Hyperbranched polymers for in vivo multimodal molecular imaging

Abstract

For the development of the next generation of polymeric nanomedicines, it is crucial to gain a fundamental understanding of their behaviour and interactions with and within biological systems. Moving beyond in vitro models, into in vivo models, earlier in the development process will greatly aid in the advancement of the next generation of nanomedicines. By moving to whole animal models, our understanding of these systems progresses beyond cell targeting and uptake, to developing mechanisms for how these materials will distribute through tissues and their pharmacokinetic profile. This information is important for truly assessing the performance of a nanomedicine. One possible set of tools for obtaining this information is molecular imaging. Molecular imaging is a field of research dedicated to the real time monitoring of biological processes in vivo, without the use of invasive techniques such as biopsies and dissections. Molecular imaging has been used extensively to follow the in vivo behaviour of a labelled material. This is advantageous because the performance of a single material in one subject can be monitored and mapped against the progression of disease. It can help to provide the pharmacokinetic information necessary for preclinical development of nanomedicines. Nanomedicines can be designed to combine molecular imaging with targeting molecules and therapeutic agents to create a theranostic, which can be used for simultaneous imaging and treatment of disease. This thesis aims to synthesise novel multimodal molecular imaging agents based on a hyperbranched polymer architecture, and to gain a deeper understanding of how these materials behave in vivo. To achieve this, biocompatible hyperbranched polymers with defined architectures were synthesised using RAFT polymerisation techniques. These materials were extensively characterised using a wide range of spectroscopic techniques to thoroughly understand their physical and chemical properties. A variety of synthetic strategies were investigated for functionalising both the α- and ω-chain ends of these polymers with multiple imaging ligands to form multimodal imaging agents. Far-red and near-infrared fluorophores provided for fluorescence imaging and radiometal chelators allowed for positron emission tomography (PET) imaging. These hyperbranched polymer systems were first evaluated as molecular imaging agents in C57 BL/6J mice using whole animal fluorescence and PET-CT imaging. It was shown that the rate of excretion was dependent on the size and level of branching of the hyperbranched polymer cores. The larger more highly branched material showed extended circulation times, making it suitable for use as a passive targeting agent for cancer. It was demonstrated in a murine model for melanoma, that the material showed significant uptake within the tumour after 24 hours and that the material was not cleared from the tissue within 72 hours. To gain a deeper understanding of the behaviour of these materials in vivo, PET imaging was combined with gadolinium contrast enhanced MRI, in order to gain both molecular and physiological information. Using this technique, we were able to show that while a folic acid targeted hyperbranched polymer did accumulate in the tumour tissue, its distribution was concentrated in highly vascularised areas of the tumour. This is the first time that this phenomenon has been demonstrated at a macroscopic level, in a living animal. This has important implications for using these materials as theranostics, because heterogeneous distribution of the nanomaterial, and therefore delivery of a therapeutic, can lead to ineffective treatment of the cancer and thus lead to tumour recurrence. In further development of these imaging agents into theranostics, targeting of the hyperbranched polymers by conjugating single chain fragment antibodies (scFv) was explored. Two potential routes to improve efficiency of conjugation were investigated. Both approaches used novel bifunctional oligoethylene glycol (OEG) linkers to introduce the required chemical functionality to either the hyperbranched polymer or scFv. The first approach utilised a heterobifunctional OEG which was synthesised with a pentafluorophenol ester at one end for coupling with amines and an ω-azide group at the other end to allow for the copper catalysed Huigsen 1,3-dipolar cycloaddition reactions. This linker was first attached to the scFv via activated ester chemistry, to provide the necessary azide functionality for coupling of the scFv to the alkyne end groups of the hyperbranched polymer. The second route used an enzymatic cross coupling approach using the sortase enzyme. In order to achieve this, a triglycine functionalised OEG ligand was synthesised and attached to the hyperbranched polymer. The triglycine could then be used as a substrate for enzymatic cross coupling to scFv bioengineered to possess the required recognition sequence (LPETG). Despite both OEG linkers being demonstrated to be able to undergo conjugation to both the hyperbranched polymers and scFv independently, further optimisation is required to achieve conjugation of the two macromolecules.