Evaluating hypoxia-targeting vectors for MRI applications

  • Nadeeja Wijesekera

Western Sydney University thesis: Doctoral thesis

Abstract

Tissue hypoxia, or oxygen deficiency in tissue, is usually the result of a locally restricted oxygen supply or an abnormal increase in oxygen consumption, or both. Hypoxia is related to a variety of pathological conditions including cancer, stroke and ischemic heart disease. In locally advanced solid tumors, hypoxic and anoxic regions are often distributed heterogeneously due to structurally and functionally abnormal vasculature. Tumor hypoxia has been closely associated with poor cancer prognosis after radiation and chemotherapy, and is known to encourage tumor proliferation factors, such as growth, apoptosis, angiogenesis and metastasis. In radiotherapy, dosage (i.e., the amount of radiation required to kill a tumorous tissue) is determined by the intrinsic radiosensitivity of the cancer cell population and the tumor microenvironment, in particular the oxygen level of the tissue. Therefore, hypoxic tissues are known to be resistant to the lethal effects of radiation. The ability to accurately identify the hypoxic levels of tumors, and regions within a tumor, can allow clinicians to alter radiotherapy treatment planning to improve treatment outcomes. Consequently, considerable efforts have been made to identify hypoxia within tumors. Of particularly interest at present is to develop chemicals, incorporating hypoxiatargeting vectors, which specifically accumulate in hypoxic tissue when injected and can then be traced/imaged by an imaging modality, thus allowing delineation of hypoxic parts of a tumor. The potential of these compounds as markers for hypoxic tumors has long been realised, and have been extensively exploited in positron-emission tomography imaging, but less so in magnetic resonance imaging. However, developing such compounds for MRI first requires determining which of the several available hypoxia-targeting vectors is most favourable. As these targeting vectors bind to macromolecular structures, a suitable model for such determinations is to probe the binding interactions of these vectors in albumin. Probing such interactions has been studied by nuclear magnetic resonance spectroscopy due to its general applicability and non-invasive nature. NMR diffusion measurements allow binding ligands to be easily differentiated from non-binding molecules by significant differences in the diffusion coefficients. The mathematical framework for analysing such data is provided by the Kärger equations. To understand limits of these equations, such as the short gradient pulse approximation, and thus the conditions under which these equations are valid, a detailed derivation of these equations for the case of two freely diffusing exchanging species is presented along with the simplifications that result in the slow and fast exchange limits. This leads to the population weighted diffusion equation, which is then used to quantitatively characterise the binding properties of the hypoxia-targeting vectors in albumin by the number of binding sites, n and the association constant, K. The binding properties of three hypoxia-targeting vectors were investigated: 2nitroimidazole, 4-nitroimidazole and 6-nitroquinoline, using NMR diffusion and relaxation measurements. For the vectors which showed binding in diffusion NMR experiments, the binding was probed at three different protein concentrations (0.23, 0.30 and 0.38 mM) with drug concentrations ranging from 0.005""0.16 M at 298 K. Simultaneously fitting the Kärger model to the data obtained from the NMR measurements (D, R1 and R2) yielded significantly more robust results than individually fitting each dataset, from which n and K were determined to be 21 ± 3 and 53 ± 4 Mâˆ'1, respectively. However, n and K were not determined for 4-nitroimidazole and 6-nitroquinoline as any binding interactions were not detected by the diffusion data. However, it is most likely the binding interactions are considerably weak based on the relaxation data, and thus the associated extremely fast exchange process was outside the time scale of the diffusion measurements. With 2-nitroimidazole possessing the strongest binding interactions of the three vectors studied, 2-nitroimidazole was conjugated to a saturated lipid chain, thus synthesising 1-(2-nitroimidazoyl)octadecane to be used as an active hypoxia-targeting amphiphilic vector. Hypoxia-specific paramagnetic liposomes were then formulated by incorporating commercially available paramagnetic amphiphilic chelates, and 1-(2nitroimidazoyl)octadecane lipids within the bilayer membranes of the 1,2-dipalmitoyl-snglycero-3-phosphocholine lipids. Two sets of hypoxia-specific paramagnetic liposomes were formulated: one containing PEGylated lipids and one without, to improve circulation times and optimise targeting, respectively. However, it is likely the absence of PEGylated lipids resulted in less stable liposomes. The relaxivity of the hypoxia-specific paramagnetic liposomes relative to the commercially available contrast agent, Magnevist, increased by a range of approximately 83 - 162%. The hypoxia-targeting ability of the liposomes was then assessed by comparing the cellular uptake of targeted and non-targeted paramagnetic liposomes in two cell lines: SHSY5Y human neuroblastoma and MCF-7 human breast cancer, in normoxic and hypoxic conditions. Inductively coupled plasma mass spectrometry revealed a three and five-fold increase in the gadolinium concentration of the hypoxic cells relative to the normoxic cells in the SH-SY5Y and MCF-7 cell lines, respectively, confirming the hypoxia-targeting ability of these liposomal formulations. However, indistinct contrast difference was observed between the MR images of the normoxic and hypoxic cells in either cell-line, thus suggesting the need to design liposomal formulations with either higher gadolinium content, or better contrast enhancement efficiency, or both. Nevertheless, the primary aim of confirming the hypoxiaspecificity of the synthesised paramagnetic liposomes was achieved. This now presents an opportunity to develop hypoxia-specific theranostic liposomes by incorporating hypoxiaactivated prodrugs in addition to the targeting and contrast components incorporated in this thesis, thereby combining diagnostic and therapeutic capabilities in a single agent.
Date of Award2018
Original languageEnglish

Keywords

  • oxygen
  • physiological transport
  • impaired oxygen delivery
  • blood-vessels
  • tumors
  • radiation
  • dosage
  • drug development.

Cite this

'