Novel CGRP receptor antagonists from central amide replacements causing a reversal of preferred chirality

Article abstract of DOI:10.1016/j.bmcl.2010.08.105

A previously utilized quinoline-for-N-phenylamide replacement strategy was employed against a central amide in a novel class of CGRP receptor antagonists. A unique and unexpected substitution pattern was ultimately required to maintain reasonable affinity

Full text of DOI:10.1016/j.bmcl.2010.08.105

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6827–6830
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel CGRP receptor antagonists from central amide replacements
causing a reversal of preferred chirality
Michael R. Wood a, , Kathy M. Schirripa a, June J. Kim a, Rodney A. Bednar a, John F. Fay a,
Joseph G. Bruno a, Eric L. Moore b, Scott D. Mosser a, Shane Roller b, Christopher A. Salvatore b,
Joseph P. Vacca a, Harold G. Selnick a,
⇑
a Department of Medicinal Chemistry, Merck & Co., Inc., PO Box 4, 770 Sumneytown Pike, West Point, PA 19486, United States
b Department of Pain Research, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A previously utilized quinoline-for-N-phenylamide replacement strategy was employed against a central
amide in a novel class of CGRP receptor antagonists. A unique and unexpected substitution pattern was
ultimately required to maintain reasonable affinity for the CGRP receptor, while at the same time predict-
ing acceptable heterocycle positioning for related analogs. Subsequently, specific quinoline and naph-
thyridine compounds were prepared which supported these structural predictions by displaying CGRP
binding affinities in the 0.037–0.15 nM range.
Received 21 July 2010
Revised 18 August 2010
Accepted 20 August 2010
Available online 26 August 2010
Keywords:
CGRP
Ó 2010 Elsevier Ltd. All rights reserved.
Calcitonin gene-related peptide
Antagonist
A recent publication from these laboratories disclosed a novel
class of CGRP receptor antagonists that were developed via
a rational design strategy initiated by the simplification of target
structures followed by a re-optimization along a different path.8
Part of this structural simplification exercise entailed the replace-
ment of a central quinoline with an N-phenylamide in order to
facilitate the more rapid production of analogs. Naturally, it would
make sense to eventually explore what benefits could be obtained
by once again reinstating the quinoline in any subsequent series of
new lead.
One of the earliest leads containing a central amide, compound
1, shown in Figure 1 as the preferred (S) enantiomer has a
Ki = 1.9 nM against a human recombinant CGRP receptor. The
straightforward production of a quinoline core analog whereby
the position of the amidic nitrogen becomes the position for the
quinoline nitrogen, according to known chemistry,9 provided
racemic compound 2. Contrary to analogous substitutions on ear-
lier lead series, which had minimal effect on CGRP receptor bind-
ing affinity, the introduction of this quinoline core resulted in a
The calcitonin gene-related peptide (CGRP) is a 37 amino acid
neuropeptide which exhibits very potent vasodilation activity and
is strongly implicated in the pathogenesis of migraine.1 In particu-
lar, IV administration of CGRP to established migraineurs induces
migraine-like headaches.2 Conversely, treatment of migraines with
sumatriptan results in a normalization of circulating CGRP levels
concomitant with pain relief.3 More compelling, however, is the
clinically demonstrated success of the two small molecule CGRP
receptor antagonists, olcegepant and telcagepant.4 Proof of concept
was first achieved with an IV formulation of olcegepant in a phase II
clinical trial, which demonstrated efficacy similar to that of the trip-
tans.5 Although the development of olcegepant appears on hold, the
oral CGRP receptor antagonist telcagepant has progressed to phase
III clinical studies. In those studies, telcagepant has been shown to
be efficacious and well-tolerated, while demonstrating a lower inci-
dence of adverse effects than zolmitriptan.6,7
CF3
O
N
O
nearly 3 log unit decrease in Ki to a value of 2.2
the two individual enantiomers of 2, neither of which would be
expected to have a CGRP receptor affinity of less than ꢀ1 M,
lM. Examining
N
N
O
H
l
N
F
NH
F
we realized that (S)-2 displayed a spatially similar connectivity
to that of compound 1 with respect to the N-acyl-benzyl amine
terminus (drawn suggestively for easy mental overlays). While
compound (R)-2 displayed the positioning of the quinoline nitro-
gen similar to known and potent quinoline CGRP receptor antag-
onists.8,9 Realization of these two critical pieces of SAR prompted
the preparation of racemic compound 3 which displayed more
N
telcagepant
⇑
Corresponding author. Tel.: +1 215 652 5334; fax: +1 215 652 7310.
E-mail addresses: harold_selnick@merck.com, hal_selnick@merck.com (H.G.
Selnick).
Present address: Department of Pharmacology, Vanderbilt University, USA.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.105

6828
M. R. Wood et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6827–6830
O
O
O
H
N
O
O
O
X
Y
NH
NH
NH
t-Bu
N
t-Bu
N
t-Bu
N
O
N
O
N
O
N
O
N
Me
Me
Me
1
2, X = N, Y = CH
3, X = CH, Y = N
4
O
O
O
N
NH
+
NH
t-Bu
N
t-Bu
N
( )-2 =
N
O
N
Me
O
N
Me
O
(S)-2
(R)-2
Figure 1. Introduction of quinoline cores to the lead structure 1.
Table 1
reasonable CGRP Ki = 35 nM. Ultimately the single (R) enantiomer
4, predicted by the above analysis of (S)-S and (R)-2, suffered a
much smaller decrease in binding affinity relative to the starting
amide analog 1 by displaying a Ki of 12 nM. The interesting
change in preferred spirohydantoin chirality between 1 and 4 re-
sults from the change in location of their amidic and quinolinic
nitrogens, respectively, but does not reflect a shift in positioning
of their N-acyl-benzyl amine termini. This would suggest that ba-
sic nitrogens or heteroatoms might be best tolerated on one side
of these quinolines, and related, structures (vide infra).
CGRP receptor binding affinities for the azaoxindoles 8–11 and 18–20
O
O
NH
t-Bu
N
R
N
X
Y
Compda
R
X
Y
CGRP Kia b
, (nM)
8
N
CH
1.3
The preparation of 4 is depicted in Scheme 1 starting from the
known hydantoin 510 and the ‘vinamidinium’ bis-tetrafluoroborate
salt 611 to give quinoline aldehyde 7. The aldehyde is then reductively
aminated and acylated under standard conditions to provide 4.
Although the binding affinity for 4 was encouraging we knew
from prior experience12 that there was the potential for an up to
10-fold improvement in Ki upon replacement of the hydantoin
with an azaoxindole to provide 8, shown in Table 1.13 Indeed, an
approximate ninefold improvement in binding affinity was
achieved to arrive at a Ki of 1.3 nM for compound 8. Applying
known SAR derived from the anilide series analogous to 18 we
were surprised to find some interesting departures. While intro-
duction of two fluorines to give 9 still improved Ki by roughly four-
fold this effect was about half of that previously observed.
However, the benefit to CGRP receptor binding affinity resulting
from the presence of a benzylic methyl in compound 10 encom-
passed an order of magnitude improvement rather than a simple
doubling of affinity as previously observed. Furthermore, the com-
bination of both fluorination and methylation was not additive in
this instance, but instead caused a decrease in CGRP receptor bind-
ing affinity for compound 11, relative to the non-fluorinated analog
10. Although the presence of two fluorines did not appear to be
generally beneficial an attempt was already underway to produce
F
9
10
11
N
N
N
CH
CH
CH
0.69
0.10
0.12
F
F
Me
F
Me
18
19
20
N
CH
N
0.037
0.11
0.15
CH
N
N
a
Values represent the numerical average of at least two experiments. Interassay
variability was 10% for the binding assays.
Ki values for competition with 125I-hCGRP determined using membranes from
b
HEK293 cells stably expressing cloned human CLR/RAMP1.
NMe2
O
Me
N
NH
-
(BF4 )2
O
O
+
Me
Me
Me
N
O
H2N
NH
N
N
F
Me Me
Me
N
N
F
5
6
a
O
O
12
b,c
NH
H
4
Figure 2. Lactam constraint applied to compound 11 to arrive at compound 12.
N
Me
O
N
7
a lactam-constrained version of 11. As shown in Figure 2, com-
pound 12 actually lost CGRP receptor affinity providing a Ki of
0.30 nM, and further demonstrated that although the exchange of
Scheme 1. Synthesis of compound 4. Reagents and conditions: (a) HOAc, KOAc,
90 °C, 40–60%; (b) benzylamine, NaHB(OAc)3, CHCl3, 70–90%; (c) pivaloyl chloride,
N-methyl morpholine, 0 °C, 85–95%.

M. R. Wood et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6827–6830
6829
a central amide for a central quinoline was tolerated at the CGRP
receptor, this modification had significantly altered preferred
structures in this new series.
shown for compound 18 (Table 1) provided a nearly threefold
improvement in Ki relative to the less conformationally biased
compound 10. This sub-nanomolar affinity for the CGRP receptor
puts compound 18 in a similar potency range as some of our best
antagonists published to date.8
Returning to the idea that nitrogen atoms might only be toler-
ated on the side opposite to that of the terminal alkyl attachment
(see compounds 2 and 3), an alternative quinoline was prepared
wherein the nitrogen is now at the Y position in the Table 1 parent
structure. This modification present in compound 19 was toler-
ated, but caused a minor decrease in the CGRP receptor binding
affinity. Further demonstrating that nitrogen atoms were tolerated
at both X and Y positions simultaneously naphthyridine 20 dis-
played a CGRP Ki similar to 19. Neither analog displayed a Ki supe-
rior to quinoline 18.
Preparation of 19, according to Scheme 3, began with the reduc-
tion of nitro-aldehyde 21. Condensation of aniline 22 with ketone
2315 gave ester 24. Direct reduction of this ester with hydride
sources proved problematic, so a two step procedure of hydrazide
formation, followed by reduction with potassium ferrocyanide
delivered aldehyde 25. The SEM protecting group on 25 was then re-
moved, followed by reductive amination of the aldehyde with (R)-1-
aminoindane and subsequent acylation to produce 19. In a similar
manner compound 20 was prepared according to Scheme 4 starting
from the commercially available iodide 26. Condensation of 26 with
ketone 2715 gave naphthyridine 28. Installation of the required alde-
hyde 29 occurred through initial vinylation, followed by osmium
tetroxide catalyzed oxidation. Reductive amination of naphthyri-
dine aldehyde 29 with (R)-1-aminoindane, was followed by tert-bu-
tyl protecting group removal, and ultimately by carefully controlled
acylation with pivaloyl chloride to give 20. This alternative order of
chemical transformations, relative to Scheme 3, was necessitated
due to the chemical instability of aldehyde-naphthyridine to the
strongly acidic conditions used to cleave the tert-butyl protecting
group.
Preparationof12initiatedwiththediester13showninScheme2.
Carefully controlled hydrolysis of the less hindered ester produced
14. Weinreb amide formation followed by aryl Grignard addition
yielded the ketone 15. The use of Ellman sulfinimide methodolo-
gies14 allowed the introduction of the primary amine in 16, concom-
itantly the methyl ester was transformed to an ethyl ester resulting
from the use of titanium tetraethoxide. Reductive alkylation of 16
with aldehyde 17 and subsequent lactamization at elevated temper-
atures provided 12.
Moving forward, cognizant that present SAR suggested that
difluorination and lactam constraints were no longer preferred
characteristics in this quinoline series of CGRP receptor antago-
nists, attempts were made to bias the position of the pendent aryl
ring in alternative fashions. The (R)-1-aminoindanyl constraint
Me
Me
Me
Me
CO2Me
CO2R
CO2Me
O
b, c
F
13, R = Me
14, R = H
a
15
F
d, e
O
O
O
OEt
NH2
NH
Me
Me
H
F
+
N
N
17
16
F
f, g
In summary, a modified quinoline for N-phenyl amide core
replacement strategy was successfully employed to arrive at a no-
vel series of CGRP receptor antagonists which displayed unique
SAR relative to their amide progenitors. Additionally, predictions
based on the tolerability of central nitrogen atoms allowed for
the preparation of similarly novel quinoline and naphthyridine
analogs which displayed sub-nanomolar affinity for the CGRP
receptor.
12
Scheme 2. Synthesis of compound 12. Reagents and conditions: (a) K2CO3, THF/
MeOH/H2O (2:3:2), rt, 54%; (b) oxalyl chloride, NEt3, DCM, MeHNOMe, 0 °C, 79%; (c)
3,5-difluorophenyl magnesium bromide, THF, 0 °C, 85%; (d) (i) (S)-tert-butyl
sulfinimide, Ti(OEt)4, THF, 60 °C; (ii) NaHB(OAc)3, 0 °C, 72%; (e) (i) HCl(g), MeOH,
0 °C, 30 min; (ii) aqueous NaHCO3, 99%; (f) NaHB(OAc)3, HOAc, CHCl3, rt, 95%; (g)
xylenes/HOAc (9:1), 140 °C, 48%.
O
X
O
SEM
N
MeO2C
N
+
H
O
O
O
t-Bu
N
I
N
N
H
21, X = NO2
22, X = NH2
23
+
a
O
N
NH2
b
26
27
a
O
SEM
N
R
e, f, g
N
O
t-Bu
N
19
R
e, f, g
N
20
N
N
24, R = CO2Me
25, R = C(H)O
c, d
28, R = I
29, R = C(H)O
b, c
Scheme 3. Synthesis of compound 19. Reagents and conditions: (a) Pd/C, H2
(1 atm.), MeOH, 45%; (b) MeOH, piperidine, 75 °C, boil dry (2Â), 50%; (c) EtOH,
hydrazine, 70 °C, quant.; (d) K4Fe(CN)6, concd NH4OH, DCM, H2O, rt, 38%; (e)
(i) HCl(g), MeOH, rt, 18 h; (ii) concentrate in vacuo; (iii) MeOH, concd NH4OH, 94%;
(f) (R)-1-aminoindane, NaHB(OAc)3, HOAc, CHCl3, 95%; (g) pivaloyl chloride, NEt3,
DCM, 0 °C, 63%.
Scheme 4. Synthesis of compound 20. Reagents and conditions: (a) EtOH,
piperidine, boil dry (3Â), 65%; (b) bis(triphenylphosphine) palladium(II) chloride,
tributyl(vinyl)stannane, dioxane, 85 °C, 18 h, 62%; (c) OsO4(cat.), NaIO4, THF, H2O,
rt, 2 h, 35%; (d) (R)-1-aminoindane, NaHB(OAc)3, HOAc, CHCl3, 29%; (e) MeSO3H
(neat), 55 °C, 2 h, quant.; (f) pivaloyl chloride, NEt3, DCM, 0 °C, 50%.

6830
M. R. Wood et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6827–6830
7. Ho, T. W.; Ferrari, M. D.; Dodick, D. W.; Galet, V.; Kost, J.; Fan, X.; Leibensperger,
Acknowledgments
H.; Froman, S.; Assaid, C.; Lines, C.; Koppen, H.; Winner, P. K. Lancet 2008, 372,
2115.
The authors thank the MRL West Point analytical chemistry,
mass spectrometry and NMR spectroscopy groups for analytical
and spectroscopic data.
8. Wood, M. R.; Schirripa, K. M.; Kim, J. J.; Quigley, A. G.; Stump, C. A.; Bell, I. M.;
Bednar, R. A.; Fay, J. F.; Bruno, J. G.; Moore, E. L.; Moser, S. D.; Roller, S.;
Salvatore, C. A.; Kane, S. A.; Vacca, J. P.; Selnick, H. G. Bioorg. Med. Chem. Lett.
2009, 19, 5787.
9. Stump, C. A.; Bell, I. M.; Bednar, R. A.; Fay, J. F.; Gallicchio, S. N.; Hershey, J. C.;
Jelley, R.; Kreatsoulas, C.; Moore, E. L.; Mosser, S. D.; Quigley, A. G.; Roller, S. A.;
Salvatore, C. A.; Sharik, S. S.; Theberge, C. R.; Zartman, C. B.; Kane, S. A.; Graham,
S. L.; Selnick, H. G.; Vacca, J. P.; Williams, T. M. Med. Chem. Lett. 2010, 20, 2572.
10. Greater experimental detail can be found in WO2007061676. All final
compounds were characterized by 1H NMR, HPLC (purity >95%) and HRMS.
11. Tom, N. J.; Ruel, E. M. Synthesis 2001, 9, 1351.
12. Stump, C. A.; Bell, I. M.; Bednar, R. A.; Bruno, J. G.; Fay, J. F.; Gallicchio, S. N.;
Johnston, V. K.; Moore, E. L.; Mosser, S. D.; Quigley, A. G.; Salvatore, C. A.;
Theberge, C. R.; Zartman, C. B.; Zhang, X.-F.; Kane, S. A.; Graham, S. L.; Vacca, J.
P.; Williams, T. M. Bioorg. Med. Chem. Lett. 2009, 19, 214.
References and notes
1. For a review on the relationship between CGRP and migraine see: Tepper, S. J.;
Stillman, M. J. Headache 2008, 48, 1259.
2. Lassen, L. H.; Haderslev, P. A.; Jacobsen, V. B.; Iversen, H. K.; Sperling, B.; Olesen,
J. Cephalalgia 2002, 22, 54.
3. Goadsby, P. J.; Edvinsson, L. Ann. Neurol. 1993, 33, 48.
4. Paone, D. V.; Shaw, A. W.; Nguyen, D. N.; Burgey, C. S.; Deng, Z. J.; Kane, S. A.;
Koblan, K. S.; Salvatore, C. A.; Hershey, J. C.; Wong, B.; Roller, S. G.; Miller-Stein,
C.; Graham, S. L.; Vacca, J. P.; Williams, T. M. J. Med. Chem. 2007, 50, 5564.
5. Olesen, J.; Diener, H.-C.; Husstedt, I. W.; Goadsby, P. J.; Hall, D.; Meier, U.;
Pollentier, S.; Kesko, L. M. N. Eng. J. Med. 2004, 350, 1104.
13. The preparation of 8, and close analogs, proceeds by direct analogy to 4.
However, further experimental details can be found in WO2007061677.
14. Wood, M. R.; Gallicchio, S. N.; Selnick, H. G.; Zartman, C. B.; Bell, I. M.; Stump, C.
A. U.S. PCT Appl. Publ. 2007265225, 2007.
6. Ho, T. W.; Mannix, L. K.; Fan, X.; Assaid, C.; Furtek, C.; Jones, C. J.; Lines, C. R.;
Rapoport, A. M. Neurology 2008, 70, 1304.
15. Ketones 23 and 27 will be the subject of future communications.