Conformational study of synthetic Δ4-uronate monosaccharides and glycosaminoglycan-derived disaccharides

Article abstract of DOI:10.1016/S0008-6215(98)00118-9

Sixteen Δ⁴-uronate monosaccharides were chemically synthesized. Their carboxy group was protected as a methyl or benzyl ester, the anomeric hydroxyl group as a benzyl glycoside and the 2 and 3 hydroxyl groups were protected with different substitution patterns as both ester and ether derivatives. Disaccharides containing Δ4-uronates were prepared from heparin using heparin lyases. Their carboxy group was unprotected or protected as a benzyl ester and the two hydroxyls in the uronate moiety were free, as O-sulfo derivatives or acylated. The conformation of these unsaturated uronate monosaccharide and disaccharide residues was studied using ¹H NMR by examining interproton vic inal coupling constants. The Δ⁴-uronate residue adopted either the ²H₁ or the ¹H₂ conformations. The equilibrium between these two conformers was shown to be controlled by substitution pattern.

Full text of DOI:10.1016/S0008-6215(98)00118-9

Carbohydrate Research 309 (1998) 135±144
Conformational study of synthetic
Á4-uronate monosaccharides and
glycosaminoglycan-derived disaccharides
Helene G. Bazin, Ishan Capila, Robert J. Linhardt*
Division of Medicinal and Natural Products Chemistry and Department of Chemical and Biochemical
Engineering, The University of Iowa, PHAR-S328, Iowa City, IA 52242, USA
Received 12 January 1998; accepted 23 April 1998
Abstract
Sixteen Á4-uronate monosaccharides were chemically synthesized. Their carboxy group was
protected as a methyl or benzyl ester, the anomeric hydroxyl group as a benzyl glycoside and the 2
and 3 hydroxyl groups were protected with di€erent substitution patterns as both ester and ether
derivatives. Disaccharides containing Á4-uronates were prepared from heparin using heparin lya-
ses. Their carboxy group was unprotected or protected as a benzyl ester and the two hydroxyls in
the uronate moiety were free, as O-sulfo derivatives or acylated. The conformation of these unsa-
1
turated uronate monosaccharide and disaccharide residues was studied using H NMR by exam-
ining interproton vicinal coupling constants. The Á4-uronate residue adopted either the 2H1 or the
1H2 conformations. The equilibrium between these two conformers was shown to be controlled by
substitution pattern. # 1998 Elsevier Science Ltd. All rights reserved
Keywords: Á4-uronates; Unsaturated saccharide residue; Synthesis of Á4-uronates; Conformational analysis;
NMR
1. Introduction
to GAG-protein binding sites, are the targets of
our synthetic program. Polysaccharide lyases break
down GAGs into oligosaccharides that contain a
non-reducing terminal Á4-uronic acid residue [2].
Our laboratory is exploring the use of these oligo-
saccharides as building blocks for the synthesis of
larger GAG oligosaccharides. This approach
requires the stereochemically controlled conversion
of the non-reducing terminal Á4-uronic acid resi-
dues to either d-glucopyranosiduronic (GlcAp) or
l-idopyranosiduronic (l-IdoAp) acids. A number
of derivatives of Á4-uronate have been synthesized
[3] that might prove useful as model compounds
for developing a general strategy for the regio and
Glycosaminoglycans (GAGs) regulate a number
of important biological events through their inter-
action with diverse proteins [1]. Synthetic oligo-
saccharide sequences of GAGs, which correspond
* Corresponding author. Fax: 319-335-6634; e-mail:
robert-linhardt@uiowa.edu
Abbreviations: GAGs, Glycosaminoglycans; GlcAp,
Glucopyranosiduronic acid; IdoAp, idopyranosiduronic acid;
DBU, 1,8-diazobiaylco[5,4,0]undec-7-ene; TBDMSCl, tert-
butyldimethylsilyl chloride; TLC, thin layer chromatography;
CE, capillary electrophoresis; DMF, dimethylformamide;
THF, tetrahydrofuran; FABMS, fast atom bombardment
mass spectrometry.
0008-6215/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII S0008-6215(98)00118-9

136
H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
stereoselective conversion of the Á4-uronic acid
residues of oligosaccharides into d-GlcAp or l-
IdoAp. In addition, Á4-uronate disaccharides
derived from GAGs have also been synthesized
[4,5].
equilibrium between the 1H2 and the 2H1 con-
formers (Fig. 1).
1
Here, the results of H NMR studies on chemi-
cally modi®ed Á4-uronates of a number of mono-
saccharides and GAG-derived disaccharides are
presented. The results of such a conformational
analysis should aid in understanding the unusual
chemistry and stereoselectivity associated with
reactions occurring at such Á4-uronates [3].
While 2,3-unsaturated pyranosides have been
extensively studied, the 4,5-unsaturated pyranoside
derivatives, major products of ꢀ-elimination reac-
tions of uronic acid containing polysaccharides,
have received less attention. Over 20 years ago,
several 4,5-unsaturated pyranosides were studied
by 1H NMR spectroscopy [6]. More recently,
Ragazzi et al. [7] reported a compilation of
1H NMR data for unsaturated acid residues of
11 GAG-derived disaccharides and oligosacchar-
ides showing that variation in interproton vicinal
coupling constants as a function of con®g-
uration, glycosidic linkage and substitution can
be unequivocally interpreted in terms of an
2. Results and discussion
Synthesis of Á4-uronates.ÐThe di€erent mono
and disaccharides synthesized are presented in
Fig. 2. Compounds 1±3, and 13 were synthesized
from the known methyl (benzyl 2,3,4-tri-O-acetyl-
ꢀ-d-glucopyranosid)uronate 1a [8], as shown in
Fig. 3. The ꢀ-elimination of 1a or 1b, using
Fig. 1. Half-chair conformations of Á4 uronates.
Fig. 2. The structure of Á4-monosaccharides and Á4-disaccharides.

H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
137
Fig. 3. Synthesis of monosaccharides 1±3 and 13. (a) NaOMe, MeOH, rt; (b) Piv-C1, Py, rt, 12 h (65±70 % from 1); (c) DBU,
CH2Cl2, rt, 7±30 h (90 %); (d) Ag2O, BnBr, DMF, rt, 48 h (40 %), (e) NEt3, MeOH, rt, 12 h (80 %).
1,8-diazabicyclo[5,4,0]undec-7-ene (DBU) [9], gave
the corresponding Á4-uronates 2 and 3 in high
yields (91%). Deacetylation of 2 using triethyla-
mine in methanol led to the free diol 1 in 87%
yield. Deacetylation [10] of 1a followed by a one
pot benzylation and ꢀ-elimination, using silver
oxide, a€orded the perbenzylated glycal 13 in 40%
yield. All the other monosaccharides were prepared
using standard methods from the diol 1. Com-
pound 4 was obtained in quantitative yield by
benzoylation of 1 using benzoyl chloride and pyr-
idine. The 2,3-di-O-silyl derivatives 5 and 6 were
synthesized by reacting 1 with respectively tert-
butyldimethylsilyl chloride (TBDMSCl) in pre-
sence of imidazole (quant.), or 1,3-dichloro-1,1,3,3-
tetraisopropyl-1,3-disiloxane in pyridine (88%
yield). Silylation of 1 with 1 equivalent of
TBDMSCl in pyridine gave a mixture of the 3-O-
tert-butyldimethylsilyl derivative 7 and 2-O-tert-
butyldimethylsilyl derivative in 82 and 5% yield,
respectively. Pivaloylation of 1 using 1 equivalent
of pivaloyl chloride gave a mixture of 2-O-pivaloyl
uronate 8, 3-O-pivaloyl uronate 9, and 2,3-di-O-
pivaloyl uronate 3 in a yield of 4, 66 and 14%,
respectively. Benzoylation of the 3-O-tert-butyldi-
methylsilyl derivative 7 followed by desilylation
using tetrabutylammonium ¯uoride led to a mix-
ture of 2-O-benzoyl derivative, 3-O-benzoyl deri-
vative 10, and 2,3-di-O-benzoyl derivative 4 in a
yield of 34, 37 and 5%, respectively. Phosphoryla-
tion [11] of 9 using triethylphosphite and iodide
a€orded 11 quantitatively. Direct benzylation of 7
using sodium hydride and benzyl chloride was
expected to result in transesteri®cation, giving a
mixture of methyl and benzyl esters. To eliminate
this potential problem, 7 was ®rst de-esteri®ed
using methanolic potassium hydroxide. The corre-
sponding carboxylic acid was then benzylated
under standard conditions to give the correspond-
ing benzyl ester 14 in 56% yield, and the perben-
zylated compound 13 in 26% yield.
Transesteri®cation of 14 using sodium methoxide,
followed by desilylation and subsequent pivaloyla-
tion led to 12 in 70% yield. Desilylation of 14 fol-
lowed by benzoylation or acidic benzylation [12]
gave respectively 15 and 16 in yields of 90 and
75%, respectively.
The disaccharides 17, 18, 20 and 21 had been
previously prepared in our laboratory [4,5].
Anomeric deacetylation of 18 using piperidine
resulted in a 73% yield of 19. The desulfated dis-
accharide 22 was obtained by enzymatic depoly-
merization of N-acetylated, O-desulfated heparin
[13], and was peracetylated under standard condi-
tions to give 23 in 80% yield.
1H NMR study.ÐVicinal coupling constants are
useful in determining the conformation of unsatu-
1
rated substances using H NMR. The chemical
shifts and coupling constants of the compounds 1±
23 are reported in Table 1. For each derivative, the
signal for H-4 appears at lowest ®eld (5.9±6.5 ppm)
due to deshielding by the double bond, and is
either a well resolved doublet, or a doublet of
doublets, due to long-range coupling with H-2. The
signal for H-1 is a doublet, or a doublet of doublets
due to the long-range coupling with H-3. The sig-
nals for H-2 and H-3 are a doublet of doublets or a
triplet, which is split to multiplet when a long-
range coupling is present. In some cases, H-2
appears as a broad singlet, making the precise
determination of J2;3 dicult. However, these
undetermined values can be estimated as being
very small (<3.0 Hz). Based on the values of J1;2
and J2;3, three di€erent groups of monosaccharides
can be distinguished.
The ®rst group includes monosaccharides 6, 7
and 14. These have high J1;2 and J2;3 values, vary-
ing from 6.3 (14) to 8.1 Hz (6) and from 5.7 (14) to

138
H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
Table 1
Chemical shifts and coupling constants of 1-23
Compounds
Chemical shifts (ppm)
Coupling constants (Hz)
H1
H2
H3
H4
J1,2
J2,3
J3,4
J1,3
J2,4
1
2
3
4
5
6
7
8
5.3(d)
5.3(dd)
5.3(dd)
5.6(dd)
5.2(d)
4.9(d)
4.9(d)
5.1(m)
5.1(d)
5.2(d)
5.5(d)
5.3(d)
5.1(d)
5.1(d)
5.3(d)
5.1(d)
5.5(dd)
5.6(d)
5.5(d)
5.5(d)
5.6(d)
4.9(d)
5.3(dd)
4.2(m)
5.2(m)
5.1(m)
5.6(m)
3.3(dt)
3.9(t)
3.7(dt)
5.5(m)
4.0(t)
4.2(m)
4.6(ddd)
3.8(t)
3.8(t)
3.6(t)
3.9(t)
3.8(t)
4.6(m)
4.7(bs)
4.6(m)
4.7(t)
4.6(t)
3.6(dd)
4.9(m)
3.9(m)
5.2(dd)
5.2(ddd)
5.6(ddd)
4.4(t)
4.5(dd)
4.4(dd)
3.9(m)
5.3(t)
6.3(dd)
6.2(dd)
6.2(dd)
6.4(dd)
6.5(dd)
5.9(d)
5.9(d)
6.4(dd)
6.1(d)
6.2(d)
6.2(dd)
6.1(d)
6.2(d)
6.0(d)
6.3(d)
6.2(d)
6.0(dd)
5.9(d)
5.9(d)
5.9(d)
6.3(d)
5.7(d)
6.1(d)
2.6
2.8
3.2
2.2
4.3
8.1
7.3
1.6
5.4
4.9
2.4
3.4
5.3
6.3
3.4
5.5
3.3
1.4
2.5
2.9
3.0
7.2
1.9
nd
2.1
2.5
1.7
3.7
7.3
6.6
nd
4.6
4.1
1.3
3.2
5.0
5.7
3.5
4.1
3.4
nd
4.6
4.6
4.2
4.7
4.2
2.5
2.9
4.7
3.8
4.0
4.8
4.6
3.7
3.2
4.4
3.7
4.3
5.0
5.0
4.4
4.5
3.0
4.8
<1
<1
1.0
1.2
0.6
1.2
<1
<1
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
5.5(t)
5.2(dd)
5.3(t)
4.1(t)
4.4(dd)
5.6(t)
4.1(t)
4.4(t)
5.5(m)
5.3(t)
5.4(m)
4.4(m)
4.1(dd)
5.2(m)
1.3
1.0
1.0
2.0
2.6
2.5
6.5
nd
7.3 Hz (6), respectively. These coupling constants
are consistent with axial or quasi-axial H-1, H-2
and H-3 protons. The small J3;4 constants for these
compounds, 2.5±3.2 Hz, also indicate the quasi-
axial orientation of H-3, characteristic for a torsion
angle between the allylic proton and the alkene of
approximately 90ꢀ [14]. These results suggest that 6
result of strong interactions between functional
groups.
The coupling constants J2;3 and J3;4 were plotted
as a function of the coupling constant J1;2 (Fig. 4).
Inspection of this ®gure shows a linear relationship
in both plots described by the equations
2
J2;3 ˆ b ‡ a ꢀ J1;2 and J3;4 ˆ d ‡ c ꢀ J1;2
solely adopts the H1 conformation while 7 and 14
primarily adopt the 2H1 conformation.
The second group includes monosaccharides 1±
4, 8 and 11. These have low values for J1;2 (1.6±
3.2 Hz) and J2;3 (1.3±2.5 Hz) characteristic of the
equatorial-quasi-equatorial orientation of the pro-
1
tons observed in the H2 conformation. Moreover,
the corresponding values of J3;4 of 4.3±4.8 Hz is
also characteristic of the quasi-equatorial orienta-
tion of H-3. Long-range couplings (J1;3 and J2;4) in
these compounds suggest a planar ``W'' arrange-
1
ment, consistent with the H2 conformation.
The third group contains 5, 9, 10, 12, 13, 15 and
16. Their J2;3 values are higher than expected
between the equatorial and the quasi-equatorial
protons H-2 and H-3. These data, together with
the absence of long-range coupling indicate that
1
either the H2 conformation may be considerably
deformed or the equilibria between the 1H2 and the
Fig. 4. Plot of J2;3 (*) and J3;4 (&) in Hz as a function of J1;2
for monosaccharides 1±16 and J2;3 (*) and J3;4 (&) in Hz as a
function of J1;2 for disaccharides 17±23.
2
2H1 may signi®cantly be shifted towards H1, as a

H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
139
Table 2
Theoretical coupling constants of 1±23
Compounds
J values calculated from
Haasnoot et al. Eq 15 (Hz)
Extrapolated theoretical J
1H2 population %
1H2 2H1
1H2
2H1
J1,2
J2,3
J1,2
J2,3
J1,2 J2,3 J3,4 J1,2 J2,3 J3,4
1
2
3
4
5
6
7
8
3.1
3.1
2.4
2.9
3.3
3.0
3.1
2.9
3.1
3.2
3.1
3.0
3.0
2.9
3.0
3.1
3.2
3.3
3.3
3.2
3.2
3.1
3.1
6.1
1.7
0.8
1.6
2.4
1.3
1.6
1.6
1.8
1.9
2.4
1.9
2.3
2.1
2.1
2.4
1.9
2.0
2.0
1.5
1.9
1.9
1.7
8.0
8.1
8.0
8.0
7.5
7.9
8.0
8.1
7.9
7.9
7.9
7.9
7.9
8.0
7.8
7.9
8.1
8.1
8.1
8.1
8.1
8.0
8.1
8.4
8.6
8.8
8.6
6.3
7.5
8.4
8.4
8.6
8.5
8.1
8.2
8.6
7.6
7.4
8.6
9.0
9.3
9.3
9.3
8.9
9.4
9.3
2.9
2.7
2.0
2.7
3.2
2.6
2.7
3.0
2.8
2.9
3.0
2.9
2.9
2.9
2.9
3.0
2.9
3.0
3.0
2.7
2.9
2.9
2.9
2.4
2.3
1.6
2.2
2.7
2.1
2.3
2.5
2.3
2.4
2.6
2.4
2.5
2.4
2.4
2.5
2.4
2.5
2.5
2.3
2.4
2.4
2.4
4.4
4.5
4.7
4.5
4.3
4.5
4.5
4.4
4.5
4.4
4.4
4.5
4.4
4.5
4.4
4.4
4.4
4.4
4.4
4.5
4.4
4.5
4.4
8.5
8.6
8.5
8.3
7.3
8.0
8.5
8.4
8.5
8.3
8.3
8.3
8.4
8.1
7.9
8.6
8.8
8.9
8.9
9.0
8.7
8.5
8.9
7.7
7.8
7.7
7.6
6.6
7.2
7.7
7.6
7.7
7.5
7.5
7.5
7.6
7.3
7.1
7.8
8.0
8.1
8.1
8.1
7.9
7.7
8.1
2.5
2.5
2.5
2.6
2.9
2.7
2.5
2.6
2.5
2.6
2.6
2.6
2.6
2.7
2.7
2.5
2.4
2.4
2.4
2.4
2.5
2.5
2.4
100
100
83
100
75
0
20
100
56
65
100
86
53
34
85
63
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
88
100
100
95
99
22
100
with a ˆ 0:94 Æ 0:05; b ˆ 0:29 Æ 0:26; c ˆ
0:34 Æ 0:04; d ˆ 5:43 Æ 0:16. These values are in
excellent agreement with those obtained for similar
plots for unsaturated disaccharides (aˆ0:98Æ0:07;
bˆ 0:44Æ0:36; cˆ 0:35Æ0:05; d ˆ 5:67 Æ 0:22;)
previously reported by Ragazzi et al. [7]. Molecular
modeling of uronates 1±16 was done using the
package SYBYL (version 6.3) from Tripos Inc., St
Louis, MO. All energy calculations were per-
formed using parameters from the Tripos force-
®eld. Charges were assigned according to the
Gasteiger±Huckel protocol. A distance dependent
dielectric constant ("=1.5) was used to compensate
for the lack of explicit solvation. Molecular
mechanics calculations led to two energy minima,
corresponding to the half-chairs 1H2 and 2H1. These
were taken as the starting points of calculation for
all the other uronate derivatives. Ragazzi et al. [7]
have also shown that for disaccharides containing
similar Á4-uronic acid residues, no other minimum
ring energy conformation is found. These observa-
tions together with the linear relationships descri-
bed in the equations above, con®rm that the
coupling constants of these di€erent compounds
can be interpreted in terms of an equilibrium
between the two conformers, 1H2 and 2H1.
The theoretical J values can be calculated from a
Karplus type equation described by Haasnoot et
al. [15] containing a correction term for ꢁ- and ꢀ-
electronegative substituents. Using this equation
and the calculated dihedral angles H±C±C±H cor-
responding to the two minimum energy conformers
of each uronate, the theoretical coupling constants
J1;2 and J2;3 were calculated (Table 2). These values
are in good agreement with the straight line, best ®t
of the experimental values as shown in Fig. 5. The
Fig. 5. Plot of the theoretical J1;2 and J2;3 values (&) calcu-
lated from the Haasnoot et al. [15] equation.

140
H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
1
Haasnoot et al. [15] equation has not been para-
meterized for unsaturated moieties and thus, the
J3;4 values could not be calculated. These J3;4
values can, however, be estimated by taking the
closest points lying on the experimental lines.
These new sets of theoretical values, determined
from the Haasnoot et al. [15] equation and from
these linear relationships (Table 2) were used for
the calculation of the conformer populations, by
least-square ®tting the experimental coupling con-
stants to the theoretical data. Since the J2;3 values
for compounds 1 and 8 could not be determined
benzoyl uronates 2 and 4 exist solely as the H2
conformer, the 2,3-di-O-pivaloyl derivative 3 and
the 2-O-benzyl-3-O-acyl derivatives 12 and 15,
1
mainly in the H2 conformation (83, 89 and 85%,
respectively). This can be explained by an allylic
e€ect [16], which consists in the stabilization of
quasi-axial polar allylic substituents. Surprisingly,
the 3-O-pivaloyl and 3-O-benzoyl derivatives 9 and
1
10 exist in 56 and 65% in the H2 conformation,
respectively. This observation can be explained by
the presence of intramolecular hydrogen bonding
between the 2-hydroxyl and the 3-O-acyl sub-
stituents, clearly shown by molecular modeling of
these two compounds.
The chemical shifts and coupling constants
obtained for the uronate residues of the dis-
accharides 17±23 are given in Table 1, and the J2;3
1
directly from the corresponding H NMR spectra,
the conformer population for these two uronates
were determined from their J2;3 values estimated
from their J1;2 values and the experimental plot
J3;4 ˆ fꢁJ1;2†. The quantitative results presented in
Table 2 are in good accordance with the previous
qualitative conclusions deduced from the 1H NMR
study. The conformational equilibria were also
determined using J1;2 and J2;3 alone, without
including J3;4. The corresponding least-square ®t-
ting gave populations very close to the ones pre-
viously calculated with a maximum deviation of
3% (except for 22 with 7% deviation).
and J3;4 values are plotted as a function of J1;2
.
(Fig. 4). These experimental values again ®t very
well with the previously described linear equations,
showing that these uronate residues in these
disaccharides exist in an equilibrium between the
1H2 and H1 conformers. For compounds 18±21
and 23, Table 1 shows small J1;2 (1.4±3.0 Hz) and
J2;3 (2.0±2.7 Hz) values, suggesting that these dis-
2
As expected, Á4-uronates 6, 7 and 14, bearing a
bulky electrorepulsive substituent such as a silyl
accharides mainly exist as the H2 conformer. The
J1;2 and J2;3 values of 7.2 and 6.5 Hz, respectively
1
2
2
group at the O-3 position, exist mainly in the H1
for 22 suggest it should primarily adopt the H1
conformation. Finally, 17 having intermediate J1;2
and J2;3 values of 3.3 and 3.4 Hz, should exist as an
conformation, in which all the substituents adopt
the more energetically favored equatorial positions.
Thus, the siloxane uronate 6 exists solely as the 2H1
conformer, and the 3-O-tert-butyldimethylsilyl
1
2
equilibrium between the H2 and H1 conformers.
The population of each conformer of these dis-
accharides was calculated as previously described
for the monosaccharides (Table 2). As expected,
2
uronate 7 mainly as the H1 conformer (80%). A
substituent bulkier than hydrogen at the O-2 posi-
tion shifts the equilibrium to the 1H2 conformer, as
seen for the 2-O-benzyl 14 which exists as 67% 2H1
conformer. This is presumably the result of unfa-
vorable steric interactions between the substituents
at the O-2 and O-3 positions. The introduction of a
second bulky tert-butyldimethylsilyl group (com-
pound 5), clearly ampli®es this e€ect towards the
1
18, 19 and 23 are exclusively in the H2 conforma-
1
tion, and 17, 20 and 21 exist mainly as the H2
conformer (88, 95 and 99%, respectively).
2
Disaccharide 22 adopts mainly the H1 conforma-
tion (77%). For these compounds, the equilibrium
is essentially controlled by electrostatic and hydro-
gen bonding interactions. It should be noted that
the calculated conformer populations of com-
pounds 17 and 23 (12 and 77% in the 1H2
conformer) are in accordance with the percen-
tages reported by Ragazzi et al. [7] (8 and 57%,
respectively).
1
1H2 conformation, H2 becoming the predominant
conformer at 75%. It was expected that the
predominant 2H1 conformation would be observed
for uronates bearing a 3-O-benzyl substituent
such as 13 and 16. However, the presence of a
second benzyl group at the O-2 position in these
two compounds generate both electronic and steric
2
interactions, shifting the equilibrium from H1 to
3. Experimental
1H2. The population of 13 and 16 in the 2H1
conformer was estimated to be only 47 and 37%,
respectively. The 2,3-di-O-acetyl and 2,3-di-O-
General methods.Ð1H NMR spectra were recor-
ded at 25 C, in deuterochloroform on a Varian
ꢀ

H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
141
Unity 500 MHz spectrometer. Chemical shifts were
recorded in ppm (ꢂ) and coupling constants in Hz,
relative to tetramethylsilane as internal standard.
®ltered and evaporated to a€ord a black oil. Flash
chromatography (ethyl acetate±hexane, v/v 1:5)
provided 2 (6.4 g, 91%) as a colorless oil.
[ꢁd]22= 103ꢀ (c 1, CHCl3); FABMS (+ve) m/z
387 [M+Na+]+; Anal. calcd for C18H20O8 (364.4)
C 59.34, H 5.53; Found: C 59.14, H 5.58.
Methyl (benzyl 4-deoxy-2,3-di-O-pivaloyl-a-l-
threo-hex-4-enopyranosid)uronate (3).ÐCompound
1b (14.5 g, 26.3 mmol) was treated as described for
2. Flash chromatography (ethyl acetate±hexanes,
v/v 1:8) a€oꢀrded 3 (6.4 g, 91%) as a colorless oil.
[ꢁd]22= 8.6 (c 2, CHCl3); FABMS (+ve) m/z
471 [M+Na+]+; Anal. calcd for C24H32O8 (448) C
64.27, H 7.19; Found: C 63.91, H 7.30.
1
The H NMR spectra were fully assigned by the
use of single frequency decoupling. Optical rota-
tions were measured with a Perkin±Elmer 141
polarimeter. Thin layer chromatography (TLC)
was performed using Merck plates of silica gel 60
with ¯uorescent indicator. Visualization was e€ec-
ted by spraying plates with Von's reagent [5] fol-
ꢀ
lowed by heating at 140 C. Reactions performed
on sulfated disaccharides were monitored by capil-
lary electrophoresis (CE) [5]. CE was performed on
a Dionex CE system (Sunnyvale, CA) equipped
with a variable wavelength detector. All analyses
used a fused silica capillary tube (75 ꢃm i.d.,
375 ꢃm o.d. and 75 cm long) from Dionex. Oper-
ating bu€er was 10 mM sodium borate, 50 mM
sodium dodecylsulfate adjusted to pH 8.8 with 1 N
HCl. Flash chromatography was conducted with
silica gel (230±430 mesh, Merck). Dichloromethane
(CH2Cl2), dimethylformamide (DMF), methanol,
pyridine and tetrahydrofuran (THF) were anhy-
drous solvents available from Aldrich, Milwaukee,
WI.
Methyl (benzyl 2,3-di-O-benzoyl-4-deoxy-a-l-
threo-hex-4-enopyranosid)uronate (4).ÐTo a solu-
tion of 1 (220 mg, 0.78 mmol) in anhydrous pyr-
ꢀ
idine (8 mL) at 0 C and under argon was added
benzoyl chloride (0.36 mL, 3.14 mmol). After 5 h at
room temperature, the reaction mixture was quen-
ched with MeOH (3 mL) and concentrated under
reduced pressure. Puri®cation of the resulting oil
by ¯ash chromatography (ethyl acetate±hexanes, v/
v 1:10) gave 4 (380 mg, quant.) as white needles;
[ꢁd]22=25.0ꢀ (c 0.1, CHCl3); Anal. calcd for
C28H24O8 (488.5) C 68.95, H 4.95; Found: C 68.73,
H 5.17.
Methyl (benzyl 4-deoxy-a-l-threo-hex-4-enopyr-
anosid)uronate (1).ÐA solution of
2 (5.0 g,
13.7 mmol) in anhydrous methanol (150 mL) con-
taining triethylamine (0.5 mL, 3.6 mmol) was stir-
red at room temperature under argon for 36 h. The
reaction mixture was concentrated under reduced
pressure and subjected to ¯ash chromatography
(acetone±hexane, v/v 1:4) to give 1 (3.4 g, 87%) as
a white amorphous solid. A portion of the pure
product was crystallized from ethanol to a€ord
white needles; mp 139±140; [ꢁd]22= 106ꢀ (ca 1,
CHCl3): Lit [17]. mp 139±140 (ethanol);
[ꢁd]22= 94.5ꢀ (c 0.5, CH3OH); FABMS (+ve)
m/z 303 [M+Na+]+; Anal. calcd for C14H16O6
(280.3) C 60.00, H 5.75; Found: C 59.76, H 5.95.
Methyl (benzyl 2,3-di-O-acetyl-4-deoxy-a-l-threo-
hex-4-enopyranosid)uronate (2).ÐDBU (3.6 mL,
24.1 mmol) was added to a solution of 1a [6] (8.2 g,
Methyl (benzyl-2,3-di-O-t-butyldimethylsilyl-4-
deoxy-a-l-threo-hex-4-enopyranosid)uronate (5).Ð
To a solution of 1 (1.0 g, 3.57 mmol) in anhydrous
DMF (10 mL) at 0 ꢀC and under argon were added
imidazole (972 mg, 14.27 mmol) and TBDMSCl
(2.15 g, 14.27 mmol). After 12 h at room tempera-
ture, the reaction mixture was poured into water
(10 mL). The aqueous layer was extracted 2 times
with CHCl3 (10 mL). The combined organic layers
were washed with water, dried over anhydrous
Na2SO4, ®ltered and concentrated under reduced
pressure to a€ord a pale yellow oil. Puri®cation by
¯ash chromatography (ethyl acetate±hexane, v/v
1:15) provided 5 (1.79 g, quant.) as a colorless oil.
[ꢁd]22=29.0ꢀ (c 0.1, CHCl3); Anal. calcd for
C26H44O6Si2 (508.8) C 61.38, H 8.72, Found: C
61.89, H8.90.
ꢀ
19.3 mmol) in anhydrous CH2Cl2 (100 mL) at 0 C
and under argon. After 10 min the ice bath was
removed and stirring continued for 7 h at room
temperature. The solution was cooled with an ice
bath and quenched with 50 mL cold saturated aq.
NH4Cl. The mixture was diluted with 150 mL
CH2Cl2 and extracted with an additional 150 mL
saturated aq. NH4Cl, then saturated aq. NaCl. The
organic layer was dried over anhydrous Na2SO4,
Methyl (benzyl 4-deoxy-2,3-tetraisopropyldisil-
oxane-a-l-threo-hex-4-enopyranosid)uronate (6).Ð
To a solution of 1 (100 mg, 0.36 mmol) in anhy-
drous pyridine (5 mL) at 0 ꢀC and under argon was
added 1,3-dichloro-1,1,3,3-tetraisopropyl-1,3-disil-
oxane (150 ꢃL, 0.46 mmol). After 12 h at room
temperature, the reaction mixture was poured into
water (5 mL). The aqueous layer was extracted 2

142
H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
times with ethyl acetate (5 mL). The combined
organic layers were washed with water, dried over
anhydrous Na2SO4, ®ltered and concentrated
under reduced pressure to give a pale yellow oil.
Puri®cation by ¯ash chromatography (ethyl acet-
ate±hexane, v/v 1:30) a€orded 6 (164 mg, 88%) as
a colorless oil. [ꢁd]22=18.0ꢀ (c 0.1, CHCl3); Anal.
calcd for C26H42O7Si2 (522.8) C 59.73, H 8.10;
Found: C 59.78, H 8.34.
Methyl (benzyl-3-O-t-butyldimethylsilyl-4-deoxy-
a-l-threo-hex-4-enopyranosid) uronate (7).ÐTo a
solution of 1 (500 mg, ꢀ1.78 mmol) in anhydrous
pyridine (10 mL) at 0 C and under argon was
added TBDMSCl (269 mg, 1.78 mmol) and the
reaction mixture was allowed to warm to room
temperature. Every 12 h an additional equivalent
of TBDMSCl was added until the disappearance of
1. The reaction mixture was poured into water
(10 mL). The aqueous layer was extracted 2 times
with chloroform (10 mL). The combined organic
layers were washed with water, dried over anhy-
drous Na2SO4, ®ltered and concentrated under
reduced pressure to give a pale yellow oil. Pur-
i®cation by ¯ash chromatography (ethyl acetate±
hexane, v/v 1:10) gave 7 (575 mg, 82%) and methyl
Methyl (benzyl 3-O-benzoyl-4-deoxy-a-l-threo-
hex-4-enopyranosid)uronate (10).ÐTo a solution
of 7 (137 mg, 0.35 mmol) in anhydrous pyridine
ꢀ
(10 mL) at 0 C and under argon was added ben-
zoyl chloride (79 ꢃL, 0.69 mmol). After 2 h at room
temperature, the reaction mixture was quenched
with MeOH (5 mL) and concentrated under
reduced pressure. The residue was dissolved under
argon in anhydrous THF (10 mL) and reacted at
room temperature with tetrabutylammonium
¯uoride (0.42 mL, 0.42 mmol). After 2 h at room
temperature, the reaction mixture was con-
centrated under reduced pressure. Puri®cation of
the resulting oil by ¯ash chromatography (ethyl
acetate±hexane, v/v 1:20) gave the methyl (benzyl
2-O-benzoyl-4-deoxy-ꢁ-l-threo-hex-4-enopyranosid)-
uronate (36 mg, 34%), 10 (39 mg, 37%) and the
dibenzoyl uronate 4 (7 mg, 5%) as colorless oils.
10: [ꢁd]22=34.0ꢀ (c 0.1, CHCl3); Anal. calcd for
C21H20O7 (384.4) C 65.62, H 5.24; Found: C 65.56,
H 5.63.
Methyl (benzyl 2-O-diethylphosphate-3-O-pivaloyl-
uronate
4-deoxy-a-l-threo-hex-4-enopyranoside)
(11).ÐIodide (265 mg, 1.04 mmol) was added to a
solution of triethylphosphate (0.2 mL, 1.14ꢀmmol)
in anhydrous dichloromethane (5 mL) at 0 C and
under argon. After 5 min, this solution was added
to a solution of 9 (345 mg, 0.95 mmol) in anhy-
drous dichloromethane (10 mL) and pyridine
(benzyl
4-deoxy-2-O-tert-butyldimethylsilyl-ꢁ-l-
threo-hex-4-enopyranosid) uronate (35 mg, 5%) as
colorless oils. [ꢁd]22=28.0ꢀ (c 0.1, CHCl3); Anal.
calcd for C20H30O6Si (394.5) C 60.89, H 7.66;
Found: C 61.28, H 7.66.
(0.31 mL, 3.8 mmol) cooled at
40 ꢀC. After
Methyl (benzyl 4-deoxy-2-O-pivaloyl-a-l-threo-
hex-4-enopyranosid)uronate (8) and methyl (benzyl
4-deoxy-3-O-pivaloyl-a-l-threo-hex-4-enopyranosid)-
uronate (9).ÐTo a solution of 1 (1.0ꢀg, 3.57 mmol)
in anhydrous pyridine (8 mL) at 0 C and under
argon was added pivaloyl chloride (0.48 mL,
3.92 mmol). After 12 h at room temperature, the
reaction mixture was quenched with water (5 mL)
and extracted 2 times with chloroform (10 mL).
The combined organic layers were washed with
water, dried over anhydrous Na2SO4, ®ltered and
concentrated under reduced pressure to give a pale
yellow oil. Puri®cation by ¯ash chromatography
(ethyl acetate±hexane, v/v 1:6) yielded the dipiva-
loyl uronate 3 (221 mg, 14%), the 2-O-pivaloyl
uronate 8 (48 mg, 4%) and the 3-O-pivaloyl uro-
nate 9 (857 mg, 66%) as colorless oils. 8:
[ꢁd]22=49.0ꢀ (c 0.1, CHCl3); Anal. calcd for
C19H24O7 (364.4) C 62.63, H 6.64; Found: C 62.25,
H 6.65. 9: [ꢁd]22= 22.0ꢀ (c 1, CHCl3); Anal. calcd
for C19H24O7 (364.4) C 62.63, H 6.64; Found: C
62.94, H 6.59.
30 min at 40 ꢀC, the reaction mixture was allowed
to warm at room temperature, diluted with
dichloromethane (10 mL), washed with saturated
aq. NaHSO4, then with phosphate bu€er (pH 7),
dried over anhydrous Na2SO4, ®ltered and con-
centrated under reduced pressure. The resulting oil
was puri®ed by ¯ash chromatography (ethyl
acetate±hexane, v/v 1:3) to give 11 (472 mg, quant.)
as a colorless oil. [ꢁd]22=8.0ꢀ (c 0.1, CHCl3); Anal.
calcd for C23H33O10P (500.5) C 55.20, H 6.65;
Found: C55.08, H 6.69.
Methyl (benzyl 2-O-benzyl-4-deoxy-3-O-pivaloyl-
a-l-threo-hex-4-enopyranosid) uronate (12).Ð14
(73.8 mg, 0.13 mmol) in methanol (5 mL) under
argon was transesteri®ed at room temperature with
sodium methoxide (0.13 mmol). After 24 h the
reaction mixture was neutralized with resin
Amberlyst IR 120 (H+), ®ltered and concentrated
under reduced pressure. The resulting residue was
dissolved under argon in anhydrous THF (5 mL)
and reacted at room temperature with tetra-
butylammonium ¯uoride (0.16 mL, 0.16 mmol).

H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
143
After 2 h, the reaction mixture was concentrated
under reduced pressure. The resulting residue was
dissolved in anhydrous pyridine (5 mL) under
argon and acylated by addition of pivaloyl chloride
(24 ꢃL, 0.20 mmol). After 12 h at room tempera-
ture, the reaction mixture was quenched with water
(2 mL) and extracted with chloroform. The com-
bined organic layers were washed with water, dried
over anhydrous Na2SO4, ®ltered and concentrated
under reduced pressure. Puri®cation by ¯ash chro-
matography (ethyl acetate±hexane, v/v 1:15) a€or-
ded 12 (41 mg, 70%) as a colorless oil. Anal. calcd
for C26H30O7 (454.5) C 68.71, H 6.65; Found: C
69.12, H 6.99.
water, dried over anhydrous Na2SO4, ®ltered and
concentrated under reduced pressure to give a pale
yellow oil. Puri®cation by ¯ash chromatography
(ethyl acetate±hexane, v/v 1:30) yielded 14 (458 mg,
56%) and 13 (204 mg, 26%) as colorless oils.
Benzyl (benzyl 2-O-benzyl-3-O-benzoyl-4-deoxy-
a-l-threo-hex-4-enopyranosid) uronate (15).Ð14
(117 mg, 0.21 mmol) was dissolved under argon in
dry THF (5 mL) and desilylated by addition of
tetrabutylammonium ¯uoride (1M in THF,
0.23 mL, 0.23 mmol). After 2 h at room tempera-
ture, the reaction mixture was concentrated under
reduced pressure. The residue was dissolved in dry
pyridine (5 mL) and acylated by addition of ben-
zoyl chloride (47 ꢃl, 0.42 mmol). After 12 h at room
temperature, the reaction mixture was quenched
with water and extracted with chloroform. The
combined organic layers were washed with water,
dried over anhydrous Na2SO4, ®ltered and con-
centrated under reduced pressure. Puri®cation by
¯ash chromatography (ethyl acetate±hexane, v/v
1:15) a€orded 15 (114 mg, 90%). Anal. calcd for
C34H30O7 (550.6) C 74.17, H 5.49. Found: C 74.37,
H 6.11.
Benzyl (benzyl 2-O-benzyl-4-deoxy-3-O-p-meth-
oxybenzyl-a-l-threo-hex-4-enopyranosid)uronate
(16).ÐTo a solution of 14 (90 mg, 0,20 mmol) in
anhydrous dichloromethane (5 mL) under argon
was added p-methoxybenzyltrichloroacetamidate
[13] (0.60 mmol) and a catalytic amount of p-
toluenesulfonic acid. After 20 h at room tempera-
ture, the reaction mixture was neutralized with
triethylamine and concentrated under reduced
pressure. Puri®cation of the resulting oil by ¯ash
chromatography (ethyl acetate±hexane, v/v 1:15)
a€orded 16 (85 mg, 75%) as a colorless oil.
Benzyl (benzyl 2,3-di-O-benzyl-4-deoxy-a-l-threo-
hex-4-enopyranosid) uronate (13).ÐSodium meth-
oxide (380 mg, 7 mmol) was added to a vigorously
stirred suspension of 1a (13.2 g, 21.2 mmol) in
methanol (100 mL) under argon with ice bath
cooling. After 22 h at room temperature, the reac-
ꢀ
tion mixture was cooled to 0 C, neutralized by
adding resin Amberlyst IR 120 (H+), ®ltered and
concentrated under reduced pressure. Flash chro-
matography (chloroform±methanol, v/v 13:1)
a€orded the corresponding methyl (benzyl ꢀ-d-
glucopyranosid)uronate (8.4 g, 90%) as a white
amorphous solid. To a solution of the previous
compound (0.18 g, 0.60 mmol) in dry DMF (4 mL)
under argon and cooled with an ice-water bath
were added benzyl bromide (0.64 mL, 5.4 mmol),
Ag2O (1.4 g, 6.0 mmol) and crushed 4 A molecular
sieves. After 48 h at room temperature, the reaction
mixture was diluted with dichloromethane (10 mL),
®ltered through a pad of celite and concentrated
under reduced pressure. Puri®cation by chromato-
graphy on silica gel (ethyl acetate±hexane, v/v 1:17)
gave 13 (0.13 g, 40%). [ꢁd]22= 30ꢀ (c 1, CHCl3).
Benzyl (benzyl 2-O-benzyl-3-O-tert-butyldimethyl-
silyl-4-deoxy-a-l-threo-hex-4-enopyranosid)uronate
(14).Ð7 (575 mg, 1.46 mmol) was reacted 2 h at
50 ꢀC with methanolic KOH 0.5 M (4 mL,
1.66 mmol). After disappearance of 7, the reaction
mixture was neutralized with resin Amberlyst IR
120 (H+), ®ltered and concentrated under reduced
pressure. The residue was dissolved in anhydrous
DMF (5 mL) under argon and reacted 30 min with
NaH (84 mg, 7.35 mmol). Benzyl chloride
(0.51 mL, 4.37 mmol) was added to the reaction
mixture. After 48 h at room temperature, the reac-
3-O-acetyl-4-deoxy-2-sulfate-a-l-threo-hex-4-eno-
pyranosyluronic
acid-(1!4)-3-O-acetyl-2-deoxy-
sulfamido-6-O-sulfo-a-d-glucopyranose, tetrasodium
salt (19).ÐThe tetrasodium salt of disaccharide 18
(90 mg, 0.11 mmol) was suspended with stirring,
under argon, in DMF (3 mL) and formamide
ꢀ
(200 ꢃL). At 4 C, piperidine (0.55 mL, 5.5 mmol)
was added and the reaction stirred at RT until CE
analysis showed the disappearance of starting
material. After 17 h, the solution was diluted with
2 mL of water and passed through a column
(1Â10 cm) of Dowex (H+) followed by a column
(1Â10 cm) of Dowex (Na+). Piperidine was still
present in the sample, indicating incomplete salt
exchange had occurred. This solution was adjusted
at pH 6.8 with dilute HCl, and the volume reduced
ꢀ
tion mixture was cooled to 0 C, quenched with
water, and extracted with chloroform (2Â5 mL).
The combined organic layers were washed with

144
H.G. Bazin et al./Carbohydrate Research 309 (1998) 135±144
to 20 mL by evaporation. The concentrated pro-
duct was decolorized with activated carbon at 40 ꢀC
and ®ltered through a celite pad. The eluent was
concentrated under reduced pressure to 6 mL and
precipitated with acetone providing a white pre-
cipitate that become tan upon standing. The
remaining supernatant was evaporated to 2 mL
and again precipitated with acetone. The combined
precipitates were taken up in water, eluted with
water from a 2.5Â25 cm Biogel P-2 column and
lyophilized. CE analysis showed a purity greater
than 95%. Neutralization of the eluent with
sodium hydroxide to pH 5.8 a€orded, after lyo-
philization 19 (0.083 mmol, 73%). FABMS (-ve)
m/z 748, 726, 704, 682 [M-5H+4Na] , [M-
4H+3Na] , [M-3H+2Na] , [M-2H+Na] , 1H
NMR (500 MHz, D2O), ꢂ 2.09, 2.11 (2s, 6H, 2
OAc), 3.46 (dd, 1H, J1;2 3.5 Hz, J2;3 10 Hz, H-2),
4.19 (dd, 1H, J3;4 10 Hz, J4;5 10 Hz, H-4), 4.24 (m,
1H, H-5), 4.31 (dd, 1H, J5;6a 11 Hz, J6a;6b 2.5 Hz,
H-6a), 4.34 (dd, J5:6b 4.5 Hz, H-6b), 4.63 (m, 1H,
H-20), 5.13 (dd, 1H, H-3), 5.31 (m, 1H, H-30), 5.40
[2] R.J. Linhardt, Analysis of glucosaminoglycans with
polysaccharide lyases in A. Varki (Ed.), Current
Protocols in Molecular Biology, Analysis of Glyco-
conjugates. Wiley Interscience, Boston, 1994.
[3] H.G. Bazin, RJ. Kerns, and RJ. Linhardt, Tetra-
hedron Lett., 38 (1997) 923±926.
[4] K.G. Rice and RJ. Linhardt, Carbohydr. Res., 190
(1985) 219±233.
[5] R.J. Kerns, I.R. Vlahov, and R.J. Linhardt, Car-
bohydr. Res., 267 (1995) 143±152.
[6] J. Alfoldi, R. Palovcõk, C. Peciar, J. Hirsch, and P.
Kovac, Carbohydr. Res., 44 (1975) 133±137.
[7] M. Ragazzi, D.R. Ferro, A. Provasoli, P. Pumilia,
A. Cassinari, G. Torri, M. Guerrini, B. Casu, HB.
Nader, and CP. Dietrich, J. Carbohydr. Chem., 12
(1993) 523±535.
[8] G.N. Bollenback, J.W. Long, D.G. Benjamin, and
JA. Lindquist, J. Am. Chem. Soc., 77 (1955) 3310±
3315.
[9] B. Glanzer, Z. Gyorgydeak, B. Bernet, and A.
Vasella, Helv. Chim Acta, 74 (1991) 343±369.
[10] M. Tanaka, M. Okita, and I. Yamatsu, Carbohydr.
Res., 241 (1993) 81±88.
[11] H.P. Wessel, T. Iversen, and D.R. Bundle, J.
Chem. Soc. Perkin Trans. I (1985) 2247±2251.
[12] C. Anderson, J. Freeman, LH. Lucas, M. Farley,
H. Dalhoumi, and T.S. Widlanski, Biochemistry,
36 (1997) 2586±2594.
0
0
(d, 1H, J1;2 3.5 Hz, H-1), 5.54 (d, 1H, J1 ;2 2:5 Hz,
H-10), 5.94 (dd, 1H, J3 ;4 5.0 Hz, J2 ;4 1 Hz, H-40).
0
0
0
0
Acknowledgements
[13] I. Nagasawa, Y. Inoue, and T. Kamata, Carbo-
hydr. Res. 58 (1977) 47±55.
[14] R.J. Ferrier, Adv. Carbohydr. Chem. Biochem., 24
(1969) 265±266.
This investigation was supported by grant number
GM38060 from the National Institutes of Health.
[15] C.A.G. Haasnoot, F.A.A.M. de Leeuw, and C.
Altona, Tetrahedron, 36 (1979) 2783±2792.
[16] R.J. Ferrier, and G.H. Sankey, J. Chem. Soc. (C),
(1966) 2345.
[17] C. Djerassi, R. Records, E. Bunnenberg, K. Mis-
low, and A. Moscowitz, J. Am. Chem. Soc., 84
(1962) 870.
References
[1] R.J. Linhardt and T. Toida, Heparin oligosacchar-
ides: New analogues development and applications in
Z. Witczak, and K. Nieforth (Eds.), Carbohydrates
in Drug Design, M. Dekker New York, 1997.