The charge effect of insulin PLGA nanoparticles on cell uptake in vitro and bioavailability in vivo

E. Czuba1, S. Sigrist1, E. Marchioni3, Y. Frere4, E. Maillard1 , M. Pinget2
1 UMR DIATHEC, EA 7294, Centre Européen d’Etude du Diabète, Université de Strasbourg (UdS), Boulevard René Leriche 67200 Strasbourg, France ;
3 Institut pluridisciplinaire Hubert Curien, faculté de pharmacie, Illkirch-Graffenstaden, France;  
4 Institut Charles Sadron (UPR 22 CNRS) 23 rue du Loess BP 84047 67034 STRASBOURG Cedex 2, France

INTRODUCTION: Last years, many progresses have been realized in insulin administration for type 1 diabetics in order to decrease painfulness for patient. However, trauma and pain associated with injections are real and causes a discomfort for patient. That’s why oral route of insulin administration have been studied. But most of peptide drugs like insulin are administrated using parenteral route due to the low bioavailability of the oral route and the degradation of the peptide through the gastrointestinal tract. In order to increase the bioavailability of peptide, we have patented (patent CeeD: FR 0304976) a unique technology based on the double encapsulation approach (combination of nanoparticles (NPs) with a gastroresistant capsule). It has observed that Physicochemical factors such as particle size, nanoparticle stability and surface charge may affect nanoparticle absorption. The aim of our work is to modify the surface of nanoparticles to increase their bioefficiency and to validate the modification in vitro and in vivo.
MATERIAL AND METHODS: Particles were prepared in presence or not of polyvinyl alcohol (PVA), with sodium dodecyl sulphate (SDS) or coated with chitosan chloride in order to play on NPs charge. In vitro internalization was tested on epithelial caco-2 by flow cytometry. NPs were then administrated thanks to a pharmaceutical complex vector in vivo by oral route (20UI and 50UI) in diabetic rats.
RESULTS: The mean size of +PVA nanoparticles measured by dynamic light scattering is 203 ± 9 nm and a neutral ζ-potential -1 ± 1. Removal of PVA (-PVA) decreased size (154 ± 24 nm) and were negatively charged (-26± 2 mV). Incorporation of SDS to these nanoparticles presented respectively sizes of 151 ± 4 nm and negative ζ-potential of -42±2 mV. With a coating agent like chitosan, size of nanoparticles increased to 162 ± 29 and 184± 11 for chitosan and SDS-chitosan NPs respectively. For ζ-potential, we obtained high positive charge for these conditions (56 ± 3 and 40 ± 3). for all formulations, encapsulation efficacy (EE) was near to 100% except for chitosan-NPs with a huge decrease to 34%.
The highest associated fluorescence was quantified on cells incubated with non-encapsulated insulin. For insulin NPs, highest associated fluorescence was quantified on cells incubated with particles contained SDS.  SDS and SDS-chitosan NPs, showed an increase of associated fluorescence compare to +PVA and –PVA NPs which are the less associated particles in both models. Chitosan-SDS showed a low reproducibility of chitosan-SDS particles but with significant increase of fluorescence compare to other group.  But without SDS, we observed for chitosan a non significate increase of uptake compare to +PVA NPs on caco-2 model. 
Administration by oral route of gastro-resistant capsule contained SDS-NPs 20 and 50UI) reduced glycaemia faster than empty vehicles and vehicles with standard NPs from 12 hours (20UI) and 14 hours  (50UI). In vitro effect on cell-uptake with negatively charged with SDS NPs were confirmed. 
CONCLUSION: Mucoadhesive particles (chitosan) formulation is definitely not the good approach to improve the bioavailability of encapsulated insulin contrary to negative NPs are the most efficient both in vitro and in vivo, and represent a promising formulation for oral insulin delivery which could be developed.