Glycobiology vol. 15 no. 11 pp. 1094–1101, 2005
doi:10.1093/glycob/cwj003
Advance Access publication on July 6, 2005
Modification of oligosaccharides by reactive oxygen species decreases
sialyl lewis x-mediated cell adhesion
Hironobu Eguchi2, Yoshitaka Ikeda3,4, Tomomi Ookawara2,
Souichi Koyota5, Noriko Fujiwara2, Koichi Honke6, Peng G.
Wang7, Naoyuki Taniguchi5, and Keiichiro Suzuki1,2
2
Department of Biochemistry, Hyogo College of Medicine, 1-1
Mukogawa, Nishinomiya, Hyogo 663-8501, Japan; 3Division of Molecular Cell Biology, Department of Biomolecular Sciences, Saga University
Faculty of Medicine, 5-1-1 Nabeshima, Saga, Saga 849-8501, Japan;
4
CREST Project, Japan Science and Technology Agency; 5Department of
Biochemistry, Osaka University Medical School, 2-2 Yamadaoka, Suita,
Osaka 565-0871, Japan; 6Department of Molecular Genetics, Kochi
University Medical School, Kohasu, Oko-cho, Nankoku, Kochi 783-8505,
Japan; 7Departments of Biochemistry and Chemistry, The Ohio State
University, 876 Biological Sciences Building, 484 West 12th Avenue,
Columbus, OH 43210
Received on March 3, 2005; revised on June 22, 2005; accepted on
June 23, 2005
Modification of cell surface oligosaccharides by reactive
oxygen species (ROS) and the biological effect of such modifications on cell adhesion were investigated. Treatment of
HL60, a human promyelocyte leukemia cell line, with ROS,
generated by a combination of hypoxanthine and xanthine
oxidase (HX/XO), decreased the sialic acid content on the
cell surface, as indicated by a flow cytometric analysis involving sialic acid-specific lectins, and a concomitant increase of
free sialic acid was observed in the supernatant. A cell adhesion assay showed that the HX/XO treatment of HL60 cells
decreases their capability of binding to human umbilical vein
endothelial cells (HUVEC), probably because of an impairment of the interaction involving E-selectin, whereas the
decrease in the binding was canceled by the addition of superoxide dismutase (SOD) and catalase. In fact, cell surface
sialyl lewis x (sLex), but not lewis x (Lex), was decreased by
HX/XO treatment. Thus, it is more likely that the impaired
interaction is based on diminished levels of the selectin ligand.
Cleavage of sialic acid by ROS was further verified by the
degradation of 4MU-Neu5Ac by HX/XO in the presence of
hydrogen peroxide and iron ion. These results indicate that
glycosidic linkage of sialic acid is a potential target for superoxide and other related ROS. It is well known that ROS cause
cellular damages such as lipid peroxidation and protein oxidation, but, as suggested by the findings reported in the literature, ROS may also regulate cell adhesion via the structural
alteration of sialylated oligosaccharides on the cell surface.
Key words: cell adhesion/reactive oxygen species/selectin/
sialic acid/sialyl lewis x
1To
whom correspondence should be addressed; e-mail: suzuki@
hyo-med.ac.jp
Introduction
Reactive oxygen species (ROS) are produced during several
pathological conditions such as in inflammation and reperfusion injury and damage various biological molecules
because of its high and nonspecific reactivity (Halliwell and
Gutteridge, 1999). As reported in earlier studies, ROS cause
oxidative damage to proteins such as Cu,Zn-superoxide
dismutase and collagen, and can also be a causal agent of
lipid peroxidation in cell membranes (Farmer et al., 1943;
Greenwald and Moy, 1979; Curran et al., 1984; Ookawara
et al., 1992; Halliwell and Gutteridge, 1999). It has been
suggested that ROS are capable of degrading glycosaminoglycans, which are important components of extracellular matrices, thus leading to an impairment of proteoglycan
functions (Moseley et al., 1995, 1997). It was also reported
that ROS are capable of modifying glycosaminoglycan
units of hyaluronic acid in vitro (Mccord, 1974; Halliwell,
1978; Greenwald and Moy, 1980; Grootveld et al., 1991;
Hawkins and Davies, 1996, 1998; Jahn et al., 1999). We
recently reported that oxidative damage by copper ion and
hydrogen peroxide (H2O2) causes the site-specific degradation of N-linked oligosaccharide at N-acetylglucosamine
residues (Eguchi et al., 2002). As indicated by these reports,
there is accumulating evidence that oligosaccharides, as
well as proteins, lipids, and DNA, may be a target molecule
of ROS.
It is known that oligosaccharides are involved in various
biological events, and their role in cell adhesion have been
intensively investigated. The endothelial glycocalyx covers the
luminal surface of the endothelium and regulates cell adhesion via its highly charged properties (Springer, 1990; Soler
et al., 1998; Sabri et al., 2000). Hyaluronic acid, a high molecular weight glycosaminoglycan consisting of a repeating
disaccharide of -glucuronic acid-β1,3-N-acetylglucosamineβ1,4-, binds to CD44, a widely distributed cell adhesion glycoprotein (Aruffo et al., 1990). The binding of CD44 to
hyaluronic acid is involved in the adherence to the extracellular matrix and cellular signaling in cells (Lesley et al., 1993;
Naot et al., 1997; Ponta et al., 2003). The sialyl lewis x (sLex)
epitope [NeuAcα2,3Galβ1,4(Fucα1,3)GlcNAc-] has been
shown to be a specific oligosaccharide ligand for E-selectin
(Phillips et al., 1990; Walz et al., 1990), a member of the
selectin family that is expressed on activated endothelial
cells. Interaction of the sLex epitope and E-selectin promotes the initial attachment and subsequent movement
of leukocytes over vessel walls, firming their adhesion
(Kansas, 1996; Vestweber and Blanks, 1999). The crystal
structure of selectin and the sLex complex revealed that
E-selectin directly recognizes the sLe x epitope and that
the sialic acid residue of the epitope is of particular
© The Author 2005. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected]
1094
Effect of reactive oxygen species on cell adhesion
importance in this binding (Graves et al., 1994; Somers
et al., 2000).
ROS facilitate the adhesion of leukocytes to endothelial
cells via inducing the expression of adhesion molecules such
as selectins and ICAM-1 (Patel et al., 1991; Palluy et al.,
1992; Lo et al., 1993). For example, P-selectin, which is
known to be present on activated platelets, is translocated
from intracellular stores to the cell surface within minutes
by ROS stimulation, and leukocytes adhesion then occurs
very soon (Patel et al., 1991). A previous study found that
treatment of human umbilical vein endothelial cells
(HUVEC) with ROS for a few minutes facilitates the cellular adhesion of colon cancer cells and neutrophils without
the de novo protein synthesis of adhesion molecules, such as
P-selectin, E-selectin, and ICAM-1 (Suzuki et al., 1999).
The findings presented in this report indicate that the
accelerated cell adhesion is the result of the modification of
glycocalyx on the surface of endothelial cells by ROS. In
fact, it has been demonstrated histochemically that
ischemia reperfusion causes the disruption of the endothelial glycocalyx (Ward and Donnelly, 1993; Czarnowska and
Karwatowska, 1995; Beresewicz et al., 1998). Therefore, the
possibility that ROS regulate cell adhesion as the result of
the direct modification of oligosaccharides on the cell surface cannot be excluded. However, details of the modification of oligosaccharide structure by ROS are not currently
well understood.
In this study, we reported on an investigation of the
ROS-mediated modification of oligosaccharides and the
biological consequence of such a modification. The data
obtained here reveal that oxidative stress due to hypoxanthine and xanthine oxidase (HX/XO) led to cleavage of the
sialic acid from oligosaccharides on the cell surface without
the activation and/or induction of sialidase, thereby leading
to an impairment in sLex-dependent cell adhesion. These
findings suggest that ROS affect cell adhesion via the direct
modification of cell surface sialylated oligosaccharides.
respectively. As shown in Figure 1, the binding of SSA and
MAM was decreased because of the loss of sialic acid residues by sialidase treatment, and a corresponding increase in
RCA lectin binding was observed, indicating that the SSA
and MAM lectins recognized the sialic acid residue and the
RCA lectin recognized the galactose at the terminal position of oligosaccharides. The binding of ConA to the HL60
cells was not affected by the HX/XO treatment whereas the
binding of MAM and SSA lectins were significantly less,
compared to non-treated controls. Because the addition of
superoxide dismutase (SOD) and catalase canceled the
impairment of their binding (data not shown), the decrease
in the reactivity of MAM and SSA lectins appeared to be
based on damage directly caused by superoxide and its
related ROS. On the other hand, the binding of the RCA
lectin was increased in contrast to the sialic acid-specific
lectin binding. This inverse correlation suggests that nonreducing terminal sialic acids are removed by ROS, which
then allows β-Gal residues to be exposed. The binding of
Lotus lectin, which binds to L-fucose, was not altered by
ROS, indicating that fucose residues were not affected by
HX/XO treatment (data not shown). As indicated by these
results, the terminal sialic acid residues appear to be most
susceptible to treatment with the HX/XO system, and thus
it is suggested that the sialic acid residues are vulnerable
target for damage by ROS.
To determine whether the loss of sialic acid residues from
the non-reducing terminus is due to the release of sialic acid
from the cell surface, we analysed the content of sialic acid
in the supernatant of HL60 cells by high-performance liquid chromatography (HPLC) using a fluorescent labeling
procedure. After the treatment of HL60 with HX/XO for
Results
The issue of how cell surface oligosaccharides are modified
by ROS was investigated. It is known that the HX/XO
system generates superoxide anion and H2O2 and is also
capable of producing hydroxyl radical via the fenton reaction, providing trace amounts of transition metal ions are
present. As judged by a cytochrome c reduction assay
(Fridovich, 1970), superoxide production by HX/XO was
comparable to the levels produced by isolated neutrophils
stimulated with either PMA or formyl methionyl leucyl
phenylalanine (fMLP): HX/XO, 5.8 nM/min; neutrophils
(107/mL) stimulated by PMA, 9.1 nM/min; and neutrophils
(107/mL) stimulated by fMLP, 3.5 nM/min. HL60 cells
were exposed to ROS using HX/XO system as a radical
source, and the resulting alteration in oligosaccharide structure was examined by flow cytometric analysis based on the
binding specificity of several lectins. The lectins used,
Maackia amurensis (MAM), Sambucus sieboldiana (SSA),
Ricinus communis (RCA), and Canavalia ensiformis
(ConA), preferentially bind to α2-3 linked sialic acid, α2-6
linked sialic acid, Galβ1–4GlcNAc, and α-D-mannose,
Fig. 1. Lectin analysis of reactive oxygen species (ROS) treated
HL60.HL60 were incubated with hypoxanthine and xanthine oxidase
(HX/XO) at 37°C for 5 min. In the sialidase treatment, HL60 were incubated with 0.1 U/mL sialidase at 37°C for 30 min. The oligosaccharide
structures of HL60 were analyzed by flow cytometry using several lectins;
(A) Sambucus sieboldiana (SSA), (B) Maackia amurensis (MAM), (C)
Ricinus communis (RCA), and (D) Canavalia ensiformis (ConA). Gray
histograms, white histograms, and dotted line histograms in each panel
show the fluorescence intensity of HX/XO treated, phosphate-buffered
saline (PBS)/XO treated and sialidase treated HL60 cells, respectively.
1095
H. Eguchi et al.
5 min at 37°C, sialic acid in the supernatant was labeled
with 1,2-diamino-4,5-methylenedioxybenzene (DMB), a
fluorogenic reagent for alpha-keto acids, and the resulting
fluorescent derivative was then assayed by reversed phase
HPLC, with a fluorescence detection. It was confirmed that
glycosidic linkage of sialic acid is not hydrolyzed under the
conditions used for DMB labeling (data not shown). As
shown in Figure 2, the analyses showed that the sialic acid
content in the supernatant of ROS-treated HL60 cells was
three times higher than that in a non-treated sample. This
increase of sialic acid was partially inhibited by the addition
of SOD, and completely abolished by the addition of SOD
and catalase. These results indicated that ROS liberate
sialic acid from the cell surface. Assuming that the amount
of sialic acid that could be cleaved by sialidase is 100%,
almost half of the total sialic acid on the cell surface was
cleaved by ROS. In addition, even if sugar chains in the
supernatant were hydrolyzed by hydrochloric acid, the
amount of sialic acid available for conjugation to DMB
was not increased. Because DMB reacts with α-keto acids,
this labeling reagent produces fluorescent derivatives only
with free sialic acid but not its glycosidically bound form.
The increase in sialic acid levels in the supernatant by ROS
appeared to result from the specific release of sialic acid but
not to any nonspecific damage to sugar chains. In addition,
Fig. 2. Sialic acid analysis of the supernatant of reactive oxygen species
(ROS) treated HL60. HL60 were incubated with hypoxanthine and xanthine oxidase (HX/XO) at 37°C for 5 min in the presence or absence of
superoxide dismutase (SOD) and catalase. In the sialidase treatment,
HL60 cells were incubated with 0.1 U/mL sialidase at 37°C for 30 min.
The sialic acid in supernatant was labeled with 1,2-diamino-4,5-methylenedioxybenzene (DMB), and measured by reversed phase highperformance liquid chromatography (HPLC). For hydrolysis assay, the
supernatant was treated with hydrochloric acid described in experimental
procedures. (mean ± SD, n = 4).
1096
because sialidase activity was not detected in HX/XO treatment (data not shown), it was unlikely that the release of
sialic acid was due to the enzymatic hydrolysis. Therefore,
these results further support the view that the glycosidic
linkage of sialic acid is labile to attack by ROS, and it
appears that the linkages of other sugars such as galactose
and mannose are stable and not affected by ROS.
To examine the biological significance of ROS-mediated
desialylation, we investigated the effect of ROS on cell
adhesion, a process that involves sialic acid residues.
E-selectin expression in HUVEC by IL-1β was confirmed
by flow cytometric analysis using a specific antibody for
E-selectin (data not shown), and it was observed that HL60
cells bind to the IL-1β-stimulated HUVEC, as shown in
Figure 3A(b). This binding was completely blocked by the
addition of either of anti-sLex or anti-E-selectin antibodies
[Figure 3A(f) and (g)], thus confirming that cellular binding
entirely depends on protein–carbohydrate interactions.
Treatment of HL60 cells with HX/XO impaired their binding to the IL-1β-stimulated HUVEC [Figure 3A(d)], as
found for the sialidase treatment of HL60 cells [Figure
3A(h)]. However, the presence of SOD and catalase rescued
the inhibitory effect of ROS [Figure 3A(e)]. As shown by
these results, ROS actually interfered with cell adhesion most
probably via damage to the oligosaccharide structures.
The issue of whether the ROS diminished sLex epitope
levels on the cell surface of HL60 cells was investigated.
After treatment of HL60 cells with HX/XO, sLex and its
desialylated form, lewis x (Lex), were analyzed by flow
cytometry using specific antibodies for sLex and Lex. Sialidase treatment of HL60 cells resulted in the decrease and
increase of the binding of both anti-sLex and Lex antibodies
to cell surface, respectively, indicating that the binding of
these antibodies were affected by the presence of sialic acid
at the terminus of oligosaccharides (Figure 4E and J).
When HL60 cells were treated by HX/XO, the decrease in
sLex epitope levels was observed along with the corresponding increase in Lex (Figure 4A, B, F, and G). Consistent with the results of the cell adhesion assay, these
structural alterations caused by ROS were almost completely
abolished in the presence of SOD and catalase (Figure 4C,
D, H, and I). Thus, ROS inhibit cell adhesion which is
involved in the interaction of selectin and its ligand, sLex, by
converting the sialylated ligand into the asialo form.
To characterize the ROS-directed desialylation in more
detail, a reaction with a model glycoside, 4MU-Neu5Ac,
was analyzed. This substrate, as well as the corresponding
galactoside, was incubated with a ROS generating system
involving HX/XO, H2O2, and Fe2+. As shown in Figure 5A,
the reaction with 4MU-Neu5Ac resulted in the release of
the fluorescent aglycan, 4MU, indicating that the glycosidic
linkage of sialic acid can also be cleaved in vitro. As
revealed by the HPLC analysis, in addition, sialic acid was
also detected as well as 4MU in the ROS treatment,
whereas no by-products were detected (data not shown). On
the other hand, the galactoside was not significantly
cleaved, consistent with the observation that treatment of
HL60 cells with ROS liberated only sialic acid residues at
the nonreducing ends of the cell surface oligosaccharides
with no further degraded. In addition, similar to the data from
experiments using the cells, the addition of SOD and catalase
Effect of reactive oxygen species on cell adhesion
Fig. 3. Adhesion assay of reactive oxygen species (ROS) treated HL60 to
human umbilical vein endothelial cells (HUVEC). (A) Impairment of the
adhesion of HL60 cells to IL-1β-activated HUVEC by the treatment of
hypoxanthine and xanthine oxidase (HX/XO). Fluorescence labeled
HL60 cells were treated as follows; (b) non treated, (c) phosphate-buffered
saline (PBS)/XO, (d) HX/XO, (e) HX/XO + SOD/CAT, (f) anti sLex
antibody, (g) anti E-selectin antibody, and (h) 0.1 U/mL sialidase. HL60
cells were then added to HUVEC and incubated at room temperature for
20 min with rotation. HUVEC were shown as negative control in (a). (B)
Fluorescence intensity of HL60 bound to HUVEC was measured by
fluorescence detection (mean ± S D, n = 4).
protected the 4MU-Neu5Ac from cleavage (Figure 5B).
Because this substrate was not cleaved solely by either of
HX/XO, H2O2, or Fe2+ but by a combination of these
Fig. 4. Effect of reactive oxygen species (ROS) on sLex and Lex structure.
HL60 were treated with (A and F) phosphate-buffered saline (PBS)/XO,
(B and G) HX/XO, (C and H) HX/XO/SOD, (D and I) HX/XO/SOD/
CAT and (E and J) sialidase at 37°C, and the sLex and Lex structure were
analyzed by flow cytometry using specific antibody for sLex (A–E) or Lex
(F–J). HL60 cells stained without the antibody against sLex or Lex were
used as the negative control (gray histogram in panel A and F).
reagents, it appeared that another, such as hydroxyl radical, produced by cooperative interactions of HX/XO,
H2O2, or Fe2+ is primarily required for the glycosidic linkage to be attacked. These results suggest that ROS is capable of directly eliminating non-reducing terminal sialic acid
residues without inducing sialidases in cells.
Discussion
The effect of ROS on the cell surface oligosaccharides was
investigated. The findings clearly show that non-reducing
1097
H. Eguchi et al.
Fig. 5. Effect of reactive oxygen species (ROS) on 4MU-Neu5Ac and
4MU-Gal. (A) Fluorescence derivatives of sialic acid and galactose were
incubated with ROS generating system involving hypoxanthine and xanthine oxidase (HX/XO), H2O2, and Fe2+ at 37°C for indicated times. (B)
4MU-Neu5Ac was treated with HX/XO/H2O2/Fe2+ (closed square), HX/
XO/H2O2/Fe2+ + SOD/CAT (closed circle), HX/XO (closed triangle),
H2O2 (closed diamond), Fe2+ (opened square) at 37°C for indicated
times. Fluorescence intensity of 4-methylumbelliferone was measured at
excitation and emission wavelengths of 365 and 448 nm, respectively.
terminal sialic acid residues are one of the most susceptible targets for ROS. The findings further suggest that
superoxide anion and the related ROS specifically cleave
and liberate the sialic acid residues. As suggested by the in
vitro study involving synthetic substrates, the cleaving
reaction appears to require transition metals such as iron.
Therefore, it is probable that hydroxyl radicals as the
result of Fenton/Haber–Weiss chemistry play a role in the
cleavage. However, additional treatment of HL60 cells
with H2O2 and iron ion with HX/XO did not further
enhance the cleavage of sialic acid (data not shown). The
culture of HL60 cells might contain trace but sufficient
amounts of iron or other transition metals to give rise to
the reaction. A similar modification of cell surface oligosaccharides was also observed in the ROS treatment of
HUVEC, which serves as a barrier between blood and tissues and thus is frequently exposed to ROS (data not
shown). Sialic acid is usually located at the terminal position of oligosaccharides and possesses a negatively
charged carboxyl group at physiological pH. Because of
its unique chemical properties, it is generally thought that
this unique sugar plays an important role in cell adhesion
(Graves et al., 1994; Varki, 1997; Razi and Varki, 1998;
Pretzlaff et al., 2000; Somers et al., 2000). For example, an
increase in α2-6 linked sialic acid by expression of the
α2,6-sialyltransferase gene decreased the VCAM-1 dependent cell adhesion because of the negative charge on the
sialic acid (Abe et al., 1999).
Many reports have indicated that ROS increase cell
adhesion via the up-regulation of adhesion molecules such
as selectins (Patel et al., 1991; Palluy et al., 1992; Lo et al.,
1993), and, thus, an interaction involving sialic acid
appears to be associated with an alteration in cell adhesion
by ROS. The induction of adhesion molecules by ROS
would require a few hours probably because protein expression is transcriptionally regulated. As indicated by this
study, however, it was found that ROS is also capable of
giving rise to altered adhesion within a few minutes via the
impairment of the interaction of sLex and E-selectin. It
seems more likely that this interaction was hindered by the
direct cleavage of a sialic acid residue of the sLex epitope
1098
without any change in protein expression. In fact, such an
effect of ROS on cell adhesion is essentially the same as the
case of the removal of sialic acids by treatment of the cells
with sialidase (Figure 3).
Considering these possibilities, the sialic acid-dependent
cell adhesion, as mediated by the interaction of selectins and
their ligands, may be regulated differentially in a manner
that is dependent on ROS levels. When the level of ROS is
relatively low, the adhesion would be facilitated by the
induction of adhesion molecules. In contrast, if ROS levels
are unusually increased, the loss of sialic acid from the oligosaccharides required for adhesion could suppress cell
adhesion. Under these conditions, the relatively rapid modification of oligosaccharides could play a more significant
role, for example, in the regulation of the excessive accumulation of neutrophils in various inflammatory states.
Because L- and P-selectin-mediated cell adhesion in which
sialic acid in the sLex epitope is involved can be also
impaired by ROS, such an oxidative damage of sLex
epitope could affect other types of cell–cell interactions.
It has been reported that ascorbic acid oxidation causes
the decarboxylation of sialic acid in glycoproteins (Ericson
and Bratt, 1987). The cleavage of sialic acid by ROS, as
found in this study, did not involve decarboxylation
because DMB reacts with the sialic acid whose carboxyl
group is intact. Thus, it is unlikely that the mechanism of
sialic acid cleavage by HX/XO is similar to the ascorbic oxidation. However, the carboxyl group of sialic acid appears
to be the most susceptible target of oxidative damage in the
structure of the sugar. A synthetic sialo-type substrate was
cleaved by ROS, whereas a galactoside substrate was essentially not damaged. In addition, as shown by flow cytometric analyses using lectins, significant changes in cell surface
oligosaccharides were not observed except for alterations
based on elimination of sialic acids. Therefore, it seems
most likely that a carboxyl group is required for a sugar to
be damaged by ROS, and thus the ROS would primarily
attack the carboxyl group of the sugar, subsequently leading to the cleavage of the glycosidic linkage. It is likely that
the radicals generated by HX/XO transfer one electron to
carboxyl group of sialic acid, and the resulting sugar-radical
in which an unpaired electron is donated by the carboxyl
group may then attack C-2 of the sialic acid, consequently
leading to cleavage of the glycosidic linkage. On the other
hand, Hawkins and Davies (1998) examined the degradation of hyaluronic acid by HOCl/OCl− and suggested that
homolysis of the chloramide species forms a nitrogencentered radical which leads to a formation of a carboncentered radical on the N-acetyl group. In the light of this
report, it is also possible that ROS generated by HX/XO
form the sugar-radicals on the N-acetyl group. Subsequent modifications of functional groups such as
hydroxyl one in sialic acid, which is known to important
in the interaction with selectin, may abrogate the binding
of HL60 cells. A significant fraction of the sLex epitope
could be immunoreactive toward the antibody used but
not functional in the selectin-binding. In the HX/XO
treatment, loss of the binding was drastic compared to
that of the antibody staining in contrast to the case of
sialidase treatment (Figures 3 and 4). However, the radical species responsible for the cleavage of the glycosidic
Effect of reactive oxygen species on cell adhesion
linkage as well as a detailed mechanism remain to be
investigated.
It is possible that the cleavage of sialic acid by ROS cause
impaired functions of glycoproteins and sugar chains
because the sialylation of many types of glycoconjugates is
often associated with various biological events (Kodama
et al., 1993; Pilatte et al., 1993; Schwarzkopf et al., 2002).
As reported, for example, oxidative damage decreases the
sialic acid content of erythropoietin, a highly glycosylated
protein and, as a result, degradation of the protein follows
(Uchida et al., 1997). Chappey et al. (1998) showed that the
low density lipoprotein (LDL) sialic acid content in patients
with advanced coronary artery disease was lower than in
healthy subjects. Tanaka et al. (1997) reported that the content of sialic acid on LDL was decreased steadily by oxidation. Furthermore, it has been shown that serum sialic acid
concentrations are elevated in cases of inflammation and
cancer, under which conditions oxidative stress tends to arise
and/or to be enhanced (Stefenelli et al., 1985; Painbeni et al.,
1997; Goswami et al., 2003). Thus, it is reasonably certain
that the loss of sialic acid residues by ROS can be a causal
factor for dysfunction in cells, tissues and organs. The this
study might provide a possible mechanism for the cleavage of sialic acid, and, in addition, explain how ROS is
associated with pathogenesis and the progression of diseases such as atherosclerosis and cancer from the point of
view of glycobiology.
Materials and methods
Chemicals
XO was purchased from Roche Diagnostics (Tokyo, Japan).
3´,6´-(Di(O-acetyl)-2´,7´-bis [N,N-bis(carboxymetyl)aminometyl] fluorescein, tetraacetoxymethyl ester (calcein-AM)
was purchased from Dojindo (Tokyo, Japan). Fluorescein
isothiocyanate (FITC) conjugated lectins, SSA, MAM, RCA,
ConA, and anti Lex monoclonal antibody were purchased
from Seikagaku Corporation (Tokyo, Japan). Sialic acid
detection kit was purchased from Takara Biochemicals (Shiga,
Japan). Sialidase from Arthrobacter ureafaclens, 4-methylumbelliferyl-N-acetylneuraminic acid (4MU-Neu5Ac) and
4-methylumbelliferyl-β-D-galactoside (4MU-Gal) were purchased from Nakalai tesque (Kyoto, Japan). Anti sLex
monoclonal antibody was purchased from Wako Pure Chemical Industries (Osaka, Japan). Mouse anti human CD62E
(E-selectin) neutralizing antibody was purchased from
Calbiochem-Novabiochem Corporation (San Diego, CA).
FITC-conjugated goat anti mouse Ig was purchased from
Dako (Kyoto, Japan). The other reagents were of the highest grade available.
Cell culture
HL 60 cells were cultured on 100-mm Petri dishes in
RPMI1640 medium containing 10% fetal calf serum (FCS).
HUVEC were isolated by dispase treatment from umbilical
vein cord as described previously (Suzuki et al., 1993).
HUVEC were grown on 24-well flat bottom microplates in
MCDB131 medium containing 10% FCS, 10 ng/mL recombinant human basic fibroblast growth factor, 1 µg/mL
hydrocortisone, 18 mU/mL heparin, and 0.25 µg/mL
amphotericin B. These cells were cultured at 37°C in a
humidified atmosphere with 5% CO2.
ROS treatment of HL60
HL60 were washed twice with phosphate-buffered saline
(PBS; pH 7.4) to remove the FCS-containing medium.
HL60 (2 × 107 cells) were treated with 2 mM HX and
50 mU XO in PBS at 37°C for 5 min. The ROS generation
was terminated by the addition of 100 µg/mL superoxide
dismutase and 100 µg/mL catalase (SOD/CAT) or FCScontaining RPMI1640 medium. ROS-treated HL60 (ROSHL60) were used immediately in under experiments.
Lectin analysis
ROS-HL60 were labeled with FITC-conjugated SSA,
MAM, RCA, or ConA lectin diluted to 1:500 with PBS at
4°C for 60 min. In the sialidase treatment, HL60 were incubated with 0.1 U/mL sialidase at 37°C for 30 min before the
addition of lectins. After removing residual unbound lectin
by washing with PBS three times, fluorescence was analyzed using flow cytometry (FACScan, Beckton Dickinson,
Franklin Lakes, NJ).
Sialic acid analysis
After treatment of HL60 with HX/XO or 0.1 U/mL sialidase, the supernatants were collected, lyophilized, and dissolved in 50 µL of water. The fluorescence labeling of sialic
acid using DMB was performed according to manufacturer’s instruction (Takara).
Fifty µL of sample was incubated with 200 µL of fluorescent reagent at 50°C for 2.5 h. For the hydrolysis
assay, samples were incubated with 2.6 µL of 1 N HCl at
80°C for 60 min before the addition of the fluorescent
reagent. The reaction was stopped by cooling the solution to 0°C. Sialic acid was analyzed by reversed phase
HPLC. A 50-µL aliquot was applied to an HPLC system
(600E, Waters, Milford, MA) equipped with a PALPAK
Type R column (4.6 mm × 250 mm, Takara). Elution was
performed at room temperature using an eluent composed
of acetonitrile (9%)–methanol (7%) at a flow rate 0.8 mL/min.
Fluorescence was monitored with excitation and emission
wavelengths of 373 and 448 nm, respectively. Sialic acid
concentrations were determined by determining the amount
of DMB group using a known amount of sialic-acid–DMB
complex as a standard.
Cell adhesion assay
A monolayer cell adhesion assay was performed as
described previously (Takada et al., 1993). HL60 cells were
washed with RPMI1640 medium and resuspended with 3 mL
of the same medium. For the fluorescence labeling of
HL60, 3 µL of calcein-AM was added and the solution was
then incubated at room temperature for 15 min in dark.
After washing with PBS, the HL60 cells were treated with
HX/XO in the presence or absence of SOD/CAT. In the
ligand antagonist assay, fluorescence labeled HL 60 were
incubated with a neutralizing antibody for sLex or E-selectin
at room temperature for 30 min. In the sialidase treatment,
1099
H. Eguchi et al.
HL60 cells were incubated with 0.1 U/mL sialidase at 37°C
for 30 min. HL60 cells were washed once with RPMI1640
medium and resuspended in the same medium. HUVEC
were incubated with 1 ng/mL IL-1β at 37°C for 4 h before
the cell adhesion experiment of HL60. After the washing
of HUVEC with RPMI1640 medium twice, HL60 were
added at a concentration of 1 × 105 in 500 µL to HUVEC
and incubated at room temperature with rotation. After
incubation for 20 min, HUVEC were washed gently with
PBS containing 1 mM CaCl2 and 1 mM MgCl2 to
remove residual unbound HL60. For the measurement of
fluorescence intensity, bound HL60 were dissolved in a
1% TritonX-100 solution at 37°C for 2 h in the dark.
Fluorescence was measured with excitation and emission
wavelengths of 490 and 520 nm, respectively. The percent
adhesion was calculated on the assumption that 500 µL
of calcein-AM labeled cell suspension represent 100%.
sLex and Lex structures on cell surface
ROS-HL60 were incubated with mouse anti sLex or Lex
antibody diluted to 1:1000 with PBS at 4°C for 60 min.
After washing three times with PBS, cells were reacted with
FITC-conjugated goat anti mouse Ig at 4°C for 30 min.
HL60 cells stained without first antibodies were used as the
negative control. Fluorescence intensity was measured
using flow cytometry.
ROS treatment of 4MU-Neu5Ac or 4MU-Gal
4MU-Neu5Ac or 4MU-Gal (0. 2 mM each) was incubated
at 37°C in 100 µL of PBS containing HX/XO, 0.1 mM
H2O2, and 0.1 mM FeCl2. The reaction was terminated by
the addition of SOD and catalase. 2.5 mL of 0.25 M
glycine–NaOH buffer (pH 10.4) was added to enhance the
fluorescence intensity of 4-methylumbelliferone. The fluorescence was measured with a fluorescence detector at excitation and emission wavelengths of 365 and 448 nm,
respectively. The released 4-methylumbelliferone concentration was determined using a known amount of 4-methylumbelliferone as a standard.
Acknowledgment
We thank Dr. Hideyuki Ihara and Dr. Ken Sasai for helpful discussions and information, and Dr. Tomohiko Taguchi
for critical reading of the manuscript. This work was supported by a Grant-in-Aid for Researchers, Hyogo College
of Medicine, a Hitech Research Center grant from the Ministry of Education, Science and Culture of Japan.
Abbreviations
4MU-Gal, 4-methylumbelliferyl-galactoside; 4MU-Neu5Ac,
4-methylumbelliferyl-N-acetylneuraminic
acid;
ConA,
Canavalia ensiformis; DMB, 1,2-diamino-4,5-methylenedioxybenzene; FITC, fluorescein isothiocyanate; HPLC, highperformance liquid chromatography; HUVEC, human
umbilical vein endothelial cells; HX, hypoxanthine; LDL,
low density lipoprotein; Lex, lewis x; MAM, Maackia
amurensis; PBS, phosphate-buffered saline; RCA, Ricinus
1100
communis; ROS, reactive oxygen species; sLex, sialyl lewis
x; SOD, superoxide dismutase; SSA, Sambucus sieboldiana;
XO, xanthine oxidase.
References
Abe, Y., Smith, C.W., Katkin, J.P., Thurmon, L.M., Xu, X., Mendoza,
L.H., and Ballantyne, C.M. (1999) Endothelial alpha 2,6-linked sialic
acid inhibits VCAM-1-dependent adhesion under flow conditions.
J. Immunol., 163, 2867–2876.
Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C.B., and Seed, B.
(1990) CD44 is the principal cell surface receptor for hyaluronate. Cell,
61, 1303–1313.
Beresewicz, A., Czarnowska, E., and Maczewski, M. (1998) Ischemic preconditioning and superoxide dismutase protect against endothelial dysfunction and endothelium glycocalyx disruption in the postischemic
guinea-pig hearts. Mol. Cell. Biochem., 186, 87–97.
Chappey, B., Beyssen, B., Foos, E., Ledru, F., Guermonprez, J.L., Gaux,
J.C., and Myara, I. (1998) Sialic acid content of LDL in coronary
artery disease: no evidence of desialylation in subjects with coronary
stenosis and increased levels in subjects with extensive atherosclerosis
and acute myocardial infarction: relation between desialylation and in
vitro peroxidation. Arterioscler. Thromb. Vasc. Biol., 18, 876–883.
Curran, S.F., Amoruso, M.A., Goldstein, B.D., and Berg, R.A. (1984)
Degradation of soluble collagen by ozone or hydroxyl radicals. FEBS
Lett., 176, 155–160.
Czarnowska, E. and Karwatowska, P.E. (1995) Ultrastructural demonstration of endothelial glycocalyx disruption in the reperfused rat
heart. Involvement of oxygen free radicals. Basic Res. Cardiol., 90,
357–364.
Eguchi, H., Ikeda, Y., Koyota, S., Honke, K., Suzuki, K., Gutteridge,
J.M., and Taniguchi, N. (2002) Oxidative damage due to copper ion
and hydrogen peroxide induces GlcNAc-specific cleavage of an Asnlinked oligosaccharide. J. Biochem., 131, 477–484.
Ericson, T. and Bratt, P. (1987) Interactions between peroxide and salivary
glycoprotein: protection by peroxidase. J. Oral Pathol., 16, 421–424.
Farmer, H., Koch, H.P., and Sutton, D.A. (1943) The course of autoxidation reactions in polyisoprenes and allied compounds. Part VII. Rearrangement of double bonds during autoxidation. J. Chem. Soc., 1943,
541–547.
Fridovich, I. (1970) Quantitative aspects of the production of superoxide
anion radical by milk xanthine oxidase. J. Biol. Chem., 245, 4053–4057.
Goswami, K., Nandakumar, D.N., Koner, B.C., Bobby, Z., and Sen, S.K.
(2003) Oxidative changes and desialylation of serum proteins in hyperthyroidism. Clin. Chim. Acta, 337, 163–168.
Graves, B.J., Crowther, R.L., Chandran, C., Rumberger, J.M., Li, S.,
Huang, K.S., Presky, D.H., Familletti, P.C., Wolitzky, B.A., and
Burns, D.K. (1994) Insight into E–selectin/ligand interaction from the
crystal structure and mutagenesis of the lec/EGF domains. Nature,
367, 532–538.
Greenwald, R.A. and Moy, W.W. (1979) Inhibition of collagen gelation
by action of the superoxide radical. Arthritis Rheum., 22, 251–259.
Greenwald, R.A. and Moy, W.W. (1980) Effect of oxygen-derived free
radicals on hyaluronic acid. Arthritis Rheum., 23, 455–463.
Grootveld, M., Henderson, E.B., Farrell, A., Blake, D.R., Parkes, H.G.,
and Haycock, P. (1991) Oxidative damage to hyaluronate and glucose
in synovial fluid during exercise of the inflamed rheumatoid joint.
Detection of abnormal low-molecular-mass metabolites by protonn.m.r. spectroscopy. Biochem. J., 273, 459–467.
Halliwell, B. (1978) Superoxide-dependent formation of hydroxyl radicals
in the presence of iron salts. Its role in degradation of hyaluronic acid
by a superoxide-generating system. FEBS Lett., 96, 238–242.
Hawkins, C.L. and Davies, M.J. (1996) Direct detection and identification
of radicals generated during the hydroxyl radical-induced degradation
of hyaluronic acid and related materials. Free Radic. Biol. Med., 21,
275–290.
Hawkins, C.L. and Davies, M.J. (1998) Degradation of hyaluronic acid,
poly- and monosaccharides, and model compounds by hypochlorite:
evidence for radical intermediates and fragmentation. Free Radic. Biol.
Med., 24, 1396–1410.
Effect of reactive oxygen species on cell adhesion
Halliwell, B. and Gutteridge, J.M.C. (1999) Free Radicals in Biology and
Medicine, 3nd ed. Oxford University Press, New York.
Jahn, M., Baynes, J.W., and Spiteller, G. (1999) The reaction of hyaluronic acid and its monomers, glucuronic acid and N-acetylglucosamine, with reactive oxygen species. Carbohydr. Res., 321, 228–234.
Kansas, G.S. (1996) Selectins and their ligands: current concepts and controversies. Blood, 88, 3259–3287.
Kodama, S., Tsujimoto, M., Tsuruoka, N., Sugo, T., Endo, T., and
Kobata, A. (1993) Role of sugar chains in the in-vitro activity of
recombinant human interleukin 5. Eur. J. Biochem., 211, 903–908.
Lesley, J., Hyman, R., and Kincade, P.W. (1993) CD44 and its interaction
with extracellular matrix. Adv. Immunol., 54, 271–335.
Lo, S.K., Janakidevi, K., Lai, L., and Malik, A.B. (1993) Hydrogen peroxide-induced increase in endothelial adhesiveness is dependent on
ICAM-1 activation. Am. J. Physiol., 264, L406–L412.
Mccord, J. (1974) Free radicals and inflammation: protection of synovial
fluid by superoxide dismutase. Science, 185, 529–531.
Moseley, R., Waddington, R., Evans, P., Halliwell, B., and Embery, G.
(1995) The chemical modification of glycosaminoglycan structure by
oxygen-derived species in vitro. Biochim. Biophys. Acta, 1244, 245–252.
Moseley, R., Waddington, R.J., and Embery, G. (1997) Degradation of
glycosaminoglycans by reactive oxygen species derived from stimulated polymorphonuclear leukocytes. Biochim. Biophys. Acta, 1362,
221–231.
Naot, D., Sionov, R.V., and Ish-Shalom, D. (1997) CD44: structure, function, and association with the malignant process. Adv. Cancer Res., 71,
241–319.
Ookawara, T., Kawamura, N., Kitagawa, Y., and Taniguchi, N. (1992)
Site-specific and random fragmentation of Cu,Zn-superoxide dismutase by glycation reaction. J. Biol. Chem., 267, 18505–18510.
Painbeni, T., Gamelin, E., Cailleux, A., Le Bouil, A., Boisdron-Celle, M.,
Daver, A., Larra, F., and Allain, P. (1997) Plasma sialic acid as a
marker of the effect of the treatment on metastatic colorectal cancer.
Eur. J. Cancer, 33, 2216–2220.
Palluy, O., Morliere, L., Gris, J.C., Bonne, C., and Modat, G. (1992)
Hypoxia/reoxygenation stimulates endothelium to promote neutrophil
adhesion. Free Radic. Biol. Med., 13, 21–30.
Patel, K.D., Zimmerman, G.A., Prescott, S.M., McEver, R.P., and
McIntyre, T.M. (1991) Oxygen radicals induce human endothelial
cells to express GMP-140 and bind neutrophils. J. Cell Biol., 112,
749–759.
Phillips, M.L., Nudelman, E., Gaeta, F.C., Perez, M., Singhal, A.K.,
Hakomori, S., and Paulson, J.C. (1990) ELAM-1 mediates cell adhesion by recognition of a carbohydrate ligand, sialyl-Lex. Science, 250,
1130–1132.
Pilatte, Y., Bignon, J., and Lambre, C.R. (1993) Sialic acids as important
molecules in the regulation of the immune system: pathophysiological
implications of sialidases in immunity. Glycobiology, 3, 201–218.
Ponta, H., Sherman, L., and Herrlich, P.A. (2003) CD44: from adhesion
molecules to signalling regulators. Nat. Rev. Mol. Cell Biol., 4, 33–45.
Pretzlaff, R.K., Xue, V.W., and Rowin, M.E. (2000) Sialidase treatment
exposes the beta1-integrin active ligand binding site on HL60 cells and
increases binding to fibronectin. Cell Adhes. Commun., 7, 491–500.
Razi, N. and Varki, A. (1998) Masking and unmasking of the sialic acidbinding lectin activity of CD22 (Siglec-2) on B lymphocytes. Proc.
Natl. Acad. Sci. U. S. A., 95, 7469–7474.
Sabri, S., Soler, M., Foa, C., Pierres, A., Benoliel, A., and Bongrand, P.
(2000) Glycocalyx modulation is a physiological means of regulating
cell adhesion. J. Cell Sci., 113, 1589–1600.
Schwarzkopf, M., Knobeloch, K.P., Rohde, E., Hinderlich, S., Wiechens, N.,
Lucka, L., Horak, I., Reutter, W., and Horstkorte, R. (2002) Sialylation
is essential for early development in mice. Proc. Natl. Acad. Sci. U. S. A.,
99, 5267–5270.
Soler, M., Desplat-Jego, S., Vacher, B., Ponsonnet, L., Fraterno, M.,
Bongrand, P., Martin, J.M., and Foa, C. (1998) Adhesion-related glycocalyx study: quantitative approach with imaging-spectrum in the
energy filtering transmission electron microscope (EFTEM). FEBS
Lett., 429, 89–94.
Somers, W.S., Tang, J., Shaw, G.D., and Camphausen, R.T. (2000)
Insights into the molecular basis of leukocyte tethering and rolling
revealed by structures of P- and E-selectin bound to SLe (X) and
PSGL-1. Cell, 103, 467–479.
Springer, T. (1990) Adhesion receptors of the immune system. Nature, 346,
425–434.
Stefenelli, N., Klotz, H., Engel, A., and Bauer, P. (1985) Serum sialic acid
in malignant tumors, bacterial infections, and chronic liver diseases.
J. Cancer Res. Clin. Oncol., 109, 55–59.
Suzuki, K., Tatsumi, H., Satoh, S., Senda, T., Nakata, T., Fujii, J., and
Taniguchi, N. (1993) Manganese-superoxide dismutase in endothelial
cells: localization and mechanism of induction. Am. J. Physiol., 265,
H1173–H1178.
Suzuki, K., Eguchi, H., Koh, Y.H., Park, Y.S., and Taniguchi, N. (1999)
Acceleration of adhesion of cancer cells and neutrophils to endothelial
cells in the absence of de novo protein synthesis: possible implication
for involvement of hydroxyl radicals. Biochem. Biophys. Res. Commun., 257, 214–217.
Takada, A., Ohmori, K., Yoneda, T., Tsuyuoka, K., Hasegawa, A., Kiso, M.,
and Kannagi, R. (1993) Contribution of carbohydrate antigens sialyl
Lewis A and sialyl Lewis X to adhesion of human cancer cells to vascular
endothelium. Cancer Res., 53, 354–361.
Tanaka, K., Tokumaru, S., and Kojo, S. (1997) Possible involvement of
radical reactions in desialylation of LDL. FEBS Lett., 413, 202–204.
Uchida, E., Morimoto, K., Kawasaki, N., Izaki, Y., Abdu Said, A., and
Hayakawa, T. (1997) Effect of active oxygen radicals on protein and
carbohydrate moieties of recombinant human erythropoietin. Free
Radic. Res., 27, 311–323.
Varki, A. (1997) Sialic acids as ligands in recognition phenomena. FASEB
J., 11, 248–255.
Vestweber, D. and Blanks, J.E. (1999) Mechanisms that regulate the function of the selectins and their ligands. Physiol. Rev., 79, 181–213.
Walz, G., Aruffo, A., Kolanus, W., Bevilacqua, M., and Seed, B. (1990)
Recognition by ELAM-1 of the sialyl-Lex determinant on myeloid and
tumor cells. Science, 250, 1132–1135.
Ward, B.J. and Donnelly, J.L. (1993) Hypoxia induced disruption of the
cardiac endothelial glycocalyx: implications for capillary permeability.
Cardiovasc. Res., 27, 384–389.
1101