N,N-Diacetylsialyl chloride-a novel readily accessible sialyl donor in reactions with neutral and charged nucleophiles in the absence of a promoter

Article abstract of DOI:10.1016/j.carres.2010.01.005

N,N-Diacetylneuraminic acid glycosyl chloride was prepared for the first time and made to react with various nucleophiles to give the corresponding α-glycosyl phosphate, β-glycosyl dibenzyl phosphate, α-glycosyl azide, α-phenyl thioglycoside and α-glycosyl xanthate in 65-82% yields and high stereoselectivity while its reactions with simple alcohols were not stereoselective. The new sialyl donor made possible the first stereoselective synthesis of sialic acid glycosyl phosphate with α-configuration and highly efficient synthesis of β-configured sialic acid glycosyl dibenzyl phosphate.

Full text of DOI:10.1016/j.carres.2010.01.005

Carbohydrate Research 345 (2010) 721–730
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
N,N-Diacetylsialyl chloride—a novel readily accessible sialyl donor in
reactions with neutral and charged nucleophiles in the absence of a promoter
*
Anna V. Orlova, Anna M. Shpirt, Nadezhda Y. Kulikova, Leonid O. Kononov
N. D. Zelinsky Institute of Organic Chemistry of the Russian Academy of Sciences, Leninsky Prosp., 47, 119991 Moscow, Russian Federation
a r t i c l e i n f o
a b s t r a c t
Article history:
N,N-Diacetylneuraminic acid glycosyl chloride was prepared for the first time and made to react with var-
Received 12 October 2009
Received in revised form 18 December 2009
Accepted 13 January 2010
Available online 18 January 2010
ious nucleophiles to give the corresponding
cosyl azide, -phenyl thioglycoside and -glycosyl xanthate in 65–82% yields and high stereoselectivity
while its reactions with simple alcohols were not stereoselective. The new sialyl donor made possible the
a-glycosyl phosphate, b-glycosyl dibenzyl phosphate, a-gly-
a
a
first stereoselective synthesis of sialic acid glycosyl phosphate with a-configuration and highly efficient
synthesis of b-configured sialic acid glycosyl dibenzyl phosphate.
Keywords:
Ó 2010 Elsevier Ltd. All rights reserved.
Glycosylation
Sialic acids
Stereoselectivity
Glycosyl phosphates
Glycosides
1. Introduction
with simple alcohols and negatively charged nucleophiles, a niche
where it surpasses other types of sialyl donors.2a
Sialic acid-containing glycoconjugates are involved in a wide
range of cell-surface recognition phenomena in living systems.1
For this reason, tremendous efforts have been made in order to de-
velop efficient methods for the synthesis of sialo-oligosaccharides.2
Thioglycoside derivatives of neuraminic acid with modified sub-
stituents at the nitrogen atom N(5), N,N-diacetyl-derivative being
the first of them,2a have recently been demonstrated to be superior
as sialyl donors in comparison with more traditional N-acetylneu-
raminic acid (Neu5Ac) glycosyl donors in terms of reactivity and
stereoselectivity.2c,d However, the reasons for success of these
modified sialyl donors still remain obscure. According to the con-
cept of supramer-mediated reactivity, which has recently been
introduced by us,3 different behaviour of N-acetyl- and N,N-diace-
tyl-sialic acid derivatives in glycosylation reactions is related to the
differences in structures of supramers formed in the solutions of
the respective sialyl donors.3a,b
2. Results and discussion
The title compound 2b was easily prepared (Scheme 1) in quan-
titative yield by treatment of the known peracetylated Neu5Ac2
glycosyl acetate 1b5 with HCl in CH2Cl2 (HCl was generated
in situ from AcCl and MeOH4b as described in the most convenient
procedure4d for the preparation of 2a). The chloride 2b was found
to be fairly stable at ambient temperature (it can be stored for sev-
eral months at ꢀ18 °C) and can be prepared on a multigram scale
without any problem.
As the first step, we studied reactions of 2b with simple alcohols
in the absence of any added promoter (see Table 1 and Scheme 2)
since it has earlier been shown4c that Neu5Ac glycosyl chloride
2a does react with neat MeOH within 1 h to give Neu5Ac
a-methyl
In order to gain deeper understanding of the influence of the
remote substituent at N(5) on the reactivity of sialyl donors we
set out to prepare previously unknown N,N-diacetylneuraminic
acid (Neu5Ac2) glycosyl chloride 2b (Scheme 1) and study its reac-
tivity in glycosylation reactions with various nucleophiles using
the known Neu5Ac glycosyl chloride 2a4 as a reference. Unlike sia-
lic acid thioglycosides, which are widely used in oligosaccharide
synthesis, sialyl chloride 2a has found application only in reactions
OAc
OAc
OAc
O
OAc
O
Cl
OAc
CO2R
HCl
AcO
AcN
AcO
AcN
CO2R
CH2Cl2
AcO
AcO
X
X
a : X = H, R = Me4
b : X = Ac, R = Me
c : X = H, R = H4c
1a,b,c
2a,b,c : ~100%
* Corresponding author. Tel.: +7 499 1377570; fax: +7 499 1376148.
E-mail addresses: kononov@ioc.ac.ru, leonid.kononov@gmail.com (L.O. Kononov).
Scheme 1. Synthesis of sialyl chlorides 2.
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2010.01.005

722
A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
Table 1
Results of reactions of glycosyl chlorides 2a and 2b with alcohols ROH
Entry
ROH
Glycosyl donor
Reaction time (h)
Yield of glycoside
Anomeric ratio (a:b)
Products
1
2
3
4
5
6
7
8
MeOHa
MeOHb
MeOHa
MeOHb
AllOHa
AllOHb
AllOHa
AllOHb
2a
2a
2b
2b
2a
2a
2b
2b
1
3
96%c,d
ꢁ30:1d
12:1
11:1
3:1
6:1
3:1
Only 7a
Only 7a
Only 7b
Only 7b
100%c
0.5
3
24
24
24
24
96%c
100%c
77%e
8a + 5a (2%e)
8a + 5a (4%e)
8a + 8b
89%e
47%c (8a) + 29%c (8b)
19%c (8a) + 58%c (8b)
1.3:1 (8a) + 1.9:1 (8b)
1.4:1 (8a) + 1.4:1 (8b)
8a + 8b
a
b
c
Reaction was performed in neat alcohol.
Reactions were performed in MeCN as the solvent.
Isolated yields.
d
e
See Ref. 4c.
Yields were calculated by taking into account signals integrals in 1H NMR spectra and the masses of the worked-up reaction mixtures.
OAc
O
P
OAc CO2MeO
P
OAc
AcO
AcN
OBn
OH
O
OAc O
O
O
O– Bu4N+
OAc
OBn
CO2Me
3a : only β, 72%4d
3b : only β, 74%
OAc
O
AcO
AcO
AcN
X
CO2Me
AcO
4a : α:β = 1:1.6, 59%
AcO
AcN
X
4b : α:β = 5:1, 60%
X
OAc
O
5a : 95%16a
5b : 92%
O
P
OBn
P
OH
–O
i-Pr2EtNH+
OH
OBn
–O
Na2HPO4
Bu4N+
OAc
OAc
OAc
OAc
O
Cl
OAc
O
OAc CO2Me
OR
CO2Me
N3
NaN3
ROH
AcO
AcN
AcO
AcN
AcO
AcN
CO2Me
O
1M NaHCO3
Bu4NHSO4
AcO
AcO
AcO
X
2a b a : X = H
,
X
X
b : X = Ac
7a,b : R = Me
8a,b : R = All
6a : only α, 94%10b
6b : only α, 82%
(see Table 1 for details)
K+ –
PhSH 1M NaHCO3
Bu4NHSO4
OEt
S
S
OAc
OAc
OAc CO2Me
OAc CO2Me
SPh
AcO
AcN
S
OEt
O
AcO
AcN
O
AcO
AcO
X
S
X
9a : only α, 80%10a
9b : α:β = 14:1, 91%
10a : α:β = 16:1, 72%8a
10b : only α, 71% (acetone)
76% (EtOH)
Scheme 2. Reactions of sialyl chlorides 2a,b with nucleophiles.
glycoside 7a4c in high yield and stereoselectivity (entry 1 in Ta-
ble 1). Based on the known higher reactivity of Neu5Ac2 thioglyco-
sides2d we expected that Neu5Ac2 glycosyl chloride 2b would be
more reactive than Neu5Ac glycosyl chloride 2a. Indeed, the reac-
tion of the chloride 2b with neat MeOH was complete within 0.5 h,
methyl glycoside 7b2e being the only product. However, consider-
which was isolated in 77% yield after additional O-acetylation
(treatment with Ac2O/Py was required since de-O-acetylation is
known to take place under the reaction conditions4c). Again, stere-
oselectivity decreased when reaction was performed in MeCN as
the solvent (entries 5 and 6 in Table 1). It is interesting to note that
similar reactions of Neu5Ac2 glycosyl chloride 2b with AllOH were
almost not stereoselective (entries 7 and 8 in Table 1) and resulted
in substantial de-N-acetylation leading to formation of N-acetyl-
sialoside 8a along with the expected N,N-diacetyl-sialoside 8b. This
additional instability of N-acetyl group in 2b/8b under reaction
conditions makes the use of glycosyl chloride 2b in reaction with
AllOH less practical than the use of more traditional glycosyl chlo-
ride 2a.
able amounts of b-anomer were formed in addition to
a-methyl
glycoside 7b (entry 3 in Table 1). The stereoselectivity decreased
even further and to a greater extent for 2b than for 2a when the
reactions with MeOH were performed in MeCN as the solvent (en-
tries 2, 4 in Table 1).
The parent N-acetylchloroneuraminic acid 2c corresponding to
the methyl ester 2a is known4c to react with neat allyl alcohol (Al-
lOH) in the absence of added promoter to give the respective allyl
Thus, although the yields of glycosides were higher than 76% in
all cases (Table 1) the stereoselectivities of the glycosylations of
simple alcohols with Neu5Ac2 glycosyl chloride 2b was lower than
those observed in glycosylations with Neu5Ac glycosyl chloride 2a
glycoside in high yield and stereoselectivity (
2a also reacted with neat AllOH to give after 24 h the respective al-
lyl glycoside 8a6 obtained as the mixture of anomers (
:b = 6:1),
a:b ꢁ 30:1). Chloride
a

A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
723
in all cases studied. This is especially clear for the reactions of 2a
and 2b with MeOH (entries 1–4 in Table 1) when the reaction
times were small. Reactions of 2a and 2b with AllOH (entries 5–8
in Table 1) were considerably slower and accompanied by the
HCl-catalyzed anomerization, which is known4c to contribute
greatly to the stereochemical outcome of unpromoted sialylation
at prolonged reaction times. This lower stereoselectivity of sialyla-
tion of simple alcohols by Neu5Ac2 glycosyl chloride 2b is espe-
cially striking considering much higher stereoselectivities
obtained in glycosylations of carbohydrate glycosyl acceptors with
Neu5Ac2 thioglycosides.2a,d,e,7 Thus the use of Neu5Ac2 glycosyl
chloride 2b in reactions with simple alcohols offers no practical
advantages over more common Neu5Ac glycosyl chloride 2a.
The next stage was to study the reactions of Neu5Ac2 glycosyl
chloride 2b with negatively charged nucleophiles (Scheme 2)
which are known2a to react with the parent Neu5Ac glycosyl chlo-
ride 2a smoothly to give the corresponding substitution products
stereoselectively.
glycosylation. Glycosylation of dibenzyl phosphoric acid ((BnO)2-
P(O)OH) with Neu5Ac glycosyl phosphite gave compound 3a in
53% yield12a while the attempted12b glycosylation of (BnO)2P(O)OH
with Neu5Ac glycosyl chloride 2a in the presence of AgOTf failed.
Stereoselective formation of b-phosphate 3b, that is, overall
retention of anomeric configuration, was apparently caused by
the in situ anomerization of the
a-phosphate, which was formed
initially in the reaction medium in an SN2-like process proceeding
with inversion of anomeric configuration. This is not surprising as
the b-isomers of Neu5Ac are known to be thermodynamically more
stable2a and dibenzyl phosphoric acid residue is a fairly good leav-
ing group (good nucleofuge). Anomerization of equatorial glycosyl
phosphates to the more thermodynamically stable isomers with
axial phosphate moiety is well known.11 Moreover, the
ured Neu5Ac and Neu5Ac2 dibenzyl phosphates ( -3a,b), prepared
a-config-
a
by a different route, were found to gradually isomerize upon stor-
age to the corresponding b-anomers.13 The ease of such anomeriza-
tion is related to the nucleofugality of phosphate anion (which is
proportional to the acidity of the conjugate phosphoric acid). For
this reason one can expect that the use of a phosphate nucleophile
with lower nucleofugality, hence lower tendency to anomerization
The reaction of Neu5Ac glycosyl chloride 2a with O-ethyl-S-
potassium dithiocarbonate (KSCSOEt) has been reported to give
the mixture of anomers of sialyl xanthate 10a (72%,
contaminated with inseparable glycal 5a (24%) in acetone and pure
a
:b = 16:1)8a
of glycosyl phosphate, may result in more a-selective reaction.
a
-xanthate 10a in anhydrous EtOH (71%)8b or under phase transfer
Indeed, a reaction of Neu5Ac glycosyl chloride 2a with
catalysis (PTC9) conditions (72% after crystallization8c). When
[Bu4N]H2PO4,14 which is a source of nucleophile (H2PO4ꢀ) with
lower nucleofugality than (BnO)2P(O)Oꢀ, in MeCN was reported15
to give Neu5Ac glycosyl phosphate 4a in 39% yield as a mixture
Neu5Ac2 glycosyl chloride 2b was treated with KSCSOEt in acetone
or in 95% EtOH a-sialyl xanthate 10b was formed in 71% and 76%
yields, respectively. It is worth mentioning that the yield of
xanthate 10b was slightly lower in acetone, incapable of reacting
with sialyl chloride 2b, than in ethanol, which might be able to
react with sialyl chloride 2b to give the corresponding ethyl
glycoside. We have never performed such a reaction since the
use of ethyl sialoside offers no advantages over the use of methyl
sialoside while lacking valuable features of allyl sialosides. By
analyzing Table 1 one can surmise that within 1 h a noticeable
amount of ethyl sialoside would be formed. However, no competi-
tion between S- and O-nucleophiles has ever been observed in
these reactions. Similar to the published reaction of sialyl chloride
2a with KSCSOEt in ethanol8b no formation of ethyl sialoside could
be detected in the case of 2b under similar conditions although a
mixture of glycosyl chloride 2b and KSCSOEt in ethanol was al-
lowed to react for 1 h at ambient temperature.
of both anomers (a:b = 2.3:1). Further research revealed that the
amount of b-isomer was in fact higher than had been reported ini-
tially and the phosphate 4a could be obtained in 59% yield
(a
:b = 1:1.6). A similar reaction of Neu5Ac2 glycosyl chloride 2b
with [Bu4N]H2PO4, resulted in formation of Neu5Ac2 glycosyl phos-
phate 4b in 60% yield and much higher -stereoselectivity
:b = 5:1). This is the first example of stereoselective synthesis of
a sialic acid glycosyl phosphate with -configuration. A related
Neu5Ac phenyl phosphate with -configuration has been isolated
as the minor product (
a
(a
a
a
a
:b ꢃ 1:6) upon reaction of Neu5Ac glycosyl
phosphite with phenyl phosphoric acid (PhOP(O)(OH)2) at
ꢀ40 °C,12a which is not surprising considering higher acidity of
PhOP(O)(OH)2 than that of H3PO4. Hence sialyl phenyl phosphate
is more prone to anomerization (
a?b) than sialyl phosphates
4a,b, resulting in lower /b ratio of anomeric phosphorylation.
a
The reactions of Neu5Ac glycosyl chloride 2a with PhSH and
Reactions of both chlorides 2a,b with [Bu4N]H2PO4 in MeCN
were also found to give phosphodiesters 11a,b as by-products
(yields 11% and 4%, respectively), which could be formed in the
reaction of glycosyl chlorides 2a,b with phosphomonoesters 4a,b
(Scheme 3).
NaN3 under PTC conditions have been reported to give phenyl
a-
thiosialoside 9a10a and -glycosyl azide 6a10b in 80% and 94%
a
yields, respectively. Application of PTC conditions (Bu4NHSO4,
AcOEt, 1 M NaHCO3)8c,10 for reactions of 2b with PhSH or NaN3 re-
sulted in clean formation of phenyl thiosialoside 9b (91%,
Neu5Ac glycosyl chloride 2a is known16a to react with much
more basic Na2HPO4 (a disubstituted salt of phosphoric acid) in
refluxing MeCN (no reaction occurs at ambient temperature) to
give exclusively an elimination product, Neu5Ac glycal 5a,16 in
quantitative yield. No product of nucleophilic substitution could
be detected. When Neu5Ac2 glycosyl chloride 2b was involved in
the reaction with Na2HPO4 under identical conditions, only
Neu5Ac glycal 5a, lacking one of the two N-acetyl groups present
a
:b = 14:1) and
a-glycosyl azide 6b (82%).
Complete inversion of anomeric configuration during these
reactions suggests SN2-like mechanism of chlorine substitution in
both sialyl chlorides 2a and 2b (see the discussion of the mecha-
nism below). The products of these reactions (glycosyl azides
6a,b, thioglycosides 9a,b and glycosyl xanthates 10a,b) are config-
urationally stable under the reaction conditions. The preparation of
glycosyl phosphates is a more demanding task due to possibility of
product anomerization.11
The reaction of Neu5Ac glycosyl chloride 2a with ethyldi(iso-
OAc
propyl)ammonium dibenzyl phosphate (½Pri NHEtꢂ½ðBnOÞ PO2ꢂÞ in
O
P
OAc CO2Me
2
2
MeCN has recently been reported4d to give the corresponding b-
phosphate 3a in 72% yield albeit contaminated with inseparable
N-acetyl-glycal 5a. A similar reaction of Neu5Ac2 glycosyl chloride
2b with ½Pri NHEtꢂ½ðBnOÞ PO2ꢂ in MeCN smoothly gave b-config-
O–
AcO
AcN
Bu4N+
O
O
AcO
X
2
2
2
ured Neu5Ac2 dibenzyl phosphate 3b in 74% yield along with the
corresponding glycal 5b (19%), which could easily be separated
by silica gel chromatography. This is the highest yield of sialic acid
glycosyl dibenzyl phosphate with b-configuration ever obtained by
11a : X = H, 11%
11b : X = Ac, 4%
Scheme 3. Structures of bis(sialyl) phosphates 11a,b.

724
A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
in 2b, was formed in quantitative yield after 3 h rather than the ex-
pected Neu5Ac2 glycal 5b. We found that the rate of de-N-acetyla-
tion is temperature dependent. Decreasing temperature resulted in
the formation of mixtures of 5a with the increasing amounts of
Neu5Ac2 glycal 5b,17 which could be prepared in pure form (92%
yield) when reaction was performed at ambient temperature
although the overall reaction rate considerably decreased (full con-
version of 2b was achieved only after 1 month). Further experi-
ments demonstrated that the N-acetyl group of Neu5Ac2 residue
could be removed only from N(5) of glycosyl chloride 2b. Glycal
5b was stable under the reaction conditions. Other sialic acid
derivatives with two N-acetyl groups (e.g., 1b or 9b) did not react
with Na2HPO4 either.
As can be seen from the results obtained, stereoselectivities
achieved in glycosylation reactions with Neu5Ac glycosyl chloride
2a or Neu5Ac2 glycosyl chloride 2b are dramatically dependent on
the nature of nucleophile used and therefore need commenting.
First of all, we have to stress that glycosylation with both sialyl
chlorides 2a and 2b leading to products, which are configuration-
ally stable under the reaction conditions (i.e., glycosyl azides
6a,b, thioglycosides 9a,b and glycosyl xanthates 10a,b), always
proceeded with inversion of configuration of anomeric centre
carbenium ions generated from glycosyl donors.18 Remarkably,
nucleophilic substitution at anomeric position involving SN1-like
mechanism would inevitably lead to a considerable loss of stere-
oselectivity. This situation indeed occurs during sialylation with
sialic acid thioglycosides where it is very difficult to achieve sub-
stantial level of stereocontrol (moderate stereoselectivities of
a
:b ꢁ 7:1 are quite common even in participating nitrile solvents).2
An important feature of these reactions is that stereoselectivity of
sialylation with thioglycosides usually does not depend on the ano-
meric configuration of the starting thioglycoside.2
Consideration of intermediate preassociation mechanism
(DN ꢄ ANint),19 which involves formation of intimate ion-pair inter-
mediate,20 could allow interpretation of the observed stereoselec-
tivities (including inversion) during sialylation with sialyl
chlorides 2a,b. However, high a-stereoselectivities in nucleophilic
substitution in sialyl chlorides 2a,b are observed even in polar sol-
vents such as acetonitrile or alcohols, where the dominant reaction
of ion-pairs is their separation to free ions.20
These considerations suggest the involvement of SN1-like
mechanism (DN + AN or DN ꢄ ANint) to be highly unlikely in unpro-
moted reactions of sialyl chlorides 2a,b and make SN2-like mecha-
nism (ANDN) of substitution fairly feasible. It would be very
interesting to study the mechanistic details of nucleophilic substi-
tution in sialyl chlorides 2a,b both theoretically and
experimentally.
and resulted in the products with a-configuration independent of
the number of acetyl groups at N(5). This observation is very
important and suggests that initially reactions of sialyl chlorides
2a and 2b with nucleophiles proceed as SN2-like processes leading
to a-configured products, which may further experience an in situ
3. Conclusions
anomerization leading to b-configured products. The possibility of
this anomerization greatly depends on the nucleofugality of agly-
con moiety and reaction conditions. Currently there is no generally
accepted rationale how the second acetyl group at N(5) in glycosyl
chloride 2b can influence the outcome of the reactions at anomeric
position at (C(2)) including glycosylation and anomerization steps
(cf. Refs.3a,3b). At the moment it is only clear that the acid-catalyzed
anomerization (HCl is formed during these promoter-free glycosy-
lations) of alkyl sialosides 7a,b and 8a,b is favored in the case of
Neu5Ac2 glycosyl derivatives, especially in more polar MeCN. On
the contrary, the presence of an additional N-acetyl group in sialyl
phosphate 4b makes this compound less prone to anomerization
than the 5-acetamido counterpart 4a under the reaction condi-
tions. It is important to note that the anomeric phosphorylation
leading to glycosyl phosphates 4a,b proceeds with no visible
change in the acidity of reaction medium probably due to the buf-
fering effect of the excess of phosphate salt used as a source of
nucleophile. For this reason one can not exclude that different
pathways for anomerization18 may be operative in these two cases.
The most amazing feature of sialyl chlorides 2a,b is their pro-
found ability to give substitution products with inverted anomeric
configuration (see the discussion above). Inversion of configuration
is usually anticipated for an SN2-like reaction, an associative mech-
anism, featuring concomitant attack of a nucleophile and loss of
the aglycon (ANDN in IUPAC notation19). However, considering
the tertiary structure of the anomeric centre in sialyl chlorides, this
is generally assumed to be a very unlikely scenario. Although the
anomeric carbon of NeuAc has three non-hydrogen substituents
and is believed to be somewhat hindered, it is very important that
ANDN-like chemistry has been identified for NeuAc in the case of
the trans-sialidase enzyme, which can utilize this pathway, despite
the bulky environment at the anomeric centre.18b
In conclusion, a novel sialyl donor N,N-diacetylneuraminic acid
glycosyl chloride 2b was prepared for the first time and its reactions
with nucleophiles were studied. Neu5Ac2 glycosyl chloride 2b
showed its best in reactions with charged nucleophiles and made
possible the first stereoselective synthesis of sialic acid glycosyl
phosphate with a-configuration and highly efficient synthesis of b-
configured sialic acid glycosyl dibenzyl phosphate. The compounds
prepared from 2b have potential for wide applications and some of
them have already been used. Thioglycoside 9b, xanthate 10b and
dibenzyl phosphate 3b are sialyl donors themselves while azide 6b
may be useful for the preparation of glycosyl amines and in ‘click
chemistry’-based approaches21 for the construction of simple
glycoside and oligosaccharide mimetics, glyco-macrocycles, glyco-
peptides, glyco-clusters and carbohydrate arrays. The novel
a-con-
figured glycosyl phosphate -4a has already been used in the first
a
synthesis of sialic acid polyprenyl glycosyl phosphate, a probable
biosynthetic intermediate of bacterial polysialic acid.22
4. Experimental
4.1. General methods
The reactions were performed with the use of commercial re-
agents (Aldrich, Fluka, Acros Organics) and distilled solvents puri-
fied according to standard procedures. Thin-layer chromatography
was carried out on Silica Gel 60 F254 plates on aluminium foil
(Merck), spots were visualized under UV light and by heating the
plates after immersion in a 1:10 (v/v) mixture of 85% aqueous
H3PO4 and 95% EtOH. The 1H and 13C NMR spectra of solutions in
CDCl3 were recorded on a Bruker AC-200 instrument (200.13,
and 50.32 MHz, respectively) and a Bruker AVANCE 600 instru-
ment (600.13 and 150.9 MHz, respectively) while 31P NMR spectra
were recorded on a Bruker AC-200 instrument (81.02 MHz). The 1H
chemical shifts are referred to the signal of the residual CHCl3 (dH
7.27 ppm), and the 13C of the 13C NMR—to the signal of CDCl3 (dC
77.0 ppm) while 31P chemical shifts of 31P NMR—to the signal of
75% H3PO4 in D2O (dP 0.0 ppm, external standard). The assignment
An alternative SN1-like mechanism (DN + AN according to IU-
PAC19) would lead to dissociative loss of the aglycon (chloride in
our case), first producing a solvent-equilibrated glycosyl oxocarbe-
nium ion, followed by nucleophilic capture. This mechanism is
considered to be quite common in nucleophilic substitution reac-
tions at tertiary centres (like in tert-butyl chloride) especially when
the respective carbenium ion is stabilized like in the case of oxa-

A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
725
of the signals in the 1H and 13C NMR spectra was made using
1H,1H-COSY 2D NMR and APT (JMODXH) experiments, respec-
tively. Mass spectra (electrospray ionization, ESI) were recorded
on a Finnigan LCQ mass spectrometer for 2 ꢅ 10ꢀ5 M solutions.
The resulting solution was added to a cold solution of glycosyl ace-
tate 1b (18.6 g, 32.1 mmol) in a mixture of anhydrous CH2Cl2
(200 mL) and AcCl (25 mL, 0.351 mmol) and the reaction mixture
was kept at +4 °C for 12 h (TLC control: Rf = 0.26 (2b), Rf = 0.23
(1b), AcOEt–petroleum ether, 1:1). Volatile components were
evaporated, CCl4 (ꢁ100 mL) was added, and volatile components
were again evaporated, the addition of CCl4 and evaporation was
repeated (3–5 times) until no smell of HCl could be detected. The
residue was dried in vacuo (oil pump) to give glycosyl chloride
2b as the slightly yellow solid (17.66 g, ꢁ100%; Rf = 0.62 AcOEt;
Rf = 0.26, AcOEt–petroleum ether, 1:1), identical to that obtained
by the small scale procedure, which was used without additional
purification.
The optical rotation ([a]D) was measured on a JASCO DIP-360
polarimeter at 20–27 °C. Reactions of glycosyl chlorides 2a,b with
nucleophiles were carried out with the use of glycosyl chlorides
2a,b, which were freshly prepared as described below from the
fully acetylated neuraminic acid methyl esters 1a or 1b and dried
for 3 h in vacuum of an oil pump. The yields are quoted with re-
spect to the amount of starting glycosyl acetates 1a,b used for
the preparation of glycosyl chlorides 2a,b since the prepared chlo-
rides 2a,b may contain coordinated HCl.4b–d
4.2. Tetrabutylammonium dihydrogen phosphate
4.3.3. Data for methyl 4,7,8,9-tetra-O-acetyl-2-chloro-5-(N,N-
diacetylamino)-2,3,5-trideoxy-b-D-glycero-D-galacto-non-2-
The title compound was prepared essentially as described in
Ref. 14. Phosphoric acid (96% aqueous solution of H3PO4, 4 mL)
was added to deionized H2O (20 mL) followed by Bu4NOH (75 mL
of 20% aqueous solution, Merck) and the resulting mixture was
stirred for 18 h at room temperature (ꢁ20 °C). Then more 96%
ulopyranosonate (2b)
½
a 2D0
ꢂ
ꢀ37.8 (c 1.7, CHCl3); 1H NMR (200.13 MHz, CDCl3): dH 1.97,
2.01, 2.05, 2.08 (4 s, 3H each, 4 ꢅ AcO), 2.19 (dd, J3ax,4 = 10.4 Hz, 1H,
H-3ax), 2.30, 2.41 (2 s, 3H each, Ac2N), 2.89 (dd, J3eq,4 = 5.1 Hz,
J3eq,3ax = 13.8 Hz, 1H, H-3eq), 3.85 (s, 3H, OMe), 4.07 (dd,
J8,9b = 5.2 Hz, J9a,9b = 12.7 Hz, 1H, H-9b), 4.30 (dd, J8,9a = 2.6 Hz,
1H, H-9a), 4.31 (t, 1H, H-5), 5.18 (ddd, 1H, H-8), 5.25 (dd,
J6,7 = 2.1 Hz, J7,8 = 7.5 Hz, 1H, H-7), 5.41 (dd, J5,6 = 10.2 Hz, 1H, H-
6), 5.89 (ddd, J4,5 = 10.3 Hz, 1H, H-4); 13C NMR (54.24 MHz, CDCl3):
dC 20.6, 20.8 (OC(O)CH3), 26.1, 27.9 (2 ꢅ NC(O)CH3), 41.8 (C-3),
53.6 (OMe), 56.1 (C-5), 61.6 (C-9), 66.4, 66.7, 69.1, 71.3 (C-4, C-6,
C-7, C-8), 95.9 (C-2), 165.8 (C-1), 169.5, 169.9, 170.5, 173.3, 174.3
(C@O).
H3PO4 was added portion-wise (3 ꢅ 20
ll) with pH monitoring.
For the determination of pH, aliquots (1 mL) of the obtained opal-
escent solution were diluted with deionized H2O (9 mL) and then
pH of the resulting clear solution was measured with a Metrohm
744 pH-meter. After pH reached 5.19 the reaction mixture was
concentrated on a rotary evaporator (at +35 °C) and anhydrous
MeCN (310 mL, Merck, c(H2O) <0.3%) was added to the residue
and the mixture was magnetically stirred for 3 days and then the
white fine crystalline precipitate was filtered off. The filtrate was
concentrated on a rotary evaporator (at +40 °C) and the residue
was dried in vacuo (oil pump) for 1.5 h to give [Bu4N][H2PO4]
(20.72 g), which was stored in a desiccator over NaOH. This salt
was used in reactions without additional drying. 1H NMR
(200.13 MHz, CDCl3): dH 0.90 (t, J = 7.0 Hz, 12H, CH3CH2CH2CH2N),
1.40 (sextet, J = 7.0 Hz, 8H, CH3CH2CH2CH2N), 1.51–1.70 (m, 8H,
CH3CH2CH2CH2N), 3.10–3.40 (m, 8H, CH3CH2CH2CH2N); 1H NMR
(200.13 MHz, D2O) dH 0.80 (t, J = 7.5 Hz, 12 H, CH3CH2CH2CH2N),
1.31 (sextet, J = 7.5 Hz, 8H, CH3CH2CH2CH2N), 1.50–1.70 (m, 8H,
CH3CH2CH2CH2N), 3.09–3.22 (m, 8H, CH3CH2CH2CH2N); 13C NMR
(54.24 MHz, CDCl3): dC 13.6 (CH3), 19.4 (CH2), 23.8 (CH2), 58.3
(NCH2); 13C NMR (54.24 MHz, D2O) dC 13.7 (CH3), 20.0 (CH2),
30.0 (CH2), 58.9 (NCH2); 31P NMR (81.02 MHz, D2O): dP +0.6.
4.4. Methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-diacetylamino)-2-O-
di(benzyloxy)phosphoryl-3,5-dideoxy-b-D-glycero-D-galacto-
non-2-ulo-pyranosonate (3b)
Ethyldi(isopropyl)amine (0.066 mL, 0.37 mmol) was added to a
solution of (BnO)2P(O)OH (117 mg, 0.42 mmol) in MeCN (1 mL).
The solution of salt thus prepared was added dropwise to a solu-
tion of glycosyl chloride 2b (prepared from 1b (52.1 mg,
0.091 mmol)) in MeCN (2 mL). The flask, in which salt was pre-
pared, was additionally rinsed with MeCN (2 mL). The course of
the reaction was monitored by TLC (Rf = 0.59 (3b), Rf = 0.62 (2b),
AcOEt). After completion of the reaction (24 h), the reaction mix-
ture was cooled in an ice-water bath, and cold saturated NaHCO3
solution (10 mL) and CHCl3 (10 mL) were added. The aqueous
phase was extracted with CHCl3 (3 ꢅ 5 mL). The organic phase
was concentrated at ambient temperature. The residue was dis-
solved in CH2Cl2 and applied onto a silica gel HPLC column (Sila-
4.3. Methyl 4,7,8,9-tetra-O-acetyl-2-chloro-5-(N,N-
diacetylamino)-2,3,5-trideoxy-b-D-glycero-D-galacto-non-2-
ulopyranosonate (2b)
sorb 600, 7.5
l
m, 15 ꢅ 250 mm). The products were eluted with
gradient of AcOEt in petroleum ether (0?100%) to give 3b as a col-
4.3.1. Small scale procedure
ourless syrup (53.7 mg, 74%; Rf = 0.59, AcOEt) and glycal 5b
Anhydrous MeOH (0.9 mL, 0.02 mol) was slowly added drop-
wise to AcCl (2.5 mL, 0.035 mol) with cooling in an ice-water bath.
The resulting solution was added to a cold solution of glycosyl ace-
tate 1b (51.2 mg, 0.088 mmol) in a mixture of anhydrous CH2Cl2
(2.5 mL) and AcCl (2.5 mL, 0.035 mol) and the reaction mixture
was kept at +4 °C for 12 h (TLC control: Rf = 0.26 (2b), Rf = 0.23
(1b), AcOEt–petroleum ether, 1:1). Volatile components were
evaporated, CCl4 (5 mL) was added, volatile components were
again evaporated and the addition of CCl4 and evaporation was re-
peated five times. The residue was dried in vacuo (oil pump) to
give glycosyl chloride 2b as a white foam (53 mg, ꢁ100%), which
was used without additional purification.
(8.7 mg, 19%, Rf = 0.62, AcOEt). ½a D27
ꢂ
ꢀ4.0 (c 1, CHCl3); 1H NMR
(600.13 MHz, CDCl3): dH 1.986, 1.989, 2.00, 2.14 (4 s, 3H each,
4 ꢅ AcO), 2.14 (from COSY) (dd, 1H, H-3ax), 2.32, 2.40 (2 s, 3H each,
Ac2N), 2.896 (dd, J3eq,4 = 5.3 Hz, J3eq,3ax = 13.7 Hz, 1H, H-3eq), 3.75
(s, 3H, OMe), 4.22 (dd, J8,9b = 6.5 Hz, J9a,9b = 12.4 Hz, 1H, H-9b),
4.31 (dd, J4,5 = 10.3 Hz, J5,6 = 10.3 Hz, 1H, H-5), 4.47 (dd,
J8,9a = 2.7 Hz, 1H, H-9a), 5.10–5.20 (m, 4H, 2 ꢅ OCH2Ph), 5.242
(ddd, 1H, H-8), 5.27 (dd, J6,7 = 1.9 Hz, J7,8 = 5.4 Hz, 1H, H-7), 5.42
(dd, 1H, H-6), 5.93 (ddd, J3ax,4 = 11.0 Hz, 1H, H-4); 13C NMR
(150.9 MHz, CDCl3): dC 20.67, 20.73, 20.9 (OC(O)CH3), 26.0, 28.0
3
(2 ꢅ NC(O)CH3), 38.0 (d, JC,P = 4.2 Hz, C-3), 53.3 (OMe), 56.5 (C-
5), 62.0 (C-9), 66.2, 67.7, 70.2, 70.7 (C-4, C-6, C-7, C-8), 69.9 (d,
2
JC,P = 4.0 Hz, OCH2Ph), 99.9 (d, JC,P = 5.3 Hz, C-2), 128.1, 128.2,
4.3.2. Large scale procedure
Anhydrous MeOH (14 mL, 0.346 mol) was slowly added drop-
wise to AcCl (25 mL, 0.351 mol) with cooling in an ice-water bath.
3
128.5 (Ph), 135.54 (d, JC,P = 8.0 Hz, C-quat. Ph), 135.60 (d,
3JC,P = 7.7 Hz, C-quat. Ph), 165.9 (C-1), 169.4, 170.0, 170.2, 170.6,

726
A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
173.5, 174.3 (C@O); 31P NMR (81.02 MHz, CDCl3): dP ꢀ6.9; ESIMS:
found m/z 815.9 [M+Na], calcd for C36H44NNaO17P: 816.2.
1 H, H-5), 4.39 (dd, J8,9a = 7.1 Hz, J9a,9b = 12.3 Hz, 1H, H-9a), 4.64
(dd, J5,6 = 10.9, J6,7 = 2.1 Hz, 1 H, H-6), 4.77 (dd, J8,9b = 7.1 Hz,
J9a,9b = 12.3 Hz, 1H, H-9b), 4.90 (br d, JNH,5 ꢃ 10 Hz, 1H, NH),
5.24–5.54 (m, 3H, H-4, H-7, H-8); 13C NMR (54.24 MHz, CDCl3):
dC 13.7 (CH3), 19.6 (CH2), 20.92, 20.95, 21.02, 21.06 (OC(O)CH3),
23.0 (NC(O)CH3), 24.0 (CH2), 37.8 (C-3), 48.6 (C-5), 52.3 (OCH3),
58.7 (NCH2), 61.8 (C-9), 69.7, 70.7, 71.4, 72.0 (C-4, C-8, C-7, C-6),
95.8 (d, J = 5.7 Hz, C-2), 169.6 (C-1), 170.35, 170.42, 170.52,
170.61, 172.3 (C@O); 31P NMR (81.02 MHz, CDCl3): dP ꢀ3.8; ESIMS:
found m/z 569.9 [MꢀBu4N]ꢀ. Calcd for C20H29NO16P: 570.12. Found
m/z 1382.1 [M2ꢀBu4N]ꢀ. Calcd for C56H94NO16P: 1382.53.
4.5. Methyl 4,7,8,9-tetra-O-acetyl-5-acetamido-2-O-
dihydroxyphosphoryl-3,5-dideoxy-
2-ulo-pyranosonate, tetrabutylammonium salt (
4,7,8,9-tetra-O-acetyl-5-acetamido-2-O-dihydroxyphosphoryl-
3,5-dideoxy-b- -glycero- -galacto-non-2-ulo-pyranosonate,
tetrabutylammonium salt (b-4a); bis[methyl (4,7,8,9-tetra-O-
acetyl-5-acetamido-3,5-dideoxy- -glycero- -galacto-non-2-ulo-
a-
D
-glycero-
D
-galacto-non-
a-4a); methyl
D
D
D
D
pyranosyl)onate]phosphate, tetrabutylammonium salt (11a)
Glycosyl chloride 2a (prepared from 1a (100 mg, 0.19 mmol))
was dissolved in MeCN (2 mL) and solid [Bu4N]H2PO4 (284 mg,
0.84 mmol) was added. The reaction mixture was stirred for 4 h
(TLC control: Rf = 0.00 (4a, 11a), Rf = 0.48 (5a), Rf = 0.50 (2a), AcOEt;
4.5.3. Data for bis[methyl (4,7,8,9-tetra-O-acetyl-5-acetamido-
3,5-dideoxy-D-glycero-D-galacto-non-2-ulo-pyranosyl)onate]
phosphate, tetrabutylammonium salt (11a)
1H NMR (200.13 MHz, CDCl3, selected signals): dH 1.02 (t,
Rf = 0.30 (
a
-4a), Rf = 0.40 (b-4a), Rf = 0.60 (11a), Rf = 0.81 (5a),
J = 6.9 Hz,
12H,
CH3CH2CH2CH2N),
1.35–1.53
(m,
8H,
CHCl3–MeOH–H2O, 65:25:4). The reaction mixture was evaporated
at ambient temperature and suspended in deionized H2O (0.5 mL)
and the suspension was applied on a C18 cartridge (16 ꢅ 20 mm,
Mega Bond Elut C18 (Varian)). The cartridge was washed with
H2O (20 mL) and MeCN (40 mL). The acetonitrile phase was evap-
orated to give a mixture (133.4 mg) of glycosyl phosphate 4a, gly-
cal 5a and anomeric mixture of phosphodiesters 11a as a dark
yellow syrup (4a:5a:11a = 5.4:1.5:1). The yields of 4a (91.0 mg,
CH3CH2CH2CH2N), 1.54–1.76 (m, 8H, CH3CH2CH2CH2N), 1.86, 1.89
(2 s, 3H each, 2 ꢅ AcN), 2.03, 2.08, 2.11, 2.13 (4 s, 24H, 8 ꢅ AcO),
3.20–3.36 (m, 8H, CH3CH2CH2CH2N), 3.75, 3.81 (2 s, 3 H each,
2 ꢅ OMe); 31P NMR (81.02 MHz, CDCl3): dP ꢀ11.0, ꢀ12.5; ESIMS:
found m/z 1043.2 [M]ꢀ. Calcd for C40H56N2O28P: 1043.28.
4.6. Methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-diacetylamino)-2-O-
dihydroxyphosphoryl-3,5-dideoxy-
ulo-pyranosonate, tetrabutylammonium salt (
4,7,8,9-tetra-O-acetyl-5-(N,N-diacetylamino)-2-O-dihydroxyphos
phoryl-3,5-dideoxy-b- -glycero- -galacto-non-2-ulo-pyrano
sonate, tetrabutylammonium salt (b-4b); bis[methyl (4,7,8,9-
a-D-glycero-D-galacto-non-2-
a:b = 1:1.6, 59%; Rf = 0.30 (
a-4a) and 0.40 (b-4a), CHCl3–MeOH–
a-4b); methyl
H2O, 65:25:4), 5a (14.7 mg, 16%), and 11a (27.7 mg, 11%;
Rf = 0.60, CHCl3–MeOH–H2O, 65:25:4) were calculated by integra-
tion of the respective signals in the 1H NMR spectrum and taking
into account the molecular masses of 4a, 5a and 11a (812.88 and
473.42, 1286.31, respectively). Analytical samples of pure glycosyl
D
D
tetra-O-acetyl-5-(N,N-diacetylamino)-3,5-dideoxy-
galacto-non-2-ulo-pyranosyl)onate]phosphate,
tetrabutylammonium salt (11b)
D-glycero-D-
phosphates a-4a and b-4a were obtained from this mixture by re-
versed phase chromatography on a C18 cartridge (16 ꢅ 20 mm,
Mega Bond Elut C18 (Varian); gradient elution (0?100%) with
MeCN in deionized H2O).
Glycosyl chloride 2b (prepared from 1b (56.5 mg, 0.098 mmol))
was dissolved in MeCN (1 mL) and solid Bu4H2PO4 (150 mg,
0.442 mmol) was added. The reaction mixture was stirred for 2 h
at ambient temperature (20 °C) (TLC control: Rf = 0.00 (4b, 11b),
4.5.1. Data for methyl 4,7,8,9-tetra-O-acetyl-5-acetamido-2-O-
dihydroxyphosphoryl-3,5-dideoxy-
2-ulo-pyranosonate, tetrabutylammonium salt (
a
-
D
-glycero-
D
-galacto-non-
-4a)
Rf = 0.62 (2b), AcOEt; 65:25:4, Rf = 0.24 (b-4b), Rf = 0.32 (a-4b),
a
Rf = 0.68 (5b), Rf = 0.73 (11b), CHCl3–MeOH–H2O). The reaction
mixture was diluted with deionized H2O (500 mL). The obtained
milky solution was filtered through a C18 cartridge (16 ꢅ 20 mm,
Mega Bond Elut C18 (Varian)). The cartridge was washed with
deionized H2O (30 mL) and the combined filtrate was filtered again
through the same cartridge. The cartridge was washed with MeCN
(100 mL) and the acetonitrile phase was evaporated to give a mix-
ture (71 mg) of glycosyl phosphate 4b, glycal 5b and 11b as a dark
1H NMR (200.13 MHz, CDCl3): dH 0.96 (t, J = 7.2 Hz, 12H,
CH3CH2CH2CH2N), 1.30–1.51 (m, 8H, CH3CH2CH2CH2N), 1.53–1.74
(m, 8H, CH3CH2CH2CH2N), 1.82 (s, 3H, AcN), 1.97, 1.98 (2 s, 6H,
2 ꢅ AcO), 2.04, 2.06 (2 s, 6H, 2 ꢅ AcO), 2.01–2.25 (m ꢁ ‘t’, 1H, H-
3ax), 3.01 (dd, J3eq,4 = 4.9 Hz, J3eq,3ax = 13.3 Hz, 1 H, H-3eq), 3.23–
3.37 (m, 8 H, CH3CH2CH2CH2N), 3.73 (s, 3 H, OMe), 3.68–3.77 (m,
1H, H-6), 3.90–4.14 (m, 2 H, H-5, H-9a), 4.27–4.37 (m ꢁ ‘dd’,
J9a,9b ꢃ 12.4 Hz, J8,9b ꢃ 2.4 Hz, 1H, H-9b), 5.03–5.20 (m ꢁ ‘ddd’,
J3eq,4 = 4.9, J4,5 = 10.3, 1H, H-4), 5.15–5.30 (m, 2 H, H-7, H-8), 5.42
(d, J5,NH = 9.7 Hz, 1 H, NH); 13C NMR (54.24 MHz, CDCl3): dC 13.7
(CH3), 19.7 (CH2), 20.8, 21.1 (C(O)CH3), 23.2 (NC(O)CH3), 24.0
(CH2), 38.0 (br s, C-3), 49.7 (C-5), 52.6 (OCH3), 59.0 (NCH2), 62.2
(C-9), 67.6, 69.4, 69.9, 72.2 (C-4, C-8, C-7, C-6), 97.67 (d,
JC,P = 3.2 Hz, C-2), 169.2 (C-1), 170.1, 170.2, 170.4, 170.8, 171.1
(C@O); 31P NMR (81.02 MHz, D2O): dP ꢀ4.9; ESIMS: found m/z
538.1 [MꢀBu4N–OMe–H]ꢀ. Calcd for C19H25NO15P: 538.10. Found
m/z 1382.1 [M2ꢀBu4N]ꢀ. Calcd for C56H94NO16P: 1382.53.
yellow syrup
(50.5 mg, :b = 4.9:1, 60%) and 5b (14.9 mg, 30%), 11b (5.6 mg,
4%) were calculated by integration of the respective signals in
the 1H NMR (for 5b and ,b-4b) and 31P NMR (for
,b-4b and
(a,b-4b:5b:11b = 13.8:2.3:1). The yields of 4b
a
a
a
11b) spectra and taking into account the molecular masses of 4b,
5b and 11b (854.92, 515.46 and 1370.88, respectively). 1H NMR
(5b:
Analytical sample of pure glycosyl phosphate
with a trace amount of glycal ( -4b:5b = 24:1) was obtained from
a
-4b:b-4b = 2.9:4.9:1), 31P NMR (
a
-4b:b-4b:11b = 4.8:1:0.43).
a-4b contaminated
a
this mixture by reversed phase chromatography on a C18 cartridge
(16 ꢅ 20 mm, Mega Bond Elut C18 (Varian); gradient elution
(0?100%) with MeCN in deionized H2O).
4.5.2. Data for methyl 4,7,8,9-tetra-O-acetyl-5-acetamido-2-O-
dihydroxyphosphoryl-3,5-dideoxy-b-D-glycero-D-galacto-non-2-
ulo-pyranosonate, tetrabutylammonium salt (b-4a)
4.6.1. Data for methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
1H NMR (200.13 MHz, CDCl3): dH 0.99 (t, J = 7.1 Hz, 12H,
CH3CH2CH2CH2N,), 1.33–1.53 (m, 8H, CH3CH2CH2CH2N), 1.55–
1.78 (m, 8H, CH3CH2CH2CH2N), 1.75–1.90 (m, 1 H, H-3ax), 1.85
(s, 3 H, AcN), 2.00, 2.01, 2.04, 2.13 (4 s, 3H each, 4 ꢅ AcO), 2.68
(dd, J3eq,4 = 4.9 Hz, J3eq,3ax = 12.9 Hz, 1H, H-3eq), 3.22–3.35 (m, 8
H, CH3CH2CH2CH2N), 3.74 (s, 3H, OCH3), 4.14 (ddd ꢁ q, J ꢃ 10.2 Hz,
diacetylamino)-2-O-dihydroxyphosphoryl-3,5-dideoxy-
glycero- -galacto-non-2-ulo-pyranosonate,
tetrabutylammonium salt ( -4b)
a-D-
D
a
1H NMR (600.13 MHz, CDCl3): dH 1.01 (t, J = 7.3 Hz, 12H,
CH3CH2CH2CH2N), 1.45 (tq, J = 7.3 Hz, 8H, CH3CH2CH2CH2N), 1.63–
1.71 (m, 8H, CH3CH2CH2CH2N), 1.957, 2.00, 2.07, 2.12 (4 s, 3H each,

A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
727
4 ꢅ AcO), 2.34–2.40(fromCOSY—signalat2.37)(m, 1H, H-3ax), 2.28,
2.36 (2 s, 3H each, Ac2N), 3.14 (dd, J3eq,4 = 5.4 Hz, J3eq,3ax = 13.3 Hz,
1H, H-3eq), 3.27–3.34 (m, 8 H, CH3CH2CH2CH2N), 3.80 (s, 3H,
OMe), 4.07 (t, J = 10.0 Hz, 1H, H-5), 4.15 (dd, J9a,9b = 12.5 Hz,
J8,9a = 4.9 Hz, 1H, H-9a), 4.36 (dd, J9a,9b = 12.5 Hz, J8,9b = 2.9 Hz, 1H,
H-9b), 4.72 (dd, J5,6 = 9.8 Hz, J6,7 = 1.5 Hz, 1H, H-6), 5.13 (dd,
J6,7 = 1.5 Hz, J7,8 = 8.0 Hz, 1H, H-7), 5.20–5.24 (m, 1H, H-8), 5.71
(ddd, J4,5 = 10.4 Hz, J3ax,4 = 10.4 Hz, J3eq,4 = 5.4 Hz, 1H, H-4); 13C
NMR (150.9 MHz, CDCl3): dC 13.7 (CH3), 19.7 (CH2), 20.76, 20.85,
20.98 (OC(O)CH3), 24.0 (CH2), 25.5, 27.9 (NC(O)CH3), 38.2 (C-3),
52.4 (OCH3), 57.9 (C-5), 58.9 (CH2N), 61.5 (C-9), 67.2, 67.8, 69.2,
69.4 (C-4, C-6, C-7, C-8), 98.0 (d, JC,P = 2.1 Hz, C-2), 168.9 (C-1),
169.5, 170.0, 170.6 (CO), 173.9, 174.7 (NCO); 31P NMR (81.02 MHz,
CDCl3): dP ꢀ5.1; ESIMS: found m/z 580.1 [MꢀBu4N–OMe–H]ꢀ. Calcd
for C21H27NO16P: 580.11.
5.54 (m, 2H, H-4 and H-7), 5.92 (d, J = 8.7 Hz, 1H, NH), 5.98 (d,
J = 2.9 Hz, 1H, H-3); 13C NMR (50.32 MHz, CDCl3): dC 20.7, 20.8
(OC(O)CH3), 23.1 (NC(O)CH3), 46.4 (C-5), 52.5 (OMe), 61.9 (C-9),
67.6 (C-4), 67.9 (C-7), 70.7 (C-8), 76.4 (C-6), 107.9 (C-3), 145.0
(C-2), 161.6 (C-1), 170.2, 170.6, 170.8 (C@O); ESIMS: found m/z
496.0 [M+Na]; calcd for C20H27NNaO12: 496.1.
4.8. Methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-diacetylamino)-3,5-
dideoxy-2,6-anhydro-D-glycero-D-galacto-non-2-enonate (5b)
Glycosyl chloride 2b (Rf = 0.62, AcOEt; prepared from 1b
(20.1 mg, 0.035 mmol)) was dissolved in MeCN (0.6 mL), then
Na2HPO4 (6 mg, 0.042 mmol) was added and the reaction mix-
ture was stirred for 30 days at ambient temperature (25–
30 °C). The reaction mixture was filtered through silica gel pad,
volatiles were evaporated and the residue was dried in vacuum
of the oil pump and applied onto a silica gel column which
was eluted with gradient of AcOEt in petroleum ether
(0?100%) to give pure 5b as a white foam (16.6 mg, 92%;
4.6.2. Data for methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-2-O-dihydroxyphosphoryl-3,5-dideoxy-b-
glycero- -galacto-non-2-ulo-pyranosonate,
tetrabutylammonium salt (b-4b)
D-
D
Rf = 0.62, AcOEt). ½a D20
ꢂ
+42.2 (c 0.6, CHCl3) (lit.17
[
a]
+51 (c
D
1H NMR (600.13 MHz, CDCl3): dH 1.01 (t, J = 7.3 Hz, 12H,
CH3CH2CH2CH2N), 1.45 (tq, J = 7.3 Hz, 8H, CH3CH2CH2CH2N),
1.63–1.71 (m, 8H, CH3CH2CH2CH2N), 1.962, 1.99, 2.08, 2.12 (4
s, 3 H each, 4 ꢅ AcO), 2.34–2.40 (from COSY—signal at 2.13)
(m, 1H, H-3ax), 2.24, 2.41 (2 s, 3H each, Ac2N), 2.93 (dd,
J3eq,4 = 5.5 Hz, J3eq,3ax = 13.2 Hz, 1H, H-3eq), 3.27–3.34 (m, 8 H,
CH3CH2CH2CH2N), 3.80 (s, 3H, OMe), 4.30 (t, J = 10.0 Hz, 1H, H-
5), 4.37–4.39 (m, 1H, H-9a), 4.61–4.64 (m, 1H, H-9b), 5.15–
5.17 (m, 1H, H-6), 5.20–5.24 (m, 1H, H-7), 5.25–5.27 (m, 1H,
H-8), 5.75 (ddd, J4,5 = 10.0 Hz, J4,3ax = 10.0 Hz, J4,3eq = 5.5 Hz, 1H,
H-4); 13C NMR (150.9 MHz, CDCl3): dC 13.7 (CH3), 19.7 (CH2),
20.71, 20.78, 20.81, 20.90 (OC(O)CH3), 24.0 (CH2), 25.4, 27.7
(NC(O)CH3), 39.1 (d, JC,P = 6.4 Hz, C-3), 52.4 (OCH3), 57.2 (C-5),
58.9 (CH2N), 61.6 (C-9), 67.6, 68.1, 68.5, 71.6 (C-4, C-6, C-7, C-
8), 99.1 (d, JC,P = 6.4 Hz, C-2), 168.9 (C-1), 169.6, 170.0, 170.1,
170.7 (CO), 174.1, 174.6 (NCO); 31P NMR (81.02 MHz, CDCl3):
dP ꢀ4.2.
0.6, CHCl3)). The 1H and 13C NMR spectra of the product 5b were
identical to those described in the literature.17 Signals of NAc2
methyl groups are broadened in 1H NMR spectrum and not pres-
ent in 13C NMR spectrum at all (probably due to exchange); 1H
NMR (600.13 MHz, CDCl3): dH 2.03, 2.04, 2.06, 2.10 (4 s, 3H each,
4 ꢅ AcO), 2.39 (br s, 6H, Ac2N), 3.81 (s, s, 3H, OMe), 4.17 (dd,
J8,9b = 6.3 Hz, J9a,9b, = 12.5 Hz, 1H, H-9b), 4.52 (dd, J8,9a = 2.8Hz,
1H, H-9a), 4.56 (dd, J4,5 = 9.3 Hz, J5,6 = 10.1Hz, 1H, H-5), 5.16
(dd, J6,7, = 1.6 Hz, 1H, H-6), 5.22 (dd, J7,8 = 6.1 Hz, 1H, H-7), 5.34
(ddd, 1H, H-8), 5.93 (d, J3,4 = 2.7 Hz, 1H, H-3), 6.08 (dd, 1H, H-
4); 13C NMR (150.9 MHz, CDCl3) dC 20.69, 20.72, 20.75, 20.77
(OC(O)CH3), 52.5 (OMe), 55.1 (C-5), 61.9 (C-9), 67.3, 67.8, 70.2,
76.3 (C-4, C-6, C-7, C-8), 108.9 (C-3), 146.4 (C-2), 161.6 (C-1),
169.7, 170.0, 170.3, 170.5 (C@O); ESIMS: found m/z 538.0
[M+Na]. Calcd for C22H29NNaO13: 538.5.
4.9. Methyl 4,7,8,9-tetra-O-acetyl-2-azido-5-(N,N-
diacetylamino)-2,3,5-trideoxy-a-D-glycero-D-galacto-non-2-
4.6.3. Data for bis[methyl (4,7,8,9-tetra-O-acetyl-5-(N,N-
ulopyranosonate (6b)
diacetylamino)-3,5-dideoxy-D-glycero-D-galacto-non-2-ulo-
pyranosyl)onate]phosphate, tetrabutylammonium salt (11b)
1H NMR (200.13 MHz, CDCl3, selected signals): dH 1.01 (t, J = 7.0
Hz, 12 H, CH3CH2CH2CH2N), 1.34–1.55 (m, 8H, CH3CH2CH2CH2N),
1.56–1.79 (m, 8H, CH3CH2CH2CH2N), 2.59 (dd, J3eq,4 = 5.3 Hz,
J3eq,3ax = 13.0 Hz, 1H, H-3eq-A), 3.13 (dd, J3eq,4 = 5.4 Hz,
J3eq,3ax = 13.0 Hz, 1H, H-3eq-B), 3.21–3.39 (m, 8H, CH3CH2CH2CH2N),
3.79, 3.85 (2 s, 6H, 2 ꢅ OMe); 31P NMR (81.02 MHz, CDCl3): dP ꢀ11.9,
ꢀ13.6.
A
solution of chloride 2b (prepared from 1b (45.1 mg,
0.078 mmol)) in CH2Cl2 (1 mL) was added to a stirred solution of
Bu4NHSO4 (26.4 mg, 0.078 mmol) and NaN3 (25.4 mg, 0.39 mmol)
in saturated aqueous NaHCO3 (1 mL). The mixture was stirred for
0.5 h (TLC control: 7:3; Rf = 0.56 (6b), Rf = 0.53 (2b), AcOEt–petro-
leum ether). Organic layer was washed successively with 2 M
H2SO4 (20 mL), saturated aqueous NaHCO3 (20 mL), and brine
(30 mL). The combined extracts were filtered through a cotton
wool plug and concentrated to give the residue, which was purified
by silica gel column chromatography using AcOEt as the eluent to
give azide 6b (35.8 mg, 82%; Rf = 0.56, AcOEt–petroleum ether).
4.7. Methyl 5-acetamido-4,7,8,9-tetra-O-acetyl-3,5-dideoxy-2,6-
anhydro-D-glycero-D-galacto-non-2-enonate (5a)
½
a 2D0
ꢂ
+74.8 (c 0.8, CHCl3).
Glycosyl chloride 2b (prepared from 1b (21.2 mg,
0.037 mmol)) was dissolved in MeCN (0.6 mL), then Na2HPO4
(6 mg, 0.042 mmol) was added and the suspension refluxed
(TLC control: Rf = 0.48 (5a), Rf = 0.62 (2b), AcOEt). After 3 h the
mixture was cooled to ambient temperature, filtered through a
silica gel pad and the volatiles were evaporated. The residue
was dried under vacuum (oil pump) to give crude glycal 5a
(17.5 mg, 100%; Rf = 0.48, AcOEt), which was almost pure accord-
ing to NMR spectroscopy and has the 1H and 13C NMR spectra
identical to those described in the literature.16 1H NMR
(200.13 MHz, CDCl3): dH 1.92 (s, 3H, AcN), 2.04, 2.05, 2.07, 2.11
(4 s, 3H each, 4 ꢅ AcO), 3.79 (s, 3H, OMe), 4.18 (dd, J = 12.4 Hz,
J = 7.1, 1H, H-9b), 4.34–4.45 (m, 2H, H-5 and H-6), 4.63 (dd,
J = 12.4 Hz, J = 3.0 Hz, 1H, H-9a), 5.28–5.39 (m, 1H, H-8), 5.45–
1H NMR (200.13 MHz, CDCl3): dH 1.76 (dd, J3ax,4 = 10.9 Hz,
J3eq,3ax = 13.1 Hz, 1H, H-3ax), 2.00, 2.03, 2.12, 2.14 (4 s, 3H each,
4 ꢅ AcO), 2.32, 2.38 (2 s, 3H each, Ac2N), 2.71 (dd, J3eq,4 = 5.3 Hz,
1H, H-3eq), 3.93 (s, 3H, OMe), 4.16 (dd, J8,9b = 5.3 Hz, J9a,9b = 12.4 Hz,
1H, H-9b), 4.23 (dd, J4,5 = 9.7 Hz, J5,6 = 10.0 Hz, 1H, H-5), 4.34 (dd,
J8,9a = 2.8 Hz, 1H, H-9a), 4.79 (dd, J6,7 = 1.4 Hz, 1H, H-6), 5.15 (dd,
J7,8 = 7.3 Hz, 1H, H-7), 5.24–5.37 (m, 1H, H-8), 5.63 (ddd, 1H, H-4);
13C NMR (54.24 MHz, CDCl3): dC 20.7, 20.9 (OC(O)CH3), 26.1, 28.0
(2 ꢅ NC(O)CH3), 37.2 (C-3), 53.5 (OCH3), 56.7 (C-5), 61.7 (C-9),
66.6, 67.1, 69.2, 71.3 (C-4, C-6, C-8, C-7), 89.1 (C-2); the carbonyl sig-
nals are not present in 13C NMR spectrum due to exchange or too
short relaxation delay during spectrum acquisition; ESIMS: found
m/z 581.0 [M+Na]. Calcd for C22H30N4NaO13: 581.2.

728
A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
4.10. Methyl [methyl 4,7,8,9-tetra-O-acetyl-5-acetamido-3,5-
dideoxy- -glycero- -galacto-non-2-ulopyranoside]onate (7a)
(entry 2 in Table 1)
4.11.4. Data for methyl [methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
D
D
diacetylamino)-3,5-dideoxy-b-
ulopyranoside]onate (b-7b)
D-glycero-D-galacto-non-2-
1H NMR (200.13 MHz, CDCl3, selected signals not overlapping
Methanol (0.6 mL, 14.8 mmol) was added to a solution of chlo-
ride 2a (prepared from 1a (19.7 mg, 0.037 mmol) in MeCN (0.6 mL)
and the mixture was stirred for 3 h at ambient temperature (20 °C).
Then the reaction mixture was concentrated to give methyl glyco-
with those of
a
-7b): dH 1.96, 2.06, 2.08, 2.14 (4 s, 3H each, 4 ꢅ AcO),
2.58 (dd, J3eq,4 = 5.5 Hz, J3eq,3ax = 12.8 Hz, 1H, H-3eq), 3.28, 3.81 (2 s,
3H each, OMe), 4.64 (dd, J8,9b = 2.1 Hz, J9a,9b = 11.4 Hz, 1H, H-9b),
5.79 (ddd, J3ax,4 = 10.4 Hz, J4,5 = 10.4 Hz, 1H, H-4).
side 7a (18.7 mg, 100%, a:b = 12:1) as a colourless syrup.
4.12. Methyl [allyl 4,7,8,9-tetra-O-acetyl-5-N-acetyl-3,5-
dideoxy-D-glycero-D-galacto-non-2-ulopyranoside]onate (8a)
4.10.1. Data for methyl [methyl 4,7,8,9-tetra-O-acetyl-5-
acetamido-3,5-dideoxy- -glycero- -galacto-non-2-
ulopyranoside]onate ( -7a)
The 1H NMR spectrum of the product
a-
D
D
a
4.12.1. Method A, entry 5 in Table 1
solution of chloride 2a (prepared from 1a (71.7 mg,
a-7a was identical to that
described in the literature.4c 1H NMR (200.13 MHz, CDCl3): dH 1.89
(s, 3 H, AcN), 1.99–2.10 (m, 1H, H-3ax), 2.03, 2.05, 2.14, 2.16 (4 s,
3H each, 4 ꢅ AcO), 2.57 (dd, J3eq,4 = 4.5 Hz, J3eq,3ax = 12.8 Hz, 1H,
H-3eq), 3.32, 3.82 (2 s, 3H each, OMe), 4.07–4.17 (m, 2H, H-5, H-
9a, H-6), 4.32 (dd, J8,9b = 2.8 Hz, JH-9a,H-9b = 12.5 Hz, 1H, H-9b),
4.85 (ddd, J4,5 = 10.1 Hz, J3ax,4 = 10.1 Hz, 1H, H-4), 5.21 (d,
JNH,5 = 9.7 Hz, 1H, NH), 5.33 (dd, J6,7 = 1.7 Hz, J7,8 = 8.7 Hz, 1H, H-
7), 5.43 (ddd, J8,9a = 5.6 Hz, 1 H, H-8).
A
0.134 mmol)) in allyl alcohol (2.3 mL, 33.8 mmol) was stirred for
24 h at ambient temperature (20 °C). Then the mixture was con-
centrated and treated with Ac2O (0.5 mL, 5.3 mmol) and pyridine
(0.5 mL, 6.2 mmol) for 24 h. The volatiles were evaporated and
coevaporated with toluene (5 ꢅ 1 mL) give the residue, which
was purified by silica gel column chromatography using AcOEt as
the eluent to give a mixture (56.1 mg) of allyl glycoside 8a with
a trace amount of glycal 5a as a colourless syrup. The yields of allyl
glycoside 8a (54.6 mg, 77%,
a:b = 6:1) and glycal 5a (1.5 mg, 2%)
4.10.2. Data for methyl [methyl 4,7,8,9-tetra-O-acetyl-5-
were calculated by integration of the respective signals in the 1H
NMR spectrum and taking into account the molecular masses of
8a and 5a (531.54 and 473.43, respectively).
acetamido-3,5-dideoxy-b-
ulopyranoside]onate (b-7a)
1H NMR (200.13 MHz, CDCl3, selected signals not overlapping
with those of -7a): dH 3.27 (OMe).
D-glycero-D-galacto-non-2-
a
4.12.2. Method B, entry 6 in Table 1
Allyl alcohol (0.6 mL, 8.8 mmol) was added to a solution of chlo-
ride 2a (prepared from 1a (30.2 mg, 0.057 mmol)) in MeCN
(0.6 mL) and the mixture was stirred for 24 h at ambient tempera-
ture (20 °C). Then the mixture was concentrated and treated with
Ac2O (0.5 mL, 5.3 mmol) and pyridine (0.5 mL, 6.2 mmol) for
24 h. The volatiles were evaporated and coevaporated with toluene
(5 ꢅ 1 mL) to give the residue, which was purified by silica gel col-
umn chromatography using AcOEt as the eluent to give a mixture
(28.1 mg) of allyl glycoside 8a with a trace amount of glycal 5a as a
colourless syrup. The yields of allyl glycoside 8a (26.9 mg, 89%,
4.11. Methyl [methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-
ulopyranoside]onate (7b)
D-glycero-D-galacto-non-2-
4.11.1. Method A, entry 3 in Table 1
solution of chloride 2b (prepared from 1b (29.9 mg,
A
0.054 mmol)) in MeOH (2 mL, 49.4 mmol) was stirred for 30 min
at ambient temperature (20 °C). Then the mixture was concen-
trated to give methyl glycoside 7b (28.4 mg, 96%,
a:b = 11:1) as a
a
:b = 3:1) and glycal 5a (1.2 mg, 4%) were calculated by integration
colourless syrup.
of the respective signals in the 1H NMR spectrum and taking into
account the molecular masses of 8a and 5a (531.54 and 473.43,
respectively).
4.11.2. Method B, entry 4 in Table 1
Methanol (0.597 mL, 14.88 mmol) was added to a solution of
chloride 2b (prepared from 1b (19.7 mg, 0.034 mmol)) in MeCN
(0.6 mL) and the mixture was stirred for 3 h at ambient tempera-
ture (20 °C). Then the reaction mixture was concentrated to give
4.12.3. Data for methyl [allyl 4,7,8,9-tetra-O-acetyl-5-N-acetyl-
3,5-dideoxy-
-8a)
The 1H and 13C NMR spectra of
a-D-glycero-D-galacto-non-2-ulopyranoside]onate
(a
methyl glycoside 7b (18.7 mg, 100%,
a:b = 3:1) as a colourless
a
-anomer a-8a were identical to
syrup.
those described in the literature.6a 1H NMR (600.13 MHz, CDCl3):
dH 1.86 (s, 3H, AcN), 1.97 (dd, J3eq,3ax = 12.8 Hz, J3ax,4 = 12.3 Hz, 1H,
H-3ax), 2.01, 2.03, 2.12, 2.13 (4 s, 3H each, 4 ꢅ AcO), 2.60 (dd,
J3eq,3ax = 12.8 Hz, J3eq,4 = 4.7 Hz, 2H, H-3eq), 3.77 (s, 3H, OMe), 3.86
(dddd, J = 12.8 Hz, J = 5.9 Hz, J = 1.3 Hz, J = 1.3 Hz, 1H, OCH2a),
4.05–4.12 (m, 3H, H-5, H-6, H-9a), 4.26 (dddd, J = 12.8 Hz,
J = 5.3 Hz, J = 1.4 Hz, J = 1.4 Hz, 1H, OCH2b), 4.31 (dd, J9a,9b = 12.4 Hz,
J8,9b = 2.7 Hz, 1H, H-9b), 4.85 (ddd, J3ax,4 = 12.3 Hz, J4,5 = 9.8 Hz,
J3eq,4 = 4.7 Hz, 1H, H-4), 5.13–5.17 (m (‘dq’, J = 10.4 Hz, J = 1.4 Hz),
1H, CH2a@CH), 5.24–5.29 (m (‘dq’, J = 17.2 Hz, J = 1.5 Hz), 1H,
CH2b@CH), 5.31 (dd, J7,8 = 8.3 Hz, J6,7 = 2.1 Hz, 1H, H-7), 5.34 (d,
J5,NH = 9.6 Hz, 1H, NH), 5.40 (ddd, J7,8 = 8.3 Hz, J8,9a = 5.8 Hz,
J8,9b = 2.7 Hz, 1H, H-8), 5.80–5.88 (m, 1H, CH2@CH); 13C NMR
(54.24 MHz, CDCl3): dC 20.8 (OC(O)CH3), 23.2 (NC(O)CH3), 38.1 (C-
3), 49.5 (C-5), 52.7 (OMe), 62.4 (C-9), 65.9 (OCH2), 67.4, 68.7, 69.1,
72.5 (C-4, C-6, C-7, C-8), 98.5 (C-2), 117.2 (CH2), 133.6 (CH), 168.4
(C-1), 170.1, 170.2, 170.6, 171.0 (C@O).
4.11.3. Data for methyl [methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy- -glycero- -galacto-non-2-
ulopyranoside]onate ( -7b)
The 1H and 13C NMR spectra of the product
a
-
D
D
a
a
-7b were identical to
those described in the literature.2e 1H NMR (200.13 MHz, CDCl3): dH
1.84 (dd, J3ax,4 = 10.9 Hz, J3eq,3ax = 12.8 Hz, 1H, H-3ax), 1.98, 2.02,
2.13, 2.15 (4 s, 3 H each, 4 ꢅ AcO), 2.29, 2.38 (2 s, 3H each, Ac2N),
2.76 (dd, J3eq,4 = 5.2 Hz, 1H, H-3eq), 3.35, 3.85 (2 s, 3H each, OMe),
4.08–4.25 (m, 2 H, H-5, H-9a), 4.33 (dd, J8,9b = 2.8 Hz, JH-9a,H-9b
=
12.5 Hz, 1H, H-9b), 4.99 (dd, J5,6 = 10.1 Hz, J6,7 = 1.7 Hz, 1H, H-6),
5.16 (dd, J7,8 = 8.0 Hz, 1H, H-7), 5.32–5.44 (m, 1 H, H-8), 5.50 (ddd,
J4,5 = 10.4 Hz, 1H, H-4); 13C NMR (54.24 MHz, CDCl3): dC 20.8
(OC(O)CH3), 26.0, 28.0 (NC(O)CH3), 38.9 (C-3), 52.3, 52.8 (OMe), 57.1
(C-5), 62.0 (C-9), 66.8, 67.3, 68.6, 69.6 (C-4, C-6, C-7, C-8), 98.9 (C-2),
167.4 (C-1), 169.7, 170.2, 170.7, 173.6 (C@O).

A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
729
4.12.4. Data for methyl [allyl 4,7,8,9-tetra-O-acetyl-5-N-acetyl-
4.13.4. Data for methyl [allyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-b- -glycero- -galacto-non-2-
ulopyranoside]onate (b-8b)
3,5-dideoxy-b-
D
-glycero-
D-galacto-non-2-ulopyranoside]onate
D
D
(b-8a)
The b-anomer b-8a was described in the literature, however no
1H NMR (600.13 MHz, CDCl3): dH 1.83 (dd, J3eq,3ax = 13.0,
J3ax,4 = 10.9 Hz, 1H, H-3ax), 1.97, 2.02, 2.05, 2.14 (4 s, 3H each,
4 ꢅ AcO), 2.31, 2.39 (2 s, 3H each, Ac2N), 2.65 (dd, J3eq,4 = 5.3 Hz,
J3eq,3ax = 13.0 Hz, 1H, H-3eq), 3.79 (s, 3H, OMe), 3.93–3.99 (m, 1H,
OCH2a), 4.06–4.10 (m, 1H, OCH2b), 4.16 (dd, J9a,9b = 12.5 Hz,
J8,9a = 5.4 Hz, 1H, H-9a), 4.200 (t, J = 10.0 Hz, 1H, H-5), 4.61 (dd,
J9a,9b = 12.5 Hz, J8,9a = 2.5 Hz, 1H, H-9b), 5.05 (dd, J5,6 = 10.0 Hz,
J6,7 = 1.8 Hz, 1H, H-6), 5.21 (dd, J7,8 = 5.4 Hz, J6,7 = 1.8 Hz, 1H, H-7),
5.20–5.25 (m, 1H, CH2a@CH), 5.31–5.35 (m, 1H, H-8), 5.41 (dq,
J = 17.2 Hz, J = 1.7 Hz, 1H, CH2b@CH), 5.82–5.94 (m, 2H, H-4, CH);
13C NMR (54.24 MHz, CDCl3): dC 20.70, 20.73, 20.88, 20.96
(OC(O)CH3), 25.9, 28.0 (NC(O)CH3), 38.7 (C-3), 52.6 (OMe), 57.2
(C-5), 62.00 (C-9), 64.5 (OCH2), 66.8, 68.2, 68.5, 70.9 (C-4, C-6, C-
7, C-8), 98.3 (C-2), 116.9 (CH2), 133.1 (CH), 167.6 (C-1), 169.5,
170.18, 170.20, 170.5 (OCO), 173.7, 174.3 (NCO).
NMR data were reported for it.6b 1H NMR (600.13 MHz, CDCl3, se-
lected signals not overlapping with those of
a-8a): dH 2.49 (dd,
J3eq,4 = 4.9 Hz, J3eq,3ax = 12.9 Hz, 1H, H-3eq), 3.79 (s, 3H, OMe),
3.95 (dd, J5,6 = 10.5 Hz, J6,7 = 2.3 Hz, 1H, H-6), 3.98–4.00 (m ꢁ ‘dt’,
J = 5.2 Hz, J = 1.4 Hz, 1H, OCH2a), 4.02–4.04 (m ꢁ ‘dt’, J = 5.4 Hz,
J = 1.4 Hz, 1H, OCH2b), 4.79 (dd, J9a,9b = 12.3 Hz, J8,9b = 2.5 Hz, 1H,
H-9b), 5.82–5.89 (m, 1H, CH2@CH); 13C NMR (54.24 MHz, CDCl3):
dC 21.1 (OC(O)CH3), 23.2 (NC(O)CH3), 37.4 (C-3), 49.4 (C-5), 52.7
(OCH3), 62.4 (C-9), 64.7 (OCH2), 68.4, 68.9, 71.2, 72.0 (C-4, C-6, C-
7, C-8), 98.3 (C-2), 117.3 (CH2), 133.2 (CH), 167.4 (C-1), 170.1,
170.3, 170.6, 170.7 (C@O).
4.13. Methyl [allyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-
ulopyranoside]onate (8b)
D-glycero-D-galacto-non-2-
4.14. Methyl [phenyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-2-thio-
2-ulopyranoside]onate (9b)
a-D-glycero-D-galacto-non-
4.13.1. Method A, entry 7 in Table 1
solution of chloride 2b (prepared from 1b (55 mg,
A
0.095 mmol)) in AllOH (1.6 mL, 23.5 mmol) was stirred for 48 h
at ambient temperature (20 °C). Then the mixture was concen-
trated and treated with Ac2O (0.5 mL, 5.3 mmol) and pyridine
(0.5 mL, 6.2 mmol) for 24 h. The volatiles were evaporated and
coevaporated with toluene (5 ꢅ 1 mL) give the residue, which
was purified by silica gel column chromatography using gradient
of AcOEt in petroleum ether (0?100%) as the eluent to give allyl
A
solution of chloride 2b (prepared from 1b (64.9 mg,
0.113 mmol)) in AcOEt (1 mL) was added to a stirred solution of
Bu4NHSO4 (38.1 mg, 0.112 mmol) in saturated aqueous NaHCO3
(1 mL). Then PhSH (0.046 mL, 0.45 mmol) was added and the mix-
ture was stirred for 1.5 h at ambient temperature (20 °C) (TLC con-
trol: Rf = 0.53 (9b), Rf = 0.62 (2b), AcOEt). Organic layer was washed
with saturated aqueous NaHCO3 (50 mL), and brine (70 mL). The
combined extracts were filtered through a cotton wool plug and
concentrated to give the residue, which was purified by silica gel
glycoside 8b as a colourless syrup (15.6 mg, 29%,
allyl glycoside 8a as a white foam (23.8 mg, 47%,
a
:b = 1.9:1) and
a:b = 1.3:1).
HPLC (Silasorb 600, 7.5
in petroleum ether (0?100%) as the eluent to give thioglycoside 9b
l
m, 15 ꢅ 250 mm) using gradient of AcOEt
4.13.2. Method B, entry 8 in Table 1
Allyl alcohol (2.3 mL, 33.8 mmol) was added to a solution of
chloride 2b (prepared from 1b (75.4 mg, 0.131 mmol)) in MeCN
(2.3 mL) and the mixture was stirred for 48 h at ambient tempera-
ture (20 °C). Then the mixture was concentrated and treated with
Ac2O (0.5 mL, 5.3 mmol) and pyridine (0.5 mL, 6.2 mmol) for
24 h. The volatiles were evaporated and coevaporated with toluene
(5 ꢅ 1 mL) give the residue, which was purified by silica gel col-
umn chromatography using gradient of AcOEt in petroleum ether
(0?100%) as the eluent to give allyl glycoside 8b as a colourless
as a colourless syrup (57.4 mg, 91%;
a:b = 14:1; Rf = 0.53, AcOEt)
and small amount of glycal 5b (5.3 mg, 9%). ½a D25
ꢂ
+32.5 (c 0.36,
CHCl3). ESIMS: found m/z 648.0 [M+Na]. Calcd for C28H35NNaO13S:
648.2.
4.14.1. Data for methyl [phenyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-2-thio-a-D-glycero-D-galacto-non-
2-ulopyranoside]onate ( -9b)
a
1H NMR (600.13 MHz, CDCl3): dH 1.93 (dd, J3ax,3eq = 12.9 Hz,
J3ax,4 = 11.2 Hz, 1H, H-3ax), 1.99, 2.029, 2.031, 2.17 (4 s, 3H each,
4 ꢅ AcO), 2.30, 2.35 (2 s, 3H each, Ac2N), 3.01 (dd, J3ax,3eq = 12.9 Hz,
J3eq,4 = 5.2 Hz, 1H, H-3eq), 3.59 (s, 3H, OMe), 4.08 (dd, J4,5 = 10.3 Hz,
J5,6 = 10.2 Hz, 1H, H-5), 4.24 (dd, J9a,9b = 12.5 Hz, J8,9a = 5.4 Hz, 1H,
H-9a), 4.34 (dd, J9a,9b = 12.5 Hz, J8,9b = 2.7 Hz, 1H, H-9b), 4.82 (dd,
J5,6 = 10.2 Hz, J6,7 = 1.6 Hz, 1H, H-6), 5.12 (dd, J7,8 = 7.1Hz,
J6,7 = 1.6 Hz, 1H, H-7), 5.19 (ddd, J7,8 = 7.1 Hz, J8,9a = 5.4 Hz,
J8,9b = 2.7 Hz, 1H, H-8), 5.50 (ddd, J3ax,4 = 11.2 Hz, J4,5 = 10.3 Hz,
J3eq,4 = 5.2 Hz, 1H, H-4), 7.33–7.37 (m, 2H, Ph), 7.39–7.43 (m, 1H,
Ph), 7.52–7.55 (m, 2H, Ph); 13C NMR (54.24 MHz, CDCl3): dC 20.8,
20.9 (OC(O)CH3), 25.9, 28.0 (2 ꢅ NC(O)CH3), 39.2 (C-3), 52.7
(OMe), 57.2 (C-5), 61.6 (C-9), 67.2, 67.5, 69.8, 72.1 (C-4, C-6, C-7,
C-8), 87.4 (C-2), 128.8, 129.9, 136.4 (Ph), 167.6 (C-1), 169.5,
169.9, 170.2, 170.6 (OCO), 173.7, 174.4 (NCO).
syrup (43.6 mg, 58%,
foam (13.1 mg, 19%,
a
:b = 1.4:1) and allyl glycoside 8a as a white
a:b = 1.4:1).
4.13.3. Data for methyl [allyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy- -glycero- -galacto-non-2-
ulopyranoside]onate ( -8b)
a-
D
D
a
1H NMR (600.13 MHz, CDCl3): dH 1.90 (dd, J3eq,3ax = 13.0 Hz,
J3ax,4 = 11.0 Hz, 1H, H-3ax), 1.99, 2.03, 2.13, 2.15 (4 s, 3H each,
4 ꢅ AcO), 2.29, 2.38 (2 s, 3H each, Ac2N), 2.81 (dd, J3eq,4 = 5.2 Hz,
J3eq,3ax = 13.0 Hz, 1H, H-3eq), 3.83 (s, 3H, OMe), 3.93–3.99 (m, 2H,
H-9a, OCH2a), 4.203 (t, J = 10.1Hz, 1H, H-5), 4.31 (dt, J = 5.2 Hz,
J = 1.5 Hz, 1H, OCH2b), 4.34 (dd, J9a,9b = 12.4 Hz, J8,9b = 2.7 Hz, 1H,
H-9b), 4.98 (dd, J5,6 = 10.1 Hz, J6,7 = 1.8 Hz, 1H, H-6), 5.15 (dd,
J7,8 = 8.0 Hz, J6,7 = 1.8 Hz, 1H, H-7), 5.15–5.18 (m, 1H, CH2a@CH),
5.29 (dq, J = 17.2 Hz, J = 1.7 Hz, 1H, CH2b@CH), 5.37 (ddd,
J7,8 = 8.0 Hz, J8,9a = 5.3 Hz, J8,9b = 2.7 Hz, 1H, H-8), 5.51 (ddd,
J3eq,4 = 5.2 Hz, J3ax,4 = 11.0 Hz, J4,5 = 10.1 Hz, 1H, H-4), 5.82–5.90 (m,
1H, CH); 13C NMR (54.24 MHz, CDCl3): dC 20.70, 20.73, 20.93, 21.06
(OC(O)CH3), 25.9, 28.0 (NC(O)CH3), 39.0 (C-3), 52.7 (OMe), 57.1 (C-
5), 61.95 (C-9), 65.6 (OCH2), 66.9, 67.3, 68.8, 69.7 (C-4, C-6, C-7, C-
8), 98.5 (C-2), 117.0 (CH2), 133.7 (CH), 167.7 (C-1), 169.6, 170.10,
170.15, 170.6 (OCO), 173.6, 174.5 (NCO).
4.14.2. Data for methyl [phenyl 4,7,8,9-tetra-O-acetyl-5-(N,N-
diacetylamino)-3,5-dideoxy-2-thio-b-
2-ulopyranoside]onate (b-9b)
D-glycero-D-galacto-non-
1H NMR (600.13 MHz, CDCl3, selected signals not overlapping
with those of
1H, H-3eq).
a-9b): dH 2.78 (dd, J3ax,3eq = 12.9 Hz, J3eq,4 = 5.2 Hz,

730
A. V. Orlova et al. / Carbohydrate Research 345 (2010) 721–730
4.15. Methyl 4,7,8,9-tetra-O-acetyl-5-(N,N-diacetylamino)-3,5-
the Russian Federation (project MK-3244.2008.3). Authors are
grateful to V. I. Bragin for registration of NMR spectra on a Bruker
AVANCE 600 spectrometer.
dideoxy-2-ethoxythiocarbonyl-2-thio-a-D-glycero-D-galacto-
non-2-ulo-pyranosonate (10b)
References
4.15.1. Method A
Glycosyl chloride 2b (prepared from 1b (100 mg, 0.174 mmol))
was dissolved in dry acetone (7 mL) and solid O-ethyl S-potassium
dithiocarbonate (63.9 mg, 0.40 mmol) was added. The mixture was
stirred for 2 h at ambient temperature (20 °C) (TLC control:
Rf = 0.50 (10b), Rf = 0.53 (2b), AcOEt–petroleum ether, 7:3) and
then the volatiles were evaporated to give the residue, which
was purified by silica gel column chromatography using gradient
of AcOEt in petroleum ether (0?100%) as the eluent to give a mix-
1. (a) Schauer, R. ‘Biochemistry of Sialic Acid Diversity’. In Carbohydrates in
Chemistry and Biology; Ernst, B., Hart, G. W., Sinaÿ, P., Eds., Part II Biology of
Saccharides, Biosynthesis and Degradation of Glycoconjugates; Wiley-VCH:
Weinheim, 2000; Vol. 3, pp 227–244.; (b) Angata, T.; Varki, A. Chem. Rev. 2002,
102, 439–469.
2. (a) Boons, G.-J.; Demchenko, A. V. Chem. Rev. 2000, 100, 4539–4565; (b) Ress, D.
K.; Linhardt, R. J. Curr. Org. Synth. 2004, 1, 31–46; (c) DeMeo, C. In Frontiers In
Modern Carbohydrate Chemistry; Demchenko, A. V., Ed.; ACS Symposium series
960; American Chemical Society: Washington, DC, 2007; Chapter 8, pp 118–
131.; (d) DeMeo, C.; Priyadarshani, U. Carbohydr. Res. 2008, 343, 1540–1552; (e)
Crich, D.; Li, W. Org. Lett. 2006, 8, 959–962; (f) Crich, D.; Li, W. J. Org. Chem.
2007, 72, 2387–2391; (g) Crich, D.; Wu, B. Org. Lett. 2008, 10, 4033–4035, and
references cited therein; (h) Tanaka, S.; Goi, T.; Tanaka, K.; Fukase, K. J.
Carbohydr. Chem. 2007, 26, 369–394, and references cited therein.
3. (a) Kononov, L. O.; Malysheva, N. N.; Kononova, E. G.; Garkusha, O. G. Russ.
Chem. Bull. 2006, 55, 1311–1313; (b) Kononov, L. O.; Malysheva, N. N.;
Kononova, E. G.; Orlova, A. V. Eur. J. Org. Chem. 2008, 3251–3255; (c) Kononov,
L. O.; Malysheva, N. N.; Orlova, A. V. Eur. J. Org. Chem. 2009, 611–616; (d)
Orlova, A. V.; Kononov, L. O.; Kimel, B. G.; Sivaev, I. B.; Bregadze, V. I. Appl.
Organomet. Chem. 2006, 20, 416–420.
ture (81.2 mg, 25:1) of a-xanthate 10b with a small amount of gly-
cal 5b as a yellow syrup. The yields of 10b (78.7 mg, 71%) and 5b
(2.5 mg, 3%) were calculated by integration of the respective sig-
nals in the 1H NMR spectrum and taking into account the molecu-
lar masses of 10b and 5b (637.67 and 515.46, respectively).
4.15.2. Method B
Glycosyl chloride 2b (prepared from 1b (24.8 mg, 0.043 mmol))
was dissolved in 95% EtOH (1 mL) and solid O-ethyl S-potassium
dithiocarbonate (10.2 mg, 0.064 mmol) was added. The mixture
was stirred for 1 h at ambient temperature (20 °C) (TLC control:
Rf = 0.50 (10b), Rf = 0.53 (2b), AcOEt–petroleum ether, 7:3) and then
the volatiles were evaporated to give the residue, which was purified
by silica gel column chromatography using gradient of AcOEt in
petroleum ether (0?100%) as the eluent to give a mixture (22 mg,
4. (a) Kuhn, R.; Lutz, P.; MacDonald, D. L. Chem. Ber. 1966, 99, 611–617; (b)
Byramova, N. E.; Tuzikov, A. B.; Bovin, N. V. Carbohydr. Res. 1992, 237, 161–175;
(c) Kononov, L. O.; Magnusson, G. Acta Chem. Scand. 1998, 52, 141–144; (d)
Shpirt, A. M.; Kononov, L. O.; Torgov, V. I.; Shibaev, V. N. Russ. Chem. Bull. 2004,
53, 717–719.
5. Demchenko, A. V.; Boons, G.-J. Chem. Eur. J. 1999, 5, 1277–1283.
6. (a) Allanson, N. M.; Davidson, A. H.; Floyd, C. D.; Martin, F. M. Tetrahedron:
Asymmetry 1994, 5, 2061–2076; (b) Tanaka, T.; Ozawa, M.; Miura, T.; Inazu, T.;
Tsuji, S.; Kajimoto, T. Synlett 2002, 1487–1490.
7. Xia, J.; Alderfer, J. L.; Piskorz, C. F.; Locke, R. D.; Matta, K. L. Carbohydr. Res. 2000,
328, 147–163.
8. (a) Martichonok, V.; Whitesides, G. M. J. Org. Chem. 1996, 61, 1702–1706;
(b) Marra, A.; Sinaÿ, P. Carbohydr. Res. 1989, 187, 35–42; (c) Tropper, F.
D.; Andersson, F. O.; Cao, S.; Roy, R. J. Carbohydr. Chem. 1992, 11, 741–
750.
9. Makosza, M. Pure Appl. Chem. 2000, 72, 1399–1403, and references cited
therein.
10. (a) Cao, S.; Meunier, S. J.; Andersson, F. O.; Letellier, M.; Roy, R. Tetrahedron:
Asymmetry 1994, 5, 2303–2312; (b) Tropper, F. D.; Andersson, F. O.; Braun, S.;
Roy, R. Synthesis 1992, 618–620.
11. (a) Shibaev, V. N.; Danilov, L. L. In Studies in Natural Products Chemistry;
Atta-ur-Rahman, Ed.; Elsevier: Amsterdam, 1991; pp 63–113; (b)
Illarionov, P. A.; Torgov, V. I.; Shibaev, V. N. Russ. Chem. Bull. 2000, 49,
1895–1898.
12. (a) Martin, T. J.; Schmidt, R. R. Tetrahedron Lett. 1993, 34, 1765–1768; (b) Sim,
M. M.; Kondo, H.; Wong, C. H. J. Am. Chem. Soc. 1993, 115, 2260–2267.
13. Shpirt, A. M.; Kononov, L. O., unpublished results.
15:1) of a-xanthate 10b with a small amount of glycal 5b as a yellow
syrup. The yields of 10b (20.9 mg, 76%) and 5b (1.1 mg, 5%) were cal-
culated by integration of the respective signals in the 1H NMR spec-
trum and taking into account the molecular masses of 10b and 5b
(637.67 and 515.46, respectively). 1H NMR (600.13 MHz, CDCl3):
dH 1.36 (t, J = 7.1 Hz, 3H, CH2CH3), 1.94 (dd, J3ax,4 = 10.3 Hz,
J3eq,3ax = 13.2 Hz, 1H, H-3ax), 1.98, 2.00, 2.10, 2.11 (4 s, 3H each,
4 ꢅ AcO), 2.28, 2.37 (2 br s, 3H each, Ac2N), 2.85 (dd, J3eq,4 = 5.0 Hz,
J3eq,3ax = 13.2 Hz, 1H, H-3eq), 3.82 (s, 3H, OMe), 4.15 (dd,
J4,5 = 10.0 Hz, J5,6 = 10.2 Hz, 1H, H-5), 4.20 (dd, J8,9a = 5.3 Hz,
J9a,9b = 12.5 Hz, 1H, H-9a), 4.32 (dd, J8,9a = 2.7 Hz, J9a,9b = 12.5 Hz,
1H, H-9a), 4.54–4.61 (m, 1H, SCH2aCH3), 4.75–4.81 (m, 1H,
SCH2bCH3), 5.11 (dd, J6,7 = 1.5 Hz, J7,8, = 7.3 Hz, 1H, H-7), 5.25 (ddd,
J7,8 = 7.3 Hz, J8,9a = 2.7 Hz, J8,9a = 5.3 Hz, 1H, H-8), 5.31 (dd,
J5,6 = 10.2 Hz, J6,7 = 1.5 Hz, 1H, H-6), 5.49 (ddd, J3ax,4 = 10.3 Hz,
J3eq,4 = 5.0 Hz, J4,5 10.0 Hz, 1H, H-4; 13C NMR (54.24 MHz, CDCl3):
dC 13.3 (OCH2CH3), 20.6, 20.7, 20.8, 20.9 (OC(O)CH3), 25.8, 27.8
(2 ꢅ NC(O)CH3), 38.2 (C-3), 53.2 (OMe), 56.9 (C-5), 61.5 (C-9), 67.0,
67.3, 69.6, 72.1 (C-4, C-6, C-7, C-8), 70.4 (OCH2CH3), 86.8 (C-2),
168.1 (C-1), 169.4, 170.0, 170.1, 170.5 (OCO), 173.4, 174.4 (NCO),
207.5 (C@S); ESIMS: found m/z 659.9 [M+Na]. Calcd for
C25H35NNaO14S2: 660.2.
14. Gotoh, T.; Koyama, T.; Ogura, K. J. Biochem. 1992, 122, 20–27.
15. Shpirt, A. M.; Kononov, L. O.; Torgov, V. I.; Shibaev, V. N. Russ. Chem. Bull. 2005,
54, 481–482.
16. (a) Kulikova, N. Y.; Shpirt, A. M.; Kononov, L. O. Synthesis 2006, 4113–
4114; (b) Okamoto, K.; Kondo, T.; Goto, T. Bull. Chem. Soc. Jpn. 1987, 60,
631–636.
17. Ercegovic, T.; Nilsson, U. J.; Magnusson, G. Carbohydr. Res. 2001, 331,
255–263.
18. (a) Garegg, P. J. Adv. Carbohydr. Chem. Biochem. 2004, 59, 69–134; (b)
Horenstein, N. A. Adv. Phys. Org. Chem. 2006, 41, 275–314.
19. Guthrie, R. D.; Jencks, W. P. Acc. Chem. Res. 1989, 22, 343–349.
20. Richard, J. P.; Amyes, T. L.; Toteva, M. M.; Tsuji, Y. Adv. Phys. Org. Chem. 2004,
39, 1–26.
21. (a) Dedola, S.; Nepogodiev, S. A.; Field, R. A. Org. Biomol. Chem. 2007, 5, 1006–
1017; (b) Dondoni, A. Chem. Asian J. 2007, 2, 700–708; (c) Meldal, M.; Tornøe, C.
W. Chem. Rev. 2008, 108, 2952–3015.
22. Shpirt, A. M.; Kononov, L. O.; Maltsev, S. D.; Torgov, V. I.; Shibaev, V. N.,
manuscript in preparation.
Acknowledgements
This work was supported by RFBR (projects No. 05-03-32579
and 08-03-00839) and the Council on Grants of the President of