Cytoplasmic Delivery, Selective Protein Labeling and Luminescence Imaging of Tb Complexes

The quantitative analysis of biomolecular activity in living cells is essential for understanding biological processes like signal transduction, or identifying potential molecular targets for diseases treatments. Optical microscopy with fluorescent proteins or small molecule probes is a critical tool for studying the dynamic distribution and activity level of proteins, lipids, mRNA and other biomolecules. Conventional fluorescence microscopy measures localized changes in fluorescence intensity that, with proper probe or sensor design, can be correlated to biochemical function. However, it is often difficult to distinguish changes in intensity that are functionally relevant from those that are simply due to variable concentration or other factors. Fluorescence lifetime imaging microscopy (FLIM) is a powerful alternative for quantitative cellular imaging because the excited state lifetime of a fluorescent species is sensitive to its local environment yet independent of its concentration. Various methods have been developed for analyzing data obtained from live-cell FLIM measurements. A major goal of the work described in this dissertation was to quantitatively assess the accuracy and photon efficiency of time-gated FLIM in combination with luminescence lanthanide probes. Organic complexes of lanthanide cations, especially Tb(III) and Eu(III), have ms-scale excited-state lifetimes and discrete, narrow emission bands that span the visible spectrum. These features enable sensitive, multi-color imaging when lanthanide probes are used in combination with a time-gated luminescence microscope. Time-gating is a method of detecting luminescence where pulsed light is used to excite the specimen, and the detector is turned on for discrete periods (or gates) after the excitation pulse. With long-lived lanthanide species, time-gating permits both rejection of non-specific, ns-scale fluorescence background and sampling of the luminescence decay curve to extract lifetime information. As with any exogenous probe molecule, controlled delivery of lanthanide complexes into the interior of living cells is a critical prerequisite for quantitative studies. The Miller laboratory has shown that conjugation to cell penetrating peptides (CPPs) enables cytoplasmic delivery of otherwise membrane-impermeant Tb complexes. In Chapter 2, experimental results are presented that quantify the CPP-mediated uptake of protein-targeted Tb complexes. Time-gated microscopy data show that cytoplasmic uptake of nonaarginine-conjugated Tb protein labels occurs in multiple cell types, most cells in a culture (>75%) are loaded with probe, and the cellular probe abundance can be controlled by varying incubation conditions. Moreover, selective, ligand-directed labeling of human alkylguanine alkyl transferase (hAGT) and Eschericia coli dihydrofolate reductase (eDHFR) fusion proteins was achieved, and this enabled two-color, Tb-to-fluorescent protein Förster resonance energy transfer (FRET) imaging. These results represent the first reported use of CPP conjugation to mediate intracellular delivery of selective protein-targeted probes, and they represent a critical advance went in the development of lanthanide probe-based microscopy including FLIM. In chapter 3, the ability to microscopically measure lanthanide emission lifetimes is assessed both theoretically and experimentally. The number of photons required to accurately estimate lifetimes by least squares fitting of time-gated FLIM data was simulated for different imaging parameters including the configuration, number, and width of the gates. The simulation results were fit to data obtained from time-gated images of a Tb complex solution, and a close correspondence between measurement and theory was observed. A unique, noise filtering aspect of the intensified charge coupled device (ICCD) detector was found to lower the number of photon counts needed to achieve a given level of lifetime estimation accuracy, and the practical implications of this result for lanthanide-based cellular FLIM are discussed. The final chapter of this dissertation presents time-gated FLIM data obtained from images of a luminescent Tb complex, Lumi4-Tb, in solution, fixed tissue slices, and living cells. The data were analyzed using three different methods, non-linear least squares (NLLS) fitting, rapid lifetime determination (RLD), and the phasor approach. Each analytical method returned the same estimate of emission lifetime for a given sample type, albeit with differing levels of photon efficiency and relative error. The results proved promising for the prospects of cellular, lanthanide-based FLIM because accurate measurements could be obtained in cells with a temporal resolution that approached that of conventional FLIM. A perspective on the directions and potential use of FRET-FLIM imaging of lanthanide biosensor activity is given at the end of Chapter 4.