Studies on the mechanism of interaction of a bioresponsive endosomolytic polyamidoamine with interfaces. 1. Micelles as model surfaces.

Polymers are appealing as pH-responsive elements of multicomponent systems designed to promote cytosolic delivery of macromolecular drugs (including proteins and genes), but so far the delivery efficiency achieved has been relatively modest. Therefore, the aim of this study was to apply several physicochemical techniques that are well established in the colloid field (surface tension measurements, small-angle neutron scattering (SANS), and electron paramagnetic resonance (EPR)) to probe the mechanism of endosomolytic polymer-surface interaction over the pH range 7.4 to 5.5 using the poly(amidoamine) (PAA) ISA23 x HCl and a series of "model" micelle surfaces. These micellar models were chosen to represent increasing complexity from simple, single surfactant sodium dodecylsulfate (SDS) micelles, surfactant mixtures containing bulky malono-bis-N-methylglucamide headgroups, or highly extended ethylene oxide headgroups. Spherical micelles composed of 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine (lyso-PC) were also used. Changes in the onset of micellization, micelle surface fluidity, and in selected cases, the overall micelle shape and size were all quantified as a function of pH in the presence and absence of ISA23 x HCl. This amphoteric PAA is negatively charged at pH 7.4 and becomes gradually more protonated on exposure to lower pH values representative of the endosomal-lysosomal pathway. As expected, the strength of polymer interaction with anionic micelles increased with a decrease in pH, while for cationic micelles the opposite was observed. Addition of bulky, nonionic surfactant headgroups led to weaker interactions. The observations from surface tension and SANS studies showed a complex pattern of interaction with both an electrostatic and hydrophobic component. Using EPR it was confirmed that ISA23 x HCl perturbed the micelle palisade layer leading to a decrease in fluidity of the interface with a lower degree of headgroup hydration, and a significant change in micelle morphology. Surprisingly, there was no interaction between ISA23 x HCl and globular micelles formed from lyso-PC (a more biologically relevant model), and this suggests that the PAA structure could be better optimized to promote rapid interaction with endosomal membranes at the physiologically relevant pH 6.5.