Songbirds learn the song from a tutor as human beings learn language. Due to this similarity, songbirds are used as an animal model to explore brain areas and networks associated with this complex behavior.
Hence, knowing the learning processes in a songbird brain will be helpful to better understand the learning and speech of language in human beings.
Several regions in a songbird brain are involved in its song learning and production as illustrated in Figure 1. Several brain regions are linked to the HVC area, which serves as the main premotor region for vocal production.
Figure 1. Schematic representation of the songbird brain with regions involved in song learning and song production. Injection was made in Area X.
This article discusses the analysis of connections within HVC on an ultrastructural level. By labeling cells projecting to the brain region ‘Area X’ with a tracer, it is possible to analyze their connectivity within HVC.
Correlative Light and Electron Microscopy (CLEM) is necessary for this application because it is important to identify the projection of cells in the HVC region through Fluorescence Light Microscopy (FLM) as well as the ultrastructure of the same cells via Scanning Electron Microscopy (SEM).
Area X of a zebra finch brain was traced by stereotaxic coordinates and inoculated with 0.5µl Alexa 488 dextran. After 5 days, a perfusion with a solution of 2% paraformaldehyde and 0.075% glutaraldehyde in phosphate buffer (0.1 M, pH 7,4) was carried out and then the brain was cut down into 60 µm thickness of sagittal sections.
After washing in cacodylate buffer (0.1 M, pH 7,4), and postfixing in 1.5% potassium ferrocyanide, 1% osmium tetroxide and in 1% osmium tetroxide for 40 min, the sections were finally postfixed in 1% uranyl acetate in distilled water for 40 min. The sections were then dehydrated and embedded in Durcupan ACM resin, followed by curing at 52°C for 48 h.
The area around HVC was resected, fixed to a blank resin block, and cut into ultrathin sections with a thickness of 60-90 nm. The sections were shifted onto cover slips coated with ITO, followed by staining with Reynold's lead citrate for 2 min and washing with ddH2O for three times for 30 sec.
The cover slip was put into the sample holder that is specially designed by Carl Zeiss for CLEM. It is possible to use this holder in LM and SEM to firmly fix the sample in the holder during the entire imaging process. The holder features three fiducial markers to enable very fast and semi-automatic calibration in the Shuttle & Find module of the AxioVision Software.
FLM of the sections was carried out with an Axio Imager.ZI from Carl Zeiss utilizing a 100x objective (EC Epiplan-Neofluar 100x/0.90 HD DIC) and a filter set with 470/40 nm excitation and 525/50 nm emission (Filter set 38HE). The microscope was then fitted with an AxioCam MRm from Carl Zeiss. After defining and selecting regions of interest (ROIs) in the fluorescence image, the sample was shifted to an SUPRA 40VP SEM from Carl Zeiss. The holder was then semi-automatically calibrated to locate the frame imaged in the FLM, followed by fine calibration to image the chosen ROIs at accuracy below 5 µm. Then, SEM imaging was performed at 1.5 kV acceleration voltage using the in-lens secondary electron detector.
A widefield FLM image of an ultrathin section from the HVC region is illustrated in Figure 2. Fluorescent spots represent cell compartments where the tracer is localized, thus identifying and selecting a neuron projecting from HVC to Area X as ROI.
Figure 2. FLM image of an ultrathin section with highlighted ROI.
The selected ROI is further demonstrated in Figure 3.
Figure 3. FLM (3a) and SEM (3b) images at ROI selected in Fig. 2.
The enlarged section of Figure 2 is shown in Figure 3a, while the SEM imaging of the structure of the selected neuron is illustrated in Figure 3b. The overlay of FLM and SEM images of this selected ROI is demonstrated in Figure 4, confirming that both microscope systems actually image the same ROI.
Figure 4. Overlay at ROI selected in Fig. 2.
The upper left part of the selected neuron is shown at a higher magnification in an overlay of FLM and SEM in Figure 5.
Figure 5. Higher magnification overlay within Fig. 4; SEM image clearly shows the structure of mitochondria (A), myelinated axons (B) or synaptic vesicles (C).
Rapid and reliable high-resolution context imaging of specifically connected neurons was possible with the Shuttle & Find interface for CLEM. After labeling with fluorescent tracers, the neurons are imaged in both FLM and SEM.
The overlay of both the information enables classification of neurons analyzed in the SEM based on data from LM, while SEM imaging of the same sections presents ultrastructural information.
Although this article demonstrates only one fluorescent color, it is possible to extend the method to multicolor tracing and imaging for characterizing the network by utilizing different colors for different types of neurons.
This information has been sourced, reviewed and adapted from materials provided by Carl Zeiss Microscopy GmbH.
For more information on this source, please visit Carl Zeiss Microscopy GmbH.