Improved anomeric selectivity for the aroylation of sugars

Article abstract of DOI:10.1016/j.carres.2004.02.015

By manipulating the solvent and using bulky TMEDA as a base, good yields and improved anomeric selectivities were obtained for the aroylation of D-glucose over similar esterifications using pyridine. The reaction has been extended to mannose and the β-ano

Full text of DOI:10.1016/j.carres.2004.02.015

Carbohydrate
RESEARCH
Carbohydrate Research 339 (2004) 1373–1376
Note
Improved anomeric selectivity for the aroylation of sugars
M. Teresa Barros,a,b, Christopher D. Maycock, Paula Rodriguesb and
c,d
*
Christine Thomassignyd
aDepartamento de Quimica, Faculdade de Ci^encias e Tecnologia, Universidade Nova de Lisboa, P-2825-114 Caparica, Portugal
bCentro de Quimica Fina e Biotecnologia/Requimte, FCT, Universidade Nova de Lisboa, P-2825-114 Caparica, Portugal
cDepartamento de Quimica e Bioquimica, Faculdade de Ci^encias, Universidade de Lisboa, Rua Ernesto de Vasconselos,
P-1700 Lisboa, Portugal
dInstituto de Tecnologia Quimica e Biologia, Universidade Nova de Lisboa, Apartado 127, P-2780-156 Oeiras, Portugal
Received 19 September 2003; received in revised form 16 February 2004; accepted 16 February 2004
Available online 24 March 2004
Abstract—By manipulating the solvent and using bulky TMEDA as a base, good yields and improved anomeric selectivities were
obtained for the aroylation of D-glucose over similar esterifications using pyridine. The reaction has been extended to mannose and
the b-anomer of pergalloylated mannose was predominantly obtained in one step by direct aroylation of the parent sugar.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Aroylation; Galloylation; Carbohydrate esterification; Anomeric esters; TMEDA
Although the aroylation of alcohols is a basic and well-
studied subject, the development of synthetic methods
for the selective aroylation of sugars is of great impor-
tance, especially when these methods are stereoselective
with respect to the anomeric centre. Unfortunately, the
known methods for direct aroylation of sugars normally
lead to a mixture of anomers where the kinetic isomer is
predominant,1;2 making the obtention of the other
anomer in important quantities a challenge.3;4 Natural
b-penta-O-galloyl glucose (b-PGG) is believed to be
responsible for the genesis of ellagitannins.5;6 The ready
availability of both anomers could help in the search for
more powerful antiviral or anticancer therapeutic
agents.7–9 We have found that by changing the solvent
and base in the reaction of galloylation, it was possible
to obtain protected PGGs with an anomeric excess
superior to 85%.
base such as pyridine, triethylamine and/or 4-DMAP.10
More recently,11 a new method for the benzoylation of
alcohols has been described using TMEDA as a base,
which gave the expected benzoates in excellent yields.
This method has been extended to other acyl halides,
allowing the obtention of alkoxycarbonates12 or pro-
tected gallates.13
A classical benzoylation of
D
-glucose in pyridine2
(Table 1, entry 1) led predominantly to the b anomer,
implying that the b-glucopyranose form had been
selectively aroylated. The result was similar with the
more hindered tri-O-methylgalloyl chloride (2) and the
reaction led to the same proportion of anomers (entry
4). In an attempt to obtain the b-PGG with a better
anomeric excess, we have replaced pyridine with TME-
DA, expecting that the bulky intermediate 411;14–16
(Fig. 1) suspected to be formed between the acyl halide
and the base, would react preferentially at the least
hindered equatorial anomeric hydroxyl to give the
As a general rule, aroylation of alcohols involves their
treatment by an aroylating agent in the presence of a
expected b-anomer of
D
-glucose (Scheme 1).4;17;18
As expected (entries 3, 5, 6 and 12), the use of
TMEDA led preferentially to the kinetic product.
Contrarily to previously published studies,11 we have
* Corresponding author. Tel.: +351-214469775; fax: +351-214469789;
e-mail: maycock@itqb.unl.pt
0008-6215/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.02.015

1374
M. Teresa Barros et al. / Carbohydrate Research 339 (2004) 1373–1376
Table 1. Aroylation of glucose (yields not optimised)
Entry
Solvent
G0Cl (equiv)
Base (equiv)a
Yield (%)
a:b
12
2
––
1
Pyridine
Pyridine
––
86
80
13
9
39:61
70:30
31:69
39:61
16:84
14:86
32:68
23:77
32:68
42:58
24:76
14:86
35:65
86:14
12:88
34:66
0:100
CH2Cl2
CH2Cl2
––
1 (8)
1 (8)
2 (8)
2 (5)
2 (5)
2 (5)
2 (5)
2 (5)
2 (5)
2 (5)
2 (8)
2 (8)
2 (8)
2 (8)
2 (8)
3 (8)
3
TMEDA (8)
Pyridine
4
5
CH2Cl2
CH2Cl2
THF
TMEDA (1)
TMEDA (5)
TMEDA (5)
TMEDA (5)
TMEDA (5)
TMEDA (5)
TMEDA (5)
TMEDA (8)
TMEDA (8)
Pyridine
6
24
41
45
20
22
29
70
54
60
52
86
61
7
8
CH2Cl2
Toluene
CH3CN
THF:CH2Cl2 (1:2)
CH2Cl2
CH2Cl2
CH2Cl2
THF
9
10
11
12
13
14
15
16
17
TMEDA (8)
TMEDA (8)
TMEDA (8)
CH3CN
CH2Cl2
aPyridine:CH2Cl2, 1:2 for entries 2, 8, 13 and 14.
prisingly, use of dichloromethane as cosolvent with
pyridine (entries 2 and 14) inverted the anomeric selec-
tivity and the a anomer was obtained as the predomi-
nant product from the reaction of glucose with 2,
yielding 86% of a-PGG (entry 14).
N
COCl
O
N
1 R = H
G'Cl =
2 R = OMe
3 R = OBn
Cl
R
R
R
R
R
R
4
Knowing that the b-anomer must be the kinetic
product of the reaction of aroylation,4 we have then
been able to deduce more aspects of the mechanism of
aroylation in the presence or absence of TMEDA. As we
have seen before, the amount of base strongly influences
the overall yield. This fact is consistent with the for-
mation of the reactive complex 4. Following the model
described by Bols and Hanson,19 we can infer that the
rate of mutarotation in dichloromethane or tetrahy-
drofuran compared with the rate of aroylation by 4,
favoured the b-O-aroylated compound, or that the base
(TMEDA) was a better catalyst for mutarotation than
pyridine. The bulky reagent 4 inhibited the reaction with
the a-hydroxyl. The use of an aroylating agent such as 2,
with its particular electron availability may enhance the
stability of the intermediate 4. The N-acylpyridinium
halide, which was probably formed on mixing the
chloride reagent and pyridine20 is probably responsible
for the obtention of the other anomer predominantly.
In order to show that the hindering of the aroylating
agent plays a crucial role in the reaction of aroylation,4
we have run the reactions in the presence of a similar but
larger aroylating agent than 2, namely tri-O-benzyl
galloylchloride 3 (Table 1, entry 17). In this case, the
combined effects of the formation of the intermediate
type 4 and its hindrance turned the reaction totally
selective and we isolated only the kinetic b-anomer. This
result can be compared with the one reported in the
literature,21 where the obtention of the same products by
direct galloylation in the presence of DCC had led to a
nearly equimolar (33:42) a=b mixture.
Figure 1. Aroylating reagents used in this work.
OH
OG'
O
OG'
G'Cl
HO
HO
O
G'O
G'O
+
G'O
G'O
O
OG'
base
OH OH
G'O
G'O
OG'
Scheme 1.
found that the amount of base was very important for
the galloylation reaction (Table 1). The yield of the
reaction was effectively higher when the number of equiv
of TMEDA was the same as for the tri-O-methylgalloyl
chloride 2 (entries 5, 6 and 12; 10 and 16). The best
yields of aroylation were obtained by adding 1.6 equiv of
chloride per hydroxyl group, and the same quantity of
base (entries 12–16).
The importance of the solvent on the anomeric
selectivity of the esterifications is also noteworthy (en-
tries 6–16). Actually, the use of dry acetonitrile provided
the best yields (entry 16), but it was with dry dichloro-
methane (entries 5, 6 and 12) or tetrahydrofuran (entry
15) that the b-selectivity was the most pronounced. The
addition of pyridine to dichloromethane (ratio 1/2, en-
tries 8 and 13) tended to decrease the proportion of
protected b-PGG in favour of its a anomer. We also
noticed (entry 4) that the use of pyridine only gave a low
(13%) yield with a ratio a : b of about 1 to 1.5. Sur-

M. Teresa Barros et al. / Carbohydrate Research 339 (2004) 1373–1376
1375
OH
OH
O
chemical shift values (d) in ppm downfield from tetra-
methylsilane. Elemental analyses were performed by the
IST analytical services in Lisbon using a combustion
apparatus.
OG'
OG'
OG'
OG'
G'Cl
HO
HO
G'O
G'O
O
+ G'O
G'O
O
OG'
base
OH
OG'
Scheme 2.
1.2. Typical procedure for the galloylation reactions
We have then extended this study to the galloylation
of mannose (C-2 epimer of glucose) (Scheme 2, Table 2,
entries 2–6) and have obtained the expected compounds
with good to very good yields. The classical benzoyl-
ation2 of mannose in pyridine afforded mainly the
a-anomer (entry 1). The addition of dichloromethane as
solvent increased the ratio in favour of the b-anomer
(entry 2), whereas substitution of pyridine by TMEDA
gave predominantly the b-anomeric form (entry 3).
We have thus managed to obtain with good diaste-
reoselectivity the b-form of the pentakis-(O-trimeth-
oxygalloyl)mannose (entry 4) by using TMEDA. The
formation of the equatorial anomer was probably due to
the kinetic anomeric effect.4
In conclusion, we have demonstrated that several
factors affect the anomeric selectivity of the aroylation
of sugars. We also describe a direct method for the
galloylation of sugars where the anomeric stereoselec-
tivity can be controlled by varying the acyl chloride
(bulkyness, electron availability), the solvent and the
base. The obtention of the challenging b-mannopyr-
To a soln of TMEDA (0.34 mL; 2.24 mmol; 8 equiv) in
dry solvent (5 mL) was added
D-glucose (50 mg;
0.28 mmol) and tri-O-methylgalloyl chloride (517 mg;
2.24 mmol; 8 equiv) dissolved in dry solvent (5 mL). The
reaction mixture was refluxed for 24 h under an inert
atmosphere. After cooling, it was quenched with a
phosphate buffer solution pH 7, extracted with CH2Cl2
(3 · 20 mL), and the organic phase washed with brine
and water, dried and concentrated. Purification of the
crude product by preparative chromatography afforded
the pure (4:1:1 Et2O–CH2Cl2–hexane) a- and b-penta-
kis(O-trimethoxygalloyl)-D-glucoses, or a mixture of the
two whose ratio have been determined by 1H NMR
spectra and/or HPLC.
1.3. 1,2,3,4,6-Pentakis[-O-(3,4,5-trimethoxybenzoyl)]-
a,b-D-glucopyranose
Purification by preparative chromatography (2:1:1
Et2O–CH2Cl2–hexane) led to the two anomers.
anoside as the predominant product from D-mannose
by a cheap, nontoxic and direct method of aroylation
has been realised.
a-Anomer: mp 96–97 ꢁC. ½aꢀ +81.2 (c 0.8, CH2Cl2).
D
IR data (KBr): mmax 1731.6 (C@O). 1H NMR data
(400 MHz, CDCl3): 7.41 (s, 2H, Gall-H), 7.29 (s, 2H,
Gall-H), 7.14 (s, 4H, Gall-H), 7.11 (s, 2H, Gall-H), 6.77
(d, J1;2 3.2 Hz, 1H, H-1), 6.26 (t, J 10.0–10.4 Hz, 1H, H-
3), 5.75 (t, J 10.0–10.4 Hz, 1H, H-4), 5.56 (dd, J1;2 3.6,
J2;3 10.0 Hz, 1H, H-2), 4.77 (dd, J5;6b 2.6 Hz, 1H, H-6b),
4.67–4.63 (m, 1H, H-5), 4.40 (dd, J5;6a 5.4, J6a;6b 12.2 Hz,
1H, H-6a), 3.94, 3.93, 3.89, 3.85, 3.83, 3.82, 3.71 (7s,
45H, OCH3). 13C NMR data (100 MHz, CDCl3): d
165.59, 164.98, 164.82, 163.87, (C@O), 153.16, 152.84
(Gall-C-3 and Gall-C-5), 143.28, 142.86, 142.75, 142.65,
142.45 (Gall-C-4), 124.34, 123.77, 123.53, 123.41, 123.30
(Gall-C-1), 107.38, 107.04, 106.91 (Gall-C-2 and Gall-C-
6), 90.29 (C-1), 70.72 (C-2), 70.52 (C-3 and C-5), 69.18
(C-4), 62.85 (C-6), 60.94, 60.80, 56.32, 56.10, 56.04,
55.88 (OCH3). Anal. Calcd for C56H62O26: C, 58.43; H,
5.43. Found: C, 58.25; H, 5.49.
1. Experimental
1.1. General methods
Reagents and solvents were purified before use. Flash
chromatography was performed on silica gel from
Macherey-Nagel (Kieselgel 60 M). Compounds were
visualised with a soln of 30% H2SO4 in EtOH and
heating, or under UV light. Optical rotations were
measured at 20 ꢁC on a Perkin–Elmer 241 polarimeter
€
(1 dm cell). Melting points were determined on a Buchi
530 apparatus and are not corrected. Anhydrous
MgSO4 was used to dry organic extracts. 1H NMR
spectra were recorded on a Bruker AMX-400 apparatus
(1H at 400 MHz and 13C at 100 MHz) in CDCl3 with
Table 2. Aroylation of mannose
Entry
Solvent
G0Cl (equiv)
Base (equiv)
Yield (%)
a:b
1
2
3
4
5
6
––
1
Pyridine
Pyridine
95
87
87
88
77
91
86:14
64:36
44:56
20:80
34:66
52:48
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
1 (8)
1 (8)
2 (8)
2 (8)
2 (10)
TMEDA (8)
TMEDA (8)
Pyridine
Pyridine

1376
M. Teresa Barros et al. / Carbohydrate Research 339 (2004) 1373–1376
b-Anomer: mp 99–100 ꢁC. ½aꢀ +18.91 (c 0.6, CH2Cl2).
(C-4), 63.19 (C-6), 60.64, 60.54, 55.89, 55.76, 55.52
(OCH3). Anal. Calcd for C56H62O26: C, 58.43; H, 5.43.
Found: C, 57.95; H, 5.53.
D
IR data (KBr): mmax 1728.7 (C@O). 1H NMR data
(400 MHz, CDCl3): d 7.33, 7.31, 7.17, 7.16, 7.11 (5s,
10H, Gall-H), 6.22 (d, J1;2 8.4 Hz, 1H, H-1), 6.03 (t, 1H,
H-3), 5.81 (dd, J1;2 8.4, J2;3 9.6 Hz, 1H, H-2), 5.74 (t, J
9.2–9.6 Hz, 1H, H-4), 4.81 (d, J6a;6b 12.2 Hz, 1H, H-6b),
4.39–4.46 (m, 2H, H-5 and H-6a), 3.92, 3.91, 3.90, 3.88,
3.87, 3.86, 3.85, 3.84, (8s, 45H, OCH3). 13C NMR data
(100 MHz, CDCl3): d 165.63, 165.48, 165.01, 164.91,
164.17, (C@O), 152.95, 152.88, 152.81 (Gall-C-3 and
Gall-C-5), 143.03, 142.78, 142.65, 142.40, (Gall-C-4),
124.42, 123.45, 123.38, 123.12, (Gall-C-1), 107.32,
107.04, 106.94 (Gall-C-2 and Gall-C-6), 92.84 (C-1),
73.16 (C-5), 72.89 (C-3), 71.08 (C-2), 69.62 (C-4), 63.06
(C-6), 60.77, 56.07, 56.00 (OCH3). Anal. Calcd for
C56H62O26: C, 58.43; H, 5.43. Found: C, 58.40; H, 5.38.
Acknowledgements
This work has been supported by NATO SFS (PO-
~
^
SUGARS) and Fundaßcao para a Ciencia e a Tecnologia
(POCTI/QUI/47973/2002).
References
1. DÕAccorso, N. B.; Thiel, I. M. E.; Schuler, M. Carbohydr.
Res. 1983, 124, 177–184.
2. Ness, R. K.; Fletcher, H. G., Jr.; Hudson, C. S. J. Am.
Chem. Soc. 1950, 72, 2200–2205.
1.4. 1,2,3,4-Pentakis-[O-(3,4,5-trimethoxybenzoyl)]-a,b-
3. For recent articles, see: (a) Crich, D.; Sun, S. Tetrahedron
1998, 54, 8321–8348; (b) Chung, S.-K.; Park, K.-H.
Tetrahedron Lett. 2001, 42, 4005–4007.
4. Boons, G.-J. In Carbohydrate Chemistry; Blackie Aca-
demic and Professional, Thomson Science: London, 1998;
pp 21–25, pp 10–14 and 108–110.
5. Feldman, K. S.; Sambandam, A. J. Org. Chem. 1995, 60,
8171–8178.
6. Gross, G. G. In Phenolic Metabolism in Plants; Stafford,
H. A., Ibrahim, R. K., Eds.; Plenum: New York, 1992;
Vol. 26, pp 297–324.
7. Kashiwada, Y.; Nonaka, G.-I.; Nishioka, I.; Chang, J.-J.;
Lee, K.-H. J. Nat. Products 1992, 55, 1033–1043.
8. Kashiwada, Y.; Nonaka, G.-I.; Nishioka, I.; Lee, K. J.-H.;
Bori, I.; Fukushima, Y.; Bastow, K. F.; Lee, K.-H.
J. Pharm. Sci. 1993, 82, 487–492.
9. Kashiwada, Y.; Nonaka, G.-I.; Nishioka, I.; Ballas, L. M.;
Jiang, J. B.; Janzen, W. P.; Lee, K.-H. Bioorg. Med. Chem.
Lett. 1992, 2, 239–244.
10. Greene, T. W.; Wuts, P. G. M. Protective Groups in
Organic Synthesis, 3rd ed.; John Wiley & Sons, 1999.
11. Sano, T.; Ohashi, K.; Oriyama, T. Synthesis 1999, 7, 1141–
1144.
12. Adinolfi, M.; Barone, G.; Guariniello, L.; Iadonisi, A.
Tetrahedron Lett. 2000, 41, 9305–9309.
D
-mannopyranoses
a-Anomer: mp 94–95 ꢁC. ½aꢀ )35.56 (c 1.0, CH2Cl2). IR
D
data (KBr): mmax 1729.1 (C@O); 1H NMR data
(400 MHz, CDCl3): 7.41, 7.33, 7.24, 7.13, 7.06, (5s, 10H,
Gall-H), 6.54 (s, 1H, H-1), 6.07 (t, J 10.0 Hz, 1H, H-4),
5.97 (dd, J2;3 2.6, J3;4 10.2 Hz, 1H, H-3), 5.88 (d, J1;2
1.6 Hz, 1H, H-2), 4.70 (dd, J5;6b 1.6 Hz, 1H, H-6b), 4.59–
4.63 (m, 1H, H-5), 4.47 (dd, J5;6a 5.4, J6a;6b 12.2 Hz, 1H,
H-6a), 3.97, 3.95, 3.94, 3.92, 3.92, 3.88, 3.88, 3.84, 3.83,
3.80, 3.80, 3.79, 3.76, 3.65 (14s, 45H, OCH3). 13C NMR
data (100 MHz, CDCl3): d 165.56, 165.09, 164.95,
164.77, 163.42 (C@O), 153.16, 153.13, 152.84, 152.76
(Gall-C-3 and Gall-C-5), 143.47, 143.04, 142.90, 142.59,
142.44 (Gall-C-4), 124.33, 123.65, 123.55, 123.36 (Gall-
C-1), 107.57, 107.35, 107.10, 107.02, 106.86 (Gall-C-2
and Gall-C-6), 91.60 (C-1), 71.00 (C-5), 69.86 (C-3),
69.75 (C-2), 66.84 (C-4), 63.09 (C-6), 60.96, 60.87, 60.78,
56.42, 56.14, 56.09, 55.94, 55.73 (OCH3). Anal. Calcd
for C56H62O26: C, 58.43; H, 5.43. Found: C, 58.29; H,
5.56.
~
13. Barros, M. T.; Maycock, C. D.; Sineriz, F.; Thomassigny,
b-Anomer: mp 90–91 ꢁC. ½aꢀ )65.14 (c 1.0, CH2Cl2).
D
C. Tetrahedron 2000, 56, 6511–6516.
14. Fersht, A. R.; Jencks, W. P. J. Am. Chem. Soc. 1970,
5442–5452.
15. King, J. A., Jr.; Bryant, G. L., Jr. J. Org. Chem. 1992, 57,
5136–5139.
16. Hubbard, P.; Brittain, W. J. J. Org. Chem. 1998, 63, 677–
683.
17. Juaristi, E.; Cuevas, G. In The Anomeric Effect; Rees, C.
W., Ed.; CRC: London, 1995.
IR data (KBr): mmax 1728.3 (C@O). 1H NMR data
(400 MHz, CDCl3): d 7.39, 7.26, 7.14, 7.13, 7.08 (5s,
10H, Gall-H), 6.36 (s, 1H, H-1), 6.05 (d, J1;2 2.8 Hz, 1H,
H-2), 6.00 (t, J4;5 9.6 Hz, 1H, H-4), 5.74 (dd, J2;3 3.2, J3;4
9.8 Hz, 1H, H-3), 4.80 (d, J6a;6b 9.2 Hz, 1H, H-6), 4.47–
4.40 (m, 2H, H-5 and H-6a), 3.88, 3.85, 3.83, 3.81, 3.80,
3.78, 3.77, 3.65, 3.63 (9s, 45H, OCH3). 13C NMR data
(100 MHz, CDCl3): d 165.34, 164.94, 164.69, 163.22
(C@O), 152.97, 152.68, 152.62, 152.57 (Gall-C-3 and
Gall-C-5), 142.78, 142.71, 142.66, 14.43, 142.14 (Gall-C-
4), 124.21, 123.78, 123.29, 122.95 (Gall-C-1), 107.17,
106.96,106.83,106.75, 106.70 (Gall-C-2 and Gall-C-6),
90.85 (C-1), 72.67 (C-5), 71.33 (C-3), 69.74 (C-2), 67.14
18. Juaristi, E.; Cuevas, G. Tetrahedron 1992, 48, 5019–5087.
19. Bols, M.; Hansen, H. C. Acta Chem. Scand. 1993, 47, 818–
822.
€
€
20. Hofle, G.; Steglich, W.; Vorbruggen, H. Angew. Chem.,
Int. Ed. Engl. 1978, 17, 569–583.
€
21. Khanbabaee, K.; Lotzerich, K. Tetrahedron 1997, 53,
10725–10732.