Discovery of 5-chloro-4-((1-(5-chloropyrimidin-2-yl)piperidin-4-yl)oxy)-1-(2-fluoro-4-(methylsulfonyl)phenyl)pyridin-2(1 H)-one (BMS-903452), an antidiabetic clinical candidate targeting GPR119

Article abstract of DOI:10.1021/jm501175v

G-protein-coupled receptor 119 (GPR119) is expressed predominantly in pancreatic β-cells and in enteroendocrine cells in the gastrointestinal tract. GPR119 agonists have been shown to stimulate glucose-dependent insulin release by direct action in the pancreas and to promote secretion of the incretin GLP-1 by action in the gastrointestinal tract. This dual mechanism of action has generated significant interest in the discovery of small molecule GPR119 agonists as a potential new treatment for type 2 diabetes. Herein, we describe the discovery and optimization of a new class of pyridone containing GPR119 agonists. The potent and selective BMS-903452 (42) was efficacious in both acute and chronic in vivo rodent models of diabetes. Dosing of 42 in a single ascending dose study in normal healthy humans showed a dose dependent increase in exposure and a trend toward increased total GLP-1 plasma levels.

Full text of DOI:10.1021/jm501175v

Drug Annotation
pubs.acs.org/jmc
Discovery of 5‑Chloro-4-((1-(5-chloropyrimidin-2-yl)piperidin-4-
yl)oxy)-1-(2-fluoro-4-(methylsulfonyl)phenyl)pyridin-2(1H)‑one (BMS-
903452), an Antidiabetic Clinical Candidate Targeting GPR119
Dean A. Wacker,* Ying Wang, Matthias Broekema, Karen Rossi, Steven O’Connor, Zhenqiu Hong,
Ginger Wu, Sarah E. Malmstrom, Chen-Pin Hung, Linda LaMarre, Anjaneya Chimalakonda, Lisa Zhang,
Li Xin, Hong Cai, Cuixia Chu, Stephanie Boehm, Jacob Zalaznick, Randolph Ponticiello, Larisa Sereda,
Song-Ping Han, Rachel Zebo, Bradley Zinker, Chiuwa Emily Luk, Richard Wong, Gerry Everlof, Yi-Xin Li,
Chunyu K. Wu, Michelle Lee, Steven Griffen, Keith J. Miller, John Krupinski, and Jeffrey A. Robl
Departments of Discovery Chemistry, Metabolic Diseases, Lead Evaluation, Computer-Assisted Drug Design, Discovery Toxicology,
Exploratory Clinical and Translational Research, and Pharmaceutical Candidate Optimization, Bristol-Myers Squibb, P.O. Box 5400,
Princeton, New Jersey 08543-5400, United States
S
* Supporting Information
ABSTRACT: G-protein-coupled receptor 119 (GPR119) is expressed predom-
inantly in pancreatic β-cells and in enteroendocrine cells in the gastrointestinal tract.
GPR119 agonists have been shown to stimulate glucose-dependent insulin release
by direct action in the pancreas and to promote secretion of the incretin GLP-1 by
action in the gastrointestinal tract. This dual mechanism of action has generated
significant interest in the discovery of small molecule GPR119 agonists as a potential
new treatment for type 2 diabetes. Herein, we describe the discovery and
optimization of a new class of pyridone containing GPR119 agonists. The potent
and selective BMS-903452 (42) was efficacious in both acute and chronic in vivo
rodent models of diabetes. Dosing of 42 in a single ascending dose study in normal healthy humans showed a dose dependent
increase in exposure and a trend toward increased total GLP-1 plasma levels.
glucose, thus lowering plasma levels. Meanwhile, nutrient
intake and the concomitant elevation in blood glucose cause
the incretin hormones to be released from the endocrine cells
in the gastrointestinal tract. These hormones further enhance
insulin secretion from the pancreas in a glucose dependent
manner. GLP-1 has the added benefit of inhibiting glucagon
release and decreasing gastric emptying to further reduce the
glucose excursion. Additionally, rodent studies with GLP-1 have
shown preservation of insulin releasing pancreatic islet β-
cells.5,6 Therefore, if a drug target could induce both insulin and
incretin hormone release, more effective control of T2DM
might be accomplished in a diabetic patient.
G-protein-coupled receptor 119 (GPR119) is a class A
GPCR with little homology to other receptors and is expressed
primarily in pancreatic β-cells and the K- and L-cells of the
gastrointestinal tract.7−10 In rodents, potent GPR119 agonists
cause the release of insulin from pancreatic islets and increase
insulin levels in vivo in a glucose dependent manner.9 Also, in
vivo studies in rodents have shown elevations in plasma levels
of GLP-1 and GIP upon oral treatment with glucose with
GPR119 agonists.10 These results indicate GPR119 agonists
have a dual mechanism for lowering plasma glucose levels and,
DISEASE TARGET
■
Diabetes is one of the fastest growing chronic diseases
worldwide, with an estimated prevalence of 382 million
patients, of which 85−95% have type 2 diabetes mellitus
(T2DM).1 T2DM is characterized by insufficient glycemic
control due to insulin resistance and insufficient insulin
secretion from the pancreas which leads to long-term
complications such as heart disease, organ failure, and lower
limb amputations.2 Even though multiple pharmacological
agents are available, less than one-third of T2DM patients
achieve the desired glycemic control.3 Furthermore, current
treatments have efficacy limitations during long-term treatment
and the potential for undesired side effects.4 Therefore, new
pharmacological treatments of T2DM with novel mechanisms
of action are needed.
MECHANISM OF ACTION
■
The human body has multiple mechanisms that work in
concert to properly control plasma glucose levels. Two of the
most important of these mechanisms are insulin release from
the pancreas and incretin hormone release (glucagon-like
peptide 1 (GLP-1) and glucose-dependent insulinotropic
polypeptide (GIP)) from the gastrointestinal tract. Elevation
of plasma glucose induces insulin release from the pancreas,
which in turn causes skeletal muscle and fat tissue to absorb
Received: July 31, 2014
Published: September 10, 2014
© 2014 American Chemical Society
7499
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
Figure 1. Pharmacophore overlays of GPR119 literature agonists and newly designed compounds.
therefore, have the potential to provide superior diabetes
control over the current therapies.
piperidine linker with a wide variety of nitrogen heterocycles
provided few compounds with improved potency (Table 1).
The incorporation of four- or five-membered rings (7−10)
provided compounds that were inactive, and the azepine (11)
was equivalent in potency to 6. The first potency improvement
came with the removal of the methylene between the oxygen
and the piperidine ring which provided a compound (5) with
COMPOUND DESIGN AND LEAD IDENTIFICATION
■
Our efforts to identify small molecule agonists of GPR119 were
initiated with a high throughput screen (HTS) of our
compound deck and pharmacophore modeling of GPR119
agonists disclosed by Prosidion Pharmaceuticals,11 exemplified
by 1, and Arena Pharmaceuticals,12 exemplified by 2. These
publications have inspired several companies to initiate
discovery programs that have led to compounds entering clinic
trials; Arena/Ortho-McNeil (APD668 and APD597),13 Metab-
olex/Sanofi-Aventis (MBX-2982),14,15 GlaxoSmithKline (GSK-
1292263),14,16 and Prosidion (PSN-821).14,17 Although our
HTS campaign identified some interesting structures, we were
intrigued by the pharmacophore model built by overlaying low
energy confirmations of both compounds (1 and 2). The
pharmacophore model was used to determine the spatial
orientation of left side aryl and right side piperidine
functionality. We then designed multiple core structures that
occupied the linker space and properly orientated the two
terminal functionalities. The most interesting of the designed
cores was the 1,3-substituted pyridone 3. As can be seen in
Figure 1, the aryl and piperidine overlay extremely well with the
corresponding functionality in 2 (overlay of peach and cyan
colored compounds) and the two carbonyls from 3 are located
near the nitrogens of the pyrimidine core. However, when we
went to the next step and analyzed the low energy
conformations of 3, we identified a flaw in our design. The
low energy conformation of 3 minimizes the overall dipole
moment of the compound and thus rotates the exocyclic amide
180° from desired conformation, as can been seen in structure
4. We therefore needed to redesign the molecule to a 1,4-
substitution of the pyridone ring, 5, to identify a low energy
confirmation that overlaid well with 2 (as can be seen in the
purple compound overlay with the cyan compound). Synthesis
of the methylene analog 6 provided our first active compound
in this series.
Table 1. In Vitro Activity of Linker Analogs
a
EC50 values were calculated from concentration−response curves. See
b
Our initial efforts focused on improving the in vitro potency
of the lead molecule 6. Replacement of the methylene
ref 18 for assay conditions. Reference compound set at 1.0 for each
assay run.
7500
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
∼13-fold greater potency. We tried to further improve potency
of the piperdine ring with conformational constraints by
bridging the ring system (13 and 14); however, in each case the
potency was reduced dramatically. Also, replacement of the
oxygen in the linker with other heteroatoms or carbon was
attempted and these modifications either dramatically reduced
or eliminated all activity of the molecules (data not shown).
Further potency optimization at the terminal groups
produced varying results. Replacement of the isopropyl
carbamate with other carbamates provided compounds (15,
16) of equivalent potency (Table 2). However, carbamate
Table 3. In Vitro Activity of Phenyl Analogs
Table 2. In Vitro Activity of Piperidine Analogs
a
EC50 values were calculated from concentration−response curves. See
b
ref 18 for assay conditions. Reference compound set at 1.0 for each
assay run.
at GPR119 (data not shown). The only changes that provided
compounds with reasonable potency either contained an amide
group (25, 26) or selected heterocyles (27, 28). The only
functional group that could replace the methyl sulfone and
retain potency was cyano (24). Although 4-cyano compounds
were potent, they typically exhibited lower aqueous solubility
than the corresponding sulfone compounds. As will be
discussed later in greater detail, solubility limited absorption
was an issue for the program, and therefore, we deemphasized
cyano compounds in the SAR program. Substitutions at the
ortho and meta positions of the aryl were tested in conjunction
with the methyl sulfone. Potency increases were not identified,
but small functionality was tolerated, such as fluoro (29, 30),
chloro, and methyl. Incorporation of large groups, such as
phenyl, resulted in a dramatic loss of all potency.
a
EC50 values were calculated from concentration−response curves. See
b
ref 18 for assay conditions. Reference compound set at 1.0 for each
With in vitro potency goals met, we wanted to identify issues
related to druglike properties for this series of compounds.
Compound 29 was chosen for evaluation in our in vitro safety
and metabolism assays in addition to pharmacokinetic
evaluation in mice. The molecule exhibited an excellent profile
in most of the safety assays: minimal inhibition (IC50 > 40 μM)
at nine different cytochrome P450 enzymes, no activation of
PXR (EC50 > 50 μM), and no cytotoxicity in a heptatic
(HEPG2) cell line at a concentration of 40 μM. Compound 29
was tested against a panel of 43 different GPCRs and enzymes
and showed minimal activity (<50% activity) at concentrations
below 20 μM. The compound weakly inhibits the hERG
channel in a patch clamp assay (47% at 10 μM). Compound 29
has excellent permeability in both the PAMPA (320 nm/s) and
Caco2 assays (220 nm/s). In a human liver microsome stability
assay, 29 has a half-life of 31 min. The compound is highly
protein bound in rodent and human plasma (99% bound) and
assay run.
replacement with a pyrimidine (17) to mimic the oxygens
increased potency by 2-fold. Incorporation of both heteroatoms
was necessary to achieve the largest potency improvement, as
shown by comparison to the pyridine (18). The substitution
pattern of the pyrimidine ring was vital for retention of in vitro
potency and efficacy, as can been seen with compounds 19 and
20. The optimal potency substitutions were either cyclopropyl
(21) or n-propyl (22) with increases of ∼3-fold for each
compound versus compound 17. Even though an almost 10-
fold increase in potency was achieved with piperidine
modifications, potency increases from changes in the aryl
substitution were more difficult to identify (Table 3).
Substitution at the para position was vital; any movement of
the group to the ortho or meta position ameliorated all activity
7501
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
has pH dependent solubility of 6 μg/mL at pH 6.5 and 630 μg/
mL at pH 1.0. With good overall in vitro characteristics, 29 was
dosed as an oral suspension to C57BL/6 mice at 30 mg/kg.
The compound achieved a maximum plasma concentration of
38.6 μM at 2 h after dose with an AUC of 245 μM·h and
terminal half-life of 2.4 h. The ability to achieve high in vivo
plasma concentrations of 29 allowed for the progression of the
compound into in vivo efficacy studies.
The effect of compound 29 on blood glucose levels was
evaluated in a 21-day study in 6-week-old db/db mice with an
oral glucose tolerance test (oGTT) on day 1 and readout on
day 21 of glycated hemoglobin (HbA1C). The compound was
orally dosed once daily (0.3, 3, 30 mg/kg) to the mice 1 h
before the glucose challenge. A dose response was observed for
glucose lowering, and statistically significant reductions were
apparent at 3 and 30 mg/kg (see Figure 2). The total glucose
increasing 1.4% in the animals treated with metformin and the
30 mg/kg dose of 29. Although we were excited to see HbA1C
lowering of 29 at 30 mg/kg in a chronic study that was
equivalent to metformin, we were concerned about the loss of
potency seen in the chronic readout (day 21 HbA1C) as
compared to the acute portion (day 1 oGTT) of the study.
These results were attributed to the short in vivo half-life of 29
in mice. Low doses (<30 mg/kg) could not provide 24 h of
coverage of the EC50 to maintain efficacy, and therefore, greater
metabolic stability was required to achieve a pharmacokinetic
profile suitable for once daily dosing.
COMPOUND OPTIMIZATION AND CLINICAL
CANDIDATE SELECTION
■
Optimization for clinical candidate selection began with
metabolite identification experiments to locate the site of
oxidation in our lead series. On the basis of experiments with
29 and several other analogs, the metabolic soft spot was
identified as the propyl side chain on the pyrimidine. During
the optimization of the pyrimidine ring substitution, we had
identified several groups that provided much better metabolic
stability, such as chloro and trifluoromethyl, but these
modifications reduced the in vitro potency by approximately
4- to 5-fold. Therefore, our efforts focused on identifying
further in vitro potency improvements, which would allow for
the replacement of the propyl group with a more metabolically
stable group. The one unexplored area of the original lead was
the pyridone ring. Substitution of the 3- or 6-position of the
pyridone was not tolerated (see Table 4). However, small
Figure 2. oGTT of compound 29 and metformin in male db/db mice
at 6 weeks of age. Metformin was dosed at 300 mg/kg. Glucose AUC
reductions were −83% for metformin, −16% for 29 at 0.3 mg/kg,
−32% for 29 at 3 mg/kg, and −38% for 29 at 30 mg/kg. ∗ indicates a
p value of <0.05.
Table 4. In Vitro Activity of Pyridone Analogs
AUC reductions were 32% and 38%, respectively, which are
half the response obtained with high dose metformin treatment
(−83%). The magnitude of the response compares well with
the maximum reduction in glucose AUC (∼40%) we have
achieved over multiple studies with multiple GPR119 agonists
and is equivalent to reductions seen with clinical DPPIV
inhibitors in this model. After 21 days of treatment, the 30 mg/
kg dose of 29 produced a statistically significant reduction in
the HbA1C levels comparable to metformin (see Figure 3). In
the vehicle-treated animals, HbA1C increased 2.8% while only
a
b
R
hGPR119 EC50, nM
hGPR119 I.A.
0.8
22
23
24
25
26
27
28
H
16
>10000
>10000
31
6-Me
3-Cl
5-Me
5-Ph
5-CN
5-Cl
0.9
>10000
12
0.8
0.8
5
a
EC50 values were calculated from concentration−response curves. See
b
ref 18 for assay conditions. Reference compound set at 1.0 for each
assay run.
substitutions were tolerated at the 5-position, such as methyl
(33) or cyano (35), but large groups, such as phenyl (34),
ameliorated all activity. The desired potency improvement was
achieved with the incorporation of a chloro group (36) which
provided a ∼3-fold increase in potency versus 29.
The combination of the metabolically stable 4-trifluorome-
thylpyrimidine and the 4-chloropyrimidine group with the 5-
chloropyridone core produced compounds with the desired
potencies (see Table 5). The 4-trifluoromethylpyrimidine (37)
and the 4-chloropyrimidine (40) combined excellent potency
with improved human liver microsome half-lives, with a greater
than 4-fold improvement for 37 and almost 2-fold for 40 over
the original in vivo compound 29. Concerned that the long
metabolic half-life for 37 would cause extended in vivo
Figure 3. Effects on HbA1C of compound 29 and metformin after 21
days of dosing in male db/db mice starting at 6 weeks of age.
Metformin was dosed at 300 mg/kg. The HbA1C reductions related to
vehicle were 1.4% for metformin, 0.6% for 29 at 0.3 mg/kg, 0.4% for
29 at 3 mg/kg, and 1.4% for 29 at 30 mg/kg. ∗ indicates a p value of
<0.05.
7502
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
a
Table 5. In Vitro Activity in Final Optimization
R1
R2
R3
hEC50 (nM)
sol., pH 6.5 (μg/mL)
sol., 50% PEG (μg/mL)
HLM T1/2 (min)
mPK T1/2 (h)
rPK T1/2 (h)
37
38
39
40
41
42
CF3
CF3
CF3
Cl
H
F
H
H
F
13
12
16
18
13
14
<1
1
3
122
31
>120
>120
>120
59
16
30
27
11
18
15
>72
>72
>48
9
H
H
H
F
<1
<1
3
H
F
7
Cl
199
41
39
12
Cl
H
<1
84
22
a
EC50 values were calculated from concentration−response curves. See ref 18 for assay conditions. All intrinsic activities for the GPR119 assay were
between 0.8 and 0.9.
exposure, we investigated the entire set of trifuoromethyl
substituted compounds in rat pharmacokinetic studies and
discovered that they had extremely long half-lives in vivo
(greater than 2 days). As we feared, this would translate to an
excessively long half-life in humans; the chloropyrimidine
derivates (40−42) were believed to demonstrate a more ideal
metabolic profile, exhibiting a good in vitro metabolic stability
across species without extended in vivo half-lives in rodents. As
such, we focused our selection of a clinical candidate from the
chloropyrimidines.
fraction affords a large safety window which was confirmed in
both rabbit and cynomolgus monkey electrophysiology studies
(data not shown) in which no in vivo effects were seen at a
plasma concentration of 20 μM. The rat pharmacokinetic
profile of 42 was tested at 1 mg/kg intravenously as a solution
and 5 mg/kg orally as a hydroxypropyl methylcellulose
(HPMC) SDD loaded at 30% and as a crystalline suspension
(see Table 6). When dosed as the SDD formulation, the
Table 6. Single Dose Plasma Pharmacokinetic Parameters of
42 in Male Sprague-Dawley Rats
Although 40 appeared to have an ideal liver microsomal
stability and in vivo half-life profile, its poor aqueous solubility
posed concerns regarding bioavailability (especially at high
doses) when administered as a crystalline suspension.
Fortunately, we found that use of amorphous spray dried
dispersion (SDD) technology19 could overcome the dissolution
issues to provide excellent bioavailability in most cases. In an
attempt to rationally apply this technology to our best analogs,
we wanted an easy method to determine which compounds had
a high likelihood of having excellent bioavailability with SDD
formulation. Therefore, in conjunction with the SDD work, we
identified a solubility test for this purpose. It was empirically
determined that compounds from this chemotype with a
solubility of greater than 20 μg/mL in a 50% PEG400/50%
PBS solution typically exhibited bioavailabilities of greater than
75%. Interestingly, when the fluorine substituted compounds
(38, 39, 41, 42) were tested in this solubility screen, all
demonstrated much higher solubilities than the des-fluoro
analogs 37 and 40, contrary to normal medicinal chemistry
dogma. Although 41 and 42 had the necessary solubility
characteristics for SSD formulation, based on its longer human
liver microsomal half-life, 42 was chosen for further evaluation
as a possible clinical candidate.
Compound 42 had an excellent profile in our vitro safety
assays. The molecule demonstrated no significant inhibition
(IC50 > 40 μM) at nine different cytochrome P450 enzymes,
did not activate PXR (EC50 > 50 μM), and exhibited no toxicity
in a hepatic (HEPG2) cell line at a concentration of 40 μM.
Compound 42 was also tested against a panel of 95 GPCRs and
enzymes in a broad receptor screening panel and was found to
exhibit negligible activity at a concentration of 10 μM.
Genotoxicity testing in both the Ames and in vitro micro-
nucleus assay was negative. While the compound inhibits hERG
in a patch clamp assay with an IC50 of 1.4 μM, it is 99.5%
protein bound in human plasma. The reduced plasma free
dose, iv/po (mg/kg)
po Tmax (h)
1/5
4
po Cmax (μM)
po T1/2 (h)
2.7
22
po AUCtot (μM·h)
po F (%)
95
33% as suspension
87% as HMPC SDD
iv Cl (mL min−1 kg−1
iv Vss (L/kg)
)
1.5
2.2
compound achieved a maximum plasma concentration of 2.7
μM at Tmax (4 h) with a terminal half-life of 16 h. As predicted
by the solubility assay, the bioavailability was high at 87%,
whereas the observed bioavailability was low at 33% when
dosed as the crystalline suspension. The superior half-life of 42
allowed us to determine if once daily low doses of a potent
GPR119 compound can achieve the desired in vivo efficacy
responses.
To understand the acute in vivo efficacy of 42, an oral
glucose tolerance test in C57/Bl6 mice was run in a dose
response mode (see Figure 4). Statistically significant
reductions in glucose were seen at the 30 and 60 min time
points for 0.1−1 mg/kg doses and at the 60 min time point for
the 0.03 mg/kg dose. The corresponding reductions in glucose
AUC were about 40% for all doses at or above 0.1 mg and were
statistically significant as compared to vehicle control. Although
the data are not presented here, these reductions are similar to
those seen with DPPIV inhibitors in this model.
We then examined the ability of 42 to release GLP-1 upon
oral glucose administration using a cannulated Sprague−
Dawley rat model (see Figure 5). The rats were treated with
drug 1 h before an oral glucose bolus, and plasma samples were
taken over the subsequent hour. Compound 42 significantly
7503
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
Figure 4. Oral glucose tolerance test of compound 42 in male C57/b6
mice. The glucose AUC reductions versus vehicle control were −14%
at 0.03 mg/kg, −37% at 0.1 mg/kg, −39% at 0.3 mg/kg, and −40% at
1 mg/kg. ∗ indicates a p value of <0.05.
Figure 6. Effect on fasting plasma glucose levels of compound 42 after
21 days of dosing in male db/db mice starting at 6 weeks of age. The
plasma glucose reductions relative to vehicle were 18% at 0.01 mg/kg,
36% at 0.03 mg/kg, 13% at 3 mg/kg, and 38% at 30 mg/kg. ∗ indicates
a p value of <0.05.
Figure 5. Effect of compound 42, DPPIV inhibitor,21 and their
combination on active GLP-1 levels in cannulated male Sprague−
Dawley rats. ∗ indicates a p value of <0.05.
Figure 7. Effect on fasting plasma insulin levels of compound 42 after
21 days of dosing in male db/db mice starting at 6 weeks of age. ∗
indicates a p value of <0.01.
increased active GLP-1 levels over vehicle at all time points
including before glucose bolus. This is consistent with literature
reports10,20 and is equivalent to the response with the DPPIV
inhibitor used in this assay. The combination of 42 with the
DPPIV inhibitor provided synergistic increases in active GLP-1
at almost all time points including prior to administration of the
glucose bolus. These results suggest diabetic patients treated
with a GPR119/DPPIV combination might experience
dramatic increases in glucose control, perhaps mirroring those
seen with GLP-1 injections, but with the convenience of an oral
dosing regimen.
The chronic efficacy of 42 was evaluated in the 3-week db/db
mouse model used previously in the evaluation of 29. However,
the efficacy end points were changed to fasting plasma glucose
and fasting plasma insulin. The once daily doses were chosen to
bracket the activity seen in the acute mouse oGTT study.
Statistically significant reductions in fasting plasma glucose on
day 21 were seen at doses as low as 0.03 mg/kg (see Figure 6).
Unfortunately, because of a handling error that required the
removal of 6 of 10 animals, the 0.1 mg/kg dose did not achieve
significance. In the measurement of the fasting plasma insulin
on day 21 (Figure 7), elevations in insulin concentration were
seen at all doses of 42, whereas the insulin levels of the vehicle
animals had dropped 2-fold, even though the glucose levels had
risen dramatically. The preservation of the animal’s ability to
secrete insulin suggests that compound 42 might be preventing
the β-cell mass depletion normally seen in this rapidly
progressing diabetes model. It should be noted that the effects
on glucose parameters were seen without any weight loss in any
of the drug treated groups. The combination of acute efficacy in
the GLP-1 elevation studies and the potent activity
demonstrated in the chronic efficacy model led us to choose
compound 42 for progression into development and ultimately
phase 1 clinical trials.
CHEMISTRY
■
The synthesis of compound 42 and analogs described above
was achieved in a straightforward four-step route (Scheme 1).
In the case of 42, the synthesis began with the coupling of
gimeracil with the protected 4-hydroxypiperidine under stand-
ard Mitsunobu conditions to yield intermediate 43. 1,2-
Difluoro-4-methylsulfonylbenzene was reacted with the sodium
salt of 43 in dimethylformamide. After the mixture was heated
to 130 °C for 2 h, intermediate 44 was isolated. The protecting
group of 44 was removed with aqueous hydrogen chloride to
yield the piperidine hydrochloride salt, 45. The synthesis was
completed by coupling 45 to 5-chloro-2-iodopyrimidine in
dimethylformamide with diisopropylethylamine to yield 42.
The compounds from this series could also be prepared by a
simple rearrangement of the reaction order, such as precoupling
the pyrimidine with the piperidine before the Mitsunobu
reaction with gimeracil followed by coupling with the
methylsulfonylphenyl group.
7504
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
Scheme 1. Synthesis of 42
was little expectation to see changes in normal healthy
volunteers who have robust glucose control mechanisms. As
expected, there was no change in the 24 h AUC for glucose
from the day before drug treatment to the first day following
drug treatment (data not shown). However, there was a
nonstatistically significant trend for increases in total GLP-1
AUC levels over the first 24 h (see Figure 9). With a dose as
CLINICAL TRIAL RESULTS
■
A phase 1 clinical trial with 42 has been conducted as a single
ascending dose study in normal healthy volunteers under a
double-blind, placebo controlled protocol to determine plasma
exposure and tolerability. The doses were escalated from 0.1 to
120 mg of drug with the first two doses given as solutions and
the remainder given as SDD suspensions. Compound 42 was
safe and well tolerated at all doses, with plasma exposure (see
Figure 8) increasing in a dose proportional manner. The
Figure 8. Pharmacokinetics of compound 42 orally dosed in normal
healthy human volunteers. The 0.1 and 0.6 mg doses were given as
oral solutions, and doses of 3 mg and above were given as oral
suspension of an amorphous spray−dried dispersion.
Figure 9. Effects of compound 42 on total GLP-1 area under the curve
over 24 h. Data are expressed as percent change in total GLP-1 AUC
between day −1 and day 1 in six patients at each dose.
terminal half-life was determined to be between 36 and 51 h.
The pharmacokinetic profile shows increases in drug plasma
concentrations at 10 and 24 h in every dose panel. These time
points correspond to meal time for the patients and indicate
possible enterohepatic recirculation of the drug. This provides
one the chance to speculate that the addition of drug to the
gastrointestinal tract every time a patient consumes glucose
could provide efficacy benefits from the activation of the
GPR119 receptors in the GI tract.
low as 0.6 mg, there is a 20% increase in total GLP-1 AUC and
a dose dependent response up to a plateau of ∼50% AUC
increase with the 10 mg and higher dose panels. Although the
data are not a robust response and have no statistical value,
even a trend in the correct direction in healthy volunteers is
encouraging and data from diabetic patients are eagerly awaited.
EXPERIMENTAL SECTION
■
General Methods. For anhydrous reactions DRISOLV solvents
from EMD were used. For other reactions reagent or HPLC grade
solvents were used. All commercially available solvents and reagents
In the phase 1 study, efficacy readouts such as glucose levels
and GLP-1 levels were measured over 24 h, even though there
7505
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
were used as received. All reactions were carried out under an argon
atmosphere unless noted otherwise. Reaction mixtures were
concentrated under reduced pressure at 40−65 °C on a rotary
evaporator. Analytical thin layer chromatography (TLC) was
performed on silica gel 60 F254 plates from Merck. Flash column
chromatography separations were preformed on RediSep normal
phase silica flash columns (230−400 mesh, 60 Å, Teledyne ISCO).
LC−MS measurements were obtained using either a Shimadzu HPLC
system with a Phenomenex Luna C18 column (5 μm, 100 Å, 4.6 mm
× 30 mm), operated at 40 °C with 0.1% TFA modified ACN/water
mobile phases or a Waters Acquity UPLC system with an Acquity
UPLC BEH C18 (1.7 μm, 130 Å, 2.1 mm × 50 mm) column operated
at 50 °C with 0.05% TFA modified ACN/water mobile phases. UV
detection was carried out with a Shimadzu SPD-10AV at 220 nm, and
mass detection was carried out with Waters ZQ single quadrupole
mass spectrometer hybrid system using positive electrospray
ionization.
Final HPLC purity determination was made with a Shimadzu
HPLC system, with a Waters Sunfire C18 (3.5 μm, 100 Å, 3.0 mm ×
150 mm) column and a Waters Xbridge phenyl (3.5 μm, 130 Å, 3.0
mm × 150 mm) column with 0.05% TFA modified ACN/water
mobile phases with gradient run from 10% to 90% ACN. UV detection
was carried out with a Shimadzu SPD-20AV at 220 and 254 nm. All
final compounds achieved a minimum of 95% purity. All
chromatography was carried out with HPLC grade organic solvents.
1H NMR spectra were obtained on a Bruker 400 MHz spectrometer
ppm 7.82−7.91 (m, 2 H), 7.62 (dd, J = 6.78, 8.53 Hz, 1 H), 7.32 (s, 1
H), 6.03 (s, 1 H), 4.57−4.64 (m, 1 H), 3.59−3.68 (m, 2 H), 3.46−3.55
(m, 2 H), 3.11 (s, 3 H), 1.84−2.02 (m, 4 H), 1.48 (s, 9 H). LC−MS:
m/z = 501.2 [M + H].
5-Chloro-1-(2-fluoro-4-(methylsulfonyl)phenyl)-4-(piperi-
din-4-yloxy)pyridin-2(1H)-one, HCl (45). A mixture of tert-butyl 4-
(5-chloro-1-(2-fluoro-4-(methylsulfonyl)phenyl)-2-oxo-1,2-dihydro-
pyridin-4-yloxy)piperidine-1-carboxylate (44) (1.31 g, 2.62 mmol) and
HCl (37% aq, 5.00 mL, 60.9 mmol) was stirred at room temperature
for 15 min. The reaction mixture was concentrated in vacuo to give 45
(1.15 g, 2.62 mmol, 100% yield) as a white solid. 1H NMR (400 MHz,
DMSO-d6) δ 8.96 (br s, 2H), 8.12 (s, 1H), 8.02 (dd, J = 1.76, 9.24 Hz,
1H), 7.88−7.94 (m, 1H), 7.79−7.87 (m, 1H), 6.31 (s, 1H), 4.80−4.92
(m, 1H), 3.35 (s, 3H), 3.16−3.26 (m, 2H), 3.03−3.16 (m, 2H), 2.10−
2.23 (m, 2H), 1.86−1.99 (m, 2H). LC−MS: m/z = 401.1 [M + H].
5-Chloro-4-(1-(5-chloropyrimidin-2-yl)piperidin-4-yloxy)-1-
(2-fluoro-4-(methylsulfonyl)phenyl)pyridin-2(1H)-one (42). To
a solution of 5-chloro-1-(2-fluoro-4-(methylsulfonyl)phenyl)-4-(piper-
idin-4-yloxy)pyridin-2(1H)-one, HCl (45) (1.15 g, 2.62 mmol) and
DIEA (1.37 mL, 7.85 mmol) in DMF (10 mL) was added 5-chloro-2-
iodopyrimidine (0.755 g, 3.14 mmol). The reaction mixture was
heated at 60 °C for 4 h. The reaction mixture was concentrated in
vacuo to give a yellow oil. The oil was dissolved in EtOAc (200 mL)
and washed with saturated NaHCO3 (200 mL), and the organic layer
was concentrated in vacuo. The residue was purified by flash column
chromatography (0−100%, EtOAc in DCM) to give the title
compound as an off-white solid. To increase the purity of the isolated
solid, the solid was added to MeOH (250 mL) and refluxed for 6 h.
The slurry was cooled to 4 °C, filtered, washed with MeOH (100 mL),
and dried in vacuo giving 42 (0.934 g, 1.82 mmol, 69.5% yield) as a
using the indicated solvent. Chemical (δ) shifts are reported in ppm
from tetramethylsilane with the residual solvent signal as the internal
standard; signals are expressed as s = singlet, d = doublet, t = triplet, q
= quartet, m = multiplet, and br = broad. Coupling constants (J) are in
hertz (Hz).
All animals were treated according to the standards of the BMS
Animal Care and Use Committee (ACUC). Animals were maintained
at the Bristol-Myers Squibb animal facilities (Hopewell, NJ) with 12 h
light/dark cycle and free access to food and water. The animals were
fed ad libitum normal chow.
1
white solid. H NMR (400 MHz, CDCl3) δ 8.25 (s, 2H), 7.82−7.94
(m, 2H), 7.63 (dd, J = 6.65, 8.41 Hz, 1H), 7.33 (d, J = 1.00 Hz, 1H),
6.07 (s, 1H), 4.70 (tt, J = 3.39, 6.53 Hz, 1H), 3.98−4.08 (m, 2H),
3.87−3.96 (m, 2H), 3.12 (s, 3H), 2.02−2.11 (m, 2H), 1.92−2.01 (m,
2H). LC−MS: m/z = 513.0 [M + H].
21-Day db/db Mouse Study. A cohort of 90 male BKS Cg-m+/+
Lepr db/db/J mice (5 weeks of age at arrival) purchased from Jackson
Laboratories (Bar Harbor, ME) was acclimated to the vivarium for 1
week prior to study start. Two days before the study date, mice were
randomized to day 1 or day 2 oGTT based on fed blood glucose and
HbA1c levels, measured by glucometer (Accu-Check) and COBAS
Mira, respectively, via tail nick. Following an overnight fast, mice were
randomized to treatment based on their blood glucose and body
weights. After 40 min in adaptation to the new group, mice were bled
again for blood glucose level at the −60 min time point, and 5 μL of
whole blood was collected into a COBAS tube containing 250 μL of
hemolysate reagent for determination of HbA1c. After the blood
collection, mice received vehicle (40% PEG 400, 10% Cremophor, and
50% water) or compound 29 orally (5 mL/kg). At 60 min after dosing,
an additional glucose measurement was taken. Immediately after the 0
min time point collection, mice were given an oral glucose challenge
with 50% dextrose (4 mL/kg). Plasma glucose was measured at 30, 60,
90, and 120 min after the glucose challenge. Mice were then returned
to home cages and given free access to food and water. From day 2 to
day 21, mice were orally dosed with the vehicle or compound once
daily at 9:00 a.m. After day 21 of dosing, mice were fasted overnight
and fasting plasma glucose and HbA1c were measured at 24 h after
dose, followed by a day 22 dosing in which trunk blood from three
mice of each group was collected for PK analysis by decapitation under
anesthesia with CO2 3.5 h after compound administration. All blood
samples (except the 5 μL blood for HbA1c) were centrifuged at 8000
rpm at 4 °C for 10 min. Remaining mice (seven mice in each group)
were returned to home cage and given access to food and water. The
24 h postdose plasma levels of compounds from trunk blood were
determined as well.
tert-Butyl 4-((5-Chloro-2-oxo-1,2-dihydropyridin-4-yl)oxy)-
piperidine-1-carboxylate (43). To an ice-cooled mixture of 5-
chloro-4-hydroxypyridin-2(1H)-one (11.86 g, 81 mmol), tert-butyl 4-
hydroxypiperidine-1-carboxylate (16.40 g, 81 mmol), and triphenyl-
phosphine (23.51 g, 90 mmol) in a mixture of DMF (55 mL) and
THF (55 mL) was added dropwise a solution of (E)-diisopropyl
diazene-1,2-dicarboxylate (17.65 mL, 90 mmol) in THF (50 mL) over
30 min. The mixture was allowed to warm to room temperature and
stirred for 2 days. The reaction mixture was concentrated in vacuo to
give a yellow oil. The oil was dissolved in EtOAc (300 mL), washed
with saturated NaHCO3 (150 mL), and the organic layer was
concentrated in vacuo. The residue was purified by flash column
chromatography. The column was initially eluted with EtOAc in DCM
(0−100%) until the impurities were off the column, and the mobile
phase was switched to MeOH in DCM (0−10%) to collect 43 (14.7 g,
1
55%) as a white solid. H NMR (400 MHz, CDCl3) δ ppm 12.98 (br
s, 1 H), 7.35 (s, 1 H), 5.92 (s, 1 H), 4.51−4.63 (m, 1 H), 3.57−3.68
(m, 2 H), 3.43−3.54 (m, 2 H), 1.79−2.00 (m, 4 H), 1.48 (s, 9 H).
LC−MS: m/z = 329.0 [M + H].
tert-Butyl 4-(5-Chloro-1-(2-fluoro-4-(methylsulfonyl)-
phenyl)-2-oxo-1,2-dihydropyridin-4-yloxy)piperidine-1-car-
boxylate (44). To a solution of tert-butyl 4-(5-chloro-2-oxo-1,2-
dihydropyridin-4-yloxy)piperidine-1-carboxylate (43) (1.93 g, 5.87
mmol) in DMF (15 mL) was added portionwise sodium hydride (60%
w/w in mineral oil, 0.282 g, 7.04 mmol) over a period of 5 min. The
mixture was stirred at room temperature for 30 min, followed by the
addition of 1,2-difluoro-4-(methylsulfonyl)benzene (1.47 g, 7.63
mmol). The reaction mixture was heated at 130 °C for 2 h, cooled
to room temperature, and quenched by the addition of H2O (200
mL). The quenched reaction mixture was extracted with EtOAc (200
mL). The organic layer was concentrated in vacuo. The residue was
purified by flash column chromatography (0−100% EtOAc in DCM),
with the unwanted O-isomer eluting before the desired N-isomer, to
give 44 (1.31 g, 45%) as a white solid. 1H NMR (400 MHz, CDCl3) δ
Mouse Acute Oral Glucose Tolerance Test Assay. Male
C57BL/6J mice (8 weeks of age at arrival) from Jackson Laboratories
(Bar Harbor, ME) were single-housed and had free access to normal
chow (Teklad 2018) and water ad libitum. They were allowed to
acclimate to the animal facilities for 2 weeks. The mice were fasted
7506
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
overnight before the study. On the morning of the experiment day, the
mice (10 weeks of age) were acclimated to the lab for 3 h before the
baseline plasma glucose values and body weights were collected.
Whole blood was obtained via tail nick, and the fasting plasma glucose
was determined via glucometer (Accu-check, Aviva, Roche). Mice were
randomized based on fasting plasma glucose and body weight and
assigned to either vehicle or one of the treatment groups (n = 8/
group). The plasma glucose baseline values were obtained at the −60
min time point immediately prior to dosing with vehicle (40% PEG
400, 10% Cremophor, and 50% water) or 42 (10 mL/kg) in vehicle.
Plasma glucose values were determined again at 60 min after initial
dose (0 time point), before a bolus of oral glucose (2 g/kg, 8 mL/kg)
was given to the mice. Plasma glucose samples were collected again at
30, 60, and 120 min after glucose challenge, respectively. At 3.5 h
postdrug dose, mice were decapitated and the trunk blood was
collected. The blood samples were centrifuged at 8000 rpm for 10 min
at 4 °C. An aliquot of 200 μL plasma sample was separated from each
blood sample and stored at −80 °C.
Blood Sample Collection and Analysis. Glucose Assay. A
COBAS clinical analyzer (Roche) was used to measure glucose levels
in the plasma in the db/db mouse study with 42. An Accu-check
glucometer (Roche) was used to determine plasma glucose from blood
samples collected from tail bleed in the C57BL/6J and SD rat studies
to get the immediate reading.
GLP-1 Assay. An aliquot (200 μL) of plasma was transferred into a
prechilled Eppendorf tube. Tubes were kept on dry ice and then stored
at −80 °C until used in assays for GLP-1 and insulin. Plasma active
GLP-1 was measured using a glucagon-like peptide-1 (active) ELISA
kit (LINCO Research, St. Charles, MO). The samples were tested
according to the manufacturer’s instruction for 100 μL per sample
protocol.
Insulin Assay. An aliquot (200 μL) of plasma was transferred into a
prechilled Eppendorf tube. Tubes were kept on dry ice and then stored
at −80 °C until used in assays for GLP-1 and insulin. Plasma insulin
was measured by Mercodia rat or mouse ultrasensitive insulin ELISA
(Mercodia, Winston-Salem, NC). The samples were tested according
to the manufacturer’s instructions for 5 μL per sample.
Cannulated Rat GLP-1 Assay. Male Sprague−Dawley (SD) rats
(body weight, 250−275g) with carotid artery cannulation (CAC) were
obtained from Charles River Laboratories (Wilmington, MA).
Additional age-matched noncannulated SD rats (donor rats) were
used to obtain blood in order to replace blood taken from CAC rats
during the study. Overnight fasted donor rats were anesthetized by
isofluorane. From each donor rat, 8−10 mL of blood was collected by
cardiac puncture. The donor blood was transferred to a vial containing
heparin solution with a final concentration of 20 U/mL. For the
glucose tolerance test, overnight fasted CAC rats (n = 10/group) were
connected to extension PE50 tubing. The lock solution was
withdrawn, and the line was flushed with heparinized saline (20 U/
mL). The animals were allowed to acclimate for 60 min before the
baseline blood sample was taken. Blood samples (0.5 mL each) were
collected at each time point (−60, 0, 2, 5, 10, 15, 30, and 60 min)
through the carotid artery line into 1 mL hematology tubes coated
with EDTA and a complex of aprotinin and protease cocktail enzyme
inhibitors (Sigma, St. Louis, MO) at a final concentration of 40 U/mL.
Plasma glucose levels were measured at these time points during the
study via glucometer (Accu-check, Aviva, Roche Diagnostics, Indian-
apolis, IN). Rats were randomized into either vehicle or one of the
treatment groups based on fasting plasma glucose values at the −60
min time point. Following randomization, rats were dosed with vehicle
(40% PEG 400, 10% Cremophor, and 50% water) or 42. One hour
later, a bolus of glucose (2 g/kg, 4 mL/kg) was given at 0 min time
point after a blood sample was taken. Blood samples were kept on ice
until processed. Blood samples were centrifuged at 8000 rpm for 10
min at 4 °C. The plasma samples were harvested and kept at −80 °C
until used for GLP-1 assessment.
Second 21-Day db/db Mouse Study. A cohort of male BKS Cg-
m+/+ Lepr db/db/J mice (5 weeks of age at arrival) was purchased
from Jackson Laboratories (Bar Harbor, ME). They were allowed to
acclimate to the animal facility for 2 weeks. They were housed 5 per
cage and had free access to normal chow (Teklad 2018) and water ad
libitum The mice (7 weeks of age) were fasted overnight. The next
morning, their fasting body weight was recorded and blood samples
were collected in EDTA tubes via tail nick for determination of plasma
glucose and insulin levels. The blood samples were centrifuged at 8000
rpm for 10 min at 4 °C. The fasting plasma glucose levels were
determined by a COBAS clinical chemistry analyzer (Roche
Diagnostics, Indianapolis, IN). The samples were frozen, and plasma
insulin was determined by ELISA at a later time. The 60 mice were
randomized into the vehicle group or one of the treatment groups (n =
8/group). Following 3 weeks of dosing with vehicle (40% PEG 400,
10% Cremophor, and 50% water) or compound 42 orally (5 mL/kg)
(after the 21st dose was given), animals were fasted. On day 22, the
fasting body weight, plasma glucose, and insulin levels were
determined 24 h after the 21st dose. In order to assess plasma
exposure of the compound, the mice were dosed again after the blood
samples were collected for glucose and insulin. Blood was collected at
3 and 24 h after dose, and plasma concentrations of 42 were
determined in these samples.
HbA1c Assay. Blood HbA1c levels were determined by a COBAS
clinical analyzer (Roche Diagnostics, Indianapolis, IN).
ASSOCIATED CONTENT
* Supporting Information
Synthetic procedures for analogs of 42. This material is
■
S
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: 609-818-5381. E-mail: dean.wacker@bms.com.
■
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS USED
■
GPR119, G-protein-coupled receptor 119; GLP-1, glucagon-
like peptide 1; GIP, glucose-dependent insulinotropic polypep-
tide; GPCR, G-protein-coupled receptor; PXR, pregnane X
receptor; HEPG2, human liver carcinoma cell line; hERG,
human ether-a-go-go-related gene; PAMPA, parallel artificial
membrane permeability assay; AUC, area under curve; oGTT,
oral glucose tolerance test; HbA1C, hemoglobin A1C; DPPIV,
dipeptidyl peptidase 4; PEG, polyethylene glycol; SDD, spray
dried dispersion; PBS, phosphate buffered saline; HPMC,
hydroxypropyl methylcellulose
REFERENCES
■
(1) International Diabetes Federation, Diabetes Atlas. http://www.
idf.org/sites/default/files/EN_6E_Atlas_Full_0.pdf, 2013
(2) Centers for Disease Control and Prevention, National Diabetes
Fact Sheet, 2011. http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.
pdf, 2011
(3) Koro, C. E.; Bowlin, S. J.; Bourgeois, N.; Fedder, D. O. Glycemic
control from 1988 to 2000 among U.S. adults diagnosed with type 2
diabetes. Diabetes Care 2004, 27, 17−20.
(4) Philippe, J.; Raccah, D. Treating type 2 diabetes: How safe are
current therapeutic agents? Int. J. Clin. Pract. 2009, 63, 321−332.
(5) Holst, J. J.; Vilsboll, T.; Deacon, C. F. The incretin system and its
role in type 2 diabetes mellitus. Mol. Cell. Endocrinol. 2009, 297, 127−
136.
(6) Brubaker, P. L. Update on incretin biology: focus on glucagon-
like peptide 1. Endocrinology 2010, 151, 1984−1989.
7507
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508

Journal of Medicinal Chemistry
Drug Annotation
(7) Fredriksson, R.; Hoglund, P. J.; Gloriam, D. E. I.; Lagerstrom, M.
C.; Schioth, H. B. Seven evolutionarily conserved human rhodopsin G
protein-coupled receptors lacking close relatives. FEBS Lett. 2003, 554,
381−388.
Parker, R. A.; Kirby, M. S.; Hamann, L. G. Potent non-nitrile
dipeptidic dipeptidyl peptidase IV inhibitors. Bioorg. Med. Chem. Lett.
2007, 17, 6476−6480.
(8) Soga, T.; Ohishi, T.; Matsui, T.; Saito, T.; Matsumoto, M.;
Takasaki, J.; Matsumoto, S.; Kamohara, M.; Hiyama, H.; Yoshida, S.;
Momose, K.; Ueda, Y.; Matsushime, H.; Kobori, M.; Furuichi, K.
Lysophosphatidylcholine enhances glucose-dependent insulin secre-
tion via an orphan G-protein-coupled receptor. Biochem. Biophys. Res.
Commun. 2005, 326, 744−751.
(9) Chu, Z.-L.; Jones, R. M.; He, H.; Carroll, C.; Gutierrez, V.;
Lucman, A.; Moloney, M.; Gao, H.; Mondala, H.; Bagnol, D.; Unett,
D.; Liang, Y.; Demarest, K.; Semple, G.; Behan, D. P.; Leonard, J. A
role for β-cell-expressed G protein-coupled receptor 119 in glycemic
control by enhancing glucose-dependent insulin release. Endocrinology
2007, 148, 2601−2609.
(10) Chu, Z.-L.; Carroll, C.; Alfonso, J.; Gutierrez, V.; He, H.;
Lucman, A.; Pedraza, M.; Mondala, H.; Gao, H.; Bagnol, D.; Chen, R.;
Jones, R. M.; Behan, D. P.; Leonard, J. A role for intestinal endocrine
cell-expressed G protein-coupled receptor 119 in glycemic control by
enhancing glucagon-like peptide-1 and glucose-dependent insulino-
tropic release. Endocrinology 2008, 149, 2038−2047.
(11) Overton, H. A.; Babbs, A. J.; Doel, S. M.; Fyfe, M. C. T.;
Gardner, L. S.; Griffin, G.; Jackson, H. C.; Procter, M. J.; Rasamison,
C. M.; Tang-Christensen, M.; Widdowson, P. S.; Williams, G. M.;
Reynet, C. Deorphanization of a G protein-coupled receptor for
oleoylethanolamide and its use in the discovery of small-molecule
hypohagic agents. Cell Metab. 2006, 3, 167−175.
(12) Jones, R. M.; Semple, G.; Xiong, Y.; Shin, Y.-J.; ren, A. S.;
Calderon, I.; Fioravant, B.; Choi, J. S. K.; Sage, C. R. Preparation of
fused aryl and heteroaryl derivatives, in particular pyrazolo[3,4-
d]pyrimidines, as modulators of G-coupled protein receptor and their
use in the prophylaxis and treatment of metabolic disorders. World
Patent Application WO 2005/007658, January 27, 2008.
(13) Semple, G.; Lehmann, J.; Wong, A.; Ren, A.; Bruce, M.; Shin, Y.
J.; Sage, C. R.; Morgan, M.; Chen, W.-C.; Sebring, K.; Chu, Z.-L.;
Leonard, J. N.; Al-Shamma, H.; Grottick, A. J.; Du, F.; Lian, Y.;
Demarest, K.; Jones, R. M. Discovery of a second generation of the
orphan G-protein coupled receptor GPR119 with an improved profile.
Biorg. Med. Chem. Lett. 2012, 22, 1750−1755.
(14) Thomas Reuters Drug News. http://drugnews.thomson-
pharma.com/ddn/home.do.
(15) McWherter, C. Presented at the 32nd Annual National
Medicinal Chemistry Symposium, Minnesota, MN, June 6−9, 2010;
Oral Session 2.
(16) Carpenter, A. J. Presented at the 32nd Annual National
Medicinal Chemistry Symposium, Minnesota, MN, June 6−9, 2010;
Oral Session 2.
(17) PSN821 structure has not been disclosed. Its close lead PSN408
was disclosed; see the following: The 32nd Annual National Medicinal
Chemistry Symposium, Minnesota, MN, June 6−9, 2010; Oral Session
2.
(18) Ye, X.-Y.; Morales, C. L.; Wang, Y.; Rossi, K. A.; Malmstrom, S.
E.; Abousleiman, M.; Sereda, L.; Apedo, A.; Robl, J. A.; Miller, K. J.;
Krupinski, J.; Wacker, D. A. Synthesis and structure−activity
relationship of dihydrobenzofuran derivatives as novel human
GPR119 agonists. Bioorg. Med. Chem. Lett. 2014, 24, 2539−2545.
(19) Friesen, D. T.; Shanker, R.; Crew, M.; Smithey, D. T.; Curatolo,
W. J.; Nightingale, J. A. S. Hydroxypropyl methylcellulose acetate
succinate-based spray-dried dispersions: an overview. Mol. Pharmaceu-
tics 2008, 5, 1003−1019.
(20) Lauffer, L. M.; Iakoubov, R.; Brubaker, P. L. GPR119 is essential
for oleoylethanolamide-induced glucagon-like peptide-1 secretion from
the intestinal enteroendocrine L-cell. Diabetes 2009, 58, 1058−1066.
(21) Compound 48 from following reference was used as the DPPIV
inhibitor in the study: Simpkins, L. M.; Bolton, S.; Pi, Z.; Sutton, J. C.;
Kwon, C.; Zhao, G.; Magnin, D. R.; Augeri, D. J.; Gungor, T.; Rotella,
D. P.; Sun, Z.; Liu, Y.; Slusarchyk, W. S.; Marcinkeviciene, J.;
Robertson, J. G.; Wang, A.; Robl, J. A.; Atwal, K. S.; Zahler, R. L.;
7508
dx.doi.org/10.1021/jm501175v | J. Med. Chem. 2014, 57, 7499−7508