Synthesis and heparanase inhibitory activity of sulfated mannooligosaccharides related to the antiangiogenic agent PI-88

Article abstract of DOI:10.1016/j.bmc.2007.10.044

A stepwise synthetic route to the mannooligosaccharides from the neutral fraction of Pichia holstii phosphomannan hydrolysate, including a tetrasaccharylamine component, was developed using only two or three readily available d-mannose building blocks. Th

Full text of DOI:10.1016/j.bmc.2007.10.044

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 699–709
Synthesis and heparanase inhibitory activity of sulfated
mannooligosaccharides related to the antiangiogenic agent PI-88
Jon K. Fairweather, Edward Hammond, Ken D. Johnstone and Vito Ferro*
Drug Design Group, Progen Pharmaceuticals Limited, PO Box 2403, Toowong, Qld 4066, Australia
Received 17 August 2007; revised 8 October 2007; accepted 12 October 2007
Available online 18 October 2007
Abstract—A stepwise synthetic route to the mannooligosaccharides from the neutral fraction of Pichia holstii phosphomannan
hydrolysate, including a tetrasaccharylamine component, was developed using only two or three readily available D-mannose build-
ing blocks. These compounds were sulfonated to give the corresponding sulfated oligosaccharides which are closely related to the
constituents of the anticancer agent PI-88. The synthetic approach is well suited to the preparation of analogues as demonstrated by
the synthesis of a series of (1 ! 3)-linked mannooligosaccharides. The inhibitory activity of the sulfated oligosaccharides against
heparanase was determined using a Microcon ultrafiltration assay. The tetra- and pentasaccharides were potent competitive inhib-
itors of heparanase (Ki = 200–280 nM) whilst the shorter di- and trisaccharides were partial competitive inhibitors and did not com-
pletely inhibit the enzyme even at very high concentrations.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
phomannan produced by the yeast Pichia holstii NRRL
Y-2448.20 This mixture is primarily composed of the
phosphorylated a(1 ! 3)/a(1 ! 2)-linked penta- (5)
and tetrasaccharide 4, which together account for
approximately 90% of the total oligosaccharide content,
with the remaining 10% comprised of phosphorylated
di- (2), tri- (3) and hexasaccharide 6.21 The minor, neu-
tral fraction from the phosphomannan hydrolysate is
made up of the corresponding non-phosphorylated oli-
gosaccharides 7–11 which are also present following
dephosphorylation of the oligosaccharide phosphate
fraction with alkaline phosphatase.21,22 An inability to
separate the oligosaccharide phosphate fraction by size
exclusion chromatography (SEC) into its individual
components, attributable to the presence of the phos-
phate group,22 has meant that structure–activity rela-
tionship studies of PI-88 have been restricted to using
partially purified fractions obtained by SEC of PI-88 it-
self19 or by using the sulfated oligosaccharides (12–
15)23,16 obtained by sulfonation of the neutral oligosac-
charides 7–10 (see Chart 1).
The heparan sulfate (HS) mimetic PI-88 (1) is a recogni-
sed inhibitor of tumour growth and metastasis currently
undergoing Phase II clinical trials in patients with ad-
vanced malignancies.1–3 It is expected to commence
Phase III clinical evaluation in 2007 as an adjuvant ther-
apy for post-resection hepatocellular carcinoma. PI-88
inhibits angiogenesis4–6 by blocking the interactions of
angiogenic growth factors such as FGF-2 and VEGF
and their receptors with HS. In addition, PI-88 is a po-
tent inhibitor of heparanase,4 an endo-b-glucuronidase
that cleaves the HS side chains of proteoglycans that
are a major component of the extracellular matrix and
basement membranes.7 Heparanase plays a key role in
both metastasis and angiogenesis and is thus an attrac-
tive target for drug development.8–11 The biological ef-
fects of PI-88 are not restricted to anticancer activity.
For example, it also displays anticoagulant,12–14 anti-
proliferative,15 antiviral16,17 and antimalarial activity.18
PI-88 is prepared19 by exhaustive sulfonation of the oli-
gosaccharide phosphate fraction obtained following
mild acid-catalysed hydrolysis of the extracellular phos-
Like PI-88, compounds 12–15 display strong affinities
for FGF-1, FGF-2 and VEGF,23 and are able to inhibit
herpes simplex virus (HSV) infection of cells and the
cell-to-cell spread of HSV-1 and HSV-2,16 with activity
increasing with increased size. Small amounts of tetra-
saccharylamine 18 have also been detected in the neutral
hydrolysate fraction21 but because of its low abundance
and the difficulty in its isolation, the corresponding
Keywords: Sulfated mannooligosaccharides; Synthesis; Heparanase
inhibitors; PI-88.
*
Corresponding author. Tel.: +61 7 3273 9150; fax: +61 7 3375
1168; e-mail: vitof@progen-pharma.com
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.044

700
J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
O
X
OR
O
O
X
OR
O
RO
RO
OR
OR
RO
RO
OR
OR
O
O
RO
RO
O
O
2
n
O
O
RO
RO
RO
RO
RO
RO
O
O
NHR
OR
1
X = PO3Na2, R = SO3Na or H (n = 0-4)
16
X = PO3Na2, R = SO3Na or H
17 X = PO3Na2, R = H
18
2 X = PO3Na2, R = H (n = 0)
3 X = PO3Na2, R = H (n = 1)
X = R = H
4
X = PO3Na2, R = H (n = 2)
5 X = PO3Na2, R = H (n = 3)
19 X = R = SO3Na or H
20 X = R = Ac
6
X = PO3Na2, R = H (n = 4)
7 X = R = H (n = 0)
8 X = R = H (n = 1)
9
X = R = H (n = 2)
10 X = R = H (n = 3)
11
X = R = H (n = 4)
12 X = R = SO3Na or H (n = 0)
13 X = R = SO3Na or H (n = 1)
14
X = R = SO3Na or H (n = 2)
15 X = R = SO3Na or H (n = 3)
Chart 1.
sulfated oligosaccharylamine 19 was not synthesised and
therefore not assessed in the previous studies. The avail-
ability of compounds 12–15 and 19 for further biological
evaluation is also limited by the need to access large
scale Pichia fermentations and phosphomannan hydrol-
ysates. To address the need for sufficient quantities of
12–15 and 19 for further testing, the total synthesis of
these compounds and several analogues containing
exclusively a(1 ! 3) linkages was undertaken. The prod-
ucts were tested for their ability to inhibit human plate-
let heparanase using an improved assay.
protecting groups. Alternatively, a non-differentiable
block (23) may be used at the non-reducing end thus
‘capping’ the oligosaccharide at a fixed length. Part of
this strategy has been easily modified to allow for the
preparation of phosphorylated PI-88-related oligosac-
charides by using a phosphorylated or differentially 6-
O-protected capping block.29
The alcohol 25 was prepared in good yield by the
sequential processes of glycosylation (of the alcohol 21
with the imidate 22, catalysed by TMSOTf) and de-O-
allylation of the resultant disaccharide 24 (Scheme 1).
The alcohol 25 served as an acceptor for further elonga-
tion and following a similar glycosylation/de-O-allyla-
tion strategy, the imidate 22 was incorporated into the
growing chain yielding first the trisaccharides 26 and
27 and then the tetrasaccharides 28 and 29 in good over-
all yields. The alcohol 29 was ultimately glycosylated
with the peracetylated imidate 23 in 87% yield to afford
the pentasaccharide 30. Compounds 25, 27, 29 and 30
were deprotected through the sequential processes of
transesterification and hydrogenolysis to yield the oligo-
2. Results and discussion
2.1. Synthesis
Du et al. have reported a convergent ‘3 + 2’ block strat-
egy for the synthesis of PI-88 pentasaccharide analogues
which was successfully applied to the preparation of a
6V-monosulfated octyl mannopentasaccharide.24
A
more general approach amenable to the synthesis of
all PI-88 oligosaccharide components (di- to hexasac-
charides) and adaptable to the preparation of analogues
is a reiterative ‘1 + 1’ strategy. A similar 1 + 1 strategy
has been reported by Ikegami for the synthesis of a pro-
tected 1-C-methyl-substituted mannosyl pentasaccha-
ride analogue of PI-88.25 Using this 1 + 1 strategy, the
target oligosaccharides can all be synthesised with the
use of only two simple and readily available mannose
building blocks (Scheme 1): a starting block (21)26 and
a differentially protected, elongating block (22).27,28
The elongating block may be partially deprotected thus
providing a site for further extension or in conjunction
with the molecule as a whole, completely liberated of
1
mannans 7–10, respectively. The products displayed H
and 13C NMR spectroscopic properties in accord with
those reported in the literature.22,30 The oligomannans
were sulfonated as previously described23 to give the sul-
fated oligomannans 12–15. Compounds 12–15 displayed
1
identical H NMR spectroscopic and capillary electro-
phoretic (CE) properties to authentic samples previously
prepared.23 The sulfonation of PI-88-related oligosac-
charides larger than disaccharides with reagents such
as sulfur trioxide pyridine or trimethylamine complex
does not go to completion. Following precedent,23,31
the target compounds were all obtained as reproducible
mixtures of highly but incompletely sulfated forms. The

J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
701
OBz
OBz
O
OAc
OAc
OBn
OH
O
O
BzO
BnO
BnO
AcO
AcO
O
O
CCl3
NH
OBn
O
CCl3
22
21
23
NH
OBz
OBz
OAc
OAc
O
O
BzO
RO
OBz
OBz
AcO
AcO
OBz
OBz
O
O
BzO
BzO
O
O
n
n
O
O
BnO
BnO
BnO
BnO
BnO
O
O
BnO
OBn
OBn
a
a
21 + 22
24
25
R = allyl, n = 0
R = H, n = 0
29 + 23
30 n = 3
10
15
c,d
e
b
c,d
e
7
12
a
25
27
22
22
+
+
26 R = allyl, n = 1
27 R = H, n = 1
c,d
c,d
e
e
b
8
9
13
14
a
b
28 R = allyl, n = 2
29 R = H, n = 2
Scheme 1. Synthesis of mannooligosaccharides 12–15. Reagents and conditions: (a) TMSOTf, 1,2-DCE, 0 ꢁC, 10 min, 84–94%; (b) PdCl2, MeOH/
1,2-DCE, 70 ꢁC, 40 min, 64–91%; (c) NaOMe, MeOH/THF, rt, o/n; (d) Pd(OH)2/C, H2 (100 psi), THF/H2O/AcOH, 3 h, 56–67%, two steps; (e) i—
SO3ÆPy, DMF, 60 ꢁC, 6 h; ii—1 M NaOH, 40–50%.
1
purity of the compounds was therefore determined by a
combination of CE and 1H NMR spectroscopy (see Sec-
tion 4).
characterised by H NMR spectroscopy and CE. The
presence of small quantities of 18 in the neutral fraction
of Pichia phosphomannan hydrolysates was previously
inferred by the isolation and NMR structure elucidation
of slightly impure peracetate 20.21 In the present study,
whilst the isolation of 18 was not required, peracetate 20
was readily obtained by hydrogenation of 32 followed
by acetylation. Flash chromatographic purification
afforded a sample of 20 of greater purity than the previ-
ously isolated sample but which was otherwise identical
The sulfated tetrasaccharylamine derivative 19 was syn-
thesised as outlined in Scheme 2. The tetrasaccharide 9
was first acetylated under standard conditions and the
resulting peracetate treated with HBr in acetic acid to
form the a-bromide 31 which was subsequently dis-
placed with azide ion to give the b-azide 32 in moderate
yield (41% from the peracetate). Azide 32 was deacety-
by H and 13C NMR spectroscopy, thus confirming the
1
´
lated under Zemplen conditions and then hydrogenated
earlier structural assignment.
with Adam’s catalyst to give, presumably, tetrasaccha-
rylamine 18, which was not isolated but instead immedi-
ately sulfonated to give the target compound 19,
The stepwise synthetic approach utilized for the synthe-
sis of 12–15 and 19 is well suited to the preparation of
OR
OR
O
OAc
OAc
O
RO
RO
AcO
AcO
OR
OR
OAc
OAc
a,b
c
e,a
O
O
9
20
RO
AcO
O
O
2
2
O
O
RO
RO
RO
AcO
AcO
AcO
O
O
R1
Br
32 R = Ac, R1 = N3
18
31
d,e
R = H, R1 = NH2
19 R = SO3Na or H, R1 = NHSO3Na
f
Scheme 2. Synthesis of tetrasaccharylamine derivatives 19 and 20. Reagents and conditions: (a) Ac2O, Py, DMAP, rt, o/n; (b) HBr/HOAc/1,2-DCE,
rt, o/n; (c) NaN3, DMF, 110 ꢁC, o/n, 41% two steps; (d) NaOMe, MeOH, rt, 4 h, 72%; (e) PtO2, H2 (100 psi), MeOH/EtOAc/AcOH, 3 h; (f) i—
SO3ÆPy, DMF, 60 ꢁC, 6 h; ii—1 M NaOH, 55% two steps.

702
J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
PI-88 analogues. A simple change to the nature of the
starting, elongating or capping blocks allows access to
a variety of different analogues. This is illustrated in
Scheme 3 whereby changing the starting block to the
alcohol 3327 gives rise to all (1 ! 3)-linked analogues
such as 41 and 43, which are useful for examining the
nature of the carbohydrate linkages on biological
activity.
from the smaller, cleaved products of heparanase catal-
ysis which are then quantified by scintillation counting.
The sulfated oligosaccharides were evaluated for their
heparanase inhibitory activity using the assay described
above (Table 1 and Fig. 1). The data indicate that the di-
and trisaccharide members of this series (12, 13, 41, 43)
deviate from the expected result for putative competitive
inhibitors. Competitive inhibitors should approach
100% inhibition of their target enzyme at saturating con-
centrations. Under these conditions, nearly all enzyme
active sites are occupied by inhibitor molecules prevent-
ing catalysis. This near 100% inactivation of the enzyme
clearly occurs with the longer compounds of the series
(14, 15 and 19) but does not for the di- and trisaccha-
rides which reach between approximately 30% and
90% inhibition at inhibitor concentrations 10-fold high-
er than their respective IC50 values. At these saturating
concentrations, the disaccharides (12 and 41) inhibit
heparanase less than the trisaccharides (13 and 43).
The IC50 values determined for the compounds were
used, along with the determined Km of heparanase for
HS (1.64 lM, data not shown), to calculate Kic values.38
The affinity of these compounds for the active site of
heparanase is very similar to that of PI-88 as indicated
by their similar Kic.
2.2. Heparanase inhibition studies
Heparanase for use in the inhibition studies was purified
from human platelets by a modification of the procedure
of Freeman and Parish32 via the sequential fractionation
of a crude platelet extract on Concanavalin A Sephar-
ose, Sulfopropyl Sepharose and Sephacryl S-200 HR
chromatography columns. This modification results in
a more efficient purification (one less chromatographic
step is required) with an improved overall yield of
protein.
Most of the heparanase assays developed to date are
suitable for qualitative assessment of heparanase inhib-
itors but not so for detailed kinetic analysis.33–36 To ad-
dress this, we developed a sensitive heparanase assay
which uses [3H]-labelled HS as substrate and allows
for the variation of substrate concentration. This assay
is similar to that developed independently by Tsuchida
et al.37 in that it uses ultrafiltration devices, in this case
a Microcon YM-10, to separate the longer native HS
The di- and trisaccharides did not conform to simple
competitive inhibition so more experiments were per-
formed using example compounds (12 and 15) to deter-
OR
OR
OBz
OBz
OBz
OBz
O
O
O
a
c
BzO
HO
RO
RO
BzO
RO
OR
OR
OBz
OBz
+ 22
O
O
OMe
RO
BzO
O
33
O
n
OMe
n
OMe
34
R = allyl, n = 1
40 R = H, n = 1
d
d
b
35 R = H, n = 1
41 R = SO3Na or H, n = 1
42
43
R = H, n = 2
R = SO3Na or H, n = 2
a
35 + 22
36
37
R = allyl, n = 2
R = H, n = 2
b
44 R = H, n = 3
a
b
37
22
+
38 R = allyl, n = 3
39 R = H, n = 3
Scheme 3. Synthesis of all a(1 ! 3)-linked mannooligosaccharides. Reagents and conditions: (a) TMSOTf, 1,2-DCE, 0 ꢁC ! rt, 30 min; (b) PdCl2,
MeOH/1,2-DCE, 70 ꢁC, 40 min, 54–70%, two steps; (c) NaOMe, MeOH/THF, rt, o/n, 85–98%; (d) i—SO3ÆPy, DMF, 60 ꢁC, 6 h; ii—1 M NaOH, 36–
48%.
Table 1. Inhibition of human platelet heparanase by sulfated oligosaccharides 12–15, 19, 41 and 43 as determined by Microcon ultrafiltration assay
Compound
IC50 (lM)
Kic (lM)
Kiu (lM)
Maximum inhibition (%)
1
12
13
14
15
19
41
43
0.98 0.11
1.57 0.31
2.69 0.46
0.80 0.06
1.14 0.18
0.99 0.11
0.25 0.12
4.15 0.36
0.24 0.03
2.23 0.20
ND
NA
100
58.6
84.3
100
100
100
34.2
91.3
68.3 22.4
ND
NA
0.20 0.02
0.28 0.04
0.24 0.03
ND
NA
NA
ND
ND
ND
Inhibition data for PI-88 (1) are given for comparison. NA, not applicable; ND, not determined.

J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
703
100
80
60
40
20
0
100
a
b
80
60
40
20
12
13
14
15
1
19
41
43
0
0
10
20
[Inhibitor] (µM)
30
40
0
10
20
[Inhibitor] (µM)
30
40
Figure 1. Inhibition of heparanase by the sulfated oligosaccharides 12–15 (a) and 1, 19, 41 and 43 (b). Heparanase assays were conducted at 0.9 lg/mL
heparanase, 5 lM [3H]HS, 40 mM sodium acetate, pH 5.0, and 0.1 mg/mL BSA at 37 ꢁC for 4 h.
0.025
0.02
0.015
0.01
0.005
0
mine their mode of inhibition and also to confirm that
the longer oligosaccharides were acting as competitive
inhibitors. Double reciprocal plot analysis confirmed
that pentasaccharide 15 behaves as a simple competitive
inhibitor of heparanase with respect to substrate HS
(Fig. 2). The linear plots intercept at the y-axis indicat-
ing that 15 increases the apparent Km but has no effect
upon Vm, traits characteristic of competitive inhibitors.
The double reciprocal plot for disaccharide 12 is similar
(Fig. 3) but this compound is not a simple competitive
inhibitor due to its inability to completely inhibit hepa-
ranase at saturating concentrations. Dixon plots of the
data are nonlinear, which, in combination with the other
observations, indicate that 12 is a partial competitive
inhibitor. The rate equation for partial competitive inhi-
bition was used to determine Kic and Kiu for compound
12 by nonlinear regression.39
0 µM
1 µM
5 µM
10 µM
20 µM
-0.05
0
0.05
0.1
0.15
0.2
0.25
1/[S]
Figure 3. Double reciprocal plot of heparanase inhibition by sulfated
disaccharide 12. Heparanase assays were conducted at 0.9 lg/mL
heparanase, various concentrations of 3[H]HS, 40 mM sodium acetate,
pH 5.0, and 0.1 mg/mL BSA at 37 ꢁC for 4 h.
0.14
0 µM
0.8 µM
1.6 µM
3 µM
0.12
0.1
The large size of the active site of heparanase may be the
reason why similar molecules have such different inter-
actions with this enzyme. During catalysis, heparanase
interacts with at least three saccharides within substrate
HS.40 These interactions are largely between negatively
charged groups on HS (sulfates and carboxylates) and
positively charged residues (lysines and arginines) flank-
ing both sides of the catalytic glutamate residues.41 The
sulfated oligosaccharides described here are likely to
interact with these positively charged residues but to
have little affinity for the negatively charged glutamate
residues. It is conceivable that the smaller oligosaccha-
rides such as 12 bind to some of the positive residues
but due to their small size leave access to the catalytic
glutamate residues allowing substrate to bind and catal-
ysis to occur. This hypothesis explains how these smaller
oligosaccharides, apparently interacting with heparan-
ase as competitive inhibitors, do not approach complete
5 µM
0.08
0.06
0.04
0.02
0
-0.05
0
0.05
0.1
0.15
0.2
0.25
1/[S]
Figure 2. Double reciprocal plot of heparanase inhibition by sulfated
pentasaccharide 15. Heparanase assays were conducted at 0.9 lg/mL
heparanase, various concentrations of 3[H]HS, 40 mM sodium acetate,
pH 5.0, and 0.1 mg/mL BSA at 37 ꢁC for 4 h.

704
J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
enzyme inactivation even at very high concentrations.
The higher affinity of 12 for free enzyme than enzyme
with substrate bound (Kic < Kiu) is consistent with this
hypothesis because substrate bound to heparanase will
restrict access to the positive residues. The interaction
between heparanase and 12 is contrasted with that of
the longer compounds such as 14 and 15 which, due to
their length, may block substrate access to the catalytic
glutamate residues and completely inhibit catalysis.
50% EtOAc/hexane) to yield the allyl ether 24 as a col-
ourless oil (1.14 g, 84%). H NMR (400 MHz, CDCl3)
1
d 3.67–3.81, 3.88–3.95, 4.06–4.15, 4.30–4.35 (4 m, 12H,
H-2I, -3I, -4I, -5I, -6aI, -6bI, -3II, -5II, -6aII, -6bII,
OCH2), 4.94–4.70 (m, 7H, CH2Ph), 4.84 (d, 1H,
JA,B = 10.8 Hz; A of AB q, CH2Ph), 4.93–4.96, 5.04–
5.09 (2 m, 2H, @CH2), 5.02 (d, 1H, J1,2 = 1.9 Hz, H-
1I), 5.24 (d, 1H, J1,2 = 1.9 Hz, H-1II), 5.59–5.69 (m,
1H, @CH), 5.72 (dd, 1H, J2,3 = 3.1 Hz, H-2II), 5.75
(dd, 1H, J3,4 = 9.8, J4,5 = 9.9 Hz, H-4II), 7.09–7.58,
7.97–8.06 (2 m, 35H, Ph); 13C NMR (100 MHz, CDCl3)
d 61.5, 63.5 (C-6I, -6II), 68.6, 69.2, 69.3, 69.5, 69.6, 71.1,
72.0, 72.6, 73.6, 74.7, 75.3, 75.4 (13 C; C-3I, -4I, -5I, -2II,
-3II, -4II, -5II, OCH2, CH2Ph), 80.0 (C-2I), 98.5, 99.6 (C-
1I, -1II), 117.7 (@CH2), 127.7–138.4 (43 C; @CH, Ph),
165.6, 165.7, 166.4 (C@O); ESMS: m/z 1072.2
[M+NH4]+.
3. Conclusions
In conclusion, a simple synthetic route to the oligosac-
charides from the neutral fraction of P. holstii phospho-
mannan hydrolysate, including the tetrasaccharylamine
component, was developed. These compounds were sul-
fonated to give the corresponding sulfated oligosaccha-
rides which are closely related to the constituents of
the anticancer agent PI-88. The synthetic approach is
well suited to the preparation of analogues as demon-
strated by the synthesis of a series of all-(1 ! 3)-linked
derivatives. The inhibitory activity of the sulfated oligo-
saccharides against human platelet heparanase was
determined using a Microcon ultrafiltration assay. The
tetra- and pentasaccharides were potent competitive
inhibitors of heparanase whilst the shorter di- and tri-
saccharides were partial competitive inhibitors and did
not completely inhibit the enzyme at saturating concen-
trations. This study now gives a clearer picture of the
biological activity of the PI-88 components and of the
size requirements of sulfated oligosaccharides for effi-
cient inhibition of heparanase. This information may
be useful in the design of new heparanase inhibitors.
4.3. Benzyl 2-O-(2,4,6-tri-O-benzoyl-a-D-mannopyrano-
syl)-3,4,6-tri-O-benzyl-a-D-mannopyranoside (25)
PdCl2 (40 mg) was added to a solution of the allyl ether
24 (1.09 g, 0.97 mmol) in MeOH (10 mL) and 1,2-DCE
(10 mL) and the combined mixture was heated (70 ꢁC,
40 min). The solvents were evaporated and the residue
subjected to FC (20–30% EtOAc/hexanes) to yield the
1
alcohol 25 as a colourless oil (0.96 g, 91%). H NMR
(400 MHz, CDCl3) d 3.68–3.81, 3.97–4.06, 4.32–4.71 (3
m, 18H, H-2I, -3I, -4I, -5I, -6aI, -6bI, -3II, -5II, -6aII, -
6bII, CH2Ph), 4.84 (d, 1H, JA,B = 12 Hz A of AB q,
CH2Ph), 5.05 (d, 1H, J1,2 = 1.9 Hz, H-1I), 5.26 (d, 1H,
J1,2 = 1.9 Hz, H-1II), 5.61 (dd, 1H, J2,3 = 3.3 Hz, H-
2II), 5.67 (dd, 1H, J3,4 = 9.8, J4,5 = 9.9 Hz, H-4II),
7.13–7.40, 7.48–7.59, 7.98–8.06 (3 m, 35H, Ph); 13C
NMR (100 MHz, CDCl3) d 60.6, 63.3 (C-6I, -6II),
69.06, 69.12, 69.2, 69.4, 70.4, 72.1, 72.6, 72.8, 73.5,
74.8, 75.47, 75.48, 76.2 (C-3I, -4I, -5I, -2II, -3II, -4II, -
5II, OCH2, CH2Ph), 79.7 (C-2I), 98.3, 99.4 (C-1I, -1II),
127.7–138.5 (42 C, Ph), 166.0, 166.4, 167.0 (C@O);
ESMS: m/z 1015.2 [M+H]+, 1037.2 [M+Na]+.
4. Experimental
4.1. General
General experimental procedures have been given previ-
ously.29 Capillary electrophoresis (CE) was performed
in reverse polarity mode on a Beckman P/ACE 5000
System equipped with a P/ACE UV absorbance detec-
tor, using 10 mM 5-sulfosalicylic acid (pH 3) as the
background electrolyte, as previously described.19
Radioactivity (3H) was counted on a Wallac 1450
MicroBeta Trilux liquid scintillation counter using Opti-
phase HiSafe 2 (Perkin-Elmer) as the scintillation fluid.
4.4. Benzyl 2-O-[(3-O-allyl-2,4,6-tri-O-benzoyl-a-D-man-
nopyranosyl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyr-
anosyl)]-3,4,6-tri-O-benzyl-a-D-mannopyranoside (26)
A mixture of the imidate 22 (742 mg, 1.01 mmol) and
the alcohol 25 (908 mg, 0.84 mmol) was treated with
TMSOTf as described for 24 to yield the allyl ether 26
1
as a colourless oil (1.26 g, 90%). H NMR (400 MHz,
CDCl3) d 3.51–3.56, 3.66–4.06, 4.23–4.27, 4.30–42,
4.47–4.72, 4.78–4.86 (6 m, 26H, H-2I, -3I, -4I, -5I, -6aI,
-6 bI, -3II, -5II, -6aII, -6bII, -3III, -5III, -6aIII, -6bIII,
OCH2, @CH2, CH2Ph), 5.04 (d, 1H, J1,2 = 1.7 Hz, H-
1I), 5.15 (dd, 1H, J1,2 = 1.8, J2,3 = 2.7 Hz, H-2II), 5.26
(d, 1H, H-1II), 5.28 (d, 1H, J1,2 = 1.7 Hz, H-1III), 5.33–
5.43 (m, 1H, @CH), 5.77–5.82 (m, 2H, H-4II, -2III),
5.92 (dd, 1H, J3,4 = 9.5, J4,5 = 9.8 Hz, H-4III), 7.00–
7.61, 7.80–8.19 (2 m, 50H, Ph); 13C NMR (100 MHz,
CDCl3) d 62.0, 62.9, 67.3, 68.2, 68.6, 68.9, 69.0, 69.1,
69.6, 70.2, 71.3, 71.7, 72.2, 73.1, 73.7, 74.5, 75.0, 75.8,
76.1, 77.2, 79.1, 97.8, 99.0, 99.4, 117.1, 127.3, 127.4,
127.5, 127.6, 127.8, 127.9, 128.1, 128.2, 128.3, 128.4,
128.5, 128.7, 129.2, 129.4, 129.5, 129.7, 129.8, 129.9,
4.2. Benzyl 2-O-(3-O-allyl-2,4,6-tri-O-benzoyl-a-D-man-
nopyranosyl)-3,4,6-tri-O-benzyl-a-D-mannopyranoside
(24)
A mixture of the imidate 22 (902 mg, 1.21 mmol)28,27
and the alcohol 21 (723 mg, 1.34 mmol)26 in 1,2-DCE
(10 mL) was stirred in the presence of mol. sieves
˚
(1.0 g of 3 A powder) under an atmosphere of argon
(30 min). The mixture was cooled (0 ꢁC) with continued
stirring (10 min) prior to the addition of TMSOTf
(219 lL, 1.21 mmol). After 10 min, Et3N (100 lL) was
introduced and the mixture was filtered. The solvent
was evaporated and the residue subjected to FC (10–

J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
705
132.5, 132.9, 133.4, 133.6, 136.8, 138.0, 164.7, 165.1,
1IV), 5.22 (dd, 1H, J1,2 = 2.1, J2,3 2.6 Hz, H-2II), 5.27–
5.29 (m, 2H, H-1III, -1IV), 5.46 (dd, 1H, J3,4 = 9.7,
J4,5 = 9.9 Hz, H-4IV), 5.79 (dd, 1H, J2,3 = 2.9 Hz, H-
2IV), 5.90–5.96 (m, 2H, H-4II, -4III), 7.01–7.56, 7.68–
8.16 (2 m, 65H, Ph).
165.5, 165.7, 165.9; ESMS: m/z 1552.5 [M+Na]+.
4.5. Benzyl 2-O-[(2,4,6-tri-O-benzoyl-a-D-mannopyrano-
syl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyranosyl)]-
3,4,6-tri-O-benzyl-a-D-mannopyranoside (27)
4.8. Benzyl 2-O-[(2,3,4,6-tetra-O-acetyl-a-D-mannopyr-
anosyl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyrano-
syl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyranosyl)-
(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyranosyl)]-3,4,6-
tri-O-benzyl-a-D-mannopyranoside (30)
The allyl ether 26 (394 mg, 241 lmol) was treated with
PdCl2 as described for 25 to yield the alcohol 27 as a col-
ourless oil (317 mg, 84%). H NMR (400 MHz, CDCl3)
1
d 3.67–3.82, 3.91–3.99, 4.01–4.21, 4.29–4.71 (4 m, 21H,
H-2I, -3I, -4I, -5I, -6aI, -6bI, -3II, -5II, -6aII, -6bII, -3III
,
-5III, -6aIII, -6bIII, CH2Ph), 4.83 (d, 1H, JA,B = 10.9 Hz,
A of AB q, CH2Ph), 5.03–5.05 (m, 2H, H-1I, -2II), 5.25–
5.28 (m, 2H, H-1II, -1III), 5.63 (dd, 1H,
J3,4 = J4,5 = 9.9 Hz, H-4II), 5.77 (dd, 1H, J1,2 = 2.0,
J2,3 = 3.1 Hz, H-2III), 5.92 (dd, 1H, J3,4 = 9.7,
J4,5 = 9.9 Hz, H-4III), 6.99–7.62, 7.80–8.16 (2 m, 50H,
Ph); ESMS: m/z 1511.4 [M+Na]+.
A mixture of the trichloroacetimidate 23 (39 mg,
78 lmol) and the alcohol 29 (85 mg, 39 lmol) was trea-
ted with TMSOTf as described for 24 to yield the penta-
saccharide 30 as a colourless oil (85 mg, 87%). 1H NMR
(400 MHz, CDCl3) d 1.82–2.04 (4 s, 3H each; CH3CO),
3.67–3.95, 4.05–4.72, 4.82–5.03, 5.21–5.28, 5.69–5.50 (m,
43 H; H-1I–IV, -2I–IV, -3I–IV, -4I–IV, -5I–IV, -6abI–IV
CH2Ph), 7.01–7.56, 7.68–8.16 (2 m, 65H0 Ar).
,
4.6. Benzyl 2-O-[(3-O-allyl-2,4,6-tri-O-benzoyl-a-D-man-
nopyranosyl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-manno-
pyranosyl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyr-
anosyl)]-3,4,6-tri-O-benzyl-a-D-mannopyranoside (28)
4.9. a-D-Mannopyranosyl-(1 ! 2)-D-mannopyranose (7)
(A) A small piece of sodium was added to a solution of
the tetrabenzyl ether 25 (86 mg, 85 lmol) in MeOH/
THF (4 mL of 1:1) and the combined mixture was stir-
red (rt, o/n). The mixture was neutralised with Dowex
50-X8 resin (H+ form) and filtered. The solvent was
evaporated and co-evaporated (MeOH) and used in
the following reaction without further purification. (B)
Pd(OH)2 (10% on C, 20 mg) was added to a solution
of the crude product from (A) in THF/H2O (4 mL of
1:1) containing a little AcOH (50 lL) and the combined
mixture was vigorously stirred under hydrogen (100 psi,
3 h). The mixture was filtered, the solvent evaporated
and the residue was subjected to size exclusion chroma-
tography (Bio-Gel P-2; H2O; 60 mL/h) to yield, after
lyophilisation, the disaccharide 7 as a colourless powder
(17 mg, 59%, two steps). The 1H NMR (400 MHz, D2O)
spectrum was consistent with that reported in the litera-
ture.22 HRMS: m/z 343.1180 [M+H]+, 365.0972
[M+Na]+.
A mixture of the imidate 22 (102 mg, 138 lmol) and the
alcohol 27 (135 mg, 86.5 lmol) was treated with
TMSOTf as described for 24 to yield the allyl ether 28
1
as a colourless oil (173 mg, 94%). H NMR (400 MHz,
CDCl3) d 3.44–3.49, 3.60–3.99, 4.05–4.16, 4.42–4.44,
4.48–4.68, 4.73–4.77 (6 m, 30H, H-2I, -3I, -4I, -5I, -6aI,
-6bI, -3II, -5II, -6aII, -6bII, -3III, -5III, -6aIII, -6bIII, -3IV,
-5IV, -6aIV, -6bIV, OCH2, @CH2, CH2Ph), 4.83 (d, 1H,
JA,B = 10.9 Hz; A of AB q, CH2Ph), 5.01–5.04 (m, 2H,
H-1I, -2III), 5.19–5.23 (m, 1H, H-2II), 5.27–5.40 (m,
4H, H-1I, -1II, -1III, @CH2), 5.61 (dd, 1H,
J3,4
=
4,5 = 9.9 Hz, H-4IV), 5.77 (dd, 1H, J1,2 = 2.0,
J2,3 = 3.1 Hz, H-2IV), 5.90–5.96 (m, 2H, H-4II, -4III),
7.01–7.56, 7.70–8.16 (2 m, 65H, Ph); 13C NMR
(100 MHz, CDCl3) d 61.9, 62.3, 62.9, 67.3, 67.3, 68.0,
68.6, 68.9, 69.0, 69.1, 69.2, 69.5, 70.2, 71.2, 71.3, 71.8,
72.3, 73.2, 73.8, 74.6, 75.1, 75.9, 76.0, 79.1, 97.8, 98.9,
99.0, 99.2, 117.1, 127.3, 127.4, 127.4, 127.7, 127.8,
127.9, 128.0, 128.2, 128.3, 128.5, 128.6, 128.7, 128.9,
129.0, 129.1, 129.5, 129.6, 129.7, 129.7, 129.8, 130.0,
132.6, 132.6, 132.8, 132.8, 133.0, 133.3, 133.5, 133.7,
136.9, 138.5, 164.7, 164.7, 165.1, 165.3, 165.6, 165.7,
166.0.
4.10. a-D-Mannopyranosyl-(1 ! 3)-a-D-mannopyranosyl-
(1 ! 2)-D-mannopyranose (8)
The trisaccharide 27 (53 mg, 36 lmol) was deprotected
as described for 7 to yield the trisaccharide 8 as a colour-
less powder (12 mg, 67%). The 1H NMR (400 MHz,
D2O) spectrum was consistent with that reported in
the literature.22 HRMS: m/z 505.1702 [M+H]+,
527.1516 [M+Na]+.
4.7. Benzyl 2-O-[(2,4,6-tri-O-benzoyl-a-D-mannopyrano-
syl)-(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyranosyl)-
(1 ! 3)-(2,4,6-tri-O-benzoyl-a-D-mannopyranosyl)]-3,4,6-
tri-O-benzyl-a-D-mannopyranoside (29)
4.11. a-D-Mannopyranosyl-(1 ! 3)-a-D-mannopyranosyl-
The allyl ether 28 (155 mg, 70.4 lmol) was treated with
PdCl2 as described for 25 to yield the alcohol 29 as a col-
ourless oil (97 mg, 64%). 1H NMR (400 MHz, CDCl3) d
3.67–3.82, 3.90–4.10, 4.24–4.68 (3 m, 26H, H-2I, -3I, -4I,
-5I, -6aI, -6bI, -3II, -5II, -6aII, -6bII, -3III, -5III, -6aIII, -
6bIII, -3IV, -5IV, -6aIV, -6bIV, CH2Ph), 4.84 (d, 1H,
(1 ! 3)-a-D-mannopyranosyl-(1 ! 2)-D-mannopyranose
(9)
The tetrasaccharide 29 (42 mg, 21 lmol) was deprotec-
ted as described for 7 to yield the tetrasaccharide 9 as
a colourless powder (9 mg, 63%). The 1H NMR
(400 MHz, D2O) spectrum was consistent with that re-
ported in the literature.22 HRMS: m/z 667.2235
[M+H]+, 689.2015 [M+Na]+.
JA,B = 11.2 Hz,
A of AB q, CH2Ph), 4.86 (d,
J1,2 = 1.8 Hz, H-1I), 4.90 (dd, 1H, J1,2 1.8,
J2,3 = 3.1 Hz, H-2III), 5.03 (d, 1H, J1,2 = 1.5 Hz, H-

706
J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
4.12. a-D-Mannopyranosyl-(1 ! 3)-a-D-mannopyranosyl-
(1 ! 3)-a-D-mannopyranosyl-(1 ! 3)-a-D-mannopyrano-
syl-(1 ! 2)-D-mannopyranose (10)
4.15. N-Acetyl 2,3,4,6-tetra-O-acetyl-a-D-mannopyrano-
syl-(1 ! 3)-2,4,6-tri-O-acetyl-a-D-mannopyranosyl-(1 ! 3)-
2,4,6-tri-O-acetyl-a-D-mannopyranosyl-(1 ! 2)-3,4,6-tri-
O-acetyl-b-D-mannopyranosylamine (20)
The pentasaccharide 30 (64 mg, 28 lmol) was deprotec-
ted as described for 7 to yield the pentasaccharide 10 as
a colourless powder (13 mg, 56%). The 1H NMR
(400 MHz, D2O) spectrum was consistent with that re-
ported in the literature.22,30 HRMS: m/z 829.2810
[M+H]+.
(A) PtO2 (20 mg) was added to a solution of the azide 32
(30 mg, 24.3 lmol) in 1:1 MeOH/EtOAc (4 mL) contain-
ing a little AcOH (50 lL) and the combined mixture was
vigorously stirred under hydrogen (100 psi, 3 h). The mix-
ture was filtered and the solvent evaporated and co-evap-
orated (CH3CN) to yield, presumably, the amine as a
colourless glass. This sample was used in the next reaction
without further purification. (B) The crude product from
(A) was treated with pyridine (1 mL), Ac2O (1 mL) and
DMAP (20 mg) and the combined mixture stirred (rt, o/
n). The mixture was treated with MeOH (1 mL) and sub-
jected to workup and FC (40–70% EtOAc/hexane) to
yield the amide 20 as a colourless powder (12 mg, 40%,
two steps). The sample was of greater purity than that pre-
4.13. Sulfonation of oligosaccharides (7–10)
Oligosaccharides 7–10 were sulfonated (SO3ÆPy com-
plex) and purified (Bio-Gel P-2) as previously de-
scribed23 to yield the sulfated oligosaccharides 12–15
as amorphous white powders (40–50%). Compounds
12–15 were identical in all respects to the previously pre-
pared samples.23 The purity of samples as determined by
CE was >95%.
1
viously isolated21 but otherwise displayed identical H
and 13C NMR spectroscopic properties.
4.14. 2,3,4,6-tetra-O-Acetyl-a-D-mannopyranosyl-(1 ! 3)-
2,4,6-tri-O-acetyl-a-D-mannopyranosyl-(1 ! 3)-2,4,6-tri-
O-acetyl-a-D-mannopyranosyl-(1 ! 2)-3,4,6-tri-O-acetyl-
b-D-mannopyranosyl azide (32)
4.16. Sulfated tetrasaccharylamine (19)
(A) A small piece of sodium metal was added to a mix-
ture of the azide 32 (30 mg, 24.3 lmol) in MeOH (3 mL)
and the combined mixture stirred (rt, 4 h). The solution
was neutralised by addition of Dowex AG50-X8 (H+)
resin and filtered. The filtrate was evaporated and the
residue lyophilised to yield, presumably, a-D-mannopyr-
anosyl-(1 ! 3)-a-D-mannopyranosyl-(1 ! 3)-a-D-man-
nopyranosyl-(1 ! 2)-b-D-mannopyranosyl azide as a
pale yellow coloured glass (12 mg, 72%). This was used
in the following reactions without further purification or
characterisation. (B) PtO2 (20 mg) was added to a solu-
tion of the azide 32 from (A) (24.3 lmol, max) in H2O
(3 mL) and the combined mixture was vigorously stirred
under hydrogen (100 psi, o/n). The mixture was filtered
and the solvent evaporated and lyophilised (H2O) to
yield, presumably, the amine 18 as a colourless glass.
This sample was used in the next reaction without fur-
ther purification. (C) The amine 18 from (B) was sulfo-
nated as described for the preparation of 12–15 to yield
19 (20 mg, 55%) as an amorphous white powder after
(A) A mixture of the tetrasaccharide 9 (126 mg,
189 lmol) and N,N-dimethylaminopyridine (10 mg)
was stirred in pyridine (3 mL) and Ac2O (2 mL) (rt,
o/n). The mixture was cooled (0 ꢁC), quenched
(MeOH) and the solvents evaporated. The residue
was subjected to workup (EtOAc) and FC (30–60%
EtOAc/hexanes) to yield
a
colourless powder
(193 mg, 81%). This sample was used in the next reac-
tion without further purification or characterisation.
(B) HBr (1.0 mL of 30% in HOAc, 3.7 mmol) was
added to a solution of the peracetate from (A)
(180 mg, 0.14 mmol) in 1,2-DCE (10 mL) and the mix-
ture stirred (rt, o/n). The mixture was poured onto an
ice-water mixture and immediately subjected to work-
up (CHCl3) to yield, presumably, the bromide 31 as a
pale yellow coloured oil. This sample was used in the
next reaction without further purification. (C) The
product from (B) (0.14 mmol, max) and NaN3
(93 mg, 1.43 mmol) in DMF (5 mL) was heated
(110 ꢁC, o/n). The mixture was cooled and the solvent
evaporated. The residue was dissolved (EtOAc) and
subjected to workup to yield the azide 32 as a colour-
1
lyophilisation. H NMR (400 MHz, D2O): d 3.70–4.51
(26H, m), 4.67–4.73 (2H, m), 4.91 (1H, m), 5.12 (1H,
m), 5.44–5.25 (4H, m), 5.68 (1H, m).
1
less oil (73 mg, 41%, two steps). H NMR (400 MHz,
4.17. Methyl 3-O-(2,4,6-tri-O-benzoyl-a-D-mannopyran-
osyl)-2,4,6-tri-O-benzoyl-a-D-mannopyranoside (35)
CDCl3) d 1.97, 2.01, 2.02, 2.06, 2.07, 2.08, 2.10,
2.10(5), 2.11(2), 2.12(7), 2.13(0), 2.16, 2.17 (13 s,
13H, OAc), 3.74 (ddd, 1H, H-5), 3.87 (ddd, 1H, H-
5), 3.93 (ddd, 1H, H-5), 3.99–4.31 (m, 11H), 4.40
(ddd, 1H, H-5I), 4.91 (d, 1H, J1,2 = 1.2 Hz, H-1),
4.95 (d, 1H, J1,2 = 2.0 Hz, H-1,), 4.97 (d, 1H,
J1,2 = 1.6 Hz, H-1), 5.00–5.03 (m, 3H), 5.16–5.33 (m,
6H); 13C NMR (100 MHz, CDCl3) d 20.7, 20.83,
20.87, 20.91, 20.94, 20.97, 21.00, 21.2, 21.3, 62.1,
62.3, 62.6, 65.86, 65.90, 67.0, 67.2, 68.6, 69.5, 69.7,
69.9, 70.0, 71.1, 71.4, 72.7, 74.8, 74.9, 75.5, 75.9,
87.1, 98.5, 99.2, 99.7, 169.2, 169.6, 169.75, 169.85,
169.96, 170.04, 170.1, 170.2, 170.3, 170.6, 170.67,
170.69, 170.8; ESMS: m/z 1260.3 [M+Na]+.
(A) A mixture of the imidate 22 (410 mg, 0.57 mmol)
and the alcohol 33 (300 mg, 0.51 mmol)27 in 1,2-DCE
˚
(6 mL) in the presence of mol. sieves (700 mg of 3 A
powder) was treated with TMSOTf (30 lL, 0.17 mmol)
and the combined mixture stirred (0ꢁ ! rt, 30 min).
Et3N (100 lL) was introduced, the mixture was filtered
and the solvent was evaporated. The residue was sub-
jected to FC (10–50% EtOAc/hexane) to yield, presum-
ably, disaccharide 34 as a colourless oil. (B) PdCl2
(40 mg) was added to a solution of the product from
(A) in MeOH (10 mL) and 1,2-DCE (10 mL) and the
combined mixture was heated (70 ꢁC, 40 min). The sol-

J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
707
vents were evaporated and the residue subjected to FC
(10–50% EtOAc/hexanes) to yield the alcohol 35 as a
colourless oil (316 mg, 68%, two steps). The 1H and
13C NMR (CDCl3) spectra were in accord with those re-
ported in the literature.27
(27 mg, 36%) as an amorphous white powder after lyo-
1
philisation. H NMR (400 MHz, D2O) d 5.26 (d, 1H,
J1,2 = 1.8 Hz, H1II), 4.98 (dd, 1H, J2,3 = 2.4 Hz, H2II),
4.87 (d, 1H, J1,2 = 1.9 Hz, H1I), 4.60–4.55 (m, 1H,
H3II), 4.53 (dd, 1H, J2,3 = 2.3 Hz, H2I), 4.41–4.19 (m,
5H, H4I, 4II, 6aI, 6aII, 6bII), 4.15 (dd, 1H, J3,4 = 9.3 Hz,
H3I), 4.06–3.91 (m, 3H, H5I, 5II, 6bI), 3.29 (s, 3H, OCH3).
4.18. Methyl 3-O-[3-O-(2,4,6-tri-O-benzoyl-a-D-manno-
pyranosyl)-2,4,6-tri-O-benzoyl-a-D-mannopyranosyl]-2,4,
6-tri-O-benzoyl-a-D-mannopyranoside (37)
4.22. Sulfated trisaccharide (43)
(A) A mixture of trichloroacetimidate 22 (269 mg,
0.37 mmol) and the alcohol 35 (306 mg, 0.31 mmol)
was treated with TMSOTf (20 lL, 0.11 mmol) as de-
scribed for 34 to yield, presumably, the trisaccharide
36 as a colourless oil. (B) The product from (A) was
deprotected as described for 35 to yield the alcohol 37
The alcohol 37 (115 mg, 0.79 mmol) was transesterified as
described for the deprotection of 25 to yield methyl a-D-
mannopyranosyl-(1 ! 3)-a-D-mannopyranosyl-(1 ! 3)-
a-D-mannopyranoside (42)45 as a colourless oil (35 mg,
86%), used without further purification in the next step;
HRMS: m/z 519.1862 [M+H]+, 541.1646 [M+Na]+. The
trisaccharide 42 (25 mg, 48 lmol) was sulfonated as de-
scribed for the preparation of 12–15 to yield 43 (36 mg,
49%) as an amorphous white powder after lyophilisa-
tion. 1H NMR (400 MHz, D2O) d 5.26 (d, 1H,
J1,2 = 1.9 Hz, H1III), 5.22 (d, 1H, J1,2 = 1.8 Hz, H1II),
5.04 (dd, 1H, J2,3 = 2.4 Hz, H2III), 4.89 (d, 1H,
J1,2 = 1.6 Hz, H1I), 4.76–4.75 (m, 1H, H2II), 4.60–4.55
(m, 1H, H3III), 4.55 (dd, 1H, J2,3 = 3.1 Hz, H2I), 4.50
(dd, 1H, J3,4 = 9.6, J4,5 = 9.7 Hz, H4III), 4.41–4.12,
4.04–3.91 (2 m, 12H, H3II, 4I, 4II, 5I–III, 6aI–III, 6bI–III),
4.10 (dd, 1H, J3,4 = 9.5 Hz, H3I), 3.29 (s, 3H, OCH3).
1
as a colourless oil (316 mg, 70%, two steps). H NMR
(400 MHz, CDCl3) d 8.14–7.22 (m, 45 H, Ph), 6.63
(dd, 1H, J1III,2III = 1.8, J2III,3III = 3.3 Hz, H2III), 5.94
(dd, 1H, J3III,4III = 10.0, J4III,5III = 10.0 Hz, H4III), 5.84
(dd, 1H, J3II,4II = 9.9, J4II,5II = 9.9 Hz, H4II), 5.48 (dd,
1H, J3I,4I = 9.8, J4I,5I = 9.8 Hz, H4I), 5.26 (d, 1H,
J1I,2I = 1.9 Hz, H1I), 5.22 (dd, 1H, J1II,2II = 2.1,
J2II,3II = 3.0 Hz, H2II), 4.91 (d, 1H, H1III), 4.90 (dd,
1H, J2I,3I = 3.2 Hz, H2I), 4.86 (dd, 1H, J1II,2II = 1.7 Hz,
H1II). 4.67–4.63 (m, 12H, H3I–III, 5I–III, 6a,bI–III); 13C
NMR (100 MHz, CDCl3) d 166.5, 166.4, 166.2, 166.1,
165.9, 165.8, 165.6, 165.19, 165.15, 133.8, 133.7,
133.61, 133.58, 133.5, 133.2, 133.1, 130.22, 130.16,
130.1, 130.05, 130.02, 129.97, 129.91, 129.88, 129.8,
129.5, 129.4, 129.3, 129.22, 129.17, 129.1, 129.0, 128.8,
128.6, 128.53, 128.50, 128.46, 99.4, 99.2, 98.7, 76.5,
76.1, 72.4, 71.8, 71.6, 69.9, 69.7, 69.0, 68.9, 68.6, 68.5,
67.8, 63.2, 62.8, 62.4, 55.7; ESMS: m/z 1373.4
[MꢀBz+H+Na]+, 1269.4 [Mꢀ2Bz+2H+Na]+.
4.23. Methyl a-D-mannopyranosyl-(1 ! 3)-a-D-manno-
pyranosyl-(1 ! 3)-a-D-mannopyranosyl-(1 ! 3)-a-D-manno-
pyranoside (44)
The alcohol 39 (127 mg, 0.66 mmol) was transesterified
as described for the deprotection of 25 to yield the tetra-
saccharide 44 as a colourless oil (45 mg, 98%). 1H NMR
(400 MHz, CD3OD): d 3.37 (1H, s, CH3), 3.49–3.89
(18H, m), 3.93–4.00 (4H, m), 4.16–4.19 (2H, m), 4.61
(1H, d, J1,2 = 1.2 Hz, H-1I), 5.03 (1H, d, J1,2 = 1.2 Hz,
H-1), 5.07 (1H, d, J1,2 = 1.2 Hz, H-1), 5.10 (1H, d,
J1,2 = 1.2 Hz, H-1); 13C NMR (100 MHz, CD3OD) d
54.0, 61.6, 61.7, 62.0, 62.1, 66.3, 66.5, 66.6, 66.8, 67.8,
70.18, 70.23, 70.9, 71.3, 73.5, 73.8, 73.9, 74.1, 78.7,
79.3, 101.5, 102.5, 102.6, 102.7; HRMS calcd for
C25H45O21 [M+H]+ 681.2453, found 681.2390.
4.19. Methyl 3-O-{3-O-[3-O-(2,4,6-tri-O-benzoyl-a-D-man-
nopyranosyl)-2,4,6-tri-O-benzoyl-a-D-mannopyranosyl]-2,
4,6-tri-O-benzoyl-a-D-mannopyranosyl}-2,4,6-tri-O-ben-
zoyl-a-D-mannopyranoside (39)
(A) A mixture of trichloroacetimidate 22 (121 mg,
0.17 mmol) and the alcohol 37 (201 mg, 0.14 mmol)
was treated with TMSOTf (1 lL, 0.06 mmol) as de-
scribed for 34 to yield, presumably, the tetrasaccharide
38 as a colourless oil. (B) The product from (A) was
deprotected as described for 35 to yield the alcohol 39
4.24. Preparation of [3H]-heparan sulfate (HS)
1
as a colourless oil (145 mg, 54%, two steps). The H
HS (18 mg) was dissolved in hydrazine hydrate (1 mL)
containing 2% hydrazine sulfate and heated at 100 ꢁC
for 170 min to remove the N-linked acetate groups.
The mixture was cooled and co-evaporated with toluene
(3· 1 mL) and the residue was dissolved in 1 M NaCl
(2 mL) and desalted into water using a PD-10 column.
The eluate from the PD-10 column was passed through
a Dowex 50 column (1 · 5 cm, Na+ form) to remove
remaining impurities. The eluate was concentrated and
the residue dissolved in 500 mM NaHCO3 (1.5 mL) con-
taining 10% MeOH, chilled to 0 ꢁC and a solution of
[3H]acetic anhydride (16 lmol, 8 mCi) in toluene
(80 lL) added. After 3 h excess unlabelled acetic anhy-
dride (50 lL) was added to ensure complete re-acetyla-
tion of amino groups and the reaction was adjusted
to alkaline pH with 0.5 M Na2CO3. The mixture was
and 13C NMR (CDCl3) spectra were in accord with
those already reported in the literature.27
4.20. Methyl a-D-mannopyranosyl-(1 ! 3)-a-D-manno-
pyranoside (40)
The alcohol 35 (10 mg, 0.10 mmol) was transesterified as
described for the deprotection of 25 to yield the disac-
charide 40 as a colourless oil (3 mg, 85%), identical by
1H NMR to that reported in the literature.42–44
4.21. Sulfated disaccharide (41)
The disaccharide 40 (25 mg, 70 lmol) was sulfonated as
described for the preparation of 12–15 to yield 41

708
J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
subjected to size exclusion chromatography (Sephacryl
S-200, 1.6 · 58 cm) equilibrated and eluting with 10%
EtOH. Fractions containing high molecular weight HS
fragments were pooled, evaporated and dissolved in
10% EtOH to an approximate concentration of 5 mg/
mL. The concentration of the [3H]HS preparation was
accurately determined (5.6 mg/mL) using the DMB as-
say for glycosaminoglycans.46 Unlabelled HS from the
same source was used as standard.
(pH 7.0) and the 100 lL transferred to a Microcon
YM-10 concentrator which was centrifuged at approxi-
mately 14,000g for 5 min. The solution that passed
through the membrane (filtrate) was retained. This sam-
ple was considered the time = 0 sample. The assays (now
80 lL in volume) were allowed to react at 37 ꢁC for 4 h
and then the filtration step was repeated for three ali-
quots of 20 lL from each assay. The time = 0 filtrate
3
and the three 4-h filtrate samples were counted for H.
The difference between the time = 0 and the averaged
4-h samples gave the amount of heparanase activity. Rel-
ative inhibition assays were run with a heparanase stan-
dard assay which was identical to the assay composition
above except no inhibitor was present and the amount of
heparanase inhibition in the other assays determined by
comparison with this standard. In assays containing a
higher concentration of HS than 5 lM the volume of as-
say mixture quenched was proportionally reduced. For
example, a 10 lM assay would have a 10 lL aliquot
quenched in 90 lL of 10 mM phosphate buffer (pH
7.0). This was done to prevent concentration effects dur-
ing ultrafiltration which can occur at high solute concen-
trations. The results are presented in Table 1.
4.25. Isolation and purification of heparanase from human
platelets
Heparanase was purified to homogeneity from human
platelets using three chromatography steps: Concanava-
lin A Sepharose, Sulfopropyl Sepharose and Sephacryl
S-200 HR. The platelets were extracted by repeatedly
freezing the cells in buffer consisting of 15 mM dimeth-
ylglutarate, pH 6.0, 500 mM NaCl. After centrifugation
to pellet the cell debris the extract was adjusted to 1 mM
CaCl2, 1 mM MnCl2 and 0.2% Triton X-100. The ex-
tract was then applied to a 100 mL Concanavalin A Se-
pharose column equilibrated with buffer consisting of
15 mM dimethylglutarate, pH 6.0, 500 mM NaCl and
0.2% Triton X-100. The Concanavalin A column was
at 4 ꢁC. Once the extract had been loaded, the column
was washed with buffer minus Triton X-100 for one col-
umn volume and the column temperature was increased
to 23 ꢁC. Once equilibrated to 23 ꢁC, glycosylated pro-
teins (including heparanase) were eluted from the col-
umn by washing the column with buffer containing
20% methyl a-D-mannopyranoside. Subsequent chroma-
tography columns in the procedure were run at room
temperature. The eluate from the Concanavalin A col-
umn was diluted 1/5 into 10 mM Na phosphate, pH
7.0, and then applied to a 150 mL Sulfopropyl Sephar-
ose column equilibrated with 20 mM Na phosphate,
pH 7.0. The column was then eluted with a 1.3 L gradi-
ent of 0–750 mM NaCl in 20 mM Na phosphate, pH
7.0. Fractions containing heparanase were pooled and
concentrated using centrifugal concentrators (Centriplus
YM-30, Millipore) to approximately 5 mL. The concen-
trated sample was loaded onto Sephacryl S-200 HR col-
umns (two 500 mL columns in series) equilibrated with
15 mM dimethylglutarate, pH 6.0, 500 mM NaCl. The
same buffer was used to elute this column. Fractions
were analysed by gel electrophoresis and those contain-
ing pure heparanase were pooled. The heparanase sam-
ple was then concentrated using Centriplus
concentrators and desalted into the final buffer
(10 mM dimethylglutarate, pH 6.0). The purified hepa-
ranase prepared this way was stored at ꢀ80 ꢁC until use.
Acknowledgment
We thank Dr. Ian Bytheway for useful discussions and
critical evaluation of this manuscript.
References and notes
1. Rosenthal, M. A.; Rischin, D.; McArthur, G.; Ribbons,
K.; Chong, B.; Fareed, J.; Toner, G.; Green, M. D.;
Basser, R. L. Ann. Oncol. 2002, 13, 770.
2. Basche, M.; Gustafson, D. L.; Holden, S. N.; O’Bryant, C.
L.; Gore, L.; Witta, S.; Schultz, M. K.; Morrow, M.;
Levin, A.; Creese, B. R.; Kangas, M.; Roberts, K.;
Nguyen, T.; Davis, K.; Addison, R. S.; Moore, J. C.;
Eckhardt, S. G. Clin. Cancer Res. 2006, 12, 5471.
3. Ferro, V.; Liu, L.; Hammond, E.; Dredge, K.; Bytheway,
I.; Li, C.; Johnstone, K.; Karoli, T.; Copeman, E.; Davis,
K.; Gautam, A. Sem. Thromb. Haemost. 2007, 33, 557.
4. Parish, C. R.; Freeman, C.; Brown, K. J.; Francis, D. J.;
Cowden, W. B. Cancer. Res. 1999, 59, 3433.
5. Hazel, S. J. J. Lab. Clin. Med. 2003, 141, 217.
6. Joyce, J. A.; Freeman, C.; Meyer-Morse, N.; Parish, C. R.;
Hanahan, D. Oncogene 2005, 24, 4037.
7. Ilan, N.; Elkin, M.; Vlodavsky, I. Int. J. Biochem. Cell
Biol. 2006, 38, 2018.
8. Hammond, E.; Bytheway, I.; Ferro, V. In New Develop-
ments in Therapeutic Glycomics; Delehedde, M., Lortat-
Jacob, H., Eds.; Research Signpost: Trivandrum, India,
2006; pp 251–282.
9. Ferro, V.; Hammond, E.; Fairweather, J. K. Mini Rev.
Med. Chem. 2004, 4, 159.
4.26. Heparanase inhibition assays
Reactions were set up in a volume of 100 lL containing
40 mM sodium acetate buffer (pH 5.0), BSA (0.1 mg/
mL), heparanase (90 ng), 5 lM [3H]HS and various con-
centrations of sulfated oligosaccharide inhibitors. Ini-
tially, all components except the [3H]HS were allowed
to equilibrate for 10 min at 22 ꢁC. The assays were then
initiated by adding the [3H]HS and immediately 20 lL
was taken, quenched with 80 lL of 10 mM phosphate
10. McKenzie, E. A. Br. J. Pharmacol. 2007, 151, 1.
11. Miao, H. Q.; Liu, H.; Navarro, E.; Kussie, P.; Zhu, Z.
Curr. Med. Chem. 2006, 13, 2101.
12. Wall, D.; Douglas, S.; Ferro, V.; Cowden, W.; Parish, C.
Thromb. Res. 2001, 103, 325.
13. Demir, M.; Iqbal, O.; Hoppensteadt, D. A.; Piccolo, P.;
Ahmad, S.; Schultz, C. L.; Linhardt, R. J.; Fareed, J. Clin.
Appl. Thromb. Hemost. 2001, 7, 131.

J. K. Fairweather et al. / Bioorg. Med. Chem. 16 (2008) 699–709
709
14. Piccolo, P.; Iqbal, O.; Demir, M.; Ma, Q.; Gerbutavicius,
R.; Fareed, J. Clin. Appl. Thromb. Hemost. 2001, 7, 149.
15. Francis, D. J.; Parish, C. R.; McGarry, M.; Santiago, F.
S.; Lowe, H. C.; Brown, K. J.; Bingley, J. A.; Hayward, I.
P.; Cowden, W. B.; Campbell, J. H.; Campbell, G. R.;
Chesterman, C. N.; Khachigian, L. M. Circ. Res. (Online)
2003, 92, e70.
16. Nyberg, K.; Ekblad, M.; Bergstro¨m, T.; Freeman, C.;
Parish, C. R.; Ferro, V.; Trybala, E. Antiviral Res. 2004,
63, 15.
30. Almond, A.; Bunkenborg, J.; Franch, T.; Gotfredsen, C.
H.; Duus, J. Ø. J. Am. Chem. Soc. 2001, 123, 4792.
31. Karoli, T.; Liu, L.; Fairweather, J. K.; Hammond, E.; Li,
C. P.; Cochran, S.; Bergefall, K.; Trybala, E.; Addison, R.
S.; Ferro, V. J. Med. Chem. 2005, 48, 8229.
32. Freeman, C.; Parish, C. R. Biochem. J. 1998, 330, 1341.
33. Nakajima, M.; Irimura, T.; Nicolson, G. L. Anal.
Biochem. 1986, 157, 162.
34. Behzad, F.; Brenchley, P. E. Anal. Biochem. 2003, 320,
207.
17. Lee, E.; Pavy, M.; Young, N.; Freeman, C.; Lobigs, M.
Antiviral Res. 2006, 69, 31.
18. Adams, Y.; Freeman, C.; Schwartz-Albiez, R.; Ferro, V.;
Parish, C. R.; Andrews, K. T. Antimicrob. Agents
Chemother. 2006, 50, 2850.
35. Huang, K. S.; Holmgren, J.; Reik, L.; Lucas-McGady, D.;
Roberts, J.; Liu, C. M.; Levin, W. Anal. Biochem. 2004,
333, 389.
36. Nardella, C.; Steinkuhler, C. Anal. Biochem. 2004, 332,
368.
¨
19. Yu, G.; Gunay, N. S.; Linhardt, R. J.; Toida, T.; Fareed,
J.; Hoppensteadt, D. A.; Shadid, H.; Ferro, V.; Li, C.;
Fewings, K.; Palermo, M. C.; Podger, D. Eur. J. Med.
Chem. 2002, 37, 783.
37. Tsuchida, S.; Podyma-Inoue, K. A.; Yanagishita, M.
Anal. Biochem. 2004, 331, 147.
38. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099.
20. Ferro, V.; Fewings, K.; Palermo, M. C.; Li, C. Carbohydr.
Res. 2001, 332, 183.
21. Ferro, V.; Li, C.; Fewings, K.; Palermo, M. C.; Linhardt,
R. J.; Toida, T. Carbohydr. Res. 2002, 337, 139.
22. Parolis, L. A. S.; Parolis, H.; Kenne, L.; Meldal, M.; Bock,
K. Carbohydr. Res. 1998, 309, 77.
23. Cochran, S.; Li, C.; Fairweather, J. K.; Kett, W. C.;
Coombe, D. R.; Ferro, V. J. Med. Chem. 2003, 46, 4601.
24. Gu, G.; Wei, G.; Du, Y. Carbohydr. Res. 2004, 339, 1155.
25. Namme, R.; Mitsugi, T.; Takahashi, H.; Ikegami, S.
Tetrahedron Lett. 2005, 46, 3033.
39. Bisswanger, H. Enzyme Kinetics: Principles and Methods;
Wiley-VCH: Weinheim, Germany, 2002.
40. Okada, Y.; Yamada, S.; Toyoshima, M.; Dong, J.;
Nakajima, M.; Sugahara, K. J. Biol. Chem. 2002, 277,
42488.
41. Hulett, M. D.; Hornby, J. R.; Ohms, S. J.; Zuegg, J.;
Freeman, C.; Gready, J. E.; Parish, C. R. Biochemistry
2000, 39, 15659.
42. Sandstro¨m, C.; Berteau, O.; Gemma, E.; Oscarson, S.;
Kenne, L.; Gronenborn, A. M. Biochemistry 2004, 43,
13926.
26. Ogawa, T.; Yamamoto, H. Carbohydr. Res. 1982, 104,
271.
27. Chen, L.; Kong, F. J. Carbohydr. Chem. 2002, 21, 341.
28. Chen, L.; Zhu, Y.; Kong, F. Carbohydr. Res. 2002, 337,
383.
43. Ogawa, T.; Sasajima, K. Carbohydr. Res. 1981, 93, 53.
44. Ogawa, T.; Sasajima, K. Carbohydr. Res. 1981, 97,
205.
45. Garegg, P. J.; Olsson, L.; Oscarson, S. Bioorg. Med.Chem.
1996, 4, 1867.
29. Fairweather, J. K.; Karoli, T.; Ferro, V. Bioorg. Med.
Chem. 2004, 12, 6063.
46. Farndale, R. W.; Buttle, D. J.; Barrett, A. J. Biochim.
Biophys. Acta 1986, 883, 173.