Uniform Linear Poly(ethylenimine) via Poly(oxazoline)s - Synthesis, Mechanism of Toxicity and the Role of the Free Polymer during Transfection

Linear poly(ethylenimine) is the current “gold standard” gene transfection agent. It is synthesised via the cationic ring opening polymerisation of 2-ethyl-2-oxazoline followed by the acid catalysed hydrolysis to the hydrochloride salt. However the synthesis of the parent polymer produces a broad distribution of molecular weights. Ideally the polymer should have a narrow molecular weight distribution - it should be “uniform”. By exploring the possible side reactions it is possible to minimise them and optimise the polymerisation of 2-oxazolines. By using a non-interfering solvent (chlorobenzene) in place of acetonitrile at a low temperature (40-45°C) the majority of side-reactions cease. Increasing the substitution on the side-chain increased steric hindrance of both the β and the 2-positions, but increased crystallinity complicated the polymerisation. When an extremely nucleofugic counterion (triflate) was used the propagation rate increased dramatically, despite no apparent change in the equilibrium between the ionic propagating species and inactive ester. The nature of propagation appears to have changed. The l-PEI derived from uniform poly(oxazoline)s together with a broad 22 kDa l-PEI simulating the commercial polymer “in-vivo jetPEITM” and a range of poly(l-lysine)s were tested on lung adenocarcinoma cells (A549) revealing a strong correlation between the molecular weight of the polymer and the degree of cell membrane disruption via LDH release. Mitochondrial activity assays carried out with short (4 hr) incubation times did not correlate with LDH release except in the case of immediate cellular destruction. In fact values much greater then 100% viability were typically observed, indicating that mitochondrial activity had actually increased. However when incubation time was increased to 24 or 48 hrs the increase in mitochondrial activity was less pronounced and the correlation with cell membrane damage was restored. This suggests that damaged cells were increasing their activity to maintain homeostatis, which obfuscates cellular toxicity in the normal MTT protocol. With the destruction of cell membranes strongly implicated in the toxicity of l-PEI and PLL (and assumidly other polycations) it was hypothesised that the mechanism of cell membrane destruction was acid catalysed hydrolysis of the ester bonds of phospholipids. This would lead to changes in cell membrane curvature resulting in the opening of pores/ holes in the cell surface and, eventually would simulate apoptotic cell death despite a necrotic origin since apoptotic “squibs” are formed by hydrolysis of phospholipid esters by an enzyme. Experiments on DOPC liposomes incubated with l-PEI revealed the concentration of DOPC falling over time with MOPC forming transiently before further hydrolysis occurred. The rate of hydrolysis is implied to be related to the MW of the polymer. Finally the observations of hydrolysis lead to a thought experiment and a new hypothesis of the underlying mechanism of polymeric transfection.Imperial Users Onl