Benzodiazepine calcitonin gene-related peptide (CGRP) receptor antagonists: Optimization of the 4-substituted piperidine

Article abstract of DOI:10.1016/j.bmcl.2006.07.044

In our continuing effort to identify CGRP receptor antagonists for the acute treatment of migraine, we have undertaken a study to evaluate alternative 4-substituted piperidines to the lead dihydroquinazolinone 1. In this regard, we have identified the piperidinyl-azabenzimidazolone and phenylimidazolinone structures which, when incorporated into the benzodiazepine core, afford potent CGRP receptor antagonists (e.g., 18 and 29). These studies produced a potent analog (18) which overcomes the instability issues associated with the lead structure 1. A general pharmacophore for the 4-substituted piperidine component of these CGRP receptor antagonists is also presented.

Full text of DOI:10.1016/j.bmcl.2006.07.044

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5052–5056
Benzodiazepine calcitonin gene-related peptide (CGRP) receptor
antagonists: Optimization of the 4-substituted piperidine
Christopher S. Burgey,a,* Craig A. Stump,a Diem N. Nguyen,a James Z. Deng,a
Amy G. Quigley,a Beth R. Norton,a Ian M. Bell,a Scott D. Mosser,b
Christopher A. Salvatore,b Ruth Z. Rutledge,b 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
Received 26 May 2006; revised 11 July 2006; accepted 13 July 2006
Available online 2 August 2006
Abstract—In our continuing effort to identify CGRP receptor antagonists for the acute treatment of migraine, we have undertaken a
study to evaluate alternative 4-substituted piperidines to the lead dihydroquinazolinone 1. In this regard, we have identified the pipe-
ridinyl-azabenzimidazolone and phenylimidazolinone structures which, when incorporated into the benzodiazepine core, afford
potent CGRP receptor antagonists (e.g., 18 and 29). These studies produced a potent analog (18) which overcomes the instability
issues associated with the lead structure 1. A general pharmacophore for the 4-substituted piperidine component of these CGRP
receptor antagonists is also presented.
Ó 2006 Elsevier Ltd. All rights reserved.
The triptan class of 5-HT1B/1D receptor agonists repre-
sents the current standard of treatment for a migraine
attack; presumably due to 5-HT1B mediated nonselec-
tive vasoconstriction,1 however, these compounds are
contraindicated in patients with cardiovascular disease.
Thus, development of a therapy devoid of these cardio-
vascular liabilities would represent a considerable thera-
peutic advance. Calcitonin gene-related peptide (CGRP)
has been implicated in the pathogenesis of migraine
headache2 and clinical proof of concept has recently
been demonstrated with the intravenous administration
of the potent receptor antagonist BIBN 4096 (olcege-
pant).3 Importantly, this efficacy was achieved without
an effect on cerebral or systemic hemodynamics.4
azepine high-throughput screening lead.5 Replacement
of the tetralone-derived spirohydantoin with a tradition-
al GPCR privileged structure,6 namely the piperidinyldi-
hydroquinazolinone, afforded a series of potent CGRP
antagonists (e.g., 1: Ki = 48 nM, cAMP IC50 = 34 nM).
A liability associated with this series of compounds
was that the dihydroquinazolinone suffered benzylic oxi-
dation under ambient laboratory conditions. Herein, we
report the development of a series of potent CGRP
antagonists containing alternative 4-substituted piperi-
dines which demonstrate improved stability profiles.
All compounds prepared were tested in the CGRP
receptor (recombinant human CLR/RAMP1) competi-
tive [125I]CGRP radioligand binding assay (Ki). Selected
compounds were subsequently tested for their functional
ability to inhibit CGRP-stimulated cAMP production in
whole cells (cAMP IC50).7
In this context, we have undertaken a research program
aimed at identifying small-molecule CGRP receptor
antagonists suitable for oral administration during a mi-
graine attack. A recent report from these laboratories
detailed the identification and initial SAR of a benzodi-
Initial focus was placed on the ring contracted ben-
zimidazolones (Fig. 1, 2), which would eliminate the oxi-
dation-prone benzylic position. As the unsubstituted
analog was previously shown to have weak affinity for
the CGRP receptor (Table 1, 4: Ki = 2250 nM),5 a series
of substituted benzimidazolones (5–17) were prepared in
an effort to improve potency. This study revealed that
Keywords: CGRP; Antagonist; Migraine.
*
Corresponding authors. Tel.: +1 215 652 2382; fax: +1 215 652 7310
(C.S.B.); tel.: +1 215 652 3087; fax: +1 215 652 3971 (T.M.W.);
e-mail addresses:
williams@merck.com
christopher_burgey@merck.com;
theresa_
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.044

C. S. Burgey et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5052–5056
5053
In an extension of this study, introduction of a nitrogen
atom into the benzimidazolone aryl ring (to form each
of the four possible pyridine isomers) revealed a pro-
found effect on potency: installation of the nitrogen at
the 4-position afforded a 100- and 30-fold improvement
in binding and functional potency, respectively (Table 1,
18). N-Oxidation of this derivative to give 22 led to a
40-fold loss in potency. Incorporation of the 4,6- and
4,7-diaza modifications gave 23 and 24, analogs which
retained good potency in the CGRP binding assay.
F3C
O
N
O
N
NH
O
NH
N
N
1
Ki = 48 nM
cAMP IC50 = 34 nM
oxidation
alternative GPCR
privileged structures
Having established the considerable potency improve-
ment upon incorporation of a nitrogen atom into the
benzimidazolone nucleus, we sought to exploit this fea-
ture within other piperidine-derived privileged struc-
tures. Unfortunately, a corresponding ‘aza binding
site’ does not appear to be operative in the context of
the dihydroquinazolinones, as preparation of all four
of the fused pyridine isomers led to decreased affinity
for the CGRP receptor (Table 2).
O
O
NH
N
NH
N
N
N
R
R
2
3
Figure 1.
Table 1. Benzimidazolone SAR
The phenylimidazolinone3 (3, Fig. 1) is an alternative
heterocycle which maintains both the cyclic urea and
aryl substituent of the parent dihydroquinazolinone
but removes the benzylic site of oxidation. Preparation
of the phenylimidazolinone derivative 29 afforded a po-
tent CGRP receptor antagonist (Table 3, Ki = 11 nM,
cAMP IC50 = 14 nM). A limited scan of phenyl ring
substituents (30–36) revealed that the addition of a
3-methoxy group produced a 10-fold improvement in
potency (31: Ki = 1.1 nM, cAMP IC50 = 5.1 nM).
Replacement of the phenyl with a 2-pyridyl ring, in an
effort to access the ‘aza binding site,’ did not lead to
an improvement in binding affinity (37: Ki = 27 nM).
F3C
O
N
O
NH
NH
N
N
N
O
4
R
7
5
6
Compound
R
H
Kia (nM)
cAMP
IC50 (nM)
4
5
2250
6900
375
1583
4-Me
4-F
4-Cl
6
373
446
7
143
Although incorporation of the piperidinyl-phenylimi-
dazolinone into the benzodiazepine-based template af-
fords very potent CGRP antagonists, this heterocycle
also experiences degradation under ambient laboratory
conditions. Decomposition products were observed
that are consistent with oxidation occurring at the
electron-rich double bond of the imidazolinone ring;8
consequently, a range of non-aryl imidazolinone sub-
stituents were explored in an attempt to stabilize this
ring system (Table 3, 38–44). Although these analogs
8
5-Me
5-CF3
5-F
6100
3800
2800
>10,000
810
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
5-Cl
5-CO2H
715
5-CONH2
5-SO2Me
6-F
1055
260
220
1396
983
7-Me
7-Cl
20,000
6400
23
4-Aza
5-Aza
49
1381
1398
1135
685
853
6-Aza
7-Aza
Table 2. Aza-dihydroquinazolinone SAR
427
4-Aza oxide
4,6-Diaza
4,7-Diaza
1000
51
F3C
430
286
O
N
156
O
NH
O
NH
a Values are means of 2 experiments.
8
6
N
N
R
7
N
5
the majority of these analogs did not lead to a significant
improvement in binding affinity (Table 1). Notable
exceptions included the 5-substituted methylsulfone
(14) and the 4-fluoro and chloro analogs (6 and 7),
which were substantially more potent in both the intrin-
sic CGRP binding and cAMP functional assays. Impor-
tantly, no decomposition of the benzimidazolones was
observed upon prolonged storage.
Compound
R
H
Kia (nM)
cAMP IC50 (nM)
1
25
26
27
28
48
295
350
320
965
34
616
8-Aza
7-Aza
6-Aza
5-Aza
224
627
1837
a Values are means of 2 experiments.

5054
C. S. Burgey et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5052–5056
Table 3. Imidazolinone SAR
Table 4. Imidazolidinone SAR
F3C
F3C
O
O
N
N
O
O
NH
O
NH
NH
NH
N
N
N
N
N
N
O
R
R
Compound
R
Kia (nM)
cAMP
IC50 (nM)
Compound
R
Kia (nM)
cAMP IC50 (nM)
49b
50b
51b
52b
53b
H
Ph
855
339
3100
790
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Ph
3-FPh
11
11
14
46
2-Pyr
3-Pyr
4-Pyr
2000
1050
720
3-MeOPh
4-FPh
1.1
5.1
107
85
11
804
4-MePh
4-MeOPh
6.3
a Values are means of 2 experiments.
b Mixture of imidazolidinone diastereomers.
205
104
19
193
431
74
4-CF3OPh
4-CHF2OPh
2-Pyr
27
37
CH2Ph
CO2Me
CO2H
CH2OH
CN
1125
110
2900
20,000
575
583
5337
450
156
bered cyclic urea, with the terminal nitrogen unalky-
lated,5 appears to represent the basic criterion for
CGRP receptor affinity (e.g., 43 and 49). To increase
potency, this core ring can be substituted with an
aryl moiety, the orientation of which is crucial. When
990
672
H
iPr
n = 0 (5-membered ring),
a pendant aryl group
(X = Ar) is required: fusion of a benzene ring onto
43 to give 4 leads to a loss in potency, while append-
age of the phenyl group (as in 29) improves potency
by 50-fold.10 As evidenced by the moderate potency
of 50, the 5-membered ring must also maintain a pla-
nar orientation to obtain full benefit of the aryl sub-
stituent. The inactivity of 47 serves to demonstrate
that when n = 1 (6-membered ring) the aryl group
should be fused distally to the piperidine ring, as in
1.
a Values are means of 2 experiments.
were qualitatively more stable than their aryl conge-
ners, they experienced a significant loss of potency
(>10-fold). In contrast, the corresponding phenyltriaz-
olinone 45 (Chart 1) is a stable ring system9 that only
forfeits 2-fold in CGRP receptor affinity. Additionally,
a range of reduced imidazolinones (i.e., imidazolidi-
nones) were prepared (Table 4, 49–53): these stable
analogs were significantly less potent than their dehy-
dro analogs.
Alternatively, substitution of the principal cyclic urea
core with
a
hydrogen bond acceptor (Fig. 2,
X = HBA) can deliver a large improvement in CGRP
receptor affinity. This strategy is demonstrated by the
comparison of 4 and 18 in which introduction of the
HBA ring nitrogen serves to improve potency by 100-
fold. Furthermore, the comparison of 49 to 48 reinforces
the concept that an aryl group is not required to deliver
potent CGRP antagonists, but can be attained with a
simple hydrogen bond acceptor.
Based on the SAR generated herein a general phar-
macophore for the 4-substituted piperidine privileged
structures can be proposed (Fig. 2). A 5- or 6-mem-
F3C
O
N
O
NH
Attempts to combine both an aryl group and a hydrogen
bond acceptor, as in 25 and 37, reveal that these potency
enhancing modifications appear to be mutually
exclusive.
NH
N
N
N
N
O
45
Ki = 26 nM
cAMP IC50 = 100 nM
The benzimidazolone SAR could be rapidly evaluated
using a modification of the literature conditions
O
O
NH
NH
N
N
N
N
O
O
NH
48
Ki = 29 nM
cAMP IC50 = 100 nM
N
N
X
X = HBA or Ar
47
inactive
n
Chart 1.
Figure 2.

C. S. Burgey et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5052–5056
5055
NO2
O
EtO2C
N
NH2
X
NO2
R
EtO2C
N
NH NO2
Cl
O
O
Na2CO3, KI
NH
THF, Et3N
NH2
cyclohexanol
BocN
NH2
BocN
NH
HO
CO2Me
R
180 ˚C, 10 min μw
55
54
OH
63
MeO2C
BocN
64
87%
Zn, 1N HCl
MeOH, 50 ˚C
CDI, CH3CN
EtO2C
N
NH NH2
O
N
Dess-Martin
Periodinane
TFA
NH
CO2Me
56
R
95%
85%
65
O
O
O
1N NaOH
100 ˚C
NH
NH
NH
HN
N
HN
N
EtO2C
N
N
CO2Me
R
R
66
57
58
Scheme 3.
Scheme 1.
(Scheme 1).11 Ethyl 4-amino-1-piperidinecarboxylate
could be added to a range of substituted o-halonitro-
aromatics 54, followed by nitro group reduction of the
product 55 with Zn metal in acidic methanol to deliv-
er the o-amino anilines 56. Ring closure with carbon-
4-nitrophenyl chloroformate. Half-reduction of ester
60 was followed by acid-catalyzed cyclodehydration to
form the imidazolinone 61. Deprotection then afforded
the desired piperidine 62.
yldiimidazole,
followed
by
ethyl
carbamate
deprotection, afforded the substituted benzimidazol-
Alternatively, substrates which contain functional
groups that are acid labile, or susceptible to Dibal-H
reduction, can be prepared via an oxidative variant
of the preceding route (Scheme 3). Amino alcohols,
such as serine methyl ester, are coupled to the 4-amin-
opiperidine 63 to afford the urea 64. Treatment with
Dess–Martin periodinane produces the imidazolinone
65 directly and does not necessitate a separate dehy-
dration step.
ones 58.12
Reductive cyclization methodology was implemented to
prepare the imidazolinones (Scheme 2).13 For example,
urea formation between the 4-aminopiperidine 59 and
phenylalanine methyl ester was accomplished using
The piperidines prepared through the preceding routes
(Schemes 1–3) were converted to the final ureas using
the previously disclosed method.5
NO2
O
In conclusion, we have identified the piperidinyl-aza-
benzimidazolone and phenylimidazolinone as alterna-
tive privileged structures which when incorporated into
the benzodiazepine core afford potent CGRP receptor
antagonists (e.g., 18 and 29). The former analog (18)
overcomes the instability issues associated with the lead
structure 1. A general pharmacophore for the 4-substi-
tuted piperidine component of these CGRP receptor
antagonists was also proposed. The optimization of this
series is currently being pursued and the results will be
disclosed in due course.
Cl
O
O
NH
THF, Et3N
NH2
BnN
NH2
BnN
NH
MeO2C
Bn
MeO2C
95%
Bn
59
60
O
1) Dibal-H
H2, Pd/C
NH
Bn
BnN
N
2) 4N HCl
36%
AcOH/MeOH
86%
61
O
Acknowledgments
NH
Bn
HN
N
We thank the analytical chemistry, mass spectroscopy,
and NMR analysis groups for their assistance, as well
as Dr. Wayne Thompson for providing several key
intermediates.
62
Scheme 2.

5056
C. S. Burgey et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5052–5056
8. A major degradation by-product formed was:
References and notes
F3C
O
1. Wackenfors, A.; Jarvius, M.; Ingemansson, R.; Edvinsson,
N
O
O
L.; Malmsjo, M. J. Cardiovasc. Pharmacol 2005, 45, 476.
2. Durham, P. L.; Russo, A. F. Pharmacol. Ther. 2002, 94,
77; Edvinsson, L. Expert Opin. Ther. Targets 2003, 7, 377;
Durham, P. L. N. Engl. J. Med. 2004, 350, 1073.
3. Rudolf, K.; Eberlein, W.; Engel, W.; Pieper, H.; Entze-
roth, M.; Hallermayer, G.; Doods, H. J. Med. Chem.
2005, 48, 5921; Rudolf, K.; Eberlein, W.; Engel, W.;
Pieper, H.; Doods, H.; Hallermayer, G.; Entzeroth, M.;
Wienen, W. WO 9811128.
NH
O
N
N
NH
N
H
.
9. Deng, J. Z.; Burgey, C. S. Tetrahedron Lett. 2005, 46,
7993.
10. Interestingly, the hybrid molecule 46 is nearly equipotent
to 29:
4. Olesen, J.; Diener, H.-C.; Husstedt, I. W.; Goadsby, P. J.;
Hall, D.; Meier, U.; Pollentier, S.; Lesko, L. N. Engl. J.
Med. 2004, 350, 1104; Petersen, K. A.; Birk, S.; Lassen, L.
H.; Kruuse, C.; Jonassen, O.; Lesko, L.; Olesen, J.
Cephalalgia 2004, 25, 139.
5. Williams, T. M.; Stump, C. A.; Nguyen, D. N.; Quigley,
A. G.; Bell, I. M.; Gallichio, S. N.; Zartman, C. B.; Wan,
B.-L.; Della Penna, K.; Kunapuli, P.; Kane, S. A.;
Koblan, K. S.; Mallee, J. J.; Mosser, S. D.; Rutledge, R.
Z.; Salvatore, C.; Fay, J. F.; Vacca, J. P.; Graham, S. L.
Bioorg. Med. Chem. Lett. 2006, 16, 2595.
F3C
O
N
O
NH
NH
N
N
N
O
46
Ki = 27 nM
cAMP IC50 = 369 nM
6. Bondensgaard, K.; Ankersen, M.; Thogersen, H.; Hansen,
B. S.; Wulff, B. S.; Bywater, R. P. J. Med. Chem. 2004, 47,
888; Patchett, A. A.; Nargund, R. P. Annu. Rep. Med.
Chem. 2000, 35, 289.
7. For the details of these assays, see: Bell, I. M.;
Graham, S. L.; Williams, T. M.; Stump, C. A. WO
2004/087649.
11. Henning, R.; Lattrell, R.; Gerhards, H. J.; Leven, M.
J. Med. Chem. 1987, 30, 814.
12. Burgey, C. S.; Stump, C. A.; Williams, T. M. WO 2005/
000807 A2.
13. Markwalder, J. A.; Pottorf, R. S.; Seitz, S. P. Synlett 1997,
521.