The design and synthesis of a potent glucagon receptor antagonist with favorable physicochemical and pharmacokinetic properties as a candidate for the treatment of type 2 diabetes mellitus

Article abstract of DOI:10.1016/j.bmcl.2013.03.014

A novel and potent small molecule glucagon receptor antagonist for the treatment of diabetes mellitus is reported. This candidate, (S)-3-[4-(1-{3,5- dimethyl-4-[4-(trifluoromethyl)-1H-pyrazol-1-yl]phenoxy}butyl)benzamido] propanoic acid, has lower molecular weight and lipophilicity than historical glucagon receptor antagonists, resulting in excellent selectivity in broad-panel screening, lower cytotoxicity, and excellent overall in vivo safety in early pre-clinical testing. Additionally, it displays low in vivo clearance and excellent oral bioavailability in both rats and dogs. In a rat glucagon challenge model, it was shown to reduce the glucagon-elicited glucose excursion in a dose-dependent manner and at a concentration consistent with its rat in vitro potency. Its properties make it an excellent candidate for further investigation.

Full text of DOI:10.1016/j.bmcl.2013.03.014

Bioorganic & Medicinal Chemistry Letters 23 (2013) 3051–3058
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The design and synthesis of a potent glucagon receptor antagonist
with favorable physicochemical and pharmacokinetic properties
as a candidate for the treatment of type 2 diabetes mellitus
⇑
Angel Guzman-Perez , Jeffrey A. Pfefferkorn, Esther C. Y. Lee, Benjamin D. Stevens, Gary E. Aspnes,
Jianwei Bian, Mary T. Didiuk, Kevin J. Filipski, Dianna Moore, Christian Perreault, Matthew F. Sammons,
Meihua Tu, Janice Brown, Karen Atkinson, John Litchfield, Beijing Tan, Brian Samas, William J. Zavadoski,
Christopher T. Salatto, Judith Treadway
Pfizer Wordwide Research and Development, Eastern Point Road, Groton, CT 06442, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel and potent small molecule glucagon receptor antagonist for the treatment of diabetes mellitus is
reported. This candidate, (S)-3-[4-(1-{3,5-dimethyl-4-[4-(trifluoromethyl)-1H-pyrazol-1-yl]phen-
oxy}butyl)benzamido]propanoic acid, has lower molecular weight and lipophilicity than historical gluca-
gon receptor antagonists, resulting in excellent selectivity in broad-panel screening, lower cytotoxicity,
and excellent overall in vivo safety in early pre-clinical testing. Additionally, it displays low in vivo clear-
ance and excellent oral bioavailability in both rats and dogs. In a rat glucagon challenge model, it was
shown to reduce the glucagon-elicited glucose excursion in a dose-dependent manner and at a concen-
tration consistent with its rat in vitro potency. Its properties make it an excellent candidate for further
investigation.
Received 8 January 2013
Revised 26 February 2013
Accepted 4 March 2013
Available online 13 March 2013
Keywords:
Glucagon receptor antagonist
Diabetes mellitus
Physicochemical properties
Ó 2013 Elsevier Ltd. All rights reserved.
Diabetes mellitus is a rapidly expanding public health problem
affecting approximately 347 million people worldwide.1 Type 2
diabetes mellitus is a metabolic disease in which fasting and post-
prandial plasma glucose is elevated due to abnormally high hepatic
glucose production, reduced glucose-stimulated insulin secretion
and insulin resistance.2 Despite there being several classes of drugs
in common clinical use, there is a significant need for new thera-
pies with improved safety and efficacy, in order to help patients
achieve appropriate glycemic control and avoid the long term com-
plications associated with the disease.3
nist, could reduce hepatic glucose production in type 2 diabetics,
resulting in improved glycemic control. This hypothesis is sup-
ported by both preclinical and clinical data.5
The identification of clinically useful small molecule glucagon
receptor antagonists for this class B G-protein coupled receptor
has been very difficult. Recently, Merck reported the discovery of
MK-0893 (1) as a clinical candidate, and several other related ser-
ies also containing a b-alanine side chain such as compound 2.6
Lilly has also recently presented phase 1 data for LY2409021, a glu-
cagon receptor antagonist of undisclosed structure.7 Most of the
reported glucagon receptor antagonists are lipophilic molecules
with relatively high molecular weight as exemplified by 1 and 2
(Fig. 1).8 In order to improve safety and biopharmaceutical proper-
ties, we focused our efforts on identifying a series of small mole-
cule glucagon receptor antagonists with reduced lipophilicity
(logD <3) and molecular weight (6500).9
Previously, we described the identification of a series of pyra-
zole ethers and amino pyrazoles, exemplified by compound 3, with
low molecular weight and logD.10 Unfortunately, despite their low
dog plasma clearance, these compounds suffered from high rat
plasma clearance which complicated rat toxicology studies and re-
duced the confidence in predictions of human pharmacokinetics.
Further, it was discovered that some of the favored compounds
from this series had undesired PPAR-d agonist activity (for exam-
ple, the PPAR-d EC50 for 3 is 440 nM). Consequently, we explored
Glucagon is a polypeptide hormone produced and secreted by
a-cells located in the endocrine portion of the pancreas. In the li-
ver, glucagon binding to the glucagon receptor triggers a signal
transduction cascade resulting in the stimulation of hepatic glu-
cose production through glycogenolysis and gluconeogenesis. In
type 2 diabetic patients, glucagon levels are inappropriately ele-
vated in the fasted state and not adequately suppressed in the
postprandial state, which contributes to the elevated hepatic glu-
cose production and plasma glucose levels seen in these patients.4
Consequently, it has been postulated that a reduction in glucagon
action in the liver, brought about by a glucagon receptor antago-
⇑
Corresponding author at present address: Therapeutic Discovery, Amgen Inc.,
360 Binney St., Cambridge, MA 02142, USA. Tel.: +1 617 444 5070.
E-mail address: aguzmanp@amgen.com (A. Guzman-Perez).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.014

3052
A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
O
O
O
O
N
H
OH
N
H
OH
OMe
N
N
NH
Cl
F
Cl
1
2
Cl
MW = 588
MW = 521
clogD = 3.9
clogD = 3.9
O
O
N
H
OH
O
CF3
N
N
3
MW = 475
clogD = 1.5
Figure 1. Structures of glucagon receptor antagonists 1–3.
whether polarity could be transposed to the aromatic ring most
distal from the acid. This led to the discovery of a new series of po-
lar and low molecular weight glucagon receptor antagonists with
improved properties.
The synthesis of the nitrogen-linked versions of these com-
pounds was accomplished as exemplified in Scheme 1. Aniline 5
was prepared in two steps starting from 4-fluoronitrobenzene
and 4-trifluoromethylpyrazole, which were coupled to afford inter-
mediate 4. The nitro functionality in 4 was reduced under transfer
hydrogenation conditions to provide 5 in quantitative yield.
Butyryl benzoate 6 was synthesized by reacting the Knochel-type
Grignard reagent prepared from ethyl-4-iodobenzoate with
butyraldehyde to afford the corresponding benzylic alcohol,11
which was subsequently oxidized under Parikh–Doering condi-
tions. Ketone 6 was an effective substrate for a reductive amination
with N-heterocyclic aniline 5 using decaborane to afford ester 7. It
is worth noting that decaborane consistently proved superior to
other reductive amination reagents for related transformations
NO2
NH2
NO2
H
N
a
b
N
+
N
N
N
N
CF3
F
CF3
CF3
5
4
CO2Et
CO2Et
c
d
I
O
6
O
CO2H
CO2R
N
H
f
HN
HN
N
N
N
N
CF3
CF3
9
7: R = Et
8: R = H
e
Scheme 1. Synthetic method for the synthesis of nitrogen-linked analogs. Reagents and conditions: (a) K2CO3, MeCN, 80 °C, 70%; (b) 10% Pd/C, EtOH, THF, NH4ꢀHCO2, quant.;
(c) (i) iPrMgClꢀLiCl, butyraldehyde, THF, ꢁ40 °C, (ii) TEA, SO3ꢀPyr., DMSO, CH2Cl2, 0 °C, quant.; (d) decaborane, MeOH, 0 °C; (e) 2:1 MeOH: 2.0 M aq LiOH, 66% for two steps;
and (f) (i) ethyl-b-alanineꢀHCl, EDCI, DIPEA, HOBt, THF, (ii) 2:1 MeOH: 2.0 M aq LiOH, 56%.

A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
3053
within this target class.12 Carboxylic acid 8 was prepared via base-
mediated hydrolysis of 7. A standard amide coupling with b-ala-
nine ethyl ester, followed by subsequent ester hydrolysis gave
the desired carboxylate 9. The racemic mixture was resolved by
chiral supercritical fluid chromatography to provide the separate
enantiomers.
or at the penultimate ester. Additionally, several analogs were pre-
pared by copper-catalyzed coupling (CuI/8-hydroxyquinoline) of
an aryl iodide with the appropriate heterocycle to provide interme-
diates similar to 16.17
The in vitro potency of test compounds was assessed using
both binding and functional assay formats. The binding affinity
of compounds for the human glucagon receptor was determined
by their ability to displace [125I] Glucagon-Cex from membranes
containing the human glucagon receptor,18 while functional activ-
ity was determined by the ability of these compounds to inhibit
glucagon-induced cAMP production in a cell line expressing the
human glucagon receptor. We were unable to perform these as-
says in the absence of serum to determine a free potency for
the antagonists, because of the high degree of non-specific bind-
ing exhibited by glucagon. Consequently, the binding assay was
performed in the presence of 0.2% bovine serum albumin (BSA),
while functional activity was determined in the presence of 4%
BSA in order to provide an estimate of the antagonists’ plasma
bound or total potency.
In addition to the molecular weight and polarity requirements
outlined above, we also set forth the following objectives: (a)
maintain a low in vivo clearance to obtain a moderate half-life de-
spite the low volume of distribution typical of acids, (b) accord-
ingly, target little to no turnover in microsomes, (c) maintain
permeability and solubility as high as possible to facilitate absorp-
tion, which can be a challenge for high molecular weight acids, and
(d) obtain a functional potency of <300 nM in the human cAMP as-
say in order to acquire an acceptable human dose prediction. For
reference, compound 1 had a Kb of 270 nM in our human glucagon
receptor (hGCGR) cAMP assay.
The corresponding ether linked compounds were synthesized
as exemplified in Scheme 2. Ketone 6 was reduced to the corre-
sponding (R)-alcohol 10 via transfer hydrogenation using an
(R,R)-Noyori catalyst.13 The 87% ee material from the Noyori
hydrogenation was then enriched to >99% ee by chiral supercritical
fluid chromatography. Hydrolysis of the ethyl ester to acid 11 en-
abled a standard amide coupling with b-alanine ethyl ester to ob-
tain alcohol 12. Separately, 4-bromo-2,6-dimethylaniline was
transformed to the corresponding hydrazine hydrochloride 13 via
diazotization and reduction with stannous chloride. The hydrazine
was converted to pyrazole 14 by base-promoted cyclization with a
b-trifluoromethyl vinamidinium salt, which was in turn synthe-
sized by a Vilsmeier–Haack reaction of 3,3,3-trifluoropropanoic
acid with phosphorus oxychloride in DMF at 70 °C followed by salt
isolation with sodium hexafluorophosphate.14 Aryl bromide 14
was then converted to phenol 15 via a palladium-catalyzed trans-
formation using tBuXPhos.15 A Mitsunobu reaction was then exe-
cuted to join enantioenriched alcohol 12 and phenol 15 to
deliver ether 16 in good yield. As expected, the reaction proceeded
with complete inversion of the stereocenter (>99% ee). Subsequent
ester hydrolysis provided the desired (ꢁ)-carboxylic acid 17.16 In
other instances, racemic alcohols similar to 10 were used to pro-
vide racemic analogs, which in several cases were resolved by chi-
ral supercritical fluid chromatography either at the final acid stage
O
O
O
O
OR1
N
H
OEt
OEt
a
c
OH
OH
O
1 = Et
11: R1 = H
6
12
10:
R
b
OH
Br
Br
d
f
g
N
N
R2
NH2
CF3
15
13:
·
R
2 = NH2NH2 HCl
N
e
14: R2
=
N
CF3
O
O
O
O
N
H
OH
N
H
OEt
h
O
O
N
N
N
N
CF3
CF3
17
16
Scheme 2. Synthetic method for the synthesis of ethers. Reagents and conditions: (a) (i) [N-[(1R,2R)-2-(amino-jN)-1,2-diphenylethyl]-4-methylbenzenesulfonamidato-
j
N]chloro[(1,2,3,4,5,6- )-1,3,5-trimethylbenzene]-ruthenium (2.5 mol %), 5:2 HCO2H/Et3N, 25 °C, 12 h, 99% yield, 86% ee, (ii) Chiralpak AD-H, SFC Mobile phase 80:20 CO2/
g
Methanol, 90% yield, 99% ee after resolution; (b) LiOHꢀH2O, THF/H2O/MeOH (1:1:1), 25 °C, 2.5 h, 93% yield, >99% ee; (c) b-alanine ethyl ester hydrochloride, HATU, DIPEA,
DMF, 25 °C, 1 h, 95%, >99% ee; (d) (i) HCl (concd)/H2O (1:2), 0 °C, 15 min then NaNO2, H2O, 0 °C, 30 min, (ii) SnCl2ꢀH2O; HCl (concd)/H2O (1:1), 0 °C, 1 h then 25 °C, 18 h, (iii)
HCl/CH3OH, EtOAc, 0 °C, 25 min, 64%; (e) (i) NaOMe, (Z)-N-(3-(dimethylamino)-2-(trifluoromethyl)allylidene)-N-methylmethanaminium hexafluorophosphate, 0–25 °C, 18 h,
(ii) TFA, 80 °C, 5 h, 70%; (f) KOH, Pd2dba3, tBuXPhos, dioxane/H2O (1:1), 95 °C, 1 h, 91%; (g) ADDP, Bu3P, THF, 25 °C, 16 h, 75% yield, >99% ee; and (h) 1 N aq NaOH, MeOH/THF
(1:1), 25 °C, 18 h, 87%, >99% ee.

3054
A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
The SAR of the nitrogen-linked analogs is summarized in Ta-
min/kg, respectively. It is likely that the main clearance pathway
of these compounds in rats is biliary excretion because: (a) there
is no observable turnover in rat liver microsomes, (b) little to no
drug was found in the urine of the rats, (c) efforts to detect glucu-
ronidation in UGT-supplemented rat liver microsomes provided
little to no turnover, (d) no plasma stability issues were detected,
and (e) following a 5 mg/kg intravenous administration of the most
potent enantiomer of 18, 57% of the dose was collected in the bile.
Interestingly, this may be rodent specific as the most potent enan-
tiomer of 18 displays a plasma clearance of 2 ml/min/kg after a
0.5 mg/kg intravenous administration to Beagle dogs, and of
151 ml/min/kg after a 1 mg/kg intravenous administration to mice.
The corresponding ether linked compounds were also investi-
gated and the results are summarized in Table 2. While the triflu-
oroimidazole ether (24) was significantly weaker than the
corresponding aniline, the trifluoromethylpyrazole ether (25) was
only twofold less potent than 19 in the functional assay. In addi-
tion, this compound possessed the desired polarity, acceptable pas-
sive permeability, no turnover in RLM, low turnover in HLM
(Clint,app 16 ml/min/kg) and good aqueous kinetic solubility at pH
ble 1. Incorporation of either a trifluoromethyl imidazole or trifluo-
romethyl pyrazole ring distal from the acid was tolerated as
observed in compounds 18 and 19. These compounds display mod-
est potency in both the binding and functional assays, display good
aqueous kinetic solubility at pH 6.5 (350–360 lM), and show no
observable turnover in human (HLM) or rat liver microsomes
(RLM). Importantly, as expected the incorporation of the heterocy-
cles results in these compounds having logD values (shake-flask
method using octanol/pH 7.4 buffer) in the range of interest. Both
of these compounds were separated into single enantiomers and
the most potent enantiomer of each was dosed intravenously to
Wistar–Han rats at 1 mg/kg to determine their pharmacokinetic
parameters. Unfortunately, the single enantiomers of compounds
18 and 19 had relatively high rat plasma clearances of 65 and
37 ml/min/kg, respectively.
Compounds 20–22 were prepared in an effort to eliminate the
aniline motif from these molecules, and to optimize potency and
reduce rat clearance. The trifluromethyl pyrazole group was cho-
sen for this purpose due to the higher passive permeability (Papp
)
observed for 19 over 18 using the MDCKII-LE assay,19 and its
slightly enhanced potency. While pyridine analog 20 provided
comparable potency and properties to 19, pyridine 21 was clearly
inferior. In contrast, pyrimidine analog 22 provided a moderate
improvement in potency while reducing the logD. This compound
6.5 (400 lM). Importantly, when 25 was separated into single
enantiomers and the most potent enantiomer dosed to Wistar–
Han rats at 1 mg/kg intravenously and 5 mg/kg orally, it provided
a clearance from plasma of 17 ml/min/kg and an oral bioavailabil-
ity of 60%. Given the favorable physicochemical, ADME and phar-
macokinetic properties of ether 25, we embarked on an effort to
improve on its potency within the ether series. Replacement of
the phenyl ring proximal to the pyrazole with a pyridine (26) or
pyrimidine ring (27) did not provide the desired improvements
in potency. This was particularly surprising in the pyrimidine case
since this had improved potency in the N-linked series.
also displays good kinetic solubility at pH 6.5 (380 lM) and no
turnover in human or rat microsomes. The dimethyl analog 23 pro-
vided a significant improvement in binding potency over 19 that
did not translate to the functional assay, and displayed moderate
turnover in rat liver microsomes, thus it was not investigated fur-
ther. Compounds 20 and 22 were separated into single enantio-
mers and the most potent enantiomer of each was used to
determine their rat plasma clearance, which was 74 and 130 ml/
Modification of the alkyl side chain of 25 did not provide the re-
quired potency improvement (Table 3). Compounds 28–31 pro-
Table 1
In vitro properties of the nitrogen-linked analogsa
O
O
N
OH
H
( )
HN
R
R
LogDb
hGCGR binding Ki (nM)
hGCGR cAMP Kb (nM)
MDCKII-LE Papp (ꢂ10ꢁ6 cm/s)
CF3
CF3
18
19
1.3
680
4700
0.6
N
N
N
1.9
2.1
1.3
1.3
160c
170
670
100
2600
1300
4100
490
4.0
4.6
0.8
0.8
N
N
N
N
CF3
CF3
CF3
N
20
21
22
N
N
N
N
N
N
CF3
23
N
2.2
12
1400
2.7
N
a
b
c
Binding and cAMP data is reported as the geometric mean of P2 determinations except where otherwise noted.
LogD values determined by shakeflask method.
Value reported is n = 1.

A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
3055
Table 2
In vitro properties of ethers 24–27a
O
O
N
H
OH
( )
O
R
R
LogDb
hGCGR binding Ki (nM)
hGCGR cAMP Kb (nM)
MDCKII-LE Papp (ꢂ10ꢁ6 cm/s)
CF3
CF3
24
1.6
220
22000
2.3
3.1
N
N
25
26
27
2.4
2.6
1.5c
890
230
5100
3700
N
N
N
N
N
N
CF3
CF3
N
3.4
N
N
1100
a
b
c
Binding and cAMP data is reported as the geometric mean of P2 determinations except where otherwise noted.
LogD values determined by shakeflask method.
Estimated using an in silico model.
vided a slight improvement in potency but at the expense of unac-
ceptable increases in lipophilicity, which were accompanied in
most cases by an increase in microsomal turnover. Compounds
32–34 resulted in weaker potency than 25 in the functional assay.
Interestingly, the ethyl analog 34, the one carbon truncation of 25,
resulted in no activity in the cAMP assay.
Our efforts to optimize the trifluromethylpyrazole group are
summarized in Table 4. Replacement of the trifluromethyl with a
chloro substituent resulted in a compound (35) with similar po-
tency and properties, with the exception of lower aqueous kinetic
of a methyl group at the 3-position (42) or fusion of a cyclohexyl
ring to the pyrazole (43) provided similar potency with higher lipo-
philicity. Benzo fusion to the pyrazole afforded compound 44 with
similar polarity and improved potency, but with moderate HLM
turnover. This compound was separated into enantiomers and
the most potent was dosed to Wistar–Han rats intravenously at
1 mg/kg and orally at 5 mg/kg. Unfortunately, although the com-
pound’s plasma clearance was 18 ml/min/kg, its oral bioavailability
was only 3%. This is presumably the result of a combination of mar-
ginal passive permeability (Papp 2.2 ꢂ 10ꢁ6 cm/s) and low solubil-
ity, which was difficult to quantify.
solubility at pH 6.5 (70 lM). Substitution with methyl or fluoro
provided compounds (36–37) with lower lipophilicity but reduced
potency in the cAMP assay. Remarkably, removal of the trifluoro-
methyl group as shown in compound 38 abolishes the functional
activity. Attempts to utilize polar groups in place of the trifluoro-
methyl (39–41) also resulted in no functional activity. Addition
Finally, we found (Table 5) that addition of two methyl groups
ortho to the pyrazole ring resulted in a robust increase in potency
in both the binding and functional assays (45). The (ꢁ)-(S)-enantio-
mer (17) was determined to be considerably more potent than (+)-
(R)-enatiomer (46). Surprisingly, the enhancement in cAMP po-
Table 3
Alkyl side chain SAR of the ether seriesa
O
O
N
H
OH
R
O
( )
N
N
CF3
R
LogDb
hGCGR binding Ki (nM)
hGCGR cAMP Kb (nM)
HLM Clint,app (ml/min/kg)
RLM Clint,app (ml/min/kg)
28
29
30
31
25
32
33
34
cyclo-Bu
iso-Bu
cyclo-pentyl
tert-Bu
n-Pr
iso-Pr
cyclo-Pr
Et
2.6
3.2
3.3
3.7
2.4
2.6
1.7
1.9
700
549
420
170
890
720
1300
1400
3000
3400
4000
4300
5100
8800
15000
>22000
26
23
22
31
16
13
<8
<8
<25
46
31
28
<36
<25
<25
<25
a
Binding and cAMP data is reported as the geometric mean of P2 determinations except where otherwise noted.
LogD values determined by shakeflask method.
b

3056
A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
Table 4
Heterocyclic ring SAR of the ether seriesa
O
O
N
H
OH
O
( )
R1
N
N
R2
R1
R2
LogDb
hGCGR binding Ki (nM)
hGCGR cAMP Kb (nM)
HLM Clint,app (ml/min/kg)
25
35
36
37
38
39
40
41
42
43
44
CF3
Cl
Me
F
H
CN
MeO
C(O)NMe2
CF3
H
H
H
H
H
H
H
H
2.4
2.6
2.1
2.0
1.7d
1.2
1.4
0.5
2.8
2.7
2.5
890
1100
1900
1700
4100c
4100
4400c
6500
1100
690
5100
6500
16
10
16
12
11000
10000
>55000c
>50000
>23000c
>30000
8300
<8
<8
<13
<10
48
Me
–CH2–CH2–CH2–CH2–
–CH@CH–CH@CH–
4200
2000
290
44
a
b
c
Binding and cAMP data is reported as the geometric mean of P2 determinations except where otherwise noted.
LogD values determined by shakeflask method.
Value reported is n = 1.
d
Estimated using an in silico model.
tency through addition of the two ortho-methyl groups was much
larger in this series than the corresponding one observed in the N-
linked series. It is also interesting to note that adding both methyl
groups is important because compound 47 with only one ortho-
methyl group afforded intermediate cAMP potency between 25
and 45.
Compound 17 has excellent properties. Its logD is essentially
the same as for 25 despite the presence of two additional methyl
groups. This is likely due to a conformational effect where the pyr-
azole is forced to be almost perpendicular to the adjacent phenyl
ring resulting in the N-2 of the pyrazole being more exposed to
the solvent surface than in 25. This additional polar surface area
essentially cancels out the added lipophilicity from the two methyl
groups. The absolute stereochemistry of 17 was confirmed to be (S)
by X-ray crystallography and examination of the Hooft parame-
ter.20 Compound 17 has acceptable passive permeability (Papp
3.6 ꢂ 10ꢁ6 cm/s), no observable turnover in HLM or RLM, and good
kinetic solubility at pH 6.5 (420 lM). In Caco-2 monolayers, it
demonstrated
a
Papp of 8.5 ꢂ 10ꢁ6 cm/s at pH 7.4 and of
30 ꢂ 10ꢁ6 cm/s at pH 6.5. The enhanced permeability at lower
pH is not surprising because the protonated form of the acid should
have higher permeability, and is likely more relevant due to the
acidic nature of the early gastrointestinal track. Its pharmacoki-
netic properties when dosed to rats and dogs are shown in Table 6.
It has low clearance and high oral bioavailability in both species.
The volume of distribution is low as expected and this results in
short to moderate half-lives. The molecular weight for compound
17 is 503, which together with a logD of 2.6 (clogD 2.3) met our
objectives. In contrast, compound 2 had an oral bioavailability of
7% and 30% after dosing at 5 mg/kg to rats and 4 mg/kg to dogs,
Table 5
Substitution at the phenyl ring of the ether seriesa
O
O
N
OH
H
O
R1
N
N
CF3
R2
R1
R2
LogDb
hGCGR binding Ki (nM)
hGCGR cAMP Kb (nM)
25
45
17
46
47
Racemic
Racemic
(ꢁ)-(S)
H
H
2.4
2.6c
2.5
2.7
2.6
890
64
14
1600
200
5100
770
270
9000
1900
Me
Me
Me
Me
Me
Me
Me
H
(+)-(R)
Racemic
a
b
c
Binding and cAMP data is reported as the geometric mean of P4 determinations except where otherwise noted.
LogD values determined by shakeflask method.
Estimated using an in silico model.

A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
3057
Table 6
Compound 17 pharmacokinetic propertiesa
Species
Plasma clearance (ml/min/kg)
Volume of distribution at steady state (L/kg)
Beta half-life (h)
Oral bioavailability (%)
Rat
Dog
2.8
1.5
0.40
0.31
1.8
2.5
100
100
a
Wistar–Han rats dosed at 1 mg/kg intravenously and 50 mg/kg orally. Beagle dogs dosed at 0.5 mg/kg intravenously and 5 mg/kg orally.
17
2
1
Figure 2. Heat map of the activity of compounds 17, 1 and 2 against a panel of pharmacological targets at 10 lM. Each column is a separate assay. Color gradient with
green = 0% effect, yellow = 50% effect and red = 100% effect.
say. Compound 17 afforded in the rat cAMP assay a Kb of
3900 nM, in the mouse cAMP assay a Kb of 1500 nM, in the dog
assay a Kb of 1200 nM, and in the monkey cAMP assay a Kb of
82 nM. As a result of the excellent rat pharmacokinetic properties
of 17 and despite its weak rat potency, it was possible to assess it
in an in vivo rat glucagon challenge model. In this study, Wistar–
Han rats (n = 6) were administered either 17 or vehicle orally fol-
VEH
200
180
160
140
120
100
0.75 mg/kg PO
7.5 mg/kg PO
75 mg/kg PO
lowed 45 min later by a 3 lg/kg intravenous challenge of gluca-
*
^
gon. At predetermined timepoints pre- and post-glucagon
administration, plasma samples were collected for the assessment
of glucose and plasma levels of 17. As shown in Figure 3, 17 sup-
pressed the glucagon induced glucose excursion in a dose propor-
tional manner, with the 75 mg/kg dose almost completely
blocking the glucose elevation. The 7.5 mg/kg dose provides
approximately 50% reduction in the glucose excursion. The bound
plasma exposure during the experiment of 17 at a dose of 7.5 mg/
kg was consistent in magnitude with the compound’s in vitro rat
Kb, demonstrating good rat in vitro–in vivo translation. The re-
sults from this model not only confirm the in vivo activity of
17, but also support our use of the in vitro cAMP assay at 4%
BSA as a good predictor of in vivo potency.
#
#
^
*
-10 0 10 20 30 40 50 60 70 80 90 100110120
IV glucagon administration
Time post-compound dose (min)
Figure 3. Effects of 17 dosed orally in the rat glucagon challenge model (N = 6,
⁄p <0.05, ^p <0.01, #p <0.001).
respectively. The low bioavailability may be, at least partially, due
In summary, we have identified compound 17 as a potent glu-
cagon receptor antagonist for the treatment of diabetes mellitus.
This compound has lower molecular weight and lipophilicity than
historical glucagon receptor antagonists, which results in excel-
lent selectivity in broad-panel screening, lower cytotoxicity, and
excellent overall in vivo safety in early pre-clinical testing. This
compound displays low in vivo clearance and excellent oral bio-
availability in both rats and dogs. In a rat glucagon challenge
model, compound 17 was shown to reduce the glucagon-elicited
glucose excursion in a dose-dependent manner and at a concen-
tration consistent with its rat in vitro potency. Taken together
the overall properties of 17 supported its selection as an early
development candidate for the treatment of type 2 diabetes
mellitus.
to the low kinetic solubility of compound 2 at pH 6.5 (23
Compound 17 has also excellent selectivity against a panel of 69
receptors, ion channels, uptake sites and enzymes tested at 10 M,
with >50% activity only identified for the rat chloride (GABA-gated)
channel (Ki = 8.2 M). This is in contrast with the numerous hits
lM).
l
l
identified when compounds 1 and 2 were screened at the same
concentration in similar panels. A comparison is best visualized
in the heat map in Figure 2. Compound 17 also displayed no ago-
nist activity at 10 lM when tested in PPAR-a, d, and c assays. It
has a binding Ki of 3160 nM for the GLP-1 receptor which provides
230-fold selectivity when compared to the glucagon binding Ki. Se-
lected compounds were also assessed in a transformed human li-
ver epithelial (THLE) cell cytotoxicity assay. This assay measures
ATP depletion as a determinant of general cellular toxicity at
72 h, and has been found to correlate with pre-clinical in vivo tox-
icity.21 Compound 17 had an IC50 of 112
toxicity than reference compounds
(IC50 = 55 M). Furthermore, compound 17 displayed an excellent
lM, suggestive of lower
Acknowledgements
1
(IC50 = 38 M) and
l
2
The authors would like to acknowledge Mei Li and David A.
Perry at Pfizer for strategic outsourcing. Additionally, we thank
Hong Li and the team at BioDuro in Beijing, China for their tech-
nical assistance running the glucagon challenge, and Jinfei Li,
Zhanshan Qian and the team at WuXi AppTec in Shanghai, China
for synthetic assistance. Finally, we thank Manthena Varma and
Charles Rotter for performing the Caco-2 permeability
determinations.
l
safety profile in 14-day exploratory toxicological studies in rats
and dogs. Together this data suggests that compound 17 may offer
a reduced safety risk profile than that of more lipophilic glucagon
receptor antagonists.
Compound 17, like most compounds investigated in this series,
showed weaker potency in the rat, mouse and dog glucagon
receptor functional assays than in the corresponding human as-

3058
A. Guzman-Perez et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3051–3058
11. Ren, H.; Krasovskiy, A.; Knochel, P. Org. Lett. 2004, 6, 4215.
References and notes
12. Bae, J. W.; Lee, S. H.; Cho, Y. J.; Yoon, C. M. J. Chem. Soc., Perkin Trans. 1 2000,
145.
13. Fujii, A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
1996, 118, 2522.
14. (a) Yamanaka, H.; Takekawa, T.; Morita, K.; Ishihara, T. Tetrahedron Lett. 1829,
1996, 37; (b) Davies, I. W.; Marcoux, J.-F.; Wu, J.; Palucki, M.; Corley, E. G.;
Robbins, M. A.; Tsou, N.; Ball, R. G.; Dormer, P.; Larson, R. D.; Reider, P. J. J. Org.
Chem. 2000, 65, 4571.
1. Danaei, G.; Finucane, M. M.; Lu, Y.; Singh, G. M.; Cowan, M. J.; Paciorek, C. J.; Lin,
J. K.; Farzadfar, F.; Khang, Y.-H.; Stevens, G. A.; Rao, M.; Ali, M. K.; Riley, L. M.;
Robinson, C. A.; Ezzati, M. Lancet 2011, 378, 31.
2. Saltiel, A. R. Cell 2001, 104, 517.
3. Krentz, A. J.; Patel, M. B.; Bailey, C. J. Drugs 2008, 68, 2131.
4. Dunning, B. E.; Gerich, J. E. Endocr. Rev. 2007, 28, 253.
5. (a) Ali, S.; Drucker, D. J. Am. J. Physiol. Endocrinol. Metab. 2009, 296, E415; (b)
Engel, S. S.; Xu, L.; Andryuk, P. J.; Davies, M. J.; Amatruda, J.; Kaufman, K.;
Goldstein, B. J. Diabetes 2011, 60, A85.
6. (a) Parmee, E. R. Abstracts of Papers, 241st National Meeting of the American
Chemical Society, Anaheim, CA; American Chemical Society: Washington, DC,
2011; Abstract MEDI-31.; (b) Xiong, Y.; Guo, J.; Candelore, M. R.; Liang, R.;
Miller, C.; Dallas-Yang, Q.; Jiang, G.; McCann, P. E.; Qureshi, S. A.; Tong, X.; Xu, S.
S.; Shang, J.; Vincent, S. H.; Tota, L. M.; Wright, M. J.; Yang, X.; Zhang, B. B.; Tata,
J. R.; Parmee, E. R. J. Med. Chem. 2012, 55, 6137; (c) Shen, D.-M.; Lin, S.; Parmee,
E. R. Exp. Opin. Ther. Patents 2011, 21, 1211.
15. Anderson, K. W.; Ikawa, T.; Tundel, R. E.; Buchwald, S. L. J. Am. Chem. Soc. 2006,
128, 10694.
16. Compound 17: 1H NMR (400 MHz, CDCl3) d: 7.88 (s, 1H), 7.73 (d, J = 8.2 Hz, 2H),
7.65 (s, 1H), 7.39 (d, J = 8.4 Hz, 2H), 6.80 (t, J = 5.9 Hz, 1H), 6.56 (s, 2H), 5.16 (dd,
J = 7.8, 5.1 Hz, 1H), 3.71 (q, J = 5.9 Hz, 2H), 2.69 (t, J = 5.8 Hz, 2H), 2.04–1.91 (m,
1H), 1.88 (s, 6H), 1.86–1.67 (m, 1H), 1.63–1.31 (m, 2H), 0.96 (t, J = 7.3 Hz, 3H).
MS (M+1): 504. Mp: 157–159 °C. ½aꢃD = ꢁ43.8 (c = 1; CHCl3). Chiral SFC.
Column: Chiralpak AD-H. Dimensions: 10 ꢂ 250 mm. Mobile phase: 80/20
CO2/2-propanol. Flow rate: 10.0 mL/min. Modifier: 0.2% isopropylamine.
Retention time: 3.23 min (first peak).
7. Kelly, R. P.; Garhyan, P.; Abu-Raddad, E. J.; Fu, H.; Lim, C. N.; Prince, M. J.;
Pinaire, J. A.; Loh, M. T.; Deeg, M. A. Diabetes 2011, 60, A84.
8. DeMong, D. E.; Miller, M. W.; Lachowicz, J. E. Annu. Rep. Med. Chem. 2008, 43,
119.
17. Filipski, K. J.; Kohrt, J. T.; Casimiro-Garcia, A.; Van Huis, C. A.; Dudley, D. A.;
Cody, W. L.; Bigge, C. F.; Desiraju, S.; Sun, S.; Maiti, S. N.; Jaber, M. R.; Edmunds,
J. J. Tetrahedron Lett. 2006, 47, 7677.
18. Li, P.; Rogers, R.; Smiley, D.; DiMarchi, R. D.; Zhang, F. Acta Crystallogr., Sect. F:
Struct. Biol. Cryst. Commun. 2007, 63, 599.
19. Di, L.; Whitney-Pickett, C.; Umland, J. P.; Zhang, H.; Zhang, X.; Gebhard, D. F.;
Lai, Y.; Federico, J. J., III; Davidson, R. E.; Smith, R.; Reyner, E. L.; Lee, C.; Feng, B.;
Rotter, C.; Varma, M. V.; Kempshall, S.; Fenner, K.; El-Kattan, A. F.; Liston, T. E.;
Troutman, M. D. J. Pharm. Sci. 2011, 100, 4974.
20. (a) Hooft, R. W. W.; Straver, L. H.; Spek, A. L. J. Appl. Crystallogr. 2008, 41, 96; (b)
Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
9. (a) Edwards, M. P.; Price, D. A. Annu. Rep. Med. Chem. 2010, 45, 381; (b) Leeson,
P. D.; Empfield, J. R. Annu. Rep. Med. Chem. 2010, 45, 393; (c) Hughes, J. D.; Blagg,
J.; Price, D. A.; Bailey, S.; DeCrescenzo, G. A.; Devraj, R. V.; Ellsworth, E.; Fobian,
Y. M.; Gibbs, M. E.; Gilles, R. W.; Greene, N.; Huang, E.; Krieger-Burke, T.; Loesel,
J.; Wagner, T.; Whiteley, L.; Zhang, Y. Bioorg. Med. Chem. Lett. 2008, 18, 4872; (d)
Price, D. A.; Blagg, J.; Jones, L.; Greene, N.; Wager, T. Exp. Opin. Drug Metab.
Toxicol. 2009, 5, 921.
10. Filipski, K. J.; Bian, J.; Ebner, D. C.; Lee, E. C. Y.; Li, J.-C.; Sammons, M. F.; Wright,
S. W.; Stevens, B. D.; Didiuk, M. T.; Tu, M.; Perreault, C.; Brown, J.; Atkinson, K.;
Tan, B.; Litchfield, J.; Pfefferkorn, J. A.; Guzman-Perez, A. Bioorg. Med. Chem. Lett.
2012, 22, 415.
21. Greene, N.; Aleo, M. D.; Louise-May, S.; Price, D. A.; Will, Y. Bioorg. Med. Chem.
Lett. 2010, 20, 5308.