Design and pharmacology of quinuclidine derivatives as M2-selective muscarinic receptor ligands

Article abstract of DOI:10.1016/S0960-894X(01)00186-X

In our search for M₂-selective muscarinic receptor antagonists, we synthesized 1,3-disubstituted indenes. The effects of different basic moieties with regard to binding and selectivity towards the five distinct muscarinic receptor subtypes were investigated. The results show that the quinuclidine series afforded the most promising compounds in terms of both receptor affinity and M₂-subtype selectivity.

Full text of DOI:10.1016/S0960-894X(01)00186-X

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1241–1243
Design and Pharmacology of Quinuclidine Derivatives as
M -Selective Muscarinic Receptor Ligands
2
a,
b
a
b
Thomas M. Bo
¨
hme, * Christine Keim, Gerd Dannhardt, Ernst Mutschler
and Gu
nter Lambrechtb
¨
aInstitute of Pharmacy, University of Mainz, Staudinger Weg 5, D-55099 Mainz, Germany
bDepartment of Pharmacology, Biocentre Niederursel, University of Frankfurt,
Marie-Curie-Straße 9, D-60439 Frankfurt, Germany
Received 3 January 2001; accepted 14 March 2001
2
Abstract—In our search for M -selective muscarinic receptor antagonists, we synthesized 1,3-disubstituted indenes. The effects of
different basic moieties with regard to binding and selectivity towards the five distinct muscarinic receptor subtypes were investi-
gated. The results show that the quinuclidine series afforded the most promising compounds in terms of both receptor affinity and
2
M -subtype selectivity. # 2001 Elsevier Science Ltd. All rights reserved.
Five distinct muscarinic receptors have been cloned
M –M ), which has made it possible to identify selec-
tive muscarinic receptor agonists and antagonists. Sev-
lipophilic group for binding into an aromatic cavity of
9
(
the receptor. Following these rules, benzofulvenes were
1
5
synthesized that fulfill the requirements mentioned
above. The 2,3-disubstituted indene, (S)-dimethindene,
has been shown to exhibit antagonist activity towards
eral M -selective compounds have been discovered (e.g.,
2
tripitramine, himbacine, AF-DX 384) and tested for in
vitro and in vivo selectivity.1 Most of these com-
ꢀ3
the muscarinic receptors with selectivity for the M
1
2
0
pounds either have a low subtype selectivity, are not
4
subtype. In our efforts to synthesize new M -selective
2
very potent, or do not penetrate into the brain. The
muscarinic antagonists, these same building blocks were
used.
objective of this investigation was to discover new M2-
selective muscarinic receptor antagonists capable of
crossing the blood–brain barrier with the potential for
The affinity profiles for the five human receptor sub-
types (M –M ) of this series of compounds have been
studied by the use of transfected Chinese hamster ovar-
detecting the loss of M receptors with positron emission
2
1
5
5
tomography (PET) in Alzheimer’s disease patients.
3
ian (CHO) cells and the displacement of [ H]-NMS (N-
1
,11
Due to the lack of crystal structures of the muscarinic
receptor proteins, the binding sites for agonists or
antagonists are still unknown. Furthermore, the five
mammalian muscarinic receptor subtypes display a high
degree of sequence identity, which makes it very difficult
methylscopolamine). The most promising com-
pounds in terms of both M receptor affinity and selec-
2
tivity contain a quinuclidine moiety. Exchanging the 2-
pyridyl moiety with 3- or 4-pyridine did not significantly
alter potency (Table 1).
6
to identify highly potent and selective compounds.
Because several studies have shown that an aspartate
residue represents a key element of the ligand binding
site, a basic amine appears to be crucial to elicit high
potency. It is believed that highly active muscarinic
receptor agonists require hydrogen bond acceptor sites,8
whereas muscarinic receptor antagonists need a bulky
The synthesis of the benzofulvenes is shown in Scheme
1. Treatment of indene (I) with butyllithium followed by
addition of (2-chloro-ethyl)-dimethylamine gave II. The
substituted indene II was condensed with pyridine car-
baldehydes to give 1–2. Treatment of I with n-butyl-
7
1
2
lithium and then dry ice yielded acid III. 3H-Indene-1-
carbonyl chloride (IV) was prepared by adding thio-
nylchloride to a suspension of 3H-indene-1-carboxylic
1
3
acid (III) in benzene. The acid chloride IV treated with
various amines and alcohols gave the corresponding
*Corresponding author. Fax:+49-69-305-84400; e-mail: thomas.
boehme@aventis.com
0
960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00186-X

1242
T. M. Bo¨hme et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1241–1243
substituted indenes V. Treatment of V with pyridine
carbaldehyde in the presence of potassium hydroxide
yielded 9–14. For 3–8 no potassium hydroxide was used
for catalysis.
selectivity. The compounds 1 and 2, compared to the
closely related structures with ester moieties 6 and 7, do
not display a significant difference in affinity, which led
to the assumption made above. The only reason for the
relatively high affinity of the compounds 3, 4, and 5 is
the quinuclidine moiety. According to double mutant
A series of benzofulvenes has been synthesized and tes-
ted for binding towards the five distinct muscarinic
receptor subtypes. Compounds 3, 4, and 5 displayed
the highest affinities for the muscarinic receptors
exhibiting a 19- to 129-fold higher affinity at M2
receptors compared to the other 1,3-disubstituted
indenes. However, compounds 3–5 exhibited a decrease
cycle analysis on M muscarinic receptors, Asp103 in
2
the third transmembrane region (TM3) is the ligand
amine counterion and therefore crucial for both agonist
1
4
and antagonist binding. Quinuclidine as the basic part
in affinity (7- to 9-fold) at M receptors compared to
2
dimethindene.
All compounds with basic moieties other than quinucli-
dine (1–2 and 6–14) show weak binding affinities at the
five different muscarinic receptor subtypes. The ester
group of compounds 3–8 most likely does not form
hydrogen bonds with certain amino acids like Asn404 of
the M receptor, because the introduction of an ester
2
8
group did not influence binding. Compounds 3, 4, and
5
receptor subtype (3: M /M =5-fold; M /M =5-fold;
only show increased selectivity for the M muscarinic
2
1
2
3
2
M /M =5-fold; M /M =3-fold). These compounds
4
2
5
2
with a quinuclidine moiety display a 37-fold increase of
affinity compared to 6, which contains the (2-dimethyl-
amino)-ethyl group. The distance between the tertiary
amine to the indene-pyridine moiety as well as the posi-
tion of the nitrogen in the pyridine residue seems to
have no significant effects with regard to affinity or
3
Scheme 1. (a) n-BuLi, Cl-(CH
2
)
2
N(CH
3
)
2
, ether; (b) R -CHO, KOH,
, ether; (d) SOCl , benzene; (e) X, THF
methanol; (c) n-BuLi, CO
2
2
0
(
X=quinuclidinol; N,N,N -trimethyl-ethane-1,2-diamine; N,N-dime-
thyl-ethane-1,2-diamine).
Table 1.
Dimethindene
1–14
b
pK
i
Example
—
R1
R2
M
1
M
2
M
3
M
4
M
5
Dimethindenec
—
6.58
5.02
4.73
5.83
6.05
5.90
5.15
4.94
4.92
4.57
4.55
4.62
4.66
4.31
7.35
4.90
4.82
6.49
6.38
6.48
4.92
5.13
4.98
4.38
4.53
4.80
5.04
4.85
4.41
6.72
4.67
4.56
5.76
5.77
5.68
4.85
5.00
4.82
4.21
4.39
4.34
4.20
4.14
4.01
6.39
4.92
4.65
5.77
5.91
5.74
4.91
4.90
4.83
4.47
4.67
4.61
4.33
4.26
4.05
6.10
4.72
4.57
6.01
5.91
5.98
4.75
4.52
4.57
4.26
4.18
4.27
4.28
4.27
4.26
a
f
1
(CH
(CH
2
)
)
2
N(CH
N(CH
3
)
)
2
2
2-Pyridine
3-Pyridine
2-Pyridine
3-Pyridine
4-Pyridine
2-Pyridine
3-Pyridine
4-Pyridine
2-Pyridine
3-Pyridine
4-Pyridine
2-Pyridine
3-Pyridine
4-Pyridine
—
2
3
4
5
6
7
8
9
2
2
3
d
CO
CO
CO
2
2
2
quinuclidinyl
quinuclidinyld
quinuclidinyld
2 2 2 3
CO (CH ) N(CH )
f
f
f
f
f
CO
CO
2
(CH
(CH
2
)
2
2
N(CH
3
N(CH )
3
)
2
2
f
2
2
)
CONH(CH
CONH(CH
COHN(CH
2
2
2
)
)
)
2
2
2
N(CH
N(CH
N(CH
N(CH
N(CH
e
3
3
3
)
)
)
2
2
2
1
1
1
1
1
0
1
2
3
4
CON(CH
CON(CH
3
)(CH
)(CH
COmethylpiperazine
Tripitramine
AF-DX 384
2
)
)
2
3
)
)
2
2
3
2
2
3
f
4.53
8.80
7.51
6.97
—
—
—
9.57
8.22
8.00
7.42
7.18
7.03
8.19
8.00
7.96
7.47
6.27
6.31
—
—
Himbacine
a
Compounds 1–14 show E-configuration determined by nuclear overhauser effect (NOE) experiments.
pK
b
i 1 5
values at muscarinic M –M receptors were determined in radioligand binding studies at CHO cell membranes. Data are given of at least three
1
11
experiments performed in duplicate. SD ꢁ0.13 in all cases. Complete protocol is described by Do
¨
rje et al. and Buckley et al.
c
Dimethindene, racemic dimethindene.
Quinuclidinyl, 3-substituted (R,S)-quinuclidine.
d
e
0
Methylpiperazine, N-substituted N -methylpiperazine.
Hill coefficients significantly different from unity.
f

T. M. Bo¨hme et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1241–1243
1243
of the molecules has clearly shown advantages over the
References
¨
. Dorje, F.; Wess, J.; Lambrecht, G.; Tacke, R.; Mutschler,
E.; Brann, M. R. J. Pharmacol. Exp. Ther. 1991, 256, 727.
. Angeli, P.; Cantalamessa, F.; Gulini, U.; Melchiorre, C.
Naunyn-Schmiedeberg’s Arch. Pharmacol. 1995, 352, 304.
. Chiarini, A.; Budriesi, M. L.; Bolognesi, A.; Minarini, A.;
other basic moieties. It exhibits not only at least 19-fold
higher affinity, but also shows selectivity for the M2
binding pocket. It is likely that the selectivity between
the muscarinic subtypes is based on conformation
1
2
1
5
rather than on single amino acid residues. The rigid
quinuclidine is not able to undergo a conformational
change and probably shows therefore the observed
selectivity. Other basic structures as the ethyl dimethy-
lamine moiety are flexible and show no selectivity for
one muscarinic receptor subtype.
3
Melchiorre, C. Br. J. Pharmacol. 1995, 114, 1507.
4. Cuello, A. C. Trends Pharmacol. Sci. (Suppl.) 1995, 16, 73.
5. Carson, R. E.; Kiesewetter, D. O.; Jagoda, E.; Der, M. G.;
Herscovitch, P.; Eckelman, W. C. J. Cerebr. Blood F. Met.
1
6
7
8
998, 18, 1130.
. Bonner, T. I. Trends Neurosci. 1989, 12, 148.
. Caulfield, M. P. Pharmacol. Ther. 1993, 58, 319.
. Heitz, F.; Holzwarth, J. A.; Gies, J.-P.; Pruss, R. M.;
Trumpp-Kallmeyer, S.; Hibert, M. F.; Guenet, C. Eur. J.
Pharmacol. 1999, 380, 183.
In conclusion, the series of fulvenes 1–14 demonstrate
affinities towards the five muscarinic receptors, and
compounds with the quinuclidine moiety clearly show
the highest affinities. More importantly, the selectivity
9. Nilsson, B. M.; Sundquist, S.; Johansson, G.; Nordvall, G.;
Glas, G. J. Med. Chem. 1995, 38, 473.
for M receptors makes quinuclidine as the ligand amine
2
very valuable in the search for new M selective mus-
2
carinic receptor antagonists, especially for related com-
pounds such as dimethindene.
10. Pfaff, O.; Hildebrandt, C.; Waelbroeck, M.; Hou, X.;
Moser, U.; Mutschler, E.; Lambrecht, G. Eur. J. Pharmacol.
1995, 286, 229.
1
1. Buckley, N. J.; Bonner, T. I.; Buckley, C. M.; Brann,
M. R. Mol. Pharmacol. 1989, 35, 469.
2. Noland, W. E.; Landucci, J. A.; Kameswaran, V. J. Org.
Chem. 1980, 45, 3456.
3. Knowles, W. S.; Kuck, J. A.; Elderfield, R. C. J. Org.
Chem. 1942, 7, 374.
4. Vogel, W. K.; Peterson, G. L.; Broderick, D. J.; Mosser,
V. A.; Schimerlik, M. I. Arch. Biochem. Biophys. 1999, 361, 283.
15. Liu, J.; Schoneberg, T.; van Rhee, M.; Wess, J. J. Biol.
Chem. 1995, 270, 19532.
1
Acknowledgements
1
The authors thank the Institute of Chemistry of the
University of Mainz, Germany for spectral and analy-
tical determinations, the Fonds der Chemischen Indus-
trie, Germany and the Deutsche Forschungsgemeinschaft
for financial support.
1
¨