| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Medicine, Section of Arthritis and Connective Tissue Diseases, and
Department of Cellular and Molecular Biology, Northwestern University Medical School, Chicago, IL 60611;
Veterans Administration Chicago Health Care System, Lakeside Division, Chicago, IL 60611;
§
Bayer Corp., West Haven, CT 06516;
¶
Department of Laboratory Medicine, Sahlgrenska Hospital, Goteborg, Sweden; and
||
Laboratory of Molecular Pathogenesis, University of Michigan Dental School, Ann Arbor, MI 48108
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The structural elements of carbohydrate moieties are built by
sequential addition of saccharide units on the luminal side of the
Golgi apparatus by specific glycosyltransferases. Blood group Ags can
be divided into type 1 and type 2 chain blood group ABH Ags, Lewis Ags,
and X/Y glycolipid Ags, each distinguished by characteristic
carbohydrate sequence, binding position, and linkage anomericity
(6). Lewis determinants are structurally related to
determinants of the ABH blood group system and are comprised of two
major Ags, Lea and Leb.
Lewis and related Ags may occur 1) as free oligosaccharides, a typical
occurrence in milk and urine; 2) protein bound by an
O-linkage via an N-acetylgalactosamine to a
serine or threonine residue; or 3) as glycolipids located mainly in the
outer leaflet of the plasma membrane (6, 7). A major site
of Lewis plasma glycolipid synthesis is the digestive tract. Here,
mucosal cells shed these Ags into the intestinal lumen, where they are
digested, reabsorbed, and transported to the plasma. Circulating Lewis
glycolipids in plasma can then be acquired by red cells, lymphocytes,
and platelets (7). Transfusion laboratories focus on
Lea and Leb as the only two
components of the Lewis blood group system; however, fucosylation of
these Ags produces Lewis isomers, which include
Ley. Numerous distinct
-(1, 3)-fucosyltransferase genes have been cloned and characterized
that can perform this function (8). Although saccharide
determinants that result from fucosyltransferase action do not appear
significant in transfusion medicine, they may play a key role in
lymphocyte traffic and inflammatory responses (9).
We have shown that the soluble form of E-selectin mediates angiogenesis
via its endothelial ligand, sialyl Lex (10, 11). To further examine the EC function in inflammation, we
raised a mAb, termed mAb 4A11, to adherent human RA synovial tissue
(ST) cells (12). 4A11 recognized the glycoconjugate
Lewisy-6/H-5-2 (Ley/H),
which is expressed mainly in inflamed and malignant disease states.
These two sugars are structurally related, in that
Ley is the result of fucose addition, by
-(1, 3)-fucosyltransferase, to the nonterminal
N-acetylglucosamine of H-5-2. To date, the species with
which Ley/H is coupled is not fully elucidated.
mAb 4A11 does not react with myeloid or lymphoid cell lines, platelets,
macrophages, lymphocytes, or fibroblasts (13). mAb 4A11
does, however, react with mature cultured human abdominal aortic ECs
and, as we show here, with a subset of human dermal microvascular ECs
(HMVECs). This mAb recognized the glycoconjugate
Lewisy-6/H-5-2 (Ley/H),
which shares some structural overlap with sialyl
Lex. Ley/H is a selective
Ag that is expressed on ECs only in synovium, lymphoid tissues, skin,
and thymus as well as on epithelial cells such as keratinocytes
(12, 14). EC Ley/H expression is
up-regulated before the ingress of inflammatory cells in vivo using a
cytokine-dependent model of human poison ivy extract-induced contact
dermatitis (12). The biological role of glycoconjugates,
such as Ley/H, which constitute the chemical
basis for blood groups, has to date been poorly defined
(6). Here we show a role for glycoconjugates as
novel mediators of angiogenesis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Anti-HLA class I was obtained from Biodesign (Kennebunk, ME).
Sulfasalazine (SASP), TNF-, and IL-1ß were obtained from
Pharmacia-Upjohn (Kalamazoo, MI). Pepsin, trypsin, chymotrypsin,
protease V8, proteinase K, ß-galactosidase, neuraminidase,
mannosidase, anti-mouse IgG, LPS, PMA, calcium ionophore A23187
(CIA), thrombin, histamine, BSA, hydrocortisone, tetramethylbenzidine,
2'-fucosyllactose (H-2g, the H analogue), anti-mouse IgM peroxidase
conjugate, and anti-mouse IgG peroxidase conjugate were obtained
from Sigma (St. Louis, MO). The 2'-fucosyllactose-BSA (H-2g-BSA)
conjugate was purchased from V-Labs (Covington, LA). IL-4 was obtained
from PeproTech (Rocky Hill, NJ). Endothelial growth medium, endothelial
basal medium (EBM), and HMVECs were obtained from Clonetics (San
Diego, CA). Mouse IgG and mouse IgM were purchased from Coulter
(Hialeah, FL). Mouse anti-human von Willebrand factor (vWF) was
obtained from PharMingen (San Diego, CA). Basic fibroblast growth
factor (bFGF) was obtained from R&D Systems (Minneapolis, MN).
Indomethacin was obtained from Cayman Chemical (Ann Arbor, MI). The
vector ABC system was purchased from Vector (Burlingame, CA).
Diaminobenzidine was obtained from Kirkegaard & Perry (Gaithersburg,
MD). Diff-Quik was purchased from Baxter (North Chicago, IL).
Lactodifucotetraose (Leyg; the
Ley analogue) and H-disaccharide (HD; a short
analogue of both Ley and H) were obtained from
Accurate Chemical (Westbury, NY). Hydron was obtained from
Interferon Sciences (New Brunswick, NJ).
1-(5-Isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H7) was
obtained from Seikagaku (Ijamsville, MD).
HMVEC stimulation for cell surface expression of Ley/H
HMVECs used in this study were between passage 5 and 11 and did not display discernable phenotypic changes when observed before each assay. Initial studies used a homogeneous population of Ley/H-expressing HMVECs sorted by flow cytometry, to optimize the sensitivity of the cell-based ELISA. Cells were incubated with or without immune modulators in RPMI plus 10% FBS for various times. Cells were fixed with 3.7% formalin in PBS (100 µl/well) for 15 min. The plates were washed with PBS. mAb 4A11, mouse IgM, or anti-HLA class I (10 µg/ml) diluted in wash buffer (PBS and 0.1% BSA) was added to each well (100 µl/well) and incubated for 2 h. Plates were washed and blocked with 3% BSA in PBS (200 µl/well) for 1 h. Plates were washed and urease-conjugated anti-mouse IgG diluted 1/1000 in wash buffer was added to each well for 1 h. Plates were washed with distilled water, combined with bromocresol purple substrate (50 µl/well), and absorbance was read at 570 nm. Origins and concentrations of stimulants are indicated in the reagents section and figure legends, respectively.
After optimizing cell-based ELISA conditions, nonsorted HMVECs were plated in 96-well plates (5000 cells/well) and were grown to confluence in endothelial growth medium. Once confluent, the medium was changed to RPMI, 10% FBS, and 2 mM glutamine for 48 h. Cells were then incubated in the absence or the presence of stimuli for 20 min, medium was removed, and the cells were immediately fixed with 3.7% formalin in PBS (100 µl/well) at 25°C for 15 min. The plates were washed with PBS and 0.05% Tween 20. Primary Ab (10 µg/ml) diluted in PBS and 0.2% BSA was added to each well (100 µl/well) and incubated for 2 h at 37°C. Primary Abs were: mAb 4A11, mouse anti-human vWF, mouse IgM (control), or mouse IgG (control). After washing, anti-mouse IgM peroxidase conjugate or anti-mouse IgG peroxidase conjugate was diluted in block buffer (100 µl/well) and incubated for 1 h at 37°C. Following another wash, tetramethylbenzidine substrate was added to each well, and the reaction was terminated with 1 N H2SO4 (50 µl/well). Absorbance was read at 450 nm for quadruplicate samples. ELISA standards included either a mucin standard or an H-2g-BSA conjugate (V-Labs). The H-2g-BSA protein carries an average of seven moieties of 2'-fucosyllactose per protein (ranging from 4–10 carbohydrate structures) as determined by laser desorption time of flight mass spectrometry.
Demonstration of Ley/H protease sensitivity
Human RA synovial tissue was digested with dispase, collagenase, and DNase for 2 h at 37°C (4). After filtration with a 40-µm pore size nylon cell strainer, cells were resuspended in RPMI and 10% FBS and plated in polylysine-coated 96-well plates (1.25 x 105 cells/well). Cells were spun down using microplate carriers and fixed with 0.25% glutaraldehyde or by air-drying at 37°C for 2 h. Cells were incubated with 0.1% BSA overnight and then treated with the following proteases for 2 h at 37°C: pepsin (280 U/ml), trypsin (308 U/ml), chymotrypsin (4.5 U/ml), protease V8 (100 U/ml), proteinase K (1.5 U/ml), ß-galactosidase (100 U/ml), neuraminidase (2 U/ml), or mannosidase (3.6 U/ml). After washing (PBS and 0.1% Tween 20), plates were back-coated with 0.1% BSA for 2 h at 37°C followed by a 1-h incubation at 37°C with mAb 4A11 (10 µg/ml). After additional washing, anti-mouse immunoperoxidase-conjugated Ab was incubated for 1 h at 37°C. After washing, and substrate addition, sulfuric acid was used to stop color development, and the plate was read at 490 nm. Each sample was assayed six times, and results are presented as the percent decrease in comparison with nonenzymatically treated tissue cells.
Detection of Ley/H by immunohistochemistry in ST
ST sections were fixed in acetone for 10 min at 4°C and rehydrated in PBS. mAb 4A11 (10 µg/ml) was used as the primary Ab, with biotinylated anti-mouse IgM as a secondary Ab. A Vector ABC system containing avidin-HRP was used with diaminobenzidine as a substrate. Slides were counterstained with Gill’s hematoxylin and Li2CO3 for bluing and were dehydrated in ethanol.
Detection of Ley/H by immunofluorescence in HMVECs
HMVECs were fixed onto gelatin-coated glass coverslips by 20-min incubation in ice-cold acetone. Coverslips were blocked with 5% goat serum and incubated with mAb 4A11, rabbit anti-human vWF (IgG; Dako, Carpinteria, CA), or rabbit anti-human P-selectin (IgG; PharMingen, San Diego, CA) overnight at 4°C. Isotypic control Abs (Coulter, Hialeah, FL; and Jackson ImmunoResearch Laboratories, West Grove, PA) were used to determine binding specificity. After washing, cells were incubated with rhodamine-conjugated goat anti-mouse IgM or fluorescein-conjugated goat anti-rabbit IgG secondary Abs (10 µg/ml; Jackson ImmunoResearch Laboratories and Kirkegaard & Perry) for 45 min at room temperature in the dark. After washing, coverslips were mounted with immunofluorescent mounting medium (Kirkegaard & Perry). A Hoechst stain (Molecular Probes, Eugene, OR) was used to show cell nuclei (blue). Images were taken with a Nikon ES 400 microscope (Nikon, Garden City, NY). The percentage of HMVECs positive for Ley/H was determined by counting images showing rhodamine positivity compared with nonstained cells (nuclear staining only). Random fields were counted until total cell counts were >350.
Detection of soluble Ley/H in synovial fluid (SF) by ELISA
ELISA plates (96-well) were coated with 100 µl/well mAb 4A11 (25 µg/ml in 50 mM H3BO3 and 120 mM NaCl, pH 8.6) for 2 h at 37°C and washed in PBS (pH 7.5) and 0.05% Tween 20. Nonspecific binding was blocked with 200 µl/well 2% BSA in PBS (block buffer), for 90 min at 37°C. After washing, specimens or standards (mucin-containing Ley/H, Sigma) were added in duplicate, and plates were incubated for 3 h at 37°C. After washing, 100 µl/well of a biotinylated form of the same 4A11 mAb (25 µg/ml in block buffer) was added for 1 h at 37°C, and plates were washed. After adding peroxidase-conjugated avidin, the plates were incubated for 30 min at 37°C, and washed. Orthophenylenediamine dichloride substrate was added for a 6- to 8-min incubation, the reaction was terminated with 1 M H2SO4 and read at 490 nm.
HMVEC chemotaxis induced by glucose analogues of Ley/H and purified H-5-2 and Ley glycolipids
Subconfluent HMVECs were fed the night before the assay. HMVECs (3.75 x 104 cells/25 µl EBM and 0.1% FBS) were placed in the bottom wells of 48-blind well chemotaxis chambers (NeuroProbe, Cabin John, MD) with gelatin-coated polycarbonate membranes (8 µm pore size; Nucleopore, Pleasant, CA). Inverted chambers were incubated at 37°C for 2 h. Test substances, PBS, or positive control bFGF (60 nM) in PBS was added to the bottom wells, and the chambers were incubated for 2 h at 37°C. The membranes were removed, fixed in methanol, and stained with Diff-Quik. Test substances include Leyg, H-2g, and HD. HD was chemically synthesized and was >95% pure. Leyg and H-2g were purified from pooled human milk and were 95 and 99.1% pure, respectively. Three high power microscope fields were counted in each well. Checkerboard analyses were performed in a similar manner as chemotaxis assays, except that the concentrations of Leyg and H-2g were varied in the upper and lower chambers.
Immunodepletion of the glucose analogues with mAb 4A11 or mouse IgM control was accomplished by preincubation for 1 h at 37°C. Glucose analogues were immunodepleted at concentrations of 1 nM and 100 µM with 50 µg/ml of Ab. Upon completion of this neutralization period, the glucose analogue/Ab combination was assayed in the HMVEC chemotaxis assay.
In addition to performing chemotaxis with the glucose analogues of Ley and H structures, we performed HMVEC chemotaxis with H-5-2 and Lewisy-6 purified from dog small intestine (15). Purified substances were dissolved in ethanol and diluted to a working concentration of 0.1% ethanol. All PBS and bFGF controls were likewise analyzed in 0.1% ethanol. HMVECs were >99% viable in 0.1% ethanol by trypan blue exclusion.
Leyg- and H-2g-induced HMVEC proliferation
HMVEC proliferation in response to Leyg, H-2g, and HD was determined using modifications of a procedure used previously (11). Briefly, 2.5 x 103 cells were allowed to adhere to 96-well plates for 2–4 h. Serial dilutions of glucose analogues or bFGF (positive control) were prepared, ranging from 0.1 pM to 100 µM in EBM and 2% FBS in a 100-µl volume. Quadruplicate samples were incubated at 37°C for 72 h. Relative cell numbers were determined using a Cell Titer 96 AQueous cell proliferation assay kit according to the manufacturer’s protocol (Promega, Madison, WI). Absorbance was read at 490 nm on an ELISA plate reader.
Leyg- and H-2g-induced angiogenesis in vivo
In vivo corneal bioassays were performed as we have previously described (5). Test substances were combined 1:1 with Hydron and implanted into the normally avascular corneal stroma of the rat several mm from the limbus (11). Corneas were perfused with colloidal carbon after 7 days to provide a permanent record of the angiogenic response. Neutralization of glucose analogue-induced angiogenesis was performed by preincubating the glucose analogue with mAb 4A11 or IgM control for 1 h at 37°C. The glucose analogue/Ab combination was added to the Hydron pellet at a 1:1 ratio. H-2g or Leyg was analyzed at a final concentration of 50 µM in combination with mAb 4A11 or control IgM at 25 µg/ml. As a control, bFGF was likewise incubated with 25 µg/ml mAb 4A11.
Statistical analysis
Statistical analysis was performed using one-way ANOVA followed by independent Student’s t test; p < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cytokines may induce EC molecules, such as E-selectin, allowing
the vasculature to play a central role in inflammation and angiogenesis
(16). We found that 42% of HMVECs were
Ley/H immunopositive by flow cytometry. To
determine whether Ley/H was inducible by
cytokines or other immune modulators,
Ley/H-positive HMVECs were treated with various
stimuli for 5, 20, or 120 min. Ley/H was rapidly
up-regulated on HMVECs by several immune modulators (Fig. 1
). Compared with nonstimulated cells,
TNF-, LPS, and PMA significantly augmented
Ley/H cell surface expression on the HMVECs after
only 5 min (Fig. 1A). After 20 min, all the stimuli
significantly induced Ley/H cell surface
expression compared with nonstimulated cells. CIA showed the greatest
induction, increasing Ley/H expression >2-fold
after 20 min of incubation (Fig. 1B). TNF-, LPS, and
IL-1ß were also potent inducers of Ley/H; each
showed about a 2-fold increase above nonstimulated cells. When the
assays were repeated at 2 h (Fig. 1C) and 4 h
(data not shown), Ley/H had returned to baseline
nonstimulated values. These results indicate that
Ley/H is a rapidly cytokine-inducible
Ag.
|
). Ley/H was
induced by CIA, LPS, PMA, histamine, thrombin, IL-1ß, IL-4, TNF-,
and bFGF. Ley/H expression was not increased
nonspecifically by indomethacin, H7, BSA, hydrocortisone, or
SASP.
|
were more potent inducers of HMVEC
Ley/H expression than of vWF expression. Storage of Ley/H in ECs
The rapid mobilization of Ley/H to the
surface of HMVECs suggested that Ley/H was
synthesized and stored ready for release upon stimulation. To
investigate this possibility, we used immunofluorescence, which
suggested mainly perinuclear vesicular storage of
Ley/H in nonstimulated HMVECs (Fig. 3A; red =
Ley/H immunopositivity, blue = Hoechst
nuclear staining). Upon stimulation with IL-1ß for 20 min,
Ley/H localization appeared more uniformly
distributed, suggesting cell membrane localization (Fig. 3B). Dual staining (Fig. 3C) demonstrated some
overlap (yellow) storage of Ley/H (red) and vWF
(green), as well as storage that appeared independent in the same
cells. A similar colocalization pattern was found for
Ley/H and P-selectin (data not shown). Negative
control stains using an irrelevant IgM (Fig. 3D) or the
combination of irrelevant IgM and IgG Abs displayed no prominent
staining (data not shown).
|
To determine whether Ley/H was attached to a
protein moiety, we tested protease sensitivity. RA synovial tissue was
enzymatically digested, and resultant cells were fixed and analyzed by
cell-based ELISA with or without protease treatment. Trypsin,
chymotrypsin, and protease V8 treatment reduced detectable levels of
cell surface Ley/H by 39, 17, and 52%,
respectively. In contrast, pepsin, proteinase K, ß-galactosidase,
neuraminidase, and mannosidase treatment had no effect on cell surface
Ley/H levels (Fig. 4, representative of three independent
experiments). These results suggest that some portion of
Ley/H is protein bound. These experiments cannot
definitively determine whether partial inhibition is the result of an
additional nonprotein Ley/H species or whether
incomplete protein digestion occurred.
|
We have shown that the soluble adhesion molecule E-selectin
mediates angiogenesis via its EC ligand sialyl
Lex (11). Because sialyl
Lex and Ley/H contain some
structural overlap, we hypothesized that Ley/H
may also be secreted in soluble form and mediate angiogenesis
(11). Using a sandwich ELISA, RA SF soluble
Ley/H was significantly increased compared with
noninflammatory OA SF (Fig. 5A; p <
0.05). Because mAb 4A11 was used for both capture and detection, this
ELISA detected molecules containing multiple carbohydrate antigenic
determinants.
|
B). Fig. 5C represents intense staining of
RA ST with mAb 4A11, while NL tissue (Fig. 5D) was typically
less immunopositive. These results indicate that the
Ley/H Ag is up-regulated in the angiogenic RA ST
and SF, suggesting that it could play a role in neovascularization.
Levels of soluble Ley/H contained in RA SF were
compared with those required to induce HMVEC chemotaxis. Mean levels of
soluble Ley/H in RA SF were equivalent to 5.2 pM
H-2g-BSA, as determined using the H-2g-BSA standard in conjunction with
the mAb 4A11 ELISA. Comparing to a standard of H-2g-BSA, RA SF contains
36 pM soluble Ley/H. Purified glycolipids
H-5-2 and Ley-6 can significantly chemoattract
HMVECs at the concentrations of 1 pM and 1 fM, respectively,
demonstrating that the levels found in RA SF are more than adequate to
induce chemotaxis. Regulation of HMVEC soluble Ley/H
To determine whether SF soluble Ley/H could
occur via endothelial shedding and release we examined stimulation of
cultured HMVECs and subsequent release of soluble
Ley/H using an ELISA (Fig. 6). Soluble Ley/H
was assayed in HMVEC-conditioned medium 5, 20, and 240 min after
stimulation with or without irrelevant stimuli such as indomethacin,
compared with stimuli such as TNF-, CIA, or IL-1ß. After 5 min,
Ley/H levels were comparable with nonstimulated
values (data not shown). After 20 min, however, TNF-, CIA, and
IL-1ß stimulation increased mean soluble Ley/H
levels by 3.8-, 9.5-, and 5.4-fold, respectively, compared with
equivalently stimulated conditioned medium after 5 min (Fig. 6A). In contrast, conditioned medium from a control (human)
skin fibroblast cell line (CRL 1635, American Type Culture Collection,
Manassas, VA), not expected to produce soluble
Ley/H, were not significantly altered compared
with the nonstimulated baseline after TNF-, CIA, or IL-1ß exposure
(1.0-, 1.0-, and 1.3-fold change, respectively). Indomethacin, an
irrelevant stimulus, had no effect on soluble
Ley/H release. After 240 min (Fig. 6B), mean soluble Ley/H levels in
conditioned medium from HMVECs were increased 3.5-, 8.6-, and
7.2-fold by TNF-, CIA, and IL-1ß stimulation, respectively,
compared with equivalently stimulated HMVECs conditioned medium. In
contrast, after 240 min, skin fibroblast production of soluble
Ley/H was not altered (1.0-, 0.4-, and 1.4-fold
change) when stimulated with TNF-, CIA, or IL-1ß. The data
represent one of three experiments. These results indicate that
shedding of soluble Ley/H from ECs may be a
potential mechanism by which soluble Ley/H is
found in RA SFs.
|
To test the hypothesis that Ley/H was
directly involved in neovascularization, we examined the involvement of
Ley/H in EC chemotaxis in vitro, a facet of the
angiogenic response, using glucose analogues of
Ley and H structures (Fig. 7) (6, 17, 18).
Leyg stimulated chemotaxis of HMVECs at
concentrations from 10 pM to 1 mM, inducing migration equivalent to 60
nM control angiogenic bFGF (Fig. 8A). H-2g was likewise
chemotactic for HMVECs from 10 pM to 1 mM, peaking at 1 µM, also
similar to the response induced by bFGF (Fig. 8B). The
shortest glucose analogue, HD, was not chemotactic for HMVECs (Fig. 8C). The inability of HD, a structurally related glucose
analogue (Fig. 7), to induce HMVEC migration, demonstrates the
specificity of the chemotactic function. Further, checkerboard analysis
indicated that the effects of Leyg and H-2g
were chemotactic (inducing directed migration), not chemokinetic
(inducing random migration), for HMVECs (Fig. 9, A and B,
respectively). These results indicate that a minimal portion of the
Ley/H structure is necessary for induction of
endothelial chemotaxis. When Leyg and H-2g were
preincubated with mAb 4A11, their chemotactic activity for HMVECs
was significantly decreased (p < 0.05; Fig. 10). As a control, we induced
chemotaxis with bFGF (60 nM) that was incubated with either mAb 4A11
(10 µg/ml) or a mouse IgM control Ab (10 µg/ml; Coulter). There was
no difference when comparing mean cells migrating in response to bFGF
and mAb 4A11 vs chemotaxis induced by bFGF and IgM (data not shown;
n = 6 and n = 4, respectively),
indicating that mAb 4A11 does not inhibit bFGF-induced EC chemotaxis.
Next, we showed that purified Ley and H likewise
induced chemotaxis of HMVECs in vitro, confirming that the results
obtained using glucose analogues of Ley/H were
similar to those obtained using purified Ley and
H (Fig. 11).
|
|
|
|
|
H-2g and Leyg did not induce HMVEC proliferation when assayed at concentrations ranging from 0.1 pM to 100 µM compared with a vehicle control (n = 3; data not shown). In contrast, a positive control concentration curve using bFGF induced significantly higher HMVEC proliferation at concentrations ranging from 0.1–100 nM, peaking at 10 nM (data not shown).
Angiogenesis induced by Ley/H
To determine whether Ley/H was angiogenic in
vivo, Leyg, H-2g, and HD were assayed in the rat
cornea (5, 11, 17). Each test substance was incorporated
into slow-release Hydron pellets and implanted into rat corneas
(5, 11). H-2g induced an angiogenic response in 12 of 12
corneas (100% positive), and Leyg induced an
angiogenic response in 15 of 16 corneas (94% positive; Fig. 12, A and C, and
Table I). When H-2g was preincubated with
mAb 4A11, the corneal angiogenic response was markedly reduced (one of
seven positive corneas, or 14% positive corneal responses; Fig. 12B). When Leyg was preincubated with
mAb 4A11, the corneal angiogenic response was absent (zero of five
positive, or 0% positive corneal responses; Fig. 12D),
indicating that mAb 4A11 immunoneutralized the effects of
Leyg and H-2g and that Leyg
and H-2g are potent angiogenic agents in vivo.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ley/H expression is stimulated on HMVECs by a number of immune modulators, indicating that Ley/H is a rapidly and transiently induced molecule, responsive to both cytokines and secretagogues. This rapid mobilization of Ley/H to the surface of HMVECs suggested that Ley/H was synthesized and stored ready for release upon stimulation, much like the adhesive EC molecule, P-selectin, which is stored in the Weibel-Palade bodies. P-selectin surface expression is rapidly induced on ECs after 5–10 min of histamine, CIA, or thrombin exposure (19, 20). We demonstrate that the Ley/H Ag is rapidly cell surface expressed after similar cell stimulation. We also show that soluble Ley/H can be released from HMVECs by similar stimuli.
Further, we demonstrate the specificity of Ley/H
expression on HMVECs using various specific immune modulators and
demonstrate its induction relative to the characterized EC Ag vWF.
Although thrombin, histamine, and bFGF induce both vWF and
Ley/H expression on HMVECs, IL-4 and LPS
up-regulate only Ley/H expression. Moreover,
cytokines such as IL-1ß and TNF- are more potent inducers of HMVEC
Ley/H than vWF expression. These results indicate
that mediators that play critical roles in inflammation, such as LPS,
IL-1ß, or TNF- also amplify the expression of HMVEC
Ley/H, suggesting a role for this Ag in
inflammation.
The significantly up-regulated Ley/H in angiogenic RA ST and soluble Ley/H in RA SF suggested that this glycoconjugate could play a role in angiogenesis, because RA ST contains a persistent angiogenic influx relative to NL ST and because RA SF is inflammatory vs OA SF. In vitro chemotaxis and checkerboard assays with HMVECs provided further evidence that Ley/H might play a role in angiogenesis. This chemotactic response in vitro and the angiogenic response in vivo were neutralized with mAb 4A11, demonstrating that this effect was specific for Ley/H.
There exists a hypothesis that while endothelium is quiescent for weeks or longer, ECs must also require a mechanism of storage of preformed regulators of angiogenesis that are capable of inducing new capillary growth within hours in response to angiogenic stimuli, such as those found in a wound or an inflamed ST (21). Despite this hypothesis, with the possible exception of bFGF, examples have not been described. Indeed, immunofluorescence data indicate a vesicular storage pattern for Ley/H. Thus, the rapid cell surface expression of Ley/H may fit this paradigm. Hence, it is likely that in an RA joint Ley/H may be stored for expedient use during times of active inflammation and subsequent angiogenesis.
Glycoconjugates (glycoproteins/glycolipids) have been known for some time to constitute the chemical basis for several blood group systems in man and to act as adhesion molecules for microbial ligands, although no physiological role has been shown for them until recently (6). An mAb MIA-15-15, detecting Ley/Lex/H, inhibited the motility of tumor cells in vitro (22). Because the potential of tumor cells for invasion is closely associated with motility, which appears to depend on specific glycosylation, it followed that patients bearing lung carcinomas identified by mAb MIA-15-15 had a strikingly worse prognosis than those whose tumors were MIA-15-15 negative (22).
The importance of glycoconjugates in the induction of autoimmunity was recently underscored by the finding that Helicobacter pylori, the micro-organism involved in gastritis, ulcers, adenocarcinoma, and lymphoma of the stomach, expresses Ley/Lex/H, which is also found in gastric mucin (23, 24). Mice bearing hybridomas making H. pylori-induced anti-Le Abs developed gastritis, pointing to a mechanism by which H. pylori participates in molecular mimicry. Hence, Abs directed against H. pylori Ley result in gastritis via an autoimmune reaction directed against gastric mucin Ley. In diseases such as RA, the inciting agent is unknown. Thus, it is possible that Ley/H may also trigger a molecular mimicry immune reaction in inflammatory angiogenic sites such as RA and psoriasis.
Glycoconjugates may play a role in cell adhesion and recruitment of cells into inflammatory sites. Sialyl Lex is a ligand for the endothelial adhesion molecules P-selectin and E-selectin (25). The importance of glycoconjugates in inflammation has recently been described by Maly et al. (26), who found that mice deficient in the enzymes that generate sialyl-Lex have a partial defect in neutrophil accumulation in response to an acute inflammatory stimulus.
The role of glycoconjugates and their ligands in angiogenesis is only recently becoming clear. Nguyen and co-workers identified a role for sialyl Lex in capillary morphogenesis (27). We have shown that soluble E-selectin can act as an angiogenic mediator via interacting with its EC ligand sialyl Lex (11). Interestingly, there is an emerging concept of angiogenesis inhibitors, which suggests that they are also often stored in a preformed manner, as cryptic parts of larger molecules that are not themselves angiogenic. This may be the case with mediators such as the 29-kDa fragment of fibronectin, the 16-kDa fragment of prolactin, and angiostatin (a fragment of plasminogen), among others (28, 29, 30, 31). However, an alternative paradigm concerning the regulation of angiogenesis by inducers and inhibitors may be that structural mimicry plays a role. Hence, a search based on crystal structure revealed that the angiogenic inhibitor endostatin was most homologous to the angiogenic mediator E-selectin (B. Olsen, data presented at the Angiogenesis in Cancer Meeting, Orlando, FL, 1998). Therefore, it is possible that a family of Le Ag-like molecules including Ley/H and their ligands, such as E-selectin, in the case of sialyl Lex, regulates angiogenesis via structural mimicry. Here, we show that glycoconjugate Ley/H, which shares some structural overlap with sialyl Lex, in the soluble form elicits an angiogenic response comparable to that induced by potent angiogenic factors such as vascular endothelial growth factor and bFGF (24, 25).
One of the final steps in the biosynthesis of ABO blood group
oligosaccharide Ags in vivo is catalyzed by
-(1, 2)-fucosyltransferase(s) (32). The H and secretor
blood group loci determine the expression of these enzymes, which
construct Fuc 162Galß-linkages (H determinants) that are essential
precursors to the A and B Ags. Erythrocytes from rare Bombay blood
group phenotype individuals are deficient in H determinants, and
consequently A and B, the likely result of homozygosity for null
alleles at the H locus (32). To date, Bombay individuals
have never been examined specifically for angiogenic deficits, probably
because no biologic relevance of the H Ag has been determined
(33). However, such a defect may not be readily apparent,
because molecular redundancy is common with regard to required biologic
functions.
In summary, we have shown that Ley/H is a rapidly inducible molecule present at higher levels in RA ST compared with nonangiogenic OA ST and in RA SF compared with nonangiogenic OA SF. This suggests that Ley/H may play a contributing role in the process of inflammatory angiogenesis. This conclusion is further supported by the ability of the glucose analogues of Ley/H to induce cell migration in vitro and induce blood vessel growth in vivo. We hypothesize that upon cytokine or other immune modulator stimulation, Ley/H is rapidly expressed, shed from ECs, and subsequently exerts a direct angiogenic effect on adjacent ECs. These results suggest a novel function for blood group glycoconjugates, such as Ley/H, as angiogenic mediators. Efforts aimed at inhibiting the production or action of the Ley/H may curb the angiogenesis that characterizes conditions such as tumor growth, psoriasis, and RA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Alisa E. Koch, Department of Medicine, Rheumatology Section, Northwestern University Medical School, 303 East Chicago Avenue, Ward Building 3-315, Chicago, IL 60611.
3 Abbreviations used in this paper: RA, rheumatoid arthritis; EC, endothelial cell; ST, synovial tissue; SASP, sulfasalazine; CIA, calcium ionophore A23187; EBM, endothelial basal medium; vWF, von Willebrand factor; bFGF, basic fibroblast growth factor; HD, H-disaccharide; H7, 1-(5-isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride; SF, synovial fluid.
Received for publication July 27, 1999. Accepted for publication February 22, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1–3 fucosyltransferases. Glycobiology 1:577.(1,3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L-, E-, and P-selectin ligand biosynthesis. Cell 86:643.[Medline]
(1,2)fucosyltransferase gene among para-Bombay individuals. Vox Sang. 72:36.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. D. Paterson, M. F. Lopes-Virella, D. Waggott, A. P. Boright, S. M. Hosseini, R. E. Carter, E. Shen, L. Mirea, B. Bharaj, L. Sun, et al. Genome-Wide Association Identifies the ABO Blood Group as a Major Locus Associated With Serum Levels of Soluble E-Selectin Arterioscler Thromb Vasc Biol, November 1, 2009; 29(11): 1958 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nosaka, Y. Ishida, A. Tanaka, T. Hayashi, T. Miyashita, C. Kaminaka, W. Eisenmenger, F. Furukawa, and A. Kimura Aberrant Expression of Histo-blood Group A Type 3 Antigens in Vascular Endothelial Cells in Inflammatory Sites J. Histochem. Cytochem., March 1, 2008; 56(3): 223 - 231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Zhu, M. A. Amin, Y. Zha, L. A. Harlow, and A. E. Koch Mechanism by which H-2g, a glucose analog of blood group H antigen, mediates angiogenesis Blood, March 15, 2005; 105(6): 2343 - 2349. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Dabelsteen and S. Gao ABO Blood-group Antigens in Oral Cancer Journal of Dental Research, January 1, 2005; 84(1): 21 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A E Koch Angiogenesis as a target in rheumatoid arthritis Ann Rheum Dis, November 1, 2003; 62(90002): ii60 - 67. [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Amin, O. V. Volpert, J. M. Woods, P. Kumar, L. A. Harlow, and A. E. Koch Migration Inhibitory Factor Mediates Angiogenesis via Mitogen-Activated Protein Kinase and Phosphatidylinositol Kinase Circ. Res., August 22, 2003; 93(4): 321 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Zhu, M. A. Amin, M. J. Kim, K. J. Katschke Jr., C. C. Park, and A. E. Koch A Novel Function for a Glucose Analog of Blood Group H Antigen as a Mediator of Leukocyte-Endothelial Adhesion via Intracellular Adhesion Molecule 1 J. Biol. Chem., June 6, 2003; 278(24): 21869 - 21877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Toivonen, S. Nishihara, H. Narimatsu, O. Renkonen, and R. Renkonen Fuc-TIX: a versatile {alpha}1,3-fucosyltransferase with a distinct acceptor- and site-specificity profile Glycobiology, June 1, 2002; 12(6): 361 - 368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Dettke, G. Palfi, E. Pursch, M. B. Fischer, and H. Loibner Increased expression of the blood group-related Lewis Y antigen on synovial fluid granulocytes of patients with arthritic joint diseases Rheumatology, September 1, 2001; 40(9): 1033 - 1037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Koch The role of angiogenesis in rheumatoid arthritis: recent developments Ann Rheum Dis, November 1, 2000; 59(90001): i65 - 71. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Fold increase over 5
min values for 20 min (A) and 240 min
(B). 
, Conditioned medium from a control fibroblast
cell type not expected to express Ley/H was also analyzed
for comparison. Quantity of stimulants included indomethacin (25 µM),
TNF-
