Sulfation and sialylation requirements for a glycoform of CD34, a major endothelial ligand for L-selectin in porcine peripheral lymph nodes

Leukocyte recruitment from blood into peripheral lymph nodes is controlled in part by a specific interaction of lymphocyte-associated L-selectin with endothelial cell receptors known as peripheral addressins. In murine lymph nodes, two peripheral addressins have been identified, Gly-CAM-1, a 50 kDa molecule that also appears as a secreted form in plasma, and CD34, a 90 kDa membrane-associated sialomucin. A predominant 105 kDa CD34 mucin-like protein has also been identified in human tonsil as peripheral addressin. We have identified a 120 kDa sialomucin as the predominant peripheral addressin in porcine lymph nodes. Validation of the 120 kDa porcine molecule as a peripheral addressins was based on its ability to bind MECA-79, a monoclonal antibody previously used to isolate peripheral addressins from mouse and human tissues, and to bind an L-selectin-Fc chimera (LS-Fc). The binding with LS-Fc was abolished in the presence of fucoidin, a sulfated polysaccharide known to inhibit L-selectin-receptor interactions. To address the possibility that the 120 kDa ligand may contain common recognition determinants for MECA-79 and L-selectin, the requirements for sialylation and sulfation were compared. Whereas desialylation of 120 kDa ligand drastically reduced its binding to LS-Fc, this treatment appeared to enhance the binding of 120 kDa ligand to MECA-79. In contrast, the binding of both MECA-79 and LS-Fc to 120 kDa ligand was drastically reduced when de novo sulfation of this ligand was reduced by including chlorate, a metabolic inhibitor of sulfation, in the culture media. N-Terminal amino acid sequences of the porcine 120 kDa protein revealed homology with human CD34. Taken together, these findings suggest that the porcine 120 kDa peripheral addressin is an L-selectin-binding glycoform of CD34.

L-selectin regulates actin polymerisation via activation of the small G-protein Rac2

L-selectin mediated adhesion to endothelial cells is a crucial step in the immune response to pathogens (1, 2) and in lymphocyte homing (3, 4). Selectin molecules mediate leukocyte rolling on endothelial cells, the initial step of adhesion (5, 6). We have previously shown that stimulation of Jurkat T-lymphocytes via L-selectin results in activation of the p21Ras pathway and synthesis of reactive oxygen intermediates (7). Here, we show that cellular stimulation via L-selectin induces a change of cytoskeleton organisation demonstrated by a tenfold increase of actin filament polymerisation. This actin polymerisation is mediated by a Ras and Rac2 regulated pathway, since inhibition of Ras by transient transfection of transdominant inhibitory N17Ras or suppression of Rac2 protein expression by antisense oligonucleotides prevents L-selectin triggered actin polymerisation. Our results point to a signaling cascade from L-selectin via Ras and Rac2 to actin filaments, which might be important for leukocyte adhesion.

Studies on selectin blockers. 4. Structure-function relationships of sulfated sialyl Lewis X hexasaccharide ceramides toward E-, P-, and L-selectin binding

In order to clarify the real ligand structure of L-selectin proposed by Rosen et al., we first synthesized 6-sulfated sLe(x) hexasaccharide ceramide 1, 6'-sulfated sLe(x) hexasaccharide ceramide 2, and 6,6'-disulfated sLe(x) hexasaccharide ceramide 3 and examined their binding avidities for L-selectin. As a result, we found that the 6'-sulfated sLe(x) hexasaccharide ceramides 1-3 have similar binding avidities to L-selectin and their binding to L-selectin appeared somewhat stronger than that of sLe(x). For P-selectin, the sulfated sLe(x) derivatives 1-3 showed a similar avidity to sLe(x). On the other hand, 6-sulfated sLe(x) 2 was recognized to E-selectin and the binding avidity was apparently weak as compared to that of sLe(x) hexasaccharide ceramide. Surprisingly, 6'-sulfated and 6,6'-disulfated sLe(x)s 1 and 3 did not bind to E-selectin at all. We constructed the E-selectin-sLe(x) complex model and investigated the binding mode. Namely, the galactose 6'-position was directed toward the negatively charged residues, Glu80 and Asp100. Our results with E-selectin indicate that the replacement of 6'-OH position from anionic charged group to cationic charged one, e.g., amino groups, could have a marked affect on E-selectin recognition. These results could provide useful information for the drug design of selectin blockers.

L-selectin from human, but not from mouse neutrophils binds directly to E-selectin

L-Selectin on neutrophils as well as inducible E- and P-selectin on endothelium are involved in the recruitment of neutrophils into inflamed tissue. Based on cell attachment assays, L-selectin was suggested to function as a carbohydrate presenting ligand for E- and P-selectin. However, previous affinity isolation experiments with an E-selectin-Ig fusion protein had failed to detect L-selectin among the isolated E-selectin ligands from mouse neutrophils. We show here that L-selectin from human neutrophils, in contrast to mouse neutrophils, can be affinity-isolated as a major ligand from total cell extracts using E-selectin-Ig as affinity probe. Binding of human L-selectin to E-selectin was direct, since purified L-selectin could be reprecipitated with E-selectin-Ig. Recognition of L-selectin was abolished by sialidase-treatment, required Ca2+, and was resistant to treatment with endoglycosidase F. Binding of L-selectin to a P-selectin-Ig fusion protein was not observed. In agreement with the biochemical data, the anti-L-selectin mAb DREG56 inhibited rolling of human neutrophils on immobilized E-selectin-Ig but not on P-selectin-Ig. No such inhibitory effect was seen with the anti-mouse L-selectin mAb MEL14 on mouse neutrophils. Rolling of E-selectin transfectants on purified and immobilized human L-selectin was inhibited by mAb DREG56. We conclude that L-selectin on human neutrophils is a major glycoprotein ligand among very few glycoproteins that can be isolated by an E-selectin affinity matrix. The clear difference between human and mouse L-selectin suggests that E-selectin-binding carbohydrate moieties are attached to different protein scaffolds in different species.

Threshold levels of fluid shear promote leukocyte adhesion through selectins (CD62L,P,E)

Leukocyte adhesion through L-selectin to peripheral node addressin (PNAd, also known as MECA-79 antigen), an L-selectin ligand expressed on high endothelial venules, has been shown to require a minimum level of fluid shear stress to sustain rolling interactions (Finger, E.B., K.D. Puri, R. Alon, M.B. Lawrence, V.H. von Andrian, and T.A. Springer. 1996. Nature (Lond.). 379:266-269). Here, we show that fluid shear above a threshold of 0.5 dyn/cm2 wall shear stress significantly enhances HL-60 myelocyte rolling on P- and E-selectin at site densities of 200/microm2 and below. In addition, gravitational force is sufficient to detach HL-60 cells from P- and E-selectin substrates in the absence, but not in the presence, of flow. It appears that fluid shear-induced torque is critical for the maintenance of leukocyte rolling. K562 cells transfected with P-selectin glycoprotein ligand-1, a ligand for P-selectin, showed a similar reduction in rolling on P-selectin as the wall shear stress was lowered below 0.5 dyn/cm2. Similarly, 300.19 cells transfected with L-selectin failed to roll on PNAd below this level of wall shear stress, indicating that the requirement for minimum levels of shear force is not cell type specific. Rolling of leukocytes mediated by the selectins could be reinitiated within seconds by increasing the level of wall shear stress, suggesting that fluid shear did not modulate receptor avidity. Intravital microscopy of cremaster muscle venules indicated that the leukocyte rolling flux fraction was reduced at blood centerline velocities less than 1 mm/s in a model in which rolling is mediated by L- and P-selectin. Similar observations were made in L-selectin-deficient mice in which leukocyte rolling is entirely P-selectin dependent. Leukocyte adhesion through all three selectins appears to be significantly enhanced by a threshold level of fluid shear stress.

G(alpha)i2-deficient mice with colitis exhibit a local increase in memory CD4+ T cells and proinflammatory Th1-type cytokines

Mice with targeted deletion of the G protein G(alpha)i2 develop an inflammatory bowel disease closely resembling ulcerative colitis. To better define disease pathogenesis, the mucosal immune system in G(alpha)i2-deficient mice was studied. Phenotypic analysis of large intestine lamina propria lymphocytes revealed a large increase in memory CD4+ T cells (CD44high, CD45RBlow, CD62Llow). Furthermore, expression of the mucosal homing receptor integrin beta7 was increased on mucosal, but not systemic, CD4+ T cells. Analysis of cytokine production revealed a marked increase in proinflammatory Th1-type cytokines in inflamed colons, as compared with wild-type mice or G(alpha)i2-deficient mice without colitis. Thus, IFN-gamma and IL-1beta levels were increased 13-fold and 30-fold, respectively, with more modest increases in IL-6 levels (5-fold) and TNF levels (2-fold). Inflamed colons of G(alpha)i2-deficient mice also demonstrated increased IL-12 p40 mRNA levels. No increase in IL-2, IL-4, IL-5, and IL-10 was seen. Large intestinal epithelial cells in G(alpha)i2-deficient mice with colitis were found by immunohistochemistry to express increased levels of both MHC class I and class II Ags. Colitis was associated with increased IgG levels (60-fold increase), predominantly IgG2a (135-fold increase), in large but not small intestinal secretions. This was shown by ELISPOT analysis to result from local production within the lamina propria.

Neutrophil apoptosis is modulated by endothelial transmigration and adhesion molecule engagement

Termination of a neutrophil-mediated inflammatory response occurs through the activation of the endogenous cell death program, apoptosis. Neutrophil apoptosis is a constitutive process that can be accelerated or delayed by signals from the microenvironment. Since cellular localization at the site of an inflammatory challenge is the critical first step in a neutrophil response, we investigated the effects of neutrophil transendothelial transmigration on the kinetic expression of apoptosis. Neutrophils isolated from rat lung following challenge with LPS demonstrated a significant delay in spontaneous apoptosis. This delay was a consequence of transmigration, since a comparable delay was seen when TNF-alpha, a potent inducer of apoptosis in vitro, was used as the inflammatory stimulus. Human neutrophils demonstrated comparable delays in apoptosis in vitro following migration across an endothelial monolayer in response to FMLP. Delayed apoptosis only occurred in cells that had first been primed by LPS, a stimulus shown to up-regulate beta2 integrins and down-regulate L-selectin. Finally, crosslinking of CD11a or CD11b, but not of CD18, with mAbs and F(ab')2 fragments produced a delay in spontaneous apoptosis, whereas crosslinking of L-selectin with mAb or its natural ligand, sulfatides, accelerated the apoptotic process. Cells in which apoptosis was inhibited demonstrated persistent functional respiratory burst activity. These observations establish a role for endothelial transmigration in the regulation of neutrophil apoptosis, and suggest that adhesion molecules serve a modulatory role in the expression of neutrophil programmed cell death.

Polymorphonuclear leukocytes L-selectin expression decreases as they age in circulation

We recently reported that L-selectin expression increases on circulating polymorphonuclear leukocytes (PMN) during active bone marrow release, which suggests that older cells in the circulation have lower levels of L-selectin than those recently released from the bone marrow. The present study was designed to test the hypothesis that L-selectin expression reduces on PMN as they age in the circulation. In vitro studies using flow cytometry showed that PMN L-selectin decreased to 14.6 +/- 2.3% of baseline during a 24-h incubation at 37 degrees C. To test this hypothesis in vivo, rabbit PMN, labeled in vivo with 5'-bromo-2-deoxyuridine (PMN-brdU) in donor animals, were infused into recipient rabbits as whole blood (n = 5) and followed over 24 h in the circulation with a double immunolabeling technique. These results showed that the fraction of the L-selectin-negative PMN-BrdU in the circulation increased with time (P < 0.001), and nearly all of the PMNBrdU in the circulation were L-selectin negative after 24 h. Removal of L-selectin from the surface of PMN-BrdU with chymotrypsin before infusion did not change their rate of removal from the circulation (half-life or t 1/2 262 vs. 296 min, P = NS). We conclude that there is a continuous loss of L-selectin from PMN during their life span in the circulation, which supports the hypothesis that L-selectin expression decreases on PMN as they age.

Expression of an L-selectin ligand on hematopoietic progenitor cells

The process of hematopoiesis is dependent on discrete cell-cell and cell-matrix interactions which are tightly regulated by expression of adhesion molecules. L-selectin, an adhesion protein best known for regulating leukocyte attachment to endothelium, is characteristically expressed on the earliest hematopoietic progenitor cells. Ligands for L-selectin have been extensively characterized on endothelial cells. We recently identified a ligand for L-selectin expressed on the human hematopoietic progenitor cell line KG1a. This molecule is an integral membrane glycoprotein which is structurally different from all ligands previously described. We hypothesize that this molecule may mediate L-selectin-specific adhesive interactions during hematopoiesis. This article discusses the biology of L-selectin and its ligands, and reviews our current understanding of the structure and distribution of the L-selectin ligand expressed on hematopoietic cells.

Neutrophil CD18-dependent arrest on intercellular adhesion molecule 1 (ICAM-1) in shear flow can be activated through L-selectin

Neutrophil emigration through endothelial cells under shear flow involves several adhesion processes including cell rolling, arrest, and transmigration. Rolling is mediated by selectins, while arrest and transmigration both require activated CD18 integrins. One mode of CD18 activation is via selectins expressed on neutrophils and endothelial cells. We have recently reported that cross-linking of L-selectin (CD62L) resulted in the rapid activation of CD18-dependent adhesion. In the current study, we examine whether binding of E-selectin (CD62E) and L-selectin can activate neutrophil CD18-dependent adhesion under shear flow. Human ICAM-1 (CD54) and E-selectin were co-transfected into L cells. Neutrophil capture, rolling, and arrest on these monolayers were quantitated in a parallel plate flow chamber at a wall shear stress of 2.0 dyne/cm2. Under these conditions, E-selectin supported cell capture and rolling on the monolayer, but did not trigger CD18-mediated cell arrest within 200 microm of rolling. However, when neutrophils were treated with anti-L-selectin mAb and cross-linked with a secondary mAb, approximately 50% of the cells arrested within 54 microm. Cell arrest was also observed in response to IL-8 stimulation. A subthreshold level of IL-8 in combination with L-selectin cross-linking potentiated the level of cell arrest due to either stimulus alone. The transition to cell arrest involved both LFA-1 (CD11a/CD18) and Mac-1 (CD11b/CD18). Blocking either subunit alone failed to reduce arrest, while blocking both molecules with mAbs reduced the number to baseline levels. These data support the conclusion that L-selectin, but not E-selectin, can signal the transition from neutrophil rolling to cell arrest under shear flow.

Expression of beta 2-integrins and L-selectin on polymorphonuclear leukocytes in septic patients

Adhesion molecules on polymorphonuclear leukocytes (PMNL) play an important role in nonspecific defense mechanisms directed at invading microorganisms. When local infection, however, cannot be controlled, a systemic inflammatory response syndrome (SIRS) ensues which may progress to septic shock and multiple organ failure, these being major determinants of the patient's outcome. In the present study, the expression of beta 2-integrins and L-selectin on blood PMNL was measured on subsequent days in patients with sepsis (n = 17) and in healthy volunteers (n = 15). beta 2-Integrins and L-selectin molecules were detected by flow cytometry, using the monoclonal antibodies IB4 (anti-CD18) and Dreg200 (anti-CD62L), respectively. Adhesion molecules were determined at baseline immediately after blood collection and also 45 min after incubation of cells in vitro at body temperature to allow for spontaneous regulation. In addition, PMNL were activated by receptor-dependent and receptor-independent stimuli to characterize stimulus-specific adhesion molecule expression. In parallel with the measurement of adhesion molecules, severity of sepsis was assessed by the Elebute score. The results demonstrate significant differences in the basal, spontaneous and stimulus-induced expression of adhesion molecules between healthy volunteers, survivors (n = 11) and nonsurvivors (n = 6). Moreover, when survivors and nonsurvivors with severe sepsis (Elebute score > 12) were compared, basal expressions of both beta 2-integrins and L-selectin were significantly lower in patients who did not survive. Thus, measurement of adhesion molecules on circulating PMNL may be useful to identify septic patients at high risk for lethal outcome.

Use of receptors immobilized on microspheres to identify ligand binding sites in tissue sections: detection of lymph node ligands for L-Selectin

Particulate microspheres bearing immobilized probes can be used to identify ligands expressed by cells and require only brightfield microscopy for detection. There are distinct advantages to using microspheres to detect low affinity interactions; microspheres require no secondary amplification or detection procedures subsequent to the binding interaction, reducing opportunities for detachment of bound probe, and concentrating probes on microspheres may greatly increase binding avidity. Selectin leukocyte-endothelial adhesion molecules undergo low affinity binding to ligands, and these interactions may be difficult to detect with standard techniques. The aim of this study was to determine if immobilizing recombinant L-Selectin on microspheres would facilitate detection of specific tissue ligands. Microspheres were incubated with sections of rabbit peripheral lymph node in a modified Stamper-Woodruff assay, and binding was assessed by brightfield microscopy. L-Selectin-IgG microspheres bound to high endothelial venules, known to be sites of expression for L-Selectin ligands. Specificity was indicated by the lack of binding of microspheres coated with control protein, and inhibition of binding by antibody to L-Selectin and by competitive antagonists of L-Selectin ligand interactions.

The faster kinetics of L-selectin than of E-selectin and P-selectin rolling at comparable binding strength

Selectins are a family of lectins that mediate tethering and rolling of leukocytes on endothelium in vascular shear flow. To test the hypothesis that the kinetics and the strength of rolling interactions can be independently varied for different selectin:ligand pairs, we have directly compared all three selectins with regard to distinct measures of selectin-mediated interactions in shear flow: tethering, rolling velocity, and strength of rolling adhesions. At comparable site densities of E-selectin, P-selectin, and the L-selectin counter-receptor CD34, neutrophils tethered with similar efficiency and developed rolling adhesions of similar strength as measured by resistance to detachment. Under the same conditions, neutrophils rolled 7.5- to 10.5-fold faster on CD34 than on E-selectin and P-selectin. These findings suggest that the kinetics of bond dissociation and bond formation are faster for L-selectin than for E- and P-selectin. We also compared the behavior of neutrophils and lymphocytes on the same selectin. Both cell types showed comparable strength of binding to CD34; however, neutrophils rolled with faster velocities than lymphocytes.

L-selectin activates the Ras pathway via the tyrosine kinase p56lck

Selectins mediate rolling, the initial step of leukocyte adhesion to endothelial cells [Springer, T. A. (1995) Annu. Rev. Physiol. 57, 827-872 and Butcher, E. C. (1991) Cell 67, 1033-1036]. In this study we show that L-selectin triggering of Jurkat cells using different antibodies or glycomimetics resulted in activation of the src-tyrosine kinase p56lck; tyrosine phosphorylation of intracellular proteins, in particular mitogen-activating protein kinase and L-selectin; and association of Grb2/Sos with L-selectin. This association correlated with an activation of p21Ras, mitogen-activating protein kinase, Rac2, and a transient increase of 2-O synthesis. Stimulation of the Ras pathway by L-selectin requires functional p56lck, since p56lck-deficient Jurkat cells (JCaM1.6) do not show tyrosine phosphorylation, association of L-selectin with Grb2/Sos, and activation of Ras upon L-selectin triggering. Transfection of JCaM1.6 cells with p56lck reconstitutes the observed signaling events. Genetic inhibition of Ras or Rac2 prevented Rac2 stimulation and 2-O synthesis, respectively. The specificity and the physiological significance of the observed signaling cascade is indicated by stimulation of L-selectin-transfected P815, L-selectin-positive CEM or peripheral blood lymphocytes resulting in the same activation events as in Jurkat cells. Our results point to a signaling cascade from L-selectin via p56lck, Grb2/Sos, Ras, and Rac2 to 2-O.

DNA aptamers block L-selectin function in vivo. Inhibition of human lymphocyte trafficking in SCID mice

Selectins participate in the initial events leading to leukocyte extravasation from the blood into tissues. Thus the selectins have generated much interest as targets for antiinflammatory agents. Therapeutic molecules based on the monomeric carbohydrate ligand sialyl Lewis X (SLe(X)) have low affinities and are not specific for a given selectin. Using SELEX (Systematic Evolution of Ligands by EXponential Enrichment) technology, we have generated aptamers specific for L-selectin that require divalent cations for binding and have low nanomolar affinity. In vitro, the deoxyoligonucleotides inhibit L-selectin binding to immobilized SLe(X) in static assays and inhibit L-selectin-mediated rolling of human lymphocytes and neutrophils on cytokine-activated endothelial cells in flow-based assays. These aptamers also block L-selectin-dependent lymphocyte trafficking in vivo, indicating their potential utility as therapeutics.

Protease-resistant L-selectin mutants. Down-modulation by cross-linking but not cellular activation

The adhesion molecule L-selectin (CD62L) is rapidly shed from the plasma membrane during leukocyte activation as a result of proteolytic cleavage between Lys321 and Ser322 within the extracellular domain. L-selectin is also down-modulated from the surface in response to cross-linking, possibly through a similar mechanism. To further characterize the mechanism of down-modulation, several L-selectin mutants were generated and transfected into COS cells. Wild-type L-selectin as well as mutants with one or two amino acid substitutions at the cleavage site were nearly quantitatively shed into the culture supernatant. However, mutants in which a nine-amino acid stretch that included the protease-sensitive site was either deleted or replaced with a polyglycine spacer or a comparable region of E-selectin were retained on the cell surface and not detected in the supernatant. These results are consistent with other reports describing protease resistant L-selectin mutants. We also demonstrate that when expressed in L1-2 pre-B cells, the L-selectin nine-amino acid deletion mutant (321del.9), but not wild-type L-selectin, is resistant to down-regulation induced by PMA. However, both wild-type and mutant 321del.9 are completely lost from the cell surface in response to cross-linking with an L-selectin Ab. PMA-induced- but not L-selectin cross-linking-induced down-modulation was inhibited by staurosporine. These data are consistent with the idea that the L-selectin protease(s) can tolerate minor structural alterations at the cleavage site, and that L-selectin down-modulation can be induced by more than one mechanism, at least one of which (cross-linking) is protein kinase C independent.

Polysulfated derivatives of beta-cyclodextrin and myo-inositol as potent inhibitors of the interaction between L-selectin and peripheral addressin: implying a requirement for highly clustered sulfate groups

We have utilized an in vitro assay that measures the binding of an L-selectin-human Fc chimera (LS-Fc) to [35S]sulfate labelled peripheral addressin (PNAd), a 120 kDa glycoprotein ligand for L-selectin in porcine lymph nodes, to evaluate inhibitory properties of a small group of sulfated derivatives of beta-cyclodextrin (beta-CD), sLe(x) and myo-inositol to their non-sulfated counterparts were studied. We found that hepta-sulfated beta-CD (IC50 = 0.2 mM) strongly inhibited the binding of L-selectin to PNAd. In contrast, the monosulfated beta-CD was a poor inhibitor, displaying < 10% inhibition at 0.5 mM and beta-CD was not active as an inhibitor. Similarly, inositol hexakissulfate, a compound containing six sulfate groups on the inositol ring displayed an inhibition of about 61% at 0.5 mM concentration, whereas the non-sulfated myoinositol was not inhibitory. These findings provide evidence that clustering of sulfate groups enhances affinity of molecules for binding to L-selectin.

Targeting of active rat alpha 2,3-sialyltransferase to the yeast cell wall by the aid of the hsp 150 delta-carrier: toward synthesis of sLe(x)-decorated L-selectin ligands

Interactions between selectins and their oligosaccharide-decorated ligands play a crucial role in the initiation of leukocyte extravasation. We have shown that synthetic multivalent sialyl Lewis x glycans inhibit strongly the adhesion of lymphocytes to endothelium at sites of inflammation. However, enzyme-assisted synthesis of these oligosaccharides si hampered by the lack of sufficient amounts of specific glycosyltransferases. We report here the construction of Saccharomyces cerevisiae strains expressing the soluble catalytic ectodomain of rat Gal(beta)1-3/4GlcNac alpha 2,3-sialyltransferase (ST3Ne) fused to the C-terminus of the hsp150 delta-carrier polypeptide. The hsp150 delta-carrier, which is an N-terminal fragmented of a natural secretory protein of yeast, is able to confer secretion-competence to several heterologous proteins, which otherwise remain in the yeast endoplasmic reticulum. The ST3Ne portion of the hsp 150 delta-ST3Ne fusion protein adopted an enzymatically active conformation and was N-glycosylated and disulfide-bonded. Hsp150 delta-ST3Ne was secreted with a half-time of about 7.5 min and remained intercalated in the cell wall, which covers the yeast plasma membrane. About 110 mU of sialyltransferase per litre was produced in 16 h. Whole live yeast cells were able to transfer sialic acid from CMP-NeuNAc to N-acetyllactosamine yielding alpha 2,3-sialyl-N-acetyllactosamine, as evidenced by paper chromatography, cleavage by linkage-specific sialidase, and NMR analysis. Our data suggest that yeast cells externalizing mammalian glycosyltransferases with the aid of the hsp150 delta-carrier could provide a source of enzymes for synthesis of valuable oligosaccharides.

Decreased neutrophil L-selectin expression in patients with systemic inflammatory response syndrome

OBJECTIVE

To determine whether the expression of L-selectin is reduced in patients with systemic inflammatory response syndrome (SIRS) compared with control subjects, to test the effect of exudation of neutrophil L-selectin expression, and to measure soluble serum concentrations of shed L-selectin in the two groups.

DESIGN

Prospective study.

SETTING

Intensive care unit in a tertiary care hospital (Royal Victoria Hospital, Montreal).

PATIENTS

Twenty-five patients with SIRS and 20 healthy, age-matched controls.

INTERVENTIONS

Collection of exudate neutrophils from skin window and circulating neutrophils from venous blood.

OUTCOME MEASURES

Neutrophil L-selectin levels, measured with the use of fluorocytometry, and soluble L-selectin levels, measured with the use of enzymelinked immunosorbent assay.

RESULTS

There is a significant reduction in L-selectin expression on circulating neutrophils in patients with SIRS, compared with control subjects. Exudation of neutrophils to an extravascular site resulted in a dramatic down-regulation of L-selectin in both groups. Serum levels of soluble L-selectin were higher in patients with SIRS than in control subjects.

CONCLUSIONS

The loss of L-selectin may be partly responsible for reduced neutrophil exudation to extravascular sites in patients with SIRS.

[Impaired neutrophil function in chronic renal failure--dysregulation of surface adhesion molecule expression and phagocytosis]

To elucidate the impaired neutrophil function in patients with chronic renal failure, we analyzed the expression of the adhesion molecules, LAM-1, LFA-1, Mac-1, gp150/95 and phagocytosis activity of neutrophils in predialysis and hemodialysis patients by flow cytometry. Further, the response to granulocyte colony-stimulating factor (G-CSF), N-formyl-methionyl-leucyl-phenylalanine (fMLP) and tumor necrosis factor alpha (TNF alpha) were investigated. In hemodialysis patients, the expression of LAM-1 was decreased and that of MAC-1 was increased, indicating the activation of neutrophils. Also in predialysis patients, the same condition of "low LAM-1, high MAC-1" was observed, but to a lesser degree. Phagocytosis activity was significantly decreased in hemodialysis patients, whereas the neutrophils of predialysis patients showed almost the same phagocytosis activity compared to the controls. The responses to G-CSF, fMLP, TNF alpha were significantly reduced both in hemodialysis and predialysis patients. The inadequate activation of neutrophils and impaired response to stimulation may play an important role in uremic patients with regard to increased susceptibility to infections.

Inhibitory action of sulfatide, a putative ligand for L-selectin, on B cell proliferation and Ig production

The interaction of L-selectin and its ligand is widely accepted to mediate leukocyte rolling and adhesion on the endothelial surface. Although L-selectin is ubiqultously expressed on lymphoid cells, its role in execution of lymphocyte functions is unknown. By flow cytometric analysis using mAb specific for sulfatide, a putative ligand for L-selectin, we found that sulfatide was selectively expressed on B cells, but not on T cells. To elucidate the involvement of L-selectin and its ligand in B cell activation, the present study was undertaken to investigate effects of sulfatide on T cell-dependent and -independent Ig production by B cells. In pokeweed mitogen-stimulated cultures, addition of sulfatide resulted in almost complete inhibition of Ig production by B cells in the presence of memory CD4+ T cells, whether L-selectin-positive or -negative. A similar inhibition of Ig production by sulfatide was found when B cells were stimulated with Staphylococcus aureus Cowan I and IL-2. Unlike sulfatide, a desulfated form of sulfatide, galactosylaceramide, did not show any effects on Ig production by B cells. Maximal inhibition of Ig production was observed when sulfatide was added at the early period of culture. Sulfatide suppressed effectively proliferation of B cells, but not of T cells. Sulfatide competed the binding of anti-L-selectin mAb to B cells, suggesting it could interfere B cell activation by blocking L-selectin function. The results suggest a novel role of the L-selectin/its ligand system in the initiation of B cell activation.

L-selectin-carbohydrate interactions: relevant modifications of the Lewis x trisaccharide

Protein-carbohydrate interactions are known to mediate cell-cell recognition and adhesion events. Specifically, three carbohydrate binding proteins termed selectins (E-, P-, and L-selectin) have been shown to be essential for leukocyte rolling along the vascular endothelium, the first step in the recruitment of leukocytes from the blood into inflammatory sites or into secondary lymphoid organs. Although this phenomenon is well-established, little is known about the molecular-level interactions on which it depends. All three selectins recognize sulfated and sialylated derivatives of the Lewis x [Le(x):Gal beta 1-->4(Fuc alpha 1-->3)GlcNAc] and Lewis a [Le(a): Gal beta 1-->3(Fuc alpha 1-->4)GlcNAc] trisaccharide cores with affinities in the millimolar range, and it is believed that variants of these structures are the carbohydrate determinants of selectin recognition. Recently it was shown that the mucin GlyCAM-1, a secreted physiological ligand for L-selectin, is capped with sulfated derivatives of sialyl Lewis x [sLe(x): Sia alpha 2-->3Gal beta 1-->4(Fuc alpha 1-->3)GlcNAc] and that sulfation is required for the high-affinity interaction between GlyCAM-1 and L-selectin. To elucidate the important sites of sulfation on Le(x) with respect to L-selectin recognition, we have synthesized six sulfated Le(x) analogs and determined their abilities to block binding of a recombinant L-selectin-Ig chimera to immobilized GlyCAM-1. Our results suggest that 6-sulfo sLe(x) binds to L-selectin with higher affinity than does sLe(x) or 6'-sulfo sLe(x) and that sulfation of sLe(x) capping groups on GlyCAM-1 at the 6-position is important for L-selectin recognition.

Sialylated, fucosylated ligands for L-selectin expressed on leukocytes mediate tethering and rolling adhesions in physiologic flow conditions

Interaction of leukocytes in flow with adherent leukocytes may contribute to their accumulation at sites of inflammation. Using L-selectin immobilized in a flow chamber, a model system that mimics presentation of L-selectin by adherent leukocytes, we characterize ligands for L-selectin on leukocytes and show that they mediate tethering and rolling in shear flow. We demonstrate the presence of L-selectin ligands on granulocytes, monocytes, and myeloid and lymphoid cell lines, and not on peripheral blood T lymphocytes. These ligands are calcium dependent, sensitive to protease and neuraminidase, and structurally distinct from previously described ligands for L-selectin on high endothelial venules (HEV). Differential sensitivity to O-sialo-glycoprotease provides evidence for ligand activity on both mucin-like and nonmucin-like structures. Transfection with fucosyltransferase induces expression of functional L-selectin ligands on both a lymphoid cell line and a nonhematopoietic cell line. L-selectin presented on adherent cells is also capable of supporting tethering and rolling interactions in physiologic shear flow. L-selectin ligands on leukocytes may be important in promoting leukocyte-leukocyte and subsequent leukocyte endothelial interactions in vivo, thereby enhancing leukocyte localization at sites of inflammation.

Increased plasma GlyCAM-1, a mouse L-selectin ligand, in response to an inflammatory stimulus

GlyCAM-1 (glycosylation-dependent cell adhesion molecule-1) is one of the sialomucin-like ligands for L-selectin, which is a member of the selectin family and mediates initial adhesion of leukocytes to specialized high endothelial venules in lymph nodes and venules at sites of inflammation. GlyCAM-1, lacking a transmembrane domain, is supposed to be secreted into the blood. To understand the functional role of secreted GlyCAM-1, we performed sandwich enzyme-linked immunosorbent assay to measure GlyCAM-1 plasma levels after inflammatory stimulus. BALB/c mice were injected with complete Freund's adjuvant (CFA) in the hind footpads; serum levels of GlyCAM-1 and L-selectin bound to GlyCAM-1 and several inflammatory cytokines, including interleukin-6 (IL-6), were measured at various intervals. IL-6 showed a significant increase 3 h after CFA stimulation. GlyCAM-1 was increased at 3 h, reached peak levels at 12 h, and gradually decreased thereafter. Levels of L-selectin bound to the plasma GlyCAM-1 changed over a similar time course, reached peak at 12 h after, and then began to decrease. The binding of L-selectin to plasma GlyCAM-1 was completely eliminated with the presence of ethyleneglycol-bis(beta-aminoethylether)-N,N'-tetraacetic acid, showing the calcium dependency of this binding. These findings show that GlyCAM-1 release is enhanced by inflammatory stimulation and also suggest that released plasma GlyCAM-1 may trap, at least in part, soluble L-selectin shed from stimulated leukocytes to neutralize each other.

L-selectin crosslinking induces integrin-dependent adhesion: evidence for a signaling pathway involving PTK but not PKC

L-selectin mediates the initial contact of leukocytes with the endothelium prior to extravasation. Here we demonstrate that L-selectin engagement can induce rapid and avid integrin-dependent T cell adhesion to recombinant immobilized cell adhesion molecules (CAMs) including ICAM-1, ICAM-3, and VCAM-1, as well as to the extracellular matrix protein fibronectin (FN). L-selectin-induced adhesion to these integrin ligands shares characteristics with CD3 mAb- or phorbol ester-induced adhesion in requiring metabolic energy, tyrosine kinase and ligand-stimulated Ca2+ channel activity. However, L-selectin-induced adhesion is distinct from that induced by phorbol ester or CD3 crosslinking in being relatively independent of protein kinase C (PKC) activity and actin polymerization. Consistent with the higher levels of L-selectin expression on CD45RA+ (naive) cells, L-selectin crosslinking induces a greater proportion of naive relative to memory cell binding to CAMs and FN. In contrast, exposure to phorbol ester or CD3 crosslinking is more effective in inducing CD45RO+ (memory) cell adhesion. Thus, in addition to its role in leukocyte capture and rolling on the endothelium, L-selectin may contribute to beta 1 and beta 2 integrin-dependent binding and arrest.

6'-Sulfo sialyl Lex but not 6-sulfo sialyl Lex expressed on the cell surface supports L-selectin-mediated adhesion

In order to determine if a sulfated oligosaccharide on the cell surface can function as an L-selectin ligand, a novel approach for in vitro transfer of oligosaccharides was utilized (Srivastava, G., Kaun, K. J., Hindsgaul, O., and Palcic, M. M. (1992) J. Biol. Chem. 267, 22356-22361). CHO cells were incubated with synthetic 6'-sulfo sialyl Lex, NeuNAcalpha2-->3(sulfate-6)Galbeta1-->4(Fucalpha1-->3) GlcNAc or 6-sulfo sialyl Lex, NeuNAcalpha2-->3Galbeta1-->4[(Fucalpha1-->3)sulfate--> 6GlcNAc] oligosaccharide linked to C-6 of a fucose residue in GDP-fucose and a milk fucosyltransferase. The resultant CHO cells expressing 6'-sulfo sialyl Lex or 6-sulfo sialyl Lex on their cell surface were tested for adhesion to E-selectin and L-selectin chimeric proteins coated on plates. The results indicate that 6'-sulfo sialyl Lex supports L-selectin-mediated adhesion much better than sialyl Lex similarly tagged on the cell surface. In contrast, 6-sulfo sialyl Lex containing a sulfate group on the N-acetylglucosamine residue did not support adhesion with either selectin. These combined results suggest that 6'-sulfo sialyl Lex is a much better ligand than sialyl Lex oligosaccharide for L-selectin.

Synthesis and biological activities of three sulfated sialyl Le(x) ganglioside analogues for clarifying the real carbohydrate ligand structure of L-selectin

Sulfated sialyl Le(x) ganglioside analogues at C-6 of D-galactose, N-acetyl-D-glucosamine, and of both D-galactose and N-acetyl-D-glucosamine residues have been synthesized, in order to clarify the structure of the real carbohydrate ligand of L-selectin. Coupling of the suitably protected N-acetyl-D-glucosaminyl-beta (1-->3)-lactose derivatives 13 and 16 with the sialyl alpha(2-->3)-D-galactopyranosyl trichloroacetimidates 10 and 12 (glycosyl donors), via glycosylation of 2-(trimethylsilyl)ethyl 4,6-O-benzylidene-beta-D-galactopyranoside (1) with the phenyl 2-thioglycoside derivative (2) of N-acetylneuraminic acid (Neu5Ac) using N-iodosuccinimide/TfOH, O-benzoylation, removal of the benzylidene group affording 5, selective 6-O-levulinoylation, O-benzoylation, removal of the 2-(trimethylsilyl)ethyl group, and imidate formation, or via O-acetylation of 5, removal of the 2-trimethylsilyl)ethyl group, then imidate formation, gave the pentasaccharides 18-20. The glycosylation of the pentasaccharide acceptors (21-23) derived from 18-20 by removal of the 4-methoxybenzyl group, with phenyl 1-thioglycoside derivative 27 of L-fucose using dimethyl(methylthio)sulfonium triflate (DMTST) afforded the corresponding hexasaccharides 28-30, which were transformed in good yields, via reductive removal of their benzyl groups, O-acetylation, selective removal of the 2-(trimethylsilyl)ethyl group, imidate formation, coupling with (2S,3R,4E)-2-azido-O-benzoyl-4-octadecene-1,3-diol (35) in the presence of boron trifluoride etherate, selective reduction of the azido group, coupling with octadecanoic acid, selective removal of the levulinoyl groups, treatment with sulfur trioxide-pyridine complex, then removal of the protecting groups, into the desired sulfated sialyl Le(x) ganglioside analogues 50-52.

L-selectin binds to P-selectin glycoprotein ligand-1 on leukocytes: interactions between the lectin, epidermal growth factor, and consensus repeat domains of the selectins determine ligand binding specificity

The selectins mediate cellular interactions by binding carbohydrate determinants present on a limited number of glycoprotein ligands. L-selectin binds multiple ligands expressed on endothelial cells, while P-selectin interacts exclusively with P-selectin glycoprotein ligand-1 (PSGL-1) on leukocytes. In this study, L-selectin was shown to bind leukocytes through the P-selectin ligand, PSGL-1, although at lower levels than P-selectin. L-selectin binding to PSGL-1 is specific since it was blocked by Abs to L-selectin or PSGL-1, required appropriate glycosylation of PSGL-1, and was Ca2+ dependent. The contributions of the extracellular domains of the selectins to ligand binding was assessed using a panel of chimeric selectins created by exchange of domains between L-selectin and P- or E-selectin. The lectin and epidermal growth factor domains of L- and P-selectin contributed significantly to binding through similar, if not identical, regions of PSGL-1. The different chimeric selectins revealed that the lectin domain was the dominant determinant for ligand binding, while cooperative interactions between the lectin, epidermal growth factor, and short consensus repeat domains of the selectins also modified ligand binding specificity. L-selectin binding to PSGL-1 expressed by leukocytes may mediate neutrophil rolling on stationary leukocytes bound to cytokine-induced endothelial cells, which was previously reported to be a L-selectin-dependent process.

Interactions through L-selectin between leukocytes and adherent leukocytes nucleate rolling adhesions on selectins and VCAM-1 in shear flow

We demonstrate an additional step and a positive feedback loop in leukocyte accumulation on inflamed endothelium. Leukocytes in shear flow bind to adherent leukocytes through L-selectin/ligand interactions and subsequently bind downstream and roll on inflamed endothelium, purified E-selectin, P-selectin, L-selectin, VCAM-1, or peripheral node addressin. Thus adherent leukocytes nucleate formation of strings of rolling cells and synergistically enhance leukocyte accumulation. Neutrophils, monocytes, and activated T cell lines, but not peripheral blood T lymphocytes, tether to each other through L-selectin. L-selectin is not involved in direct binding to either E- or P-selectin and is not a major counterreceptor of endothelial selectins. Leukocyte-leukocyte tethers are more tolerant to high shear than direct tethers to endothelial selectins and, like other L-selectin-mediated interactions, require a shear threshold. Synergism between leukocyte-leukocyte and leukocyte-endothelial interactions introduces novel regulatory mechanisms in recruitment of leukocytes in inflammation.

GlyCAM-1, a physiologic ligand for L-selectin, activates beta 2 integrins on naive peripheral lymphocytes

Naive T cells are selectively recruited from the blood into peripheral lymph nodes during lymphocyte recirculation. L-selectin, a lectin-like receptor, mediates the initial attachment of lymphocytes to high endothelial venules (HEV) in lymph nodes. A subsequent step involving the activation of beta 2 integrins has been proposed to facilitate firm adhesion, but the activating signals are poorly understood. We report here that either antibody-mediated cross-linking of L-selectin on human lymphocytes or treatment of the cells with GlyCAM-1, an HEV-derived, secreted ligand for L-selectin, stimulates their binding to ICAM-1 through the beta 2 integrin pathway. Furthermore, GlyCAM-1 causes the rapid expression of a neoepitope on beta 2 integrins associated with a high-avidity state. Naive (CD45RA+), but not memory (CD45R0+) lymphocytes, respond to L-selectin cross-linking or GlyCAM-1 treatment. Thus, the complexing of L-selectin by specific ligands may provide key signals to naive lymphocytes, contributing to their selective recruitment into peripheral lymphoid organs.

Antigen-driven shedding of L-selectin from human gamma delta T cells

Activation of lymphocytes leads to the modulation of a number of surface molecules. We have investigated the expression of one such molecule. L-selectin, following activation of gamma delta T cells with Mycobacterium tuberculosis. L-selectin is modulated during lymphocyte entry into lymph nodes; this modulation reflects the recirculation and homing potential of lymphocytes. We find that stimulation of gamma delta T cells by M. tuberculosis antigens results in shedding of L-selectin from gamma delta T cells. Re-expression of L-selectin occurs on removal of antigen suggesting that the regulation of expression is controlled by the presence or absence of antigen. The gamma delta T-cell receptor (TCR)-positive, L-selectin negative population of peripheral blood lymphocytes appears to be resting cells, as assessed by forward- and light-scatter analysis. We further find that gamma delta T cells isolated from a site of infection, the pleural fluid of a tuberculosis patient, are L-selectin negative, and that L-selectin is re-expressed following culture of the pleural fluid gamma delta T cells in the absence of antigen. These results demonstrate that, in addition to stimulation with polyclonal mitogens, antigen stimulation can also promote the surface shedding of L-selectin and that gamma delta T cells have the potential to home to sites of infection supporting their role in the immunological defence against infectious micro-organisms.

Phenotype, and migration properties of three major subsets of tissue homing T cells in sheep

T cells show a bias in their migration pathways: some T cells preferentially migrate to peripheral lymph nodes (LN), some to mucosal tissues, and some to peripheral tissues such as skin. These recirculation pathways were examined in sheep by collecting lymph draining into and out of peripheral and intestinal LN, and using fluorescent dyes to trace the recirculation of the lymph cells. Monoclonal antibodies (mAb) to alpha 4, beta 1, and beta 7 integrins, and L-selectin, were used to define three major populations of recirculating T cells. Naive-type T cells (L-selectin+, alpha 4 beta 1lo beta 7lo) migrated preferentially through peripheral LN. Two memory populations could be defined: alpha 4 beta 1hi beta 7- and alpha 4 beta 7hi beta 1lo. alpha 4 beta 1hi beta 7- T cells were present in lymph draining from the skin. T cells migrating preferentially through intestinal LN were alpha 4 beta 7hi beta 1lo. Consistent with this migration pattern, the endothelial receptor for alpha 4 beta 7, mucosal addressin cell adhesion molecule-1 (MAdCAM-1) was detected on high endothelial venules within intestinal LN and Peyer's patches, but only weakly on high endothelial venules within peripheral LN. Thus, there are at least three easily definable subsets of T cells, based on integrin expression, which show distinct migration preferences.

A developmental switch in lymphocyte homing receptor and endothelial vascular addressin expression regulates lymphocyte homing and permits CD4+ CD3- cells to colonize lymph nodes

IN adult mice, the dominant adhesion molecules involved in homing to lymph nodes are L-selectin homing receptors on lymphocytes and the peripheral lymph node addressins on specialized high endothelial venules. Here we show that, from fetal life through the first 24 hr of life, the dominant adhesion molecules are the mucosal addressin MAdCAM-1 on lymph node high endothelial venules and its counterreceptor, the Peyer's patch homing receptor, integrin alpha 4 beta 7 on circulating cells. Before birth, 40-70% of peripheral blood leukocytes are L-selectin-positive, while only 1-2% expresses alpha 4 beta 7. However, the fetal lymph nodes preferentially attract alpha 4 beta 7-expressing cells, and this can be blocked by fetal administration of anti-MAdCAM-1 antibodies. During fetal and early neonatal life, when only MAdCAM-1 is expressed on high endothelial venules, an unusual subset of CD4 + CD3- cells, exclusively expressing alpha 4 beta 7 as homing receptors, enters the lymph nodes. Beginning 24 hr after birth a developmental switch occurs, and the peripheral node addressins are upregulated on high endothelial venules in peripheral and mesenteric lymph nodes. This switch in addressin expression facilitates tissue-selective lymphocyte migration and mediates a sequential entry of different cell populations into the lymph nodes.

Engagement of L-selectin impairs the actin polymerizing capacity of beta 2-integrins on neutrophils

A sequential activation of L-selectin and beta 2-integrins on neutrophils is crucial for the rolling, adherence and subsequent migration of these cells on the endothelium. However, little is known about a possible interplay between these adhesion receptors in the final regulation of cell motility. The results presented here show that sulfatides themselves (here used as tools to activate L-selectins), have no major effect on the cellular content of filamentous actin (F-actin), but cause a time-related decrease in the beta 2-integrin-induced formation of F-actin. This effect of sulfatides was abolished in cells lacking L-selectin as a result of pretreatment with chymotrypsin. A similar sulfatide-induced activation of L-selectin also caused a pronounced and time-related decrease of a subsequent chemotactic peptide-induced F-actin response. The effect of sulfatides on both beta 2-integrin- and chemotactic peptide-induced F-actin were abolished if L-selectin were blocked by preincubating the cells with specific antibodies to L-selectin. These effects of L-selectin engagement on cellular F-actin content were neither abolished by blocking the cytosolic free Ca2+ signal with bis-(2-amino-5-methylphenoxy)ethane-N,N,N',N'-tetraaceticacid tetraacetoxymethyly ester (MAPT/AM) nor by blocking a cAMP-induced activation of protein kinase A by pretreating the cells with adenosine-3′,5′-cyclic monophos-phorothioate (Rp-cAMPS). Instead we found that L-selectin engagement impaired an early beta 2-integrin-induced tyrosine kinase activation, an event shown to be necessary for a normal beta 2-integrin-mediated F-actin response. The present demonstration of a negative feed-back function of L-selectin on beta 2-integrin-induced modulations of the actin cytoskeleton, suggests that the relative distribution and/or density of the respective L-selectin and beta 2-integrin ligands on endothelial cells might be important factors in determining the final site of firm adhesion and extravasation of neutrophils.

Neutrophil adhesion molecule expression and response to stimulation with bacterial wall products in humans is unaffected by parenteral nutrition

1. Total parenteral nutrition is associated with a high incidence of septic complications. This may be partly due to neutrophil dysfunction induced by the parenteral nutrition. 2. Neutrophil adhesion molecule expression and the expression of CD11b in response to stimulation with formylmethionyl-leucyl-phenylalanine and lipopolysaccharide were determined before and after 24 h of lipid-containing parenteral nutrition. Eighteen adult patients referred for parenteral nutrition were studied. 3. There was no change in the expression of neutrophil L-selectin (CD62L), CD11a, CD11b, CD11c or CD15. Neutrophil response to stimulation with formylmethionyl-leucyl-phenylalanine and lipopolysaccharide as determined by CD11b expression was unaffected by parenteral nutrition. 4. This study has shown no evidence of parenteral nutrition-induced neutrophil dysfunction.

Constitutive class I-restricted exogenous presentation of self antigens in vivo

Ovalbumin (OVA)-specific CD8+ T cells from the T cell receptor-transgenic line OT-I (OT-I cells) were injected into unirradiated transgenic RIP-mOVA mice, which express a membrane-bound form of OVA (mOVA) in the pancreatic islet beta cells and the renal proximal tubular cells. OT-I cells accumulated in the draining lymph nodes (LN) of the kidneys and pancreas and in no other LN. They displayed an activated phenotype and a proportion entered cell cycle. Unilateral nephrectomy 7-13 d before inoculation of OT-I cells into RIP-mOVA mice allowed the injected T cells to home only to the regional LN of the remaining kidney (and pancreas), but when the operation was performed 4 h before injecting the T cells, homing to the LN of the excised kidney was evident. When the bone marrow of RIP-mOVA mice was replaced with one of a major histocompatibility haplotype incapable of presenting OVA to OT-I cells, no homing or activation was detectable. Therefore, OT-I cells were activated by OVA presented by short-lived antigen-presenting cells of bone marrow origin present in the draining LN of OVA-expressing tissue. These results provide the first evidence that tissue-associated "self" antigens can be presented in the context of class I via an exogenous processing pathway. This offers a constitutive mechanism whereby T cells can be primed to antigens that are present in nonlymphoid tissues, which are not normally surveyed by recirculating naive T cells.

Differential regulation of leucocyte L-selectin (CD62L) expression in normal lymphoid and inflamed extralymphoid tissues

AIMS

To study tissue expression of L-selectin, a leucocyte cell surface molecule that is considered to be involved in adhesion to certain endothelia, particularly in peripheral lymph nodes and during inflammation, and is shed upon leucocyte activation.

METHODS

Leucocytes were examined by immunohistochemistry and double immunofluorescence staining in various lymphoid sites and normal and inflamed extralymphoid tissues.

RESULTS

L-selectin was present on mantle zone B lymphocytes in different lymphoid sites, including in intestinal lymphoid tissue, but was absent on germinal centre B cells. Splenic white pulp B cells also expressed L-selectin. The proportion of T lymphocytes expressing L-selectin depended on the site under study, being greatest in peripheral lymph nodes (mean 48% of T cells positive), and lower in mucosal lymphoid sites and spleen (9 and 11% positive, respectively). Non-lymphocytic L-selectin staining was observed on follicular dendritic cells in tonsils and on macrophages in thymus. L-selectin positive leucocytes were rare in normal extralymphoid tissues, and relatively few were seen in most inflammatory settings. However, in rejecting renal transplants, a higher proportion (30%) of leucocytes expressed L-selectin.

CONCLUSIONS

Overall, the results indicate how the degree of L-selectin expression by leucocytes in particular tissues may reflect a requirement for L-selectin expression for entry into those tissues and the activation state of leucocytes once localised there.

Neutrophil-neutrophil interactions under hydrodynamic shear stress involve L-selectin and PSGL-1. A mechanism that amplifies initial leukocyte accumulation of P-selectin in vitro

Leukocytes attach to and roll on inflamed endothelium and on leukocyte monolayers that form on the endothelial cells. Leukocyte-leukocyte interactions occurring under hydrodynamic shear stress are mediated by binding of L-selectin to unknown sialomucin-like glycoproteins. We show that purified neutrophil PSGL-1, a sialomucin glycoprotein that serves as a ligand for both P- and E-selectin, can also support the attachment and rolling of free flowing neutrophils in vitro. Neutrophil rolling on PSGL-1 was abolished by the anti-L-selectin mAb DREG200 and by the anti-PSGL-1 mAb PL1, indicating that L-selectin can interact directly with PSGL-1. Neutrophil rolling on neutrophil monolayers was also blocked by PL1 (60 +/- 9% SEM inhibition); however, DREG200 blocked more efficiently (93 +/- 7% SEM inhibition), suggesting that other L-selectin ligands may exist on the neutrophil surface. These studies demonstrate that PSGL-1 on the neutrophil surface is a major functional ligand for L-selectin. The avidity of this L-selectin-dependent adhesion event was sufficient to allow individual neutrophils rolling on P-selectin to capture free flowing neutrophils, which progressed to form linear strings and discrete foci of rolling neutrophils. Neutrophil accumulation on P-selectin accelerated with time as a result of neutrophil-assisted capture of free flowing neutrophils. When neutrophil-neutrophil interactions were blocked by DREG200, neutrophils accumulated on P-selectin in a random pattern and at a uniform rate. Thus, leukocyte-assisted capture of flowing leukocytes may play an important role in amplifying the rate of initial leukocyte recruitment at sites of inflammation.

Sulfatide-induced L-selectin activation generates intracellular oxygen radicals in human neutrophils: modulation by extracellular adenosine

The sulfated form of galactocerebrosides (sulfatides) have recently been established as ligands for L-selectin. In this study we show that exposure of human neutrophils to sulfatides induces a transient generation of oxygen radicals, revealed by the luminol-enhanced chemiluminescence (CL) technique. The CL response was mainly located intracellularly, and was dependent on sulfation of the galactose ring, since non-sulfated galactocerebrosides had no effect. Sulfatides also dramatically amplified the CL response triggered by the chemotactic peptide formylmethionyl-leucyl-phenylalanine (fMLP). This effect was primarily due to an increased (up to 10-fold) intracellular generation of oxygen metabolites. Removal or blocking of L-selectin with chymotrypsin and monoclonal antibodies, respectively, markedly reduced the effects of sulfatides. Furthermore, sulfatides amplified the CL response triggered by ionomycin, whereas the response induced by phorbol-12-myristate-13-acetate was slightly reduced. The tyrosine kinase inhibitor, genistein, markedly inhibited the oxygen radical production induced by sulfatides, and totally abolished the potentiating effects of sulfatides in fMLP- and ionomycin-stimulated neutrophils. Sulfatides also triggered a transient rise in the intracellular free calcium concentration, [Ca2+]i. Consequently, L-selectin activation through sulfatides appear to affect oxidase activity through a Ca(2+)-dependent pathway involving tyrosine phosphorylation. Adenosine is an anti-inflammatory agent predominately released from the vascular endothelium which might suppress an inappropriate activation of the oxidase during L-selectin-mediated rolling of neutrophils. Indeed, we found that adenosine inhibited the oxidative burst induced by sulfatides, mainly by attenuating the intracellular generation of oxygen radicals. However, 10-100 times higher concentration of exogenous adenosine was required to inhibit the CL response induced by sulfatides to the same extent as the adenosine-mediated inhibition of the fMLP-induced response. This difference in sensitivity to adenosine could be explained by various expression of extracellular adenosine deaminase (ADA), since we found that the ADA-inhibitor erythro-9-(2-hydroxy-3-nonyl)-adenine (EHNA) markedly reduced the oxygen radical production caused by sulfatides and almost totally abolished the potentiating effects of sulfatides on the fMLP-induced respiratory burst. In contrary, EHNA only slightly reduced the fMLP-triggered CL response. We suggest that the initial activation of L-selectin prepare the neutrophil for an effective microbicidal activity in the extravascular space. This process might be dependent on a L-selectin-mediated increase in the expression and activity of ADA, which locally reduces the extracellular level of adenosine.

Cloning of the cDNA for rabbit L-selectin and expression of recombinant protein with a kinase target site to facilitate radiolabeling

The cDNA encoding rabbit L-Selectin has been cloned from a cDNA library, utilizing a PCR-derived probe. It encodes a peptide of 377 amino acids, including a signal peptide of 38 amino acids. Sequence analysis demonstrated extensive homology with L-Selectin's from other species. Recombinant rabbit L-Selectin protein was expressed in eukaryotic cells in a chimeric construct incorporating the entire extracellular portion of the protein coding region, a phosphokinase target site to allow high activity radiolabeling with 32P, and the constant region of the heavy chain of human IgG1 to facilitate purification and detection. Amino-terminal peptide sequencing of recombinant L-Selectin confirmed that the signal peptide had been removed at the expected site.

Release of histamine in whole blood by oxygen radicals: division between specific and unspecific processes

Oxygen derived free radicals are involved in many pathological processes such as postischemic reperfusion injuries, hepatotoxicity of drugs and inflammatory processes. Thereby these oxygen radicals induce lipid peroxidation and perturbation of cellular membranes. The aim of our present study was to determine whether oxygen radicals generated by the xanthine oxidase/ hypoxanthine system cause a release of histamine in human blood cell cultures. Stimulation of blood cell cultures with oxygen radicals induced a histamine liberation which was mainly due to calcium independent processes during the first 30 min, whereas then calcium requiring processes took part in the release of histamine. The regulation of the leukocyte selection LECAM-1 was altered by oxygen radicals whereas histamine, which is known to modulate vascular selectin expression, did not affect the expression of LECAM-1. Our data indicate that oxygen radicals induce a direct calcium independent release of histamine which is due to membrane pertubating processes during the first phase but also induce a specific reaction leading to a further indirect histamine liberation which is probably mediated by PAF.

Carbohydrate- and CD18-dependent neutrophil adhesion to cardiac myocytes: effects of adenosine

OBJECTIVE

Adenosine inhibits neutrophil adhesion and injury to isolated cardiac myocytes. In the present study, the contribution of selectin and CD18 interactions to neutrophil-myocyte adhesion and their sensitivity to adenosine were assessed.

METHODS

Activated human neutrophils and canine myocytes were incubated with inhibitors of CD18 or selectin binding, adenosine, or combinations of both for 30-50 min at 37 degrees C. Neutrophils were pretreated with 0.1 microM fMLP for 10 min to study L-selectin-independent adhesion. Adhesion was measured by phase contrast microscopy.

RESULTS

Anti-L-selectin mAb and the selectin-blocking carbohydrates sialyl Lewisx or mannose-6-phosphate, as well as anti-CD18 or anti-ICAM-1 mAbs, inhibited cell adhesion (by 84-99%, P < 0.05). CD11a, but not CD11b, was responsible for most of the CD18-mediated binding. An L-selectin-independent interaction between neutrophils and cardiac myocytes was observed that was delayed (peak adhesion at 40-50 min, rather than 30 min), but still inhibited by anti-CD18 mAb (by 65 +/- 11%, P < 0.05) and carbohydrates (by 87-97%, each P < 0.05). Adenosine (100 nM) inhibited this late CD18-dependent/L-selectin-independent phase of adhesion (by 61 +/- 14%, P < 0.05). The combination of adenosine and anti-CD18 mAb was additive such that adhesion was completely blocked (P < 0.05, compared to either agent alone). Inhibition of adhesion by adenosine was prevented by the A2 antagonist, DMPX (100 nM), and mimicked by the A2 agonist, CGS-21680 (10 nM) or the adenosine regulating agents, acadesine (100 microM) or GP531 (10 microM).

CONCLUSION

Neutrophil-myocyte adhesion involved both L-selectin-dependent and L-selectin-independent carbohydrate binding as well as CD11a/CD18. Inhibition of adhesion by adenosine interferes with L-selectin-independent carbohydrate binding and possibly CD18.

A highly practical synthesis of the sialyl Lewis X pentasaccharide and an investigation of binding to E-, P-, and L-selectins

A practical synthesis of the sialyl Lewis X (sLex) pentasaccharide, NeuAc alpha 2-3Gal beta 1-4(Fuc alpha 1-3)GlcNAc beta 1-3Gal beta OEt (1), as a potential blocker for E-selectin has been described. The glycosylation of a trisaccharide acceptor, Fuc alpha (1-3)GlcNAc beta (1-3)Gal beta OEt, with a disaccharide donor, NeuAc alpha (2-3)Gal beta SMe, did not yield the desired sLex pentasaccharide 1 at all. However, the glycosylation of a disaccharide acceptor, GlcNAc beta (1-3)Gal beta OEt, with a disaccharide donor, NeuAc alpha (2-3)Gal beta SMe, quantitatively yielded the tetrasaccharide NeuAc alpha (2-3)Gal beta (1-4)GlcNAc beta (1-3)Gal beta OEt. This tetrasaccharide is readily converted to the title compound in a high yield by fucosylation, followed by deprotection. The inhibitory activities of compound 1 toward the binding of the natural ligand (sLex) with the E-, P-, and L-selectins were stronger than those of the sLex tetrasaccharide.

Renal ischemic-reperfusion injury in L-selectin-deficient mice

L-selectin on leukocyte surfaces mediates cell rolling on endothelium. L-selectin blockade with antibodies attenuated ischemic-reperfusion injury (IRI) in heart and skeletal muscle, but its role in renal IRI is unknown. We evaluated the role of L-selectin in renal IRI using L-selectin-deficient mice. Neutrophil migration to chemically inflamed peritoneum was reduced by 47% (P < 0.01) in L-selectin-deficient mice. Ischemia was induced by bilateral renal pedicle clamping for 30 min. Control and L-selectin groups had similar elevations of serum creatinine (1.8 +/- 0.3 vs. 1.7 +/- 0.2 mg/dl) and blood urea nitrogen (111 +/- 17 vs. 128 +/- 12 mg/dl) 24 h postischemia. Pathological assessment showed comparable degrees of tubular necrosis at 24 h. The postischemic increase in peritubular neutrophils per 10 high-power field was similar in control and L-selectin-deficient groups at 4 (28 +/- 10 vs. 22 +/- 5), 12 (245 +/- 80 vs. 236 +/- 78), and 24 h (130 +/- 12 vs. 156 +/- 18). These data argue against a significant role for L-selectin in renal IRI. Patho-physiological roles of L-selectin in vivo appear to be more complex than in vitro data would suggest.

CD45 engagement induces L-selectin down-regulation

The CD45 glycoprotein isoforms exhibit a receptor-like composition and display intracellular protein tyrosine phosphatase (PTPase) activity. The present study links CD45 to the regulation of L-selectin (CD62L), a leucocyte glycoprotein important for extravasation and homotypic aggregation. Monoclonal antibodies (MoAbs) IOL1b and AICD45.2, but not GAP8.3, all of which are directed against common CD45 epitopes, were found to elicit lymphocyte L-selectin down-regulation. Lymphocyte L-selectin down-regulation in response to anti-CD45 MoAbs was enhanced by high cell density and partially antagonized by the protein tyrosine kinase (PTK) inhibitor, herbimycin A. The MoAbs IOL1b, AICD45.2 and GAP8.3 recognized granulocyte-expressed CD45 but did not induce loss of L-selectin expression of granulocytes. In contrast, the CD45 PTPase inhibitor, vanadate, induced L-selectin down-regulation both in lymphocytes and granulocytes. The PTPase activation by nitric oxide (NO) or the NO-generating compound, sodium nitroprusside, did not affect L-selectin surface expression. Increased concentrations of soluble L-selectin were detected after anti-CD45 or vanadate-induced down-regulation of L-selectin surface expression. While activation of protein kinase C (PKC) by phorbol 12-myristate 13-acetate (PMA) induces rapid L-selectin down-regulation of L-selectin surface expression in both lymphocytes and granulocytes, the PKC inhibitor, H 7, was also found to down-regulate lymphocyte and granulocyte L-selectin surface expression. The inhibitor H 7 synergized with vanadate in down-regulating lymphocyte L-selectin surface expression, but partially inhibited vanadate-induced granulocyte L-selectin down-regulation. The results suggest that in a cell type-specific fashion the PKC system and tyrosine phosphorylation and dephosphorylation cascades are involved in the regulation of L-selectin surface expression.

Glucocorticoid effects in an endotoxin-induced rat pulmonary inflammation model: differential effects on neutrophil influx, integrin expression, and inflammatory mediators

To understand the basis for the refractory nature of acute respiratory distress syndrome (ARDS) to glucocorticoids, the effects of dexamethasone pretreatment (DEX, 2 mg/kg, intraperitoneally) on the kinetics of airway tumor necrosis factor-alpha (TNF alpha) and macrophage inflammatory protein 2 (MIP-2) production, and polymorphonuclear leukocyte (PMN) influx after intratracheal lipopolysaccharide (LPS) (1 mg/kg) in rats were investigated. In the absence of exogenous glucocorticoids, TNF alpha and MIP-2 levels in bronchoalveolar lavage (BAL) fluid peaked at 21 and 300 ng, respectively, by 3 h. DEX pretreatment resulted in a 74% reduction in BAL TNF alpha, yet MIP-2 accumulation was unchanged. In addition, DEX reduced PMN influx at 5 h by 58.4% to 4.1 +/- 0.7 x 10(6) PMN (n = 5). DEX, however, did not mitigate the 3-fold increase in total BAL protein observed at 5 h, attributable to albumin influx. The effects of subacute DEX treatment (3.8 mg/kg per day, for 3 days) on cell-surface expression of the adhesion molecules CD11a, CD11b, and L-selectin were determined by flow cytometric analysis of peripheral blood and autologous BAL PMN. Compared with peripheral blood PMN, exudative PMN had 4-fold greater CD11b expression, no change in CD11a, and loss of L-selectin immunoreactivity 5 h after LPS challenge. The upregulation of CD11b on exudative PMN was insensitive to DEX pretreatment, which, together with a failure to suppress MIP-2 levels, provides a possible explanation for the lack of efficacy of steroids in the management of ARDS.

Defective Peyer's patch organogenesis in mice lacking the 55-kD receptor for tumor necrosis factor

Lymphotoxin alpha (LT-alpha) may form secreted homotrimers binding to p55 and p75 tumor necrosis factor (TNF) receptors or cell surface-bound heterotrimers with LT-beta that interact with the LT-beta receptor. Genetic ablation of LT-alpha revealed that mutant mice have no detectable lymph nodes or Peyer's patches and that the organization of the splenic white pulp in T and B cell areas is disturbed. In this report we describe a novel function for the p55 TNF receptor during ontogeny and demonstrate that mice deficient for p55 completely lack organized Peyer's patches. In contrast, lymph nodes and spleen are present in p55-deficient mice and lymphocytes segregate normally into B and T cell areas in these organs. Lamina propria and intraepithelial lymphocytes of the small intestine were detected in normal number and distribution in p55 mutant mice. Lymphocytes and endothelial cells from p55-deficient mice express normal levels of adhesion molecules considered important for lymphocyte migration to mucosal organs; this indicates that the lack of Peyer's patches does not result from a defect in lymphocyte homing. In summary, the p55 receptor for TNF selectively mediates organogenesis of Peyer's patches throughout ontogeny, suggesting that the effects of LT-alpha on the development of lymphoid organs may be mediated by distinct receptors, each functioning in an organ-specific context.

In vivo T cell response to viral superantigen. Selective migration rather than proliferation

Superantigens induce T cell activation and proliferation in vitro, and some also induce cell activation in vivo. MMTV(SW) is an infectious mouse mammary tumor virus (MMTV) encoding a superantigen with the same Vbeta specificity as MIs-1a (Mtv-7), which induces a strong local response in vivo. injection of MMTV(SW) into mouse footpads leads to accumulation of superantigen-reactive T cells (Vbeta6+CD4+) and B cells in the draining lymph nodes (LN). We investigated the kinetics of this cell accumulation by measuring cell activation (blastogenesis, CD25 and CD69 expression), cell migration (using syngenic FITC-labeled CD4+ cells and L-selectin detection), and cell proliferation (using in vivo labeling with bromodeoxyuridine). Specific T cells selectively migrated to the draining LN. Accumulating Vbeta6+CD4+ T cells were large CD69+ cells, but remained CD25 negative and showed down-regulated L-selectin expression. Their DNA synthesis rate, studied by pulse labeling and continuous administration of bromodeoxyuridine, was increased, but remained too low to explain the draining LN hyperplasia. These data show that the local T cell response to MMTV(SW) mainly consists of selective migration followed by local activation of reactive T cells, and that cell proliferation is only a minor component of the response. By contrast, the optimal dose of staphylococcal enterotoxin B that, nevertheless, leads to a lower reactive T cell accumulation in the draining LN induces a very high proliferation rate.

Calcium-dependent oligonucleotide antagonists specific for L-selectin

The selectins are calcium-dependent C-type lectins that recognize complex anionic carbohydrate ligands, initiating many cell-cell interactions in the vascular system. Selectin blockade shows therapeutic promise in a variety of inflammatory and postischemic pathologies. However, the available oligosaccharide ligand mimetics have low affinities and show cross-reaction among the three selectins, precluding efficient and specific blockade. The SELEX (systematic evolution of ligands by exponential enrichment) process uses combinatorial chemistry and in vitro selection to yield high affinity oligonucleotides with unexpected binding specificities. Nuclease-stabilized randomized oligonucleotides subjected to SELEX against recombinant L-selectin yielded calcium-dependent antagonists with approximately 10(5) higher affinity than the conventional oligosaccharide ligand sialyl LewisX. Most of the isolated ligands shared a common consensus sequence. Unlike sialyl LewisX, these antagonists show little binding to E- or P-selectin. Moreover, they show calcium-dependent binding to native L-selectin on peripheral blood lymphocytes and block L-selectin-dependent interactions with the natural ligands on high endothelial venules.