Nano-to-micro scale dynamics of P-selectin detachment from leukocyte interfaces. III. Numerical simulation of tethering under flow

Abstract

Transient capture of cells or model microspheres from flow over substrates sparsely coated with adhesive ligands has provided significant insight into the unbinding kinetics of leukocyte:endothelium adhesion complexes under external force. Whenever a cell is stopped by a point attachment, the full hydrodynamic load is applied to the adhesion site within an exceptionally short time-less than the reciprocal of the hydrodynamic shear rate (e.g., typically <0.01 s). The decay in numbers of cells or beads that remain attached to a surface has been used as a measure of the kinetics of molecular bond dissociation under constant force, revealing a modest increase in detachment rate at growing applied shear stresses. On the other hand, when detached under steady ramps of force with mechanical probes (e.g., the atomic force microscope and biomembrane force probe), P-selectin:PSGL-1 adhesion bonds break at rates that increase enormously under rising force, yielding 100-fold faster off rates at force levels comparable to high shear. The comparatively weak effect of force on tether survival in flow chamber experiments could be explained by a possible partition of the load amongst several bonds. However, a comprehensive understanding of the difference in kinetic behavior requires us to also inspect other factors affecting the dynamics of attachment-force buildup, such as the interfacial compliance of all linkages supporting the adhesion complex. Here, combining the mechanical properties of the leukocyte interface measured in probe tests with single-bond kinetics and the kinetics of cytoskeletal dissociation, we show that for the leukocyte adhesion complex P-selectin:PSGL-1, a detailed adhesive dynamics simulation accurately reproduces the tethering behavior of cells observed in flow chambers. Surprisingly, a mixture of 10% single bonds and 90% dimeric bonds is sufficient to fully match the data of the P-selectin:PSGL-1 experiments, with the calculated decay in fraction of attached cells still appearing exponential.

FIGURE 1

(A) Side-view outline of a tethered cell at a wall shear stress of 1.0 dyn/cm², initially and after 1 s. The length and orientation of the tether linking the cell to the wall is plotted at intervals of 1/30 s. (B) Closeup of the bond region.

FIGURE 2

Time course of the bond force (A), tether length (B), and cell deformation (C) during individual cell tethering events at a wall shear stress of 0.7 dyn/cm². In this case, tethers are mediated by single P-selectin:PSGL-1 bonds. Histograms of cytoskeletal and P-selectin:PSGL-1 dissociation events are given in panel D. Nine-hundred individual adhesion events are overlaid in panels A–C.

FIGURE 3

Time course of the bond force (A), tether length (B), and cell deformation (C) during individual cell tethering events at a wall shear stress of 0.7 dyn/cm². In this case, tethers are mediated by dimeric P-selectin:PSGL-1 bonds that share force equally. Histograms of cytoskeletal and P-selectin:PSGL-1 dissociation events are given in panel D. Four-hundred individual adhesion events are overlaid in panels A–C.

FIGURE 4

Pause-time distributions used to calculate the “apparent” bond dissociation constants of neutrophils interacting with a P-selectin substrate. Simulated data for either single-bond tethers (A) or double-bond tethers (B) at different wall shear stresses are presented in the same form as in the experimental study of Park et al. (2002).

FIGURE 5

Dependence of the apparent off rate of the P-selectin:PSGL-1 bond on wall shear stress, as determined from the interpretation of real or simulated tethering experiments. Squares denote the experimental data of Park et al. (2002), and are compared with simulation results for either single or double bonds. Note that a mixture of single and double bonds, with a double-bond fraction of 90% agrees well with the experiments of Park et al. (2002). Also plotted for comparison is a relative weighting of 60% single bonds, 30% double bonds, and 10% triple bonds, corresponding to a Poisson distribution in bond number.