1
840
F. S. Ekholm et al. / Tetrahedron Letters 52 (2011) 1839–1841
Table 1
A summary of the performed deprotections13
Entry
Substrate
Product
Yielda (%)
95
OBn
O
OH
O
BnO
HO
BnO
HO
1
2
3
4
HO
BnO
OMe
OBn
OMe
OH
TBDMSO
BnO
TBDMSO
HO
O
O
BnO
HO
95
95
90
Figure 1. Schematic representation of the CF reactor.
OMe
OMe
OAc
OAc
O
Ph
Ph
O
HO
HO
not influence other functional groups.1,2 One of the major concerns
O
O
AcO
AcO
associated with the deprotection of carbohydrate derivatives is the
extreme change in the nature of the molecules brought on by the
cleavage of protecting groups. While the protected carbohydrates
are relatively hydrophobic, the deprotected ones are hydrophilic.
Therefore, the choice of the solvent system turned out to be crucial,
in order to maintain the solubility of both the starting material and
the product but simultaneously avoiding rapid solidification of the
material, which would cause stocking of tubes and lines. A solvent
system containing MeOH:EtOAc 1:1 was found to be suitable when
the sample concentration was 1 mg/ml.
OMe
OMe
OH
OBn
O
O
HO
HO
O
O
BnO
HO
OMe
OMe
a
Isolated yield.
Because the yields were high in all cases no additional efforts to
further optimize the reaction conditions were made. It should be
noted that all experiments were carried out with a single run
and the easiest way of obtaining higher conversions, in the case
of low conversions, would be to perform additional runs. In addi-
tion, it is worth to mention that the CF reactor was a small scale
model and, therefore, a further up-scaling of the reactions to gram
or kilogram scale should be possible by using a larger version of the
Most debenzylation reactions described have been performed
by the use of acidic additives.8 Due to the presence of acid labile
protecting groups the use of additional acids was avoided. There-
fore, new reaction conditions needed to be developed and opti-
mized. Since Pd/C is commonly used as
a catalyst when
debenzylations are performed in batch reactors, it was also our ini-
tial choice of catalyst and a standard 30 mm long holder was filled
9
reactor. Future work will focus on the one-pot removal of other
types of protective groups, for example, base- and acid labile pro-
tective groups, with the subsequent hydrogenolyses of benzyl and
benzylidene groups. This extension of the methodology is expected
to find wide use as it will further decrease the amount of synthetic
steps needed to obtain fully deprotected natural products.
To conclude, a simple, rapid and efficient methodology for the
deprotection of benzyl- and benzylidene protected carbohydrates
has been developed utilizing a CF hydrogenation reactor. The
method tolerates both acid- and base sensitive functional groups
as shown by the deprotection of the derivatives containing acetyl
and silyl protective groups. The high efficiency, simple work-up,
and short reaction time should make this method appealing to a
researcher working with carbohydrate chemistry and total
synthesis.
6
,11
with it.
Compound 1 was selected as a model substrate and
-pres-
optimal reaction parameters were found utilizing 40 bar H
2
sure, a reactor temperature of 80 °C, and a flow velocity of 1 ml/
min. Another advantage of performing the reactions in a CF reactor
is that the reaction is followed by simple evaporation as opposed to
the batch reactors where filtration through celite is necessary be-
4
c,6a,12
fore evaporation.
Deprotection of 1 proceeded smoothly un-
der the mentioned conditions, thereby resulting in methyl
a-D-
mannopyranoside in almost quantitative yield. 1H NMR spectros-
copy was found to be a simple and convenient method for monitor-
ing of the reaction, especially following the disappearance of the
1
aromatic signals from the H NMR spectra.
With compounds 2 and 3, the reactions proceeded with full
conversions and the compounds could be isolated in high yields
(
Table 1). In both cases the additional protective groups remained
Acknowledgments
intact, thereby showing that these groups remain unaltered under
the employed reaction conditions. While the reactions were mainly
performed on a 20 mg scale, deprotection of compound 2 was suc-
cessfully upscaled to 200 mg scale. Deprotection of compound 4
Financial support from the Academy of Finland (Projects
#121334 and #121335) and the Finnish Glycoscience Graduate
School is gratefully acknowledged.
gave methyl
a-
D-mannopyranoside with 95% conversion. In order
to further optimize the conditions for this substrate, the H
2
-pres-
References and notes
sure was increased (50, 60 and 70 bar), however, upon increased
pressure the conversion dropped dramatically. This decrease of
conversion may be due to a rapid dissociation of the substrate from
the metal surface thus hampering the reaction. It was also noted
1. See, for example: (a) Nicolaou, K. C.; Mitchell, H. J. Angew. Chem., Int. Ed. 2001,
4
0, 1576–1624; (b) Thorson, J. S.; Hosted, T. J., Jr.; Jiang, J.; Biggins, J. B.; Ahlert,
J. Curr. Org. Chem. 2001, 5, 139–167; (c) Weymouth-Wilson, A. C. Nat. Prod. Rep.
997, 14, 99–110.
1
that lower H
2
-pressures (<40 bar) resulted in decreased conver-
2. See, for example: (a) Werz, D. B.; Seeberger, P. H. Chem. Eur. J. 2005, 11, 3194–
3206; (b) Imberty, A.; Varrot, A. Curr. Opin. Struct. Biol. 2008, 18, 567–576; (c)
Carmona, A. T.; Moreno-Vargas, A. J.; Robina, I. Curr. Org. Chem. 2008, 5, 33–60.
sions. The reason for this is most likely that the H
these cases was not sufficient.
2
-pressure in
3.
See, for example: (a) Robertson, J.; Stafford, P. M. Carbohydrates In Osborn, H.
M. I., Ed.; Academic Press-Elsevier: Oxford, 2003. pp 9–65; (b) Kanie, O.;
Hindsgaul, O. Curr. Opin. Struct. Biol. 1992, 2, 674–681; (c) Whitfield, D. M.;
Douglas, S. P. Glycoconjugate J. 1996, 13, 5–17.
4.
See, for example: (a) Garegg, P. J. Acc. Chem. Res. 1992, 25, 575–580; (b)
Seeberger, P. H. J. Carbohydr. Chem. 2002, 21, 613–643; (c) Ekholm, F. S.;
Sinkkonen, J.; Leino, R. New J. Chem. 2010, 34, 667–675.
5
.
.
Kocie n´ ski, P. J. In Protecting Groups; Thieme, Georg, Ed.; Verlag: Stuttgart, 2005.
Vol. 3, pp 137–155 and 241–257.
See for example: (a) Crich, D.; Banerjee, A.; Yao, Q. J. Am. Chem. Soc. 2004, 126,
6
14930–14934; (b) Hewitt, M. C.; Seeberger, P. H. J. Org. Chem. 2001, 66, 4233–
Figure 2. The substrates utilized in the deprotection study.
4243; (c) Oscarson, S.; Sehgelmeble, F. W. J. Org. Chem. 2002, 67, 8457–8462.