Organic Letters
Letter
sugar motif. Retrosynthetically (Scheme 1), fusaroside 1 can be
obtained via aldol condensation of trehalose derivative 2 and
benzyl groups. Among the silyl groups, the TMS groups are
labile under acidic as well as basic conditions, whereas
placement of TBS groups on adjacent equatorial hydroxyl
(OH) groups of glucose has been shown to cause conforma-
tional changes in the ring.12 We therefore decided to protect
the trehalose OH groups with PMB ether, which can be
deprotected under acidic conditions without affecting the other
functional groups in the molecule. For this purpose,
monoanisaldehyde acetal formation of trehalose (52%)
followed by PMB etherification of the remaining free hydroxyl
groups (97%) and subsequent regioselective reductive ring
opening of the acetal at the O-4 position using BH3·THF (5
equiv) and dibutylboryl trifluoromethanesulfonate (2.5 equiv)
at −78 °C13 furnished compound 4 in 91% yield, exclusively.
With the requisite 4-OH trehalose derivative 4 in hand, the
stage was set for acylation. First, we tried DCC-mediated
coupling of alcohol 4 and acid 511 to obtain the desired
compound 2. However, the coupling did not occur, and we
recovered compound 4 as such. Reaction with a different
coupling reagent, EDC·HCl, or with an acid chloride also did
not give the desired compound 2. We rationalized that perhaps
due to the steric hindrance and concomitant decarboxylation
of acid 5, the acylation reaction was unsuccessful.
Scheme 1. Retrosynthetic Analysis
Next, to prevent the possibility of decarboxylation, we
decided to perform acylation with acetoacetic acid first and to
introduce the gem-dimethyl group later. Accordingly, a DCC-
mediated acylation of acetoacetic acid14 with compound 4
followed by geminal dimethylation using methyl iodide and
sodium hydride gave our desired compound 2 in 90% yield
over two steps. After the successful synthesis of compound 2,
the stage was set for the aldol reaction with known15 aldehyde
11. Initial attempts to carry out the aldol reaction using LDA
did not give the desired product. Instead, the self-aldol product
of aldehyde 11 was obtained.16 Likewise, aldol reaction
attempted using titanium tetrachloride led to the formation
of PMB-hydrolyzed products of compound 2. On the contrary,
the reaction using dibutylboryl trifluoromethanesulfonate did
not progress and the starting compound 2 was recovered.
As the aldol reaction on a sugar scaffold failed to give
compound 12, we decided to carry out the aldol reaction first
to synthesize the fully functionalized carbon chain, which we
thought could be attached with the C-4-OH trehalose
derivative 4 via acylation at a later stage. This time, the aldol
reaction proceeded smoothly to deliver the corresponding
aldehyde 3, followed by selective removal of p-methoxybenzyl
(PMB) groups. Compound 2 can be synthesized by carrying
out esterification of trehalose 4-OH derivative 4 with known11
acid 5. Aldehyde 3 can be synthesized in three steps by
sequential Julia−Kocienski olefination of sulfone 7 with α,β-
unsaturated monoaldehyde 6 followed by tert-butyldimethyl-
silyl (TBS) deprotection and pyridinium chlorochromate
(PCC) oxidation of the corresponding alcohol. Compounds
6 and 7 in turn could be synthesized from commercially
available 1,7-heptanediol 8 and 1,5-pentanediol 9, respectively.
Accordingly, our immediate goal was to couple acid 5 with
the trehalose core at the C-4-OH position and try out
conditions for aldol condensation to synthesize a structurally
simple analogue of fusaroside (Scheme 2). The presence of an
ester linkage and various double bonds in the target molecule
precluded the use of other ester type groups or benzyl groups,
as we could no longer use basic conditions, or hydrogenation/
Birch reduction for their selective removal. This limited the
choice of permanent protecting groups to silyl and substituted
Scheme 2. Attempted Synthesis of Glycolipid Derivative 12
1665
Org. Lett. 2021, 23, 1664−1668