Synthesis and inhibitory activity of a di- and a trisaccharide corresponding to an erythrocyte glycolipid responsible for the nor polyagglutination

Article abstract of DOI:10.1016/S0008-6215(02)00222-7

The polyagglutinable erythrocytes NOR contain unusual neutral glycolipids reactive with anti-NOR antibodies. The disaccharide α-D-Galp-(1→4)-D-GalpNAc and the trisaccharide α-D-Galp-(1→4)-β-D-GalpNAc-(1→3)-D-Gal corresponding to the non-reducing end of a NOR glycolipid (NOR1) were chemically synthesized. The syntheses were based on a common (1→4)-β-D-GalNAc precursor, and utilized benzyl glycoside and benzyl ether functions for persistent blocking of hydroxyls. The α-D-Galp-(1→4)-β-D-GalpNAc structural element has been found only recently in Nature, and derivatives thereof have not been synthesized before. Both the synthesized oligosaccharides inhibited specifically human anti-NOR antibodies, the trisaccharide being 300 times more active than the disaccharide.

Full text of DOI:10.1016/S0008-6215(02)00222-7

Carbohydrate Research 337 (2002) 1517–1522
www.elsevier.com/locate/carres
Synthesis and inhibitory activity of a di- and a trisaccharide
corresponding to an erythrocyte glycolipid responsible for the nor
polyagglutination
Ulrika Westerlind,a Per Hagback,a Maria Duk,b Thomas Norberga,*
aDepartment of Chemistry, Swedish Uni6ersity of Agricultural Sciences, PO Box 7015, S-750 07 Uppsala, Sweden
bDepartment of Immunochemistry, Ludwik Hirszfeld Institute of Immunology and Experimental Therapy, Polish Academy of Sciences,
PL-53114 Wroclaw, Poland
Received 2 May 2002; accepted 22 July 2002
Abstract
The polyagglutinable erythrocytes NOR contain unusual neutral glycolipids reactive with anti-NOR antibodies. The disaccha-
ride a-
non-reducing end of a NOR glycolipid (NOR1) were chemically synthesized. The syntheses were based on a common
(1“4)-b- -GalNAc precursor, and utilized benzyl glycoside and benzyl ether functions for persistent blocking of hydroxyls. The
a- -Galp-(1“4)-b- -GalpNAc structural element has been found only recently in Nature, and derivatives thereof have not been
D
-Galp-(1“4)-
D
-GalpNAc and the trisaccharide a-
D
-Galp-(1“4)-b-D-GalpNAc-(1“3)-D-Gal corresponding to the
D
D
D
synthesized before. Both the synthesized oligosaccharides inhibited specifically human anti-NOR antibodies, the trisaccharide
being 300 times more active than the disaccharide. © 2002 Elsevier Science Ltd. All rights reserved.
Keywords: NOR polyagglutination; Erythrocytes; Oligosaccharide synthesis
1. Introduction
than oligosaccharide derivatives with a spacer attached.
Should spacer derivatives be needed in the future (e.g.
for affinity purification of anti-NOR antibodies), they
can be prepared from the free oligosaccharides by
several methods. The preparative details of the free
oligosaccharide synthesis and the inhibitory activity of
the oligosaccharides are hereby reported.
The rare NOR polyagglutination was recently shown
to be associated with the occurrence on the erythrocytes
of unusual neutral glycolipids.1 The structure of one of
them (NOR1) has been recently established2 as
Gal(a1“4)GalNAc(b1“3)Gal(a1“4)Gal(b1“4)Glc-
Ceramide.
The non-reducing terminal end of this structure con-
tains the unusual structural element Gal(a1“
4)GalNAcb which has hitherto only been demonstrated
in NOR glycolipids and in amphibium oviductal
mucins.3 To our knowledge, compounds with this struc-
tural element have not been synthesized before. In
order to gain access to material for further immunolog-
ical work, a chemical synthesis of the non-reducing end
di- and trisaccharide structures (9 and 15) correspond-
ing to the above glycolipid was carried out. We chose
to synthesize the free reducing oligosaccharides, rather
2. Results and discussion
Synthesis of the oligosaccharides.—The syntheses
were carried out using compounds 3, 6 and 10 as
monosaccharide building blocks. The terminal Gal and
“3Gal building blocks (6 and 10, respectively) were
prepared by slightly modified literature synthesis proce-
dures,4,5 whereas the “4GalNAc building block (3)
was prepared from
a corresponding glucosamine
derivative using the following modified literature proce-
dure:6 treatment of 17 with triflic anhydride gave the
4-triflate 2, which was treated in situ with cesium
acetate in N,N-dimethylformamide to produce the
* Corresponding author. Tel.: +46-18-671578; fax: +46-
18-673477
E-mail address: thomas.norberg@kemi.slu.se (T. Norberg).
0008-6215/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(02)00222-7

1518
U. Westerlind et al. / Carbohydrate Research 337 (2002) 1517–1522
Table 1
1H (upper row) and 13C NMR data for 9 (coupling constants in parentheses). Comments and explanations, see experimental part,
prepn. of 9
1
2
3
4
5
6a
6b
Gal(a)
4.96 (3.9)
101.72
4.94 (3.9)
101.72
5.22 (3.3)
92.30
4.66 (8.2)
96.70
3.81
69.77
3.81
69.77
4.14
51.75
3.87
3.92
70.13
3.92
70.13
3.94
68.28
3.72
3.99
70.20
3.99
70.20
4.06
79.85
3.99
4.25
72.18
4.25
72.18
4.10
71.97
3.68
3.65
61.76
3.65
61.76
3.65
61.76
3.79
3.65
3.65
3.65
3.79
Gal(b)
GalNAc(a)
GalNAc(b)
55.95
71.97
78.29
76.09
61.54
galactosamine derivative 3 (72% yield from 1). Com-
pound 3 carries a protecting group pattern suitable for
4-substitution, and was subsequently used in b-glycosy-
lation of either benzyl alcohol or the monosaccharide
10 to give monosaccharide 4 (95% yield) or the disac-
charide 11, respectively. The latter compound was ob-
tained only in impure form after chromatography.
Final purification was, however, achieved in the next
step. Compounds 4 and 11 were both deacylated in the
4-position to give compounds 5 (95%) and 12 (37%,
from 3), respectively. The latter relatively low yield of
12 was partly caused by the difficulties in chromato-
graphic purification of 11. However, the glycosylation
yield in the reaction between 3 and 10 was consistently
less than 50% (TLC estimation) during several different
runs, and the reason for this is currently not clear.
Compounds 5 and 12 were both a-glycosylated with
phenyl 2,3,4,6-tetra-O-benzyl-1-thio-b-D-galactopyran-
oside (6)4 to give the disaccharide 7 and trisaccharide 13
in 66 and 70% yield, respectively. Phthalimido depro-
tection (hydrazine hydrate) and N-acetylation (acetic
anhydride) gave 8 and 14 (93 and 97% yield), catalytic
hydrogenation of which produced the free di- and
trisaccharides 9 and 15 in 68 and 61% yields, respec-
tively. The structures of both oligosaccharides was
confirmed by complete assignment of the proton and
carbon NMR spectra (Tables 1 and 2, respectively), as
well as by high-resolution mass spectroscopy.

U. Westerlind et al. / Carbohydrate Research 337 (2002) 1517–1522
1519
Table 2
1H (upper row) and 13C NMR data for 15 (coupling constants in parentheses). Comments and explanations, see experimental part,
prepn. of 15
1
2
3
4
5
6a
6b
Gal(a+b)
GalNAc(a)
GalNAc(b)
Gal*(a)
4.96(3.7)
101.34
4.70 (8.2)
103.92
4.66 (8.4)
103.92
5.19 (3.7)
93.20
4.53 (7.8)
97.34
3.82
69.46
3.95
53.71
3.90
53.71
3.84
68.33
3.52
71.92
3.94
69.89
3.78
71.34
3.79
71.34
3.89
79.83
3.68
82.96
4.01
69.94
4.02
78.10
4.02
78.10
4.13
69.94
4.08
69.32
4.29
71.80
3.69
75.75
3.69
75.75
3.64
70.91
3.64
75.46
3.63
61.84
3.83
61.49
3.83
61.49
3.70
61.67
3.70
61.17
3.68
3.86
3.86
3.70
3.70
Gal*(b)
Inhibitory acti6ity of the oligosaccharides.—The abil-
ity of the oligosaccharides 9 and 15 to react with
anti-NOR antibodies isolated from human sera was
tested by hemagglutination inhibition, using papain-
treated NOR erythrocytes (Table 3). The antibodies
were weakly inhibited by galactose, and 8- and 2400-
times more strongly by the di- and trisaccharide, respec-
tively. They were not inhibited by the reference
disaccharide a-
cificity of these reactions, the inhibition of anti-a-
Galp-(1“3)- -Gal human antibodies was compared.
The latter antibodies were also inhibited weakly by
galactose, slightly more strongly by the a- -Galp-(1“
3)- -Gal disaccharide, but were not inhibited by
D
-Galp-(1“3)-D-Gal. To assess the spe-
D
-
D
D
D
oligosaccharides 9 and 15 (Table 3). In conclusion,
these results show that the disaccharide 9 and trisaccha-
ride 15 are specifically recognized by anti-NOR anti-
bodies and that the galactose reducing end of trisaccha-
ride 15 is important for an effective binding to the
antibodies.
3. Experimental
General methods.—Concentrations were performed
at reduced pressure (bath temperature B40 °C). NMR
spectra were recorded for solutions in CDCl3 (internal
Table 3
Inhibition of human anti-NOR and anti-Gal(a1-3)Gal antibodies by galactose and oligosaccharides
Inhibitor
Anti-NORa (mM)b
Anti-a-D-Galp-(1“3)-D
-Gala (mM)b
Galactose
25
3
0.01
\100
25
\50
\50
6
a-D
a-D
a-D
-Galp-(1“4)-
-Galp-(1“4)-b-
-Galp-(1“3)- -Gal
D
-GalNAcp (9)
D
-GalNAcp-(1“3)-
D
-Gal (15)
D
a Anti-NOR were tested with papain-treated NOR erythrocytes, and anti-Gal(a1-3)Gal with untreated rabbit erythrocytes.
b Minimal concentration of inhibitor required for inhibition of four hemagglutinating units of each antibody

1520
U. Westerlind et al. / Carbohydrate Research 337 (2002) 1517–1522
Me4Si, l=0.00) or D2O (internal acetone, 1H l=
2.225, 13C l=30.7) at 303 K unless otherwise stated
with a Bruker DRX 400 spectrometer. Only selected
NMR data are reported. Assignments were corrobo-
rated by appropriate 2D experiments. Coupling con-
stants (in Hz) are given within brackets. In
oligosaccharide derivatives, sugar protons are deno-
tated as H-1, H-1% or H-1%%, in order of increasing
distance from the reducing end. The FAB-MS spectra
were recorded with a JEOL JMS-SX/SX-102A instru-
ment. Ions were produced by a beam of Xe-atoms (6
keV), using a matrix of glycerol or m-nitrobenzyl alco-
hol. For HRMS, PEG was used as an internal stan-
dard. TLC was performed on Silica Gel F254 (E. Merck,
Darmstadt, Germany) with detection by UV-light and
by staining with 5% H2SO4 in EtOH. Column chro-
washed with aq 2 M H2SO4, aq 1 M NaHCO3, dried
and concentrated. Column chromatography (5:2
petroleum ether–EtOAc) of the residue gave 4 (421 mg,
0.677 mmol, 97%) as a colorless syrup. NMR data
1
(CDCl3): H, l 5.58 (dd, 1 H, J3,4 3.3, J4,5 0.5, H-4),
5.05 (d, 1 H, J1,2 8.6, H-1), 4.73 (d, 1 H, J 12.3,
PhCH2), 4.56–4.39 (m, 4 H, PhCH2), 4.36 (dd, 1 H, J2,3
11.0, H-2), 4.21 (dd, 1 H, H-3), 4.15 (d, 1 H, J 12.5,
PhCH2), 3.84 (m, 1 H, H-5), 3.59 (dd, 1 H, J6a,6b 9.5,
J6a,5 5.9, H-6a), 3.53 (dd, 1 H, J6b,5 7.0, H-6b). [HRMS:
Calc. for C37H35NO8Na: 644.2260 (M +Na+). Found:
644.2326 (M +Na+)].
Benzyl 3,6-di-O-benzyl-2-deoxy-2-phthalimido-i-D-
galactopyranoside (5).—To a solution of 4 (350 mg,
0.563 mmol) in dry MeOH (43 mL) was added 1 M
NaOMe (2.15 mL). The mixture was stirred at room
temperature for 8 h, then neutralized with Dowex-50
(H+), and filtered. A drop of conc. aq ammonia was
added and the solution was then concentrated and
co-evaporated several times with EtOAc and once with
toluene to give 5 (309 mg, 0.53 mmol, 95%). NMR data
,
matography was performed on Matrex Silica Gel 60 A
,
(35–70 mm, Amicon). Molecular sieves (powdered 4 A)
were dried at 280 °C/0.5 torr overnight.
Dichloromethane was distilled from P2O5 when neces-
sary. Dimethyl(thiomethyl)sulfonium triflate (DMTST)
was prepared8 by mixing equimolar amounts of
dimethyl disulfide and methyl triflate in dry CH2Cl2 and
then filtering off the precipitated (fridge) crystals. The
material was stored under dry nitrogen at −20 °C. The
1
(CDCl3): H, l 5.02 (d, 1 H, J1,2 8.6, H-1), 4.72 (d, 1 H,
PhCH2), 4.54 (s, 2 H, PhCH2), 4.51 (d, 1 H, PhCH2),
4.43 (dd, 1 H, J2,3 11.0, H-2), 4.40 (d, 1 H, PhCH2),
4.20 (d, 1 H, PhCH2), 4.16 (dd, 1 H, J3,4 3.3, H-3), 4.06
(dd, 1 H, J3,4 3.3, J4,5 0.5, H-4). [HRMS: Calc. for
C35H33NO7Na: 602.2155 (M +Na+). Found: 602.2134
(M+Na+)].
disaccharide a-
from Glycorex (Lund, Sweden).
Ethyl 4-O-acetyl-3,6-di-O-benzyl-2-deoxy-2-phthal-
imido-1-thio-i- -galactopyranoside (3).—Triflic anhy-
D
-Galp-(1“3)-D-Gal was purchased
D
Benzyl O-(2,3,4,6-tetra-O-benzyl-h-
osyl)-(1“4)-3,6-di-O-benzyl-2-deoxy-2-phthalimido-
i- -galactopyranoside (7).—A mixture of 5 (287 mg,
0.495 mmol), phenyl 2,3,4,6-tetra-O-benzyl-1-thio-b-
galactopyranoside 64 (477 mg, 0.753 mmol) and pow-
D-galactopyran-
dride (1.2 mL, 7.0 mmol) was added dropwise during 5
min at 0 °C to a solution of 17 (1.52 g, 2.85 mmol) in
dry 2:1 CH2Cl2–pyridine (10 mL). The mixture was
stirred for 1 h at 0 °C and 1.5 h at room temperature
after which TLC (2:1 petroleum ether–EtOAc) indi-
cated complete conversion. The mixture was diluted
with CH2Cl2, washed with aq 1 M H2SO4, aq 1 M
NaHCO3, water, dried and concentrated. The residue,
containing the 4-triflate 2, was dissolved in dry DMF (7
mL) and added to a mixture of CsOAc (2.73 g, 14.2
mmol) and dry DMF (17 mL). The reaction mixture
was stirred at room temperature overnight and was
then partitioned between toluene and water. The or-
ganic layer was washed with water, dried and concen-
trated. Column chromatography (5:2 petroleum
ether–EtOAc) of the residue gave 3 (1.18 g, 2.05 mmol,
72%) as a syrup. The NMR data were as published.6
Benzyl 4-O-acetyl-3,6-di-O-benzyl-2-deoxy-2-phthal-
D
D
-
,
dered
4
A
molecular sieves (1.71 g) in dry
CH2Cl2–diethyl ether–toluene (3.8–2.7–2.7 mL) was
stirred at room temperature for 5 min. Solid DMTST
(404 mg, 1.56 mmol) was added in portions over 2 h
and then pyridine (360 mL) was added to quench the
reaction. The mixture was diluted with diethyl ether
and the solids were filtered off. The filtrate was then
washed with aq 2 M H2SO4, aq 1 M NaHCO3, dried
and concentrated. Column chromatography (5:2
petroleum ether–EtOAc) of the residue gave 7 (360 mg,
0.327 mmol, 66%) as a colorless syrup. NMR data
1
(CDCl3): H, l 5.17 (d, 1 H, J1,2 8.6, H-1), 5.08 (d, 1 H,
J1,2 3.1, H-1%), 4.67 (1 H, H-2), 4.28 (1 H, H-3%), 4.16 (1
H, H-2%). [HRMS: Calc. for C69H67NO12Na: 1124.4561
(M+Na+). Found: 1124.4530 (M+Na+)].
imido-i-
D
-galactopyranoside (4).—A mixture of 3 (400
mg, 0.695 mmol), benzyl alcohol (114 mg, 1.06 mmol)
Benzyl O-(2,3,4,6-tetra-O-benzyl-h-
osyl)-(1“4)-2-acetamido-3,6-di-O-benzyl-2-deoxy-i-
-galactopyranoside (8).—Hydrazine hydrate (200 mL,
5.16 mmol) was added to a mixture of 7 (354 mg, 0.321
mmol) in 2:3 toluene–EtOH (15 mL). The mixture was
refluxed for 36 h, and then cooled and evaporated. The
residue was partitioned between 1:1 CH2Cl2–EtOH and
water. The organic layer was concentrated and then
D-galactopyran-
,
and powdered 4 A molecular sieves (2.40 g) in dry
CH2Cl2–diethyl ether–toluene (5.3–3.8–3.8 mL) was
stirred at room temperature for 5 min. Solid DMTST
(567 mg, 2.20 mmol) was added in portions over 2 h
and then pyridine (0.5 mL) was added to quench the
reaction. The mixture was diluted with diethyl ether
and the solids were filtered off. The filtrate was then
D

U. Westerlind et al. / Carbohydrate Research 337 (2002) 1517–1522
1521
dissolved in 1:1 CH2Cl2–MeOH (15 mL) and treated
with acetic anhydride (0.75 mL) at room temperature.
After 2 h, the mixture was concentrated. Column chro-
matography (2:1 petroleum ether–EtOAc) gave 8 (302
mmol, 37% calculated from 10). NMR data (CDCl3):
1H, l 5.37 (d, 1 H, J1,2 8.5, H-1%), 5.38 (1 H, PhCH2),
4.47 (1 H, H-2%), 4.16 (1 H, H-6a), 4.15 (1 H, H-H-3%),
3.78 (1 H, H-6a), 3.17 (1 H, H-5). [HRMS: Calc. for
C55H53NO12Na: 942.3465 (M+Na+). Found: 942.3383
(M+Na+)].
1
mg, 0.298 mmol, 93%). NMR data (CDCl3): H, l 5.20
(d, 1 H, JNH,2 7.6, NH), 4.99 (d, 1 H, J1,2 2.5, H-1%),
4.83 (d, 1 H, H-1), 4.03 (1 H, H-2%), 4.01 (1 H, H-4),
3.91 (1 H, H-3), 3.54 (1 H, H-2), 1.98 (CH3CONH).
[HRMS: Calc. for C63H67NO11Na: 1036.4612 (M+
Na+). Found: 1036.4584 (M+Na+)].
Benzyl O-(2,3,4,6-tetra-O-benzyl-h-
osyl)-(1“4)-O-(3,6-di-O-benzyl-2-deoxy-2-phthal-
imido-i- -galactopyranosyl)-(1“3)-2-O-benzyl-4,6-O-
benzylidene-i- -galactopyranoside (13).—A mixture of
12 (185 mg, 0.201 mmol), 64 (199 mg, 0.314 mmol) and
D-galactopyran-
D
D
O - (h -
D
- Galactopyranosyl) - (1“4) - 2 - acetamido - 2-
deoxy- -galactopyranose (9).—A suspension of Pd/C
D
,
powdered 4 A molecular sieves (0.71 g) in dry CH2Cl2–
(10%, 150 mg) in EtOH (6 mL) was added to a solution
of 8 (200 mg, 0.197 mmol) in EtOH (6 mL). The
mixture was flushed with nitrogen and then with hydro-
gen and stirred overnight at rt under a hydrogen atmo-
sphere. The solution was filtered and the filtrate
concentrated and purified on a Biogel P-2 column,
using 95:5 water–n-butanol as eluant, to give 9 (51 mg,
0.133 mmol, 68%), [a]D +143° (c 0.6, water). [HRMS:
Calc. for C14H26NO11: 384.1506 (M+H+). Found:
384.1540 (M+H+)].
diethyl ether–toluene (1.6–1.1–1.1 mL) was stirred at
room temperature for 5 min. Solid DMTST (169 mg,
0.654 mmol) was added in portions for 2 h and then
pyridine (150 mL) was added to quench the reaction.
The mixture was diluted with diethyl ether and the
solids were filtered off. The filtrate was then washed
with aq 2 M H2SO4, aq 1 M NaHCO3, dried and
concentrated. Column chromatography (4:1 toluene–
EtOAc) of the residue gave 13 (203 mg, 0.141 mmol,
1
70%). NMR data (CDCl3): H, l 5.43 (d, 1 H, J1,2 8.4,
NMR data (D2O, 50 °C) are given in Table 1. Each
spectrum of 9 consists of two subspectra corresponding
to (a) or (b) configurations at the reducing end. The
two subspectra are distinguished by the suffixes (a) or
(b). The (a+b) N-acetyl signals were at 1.99 and
175.80/23.00. In the Gal spin system, the a and b
reducing end assignments (if unequal) may be reversed.
The (1“4) linkage was evident from the lC increase of
the substituted position as compared to an unsubsti-
tuted reference compound, as well as from 2D HMBC
experiments.
H-1%), 5.31 (s, 1 H, PhCH), 4.93 (d, 1 H, J1,2 3.3, H-1%%),
4.83 (d, 1 H, PhCH2), 4.79 (d, 1 H, PhCH2), 4.77 (d, 1
H, PhCH2), 4.64 (1 H, H-2%), 4.62 (d, 1 H, PhCH2), 4.49
(d, 1 H, PhCH2), 4.45 (d, 1 H, PhCH2), 4.43 (1 H,
H-5%%), 4.18 (1 H, H-3%%), 4.14 (1 H, H-6a), 4.08 (1 H,
H-3%), 4.06 (1 H, H-2%%), 3.88 (dd, 1 H, J5,6a 5.4, J6a,6b
8.5, H-6%b), 3.71 (dd, 1 H, J5,6a 1.5, J6a,6b 12.1, H-6b),
3.48 (t, 1 H, J 9.1, H-6%%a), 3.15 (s, 1 H, H-5), 3.10 (dd,
1 H, J5,6a 4.9, J6a,6b 8.8, H-6%%b). [HRMS: Calc. for
C89H87NO17Na:
1464.5872
(M+Na+).
Found:
1464.5845 (M+Na+)].
Benzyl O-(3,6-di-O-benzyl-2-deoxy-2-phthalimido-i-
Benzyl O-(2,3,4,6-tetra-O-benzyl-h-D-galactopyran-
D
-galactopyranosyl)-(1“3)-2-O-benzyl-4,6-O-benzyl-
osyl)-(1“4)-O-(2-acetamido-3,6-di-O-benzyl-2-deoxy-
i- -galactopyranosyl)-(1“3)-2-O-benzyl-4,6-O-ben-
zylidene-i- -galactopyranoside (14).—Hydrazine hy-
idene-i- -galactopyranoside (12).—A solution of
D
D
DMTST (910 mg, 3.52 mmol) in dry CH2Cl2 (5 mL)
was added to a stirred mixture of 3 (607 mg, 1.06
mmol), 10 (475 mg, 1.06 mmol) and molecular sieves 4
D
drate (127 mL, 3.27 mmol) was added to a mixture of 13
(228 mg, 0.158 mmol) in 2:3 toluene–EtOH (7 mL).
The mixture was refluxed for 36 h, and then cooled and
evaporated. The residue was partitioned between 1:1
CH2Cl2–EtOH and water. The organic layer was con-
centrated and then dissolved in 1:1 CH2Cl2–MeOH (6
mL) and treated with acetic anhydride (0.35 mL) at
room temperature. After 2 h, the mixture was concen-
trated. Column chromatography (2:1 petroleum ether–
EtOAc) gave 14 (207 mg, 0.153 mmol, 97%) as a solid.
,
A (3.85 g) in dry CH2Cl2 (10 mL) at 0 °C. The mixture
was stirred for 3 h at rt and then pyridine (0.77 mL)
was added. The mixture was diluted with diethyl ether
and the solids were filtered off. The filtrate was then
washed with aq 2 M H2SO4, aq 1 M NaHCO3, dried,
and concentrated. Column chromatography (2:1
petroleum ether–EtOAc) of the residue gave 11, con-
taminated with an unknown compound. Fractions con-
taining 11 were pooled, concentrated and dissolved in
dry MeOH (43 mL), and then 1 M NaOMe (2.15 mL)
was added. The mixture was stirred at room tempera-
ture for 12 h, then neutralized with Dowex-50 (H+),
and filtered. A drop of conc. ammonia (aq) was added
and the solution was concentrated and the residue was
purified by column chromatography (4:1 toluene–
EtOAc). Appropriate homogenous fractions were
pooled and concentrated to give 12 (357 mg, 0.388
1
NMR data (CDCl3): H, l 5.30 (s, 1 H, PhCH), 5.13 (1
H, J2,NH 7.4, NH), 5.09 (1 H, J1,2 8.3, H-1%), 4.93 (d, 1
H, H-1%%), 4.27 (1 H, 5%%), 4.17 (1 H, H-6a), 4.06 (1 H,
H-3%), 4.01 (1 H, H-2%%), 3.75 (1 H, H-6b), 3.42 (t, J 8.8,
H-6%%a), 3.42 (1 H, H-2%), 3.17 (s, 1 H, H-5), 3.14 (dd, 1
H, J5,6a 5.1, J6a,6b 8.8, H-6%%b), 1.95 (CH3CONH).
[HRMS: Calc. for C83H87NO16Na: 1376.5923 (M+
Na+). Found: 1376.5935 (M+Na+)].

1522
U. Westerlind et al. / Carbohydrate Research 337 (2002) 1517–1522
charide solutions. The agglutinating activity of these
samples was determined in U-shaped microtiter plates,
using 2% suspensions of NOR (for anti-NOR) or rabbit
O-(h-
deoxy-i-
D
-Galactopyranosyl)-(1“4)-O-(2-acetamido-2-
-galactopyranosyl)-(1“3)- -galactopyranose
D
D
(15).—A suspension of Pd/C (10%, 150 mg) in EtOH (6
mL) was added to a solution of 14 (110 mg, 81.2 mmol)
in EtOH (6 mL). The mixture was flushed with nitrogen
and then with hydrogen and stirred overnight at rt
under a hydrogen atmosphere. The solution was filtered
and the filtrate concentrated and purified on a Biogel
P-2 column, using 95:5 water–n-butanol as eluant, to
give 15 (27 mg, 49.5 mmol, 61%), [a]D +101° (c 0.3,
water). [HRMS: Calc. for C20H36NO16: 546.2034 (M+
H+), C20H35NO16Na: 568.1854 (M+Na+). Found:
546.2061 (M+H+), 568.1876 (M+Na+)].
erythrocytes (for anti-a-
D
-Galp-(1“3)-D-Gal), as al-
ready described.1
Acknowledgements
This work was supported by grant 4PO5A 124 18
from the Committee for Scientific Research (KBN,
Warsaw).
NMR data (D2O, 55 °C) are given in Table 2. Each
spectrum of 15 consists of two subspectra correspond-
ing to (a) or (b) configurations at the reducing end. The
two subspectra are distinguished by the suffixes (a) or
(b). Gal* denotes the reducing end Gal. The (a+b)
N-acetyl signals were at 1.998 and 175.99/23.06. In spin
systems one or more rings away from the reducing end,
the a and b reducing end assignments (if unequal) may
be reversed. The (1“4), (1“3) linkage pattern was
evident from the lC increase of the substituted position
as compared to unsubstituted reference compounds, as
well as from 2D HMBC experiments.
References
1. Kusnierz-Alejska, H.; Duk, M.; Storry, J. R.; Reid, M. E.;
Wiecek, B.; Seyfried, H.; Lisowska, E. Transfusion 1999,
39, 32–38.
2. Duk, M.; Reinhold, B. B.; Reinhold, V. N.; Kusnierz-Ale-
jska, G.; Lisowska, E. J. Biol. Chem 2001, 276, 40574–
40582.
3. Mourad, R.; Morelle, W.; Neveu, A.; Strecker, G. Eur. J.
Biochem. 2001, 268, 1990–2003.
4. Garegg, P. J.; Hultberg, H.; Lindberg, C. Carbohydr. Res.
1980, 83, 157–162.
5. David, S.; Thieffry, A.; Veyrieres, A. J. Chem. Soc., Perkin
Trans. 1 1981, 1796–1801.
6. van Seeventer, P. B.; Corsten, M. A.; Sanders, M. P.;
Kamerling, J. P.; Vliegenthart, J. F. G. Carbohydr. Res.
1997, 299, 171–179.
7. Nilsson, M.; Norberg, T. J. Carbohydr. Chem. 1990, 9,
1–14.
8. Ravenscroft, M.; Roberts, R. M. G.; Tillett, J. G. J. Chem.
Soc., Perkin Trans. 2 1982, 1569–1572.
Inhibition assay.—Anti-NOR and anti-a-
D
-Galp-
(1“3)-
D-Gal antibodies were isolated from human sera
by affinity procedures.2 The NOR erythrocytes were
pretreated with papain to enhance the agglutination.1
The aliquots of antibodies diluted to a titer 4 were
mixed with equal volumes of serially diluted oligosac-