Nanoscopy inside living brain slices
by Nicolai Thomas Urban
Date of Examination:2012-11-01
Date of issue:2014-10-30
Advisor:Prof. Dr. Stefan Hell
Referee:Prof. Dr. Stefan Hell
Referee:Prof. Dr. Jörg Enderlein
Referee:Prof. Dr. Erwin Neher
Persistent Address:
http://dx.doi.org/10.53846/goediss-4761
Files in this item
Name:Nicolai_T_Urban_Dissertation.pdf
Size:32.6Mb
Format:PDF
The following license files are associated with this item:
Abstract
English
In order to understand how memory is processed and stored in the brain, it would be helpful to observe the ongoing memory processes in action. This is no easy task, however, if it is to be done at the synaptic level. It requires high spatial resolution (preferably in all three dimensions) to observe the synapses in nanoscopic detail, and a fast time resolution (seconds or faster) to observe rapid dynamic processes. On the other hand the method must be noninvasive, so as to not disturb the natural dynamics of the brain and to permit observations over hours or longer without damaging the sensitive tissue. Finally, it must be able to image deep inside living brain tissue, in order to observe synapses which are still part of an intact, functioning neural network. Until now it was either possible to image living neurons (using fluorescence confocal or two-photon microscopy), but limited in spatial resolution by diffraction, or to image with nanometer resolution (using electron microscopy), but restricted to fixed, and therefore dead, tissue. In this dissertation we show that the optical super-resolution techniques of STED (STimulated Emission Depletion) and RESOLFT (REversible Saturable OpticaL Fluorescent Transition) nanoscopy are well suited to the task, and we demonstrate this by, for the first time, observing morphological changes of postsynaptic structures belonging to healthy neurons embedded deep inside the intact neural network of a living organotypic hippocampal brain slice. First, we showed that simple aberration correcting techniques are sufficient to preserve the 60 nm spatial resolution of our STED nanoscope even in depths of >90 µm inside living hippocampal brain slices. With this, we were able to measure the neck diameters of dendritic spines belonging to healthy CA1 pyramidal neurons with high precision, without danger of possible fixation artifacts. We observed that dendritic spines can be highly dynamic, exhibiting both movement and morphological changes often smaller than the diffraction limit. This motility could be fast (seconds), or occur more gradually over hours. Motility was evident at room temperatures, but both frequency and magnitude increased upon heating to physiological temperatures. Furthermore, we observed the distribution of the cytoskeletal protein actin over time, as well as the effects of the actin depolymerizing drug Latrunculin A on the shape of dendritic spines. We demonstrate the strength of this approach by monitoring spine neck diameters after stimulating neurons using a long-term potentiation (LTP) protocol designed to elicit synaptic strengthening. On average the spine neck diameters of stimulated neurons increased by 30% and these changes remained stable for hours. Neurons in unstimulated slices showed no such behavior. Using RESOLFT nanoscopy we could observe the same spontaneous and stimulated dynamics as with STED, although using illumination intensities that were five to six orders of magnitude less than for STED. The low light levels inherent to this RESOLFT approach allowed us to image entire stretches of spiny dendrites continuously for hours, unhindered by bleaching and devoid of any signs of light-induced effects or phototoxic stress. This also alleviated concerns as to whether or not the imaging process itself disturbed or influenced the sensitive synaptic structures. By combining two differently patterned de-excitation beams, we achieved a threefold isotropic resolution increase over diffraction limited confocal microscopy. In conclusion, we demonstrate how optical nanoscopy techniques can be used to examine hitherto unobservable dynamic brain phenomena as they occur deep inside intact brain tissue.
Keywords: STED; RESOLFT; nanoscopy; super-resolution; microscopy; aberration correction; spherical aberrations; deep tissue imaging; penetration depth; RSFP; time-lapse; dendritic spines; dendrites; actin; cytoskeleton; LTP; long-term potentiation; synaptic plasticity; morphological changes; postsynaptic activity; motility; hippocampus; CA1; pyramidal neuron; living brain; organotypic brain slice