Calcitonin gene-related peptide (CGRP) receptor antagonists: Pyridine as a replacement for a core amide group

Article abstract of DOI:10.1016/j.bmcl.2012.02.065

In our continuing efforts to identify CGRP receptor antagonists that can be dosed orally for the treatment of migraine headache, we have investigated a pyridine bioisosteric replacement of a polar amide portion of a previous lead compound, BMS-694153. Pyr

Full text of DOI:10.1016/j.bmcl.2012.02.065

Bioorganic & Medicinal Chemistry Letters 22 (2012) 2917–2921
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Calcitonin gene-related peptide (CGRP) receptor antagonists: Pyridine
as a replacement for a core amide group
⇑
Guanglin Luo , Ling Chen, Rita Civiello, Sokhom S. Pin, Cen Xu, Walter Kostich, Michelle Kelley,
Charles M. Conway, John E. Macor, Gene M. Dubowchik
Molecular Sciences and Candidate Optimization, Neuroscience Biology, Bristol-Myers Squibb Research & Development, 5 Research Parkway, Wallingford, CT 06492, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 3 March 2012
In our continuing efforts to identify CGRP receptor antagonists that can be dosed orally for the treatment
of migraine headache, we have investigated a pyridine bioisosteric replacement of a polar amide portion
of a previous lead compound, BMS-694153. Pyridine derivatives were discovered and their SAR was stud-
ied. Some of them showed excellent binding potency. However, oral bioavailability was low, even for
compounds with good Caco-2 cell permeability.
Keywords:
CGRP antagonist
Migraine
Ó 2012 Elsevier Ltd. All rights reserved.
Migraine, as a very common and disabling medical illness, af-
fects approximately 12% of the adult population in the Western
society.1 Although the precise mechanisms remain elusive, mi-
graine pathophysiology is currently understood to primarily in-
volve the trigeminovascular system with dilation of intracranial
blood vessels playing an important role.2 As a current standard
treatment, the 5-HT1B/1D agonists (‘triptans’) act primarily through
constriction of intracranial arteries.3 However, these nonselective
vasoconstrictors are also associated with unpleasant cardiovascu-
lar side effects and are contraindicated in patients with hyperten-
sion or ischemic heart disease.4
As a 37 amino acid neuropeptide, calcitonin gene-related pep-
tide (CGRP) is an extremely potent vasodilator and has been impli-
cated in the pathogenesis of migraine.5 Therefore, a CGRP receptor
antagonist is an attractive therapeutic target for the treatment of
migraine.6 Clinical proof of concept has recently been achieved
with the intravenous administration of the potent receptor antag-
onist BIBN 4096 (olcegepant) with alleviation of pain in migra-
neurs.7 Other small-molecule CGRP receptor antagonists have
also been reported, most prominently the orally available MK-
0974 (telcagepant) which was recently discontinued following
completion of a large safety study.8
replacing the piperidinopiperidinyl amide portion of BMS-694153
with substituted pyridines. Herein, we report some SAR around
these pyridine derivatives (Fig. 1).
In a previous publication, we reported that substitution of the
core urea NH of BMS-694153 with a methylene group fully re-
tained the CGRP binding potency with improved intrinsic perme-
ability.10 For this reason, we first incorporated the same carbon
substitution into our pyridine derivatives. Synthesis of the carbon
analogs 7 and 8 is shown in Scheme 1, starting with aldehyde
1.11 Wittig reaction gave 2 which afforded the aldehyde 3 upon
treatment with strong acid. Horner–Emmons reaction of 3 afforded
acrylate 4. Regioselective protection of 4 was achieved for prepara-
tion of 5 using our previously reported conditions.11 Followed a
published procedure,12 freshly made pyridyl cuprate reacted with
5 to afford 6. Basic hydrolysis of 6 followed by amide coupling13
and SEM deprotection with TFA, afforded the desired carbon ana-
logs 7a–h and 8 (Scheme 1).
F
H
O
N
N
X = CH2, NH, O
O
A recent publication from our laboratories disclosed a potent
indazole-containing CGRP receptor antagonist (BMS-694153,
Fig. 1) with rapid and efficient intranasal exposure.9 However, with
poor intrinsic permeability, BMS-694153 did not have significant
oral bioavailability in either the cynomolgus monkey or rat. One
approach we have taken to increase oral bioavailability was to
try and improve bilayer permeability of our compounds by
H
N
N
N
N
N
X
O
R
N
O
N
N
Me
N
H
N
Me
H
BMS-694153
hCGRP Ki = 0.013 nM
Pyridines
⇑
Corresponding author.
E-mail address: guanglin.luo@bms.com (G. Luo).
Figure 1. Novel pyridine derivatives.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2012.02.065

2918
G. Luo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2917–2921
Me
Me
Me
Me
R
R
H
N
H
N
H
N
a
b
b, c
N
a
O
P
O
+
N
N
SEM
OPh
OPh
N
N
N
N
H
OHC
OHC
O
O
NHPh
1
2
3
SEM
9
10
11
N
N
SEM
O
N
HO
N
SEM
SEM
N
N
N
N
Me
d
N
N
HN
N
Me
Me
R
c
d
e
R
Me
Me
OEt
12
13
OEt
OEt
O
N
O
N
R
H2N
N
R
SEM
O
O
N
N
N
SEM
O
N
N
f
e
4
5
6
X
R
N
HN
H
Me
O
N
15
Me
Me
14
H
O
N
N
f, g, h
H
N
N
O
N
H
8
g, h
N
N
O
O
7a-h
N
N
R
N
H
N
R
Me
16
Scheme 1. Reagents and conditions: (a) (methoxymethyl)triphenyl-phosphonium
chloride, KOtBu, THF, 93%: (b) HClO4 (60%), THF; (c) (carbethoxymethylene)tri-
phenylphosphorane, THF, 33% for two steps; (d) SEM-Cl, N,N-dicyclohexylmethyl-
amine, THF, 93%; (e) substituted 2-bromopyridine, n-BuLi, CuI, n-Bu2S, ether; (f)
LiOH, THF; (g) 3-(diethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one, 4-substi-
tuted piperidines, Et3N, CH2Cl2; (h) TBAF, THF, 60 °C.
i
Scheme 2. Reagents and conditions: (a) phenylamine, diphenyl phosphate, PrOH;
(b) Cs2CO3, THF/iPrOH (4:1); (c) HCl (1 N); (d) NaBH4, MeOH; (e) phthalimide, Ph3P,
DIAD, CH2Cl2; (f) hydrazine, MeOH, 70 °C; (g) CDI, 4-substituted piperidine, iPr2NEt,
CH2Cl2; (h) TBAF, THF, 60 °C.
with broad substitution on the pyridine ring for optimizing com-
pound properties, we decided to investigate the corresponding
urea and carbamate analogs.
All compounds prepared were tested in
a competitive
[
125I]CGRP binding assay (hCGRP IC50) and selected compounds
were tested for antagonist activity (cAMP IC50) in a whole-cell as-
say following previously published protocols.9 All compounds
tested for functional activity were found to be full antagonists of
the human CGRP receptor. The data for compounds 7a–h and 8 is
summarized in Table 1. The methylene-linked pyridine derivatives
generally showed potent activity. As we had hoped, some com-
pounds (7a–c) had good Caco-2 permeability (>100 nm/sec). How-
ever, rat oral bioavailability of these compounds was low (<4%),
most likely due to metabolic instability and/or P-gp efflux. With
these issues and synthetic chemistry that was largely incompatible
For the synthesis of urea and carbamate derivatives, we envi-
sioned that both would be derived from a common intermediate
pyridinyl carbinol 13 as shown in Scheme 2. A published procedure
for the synthesis of substituted phenylacetylpyridines starting
with unsubstituted pyridine-carboxaldehyde14 was successfully
adapted for the synthesis of a variety of ketones 12. In this case,
substituted pyridine-2-carboxyaldehye 9 was converted to the
phenylaminophosphonate 10, which reacted with aldehyde 1111
followed by acid treatment to afford 12. Simple reduction afforded
Table 1
Selected data for compounds 7 and 8
X
H
N
HN
O
N
Me
N
N
H
N
O
O
7
8
N
R
a
a
Compds
X
R
hCGRP IC50 (nM)
cAMP IC50 (nM)
Caco-2 (nm/sec) (rat% Fpo)
7a
7b
7c
7d
7e
7f
H
F
H
H
4-Et
3-Me
5-Me
5-CH2OH
5-CHO
8.6
35
10
4.4
12
7.9
1.1
8.7
20
13
8.2
18
6.0
0.94
120 (3%)
210 (3%)
140 (2%)
H
H
H
H
H
64 (1%)
7g
5
7h
8
H
—
1.8
24
1.3
30
<15
34
N
H
a
Values are means of at least two experiments.

G. Luo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2917–2921
2919
R1
Table 3
Selected data for compounds 19l–r
R1
O2N
OCOCl
N
O
H
N
O
HN
a
O
O2N
N
17
A
SEM
O
N
N
HN
N
N
N
H
O
R
Me
Me
N
H
O
N
N
R1
R1
17
b
Me
c
13
B
O
N
N
O
N
O
O
N
Compds
R
GPCR
element
hCGRP
IC50
cAMP
IC50
Caco-2
a
a
R
R
(nm/
sec)
(nM)
(nM)
0.099
0.020
19
18
N
N
19l
A
0.28
Scheme 3. Reagents and conditions: (a) Et3N, CH2Cl2; (b) NaH, THF; (c) TBAF, THF,
60 °C, or TFA, CH2Cl2.
O
19m
19n
A
B
0.10
12
H
Table 2
19o
19p
19q
N
B
B
B
0.16
0.24
1.2
0.10
85
49
99
Selected data for compounds 16a–f and 19a–k
H
N
O
N
N
O
N
N
X
N
N
19r
B
0.095
0.042
28
N
O
N
H
a
Me
Values are means of at least two experiments.
R
a
a
Compds
X
R
hCGRP IC50 (nM)
cAMP IC50 (nM)
16a
19a
16b
19b
16c
19c
16d
19d
16e
16f
19e
19f
19g
19h
19i
NH
O
NH
O
NH
O
NH
O
NH
NH
O
O
O
H
H
6-Br
6-Br
5,6-Benzo
5,6-Benzo
4,5-Benzo
4,5-Benzo
6-Ph
6-Me
4-CH2OH
4-CHO
4-CO2H
4-CO2Me
4-CN
4-CONMe2
4-(4-Py)
3.1
1.5
0.49
0.38
2.0
4.1
1.9
0.82
0.46
1.4
4.8
1.0
0.63
14
3.4
2.1
0.28
28
N
HN
Me
H
O
N
R
N
Me
Me
1.4
H
N
H
N
0.82
0.46
12
3.8
2.0
0.21
30
4.4
3.0
0.33
0.27
*
O
N
a
N
N
O
OHC
OHC
N
R
1
R = Br or Cl
22a,b
SEM
O
SEM
N
O
O
O
O
N
SEM
N
N
N
H
N
19j
19k
Me
O
Me
O
N
N
a
Values are the means of at least two experiments.
N
O
N
b
O
N
O
N
the key intermediate pyridine-2-carbinol 13. Mitsunobu reaction
with phthalimide afforded 14, which upon reaction with hydra-
zine, produced primary amine 15. After activation with CDI, 15 re-
acted with various 4-substituted piperidines to afford the urea
analogs 16 (Scheme 2).
For the synthesis of carbamate analogs, activation of the hydro-
xyl group in 13 proved to be unsuccessful owing to facile hydroly-
sis of the active intermediate with the assistance of the
neighboring pyridine nitrogen. On the other hand, reaction of 13
with p-nitrophenyl carbamate 17 in the presence of NaH proved
to be very efficient leading to carbamates 18 (Scheme 3). Deprotec-
tion of 18 generated the desired pyridyl carbamate derivatives
19.15
18
20
SEM
N
N
SEM
N
N
HN
H
N
Me
O
Me
O
CHO
R
N
N
O
N
N
O
N
N
c
d, e
O
O
21
22d-g
Binding and selected functional data for ureas 16 and carba-
mates 19 are summarized in Tables 2 and 3. Both showed
comparable binding potency. Initially, a series containing the
piperidine-linked dihydroquinazolinone GPCR recognition element
Scheme 4. Reagents and conditions: (a) NBS or NCS, CH2Cl2; Ã as per Scheme 3; (b)
SEM-Cl, NaH, THF; (c) n-BuLi, DMF, THF; (d) Secondary amines, Na(OAc)3BH,
ClCH2CH2Cl; (e) TFA, CH2Cl2.

2920
G. Luo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2917–2921
Table 4
to the hydrogen bond accepting capability of these substituents
mimicking the second piperidine in BMS-694153.
Selected data for compounds 22
H
N
Substitution with piperidylcarboxyl (19l) and piperidylmethyl
(19m) groups maintained excellent binding potency (Table 3). Be-
cause the dihydroquinazolinone ring has been reported to undergo
facile air oxidation,16 we decided to replace it with the quinolone
group (GPCR element B in Table 3). It should be noted that, the
example with no pyridyl substitution (19n) loses ca. eightfold in
binding potency in comparison with 19a. However, the piper-
idylmethyl analog (19o) largely regains activity. In addition,
Caco-2 permeability was significantly improved over other mem-
bers of the series. Other analogs with saturated heterocyclic sub-
stituents either showed diminished permeability (19p and 19r)
or binding potency (19q). Unfortunately, all of these compounds
were rapidly cleared in human and rat liver microsomes.
O
N
R
N
O
N
N
O
N
H
Me
a
Compds
R
hCGRP IC50 (nM)
19a
22a
22b
22c
22d
H
Br
Cl
Ph
CHO
1.5
0.33
0.39
0.24
1.3
Based on SAR from a different chemotype, we also investigated
limited substitution at the 3-position of the indazole core. Synthe-
sis of these derivatives were either from direct halogenation of
aldehyde 1 at the electron-rich 3-C, followed by elaboration as
shown in Schemes 2 and 3,17 or SEM directed formylation11 fol-
lowed by reductive amination to incorporate polar tertiary amine
groups (Scheme 4). CGRP binding data for representative com-
pounds 22a–g is shown in Table 4. In comparison with 19a, halo-
gen and aryl substitution (22a–c) further increases binding
potency, while incorporation of polar groups (22e–g) tends to re-
duce it. Unfortunately, the very low aqueous solubilities of the
more potent compounds precluded them from further study.
Finally, a small set of 6-chloro-4-N,N-dimethylamidopyridines
with selected known right side GPCR recognition elements (R in
25)7,8 was prepared as shown in Scheme 5. Commercially available
acid chloride 23 was converted to the aldehyde 24 in a 3-step se-
quence following published procedures.18 Aldehyde 24 was then
converted to the target compounds 25a–c following the protocols
illustrated in Schemes 2 and 3. Selected data is shown in Table 5.
Quinolone 25a showed sub-nanomolar binding potency, with
activity residing, as expected, in the R-enantiomer.19 The com-
pound had modest Caco-2 cell penetration, and demonstrated
low oral bioavailability in rats. Introduction of the more lipophilic
diazepinone component in 25b retained binding potency but abol-
ished Caco-2 permeability. The azabenzimidazolone 25c, also
highly potent, showed barely measurable Caco-2 permeability
and no detectable oral exposure. It is worth noting, however, that
the latter possessed a very high plasma free fraction (21%) in com-
parison with 25a (6%).
22e
22f
22g
N
6.6
1.9
6.8
O
N
N
a
Values are means of at least two experiments.
N
HN
Me
R
O
Cl
O
NMe2
Scheme 2, 3
a, b, c
O
N
N
O
Cl
N
Me
Cl
N
CHO
O
Cl
23
24
NMe2
25a-c
Scheme 5. Reagents and conditions: (a) Me2NH (in THF), CH2Cl2, 96%; (b) tBu-ONO,
KOtBu (1 M in THF), THF, 90%; (c) acetone, HCl, water/THF (1:1), 83%.
was explored (Table 2). Quinolines (16c and 19c) were slightly less
potent than isoquinolines (16d and 19d). Although a lipophilic bro-
mine at the 6-position of the pyridine conferred some advantage
(16b and 19b), this did not extend to methyl (16f) and phenyl
(16e) substitution. In the carbamate series, 4-substitution with
aldehyde (19f), N,N-dimethylamide (19j) and 4-pyridyl (19k) gave
compounds with sub-nanomolar binding potency, possibly owing
In conclusion, we have discovered a novel class of CGRP antag-
onists in which a core amide group found in previous series has
Table 5
Selected data for compounds 25a–c
a
a
Compds
R
hCGRP IC50 (nM)
cAMP IC50 (nM)
Protein Adj. Kib (nM)
Caco-2 (nm/sec) Fpo (rat)
52 (4.5%)
O
NH
25a
0.28
0.34
3.3
(R)-25a
25b
0.077
0.093
0.064
H
O
N
<15
N
O
NH
N
25c
0.071
0.18
16 (BBLQ)
N
a
Values are means of at least two experiments.
Plasma protein binding was factored in.
b

G. Luo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 2917–2921
2921
been replaced by pyridine. As a backbone linker moiety, the ure-
thane moiety was most compatible with existing synthetic chem-
istry, and afforded sub-nanomolar antagonists with appropriate
pyridyl substitution. However, oral bioavailability remained low,
emphasizing the difficulties we and others have encountered in
discovering development candidates for this difficult Class B GPCR
target. Further optimization towards orally available CGRP recep-
tor antagonists will be reported in due course.
16. Degnan, A. P.; Conway, C. M.; Dalterio, R. A.; Macci, R.; Mercer, S. E.; Schartman,
R.; Xu, C.; Dubowchik, G. M.; Macor, J. E. Bioorg. Med. Chem. Lett. 2009, 19, 3555.
17. Compound 22c was prepared by
a Suzuki reaction (phenylboronic acid,
Pd(Ph3P)4, K2CO3, toluene) with an intermediate for 22a as follows:
N
N
N
HN
HN
HN
H
N
Me
Me
Me
O
Br
Ph
Ph
O
N
References and notes
OH
OH
N
O
N
N
Suzuki
Scheme 3
1. Ferrari, M. D. Lancet 1998, 351, 1043.
N
2. (a) De Vries, P.; Villalon, C. M.; Saxena, P. R. Eur. J. Pharmacol. 1999, 375, 61; (b)
Villalon, C. M.; Centurion, D.; Valdivia, L. F.; De Vries, P.; Saxena, P. R. Proc. West
Pharmacol. Soc. 2002, 45, 199.
22c
3. Edvinsson, L.; Uddman, E.; Wackenfors, A.; Davenport, A.; Longmore, J.;
Malmsjo, M. Clin. Sci. 2005, 109, 335.
4. Goadsby, P. J.; Lipton, R. B.; Ferrari, M. D. N. Engl. J. Med. 2002, 346, 257.
5. (a) Goadsby, P. J. Drugs 2005, 65, 2557; (b) Edvinsson, L. Cephalalgia 2004, 24,
611; (c) Williamson, D. J.; Hargreaves, R. J. Microsc. Res. Tech. 2001, 53, 167.
6. (a) Edvinsson, L. Exper. Opin. Ther. Targets 2003, 7, 377; (b) Durham, P. L. N. Engl.
J. Med. 2004, 350, 1073.
7. (a) Oleson, 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; (b) Rudolf, K.;
Eberlein, W.; Engel, W.; Pieper, H.; Entzeroth, M.; Hallermayer, G.; Doods, H. J.
Med. Chem. 2005, 48, 5921.
18. Frey, L. F.; Marcantonio, K.; Frantz, D. E.; Murry, J. A.; Tillyer, R. D.; Grabowski,
E. J. J.; Reider, P. J. Tetrahedron Lett. 2001, 42, 6815.
19. Compound 25a was resolved through chiral HPLC column separation and (R)-
25a with 96%ee was also prepared through the following chiral reduction
reaction (See: Brwon, H. C.; Chandrasekharan, J.; Ramachandran, P. V. J. Am.
Chem. Soc. 1988, 110, 1539):
8. (a) Paone, D. V.; Shaw, A. W.; Nguyen, D. N.; Burgey, C. S.; Deng, J. Z.; Kane, S. A.;
Koblan, K. S.; Salvatore, C. A.; Mosser, S. D.; Johnston, V. K.; Wong, B. K.; Miller-
Stein, C. M.; Hershey, J. C.; Graham, S. L.; Vacca, J. P.; Williams, T. M. J. Med.
Chem. 2007, 50, 5564; (b) Nguyen, D. N.; Paone, D. V.; Shaw, A. W.; Burgey, C. S.;
Mosser, S. D.; Johnston, V.; Salvatore, C. A.; Leonard, Y. M.; Miller-Stein, C. M.;
Kane, S. A.; Koblan, K. S.; Vacca, J. P.; Gaham, S. L.; Williams, T. M. Bioorg. Med.
Chem. Lett. 2008, 18, 755; (c) Hewitt, D. J.; Martin, V.; Lipton, R. B.; Brandes, J.;
Ceesay, P.; Gottwald, R.; Schaefer, E.; Lines, C.; Ho, T. W. Headache 2011, 51,
533. and references cited therein; (d) Tfelt-Hansen, P. Headache 2011, 51, 118.
9. Degnan, A. P.; Chaturvedula, P. V.; Conway, C. M.; Cook, D. A.; Davis, C. D.;
Denton, R.; Han, X.; Macci, R.; Mathias, N. R.; Moench, P.; Pin, S. S.; Ren, S. X.;
Schartman, R.; Signor, L. J.; Thalody, G.; Wikmann, K. A.; Xu, C.; Macor, J. E.;
Dubowchik, G. M. J. Med. Chem. 2008, 51, 4858.
SEM
SEM
N
N
N
N
Me
Me
(+)-DIP-Cl
THF
O
OH
(R)
N
N
O
O
Cl
Cl
NMe2
NMe2
O
10. Luo, G.; Chen, L. Pin, S. S.; Xu, C.; Conway, C. M.; Macor, J. E.; Dubowchik, G. M.
Bioorg. Med. Chem. Lett. 2012, doi:10.1016/j.bmcl.2012.02.066.
H
N
11. Luo, G.; Chen, L.; Dubowchik, G. J. Org. Chem. 2006, 71, 5392.
12. Malmberg, H.; Nilsson, M. Tetrahedron 1982, 38, 1509. For unsubstituted
pyridine, this reaction gave 62% yield with 5, but without SEM protection, the
yield was only 11% with 4. For other substituted pyridines, the yields were
significantly lower.
13. Li, H.; Jiang, X.; Ye, Y.; Fan, C.; Romoff, T.; Goodman, M. Org. Lett. 1999, 1, 91.
14. Journet, M.; Cai, D.; Larsen, R. D.; Reider, P. J. Tetrahedron Lett. 1998, 39, 1717.
15. Some of the 4 or 6-substituted pyridines were prepared by further derivatizing
functional groups such as ester or bromide at either intermediate or final
compound stage. For example:
O
N
N
O
N
N
H
Me
Cl
NMe2
O
(R)-25a
CO2Me SEM
HO
N
CO2Me
N
N
DIBAL
CO2Me
N
CHO
N
N
CO2Me
Me
Me
N
HN
R1
SEM
HO
N
N
DIBAL
O
N
N
CHO
O
Me
19
R