particular, our ability to partially reduce compound 1 (Figure
1).3 We sought a method of varying the stereochemistry
obtained after reduction and, in addition, the means to
produce either cis- or trans-diols at the C-3,4 positions. We
view this flexible approach to the simple pyrrolidine skeleton
(see DMDP4) as an introduction to the synthesis of the more
complex pyrrolizidine analogues such as australine, alexine
and hyacinthacine C1.1
This program of research relies on an efficient synthesis
of the diester starting material 1. After some experimentation,
we found that commercially available N-Boc pyrrole could
be doubly lithiated with lithium 2,2,6,6-tetramethylpiperidide
(LiTMP), followed by a quench with methyl chloroformate,
Scheme 1.5 With a multigram, one-step synthesis of 1 in
Figure 2.
Scheme 1a
The mechanism of reduction can be explained by forma-
tion of a relatively stable dianion A after addition of two
electrons to the starting material 1 (Figure 2). We suggest
that reaction with a bulky acid such as 2,6-di-tert-butylphenol
means that, after a single protonation, the resulting mo-
noenolate B reacts from the least hindered face to produce
cis-2. The reason for the trans selectivity during (kinetic)
protonation in ammonia is more obscure. At this juncture,
we wish only to point out the differences between reduction/
protonation in ammonia versus that in THF: (1) the
aggregation state of the enolate intermediate B is likely to
be very different in ammonia versus THF;8 (2) the geometry
of enolate B is not known and may differ depending on the
amount of chelation between the enolate OLi and the Boc
group; and (3) ammonia solvent has the ability to act as an
acid whereby the proton that becomes attached to the enolate
carbon may not be the same one as that added to the reaction
mixture as ammonium chloride.9
With the synthesis of cis- and trans-2 in hand, we
attempted to prepare the substitution pattern of the pyrrolidine
natural products. Ideally, we want to be able to set the
stereochemistry of the C-3,4 hydroxyl groups as either cis
or trans, starting from both cis- and trans-2.
Initially, the protected diols 3 and 6 were prepared via a
two-step sequence from the corresponding esters, Scheme
2. X-ray structural analysis confirmed the stereochemistry
of trans-2.
a Reagents and conditions: (a) LiTMP, MeOCOCI; (b) Li, NH3,
THF, then NH4Cl; (c) Li, cat. DBB, THF, then 2,6-di-tert-
butylphenol.
hand, we then examined the partial reduction of this com-
pound, searching for conditions that would influence the
stereochemistry of this process.
By variation of the reaction conditions, we discovered that
pyrrole 1 could be reduced to give the trans isomer of 2
with good diastereoselectivity, using lithium in ammonia and
quenching with ammonium chloride, Scheme 1.3 Remark-
ably, reduction under “ammonia-free” conditions (Li, cata-
lytic DBB, THF)6 followed by protonation with 2,6-di-tert-
butylphenol gave cis-2 exclusively.7 Control experiments
showed that each isomer was formed under kinetic control
and that cis-2 did not equilibrate under ammonia conditions.
This chemistry effectively achieves our first goal and allows
us to set the stereochemistry at C-5,6 at will.
The corresponding C-3,4 cis-diol motif was easily prepared
by dihydroxylation with catalytic osmium tetroxide. The
facial selectivity of oxidation is not an issue during formation
of 4 because compound 3 can only form a single cis
diastereoisomer. However, the oxidation of 6 was diaste-
reoselective upon dihydroxylation; the product 7 was shown
to be the C-2,3 anti compound by X-ray crystallography on
the N-Boc tetrol 9 derived from deprotection of 7. In both
(2) (a) Donohoe, T. J.; Guyo, P. M. J. Org. Chem. 1996, 61, 7664. (b)
Donohoe, T. J.; Guyo, P. M.; Beddoes, R. L.; Helliwell, M. J. Chem. Soc.,
Perkin Trans. 1 1998, 667.
(3) Donohoe, T. J.; Harji, R. R.; Cousins, R. P. C. Tetrahedron Lett.
2000, 41, 1327.
(4) Evans, S. V.; Fellows, L. E.; Shing, T. K. M.; Fleet, G. W. J.
Phytochemistry 1985, 24, 1953.
(5) Hasan, I.; Marinelli, E. R.; Lin, L.-C. C.; Fowler, F. W.; Levy, A. B.
J. Org. Chem. 1981, 46, 157.
(6) (a) Donohoe, T. J.; Harji, R. R.; Cousins, R. P. C. Tetrahedron Lett.
2000, 41, 1331. (b) Donohoe, T. J.; House, D. J. Org. Chem. 2002, 67, 5015.
(7) The diethyl ester analogue of 1 has previously been reduced and
reductively alkylated in ammonia.3 While reductive alkylation proceeds to
give the trans isomers exclusively, protonation in ammonia had been reported
to give a (7:3) mixture of cis and trans isomers. This ratio is our error, and
further study confirms that reduction (followed by protonation) in ammonia
definitely proceeds to give the trans isomer as the major compound (6:1).
(8) Schultz, A. G.; Macielag, M.; Sundararaman, P.; Taveras, A. G.;
Welch, M. J. Am. Chem. Soc. 1988, 110, 7828.
(9) See: Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624. This
is reminiscent of the Grotthus mechanism for proton transfer in water; see:
Pilling, M. J.; Seakins, P. W. Reaction Kinetics; Oxford University Press:
New York, 1995.
1000
Org. Lett., Vol. 5, No. 7, 2003