Glycosylation of Nα-lauryl-O-(β-d-xylopyranosyl)-l- serinamide as a saccharide primer in cells

Article abstract of DOI:10.1016/j.carres.2012.08.003

N^α-Lauryl-O-(β-d-xylopyranosyl)-l-serinamide (Xyl-Ser-C12) was synthesized as a saccharide primer to obtain oligosaccharides of glycosaminoglycan using the glycan biosynthetic potential of mouse osteosarcoma FBJ-S1 cells and Chinese hamster ovary (CHO) cells. The glycosylated products secreted into the culture medium were collected and analyzed by liquid chromatography-mass spectrometry and glycosidase digestion. The structure of the Xyl-Ser-C12 derivatives was investigated. Several glycosaminoglycan-type oligosaccharides, such as GalNAc-(GlcA-GlcNAc) _n-GlcA-Gal-Gal-Xyl-Ser-C12, were detected, and identified as intermediates of the biosynthesis of heparan sulfate glycosaminoglycans. Xyl-Ser-C12 exhibited greater acceptor activity for the glycosylation of glycosaminoglycan-type oligosaccharides than p-nitrophenyl-β-d- xylopyranoside.

Full text of DOI:10.1016/j.carres.2012.08.003

Carbohydrate Research 361 (2012) 33–40
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
a
Glycosylation of N -lauryl-O-(b-D-xylopyranosyl)-L-serinamide as a
saccharide primer in cells
Yinan Wang a, Tomoaki Kumazawa a, Keisuke Shiba a, Kenji Osumi b, Mamoru Mizuno b, Toshinori Sato a,
⇑
a Department of Biosciences and Informatics, Keio University, Hiyoshi, Yokohama 223-8522, Japan
b Laboratory of Glyco-organic Chemistry, The Noguchi Institute, Kaga 1-8-1, Itabashi-Ku, Tokyo 173-0003, Japan
a r t i c l e i n f o
a b s t r a c t
a
Article history:
Received 2 July 2012
Received in revised form 8 August 2012
Accepted 9 August 2012
Available online 17 August 2012
N -Lauryl-O-(b-
D
-xylopyranosyl)-
L
-serinamide (Xyl-Ser-C12) was synthesized as a saccharide primer to
obtain oligosaccharides of glycosaminoglycan using the glycan biosynthetic potential of mouse osteosar-
coma FBJ-S1 cells and Chinese hamster ovary (CHO) cells. The glycosylated products secreted into the cul-
ture medium were collected and analyzed by liquid chromatography–mass spectrometry and glycosidase
digestion. The structure of the Xyl-Ser-C12 derivatives was investigated. Several glycosaminoglycan-type
oligosaccharides, such as GalNAc-(GlcA-GlcNAc)n-GlcA-Gal-Gal-Xyl-Ser-C12, were detected, and identi-
fied as intermediates of the biosynthesis of heparan sulfate glycosaminoglycans. Xyl-Ser-C12 exhibited
greater acceptor activity for the glycosylation of glycosaminoglycan-type oligosaccharides than p-nitro-
Keywords:
Glycosaminoglycan
Saccharide primer
Oligosaccharide
FBJ osteosarcoma
phenyl-b-D-xylopyranoside.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
The biosynthesis of GAGs is initiated by the formation of a tetra-
saccharide linkage region composed of glucuronic acid-galactose-
galactose-xylose (b-GlcA-(1?3)-b-Gal-(1?3)-b-Gal(1?4)-b-Xyl-),
where Xyl is attached to a serine residue in the core protein.10 It
Oligosaccharides, which play an important role in cellular func-
tions, such as recognition, adhesion, and division, are attracting
increasing interest in biotechnology and drug development.1 This
expanding interest has led to an increased demand for the construc-
tion of saccharide libraries to provide research materials. Several
strategies involving direct extraction from natural resources, organic
synthesis, and enzymatic synthesis, have been developed in order to
construct oligosaccharide libraries.2 Recently, a way to obtain oligo-
saccharides by administering saccharide primers to animal cells has
been established. Dodecyl b-lactoside (Lac-C12),3–5 dodecyl-acetam-
has been found that exogenous b-
D-Xylosides such as p-nitro-
phenyl-b-
D
-xylopyranoside11,12 (Xyl-pNP), methylumbelliferyl-b-
D
-xylopyranoside (Xyl-MU),13 b-estradiol xylopyranoside,14 and
naphthol xyloside derivatives (XylNapOH)15–17 could act as artifi-
cial initiators for the elongation of glycosaminoglycan chains, and
primed the synthesis of glycosaminoglycan-type oligosaccharides
in several types of cells. Furthermore b-D-xylosides have been found
to inhibit the endogenous biosynthesis of glycosaminoglycan.12,18
a
ido-2-deoxy-b-
D
-glucopyranoside (GlcNAc-C12), dodecyl-b-
D-galac
In the present study, a novel saccharide primer, N -lauryl-O-(b-
topyranosyl-(1(4)-2-acetamido-2-deoxy-b-
D
-glucopyranoside
D-xylopyranosyl)-L-serinamide (Xyl-Ser-C12, Fig. 1), was employed
(LacNAc-C12),6 and azidododecyl b-lactoside (Lac-C12N3)7 have
been developed as saccharide primers to produce glycan libraries
involving ganglio-, globo-, and neolacto-series oligosaccharides.
Glycosaminoglycans (GAGs) are known to participate in specific
biological and physiological functions in cancer progression, angi-
ogenesis, cell adhesion, and anticoagulation.8,9 However, due to the
heterogeneity and diversity of GAG structures, the relationship be-
tween their biological function and structure is not well under-
as an initiator to synthesize glycosaminoglycan (GAG) oligosaccha-
rides using the GAG biosynthesis system of cells. Xyl-Ser-C12,
mimicking the region where xylose attached to the specific serine
residues of the core protein, was suggested to prime the elongation
of GAG chains. The glycosylation of Xyl-Ser-C12 by mammalian
cells was examined.
2. Results
stood.10 Therefore, it is important to develop
a method of
reconstructing GAG oligosaccharides and to know the GAG biosyn-
thetic pathway in cells for understanding their structure–function
relationships.
2.1. Chemical synthesis of Xyl-Ser-C12
a
The chemical synthesis of N -lauryl-O-(b-
D-xylopyranosyl)-
L-
serinamide (5) was shown in Scheme 1. 2,3,4-Tri-O-acetyl-
D
-
xylopyranosyl-trichloroacetimidate (1) served as the glycosyl
donor. Glycosylation of 1 with N- -carbobenzoxy- -serinamide
⇑
Corresponding author. Fax: +81 45 566 1447.
E-mail address: sato@bio.keio.ac.jp (T. Sato).
a
L
0008-6215/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2012.08.003

34
Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
spectra of SXN1 revealed peaks at m/z 178.7 corresponding to
the C1 fragment ion indicating that SXN1 was Hex-Xyl-Ser-C12.
The C2 fragment ion with m/z 340.6 in the MS/MS spectra of
SXN2, suggested the existence of two hexose residues in SXN2 (Ta-
ble 2). The non-reducing hexose residues of SXN1 and SXN2 were
cleaved by an exo-b-galactosidase (Supplementary Fig. 3A). Thus,
SXN1 and SXN2 were identified as Galb-Xyl-Ser-C12 and Galb-
Galb-Xyl-Ser-C12, respectively. Six products with m/z 659.7
[MꢁH]ꢁ, 870.8 [MꢁH]ꢁ, 1032.9 [MꢁH]ꢁ, 917.9 [MꢁH]ꢁ, 1120.8
[MꢁH]ꢁ, and 749.4 [Mꢁ2H]2ꢁ were detected in the acidic fraction
(Table 1). SXA1 with m/z 659.7 was suggested to be the product
having a sulfated group in Gal-Xyl-Ser-C12. The B1 fragment with
m/z 240.7 of SXA1 showed that the Gal residue was sulfated (Table
2). SXA2 was considered to be NeuAc-Gal-Xyl-Ser-C12 based on
the B1 fragment (m/z 290.2) and the B2 fragment (m/z 451.5).
SXA3 was deduced to be NeuAc-Gal-b-Gal-b-Xyl-Ser-C12 from
the B1 fragment (m/z 289.9) and the B2 fragment (m/z 451.5). Both
of the N-acetylneuraminic acid residues of SXA2 and SXA3 were
a
Figure 1. The structure of N -lauryl-O-(b-D-xylopyranosyl)-L-serinamide (Xyl-Ser-
C12).
(Z-Ser-NH2) was carried out in the presence of BF3ꢀEt2O. The
obtained product 2 was considered to be an orthoester, and was
subjected to TMSOTf-catalyzed isomerization to obtain b-anomer
3 predominantly. Conversion of 3 into 4 was carried out in a
two-step synthesis involving the removal of a carbobenzyloxy
group (Cbz) by catalytic hydrogenation over Pd/C and condensa-
tion with lauric acid in 41% overall yield for the two steps. Finally,
a
N -lauryl-O-(b-
D
-xylopyranosyl)-
L-serinamide (5) was obtained by
cleaved by
linkages of N-acetylneuraminic acid to galactose in SXA2 and SXA3
were determined to be -(2?3). SXA4 was deduced to be GlcA-b-
a-(2?3)-neuraminidase (Supplementary Fig. 3B). Thus,
O-deacetylation with methanolic NaOMe and purification by silica
gel chromatography.
a
Gal-b-Gal-b-Xyl-Ser-C12, based on the C2 fragment with m/z 354.8
and the digestion by b-glucuronidase (Supplementary Fig. 3C). In
the MS/MS spectra of SXA5, the elongation of HexNAc and HexA
to SXN2 was indicated by the Z4 fragment with m/z 899.6 and
the Y3 fragment with m/z 742.0 (Table 2). The structure of SXA5
was considered as HexNAc-HexA-Gal-b-Gal-b-Xyl-Ser-C12. SXA6
with m/z 749.4 [Mꢁ2H]2ꢁ was determined to be (HexNAc-
GlcA)2-Gal-b-Gal-b-Xyl-Ser-C12 from the Z6 fragment with m/z
1278.5 and the Y5 fragment with m/z 1120.6 (Table 2).
2.2. Glycosylation of Xyl-Ser-C12 by FBJ-S1 cells
The murine osteosarcoma cell line FBJ-S119,20 was employed to
investigate the extent of the glycosylation of Xyl-Ser-C12. FBJ-S1
cells were incubated with 50 lM Xyl-Ser-C12 for 48 h, and the gly-
cosylated products were isolated from the culture medium using a
Sep-Pak C18 cartridge. These products were separated into neutral
and acidic fractions with an aminopropyl SPE cartridge. The struc-
tures of the neutral and acidic products were analyzed by LC–MS.
As shown in Table 1, two neutral products and six acidic products
were detected and their structures were predicted based on MS/MS
assignments (Tables 1 and 2 and Supplementary Fig. 1) and glyco-
sidase digestion (Supplementary Fig. 3).
To identify the isomer of the non-reducing HexNAc in SXA5 and
SXA6 and the linkage between the HexNAc and HexA, digestion
experiments using exo-type b-acetylhexosaminase and a-acetylga-
lactosaminidase were carried out. Since SXA5 and SXA6 were not
digested by b-acetylhexosaminase, it was suggested that HexNAc
was conjugated to HexA by an
Fig. 3D). Furthermore, since SXA5 and SXA6 were also hydrolyzed
by -acetylgalactosaminidase, HexNAc of the non-reducing end
was suggested to be -GalNAc (Supplementary Fig. 3E). The major
a-linkage (Supplementary
2.3. Analyses of the structures of the glycosylated products
derived from FBJ-S1 cells
a
a
As shown in Table 1, two neutral products with m/z 615.9
[M+Cl]ꢁ (SXN1) and 777.9 [M+Cl]ꢁ (SXN2) were detected. MS/MS
product was SXA5 (GalNAc-
a
-GlcA-b-Gal-b-Gal-b-Xyl-Ser-C12).
O
AcO
AcO
O
O
BF3•Et2O, Z-Ser-NH2
O
TMSOTf
CH2Cl2, 0°C
NHCbz
AcO
AcO
O
O
C
NH
CCl3
CH3CN, -20 °C~0 °C
AcO
H2N
O
1
2
O
AcO
AcO
DIEA,
PyBOP,
HO
O
H2, Pd/C
AcO
O
H2N
DMF, r.t.
THF / EtOH, r.t
NHCbz
O
3
4
O
O
HO
HO
AcO
AcO
O
O
HO
AcO
NaOMe
H2N
H2N
N
H
C
O
N
H
C
O
MeOH, r.t
O
O
5
a
Scheme 1. Synthesis of N -lauryl-O-(b-
D-xylopyranosyl)-L-serinamide (Xyl-Ser-C12).

Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
35
Table 1
Xyl-Ser-C12 glycosylated products from FBJ cells detected by LC–MS
Product
Retention time (min)
Predicted Structures
Ion type
Observed
mass (m/z)
Calculated
mass (m/z)
Proportiona
(%)
SXN1
SXN2
25.3
32.0
Galb-Xyl-Ser-C12
Galb-Galb-Xyl-Ser-C12
[M+Cl]ꢁ
[M+Cl]ꢁ
615.9
777.9
615.3
777.3
51.8
48.2
SXA1
SXA2
SXA3
SXA4
SXA5
SXA6
29.7
33.0
34.1
34.3
37.6
39.4
GalbS-Xyl-Ser-C12
[MꢁH]ꢁ
[MꢁH]ꢁ
[MꢁH]ꢁ
[MꢁH]ꢁ
[MꢁH]ꢁ
[Mꢁ2H]2ꢁ
659.7
870.8
1032.9
917.9
1120.8
749.4
659.3
870.8
1032.8
917.6
1120.5
749.3
2.6
NeuAc
NeuAc
a
a
-Galb-Xyl-Ser-C12
-Galb-Galb-Xyl-Ser-C12
25.5
10.8
5.1
52.5
3.5
GlcAb-Galb-Galb-Xyl-Ser-C12
GalNAc
GalNAc
a
a
-GlcAb-Galb-Galb-Xyl-Ser-C12
-(GlcAb-GlcNAc )-GlcAb-Galb-Galb-Xyl-Ser-C12
a
a
The proportion of each product represents the ratio of each chromatogram area to total area (Normalized by protein amount). The data for the neutral products and the
acidic products were calculated separately.
Table 2
MS/MS assignment of Xyl-Ser-C12 glycosylated products from FBJ-S1 cells
Product
Fragments
SXN1
SXN2
SXA1
SXA2
SXA3
SXA4
SXA5
579.1 ([MꢁH]ꢁ), 310.9 ([C2ꢁH]ꢁ), 267.1 ([Z0ꢁH]ꢁ), 178.7 ([C1ꢁH]ꢁ)
742.3 ([MꢁH]ꢁ), 579.0 ([Y2ꢁH]ꢁ), 472.7 ([C3ꢁH]ꢁ), 340.6 ([C2ꢁH]ꢁ), 322.9 ([B2ꢁH]ꢁ)
390.9 ([C2ꢁH]ꢁ), 372.9 ([B2ꢁH]ꢁ), 258.9 ([C1ꢁH]ꢁ), 240.7 ([B1ꢁH]ꢁ)
601.9 ([C3ꢁH]ꢁ), 451.4 ([B2ꢁH]ꢁ), 290.2 ([B1ꢁH]ꢁ), 267.3 ([Z0ꢁH]ꢁ)
764.2 ([C4ꢁH]ꢁ), 742.3 ([Y3ꢁH]ꢁ), 614.2 ([B3ꢁH]ꢁ), 578.4([Y2ꢁH]ꢁ), 451.5 ([B2ꢁH]ꢁ), 289.8 ([B1ꢁH]ꢁ)
740.2 ([Y3ꢁH]ꢁ), 649.4 ([C4ꢁH]ꢁ), 517.1 ([C3ꢁH]ꢁ), 498.8 ([B3ꢁH]ꢁ), 354.8 ([C2ꢁH]ꢁ), 336.8 ([B2ꢁH]ꢁ)
918.0 ([Y4ꢁH]ꢁ), 899.6 ([Z4ꢁH]ꢁ), 852.1 ([C5ꢁH]ꢁ), 742.0 ([Y3ꢁH]ꢁ), 720.3 ([C4ꢁH]ꢁ)
558.9 ([C3ꢁH]ꢁ), 395.4 ([C2ꢁH]ꢁ)
SXA6
1278.5 ([Z6ꢁH]ꢁ), 1231.6 ([C7ꢁH]ꢁ), 1120.6 ([Y5ꢁH]ꢁ), 1099.4 ([C6ꢁH]ꢁ), 937.4 ([C5ꢁH]ꢁ), 899.5 ([Z4ꢁH]ꢁ)
599.1 ([C3ꢁH]ꢁ), 395.7 ([C2ꢁH]ꢁ)
FBJ-S1 cells were also treated with 50
l
M p-nitrophenyl-b-
D
-
m/z 178.7 and b-galactosidase digestion (Supplementary Fig. 6A).
CXN2 was deduced to be Gal-b-Gal-b-Xyl-Ser-C12 from MS/MS
spectra (B2 and Y2 fragments) and b-galactosidase digestion (Sup-
plementary Fig. 6A).
Among the acidic products, CXA1 with m/z 659.7 was considered
to be the structure elongating the sulfated Gal residue to Xyl-Ser-
C12 from the B1 fragment with m/z 240.7. CXA2 was deduced to be
NeuAc-Gal-Xyl-Ser-C12. The B1 fragment (m/z 289.6) and the B2
fragment (m/z 451.9) suggested the presence of NeuAc-Gal in
CXA2. CXA3 was deduced to be NeuAc-Gal-b-Gal-b-Xyl-Ser-C12,
based on the B1 fragment (m/z 290.0) and the B3 fragment (m/z
614.0). The linkage of NeuAc-Gal in CXA2 and CXA3 was determined
xylopyranoside, for 2 days. Five glycosylated products were de-
tected by LC–MS (Table 3) and their structures were deduced based
on MS/MS spectra (Table 4). SPN1 with m/z 468.7 [M+Cl]ꢁ was sug-
gested to be Hex-Xyl-pNP based on the B1 fragment with m/z
160.8. The B3 fragment ion with m/z 455.1 in MS/MS spectra of
SPN2, suggested that the structure of SPN2 is Hex-Hex-Xyl-pNP.
SPA1 with m/z 723.9 was considered to be NeuAc-Hex-Xyl-pNP
based on the B1 fragment (m/z 290.0) and the B3 fragment (m/z
566.1). MS/MS spectra of SPA2 revealed a Z4 fragment (m/z
752.1), Y3 fragment (m/z 595.1), C2 fragment (m/z 396.1), and C3
fragment (m/z 558.1), suggesting that SPA2 was HexNAc-HexA-
Hex-Hex-Xyl-pNP. For the structure of SPA3, the elongation of Hex-
NAc-HexA to SPA2 was indicated by the Z6 fragment (m/z 1131.3)
and Z5 fragment (m/z 956.2) in the MS/MS spectra. SPA3 was deter-
mined as (HexNAc-HexA)2-Hex-Hex-Xyl-pNP.
as a-(2?3) with hydrolysis using a2-3 neuraminidase (Supplemen-
tary Fig. 6B). CXA4 was deduced to be HexA-Gal-Gal-Xyl-Ser-C12
based on the C2 fragment with m/z 355.0. The HexA residue was
determined as GlcA by the hydrolysis with b-glucuronidase (Supple-
mentary Fig. 6C). The structure of CXA5 was identified as HexNAc-
HexA-Gal-Gal-Xyl-Ser-C12 from the Z4 fragment (m/z 899.5) and
Y3 fragment (m/z 742.2). A sulfated product, CXA6, was also de-
tected, and the sulfated position was determined as Gal linked to
Xyl from the Y4 fragment (m/z 997.3), Z4 fragment (m/z 979.2), Y3
fragment (m/z 820.9), and Y2 fragment (m/z 659.8). CXA7 with m/z
648.3 was deduced to be GlcA-HexNAc-GlcA-Gal-Gal-Xyl-Ser-C12
from the Z5 fragment (m/z 1120.2) and C1 fragment (m/z 192.8).
2.4. Glycosylation of Xyl-Ser-C12 by CHO cells
The glycosylation of Xyl-Ser-C12 was also examined using CHO
cells. Two neutral products and 14 acidic products were detected,
and their structures were determined as shown in Table 5. MS/
MS assignments are shown in Table 6. CXN1 (m/z 615.9 [M+Cl]ꢁ)
was deduced to be Gal-b-Xyl-Ser-C12 from the C1 fragment with
Table 3
Xyl-pNP glycosylated products from FBJ-S1 cells detected by LC–MS
Product
Retention time (min)
Predicted structures
Ion type
Observed mass (m/z)
Calculated mass (m/z)
Proportiona (%)
SPN1
SPN2
20.6
29.1
Hex-Xyl-pNP
Hex-Hex-Xyl-pNP
[M+Cl]ꢁ
[M+Cl]ꢁ
468.7
630.8
468.1
630.1
18.5
81.5
SPA1
SPA2
SPA3
33.1
35.1
41.2
NeuAc-Hex-Xyl-pNP
HexNAc-HexA-Hex-Hex-Xyl-pNP
HexNAc-HexA-HexNAc-HexA-Hex-Hex-Xyl-pNP
[MꢁH]ꢁ
[MꢁH]ꢁ
[Mꢁ2H]2ꢁ
723.9
974.0
676.9
723.2
973.3
676.7
6.1
92.5
1.5
a
The proportion of each product represents the ratio of each chromatogram area to total area (Normalized by protein amount). The data for the neutral products and the
acidic products were calculated separately.

36
Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
Table 4
MS/MS assignment of Xyl-pNP glycosylated products from FBJ-S1 cells
Product
Fragments
SPN1
SPN2
SPA1
SPA2
SPA3
432.8 ([MꢁH]ꢁ), 160.8 ([B1ꢁH]ꢁ)
594.7 ([MꢁH]ꢁ), 455.1 ([B3ꢁH]ꢁ)
583.9 ([C3ꢁH]ꢁ), 566.1 ([B3ꢁH]ꢁ), 290.0 ([B1ꢁH]ꢁ)
835.3 ([C5ꢁH]ꢁ), 752.1 ([Z4ꢁH]ꢁ), 720.2 ([C4ꢁH]ꢁ), 595.1 ([Y3ꢁH]ꢁ), 558.1 ([C3ꢁH]ꢁ), 396.1 ([C2ꢁH]ꢁ)
1213.6 ([B7ꢁH]ꢁ), 1131.3 ([Z6ꢁH]ꢁ), 1099.2 ([C6ꢁH]ꢁ), 956.2 ([Z5ꢁH]ꢁ), 937.2 ([C5ꢁH]ꢁ), 752.2 ([Z4ꢁH]ꢁ), 598.7 ([C3ꢁH]ꢁ)
Table 5
Xyl-Ser-C12 glycosylated products from CHO cells detected by LC–MS
Product
Retention
Predicted structures
Ion type
Observed
Calculated
Proportiona (%)
time (min)
mass (m/z)
mass (m/z)
CXN1
CXN2
25.1
30.4
Galb-Xyl-Ser-C12
Galb-Galb-Xyl-Ser-C12
[M+Cl]ꢁ
[M+Cl]ꢁ
615.9
777.9
615.3
777.3
59.7
40.3
CXA1
CXA2
CXA3
CXA4
CXA5
CXA6
CXA7
CXA8
CXA9
CXA10
CXA11
CXA12
CXA13
CXA14
30.6
31.6
32.8
33.1
33.9
34.7
35.3
39.1
41.1
44.9
46.2
47.1
49.7
50.7
GalbS-Xyl-Ser-C12
[MꢁH]ꢁ
659.7
870.9
1032.9
917.8
1121.0
600.3
648.3
749.9
789.9
837.9
939.2
979.2
1027.3
1128.9
659.3
870.4
1032.5
917.4
1120.5
599.7
647.8
749.3
789.3
837.3
938.8
978.9
1026.9
1128.4
3.3
NeuAc
NeuAc
a
-Galb-Xyl-Ser-C12
[MꢁH]ꢁ
14.0
14.8
23.6
24.6
3.7
3.8
9.7
0.1
0.4
1.3
0.1
0.2
0.5
a-Galb-Galb-Xyl-Ser-C12
[MꢁH]ꢁ
GlcAb-Galb-Galb-Xyl-Ser-C12
[MꢁH]ꢁ
GalNAc
GalNAc
a
a
-GlcAb-Galb-Galb-Xyl-Ser-C12
-GlcAb-Galb-GalbS-Xyl-Ser-C12
[MꢁH]ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
[Mꢁ2H]2ꢁ
GlcA-GlcNAc
a
-GlcAb-Galb-Galb-Xyl-Ser-C12
-(GlcA-GlcNAc )-GlcAb-Galb-Galb-Xyl-Ser-C12
S-(GlcA-GalNAc )-GlcAb-Galb-Galb-Xyl-Ser-C12
) 2-GlcAb-Galb-Galb-Xyl-Ser-C12 [
-(GlcA-GalNAc )2-GlcAb-Galb-Galb-Xyl-Ser-C12
S-(GlcA-GalNAc )2-GlcAb-Galb-Galb-Xyl-Ser-C12
) 3-GlcAb-Galb-Galb-Xyl-Ser-C12
-(GlcA-GalNAc )3-GlcAb-Galb-Galb-Xyl-Ser-C12
GalNAc
GalNAc
(GlcA-GalNAc
GalNAc
GalNAc
(GlcA-GalNAc
a
a
a
a
a
a
a
a
a
a
GalNAc
a
a
a
The proportion of each product represents the ratio of each chromatogram area to total area (Normalized by protein amount). The data for the neutral products and the
acidic products were calculated separately.
Table 6
MS/MS assignment of Xyl-Ser-C12 glycosylated products from CHO cells
Product
Fragments
CXN1
CXN2
CXA1
CXA2
CXA3
CXA4
CXA5
CXA6
579.2 ([MꢁH]ꢁ), 310.8 ([C2ꢁH]ꢁ), 267.0 ([Z0ꢁH]ꢁ), 178.7 ([C1ꢁH]ꢁ)
742.3 ([MꢁH]ꢁ), 579.2([Y2ꢁH]ꢁ), 473.2 ([C3ꢁH]ꢁ), 341.3 ([C2ꢁH]ꢁ), 322.9 ([B2ꢁH]ꢁ)
390.9 ([C2ꢁH]ꢁ), 373.4 ([B2ꢁH]ꢁ), 259.3 ([C1ꢁH]ꢁ), 240.7 ([B1ꢁH]ꢁ)
602.1 ([C3ꢁH]ꢁ), 451.9 ([B2ꢁH]ꢁ), 289.6 ([B1ꢁH]ꢁ), 267.2([Z0ꢁH]ꢁ)
764.0 ([C4ꢁH]ꢁ), 742.2([Y3ꢁH]ꢁ), 614.0 ([B3ꢁH]ꢁ), 578.2 ([Y2ꢁH]ꢁ), 290.0 ([B1ꢁH]ꢁ)
739.7 ([Y3ꢁH]ꢁ), 649.1 ([C4ꢁH]ꢁ), 516.9 ([C3ꢁH]ꢁ), 498.8 ([B3ꢁH]ꢁ), 355.0 ([C2ꢁH]ꢁ), 336.7 ([B2ꢁH]ꢁ)
918.1 ([Y4ꢁH]ꢁ), 899.4 ([Z4ꢁH]ꢁ), 852.0 ([C5ꢁH]ꢁ), 742.2 ([Y3ꢁH]ꢁ), 720.4 ([C4ꢁH]ꢁ), 559.1 ([C3ꢁH]ꢁ)
1120.6 ([Y5ꢁH]ꢁ), 1099.4 ([C6ꢁH]ꢁ), 997.3 ([Y4ꢁH]ꢁ), 979.2 ([Z4ꢁH]ꢁ]), 932.2 ([C5ꢁH]ꢁ), 899.8 ([Z4ꢁH]ꢁ),
820.9 ([Y3ꢁH]ꢁ), 659.8 ([Y2ꢁH]ꢁ), 620.0 ([B3ꢁH]ꢁ), 539.5 ([B3ꢁH]ꢁ), 458.0 ([B2ꢁH]ꢁ), 378.0 ([B2ꢁH]ꢁ)
1120.2 ([Y5ꢁH]ꢁ), 1028.3 ([C6ꢁH]ꢁ), 899.5 ([Z4ꢁH]ꢁ), 880.3 ([B5ꢁH]ꢁ), 734.2 ([C4ꢁH]ꢁ)
554.0 ([B3ꢁH]ꢁ), 396.0 ([C2ꢁH]ꢁ), 192.8 ([C1ꢁH]ꢁ)
CXA7
CXA8
1278.6 ([Z6ꢁH]ꢁ), 1231.2 (C7ꢁH]ꢁ), 1120.6 ([Y5ꢁH]ꢁ), 1100.2 ([C6ꢁH]ꢁ), 937.4 ([C5ꢁH]ꢁ), 899.4 ([Z4ꢁH]ꢁ),
775.7 ([C4ꢁH]ꢁ), 740.7 ([Y3ꢁH]ꢁ), 599.2 ([C3ꢁH]ꢁ), 396.1 ([C2ꢁH]ꢁ)
CXA9
1296.4 ([Y6ꢁH]ꢁ), 1120.3 ([Y5ꢁH]ꢁ), 917.2 ([Y4ꢁH]ꢁ), 899.5 ([Z4ꢁH]ꢁ), 775.6 ([C4ꢁH]ꢁ)
661.2 ([B3ꢁH]ꢁ), 458.0 ([B2ꢁH]ꢁ), 299.6 ([C1ꢁH]ꢁ), 281.9 ([B1ꢁH]ꢁ)
CXA10
CXA11
CXA12
CXA13
CXA14
1500.4 ([Y7ꢁH]ꢁ), 1407.3 ([C8ꢁH]ꢁ), 1296.5 ([Y6ꢁH]ꢁ), 1120.3 ([Y5ꢁH]ꢁ), 1113.4 ([C6ꢁH]ꢁ), 933.2 ([B5ꢁH]ꢁ),
899.2 ([Z4ꢁH]ꢁ), 775.3 ([C4ꢁH]ꢁ), 741.6 ([Y3ꢁH]ꢁ), 572.0 ([C3ꢁH]ꢁ), 396.2 ([C2ꢁH]ꢁ)
1658.6 ([Z8ꢁH]ꢁ), 1499.7 ([Y7ꢁH]ꢁ), 1316.7 ([C7ꢁH]ꢁ), 1296.5 ([Y6ꢁH]ꢁ), 1136.2 ([B6ꢁH]ꢁ), 1120.5 ([Y5ꢁH]ꢁ),
978.3 ([C5ꢁH]ꢁ), 917.5 ([Y4-H]ꢁ), 775.1 ([C4ꢁH]ꢁ), 741.6 ([Y3ꢁH]ꢁ), 599.3 ([C3ꢁH]ꢁ), 396.2 ([C2ꢁH]ꢁ)
1658.6 ([Z8ꢁH]ꢁ), 1499.6 ([Y7ꢁH]ꢁ), 1378.6 ([B7-H]ꢁ), 1296.5 ([Y6ꢁH]ꢁ), 1234.6 ([C6ꢁH]ꢁ), 1120.3 ([Y5ꢁH]ꢁ),
1040.2 ([B5ꢁH]ꢁ), 917.3 ([Y4ꢁH]ꢁ), 869.1 ([Z8ꢁ2H]2ꢁ), 837.2 ([B4ꢁH]ꢁ), 789.2 ([Y7ꢁ2H]2ꢁ), 458.1 ([B2ꢁH]ꢁ)
1879.5 ([Y9ꢁH]ꢁ), 1675.5 ([Y8ꢁH]ꢁ), 1499.7 ([Y7ꢁH]ꢁ), 1312.4 ([C7ꢁH]ꢁ), 1296.5 ([Y6ꢁH]ꢁ), 1154.2 ([C6ꢁH]ꢁ),
1120.5 ([Y5ꢁH]ꢁ), 950.9 ([C5ꢁH]ꢁ), 917.2 ([Y4ꢁH]ꢁ), 776.1 ([C4ꢁH]ꢁ), 571.7 ([C3ꢁH]ꢁ), 396.0 ([C2ꢁH]ꢁ)
1878.9 ([Y9ꢁH]ꢁ), 1675.7 ([Y8ꢁH]ꢁ), 1515.3 ([B8ꢁH]ꢁ), 1499.7 ([Y7ꢁH]ꢁ), 1339.3 ([B7ꢁH]ꢁ), 1296.5 ([Y6ꢁH]ꢁ),
1136.2 ([B6ꢁH]ꢁ), 1120.4 ([Y5ꢁH]ꢁ), 960.4 ([B5ꢁH]ꢁ), 917.3 ([Y4ꢁH]ꢁ), 757.2 ([B4ꢁH]ꢁ), 581.3 ([B3ꢁH]ꢁ), 396.1 ([C2ꢁH]ꢁ)
CXA8 with m/z 749.9 was deduced to be (HexNAc-GlcA)2-Gal-Gal-
Xyl-Ser-C12 from MS/MS spectra with the Z6 fragment (m/z
1278.6) and Y5 fragment (m/z 1120.2). Longerglycosylatedproducts,
such as CXA11-14 and (HexNAc-GlcA)n-Gal-Gal-Xyl-Ser-C12, were
detected for CHO cells. Sulfated products such as CXA9 and CXA12
were also detected. The terminal HexNAc of CXA9 was considered
to be sulfated based on the Y6 fragment (m/z 1296.4) and B1 frag-
ment (m/z 281.9). The terminal HexNAc of CXA12 was deduced to
be sulfated from the Z8 fragment with m/z 1658.6 and the dominant
presence of the Y7 fragment with m/z 1499.6.

Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
37
Table 7
Xyl-pNP glycosylated products from CHO cells detected by LC–MS
Product
Retention time (min)
Predicted structures
Ion type
Observed mass (m/z)
Calculated mass (m/z)
Proportiona (%)
CPN1
CPN2
19.0
28.5
Hex-Xyl-pNP
Hex-Hex-Xyl-pNP
[M+Cl]ꢁ
[M+Cl]ꢁ
468.8
630.9
468.1
630.1
12.4
87.6
CPA1
CPA2
CPA3
32.6
35.4
42.5
NeuAc-Hex-Xyl-pNP
HexNAc-HexA-Hex-Hex-Xyl-pNP
HexNAc-HexA-HexNAc-HexA-Hex-Hex-Xyl-pNP
[MꢁH]ꢁ
[MꢁH]ꢁ
[Mꢁ2H]2ꢁ
723.4
973.6
676.5
723.2
973.3
676.7
8.5
90.8
0.7
a
The proportion of each product represents the ratio of each chromatogram area to total area (Normalized by protein amount). The data for the neutral products and the
acidic products were calculated separately.
Table 8
MS/MS assignment of Xyl-pNP glycosylated products from CHO cells
Product
Fragments
CPN1
CPN2
CPA1
CPA2
CPA3
432.7 ([MꢁH]ꢁ), 178.8 ([C1ꢁH]ꢁ)
594.0 ([MꢁH]ꢁ), 455.0 ([B3ꢁH]ꢁ)
583.9 ([C3ꢁH]ꢁ), 470.1 ([C2ꢁH]ꢁ), 289.9 ([B1ꢁH]ꢁ)
834.1 ([C5ꢁH]ꢁ), 752.1 ([Z4ꢁH]ꢁ), 719.9 ([C4ꢁH]ꢁ), 595.0 ([Y3ꢁH]ꢁ), 558.1 ([C3ꢁH]ꢁ), 396.0 ([C2ꢁH]ꢁ)
1213.4 ([B7ꢁH]ꢁ), 1131.1 ([Z6ꢁH]ꢁ), 1099.4 ([C6ꢁH]ꢁ), 956.2 ([Z5ꢁH]ꢁ), 937.2 ([C5ꢁH]ꢁ), 752.2 ([C4ꢁH]ꢁ)
598.7 ([Z3ꢁH]ꢁ), 396.9 ([C2ꢁH]ꢁ)
HexNAc residues of the non-reducing end in products such as
CXA5 derived from CHO cells were determined to be -GalNAc
by hydrolysis with -acetylgalactosaminidase (Supplementary
Fig. 6E). Since products with several repeating units of HexNAc-
GlcA, such as CXA8, were cleaved by heparitinase, it was indicated
that Xyl-Ser-C12 was glycosylated by the biosynthesis of heparan
sulfate (Supplementary Fig. 6F). The major products were CXA4
In the present study, Xyl-Ser-C12 and Xyl-pNP were added to
cells, and the glycosylated products were isolated and analyzed.
The Xyl-Ser-C12 primer produced more and longer species of gly-
cosylated products, than did Xyl-pNP, especially in CHO cells. We
hypothesized that the hydrophobicity of the aglycones may influ-
ence the uptake of the b-Xylosides by cells, the localization in sub-
cellular compartments, and the secretion of the glycosylated
products,23,24 consequently affecting the composition of the glycos-
ylated products, though the current study focused on the synthesis
of Xyl-Ser-C12 and the structures of the glycosylated products.
Longer products (CXA11-14) having several repeating units of
disaccharides and more sulfated products (CXA9, 12) were detected
with Xyl-Ser-C12 than Xyl-pNP. The profile of Xyl-pNP-primed oli-
gosaccharides follows a similar pattern to that reported in the liter-
ature.21 When FBJ-S1 cells were treated with Xyl-Ser-C12, eight
glycosylated products were detected. In contrast, only five products
were detected after treatment with Xyl-pNP. Sulfated Gal-Xyl-R
(SXA1) was detected in products from Xyl-Ser-C12 but not in prod-
ucts from Xyl-pNP. These results suggested that Xyl-Ser-C12 exhib-
ited better acceptor activity for the glycosylation of xyloside to
obtain GAG-type oligosaccharide or intermediates of GAG chains.
The differences in the variety of glycosylated products between
CHO and FBJ-S1 cells may partially reflect the profile of biosynthesis
in these cells and the activity of glycoside transferases.
a
a
(GlcA-b-Gal-b-Gal-b-Xyl-Ser-C12, 23.6%) and CXA5 (GalNAc-
a-
GlcA-b-Gal-b-Gal-b-Xyl-Ser-C12, 24.6%).
The glycosylation reaction in CHO cells was also examined
using Xyl-pNP. Under reaction conditions similar to those for
Xyl-Ser-C12, five products were detected (Table 7). Their struc-
tures were predicted based on MS/MS assignments (Table 8) and
the exploration of structures of Xyl-pNP-initiated products by
Salimath et al.21 Only short products were detected for Xyl-pNP
compared to Xyl-Ser-C12. The major product was CPA2, GalNAc-
GlcA-Gal-Gal-Xyl-pNP (92.5% of acidic products). A sialyted prod-
uct (CPA1) was also detected.
2.5. The viability of cells treated with Xyl-Ser-C12
FBJ-S1 or CHO cells were treated with 50
lM Xyl-Ser-C12,
50
lM Xyl-pNP, or an equal volume of DMSO in culture medium
for 2 days. Cytotoxicity was investigated by cell morphology and
WST assays. Compared with the control without saccharide prim-
ers, no obvious toxic effects were observed (Supplementary Fig. 7).
It has been reported that the sialyted structure a-NeuAc-(2?3)-
Gal-Xyl-R, which is unrelated to the biosynthesis pathway of GAG,
was produced when various cells were fed with b-xylosides, such
as Xyl-pNP and Xyl-MU.21,25–27 Etchison et al. demonstrated that
3. Discussion
a
-(2?3)-sialyltransferase was co-localized with GAG core-specific
galactosyltransferase I at the same Golgi apparatus.10,28 Therefore
NeuAc-Gal-Xyl was synthesized in cells by the recognition of
A series of b-
D
-xylosides including p-nitrophenyl-b-
D-xylopy-
ranoside11 (Xyl-pNP), methylumbelliferyl-b-
D
-xylopyranoside,13
b-estradiol xylopyranoside,14 and naphthol xyloside derivatives
(XylNapOH)15–17 have been developed as initiators of GAGs or
inhibitors of endogenous GAGs. Most of these xylosides primed
the production of short oligosarccharides, mainly consisting of
xylosides with Gal-Xyl-R, Gal-Gal-Xyl-R, and NeuAc-Gal-Gal-Xyl-
R. High molecular weight GAG chains usually comprise minor xylo-
side products.22 Thus, we attempted to develop a novel saccharide
primer (Xyl-Ser-C12) for synthesizing GAG-type oligosaccharides.
The saccharide primer synthesized in the present study contained
xylose, serine, and dodecanoyl groups. Such a chemical structure
was expected to function as a good artificial substrate for glycosyl-
transferases to obtain GAG-type oligosaccharides.
a-(2?3)-sialyltransferase. However, another sialyted product
(SXA3 or CXA3), -NeuAc-(2?3)-Gal-Gal-Xyl-R, has not been
a
reported. Detection of these products may suggest the co-local-
ization of GAG core-specific glucuronic acid transferase I and
a
-(2?3)-sialyltransferase.10,28
a
-GalNAc-capped xylosides, such as SXA5, SXA6, CXA5, and
CXA8, were also detected in the glycosylated products. Miura et al.
have previously reported that -GalNAc-capped structures were
produced by several cell lines when incubated with 4-methylum-
belliferyl (MU) or p-nitrophenyl (pNP) b-
-xylosides.21,25 Kitagawa
et al. have clarified that glycosyltransferase EXTL2, which transfers
-GlcNAc to GlcA of the GAG linker region in the biosynthesis of
a
D
a

38
Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
heparan sulfate, also transferred
a
-GalNAc.29,30 Though the signifi-
Chemical Industries) were added into a 2-neck flask. The flask
was evacuated, dried, and refilled with argon. After 7.0 mL of anhy-
drous acetonitrile was added, the reaction system was chilled to
cance of the
that the addition of
sion of the GAG chain.
Artificial b- -Xylosides such as Xyl-pNP and Xyl-MU preferen-
a-GalNAc-cap’s modification is still unknown it seems
a-GalNAc resulted in termination of the exten-
ꢁ20 °C. Then 2.9
l
L of BF3ꢀEt2O (0.023
lmol) was added drop wise
D
to trigger the reaction. The resulting mixture was stirred at ꢁ20 °C
overnight. The reaction was stopped by adding a drop of triethyl-
amine. After celite filtration, the reaction solution was washed with
saturated aqueous NaHCO3 three times and dried over Na2SO4 over-
night. The mixture was filtered through a cotton plug. After removal
of the solvent by vacuum distillation, purification of the crude prod-
tially primed the elongation of chondroitin sulfate (CS) and only
weakly initiated the synthesis of heparan sulfate chains (HS).11,27
It has been reported that the aglycone structure of these xylosides
could affect the proportion of CS or HS made on these b-
sides. Xylosides with hydrophobic aglycone structures, such as
estradiol b- -xyloside and naphthol b- -xyloside, could initiate
D-xylo-
D
D
uct by silica column chromatography (2.5 cm
ethyl acetate = 2/5) afforded 2. The compound was considered to
U
ꢂ 15 cm, hexane/
the synthesis of HS effectively in various types of cells.14,17,24 How-
ever, a clear correlation between the aglycone structure and the
composition of products (proportion of CS/HS) has not been found.
In the present study, products such as SXA6 and CXA8 were hydro-
lyzed by heparitinase, suggesting that Xyl-Ser-C12 could stimulate
the elongation of HS GAGs. On the other hand, products such as
SXA2 and CXA2, in which galactose residues were sulfated, were
detected. It has been reported that sulfation on the galactose of
the linker region occurred in the biosynthesis of CS, but not
HS.10,31,32 The endogenous expression profiles of CS and HS have
been extensively reported.33 It also could be inferred that the bio-
synthesis of both CS and HS co-exist in mouse osteosarcoma cells
based on Basappa et al.’s study34 and the detection of the expres-
sion of GAG/proteoglycans and GAG-related glycosyltransferases
in mouse osteosarcoma cells (unpublished data by Wang et al.).
Thus, it is reasonable to assume that the products obtained from
Xyl-Ser-C12 were synthesized by the two pathways for HS and CS.
Because glycans play important roles in living systems, there is
an urgent need to develop high-throughput and large-scale glyco-
mics for analysis of the structure and functions of sugar chains. The
glycan microarray using synthetic or natural oligosaccharide li-
braries immobilized on solid surfaces is a useful tool for glycomics
research. The saccharide primer method is an efficient comple-
mentary method for oligosaccharide synthesis. Introduction of an
azido group into the alkyl chain of saccharide primers could be
used in immobilization on sensor chips, and facilitate the achieve-
ment of an efficient high-throughput glycan array.23,35 Xyl-Ser-
C12-initiated oligosaccharides are also expected to be applied in
glycan arrays after chemical modification.
be an orthoester formed during glycosylation, and was subjected
to TMSOTf-catalyzed isomerization. TMSOTf (22 lL, 0.120 lmol,
0.4 equiv) was added to a solution of 0.740 g of 2 (0.081 mmol,
1.0 equiv) in 2 mL of anhydrous dichloromethane in an ice bath
with argon protection. The next day, the reaction mixture was
washed with saturated aqueous NaHCO3, dried over MgSO4, and
subjected to reduced pressure to remove the solvent and to purifi-
cation by silica column chromatography (2.5 cm
U
ꢂ 15.0 cm, hex-
ane/ethyl acetate = 1/3) to yield compound 3 (0.013 g, 31.5%). 1H
NMR(CDCl3): d 7.36 (m, 5H, CbzꢀC6H5), 6.30 (s, 1H, NH2), 5.68–
5.70 (d, JCH,NH 6.2 Hz, 1H, SerꢀNH), 5.47 (s, 1H, NH2), 5.16–5.19
(dd, J2,3 8.3 Hz, J3,4 8.9 Hz, 1H, H-3), 5.12 (m, 2H, CbzꢀCH2), 4.94–
4.96 (ddd, J3,4 8.9 Hz, J4,5a 6.2 Hz, J4,5e 10.3 Hz, 1H, H-4), 4.90–4.93
(dd, J1,2 6.2 Hz, J2,3 8.2 Hz, 1H, H-2), 4.53–4.54 (d, J1,2 6.2 Hz, 1H,
H-1), 4.41 (m, J6a,7 3.4 Hz, 1H, H-7), 4.17–4.20 (dd, J6a,7 3.4 Hz, Jgem
10.3 Hz, 1H, H-6a), 4.11–4.14 (J4,5a 6.2 Hz, Jgem 10.3 Hz, 1H, H-5a),
3.66–3.69 (dd, Jgem 10.3 Hz, 1H, H-6b), 3.36–3.40 (dd, J4,5e 10.3 Hz,
Jgem 10.3 Hz, 1H, H-5e), 2.03–2.05 (s, 9H, -OAc).
a
4.1.2. N -Lauryl-O-(2,3,4-tri-O-acetyl-b-
D-xylopyranosyl)-
L-
serinamide (4)
To a solution of 0.050 g of compound 3 (0.100 mmol) dissolved in
10 mL of THF and 5.0 mL of EtOH, was added slowly 0.200 g of Pd/C
(Aldrich) with stirring. This mixed solution was subjected to de-pro-
tection by hydrogen bubbling for 20 min (H2 was generated by a
hydrogen generator, GL Sciences). When TLC indicated the comple-
tion of the reaction, the reaction was terminated by stopping
hydrogen generation. Then 28.0 lL of 4 M hydrogen chloride in
In summary, Xyl-Ser-C12 was synthesized to obtain GAG-type
oligosaccharides using mammalian cells and exhibited good ability
to prime the elongation of oligosaccharides. The glycosylated prod-
ucts were detected by LC–MS, and the sequences of the products
were determined by enzymatic digestion and MS/MS. This study
should contribute to the comparison of glycosaminoglycan expres-
sion between different cell lines and the construction of glycosami-
noglycan saccharide libraries.
1,4-dioxane (Kokusan Chemical) was added to the mixture. Then
after the removal of Pd/C through celite filtration, the filtrate was
evaporated and condensed to obtain a white crystal. Without further
purification, the product was subjected to the next reaction: 0.072 g
of the compound obtained above and 0.050 mg of lauric acid
(0.250 mmol) were added in a 2-neck flask and dissolved in 5.0 mL
of anhydrous DMF under argon protection. To this solution, 44
lL
of triethylamine (0.320 mol) was added. Then 0.038 g of EDC (1-
l
ethyl-3-(3-dimethyl-aminopropyl) carbodiimide, 0.198 mmol, Doj-
indo), and 0.055 g of HOBt (1-hydroxybenzotriazole, 0.407 mmol,
Nova Biochem) were added to start the reaction in an ice bath. After
confirmation of the disappearance of the reactant by TLC, the reac-
tion mixture was evaporated to remove DMF. The residue was dis-
solved in chloroform, washed with water, and dried over Na2SO4,
followed by removal of the solvent by vacuum distillation. The crude
product was purified by silica column chromatography (2.5 cm
4. Experimental
a
4.1. Synthesis of N -lauryl-O-(b-
(Xyl-Ser-C12)
D
-xylopyranosyl)-
L
-serinamide
a
4.1.1. N -Benzyloxycarbonyl-O-(2,3,4-tri-O-acetyl-b-
D
-xylopyr
U
ꢂ 15.0 cm, hexane/ethyl acetate = 1/20) to obtain 4. Yield: 41.1%
anosyl)-L-serinamide (3)
(0.035 g). 1H NMR (CDCl3): d 6.32–6.34 (d, JCH,NH 6.8 Hz, 2H, NH2,
NH), 5.35 (s, 1H, NH2), 5.01–5.17 (dd, J2,3 8.9 Hz, J3,4 9.2 Hz, 1H,
H-3), 4.87–4.93 (ddd, J3,4 9.2 Hz, J4,5 9.0 Hz, J4,5b 5.5 Hz, 1H, H-4),
4.84–4.85 (dd, J1,2 7.0 Hz, J2,3 8.9 Hz, 1H, H-2), 4.53–4.61 (m, 1H,
H-7), 4.51–4.53 (d, J1,2 7.2 Hz, 1H, H-1), 4.08–4.12 (dd, J4,5b 5.9 Hz,
Jgem 11.6 Hz, 1H, H-5e), 4.03–4.08 (dd, Jgem 11.0 Hz, 1H, H-6a),
3.53–3.59 (dd, Jgem 10.3 Hz, 1H, H-6b), 3.31–3.38 (dd, J4,5a 9.4 Hz, Jgem
11.7 Hz, 1H, H-5a), 2.14–2.19 (t, 2H, COCH2), 1.93–2.00 (m, 9H, acetyl
Xylose was acetylated with acetic anhydride and sodium
acetate.36 Selective 1-O-deacetylation of acetylated xyloside37,38
was carried out, and the glycosyl donor, 2,3,4-tri-O-acetyl-D-xylo-
pyranosyl-tricholoroacetimidate (1), was synthesized by Schmidt
activation.39,40 The glycosyl donor 1 was identified from NMR
spectroscopy.39
0.100 g of 139,40 (0.238 mmol, 1.0 equiv) and 0.068 g of Z-Ser-
NH2 (N-a-Carbobenzoxy-L-serinamide, 0.285 mmol, Watanabe

Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
39
group), 1.47–1.62 (m, 2H, COCH2CH2), 1.15–1.30 (m, 16H,
(CH2)8CH3), 0.78–0.88 (t, 3H, COCH2CH2 (CH2)8CH3).
both the positive and the negative ion modes. The glycosylated
products were loaded onto a 2 mm ꢂ 150 mm UK-silica column
(imtakt, Japan) equilibrated with solvent A [chloroform/metha-
nol/water (C/M/W) containing 50 mM TEA acetate buffer, pH 4.2
(83/16/1, v/v/v)] and eluted with a 0–100% linear gradient of sol-
vent B [M/W containing 50 mM TEA acetate buffer, pH 4.2, (3/1,
a
4.1.3. N -lauryl-O-(b-
D
-xylopyranosyl)-
L
-serinamide (5)
0.530 g of compound 4 (0.973 mmol) was mixed with 10.0 mL
of methanol. To this mixture was added 10 drops of sodium meth-
oxide methanol solution. After 45 min when TLC indicated disap-
pearance of the reactants, activated amberlite was added to
neutralize the reaction system. Then the reaction mixture was fil-
tered through a cotton plug to remove amberlite, and vacuum dis-
tilled to remove the solvent. The product was purified by
v/v)] in 45 min at a flow rate of 100 lL/min. The ESI parameters
were as follows: nebulizer, 10.0 psi; dry gas (N2), 4.0 L/min; dry
temperature, 250 °C; negative ion mode. The mass recorded ranged
from 150 to 3000 m/z.
chromatography on a silica column (3.0 cm
U
ꢂ 25.0 cm, chloro-
4.5. Cell viability assay
form/methanol = 6/1), yielding 5 (0.249 g, 61.1%). The characteris-
tics of 5 are as follows: Melting point: 151.9 °C (Barnstead
International, Melting-point apparatus 1201D); Optical rotation:
FBJ-S1 cells (1.5 ꢂ 104) or CHO cells (3 ꢂ 103) in a 96-well
microplate were incubated with 50 lM Xyl-Ser-C12 or 50 lM
½
a 2D4:7
ꢃ
41.2 (JASCO, Polarimeter P-1020) (cell length: 100 mm,
Xyl-pNP for 48 h. Cells treated with an equal volume of DMSO
served as a negative control. Ten microliters of WST-1 dye solution
(10 mM WST-1 and 0.2 mM 1-methoxy PMS, Dojindo Laboratories)
per well was added to each well. After 3 h, absorbance at 450 nm
with a reference wavelength of 690 nm was measured using a
microplate reader (Multiskan, Labsystem).
24.7 °C, c = 1.0 g/mL, k = 589 nm). 1H NMR (CD3OD): d 4.46–4.49
(dd, JCH,CH2 4.8 Hz, 1H, H-7), 4.11–4.14 (d, J1,2 7.3 Hz, 1H, H-1),
4.02–4.07 (dd, JCH,CH2 4.8 Hz, Jgem 5.3 Hz, 1H, H-6b), 3.73–3.79
(dd, J5a,5b 11.4 Hz, 1H, H-5b), 3.59–3.63 (dd, JCH,CH2 4.8 Hz, Jgem
5.3 Hz, 1H, H-6a), 3.33–3.44 (m, 2H, H-3, H-4), 3.06–3.13 (m, J1,2
7.3 Hz, J5a,5b 11.4 Hz, 2H, H-2, H-5a), 2.16–2.21 (dd, 2H, COCH2),
1.52 (m, 2H, COCH2CH2), 1.20 (m, 16H, (CH2)8CH3), 0.78–0.82 (t,
3H, CH3). 13C NMR (CD3OD): d 105.35 (C-1), 77.75 (C-3), 77.74
(C-2), 71.16 (C-4), 70.61 (Ser CH2), 67.06 (C-5), 54.37 (Ser CH),
36.93 (COCH2), 33.12, 30.79, 30.69, 30.55, 30.52, 30.41, 23.77
((CH2)8CH3), 26.81 (COCH2CH2), 14.48 (CH3). HRESIMS Cald. for
4.6. Glycosidase digestion of glycosylated products
Glycosidase digestion of glycosylated products was carried out
in 100 mM phosphate buffer (pH 7.2) containing b-galactosidase
(100 U, Wako, E.C. 3.2.1.23); 50 mM sodium acetate buffer (pH
7.0) containing 5 mM calcium acetate and heparitinase (1 mU,
Seikagaku, E.C. 4.2.2.8); 50 mM sodium citrate buffer (pH 5.0) con-
taining b-N-acetylhexosaminidase (100 mU, Seikagaku, 100094,
E.C. 3.2.1.52); 50 mM sodium citrate buffer (pH 4.3) containing
C
20H38N2O7: [M+Na]+, 441.2571. Found: [M+Na]+, 441.2565.
4.2. Cell lines and culture
The mouse osteosarcoma cell line FBJ-S1 was produced from a FBJ
virus-induced osteosarcoma of the BALB/c mouse.20,41 The cells were
maintained in RPMI-1640 medium (Nacalai, Japan) supplemented
with 10% FBS (GIBCO, USA) as described previously.42 CHO (Chinese
hamster ovary) cells were cultured in Ham’s F-12 medium (Nacalai,
Japan) supplemented with 10% heat-inactivated fetal bovine serum,
0.1 g/L streptomycin, and 50,000 units/L penicillin G potassium at
37 °C in a humidified atmosphere containing 5% CO2.
a-N-acetylgalactosaminidase (100 mU, Seikagaku, 100086, E.C.
3.2.1.30); 20 mM phosphate buffer (pH 6.5) containing b-Glucuron-
idase (100 mU, Wako, 075-04631, E.C. 3.2.1.31); and 50 mM so-
dium phosphate buffer (pH 6.0) containing
a2-3 neuraminidase
(15 mU, Sigma, N7271, E.C. 3.2.1.1.8). The samples were recovered
with a Sep-Pak C18 column and subjected to LC–MS/MS. Peak areas
of extracted ion chromatograms of the indicated products treated
with enzyme were measured, and compared with those without
the enzyme treatment. The validity of enzymatic cleavage was con-
firmed with a positive control.
4.3. Administration of saccharide primers
FBJ-S1 cells (2 ꢂ 106) or CHO cells (2 ꢂ 106) seeded into 100-
mm culture dishes were incubated with serum-free and phenol
Acknowledgements
red-free RPMI-1640 medium (Gibco) containing 50 lM Xyl-Ser-
C12 or Xyl-pNP (Seikagaku), 5 mg/L of transferrin, 5 mg/L of insu-
lin, and 30 nM selene dioxide for 48 h. The glycosylated products
were collected from the culture medium using a Sep-Pak C18 car-
tridge (Waters, Ireland). The products were eluted with methanol.
For the analysis by LC–MS, the glycosylated products were dis-
solved in 1 ml of chloroform/methanol (C/M) (19/1, v/v) and ad-
sorbed to a discovery DSC–NH2 cartridge (SUPELCO, USA). After
washing of the cartridge with chloroform, the acidic products were
eluted with 3% acetic acid/4% triethylamine in methanol. The elu-
We thank Professor Tatsuya Yamagata for kindly providing FBJ-
S1 cells and valuable discussions. This work was partly supported
by the Global COE Project ‘‘Global COE for Human Metabolomic
Systems Biology’’ (YW) from the Ministry of Education, Culture,
Sports, Science and Technology and a Grant-in-aid for Challenging
and Exploratory Research (2265108, TS) from the Japan Society for
the Promotion of Science.
Supplementary data
ates were passed through
a minisart RC4 filter (pore size;
0.2 m) (Satorius Stedim Biotech, Germany), evaporated, and dis-
solved in C/M (9:1).
l
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carres.2012.08.003.
4.4. Mass spectrometry
References
The glycosylated products were subjected to LC–MS as de-
scribed previously.43 Briefly, mass spectrometric analyses were
performed using electrospray ionization (ESI)/an ion trap type
mass spectrometer (Esquire 3000 plus, Bruker Daltonics, USA),
coupled online by the Agilent 1100 series LC system (Agilent Tech-
nologies, USA) with ion extraction, separation, and detection in
1. Varki, A. Glycobiology 1993, 3, 97–130.
2. Seeberger, P. H.; Werz, D. B. Nature 2007, 446, 1046–1051.
3. Nakajima, H.; Miura, Y.; Yamagata, T. J. Biochem. 1998, 124, 148–156.
4. Miura, Y.; Yamagata, T. Biochem. Biophys. Res. Commun. 1997, 241, 698–703.
5. Miura, Y.; Arai, T.; Yamagata, T. Carbohydr. Res. 1996, 289, 193–199.
6. Sato, T.; Takashiba, M.; Hayashi, R.; Zhu, X.; Yamagata, T. Carbohydr. Res. 2008,
343, 831–838.

40
Y. Wang et al. / Carbohydrate Research 361 (2012) 33–40
7. Kasuya, M. C. Z.; Wang, L. X.; Lee, Y. C.; Mitsuki, M.; Nakajima, H.; Miura, Y.;
25. Miura, Y.; Freeze, H. H. Glycobiology 1998, 8, 813–819.
26. Takagaki, K.; Nakamura, T.; Shibata, S.; Higuchi, T.; Endo, M. J. Biochem. 1996,
119, 697–702.
Sato, T.; Hatanaka, K.; Yamagata, S.; Yamagata, T. Carbohydr. Res. 2000, 329,
755–763.
8. Handel, T. M.; Johnson, Z.; Crown, S. E.; Lau, E. K.; Proudfoot, A. E. Annu. Rev.
Biochem. 2005, 74, 385–410.
27. Takagaki, K.; Nakamura, T.; Kon, A.; Tamura, S.; Endo, M. J. Biochem. 1991, 109,
514–519.
9. Bernfield, M.; Gotte, M.; Park, P. W.; Reizes, O.; Fitzgerald, M. L.; Lincecum, J.;
Zako, M. Annu. Rev. Biochem. 1999, 68, 729–777.
10. Prydz, K.; Dalen, K. T. J. Cell Sci. 2000, 113(Pt 2), 193–205.
11. Schwartz, N. B.; Galligani, L.; Ho, P. L.; Dorfman, A. Proc. Natl. Acad. Sci. U. S. A.
1974, 71, 4047–4051.
28. Etchison, J. R.; Srikrishna, G.; Freeze, H. H. J. Biol. Chem. 1995, 270, 756–764.
29. Kitagawa, H.; Kano, Y.; Shimakawa, H.; Goto, F.; Ogawa, T.; Okabe, H.;
Sugahara, K. Glycobiology 1999, 9, 697–703.
30. Kitagawa, H.; Shimakawa, H.; Sugahara, K. J. Biol. Chem. 1999, 274, 13933–
13937.
12. Okayama, M.; Kimata, K.; Suzuki, S. J. Biochem. 1973, 74, 1069–1073.
13. Izumi, J.; Takagaki, K.; Nakamura, T.; Shibata, S.; Kojima, K.; Kato, I.; Endo, M. J.
Biochem. 1994, 116, 524–529.
14. Lugemwa, F. N.; Esko, J. D. J. Biol. Chem. 1991, 266, 6674–6677.
15. Siegbahn, A.; Aili, U.; Ochocinska, A.; Olofsson, M.; Ronnols, J.; Mani, K.;
Widmalm, G.; Ellervik, U. Bioorg. Med. Chem. 2011, 19, 4114–4126.
16. Mani, K.; Belting, M.; Ellervik, U.; Falk, N.; Svensson, G.; Sandgren, S.; Cheng, F.;
Fransson, L. A. Glycobiology 2004, 14, 387–397.
17. Mani, K.; Havsmark, B.; Persson, S.; Kaneda, Y.; Yamamoto, H.; Sakurai, K.;
Ashikari, S.; Habuchi, H.; Suzuki, S.; Kimata, K.; Malmstrom, A.; Westergren-
Thorsson, G.; Fransson, L. A. Cancer Res. 1998, 58, 1099–1104.
18. Garud, D. R.; Tran, V. M.; Victor, X. V.; Koketsu, M.; Kuberan, B. J. Biol. Chem.
2008, 283, 28881–28887.
19. Hyuga, S.; Yamagata, S.; Tai, T.; Yamagata, T. Biochem. Biophys. Res. Commun.
1997, 231, 340–343.
20. Yamagata, S.; Miwa, M.; Tanaka, K.; Yamagata, T. Biochem. Biophys. Res.
Commun. 1982, 105, 1208–1214.
21. Salimath, P. V.; Spiro, R. C.; Freeze, H. H. J. Biol. Chem. 1995, 270, 9164–9168.
22. Freeze, H. H.; Sampath, D.; Varki, A. J. Biol. Chem. 1993, 268, 1618–1627.
23. Sato, T.; Hatanaka, K.; Hashimoto, H.; Yamagata, T. Trends Glycosci. Glyc. 2007,
19, 1–17.
24. Fritz, T. A.; Lugemwa, F. N.; Sarkar, A. K.; Esko, J. D. J. Biol. Chem. 1994, 269,
300–307.
31. Tone, Y.; Pedersen, L. C.; Yamamoto, T.; Izumikawa, T.; Kitagawa, H.; Nishihara,
J.; Tamura, J.; Negishi, M.; Sugahara, K. J. Biol. Chem. 2008, 283, 16801–16807.
32. Sugahara, K.; Kitagawa, H. Curr. Opin. Struct. Biol. 2000, 10, 518–527.
33. Stow, J. L.; Farquhar, M. G. J. Cell Biol. 1987, 105, 529–539.
34. Basappa; Murugan, S.; Sugahara, K. N.; Lee, C. M.; ten Dam, G. B.; van
Kuppevelt, T. H.; Miyasaka, M.; Yamada, S.; Sugahara, K. Glycobiology 2009, 19,
735–742.
35. Sato, T.; Fujita, S.; Kasuya, M. C. Z.; Hatanaka, K.; Yamagata, T. Chem. Lett. 2004,
33, 580–581.
36. Gruzman, A.; Shamni, O.; Ben Yakir, M.; Sandovski, D.; Elgart, A.; Alpert, E.;
Cohen, G.; Hoffman, A.; Katzhendler, Y.; Cerasi, E.; Sasson, S. J. Med. Chem. 2008,
51, 8096–8108.
37. Utille, J.-P.; Gagnaire, D. Carbohydr. Res. 1982, 106, 43–57.
38. Rowell, R. M.; Feather, M. S. Carbohydr. Res. 1967, 4, 486–491.
39. Johnsson, R.; Ellervik, U. Synlett 2005, 2939–2940.
40. Mori, M.; Ito, Y.; Ogawa, T. Carbohydr. Res. 1990, 195, 199–224.
41. Yamagata, S.; Tanaka, R.; Ito, Y.; Shimizu, S. Biochem. Biophys. Res. Commun.
1989, 158, 228–234.
42. Wang, L.; Takaku, S.; Wang, P.; Hu, D.; Hyuga, S.; Sato, T.; Yamagata, S.;
Yamagata, T. Glycoconj. J. 2006, 23, 303–315.
43. Ogasawara, N.; Katagiri, Y. U.; Kiyokawa, N.; Kaneko, T.; Sato, B.; Nakajima, H.;
Miyagawa, Y.; Kushi, Y.; Ishida, H.; Kiso, M.; Okita, H.; Sato, T.; Fujimoto, J. J.
Biochem. 2011, 149, 321–330.