Design and evaluation of a 2-(2,3,6-trifluorophenyl)acetamide derivative as an agonist of the GPR119 receptor

Article abstract of DOI:10.1016/j.bmcl.2011.01.088

The design and synthesis of a GPR119 agonist bearing a 2-(2,3,6- trifluorophenyl)acetamide group is described. The design capitalized on the conformational restriction found in N-β-fluoroethylamide derivatives to help maintain good levels of potency while

Full text of DOI:10.1016/j.bmcl.2011.01.088

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1306–1309
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and evaluation of a 2-(2,3,6-trifluorophenyl)acetamide derivative as
an agonist of the GPR119 receptor
⇑
Vincent Mascitti , Benjamin D. Stevens, Chulho Choi, Kim F. McClure, Cristiano R. W. Guimarães,
Kathleen A. Farley, Michael J. Munchhof, Ralph P. Robinson, Kentaro Futatsugi, Sophie Y. Lavergne,
Bruce A. Lefker, Peter Cornelius, Paul D. Bonin, Amit S. Kalgutkar, Raman Sharma, Yue Chen
Pfizer Global Research & Development, Groton Laboratories Eastern Point Rd, Groton, CT 06340, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The design and synthesis of a GPR119 agonist bearing a 2-(2,3,6-trifluorophenyl)acetamide group is
described. The design capitalized on the conformational restriction found in N-b-fluoroethylamide deriv-
atives to help maintain good levels of potency while driving down both lipophilicity and oxidative metab-
olism in human liver microsomes. The chemical stability and bioactivation potential are discussed.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 13 December 2010
Revised 19 January 2011
Accepted 20 January 2011
Available online 25 January 2011
Keywords:
GPR119
Diabetes
Gauche effect
Configurational entropy
Metabolism
S
GPR119 is a G-protein coupled receptor found in human enteric
L-cells and in pancreatic b-cells. Its role in glucose homeostasis has
been the subject of numerous studies that suggest agonism of
GPR119 may be a viable and highly desirable approach to the treat-
ment of type 2 diabetes.1,2 By way of a unique glucose-dependent
mechanism, GPR119 agonism is anticipated to promote glycemic
control by stimulating the secretion of insulin from the pancreas
as well as gastrointestinal hormones (e.g., the incretins GLP-1
and GIP) from the gut.3
N
N
O
N
N
N
N
N
1, MBX-2982
PhaseIIa
N
N
O
S
N
N
O
N
O
O
Ongoing research programs across the pharmaceutical industry
have led to the advancement of small molecule GPR119 agonists
into clinical trials (e.g., clinical candidates 1 and 2; Fig. 1).4 In this
Letter, we report the design and synthesis of compound 3 (Fig. 2), a
novel GPR119 agonist bearing an unprecedented 2-(2,3,6-trifluor-
ophenyl)acetamide group.
Initial interest in this series originated from the identification of
compound 4, which exhibited excellent activity in our c-AMP func-
tional assay (13 nM, 87% intrinsic activity; Table 1); however, 4
was highly lipophilic (E Log D = 4.3)5 which resulted in high oxida-
tive turnover in human liver microsomes (HLM). Interestingly, as
exemplified by compounds 5 and 6 (Table 1), replacement of the
2,3,6-trifluorophenyl group in 4 with more classical GPR119 phar-
macophores led to a decrease in potency. Systematic substitution
of any of the fluorine atom on the aromatic ring by a hydrogen
2, GSK1292263A
PhaseIIa
Figure 1. Structure of some known GPR119 agonists in the clinic.
F
F
H
N
F
F
O
N
O
3
F
O
O
F
Me
H
N
O
N
O
F
4
⇑
Corresponding author. Tel.: +1 860 686 3455; fax: +1 860 715 0310.
E-mail address: vincent.mascitti@pfizer.com (V. Mascitti).
Figure 2. 2-(2,3,6-Trifluorophenyl)acetamide GPR119 agonists.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.088

V. Mascitti et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1306–1309
1307
Table 1
Representative examples of analogs and their respective properties
R2
H
N
R3
O
N
O
R1
O
Entry
Compound
R1
R2a
R3
EC50 SDb nM
IA SDc
%
CLint (HLM)d (mL/min/kg)
E Log D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
3
4
5
6
7
1-MecPr
tert-Bu
tert-Bu
tert-Bu
tert-Bu
tert-Bu
tert-Bu
tert-Bu
tert-Bu
tert-Bu
i-Pr
1-MecPr
1-MecPr
i-Pr
1-MecPr
1-MecPr
F (S)
Me
Me
Me
Me
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
p-PhSO2Me
1-p-Ph-1H-tetrazole
2,3-Difluorophenyl
2,5-Difluorophenyl
2,6-Difluorophenyl
2-Fluorophenyl
80 41 (39)
13 8 (4)
884 (2)
267 235 (4)
75 (2)
78 (2)
120 28 (3)
808 (2)
107 10 (39)
87 12 (4)
47 (2)
72 14 (4)
107 (2)
104 (2)
101 9 (3)
97 (2)
42.4
230
136
>270
>300
>300
nd
>300
>300
247
205
>300
187
nd
3.4
4.3e
2.8e
3.1e
4.2e
4.2e
3.9
8
9
Me
Me (S)f
Me
10
11
12
13
14
15
16
17
21
4.1e
4.2e
4.0e
4.0
3.9
3.9
3.4
3.4
3.5
Me
Me
Me
3-Fluorophenyl
Phenyl
340 (2)
5980 (1)
102 (2)
100 (1)
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
2,3,6-Trifluorophenyl
32 14 (4)
44 15 (5)
236 44 (5)
409 285 (4)
1380 (1)
88 8 (4)
124 16 (5)
97 7 (5)
68 9 (4)
100 (1)
Me (S)f
Me (R)f
H
F (R)
H
63.8
nd
868 301 (3)
111 11 (3)
nd: not determined.
a
The configuration of the stereocenter is indicated in parentheses.
Potency in the human recombinant cell-based c-AMP functional assay. See online Supplementary data for more details. Number of runs indicated in parentheses. SD:
b
standard deviation of the mean (for n >2).
c
Intrinsic Activity in the c-AMP assay. Number of runs indicated in parentheses. SD: standard deviation of the mean (for n >2).
CLint refers to total intrinsic clearance obtained from scaling in vitro HLM half-lives.
Calculated Log D.
The configuration was arbitrarily assigned.
d
e
f
atom also led to compounds of increasingly suboptimal potency
(compare 4 vs 7–12). Furthermore, compounds 7–12 had high met-
abolic turnover in HLM as anticipated by their lipophilicity similar
to 4. Replacement of the acid labile tert-butyl carbamate by an iso-
propyl (i-Pr) or a 1-methylcyclopropyl (1-MecPr) carbamate also
led to compounds of similar lipophilicity and high clearance
in vitro (compounds 13–15).
influence on potency. Thus, although a productive interaction with
the receptor cannot be ruled out, it appeared reasonable to suggest
that the stereocenter helps reduce the configurational entropic
penalty upon binding.6,7 Mindful of the gauche effect found in N-
b-fluoroethylamides,8 we proposed the synthesis of conformation-
ally restricted compounds such as 3. The fluorine-containing ste-
reocenter was expected to positively impact the conformational
and/or vibrational component of the configurational entropy, and
thereby help maintain potency.7 In addition, this fluorine atom
could help reduce lipophilicity and metabolic turnover.
The presence of the methyl group (compare 13 vs 16) and the
configuration of the stereocenter (compare 14 vs 15) both had an
Compound 3 was synthesized in high yield and high enantiose-
lectivity as shown in Scheme 1.
dineacetaldehyde using MacMillan’s organocatalytic methodology
a-Fluorination of N-Boc-4-piperi-
O
N
Boc
N
Boc
Boc
N
N
Ph
N
1)
H
DPPA
THF, 23 °C
THF, i-PrOH, -15 °C
H
F
F
PhO2S
SO2Ph
76%
N
F
O
OH
N3
19
18
2) NaBH4, CH2Cl2
89 %
NBoc
F
1) TFA, CH2Cl2, 23 °C
1) Pd/C, H2, MeOH, 23 °C
3
HN
O2N
2) EDCI, HOBt, Et3N,
CH2Cl2, 23 °C
2)
O
O
F
O
O
F
F
THF, 60 °C
HO
O
F
89 %, 2 steps
F
F
20
86 %, 2 steps
Scheme 1. Synthesis of compound 3.
Figure 3. Averaged conformation of 3 in solution on the NMR timescale.

1308
V. Mascitti et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1306–1309
produced compound 18 (98% ee),9 which was then converted to
the -fluoro azide 19 in 76% yield via a Mitsunobu-type reaction.
Reduction of the azide to the amine followed by amide bond cou-
pling under classical conditions (EDCI/HOBt/Et3N) gave intermedi-
ate 20. This intermediate was converted to the desired product 3
by BOC removal and subsequent installation of the 1-MecPr
carbamate.10
Compound 3 proved to be a full agonist in our functional c-AMP
assay and its EC50 remained within two fold of the methyl analog
14. Importantly, 3 was markedly more stable in HLM incubations
than the corresponding methyl-substituted compounds 14 and
15 as well as less lipophilic.
OH
NH2
a
O
O
O
S
HN
OH
F
F
H
N
NH
OH
O
N
O
F
24
O
O
Figure 4. Representative structure of glutathione adduct 24.
We next sought to confirm the gauche effect hypothesis by
studying the conformation of compound 3 in solution. Indeed, as
depicted in Figure 3, NMR spectroscopic analysis (600 MHz/
CDCl3/300 K) showed that the N-b-fluoroethylamide favors adopt-
ing a gauche conformation, which constitutes about 75% of the con-
former population on the NMR timescale.11
glycine); m/z = 591 (loss of glutamate).14,15 A proposed structure
for the GSH conjugate of 3 that is consistent with the observed
mass spectrum is compound 24 shown in Figure 4.16 This GSH ad-
duct is postulated to occur via an ipso substitution of one of the
pendant fluorine atoms in the course of P450 catalyzed phenyl ring
epoxidation.17 The formation of 24 was NADPH-dependent, sug-
gesting the involvement of cytochrome P450 in the bioactivation
of 3. Also, inclusion of the specific CYP3A4 inhibitor ketoconazole
in the HLM incubations eliminated conjugate formation and over-
all metabolism, implicating that CYP3A4 was responsible for the
oxidative metabolism/bioactivation of 3.
In conclusion, we have described the design and synthesis of
compound 3, a novel GPR119 agonist bearing a N-b-fluoroethyla-
mide motif as key element of design. This group helped maintain
good agonist potency while reducing the lipophilicity and oxida-
tive metabolism in HLM in vitro. However, recognizing the poten-
tial for immune-mediated toxicity due to reactive metabolite
formation, further pharmacologic and pharmacokinetic profiling
of 3 was suspended.18
As observed previously in the methyl case (compounds 14 and
15), the configuration of the stereocenter in 3 had an impact on po-
tency in the functional c-AMP assay (compare 3 vs 17). Again, since
the topology of the receptor is not known, one can only speculate
around the causes of this selectivity. Although it could be the result
of the conformational bias induced by the stereocenter,11 a produc-
tive interaction between the fluorine atom of 3 and the receptor
(and/or an unfavorable one in the case of 17) cannot be ruled
out. The decrease in potency from 3 to 21 is also worth mentioning.
In this case, conformational analysis of simplified versions of com-
pounds 3 and 21 revealed energetically similar conformer distribu-
tions.11 The low energy conformers adopted by the fluoro-
substituted analog 3, including the most stable conformation
shown in Figure 3, are easily accessible to the unsubstituted analog
21. This suggests that the energy required to achieve the bioactive
conformation, whatever this might be, is not the main source of the
potency increase from 21 to 3. While the conformer distribution
does not change significantly, torsional scans for the dihedral an-
gles flanking the F atom revealed energy wells that are narrower
for the fluoro-substituted analog 3 compared to 21.11 Therefore,
it is conceivable that 3 experiences a less significant loss of vibra-
tional entropy upon binding and that this factor contributes to its
increase of potency over 21.7 However, in this case too, a produc-
tive interaction between the fluorine atom and the receptor cannot
be ruled out.
Supplementary data
Supplementary data (details related to the in vitro c-AMP func-
tional assay, NMR and computational studies around compound 3,
collision-induced dissociation spectrum of compound 24) associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2011.01.088.
References and notes
1. (a) Jones, R. M.; Leonard, J. N. Annu. Rep. Med. Chem. 2009, 44, 149; (b) Semple,
G.; Fioravanti, B.; Pereira, G.; Calderon, I.; Uy, J.; Choi, K.; Xiong, Y.; Ren, A.;
Morgan, M.; Dave, V.; Thomsen, W.; Unett, D. J.; Xing, C.; Bossie, S.; Carroll, C.;
Chu, Z.-L.; Grottick, A. J.; Hauser, E. K.; Leonard, J.; Jones, R. M. J. Med. Chem.
2008, 51, 5172.
In considering further profiling of 3, we were initially concerned
about the formation of a potentially reactive aziridine such as 22
under physiologic conditions (Scheme 2). However, exposure of 3
to various conditions expected to promote cyclization did not lead
to the formation of 22 or dihydrooxazole 23.12,13
Although the risk of aziridine formation appeared low, subse-
quent studies assessing the bioactivation potential of 3 in NADPH-
and glutathione (GSH)-supplemented HLM revealed a risk of reac-
tive species formation by metabolic activation. Formation of a GSH
adduct was clearly observed: m/z = 720 (MH+); m/z = 645 (loss of
2. Shah, U. Curr. Opin. Drug Discov. Dev. 2009, 12, 519.
3. Lauffer, L.; Iakoubov, R.; Brubaker, P. L. Endocrinology 2008, 149, 2035.
4. Jones, R. M.; Leonard, J. N.; Buzard, D. J.; Lehmann, J. Expert Opin. Ther. Patents
2009, 19, 1339. For MBX-2982 see: McWherter, C. Presented at the 32nd
Annual National Medicinal Chemistry Symposium, Minnesota, MN, June 6–9,
2010; oral session 2. For GSK1292263A see: Carpenter, A. J. Presented at the
32nd Annual National Medicinal Chemistry Symposium, Minnesota, MN, June
6–9, 2010; oral session 2. Peckham, G. E. Presented at the 240th National
Meeting of the American Chemical Society, Boston, MA, August 2010; poster
MEDI 199.
5. For a discussion around E Log D for lipophilicity determination see: Lombardo,
F.; Shalaeva, M. Y.; Tupper, K. A.; Gao, F. J. Med. Chem. 2001, 44, 2490.
6. Hoffmann, R. W. Angew. Chem., Int. Ed. 2000, 39, 2054.
7. Chang, C. A.; Chen, W.; Gilson, M. K. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 1534.
8. O’Hagan, D.; Bilton, C.; Howard, J. A. K.; Knight, L.; Tozer, D. J. J. Chem. Soc.,
Perkin Trans. 2 2000, 605.
9. Beeson, T. D.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127, 8826.
10. For the preparation of 1-methylcyclopropyl 4-nitrophenyl carbonate see:
Azimioara, M.; Cow, C.; Epple, R.; Jiang, S.; Lelais, G.; Mutnick, D.; Wu, B.
WO09105717, 2009.
11. See online Supplementary data.
12. Hudlicky, M. Israel J. Chem. 1978, 17, 80.
13. Some of the conditions tried were: (a) heating in DMSO for 30 min at
temperatures up to 200 °C in presence of i-Pr2NEt; (b) heating in DMSO for
60 min at temperatures up to 100 °C in presence of NaHMDS.
F O
O
N
N
N
O
22
23
F
F
3
X
F
F
O
O
O
N
F
Scheme 2. Attempts of intramolecular displacement of fluorine in 3.

V. Mascitti et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1306–1309
1309
14. Compound 3 (10
lM) was incubated in phosphate buffer (0.1 M, pH 7.4)
Post-column flow passed through the PDA detector to provide UV (k = 200–
containing MgCl2 (3.3 mM), HLM (2.0 mg/mL), NADPH (1.3 mM) and GSH
(5 mM). Reactions were initiated with the addition of microsomes. Control
incubations were run in parallel in the absence of NADPH and/or GSH. All
incubations were conducted in a shaking water bath maintained at 37 °C open
to the air. After 60 min, the incubations were terminated by addition of ice-
cold acetonitrile containing 0.1% formic acid, mixed vigorously, and the
precipitated materials were removed by spinning in a centrifuge (3000g) for
5 min. Aliquots of the supernatants were analyzed for metabolite formation by
liquid chromatography tandem mass spectrometry (LC–MS/MS). The HPLC
system consisted of an Accela quaternary solvent delivery pump and
400 nm) detection prior to being split to the mass spectrometer such that
mobile phase was introduced into the electrospray source at a rate of 50 lL/
min. The LC system was interfaced to a Thermo Orbitrap mass spectrometer
(Thermo Fisher Scientific, Bremen, Germany). Xcalibur software version 2.0
was used to control the HPLC/MS system. Mass spectroscopy analyses were
carried out in the positive ion mode using full-scan MS with a mass range of
100–1000 Da. Full scan data and data-dependent MS/MS acquisition on the
two most intense ions were collected at 15,000 resolutions. All experimental
data were acquired using external calibration prior to data acquisition.
15. For the collision-induced dissociation spectrum of compound 24, see
Supplementary data.
autoinjector,
Electron Corporation, Waltham, MA). Chromatography was performed on a
Phenomenex, Synergy RP column, 150 Â 4.6 mm, (Phenomenex,
Torrance, CA). LC analysis was performed at a constant flow rate of 1000 L/
a Surveyor PDA Plus photodiode array detector (Thermo
16. The regiochemistry of GSH attachment was not investigated. The
regiochemistry as shown in Figure 4 is arbitrary and for illustrative purposes
only.
5 lm
l
min using a binary solvent system: Solvent A, 5 mM ammonium formate buffer
(pH ꢀ3.0) with 0.1% formic acid and Solvent B, acetonitrile. The initial HPLC
gradient system was held at 5% B for 3 min and linearly increased to 80% B in
35 min, followed by a return to initial conditions for column re-equilibration.
17. Samuel, K.; Yin, W.; Stearns, R. A.; Tang, Y. S.; Chaudhary, A. G.; Jewell, J. P.;
Lanza, T., Jr.; Lin, L. S.; Hagmann, W. K.; Evans, D. A.; Kumar, S. J. Mass Spectrom.
2003, 38, 211.
18. Kalgutkar, A. S.; Soglia, J. R. Expert Opin. Drug Metab. Toxicol. 2005, 1, 91.