TY  - GEN
ID  - heidok23026
N2  - Single molecule localization microscopy (SMLM) has been established as a powerful technique to investigate biological samples with a resolution well below the diffraction limit of light microscopy. However, the limits of these new techniques have not yet been probed. In particular, the localization of multiple targets using different fluorescent dyes is difficult due to chromatic aberrations. Hence, the aim of this project was to build a microscope based on the principle of stochastic optical reconstruction microscopy (STORM), which can be used to simultaneously acquire 3D aberration-free images of two fluorophore-labeled structures, establish an automated imaging workflow and systematically probe the possibilities and limits of SMLM. 
The microscope was built and used to record images of F-actin and synaptophysin, a protein enriched at synaptic vesicles, in the rat?s calyx of Held, a glutamatergic model synapse. Contrary to our expectations, the experiment resolved neither actin filaments nor clearly distinct synaptic vesicles. To validate the experimental results and to define the limits of SMLM more precisely, a realistic simulation tool for SMLM experiments called SuReSim (super-resolution simulation) was developed. In SuReSim, 3D models of ground truth structures (e.g. filaments or organelles) can be imported and the expected outcome of a SMLM experiment can be simulated taking into consideration a multitude of parameters. Two options for the resulting output are available: either a list of simulated localizations, compatible with SMLM-specific 3D viewers or a realistic Tiff stack resembling raw data as recorded during a SMLM measurement. SuReSim was used to simulate realistic F-actin and synaptophysin imaging results based on model structures derived from electron microscopy. The simulations confirmed the experimental results, no distinct structures could be resolved. Additional simulations were performed on other biological samples to find the limitations of SMLM microscopy. The packing density of the structure, the density of binding sites, the localization precision, the labeling efficiency and the label length were identified as limiting factors for SMLM measurements.
Due to the many factors contributing to a SMLM measurement, it is not possible to find universal limits, but only certain sets of parameters that are necessary for a successful experiment. These limits have to be identified individually for each target structure. SuReSim can help to make the decision, whether or not SMLM is the right tool to address a specific research question.
A1  - Herrmannsdörfer, Frank
UR  - https://archiv.ub.uni-heidelberg.de/volltextserver/23026/
AV  - public
Y1  - 2017///
TI  - Exploring the Limits of Single Molecule Localization Microscopy using Realistic Simulations with a Focus on Presynaptic Nerve Terminals
ER  -