Determination of the configuration of ribitol in the C-polysaccharide of Streptococcus pneumoniae using a synthetic approach

Article abstract of DOI:10.1139/v99-056

The C-polysaccharide is an antigen common to all known serotypes of Streptococcus pneumoniae bacteria and is a potential target for vaccine preparation. The final uncertainty in the structure of its repeating unit, a pentasaccharide phosphate containing two phosphorylcholine side chains, has been resolved by determining the configuration of ribitol. Assignment of configuration was performed by synthesis of the two trisaccharide phosphate fragments that have either D- or L-ribitol at their centers and comparison of their ¹H and ¹³C NMR spectral data with that of the natural polysaccharide. The syntheses employed common synthons added in different orders to an asymmetrically substituted chiral ribitol derivative to obtain opposite chiralities in the ribitol segments. The data for the trisaccharide containing D-ribitol was almost identical to that of the natural material while that for the trisaccharide containing L-ribitol differed significantly. In particular, the chemical shift differences between the two protons of the primary carbons of ribitol units directly attached to the β-2-acetamido-2-deoxy-galactopyranosyl residue were 0.10, 0.12, and 0.33 ppm, in the natural polysaccharide, the D-ribitol-containing trisaccharide, and the L-ribitol-containing trisaccharide, respectively. The average difference between the ¹³C NMR chemical shifts of corresponding ribitol carbons from the natural polysaccharide and the D-ribitol-containing trisaccharide phosphate was 0.034 ppm. This evidence indicates that ribitol in the C-polysaccharide has the D-configuration and that a very similar mixture of conformations in the ribitol portions is present for the natural polysaccharide and the D-ribitol-containing trisaccharide phosphate.

Full text of DOI:10.1139/v99-056

481
Determination of the configuration of ribitol in
the C-polysaccharide of Streptococcus
pneumoniae using a synthetic approach
Huiping Qin and T. Bruce Grindley
Abstract: The C-polysaccharide is an antigen common to all known serotypes of Streptococcus pneumoniae bacteria
and is a potential target for vaccine preparation. The final uncertainty in the structure of its repeating unit, a
pentasaccharide phosphate containing two phosphorylcholine side chains, has been resolved by determining the
configuration of ribitol. Assignment of configuration was performed by synthesis of the two trisaccharide phosphate
1
fragments that have either D- or L-ribitol at their centers and comparison of their H and 13C NMR spectral data with
that of the natural polysaccharide. The syntheses employed common synthons added in different orders to an
asymmetrically substituted chiral ribitol derivative to obtain opposite chiralities in the ribitol segments. The data for
the trisaccharide containing D-ribitol was almost identical to that of the natural material while that for the trisaccharide
containing L-ribitol differed significantly. In particular, the chemical shift differences between the two protons of the
primary carbons of ribitol units directly attached to the β-2-acetamido-2-deoxy-galactopyranosyl residue were 0.10,
0.12, and 0.33 ppm, in the natural polysaccharide, the D-ribitol-containing trisaccharide, and the L-ribitol-containing
trisaccharide, respectively. The average difference between the 13C NMR chemical shifts of corresponding ribitol
carbons from the natural polysaccharide and the D-ribitol-containing trisaccharide phosphate was 0.034 ppm. This
evidence indicates that ribitol in the C-polysaccharide has the D-configuration and that a very similar mixture of
conformations in the ribitol portions is present for the natural polysaccharide and the D-ribitol-containing trisaccharide
phosphate.
Key words: C-polysaccharide, Streptococcus pneumoniae, ribitol, dibutylstannylene acetal, determination of configura-
tion.
Résumé : Le polysaccharide-C est un antigène commun à tous les sérotypes connus de la bactérie Streptococcus
pneumoniae et est un objectif potentiel pour la préparation de vaccins. En déterminant la configuration du ribitol, on a
levé la dernière incertitude dans la structure de son unité constitutive, un phosphate de polysaccharide contenant deux
phosphorylcholine comme chaînes latérales. On a attribué la configuration en effectuant la synthèse des deux fragments
de phosphate de trisaccharide comportant soit le D- soit le L-ribitol comme centres et en comparant les données de
1
leurs spectres RMN du H et du 13C avec celles du polysaccharide naturel. On a réalisé la synthèse des trisaccharides
en utilisant des synthons communs et un dérivé chiral du ribitol substitué d’une façon asymétrique qui ont été ajouté
dans des ordres différents dans le but d’obtenir une chiralité différente dans ce segment. Les données obtenues avec le
trisaccharide contenant du D-ribitol sont pratiquement identiques à celles du produit naturel alors que celles obtenues
avec le trisaccharide contenant du L-ribitol présentent des différences significatives. En particulier, les différences de
déplacements chimiques entre les deux protons des carbones primaires des unités ribitol attachées directement au résidu
β-2-acétamido-2-désoxy-galactopyranosyle sont égales respectivement à 0,10, 0,12 et 0,33 ppm pour le polysaccharide
naturel, le trisaccharide contenant le D-ribitol et le trisaccharide contenant le L-ribitol. La différence moyenne entre les
déplacements chimiques en RMN du 13C pour les carbones correspondants du ribitol du polysaccharide naturel et du
trisaccharide contenant le D-ribitol sont de 0,034 ppm. Cette différence indique que le polysaccharide-C possède la
configuration D et que le polysaccharide naturel et le phosphate du trisaccharide contenant le D-ribitol comportent des
mélanges très semblables de conformations dans les portions du ribitol.
Mots clés : polysaccharide-C, Streptococcus pneumoniae, ribitol, acétal de dibutylstannylène, détermination de configu-
ration.
[Traduit par la Rédaction] Qin and Grindley 494
Received October 16, 1998.
H. Qin and T.B. Grindley.1 Department of Chemistry, Dalhousie University, Halifax, NS B3H 4J3, Canada.
1Author to whom correspondence may be addressed. Telephone: (902) 494-2041. Fax: (902) 494-1310.
e-mail: grindley@chem1.chem.dal.ca
Can. J. Chem. 77: 481–494 (1999)
© 1999 NRC Canada

482
Can. J. Chem. Vol. 77, 1999
Scheme 1. The structure of the C-polysaccharide of Streptococcus pneumoniae with the configuration of ribitol shown as D.
NH2
O
[
N+
O
O
HO
HO
O
P
O
O
N+
O
OH
AcH
N
O
O
O
O
O
O
P
OH
HO
O
O
AcH
N
O
O
NHAc
OH
OH
OH
O
O
P
O
Na+
]
1
n
O
to have the D-configuration in the capsular polysaccharide
from type b Haemophilus influenzae, a Gram-negative or-
ganism (23), in a compound isolated from a lichen (24), and
in CDP-ribitol (25). Glycerol-1-phosphate normally has the
D-configuration (15, 17, 19–21) in agreement with the scheme
proposed for its biosynthesis, but L-glycerol-1-phosphate is
also known, in the membrane-bound teichoic acid from
Streptococcus lactis (16) and in the cell wall of Bacillus ce-
reus (22). It was discovered that the biosynthetic pathway
leading to L-glycerol-1-phosphate was different from that
proposed for its D-analogue (26) and the question naturally
arises as to whether only one biosynthetic pathway leads to
ribitol-5-phosphate.
Ribitol-5-phosphate is present in several of the capsular
polysaccharides of Streptococcus pneumoniae as well as in the
C-polysaccharide, and its configuration in these compounds
has also been assumed to be D (27). In this publication, we
outline the determination of the configuration of ribitol in
the C-polysaccharide of Streptococcus pneumoniae through
synthesis of the two trisaccharide portions of the natural C-
polysaccharide that have D- and L-ribitol at the center. A
tetrasaccharide portion of the repeating unit synthesized pre-
viously had D-ribitol at the nonreducing terminus, but no at-
tempt was made to determine whether the configuration of
ribitol matched that in the natural product (28).
Introduction
The Gram-positive microorganism Streptococcus pneumoniae
induces infections such as pneumonia, bacteremia, otitis me-
dia, and meningitis in human beings (1), and resistance to
drug treatment is becoming an increasing problem. Ninety
serologically defined types of these capsular bacteria have
been described to date (2). The capsule of each serotype is a
unique polysaccharide that extends outwards from the cell
wall and current vaccines against this organism contain a
mixture of 23 of the capsular polysaccharides of the most
common serotypes. The C-polysaccharide, a common anti-
gen of all the known serotypes of Streptococcus Pneumoniae,
is distributed on both the inside and outside of the cell walls
(3). A polysaccharide with the same repeating unit is cova-
lently bonded to a lipid moiety in the plasmic membrane
through a phosphodiester linkage (4), in which form it is
called the F antigen. Its presence in all serotypes makes it an
attractive target for vaccine preparation but attempts to date
have not resulted in sufficient activity to yield commercial
vaccines (1c, 5, 6). Structural studies (7–12) showed that the
polysaccharide contained the pentasaccharide repeating unit
shown in Scheme 1.
The configuration of ribitol in the structure was assumed
to be D, based on schemes proposed for the biosynthesis of
teichoic acids where the donors of glycerol and ribitol phos-
phate residues are CDP-glycerol and CDP-ribitol, respec-
tively (8, 9, 13). Teichoic acids are major constituents of
Gram-positive bacterial cell walls and consist of polymers of
ribitol-5-phosphate or glycerol-3-phosphate with carbohydrates
and alanine as substituents (13) and the C-polysaccharide
was classified as a teichoic acid (8). The experimental evi-
dence supporting this scheme is that CDP-ribitol is incorpo-
rated in polyribitol phosphate by cell-free enzymic
preparations from various Gram-positive bacteria (14). It is
not certain that this scheme would apply to the ribitol phos-
phate in the C-polysaccharide or other polymers where ribitol
phosphate constitutes a small part of the polymer.
Results and discussion
Synthetic strategy
To simulate the environment of the ribitol unit in the C-
polysaccharide, two trisaccharide phosphate targets (18 and
24) were designed. In 18 and 24, instead of linking to an-
other 2-acetamido-2-deoxy-D-galactopyranose residue as in
′′
the C-polysaccharide, the 3 position was substituted with a
methyl group. A synthetic strategy to obtain 18 and 24 was
developed that made the maximum use of synthons that were
common to both trisaccharides and is shown in Scheme 2.
Although the absolute configuration of glycerol has been
determined a number of times (15–22), the configuration of
ribitol has never been determined for any compounds de-
rived from Gram-positive bacteria. In fact, its configuration
has only been determined three times; it was demonstrated
Synthesis of trisaccharide phosphates containing the
D-ribitol unit (18)
Compound 12 was obtained from the known compound,
tert-butyldimethylsilyl 2-azido-4,6-O-benzylidene-2-deoxy-β-D-
galactopyranoside (29), as shown in Scheme 3. The anomeric
© 1999 NRC Canada

Scheme 2. Retrosynthetic analysis of synthesis of trisaccharide phosphates 18 and 24.
OH
OH
OpMBn
OBn
Ph
O
O
CH3O
OBn
O
OBn
NHAc
O
HO
BnO
OH
OH
OH
O
P
O
O
O
P
O
+
10
BnO
OCE
BnO
BnO
+
O
O
O
CH3O
BnO
O
Na+
OCH3
N3
NH
pMBnO
O
O
OBn
HO
HO
12
14
OCH3
8
CCl3
18
OH
OH OH
Ph
O
CH3O
O
O
N(iPr)2
O
O
P
NHAc
HO
HO
HO
+ 8
12
OCE
OCH3
O
O
BnO
BnO
O
P
+
O
CH3O
O
N3
OBn
O
BnO
BnO
BnO
+
O
Na
10
19
O
HO
HO
OCH3
24
pMBnO
OH

484
Can. J. Chem. Vol. 77, 1999
Scheme 3. Synthesis of glycosyl donor 12.
Ph
O
Ph
Ph
O
O
O
O
O
1. TBAF, CH2Cl2
O
CH3I, NaH, DMF
O
O
CH3O
O
2. Cl3CCN, NaH
CH3O
97% yield
SitBu(CH3)2
O
HO
SitBu(CH3)2 69% yield
N3
NH
N3
N3
O
11
12
CCl3
Scheme 4. Synthesis of key ribitol-containing intermediate 8.
EtS
EtS
SEt
SEt
1. BnBr, NaH, DMF
OH
OH
OH
OH
OH
OH
OH
2. HgCl2, CaCO3, CH3CN/H2O
1. Bu2SnO, Toluene
3. NaBH4, CH3OH/H2O
50% Yield
2. AllBr, CsF, DMF
68% Yield
OAll
2
3
OpMBn
OBn
OBn
OBn
OH
OH
HO
1. pMBnCl, NaH, DMF
89% Yield
OBn
OBn
OBn
OAll
BnO
BnO
2. Pd/C, H+, CH3OH/H2O
69% Yield
BnO
pMBnO
6
8
tert-butyldimethylsilyl group was stable in the presence of
sodium hydride, allowing O-3 to be methylated in 97% yield
with methyl iodide to provide 11. Treatment of 11 with
TBAF in dichloromethane successfully afforded the desired
desilylated product. Subsequent reaction with trichloroace-
tonitrile in the presence of sodium hydride afforded the α-
trichloroacetimidate 12 selectively in a 69% yield for the
two steps.
Compound 8, the key intermediate, could be converted
into either D- or L-ribitol derivatives, depending on the order
in which the differentially protected primary hydroxyl groups
were functionalized. Its synthesis started with D-ribose di-
ethyl dithioacetal (2), which was selectively allylated at the
primary position via a stannylene acetal intermediate to give
3. This is the first time that a dialkyldithioacetal derivative
has been selectively protected at the primary position via a
dibutylstannylene acetal intermediate (30) and this approach
replaces a four-step preparation from D-ribose (31). As shown
in Scheme 4, following benzylation of the selectively allyl-
ated 3, the diethyl dithioacetal group was removed and re-
duction gave 6. Compound 6 had previously been prepared
in six steps from γ-ribonolactone, a relatively expensive starting
material (32). Following p-methoxybenzylation of the free
hydroxy group and deallylation, ribitol derivative 8 was ob-
tained. Compound 8 is an L-derivative if the free hydroxyl
group is at O-1 as numbered in Scheme 4 but a D-derivative
if the p-methoxybenzyl group is at O-1 as numbered for 6.
The D-ribitol-containing trisaccharide phosphate 18 was
synthesized by glycosylation of a phosphorus-linked disac-
charide 14 with trichloroacetimidate 12. As shown in
Scheme 5, the synthesis of the disaccharide glycosyl accep-
tor 14 was performed by phosphitylation of L-ribitol deriva-
tive 8 with 10. Compound 10 was made from methyl 2,3,4-
tri-O-benzyl-6-O-trityl-β-D-glucopyranoside (33) by a sequence
that started with detritylation in 75% yield using 30% HBr
in acetic acid, an improved method. The product, methyl
2,3,4-tri-O-benzyl-β-D-glucopyranoside (9), was reacted with
chloro 2-cyanoethyl (N,N-diisopropyl)phosphoramidite (34)
in the presence of diisopropylethylamine (DIPEA) to give
10 in an 83% yield. The reaction of 10 with 8 was followed
by oxidation with m-chloroperbenzoic acid (MCPBA) in the
same pot. The p-methoxybenzyl group survived the acidic
conditions provided by both tetrazole and MCPBA, and was
removed with 2,3-dichloro-4,5-dicyanoquinone (DDQ).
These two steps afforded 14 in a 57% yield. The formation
of the phosphorus linkage was clearly evident in the cou-
pling of the phosphorus to the C-4 (doublet) and C-5 (dou-
blet) of the ribitol unit and the disappearance of isopropyl
group signals (δ 24, 43) in the 13C NMR spectrum. Com-
pound 14 was a mixture of epimers at phosphorus as in-
dicated by the two signals (δ –0.72, –0.77) in the 31P
NMR spectrum.
Glycosylation of the disaccharide glycosyl acceptor 14 by
12 was successfully performed in an 82% yield. As antici-
© 1999 NRC Canada

Qin and Grindley
485
Scheme 5. Synthesis of D-ribitol-containing trisaccharide phosphate 18.
HO
OBn
OBn
OBn
HO
N(iPr)2
O
P
1. Tetrazole, CH3CN
2. MCPBA
O
BnO
BnO
OCE
O
O
+
BnO
P
O
3. DDQ, CH2Cl2\H2O
57% yield
OCH3
BnO
BnO
OCE
OBn
O
pMBnO
BnO
BnO
OCH3
OBn
10
8
14
Ph
O
Ph
O
Ph
O
O
O
O
+
O
O
O
CH3O
O
CH3O
CH3O
NH
O
NHAc
N
3O
N3
OBn
OBn
OBn
OBn
AcSH (75% yield)
OBn
OBn
O
12
O
CCl3
O
P
O
TMSOTf, CH3CN
82% yield
P
O
O
OCE
OCE
O
16
BnO
BnO
O
15
BnO
BnO
OCH3
OCH3
OBn
OBn
OR'
OR'
O
CH3O
O
1. TEA\Py\H2O
2. Pd/C, H2
NHAc
OR
OR
OR
O
3. Na+ form resin
P
O
O
M
O
O
RO
RO
OCH3
OR
17 R = Bn, 2R' = CHPh, M = Et3N+
+
18 R = R' = H, M = Na
pated, solvent participation by acetonitrile (35) caused the
β-linkage to be formed selectively, which was clearly shown
by the observation of a large vicinal coupling constant of
7.8 Hz between H-1 and H-2. No products having an α-link-
age were observed. In the transformation of the azido group
to the acetamido group, the benzylidene ring surprisingly
survived overnight treatment with thiolacetic acid at room
temperature to give the reduction product in 75% yield.
Due to the stereogenic centre at phosphorus, 10, 14, and
15 are diastereomeric mixtures, as shown by their 31P NMR
spectra, each of which contained two peaks. This is also re-
flected in the appearance of two peaks in the 13C NMR spec-
tra for most of the carbons close to the phosphorus atoms.
Surprisingly, on treatment of 15 with thiolacetic acid, an-
other pair of peaks appeared. In the 31P NMR spectrum of
16, there are two large peaks (δ –0.68, –0.96) and two small
peaks (δ 0.17, –0.73). A possible explanation is that both E
and Z acetamido rotamers are being observed, perhaps due
to hydrogen bond formation between the amide hydrogen
and a phosphate oxygen in the normally less stable con-
former.
The cyanoethyl (CE) group on phosphorus was removed
by treatment with a triethylamine, pyridine, and water mix-
ture for 2 h at room temperature and the product had only a
single peak in its 31P NMR spectrum (1.41 ppm), as ex-
pected for a product with no chirality at phosphorus. The
disappearance of the CN (116 ppm) and two CH2 (19 and
61 ppm) signals from the 13C NMR spectrum confirmed the
removal of this group. Hydrogenation removed the 4,6-O-
benzylidene ring and all of the O-benzyl groups, as con-
firmed by disappearance of all of the signals assigned to aro-
1
matic proton and carbon signals in the H and 13C NMR
© 1999 NRC Canada

486
Can. J. Chem. Vol. 77, 1999
Scheme 6. Synthesis of the L-ribitol-containing trisaccharide phosphate 24.
Ph
Ph
O
O
O
O
HO
O
BnO
BnO
TMSOTf
CH3CN
CH3O
O
O
+
CH3O
N3
BnO
BnO
BnO
RO
BnO
N
NH
19 R = pMBn 17%
20 R = H 16%
3O
pMBnO
CCl3
12
8
Ph
O
Ph
N(iPr)2
O
P
O
O
CEO
O
O
O
BnO
O
CH3O
O
+
CH3O
AcSH
O
BnO
OCH3
22
NHAc
BnO
N3
OBn
BnO
BnO
BnO
10
BnO
BnO
O
P
O
P
1. Tetrazole, CH3CN
2. MCPBA
O
O
O
O
OCE
84%
OCE
O
BnO
BnO
O
BnO
BnO
OCH3
21
OCH3
OBn
OBn
OR'
OR'
O
CH3O
O
NHAc
RO
RO
RO
1. TEA\Py\H2O
2. Pd/C, H2
O
O
P
O
M
O
3. IR-120 (Na+)
O
RO
RO
OCH3
OR
23 R = Bn, 2R' = CHPh, M = Et3N+
24 R = R' = H, M = Na+
spectra, respectively. Treatment with Amberlite IR-120 (Na+)
resin afforded the target trisaccharide phosphate 18 in a quan-
titative yield. The disappearance of the signals for the ethyl
group in the NMR spectra indicated the removal of the tri-
ethylammonium ion and the completion of the ion exchange
reaction.
that formation of the glycosidic linkage was faster than the
removal of the primary p-methoxybenzyl group under the
coupling conditions. If the p-methoxybenzyl group had been
cleaved at a rate comparable with or faster than glycoside
formation, 20 and its diastereomer with the ribitol unit having
the D-configuration would have been formed. The phospho-
triester linkage between the disaccharide 20 and the phos-
phoramidite derivative 10 was formed as previously in an
84% yield. A diastereomeric mixture 21 having both config-
urations at phosphorus was obtained. The transformation of
the azido group in 21 to an acetamido group and the subsequent
transformation from 22 to 24, as shown in Scheme 6, fol-
lowed the same procedure as in Scheme 5.
Synthesis of trisaccharide phosphate containing the
L-ribitol unit (24)
The synthesis of the target trisaccharide phosphate 24 fol-
lowed the reverse combination route outlined in Scheme 2.
Though extra caution was exercised, coupling of 12 and 8
gave a low yield of a mixture of the expected product (19)
and the product in which the p-methoxybenzyl group had
been removed (20). Compound 19 was transformed into 20
by treatment with DDQ (63% yield) at room temperature.
This observation confirms that compound 20 isolated from
the coupling reaction was formed from 19 and, therefore,
Assignment of configuration
1
Complete assignments of all H and 13C NMR signals in
the 500 and 125 MHz NMR spectra of the two trisaccharides
18 and 24 were made using a variety of 1D and 2D NMR
© 1999 NRC Canada

Qin and Grindley
487
1
Table 1. H NMR chemical shifts (ppm) for PnC (11), 18, and 24.
Comp.
Unit
H1a, H1b
H2
H3
H4
H5a, H5b
H6a, H6b
4.06, 4.06
PnC
GalNAc
Rib
Glc
GalNAc
Rib
4.64
3.86, 3.96
4.619
4.49
3.83, 3.95
4.39
–0.03, –0.01
4.48
3.740, 4.07
4.4
4.1
3.85
3.76
3.5
3.44
3.77
3.5
+0.01
3.45
3.744
3.5
4.17
3.88
3.54
4.22
3.89
3.5
+0.01
4.22
3.93
3.5
3.83
3.98, 4.04
3.55
3.66
3.99, 4.06
3.58
+0.01, +0.02
3.66
3.98, 4.08
3.58
3.99
3.31
3.93
3.95
3.28
0
3.94
3.95
3.28
0
4.11, 4.11
3.77, 3.84
18
24
Glc
4.07, 4.17
3.78, 3.83
4.09, 4.18
∆δRiba
GalNAc
Rib
Glc
∆δRib
–0.12, +0.11
0
+0.05
0.00, +0.04
a∆δRib refers to the chemical shift difference of the ribitol unit between 18, 24, and PnC. Digital resolution: 0.22 Hz/point for 18, and
0.43 Hz/point for 24.
Table 2. 13C NMR chemical shifts (ppm) for PnC (11), 18, and 24.
Comp.
Unit
C1
C2
C3
C4
C5
C6
PnC
GalNAc
Rib
Glc
GalNAc
Rib
102.08
71.46
104.81
102.67
71.41
104.12
–0.05
103.15
71.7
51.46
71.35
73.79
51.99
71.34
73.86
–0.01
51.98
71.34
73.89
–0.01
75.49
72.23
76.37
80.7
72.19
76.36
0
80.63
72.41
76.39
+0.18
64.07
71.46
69.68
64.35
71.48
69.87
+0.02
64.3
74.3
65.33
67.46
75.22
75.78
67.41
75.38
–0.1
75.77
67.36
75.38
–0.1
64.92
61.88
18
24
Glc
64.97
61.88
64.96
∆δRiba
GalNAc
Rib
71.67
69.87
+0.21
Glc
∆δ Rib
104.15
+0.24
a∆ δRib refers to the chemical shift difference of the ribitol unit between 18, 24, and PnC. Digital resolution: 0.015 ppm/point for 18,
0.0076 ppm/point for 24.
techniques, including JMOD, COSY, TOCSY, HMBC, and
HMQC experiments.2 In accordance with the literature re-
port detailing the spectral data for the C-polysaccharide (11),
acetone was used as an internal reference, using chemical
obtained by simulation using the program LAME8 (36), be-
cause the chemical shifts of H-1b′ and H-2′ were virtually
identical. The chemical shift differences (∆δ’s) between the
signals of the two H-1′s from the ribitol unit in the C-poly-
saccharide (PnC) and those from 18, the D-ribitol derivative,
were small, –0.03 ppm and –0.01 ppm; in contrast, the com-
parable differences between the data from PnC and that
from 24, the L-ribitol derivative, were much larger,
–0.12 ppm and +0.11 ppm. Probably the most significant ob-
servations concern the chemical shift differences between
the two H-1′ signals for each compound, being 0.10 ppm for
PnC, 0.12 ppm for the D-ribitol derivative 18, but much
larger, 0.33 ppm, for the L-derivative 24. The chemical shift
differences were also somewhat larger for the signals of H-4′
and H-5b′ for 24 with respect to PnC than for 18.
These observations clearly demonstrate that the data for
the D-ribitol derivative 18 was very similar to that of PnC,
whereas that of the L-ribitol derivative 24, while still fairly
similar to that of PnC, was less so. In fact, the chemical
shift for every proton from the ribitol part of 18 is almost
identical to that of the comparable proton in the natural
polysaccharide; the largest ∆δ was –0.04 ppm for H-2′.
As stated above, the most significant observation is that
the chemical shift differences between the two C-1′ protons
1
shifts of the methyl group at δ 2.225 and δ 31.07 in the H
and 13C NMR spectra, respectively, as references and sam-
ple concentrations were also the same. A detailed outline of
the procedures followed for signal assignment is provided in
the supplementary material. Assignments are given for the
1H and 13C NMR spectra of 18, 24, and the corresponding
segments of the natural polysaccharide (PnC) in Tables 1
and 2, respectively.
The 1H NMR spectra of the final products 18 and 24 were
similar, but there were small but significant differences in
the section of the spectra where the ribitol signals appear.
The chemical shifts for the galactosamine (GalNAc) and
glucose (Glc) parts of 18 and 24 were the same within
0.02 ppm (see Table 1), which confirmed that the careful at-
tention given to sample preparation, data point resolution,
and referencing resulted in highly reproducible chemical shifts.
The largest differences were observed for the signals of H-
1a and H-1b of ribitol. Most chemical shifts were obtained
from first-order analysis and from 2D experiments; those for
the D-ribitol segment of the 500 MHz spectrum of 18 were
2 Discussion of spectral assignments of 18 and 24 with some of the supporting JMOD, TOCSY, and HMQC spectra (7 pages) is available as
supplementary material and can be ordered from: The Depository of Unpublished Data, Document Delivery, CISTI, National Research
Council of Canada, Ottawa, ON K1A 0S2, Canada.
© 1999 NRC Canada

488
Can. J. Chem. Vol. 77, 1999
1
are virtually identical for PnC and the D-ribitol derivative
18, being 0.10 and 0.12 ppm, respectively, whereas that for
the L-ribitol derivative 24 is much larger, 0.33 ppm. This
section of the spectrum of the natural polysaccharide might
be different than that of the trisaccharide containing a ribitol
segment of the same configuration if the size of the poly-
saccharide and the presence of other charged groups (phos-
phorylcholine groups) caused the composition of the
conformational mixture for this region of the polysaccharide
to be different than that present for the trisaccharide. The
conformations and configuration of the ribitol segment and
the conformation of the attached 2-acetamido-2-deoxy-β-D-
galactopyranosyl unit will be most important in determining
this chemical shift difference. The latter will be fixed in a
4C1 conformation; the former, as an alditol with a β-1,3-O-O
interaction in its Fischer projection, will be highly flexible
(37). The ribitol segment should maintain this relative flexi-
bility even if the remainder of the polysaccharide is less mo-
shifts of the corresponding H and 13C NMR signals. Of the
two configurations, only one can be that present in the natu-
ral polysaccharide. The configuration for which the NMR
data was in closer agreement to that of the corresponding
part of the C-polysaccharide (PnC) should be the one pres-
ent in the natural material as long as there is reasonable
agreement. The H and 13C NMR chemical shift data for the
1
D-ribitol derivative 18 showed remarkable agreement with
that for PnC, with average chemical shift differences for the
corresponding signals being 0.02 and 0.034 ppm, respec-
tively; the agreement for the L-ribitol derivative 24 was sig-
nificantly worse. Perhaps the most significant observation
was that the chemical shift differences between the signals
of the two primary hydrogens on the ribitol carbon closest to
the anomeric center of the N-acetyl-β-D-galactopyranosyl
unit were 0.10 ppm in the natural polysaccharide, 0.12 ppm
in the D-ribitol derivative, but 0.33 ppm in the L-ribitol de-
rivative. In view of the conformational flexibility of ribitol
1
1
bile. The near identity of the H NMR chemical shifts of
derivatives and the near identity of all H and 13C NMR
PnC and 18 supports the contention that the conformational
mixtures present for both are almost identical, and the simi-
lar situation for 13C chemical shifts, discussed in the follow-
ing paragraph, provides further support. Therefore, the major
factors influencing the H-1′ chemical shift differences will
be the relative configurations at the anomeric centre of the
2-acetamido-2-deoxy-β-D-galactopyranosyl unit and those in
ribitol, primarily at C-2′ but also at C-3′, and C-4′. These ob-
servations indicate that the configuration of ribitol in the
natural material is D.
chemical shifts in the D-ribitol derivative and the natural
polysaccharide, it seems highly unlikely that there could be
a conformational effect in the natural polysaccharide that
could result in the two C-1 protons from the natural
polysaccharide having a chemical shift difference markedly
different than that of the trisaccharide containing ribitol with
the same configuration. Thus, the ribitol unit in the C-
polysaccharide of Streptococcus pneumoniae can now be re-
liably assigned the D-configuration.
The 13C NMR spectra of 18 and of 24 were also quite
similar. The largest differences lie in the chemical shifts of
C-1′ and C-4′ of the ribitol unit. The carbons of the glucopy-
ranose units of 18 and 24 have almost the same chemical shifts,
which indicates that there were no systematic shift differ-
ences between results from different samples (see Table 2).
It was found that the chemical shifts of the carbons of ribitol
in the synthetic D-trisaccharide phosphate 18 were almost
identical to those from the corresponding part of the PnC;
the average absolute deviation was 0.034 ppm. The chemical
shifts for the same part of the L-trisaccharide phosphate 24
had an average absolute deviation of 0.15 ppm, with the
largest differences at C-1 (+0.24 ppm), C-3 (+0.18 ppm),
and C-4 (+0.21 ppm), respectively. All of these were less
shielded than in the C-polysaccharide. 13C NMR chemical
shifts of carbons in acyclic systems are known (37a, 37b, 38)
to be strongly dependent on the mixture of conformations
present, particularly on the relative proportion of gauche and anti
conformers. The close similarity of the chemical shifts of the
corresponding carbons from the ribitol segments of PnC, 18,
and 24 to those from the first two, being identical within ex-
perimental error, strongly supports the contention that the
conformational mixtures present for the ribitol segments of
all three are identical. Therefore, this evidence supports the
above conclusion that the ribitol unit in the C-polysaccha-
ride of Streptococcus pneumoniae has the D-configuration.
Experimental section
General methods
Melting points were determined with a Fisher–Johns melt-
ing point apparatus and are uncorrected. Specific rotations
were measured on Perkin–Elmer model 141 or 243
polarimeters. TLC was carried out on 0.20 mm thick Merck
silica gel 60F-254 aluminum sheets cut to be approximately
7 cm long. Components were located by spraying with 2%
ceric sulfate in 1 M sulfuric acid and heating on a hot plate
until coloration occurred. Dry flash column chromatography
was performed on silica gel 60 PF-254 for TLC. Low-
pressure flash column chromatography was performed on
1
silica gel Merck grade 9385, 230–400 mesh, 60 Å. The H,
13C, and 31P NMR spectra were recorded on a Bruker AC-
250 NMR spectrometer, unless otherwise noted. Some spec-
tra were recorded on Bruker AMX-400 or AMX-500 NMR
spectrometers. Solutions for NMR spectra were prepared in
chloroform-d unless otherwise stated. Chemical shifts are re-
ported in ppm downfield from internal TMS for H and 13C
1
NMR spectra. The central peak for chloroform-d, at δ 77.0,
was used as a secondary reference for 13C NMR spectra.
Phosphoric acid (25%) was used as an external chemical
shift reference for 31P NMR (δ 0.00). For 18 and 24, acetone
was used as an internal chemical shift reference for 1H NMR
(δ 2.225) and for 13C NMR (δ 31.07). Exact masses for 3, 5,
6, and 7 were measured on a JEOL AX505H mass spectrom-
eter using fast atom bombardment (FAB) at the NRC in Ot-
tawa. Exact masses for 10, 11, and 24 were measured on a
Kratos IIH mass spectrometer using fast atom bombardment
(FAB) at the University of Ottawa. The fast atom beam was
provided by a cesium gun. Solvents for the matrix were
Conclusions
From the differences in the NMR data of 18 and 24, it
was demonstrated that the configuration of the ribitol unit in
these molecules had a significant influence on the chemical
© 1999 NRC Canada

Qin and Grindley
489
glycerol and (or) 3-nitrobenzyl alcohol and combinations
thereof. Exact masses for 15, 16, 17, 18, 21, and 23 were
measured on a Kratos MS50 mass spectrometer using
electrospray ionization (ESI) at the University of Alberta.
Electrospray ionization mass spectra for 8, 14, and 20 were
run on a VG Micromass Quattro spectrometer in aceto-
nitrile–water (1:1) using a loop injector. The solution con-
centration was approximately 5.21 × 10–4 M and injection
was 10 µL. The sample capillary was at 3500 V, the sample
cone was at 20 V, and the source temperature was at 100°C.
1H NMR, same as ref. 31; 13C NMR δ: 14.6 (2 CH3 SEt),
24.8, 26.1 (2 CH2 SEt), 53.8 (C-1), 69.3 (C-5), 72.0, 72.4,
73.4, 74.5 (3 CH2Ph, 1 CH2 All), 78.9, 79.8, 82.1 (C-2, C-3,
C-4), 116.4 (CH2ϭ), 134.7 (-CHϭ), 127.2-128.7, 138.2, 138.3,
138.7 (C-Ph).
5-O-Allyl-2,3,4-tri-O-benzyl-D-ribose, 5
A solution of compound 4 (0.310 g, 0.55 mmol) in aque-
ous acetonitrile (7.25 mL CH3CN + 0.72 mL H2O) was
treated with mercury(II) chloride (0.604 g, 2.21 mmol) and
calcium carbonate (0.229 g, 2.29 mmol). The resulting mix-
ture was stirred at room temperature for 1 h when TLC
showed that all starting material had been consumed. The re-
action mixture was filtered through Celite and the filtrate
was concentrated to a residue that was taken up in dichloro-
methane (100 mL). The solution was subsequently washed
with potassium iodide solution (1 M, 3 × 100 mL) and so-
dium thiosulfate solution (30%, 2 × 100 mL) and then dried
(Na2SO4) and concentrated. The title compound (5) was ob-
D-Ribose diethyl dithioacetal, 2
To a solution of D-ribose (30.0 g, 0.20 mol) and concen-
trated hydrochloric acid (36 mL) was added ethanethiol
(36 mL) dropwise at 0°C over 15 min and the reaction mix-
ture was stirred for another 30 min. Then a saturated sodium
bicarbonate solution was added dropwise until the solution
was neutral. The resulting solution was concentrated to a
residue that was extracted with ethyl acetate. The insoluble
inorganic salts were removed by filtration. The ethyl acetate
filtrate was concentrated to solid 2, and recrystallized from
ethanol as needles (38.61 g, 75% yield): mp 82–83°C (lit.
(39) mp 82–83°C).
23
tained as a colourless liquid (0.216 g, 86% yield); [α]D
1
20.5 (c 0.48, CHCl3); H NMR δ: 3.53 (dd, 1H, H-5a, J5a,5b
= –10.7 Hz, J5a,4 = 4.6 Hz), 3.67 (dd, 1H, H-5b, J5b,4
=
2.1 Hz), 3.87–4.01 (m, 4H, H-3, H-4, CH2 All), 4.09 (br d,
1H, H-2), 4.47–4.82 (m, 6H, 3 CH2Ph), 5.11–5.26 (m, 2H,
CH2ϭ), 5.78–5.93 (m, 1H, -CHϭ), 7.22–7.33 (m, 15H, H-
Ph), 9.47 (br s, 1H, -CHϭO); 13C NMR δ: 69.0 (C-5), 72.1
(CH2 All), 72.5, 72.7, 79.9 (3 CH2Ph), 76.5, 80.3 (2C C-3,
C-4), 82.2 (C-2), 116.8 (CH2ϭ), 134.6 (-CHϭ), 127.4–
128.3, 137.3, 137.5, 137.9 (C-Ph), 200.9 (C-1, HCϭO). Ex-
act Mass (FAB) calcd. for C29H32O5Na [M + Na]+:
483.2148; found: 483.2195.
5-O-Allyl D-ribose diethyl dithioacetal, 3
Dibutyltin oxide (0.248 g, 1.0 equiv.) and compound 2
(0.256 g, 1.0 mmol) were refluxed in toluene (20 mL) for
4 h in a Dean–Stark apparatus for the azeotropic removal of
water. Then the solvent was evaporated and cesium fluoride
(0.222 g, 1.45 equiv.) added. After the mixture had been
kept under vacuum for 1 h, a solution of allyl bromide
(0.12 mL, 1.3 equiv.) in DMF solution (6 mL) was added.
The reaction mixture was stirred at room temperature for 1 h
when TLC showed that all starting material had been con-
sumed. The reaction solution was concentrated to a residue
that was taken up in ethyl acetate (50 mL). The solution was
dried (Na2SO4) and concentrated to a liquid residue that was
purified by dry flash column chromatography to yield the ti-
tle compound (3): yield 0.202 g (68%); [α]D24 –17.4 (c 1.16,
5-O-Allyl-2,3,4-tri-O-benzyl-D-ribitol, 6
A solution of compound 5 (0.215 g, 0.47 mmol) in aque-
ous methanol (methanol:water 1:1, 3 mL) was stirred with
sodium borohydride (0.032 g, 0.81 mmol) for 10 h. Then
water (10 mL) was added and the resulting mixture was ex-
tracted with chloroform (3 × 10 mL). The combined extracts
were dried (Na2SO4) and concentrated to a liquid residue
that was purified with dry flash column chromatography
1
CHCl3), H NMR, same as ref. 31; 13C NMR δ: 14.4, 14.5
23
(2 CH3), 25.6 (2 CH2S), 54.4 (C-1), 71.1 (C-4), 71.5 (C-5),
72.4 (CH2-All), 72.8 (C-3), 75.0 (C-2), 117.6 (CH2ϭ), 133.9
(-CHϭ). Exact Mass (FAB) calcd. for C12H24O4S2Na [M +
Na]+: 319.1014; found: 319.1018.
to yield the title compound (6) (0.194 g, 90% yield); [α]D
23
–17 (c 2.52, CHCl3) (lit. (31) [α]D –23.2; lit. (32) [α]D
–15.0); H (31, 32) and 13C (32) NMR spectra same as liter-
1
ature values. Exact Mass (FAB) calcd. for C34H38O6Na [M +
Na]+:565.2566; found: 565.2544.
5-O-Allyl-2,3,4-tri-O-benzyl-D-ribose diethyl dithioacetal,
4
5-O-Allyl-2,3,4-tri-O-benzyl-1-O-p-methoxybenzyl-D-
ribitol, 7
A solution of compound 3 (0.425 g, 1.44 mmol) in DMF
(5 mL) was treated with excess NaH (0.42 g, 60% suspen-
sion in mineral oil, washed with hexanes 3 times before use)
at 0°C. The resulting mixture was stirred for 30 min. Then a
solution of benzyl bromide (1.36 mL) in DMF (2 mL) was
added dropwise. The reaction solution was stirred at 0°C for
3.5 h when TLC showed that all the starting material had
been consumed. Cold water was added until the total volume
was about 70 mL. The water layer was extracted with pentane
(5 × 40 mL). The pentane solution was dried (Na2SO4) and
concentrated to a liquid residue that was separated with dry
flash column chromatography using a gradient elution from
hexane to ethyl acetate. The title compound (4) was obtained
A solution of 6 (0.150 g, 0.325 mmol) in DMF (3 mL)
was stirred with sodium hydride (0.06 g, 60% suspension in
mineral oil, washed with hexanes before use) at 0°C for
20 min and a solution of p-methoxybenzyl chloride (300 µL,
2.21 mmol) in DMF (2 mL) was added dropwise in 6 min.
The resulting mixture was stirred at room temperature for
4 h when TLC showed that all starting material had been
consumed. Water (15 mL) was added and the resulting mix-
ture was extracted with pentane (6 × 20 mL). The combined
extracts were dried (Na2SO4) and concentrated to a liquid
residue that was purified with dry flash column chromatog-
raphy using a gradient elution from hexane to ethyl acetate
23
24
as a syrup (0.514 g, 63% yield); [α]D 24 (c 6.76, CHCl3);
to yield 7 (0.169 g, 89% yield); [α]D –0.11 (c 2.48,
© 1999 NRC Canada

490
Can. J. Chem. Vol. 77, 1999
1
CHCl3); H NMR δ: 3.63–3.70 (m, 4H, H-1, H-5), 3.75 (s,
3H, OCH3), 3.85–3.95 (m, 5H, H-2, H-3, H-4, CH2 All),
(34) (2.22 g, 1.3 equiv.). The reaction was complete in
30 min. The reaction solution was diluted with dichloro-
methane (30 mL), then washed with saturated sodium chlo-
ride containing 5% triethylamine (3 × 30 mL), and with
tetraethylammonium bromide (TEAB, 2 M, 2 × 30 mL). The
organic solution was dried (MgSO4) and concentrated to a
syrup that was purified with dry flash column chromatogra-
phy using as eluant hexane:ethyl acetate 5:1 containing 4%
triethylamine. Compound 10 was obtained as a colourless
syrup (3.86 g, 83% yield); [α]D 4.4 (c 3.03, CHCl3); H
NMR δ: 1.14 (m, 12H, CH3 iPr), 2.48–2.60 (m, 2H,
OCH2CH2CN), 3.36–4.05 (m, 13H, H-2–H-6, 2 CH iPr,
OCH2CH2CN, OCH3), 3.53 (s, OCH3), 4.31 (d, 1H, H-1,
J1,2 = 7.6 Hz), 4.64–4.97 (m, 6H, 3 CH2Ph), 7.15–7.37 (m,
15H, H-Ph); 13C NMR δ: 19.9, 20.0 (OCH2CH2CN), 24.3,
24.4 (CH3 iPr), 42.7, 42.9, 43.1 (CH iPr), 56.5, 56.6
(OCH3), 58.1, 58.4 (OCH2CH2CN), 62.2, 62.4, 62.7 (C-6),
74.3, 74.6, 75.3 (3 CH2Ph, C-5), 77.5 (C-3), 82.1 (C-2),
84.26, 84.34 (C-4), 104.3 (C-1), 117.3, 117.5 (CN), 127.3–
128.1, 137.9, 138.3 (C-Ph); 31P NMR δ: 150.0, 149.4. Exact
Mass (FAB) calcd. for C37H50N2O7P [M + H]+: 665.3356,
C37H48N2O7P [M – H]+: 663.3199; found: 665.3323 [M +
H]+, 663.3210 [M – H]+.
4.42 (s, 2H, CH2PhOCH3-p), 4.57–4.72 (m, 6H,
3
CHAHBPh, JA,B = –11.6 Hz, –11.9 Hz), 5.10–5.28 (m, 2H,
CH2ϭ), 5.80–5.93 (m, 1H, -CHϭ), 6.82 (d, 2H, ½ AA′BB′
pattern, pMPh, J = 8.5 Hz), 7.18–7.25 (m, 17H, H-Ph); 13C
NMR δ: 55.1 (OCH3), 69.6, 70.0 (C-1, C-5), 72.1, 72.2,
72.3, 73.7 (3 CH2Ph, 1 CH2 All), 72.8 (CH2PhOCH3-p),
78.4, 78.6 (C-2, C-3, C-4), 113.6 (2C o-CH pMBn), 116.6
(CH2ϭ), 134.8 (-CHϭ), 127.3–129.2, 130.4, 138.6, 159.0
(C-Ph). Exact Mass (FAB) calcd. for C37H42O6Na [M +
Na]+: 605.2879; found: 605.2936.
20
1
2,3,4-Tri-O-benzyl-5-O-p-methoxybenzyl-L-ribitol, 8
A solution of 7 (6.052 g, 10.4 mmol) in aqueous methanol
(methanol:water 5:1, 55 mL) was stirred with Pd/C (10%,
0.60 g) at 60°C for 10 h. The reaction mixture was filtered
through Celite and diluted with water (30 mL). The resulting
mixture was extracted with dichloromethane (4 × 50 mL).
The combined extracts were dried (Na2SO4), and concen-
trated to a liquid residue that was purified with dry flash col-
24
umn chromatography to yield 8 (3.88 g, 69% yield); [α]D
11 (c 2.29, CHCl3); H NMR δ: 3.5–4.0 (m, 7H, H-1–H-5),
1
3.8 (s, 3H, OCH3), 4.4 (s, 2H, CH2PhOCH3-p), 4.5–4.8 (m,
6H, 3 CH2Ph), 6.82 (½ AA′BB′ pattern, 2H, o-CH pMBn, J
= 8.5 Hz), 7.25 (m, 17H, H-Ph); 13C NMR δ: 55.3 (OCH3),
61.4 (C-1), 69.3 (C-5), 71.9, 72.4, 73.0, 74.0 (3 CH2Ph, 1
CH2PhOCH3-p), 78.2, 78.9, 79.1 (3C, C-2, C-3, C-4), 113.8
(2C o-CH pMBn), 127.7–129.4, 130.3, 138.2, 138.3, 159.2
(C-Ph); ESI-MS (g/mol): 565 for [C34H38O6Na]+ [M + Na]+
(100), isotopic peaks of [M + Na]+: 566 (calcd. 38.6, mea-
sured 38.4), 567 (calcd. 8.5, measured 8.5), 568 (calcd. 1.3,
measured 1.2); 581 for [C34H38O6K]+ [M + K]+; calcd. mass
542.3 (M); ESI-MS (g/mol) (in the presence of trace amount
of formic acid): 543 for C34H39O6 [M + H]+, 565 for
C34H38O6Na [M + Na]+, 581 for C34H38O6K [M + K]+.
(tert-Butyldimethyl)silyl 2-azido-4,6-O-benzylidene-2-
deoxy-3-O-methyl- -D-galactopyranoside, 11
A solution of tert-butyldimethylsilyl 2-azido-4,6-O-benzyl-
idene-2-deoxy-β-D-galactopyranoside (29) (3.14 g 7.71 mmol)
in DMF (50 mL) was stirred with NaH (0.78 g, 60% suspen-
sion in mineral oil, washed with pentane 3 times before use)
at 0°C for 0.5 h before methyl iodide (1.2 mL) was added.
The reaction mixture was stirred at 0°C for 15 min, after
which cold water (2 mL) was added. The resulting mixture
was extracted with dichloromethane (3 × 30 mL). The com-
bined extracts were dried (Na2SO4) and concentrated to a
syrup that was purified by low-pressure flash column chro-
matography using as eluent hexane:ethyl acetate 4:1 to yield
20
11 as a syrup (3.14 g, 97% yield); [α]D 8.5 (c 2.31,
Methyl 2,3,4-tri-O-benzyl- -D-glucopyranoside, 9
CHCl3); 1H NMR (400 MHz) δ: 0.10, 0.12 (2 s, 6H, CH3Si),
0.88 (s, 9H, SiC(CH3)3), 3.03 (dd, 1H, H-3, J2,3 = 10.5 Hz,
J3,4 = 3.7 Hz), 3.25 (br d, 1H, H-4,), 3.41 (s, 3H, OCH3),
3.62 (dd, 1H, H-2, J1,2 = 7.8 Hz), 3.97 (dd, 1H, H-6a, J6a,6b
= 12.2 Hz, J6a,5 = 2.0 Hz), 4.13–4.20 (m, 2H, H-5, H-6b),
4.47 (d, 1H, H-1), 5.50 (s, 1H, CHPh), 7.25–7.45 (m, 5H, H-
Ph); 13C NMR δ: –4.87, –4.14 (2 CH3Si), 18.0 (SiC(CH3)3),
25.6 (SiC(CH3)3), 57.2 (OCH3), 64.3 (C-2), 66.5 (C-4), 69.3
(C-6), 71.7 (C-5), 79.6 (C-3), 97.3 (C-1), 101.2 (CHPh),
126.4, 128.7, 129.0, 137.7 (C-Ph). Exact Mass (FAB) calcd.
for C20H30N3O5Si [M – H]+: 420.1955; found: 420.196.
A solution of methyl 2,3,4-tri-O-benzyl-6-O-trityl-β-D-
glucopyranoside (40) (19.52 g, 27.6 mmol) in glacial acetic
acid (35 mL) was treated with a solution of HBr in acetic
acid (30% w/v, 7.5 mL, 1.0 equiv.) at 0oC. The resulting pre-
cipitate was removed by filtration and washed with glacial
acetic acid. The combined filtrate and washings were poured
onto a mixture of ice and solid sodium carbonate and the re-
sulting mixture was extracted with dichloromethane (3 ×
100 mL). The combined extracts were washed with saturated
sodium hydrogen carbonate solutions (2 × 100 mL), and wa-
ter (3 × 100 mL), then dried (Na2SO4) and concentrated to a
solid that was purified by flash chromatography using hex-
ane and ethyl acetate (2:1) as eluent. The resulting solid
(9.61 g, 75% yield) was recrystallized from ether and hex-
2-Azido-4,6-O-benzylidene-2-deoxy-3-O-methyl- -D-
galactopyranosyl trichloroacetimidate, 12
20
ane to give fine needles of 9: mp 90–91oC; [α]D +8.8
A solution of 11 (1.53 g, 3.64 mmol) in dichloromethane
(100 mL) was cooled to –25°C under argon and a solution of
tetrabutylammonium fluoride (TBAF) in THF (1 M,
18.2 mL, 5 equiv.) was added dropwise. TLC analysis showed
that all the starting material had been consumed in 10 min.
The reaction mixture was poured into ice-water and the re-
sulting mixture was extracted with dichloromethane (3 ×
100 mL). The combined extracts were dried (Na2SO4) and
concentrated to a residue, a mixture of anomers of 2-azido-
20
(c 0.34, CHCl3) (lit. (40) mp 91–92oC; [α]D +9.9).
(Methyl 2,3,4-tri-O-benzyl- -D-glucopyranosid-6-yl) 2-
cyanoethyl (N,N-diisopropyl) phosphoramidite, 10
To a solution of 9 (3.25 g, 7.00 mmol) in dichloromethane
(70 mL), stirred under argon at room temperature, were
added diisopropylethylamine (DIPEA, 3.4 mL, 2.8 equiv.)
and chloro 2-cyanoethyl (N,N-diisopropyl)phosphoramidite
© 1999 NRC Canada

Qin and Grindley
491
4,6-O-benzylidene-2-deoxy-3-O-methyl-D-galactopyranose.
To a solution of the anomeric mixture in dichloromethane
(50 mL) was added trichloroacetonitrile (4.5 mL, 45 mmol).
The solution was cooled to 0°C and sodium hydride (0.19 g,
60% suspension, washed with pentane three times before
use) was added. After 1 h the reaction mixture was filtered
through Celite and concentrated to a foam that was purified
with low-pressure flash column chromatography, using hex-
(m, 3H, H-5, H-6′, H-1′, J1′,2′ = 7.9 Hz), 4.53–4.96 (m, 12H,
CH2Ph), 7.20–7.70 (m, 30H, H-Ph); 13C NMR δ: 19.14
(OCH2CH2CN), 57.03 (OCH3), 61.00, 61.05 (C-1), 61.62,
61.77 (OCH2CH2CN), 66.58, 67.23, 67.37 (C-5, C-6′),
72.00, 72.32, 72.43 (CH2Ph), 73.30 (C-5′), 73.79, 73.90,
74.85, 74.91, 75.45 (CH2Ph), 76.92 (C-4′), 77.73, 77.79,
78.03, 78.75, 78.85 (C-2, C-3, C-4), 82.06 (C-2′), 84.18,
84.25 (C-3′), 104.5 (C-1′), 116.3 (CN), 127.7–128.4, 137.6–
138.3 (C-Ph); 31P NMR δ: –0.72, –0.77; ESI-MS (g/mol)
1024 for [C57H64O13NPNa]+ [M + Na]+ (100), isotopic
peaks: 1025 (calcd. 65.2, measured 64.9), 1026 (calcd. 23.5,
measured 23.5), 1027 (calcd. 5.7, measured 5.8); 1040 for
[C57H64O13NPK]+ [M + K]+; calcd. mass 1001.4 (M).
Following the same procedure, starting from 10 (1.389 g,
2.09 mmol) and 8 (1.133 g, 1.0 equiv.), 13 was obtained
quantitatively and was treated with water and DDQ directly
without further purification. Compound 14 was obtained in a
57% yield (1.190 g) based on 10.
ane:ethyl acetate 5:1 as eluent, to yield 12 as a foam (1.14 g,
24
69% yield); [α]D
128 (c 0.84, CHCl3); 1H NMR
(400 MHz) δ: 3.56 (s, 3H, OCH3), 3.87 (m, 2H, H-3, H-5),
4.08 (dd, 1H, H-6a, J6a,6b = -12.7 Hz, J6a,5 = 1.5 Hz), 4.15
(dd, 1H, H-2, J2,3 = 10.3 Hz), 4.32 (dd, 1H, H-6b, J5,6b
=
1.5 Hz), 4.48 (br d, 1H, H-4, J4,3 = 2.9 Hz), 5.59 (s, 1H,
CHPh), 6.51 (d, 1H, H-1, J1,2 = 3.3 Hz), 7.34–7.53 (m, 5H,
H-Ph), 8.52 (s, 1H, NH); 13C NMR δ: 56.60 (OCH3), 58.20
(C-2), 65.23(C-4), 69.04 (C-6), 71.78 (C-5), 76.26 (C-3),
95.64 (C-1), 101.1 (CHPh), 126.2, 128.2, 129.1, 137.4 (C-
Ph), 160.59 (CϭNH).
(Methyl 2,3,4-tri-O-benzyl- -D-glucopyranosid-6-yl) 2-
cyanoethyl 1-(2-azido-4,6-O-benzylidene-2-deoxy-3-O-
methyl- -D-galactopyranosyl) 2,3,4-tri-O-benzyl-D-
ribityl-5-phosphate, 15
(Methyl 2,3,4-tri-O-benzyl- -D-glucopyranosid-6-yl) 2-
cyanoethyl 2,3,4-tri-O-benzyl-D-ribityl-5-phosphate, 14
To a stirred solution of 10 (0.412 g, 0.62 mmol) and 8
(0.343 g, 1.02 equiv.) in acetonitrile (10 mL) under argon
was added tetrazole (77.5 mg, 1.8 equiv.). After 1 h, the
reaction mixture was cooled to 0°C and a solution of m-
chloroperbenzoic acid (MCPBA, 0.544 g, 2.6 equiv.) in di-
chloromethane was added dropwise. After 20 min, the reac-
tion mixture was diluted with dichloromethane (20 mL) and
the resulting solution was washed with TEAB (2 M, 1 ×
20 mL) and water (1 × 30 mL). The dichloromethane solu-
tion was dried (MgSO4) and concentrated to a residue that
was purified by low-pressure flash column chromatography
using hexane:ethyl acetate 1:1 as eluent. (Methyl 2,3,4-tri-O-
benzyl-β-D-glucopyranosid-6-yl) 2-cyanoethyl 1-p-methoxy-
benzyl-2,3,4-tri-O-benzyl-D-ribityl-5-phosphate (13) was ob-
A solution of 14 (0.845 g, 0.844 mmol) and 12 (0.568 g,
1.2 equiv.) in acetonitrile (30 mL) was stirred with 4 Å mo-
lecular sieves for 2 h, then cooled to -40°C and treated with
TMS triflate (120 µL, 0.74 equiv.). After 30 min, pyridine
(1 mL) was added and the reaction mixture was filtered
through Celite. The filtrate was washed with water (3 ×
80 mL), dried (Na2SO4), and concentrated to a syrup. The
product was purified with low-pressure flash column chro-
matography using hexane:ethyl acetate 1:1.3 as eluent. The
title compound 15 was obtained as a colourless syrup
20
1
(0.892 g, 82% yield); [α]D 8.2 (c 3.40, CHCl3); H NMR
δ: 2.39 (t, 2H, OCH2CH2CN, J = 6.4 Hz), 3.13 (dd, 1H,
′′
H-3 , J2′′,3′′ = 10.27 Hz, J3′′,4′′ = 3.4 Hz), 3.18 (br s, 1H,
H-5 ), 3.38 (t, 1H, H-2, J1,2 = 8.3, J2,3 = 8.8 Hz), 3.45 (m,
20
′′
tained as a colorless syrup (0.480 g, 69% yield); [α]D 6.5
(c 3.72, CHCl3); 1H NMR (400 MHz) δ: 2.41 (m, 2H,
OCH2CH2CN), 3.30–4.12 (m, H-1–H-5, H-1′–H-6′, OCH2CH2CN,
1H, H-5), 3.48, 3.50 (2 s, 6H, 2 OCH3), 3.51 (m, 1H, H-4),
′′
3.63 (t, 1H, H-3, J3,4 = 8.8 Hz), 3.77 (dd, 1H, H-2 , J1′′,2′′
=
=
7.8 Hz), 3.87 (dd, 1H, H-1′A, J1′A,1′B = 10.8 Hz, J1′A,2′
3.7 Hz), 3.98 (m, H-2, H-3, H-4, H-6′, OCH2CH2CN), 4.16
OCH3), 3.49, 3.51, 3.75 (3s, OCH3), 4.39 (d, 1H, H-1, J1,2
=
8.3 Hz), 4.15–4.95 (m, 14H, CH2Ph), 6.79–6.84, 7.16–7.35
(m, 34H, H-Ph); 13C NMR δ: 19.00 (OCH2CH2CN), 55.14,
57.03 (2 × OCH3), 61.64 (OCH2CH2CN), 66.57, 67.45, 67.76
(C-5, C-6′), 69.20, 69.30 (C-1), 72.38, 72.89 (CH2Ph), 73.35
(C-5′), 73.60, 73.69, 74.62, 74.90, 75.48 (CH2Ph), 78.03 (C-
2, C-3, C-4), 77.00, 82.11, 84.26, 84.33 (C-2′, C-3′, C-4′),
104.6 (C-1′), 116.3 (CN), 113.6, 127.6-129.2, 130.2, 137.8–
138.4, 159.1 (C-Ph); 31P NMR δ: –0.73, –0.81.
′′
(dd, 1H, H-1′b, J1′b,2′ = 3.7 Hz), 4.21 (m, H-5′a, H-4 ), 4.26
′′
(d, H-1 ), 4.29 (d, H-1), 4.30 (m, H-6a), 4.37 (m, H-5′b), 4.48
(m, H-6b), 4.55–4.95 (m, 6 CH2Ph), 5.53 (s, 1H, CHPh),
7.2–7.6 (m, 35H, H-Ph); 13C NMR δ: 19.0 (OCH2CH2CN),
′′
′′
′′
57.1 (2 × OCH3), 61.6 (OCH2CH2CN), 62.2 (C-2 ), 66.4 (C-5 ),
′′
66.6 (C-5′), 67.4 (C-6), 68.3 (C-1′), 69.0 (C-6 ), 71.5 (C-4 ),
72.2 (2 CH2Ph), 73.3 (C-5), 73.6, 74.6, 74.9, 75.4 (4 CH2Ph),
′′
77.0 (C-4), 77.8 (3C, C-2′, C-3′, C-4′), 79.9 (C-3 ), 82.1 (C-
To a solution of 13 (0.330 g, 0.294 mmol) in dichloro-
methane (9 mL) was added water (0.5 mL) and 2,3-dichloro-
5,6-dicyanoquinone (DDQ, 0.082 g, 1.2 equiv.) and the mix-
ture was stirred vigorously for 4 h. The organic layer was
extracted with saturated sodium hydrogen carbonate (3 ×
50 mL), then dried (Na2SO4) and concentrated to a residue
that was purified with low-pressure flash column chroma-
tography to yield 14 as a colourless syrup (0.185 g, 63%
′′
2), 84.3 (C-3), 101.1 (CHPh), 101.9 (C-1 ), 104.5 (C-1),
116.3 (CN), 126.3–128.9, 137.5–138.4 (C-Ph); 31P NMR δ:
–0.73, –0.83. Exact Mass (ESI) calcd. for C71H79N4O17PNa
[M + Na]+: 1313.5076; found: 1313.5092.
(Methyl 2,3,4-tri-O-benzyl- -D-glucopyranosid-6-yl) 2-
cyanoethyl 1-O-(2-acetamido-4,6-O-ben-zylidene-2-
deoxy-3-O-methyl-β-D-galactopyranosyl) 2,3,4-tri-O-
benzyl-D-ribityl-5-phosphate, 16
A solution of 15 (0.651 g, 0.505 mmol) in thiolacetic acid
(4 mL) was stirred under argon at room temperature for 17 h
and then was concentrated to a liquid residue. Low-pressure
20
1
yield); [α]D 5.3 (c 1.36, CHCl3); H NMR (400 MHz) δ:
2.24, 2.29 (2 br t, 1H, 1-OH), 2.49 (m, 2H, OCH2CH2CN),
3.39 (m, 1H, H-2′), 3.45 (m, 1H, H-5′), 3.50, 3.52 (2s, 3H,
OCH3), 3.55 (br s, 1H, H-4′), 3.65 (m, 1H, H-3′), 3.86–3.95
(m, 1H, H-1), 3.99–4.05 (m, 2H, OCH2CH2CN), 4.18–4.50
© 1999 NRC Canada

492
Can. J. Chem. Vol. 77, 1999
flash column chromatography using ethyl acetate as eluent
4.4 Hz, J5′′,6b′′ = 8.1 Hz), 3.77 (dd, 1H, H-3′, J2′,3′ = 7.0 Hz,
J3′,4′ = 4.8 Hz), 3.77 (dd, 1H, H-6 a, J6a′′,6b′′ = 11.4 Hz), 3.83
20
′′
yielded 16 as a foam (0.493 g, 75% yield); [α]D +22.3 (c
1
′′
0.35, CHCl3); H NMR (400 MHz) δ: 1.80, 1.81 (2 s, 3H,
(dd, 1H, H-2′, J1′.2′ = 6.4 Hz), 3.84 (dt, H-6 b), 3.89 (m, 1H,
′′
CH3CO), 2.32–2.36, 2.46-2.50 (m, 2H, OCH2CH2CN), 3.41,
3.48, 3.51 (3 s, OCH3), 5.527, 5.533, 5.539 (3s, CHPh),
6.10, 6.38 (2 d, 1H, NH, JNH,2′′ = 6.8, 7.3 Hz), 7.20–7.48 (m,
35H, H-Ph); 13C NMR δ: 18.99, 19.08, 19.15
H-4′), 3.93 (dd, 1H, H-2 ), 3.95 (d, 2H, H-1′), 3.99 (t, 1H, H-
5′a, J5′a,4′ = 6.1 Hz), 4.06 (m, 1H, H-5′b, J5′b,5′a = 10.4 Hz,
J5′b,4′ = 5.4 Hz), 4.07 (m, 1H, H-6a), 4.17 (dddd, 1H, H-6b,
J6b,6a = 11.6 Hz, J6b,5 = 5.2 Hz, J6b,P = 2.1 Hz), 4.22 (d, 1H,
′′
′′
(OCH2CH2CN), 23.49 (CH3CO), 54.31, 54.50, 54.62 (C-2 ),
H-4 , J4′′,3′′ = 3.3 Hz), 4.39 (d, 1H, H-1, J1,2 = 8.0 Hz), 4.49
56.89, 57.06 (OCH3), 61.78 (OCH2CH2CN), 66.39 (C-5 ),
(d, 1H, H-1 , J1′′,2′′ = 8.6 Hz); 13C NMR (D2O) δ: 23.94
′′
′′
′′
′′ ′′
66.67 (C-5′), 67.60 (C-6), 68.41 (C-1′), 69.35 (C-6 ), 71.89
(CH2Ph), 72.36 (C-4 ), 72.68 (CH2Ph), 73.36 (C-5), 73.47,
(CH3CO), 51.99 (C-2 ), 57.18 (3 -OCH3), 58.10 (1-OCH3),
′′
′′
′′
61.88 (C-6 ), 64.35 (C-4 ), 64.97 (d, C-6, J6,P = 4.8 Hz),
67.41 (d, C-5′, J5′,P = 4.8 Hz), 69.87 (C-4), 71.34 (C-2′),
71.41 (C-1′), 71.48 (d, C-4′, J4′,P = 7.6 Hz), 72.19, (C-3′),
74.93, 75.09, 75.49 (4 CH2Ph), 77.0 (C-4), 77.61, 77.95,
′′
78.20, 78.51, 78.66 (C-2′, C-3′, C-4′, C-3 ), 82.13 (C-2),
′′
′′
73.86 (C-2), 75.38 (d, C-5, J5,P = 7.6 Hz), 75.78 (C-5 ),
84.27, 84.35 (C-3), 99.61, 99.89 (C-1 ), 101.0 (CHPh), 104.6
′′
′′
(C-1), 116.5 (CN), 126.2–128.9, 137.8–138.4 (C-Ph), 170.9,
171.0 (CH3CO); 31P NMR δ: 0.17, –0.68, –0.73, –0.96. Ex-
act Mass (ESI) calcd. for C73H83N2O18PNa [M + Na]+:
1329.5276; found: 1329.5275.
76.36 (C-3), 80.70 (C-3 ), 102.67 (C-1 ), 104.12 (C-1),
175.65 (C = O); 31P NMR (D2O) δ: 1.73. Exact Mass (ESI)
calcd. for C21H40NO18PNa [M + H]+: 648.1881; found:
648.1900; calcd. for C21H39NO18PNa2 [M + Na]+: 670.1700;
found: 670.1697; calcd. for C21H38NO18PNa3 [M – H + 2Na]+:
692.1519; found: 692.1536.
Triethylammonium (methyl 2,3,4-tri-O-benzyl-β-D-
glucopyranosid-6-yl) 1-O-(2-acetamido-4,6-O-
benzylidene-2-deoxy-3-O-methyl-β-D-galactopyranosyl)
2,3,4-tri-O-benzyl-D-ribityl-5-phosphate, 17
1-O-(2-Azido-4,6-O-benzylidene-2-deoxy-3-O-methyl- -D-
galactopyranosyl) 5-O-p-methoxybenzyl-2,3,4-tri-O-
benzyl-L-ribitol, 19, and 1-O-(2-azido-4,6-O-benzylidene-
2-deoxy-3-O-methyl- -D-galactopyranosyl) 2,3,4-tri-O-
benzyl-L-ribitol, 20
A solution of 16 (0.312 g, 0.239 mmol) in a mixture of
triethylamine, pyridine, and water (3:1:1, 20 mL) was stirred
at room temperature for 2 h. The reaction solution was con-
centrated to a foam of 17 (0.323 g, 99.6% yield); [α]D25 –8.2
A solution of 12 (0.460 g, 1.02 mmol) and 8 (0.554 g,
1.02 mmol) in acetonitrile (20 mL) was stirred with 4 Å mo-
lecular sieves (1 g) for 1 h at room temperature. Then the
solution was cooled to –50°C and trimethylsilyl tri-
fluoromethanesulfonate (TMS triflate, 130 µL, 0.66 equiv.)
was added. The reaction solution was stirred at –50°C
for 40 min after which pyridine (1 mL) was added and the
reaction mixture was filtered through Celite. The filtrate was
washed with water (3 × 80 mL), dried (Na2SO4), and con-
centrated to a brownish syrup that was purified with low-
pressure flash column chromatography using hexane:ethyl
acetate 2:1 as eluent. Compound 19 was obtained as a syrup
1
(c 0.48, CH3OH); H NMR (400 MHz, CD3OD) δ: 1.22 (t,
3H, CH3CH2), 1.84 (s, 3H, CH3CO), 2.95 (q, 2H, CH3CH2),
′′
′′
3.15–3.55 (m, 12H, H-5 , H-2, H-3, H-3 , H-4, H-5, 2
OCH3), 3.34, 3.38 (2 s, 2 × OCH3), 3.88–4.36 (m, 14H, H-
′′
′′
′′
′′
1′–H-5′, H-6 , H-6, H-2 , H-4 , H-1), 4.23 (d, H-1, J1,2
=
7.3 Hz)), 4.05(br dd, H-2 , J1′′,2′′ = 8.8 Hz, J2′′,3′′ = 10.3 Hz),
′′
4.53 (d, 1H, H-1 ), 4.54–4.83 (m, 6 CH2Ph), 5.54 (s, 1H,
CHPh), 7.10–7.48 (m, 35H, H-Ph); 13C NMR (CD3OD) δ:
9.31 (CH3CH2), 23.56 (CH3CO), 47.72 (CH3CH2), 52.77
′′
(C-2 ), 57.20, 57.52 (2 × OCH3), 65.70, 66.32 (C-5′, C-6),
′′
′′
′′
67.95 (C-5 ), 69.81 (C-1′), 70.45 (C-6 ), 72.91 (C-4 ), 73.53,
74.86, 75.58 (4 CH2Ph), 75.77 (C-5), 75.95, 76.48 (2
CH2Ph), 79.22 (C-4), 80.10, 80.28, 80.60 (C-2′, C-3′, C-4′,
C-3 ), 83.47 (C-2), 85.73 (C-3), 102.2 (CHPh), 102.8 (C-1 ),
105.9 (C-1), 127.2–129.9, 139.8–140.2 (C-Ph), 173.6 (C =
O); 31P NMR (CD3OD) δ: 1.41. Exact Mass (ESI) calcd. for
C70H79NO18PNa2 [M – NHEt3 + 2Na]+: 1298.4830; found:
1298.4814.
20
1
(0.145 g, 17% yield); [α]D 2.2 (c 0.68, CHCl3); H NMR
(400 MHz) δ: 3.11 (dd, 1H, H-3′, J2′,3′ = 10.3 Hz, J3′,4′
=
′′
′′
3.4 Hz), 3.20 (br s, 1H, H-5′), 3.49 (s, 3H, OCH3), 3.61–4.06
(m, 11H, H-2, H-3, H-4, H-6′a, H-1a, H-5, H-2′, OCH3),
3.75 (s, OCH3), 4.18–4.34 (m, 4H, H-4′, H-6′b, H-1b, H-1′,
J1′,2′ = 8.3 Hz), 4.42 (t, 2H, CH2PhOMe-p), 4.55–4.84 (m,
6H, 3 CH2Ph), 5.53 (s, 1H, CHPh), 6.82 (d, 2H, AA′BB′ pat-
tern, Ho-pMPh, J = 8.8 Hz), 7.15–7.55 (m, 22H, H-Ph); 13C
NMR δ: 55.12, 57.14 (2 × OCH3), 62.16 (C-2′), 66.34 (C-
5′), 69.07, 69.44, 70.31 (C-1, C-6′, C-5), 71.63 (C-4′), 72.34,
72.56, 72.84, 73.53 (4 CH2Ph), 78.22, 78.91, 79.03 (C-2, C-
3, C-4), 80.07 (C-3′), 101.1 (CHPh), 102.3 (C-1′), 113.7,
127.2-129.7, 130.4, 137.6–138.7, 159.1 (C-Ph).
Sodium (methyl β-D-glucopyranosid-6-yl) 1-O-(2-
acetamido-2-deoxy-3-O-methyl-β-D-galacto-pyranosyl)-D-
ribityl-5-phosphate, 18
A solution of 17 (0.323 g, 0.238 mmol) in a mixture of
ethyl acetate, methanol, and acetic acid (10:14.5:0.5, 25 mL)
was stirred with Pd/C (10%, 0.38 g) under hydrogen (1 atm
(= 101.3 kPa)) for 70 h. The reaction mixture was filtered
through Celite and the filtrate was concentrated. A solution
of the residue in water (3 mL) was stirred with Amberlite
IR-120 (Na+) resin overnight. The filtrate was concentrated
Another product (20) was obtained as a syrup (0.118 g,
24
16% yield); [α]D
–3.5 (c 0.31, CHCl3); 1H NMR
(400 MHz) δ: 2.29 (br t, 5-OH), 3.14 (dd, 1H, H-3′, J2′,3′
=
10.3 Hz, J3′,4′ = 3.4 Hz), 3.26 (br s, 1H, H-5′), 3.50 (s, 3H,
OCH3), 3.65–4.07 (m, 8H, H-2, H-3, H-4, H-6′a, H-1a, H-5,
H-2′), 4.18–4.34 (m, 4H, H-4′, H-6′b, H-1b, H-1′), 4.53–4.86
(m, 6H, 3 CH2Ph), 5.58 (s, 1H, CHPh), 7.15–7.55 (m, 20H,
H-Ph); 13C NMR δ: 57.19 (OCH3), 61.25 (C-5), 62.16 (C-
2′), 66.44 (C-5′), 69.45, 69.74 (C-1, C-6′), 71.63 (C-4′),
72.07, 72.70, 73.81 (3 CH2Ph), 78.78, 79.09 (3C C-2, C-3,
to give 18 quantitatively; [α]D –9.9 (c 0.68, H2O); 1H
20
NMR (500 MHz, D2O) δ: 2.04 (s, 3H, CH3CO), 3.28 (m,
′′
1H, H-2, J2,3 = 9.4 Hz), 3.40 (s, 3H, 3 -OCH3), 3.44 (dd,
′′
1H, H-3 , J2′′,3′′ = 10.9 Hz), 3.50 (m, 2H, H-3,4), 3.57 (s, 3H,
′′
1-OCH3), 3.58 (m, 1H, H-5), 3.66 (br dd, 1H, H-5 , J5′′,6a′′
=
© 1999 NRC Canada

Qin and Grindley
493
C-4), 80.07 (C-3′), 101.2 (CHPh), 102.2 (C-1′), 126.3–128.3,
137.6–138.4 (C-Ph); ESI-MS (g/mol): 706 for [M + Na –
N2]+, 734 for [M + Na]+, 750 for [M + K]+; calculated mass:
711.3 (M).
4-tri-O-benzyl-L-ribityl-5-phosphate (22) was obtained as a
colourless foam (0.066 g, 47% yield); H NMR (400 MHz)
1
δ: 1.80, 1.84 (2 s, CH3CO), 2.34–2.46 (m, OCH2CH2CN),
′′
′′
3.30–4.5 (m, H-1–H-6, H-1′–H-5′, H-1 –H-6 , OCH3,
OCH2CH2CN), 3.41 (s, OCH3), 4.52–4.96 (m, 6 CH2Ph),
5.54, 5.55 (2 s, CHPh), 5.9, 6.2, 6.5 (2 br, NH), 7.2–7.6 (m,
35H, H-Ph); 13C NMR δ: 19.13 (OCH2CH2CN), 23.54
A solution of 19 (0.107 g, 0.128 mmol) in dichloro-
methane (5 mL) was stirred with water (0.4 mL) and 2,3-
dichloro-5,6-dicyano-quinone (DDQ, 0.0327 g, 1.1 equiv.)
vigorously for 4 h. The reaction mixture was extracted with
saturated sodium hydrogen carbonate (3 × 20 mL), then
dried (Na2SO4) and concentrated to a syrup that was purified
with low-pressure column chromatography using ethyl ace-
tate and hexane (1:1.2) as eluent to yield 20 as a colourless
′′
(CH3CO), 54.36 (C-2 ), 56.88, 57.16 (2 × OCH3), 61.91
′′
(OCH2CH2CN), 66.46 (C-5 ), 66.70, 67.30, 67.84 (C-5′, C-
′′
′′
6), 68.63, 68.99, 69.40 (C-6 , C-1′), 72.23 (C-4 ), 72.49
(CH2Ph), 73.32, 73.39 (C-5), 73.53, 74.59, 74.94, 75.50 (4
CH2Ph), 76.16, 77.00, 77.62, 78.15, 78.33, 78.49 (C-4, C-2′,
1
syrup (0.059 g, 63% yield). The H and 13C NMR spectra of
′′
C-3′, C-4′, C-3 ), 82.10 (C-2), 84.28 (C-3), 99.52, 99.60
′′
(C-1 ), 100.9 (CHPh), 104.6 (C-1), 116.5 (CN), 126.2–128.8,
this product were identical in all respects to that of 20 above.
137.8–138.4 (C-Ph), 171.2 (CϭO); 31P NMR δ: –0.03,
–0.58, –1.49, –1.50.
(Methyl 2,3,4-tri-O-benzyl- -D-glucopyranosid-6-yl) 2-
cyanoethyl 1-O-(2-azido-4,6-O-benzylidene-2-deoxy-3-O-
methyl- -D-galactopyranosyl) 2,3,4-tri-O-benzyl-L-
ribityl-5-phosphate, 21
A solution of 22 (0.066 g, 0.051 mmol) in a mixture of
triethylamine, pyridine, and water (3:1:1, 5 mL) was stirred
at room temperature for 2 h. The solution was concentrated
20
to a foam of 23 (0.068 g, 99.6% yield); [α]D –7.9 (c 0.30,
To a solution of 10 (0.105 g, 1.1 equiv.) and 20 (0.102 g,
0.144 mmol) in acetonitrile (20 mL) was added tetrazole
(23.2 mg, 2.3 equiv.). The reaction solution was stirred un-
der argon for 40 min. The reaction mixture was cooled to
0°C and a solution of m-chloroperbenzoic acid (MCPBA,
0.142 g, 2.6 equiv.) in dichloromethane (5 mL) was added
dropwise. The reaction was complete in 20 min. The reac-
tion mixture was diluted with dichloromethane (30 mL) and
the organic layer was washed with TEAB (2 M, 1 × 30 mL)
and water (1 × 30 mL), then dried (MgSO4) and concen-
trated. The residue was purified with low-pressure flash col-
umn chromatography using hexane:ethyl acetate 1:2 as eluent.
The title compound 21 was obtained as colourless foam
CH3OH); 1H NMR (400 MHz, CD3OD) δ: 1.18 (t, 3H,
CH3CH2), 1.75 (s, 3H, CH3CO), 2.97 (q, 2H, CH3CH2),
3.22 (dd, 1H, H-2, J1,2 = 7.8 Hz, J2,3 = 9.29 Hz), 3.35–3.45
′′
′′
(m, H-4,5,5 ), 3.39, 3.43 (2 s, 2 × OCH3), 3.51 (m, H-3 ,3),
3.75 (dd, 1H, H-1′, J1′A,1′B = –10.8 Hz, J1′A,2′ = 6.4 Hz), 3.92
(m, H-3′,4′), 4.02 (m, H-2′,6), 4.25 (d, 1H, H-1), 4.40 (br d,
′′
′′
1H, H-4 , J3′′,4′′ = 3.4 Hz), 4.52 (d, 1H, H-1 , J1′′,2′′ = 8.8 Hz),
4.57–4.83 (m, 6 CH2Ph), 5.60 (s, 1H, CHPh), 7.1–7.5 (m,
35H, H-Ph); 13C NMR (CD3OD) δ: 9.40 (CH3CH2), 23.34
′′
(CH3CO), 47.67 (CH3CH2), 52.83(C-2 ), 57.16, 57.57 (2 ×
OCH3), 65.85 (C-6), 66.39 (C-5′), 68.04 (C-5 ), 70.54 (C-6 ),
′′ ′′
′′
71.37 (C-1′), 72.93 (C-4 ), 73.56, 73.77, 75.00, 75.50 (4
25
1
CH2Ph), 75.73 (C-5), 75.96, 76.41 (2 CH2Ph), 79.27 (C-4),
(0.155 g, 84% yield); [α]D 6.6 (c 0.33, CHCl3); H NMR
(400 MHz) δ: 2.42–2.46 (m, 2H, OCH2CH2CN), 3.14 (m,
′′
80.23 (3C C-3 , 2′, 3′, or 4′), 80.55 (C-3′ or 4′), 83.39 (C-2),
′′
85.71 (C-3), 102.2 (CHPh), 103.1 (C-1 ), 105.9 (C-1),
′′
′′
H-3 ), 3.35 (br d, 1H, H-5 ), 3.37 (m, 1H, H-2), 3.49 (m, 1H,
127.7–130.0, 138.5–140.5 (C-Ph), 173.5 (CϭO); 31P NMR
(CD3OD) δ: 1.22. Exact Mass (ESI) calcd. for
C70H79NO18PNa2 [M – NHEt3 + 2Na]+: 1298.4830; found:
1298.4808.
H-4), 3.50 (s, OCH3), 3.63 (m, 1H, H-3), 3.75–3.86 (m, 2H,
′′
′′
H-1′a, H-2 ), 3.88–4.15 (m, H-2′, H-3′, H-4′, H-6 a,
′′
′′
′′
OCH2CH2CN), 4.16–4.53 (m, H-1′b, H-6 b, H-5′, H-4 , H-1 ,
H-1, H-6), 4.54–4.96 (m, 6 CH2Ph), 5.54 (s, 1H, CHPh),
7.2–7.6 (m, 35H, H-Ph); 13C NMR δ: 19.24 (OCH2CH2CN),
57.11, 57.24 (2 × OCH3), 61.96 (OCH2CH2CN), 62.15
Sodium (methyl -D-glucopyranosid-6-yl) 1-O-(2-
acetamido-2-deoxy-3-O-methyl- -D-galactopyranosyl)-L-
ribityl-5-phosphate, 24
′′
′′
′′
(C-2 ), 66.45 (C-5 ), 66.83 (C-5′), 67.39, 67.69 (C-6), 69.08
′′
(C-6 ), 69.59 (C-1′), 71.63 (C-4 ), 72.46, 72.65 (2 CH2Ph),
73.30 (C-5), 73.56, 74.61, 74.94, 75.49 (4 CH2Ph), 77.0 (C-
4), 77.62, 78.18, 78.45 (C-2′, C-3′, C-4′), 80.06 (C-3"), 82.06
A solution of 23 (0.323 g, 0.238 mmol) in a mixture of
ethyl acetate, methanol, and acetic acid (10:14.5:0.5, 25 mL)
was stirred with Pd/C (10%, 0.38 g) under hydrogen (1 atm)
for 50 h. The reaction mixture was filtered through Celite
and the filtrate was concentrated. A solution of the residue in
water (3 mL) was stirred with Amberlite IR-120 (Na+) resin
overnight. The filtrate was concentrated to give 24 quantita-
′′
(C-2), 84.26 (C-3), 101.1 (CHPh), 102.2 (C-1 ), 104.5 (C-1),
116.2, 116.4 (CN), 126.2–134.4, 137.4–138.3 (C-Ph); 31P
NMR δ: –0.77, –1.28. Exact Mass (ESI) calcd. for
C71H79N4O17PNa [M + Na]+: 1313.5076; found: 1313.5102.
20
Triethylammonium (methyl 2,3,4-tri-O-benzyl- -D-
glucopyranosid-6-yl) 1-O-(2-acetamido-4,6-O-
benzylidene-2-deoxy-3-O-methyl-β-D-galactopyranosyl)
2,3,4-tri-O-benzyl-L-ribityl-5-phosphate, 23
A solution of 21 (0.139 g, 0.108 mmol) in thiolacetic acid
(2 mL) was stirred under argon at room temperature for
17 h. Then the reaction mixture was concentrated and was
purified by low-pressure flash column chromatography us-
ing ethyl acetate as eluent. (Methyl 2,3,4-tri-O-benzyl-β-D-
glucopyranosid-6-yl) 2-cyanoethyl 1-O-(2-acetamido-4,6-O-
benzylidene-2-deoxy-3-O-methyl-β-D-galactopyranosyl) 2,3,
tively; [α]D –14.2 (c 0.12, H2O); 1H NMR (500 MHz)
(D2O) δ: 2.04 (s, 3H, CH3CO), 3.28 (m, 1H, H-2, J2,3
′′
9.4 Hz), 3.41 (s, 3H, 3"-OCH3), 3.45 (dd, 1H, H-3 , J2′′,3′′
=
=
11.0 Hz), 3.50 (m, 2H, H-3,4), 3.57 (s, 3H, 1-OCH3), 3.58
′′
(m, 1H, H-5), 3.66 (br dd, 1H, H-5 , J5′′,6a′′ = 4.3 Hz, J5′′,6b′′
=
=
6.9 Hz), 3.740 (dd, 1H, H-1′a, J1′a,1′b = 11.6 Hz, J1′a,2′
6.3 Hz), 3.744 (t, 1H, H-3′, J2′,3′ = J4′,3′ = 6.4 Hz), 3.78 (dd,
′′
1H, H-6 a, J6a′′,6b′′ = 11.8 Hz, J6′′a,5′′ = 4.3 Hz), 3.83 (dd, 1H,
′′
′′
H-6 b, J6′′b,5′′ = 7.9 Hz), 3.93 (m, 1H, H-4′), 3.94 (dd, H-2 ,
J2′′.,3′′ = 10.6 Hz), 3.95 (m, 1H, H-2′), 3.98 (m, 1H, H-5′a),
4.07 (dd, 1H, H-1′b, J1′b,2′ = 5.2 Hz), 4.08 (m, H-5′b), 4.09
© 1999 NRC Canada

494
Can. J. Chem. Vol. 77, 1999
2684 (1974); (d) F. Fiedler and L. Glaser. J. Biol. Chem. 249,
2690 (1974).
15. A.R. Archibald and H.E. Coapes. Biochem. J. 124, 449 (1971).
16. H.U. Koch and W. Fischer. Biochemistry, 17, 5275 (1978).
17. W. Fischer, M.A. Schmidt, B. Jann, and K. Jann. Biochemis-
try, 21, 1279 (1982).
18. J.C. Richards and R.A. Leitch. Can. J. Chem. 68, 1574 (1990).
19. (a) L.M. Beynon, J.C. Richards, and M.B. Perry. Eur. J.
Biochem. 210, 119 (1992); (b) L.M. Beynon, M.B. Perry, and
J.C. Richards. Carbohydr. Res. 209, 211 (1991).
20. L.M. Beynon and J.C. Richards. Carbohydr. Res. 252, 263
(1994).
21. (a) T. Rundlöf and G. Widmalm. Anal. Biochem. 243, 228
(1996); (b) T. Rundlof, A. Weintraub, and G. Widmalm.
Carbohydr. Res. 291, 127 (1996).
22. N. Kojima, Y. Araki, and E. Ito. Eur. J. Biochem. 148, 479
(1985).
23. P. Branefors-Helander, C. Erbing, L. Kenne, and B. Lindberg.
Carbohydr. Res. 56, 117 (1977).
24. P.J. Garegg, B. Lindberg, K. Nilsson, and C.-G. Swahn. Acta
Chem. Scand. 27, 1595 (1973).
25. J. Baddiley, J.G. Buchanan, and B. Carss. J. Chem. Soc. 1869
(1957).
(dd, 1H, H-6a, J6a,5 = 2.9 Hz), 4.18 (dddd, 1H, H-6b, J6b,6a
=
′′
11.6 Hz, J6b,5 = 5.1 Hz, J6b,P = 2.1 Hz), 4.22 (d, 1H, H-4 ,
J4′′,3′′ = 3.2 Hz), 4.40 (d, 1H, H-1, J1,2 = 8.0 Hz), 4.48 (d, 1H,
H-1 , J1′′,2′′ = 8.5 Hz); 13C NMR δ: 23.10 (CH3CO), 51.98
′′
′′
′′
(C-2 ), 57.13 (3"-OCH3), 58.14 (1-OCH3), 61.88 (C-6 ), 64.30
(C-4 ), 64.96 (d, C-6, J6,P = 5.0 Hz), 67.36 (d, C-5′, J5′,P
′′
=
=
6.3 Hz), 69.87 (C-4), 71.34 (C-2′), 71.67 (d, C-4′, J4′,P
7.7 Hz), 71.70 (C-1′), 72.41, (C-3′), 73.89 (C-2), 75.38 (d, C-
′′
′′
5, J5,P = 7.8 Hz), 75.77 (C-5 ), 76.39 (C-3), 80.63 (C-3 ),
103.15 (C-1 ), 104.15 (C-1), 175.72 (CϭO); 31P NMR (D2O)
′′
δ: 1.71. Exact Mass (FAB) calcd. for C21H40NO18PNa [M +
H]+: 648.1881; found: 648.188.
Acknowledgments
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support, and Dr.
M.D. Lumdson and Dr. I. Burton for recording some of the
NMR spectra. We also thank the National Research Council
in Ottawa, the University of Ottawa, and the University of
Alberta for recording high-resolution MS spectra. Special
thanks go to Dr. O. Hindsgaul for arranging for the MS to be
run at the University of Alberta.
26. M.W. Ganfield and R.A. Pieringer. J. Biol. Chem. 255, 5164
(1980).
27. (a) J.E.G. van Dam, J. Breg, R. Komen, J.P. Kamerling, and
J.F.G. Vliegenthart. Carbohydr. Res. 187, 267 (1989); (b) L.
Kenne, B. Lindberg, and J.K. Madden. Carbohydr. Res. 73,
175 (1979); (c) C. Jones. Carbohydr. Res. 269, 175 (1995);
(d) L. Kenne and B. Lindberg. Carbohydr. Res. 184, 288 (1988);
(e) L.M. Beynon, J.C. Richards, and M.B. Perry. Can. J. Chem.
73, 41 (1995).
28. P. Smid, M. de Swart, W.P.A. Jörning, G.A. van der Marel,
and J.H. van Boom. J. Carbohydr. Chem. 12, 1073 (1993).
29. G. Grundler and R.R. Schmidt. Liebigs Ann. Chem. 1826 (1984).
30. (a) T.B. Grindley. Adv. Carbohydr. Chem. Biochem. 53, 17
(1998); (b). In Synthetic oligosaccharides: indispensable probes
for the life sciences. Edited by P. Kováꢀ. ACS, Washington.
1994. pp. 51–76; (c) S. David and S. Hanessian. Tetrahedron,
41, 643 (1985).
31. L. Chan and G. Just. Tetrahedron, 46, 151 (1990).
32. J.P.G. Hermans, L. Poot, M. Kloosterman, G.A. van der Marel,
C.A.A. van Boeckel, D. Evenberg, J.T. Poolman, P. Hoogerhout,
and J.H. van Boom. Recl. Trav. Chim. Pays-Bas, 106, 498 (1987).
33. B. Bernet and A. Vasella. Helv. Chim. Acta, 62, 1990 (1979).
34. N.D. Sinha, J. Biernat, J. McManus, and H. Köster. Nucleic
Acids Res. 12, 4539 (1984).
References
1. (a) J.B. Robbins, R. Austrian, C.-J. Lee, S.C. Rastogi, G.
Schiffman, J. Henrichsen, P.H. Makela, C.V. Broome, R.R.
Facklam, R.H. Tiesjema, and J.C. Parke, Jr. J. Infect. Dis. 148,
1136 (1983); (b) E. Alonsodevelasco, A.F.M. Verheul, J. Verhoef,
and H. Snippe. Microbiol. Rev. 59, 591 (1995); (c) K.M. Zangwill,
C.M. Vadheim, A.M. Vannier, L.S. Hemenway, D.P. Greenberg,
and J.I. Ward. J. Infect. Dis. 174, 752 (1996).
2. J. Henrichsen. J. Clin. Microbiol. 2759 (1995).
3. (a) U.B.S. Sorensen and J. Henrichsen. J. Clin. Microbiol. 25,
1854 (1987); (b) U.B.S. Sorensen, J. Blom, A. Birch-Andersen,
and J. Henrichsen. Infect. Immun. 56, 1890 (1988).
4. (a) T. Behr, W. Fischer, J. Peter-Katalinic, and H. Egge. Eur. J.
Biochem. 207, 1063 (1992); (b) R.A. Klein, R. Hartmann, H.
Egge, T. Behr, and W. Fischer. Carbohydr. Res. 256, 189 (1994).
5. S.C. Szu, R. Schneerson, and J.B. Robbins. Infect. Immun. 54,
448 (1986).
6. C.A. Laferriere, R.K. Sood, J.M. Demuys, F. Michon, and H.J.
Jennings. Vaccine, 15, 179 (1997).
7. (a) W.F. Goebel, T. Shedlovsky, G.I. Lavin, and M.H. Adams.
J. Biol. Chem. 148, 1 (1943); (b) E.C. Gotschlich and T.-Y.
Liu. J. Biol. Chem. 242, 463 (1967).
8. D.E. Brundish and J. Baddiley. Biochem. J. 110, 573 (1968).
9. I.R. Poxton, E. Tarelli, and J. Baddiley. Biochem. J. 175, 1033
(1978).
10. H.J. Jennings, C. Lugowski, and N.M. Young. Biochemistry,
19, 4712 (1980).
11. M. Kulakowska, J.R. Brisson, D.W. Griffith, N.M. Young, and
H.J. Jennings. Can. J. Chem. 71, 644 (1993).
12. W. Fischer, T. Behr, R. Hartmann, J. Peter-Katalinic, and H.
Egge. Eur. J. Biochem. 215, 851 (1993).
13. (a) J. Baddiley. Acc. Chem. Res. 3, 98 (1970); (b) H. Hussey
and J. Baddiley. In The enzymes of biological membranes.
Vol. 2. Edited by A. Martonosi. Plenum, New York. 1976.
pp. 227–326.
14. (a) L. Glaser. J. Biol. Chem. 239, 3178 (1964); (b) T. Chin,
M.M. Burger, and L. Glaser. Arch. Biochem. Biophys. 116,
358 (1966); (c) F. Fiedler and L. Glaser. J. Biol. Chem. 249,
35. R.R. Schmidt and W. Kinzy. Adv. Carbohydr. Chem. Biochem.
50, 21 (1994).
36. C.W. Haigh. Annu. Rep. NMR Spectrosc. 4, 311 (1971).
37. (a) S.J. Angyal, J.K. Saunders, C.T. Grainger, R. Le Fur, and
P.G. Williams. Carbohydr. Res. 150, 7 (1986); (b) S.J. Angyal
and R. Le Fur. Carbohydr. Res. 84, 201 (1980); (c) G.E. Hawkes
and D. Lewis. J. Chem. Soc. Perkin Trans. 2, 2073 (1984);
(d) G.A. Jeffrey and H.S. Kim. Carbohydr. Res. 14, 207 (1970);
(e) D. Lewis and S.J. Angyal. J. Chem. Soc. Perkin Trans. 2,
1763 (1989); (f) J.R. Synder. Carbohydr. Res. 198, 1 (1990).
38. (a) G.W. Schnarr, D.M. Vyas, and W.A. Szarek. J. Chem. Soc.
Perkin Trans. 1, 496 (1979); (b) J.K. Whitesell, T. LaCour,
R.L. Lovell, J. Pojman, P. Ryan, and A. Yamada-Nosaka. J.
Am. Chem. Soc. 110, 991 (1988).
39. G.W. Kenner, H.J. Rodda, and A.R. Todd. J. Chem. Soc. 1613
(1949).
40. D. Kegleviƒ and D. Ljevakovic. Carbohydr. Res. 64, 319 (1978).
© 1999 NRC Canada