Stereocontrolled Total Synthesis of Alkaloid G via the Oxy-anion Cope Rearrangement and Improved Total Synthesis of (+)-Ajmaline

Article abstract of DOI:10.1021/ol000331g

(Matrix Presented) The oxy-anion Cope rearrangement followed by protonation of the enolate which resulted under conditions of kinetic control has been employed to generate the key asymmetric centers at C(15), C(16), and C(20) in alkaloid G (1) and (+)-ajm

Full text of DOI:10.1021/ol000331g

ORGANIC
LETTERS
2001
Vol. 3, No. 3
345-348
Stereocontrolled Total Synthesis of
Alkaloid G via the Oxy-anion Cope
Rearrangement and Improved Total
Synthesis of (+)-Ajmaline
Tao Wang, Qingge Xu, Peng Yu, Xiaoxiang Liu, and James M. Cook*
Department of Chemistry, UniVersity of WisconsinsMilwaukee,
Milwaukee, Wisconsin 53201
capncook@csd.uwm.edu
Received November 1, 2000
ABSTRACT
The oxy-anion Cope rearrangement followed by protonation of the enolate which resulted under conditions of kinetic control has been employed
to generate the key asymmetric centers at C(15), C(16), and C(20) in alkaloid G (1) and (+)-ajmaline (2) in a highly stereocontrolled fashion.
The aldehyde 7b from this process has been converted into alkaloid G (1) and (+)-ajmaline (2) in 36% and 13% overall yields (11 reaction
vessels from 3), respectively.
Alkaloid G (1) was isolated from plant cell cultures of
Rauwolfia serpentina Benth by Sto¨ckigt et al.1 after feeding
experiments with (+)-ajmaline (2); the latter alkaloid was
isolated from the roots of R. serpentina in 1931.2 Alkaloid
G (1) is of interest because of the novel C(6)-oxygen, C(17)-
hemiacetal bond illustrated in Figure 1, while (+)-ajmaline
(1) Endreb, S.; Takayama, H.; Suda, S.; Kitajima, M.; Aimi, N.; Sakai,
S.-I.; Sto¨ckigt, J. Phytochemistry 1993, 32, 725.
(2) Siddiqui, S.; Siddiqui, R. H. J. Ind. Chem. Soc. 1931, 8, 667.
(3) Brugada, J.; Brugada, P. Am. J. Cardiol 1996, 78(5A), 69.
(4) Slowinski, S.; Rajch, D.; Zabowka, M. Przegl. Lek. 1996, 53, 196.
(5) Chen, X.; Borggrefe, M.; Martinez, R. A.; Hief, C.; Haverkamp, W.;
Hindricks, G.; Breithardt, G. J. CardioVasc. Pharmacol 1994, 24, 664.
(6) Mozo, d. R. F.; Moreno, J.; Bodegas, A.; Melchor, J.; Fernandez, L.
L.; Aranguren, G. Eur. J. Obstet. Gynecol. Reprod. Biol. 1994, 56, 63.
(7) Creasey, W. A. The Monoterpenoid Indole Alkaloids. In Heterocyclic
Compounds; Indole Series; Saxton, J. E., Ed.; John Wiley and Sons: New
York, 1983; Vol. 25, p 783.
(8) Best, H. J.; Winkler, J.; Foerster, W. Acta Biol. Med. Ger. 1977, 36,
1193.
Figure 1.
(9) Koskinen, A.; Lounasmaa, M. The Sarpagine-Ajmaline Group of
Indole Alkaloids. In Progress in the Chemistry of Organic Natural Products;
Herz, W., Grisebach, H., Kirby, G. W., Ed.; Springer-Verlag: New York,
1983; Vol. 43, p 267.
(10) Bi, Y.; Hamaker, L. K.; Cook, J. M. The Synthesis of Macroline
Related Alkaloids. In Studies in Natural Products Chemistry, BioactiVe
Natural Products, Part A; Basha, F. Z., Rahman, A., Ed.; Elsevier
Science: Amsterdam, 1993; Vol. 13, p 383.
(2) contains four heteroatoms, six rings, and nine asymmetric
centers. It is a clinically important indole alkaloid3-8 with
historical significance2,9 and is related to the sarpagine
bases.10,11 “The most prominent action of ajmaline is an
10.1021/ol000331g CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/09/2001

antiarrhythmic effect on the heart” as was reviewed by
Creasey, “that is less pronounced than that of propranolol,7
but is superior in terms of the ratio of the refractory phase
over reduced conduction to that of procaine amide and
quinidine”.7,12 Both alkaloid G (1) and (+)-ajmaline (2) are
structurally related by the presence of the quinuclidine ring
and the C(5)-C(16) bond linkage as well as the identical
absolute configurations of the stereogenic centers at C(3),
C(5), C(15), C(16), C(20), and C(21) as shown in Figure 1.
Three important reports on the synthesis of 2 have
appeared previously,13-16 and an enantiospecific total syn-
thesis of (+)-2 as well as alkaloid G (1) has recently been
reported.17,18 The chiral centers of C(15), C(16), and C(20)
were established selectively in earlier work17,18 based on a
Barbier-Grignard process17-19 (Mg metal) with a seven-
carbon-atom pseudosymmetric carbanion, followed by an
oxy-anion Cope rearrangement. This approach resulted in
the formation of several diastereomers at C(20), the majority
of which could be employed in the synthesis of (+)-2.17,18
Recently, this seven-carbon fragment has been replaced with
a five-carbon unit (trans-1-bromo-2-pentene) employing the
barium chemistry of Yamamoto.20,21 This provided a ho-
moallylic alcohol which underwent the oxy-anion Cope
rearrangement with high diastereoselectivity. Kinetic proto-
nation of the enolate, which resulted, gave the desired
intermediate with the asymmetric centers at C(15), C(16),
and C(20) in high diastereoselectivity (43:1). This improve-
ment provides the first facile entry into the desired absolute
configuration of the ethyl group at C(20) of the ajmaline-
related alkaloids and resulted in an enantiospecific total
synthesis of alkaloid G (1) and an improved synthesis of
(+)-2, the details of which follow below.
Scheme 1a
a (a) ClCH2SOPh, LDA/THF, -78 °C, KOH (aq), rt; LiClO4/
dioxane, reflux, 24 h, on 50 g scale, 87% overall yield; (b) Li/
biphenyl/Bal2/THF, -78 °C, 4 and trans-1-bromo-2-pentene, 90%;
(c) KH/dioxane/18-crown-6, 100 °C, 14 h, 85%; (d) NaOMe/
MeOH, 95%; (e) KH/dioxane/18-crown-6, 100 °C, 14 h; CH3OH,
0 °C f rt, 4 h, 85%.
tion of the chemistry of Yamamoto.20,21 When trans-1-bromo-
2-pentene was stirred with barium metal under normal
reaction conditions, none of the desired olefinic alcohol 5
was observed. However, when aldehyde 4 and trans-1-
bromo-2-pentene were premixed and added to a solution of
preformed barium metal at -78 °C, analogous to a Barbier-
Grignard process, a 90% yield of the desired homoallylic
alcohol 5 was obtained. Allylic rearrangement of the barium-
stabilized carbanion did not occur at -78 °C.
The starting (-)-Na-H, Nb-benzyl tetracyclic ketone 3 has
been synthesized via a two-pot process on multihundred gram
scale in our laboratory.17,18,22-25 Conversion of the carbonyl
function of (-)-3 into the R,â-unsaturated aldehyde moiety
of 4 was achieved through the spirooxiranophenylsul-
foxide26,27 in 87% yield, as shown in Scheme 1. A key
improvement in the synthesis of 2 came on conversion of
aldehyde 4 into the allylic alcohol 5 employing a modifica-
When the homoallylic alcohol 5 was heated to 100 °C
under the conditions of the oxy-anion Cope rearrangement
(KH, 18-crown-6, dioxane), the process took place from the
R-face of the olefinic system to furnish the desired stereo-
chemistry at C(15) and C(20) with high diastereoselectivity
(>30:1). Only a trace of the epimeric diastereomer at C(15)
was ever observed; moreover, the correct chirality of the ethyl
function as 20(S) required for 1 and 2 had been established.
The two epimers 7a and 7b at C(16), originally isolated as
a 4:1 mixture, could be converted entirely into the sarpagine
stereochemistry at C(16) on treatment with base.24
From examination of MM228 calculations and epimeriza-
tion experiments, it was clear that epimer 7b with the
ajmaline aldehydic stereochemistry at C(16) was the ther-
modynamically less stable isomer. In earlier work, epimer-
ization of the aldehydic group at C(16) from the R to the S
(ajmaline) configuration was reported as 43:7 to 7:3,13,14,16
(11) Hamaker, L. K.; Cook, J. M. The Synthesis of Macroline Related
Sarpagine Alkaloids. In Alkaloids: Chemical and Biological PerspectiVes;
Pelletier, S. W., Ed.; Elsevier Science: New York, 1995; Vol. 9, p 23.
(12) Benthe, H. F. Naunyn-Schmiedebergs Arch. Exptl. Pathol. Phar-
makol. 1956, 229, 82.
(13) Masamune, S.; Ang, S. K.; Egli, C.; Nakatsuka, N.; Sarkar, S. K.;
Yasunari, Y J. Am. Chem. Soc. 1967, 89, 2506.
(14) Mashimo, K.; Sato, Y. Tetrahedron Lett. 1969, 901.
(15) Van Tamelen, E. E.; Oliver, L. K. J. Am. Chem. Soc. 1970, 92,
2136.
(16) Van Tamelen, E. E.; Oliver, L. K. Bioorg. Chem. 1976, 5, 309.
(17) Li, J.; Cook, J. M. J. Org. Chem. 1998, 63, 4166.
(18) Li, J.; Wang, T.; Yu, P.; Peterson, A.; Weber, R.; Soerens, D.;
Grubisha, D.; Bennett, D.; Cook, J. M. J. Am. Chem. Soc. 1999, 121, 6998.
(19) Fu, X.; Cook, J. M. J. Am. Chem. Soc. 1992, 114, 6910.
(20) Yanagisawa, A.; Habaue, S.; Yamamoto, H. J. Am. Chem. Soc. 1991,
113, 8955.
(21) Yanagisawa, A.; Yamamoto, H. ActiVe Metals; Fu¨rstner, A., Ed.;
VCH Verlagsgesellschaft: Weinheim, 1996; p 61.
(22) Yu, P.; Cook, J. M. J. Org. Chem. 1998, 63, 9160.
(23) Wang, T.; Yu, P.; Li, J.; Cook, J. M. Tetrahedron Lett. 1998, 8009.
(24) Yu, P.; Wang, T.; Li, J.; Cook, J. M. J. Org. Chem. 2000, 65, 3173.
(25) Wang, T.; Cook, J. M. Org. Lett. 2000, 2, 2057.
(26) Taber, D. F.; Gunn, B. P. J. Org. Chem. 1979, 44, 450.
(27) Reutrakul, V.; Kanghae, W. Tetrahedron Lett. 1977, 16, 1377.
(28) Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127.
Org. Lett., Vol. 3, No. 3, 2001
346

with the desired 16(S) material as the minor component. A
study of the structure of the enolate (see 6, Figure 2) indicated
was protected as the acetal with ethylene glycol/p-TSA with
no epimerization at C(16). This provided a stable stereocenter
at C(16) which permitted the conversion (NaH, CH3I) of 8
into the Na-methyl derivative 9 in 94% yield. This improve-
ment resulted in the replacement of Na-methyl-D-tryptophan
with the cheaper D-(+)-tryptophan employed in the asym-
metric Pictet-Spengler reaction.29 Oxidative removal of the
methylene group of the latent aldehyde in 9 with OsO4/NaIO4
furnished aldehyde 10 in 91% yield (see Scheme 2 for
Scheme 2a
Figure 2.
that protonation from the less hindered (bottom) face (kinetic
protonation) might provide the desired 16(S) stereochemistry
directly in a one-pot fashion. Consequently, a number of
experiments were carried out to study the effect of kinetic
protonation in this system (see 6). After many attempts, only
a few of which are listed in Table 1, this kinetic protonation
Table 1. Kinetic Protonation versus Thermodynamic
Protonation
temp,
°C
ratio
(7a :7b)a
entry
solvent
dioxane
dioxane
dioxane
dioxane
dioxane/THF
reagent for quench
1
2
3
4
5
6
0
0
0
0
CH3OH; rt, 4 h
CH3OH
2,6-diisopropylphenol
1 N HCl (aq)
100:0
4:1
3:2
5:6
1:27
1:43
a (a) ethylene glycol/p-TSA/benzene, ∆, 20 h, 92%; (b) NaH/
THF, CH3I, 94%; (c) OsO4/py/THF, 0 °C, 16 h; NaHSO3(aq), rt, 4
h; NalO4/CH3OH/H2O, 0 °C, 16 h, 91%; (d) Pd/C/H2, DME, 2 d;
Ac2O/DMAP, 2 h, 91%; (e) p-TSA/acetone, rt, 12 h, 89%; (f) DDQ/
THF/H2O, rt, 94%; (g) CH3OH, K2CO3 (5% aq), 0 °C, 2 h, 92%.
-78 1 N TFA (THF solution)
dioxane/THF -100 1 N TFA (THF solution)
1
a The ratio was determined by integration of the H NMR spectrum of
the crude product.
details). This was followed by catalytic removal of the benzyl
function (Pd/C, H2) and acetylation of the carbinolamine
which resulted in a one-pot process to provide acetate 11.
The acetal moiety of 11 was removed under standard
conditions to release the aldehyde 12. This aldehyde was
oxidized at C(16) employing the conditions of Yonemitsu30,31
later modified in our laboratory (DDQ, aqueous THF)32,33
was realized. When the enolate 6 was quenched under acidic
conditions at lower temperature, the ratios of the desired 7b
(16S) to 7a began to increase [compare entries 4 and 5, Table
1, to entry 1 (basic conditions)]. When the enolate was
quenched by pouring it (6, THF/dioxane, -100 °C) into a
(1 N) TFA/THF solution at -100 °C, the ratio of 7b to 7a
improved dramatically to greater than 43:1. This one-pot
conversion of 5 into 7b can now be accomplished in high
yield. This, for the first time, provided a highly stereocon-
trolled route to the key chiral centers at C(15), C(16), and
C(20) of the ajmaline series.
(29) Cox, E. D.; Hamaker, L. K.; Li, J.; Yu, P.; Czerwinski, K. M.; Deng,
L.; Bennett, D. W.; Cook, J. M. J. Org. Chem. 1997, 62, 44.
(30) Oikawa, Y.; Yonemitsu, O. J. Org. Chem. 1977, 42, 1213.
(31) Oikawa, Y.; Yoshioka, T.; Kunihiko, M.; Yonemitsu, O. Hetero-
cycles 1979, 12, 1457.
(32) Campus, O.; Diperro, M.; Cain, M.; Mantei, R.; Gawish, A.; Cook,
J. M. Heterocycles 1980, 14, 975.
The enantiospecific total synthesis of alkaloid G (1) (from
7b) was completed as follows. The aldehydic group in 7b
(33) Cain, M.; Mantei, R.; Cook, J. M. J. Org. Chem. 1982, 47, 4933.
Org. Lett., Vol. 3, No. 3, 2001
347

to provide the acetal 13, stereospecifically (94%). A number
of mechanisms for the generation of 13 are possible,30-37
only one of which is depicted here (Scheme 3).
Scheme 4a
Scheme 3
It has been proposed in the Na-H indole system that
indolenine/diene intermediates such as 14 are involved.30-37
Although higher in energy, by analogy it is conceivable that
the Na-methyl-substituted indolenine 14 was involved which
could only be attacked by the aldehyde from the top face of
the olefinic system. This would provide the stereochemistry
at C(6) illustrated in 14. Other pathways or equilibrium
processes might well be involved; however, if they proceed
via 14, the stereochemistry of the hydroxyl group at C(6)
would be set as â. Hydrolysis of 13 under standard conditions
provided 1 in enantiospecific fashion and in 36% overall
yield (from 3).
The readily available aldehyde 7b was also converted via
11 (Scheme 2) into (+)-ajmaline 2 (Scheme 4) by following
the chemistry of Li17,18 earlier reported from our laboratory.
This has now provided an enantiospecific route to (+)-2 in
13% overall yield (from 3, 8.9% overall yield from D-(+)-
tryptophan methyl ester).
a (a) HOAc/HCl(concd), 3 h; Ac2O/HCl(g), 18 h, aq workup,
85%; (b) BF3OEt2, CH2Cl2, PtO2, H2, 14 h, 89% (16a + 16b); (c)
CH3OH, 5% aq K2CO3, 93%.
prepared by a Barbier modification of the barium chemistry
of Yamamoto20,21 again in high yield. Because the DDQ
(aqueous THF)32,33 mediated oxidation at C(6) of aldehyde
12 provided the hemiacetal as the sole diastereomer, the route
to 1 is highly efficient. This modification 5 f [6] f 7b f
f 10 now provides a route to (+)-ajmaline 2 (13% from 3)
in much higher overall yield (8.9% from D-(+)-tryptophan
methyl ester) than previously reported and in fewer steps
(11 reaction vessels).13-18
The combination of the cyclization reactions described
above has permitted the use of D-(+)-tryptophan as both the
chiral auxiliary and the starting material in the synthesis of
1 and (+)-2 with extremely high stereocontrol.
In summary, the oxy-anion Cope rearrangement of 5,
followed by quenching of the enolate under kinetically
controlled conditions at -100 °C, has provided the C(15),
C(16), and C(20) asymmetric centers in 7b in a highly
stereoselective manner. The homoallylic alcohol 5 was
(34) Walker, D.; Hiebert, J. D. Chem. ReV. 1967, 67, 153.
(35) Braude, E. A.; Jackman, L. M.; Linstead, R. P. J. Chem. Soc. 1954,
3548.
Acknowledgment. We wish to thank Dr. Ivy Carroll for
(36) Jackman, L. M. AdV. Org. Chem. 1960, 2, 329.
(37) Bergman, J.; Carlsson, R.; Misztal, S. Acta Chem. Scand. B 1976,
30, 853.
helpful suggestions and the NIMH for support of this work.
OL000331G
348
Org. Lett., Vol. 3, No. 3, 2001