Synthesis of potent sigma-1 receptor ligands via fragmentation of dextromethorphan

Article abstract of DOI:10.1016/j.bmcl.2003.12.105

Treatment of dextromethorphan 1 with various alkylating agents followed by base treatment led to Hoffman-type elimination reactions to produce a series of tricyclic derivatives, 6. These derivatives were characterized in vitro as sigma-1 receptor ligands.

Full text of DOI:10.1016/j.bmcl.2003.12.105

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1807–1809
Synthesis of potent sigma-1 receptor ligands via fragmentation
of dextromethorphan
Mark P. Arrington, Claire Brown and C. Eric Schwartz*
UCB Research, Inc., 840 Memorial Drive, Cambridge, MA 02139, USA
Received 29 October 2003; accepted 19 December 2003
Abstract—Treatment of dextromethorphan 1 with various alkylating agents followed by base treatment led to Hoffman-type elim-
ination reactions to produce a series of tricyclic derivatives, 6. These derivatives were characterized in vitro as sigma-1 receptor
ligands.
#
2004 Elsevier Ltd. All rights reserved.
Dextromethorphan (1, Scheme 1) is one of the most
widely used antitussive agents despite the fact that its
precise mechanism of action has not been fully eluci-
dated. Dextromethorphan displays antagonist activity
While dextromethorphan is a nonnarcotic antitussive
4
with a cleaner profile than codeine, we were interested
in preparing zwitterionic derivatives in an attempt to
limit CNS penetration and related side effects while
maintaining antitussive activity via activity at peripheral
1
at the cationic NMDA receptor channel and agonist
2
5
6
activity at sigma receptors, suggesting that the anti-
sigma receptors. Accordingly, we treated dextrorphan
tussive activity may involve action at either or both of
these receptors. Indeed, the antitussive effect of dex-
tromethorphan in animal models of citric acid-induced
cough was reversed by rimcazole, a sigma receptor
antagonist, suggesting a role for the sigma receptor in
modulating the cough response.
(2) with t-butyl bromoacetate in the presence of cesium
carbonate with the expectation of isolating the glycolate
3 (Scheme 1); subsequent deprotection of the t-butyl
ester would then provide the desired zwitterion. How-
ever, the major product from this reaction was the tri-
cycle 5, isolated in modest yield (ca. 35%). This product
3
Scheme 1.
*
Corresponding author. Fax: +1-617-547-8481; e-mail: eric.schwartz@ucb-group.com
0
960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.12.105

1808
M. P. Arrington et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1807–1809
Scheme 2.
Scheme 3.
presumably arises from O- and N-alkylation to form the
quaternary ammonium salt 4 followed by a based-
of 5 and was derived from levorphanol as shown in
Scheme 4.
7
induced fragmentation to afford 5. Somewhat surpris-
8
ingly, 5 was an extremely potent sigma-1 receptor
9
All the tricycles 6 were examined in vitro for their sigma-1
9
ligand with a K =14 nM. This finding prompted us to
i
receptor affinity. The results are shown in Table 1,
1
3a
prepare additional analogues and examine their in vitro
affinities for the sigma-1 receptor.
along with the values for 5 and dextromethorphan
(1), and some trends are readily observable. The aryl
glycolate moiety of 5 is not necessary for high affinity as
the aryl methyl ether 6a retains significant affinity for
the receptor. Removal of the ester group from 6a results
in a 6.5-fold loss of affinity (6b), but this can be regained
by addition of small lipophilic substituents to the nitro-
gen atom (6c,d). However, the ethoxybenzyl-substituted
analogue 6e shows a loss of affinity suggesting that the
phenyl ring of this substituent is not well accom-
modated by the receptor.
We found that the yield of 5 could be improved to 55%
when the transformation was carried out in two steps:
alkylation to form the diastereomeric quaternary
1
0
ammonium salts 4 (BrCH CO t-Bu, EtOAc) followed
2
2
ꢀ
to provide 5. Similar reactions with dextromethorphan,
by base-induced fragmentation (DBU, CH CN, 80 C)
3
1, and various electrophiles allowed easy access to a
variety of tertiary amine-substituted tricycles (6,
Scheme 2).
The nature of the nitrogen atom is also important for
receptor binding. The carbamate 6f is inactive, demon-
strating the need for a basic nitrogen atom. The sec-
ondary amine 6g is less potent than the corresponding
tertiary amine 6b, and the quaternary ammonium salt
6h loses all affinity for the receptor. The tertiary amine
The dimethylamine 6b was used to generate two addi-
tional analogues, as shown in Scheme 3. Demethylation
1
1
with ethyl chloroformate afforded the carbamate 6f,
while demethylation using a-chloroethyl chloroformate
1
2
(
ACE-Cl) provided the secondary amine 6g. Addi-
tional analogues were generated from the ester 6a via N-
alkylation (6h), ester reduction (6i), and olefin hydro-
genation (6j). The last analogue, 6k, is the enantiomer
Table 1. Sigma-1 receptor binding affinity
Compd
K
i
(nM)
Compd
i
K (nM)
1
5
348 (n=5)
14 (n=6)
37 (n=8)
245 (n=5)
49 (n=4)
49 (n=5)
241 (n=4)
6f
6g
6h
6i
6j
6k
>3300 (n=4)
456 (n=3)
>3300 (n=4)
161 (n=4)
373 (n=3)
6a
6b
6c
6d
6e
115 (n=3)
Scheme 4.

M. P. Arrington et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1807–1809
1809
6
i shows that a small, polar N-substituent (HOCH CH –)
2
5. Antitussive activity of sigma-1 receptor agonists in the
Guinea pig: Brown, C.; Fezoui, M.; Selig, W. M.;
Schwartz, C. E.; Ellis, J. L. Br. J. Pharmacol. 2004, 141,
2
is tolerated but with some loss of affinity (compare to
c,d).
6
2
. Dextrorphan was produced via O-demethylation of dex-
33.
6
Saturation of the olefin in 6a leads to 6j, with a corres-
ponding 10-fold loss of binding affinity, indicating that
conformation and perhaps rigidity are key elements for
recognition of this scaffold by the receptor. The inter-
action is also stereoselective as the enantiomer of 5 (6k)
is a less potent ligand, albeit only by a factor of eight.
ꢀ
tromethorphan hydrobromide: excess BBr , CH Cl , 0 C
3
2
2
to rt overnight, >90% yield.
7
. (a) Similar fragmentations have been described: Anasta-
sia, L.; Cighetti, G.; Allevi, P. J. Chem. Soc., Perkin
Trans. 1 2001, 2398. (b) Pevarello, P.; Traquandi, G.;
Bonsignori, A.; McArthur, R. A.; Maj, R.; Caccia, C.;
Salvati, P.; Varasi, M. Bioorg. Med. Chem. Lett. 1999, 9,
1783.
In summary, we have reported the preparation of a
novel series of potent sigma-1 receptor ligands and have
established an initial SAR for this series. Further ana-
logues as well as functional characterization (agonism
versus antagonism) should provide additional insights
8
9
. Structurally related octahydrobenzoquinolines have been
examined as sigma-1 ligands: Wikstrom, H.; Andersson,
B.; Elebring, T.; Svensson, K.; Carlsson, A.; Largent, B.
J. Med. Chem. 1987, 30, 2169.
. Sigma-1 receptor binding assay (guinea pig membrane
preparation) performed as described by: Bowen, W. D.;
de Costa, B. R.; Hellewell, S. B.; Walker, M.; Rice, K. C.
Mol. Neuropharm. 1993, 3, 117.
1
3b,14
into the sigma-1 binding site.
References and notes
1
0. A mixture of quaternary ammonium salts could be
1
observed by H NMR after the alkylation step, as well as
from many of the previous reactions of 1 or 2 with t-butyl
bromoacetate.
1
2
. (a) Grant, K. A.; Colombo, G.; Grant, J.; Rogawski,
M. A. Neuropharmacology 1996, 238, 1709. (b) Wong,
B. Y.; Coulter, D. A.; Choi, D. W.; Prince, D. A. Neu-
rosci. Lett. 1988, 85, 261.
. (a) Deutsch, S. I.; Weizman, A.; Goldman, M. E.; Mor-
ihisa, J. M. Clin. Neuropharmacology 1988, 11, 105. (b)
Walker, M.; Bowen, W. D.; Walker, F. O.; Matsumoto,
R. R.; De Costa, B.; Rice, K. C. Pharmacol. Rev. 1990,
11. (a) Thompson, P. E.; Steer, D. L.; Aguilar, M.-I.; Hearn,
M. T. Bioorg. Med. Chem. Lett. 1998, 8, 2699. (b) Xu, R.;
Chu, G.; Bai, D. Tetrahedron Lett. 1996, 37, 1463.
12. Gubert, S.; Braojos, C.; Sacristan, A.; Ortiz, J. A. Synth-
esis 1991, 318.
13. (a) The K for dextromethorphan is similar to that repor-
i
42, 355.
ted (K =245 nM) in ref 2b. (b) The SAR for the com-
i
3
. (a) Guinea pig model: Kotzer, C. J.; Hay, D. W. P.;
Dondio, G.; Giardina, G.; Petrillo, P.; Underwood, D. C.
J. Pharmacol. Exp. Ther. 2000, 292, 803. (b) Rat and
mouse model: Kamei, J.; Iwamoto, Y.; Misawa, M.;
Kasuya, Y. Eur. J. Pharmacol. 1993, 242, 209.
pounds described in the present paper follows trends
similar to those described in the same review article,
which provides SAR summaries for a variety of sigma
receptor ligands.
14. (a) Binding & pharmacophore models: Ablordeppy, S. Y.;
Fischer, J. B.; Law, H.; Glennon, R. A. Bioorg. Med.
Chem. 2002, 10, 2759. (b) Ablordeppy, S. Y.; Fischer,
J. B.; Glennon, R. A. Bioorg. Med. Chem. 2000, 8, 2105.
(c) Ablordeppy, S. Y.; El-Ashmawy, M. B.; Glennon,
R. A. Med. Chem. Res. 1991, 1, 245.
4
. Side effects of codeine include sedation, constipation,
respiratory depression, and abuse liability, see: Way,
W. L.; Way, E. L.; Fields, H. L. In Basic & Clinical
Pharmacology; Katzung, B. G., Ed.; Appleton & Lange:
Norwalk, CT, 1995; p 460.