Ammonia‐scavenging transmembrane pH‐gradient poly(styrene)‐b‐poly(ethylene oxide) polymersomes are investigated for the oral treatment and diagnosis of hyperammonemia, a condition associated with serious neurologic complications in patients with liver disease as well as in infants with urea cycle disorders. While these polymersomes are highly stable in simulated intestinal fluids at extreme bile salt and osmolality conditions, they unexpectedly do not reduce plasmatic ammonia levels in cirrhotic rats after oral dosing. Incubation in dietary fiber hydrogels mimicking the colonic environment suggests that the vesicles are probably destabilized during the dehydration of the intestinal chyme. The findings question the relevance of commonly used simulated intestinal fluids for studying vesicular stability. With the encapsulation of a pH‐sensitive dye in the polymersome core, the local pH increase upon ammonia influx could be exploited to assess the ammonia concentration in the plasma of healthy and cirrhotic rats as well as in other fluids. Due to its high sensitivity and selectivity, this polymersome‐based assay could prove useful in the monitoring of hyperammonemic patients and in other applications such as drug screening tests.

The molecular weight distributions of poly(lactic acid) produced by ring-opening polymerization of L,L-lactide in bulk melt are measured and compared with the ones predicted using a kinetic model accounting for reversible catalyst activation, reversible propagation, reversible chain transfer to co-catalyst and inter-molecular transesterification. The same values of the model parameters as evaluated in previous works are used without any adjustment, i.e. the model is used in fully predictive way. In order to calculate the complete molecular weight distribution, the model equations are solved through two different numerical methods, "direct integration" of the population balances at all values of chain length, and "fractionated moments", where the chains are artificially classified into two different categories depending upon the experienced reaction steps. The accuracy of the molecular weight distributions calculated in the latter case is evaluated by comparison with those computed by solving the model equations with the "direct integration" method. It is found that the "fractionated moments" method provides enough accuracy and much smaller computational effort, thus representing an optimal tool for most modeling applications. Finally, the model predictions are compared with the experimental molecular weight distributions measured experimentally in bulk melt at 130°C and various initial concentrations of catalyst and co-catalyst. The generally good agreement verified between model and experiment after correcting for peak broadening, represents a convincing confirmation of the model reliability.

COMMUNICATION 1700584 (1 of 8)