Calcitonin gene-related peptide (CGRP) receptor antagonists: Investigations of a pyridinone template

Article abstract of DOI:10.1016/j.bmcl.2007.11.042

In our effort to find potent, orally bioavailable CGRP receptor antagonists for the treatment of migraine, a novel series based on a pyridinone template was investigated. After optimizing the privileged structure and the placement of the attached phenyl r

Full text of DOI:10.1016/j.bmcl.2007.11.042

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 755–758
Calcitonin gene-related peptide (CGRP) receptor
antagonists: Investigations of a pyridinone template
Diem N. Nguyen,a,* Daniel V. Paone,a Anthony W. Shaw,a Christopher S. Burgey,a
Scott D. Mosser,b Victor Johnston,b Christopher A. Salvatore,b Yvonne M. Leonard,c
Cynthia M. Miller-Stein,c Stefanie A. Kane,b Kenneth S. Koblan,b Joseph P. Vacca,a
Samuel L. Grahama and Theresa M. Williamsa
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bDepartment of Pain Research, Merck Research Laboratories, West Point, PA 19486, USA
cDepartment of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486, USA
Received 10 October 2007; revised 9 November 2007; accepted 12 November 2007
Available online 17 November 2007
Abstract—In our effort to find potent, orally bioavailable CGRP receptor antagonists for the treatment of migraine, a novel series
based on a pyridinone template was investigated. After optimizing the privileged structure and the placement of the attached phenyl
ring, systematic SAR was carried out on both the N-alkyl and C-5 aryl substituents. Several analogs with good potency and phar-
macokinetic profiles were identified.
Ó 2007 Elsevier Ltd. All rights reserved.
Calcitonin gene-related peptide (CGRP) is a 37-amino
acid neuropeptide that is expressed in trigeminal ganglia
nerves. Activation of trigeminal sensory nerves and
CGRP release play a central role in the pathophysiology
of migraine.1,2 Current standard treatment for migraine
is the triptan class of 5-HT1B/1D receptor antagonists.
Though effective, these triptans are contraindicated in
patients with cardiovascular disease.3,4 When dosed iv,
the potent CGRP receptor antagonist BIBN4096BS
(olcegepant) was revealed in clinical trials to have simi-
lar efficacy against migraine attacks to the triptans but
without the cardiovascular or other serious ancillary ef-
fects.5,6 In this context, our program focused on finding
orally bioavailable CGRP receptor antagonists for the
treatment of migraine headache.
R1
N
O
R1
N
O
O
O
N
N
H
H
R2
R2
Figure 1.
aryl group, and urea-linked substituent was a key con-
tributing factor to the potency of the compound. When
exploring alternative templates, we reasoned that use of
a central pyridinone substructure would allow for a pla-
nar, spatial orientation of the appended substituents
similar to the caprolactam. In addition, the achiral
pyridinone core would greatly simplify synthesis, thus
facilitating rapid SAR studies. The inherent rigidity
and the degree of unsaturation in the pyridinone hetero-
cycle could improve pharmacokinetic profiles versus the
caprolactams due to less oxidative metabolism. Herein
we describe the syntheses and the inhibitory and phar-
macokinetic profiles of compounds from this novel
pyridinone series.
In prior work, the discovery of the amino caprolactam
template gave rise to a series of lower molecular weight,
non-peptide CGRP receptor antagonists7,8 (Fig. 1). Be-
cause the trans isomer was significantly more potent
than the cis isomer, we thought that perhaps the pseu-
do-equatorial arrangement of the N-alkyl side chain,
We assessed the potency of these compounds using re-
combinant human CL-receptor/RAMP1 in a [125I]-
CGRP radioligand binding assay (Ki). Selected com-
Keywords: CGRP; CGRP receptor antagonist; Migraine.
*
Corresponding author. Tel.: +1 215 652 9925; fax: +1 215 652
3971; e-mail: diem_nguyen@merck.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.042

756
D. N. Nguyen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 755–758
R1
N
pounds were subsequently tested for their functional
ability to inhibit CGRP-stimulated cAMP production
in whole cells (cAMP IC50). The functional assay can
also be run in the presence of 50% human serum to give
an indication of the extent of protein binding.
1. R1-X, Cs2CO3, DMF
(50-60%)
N
OH
O
Br
NO2
Br
NH2
2. SnCl2, EtOAc
(60-75%)
4
5
O
R1
N
O
COCl2, THF, Et3N
Initially we wanted to determine the optimal position of
the aryl group for potency (Fig. 2). The incorporation of
the phenyl ring at the C-6 position (1) resulted in mod-
erate CGRP receptor affinity (Ki = 3400 nM). However,
shifting the phenyl to the C-5 position (2) increased po-
tency by 43-fold (Ki = 80 nM). Additionally, this analog
O
NH
N
N
N
O
N
H
NH
HN
N
Br
7
N
6
(65-75%)
O
R1
N
O
was
a
potent functional antagonist (cAMP
O
NH
N
R2-B(OH)2, P(
m
-SO3PhNa)3
-Pr2NH
N
N
IC50 = 360 nM). This result correlated well with the cap-
rolactam series where the C-6 phenyl compound was sig-
nificantly more potent than the C-7 phenyl analog.7
Although the phenylimidazolinone privileged structure9
provided the best potency in the caprolactam series,
additional studies indicated that this substructure was
chemically unstable due to air oxidation of the imidaz-
olinone ring.8,10 As an alternative, the azabenzimidazo-
lone was shown to be stable and therefore was used in
N
H
Pd(OAc)2,
i
R2
DMF/H2O, 80 oC
8
(50-60%)
Scheme 1.
boronic acids were not commercially available, the bro-
mide intermediate 7 could be converted to the corre-
sponding boronic acid12 prior to coupling with
different aryl bromides.
the pyridinone series to provide compound
3
(Ki = 170 nM). Though the binding affinity of 3 was 2-
fold less than 2, its functional potency was slightly better
(cAMP IC50 = 302 nM).
Initial SAR studies were aimed at optimizing potency by
varying aryl groups at the C-5 position while keeping a
fixed N-methyl group (Table 1). The parent phenyl ana-
log 9 had similar potency in both binding and functional
assays (Ki = 498 nM and cAMP = 495 nM). Com-
pounds 10–13 had low to moderate binding affinity at
the CGRP receptor (Ki = 2 lM–9 lM). In general,
4-substituted phenyls were not well tolerated with the
exception of the 4-hydroxyphenyl analog (16) (Ki =
43 nM, cAMP IC50 = 88 nM). Although 2-hydroxy-
phenyl analog 14 was not as active as 16, 3-hydroxy-
phenyl derivative 15 had similar activity in both the
intrinsic binding and cAMP assay (Ki = 42 nM, cAMP
IC50 = 142 nM). 4-Methoxyphenyl 17, differing only by
a methyl group from 16, suffered large losses in potency
(Ki = 495 nM, cAMP IC50 = 1528 nM). The 3-pyridyl
compound 18 was potent in the binding assay
(Ki = 270 nM) but not as active in the functional assay
(cAMP IC50 = 1066 nM). Comparing 5-membered
heterocycles, the 3-pyrazole analog (19) was 29-fold
more potent (Ki = 71 nM, cAMP IC50 = 184 nM) than
the 4-imidazole analog (12), while the 3-thiophene deri-
vative (20) was potent in both the binding and func-
tional assays (Ki = 75 nM, cAMP IC50 = 330 nM).
Unlike the caprolactam series, where the 2,3-difluoro-
phenyl derivative was 10-fold more potent than either
the 2-fluorophenyl or 3-fluorophenyl analog indepen-
dently,8 compounds 21, 22, and 23 had similar potencies.
Though compound 1 was synthesized according to a lit-
erature procedure,11 most of the pyridinone compounds
described were synthesized as shown in Scheme 1. 5-Bro-
mo-3-nitropyridin-2-ol (4) was first converted to pyridi-
none 5 by N-alkylation with alkylhalides using cesium
carbonate as base to minimize O-alkylation. The nitro
group was then reduced to the amine with tin(II) chlo-
ride at ambient temperature. Urea coupling between 5
and azabenzimidazolone piperidine 6 was accomplished
using phosgene and afforded 7 in good yield. Bromide 7
underwent Suzuki coupling with various substituted
arylboronic acids to arrive at final compounds 8. The
water-soluble ligand 3,30,300-phosphinidynetris(benzene-
sulfonic acid) trisodium salt used in the palladium-cata-
lyzed coupling helped facilitate the isolation of the
compounds, as most of the reactions were simply loaded
directly onto a reverse phase purification system without
filtration. For desired compounds in which the aryl
O
O
O
NH
N
N
N
6
R1
N
H
5
R2
1 R1 = Ph, R2 = H; Ki = 3400 nM
2 R1 = H, R2 = Ph; Ki = 80 nM, cAMP IC50 = 360 nM
O
In an effort to derive more potency, different N-alkyl
substituents were explored. Two of the more potent
groups based on the caprolactam series were 2-methoxy-
ethyl and 2,2,2-trifluoroethyl. Each of the two groups
was integrated with the more potent aryl groups from
the SAR study noted above. With the N-2-methoxyethyl
derivatives, the phenyl analog 24 exhibited 2-fold better
binding affinity (Ki = 228 nM) compared to 9, though it
O
O
NH
N
N
N
N
N
H
3
Ki = 170 nM, cAMP IC50 = 302 nM
Figure 2.

D. N. Nguyen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 755–758
Table 1. SAR at N-1 alkyl and C-5 aryl position of the pyridinones
757
O
R1
N
O O
NH
NH
N
N
N
R2
a
b
Compound
R1
R2
Ki (nM)
cAMP IC50
(nM)
cAMP + 50% human
b
serum IC50 (nM)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
CH3
CH3
Phenyl
498
9000
1900
2050
1750
220
42
495
—
10,900
—
—
4-Benzoic acid
2,6-Pyrimidine
4-Imidazole
CH3
CH3
—
—
—
—
CH3
CH3
4-(Methanesulfonyl)phenyl
2-Hydroxyphenyl
3-Hydroxyphenyl
4-Hydroxyphenyl
4-Methoxyphenyl
3-Pyridyl
—
827
142
88
—
CH3
CH3
715
—
43
CH3
CH3
495
270
71
1528
1066
184
330
477
337
567
665
540
1190
307
170
164
209
345
651
550
282
186
119
18
—
—
CH3
CH3
3-Pyrazole
3-Thiophene
872
—
75
CH3
CH3
2-Fluorophenyl
3-Fluorophenyl
2,3-Difluorophenyl
Phenyl
213
155
166
228
390
695
140
74
—
—
CH3
—
—
2-Methoxyethyl
2-Methoxyethyl
2-Methoxyethyl
2-Methoxyethyl
2-Methoxyethyl
2-Methoxyethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
2,2,2-Trifluoroethyl
3-Thiophene
3-Pyridyl
3-Pyrazole
—
—
394
280
1004
—
4-Hydroxyphenyl
2,3-Difluorophenyl
Phenyl
41
95
3-Thiophene
3-Pyridyl
3-Pyrazole
105
635
205
96
—
—
1421
971
—
4-Hydroxyphenyl
2-Fluorophenyl
3-Fluorophenyl
2,3-Difluorophenyl
54
42
—
811
17
a Values are means of at least two determinations.
b The coefficient of variation of the cell-based assays is <0.3.
had similar potency in the functional assay (cAMP
IC50 = 665 nM). Other analogs in this group were either
similar or slightly less potent compared to their
N-methyl counterparts with the exception of 29 which
was 4-fold more potent than 23 in the binding assay
(Ki = 41 nM). In general, the SAR of the N-2,2,2-trifluo-
roethyl analogs did not correlate with those of the
N-methyl analogs. Phenyl derivative 30 was 5-fold more
potent than 9 in the binding assay. Although the 3-pyr-
idyl derivative (32) was 2-fold less potent in the binding
assay (Ki = 635 nM) compared to its N-methyl counter-
part, it was 2-fold better in the functional assay (cAMP
IC50 = 651 nM). The 3-pyrazole (33) and the 4-hydroxy-
phenyl (34) analogs were approximately 3-fold less po-
tent than 19 and 16, respectively. The 2-fluorophenyl
analog (35) and 3-fluorophenyl analog (36) were both
more potent than 30. As was observed within the capro-
lactam series, the 2,3-difluorophenyl group added more
potency than the 2-fluorophenyl or 3-fluorophenyl
groups did separately.8 Compound 37 had excellent po-
tency in both CGRP binding and functional cAMP as-
says (Ki = 17 nM, cAMP IC50 = 18 nM). Several
compounds were also tested in the cAMP assay in the
presence of 50% human serum. In general, serum shifts
increased with increasing lipophilicity of the C-5 aryl
group (difluorophenyl 37 = 45-fold shift), whereas they
decreased with increasing polarity of these aryl groups
(pyrazole 27 = 1.3-fold shift, hydroxyphenyl 28 = 1.6-
fold shift). When comparing amongst the N-alkyl
groups, the N-methyl compounds show the greatest ser-
um shifts while the N-2-methoxyethyl analogs generally
show lower shifts.
Pharmacokinetic profiles of selected compounds were
obtained in either rat or dog (Table 2). All the com-
pounds tested had low to moderate clearances and rea-
sonable iv half-lives for an acute migraine indication
(1–6 h). The N-methyl derivatives had low oral bioavail-
ability with the exception of the phenyl analog (9) (rat
F = 40%) and the 3-thiophene analog (20) (rat
F = 47%). The N-2-methoxyethyl derivatives generally
had higher clearances which may have contributed to
lower bioavailability with the exception of 24 (rat
F = 37%). Several compounds of the N-2,2,2-trifluoro-
ethyl group had very good oral bioavailability, including
phenyl analog 30 (rat F = 31%, dog F = 58%), 3-thio-
phene 31 (rat F = 44%), and 3-pyridyl 32 (rat
F = 56%). Compound 37 had very good oral bioavail-
ability in both rat (F = 34%) and dog (F = 65%) with
long half-lives and extremely low clearances.

758
D. N. Nguyen et al. / Bioorg. Med. Chem. Lett. 18 (2008) 755–758
Table 2. Pharmacokinetic profiles in rata and dogb
NMR analysis and animal dosing groups, as well as
the Department of Drug Metabolism for their
assistance.
Compound
Species
t1/2
Cl
(mL/min/kg)
Cmax
F
(h)
(lM)
(%)
9
16
16
19
20
24
25
30
30
31
32
37
37
Rat
Rat
Dog
Dog
Rat
Rat
Rat
Rat
Dog
Rat
Dog
Rat
Dog
3.0
1.1
2.9
1.2
1.0
0.5
2.8
2.5
3.7
0.9
5.4
6.3
3.3
1.3
16.1
0.6
20.3
0.1
0.6
0.3
7.4
1.3
0.08
5.3
10.3
5.1
7.4
9.4
9.4
40
2
References and notes
3
2.8
8.9
6
1. Edvinsson, L. Expert Opin. Ther. Targets 2003, 7, 377.
2. Durham, P. L. N. Engl. J. Med. 2004, 350, 1073.
3. Silberstein, S. D. Lancet 2004, 363, 381.
47
37
1
25.3
23.5
2.3
4. J. Headache Pain 2001, 120–146.
31
58
44
56
34
65
5. Olesen, J.; Diener, H.-C.; Husstedt, I. W.; Goadsby, P. J.;
Hall, D.; Meier, U.; Pollentier, S.; Lesko, L. M. N. Engl. J.
Med. 2004, 350, 1104.
6. Petersen, K. A.; Birk, S.; Lassen, L. H.; Krusse, C.;
Jonassen, O.; Lesko, L.; Olesen, J. Cephalalgia 2004, 25,
139.
7. Shaw, A. W.; Paone, D. V.; Nguyen, D. N.; Stump, C. A.;
Burgey, C. S.; Mosser, S. D.; Salvatore, C. A.; Rutledge,
R. Z.; Kane, S. A.; Koblan, K. S.; Graham, S. L.; Vacca,
J. P.; Williams, T. M. Bioorg. Med. Chem. Lett. 2007, 17,
4795.
0.2
9.6
0.3
0.8
0.2
a Dosed at 2.0 mpk IV as DMSO solution and 10 mpk PO as 1%
methylcellulose suspension.
b Dose at 0.5 mpk IV as DMSO solution and 1.0 mpk PO as 1%
methylcellulose suspension.
8. Paone, D. V.; Shaw, A. W.; Nguyen, D. N.; Burgey, C. S.;
Deng, J. Z.; Kane, S. A.; Koblan, K. S.; Salvatore, C. A.;
Hershey, J. C.; Wong, B.; Roller, S.; Miller-Stein, C. M.;
Graham, S. L.; Vacca, J. P.; Williams, T. M. J. Med.
Chem. 2007, 50, 5564.
9. Rudolf, K.; Eberlein, W.; Engel, W.; Pieper, H.; Doods,
H.; Hallermayer, G; Entzeroth, M.; Wienen, W. WO 1998/
9811128.
10. Burgey, C. S.; Stump, C. A.; Nguyen, D. N.; Deng, J. Z.;
Quigley, A. G.; Norton, B. R.; Bell, I. M.; Mosser, S. D.;
Salvatore, C. A.; Kane, S. A.; Koblan, K. S.; Vacca, J. P.;
Graham, S. L.; Williams, T. W. Bioorg. Med. Chem. Lett.
2006, 16, 5052.
11. Damewood, J. R. Jr.; Edwards, P. D.; Feeney, S.; Gomes,
B. C.; Steelman, G. B.; Tuthill, P. A.; Williams, J. C.;
Warner, P.; Woolson, S. A.; Wolanin, D. J.; Veale, C. A.
J. Med. Chem. 1994, 37, 3303.
In conclusion, we have discovered the pyridinone as an
alternative template for the CGRP receptor antagonist
program. These pyridinones are achiral and structurally
simple, and therefore provided easy access for either
N-alkyl or C-5 aryl substitution. SAR studies revealed
many compounds with good potency in both the intrin-
sic binding assay and functional cAMP assay. In gen-
eral, compounds containing polar groups, especially
H-bond donors such as 4-hydroxyphenyl and 3-pyra-
zole, had poor bioavailability but exhibited small serum
shifts. Analogs containing more lipophilic aryl groups
had greatly improved oral bioavailability; however,
these compounds displayed higher serum shifts. Most
notable was 37 which possessed the best balance of
potency and pharmacokinetic profiles.
12. Initially, the use of bis(pinacolato)diboron under palla-
dium catalysis failed to convert bromide 7 to its corre-
sponding boronate ester. However, when ethylmagnesium
bromide, tert-butyllithium and trimethylborate were used
at ꢀ78 °C, the corresponding boronic acid was obtained
after an aqueous workup.
Acknowledgments
We thank the West Point Department of Medicinal
Chemistry mass spectroscopy, analytical chemistry,