6806
L. M. Antunes, M. G. Organ / Tetrahedron Letters 44 (2003) 6805–6808
Scheme 1.
Bupleurynol is characterized by a fully conjugated
system consisting of two alkynes and three olefins,
the E/Z geometry of which is rigorously defined. We
saw this compound as an opportunity to demonstrate
the utility of an olefin template, such as (E) 2-iodo-1-
chloroethylene (5),3 used in the stereodefined synthesis
of disubstituted olefins.4 In this strategy, poly func-
tionalized olefin building blocks (templates) are used
whereby Pd can selectively and sequentially activate
these halides (or the corresponding organometallic
moieties) whose activities are electronically and/or
sterically differentiated. Thus, as demonstrated in Fig-
ure 1, the synthesis of 1 can be realized entirely by
an ordered sequence of transition metal-mediated cou-
pling reactions of simple compounds, all of which are
commercially available or obtainable in one simple
synthetic transformation.5 Herein we describe the first
total synthesis of Bupleurynol (1) in a convergent and
stereospecific manner.
ester (11, Scheme 2). Although the corresponding
boronic acid cross coupled well enough with 5 in the
presence of a Pd catalyst, the fact that the free acid
existed in equilibrium with the cyclic trimer meant
that stoichiometry could not be controlled as accu-
rately as desired. Compound 11 was coupled cleanly
at the iodide site of 5, the selectivity of which is due
to the weaker CꢀI bond. However, while a CꢀCl
bond activates slowly as a coupling partner in such
reactions, it is much more reactive in alkyne-related
coupling reactions, something that has been observed
by others.10 Thus, Sonogashira coupling with 6 pro-
duced 12 that, upon removal of the TMS group and
iodination, completed the synthesis of the left-hand
piece.
The final alkyne/alkyne coupling was not as straight-
forward to carry out operationally as it might appear
at first glance.11 The desired coupling between 2 and
3 is complicated by two competing dimerization pro-
cesses between the starting materials leading to the
formation of 13 and 14, respectively (Scheme 3). The
dimerization of 2 is likely to involve Pd whereas the
dimerization of simple terminal alkynes by the Cu is
known to be a competing process in Sonogashira
coupling reactions.12 When 2 and 3 were used in this
reaction in a ratio of 1:1.2, 113 was obtained in 40%
yield and dimers 13 and 14 accounted for the remain-
der of the mass balance. Presumably, oxidative addi-
tion of alkynyliodide 2 was the fastest event in the
reaction flask and self coupling (dimer formation) and
cross coupling with 3 were somewhat competitive pro-
cesses. Once all of 2 was gone, the only reaction pos-
sible for 3 is dimerization.
Construction of the right-hand piece 3 began from
propargyl alcohol. As shown in Scheme 1, iodination
at the terminal acetylenic position of 7 was followed
by reduction. Hydroboration of 3-iodo propargyl
alcohol actually proceeded well with an excess of the
boron reagent.6 However, byproducts of the cyclohex-
ane ligand co chromatographed and co distilled with
the desired product under all conditions tried, thus
we opted to protect the alcohol to make the separa-
tion more facile.7 Hydroboration/protonolysis8 and
deprotection gave 9, which was followed by Sono-
gashira coupling9 with 6, that upon deprotection pro-
vided 3. The TMS group can be removed during the
work up as well, thus eliminating one step. In this
instance, the TMS group is not a protecting group,
but rather is required for ease of handling (i.e. acetyl-
ene is a gas).
In summary, we have developed and completed a
synthetic route for the total synthesis of Bupleurynol.
This synthesis is based on a modular approach com-
prised of individual building blocks that were all
assembled using Pd catalysis. This represents the first
synthesis of this highly conjugated structure.
1-Octyne (10) was regioselectively hydroborated with
dibromoborane and hydrolyzed in situ to the boronic
acid that was then converted directly to the borate