INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE DISCREPANCY BETWEEN THE size of polymorphonuclear leukocytes (PMN) and the size of pulmonary capillary segments restricts the passage of PMN through the pulmonary vascular bed. PMN have a mean diameter larger than the mean diameter of pulmonary capillary segments (6, 16). Wiggs and colleagues (32) have shown that PMN of ~7 µm in diameter negotiate the pulmonary capillary bed in the same way as 4-µm nondeformable beads. This indicates that the PMN must deform to pass through the estimated 60-100 capillary segments encountered as they travel from the arterial to venous side of the pulmonary circulation (16-18).

Red blood cells (RBCs) have a similar maximum diameter as PMN, but their ability to deform rapidly by folding allows them to cross the pulmonary capillary bed ~100 times faster than PMN (19, 20). The pulmonary capillary bed is composed of short, interconnected segments that allow the faster moving RBC to stream around the slower moving PMN. The resulting increase in PMN concentration with respect to RBC accounts for the marginated pool of PMN in the lung (17). This marginated pool may be one to six times larger than the circulating pool (5, 17, 18, 20). Several in vitro and in vivo studies have shown that biophysical properties (stiffness and deformability) of PMN are the major determinants of the magnitude of PMN sequestration in the pulmonary capillaries (9-11, 29). This sequestration of PMN is the initial step in the recruitment of PMN into an inflammatory site in the lung. The accumulation of activated PMN within the lung microvessels may also play a key role in the pathogenesis of the acute lung injury characteristic of acute respiratory distress syndrome.

Kitagawa and colleagues (23) have shown that passive mechanical deformation of PMN results in an increase in the cell-surface expression of the adhesion molecules CD11b/CD18, an effect that can be enhanced by priming the PMN. Filtration of PMN through smaller 3-µm-pore-size filters had a greater effect than filtration through the 5-µm-pore-size filters, suggesting that the degree of deformation or biophysical stress is related to the increased adhesion molecule expression (23). The functional consequence of this deformation-induced increase in CD11b/CD18 expression is still unclear. The PMN adhesion molecules CD11b/CD18 facilitate PMN adhesion to vascular walls by binding to their inducible endothelial ligand, intercellular adhesion molecule-1 (ICAM-1) (1). This interaction is responsible for firm adhesion of PMN to the endothelium and also facilitates PMN migration into tissues (1, 28).

Our working hypothesis is that mechanical deformation of PMN, similar to that which occurs in pulmonary capillaries, enhances cell adhesion via the CD18/ICAM-1 interaction. To test this hypothesis, we used an in vitro filtration system to simulate the in vivo deformation of PMN in pulmonary capillaries. This system has been used by others (25), and we have modified it to study the effect of mechanical deformation on the CD18-dependent PMN adhesion. Adhesion assays were performed by using 96-well plates coated with recombinant ICAM-1 (rICAM-1), and flow cytometry was used to quantify the changes in adhesion molecule expression caused by mechanical deformation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials

Monoclonal antibodies and agonists. Conjugated antibodies against human CD11b [phycoerythrin (PE)-conjugated monoclonal mouse anti-human C3bi receptor, CD11b], IgG1/PE, and IgG1/FITC were purchased from DAKO (Dakopatts, Glostrup, Denmark). The conjugated antibody for L-selectin (CD62L, LECAM-1/FITC) was purchased from Immunotech. Blocking antibodies R15.7 (anti-CD18) were generously provided by Dr. Scott I. Simon (Baylor College of Medicine, Houston, TX). A soluble ICAM-1 construct consisting of cDNA coding for the five extracellular domains of ICAM-1 fused to a gene for part of the heavy chain constant region of mouse IgG2b was kindly provided by Dr. H. Hedman. Chinese hamster ovary K-1 cells transfected with the soluble ICAM-1 construct were grown up, and rICAM was purified from the cell culture as previously described (15).

Methods

Cell preparation. Human leukocyte-rich plasma (LRP) was prepared from citrate anti-coagulated venous blood obtained from infection-free healthy volunteers by dextran (molecular wt, 100,000-200,000, 4% final concentration; Sigma Chemical) sedimentation of RBC for 25-30 min. Resulting LRP was diluted in PMN buffer [(in mM) 138 NaCl, 2.7 KCl, 8.1 Na₂HPO₄ · 7H₂O, 1.5 KH₂PO₄, and 5.5 glucose, pH 7.4], and PMN were purified by hypotonic lysis of residual RBC with sterile water and 2× PBS (2× PBS is 27 nM Na₂HPO₄, 132 mM KH₂PO₄, and 2.74 M NaCl). PMN were then separated from the mononuclear cells by centrifugation in Histopaque (Sigma Chemical), with a density of 1.077 g/ml at 1,000 rpm for 13 min and were resuspended in PMN buffer. The isolated PMN were 95-98% pure, with a viability of 98% determined by trypan blue exclusion.

In vitro filtration of PMN. Purified PMN were filtered by using a method previously described (23, 25). Briefly, a 35-ml polypropylene syringe (Sherwood Medical, St. Louis, MO) was filled with cell solution (0.5 × 10⁶ cells/ml in 0.5% human albumin, 1× PBS) and then filtered through polycarbonate membrane filters (Poretics, Livermore, CA) with defined pore size (pore diameter 5 µm, length 10 µm, and pore density 4 × 10⁵ cm²; or pore diameter 3 µm, length 9 µm, and pore density 2 × 10⁶/cm²; manufacturer's data) by using a syringe-infusion pump (pump 22, Harvard Apparatus, Mills, MA) that provided a constant flow rate (3.0 ml/min) of solution across the filter. Hydrostatic pressure was monitored upstream by using a pressure transducer (Validyne Engineering, Northridge, CA) connected to a recording system. The system was calibrated by using water manometer under conditions of no flow before each filtration.

Adhesion assay. Ninety-six-well plates (Nunc, Immuno) were coated with 100 µl of 10 µg/ml concentration of rICAM-1 in Tris buffer and incubated overnight at 4°C. In pilot experiments, the homogeneity of coated wells was verified by incubating the wells with mouse anti-human ICAM-1 monoclonal antibodies (Sigma Chemical) and staining with the alkaline phosphatase anti-alkaline phosphatase method (4).

Blocking antibody studies. To test the role of the CD18/ICAM-1-interaction in PMN adhesiveness after filtration, an excess of blocking antibodies for CD18 (R15.7, 40 µl of a 20 µg/ml solution) was added to both filtered (3-µm-pore filters) and unfiltered cell solutions and incubated for 5 min before adhesion assays were performed. To block the ICAM-1 on coated plates, an excess of blocking antibody (anti-CD54, clone BBA3 from R&D Systems, Minneapolis, MN; 40 µl of a 20 µg/ml solution) was added to each well coated with rICAM-1 and incubated for 10 min. The plates were inverted to remove excess antibody before the addition of cell solution. In each experiment, untreated cells were added to rICAM-1 wells to act as controls.

PMN deformation as a priming stimulus. To test whether PMN deformation primed cells for enhanced adhesion, 5% ZAP was used to stimulate both filtered (3- and 5-µm-pore filters) and unfiltered PMN for 3 min before the adhesion assay. In each experiment, untreated cell solutions were used as a control.

Flow cytometry. PMN expression of CD11b and L-selectin in both LRP (23) and purified PMN was measured before and after filtration by indirect immunofluorescence. One hundred microliters of cell solution were incubated with 200 µl of PBS (pH 7.3) containing the FITC-conjugated antibodies against human CD11b (2 µg/ml), L-selectin (1.25 µl/ml), and the negative controls IgG1/R-PE (1 µg/ml) and IgG1/FITC (1 µg/ml). Cells were incubated in the dark for 10 min at room temperature, washed with PBS, fixed with 1% paraformaldehyde, and stored in the dark at 4°C. Adhesion molecule expression was measured by using flow cytometry (Profile Epics II, Coulter Electronics, Haleah, FL). Analysis gates for PMN were established by using distinctive forward- and side-scatter profiles, and results were expressed as mean fluorescence intensity (log) of 3,000-6,000 cells. Cells incubated with 10 nM fMLP served as a positive control.

Statistical analysis. Differences between prefiltered or prestimulated values for each condition were treated as individual parametric values, and groups of these values were compared with postfiltered or poststimulated values by using paired t-tests. Changes over time were analyzed by using ANOVA for repeated measures. Corrections for multiple tests and comparisons were performed by using the sequential rejective Bonferroni procedure (21). Corrected P values < 0.05 were considered significant. All values are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Pressures During PMN Filtration

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Pressure changes when purified polymorphonuclear leukocytes (PMN) were filtered through a 5-µm polycarbonate membrane. PMN (5 × 105 cell/ml) were filtered through polycarbonate filters before or after priming [N-formyl-methionyl-leucyl-phenylalanine (fMLP), 0.5 nM] or stimulation (fMLP, 10 nM), and pressure was measured continuously for 300 s. Both priming and stimulation caused a significant increase in plateau pressures (P < 0.01). Pressure tracings are representative of 7 experiments.

Flow Cytometry

Adhesion of Deformed PMN

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of deformation of PMN on their adhesion to intercellular adhesion molecule-1 (ICAM-1). Purified PMN were filtered through 5-µm (n = 7) and 3-µm (n = 5) polycarbonate filters, and their adhesion to recombinant human ICAM-1 (rICAM-1) was measured. Filtration through both 5-µm- (P < 0.04) and 3-µm-pore-size filters (P < 0.01) resulted in increased adhesion to ICAM-1. The nos. beneath the bars represent the calculated no. of PMN adherent to rICAM-1. All values are means ± SE.

To test the role of the CD18/ICAM-1 interaction in PMN adhesion after 3-µm filtration, blocking antibodies were added to filtered PMN- and ICAM-1-coated wells (Fig. 3). Filtration of PMN through 3-µm-pore-size filters caused a significant increase in adhesion to the ICAM-1-coated plates (232.4 ± 20.9%, postfiltered vs. baseline; P < 0.01; n = 5). Adding anti-CD18 blocking antibodies to filtered PMN reduced this deformation-induced adhesion to 140.1 ± 29.1% from baseline (P < 0.05; postfiltered vs. anti-CD18; n = 5), and blocking of rICAM-1 in the wells reduced adhesion to 10.1 ± 2.1% from baseline (P < 0.01; postfiltered vs. anti-ICAM-1; n = 5). Blocking of both CD18 and ICAM-1 did not reduce the adhesion more than blocking of ICAM-1 alone (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of blocking antibodies against CD18 and ICAM-1 on the adhesion of purified PMN deformed through 3-µm-pore-size filters (n = 5). PMN filtration through 3-µm filters caused an increase in their adhesion to rICAM-1 (P < 0.01) that was partially blocked by monoclonal antibodies to CD18 (P < 0.05) and nearly completely blocked by antibodies to ICAM-1 (P < 0.001). The nos. beneath the bars represent the no. of PMN adherent to rICAM-1. All values are means ± SE.

Effect of Priming on the Adhesion of Deformed PMN

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of priming PMN (fMLP, 0.5 nM) before deformation through 5-µm filters on their adhesion to rICAM-1 (n = 7). Priming of PMN alone does not increase adhesion to ICAM-1, but deformation of primed PMN significantly increases adhesion (P < 0.05). The nos. beneath the bars represent the no. of PMN adherent to rICAM-1. All values are means ± SE.

Figure 5 shows that deformation of PMN enhanced adhesion on subsequent stimulation (5% ZAP). Treatment of prefiltered PMN with 5% ZAP resulted in an increase in PMN adhesion (50.6 ± 4%). Stimulation of PMN after 5-µm filtration resulted in an increase in PMN adhesion compared with untreated filtered PMN (148.4 ± 26.7%; P < 0.05; n = 5). A further increase was observed when PMN were stimulated after 3-µm filtration (249.5 ± 24.3%; P < 0.01; n = 12). Filtration through both 5- and 3-µm filters caused a greater increase in adhesion of PMN to ICAM-1 than with ZAP stimulation alone (P < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Effect of PMN stimulation with 5% zymosan-activated plasma (ZAP) on adhesion to ICAM-1 after deformation through 5- and 3-µm-pore-size filters. Stimulation of PMN increased their adhesion to ICAM-1 (n = 5; P < 0.05) and was enhanced by prior deformation through both 5-µm (n = 5: P < 0.05) and 3-µm (n = 12; P < 0.01) filters. The adhesion was significantly more enhanced by 3- compared with 5-µm deformation (P < 0.05). The nos. beneath the bars represent the no. of PMN adherent to rICAM-1. All values are means ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Neutrophils circulating through the pulmonary microvessels must negotiate ~60 capillary segments as they travel from the arterial to the venous side of the pulmonary vascular bed. These segments have an average diameter of 7.4 ± 2.3 µm (humans) and range between 2 and 15 µm (6, 17). In this study, we have used 3- and 5-µm-pore-size filters to simulate the smaller 20-30% of pulmonary capillary segments and to determine the effect of deformation on PMN adhesion characteristics. The deformation procedure has been previously reported by our laboratory (23), and a similar procedure has been used by several other investigators (8, 10, 11, 13, 25, 29). Downey and colleagues (11) calculated that the shear stress in this in vitro filtration system was remarkably similar to the estimated wall shear stress in vivo and that flow rates of 3.0 ml/min in vitro are comparable to flow rates in the pulmonary microcirculation (6, 11, 20). All filters were coated with albumin, and cells were prepared and filtered by using techniques to eliminate any significant exposure of cells to endotoxin that could result in cell activation. Studies from our laboratory have shown that passive deformation of PMN by using this system causes upregulation of the adhesion molecule CD11b/CD18 (23). The hypothesis of this study was to determine whether these changes in CD11b also enhanced PMN adhesion properties.

The expression of adhesion molecules acts as a component of a molecular cascade in leukocyte-endothelial interaction and is regulated according to the activation status of PMN (1, 28). Known activation stimuli, such as chemoattractants, result in shedding of L-selectin and translocation of CD11b/CD18 (Mac-1) from intracellular storage pools to the cell surface (22). Therefore, the quantitative evaluation of these molecules reflects the PMN activation status. Similar to our laboratory's previous study using LRP (23), CD11b expression of PMN increased with filtration through 3-µm filters (Table 1). The purification process of PMN was associated with an increase in CD11b expression (Table 1) that blunted the quantitative changes in CD11b expression of PMN induced by deformation. Kuijpers and colleagues (24) also showed that changes in CD11b expression, resulting from the PMN purification process, were largely induced by the density centrifugation step. L-selectin expression, on the other hand, did not change as a result of the purification process (results similar to those of Kuijpers and colleagues) or filtration, suggesting that this adhesion molecule does not change with PMN deformation.

This differential effect of PMN deformation on cell adhesion regulation suggests that deformation does not affect the signaling pathways involved in L-selectin shedding. Molad and colleagues (27) recently demonstrated that immune complexes induced a similar type of differential effect on PMN adhesion molecule expression (impaired L-selectin shedding after stimulation). They postulated that upregulation of CD11b/CD18 with impaired L-selectin shedding causes a tighter adhesion between PMN and microvascular endothelium, delaying migration and causing vascular endothelial injury if associated with production of active oxygen metabolites (27). Furthermore, Simon and colleagues (14, 30) have recently shown that L-selectin-mediated adhesion may signal and promote subsequent CD11b/CD18 activity. The lack of change in L-selectin after PMN deformation suggests that the increased CD11b/CD18 adhesion that we have observed after PMN deformation is not mediated via L-selectin signaling.

Filtration of PMN through both 5- and 3-µm filters caused an increased adhesion of cells to microtiter plates coated with rICAM-1 (Fig. 2). The lack of increase in CD11b expression on PMN after filtration (using purified PMN) suggests that the increase in adhesion could be due to conformational changes in the CD11b/CD18 molecule. Several studies have shown that an increase in CD11b/CD18-dependent adhesion does not necessarily translate into an increase in surface expression of the molecule (31). The increase in PMN adhesiveness after passive deformation suggests that deformation causes conformational changes in CD11b/CD18 that augment their interaction with ICAM-1. Similarly, CD11a (lymphocyte function-associated antigen-1) on PMN can also bind to the ICAM-1 (26) and contribute to the adhesion of deformed PMN. Clearly further studies are needed to determine whether quantitative changes and/or conformational changes to the -subunits lymphocyte function-associated antigen-1 (CD11a) and Mac-1 (CD11b) contribute to the deformation-induced adhesion of PMN to the ICAM-1.

Integrin ligand binding is tightly regulated by cellular signaling mechanisms, referred to as integrin activation or inside-out signaling (3). This induced inside-out signaling could be initiated by cell deformation and may be responsible for the conformational changes and activation of the CD11b/CD18 receptor, resulting in increased adhesiveness. CD11b/CD18 receptors are linked to cytoskeletal elements in the cell (2). Previous studies have shown that passive deformation of PMN causes an increase in F-actin assembly and microtubule organizing center (23). These changes in the cytoskeletal elements may serve as a pathway for inside-out signaling and increased CD18/ICAM-1 interaction.

The specificity of the CD18 and ICAM-1 interaction was tested by using specific monoclonal antibodies against CD18 and ICAM-1 (Fig. 3). Anti-ICAM-1 antibodies nearly abolish all increased adhesion induced by PMN deformation, and incubating PMN with anti-CD18 reduced deformation-induced PMN adhesiveness by ~50% (Fig. 3). The anti-CD18 antibodies did not block all of the adhesive activity induced by cell deformation. The adhesive interaction(s) responsible for the remaining adhesiveness not blocked by CD18 antibodies suggests a CD18-independent adhesive interaction. The nature of this interaction is presently unclear and clearly needs further study.

Previous studies from our laboratory have shown that deformation of fMLP-primed PMN by 5-µm filtration causes an increase in CD11b/CD18 expression, in contrast to deformation of naive PMN (23). Our data show that priming PMN with the receptor agonist (fMLP) causes an ~100% increase in the adhesion of deformed PMN to ICAM-1, in contrast to an ~63% increase of naive PMN (Fig. 4). This suggests that priming of intravascular PMN by circulating inflammatory mediators promotes their adhesion in the lung microvessels. To test the hypothesis that deformation of PMN per se primes the cells for an increased response to subsequent stimulation, we deformed PMN through 3-µm filters before stimulation with complement fragments. The results clearly show that deformation before stimulation enhances PMN adhesiveness (Fig. 5) and that this response was dependent on the magnitude of deformation. Interestingly, only 1-2% of PMN that are delivered to a pneumonic region actually migrate into the alveolar spaces (7). The mechanism described here may be relevant to the selection of PMN that migrate in the lung. Deformation of these cells during their transit through the lung capillaries may sensitize them for subsequent stimulation and result in enhanced adhesion to the endothelium and preferential migration into the alveolar space.

In summary, this study shows that deformation of PMN similar to that caused by transit through the smaller 20% of the pulmonary capillary segments (5 µm in diameter) results in enhanced adhesion to ICAM-1, and the degree of enhancement is dependent on the magnitude of PMN deformation. It also shows that deformation of PMN enhanced their adhesive response to subsequent stimulation by a mechanism that is mediated predominantly through the CD11b/CD18 and ICAM-1 interaction. We conclude that deformation of PMN is an important step in the sequential events that lead to PMN migration from lung microvessels into alveolar spaces.