Organosilicon-mediated regioselective acetylation of carbohydrates

Article abstract of DOI:10.1039/c2cc31556d

Organosilicon-mediated, regioselective acetylation of vicinal- and 1,3-diols is presented. Methyl trimethoxysilane or dimethyl dimethoxysilane was first used to form cyclic 1,3,2-dioxasilolane or 1,3,2-dioxasilinane intermediates, and subsequent acetate-c

Full text of DOI:10.1039/c2cc31556d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 5370–5372
www.rsc.org/chemcomm
COMMUNICATION
Organosilicon-mediated regioselective acetylation of carbohydratesw
a
b
a
¨
Yixuan Zhou, Olof Ramstrom and Hai Dong*
Received 1st March 2012, Accepted 3rd April 2012
DOI: 10.1039/c2cc31556d
Organosilicon-mediated, regioselective acetylation of vicinal- and
1,3-diols is presented. Methyl trimethoxysilane or dimethyl
dimethoxysilane was first used to form cyclic 1,3,2-dioxasilolane
or 1,3,2-dioxasilinane intermediates, and subsequent acetate-
catalyzed monoacylation was efficiently performed by addition
of acetic anhydride or acetyl chloride under mild conditions. The
reaction exhibited high regioselectivity, resulting in the same
protection pattern as in organotin-mediated schemes.
Scheme 1 Organosilicon-mediated acetylation of diols. R = Me or
Regioselective protection of vicinal diols under mild and
sustainable conditions remains a central challenge in chemistry.
This is particularly the case in carbohydrate chemistry, where
efficient protecting group strategies are required for the preparation
of value-added intermediates and building blocks for oligo-
saccharide synthesis. One such regioselective protection
method is based on organotin reagents,1 enhancing the oxygen
reactivity differences through the tin intermediates formed
with the parent diol. Despite the high utility of organotin-mediated
chemistry in this respect, the main drawback with these reagents is
the potential inherent toxicity of organotin compounds.2 For this
reason, nontoxic alternatives to common organotin reagents are
attractive, provided that they result in the same regioselective
protection patterns with similar yields. For example, the use
of catalytic amounts of dimethyltin dichloride,3 diarylborinic
acid-derived catalysts,4 enzyme-catalyzed5 and organo-catalytic
methods,6 silver(I) oxide mediated acylation,7 molybdenum-based
processes,8 and tandem catalytic reactions of persilylated sugar
derivatives9 have been exemplified, all of which with their respective
advantages and shortcomings. In the present study, organosilicon
reagents have been applied to regioselective protection of vicinal
and 1,3-diols at very mild conditions (Scheme 1). In comparison to
other methods, the organosilicon-mediated acylation protocol is
advantageous. The used reagents are associated with lower cost,
lower toxicity, and are more easily acquired.
OMe.
organosilicon reagents and either vicinal- or 1,3-diols.11 In a very
recent study by us, the origin of the resulting regioselectivities for
organotin-mediated protection has been attributed to the inherent
structure of the pyranoside.12 Thus, analogous selectivities could in
principle be obtained when the organotin reagent is replaced by
other reagents that can form cyclic dioxolane-type intermediates
with hydroxyl groups. These studies intrigued us to develop a
novel organosilicon-mediated regioselective carbohydrate protec-
tion method by making use of dioxasilolane or dioxasilinane
intermediates in analogy to the commonly used organotin struc-
tures. To our knowledge, regioselective introduction of protecting
groups through dioxasilolane- or dioxasilinane-type intermediates
has never been reported. Herein, we report a novel organosilicon-
mediated carbohydrate acylation method, where discrete hydroxyl
groups were regioselectively acetylated under very mild conditions,
using common acylation reagents and anionic activation.
Two common organosilicon reagents, dimethyl dimethoxysilane
and methyl trimethoxysilane, were chosen to test the reactions. It
was assumed that dioxasilolane- or dioxasilinane rings would be
formed when dimethyl dimethoxysilane or methyl trimethoxysilane
reacts with vicinal- or 1,3-diols in acetonitrile at 80 1C, with
continuous removal of the methanol formed in the reaction.
Regioselective hydroxyl group protection was subsequently
attempted through the preferential cleavage of one of the Si–O
bonds of the dioxasilolane- or dioxasilinane rings (Scheme 1).
In contrast to the analogous Sn–O bonds, addition of anions
proved necessary for inducing cleavage of the relatively strong
Si–O bond. In line with our recent studies of anionic reagents,13
tetrabutylammonium acetate was thus primarily chosen to activate
the cyclic intermediates.
Organosilicon reagents are widely used as protecting groups
in organic synthesis.10 It has also been observed that cyclic
1,3,2-dioxasilolane-, 1,3,2-dioxasilinane-, or 1,3,5,2,4-trioxadi-
silepane-type intermediates can be formed between certain
a Huazhong University of Science & Technology, School of Chemistry
& Chemical Engineering, Luoyu Road 1037, 430074 Hongshan,
Wuhan, P.R. China. E-mail: hdong@mail.hust.edu.cn;
Fax: +86 27 87793242
In principle, there are three possible pathways for the
acylation reaction under these conditions, and these were
initially evaluated (Scheme 2). Firstly (I), when the acylation
reagent and the anionic activation reagent were allowed to
b Royal Institute of Technology (KTH), Department of Chemistry,
Teknikringen 30, S-10044 Stockholm, Sweden
w Electronic supplementary information (ESI) available: General
methods, syntheses and 1H NMR data. See DOI: 10.1039/c2cc31556d
c
5370 Chem. Commun., 2012, 48, 5370–5372
This journal is The Royal Society of Chemistry 2012

View Article Online
a trans-diol where the substituent adjacent to the 2-OH is axial
and the substituent adjacent to the 3-OH is equatorial; methyl
2,6-di-O-benzyl-b-D-galactoside 6, a cis-diol; and methyl 2,3-
di-O-benzyl-a-D-glucoside 8 and methyl 2,3-di-O-benzyl-a-D-
galactoside 10, two 1,3-diols with one primary hydroxyl
group. These compounds were allowed to react with dimethyl
dimethoxysilane or methyl trimethoxysilane in acetonitrile.
After removal of most of the solvent, the residues were allowed
to react with acetic anhydride in acetonitrile in the presence of
tetrabutylammonium acetate. The results (cf. Table 1) showed
high regioselectivities and generally fair to good isolated
yields. One of the hydroxyl groups was thus regioselectively
acetylated in each case, and the recorded product pattern
exclusively followed the product outcome from organotin-
mediated protection. Thus, the hydroxyl groups adjacent to
axial substituents were acetylated in the trans-diols 1 and 4; the
equatorial hydroxyl groups were acetylated in the cis-diols 6;
and the primary hydroxyl group was acetylated in the 1,3-diols
8 and 10.
Scheme 2 Analysis of organosilicon-mediated acetylations. R = OMe.
react with methyl 4,6-O-benzylidene-b-D-galactoside 1 in the
absence of an organosilicon reagent, no acetylation was
recorded within the timeframe of the reaction. Reactions with
excess triethylamine generally resulted in mixtures of compounds,
and low conversions even after prolonged reaction times at higher
temperature (cf. Table S3 in ESIw). Secondly (II), when the
organosilicon reagent, dimethyl dimethoxysilane or methyl tri-
methoxysilane, was first allowed to react with methyl 4,6-O-
benzylidene-b-D-galactoside 1 in acetonitrile at 80 1C for 5 hours,
followed by removal of most of the solvent, it could be observed
by 1H-NMR analysis that two of the OMe groups were
removed (cf. Fig. S1 and S2 in ESIw), indicating formation
of the dioxasilolane intermediate. Regioselective acetylation
subsequently occurred upon addition of the acylation reagent
and the anionic activation reagent. 1H-NMR analysis together
with the absence of acyl group migration14 finally disproved
the occurrence of the silyl ether formation (III), where the final
major product would be galactoside 3. These observations
support the necessity of intermediate dioxasilolane formation
in order to obtain galactoside 2 as the major product. Further
experiments indicated that the organosilicon-mediated protection
did not take place without the addition of acetate anions, and
anionic activation was thus necessary for the cleavage of the Si–O
bond of the dioxasilolane- or dioxasilinane rings.
The regioselectivity is affected by the nature of the acylating
reagent, for example reported in acylations of polyols.19 For
organotin-mediated protection, opposing regioselectivities
were observed when acetic anhydride or acetic chloride was
used as acetylation reagents, respectively.20 A proposed
mechanism suggested that migration is blocked with acetic
Table 1 Organosilicon-mediated acetylation of 1,2- and 1,3-diolsa
Entry Reactant
Product
Condition Yieldb (%)
A
B
C
75
74
76
1
A
B
C
65
63
64
2
3
A
C
79
78
For organotin regioselective protection, 2,2-dibutyl-1,3,2-
dioxastannolanes formed from vicinal cis-diols of pyranoside
rings give major products protected on the equatorial hydroxyl
groups.15 The dioxastannolanes of the corresponding trans-
diols, on the other hand, result in mixtures if the adjacent
substituents are either both equatorial or both axial,16 but give
major products protected on the hydroxyl group adjacent to
the axial substituent if one adjacent substituent is equatorial
and one is axial.17 Dioxastannolanes or dioxastanninanes
formed from terminal diols with one primary hydroxyl group
normally give major products protected on the primary position.18
In the present study, a range of structures were studied in order to
compare the selectivities observed with organosilicon reagents
with those of organotin-mediated methods (cf. Table 1): 4,6-O-
benzylidene-b-D-galactoside 1, a trans-diol where the substituent
adjacent to the 2-OH is equatorial and the substituent adjacent to
the 3-OH is axial; methyl 4,6-O-benzylidene-a-D-glucoside 4,
A
C
80
78
4
5
B
B
84
85
6
a
Reaction conditions: reactant (100 mg), TBAOAc (0.3 eq.),
(A) Ac2O (1.1 eq.) and Me2Si(OMe)2 (2.0 eq.); (B) MeSi(OMe)3
(1.2 eq.) and Ac2O (1.1 eq.); (C) Me2Si(OMe)2 (2.0 eq.) and AcCl
b
(1.1 eq). Isolated yield.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5370–5372 5371

View Article Online
anhydride due to bidentate coordination of the produced
acetate to the tin center, whereas this is possible with acetyl
chloride where monodentate coordination takes place. For
organosilicon protection, migration should be less pronounced
since the Lewis acidity is lower, resulting in similar selectivities
using acetyl chloride and acetic anhydride. This effect was thus
further addressed, and the results are displayed in Table 1.
As can be seen, the expectations were fulfilled, and acetyl
chloride produced very similar yields and selectivities as acetic
anhydride in the acetylation reactions. When the method was
expanded to protect the primary hydroxyl group of 1-phenyl-
1,2-ethanediol 12, good yields and selectivities were also
obtained.
This study was supported by HUST, the Chutian Project-
Sponsored by Hubei Province and the Project-Sponsored by
the SRF for ROCS, SEM. The authors are also grateful to the
staffs in the Analytical and Test Center of HUST for support
with the NMR instruments.
Notes and references
1 T. B. Grindley, Adv. Carbohydr. Chem. Biochem., 1998, 53, 17–142.
2 (a) M. M. Whalen, B. G. Loganathan and K. Kannan, Environ.
Res., 1999, 81, 108–116; (b) S. M. Jenkins, K. Ehman and
S. J. Barone, Dev. Brain Res., 2004, 151, 1–12 42, 213–224.
3 Y. Demizu, Y. Kubo, H. Miyoshi, T. Maki, Y. Matsumura,
N. Moriyama and O. Onomura, Org. Lett., 2008, 10, 5075–5077.
4 D. Lee and M. S. Taylor, J. Am. Chem. Soc., 2011, 133, 3724–3727.
5 (a) M. Therisod and A. M. Klibanov, J. Am. Chem. Soc., 1987, 109,
3977–3981; (b) Y.-F. Wang, J. J. Lalonde, M. Momongan, D. E.
Bergbreiter and C.-H. Wong, J. Am. Chem. Soc., 1988, 110, 7200–7205.
6 (a) K. S. Griswold and S. J. Miller, Tetrahedron, 2003, 59,
8869–8875; (b) G. Hu and A. Vasella, Helv. Chim. Acta, 2002, 85,
4369–4391; (c) T. Kawabata, W. Muramatsu, T. Nishio, T. Shibata
and H. Schedel, J. Am. Chem. Soc., 2007, 129, 12890–12895.
7 H. Wang, J. She, L.-H. Zhang and X.-S. Ye, J. Org. Chem., 2004,
69, 5774–5777.
8 E. V. Evtushenko, J. Carbohydr. Chem., 2010, 29, 369–378.
9 (a) C. C. Wang, J. C. Lee, S. Y. Luo, S. S. Kulkarni, Y. W. Huang,
C. C. Lee, K. L. Chang and S. C. Hung, Nature, 2007, 446,
896–899; (b) M. A. Witschi and J. Gervay-Hague, Org. Lett.,
2010, 12, 4312–4315; (c) Y. Bourdreux, A. Lemetais, D. Urban
and J. M. Beau, Chem. Commun., 2011, 47, 2146–2148.
10 P. G. M. Wuts and T. W. Greene, Green’s Protective Groups
in Organic Synthesis, John Wiley & Sons, Inc., Hoboken, NJ,
4th edn, 2007.
In order to further compare the selectivities observed with
organosilicon reagents with those of organotin-mediated
1
methods, more structures were tested and the H NMR ratio
of products was recorded (cf. Table S1 and S2 in ESIw). It can
be seen that the protection results showed the same selectivity
pattern as for traditional organotin-mediated protection. It is
worth noting the acetylation of the trans-diols 19 and 22 where
the adjacent substituents are either both equatorial or both
axial. As expected, inferior selectivities were observed (entries
5 and 6 in Table S1, ESIw). The NMR ratios were recorded as
57/43 for compound 19 and as 56/44 for compound 22,
respectively. High regioselectivities were on the other hand
observed in the acetylation of 1,3-diols (entries 7–14 in Table S1,
ESIw). The primary hydroxyl group was in these cases acetylated
with NMR ratios of 490%.
Thus, this straightforward two-step reaction, with simple
intermittent solvent removal, proceeded efficiently with all
reactants tested, producing the predicted monoacylated products
in all cases. Anhydrous conditions were generally required, since
moisture disrupted the formation of the dioxasilolane or dioxa-
silinane rings, potentially resulting in lower reaction yields. Since
analogous selectivities were obtained when the organotin reagent
is replaced by the organosilicon reagent that can form cyclic
dioxolane-type intermediates with hydroxyl groups, the results
further point to the fact that the regioselectivity of organotin-
mediated protection is not related to the complex structure of the
dioxastannolane (or dioxastanninane) intermediate, but mainly
depends on the inherent structure of the pyranoside. The scope
of the organosilicon-based methodology is currently under
investigation, where for example organosilicon-mediated
benzylations, glycosylations, tosylations, and silylations are under
development.
11 (a) X. l. Wang, Q. l. Meng, N. R. Perl, Y. Xu and J. L. Leighton,
J. Am. Chem. Soc., 2005, 127, 12806–12807; (b) R. Tacke,
R. Bertermann, C. Burschka and S. Dragota, Angew. Chem., Int.
Ed., 2005, 44, 5292–5295; (c) B. J. Ager, L. E. Bourque,
K. M. Buchner and K. A. Woerpel, J. Org. Chem., 2010, 75,
5729–5732.
12 H. Dong, Y. X. Zhou, X. L. Pan, F. C. Cui, W. Liu, J. Y. Liu and
O. Ramstrom, J. Org. Chem., 2012, 77, 1457–1467.
¨
13 H. Dong, Z. C. Pei and O. Ramstrom, Chem. Commun., 2008,
¨
1359–1361.
14 S. L. Deng, U. Gangadharmath and C. W. T. Chang, J. Org.
Chem., 2006, 71, 5179–5185.
15 (a) Z. Zhang and G. Magnusson, J. Org. Chem., 1996, 61,
2383–2393; (b) F.-Y. Dupradeau, J. Prandi and J. M. Beau, Tetra-
hedron, 1995, 51, 3205–3220; (c) S.-H. Kim, D. Augeri, D. Yang
and D. Kahne, J. Am. Chem. Soc., 1994, 116, 1766–1775.
16 (a) H. Qin and T. B. Grindley, J. Carbohydr. Chem., 1996, 15,
95–108; (b) T. B. Grindley and R. Thangarasa, Can. J. Chem.,
1990, 68, 1007–1019; (c) F. Dasgupta and P. J. Garegg, Synthesis,
1994, 1121–1123.
17 (a) D. D. Manning, X. Hu, P. Beck and L. L. Kiessling, J. Am.
Chem. Soc., 1997, 119, 3161–3162; (b) S. David and A. Malleron,
Carbohydr. Res., 2000, 329, 215–218; (c) F. Peri, L. Cipolla and
F. Nicotra, Tetrahedron Lett., 2000, 41, 8587–8590.
18 (a) D. A. Leigh, R. P. Martin, J. S. Smart and A. M. Truscello, J. Chem.
Soc., Chem. Commun., 1994, 1373–1374; (b) S. Knapp, C. Jaramillo and
B. Freeman, J. Org. Chem., 1994, 59, 4800–4804; (c) J. T. Randolph and
S. J. Danishefsky, J. Am. Chem. Soc., 1995, 117, 5693–5700.
19 M. Flores-Mosquera, M. Martin-Lomas and J. L. Chiara, Tetrahedron
Lett., 1998, 39, 5085–5088.
A novel organosilicon-mediated regioselective acetylation reac-
tion proceeding through cyclic dioxasilolane- or dioxasilinane-type
intermediates has been developed. The resulting regioselectivities
were dependent on the parent diol structure, showing the same
product pattern as for traditional organotin-mediated approaches.
The method proved straightforward, representing an affordable,
benign alternative to organotin protocols.
20 H. Dong, Z. C. Pei, S. Bystrom and O. Ramstrom, J. Org. Chem.,
2007, 72, 1499–1502.
¨
¨
c
This journal is The Royal Society of Chemistry 2012
5372 Chem. Commun., 2012, 48, 5370–5372