Tuning effect of silyl protecting groups on the glycosylation reactivity of arabinofuranosyl thioglycosides

Article abstract of DOI:10.1021/ol401166x

The tuning effect of silyl protecting groups on the glycosylation reactivity of arabinofuranosyl phenyl thioglycoside donors is presented. Silyl ethers on the 3-, 5-, and 3,5-positions of the arabinofuranose ring are found to have an arming effect on the

Full text of DOI:10.1021/ol401166x

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Tuning Effect of Silyl Protecting Groups
on the Glycosylation Reactivity of
Arabinofuranosyl Thioglycosides
Xing-Yong Liang,†,§ Hua-Chao Bin,†,§ and Jin-Song Yang*,†,‡
Key Laboratory of Drug Targeting and Drug Delivery Systems of the Ministry of
Education, Department of Chemistry of Medicinal Natural Products, West China School
of Pharmacy, and State Key Laboratory of Biotherapy, West China Hospital,
Sichuan University, Chengdu 610041, China
yjs@scu.edu.cn
Received April 25, 2013
ABSTRACT
The tuning effect of silyl protecting groups on the glycosylation reactivity of arabinofuranosyl phenyl thioglycoside donors is presented. Silyl
ethers on the 3-, 5-, and 3,5-positions of the arabinofuranose ring are found to have an arming effect on the donor reactivity, whereas the cyclic
3,5-acetal type protecting groups reduce the reactivity.
The stereoelectronic characteristics of protecting groups
play a significant role in carbohydrate reactivity.1 The armedÀ
disarmed concept, which has proven to be a powerful tool in
the concise synthesis of a cisÀtrans linked trisaccharide
sequence, relies upon the fact that benzylated (armed) sugars
are always more reactive than their acylated (disarmed)
analogues.2
In order to facilitate the synthesis of complex oligosac-
charides and glycoconjugates, a wider range of saccharide
building blocks with different anomeric reactivity is re-
quired. In this respect, Wong and co-workers quantified
the relative reactivity of numerous thioglycoside donors3
and developed a computerized program for the design of
synthetic schemes for one-pot multistep chemoselective
assemblies of many important oligosaccharides.4 They re-
cently further studied the graded arming of p-tolyl thiogluco-
sides by variously positioned silyl groups and the application
of these findings in the reactivity-based one-pot assembly of
linker-attached Lc4 and IV2Fuc-Lc4, which are components
of the human embryonic stem cell surface.5 Bols et al.
reported that bulky silyl group substituted pyranose thiogly-
cosides have superior reactivity compared with the benzyl-
ated thioglycosides.6 The enhanced reactivity can be ex-
plained by the stereoelectronic effects associated with the
conformational change caused by the silyl protection.
On the other hand, cyclic protecting groups were found
to have a torsionally deactivating effect on pyranoside
reactivity. Fraser-Reid and co-workers first disclosed that
the oxidative hydrolysis of a series of acetalated pyranosyl
n-pentenyl glycosides proceeds less readily than for their
torsion-free analogues.7 Later, the Ley group showed that
† West China School of Pharmacy.
‡ West China Hospital.
§ These authors contributed equally.
ꢀ
ꢀ
(1) Fraser-Reid, B.; Jayaprakash, K. N.; Lopez, J. C.; Gomez, A. M.;
Uriel, C. In Frontiers in Modern Carbohydrate Chemistry; Demchenko,
A. V., Ed.; American Chemical Society: Washington, DC, 2007; p 91.
(2) (a) Mootoo, D. R.; Konradsson, P.; Udodong, U.; Fraser-Reid,
B. J. Am. Chem. Soc. 1988, 110, 5583. (b) Fraser-Reid, B.; Wu, Z.;
Udodong, U.; Ottosson, H. J. Org. Chem. 1990, 55, 6068.
(5) Hsu, Y.; Lu, X.-A.; Zulueta, M. M. L.; Tsai, C.-M.; Lin, K.-I.;
Hung, S.-C.; Wong, C.-H. J. Am. Chem. Soc. 2012, 134, 4549.
(6) (a) Jensen, H. H.; Pedersen, C. M.; Bols, M. Chem.;Eur. J. 2007,
13, 7576. (b) Pedersen, C. M.; Marinescu, L. G.; Bols, M. Chem.
Commun. 2008, 2465. (c) Pedersen, C. M.; Nordstrøm, L. U.; Bols, M.
J. Am. Chem. Soc. 2007, 129, 9222.
(3) Zhang, Z.; Ollmann, I. R.; Ye, X.-S.; Wischnat, R.; Baasov, T.;
Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734.
(4) Selected examples: (a) Polat, T.; Wong, C.-H. J. Am. Chem. Soc.
2007, 129, 12795. (b) Duron, S. G.; Polat, T.; Wong, C.-H. Org. Lett.
2004, 6, 839. (c) Mong, T. K.-K.; Lee, H.-K.; Duron, S. G.; Wong, C.-H.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 797.
(7) Fraser-Reid, B.; Wu, Z.-F.; Andrews, C. W.; Skowronski, E.
J. Am. Chem. Soc. 1991, 113, 1434.
r
10.1021/ol401166x
XXXX American Chemical Society

Table 1. Competition Glycosylation Reactionsa
a Glycosylations were run with donors A/B, acceptor 3a (for entries 1À5, 7À8, and 11À12) or 3b (for entries 6 and 9À10), NIS (1.2 equiv), and TfOH
(0.1 equiv), 4 A molecular sieves (MS) in dry CH2Cl2 in 1 h. b Isolated yield. c R/β = 1/1. The R/β ratios in entries 1, 2, and 7 were determined by 1H NMR
˚
analysis of the corresponding anomer mixtures. d R/β = 3/1. e R/β = 3/1.
thioglycosides, bearing a dispiroketal (dispoke) or cyclo-
hexane-1,2-diacetal (CDA) protecting group, have reac-
tivities between armed and disarmed thioglycosides.8
Furthermore, Boons et al. reported that 2,3-O-carbonate-
masked ethyl thioglucosides can act as acceptors in che-
moselective glycosylations with disarmed thioglycosides
due to their reduced reactivity.9
Arabinofuranosides are widely distributed in living
organisms.10 Both D- and L-arabinosides are found, with
the D-form being the major component of mycobacterial
cell walls and the L-form being an important constituent of
plant cell walls. Because of their biological relevance, the
synthesis of arabinose-containing oligosaccharides has
received considerable attention in the past decade.11 But
to date, systematical study of the effect of protecting groups
on the glycosylation reactivity of furanosyl donors has yet to
occur.12 Establishment of the relative reactivity of glycosyl
building blocks can not only provide valuable insight into
the nature of a given furanose donor but also pave the way
to the development of new furanosylation chemistry. In our
search for new methods to synthesize oligofuranoses,13 we
recently reported a novel regioselective furanosylation ap-
proach and its application in the one-pot synthesis of
natural oligoarabino- and galactofuranosides.13a We now
present, in this Letter, the tuning effect of silyl protections
on the reactivity of arabinosyl thioglycoside donors. This
work extends the conventional armedÀdisarmed concept.
Our studies began with the design and preparation of
a set of D- and L-arabinofuranose phenyl thioglycoside
(8) (a) Ley, S. V.; Priepke, H. W. M. Angew. Chem., Int. Ed. Engl.
€
1994, 33, 2292. (b) Douglas, N. L.; Ley, S. V.; Lucking, U.; Warriner,
S. L. J. Chem. Soc., Perkin Trans. 1 1998, 51.
(9) Zhu, T.; Boons, G.-J. Org. Lett. 2001, 3, 4201.
(10) Lowary, T. L. In Glycoscience: Chemistry and Chemical Biology;
Fraser-Reid, B., Tatsuta, K., Thiem, J., Eds.; Springer: Berlin, 2001; p 2005.
(11) For reviews, see: (a) Lowary, T. L. Curr. Opin. Chem. Biol. 2003,
7, 749. (b) Peltier, P.; Euzen, R.; Daniellou, R.; Nugier-Chauvin, C.;
(12) Lowary and co-workers reported the influence of the reactivity
of D-arabinofuranosyl thioglycosides on the glycosylation stereoselec-
tivity; see: Yin, H.-F.; Lowary, T. L. Tetrahedron. Lett. 2001, 42, 5829.
(13) (a) Deng, L.-M.; Liu, X.; Liang, X.-Y.; Yang, J.-S. J. Org. Chem.
2012, 77, 3025. (b) Zhu, S.-Y.; Yang, J.-S. Tetrahedron 2012, 68, 3795. (c)
Liang, X.-Y.; Deng, L.-M.; Liu, X.; Yang, J.-S. Tetrahedron 2010, 66, 87.
ꢁ
Ferrieres, V. Carbohydr. Res. 2008, 343, 1897. (c) Imamura, A.; Lowary,
T. L. Trends Glycosci. Glycotech. 2011, 23, 134.
B
Org. Lett., Vol. XX, No. XX, XXXX

derivatives protected as silyl ethers and 3,5-O-acetals
(Table 1, 1aÀg, L-1b, andL-1f) by known or slightly modified
experimental procedures (see the Supporting Information).
With these compounds, a series of three-component compe-
tition glycosylations were run to assess the influence of silyl
protecting groups on glycosylation reactivity. In these experi-
ments, both a silylated thioarabinoside (donor A, 1.0 equiv)
and a nonsilylated armed or disarmed thioarabinoside
(donor B, 1.0 equiv) would be mixed in the same reaction
vessel in dry dichloromethane with a model acceptor arabi-
noside 3a13c or 3b14 (0.9 equiv), respectively. Upon addition
of the N-iodosuccinimide (NIS, 1.2 equiv)/trifluoromethane-
sulfonic acid (TfOH, 0.1 equiv) promoter system,15 the two
donors would then compete to couple with the acceptor to
form disaccharide products. The product ratio will reflect the
inherent reactivity of the donors. The results of these compe-
tition glycosylations are summarized in Table 1.
In the first instance, the competitive reaction between
the fully tert-butyldimethylsilyl (TBS) protected donor 1a
and the fully benzylated donor 2a16 with 5-OH acceptor 3a
gave (1f5)-linked disaccharide 4 as a mixture of anomers
(R/β = 1/1) in 85% yield (Table 1, entry 1). No coupling
product between 2a and 3a was detected under these
reaction conditions. More than 50% of the added 2a was
recovered after workup. Next, the glycosylation reaction
of the donors with a mixed protecting group pattern was
examined. Predominant activation of 3,5-di-O-TBS-2-O-Bz
protected 1b over 2a took place, giving 5a as the major
product along with a minor quantity of 5b (entry 2). So,
these results can be attributed to the greater anomeric
reactivity of the per- and disilylated donors 1a,b relative to
that of the per-benzylated armed donor 2a. Subsequently,
when compared with the less armed 2b having 3,5-dibenzyl-
2-Bz substituents, 1b and its 3,5-di-O-tert-butyldiphenylsilyl
(TBDPS) protected analogue 1c could be exclusively acti-
vated to glycosylate with 3a, to afford solely the disaccharides
5a and 5c with complete stereoselectivity in good yield,
respectively (entries 3À4). The mono-TBS masked substrates
1d,e were observed to be more reactive toward glycosylation
with 3a or 3b than the corresponding 3- or 5-O-Bn blocked
counterparts 2c,d (entries 5À6, respectively).
1f was compared with per-benzoylated 2e,13c a 1:1.9 mix-
ture of 8a and 8b18 was obtained (entry 8). Judging from
the results of both reactions, it is clear that the 3,5-silylene
acetal function remarkably decreases the donor reactivity,
rendering 1f even less reactive than the fully disarmed
donor 2e. The deactivation effect of an annulated 3,5-
acetal moiety was further demonstrated by the TIPDS-
protected 1g, which was confirmed to exhibit reactivity
that is between moderately armed (2b) and disarmed (2e)
donors (entries 9À10). The different glycosylation behav-
ior of 1f and 1g is probably due to the different torsional
strains induced by the respective 3,5-O-acetal groups in the
sugar rings. Compared with the TIPDS unit, the more
conformationally rigid DTBS group further limits forma-
tion of the corresponding arabinofuranosyl oxacarbenium
ion, thereby resulting in a decline in reactivity for 1f.
Similar observations were made in the L-arabinose series.
L-1B and L-1f were verified by the competition reactions to
be more and slightly less reactive than the corresponding
armed L-2b and disarmed L-2e (Table 1, entries 11À12).
The above silylated thioglycosyl donors may prove useful
for the synthesis of oligoarabinoses. Of particular promise
should be the 3,5-di-TBS/TBDPS (1b/c) and 3,5-DTBS (1f)
protected thioglycosides because of their exceedingly high
or low reactivity which widens the spectrum of furanosyl
donor reactivity. However, their abilities to ensure stereo-
selective construction of 1,2-trans glycosidic linkages via
neighboring group participation are equally important.
According to Scheme 1, eq 1, the coupling between 1b
and the armed 5-OH thioglycoside 11 proceeded smoothly
to furnish R-linked disaccharide thioglycoside 12 as the
only condensation product in good yield. It could undergo
chemoselective activation in the presence of 2a, reacting
with the hindered 1-adamantanol, to yield the correspond-
ing 1,2-trans glycoside 13 in 89% yield (eq 2).
Scheme 1. Selective Activation of 1b
We further sought to test the reactivity of thioglycosides
1f13c and 1g protected as cyclic 3,5-O-di-tert-butylsilylene
(DTBS) or 3,5-O-tetraisopropyldisiloxanylidene (TIPDS)
acetals. In previous work by Crich et al., a slow activation
for 3,5-acetalated p-tolyl thioglycoside and sulfoxide do-
nors was observed.17 Here, completely selective activation
was still retained in the three-component reaction of 1f, 2a,
and 3a. Only disaccharide 5b (R/β = 3/1), the coupling
product of 2a with 3a, was formed (Table 1, entry 7). When
It is now possible to perform one-pot glycosylations that
were previously not possible by the use of these super-
armed/disarmed glycosyl donors. Two such examples are
outlined in Scheme 2. It is worth pointing out that these
reactions were carried out as a true automated one-pot
glycosylation, as in both cases the three glycosylating
agents and the activating reagents are mixed simulta-
neously, and not added sequentially. Thus, on activation
(14) Wang, Z.-W.; Prudhomme, D. R.; Buck, J. R.; Park, M.; Rizzo,
C. J. J. Org. Chem. 2000, 65, 5969.
(15) NIS/TfOH has been screened as the appropriate activator for
discriminating the reactivity levels of the donor substrates. Other
activator systems, such as NIS, NIS/AgOTf, or DMTST, did not give
results as those of NIS/TfOH.
(16) Hiranuma, S.; Kajimoto, T.; Wong, C.-H. Tetrahedron. Lett.
1994, 35, 5257.
(17) Crich, D.; Pedersen, C. M.; Bowers, A. A.; Wink, D. J. J. Org.
(18) Kawabata, Y.; Kaneko, S.; Kusakabe, I.; Gama, Y. Carbohydr.
Res. 1995, 267, 39.
Chem. 2007, 72, 1553.
Org. Lett., Vol. XX, No. XX, XXXX
C

Table 2. Relation between the Chemical Shifts of the Anomeric
Protons of Thioarabinosides and Their Reaction Reactivity
Scheme 2. Automated One-Pot Glycosylation
with NIS (2.5 equiv)/TfOH (0.1 equiv) at À80 °C and
warming to 0 °C in CH2Cl2, triarabinoside 15 was rapidly
assembled within 1.5 h from nonreducing end to reducing
end in 88% yield, based on the reactivity difference of the
three reactants: the superarmed donor 1b, the disarmed
thioglycoside acceptor 14,13c and the 3-OH acceptor 3b
(eq 1). On the other hand, methyl glycoside acceptor 3a
was coupled to disarmed 3-OH thioglycoside acceptor
1613a and then the superdisarmed donor 1f to produce an
excellent 91% yield of trisaccharide 17 together with a
trace amount of tetrasaccharide 18 (eq 2). A trisaccharide
with the same sugar frame as that of 17 has been prepared
in our earlier study through a one-pot glycosylation pro-
tocol which was based on our developed regioselective
glycosylation strategy.13a In that synthesis, a standard
disarmed phenyl 2,3,5-tri-O-acetyl-R-D-thioarabinofura-
noside donor rather than 1f was employed to glycosylate
with 16 and 3a. But the total yield was in lower 70% yield.
We reason that the improved yield of the automated one-
pot procedure is attributable largely to the increased
difference in reactivity between the donors 1f and 16.
Interestingly, during our study of arabinose thioglyco-
side reactivity, we observed a good correlation between the
chemical shifts of anomeric protons and the corresponding
silyl ether protected donor reactivity (Table 2). For exam-
ple, the anomeric H-1 resonance of the most reactive 1a is
significantlyfurtherupfieldthanthatofthe lessreactive2a,
while the H-1 resonance of the latter is more upfield than
that of the least reactive 2e(δH 5.324 vs5.610 vs 5.849 ppm,
respectively). The same correlation is observed as well for
each set of mixed protected thioglycosides.19 As the stabil-
ity of the positively charged oxacarbenium species seems to
be mainly affected by the electron density at the anomeric
center, the chemical shift of anomeric proton might be
useful in explaining the observed reactivity order of these
furanose thioglycosides.
In conclusion, the relative anomeric reactivity of a
variety of silylated arabinose thioglycosides has been
experimentally determined. Silyl ethers on the furanose
ring are found to effectively enhance the reactivity. The
highly reactive 3,5-di-TBS-2-Bz protected thioarabinose is
able to condense with the traditionally armed thioglyco-
side as well as the very unreactive acceptor. A 3,5-cyclic
acetal protection generally has a disarming effect on the
reactivity. These findings offer a wider range of sugar
building blocks of different anomeric reactivity and have
allowed us to realize the first facile automated one-pot
furanosylation. In addition, we observed a good correla-
tion between the reactivity and the chemical shift of the
1
anomeric hydrogen by H NMR spectroscopy. Further
exploration of these novel superarmed/disarmed thiogly-
coside molecules in the synthesis of oligoarabinoses and
extensive investigation into the mechanism of the silyl
protection mediated reactivity tuning are now in progress.
Acknowledgment. The financial support from the NSFC
(21172156, 21021001), the 973 program (2010CB833202),
and the Ministry of Education (NCET-08-0377, 20100181-
110082) is highly appreciated.
Supporting Information Available. Experimental details,
1H and 13C NMR spectra for all new compounds, 2D NMR
spectra for compounds 12, 15, and 17. This material is
available free of charge via the Internet at http://pubs.acs.org.
(19) For a similar correlation between chemical shifts of anomeric
hydrogens and pyranosyl thioglycosides reactivity, see ref 3.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX