| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Glycobiology Research and Training Center, Department of Medicine, University of California, and
The Burnham Institute, La Jolla, CA 92037
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
BAP was prepared as described previously (5, 6).
Recombinant streptavidin, bovine lung acetone powder, glutaraldehyde,
Ig isotyping kit, protein G-Sepharose, platelet-activating factor
(PAF), PMA, TNF-
, anti-human platelet-endothelial cell adhesion
molecule (PECAM; CD31), monoclonal anti-BSA, R-PE-conjugated
anti-mouse IgG, exoglycosidases, and zymosan were obtained from
Sigma (St. Louis, MO). The following materials were obtained from the
indicated sources: ionomycin, Calbiochem (La Jolla, CA);
peptide-N-glycosidase (PNGase) F, New England Biolabs
(Beverly, MA); PNGase A, a gift from Seikagaku America (Falmouth, MA);
proteinase K, Roche (Indianapolis, IN); Bio-Gel P2 and
peroxidase-linked goat anti-mouse IgG, Bio-Rad (Richmond, CA);
1-ethyl-3(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) and
N-hydroxysuccinimide (NHS), Pierce (Rockford, IL); alkaline
phosphatase-conjugated goat anti-mouse IgG, Promega (Madison, WI);
peroxidase-linked streptavidin, The Binding Site (San Diego, CA); Cy3
mAb labeling kit, Amersham Pharmacia (Arlington Heights, IL); Monopoly
Resolving Medium, ICN Biomedicals (Aurora, OH); Nycoprep, Nycomed
Pharma (Oslo, Norway); FITC anti-mouse Ly-6G (Gr-1) and PE
anti-mouse CD11b (Mac-1), PharMingen (San Diego, CA); covalent
binding plates, Xenopore (Hawthorne, NJ); bovine tissues, Pel-Freeze
(Rogers, AR); and human tissue sections, Histology Core Facility of the
Cancer Center, University of California (San Diego, CA). HUVECs were
obtained from Cell Applications System (San Diego, CA), and calf
pulmonary artery endothelial (CPAE) cells were obtained from
American Type Culture Collection (Manassas, VA).
Preparation of neoglycoproteins for immunization and screening
Anionic bovine lung glycans of moderate negative charge coupled
to BAP (D. K. Toomre et al., unpublished observations) were
treated with Arthrobacter ureafaciens sialidase (10 mU),
jack bean 
-N-acetylhexosaminidase (53 mU), bovine
testicular -galactosidase (2 mU), coffee bean -galactosidase (5
mU), and jack bean -mannosidase (2–5 mU) in 
20 µl of sodium
citrate buffer (pH 4.5). These multiple exoglycosidases were
inactivated by heating at 100°C for 5–10 min, and the sample was
fractionated on a DEAE-2SW HPLC column as previously described
(6). Specific subfractions from the DEAE column were also
analyzed by reverse phase HPLC as previously described
(6). The mixed BAP-coupled oligosaccharides enriched for
the carboxylate residue (see Fig. 1
) were mixed with streptavidin in
PBS at a ratio of 1:3, and the mixture was kept at 4°C for 1 h
before immunization of mice. To generate BSA neoglycoproteins for
screening, glycopeptides were prepared from bovine lung acetone powder
by proteinase K digestion followed by purification on a Bio-Gel P2
column run in 0.1 M ammonium formate, pH 6.5. Glycopeptides eluting in
the void volume were conjugated to BSA using either glutaraldehyde or
EDC following standard protocols. The BSA conjugates (henceforth
referred to as BSA neoglycoproteins for the glutaraldehyde-coupled
glycopeptides, unless otherwise stated) were analyzed by PAGE gels and
by neutral sugar estimation (phenol sulfuric acid assay), which
indicated an average of 2.5 mol of N-glycans coupled/mol of
protein (assuming a typical biantennary N-glycan; data not
shown).
|
Glycopeptides generated from bovine lung acetone powder as described above were desialylated by mild acid treatment (10 mM HCl, 30 min at 100°C) and lyophilized. (These glycopeptides are henceforth referred to as asialo-COO--glycopeptides). Five hundred nanomoles of such glycopeptides (by neutral sugar estimation) were dissolved in 50 µl of 50 mM MES (2-[N-morpholino]ethanesulfonic acid) buffer, pH 5.5, followed by addition of 100 µl of 1 M methylamine. Fifty microliters of EDC/NHS from a freshly prepared stock solution of 100 mg of EDC and 50 mg of NHS/ml water was added and incubated at 37°C. After 1 h, another 50 µl of fresh EDC/NHS was added, and the incubation was continued for another 2 h. Control glycopeptides were treated identically, except that EDC/NHS solution was replaced with water. These carboxylate-neutralized glycopeptides (henceforth referred to as asialo-CONHMe-glycopeptides) were then dialyzed against water overnight to remove excess methylamine, and coupling to BSA was conducted as described above. Carboxylate neutralization using EDC-NHS/methylamine caused about 70–80% of the negative charges to be masked as determined by QAE-Sephadex chromatography (data not shown).
Immunization procedures and hybridoma establishment
Two- to 4-wk-old female BALB/c mice were primed by s.c. injection of 10 µg of native streptavidin in CFA. Seven days later they were immunized with 10 µg of streptavidin neoglycoconjugates in IFA and then boosted i.p. with PBS at 2-wk intervals until significant titers of serum Abs against BSA neoglycoproteins were generated. Splenocytes from the mice were isolated and fused with Ag8.653 mouse myeloma cells using standard protocols. Hybridomas were cultured in DMEM (high glucose; Life Technologies, Gaithersburg, MD) supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine, 20% FBS, and 100 µM hypoxanthine/16 µM thymidine. Screening was performed by ELISAs against bovine lung glycopeptides coupled to BSA, thus selecting for Abs directed against glycans and avoiding any against streptavidin and BAP. Seven highly reactive and 18 moderately reactive hybridomas were cloned by limiting dilution and subcloned as necessary. They were further screened for differential reactivity against BSA-coupled asialo-COO--glycopeptides and asialo-CONHMe-glycopeptides. Selected hybridoma clones/subclones were injected i.p. into BALB/c mice for the production of ascitic fluid, and IgGs were purified from culture supernatants or ascitic fluid on protein G-Sepharose. One of the clones (mAb GB3.1, Ig subclass IgG2b) was chosen for further characterization, because it showed a marked difference in reactivity between asialo-COO--glycopeptides and asialo-CONHMe-glycopeptides, and functioned well in both ELISAs and immunoblots.
ELISAs
Ninety-six-well plates were coated with 250 ng of BSA neoglycoproteins for 6 h at room temperature or 4 h at 37°C. Plates were washed and blocked with 3% BSA in PBS overnight and incubated with hybridoma culture supernatants or purified IgG diluted in PBS containing 1% BSA and 0.05% Tween-20. Incubations were performed for 1–2 h at room temperature or overnight at 4°C. The plates were then washed and incubated with alkaline phosphatase-conjugated goat anti-mouse IgG, followed by development with p-nitrophenyl phosphate substrate. Absorbance was measured at 405 nm in an ELISA plate reader.
Deglycosylation of papain-released bovine lung glycopeptides with PNGase A/PNGase F
Bovine lung acetone powder was digested with papain in 0.1 M
sodium acetate, pH 5.5, in the presence of 5 mM L-cysteine
and 1 mM EDTA at 60°C for 48 h. The mixture was heat
inactivated, centrifuged at 10,000 x g to remove
undigested residue, and dialyzed using 3000 m.w. cut-off tubing.
Glycopeptides (100 nmol, as determined by neutral sugar quantitation
using phenol-sulfuric acid) were digested with PNGase A in 0.1 M sodium
acetate or 0.1 M citrate phosphate, pH 5.5, in the ratio of 2 µU/nmol
of glycopeptide or with PNGase F in 50 mM Tris, pH 8.0, for 16 h
at 37°C, both in duplicate. Control tubes had no added enzyme. After
digestion, the mixtures were heat inactivated. The glycopeptides with
or without treatments were coated onto covalent binding plates, and an
ELISA against mAb GB3.1 was conducted as described above.
Immunostaining of cells and tissues
Tissue sections were fixed in 10% buffered formalin for 20 min, followed by blocking of nonspecific binding sites with 10% normal goat serum in PBS containing 1% BSA. After washing, Abs were overlaid onto serial tissue sections at predetermined dilutions (usually between 1 and 10 µg/ml), and the slides were incubated in a humid atmosphere for 30 min at room temperature or overnight at 4°C. The labeled streptavidin-biotin kit (Dako, Carpenteria, CA) was used following the manufacturer’s instructions with PBS or Tris-buffered saline washes between every step. Biotinylated anti-mouse IgG was applied for 10 min, followed by either alkaline phosphatase or peroxidase-linked streptavidin for 10 min. After another wash, the appropriate substrate was added, and the slides were incubated in the dark for 20 min. After a wash in buffer, slides were counterstained with hematoxylin, mounted, and viewed using an Olympus BH2 microscope (New Hyde Park, NY).
CPAE or HUVEC cells for cell surface staining were grown on Lab-Tek Chamber glass slides (Nunc, Naperville, IL), washed with PBS, and fixed in 10% buffered formalin for 20 min. The slides were blocked with 3% BSA in PBS for 20 min and then incubated with 10 µg/ml of one of the following Cy3-labeled Abs, labeled according to the manufacturer’s instructions: purified mAb GB3.1 or anti-human CD31 (positive control for endothelial cell staining) or CAB4 (negative control) (7). After overnight incubation at 4°C, the slides were washed and observed using epifluorescence microscopy.
Western blot analysis
Tissue or cell proteins were subjected to SDS-PAGE in 12% polyacrylamide gels under reducing conditions and transferred to nitrocellulose membranes. The membranes were blocked overnight with 10% skimmed milk or 3% BSA in PBS, washed with TBS containing 0.05% Tween 20, and incubated for 1–2 h at room temperature with Ab from hybridoma culture supernatants (or purified IgG) diluted in PBS containing 1% BSA and 0.05% Tween 20. Membranes were then reacted with alkaline phosphatase-conjugated goat anti-mouse IgG, and reactive proteins were visualized by incubation with the 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium substrate.
Isolation and activation of human peripheral blood leukocytes
Human neutrophils were isolated from fresh peripheral
(EDTA-anticoagulated) blood of healthy volunteers by differential
migration through monopoly resolving medium (Ficoll-Hypaque; density,
1.114) followed by hypotonic lysis of residual erythrocytes. The
preparations were >98% granulocytes as confirmed using Turk’s
staining (0.01% crystal violet in 3% acetic acid). Cells were
resuspended at 5 x 106 neutrophils/ml in
HBSS buffer containing 0.2% BSA and were used within 30 min after
isolation for adhesion assays. For flow cytometric assays, mononuclear
cells from the above separation were further fractionated into
monocytes and lymphocytes using Nycoprep medium (cells 80% pure).
For activation, cells were resuspended at 3 x
106/ml in 10 mM HEPES buffer, pH 7.5, containing
150 mM NaCl, 5 mM KCl, 1.2 mM MgCl2, 1.3 mM
CaCl2, and 5.5 mM glucose and were incubated in
the presence or the absence of 1 µM PAF (neutrophils) or 100 ng/ml
PMA/1 µM ionomycin (neutrophils and mononuclear cells) for 15 min at
37°C. Cells were pelleted at 300 x g for 10 min and
analyzed as described below.
Flow cytometric assays
Activated and unactivated peripheral blood leukocyte populations were incubated with 50 µg/ml BSA-conjugated asialo-COO--glycopeptides or asialo-CONHMe-glycopeptides in HBSS buffer for 30 min on ice, followed by mouse monoclonal anti-BSA and R-PE-conjugated anti-mouse IgG, each for similar lengths of time on ice and with two washes in between. Cells stained with second- and third-stage reagents only or with BSA followed by second- and third-stage reagents, were used as negative controls. They were analyzed by flow cytometry employing a FACScan (Becton Dickinson, Mountain View, CA) equipped with CellQuest software and were gated by side scatter and forward scatter filters.
Neutrophil adhesion assays with immobilized carboxylated glycans
Ninety-six-well microtiter plates were coated with 250 ng/well of BSA-coupled bovine lung asialo-COO--glycopeptides for 4 h at 37°C. Plates were washed and blocked with 3% BSA in PBS overnight. Control wells were coated with BSA alone. Neutrophils (0.5 x 106) in 100 µl of HBSS buffer containing 0.2% BSA and 1 µM PAF were added to each well and incubated for 30 min at 37°C in the presence or the absence of varying concentrations of mAb GB3.1, control Ab, asialo-COO-, or asialo-CONHMe-glycopeptides in a total volume of 200 µl. PAF was excluded from the incubation mixture in some wells. Unbound cells were removed by flicking out the plates and washing five times with HBSS. Bound cells were quantitated by assaying myeloperoxidase. Briefly, adherent cells were solubilized in 100 µl of 1% Triton X-100 and incubated with 25 µl of 0.5% hydrogen peroxide and 25 µl of 4.7 mM orthodianisidine for 30 min at room temperature. The reaction was stopped with 0.4% sodium azide, and absorbance was read at 450 nm. Assays were performed in duplicate.
Neutrophil adhesion to endothelial cells
HUVECs were grown to confluence in endothelial cell growth medium (Cell Applications, San Diego, CA) at 37°C in 5% CO2. Cells at passages 2–3 that were positive for cell staining and Western blotting with mAb GB3.1 were used in the adhesion assays. At confluence, HUVECs were detached and seeded on fibronectin (1 µg/cm2)-coated 96-well culture wells. Confluent EC monolayers were treated with medium containing 1 µM PAF for 4 h at 37°C and washed. In some experiments unstimulated endothelial cells were also tested. Neutrophils isolated and suspended in adhesion buffer as described above were added and incubated under static conditions in the presence of PAF for 30 min at 37°C in 5% CO2 in the presence or the absence of modifiers. Nonadherent cells were removed by washing, and adherent cells were quantitated using myeloperoxidase assay as described above.
Zymosan-induced acute peritoneal inflammation
Female (BALB/c) mice (Harlan Sprague Dawley, Indianapolis, IN;
16–18 g body weight) were maintained on a standard chow pellet diet
with tap water ad libitum and were used 2–3 days after arrival. The
experimental protocols followed the criteria of the institutional
animal care and use committee at the Burnham Institute (La Jolla, CA).
Peritonitis was induced by injection of 1 mg of zymosan i.p. in 0.5 ml
of PBS. Animals were injected i.v. with saline, mAb GB3.1, or isotype
control mouse IgG at the time of zymosan injection. Mice were
euthanized at different time points, and peritoneal cavities were
lavaged with RPMI medium containing 2% FBS, 3 mM EDTA, and 25 U/ml
heparin. Lavage fluids were centrifuged, aliquots of cells were stained
with Turk’s solution, and differential cell counts were performed with
a Neubauer hemocytometer. In parallel experiments, the peritoneal
exudate cells were also immunostained with either FITC-labeled anti
mouse Gr-1 (Ly6G) or PE-labeled anti-mouse Mac-1 for 20–30 min on
ice, washed, and analyzed by flow cytometry. The mesenteries obtained
from these animals were fixed in 10% buffered formalin, embedded in
paraffin, stained with hematoxylin and eosin, and evaluated by
brightfield microscopy using a Nikon microscope (Melville, NY). Digital
images were captured on a CCD camera at different magnifications and
rendered with Adobe Photoshop (Abacus Concepts, Berkeley, CA).
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
We had previously shown that BAP-conjugated glycans presented in
multivalent arrays on streptavidin can evoke an IgG serum immune
response in mice (5, 6). To generate Abs directed against
the novel carboxylate-associated negative charge, a mixture of anionic
hydrazine-released bovine lung glycans was coupled to BAP (D. K.
Toomre and A. Varki, unpublished observations, details to be reported
elsewhere). To enrich for coupled glycans bearing the novel
carboxylate, we applied multiple exoglycosidase treatment to molecules
with more than two negative charges that had been obtained by anion
exchange chromatography on DEAE-HPLC. The exoglycosidases chosen (see
Materials and Methods) were highly specific and did not
affect the negative charge of the carboxylate (data not shown).
Following the mixed exoglycosidase treatment, we collected the
fractions that shifted to an elution position on DEAE-HPLC that is
typical for glycans with a single negative charge (Fig. 1
A, pool II). Because
sialidase was included in the digestions, the mixture was expected to
be enriched for molecules carrying one copy each of the novel
carboxylate modification. Indeed, this material eluted from reverse
phase HPLC (Fig. 1B) in the general region expected for
BAP-coupled N-glycans with one or two negative charges, such
as mono- or bisialylated biantennary N-glycans. This mixture
of BAP-coupled glycans was bound to streptavidin to form multivalent
pseudoneoglycoproteins. Injection of these complexes into mice elicited
a serum IgG response that was at least partly directed against the
oligosaccharides, as detected by reactivity in ELISA with BSA-coupled
bovine lung glycopeptides or whole bovine lung homogenates (data not
shown).
Monoclonal Abs that recognize the novel carboxylated epitope on bovine lung glycans
We screened specifically for IgG secreting hybridomas that reacted with total bovine lung glycopeptides coupled to BSA. Because the original immunogen contained only bovine glycans and no peptides, this strategy selectively detected only Abs directed against the oligosaccharides, while avoiding detection of Abs directed against streptavidin or BAP. Early on, we noted that some of the Abs reacted differentially depending on whether the original coupling of the glycopeptides to BSA was performed with glutaraldehyde (which reacts with amino groups) or carbodiimide (which reacts with carboxyl groups). We reasoned that the decreased reactivity with carbodiimide-coupled glycopeptides might result from a carbodiimide-induced modification of the novel carboxylate on the glycans. In fact, we found that binding of four of the most reactive Abs was also substantially abrogated by direct methylamidation of the carboxylates on the target glycopeptides (data not shown). Of these, we chose to focus on mAb GB3.1, of the IgG2b subclass, because it gave the best reactivity in blots and ELISAs.
Characterization of mAb GB3.1 specificity
mAb GB3.1 binding to immobilized BSA neoglycoproteins could be
blocked by asialo-COO--glycopeptides in
solution, but not by asialo-CONHMe-glycopeptides (Fig. 2A). We further checked the
consequences of protonation of negative charges on binding by studying
reactivity at different pH values. As shown in Fig. 2B,
there was marked loss of reactivity below pH 6.0, whereas the
reactivities of two control mAbs against their respective Ags were
unaffected. Therefore, the pH dependence of mAb GB3.1 reactivity
suggests either that the target carboxyl group has a pKa of 5.8, or
that the Ab binding site has a critical residue with a pKa in this
range. Regardless, we pursued the possibility that the Ab reactivity
might be blocked by other carboxylic acids. Indeed, as shown in Fig. 2C, reactivity is inhibited by low millimolar concentrations
of several aliphatic carboxylate-containing compounds, such as acetate,
succinate, citrate, isocitrate, and EDTA, roughly in proportion to the
number of carboxyl groups on each (all compounds were tested at pH 7.5
where the respective carboxyl groups are ionized). In contrast, similar
concentrations of glucuronic and galacturonic acids (Fig. 2C) or sialic, lactic, or pyruvic acids (data not shown) did
not inhibit binding. Overall, it appears that molecules carrying
multiple aliphatic carboxylate groups were the most effective
inhibitors.
|
In other studies we had noted that a significant part of the sialidase-resistant negative charge in the bovine lung glycans can be explained by sulfate esters (D. K. Toomre and A. Varki, unpublished observations). These would not be neutralized by the methylamidation procedure. However, to verify that mAb GB3.1 did not cross-react with sulfated glycans, Western blots were conducted on thyroglobulin, which carries terminal Gal-3-sulfate and internal GlcNAc-6-sulfate, and on N-CAM, which is known to express the HNK-1 epitope carrying a terminal glucuronic acid-3-sulfate. mAb GB3.1 did not react with either of these glycoproteins. Chondrosamine, chondroitin sulfate (tetramer and octamer), and hyaluronic acid (tetramer and octamer) at 1–5 mM also do not inhibit binding, suggesting that glycosaminoglycan-like epitopes do not cross-react. Reactivity in the ELISA was unaffected by sulfated or phosphorylated sugars such as Glc-1-phosphate, Glc-6-phosphate, Man-6-phosphate, Gal-6-phosphate, GlcNAc-1-and GlcNAc-6-phosphates, Glc-6-sulfate, and Gal-6-sulfate (when tested at or below 5 mM; data not shown).
A major fraction of the mAb GB3.1 reactive-epitope is carried on N-linked glycans, including PNGase F-resistant ones
Bovine lung papain-released glycopeptides were digested with
PNGase F or PNGase A and tested for their reactivity in an ELISA
against mAb GB3.1 before and after digestion. Intact glycoproteins are
poor substrates for PNGase A, but both enzymes can act well on
glycopeptides released by thermolysin, pepsin, trypsin, or papain
(8). The deglycosylated peptides showed partial loss of Ab
reactivity compared with undigested glycopeptides (incubated in the
absence of enzymes). This loss of activity was greater with PNGase A
digestion than with PNGase F (Fig. 3).
This may be because some of the carboxylate epitopes are present on
PNGase F-resistant oligosaccharides carrying unusual core modifications
(G. Srikrishna and H. Freeze, unpublished observations).
|
Western blots showed that several proteins from bovine lung and
other bovine tissues reacted with mAb GB3.1. Reactivity was abrogated
in the presence of lung
asialo-COO--glycopeptides, but not by
asialo-CONHMe-glycopeptides (data not shown). There was also widespread
reactivity in several human tissues (Fig. 4, top). Binding to a majority
of bands, except for the 38-/40-kDa doublet bands in muscle tissue, was
blocked by asialo-COO--glycopeptides (Fig. 4, bottom). The somewhat lower staining seen with human lung
compared with bovine lung was reproducible and may be due to relative
species difference. Western blots of a variety of mouse and rat tissues
were also positive, with almost all binding being blocked by
asialo-COO--glycopeptides (data not shown). This
apparently widespread distribution of reactive epitopes is actually
explained by the fact that they are concentrated on endothelial cells,
which are, of course, present in all tissues (see below).
|
In frozen sections of bovine lung, the modification was localized
mainly to the vascular endothelium, showing that the novel
glycopeptides we had been studying from whole bovine lung had primarily
originated from the endothelium. Again, binding was almost completely
blocked by asialo-COO--glycopeptides (data not
shown), confirming the specificity of the interactions. A more
extensive survey of various human tissues showed that Ab reactivity was
again predominantly localized to the vascular endothelium of most blood
vessels (see examples in Fig. 5). In the
spleen and placenta almost every blood vessel was stained. At higher
magnification of the skin vessels, it was evident that the endothelial
cells were specifically stained (data not shown). The only positive
nonendothelial cells were some lung macrophages. Staining of cultured
human and bovine endothelial cells with Cy3-labeled mAb GB3.1 showed
predominant localization of the epitope on the cell surface (see
example in Fig. 6).
|
|
Because most of the Ab reactivity was localized to endothelial
cells, we reasoned that cognate lectins for these molecules might be
found on peripheral blood leukocytes. Therefore, we probed freshly
isolated human leukocyte populations (neutrophils, monocytes, and
lymphocytes) with the BSA-coupled
asialo-COO--glycopeptides (to enhance
multivalency, which is frequently required for lectin-carbohydrate
interactions). Desialylated glycopeptides were used to eliminate any
sialic acid-dependent binding. We found a strong expression of cell
surface binding sites for the novel glycans on the entire population of
activated, but not resting, neutrophils within 15 min of activation
with PAF or PMA/ionomycin (Fig. 7). The
specificity of this binding was proven by selective binding of
BSA-coupled asialo-COO--glycopeptides, but
>80% reduction in the binding of BSA-coupled
asialo-CONHMe-glycopeptides (Fig. 7). The residual binding of
neutralized glycopeptides is probably explained by the fact that
carboxylate neutralization by methylamidation does not achieve 100%
completion even after two rounds of treatments. In addition to these
surface binding sites, secretions elicited upon activation with other
physiologically relevant inflammatory mediators such as TNF- or fMLP
contained proteins that bound to the novel glycans in a
carboxylate-dependent manner (data not shown). Also, induction of
binding sites on monocytes was observed under the same conditions,
albeit to a lesser extent (data not shown).
|
Because a mouse mAb had to be used to detect the BSA to which the
glycopeptides were coupled, we could not test the effects of the mouse
mAb GB3.1 in the above flow cytometric assay. To address this issue and
to obtain further evidence for specific binding, we incubated freshly
isolated neutrophils with BSA-conjugated
asialo-COO--glycopeptides coated on plastic. As
shown in Fig. 8, activated human
neutrophils bound to the carboxylated glycans, while minimal binding
was observed to BSA itself. Adhesion was completely inhibited by mAb
GB3.1 at various concentrations, whereas a control
anti-carbohydrate Ab did not have any effect. Binding was also
progressively and completely blocked by adding soluble bovine lung
asialo-COO--glycopeptides, but not by
asialo-CONHMe-glycopeptides.
|
To explore the potential role of the novel glycans in neutrophil adherence to vascular endothelium, we measured neutrophil adhesion to HUVECs in the presence of various modulators. An isotype-specific control Ab did not have any effect. In contrast, increasing amounts of mAb GB3.1 enhanced adhesion by 2- to 3-fold (data not shown). This effect was seen regardless of whether the endothelial cells were prestimulated or the endothelial cells were incubated with the Ab before the neutrophils were allowed to adhere. Adhesion was also increased in the presence of asialo-COO--glycopeptides, but not by asialo-CONHMe-glycopeptides. The reasons for the diametrically opposite effects of mAb GB3.1 and the soluble asialo-COO--glycopeptides on adherence of neutrophils to plastic-immobilized glycans vs endothelial cells are unclear. Presumably, the situation involving intact endothelial cells is more biologically complex and involves other signaling circuits and/or other adhesion pathways. For example, the interaction of activated neutrophils with the endothelial glycans could trigger cis or trans signaling in either or both of these cells, eventually resulting in inhibition of the overall interactions. Regardless, the results clearly indicate that these novel glycans on endothelial cells can modulate interactions with activated neutrophils.
Monoclonal Ab GB3.1 inhibits leukocyte influx in a mouse model of acute peritonitis by reducing transmigration of cells across the endothelium
Based on the above results and to further understand the in vivo
role of the novel glycans in the dynamic interaction between
endothelial cells and leukocytes, we checked the effects of mAb GB3.1
on the pathophysiology of acute inflammation. Because the novel glycan
is also expressed in murine tissues, we reasoned that the inflamed
mouse peritoneum would be a convenient model to quantitatively examine
such effects. Zymosan-induced peritonitis in mice was characterized by
a time-dependent accumulation of cells in the peritoneal cavity.
Neutrophil content in peritoneal cavities rose from about 0.75 million
cells in saline-treated mice to about 12 million cells within 4 h
after zymosan challenge. Neutrophil influx was followed by monocyte
infiltration, which rose from 10 million cells in control mice to
50 million cells 16 h after zymosan treatment. Intravenous
injection of mAb GB3.1 immediately before the induction of peritonitis
resulted in a dose-dependent reduction in the extent of neutrophil and
monocyte accumulation, while an isotype-matched mouse IgG or mAbAD7.5
or an unrelated anti-carbohydrate Ab (9) had minimal
effect (Fig. 9). Titers of mAb GB3.1
checked at various time points after injection in control mice showed
that Ab levels stabilized by 2 h and remained steady for up to
24 h in the circulation (data not shown). The dose of 20 µg/g
body weight, which gave about 65% reduction in infiltration was chosen
to further confirm the patterns of cells in the inflamed peritoneum by
flow cytometry. Peritoneal cells from untreated and zymosan-treated
mice were stained with FITC-labeled anti-mouse Gr-1
(anti-neutrophil) or PE-labeled anti-mouse Mac-1 (monocytes and
neutrophils) and analyzed by flow cytometry. Four hours after zymosan
injection, Gr-1high neutrophils were the most
abundant in the peritoneum, followed by recruitment of
Gr-1low/Mac-1high monocytes
at 16 h (data not shown). Again, injection of mAb GB3.1 caused a
substantial reduction in appearance of both types of cells, while a
control Ab had no effect (data not shown).
|
, showing the 16 h point).
Second, this was unaffected by injection of a control Ab. Third,
increased leukocyte adherence to the mesenteric microvasculature could
also be detected in saline- or control mAb-treated mice; however, only
mAb GB3.1-treated mice showed a clear reduction in leukocyte
recruitment into the inflamed tissues. This effect of the mAb was
associated with marked increase in adherence of cells to the
endothelium in over half of randomly examined mesenteric vessels. This
adherence was more dramatic at the 16 h point compared with the
4 h point. Thus, the reduction in cell efflux into the peritoneal
cavity can be explained by a reduction in transmigration across the
vessel walls. These findings suggest that the novel glycans can
selectively modulate adhesion and transmigration of activated
leukocytes across the vascular endothelium in the setting of acute
inflammation.
|
2 integrins on the
neutrophils are largely responsible for stronger adhesive interactions
with intercellular adhesion molecules (ICAM-1 and ICAM2). ICAM-1 is
constitutively expressed at low levels by endothelial cells, but is
rapidly up-regulated during inflammation (10, 11).
Although rolling and firm adhesion of leukocytes are well understood,
there is much less information regarding the in vivo mechanisms
mediating transmigration. Several in vitro and in vivo studies have
shown that PECAM-1 (CD31) is critically involved in
transendothelial migration (15). Interestingly, Abs
against PECAM-1 inhibit leukocyte extravasation from mesenteric vessels
(16, 17), similar to the effects of mAb GB3.1 seen in our
studies. However, the expression patterns of CD31 on the surface of
most leukocytes, platelets, and intercellular junctions of endothelial
cells (15) is different from that of the GB3.1 Ags.
Collectively, our findings indicate the carboxylated glycans
participate in yet another set of novel mechanisms involved in the
regulation of acute inflammation. Conclusions and perspectives
Here we have used a novel approach to generate an IgG mAb that detects unusual carboxylated N-glycans that are predominantly localized in the vascular endothelium of a variety of mammalian tissues. The vascular endothelium is a dynamic and complex system that possesses many secretory, synthetic, immunologic, and metabolic functions (18). The contribution of carbohydrates to endothelial function is best exemplified by the selectin family of adhesion molecules, which recognize sialyl Lewisx- and sialyl Lewisa-containing structures (12, 13, 14). Also, unusual anionic oligosaccharides such as sialyl Lewisx/a as well as sulfosialyl Lewisx/a, Man-6-phosphate (19), polysialic acid (20), and GalNAc-4-sulfate (21) have been noted to be more involved in mediating specific biological actions than the more common neutral glycans. Therefore, the predominant localization of these novel carboxylated glycans on the vascular endothelium raised the intriguing possibility that they could participate in endothelial functions or in interactions with cells or proteins in the blood. The functional studies described here clearly indicate that they can mediate interactions with activated neutrophils and modulate inflammatory responses. Further studies of the structure and biosynthesis of the carboxylated glycans and the identification of their cognate receptors on activated neutrophils are now underway.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: European Molecular Biology Laboratory, Meyerhofstrasse 1, D-69117 Heidelberg, Germany.
3 Current address: Nextran, Inc., A Division of Baxter Healthcare, San Diego, CA 92093.
4 Address correspondence and reprint requests to Dr. Ajit Varki, CMM East, Room 1065, University of California at San Diego School of Medicine, La Jolla, CA 92093.
5 Abbreviations used in this paper: BAP, biotinylated diaminopyridine; CPAE, calf pulmonary artery endothelial; EDC, 1-ethyl-3(3-dimethylaminopropyl)carbodiimide; NHS, N-hydroxysuccinimide; PAF, platelet-activating factor; PNGase, peptide-N-glycosidase; PECAM, platelet-endothelial cell adhesion molecule.
Received for publication July 28, 2000. Accepted for publication October 10, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
(1–6)dextran. J. Immunol. 142:863.[Abstract]
(1 replaced by 6) dextran. J. Immunol. 128:1350.[Abstract]
1,6GlcNAc in the core of N-linked glycans: localized expression of core-fucosylated glycoconjugates in human tissues. J. Biol. Chem. 272:25743.This article has been cited by other articles:
|
|
 |
|
  P. Sinha, C. Okoro, D. Foell, H. H. Freeze, S. Ostrand-Rosenberg, and G. Srikrishna Proinflammatory S100 Proteins Regulate the Accumulation of Myeloid-Derived Suppressor Cells J. Immunol., October 1, 2008; 181(7): 4666 - 4675. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Turovskaya, D. Foell, P. Sinha, T. Vogl, R. Newlin, J. Nayak, M. Nguyen, A. Olsson, P. P. Nawroth, A. Bierhaus, et al. RAGE, carboxylated glycans and S100A8/A9 play essential roles in colitis-associated carcinogenesis Carcinogenesis, October 1, 2008; 29(10): 2035 - 2043. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Srikrishna, O. Turovskaya, R. Shaikh, R. Newlin, D. Foell, S. Murch, M. Kronenberg, and H. H. Freeze Carboxylated Glycans Mediate Colitis through Activation of NF-{kappa}B J. Immunol., October 15, 2005; 175(8): 5412 - 5422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Sonnenburg, H. van Halbeek, and A. Varki Characterization of the Acid Stability of Glycosidically Linked Neuraminic Acid. USE IN DETECTING DE-N-ACETYL-GANGLIOSIDES IN HUMAN MELANOMA J. Biol. Chem., May 10, 2002; 277(20): 17502 - 17510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Srikrishna, K. Panneerselvam, V. Westphal, V. Abraham, A. Varki, and H. H. Freeze Two Proteins Modulating Transendothelial Migration of Leukocytes Recognize Novel Carboxylated Glycans on Endothelial Cells J. Immunol., April 1, 2001; 166(7): 4678 - 4688. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
