Conversion of β-glycopyranoside to α-glycopyranoside by photo-activated radical reaction

Article abstract of DOI:10.1016/j.tetlet.2016.04.064

By using carbon tetrachloride as the chloride radical and boron trifluoride etherate as the Lewis acid, the halogen-light-activated anomeric inversion of glycoside was achieved. This reaction is a novel guide to invert the glycosidic bond from a β-anomer to an α-anomer.

Full text of DOI:10.1016/j.tetlet.2016.04.064

Tetrahedron Letters 57 (2016) 2474–2477
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Conversion of b-glycopyranoside to a-glycopyranoside
by photo-activated radical reaction
⇑
Yu-Chen Lai, Chin-Hung Luo, Hsin-Chun Chou, Cheng-Jhang Yang, Le Lu, Chien-Sheng Chen
Department of Chemistry, Fu-Jen Catholic University, New Taipei City 24205, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history:
By using carbon tetrachloride as the chloride radical and boron trifluoride etherate as the Lewis acid, the
halogen-light-activated anomeric inversion of glycoside was achieved. This reaction is a novel guide to
invert the glycosidic bond from a b-anomer to an a-anomer.
Received 17 February 2016
Revised 14 April 2016
Accepted 19 April 2016
Available online 20 April 2016
Ó 2016 Elsevier Ltd. All rights reserved.
Keywords:
Radical reaction
Quasi-anomeric effect
Anomeric inversion
Halogen
Introduction
carbon is covalently bound to two alkoxyl oxygens, resulting in a
more electron-withdrawing bonding structure such that the
anomeric proton is slightly more acidic than the others. A suitable
radical can be expected to absorb the anomeric proton, yielding a
Radical intermediates play an important role in modern syn-
thetic chemistry, and the reaction of the anomeric radical of carbo-
hydrates has been extensively studied.1,2 In general, radical Barton
decarboxylation of 1-carboxylated glycosides is effective in con-
structing b-glycosidic bonds,3–9 via a mechanism in which abstrac-
tion of the axial hydrogen is eight times faster than that of the
corresponding equatorial hydrogen.10 Giese and co-workers11–14
significantly contributed to the development of an intermolecular
p-radical and enrichment of the a-anomer. The stereoselective b-
glycosidation of a sugar residue can be controlled by the neighbor-
ing group participation of 2-O-acetyl or by the O-benzoyl protect-
ing group,22 whereas in the alternative
a-glycoside route synthesis
involves direct inversion of the b-anomer. Here, we report the
photo-activated anomeric inversion of protected hexose using car-
bon tetrachloride or N-chlorosuccinimide (NCS) as the chlorine
radical source.
radical C-glycosylation reaction with
a-selectivity and reflecting
a quasi-anomeric effect.15,16 This stereoelectronic effect on the
anomeric radical was shown to be due to the periplanar arrange-
ment, with the nonbonding electrons on the ring oxygen and the
anti-bonding orbital on the glycosidic 2-substituent.17,18 In addi-
tion, radical hydrogen-atom transfer reactions for oligosaccharides
have been developed to modify cyclodextrin19 and to photode-
Results and discussion
As shown in Table 1, the radical reaction can be activated with
halogen light and BF3ÁOEt2, using NCS as the chloride radical
grade the target
Study with pyranosyl radical preparation is the conversion of
the protected
-glycopyranosyl bromide using AIBN/Bu3SnH,21
which eliminates the protected alkoxyl group attached to the pyra-
nose. To overcome this problem, we developed a method of radical
generation from a protected carbohydrate that does not involve
dismantling the hydroxyl group and that allows for anomeric
inversion. In contrast to the pyranosyl carbons, the anomeric
D
-galactofuranose residue20 selectively.
source. The anomeric conversion of b-D-galactopyranosyl acetate
1b into
an /b ratio of 5.7/1 (Table 1, entry 1). NMR analysis showed that
the /b ratio was almost identical when -galactopyranosyl acet-
ate 1 and b- -glucopyranosyl acetate 2b were tested (Table 1,
entries 2 and 3). The /b conversion ratio of methyl glucopyra-
a-major product resulted in a good yield within 1 h, with
D
a
a
a-D
a
D
a
noside 3b was 2.9/1, corresponding to a yield of 70%, after 1 h of
activation, and remained nearly the same in the presence of excess
NCS or with a longer reaction time. Anomeric conversions of other
glucopyranosides (Table 1, entries 4, 6–8) with NCS were not
observed; negative results were also obtained with the bromine
radical reagent N-bromosuccinimide (NBS).
⇑
Corresponding author. Tel.: +886 2 29052475; fax: +886 2 29023209.
E-mail address: 092847@mail.fju.edu.tw (C.-S. Chen).
http://dx.doi.org/10.1016/j.tetlet.2016.04.064
0040-4039/Ó 2016 Elsevier Ltd. All rights reserved.

Y.-C. Lai et al. / Tetrahedron Letters 57 (2016) 2474–2477
2475
Table 1
Photo-activated glycosidic inversion of hexoses to the corresponding
a
-glycosides with N-chlorosuccinimide26
50w hv, NCS
O
O
OR
(PO)
(PO)
BF3.OEt2, Ac2O
CH3CN
OR
P or R = Ac or Me
Entry
Hexose
Time (h)
Yield (%)
a
/ba
1
2
3
4
5
6
7
8
1b
1
1
1
1
1
1
1
1
88
99
5.7/1
6.3/1
6.0/1
—
2.9/1
—
1
a
2b
76
2
a
NRb
70
3b
3
a
NRb
NRb
NRb
4b
—
—
4
a
a
NMR result, see Supporting information Figure S1.
No reaction.
b
AcO
AcO
OAc
O
AcO
OAc
OAc
O
OAc
O
O
AcO
AcO
AcO
AcO
OAc
AcO
OAc
AcO
AcO
AcO
AcO
OAc
OAc
1
1α
2
β
β
2α
OMe
O
OMe
O
OAc
O
OAc
O
MeO
MeO
MeO
MeO
OMe
AcO
AcO
AcO
AcO
OMe
MeO
MeO
AcO
AcO
OMe
OMe
3β
3α
4β
4α
Table 2
a. radical initiation
Photo-activated glycosidic inversion of b-galactose to the corresponding
topyranose with CCl4
a-galac-
27
hν
Cl
CCl3
+
Cl4C
BF3 OEt2
AcO
AcO
OAc
O
AcO
AcO
OAc
O
50w hv
b. anomeric inversion
CCl4, BF3.OEt2
OAc
AcO
AcO
AcO
OAc
O
OAc
O
CH3CN
AcO
AcO
1β
HCl
Cl
OAc
1α
AcO
OAc
OAc
AcO
1β
I
Entry
BF3ÁOEt2 (equiv)
CCl4 (equiv)
Time (min)
Yield (%)
a
/bd
AcO
1
2
3
4
5
6
1.1
1.1
1.1
0.55
0.55
0.55
1.1
1.1
1.1
1.2
1.2
1.2
1.2
0.5
0.1
1.2
1.2
1.2
1
2
86
91
66
59
51
77
5.9/1
6.1/1
6.2/1
3.9/1
2.1/1
1/2
5.7/1
—
—
20
20
20
20
20
30
30
AcO
AcO
AcO
OAc
OAc
O
Cl
O
AcO
II
7a
8b
9c
69
OAc
HCl
AcO
NRe
NRe
OAc
1
α
AcO
c. radical propagation
HCl CCl3
a
b
c
Without CCl4, with CBr4.
Without 50 W of light.
With 1 equiv DMSO.
NMR result, see Supporting information Figure S2.
No reaction.
+
CHCl3
Cl
+
d
e
Scheme 1. Postulated photo-activated radical mechanism for anomeric inversion.
The isolated yield was 66% after 20 min, during which pyranosyl
acetate underwent anomeric hydrolysis. A longer reaction time
Table 1 also shows that b-D-galactopyranosyl acetate 1b was a
more active monosaccharide; it was therefore chosen for further
study. The acidic conditions were similar and carbon tetrachloride
was used as the chlorine radical source. To optimize the conversion
conditions, CCl4 (1.2 equiv) and BF3ÁOEt2 (1.1 equiv) were added to
a solution of hexose 1b in Pyrex glass (Table 2). The anomeric
provided a nearly identical
a/b ratio of 6.2/1, and the nature of
the quasi-anomeric effect seemed to influence the selectivity of
the reaction. Decreasing the quantity of BF3ÁOEt2 from 1.1 to
0.55 equiv led to an a/b ratio of 3.9/1, while following a reduction
of the amount of CCl4 from 1.1 to 0.5 or 0.1 equiv the
2.1/1 and 1/2, respectively. With CBr4 (Table 2, entry 7) treatment,
the conversion of b-anomer 1b into a mixture of 1 and 1b pro-
vided the same results as obtained with CCl4. The presence of
a/b ratio was
inversion started immediately under halogen light. The
was 5.9/1 within 1 min, with an 86% yield. Similarly, within
2 min the /b ratio and yield were 6.1/1 and 91%, respectively.
a/b ratio
a
a

2476
Y.-C. Lai et al. / Tetrahedron Letters 57 (2016) 2474–2477
Table 3
27
Photo-activated glycosidic inversion of hexoses to the corresponding
a
-glycosides with CCl4
O
O
50w hv
OR
(PO)
(PO)
CCl4, BF3. OEt2
P or R = Ac or Me
CH3CN
OR
Entry
Hexose
Time (min)
Yield (%)
a
/bb
1
2
3
4
5
6
7
8
1
1
2
2
2b
2b
3b
3b
a
a
a
a
2
30
2
30
2
30
2
30
2
30
30
30
15
30
30
15
30
15
30
30
30
30
94
95
98
94
96
60
62
58
6.4/1
6.2/1
12.5/1
7.1/1
1/1.1
7.3/1
3.1/1
2.2/1
3.2/1
2.3/1
—
9
3
3
a
a
58
10
11
12
13a
14a
15a
16
17
18
19
20
21a
22a
57
4b
NRc
NRc
70
4a
—
4b
4b
1/2.2
1/2.2
—
1/2.1
4.9/1
12.5/1
2.3/1
—
70
4a
NRc
70
5b
5b
60
77
5
5
a
a
61
6b
6b
NRc
90
1/3.2
—
6a
NRc
a
b
c
75 W hm.
NMR result, see Supporting information Figure S3.
No reaction.
OBz
O
OBn
O
OBn
O
OBz
O
BzO
BnO
BnO
BnO
OMe
BzO
BzO
OMe
BzO
BnO
BzO
OBn
BnO
OBz
OMe
OMe
5α
6α
5β
6β
DMSO or performing the reaction in the dark prevented the radi-
cal-mediated inversion, as CCl4 forms a radical complex with
DMSO23–25 such that the activity of the chlorine radical is
restricted.
A postulated mechanism is illustrated in Scheme 1. A chlorine
radical initially generated and activated from CCl4 under halogen
light reacts with anomeric proton to yield the corresponding galac-
the same conditions to pyranosyl derivatives 1–6. The results
are presented in Table 3. As shown in entries 1–4 of the table,
a-galactose 1a and a-glucose 2a provided the corresponding a/
b anomers in ratios of 6.4/1 and 12.5/1 in 2 min and in ratios
of 6.2/1 and 7.1/1 in 30 min, with good yields. b-Glucose acetate
2b afforded anomeric ratios of 1/1 and 7.3/1 in 2 min and 30 min
(Table 3, entries 5 and 6) with 95% and 60% yields, respectively.
As shown in Table 2, the hydrolytic by-product was observed
with a longer reaction time. For the O-methyl glucose methoxy-
topyranos-1-yl p-radicals I and II with a boat 2B5 transition struc-
ture. Owing to a quasi-anomeric effect, the lone electron pair of the
pyranosyl ring-oxygen and the antibonding orbital r⁄ of the 2-ace-
toxyl group can stabilize the boat structure II, such that it can
accept a hydrogen radical from the Si-face to predominately pro-
lates 3b and 3a (entries 7–10), the a/b ratios obtained with the
anomeric conversions were 3.1/1 and 3.2/1 in 2 min, and 2/1 in
30 min. For O-benzyl derivatives 5 (entries 16–19), the resulting
duce the
a
-anomer. The presence of boron trifluoride extends the
a/b ratios were 4.9/1 for 5b and 2.3/1 for 5a in 30 min. With
half-life of the chlorine radical, allowing it to participate in subse-
quent radical reactions. The trichloromethyl radical likely causes
greater steric hindrance than chloride. Indirect evidence (see the
Fig. S4, Supporting information) comes from an examination of
the proton and carbon nuclear signals of chloroform, which sug-
gests that the trichloromethyl radical accepts a proton from a pro-
tic source. By contrast, a series of photo-activated reactions carried
out using AIBN over 4 h generated the isobutyronitrile radical and
only the inverted glucopyranosyl hexoses 2b and 3b (see Table S1,
Supporting information). With a larger steric axial acetoxyl group
at the 4-position, the anomeric inversion of b-galactopyranosyl
acetate 1b was not activated by the radical in the presence of
AIBN/BF3ÁOEt2.
methyl or benzyl protection groups, the radical inversion of reac-
tants 3 and 5 showed lower yields (ꢀ60%). The presence of the
involuted hydrolytic by-product indicated that photo-activated
chlorine and trichloromethyl radicals cleave ether-linked protect-
ing groups. The reaction was hindered when tetra-O-acetyl-glu-
copyranosyl methoxylates 4b and 4a (entries 11 and 12) and
tetra-O-benzoyl-glucopyranosylmethoxylates 6b (entry 20) were
used. To attain conversion, brighter (75 W) halogen light irradia-
tion was applied, which allowed the inversion of hexose
methoxylates 4b and 6b with
a/b ratios of 1/2.2 and 1/3.2
(entries 14 and 21). This result demonstrated that a stronger
halogen light source/irradiation would generate more chlorine
radicals and thereby enhance the reactivity of anomeric inver-
sion, while the results obtained using bulky and aromatic benzoyl
After the successful photo-activated anomeric inversion of b-
galactopyranose 1b using CCl4 as the radical source, we applied
groups and a smaller acetyl group were the same. Anomers 4
a

Y.-C. Lai et al. / Tetrahedron Letters 57 (2016) 2474–2477
2477
2. Binkley, E. R.; Binkley, R. W. Carbohydrate Photochemistry; American Chemical
Society: Washington, DC, 1998.
and 6b were not reactive and the starting material was recovered
after 30 min (entries 15 and 22).
3. Crich, D.; Hermann, F. Tetrahedron Lett. 1993, 34, 3385.
4. Crich, D.; Hwang, J.-T.; Yuan, H. J. Org. Chem. 1996, 61, 6189.
5. Crich, D.; Lim, L. B. J. J. Chem. Soc., Perkin Trans. 1 1991, 2209.
6. Crich, D.; Lim, L. B. J. J. Chem. Soc., Perkin Trans. 1 1991, 2205.
7. Crich, D.; Ritchie, T. J. Chem. Soc., Chem. Commun. 1988, 1461.
8. Crich, D.; Ritchie, T. J. Chem. Soc., Perkin Trans. 1 1990, 945.
9. Crich, D.; Sun, S.; Brunckova, J. J. Org. Chem. 1996, 61, 605.
10. Crich, D. J. Org. Chem. 2011, 76, 9193.
11. Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry of Radical Reactions; VCH
Press: Weinheim, 1996.
12. Giese, B.; Dupes, J. Tetrahedron Lett. 1984, 25, 1349.
13. Giese, B. D. J.; Leising, M.; Nix, M.; Lindner, H. J. Carbohydr. Res. 1987, 171, 329.
14. Giese, B. Angew. Chem., Int. Ed. 1989, 28, 969.
15. Sustmann, R.; Korth, H.-G. J. Chem. Soc., Faraday Trans. 1 1987, 83, 95.
16. Héberger, K.; Lopata, A. J. Chem. Soc., Perkin Trans. 2 1995, 91.
17. Abe, H.; Terauchi, M.; Matsuda, A.; Shuto, S. J. Org. Chem. 2003, 68, 7439.
18. Rychnovsky, S. D.; Powers, J. P.; LePage, T. J. J. Am. Chem. Soc. 1992, 114, 8375.
19. Alvarez-Dorta, D.; Leon, E. I.; Kennedy, A. R.; Martin, A.; Perez-Martin, I.;
Suarez, E. Angew. Chem., In.t Ed. 2015, 54, 3674.
20. Takahashi, D.; Hirono, S.; Hayashi, C.; Igarashi, M.; Nishimura, Y.; Toshima, K.
Angew. Chem., Int. Ed. 2010, 49, 10096.
21. Korth, H.-G.; Sustmann, R.; Dupuis, J.; Giese, B. J. Chem. Soc., Perkin Trans. 2
1986, 1453.
Conclusion
In summary, a chlorine radical was activated from photo-acti-
vated CCl4 using BF3ÁOEt2 as the Lewis acid and then used to
remove anomeric hydrogen and induce anomeric inversion. The
nature of the quasi-anomeric effect could be explained by the
greater stability of the pyranos-1-yl b-radical intermediate than
the
vated condition was also examined using b-cellobiose acetate,
but a significant inversion signal for the -anomer was not
a-radical, such that the a-product predominated. This acti-
a
observed. This chlorine radical reaction has numerous practical
applications, and further studies with other fluorogen or halogens
to achieve anomeric inversion from one b-anomer to the other are
warranted. Such studies are currently being carried out in our lab-
oratory and their results will be reported in due course.
22. Gervay-Hague, J. Acc. Chem. Res. 2016, 49, 35.
Acknowledgments
23. Sumiyoshi, T.; Katayama, M. Bull. Chem. Soc. Jpn. 1990, 63, 1293.
24. Sumiyoshi, T.; Minegishi, H.; Fujiyoshi, R.; Sawamura, S. Rad. Phys. Chem. 2006,
75, 195.
25. Sumiyoshi, T.; Watanabe, K.; Syogen, S.; Kawasaki, M.; Katayama, M. Bull.
Chem. Soc. Jpn. 1990, 63, 1584.
This work was supported by Ministry of Science and Technology
and Fu-Jen University, Taiwan (ROC).
26. To a solid of per-O-Ac hexose (0.27 mmol) and NCS (3.5 mg, 0.027 mmol) in
Pyrex tube was added CH3CN (1 mL), the mixture was added BF3ÁOEt2 complex
Supplementary data
(37 lL, 0.29 mmol) at 0 °C. The mixture was degassed under vacuum and
refilled with Argon gas for three times, and the reaction was stirred and
irradiated with halogen light (50 W, 110 V, Philips) at room temperature. The
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2016.04.
064.
progress of the reaction was monitored with NMR by removing a 20-
subsample and mixing it with CDCl3 (450 L). The reaction was quenched with
NEt3 (100 L) and cyclohexene (100 L). The crude mixture was purified by
lL
l
l
l
flash column chromatography (EA/Hex = 1/2) to afford the product.
27. To a solid of hexose (0.27 mmol) in Pyrex tube was added CH3CN (0.8 mL) and
References and notes
a solution of CCl4 (37 lL, 0.29 mmol) in CH3CN (0.2 mL), the mixture was
added BF3ÁOEt2 complex (37 lL, 0.29 mmol) at 0 °C. Following procedure could
1. Binkley, R. W.; Binkley, E. R. Radical Reaction of Carbohydrate; http://
refer to Ref. 26.
www.carborad.com/default.html 2013.