α- and β-Glycosyl sulfonium ions: Generation and reactivity

Article abstract of DOI:10.1002/chem.200802293

A study was conducted to observe the generation and reactivity of α- and β-glycosyl sulfonium ions. Preparation of a glycosyl triflate of 2-deoxy-2-amino sugar was done from a thioglycoside using glycosyl triflate pool method. The triflate oxygen was boun

Full text of DOI:10.1002/chem.200802293

DOI: 10.1002/chem.200802293
a- and b-Glycosyl Sulfonium Ions: Generation and Reactivity
Toshiki Nokami,[a] Akito Shibuya,[a] Shino Manabe,*[b] Yukishige Ito,[b] and
Jun-ichi Yoshida*[a]
Long-standing interest in glycosyl onium ions such as am-
monium, imidazoylium, phosphorium, and sulfonium ions in
carbohydrate chemistry[1] has recently been stimulated by
their utilitiy in the control of the stereoselectivity of glycosy-
lation reactions.[2] Schuerch reported the first example of
stereoselective glycosylation reactions using a glucosyl sulfo-
nium ion. A glucopyranosyl dimethylsulfonium bromide was
supposed to be generated by the reaction of the correspond-
ing glucosyl bromide with Me2S, although its characteriza-
tion was not reported.[2a] Boons and co-workers reported the
generation of b-glycosyl sulfonium ion intermediates by
inter- and intramolecular reactions of glycosyl donors with
sulfides.[3] The NMR analyses revealed that only the b-glyco-
syl sulfonium ion was generated, which gave a-glycosides by
the action of glycosyl acceptors. These fascinating findings
prompted us to investigate the generation, structures, and
reactivity of glycosyl sulfonium ions in further detail. We
report herein the results of our studies on a- and b-glycosyl
sulfonium ions.
We began with the preparation of a glycosyl triflate of 2-
deoxy-2-amino sugar from a thioglycoside using our glycosyl
triflate pool method. We chose thioglycoside 1 as a starting
material, because the 2-azido group has already been shown
to function as a protecting group without neighboring group
participation. This is an important property of the 2-azido
group in terms of the ability to examine the stereochemistry
of glycosylations. Electrochemical oxidation of thioglycoside
1 in the presence of Bu4NOTf (0.1m) in CD2Cl2 at À788C
gave a solution of the corresponding glycosyl triflate 2
(Scheme 1).[8] As shown in Figure 1, the anomeric proton
signal, which was observed at d=6.13 ppm (d, J=3.4 Hz),
indicates an a-configuration at the anomeric center.[5a] The
13C NMR signal observed at d=103.6 ppm was assigned to
the anomeric carbon based on HMQC (heteronuclear multi-
ple quantum coherence) analysis. The chemical shift of the
anomeric carbon indicated that the triflate oxygen was
bound covalently to the anomeric carbon.
The present work stems from our earlier finding[4] that
glycosyl triflates[5] were effectively generated by low-temper-
ature electrochemical oxidation[6,7] of thioglycosides (the
glycosyl triflate pool methodology). Because only intact
Bu4NOTf and disulfide are present in the solution besides
glycosyl triflates, we envisioned that the electrochemically
generated glycosyl triflates serve as excellent precursors of
glycosyl sulfonium ions for mechanistic studies by using
spectroscopic methods.
Scheme 1. Electrochemical generation of glycosyl triflate.
[a] Dr. T. Nokami, A. Shibuya, Prof. Dr. J. Yoshida
Department of Synthetic Chemistry and Biological Chemistry
Graduate School of Engineering Kyoto University
Nishikyo-ku, Kyoto 615-8510 (Japan)
Fax : (+81)75-383-2727
E-mail: yoshida@sbchem.kyoto-u.ac.jp
[b] Dr. S. Manabe, Dr. Y. Ito
RIKEN (The Institute of Physical and Chemical Research)
Hirosawa, Wako, Saitama 351-0198 (Japan)
E-mail: smanabe@riken.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802293.
Figure 1. 1H NMR spectrum of the glycosyl triflate 2 (600 MHz, CD2Cl2,
À808C).
2252
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2252 – 2255

COMMUNICATION
The conversion of glycosyl triflate 2 to the glycosyl sulfo-
nium ion was performed by addition of Me2S (5 equiv) to a
solution of 2 at À788C followed by raising the temperature
to 08C (Scheme 2). Signals from two novel species appeared
at the expense of signals of 2 at À608C (Figure 2). The spe-
Scheme 2. Generation of glycosyl sulfonium ions by the action of glycosyl
triflate with dimethyl disulfide.
Figure 3. Cold-spray mass spectra of glycosyl sulfonium ions 3a/3b.
(spray temperature 08C).
tion.[1,9] Therefore, 3a seems to be less stable than 3b. In
fact, molecular orbital calculations (HF/6-31G*) indicated
that 3a is distorted from the chair conformation and
3.2 kcalmolÀ1 less stable than 3b (see the Supporting Infor-
mation). However, in the present case, 3a was produced in
a significant amount (3a/3b 45:55). The a/b ratio did not
change on varying the temperature (À808C to 08C). Based
on these arguments, it seems to be reasonable to consider
that the observed stereoselectivity is based on kinetics in the
present case.
Next, the reactivities of glycosyl sulfonium ions 3a and
3b were investigated (Scheme 3). When MeOH (5 equiv)
Figure 2. 1H NMR spectra of the glycosyl sulfonium ions 3a/3b
(600 MHz, CD2Cl2, 08C). The blue peaks indicate H-1, H-2, and H-4 of
the a-glycosyl sulfonium ion 3a. The red peaks indicate H-1, H-2, and
H-4 of the b-glycosyl sulfonium ion 3b.
cies that exhibited a signal from the anomeric proton at d=
6.27 ppm (d, J=4.8 Hz) was assigned to the a-glycosyl sulfo-
nium ion 3a based on the NOE (nuclear Overhauser effect)
correlation between H-1 and H-2 (3.3%). HMBC (hetero-
nuclear multiple-bond connectivity) analysis and the NOE
correlation between H-1 and the methyl protons of Me2S in-
dicate the presence of a covalent bond between the anome-
ric carbon and sulfur. The larger coupling constant of the
anomeric proton (J=4.8 Hz) and the relatively small cou-
pling constant between H-3 and H-4 (J=6.9 Hz) suggested
that the pyranose ring conformation was distorted. To the
best of our knowledge, this is the first observation of an a-
glycosyl sulfonium ion. The species that exhibited a signal
from the anomeric proton at d=5.43 ppm (d, J=10.3 Hz)
was assigned to the b-glycosyl sulfonium ion 3b.[3a] A cold-
spray mass spectrum of the mixture showed two major
peaks (Figure 3). The first peak was assigned to the glycosyl
cation (m/z calcd for C12H16N3O3: 314.10; found: 314.16),
and the second peak was assigned to the glycosyl sulfonium
ion (m/z calcd for C14H22N3O7S: 376.12; found: 376.19). The
glycosyl cation seemed to be produced by fragmentation of
the glycosyl sulfonium ion in the mass spectrometer, be-
cause it was not observed by NMR spectroscopy.
Scheme 3. Glycosylation of the glycosyl sulfonium ion 3a/3b with
MeOH.
was added to a solution of 3a and 3b at room temperature,
only 3a was consumed after 1 h (Figure 4). At this stage,
methyl glycosides 4a and 4b were produced in the ratio of
40:60.[10] After consumption of 3a, 3b began to react with
MeOH. Eventually, both 3a and 3b were consumed, and
methyl glycosides 4a and 4b were produced in the ratio of
41:59 after 24 h. The present time-course NMR study clearly
shows higher reactivity of 3a than 3b. The stereochemical
outcome cannot be explained by a simple SN2 displacement
at the anomeric carbon atom. Rather, the reaction may pro-
ceed by intermediacy of a glycosyl cation.[11] If we assume
such a mechanism, the higher reactivity of 3a can be ex-
plained by the relative stability of 3a and 3b, because the
energy required for ionization of less stable 3a should be
smaller than that for more stable 3b.
It is also noteworthy that the 4a/4b ratio observed for the
glycosylation of glycosyl sulfonium ions 3a and 3b with
MeOH is different from that observed for the glycosylation
It is known that a cationic substituent such as Me2S+ at
an anomeric center prefers to occupy the equatorial posi-
Chem. Eur. J. 2009, 15, 2252 – 2255
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2253

S. Manabe, J.-i. Yoshida et al.
various temperatures (from À808C to 08C), indicated that 2 was convert-
ed to glycosyl sulfonium ions as a mixture of a- and b-isomers (3a/3b
45:55). The reaction mixture was warmed to room temperature and then
MeOH (2 mL, 0.05 mmol) was added under an argon atmosphere. During
the NMR measurement, which was carried out at room temperature for
1 day, the glycosyl sulfonium ion 3a and 3b was converted to a mixture
of a- and b-isomers of the methyl glycoside (4a/4b 41:59).
(3,4,6-Tri-O-acetyl-2-azido-2-deoxy-a-d-glucopyranosyl)dimethylsulfon-
AHCTUNGTRENNUNG
ium ion (3a): Selected data for 3a (6.5–2.9 ppm for 1H NMR at 08C,
100–55 ppm for 13C NMR at À408C). 1H NMR (CD2Cl2, 600 MHz): d=
6.26 (d, J=4.8 Hz, 1H, H-1), 5.33 (dd, J=8.3, 6.9 Hz, 1H, H-3), 5.04 (dd,
J=7.6, 6.9 Hz, 1H, H-4), 4.62 (dd, J=7.6, 4.8 Hz, 1H, H-2), 4.24–4.21
(m, 1H, H-6), 4.16 (dd, J=12.4, 2.0 Hz, 1H, H-6’), 4.13–4.11 (m, 1H, H-
5), 3.16 (s, 3H, SCH3), 2.94 ppm (s, 3H, SCH3); 13C NMR (CD2Cl2,
150 MHz): d=90.0 (C-1), 73.9 (C-5), 70.5 (C-3), 66.3 (C-4), 61.6 (C-6),
58.5 ppm (C-2).
(3,4,6-Tri-O-acetyl-2-azido-2-deoxy-b-d-glucopyranosyl) dimethylsulfon-
AHCTUNGTRENNUNG
ium ion (3b): Selected data for 3b (6.5–2.9 ppm for 1H NMR at 08C,
100–55 ppm for 13C NMR at À408C). 1H NMR (CD2Cl2, 600 MHz): d=
5.42 (d, J=10.3 Hz, 1H, H-1), 5.38 (pseudo t, J=9.6 Hz, 1H, H-3), 5.12–
5.09 (m, 1H, H-4), 4.24–4.21 (m, 3H), 4.02 (dd, J=10.3, 9.6 Hz, 1H, H-
2), 3.10 (s, 3H, SCH3), 2.96 ppm (s, 3H, SCH3); 13C NMR (CD2Cl2,
150 MHz): d=81.4 (C-1), 76.7 (C-5), 73.9 (C-3), 66.5 (C-4), 60.7 (C-6),
58.6 ppm (C-2); LRMS (CS): m/z: calcd for C14H22N3O7S [M]+, 376.12;
found, 376.19. Further details are given in the Supporting Information.
Figure 4. The time-course 1H NMR spectra of the reaction between the
glycosyl sulfonium ions 3a/3b and MeOH (600 MHz, CD2Cl2, room tem-
perature). The blue peaks indicate H-1, H-2, and H-4 of the a-glycosyl
sulfonium ion 3a. The red peaks indicate H-1, H-2, and H-4 of the b-gly-
cosyl sulfonium ion 3b.
of glycosyl triflate 2 (4a/4b 13:87).[8] On the other hand, the
reaction may proceed by a contact ion pair (CIP) mechanis-
m[11a] or an SN2 like mechanism in the case of the glycosyla-
tion of glycosyl triflate 2, although more data should be ac-
cumulated before the elucidation of a detailed mechanism.
In conclusion, both a- and b-glycosyl sulfonium ions were
successfully produced from an electrochemically generated
glycosyl triflate and were characterized by NMR spectrosco-
py and mass spectrometry. The time-course NMR study for
the reaction with MeOH clearly revealed that the a-glycosyl
sulfonium ion is more reactive than the b-glycosyl sulfonium
ion. The stereochemical outcome indicates that the reaction
does not proceed by a simple SN2 mechanism. Rather, the
reaction seems to proceed via an glycosyl cation intermedi-
ate. We believe that the present observations will make a
significant contribution to mechanistic studies on glycosyla-
tion reactions. Further work aimed at the elucidation of the
stereochemistry and reactivity of various glycosyl sulfonium
ions is currently in progress.
Acknowledgements
This research was partially supported by Grants-in Aid for Scientific Re-
search (Grant No. 19590032, 17205012, and 20245008) from the JSPS, In-
centive Research Grant from RIKEN (S.M.), and a Grant-in-Aid for the
Global COE Program, from the MEXT, Japan. T.N. thanks Meiji Seika
Kaisha, Ltd. for financial support. S.M. thanks Ms. Akemi Takahashi for
her technical assistance.
Keywords: carbohydrates
·
glycosylation
·
NMR
spectroscopy · oxidation · stereochemistry
[1] a) P. Finch, A. G. Nagpurkar, Carbohydr. Res. 1976, 49, 275–287;
b) A. R. Vaino, S. S. C. Chan, W. A. Szarek, G. R. J. Thatcher, J.
Org. Chem. 1996, 61, 4514–4515; c) P. G. Jones, A. J. Kirby, I. V. Ko-
marov, P. D. Wothers, Chem. Commun. 1998, 1695–1696; d) R. J.
Batchelor, D. F. Green, B. D. Johnston, B. O. Patrick, B. M. Pinto,
Carbohydr. Res. 2001, 330, 421–426; e) E. Skorupowa, M. Kurszew-
ska, A. Konitz, W. Wojnowki, A. Wisniewki, Carbohydr. Res. 2001,
331, 343–346.
[2] a) A. C. West, C. Schuerch, J. Am. Chem. Soc. 1973, 95, 1333–1335;
b) W. G. Dauben, P. Kçhler, Carbohydr. Res. 1990, 203, 47–56.
[3] a) J.-H. Kim, H. Yang, J. Park, G.-J. Boons, J. Am. Chem. Soc. 2005,
127, 12090–12097; b) J. Park, S. Kawatkar, J.-H. Kim, G.-J. Boons,
Org. Lett. 2007, 9, 1959–1962.
[4] T. Nokami, A. Shibuya, H. Tsuyama, S. Suga, A. A. Bowers, D.
Crich, J. Yoshida, J. Am. Chem. Soc. 2007, 129, 10922–10928.
[5] Glycosyl triflate observation; a) D. Crich, S. Sun, J. Am. Chem. Soc.
1997, 119, 11217–11223; b) D. Crich, J. Carbohydr. Chem. 2002, 21,
667–690; c) S. Yamago, T. Yamada, T. Maruyama, J. Yoshida,
Angew. Chem. 2004, 116, 2197–2200; Angew. Chem. Int. Ed. 2004,
43, 2145–2148; d) P. Wei, R. J. Kerns, J. Org. Chem. 2005, 70, 4195–
4198; e) A. Rencurosi, L. Lay, G. Russo, E. Caneva, L. Poletti, Car-
bohydr. Res. 2006, 341, 903–908; f) J.-Y. Baek, T.-J. Choi, H.-B.
Jeon, K.-S. Kim, Angew. Chem. 2006, 118, 7596–7600; Angew.
Chem. Int. Ed. 2006, 45, 7436–7440.
Experimental Section
The anodic oxidation was carried out in an H-type divided cell (4G glass
filter) equipped with a carbon felt anode and a platinum plate cathode.
In the anodic chamber were placed the thioglycoside
1 (43.4 mg,
0.0992 mmol) and 0.1m Bu4NOTf in CD2Cl2 (5.0 mL). In the cathodic
chamber were placed trifluoromethanesulfonic acid (22 mL, 0.25 mmol)
and 0.1m Bu4NOTf in CD2Cl2 (5.0 mL). The constant current electrolysis
(4.0 mA) was carried out at À788C with magnetic stirring. After
1.5 FmolÀ1 of electricity was consumed, the reaction mixture in the
anodic chamber was transferred to a 5 mm NMR tube with a septum cap
under an argon atmosphere at À788C. After the NMR measurement of
triflate 2, dimethyl sulfide (5 mL, 0.05 mmol) was added under an argon
atmosphere at À788C. The NMR measurement, which was carried out at
2254
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2252 – 2255

a- and b-Glycosyl Sulfonium Ions
COMMUNICATION
[6] Electrochemical oxidations: a) K. D. Moeller, Tetrahedron 2000, 56,
9527–9554; b) J. Yoshida, K. Kataoka, R. Horcajada, A. Nagaki,
Chem. Rev. 2008, 108, 2265–2299.
[9] a) J.-P. Praly, R. U. Lemieux, Can. J. Chem. 1987, 65, 213–223;
b) C. L. Perrin, Tetrahedron 1995, 51, 11901–11935; c) C. L. Perrin,
M. A. Fabian, J. Brunckova, B. K. Ohta, J. Am. Chem. Soc. 1999,
121, 6911–6918; d) K. D. Randell, B. D. Johnson, D. F. Green, B. M.
Pinto, J. Org. Chem. 2000, 65, 220–226, and references therein.
[10] There is no equilibrium between methyl glycoside 3a and 3b under
these conditions.
[11] a) D. Crich, N. S. Chandrasekera, Angew. Chem. 2004, 116, 5500–
5503; Angew. Chem. Int. Ed. 2004, 43, 5386–5389; b) M. H. El-
Badri, D. Willenbring, D. Tantillo, J. Gervay-Hague, J. Org. Chem.
2007, 72, 4663–4672; c) M. G. Beaver, S. B. Billings, K. A. Woerpel,
J. Am. Chem. Soc. 2008, 130, 2082–2086.
[7] Electrochemical glycosylations: a) R. Noyori, I. Kurimoto, J. Org.
Chem. 1986, 51, 4320–4322; b) C. Amatore, A. Jutand, J.-M. Mallet,
G. Meyer, P. Sinaꢁ, J. Chem. Soc. Chem. Commun. 1990, 718–719;
c) G. Balavoine, A. Gref, J.-C. Fischer, A. Lubineau, Tetrahedron
Lett. 1990, 31, 5761–5764; d) S. Yamago, K. Kokubo, J. Yoshida,
Chem. Lett. 1997, 111–112; e) S. Suzuki, K. Matsumoto, K. Kawa-
mura, S. Suga, J. Yoshida, Org. Lett. 2004, 6, 3755–3758; f) R. R.
France, N. V. Rees, J. D. Wadhawan, A. J. Fairbanks, R. G. Compton,
Org. Biomol. Chem. 2004, 2, 2188–2194; g) K. Mitsudo, T. Kawagu-
chi, S. Miyahara, W. Matsuda, M. Kuroboshi, H. Tanaka, Org. Lett.
2005, 7, 4649–4652; h) N. Tanaka, F. Ohnishi, D. Uchihara, S. Torii,
J. Nokami, Tetrahedron Lett. 2007, 48, 7383–7387.
[8] After addition of MeOH (5 equiv) to a solution of glycosyl triflate
2, the corresponding methyl glycoside 4 was obtained in 88% yield
(4a/4b 13:87).
Received: November 5, 2008
Revised: December 2, 2008
Published online: January 20, 2009
Chem. Eur. J. 2009, 15, 2252 – 2255
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2255