| Summary: | The work reported in this thesis aims to develop and apply new assays for high content analysis (HCA) based on novel automated fluorescence lifetime imaging microscopy (FLIM) technology adapted for multiwell plate readers and evaluate their potential for drug discovery. Two such FLIM multiwell plate readers were investigated, one based on a custom-modified commercially available plate reader (GE Healthcare In Cell 1000) and the other based on an Olympus IX-81 wide-field microscope adapted for use as an automated multiwell plate imaging system. To evaluate the potential for drug discovery, an exemplar assay of HIV-1 Gag protein aggregation was developed and used to evaluate the performance of the multiwell plate readers. HIV-1 Gag is the major structural protein within HIV-1 virions and is thought to interact with other viral proteins, the viral genome and with a large number of host cell factors to orchestrate the formation of new virions. HIV-1 Gag protein oligomerisation is a precursor to virion production at the plasma membrane of the target cell during the HIV virus life cycle and so represents a potential readout for testing the efficacy of anti HIV drugs. The expression of HIV-1 Gag alone within living cells leads to the formation of virus-like particles (VLPs), which provide a convenient and safe means to study this late stage of the HIV cycle. This exemplar assay is based on Förster Resonance Energy Transfer (FRET) between appropriately (fluorescently) labelled HIV-1 Gag proteins. By tagging HIV-1 Gag proteins with either a donor fluorophore or an acceptor fluorophore, a FRET signal can be utilised to indicate when the oligomerisation brings the donor and acceptor within close ( < ~ 10 nm) proximity and this can be read out and mapped using FLIM to observe the decrease in donor fluorescence lifetime that is a consequence of FRET. In the first instance the Gag proteins were stochastically labelled with either CFP or YFP and FRET was mapped by imaging the CFP lifetime. The assay could also be implemented by labelling the Gag protein with CFP only and detecting the small change in lifetime that occurs during homo-FRET of CFP. To evaluate and validate the assay, biological controls were developed using mutants of the Gag protein that lacked the ability to be myristoylated, a pre-requisite for the Gag protein to assemble at the cell plasma membrane where the VLP are formed. Comparisons were made using both myristoylated (WT) and non- myristoylated (mutated) HIV-1 Gag proteins to demonstrate each plate reader's ability to read out levels of HIV-1 Gag protein aggregation. To further characterise the performance of the assay and the plate readers, a dose response study was undertaken using an inhibitor of the enzyme responsible for myristoylation in eukaryotic cells and the assay was fully characterised following standard pharmaceutical industry guidelines. Accounting for experimental factors such as pipetting errors, plate edge effects, spatial uniformity and drift over time, these characterisations and dose response studies yielded Z' factors to reflect the practical quality of the assays and thereby provided a robust means to compare different approaches, including different labelling strategies (e.g. hetero-FRET v. homo-FRET), imaging strategies (e.g. wide-field v. optically sectioned) and data analysis strategies (e.g. fitting models, image segmentation). To my knowledge this represents the first such robust and systematic evaluation of FLIM assays, e.g. using Z' from dose response curves, and is therefore of value to the pharmaceutical industry and other potential users of FLIM HCA.
|
|---|
