Valency-affinity mapping of multivalent liposomes for tunable target cell discrimination.

Multivalency can drive high-avidity binding of ligand-functionalized nanoparticles to cells with high target receptor expression, but it can also contribute to off-target binding to low-expression non-target cells. We explored how ligand affinity and liposome valency shape the resulting binding performance index (BPI), defined as the product of the proportion of liposome-bound target cells and that of non-bound non-target cells. Designed ankyrin repeat proteins (DARPins) spanning a wide range of HER2-binding affinities were tethered onto PEGylated liposomes at varying concentrations. BPI was initially evaluated in mixed-cell suspensions of HER2^high SKBR3 (target) cells and HER2^low T47D (non-target) cells, with the highest BPI (> 0.8) observed for high-valency liposomes displaying high-affinity DARPins. To further map the BPI landscape, we measured particle binding to HEK293T cells transiently transfected with HER2-EGFP, leveraging the inherent transfection heterogeneity to generate continuous binding response curves as a function of HER2 expression. HER2^high (target) and HER2^low (non-target) populations were defined by a HER2 threshold, which was varied across the range of HER2 expression to determine maximum BPI values (> 0.85) and corresponding HER2 threshold optima (HER2_OPT). BPI generally tracks with traditional binding selectivity, but BPI is more sensitive to off-target effects or poor on-target binding and thus may better assess particle performance. We further demonstrate that HER2_OPT can be rationally increased or decreased by adjusting DARPin valency and affinity (separately or synergistically) to lower or higher values, respectively. The approach outlined here enables rapid testing and optimization of ligand parameters for nanoparticle binding toward a given therapeutic target.