The JI PBL Intereron Source
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Abe, Y.
Right arrow Articles by Ballantyne, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Abe, Y.
Right arrow Articles by Ballantyne, C. M.
The Journal of Immunology, 1999, 163: 2867-2876.
Copyright © 1999 by The American Association of Immunologists

Endothelial {alpha}2,6-Linked Sialic Acid Inhibits VCAM-1- Dependent Adhesion Under Flow Conditions1

Yasunori Abe*, C. Wayne Smith*,{dagger}, Julie P. Katkin{ddagger}, Lisa M. Thurmon*, Xudong Xu{ddagger}, Leonardo H. Mendoza* and Christie M. Ballantyne2,*,§

* Speros Martel Section of Leukocyte Biology, Department of Pediatrics, {dagger} Department of Microbiology and Immunology, {ddagger} Pulmonary Medicine, Department of Pediatrics, and § Section of Atherosclerosis, Department of Medicine, Baylor College of Medicine, Houston, TX 77030


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We have previously shown that costimulation of endothelial cells with IL-1 + IL-4 markedly inhibits VCAM-1-dependent adhesion under flow conditions. We hypothesized that sialic acids on the costimulated cell surfaces may contribute to the inhibition. Northern blot analyses showed that Galß1-4GlcNAc {alpha}2,6-sialyltransferase (ST6N) mRNA was up-regulated in cultured HUVEC by IL-1 or IL-4 alone, but that the expression was enhanced by costimulation, whereas the level of Galß1-4GlcNAc/Galß1-3GalNAc {alpha}2,3-sialyltransferase (ST3ON) mRNA was unchanged. Removing both {alpha}2,6- and {alpha}2,3-linked sialic acids from IL-1 + IL-4-costimulated HUVEC by sialidase significantly increased VCAM-1-dependent adhesion, whereas removing {alpha}2,3-linked sialic acid alone had no effect; adenovirus-mediated overexpression of ST6N with costimulation almost abolished the adhesion, which was reversible by sialidase. The same treatments of IL-1-stimulated HUVEC had no effect. Lectin blotting showed that VCAM-1 is decorated with {alpha}2,6- but not {alpha}2,3-linked sialic acids. However, overexpression of {alpha}2,6-sialyltransferase did not increase {alpha}2,6-linked sialic acid on VCAM-1 but did increase {alpha}2,6-linked sialic acids on other proteins that remain to be identified. These results suggest that {alpha}2,6-linked sialic acids on a molecule(s) inducible by costimulation with IL-1 + IL-4 but not IL-1 alone down-regulates VCAM-1-dependent adhesion under flow conditions.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Vascular cell adhesion molecule-1 is expressed on the surfaces of endothelial cells following stimulation with endotoxin and cytokines such as IL-1, TNF, IL-4, and IL-13 (1, 2), as well as oxidized low-density lipoprotein (3, 4) and advanced glycation end products (5), which are postulated to mediate atherosclerosis. VCAM-1 supports adhesion of leukocytes by interacting with {alpha}4 integrins (i.e., {alpha}4ß1 and {alpha}4ß7), which are constitutively expressed on monocytes, lymphocytes, eosinophils, and basophils but not on neutrophils in humans (1). For leukocytes to adhere to endothelial cells under physiological flow conditions, leukocytes must be captured from the bloodstream. Our laboratory and others have shown that interactions between VCAM-1 and {alpha}4 integrins are necessary and sufficient to support primary capture of mononuclear leukocytes under flow conditions in vitro (6, 7, 8, 9). This is in contrast to adhesion of neutrophils under flow conditions, in which selectins are indispensable for primary capture (1). A major isoform of VCAM-1 that is inducible on endothelial cells consists of seven Ig-like domains, and the binding sites for {alpha}4 integrins reside in domains 1 and 4 (10, 11, 12). For primary capture, only domain 1 of VCAM-1 is required, whereas both domains 1 and 4 are involved in adhesion under static conditions (8).

IL-4 is produced by Th2-type T cells, mast cells, basophils, and a subset of NK cells (13). It down-regulates endothelial cell expressions of E-selectin and ICAM-1 induced by IL-1 or TNF (14), and it enhances the expression of VCAM-1 (15, 16). However, we have previously shown that under flow conditions, adhesion of monocytes and T cells to cultured endothelial cells stimulated with IL-1 and IL-4 in combination is almost abolished (6, 8). Thus, despite enhanced VCAM-1 expression, under physiological flow conditions the net effect of the combination of IL-1 + IL-4 on endothelial cells is anti-adhesive. Although this might be partially explained by down-regulation of E-selectin or other selectin-like molecules, we have found that VCAM-1-dependent adhesion under flow conditions is markedly inhibited (8). This latter finding is of particular interest because the VCAM-1 expressed following IL-1 + IL-4 costimulation supports adhesion under static conditions as efficiently as VCAM-1 induced by IL-1 alone (8). These results strongly suggest that either the structure of VCAM-1 is somehow altered, interfering with the function of domain 1, or an antiadhesive mechanism(s) that works under flow conditions but not static conditions is up-regulated by the costimulation.

Sialic acids are 9-carbon monosaccharides that link to the terminal galactose, N-acetylgalactosamine, or other sialic acids in carbohydrate chains that are attached to glycoproteins or glycolipids (17, 18). Because of their terminal location, sialic acids on the cell surface are among the first molecules encountered by other cells coming in contact with the cell (17, 18). In addition, the first carbon of sialic acids is ionized at physiological pH (17, 18). This makes sialic acid the only sugar in glycoproteins that bears a net negative charge (17, 18). Because of their location and negative charge, sialic acids have the potential to inhibit interactions between molecules (17, 18). There are several reports that document the antiadhesive nature of sialic acids (19, 20, 21).

Recently, Hanasaki et al. (22, 23) reported that Galß1-4GlcNAc{alpha}2,6-sialyltransferase (ST6N)3 is up-regulated in endothelial cells following stimulation with IL-1, TNF, and IL-4 and that VCAM-1 molecules are decorated with {alpha}2,6-linked sialic acids. These reports prompted us to examine the following questions: 1) are sialic acids on the cell surfaces of HUVEC required for the inhibition of VCAM-1-dependent adhesion by costimulation with IL-1 + IL-4, 2) does overexpression of sialic acids lead to the inhibition of VCAM-1-dependent adhesion, and 3) if the above hypotheses are correct, is the inhibition due to increased sialylation of VCAM-1 molecules?

In this paper we demonstrate for the first time that {alpha}2,6-linked sialic acids on endothelial cells can inhibit VCAM-1-dependent adhesion of leukocytes under flow conditions. However, this inhibitory effect is neither due to increased sialylation of VCAM-1 nor related to the total amount of sialic acids present on the cell surfaces. Rather, to exhibit their inhibitory effect, sialic acids must be associated with a molecule(s) that is inducible by stimulation with IL-1 and IL-4 in combination but not IL-1 alone. We propose a novel inducible mechanism on endothelial cells that negatively regulates leukocyte adhesion under flow conditions.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture

HUVEC were harvested from 5 to 10 umbilical cords by collagenase digestion as previously described (24). Primary cultured cells were passaged with 0.05% trypsin and 0.53 mM EDTA (Life Technologies, Gaithersburg, MD) and seeded onto tissue culture dishes or flasks (Corning, Cambridge, MA) coated with 1% gelatin (Sigma, St. Louis, MO). The cells were grown in M199 (Life Technologies) supplemented with 10% FCS (HyClone, Logan, UT), 10% CS (HyClone), 0.1 mg/ml heparin (Sigma), 50 µg/ml endothelial cell growth supplement (Collaborative Research, Bedford, MA), and 1% penicillin-streptomycin (Life Technologies). First-passage HUVEC were used in all experiments.

Ramos cells (CRL-1596) and 293 cells (CRL-1573) were obtained from the American Type Culture Collection (Manassas, VA) and maintained in RPMI 1640 medium (Life Technologies) and Minimum Essential Medium (Life Technologies), respectively, containing 10% FCS and 1% penicillin-streptomycin.

RT-PCR and PCR cloning of {alpha}2,3-sialyltransferase cDNA

Based on conserved motifs of three published human {alpha}2,3-sialyltransferases (25, 26, 27), three pairs of oligonucleotides were synthesized at the Child’s Health Research Center Core Laboratory, Baylor College of Medicine. The sequences of sense and antisense primers were: Galß1-3(4)GlcNAc {alpha}2,3-sialyltransferase (ST3N) (25), 5'-GCT GCC GCC GCT GCA TCA TCG TGG GCA A (sense) and 5'-GAT GAA ATA TGG GTT GAG GAT (antisense); Galß1-3GalNAc {alpha}2,3-sialyltransferase (ST3O) (26), 5'-GCT GCC GGC GCT GCG CCG TTG TGG GCA A (sense) and 5'-GAT GAA GGC TGG GTG GTA GAT (antisense); and Galß1-4GlcNAc/Galß1-3GalNAc {alpha}2,3-sialyltransferase (ST3ON) (27), 5'-GGT GCC GCC GCT GTG TGG TCG TGG (sense) and 5'-CCG GCA ATG TGC ACC AAG TCA CAG (antisense). Total RNA was isolated from unstimulated and IL-1-, IL-4-, and IL-1 + IL-4-stimulated HUVEC by acid guanidium thiocyanate-phenol-chloroform extraction (28). Reverse transcriptase (RT) reaction was performed by incubating 2 µg of total RNA with 1.5 µM antisense oligonucleotides and 1.5 mM dNTP (Promega, Madison, WI) in 10 µl of RT buffer (Promega) at 80°C for 5 min, followed by mixing with 23 U of avian myeloblastosis virus (AMV)-RT (Promega) and 40 U of RNase inhibitor (Promega) in 5 µl of RT buffer, and incubating at 42°C for 2 h. Then, the RT products were diluted eight times with PCR buffer (Fisher Scientific, Pittsburgh, PA), and the subsequent PCR was performed with 0.2 µM sense oligonucleotides, 0.2 mM dNTP, 2–5 mM MgCl2, and 5 U/50 µl of Taq polymerase (Fisher Scientific) for 33 cycles. PCR product was gel purified and cloned into pGEM-T Easy vector (Promega). Sequence analysis and database searches were performed by using the Molecular Biology Information Resources Center at Baylor College of Medicine.

Northern blot analysis

Total RNA was isolated from HUVEC by acid guanidium thiocyanate-phenol-chloroform extraction (28). Subsequently, 10 µg of total RNA was electrophoresed in a 1% agarose/formaldehyde gel. The RNA was then transferred to a charged nylon membrane (GeneScreenPlus, NEN, Boston, MA). Equal loading of lanes was assessed by photographs of the ethidium bromide-stained gels as well as by probing of the membranes with human GAPDH cDNA (29). cDNA probes for human VCAM-1 (30) and ST6N (31) were gel purified with QIAEX II (Qiagen, Chatsworth, CA) and were labeled with [32P]-dCTP using the random-hexamers (Prime-it II; Stratagene, La Jolla, CA) according to the manufacturer’s protocol. The membrane was prehybridized with QuikHyb solution (Stratagene), and hybridization was performed as previously described (29). To quantify radioactivity, ß-emissions from respective bands were measured by the Molecular Image Scanner (Bio-Rad, Hercules, CA) and the values were normalized by those from the bands hybridized with GAPDH cDNA. Radioactivity was normalized and compared as follows: relative radioactivity compared with unstimulated HUVEC = [(counts from respective ST6N band/counts from the GAPDH of the same sample lane) ÷ (counts from the ST6N band of unstimulated HUVEC/counts from the GAPDH band of unstimulated HUVEC)].

Detection of cell surface expression of {alpha}2,3- and {alpha}2,6-linked sialic acids

Expression of {alpha}2,3- and {alpha}2,6-linked sialic acids on endothelial cell surfaces was detected by cell-surface ELISA using lectins. HUVEC were seeded onto 1% gelatin-coated 96-well tissue culture plates (Becton Dickinson, San Jose, CA). Two or 3 days after passage, cells were stimulated with or without culture medium containing IL-1 (Genzyme, Cambridge, MA), IL-4 (Genzyme), and IL-1 + IL-4 at 37°C with 5% CO2 for 22.5 h. In some wells, 0.1 U/ml of sialidase from Arthrobacter ureafaciens (Boehringer Mannheim, Indianapolis, IN), which cleaves both {alpha}2,3- and {alpha}2,6-linked sialic acid (32), or 20 U/ml of recombinant sialidase L (V-LABS, Covington, LA), which specifically cleaves {alpha}2,3-linked sialic acid (33), was added to the culture medium and incubated for another 90 min. After washing four times with Dulbecco’s PBS (D-PBS), cells were fixed with D-PBS containing 2% paraformaldehyde and 0.2% glutaraldehyde for 20 min. After monolayers were blocked with Tris-buffered saline containing 0.5% casein (Boehringer Mannheim), {alpha}2,6- and {alpha}2,3-linked sialic acids were detected by Sambucus nigra (SNA) and Maackia amurensis (MAA) agglutinins labeled with digoxigenin (Boehringer Mannheim), respectively, in Tris-buffered saline (pH 7.5) containing 1 mM CaCl2, 1 mM MgCl2, and 1 mM MnCl2 (34, 35). Bound lectins were detected by alkaline phosphatase-conjugated goat anti-digoxigenin Ab (Boehringer Mannheim) and p-nitrophenyl phosphate (Sigma) as a substrate. The optical absorbance was measured at 405 nm by an automatic microplate reader (Cambridge Technology, Watertown, MA). Values of optical absorbance at 405 nm were subtracted with corresponding controls (wells in which lectins were not added).

Flow adhesion assay

The adhesion of Ramos cells to HUVEC under hydrodynamic flow conditions was studied as previously described (8, 36). Briefly, cell monolayers in 35-mm tissue culture dishes were mounted in a parallel-plate flow system. Ramos cells suspended at 106 cells/ml in 37°C D-PBS containing Ca2+, Mg2+, and 11 mM glucose were passed through the parallel plate flow chamber at 1.0 dyne/cm2 under a phase contrast microscope (Diaphot TMD, Nikon, Garden City, NY) for 9 min at 37°C. At the end of the perfusion, four different fields were videotaped at 15-s intervals. Videotaped images were analyzed with Optimas image analysis software (Bioscan, Edmonds, WA). The number of stably adherent cells was quantitated as the average number of leukocytes remaining on the monolayer in the four fields.

Construction of a recombinant adenovirus carrying ST6N cDNA

A recombinant type 5 adenovirus (Ad5) vector expressing the ST6N gene was constructed according to standard methods. Briefly, an expression cassette containing the ST6N cDNA (a generous gift of Dr. I. Stamenkovic, Harvard Medical School) under control of the human elongation factor-1{alpha} promoter (EF-1{alpha} promoter) (37), and the bovine growth hormone polyadenylation signal was subcloned into the shuttle plasmid p{delta}E1sp1B (Microbix Biosystems, Toronto, Canada), which contains the 5' Ad5 sequence with a deletion in the E1 region (38). The transformed shuttle plasmid DNA was cotransfected into Ad5-transformed 293 cells (39) with the circular Ad5 genome plasmid pBHGE3 (40) (Microbix Biosystems) to generate recombinant adenovirus carrying ST6N cDNA (Ad26st). Recombinant viruses were isolated by plaque purification and characterized by PCR. A plaque containing the correct recombinant vector was propagated to high titer by serial passage on 293 cells and purified on cesium chloride gradients using standard methods (41, 42). The titer was determined by identifying the presence of viral cytopathic effect in 293 cells infected with serial dilutions of each preparation. The preparation used for these experiments had a titer of 2 x 1011 infectious units/ml. The EF-1{alpha} promoter used in this construct has generated high levels of transgene expression from other recombinant vectors produced in our laboratory (43).

Transduction of HUVEC with adenovirus vectors

After HUVEC were grown to 80–90% confluence, they were incubated with medium containing recombinant virus at 37°C with 5% CO2 for 24 h. For overexpression of {alpha}2,6-sialyltransferase and cytokine stimulation, cells were first incubated with Ad26st in medium at 37°C with 5% CO2 for 17 h, and then 10 U/ml of IL-1 and 10 ng/ml of IL-4, or 10 U/ml of IL-1 + 10 ng/ml of IL-4 was added directly to the medium and the incubation was continued for another 24 h.

Detection of ß-galactosidase in HUVEC following transduction with adenovirus carrying ß-galactosidase cDNA

Following transduction of HUVEC with recombinant adenovirus carrying ß-galactosidase cDNA (Adgal) for 24 h, the cells were stained with 5-bromo-4-chloro-3-indolyl ß-D-galactoside (X-Gal) as follows. Cell monolayers were fixed with 2% paraformaldehyde and 0.2% glutaraldehyde in D-PBS for 10 min. After rinsing twice with D-PBS containing 50 mM glycine, cells were stained with 5 mM K4Fe(CN)6, 5 mM K3Fe(CN)6, 1 mg/ml X-Gal (Sigma), 2 mM MgCl2, 0.1% deoxycholic acid, and 0.2% Nonidet P-40 in D-PBS at 37°C for 2 h.

Lectin blotting of whole cell lysates and immunoprecipitated VCAM-1

Lectin blotting of whole cell lysates and immunoprecipitated VCAM-1 was performed as previously described (22). After washing HUVEC monolayers four times with ice-cold D-PBS, the cells were lysed with 1 ml of 20 mM Tris/HCl (pH 8.0), 150 mM NaCl containing 1% Nonidet P-40, 1 mM EDTA, 1 mM PMSF (Sigma), 10 µg/ml aprotinin (Boehringer Mannheim), and 2 µg/ml of leupeptin (Boehringer Mannheim) at 4°C for 30 min. The lysates were centrifuged at 10,000 x g at 4°C for 15 min to remove debris. For lectin blotting of whole cell lysates, protein concentration of cell lysates was measured by the BCA protein assay kit (Pierce, Rockford, IL), and 20 µg of total protein per lane was separated by 7.5% SDS-PAGE under reducing conditions and transferred to nitrocellulose membranes (Schleicher & Schuell, Keene, NH) by standard methods. For immunoprecipitation of VCAM-1, after preclearing with protein A-Sepharose (Pierce) at 4°C for 2 h, 400 µg of total protein was incubated with 15 µg/ml of anti-VCAM-1 mAb, 4B9 (a generous gift from Dr. Roy Lobb, Biogen) (44), at 4°C for 2 h. The immune complexes were then absorbed with protein A-Sepharose at 4°C for 2 h. The protein A-Sepharose beads were washed with lysis buffer, and the beads were boiled with Laemmli sample buffer containing 100 mM dithiothreitol (Sigma). Supernatants were separated by 7.5% SDS-PAGE and transferred to nitrocellulose membranes. For detection of sialic acids, after blots were blocked with 0.5% casein, {alpha}2,6- and {alpha}2,3-linked sialic acids were detected by digoxigenin-labeled SNA (Boehringer Mannheim) (34) and MAA (Boehringer Mannheim) (35), respectively. Bound lectins were detected by alkaline phosphatase-conjugated sheep anti-digoxigenin Ab (Boehringer Mannheim). For detection of VCAM-1, after blots were blocked with 3% BSA in Tris-buffered saline, VCAM-1 was probed with goat anti-VCAM-1 sera (R&D Systems, Minneapolis, MN) and alkaline phosphatase-conjugated rabbit anti-goat IgG (Pierce). Bound alkaline phosphatase-conjugated Abs were detected by bromochloroindoyl phosphate/nitro blue tetrazolium chloride (Boehringer Mannheim).

Statistics

Results are presented as mean ± SD. Statistical assessments were made using Student’s t test or one-way ANOVA with Bonferroni multiple comparisons. The p values that exceeded 0.05 were not considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Expression of {alpha}2,6-sialyltransferase and {alpha}2,3-sialyltransferase mRNA in HUVEC

Human ST6N cDNA has been identified and cloned by three independent groups from different cell types (31, 45, 46). This enzyme is expressed in a wide variety of tissues, and the size of the mRNA varies in cell types due to differences in the 5' untranslated regions (22, 26, 46). As shown in Fig. 1GoA, ST6N mRNA was detected in unstimulated HUVEC; the size of the mRNA was ~4.2 kb, which is consistent with Hanasaki et al. (22). Because we have previously shown that the anti-adhesive effect of IL-1 + IL-4 costimulation appears later than 8 h and remains at 24 h after incubation (8), we were interested in the levels of ST6N mRNA at these time points. After 22 h of incubation, ST6N mRNA was up-regulated by either IL-1 or IL-4 alone. However, up-regulation was further enhanced by the combination of IL-1 + IL-4 (see Fig. 1GoA). The relative radioactivities of bands identified with ST6N cDNA compared with those of unstimulated HUVEC were 1.25 ± 0.35 (IL-1), 1.64 ± 0.27 (IL-4), and 2.27 ± 0.48 (IL-1 + IL-4).



View larger version (62K):
[in this window]
[in a new window]
  		FIGURE 1. Northern blot analyses of ST6N and ST3ON mRNA expression in HUVEC after cytokine stimulation. HUVEC monolayers were stimulated with 10 U/ml IL-1, 10 ng/ml IL-4, and 10 U/ml IL-1 + 10 ng/ml IL-4 at 37°C with 5% CO2 for 22 h. Total RNA (10 µg/lane) was hybridized with cDNA probes for human {alpha}2,6 sialyltransferase, ST6N (ST6) (A, upper panel), and human {alpha}2,3 sialyltransferase, ST3ON (ST3) (B lower panel) as described in Materials and Methods. For the detection of GAPDH mRNA, ST6N and ST3ON probes were stripped from the filters and the same filter was reprobed with GAPDH cDNA.

 
To examine which {alpha}2,3-sialyltransferase is expressed in human endothelial cells, we performed RT-PCR using primers synthesized according to three published human {alpha}2,3-sialyltransferases—ST3N, ST3O, and ST3ON—which use glycoproteins as acceptor substrates. Although we tried using different annealing temperatures and Mg2+ concentrations, PCR product was only obtained with primers made from the sequence of ST3ON. The PCR product was a single band on gel electrophoresis, and this 493-bp PCR product was cloned into pGEM-T Easy vector. Four clones randomly picked were sequenced and found to be identical with the published sequence of ST3ON cDNA (27). Northern blot analysis using this clone as a probe demonstrated that ST3ON mRNA was constitutively expressed in HUVEC (Fig. 1GoB). The size of the RNA was ~2.0 kb, which is consistent with a previous report (27). However, unlike ST6N, the levels of ST3ON mRNA were not regulated by IL-1, IL-4, or IL-1 + IL-4 (see Fig. 1GoB).

Surface expression of {alpha}2,3- and {alpha}2,6-linked sialic acids on HUVEC

MAA and SNA lectins recognize {alpha}2,3- and {alpha}2,6-linked sialic acids, respectively (34, 35). In spite of the increased expression of ST6N, the binding of SNA lectin to HUVEC stimulated with IL-1, IL-4, and IL-1 + IL-4 for 24 h was not significantly different from unstimulated HUVEC (Fig. 2GoA). After treatment of HUVEC with sialidase from A. ureafaciens, which cleaves all three linkages of sialic acids (32), the binding of SNA lectin was markedly reduced (see Fig. 2GoA). The reduction was similar among different cytokine stimulations. This suggests that up-regulation of ST6N has little effect on total amount of {alpha}2,6-linked sialic acids on cell surfaces. Similarly, MAA binding to HUVEC was not affected by stimulation with IL-1, IL-4, or IL-1 + IL-4 for 24 h. This binding was almost eliminated by sialidase (Fig. 2GoB).



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 2. Binding of SNA and MAA lectins to HUVEC and the effects of sialidase. HUVEC were seeded onto 96-well tissue culture plates. After cells were grown to confluence, the monolayers were stimulated with 10 U/ml IL-1, 10 ng/ml IL-4, or 10 U/ml IL-1 + 10 ng/ml IL-4 at 37°C with 5% CO2 for 24 h. In some wells, 0.1 U/ml of sialidase from A. ureafaciens was added to the culture medium and incubated for the last 90 min. Sialic acids on the cell surfaces were detected as described in Materials and Methods. Values represent mean absorbance at 405 nm ± SD from triplicate experiments. SNA lectin detects {alpha}2,6-linked sialic acids (A), and MAA lectin detects {alpha}2,3-linked sialic acids (B).

 
Effects of sialidase treatment of HUVEC on Ramos cell adhesion under flow conditions

Ramos cells are able to tether and adhere stably to cytokine-stimulated HUVEC under shear stresses up to 2.0 dyne/cm2 (8). This tethering and stable adhesion are mediated exclusively by interactions between {alpha}4 integrins on Ramos cells and VCAM-1 on HUVEC (8). To examine the role of sialic acids on the surfaces of HUVEC in {alpha}4/VCAM-1-dependent adhesion under flow conditions, we studied Ramos cell adhesion at 1.0 dyne/cm2 to HUVEC that were stimulated with cytokines and then incubated with or without sialidase. Because most Ramos cells that tethered stably adhered, we examined only the number of stably adherent Ramos cells. Treatment of HUVEC with sialidase after 22.5 h of cytokine stimulation, for the last 90 min of cytokine stimulation, did not cause any visible change in the monolayers under phase contrast microscopy. Although incubation for 90 min at neutral pH in culture medium might not be optimal for sialidase, this was sufficient to remove sialic acids from surfaces of HUVEC monolayers as shown in Fig. 2Go. Under flow conditions, costimulation of HUVEC with IL-1 + IL-4 markedly inhibited {alpha}4 integrin/VCAM-1-dependent adhesion compared with stimulation with IL-1 alone (Fig. 3GoB), although the surface expression of VCAM-1 was higher in the costimulated HUVEC, as was seen in our previous report (8). Treatment of IL-1 + IL-4-costimulated HUVEC with sialidase significantly increased adhesion of Ramos cells, and this increase was completely inhibited by the anti-VCAM-1 mAb 4B9 (44) and the anti-{alpha}4 integrin mAb HP2/1 (47) (Fig. 3GoA). In contrast, the same treatment of either IL-1- or IL-4-stimulated HUVEC failed to increase the adhesion of Ramos cells (see Fig. 3GoA). Thus, sialic acids are required for the inhibition of VCAM-1-dependent adhesion in IL-1 + IL-4-costimulated HUVEC. However, this is not related solely to the total amount of sialic acids on cell surfaces, because removing a high percentage of sialic acids from the surfaces of HUVEC stimulated with either IL-1 or IL-4 had no significant effect on adhesion.



View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 3. Effects of sialidase on Ramos cell adhesion to HUVEC under flow conditions. HUVEC were seeded onto 35-mm culture dishes. After cells were grown to confluence, the monolayers were stimulated with 10 U/ml IL-1, 10 ng/ml IL-4, and 10 U/ml IL-1 + 10 ng/ml IL-4 at 37°C with 5% CO2 for 24 h. In some experiments, 0.1 U/ml sialidase from A. ureafaciens was added to the culture medium and incubated for the last 90 min. For blocking of VCAM-1, 20 µg/ml of mAb 4B9 was added to the culture medium for 20 min immediately before assay and to the perfusion buffer to maintain saturation during assay. For blocking of {alpha}4 integrins, Ramos cells were preincubated with 10 µg/ml of HP2/1 at room temperature for 10 min immediately before assay. Adhesion assays were performed at 1.0 dyne/cm2 as described in Materials and Methods. For each cytokine stimulation, the number of stably adherent Ramos cells after sialidase treatment of HUVEC was compared with the number of Ramos cells stably adherent to HUVEC without sialidase to provide percent adhesion (A). The actual number of stably adherent Ramos cells is shown in B. Values represent the mean ± SD from four to eight experiments except blocking studies using Abs (n = 2). *, p < 0.001 as compared with the group not treated with sialidase.

 
Effects of general versus {alpha}2,3-linkage-specific sialidase on lectin binding and on Ramos cell adhesion under flow conditions

Because sialidase isolated from A. ureafaciens cleaves both {alpha}2,6- and {alpha}2,3-linked sialic acids as shown in Fig. 2Go, the increased adhesion of Ramos cells following the treatment of IL-1 + IL-4-costimulated HUVEC with the sialidase may have been due to the removal of {alpha}2,6- or {alpha}2,3-linked sialic acids. Although to our knowledge sialidase specific to {alpha}2,6-linked sialic acid is not known, it has been reported that sialidase from Macrobdella decora leech (sialidase L) specifically cleaves {alpha}2,3-linked sialic acids (33). We examined whether this sialidase cleaves {alpha}2,3-linked sialic acids from the surfaces of HUVEC under the same conditions used for sialidase from A. ureafaciens. Confluent monolayers of HUVEC grown in 96-well plates were costimulated with IL-1 (10 U/ml) and IL-4 (10 ng/ml) for 24 h; in some wells, 20 U/ml of recombinant sialidase L or 0.1 U/ml of sialidase from A. ureafaciens was added to the culture medium and incubated for the last 90 min. {alpha}2,6- and {alpha}2,3-linked sialic acids on the surfaces of HUVEC were detected by the binding of SNA and MAA lectins, respectively, as described in Materials and Methods. As shown in Fig. 4GoB, sialidase L significantly reduced the level of {alpha}2,3-linked sialic acid on IL-1 + IL-4-costimulated HUVEC, whereas it did not change the level of {alpha}2,6-linked sialic acid (Fig. 4GoA). To determine whether removing {alpha}2,3-linked sialic acid from IL-1 + IL-4-costimulated HUVEC increases Ramos cell adhesion, HUVEC were costimulated with IL-1 + IL-4 for 24 h and treated with sialidases for the last 90 min, and the adhesion of Ramos cells to HUVEC was compared under flow conditions (Fig. 4GoC). Removal of {alpha}2,3-linked sialic acids by sialidase L treatment of IL-1 + IL-4-costimulated HUVEC did not show any increase in Ramos cell adhesion, whereas treatment with sialidase from A. ureafaciens markedly increased the adhesion. These results suggest that {alpha}2,6-linked sialic acid is required for the inhibition of VCAM-1-dependent adhesion by IL-1 and IL-4.



View larger version (12K):
[in this window]
[in a new window]
  		FIGURE 4. Effects of general vs {alpha}2,3-linkage-specific sialidase on lectin binding and on Ramos cell adhesion under flow conditions. A and B, SNA and MAA lectin binding. HUVEC were seeded onto 96-well tissue culture plates. After cells were grown to confluence, the monolayers were stimulated with 10 U/ml IL-1 + 10 ng/ml IL-4 at 37°C with 5% CO2, for 24 h. In some wells, 20 U/ml of sialidase L or 0.1 U/ml of sialidase from A. ureafaciens was added to the culture medium and incubated for the last 90 min. Sialic acids on the cell surfaces were detected as described in Materials and Methods. Values represent mean absorbance at 405 nm ± SD from triplicate experiments. SNA lectin detects {alpha}2,6-linked sialic acids (A), and MAA lectin detects {alpha}2,3-linked sialic acids (B). C, Ramos cell adhesion. HUVEC were seeded onto 35-mm culture dishes. After cells were grown to confluence, the monolayers were stimulated with 10 U/ml IL-1 + 10 ng/ml IL-4 at 37°C with 5% CO2 for 24 h. In some experiments, 20 U/ml of sialidase L or 0.1 U/ml sialidase from A. ureafaciens was added to the culture medium and incubated for the last 90 min. Adhesion assays were performed at 1.0 dyne/cm2 as described in Materials and Methods. The number of stably adherent Ramos cells after sialidase treatment of HUVEC was compared with the number of Ramos cells stably adherent to HUVEC without sialidase to provide percent adhesion. Values represent the mean ± SD from four separate experiments. *, p < 0.0001; {dagger}, p < 0.005 as compared with the group not treated with sialidase.

 
Overexpression of human {alpha}2,6-sialyltransferase in HUVEC by adenovirus vector

To investigate whether overexpression of ST6N in HUVEC inhibits VCAM-1-dependent adhesion, ST6N cDNA was introduced into HUVEC by means of a recombinant adenovirus vector. To confirm the expression of ST6N mRNA by the virus, HUVEC were incubated with increasing concentrations of Ad26st for 24 h, and Northern blot analysis was performed. The transcript length of ST6N expressed by Ad26st was 2.7 kb and was expressed at high levels at multiplicity of infection (MOI) >=100 infectious units (Fig. 5GoA). Wild-type ST6N mRNA, which has a longer untranslated region and is ~4.2 kb, was not evident because of the very short exposure time of the x-ray film (4 h).



View larger version (77K):
[in this window]
[in a new window]
  		FIGURE 5. Expression of ST6N mRNA following transduction of HUVEC with Ad26st and the effect on the expression of VCAM-1 mRNA. A; After HUVEC were grown to 80–90% confluence, the monolayers were incubated with medium alone, Adgal at 500 MOI, or increasing concentrations of Ad26st in HUVEC medium at 37°C with 5% CO2 for 24 h. Total RNA was extracted and hybridized with ST6N cDNA probe. High levels of the 2.7-kb transcript of ST6N (ST6) expressed by Ad26st were observed for MOI of >=100. Endogenous ST6N mRNA, which is 4.2 kb in length, was not evident because of the very short exposure time of the x-ray film (4 h) for this figure. B; Incubation of HUVEC monolayers with medium alone, Adgal at 250 MOI and 500 MOI, or increasing concentrations of Ad26st or 10 U/ml IL-1. Total RNA was extracted and hybridized with VCAM-1 cDNA probe as described in Materials and Methods. Increased levels of VCAM-1 mRNA were not observed at MOI of <500 compared with controls.

 
Because HUVEC constitutively express {alpha}2,6-linked sialic acids and ST6N, it is difficult to distinguish the expression of ST6N induced by gene transfer from that of wild type in each cell. Therefore, instead of ST6N, we examined the expression of ß-galactosidase after the cells were transduced with increasing concentrations of recombinant Ad5 carrying ß-galactosidase cDNA for 24 h. Expression was confirmed by visual inspection of infected monolayers after exposure to X-Gal. In this study, 100% of cells expressed the transgene following infection at an MOI of 100 (data not shown).

To determine whether transfection of the adenovirus vector itself may induce expression of cell adhesion molecules, Northern blot analysis of VCAM-1 mRNA was performed (Fig. 5GoB). Although VCAM-1 mRNA was up-regulated by the virus at concentrations >500 MOI, the level of mRNA expression was not different from controls at concentrations <250 MOI (see Fig. 5GoB). Based on these results, we selected an MOI of 100 for use in subsequent experiments, similar to that previously published for HUVEC (48).

Effects of overexpression of human {alpha}2,6-sialyltransferase in HUVEC on Ramos cell adhesion under flow conditions

To examine the effects of overexpression of ST6N on VCAM-1-dependent adhesion, adhesion of Ramos cells to HUVEC stimulated with cytokines in association with overexpression of ST6N was studied under flow conditions. In HUVEC stimulated with either IL-1 or IL-4 alone for 24 h, overexpression of ST6N did not have a significant effect on Ramos cell adhesion (Fig. 6GoA). Although the adhesion of Ramos cells to IL-1 + IL-4-costimulated HUVEC was significantly inhibited compared with IL-1 stimulation alone, the inhibition was not complete and VCAM-1-dependent adhesion was still observed (see Fig. 3Go) (8). However, overexpression of ST6N in IL-1 + IL-4-costimulated HUVEC led to further inhibition of Ramos cell adhesion, and adhesion was nearly eliminated (Fig. 6Go, B and C), whereas transduction of HUVEC with Adgal at the same MOI had no effect (Fig. 6GoB). This inhibition was reversed by removing sialic acids from HUVEC (see Fig. 6Go, B and C), and the adhesion was completely VCAM-1 dependent (see Fig. 6Go, B and C).



View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 6. Effects of overexpression of ST6N on adhesion of Ramos cells to HUVEC under flow conditions. A; HUVEC were seeded onto 35-mm culture dishes. After cells were grown to 80–90% confluence, they were incubated with or without 100 MOI of Ad26st for 17 h. Then, 10 U/ml IL-1, 10 ng/ml IL-4, or 10 U/ml IL-1 + 10 ng/ml IL-4 were added to the culture medium for 24 h. Adhesion assays were performed at 1.0 dyne/cm2. Values represent percent adhesion of Ramos cells to ST6N-overexpressed HUVEC compared with those without the overexpression. B; HUVEC were incubated with or without 100 MOI of Ad26st or Adgal for 17 h. Then, 10 U/ml IL-1 + 10 ng/ml IL-4 were added to the culture medium for 24 h. In some experiments, IL-1 + IL-4-costimulated HUVEC were treated with 0.1 U/ml sialidase from A. ureafaciens for the last 90 min of the cytokine stimulation. For blocking of VCAM-1, 20 µg/ml of mAb 4B9 was added to the culture medium and to the perfusion buffer. The number of adherent cells in each condition was compared to that in the control to provide a percentage. Values represent mean ± SD from three to five experiments except blocking studies using 4B9 (n = 2). *, p < 0.001 compared with control; {dagger}, p < 0.001 compared with the group with Ad26st without sialidase. C; Images were taken from one representative set of experiments, which correspond to the conditions described in B: IL-1 + IL-4 costimulation alone (N), overexpression of ST6N and IL-1 + IL-4 costimulation (ST), sialidase treatment following overexpression of ST6N and IL-1 + IL-4 costimulation (SA), and blocking of VCAM-1 and sialidase treatment following overexpression and IL-1 + IL-4 costimulation (SB). The white round cells are Ramos cells attached to HUVEC, which are seen as dark cells in the background; the white blurs are the paths of flowing Ramos cells.

 
Lectin blotting of VCAM-1 immunoprecipitated from HUVEC

We wished to determine whether VCAM-1 induced by IL-1, IL-4, and IL-1 + IL-4 is decorated by {alpha}2,6-linked sialic acids and whether overexpression of ST6N up-regulates the levels of {alpha}2,6-linked sialic acids on VCAM-1. VCAM-1 molecules immunoprecipitated from HUVEC stimulated with IL-1 and IL-1 + IL-4 showed two distinct bands in SDS-PAGE. The lower band and upper band correspond to the precursor and mature forms of VCAM-1, respectively (Fig. 7Go, A and B) (49). The amount of VCAM-1 immunoprecipitated from IL-1 + IL-4-costimulated HUVEC was higher than from IL-1 stimulation alone, as reported previously (see Fig. 7GoB) (8). VCAM-1 immunoprecipitated from HUVEC costimulated with IL-1 + IL-4 was decorated with {alpha}2,6- but not {alpha}2,3-linked sialic acids (see Fig. 7GoA), and {alpha}2,6-linked sialic acids were also detected on VCAM-1 molecules induced by IL-1 (see Fig. 7GoB). However, overexpression of ST6N did not increase the levels of {alpha}2,6-linked sialic acids on VCAM-1 induced by either IL-1 or IL-1 + IL-4 (see Fig. 7GoB).



View larger version (56K):
[in this window]
[in a new window]
  		FIGURE 7. {alpha}2,6-linked sialic acids on VCAM-1 following the overexpression of ST6N. A; After HUVEC were grown to confluence, the cells were costimulated with 10 U/ml IL-1 + 10 ng/ml IL-4 for 24 h. Then, VCAM-1 was immunoprecipitated with anti-VCAM-1 mAb 4B9 from detergent lysates, and {alpha}2,6- and {alpha}2,3-linked sialic acids on VCAM-1 were detected by SNA and MAA lectin respectively as described in Materials and Methods. VCAM-1 was also detected by a polyclonal anti-VCAM-1 Ab. B; After HUVEC were grown to 80–90% confluence, they were incubated with or without 100 MOI of Ad26st for 17 h. Then, 10 U/ml IL-1 or 10 U/ml IL-1 + 10 ng/ml IL-4 were added to the culture medium and the cells were incubated for another 24 h. The amount of {alpha}2,6-linked sialic acids on VCAM-1 was not increased after incubation with Ad26st.

 
Lectin blotting of whole cell lysates

Because the level of {alpha}2,6-linked sialic acid on VCAM-1 was unchanged despite the overexpression of {alpha}2,6-sialyltransferase mRNA by adenovirus vector, we examined whether the increased {alpha}2,6-sialytransferase mRNA resulted in an increased level of {alpha}2,6-linked acid on other proteins. Detergent lysates of HUVEC stimulated with IL-1 + IL-4 had numerous proteins that were detected by SNA lectin (Fig. 8Go). Although we did not see any difference in {alpha}2,6-linked sialic acids on VCAM-1, we did see an increase of {alpha}2,6-linked sialic acids on some other proteins, which are larger than 90 kDa, following overexpression of ST6N in IL-1 + IL-4-costimulated HUVEC (see Fig. 8Go). This finding suggests that transduction of HUVEC with Ad26st not only increased ST6N mRNA but also increased enzyme activity, which resulted in an increase in {alpha}2,6-linked sialic acids on at least several proteins.



View larger version (32K):
[in this window]
[in a new window]
  		FIGURE 8. Effects of the overexpression of ST6N on SNA binding to proteins expressed by HUVEC. After HUVEC were grown to 80–90% confluence, they were incubated with or without 100 MOI of Ad26st at 37°C with 5% CO2 for 17 h. Then, 10 U/ml IL-1 or 10 U/ml IL-1 + 10 ng/ml IL-4 were added to the culture medium and the cells were incubated at 37°C with 5% CO2 for another 24 h. The cells were lysed, and SNA lectin blotting, which detects {alpha}2,6-linked sialic acids, showed increased binding of SNA to several high-molecular weight proteins.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Published evidence suggests that adhesion of mononuclear leukocytes to endothelial cells is mediated by cell adhesion molecules, including {alpha}4 integrins and VCAM-1, in a ligand and receptor-specific manner (1). Adhesion of leukocytes can be regulated by the levels of expression and activation state of cell adhesion molecules (1). In this study, we have demonstrated a novel regulatory mechanism of leukocyte adhesion to endothelial cells distinct from the above mechanisms. This inducible mechanism on endothelial cells negatively regulates leukocyte adhesion under flow conditions. For the following reasons, we concluded that at least part of this mechanism is attributable to {alpha}2,6-linked sialic acids that are associated with a molecule(s) inducible by costimulation of HUVEC with IL-1 + IL-4: 1) removing both {alpha}2,3- and {alpha}2,6-linked sialic acids increased VCAM-1-dependent adhesion in IL-1 + IL-4-costimulated HUVEC but not IL-1-stimulated HUVEC; 2) removing {alpha}2,3-linked sialic acid without removing {alpha}2,6-linked sialic acid from IL-1 and IL-4-costimulated HUVEC did not affect VCAM-1-dependent adhesion; 3) {alpha}2,6-sialyltransferase was highly up-regulated in IL-1 + IL-4-costimulated HUVEC, whereas {alpha}2,3-sialyltransferase was not regulated by the cytokines; and 4) overexpression of {alpha}2,6-sialyltransferase inhibited VCAM-1-dependent adhesion in IL-1 + IL-4-costimulated HUVEC but not IL-1-stimulated HUVEC.

Although our results and others suggest that {alpha}2,6-linked sialic acids are common among glycoproteins expressed by HUVEC (22), little is known about their function. A great deal of evidence suggests that sialic acids support cell adhesion (18). Sialic acids and their related structures bind to numerous microbial lectins that support adhesion of microorganisms to their hosts (18). Selectins and sialoadhesin family molecules mediate cell-cell adhesion by binding to sialic acids and their related structures in a ligand and receptor-specific manner (18). Several reports indicate that {alpha}2,6-linked sialic acids may serve as a ligand for CD22, which is expressed by mature B cells to support adhesion of these lymphocytes (50, 51, 52). A recent study in ST6N-deficient mice suggests that ST6N is essential in immune responses of B cells (53). Ramos cells are a B cell line that is reported to express CD22, although unlike mature B cells, most of the CD22 in Ramos cells is present in the cytoplasm instead of on the cell surface (54). However, our previous report and the current study clearly indicate that the adhesion of Ramos cells to cytokine-stimulated HUVEC under both static and flow conditions is dependent exclusively on {alpha}4 integrins and VCAM-1 (8). There is no evidence that Ramos cells adhere to HUVEC in a CD22 and {alpha}2,6-linked sialic acid-dependent manner in our system, and in fact, overexpression of ST6N in HUVEC led to reduced Ramos cell adhesion to costimulated HUVEC.

A few reports suggest that sialic acids may inhibit intercellular interactions as well. For instance, the rosette formation of T cells with erythrocytes can be enhanced by removing sialic acids from T cells (19). The mucin-like molecule CD43 inhibits ß2 integrin-dependent adhesion of T cells to ICAM-1, and this inhibitory effect is in part attributable to sialic acids (21). Polysialic acid, a diverse polymer of sialic acids attached to neural cell adhesion molecules (NCAM), negatively regulates NCAM-dependent cell adhesion (20).

In eukaryotic cells, sialic acids are attached to terminal galactose or N-acetylgalactosamine with {alpha}2,3- or {alpha}2,6-linkages (55). Transfer of sialic acids and formation of these linkages are catalyzed by a family of sialyltransferases (55). Sialyltransferases transfer CMP-sialic acids to acceptor substrates, namely galactose and N-acetylgalactosamine, on carbohydrate chains that are attached to glycoproteins and glycolipids (55). So far, a number of sialyltransferases have been identified, which differ in specificities for acceptor substrates and linkages they create (55). In N-linked glycoproteins, sialic acids are attached to Galß1-4GlcNAc or Galß1-3GlcNAc with {alpha}2,6 and {alpha}2,3 linkages (55). In humans, ST6N (31, 45, 46), ST3N (25), and ST3ON (27) have been identified. ST6N creates {alpha}2,6 linkages to Galß1-4GlcNAc, whereas ST3N and ST3ON create {alpha}2,3 linkages to Galß1-4GlcNAc in N-linked glycoproteins. ST3N also uses Galß1-3GlcNAc as an acceptor substrate. In O-linked glycoproteins, sialic acids are attached to Galß1-3GalNAc and GalNAc. In humans, ST3O (26) and ST3ON (27) have been identified. Both ST3O and ST3ON catalyze the formation of {alpha}2,3 linkages to Galß1-3GalNAc. Although GalNAc {alpha}2,6-sialyltransferase (ST6O), which creates {alpha}2,6 linkages to GalNAc, has been cloned in rats (56), to our knowledge the human counterpart of this enzyme has not yet been identified.

Hanasaki et al. (22, 23) reported that the activity and mRNA levels of ST6N, which creates {alpha}2,6 linkages of sialic acids to N-linked and possibly O-linked glycoproteins, are up-regulated in HUVEC by IL-1, TNF, and IL-4. We found that either IL-1 or IL-4 up-regulates ST6N mRNA in HUVEC, and the combination of the two cytokines further enhances the expression of ST6N. Western blot analysis of whole cell lysates using MAA lectin (data not shown) and the binding of MAA lectin (see Fig. 2GoB) to the surfaces suggested that there are glycoproteins in HUVEC that are decorated with {alpha}2,3-linked sialic acids. Therefore, we sought to identify the enzyme(s) responsible for the formation of {alpha}2,3 linkages to glycoproteins in HUVEC. The only transcript identified by PCR was ST3ON, which uses both N- and O-linked glycoproteins as acceptor substrates (27). In contrast to ST6N, ST3ON mRNA was constitutively expressed in HUVEC and was not regulated by the cytokines we tested.

Sialidase isolated from A. ureafaciens cleaves all three linkages of sialic acids from carbohydrate chains (32). Removal of sialic acids from IL-1 + IL-4-costimulated HUVEC by sialidase increased VCAM-1-dependent adhesion under flow conditions, whereas the same treatment of IL-1-stimulated HUVEC did not. Although the level of ST6N in costimulated HUVEC was extremely high compared with IL-1 stimulation alone, the levels of total {alpha}2,6-linked sialic acids on the cell surfaces were similar. Therefore, the inhibitory effect of sialic acids on VCAM-1-dependent adhesion under flow conditions does not appear to be related solely to the total amount of sialic acids on the cell surfaces. In contrast with sialidase from A. ureafaciens, sialidase L cleaves {alpha}2,3-linked sialic acid without changing the level of {alpha}2,6-linked sialic acid, and removing {alpha}2,3-linked sialic acid alone from IL-1 + IL-4-costimulated HUVEC by using this enzyme did not increase VCAM-1-dependent adhesion. Therefore, {alpha}2,6-linked sialic acid is required for the inhibition of VCAM-1-dependent adhesion by IL-1 + IL-4 costimulation. To examine further whether increased expression of ST6N might be mechanistically related to the inhibition of VCAM-1-dependent adhesion, we overexpressed ST6N in HUVEC. We chose adenovirus-mediated gene transfer for the following reasons: 1) because HUVEC are primary cultured cells, stably transfected HUVEC cannot be established unless cells are transformed to become immortal; and 2) transient gene transfer by conventional methods is not efficient enough to determine the effect in a flow adhesion assay. Overexpression of ST6N inhibited Ramos cell adhesion only when ST6N was overexpressed in HUVEC costimulated with IL-1 + IL-4. These results suggest that to exhibit the inhibitory effect, {alpha}2,6-linked sialic acids must be associated with a molecule(s) that is inducible by IL-1 + IL-4 but not IL-1 alone.

VCAM-1 has seven potential N-linked glycosylation sites (57), and ~30% of the molecular weight of VCAM-1 expressed in HUVEC is attributed to N-linked carbohydrates (49). Our previous report demonstrated that VCAM-1 immunoprecipitated from HUVEC after stimulation with IL-1, IL-4, or IL-1 + IL-4 migrated similarly in SDS-PAGE (8). This result suggests that there is no marked difference in glycosylation among these VCAM-1 molecules, although the possibility of minor modification of carbohydrate chains could not be excluded. Because ST6N mRNA in IL-1 + IL-4-costimulated HUVEC was highly up-regulated and VCAM-1 molecules were decorated with {alpha}2,6-linked sialic acids, we examined the hypothesis that costimulation of HUVEC with IL-1 + IL-4 increases {alpha}2,6-linked sialic acids on VCAM-1, which hampers the ability of VCAM-1 to interact with {alpha}4 integrins under flow conditions. This hypothesis was not supported by the experimental results, because overexpression of ST6N in IL-1 + IL-4-costimulated HUVEC 1) did not increase {alpha}2,6-linked sialic acids on VCAM-1; 2) did reduce adhesion of Ramos cells to HUVEC; and 3) led to increased {alpha}2,6-linked sialic acids on other proteins. These results suggest that the inhibitory effect of {alpha}2,6-linked sialic acid may be due to the expression of {alpha}2,6-linked sialic acids on some other molecule(s) that remains to be identified.

There are reports that certain cell surface molecules are able to inhibit cell adhesion; the molecules that have so far been identified are mucins such as leukosialin (CD43) on T cells (21), episialin and epiglycanin on epithelial cells (58, 59), and polysialic acids, which are attached to NCAM on neuronal cells (20). These molecules have two features in common: 1) they are all very large, and 2) they have a high sialic acid content. One of the proposed mechanisms for sialic acid inhibition of cell adhesion is the negative charge, which generates repulsive forces (18). However, a negative charge per se does not explain the anti-adhesive effect in our experiments, because neither removing sialic acids from IL-1-stimulated HUVEC nor overexpressing them on IL-1-stimulated HUVEC had a significant effect on adhesion. In addition, most negative charges on endothelial cells are attributable to glycosaminoglycans rather than sialic acids (60). Because it is reported that the anti-adhesive effect of the mucins can be abrogated by capping these molecules with Abs (58, 59), spatial relationships between adhesion and anti-adhesion molecules appear to be critical in their anti-adhesive effects.

In summary, {alpha}2,6-linked sialic acids play an important role in the anti-adhesive mechanism by which costimulation of HUVEC with IL-1 + IL-4 negatively regulates leukocyte adhesion under flow conditions. We propose a novel inducible mechanism by which {alpha}2,6-linked sialic acids associated with an endothelial cell molecule(s) negatively regulates leukocyte adhesion under flow conditions. Identification of this molecule would provide new insight into the understanding of the regulation of leukocyte and endothelial cell interactions by cytokines.


   Acknowledgments
 
We thank Drs. Roy Lobb and Ivan Stamenkovic for valuable reagents, Kerrie Jara for editorial assistance, Drs. Priya Gopalan and Samina Kanwar for critical review of this manuscript, and Amy Turner and Nelson Bennett for technical assistance. We also thank Jestina Mason and the Child’s Health Research Center Core Laboratory, Baylor College of Medicine, for oligonucleotide synthesis and DNA sequencing, and Texas Women’s and Ben Taub General Hospitals for providing umbilical cords.


   Footnotes
 
1 This work was supported by National Institutes of Health Grants HL-42550 (to C.W.S. and C.M.B.), National Institutes of Health Grants ES-06091 and AI-19031 (to C.W.S), and an American Heart Association Established Investigator Award (to C.M.B.). Back

2 Address correspondence and reprint requests to Dr. Christie M. Ballantyne, Baylor College of Medicine, 6565 Fannin, M.S. A-601, Houston, TX 77030. E-mail address: Back

3 Abbreviations used in this paper: ST6N, Galß1-4GlcNAc {alpha}2,6-sialyltransferase; Ad26st, adenovirus carrying ST6N cDNA; Ad5, type 5 adenovirus; Adgal, adenovirus carrying ß-galactosidase cDNA; D-PBS, Dulbecco’s PBS; MAA, Maackia amurensis agglutinin; MOI, multiplicity of infection; NCAM, neural cell adhesion molecules; SNA, Sambucus nigra agglutinin; ST3N, Galß1-3(4)GlcNAc {alpha}2,3-sialyltransferase; ST3O, Galß1-3GalNAc {alpha}2,3-sialyltransferase; ST3ON, Galß1-4GlcNAc/Galß1-3GalNAc {alpha}2,3-sialyltransferase; X-Gal, 5-bromo-4-chloro-3-indolyl ß-D-galactoside. Back

Received for publication February 11, 1999. Accepted for publication June 14, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Springer, T. A.. 1995. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57:827.[Medline]
  2. Bochner, B. S., D. A. Klunk, S. A. Sterbinsky, R. L. Coffman, R. P. Schleimer. 1995. IL-13 selectively induces vascular cell adhesion molecule-1 expression in human endothelial cells. J. Immunol. 154:799.[Abstract]
  3. Kume, N., M. I. Cybulsky, Jr M. A. Gimbrone. 1992. Lysophosphatidylcholine, a component of atherogenic lipoproteins, induces mononuclear leukocyte adhesion molecules in cultured human and rabbit arterial endothelial cells. J. Clin. Invest. 90:1138.
  4. Khan, B. V., S. S. Parthasarathy, R. W. Alexander, R. M. Medford. 1995. Modified low density lipoprotein and its constituents augment cytokine-activated vascular cell adhesion molecule-1 gene expression in human vascular endothelial cells. J. Clin. Invest. 95:1262.
  5. Schmidt, A. M., O. Hori, J. X. Chen, J. F. Li, J. Crandall, J. Zhang, R. Cao, S. D. Yan, J. Brett, D. Stern. 1995. Advanced glycation endproducts interacting with their endothelial receptor induce expression of vascular cell adhesion molecule-1 (VCAM-1) in cultured human endothelial cells and in mice: a potential mechanism for the accelerated vasculopathy of diabetes. J. Clin. Invest. 96:1395.
  6. Jones, D. A., L. V. McIntire, C. W. Smith, L. J. Picker. 1994. A two-step adhesion cascade for T cell/endothelial cell interactions under flow conditions. J. Clin. Invest. 94:2443.
  7. Alon, R., P. D. Kassner, M. W. Carr, E. B. Finger, M. E. Hemler, T. A. Springer. 1995. The integrin VLA-4 supports tethering and rolling in flow on VCAM-1. J. Cell Biol. 128:1243.[Abstract/Free Full Text]
  8. Abe, Y., C. M. Ballantyne, C. W. Smith. 1996. Functions of domain 1 and 4 of vascular cell adhesion molecule-1 in {alpha}4 integrin-dependent adhesion under static and flow conditions are differentially regulated. J. Immunol. 157:5061.[Abstract]
  9. Kukreti, S., K. Konstantopoulos, C. W. Smith, L. V. McIntire. 1997. Molecular mechanisms of monocyte adhesion to interleukin-1ß-stimulated endothelial cells under physiologic flow conditions. Blood 89:4104.[Abstract/Free Full Text]
  10. Cybulsky, M. I., J. W. Fries, A. J. Williams, P. Sultan, V. M. Davis, Jr M. A. Gimbrone, T. Collins. 1991. Alternative splicing of human VCAM-1 in activated vascular endothelium. Am. J. Pathol. 138:815.[Abstract]
  11. Vonderheide, R. H., T. A. Springer. 1992. Lymphocyte adhesion through very late antigen 4: evidence for a novel binding site in the alternatively spliced domain of vascular cell adhesion molecule 1 and an additional {alpha}4 integrin counter-receptor on stimulated endothelium. J. Exp. Med. 175:1433.[Abstract/Free Full Text]
  12. Osborn, L., C. Vassallo, B. G. Browning, R. Tizard, D. O. Haskard, C. D. Benjamin, I. Dougas, T. Kirchhausen. 1994. Arrangement of domains, and amino acid residues required for binding of vascular cell adhesion molecule-1 to its counter-receptor VLA-4 ({alpha}4 ß1). J. Cell Biol. 124:601.[Abstract/Free Full Text]
  13. Paul, W. E., R. A. Seder. 1994. Lymphocyte responses and cytokines. Cell 76:241.[Medline]
  14. Thornhill, M. H., D. O. Haskard. 1990. IL-4 regulates endothelial cell activation by IL-1, tumor necrosis factor, or IFN-{gamma}. J. Immunol. 145:865.[Abstract]
  15. Masinovsky, B., D. Urdal, W. M. Gallatin. 1990. IL-4 acts synergistically with IL-1ß to promote lymphocyte adhesion to microvascular endothelium by induction of vascular cell adhesion molecule-1. J. Immunol. 145:2886.[Abstract]
  16. Thornhill, M. H., S. M. Wellicome, D. L. Mahiouz, J. S. Lanchbury, U. Kyan-Aung, D. O. Haskard. 1991. Tumor necrosis factor combines with IL-4 or IFN-{gamma} to selectively enhance endothelial cell adhesiveness for T cells: the contribution of vascular cell adhesion molecule-1-dependent and -independent binding mechanisms. J. Immunol. 146:592.[Abstract]
  17. Schauer, R., K. Sörge, G. Reuter, P. Roggentin, L. Shaw. 1995. Biochemistry and role of sialic acids. L. Shaw ed. Biology of the Sialic Acids 7. Plenum Press, New York.
  18. Varki, A.. 1997. Sialic acids as ligands in recognition phenomena. FASEB J. 11:248.[Abstract]
  19. Plunkett, M. L., M. E. Sanders, P. Selvaraj, M. L. Dustin, T. A. Springer. 1987. Rosetting of activated human T lymphocytes with autologous erythrocytes. Definition of the receptor and ligand molecules as CD2 and lymphocyte function-associated antigen 3 (LFA-3). J. Exp. Med. 165:664.[Abstract/Free Full Text]
  20. Acheson, A., J. L. Sunshine, U. Rutishauser. 1991. NCAM polysialic acid can regulate both cell-cell and cell-substrate interactions. J. Cell Biol. 114:143.[Abstract/Free Full Text]
  21. Ardman, B., M. A. Sikorski, D. E. Staunton. 1992. CD43 interferes with T-lymphocyte adhesion. Proc. Natl. Acad. Sci. USA 89:5001.[Abstract/Free Full Text]
  22. Hanasaki, K., A. Varki, I. Stamenkovic, M. P. Bevilacqua. 1994. Cytokine-induced ß-galactoside {alpha}-2,6-sialyltransferase in human endothelial cells mediates {alpha}2,6-sialylation of adhesion molecules and CD22 ligands. J. Biol. Chem. 269:10637.[Abstract/Free Full Text]
  23. Hanasaki, K., A. Varki, L. D. Powell. 1995. CD22-mediated cell adhesion to cytokine-activated human endothelial cells. Positive and negative regulation by {alpha}2-6-sialylation of cellular glycoproteins. J. Biol. Chem. 270:7533.[Abstract/Free Full Text]
  24. Abbassi, O., T. K. Kishimoto, L. V. McIntire, D. C. Anderson, C. W. Smith. 1993. E-selectin supports neutrophil rolling in vitro under conditions of flow. J. Clin. Invest. 92:2719.
  25. Kitagawa, H., J. C. Paulson. 1993. Cloning and expression of human Galß1,3(4)GlcNAc {alpha}2,3-sialyltransferase. Biochem. Biophys. Res. Commun. 194:375.[Medline]
  26. Kitagawa, H., J. C. Paulson. 1994. Differential expression of five sialyltransferase genes in human tissues. J. Biol. Chem. 269:17872.[Abstract/Free Full Text]
  27. Kitagawa, H., J. C. Paulson. 1994. Cloning of a novel {alpha}2,3-sialyltransferase that sialylates glycoprotein and glycolipid carbohydrate groups. J. Biol. Chem. 269:1394.[Abstract/Free Full Text]
  28. Chomczynski, P., N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156.[Medline]
  29. Sampath, R., G. L. Kukielka, C. W. Smith, S. G. Eskin, L. V. McIntire. 1995. Shear stress-mediated changes in the expression of leukocyte adhesion receptors on human umbilical vein endothelial cells in vitro. Ann. Biomed. Eng. 23:247.[Medline]
  30. Kumar, A. G., X. Y. Dai, C. A. Kozak, M. P. Mims, A. M. Gotto, C. M. Ballantyne. 1994. Murine VCAM-1: molecular cloning, mapping, and analysis of a truncated form. J. Immunol. 153:4088.[Abstract]
  31. Stamenkovic, I., H. C. Asheim, A. Deggerdal, H. K. Blomhoff, E. B. Smeland, S. Funderud. 1990. The B cell antigen CD75 is a cell surface sialyltransferase. J. Exp. Med. 172:641.[Abstract/Free Full Text]
  32. Uchida, Y., Y. Tsukada, T. Sugimori. 1979. Enzymatic properties of neuraminidases from Arthrobacter ureafaciens. J. Biochem. 86:1573.[Abstract/Free Full Text]
  33. Chou, M.-Y., S.-C. Li, M. Kiso, A. Hasegawa, Y.-T. Li. 1994. Purification and characterization of sialidase L, a NeuAc{alpha}2->3Gal-specific sialidase. J. Biol. Chem. 269:18821.[Abstract/Free Full Text]
  34. Shibuya, N., I. J. Goldstein, W. F. Broekaert, M. Nsimba-Lubaki, B. Peeters, W. J. Peumans. 1987. The elderberry (Sambucus nigra L.) bark lectin recognizes the Neu5Ac({alpha}2-6)Gal/GalNAc sequence. J. Biol. Chem. 262:1596.[Abstract/Free Full Text]
  35. Knibbs, R. N., I. J. Goldstein, R. M. Ratcliffe, N. Shibuya. 1991. Characterization of the carbohydrate binding specificity of the leukoagglutinating lectin from Maackia amurensis: comparison with other sialic acid-specific lectins. J. Biol. Chem. 266:83.[Abstract/Free Full Text]
  36. Gopalan, P. K., D. A. Jones, L. V. McIntire, and C. W. Smith. 1996. Cell adhesion under hydrodynamic flow conditions. In Current Protocols in Immunology. J. E. Coligan, D. H. Margulies, E. M. Shevac, and W. Strober, eds. John Wiley & Sons, New York, p. 7.29.21.
  37. Mizushima, S., S. Nagata. 1990. pEF-BOS, a powerful mammalian expression vector. Nucleic Acids Res. 18:5322.[Free Full Text]
  38. Morsy, M. A., E. L. Alford, A. Bett, F. L. Graham, C. T. Caskey. 1993. Efficient adenoviral-vector mediated ornithine transcarbamylase expression in deficient mouse and human hepatocytes. J. Clin. Invest. 92:1580.
  39. Graham, F. L., J. Smiley, W. C. Russell, R. Nairn. 1977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J. Gen. Virol. 36:59.[Abstract/Free Full Text]
  40. Bett, A. J., W. Haddara, L. Prevec, F. L. Graham. 1994. An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3. Proc. Natl. Acad. Sci. USA 91:8802.[Abstract/Free Full Text]
  41. Graham, F., L. Prevec. 1991. Manipulation of adenovirus vectors. L. ShawL. Prevec ed. Methods in Molecular Biology 109. Humana Press, Clifton, NJ.
  42. Graham, F. L.. 1992. Adenovirus-based expression vectors and recombinant vaccines. L. ShawL. PrevecF. L. Graham ed. Vaccines: New Approaches to Immunological Problems 363. Butterworth-Heinemann, Boston.
  43. O’Donovan, D. J., J. P. Katkin, T. Tamura, R. Husser, X. Xu, C. V. Smith, S. E. Welty. 1999. Gene transfer of mitochondrially targeted glutathione reductase protects H441 cells from t-butyl hydroperoxide-induced oxidant stresses. Am. J. Respir. Cell. Mol. Biol. 20:256.[Abstract/Free Full Text]
  44. Carlos, T. M., B. R. Schwartz, N. L. Kovach, E. Yee, M. Rosso, L. Osborn, G. Chi-Rosso, B. Newman, R. Lobb, J. M. Harlan. 1990. Vascular cell adhesion molecule-1 mediates lymphocyte adherence to cytokine-activated cultured human endothelial cells. Blood 76:965.[Abstract/Free Full Text]
  45. Grundmann, U., C. Nerlich, T. Rein, G. Zettlmeissl. 1990. Complete cDNA sequence encoding human ß-galactoside {alpha}-2,6-sialyltransferase. Nucleic Aci