Synthesis of a 3-deoxy-L-iduronic acid containing heparin pentasaccharide to probe the conformation of the antithrombin III binding sequence

Article abstract of DOI:10.1016/S0968-0896(98)00127-8

We report in this work the total synthesis of a close analogue of the pentasaccharide active site of heparin, in which the l-iduronic acid residue has been deoxygenated at position three. ¹H NMR studies demonstrated that, as anticipated, such a modification induces a shift of the conformational equilibrium toward ¹C₄ (contribution to the conformational equilibrium rises from 37% to 65%) and a substantial decrease of the affinity for antithrombin III (K(d) 0.154μM versus 0.050μM). Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00127-8

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1337±1346
Synthesis of a 3-Deoxy-L-iduronic Acid Containing Heparin
Pentasaccharide to Probe the Conformation of the
Antithrombin III Binding Sequencey
Ping-Sheng Lei,a Philippe Duchaussoy,b Philippe Sizun,c Jean-Maurice Mallet,a
Maurice Petitoub,* and Pierre Sinaya,{
a
Â
Ecole Normale Superieure, Departement de Chimie, URA CNRS 1686, 24 Rue Lhomond, 75231 Paris Cedex 05, France
Â
Â
Sano® Recherche, Ligne Hemobiologie, 195 Route d'Espagne, 31036 Toulouse Cedex, France
b
Â
cSano® Recherche, Recherche Exploratoire, 371 rue du Professeur Joseph Blayac, 34184 Monpellier Cedex 04, France
Received 17 February 1998; accepted 24 June 1998
AbstractÐWe report in this work the total synthesis of a close analogue of the pentasaccharide active site of heparin, in
1
which the l-iduronic acid residue has been deoxygenated at position three. H NMR studies demonstrated that, as
anticipated, such a modi®cation induces a shift of the conformational equilibrium toward C4 (contribution to the
1
conformational equilibrium rises from 37% to 65%) and a substantial decrease of the anity for antithrombin III (Kd
0.154 mM versus 0.050 mM). # 1998 Elsevier Science Ltd. All rights reserved.
Introduction
l-iduronic acid conformation is highly in¯uenced by the
substituents and the nature of the neighbouring units.8
Thus, while 1C4 was found to represent the predominant
conformer in heparin, a signi®cant shift of the con-
Heparin, a complex anionic polysaccharide,1 inhibits
blood coagulation through activation of antithrombin
III2 (AT III), a physiological inhibitor of coagulation.
Binding of heparin to AT III is mediated by the penta-
saccharide sequence 1.3 Chemical synthesis of this
pentasaccharide,4 and its methyl glycoside 2,5 allowed
2
formational equilibrium toward S0 was observed when
l-iduronic acid was next to a 3-O-sulfated glucosamine
unit (unit F).9 Since this extra-sulfate, at position 3 of
unit F, is the key structural element responsible for the
binding to AT III10 the question arises whether its
function is to directly interact with AT III or to drive l-
1
careful analysis of the H NMR coupling constants for
the l-iduronic acid unit (unit G) which ®nally led us to
conclude that the conformational equilibrium of this
monosaccharide could not be explained by the presence
of the two well known conformers 4C1 and 1C4 only, but
that a third one, 2S0, had also to be considered to
account for the observed pattern of coupling constants.6
The contribution of each of the conformers to the
conformational equilibrium could be computed from
1H NMR coupling constants7 and it appeared that
2
iduronic toward an `active' S0 conformation, or both.
The same issue is raised by the extra sulfate group
introduced on the H unit in various oligosaccharides,11
which confers to these compounds a higher anity for
AT III12 and strongly favours the 2S0 conformation. To
assess the role of the conformation itself in the inter-
action, it is necessary to compare compounds that contain
the same potential interaction sites (same pattern of
sulfates and carboxylates) yet di€er in their conforma-
tion. We reasoned that deoxygenation at position 3 of
unit G in 2 should shift the conformational equilibrium
Key words: l-Iduronic acid; pentasaccharide; heparin; anti-
thrombin III.
1
toward C4, by decreasing the non bonding interactions
*Corresponding author. Tel: +33 (0) 5 61 16 23 90; Fax: +33
(0) 5 61 16 22 86; E-mail: Maurice.Petitou@tls1.elfsano®.fr
yThis paper is dedicated to Professor Stuart Schreiber.
{Tel: +33 (0)1 44 32 33 90; Fax: +33 (0)1 44 32 33 97; E-mail:
pierre.sinay@ens.fr
occurring between O-3 and the other axial substituents
in this conformer. In the present article,13 we report the
synthesis of 30 (deoxygenated 2) and the comparison of
the biophysical and biochemical properties of 2 and 30.
0968-0896/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00127-8

1338
P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
Scheme 2. Reagents and conditions: (a) NiCl2, NaBH4, DMF,
96%; (b) 60%, AcOH, rt, 72 h, 78%; (c) MsCl, pyr, rt, 16 h,
99%; (d) AcOK, DMF, 80 ꢀC, 48 h, 63%; (e) tBuOH, CH2Cl2,
100%; (f) 0.1 N H2SO4, 100%; (f) 0.1 N H2SO4, 50 ꢀC, 2 h,
100%.
Results and Discussion
General strategy. The general selected synthetic strategy
is outlined in Scheme 1. The known11b trisaccharide
imidate 25 had already been successfully used as a gly-
cosyl donor for the synthesis of heparin pentasacchari-
dic fragments.11b,12 The major task of the present work
was thus the selective synthesis of the disaccharidic
building block 24, the glycosylation of which by 25
should then provide the protected pentasaccharide 26,
that would be easily converted into the ®nal product 30
following a well established methodology.4,5
to d-glucose (which does not give rise to furanoid forms
at equilibrium in aqueous solution) arises from the lack,
in the 3-deoxy compound, of the most unfavourable
interaction between the side chain at C-4 and the oxygen
atom at C-3 which are cis to each other in glucofuranose.
This trend is even more pronounced in the case of 3-
deoxy-l-lyxo-hexose because of the lower stability of the
pyranoid forms of l-idose itself. The previously repor-
ted18 strategy for the synthesis of pyranoid l-iduronic
acid derivatives is thus not applicable for the preparation
of 3-deoxy analogues. Acetylation of 3-deoxy-l-idose
would mainly provide an unwanted furanoid derivative.
Chemical synthesis of the disaccharidic building block
24. We initially decided to prepare the disaccharide 24
starting from 3-deoxy-l-idose (3-deoxy-l-lyxo-hexose)
eventually obtained from diacetone glucose via the
known14 3 as shown in the self explanatory Scheme 2.
We thus explored a second route starting from the
known19 methyl 2-O-benzyl-3-deoxy-a-d-ribo-hexo-
pyranoside 11 wherein, in contrast to the ®rst approach,
the pyranoid form is locked throughout the synthesis
(Scheme 3). The conversion of the iodo derivative 12
into the hex-5-enopyranoside 13 was ®rst attempted
using 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) either
in tetrahydrofuran,20 acetonitrile,21 or dimethylsulf-
oxide;22 in all cases 13 was only isolated in poor yields.
Finally, a satisfactory reaction (79%) was observed
upon treatment of 12 with a re¯uxed solution of sodium
methoxide in methanol although some substituted pro-
duct 14 was also isolated (17%).
Anomeric protons corresponding to the a and b anom-
ers of 9 and 10 were identi®ed in the 1H NMR spectrum
in deuterium oxide solution of the mixture, and the
proportion of the various constituents at equilibrium
could be easily determined. As shown in Table 1 the
furanoid forms are predominant in the case of 3-deoxy-
l-lyxo-hexose (3-deoxy-l-idose).
This feature can be easily explained. The noteworthy
known17 proportion (see Table 1) of furanoid forms in the
aqueous solution of 3-deoxy-d-ribo hexose as compared
Scheme 1. Synthetic route to 30.

P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
1339
Table 1. Proportion(%) of pyranoid and furanoid forms at equilibrium in deuterium oxide solution
Pyranoid forms
Furanoid forms
a
b
total
a
b
total
d-Glucose (at 31 ꢀC)17
38
62
55
36
12
100
79.5
74.5
24
±
5
±
15.5
14
±
3-Deoxy-d-ribo-hexose (at 31 ꢀC)17 (3-deoxy-d-glucose)
d-Idose (at 31 ꢀC)17
24.5
38.5
12
20.5
25.5
76
11.5
57
3-Deoxy-l-lyxo-hexose (3-deoxy-l-idose) (at 27 ꢀC)
19
1H NMR analysis of pentasaccharide 30. H NMR data
for 30 (Table 2) were obtained at 500 MHz.27 Signals
were assigned through a phase sensitive COSY experi-
ment.28 The spectrum was acquired as 4 KÂ512 (sine
bell weighting was applied in each dimension), zero-®l-
led to 4 KÂ2 K prior to double Fourier Transform, not
symmetrised. Signal assignments are in total agreement
with the structure. The 3J coupling constants were
obtained from the 1-D spectrum. The coupling con-
stants observed for the anomeric protons con®rmed the
expected con®gurations at these carbons. Concerning
the l-iduronic acid unit G, most noticeable is the
expected up®eld shift of the two protons G-3ax and G-
3eq of the deoxy position observed at 2.29 and 2.24 ppm
1
A detrimental extensive migration of the 4-O-acetyl
group to the primary position was observed during the
removal of the trityl group of 17 under various classical
conditions23 (Scheme 4). We found that a 2 min
treatment at 0 ꢀC of a dichloromethane solution of 17
with 60% aqueous perchloric acid limited the migration
(10%) and provided the expected primary alcohol 18 in
satisfactory yield (70%). The primary alcohol function
and the benzyl ether group were simultaneously oxidised
according to Sharpless24 to give, after treatment with
diazomethane, the l-iduronic acid derivative 20 as a
crystalline product. 1H NMR data at 250 MHz of 20
(CDCl3 solution) are in agreement with a 1C4 con-
formation. Formation of the glycosyl chloride 21 was
®nally achieved using dichloromethyl methyl ether in
the presence of zinc chloride25 at 60 ꢀC. When this reac-
tion was conducted at room temperature, anomerization
(88%) to the a-methyl glycoside was observed.
4
versus 4.17 for G-3 in 2. The observation of a J long-
range coupling constant between the anomeric proton
of l-iduronic acid residue G (observed as a broad sing-
let) and one of the two protons at position 3 of the same
unit is an indication for predominance of 1C4 con-
formation for the G unit. The equatorial position was
The glycosyl donor 21 was now reacted at 15 ꢀC in
dichloromethane with the known26 alcohol 22 in the
presence of silver tri¯ate to give selectively, albeit in
limited yield (36%), the crystalline disaccharide 23
(Scheme 5). The expected alcohol 24 was ®nally
obtained by selective de-O-acetylation of 23.
Preparation of the pentasaccharide 30. The alcohol 24
was condensed with the previously reported11b imidate
25 to give the protected pentasaccharide 26 (Scheme 6).
The ®nal deprotecting steps were achieved as extensively
developed by us in this ®eld.4,5
Scheme 4. Reagents and conditions: (a) TrCl, pyridine, 80 ꢀC,
16 h; (b) Ac2O, Pyr, rt, 80%; (c) 60% aq HClO4, 0 ꢀC, 5 min,
18: 70% 19: 10%; (d) RuCl3/NaIO4, H2O/CH3CN/CCl4, rt,
24 h then CH2N2/ether, 60%; (e) CHCl2OCH3, ZnCl2, 60 ꢀC,
5 h, 70%.
Scheme 3. Reagents and conditions: (a) Ph3P, I2, imidazole,
toluene, 70 ꢀC, 3 h, 83%; (b) MeONa/MeOH, re¯ux, 16 h, 13:
79%, 14: 18%; (c) BH3/THF then H2O2/NaOH, 78%, ido/
gluco (5.5/1).
Scheme 5. Reagents and conditions: (a) AgOTf, CH2Cl2,
15 ꢀC, 3 h, 36%; (b) HCl/MeOH, CH2Cl2, 3 days, 33%.

1340
P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
Scheme 6. Reagents and conditions: (a) TMSOTf, CH2Cl2, 20 ꢀC, 64%; (b) LiOOH, THF; (c) Et3N:SO3, DMF; (d) H2, Pd/C; (e)
Pyr:SO3, water (35% from 24).
assigned to the proton at 2.24 ppm on the basis of this
small long-range coupling. It is noteworthy that such a
long-range coupling is not observed in the proton 1H
NMR spectrum of 1 and 2 where the anomeric proton
of the l-iduronic acid residue appears as a doublet. No
such coupling was detected for the other proton at G-3.
The 3J coupling constants for all ring protons are shown
in Table 2.
weak di€erence between J2,3 and J3,4 (Table 2) indicates
that the 2S0 conformer is present in 30 although to a
rather low extent.
To obtain a more precise evaluation we used the method
reported by Ferro et al.,7 where the participation of each
conformer to the conformational equilibrium is com-
puted from the interproton coupling constants of the
iduronate ring. The values obtained in the present case
for 30 are depicted in Table 3. As expected, the pre-
Conformation of L-iduronic acid and biological activity.
It is now well documented6±9 that in heparin and
heparin related oligosaccharides, l-iduronic acid stands
in equilibrium between the three di€erent conformers
1C4, 4C1 and 2S0 depicted in Scheme 7. The pre-
dominance of the 2S0 conformer is easily revealed by
comparison of the J2,3 and J3,4 set of coupling con-
stants. Thus while J2,3 and J3,4 have similar values in the
presence of the chair forms only, J2,3 becomes much
1
dominant conformer is C4, the participation of which
amounts to 65%. The 2S0 conformer, which participates
to the extent of 64 % in 27, only represent 24% in 30
(Table 3).
2
The presence of the S0 conformer can also be probed
through NOE experiments.8 Thus, whereas the H-2/H-5
distance in the chair conformers is too large to give rise
to an observable NOE, these two protons are much
closer (2.3 A) in the case of the skewed boat (Scheme 7),
and a NOE is observed. Since the 2S0 conformer
participates to a greater extent to the conformational
equilibrium in 2, a stronger NOE should be observed in
this compound, compared to 30. NOE build-up experi-
ments on these two pentasaccharides were thus
designed, and four similar phase sensitive30 NOESY
spectra were acquired at 283 K (2 KÂ400 pts) with mix-
ing periods of 100, 150, 200, and 250 ms. NOESY
crosspeaks intensities were evaluated using the
UXNMR program. Some pairs of protons give rise to
NOE that can be easily observed and quanti®ed, the
results being reported in Table 4. It clearly appears that,
while similar values are obtained for various pairs of
proton in 2 and 30, an obvious di€erence is noted con-
cerning the H-2G/H-5G NOE and a much stronger
e€ect is observed in 2 compared to 30.
2
larger than J3,4 (and can reach 9 Hz) when the S0 con-
former contributes signi®cantly to the conformational
equilibrium. Thus, from a qualitative standpoint, the
Table 2. Chemical shifts (ppm) and coupling constants (Hz)
observed for 30
d
D
E
F
G
H
H-1
H-2
H-3
H-4
H-5
H-6
H-60
5.60 4.59
3.22 3.37
3.58 3.82a
3.54 3.81a
3.86 3.73
4.34
5.17
3.41
4.38
3.92
4.05
4.47
4.18
5.30
4.35
4.99
3.24
3.63
3.72
3.96
4.34
4.24
2.29 ax 2.24 eq
4.20
4.65
4.12
3J1,2
3.8
7.9
3.5
2.6
3.6
4J1,3eq
ꢁ1
3J2,3
3J3,4
10.0
9.2
9.4
9.2
10.6
9.1
4.6
4.6
5.5
4.0
10.3
8.7
2
0
J3,3
15.1
2.7
1
In conclusion the present H NMR data indicate that,
3J4,5
3J5,6
9.6
2.2
9.6
10.0
1.7
9.9
2.1
as expected, the removal of the oxygen at C-3 of the
iduronate ring induces a shift of the conformational
equilibrium of this unit at the expense of the 2S0
conformer. In order to evaluate the in¯uence of this
conformational change on the anity for AT III of such
3
0
J5,6
2.0
11.1
1.6
11.2
ꢁ1
5.6
11.7
2
0
J6,6
aAssignments of these two protons can be reversed.

P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
1341
Scheme 7.
pentasaccharides, we ®nally compared the binding con-
stants of 2 and 30. These constants were measured by
¯uorescence spectroscopy using a technique already
reported.31
according to standard procedures. Reactions were
monitored by TLC on silica gel 60 F254 (Merck) with
detection by charring with H2SO4. Unless otherwise
stated, column chromatography was performed on silica
gel 60 (E. Merck 63-200 mm). 1H NMR spectra were
recorded with Bruker AM100, AC 250, AM400 or AM
500 instruments for solution in CDCl3 (internal Me4Si)
unless otherwise stated. MS analyses were performed on
Nermag R10-10 instrument using chemical ionisation
(NH3) and detection of positive ions. Melting points
were determined in capillary tubes with a Buchi 510
apparatus, and are uncorrected. Optical rotations were
measured with a Perkin-Elmer Model 141 polarimeter at
23‹3 ꢀC. Elemental analyses were carried out at the
Service Central d'Analyses (C.N.R.S., Vernaison,
France)
As shown in Table 5, the pentasaccharide 2 has a much
higher anity for antithrombin. Since 2 and 30 have the
same sulfation pattern, and considering that the inter-
action with the protein mainly involves the sulfate and
the carboxylate groups,12,32 we conclude that the anity
for the protein is highly in¯uenced by the conformation
2
of G. This anity is higher when the S0 conformation
at l-iduronic acid predominates. This conclusion was
recently strengthened by the ®nding that
a con-
formationally constrained heparin-like pentasaccharide
where the iduronic acid unit is ®xed in the 1C4 con-
formation displays a very low activity in an AT III
mediated anti-factor Xa assay.33 However since the
compounds used in the present study can reach several
conformations, one cannot tell whether the pre-
dominance of the 2S0 state of l-iduronate is impor-
tant for initial recognition by the protein or if it
favours the subsequent induced ®t known to occur dur-
ing the assembling of the antithrombin±pentasaccharide
complex.
3-Deoxy-1,2:5,6-di-O-isopropylidene-ꢀ-D-ribo-hexofuran-
ose (4). To a solution of 3-deoxy-3-iodo-1,2:5,6-di-O-
isopropylidene-a-d-allofuranose (2 g, 5.5 mmol), and
NaBH4 (0.82 g, 21.5 mmol) in anhydrous DMF (40 mL),
wad added NiCl2 (1.7 g, 13.2 mmol) at 0 ꢀC. After
30 min at room temperature, the solution was con-
centrated. A solution of the residue in chloroform was
washed with water, dried (MgSO4) and concentrated to
yield 4 (1.26 g, 96%), which was directly engaged in the
next step.
Experimental
3-Deoxy-1,2-O-isopropylidene-ꢀ-D-ribo-hexofuranose (5).
A solution of 3-deoxy-1,2:5,6-di-O-isopropylidene-a-d-
ribo-hexofuranose (4) (107 mg, 0.44 mmol) in aq acetic
General. All solvents and reagents were of the best
commercially available grade or were puri®ed and dried
Table 3. Participation of the three conformers to the
conformational equilibrium of 28 and 30
Table 5. Dissociation constants of antithrombin III and
compounds 2 and 30
1C4
4C1
2S0
Kd (mM)
30
2
65
36
11
0
24
64
30
2
0.154
0.050
Table 4. Relative intensities of NOEs (%) observed for some pairs of protons in 2 and 30
H-2G/H-5G
H-5G/H-4G
H-1E/H-6F
H-1E/H-5E
H-5F/H-6F
H-1E/H-4F
30
8.0
13.6
0.59
33.4
34.6
0.97
30.5
32.5
0.94
49
52.5
0.93
25
26.4
23.2
23.2
1.00
2
30/2
0.95

1342
P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
acid (60%, 1 mL) was stirred for 60 h at room tempera-
ture, and concentrated. The residue was taken up in a
chloroform/hexane mixture and 5 (70 mg, 78%) was ®l-
tered o€. Mp 82±83 ꢀC (chloroform/hexane). Lit.15a,b
mp 84 ꢀC.
3.03 (m, 1H, H-5), 2.80 (m, 2H, H-6a,6b), 2.16 (dd, 1H,
J3a,3e=14 Hz, H-3e), 1.85 (ddd, 1H, H-3a), 1.49 and
1.31 (2s, 6H, 2CH3). MS: 204 (M+18), 187 (M+1),
171, 146. Anal. calcd for C9H14O4 (186.209): C, 58.05;
H, 7.58. Found: C, 58.03; H, 7.65.
3-Deoxy-1,2-O-isopropylidene-5,6-di-O-methanesulfonyl-
ꢀ-D-ribo-hexofuranose (6). Methanesulfonyl chloride
(0.48 mL, 6.2 mmol) was added at 0 ꢀC to a solution of 5
(500 mg, 2.4 mmol) in dry pyridine (5 mL). The resulting
solution was stirred for 24 h at room temperature, and
concentrated. A solution of the residue in dichloro-
methane was washed with water (3Â30 mL), dried
(MgSO4) and concentrated. Column chromatography
(chloroform/ethyl acetate, 3/1) gave 6 (874 mg, 99%),
mp 90±91 ꢀC (acetone/water). [a]d 1 (c 1.1, chloro-
3-Deoxy-L-lyxo-hexose (9) and (10). A solution of 7
(100 mg, 0.5 mmol) in aq sulfuric acid (0.1 N, 2 mL) was
stirred at 50 ꢀC for 2 h, neutralised by 1 N aq sodium
hydroxide, and concentrated. Column chromatography
(chloroform/methanol, 10/1) gave 3-deoxy-l-lyxo-hex-
ose (70 mg, 80%). 1H NMR (D2O): d 5.31 (d, 1H,
J1,2=4 Hz, H-1 b-fur), 5.24 (s, H-1 a-fur), 5.11 (s, H-1
b-fur), 4.5±3.5 (m, 5H, H-2, H-4, H-5, H-6a,6b), 2.19±
1.89 (m, 2H, H-3a,3b); 13C NMR (D2O): d 102.85 (C-1,
a-fur), 98.04 (C-1, b-pyr) 35.51 et 31.43 (C-3, a- and b-
pyr), 33.87 (C3, a-fur), 33.83 (C-3, b-fur).
form). Lit.16 mp 84.5±85.5 ꢀC (ether/pentane) [a]26
1ꢀ
d
(c 2.46, chloroform). 1H NMR:
d 5.88 (d, 1H,
J1,2=3.5 Hz, H-1), 4.98 (m, 1H, J4,5=6 Hz, J5,6a=3 Hz,
J5,6b=5 Hz, H-5), 4.83 (dd, 1H, J2,3a=4.5 Hz, H-2), 4.58
(dd, 1H, J6a,6b=12 Hz, H-6a), 4.42 (dd, 1H, H-6b), 4.40
(dt, 1H, J3e,4=4.5 Hz, J3a,4=12 Hz, H-4), 3.17 and 3.12
(2s, 6H, 2 CH3S), 2.30 (dd, 1H, J3a,3e=14 Hz, H-3e),
2.00±1.84 (ddd, 1H, H-3a), 1.50 and 1.30 (2s, 6H,
2CH3). MS: 378 (M+18), 361 (M+1), 345, 320, 284.
Methyl 2-O-benzyl-3,6-dideoxy-6-iodo-ꢀ-D-ribo-hexo-
pyranoside (12). A solution of 11 (64 mg, 0.24 mmol),
triphenylphosphine (173 mg, 0.66 mmol, 3 equiv), imi-
dazole (45 mg, 0.66 mmol, 3 equiv), and iodine (74 mg,
0.29 mmol, 1.3 equiv) in anhydrous toluene (2 mL) was
stirred at 70 ꢀC for 3 h. The reaction mixture was
washed with aq Na2S2O3 (10%, 2 mL), aq saturated
NaHCO3, water, dried (MgSO4), and concentrated.
Column chromatography (chloroform/acetone, 5/1)
6-O-Acetyl-3-deoxy-1,2-O-isopropylidene-5-O-methane-
sulfonyl-ꢀ-D-ribo-hexofuranose (7). A mixture of 6
(120 mg, 0.33 mmol) and potassium acetate (74 mg,
0.83 mmol) in dry DMF (1 mL) was stirred for 60 h at
80 ꢀC and concentrated. A solution of the residue in
dichloromethane, was washed with water, dried
(MgSO4), and concentrated. Column chromatography
(toluene/ethyl acetate, 4/1) gave 7 (68 mg, 63%). [a]d
+4 (c 0.8, chloroform). 1H NMR: d 5.82 (d, 1H,
J1,2=3.5 Hz, H-1), 4.98 (m, 1H, J4,5=4 Hz,
J5,6a=3.0 Hz, J5,6b=7 Hz, H-5), 4.79 (dd, 1H,
J2,3a=4.5 Hz, H-2), 4.44 (dd, 1H, J6a,6b=12 Hz, H-6a),
4.40±4.36 (dt, 1H, J3e,4=4.5 Hz, J3a,4=12 Hz, H-4), 4.14
(dd, 1H, H-6b), 3.10 (s, 3H, CH3S), 2.08 (dd, 1H,
J3a,3e=14 Hz, H-3e), 2.01 (s, 3H, OAc), 1.96 (ddd, 1H,
H-3a), 1.50 and 1.30 (2s, 6H, 2CH3). MS: 342 (M+18),
325 (M+1), 309, 284. Anal. calcd for C12H20O8S
(324.352): C, 44.44; H, 6.21. Found: C, 44.80; H, 6.28.
1
gave 12 (75 mg, 83%). [a]d +59 (c 1.4, chloroform). H
NMR: d 7.40 (m, 5H, arom), 4.86 (d, 1H, J1,2=3.5 Hz,
H-1), 4.67 (AB, 2H, CH2Ph), 3.67±3.30 (m, 5H, H-
2,4,5,6a,6b), 3.56 (s, 3H, OCH3), 2.30 (m, 1H, H-3e),
1.98 (q, 1H, H-3a). MS: 396 (M+18), 379 (M+1), 364
(-CH3), 347 (-OH). Anal. calcd for C14H19O4I (378.208):
C, 44.46; H, 5.06. Found: C, 44.26; H, 5.11.
Methyl 2-O-benzyl-3,6-dideoxy-ꢀ-D-ribo-hex-5-enopyrano-
side (13). Sodium methoxide (1 M) in methanol
(10 mL) was added to a solution of 12 (720 mg,
1.9 mmol) in methanol (10 mL). The mixture was stirred
at 80 ꢀC for 12 h, cooled, and diluted with water (25 mL)
and dichloromethane. The organic layer was washed
with cold aq diluted HCl, water, dried (MgSO4), and
concentrated. Column chromatography (chloroform/
acetone, 5/1) gave ®rst 13 (375 mg, 79%), mp 59±60 ꢀC
(ethyl acetate/hexane). [a]d +52 (c 2, chloroform). IR
(chloroform) Vmax (cm 1): 1665 (C=C) 2940, 3020
(benzyl). 1H NMR: d 7.45±7.35 (m, 5H, arom), 4.80±4.60
(m, 5H, H-1,6a,6b, CH2Ph), 4.10 (m, 1H, J3e,4=5.5 Hz,
J3a,4=11 Hz, J4,OH=8 Hz, H-4), 3.75±3.45 (ddd,
J2,3e=4.5 Hz, J1,2=3 Hz, J2,3a=11 Hz, H-2), 3.50 (s, 3H,
OCH3), 2.30±2.15 (m, 1H, J3a,3e=11 Hz, H-3e), 2.05±
1.90 (q, 1H, H-3a). MS: 268 (M+18), 234, 219. Anal.
calcd for C14H18O4 (250.296): C, 67.18; H, 7.15. Found:
C, 66.84; H, 7.44.
5,6-Anhydro-3-deoxy-1,2-O-isopropylidene-ꢁ-L-lyxo-hexo-
furanose (8). A solution of potassium tert-butoxide
(179 mg, 1.6 mmol, 2 equiv) in tert-butanol (3 mL) was
added, at 0 ꢀC, to a solution of 7 (254 mg, 0.8 mmol) in
anhydrous dichloromethane (5 mL). The reaction mix-
ture was stirred for 6 h at room temperature, ®ltered,
and concentrated. Column chromatography (chloro-
form/ethyl acetate, 10/1) gave 8 (345 mg, 95%). [a]d
5
5.80 (d, 1H,
(c 2.7, chloroform). 1H NMR:
J1,2=3.5 Hz, H-1), 4.74 (dd, 1H, J2,3a=4 Hz, H-2), 4.15
(dt, 1H, J3a,4=11 Hz, J3e,4=4.5 Hz, J4,5=4 Hz, H-4),
d
Then 14 was eluted (4 mg, 17.5%).

P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
1343
Methyl 2-O-benzyl-3-deoxy-6-O-methyl-ꢀ-D-ribo-hexo-
pyranoside (14). [a]d +35ꢀ (c 0.7, chloroform). 1H
NMR: d 7.40 (m, 5H, arom), 4.67 (d, 1H, J1,2=4 Hz, H-
1), 4.55 (AB, 2H, CH2Ph), 3.70±3.45 (m, 5H, H-
2,4,5,6a,6b), 3.40 (2s, 6H, 2CH3), 2.20 (m, 1H, H-3e),
1.80 (m, 1H, H-3a). Anal. calcd for C15H22O5 (282.338):
C, 63.81; H, 7.85. Found: C, 63.74; H, 7.92.
J2,3a=4 Hz, H-2), 3.50 (s, 3H, OCH3), 3.45±3.30 (m, 2H,
J6a,6b=15 Hz, H-6a,6b), 2.40 (dt, 1H, J3a,3e=15 Hz, H-
3e), 1.75 (dt, 1H, H-3a). Anal. calcd for C35H36O6
(552.669): C, 76.06; H, 6.57. Found: C, 75.58; H, 6.97.
Methyl 4-O-acetyl-2-O-benzyl-3-deoxy-ꢁ-L-lyxo-hexo-
pyranoside (18). Sixty percent aq perchloric acid
(6 mmol, 10 equiv) was added to a solution of 17
(330 mg, 0.6 mmol) in dichloromethane (12 mL), at 0 ꢀC.
The reaction mixture was stirred for 2 min then poured
into a dropping funnel containing cold water and
dichloromethane. The organic layer was washed with
water, dried (MgSO4), and concentrated. Column chro-
matography (chloroform/acetone, 10/1) gave ®rst 18
Methyl 2-O-benzyl-3-deoxy-ꢁ-L-lyxo-hexopyranoside (15).
A 1 M solution of BH3 in THF (4.8 mL) was added to a
solution of 13 (300 mg, 1.2 mmol) in dry THF (10 mL).
The reaction mixture was stirred for 1 h then diluted
with ethanol. Aqueous NaOH, (3 M, 3 mL) and H2O2
(3 mL) were added, and after 2 h stirring at 50 ꢀC, the
solution was diluted with chloroform, washed with
water, dried (MgSO4), and concentrated. Column chro-
matography (toluene/methanol, 8/1) gave 15 (212 mg,
66%) then 11 (38 mg, 12%). 15: mp 63 ꢀC (ethyl acetate/
hexane). [a]d +94 (c 2.4, chloroform). 1H NMR: d
7.45±7.30 (m, 5H, arom), 4.80 (AB, 2H, CH2Ph), 4.42 (d,
1H, J1,2=0.5 Hz, H-1), 4.03 (dd, 1H, J6a,6b=9 Hz,
J5,6a=4.5 Hz, H-6a), 3.85 (dd, 1H, J5,6b=3 Hz, H-6b),
3.75 (m, 2H, H-2,4), 3.60 (2s, 4H, H-5 and OCH3), 2.20
(dt, 1H, J3a,3e=10 Hz, J2,3e=3 Hz, J3e,4=3 Hz, H-3e),
1.70 (dt, 1H, J2,3a=3 Hz, J3a,4=3 Hz, H-3a). MS: 286
(M+18), 269 (M+1), 254 (-OCH3), 237, 108. Anal.
calcd for C14H20O5 (268.311): C, 62.67; H, 7.51. Found:
C, 62.76; H, 7.62.
1
(130 mg, 70%). [a]d +58 (c 0.6. chloroform). H NMR:
d 7.45±7.35 (m, 5H, arom), 4.95 (m, 1H, J3e,4=4.5 Hz,
J3a,4=4 Hz, J4,5=2.5 Hz, H-4), 4.72 (AB, 2H, CH2Ph),
4.50 (d, 1H, J1,2=2 Hz, H-1), 3.90 (m, 1H, J5,6a=5 Hz,
J6a,6b=12 Hz, H-6a), 3.83 (1H, J5,6b=6 Hz, H-5), 3.70
(1H, H-6b), 3.65 (1H, J2,3e=4.5 Hz, J2,3a=4 Hz, H-2),
3.62 (s, 3H, OCH3), 2.52 (s, 1H, OH), 2.40 (dt, J3a,3e
15 Hz, H-3e), 2.10 (s, 3H, OAc), 1.70 (dt, H-3a). MS:
328 (M+18), 311 (M+1), 300, 296, 279. Anal. calcd for
C16H22O6 (310.348): C, 61.92; H, 7.15. Found: C, 61.93;
H, 7.20.
Then 19 was eluted (18 mg, 10%).
Methyl 6-O-acetyl-2-O-benzyl-3-deoxy-ꢁ-L-lyxo-hexo-
pyranoside (19). Melting point 50±51 ꢀC (methanol). [a]d
+72 (c 0.7, chloroform). 1H NMR: d 7.45±7.30 (m, 5H,
arom), 4.80 (AB, 2H, CH2Ph), 4.45±4.35 (m, 3H, H-
1,6a,6b), 3.80±3.70 (m, 3H, H-2,4,5), 3.60 (s, 3H,
Methyl 2-O-benzyl-3-deoxy-6-O-trityl-ꢁ-L-lyxo-hexo-
pyranoside (16). A solution of 15 (340 mg, 1.26 mmol)
and trityl chloride (432 mg, 1.55 mmol, 1.2 equiv) in dry
pyridine (8 mL) was stirred at 80 ꢀC until TLC (chloro-
form/ethyl acetate, 10/1) showed complete conversion of
15. After concentration, column chromatography
OCH3), 2.35 (dt, 1H, J3a,3e=15 Hz, J2,3e=2.5 Hz, J3e,4
2.5 Hz, H-3e), 2.10 (s, 3H, OAc), 1.70 (dt, 1H, J2,3a
=
=
1
(chloroform/acetone, 10/1) gave 16. H NMR: d 7.60±
2.7 Hz, J3a,4=2.7 Hz, H-3a). Anal. calcd for C16H22O6
(310.348): C, 61.92; H, 7.15. Found: C, 62.04; H, 7.28.
7.20 (m, 20H, arom), 4.80 (AB, 2H, CH2Ph), 4.40 (d,
1H, J1,2=1 Hz, H-1), 3.80±3.40 (m, 5H, H-2,4,5,6a,6b),
3.65 (s, 3H, OCH3), 2.30 (dt, J3a,3e15 Hz, J2,3e=3 Hz,
J3e,4=3 Hz, H-3e), 1.65 (dt, 1H, J2,3a=3 Hz,
J3a,4=3 Hz, H-3a).
Methyl (methyl 4-O-acetyl-2-O-benzoyl-3-deoxy-ꢁ-L-
lyxo-hexopyranosyl)uronate (20).
A mixture of 18
(220 mg, 0.7 mmol), ruthenium (III) chloride hydrate
(3 mg, 0.015 mmol, 2.2% equiv), sodium periodate
(600 mg, 2.8 mmol, 4 equiv), water (9 mL), acetonitrile
(6 mL) and tetrachloromethane (6 mL) was vigorously
stirred for 30 h at room temperature then diluted with
chloroform. The organic layer was separated, the aq
layer was treated twice with chloroform. The organic
solutions were pooled and concentrated. A solution of
the residue in ether (5 mL) was treated with an excess of
diazomethane in ether, at 0 ꢀC, and concentrated. Col-
umn chromatography (ethyl acetate/hexane, 1/1) gave
20 (135 mg, 54%), mp 175 ꢀC (ethyl acetate). [a]d +122
(c 0.5, chloroform). 1H NMR: 8.20 and 7.60±7.50 (m,
5H, arom), 5.35±5.30 (m, 2H, H-2,4), 4.67 (d, 1H,
J1,2=1.5 Hz, H-1), 4.43 (d, 1H, J4,5=2 Hz, H-5), 3.85 (s,
3H, CO2CH3), 3.65 (s, 3H, OCH3), 2.60 (dt, 1H,
Methyl 4-O-acetyl-2-O-benzyl-3-deoxy-6-O-trityl-ꢁ-L-
lyxo-hexopyranoside (17). Acetic anhydride (5.2 mmol,
4 equiv) was added to a solution of 16 (corresponding to
1.26 mmol of diol 15) in dry pyridine (8 mL) at 0 ꢀC. The
reaction mixture was stirred for 3 h at rt then con-
centrated. A solution of the residue in chloroform was
washed with water, dried (MgSO4), and concentrated.
Column chromatography (chloroform/acetone, 10/1)
gave 17 (696 mg, 100%), mp 108±109 ꢀC (ethyl acetate/
hexane). [a]d +80 (c 0.7, chloroform). 1H NMR: d
7.50±7.25 (m, 20H, arom), 5.25 (m, 1H, J3a,4=4 Hz,
J3e,4=4 Hz, J4,5=2 Hz, H-4), 4.65 (AB, 2H, CH2Ph),
4.45 (d, 1H, J1,2=1.5 Hz, H-1), 3.95 (m, 1H,
J5,6a=9 Hz, J5,6b=5 Hz, H-5), 3.55 (m, 1H, J2,3e=4 Hz,

1344
P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
J3a,3e=15 Hz, J2,3e=3.5 Hz, J3e,4=3.5 Hz, H-3e), 2.13
(dt, 1H, J2,3a=4 Hz, J3a,4=4 Hz, H-3a), 1.90 (s, 3H,
OAc). Anal. calcd for C17H20O8 (352,342): C, 57.95; H,
5.72. Found: C, 58.03; H, 5.68.
ture was stirred for 3 days at room temperature, washed
with aq saturated NaHCO3, water, dried (MgSO4), and
concentrated. Column chromatography on silica gel
(ethyl acetate/hexane, 1/1) gave 24, (220 mg, 33%). [a]d
1
+47 (c 0.6, chloroform). H NMR: d 8.08, 7.88, 7.55±
Methyl 4-O-acetyl-2-O-benzoyl-1-chloro-1,3-di-deoxy-ꢀ-
L-lyxo-hexopyranosyl uronate (21). A solution of 20
(100 mg, 0.28 mmol) and a catalytic amount of anhy-
drous zinc chloride in dichloromethyl methyl ether
(5 mL) was heated at 60 ꢀC for 5 h, diluted with toluene,
®ltered and concentrated. The residue was chromato-
graphed to give 21 (68 mg, 67%). 1H NMR: d 8.15±7.53
(m, 5H, arom), 6.40 (s, 1H, H-1), 5.47 (m, 1H, H-4),
5.25 (m, 1H, H-2), 4.92 (d, J4,5=2 Hz, H-5), 3.85 (s, 3H,
CO2CH3), 2.65 (dt, 1H, Jgem=16 Hz, J2,3a=3.5 Hz, H-
3a), 2.46 (dm, 1H, Jgem=16 Hz, H-3e), 1.97 (s, 3H, OAc).
7.25 (m, 20H, arom), 5.34 (s, 1H, H-10), 5.03±5.01 (m,
3H, CO2CH2Ph, NH,), 5.01 (m, 1H, H-20), 4.90 (dd,
1H, Jgem=12 H0z, J5,6a=2.0 Hz, H-6a), 4.83 (d, 1H,
0
0
J4 ,5 =1 Hz, H-5 ), 4.78±4.60 (AB, CH2Ph), 4.69 (d, 1H,
J1,2=3.5 Hz, H-1), 4.45 (dd, 1H, Jgem=12 Hz,
J5,6b=4.5 Hz, H-6b), 4.16 (dt, 1H, J1,2=3.5 Hz,
J2,3=10 Hz, J2,NH=10 Hz, H-2), 4.10 (t, 1H,
J3,4=10 Hz, J4,5=10 Hz, H-4), 3.98 (ddd, 1H,
J4,5=10 Hz, J5,6a=2.0 Hz, J5,6b=4.5 Hz, H-5), 3.94
(m, 1H, H-40), 3.78 (t, 1H, J2,3=10 Hz, J3,4=10 Hz, H-3),
3.53 (s, 3H, CO2CH3), 3.38 (s, 3H, OCH3), 2.23 (m, 2H,
H-30e, H-30a). MS: 817 (M+17), 801 (M+1), 769, 522.
Anal. calcd for C43H45O14N (799.831): C, 64.57; H,
5.67; N, 1.75. Found: C, 64.54; H, 5.62; N, 1.70.
Methyl 6-O-benzoyl-3-O-benzyl-2-(benzyloxycarbonyl)-
amino-2-deoxy-4-O-(methyl-4-O-acetyl-2-O-benzoyl-3-
deoxy-ꢀ-L-lyxo-hexopyranosyluronate)-ꢀ-D-glucopyrano-
side (23). A mixture of 21 (prepared and directly
engaged, from 480 mg, 1.4 mmol of 20), methyl 6-O-
benzoyl-3-O-benzyl-2-(benzyloxycarbonyl)amino-2-
deoxy-a-d-glucopyranoside 22 (1.54 mmol, 1.1 equiv),
and 4 A molecular sieves, in anhydrous dichloromethane
(5 mL) was stirred for 40 min at room temperature under
argon then cooled to 15 ꢀC. Silver tri¯ate (400 mg,
1.54 mmol) was added, and the reaction mixture was
stirred in the dark for 3 h, diluted with CH2Cl2, ®ltered,
and concentrated. Column chromatography (chloro-
form/acetone, 50/1) and crystallisation from ethyl acet-
ate gave 23, (413 mg, 35% from 20), mp 136 ꢀC (ethyl
acetate). [a]d +52 (c 0.4, chloroform). 1H NMR: d
8.08, 7.98, 7.53±7.28 (m, 20H, arom), 5.33 (s, 1H, H-1),
5.03 (m, 4H, CO2CH3, NH, H-40), 4.95 (m, 1H, H-20),
Methyl O-(6-O-acetyl-2-azido-3,4-di-O-benzyl-2-deoxy-
ꢀ-D-glucopyranosyl)-(1!4)-O-(methyl-2,3-di-O-benzyl-
ꢁ-D-glucopyranosyluronate)-(1!4)-O-(3,6-di-O-acetyl-2-
azido-2-deoxy-ꢀ-D-glucopyranosyl)-(1!4)-O-(methyl 2-
O-benzoyl-3-deoxy-ꢀ-L-lyxo-hexopyranosyluronate)-(1!4)-
6-O-benzoyl-3-O-benzyl-2-(benzyloxycarbonyl)amino-2-
deoxy-ꢀ-D-glucopyranoside (26).
A solution of 25
(340 mg, 0.28 mmol) and 24 (225 mg, 0.28 mmol), in dry
CH2Cl2 (7 mL) containing 4 A molecular sieves (500 mg),
was stirred 30 min at room temperature under argon then
cooled to 20 ꢀC. A 0.04 M solution of trimethylsilyl tri-
¯uoromethanesulfonate (490 mL, 0.019 mmol) in CH2Cl2
was added. After 30 min the mixture was neutralised
with solid NaHCO3, ®ltered, and concentrated. The
residue was chromatographied on Sephadex LH 20
(CHCl3/methanol, 1/1). Further puri®cation was
achieved on silica gel (CHCl3/EtOAc, 7/1) yielding 26
(334 mg, 64%) as a white foam. [a]d + 81 (c 1.2,
CHCl3). 1H NMR: d 7.42±7.28 (m, H, arom.), 5.53 (d, 1
H, J1,2=3.8 Hz, H-1D), 5.46 (d, 1 H, J1,2=2.1 Hz, H-
1G), 5.33 (dd, 1 H, J2,3=10.8 Hz, J3-4=9.1 Hz, H-3F),
5.03 (2d, 2 H, J1,2=3.5 Hz, H-1F and H-1H), 4.36 (d, 1
H, J1,2=7.8 Hz, H-1E), 3.86 (d, 1 H, J4-5=9.7 Hz, H-
5E), 3.76 (s, 3 H, COOMe), 3.49 (dd, 1 H, J2,3=9.1 Hz,
H-2E), 3.45 (s, 6 H, OMe and COOMe), 3.34 (dd, 1 H,
J2,3=10.4 Hz, H-2D), 3.21 (dd, 1 H, J2,3=10.8 Hz, H-
2F), 2.50±2.42 (m, 1 H, H-30G); 2.14±2.10 (m, 1 H, H-
3G), 2.10, 2.06, 2.05 (3 s, 9 H, Ac). MS-ESI, positive
mode: (M+Na)+ m/z 1873.9, (M+2Na)2+ m/z 948.6.
Anal. calcd. for C96H103N7O31: C, 62.30; H, 5.61; N,
5.30. Found: C, 62.03; H, 5.62; N, 5.33.
4.90 (d, 1H, J4 ,5 =2.2 Hz, H-50), 4.86 (dd, 1H,
0
0
Jgem=12 Hz, J5,6a=2.2 Hz, H-6a), 4.73 (AB, CH2Ph),
4.69 (d, 1H, J1,2=3.5 Hz, H-1), 4.44 (dd, 1H, Jgem
12 Hz, J5,6b=4.5 Hz, H-6b), 4.15 (dt, 1H, J1,2=3.5 Hz,
J2,3=10 Hz, J2,NH=10 Hz, H-2), 4.10 (t, 1H, J3,4
=
=
10 Hz, J4,5=10 Hz, H-4), 3.97 (ddd, 1H, J4,5=10 Hz,
J5,6a=2.2 Hz, J5,6b=4.5 Hz, H-5), 3.77 (t, 1H,
J2,3=10 Hz, J3,4=10 Hz, H-3), 3.45 (s, 3H, CO2CH3),
3.40 (s, 3H, OCH3), 2.32 (dm, 1H, Jgem=16 Hz, H-30e),
2.21 (dt, 1H, Jgem=16 Hz, J2,3a=4 Hz, J3a,4=4 Hz, H-
30a), 1.91 (s, 3H) OAc). MS: 859 (M+17), 842 (M), 828,
752, 391. Anal. calcd for C45H47O15N (841.869): C, 64.20
H, 5.63. Found: C, 64.33 H, 5.66.
Methyl 6-O-benzoyl-3-O-benzyl-2-(benzyloxycarbonyl)-
amino-2-deoxy-4-O-(methyl-2-O-benzoyl-3-deoxy-ꢀ-L-
lyxo-hexopyranosyluronate)-ꢀ-D-glucopyranoside (24). A
solution of hydrogen chloride in methanol (prepared
from 1.5 mL of acetyl chloride and 2.2 mL of methanol)
was added to a solution of 23 (700 mg, 0.83 mmol) in
anhydrous dichloromethane (12 mL). The reaction mix-
Methyl O-(2-deoxy-2-N-sulfonato-6-O-sulfonato-ꢀ-D-
glucopyranosyl)-(1!4)-O-(ꢁ-D-glucopyranosyluronate)-
(1!4)-O-(2-deoxy-2-N-sulfonato-3,6-di-O-sulfonato-ꢀ-D-
glucopyranosyl)-(1!4)-O-(3-deoxy-2-O-sulfonato-ꢀ-L-

P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
1345
lyxo-hexopyranosyluronate)-(1!4)-O-2-N-sulfonato-6-O-
sulfonato-ꢀ-D-glucopyranoside, decasodium salt (30). (i)
Saponi®cation. An aq solution of H2O2 (30%, 8.5 mL)
was added to a cooled ( 5 ꢀC) solution of 26 (0.387 g,
0.21 mmol) in THF (21 mL), and aq LiOH (0.7 M,
5 mL) was introduced dropwise. After 16 h at room
temperature methanol (20 mL) and aq NaOH (4 M,
5.5 mL) were added. Twenty-four hours later the mix-
ture was acidi®ed (6 M aq HCl) and diluted with H2O.
The compound was extracted with CH2Cl2, washed with
10% aq Na2SO3, H2O, dried, and concentrated to give
white a powder (0.282 g,90%). Tlc: Rf=0.55, EtOAc/
pyridine/acetic acid/water, 12/2/0.6/1. (ii) O-Sulfona-
tion. A solution of the above compound in dry DMF
(8.0 mL) and Et3N:SO3 complex (0.860 g, 4.75 mmol)
was heated at 50 ꢀC for 20 h. After cooling aq NaHCO3
(1.6 g in 19 mL of H2O) was added. After 16 h stirring,
evaporation to dryness gave a white residue that was
treated with CH2Cl2/methanol, 1/1 (40 mL). After ®l-
tration and concentration, the solid residue was treated
with CH2Cl2/methanol, 8/2 (25 mL), and after ®ltration
the solution was concentrated. (iii) Hydrogenation. A
solution of the O-sulfonated compound in tert-butyl
alcohol/water, 13/20 (33 mL) was stirred with 10% Pd/C
(300 mg), under hydrogen, for 48 h then ®ltered and
concentrated. The treatment was repeated (at which
point no signals of benzyl group were detected by 1H
NMR). (iv) N-sulfonation. The hydrogenated com-
pound was dissolved in water (10 mL) and the pH was
adjusted to 9.5 with 2 M NaOH. Sulfur trioxide±pyr-
idine complex (89.0 mg, 0.57 mmol) was added, and the
pH was maintained to 9.5 by addition of 2 M NaOH.
After 1 h at room temperature more complex (89.0 mg,
0.57 mmol) was added. The mixture was chromato-
graphed on Sephadex G-25 (500 mL) equilibrated with
0.2 N NaCl. The fractions containing the expected pro-
duct were pooled and puri®ed on Q Sepharose fast-¯ow
(100 mL) using a NaCl gradient (0.75±1.2 M). Pooled
fractions were desalted on Sephadex G-25 (500 mL) and
lyophilised to give 30 (113 mg, 35% over four steps).
Lederman, I.; Choay, J.; Torri, G. Carbo. Res. 1984, 132, C5.
(b) Petitou, M.; Duchaussoy, P.; Lederman, I.; Choay, J.;
Sinay, P.; Jacquinet, J.-C.; Torri G. Carbo. Res. 1986, 147, 221.
5. Petitou, M.; Duchaussoy, P.; Lederman, I.; Choay, J.; Jac-
quinet, J.-C.; Sinay, P.; Torri, G. Carbo. Res. 1987, 167, 67.
6. Torri, G.; Casu, B.; Gatti, G.; Petitou, M.; Choay, J.; Jac-
quinet, J.-C.; Sinay, P. Biochem. Biophys. Res. Commun. 1985,
128, 134.
7. Ferro, D. R.; Provasoli, A.; Ragazzi, M.; Torri, G.; Casu,
B.; Gatti, G.; Jacquinet, J.-C.; Sinay, P.; Petitou, M.; Choay, J.
J. Am. Chem. Soc. 1986, 108, 6773.
8. Ferro, D. R.; Provasoli, A.; Ragazzi, M.; Casu, B.; Torri,
G.; Bossennec, V.; Perly, B.; Sinay, P.; Petitou, M.; Choay, J.
Carbo. Res. 1990, 195, 157.
9. Casu, B.; Choay, J.; Ferro, D. R.; Gatti, G.; Jacquinet, J.-
C.; Petitou, M.; Provasoli, A.; Ragazzi, M.; Sinay, P.; Torri, G.
Nature (London) 1986, 322, 215.
10. Petitou, M.; Duchaussoy, P.; Lederman, I.; Choay, J.;
Sinay, P. Carbo. Res. 1988, 179, 163.
11. (a) van Boeckel, C. A. A.; Beetz, T.; van Aelst, S. F. Tet-
rahedron Lett. 1988, 29, 803. (b). Petitou, M.; Jaurand, G.;
Derrien, M.; Duchaussoy, P.; Choay, J. Bioorg. Med. Chem.
Lett. 1991, 1, 95.
12. For a review see: van Boeckel, C. A. A.; Petitou, M.
Angew. Chem. Int. Ed. Engl. 1993, 32, 1671.
13. A preliminary report on this work has been disclosed: Lei,
P. S.; Duchaussoy, P.; Petitou, M.; Sinay, P. Abstr. XVIIth Int.
Carbo. Symp., Ottawa 1994, p 378.
14. (a) Binkley, R. W.; Hehemann, D. G. J. Org. Chem. 1978,
43, 3244. (b) Kunz, H.; Schmidt, P. Tetrahedron Lett., 1979, 20,
2123. (c) Garegg, P. J.; Samuelsson, B. J. Chem. Soc. Perkin I
1980, 2866.
15. (a) Cerny, M.; Pacak, J. Chem. Listy 1955, 49, 1848. (b)
Hedgley, E. J.; Overend, W. G.; Rennie, R. A. C. J. Chem. Soc.
1963, 4701.
16. Gillard, F.; Heissler, D.; Riehl, J.-J. J. Chem. Soc. Perkin I
1988, 2291.
17. (a) Angyal, S. J.; Pickles, V. A. Austr. J. Chem. 1972, 25,
1695. (b) ibid. 1711.
18. van Boeckel, C. A. A.; Beetz, T.; Vos, J. N.; de Jong, A. J.
M.; van Aelst, S. F.; van den Bosch, R. H.; Mertens, J. M. R.;
van der Vlugt, F. A. J. Carbo. Chem. 1985, 4, 293.
1
[a]d +63 (c 0.2, H2O). H NMR: see Table 2. MS-ESI,
negative mode: monoisotopic mass: 1710.7; chemical
mass: 1712,09; experimental mass: 1711.6.
19. Petrakova, E.; Glaudemans, C. P. J. Carbo. Res. 1996, 284,
191.
20. Sakairi, N.; Kuzuhara H. Tetrahedron Lett. 1982, 23, 5327.
21. Gallagher, T. F.; Horton, D. Carbo. Res. 1893, 116, 227.
22. Sato, K.-i.; Kubo, N.; Takada, R.; Aqeel, A.; Hashimoto,
H.; Yoshimura, J. Chem. Lett. 1988, 1703.
References and Notes
1. Casu, B. Adv. Carbo. Chem. Biochem. 1985, 13, 51.
23. Greene, T. W.; Wuts, P. G. M. In Protective Groups in
Organic Synthesis; Wiley & Sons: New York, 1991; pp 60±61.
2. Rosenberg, R. D.; Damus, P. S. J. Biol. Chem. 1973, 248,
6490.
24. (a) Carlsen, H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K.
B. J. Org. Chem. 1981, 46, 3936. (b) Schuda, P. F.; Cichowicz,
M. B.; Heimann, M. R. Tetrahedron Lett. 1983, 24, 3829.
3. (a) Choay, J.; Lormeau, J.-C.; Petitou, M.; Sinay, P.;
Fareed, J. Ann. N.Y. Acad. Sci. 1981, 370, 644. (b) Thunberg,
L.; Backstrom, G.; Lindahl, U. Carbo. Res. 1982, 100, 393. (c)
Choay, J.; Petitou, M.; Lormeau, J.-C.; Sinay, P.; Casu, B.;
Gatti, G. Biochem. Biophys. Res. Commun. 1983, 116, 492.
25. Kovac, P.; Whittaker, N. F.; Glaudemans, C. P. J. J.
Carbo. Chem. 1985, 4, 243.
26. Jacquinet, J.-C.; Petitou, M.; Duchaussoy, P.; Lederman,
I.; Choay, J.; Torri, G.; Sinay, P. Carbo. Res. 1984, 130, 221.
4. (a) Sinay, P.; Jacquinet, J.-C.; Petitou, M.; Duchaussoy, P.;

1346
P.-S. Lei et al./Bioorg. Med. Chem. 6 (1998) 1337±1346
27. Pentasaccharide (12 mg) were dissolved in 0.5 mL of D2O
1
30. Bodenhausen, G.; Kogler, H.; Ernst, R. R. J. Magn. Reson.
1984, 58, 370.
(99.96%, C.E.A., Saclay France). H NMR spectra (500 MHz)
were obtained with a Bruker DRX 500 spectrometer, at 300 K.
Typically 32 K data points were used, with a spectral width of
6000 Hz. Resolution enhancement of free induction decay
(FID), was performed by the Lorentzien±Gaussian method.
31. Atha, D. H.; Lormeau, J.-C.; Petitou, M.; Rosenberg, R.
D.; Choay, J. Biochemistry 1987, 26, 6454.
32. Olson, S. T.; Bjork, I.; She€er, R.; Craig, P. A.; Shore, J.
D.; Choay, J. J. Biol. Chem. 1992, 267, 12528.
28. Marion, D.; Wuthrich, K. Biochem. Biophys. Res. Com-
mun. 1983, 113, 967.
33. Sakairi, N.; Basten, J. E. M. ; van der Marel, G. A.; van
Boeckel, C. A. A.; van Boom, J. H. Chem. Eur. J. 1996, 2,
1007.
29. Ragazzi, M.; Ferro, D. R.; Perly, B.; Sinay, P.; Petitou, M.;
Choay, J. Carbo. Res. 1990, 195, 169.