Synthesis and biology of bis-xylosylated dihydroxynaphthalenes

Article abstract of DOI:10.1016/j.bmc.2007.02.020

The 10 analogous bis-xylosylated dihydroxynaphthalenes have been synthesized and their chemical and biological properties investigated. The yield of the xylosylation reactions can be correlated to the electrostatic potential, and thus to the nucleophilicity, for the oxygen atoms of the dihydroxynaphthalenes. The bis-xylosylated compounds were more stable compared to the mono-xylosylated ones. They initiate priming of glycosaminoglycan chains to less extent but the priming proceeds in two directions. Contrary to the mono-xylosylated analogs, the tested compounds did not show any antiproliferative properties.

Full text of DOI:10.1016/j.bmc.2007.02.020

Bioorganic & Medicinal Chemistry 15 (2007) 2868–2877
Synthesis and biology of bis-xylosylated dihydroxynaphthalenes
Richard Johnsson,a Katrin Manib and Ulf Ellervika,*
aOrganic Chemistry, Lund University, PO Box 124, SE-221 00 Lund, Sweden
bDepartment of Experimental Medical Science, Division of Neuroscience, Lund University,
Biomedical Center A13, SE-221 84 Lund, Sweden
Received 31 October 2006; revised 1 February 2007; accepted 9 February 2007
Available online 13 February 2007
Abstract—The 10 analogous bis-xylosylated dihydroxynaphthalenes have been synthesized and their chemical and biological
properties investigated. The yield of the xylosylation reactions can be correlated to the electrostatic potential, and thus to the
nucleophilicity, for the oxygen atoms of the dihydroxynaphthalenes. The bis-xylosylated compounds were more stable compared
to the mono-xylosylated ones. They initiate priming of glycosaminoglycan chains to less extent but the priming proceeds in two
directions. Contrary to the mono-xylosylated analogs, the tested compounds did not show any antiproliferative properties.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
sulfate (HS) chains is low or undetectable. It has recent-
ly been shown that in polarized cells xyloside based
GAG chains secreted basolaterally are more intensely
sulfated than their apical counterparts.1 Increased yields
of HS can be obtained when the aglycon of the xylosides
comprises aromatic, polycyclic structures, such as naph-
thol-derivatives. The xyloside-primed GAG chains can
be retained inside the cells but are usually mainly secret-
ed into the medium and possibly also re-internalized.
b-D-Xyloside-primed HS chains have interesting biolog-
ical properties, such as activation of basic fibroblast
growth factor,2 antithrombotic effects,3 and growth inhi-
bition of transformed cells.4–6
Proteoglycans are composed of glycosaminoglycan
(GAG) chains covalently attached to a core protein
(Fig. 1). The first step in GAG assembly is xylosylation
of a serine residue and a specific linker tetrasaccharide,
GlcA(b1-3)Gal(b1-3)Gal(b1-4)Xylb, is assembled and
serves as an acceptor for elongation of GAG chains.
The resultant GAG precursor polymers are subsequent-
ly modified, presumably during chain elongation,
through a series of reactions such as N-deacetylation/
N-sulfation, epimerization of some GlcA residues to
IdoA units, and also O-sulfation at various positions.
By using xylosides with hydrophobic aglycon, that can
penetrate cell membranes, as primers, the biosynthesis
of GAG chains can be artificially initiated independently
of core protein synthesis. The xylosides initiate GAG
synthesis by serving as acceptors in the first galactosyla-
tion step. The composition of the GAG chains assem-
bled on the xyloside primers depends on the structure
of the aglycon, which may reflect selective partitioning
of primers into different intracellular compartments or
into different branches of biosynthetic pathways. In
most cases, priming of chondroitin sulfate/dermatan sul-
fate (CS/DS) dominates and synthesis of free heparan
The HS-priming 2-(6-hydroxynaphthyl)-b-D-xylopyr-
anoside (5d) has previously been reported to selectively
inhibit growth of transformed or tumor derived cells
in vitro as well as in vivo.5 Treatment with this xyloside
at a pharmacologically relevant dose reduced the aver-
age tumor load by 70–97% in SCID mice. Attempts to
determine the mechanism for the selective growth inhibi-
tion have also been made.4,5 These results suggest that
(i) the priming of HS synthesis is required for selective
growth inhibition and (ii) the effect on transformed cells
is not caused by the xyloside itself but by products de-
rived from the priming of HS on the xyloside. Further-
more, the bioactivity is dependent on (iii) the hydroxyl
substitution pattern in the naphthalene rings of the xylo-
side and (iv) nuclear targeting of the xyloside-derived
products. However, the knowledge of structure recogni-
tion of the proteoglycan synthesizing enzymes is still
scarce.
Keywords: Carbohydrates; Dihydroxynaphthalene; Glycosylation;
Xylose.
*
Corresponding author. Tel.: +46 46 2228220; fax: +46 46 2228209;
e-mail: ulf.ellervik@organic.lu.se
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.02.020

R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
2869
HN
HN
OH
OH
OH
OH
OH
OH
COOH
COOH
O
O
O
O
O
O
O
O
O
O
O
HO
O
NH
O
HO
O
HO
AcHN
OH
OH
OH
OH
OH
n
O
Linker tetrasaccharide
Core Protein
Figure 1. Glycosaminoglycan chains consist of a linker tetrasaccharide unit (GlcA(b1-3)Gal(b1-3)Gal(b1-4)Xylb) coupled to serine residues of the
protein chain. The general structure is then modified through N-deacetylation/N-sulfation, epimerization, and O-sulfation.
A series of mono-xylosylated dihydroxynaphthalenes
(1–10d, e) which gave insight into the mechanisms of
priming and antiproliferative properties of these com-
pounds have been reported earlier.7 In that study a cor-
relation between the antiproliferative effect toward some
cell lines and the polarity of the compounds was found.
However, some compounds (e.g., 2-(6-hydroxynaph-
thyl)-b-D-xylopyranoside) were clearly different and
gave a selective antiproliferative effect toward trans-
formed cells (e.g., T24). It was also found that some
compounds, especially the 1-(4-hydroxynaphthyl)-b-D-
xylopyranoside (1d), were further hydroxylated in the
Ham’s F-12 medium used for cell experiments. In the
present study, we investigate the stability, the GAG
priming capability, and the antiproliferative effects of a
series of bis-xylosylated compounds. The bis-xylosylated
compounds expand the variation in properties, such as
polarity, and also give the possibility for priming at
two different positions in the same compound.
anoside) (4b), and 1,2-dihydroxynaphthalene bis(2,3,4-
tri-O-acetyl-b-D-xylopyranoside) (7b) could not be syn-
thesized by this method. In the syntheses of compounds
1b, 2b, 7b, and 8b we observed mono-xylosylated mate-
rial, whereas the attempted syntheses of compounds 3b
and 4b gave complete degradation of the starting mate-
rial (Chart 1).
To further investigate xylosylation of dihydroxynaph-
thalenes, the electrostatic potentials and the distances
between the phenolic oxygen atoms for the ten dihydr-
oxynaphthalenes were calculated using density function-
al theory at the B3LYP/6-31G* level and default settings
in Spartan ’02 for Macintosh (Table 1).10 As expected,
the average electrostatic potential was correlated to the
yield in the glycosylation reactions, that is, higher elec-
trostatic potential, which correlates to the nucleophilic-
ity of the oxygen atoms, gave better yields in the
reaction (Fig. 2). However, the isolation of mono-
xylosylated material in some cases also indicates steric
effects.
2. Results and discussion
To xylosylate the unreactive compounds 3a, 4a, and 7a,
a trichloroacetimidate donor, activated with TMSOTf,
was used.11,12 Trichloroacetimidate donors are well
known to be more active compared to the analogous
acetylated compounds and thus better suited for the less
reactive phenols.8 This gave 4b in 58% and 7b in 33%,
but still 3b could not be synthesized. The problem is
probably the strong hydrogen bond between the two
hydroxyl groups of the 1,8-dihydroxynaphthalene which
renders this compound less nucleophilic. Since com-
pound 3a is highly acidic (pKa 6.71),13 the synthesis
was instead performed under basic conditions. 3b was
then synthesized using the corresponding bromosugar
and phase transfer conditions, with NaOH as base, to
give a yield of 13%.14–16 The synthesis of 7b was also
performed under these conditions giving 14% yield. This
indicates that 3a is the substance most affected by the
acidity of the hydroxyl groups. The deprotection was
performed using methanolic NaOMe (0.05 M). Prob-
lems with high crystallinity affected the purification of
some compounds. 1c, 2c, and 5c could thus not be puri-
fied using normal phase column chromatography since
they were virtually insoluble in most used solvents (with
the exception of DMSO), and therefore they were puri-
fied using reverse phase HPLC. All compounds were
additionally purified on reverse phase HPLC prior to
biological evaluation.
Glycosylation of dihydroxynaphthalenes is a difficult
task due to the strongly activated aromatic system.8
Depending on the substitution pattern, electrophilic aro-
matic substitution competes with glycosylation of the
hydroxyl groups. Another problem is steric hindrance
as well as intramolecular hydrogen bonds which lower
the activity toward glycosylation in the 1,8-, 2,3-, and
1,2-dihydroxynaphthalenes (3a, 4a, 7a). The synthesis
of the acetylated bis-xylosides was therefore first per-
formed using peracetylated xylose activated by
BF3OEt2, with addition of Et3N to minimize anomeriza-
tion, in CH2Cl2.9 This gave excellent yields (>80%) for
2,6-dihydroxynaphthalene
xylopyranoside) (5b),
bis(2,3,4-tri-O-acetyl-b-D-
2,7-dihydroxynaphthalene
bis(2,3,4-tri-O-acetyl-b-D-xylopyranoside) (6b), 1,6-di-
hydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-D-xylopyr-
anoside) (9b), and 1,7-dihydroxynaphthalene bis(2,3,
4-tri-O-acetyl-b-D-xylopyranoside) (10b). However, 1,4-
dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-D-xylo-
pyranoside) (1b), 1,5-dihydroxynaphthalene bis(2,3,
4-tri-O-acetyl-b-D-xylopyranoside) (2b), and 1,3-dihy-
droxynaphthalene
bis(2,3,4-tri-O-acetyl-b-D-xylopyr-
anoside) (8b) were synthesized in low to moderate
yields (22–64%), whereas 1,8-dihydroxynaphthalene
bis(2,3,4-tri-O-acetyl-b-D-xylopyranoside) (3b), 2,3-di-
hydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-D-xylopyr-

2870
R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
OR1
OR1
OR1 OR2
R1O
R2O
OR2
OR2
2a: R1 = R2 = H
1a: R1 = R2 = H
3a: R1 = R2 = H
4a: R1 = R2 = H
1b: R1 = R2 = Xyl(OAc)3 2b: R1 = R2 = Xyl(OAc)3 3b: R1 = R2 = Xyl(OAc)3 4b: R1 = R2 = Xyl(OAc)3
1c: R1 = R2 = Xyl
2c: R1 = R2 = Xyl
3c: R1 = R2 = Xyl
4c: R1 = R2 = Xyl
1d: R1 = Xyl, R2 = H
2d: R1 = Xyl, R2 = H
3d: R1 = Xyl, R2 = H
4d: R1 = Xyl, R2 = H
OR1
OR1
R1O
OR2
R2O
R1O
R2O
8a: R1 = R2 = H
OR2
5a: R1 = R2 = H
6a: R1 = R2 = H
7a: R1 = R2 = H
5b: R1 = R2 = Xyl(OAc)3 6b: R1 = R2 = Xyl(OAc)3 7b: R1 = R2 = Xyl(OAc)3 8b: R1 = R2 = Xyl(OAc)3
5c: R1 = R2 = Xyl
6c: R1 = R2 = Xyl
7c: R1 = R2 = Xyl
8c: R1 = R2 = Xyl
5d: R1 = Xyl, R2 = H
6d: R1 = Xyl, R2 = H
7d: R1 = Xyl, R2 = H
7e: R1 = H, R2 = Xyl
8d: R1 = Xyl, R2 = H
8e: R1 = H, R2 = Xyl
OR1
OR1
OR2
O
HO
Xyl =
HO
OR2
OH
9a: R1 = R2 = H
10a: R1 = R2 = H
9b: R1 = R2 = Xyl(OAc)3 10b: R1 = R2 = Xyl(OAc)3
O
OAc
AcO
9c: R1 = R2 = Xyl
10c: R1 = R2 = Xyl
Xyl(OAc)3
=
AcO
9d: R1 = Xyl, R2 = H
9e: R1 = H, R2 = Xyl
10d: R1 = Xyl, R2 = H
10e: R1 = H, R2 = Xyl
Chart 1. Dihydroxynaphthalenes (1–10a) and the analogous bis-xylosylated compounds (1–10b, c) as well as the mono-xylosylated compounds
(1–10d, e) described in this paper.
Table 1. Yield of bis-xylosylation of dihydroxynaphthalenes, average
electrostatic potential and distance between the two oxygen atoms of
the naphthalene unit
Yield (%)
100
Compound Yield of
Average
bis-xylosylation electrostatic
Distance
(nm)
80
60
40
20
0
(%)
potential (kcal/mol)
1a
2a
3a
4a
5a
6a
7a
8a
9a
10a
64
55
0
ꢀ30.3
ꢀ28.0
ꢀ23.6
ꢀ23.1
ꢀ32.0
ꢀ30.9
ꢀ23.7
ꢀ27.5
ꢀ29.9
ꢀ28.4
0.56
0.60
0.26
0.27
0.78
0.73
0.27
0.48
0.63
0.51
0
82
81
0
22
95
96
-34
-32
-30
-28
-26
-24
-22
Average electrostatical potential (kcal/mol)
Figure 2. Yield of the bis-xylosylation reactions versus average
electrostatic potential (kcal/mol) of the two oxygen atoms of the
naphthalene unit.
Gradient HPLC retention times can be used to substi-
tute log P values in biological evaluations.17 The gradi-
ent HPLC retention times for the compounds were
measured using a C-18 column and a mobile phase of
water (0.1% TFA) with a gradient of MeCN from
1 min increasing by 1.2% per minute. The retention
times were measured for three separate runs per com-
pound, and the calculated mean retention times are pre-
sented in Table 2. As a comparison retention times for
the mono-xylosylated compounds 1–10d, e are shown.
As expected, the bis-xylosylated compounds, carrying
two polar carbohydrate residues, show shorter retention
times. A substantial variation of the polarity due to the
substitution pattern can also be seen. For example, the
compounds with 1,2-, 1,3- or 1,8-substitution are more
unpolar, while 1,4- and 2,6-substitution gives polar
compounds. Obviously, the polarity is connected to
the distance between the polar groups, that is, the carbo-
hydrate residues. When the distance between the two
phenolic oxygen atoms of the naphthalene systems was
plotted against the HPLC retention times a correlation
was found as indicated in Figure 3.
To evaluate the stability of the compounds, the xylosides
were dissolved in DMSO/water (1:1, 20 mM), diluted
with cell growth medium (Ham’s F-12 medium) to
0.2 mM, and incubated at 37 ꢁC. Samples were taken
at 0, 1, 3, 24, 48, 72, and 96 h, and the amounts of xylo-
side, as compared to 4,40-dihydroxybiphenyl used as an

R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
2871
Table 2. HPLC retention times (min) for bis- and mono-xylosylated
dihydroxynaphthalenes
compounds 8d and 10e showed slight decomposition
and compounds 1d and 7e were totally decomposed after
less than 80 h.7 The addition of the second xylose moiety
thus seems to give a shielding effect that prevents oxida-
tion as compared to the mono-xylosylated analogs.
Bis-xylosylated Retention
compound
Mono-xylosylated Retention
compound
time (min)
time (min)
1c
2c
3c
4c
5c
6c
7c
15.93 0.00
17.03 0.04
24.17 0.03
23.54 0.07
15.86 0.01
18.09 0.01
24.52 0.02
1d
2d
3d
4d
5d
6d
7d
7e
8d
8e
9d
9e
19.60 0.01
18.70 0.01
28.54 0.02
25.73 0.03
17.94 0.09
20.25 0.01
27.93 0.02
27.97 0.03
21.99 0.02
24.33 0.01
20.12 0.06
19.03 0.04
24.01 0.03
22.44 0.01
To test the xyloside’s ability to prime GAG synthesis,
T24 cells (human bladder carcinoma cells) were incubat-
ed with 100 lM xyloside and [35S]sulfate followed by
isolation and size separation of free GAG chains. All
cells secreted alkali sensitive proteoglycans (example
shown in Fig. 4, Pool I). However, treatment with some
xylosides also initiated synthesis of free GAG chains
(Fig. 4, Pool II). The proportion of GAG-priming is giv-
en as the integrated value of fractions in Pool II (Table
3) divided by the integrated value for fractions of
untreated cells. In addition we have also tested selected
mono-xylosylated compounds and the results are shown
in Table 3. Apart from compounds 2c and 5c, the bis-
xylosylated compounds seem to be less capable of
GAG priming compared to the mono-xylosylated ones.
No correlations between the GAG priming capability
and the polarity of the compounds or the distance be-
tween the xylose residues have been found.
8c
9c
20.56 0.00
18.02 0.01
10c
22.33 0.01 10d
10e
HPLC retention time (min)
30
a
25
20
15
The separation procedure of the GAG chains also indi-
cates the relative size of different GAG chains, that is,
longer chains are eluted faster from the column. As indi-
cated in Figure 5 and Table 3, the GAG chains from
bis-xylosylated dihydroxynaphthalenes elute faster com-
pared to the mono-xylosylated analogs, that is, the
GAG chains are longer. This is probably due to priming
in two directions, even if it is not possible from this
experiment to exclude priming of longer chains in only
one direction.
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Distance between phenolic oxygen atoms (nm)
In order to verify bidirectional priming by dihydroxy-
naphthalenes, the xyloside-primed GAG chains synthe-
sized by compounds 2c, 5c, and 9c were subjected to
periodate oxidation-alkaline elimination that would
selectively cleave the xylose structure. The samples were
then rechromatographed on Superose 6. As expected,
HPLC retention times (min)
30
b
25
20
15
Radioactivity
(dpm)
2500
Pool II
Pool I
2000
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1500
1000
500
0
Distance between phenolic oxygen atoms (nm)
Figure 3. HPLC retention time correlated to the distance between the
phenolic oxygen atoms for (a) bis-xylosylated compounds and (b)
mono-xylosylated compounds.
internal standard, were measured by HPLC using a C-18
column and a mobile phase of water (0.1% TFA) with a
gradient of MeCN from 1 min increasing by 1.2% per
minute. No decomposition of any xyloside was observed,
even after 96 h. This is to be compared to earlier data for
the analogous mono-xylosylated compounds where
10
20
30
40
50
60
Fraction
Figure 4. Priming of GAG chains in T24 cells incubated with 5c. Pool
II contains xyloside-primed GAG chains. The dashed line show the
result for untreated cells.

2872
R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
Table 3. GAG-Priming of bis- and mono-xylosylated compounds and antiproliferative activity (ED50, lM) of bis-xylosylated dihydroxynaphtha-
lenes towards T24 cells
Bis-xylosylated
compound
Priming
Peak maximum
(fraction)
Peak maximum
(periodate cleavage)
Antiproliferative
activity (ED50, lM)
Mono-xylosylated
compound
Priming
Peak
maximum
(fraction)
1c
2c
3c
4c
5c
6c
7c
8c
1.2
8.7
1.6
2.5
8.8
1.3
1.0
1.4
35
35
35
33
34
33
32
35
400
400
1d
2d
7.5
7.7
42
44
42
44
ꢁ400
5d
6d
7.2
5.7
43
42
450
ꢁ400
8d
8e
2.7
4.3
38
40
9c
10c
3.6
2.7
35
35
43
priming p-nitrophenyl b-D-xylopyranoside protects nor-
mal fibroblasts against apoptosis through PDGF-BB-in-
duced GAG secretion.18 The structure of the GAG
chains initiated on the antiproliferative xylosides may
be different depending on the aglycon structure and this
may endow the GAG products with the ability to enter
the nucleus and exert an antiproliferative response.
Another explanation for the lack of antiproliferative
activity of these compounds could be due to either too
high polarity or too big size that makes the transport
to the nuclei impossible. Polyhydroxylated naphthalenes
are well known for their toxic effects, caused by redox cy-
cling between semiquinones and quinones resulting in
superoxide radicals and apoptosis.19–21 Another possibil-
ity is thus that the bis-xylosylated analogs, due to the
lack of a free hydroxyl group, may not be able to induce
toxic effects and therefore are less growth inhibiting.
Figure 5. Priming of GAG chains in T24 cells incubated with bis-
xyloside 5c (solid line) compared to the mono-xyloside 5d (dashed
line). The dotted line shows the result from periodate oxidation-
alkaline elimination of GAG chains primed by 5c.
3. Conclusions
We have synthesized a collection of the ten analogous
bis-xylosylated dihydroxynaphthalenes and investigated
their chemical and biological properties. The yield of the
xylosylation reactions could be correlated to the electro-
static potential for the oxygen atoms of the dihydroxy-
naphthalenes. All bis-xylosylated compounds were
more polar, compared to the mono-xylosylated analogs,
and the polarity could be correlated with the distance
between the polar carbohydrate residues. The bis-
xylosylated compounds were also more stable compared
to the mono-xylosylated ones. Our results also indicate
that these compounds, except for compounds 2c and
5c, initiate priming to a lesser extent than their mono-
xylosylated analogs but that the priming proceeds in
two directions. None of the tested compounds showed
antiproliferative properties.
this resulted in shorter chains, similar to those from
monodirectional priming (Fig. 5 and Table 3).
Finally we investigated the growth inhibiting properties
of a selection of the bis-xylosylated dihydroxynaphtha-
lenes on T24 cells. Despite the fact that several of the
mono-xylosylated analogs show antiproliferative activi-
ties in the order of 25–100 lM, none of the bis-xylosylat-
ed analogs gave any strong growth inhibition (Table 3).
The reasons why bis-xylosylated analogs did not exert
antiproliferative activity could be several fold. It may
be due to the type and the composition of the GAG
chains produced by the xylosides. It has earlier been
shown that priming of HS is required but insufficient
for the antiproliferative activity. It has also been shown
that HS chains primed by the antiproliferative xylosides
are degraded by nitric oxide at GlcNH3þ. That results in
anhydromannose containing degradation products
localized to the nuclei in growth inhibited cells. The anti-
proliferative effect has also been shown to be accompa-
nied by increased apoptosis in tumor derived cells.7
Furthermore, other studies have shown that CS/DS
4. Experimental
4.1. Chemistry
NMR spectra were recorded with a Bruker DRX
400 MHz. 1H NMR spectra were assigned using

R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
2873
2D-methods (COSY). Chemical shifts are given in ppm
downfield from the signal for Me4Si, with reference to
residual CHCl3 or DMSO. Reactions were monitored
by TLC using alumina plates coated with silica gel and
visualized using either UV light or by charring with
para-anisaldehyde. Preparative chromatography was
OAc). 13C NMR (CDCl3): d 170.03, 169.98, 169.7,
152.4, 127.1, 125.8, 116.6, 109.7, 98.8, 70.6, 70.1, 68.5,
62.0, 21.0, 20.9. HRMS calcd for C32H37O16 (M+H):
677.2082; found: 677.2089.
4.1.4. 1,5-Dihydroxynaphthalene bis(b-D-xylopyranoside)
(2c). Same method as for 1c. Purified using HPLC and
˚
performed with silica gel (35–70 lm, 60 A). CH2Cl2
22
D
was dried on Al2O3. Known and commercially available
compounds were in agreement with previously pub-
lished data (e.g., NMR). Compounds 1–10d, e were syn-
thesized as described elsewhere.22
then lyophilized to give 2c (13 mg, 49%). ½aꢂ ꢀ77 (c
1
0.41, DMSO-d6). H NMR (DMSO-d6): d 7.94 (d, 2H,
J = 8.5 Hz, H-40, H-80), 7.39 (t, 2H, J = 7.8 Hz, H-30,
H-70), 7.14 (d, 2H, J = 7.4 Hz, H-20, H-60), 5.49 (d,
2H, J = 5.5 Hz, OH), 5.16 (d, 2H, J = 4.9 Hz, OH),
5.09 (d, 2H, J = 4.9 Hz, OH), 5.01 (d, 2H, J = 7.5 Hz,
H-1), 3.79 (dd, 2H, J = 11.1, 5.1 Hz, H-5), 3.38–3.46
(m, 4H, H-2, H-4), 3.25–3.33 (m, 4H, H-3, H-5). 13C
NMR (DMSO-d6): d 152.5, 126.5, 125.3, 115.9, 109.7,
101.6, 76.3, 73.2, 69.4, 65.7. HRMS calcd for
C20H25O10 (M+H): 425.1448; found: 425.1442.
4.1.1. 1,4-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-
D-xylopyranoside)
(1b). 1,4-Dihydroxynaphthalene
(50 mg, 0.31 mmol) and 1,2,3,4-tetra-O-acetyl-b-D-xylo-
pyranoside (397 mg, 1.25 mmol) were dissolved in
CH2Cl2 (2.5 mL) and Et3N (0.03 mL, 0.21 mmol) was
added. The mixture was stirred under nitrogen at rt
BF3ÆOEt2 (0.190 mL, 1.50 mmol) dissolved in CH2Cl2
(0.625 mL) was added dropwise during 10 min. After
2 h, the reaction mixture was neutralized with Et3N
and concentrated. The residue was chromatographed
(SiO2, heptane/EtOAc 3:4) to give 1b (128 mg, 64%) as
4.1.5. 1,8-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-
D-xylopyranoside)
(3b). 1,8-Dihydroxynaphthalene
(102 mg, 0.64 mmol), 2,3,4-tri-O-acetyl-a-D-xylopyrano-
syl bromide (471 mg, 1.39 mmol), and tetrabutylammo-
nium bromide (244 mg, 0.76 mmol) were dissolved in
CH2Cl2 (14 mL) and NaOH (aq, 1 M, 14 mL) was add-
ed. The mixture was stirred vigorously for 2 h and then
diluted with EtOAc (70 mL) and the phases were sepa-
rated. The organic phase was washed with NaOH
(1 M) three times, H2O three times, brine once, dried
(Na2SO4), and concentrated. The residue was chromato-
an amorphous brown solid. An analytical sample was
23
D
recrystallized from ethanol. Mp 113.1–114.4 ꢁC. ½aꢂ
1
ꢀ97 (c 0.35, CHCl3). H NMR (CDCl3): d 8.09 (dd,
2H, J = 6.5, 3.3 Hz, H-50, H-80), 7.51 (dd, 2H, J = 6.5,
3.3 Hz, H-60, H-70), 6.99 (s, 2H, H-20, H-30), 5.36 (dd,
2H, J = 8.0, 6.1 Hz, H-2), 5.26–5.31 (m, 4H, H-1, H-
3), 5.03–5.08 (m, 2H, H-4), 4.26 (dd, 2H, J = 12.1,
4.7 Hz, H-5), 3.55 (dd, 2H, J = 12.1, 7.8 Hz, H-5), 2.13
(s, 6H, OAc), 2.09 (s, 12H, OAc). 13C NMR (CDCl3):
d 170.1, 170.0, 148.3, 126.8, 126.7, 121.8, 109.1, 99.4,
70.8, 70.3, 68.6, 62.1, 20.9. HRMS calcd for
C32H37O16 (M+H): 677.2082; found: 677.2081.
graphed (SiO2, heptane/EtOAc 4:3) to give 3b (51 mg,
23
13%) as an amorphous white solid. ½aꢂ ꢀ133 (c 0.50,
D
CHCl3). 1H NMR (CDCl3): d 7.49 (dd, 2H, J = 8.3,
0.8 Hz, H-40, H-50), 7.35 (dd, 2H, J = 8.1, 7.8 Hz, H-
30, H-60), 7.09 (dd, 2H, J = 7.7, 0.8 Hz, H-20, H-70),
5.62 (d, 1H, J = 7.6 Hz, H-2), 5.59 (d, 1H, J = 7.6 Hz,
H-2), 5.22–5.27 (m, 4H, H-3, H-4), 5.07 (d, 2H,
J = 7.6 Hz, H-1), 4.32 (dd, 2H, J = 12.0, 4.8 Hz, H-5),
3.52 (dd, 2H, J = 11.9, 8.3 Hz, H-5), 2.09, 2.07, 1.93 (s,
6H each, OAc). 13C NMR (CDCl3): d 170.7, 170.1,
169.7, 155.1, 137.0, 126.6, 123.0, 110.9, 102.1, 72.7,
70.7, 69.9, 21.0, 20.95, 20.90. HRMS calcd for
C32H37O16 (M+H): 677.2082; found: 677.2074.
4.1.2. 1,4-Dihydroxynaphthalene bis(b-D-xylopyranoside)
(1c). Compound 1b (20 mg, 0.031 mmol) was stirred in
methanolic NaOMe (0.05 M, 2 mL) for 2 h and then
acidified with HOAc and concentrated. The residue
was purified using HPLC and then lyophilized to give
22
1c (7 mg, 53%). ½aꢂ ꢀ66 (c 0.28, DMSO-d6). 1H
NMR (DMSO-d6): dD8.28 (dd, 2H, J = 6.5, 3.3 Hz, H-
50, H-80), 7.52 (dd, 2H, J = 6.5, 3.3 Hz, H-60, H-70),
7.05 (s, 2H, H-20, H-30), 5.50 (d, 2H, J = 5.5 Hz, OH),
5.14 (d, 2H, J = 4.8 Hz, OH), 5.07 (d, 2H, J = 5.0 Hz,
OH), 4.87 (d, 2H, J = 7.6 Hz, H-1), 3.77 (dd, 2H,
J = 11.2, 5.2 Hz, H-5), 3.37–3.45 (m, 4H, H-2, H-4),
3.22–3.33 (m, 4H, H-3, H-5). 13C NMR (DMSO-d6): d
147.8, 126.3, 125.8, 122.0, 110.0, 102.5, 76.3, 73.3,
69.4, 65.7. HRMS calcd for C20H24O10Na (M+Na):
447.1267; found: 447.1261.
4.1.6. 1,8-Dihydroxynaphthalene bis(b-D-xylopyranoside)
(3c). Same method as for 1c. Chromatographed (SiO2,
CH2Cl2/MeOH 3:1) to give 3c (23 mg, 82%) as an amor-
phous white solid. The product was additionally purified
using HPLC and then lyophilized before biological tests.
22
D
1
½aꢂ ꢀ58 (c 0.19, DMSO-d6). H NMR (DMSO-d6): d
7.55 (d, 2H, J = 7.5 Hz, H-40, H-50), 7.41 (t, 2H,
J = 7.9 Hz, H-30, H-60), 7.16 (d, 2H, J = 7.2 Hz, H-20,
H-70), 5.01, (d, 2H, J = 7.6 Hz, H-1), 4.25–5.25 (br s,
6H, OH), 3.74 (dd, 2H, J = 11.2, 5.2 Hz, H-5), 3.24–
3.47 (m, 8H, H-2, H-3, H-4, H-5). 13C NMR (DMSO-
d6): d 154.1, 126.5, 122.3, 111.5, 103.2, 75.9, 73.4, 69.2,
65.9. HRMS calcd for C20H25O10 (M+H): 425.1448;
found: 425.1446.
4.1.3. 1,5-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-
D-xylopyranoside) (2b). Same method as for 1b. Chro-
matographed (SiO2, heptane/EtOAc 3:4 ! 1:3) to give
23
2b (112 mg, 55%) as an amorphous white solid. ½aꢂ
D
ꢀ90 (c 0.50, CHCl3). 1H NMR (CDCl3): d 7.84 (d,
2H, J = 8.6 Hz, H-40, H-80), 7.38 (t, 2H, J = 8.0 Hz, H-
30, H-70), 7.10 (d, 2H, J = 7.5 Hz, H-20, H-60), 5.27–
5.38 (m, 6H, H-1, H-2, H-3), 5.03–5.08 (m, 2H, H-4),
4.27 (dd, 2H, J = 12.2, 4.6 Hz, H-5), 3.59 (dd, 2H,
J = 12.1, 7.5 Hz, H-5), 2.14, 2.10, 2.07 (s, 6H each,
4.1.7. 2,3-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-
D-xylopyranoside)
(4b). 2,3-Dihydroxynaphthalene
(31 mg, 0.19 mmol), 2,3,4-tri-O-acetyl-a/b-D-xylopyr-
anosyl trichloroacetimidate (262 mg, 0.62 mmol), and

2874
R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
MS 300AW (0.3 g) were dissolved in CH2Cl2 (3 mL) and
stirred at r.t. under nitrogen for 40 min. TMSOTf
(0.05 M in CH2Cl2, 0.400 mL) was added dropwise to
the mixture which was then stirred for 1 h. Et3N was
added (pH 8) and the mixture was concentrated. The
residue was chromatographed (SiO2, heptane/EtOAc
3:4) to give 4b (72 mg, 58%) as an amorphous white sol-
69.4, 65.8. HRMS calcd for C20H24O10Na (M+Na):
447.1267; found: 447.1267.
4.1.11. 2,7-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-
b-D-xylopyranoside) (6b). Same method as for 1b. Chro-
matographed (SiO2, heptane/EtOAc 3:4) to give 6b
23
(158 mg, 81%) as an amorphous white solid. ½aꢂ ꢀ27
D
id. An analytical sample was recrystallized from hep-
(c 0.53, CHCl3).1H NMR (CDCl3): d 7.71 (d, 2H,
J = 8.9 Hz, H-40, H-50), 7.27 (s, 2H, H-10, H-80), 7.08
(dd, 2H, J = 8.9, 2.4 Hz, H-30, H-60), 5.31 (d, 2H,
J = 5.6 Hz, H-1), 5.21–5.28 (m, 4H, H-2, H-3), 5.01–
5.06 (m, 2H, H-4), 4.27 (dd, 2H, J = 12.1, 4.6 Hz, H-5),
3.60 (dd, 2H, J = 12.1, 7.5 Hz, H-5), 2.11 (s, 6H, OAc),
2.10 (s, 12H, OAc). 13C NMR (CDCl3): d 170.1, 170.0,
169.6, 155.2, 135.3, 129.6, 126.7, 117.6, 110.9, 98.6,
70.7, 70.2, 68.6, 62.0, 21.0, 20.92, 20.89. HRMS calcd
for C32H36O16Na (M+Na): 699.1901; found: 699.1931.
23
tane/toluene. Mp 99.3–101.1 ꢁC. ½aꢂ ꢀ47 (c 0.49,
D
CHCl3). 1H NMR (CDCl3): d 7.69 (dd, 2H, J = 6.1,
3.3 Hz, H-50, H-80), 7.44 (s, 2H, H-10, H-40), 7.39 (dd,
2H, J = 6.2, 3.3 Hz, H-60, H-70), 5.35 (d, 2H,
J = 5.6 Hz, H-1), 5.23–5.26 (m, 4H, H-2, H-3), 5.02–
5.06 (m, 2H, H-4), 4.30 (dd, 2H, J = 12.2, 4.6 Hz, H-
5), 3.59 (dd, 2H, J = 12.2, 7.2 Hz, H-5), 2.10 (s, 18H,
OAc). 13C NMR (CDCl3): d 170.2, 170.0, 169.4, 146.9,
130.4, 127.0, 125.5, 114.5, 99.6, 70.6, 70.3, 68.6, 62.1,
21.0, 20.94, 20.88. HRMS calcd for C32H36O16Na
(M+Na): 699.1901; found: 699.1934.
4.1.12. 2,7-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (6c). Same method as for 1c. Chromatographed
(SiO2, 60:35:5 CH2Cl2/MeOH/H2O) to give 6c(13 mg,
99%) as an amorphous white solid. The product was
4.1.8. 2,3-Dihydroxynaphthalene bis(b-D-xylopyranoside)
(4c). Same method as for 1c. Chromatographed (SiO2,
60:35:5 CH2Cl2/MeOH/H2O) to give 4c (11 mg, 86%)
as an amorphous white solid. The product was addi-
tionally purified using HPLC and then lyophilized be-
additionally purified using HPLC and then lyophilized
22
before biological tests. ½aꢂ +12 (c 0.38, DMSO-d6).
D
1H NMR (DMSO-d6): d 7.77 (d, 2H, J = 8.9 Hz, H-40,
H-50), 7.35 (d, 2H, J = 2.4 Hz, H-10, H.80), 7.08 (dd,
2H, J = 8.9, 2.4 Hz, H-30, H-60), 5.01 (d, 2H,
J = 7.3 Hz, H-1), 5.00–5.50 (br s, 6H, OH), 3.79 (dd,
2H, J = 9.5, 3.5 Hz, H-5), 3.23–3.43 (m, 8H, H-2, H-3,
H-4, H-5). 13C NMR (DMSO-d6): d 155.5, 129.0,
116.9, 109.6, 100.9, 76.5, 73.1, 69.4, 65.8. HRMS calcd
for C20H24O10Na (M+Na): 447.1267; found: 447.1278.
22
D
1
fore biological tests. ½aꢂ ꢀ16 (c 0.18, DMSO-d6). H
NMR (DMSO-d6): d 7.77 (dd, 2H, J = 6.1, 3.3 Hz, H-
50, H-80), 7.52 (s, 2H, H-10, H-40), 7.36 (dd, 2H,
J = 6.2, 3.3 Hz, H-60, H-70), 5.22 (d, 2H, J = 4.9 Hz,
OH), 5.14 (d, 2H, J = 4.8 Hz, OH), 5.08 (d, 2H,
J = 5.0 Hz, OH), 4.99 (d, 2H, J = 7.3 Hz, H-1), 3.79
(dd, 2H, J = 11.2, 5.0 Hz, H-5), 3.23–3.45 (m, 8H,
H-2, H-3, H-4, H-5). 13C NMR (DMSO-d6): d 147.1,
129.5, 126.6, 124.7, 113.4, 101.9, 76.1, 73.1, 69.3, 65.7.
HRMS calcd for C20H24O10Na (M+Na): 447.1267;
found: 447.1257.
4.1.13. 1,2-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-
b-D-xylopyranoside) (7b). 1,2-Naphthoquinone (28 mg,
0.18 mmol) was dissolved in ethanol (3 mL). NaBH4
(73 mg, 1.93 mmol) suspended in ethanol (3 mL) was
added dropwise. The mixture was stirred at r.t. under
nitrogen. After 1.5 h, the mixture was poured into ice-
water (50 mL), acidified with HCl (0.5 M), extracted with
CH2Cl2, dried (MgSO4), and concentrated to give crude
1,2-dihydroxynaphthalene. The 1,2-dihydroxynaphtha-
lene, 2,3,4-tri-O-acetyl-a/b-D-xylopyranosyl trichloro-
acetimidate (268 mg, 0.64 mmol), and MS 300AW
(0.3 g) were dissolved in CH2Cl2 (3 mL) and stirred at rt
under nitrogen for 80 min. TMSOTf (0.05 M in CH2Cl2,
0.400 mL) was added dropwise to the mixture. After 1 h,
TMSOTf (0.05 M in CH2Cl2, 0.080 mL) was added and
the mixture was stirred for another 80 min. Et3N was add-
ed (pH 8) and the mixture was concentrated. The residue
was chromatographed (SiO2, heptane/EtOAc 3:4) to give
7b (38 mg, 33%) as an amorphous white solid. An analyt-
4.1.9. 2,6-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-b-
D-xylopyranoside) (5b). Same method as for 1b. Chro-
matographed (SiO2, heptane/EtOAc 3:4) to give 5b
23
(164 mg, 82%) as an amorphous white solid. ½aꢂ ꢀ35
D
(c 0.50, CHCl3). 1H NMR (CDCl3): d 7.67 (d, 2H,
J = 8.9 Hz, H-40, H-80), 7.33 (d, 2H, J = 2.5 Hz, H-10,
H-50), 7.17 (dd, 2H, J = 8.8, 2.4 Hz, H-30, H-70), 5.20–
5.30 (m, 6H, H-1, H-2, H-3), 5.01–5.06 (m, 2H, H-4),
4.27 (dd, 2H, J = 12.1, 4.6 Hz, H-5), 3.58 (dd, 2H,
J = 12.1, 7.5 Hz, H-5), 2.10 (s, 12H, OAc), 2.09 (s, 6H,
OAc). 13C NMR (CDCl3): d 170.1, 170.0, 169.6, 153.7,
130.8, 129.0, 119.7, 111.7, 98.8, 70.7, 70.3, 68.6, 62.0,
20.94, 20.92, 20.89. HRMS calcd for C32H37O16
(M+H): 677.2082; found: 677.2068.
4.1.10. 2,6-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (5c). Same method as for 1c. Purified using HPLC
ical sample was purified on Sephadex LH-20 (1:1 CH2Cl2/
23
MeOH). ½aꢂ ꢀ25 (c 0.61, CHCl3). 1H NMR (CDCl3): d
D
22
D
and then lyophilized to give 5c (11 mg, 74%). ½aꢂ +2 (c
8.18 (d, 1H, J = 8.5 Hz, H-80), 7.77 (d, 1H, J = 8.1 Hz,
H-50), 7.61 (d, 1H, J = 9.0 Hz, H-40), 7.49 (ddd, 1H,
J = 8.4, 6.9, 1.2 Hz, H-70), 7.41 (ddd, 1H, J = 8.1, 6.9,
1.3 Hz, H-60), 7.33 (d, 1H, J = 9.0 Hz, H-30), 5.51 (d,
1H, J = 6.2 Hz, H-1), 5.36 (dd, 1H, J = 8.1, 6.2 Hz, H-
2), 5.22–5.31 (m, 4H, H-1, H-2, H-3), 5.03–5.09 (m, 2H,
H-4), 4.30 (dd, 1H, J = 12.1, 4.7 Hz, H-5), 4.16 (dd, 1H,
J = 12.1, 4.8 Hz, H-5), 3.53 (dd, 1H, J = 12.1, 7.6 Hz,
H-5), 3.36 (dd, 1H, J = 12.1, 8.0 Hz, H-5), 2.13, 2.11,
1
0.43, DMSO-d6). H NMR (DMSO-d6): d 7.76 (d, 2H,
J = 8.9 Hz, H-40, H-80), 7.39 (d, 2H, J = 2.5 Hz, H-10,
H-50), 7.20 (dd, 2H, J = 8.9, 2.4 Hz, H-30, H-70), 5.36
(d, 2H, J = 4.8 Hz, OH), 5.12 (d, 2H, J = 4.4 Hz, OH),
5.08 (d, 2H, J = 4.7 Hz, OH), 4.97 (d, 2H, J = 7.2 Hz,
H-1), 3.78 (dd, 2H, J = 10.5, 4.5 Hz, H-5), 3.22–3.41
(m, 8H, H-2, H-3, H-4, H-5). 13C NMR (DMSO-d6): d
153.7, 129.8, 128.3, 119.3, 110.6, 101.2, 76.5, 73.1,

R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
2875
23
D
2.10, 2.09, 2.08, 2.05 (s, 3H each, OAc). 13C NMR
(CDCl3): d 170.1, 169.9, 144.8, 127.8, 126.8, 125.6,
125.5, 122.3, 118.5, 100.7, 100.1, 71.4, 71.3, 71.0, 70.5,
68.8, 62.5, 62.4, 31.2, 21.0. HRMS calcd for
C32H36O16Na (M+Na): 699.1901; found: 699.1870.
(187 mg, 95%) as an amorphous white solid. ½aꢂ ꢀ73
(c 0.50, CHCl3). 1H NMR (CDCl3): d 8.07 (d, 1H,
J = 9.2 Hz, H-80), 7.43 (d, 1H, J = 8.3 Hz, H-40), 7.36
(m, 1H, H-30), 7.33 (d, 1H, J = 2.4 Hz, H-50), 7.17 (dd,
1H, J = 9.2, 2.5 Hz, H-70), 6.97 (dd, 1H, J = 7.6,
0.7 Hz, H-20), 5.21–5.39 (m, 6H, H-1, H-2, H-3), 5.01–
5.08 (m, 2H, H-4), 4.26 (ddd, 2H, J = 12.1, 4.6,
1.7 Hz, H-5), 3.59 (ddd, 2H, J = 12.1, 7.4, 4.4 Hz, H-
5), 2.13, 2.11 (s, 3H each, OAc), 2.10 (s, 9H, OAc),
2.07 (s, 3H, OAc). 13C NMR (CDCl3): d 170.1, 170.01,
169.98, 169.97, 169.7, 169.5, 155.1, 152.6, 135.7, 126.8,
123.9, 122.3, 122.1, 118.6, 111.2, 107.7, 98.8, 98.5,
70.7, 70.6, 70.20, 70.16, 68.6, 62.1, 62.0, 20.93, 20.91,
20.87. HRMS calcd for C32H37O16 (M+H): 677.2082;
found: 677.2086.
4.1.14. 1,2-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (7c). Same method as for 1c. Chromatographed
(SiO2, 5:1 CH2Cl2/MeOH) to give 7c (8 mg, 70%) as
an amorphous white solid. The product was additionally
purified using HPLC and then lyophilized before biolog-
22
ical tests. ½aꢂ ꢀ47 (c 0.23, DMSO-d6). 1H NMR
D
(DMSO-d6): d 8.19 (d, 1H, J = 8.5 Hz, H-80), 7.84 (d,
1H, J = 8.1 Hz, H-50), 7.70 (d, 1H, J = 8.9 Hz, H-40),
7.39–7.51 (m, 3H, H-30, H-60, H-70), 5.07 (d, 1H,
J = 7.6 Hz, H-1), 4.92 (d, 1H, J = 7.5 Hz, H-1), 4.90–
5.90 (br s, 6H, OH), 3.80 (dd, 1H, J = 11.2, 5.2 Hz, H-
5), 3.62 (dd, 1H, J = 11.3, 5.3 Hz, H-5), 3.21–3.46 (m,
7H, H-2, H-3, H-4, H-5), 2.95 (t, 1H, J = 11.3 Hz, H-
3). 13C NMR (CDCl3): d 145.5, 139.9, 130.0, 129.4,
127.3, 126.1, 124.8, 122.5, 118.9, 104.4, 102.9, 76.3,
75.7, 74.0, 73.5, 69.5, 69.4, 65.9. HRMS calcd for
C20H25O10 (M+H): 425.1448; found: 425.1442.
4.1.18. 1,6-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (9c). Same method as for 1c. Chromatographed
(SiO2, 60:35:5 CH2Cl2/MeOH/H2O) to give 9c (9 mg,
85%) as an amorphous white solid. The product was
additionally purified using HPLC and then lyophilized
22
D
before biological tests. ½aꢂ ꢀ45 (c 0.31, DMSO-d6).
1H NMR (DMSO-d6): d 8.22 (d, 1H, J = 9.2 Hz, H-
80), 7.45 (d, 1H, J = 8.3 Hz, H-40), 7.39 (d, 1H,
J = 2.5 Hz, H-50), 7.37 (t, 1H, J = 8.1 Hz, H-30), 7.19
(dd, 1H, J = 9.2, 2.4 Hz, H-70), 6.98 (d, 1H,
J = 7.3 Hz, H-20), 5.03 (d, 1H, J = 7.2 Hz, H-1), 4.99
(d, 1H, J = 7.6 Hz, H-1), 4.80–5.50 (br s, 6H, OH),
3.76–3.80 (m, 2H, H-5), 3.24–3.45 (m, 8H, H-2, H-3,
H-4, H-5). 13C NMR (DMSO-d6): d 155.4, 152.9,
135.3, 126.7, 123.9, 121.3, 120.7, 118.0, 109.9, 107.4,
101.6, 100.8, 76.5, 76.3, 73.2, 73.1, 69.4, 65.8, 65.7.
HRMS calcd for C20H25O10 (M+H): 425.1448; found:
425.1450.
4.1.15. 1,3-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-
b-D-xylopyranoside) (8b). Same method as for 1b. Chro-
matographed (SiO2, heptane/EtOAc 3:4) to give 8b
23
(45 mg, 22%) as an amorphous white solid. ½aꢂ ꢀ45
D
(c 0.50, CHCl3). 1H NMR (CDCl3): d 8.04 (d, 1H,
J = 8.4 Hz, H-80), 7.69 (d, 1H, J = 8.2 Hz, H-50), 7.47
(ddd, 1H, J = 8.1, 6.9, 1.2 Hz, H-60), 7.37 (ddd, 1H,
J = 8.2, 6.9, 1.1 Hz, H-70), 7.06 (d, 1H, J = 1.9 Hz, H-
40), 6.80 (d, 1H, J = 2.2 Hz, H-20), 5.20–5.36 (m, 6H,
H-1, H-2, H-3), 5.02–5.05 (m, 2H, H-4), 4.27 (ddd,
2H, J = 12.1, 4.6, 1.4 Hz, H-5), 3.58–3.65 (m, 2H, H-
5), 2.15 (s, 3H, OAc), 2.11 (s, 9H, OAc), 2.10, 2.08 (s,
3H each, OAc). 13C NMR (CDCl3): d 170.1, 170.0,
169.9, 169.63, 169.56, 154.4, 153.5, 134.8, 127.6, 127.1,
124.6, 122.7, 121.8, 105.7, 102.6, 98.5, 98.4, 70.7, 70.2,
69.8, 68.6, 68.4, 62.1, 61.8, 20.93, 20.90. HRMS calcd
for C32H36O16Na (M+Na): 699.1901; found: 699.1874.
4.1.19. 1,7-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-
b-D-xylopyranoside) (10b). Same method as for 1b.
Chromatographed (SiO2, heptane/EtOAc 3:4) to give
23
D
10b (194 mg, 96%) as an amorphous white solid. ½aꢂ
ꢀ71 (c 0.50, CHCl3). 1H NMR (CDCl3): d 7.74 (d,
1H, J = 9.0 Hz, H-50), 7.61 (d, 1H, J = 2.5 Hz, H-80),
7.49 (d, 1H, J = 8.3 Hz, H-20), 7.26–7.30 (m, 1H, H-
30), 7.18 (dd, 1H, J = 8.9, 2.5 Hz, H-60), 7.03 (d, 1H,
J = 7.4 Hz, H-40), 5.42–5.46 (m, 2H, JH-1 = 6.7 Hz, H-
1, H-2), 5.31–5.37 (m, 2H, H-3), 5.22–5.26 (m, 2H,
4.1.16. 1,3-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (8c). Same method as for 1c. Chromatographed
(SiO2, 60:35:5 CH2Cl2/MeOH/H2O) to give 8c (24 mg,
88%) as an amorphous white solid. The product was
J
H-1 = 6.7 Hz, H-1, H-2), 5.05–5.14 (m, 2H, H-4),
additionally purified using HPLC and then lyophilized
4.22–4.31 (m, 2H, H-5), 3.81 (dd, 1H, J = 11.9,
9.0 Hz, H-5), 3.61 (dd, 1H, J = 11.9, 8.6 Hz, H-5),
2.13 (s, 3H, OAc), 2.09 (s, 9H, OAc), 2.07, 2.06 (s, 3H
each, OAc). 13C NMR (CDCl3): d 170.3, 170.2,
170.03, 170.00, 169.8, 169.6, 154.6, 152.2, 131.0, 129.6,
126.5, 124.3, 122.5, 119.7, 109.1, 104.4, 99.8, 97.8,
71.7, 71.2, 70.9, 70.7, 69.0, 68.8, 62.6, 62.3, 21.1, 21.0,
20.9, 20.8. HRMS calcd for C32H37O16 (M+H):
677.2082; found: 677.2049.
22
before biological tests. ½aꢂ ꢀ35 (c 0.28, DMSO-d6).
D
1H NMR (DMSO-d6): d 8.18 (d, 1H, J = 8.4 Hz, H-
80), 7.77 (d, 1H, J = 8.2 Hz, H-50), 7.47 (ddd, 1H,
J = 8.1, 6.8, 1.1 Hz, H-60), 7.35 (ddd, 1H, J = 8.2, 6.9,
1.1 Hz, H-70), 7.11 (d, 1H, J = 1.9 Hz, H-40), 6.83 (d,
1H, J = 2.1 Hz, H-20), 5.01 (d, 1H, J = 7.5 Hz, H-1),
4.99 (d, 1H, J = 7.6 Hz, H-1), 4.50–5.50 (br s, 6H,
OH), 3.77–3.84 (m, 2H, H-5), 3.24–3.46 (m, 8H, H-2,
H-3, H-4, H-5). 13C NMR (DMSO-d6): d 155.1, 153.7,
134.4, 127.0, 126.7, 121.7, 104.0, 102.8, 101.7, 101.0,
76.5, 76.2, 73.14, 73.08, 69.4, 69.3, 65.8. HRMS calcd
for C20H24O10Na (M+Na): 447.1267; found: 447.1269.
4.1.20. 1,7-Dihydroxynaphthalene bis(b-D-xylopyrano-
side) (10c). Same method as for 1c. Chromatographed
(SiO2, 60:35:5 CH2Cl2/MeOH/H2O) to give 10c
(15 mg, 98%) as an amorphous white solid. The product
4.1.17. 1,6-Dihydroxynaphthalene bis(2,3,4-tri-O-acetyl-
b-D-xylopyranoside) (9b). Same method as for 1b. Chro-
matographed (SiO2, heptane/EtOAc 3:4) to give 9b
was additionally purified using HPLC and then lyophi-
22
D
lized before biological tests. ½aꢂ ꢀ62 (c 0.28, DMSO-
1
d6). H NMR (DMSO-d6): d 7.82 (d, 1H, J = 9.0 Hz,

2876
R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
H-50), 7.75 (d, 1H, J = 2.5 Hz, H-70), 7.50 (d, 1H,
J = 8.3 Hz, H-40), 7.28 (t, 1H, J = 7.9 Hz, H-30), 7.24
(dd, 1H, J = 8.9, 2.6 Hz, H-60), 7.08 (d, 1H,
J = 7.3 Hz, H-20), 5.57 (d, 1H, J = 5.5 Hz, OH), 5.41
(d, 1H, J = 4.9 Hz, OH), 5.17 (d, 1H, J = 4.9 Hz, OH),
5.14 (d, 1H, J = 4.6 Hz, OH), 5.11 (d, 1H, J = 4.4 Hz,
OH), 5.09 (d, 1H, J = 7.1 Hz, H-1), 5.08 (d, 1H,
J = 5.0 Hz, OH), 4.96 (d, 1H, J = 7.5 Hz, H-1), 3.79
(dd, 2H, J = 11.0, 5.1 Hz, H-5), 3.37–3.47 (m, 4H, H-
2, H-4), 3.26–3.35 (m, 4H, H-3, H-5). 13C NMR
(DMSO-d6): d 154.6, 151.8, 130.1, 129.0, 126.3, 123.9,
121.4, 118.9, 110.1, 104.8, 101.6, 100.5, 76.6, 76.3,
73.2, 73.1, 69.4, 65.7, 65.6. HRMS calcd for
C20H24O10Na (M+Na): 447.1267; found: 447.1261.
PBS (0.137 M NaCl, 3 mM KCl, 8 mM Na2HPO4,
2 mM KH2PO4, pH 7.5).
4.2.2. Isolation of xyloside-primed radiolabeled GAG.
The procedures have been described in detail previous-
ly.4 [35S]sulfate labeled polyanionic macromolecules
were isolated from the culture medium by ion ex-
change-chromatography on DEAE-cellulose at 4 ꢁC.
Samples were mixed with 1.3 vol of 7 M urea, 10 mM
Tris, pH 7.5, 0.1% Triton X-100, 10 mM NEM, and
passed over a 1 mL-column of DEAE equilibrated with
6 M urea, 0.5 M NaOAc, pH 5.8, 5 lg/mL ovalbumin,
0.1% Triton X-100. After sample application, the col-
umns were washed successively with 10 mL-portions
of (a) equilibration buffer (see above), (b) 6 M urea,
10 mM Tris, pH 8.0, 5 lg/mL ovalbumin, 0.1% Triton
X-100, and (c) 50 mM Tris, pH 7.5. Bound material
was eluted with 5· 1 mL of 4 M guanidine–HCl,
50 mM NaOAc, pH 5.8, 5 lg/mL ovalbumin, 0.2% Tri-
ton X-100. Radioactive fractions were pooled, precipi-
tated with 5 vol of 95% ethanol overnight at ꢀ20 ꢁC
using 100 lg of dextran as carrier. After centrifugation
in a Beckman JS-7.5 at 4000 rpm and 4 ꢁC for 45 min,
material was dissolved in 4 M guanidine–HCl, 50 mM
NaOAc, pH 5.8 and free xyloside-primed GAG chains
were separated from PG by hydrophobic interaction
chromatography on Octyl-Sepharose followed by gel
permeation FPLC on Superose 6. Radioactivity was
determined in a b-counter.
4.1.21. HPLC of naphthoxylosides. High-performance
liquid chromatography was run on a Hewlett–Packard
Series II 1090 Liquid Chromatograph and a YMC
Hydrosphere C18 column (15 cm · 4.6 mm, 5 lm).
The system was controlled by the Hewlett–Packard
ChemStation for LC software suite. The mobile phase
consisted of H2O + 0.1% trifluoroacetic acid (TFA)
with a gradient of acetonitrile from 1 min increasing
by 1.2% per minute until 30 min. The mean retention
times were calculated from three separate runs per
compound.
4.1.22. Stability of naphthoxylosides. A solution of xylo-
side (20 mM in DMSO/water 1:1) was diluted in Ham’s
F-12 medium to about 0.20 mM. 4,40-Dihydroxybiphenyl
(0.1 mM) was used as an internal standard. The samples
were heated to 37 ꢁC, and analytical samples were taken
at t = 0, 1, 3, 24, 48, 72, and 96 h and analyzed by HPLC.
4.2.3. Degradative method. Xyloside-primed GAG
chains separated by Superose 6 were pooled and
precipitated with 5 vol of 95% ethanol overnight at
ꢀ20 ꢁC using 100 lg of dextran as carrier. After centri-
fugation in a Beckman JS-7.5 at 4000 rpm and 4 ꢁC for
45 min, material was subjected to periodate oxidation
using 0.02 M NaIO4, 0.1 M sodium formate, pH 3.0,
at 4 ꢁC for 24 h.23 The reaction was stopped by addi-
tion of mannitol. The samples were then desalted using
PD10 columns, lyophilized, and subjected to alkali
treatment at pH 12 for 30 min.23
4.2. Biology
The human bladder carcinoma cell-line T24 was ob-
tained from ATCC, Rockville, MD. Regular cell culture
media, L-glutamine, penicillin–streptomycin, trypsin,
and donor calf serum were obtained from Life Technol-
ogies. Dulbecco’s modified Eagle’s medium (DMEM)
and Ham’s F-12 medium were purchased from Sigma.
Na235SO4 (1310 Ci/mmol) was obtained from Amer-
sham International, UK. Epidermal growth factor was
purchased from Genzyme, Cambridge, MA and crystal
violet from Merck, Germany. The prepacked Superose
6 HR 10/30 and Dextran T-500 were from Pharmacia-
LKB, Sweden, and DE-53 DEAE-cellulose was from
Whatman. Water for HPLC-analysis was from a Milli-
pore Milli-Q system.
4.2.4. In vitro growth assay using crystal violet
method. The procedure has been described else-
where.4 Cells were seeded into 96-well microculture
plates at 3000 cells/well in DMEM supplemented
with insulin (10 ng/mL), transferrin (25 ng/mL), and
10% fetal calf serum. After 4 h of plating the cells
were placed in serum-free Ham’s F-12 medium sup-
plemented with insulin (10 ng/mL) and transferrin
(25 ng/mL) for an additional 24 h. Cells were then al-
lowed to proliferate supported by 10 ng/mL of epi-
dermal growth factor in the presence of 0.025, 0.05,
0.1, 0.2 and 0.4 mM of xylosides. Controls without
xylosides were included. The total exposure-time
was 96 h. Cells were then fixed in 1% glutaraldehyde
dissolved in Hanks’ balanced salt solution (NaCl
80 g/L, KCl 4 g/L, glucose 10 g/L, KH2PO4 600 mg/
L, NaHPO4 475 mg/L) for 15 min, then cell nuclei
were stained with 0.1% crystal violet. After washing
and cell lysis for 24 h in Triton X-100, the amount
of bound dye was measured at A600 in a microplate
photometer (Titertek multiscan). The inhibitory
4.2.1. Cell culture and radiolabeling. Cells were cultured
as monolayers in modified Eagle’s medium (MEM)
supplemented with 10% (v/v) fetal bovine serum,
2 mM L-glutamine, penicillin (100 U/mL), and strepto-
mycin (100 lg/mL) in an incubator with humidified
atmosphere and 5% CO2 at 37 ꢁC. Confluent cells were
preincubated for 1 h in low-sulfate, MgCl2-labeling
medium supplemented with 2 mM glutamine. The
pre-incubation medium was replaced by fresh medium
containing 50 mCi/mL of [35S]sulfate and 0.1 mM xylo-
side. After the incubation period, culture medium was
collected and pooled with two washings of ice-cold

R. Johnsson et al. / Bioorg. Med. Chem. 15 (2007) 2868–2877
2877
6. Belting, M.; Borsig, L.; Fuster, M. M.; Brown, J. R.;
˚
Persson, L.; Fransson, L.-A.; Esko, J. D. Proc. Natl. Acad.
effect of the compounds is expressed as the percent-
age of growth in the absence of drugs.
Sci. U.S.A. 2002, 99, 371.
7. Jacobsson, M.; Ellervik, U.; Belting, M.; Mani, K. J. Med.
Chem. 2006, 49, 1932.
8. Jacobsson, M.; Malmberg, J.; Ellervik, U. Carbohydr. Res.
2006, 341, 1266.
Acknowledgments
The Swedish Research Council, Association for Interna-
tional Cancer Research, the Cancer Fund, the Swedish
Cancer Society, the American Cancer Society, the
Crafoord Foundation, the Thelma Zoegas Foundation,
and the Bergvall Foundation supported this work.
9. Lee, Y. S.; Rho, E. S.; Min, Y. K.; Kim, B. T.; Kim, K. H.
J. Carbohydr. Chem. 2001, 20, 503.
10. Spartan ’02 for Macintosh, Wavefunction, Inc. Irvine,
CA.
11. Johnsson, R.; Ellervik, U. Synlett 2005, 2939.
12. Mori, M.; Ito, Y.; Ogawa, T. Carbohydr. Res. 1990, 195,
199.
13. Musso, H.; Matthies, H.-G. Chem. Ber. 1961, 94,
356.
Supplementary data
14. Kartha, K. P. R.; Jennings, H. J. J. Carbohydr. Chem.
1990, 9, 777.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2007.02.020.
´
´
15. Satge, C.; Le Bras, J.; Henin, F.; Muzart, J. Tetrahedron
2005, 61, 8405.
16. Kro¨ger, L.; Thiem, J. J. Carbohydr. Chem. 2003,
22, 9.
´
17. Valko, K.; Plass, M.; Bevan, C.; Reynolds, D.; Abraham,
References and notes
M. H. J. Chromatogr. A 1998, 797, 41.
18. Cartel, N. J.; Post, M. Am. J. Physiol. Lung Cell. Mol.
Physiol. 2005, 288, L285.
19. Foti, M. C.; Johnson, E. R.; Vinqvist, M. R.; Wright, J. S.;
Barclay, L. R. C.; Ingold, K. U. J. Org. Chem. 2002,
67, 5190.
20. O’Brien, P. J. Chem. Biol. Interact. 1991, 80, 1.
21. Scott, L. T.; Rozeboom, M. D.; Houk, K. N.; Fukunaga,
T.; Lindner, H. J.; Hafner, K. J. Am. Chem. Soc. 1980,
102, 5169.
1. Vuong, T. T.; Prydz, K.; Tveit, H. Glycobiology 2006, 16, 326.
2. Miao, H.-Q.; Fritz, T. A.; Esko, J. D.; Zimmermann, J.;
Yayon, A.; Vlodavsky, I. J. Cell. Biochem. 1995, 57, 173.
3. Martin, N. B.; Masson, P.; Sepulchre, C.; Theveniaux, J.;
Millet, J.; Bellamy, F. Semin. Thromb. Hemostat. 1996, 22, 247.
4. Mani, K.; Havsmark, B.; Persson, S.; Kaneda, Y.;
Yamamoto, H.; Sakurai, K.; Ashikari, S.; Habuchi, H.;
Suzuki, S.; Kimata, K.; Malmstro¨m, A.; Westergren-
˚
Thorsson, G.; Fransson, L.-A. Cancer Res. 1998, 58,
1099.
5. Mani, K.; Belting, M.; Ellervik, U.; Falk, N.; Svensson,
22. Jacobsson, M.; Ellervik, U. Tetrahedron Lett. 2002, 43,
6549.
˚
23. Fransson, L.-A.; Silverberg, I.; Carlstedt, I. J. Biol. Chem.
1985, 260, 14722.
˚
G.; Sandgren, S.; Cheng, F.; Fransson, L.-A. Glycobiology
2004, 14, 387.