Discovery of Potent and Orally Bioavailable Dihydropyrazole GPR40 Agonists

Article abstract of DOI:10.1021/acs.jmedchem.7b00982

G protein-coupled receptor 40 (GPR40) has become an attractive target for the treatment of diabetes since it was shown clinically to promote glucose-stimulated insulin secretion. Herein, we report our efforts to develop highly selective and potent GPR40 agonists with a dual mechanism of action, promoting both glucose-dependent insulin and incretin secretion. Employing strategies to increase polarity and the ratio of sp³/sp² character of the chemotype, we identified BMS-986118 (compound 4), which showed potent and selective GPR40 agonist activity in vitro. In vivo, compound 4 demonstrated insulinotropic efficacy and GLP-1 secretory effects resulting in improved glucose control in acute animal models.

Full text of DOI:10.1021/acs.jmedchem.7b00982

Subscriber access provided by UNIV OF DURHAM
Article
Discovery of Potent and Orally Bioavailable Dihydropyrazole GPR40 Agonists
Jun Shi, Zhengxiang Gu, Elizabeth Anne Jurica, Ximao Wu, Lauren E Haque, Kristin N. Williams, Andres S.
Hernandez, Zhenqiu Hong, Qi Gao, Marta Dabros, Akin H. Davulcu, Arvind Mathur, Richard A. Rampulla,
Arun Kumar Gupta, Ramya Jayaram, Atsu Apedo, Douglas B. Moore, Heng Liu, Lori K. Kunselman, Edward J
Brady, Jason J. Wilkes, Bradley A. Zinker, Hong Cai, Yue-Zhong Shu, Qin Sun, Elizabeth A. Dierks, Kimberly
A. Foster, Carrie Xu, Tao Wang, Reshma Panemangalore, Mary Ellen Cvijic, Chunshan Xie, Gary G. Cao,
Min Zhou, John Krupinski, Jean M Whaley, Jeffrey A Robl, William R Ewing, and Bruce Alan Ellsworth
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00982 • Publication Date (Web): 09 Jan 2018
Downloaded from http://pubs.acs.org on January 10, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Xie, Chunshan; Bristol-Myers Squibb
Cao, Gary; Bristol-Myers Squibb
Zhou, Min; Bristol-Myers Squibb Company, Discovery Chemistry
Krupinski, John; Bristol-Myers Squibb, Diabetes Drug Discovery
Whaley, Jean; Bristol-Myers Squibb Company, Discovery Chemistry
Robl, Jeffrey; Bristol-Myers Squibb Company, Discovery Chemistry
Ewing, William; Bristol-Myers Squibb, Discovery Chemistry
Ellsworth, Bruce; Bristol-Myers Squibb, Discovery Chemistry
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Discovery of Potent and Orally Bioavailable
Dihydropyrazole GPR40 Agonists
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Jun Shi,* Zhengxiang Gu, Elizabeth Anne Jurica, Ximao Wu, Lauren E. Haque, Kristin N.
Williams, Andres S. Hernandez, Zhengqiu Hong, Qi Gao, Marta Dabros, Akin H. Davulcu,
Arvind Mathur, Richard A. Rampulla, Arun Kumar Gupta, Ramya Jayaram, Atsu Apedo,
Douglas B. Moore, Heng Liu, Lori K. Kunselman, Edward J. Brady, Jason J. Wilkes, Bradley
A. Zinker, Hong Cai, Yue-Zhong Shu, Qin Sun, Elizabeth A. Dierks, Kimberly A. Foster, Carrie
Xu, Tao Wang, Reshma Panemangalore, Mary Ellen Cvijic, Chunshan Xie, Gary G. Cao, Min
Zhou, John Krupinski, Jean M. Whaley, Jeffrey A. Robl, William R. Ewing, Bruce Alan Ellsworth
Research and Development, BristolꢀMyers Squibb, Co., P.O. Box 4000, Princeton, New Jersey
08540ꢀ4000.
KEYWORDS: GPR40, FFAR1, GPCR agonist, diabetes therapy, dual MOA, GSIS, insulin
secretagogue, GLPꢀ1 secretagogue, BMSꢀ986118
Abstract
G proteinꢀcoupled receptor 40 (GPR40) has become an attractive target for the treatment of
diabetes since it was shown clinically to promote glucoseꢀstimulated insulin secretion. Herein,
we report our efforts to develop highly selective and potent GPR40 agonists with a dual
mechanism of action, promoting both glucoseꢀdependent insulin and incretin secretion.
ACS Paragon Plus Environment
1

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
2
Employing strategies to increase polarity and the ratio of sp /sp character of the chemotype, we
identified BMSꢀ986118 (compound 4), which showed potent and selective GPR40 agonist
activity in vitro. In vivo, compound 4 demonstrated insulinotropic efficacy and GLPꢀ1 secretory
effects resulting in improved glucose control in acute animal models.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
Type 2 diabetes, representing over 85% of diabetes cases, is a progressive disease characterized
by insulin resistance resulting in insufficient insulin action, eventually leading to a loss of
glucose homeostasis. The rapidly growing prevalence of type 2 diabetes mellitus (T2DM) is a
worldwide health concern due to its contribution towards cardiovascular coꢀmorbidities (e.g.
heart failure, stroke) and microvascular damage leading to blindness, kidney failure, foot ulcers,
1
neuropathathic pain and more. According to the WHO (World Health Organization), over 300
2
million people worldwide were estimated to have diabetes in 2013, and that number is projected
3
4
to rise to 629 million by 2045. Metformin is the firstꢀline treatment of type 2 diabetes. Other
glucose independent therapies for the treatment of diabetes include insulin secretagogues, such
2
,3,4
as sulfonylureas (lowering HbA1c by ~1.5%) and glinides (HbA1c lowering by ~1.5%).
While effective in the short term, chronic use of these treatments can lead to hypoglycemia and
βꢀcell exhaustion. These limitations often lead to disease progression, resulting in the need for
4
polypharmacology or transition to insulin therapies. As an alternative mechanism to treat
diabetes, incretin based therapies are appealing due to their ability to stimulate insulin in a
glucoseꢀdependent fashion (glucose–stimulated insulin secretion, GSIS). DPP4 inhibitors and
GLPꢀ1 mimetics are among these treatment options, and they are effective at lowering HbA1c
5
,6
with the benefit of a reduced risk for hypoglycemia. There remains potential opportunity for
additional incretinꢀbased therapies, since DPP4 inhibition exhibits modest HbA1c lowering
ACS Paragon Plus Environment
2

Journal of Medicinal Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
(
~0.7%) and GLPꢀ1 mimetics can have some safety/tolerability limitations. Peroxisome
proliferatorꢀactivated receptor γ (PPARγ) agonists increase insulin sensitivity and infrequently
result in hypoglycemia but concerns around their side effects (significant weight gain, and
7
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
possible cardiovascular and bladder cancer risks) have limited their clinical use. A promising
new class of treatments for diabetes, SGLTꢀ2 inhibitors, demonstrates insulinꢀindependent
glucose lowering while providing additional benefits such as moderate weight loss and bloodꢀ
pressure lowering; however, modest efficacy (lowering HbA1c ~0.5ꢀ0.9%) leaves opportunity
for other therapies that further reduce HbA1c without risk of hypoglycemia. Lastly, many
patients progress to the use of injected insulin, which is an effective therapy to manage T2DM,
4
but it must be carefully titrated to avoid incidence of hypoglycemia and weight gain.
G proteinꢀcoupled receptor 40 (GPR40), also known as free fatty acid receptor 1 (FFAR1), is a
G protein coupled receptor (GPCR) that is predominately expressed in pancreatic
β
ꢀcells and is
q
9
,10
also found in the GI tract and brain.
GPR40 is activated by free fatty acids (FFAs) such as
docosahexaenoic acid (DHA) and linoleic acid (LA). In pancreatic
β
ꢀcells, glucose metabolism
+
+
induces the closure of ATPꢀsensitive K channels in the plasma membrane, which reduces K
2
+
efflux and depolarizes the membrane, opening Ca channels. Activation of GPR40 causes
2+
increases in intracellular Ca concentrations via the inositol 1,4,5ꢀtrisphosphate (IP ) pathway,
3
2
+
and increased Ca stores leads to amplified insulin release after the glucoseꢀinduced membrane
1
1
depolarization. Since the insulinotropic effect is glucoseꢀdependent, GPR40 agonists have the
therapeutic potential to treat diabetes with a low risk of hypoglycemia. In addition, GPR40–null
mutant mice exhibit reduced GLPꢀ1 and glucoseꢀdependent insulintropic peptide (GIP) secretion
1
2
suggesting that GPR40 may be involved in FFAꢀmediated incretin secretion. Several GPR40
agonists have been reported to stimulate both GLPꢀ1 and GIP secretion in vivo, providing a
ACS Paragon Plus Environment
3

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13,14,15
second mechanism to control GSIS.
The constellation of GPR40ꢀmediated effects makes it
an attractive target for the treatment of typeꢀ2 diabetes.
Figure 1. Structures of dihydropyrazole and pyrrolidine GPR40 agonists
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Recently, we disclosed potent GPR40 agonists 1 and 2 (Figure 1), featuring a pyrrolidine ring
1
6,17
and a dihydropyrazole ring, respectively.
acids, and we found that ligands with 4 rings (AꢀD ring, as depicted in Figure 1 structures 1 and
) yielded optimal agonist activity. Both compounds exhibited potent GPR40 agonist activity
GPR40 is a receptor that binds longꢀchain fatty
2
and glucoseꢀlowering efficacy in several animal models of diabetes, however, each demonstrated
in vivo offꢀtarget liabilities that precluded their advancement: compound 1 caused hyperactivity
in rats and compound 2 induced undesired bloodꢀpressure and heart rate increases in rodents and
1
8
monkeys at high doses. Both compounds also demonstrated PPARγ activity in vitro. It is
known that PPARγ agonists improve insulin sensitization and this activity may provide
additional therapeutic benefit in combination with GPR40 agonists. However, since PPARγ
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
agonism may introduce increased safety risks, and the safety profile of GPR40 agonists has not
been fully deꢀrisked in clinical development, we decided that it would be prudent to develop a
7
GPR40 selective agonist that was devoid of PPARγ activity. The undesired hyperactive and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
cardiovascular effects of compounds 1 and 2 also caused concern for advancement of these
1
2
compounds, especially in a population that is prone to cardiovascular (CV) disease. We found
that the CV liabilities of 2 were due to offꢀtarget pharmacology since hemodynamic changes
1
7
were observed when compound 2 was administered to GPR40 knockout mice. Despite
extensive in vitro profiling efforts, we were unable to identify the origin of the undesired CV
1
7
effects. The rather high log P and associated lipophilicity of these compounds may have
contributed to the promiscuity of their actions in vivo. We therefore adopted a strategy to reduce
log P and fractional aromaticity of the ligands, believing that this approach would lead to
compounds that matched the spatial/electronic requirements for GPR40 agonism while being less
1
9,20
likely to match those of any unidentified targets.
Herein, we describe the identification of a series of pyrrolidine and dihydropyrazole
derivatives leading to compound 4, a novel GPR40 agonist with an improved potency,
pharmacokinetics, glucoseꢀlowering efficacy, and safety profile.
Chemistry
The synthesis of dihydropyrazole analogs is detailed in Scheme 1. The preparation of
dihydropyrazole phenol intermediate 14 began with acylation of hydrazine 5 with TFAA to
afford hydrazide 6. Treatment of 6 with benzenesulfonyl chloride and DIEA afforded a
hydrazonoyl chloride that was subjected to a [3+2] cycloaddition with alkene 7 providing
dihydropyrazole 8 with 2:1 diastereoselectivity (the stereochemistry of 8 was confirmed by Xꢀ
2
1
ray analysis). Intermediate 8 was homologated to ester 12 by a fourꢀstep sequence. First,
ACS Paragon Plus Environment
5

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intermediate 8 was reduced with NaBH in THF and H O to give alcohol 9. The resulting alcohol
4
2
was converted to mesylate 10 via reaction with MsCl and Et N in DCM at room temperature.
3
Displacement of 10 via treatment with potassium cyanide in DMSO at 40 ºC generated nitrile 11
that was subsequently hydrolyzed to the corresponding methyl ester 12 using 3.6 M HCl in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
22
MeOH at room temperature. Miyaura borylation of 12 afforded boronate 13 that was subjected
to peroxideꢀmediated (H O ) oxidation to produce key intermediate, phenol 14. Synthesis of Dꢀ
2
2
ring analogs (see Figure 1 for ring nomenclature) was accomplished via coupling between 4ꢀ
st
hydroxyl piperidine 15 and various aryl bromides using 1 generation SꢀPhos precatalyst and
LiHMDS in THF at 60 °C. Tosylation of alcohols 16a-ac with TsCl and pyridine afforded the
corresponding tosylates 17a-ac. Displacement of the tosyl group with dihydropyrazole phenol 14
under basic conditions generated substituted methyl esters 18a-ac. Finally, hydrolysis of esters
1
8a-ac furnished the desired dihydropyrazole analogs 3a and 19a-ac with different
aromatic/heteroaromatic Dꢀrings.
a
Scheme 1. Synthesis of dihydropyrazole Dꢀring analogs
A. Synthesis of dihydropyrazole AꢀB ring
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: a) TFAA, DCM, 0 ºC to rt, 81%; b) benzenesulfonyl chloride, DIEA,
DCM, 0 ºC to rt, 90%; c) Ag CO , dioxane, 65 ºC, 46%; d) NaBH , THF/H O, rt, 94%; e) MsCl,
2
3
4
2
Et N, DCM, rt, 98%; f) KCN, DMSO, 40 ºC, 92%; g) HCl (3.6 M in MeOH), rt, 82%; h)
3
Pd(dppf)Cl , bis(pinacolato)diboron, KOAc, DMF, 80 ºC, 90%; i) 30% aq. H O , EtOAc, rt,
2
2
2
8
5%;
a
B. Synthesis of Dꢀring analogs
ACS Paragon Plus Environment
7

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ar =
1
1
1
6a Ph
16p 2ꢀFꢀ5ꢀiPrꢀPh
16q 2ꢀFꢀ5ꢀClꢀPh
16r 2ꢀClꢀ5ꢀFꢀPh
16s 2,5ꢀdiꢀClꢀPh
16t 2,5ꢀFꢀPh
6b 2ꢀClꢀPh
6c 2ꢀMeꢀPh
16d 2ꢀEtꢀPh
1
1
1
1
6e 3ꢀMeOꢀPh
6f 4ꢀClꢀPh
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16u 2ꢀClꢀ5ꢀMeOꢀPh
16v 2ꢀClꢀ5ꢀMeꢀPh
16w 2ꢀCNꢀ5ꢀMeOꢀPh
16x 2ꢀPy
6g 4ꢀFꢀPh
6h 3ꢀFꢀ4ꢀMeOꢀPh
16i 2ꢀFꢀ6ꢀMeOꢀPh
1
1
1
1
6j 2ꢀFꢀ6ꢀMeꢀPh
6k 2,6ꢀdiꢀMeꢀPh
16y 6ꢀFꢀ2ꢀPy
16z 3ꢀMeꢀ6ꢀFꢀ2ꢀPy
6l 2,6ꢀdiꢀMeꢀ4ꢀMeOꢀPh 16aa 3ꢀPy
6m 2ꢀFꢀ5ꢀEtOꢀPh 16ab 2ꢀFꢀ5ꢀMeOꢀ4ꢀPy
16ac 2ꢀClꢀ5ꢀMeOꢀ4ꢀPy
16n 2ꢀFꢀ5ꢀMeꢀPh
1
6o 2ꢀFꢀ5ꢀEtꢀPh
a
st
Reagents and conditions: j) ArBr, 1 generation SꢀPhos precat., LiHMDS, THF, 60 ºC, 30%ꢀ
70%; k) TsCl, pyridine, DMAP, rt, DCM, 55%ꢀ80%; l) 14, Cs CO , DMF, 60 ºC, 30%ꢀ60%; m)
LiOH, H O, THF, rt, 70%ꢀ95%.
2
3
2
The synthesis of compounds where the Cꢀring (see Figure 1) is substituted is depicted in
Scheme 2. In order to introduce a 3ꢀsubstituted piperidine ring (Scheme 2), the synthesis began
with commercially available racemic 3ꢀfluoroꢀpiperidinꢀ4ꢀone (20a) or 3ꢀmethylꢀsubstituted
piperidinꢀ4ꢀone (20b). LꢀSelectrideꢀmediated reduction of ketones 20a-b (with
diastereoselectivity over 20:1) followed by TBS protection (using TBSꢀOTf) of the resulting cisꢀ
alcohol afforded intermediates 21a-b. Hydrogenolysis of the benzyl group using 10% palladium
on carbon under 1 atm H provided amines of structure 22a-b. Subsequent SN reaction of the
2
Ar
resulting amines 22a-b with 4ꢀbromoꢀ5ꢀchloroꢀ2ꢀmethoxypyridine and K CO in DMSO at 100
2
3
°
C gave a racemic adduct. At this point in the synthesis, the resulting racemic alcohols were
subjected to chiral separation to give chiral alcohols 23a-b. Next, displacement of the hydroxyl
group in intermediates 23a-b with phenol 14 was accomplished by Mitsunobu reaction using
Bu P and ADDP in toluene at elevated temperature (50 °C). The resulting methyl esters 24a-b
3
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were hydrolyzed using LiOH in THF and H O at room temperature to generate the
2
corresponding fluoro (25) and methyl (4) analogs.
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of Cꢀring analogs
a
Reagents and conditions: a) LꢀSelectride, THF, ꢀ78 ºC, 88%ꢀ98%; b) TBSꢀOTf, Et N, DCM, 0
3
ºC to rt, 76%ꢀ92%; c) H (1 atm), Pd/C, MeOH, rt, 97%ꢀ100%; d) K CO , 100 ºC, DMSO,
2
2
3
TBAF, THF, rt; followed by chiral separation, 30ꢀ32%; e) Bu P, ADDP, toluene, rt, 14, 54ꢀ
3
6
4%; f) LiOH, H O, THF, rt, 78ꢀ80%.
2
Results and Discussion
Although the pharmacological origins of blood pressure and heart rate elevations observed with
compound 2 (as well as hyperactivity associated with compound 1) were unclear, we envisioned
3
2
that increasing polarity and the sp /sp carbon ratio of our ligands may be beneficial in reducing
19,20
offꢀtarget activities.
Previous SARs demonstrated that the Cꢀring was promiscuous for
1
6
GPR40 activity, so we decided to focus on replacing this portion of the molecule with a series
of saturated heterocycles that were designed to maintain the distance and constrained nature of
ACS Paragon Plus Environment
9

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the phenyl Cꢀring (Structure A). As shown in Table 1, pyrrolidine 26 (see Supplemental
Information for the synthesis of pyrrolidine analogs) demonstrated moderate but encouraging
1
6,17
hGPR40 agonist activity. Following our previously described SAR for the series,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
incorporation of a 4ꢀ(R)ꢀCF substituent to the pyrrolidine ring led to significantly (7ꢀfold)
3
improved in vitro potency (e.g. 3b) in this oxygenated piperidine Cꢀring series. Carbonꢀlinked
analogs (e.g. 27) were less potent than the oxygenꢀlinked analogs (e.g. 3b) which was surprising
given that carbonꢀlinked analogs were more potent when the Cꢀring was phenyl (structures 1 and
2
). To determine if the pyrrolidine Cꢀring unit was optimal, pyrrolidine (28-31) analogs
(generated following the same general chemical protocols as outlined in Scheme 1) were tested,
but their GPR40 agonist activities were inferior to the oxygenated piperidine Cꢀring analog.
When the Cꢀring piperidine was incorporated into the dihydropyrazole series (Table 2), we
discovered lead 3a which exhibited not only potent hGPR40 activity but was devoid of PPARγ
activity as well (3a: PPARγ EC > 50 µM, 1: PPARγ EC = 4.4 µM and 2: PPARγ EC = 2.6
5
0
50
50
µM ). Unfortunately the electron donating nature of the piperidine nitrogen affected the
metabolic fate of the Dꢀring, rendering it less stable to oxidative metabolism and glutathione
(GSH) addition in liver microsomes. Compound 3a also demonstrated increased (vide infra,
Table 5) clearance in rodents due to demethylation of the methoxy group. We therefore set out to
identify a compound with an improved PK profile and a reduced propensity for GSH adducts
while maintaining the potency and selectivity profile of 3a.
Table 1. CꢀRing SAR with a pyrrolidine A ring
ACS Paragon Plus Environment
10

Journal of Medicinal Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FLIPR hGPR40
CLogP
Compound
Cꢀring linker
Y
a,b
EC (µM)
5
0
26
3b
2.49
4.55
H
1.20, 1.60
0.21±0.06
0.53±0.14
2.83
CF3
CF3
CF3
2
2
2
7
8
9
5.73
4.48
CF3
5.36
4.48
4.60
4.60
3
3
0
1
CF3
CF3
0.94
0.63
a
S.D.: standard deviation, provided for results of n>3 tests; other EC ’s reported as the results
5
0
of each separate test occasion.
b
all compounds showed full activation as compared to compound 2 and DHA in this assay.(See
SI for detailed information)
ACS Paragon Plus Environment
11

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Since we identified the Dꢀring as the primary site of metabolism of 3a (vide infra, Figure 2),
emphasis was placed on exploration of this portion of the molecule, focusing first on monoꢀ
substituted Dꢀring analogs (Table 2). Compared to unsubstituted analog 19a, orthoꢀsubstitution
on the Dꢀring (19b-d) resulted in improved GPR40 activity, in contrast to paraꢀsubstitution (19f,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9g) which decreased activity. Both oꢀchloro (19b) and mꢀmethoxy (19e) substituents exhibited
a 4ꢀfold improved activity, indicating the benefit of having a chloroꢀsubstituent at the orthoꢀ
position and a methoxyꢀsubstituent at the metaꢀposition relative to the piperidine nitrogen
attachment. A survey of diꢀ and triꢀsubstituted Dꢀring analogs revealed that 2,6ꢀ (19i-j), 3,4ꢀ
(19h) and 2,4,6ꢀsubstituted analogs (19l) were not in the desired potency range. Interestingly,
2,5ꢀdisubstitutions (19m-w) showed a positive effect on agonist activity and this substitution
pattern was deemed to be optimal in this series.
Unfortunately the Dꢀring substituents did not adequately address metabolism concerns.
However, electron deficient heterocycles were introduced in an effort to reduce the potential for
Dꢀring GSH adduct formation. As shown in Table 2, replacing the Dꢀring phenyl, as in
compound 19a, pyridine analogs (19y, 19z, 19ab and 19ac) showed a tendency towards
improved potency. Surprisingly, the 2ꢀchloroꢀ5ꢀmethoxyꢀ4ꢀpyridine analog (19ac) maintained
similar in vitro potency as the lead compound 3a despite different electronic characteristics of
the Dꢀring.
Since the piperidine Cꢀring seemed to provide preferred spatial placement of the B and D
rings, we decided to evaluate conformational constraints in the piperidine group in an attempt to
further increase potency. 3ꢀSubstituted piperidines were synthesized as shown in Table 3. The
introduction of fluorine (32, 25) or/and methyl (33-35, 4) group on the Cꢀring provided a
significant improvement in EC values. Methyl substitution (4) improves potency by ~2 fold as
5
0
ACS Paragon Plus Environment
12

Journal of Medicinal Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compared to 3a. (Analogs are counterscreened against a proprietary panel of GPCRs (including
the related FFARs), NHRs, transporters, ion channels, and hERG. Compound 4 had no
detectable activities below 10 µM against any counterscreen with the exception of MAO B (IC50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
=
6 µM).) With several analogs in the desired potency range in hand, biotransformation studies
were initiated as a means to further differentiate between preferred compounds.
Table 2. In Vitro Activities of substituted Dꢀring Analogs
FLIPR hGPR40
Compound
Dꢀring
CLogP
a
EC (µM)
5
0
1
9a
3a
9b
Ph
1.93
4.22
5.39
6.02
4.63
5.12
4.73
5.34
4.77
4.95
5.39
2ꢀFꢀ5ꢀMeOꢀPh
2ꢀClꢀPh
0
.11±0.04
1
0.45
0.50
1.18
0.62
6.27
1
9c
2ꢀMeꢀPh
19d
2ꢀEtꢀPh
1
9e
9f
19g
9h
9i
3ꢀMeOꢀPh
4ꢀClꢀPh
1
4ꢀFꢀPh
2
.73, 2.55
.62
5.90
1
3ꢀFꢀ4ꢀMeOꢀPh
3
2
ꢀFꢀ6ꢀMeOꢀPh
1
ACS Paragon Plus Environment
13

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
ꢀFꢀ6ꢀMeꢀPh
,6ꢀdiꢀMeꢀPh
2,6ꢀdiꢀMeꢀ4ꢀMeOꢀPh
ꢀFꢀ5ꢀEtOꢀPh
ꢀFꢀ5ꢀMeꢀPh
2ꢀFꢀ5ꢀEtꢀPh
ꢀFꢀ5ꢀiPrꢀPh
1
9j
19k
9l
9m
9n
0.80
1.34
6.06
5.02
4.49
5.75
6.06
6.51
6.79
6.16
6.16
6.63
5.68
5.87
6.44
5.10
3.20
4.34
4.02
3.07
2
1
3.06
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
1
0.22
2
1
0.44
1
9o
9p
9q
0.45, 0.72
0.60
2
1
2
ꢀFꢀ5ꢀClꢀPh
1
0.51
2
ꢀClꢀ5ꢀFꢀPh
19r
9s
9t
0.36
2
,5ꢀdiꢀClꢀPh
,5ꢀdiꢀFꢀPh
ꢀClꢀ5ꢀMeOꢀPh
1
0.25
2
1
0.48
2
19u
19v
0.10
2ꢀClꢀ5ꢀMeꢀPh
0
.36
0.68
.61
2
ꢀCNꢀ5ꢀMeOꢀPh
1
9w
b
1
1
9x
2ꢀPyridyl
2
6ꢀFꢀ2ꢀPyridyl
9y
0.39, 0.86
0.37
3
ꢀMeꢀ6ꢀFꢀ2ꢀPyridyl
ꢀPyridyl
2ꢀFꢀ5ꢀMeOꢀ4ꢀPyridyl
ꢀClꢀ5ꢀMeOꢀ4ꢀPyridyl
1
9z
9aa
9ab
9ac
3
1
7.68
1
0.15, 0.18
0.11
4.21
4.27
2
1
a
S.D.: standard deviation, provided for results of n>3 tests; other EC ’s reported as the results of
50
each separate test occasion.
ACS Paragon Plus Environment
14

Journal of Medicinal Chemistry
Page 16 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Reꢀexploration of Cꢀring SAR
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FLIPR hGPR40
CLogP
Compound
Cꢀring
Dꢀring
a
EC (µM)
5
0
5.61
32
2ꢀFꢀ5ꢀMeOꢀPh
0
0
0
.07±0.15
4.50
2
5
3
2ꢀClꢀ5ꢀMeOꢀPyridyl
2ꢀFꢀ5ꢀEtOꢀPh
0.18
6.23
5.07
5.13
3
.15±0.08
34
2ꢀFꢀ5ꢀEtOꢀ4ꢀPyridyl
2ꢀClꢀ5ꢀEtOꢀ4ꢀPyridyl
0.28
3
5
.25, 0.19
2ꢀClꢀ5ꢀMeOꢀ4ꢀ
Pyridyl
4
0.07±0.04
4.76
ACS Paragon Plus Environment
15

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2ꢀClꢀ5ꢀMeOꢀ4ꢀ
Pyridyl
4
2
0
.32
4.76
a
S.D.: standard deviation, provided for results of n>3 tests; other EC ’s reported as the results
5
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
of each separate test occasion.
Biotransformation
In vitro biotransformation studies in liver microsomes on three advanced compounds (3a, 33
and 4) highlighted significant differences with respect to metabolic stability and GSH adduct
formation. The proposed pathways that lead to the formation of deꢀalkylated and oxidized
metabolites and glutathione conjugates are outlined in Figure 2. Compounds 3a, 33 and 4
underwent CYP450ꢀmediated oxidative deꢀalkylation to form phenolꢀtype metabolites (36, 37
and 38, respectively) and oxidative GSH addition forms GSH adducts (39 and 40).As shown in
Table 4, compound 3a exhibited high amounts of deꢀmethylated and GSHꢀconjugated
metabolites both in rat liver microsomes (RLM) and human liver microsomes (HLM) while
compounds 33 and 4 showed a much lower percentage of the corresponding deꢀalkylated/GSHꢀ
adducts in both (human and mouse) microsomal preparations. We attribute the lack of GSH
adducts to the introduction of an electron deficient pyridine (4) vs phenyl (3a) Dꢀring. Since high
23
levels of GSH adducts are associated with an increased risk of idiosyncratic drug reactions, the
metabolism profiles for 33 and 4 proved especially desirable. As a result of their improved
metabolic profiles and potent GPR40 agonist activities, we selected these compounds for
advanced in vivo evaluations.
Figure 2. Formation of oxidation metabolites and related GSH conjugates
ACS Paragon Plus Environment
16

Journal of Medicinal Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 4. Metabolic fate of compounds 3a, 33 and 4 in liver microsome incubations
Human liver microsomes
Rat liver microsomes
%
parent
% Dꢀring
% GSH
T1/2
% parent
% Dꢀring
% GSH
T1/2
compound Compound dealkylation adducts
remaining metabolites
(min)
Compound dealkylation adducts
remaining metabolites
(min)
3
a
42
93
89
65
1.5
6.4
4
48
95
27
79
75
36
15
16
3ꢀ14
<1
23
37
81
33
4
<1
ND
102
ND
a
The incubation mixtures contained test compound (1 or 3 µM), microsomal protein (1 mg/mL),
NADPH (1 mM), phosphate buffer (100 mM, pH 7.4), and magnesium chloride (6.7 mM). The
reaction, conducted in triplicate, was initiated by the addition of NADPH followed by incubation
at 37°C for 20 minutes. (See Supporting Information for more detailed experimental procedure.)
Pharmacokinetics
The pharmacokinetic (PK) profiles of 3a, 4 and 33 across several species are shown in Table 5.
As predicted by their improved microsomal stability profiles, compounds 4 and 33 showed
improved PK profiles as compared to 3a. Compounds 4 and 33 shared similar PK profiles with
low clearance, moderate volume of distribution (V ) and moderate halfꢀlife in several preclinical
ss
ACS Paragon Plus Environment
17

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
species. However, in cynomologous monkeys, 33 showed higher clearance and Vss, and shorter
halfꢀlife than 4, introducing some uncertainty over the projected human clearance of 33 relative
to 4. As such, compound 4 was selected for further advancement due to its superior crossꢀspecies
PK profile.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 5. Pharmacokinetic properties of compounds 3a, 4 and 33 in mouse, rat, dog and monkey
mouse
rat
33
monkey
dog
PK parameters
4
33
3a
4
3a
4
33
3a
4
33
dose iv/po (mg/kg) 1/1 1/ꢀ 1/3
1/3
1/3
3.2
1.7
3.4
5.0
0.8
89
1/3 0.5/1 0.5/ꢀ 0.2/0.5 0.5/1 0.5/ꢀ
*
Cmax (µM)
Tmax (h)
T (h)
6.0 n.d. 3.4 3.5
1.1
0.7
1.4
17
1.4
5
n.d.
n.d.
9.4
0.2
2
1.2
0.8
5.2
2.2
0.6
62
n.d.
n.d.
7.3
1
n.d.
3.1 3.3 1.5 4.0
2.0 1.6 3.4
0.4 0.7 0.8 0.3
100 n.d. 96 47
1
1.3
13
0.7
0.7
61
5.3
6.6
1.7
66
1
/2
Cl (mL/min/kg)
6
4.0
2.6
V (L/kg)
ss
0.6
36
2.2
0.6
F%
n.d.
n.d.
*
n.d. = not determined
Pharmacodynamics
To assess the pharmacologcial effects of 4, we performed an intraperitoneal glucose tolerance
test (IPGTT) in 5 h fasted male mice, where compound (0.03 to 1 mg/kg) was administered
orally 1 h prior to an intraperitoneal glucose challenge. As shown in Figure 3, compound 4 doseꢀ
dependently suppressed plasma glucose excursion similar to that of our earlier advanced lead 1.
Administration of compound 4 as low as 0.1 mg/kg produced a glucoseꢀlowering effect.
ACS Paragon Plus Environment
18

Journal of Medicinal Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Hypoglycemia was not observed at higher doses, consistent with the glucoseꢀdependent nature of
GPR40 action.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Glucose lowering effects of compound 4 in mice during an intraperitoneal glucose
tolerance test (IPGTT)
Plasma Glucose Levels
600
5
4
3
00
00
00
vehicle
IPGTT
Cmpd 4, 0.03 mg/kg
284 .03mpk
2 C8 4m .p1 dm4 p, k0 .1 mg/kg
*
*
**
*
2
C8 4m .p3 dm4 p, k0 .3 mg/kg
C8 4m 1p dm 4p ,k1 mg/kg
9 C7 5m 7 p3 d1 1, m1 pk mg/kg
*
** *
200
*
**
2
***
100
0
-60 -40 -20 0 20 40 60 80 100120
Time (min)
*, *** P<0.05, 0.001 vs. veh.
To assess the ability of 4 to elicit GLPꢀ1 secretion in vivo, an important part of the desired dual
mechanism of action, we performed an oral glucose tolerance test (OGTT) in 5 h fasted mice in
2
4
the presence and absence of the dipeptidylpeptidaseꢀ4 (DPP4) inhibitor BMSꢀ538305. As GLPꢀ
is known to be rapidly cleaved in plasma, DPP4 inhibition is required to preserve the effect on
1
compound induced active GLPꢀ1 secretion. Any changes in active plasma GLP secretion upon
coꢀadministration may also suggest that pairing a GPR40 agonist with a DPP4 inhibitor could
result in clinically relevant benefits on glucose homeostasis. As shown in Figure 4a and 4c, an
ACS Paragon Plus Environment
19

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
oral dose of 3 mg/kg of compound 4 alone increased insulin levels and reduced plasma glucose
excursion significantly but had a minimal effect on active GLPꢀ1 levels (Figure 4b) due to rapid
enzymatic proteolysis via DPP4. In combination with a DPP4 inhibitor, both active GLPꢀ1 was
increased and plasma glucose excursion was attenuated to an extent greater than either
monotherapies, prior toꢀ and after the OGTT. Our results indicate that compound 4 not only
enhances GSIS but also it promotes GLPꢀ1 secretion in vivo. In summary, compound 4 showed
significant glucose lowering effects via increased insulin and GLPꢀ1 levels, demonstrating a
desired dual mechanism of action in vivo.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4a. Plasma glucose levels of compound 4 in mice during an oral glucose tolerance test
(OGTT)
kg
Figure 4b. Plasma active GLPꢀ1 levels of compound 4 in mice during an OGTT
ACS Paragon Plus Environment
20

Journal of Medicinal Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4c. Plasma insulin levels of compound 4 in mice during an OGTT
mg/kg
mg/kg
Conclusions
ACS Paragon Plus Environment
21

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
With the aim of developing highly selective and potent GPR40 agonists with favorable ADME
3
2
properties, we employed strategies to increase polarity and the sp /sp carbon ratio, and to reduce
undesired compound metabolism, leading to the identification of compound 4. This compound
showed potent and selective GPR40 full agonist activity with dual insulinotropic and GLPꢀ1
secretory effects, resulting in robust plasma glucose lowering effects in acute animal models.
Compound 4 also exhibited a favorable PK profile across preclinical species. The incorporation
of a Cꢀring piperidine group paired with an optimized pyridyl Dꢀring functionality reduced GSH
adduct formation, providing a path to compounds with reduced potential for idiosyncratic
adverse reactions or offꢀtarget PPARγ activity. Based on its highly desirable drug developability
profile, compound 4 was selected as a candidate for clinical development. Additional data on this
compound will be reported in due course.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental Section
Chemistry. All anhydrous reactions were carried out using oven dried glassware under an
atmosphere of argon or nitrogen. All reagents and solvents were obtained from commercial
1
13
vendors and used without further purification unless otherwise indicated. NMR spectra ( H, C)
were recorded on JEOL JNMꢀECP500, JEOL GSX400, and Bruker 400 spectrometers. Chemical
shifts are given in parts per million (ppm) downfield from internal reference tetramethylsilane
standard; coupling constants (J values) were given in hertz (Hz). Elemental analyses were
performed by the BristolꢀMyers Squibb Discovery Analytical Science department. Melting
points were taken on a Hoover Unimelt melting point apparatus and were uncorrected. LC/MS
data were recorded on a Shimadzu LCꢀ10AT equipped with a SILꢀ10A injector, a SPDꢀ10AV
ACS Paragon Plus Environment
22

Journal of Medicinal Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
detector normally operating at 220 nm, and interfaced to a Micromass ZMD mass spectrometer.
LC/MS or HPLC retention times were reported using a Phenomenex Luna Cꢀ18 4.6 mm × 50
mm column eluting with a 4 min gradient of 0−100% solvent B, where solvent A was 10:90:0.1
CH OH−H O−TFA and solvent B was 90:10:0.1 CH OH−H O−TFA. Reactions were monitored
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
2
3
2
by TLC using 0.25 mm E. Merck silica gel plates (60 F254) and were visualized using UV light
and 5% phosphomolybdic acid in 95% EtOH. All compounds were found to be >95% pure by
HPLC analysis unless otherwise noted.
General procedure A - Compound 3a :
1
-(2-Fluoro-5-methoxyphenyl)piperidin-4-ol. A roundꢀbottomed flask containing a mixture of
2
ꢀbromoꢀ1ꢀfluoroꢀ4ꢀmethoxybenzene (650 mg, 3.2 mmol) and piperidinꢀ4ꢀol (800 mg, 7.9
st
mmol) in THF (3.2 mL) was purged with argon. 1 generation SꢀPhos precat. ((2ꢀ
Dicyclohexylphosphinoꢀ2′,6′ꢀdimethoxyꢀ1,1′ꢀbiphenyl)[2ꢀ(2ꢀaminoethylphenyl)]palladium(II)
chloride) (48 mg, 0.063 mmol) was added to the vessel followed by lithium
bis(trimethylsilyl)amide (1 N in THF, 15 mL, 15 mmol). The flask was purged with argon at
room temperature for several min and then the reaction mixture was heated at 70 ºC for 2.5 h.
The resulting solution was diluted with aqueous NaHCO and EtOAc, and the layers were
3
extracted. The organic extract was washed successively with H O and brine, and the organic
2
layer was dried (MgSO ) and concentrated. Purification of the residue via silica gel
4
1
chromatography gave the desired product as an oil (330 mg, 46% yield). H NMR (400 MHz,
CDCl ) δ ppm 6.92 (1 H, dd, J=12.09, 8.79 Hz), 6.51 (1 H, dd, J=7.15, 3.30 Hz), 6.40 (1 H, dt,
3
J=8.79, 3.02 Hz), 3.85 (1 H, td, J=8.52, 4.40 Hz), 3.76 (3 H, s), 3.27 ꢀ 3.46 (2 H, m), 2.74 ꢀ 2.91
(2 H, m), 1.95 ꢀ 2.14 (2 H, m), 1.75 (2 H, dtd, J=12.85, 9.10, 9.10, 3.57 Hz), 1.45 (1 H, d,
+
J=4.40 Hz). m/z obs 226.2 [M + H] .
ACS Paragon Plus Environment
23

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
-(2-Fluoro-5-methoxyphenyl)piperidin-4-yl 4-methylbenzenesulfonate. To a solution of 1ꢀ
(
2ꢀfluoroꢀ5ꢀmethoxyphenyl)piperidinꢀ4ꢀol (1.0 g, 4.4 mmol) and 4ꢀmethylbenzeneꢀ1ꢀsulfonyl
chloride (1.7 g, 8.9 mmol) in CH Cl (10 mL) at room temperature, was dropwise added
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pyridine (3.5 g, 44 mmol). The reaction mixture was stirred at room temperature for 16 h. The
reaction mixture was diluted with EtOAc, and the organic layer was washed successively with
H O and brine. The resulting organic layer was dried (MgSO ) and concentrated. Purification of
2
4
the residue via silica gel chromatography gave the desired product (light yellow solid, 1.5 g, 87%
1
yield). H NMR (400 MHz, CDCl ) δ 7.91 ꢀ 7.76 (m, J=8.2 Hz, 2H), 7.40 ꢀ 7.30 (m, J=8.2 Hz,
3
2H), 6.91 (dd, J=12.1, 8.8 Hz, 1H), 6.52 ꢀ 6.33 (m, 2H), 4.79 ꢀ 4.58 (m, 1H), 3.75 (s, 3H), 3.22
(
ddd, J=11.8, 7.4, 3.8 Hz, 2H), 2.91 (ddd, J=11.8, 7.4, 3.8 Hz, 2H), 2.46 (s, 3H), 2.05 ꢀ 1.84 (m,
+
4
H). m/z obs 380.1 [M + H] .
Methyl
2-((4S,5S)-1-(4-((1-(2-fluoro-5-methoxyphenyl)piperidin-4-yl)oxy)phenyl)-4-
methyl-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazol-5-yl)acetate. To a solution of 1ꢀ(2ꢀ
fluoroꢀ5ꢀmethoxyphenyl)piperidinꢀ4ꢀyl 4ꢀmethylbenzenesulfonate (1.22 g, 3.0 mmol) and 14
(
0.94 g, 3.0 mmol) in DMF (7 mL) was added cesium carbonate (2.2 g, 6.8 mmol). The reaction
mixture was heated at 60 ºC for 16 h. The reaction mixture was diluted with EtOAc, washed with
H O, brine, dried (MgSO ) and concentrated. Purification via silica gel chromatography gave the
2
4
1
desired product as a gum (660 mg, 42% yield). H NMR (400 MHz, CDCl ) δ 7.12 ꢀ 6.99 (m,
3
2
H), 6.99 ꢀ 6.83 (m, 3H), 6.53 (dd, J=7.1, 3.3 Hz, 1H), 6.41 (dt, J=8.8, 3.3 Hz, 1H), 4.38 (tt,
J=7.1, 3.6 Hz, 2H), 3.81 ꢀ 3.73 (m, 3H), 3.72 ꢀ 3.67 (m, 3H), 3.33 (ddd, J=11.4, 7.6, 3.6 Hz,
2
2
H), 3.26 ꢀ 3.12 (m, 1H), 2.97 (ddd, J=11.7, 8.1, 3.3 Hz, 2H), 2.80 (dd, J=15.9, 3.3 Hz, 1H),
.49 ꢀ 2.33 (m, 1H), 2.19 ꢀ 2.04 (m, 2H), 2.04 ꢀ 1.87 (m, 2H), 1.34 (d, J=6.6 Hz, 3H). m/z obs
+
524.2 [M + H] .
ACS Paragon Plus Environment
24

Journal of Medicinal Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
-((4S,5S)-1-(4-((1-(2-Fluoro-5-methoxyphenyl)piperidin-4-yl)oxy)phenyl)-4-methyl-3-
(
trifluoromethyl)-4,5-dihydro-1H-pyrazol-5-yl)acetic acid (3a). To a solution of methyl 2ꢀ
((4S,5S)ꢀ1ꢀ(4ꢀ((1ꢀ(2ꢀfluoroꢀ5ꢀmethoxyphenyl)piperidinꢀ4ꢀyl)oxy)phenyl)ꢀ4ꢀmethylꢀ3ꢀ
trifluoromethyl)ꢀ4,5ꢀdihydroꢀ1Hꢀpyrazolꢀ5ꢀyl)acetate (530 mg, 1.0 mmol) in THF (20 mL) was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
added lithium hydroxide (1N in water, 1.4 mL, 1.4 mmol) and the mixture was stirred at room
temperature. After 16 h, lithium hydroxide (1 N, 2.6 mL, 2.6 mmol) was added and the reaction
mixture was stirred at room temperature for 3 h. The reaction mixture was cooled with an iceꢀ
water bath and the solution was acidified with 1 N HCl. The mixture was extracted with CH Cl .
2
2
The organic extracts were dried (MgSO ), filtered and concentrated, and the residue was purified
4
via reverse phaseꢀPrep HPLC. The productꢀcontaining fractions were evaporated, and the residue
was treated with CH CN and 1 N HCl, and the mixture was lyophilized to give 3a as a light
3
1
yellow solid (480 mg, 84% yield). H NMR (500 MHz, CD CN) δ ppm 7.13 ꢀ 7.07 (m, 2H), 7.07
3
ꢀ 7.03 (m, 1H), 7.01 ꢀ 6.96 (m, 2H), 6.74 (dd, J=7.2, 3.0 Hz, 1H), 6.61 (dt, J=8.9, 3.3 Hz, 1H),
5
.67 (br.s, 1H), 4.53 ꢀ 4.47 (m, 1H), 4.47 ꢀ 4.42 (m, 1H), 3.79 (s, 3H), 3.46 (ddd, J=11.6, 7.6, 3.6
Hz, 2H), 3.40 ꢀ 3.32 (m, 1H), 3.11 (ddd, J=11.9, 8.2, 3.3 Hz, 2H), 2.75 (dd, J=16.5, 3.0 Hz, 1H),
2.52 (dd, J=16.5, 9.6 Hz, 1H), 2.21 ꢀ 2.13 (m, 2H), 2.00 (s, 1H), 1.96 ꢀ 1.89 (m, 2H), 1.33 (d,
+
J=7.2 Hz, 3H). m/z obs 510.2 [M + H] .
General procedure B - Compound 4
2
-((4S,5S)-1-(4-(((3R,4R)-1-(5-Chloro-2-methoxypyridin-4-yl)-3-methylpiperidin-4-
yl)oxy)phenyl)-4-methyl-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazol-5-yl)acetic acid (4). To
a stirred solution of methyl 2ꢀ((4S,5S)ꢀ1ꢀ(4ꢀ(((3R,4R)ꢀ1ꢀ(5ꢀchloroꢀ2ꢀmethoxypyridinꢀ4ꢀyl)ꢀ3ꢀ
methylpiperidinꢀ4ꢀyl)oxy)phenyl)ꢀ4ꢀmethylꢀ3ꢀ(trifluoromethyl)ꢀ4,5ꢀdihydroꢀ1Hꢀpyrazolꢀ5ꢀ
ACS Paragon Plus Environment
25

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
yl)acetate (5.5 g, 9.9 mmol) in THF (90 mL) and water (9 mL) at room temperature was added
2
N LiOH solution (12 mL, 24 mmol). The reaction mixture was stirred at room temperature for
16 h and 1N HCl (25 mL, 25 mmol) was added at 0 °C to pH = 4ꢀ5. The solvent was evaporated
and the residue was extracted 3X with EtOAc. The organic extracts were dried over Na SO ; the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
solution was filtered and concentrated. The residue was recrystallized from isopropanol to give 4
1
(
neutral form) as white solid (4.3 g, 7.7 mmol, 78% yield). H NMR (500 MHz, DMSOꢀd6) δ
ppm 12.52 (br. s., 1H), 8.01 (s, 1H), 7.05 (d, J=9.1 Hz, 2H), 6.96 (d, J=9.1 Hz, 2H), 6.40 (s,
1
H), 4.49 ꢀ 4.33 (m, 1H), 4.02 (td, J=8.8, 4.1 Hz, 1H), 3.80 (s, 3H), 3.56 ꢀ 3.39 (m, 2H), 3.37 ꢀ
.29 (m, 1H), 2.94 ꢀ 2.85 (m, 1H), 2.72 ꢀ 2.66 (m, 1H), 2.64 (dd, J=16.1, 2.9 Hz, 1H), 2.49 ꢀ 2.41
3
(m, 1H), 2.22 ꢀ 2.05 (m, 1H), 2.01 ꢀ 1.86 (m, 1H), 1.68 ꢀ 1.50 (m, 1H), 1.25 (d, J=7.2 Hz, 3H),
13
1
.03 (d, J=6.9 Hz, 3H). C NMR (126MHz, DMSOꢀd6) δ 171.5, 163.7, 157.1, 152.5, 146.3,
39.7 (q, J=34.7 Hz), 136.2, 121.7 (q, J=269.3 Hz), 117.3, 117.2, 116.0, 100.4, 78.9, 65.6, 54.2,
1
1
9
53.4, 47.8, 44.2, 36.0, 34.9, 29.5, 17.4, 15.3. F NMR (471MHz, DMSOꢀd6) δ ꢀ61.94 (s, 3F).
2
0
+
Optical rotation: [α]D −11.44 (c 2.01, MeOH). HRMS (ESI/HESI) m/z: [M + H] Calcd for
C H ClF N O 541.1824; Found 541.1813. HPLC (Orthogonal method, 30% Solvent B start):
25
29
3
4
4
+
RT = 11.9 min, HI: 97%. m/z obs 541.0 [M + H] . Melting Point = 185.5 °C.
N'-(4-Bromophenyl)-2,2,2-trifluoroacetohydrazide (6). To a stirred suspension of (4ꢀ
bromophenyl)hydrazine hydrochloride (100 g, 447 mmol) in DCM (1.5 L) at room temperature
was added dropwise a solution of TFAA (68.4 mL, 492 mmol) in DCM (150 mL). The reaction
mixture was stirred at room temperature for 2 h and additional TFAA (45 mL, 324 mmol) was
added slowly to the reaction mixture. After 5 min, the homogeneous reaction was judged
complete by LCMS. The reaction mixture was concentrated and diluted with hexane (~1L),
ACS Paragon Plus Environment
26

Journal of Medicinal Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
resulting in an off white crystalline precipitate. The mixture was filtered and the solids were
rinsed with hexanes. The resulting solid was dried under highꢀvacuum for 12 h at 50 °C to give 6
1
(103 g, 364 mmol, 81 % yield) as an off white crystalline compound. H NMR (400 MHz,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CDCl ) δ 8.10 (br. s., 1H), 7.46 ꢀ 7.33 (m, 2H), 6.79 ꢀ 6.67 (m, 2H), 6.06 (br. s., 1H). m/z obs
3
+
3
05 [M + Na] .
(S,E)-3-But-2-enoyl-4-phenyloxazolidin-2-one (7). To a solution of (S)ꢀ(+)ꢀ4ꢀphenylꢀ2ꢀ
oxazolidinone (84 g, 517 mmol), lithium chloride (23 g, 540 mmol) and triethylamine (79 mL,
5
5
60 mmol) in THF (1.2 L) in a 5L 3ꢀneck flask at ꢀ 5 °C was added crotonic anhydride (76 mL,
12 mmol) over 5 min. The reaction temperature increased slightly during the addition. The
mixture was allowed to warm to room temperature, resulting in a white milky solution. The
reaction mixture was stirred at room temperature overnight. Upon completion, as monitored by
LCMS, the reaction was quenched with 0.2M aqueous HCl (600 mL) to give an aqueous layer of
pH~7. The two layers were separated and the aqueous layer was extracted with EtOAc (400 mL).
The combined EtOAc and THF layers were washed with brine, dried over Na SO and
2
4
concentrated to give a light brown oil (200 mL). To the oil was added MeOH (200 mL). Crystals
formed slowly at the beginning and continued to precipitate over 15 min. The product was
collected by filtration and rinsed with MeOH (200 mL) to give a white solid. Additional product
was collected from the MeOH filtrate, and the materials were combined and dried to give 102 g
1
of 7 (85%, 439 mmol) as white crystals. H NMR (400 MHz, CDCl ) δ 7.57 ꢀ 7.22 (m, 6H),
3
7
.21 ꢀ 6.84 (m, 1H), 5.52 (dd, J=8.8, 4.0 Hz, 1H), 4.73 (t, J=8.8 Hz, 1H), 4.42 ꢀ 4.19 (m, 1H),
+
2
.11 ꢀ 1.88 (m, 3H). m/z obs 231.9 [M + H] .
ACS Paragon Plus Environment
27

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
S)-3-((4S,5R)-1-(4-Bromophenyl)-4-methyl-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazole-
5
-carbonyl)-4-phenyloxazolidin-2-one (8). To a solution of 6 (110 g, 390 mmol) in EtOAc
(1500 mL) at room temperature was added dropwise benzenesulfonyl chloride (54 mL, 410
mmol). The mixture was cooled to 0 °C and Hunig's Base (73 mL, 410 mmol) was added
dropwise. The resulting mixture was stirred at room temperature overnight, and the reaction was
judged complete as monitored by LCMS. The mixture was diluted with EtOAc and the organic
layer was washed with water (3 x 2L). The combined organic extracts were dried over sodium
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sulfate,
filtered
and
concentrated
to
give
(Z)ꢀN'ꢀ(4ꢀbromophenyl)ꢀ2,2,2ꢀ
trifluoroacetohydrazonoyl chloride (120 g, 390 mmol) as a yellow liquid, which was used for
next step without further purification. To A solution of (Z)ꢀN'ꢀ(4ꢀbromophenyl)ꢀ2,2,2ꢀ
trifluoroacetohydrazonoyl chloride (94 g, 310 mmol) and 7 (65 g, 280 mmol) in 1,4ꢀdioxane
(
780 mL) under N at 1 atm was added silver(I) carbonate (100 g, 380 mmol) and the mixture
2
was stirred at 50 °C overnight. The reaction was monitored via LCMS until the starting material
®
was no longer detected. The reaction mixture was filtered through Celite , and the filter cake
was washed with EtOAc. The filtrate was concentrated to give an oil. The residue was purified
via crystallization from hexanesꢀether (7:3) to give 8 (72 g, 140 mmol, 46 % yield) as a white
1
crystalline solid. H NMR (400 MHz, CDCl ) δ 7.47 ꢀ 7.31 (m, 5H), 7.29 ꢀ 7.20 (m, 3H), 6.93 ꢀ
3
6
.81 (m, 2H), 5.81 (d, J=2.2 Hz, 1H), 5.39 (dd, J=8.8, 4.2 Hz, 1H), 4.82 (t, J=9.0 Hz, 1H), 4.43
(dd, J=9.2, 4.2 Hz, 1H), 3.16 (dd, J=7.0, 1.5 Hz, 1H), 1.50 (d, J=7.0 Hz, 3H). m/z obs 496.0 [M
+
+
H] .
((4S,5R)-1-(4-Bromophenyl)-4-methyl-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazol-5-
yl)methanol (9). To a solution of 8 (94g, 190 mmol) in THF (1.4 L) was added a solution of
NaBH (7.0 g, 190 mmol) in water (57 mL) in two portions. The resulting mixture was stirred at
4
ACS Paragon Plus Environment
28