Signaling Pathways Involved in IL-8-Dependent Activation of Adhesion Through Mac-1¹

Cellular adhesion is a critical event required for the full activation and physiologic responses of neutrophils, including protease release, reduction of molecular oxygen to superoxide anion, initiation of the respiratory burst, and reorganization of the actin-based cytoskeleton (1, 2, 3, 4, 5). Neutrophil adhesion is dependent upon the expression, activation, and binding of the β₂ integrin family members LFA-1 and Mac-1 (1). Adhesion through the β₂ family of integrins is a complex process in which activating stimuli result in enhanced avidity of the β₂ integrins for their ligands (6, 7, 8, 9).

Integrin-dependent binding is stimulated through inside-out signaling pathways that are triggered by ligand binding to an array of receptors such as those for the chemokines IL-8, C5a, and FMLP, and cytokine receptors such as TNF (7, 8, 9, 10). Enhanced adhesion through LFA-1 and Mac-1 can also be achieved by direct activation of protein kinase C (PKC)³ by the tumor-promoting agent PMA (7, 8, 9, 10). Engagement of inflammatory receptors as well as activation of PKC initiates a signal transduction cascade that results in inside-out activation of the integrin and enhanced cellular adhesion (11).

IL-8 is an important endothelial (12)- and epithelial (13, 14, 15)-derived inflammatory mediator that induces neutrophil chemotaxis and stimulates neutrophil transmigration. IL-8 initiates its effects by binding to specific receptors that are members of the seven-transmembrane or serpentine receptor family expressed on the surface of neutrophils (16, 17). IL-8 has been shown to trigger increased expression of CR1 (18) and β₂ integrin complexes in neutrophils (19, 20). In addition, IL-8 has been shown to stimulate β₂ integrin-mediated arrest of rolling neutrophils in vivo (21, 22, 23).

Although the cellular functions induced by IL-8 in neutrophils have been established, the signal transduction pathways that mediate these activities remain to be defined. IL-8 induces the mobilization of calcium in neutrophils (24, 25), and this may occur through the activation of phospholipase C-β (26). There is also evidence that IL-8 stimulates phospholipase D activity (27), phosphatidylinositol-3 kinase (PI3K) activity (28), and GTP loading of RhoA (29) in human neutrophils. IL-8 has been shown to activate mitogen-activated protein kinase (MAPK) in human neutrophils as well as in enucleate neutrophil cytoplasts (28, 30, 31). In addition, IL-8 has been shown to stimulate GTP loading of Ras and Raf activation in human neutrophils with a time course that indicates that these events are upstream of MAPK activation. All three events, Ras, Raf, and MAPK activation, appear to be dependent on PI3K activation (28). Both MAPK/extracellular signal-related kinase (ERK)-activating enzyme-1 (MEK-1) and MEK-2 isoforms are activated in human neutrophils in response to the chemokine FMLP (32, 33); this has not been directly demonstrated in response to IL-8.

MAPK activation in response to IL-8 stimulation of neutrophils has not yet been well targeted to specific physiologic or biochemical responses in the cell. There is evidence that IL-8-stimulated neutrophil migration across filters is independent of the MAPK pathway, but is dependent upon PI3K activation (34). Homotypic aggregation of neutrophils stimulated by arachidonic acid or the chemotactic factor FMLP appears to correlate with ERK kinase activity in these cells (35). Salicylates and the MEK inhibitor PD098059 have been shown to inhibit FMLP-stimulated ERK activation as well as neutrophil aggregation and adhesion to endothelial cell monolayers (36). MAPK activation appears to be required for FMLP-activated phagocytosis in human neutrophils (37); however, PMA-induced oxidative burst and FMLP-induced degranulation appear to be independent of the MAPK cascade (38, 39).

In this study, we determined the specific signaling pathways activated by IL-8 in neutrophils and their association with Mac-1-dependent adhesion. We demonstrate that both MAPK and PKC are rapidly activated upon IL-8 stimulation and that both of these enzyme activities are critical for Mac-1 activation and neutrophil adhesion. PI3K lies upstream of MAPK, but preservation of its activity does not appear critical to pathways that directly activate PKC. The MAPK and PKC pathways involved in integrin-mediated adhesion appear to be discrete, but may interact and cross over at some level, possibly fine-tuning the signals and impacting the strength of adhesion, transmigration, or other physiologic functions of the neutrophil.

The β₂-specific stimulatory Ab CBR LFA-1/2 is a mouse anti-human mAb that was isolated and purified, as described previously (40). The anti-phosphotyrosine Ab 4G10 was purchased from Upstate Biotechnology (Lake Placid, NY). The inhibitory anti-CD11b (Mac-1 α subunit) A44 mAb was a generous gift from Dr. R. Todd III (University of Michigan, Ann Arbor, MI) and TS1/22, anti-LFA-1α-inhibitory mAb, was obtained from Dr. T. Springer (Center for Blood Research, Harvard Medical School, Cambridge, MA). Human IL-8 (72 and 77 aa) was purchased from PeproTech (Princeton, NJ). Antiactivated pan-PKC rabbit polyclonal Ab, δ-PKC-specific blocking peptide, leupeptin, aprotinin, Triton X-100, bisindolylmaleimide (BIM-I), and PMA were obtained from Calbiochem (La Jolla, CA). Benzamidine, LY294002, wortmannin, FMLP, and fibrinogen were purchased from Sigma-Aldrich (St. Louis, MO). The rabbit polyclonal Ab against the activated δ-PKC was purchased from Cell Signaling (Beverly, MA). Antiactivated p44/ERK1 and p42/ERK2 (MAPK) rabbit polyclonal Ab was purchased from Promega (Madison, WI). HRP-linked goat anti-rabbit IgG was purchased from Bio-Rad (Hercules, CA), and the HRP-linked goat anti-mouse IgG Ab was obtained from Life Technologies (Gaithersburg, MD). PD098059 was a gift from Parke Davis, now Pfizer Pharmaceuticals (Ann Arbor, MI).

Human neutrophils were isolated from whole blood after dextran sedimentation, Ficoll gradient centrifugation, and hypotonic lysis of RBCs, as previously described (41). Neutrophils (5 × 10⁶ cells/ml) were resuspended in HBSS supplemented with 10 mM HEPES and 2 mM MgCl₂, as described previously (42).

Purified human fibrinogen (0.5 mg/ml in PBS) or purified human ICAM-1 in PBS was adsorbed to polystyrene 96-well plates (Linbro/Titertek; ICN, Aurora, OH) for 1.5 h at room temperature. Unbound protein was aspirated, and the plate was rinsed with PBS containing 1% Tween 20. After 2 min, the plates were washed three times with PBS. Human neutrophils (2 × 10⁶/ml) were labeled by preincubation with 1.7 μg/ml 2′,7′-bis(2-carboxyethyl)-5(and -6)-carboxyfluorescein acetoxymethyl ester (Molecular Probes, Eugene, OR) for 30 min at room temperature. For blocking studies, cells were preincubated for 10 min at room temperature and subsequently stimulated in the presence of A44 ascites (1:100), TS1/22 (10 μg/ml), PD098059 (25 μM), EDTA (10 mM), or LY294002, wortmannin, or BIM-I at the concentrations indicated in the figure legends.

Neutrophils were added to wells containing PMA (50 ng/ml), CBR LFA-1/2 (25 μg/ml), human IL-8 (200 ng/ml), or FMLP (1 μM) with or without inhibitor and centrifuged at 60 × g for 1 min. The total fluorescent content of the cells in each well was assessed in a fluorescent concentration analyzer (Cytofluor; PerSeptive Biosystems, Framingham, MA), and plates were incubated as indicated in the figure legends. Unbound cells were removed by the addition of HBSS supplemented with 2 mM MgCl₂ and 10 mM HEPES (pH 7.5) and aspirated with a 21-gauge needle four times at 90° intervals around the well at room temperature. Bound cells were quantitated in the fluorescent concentration analyzer, and data are expressed as percentages of bound to total input cells per well. Each binding condition was assessed in triplicate.

Fibrinogen (0.5 ml; 0.5 mg/ml) in PBS was adsorbed to 24-well, 17-mm/well polystyrene plates (Falcon 3847; BD Labware, Lincoln Park, NJ) for 1.5 h at room temperature, as described above. Neutrophils (0.5 ml; 5 × 10⁶ cells/ml) were layered onto each plate, and cells were stimulated by the addition of PMA (50 ng/ml), CBR LFA-1/2 (25 μg/ml), IL-8 (200 ng/ml), or FMLP (1 μM) at 37°C for 2–60 min, as indicated in the legends. For blocking studies, cells were preincubated for 10 min at room temperature, and subsequently stimulated in the presence of A44 ascites (1:100), TS 1/22 (10 μg/ml), PD098059 (25 μM), EDTA (10 mM), LY294002, wortmannin, or BIM-I at indicated concentrations.

After stimulation, adherence was confirmed visually by light microscopy. Unbound cells were aspirated and collected by centrifugation for 2 min in a microcentrifuge at 350 × g. Bound cells were removed by incubation with 2× SDS sample buffer (120 mM Tris, pH 6.8, 10% glycerol, 3.4% SDS, and 5% 2-ME) at 100°C, and lysates were added to the corresponding unbound cells to maintain an equal cell number in each sample. After 10 min at 100°C, samples were centrifuged at 17,500 × g in an Eppendorf microcentrifuge at 4°C for 15 min, and the supernatant was subjected to 10% SDS-PAGE. The proteins were transferred to nitrocellulose, and the filter was analyzed by immunoblotting with antiactivated MAPK or antiactivated PKC Abs. Bound Ab was detected with HRP-linked secondary Ab and ECL according to the manufacturer’s directions (Amersham, Arlington Heights, IL).

Graphic data are expressed as mean ± SE. Statistical analysis was performed using Student’s t test. All p values <0.05 were considered to be significant.

Although both of the β₂ integrins, Mac-1 and LFA-1, are present on the cell surface of neutrophils, there is emerging evidence to suggest that they play separate and distinct roles in the recruitment of neutrophils from the circulation (43). To elucidate the relative contribution of Mac-1 and LFA-1 in neutrophil adhesion to fibrinogen, a reported Mac-1 ligand, neutrophils were layered onto fibrinogen-coated polystyrene plates and stimulated with PMA, the β₂-activating Ab CBR LFA-1/2, or IL-8. IL-8 markedly enhanced neutrophil adhesion to fibrinogen, and adhesion was blocked by Ab to CD11b, indicating that the IL-8 effect is dependent upon the engagement of Mac-1 (Fig. 1,A). The LFA-1-specific blocking Ab TS1/22 did not affect neutrophil adhesion to fibrinogen with any of the stimuli examined, confirming that LFA-1 most likely does not play a role in neutrophil adhesion to fibrinogen (Fig. 1 B). IL-8-stimulated adhesion to purified ICAM-1, a ligand recognized by both LFA-1 and Mac-1, was dramatically blocked by A44, with only a small additional inhibition when TS1/22 was added to block LFA-1. This suggests that LFA-1 may contribute in part to the complete adhesive event in this setting. These data confirm that stimulation of human neutrophils with IL-8 activates both β₂ integrins expressed on the cell surface, and that the individual role of each integrin in neutrophil adhesion may depend upon the ligand available. A similar pattern of adhesion was observed when cells were allowed to settle on fibrogen before stimulation (data not shown). To isolate the effect of IL-8 stimulation on the activity of Mac-1 specifically and to elucidate the intracellular signals that play a role, the remaining experiments were conducted on fibrinogen-coated surfaces.

MAPK activity was assessed in lysates of neutrophils stimulated with PMA, CBR LFA-1/2, FMLP, or IL-8. PMA, IL-8, and FMLP stimulated rapid activation of MAPK in these cells. The MAPK activation was completely abolished in the presence of the MEK inhibitor, PD098059, indicating the involvement of MEK as an upstream component of the MAPK cascade (Fig. 2,A). IL-8-stimulated MAPK activation was detected at 10 min and diminished by 30 min (Fig. 2,B). In contrast, PMA-stimulated MAPK activation was detected at 10 min, enhanced at 30 min (Fig. 2 B), and sustained at 60 min, the longest time point analyzed (data not shown). The β₂-activating Ab, that stimulates integrin-dependent adhesion independent of intracellular signals, did not stimulate MAPK activation. This suggests that MAPK activity is more important for inside-out signals that activate the integrin than for outside-in signals that are associated with integrin-dependent adhesion. When we examined other related signaling pathways, we noted that p38 MAPK was activated in unstimulated neutrophils and we detected only a slight and inconsistent enhancement of the activated form JNK. Neither of these activities was affected by PD098059 (data not shown).

To determine whether MAPK activation is required for neutrophil adhesion, polymorphonuclear neutrophils (PMN) were preincubated with and stimulated in the presence of the MEK inhibitor that was shown to inhibit ERK activation in neutrophils at 25 μM. Adhesion to fibrinogen was inhibited 50% by PD098059 in cells stimulated with either FMLP or IL-8. In contrast, the MEK inhibitor did not affect neutrophil adhesion stimulated by PMA or CBR LFA-1/2 (Fig. 3). These data support a role for MAPK in mediating Mac-1-dependent neutrophil adhesion stimulated by IL-8. Although PMA treatment of cells results in the activation of MAPK, PMA-stimulated neutrophil adhesion was not affected by inhibition of MAPK. This may be due to the fact that PMA, as a potent pharmacologic stimulator of PKC, overrides other signals in the cell in a manner less sensitive to physiologic inhibition. Alternatively, PKC may have a direct stimulatory effect on the integrin receptor that is independent of MAPK activation. The active form of p38 MAPK was detected in the absence and presence of activators; inhibition with SB203580 had no effect on adhesion (data not shown).

The activation of ERK in human neutrophils has been shown to be dependent upon PI3K activity (34), and PI3K has been implicated in the activation of cell migration in platelets (44), renal epithelial cells (45), and human T lymphocytes (46). Although neutrophil migration requires PI3K activity, it appears to be independent of MAPK activity, suggesting a dissociation of the pathways at the level of PI3K (34). To further clarify the role of PI3K and the relationship of its activation to MAPK activity in IL-8-stimulated neutrophil adhesion on fibrinogen, we used LY294002 as a specific competitive inhibitor of PI3K in the adhesion assay. PI3K inhibition diminished β₂-activating mAb-stimulated neutrophil adhesion to ∼60% of control, and reduced IL-8-stimulated adhesion to 25% of control. PMA-stimulated neutrophil adhesion was not affected (Fig. 4,A). These data demonstrate a role for PI3K in the inside-out signals mediating IL-8-stimulated integrin activation. In addition, both LY294002 and the fungal metabolite, wortmannin, a somewhat less specific inhibitor of PI3K, inhibited IL-8-stimulated MAPK activation in these cells (Fig. 4, B and C). Of note is that inhibition of PI3K did not affect PMA-stimulated MAPK activation (Fig. 4 B). These data place PI3K upstream to MAPK in the IL-8-stimulatory pathway mediating integrin activation in neutrophils.

MAPK inhibition decreased IL-8-stimulated neutrophil adhesion by ∼50%. Thus, it remained likely that an additional signaling pathway that could modulate Mac-1-dependent adhesion was stimulated by IL-8. Since direct activation of PKC is well known to stimulate β₂ integrin-dependent adhesion and IL-8 is known to activate PKC, we next examined whether this pathway was playing a role in IL-8-stimulated β₂ integrin activation. Using BIM-I to inhibit PKC activity, we found that both PMA- and IL-8-stimulated neutrophil adhesion was inhibited in a dose-dependent manner (Fig. 5,A). It should be noted that in vitro concentrations of BIM-I greater than 2 μM are not fully specific for inactivation of PKC; thus, the near total blockade of neutrophil adhesion in the presence of 5 μM BIM-I may reflect inhibition of other pathways. Although inhibition of either MAPK or PKC resulted in partial inhibition of IL-8-stimulated PMN adhesion, the combination of BIM-I and PD098059 at kinase-specific concentrations completely abolished adhesion to ligand (Fig. 5 B). These data implicate a role for both pathways in the observed enhanced adhesion in response to IL-8.

PKC activity was assessed in lysates of neutrophils utilizing Ab specific for the activated phosphorylated isoforms of PKC. Cells were stimulated with PMA, CBR-LFA-1/2, or IL-8. PMA and IL-8 stimulated rapid activation of a 74-kDa species recognized by the Ab directed against the active form of PKC (Fig. 6, A and B). The IL-8 response is present after 2 min of stimulation, diminished by 5 min, and markedly reduced by 30 min. PMA-stimulated PKC activity, also present at 2 min, is sustained for up to 30 min (Fig. 6,C). This 74-kDa protein may be a novel or atypical isoform of PKC, as it is activated in cells stimulated in Ca²⁺-free medium, and is not perturbed in the presence of the calcium chelator EDTA (Fig. 7,C). This 74-kDa species is also recognized by an Ab specific for the activated form of δ-PKC in immunoblots (Fig. 6,D). In addition, the 74-kDa band is absent when the immunoblot is performed in the presence of a purified δ-PKC-blocking peptide (Fig. 6 D).

Our data demonstrate that PI3K activation is necessary to achieve MAPK activation in IL-8-stimulated integrin-dependent adhesion of neutrophils (Fig. 4, A and B). To assess the relationship between PKC activation and PI3K activity, LY294002 was used as a specific competitive inhibitor of PI3K and lysates were assayed for the appearance of the activated phosphorylated form of PKC. Inhibition of PI3K in this manner did not affect either PMA- or IL-8-stimulated PKC activation in these cells (Fig. 8,C). Inhibition of PI3K by LY294002 did not affect PMA-stimulated MAPK activation in these cells (Fig. 4 B). Thus, the MAPK and PKC pathways may branch before PI3K activation. These data also leave open the possibility that MAPK can be activated via PKC independently of the IL-8-PI3K pathway.

We have demonstrated that both MAPK and PKC activity are necessary for stimulation of Mac-1-mediated neutrophil adhesion in response to IL-8 (Figs. 3 and 5). We hypothesized that MAPK and PKC activity are components of the inside-out signals generated at the IL-8R and targeted toward integrin activation; this is suggested by the observation that the β₂-activating Ab, which bypasses intracellular signals to activate the integrin, does not activate MAPK or PKC (Figs. 2 and 6). However, as integrins have been shown to generate outside-in signals subsequent to integrin-ligand binding, it was important to further clarify the roles of MAPK and PKC as inside-out signals leading to integrin activation or outside-in signals mediating cellular responses subsequent to integrin-ligand binding. We used two agents that disrupt integrin-ligand binding, A44, the Mac-1-specific blocking Ab, and EDTA, which chelates divalent cations, that are required for integrin-ligand binding (47, 48, 49, 50). Integrin-mediated neutrophil adhesion to fibrinogen is diminished to baseline when cells are treated with A44 (Fig. 1, A and B) or EDTA (data not shown). In addition, both of these agents have been shown to reverse protein tyrosine phosphorylation resulting from integrin-ligand binding (31). Neither MAPK (Fig. 7, A and B) nor PKC (Fig. 7, C and D) activity was significantly diminished in lysates of neutrophils preincubated and stimulated in the presence of these agents. These data support that both the MAPK and PKC pathways are components of the inside-out signals generated at the IL-8R, rather than outside-in signals initiated subsequent to integrin-ligand binding.

To assess the possible interactions between these two pathways, we first looked at MAPK activity in lysates of neutrophils stimulated in the presence of 10 μM of the PKC inhibitor BIM-I. BIM-I abolished PMA-stimulated MAPK activity. In contrast, treatment with BIM-I did not diminish IL-8-stimulated MAPK and may even sustain it, resulting in an enhanced signal at 30 min (Fig. 8, A and B). In parallel experiments, we examined PKC activity in cells treated with PD098059 to inhibit the MAPK cascade. PMA-stimulated PKC activation was not affected by MAPK inhibition, whereas IL-8-stimulated PKC activity appeared diminished (Fig. 8, C and D). Thus, while IL-8-stimulated MAPK pathway appears to be independent of PKC activation, the activation or preservation of the 74-kDa PKC isoform may require active MAPK.

The signaling mechanisms through which IL-8 stimulates integrin activation and subsequent cellular adhesion of human neutrophils have yet to be defined and were probed in this work. Previous work has demonstrated that chemokine receptor activation results in enhanced adhesion through both LFA-1 and Mac-1 (43). In this study, we demonstrate that IL-8 activates adhesion on fibrinogen and ICAM-1, and that adhesion to the former occurs through Mac-1. Adhesion to the latter was abolished only by the simultaneous addition of blocking mAb to Mac-1 and LFA-1, confirming that IL-8 activates both integrins.

IL-8 stimulation results in rapid activation of MAPK that is dependent upon activation of MEK, as supported by the observation that ERK1 and 2 (but not p38 or JNK) phosphorylation in response to IL-8 was completely abolished by the MEK inhibitor PD098059. Our results support a model in which PI3K lies upstream to MAPK in the IL-8-stimulatory pathway (Fig. 4,B) and in which Mac-1-mediated neutrophil adhesion to fibrinogen in response to IL-8 is dependent upon the activity of both of these kinases (Figs. 3 and 4 A). This was supported by the observation that inhibition of PI3K activity by the inhibitor LY294002 completely abolished activation of MAPK by IL-8. This is somewhat in contrast to the findings of Knall et al. (28, 34), who demonstrated that neutrophil migration across membranes coated with HSA is independent of MAPK activity. Transmigration is a complex process involving both integrin-mediated adhesion as well as subsequent deadhesion. We therefore chose to isolate the adhesive phenomenon and use a specific integrin ligand to dissect the signaling pathways important for this integrin-mediated component of neutrophil function. Thus, MAPK activation may be an important initial signal required for activation of Mac-1 and neutrophil adhesion, both of which are necessary for transmigration. Similarly, although Jones et al. (51) show that expression of activated Mac-1 on the cell surface of neutrophils stimulated with IL-8 in suspension is PI3K independent, we have focused our work on Mac-1-mediated adhesion to a specific ligand. This event may require alterations in cellular morphology and distinct signaling pathways.

As noted in our results, we demonstrated ∼50% inhibition of IL-8-stimulated neutrophil adhesion in the presence of concentrations of the MEK inhibitor that reduced MAPK activation to unstimulated levels (Figs. 2,A and 3). This suggests that a parallel pathway may contribute to the signals generated by engagement of the IL-8R that lead to integrin activation. We therefore probed PKC as a candidate and have demonstrated that PKC activity is critical for the full adhesive response to IL-8 stimulation of the neutrophil. The MAPK and PKC pathways appear to work independently in that only inhibition of both of these pathways brings cellular adhesion down to baseline levels (Fig. 5,B). Although Laudanna et al. (52) have shown that PKCζ is important for IL-8-stimulated neutrophil adhesion to fibrinogen, this is the first study to date that examines the synergy between MAPK and PKC pathways in this setting. The PKC activity in IL-8- and PMA-stimulated neutrophil lysates migrates at a molecular mass of 74 kDa, and activation is independent of exogenous Ca²⁺ (Figs. 6,A and 7C). Thus, we may be detecting a novel or atypical PKC, and future studies will be directed at identifying this isoform. The 74-kDa species recognized by the Ab directed against pan-activated PKC is also recognized by Ab specific for the activated form of δ-PKC, and recognition by the Ab is blocked in the presence of a δ-PKC-blocking peptide. These data suggest that the 74-kDa band (Fig. 6 D) may be a cleavage product of δ-PKC.

Both MAPK and PKC activation appear to be inside-out signals resulting in integrin-dependent adhesion, rather than outside-in signals initiated by integrin-ligand binding (Fig. 7). The MAPK and PKC pathways are discrete; inhibition of both of these pathways is required to decrease cellular adhesion to baseline levels. However, the two pathways may interact and cross over at some level. We have found that when PKC is inhibited, IL-8-stimulated MAPK activity is sustained in the cell (Fig. 8,A). There are at least two possible explanations for this: 1) the inhibition of the PKC pathway leads to enhanced utilization of the MAPK pathway, or 2) the PKC pathway is responsible for the activation of a phosphatase that contributes to the dephosphorylation of activated MAPK. Interestingly, inhibition of MAPK with the MEK inhibitor diminishes PKC activation in response to IL-8 (Fig. 8, C and D). Thus, MAPK activity may be important for the preservation or production of the 74-kDa PKC species recognized by specific Abs in activated cells.

We have shown that IL-8 stimulation of human neutrophils results in activation of a complex array of signals that are important for integrin activation and subsequent cellular adhesion. Two major components of this array are MAPK and PKC. PI3K lies upstream to MAPK (Fig. 4,B), but not PKC (Fig. 8 C) activation, suggesting that there is a branch point that precedes PI3K activation. Activation of MAPK by IL-8 is dependent on the preservation of PI3K activity, whereas activation by PMA is not. Although the MAPK and PKC pathways can be stimulated independently of each other, there may be some subtle communication between the pathways. This may be mediated by phosphatases that fine-tune the signals and impact on the strength of adhesion, transmigration, or specific cellular morphologies pertinent to neutrophil function.

We are grateful to Dr. Robert F. Todd III for his generous gift of mAb A44 and to Lisa Cummins and Sue Scott for preparation of this manuscript.