A general approach to the galanthamine ring system

Article abstract of DOI:10.1016/S0040-4039(01)00111-3

An approach to the galanthamine ring system is reported, which utilises an intramolecular Heck reaction to establish the benzofuran ring system present in both galanthamine and morphine.

Full text of DOI:10.1016/S0040-4039(01)00111-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2209–2211
A general approach to the galanthamine ring system
Philip J. Parsons,a,* Mark D. Charles,a Darren M. Harvey,a L. Ravi Sumoreeah,a Adrian Shell,a
Grant Spoors,a Adrian L. Gillb and Steve Smithc
aSussex University, CPES, Falmer, Brighton, East Sussex BN1 9QJ, UK
bRoche Discovery Welwyn, 40 Broadwater Road, Welwyn Garden City, Hertfordshire AL7 3AY, UK
cGlaxo Wellcome Research and Development, Medicines Research Centre, Gunnels Wood Road, Stevenage,
Hertfordshire SG1 2NY, UK
Received 13 October 2000; accepted 17 January 2001
Abstract—An approach to the galanthamine ring system is reported, which utilises an intramolecular Heck reaction to establish
the benzofuran ring system present in both galanthamine and morphine. © 2001 Elsevier Science Ltd. All rights reserved.
Galanthamine (1) is an alkaloid with a structure related
to morphine (2). It is isolated from the Cauasian snow-
drop, Galanthaus woronowii1 as well as the bulbs of
Amaryllidaceae. Galanthamine has been shown to
exhibit potent acetylcholinesterase inhibitory activity2
and as such is a primary candidate in the management
of Alzheimer’s disease. A recent publication by Fels et
al.,3 has prompted us to comment upon our research in
this area.
Our approach to both galanthamine and morphine
relies on an efficient route to alkenes 6a and 6b which
were synthesised using Heck methodology (Scheme 1).
We employed a similar synthetic scheme to Fels et al.,3
utilising a Mitsunobu4 coupling of phenol 3;5 however,
we found that using tributylphosphine and the morpho-
line derivative of DEAD made purification easier. In
our case, intramolecular Heck6 cyclisation proceeded in
lower yields with formation of the isomeric alkene (7)
from palladium(0) mediated isomerisation when potas-
sium carbonate is used as a base. The use of silver(I)
salts in Heck reactions prevents double bond isomerisa-
tion as described by Overman et al.7 Indeed, in our
case, changing the base to silver(I) carbonate gave
exclusively alkene 6 in 75% yield with no evidence of
isomeric alkene 7.
MeO
HO
NMe
O
O
NMe
H
HO
OH
(1)
(2)
MeO
O
MeO
R
OH
I
R
R
i
ii
+
CONMe2
O
I
CONMe2
OH
Me2NOC
OMe
(3)
(4)
(5)
(6)
(3a) R = CHO, (3b) R = -CH2CH2NO2
(5a) R = CHO, (5b) R = -CH2CH2NO2
(6a) R = CHO, (6b) R = -CH2CH2NO2
Scheme 1. Reagents: (i) azodicarboxylic dimorpholide, Bu3P, THF, 65%. (ii) Pd(OAc)2, Ag2CO3, ethylenebis(diphenylphosphine),
DMF, reflux 2 hours, 75%.
Keywords: galanthamine; morphine; Mitsunobu; intramolecular Heck; allylic oxidation.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00111-3

2210
P. J. Parsons et al. / Tetrahedron Letters 42 (2001) 2209–2211
amine was heated under high vacuum to produce the
MeO
O
galanthamine ring system (14) in 92% yield (Scheme
3).11 This is the same intermediate that Fels and co-
workers reported and are intending to manipulate at
the allylic position. However, we wish to report our
findings in this area, as we have found allylic oxidation
of 14 to be problematic as shown in Table 1.
R
CONMe2
(7)
In addition we have also attempted to introduce the
desired functionality via various other methods. These
include using NBS to produce an allylic bromide which
could be manipulated using the procedure of Lalle-
mand et al.12 However, only a complex mixture was
obtained. The use of molecular bromine as an alterna-
tive resulted in bromination of the aromatic ring. Epox-
idation of 14 and base mediated ring opening was also
attempted. Although the epoxide could be produced
from 1,1,1-trifluoromethylmethyldioxirane,13 ring open-
ing could not be effected with a variety of bases.
Our allylic alcohol 4 was synthesised by a reliable and
highly efficient route utilising Birch8 reduction chem-
istry (Scheme 2). Birch reduction of methyl 2-methoxy-
benzoate under standard conditions gave the
cyclohexadiene (8) in 83% yield. Deprotonation with
LDA, and subsequent quenching with the bromoac-
etamide (9),9,10 cleanly produced 10 in almost quantita-
tive yield. Saponification of the ester (10) and
hydrolysis with concentrated hydrochloric acid gener-
ated the b-keto acid, which spontaneously decarboxyl-
ated to furnish isomeric cyclohexenes 11 and 12. These
were converted into the 2-ene isomer (11) by heating in
THF and 1 M HCl in an overall yield of 92%. The
enone (11) was reduced with sodium borohydride,
interestingly with no 1,4 reduction occurring, to yield
allylic alcohol 4.
We are currently investigating a synthetic route to
morphine utilising intermediate 6b in a [3+2] nitrile
oxide cycloaddition. The nitroalkane (6b) can be
smoothly converted with phenylisocyanate in DCM
and catalytic triethylamine to the isoxazoline (17) in
80% yield.
In summary we have developed a route to the galan-
thamine ring system by a conceptually similar route to
that of Fels et al., and we have shown that manipula-
tion of the allylic position is a far from trivial matter.
We are also continuing our research efforts toward
In order to obtain the galanthamine ring system (14)
the aldehyde (6a) was condensed with methylamine,
and subsequent sodium borohydride reduction afforded
the secondary amine (13). The hydrochloride salt of this
CO2Me
OMe
CO2Me
OMe
CO2Me
NMe2
Me2NOC
OMe
Br
ii
i
+
O
(8)
(9)
(10)
O
O
OH
v
iii
CONMe2
CONMe2
(11)
(4)
iv
CONMe2
(12)
Scheme 2. Reagents: (i) K, liq. NH3, THF, t-BuOH; NH4Cl, 83%. (ii) LDA, THF, 9, 95%. (iii) NaOH, H2O; conc. HCl. (iv) 1
M HCl, THF, 92%. (v) NaBH4, MeOH, 75%.
MeO
O
MeO
ii
i
NHMe
N
6a
O
CONMe2
O
(13)
(14)
Scheme 3. Reagents: (i) MeNH2, EtOH, rt; NaBH4, MeOH, 90%. (ii) HCl, MeOH; vacuum, 120°C, 92%.

P. J. Parsons et al. / Tetrahedron Letters 42 (2001) 2209–2211
Table 1. Results from attempted allylic oxidation of 14
2211
Reagents and conditions
Starting material (%)
Cleaved lactam 15 (%)
Imide 16 (%)
Complex mixture
SeO2, EtOH, heat
100
100
60
–
–
–
–
–
90
–
–
–
–
–
–
–
–
–
70
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ã
–
SeO2, tert-BuOH, 150°C
SeO2, HCO2H, dioxane
Cr(CO)6, TBHP, MeCN
MnO2, DCM, rt
Pd(OAc)2, MnO2, AcOH, heat
CuI, TBHP, benzene, heat
CuBr, TBHP, benzene, heat
Pb(OAc)4, AcOH, rt
CrO3, t-BuOH, rt
CrO3, DMP, −20°C
PDC, TBHP, benzene, rt
[Rh3O(OAc)6CH2O]OAc, TBHP, AcOH,
90
20
70
80
70
–
60
40
–
–
50
–
–
100
–
–
MeO
MeO
O
MeO
O
O
NMe
H
CONMe2
O
O
N
N
H
O
O
O
(15)
(16)
(17)
galanthamine, by utilising an alternative procedure for
introduction of the oxygen functionality. In addition
research toward morphine is continuing and will be
reported at a later date.
3. Pilger, C.; Westermann, B.; Flo¨rke, U.; Fels, G. Synlett
2000, 1163.
4. Mitsunobu, O. Synthesis 1981, 1.
5. Markovich, K. M.; Tantishaiyakul, U.; Hamada, A.;
Miller, D. D.; Romstedt, K. J.; Shams, G.; Shin, Y.;
Fraundorfer, P. F.; Doyle, K.; Feller, D. R. J. Med.
Chem. 1992, 35, 466.
6. Heck, R. F. J. Am. Chem. Soc. 1968, 90, 5518.
7. Abelman, M. M.; Oh, T.; Overman, L. E. J. Org. Chem.
1987, 52, 4130.
Acknowledgements
We thank Dr. Bonnie Davis for helping to support this
work and to Professor Madeleine Joullie´ for her contin-
ued interest. We also thank Roche Discovery Welwyn
and Glaxo Wellcome for financial support.
8. Birch, A. J. J. Chem. Soc. 1950, 1551.
9. Hook, J. M.; Mander, L. N.; Woolias, M. Tetrahedron
Lett. 1982, 23, 1095.
10. Bartlett, P. A.; Carruthers, N. I.; Winter, B. M.; Long, K.
P. J. Org. Chem. 1982, 47, 1284.
References
11. Harvey, D. M. PhD Thesis, University of Sussex, 1999.
12. Bouchard, H.; Renard, P. Y.; Lallemand, J. Y. Tetra-
hedron Lett. 1991, 32, 5953.
13. Yang, D.; Wong, M. K.; Yip, C. J. Org. Chem. 1995, 60,
3887.
1. Proskurnina, N. F.; Yakovleva, A. P. J. Gen. Chem.
1952, 22, 1899.
2. Thomsen, T.; Kewitz, H. Life Sci. 1990, 46, 1553.
.