While a number of PEGylated proteins have been studied for injectable applications and reviewers have used this data to speculate possible oral delivery improvements, a detailed investigation of PEGylated insulin for oral delivery and the development of an optimized pH-sensitive carrier for PEGylated insulin conjugates had yet to be accomplished. In order to proceed with oral delivery study, improvements in yield, with respect to previous PEGylation methods were necessary to enable the completion of high throughput drug delivery studies. Subsequently, a reaction scheme for the covalent attachment of PEG to insulin using nitrophenyl carbonate-PEG was developed. It was demonstrated that this reaction occurred at a 1:1 ratio and was site specific at the B29Lys position. A P(MAA-g-EG) hydrogel carrier was developed to optimize loading and release behavior for PEGylated insulin. It was demonstrated that the density and length of polymer grafts affected both loading and release behavior of PEGylated insulin. The best performing grafted polymers had a 3:1 methacrylic acid: ethylene glycol (MAA:EG) ratio and achieved loading efficiencies from 96% to nearly 100%. With respect to release, polymer particles containing fewer, but longer grafts shown to release faster than polymers with shorter grafts with the same MAA:EG ratio. Finally, the effects of PEGylation on intestinal absorption was investigated using an intestinal epithelial model as well as a rat model. It was demonstrated that PEGylated insulin in the presence of P(MAA-g-EG) microparticles did not significantly alter the tight junctions over unmodified insulin. However, the conjugate permeabilities across the membrane were reduced. The pharmacological availability (PA) was then verified by injecting the insulin conjugates subcutaneously in fasted Sprague-Dawley rats. It was determined that PEG 1000 insulin (1KPI) had a PA roughly equivalent to insulin, while it was reduced by 59% for 2KPI and by 81% for 5KPI. The effectiveness of utilizing PEGylated insulin as an oral drug delivery candidate was evaluated with a closed loop intestinal study, in which PEGylated insulin or insulin in solution was delivered directly to the jejunum. It was shown that 1KPI and insulin performed identically; with a pharmacological availability of 0.56%. 2KPI, however improved the pharmacological availability of insulin by 2.8 times. These results demonstrate that PEGylation holds promise for improving the oral delivery of proteins. / text

Les S-nitrosothiols (RSNO) comme le S-nitrosoglutathion (GSNO) sont des donneurs de monoxyde d’azote (NO) prometteurs pour le traitement des maladies cardiovasculaires. Cependant, ce sont des candidats médicaments peu stables. Précédemment, des nanoparticules chargées en GSNO (GSNO-NP) ont été incluses dans une matrice d’alginate/chitosan. Les particules composites ainsi produites avaient une bonne encapsulation et une libération prolongée de GSNO. De plus, leur administration orale à des rats produisait un stock de NO au niveau de la paroi de l’aorte. Elles avaient cependant plusieurs limitations : préparation et caractérisation longues, manque de stabilité et de reproductibilité. Ce travail avait donc trois objectifs : (1) déterminer le mécanisme d’absorption intestinale des RSNO non formulés ; (2) évaluer la capacité des RSNO non formulés à créer un stock vasculaire de NO ; 3) optimiser la formulation de GSNO. Nous avons montré, dans un modèle in vitro de barrière intestinale, que la perméabilité intestinale de GSNO, S-nitroso-N-acétylcystéine (NACNO) et S-nitroso-N-acetylpénicillamine (SNAP) se fait par un mécanisme passif, principalement par voie transcellulaire (également paracellulaire pour SNAP), avec une perméabilité moyenne. Après avoir traversé la barrière intestinale, les RSNO atteindront les vaisseaux sanguins. Pour comparer leur capacité à former un stock vasculaire de NO dans des aortes (avec endothélium intact ou retiré), nous avons quantifié le stock, vérifié sa biodisponibilité pour la vasorelaxation et évalué son impact sur une vasoconstriction induite par la phénylephrine (PHE). L’incubation des aortes avec les RSNO augmente le stock basal de NO par un facteur trois à cinq. Ce stock est mobilisable pour induire la vasorelaxation et efficace pour diminuer la réactivité vasculaire à la PHE (NACNO>GSNO = SNAP), seulement dans les aortes dont l’endothélium a été retiré. Comme la perméabilité intestinale des RSNO est moyenne, l’intégration du GSNO dans une formulation appropriée est nécessaire. Vu l’impossibilité de résoudre les problèmes liés aux particules composites, le protocole de production des GSNO-NP a été modifié pour produire des microparticules (deux types selon l’état liquide ou solide de GSNO dans la phase interne de l’émulsion). Les deux types de microparticules avaient une libération de GSNO ralentie par rapport aux GSNO-NP. Les nano- comme les micro-particules ont pu être stabilisées par lyophilisation, et amélioraient la perméabilité intestinale de GSNO (jusqu’à une forte perméabilité avec les microparticules). Par conséquent, une administration orale de nano/microparticules chargées en GSNO/RSNO pourrait représenter une nouvelle approche thérapeutique pour les maladies cardiovasculaires. / S-nitrosothiols (RSNOs) such as S-nitrosoglutathione (GSNO) are promising nitric oxide (NO) donors for cardiovascular diseases treatment. However, they are poorly stable drug candidates. In previous studies, GSNO-loaded nanoparticles (GSNO-NP) were embedded into an alginate/chitosan matrix. Resulting nanocomposite particles showed high encapsulation and sustained release of GSNO, and led to the formation of a NO store in the wall of aorta after a single oral administration to rats. However, these nanocomposite particles have several limitations such as time-consuming preparation, lack of both stability and reproducibility. This thesis work aimed at: 1) Elucidate the mechanism of free RSNOs intestinal absorption; 2) Evaluate ability of free RSNOs to form a vascular NO store; 3) Optimize the GSNO formulation. In this study, we showed that the intestinal permeability (in vitro model of intestinal barrier) of GSNO, S-nitroso-N-acetylcysteine (NACNO) and S-nitroso-N-acetylpenicillamine (SNAP) was a passive diffusion, following the transcellular pathway (and also the paracellular way for SNAP) and belonging to the medium permeability class. After crossing the intestinal barrier, RSNOs will reach the vasculature. In order to compare the ability of free RSNOs to form a vascular store of NO either in endothelium-intact or endothelium-removed aortae, we quantified the store, verified its bioavailability for vasorelaxation and evaluated its impact on phenylephrine (PHE)-induced vasoconstriction. Incubation with RSNOs increased the basal NO store three to five times. This store is still bioavailable to induce vasorelaxation and efficient to induce vascular hyporeactivity to PHE (NACNO> GSNO = SNAP) only in endothelium-removed aortae. As intestinal permeability of RSNOs was in the medium class, the integration of GSNO into an appropriate delivery system is essential. Limitations of previously developed nanocomposites particles were impossible to bypass so the production process of GSNO-NP was modified (liquid or solid GSNO in the internal phase of the emulsion) to produce microparticles. Both kinds of microparticles exhibited a slower release of GSNO than GSNO-NP. Nano-and micro-particles were stable after lyophilization and presented an enhancement of GSNO intestinal permeability (up to high permeability class for microparticles). Thus, oral administration of GSNO/RSNO loaded nano/micro particles seems to be a promising avenue for the treatment of cardiovascular diseases.