Abstract
Introduction: Drug delivery to the brain to treat neurological disorders is a challenge due to the blood-brain barrier, which physically and actively restricts the transport of molecules into the brain. Implantable drug delivery systems (e.g. hydrogels and nano/micro-particles) have been developed to overcome the blood-brain barrier issues. However, limitations such as excessive swelling or disintegration of the hydrogel, retaining the particles at the site of action, low loading efficiency and burst release of the drug can occur. Therefore, the overarching aim of this thesis was to develop and characterise implantable nanofibres (NF) for drug delivery to treat neurological disorders using low toxic solvent systems.
Methods: NFs were prepared using the technique of electrospinning using biodegradable polymers. For the first time, a new, low toxicity solvent system consisting of acetone (ACE) and ethyl acetate (EtAc) (Class III solvents) was used to prepare poly-(lactic-coglycolic acid) (PLGA) NFs. The electrospun PLGA NFs were characterised based on morphology, solid-state characteristics, tensile strength and hydrophobicity and compared to NFs electrospun using more toxic solvent systems of dichloromethane (DCM) and dimethylformamide (DMF) (Class II solvents). The incorporation and release kinetics of drugs with different molecular weights (paracetamol 151.17 g/mol, ampicillin 349.41 g/mol and tetracycline 480.90 g/mol) were investigated.
Poly-(caprolactone) (PCL) was also investigated as a NF drug delivery system and was co-electrospun with chitosan (CH), gelatine (GEL) or poly-(ethylene oxide) (PEO). The NFs were compared to determine if the blended polymers improved the chemical and physical properties of PCL NFs. In vitro degradation studies were conducted and the drug release kinetics of L-655,708 from PCL and the PCL/blend NFs was investigated. Further release studies from PCL/CH NFs were undertaken, with a comparison between the neuroactive-drug L-655,708 and indomethacin. The in vitro cytotoxicity and immune response assays were conducted to investigate the biocompatibility of the NFs.
Results: Compared to PLGA NFs produced using the conventional solvents DCM and DMF, the new ACE:EtAc solvent system resulted in NFs with a decrease in fibre diameter uniformity and an increased hydrophobic surface as determined by field emission scanning electron microscopy (FE-SEM) and water contact angle measurements, respectively. NFs made using both solvent systems were amorphous, however, the residual solvent could be detected by thermogravimetric analysis. Attenuated total reflectance-Fourier transform infrared spectroscopy analysis showed there was no change in PLGA chemical composition after electrospinning. Tensile strength studies showed that NFs prepared using the less toxic ACE:EtAc solvent system was lower in tensile strength and underwent a different mechanical deformation process compared with DCM:DMF NFs.
The solid-state analysis confirmed the incorporation of the drug in the PLGA NFs, with FE-SEM and hot stage microscopy identifying drug on the surface of the NFs. The drug release kinetics of tetracycline and paracetamol were similar from both NFs with an initial burst release, followed by a gradual release over 504 h. Assessment, however, of the release profile for ampicillin revealed a smaller initial bolus release with the release from PLGA ACE:EtAc being slower than PLGA DCM:DMF. The release kinetics indicated that the molecular weight of the incorporated drug, nor fibre diameter influenced drug release.
FE-SEM analysis of PCL and PCL/blend NFs showed smooth, bead-free NFs with average fibre diameter in the order of PCL/CH < PCL < PCL/PEO < PCL/GEL. The polymers incorporated into the NFs were chemically unchanged by the electrospinning process and the NFs did not appear to have residual solvent present. The PCL component of PCL/CH and PCL/GEL NFs had reduced crystallinity, while PCL/PEO NF was the most thermally stable and had the highest mechanical strength. The hydrophobic surface properties of the PCL NF were decreased when blended with PEO and GEL.
An in vitro degradation study revealed a reduction in mass of 44% and 38% from PCL/PEO and PCL/GEL, with negligible loss over the 4-weeks from PCL and PCL/CH. With polymer degradation said to influence drug release, it was interesting to observe that the degradation study did not translate into the faster release of L-655,708 from PCL/PEO or PCL/GEL, as PCL/CH showed the fastest release with 100% release after 24 h. The drug release from PCL/CH was further investigated as a charge interaction was postulated to be the mechanism of release, however, the release of the neutral drug indomethacin was similar to the release of L-655,708. Two theories were put forward to explain the release kinetics from the PCL/CH NF; 1) phase separation of the polymers occurred during electrospinning, forming two distinct fibres with the hydrophobic drugs favouring the lesser hydrophobic interaction with CH resulting in drug diffusion out of the CH NF. Alternatively, a core/sheath NF occurred where the drug was located in the CH sheath around a PCL core resulting in drug distributed at the surface of the NF. 2) During the electrospinning process, nano-pores were created on the surface of the NF that facilitated the diffusion of the drugs from the NF. In vitro cell culture studies showed PCL/GEL had the greatest cell proliferation over 24 h and the highest cell viability over 3 days. NFs with fibroblast cells cultured on the surface did not elicit an immune response, however further studies need to be conducted to confirm biocompatibility.
Conclusion: This thesis demonstrates the development of a drug delivery system using the electrospinning process that utilises less toxic solvent systems. Drug release was sustained over 21 days and supports the use of a NF as a promising candidate for the local delivery to treat neurological disorders.