Synthesis of bromoindolyl 4,7-di-O-methyl-Neu5Ac: Specificity toward influenza A and B viruses

Article abstract of DOI:10.1016/S0008-6215(99)00058-0

N-Acetylneuraminic acid (Neu5Ac) was converted into the methyl ester methyl ketoside-8,9-epoxy derivative (8). Methylation of 8 followed by deprotection gave 4,7-di-O-methyl-Neu5Ac (10). Compound 10 was converted into the corresponding methyl ester-chloroacetate derivative, which was subsequently coupled to 5-bromo-indol-3-ol to give the chromogenic product (13). Deprotection of 13 gave 5-bromo-indol-3-yl 4,7-di-O-methyl-Neu5Ac (5). The product 5 was specifically cleaved by sialidase from either influenza A or influenza B virus to give an indigo-blue precipitate, but was not cleaved by several bacterial or viral sialidases tested. The properties of product 5 relative to a fluorescent substrate for sialidase were also documented. Copyright (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0008-6215(99)00058-0

www.elsevier.nl/locate/carres
Carbohydrate Research 317 (1999) 198–203
Note
Synthesis of bromoindolyl 4,7-di-O-methyl-Neu5Ac:
specificity toward influenza A and B viruses
Avraham Liav *, Joyce A. Hansjergen, Komandoor E. Achyuthan,
Craig D. Shimasaki
ZymeTx Inc., 800 Research Parkway c100, Oklahoma City, OK 73104, USA
Received 7 December 1998; accepted 16 February 1999
Abstract
N-Acetylneuraminic acid (Neu5Ac) was converted into the methyl ester methyl ketoside-8,9-epoxy derivative (8).
Methylation of 8 followed by deprotection gave 4,7-di-O-methyl-Neu5Ac (10). Compound 10 was converted into the
corresponding methyl ester–chloroacetate derivative, which was subsequently coupled to 5-bromo-indol-3-ol to give
the chromogenic product (13). Deprotection of 13 gave 5-bromo-indol-3-yl 4,7-di-O-methyl-Neu5Ac (5). The product
5 was specifically cleaved by sialidase from either influenza A or influenza B virus to give an indigo-blue precipitate,
but was not cleaved by several bacterial or viral sialidases tested. The properties of product 5 relative to a fluorescent
substrate for sialidase were also documented. © 1999 Elsevier Science Ltd. All rights reserved.
Keywords: Chromogenic substrate; Sialidase; Influenza; di-O-Methyl-Neu5Ac; Bromoindolyl
It has also been shown [3] that a 4-O-acetyl
For a number of years we have been in-
volved in the development of influenza diag-
nostic agents [1]. Our approach is based on
the enzymatic cleavage of chromogenic ke-
tosides of N-acetylneuraminic acid (Neu5Ac)
by influenza virus sialidase, and as such, it
requires the utilization of a substrate that will
exhibit specificity towards the viral enzyme.
The specificity of various sialidase sub-
strates has been studied in the past, and it has
already been reported [2] that 4-O-methyl-
Neu5Ac bound to fetuin showed resistance
toward Vibrio cholerae sialidase, but was
cleaved rapidly by fowl plague virus sialidase.
substitution in sialic acids blocks the action of
bacterial sialidases for substrates containing
these derivatives, while viral enzymes show
significant activity. In another study it was
found that the sialidase of Streptococcus san-
guis, from the oral cavity, did not cleave 4-O-
acetylated sialic acid derivatives [4].
Accordingly, we proceeded to develop a
practical synthesis of 4-O-methyl-N-acetyl-
neuraminic acid (1) [5] from the acetal 2, and
when this was accomplished, the product was
coupled to 1-acetyl-5-bromoindo-3-ol (3c).
Deprotection of the coupled product gave the
substrate 5-bromo-indol-3-yl 4-O-methyl-N-
acetylneuraminic acid (4) which was purified
by HPLC (Liav, unpublished results). In the
process of purifying 4, a byproduct was iso-
* Corresponding author. Tel.: +1-405-271-1314; fax: +1-
405-271-1944.
E-mail address: liava@zymetx.com (A. Liav)
0008-6215/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(99)00058-0

A. Lia6 et al. / Carbohydrate Research 317 (1999) 198–203
199
lated. Examination of the structure of this
to the substrate 5-bromoindo-3-yl 4-O-
methyl-Neu5Ac (4).
1
product by mass spectrometry as well as H
13
and C NMR showed it to be 5-bromoindol-
Treatment of 2 with sodium hydride and
dimethyl sulfate under the conditions de-
scribed before [5] failed to give satisfactory
results. When slightly more than two molar
equivalents of sodium hydride and dimethyl
sulfate were used, the major product obtained
was the 4-methoxy derivative 6, and the sepa-
ration of the monomethoxy and dimethoxy
derivatives at this stage was tedious. With
larger amounts of the catalyst and dimethyl
sulfate, employing DMF as the solvent, hy-
drolysis of the methyl ester took place, and
the formation of the undesired N-methyl-N-
acetyl product could also be observed.
In an attempt to alleviate the apparent
steric hindrance associated with OH-7 in 2
(difficulties in the chemical synthesis of 7-sub-
stituted-Neu5Ac derivatives have already been
reported [8]), we sought to replace the acetal
group with a less restrictive system, such as
the 8,9-epoxy moiety (compounds 8 and 9).
Treatment of methyl (methyl 5-acetamido-
3-yl 4,7-di-O-methyl-N-acetylneuraminic acid
(5) (the presence of this product is most likely
the result of some dimethylation of the acetal
2).
Incubation of the two bromoindolyl gly-
cosides, 4 and 5, with sialidase from influenza
A or influenza B virus (see Section 1) resulted
in the cleavage of the aglycon, as indicated by
the development of the typical indigo-blue
color [7]. However, when sialidases from other
viruses, such as mumps and parainfluenza,
were employed, only the monomethoxy
derivative, 4, was cleaved, and no color was
observed in the case of the dimethoxy product
5. In addition, non-sialidase-producing viruses
which could potentially be isolated from res-
piratory specimens were also unreactive to-
ward 5. The specificity of 5 as a substrate for
influenza virus sialidase was also confirmed
when sialidases from several different bacterial
sources did not cleave 5, even when these
enzymes were tested at activity levels that were
5–50-fold higher than influenza virus sialidase
activity (Table 1).
The remarkable specificity of 5 toward infl-
uenza A and B viruses prompted us to devise
a synthetic scheme for this product. The most
logical approach appeared to be the dimethy-
lation of the readily available acetal 2, by
employing larger amounts of the catalyst and
methylating agent than those used in the syn-
thesis of 4-O-methyl-Neu5Ac, and following
the same sequence of reactions which had led
3,5-dideoxy-9-O-p-tolylsulfonyl-
D
-glycero-a- -
D
galacto-2-nonulopyranosid)onate (7) [9] with
M sodium methoxide solution in methanol
gave the 8,9-epoxide 8 in good yield. Methyla-
tion of 8 with dimethyl sulfate as described
before [5] proceeded smoothly, and the 4,7-
dimethoxy product 9 was isolated after
column chromatography in 70% yield. The
deprotection of 9 was carried out in two steps:
First, the ester was hydrolyzed by treatment
with M sodium hydroxide solution, and then
Table 1
Bromoindolyl-4,7-di-O-methyl-Neu5Ac is an influenza virus sialidase-specific substrate
Sialidase source
Enzyme activity 4-MU-
Fold
Reactivity to
bromoindolyl-4,7-di-O-
methyl-Neu5Ac chromogenic substrate
Neu5Ac (mol. min−1
)
Influenza A virus
Arthrobacter ureafaciens
Streptococcus mitis
Clostridium perfringens
Salmonella typhimurium
0.2090.005×10−9
10.5090.500×10−9
1.0690.180×10−9
10.0090.950×10−9
10.5590.350×10−9
1.00
52.50
5.30
50.00
52.75
positive
negative
negative
negative
negative

200
A. Lia6 et al. / Carbohydrate Research 317 (1999) 198–203
compared to 5 (Km =2.6390.53×10−4 M;
n=4) (Scheme 1).
the methyl ketoside was treated with Dowex
50 (H+) resin in 0.025 M hydrochloric acid.
The product, 4,7-di-O-methyl-Neu5Ac (10),
1
was characterized by H and 13C NMR spec-
1. Experimental
troscopy. Treatment of 10 with Dowex 50
(H+) in dry methanol gave the practically
pure methyl ester 11. Acetylation of 11 with
acetic anhydride and pyridine gave a mixture
of the anomers, which was purified by column
chromatography. Conversion of the acetates
mixture into the corresponding chloride, fol-
lowed by treatment with 1-N-acetyl-5-bro-
moindo-3-ol (3c; prepared from the 3-acetoxy
derivative 3a by acetylation to give the diac-
etate 3b and subsequent treatment with 90%
sulfuric acid [10]) gave a mixture of the bro-
moindolyl di-O-methyl-Neu5Ac product 13,
and a byproduct, most likely the 2,3-dehydro
derivative 14. Column chromatography re-
moved the unchanged chromogen 3c and mi-
nor impurities, but the mixture of the two
products was inseparable. Partial deprotection
of the mixture gave the esters 15 and 16,
which still could not be separated by chro-
matography. Finally, deesterification of 15
and 16 gave a mixture which could be easily
separated by HPLC, yielding the desired bro-
moindolyl di-O-methyl-Neu5Ac derivative 5.
In order to further evaluate 5 as a substrate
for sialidase, the catalytic properties of influ-
enza virus sialidase toward 5 were compared
to that of the well-known fluorescent substrate
4-methylumbelliferyl Neu5Ac (4-MU Neu5-
Ac). The optimal pH for influenza virus siali-
dase toward 5 was 5.4. Nearly 90% of maxi-
mal activity was also observed between pH
values of 5.2 and 4.8. However, enzymatic
activity rapidly declined above pH 5.4 to
nearly one-half maximal activity at a pH value
of 6.2. In contrast, near maximal (95–100%)
influenza virus sialidase activity was observed
between pH values of 4.8 and 6.2 when 4-MU
Neu5Ac was employed as a substrate. Differ-
ences exist in the pH optima of influenza virus
sialidase toward different substrates [6]. The
activity of influenza virus sialidase upon 4-
MU Neu5Ac was higher (13.2×10−6 mol
General methods.—Optical rotations were
measured with a Jasco P-1020 polarimeter at
24 °C. NMR spectra were measured with a
Jeol FX-90QM multinuclear spectrometer un-
less otherwise stated. Column chromatogra-
phy was performed with silica gel 70–230
mesh (Aldrich). Thin layer chromatography
was performed on Analtech Uniplate silica gel
plates. HPLC purification was performed on a
Waters system and a C18 reverse-phase column
(2.2×25 cm) using 1–20% acetonitrile gradi-
ent. Microanalyses were performed by Gal-
braith Laboratories Inc., Knoxville, TN.
Detection of influenza A or influenza B
6irus.—A throat swab specimen collected
from a patient with influenza-like symptoms is
placed in a tube containing 1.3 mL of a
buffered solution (pH 5.4), and thoroughly
squeezed out into the solution to extract the
specimen. The swab is then discarded. A filter
tip is then placed on the tube and the speci-
men contents are filtered into a vial containing
1 mg of the bromoindolyl substrate 5. The vial
is then incubated for 30 min at 41 °C, after
which time an alkaline stop solution (0.5 mL)
is added. The mixture is then transferred into
a funnel filtration device [1] and the blue
precipitate (indicative of the presence of either
influenza A or influenza B virus) is collected
on a membrane at the bottom of the device.
Enzymatic assays.—Purified or partially
purified sialidases from Arthrobacter ureafa-
ciens, Clostridium perfringens, and Salmonella
typhimurium were purchased from Sigma. Bac-
teria cultured from human volunteer throat
swab extracts and identified as Streptococcus
mitis were lysed and then used as a sialidase
source. Sialidases were assayed at pH 5.4,
except during experiments where the effect of
pH on activity was being determined. Enzy-
matic activity was quantitated using 4-MU
Neu5Ac that was synthesized ‘in house’ (Liav,
unpublished data). The fluorescence of cleaved
4-methylumbelliferone was measured using ex-
citation and emission wavelengths u of 365 and
450 nm, respectively, with an Hitachi F-3010
min−1
L
−1) relative to 5 (2.3×10−6 mol
min−1
L
−1). Likewise, the affinity of 4-MU
Neu5Ac for influenza virus sialidase was also
higher (Km =4.6790.97×10−5 M; n=3)

A. Lia6 et al. / Carbohydrate Research 317 (1999) 198–203
201
Scheme 1.
Fluorescence Spectrophotometer. The activity
of various sialidases toward 5 was quantitated
by measuring the released N-acetylneuraminic
acid [11]. The kinetic constant Km was calcu-
lated using EnzymeKinetics (Trinity Software,
v. 1.2).
In-house evaluation of the substrates for
sialidase specificity was conducted with char-
acterized virus strains obtained from the
American Type Culture Collection. Influenza
A and B virus strains, as well as other siali-
dase-producing viruses such as parainfluenzas
and mumps were tested with the substrate. In
addition, non-sialidase-producing viruses
which could potentially be isolated from res-
piratory specimens were also tested. All test-
ing was done from virus cultures grown to
maximal titers to obtain the most stringent
evaluation of the substrates. Bacterial siali-
dases were tested with 5 at activity levels of
5–50-fold higher than influenza virus sialidase
to demonstrate specificity.
Methyl (methyl 5-acetamido-8,9-anhydro-
3,5-dideoxy-
pyranosid)onate (8).—Methyl (methyl 5-ac-
etamido-3,5-dideoxy-9-O-p-tolylsulfonyl-
-galacto-2-nonulopyranosid)onate
D
-glycero-h-
D
-galacto-2-nonulo-
D
-
glycero-a-
D
(7, 1.66 g) [9] was treated with M NaOMe
solution (3.5 mL) in MeOH (20 mL) for 45
min at room temperature (rt). It was neutral-

202
A. Lia6 et al. / Carbohydrate Research 317 (1999) 198–203
ized with Dowex 50 (H+) resin, and the resin
was filtered off and washed with methanol.
The filtrate was evaporated, and the residue
was dried and extracted with acetone. The
insoluble solids were filtered off and washed
with acetone, and the filtrate was evaporated.
The residue was dried and chromatographed
on silica gel. Elution with 9:1 CH2Cl2–MeOH
removed minor impurities. Continued elution
with the same solvent system yielded the
amorphous product 8 (1.02 g, 95%); [h]D −
62° (c 1.0, CH2Cl2). Anal. Calcd for
C13H21NO8: C, 48.89; H, 6.63; N, 4.38. Found:
C, 48.94; H, 6.92; N, 4.02.
rated and the residue was subsequently treated
with Dowex 50 (H+, 2.5 g) and 0.025 M HCl
solution (35 mL) for 3 h at 95–100 °C. The
resin was filtered off and washed with MeOH,
and the filtrate was evaporated to give the
practically pure 4,7-di-O-methyl-Neu5Ac
1
product (10), isolated as a solid. H NMR
(D2O, 300 MHz): l 3.96 (bd, 1 H, J6,7 1.0, J6,5
10.4 Hz, H-6), 3.83 (t, 1 H, J 10.4 Hz, H-5),
3.70–3.63 (m, 2 H, H-9 and H-9%), 3.53 (m,
partially overlapped, 1 H, H-4), 3.44 (bdd, 1
H, J8,9 6.3, J8,9% 12.5 Hz, H-8), 3.27 (s, 3 H,
OMe), 3.25 (m, 1 H, partially overlapped,
H-7), 3.22 (s, 3 H, OMe), 2.30 (dd, 1 H, J3eq,3ax
13.2 Hz, J3eq,4 4.8 Hz, H-3eq), 1.90 (s, 3 H,
NAc), 1.62 (dd, 1 H, J3ax,4 13.2 Hz, H-3ax).
Methyl (5-acetamido-3,5-dideoxy-4,7-di-O-
Methyl (methyl 5-acetamido-8,9-anhydro-
3,5-dideoxy-4,7-di-O-methyl-
D
-glycero-h- -
D
galacto-2-nonulopyranosid)onate (9).—A cold
(ice-bath) solution of 8 (1.36 g) in MeCN (15
mL) was saturated with nitrogen. Sodium hy-
dride (80% dispersion in oil, 280 mg) was
added, and the mixture was stirred under ni-
trogen for 15 min. Dimethyl sulfate (2.5 mL)
was added and stirring in an ice bath, under
nitrogen, was continued for an additional 30
min. The mixture was filtered through Celite,
and the precipitate washed with dry acetoni-
trile. The filtrate was evaporated and the
residue was dried and extracted with acetone.
The insoluble solid material was filtered off
and washed with acetone and the filtrate was
evaporated. The residue was dried and chro-
matographed on silica gel. Elution with 15:1
CH2Cl2 –MeOH removed the oily material.
Continued elution with the same solvent sys-
tem gave the amorphous dimethoxy derivative
9 (1.03g, 69%); [h]D −27.9° (c 1.0, CH2Cl2);
1H NMR: l 3.80 (s, 3 H, CO2Me), 3.46 (s, 3
H, OMe-4), 3.34 (S, 3 H, OMe-7), 3.28 (s, 3
H, OMe-2), 2.56 (dd, 1 H, J3eq,4 5.3 Hz, J3eq,3ax
13.2 Hz, H-3eq), 2.01 (s, 3 H, NAc), 1.77 (dd,
1 H, H-3ax). Anal. Calcd for C15H25NO8: C,
51.86; H, 7.25; N, 4.03. Found: C, 51.50; H,
7.20; N, 4.02.
methyl-
D
-glycero-h,i- -galacto-nonulopyran-
D
osid)onate (11).—The crude 10 (obtained
from 2.66 g of 9) was treated with dry Dowex
50 (H+, 1.6 g) in MeOH (35 mL). The mix-
ture was boiled under reflux for 90 min, and
the resin was filtered off and washed with
MeOH. The filtrate was evaporated and the
residue was dried to give the crude amorphous
ester 11.
Methyl (5-acetamido-2,8,9-tri-O-acetyl-3,5-
dideoxy-4,7-di-O-methyl-
D
-glycero-h,i- -gal-
D
acto-nonulopyranosid)onate (12).—The methyl
ester 11 was treated with acetic Ac2O (8 mL)
and pyridine (16 mL) in the catalytic presence
of 4-dimethylaminopyridine at rt overnight.
The mixture was evaporated and the residue
was dried and chromatographed on silica gel.
Elution with 25:1 CH2Cl2 –MeOH gave the
product 12 as a syrupy anomeric mixture (1.86
g, 51%, based on the anhydro derivative 9);
1
[h]D −8.1° (c 1.0. CH2Cl2); H NMR: l 4.65
(dd, 1 H, J9,9% 12.5, J9,8 2.1 Hz, H-9), 4.24 (dd,
1 H, J9%,8 6.5 Hz, H-9%), 4.11 (dd, 1 H, H-8),
3.98–3.68 (m, partially overlapped with
CO2Me), 3.51 (s, 3 H, OMe), 3.35 (s, 3 H,
OMe), 2.65 (dd, 1 H, J3eq,3ax 13.4, J3eq,4 4.8 Hz,
H-3eq), 2.13, 2.06, 2.04 (s, 9 H, OAc), 2.03 (s,
3 H, NAc), 1.75 (dd, 1 H, J3ax,4 11.0 Hz,
H-3ax) Anal. Calcd For C20H31NO12: C,
50.30; H, 6.54; N, 2.93. Found: 49.89; H, 6.54;
N, 2.96.
5-Acetamido-3,5-dideoxy-4,7-di-O-methyl-
D
-glycero- -galacto-nonulopyranosidonic acid
D
(4,7-di-O-methyl-Neu5Ac, 10).—Compound 9
(2.66 g) was treated with M NaOH solution
(10 mL) in 50% aq MeOH (30 mL) for 30 min
at rt. The mixture was neutralized with Dowex
50 (H+) and the resin was filtered off and
washed with methanol. The filtrate was evapo-
Methyl (5-bromoindol-3-yl 5-acetamido-8,9-
di-O-acetyl-4,7-di-O-methyl-3,5-dideoxy-
glycero - h -
D
-
D
- galacto - nonulopyranosid)onate

A. Lia6 et al. / Carbohydrate Research 317 (1999) 198–203
203
(13).—A solution of 12 (380 mg) in CH2Cl2 (5
mL) and AcCl (0.5 mL) was saturated with
dry HCl. The mixture was then stirred for 5h
at rt, and evaporated. The residue was dried
under vacuum. A solution of the crude chlo-
ride in acetone (5 mL) was saturated with
nitrogen at rt. 1-Acetyl-5-bromo-3-indol-ol
(3c, 208 mg) was added. Stirring under nitro-
gen was continued, and 5 min later M NaOH
solution was added. After an additional 5 min,
more NaOH solution (0.2 mL) was added,
and at this point the reaction mixture turned
green. The mixture was stirred for 30 min
more, and evaporated. The residue was dried
and chromatographed on silica gel. Elution
with 25:1 CH2Cl2–MeOH gave the unchanged
indole derivative 3c. Continued elution with
the same solvent system gave a syrupy mixture
of the coupled product 13 and 2,3-dehydro
derivative 14, as indicated by NMR. Attempts
to separate the two products by using differ-
ent conditions were unsuccessful.
(s, 3 H, OMe), 3.49 (s, 3 H, OMe), 2.15 (s, 3
H, NAc). The desired product 5 (61 mg,
14.5%, based on the dimethoxy ester 11) had a
retention time of 60 min, and was isolated as
1
a solid, [h]D−41° (c 1.0, H2O); H NMR data
(in D2O, 300 MHz): l 7.85 (d) and 7.07–7.02
(m) (4 H, indolyl), 4.03 (t, J5,4 9.5, J5,6 9.5 Hz,
H-5), 3,90 (dd, J6,7 1.9 Hz, H-6), 3.85 (J8,7 8.3,
J8,9 2.3, J8,9% 4.6 Hz, H-8), 3.78 (dd, J9,9% 12.4
Hz, H-9), 3.63 (dd, H-9%), 3.43 (s, 3 H, OMe),
3.38 (dd, H-7), 3.34 (s, 3 H, OMe), 3.29 (m,
J4,3eq 4.7, J4,3ax 11.8 Hz, H-4), 3.03 (dd, J3eq,3ax
11.8 Hz, H-3eq), 1.95 (3, 3 H, NHAc), 1.68 (t,
H-3 ax); 13C NMR (in D2O with 1,4-dioxane
as the internal standard): l 175.0 (CO2H);
173.4 (CO–NAc); 132.9, 132.1, 125.4, 124.2,
121.7, 118.3; 114.4; 112.7 (indolyl); 104.9 (C-
2); 79.5 (C-4); 78.8 (C-8); 74.2 (C-6); 72.1
(C-7); 63.1 (C-9); 61.1 (OMe-7); 57.6 (OMe-4);
50.8 (C-5); 37.7 (C-3); 23.1 (Me–NAc).
5-Bromoindol-3-yl 5-acetamido-3,5-dideoxy-
References
4,7-di-O-methyl-
D
-glycero- -galacto-nonulo-
D
pyranosidonic acid (5).—The foregoing mix-
ture of 13 and 14 was treated with M NaOMe
solution (0.2 mL) in MeOH (5 mL) for 40 min
at rt. It was neutralized with Dowex 50 (H+)
and the resin was filtered off and washed with
MeOH. The filtrate was evaporated and the
residue was dried and chromatographed on
silica gel. Elution with 5:1 CH2Cl2–MeOH
removed residual yellow color. Continued elu-
tion with the same solvent system yielded a
mixture of the methyl esters 15 and 16 which
was still inseparable. Finally, this mixture was
treated with M NaOH solution (0.5 mL) in
50% aq MeOH (5 mL) for 1 h at rt. The
mixture was evaporated and the residue was
purified by reverse-phase HPLC. The 2,3-de-
hydro derivative 17 was eluted first (retention
[1] A. Liav, J.F. Maher, C.D. Shimasaki, C. Worth
Clinkscales, M.D. Roark, United States Patent No.
5,252,458 (Oct. 12, 1993).
[2] J.M. Beau, R. Schauer, Eur. J. Biochem., 106 (1980)
531–540.
[3] A.P. Corfield, M. Sander-Wewer, R.W. Veh, M. Wem-
ber, R. Schauer, Biol. Chem. Hoppe-Seyler, 367 (1986)
433–439.
[4] A. Varki, S. Diaz, J. Biol. Chem., 258 (1983) 12465–
12471.
[5] A. Liav, Carbohydr. Res., 271 (1995) 241–245.
[6] R. Drzeniek, Curr. Top. Microbiol. Immunol., 59 (1972)
35–74.
[7] V. Eschenfelder, R. Brossmer, Glyconjugate J., 4 (1987)
171–178.
[8] D.C.M. Kong, M. von Itzstein, Tetrahedron Lett., 36
(1995) 957–960.
[9] R. Isecke, R. Brossmer, Tetrahedron, 50 (1994) 7445–
7460.
[10] J.P. Horwitz, J. Shua, R.J. Curby, A.J. Tomson, M.A.
Da Rouge, B.E. Fisher, J. Mauricio, I. Klundt, J. Med.
Chem., 7 (1964) 574–575.
[11] J.I. Murayama, M. Tomita, A. Tsuji, A. Hamada, Anal.
Biochem., 73 (1976) 535–538.
1
time 10 min); [h]D +22° (c 1.0, H2O); H
NMR (D2O): l 6.04 (d, J 1.8 Hz, H-3), 3.51
.