Predicting different adhesive regimens of circulating particles at blood capillary walls
For researchers designing vascular-targeted drug delivery particles, this model provides a tool to predict adhesion behavior without experimental trial-and-error.
This study presents a computational model combining Lattice Boltzmann and Immersed Boundary methods to predict the adhesion dynamics of circulating particles at blood vessel walls. The model accurately predicts rolling velocities of tumor cells and identifies four adhesion regimens (not adhering, rolling, sliding, firmly adhering) based on particle geometry, ligand density, bond strength, and Reynolds number.
A fundamental step in the rational design of vascular targeted particles is the firm adhesion at the blood vessel walls. Here, a combined Lattice Boltzmann Immersed Boundary model is presented for predicting the near wall dynamics of circulating particles. A moving least squares algorithm is used to reconstruct the forcing term accounting for the immersed particle, whereas ligand-receptor binding at the particle- wall interface is described via forward and reverse probability distributions. First, it is demonstrated that the model predicts with good accuracy the rolling velocity of tumor cells over an endothelial layer in a microfluidic channel. Then, particle-wall interactions are systematically analyzed in terms of particle geometries (circular, elliptical with aspect ratios 2 and 3), surface ligand densities (0.3, 0.5, 0.7 and 0.9), ligand-receptor bond strengths (1 and 2), and Reynolds numbers (Re = 0.01, 0.1 and 1.0). Depending on these conditions, four different particle-wall interaction regimens are identified, namely not adhering, rolling, sliding and firmly adhering particles. The proposed computational strategy can be efficiently used for predicting the near wall dynamics of particles with arbitrary geometries and surface properties and represents a fundamental tool in the rational design of particles for the specific delivery of therapeutic and imaging agents.