Evaluation of the GCPQ as drug delivery system using atomistic molecular dynamics simulation and experimental methods

This work applies atomistic molecular dynamics (MD) simulations to the investigation of quaternary ammonium palmitoyl glycol chitosan (GCPQ) as a drug delivery system, with the aim of predicting its molecular structure at the atomic level, elucidating the mechanisms of GCPQ self-assembly and drug encapsulation, and providing guidelines for rational polymer design and drug – carrier selection. After building the initial atomistic models (Chapter 2), the ab initio GCPQ polymer model is introduced (Chapter 3) with a developed protocol allowing the acceptable MD simulation time and real time consumption while removing the bias of the start-ing polymer conformation and system configuration, and the findings expanded to explore the effect of various parameters on the self-assembly of GCPQ polymers (Chapter 4), including polymer architecture (distribution of hydrophobic moieties), composition (degree of palmitoylation), and environmental factors (pH and salt concentration). The results suggest that GCPQ polymers self-assemble into a physically crosslinked hydrogel-like matrix with absorbed water, rather than forming the typical core-shell micelles previously assumed. The self-assembly is primarily governed by the pH, with additional contributions from polymer structure, modification levels, and ionic conditions in a complex non-monotonic way. The resulting assemblies are explored further (Chapter 5) by screening their interaction with different drug molecules (drug-to-polymer molar ratio 1:2) to evaluate the nature of binding, affinity, impact on morphology and loading performance. Our findings indicate that GCPQ can encapsulate a wide range of drugs, both hydro-philic and hydrophobic, not so much by the provision of separate hydrophilic or hydrophobic domains but rather at molecular level through the formation of individual dynamic binding pockets which is accompanied by polymer chain conformation changes maximising these interactions. A more detailed investigation of the specific effects of chemical and physical parameters, including degree of palmitoylation, molecular weight, pH, and drug-to-polymer ratio, on drug loading is carried out using curcumin as the model drug (Chapter 6). The simulations reveal that loading performance is influenced by mul-tiple factors, suggested by Pearson correlation coefficients (~0.7, DP vs number of hydrophobic contacts; -0.7~-0.9, pH vs polymer conformations), requiring a balance between hydrophobicity, solubility, and structural flexibility for optimal design. Finally, Chapter 7 reports experimental data looking to explore to what extent the atomistic MD simulations accurately reflect these processes accurately. Specifically, different GCPQ polymers with structural modifications similar to those explored in the models were synthesised and characterised alongside GCPQ-curcumin formulations, enabling direct comparison with computational predictions.