Overlapping roles for L-selectin and P-selectin in antigen-induced immune responses in the microvasculature

Although L-selectin mediates lymphocyte attachment to endothelial venules of peripheral lymph nodes, its role in leukocyte recruitment into tissues following Ag challenge is less well established. The objective of this study was to systematically examine the role of L-selectin in leukocyte rolling in the peripheral microvasculature during the first 24 h of an immune response. A type I hypersensitivity response was elicited in wild-type (C57BL/6) and L-selectin-deficient mice by systemic (i.p.) sensitization and intrascrotal challenge with chicken egg OVA. The cremaster microcirculation was observed in untreated and sensitized mice 4, 8, and 24 h post-Ag challenge by intravital microscopy. Leukocyte recruitment in L-selectin-deficient mice and wild-type mice treated with an L-selectin function-blocking mAb was examined at each time point. Ag challenge induced a significant increase in leukocyte rolling (60 cells/min/venule to approximately 300 cells/min/venule) in wild-type mice at 4-24 h. This response was reduced by approximately 60-70% in L-selectin-deficient mice and in wild-type mice treated with an L-selectin-blocking mAb. P-selectin blockade by Ab completely inhibited leukocyte rolling at 4-24 h in wild-type animals and also blocked the residual rolling seen in L-selectin-deficient mice. Blocking E-selectin function had no effect on leukocyte rolling flux at any time point in wild-type or L-selectin-deficient mice. Despite reduced rolling, leukocyte adhesion and emigration were not measurably reduced in the L-selectin-deficient mice in this vascular bed. In conclusion, leukocyte rolling is L-selectin-dependent post-Ag challenge with L-selectin and P-selectin sharing overlapping functions.     

Effects of CXC chemokines on neutrophil activation and sequestration in hepatic vasculature

The initiating step of neutrophil-induced cytotoxicity in the liver is the recruitment of these phagocytes into sinusoids. The aim of our study was to compare the efficacy of systemic exposure with individual inflammatory mediators on neutrophil activation and sequestration in the hepatic vasculature of C3Heb/FeJ mice as assessed by flow cytometry and histochemistry, respectively. The CXC chemokine macrophage inflammatory protein-2 (MIP-2; 20 microg/kg) induced a time-dependent upregulation of Mac-1 (318% at 4 h) and shedding of L-selectin (41% at 4 h). MIP-2 treatment caused a temporary increase of sinusoidal neutrophil accumulation at 0.5 h [97 +/- 6 polymorphonuclear leukocytes (PMN)/50 high-power fields (HPF)], which declined to baseline (8 +/- 2) at 4 h. The CXC chemokine KC was largely ineffective in activating neutrophils or recruiting them into the liver. Cytokines (tumor necrosis factor-alpha and interleukin-1alpha) and cobra venom factor substantially increased Mac-1 expression and L-selectin shedding on neutrophils and caused stable sinusoidal neutrophil accumulation (170-220 PMN/50 HPF). Only cytokines induced venular neutrophil margination. Thus CXC chemokines in circulation are less effective than cytokines or complement in activation of neutrophils and their recruitment into the hepatic vasculature in vivo.     

SC-41930: an inhibitor of leukotriene B4-stimulated human neutrophil functions

SC-41930 was evaluated for effects on human neutrophil chemotaxis and degranulation. At concentrations up to 100 microM, SC-41930 alone exhibited no effect on neutrophil migration, but dose-dependently inhibited neutrophil chemotaxis induced by leukotriene B4 (LTB4) in a modified Boyden chamber. Concentrations of SC-41930 from 0.3 microM to 3 microM competitively inhibited LTB4-induced chemotaxis with a pA2 value of 6.35. While inactive at 10 microM against C5a-induced chemotaxis, SC-41930 inhibited N-formyl-methionyl-leucyl-phenylalanine (fMLP)-induced chemotaxis, with 10 times less potency than against LTB4-induced chemotaxis. SC-41930 inhibited [3H]LTB4 and [3H]fMLP binding to their receptor sites on human neutrophils with KD values of 0.2 microM and 2 microM, respectively. SC-41930 also inhibited neutrophil chemotaxis induced by 20-OH LTB or 12(R)-HETE. At concentrations up to 10 microM, SC-41930 alone did not cause neutrophil degranulation, but inhibited LTB4-induced degranulation in a noncompetitive manner. SC-41930 also inhibited fMLP- or C5a-induced degranulation, but was about 8 and 10 times less effective for fMLP and C5a, respectively. The results indicate that SC-41930 is a human neutrophil LTB4 receptor antagonist with greater specificity for LTB4 than for fMLP or C5a receptors.     

Expression of L-selectin, but not CD44, is required for early neutrophil extravasation in antigen-induced arthritis

L (leukocyte)-selectin (CD62L) and CD44 are major adhesion receptors that support the rolling of leukocytes on endothelium, the first step of leukocyte entry into inflamed tissue. The specific contribution of L-selectin or CD44 to the regulation of cell traffic to joints in arthritis has not been investigated. We used CD44-deficient, L-selectin-deficient, and CD44/L-selectin double knockout mice to determine the requirement for these receptors for inflammatory cell recruitment during Ag-induced arthritis. Intraperitoneal immunization resulted in similar activation status and Ag-specific responses in wild-type and gene-targeted mice. However, extravasation of neutrophil granulocytes, but not the emigration of T cells, into the knee joints after intra-articular Ag injection was significantly delayed in L-selectin-deficient and double knockout mice. Intravital videomicroscopy on the synovial microcirculation revealed enhanced leukocyte rolling and diminished adherence in mice lacking either CD44 or L-selectin, but CD44 deficiency had no significant effect on the recruitment of L-selectin-null cells. Compared with wild-type leukocytes, expression of L-selectin was down-regulated in CD44-deficient cells in the spleen, peripheral blood, and inflamed joints, suggesting that reduced expression of L-selectin, rather than the lack of CD44, could be responsible for the delayed influx of granulocytes into the joints of CD44-deficient mice. In conclusion, there is a greater requirement for L-selectin than for CD44 for neutrophil extravasation during the early phase of Ag-induced arthritis.     

Surfactant protein A differentially regulates peripheral and inflammatory neutrophil chemotaxis

Surfactant protein A (SP-A), a pulmonary lectin, plays an important role in regulating innate immune cell function. Besides accelerating pathogen clearance by pulmonary phagocytes, SP-A also stimulates alveolar macrophage chemotaxis and directed actin polymerization. We hypothesized that SP-A would also stimulate neutrophil chemotaxis. With the use of a Boyden chamber assay, we found that SP-A (0.5-25 microg/ml) did not stimulate chemotaxis of rat peripheral neutrophils or inflammatory bronchoalveolar lavage (BAL) neutrophils isolated from LPS-treated lungs. However, SP-A affected neutrophil chemotaxis toward the bacterial peptide formyl-met-leu-phe (fMLP). Surprisingly, the effect was different for the two neutrophil populations: SP-A reduced peripheral neutrophil chemotaxis toward fMLP (49 +/- 5% fMLP alone) and enhanced inflammatory BAL neutrophil chemotaxis (277 +/- 48% fMLP alone). This differential effect was not seen for the homologous proteins mannose binding lectin and complement protein 1q but was recapitulated by type IV collagen. SP-A bound both neutrophil populations comparably and did not alter formyl peptide binding. These data support a role for SP-A in regulating neutrophil migration in pulmonary tissue.     

Characterization of the plasma membrane bound GTPase from rabbit neutrophils. I. Evidence for an Ni-like protein coupled to the formyl peptide, C5a, and leukotriene B4 chemotaxis receptors

We have characterized the GTPase activity of the Ni-like guanine-nucleotide-binding regulatory protein in rabbit neutrophil plasma membranes. The low Km (3.64 +/- 0.87 X 10(-7) M) GTPase copurified with the formyl peptide receptor in the plasma membrane fraction obtained by discontinuous sucrose density gradient centrifugation. The Vmax (23.9 +/- 2.91 pmol/mg/min) and Km of the unstimulated enzyme were similar to those reported for Ni in other cell types. The activity of the unstimulated enzyme was both magnesium and sodium dependent and linear over the first 4 min of the assay. The chemoattractants, formyl-methionyl-leucyl-phenylalanine (fMLP), C5a, and leukotriene B4 (LTB4) stimulated the GTPase in purified neutrophil plasma membrane preparations, whereas other secretagogues, such as A23187 and PMA, were without effect. Lineweaver-Burk analysis showed an fMLP-induced increase in Vmax (31.94 +/- 4.80 pmol/mg/min) (33.1 +/- 9.5%) but not in Km. The dose-response curve for fMLP stimulation showed an ED50 of 4.1 +/- 1.0 X 10(-8) M and an overall 22.2 +/- 3.1% maximal stimulation. C5a (30 micrograms/ml) increased the activity of the GTPase 21.3 +/- 5.7% and 10(-7) M LTB4 produced a 32.2 +/- 5.4% increase. Activated pertussis toxin treatment of neutrophil plasma membranes inhibited by 72.5 +/- 14.3% the stimulation of GTPase activity induced by fMLP; however, activated cholera toxin had no effect on the inhibition of fMLP stimulation, suggesting a direct role for an Ni-like protein in the coupling process. In contrast to the lack of inhibition of fMLP stimulation by activated cholera toxin treatment of plasma membranes, both pertussis toxin and to a lesser extent cholera toxin treatment reduced fMLP, C5a, and LTB4 stimulation of the GTPase in sonicates prepared from pretreated whole cells. Pertussis toxin inhibited fMLP stimulation of the GTPase by 75 +/- 7%, C5a stimulation was inhibited by 83 +/- 13%, and LTB4 stimulation was inhibited completely. Sonicates prepared from neutrophils treated similarly with cholera toxin showed a smaller inhibition of GTPase activity (50 +/- 4% and 14 +/- 9% for fMLP and LTB4, respectively) with the exception of C5a, where CT inhibition (81 +/- 32%) equaled pertussis toxin inhibition. Similarly, pertussis toxin completely inhibited the release of the granule enzyme N-acetyl-glucosaminidase by all three chemoattractants, whereas cholera toxin, except with C5a stimulation, had little or no effect.(ABSTRACT TRUNCATED AT 400 WORDS)     

A functional role for circulating mouse L-selectin in regulating leukocyte/endothelial cell interactions in vivo

L-selectin mediates the initial capture and subsequent rolling of leukocytes along inflamed vascular endothelium and mediates lymphocyte migration to peripheral lymphoid tissues. Leukocyte activation induces rapid endoproteolytic cleavage of L-selectin from the cell surface, generating soluble L-selectin (sL-selectin). Because human sL-selectin retains ligand-binding activity in vitro, mouse sL-selectin and its in vivo relevance were characterized. Comparable with humans, sL-selectin was present in adult C57BL/6 mouse sera at approximately 1.7 micro g/ml. Similar levels of sL-selectin were present in sera from multiple mouse strains, despite their pronounced differences in cell surface L-selectin expression levels. Adhesion molecule-deficient mice prone to spontaneous chronic inflammation and mice suffering from leukemia/lymphoma had 2.5- and 20-fold increased serum sL-selectin levels, respectively. By contrast, serum sL-selectin levels were reduced by 70% in Rag-deficient mice lacking mature lymphocytes. The majority of serum sL-selectin had a molecular mass of 65-75 kDa, consistent with its lymphocyte origin. Slow turnover may explain the relatively high levels of sL-selectin in vivo. The t(1/2) of sL-selectin, assessed by transferring sera from wild-type mice into L-selectin-deficient mice and monitoring serum sL-selectin levels by ELISA, was >20 h, and it remained detectable for longer than 1 wk. Short-term in vivo lymphocyte migration assays demonstrated that near physiologic levels ( approximately 0.9 micro g/ml) of sL-selectin decreased lymphocyte migration to peripheral lymph nodes by >30%, with dose-dependent inhibition occurring with increasing sL-selectin concentrations. These results suggest that sL-selectin influences lymphocyte migration in vivo and that the increased sL-selectin levels present in certain pathologic conditions may adversely affect leukocyte migration.     

Release of O2- by human umbilical cord blood-derived eosinophils: role of intra- and extracellular calcium.

The aim of our study was to investigate the physiologic mechanisms involved in eosinophil activation as an essential prerequisite to disrupting the biochemical cascade that triggers inflammation, thereby attenuating the effect of this activation or, ideally, preventing it from occurring. We have, therefore, examined the nature of the fMLP- and PAF-induced [Ca2+]i rise and the relationship between the [Ca2+]i rise and O2- production in human umbilical cord blood-derived eosinophils cultured in the presence of IL-3 and IL-5. These cells responded to fMLP or PAF (1 microM each) with an increase in [Ca2+]i (217.3 +/- 22.1 and 197.8 +/- 22.1 nM respectively) which was associated with production of O2- (40.2 +/- 8.2 and 35.2 +/- 7.6 pmol/min/10(6) cells respectively). The role of Ca2+ in the induced respiratory burst was studied by changing the availability of Ca2+ in the intra- and extracellular compartments. Removal or chelation of extracellular Ca2+ induced a reduction of both the fMLP and PAF-induced [Ca2+]i rise and O2- production. Chelation of intracellular Ca2+ induced a concentration-dependent inhibition of fMLP- and PAF-induced [Ca2+]i rise and caused a decrease in O2- production. SK&F 96365 had a stimulatory effect on PAF-induced [Ca2+]i rise and on fMLP-induced O2- production, this phenomenon was not observed with extracellular Ca2+ removal or chelation. Furthermore, Ni2+ exhibited an inhibition of both fMLP and PAF-induced [Ca2+]i rise and O2- production. Finally, both fMLP and PAF induced an increase in divalent cation influx that was further augmented by thapsigargin. Our results indicate that fMLP and PAF dependent O2- production in human eosinophils require intra- and extracellular Ca2+ and that Ca2+ influx is necessary for optimal activation.     

ONO-4057, a novel, orally active leukotriene B4 antagonist: effects on LTB4-induced neutrophil functions.

ONO-4057(5-[2-(2-Carboxyethyl)-3-[6-(4-methoxyphenyl)-5E- hexenyl]oxyphenoxy]valeric acid), an orally active leukotriene B4(LTB4) antagonist, displaced the binding of [3H] LTB4 to the LTB4 receptor in human neutrophil (Ki = 3.7 +/- 0.9 nM). ONO-4057 inhibited the LTB4-induced rise in cytosolic free calcium (the concentration causing 50% inhibition (IC50) = 0.7 +/- 0.3 microM) and inhibited human neutrophil aggregation, chemotaxis or degranulation induced by LTB4 (IC50 = 3.0 +/- 0.1, 0.9 +/- 0.1 and 1.6 +/- 0.1 microM) without showing any agonist activity at concentration up to 30 microM. ONO-4057 did not inhibit fMLP or C5a-induced neutrophil activation at concentrations up to 30 microM. In the in vivo study, ONO-4057 given orally, prevented LTB4-induced transient neutropenia or intradermal neutrophil migration in guinea pig (the dose causing 50% efficacy (ED50) = 25.6mg/kg or 5.3mg/kg). Furthermore, ONO-4057 given topically, suppressed phorbol-12-myristate-13-acetate (PMA)-induced neutrophil infiltration in guinea pig ear (the effective dose = 1 mg/ear). These results indicate that ONO-4057 is a selective and orally active LTB4 antagonist and may be a potential candidate for the treatment of various inflammatory diseases.     

L-selectin facilitates emigration and extravascular locomotion of leukocytes during acute inflammatory responses in vivo

L-selectin has been shown to be important in mediating leukocyte recruitment during inflammatory responses. Although there are numerous in vitro studies demonstrating that engagement of L-selectin leads to the activation of several signaling pathways potentially contributing to subsequent adhesion, emigration, or even migration through the interstitium, whether this actually induces cellular events in vivo is completely unknown. Therefore, we used intravital microscopy to visualize the role of L-selectin in downstream leukocyte adhesion, emigration, and interstitial migration events in wild-type and L-selectin-deficient (L-selectin(-/-)) mice. The cremaster muscle was superfused with the chemotactic inflammatory mediators platelet-activating factor or KC. Leukocyte rolling, adhesion, and emigration in postcapillary venules were examined, and the migration of emigrated leukocytes was recorded continuously using time-lapse videomicroscopy. Platelet-activating factor increased leukocyte adhesion to a similar level in both wild-type and L-selectin(-/-) mice. In contrast, both the number of emigrated leukocytes and the distance of extravascular migration were significantly reduced in L-selectin(-/-) mice. A similar pattern was observed in response to the superfusion of KC. Because superfusion of these mediators induced chemokinesis, we developed a new in vivo chemotaxis assay using slow release of KC from an agarose gel positioned 350 microm from a postcapillary venule. These experiments showed that L-selectin(-/-) leukocytes were also severely impaired in their ability to respond to a directional cue. These findings indicate that L-selectin is important in enabling leukocytes to respond effectively to chemotactic stimuli in inflamed tissues.     

Role of alveolar macrophage and migrating neutrophils in hemorrhage-induced priming for ALI subsequent to septic challenge

Acute lung injury (ALI) is identified with the targeting/sequestration of polymorphonuclear leukocytes (PMN) to the lung. Instrumental to PMN targeting are chemokines [e.g., macrophage inflammatory protein-2 (MIP-2), keratinocyte-derived chemokine (KC), etc.] produced by macrophage, PMN, and other resident pulmonary cells. However, the relative contribution of resident pulmonary macrophages as opposed to PMN in inducing ALI is poorly understood. We therefore hypothesize that depletion of peripheral blood PMN and/or the oblation of a macrophage-mediated PMN chemokine signal (via macrophage deficiency) will reduce the inflammation and ALI observed in mice following hemorrhage (Hem) and subsequent sepsis (CLP) in our murine model of ALI. To examine this we pretreated mice with either 500 microg anti-mouse Gr1 antibody/animal (to deplete PMN) or subjected mice deficient in mature macrophage (B6C3Fe-a/a-CsF1op) to Hem (90 min at 35 +/- 5 mmHg) followed by resuscitation. Twenty-four hours post-Hem, mice were subjected to CLP and killed 24 h later, and lung tissue samples were collected. Our data showed that in the absence of either peripheral blood PMN or mature tissue macrophages there was a suppression of IL-6, KC, and MIP-2 levels in lung tissue from Hem/CLP mice as well as a reduction in PMN influx to the lung and lung injury (bronchoalveolar lavage fluid protein). In contrast, lung tissue IL-10 and TNF-alpha levels were suppressed in the macrophage-deficient Hem/CLP mice compared with PMN-depleted Hem/CLP mice. Together, these data suggest that both the PMN and the macrophage are required to induce inflammation seen here, however, macrophage not PMN regulate the release of IL-10, independent of local changes in TNF.     

Abnormal directed migration of blood polymorphonuclear leukocytes in rheumatoid arthritis. Potential role in increased susceptibility to bacterial infections.

Rheumatoid arthritis (RA) patients are at higher risks of bacterial infection than healthy subjects. Polymorphonuclear leukocytes (PMN) are the first line of nonspecific cellular defence against these infections. We tested the hypothesis that abnormal directed migration of PMN may be one reason for the increased infection rate of RA patients. PMN migration was investigated in 68 peripheral blood samples of 15 RA patients compared with 64 samples of healthy controls in a novel whole blood in vitro membrane filter assay. The migration of PMNs from RA patients and controls was stimulated using the bacterial chemoattractant N-formyl-methionyl-leucyl-phenylalanine (fMLP). Unstimulated PMN migration of RA patients was increased compared with healthy controls as measured by the following parameters: (a) absolute number of migrant PMNs (1954+/-87 vs. 1238 +/-58 PMN/mm2), (b) percentage of PMNs migrated into the filter (total migration index, TMI) (28.6+/-0.9 vs. 24.0+/-0.8%), (c) the distance half the migrating PMNs had covered (distribution characteristic, DC) (22.6+/-1.1 vs. 16.1+/-0.6 mm) and (d) the product of TMI and DC (neutrophil migratory activity, NMA) (669.0+/-45.0 vs. 389.0+/-18.9). fMLP stimulated PMNs of RA patients showed defective migration compared to unstimulated samples as shown by (a) a reduced number of migrant PMNs (1799+/-93 PMN/mm2), (b) lower TMI (26.1+/-0.9%), (c) unremarkable altered distribution characteristic (22.9+/-0.8 mm) and (d) significant reduced migratory activity (600.0+/-30.0). Our data suggest that the high incidence of infections in RA patients may partly be caused by defective migratory activity of PMNs to bacterial chemoattractants as demonstrated by fMLP.     

Expression of the cell adhesion molecules on leukocytes that demarginate during acute maximal exercise.

The pulmonary vascular bed is an important reservoir for the marginated pool of leukocytes that can be mobilized by exercise or catecholamines. This study was designed to determine the phenotypic characteristics of leukocytes that are mobilized into the circulation during exercise. Twenty healthy volunteers performed incremental exercise to exhaustion [maximal O2 consumption (VO2 max)] on a cycle ergometer. Blood was collected at baseline, at 3-min intervals during exercise, at VO2 max, and 30 min after exercise. Total white cell, polymorphonuclear leukocyte (PMN), and lymphocyte counts increased with exercise to VO2 max (P < 0.05). Flow cytometric analysis showed that the mean fluorescence intensity of L-selectin on PMN (from 14.9 +/- 1 at baseline to 9.5 +/- 1.6 at VO2 max, P < 0.05) and lymphocytes (from 11.7 +/- 1.2 at baseline to 8 +/- 0.8 at VO2 max, P < 0.05) decreased with exercise. Mean fluorescence intensity of CD11b on PMN increased with exercise (from 10.2 +/- 0.6 at baseline to 25 +/- 2.5 at VO2 max, P < 0.002) but remained unchanged on lymphocytes. Myeloperoxidase levels in PMN did not change with exercise. In vitro studies showed that neither catecholamines nor plasma collected at VO2 max during exercise changed leukocyte L-selectin or CD11b levels. We conclude that PMN released from the marginated pool during exercise express low levels of L-selectin and high levels of CD11b.     

Importance of extracellular divalent cations to polarisation of polymorphonuclear leukocytes induced by plasma

The roles of extracellular calcium and magnesium ions in the polarisation of human peripheral blood polymorphonuclear leukocytes (PMN) induced by autologous fresh heparinised plasma were investigated by studying the effects of 5 mM chelators of divalent cations [ethylenediamine tetra-acetic acid (EDTA), ethylenebis-(oxyethylene-nitrilo)-tetra-acetic acid (EGTA) or disodium hydrogen citrate]. In addition, the effects of a blocker of membrane calcium channels (verapamil) were studied. Polarisation of PMN suspended in plasma (84.1 +/- 11.9%) was reduced by each chelating agent over 30 min (to 20.0 +/- 15.6% by EDTA, to 42.5 +/- 19.3% by EGTA and to 29.4 +/- 22.9% by citrate). Polarisation of PMN suspended in plasma treated with EDTA or EGTA was restored by inclusion of equimolar additional Ca2+ ions, and in plasma treated with EDTA, EGTA or citrate, by equimolar additional Mg2+ ions. Additional Mg2+ had no effect on the spherical shape of PMN in Hanks' solution and additional cations had no effects on the polarisation of PMN induced by fMLP. Cells rendered spherical by each chelating agent in plasma for 30 min retained their ability to polarise on addition of fMLP to the plasma-chelator medium. Verapamil (10(-4) M) markedly reduced polarisation in plasma (to 52 +/- 11.3%) but the same drug (10(-5) M) had no such effect. In contrast to the polarisation of cells in plasma, the polarisation response of PMN to N-formyl-methionyl-leucyl-phenylalanine (fMLP, 10(-8) M) in buffered Hanks' solution was not affected by any of the chelating agents or by verapamil, even in high concentration. These results indicate that extracellular divalent cations are necessary for the polarisation of PMN suspended in autologous plasma and that the mechanism of polarisation of PMN in plasma may be different to that of polarisation induced by fMLP.     

The bacterial peptide N-formyl-Met-Leu-Phe inhibits killing of Staphylococcus epidermidis by human neutrophils in fibrin gels.

To study human neutrophil (polymorphonuclear leukocyte (PMN)) migration and killing of bacteria in an environment similar to that found in inflamed tissues in vivo, we have used fibrin gels. Fibrin gels (1500 microm thick) containing Staphylococcus epidermidis were formed in Boyden-type chemotaxis chambers. PMN migrated < 300 microm into these gels in 6 h and did not kill S. epidermidis when the gels contained heat-inactivated serum, C5-deficient serum, a streptococcal peptidase specific for a fragment of cleaved C5 (C5a), or anti-C5aR IgG. In contrast, in gels containing normal human serum, PMN migrated approximately 1000 microm into the gels in 4 h and into the full thickness of the gels in 6 h, and killed 90% of S. epidermidis in 6 h. fMLP reduced PMN migration into fibrin gels and allowed S. epidermidis to increase by approximately 300% in 4 h, whereas leukotriene B(4) stimulated PMN to migrate the full thickness of the gels and to kill 80% of S. epidermidis in 4 h. We conclude that both complement opsonization and C5a-stimulated chemotaxis are required for PMN bacterial killing in fibrin gels, and that fMLP inhibits PMN bactericidal activity in fibrin gels. The latter finding is surprising and suggests that in the presence of fibrin fMLP promotes bacterial virulence.     

Exudation induces clustering of CR1 receptors at the surface of human polymorphonuclear leukocytes

The complement receptor type 1 (CR1) surface distribution, density and immune adherence efficiency were determined in circulating PMN activated by fMLP, NAP-1/IL-8, TNF, GM-CSF and C5a, or exudate PMN harvested from skin-blisters. These observations were compared with those observed on resting peri-pheral blood PMN. PMN activators known to upregulate CR1 expression did not induce a significant increase in CR1 clustering, or immune adherence efficiency towards opsonized immune complexes. By contrast, increase in CR1 density at the surface of exudated PMN was accompanied by an increased clustering. This clustering was however insufficient to increase the binding efficiency for immune complexes. Eventually, CR1 expression of exudated neutrophil could not be increased further by stimulation with fMLP or PMA. These results indicated that clustering of CR1 on PMN may occur in vivo. Such reaction might determine the phagocytic potential of the cell for opsonized micro-organisms or debris. This clustering could not be attributed to one of the PMN activators tested.     

The cutaneous reverse Arthus reaction requires intercellular adhesion molecule 1 and L-selectin expression

The deposition of immune complexes (IC) induces an acute inflammatory response with tissue injury. IC-induced inflammation is mediated by inflammatory cell infiltration, a process highly regulated by expression of multiple adhesion molecules. To assess the role of L-selectin and ICAM-1 in this pathogenetic process, the cutaneous reverse passive Arthus reaction was examined in mice lacking L-selectin (L-selectin(-/-)), ICAM-1 (ICAM-1(-/-)), or both (L-selectin/ICAM-1(-/-)). Edema and hemorrhage, which peaked 4 and 8 h after IC challenge, respectively, were significantly reduced in L-selectin(-/-), ICAM-1(-/-), and L-selectin/ICAM-1(-/-) mice compared with wild-type littermates. In general, edema and hemorrhage were more significantly inhibited in ICAM-1(-/-) mice than in L-selectin(-/-) mice, but were most significantly reduced in L-selectin/ICAM-1(-/-) mice compared with ICAM-1(-/-) or L-selectin(-/-) mice. Decreased edema and hemorrhage correlated with reduced neutrophil and mast cell infiltration in all adhesion molecule-deficient mice, but leukocyte infiltration was most affected in L-selectin/ICAM-1(-/-) mice. Reduced neutrophil and mast cell infiltration was also observed for all mutant mice in the peritoneal Arthus reaction. Furthermore, cutaneous TNF-alpha production was inhibited in each deficient mouse, which paralleled the reductions in cutaneous inflammation. These results indicate that ICAM-1 and L-selectin cooperatively contribute to the cutaneous Arthus reaction by regulating neutrophil and mast cell recruitment and suggest that ICAM-1 and L-selectin are therapeutic targets for human IC-mediated disease.     

Dissociation between alveolar transmigration of neutrophils and lung injury in hyperoxia

The objective of this study was to quantitatively assess changes in cell adhesion molecule (CAM) expression on the pulmonary endothelial surface during hyperoxia and to assess the functional significance of those changes on cellular trafficking and development of oxygen-induced lung injury. Mice were placed in >95% O(2) for 0-72 h, and pulmonary injury and neutrophil (PMN) sequestration were assessed. Specific pulmonary CAM expression was quantified with a dual-radiolabeled MAb technique. To test the role of CAMs in PMN trafficking during hyperoxia, blocking MAbs to murine P-selectin, ICAM-1, or platelet-endothelial cell adhesion molecule-1 (PECAM-1) were injected in wild-type mice. Mice genetically deficient in these CAMs and PMN-depleted mice were also evaluated. PMN sequestration occurred within 8 h of hyperoxia, although alveolar emigration occurred later (between 48 and 72 h), coincident with rapid escalation of the lung injury. Hyperoxia significantly increased pulmonary uptake of radiolabeled antibodies to P-selectin, ICAM-1, and PECAM-1, reflecting an increase in their level on pulmonary endothelium and possibly sequestered blood cells. Although both anti-PECAM-1 and anti-ICAM-1 antibodies suppressed PMN alveolar influx in wild-type mice, only mice genetically deficient in PECAM-1 showed PMN influx suppression. Neither CAM blockade, nor genetic deficiency, nor PMN depletion attenuated lung injury. We conclude that early pulmonary PMN retention during hyperoxia is not temporally associated with an increase in endothelial CAMs; however, subsequent PMN emigration into the alveolar space may be supported by PECAM-1 and ICAM-1. Blocking PMN recruitment did not prevent lung injury, supporting dissociation between PMN infiltration and lung injury during hyperoxia in mice.     

Role of C5a in multiorgan failure during sepsis

In humans with sepsis, the onset of multiorgan failure (MOF), especially involving liver, lungs, and kidneys, is a well known complication that is associated with a high mortality rate. Our previous studies with the cecal ligation/puncture (CLP) model of sepsis in rats have revealed a C5a-induced defect in the respiratory burst of neutrophils. In the current CLP studies, MOF occurred during the first 48 h with development of liver dysfunction and pulmonary dysfunction (falling arterial partial pressure of O(2), rising partial pressure of CO(2)). In this model an early respiratory alkalosis developed, followed by a metabolic acidosis with increased levels of blood lactate. During these events, blood neutrophils lost their chemotactic responsiveness both to C5a and to the bacterial chemotaxin, fMLP. Neutrophil dysfunction was associated with virtually complete loss in binding of C5a, but binding of fMLP remained normal. If CLP animals were treated with anti-C5a, indicators of MOF and lactate acidosis were greatly attenuated. Under the same conditions, C5a binding to blood neutrophils remained intact; in tandem, in vitro chemotactic responses to C5a and fMLP were retained. These data suggest that, in the CLP model of sepsis, treatment with anti-C5a prevents development of MOF and the accompanying onset of blood neutrophil dysfunction. This may explain the protective effects of anti-C5a in the CLP model of sepsis.     

Neutrophil caveolin-1 expression contributes to mechanism of lung inflammation and injury

Caveolin-1 present in immune cells may be involved in regulation of the inflammatory response. Here, using caveolin-1-null (Cav-1(-/-)) mice, we addressed the role of caveolin-1 in polymorphonuclear neutrophils (PMNs) in regulating PMN activation-mediated lung injury. In lungs of wild-type (Cav-1(+/+)) mice perfused at constant flow with Krebs-Henseleit solution, addition of Cav-1(+/+) PMNs (4 x 10(6) cells) into the perfusate followed by their activation with formyl-Met-Leu-Phe (fMLP, 1.0 muM) plus platelet-activating factor (1.0 nM) increased pulmonary microvessel filtration coefficient by 150% and wet-to-dry lung weight ratio by 50% as well as PMN accumulation in lungs. These responses were markedly reduced in lungs perfused with Cav-1(-/-) PMNs followed by addition of the same activating agents. fMLP-stimulated adhesion of Cav-1(-/-) PMNs to pulmonary microvascular endothelial cells and migration of Cav-1(-/-) PMNs across endothelial monolayers were also impaired compared with Cav-1(+/+) PMNs. Cav-1(-/-) PMNs showed 50-80% reduction in PMA- or fMLP-stimulated superoxide production compared with Cav-1(+/+) PMNs. In addition, Cav-1(-/-) PMNs had decreased migratory activity (50%) and adhesion to fibrinogen (40%) in response to fMLP. Rac1 and Rac2 were activated in Cav-1(+/+) PMNs after stimulation of fMLP but not in Cav-1(-/-) PMNs. Exogenous expression of caveolin-1 in COS-phox cells augmented the fMLP-induced Rac1 activation and superoxide production, indicating a direct role of caveolin-1 in the mechanism of superoxide production. Thus caveolin-1 expression in PMNs plays a key role in mediating PMN activation, adhesion, and transendothelial migration and in PMN activation-induced lung inflammation and vascular injury.