| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institute for Medicine and Engineering, Department of Bioengineering, and Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania
Correspondence: Address reprint requests to S. L. Diamond, Tel.: 215-573-5702; E-mail: sld{at}seas.upenn.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
), membrane tethering fraction (f), membrane tether growth dynamics, and PSGL-1/P-selectin binding lifetime. For 1391 collisions analyzed over venous wall shear rates from 25 to 200 s–1, decreased from 0.17 to 0.004, whereas f increased from 0.15 to 0.70, and the average projected membrane tether length, 
increased from 0.35 µm to 
2.0 µm over this shear range. At all shear rates tested, adhesive collisions lacking membrane tethers had average bond lifetimes less than those observed for collisions with tethers. For adhesive collisions that failed to form membrane tethers, the regressed Bell parameters (consistent with single bond Monte Carlo simulation) were zero-stress off-rate, koff(0) = 0.56 s–1 and reactive compliance, r = 0.10 nm, similar to published atomic force microscopy (AFM) measurements. For all adhesion events (± tethers), the bond lifetime distributions were more similar to those obtained by rolling assay and best simulated by Monte Carlo with the above Bell parameters and an average of 1.48 bonds (n = 1 bond (67%), n = 2 (22%), and n = 3–5 (11%)). For collisions at 100 s–1, pretreatment of neutrophils with actin depolymerizing agents, latrunculin or cytochalasin D, had no effect on , but increased 
by 1.74- or 2.65-fold and prolonged the average tether lifetime by 1.41- or 1.65-fold, respectively. Jasplakinolide, an actin polymerizing agent known to cause blebbing, yielded results similar to the depolymerizing agents. Conversely, cholesterol-depletion with methyl-ß-cyclodextrin or formaldehyde fixation had no effect on , but reduced by 66% or 97% and reduced the average tether lifetime by 30% or 42%, respectively. The neutrophil-bead collision assay combines advantages of atomic force microscopy (small contact zone), aggregometry (discrete interactions), micropipette manipulation (tether visualization), and rolling assays (physiologic flow loading). Membrane tether growth can be enhanced or reduced pharmacologically with consequent effects on PSGL-1/P-selectin lifetimes. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
The Bell model is an empirically useful approach for quantifying the force sensitivity of bond lifetimes where koff = koff(0) exp(rFB/kBT) and koff is the dissociation constant; FB is the force on the bond; koff(0) and r are the experimentally determined unstressed off-rate and reactive compliance, respectively; kB is the Boltzmann constant; and T is the temperature. Bell model parameters have been measured for selectin-ligand pairs in a variety of ways, including rolling cells and/or rolling beads (1
–5); atomic force microscopy (AFM (6–8)); molecular force probe (9,10); laser trap (11); and dynamic force spectroscopy (DFS (12–14)). Values of koff(0) determined by extrapolation to zero force for PSGL-1/P-selectin are consistent with those obtained in solution phase using plasmon resonance (15). These different mechanical methods create various force loading profiles to drive bond rupture dynamics (16). At venous wall shear rates of 25–200 s–1, physiological loading during neutrophil capture is characterized by rapid loading (1–50 ms) to a peak value (10 pN to 100s pN) with subsequent force relaxation due to tether pulling (17–19). Experimental challenges include whether only single bond interactions are being evaluated (20), the requirement of estimating forces and geometries on both microscopic and macroscopic length scales, the integrity of purified molecule presentation, and the temporal/spatial resolution of each method. In general, the koff(0) of L-selectin is considered to be greater than that of P-selectin or E-selectin, consistent with the fact that neutrophils roll on L-selectin at a velocity 7.5–11.5 times faster than they do on either P or E-selectin (21).
The dimeric PSGL-1 molecule is located primarily on the microvilli tips of neutrophils. Microvilli extend 0.25–0.35 µm from the neutrophil surface as measured by electron microscopy or micropipette aspiration (17), and as calculated based upon comparison of the off-rates of fixed cells and beads (3). Micropipette aspiration techniques have been developed to explore the mechanics of tether formation from neutrophils (17,22,23) and from endothelial cells (24). Membrane tether formation using micropipette manipulation has revealed two regimes of tether growth: microvilli extension, characterized by linear elastic deformation of the neutrophil (FB < 34 pN); and membrane tether formation, in which lipid membrane flows freely and at a constant velocity when FB is >45–61 pN (17,19,22). Between 34 and 61 pN, the formation of a membrane tether versus an extended microvillus depends upon the degree of interaction between cytoskeleton and membrane (17). Membrane tether pulling during flow perfusion over adherent platelets or selectin-coated surfaces is visualized as translation of the cell downstream of an adhesion point at a constant velocity well below that of the free-stream velocity (18).
Pharmacological agents targeting actin polymerization can be used to study the relationship between the cytoskeleton function and cell mechanics. Actin polymerization and membrane fluidity in the leukocyte can be altered using actin-polymerizing agents such as jasplakinolide or depolymerizing agents such as cytochalasins or latrunculin (25–27). Cytochalasins depolymerize actin, although their modes of action and specificity are varied (26). Latrunculin A is a more specific actin depolymerizing agent that binds to actin monomer and prevents incorporation into F-actin. Jasplakinolide increases actin polymerization and stabilization by decreasing the amount of sequestered actin and enhancing F-actin nucleation (28–29) with some observed cytotoxicity (30). Neutrophils treated with cytochalasin B or D exhibit a decrease in cortical tension and cytoplasmic viscosity during aspiration into micropipettes (31–33) or tether extraction by micropipette (23). Similar increases in deformability after treatment with cytochalasin B were also observed in eosinophils (27). Perfusion of neutrophils treated with cytochalasin is associated with localization of PSGL-1 to the pointed ends of teardrop-shaped cells (34). Actin depolymerization is also associated with a loss of membrane ruffles, as evidenced by 1), the ability of cytochalasin B to abolish tether formation (35); and 2), the smooth shape of latrunculin A-treated neutrophils under scanning electron microscope (SEM) (36). In contrast, jasplakinolide treatment of neutrophils results in an increase in global rigidity as measured by aspiration into a micropipette (37). However, AFM elasticity studies of jasplakinolide, cytochalasin, or latrunculin-treated neutrophils suggests that all three reagents reduced the local Young's modulus of the plasmalemma, possibly due to the effect that jasplakinolide caused disaggregation but not disassembly of the actin fibers (38). Also, jasplakinolide can cause membrane blebbing (37). In contrast, the membrane cholesterol content of a cell can also be manipulated to study the effects of membrane fluidity on cell adhesion. Methyl-ß-cyclodextrin (MßCD) can be used alone or in complexes with cholesterol to deplete or enrich the plasmalemma with cholesterol, respectively (39). Cholesterol depletion in neutrophils has been shown to be associated with resistance to deformation during aspiration (33), as does chemical fixation. Unexpectedly, MßCD has been reported to reduce neutrophil actin polymerization (40).
To study the effect of neutrophil membrane tether formation on selectin bond mechanics, we developed a new method of probing interactions between a flowing cell and a small point source of ligand presented on a 1-µm-diameter adherent bead. The advantages of this approach:
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
Human blood was obtained via venipuncture from healthy adult donors who had not taken any medications in the prior 10 days. Blood was mixed with anticoagulant sodium citrate (9 parts blood to 1 part citrate). Neutrophils were isolated by centrifugation with a separation medium (Cedarlane Laboratories, Hornby, Ontario) as previously described (18,41). Neutrophils were counted and diluted with a 2% solution of human serum albumin in Hank's Balanced Salt Solution (HBSS; Invitrogen, Carlsbad, CA) with Ca2+ to a final concentration of 1–2 x 106 cells/ml. In some cases, neutrophils were treated with jasplakinolide (10 µM), cytochalasin D (15 µM), latrunculin (0.3 µM), or MßCD (10 mM) for 30 min at room temperature before perfusion. In the case of treatment with formaldehyde, cells were fixed for 15 min with a 3% formaldehyde solution. The cells were then washed and resuspended in HBSS with Ca2+ and 2% HSA.
Neutrophil-bead collision assay
Microspheres coated with Protein A were labeled with P-selectin by the method described by Rodgers, et al. (42). Beads were incubated with periodic vortexing for 45 min at room temperature with up to 0.2 µg/ml P-selectin-IgG chimera. This yielded a Protein A-saturated surface density of 5.4 x 103 IgG-chimera/µm2. After labeling, the microspheres were then washed three times with a solution of 1% HSA in HBSS. In preparation for attachment to the chamber, the beads were then incubated for 4 h at 4°C with 200 µg/ml fibrinogen. Rectangular glass capillaries (Vitrocom, Mountain Lakes, NJ) having a cross-section of 0.2 x 2.0 mm, a length of 7 cm, and a wall thickness of 0.15 mm were used as flow chambers, as described previously (18,41). Chambers were incubated with fibrinogen (100 µg/ml) for 120 min at room temperature. The microspheres were perfused into the chambers, and the bead-coated chambers were incubated at room temperature overnight in order for the beads to have full adhesion to the capillary surface. Bead-coated flow chamber surfaces were then washed and blocked with HBSS with 2% HSA for 5 min before perfusion studies. Fibrinogen caused no ß2-integrin mediated adhesion since tests employing anti-PSGL-1, calcium-free buffer, or P-selectin-free beads all resulted in zero detectable adhesion (see Results), and resting neutrophils were used that failed to adhere to the chamber walls even at 
w = 25 s–1.
Neutrophils were perfused into the chambers using a syringe pump (Harvard Apparatus, Holliston, MA) at various wall shear rates (w) as calculated via the Navier-Stokes equation for laminar flow of a Newtonian fluid: w = 6Q/B2W, where Q represents the flow rate (cm3/s), B is the total plate separation (0.02 cm), and W is the width (0.2 cm). The wall shear stress can be calculated from 
w = (6Qµ)/(B2W), where µ is the buffer viscosity (0.01 Poise at room temperature). For differential interference contrast microscopy, a Zeiss Axiovert 135 microscope with a 63x (NA 1.40) oil immersion objective lens (Plan Apochromat) was used (Carl Zeiss, Jena, Germany). For high-speed imaging, images were captured using a MotionCorder Analyzer high-speed digital camera (Eastman Kodak, Rochester, NY) at an imaging rate of 240 frames per second (fps). The images were played back at 5 fps or frame-by-frame to videotape for analysis.
During frame-by-frame analysis, adhesive interactions were identified as those collisions that had a visible pause in neutrophil motion lasting for at least one frame, and had velocities that had decreased below the hydrodynamic velocity (Fig. 1, A–D). An adhesion collision efficiency () was calculated from these observations as the number of collisions resulting in an adhesive interaction, divided by the total number of collisions observed. The lifetimes of all adhesive interactions were recorded. These values always greatly exceeded those of simple hydrodynamic collision without bonding. Adhesive neutrophils were defined as tether-forming when they translated in the direction of flow at a velocity below the hydrodynamic stream velocity (18). Membrane tethering fraction (f) was defined as the ratio of tether-forming events to the total number of adhesive interactions. Tether lengths were measured using Scion Image or ImageJ software (National Institutes of Health, Bethesda, MD) from the center of the adhesive bead to the lagging edge of the neutrophil. Approach and release velocities were determined by collecting images of neutrophils as they approached, interacted, and released from a bead at 240 fps.
|
),
 | (1) |
 | (2) |
is the measured projected tether length between the top of the bead and the tether attachment point on the neutrophil. In the case of multiple bond formation, the force was considered equally shared among the bonds. The length of the lever arm, l, and the angle of attachment 
was calculated. A radius R = 4.25 µm was used for the neutrophil (11,18). For the case of neutrophils that formed adhesive interactions with tethers of a length less than the minimum discernible length of 500 nm (Fig. 1 A), we estimated the lever arm to be the radius of the cell, R. This approximation yields a maximum error of 24.0% when the cell is considered to be an incompressible sphere, and this error decreases logarithmically as the lever arms increases.
Goldman's equations (43) for the relationship between shear stress and the forces experienced by a flowing sphere as it flows parallel to a plane wall were used to solve for Fs and Ts:
 | (3) |
 | (4) |
, where is the gap distance between the neutrophil and the chamber surface (Fig. 1 E). The values C and D are constants that are determined from the ratio of h/R. Release velocities of adherent neutrophils at 100 s–1 were determined from distance versus time plots (Fig. 1 D), and were used to determine the average gap distance between neutrophil and chamber surface, release = 0.38 µm. This gap distance is similar to that of neutrophils releasing from adherent platelets (0.2–0.3 µm; (18)), and yielded values of C and D of 1.64 and 0.95, respectively, based upon data given in Goldman et al. (43). In addition, approach gap distances for the nonadhesive interactions ( = 0.86 ± 0.24 µm; n = 19 interactions) were similar to those for the adhesive interactions ( = 0.86 ± 0.29 µm; n = 25) for the portion of interactions observed. Using the measured projected tether length, Eqs. 1–4 were solved for and FB. For neutrophils that formed tethers of the minimum discernible length (
), the angle was calculated to be 53.6°.
Kinetic Monte Carlo (MC)
Simulations were performed as previously described (44) using Gillespie's algorithm for stochastic reactions (45). Four possible event types were allowed at a given time-step: 1), breaking a P-selectin/PSGL-1 bond; 2), creating a new bond; 3), growth of the tether; and 4), departure of the neutrophil from the bead after the final bond broke. A bond was defined with respect to the surface densities on a per-protein interface basis such that a total of two bonds could be counted between one PSGL-1 and one P-selectin molecule, since both are dimers. According to the stochastic theory, each of the events has the following probability: 1), 
The contact area (0.071 µm2) was assumed to be a circle with the diameter of the tether (dt 300 nm; (46)). We assumed that the initial conditions for the simulations consisted of n = 1 or 2 bonds, XPSGL–1 = 100 free PSGL-1 and XP–sel = 770 P-selectins (for 50 PSGL-1 dimers per microvillus (47,48), for 10–15 thousand dimers on 252 microvilli per cell (48,49); 5.4 x 103 P-selectin chimeras/µm2). For each event simulated, the tether growth velocity was randomly selected from the experimental mean tether growth rate (
) distribution obtained at each wall shear rate (). The value n,0 is the -function and equals 1 when no bonds are present, thus allowing escape of the neutrophil.
Using MC, the values of kon, 
and r were determined by minimizing the global error between the simulated and measured bond lives at all shear rates. The Bell model was used for determining the off-rate 
The simulated data was compared to experimental data at each shear rate, and the global weighted fitness over all shear rates was evaluated by minimizing
 | (5) |
,ex is the number of experimental data points at a given shear rate; Ntot,ex is the total number of experimental data points; and 
and 
are the experimental and simulated mean times of the adhesion events. For other simulations with a variable number of initial bonds, the bonds were distributed exponentially with up to five bonds initially present via
 | (6) |
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
We measured the adhesion collision efficiency for neutrophil collisions with 1.0-µm P-selectin-coated beads at various shear rates for three separate donors. The collision efficiency decreased from = 0.17 to 0.0043 as w was increased from 25 to 200 s–1 (Fig. 2 A). The specificity of bonding interaction was verified at 100 s–1 through the use of anti-PSGL-1 monoclonal antibody, adherent beads containing no P-selectin chimeric IgG, and calcium-free HBSS. In each of these three control cases, no adhesive collisions between neutrophils and beads were observed in over 150 collisions ( = 0) monitored at each condition (data not shown), compared to = 0.079 for P-selectin-coated beads.
|
w = 200 s–1, where N = 2 due to low capture efficiency ( = 0.0043). Neutrophils were least likely to form membrane tethers at the lowest shear rates, where membrane tether formation fraction was 0.15 ± 0.08 at w = 25 s–1. The tethering fraction increased to a plateau of 0.62–0.70 at w 
75 s–1, indicating that neutrophil rolling at venous flows is typically accompanied by tether pulling.
We measured the projected membrane tether length, for each interaction and at each shear rate. For adhesive interactions that formed tethers of length <500 nm (below detection), we assigned a length of 0 µm. For average membrane tethers lengths calculated for all adhesive interactions, increased from a minimum of 0.35 ± 0.9 µm at w = 25 s–1 to a maximum of 2 µm at w 75 s–1 (Fig. 2 C), in agreement with prior estimates obtained by reconciling neutrophil and rigid bead pause time distributions (3).
At all shear rates tested, the longevity of binding events that had tethers exceeded the lifetime of events lacking tethers (p < 0.005) (Fig. 2 D). We found that the average overall lifetime of adhesion decreased with increasing wall shear rate from 1.38 ± 0.96 s at 25 s–1 to 0.14 ± 0.11 s at 200 s–1 (Fig. 2 E). The average lifetime for interactions with and without membrane tether formation exhibited the same trend, decreasing with an increase in wall shear rate (Fig. 2 D). Due to the overall long lifetimes of adhesive bonds at low shear rates, the effect of tether formation at w < 50 s–1 was to increase the overall lifetime of the bond by 1.5-fold. However, at wall shear rates 100 s–1, adhesive interactions with membrane tethers experienced lifetimes 3–3.2-fold greater than those without membrane tether formation. These prolongations are consistent with the original predictions of Shao et al. (17). Under all shear conditions, the life of adhesive events greatly exceeded that of nonadhesive hydrodynamic collisions (Fig. 2 E).
From tether lifetime and length data, we calculated a membrane tether growth rate (vtether) for each membrane-tether-forming interaction. These growth rates were averaged at each shear rate tested, and the resultant velocities were plotted as a function of shear rate (Fig. 3). The tether growth rate increased linearly from an average vtether = 1.4–21.3 µm/s over the shear range tested. These results agree with those previously obtained for neutrophil interactions with adherent platelets (18). The critical wall shear rate (x-intercept of Fig. 3) to initiate tether growth was calculated as w = 24 s–1 corresponding to a critical force of Fcrit = 24.3 pN (±10%), slightly lower than prior measurements made using micromanipulated neutrophils (22,23).
|
for each event). In reality, the set of events collected at each wall shear rate was heterogeneous, since the applied force depends on the length of the tether for each event. For dissociation of the PSGL-1/P-selectin adhesions following first-order kinetics, the decay in number of attached cells was used to graphically fit the apparent kinetic off-rate at each shear rate analyzed (Fig. 4 A), which agreed with the statistical point estimate (SPE) of koff = 1/ 
(44). Despite the observed variation in tether lengths (and thus force loading) across the events measured at each wall shear rate, the regression plots were highly linear (R2 = 0.98 or 0.99), indicating that linearity is not necessarily proof that a single class of events is being analyzed. The average forces were then used to regress the Bell parameters from values of koff (from the slopes from Fig. 4 A) to obtain the apparent unloaded off-rates and reactive compliance for all data (± tethers) (Fig. 4 B) or events lacking tethers (Fig. 4 C). The koff at w = 200 s–1 was determined from only 20 events (Fig. 4 B) or eight events (Fig. 4 C) due to the extremely low capture efficiency at this shear rate (approximately four in every 1000 collisions generated an adhesive event). Thus, we also regressed the graph of FB versus koff with (SPE 2 in Fig. 4, B or C) and without (SPE 1 in Fig. 4, B or C) inclusion of koff at 200 s–1. The weighted Monte Carlo simulation of all shear rates corresponded best with the SPE regression where data at w = 200 s–1 was neglected. Cell flattening may also be more pronounced at 200 s–1, thus reducing force loading at this higher level of flow.
|
±< 8%) for all events (± tethers) and for w = 25–125 s–1. In comparing these Bell parameters with earlier literature reports (curves 2–14 in Fig. 5), we then evaluated the Bell parameters (by SPE and by MC) for only those events that displayed no detectable membrane tether since the force applied on the adhesion is the same from event to event at a given wall shear rate. These events yielded Bell parameters (koff(0) = 0.53 s–1, r = 0.10 nm) that 1), were best modeled by single-bond MC (where almost no additional bonds formed); 2), were very poorly fit by a multiple bonding hypothesis (n = 2 bonds initially); and 3), were quite consistent with literature values shown in curves 2–8 in Fig. 5 and the recent DFS data of Heinrich et al. (14).
|
|
w = 200 s–1 was quite sparse as a consequence of the extremely low capture efficiency at this shear rate and contributed little to the MC determination of the Bell parameters. Using the Bell parameters obtained for no tether growth and kon = 0, the initial condition of n = 2 bonds (a fully engaged dimer interaction) allowed generation of a set of pause times (curve 1c of Fig. 5) quite consistent with the data of Park et al. (3), a result also obtained by King et al. (50). This simulation matched curves 12–14 and was then fit with apparent parameters of (koff)app = 0.76 s–1 and rapp = 0.05 nm. Figs. 4 and 5 indicate that the neutrophil-bead collision assay allows accurate measurement of lifetime distributions that are consistent with prior measurements and analyses for the PSGL-1/P-selectin interaction.
Chemical treatment of perfused neutrophils
Neutrophils were treated with cytochalasin D (15 µM), latrunculin (0.3 µM), jasplakinolide (10 µM), MßCD (10 mM), or formaldehyde (3% v/v) for 30 min. Latrunculin-treated neutrophils that were perfused at 100 s–1 formed an unusual teardrop shape (Fig. 1 F). This effect was not observed with any of the other treatments.
None of the treatments significantly altered the primary capture efficiency of the neutrophils with the P-selectin-coated bead (Fig. 6 A), indicating that the reagents had minimal effect on the concentration or activity of PSGL-1 on the neutrophil surface. Only latrunculin treatment caused a slight increase in the fraction of cells that formed membrane tethers (p = 0.024; Fig. 6 B) due to the extreme softness of the cells. Cytochalasin D and jasplakinolide had no significant effect on the membrane tether formation fraction. Cholesterol-depleting MßCD significantly reduced the likelihood that cells would form membrane tethers at w = 100 s–1 (p = 0.0007), as did fixation of the cells with formaldehyde (p < 0.0001). All of the treatments significantly altered the average membrane tether length pulled from the cell (Fig. 6 C), with actin-depolymerizing reagents associated with an overall increase in membrane tether length, and cholesterol depletion and fixation with decreased membrane tether lengths. Interestingly, jasplakinolide, which acts to polymerize actin in the cell, had an affect similar to the depolymerizing reagents—likely due to the membrane blebbing observed.
|
The tether growth rates (vtether) for membrane-tether-forming interactions of treated neutrophils with P-selectin-coated beads were also analyzed (Fig. 7). These results were compared against the average tether growth rate for untreated cells perfused over beads at 100 s–1 (average vtether = 8.1 ± 4.9 µm/s), and Student's t-tests were performed comparing the set of control cell growth rates to those for the treated cells. In all cases except latrunculin (where cells had an unusual morphology), treatment with actin- or cholesterol-affecting reagents caused significant differences between the two samples. Cytochalasin and jasplakinolide both caused tethers to grow faster than untreated cells, whereas cholesterol depletion decreased the tether growth rate relative to untreated cells. In comparing the tether growth rate velocity for all events (solid bar, Fig. 7 A) with the measured lifetimes of those events (Fig. 6 E), it was clear that increases in the tether growth velocity caused prolongation in the PSGL-1/P-selectin lifetime due to decreased forces on the bonds (Fig. 7 B). It would be highly unlikely that the lifetime of the tether attachment to the bead would alter the tether growth rate velocity, a process that is dictated by lipid flow from the neutrophil.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
A central issue addressed by this technique was whether tether growth shields the bond from force loading, and to what extent. The data in Fig. 2 D comparing the average lifetime of populations with and without tethers at each shear rate suggests, but does not prove, that tether growth results in longer lived bonds. Long-lived bonds allow more time for tether growth and consequently longer tethers. Short-lived bonds provide less time for tether growth and thus result in shorter tethers. We have used five different pharmacological treatments to alter membrane and cell mechanics. The most extreme treatment was formaldehyde fixation, which obliterated tether formation and reduced the bond lifetime to a level identical to that found for events that did not form tethers using unfixed neutrophils (Fig. 6 D). Since the collision efficiency was not reduced with formaldehyde, this reduction in the overall lifetime of all events by formaldehyde was not likely due to a defect in the bond or the level of PSGL-1 on the fixed neutrophil. This result extends the prior understanding of formaldehyde effects (3) by simultaneously measuring both tether length and bond life for each adhesion event. A less extreme method of reducing tether growth was achieved through the use of MßCD to extract cholesterol out of the neutrophil membrane. MßCD reduced the fraction of cells forming tethers and reduced the tether length and the tether growth velocity, but not to the extent achieved with formaldehyde. Consistent with tether length shielding the bond from loading, MßCD reduced the overall lifetime of bonds. Since the collision efficiency was not reduced with MßCD, this reduction in the overall lifetime of all events by MßCD was not likely due to a defect in the bond or the level of PSGL-1 on the surface of cholesterol-depleted neutrophils. MßCD has previously been reported to increase leukocyte rolling velocity (33). On a local scale of the plasmalemma at the base of the tether, neutrophils depleted of cholesterol exhibit the rigidifying effects that have previously been demonstrated on a global whole-cell scale using micropipette aspiration (33).
Conversely, agents that caused longer tethers to form (cytochalasin D, latrunculin, and jasplakinolide) also prolonged the average lifetime of the adhesive interaction (Fig. 6 D). Latrunculin softens the entire cell body and causes an unusual tear-shape morphology in the bead collision assay, which may lead to cell flattening and reduction of force loading at a given shear rate. In contrast, jasplakinolide hardens the overall cell but causes plasmalemma blebbing, which in turn facilitates membrane tether formation (an increase in f) because lipid is already detached from the cytoskeleton and available for flow into the tether. Jasplakinolide serves as an example where whole cell mechanics, i.e., cell hardening as detected by micropipette aspiration (37), does not predict local plasmalemma dynamics during PSGL-1-mediated bonding. In the study of Sheikh et al. (37), it was only the neutrophil core that would resist entry into the pipette; membrane blebs that had formed on the surface of the cells would easily enter. We also observed membrane blebs on the surface of neutrophils treated with this reagent (not shown), which indicated that the lipid bilayer separated from the neutrophil at these locations (51). This separation of bilayer and cytoskeleton reduces the minimum force necessary to pull a tether by eliminating the membrane-cytoskeletal adhesion portion of tether adhesion energy (51), thus rendering the cell locally as less rigid than untreated cells.
Micropipette studies looking at global effects on rigidity due to cytochalasin D (32–34) through measurement of the cortical viscosity previously pointed to depolymerization as a mechanism of neutrophil softening. Other studies looking specifically at tether growth after treatment with either cytochalasin D (23) or latrunculin A (36) have shown that the membrane slip between the lipid bilayer and cytoskeleton is a major contributor to energy dissipation during tether formation. In our study, on average, membrane tethers generated by cells treated with cytochalasin D or latrunculin A had faster tether growth rates (Fig. 7) than untreated cells, and consequently longer bond lifetimes than untreated cells (Fig. 6). These results point to a greater ease of lipid flow into the tether during formation under venous flow conditions. Neutrophils that were fixed or treated with MßCD had opposite effects to those seen with depolymerizing agents, and formed tethers with significantly reduced growth rates compared to untreated cells and consequently had shorter lifetimes (Figs. 6 and 7). A force as low as 25.6 pN at a wall shear rate of 25 s–1 was detected to cause membrane tether pulling. This force is 50% of that predicted by micropipette studies, where micromanipulation of the cell may lead to low level cellular activation and consequent slight cortical stiffening.
Since this study focused on both on-rate and off-rate processes, a high saturating level of P-selectin IgG chimera was used in these studies to allow a reliable determination of the collision efficiency as a function of shear rate (Fig. 2 A). Since the contact area is so small at the tip of the tether and the interaction time so short during the actual neutrophil-bead collision (Fig. 2 E), the method resulted in events dominated by single bond interactions (estimated at >65%) and provided Bell parameters that were consistent with those obtained from rolling experiments (curve 1b in Fig. 5). A subset of all events, those without tethers where forces are most uniform and accurately estimated at each w (curve 1a in Fig. 4 B), provided Bell parameters consistent with single-bond MC and prior AFM and rolling beads experiments (curves 2–8 in Fig. 5). Rebonding processes (kon in the MC) did not appear significant in the data sets and this suggests that the contact area at the tip of the tether, when pulled, becomes exceedingly small, possibly approaching the dimensions of an AFM cantilever tip. Future work employing sparsely-coated beads that is focused solely on off-processes (where is not determined) should allow the event distribution to be dominated by single bonds.
Catch-bonding has been detected by AFM and in flow chambers for the PSGL-1/P-selectin bond at force loadings in the 5–20 pN range (7). Catch-bonding may be one of several underlying mechanisms to explain the unexpected hydrodynamic-thresholding observed for neutrophil rolling on selectins (35). Platelet translocation on vWF A1 domain also displays thresholding at 1 dyn/cm2 (44) with r > 0 at Fbond > 36 pN. Neutrophils rolling on selectin-coated surfaces display biphasic force-enhanced adhesion at low shears, and force-decreased adhesion at shear rates above a certain maximum in shear (52). The direct observation of a decrease in off-rate at low shear rates that switches to an increase at higher shear rates provided the first experimental observation of catch-bonding in the P-selectin/PSGL-1 system (7). Catch-bonding has also been reported for L-selectin/PSGL-1 at low forces (8,53). In the current study, we did not look at the force regime below the catch-bond threshold for P-selectin/PSGL-1 (FB < 10–20 pN, w < 15 s–1), and thus our results are for those interactions in the slip-bond regime only.
The neutrophil-bead collision assay may prove useful in future studies, including: determination of the on-rate as a function of shear rate and ligand density; thresholding studies at low flows where contaminating flow fluctuations would have to be coincident with the bead collision and thus hydrodynamic fluctuation effects are better filtered out of the assay; studies of activated neutrophils to probe VCAM-1 or ICAM-1 binding lifetimes; or the analysis of patient neutrophils where neutrophil membrane mechanics and inflammation are potential cofactors in vaso-occlusive phenomena.
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
65–75% of the tethers having only a single bond present.
|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
This work was supported by National Institutes of Health grant No. HL56621 (to S.L.D.). K.E.E. is a recipient of an Ashton Fellowship.
Submitted on May 10, 2005; accepted for publication August 3, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONAPPENDIXACKNOWLEDGEMENTSREFERENCES |
|---|
2. Chen, S., and T. A. Springer. 2001. Selectin receptor-ligand bonds: Formation limited by shear rate and dissociation governed by the Bell model. Proc. Natl. Acad. Sci. USA. 98:950–955.
3. Park, E. Y. H., M. J. Smith, E. S. Stropp, K. R. Snapp, J. A. DiVietro, W. F. Walker, D. W. Schmidtke, S. L. Diamond, and M. B. Lawrence. 2002. Comparison of PSGL-1 microbead and neutrophil rolling: microvillus elongation stabilizes P-selectin bond clusters. Biophys. J. 82:1835–1847.
4. Ramachandran, V., T. Yago, T. K. Epperson, M. M. A. Kobzdej, M. U. Nollert, R. D. Cummings, C. Zhu, and R. P. McEver. 2001. Dimerization of a selectin and its ligand stabilizes cell rolling and enhances tether strength in shear flow. Proc. Natl. Acad. Sci. USA. 98:10166–10171.
5. Smith, M. J., E. L. Berg, and M. B. Lawrence. 1999. A direct comparison of selectin-mediated transient, adhesive events using high temporal resolution. Biophys. J. 77:3371–3383.
6. Fritz, J., A. G. Katopodis, F. Kolbinger, and D. Anselmetti. 1998. Force-mediated kinetics of single P-selectin/ligand complexes observed by atomic force microscopy. Proc. Natl. Acad. Sci. USA. 95:12283–12288.
7. Marshall, B. T., M. Long, J. W. Piper, T. Yago, R. P. McEver, and C. Zhu. 2003. Direct observation of catch bonds involving cell-adhesion molecules. Nature. 423:190–193.[CrossRef][Medline]
8. Sarangapani, K. K., T. Yago, A. G. Klopocki, M. B. Lawrence, C. B. Fieger, S. D. Rosen, R. P. McEver, and C. Zhu. 2004. Low force decelerates L-selectin dissociation from P-selectin glycoprotein ligand-1 and endoglycan. J. Biol. Chem. 279:2291–2298.
9. Hanley, W., O. McCarty, S. Jadhav, Y. Tseng, D. Wirtz, and K. Konstantopoulos. 2003. Single molecule characterization of P-selectin/ligand binding. J. Biol. Chem. 21:10556–10561.
10. Hanley, W., D. Wirtz, and K. Konstantopoulos. 2004. Distinct kinetic and mechanical properties govern selectin-leukocyte interactions. J. Cell Sci. 117:2503–2511.
11. Rinko, L. J., M. B. Lawrence, and W. H. Guilford. 2004. The molecular mechanics of P- and L-selectin lectin domains binding to PSGL-1. Biophys. J. 86:544–554.
12. Evans, E., A. Leung, D. Hammer, and S. Simon. 2001. Chemically distinct transition states govern rapid dissociation of single L-selectin bonds under force. Proc. Natl. Acad. Sci. USA. 98:3784–3789.
13. Evans, E., A. Leung, V. Heinrich, and C. Zhu. 2004. Mechanical switching and coupling between two dissociation pathways in a P-selectin adhesion bond. Proc. Natl. Acad. Sci. USA. 101:11281–11286.
14. Heinrich, V., A. Leung, and E. Evans. 2005. Nano- to microscale dynamics of P-selectin detachment from leukocyte interfaces. II. Tether flow terminated by P-selectin dissociation from PSGL-1. Biophys. J. 88:2299–2308.
15. Mehta, P., R. D. Cummings, and R. P. McEver. 1998. Affinity and kinetic analysis of P-selectin binding to P-selectin glycoprotein ligand-1. J. Biol. Chem. 273:32506–32513.
16. Marshall, B. T., K. K. Sarangapani, J. Lou, R. P. McEver, and C. Zhu. 2005. Force history dependence of receptor-ligand dissociation. Biophys. J. 88:1458–1466.
17. Shao, J. Y., H. P. Ting-Beall, and R. M. Hochmuth. 1998. Static and dynamic lengths of neutrophil microvilli. Proc. Natl. Acad. Sci. USA. 95:6797–6802.
18. Schmidtke, D. W., and S. L. Diamond. 2000. Direct observation of membrane tethers formed during neutrophil attachment to platelets or P-selectin under physiological flow. J. Cell Biol. 149:719–729.
19. Evans, E., V. Heinrich, A. Leung, and K. Kinoshita. 2005. Nano- to microscale dynamics of P-selectin detachment from leukocyte interfaces. I. Membrane separation from the cytoskeleton. Biophys. J. 88:2288–2298.
20. Zhu, C., M. Long, S. E. Chesla, and P. Bongrand. 2002. Measuring receptor/ligand interaction at the single-bond level: experimental and interpretive issues. Ann. Biomed. Eng. 30:305–314.[CrossRef][Medline]
21. Puri, K. D., E. B. Finger, and T. A. Springer. 1997. The faster rolling kinetics of L-selectin than E-selectin and P-selectin rolling at comparable binding strength. J. Immunol. 158:405–413.[Abstract]
22. Shao, J. Y., and R. M. Hochmuth. 1996. Micropipette suction for measuring picoNewton forces of adhesion and tether formation from neutrophil membranes. Biophys. J. 71:2892–2901.
23. Shao, J. Y., and J. Xu. 2002. A modified micropipette aspiration technique and its application to tether formation from human neutrophils. J. Biomech. Eng. Trans. ASME. 124:388–396.[CrossRef][Medline]
24. Girdhar, G., and J. Y. Shao. 2004. Membrane tether extraction form human umbilical vein endothelial cells and its implication in leukocyte rolling. Biophys. J. 87:3561–3568.
25. Spector, I., F. Braet, N. R. Shochet, and M. R. Bubb. 1999. New anti-actin drugs in the study of the organization and function of the actin cytoskeleton. Microsc. Res. Tech. 47:18–37.[CrossRef][Medline]
26. Torres, M., and T. D. Coates. 1999. Function of the cytoskeleton in human neutrophils and methods for evaluation. J. Immunol. Methods. 232:89–109.[CrossRef][Medline]
27. McCarty, O. J. T., N. Tien, B. S. Bochner, and K. Konstantopoulos. 2003. Exogenous eosinophil activation converts PSGL-1-dependent binding to CD18-dependent stable adhesion to platelets in shear flow. Am. J. Physiol. Cell Physiol. 284:C1223–C1234.
28. Bubb, M. R., A. M. J. Senderowicz, E. A. Sausville, K. L. K. Duncan, and E. D. Korn. 1994. Jasplakinolide, a cytotoxic natural product, induces actin polymerization and competitively inhibits the binding of phalloidin to F-actin. J. Biol. Chem. 269:14869–14871.
29. Bubb, M. R., I. Spector, B. B. Beyer, and K. M. Fosen. 2000. Effects of jasplakinolide on the kinetics of actin polymerization. J. Biol. Chem. 275:5163–5170.
30. Senderowicz, A. M., G. Kaurm, E. Sainz, C. Laing, W. D. Inman, J. Rodriguez, P. Crews, L. Malspeis, M. R. Grever, and E. A. Sausville. 1995. Jasplakinolide's inhibition of the growth of prostate carcinoma cells in vitro with disruption of the actin cytoskeleton. J. Nat. Can. Inst. 87:46–51.[CrossRef]
31. Tsai, M. A., R. S. Frank, and R. E. Waugh. 1994. Passive mechanical behavior of human neutrophils: effect of cytochalasin B. Biophys. J. 66:2166–2172.
32. Ting-Beall, H. P., A. S. Lee, and R. M. Hochmuth. 1995. Effect of cytochalasin D on the mechanical properties and morphology of passive human neutrophils. Ann. Biomed. Eng. 23:666–671.[Medline]
33. Yago, T., A. Leppanen, H. Qiu, W. D. Marcus, M. U. Nollert, C. Zhu, R. D. Cummings, and R. P. McEver. 2002. Distinct molecular and cellular contributions to stabilizing selectin-mediated rolling under flow. J. Cell Biol. 4:787–799.
34. Sheikh, S., and G. B. Nash. 1998. Treatment of neutrophils with cytochalasins coverts rolling to stationary adhesion on P-selectin. J. Cell. Physiol. 174:206–216.[CrossRef][Medline]
(the nearly constant rupture force) calculated from the applied pulling rate (
pull in µm/s) by f
and r = 0.10 nm. Two-thousand simulations were averaged at each data point shown.