Xylosylated naphthoic acid-amino acid conjugates for investigation of glycosaminoglycan priming

Article abstract of DOI:10.1016/j.carres.2008.04.006

Three different series of xylosylated naphthoic acid-amino acid conjugates containing one or two amino acid residues were synthesized for the investigation of glycosaminoglycan priming and potential use as anti-tumor drugs. All xylosylated naphthoic acid-conjugates inhibited the growth of normal lung fibroblasts to some extent, whereas the growth of tumor derived T24 carcinoma cells was not affected. There was no correlation between amino acid conjugation, retention time and the antiproliferative activity. Only one compound initiated the priming of glycosaminoglycans. Modification of the naphthalene ring with one or two amino acid residues did not have any effect on proteoglycan biosynthesis or glycosaminoglycan priming in T24 carcinoma cells.

Full text of DOI:10.1016/j.carres.2008.04.006

Available online at www.sciencedirect.com
Carbohydrate Research 343 (2008) 1473–1477
Note
Xylosylated naphthoic acid–amino acid conjugates
for investigation of glycosaminoglycan priming
Carl-Olof Abrahamsson,a Ulf Ellervik,a,* Johan Eriksson-Bajtner,b
a
c,
*
˚
Marten Jacobsson and Katrin Mani
aOrganic Chemistry, Lund University, PO Box 124, SE-221 00 Lund, Sweden
bDuPont Chemoswed, PO Box 139, SE-201 80 Malmo¨, Sweden
cDepartment of Experimental Medical Science, Division of Neuroscience, Glycobiology Group, Lund University,
Biomedical Center C13, SE-221 84 Lund, Sweden
Received 29 February 2008; received in revised form 2 April 2008; accepted 4 April 2008
Available online 10 April 2008
Abstract—Three different series of xylosylated naphthoic acid–amino acid conjugates containing one or two amino acid residues
were synthesized for the investigation of glycosaminoglycan priming and potential use as anti-tumor drugs. All xylosylated naph-
thoic acid-conjugates inhibited the growth of normal lung fibroblasts to some extent, whereas the growth of tumor derived T24 car-
cinoma cells was not affected. There was no correlation between amino acid conjugation, retention time and the antiproliferative
activity. Only one compound initiated the priming of glycosaminoglycans. Modification of the naphthalene ring with one or two
amino acid residues did not have any effect on proteoglycan biosynthesis or glycosaminoglycan priming in T24 carcinoma cells.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Glycosaminoglycan; Priming; Xylopyranoside; Amino acid; Naphthoic acid
Glycosaminoglycan (GAG) chains anchored to core
proteins form proteoglycans (PG), a class of extracellu-
lar macromolecules having functions ranging from bulk
construction material to involvements in cell–cell inter-
actions. The biosynthesis of GAG chains start with the
formation of a glycosidic bond between serine residues
of the core protein and the unique xylose residue of
the resulting GAG chain.1 A specific linker tetrasaccha-
ride, GlcA(b1-3)Gal(b1-3)Gal(b1-4)Xylb is then assem-
bled on which GAG chains are elongated and further
modified through N-deacetylation/N-sulfation, O-sulfa-
tion and epimerization (Chart 1). Addition of GlcNAc
or GalNAc to the linker defines whether heparan sulfate
(HS) or chondroitin sulfate/dermatan sulfate (CS/DS) is
formed.
ing on the structure of the aglycon, different GAG
chains are formed. We have previously shown that
the GAG-priming 2-(6-hydroxynaphthyl) b-D-xylo-
pyranoside (1, Chart 1) selectively inhibits the prolifer-
ation of transformed or tumor-derived cells in vitro as
well as in vivo.3 Also, treatment with this xyloside re-
duced the average tumor load by 70–97% in a SCID
mice model.4 Toxicity studies of the 14 isomeric xylosy-
lated dihydroxynaphthalenes revealed differences in
anti-proliferative activity depending on the aglycon
structure.5,6
Our results suggest that the priming of HS chains is
required for selective growth inhibition,3,4 and new xylo-
side analogues, which to a great extent prime HS as
opposed to CD/DS, would thus be of great interest. It
is still unclear what determines whether HS or CS/DS
chains are attached to the core protein, but repetitive
Ser-Gly sequences and a high proportion of Phe, Tyr
or Trp promote the formation of HS.7 Coupling of a
Ser residue directly to the xylose has been investigated,8
and whilst such xylosides prime GAG synthesis they
Xylosides carrying hydrophobic aglycon can enter
cells and serve as primers of GAG synthesis.2 Depend-
*
Corresponding authors. Tel.: +46 46 222 82 20; fax: +46 46 222 82 09
(U.E.); e-mail: ulf.ellervik@organic.lu.se
0008-6215/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2008.04.006

1474
C.-O. Abrahamsson et al. / Carbohydrate Research 343 (2008) 1473–1477
a)
Core Protein
HN
COOH
OH
O
O
O
HO
O
HO
OH
OH
OH
OH
COOH
OH
O
AcHN
O
O
O
O
O
HO
O
HO
O
NH
O
O
OH
OH
OH
OH
O
b)
HN
O
HO
HO
O
OH
1
OH
Chart 1. (a) Biosynthesis of glycosaminoglycan (GAG) chains start with the xylosylation of a serine residue after which a linker tetrasaccharide
(GlcA(b1-3)Gal(b1-3)Gal(b1-4)Xylb) is assembled. (b) Xylosides carrying hydrophobic aglycon, such as compound 1, can enter cells and serve as
primers of GAG chains.
lack the naphthalenic moiety needed for increased HS
priming.
sugar moiety and the acid according to the two-step pro-
cedure of Grynkiewicz (Scheme 1).12
We therefore looked into the synthesis of naphthoxy-
loside conjugates carrying different amino acid residues.
In a first stage, we envisaged coupling amino acids with
a carboxylic acid group on the naphthalene moiety of a
naphthoxyloside. Since the naphthalene moiety would
not carry two oxygen atoms, its inherent toxicity would
most likely be low. This would enable us to study the
effect of the amino acid on priming with minimal effect
on cell proliferation.
Our first attempts to xylosylate the free hydroxynaph-
thoic acids failed due to low solubility in CH2Cl2. In-
stead, the synthesis started from the known methyl
6-hydroxy-1-naphthoate (2)9 and methyl 6-hydroxy-2-
naphthoate (7).10 To our surprise, xylosylation using tri-
chloroacetimidate donor gave low yields (below 50%),
and we therefore turned to the use of peracetylated
xylose with the addition of triethylamine to suppress
anomerization, 11 which gave excellent conversions.
Selective deprotection of the carboxylic acid turned
out to be difficult and instead we deprotected both the
Selective protection of the sugar was then performed
using Ac2O in pyridine at room temperature, as heating
in acetic anhydride gave substantial anhydride formation.
A selection of ester-protected amino acids (glycine t-butyl
ester hydrochloride, L-alanine methyl ester hydrochlo-
ride, L-phenylalanine methyl ester hydrochloride) were
chosen for solution phase synthesis and couplings of the
xylosylated
hydroxynaphthoic
acids
proceeded
smoothly. NaOMe–MeOH–CH2Cl2 was used to depro-
tect the sugar moiety, and full deprotection was then
performed using NaOH. The conjugates were further
purified using semi-preparative HPLC and lyophilized.
To synthesize a series of xylosylated naphthoic acid-
dipeptide conjugates, we turned to solid-phase method-
ology, where preloaded resins allow for fast synthesis.
The synthesis started from the commercially available
Merrifield resins with Boc-protected glycine (0.5 mmol/
g) and Boc-protected L-alanine (0.7 mmol/g). After
deprotection of the amino group using TFA in CH2Cl2,
a dipeptide was formed by peptide coupling with Boc-
HO
O
O
HO
HO
HO
O
O
a-c
d-h
HO
OH
OH
R
3, 76%
4: R = H, 82%
5: R = Me, 75%
6: R = Bn, 61%
2
O
OMe
O
OH
O
N
COOH
H
O
O
O
HO
HO
HO
HO
HO
O
a-c
OMe
d-h
OH
OH
H
N
R
OH
O
O
O
COOH
7
8, 85%
9: R = H, 71%
10: R = Me, 65%
11: R = Bn, 54%
Scheme 1. Synthesis of xylosylated naphthoic acid–amino acid conjugates. Reagents and conditions: (a) 1,2,3,4-tetra-O-acetyl-b-D-xylopyranose,
BF3ÁOEt2, CH2Cl2, rt; (b) NaOMe, MeOH, rt; (c) NaOH, MeOH, rt; (d) Ac2O, pyridine, rt; (e) DMAP, protected amino acid, CH2Cl2, rt; (f) DIC in
CH2Cl2, rt; (g) NaOMe, MeOH, CH2Cl2, rt; (h) NaOH (1 M, aq), MeOH, rt.

C.-O. Abrahamsson et al. / Carbohydrate Research 343 (2008) 1473–1477
1475
O
14: R1 = H, R2 = CH2CH(CH3)2, 61%
O
a-h
15: R1 = H, R2 = Bn, 66%
BocHN
HO
HO
O
O
16: R1 = H, R2 = CH2CH2SCH3, 78%
17: R1 = Me, R2 = CH2CH(CH3)2, 88%
18: R1 = Me, R2 = Bn, 95%
R1
OH
R2
O
H
12: R1 = H
19: R1 = Me, R2 = CH2CH2SCH3, 81%
N
(S)
13: R1 = Me
O
N
OH
H
R1
O
Scheme 2. Solid-phase synthesis of xylosylated naphthoic acid–amino acid conjugates containing two amino acid residues. Reagents and conditions:
(a) TFA (80% in CH2Cl2 containing 0.4 M thiophenol), rt; (b) DMAP, protected amino acid, CH2Cl2, rt; (c) DIC in CH2Cl2, rt; (d) TFA (80% in
CH2Cl2 containing 0.4 M thiophenol),rt; (e) DMAP, 6-(1-carboxynaphthyl) 2,3,4-tri-O-acetyl-b-D-xylopyranoside, CH2Cl2,rt; (f) DIC in CH2Cl2,rt;
(g) NaOMe, MeOH, CH2Cl2, rt; (h) NaOH (1 M, aq), MeOH, rt.
protected amino acids (N-a-t-Boc-L-leucine, N-a-t-Boc-
L-phenylalanine and N-a-t -Boc-L-methionine) using
DIC/DMAP (Scheme 2). Unbound peptide was re-
moved by washing and the terminal Boc-group was
removed prior to coupling with the xylosylated
hydroxynaphthoic acid. Cleavage from the resin and
deprotection of the conjugates gave 14–19 in good yields
with no noticeable epimerization.
It has been shown that gradient HPLC retention
times, in contrary to isocratic retention times, can be
treated as linear free-energy related parameters, that is,
gradient HPLC retention times can be used to substitute
logP values in biological evaluations.13 The gradient
HPLC retention times for the conjugates were measured
using a C-18 column and a mobile phase of water (0.1%
trifluoroacetic acid) with a gradient of acetonitrile from
1 min increasing by 1.2% per min up to 30 min. The
retention times were measured for two separate runs
per compound, and the calculated mean retention times
are presented in Table 1.
For the determination of the antiproliferative activity,
normal HFL-1 cells (human fetal lung fibroblasts) and
T24 cells (human bladder carcinoma cells) were used.
The xylosides were added to the growth medium at var-
ious concentrations and cell proliferation was recorded
using the crystal violet method.3 The inhibitory effect
of the compounds is expressed as ED50 (lM) scored
after 96 h of exposure relative to untreated cells (Table
1). As expected, the antiproliferative activities of naph-
thoxylosides carrying one or two amino acid residues
were low in both normal lung fibroblasts and tumor
derived T24 carcinoma cells. Interestingly, normal lung
fibroblasts were at least two times more sensitive to
these compounds with an ED 50 value ranging between
200 and 300 lM compared to T24 cells that grew well at
these concentrations.
To test the GAG priming ability of the xylosides,
T24 cells were incubated with 100 lM xyloside and
[35S]sulfate. GAG chains were then isolated from the
medium and subsequently analyzed by size separation
Table 1. Retention times, GAG-priming and antiproliferative activity (ED50, lM)
Compound
Retention time (min)
GAG-priming
(T24 cells)a
Antiproliferative activity
(T24 cells, ED50, lM)
Antiproliferative activity
(HFL-1 cells, ED50, lM)
Series 1
3
4
5
6
18.84 0.09
13.49 0.00
15.28 0.01
25.04 0.06
0.75
0.62
0.60
0.93
>500
>500
>500
>500
210
220
270
240
Series 2
8
9
18.44 0.01
15.74 0.02
18.08 0.00
26.43 0.10
0.63
0.63
0.66
1.43
>500
>500
>500
>500
250
200
250
220
10
11
Series 3
14
15
21.64 0.00
22.86 0.01
18.91 0.02
22.31 0.00
23.74 0.01
22.70 0.03
0.47
0.75
0.43
0.45
0.36
0.54
>500
>500
>500
>500
>500
>500
270
300
280
250
220
300
16
17
18
19
a The proportion of GAG-priming is given as the integrated value of [35S]sulfate detected per min for the fractions containing GAG divided by the
integrated value for fractions of untreated cells.

1476
C.-O. Abrahamsson et al. / Carbohydrate Research 343 (2008) 1473–1477
chromatography.4 All treated cells secreted alkali sensi-
tive proteoglycans to the extracellular space, and modi-
fications of naphthalene ring with one or two amino acid
residues did not affect proteoglycan biosynthesis or
GAG priming ability. Only compound 11 initiated the
synthesis of free GAG chains (Table 1). As a compari-
son, the GAG-priming capability of compound 1, in
T24 cells, was 7.2.14
was stirred for 19 h at rt. The reaction was neutralized
with Amberlite IR-120 H+, filtered and evaporated
to dryness to give 3 as a white solid (112 mg, quant.).
HRESIMS calcd for C16H16O7Na [M+Na]+: 343.0794;
found 343.0811.
1.3. 6-(2-Carboxynaphthyl) b-D-xylopyranoside (8)
The structure of the aglycon has been shown to play
an important role for the priming ability, and for the
structure of the GAG chains synthesized by the xylopyr-
anosides.15,16 The compounds used in this study may be
internalized but not distributed to correct intracellular
compartments to take part in GAG biosynthesis or
may not be recognized by the GAG-priming enzymes.
The results are similar to earlier studies using fluorescent
analogs,17 and it may be assumed that these compounds
are unsuitable to function as GAG-primers. However,
we note that compound 11, which did initiate priming,
is the most nonpolar (as indicated by HPLC retention
times) of the investigated analogs and it is reasonable
to assume that the polarity is important for the cellular
transportation and targeting.
Compound 8 was synthesized according to the proce-
dure for 4 to give a white amorphous solid. HRESIMS
calcd for C16H16O7Na [M+Na]+: 343.0794; found
343.0806.
1.4. General procedure for the synthesis of compounds
4–6 and 9–11
Compounds 3 or 8 was dissolved in pyridine (15 mL).
Ac2O (0.71 mL, 7.5 mmol) was added, and the mixture
was stirred for 4 h at rt. The soln was concentrated and
the residue was dissolved in EtOAc and washed three
times with HCl (10% aq) and the organic layer was dried
with MgSO4 and concentrated. The intermediate
(10 mg, 0.022 mmol) and glycine tBu-ester hydrochlo-
ride (4.9 mg, 0.029 mmol) were dissolved in CH2Cl2
(1.5 mL). DMAP (4.1 mg, 0.034 mmol) was added
and the soln was stirred for 15 min. DIC (4.4 lL,
0.028 mmol) in CH2Cl2 (0.1 mL) was added and the
soln was stirred under N2 at rt for 3 h. The mixture
was concentrated and chromatographed (SiO2, 1:10 hep-
tane–EtOAc). The intermediate (11.2 mg, 0.020 mmol)
was dissolved in MeOH (2 mL). NaOH (0.40 mL,
1 M) was added and the mixture was stirred for 5 h.
The reaction was quenched with Amberlite IR-120 H+
and concentrated to give 4 as white amorphous solid.
1. Experimental
1.1. General methods
CH2Cl2 for reactions was dried by passing through a
column of Al2O3 (neutral, activity grade I). NMR-spec-
tra were collected at 400 MHz (1H) and 100 MHz (13C)
on a Bruker DRX 400 spectrometer. Chemical shifts are
reported in ppm with the residual solvent peaks (1H)
and solvent signals (13C) as reference. High-resolution
mass-spectra were collected using a Micromass Q-Tof
ESI or aFABMS JEOL SX-102. All the title compounds
were further purified using reverse phase preparative
HPLC using a Waters C18 symmetry column.
1.5. General procedure for solid-phase dipeptide synthesis
Dipeptides were synthesized using dry CH2Cl2 as a sol-
vent in a mechanically agitated reactor tube. Pre-loaded
Merrifield resins with Boc-protected glycine (0.5 mmol/
g, 0.020 mmol) and Boc-protected alanine (0.7 mmol/g,
0.021 mmol) were used. The resins were swelled for 10
min followed by Boc-deprotection. The deprotection
was performed twice using 4:1 TFA–CH2Cl2 (2 mL).
The resins were then washed with CH2Cl2. The second
Boc-protected amino acid was added alongside with
DMAP and CH2Cl2 and the tube was agitated for
15 min and DIC was added. The mixture was agitated
for 3 h after which the resins were washed with CH2Cl2
and MeOH. The Boc-group on the second amino acid
was removed as described earlier. The resins were
swelled for 10 min, followed by the addition of 4
(10.7 mg, 0.024 mmol), DMAP and CH2Cl2 and
agitated for 15 min. DIC was added and the tube was
agitated for 3 h. The resins were then washed as previ-
ously described. Swelling in CH2Cl2, followed by cleav-
1.2. 6-(1-Carboxynaphthyl) b-D-xylopyranoside (3)
Compound 29 (95 mg, 0.47 mmol) and 1,2,3,4-tetra-O-
acetyl-b-D-xylopyranose (299 mg, 0.94 mmol) were dis-
solved in CH2Cl2 (10 mL) under N2. Et3N (0.07 mL,
0.47 mmol) followed by BF3ÁOEt2 (0.295 mL, 2.34
mmol) were added. The soln was stirred at rt for 3 h
and quenched by the addition of Et3N, concentrated
and chromatographed (SiO2, 40:1 CH2Cl2–acetone).
The intermediate (178.7 mg, 0.39 mmol) was dissolved
in MeOH (10 mL). NaOMe (0.62 mL, 0.25 M) was
added to the soln and stirred for 45 min at rt. The reac-
tion was neutralized with solid CO2, concentrated and
chromatographed (SiO2, 1:1 CHCl3–MeOH). The inter-
mediate (117.2 mg, 0.35 mmol) was dissolved in MeOH
(15 mL). NaOH (7 mL, 1 M) was added and the soln

C.-O. Abrahamsson et al. / Carbohydrate Research 343 (2008) 1473–1477
1477
2. Okayama, M.; Kimata, K.; Suzuki, S. J. Biochem. 1973,
74, 1069–1073.
age from the resin using NaOMe (0.5 mmol in 2:3
CH2Cl2–MeOH, 2.5 mL) for 10 min. The reactions were
quenched with Amberlite IR-120 H+ and concentrated.
Deprotection of the ester was conducted using NaOH
for 5 h. The reactions were neutralized with Amberlite
IR-120 H+ followed by concentration and purification
by preparative HPLC.
3. 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,
1099–1104.
4. Mani, K.; Belting, M.; Ellervik, U.; Falk, N.; Svensson,
˚
G.; Sandgren, S.; Cheng, F.; Fransson, L.-A. Glycobiology
2004, 14, 387–397.
5. Jacobsson, M.; Ellervik, U. Tetrahedron Lett. 2002, 43,
6549–6552.
6. Jacobsson, M.; Ellervik, U.; Belting, M.; Mani, K. J. Med.
Chem. 2006, 49, 1932–1938.
Acknowledgements
7. Zhang, L.; David, G.; Esko, J. D. J. Biol. Chem. 1995, 270,
27127–27135.
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,
The Royal Physiographic Society in Lund and the
Bergvall Foundation supported this work.
´
8. Helting, T.; Roden, L. J. Biol. Chem. 1969, 244, 2790–
2798.
9. Anderson, L. C.; Thomas, D. G. J. Am. Chem. Soc. 1943,
65, 234–238.
`
10. Jullien, L.; Canceill, J.; Valeur, B.; Bardez, E.; Lefevre,
J.-P.; Lehn, J.-M.; Marchi-Artzner, V.; Pansu, R. J. Am.
Chem. Soc. 1996, 118, 5432–5442.
11. Lee, Y. S.; Rho, E. S.; Min, Y. K.; Kim, B. T.; Kim, K. H.
J. Carbohydr. Chem. 2001, 20, 503–506.
12. Grynkiewicz, G.; Achmatowicz, O.; Hennig, J.; Indulski,
J.; Klessig, D. F. Pol. J. Chem. 1993, 67, 1251–1254.
Supplementary data
´
13. Valko, K.; Plass, M.; Bevan, C.; Reynolds, D.; Abraham,
Supplementary data (complete experimental procedures,
product characterization and spectral data) associated
with this article can be found, in the online version, at
doi:10.1016/j.carres.2008.04.006.
M. H. J. Chromatogr. A 1998, 797, 41–55.
14. Johnsson, R.; Mani, K.; Ellervik, U. Bioorg. Med. Chem.
2007, 15, 2868–2877.
15. Schwartz, N. B.; Galligani, L.; Ho, P.-L.; Dorfman, A.
Proc. Nat. Acad. Sci. U.S.A. 1974, 71, 4047–4051.
16. Sobue, M.; Habuchi, H.; Ito, K.; Yonekura, H.; Oguri, K.;
Sakurai, K.; Kamohara, S.; Ueno, Y.; Noyori, R.; Suzuki,
S. Biochem. J. 1987, 241, 591–601.
References
17. Johnsson, R.; Mani, K.; Ellervik, U. Bioorg. Med. Chem.
Lett. 2007, 17, 2338–2341.
1. Schwartz, N. B. Trends Glycosci. Glycotechnol. 1995, 7,
429–445.