Synthesis and SAR of 1,2,3,4-tetrahydroisoquinolin-1-ones as novel G-protein-coupled receptor 40 (GPR40) antagonists

Article abstract of DOI:10.1016/j.bmcl.2009.03.082

The development of a series of novel 1,2,3,4-tetrahydroisoquinolin-1-ones as antagonists of G protein-coupled receptor 40 (GPR40) is described. The synthesis, in vitro inhibitory values for GPR40, in vitro microsomal clearance and rat in vivo clearance da

Full text of DOI:10.1016/j.bmcl.2009.03.082

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2400–2403
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of 1,2,3,4-tetrahydroisoquinolin-1-ones as novel
G-protein-coupled receptor 40 (GPR40) antagonists
*
Paul S. Humphries , John W. Benbow, Paul D. Bonin, David Boyer, Shawn D. Doran, Richard K. Frisbie,
David W. Piotrowski, Gayatri Balan, Bruce M. Bechle, Edward L. Conn, Kenneth J. Dirico,
Robert M. Oliver, Walter C. Soeller, James A. Southers, Xiaojing Yang
Pfizer Global R&D, Eastern Point Road, Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The development of a series of novel 1,2,3,4-tetrahydroisoquinolin-1-ones as antagonists of G protein-
coupled receptor 40 (GPR40) is described. The synthesis, in vitro inhibitory values for GPR40, in vitro
microsomal clearance and rat in vivo clearance data are discussed. Initial hits displayed high rat
in vivo clearances that were higher than liver blood flow. Optimization of rat in vivo clearance was
achieved and led to the identification of 15i, whose rat oral pharmacokinetic data is reported.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 4 February 2009
Revised 16 March 2009
Accepted 20 March 2009
Available online 25 March 2009
Keywords:
Diabetes
GPR40
Tetrahydroisoquinolinones
Ligand efficiency
Type 2 diabetes (T2D), the most commonly occurring form of
diabetes, is a condition in which the body resists the insulin that
is produced by the pancreas and may fail to make enough insulin
to maintain normal glucose levels.1 The incidence of T2D has in-
creased worldwide in recent years, largely because of growing
rates of obesity, and is likely to grow to greater than 366 million
by the year 2030.2
One of the recently characterized G-protein-coupled receptor
(GPCR) families is the GPR40–43 family,3 comprising GPR40, 41
and 43.4 These three family members share ꢀ 30–40% sequence
identity. Three independent groups have identified GPR40 as a
receptor for medium- (C6–C12) and long-chain (C14–C24) fatty
acids (FAs).5 GPR40 is preferentially expressed in the pancreas with
elevated levels reported in the islets and also in the pancreatic
b-cell lines.6 GPR40-deficient b-cells secrete less insulin in
response to FAs, and loss of GPR40 protects mice from obesity-in-
duced hyperinsulinemia, increased hepatic glucose output,
yperglycemia and glucose intolerance.7 Conversely, overexpression
of GPR40 in b-cells of mice leads to impaired cell function, hypoin-
sulinemia and diabetes. These results suggest that GPR40 plays a
critical role in linking obesity and T2D.
Figure 1. Initial hits from high-throughput screening.
fied compounds 1 and 2 (Fig. 1) as antagonists of GPR40. These hits
were extremely attractive based on their high ligand efficiency,8
low human liver microsomal (HLM) clearance and promising selec-
tivity in an initial selectivity panel. The above properties motivated
an initiation of hit-to-lead chemistry to explore the activity of this
class of compounds as GPR40 antagonists.
Initial efforts involved exploration of the SAR of the terminal
cyclohexyl and phenyl moieties of 1 and 2. In order to do this in
an efficient manner, a concise synthesis of phenol intermediate 7
was required. Intermediate 7 could then be utilized for late stage
diversification and efficient SAR exploration of this region of the
molecule.
Phenol 7 was accessed in a straightforward fashion following
the five step protocol below (Scheme 1). 4-Benzyloxybenzaldehyde
3 was condensed with propylamine to afford imine 4 in excellent
yield. Treatment of imine 4 with homophthalic anhydride resulted
in the formation of cis-lactam 5 in moderate yield along with minor
Efforts toward the identification of 1,2,3,4-tetrahydroisoquino-
lin-1-ones as novel GPR40 antagonists are reported here. A high-
throughput screen (HTS) of the Pfizer compound collection identi-
* Corresponding author. Tel.: +1 860 705 0559.
E-mail address: phumphri@gmail.com (P.S. Humphries).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.082

P. S. Humphries et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2400–2403
2401
Table 1
GPR40 activity, human and rat liver microsome data for antagonists 1, 2 and 8–10
Compound R/Ar/HetAr
GPR40 IC50
(nM)a
HLM Cl (
min/kg)
l
L/
RLM Cl (
min/kg)
lL/
1
2
8a
8b
9a
9b
—
—
Me
CH2Ph
3,5-Di-Me–Ph
3,4-Di-Me–Ph
10
12
2100
96
<12
<9
<8
<8
21
<8
274
97
<14
<16
46
2
3
<14
10a
29
<8
<14
a
Values are means of two or more independent experiments.
0
dent in vitro clearance of this series was required, while maintain-
ing our GPR40 potency and excellent human in vitro clearance.
Significant reduction in MW (e.g., 8a) resulted in a concomitant
drop-off in potency, but interestingly this compound possessed
low rat microsomal clearance. Benzyl substitution (e.g., 8b) re-
sulted in a compound with moderate GPR40 activity and low hu-
man and rat in vitro clearance. Biaryl ethers (e.g., 9b and 10a)
also provided excellent potency and in vitro clearance.
Scheme 1. Reagents and conditions: (a) C3H7NH2, 4 ÅA MS, CH2Cl2, rt, 16 h, 100%; (b)
homophthalic anhydride, CH3CN, 60 °C, 16 h, 50%; (c) AcOH, 120 °C, 16 h, 95%; (d)
MeI, K2CO3, Me2CO, rt, 16 h, 90%; (e) BBr3, CH2Cl2, rt, 16 h, 90%.
Compound 8b was then selected for rat pharmacokinetic (PK)
studies.15 Administration to male Sprague–Dawley (SD) rats sur-
prisingly resulted in iv clearance that was greater than liver blood
flow (82.8 mL/min/kg). In order to see whether this high rat iv
clearance held true across the series, a number of other compounds
were tested and all possessed high rat in vivo clearance.
In an effort to better understand the reason for the higher than
predicted in vivo rat clearance, four compounds were cassette
Table 2
Rat in vivo clearance and biliary excretion results
Structure
Rat iv Cl (mL/min/kg)
Percent dose in bile (%)
111
13.9
Scheme 2. Reagents and conditions: (a) ROH, PPh3, DIAD, THF, DMF, 80 °C, 16 h. (b0)
LiOH, THF, MeOH, H2O, rt, 16 h; (c) ArB(OH)2, Et3N, Cu(OAc)2, Py, 1,2-DCE, DMF, 4 ÅA
MS, air, 50 °C, 16 h; (d) 2-Br-HetAr, CuI (cat.), ligand 1 from ref.14 (cat.), K3PO4,
MeCN, 100 °C, 16 h.
84.5
22.9
69.9
64.6
amounts of trans-lactam 6.9 The thermodynamically less stable iso-
mer 5 could be epimerized under acidic conditions, resulting in the
exclusive formation of trans-lactam 6 in excellent yield.10 Esterifi-
cation of acid 6 with iodomethane afforded the intermediate
methyl ester, which was subsequently debenzylated utilizing bor-
on tribromide to yield the required phenol intermediate 7.11
Phenol 7 underwent a Mitsunobu reaction, followed by saponi-
fication, to afford required products 8 in a parallel fashion (Scheme
2).12 Diaryl ethers 9 were also accessed from phenol 7 utilizing ele-
gant methodology developed by Evans.13 SNAr reaction of phenol 7
with 2-bromoheteroaryls yielded products 10 via a copper-cata-
lyzed Ullman-type coupling.14
136
32.7
All the compounds were tested in a GPR40 FLIPR calcium mobi-
lization assay.5 Simultaneously, compounds were screened in hu-
man (HLM) and rat liver microsome (RLM) assays. The effect of
different terminal substitution (e.g., replacements for cyclohexyl
and phenyl moieties in 1 and 2) was studied first, and these results
are summarized in Table 1. Both compounds 1 and 2 have good
GPR40 activity and low human microsomal clearance, but high
rat microsomal clearance. Knowing that the desired pre-clinical
diabetic model study would be in a rat, improvements in the ro-
Scheme 3. Reagents and conditions: (a) MeOH, rt, 16 h then DMF, homophthalic
anhydride, rt, 16 h then 1 N aq NaOH, rt, 16 h.

2402
P. S. Humphries et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2400–2403
Table 3
Table 4
GPR40 activity and ligand efficiency for antagonists 2 and 15
GPR40 activity and ligand efficiency for antagonists 15
Compound
R1
R2
GPR40 IC50 (nM)a
RLM Cl
L/min/kg)
Compound
R1
R2
GPR40 IC50 (nM)a
RLM Cl
(lL/min/kg)
(
l
2
4-PhOꢀ-Ph
4-PhO-Ph
4-PhO-Ph
3,4-Di-Cl–Ph
3-Cl–Ph
Et
Me
CH2(c-C3H5)
C3H7
C3H7
12
156
4
89
228
97
27
120
<14
<14
15f
15g
15h
3,4-di-Cl–Ph
3,4-di-Cl–Ph
13
10
10
<14
27
15a
15b
15c
15d
CH2CH(Et)2
CH2CH(Et)2
<18
15e
C3H7
119
18
15i
20
<14
a
Values are means of two or more independent experiments.
a
Values are means of two or more independent experiments.
dosed in a bile duct cannulated (BDC) rat study (Table 2).15 A wide
variation in the percentage of dose found in the bile (even with
structurally similar compounds) was observed, which may indicate
that (i) biliary excretion is not a series-wide issue, and (ii) this set
of compounds suffers from high in vivo clearance for multiple
reasons.
In parallel to the above rat BDC study, one representative com-
pound, 12, was chosen for a dog iv PK study in order to assess
whether this was a rodent specific phenomenon.15 Acid 12 dis-
compounds that possessed low rat in vitro clearance and good
GPR40 activity (Table 4).
Based on good GPR40 activity, good in vitro permeability (CaCo-
2 AB = 12.0 and BA = 18.7 ꢁ 10ꢂ6 cm/sec), human in vitro clearance
(HLM Cl < 8.0
Cl < 14.1
lL/min/mg) and rat in vitro clearance (RLM
l
L/min/mg) 15i was selected for rat PK studies.15 Admin-
istration to male SD rats resulted in satisfactory PK parameters – iv
clearance = 4.3 mL/min/kg, Vd = 1.1 L/kg and iv half-life of 5.4 h.
In summary, a novel series of 1,2,3,4-tetrahydroisoquinolin-1-
ones have been identified as antagonists of GPR40. Initial hits dis-
played high rat in vivo clearances that were higher than liver blood
flow. Optimization of rat in vivo clearance was achieved and led to
the identification of 15i, which showed satisfactory PK parameters.
played low dog microsomal clearance (DLM Cl = 2.99 lL/min/mg)
and low dog in vivo clearance (7.4 mL/min/kg). The fact still re-
mained that we required access to a tool GPR40 antagonist that
displayed low rat in vivo clearance and thus needed to break into
different chemical space.
Readily accessible regions of the molecule were varied in paral-
lel to modulate metabolism and reduce clearance, while retaining
potency. In order to achieve this, optimization of the chemistry
in Scheme 1 was required in order to allow for it to be utilized in
a number of parallel arrays (Scheme 3). Two of the most immediate
challenges were the solubility of the reagents/products in the
existing solvents and the requirement for Dean-Stark conditions
on formation of the imine (e.g., 4) Solutions to these challenges in-
volved utilizing methanol as the solvent of choice for imine forma-
tion. The poor solubility of homophthalic anhydride necessitated
the use of DMF as the solvent for the formation of the cis- and
trans-lactams (e.g., 5 and 6). The resulting mixture was converted
to all trans-lactam (e.g., 6) by subjection to strong base, rather than
the original acidic conditions.
Initially, two libraries were executed utilizing this synthetic se-
quence. R1 was fixed as 4-phenoxyphenyl in 13 and the R2 moiety
was varied in 14. Following this, R2 was fixed as propyl in 14 and
the R1 moiety was varied in 13. The results from this exercise
can be seen in Table 3. Analysis of the amine diversity library re-
vealed that only five compounds had LE > 0.35 (e.g., 15b), which
led us to conclude that the original N-ethyl and N-propyl moieties
were already close to optimal. Interestingly, moving to N-methyl
(e.g., 15a) resulted in a significant loss of GPR40 activity, although
it displayed low rat in vitro clearance. Analysis of the aldehyde
diversity library revealed that 16 compounds had LE > 0.35. Mono-
aryl moieties at R1 (e.g., 15c and 15d) resulted in much improved
rat in vitro clearance with moderate GPR40 activity.
Acknowledgments
The authors gratefully acknowledge all our colleagues in the
GPR40 program for their technical support in the evaluation of
the compounds presented in this manuscript. We also thank Yue
Chen for the rat BDC study and Bruce Lefker and Joseph Warmus
for stimulating discussions and feedback on this manuscript.
References and notes
1. (a) Taylor, S. I. Cell 1999, 97, 9; (b) Jovanovich, L.; Gondos, B. Ann. Clin. Lab. Sci.
1999, 29, 33.
2. Wild, S.; Roglic, G.; Green, A.; Sicree, R.; King, H. Diabetes Care 2004, 27, 1047.
3. (a) Hirasawa, A.; Hara, T.; Katsuma, S.; Adachi, T.; Tsujimoto, G. Biol. Pharm. Bull.
2008, 31, 1847; (b) Bernard, J. Curr. Opin. Invest. Drugs 2008, 9, 1078; (c)
Telveker, V. N.; Kundaikar, H. S. Curr. Drug Targets 2008, 9, 899; (d) Costanzi, S.;
Neumann, S.; Gershengorn, M. C. J. Biol. Chem. 2008, 283, 16269; (e) Winzell, M.
S.; Ahren, B. Pharmacol. Ther. 2007, 116, 437; (f) Rayasam, G. V.; Tulasi, V. K.;
Davis, J. A.; Bansal, V. S. Exp. Opin. Ther. Targets 2007, 11, 661.
4. Sawzdargo, M.; George, S. R.; Nguyen, T.; Xu, S.; Kolakowski, L. F.; O’Dowd, B. F.
Biochem. Biophys. Res. Commun. 1997, 239, 543.
5. (a) Itoh, Y.; Kawamata, Y.; Harada, M. Nature 2003, 422, 173; (b) 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. 2003, 278, 11303; (c) Kotarsky,
K.; Nilsson, N. E.; Flodgren, E.; Owman, C.; Olde, B. Biochem. Biophys. Res.
Commun. 2003, 301, 406.
6. Turnbaugh, P. J.; Ley, R. E.; Mahowald, M. A.; Magrini, V.; Mardis, E. R.; Gordon,
J. I. Nature 2006, 444, 1027.
7. Steneberg, P.; Rubins, N.; Bartoov-Shifman, R.; Walker, M. D.; Edlund, H. Cell
Metab. 2005, 1, 245.
8. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
9. (a) Ryckebusch, A.; Garcin, D.; Lansiaux, A.; Goossens, J.-F.; Baldeyrou, B.;
Houssin, R.; Bailly, C.; Henichart, J.-P. J. Med. Chem. 2008, 51, 3617; (b) Morrell,
A.; Placzek, M.; Parmley, S.; Antony, S.; Dexheimer, T. S.; Pommier, Y.;
Cushman, M. J. Med. Chem. 2007, 50, 4419; (c) Ng, P. Y.; Tang, Y.; Knosp, W.
M.; Stadler, H. S.; Shaw, J. T. Angew. Chem., Int. Ed. 2007, 46, 5352; (d) Morrell,
A.; Antony, S.; Kohlhagen, G.; Pommier, Y.; Cushman, M. J. Med. Chem. 2006, 49,
7740.
At this point two low MW acids, 15d and 15e, were chosen for
rat PK studies, with both compounds displaying low rat in vitro
clearance.15 Administration of 15d and 15e to male SD rats re-
sulted in low iv clearance (15d—2.96 mL/min/kg, 15e—3.0 mL/
min/kg). This data provided evidence that low rat in vivo clearance
could be achieved within this series.
Finally, a full 2D matrix parallel array was executed where the
five most ligand efficient amines were crossed with the 20 most li-
gand efficient aldehydes. This exercise resulted in a number of
10. (a) Cushman, M.; Gentry, J.; Dekow, F. . J. Org. Chem. 1977, 42, 1111; (b)
Cushman, M.; Mohan, P. Tetrahedron Lett. 1985, 26, 4563.

P. S. Humphries et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2400–2403
2403
11. Paliakov, E.; Strekowski, L. Tetrahedron Lett. 2004, 45, 4093.
12. (a) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40, 2380; (b)
Mitsunobu, O. Synthesis 1981, 1; (c) Hughes, D. L. Org. React. 1992, 42, 335; (d)
Hughes, D. L. Org. Prep. Proc. Int. 1996, 28, 127.
13. (a) Evans, D. A.; Katz, J. L.; West, T. R. Tetrahedron Lett. 1998, 39, 2937; (b)
Decicco, C. P.; Song, Y.; Evans, D. A. Org. Lett. 2001, 3, 1029.
14. Cristau, H.-J.; Cellier, P. P.; Hamada, S.; Spindler, J.-F.; Taillefer, M. Org. Lett.
2004, 6, 913.
15. All in vivo procedures were approved by the Pfizer Groton Institutional Animal
Care and Use Committee, and principles of laboratory care (NIH Publication No.
85-23, revised 1985) were followed.