ABSTRACT
Most antiviral and anticancer nucleosides are prodrugs that require stepwise phosphorylation to their triphosphate nucleotide form for biological activity. Monophosphorylation may be rate-limiting, and the nucleotides may be unstable and poorly internalized by target cells. Effective targeting and delivery systems for nucleoside drugs, including oligonucleotides used in molecular therapeutics, could augment their efficacy. The development of a carrier designed to effect selective transmembrane internalization of nucleotides via the asialoglycoprotein receptor (ASGPr) is now reported. In this work, the polycationic, polygalactosyl drug delivery carrier heptakis[6-amino-6-deoxy-2-O-(3-(1-thio-ß-D-galactopyranosyl)-propyl)]-ß-cyclodextrin hepta-acetate salt (GCyDAc), potentially a bifunctional carrier of (poly)nucleotides, was modeled by molecular docking in silico as an ASGPr-ligand, then synthesized for testing. The antivirals arabinosyl adenine (araA, vidarabine, an early generation antiviral nucleoside), arabinosyl adenine 5'-monophosphate (araAMP), and 12-mer-araAMP (p-araAMP) were selected for individual formulation with GCyDAc to develop this concept. Experimentally, beta cyclodextrin was decorated with seven protonated amino substituents on the primary face, and seven thiogalactose residues on its secondary face. AraA, araAMP, and p-araAMP were individually complexed with GCyDAc and complex formation for each drug was confirmed by differential scanning calorimetry (DSC). Finally, the free drugs and their GCyDAc complexes were evaluated for antiviral activity using ASGPr-expressing HepAD38 cells in cell culture. In this model, araA, araAMP, and p-araAMP showed relative antiviral potencies of 1.0, 1.1, and 1.2, respectively. In comparison, GCyDAc-complexes of araA, araAMP, and p-araAMP were 2.5, 1.3, and 1.2 times more effective than non-complexed araA in suppressing viral DNA production. The antiviral potencies of these complexes were minimally supportive of the hypothesis that ASGPr-targeted, CyD-based charge-association complexation of nucleosides and nucleotides could effectively enhance antiviral efficacy. GCyDAc was non-toxic to mammalian cells in cell culture, as determined using the MTS proliferation assay.
ABSTRACT
PURPOSE: Iododeoxyuridine (IUdR) has a very short in vivo half-life and consequently achieves low target-tissue concentrations with concomitant lower efficacy than would be predicted from in vitro studies. This work reports the preparation of IUdR:beta-cyclodextrin (beta-CyD) inclusion complexes designed to reduce in vivo inactivation of IUdR. METHODS: IUdR was derivatized with either 1-adamantanecarbonyl chloride or 4-(1-adamantyl-carbamoyl)butanoic acid, to prepare 5'-O-(1-adamantoyl)-5-iodo-2'-deoxyuridine 1 and 5'-O-(4-(1-adamantylcarbamoyl)butoyl)-5-iodo-2'-deoxy-uridine 4, respectively. beta-CyD complexes 5 and 6 were formed by vigorous stirring of 1:1 solutions of beta-CyD and 1 or 4, respectively, in D2O under argon. Complexation was inferred from DSC, powder x-ray diffractometry and NMR spectrometry. The dissociation of 5 in water and under cholesterol challenge, and the effect of complexation on the stability of 1 was determined by incubation in plasma. RESULTS: IUdR coupling with adamantanecarbonyl chloride proceeded smoothly to afford 1 (69 %) and the di-substituted derivative, 3',5'-di-O-(1-adamantoyl)-5-iodo-2'-deoxyuridine 2 (8 %); 4 was obtained in 42 % yield. The formation of 1:1 complexes 5 and 6 was inferred from NMR chemical shift data. In serum, 1 was 90 % hydrolyzed to IUdR in 30 min, compared to 10 % hydrolysis of 1 to IUdR when from complex 5. CONCLUSIONS: Inclusion complexes were formed between beta-CyD and adamantamine-IUdR conjugates at 1:1 molar ratios. The complex 5 was resistant to dissociation by cholesterol challenge, and 5 was more slowly converted to IUdR than non-complexed 1. In vivo studies are required to further exploit the beta-CyD inclusion complex approach for improved delivery of nucleoside derivatives.
