Probing Lipid Diffusion in Curved and Planar Membranes with Fluorescence Microscopy
by Jan Thiart
Date of Examination:2017-08-31
Date of issue:2017-09-05
Advisor:Prof. Dr. Jörg Enderlein
Referee:Prof. Dr. Jörg Enderlein
Referee:Prof. Dr. Claudia Steinem
Persistent Address:
http://dx.doi.org/10.53846/goediss-6463
Files in this item
Name:phd_thesis_final_ediss.pdf
Size:3.17Mb
Format:PDF
The following license files are associated with this item:
Abstract
English
Diffusion is the most important transport mechanism in biological membranes and essential for processes such as signalling or trafficking. Many different techniques have given insight into this matter, most of which are based on fluorescence microscopy. Applying these to synaptic vesicles or nanoscopic membrane domains, which are much smaller than the diffraction-limited resolution of a light microscope, is an ambitious task. In this thesis, two methods based on fluorescence correlation spectroscopy (FCS) are presented which circumvent the resolution limit and enable diffusion coefficient estimations in vesicles less than 200 nm in diameter. At this scale, the influence of membrane curvature on viscosity, diffusion speed, or lipid composition becomes dominant. The first approach, dynamicMIET, exploits the interaction energy transfer of a fluorescent lipid in close proximity to a thin metal sheet. The resulting fluctuation in fluorescence intensity strongly depends on their distance to each other, which can in turn be obtained by FCS, making it possible to extract the membrane diffusion coefficient of a surface-tethered vesicle of arbitrary size. A robust and highly specific binding assay has been developed, but the organic dyes used for labelling were neither bright nor stable enough to obtain proper correlation curves. A detailed analysis revealed drastically increased dark state transitions and photo-bleaching of the lipid-conjugated dyes compared to their free counterparts. The second technique is based on measuring the polarisation-resolved rotational diffusion of a fluorescent lipid within a vesicle bilayer. By fixing the dye-to-membrane orientation, rotational and translational diffusion components can be measured and extracted separately. A 3D diffusion model incorporating the vesicle size distribution fits the correlation curves very well, but the obtained diffusion coefficients are biased towards higher values, especially for larger liposomes. However, the fluorophore used in these experiments was found to have excellent photo-physical characteristics which could help to resolve the issues encountered in the dynamicMIET measurements. To handle the data evaluation for many of the control experiments, I developed TrackNTrace, an open-source framework for fluorescence microscopy image analysis. TNT was originally designed as a localisation microscopy and particle tracking tool, but is extendible through a simple plugin system. It provides many state-of-the-art implementations of important algorithms and is aimed at novices as well as experienced researchers. An extensive visual feedback mechanism allows inspecting the program's output at all times, facilitating parameter optimisation and error recognition. These concepts were validated by comparing TrackNTrace against similar programs. The software has been a great help in analysing many of the experiments presented in this thesis and will hopefully turn out to be similarly beneficial for other scientists.
Keywords: Fluorescence; Microscopy; Membrane; Diffusion; Lipid bilayer; Particle Tracking; MIET; FCS