The objective of this thesis project was to develop an in vivo microprism preparation that provides long-term access to the brain for 2-photon laser scanning microscopy, and combine it with a head-fixed fear conditioning paradigm. The aim was to longitudinally study the activity patterns of L2/3 pyramidal and L1 5HT3aR expressing interneurons in the cingulate cortex in order to better understand their possible roles in the learning and expression of fear.
In vivo calcium imaging techniques combined with fear conditioning are potentially powerful methods to model and study neuronal and behavioural processes during fear. Previously this has contributed to the identification of disinhibitory microcircuit motifs, constituted by superficial inhibitory neurons that gate the excitation of pyramidal neurons in various cortical regions during fear conditioning. However, it is not clear if such microcircuits are possibly involved in the regulation of fear through the ACC, partly because a good preparation for high-resolution imaging of neuronal activity in the ACC was lacking.
We first wanted to provide a proof of principle for the feasibility and limits of a microprism-based chronic cranial window in mice. For this purpose, we established a surgical protocol for the microprism implantation and imaged diverse cortical elements in the anterior cingulate cortex. We achieved to obtain a clean optical access for months with a high resolution allowing the imaging of layer 1, layer 2/3 and layer 5 somatic calcium events, dendritic spines and axonal boutons. In addition, we performed behavioural controls to establish the safety and functional impact of such a surgery on the animals behavior. These results are described in part 1and 2 of the Results section and are in preparation for publication.
Next, we optimised a head fixed tone cued fear conditioning paradigm to adapt it to the 2- photon laser microscope set-up. Freezing and pupillometry are very reliably used behavioural read outs for animal and human fear respectively. Therefore, we chose to try and combine both methods to read out fear behaviour while the mouse is head restraint. This was challenging, in part because for monitoring pupil size in mice that are in a dark environment we needed to force a slight myosis. We resolved these issues using multiple cameras and an ultraviolet light source reducing the pupil size and allowing for pupil fluctuation tracking. These results are described in part 3 of the Results section.
Finally, using the above mentioned tools and a combination of transgenic animals and virus based transduction, we successfully expressed the calcium indicator GCaMP6 in a large population of layer 2/3 pyramidal neurons and layer 1 5HTaR expressing interneurons, but also specifically labelled neurogliaform cells and successfully performed longitudinal imaging during a fear recall protocol. We observed a large proportion of neuronal calcium responses triggered by the tone cue before conditioning and these responses were positively and negatively modulated by conditioning. More specifically among the layer 1 interneurons, the NGC subpopulation showed a net increase of their tone-evoked responses. These results are described in part 3 and 4 of the Results section and have been submitted for publication (Bimodal modulation of L1 interneuron activity in anterior cingulate cortex during fear conditioning, authors: Gi
