Use this URL to cite or link to this record in EThOS:
Title: Quantitative optical microscopy of bio-nanostructures
Author: Williams, Joseph Marc Douglas Bleddyn
Awarding Body: Cardiff Univesity
Current Institution: Cardiff University
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Thesis embargoed until 13 May 2021
Access from Institution:
The work in this thesis was motivated by three main questions, namely: i) how far can quantitative differential interference contrast (DIC) microscopy be expanded and improved to accurately measure the thickness of single lipid bilayers, ii) do Gold nanoparticle-fluorophore conjugates provide a stable and reliable dual probe for correlative imaging studies, iii) can DIC offer a label-free quantitative method to accurately measure the surface temperature of a Gold nanoparticle (AuNP). To address the first question, a software has been developed which extracts optical phase information from a specimen imaged with differential interference contrast. A Wiener filtering approach is used to integrate the differential phase contrast image, to obtain an optical phase image of the specimen. The quality of Wiener filtered images is improved through the use of an apodization process to produce a two dimensional window around the differential phase image similar to the Hann window often used in one dimensional Fast Fourier Transform algorithm. Additionally, an energy minimisation routine further improves the quality of the retrieved phase images. From results presented in this thesis, it has been shown that one can perform the energy minimisation algorithm on multiple images in parallel with varied parameters. Images reach a converged status after approximately 107 iterations. To perform 107 iterations on graphics processing units (GPUs), it takes approximately 72 hours. To address the second question, following the acquisition of a series of confocal fluorescence and four wave mixing (FWM) amplitude images from bioconjugated fluorescently labelled AuNPs in Naya Giannakopoulou’s thesis [1], a cross-correlation algorithm has been developed to quantify the Pearson’s coefficient for lateral shifts between these images. To accommodate for any angular discrepancies between these images, a further algorithm was developed in order to rotate images relative to one another so that cross the images may be cross correlated. Pearson’s coefficients from cross correlated images showed a lack of colocalisation between FWM amplitude and fluorescence images of HeLa cells loaded with AuNPfluorophore constructs, generally resulting in values rP < 0.05 (apart from one instance where rP = 0.15) which confirms what was qualitatively reported in [1]. A direct study of the correlation between extinction and epi-fluorescence images of AuNP-fluorophore constructs prepared (washed) as per manufacturer specification and deposited onto a glass surface was subsequently performed. The cross cor- – ix – relation of extinction and epi-fluorescence images from a 10nm AuNP-fluorophore construct (10nmAuNP-SA(A-488)) gave Pearson’s coefficients rP < 0.06 for 1× and 3× washes, clearly indicating that the supposedly attached fluorophores are not reliable reporters of the NP location. Conversely, 20nm diameter AuNPs covalently bound to fluorescently labelled antibodies revealed a good degree of colocalisation with the fluorophore (rP = 0.476) after 3× washes. To answer the third question, a AuNP heating set-up was developed, using the qDIC imaging method outlined in this thesis. Briefly, a lipid bilayer was overlaid above a AuNP bound to a coverslip, and was photothermally heated at the localised surface plasmon resonance (LSPR). The heat which dissipates from the AuNP induces a phase transition in the lipid bilayer in its vicinity. By locating the phase boundary in the lipid bilayer, one can quantitatively measure the surface heating of the AuNP. In principle, the precise location of a phase boundary in a lipid bilayer is observable in qDIC. However, the dynamics observed through acquiring a qDIC time course did not resemble the behaviour previously observed in fluorescence measurements [2]. In particular, it is hypothesized that a blister effect occurred during resonant photothermal heating of a 50nm AuNP at the LS.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Q Science (General)