Fluid Shear-Induced Activation and Cleavage of CD18 During Pseudopod Retraction by Human Neutrophils
Surface membrane expression and conformational activation of CD18 integrins into an open molecular configuration play critical roles in neutrophil ligand binding, membrane attachment, spreading on the endothelium, and cell migration to sites of inflammation. Previously, we observed pseudopod retraction and concomitant cleavage of CD18 by human neutrophils upon exposure to fluid shear stress. But the underlying cellular mechanism(s) linking these phenomena remains unknown. We hypothesize here that activation of CD18 under the influence of fluid shear stress leads to its increased susceptibility to proteolytic cleavage by lysosomal proteases such as cathepsin B and is a requirement for CD18 cleavage and subsequent pseudopod retraction. Specifically, we report conformational changes in the CD18 extracellular domain on neutrophils exposed to physiological fluid shear stresses. Western blot analysis using a CD18 antibody targeted against the intracellular domain revealed reduced levels of full-length CD18 after stimulation of neutrophils with either fluid shear stress or with the Ca2+ ionophore phorbol 12-myristate 13-acetate (PMA; 100 nM) in the presence of exogenous cathepsin B (0.5 U/ml).
Moreover, we identified cathepsin B as one protease that may be released by neutrophils under flow and required for shear-induced pseudopod retraction. These results suggest that a putative mechanotransduction mechanism involving shear-induced changes in the conformation of CD18 and its subsequent cleavage from the cell surface serves to regulate pseudopod activity of neutrophils under physiologic shear stress.
In addition to biochemical mediators, fluid shear stress (the tangential force per unit area; dyn/cm2) resulting from blood flow, regulates neutrophil activity. The role of fluid shear stress to maintain leukocytes in a spherical shape as well as to minimize cell stiffness (Moazzam et al., 1997; Fukuda et al., 2000) and F-actin polymerization (Shive et al., 2000) in order to ensure their passage through the microcirculation is consistent with reported evidence demonstrating increased entrapment of activated leukocytes within the microcirculation due to the effects of inflammatory stimulation on cell deformability (Worthen et al., 1989; Mazzoni and Schmid-Scho¨ nbein, 1996). Furthermore, an impaired pseudopod retraction response to shear, such as by stimulation with inflammatory mediators (Fukuda et al., 2000), has been associated with leukocyte entrapment in the microcirculation of normotensive rats (Fukuda et al., 2004a) as well as increased microvascular resistance in spontaneously hypertensive (Fukuda et al., 2004b) and glucocorticoid-treated rats (Fukuda et al., 2004a). These results indicated a role for shear stress in regulating neutrophil activity in a non-inflamed environment, specifically, a possible control mechanism that, in the absence of agonists, actively prevents leukocyte adhesion and pseudopod formation under physiological conditions.
To date, details regarding the ability of neutrophils to sense and respond to local fluid stresses remain unclear. The fact that neutrophils respond to fluid shear stress at magnitudes far below those required for passive viscoelastic deformation (Sugihara-Seki and Schmid-Scho¨ nbein, 2003) points to transducing elements on the cell surface that convert shear stresses or their gradients into biological cues (i.e., mechanotransduction) and subsequently coordinate diverse cell functions (e.g., membrane adhesion, migration). In line with previously-reported evidence for fluid flow-induced activation of vascular endothelial cell growth factor receptor-2 (VEGFR- 2) (Chen et al., 1999; Jin et al., 2003) or Tie-2 (Lee and Koh, 2003) on endothelial cells, we hypothesized that surface proteins/receptors may also serve as shear stress sensors on the neutrophil surface. Recent results demonstrate a shift in the activity of G-protein-coupled receptors (GPCRs) (Makino et al., 2006) and small rho-family GTPases (Makino et al., 2005) in human leukocytes exposed to fluid flow suggesting a role of membrane receptors as shear stress sensors. These studies also serve to elucidate a mechanism for the rapid actin depolymerization associated with pseudopod retraction by neutrophils during shear.
In addition to actin depolymerization, pseudopod retraction by adhered neutrophils in response to shear stress requires an integrin-mediated detachment mechanism that promotes de- binding of membrane-substrate attachments. CD18 integrins are one class of adhesion molecules expressed on the cell surface that when conformationally activated (e.g., by chemokine stimulation) play a critical role in the adhesion of neutrophils to substrate surfaces during recruitment to sites of acute inflammation (Simon and Green, 2005). Recent evidence implicates CD18 in the responses of neutrophils to fluid shear stress. For instance, neutrophils attached to and migrating on glass slides via CD18, but not b1, integrins exhibited pseudopod retraction (Marschel and Schmid-Scho¨ nbein, 2002), a characteristic response of these cells to fluid shear stress (Moazzam et al., 1997; Fukuda et al., 2000). This evidence is in line with the established role of CD18 in mediating cytokine- stimulated neutrophil recruitment by serving as an anchoring point for either cell aggregation, capture and/or migration and thus providing an adherent cell with the capability to withstand the forces imposed by fluid flow (Simon and Goldsmith, 2002).
Exposure of naive neutrophils to shear stress regimes, similar to those reported to induce pseudopod retraction, also elicit redistribution and cleavage of CD18 on the cell surface through a mechanism that involves extracellular enzymatic activity characteristic of cysteine proteases including cathepsin B (Fukuda and Schmid-Scho¨ nbein, 2003), a lysosomal protease expressed by a wide variety of cells (Mort and Buttle, 1997). The cellular mechanism(s) linking fluid shear stress, CD18 cleavage, and pseudopod retraction, however, remains to be elucidated. The present study was designed to investigate this link and shed light on a cellular mechanism that may be associated with fluid flow-induced integrin detachment involving both changes in the conformation and subsequent cleavage of CD18.
Materials and Methods
Cells
To investigate shear-induced CD18 activation and cleavage by migrating cells, neutrophils in buffy coat fractions of whole blood were collected from asymptomatic human volunteers by 1 × G red cell sedimentation (Moazzam et al., 1997) and subsequently diluted 1:20 (v/v) in Plasma-Lyte buffer (Baxter, Deerfield, IL) supplemented with 2.5 mM CaCl2 (Sigma–Aldrich, St. Louis, MO). Purified populations of neutrophils in suspension were also used to biochemically examine shear-induced CD18 cleavage. For these studies, human neutrophils were harvested from asymptomatic volunteers by venipuncture, isolated by two-step Histopaque- Percoll gradient centrifugation (Rainger et al., 1999) and resuspended in Plasma-Lyte buffer containing 2.5 mM CaCl2 (reagents purchased from Sigma–Aldrich).
For examination of shear-induced activation of CD18 on individual cells using fluorescence resonance energy transfer (FRET), K562 cells stably transfected with a FRET reporter of leukocyte function-associated antigen-1 (i.e., LFA-1 or CD11a/ CD18) activation (kindly provided by Dr. T. Springer, Harvard University) were cultured in RPMI-1640 media (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Mediatech, Herndon, VA) and 1 mg/ml G412 (InvivoGen, San Diego, CA) (Kim et al., 2003). For experiments, suspensions of K562 transfectants in culture medium were diluted 1:20 (v/v) in Plasma-Lyte supplemented with 2.5 mM CaCl2.
Anti-protease treatment of migrating cells
To examine shear-induced CD18 cleavage and pseudopod retraction, neutrophils from the buffy coat fractions of whole blood were allowed to adhere to and migrate on plasma proteins adsorbed on borosilicate glass slides (Fisher Scientific, Pittsburgh, PA) for 10 min, incubated with 50 mM CA074Me (a cell-permeable protease inhibitor selective for lysosomal cathepsins B and L; EMD Biosciences, San Diego, CA) for 10 min (Fukuda and Schmid-Scho¨ nbein, 2003) and subsequently washed with excess Plasma-Lyte buffer containing 2.5 mM CaCl2 prior to shear stress application. Untreated cells were preincubated in fresh Plasma-Lyte containing 2.5 mM CaCl2. In addition, for some experiments, CD18 cleavage was assessed for neutrophils exposed to fluid shear stress while in the presence of 5 mM CA074 (cell-impermeable cathepsin B inhibitor; EMD Biosciences) for 5 min.Under all protease treatments the neutrophils exhibited approximately 90% viability as assessed using trypan blue exclusion.
Analysis of conformational changes in CD18 integrins
Neutrophils migrating on borosilicate glass coverslips were exposed to 0 (control), 1.8, or 5.3 dyn/cm2 fluid shear stress (in Plasma-Lyte buffer supplemented with 2.5 mM CaCl2) at room temperature for 1.5 min using a parallel-plate flow chamber (Makino et al., 2005). Positive-control experiments for CD18 activation were also conducted by stimulating migrating neutrophils with 10—8 M FMLP (formyl peptide; a known neutrophil activator; Sigma–Aldrich) in the absence of shear. After experiments, the cells were fixed with 1% para-formaldehyde (ICN Biomedicals, Irvine, CA) in 0.1 M phosphate buffer (pH 7.0; Fisher Scientific), labeled with 1 mg/ml Alexa 488-conjugated 327c monoclonal antibodies that only bind activated CD18 in the I-like domain near the amino terminus as previously described (Beals et al., 2001; Lum et al., 2002), and viewed with an epi-fluorescence microscope (Olympus, Center Valley, PA) (Fukuda and Schmid- Scho¨ nbein, 2003). To quantify 327c binding (as a measure of the amount of activated CD18 on the cell surface), the total fluorescence associated with each individual cell was determined digitally (box tool function, Image J software; National Institutes of Health).
For single-cell analyses of CD18 conformational activity, we conducted shear stress experiments using individual K562 myeloleukemic cells stably transfected with a FRET reporter for LFA-1 activation, consisting of CD11a (integrin aL) tagged with cyan fluorescent protein (CFP) and CD18 fused to yellow fluorescent protein (YFP). These transfectants were exposed to ~2 dyn/cm2 shear stress for 1 min using a micropipette shear setup (Moazzam et al., 1997). FRET activity of LFA-1 molecules along the cell periphery was monitored using an Olympus fluorescence microscope. CFP and YFP fluorescence intensity measurements (ICFP and IYFP, respectively) were determined on individual cells at 30-sec intervals for up to 1 min and used to calculate FRET activity as the ratio between YFP and CFP fluorescence (IYFP/ICFP). This FRET ratio was normalized by the FRET ratio of the same cell immediately (within 5 sec) after onset of shear stress. This normalization allows comparison of the LFA-1 activity between cells with or without exposure to shear stress by serving to account for differences in the baseline/background FRET fluorescence intensities between individual cells.
Analysis of CD18 cleavage
To assess the involvement of cathepsin B in shear-induced CD18 cleavage, migrating neutrophils were exposed to shear stress in a parallel plate flow chamber, fixed, and analyzed using immunofluorescence (Fukuda and Schmid-Scho¨ nbein, 2003) with 1:100 (v/v) dilutions of monoclonal antibodies to CD18 (clone 6.7; Pharmingen, La Jolla, CA) that binds to the stalk region located in the ectodomain (Lu et al., 2001). Unlike monoclonal antibody 327c, antibody clone 6.7 exhibits binding affinity for both activated and inactivated CD18 integrins (Lu et al., 2001).
CD18 cleavage was also examined for suspended neutrophil populations. For these experiments, purified populations of human neutrophils in suspension were either maintained under no-flow conditions or exposed to 3 dyn/cm2 fluid shear stress in the absence and presence of 0.5 U/ml exogenous human liver cathepsin B (Sigma–Aldrich) using a cone-plate viscometer and established procedures (Fukuda and Schmid-Scho¨ nbein, 2003). In addition, some neutrophils were stimulated under no-flow conditions with 12-phorbol 12-myristate acetate (PMA; Sigma–Aldrich, 100 nM) with and without cathepsin B (0.5 U/ml for 30 min). After the experiments, some of the cell populations were immediately fixed and labeled with monoclonal antibody against CD18 (clone 6.7) according to standard immunofluorescence procedures. For Western blot analysis of CD18 proteolysis, cellular extracts were harvested, as previously described (Chen et al., 2004). Aliquots of these samples (containing equal amounts of total protein) were analyzed by Western blot analysis with 1 mg/ml goat polyclonal antibody (clone C-20, Santa Cruz, CA) to CD18, which recognize epitopes in the cytosolic domain.
Assay of shear-induced pseudopod retraction
Migrating neutrophils with extended pseudopod(s) were either maintained under no-flow (control) conditions or exposed to ~2 dyn/cm2 shear stress for 2 min using a micropipette shear device, as described previously (Moazzam et al., 1997). A sequence of images of individual neutrophils was recorded with an inverted brightfield microscope (Olympus) during the 2-min control and shear experiments and analyzed using digital imaging software (Image J, National Institutes of Health). As a measure for the degree of pseudopod extension/retraction, the length of the major axis for each cell was measured at 15 sec intervals and normalized by its length at 5 sec (t = —5 sec) prior to the onset of fluid shear.
Neutrophil migration assay
Digital imaging software (Image J) was used to collect time-lapse image sequences of untreated neutrophils as well as cells pretreated with 50 mM CA074Me and subsequently allowed to migrate in the absence or presence of 0.5 U/ml cathepsin B in Plasma-Lyte buffer containing 2.5 mM CaCl2 for 10 min. Migration of these neutrophils was assessed by measuring the displacement of each cell centroid during 30-sec time intervals and by determining its path length.
Statistitical analysis
Comparisons between different fluid shear exposure times were assessed using parametric, paired Student’s t-test. Comparisons between multiple treatments were conducted using Student’s t-test with Bonferroni’s corrected P-value.
Results
Shear stress induces conformational changes in CD18 integrins Fluid shear stress modulates the conformational activity of CD18 on the cell surface. Specifically, neutrophil populations either maintained under control (no-flow) conditions or exposed to 5.3 dyn/cm2 fluid shear stress for 1.5 min, exhibited punctate as well as random clusters of activated CD18 (Fig. 1A). The average surface levels of activated CD18 were significantly increased ( P < 0.05) after exposure of neutrophils to 5.3 dyn/ cm2 (Fig. 1A,B) as well as to 1.8 dyn/cm2 (Fig. 1B). The approximately 1.8-fold increase in CD18 activation by exposure of neutrophils to fluid shear stress was comparable in magnitude to that of cells stimulated with 10—8 M FMLP for 2 min (Fig. 1C). To confirm our observations regarding the influence of shear stress on the CD18 conformation, we examined the FRET activity of LFA-1 reporter constructs stably transfected in K562 cells that had been either maintained under no-flow (control) conditions or exposed to ~2 dyn/cm2 shear stress for up to 60 sec. Under control conditions, K562 stable transfectants exhibited a constitutive FRET activity along the peripheral membrane over a 60-sec time period (Fig. 2A). The constitutive FRET activity, however, was decreased (as denoted by an increase in red pseudo-color intensity) along the cell membranes of K562 stable transfectants exposed to fluid shear stress for 60 sec (Fig. 2A). Compared to controls, statistically significant reductions in the FRET activity of K562 transfectants were observed only after exposure to shear stress for 60, but not for 30 sec (Fig. 2B). These results provided evidence that shear stress elicits conformational changes in the CD18 integrin, LFA-1. Fig. 1. Exposure of migrating neutrophils to fluid shear stress elicits increased levels of activated CD18 on the cell surface. Part A: Micrographs depicting the distribution of conformationally active CD18 on representative human neutrophils either maintained under control conditions (Control) or exposed to 5.3 dyn/cm2 shear stress (Shear) for 1.5 min. Images depict three cells from control experiments and three different cells from shear experiments. CD18 conformational changes were assessed using Alexa 488-conjugated monoclonal antibody 327c with specific binding affinity for the active conformation of CD18. Imafes depict three cells from control experiments and three different cells from shear experiments. White horizontal bars U 5 mm. Part B: Compared to controls (0 dyn/cm2), PMNLs exposed to 1.8 and 5.3 dyn/cm2 fluid shear stress exhibited increased surface levels of open-extended (active) conformation of CD18. Part C: Compared to controls, neutrophils stimulated with 10—8 M FMLP for 2 min exhibited significantly increased binding affinity for monoclonal antibody 327c. Black vertical bars are mean W SEM from n ‡ 3 independent experiments; *P < 0.05 compared to controls (Bonferroni-adjusted Student’s t-test). Fig. 2. Fluid shear stress elicits time-dependent conformational activation of CD18. Part A: Individual K562 stable transfectants were either maintained under no-flow conditions (Control) or exposed to a non-uniform fluid shear stress (Shear; tmax U ~2.0 dyn/cm2) using a micropipette shear setup. Whitehorizontalbars U 5 mm. Colorbarrepresentsnormalized FRET (FRETt/FRETtU0 sec) activity. Part B: Normalized FRET activity ontheperipheralcellmembraneof K562 transfectantswasassessedforcellsmaintainedunderno-flowconditionsfor 60 sec(Control) aswellascells exposed to a non-uniform shear stress field (tmax on the cell surface U ~2 dyn/cm2) for either 30 (Shear, 30 sec) or 60 (Shear, 60 sec) sec. Black vertical bars are mean W SEM from n ‡ 6 cells; MP < 0.05 compared to normalized FRET activity at t U 0 sec (Student’s t-test). Shear stress induces reductions in cell surface-associated CD18 by cathepsin B-like proteolytic activity Buffy coat populations of neutrophils, migrating on glass substrates under either no-flow conditions or while subjected to 5.3 dyn/cm2 steady fluid shear stress, exhibited uniform and constitutive levels of CD18 along the surface membrane (Fig. 3A) as detected by monoclonal antibody 6.7 (that binds to both activated and inactivated protein conformations). Furthermore, exposure of migrating neutrophils to fluid shear stress resulted in significant reductions in the amount of surface-associated CD18 (Fig. 3B). This shear-induced reduction of CD18, however, was blocked by pretreatment of the cells with CA074Me for 10 min. Treatment of cells with CA074Me for 10 min without shear had no effect on CD18 surface expression levels (Fig. 3C) confirming that shear- induced CD18 cleavage was protease-dependent and independent of integrin surface expression due to degranulation. Suspended neutrophils also exhibited CD18 cleavage under the influence of shear (Fig. 3D). Briefly, compared to controls, populations of suspended neutrophils exposed to 3 dyn/cm2 shear stress was associated with significantly decreased levels of full-length CD18 in cell lysates (Fig. 3D). In contrast, levels of cell-associated full-length CD18 in cells either maintained under no-flow conditions or exposed to shear stress in the presence of CA074, the cell-impermeable parent analog of CA074Me, were not significantly different from those of untreated controls (Fig. 3D). Additionally, suspended neutrophils exhibited a uniform distribution of CD18 along the surface membrane (Fig. 4A) that was significantly reduced after exposure to 3 dyn/cm2 fluid shear stress for either 2 or 5 min (Fig. 4A,B). The CD18 reduction was not enhanced by shearing cells in the presence of exogenous cathepsin B (Fig. 4B). Moreover, CD18 surface levels were unchanged after treatment of migrating neutrophils with exogenous cathepsin B under control (no-flow) conditions (Fig. 4B). Whereas neutrophils maintained under control conditions in either the absence of presence of exogenous cathepsin B exhibited similar levels of full-length CD18, neutrophils exposed to fluid shear stress alone for 5 min exhibited decreased levels of full-length CD18 (Fig. 4C). In the presence of exogenous cathepsin B after exposure of neutrophils to fluid shear stress, reductions in cellular levels of full-length CD18 were not enhanced (Fig. 4C). To test whether cathepsin B-mediated CD18 cleavage depends on conformational changes in the CD18 extracellular domain, we treated cells with PMA as a chemical approach to induce conformational activation. Compared to untreated neutrophils as well as to cells incubated with either PMA or cathepsin B alone, neutrophils co-incubated with PMA and cathepsin B exhibited significant reductions in cellular levels of full-length CD18 (Fig. 4D). Proteolysis plays a role in shear-induced pseudopod retraction but not migration under no-flow conditions Compared to controls (cells maintained under no-flow conditions), migrating neutrophils exposed to ~2 dyn/cm2 fluid shear for 2 min under micropipette flow retracted their pseudopods and assumed a more rounded morphology (Fig. 5A) characterized by reductions in the lengths of their major axes (Fig. 5B). In contrast, neutrophils exposed to shear stress immediately after pretreatment with CA074Me continued to extend and/or retract pseudopods (Fig. 5A) with fluctuations of major axis cell length (Fig. 5B). In the case of neutrophils maintained under no-flow conditions, cathepsin B modulated cell migration over a 10-min time period (Fig. 5C,D). Although blockade of lysosomal cathepsins, predominantly cathepsins B and L, by pretreating cells with CA074Me had no effect on random cell migration, neutrophils in the presence of exogenous cathepsin B exhibited increased migration path lengths (Fig. 5C,D). In a similar fashion, neutrophils pretreated with CA074Me and subsequently allowed to migrate in the presence of cathepsin B exhibited enhanced migration rates as compared to those of untreated controls (Fig. 5C,D). Discussion Increasing evidence points to a role for mechanical stresses in regulating adhesion and migration of human neutrophils. The current results suggest that application of fluid shear stress to neutrophils at physiological levels induces a change in CD18 protein structure and that this shift in the conformation of CD18 increases its susceptibility to proteolysis by lysosomal proteases such as cathepsin B that is released by the cells under the influence of fluid flow. Relationship between fluid shear stress and CD18 activation The present study provides evidence, by two independent, experimental approaches, that fluid shear stress elicits conformational changes in surface-associated CD18 integrins. Although we saw a 50% to 100% increase in surface binding of 327c antibody to sheared neutrophils relative to controls (Fig. 1), we only detected a relatively small decrease in FRET activity for K562 cells stably transfected with the LFA-1 FRET reporter (Fig. 2). This small but significant decrease in FRET activity (Fig. 2B) may have resulted from the lower sensitivity of the FRET construct. In fact, stimulation of the K562 transfectants with near-maximal concentrations (100 ng/ml) of PMA resulted only in an ~25% decrease in FRET efficiency (Kim et al., 2003) using the more sensitive acceptor/photo-bleach method which measures the ratio of CFP fluorescence when energy transfer was permitted to the total amount of CFP present on the cell surface when energy transfer to YFP was blocked by photo-bleaching. Since we could not measure CFP fluorescence before and after photo-bleaching as it would have required exposure of the cells to shear stress for two sequential time periods, we normalized the ratio of YFP/CFP before shear exposure to the same ratio after terminating shear stress application. This approach served to not only increase the sensitivity of our measurements but at the same time take into account any differences in fluorescence intensity background when comparing FRET activity between different cells. Another possibility which may explain the small difference in FRET efficiency between sheared and unsheared K562 transfectants (Fig. 2) is that fluid shear may be less potent an activator of CD18 than PMA. Regardless of the technical limitations, our results demonstrated using two independent experimental approaches conducted on two separate cell lines that fluid shear is capable of eliciting conformational changes in CD18. Conformational activity of CD18, however, did not appear to correlate spatially with the direction of flow over the cell surface (Figs. 1 and 2). For adhesive leukocytes, the shear stress due to flow from either a parallel plate flow chamber or micropipette tends to be low in the contact region and reach a maximum at the top of the cell exposed to the flow (Sugihara- Seki and Schmid-Scho¨ nbein, 2003). In addition, there are non- uniformities in shear stress on the neutrophil surface resulting from the numerous membrane folds for both migrating and suspended cells subjected to flow. In spite of this, we did not detect any correlation between CD18 conformational activity and areas of maximum shear stress (Fig. 1A). One possibility is that shear-induced conformational changes in CD18 are downstream of an as yet unidentified mechanosensor. Alternatively, the conformation of membrane-bound CD18 is insensitive to the small variations in membrane shear stresses resulting from the geometry-dependent variations of the cell surface. In fact the differences in amounts of activated CD18 on neutrophil surfaces exposed to either 1.8 or 5.3 dyn/cm2 (Fig. 1B) were not statistically significant suggesting that flow-induced changes in CD18 conformations require greater than approximately threefold increases in shear magnitudes. Fig. 4. Shear and PMA facilitates cathepsin B-related cleavage of surface-associated CD18 on purified populations of neutrophils in suspension. Part A: Representative human neutrophils in suspension eithermaintainedunder no-flow (Control) conditions orexposed to 3 dyn/cm2 fluid shear stress for 2 min(Shear, 2 min) and 5 min (Shear, 5 min) using a cone-plateviscometer. Expression of CD18 on the cell surface was assessed for fixed, but non-permeabilized, cells using monoclonal antibody 6.7 directed against the extracellular domain. White horizontal bars U 5 mm. Part B: Compared to controls, neutrophils exposed to 3 dyn/cm2 fluid shear stress for either 2 or 5 min exhibited decreased levels of CD18. Incubation of controls (Cath. B) orneutrophils exposed tofluid shear for 5 min with 0.5 U/ml exogenous cathepsin B (Shear, 5 min R Cath. B) did notsignificantly affect overall levels of CD18 relative to respective untreated cells. Black vertical bars are mean W SEM from n ‡ 3 independent experiments; M,zP < 0.05 compared to control or Cath. B, respectively (Bonferroni-adjusted Student’s t-test). Part C: CD18 cleavage was assessed in suspended human neutrophils exposed to 3 dyn/cm2 fluid shear stress in the absence or presence of cathepsin B for 5 min using Western blot analysis with polyclonalantibody(clone C-20) tothecytoplasmicdomainof CD18. Part D: CD18 cleavage wasassessedinsuspendedneutrophilsstimulatedwith 100 nM PMA in the absence (PMA) or presence of 0.5 U/ml exogenous cathepsin B (PMA R Cath. B). Controls were unstimulated neutrophils (Control) or neutrophils treated with exogenous cathepsin B (Cath. B) for 30 min. Black vertical bars are mean W SEM from n ‡ 3 independent experiments; M,#P < 0.05 compared to control or PMA, respectively (Bonferroni-adjusted Student’s t-test). CD18 cleavage under shear stress Our results demonstrating shear-induced CD18 cleavage by both immunofluorescence analyses targeted for the extracellular domain of CD18 and Western blot analyses targeted for the cell-associated CD18 fragments provide strong support for the hypothetical role of shear stress in downregulating surface levels of full-length CD18. We were unable, however, to repeatedly detect CD18 fragments released into the supernatant (i.e., extracellular domain fragments) or ‘‘left behind’’ on the cell surface (C-terminus- containing peptides). Because of the low number of cells available for our shear experiments, the observed reduction of cell-associated full-length CD18 (Fig. 4C) likely resulted in generating only a marginal amount CD18 fragments beyond the detection limits of our protein analysis. There is also a possibility that the cleaved fragments may have been subject to further proteolytic degradation. Our shear experiments, however, were conducted for only 2–5 min (Fig. 4B), which is well below the time period necessary for cells to either synthesize additional CD18 or internalize and substantially degrade fragmented CD18. Moreover, Western blot analyses of total cell lysates are incapable of differentiating between internalized and surface-associated CD18 fragments. The observed reductions in cell-associated full-length CD18 were, therefore, most likely the result of shear-induced proteolysis. The involvement of an active proteolytic process in regulating CD18 surface expression on freely suspended neutrophils exposed to fluid shear stress was first proposed by Fukuda and Schmid-Scho¨ nbein (2003). Specifically, these investigators (Fukuda and Schmid-Scho¨ nbein, 2003) identified lysosomal cysteine proteases, including cathepsin B, as candidate enzyme(s) involved in the CD18 cleavage response of neutrophils to shear stress using a panel of broad-spectrum protease inhibitors as well as the selective inhibitor, CA074Me. It is possible, however, that CD18 cleavage under the influence of shear stress may not involve cathepsin B or only cathepsin B in light of evidence demonstrating the inhibitory activity of CA074Me on other related lysosomal cathepsins (Bogyo et al., 2000), particularly cathepsin L (Montaser et al., 2002). Furthermore, we were unable to detect changes in the levels of cathepsin B protein (as assessed by Western blot using antigen- specific monoclonal antibodies) or activity (as measured using the cathepsin B-selective fluorogenic substrate Z-Arg-Arg- AMC (Hulkower et al., 2000)) in the supernatants and the whole-cell lysates of neutrophils exposed to shear (data not shown), but this may have been due to the small amounts of protease released over the 5-min duration of experiments. In MP < 0.05 compared to controls and zP < 0.05 compared to CA074Me using Bonferroni-adjusted Student’s t-test. Part C: Representative micrographs of cells migrating on glass slides under no-flow conditions for 10 min. Images depict the cell position at t U 0 (shaded in black) and at t U 10 minconnectedbyawhitepath. Part D: Migrationpathlengthswerequantifiedforuntreatedcells(Untreated) aswellasforcellsthathadbeen pretreatedwithprotease inhibitor (CA074Me) and allowed tomigrateineitherthe absence orpresence of 0.5 U/mlcathepsin B (Cath. B). Bars are mean path length W SEM from n U 9–20 cells. #P < 0.05 compared to untreated cells (Bonferroni-adjusted Student’s t-test). The mechanism by which cathepsin B or other cathepsins (e.g., cathepsin L) exerts their activity on the extracellular domain of CD18 under the influence of shear stress exposure remains to be determined. It is known that cathepsins B and L play more than just a role in protein turnover in lysosomes (Turk et al., 2000). For example, these cathepsins can be secreted and contribute to protein degradative processes outside the cell such as cleavage of extracellular matrix proteins (Buck et al., 1992; Felbor et al., 2000; Sameni et al., 2000; Mai et al., 2002). Furthermore, lysosomal cathepsins such as cathepsin B reportedly activates soluble and receptor-bound urokinase-type plasminogen activator through a cleavage process at the surfaces of tumor cells (Kobayashi et al., 1991, 1993). This evidence in conjunction with the current results (Fig. 3D) using the cell-impermeable, cathepsin B-specific inhibitor, CA074, is consistent with the possibility that shear stress-induced CD18 cleavage involves secretion of cysteine proteases such as cathepsin B. Exogenous cathepsin B added to cell suspensions, however, was not sufficient to induce detectable reductions in surface levels of CD18 on neutrophils under no-flow conditions even after incubation for 30 min (Fig. 4B,C). This observation points to the requirement for fluid shear stress to facilitate proteolysis. Interestingly, exposure of neutrophils to 3 dyn/cm2 shear stress for 5 min in the presence of exogenous cathepsin B did not further reduce membrane surface levels of full-length CD18 (Fig. 4B) suggesting the presence of a cellular mechanism(s) that maintains threshold levels of CD18 integrins on the surface and/or the existence of a population of CD18 integrins not cleaved by shear exposure. Another possibility is that cleavage of CD18 is most efficient when cathepsin B is transported to the cell membrane from intracellular rather than extracellular sources. In fact, cathepsin B is thought to translocate to the cell surface and act on extracellular or surface-associated proteins after transport from lysosomes for tumor cells (Hulkower et al., 2000; Mai et al., 2000a,b) and T-lymphocytes (Balaji et al., 2002). Furthermore, activity of cathepsin B is reportedly enhanced by its close association with heparan sulfate proteoglycans characteristic of those found on the surfaces of cells (Almeida et al., 2001). The evidence in our studies and in these reports suggests that even low levels of extracellular cathepsin B activity may elicit cleavage of CD18 integrins and that this proteolytic process may depend on a component on the cell membrane that retains cathepsin B released from endogenous sources or, at least, increases its residence time after its translocation to the surface under the influence of shear stress. Furthermore, this shear-related mechanism is tied into cathepsin B release from lysosomes but not accessible by exogenous cathepsin B. It is also possible that cathepsin B or other cathepsin-related proteases are only released in regions on the cell surface that experiences a threshold level of membrane stress. In the case of suspended cells, CD18 activation and cleavage presumably resulted only from shear-activated CD18 with cathepsin B release over the entire surface of the cell. On migrating cells, regions of high fluid stress would coincide with regions at the top of the cell (Su, 2007) and its microvilli. Shear stresses at the top of the cell are balanced by stresses in the F-actin cytoskeleton and by stresses in the membrane contact region where a fraction (estimated to be ~10%) of CD18 (in the activated state) mediates the cell adhesion and facilitates migration (Lum et al., 2002). This would explain the small fraction of CD18 that is activated and acted upon by proteases such as cathepsin B. Relationship between shear-induced CD18 activation and CD18 cleavage Activation of CD18 integrins, such as CD11a/CD18 (lymphocyte function adhesion molecule-1) and CD11b/CD18 (Mac-1; the most abundant CD18 receptor on the neutrophil) is accompanied by a structural switch from a close-folded to an open-extended protein conformation (Simon and Goldsmith, 2002; Kim et al., 2003) that decreases steric hindrance (Carman and Springer, 2003) in the ectodomain required for efficient binding to ligands (e.g., ICAM-1). Based on this information, we hypothesized that an extended conformation of CD18 also plays a role in its cleavage under shear by facilitating access of cathepsin B or other proteases to cleavage sites on the extracellular domain. To test our hypothesis, we explored the ability of cathepsin B to cleave CD18 in the absence or presence of PMA, an inflammatory mediator known to induce activation of CD18 integrins (Beals et al., 2001; Kim et al., 2003). Our results suggest that shear stress may be linked to cathepsin B-mediated cleavage by altering the conformational state of CD18 since stimulation of neutrophils with PMA significantly reduced levels of full-length CD18 only upon co-incubation with cathepsin B (Fig. 4D). Since co-incubation of neutrophils with cathepsin B and PMA reduced full-length CD18 levels to a lesser extent than when cells were exposed to fluid shear stress (Fig. 4B,C) it is possible that there exists either a population of CD18 that is protected from cathepsin B cleavage or cellular mechanism(s) that enhances the efficiency of cathepsin B-mediated proteolysis at the plasma membrane. The small, but significant, reduction in cellular levels of full-length CD18 resulting from co-incubation of cells with PMA and cathepsin B (Fig. 4D), however, provides evidence that conformational changes in CD18 contribute to its susceptibility to proteolytic cleavage. Relationship between CD18 cleavage and pseudopod retraction We provide evidence that shear-induced pseudopod retraction depends on CD18 cleavage by cathepsin B and/or other lysosomal proteases released under the influence of fluid shear stress (Fig. 5A,B). Our evidence centers around the ability of CA074Me pretreatments to block shear pseudopod retraction on the neutrophil (Fig. 5A,B). Since CA074Me also exhibits activity for other cathepsins, for example, cathepsin L, we cannot rule out the possibility that cathepsin L plays some role in this process. Regardless, this finding (Fig. 5A,B), in conjunction with the observation that CD18-mediated attachments are required for shear-induced pseudopod retraction by migrating neutrophils (Marschel and Schmid- Scho¨ nbein, 2002), are consistent with the requirement for active cleavage of CD18 as a key step in the underlying cellular mechanism. Furthermore, the actions of lysosomal cysteine proteases such as cathepsin B and L seem to be limited to neutrophil pseudopod retraction during shear stress exposure since CA074Me pretreatments did not alter neutrophil migration rates in the absence of shear stress (Fig. 5C,D). Surprisingly, addition of exogenous cathepsin B increased the migration rate of neutrophils (Fig. 5C,D) presumably due to enhanced rates of integrin cleavage that cooperates with other physiological migration mechanisms involved in neutrophil motility. This is in agreement with previous reports (Palecek et al., 1996) indicating that cell migration is limited not only by rates of pseudopod extension, but also by rates of pseudopod or uropod retraction involving intracellular cleavage of integrin- cytoskeletal connections as well as adhesion receptor debinding pathways. Release of cathepsin B or other related proteases may, therefore, serve as a mechanism to enhance retraction of pseudopods only under shear stress at sites of CD18 mediated attachments where these adhesion receptors adopt a conformation susceptible to cleavage by proteases released by the cells. Conclusion Fluid shear stress on migrating leukocytes causes retraction of pseudopods and reduced cell stiffness thereby turning an ‘‘activated’’ cell with cellular projections in the circulation into a rounded, ‘‘inactivated’’ cell that passes freely through the small vessels of the microcirculation. In fact, lack of a shear stress response (e.g., pseudopod retraction) associated with cytokine-stimulated neutrophils as well as neutrophils from either spontaneously hypertensive or glucocorticoid-treated rats leads to an increased number of neutrophils adhered to or entrapped within the vessels of the microcirculation (Fukuda et al., 2004a,b). Due to the important role of CD18 in mediating adhesion and migration of neutrophils during initial recruitment from the bloodstream (Springer et al., 1984; Walzog et al., 1999), it is likely that a 10–20% reduction in surface expression of this receptor could also have a significant impact in modulating the attachment of the neutrophil to endothelium under blood flow (Simon and Goldsmith, 2002). The salient finding of the present study is that shear stress stimulation increases the susceptibility of select CD18 populations to proteolysis through its actions on receptor conformation. This cellular mechanism may play a critical role in facilitating a membrane detachment process that, in addition to depolymerization of cytoskeletal F-actin due to deactivation of G-protein-couple receptors (GPCR) and small GTPases (Rac) (Makino et al., 2005, 2006), contributes to pseudopod retraction by neutrophils under the influence of shear stress. Moreover, this mechanism may serve to keep neutrophils in the circulation without attaching and spreading on the endothelium under normal physiological conditions. This concept is consistent with recent evidence demonstrating Mac-1 (Youker et al., 2000) and LFA-1 (Evans et al., 2006) shedding by neutrophils as putative mechanisms to detach integrins from their counter-receptors or prevent further migration at sites of inflammation. The proposed mechanotransduction mechanism may, therefore, be critical for regulating neutrophil adhesion on the microvascular wall and serve as an anti-inflammatory measure. In this regard, the fluid shear stress response has to be regarded as a key player, in addition to traditional biochemical mediators, that contributes to dysregulated inflammation in the pathogenesis of cardiovascular diseases (e.g., hypertension) that lead to such complications as increased peripheral vascular resistance, capillary rarefaction,CA-074 methyl ester and organ injury.