Dynamics of bionanoparticle targeting in mixtures of human tumour cells by validated population balance modelling
Abstract
Preferential aggregation of bionanoparticles to different human suspension cell lines preceding cellular uptake by endocytosis was investigated by a 3-dimensional discrete population balance model. Property space was reduced by excluding intercellular aggregation in accordance with experimental evidence and by establishing an adaptive grid with full property resolution in areas of high particle or cell densities. The aggregation rates are based on deterministic models from colloidal sciences, adapted to include biologically specific aggregation in addition to unspecific aggregation. While the collision frequency is described by the Smoluchowski kernel for Brownian motion, the collision efficiency is modelled by a kernel considering inhomogeneous surface patches. The patch encounter probability and the interaction potential energies embedded in the stability ratio gain dominant influence on the aggregation kinetics.
Given a cell specific receptor density as input parameter, the preferential aggregation of bionanoparticles to U-937 cells in mixture with KARPAS-299 cells was successfully simulated. Fast specific biological aggregation is superimposed on slow unspecific aggregation. The distribution of adsorbed bionanoparticles per cell initially broadens considerably until all receptors are saturated, then narrows down to a quasi-equilibrium state where only unspecific aggregation persists. A low probability of bionanoparticle-receptor encounters, based on considerations concerning cell and receptor geometries, causes a rate limitation of biologically specific aggregation despite the high bionanoparticle to cell concentration ratio and purely attractive interaction potentials.
Validation experiments with IgG1, CD13 and CD33 antibodies were performed by