New infrared and fluorescence techniques for the elucidation of cellular ultrastructure

The underlying philosophy of this doctoral thesis is that application of advanced experimental techniques to address open research questions while synchronously working to benchmark and augment the instrumentation and methodologies, produces an enhanced appreciation and understanding of both the experiments undertaken and the resultant data. This approach guided and informed the research carried out during the thesis which sought to investigate the constituents, concentrations, structures and functions of biological samples using two state-of-the-art techniques: Fourier transform infrared (FTIR) spectroscopy and super resolution single molecule localisation microscopy (SMLM). Significant contributions to a diverse range of research areas as well as improvements to experimental methodologies were achieved demonstrating the soundness of this approach and the efficacy of both FTIR spectroscopy and SMLM. New methodologies for benchtop and synchrotron source (S-)FTIR spectroscopy of live and hydrated samples were developed and applied to detect a previously unknown en masse reversible B- to A-DNA conformational transition in single cells. This was in counterpoint to the long-standing assumption that the majority of DNA exists persistently in the B-form. Subsequent detection of the B-A-B transition in functional prokaryotes during desiccation and subsequent rehydration demonstrated a possible role for the A-form helix as a defence mechanism against various forms of damage and as key in the evolution of nucleic acids. Successful detection of DNA conformation in cells using S-FTIR was also contrary to previous FTIR spectroscopic work that hypothesised that absorption of infrared light by compacted DNA did not obey Beer-Lambert’s Law and, as a result, dictated that DNA absorptions from cell samples could not be regarded quantitatively. FTIR spectra of standard mixtures of protein and DNA were used to produce multivariate regression models, which were then used to accurately estimate the DNA content of simple cells. As well as debunking the “Dark DNA” hypothesis which had hampered FTIR biospectroscopic research for over a decade, these experiments clearly demonstrated improved detection sensitivity in hydrated cell work. This motivated application of the developed live-cell S-FTIR spectroscopy methodologies to investigate the changing biochemistry of cells as they progressed through the cell cycle. Multivariate cluster analysis of spectra collected from COS-7 cells over the course of the cell cycle allowed differentiation of cells only two hours apart within the same phase. Furthermore, through analysis of the associated causes for variance, it was demonstrated that the spectroscopic detection of subtle changes in relative DNA, protein and lipid concentrations was possible. This level of sensitivity had previously been unreachable with FTIR spectra of dehydrated or fixed cells and is now envisioned as stimulating further live-cell spectroscopic research and allowing detection of previously unknown biochemistry. Concurrent to this work, the first home-built SMLM setup in Australia was established to provide a complementary technique for the FTIR spectroscopic work being undertaken. As a very new technique, development and application of SMLM involved design and construction of the necessary instrumentation from scratch as well as extensive optimisation of sample preparation protocols for imaging of microtubules, actin and mitochondria. To achieve this various parameters involved in cell fixation, data acquisition and image rendering were investigated systematically and the numerous encountered causes of image artefacts were characterised. Artefacts were shown to arise from all aspects of the experiment: hardware, software, and particularly from the sample. Importantly, it was shown that a commonly employed permeabilisation step in the preparation of microtubules, compromises biological relevance in pursuit of a ‘better’ image. Application of the protocols developed for labelling of microtubules enabled identification of a novel microtubule bundling response in cells expressing a Rabies protein. A number of mutants of the protein were investigated and it was found that the extent of microtubule bundling correlated with both the pathogenicity of virus in previous live mouse experiments and the antagonism to the innate interferon response. These results demonstrate a previously unknown interaction between the viral protein, the host cell immune response, and the cytoskeletal architecture, and have inspired much future research aimed at elucidating the mechanisms at play. SMLM was also applied to the imaging of fluorescently labelled 100 and 20 nm silica nanospheres and the exemplary spatial resolution achievable using the set-up was convincingly demonstrated. These nanospheres were also established as a proof-of-principle material for correlative electron microscopy-SMLM. Finally, SMLM was successfully used to image extracted double stranded DNA exhibiting the potential for future use of this technique to reveal sub-diffraction chromatin and DNA structure. This research underscores the potential of FTIR spectroscopy and SMLM both independently and for future complementary use, as well as highlights the importance of ongoing technique augmentation in parallel with the application of experimental methods to novel research endeavours.<div><br></div><div> Awards: Vice-Chancellor’s Commendation for Doctoral Thesis Excellence in 2014.