The application of the intermediate 2-methyl-glyco-[2,1-d]-2-oxazolines for glycoside synthesis

Article abstract of DOI:10.1016/S0008-6215(00)00237-8

The synthesis of 2-acylamino-2-deoxysugars 1,2-trans-glycosides is described via the oxazolinium salt generated from an O,N-acetylated 1,2-cis-glycosyl halide of 2-amino-2-deoxysugar under the conditions of halide-anion catalysis. This salt was then inter

Full text of DOI:10.1016/S0008-6215(00)00237-8

www.elsevier.nl/locate/carres
Carbohydrate Research 329 (2000) 895–899
Note
The application of the intermediate
2-methyl-glyco-[2,1-d]-2-oxazolines for glycoside
synthesisꢀ
Sergey S. Pertel *, Vasily Ya. Chirva, Andrey L. Kadun, Elena S. Kakayan
Department of Organic Chemistry, National Taurida Uni6ersity, Yaltinskaya 4, Simferopol, Crimea 95007, Ukraine
Received 5 August 1999; received in revised form 12 January 2000; accepted 4 August 2000
Abstract
The synthesis of 2-acylamino-2-deoxysugars 1,2-trans-glycosides is described via the oxazolinium salt generated
from an O,N-acetylated 1,2-cis-glycosyl halide of 2-amino-2-deoxysugar under the conditions of halide-anion
catalysis. This salt was then interacted with alcohol to form the corresponding 1,2-trans-glycoside. A method for
removing the generated hydrogen chloride is described. The conditions of this synthesis allow glycosides with
acid-labile functional groups to be obtained. Suppression of the anomerisation of 1,2-trans-glycosides was achieved
by the introduction of N,N%-dicyclohexyl urea into the reaction medium. © 2000 Elsevier Science Ltd. All rights
reserved.
Keywords: Glycosyl halides; Halide-ion catalysis; 1,2-Trans-glycosides of 2-acylamino-2-deoxysugars; Glyco-[2,1-d]-2-oxazolines;
1,2-Trans-glycosaminides anomerisation
The 2-substituted glyco-[2,1-d]-2-oxazolines
are widely used in the synthesis of N-acetyl-
ated 1,2-trans-glycosaminides [1,2]. Fixation
of the anomeric centre configuration during
bicyclic system formation is an important ad-
vantage of the oxazoline method, providing
glycosylation stereospecificity [3]. The sugars’
oxazoline derivatives are comparatively un-
stable. Like other imidoesters, they are easily
hydrolysed at the CꢀN bond when traces of
protic acids are present [4]. Glycooxazolines
are also easily subjected to elimination to
form 2-acylaminoglycals [5,6]. As a rule, oxa-
zoline stability decreases with the increase of
their glycosylating activity. Therefore the gly-
coside synthesis method, with oxazolines be-
ing generated from a stable precursor directly
in the reaction medium, would be an
advantage.
2-Methylglycooxazolines can be easily ob-
tained from N-acylglycosyl halides of 2-
amino-2-deoxysugars by the Lemieux method
[7]. In the conditions of halide-ion catalysis,
the relatively stable 1,2-cis-glycosyl halides
which are obtained are transformed into the
reactive 1,2-trans-glycosyl halides. Treatment
of these trans-halides by weak bases, sodium
hydrogen carbonate for example, induces an
amidic carbonyl intramolecular nucleophilic
attack on the anomeric centre. As a result, the
corresponding oxazoline derivatives are
ꢀ Part of this work was presented at the XVIII Interna-
tional Carbohydrate Symposium, Milan, Italy, July 1996.
* Corresponding author. Fax: +380-652-232310.
E-mail address: orgchem@ccssu.crimea.ua (S.S. Pertel).
0008-6215/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(00)00237-8

896
S.S. Pertel et al. / Carbohydrate Research 329 (2000) 895–899
formed. The glycosylation of alcohols by
sugar oxazoline derivatives is catalysed by
protic acids [8,9] and Lewis acids [10–12]. The
synthesis of glycooxazolines by the Lemieux
method provides a base presence in the reac-
tion medium, and therefore can not be com-
bined with the glycoside synthesis.
glycosides by refluxing of the reagent solution.
The bases usually used for this are unsuitable
as they deprotonate the oxazolinium ion and
therefore
decrease
glycosylating
agent
concentration.
Thus the 1,2-trans-glycosaminide synthesis
was realised by refluxing the solution contain-
ing the N-acylglycosyl halide, halide-anion
with the corresponding alcohol. As high tem-
peratures and polar solvents promote rear-
We have found that oxazoline cycle forma-
tion also occurs in the absence of a base due
to 1,2-trans-glycosyl halide spontaneous rear-
rangement (see Scheme 1). As a result, the
equilibrium mixture contains the initial glyco-
syl halide and glyco-[2,1-d]-2-oxazoline hydro-
chloride when examined by TLC. The
rangement
of
the
oxazolines
into
2-acylaminoglycals [5,6], the most suitable sol-
vent was of low polarity with a boiling point
less than 80 °C in which HCl hardly dissolves,
e.g., chloroform, methylene chloride, 1,2-
dichloroethane, benzene, carbon tetrachloride.
A quaternary ammonium salt, e.g., triethyl-
benzylammonium chloride, was used as a
halide anion source readily soluble in these
media. It should be noted that the alternatives
hydrogen bromide and hydrogen iodide cause
a greater destruction of carbohydrate deriva-
tives than hydrogen chloride and are readily
soluble in the solvents used (Scheme 2).
The synthesis of compounds 4–6 demon-
strated the described method for obtaining
1,2-trans-glycosides of N-acetylated 2-amino-
2-deoxysugars. Benzylglycoside 4 and disac-
charide 5 were obtained by interaction of
glycosyl chloride 1 with 1.0–1.2 equivalents of
the corresponding alcohol in the presence of
2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-a- -
D
glucopyranosyl chloride 1 interaction with wa-
ter in the presence of 0.5 equivalents of
benzyltriethylammonium chloride in ni-
tromethane led to the stereospecific formation
of 2-amino-1,3,4,6-tetra-O-acetyl-2-deoxy-a-
D-glucopyranose hydrochloride 3, which crys-
tallised from the solution with 90% yield (see
Section 1). It was evident that this transforma-
tion could be an effective means of 2-acy-
lamino-2-deoxysugar derivatives selective
N-deacylation in mild conditions.
The oxazolinium salt 2, an active glycosy-
lating agent, gives the corresponding 1,2-
trans-glycoside when alcohol is added instead
of water. In this case, hydrogen halide is also
a product that was removed from the reaction
mixture to avoid destruction of the forming
Scheme 1.

S.S. Pertel et al. / Carbohydrate Research 329 (2000) 895–899
897
Scheme 2.
0.5 equivalents of triethylbenzylammonium
chloride. The reaction proceeded over 15–18
h with refluxing of the reagent chloroform
solution. Products 4 and 5 were obtained in 77
and 71% yields, respectively. For the tert-bu-
tyl glycoside 6, a fourfold surplus of tert-bu-
tanol and 0.2 equivalents of triethylbenzyl-
ammonium chloride were used. Refluxing in
chloroform solution for 24 h afforded the
desired glycoside 6 in 55% yield.
sary to bind the remaining hydrogen halide
with substituted ureas which are weak enough
not to deprotonate the oxazolinium ion.
N,N%-Dicyclohexyl urea (DCU) was the
most effective agent and not the widely used
tetramethyl urea. N,N%-Dicyclohexyl urea
(0.2–0.5 equivalents) suppressed the anomeri-
sation completely in the reaction medium and
promoted an increase in the yield of desired
glycoside. For example, the yield of tert-butyl
glycoside 6 increased from 55 to 68%, and the
yield of benzyl glycoside 4 increased from 77
to 82% (see Section 1). The formation of
anomeric 1,2-cis-glycosides was not observed
using this glycosylation procedure, even with
bromide as a catalyst. In most cases, N,N%-di-
cyclohexyl urea crystallised from the reaction
medium after the solution was cooled. It was
necessary to carefully control complete gly-
coside separation from the N,N%-dicyclohexyl
urea. Glycosylation in the presence of N,N%-
dicyclohexyl urea was optimum as anomerisa-
tion of 1,2-trans-glycoside was eliminated.
The use of a large catalyst concentration in
the glycosylation of alcohols with low reactiv-
ity promoted the formation of a significant
amount of glycal, even at low (B80 °C) tem-
peratures. At high temperatures, this sec-
ondary reaction became predominant. When
the boiling point of the being glycosylated
alcohols was lower than that of the chosen
solvent, it was necessary to use an excess of
being glycosylated substance or multiple addi-
tional introductions into the reaction mixture.
Analysis of the reaction medium showed
that in some cases, e.g., in the synthesis of
tert-butyl glycoside 6, the corresponding
anomeric glycoside is present. As the oxazo-
line synthesis is highly stereospecific, we sup-
posed that the 1,2-cis-glycoside is generated
because of the previously formed 1,2-trans-
glycoside anomerisation. The anomerisation
catalysis was provided by traces of the protic
acid being present in the reaction medium.
Hydrogen bromide is a stronger acid than
hydrogen chloride, and the yield of the 1,2-
cis-glycoside increased when bromide ion was
used as a glycosylation catalyst. To prevent
this undesirable anomerisation it was neces-
1. Experimental
General methods.—1H NMR spectra were
recorded at 200 MHz with a Bruker WP-200
spectrometer in CDCl3. Chemical shifts are
expressed in ppm downfield from Me4Si.
Thin-layer chromatography was performed on
Silufol UV-254 plates — aluminium plates
precoated with silica gel Silpearl UV-254
(Kavalier, Czech Republic). Detection was ac-

898
S.S. Pertel et al. / Carbohydrate Research 329 (2000) 895–899
complished by heating to 300 °C. Column
chromatography was performed with silica gel
L-40/100 (Lachema, Czech Republic) using
gradient elution from CHCl3 to CHCl3–EtOH
(A) 100:0.5 (v/v), (B) 100:1 (v/v). Optical rota-
tion was measured with a Polamat-S polar-
imeter. Melting points were determined in
capillaries and were uncorrected.
(186 mg, 77%): mp 164–165 °C, lit. 165 °C
30 1
[14]; [h]
−59° (c 1.3, CH2Cl2); H NMR
546
(CDCl3): l 7.38–7.22 (m, 5 H, Ar-H), 5.40 (d,
1 H, JNH,2 8 Hz, NH), 5.21 (dd, 1 H, J3,4 9.3
Hz, H-3), 5.09 (dd, 1 H, J4,5 9.3 Hz, H-4), 4.9
(d, 1 H, JPhCHa,PhCHb 12 Hz, PhCHa), 4.63 (d,
1 H, J1,2 8.1 Hz, H-1), 4.6 (d, 1 H, PhCHb),
4.37 (dd, 1 H, J6a,6b 12 Hz, H-6a), 4.13 (dd, 1
H, J6b,5 2.4 Hz, H-6b), 3.95 (ddd, 1 H, J2,3 9.6
Hz, H-2), 3.66 (ddd, 1 H, J5,6a 4.5 Hz, H-5),
2.11, 2.02, 2.015 and 1.91 (4 s, 12 H, 4 Ac).
Method (b). The synthesis of benzyl-2-acet-
1,3,4,6-Tetra-O-acetyl-2-amino-2-deoxy-h-
D
-glucopyranose hydrochloride (3).—2-Acet-
amido-3,4,6-tri-O-acetyl-2-deoxy-a- -glucopy-
D
ranosyl chloride 1 (200 mg, 0.55 mmol) and
triethylbenzylammonium chloride (25 mg, 0.2
equiv) were dissolved in nitromethane (2 mL)
and water (20 mL, 2 equiv) was added to the
solution. After 5 h, when TLC (10:1 (v/v)
CHCl3–EtOH) showed that the reaction was
complete, the precipitate of the hydrochloride
3 was filtered. The filtrate and washings were
then combined and evaporated. Nitromethane
(2 mL) was added to the dry residue. After
filtration of this mixture, an additional quan-
tity of compound 3 was obtained. The total
amido-3,4,6-tri-O-acetyl-2-deoxy-b- -glu-
D
copyranoside was carried out as described in
method (a). In addition, N,N%-dicyclohexyl
urea (DCU, 24 mg, 0.2 equiv) was introduced
in the reaction medium. After reaction com-
pletion and cooling of the solution, the DCU
was removed by filtration. Purification on a
column of silica gel followed by crystallisation
according to method (a) gave 197 mg (82%) of
the title compound.
6 - O - (2 - Acetamido - 3,4,6 - tri - O - acetyl - 2-
yield of 3 was 188 mg (90%): mp 186 °C, lit.
deoxy-i-
propylidene-h-
amido-3,4,6-tri-O-acetyl-2-deoxy-a-
pyranosyl chloride 1 (210 mg, 0.57 mmol),
1,2:3,4-di-O-isopropylidene-a- -galactopy-
D
-glucopyranosyl)-1,2:3,4-di-O-iso-
-galactopyranose (5).—2-Acet-
-gluco-
21
546
185 °C [13]; [h]
+171° (c 2.0, water), lit.
D
[h]2D0 +140° [13]; 1H NMR (Me2SO-d6): l 8.75
(bs, 3 H, NH+3 ), 6.20 (d, 1 H, J1,2 3.4 Hz,
H-1), 5.25 (t, 1 H, J3,4 10 Hz, H-3)1, 5.00 (t, 1
H, J4,5 10 Hz, H-4)1, 4.18 (dd, 1 H, J6a,6b12
Hz, H-6a), 4.13 (ddd, 1 H, J5,6a 3.6 Hz, H-5),
3.97 (dd, 1 H, J6a,5 2 Hz, H-6b), 3.86 (dd, 1 H,
J2,3 10 Hz, H-2), 2.18, 2.03, 1.99 and 1.97 (4 s,
12 H, 4 Ac).
D
D
ranose (150 mg, 1 equiv), and triethylbenzyl-
ammonium chloride (65 mg, 0.5 equiv) were
dissolved in abs CHCl3 (8 mL) and the solu-
tion was half evaporated. The mixture was
then boiled under reflux for 18 h with reaction
process monitoring by TLC using a 10:0.5
(v/v) mixture of CHCl3–EtOH. After comple-
tion, the solvent was removed and the product
was chromatographed on a column of silica
gel using CHCl3“mixture B, followed by
crystallisation from a CHCl3–Et2O mixture,
Benzyl-2-acetamido-3,4,6-tri-O-acetyl-2-de-
oxy-i- -glucopyranoside (4)
D
Method (a). 2-Acetamido-3,4,6-tri-O-acetyl-
2-deoxy-a- -glucopyranosyl chloride 1 (200
D
mg, 0.55 mmol) and triethylbenzylammonium
chloride (62 mg, 0.5 equiv) were dissolved in
abs CHCl3 (6 mL). Then 3 mL of CHCl3 was
evaporated to remove trace water. Absolute
benzyl alcohol (71 mL, 1.2 equiv) was added to
the cooled solution and the mixture was
boiled under reflux for 15 h. The reaction was
monitored by TLC with 10:0.5 (v/v) CHCl3–
EtOH. After completion (:15 h), the solvent
was evaporated. The dry residue was chro-
matographed on a column of silica gel using
CHCl3“mixture A, followed by crystallisa-
tion from a CHCl3–Et2O mixture to give 4
to give 5 (240 mg, 71%): mp 100–101 °C, lit.
30
99–102 °C [15]; [h]
−71° (c 1.4, CH2Cl2),
546
1
lit. [h]D −66° [15]; H NMR (CDCl3): l 5.53
(d, 1 H, JNH,2% 9 Hz, NH), 5.47 (d, 1 H, J1,2
Hz, H-1), 5.06 (m, 2 H, H-3%+H-4%), 4.64
(d, 1 H, J1%,2% 9 Hz, H-1%), 4.51 (dd, 1 H, J3,2
4
2
Hz, H3), 4.25 (dd, H-2), 4.20 (dd, J6a%,5%
4.5, J6a%,6b% 13 Hz, H-6a%), 4.07 (dd, J4,5 1.8 Hz,
J4,3 8 Hz, H-4), 4.06 (dd, J6b%,5% 2.4 Hz, H-6b%),
3.93 (m, 2 H, H-2%+H-5%), 3.91 (dd, 1 H, J6a,6b
12.4 Hz, J6a,5 2.4 Hz, H-6a), 3.67 (dd, 1 H,
J6b,5 8.8 Hz, H-6b), 3.62 (ddd, 1 H, H-5),
1 The assignment can be inverse.

S.S. Pertel et al. / Carbohydrate Research 329 (2000) 895–899
899
pletion of the reaction and cooling of the
solution, the DCU was removed by filtration.
The product was obtained as described above,
to afford 6 (150 mg, 68%).
2.02, 1.96, 1.95 and 1.90 (4 s, 12 H, 4 Ac),
1.44, 1.38 and 1.25 (3 s, 12 H, 2 Me2C).
Tert-butyl-2-acetamido-3,4,6-tri-O-acetyl-2-
deoxy-i-
Method (a). 2-Acetamido-3,4,6-tri-O-acetyl-
2-deoxy-a- -glucopyranosyl chloride 1 (200
D
-glucopyranoside (6)
D
Acknowledgements
mg, 0.55 mmol) and triethylbenzylammonium
chloride (25 mg, 0.2 equiv) were dissolved in
abs CHCl3 (6 mL) and the solution was half
evaporated. Tert-butanol (102 mL, 2 equiv)
was added to the solution, and the mixture
was boiled under reflux for 12 h. More tert-
butanol (102 mL) was then added to the reac-
tion mixture and the refluxing was continued
for a further 12 h. The reaction was moni-
tored by TLC using a 10:0.5 (v/v) mixture of
CHCl3–EtOH. The solution was evaporated
to dryness. The dry residue was then chro-
matographed on a column of silica gel using
CHCl3“mixture A, followed by crystallisa-
The authors are grateful to Michael Songer
for his considerable assistance in preparing the
text of the present paper.
References
[1] S.E. Zurabyan, A.Ya. Khorlin, Russ. Chem. Re6., 43
(1974) 887–904.
[2] R. Kaifu, T. Osawa, Carbohydr. Res., 52 (1976) 179–185.
[3] A.F. Bochkow, G.E. Zaikov, Chemistry of the O-Glyco-
sidic Bond: Formation and Clea6age, Pergamon, Oxford,
1979, p. 48.
[4] S.S. Pertel, A.L. Kadun, V.Ya. Chirva, Bioorgan. Khim.,
21 (1995) 226–229.
[5] W.L. Salo, H.G. Fletcher, Jr., J. Org. Chem., 34 (1969)
3189–3191.
[6] S. David, A. Veyrieres, Carbohydr. Res., 40 (1975) 23–29.
[7] R.U. Lemieux, H. Driguez, J. Am. Chem. Soc., 97 (1975)
4063–4069.
[8] R. Kaifu, T. Osawa, R.W. Jeanloz, Carbohydr. Res., 40
(1975) 111–117.
[9] S. David, A. Veyrieres, Carbohydr. Res., 40 (1975) 23–29.
[10] M.A. Nashed, M. Kiso, C.W. Slife, L. Anderson, Carbo-
hydr. Res., 90 (1981) 71–82.
[11] T. Ogawa, K. Beppu, S. Nakabayashi, Carbohydr. Res.,
93 (1981) C6–C9.
tion from a CHCl3 –Et2O mixture, to afford 6
30
(122 mg, 55%): mp 169–171 °C; [h] −2° (c
546
1
1.5, CH2Cl2); H NMR (CDCl3): l 5.52 (d, 1
H, JNH,2 8.4 Hz, NH), 5.39 (dd, 1 H, J3,4 9.2
Hz, H-3), 4.90 (dd, 1 H, J4,5 9.8 Hz, H-4), 4.88
(d, 1 H, J1,2 8.4 Hz, H-1), 4.17 (dd, 1 H, J6a,6b
12.2 Hz, H-6a), 4.00 (dd, 1 H, J6b,5 2.6 Hz,
H-6b), 3.67 (ddd, 1 H, J5,6a 5.8 Hz, H-5), 3.51
(ddd, 1 H, J2,3 10.6 Hz, H-2), 1.99, 1.965, 1.96
and 1.86 (4 s, 12 H, 4 Ac), 1.16 (s, 9 H,
C(CH3)3).
Method (b). The synthesis of the glycoside 6
was carried out as described in method (a). In
addition, DCU (61 mg, 0.5 equiv) was intro-
duced into the reaction medium. After com-
[12] J. Dahmen, G. Gnosspelius, A.-C. Larsson, T. Lave, G.
Noori, K. Pa˚lsson, Carbohydr. Res., 138 (1985) 17–28.
[13] D. Horton, Ad6. Carbohydr. Chem., 15 (1960) 159–200.
[14] H.H. Zu, L.N. Congson, Can. J. Chem., 70 (1992) 2607–
2617.
[15] P. Rollin, P. Sinay, J. Chem. Soc., Perkin Trans. 1, (1977)
2513–2517.
.