Bone tissue engineering is promising as a potential alternative in the treatment of non-healing bone fractures. In this approach, nanocomposite thermoplastic poly(ε-caprolactone) (PCL) - nanohydroxyapatite (HAP) scaffolds show encouraging properties; however, challenges still exist in fabricating matrices with optimal bioactivity and suitable mechanical properties due to problems in dispersion of HAP. To this end, a novel surface modification of covalent attachment of polyelectrolytes to HAP was conducted and an understanding of the optimal reaction conditions was sought. The aim was to design a surface modification method that would have the ability of improving the dispersion of HAP in PCL solution and scaffold, and additionally be able to control the release of the osteoinductive growth factor bone morphogenetic protein-2 (BMP-2), while maintaining its protein activity. For this purpose, heparin was selected as the polyelectrolyte.
HAP modification using 3-aminopropyltriethoxysilane (APTES) was carried out in order to enable the attachment of a polyelectrolyte to the surface covalently and it was hypothesised that the formation of an APTES layer would serve as a platform to do this. Reaction of APTES with HAP was conducted in toluene and the effect of reaction time and post-reaction curing temperature on the formed surface layer was investigated through X-ray photoelectron (XPS), Fourier transform infrared (FT-IR) and solid-state nuclear magnetic resonance (NMR) spectroscopy. Results showed that the chemical composition of the APTES layer formed was significantly influenced by the curing temperature and it was concluded that there was a trade-off between functionality and stability for these APTES-modified HAP particles. Subsequent attachment of PAA via both ionic interaction and covalent bonding using carbodiimide chemistry (as verified from XPS, FT-IR and NMR) resulted in particles which were more stable to dispersion in aqueous solution, both with respect to chemical composition and aggregation. Higher graft density of PAA was achieved, compared to physisorption onto HAP. The study of the covalent attachment of PAA served as a model for the biologically functional heparin.
Investigation of the attachment of heparin revealed that the grafting density achieved was dependent on the curing temperature used in the fabrication of APTES-modified HAP. Higher grafting density was achieved on APTES-modified HAP compared to direct physisorption of heparin to HAP. Although comparable amounts of heparin were attached via both covalent and ionic grafting to the APTES-modified particles, the surface characterization by zeta potential measurements and heparin release studies indicated that the conformation of the heparin on the surface was dependent on the method of attachment used. BMP-2 was adsorbed on to the particles at a BMP-2 to HAP mass ratio of 1:4000. It was determined that after a 7 day period in bovine serum albumin containing phosphate-buffered saline, 31 % of the bound BMP-2 was released from the particles with the heparin attached covalently via the silane layer, compared with 16 % from the particles with the heparin attached ionically via the silane layer and 5 % when bound to HAP or HAP with physisorbed heparin, confirming that by altering the mode of attachment of heparin to HAP the release of BMP-2 could be manipulated. Importantly, the BMP-2 released from the heparin attached particles containing an intermediate APTES layer was found to be biologically active.
Investigation of the dispersion stability of the particles was carried out in 1,4-dioxane (DO), 1 mM KCl, DO and water mixture, and a water-containing PCL solution. It was determined, by DLS size measurements, that the modification of HAP with heparin (by both physical adsorption and covalent attachment via an APTES layer) improved the dispersion stability of the HAP in 1 mM KCl and the PCL solution, while no difference was observed in DO. PCL composite scaffolds were fabricated using the thermally induced phase separation technique in DO and water/DO mixed solvent. The distribution of the particles within the PCL matrix (using transmission electron microscopy) demonstrated that the particles were well dispersed in the scaffolds made using DO as solvent and were segregated to the polymer surface in the scaffolds made using the mixed solvent, significantly different to the behaviour in solution. There was no significant difference in the distribution of the particles in the vertical direction (determined using a Ca ion selective electrode). The distribution of the HAP particles was determined to be significantly affected by the solvent system used, the polymer used and the phase separation mechanism that occurred and it was concluded that a direct relationship between HAP solution stability and dispersion in the scaffold did not exist, contrary to previous claims. Although the inclusion of HAP did not significantly improve the compressive moduli of the PCL scaffolds, it did not significantly deteriorate the moduli and the reasons for this are discussed in the thesis. Importantly, the composite scaffolds had significantly greater bioactivity (determined by immersion in simulated body fluid) than the pristine scaffolds.