Discovery of AM-1638: A potent and orally bioavailable GPR40/FFA1 full agonist

Article abstract of DOI:10.1021/ml300133f

GPR40 (FFA1) is a G-protein-coupled receptor, primarily expressed in pancreatic islets, the activation of which elicits increased insulin secretion only in the presence of elevated glucose levels. A potent, orally bioavailable small molecule GPR40 agonist

Full text of DOI:10.1021/ml300133f

Letter
pubs.acs.org/acsmedchemlett
Discovery of AM-1638: A Potent and Orally Bioavailable GPR40/FFA1
Full Agonist
Sean P. Brown,* Paul J. Dransfield,* Marc Vimolratana, XianYun Jiao, Liusheng Zhu, Vatee Pattaropong,
Ying Sun, Jinqian Liu, Jian Luo, Jane Zhang, Simon Wong, Run Zhuang, Qi Guo, Frank Li,
Julio C. Medina, Gayathri Swaminath, Daniel C.-H. Lin, and Jonathan B. Houze
Departments of Therapeutic Discovery, Metabolic Disorders, Pharmaceutics and Pharmacokinetic and Drug Metabolism, Amgen Inc.,
1120 Veterans Boulevard, South San Francisco, California 94080, United States
S
* Supporting Information
ABSTRACT: GPR40 (FFA1) is a G-protein-coupled recep-
tor, primarily expressed in pancreatic islets, the activation of
which elicits increased insulin secretion only in the presence of
elevated glucose levels. A potent, orally bioavailable small
molecule GPR40 agonist is hypothesized to be an effective
antidiabetic posing little or no risk of hypoglycemia. We
recently reported the discovery of AMG 837 (1), a potent
partial agonist of GPR40. Herein, we present the optimization
from the GPR40 partial agonist 1 to the structurally and
pharmacologically distinct GPR40 full agonist AM-1638 (21).
Moreover, we demonstrate the improved in vivo efficacy that
GPR40 full agonist 21 exhibits in BDF/DIO mice as compared
to partial agonist 1.
KEYWORDS: GPR40, full agonist, AM-1638, AMG 837, insulin secretagogue, FFA1
To provide a greater dynamic range with which to assess
improvements in intrinsic efficacy, we chose to reevaluate
compounds previously synthesized toward discovery of partial
agonist 1 in CHO cells transfected with lower levels of GPR40
expression plasmid [from 5.0 (Figure 1A) to 0.05 μg (Figure
1B)].9 Under the original 5.0 μg plasmid conditions, partial
agonist 1 demonstrates 75% of the response (Emax) shown by
the natural free fatty acid ligand docosahexaenoic acid (DHA)
(Figure 1A). In contrast, reducing the expression plasmid to
0.05 μg affords an assay with the appropriate dynamic range to
distinguish GPR40 partial and full agonists. As depicted in
Figure 1B, partial agonist 1 displays 25% Emax as compared to
DHA. For routine evaluation of compounds, a stable cell line
was developed that displays lower GPR40 expression.
Using the revised assay conditions in the presence of 0.1%
human serum, we discovered that removing the trifluoromethyl
moiety of partial agonist 1 and relocating the aryl−aryl linkage
of the biphenyl from the meta to the para orientation provides
an increase in intrinsic efficacy from 20 to 105%, albeit with a
substantial loss of potency [Table 1, ( )-2, 7.7 μM]. The
addition of a 3-methoxy group to the terminal aryl ring of the
biphenyl provides a significant increase in potency while
maintaining 98% intrinsic efficacy [Table 1, ( )-3, 2.3 μM].
ype II diabetics lose their ability to maintain glucose
Thomeostasis due to defects in both insulin secretion and
action.1 GPR40 (FFA1) is a G-protein-coupled receptor,
primarily expressed in pancreatic islets.2 When activated by
medium to long chain fatty acids, GPR40 elicits increased
insulin secretion only in the presence of elevated glucose
levels.3
This alluring mechanism to treat type II diabetes presents the
potential of little or no risk of hypoglycemia and has been
investigated by multiple groups, leading to the discovery of
several clinical candidates.4−7 We previously described the
discovery of AMG 837 (1),8−10 a small molecule partial agonist
of GPR40 that displays oral efficacy in a variety of rodent
diabetic models without exhibiting hypoglycemia. Because of
the robust antidiabetic activity and favorable pharmacokinetic
properties, 1 was selected for clinical evaluation. Because the
ability of partial agonist 1 to maintain glycemic control was
being tested in a clinical setting, we became interested in
interrogating GPR40 with full agonists. We hypothesized that a
GPR40 full agonist should have a greater ability to induce
insulin secretion and thus provide greater glycemic control. In
this letter, we describe in detail the structure−activity
relationship (SAR) studies that started from the GPR40 partial
agonist 1 and culminate with the identification of GPR40 full
agonist AM-1638 (21) and provide further evidence that
GPR40 full agonists demonstrate superior efficacy over partial
agonists when evaluated in vivo.
Received: May 29, 2012
Accepted: August 10, 2012
Published: August 15, 2012
© 2012 American Chemical Society
726
dx.doi.org/10.1021/ml300133f | ACS Med. Chem. Lett. 2012, 3, 726−730

ACS Medicinal Chemistry Letters
Letter
Table 1. Initial Identification of GPR40 Full Agonists
Figure 1. (A) Effect of 1 and DHA in CHO cells transfected with 5.0
μg of GPR40 expression plasmid. (B) Effect of 1 and DHA in CHO
cells transfected with 0.05 μg of GPR40 expression plasmid.
Evaluation of the single enantiomers of full agonist lead 3
revealed that while the majority of the potency resided in (S)-4
(Table 1, 0.65 μM), compounds in this enantioseries are partial
agonists (Table 1, 4, 47%). Thus, we began a chemical
optimization of the less active enantiomer (R)-5 due to its
demonstration of full agonism (Table 1, 4.0 μM, 99%). Further
evaluation of the terminal aryl ring revealed that a 2-fluoro-5-
methoxy substitution imparts an increase in potency (Table 1,
6, 2.3 μM, 97%). The observation that the chirality of R1
determined whether GPR40 agonists were partial or full
prompted us to remove the methylalkyne functionality. To our
delight, a simple hydrogen substitution at R1 maintains full
agonism and increased potency (Table 1, 7, 1.9 μM, 96%).
Relocating the 4-phenylbenzyloxy group from A4 on the A ring
to A3 affords an additional improvement in potency (Table 1,
8, 1.2 μM, 104%).
Introduction of a tert-butyl substituent on the B ring ortho to
the biaryl linkage supplies a further increase in potency (Table
2, 9, 0.9 μM, 100%). We examined substitutions at R1 on this
more potent scaffold with the knowledge that compounds in
the same enantioseries as alkynes 5 and 6 maintain full
agonism. The general synthesis of agonists 10−17 was carried
out via a high yielding asymmetric rhodium-catalyzed conjugate
addition11 to construct the requisite phenol (Scheme 1). The
biphenyl component was constructed using Suzuki−Miyaura
coupling and then reduction of the methyl ester, and the
resulting alcohol was exchanged for a chlorine. Facile
etherification of the phenol with the biphenyl methyl chloride
a
b
Mean of at least two runs. Percent compared to reference agonist
21.
41%). Phenyl and trifluoromethyl substitutions also led to a
decrease in potency (Table 2; 18, 1.2 μM; 19, 2.1 μM).
After identification of the cyclopropane moiety as the favored
substituent at R1, we returned to optimization of the biphenyl
group. We previously observed exchanging a hydrogen at R2 for
a tert-butyl substituent leads to a modest increase in potency in
an assay run in the presence of 0.1% human serum (Table 3;
20, 0.48 μM; 14, 0.37 μM). However, when the assay is run in
the presence of 100% human serum, this modification affords a
20-fold improvement in potency (Table 3; 20, 38 μM; 14, 1.9
μM). Further increasing the size of the R2 substituent to 5,5-
dimethylcyclopentenyl moiety provides a GPR40 full agonist
that displays the requisite potency in the presence of 100%
human serum (Table 3; 21, 0.71 μM) to warrant evaluation in
vivo. In contrast, the enantiomer of full agonist 21 is both less
potent and a partial agonist (Table 3; 22, 34 μM, 71%).
followed by hydrolysis yields the desired final compounds.12
A
modest increase in potency was observed upon systematically
increasing the size of R1 from hydrogen to methyl and then
ethyl (Table 2; 9, 0.90 μM; 10, 0.77 μM; 11, 0.68 μM). A
further increase in the size of R1 to n-propyl or iso-propyl did
not lead to any additional gain in potency (Table 2; 12, 0.69
μM; 13, 2.2 μM). Realizing that small alkyl substituents were
preferred at R1, we evaluated a series of alkyl rings. Cyclopropyl
and cyclobutyl provided significant increases in potency (Table
2; 14, 0.37 μM; 15, 0.42 μM), while larger rings such as
cyclopentyl and cyclohexyl led to a substantial loss in potency
and intrinsic efficacy (Table 2; 16, 1.31 μM, 59%; 17, 2.1 μM,
727
dx.doi.org/10.1021/ml300133f | ACS Med. Chem. Lett. 2012, 3, 726−730

ACS Medicinal Chemistry Letters
Letter
a
Table 2. Exploration of R1 SAR
Scheme 1.
a
Reagents and conditions: (a) 3-Hydroxybenzene boronic acid,
Rh(I)(OH)/(S)-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl dimer,
H2O, 1,4-dioxane, 86−96%, 83−96% ee. (b) Chiral resolution to
remove minor isomer: Chiracel OD-H column, 3% iPrOH/hexane,
220 nm. (c) (Tf)2O, Et3N, 4-dimethylaminopyridine, CH2Cl2, 98%.
(d) 2-Fluoro-5-methoxybenzene boronic acid, Pd(PPh3)4, K2CO3,
DMF, 100 °C, 71%. (e) LiAlH4, THF, 0 °C, 72%. (f) SOCl2, CH2Cl2,
74%. (g) A, Cs2CO3, DMF. (h) LiOH, EtOH, H2O.
Table 3. Effect of Biphenyl Modification on GPR40 Activity
in the Presence 100% Human Serum
a
b
Mean of at least two runs. Percent compared to reference agonist
21.
To supply quantities of full agonist 21 sufficient for in vivo
studies, a scalable synthesis was developed (Scheme 2).
Protection of phenol 23 as a THP ether allowed introduction
of the 5,5-dimethylcyclopentenyl group by Suzuki−Miyaura
coupling utilizing the Buchwald biarylphospine 2-dicyclohex-
ylphosphino-2′,6′-dimethoxybiphenyl (S-Phos) as a supporting
ligand.13 Mild removal of the THP group under acidic
conditions, followed by activation of the phenol as a triflate
and subsequent palladium-catalyzed Suzuki−Miyaura coupling,
again employing S-Phos, yielded biphenyl 24. Reduction of the
ester and chlorination delivered chloride 26. Rhodium-
catalyzed asymmetric 1,4-addition of 3-hydroxybenzene bor-
onic acid to α,β-unsaturated ester 27 delivered phenol 28 in
good yield and enantioselectivity. Alkylation of enantiomeri-
cally pure phenol 28 with chloride 26, followed by ester
hydrolysis, affords full agonist 21 in 36% overall yield.
a
b
Mean of at least two runs. Percent compared to reference agonist
21.
Moreover, oral administration of full agonist 21 demonstrates
excellent oral bioavailability (mouse, >100%; rat, 72%; and
cyno, 71%), affording the GPR40 full agonist 21 and the
appropriate pharmacokinetic properties for comparison in vivo
to the partial agonist 1.
Full agonist 21 and partial agonist 1 were compared at a dose
of 60 mg/kg for their ability to improve glycemic control in
BDF mice with diet-induced obesity (DIO), a model of type II
diabetes that develops elevated blood glucose and impaired
glucose tolerance.14 Both compounds reduce blood glucose
excursion (Figure 2A; 1 [glucose]max 317 mg/dL; 21,
Importantly, full agonist 21 exhibits moderate cross-species
plasma clearance and volume of distribution, resulting in
plasma half-lives suitable for evaluation of its antidiabetic
properties in mouse, rat, and cynomologus monkey (Table 4).
728
dx.doi.org/10.1021/ml300133f | ACS Med. Chem. Lett. 2012, 3, 726−730

ACS Medicinal Chemistry Letters
Letter
a
Scheme 2.
a
Reagents and conditions: (a) Dihydropyran, pyridinium p-toluene-
sulfonate, 90%. (b) 2-(5,5-Dimethylcyclopent-1-enyl)-4,4,5,5-tetra-
methyl-1,3,2-dioxaborolane, Pd(OAc)2, S-Phos, K3PO4, DMF/H2O,
60 °C, 80%. (c) Pyridinium p-toluenesulfonate, MeOH, 90%. (d)
PhN(Tf)2, Et3N, CH2Cl2, 88%. (e) 2-Fluoro-5-methoxybenzene
boronic acid, Pd(PPh3)4, K2CO3, DMF, 90 °C, 91%. (f) LiAlH4,
THF, 0 °C. (g) SOCl2, CH2Cl2, 86% over two steps. (h) 3-
Hydroxybenzene boronic acid, Rh(I)(OH)/(S)-2,2′-bis-
(diphenylphosphino)-1,1′-binaphthyl dimer, H2O, 1,4-dioxane, 60%,
85% ee. (i) Chiral resolution to remove minor isomer: Chiracel OD-H
column, 3% iPrOH/hexane, 220 nm, rt, 40 min. (j) Compound 26,
Cs2CO3, DMF. (k) LiOH, EtOH, H2O, 80% over two steps.
a
Table 4. Pharmacokinetic Properties of Full Agonist 21
oral Cmax
(μM)
tmax
(h)
Cl
Vdss
(μM)
iv t1/2
(h)
species
(L/h/kg)
% F
mouse
rat
14
1
0.18
0.91
0.81
0.45
1.1
2.0
1.8
2.1
>100
72
Figure 2. The effect of full agonist 21 and partial agonist 1 during an
OGTT in BDF/DIO mice. Time-dependent changes in (A) plasma
glucose and (C) plasma insulin after oral administration of either full
agonist 21 or partial agonist 1 followed by 1 g/kg oral glucose
challenge. (B) Incremental AUC0−120 min of plasma glucose levels. (D)
Incremental AUC−60−60 min of plasma insulin levels. Values are means
3.6
5.1
3.5
2
cyno
2.0
71
a
Administered at a dose of 1 mg/kg, iv; 5 mg/kg, po, in mice.
Administered at a dose of 0.5 mg/kg, iv; 2 mg/kg, po, in rats.
Administered at a dose of 0.5 mg/kg, iv; 2 mg/kg, po, in cyno. Data
are expressed as mean values (n = 3).
#
SDs (n = 6). p ≤ 0.025 as compared with control by one-tailed
Williams’ test.
[glucose]max 204 mg/dL) and area under the curve of blood
glucose (AUC 0−60 min) (Figure 2B) upon administration 1 h
before an oral glucose tolerance test (OGTT) as compared to
the control (Figure 2A; control [glucose]max 462 mg/dL).
However, full agonist 21 shows greater efficacy in blunting
glucose excursion than partial agonist 1. Moreover, full agonist
21 imparts greater 46% improvement in AUCglucose, while
partial agonist 1 provides 34%. Partial agonist 1 does not
display a statistically significant increase in plasma insulin as
compared to the control. However, full agonist 21 elicits a
statistically significant increase in both plasma insulin at all time
points and in the AUCinsulin (Figure 2C,D). This is in accord
with our original hypothesis and further confirms that the
greater intrinsic efficacy observed in vitro for full agonist 21
affords advantages in vivo for maintaining plasma glucose
homeostasis as compared to partial agonist 1.
In conclusion, we have described the SAR leading from the
GPR40 partial agonist 1 to a structurally and pharmacologically
distinct series of GPR40 agonists, exemplified by the orally
bioavailable GPR40 full agonist 21. The antidiabetic activity
that full agonist 21 exhibits in BDF/DIO mice provides
729
dx.doi.org/10.1021/ml300133f | ACS Med. Chem. Lett. 2012, 3, 726−730

ACS Medicinal Chemistry Letters
Letter
(11) Hayashi, T.; Yamasaki, K. Rhodium-catalyzed asymmetric 1,4-
addition and its related asymmetric reactions. Chem. Rev. 2003, 103,
2829−2844.
(12) Walker, S. D.; Borths, C. J.; DiVirgilio, E.; Huang, L.; Liu, P.;
Morrison, H.; Sugi, K.; Tanaka, M.; Woo, J. C. S.; Faul, M. M.
Development of a scalable synthesis of a GPR40 receptor agonist. Org.
Process Res. Dev. 2011, 15, 570−580.
(13) Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L.
Catalysts for Suzuki−Miyaura coupling processes: Scope and studies
of the effect of ligand structure. J. Am. Chem. Soc. 2005, 127, 4685−
4696.
(14) Karasawa, H.; Nagata-Goto, S.; Takaishi, K.; Kumagae, Y. A
novel model of type 2 diabetes mellitus based on obesity induced by
high-fat diet in BDF1 mice. Metabolism 2009, 58, 296−303.
compelling evidence that GPR40 full agonists afford access to a
powerful mechanism for maintaining glycemic control and
great potential for the treatment of type II diabetic patients.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and analytical data. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: sebrown@amgen.com (S.P.B.) or pdransfi@amgen.
com (P.J.D.).
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Defronzo, R. A. From the triumvirate to the ominous octet: a
new paradigm for the treatment of type 2 diabetes. Diabetes 2009, 58,
773−795.
(2) Itoh, Y.; Kawamata, Y.; Harada, M.; Kobayashi, M.; Fujii, R.;
Fukusumi, S.; Ogi, K.; Hosoya, M.; Tanaka, Y.; Uejima, H.; Tanaka,
H.; Maruyama, M.; Satoh, R.; Okubo, S.; Kizawa, H.; Komatsu, H.;
Matsumura, F.; Noguchi, Y.; Shinohara, T.; Hinuma, S.; Fujisawa, Y.;
Fujino, M. Free fatty acids regulate insulin secretion from pancreatic β
cells through GPR40. Nature 2003, 422, 173−176.
(3) Stoddart, L. A.; Smith, N. J.; Milligan, G. International union of
pharmacology. LXXI. Free fatty acid receptors FFA1, -2, and -3:
Pharmacology and pathophysiological functions. Pharmacol. Rev. 2008,
60, 405−417.
(4) For a review, see Medina, J. C.; Houze, J. B. GPR40 (FFAR1)
modulators. Annu. Rep. Med. Chem. 2008, 43, 75−85.
(5) For a review, see Bharate, S. B.; Nemmani, K. VS.; Vishwakarma,
R. A. Progress in the discovery and development of small-molecule
modulators of G-protein-coupled receptor 40 (GPR40/FFA1/
FFAR1): An emerging target for type 2 diabetes. Expert Opin. Ther.
Pat. 2009, 19, 237−264.
(6) Negoro, N.; Sasaki, S.; Mikami, S.; Ito, M.; Suzuki, M.; Tsujihata,
Y.; Ito, R.; Harada, A.; Takeuchi, K.; Suzuki, N.; Miyazaki, J.; Santou,
T.; Odani, T.; Kanzaki, N.; Funami, M.; Tanaka, T.; Kogame, A.;
Matsunaga, S.; Yasuma, T.; Momose, Y. Discovery of TAK-875: A
potent, selective, and orally bioavailable GPR40 agonist. ACS Med.
Chem. Lett. 2010, 1, 290−294.
(7) Christiansen, E.; Due-Hansen, M. E.; Urban, C.; Merten, N.;
Pfleiderer, M.; Karlsen, K. K.; Rasmussen, S. S.; Steensgaard, M.;
Hamacher, A.; Schmidt, J.; Drewke, C.; Petersen, R. K.; Kristiansen,
K.; Ullrich, S.; Kostenis, E.; Kassack, M. U.; Ulven, T. Structure-
activity study of dihydrocinnamic acids and discovery of the potent
FFA1 (GPR40) agonist TUG-469. ACS Med. Chem. Lett. 2010, 1,
345−349.
(8) Woo, J. C. S.; Cui, S.; Walker, S. D.; Faul, M. M. Asymmetric
syntheses of a GPR40 receptor agonist via diastereoselective and
enantioselective conjugate alkynylation. Tetrahedron 2010, 66, 4730−
4737.
(9) Lin, D. C-H.; Zhang, J.; Zhuang, R.; Li, F.; Nguyen, K.; Chen, M.;
Tran, T.; Lopez, E.; Lu, J. Y. L.; Li, X. N.; Tang, L.; Tonn, G. R.;
Swaminath, G.; Reagan, J. D.; Chen, J.-L.; Tian, H.; Lin, Y.-J.; Houze, J.
B.; Luo, J. AMG 837: A novel GPR40/FFA1 agonist that enhances
insulin secretion and lowers glucose levels in rodents. PLoS One 2011,
6, e27270.
(10) Houze, J. B.; Zhu, L.; Sun, Y.; Akerman, M.; Qiu, W.; Zhang, A.
J.; Sharma, R.; Schmitt, M.; Wang, Y.; Liu, J.; Liu, J.; Medina, J. C.;
Reagan, J. D.; Luo, J.; Tonn, G.; Zhang, J.; Lu, J. Y-L.; Chen, M.;
Lopez, E.; Nguyen, K.; Yang, L.; Tang, L.; Tian, H.; Shuttleworth, S. J.;
Lin, D. C-H. AMG 837: A potent, orally bioavailable GPR40 agonist.
Bioorg. Med. Chem. Lett. 2012, 22, 1267−1270.
730
dx.doi.org/10.1021/ml300133f | ACS Med. Chem. Lett. 2012, 3, 726−730