Enantioselective total synthesis of (-)-galanthamine [19]

Full text of DOI:10.1021/ja002231b

11262
J. Am. Chem. Soc. 2000, 122, 11262-11263
Scheme 1. Asymmetric Synthesis of Tricyclic Core
Enantioselective Total Synthesis of (-)-Galanthamine
Barry M. Trost* and F. Dean Toste
Department of Chemistry
Stanford UniVersity
Stanford, California 94305
ReceiVed June 22, 2000
(-)-Galanthamine (1),1 the parent member of the galanthamine-
type Amaryllidaceae alkaloids, has recently received significant
attention as a selective acetylcholinesterase inhibitor and, con-
sequently, for its potential clinical application for the treatment
of Alzheimer’s disease.2 An efficient synthesis of 1 is needed
since its extraction from daffodils is low-yielding (0.1-2% dry
weight)3 and it has been reported that native sources are
threatened.4 To date, all syntheses of galanthamine have utilized
a biomimetic oxidative bisphenol coupling5 to create the critical
spiro quaternary carbon6 of narwedine 2, which is converted into
1 by diastereoselective reduction. The only reported asymmetric
synthesis of a galanthamine alkaloid relies on the oxidative
coupling of L-tyrosine to prepare unnatural (+)-galanthamine.7,8
Herein, we report a versatile approach for the enantioselective
synthesis of the galanthamine alkaloids.
(a) 3% 6, 1% [η3-C3H5PdCl]2, (C2H5)3N, CH2Cl2, rt. (b) PhCH3, -78°.
(c) 2,6-Lutidine, CH2Cl2, rt. (d) Proton sponge, DMA, 80°.
ortho-halophenol and then employ an intramolecular Heck
reaction10 to prepare the crucial quaternary center. To this end,
palladium-catalyzed reaction of 2-bromovanillin (4)11 with car-
bonate 512 (available in two steps from glutaraldehyde and the
Emmons-Wadsworth-Horner reagent) in the presence of ligand
613 furnished the required aryl ether (7) in 72% yield and with
88% enantiomeric excess on a 24 mmol scale (Scheme 1).
Gratifyingly, good enantioselectivity could be obtained despite
the fact that phenol 4 is ortho-disubstituted and that cyclohexene
5 bears a C2-substitutent. Surprisingly, the aryl ethers obtained
from the palladium-catalyzed AAA of 5 are of the opposite
absolute stereochemistry to those obtained in the AAA reactions
of unsubstituted cyclohexenyl carbonates.14 Investigation of the
reasons for this reversal are ongoing and will be reported in due
course.
All attempts to effect the intramolecular Heck reaction of aryl
ether 7 failed, resulting primarily in ionization of phenol 4. An
earlier report15 suggested that the presence of electron-withdrawing
subsitutents on the phenol favored the palladium-catalyzed
ionization over the intramolecular Heck reaction. Therefore, 7
was reduced with DIBAL-H and the resulting diol protected with
TBDMS-triflate to afford bis-TBDMS ether 8 in 89% yield from
7. Heck reaction of 8 did not prove straightforward. The conditions
developed by Overman10,16 for formation of quaternary centers
produced only poor yields of 9 as did the Jeffery-type conditions
reported by Larock.15 Under these conditions, ionization of the
phenol persisted as the major pathway. Utilizing tri(tolyl)-
phosphine as ligand (20% Pd(OAc)2, 40% (otol)3P or 20% of the
preformed palladacycle17) suppressed the ionization reaction;
Our strategy was to first form the O5-C4a bond by a
palladium-catalyzed asymmetric allylic alkylation (AAA)9 of an
(1) For reviews see: (a) Hoshino, O. In The Alkaloids; Cordell, G. A.,
Ed.; Academic Press: New York, 1998; Vol. 51, pp 323-424. (b) Martin, S.
F. In The Alkaloids; Brossi, A., Ed.; Academic Press: New York, 1987; Vol.
30, pp 251-376.
(2) (a) Rainer, M. Drugs Today 1997, 33, 273. (b) Mucke, H. A. M. Drugs
Today 1997, 33, 251. (c) Gaicobini, E. Neurochem. Int. 1998, 32, 413. (d)
Weinstock, M. CNS Drugs 1999, 12, 307. (e) Unni, K. CNS Drugs 1998, 10,
447. (f) Nordberg, A.; Svensson, A. L. Drug Safety 1998, 19, 45.
(3) Shieh, W.-C.; Carlson, J. A. J. Org. Chem. 1994, 59, 5463.
(4) Poulev, A.; Deus-Nuemann, B.; Zenk, M. H. Planta Med. 1993, 59,
442.
(5) (a) Barton, D. H. R.; Kirby, G. W. J. Chem. Soc. 1962, 806. (b)
Kametani, T.; Yamaki, K.; Terui, T. J. Heterocycl. Chem. 1973, 10, 35. (c)
Vlahov, R.; D. Krikorian, D.; Spassov, G.; Chinoua, M.; Ulahov, F.; Parushev,
S.; Snatzke, G.; Ernst, L.; Klieslich, K.; Abraham, W.-R.; Sheldrick, W. S.
Tetrahedron 1988, 45, 3329. (d) Szewczyk, J.; Wilson, J. W.; Lewin, A. H.;
Carroll, F. I. J. Heterocycl. Chem. 1995, 32, 195. (e) Chaplin, D. A.; Fraser,
N.; Tiffin, P. D. Tetrahedron Lett. 1997, 38, 7931. (f) Czollner, L.; Frantsits,
W.; Ku¨enberg, B.; Hedenig, U.; Fro¨hlich, J.; Jordis, U. Tetrahedron Lett. 1998,
39, 2087. (g) Kita, Y.; Arisawa, M.; Gyoten, M.; Nakajima, M.; Hamada, R.;
Tohma, H.; Takada, T. J. Org. Chem. 1998, 63, 6625. (h) Ku¨enburg, B.;
Czollner, L.; Fro¨hlich, J.; Jordis, U. Org. Process Res. ReV. 1999, 3, 425.
(6) A variety of methods have been used to prepare this center in syntheses
of (()-lycoramine 3: (a) Gras. E.; Guillou, C.; Thal., C. Tetrahedron Lett.
1999, 40, 9243. (b) Ishizaki, M.; Ozaki, K.; Kanematsu, A.; Isoda, T.; Hoshino,
O. J. Org. Chem. 1993, 58, 3877. (c) Parker, K. A.; Kim, H.-J. J. Org. Chem.
1992, 57, 752. (d) Holton, R. A.; Sibi, M. P.; Murphy, W. S. J. Am. Chem.
Soc. 1988, 110, 314. (e) Ackland, D. J.; Pinhey, J. T. J. Chem. Soc., Perkin
Trans. 1 1987, 2695. (f) Sa˜nchez, I. H.; Soria, J. J.; Lo´pez, F. J.; Larraza, M.
I.; Flores, H. J. J. Org. Chem. 1984, 49, 157 (g) Martin, S. F.; Garrison, P. J.
J. Org. Chem. 1982, 47, 1512. (h) Schultz, A. G.; Yee, Y. K.; Berger, M. H.
J. Am. Chem. Soc. 1977, 99, 8065.
(9) (a) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1998, 120, 815. (b)
Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1998, 120, 9074. (c) Trost, B.
M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 3543. (d) Trost, B. M.; Toste,
F. D. J. Am. Chem. Soc. 1999, 121, 4545. (e) Trost, B. M.; Tsui, H.-C.; Toste,
F. D. J. Am. Chem. Soc. 2000, 122, 3534.
(10) For recent examples of intramolecular Heck reaction of allyl aryl ethers
to form quaternary centers, see: (a) Frey, D. A.; Duan, C.; Hudlicky, T. Org.
Lett. 1999, 1, 2085. (b) Liou, J.-P.; Cheng, C. Y. Tetrahedron Lett. 2000, 41,
915. For a review of the intramolecular Heck reaction, see: (c) Gibson, S.
E.; Middleton, R. J.Contemp. Org. Synth. 1996, 3, 447.
(11) Toth, J. E.; Hamann, P. R.; Fuchs, P. L. J. Org. Chem. 1988, 53,
4694.
(12) (a) Villie´ras, J.; Rambaud, M.; Graff, M. Synth. Commun. 1986, 16,
149. (b) Amri, H.; Rambaud, M.; Villie´ras, J. Tetrahedron 1990, 49, 3535.
(13) Trost, B. M.; van Vranken, D. L.; Bingel, C. J. Am. Chem. Soc. 1992,
114, 9327.
(7) (a) Shimizu, K.; Tomioka, K.; Yamada, S.; Koga, K. Chem. Pharm.
Bull. 1978, 26, 3765. (b) Shimizu, K.; Tomioka, K.; Yamada, S.; Koga, K.
Heterocycles 1977, 8, 277.
(8) For resolution of narwedine, see: (a) Kametani, T.; Premila, M. S.;
Fukumoto, K. Heterocycles 1976, 4, 1111. (b) Chaplin, D. A.; Johnson, N.
B.; Paul, J. M.; Potter, G. A. Tetrahedron Lett. 1998, 39, 6777.
(14) The assignment of the absolute configuration derives by analogy to
the reaction of 5 with p-methoxyanisole for which the product was correlated
to a known compound and by completion of the synthesis to (-)-galanthamine.
(15) Larock, R. C.; Stinn, D. E. Tetrahedron Lett. 1988, 29, 4687.
(16) Hong, C. Y.; Kado, N.; Overman, L. E. J. Am. Chem. Soc. 1993, 115,
11028.
10.1021/ja002231b CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/25/2000

Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 45, 2000 11263
Scheme 3. Adjustment of Oxidation Level
Scheme 2. Formation of Azepine Ring
(a) CH2Cl2, rt. (b) C2H5OH, 80°. (c) THF, H2O, rt then CHCl3. (d)
TsOH, PhH, 60°.
(a) THF, rt. (b) CH3COCH3, rt. (c) CH3OH, reflux then CH2Cl2,
(C2H5)3N, rt. (d) NaHCO3, CH2Cl2, rt. (e) NaN(TMS)2, THF, 0°. (f)
CH2Cl2, rt then CH3OH, 4 Å MS, 0°.
but immediately treated with sodium cyanoborohydride to afford
(-)-3-deoxygalanthamine (14) in 53% overall yield.
however, tetrahydrobenzofuran 9 was isolated as a mixture of
olefin isomers. On the other hand, use of 20% palladium acetate
with bidentate ligands (18% dppb, dppe, dppf, or BINAP)
prevented the olefin isomerization, but once again, phenol
ionization dominated.
Attempts to convert 14 to (-)-galanthamine (1) by direct allylic
oxidation (SeO2 or tert-butylperbenzoate, CuBr) were not suc-
cessful. Therefore, we resorted to a four-step protocol to introduce
the C3 hydroxyl group (see Scheme 3). Reaction of the tosyl-
ammonium salt of tertiary amine 14 with dimethyldioxirane20
afforded a mixture of epoxide 15 and the corresponding R-hy-
droxy tosylate. Treatment of the crude mixture with DBU
converted R-tosyl alcohol to epoxide 15 which was isolated in
45% from 14. An nOe between the hydrogen on C-1 and one of
the benzylic hydrogens (i.e., on C-9) established the stereochem-
istry as depicted in 15. Regioselective opening of epoxide 15 with
sodium phenylselenide produced R-hydroxy selenide 16 in 98%
yield. Chemoselective oxidation of selenide 16 was readily
achieved by treatment with sodium periodate to produce a 1:1
mixture of diastereomeric selenoxides. These selenoxides did not
eliminate at room temperature but required heating (70 °C) to
effect the desired conversion into (-)-isogalanthamine (17).
Reaction of 17 with Osborn’s rhenium(VII) catalyst21 produced
(-)-galanthamine ([R]D ) -112.8° (c ) 0.5, EtOH), lit.22 [R]D
) -131.4° (c ) 0.6, EtOH)) with spectral characteristics identical
to those of a sample of the natural product.
These results led us to the hypothesis that a sterically
demanding bidentate phosphine ligand would suppress both the
olefin isomerization reaction and the phenol ionization reaction.
We thus turned our attention to bidentate alkylphosphine ligands.18
1,2-bis(Dicylohexylphosphino)ethane (dcpe) proved to be an
excellent ligand for the intramolecular Heck reaction, affording
benzofuran 9 in 50% yield along with 19% of the monodepro-
tected product 10 without detectable amounts of ionization (total
yield of 69%). This selectivity may derive from a combination
of factors. First, alkylphosphines are more electron-rich than their
aryl counterparts, which has been shown to increase the rate of
oxidative addition of palladium(0) to aryl halides.19 Electron-rich
ligands may also decrease the rate of coordination of palladium(0)
to the trisubstituted olefin of 8 required for phenol ionization.
Second, the steric hindrance of the dicyclohexylphosphine ligands
is significantly greater than that of the corresponding diarylphos-
phine ligands. This factor may further decrease the rate of
coordination of a trisubstituted olefin. This combination of
electronic and steric factors present in the dicyclohexylphosphine
ligands is presumably responsible for the sharp increase in the
rate of the Heck reaction relative to the ionization.
The hydrobenzazepine ring of galanthamine was prepared in
a six-step sequence (see Scheme 2). The TBAF deprotection of
the mixture of 9 and 10 followed by chemoselective manganese
dioxide oxidation of the resulting diol afforded aldehyde 11.
Reductive amination of aldehyde 11 was accomplished by
formation of the imine with methylamine followed by reduction
with sodium cyanoborohydride. The resulting secondary amine
was protected without isolation to supply tert-butyl carbamate
12 in 83% yield. Dess-Martin oxidation of the amino alcohol
12 gave the amino aldehyde in 93% yield. The necessary one-
carbon homologation was achieved by a Wittig olefination of the
aldehyde with the ylide derived from methoxymethyltriphenyl-
phosphonium bromide and NaHMDS producing 13 in 64% yield
as a 10:1 mixture of olefin isomers. Concomitant deblocking of
the secondary amine and hydrolysis of the methoxyvinyl group
gave a seven-membered ring hemiaminal which was not isolated
The asymmetric synthesis disclosed herein is the first total
synthesis which does not utilize an oxidative phenol coupling to
construct the quaternary center of (-)-galanthamine (1). Notably,
aside from resolution, the sequential palladium-catalyzed AAA
and intramolecular Heck reaction provides the only entry into
the natural enantiomer of the (-)-galanthamine-type alkaloids.
This strategy should allow for entry into a variety of clinically
important natural and unnatural galanthamine-type Amaryllidaceae
alkaloids.
Acknowledgment. We thank Professor Madeleine Joullie´ for a
generous sample of (-)-galanthamine hydrobromide. We thank the NIH-
GM for their generous support of our programs. Mass spectra were
obtained from the Mass Spectrometry Regional Center of the University
of California-San Francisco supported by the NIH Division of Research
Resources. We thank ChiroTech and NSC Technologies for the generous
gifts of ligands.
Supporting Information Available: Experimental details (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
JA002231B
(17) Herrmann, W. A.; Brossmer, C.; Ofele, K.; Reisenger, C. P.;
Priermeier, T.; Beller, M.; Fischer, H. Angew. Chem., Int. Ed. Engl. 1995,
34, 1844.
(20) (a) Asensio, G.; Mello, R.; Boix-Bernardini, C.; Gonza´lez-Nu´n˜ez, M.
E.; Castellano, G. J. Org. Chem. 1995, 60, 3692. (b) Asensio, G.; Boix-
Bernardini, C.; Andreu, C.; Gonza´lez-Nu´n˜ez, M. E.; Mello, R.; Edwards, J.
O.; Carpenter, G. B. J. Org. Chem. 1995, 60, 4705.
(21) (a) Bellemin-Lapponnaz, S.; Le-Ny, J. P.; Osborn, J. A. Tetrahedron
Lett. 2000, 41, 1549. (b) Bellemin-Laponnaz, S.; Gisie, H.; Le-Ny, J. P.;
Osborn, J. A. Angew. Chem. 1997, 36, 976.
(22) Han, S. Y.; Sweeney, J. E.; Bachman, E. S.; Schweiger, E. J.; Forloni,
G.; Coyle, J. T.; Davis, B. M.; Joullie´, M. M. Eur. J. Med. Chem. 1992, 27,
673.
(18) Hartwig has shown 1,1′-(dicyclohexyphosphino) ferrocene (dcpf) to
be an excellent ligand for the palladium-catalyzed amination of aryl halides.
Hamann, B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1998, 120, 7369.43. On the
other hand, Fu found that utilizing dcpe did not lead to a effective catalyst
for the intermolecular Heck reaction of aryl chlorides. Littke, A. F.; Fu, G.
C. J. Org. Chem. 1999, 64, 10.
(19) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1665.