The role of pH changes during remote loading of setmelanotide in PLGA microspheres

The remote loading method enables aqueous, post-fabrication encapsulation of cationic peptides into PLGA microspheres via ion pairing between the deprotonated polymer carboxylate end groups and the basic amino acid side chains and/or alpha amino group of the peptide. This encapsulation method can produce peptide loaded PLGA microspheres with competitive in vitro/in vivo performance to successful commercial PLGA-peptide products. However, this ion pairing loading technique also can be pH-dependent owing to the exchange of polymer protons with the peptide and the resulting pH effects on the polymer (and potentially peptide). Here, we investigated how pH dynamics govern remote loading performance for the model peptide setmelanotide, a permanent di-cation at neutral pH, by varying the buffer strength of the peptide loading solution (0.05 - 0.30 M HEPES) and microsphere attributes (porosity and PLGA molecular weight). The drug loading, first-week in vitro burst release, and microsphere surface morphology were examined. Increasing buffer concentration attenuated the pH drop during remote loading, preventing premature termination of ion pairing and thereby increasing encapsulation efficiency; lower molecular weight PLGA provided significantly higher loading and encapsulation efficiency across buffers (∼10% w/w with near-complete encapsulation with > 0.1 M buffer concentration). The largest pH decrease at 0.05 M HEPES (∼2 units) was associated with a substantially reduced initial burst (∼5% in the first week) irrespective of microsphere initial surface porosity, whereas the smallest pH change at 0.30 M buffer concentration yielded the highest burst (∼20% in the first week), consistent with observed surface polymer healing changes before and after remote loading. Guided by the findings, a low pH rinse (0.2 M MES buffer, pH ∼6.0) applied to a high-burst formulation was shown to reduce the initial burst release in vitro in the first 24 h from ∼7% to ∼3%. Overall, this work identifies process and material parameters that can be used to optimize pH effects of remote loading to minimize burst release while maximizing peptide loading via the remote loading method.