Jörg Schnitzbauer

Fluorescence Microscopy of Single Molecules: Advances and Applications

ZNN, 2. OG, Seminarraum

21.01.2015, 10:00

Fluorescence microscopy is a powerful concept employed by numerous methods in biomedical research. Recent developments even enable the localization of single molecules with a precision of only a few nanometers. Single molecule localizations can be used to generate temporal molecule trajectories or super-resolution images as in PALM or STORM. During my talk, I present three applications and technical advances in the field of single molecule localization.

First, I show the technical effort to improve three-dimensional localization of single molecules. While two-dimensional localization is fairly straightforward, common axial localization schemes suffer from one of two issues: (1) axial localization is substantially worse than lateral; or (2) complex instrumentation limits the feasibility of routine experiments. Here, I show how axial localization precision can be improved by placing a mirror on top of the sample. Thereby, fluorescence is collected from two sides of the sample with a single objective. Self-interference of direct and reflected fluorescence thus enable high axial localization precision with relatively simple instrumentation.

Second, I demonstrate the power of molecular tracking with the movement analysis of single genomic loci in living human cells. So far, studying molecular dynamics of the genome has been limited by the lack of appropriate labeling methods. However, the specificity and live cell compatibility. Here, I extracted diffusion parameters from time-lapse CRISPR images and discuss possible implications for chromosome dynamics. For example, I show that the diffusion speed of telomeres increases after disrupting their protective cap which might indicate the activation of DNA repair machinery. 

Finally, I derive and discuss new analysis theory for localization-based super-resolution microscopy. Biological interpretation of single molecule images is still an underdeveloped area, because image information is stored as a set of coordinates rather than on the conventional pixel grid. Consequently, traditional image analysis algorithms have only limited capabilities for this new type of microscopy. I will show that the derived analysis methods can measure diffusion of membrane lipids, co-localization of Golgi associated proteins and improve the quality of images by rotation alignment of molecular structures.