Living cell super resolution microscopy of intact cardiac nanodomains in a GFP-reporter-mouse-model
by Johannes Moritz Riebeling
Date of Examination:2025-06-03
Date of issue:2025-05-26
Advisor:Stephan Elmar Lehnart
Referee:Prof. Dr. Stephan E. Lehnart
Referee:Prof. Dr. Stefan Jakobs
Persistent Address:
http://dx.doi.org/10.53846/goediss-11265
Files in this item
Name:Riebeling_Johannes Moritz_Dissertation_sub.pdf
Size:2.64Mb
Format:PDF
This file will be freely accessible after 2026-06-02.
Abstract
English
The aim of this thesis is to validate the first application of RESOLFT imaging in a transgenic mouse model in order to confirm a novel transgenic-microscopic approach and from there on address relevant questions in cardiac research. Based on a knock-in of rsEFP2 as a molecular tag to Caveolin 3 (CAV3), isolated murine ventricular myocytes (VMs) shall get assessable for living cell RESOLFT microscopy. After validating the novel transgenic-microscopic approach on multiple levels, from cardiac function to nanoscopic distribution of reporter proteins, questions of CAV3 membrane physiology were addressed. This study in detail focused on the interplay of CAV3 with lipids, the spatio-temporal dynamics of CAV3 in regard to lateral mobility of CAV3-clusters, cellular contraction, electrical stimulation and receptor interactions. Preliminary analysis of cardiac function by echocardiography indicated no differences in EF, heart weight to body weight ratio, anterior wall thickness, and End-diastolic LV-area among heterozygous (het/cre) mice, driver (wt/cre) mice and wt mice. These results, in combination with cell size measurements, showed no systemic features of a cardiac failure phenotype in reporter animals. Microscopic phenotyping of the reporter construct was a three-step process: a) analysis of intracellular CAV3 distribution in reporter mice by confocal and STED immunofluorescence (STED-IF), b) relative comparison of cluster sizes of CAV3 signals in het/cre, cre, and wt mice, and c) analysis of trafficking and oligomerization of CAV3 in het/cre and hom/cre mice with simultaneous confirmation of antibody specificity by dual color STED-IF. The intracellular distribution of CAV3 showed a regular pattern in het/cre and wt mice, while het/cre mice further showed typical subsurface nanoscopic patterns in STED-IF. Deeper internal CAV3-STED-IF-signals in het/cre- VMs followed a recti-linear-distribution, resembling TATS-architecture, arguing against accumulation of CAV3 in endosomes or other organelles. Additionally, the size distribution of CAV3 signals in STED-IF samples revealed specific differences between het/cre, wt/cre, and wt animals after statistical comparison of the spatial density of small, intermediate, and large clusters. For smaller clusters, no difference was observed between all 3 genotypes while the spatial density of intermediate clusters reduced in wt/cre mice. Importantly the spatial density of all CAV3 cluster subtypes of het/cre mice was almost indistinguishable from wt mice, which was consistent with the described healthy cardiac phenotype characterized by echocardiographic measurements. Interestingly, homozygous mice showed distinct trafficking and targeting deficits: a vesicular surface pattern of CAV3 signals instead of the complex pattern described above in het/cre, as well as bright unpaired GFP signals occurred, indicating lysosomal degradation or termination processes. This hypothesis was strengthened by confocal images showing colocalization of CAV3 signals with p115-labeled Golgi signals in homozygous mice but not in heterozygous mice. This suggests defective oligomerization of rsEGFP2-CAV3 in hom/cre mice, which is key for Golgi exit of the reporter construct. There it seems likely that the trafficking deficiencies are rescued by molecular interaction and hetero oligomerization with endogenous CAV3 in het/cre mice. Accordingly, CAV3 hetero-oligomers in het/cre-mice showed trafficking and cluster formation almost indistinguishable from wt CAV3 alone justifying the reporter protein approach presented in this thesis. From there living cell RESOLFT microscopy could readily reproduce above-described signal patterns as from anti-CAV3 IF-STED imaging, including small single domains of 80-100 nm diameter, ring structures with diameters of up to 200 nm and large assemblies at the subsurface of living het/cre VMs. However, it must be considered that imaging of TATS compartments could not be realized because signal noise increased when imaging at deeper intracellular levels, most likely due to mitochondrial autofluorescence. Spatiotemporal dynamics of CAV3 in living primary cardiomyocytes could be visualized by RESOFLT microscopy timelapse recordings at frame rates up to 9 µm2/min without obvious signs of major cellular damage due to the applied laser intensities. Interestingly imaging of singular vesiculation and transport processes at the time scale of minutes became feasible by applied microscopy protocols. There, negligible xy-motions and no obvious z-plane fluctuations were observed. Z-level monitoring by wheat germ agglutinin (WGA) -labeling supported those findings, whereby WGA in part served as a z-plane indicator as even small fluctuations of WGA signals itself occurred. The stability of CAV3 domains was also confirmed during electrical stimulation, as temporal plasticity before and after stimulation showed no obvious fluctuations in CAV3 signals. Nevertheless, minor fluctuations occurred at the edges of large CAV3 signals. It is currently unclear whether these fluctuations are due to CAV3 movements or to z-plane undulations. Also, a novel combination of dual color STED-RESOLFT super-resolution microscopy was developed based on monitoring the distribution of cholesterol and CAV3 in living primary VMs. For this purpose, further application development towards combined RESOLFT-STED microscopy was achieved. From there, the feasibility of monitoring cholesterol depletion on the level of individual CAV3-domains in living cells was proven. Dissociation of large clusters into a vesicular pattern was observed for the first time in living cells. By implementing a novel imaging protocol, based on sequential frame acquisition of fluorescently labeled Transferrin by STED and rsEGFP2-CAV3 by RESOLFT imaging, lateral interaction of CAV3 domains with transferrin (TF)-transferrin-receptor-1 (TfR-1) complexes could be demonstrated. Since TF-TFR-1 complexes showed increasing lateral association with CAV3 complexes, docking and regulation of TF-TFR-1 complexes at the rims of CAV3 domains can be suggested. Since this work demonstrates a blueprint for validation of reporter constructs by RESOLFT microscopy the described microscopy technique should be considered, when future experiments address questions of reporter validation in various cell-systems, as long as RESOLFT-suitable fluorophores are used as protein-tags. Especially in adipocytes, cardiac and skeletal muscle cells, the experimental system seems to be an interesting solution for the characterization of caveolins and caveolae, provided that the cell-specific cre-expression is respected.
Keywords: Caveolin 3, STED, RESOLFT, superresolution microscopy, cardiomyocytes, rsEGP2
