Discovery of novel benzo[b]thiophene tetrazoles as non-carboxylate GPR40 agonists

Article abstract of DOI:10.1016/j.bmcl.2017.12.022

GPR40 partial agonism is a promising new mechanism for the treatment of type 2 diabetes mellitus with clinical proof of concept. Most of the GPR40 agonists in the literature have a carboxylic acid functional group, which may pose a risk for idiosyncratic drug toxicity. A novel series of GPR40 agonists containing a tetrazole as a carboxylic acid bioisostere was identified. This series of compounds features a benzo[b]thiophene as the center ring, which is prone to oxidation during phase 1 metabolism. Following SAR optimization targeting GPR40 agonist activity and intrinsic clearance in microsomes (human and rat), potent and metabolically stable compounds were selected for in vivo evaluation. The compounds are efficacious at lowering blood glucose in a SD rat oGTT model.

Full text of DOI:10.1016/j.bmcl.2017.12.022

Accepted Manuscript
Discovery of Novel Benzo[b]thiophene Tetrazoles as Non-Carboxylate GPR40
Agonists
Hui Huang, Michael P. Winters, Sanath K. Meegalla, Eric Arnoult, S. Paul Lee,
Shuyuan Zhao, Tonya Martin, Brian Rady, Jianying Liu, Meghan Towers,
Monicah Otieno, Fran Xu, Heng Keang Lim, Jose Silva, Alessandro Pocai, Mark
R. Player
PII:
S0960-894X(17)31182-4
https://doi.org/10.1016/j.bmcl.2017.12.022
BMCL 25478
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
5 September 2017
8 December 2017
12 December 2017
Please cite this article as: Huang, H., Winters, M.P., Meegalla, S.K., Arnoult, E., Paul Lee, S., Zhao, S., Martin, T.,
Rady, B., Liu, J., Towers, M., Otieno, M., Xu, F., Keang Lim, H., Silva, J., Pocai, A., Player, M.R., Discovery of
Novel Benzo[b]thiophene Tetrazoles as Non-Carboxylate GPR40 Agonists, Bioorganic & Medicinal Chemistry
Letters (2017), doi: https://doi.org/10.1016/j.bmcl.2017.12.022
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
Discovery of novel benzo[b]thiophene
tetrazoles as non-carboxylate GPR40
agonists
Leave this area blank for abstract info.
Hui Huang, Michael P. Winters, Sanath K. Meegalla, Eric Arnoult, S. Paul Lee, Shuyuan Zhao, Tonya
Martin, Brian Rady, Jianying Liu, Meghan Towers, Fran Xu, Monicah Otieno, Heng Keang Lim, Jose
Silva,
Alessandro Pocai, Mark R. Player
Janssen Research & Development, Welsh and McKean Roads, Spring House, PA 19477-0776, US

Bioorganic & Medicinal Chemistry Letters
Discovery of Novel Benzo[b]thiophene Tetrazoles as Non-Carboxylate GPR40
Agonists
Hui Huang, Michael P. Winters, Sanath K. Meegalla, Eric Arnoult, S. Paul Lee, Shuyuan Zhao, Tonya
Martin, Brian Rady, Jianying Liu, Meghan Towers, Monicah Otieno, Fran Xu, Heng Keang Lim, Jose
Silva, Alessandro Pocai, Mark R. Player*
Janssen Research & Development, Welsh and McKean Roads, Spring House, PA 19477-0776, USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
GPR40 partial agonism is a promising new mechanism for the treatment of type 2 diabetes mellitus
with clinical proof of concept. Most of the GPR40 agonists in the literature have a carboxylic acid
functional group, which may pose a risk for idiosyncratic drug toxicity A novel series of GPR40
agonists containing a tetrazole as a carboxylic acid bioisotere was identified. This series of compounds
features a benzo[b]thiophene as the center ring, which is prone to oxidation during phase 1
metabolism. Following SAR optimization targeting GPR40 agonist activity and intrinsic clearance in
microsomes (human and rat), potent and metabolically stable compounds were selected for in vivo
evaluation. The compounds are efficacious at lowering blood glucose in a SD rat oGTT model.
Keywords:
GPR40 agonist, tetrazole
type 2 diabetes, GPCR, GSIS
©
2017 Elsevier Ltd. All rights reserved.
*
Corresponding author. Tel.: +1 215 628 7860
E-mail address: mplayer@its.jnj.com (M.R. Player)
According to WHO 2014 estimates, diabetes continues to
present an increasing health risk to the global population,
affecting 422 million individuals worldwide. Type 2 diabetes
1
There are numerous GPR40 partial agonists reported in the
(
T2D), which affects the vast majority (ca. 90%) of the diabetic
literature, several of which have been explored in clinical settings
Fig. 1).
The most advanced compound was TAK-875 from Takeda in
population, occurs when the body cannot effectively utilize the
insulin that is being produced. The direct effect of uncontrolled
diabetes is raised blood sugar, or hyperglycemia. Over time,
patients with T2D are at risk of developing other diseases, such as
cardiovascular diseases, stroke, kidney failure, blindness,
(
8
9
10
Phase III. Other companies such as Amgen, Eli Lilly, and
1
1
Japan Tobacco had their GPR40 agonists in Phase I and Phase II
clinical studies. The exact structure of JTT-851 has not been
disclosed, and the structure shown in Fig.1 is one of the potent
compounds that were disclosed in the only GPR40 patent
2
peripheral neuropathy, and several forms of cancer. Most of
current available therapies for T2D are associated with
undesirable side effects including hypoglycemia, weight gain,
1
2
3
,4
application published by Japan Tobacco.
urinary tract infections, and gastric symptoms. Thus there is still
a significant unmet medical need for new antidiabetic agents,
which provide glycemic control with extended durability and
improved safety.
Through binding and activating several free fatty acid
receptors (FFARs), free fatty acids (FFAs) are involved in the
regulation of metabolic homeostasis, and play an important role in
the development of many metabolic diseases, including T2D,
5
obesity, and atherosclerosis. GPR40, also known as FFAR1,
belongs to the class A family of G-protein coupled receptors, and
is expressed predominantly in pancreatic

cells and
6,7
enteroendocrine cells. Activation of the GPR40 receptor by
medium and long chain FFAs, leads to enhancement of insulin
secretion in a glucose-dependent manner, which makes it an
attractive therapeutic target to treat T2D with minimal risk of
hypoglycemia and weight gain.
Figure 1. GPR40 partial agonists reported to be in clinical development.

TAK-875 is a GPR40 partial agonist that has shown
benzo[b]thiophene center ring and a 2-propynyl substitution at
the -carbon retained the potency of the acid. In calcium flux
assays in cells overexpressing human and rat GPR40 receptors,
tetrazole 1 demonstrated an EC50 of 9 nM and 7 nM for human
and rat GPR40, respectively (Table 1).
clinical efficacy comparable to sulfonylureas but no propensity to
1
3,14
cause hypoglycemia and weight gain.
However, TAK-875
was terminated during Phase III trials due to drug-induced liver
injury (DILI) that was presumably idiosyncratic in nature. It is
proposed that the formation and accumulation of a reactive acyl
glucuronide (AG) in hepatocytes. We found that the inhibition of
elimination of TAK-875 and its AG by MRP2/3/4 transporters
are key contributors to TAK-875-mediated DILI and thus sought
To date, there are two major types of GPR40 modulators
reported: partial agonists, e.g. TAK-875, and full agonists, e.g.
2
7
AM-1638. The partial agonists only signal through the Gq
pathway, while the full agonists can signal through not only
1
5
2
8
to mitigate this in our series.
Gqbut also Gs and cAMP pathways. The different levels of
receptor activation afforded by partial vs. full agonists are evident
in transiently transfected CHO cells expressing relatively low
Most of the GPR40 agonists in the literature have a
carboxylic acid functional group that mimics the acid of the FFA
in the binding pocket. Although there are many carboxylic acid-
containing drugs that have been marketed worldwide, some of
them have been withdrawn from the market due to idiosyncratic
drug toxicities arising from the metabolism of the carboxylic acid
2
9
levels of hGPR40.
monophosphate (IP1) turnover experiments, both partial agonist
TAK-875) and full agonist (AM-1638) stimulated IP1
accumulation, however, the full agonist showed a higher Emax
In the Gq-mediated inositol
(
1
6,17
moiety (e.g. AG).
Reactive AG metabolites can covalently
(
(
Fig. 3). When cAMP levels were measured, only the full agonist
AM-1638) stimulated cAMP production. Since tetrazole 1
modify endogenous proteins, which may initiate toxicity/immune
1
8
responses. There are two ways to mitigate the problem:
carboxylic acid-containing compounds with more stable/less
reactive AG metabolites or replacing the carboxylic acid with a
bioisostere. A few companies have explored the second option
for non-carboxylate GPR40 agonists. For example, Astellas has
behaved similarly to TAK-875 in these assays, we consider it a
partial GPR40 agonist.
1
9
20
identified oxadiazolidines and tetrazoles, Merck has used
2
1
thiazolidine-2,4-diones, and Mochida has disclosed compounds
2
2
with 1,2-thiazinan-3-one 1,1-dioxides (Fig. 2). Generally, these
non-carboxylate GPR40 agonists are less potent than their
carboxylic acid counterparts, and are still in the preclinical testing
stage. Reported herein is our effort on discovery, optimization,
and pharmacologic characterization of non-carboxylate GPR40
agonists with a tetrazole as the carboxylic acid bioisostere.
.
Figure 3. Compound 1 is a GPR40 partial agonist by IP1 and cAMP in a low-
expressing cell line.
Figure 2. Non-carboxylate GPR40 agonists in preclinical development.
The carboxylic acid and its tetrazole bioisosteric pair exhibit
very similar H-bond environments in crystal structures from the
CSD, and the attractive energies of these H-bonds are very
1
H-Tetrazole is one of the most commonly used
bioisosteres for carboxylic acids arising from their similarity in
pKa (ca. 4.5-4.9 vs 4.2-4.4, respectively).
tetrazoles may exhibit an advantage over carboxylic acids
2
3,24
30
Metabolically,
similar. Recently, the high resolution X-ray structure of GPR40
3
1
in complex with the partial agonist, TAK-875 was published.
because they form N-glucuronides which are not as chemically
reactive as the O-glucuronides derived from carboxylic acids.
We used this co-crystal structure of hGPR40 (PDB ID: 4PHU) to
predict the binding mode of tetrazole 1. The binding site was
defined as the set of residues in a range of 5 Å around the co-
crystallized ligand, TAK-875. The best docking pose of tetrazole
1 in the “pocket #1” binding site, predicted by GOLD (version
2
5
Tetrazole-bearing drugs have not been linked to toxic effects in
human due to N-glucuronidation. Our earlier efforts on discovery
of potent GPR40 agonists led to a novel series of carboxylic acid
2
6
32
compounds with benzo[b]thiophene as the center ring. Based on
this scaffold, various carboxylic acid bioisosteres were explored,
and tetrazole 1 was identified as the most potent GPR40 agonist
among them. Replacement of the carboxylate on different
chemical scaffolds of GPR40 agonists with tetrazole frequently
resulted in loss of potency. However, the series with a
5.4.1) , is shown in Figure 4. The o-tolyl substituted
benzo[b]thiophene part of the molecule is exposed to the lipid
bilayer of the membrane, in the same area where the
(methylsulfonyl)ethoxy biphenyl of TAK-875 projects. The
tetrazole group of 1 is buried into the binding site, making ionic
H-bonds with key binding site residues Arg183 and Arg258,

while the carboxylate in TAK875 involves Tyr91. This tetrazole
is also stabilized by pi-pi (π – π) stacking interactions with
Phe87. Such interactions are crucial as they serve as anchors for
the compound in the binding site. The propyne group is making
mainly hydrophobic interactions in a small pocket formed by
Tyr91, Leu138, Ala182 and His137.
R- and S- enantiomers (at sulfur atom, about 1:1 ratio). and LC-
MS confirmed that it was the same compound as M4. The
sulfoxide 2 was weakly active in the in vitro assay with an EC50
of 1.4 M for hGPR40, and it was also metabolically unstable in
microsomes (Table 1). We initiated an effort to search for
benzo[b]thiophene tetrazoles with improved metabolic stability.
Table 1. Effect of 2-benzo[b]thiophene substitutions on GPR40 activity and
intrinsic clearance.
a
a
b
b
1
hGPR40
rGPR40
hClint
T
1/2
rClint T
1/2
Cpd
1
R
EC50 (nM)
EC50 (nM)
(min)
(min)
H
H
9
7
5
16
2
1411
20
301
10
28
4
11
29
22
65
55
(sulfoxide)
3
CH
3
Figure 4. Predicted binding mode of tetrazole 1 (shown in yellow sticks) in
the partial agonist binding site (PDB ID:4PHU), overlaying with TAK-875
4
F
15
14
4
(shown in pink sticks). The deprotonated NH of 1 is making ionic H-bonds
with key binding site residues Arg183 and Arg258, while the carboxylate in
TAK875 involves Tyr91, additionally. The picture was generated with Pymol.
5
Cl
Br
20
20
10.5
9.5
3
3
6
19
19
Despite good potency toward GPR40 receptors, 1 had poor
microsomal stability with an intrinsic clearance T1/2 of 5 min and
6 min in human and rat microsomes, respectively (Table 1). To
7
I
28
33
10
29
1
identify the metabolically labile part of the molecule, a metabolic
ID study was carried out in rat, dog and human liver S9 fractions.
The major metabolite of 1 was the mono-oxidation product M4 in
all three species (Table 2). The turnover was higher in human and
rat than in dog S9 mix. Only 14.1 % of unchanged drug (UD)
was remaining in human S9 after 1 hour of incubation. The phase
II metabolite (M17, N-glucuronide of the tetrazole moiety) was
only 1% in human S9, while the percentage was higher in dog S9
8
9
CN
CF
16
15
19
12
2
32
91
3
12
1
0
672
128
5
20
a
GPR40 activity from FLIPR-based intracellular calcium mobilization assay.
b
All reported values are averages of duplicate experiments. The intrinsic
clearance (Clint) was determined using human and rat liver microsomes.
(7.6%). It was suspected that the mono-oxidation metabolite M4
was the benzo[b]thiophene 1-oxide 2 shown in Table 2.
Compound 2 was synthesized by treating 1 with hydrogen
peroxide and TFA, and NMR indicated that it was a mixture of
Table 2. Metabolite ID study identifies the major metabolite of 1 as the mono-oxidation product M4.
b
b
RT
% of Reconstructed MS
a
+
Metabolites
, UD
[M+H]
465
271
481
481
497
641
463
(min)
26.64
21.73
23.15
24.24
21.18
25.22
24.62
Rat S9
26.1
0.3
24.1
8.2
10.6
0.8
14.6
Dog S9
60.9
0.1
27.2
0.3
0.0
7.6
0.7
Human S9
14.1
9.5
1
M1, Monooxy-O-Desalkyl-UD
M4, Monooxy-UD
M5, Monooxy-UD
M7, Diooxy-UD
M17, Glucuronide of UD
M19, Dehydrogenated UD
46.9
7.4
2.7
1.0
12.4
a
Compound was incubated with cofactor-fortified rat, dog and human liver S9 fractions at 37 °C for 1 h. The cofactors include NADPH (for
Phase I), and UDPGA, UDPAG, DTT, PAPS (for Phase II). LC-MS/MS analysis was used to separate and identify the metabolites.
b

Scheme 1. Synthesis of tetrazole 1: (a) isopropylmagnesium chloride, DMF, THF, rt (68%); (b) bromine, acetic acid, 15 °C (68%); (c) o-tolylboronic acid,
Pd(dppf)Cl , Cs CO , 1,4-dioxane/H O, 80 °C (87%); (d) NaBH , THF, MeOH, 0 °C (89%); (e) thionyl chloride, DCM, DMF, 0 °C; (f) methyl (S)-3-(4-
hydroxyphenyl)hex-4-ynoate, Cs CO , acetonitrile, rt (87%, 2 steps); (g) LiOH, THF, MeOH, H O, rt (57%); (h) thionyl chloride, DCM, DMF, 0 °C; (i) NH
ether, rt (quantitative, 2 steps); (j) 2,2,2-trifluoroacetic anhydride, trimethylamine, DCM, 0 °C-rt (96%); (k) TMS-N , Bu SnO, toluene, 110°C (33%); (l) 3-
aminopropanenitrile, HATU, diisopropylethylamine, DMF, rt (quantitative); (m) PCl , pyridine, DCM, 40 °C; (n) TMSN , DCM, rt (70%, 2 steps); (o) NaOH,
2
2
3
2
4
2
3
2
3
,
3
2
5
3
H
2
O, THF, i-PrOH, rt (69%).
A general synthesis of the tetrazole 1 is shown in Scheme 1.
occurred when the substituent was as large as the methylsulfonyl
group (10). There is no major change with the intrinsic clearance,
especially in human microsomes. However, there is a trend for
improvement in rat microsomal stability with substituents like Cl
First, a magnesium-halogen exchange reaction quenched with
DMF afforded benzo[b]thiophene-5-carbaldehyde S2, which was
further decorated with a bromine at the 3-position (S3). A Suzuki
reaction with o-tolylboronic acid followed by reduction of the
aldehyde gave the alcohol intermediate S5, which was coupled
with the right-side phenolic ester through the chloride S6.
Following ester hydrolysis, the carboxylic acid S8 can be
converted to the tetrazole 1 through two routes. The first route
utilizes a 1,3-dipolar cycloaddition on the nitrile S10 (obtained
from the amide S9) with TMS-azide. However this route often
gave low yields of tetrazoles. A second more consistent yielding
(5), Br (6), and CF (9).
3
Another modification that may be useful in blocking the
phase
1 oxidation on sulfur is the 7-position of the
benzo[b]thiophene. Analogues with methyl and small halogens
(F- and Cl-) substituted at the 7-position were prepared
(Supplementary Data), and their data are shown in Table 3. The
methyl compound (11) was much less active and equally unstable
metabolically as compared to 1. The 7-F compounds (12, 13)
were more potent than the methyl (11) and the Cl compounds
(14, 15), probably due to their smaller size. The compounds (13,
15) with a methoxy substituent on the distal o-methylphenyl
group were designed to reduce the lipophilicity of the molecule.
They were generally more active against the GPR40 receptors
and more stable in microsomes, especially human microsomes,
than the unsubstituted ones (12, 14).
3
4
route was developed . The acid S8 was converted to the
cyanoethyl amide S11, which reacted with TMS-azide via the
imidoyl chloride intermediate to provide the cyanoethyl-protected
tetrazole S12. Deprotection with aqueous hydroxide under mild
conditions led to the tetrazole 1 (experimental details and
characterization of 1 can be found in Supplementary Data).
Since 1 was found metabolically unstable due to oxidation of
the benzo[b]thiophene sulfur, we synthesized the benzofuran
analog of 1. The compound showed similar GPR40 activity
against both species as 1, and much more stable in rat
microsomes (T1/2 = 49 min). However, it was still unstable in
human microsomes (T1/2 = 6 min). So, we decided to focus on the
optimization of the benzo[b]thiophene scaffold. The first strategy
was to incorporate substation groups at the 2-position of
benzo[b]thiophene to reduce formation of the sulfoxide. 2-Methyl
Table 3. Effect of 7-benzo[b]thiophene substitutions on GPR40 activity and
intrinsic clearance.
substituted compound
commercially available
3
was synthesized utilizing the
methyl 3-bromo-2-
methylbenzo[b]thiophene-5-carboxylate according to the
methods outlined in Scheme 1, first route. The synthesis of other
a
a
b
b
hGPR40
EC50
(nM)
rGPR40
EC50
(nM)
hClint
rClint
T
1/2
2
-substituted benzo[b]thiophenes is described in Supplementary
Data.
The GPR40 activity and intrinsic clearance data of these 2-
2
3
Cpd
R
R
T
1/2
(min)
(min)
1
1
1
1
2
3
CH
3
H
H
134
23
213
30
<4
<4
substituted benzo[b]thiophenes are shown in Table 1. The methyl
group (3) had minimal effect on either potency or intrinsic
clearance. For electron-withdrawing substituents, size of the
group affects the potency. For small groups, there was a 2- to 3-
fold potency loss as compared to 1. A significant potency drop
F
5
6
F
-OCH
3
10
39
30
20

1
1
4
5
Cl
Cl
H
63
25
108
71
13
88
7
2
2
7
8
83
21
16
<4
11
16
6
-OCH
3
20
107
858
46
a
GPR40 activity from FLIPR based intracellular calcium mobilization assay;
All reported values are averages of duplicate experiments. The intrinsic
clearance (Clint) was determined using human and rat liver microsomes.
b
29
239
103
a
Because methoxy substitution on the distal phenyl group
improved the human microsomal stability, a variety of different
substituents were explored at the 3-position of the
benzo[b]thiophene (Table 4). Polar groups were introduced on
the o-methylphenyl ring through alkylation of the corresponding
GPR40 activity from FLIPR based intracellular calcium mobilization assay;
b
All reported values are averages of duplicate experiments. The intrinsic
clearance (Clint) was determined using human and rat liver microsomes.
stability. When the o-methylphenyl ring was replaced with
heteroaryls (24-26) or saturated rings (27-29), the compounds are
generally substantially less potent, yet not stable, especially in
human microsomes. These compounds (24-29) were synthesized
in the same manner as described in Scheme 1, and the 3-
substitutions were added by Suzuki coupling (24-26, 29),
Buchwald reaction (27) or Negishi coupling (28).
2
6
phenols. Many of them maintained good GPR40 activity, and
were beneficial for both human and rat microsomal stability (16-
2
2). Compound 16 was equally potent as 1, but much more stable
in human microsomes, although its intrinsic clearance in rat
microsomes was still quite rapid. Compound 17 and 18 are close
analogues of 16 with a 2-methoxyethoxyl group substituted at the
Having shown that some of substitutions (Cl, CF ) at the 2-
3
3
- and 5-position of the o-methylphenyl ring, respectively. They
position of the thiophene improved rat microsomal stability, and
polar groups at the 4-position of the aryl ring had a beneficial
effect on stability in both human and rat microsomes, we
combined these features in a new set of analogues. Methyl (30),
fluoro (31), and chloro (32) substitution at the 2-position of the
benzo[b]thiophene further improves the human microsomal
stability of 16, but with a 3- to 8-fold loss of GPR40 potency
were 9- to 14-fold less potent than 16 in human GPR40 receptors.
The sulfone-containing compounds, 21 and 22 had reasonable
GPR40 activity, with improved stability in human and rat
microsomes. The replacement of the o-methyl group with o-
trifluoromethyl group (23) had little effect on either activity or
Table 4. Effect of substitutions on 3-benzo[b]thiophene on GPR40 activity
and intrinsic clearance.
(
Table 5). However, trifluoromethyl (33) substitution not only
maintained good GPR40 activity, but also greatly increased
microsomal stability in both species, with T1/2 greater than 3
hours.
Table 5. Effect of substitutions on 2-benzo[b]thiophene of compound 16 on
GPR40 activity and intrinsic clearance.
*
a
a
b
b
hGPR40
EC50
rGPR40
EC50
hClint
rClint
T
1/2
4
Cpd
R
T
1/2
(nM)
(nM)
(min)
(min)
1
6
9
4
43
13
1
1
1
7
8
9
54
85
29
14
26
19
150
52
7
a
a
b
b
hGPR40
EC50
rGPR40
EC50
(nM)
hClint
rClint
T
1/2
5
Cpd
R
T
1/2
22
76
(nM)
(min)
(min)
3
3
0
1
CH
3
47
32
24
21
14
89
7
73
F
18
39
13
117
121
21
25
2
0
8
4
57
62
32
Cl
3
3
CF
3
>180
>180
2
2
1
2
12
30
9
6
77
52
89
a
GPR40 activity from FLIPR based intracellular calcium mobilization assay;
b
All reported values are averages of duplicate experiments. The intrinsic
clearance (Clint) was determined using human and rat liver microsomes.
122
Several compounds were selected for pharmacokinetic (PK)
profiling in rat (Table 6). As expected from its low microsomal
stability in rat, 1 has high clearance and low oral exposure.
Fluorination of the benzo[b]thiophene ring (4) led to a 3-fold
increase with respect to oral exposure, but the clearance was still
high. Both 2-chloro-benzo[b]thiophene (5) and 2-trifluoromethyl-
benzo[b]thiophene (9) showed low clearance and higher oral
exposure, consistent with their relatively good stability in rat
microsomes. Despite that the sulfone-containing compounds, 21
and 22 had good in vitro microsomal stability in human and rat,
their rat PK profiles are not optimal, with very high clearance and
low oral exposure. It is worth mentioning that the molecular
2
2
3
4
13
88
2
<4
<4
10
11
20
2
2
5
6
37
67
18
32
14
9
20
33

weights of 21 and 22 are 601 and 613, respectively, which may
be one of the contributing factors to their poor PK profiles. We
did not run rat PK on the in vitro optimized compound 33.
Instead it was tested directly in vivo for pharmacological activity
with measurement of oral exposure, which will be discussed
later.
Figure 5, the total AUC % change vs vehicle for 1 is -8.3 and -19
at the 1 mg/kg dose and 10 mg/kg dose, respectively. With a
similar PK profile as 1, 4 is also efficacious in vivo at the oGTT
model at a dose of 10 mg/kg. Surprisingly, compounds with low
clearance and higher oral exposure in PK, such as 5 and 9, are
slightly less potent in vivo despite higher compound
concentrations, which may be explained by the 3-4 fold potency
difference. There seems to be no correlation between PK profile
and in vivo efficacy. Therefore, the metabolically stable 33 was
tested directly in vivo with a 3- time point measurement of
plasma concentrations. Even though there is moderate amount of
compound measured at all three time points, 33 is less efficacious
than 1 at lowering blood glucose in the oGTT assay.
Table 6. Rat pharmacokinetic properties of selected compounds
Table 7. Normal Sprague Dawley rat oGTT results of selected compounds
a
b
po Cmax
(
po t1/2
(h)
Iv t1/2
Cl
Vdss
(L/kg)
Cpd
% F
43.2
62.3
72
ng/mL)
(h)
(mL/min/kg)
a
b
Doses
Total AUC % Change
c
Cpd
Conc. (M)
0.32  0.08
1.85  0.80
3.74  1.32
8.09  3.09
7.42  2.67
1
4
5
9
576
2.5
2.7
2.2
3.2
2.8
3.3
1.9
29.2
23.2
4.4
1.5
2.2
0.8
1.1
5.1
5.9
(mg/kg)
vs Vehicle
1
1
4
5
9
1
-8.3  2.9
1590
4220
2520
417
4.3
2.7
3.1
1.5
1.9
10
10
10
10
-19  3.2
-19  6.2
-17  2.4
-16  2.9
5.3
82.1
56.8
50.3
2
2
1
2
65
356
72
d
5
.66  1.85
e
a
b
1
0 mg/kg po dose formulated in 0.5% methocel; 2 mg/kg iv dose
3
3
10
-9  3
3.90  0.85
formulated in 20% HPCD.
f
4
.41  1.40
a
The rats were dosed compound (in 0.5% methocel) or vehicle 40 min before
b
Compound 1 was selected for an in vivo evaluation based on
its good rGPR40 activity (EC = 5 nM) and acceptable T (2.5
hour) for oral dosing. Oral administration of 1 in normal Sprague
Dawley (SD) rats 40 min prior to glucose challenge in an oral
glucose tolerance test (oGTT) significantly reduced blood
glucose excursion in a dose-dependent manner (Fig. 5). At a dose
of 10 mg/kg, 1 demonstrated comparable efficacy to TAK-875
with a much lower exposure (1.85  0.80 M) than TAK-875
they were given glucose (2 g/kg) orally; AUC: integrated area under the
glucose excursion curve from t=0 to t=120 minutes; Compound
c
5
0
1/2
concentration in plasma samples taken 160 minute post-compound dose;
d
e
Plasma samples were taken 35 min post-compound dose; Plasma samples
f
were taken 95 min post-compound dose; Plasma samples were taken 155 min
post-compound dose.
3
5
In conclusion, a series of tetrazoles with a benzo[b]thiophene
center ring were discovered as non-carboxylate GPR40 agonists.
Following SAR optimization targeting GPR40 agonist activity
and intrinsic clearance in microsomes (human and rat), potent
and metabolically stable compounds were selected for in vivo
evaluation. However, there was a disconnect between compound
exposure level and pharmacodynamic efficacy. Various reasons
could contribute to this phenomenon. For instance, high plasma
protein binding may result in less drug available to the target
tissue. Unfortunately, our effort to obtain rat plasma protein
binding data for the compounds failed due to their extremely long
equilibration times and high non-specific binding to the
equilibrium dialysis device. Nevertheless, tetrazole 1 has shown
good in vivo efficacy at lowering blood glucose in Sprague-
Dawley rats even at a low exposure. We have demonstrated that
non-carboxylate GPR40 agonists have the potential to be next
generation glucose lowering agents without the risk of forming
potentially toxic AG metabolites.
(
35.3  7.6 M).
Figure 5. Normal SD rat oGTT results of Cpd 1 with TAK-875 as a positive
References and notes
control.
1
2
.
.
Global Report on Diabetes. World Health Organization (WHO),
Geneva, 2016.
Amos, A. F.; McCarty, D. J.; Zimmet, P. Diabet. Med. 1997, 14 (Suppl.
Several other in vitro optimized compounds were also
tested in oGTT studies. Table 7 lists the in vivo efficacy of these
compounds in terms of total AUC (area under curve) percentage
changes vs. vehicle, which measures the compound’s ability to
lower blood glucose excursion within 2 hours post oral glucose
challenge. The concentrations of the compounds in plasma taken
5
): S1-85.
3
4
.
.
Bailey, C. J. Clin. Pharmacol. Ther. 2015, 98, 185.
Stein, S. A.; Lamos, E. M.; Davis, S. N. Expert Opin. Drug Saf. 2013,
12, 153.
5. McGarry, J. D.; Dobbins, R. L. Diabetologia. 1999, 42, 128.
1
60 min after compound dose are also shown here. Derived from

6
7
.
.
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. Nature, 2003, 422, 173.
Briscoe, C. P.; Tadayyon, M.; Andrews, J. L.; Benson, W.
G.; Chambers, J. K.; Eilert, M. M.; Ellis, C.; Elshourbagy, N.
A.; Goetz, A. S.; Minnick, D. T.; Murdock, P. R.; Sauls, H.
R.; Shabon, U.; Spinage, L. D.;Strum, J. C.; Szekeres, P. G.; Tan, K.
B.; Way, J. M.; Ignar, D. M.; Wilson, S.; Muir, A. I. J. Biol. Chem.
blood glucose and body weight. Rats are dosed with vehicle (0.5%
methocel) or compounds (10 mg/kg, po) 40 minutes prior to the oGTT
(glucose, 2g/kg, po). Blood is collected from the tail vein at 0, 10, 30,
60 and 120 minutes after glucose challenge to measure blood glucose;
The area under the curve for blood glucose excursion was calculated
from t= 0 to t = 120 minutes. Percent lowering of glucose was
calculated from the AUC data with respect to the vehicle-treated group.
2
003, 278, 11303.
8
9
.
.
Kaku; K.; Enya, K.; Nakaya, R.; Ohira, T.; Matsuno, R. Diabetes Obes
Metab. 2015, 17, 675.
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.; Shutt
leworth, S. J.; Lin, D. C. H. Bioorg. Med. Chem. Lett. 2012, 22, 1267.
th
1
0. Miller, A. R. Presented at the 76 American Diabetes Association
(ADA) Annual Meeting Scientific Sessions, New Orleans, LA, June
2
016; Poster 120-LB.
1
1
1. NCT01699737 on www. clinicaltrials.gov [internet]. 2013.
2. Shimada, T.; Ueno, H.; Tsutsumi, K.; Aoyagi, K.; Manabe, T.; Sasaki,
S.-Y.; Katoh, S. WO2009054479 2009.
1
1
1
3. Araki, T., Hirayama, M.; Hiroi, S.; Kaku, K. Diabetes Obes. Metab.
2
012, 14, 271.
4. Burant, C. F.; Viswanathan, P.; Marcinak, J.; Cao, C.; Vakilynejad, M.;
Xie, B.; Leifke, E. Lancet 2012, 379, 1403.
5. Otieno, M. A.; Snoeys, J.; Lam, W.; Ghosh, A.; Player, M. R.; Pocai,
A.; Salter, R.; Simic, D.; Skaggs, H.; Singh, B.; Lim, H. K. Toxicol. Sci.
2
017, 1-11.
1
6. Lassila, T.; Hokkanen, J.; Aatsinki, S.-M.; Mattila, S.; Turpeinen, M.;
Tolonen, A. Chem. Res. Toxicol., 2015, 28, 2292.
1
1
1
7. Fung, M.; Thornton, A.; Mybeck, K. Drug Inf. J. 2001, 35, 293.
8. Bailey, M.J.; Dickinson, R.G. Chem. Biol. Interact. 2003, 145, 117.
9. Negoro, K.; Iwasaki, F.; Ohnuki, K.; Kurosaki, T.; Yonetoku, Y.; Asai,
N.; Yoshida, S.; Soga, T. WO2007123225 2007.
2
0. Negoro, K.; Ohnuki, K.; Yonetoku, Y.; Kuramoto, K.; Urano, Y.;
Iwasaki, F. WO2010123017 2010.
2
2
2
1. Ge, M.; Lin, S.; Yang, L.; Zhou, C. WO2008054674 2008.
2. Ohkouchi, M. WO2013154163. 2013.
3. Butler, R. N. In Comprehensive Heterocyclic Chemistry II; Katritzky,
A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon: Oxford, 1996;
Vol. 4, p. 621.
2
2
4. Herr, R. J. Bioorg. Med. Chem. 2002, 10, 3379.
5. (a) Green, M. D.; Tephly, T. R. Drug Metab. Dispos. 1998, 26, 860. (b)
Huskey, S. W.; Doss, G. A.; Miller, R. R.; Schoen, W. R.; Chiu, S. L.
Drug Metab. Dispos. 1994, 22, 651.
2
2
6. Kuo,G.-H.; Player, M. R.; Yang, S.-M.; Zhang, Y.-M.; Huang, H.
WO2016057731 2016.
7. Brown, S. P.; Dransfield, P. J.; Vimolratana, M.; Jiao, X-Y.; Zhu, L.;
Pattaropong, V.; Sun, Y.; Liu, J.; Luo, J.; Zhang, J.; Wong, S.; Zhuang,
R.; Guo, Q.; Li, F.; Medina, J. C.; Swaminath, G.; Lin, D. C-H.; Houze,
J. B. ACS Med. Chem. Lett. 2012, 3, 726-730.
2
2
8. Hauge, M.; Vestmar, M. A.; Husted, A. S.; Ekberg, J. P.; Wright, M. J.;
Salvo, J. D.; Weinglass, A. B.; Engelstoft, M. S.; Madsen, A. N.;
Lückmann, M.; Miller, M. W.; Trujillo, M. E.; Frimurer, T. M.; Holst,
B.; Howard, A. D.; Schwartz, T. W. Mol. Metab. 2015, 4, 3-14.
9. Du, X.; Dransfield, P. J.; Lin, D. C-H.; Wong, S.; Wang, Y.; Wang, Z.;
Kohn, T.; Yu, M.; Brown, S. P.; Vimolratana, M.; Zhu, L.; Li, A-R.; Su,
Y.; Jiao, X.; Liu, J.; Swaminath, G.; Tran, T.; Luo, J.; Zhuang, R.;
Zhang, J.; Guo, Q.; Li, F.; Connors, R.; Medina, J. C.; Houze, J. B. ACS
Med. Chem. Lett. 2014, 5, 384-389.
3
3
0. Allen, F. H.; Groom, C. R.; Liebeschuetz, J.W.; Bardwell, D.A.; Olsson,
T.S.; Wood, P.A. J. Chem. Inf. Model. 2012, 52 (3), 857-866.
1. Srivastava, A.; Yano, J.; Hirozane, Y.; Kefala, G.; Gruswitz, F.; Snell,
G.; Lane, W.; Ivetac, A.; Aertgeerts, K.; Nguyen, J.; Jennings, A.;
Okada, K. Nature 2014, 513, 124-127.
3
3
2. Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol.
1
997, 267, 727-748.
3. The PyMOL Molecular Graphics System, Version 1.8 Schrödinger,
LLC.
3
3
4. Kennedy, L. J. Tetrahedron Lett. 2010, 51, 2010-2013.
5. Oral Glucose Tolerance Test (oGTT): Male SD rats (200-250 g) are
housed 2 per cage in a temperature-controlled room with a 12-hour
light/dark cycle. They are allowed ad libitum access to water and fed
with normal rodent chow. The night before the oGTT, the rats are
transferred to clean cages and fasted overnight. On the morning of the
oGTT, the rats are weighed and randomized into groups based on fasted