Abstract
The work presented in this thesis focuses on the development of carbohydrate-functionalized nanoparticles (NPs) and their diverse biomedical applications such as cell-specific targeting, imaging and cancer therapy. The asialoglycoprotein receptors (ASGP-Rs), which can specifically interact with galactose (Gal) or N-acetyl-galactosamine (GalNAc), were chosen as a way to target HepG2 cells. Targeting ligands were presented on NP carriers such as magnetic nanoparticles (MNPs), silica oxide nanoparticles (SiO2NPs) and mesoporous silica nanoparticles (MSNs). Moreover, a dendrimer-like multivalent galactosyl carborane (dendritic glyco-borane, DGB) was developed for potential application in boron neutron capture therapy (BNCT). Multivalent carriers such as NPs or dendrimer-like molecules decorated with trivalent-Gal moieties are good systems for HepG2 cell-specific targeting via receptor-mediated endocytosis. The fluorescent NPs (MNP, SiO2NPs and MSN) were fabricated by a competition method to incorporate Cy3 without the loss of the original surface amine density that allows for the loading of high concentrations of targeting ligand. The fluorescent dye Cy3 and galactose derivatives were covalently assembled with different ratios on the surfaces of MNPs to produce multifunctional HepG2 cancer cell–targeting agents. We found that the specific uptake of galactosyl-conjugated MNP by HepG2 cell via receptor-mediated endocytosis and T-Gal-s-Cy3@MNP was the most efficiently ingested MNP tested. Moreover, we found that by adjusting the spatial arrangement of the ligands on MNPs to match the distance between carbohydrate binding sites on the receptor, the increase in cellular uptake by multivalent presentation of the ligand could be maximized. All the glyco Cy3@MNPs are not cytotoxic, indicating that they may potentially be used for in vivo applications. We also used sequential double click chemistry (SDCC) involving strain-promoted azide-alkyne cycloaddition (SPAAC) and Cu(I)-catalyzed azide-alkyne cycloaddition to assemble an anticancer drug (paclitaxel, PTX) and a targeting ligand (trivalent galactosside, TGal) on a fluorescent silicon oxide nanoparticle (SiO2NP) by using a di-alkyne linker as a bridge which can increase the surface availability for further functionalization. The expensive compound used in SPAAC can be easily recovered due to the absence of other reagents in the reaction mixture. The use of a trivalent galactosyl ligand, which interacts with the ASGP-Rs on the surface of HepG2 cells, not only provides a targeting function, but also overcomes the inherent low water solubility of PTX. The presence of a fluorescent probe, a targeting ligand, and an anticancer drug on the multifunctional TGal-PTX@Cy3SiO2NP allows for real-time imaging, specific cancer-cell targeting and cell-killing effects that are similar to PTX. Boron neutron capture therapy (BNCT) relies on the uptake of a sufficient number of 10B atoms by the target cell. The cell is then being irradiated with neutrons and the absorption of neutrons by 10B atoms leads to the release of energy and finally to the death of the cell. The success of BNCT requires a sufficient number of 10B atoms to be delivered to the targeted cells and the main challenges often arise from the low water solubility of boron compounds, the unselective uptake of the cancer cell, and the toxicity of boron. Two types (dendritic glyco-borane, DGB and T-Gal-B-Cy3@MSN) of boron neutron capture therapy (BNCT) agents were design as third generation BNCT agents. DGB which possesses trivalent Gal moieties and trivalent carboranes was synthesized and tested as a potential cell-targeting agent in BNCT with HepG2 cells. DGB improved the delivery of boron to HepG2 cells, and neutron irradiation data show that DGB exhibits a ten-fold improvement at killing HepG2 cells compared to BSH, an FDA-approved drug. Another strategy we pursued was to use mesoporous silicon NPs (MSNs) as 10B carriers for a “Trojan horse” type approach. T-Gal-B-Cy3@MSN as BNCT agent has large pore volumes which allow for the loading of o-carborane (almost 50% boron atoms per MSNs particle). This resolves previous limitations concerning the drug release kinetics of MSNs. Moreover, the trivalent Gal moiety serves as a targeting ligand for the targeting of HepG2 cells. We believe that our approach provides new insights on the development of dendrimer- and MSN-type BNTC agents.