Lymphocyte transendothelial migration (TEM) is promoted by fluid shear signals and apical endothelial chemokines. Studying the role of these signals in neutrophil migration across differently activated HUVEC in a flow chamber apparatus, we gained new insights into how neutrophils integrate multiple endothelial signals to promote TEM. Neutrophils crossed highly activated HUVEC in a β2 integrin-dependent manner but independently of shear. In contrast, neutrophil migration across resting or moderately activated endothelium with low-level β2 integrin ligand activity was dramatically augmented by endothelial-presented chemoattractants, conditional to application of physiological shear stresses and intact β2 integrins. Shear stress signals were found to stimulate extensive neutrophil invaginations into the apical endothelial interface both before and during TEM. A subset of invaginating neutrophils completed transcellular diapedesis through individual endothelial cells within <1 min. Our results suggest that low-level occupancy of β2 integrins by adherent neutrophils can mediate TEM only if properly coupled to stimulatory shear stress and chemoattractant signals transduced at the apical neutrophil-endothelial interface.

The protective host response against pathogens and injuries relies on the rapid recruitment of circulating neutrophils from blood through postcapillary venules (1). Although intensely investigated, the mechanisms underlying neutrophil migration across inflamed endothelial barriers are still largely elusive (2). Intravital microscopy experiments have led to a fundamental understanding of the adhesive cascades that regulate neutrophil arrest on target endothelial sites during various inflammatory processes (3). However, the molecular basis of neutrophil diapedesis at these sites has been difficult to assess in these models, because leukocyte extravasation involves both adhesive and migratory steps. Therefore, in vitro models were developed to monitor neutrophil adhesion and migration across specifically stimulated endothelial monolayers at a single-cell level (4, 5). However, most of these studies were conducted in the absence of physiological flow conditions and thus did not address whether the hydrodynamic environment of endothelial-adherent leukocytes modulates their ability to adhere and pass through defined endothelial barriers. To address whether and how shear flow affects the ability of adherent leukocytes to cross endothelial barriers, in vitro flow chamber systems have been introduced recently (6, 7). Using shear flow assays, we have demonstrated a key role played by fluid shear stress in promoting the migration of PBL across cytokine-stimulated endothelial barriers presenting chemokines on their apical surface (8). This study also indicated that chemokine gradients across the endothelial barrier are not required for lymphocyte transendothelial migration (TEM).3 Rather, shear-triggered mechanosignals are necessary to direct adherent lymphocytes to cross the endothelial barrier, a process we termed “chemo-rheo-taxis” (8). A similar mechanism is apparently used by eosinophils (9). Shear flow was also shown to enhance neutrophil migration across cytokine-activated HUVEC (10). However, in contrast to lymphocyte and eosinophil TEM, neutrophil TEM occurred also under shear-free conditions and was only partially accelerated in the presence of shear flow. Other studies indicated that monocyte migration across inflamed endothelium can also take place in shear-free environments (11, 12). Taken together, these findings suggested that different types of leukocytes use distinct TEM mechanisms with different requirements for shear stress signals.

In the present study, we investigated the dependence on shear stress signals of neutrophil TEM by following neutrophil migration across distinct HUVEC barriers. Our results indicate that neutrophil migration across differently activated HUVEC, although β2 integrin dependent, varies in its dependence on shear stress signals. Shear contribution to TEM is tightly governed by the endothelial cell (EC) activation state and is dictated by the level of β2 integrin occupancy at apical neutrophil-endothelial interfaces. Shear and chemoattractant signals appear required only for neutrophil migration across nonactivated or weakly cytokine-stimulated ECs, whereas chronically activated ECs expressing elevated β2 integrin ligands and E-selectin promote rapid shear-independent neutrophil TEM.

Human neutrophils and PBL were isolated from citrate-anticoagulated whole blood from healthy donors as described (13). Cells were stored in binding medium (H/H medium supplemented with Ca2+ and Mg2+, both at 1 mM) and used within 3 h. HUVEC were isolated from umbilical cord veins, and established as primary cultures as described (8).

Ligand expression on apical HUVEC surfaces was assayed by FACS analysis as described (14). Briefly, HUVEC were cultured in fibronectin-coated 96-well plates. Confluent cells were left untreated or stimulated with different doses of TNF-α for 4–24 h, washed, and shortly overlaid with platelet-activating factor (PAF; 10−7 M; Sigma-Aldrich, St. Louis, MO). Cells were then washed and incubated with the following mAbs (at 10 mg/ml, 30 min, 4°C): anti-ICAM-1 (clone 6.5B5; DakoCytomation, Glostrup, Denmark), anti-ICAM-2 (B-T1; IQP Corporation, Groningen, The Netherlands), or anti-E-selectin (H4/18) (Ref.15 ; a gift of Dr. F. Luscinskas (Brigham and Women’s Hospital, Boston, MA)). mAb binding was assessed by PE-labeled goat anti-mouse Abs (Jackson ImmunoResearch Laboratories, West Grove, PA). Cells were removed from the plates using cold EDTA and immediately analyzed on FACScan (BD Biosciences, San Jose, CA).

Primary HUVEC (passage 2 or 3) were plated at confluence on petri dishes or glass slides spotted with human fibronectin. Cells were left intact for 24 h, followed by stimulation with 2 U/ml TNF-α for 4 h (moderately activated HUVEC) or medium control (nonactivated HUVEC), or were stimulated with 200 U/ml TNF-α for 24 h before experimentation (highly activated HUVEC). ECs were extensively washed before being assembled in the flow chamber. All flow experiments were performed at 37°C. The TEM assays under shear flow were previously described (16). PAF (10−7 M), and the chemokine growth-related oncogene α (Groα; CXCL1; 1 μg/ml) or IL-8 (CXCL8; 0.05–10 μg/ml), both from R&D Systems (Minneapolis, MN), were overlaid for 5 min on a HUVEC monolayer assembled in the flow chamber and washed extensively. Neutrophils were perfused over the EC monolayer at 0.75 dyn/cm2 for 40 s (accumulation phase, counted from the time the first neutrophil landed on the HUVEC surface), and then left in either shear-free conditions or under constant shear (5 dyn/cm2, the mid range of stresses in postcapillary venules, found to support optimal TEM in these settings) throughout the assay. All images were videorecorded through a ×20 phase contrast objective at 1 frame/s (a field of 540 × 400 μm). Endothelial-attached neutrophils, which accumulated in the field of view during the first 40-s phase, were individually followed throughout the assay and then categorized into four groups: leukocytes that rolled away or detached during the assay (detaching); leukocytes that remained stationary throughout the assay (arrested); leukocytes that locomoted over the HUVEC surface without crossing the monolayer (locomoting); and neutrophils that underwent stepwise darkening of their leading edge and remained motile underneath the EC (transmigrating; videos 2–4). The different categories were expressed as fractions of originally accumulated neutrophils. Adherent neutrophils were also analyzed 10 s after elevation of shear stress, and those that did not detach or roll away immediately after shear elevation were considered firmly arrested. PBL migration over stromal cell-derived factor-1-presenting activated HUVEC was assayed as previously described (8). Leukocyte locomotion over the HUVEC monolayer before TEM was determined for all accumulated leukocytes by measuring distances passed from the initial point of arrest during 1.5 or 3 min (for neutrophils or lymphocytes, respectively).

For Ab treatment, neutrophils were preincubated (5 min; room temperature) with 20 μg/ml of the appropriate mAbs, and perfused through the flow chamber in binding medium containing 1 μg/ml of the mAbs. The mAbs TS 1/18, TS1.22, and CBRM1/2 (gifts of Dr. T. Springer (Harvard, Cambridge, MA)) and the mAb Bear-1 (a gift of Dr. Y. Van Kooyk, Vrije Universiteit Medical Center, Amsterdam, The Netherlands) were used to functionally block the β2, αL, and, αM integrin chains, respectively. The anti-P-selectin glycoprotein ligand-1 mAb KPL-1 (BD Pharmingen, San Diego, CA) was used as control mAb. The mAb CBRM1/5 was used to monitor Mac-1 activation on neutrophils. Neutrophil PAF-R was blocked with the specific antagonist WEB-2086 (Ref.17 ; a kind gift of Boehringer Ingelheim (Biberach, Germany)) at 10 μM for 15 min. The inhibitor was also included in the perfusion solution throughout the assay. PAF-R on HUVEC was blocked with 10 μM WEB-2086 before and during PAF overlay.

Chemokine-overlaid endothelial monolayers with adhering neutrophils were fixed with 3% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.2). Samples were postfixed as described (8). Ultrathin (70- to 100-nm) and semithin (150- to 200-nm) cross sections were prepared using an Ultracut ultramicrotome (UCT; Leica, Vienna, Austria) and collected on 200 mesh or slot (for serial sections) copper grids. Sections were stained with lead citrate and viewed by transmission electron microscopy (Technai-12; Phillips, Eindhoven, The Netherlands) at 120 kV. Electron micrographs were processed using CCD Megaview II with AnalySIS software (Soft Imaging System, Munster, Germany).

Statistical comparison of means was performed using the paired, two-tailed Student’s t test.

We previously developed an in vitro model to monitor lymphocyte TEM through TNF-α-activated HUVEC under physiological shear flow (8). Lymphocyte migration across this endothelial barrier was shown to require both high levels of apically displayed chemokines and continuously applied shear flow. In the present study, neutrophil migration across differentially TNF-α-activated HUVEC was analyzed. Unlike lymphocytes, the majority of neutrophils arrested on highly activated HUVEC could transmigrate the monolayers within minutes, without exogenous introduction of chemoattractants and independent of applied shear stress (Fig. 1,A). Neither the magnitude nor the rate of neutrophil TEM was significantly enhanced by the presence of shear flow (Fig. 1,B). Neutrophil TEM monitored by real-time phase-contrast videomicroscopy was dependent on both Mac-1 and LFA-1 (data not shown). Because this cytokine-activated HUVEC expressed high levels of the major β2 integrin ligands ICAM-1 and ICAM-2 (Fig. 1 C), high occupancy of both LFA-1 and Mac-1 on neutrophils adhered to this HUVEC substrate appeared to provide optimal TEM-promoting signals, even in the absence of external shear force signals.

FIGURE 1.

Highly activated HUVEC support Mac-1-dependent, shear-independent neutrophil TEM. A, Effect of shear stress application on neutrophil migration through HUVEC prestimulated with 200 U/ml TNF-α for 24 h. The four indicated migratory phenotypes of neutrophils were determined by videomicroscopy as outlined in Materials and Methods. Results are the mean ± SEM of three independent experiments with separate HUVEC preparations. B, Time course of neutrophil TEM. Values represent mean TEM results, averaged from all independent experiments depicted in A. Time zero was set at the end of the accumulation phase (see Materials and Methods). C, Surface expression of ICAMs and E-selectin on differentially stimulated HUVEC monolayers probed by FACS. Results are the means of three experiments.

FIGURE 1.

Highly activated HUVEC support Mac-1-dependent, shear-independent neutrophil TEM. A, Effect of shear stress application on neutrophil migration through HUVEC prestimulated with 200 U/ml TNF-α for 24 h. The four indicated migratory phenotypes of neutrophils were determined by videomicroscopy as outlined in Materials and Methods. Results are the mean ± SEM of three independent experiments with separate HUVEC preparations. B, Time course of neutrophil TEM. Values represent mean TEM results, averaged from all independent experiments depicted in A. Time zero was set at the end of the accumulation phase (see Materials and Methods). C, Surface expression of ICAMs and E-selectin on differentially stimulated HUVEC monolayers probed by FACS. Results are the means of three experiments.

Close modal

Cultured HUVEC not exposed to primary cytokines express nearly 70-fold lower levels of the major endothelial β2 integrin ligand, ICAM-1, compared with highly TNF-α-activated HUVEC (Fig. 1,C). We next asked whether apical presentation of a chemoattractant to tethered neutrophils (Fig. 2,A) could stimulate their adhesion to, and migration across this HUVEC. Notably, despite high-level constitutive expression of the LFA-1 coligand, ICAM-2, by resting cultured HUVEC (Fig. 1,C), neutrophils failed to firmly arrest on these ECs (Fig. 2,B). These nonactivated HUVEC also failed to bind exogenously overlaid prototypic chemoattractants, like fMLP, IL-8 (CXCL8), or Groα (CXCL1), at levels sufficient to induce neutrophil arrest (data not shown). However, when the HUVEC monolayer was briefly overlaid with the prototypic lipid chemoattractant PAF, neutrophil accumulation and arrest on the HUVEC monolayer were markedly enhanced (Fig. 2,B). PAF-stimulated neutrophil adhesion was β2 integrin dependent (Fig. 2,B), but did not involve de novo induction of ICAM-1 or the endothelial selectins, E- and P-selectin (Fig. 1,C and data not shown). Interestingly, although promoting neutrophil chemokinesis (locomotion), PAF could not support any neutrophil diapedesis across this barrier under shear-free conditions (Fig. 2,C, treatment 2; supplemental video 1).4 Yet, in the presence of continuous physiological shear stress, robust neutrophil migration through PAF-presenting HUVEC took place within minutes (Fig. 2,C; supplemental video 2). Neutrophil pretreatment with the PAF-R antagonist WEB-2086 (17) abrogated both PAF-triggered adhesion and subsequent TEM (Fig. 2,C, treatment 4). Although HUVEC also express PAF-R, their pre-exposure to the PAF-R blocker did not interfere with neutrophil TEM (Fig. 2,C, treatment 6). Thus, PAF-induced neutrophil TEM is exclusively mediated by the PAF-R on the neutrophils rather than on the endothelial barrier. Extensive perfusion of PAF-bound endothelium did not promote any chemotactic neutrophil TEM once neutrophils were settled on the endothelium under shear-free conditions (not shown). This ruled out the possibility that shear flow promoted TEM by redistributing apically overlaid PAF to abluminal HUVEC compartments, thereby forming a promigratory PAF gradient. Interestingly, neutrophil pretreatment with soluble PAF (Fig. 2,A), although enhancing β2-mediated neutrophil arrest to this HUVEC (C, right panel), did not promote any neutrophil TEM, even at saturating doses (C, left panel), although robustly activating neutrophil β2 integrin conformations associated with high affinity to ligands (data not shown). Thus, global stimulation of integrin adhesiveness by soluble PAF as well as by other soluble neutrophil chemoattractants, including fMLP, IL-8, or Groα (data not shown), was insufficient to promote TEM. Taken together, these findings demonstrate that neutrophil TEM can be promoted by shear flow signals only when coupled to in situ PAF signals at the apical endothelial interface engaged by the migrating leukocyte (Fig. 2 A).

FIGURE 2.

Apical endothelial-bound PAF promotes neutrophil TEM through nonactivated HUVEC under shear flow. A, A scheme depicting the two modes of neutrophil activation by PAF tested in this section. Neutrophils are either pretreated with a saturating dose of PAF and washed before accumulating on the endothelial monolayer (prior global activation) or perfused over the monolayer immediately after PAF was overlaid (in situ apical GPCR activation). B, Neutrophil arrest and development of firm shear-resistant adhesion (firm arrest) on nonactivated HUVEC alone or in the presence of endothelial-bound PAF. To assay initial strength of contacts, neutrophils were perfused for 40 s at 0.75 dyn/cm2 over HUVEC monolayers and subjected to a shear stress of 5 dyn/cm2 for 10 s. Where indicated, neutrophils were pretreated with specific integrin-blocking mAbs. Data represent mean ± range of four fields of view taken from two independent experiments. C, Effect of endothelial-overlaid PAF and shear stress on neutrophil TEM through nonactivated HUVEC monolayers. Neutrophils accumulated over the indicated monolayers (intact or previously overlaid with PAF) were subjected to shear stress of 5 dyn/cm2 or left under shear-free conditions for 10 min. Neutrophils treated with the PAF-R antagonist (N), WEB-2086, temporarily arrested but failed to exhibit TEM. Pretreatment of the PAF-overlaid HUVEC (EC) with the antagonist had no effect on neutrophil accumulation or TEM. For soluble PAF stimulation, neutrophils were exposed to PAF (10−7 M; 1 min), washed, and immediately perfused over the monolayer. The different treatments are numbered from 1 to 7. The four indicated migratory phenotypes of neutrophils were determined as in Fig. 1. Right panel, Effect of HUVEC-bound PAF vs soluble PAF on temporary and firm neutrophil arrest. Note that a similar number of neutrophils arrested on the HUVEC irrespective of their mode of PAF exposure, but TEM was triggered only by endothelial-bound PAF. Results are the mean ± SEM of 5–10 independent experiments, each using different neutrophil donors and EC preparations. Sol., Soluble.

FIGURE 2.

Apical endothelial-bound PAF promotes neutrophil TEM through nonactivated HUVEC under shear flow. A, A scheme depicting the two modes of neutrophil activation by PAF tested in this section. Neutrophils are either pretreated with a saturating dose of PAF and washed before accumulating on the endothelial monolayer (prior global activation) or perfused over the monolayer immediately after PAF was overlaid (in situ apical GPCR activation). B, Neutrophil arrest and development of firm shear-resistant adhesion (firm arrest) on nonactivated HUVEC alone or in the presence of endothelial-bound PAF. To assay initial strength of contacts, neutrophils were perfused for 40 s at 0.75 dyn/cm2 over HUVEC monolayers and subjected to a shear stress of 5 dyn/cm2 for 10 s. Where indicated, neutrophils were pretreated with specific integrin-blocking mAbs. Data represent mean ± range of four fields of view taken from two independent experiments. C, Effect of endothelial-overlaid PAF and shear stress on neutrophil TEM through nonactivated HUVEC monolayers. Neutrophils accumulated over the indicated monolayers (intact or previously overlaid with PAF) were subjected to shear stress of 5 dyn/cm2 or left under shear-free conditions for 10 min. Neutrophils treated with the PAF-R antagonist (N), WEB-2086, temporarily arrested but failed to exhibit TEM. Pretreatment of the PAF-overlaid HUVEC (EC) with the antagonist had no effect on neutrophil accumulation or TEM. For soluble PAF stimulation, neutrophils were exposed to PAF (10−7 M; 1 min), washed, and immediately perfused over the monolayer. The different treatments are numbered from 1 to 7. The four indicated migratory phenotypes of neutrophils were determined as in Fig. 1. Right panel, Effect of HUVEC-bound PAF vs soluble PAF on temporary and firm neutrophil arrest. Note that a similar number of neutrophils arrested on the HUVEC irrespective of their mode of PAF exposure, but TEM was triggered only by endothelial-bound PAF. Results are the mean ± SEM of 5–10 independent experiments, each using different neutrophil donors and EC preparations. Sol., Soluble.

Close modal

To test whether apically displayed chemoattractants and shear stress signals promote neutrophil TEM only across nonactivated ECs, we next followed neutrophil TEM through HUVEC prestimulated for 4 h with low-level TNF-α, herein termed “moderately activated HUVEC.” This cytokine stimulation was sufficient to induce both E-selectin and ICAM-1 expression to levels sufficient to supported instantaneous neutrophil capture and arrest under shear flow (compare Figs. 2,A and 3,A, right panel). Firm neutrophil adhesion could be still significantly increased by endothelial-overlaid PAF (Fig. 3,A, right panel), although this short endothelial pretreatment with the chemoattractant did not alter either E-selectin or ICAM-1 expression (Fig. 1,C). The moderately TNF-α-stimulated ECs supported significant levels of neutrophil TEM, which remained unaffected by the application of shear on adherent neutrophils (Fig. 3, A, treatments 1 and 3, and B). Notably, this shear-independent TEM was neither antagonized nor augmented by PAF overlay on the endothelial surface (Fig. 3,A, treatment 2). However, when neutrophils adhered to PAF-presenting HUVEC were exposed to continuous shear flow, robust TEM was augmented over that observed in the absence of shear (Fig. 3, A, treatment 4, and B). Reminiscent of our results on resting HUVEC, neutrophil pre-exposure to soluble PAF, although stimulating locomotion over the ECs, did not stimulate TEM (Fig. 3 A, treatment 5). These results collectively suggest that neutrophil adhesion on endothelial interfaces prestimulated by moderate TNF-α signals is sufficient to prime shear-independent neutrophil TEM. Apical PAF signals can further augment this TEM, only if coupled to shear flow signals.

FIGURE 3.

Apical endothelial-bound PAF enhances neutrophil TEM through moderately TNF-α-activated HUVEC in the presence of shear flow. A, Effect of endothelial-overlaid PAF and shear stress application on neutrophil TEM through HUVEC prestimulated with 2 U/ml TNF-α for 4 h. Neutrophil arrest (right panel) and migration phenotypes were determined as in Fig. 2. Control neutrophils were prestimulated with soluble PAF, allowed to accumulate on the HUVEC monolayer, and subjected to continuous shear flow as in Fig. 1. The different treatments are numbered from 1 to 5. Results represent the mean ± SEM of 5–10 independent experiments. ∗, p < 0.002 compared with neutrophil TEM in the absence of shear flow. Sol., Soluble. B, Time course of neutrophil TEM as depicted in A across PAF-free and PAF-overlaid TNF-α-stimulated HUVEC under shear flow and in shear-free conditions. Percentage of accumulated leukocytes successfully transmigrating under the indicated conditions was determined as in previous figures. C, Effect of Mac-1 or LFA-1 blocking on adhesive and migratory properties of neutrophils accumulated on the PAF-presenting, TNF-α-stimulated HUVEC. The integrin-blocking mAbs were constantly present in the perfusion medium to ensure blockage of Mac-1 newly mobilized to the neutrophil surface. Results are the mean ± SEM of four experiments. ∗, p < 0.02, compared with TEM of neutrophils treated with control mAb (anti-P-selectin glycoprotein ligand-1).

FIGURE 3.

Apical endothelial-bound PAF enhances neutrophil TEM through moderately TNF-α-activated HUVEC in the presence of shear flow. A, Effect of endothelial-overlaid PAF and shear stress application on neutrophil TEM through HUVEC prestimulated with 2 U/ml TNF-α for 4 h. Neutrophil arrest (right panel) and migration phenotypes were determined as in Fig. 2. Control neutrophils were prestimulated with soluble PAF, allowed to accumulate on the HUVEC monolayer, and subjected to continuous shear flow as in Fig. 1. The different treatments are numbered from 1 to 5. Results represent the mean ± SEM of 5–10 independent experiments. ∗, p < 0.002 compared with neutrophil TEM in the absence of shear flow. Sol., Soluble. B, Time course of neutrophil TEM as depicted in A across PAF-free and PAF-overlaid TNF-α-stimulated HUVEC under shear flow and in shear-free conditions. Percentage of accumulated leukocytes successfully transmigrating under the indicated conditions was determined as in previous figures. C, Effect of Mac-1 or LFA-1 blocking on adhesive and migratory properties of neutrophils accumulated on the PAF-presenting, TNF-α-stimulated HUVEC. The integrin-blocking mAbs were constantly present in the perfusion medium to ensure blockage of Mac-1 newly mobilized to the neutrophil surface. Results are the mean ± SEM of four experiments. ∗, p < 0.02, compared with TEM of neutrophils treated with control mAb (anti-P-selectin glycoprotein ligand-1).

Close modal

We next tested how specific blockage of either LFA-1 or Mac-1 affects neutrophil adhesion, locomotion, and subsequent TEM through the PAF-presenting moderately activated HUVEC. Notably, the specific promigratory roles of these integrins could not be tested on the PAF-overlaid nonactivated endothelium, because, in that endothelial model, each integrin played a key role already in the earliest adhesion-strengthening events of neutrophils (Fig. 2,A). In contrast, arrest and adhesion-strengthening events of neutrophils accumulated on PAF-presenting, moderately activated HUVEC were not inhibited by blocking either LFA-1 or Mac-1 (Fig. 3,C, detachment category). This allowed us to assess the promigratory properties of each neutrophil integrin independent of its role in earlier accumulation and arrest on the endothelial surface. Reminiscent of neutrophil TEM through highly activated HUVEC (Fig. 1), Mac-1 blocking attenuated neutrophil TEM through PAF-presenting, moderately activated HUVEC under shear flow to a larger extent than blocking LFA-1 (Fig. 3 C). Thus, neutrophil TEM through PAF-presenting, moderately activated HUVEC is triggered by shear flow signals, and is dependent on the Mac-1 more than on the LFA-1 integrin.

Having demonstrated that neutrophils can incorporate TEM-promoting signals from the lipid chemoattractant PAF after arresting on β2 integrin ligands expressed by differently activated HUVEC, we next asked whether other prototypic neutrophil chemokines like IL-8 and Groα, found on inflamed endothelia (18), may also promote neutrophil TEM under shear flow. Although neither IL-8 nor Groα adsorbed on nonactivated HUVEC at appreciable levels sufficient to stimulate neutrophil adhesion, moderately TNF-α-stimulated HUVEC readily presented these chemokines in functional proadhesive states. Reminiscent of PAF, Groα, a ligand for the neutrophil GPCR, CXCR2, triggered firm neutrophil arrest, locomotion, and subsequent TEM in the presence of shear stress (Fig. 4). Thus, the ability to induce neutrophil TEM under shear flow is shared among lipid and protein chemoattractants. Similar to Groα, IL-8, which signals via both neutrophil CXCR1 and CXCR2 (7), triggered firm neutrophil arrest and subsequent locomotion over the endothelial surface (Fig. 4). Nevertheless, IL-8 failed to trigger any TEM, regardless of the shear flow conditions applied or the chemokine dose tested (Fig. 4 and data not shown). IL-8 also attenuated the spontaneous neutrophil TEM triggered in the absence of exogenous chemoattractants (Fig. 4). Furthermore, when either GROα or PAF were first immobilized on the apical endothelial surface at doses supporting optimal TEM, the subsequent addition of IL-8 still totally inhibited their promigratory effect (data not shown). IL-8 did not interfere with Mac-1 translocation or activation, because surface up-regulation of Mac-1 and induction of the activation neoepitope CBRM1/5 (19) were similarly induced by PAF, Groα, and IL-8 (data not shown). These results are consistent with known inhibitory roles of IL-8 on neutrophil migration through endothelial surfaces (20). This is also a first in vitro confirmation of a report by Ley et al. (21) in which i.v. introduced IL-8 attenuated granulocyte TEM through mesenteric venules without inhibiting leukocyte rolling and arrest.

FIGURE 4.

Apical Groα, but not IL-8, promotes neutrophil TEM under shear flow. Effects of endothelial overlaid Groα or IL-8 on firm arrest and migratory phenotypes of neutrophils accumulated on HUVEC prestimulated with TNF-α (2 U/ml; 4 h). ∗, p < 0.001, compared with neutrophil TEM in the absence of exogenous chemoattractants. Right panel, Effect of HUVEC-bound PAF, Groα, or IL-8 on firm neutrophil arrest. All experiments were conducted under shear flow as in Fig. 3. Results are the mean ± SEM of five to eight independent experiments.

FIGURE 4.

Apical Groα, but not IL-8, promotes neutrophil TEM under shear flow. Effects of endothelial overlaid Groα or IL-8 on firm arrest and migratory phenotypes of neutrophils accumulated on HUVEC prestimulated with TNF-α (2 U/ml; 4 h). ∗, p < 0.001, compared with neutrophil TEM in the absence of exogenous chemoattractants. Right panel, Effect of HUVEC-bound PAF, Groα, or IL-8 on firm neutrophil arrest. All experiments were conducted under shear flow as in Fig. 3. Results are the mean ± SEM of five to eight independent experiments.

Close modal

A major fraction of neutrophils arrested on PAF overlaid on nonactivated or moderately activated HUVEC spread and migrated across it within less than a cell diameter from their original arrest site (supplemental videos 2 and 3). Analysis of neutrophil locomotion over the EC surface at a time frame preceding the first detectable TEM events (Fig. 2,C) unexpectedly revealed that applied shear forces restricted the motility of adherent neutrophils over the PAF-presenting nonactivated HUVEC by up to 3-fold relative to shear-free conditions (Fig. 5,A). These observations collectively argued against a role for shear in transporting neutrophils to sites of diapedesis and suggested a novel mechanoregulatory role for shear flow in modulating β2 integrin-dependent leukocyte associations with apical endothelial surfaces. To gain ultrastructural insights into this phenomenon, we next used electron microscopic analysis of neutrophils arrested on PAF-presenting nonactivated HUVEC and immediately exposed to either shear-free or shear flow conditions. Under shear-free conditions, the vast majority (>80%) of EC-adherent neutrophils remained tethered to the endothelial surface through small microvillar contacts (Fig. 5,BI, arrowheads) or flattened on the endothelial membrane (Fig. 5,BI, asterisks). In contrast, in the presence of shear flow a major fraction of EC-adherent neutrophils extended invaginations into apical endothelial compartments remote from endothelial junctions (Fig. 5, B, II and III; and C, intracellular protrusions). A fraction of these neutrophils simultaneously sent out invaginations through both apical endothelial surfaces and endothelial junctions (Fig. 5,BIII). In addition, neutrophils crossing through endothelial junctions extended small invaginations into these junctions (Fig. 5 BIV). These results suggest that shear-transduced signals rapidly promote neutrophil invaginations into both junctional and nonjunctional PAF-presenting endothelial interfaces during initial stages of TEM.

FIGURE 5.

Shear stress triggers neutrophil invaginations into apical regions of PAF-presenting nonactivated ECs. A, Effect of shear flow on neutrophil locomotion over PAF-presenting nonactivated HUVEC. Neutrophils accumulated on the monolayer and subjected to shear-free conditions or to shear flow for a 90-s period, a time frame preceding TEM initiation (Fig. 1C). Lateral movements of all adherent neutrophils were determined by image analysis tracking (42 ). Data shown are the mean locomotion speed ± SEM of five independent experiments, with 15–30 neutrophils in each experiment. ∗, p < 0.02, compared with the velocity under shear-free conditions. B, Ultrastructural analysis of neutrophils adherent on PAF overlaid HUVEC left under shear-free conditions (I), or subjected to shear flow as in A (II–IV). Monolayers were fixed and processed for transmission electron microscopy. Shown are ultrathin cross sections of representative neutrophils spread on the apical side of the endothelium without detectable protrusions (I), or with multiple apical invaginations, i.e., intracellular protrusions sent into the apical endothelial membrane (II and III). A cross section of a neutrophil sending out a protrusion through an interendothelial junction (junctional protrusion) is shown in IV. In II and III, the neutrophil-endothelial interfaces marked in rectangles are magnified. Note, in III, that the neutrophil invaginating the nonjunctional endothelial compartment also sends outs simultaneous projections underneath the interendothelial junction. Bar, 1 μm. ∗, Closely opposed neutrophil endothelial membranes; arrowheads, neutrophil dot contacts with the endothelial membrane. E1 and E2, ECs 1 and 2. Arrows, Interendothelial junctions. C, The various types of protruding (intracellular or junctional) or non protruding neutrophils were expressed as fractions of the total adherent neutrophils on the PAF-presenting HUVEC monolayers analyzed in B. In this experiment, 12 and 24 neutrophils were ultrastructurally analyzed under shear-free and shear flow conditions, respectively.

FIGURE 5.

Shear stress triggers neutrophil invaginations into apical regions of PAF-presenting nonactivated ECs. A, Effect of shear flow on neutrophil locomotion over PAF-presenting nonactivated HUVEC. Neutrophils accumulated on the monolayer and subjected to shear-free conditions or to shear flow for a 90-s period, a time frame preceding TEM initiation (Fig. 1C). Lateral movements of all adherent neutrophils were determined by image analysis tracking (42 ). Data shown are the mean locomotion speed ± SEM of five independent experiments, with 15–30 neutrophils in each experiment. ∗, p < 0.02, compared with the velocity under shear-free conditions. B, Ultrastructural analysis of neutrophils adherent on PAF overlaid HUVEC left under shear-free conditions (I), or subjected to shear flow as in A (II–IV). Monolayers were fixed and processed for transmission electron microscopy. Shown are ultrathin cross sections of representative neutrophils spread on the apical side of the endothelium without detectable protrusions (I), or with multiple apical invaginations, i.e., intracellular protrusions sent into the apical endothelial membrane (II and III). A cross section of a neutrophil sending out a protrusion through an interendothelial junction (junctional protrusion) is shown in IV. In II and III, the neutrophil-endothelial interfaces marked in rectangles are magnified. Note, in III, that the neutrophil invaginating the nonjunctional endothelial compartment also sends outs simultaneous projections underneath the interendothelial junction. Bar, 1 μm. ∗, Closely opposed neutrophil endothelial membranes; arrowheads, neutrophil dot contacts with the endothelial membrane. E1 and E2, ECs 1 and 2. Arrows, Interendothelial junctions. C, The various types of protruding (intracellular or junctional) or non protruding neutrophils were expressed as fractions of the total adherent neutrophils on the PAF-presenting HUVEC monolayers analyzed in B. In this experiment, 12 and 24 neutrophils were ultrastructurally analyzed under shear-free and shear flow conditions, respectively.

Close modal

We next tested, by ultrastructural analysis, whether shear- and GPCR-promoted neutrophil TEM across moderately activated HUVEC also translate into apical endothelial invaginations. Neutrophils were allowed to migrate for 3 min across PAF-presenting, moderately activated HUVEC, and at this time point, 50% of the total transmigratory neutrophil pool was either in the middle of TEM or had recently completed diapedesis (Fig. 3,B). Thus, concomitant neutrophil associations with both the apical and sublumenal endothelial compartments could be analyzed. In the absence of shear flow, neutrophils either remained tethered to the apical surface (Fig. 6,I) or transmigrated the PAF-presenting HUVEC (Fig. 6, II–IV), but no invaginations into the apical endothelial surfaces were observed. In sharp contrast, neutrophils transmigrating identical HUVEC monolayers under shear flow conditions sent out multiple invaginations into the apical endothelial surfaces, both at junctional and nonjunctional sites (Figs. 7,A and S1). However, non-transmigrating neutrophils subjected to shear flow conditions lacked apical invaginations (data not shown). Strikingly, a small subset of these invaginations (approximated at 5%) resulted in transcellular migration (Figs. 7,A and S1). Serial sections of transcellularly migrating neutrophils indicated extensive endothelial remodeling in the vicinity of the endothelial nuclei (Figs. 7,A and S1). A fraction of these transmigrating neutrophils could be visualized by phase contrast videomicroscopy in real time, crossing at sites remote from endothelial junctions (see supplemental video 3). Notably, neutrophil transcellular migration also occurred across non-confluent HUVEC, where adherent neutrophils were detected crossing through single ECs under shear flow within a 1-min period (supplemental video 4). In contrast, neutrophils adherent to IL-8-presenting, moderately activated HUVEC, were unable toinvaginate the apical endothelial surface (Fig. 7,B), which closely agreed with their inability to migrate across this monolayer (Fig. 4). Taken together, these data indicate shear signals to promote neutrophil transmigration across nonactivated and moderately activated HUVEC by inducing extensive leukocyte invaginations into apical chemoattractant-presenting endothelial surfaces.

FIGURE 6.

Neutrophil TEM through PAF-presenting TNF-α-activated HUVEC under shear-free conditions occurs without apical invaginations. Ultrathin cross sections of neutrophils migrating across PAF-presenting HUVEC prestimulated with 2 U/ml TNF-α for 4 h in the absence of shear flow. Cell samples were fixed 3 min after neutrophil adherence on the HUVEC monolayer and processed for transmission electron microscopy. N1, N2, and E1, E2, depict individual neutrophils and ECs, respectively. Arrow, Interendothelial junctions. Depicted are 4 of 34 analyzed neutrophils. Bar, 1 μm.

FIGURE 6.

Neutrophil TEM through PAF-presenting TNF-α-activated HUVEC under shear-free conditions occurs without apical invaginations. Ultrathin cross sections of neutrophils migrating across PAF-presenting HUVEC prestimulated with 2 U/ml TNF-α for 4 h in the absence of shear flow. Cell samples were fixed 3 min after neutrophil adherence on the HUVEC monolayer and processed for transmission electron microscopy. N1, N2, and E1, E2, depict individual neutrophils and ECs, respectively. Arrow, Interendothelial junctions. Depicted are 4 of 34 analyzed neutrophils. Bar, 1 μm.

Close modal
FIGURE 7.

Shear stress induces massive apical invaginations and transcellular neutrophil TEM through PAF-presenting, TNF-α-activated HUVEC. A, Ultrastructural analysis of neutrophils migrating under shear flow across PAF-presenting HUVEC prestimulated with 2 U/ml TNF-α for 4 h. Shown are serial cross sections (5, 8, 12, 15, and 18 from a total of 19) of two neutrophils adhering to and migrating across neighboring ECs. Samples were fixed 3 min after neutrophil adherence as in Fig. 6. B, A representative neutrophil left under shear on IL-8-presenting HUVEC prestimulated with 2 U/ml TNF-α, identically as in A. Symbols and bars are identical with those in Fig. 6. At least 25 neutrophils were analyzed in each experimental condition. Bar, 1 μm.

FIGURE 7.

Shear stress induces massive apical invaginations and transcellular neutrophil TEM through PAF-presenting, TNF-α-activated HUVEC. A, Ultrastructural analysis of neutrophils migrating under shear flow across PAF-presenting HUVEC prestimulated with 2 U/ml TNF-α for 4 h. Shown are serial cross sections (5, 8, 12, 15, and 18 from a total of 19) of two neutrophils adhering to and migrating across neighboring ECs. Samples were fixed 3 min after neutrophil adherence as in Fig. 6. B, A representative neutrophil left under shear on IL-8-presenting HUVEC prestimulated with 2 U/ml TNF-α, identically as in A. Symbols and bars are identical with those in Fig. 6. At least 25 neutrophils were analyzed in each experimental condition. Bar, 1 μm.

Close modal

Neutrophil adhesion and TEM through variably activated ECs are orchestrated by the two major β2 integrins, LFA-1 and Mac-1 (22). Our results in this in vitro model reveal the complex interplay between these β2 integrins, shear stress mechanosignals, and GPCR-stimulatory signals transduced by prototypic chemoattractants expressed by injured or inflamed endothelial barriers. Neutrophils, as other leukocytes, must integrate both adhesive and chemotactic signals at apical endothelial interfaces to initiate productive TEM (23). The present study demonstrates that neutrophils, unlike resting lymphocytes (8), translate shear stress signals into productive TEM depending on the state of endothelial activation, the level of β2 integrin ligands, and the type of activating chemoattractants displayed on the apical aspect of the endothelial barrier. Neutrophil migration across highly activated HUVEC, which expresses high levels of the major β2 integrin ligands, ICAM-1 and ICAM-2, as well as potential integrin-activating receptors such as E-selectin (24), can take place independent of shear signals. In contrast, neutrophil migration across nonactivated HUVEC expressing low-level β2 integrin ligands cannot take place unless the neutrophils encounter proper GPCR activation and shear signals (Fig. 8). Because short pretreatment of neutrophils with soluble rather than endothelial-presented promigratory chemoattractants failed to promote any TEM (Figs. 2 and 3), we conclude that, to promote TEM, in situ-activated leukocyte GPCRs and shear stress signals must both operate on their integrin targets in a polarized fashion at the leukocyte endothelial interface (Fig. 2 A).

FIGURE 8.

A suggested model for the relationship between shear and GPCR signals, β2 integrin ligands, and successful diapedesis across different endothelial barriers. Neutrophil TEM through weakly stimulated HUVEC, expressing low-level β2 integrin ligands and presenting high levels of promigratory chemoattractants (PAF, Groα, but not IL-8) is highly shear dependent and correlates with restricted locomotion on β2 ligands and enhanced neutrophil invaginations into apical endothelial surfaces. Since β2 ligand activity is elevated on the highly TNF-α-activated HUVEC (right arrowhead), neutrophil TEM is primed independently of shear stress signals. In contrast, T cell TEM through highly TNF-α-activated HUVEC requires both high-level integrin ligands, stimulation by apical endothelial chemokines, and continuous shear stress signals (8 ).

FIGURE 8.

A suggested model for the relationship between shear and GPCR signals, β2 integrin ligands, and successful diapedesis across different endothelial barriers. Neutrophil TEM through weakly stimulated HUVEC, expressing low-level β2 integrin ligands and presenting high levels of promigratory chemoattractants (PAF, Groα, but not IL-8) is highly shear dependent and correlates with restricted locomotion on β2 ligands and enhanced neutrophil invaginations into apical endothelial surfaces. Since β2 ligand activity is elevated on the highly TNF-α-activated HUVEC (right arrowhead), neutrophil TEM is primed independently of shear stress signals. In contrast, T cell TEM through highly TNF-α-activated HUVEC requires both high-level integrin ligands, stimulation by apical endothelial chemokines, and continuous shear stress signals (8 ).

Close modal

The endothelial monolayer is not just an adhesive surface for leukocytes recruited to target vessel wall sites under shear flow, but also an active participant in leukocyte TEM (25). Recent studies highlight the key regulatory roles played by interactions of integrins on lymphocytes and monocytes with respective endothelial ligands for correct activation of the endothelial machineries critical for TEM (26, 27). Neutrophils migrating across ECs trigger local transient elevations of endothelial Ca2+, and retraction signals that are thought to enhance endothelial gap formation near the migrating leukocytes (28, 29, 30). Our results suggest that this endothelial remodeling may depend on the ability of neutrophils to extend invaginations into apical endothelial compartments. This capacity may also underlie the initiation of transcellular diapedesis, imaged here for the first time by videomicroscopy. By monitoring neutrophil TEM through both single and confluent ECs, we were able to unambiguously follow full transcellular TEM routes and record them in real time (supplemental videos 3 and 4). This live imaging and the complementary electron microscopic analysis are fully consistent with previous in vivo studies (1, 31). These works indicated that in vivo injection of specific chemoattractants can activate neutrophils to cross blood vessels via transcellular routes, but did not note a specific requirement for these or other chemoattractants to be deposited on vessel walls to exert their promigratory functions. Our real-time tracking suggests that it takes neutrophils less than a minute to cross through EC bodies, a remarkably short period considering the extensive endothelial remodeling required for this unique migratory mechanism (supplemental videos 3 and 4; Figs. 7 and S1). Transcellular migration often takes place near EC junctions (1) and is therefore hard to follow by either light microscopy or electron microscopy. In fact, the paracellular and transcellular routes are not mutually exclusive. Neutrophils migrating across endothelial junctions often extended transcellular protrusions across apical endothelial compartments nearby the junction (Fig. 5 BIII). Thus, the actual fraction of transcellular migration routes approximated at 5% in our in vitro assays could in fact be higher.

Neutrophils arrested on vascular endothelium must embark on a series of cytoskeleton-remodeling events to locomote to and migrate through the endothelial barrier. The major candidate to regulate these processes is the small GTPase Rap1 and its LFA-1-associated effector, RAPL, recently shown to regulate lymphocyte polarization and TEM (32, 33). Interestingly, activated Rap1 can promote lymphocyte TEM only in the presence of shear stress signals (33). Furthermore, the Rap1 GTPase is stored in endosomes that rapidly fuse with and locally extend the plasma membranes of activated neutrophils and lymphocytes (34). Mac-1 is also up-regulated during particular neutrophil TEM processes (35) through fusion of the specific granules with the neutrophil plasma membrane. Granule fusion may thus have multiple roles at sites containing low levels of β2 integrins, but high GPCR-activating signals, where maximal invaginations and shear-triggered TEM were observed (Fig. 8). The granule fusion process may translocate and localize de novo-activated Mac-1 and possibly recycled LFA-1 (36) to extend the neutrophil plasma membrane and thereby provide additional contact area to rapidly forming de novo invaginations. A particularly intriguing possibility is that tertiary neutrophil granules containing both Rap-1 and Mac-1 fuse together at the plasma membrane of neutrophils at endothelial sites presenting chemoattractants such as PAF and Groα. Thus, localized Rap-1 activation, in situ stimulation of Mac-1 and LFA-1, and localized membrane extension by fused neutrophil granules may all trigger the neutrophil invaginations observed by us to correlate with enhanced TEM. Because the Mac-1 integrin is also a key complement receptor involved in various phagocytic processes through activation of the RhoA GTPase (37, 38), RhoA regulated contractility events in neutrophils may translate Mac-1 activation by apical chemoattractants to productive formation of neutrophil invaginations underlying TEM.

Why, then, can neutrophils transmigrate across cytokine-activated HUVEC expressing high-level β2 integrin ligands in vitro even in the absence of shear signals, whereas lymphocytes transmigrate across similar HUVEC only if exposed to both chemokine and shear flow signals at apical endothelial compartments (8)? It appears that very high levels of endothelial ligand binding to both Mac-1 and LFA-1 on neutrophils but not on lymphocytes is sufficient to elicit TEM cues without shear signals. One intriguing possibility that deserves future experimentation is that E-selectin may activate these β2 integrin members in neutrophils much more effectively than it activates VLA-4 and LFA-1 in resting lymphocytes, because these cells express fewer E-selectin ligands than neutrophils (24, 39). The activated β2 integrins may then readily occupy the high levels of β2 integrin ligands, and this may trigger bidirectional outside-in signaling both in the neutrophils and their endothelial counterparts, driving neutrophil TEM without additional shear signals. The failure of lymphocytes to transmigrate highly activated endothelial barriers despite high-level expression of both β1 and β2 ligands may also reflect different outside-in signaling machineries triggered by neutrophil and lymphocyte integrin occupancy (40, 41). Similar to neutrophils, monocytes also show considerable capacity to transmigrate highly activated EC barriers even without shear stress signals (11, 12). Thus, lymphocyte integrins occupied by their respective endothelial ligands may be incapable of triggering their outside-in signaling machineries unless in situ activated at the apical endothelial interface by correct chemokines and shear stress signals. Taken together, it appears that shear usage by migrating neutrophils is more restricted than in migrating lymphocytes. It appears to depend on both the level of apical β2 integrin ligands and the activation state of the β2 integrins and so on the magnitude of β2 integrin occupancy. Further elucidation of how shear signals differentially regulate the β2 integrin-associated elements critical for neutrophil, monocyte, and lymphocyte TEM may potentially introduce new specific targets for controlling myeloid and lymphoid cell emigration processes in various settings of inflammation and injury.

We thank Y. Van Kooyk, T. Springer, and F. W. Luscinskas for gifts of reagents, and Dr. S. Schwarzbaum for editorial assistance. We also thank F. W. Luscinskas for very helpful discussions.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Parts of this study were supported by the Israel Science Foundation founded by the Israel Academy of Sciences and Humanities, the Abisch-Frenkel Foundation, and the Crown Foundation. R.A. is the incumbent of the Tauro Career Development Chair in Biomedical Research.

3

Abbreviations used in this paper: TEM, transendothelial migration; EC, endothelial cell; PAF, platelet-activating factor; Groα, growth-related oncogene α; GPCR, G protein-coupled receptor.

4

The on-line version of this article contains supplemental material.

1
Feng, D., J. A. Nagy, K. Pyne, H. F. Dvorak, A. M. Dvorak.
1998
. Neutrophils emigrate from venules by a transendothelial cell pathway in response to fMLP.
J. Exp. Med.
187
:
903
.
2
Johnson-Leger, C., M. Aurrand-Lions, B. A. Imhof.
2000
. The parting of the endothelium: miracle, or simply a junctional affair?.
J. Cell Sci.
113
:
921
.
3
Ley, K..
2001
. Pathways and bottlenecks in the web of inflammatory adhesion molecules and chemoattractants.
Immunol. Res.
24
:
87
.
4
Worthylake, R. A., K. Burridge.
2001
. Leukocyte transendothelial migration: orchestrating the underlying molecular machinery.
Curr. Opin. Cell Biol.
13
:
569
.
5
Mamdouh, Z., X. Chen, L. M. Pierini, F. R. Maxfield, W. A. Muller.
2003
. Targeted recycling of PECAM from endothelial surface-connected compartments during diapedesis.
Nature
421
:
748
.
6
Burns, A. R., R. A. Bowden, S. D. MacDonell, D. C. Walker, T. O. Odebunmi, E. M. Donnachie, S. I. Simon, M. L. Entman, C. W. Smith, E. S. Brown, et al
2000
. Analysis of tight junctions during neutrophil transendothelial migration: neutrophil transendothelial migration is independent of tight junctions and occurs preferentially at tricellular corners.
J. Cell Sci.
113
:
45
.
7
Luu, N. T., G. E. Rainger, G. B. Nash.
2000
. Differential ability of exogenous chemotactic agents to disrupt transendothelial migration of flowing neutrophils.
J. Immunol.
164
:
5961
.
8
Cinamon, G., V. Shinder, R. Alon.
2001
. Shear forces promote lymphocyte migration across vascular endothelium bearing apical chemokines.
Nat. Immunol.
2
:
515
.
9
Cuvelier, S. L., K. D. Patel.
2001
. Shear-dependent eosinophil transmigration on interleukin 4-stimulated endothelial cells: a role for endothelium-associated eotaxin-3.
J. Exp. Med.
194
:
1699
.
10
Kitayama, J., A. Hidemura, H. Saito, H. Nagawa.
2000
. Shear stress affects migration behavior of polymorphonuclear cells arrested on endothelium.
Cell. Immunol.
203
:
39
.
11
Weber, K. S., P. von Hundelshausen, I. Clark-Lewis, P. C. Weber, C. Weber.
1999
. Differential immobilization and hierarchical involvement of chemokines in monocyte arrest and transmigration on inflamed endothelium in shear flow.
Eur. J. Immunol.
29
:
700
.
12
Schenkel, A. R., Z. Mamdouh, W. A. Muller. Locomotion of monocytes on endothelium is a critical step during extravasation.
Nat. Immunol.
5
:
393
.
13
Dwir, O., F. Shimron, C. Chen, M. Singer, S. D. Rosen, R. Alon.
1998
. GlyCAM-1 supports leukocyte tethering and rolling: evidence for a greater dynamic stability of L-selectin rolling of lymphocytes than of neutrophils.
Cell Adhes. Commun.
6
:
349
.
14
Kluger, M. S., D. R. Johnson, J. S. Pober.
1997
. Mechanism of sustained E-selectin expression in cultured human dermal microvascular endothelial cells.
J. Immunol.
158
:
887
.
15
Bevilacqua, M. P., J. S. Pober, D. L. Mendrick, R. S. Cotran, M. A. Gimbrone, Jr.
1987
. Identification of an inducible endothelial-leukocyte adhesion molecule.
Proc. Natl. Acad. Sci. USA
84
:
9238
.
16
Cinamon, G., R. Alon.
2003
. A real time in vitro assay for studying leukocyte transendothelial migration under physiological flow conditions.
J. Immunol. Methods
273
:
53
.
17
Ostrovsky, L., A. J. King, S. Bond, D. Mitchell, D. E. Lorant, G. A. Zimmerman, R. Larsen, X. F. Niu, P. Kubes.
1998
. A juxtacrine mechanism for neutrophil adhesion on platelets involves platelet-activating factor and a selectin-dependent activation process.
Blood
91
:
3028
.
18
Mantovani, A., F. Bussolino, M. Introna.
1997
. Cytokine regulation of endothelial cell function: from molecular level to the bedside.
Immunol. Today
18
:
231
.
19
Diamond, M. S., T. A. Springer.
1993
. A subpopulation of Mac-1 (CD11b/CD18) molecules mediates neutrophil adhesion to ICAM-1 and fibrinogen.
J. Cell Biol.
120
:
545
.
20
Gimbrone, M. A., Jr, M. S. Obin, A. F. Brock, E. A. Luis, P. E. Hass, C. A. Hebert, Y. K. Yip, D. W. Leung, D. G. Lowe, W. J. Kohr, et al
1989
. Endothelial interleukin-8: a novel inhibitor of leukocyte-endothelial interactions.
Science
246
:
1601
.
21
Ley, K., J. B. Baker, M. I. Cybulsky, M. A. Gimbrone, Jr, F. W. Luscinskas.
1993
. Intravenous interleukin-8 inhibits granulocyte emigration from rabbit mesenteric venules without altering L-selectin expression or leukocyte rolling.
J. Immunol.
151
:
6347
.
22
Henderson, R. B., L. H. Lim, P. A. Tessier, F. N. Gavins, M. Mathies, M. Perretti, N. Hogg.
2001
. The use of lymphocyte function-associated antigen (LFA)-1-deficient mice to determine the role of LFA-1, Mac-1, and α4 integrin in the inflammatory response of neutrophils.
J. Exp. Med.
194
:
219
.
23
Alon, R., F. W. Luscinskas. Crawling and INTEGRating apical cues.
Nat. Immunol.
5
:
351
.
24
Simon, S. I., Y. Hu, D. Vestweber, C. W. Smith.
2000
. Neutrophil tethering on E-selectin activates β2 integrin binding to ICAM-1 through a mitogen-activated protein kinase signal transduction pathway.
J. Immunol.
164
:
4348
.
25
Muller, W. A..
2003
. Leukocyte-endothelial-cell interactions in leukocyte transmigration and the inflammatory response.
Trends Immunol.
24
:
327
.
26
Barreiro, O., M. Yanez-Mo, J. M. Serrador, M. C. Montoya, M. Vicente-Manzanares, R. Tejedor, H. Furthmayr, F. Sanchez-Madrid.
2002
. Dynamic interaction of VCAM-1 and ICAM-1 with moesin and ezrin in a novel endothelial docking structure for adherent leukocytes.
J. Cell Biol.
157
:
1233
.
27
Carman, C. V., C. D. Jun, A. Salas, T. A. Springer.
2003
. Endothelial cells proactively form microvilli-like membrane projections upon intercellular adhesion molecule 1 engagement of leukocyte LFA-1.
J. Immunol.
171
:
6135
.
28
Huang, A. J., J. E. Manning, T. M. Bandak, M. C. Ratau, K. R. Hanser, S. C. Silverstein.
1993
. Endothelial cell cytosolic free calcium regulates neutrophil migration across monolayers of endothelial cells.
J. Cell Biol.
120
:
1371
.
29
Pfau, S., D. Leitenberg, H. Rinder, B. R. Smith, R. Pardi, J. R. Bender.
1995
. Lymphocyte adhesion-dependent calcium signaling in human endothelial cells.
J. Cell Biol.
128
:
969
.
30
Su, W. H., H. Chen, J. Huang, C. J. Jen.
2000
. Endothelial [Ca2+]i signaling during transmigration of polymorphonuclear leukocytes.
Blood
96
:
3816
.
31
Hoshi, O., T. Ushiki.
1999
. Scanning electron microscopic studies on the route of neutrophil extravasation in the mouse after exposure to the chemotactic peptide N-formyl-methionyl-leucyl-phenylalanine (fMLP).
Arch. Histol. Cytol.
62
:
253
.
32
Katagiri, K., A. Maeda, M. Shimonaka, T. Kinashi.
2003
. RAPL, a Rap1-binding molecule that mediates Rap1-induced adhesion through spatial regulation of LFA-1.
Nat. Immunol.
4
:
741
.
33
Shimonaka, M., K. Katagiri, T. Nakayama, N. Fujita, T. Tsuruo, O. Yoshie, T. Kinashi.
2003
. Rap1 translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow.
J. Cell Biol.
161
:
417
.
34
Bivona, T. G., H. H. Wiener, I. M. Ahearn, J. Silletti, V. K. Chiu, M. R. Philips.
2004
. Rap1 up-regulation and activation on plasma membrane regulates T cell adhesion.
J. Cell Biol.
164
:
461
.
35
Kuijpers, T. W., M. Hoogerwerf, D. Roos.
1992
. Neutrophil migration across monolayers of resting or cytokine-activated endothelial cells: role of intracellular calcium changes and fusion of specific granules with the plasma membrane.
J. Immunol.
148
:
72
.
36
Dustin, M. L., T. G. Bivona, M. R. Philips.
2004
. Membranes as messengers in T cell adhesion signaling.
Nat. Immunol.
5
:
363
.
37
Caron, E., A. Hall.
1998
. Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases.
Science
282
:
1717
.
38
Olazabal, I. M., E. Caron, R. C. May, K. Schilling, D. A. Knecht, L. M. Machesky.
2002
. Rho-kinase and myosin-II control phagocytic cup formation during CR, but not FcγR, phagocytosis.
Curr. Biol.
12
:
1413
.
39
Green, C. E., D. N. Pearson, R. T. Camphausen, D. E. Staunton, S. I. Simon.
2004
. Shear-dependent capping of L-selectin and P-selectin glycoprotein ligand 1 by E-selectin signals activation of high-avidity β2 integrin on neutrophils.
J. Immunol.
172
:
7780
.
40
Vines, C. M., J. W. Potter, Y. Xu, R. L. Geahlen, P. S. Costello, V. L. Tybulewicz, C. A. Lowell, P. W. Chang, H. D. Gresham, C. L. Willman.
2001
. Inhibition of β2 integrin receptor and Syk kinase signaling in monocytes by the Src family kinase Fgr.
Immunity
15
:
507
.
41
Perez, O. D., D. Mitchell, G. C. Jager, S. South, C. Murriel, J. McBride, L. A. Herzenberg, S. Kinoshita, G. P. Nolan.
2003
. Leukocyte functional antigen 1 lowers T cell activation thresholds and signaling through cytohesin-1 and Jun-activating binding protein 1.
Nat. Immunol.
4
:
1083
.
42
Franitza, S., R. Hershkoviz, N. Kam, N. Lichtenstein, G. G. Vaday, R. Alon, O. Lider.
2000
. TNF-α associated with extracellular matrix fibronectin provides a stop signal for chemotactically migrating T cells.
J. Immunol.
165
:
2738
.