Discovery of cyclic guanidines as potent, orally active, human glucagon receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2011.09.085

In the course of the development of an aminobenzimidazole class of human glucagon receptor (hGCGR) antagonists, a novel class of cyclic guanidine hGCGR antagonists was discovered. Rapid N-dealkylation resulted in poor pharmacokinetic profiles for the benc

Full text of DOI:10.1016/j.bmcl.2011.09.085

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7131–7136
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of cyclic guanidines as potent, orally active, human glucagon
receptor antagonists
⇑
Christopher Sinz , Jiang Chang, Ashley Rouse Lins, Ed Brady, Mari Candelore, Qing Dallas-Yang,
Victor Ding, Guoqiang Jiang, Zhen Lin, Steven Mock, Sajjad Qureshi, Gino Salituro, Richard Saperstein,
Jackie Shang, Deborah Szalkowski, Laurie Tota, Stella Vincent, Michael Wright, Shiyao Xu,
Xiaodong Yang, Bei Zhang, James Tata, Ronald Kim, Emma Parmee
Discovery and Preclinical Sciences, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 29 September 2011
In the course of the development of an aminobenzimidazole class of human glucagon receptor (hGCGR)
antagonists, a novel class of cyclic guanidine hGCGR antagonists was discovered. Rapid N-dealkylation
resulted in poor pharmacokinetic profiles for the benchmark compound in this series. A strategy aimed
at blocking oxidative dealkylation led to a series of compounds with improved rodent pharmacokinetic
profiles. One compound was orally efficacious in a murine glucagon challenge pharmacodynamic model
and also significantly lowered glucose levels in a murine diabetes model.
Keywords:
Diabetes
Glucagon
Antagonist
Guanidine
Ó 2011 Elsevier Ltd. All rights reserved.
Glucagon and insulin are the primary hormones responsible for
regulating hepatic glucose production.1 In response to falling plas-
conversion of both isomers to the corresponding aminotetrazole
amides 6 and 7, we were pleased to find that both benzimidazole
6 and the cyclic guanidine isomer 7 showed high affinity for the
human glucagon receptor as measured by inhibition of 125I-gluca-
gon binding to hGCGR expressed in CHO cell membranes (hGCGR
IC50).6 Not only did compound 7 show enhanced binding affinity
(hGCGR IC50 = 4 nM) relative to isomer 6 (hGCGR IC50 = 19 nM),
but it also demonstrated a greater ability to block glucagon-induced
cAMP accumulation in hGCGR-transfected CHO cells (cAMP
IC50 = 23 nM; cf. cAMP IC50 = 670 nM for 6).6 This potent in vitro
activity prompted us to test the effects of compound 7 in vivo,
and to further develop this novel class of hGCGR antagonists.
Compound 7 showed excellent activity in a murine PD model.
Oral administration of 7 at 1 mg/kg to transgenic mice expressing
the human glucagon receptor 1 h prior to IP injection of glucagon
significantly inhibited glucagon-dependent glucose excursion
(Fig. 1).7 Robust pharmacodynamic activity was observed despite
a poor pharmacokinetic (PK) profile in the mouse (Table 1). While
7 showed an acceptable PK profile in the dog, poor oral exposure
was achieved in the mouse, the rat, and particularly in the rhesus.
Furthermore, poor metabolic stability was evident upon incubation
of compound 7 with human liver microsomes (0% parent com-
pound remaining after 30 min), with N-demethylation as the pri-
mary pathway for metabolism.8
ma glucose levels, glucagon is secreted from pancreatic
a-cells.
This hormone activates its receptor, a G-protein coupled receptor
located predominately in the liver, leading to increased hepatic
glucose production via glycogenolysis and gluconeogenesis. In
db/db2 and ob/ob3 mouse models of diabetes, antidiabetic effects
were observed upon reduction of glucagon receptor expression
with antisense oligonucleotides. In patients with type 2 diabetes,
inappropriately elevated glucagon levels lead to excessive hepatic
glucose output, a key contributing factor to hyperglycemia. Based
on this strong biological rationale, both non-small molecule ap-
proaches and small molecule glucagon receptor antagonists have
been pursued as potential treatments for type 2 diabetes.4 Further
interest was stimulated when Bayer reported the small molecule
BAY-27-9955 blocked glucagon-induced glucose production in
healthy human volunteers.5 In this Letter, we disclose a novel class
of cyclic guanidine human glucagon receptor antagonists.
A recent report by Merck co-workers detailed the discovery of an
aminobenzimidazole class of hGCGR antagonists.4b In the course of
that work, an isomeric structural class was discovered. Shown in
Scheme 1, the synthetic route to these aminobenzimidazoles in-
volved sodium hydride mediated alkylation of benzimidazoles such
as 3 with methyl 4-bromomethylbenzoate. In this case, a 1:1
mixture of alkylation regioisomers 4 and 5 was obtained. After
Incorporation of metabolically robust N3-substituents or steric
blockade of N3 oxidative metabolism pathways by introduction of
the substitution at C4 were each considered as potential strategies
for improving metabolic stability. Before either of these strategies
⇑
Corresponding author.
E-mail address: christopher_sinz@merck.com (C. Sinz).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.09.085

7132
C. Sinz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7131–7136
OCF3
NCS
2
Cl
Cl
NH
F3CO
Cl
Cl
N
N
NH
a
NH2
3
1
OCF3
OCF3
b
Cl
Cl
N
Cl
Cl
N
N
N
N
N
MeO2C
MeO2C
4
5
OCF3
OCF3
c
Cl
Cl
N
N
Cl
Cl
N
N
N
N
H
N
HN
HN
N
O
HN
N
O
N
N
N
N
6
7
Scheme 1. Reagents and conditions: (a) CH2Cl2, 40 °C, then Hg(OTFA)2, DMF, rt; (b) NaH, methyl 4-bromomethylbenzoate, DMF (c) (i) LiOH, dioxane/water, 40 °C;
(ii) 5-aminotetrazole, EDC, HOBt, DIEA, DMF.
Table 1
1500
Pharmacokinetic properties of compound 7
*
1250
1000
750
500
250
0
Species
Clp (mL/min/
kg)
Vd (L/
kg)
t1/2
(h)
PO AUCn
dose
(
l
M h mg/kg)/
F
(%)
a
Mousea
Ratb
18
16
1.6
1.2
0.9
0.2
1.6
1.6
2.2
1.1
0.09
0.14
4.6
5
8
27
N/A
Dogc
a
Rhesusd 15
0.29
<LOQ
a
Compound was dosed at 1 mpk IV, 2 mpk PO in a 15:35:55 mixture of DMSO,
water, and PEG400.
Compound was dosed at 1 mpk IV, 2 mpk PO in a 15:35:55 mixture of DMSO,
water, and PEG400.
Compound was dosed at 0.5 mpk IV in EtOH/PEG/water (2:5:3) and 1 mpk PO in
0.5% methylcellulose.
Compound was dosed at 0.5 mpk IV in EtOH/PEG/water (2:5:3) and 2 mpk PO in
0.5% methylcellulose.
a
b
c
d
Vehicle Glucagon
Cpd 7
Cpd 7
Cpd 7
1 mpk 3 mpk 10 mpk
* p <0.05 vs. vehicle
a p <0.05 vs. glucagon
Figure 1. Effect of compound 7 on glucagon-induced glucose excursion in hGCGR
mice. Compound was administered orally in 0.25% methylcellulose.
is reversed, providing cyclic guanidine 11 exclusively. Modification
at C4, for example by palladium-catalyzed aryl ether formation,
provides compounds such as ether 12.9 Saponification and EDC/
HOBt mediated coupling with 5-aminotetrazole then gives rise to
hGCGR antagonist 13.
The cyclic guanidines could also be synthesized by a one-pot, 3-
component coupling reaction of 2-bromobenzimidazoles with ben-
zylic bromides and anilines (Scheme 3). Toward this end, aniline 9
could be converted to benzimidazole 14 in two steps. Exposure of
14 to n-BuLi provides a 2-lithio intermediate which could be
quenched with bromine to generate 15. To effect the three-
component coupling, thermal reaction of 15 with methyl 4-
bromomethylbenzoate provides a benzimidazolium ion which
could be evaluated efficiently, an improved synthesis was neces-
sary; specifically, a method for selective benzimidazole alkylation
was required (Scheme 2). Toward this end, a variety of alkylation
conditions were screened, utilizing benzimidazole 10, which is
available in three steps from nitroaniline 8. For substrates bearing
a C4-substituent, sodium hydride or cesium carbonate mediated
alkylation heavily favors undesired reaction at the exocyclic nitro-
gen. However, upon thermal reaction of 10 with methyl 4-
bromomethylbenzoate in the absence of base, this regioselectivity

C. Sinz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7131–7136
7133
Br
H
Br
c
N
NH2
NO2
a, b
F3C
NH2
F3C
9
8
OCF3
OCF3
Br
Br
N
N
e
d
N
N
N
F3C
NH
F3C
11
10
MeO2C
OCF3
OCF3
O
O
N
N
N
N
N
f
N
F3C
F3C
H
N
MeO2C
HN
N
O
12
13
N
N
Scheme 2. Reagents and conditions: (a) NaH, MeI, DMF, rt; (b) SnCl2ꢀ2H2O, DMF/water, 40 °C (c) 4-trifluoromethoxyphenyl isothiocyanate, CH2Cl2, 40 °C, then Hg(OTFA)2,
DMF, rt; d. methyl 4-bromomethylbenzoate, CH3CN, 110 °C (e) n-propanol, Cs2CO3, Pd(OAc)2, 2-(di-tert-butylphosphino-1,10-binaphthyl, toluene, 70 °C; (f) (i) LiOH, dioxane/
water, 40 °C; (ii). 5-aminotetrazole, EDC, HOBt, DIEA, DMF.
Br
O
H
N
a, b
c
N
N
F3C
NH2
F3C
14
9
CF3
O
O
N
N
N
N
O
Br
d
N
F3C
N
N
F3C
Br
F3C
15
MeO2C
MeO2C
16
17
Scheme 3. Reagents and conditions: (a) formic acid, 100 °C; (b) n-propanol, Cs2CO3, Pd(OAc)2, 2-(di-tert-butylphosphino-1,10-binaphthyl, toluene, 70 °C (c) (i) n-BuLi, THF,
ꢁ78 °C; (ii) Br2, THF; (d) (i) methyl 4-bromomethylbenzoate, CH3CN, 110 °C; (ii) 4-trifluoromethylaniline, CH3CN.
can undergo addition/elimination with anilines to provide cyclic
guanidines such as 17.
substituents (as in compounds 20 and 21) led to reduced in vitro
activity.
Table 2 summarizes the effect of the N3-substituent (R) on met-
abolic stability, binding affinity, and functional activity (cAMP
IC50). Increasing steric bulk of the N3-substituent (R) indeed ap-
peared to block N-dealkylation, as compound 18 showed minimal
turnover after 60 min incubation with human liver microsomes.
This modification, however, led to reduced in vitro potency which
was further diminished as the steric demand of the N3-substituent
(R) increased. Further homologation of the N3-substituent in
compound 19 gave comparable potency, while branched or polar
We then turned our attention to a strategy aimed at using steric
effects to prevent oxidative N-dealkylation, by introduction of
substitution at C4. To set the stage for this approach, a brief survey
of phenylenediamine ring substituents was conducted (Table 3).
Introduction of 4-n-propoxy group (compound 22) led to a 10-fold
loss in functional activity, while 6-Cl and 6-CF3 analogs, 23 and 25
maintained most of the binding and functional activity of com-
pound 7. We therefore targeted 4,6-disubstituted analogs, to
examine whether this substitution pattern provided compounds

7134
C. Sinz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7131–7136
Table 2
Survey of nitrogen substituents
N
N
N
O
HN
N
H
OCF3
Cl
Cl
N
N
N
R
Compound
R
hGCGR BNDa IC50 (nM) (n)
0.6 (2)
18 6.4 (2)
14
64
160
hGCGR cAMPa IC50 (nM) (n)
HLM (% parent at 60 min)
7
Me
Et
Pr
iPr
4
23 3.6 (7)
73 12 (5)
61 16 (3)
210 27 (2)
640 300 (2)
0
18
19
20
21
95
40
Nd
Nd
CH2CH2OMe
a
Each IC50 in this Letter is reported as an average rounded to two significant figures standard error of the mean when more than one measurement was made.
Table 3
Effect phenylenediamine ring substituents
N
N
N
O
HN
N
H
OCF3
6
5
N
N
R
N
4
Compound
R
hGCGR BNDa IC50 (nM) (n)
hGCGR cAMPa IC50 (nM) (n)
22
23
24
25
26
27
4-OnPr
6-CF3
6-OCF3
6-Cl
6-OnPr
5-Cl
42 21 (2)
13 2.3 (3)
24
18
55
38
250 130 (2)
49 12 (8)
180 53 (2)
96 7.7 (2)
1200 480 (2)
440 91 (2)
a
Each IC50 in this Letter is reported as an average rounded to two significant figures standard error of the mean when more than one measurement was made.
with in vitro potency similar to compound 7, but with enhanced
metabolic stability.
Incorporation of alkoxy substituents at the 4-position proved to
be much more beneficial (Table 4). While incorporation of 4-OMe
groups as in compounds 34 and 35 led to moderate binding and
functional potency, incorporation of larger alkoxy substituents
led to enhanced in vitro activity. Consistent with our hypothesis
of steric blockade of N3-oxidative metabolism, these larger C4-
substituents engendered enhanced metabolic stability. In human
liver microsomes, compounds bearing n-propoxy or n-butoxy sub-
stituents at C4 all demonstrated improved metabolic stability
(>40% parent compound remaining after 60 min incubation with
human liver microsomes), while compounds bearing smaller
C4-substituents (34 and 35) showed more modest metabolic sta-
bility. The combination of an n-propoxy substituent at R2 and a
trifluoromethyl group at R1 (e.g., compound 13) appeared to
provide an optimal combination of binding activity, functional
activity, and metabolic stability. A survey of R3 substituents
A series of compounds bearing a 6-CF3 substituent and aliphatic
substituents at C4 was prepared (Table 4). Increasing steric
demand at the R2 position gave incremental enhancements in po-
tency, with compound 30 displaying greater potency than the ini-
tial lead 7. A similar trend emerged with respect to the R3
substituent on the N2-aromatic ring, with sterically demanding
groups (OCF3, t-Bu), providing greater potency than smaller sub-
stituents, (cf. compounds 32 and 33). Despite the enhanced activity
in vitro, and improved rodent PK profiles (vide infra) displayed by
these 4-alkyl substituted analogs, these compounds showed only
modest activity in vivo. Compounds 29 and 32 each required a
10 mg/kg dose in order to effect a significant blockade of gluca-
gon-dependent glucose excursion in the hGCGR transgenic murine
PD model (data not shown).

C. Sinz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7131–7136
7135
Table 4
SAR of 4,6-disubstituted cyclic guanidines
N
N
N
O
HN
N
H
3
4
R3
2
R1
N
N
N
R2
Compound
R1
R2
R3
hGCGR BNDa IC50 (nM) (n)
hGCGR cAMPa IC50 (nM) (n)
HLM (% parent at 60 min)
28
29
30
31
32
33
34
35
36
37
13
38
39
40
41
42
43
44
45
46
47
CF3
CF3
CF3
CF3
CF3
CF3
CF3
Cl
CF3
Cl
CF3
Cl
CF3
Cl
Cl
CF3
CF3
CF3
CF3
CF3
CF3
Me
Et
nBu
Et
Et
4-OCF3
4-OCF3
4-OCF3
4-Me
4-t-Bu
4-F
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-OCF3
4-t-Bu
4-CF3
30 1.4 (2)
11 1.4 (2)
1.7 0.6 (2)
80
4.4 0.4 (2)
110
8.3 1.3 (2)
12 0.6 (2)
2.8 0.5 (3)
7.0 1.2 (2)
1.3 0.4 (2)
8.3 0.7 (2)
3.1 1.1 (2)
130 17 (2)
22 1.7 (3)
7.6 1.7 (3)
230 50 (2)
17 2.3 (3)
1200 490 (2)
41 1.9 (2)
73 25 (2)
14.4 1.3 (3)
19 2.9 (4)
16.3 3.1 (4)
19 2.7 (3)
28 8.5 (4)
25 3.5 (4)
29 2.7 (3)
13 1.1 (2)
14 1.8 (9)
18 2.5 (4)
22 6.9 (3)
19 5.5 (2)
14 3.7 (2)
Et
OMe
OMe
OEt
15
18
34
42
60
46
76
52
OEt
OnPr
OnPr
OnBu
OnBu
OiPr
OnPr
OnPr
OnPr
OnPr
OnPr
OnPr
7
0.9 (2)
9.3 0.3 (2)
2.2 0.1 (2)
58
31
6
1.6 (2)
3-OCF3
3-CF3
3-CF3, 4-Me
3-CF3, 4-F
7.0 1.4 (2)
31 9.8 (3)
14 4.6 (2)
7.7 0.4 (2)
a
Each IC50 in this Letter is reported as an average rounded to two significant figures standard error of the mean when more than one measurement was made.
Table 5
Selectivity of representative compounds
Compound
hGLP-1a cAMP IC50 (nM) (n)
hGIPa cAMP IC50 (nM) (n)
IKr bindinga MK499 IC50 (nM) (n)
13
38
42
43
44
45
4700
6700
5800
2600
2300
2300
370 72 (7)
510
290
1500
1600
2200 61 (2)
3000
280 14 (2)
>10000
250 5.8 (2)
570 11 (2)
1100
a
Each IC50 in this Letter is reported as an average rounded to two significant figures standard error of the mean when more than one measurement was made.
(compounds 44–47) provided compounds with reduced binding
affinity, but nearly equivalent functional activity to compound
13. While these compounds showed largely similar activity on
the human glucagon receptor, they exhibited distinct off-target
profiles.
Off-target profiles of selected compounds are shown below
(Table 5). Selectivity for the human glucagon receptor over the
closely related human GLP-1 (hGLP-1) and human GIP (hGIP)
receptors were evaluated, as these receptors have important roles
in glucose-dependent insulin secretion. These compounds typically
Table 6
Pharmacokinetic properties of select compounds
Compound
Species
Clp (mL/min/kg)
Vd (L/kg)
t1/2 (h)
PO AUCn
(lM h mg/kg)/dose
F (%)
13
38
43
43
Rata
Rata
Rata
Dogb
29
24.2
39
6.6
1.9
9.5
0.56
2.9
1.9
3.6
4.7
0.33
0.18
0.17
5.0
36
25
24
35
1.6
a
Compounds were dosed at 1 mpk IV, 2 mpk PO in a 5:10:85 mixture of DMSO, Tween 80, and water.
Compound was dosed at 0.5 mpk IV in EtOH/PEG/water (2:5:3) and 1 mpk PO in 0.5% methylcellulose.
b

7136
C. Sinz et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7131–7136
in an hGCGR murine PD model. Oral administration of 13 to hGCGR
mice 1 h prior to a glucagon challenge resulted in significant block-
ade of glucose excursion, at doses as low as 1 mg/kg (Fig. 2). To
demonstrate efficacy in a murine diabetes model, oral administra-
tion of 43 to ob/ob mice expressing only the human glucagon
receptor at doses as low as 1 mg/kg led to significant reduction
in ambient glucose levels over a 6 h period (Fig. 3).
We report herein a novel, orally active cyclic guanidine class of
hGCGR antagonists. Compound 7, the initial lead for this series, dis-
played promising in vitro potency and efficacy in vivo but suffered
from unacceptable PK profiles in preclinical species and poor hu-
man liver microsome stability, largely due to N-demethylation. A
strategy aimed at improving metabolic stability by introducing ste-
ric bulk near the site of metabolism resulted in a series of 4,6-
disubstituted cyclic guanidine analogs which demonstrate vastly
improved metabolic stability, good rodent PK profiles, and excel-
lent pharmacodynamic activity. Highlighting the potential of cyclic
guanidine human glucagon receptor antagonists, compound 43 ef-
fected robust blockade of glucagon induced glucose excursion in
murine PD model, and also effected significant reduction in ambi-
ent glucose levels in ob/ob/hGCGR efficacy model.
Figure 2. Effect of compound 43 on glucagon-induced glucose excursion in hGCGR
mice. Compound was administered orally in 0.25% methylcellulose.
References and notes
1. Inhibition of hepatic glucose production as an approach to type 2 diabetes has
been reviewed: (a) Kurukulasuriya, R.; Link, J. T.; Pei, Z.; Rohde, J. J.; Richards, S.
J.; Souers, A. J.; Szczepankiewicz, B. G. Curr. Med. Chem. 2002, 10, 99; (b)
Kurukulasuriya, R.; Link, J. T.; Madar, D.; Pei, Z.; Richards, S. J.; Rohde, J. J.;
Souers, A. J.; Szczepankiewicz, B. G. Curr. Med. Chem. 2002, 10, 123; For a recent
review of glucagon antagonism for type 2 diabetes, see: (c) Ali, S.; Drucker, D. J.
Am. J. Physiol. Endocrinol. Metab. 2009, 296, E415; For
a review of small
molecule glucagon antagonists, see: (d) Sloop, K. W.; Michael, M. D.; Moyers, J.
S. Expert Opin. Ther. Targets 2005, 9, 593.
2. Liang, Y.; Osborne, M. C.; Monia, B. P.; Bhanot, S.; Gaarde, W. A.; Reed, C.; She,
P.; Jetton, T. L.; Demarest, K. T. Diabetes 2004, 53, 410.
3. Sloop, K. W.; Cao, J. X.-C.; Siesky, A. M.; Zhang, H. Y.; Boenmiller, D. M.; Cox, A.
L.; Jacos, S. J.; Moyers, J. S.; Owens, R. A.; Showalter, A. D.; Brenner, M. B.; Raap,
A.; Gromada, J.; Berridge, B. R.; Moneteith, D. K. B.; Porksen, N.; McKay, R. A.;
Monia, B. P.; Bhanot, S.; Watts, L. M.; Doson, M. J. Clin. Invest. 2004, 113, 1571.
4. (a) Some aspects of this work have recently been presented: Kim, R.; Chang, J.;
Lins, A. R; Sinz, C. J.; Beconi, M.; Brady, E. J.; Candelore, M. R.; Ding, V.; Jiang, G.;
Lin, Z.; Qureshi, S.; Salituro, G.; Saperstein, R.; Shang, J.; Tota, L.; Wright, M.;
Yang, X.; Tata, J.; Zhang, B.; Parmee, E. Abstracts of Papers, 232nd National
Meeting of the American Chemical Society: San Francisco, CA, 2006; American
Chemical Society: Washington, DC, 2009; MEDI-294; (b) Kim, R. M.; Chang, J.;
Lins, A. R.; Brady, E.; Candelore, M. R.; Dallas-Yang, Q.; Ding, V.; Dragovic, J.;
Iliff, S.; Jiang, G.; Mock, S.; Qureshi, S.; Saperstein, R.; Szalkowski, D.;
Tamvakopoulos, C.; Tota, L.; Wright, M.; Yang, X.; Tata, J. R.; Chapman, K.;
Zhang, B. B.; Parmee, E. R. Bioorg. Med. Chem. Lett. 2008, 18, 3701; (c) Madsen,
P.; Kodra, J. T.; Behrens, C.; Nishimura, E.; Jeppesen, C. B.; Pridal, L.; Andersen,
B.; Knudsen, L. B.; Valcare-Aspegren, C.; Guldbrandt, M.; Christensen, I. T.;
Jørgensen, A. S.; Ynddal, L.; Brand, C. L.; Bagger, M. A.; Lau, J. J. Med. Chem. 2009,
52, 2989.
Figure 3. Acute lowering of ambient glucose levels by compound 43 in an hGCGR/
ob/ob mouse model. Compound was administered orally in 0.5% methylcellulose.
showed >250-fold selectivity over the hGLP-1 receptor and 20 to
100-fold selectivity over the hGIP receptor. Compounds 13 and
38, which differ only by their C6-substituents, exhibited similar
selectivity profiles. Aniline substitution had a more dramatic im-
pact on selectivity, with a 4-tert-butyl (compound 42) substituent
resulting in poor selectivity over hGIP and IKr. Interestingly, while
the 3-trifluoromethoxy and 3-trifluoromethyl substituents on
compounds 44 and 45 engendered potent IKr binding, the 4-trifluo-
romethyl analog 43 displayed excellent selectivity.10
Compounds 13, 38, and 43 all displayed acceptable selectivity,
and were chosen for further profiling in vivo. The PK profiles for
these selected compounds in preclinical species are outlined in Ta-
ble 6. In all cases, C4-substitution led to enhanced oral bioavailabil-
ity relative to the benchmark compound 7, while compounds 13 and
43 also displayed longer half-lives. Compound 43 was also found to
have a favorable PK profile in the dog (t1/2 = 4.7 h, F = 35%).
5. Petersen, K. F.; Sullivan, J. T. Diabetologia 2001, 44, 2018.
6. Qureshi, S. A.; Candelore, M. R.; Xie, D.; Yang, X.; Tota, L. M.; Ding, V. D.-H.; Li,
Z.; Bansal, A.; Miller, C.; Cohen, S.; Jiang, G.; Brady, E.; Saperstein, R.; Duffy, J. L.;
Tata, J. R.; Chapman, K. T.; Moller, D. E.; Zhang, B. Diabetes 2004, 53, 3267.
7. (a) Shiao, L.-L.; Cascieri, M. A.; Trumbauer, M.; Chen, H.; Sullivan, K. A.
Transgenic Res. 1999, 8, 295; (b) Dallas-Yang, Q.; Shen, X.; Strowski, M.; Brady,
E.; Saperstein, R.; Gibson, R. E.; Szalkowski, D.; Candelore, M. R.; Fenyk-Melody,
J. E.; Parmee, E. R.; Zhang, B. B.; Jiang, G. Eur. J. Pharmacol. 2004, 501, 225.
8. For metabolic stability studies, test compounds (1 lM) were incubated with
microsomal proteins (0.25 mg/mL) containing NADPH (1 mM) for 1 h at 37 °C.
9. Torraca, K. E.; Huang, X.; Parrish, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 2001,
123, 10770.
10. IKr binding data were obtained by measuring displacement of
[
35S]-
radiolabeled MK-499 from HEK cells stably expressing hERG: Lynch, J. J., Jr.;
Wallace, A. A.; Stupienski, R. F., III; Baskin, E. P.; Beare, C. M.; Appleby, S. D.;
Salata, J. J.; Jurkiewicz, N. K.; Sanguinetti, M. C.; Stein, R. B.; Gehret, J. R.;
Kothestein, T.; Claremon, D. A.; Elliott, J. M.; Butcher, J. W.; Remy, D. C.;
Baldwin, J. J. J. Pharmacol. Exp. Ther. 1994, 269, 541.
Based on its potent hGCGR antagonist activity, good selectivity
and favorable rodent PK profile, compound 43 selected for testing