagliari, A.; Andrade, C. M.; Margis-Pinheiro, M.;
Margis, R. Evolutionary view of acyl-CoA diacylglycerol acyltransferase
1
173. (d) Wurie, H. R.; Buckett, L.; Zammit, V. A. Diacylglycerol
acyltransferase 2 acts upstream of diacylglycerol acyltransferase 1 and
utilizes nascent diglycerides and de novo synthesized fatty acids in
HepG2 cells. FEBS J. 2012, 279 (17), 3033−3047. (e) Qi, J.; Lang, W.;
Geisler, J. G.; Wang, P.; Petrounia, I.; Mai, S.; Smith, C.; Askari, H.;
Struble, G. T.; Williams, R.; Bhanot, S.; Monia, B. P.; Bayoumy, S.;
Grant, E.; Caldwell, G. W.; Todd, M. J.; Liang, Y.; Gaul, M. D.;
Demarest, K. T.; Connelly, M. A. The use of stable isotope-labeled
glycerol and oleic acid to differentiate the hepatic functions of DGAT1
and −2. J. Lipid Res. 2012, 53 (6), 1106−1116.
(15) During the preparation of this manuscript, a patent application
was published by Eli Lilly on pyrimidine-based DGAT2 inhibitors
showing in vivo pharmacology in rodent models . Camp, N. P.; Naik,
M. Novel DGAT2 inhibitors. PCT Int. Appl. WO2015077299, 2015.
(16) Yang, D.; Fokas, D.; Li, J.; Yu, L.; Baldino, C. M. A Versatile
Method for the Synthesis of Benzimidazoles from o-Nitroanilines and
Aldehydes in One Step via a Reductive Cyclization. Synthesis 2005, 1,
47−56.
(
2
DGAT), a key enzyme in neutral lipid biosynthesis. BMC Evol. Biol.
011, 11, 263.
8) (a) 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 (42), 38870−38876. (b) Yen, C. L. E.;
Monetti, M.; Burri, B. J.; Farese, R. V., Jr. The triacylglycerol synthesis
enzyme DGAT1 also catalyzes the synthesis of diacylglycerols, waxes,
and retinyl esters. J. Lipid Res. 2005, 46 (7), 1502−1511.
(
9) (a) Devita, R. J.; Pinto, S. Current Status of the Research and
Development of Diacylglycerol O-Acyltransferase 1 (DGAT1)
Inhibitors. J. Med. Chem. 2013, 56 (24), 9820−9825. (b) Naik, R.;
Obiang-Obounou, B. W.; Kim, M.; Choi, Y.; Lee, H. S.; Lee, K.
Therapeutic Strategies for Metabolic Diseases: Small-Molecule
Diacylglycerol Acyltransferase (DGAT) Inhibitors. ChemMedChem
2
014, 9 (11), 2410−2424.
L
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(
17) (a) Ivanova, N. V.; Sviridov, S. I.; Stepanov, A. E. Parallel
1A9 substrates but not UGT1A1 and UGT1A6 activities. Drug Metab.
Dispos. 2008, 36 (6), 1056−1062.
solution-phase synthesis of substituted 2-(1,2,4-triazol-3-yl)-
benzimidazoles. Tetrahedron Lett. 2006, 47 (46), 8025−8027.
b) Lin, S.; Yang, L. A simple and efficient procedure for the synthesis
(25) Kaivosaari, S.; Finel, M.; Koskinen, M. N-glucuronidation of
drugs and other xenobiotics by human and animal UDP-glucuronosyl-
transferases. Xenobiotica 2011, 41 (8), 652−669.
(
of benzimidazoles using air as the oxidant. Tetrahedron Lett. 2005, 46
25), 4315−4319.
18) Lin, S.-Y.; Isome, Y.; Stewart, E.; Liu, J.-F.; Yohannes, D.; Yu, L.
(
(26) Lovering, F. Escape from Flatland 2: complexity and
promiscuity. MedChemComm 2013, 4 (3), 515−519.
(
(27) Rose, K.; Yang, Y.-S.; Sciotti, R.; Cai, H. Structure-activity
Microwave-assisted one step high-throughput synthesis of benzimida-
relationship (SAR): effort towards blocking N-glucuronidation of
indazoles (PF-03376056) by human UGT1A enzymes. Drug Metab.
Lett. 2009, 3 (1), 28−34.
zoles. Tetrahedron Lett. 2006, 47 (17), 2883−2886.
(
19) Guillaume, D.; Brum-Bousquet, M.; Aitken, D. J.; Husson, H.-P.
Cyclopropanation of dibenylaminoacetonitrile: evaluation of 1,2-
dibromides and cyclic 1,2-sulfates as dielectrophiles. Bull. Soc. Chim.
Fr. 1994, 131, 391−396.
(28) A subset of analogues of compounds 8 and 9 that incorporate
other substituents at 4 position of the pyrazole ring, such as cyano,
methyl, cyclopropyl, or trifluoromethyl, showed no improvement of
potency and LipE over compound 8 and 9 among tested, while
showing measurable metabolic turnover in HLM.
(
20) (a) McFie, P. J.; Stone, S. J. A fluorescent assay to quantitatively
measure in vitro acyl CoA:diacylglycerol acyltransferase activity. J.
Lipid Res. 2011, 52 (9), 1760−1764. (b) Harris, C. A.; Haas, J. T.;
Streeper, R. S.; Stone, S. J.; Kumari, M.; Yang, K.; Han, X.; Brownell,
N.; Gross, R. W.; Zechner, R.; Farese, R. V., Jr. DGAT enzymes are
required for triacylglycerol synthesis and lipid droplets in adipocytes. J.
Lipid Res. 2011, 52 (4), 657−667. (c) Cao, J.; Cheng, L.; Shi, Y.
Catalytic properties of MGAT3, a putative triacylgycerol synthase. J.
Lipid Res. 2007, 48 (3), 583−591. (d) Yen, C. L.; Farese, R. V., Jr.
MGAT2, a monoacylglycerol acyltransferase expressed in the small
intestine. J. Biol. Chem. 2003, 278 (20), 18532−18537. (e) Yen, C. L.;
Stone, S. J.; Cases, S.; Zhou, P.; Farese, R. V., Jr. Identification of a
gene encoding MGAT1, a monoacylglycerol acyltransferase. Proc. Natl.
Acad. Sci. U. S. A. 2002, 99 (13), 8512−8517.
(29) 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-
0
4620110, a Potent, Selective, and Orally Bioavailable Inhibitor of
DGAT-1. ACS Med. Chem. Lett. 2011, 2 (5), 407−412.
30) Ishibashi, S.; Brown, M. S.; Goldstein, J. L.; Gerard, R. D.;
(
Hammer, R. E.; Herz, J. Hypercholesterolemia in low density
lipoprotein receptor knockout mice and its reversal by adenovirus-
mediated gene delivery. J. Clin. Invest. 1993, 92 (2), 883−893.
(31) Cuchel, M.; Bloedon, L. T.; Szapary, P. O.; Kolansky, D. M.;
Wolfe, M. L.; Sarkis, A.; Millar, J. S.; Ikewaki, K.; Siegelman, E. S.;
Gregg, R. E.; Rader, D. J. Inhibition of microsomal triglyceride transfer
protein in familial hypercholesterolemia. N. Engl. J. Med. 2007, 356
(
21) More details on the in vitro assay development and mechanistic
characterizations of DGAT2 inhibitors will be published in the future.
22) (a) Shultz, M. D. Setting expectations in molecular
(
(
2), 148−156.
32) PF-06424439 is commercially available from Sigma Aldrich
catalogue number PZ0233).
33) Gerdes, L. U.; Gerdes, C.; Klausen, I. C.; Faergeman, O.
optimizations: Strengths and limitations of commonly used composite
(
parameters. Bioorg. Med. Chem. Lett. 2013, 23 (21), 5980−5991.
(
(
b) Freeman-Cook, K. D.; Hoffman, R. L.; Johnson, T. W. Lipophilic
(
efficiency: the most important efficiency metric in medicinal chemistry.
Future Med. Chem. 2013, 5 (2), 113−115. (c) Tarcsay, A.; Nyiri, K.;
Keseru, G. M. Impact of Lipophilic Efficiency on Compound Quality.
J. Med. Chem. 2012, 55 (3), 1252−1260. (d) Leeson, P. D.; Empfield,
J. R. Reducing the risk of drug attrition associated with
physicochemical properties. Annu. Rep. Med. Chem. 2010, 45, 393−
Generation of analytic plasma lipoprotein profiles using two prepacked
superose 6B columns. Clin. Chim. Acta 1992, 205 (1−2), 1−9.
(34) Folch, J.; Lees, M.; Sloane Stanley, G. H. A simple method for
the isolation and purification of total lipides from animal tissues. J. Biol.
Chem. 1957, 226 (1), 497−509.
(35) Kaluzny, M. A.; Duncan, L. A.; Merritt, M. V.; Epps, D. E. Rapid
separation of lipid classes in high yield and purity using bonded phase
columns. J. Lipid Res. 1985, 26 (1), 135−140.
(36) El-Hamdy, A. H.; Christie, W. W. Separation of non-polar lipids
by high performance liquid chromatography on a cyanopropyl column.
J. High Resolut. Chromatogr. 1993, 16 (1), 55−57.
4
07. (e) Edwards, M. P.; Price, D. A. Role of Physicochemical
Properties and Ligand Lipophilicity Efficiency in Addressing Drug
Safety Risks. Annu. Rep. Med. Chem. 2010, 45, 380−391.
(
23) Lombardo, F.; Shalaeva, M. Y.; Tupper, K. A.; Gao, F.
ElogD(oct): a tool for lipophilicity determination in drug discovery. 2.
Basic and neutral compounds. J. Med. Chem. 2001, 44 (15), 2490−
2
497.
(
24) (a) Di, L.; Keefer, C.; Scott, D. O.; Strelevitz, T. J.; Chang, G.;
Bi, Y. A.; Lai, Y.; Duckworth, J.; Fenner, K.; Troutman, M. D.; Obach,
R. S. Mechanistic insights from comparing intrinsic clearance values
between human liver microsomes and hepatocytes to guide drug
design. Eur. J. Med. Chem. 2012, 57, 441−448. (b) Walsky, R. L.;
Bauman, J. N.; Bourcier, K.; Giddens, G.; Lapham, K.; Negahban, A.;
Ryder, T. F.; Obach, R. S.; Hyland, R.; Goosen, T. C. Optimized assays
for human UDP-glucuronosyltransferase (UGT) activities: altered
alamethicin concentration and utility to screen for UGT inhibitors.
Drug Metab. Dispos. 2012, 40 (5), 1051−1065. (c) Gill, K. L.;
Houston, J. B.; Galetin, A. Characterization of in vitro glucuronidation
clearance of a range of drugs in human kidney microsomes:
comparison with liver and intestinal glucuronidation and impact of
albumin. Drug Metab. Dispos. 2012, 40 (4), 825−835. (d) Kilford, P. J.;
Stringer, R.; Sohal, B.; Houston, J. B.; Galetin, A. Prediction of drug
clearance by glucuronidation from in vitro data: use of combined
cytochrome P450 and UDP-glucuronosyltransferase cofactors in
alamethicin-activated human liver microsomes. Drug Metab. Dispos.
2
009, 37 (1), 82−89. (e) Rowland, A.; Knights, K. M.; Mackenzie, P.
I.; Miners, J. O. The ″albumin effect″ and drug glucuronidation:
bovine serum albumin and fatty acid-free human serum albumin
enhance the glucuronidation of UDP-glucuronosyltransferase (UGT)
M
DOI: 10.1021/acs.jmedchem.5b01006
J. Med. Chem. XXXX, XXX, XXX−XXX