Regioselective glycosylation reactions based on computational predictions

Article abstract of DOI:10.1016/j.tetlet.2010.01.044

Regioselectivity is a major issue in glycosylation reactions. Better understanding of regioselectivity of acceptors can greatly facilitate and simplify the syntheses of oligosaccharides. The reactivity of diol acceptors is often affected by stereochemistry and protecting groups, which make prediction the regioselectivity of diol acceptors extremely difficult. Quantum mechanic methods were used to study the relationship between protecting groups and reactivity of diol acceptor and a correlation between Fukui function and regioselectivity is established through series of glycosylation reactions.

Full text of DOI:10.1016/j.tetlet.2010.01.044

Tetrahedron Letters 51 (2010) 1550–1553
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Regioselective glycosylation reactions based on computational predictions
*
Jane Kalikanda, Zhitao Li
Department of Chemistry, State University of New York, PO Box 6000, Binghamton, NY 13902, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Regioselectivity is a major issue in glycosylation reactions. Better understanding of regioselectivity of
acceptors can greatly facilitate and simplify the syntheses of oligosaccharides. The reactivity of diol
acceptors is often affected by stereochemistry and protecting groups, which make prediction the regiose-
lectivity of diol acceptors extremely difficult. Quantum mechanic methods were used to study the rela-
tionship between protecting groups and reactivity of diol acceptor and a correlation between Fukui
function and regioselectivity is established through series of glycosylation reactions.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 10 December 2009
Revised 8 January 2010
Accepted 12 January 2010
Available online 18 January 2010
Keywords:
Regioselectivity
Glycosylation
Fukui function
Computational chemistry
Oligosaccharides, the third major class of biopolymers after
peptides and oligonucleotides, play very important roles in various
biological processes, such as cell adhesion, cell recognition, immu-
nization and so on.1–4 The synthesis of oligosaccharides, however,
is far more difficult than other biopolymers, such as peptides and
oligonucleotides due to the high regio- and stereoselectivity re-
quired in the formation of glycosidic linkages. What makes the
chemical synthesis of oligosaccharides even more difficult is the
low predictability of glycosylation reactions, where little changes
in the structure of glycosyl monomers (such as protecting groups
or stereochemistry) can often cause dramatic change in reaction
results.5–7 Prediction of glycosylation reaction and rational design
of oligosaccharide synthesis are therefore of particular interest.8,9
Here we report a successful prediction of the reactivity of a series
of 2,3-diol mannosyl acceptors using computational chemistry
methods.
such as change of protecting groups. Better understanding of the
relative reactivity of hydroxyl groups in diol or triol acceptors,
especially the relationship between protecting groups and reactiv-
ity, is therefore very important for prediction of glycosylation reac-
tions and rational design of oligosaccharide synthesis.
Changing protection groups can introduce both steric and elec-
tronic effects to the acceptors, whereas electronic effects are what
we cannot predict based on our current knowledge. Since compu-
tational chemistry is good at explaining and predicting properties
related to electron distribution in organic molecules, we feel appli-
cation of computational chemistry in carbohydrate chemistry
could be a good way to help us understand the influence of protect-
ing groups on acceptor reactivity. Fukui function, first introduced
by Parr and Yang, is defined as the differential change in electron
density due to an infinitesimal change in the number of elec-
trons.11–13 It can be expressed in a condensed form as:
Regioselectivity is a major issue in carbohydrate synthesis. Bet-
ter understanding and control of regioselectivity can not only facil-
itate regioselective protection of acceptors, but may also realize
regioselective glycosylations without or with only minimum pro-
tections. Currently, reactivity of acceptor hydroxyls can only be
predicted based on a few empirical rules, such as: primary hydro-
xyl groups are more reactive than secondary hydroxyl groups and
equatorial hydroxyl groups are more reactive than axial hydroxyl
groups.10 However, these empirical rules have limitations. First, it
is difficult to predict reactivity when both hydroxyl groups are
equatorial or axial. Second, it cannot be used to estimate the differ-
ence between acceptors with only subtle difference in structures,
fkꢀ ¼ qkðNÞ ꢀ qkðN ꢀ 1Þ
fkþ ¼ qkðN þ 1Þ ꢀ qkðNÞ
where qkðNÞ is Mulliken’s charge on the k-atom of the molecule
with N electrons and qkðN þ 1Þ, qkðN ꢀ 1Þ are the charges on the
k-atom of the molecule with (N + 1) and (N ꢀ 1) electrons in a fro-
zen orbital approximation, respectively.
Fukui function is the DFT analogue of the frontier orbital regi-
oselectivity for nucleophilic (f+) and electrophilic (fꢀ) attack.14–16
Fukui function has been successfully used in explaining and pre-
dicting regioselectivity in organic reactions, especially reactions
involving nucleophiles and electrophiles, such as dipole addition
reactions and Michael reaction.16–18 Since glycosylation of glycosyl
acceptors can be considered as an electrophilic attack of the OH
ð1Þ
ð2Þ
* Corresponding author. Tel.: +1 607 777 4825.
E-mail address: zli@binghamton.edu (Z. Li).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.01.044

J. Kalikanda, Z. Li / Tetrahedron Letters 51 (2010) 1550–1553
1551
f -
=
=
0.015
0.007
f -M
f-
=
=
0.050
0.040
BnO
BnO
OH
f-M
MeO
OH
O
N
O
N
MeO
HO
HO
OMe
3
2
OMe
f -
=
=
0.078
0.077
OH
f -
=
=
0.042
0.039
RO
RO
f -M
f -M
O
N
N
HO
f -
=
=
0.078
0.084
f -
=
=
0.025
0.021
OMe
f -M
f -M
AcO
OH
OH
1
Ph
O
N
O
O
N
O
AcO
HO
HO
4
OMe
OMe
5
f -
=
=
0.052
0.049
f -
=
=
0.070
0.079
f -M
f -M
N
N
Figure 1. Mannose 2,3-diol acceptors.
OH
OH
O
OAc
RO
RO
RO
O
O
pyridine
RO
+
RO
RO
+
Ac2O
HO
AcO
HO
7
6
OMe
OMe
OMe
R
ratio 6 : 7
1:6
R = Bn
R = Ac
2:3
R = -CHPh-
7 only
Figure 2. Acetylation of diol acceptors.
group by an oxocarbenium ion intermediate, fꢀ could be a good
indicator for the regioselectivity of hydroxyl groups in diol glycosyl
acceptors.
not in compound 4. From the steric point of view, 3-OH (equatorial)
should be more reactive than 2-OH (axial), but this is not the case in
compound 4. This result is probably due to the fact that the electro-
phile used in the experiment is relatively small and less sterically
demanding. Under this circumstance, the Fukui function fꢀ is an
effective indicator to predict the correlation between the protecting
groups and the relative reactivity.
Encouraged by the acetylation reaction results, we further tested
theregioselectivityof thesediol acceptorsin glycosylation reactions.
We first tested mannose donors in glycosylation, because the disac-
charides obtained from these reactions are useful intermediates for
the syntheses of the core structure of N-glycan and some other types
of oligosaccharides.23 Three tetraacetylmannosyl donors (8a, 8b, 8c)
were tested in glycosylation reactions (Fig. 3). Similar to acetylation
reactions, acceptor 3 and 5 showed great regioselectivity. Only
1,3-disaccharides 10were isolated in reactions betweenall three do-
nors and compound 3 and 5, no 1,2-disaccharides 9 were observed
(Table 1). The difference between compound 3 and 5 is that
compound 3 gave a small amount of trisaccharide products in reac-
tions with bromide donor and trichloroacetimmidate donor. In case
of acceptor 4, both 1,2- and 1,3-disaccharides were isolated in the
reactions, together with a small amount of trisaccharide.
The main difference between glycosylation and acetylation of the
diol acceptors is the regioselectivity of compound 4, which showed
no selectivity in acetylation, whereas a substantial selectivity (1:3)
in glycosylation. This difference can be explained by steric effect.
The equatorial hydroxyl group is generally considered less sterically
hindered and is thus more reactive than the axial hydroxyl group.
However, when the electrophile is small, like acetyl group, the steric
effect is not strong enough to control the reaction outcome. The elec-
tronic effect becomes the key factor under this circumstance and the
Fukui function is a better indicator for the regioselectivity. When the
electrophile is bigger, like glycosyl donor, the steric effect is more
substantial and contributes more to the control of the regioselectiv-
As part of our effort to develop a regioselective synthesis of high
mannose type N-glycan oligosaccharides, we needed a regioselec-
tive diol acceptor like compound 1 (Fig. 1). To search for such a diol
acceptor, a set of 2,3-diol mannose acceptors (Fig. 1 and 2–5) were
selected as candidates for a preliminary study. Fukui function was
calculated using Q-CHEM 3.2, at B3LYP/6-31+G* level for all four
diol acceptors.19 Both Mulliken charges and NBO20 charges were
used for the calculation. Two different charges gave different but
comparable Fukui function values. All fꢀ values (fMꢀ (Mulliken
charge) and fNꢀ (NBO charge)) for oxygen atom of each hydroxyl
group are shown in Figure 1. Since fꢀ is used as the indicator for
electrophilic attack, we expect oxygen atoms with higher Fukui
function value to be the preferred reaction sites. In compound 2,
Fukui function at 3-OH is about 1.5 times of 2-OH. In compound
3, Fukui function at 3-OH is about 2.8 times of 2-OH. In compound
4, 2-OH and 3-OH have about the same Fukui function values. In
compound 5, the Fukui function at 3-OH is about 2.1 times of 2-
OH. These results suggest that 3-OHs of compound 2, 3, and 5
are the more reactive sites when react with electrophiles, whereas
2-OH and 3-OH in compound 4 are of similar reactivity.
To test how the prediction based on Fukui function correlates to
real reactivity of the hydroxyl groups towards electrophiles, com-
pounds 3, 4, and 5 were tested in acetylation reactions (Fig. 2).21,22
Compound 2 was not tested in reactions because methyl group is
not a commonly used protecting group in glycosylation reactions.
In the acetylation reactions, acceptor 3 and 5 showed great regiose-
lectivity with mostly 3-O-acetyl products isolated in the reactions.
Compound 4 didn’t show significant regioselectivity, with similar
amount of 2-O- and 3-O-acetylation products isolated. These results
are clearly consistent with the predictions based on Fukui function,
which suggest that 3-OH is more reactive in compound 3 and 5, but

1552
J. Kalikanda, Z. Li / Tetrahedron Letters 51 (2010) 1550–1553
OAc
O
AcO
AcO
OAc
O
AcO
AcO
AcO
OH
AcO
AcO
OAc
O
RO
OH
O
RO
RO
HO
O
RO
RO
AcO
RO
O
O
+
+
HO
AcO
O
X
OMe
3
R = Bn
OMe
OMe
8a
X=Br
9
10
8b X=SPh
4 R = Ac
8c
X=trichloroacetoimidate
5
R = -CHPh-
Figure 3. Glycosylation of diol acceptors.
References and notes
Table 1
Glycosylation results
1. Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357–2364.
2. B. Ernst, G.W. Hart, P. Sinay, Carbohydrates in Chemistry and Biology, 2000.
3. Schofield, L. Nature 2002, 418, 785–789.
4. A. Varki, Essentials of Glycobiology, 1999.
5. Agnihotri, G.; Misra, A. K. Carbohydr. Res. 2006, 341, 2420–2425.
6. Huang, L.; Huang, X. Chem. Eur. J. 2007, 13, 529–540.
7. Leigh, D. A.; Smart, J. P.; Truscello, A. M. Carbohydr. Res. 1995, 276, 417–424.
8. Zhang, Z.; Ollmann, I. R.; Ye, X.-S.; Wischnat, R.; Baasov, T.; Wong, C.-H. J. Am.
Chem. Soc. 1999, 121, 734–753.
9. Huang, X.; Huang, L.; Wang, H.; Ye, X.-S. Angew. Chem., Int. Ed. 2004, 43, 5221–
5224.
Compound 3
Compound 4
Compound 5
Donor
8 X = Br
9 X = SPh
9:10
9:10
1:3c
1:3c
1:3c
9:10
10 only
10 only
10 only
10 only
10 onlya
10 X = trichloroacetoimidate
10 onlyb
a
b
c
With about 10% of trisaccharide.
With about 15% of trisaccharide.
Trisaccharides are present at similar amount to 1,2-disaccharides.24
10. Z.J. Witczak, in: B. Fraser-Reid, K. Tasuta, J. Thiem (Eds.), Glycoscience,
Chemistry and Chemical Biology, 2001.
11. Ayers, P.; Melin, J. Theor. Chem. Acc. 2007, 117, 371–381.
12. Fuentealba, P.; Chamorro, E.; Cárdenas, C. Int. J. Quantum Chem. 2007, 107, 37–
45.
13. Parr, R. G.; Yang, W. J. Am. Chem. Soc. 1984, 106, 4049–4050.
14. Cohen, M. H.. In Density Functional Theory IV-Theory of Chemical Reactivity;
Springer-Verlag: Heidelberg, 1986; vol. 183.
15. Gazquez, J. L.; Vela, A.; Galwan, M.. In Electronegativity; Springer-Verlag:
Heidelberg, 1987; vol. 66.
ity. Under this circumstance, both steric effect and electronic effect
must be considered to explain the regioselectivity. Diol acceptor 3
and 5, where both steric and electronic effects favor 3-OH are there-
fore more regioselective than diol acceptor 4, which is only favorable
when considering the steric effect.
Another factor that can affect the regioselectivity of diol acceptor
is the intramolecular hydrogen bonding.25 Even though both 2-OH
and 3-OH are trans to the neighboring groups, which is considered
unfavorable for the formation of intramolecular hydrogen bond, 3-
OH can still form hydrogen bond with neighboring group (at 4-posi-
tion), because it is equatorial. At the same time 2-OH cannot form
hydrogen bond with neighboring group (at 1-position) due to its ax-
ial orientation. This can explain why 3-OH is more reactive than 2-
OH, because it has been suggested that hydroxyl group involved in
stronger hydrogen bond is more reactive.26 However, this effect can-
not explain the difference between acetyl- and benzyl-protected
acceptors. Acetyl is expected to form a stronger hydrogen bonding
and should show a higher regioselectivity, which is opposite to the
experimental observation. There must be other factors that make
3-OH of compound 2 and 3 more reactive, which could be the more
favorable Fukui function value at the 3-position. A comprehensive
consideration of Fukui functions and other factors (like steric factors
and hydrogen bondings) could therefore give more accurate predic-
tion of the regioselectivity of diol acceptors.
In summary, Fukui functions were calculated for a series of
mannose diol acceptors and successfully applied in predicting the
regioselectivity of these acceptors in reactions with electrophiles.
Computational chemistry and Fukui function could be a good indi-
cator for predicting the reactivity of glycosyl acceptors, especially
the electronic effect and the influence of protecting groups. More
studies will be conducted to further understand and extend the
application to more glycosyl acceptors.
16. Madjarova, G.; Tadjer, A.; Cholakova, T. P.; Dobrev, A. A.; Mineva, T. J. Phys.
Chem. A 2004, 109, 387–393.
17. Luis, R. D.; Eduardo, C.; Patricia, P. Eur. J. Org. Chem. 2009, 2009, 3036–3044.
18. Mendez, F.; Tamariz, J.; Geerlings, P. J. Phys. Chem. A 1998, 102, 6292–6296.
19. Shao, Y.; Fusti-Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S. T.;
Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.; O’Neill, D. P.; Distasio, R. A.,
Jr; Lochan, R. C.; Wang, T.; Beran, G. J. O.; Besley, N. A.; Herbert, J. M.; Lin, C. Y.;
Van Voorhis, T.; Chien, S. H.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.;
Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.; Dachsel, H.;
Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney,
S. R.; Heyden, A.; Hirata, S.; Hsu, C.-P.; Kedziora, G.; Khalliulin, R. Z.; Klunzinger,
P.; Lee, A. M.; Lee, M. S.; Liang, W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.;
Pieniazek, P. A.; Rhee, Y. M.; Ritchie, J.; Rosta, E.; Sherrill, C. D.; Simmonett, A.
C.; Subotnik, J. E.; Woodcock, H. L., III; Zhang, W.; Bell, A. T.; Chakraborty, A. K.;
Chipman, D. M.; Keil, F. J.; Warshel, A.; Hehre, W. J.; Schaefer, H. F., III; Kong, J.;
Krylov, A. I.; Gill, P. M. W.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2006, 8,
3172–3196.
20. Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899–926.
21. Chowdhary, M. S.; Jain, R. K.; Rana, S. S.; Matta, K. L. Carbohydr. Res. 1986, 152,
323–328.
22. Kong, F.; Schuerch, C. Carbohydr. Res. 1983, 112, 141–147.
23. Evgeny, V.; Malcolm, B. P.; Conlan, J. W. Eur. J. Biochem. 2002, 269, 6112–6118.
24. General procedure for acetylation reactions: The diol (0.27 mmol) and pyridine
(0.27 mmol) were dissolved in anhydrous CH2Cl2 (2 mL) and cooled to 0 °C.
Acetic anhydride (0.27 mmol, in 1 mL of CH2Cl2) was added dropwise within
30 min. The reaction was allowed to warm up to room temperature and stirred
overnight. The reaction mixture was then concentrated and purified by flash
chromatography to give the products. General glycosylation procedure for
glycosyl bromide donors: The diol (0.27 mmol) and glycosyl donor (0.27 mmol),
in anhydrous CH2Cl2 (5 mL) were stirred at room temperature in flame-dried
molecular sieves 3 Å under nitrogen atmosphere for 30 min. The mixture was
cooled to ꢀ30 °C, silver triflate (0.27 mmol) was added and the reaction was
allowed to warm up to room temperature. After 30 min, it was quenched with
triethylamine, concentrated, and purified by flash chromatography to give the
products. General glycosylation procedure for thioglycoside donors: The diol
(0.27 mmol) and glycosyl donor (0.27 mmol) in anhydrous CH2Cl2 (5 mL) were
stirred at room temperature in flame-dried molecular sieves 3 Å under
nitrogen atmosphere for 30 min. The mixture was cooled to ꢀ30 °C, NIS
(0.41 mmol) was added, after stirring for 5 min, BF3ꢁEt2O (0.27 mmol) was
added and the reaction was allowed to warm up to room temperature. After
30 min, it was quenched with saturated aqueous sodium bicarbonate and 10%
aqueous sodium thiosulphate, extracted with dichloromethane, and washed
with brine. It was dried over sodium sulfate, concentrated, and purified by
flash chromatography to give the products. General glycosylation procedure for
trichloroacetimidate donors: The diol (0.27 mmol) and glycosyl donor
Acknowledgments
We thank Professor Paul W. Ayers of McMaster University for
interesting discussion on using Fukui function for reaction predic-
tion. We also thank James Switzer and Professor Jim Dix of the
Department of Chemistry for help in setting up our computational
program. The financial support is provided by Binghamton Univer-
sity Research Foundation.

J. Kalikanda, Z. Li / Tetrahedron Letters 51 (2010) 1550–1553
1553
(0.27 mmol) in anhydrous CH2Cl2 (5 mL) were stirred at room temperature in
flame-dried molecular sieves 3 Å under nitrogen atmosphere for 30 min. The
mixture was cooled to ꢀ78 °C, BF3ꢁEt2O (0.14 mmol) was added, after 20 min
the reaction was quenched with triethylamine concentrated and purified by
flash chromatography to give the products.
25. Muddasani, P. R.; Bernet, B.; Vasella, A. Helv. Chim. Acta 1994, 77, 334–350.
26. Cid, M. B.; Alfonso, F.; Alonso, I.; Martin-Lomas, M. Org. Biomol. Chem. 2009, 7,
1471–1481.