Alkyne-quinuclidine derivatives as potent and selective muscarinic antagonists for the treatment of COPD

Article abstract of DOI:10.1016/j.bmcl.2008.03.024

SAR around alkyne-quinuclidine derivatives allowed the discovery of highly potent muscarinic antagonists displaying interesting preferential slow off-rates from the M3 receptor.

Full text of DOI:10.1016/j.bmcl.2008.03.024

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2675–2678
Alkyne–quinuclidine derivatives as potent and selective
muscarinic antagonists for the treatment of COPD
Jean-Philippe Starck, Laurent Provins,* Bernard Christophe, Michel Gillard,
´ ´
Sophie Jadot, Patrick Lo Brutto, Luc Quere, Patrice Talaga and Michel Guyaux
UCB, R&D, Chemin du Foriest, B-1420 Braine-L’Alleud, Belgium
Received 24 January 2008; revised 5 March 2008; accepted 6 March 2008
Available online 18 March 2008
Abstract—SAR around alkyne–quinuclidine derivatives allowed the discovery of highly potent muscarinic antagonists displaying
interesting preferential slow off-rates from the M3 receptor.
Ó 2008 Elsevier Ltd. All rights reserved.
Chronic obstructive pulmonary disease (COPD) is a ma-
jor cause of morbidity and mortality worldwide. This
pathology is foreseen to become the third leading cause
of death and the fifth leading cause of morbidity over
the next 10 years. Current treatments are essentially pal-
liative, focusing on relieving symptoms, preventing
acute disease exacerbations and improving quality of
life. The understanding of the molecular and cellular
mechanisms involved in COPD allowed the identifica-
tion of many potential therapeutic approaches based
on a variety of biological targets. For example, the b2-
adrenergic or the cholinergic system, several phosphodi-
esterases or the chemokines-interleukines pathways have
been extensively investigated.1–3
M3 receptor mediates the contractile response in air-
ways smooth muscles as well as secretion from submu-
cosal glands.4
The M1 and M3 receptors are the main receptor sub-
types present in the human lungs.5 The M2 receptor is
also postulated to exist in bronchi6 although it has not
been autoradiographically visualized in human lungs.5
Current anticholinergic drugs are non-selective musca-
rinic antagonists. Their bronchorelaxing effects are
thought to be mainly mediated by the blockade of M3
receptors on lung smooth muscles although an addi-
tional contribution of M1 receptors is not excluded.
Anticholinergics drugs alone or in combination with b2
agonists are reported to be the preferred choice for the
management of COPD, at all stages of the disease. Vag-
ally-mediated reversible bronchoconstriction is an
important component of airway obstruction in COPD
patients. Three muscarinic receptors subtypes (M1R,
M2R and M3R) have been identified in the respiratory
system with each of these having been experimentally in-
volved in specific effects and physiological responses.
The M1 receptor is thought to facilitate the cholinergic
neurotransmission in parasympathetic ganglia, the M2
receptor provides negative feedback modulation on ace-
tylcholine release on postganglionic nerves whilst the
A long duration of action is an important feature to
treat chronic illnesses such as COPD. The measurement
of kinetics of drug interaction to the target protein offers
another approach to improve the duration of action of a
drug besides pharmacokinetic factors.7
This strategy has been used to discover Tiotropium bro-
mide (SpirivaÒ), the current gold standard anticholiner-
gic drug for the treatment of COPD. Contrary to
Ipratropium bromide (AtroventÒ), Tiotropium is
characterized by a slow dissociation rate from the M3
(t1/2off = 3308 min) and M1 receptors (t1/2off = 876 min).
Thanks to this remarkable property, this drug displays
a very long duration of action allowing a once-a-day
administration as an inhaled dry powder.8,9
Keywords: Muscarinic receptor(s); COPD; Quinuclidine.
*
Interestingly, Tiotropium bromide also demonstrates a
receptor-subtype kinetic selectivity with a far shorter
Corresponding author. Tel.: +32 2 3863217; fax: +32 2 3863669;
e-mail: laurent.provins@ucb-group.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.03.024

2676
J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2675–2678
dissociation from the M2 receptor (t1/2off = 88 min).
Even so the clinical impact of this latest property has
not been fully demonstrated, it is nevertheless seen as
a potential advantage to avoid a putative increased re-
lease of acetylcholine in airways but also to reduce the
risk of cardiovascular adverse effects (mainly tachycar-
dia) associated to blockade of M2 receptors on the heart
(Fig. 1).
Compounds H and I bearing, respectively, a fluorine or
a hydrogen atom instead of the hydroxyl group, were
obtained starting from D. Introduction of fluorine was
performed using DAST as the fluorinating agent, while
introduction of the hydrogen atom was carried out un-
der reductive conditions in the presence of Et3SiH and
BF3ÆOEt2.15
Finally, the bis-O-methylated analogue J was obtained
by methylation of the amino-borane complex, followed
by the deprotection of the nitrogen under acidic
conditions.
During the course of a work directed towards the iden-
tification of a potent M3 receptor antagonist as a poten-
tial new drug for the treatment of overactive bladder, we
have surprisingly discovered a new class of compounds
displaying a dissociation rate profile quite similar to
the one of tiotropium bromide.
All compounds were tested for their binding affinity16
and dissociation rates (t1/2off) from the M2 and M3 mus-
carinic receptors.17
In this article, we wish to report the structure–activity
relationships around these compounds focusing on the
modulation of R1, R2 and R3 groups that confer either
binding equilibrium or kinetic selectivity for the M3
receptor (Fig. 2). It has to be interestingly noticed that
close scaffold elements are also found in other reported
long-acting muscarinic receptor antagonists.10
This investigation was prompted by the observation that
compound 1 showed a markedly slower dissociation rate
from M3 receptor compared to M2 subtype. In order to
try to increase this selectivity within this series of mole-
cules, several analogues with different substituents at R1/
R2/R3 were prepared (Table 1).
The general method of synthesis of compounds listed in
Table 1 is outlined in Scheme 1. The majority of the
quinuclidine derivatives reported in this manuscript
were prepared according to the synthetic methodology
described in a previous paper.11
On the basis of this initial result, we speculated that the
lipophilic moiety could contribute to modulate the dis-
sociation rate from the muscarinic receptor subtypes.
First, the introduction of small cycloalkyl groups (cyclo-
butyl 2 or cyclopentyl 3) at the R3 position yielded com-
pounds displaying rather short t1/2off values, comparable
to the one of the corresponding diphenyl derivative 1.
Nevertheless, the replacement of one of the two aro-
matic rings by a cyclohexyl ring yielded 4 exhibiting
more than 15-fold t1/2off ÀM3/M2 selectivity.
Briefly, the alkyne derivative C is coupled with selected
ketones (Scheme 1) yielding D, obtained as a mixture of
diastereoisomers or as pure products after separation by
chiral chromatography.
The replacement of the hydroxyl group by different
chemical moieties has also been explored. Thus, com-
pounds E and F bearing a methyl or a cyano group were
prepared starting from the corresponding primary al-
kynes G12–14 and 3-quinuclidinone A in three steps, first
involving nitrogen protection (via an amino-borane
complex), followed by methylation and finally deprotec-
tion of the quinuclidine moiety.
This result prompted us to perform further modifica-
tions by introducing larger cycloalkyl and/or bulkier
substituents at the same position. The cycloheptyl deriv-
ative 7 dissociated 50 times more slowly from the M3
receptors than from the M2 receptors. An even better
profile was obtained with the cyclooctyl analogue 10.
The stereochemistry proved to have a large impact on
both the binding affinities (about 10-fold difference)
and the dissociation rates from the M3 receptor. In all
cases, compounds characterized by the (1R, 3R) stereo-
chemistry displayed the highest t1/2off values and, as ex-
pected, the best affinity (compare 5 and 6, 8 and 9, 11
and 12).
N+
Br-
O
S
S
OH
O
N+
O
O
HO
Ph
Br
O
However, modulating the bulkiness of the R3 substitu-
ent by introducing a linear (n-butyl, 14), a branched al-
kyl moiety (15) or an adamantyl group (13) resulted in a
faster dissociation from the M3 receptor while keeping a
rather good affinity. Of interest, the introduction of
a sulfur-containing cycloalkyl group (16) resulted in a
somewhat lower binding affinity and quite reduced dis-
sociation rates.
Ipratropium
Tiotropium
Figure 1. Reference anticholinergic agents.
O
MeO
N
OH
N
R3
R1
R2
Next, we evaluated the impact of the nature of the group
at the R1 position. Thus, replacement of the hydroxyl
group by a hydrogen atom (21), by a methyl (19), or
by a nitrile moiety (18) reduced the t1/2off value in all
N
Figure 2. Alkyne–quinuclidine scaffold.

J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2675–2678
2677
Table 1. Binding affinities and dissociation constant (t1/2off) for M3 and M2 muscarinic receptors
MeO
3
R3
1
N
R2
R1
Compound
Stereochemistry
R1
R2
R3
M3
t1/2off (min)
M2
t1/2off (min)
pKi
pKi
1
2
3R
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
9.9
9.0
9.6
9.3
9.7
8.4
9.4
9.5
8.7
8.8
9.5
8.6
8.5
8.7
58
11
9.1
7.9
8.2
8.2
8.8
7.6
8.4
8.6
7.7
7.6
8.5
7.5
7.5
7.3
6
Mixt.
Mixt.
Mixt.
1R, 3R
1S, 3R
Mixt.
1R, 3R
1S, 3R
Mixt.
1R, 3R
1S, 3R
Mixt.
Mixt.
Cyclobutyl
Cyclopentyl
Cyclohexyl
Cyclohexyl
Cyclohexyl
Cycloheptyl
Cycloheptyl
Cycloheptyl
Cyclooctyl
Cyclooctyl
Cyclooctyl
1-Adamantyl
n-Butyl
<1
<1
9
3
44
4
147
210
25
5
10
2
6
7
404
830
12
8
8£
9
5
<1
10
20
5
10
11$
12
13
14
444
1418
58
51
10
5
<1
15
16
Mixt.
Mixt.
OH
OH
Ph
Ph
9.4
8.9
42
42
8.2
7.4
3
*
*
<1
S
17
18
19
20
21
22
3R
OMe
CN
Me
F
Ph
Ph
Ph
Ph
Ph
Ph
Ph
8.0
8.6
7.8
8.8
7.9
9.5
7.4
21
6.8
8.2
7.4
8.2
7.6
8.9
<1
<1
5
Racemate
Racemate
Mixt.
Ph
Cyclohexyl
Cyclohexyl
10
147
32
9
Mixt.
H
14
3.1
Ipratropium
62
bromide
Tiotropium
bromide
23
9.7
3308
9.5
88
Except racemates, all compounds possess the 3R configuration. Mixt.: 1:1 mixture of 1R, 3R and 1S, 3R.
£: Compound 8: pA2 M3 = 8.2, pA2 M2 = 8.3.
$: Compound 11: pA2 M3 = 8.2, pA2 M2 = 7.5.
MeO
HO
MeO
O
b, c, d
a
e
N
HO
R1
N
N
N
R1
R2
R2
O
(A)
(B)
(C)
(D)
R3
Ph
(G)
Ph
g or h
b, i, j
f, b, c, d
MeO
MeO
MeO
R3
N
N
N
R3
R3
R2
R3 = H (H)
R2
R3 = OMe (J)
Ph
R1
R1
Ph
R3 = Me (E)
R3 = CN (F)
R3 = F (I)
Scheme 1. Synthesis of compounds. Reagents and conditions: (a) Li acetylide (ethylene diamine complex), THF; (b) BH3ÆTHF, THF, À10 °C; (c)
NaH, NBu4I, MeI, THF, rt; (d) 5 M HCl, acetone/Et2O, rt; (e) n-BuLi, ketone, THF, À78 °C to rt; (f) n-BuLi, alkyne, THF, À70 °C; (g) DAST,
CH2Cl2, À30 °C to À10 °C; (h) Et3SiH, BF3ÆOEt2, CH2Cl2, À40 °C; (i) NaH, MeI, THF; (j) TFA, acetone/Et2O.

2678
J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2675–2678
cases. Replacement of the hydroxyl group by fluorine
(20) was however tolerated, yielding a compound with
a t1/2off of 147 min. Methylation of the hydroxyl group
(17) also dramatically reduced the dissociation rate for
both muscarinic receptor subtypes.
Dr Christophe Genicot for careful reading of the
manuscript.
References and notes
1. Barnes, P. J. Pharmacol. Rev. 2004, 56, 515.
2. Barnes, P. J.; Hansel, T. T. Lancet 2004, 364, 985.
3. Krishna, G.; Sankaranarayanan, V.; Chitkara, R. K.
Expert Opin. Investig. Drugs 2004, 13, 255.
4. Barnes, P. J. Life Sci. 1993, 52, 521.
5. Mak, J. C.; Barnes, P. J. Am. Rev. Respir. Dis. 1990, 141,
1559.
6. Minette, P. A.; Barnes, P. J. J. Appl. Physiol. 1988, 64,
2532.
7. Dowling, M. R.; Charlton, S. J. Br. J. Pharmacol. 2006,
148, 927.
8. Panning, C. A.; DeBisschop, M. Pharmacotherapy 2003,
23, 183.
9. Hvizdos, K. M.; Goa, K. L. Drugs 2002, 62, 1195.
10. See for example: Norman, P. Expert Opin. Ther. Patents
2006, 16, 1315, and references cited therein.
The antagonistic properties of all the compounds have
been confirmed (see Table 1, data for compounds 8
and 11) using the isolated guinea pig trachea or left at-
rium both stimulated by carbachol as functionally inte-
grated M3 and M2 assays, respectively. These
compounds also induced a very slowly reversible inhibi-
tion (recovery of 20% of the maximal effect >280 min) of
the twitch contraction induced by electrical field stimu-
lation of the isolated guinea pig trachea, possibly con-
firming the long t1/2off values.
In summary, this series of quinuclidine derivatives led to
potent muscarinic antagonists, displaying interesting
preferential slow off-rate from the muscarinic M3 recep-
tor, versus the M2 subtype. The reasons for that behav-
iour remain unclear though a link to lipophilicity of the
R3 substituent may be evoked. We are working towards
the elucidation of these findings.
´ ´
11. Starck, J.-P.; Talaga, P.; Quere, L.; Collart, P.; Chris-
tophe, B.; Chimmanamada, D.; Zanda, M.; Wagner, A.;
Mioskowski, C.; Guyaux, M. Bioorg. Med. Chem. Lett.
2006, 16, 373.
12. Dehmlow, E. Tetrahedron Lett. 1971, 563.
13. Arumugam, S.; Verkade, J. G. J. Org. Chem. 1997, 62, 4827.
14. Norman, R. O. C.; Thomas, C. B. J. Chem. Soc. B 1967,
598.
15. Batt, D. G.; Maynard, G. D.; Petraitis, J. J.; Shaw, J. E.;
Galbraith, W.; Harris, R. R. J. Med. Chem. 1990, 33, 360.
16. Affinities for the M3 and M2 muscarinic receptors were
determined by competition binding experiments with [3H]-
N-methylscopolamine performed in CHO cells membranes
expressing recombinant human receptors.
Compounds 8 and 11 displaying dissociation rates and
selectivity comparable to tiotropium bromide have been
selected for extensive biological and pharmacological
evaluation, as potential therapeutic candidates for the
treatment of COPD. These results will be reported in
due course.
Acknowledgements
17. Koff on both receptors are measured following 1 h asso-
ciation at 25 °C followed by dissociation by atropine
(10À6M). t1/2off are calculated according to the equation:
´
The authors thank Dr Veronique Pinilla and her team
for chiral chromatography, Dr Corinne Audouin and
t1/2off = Ln2/Koff
.