|dc.description.abstract||Episodic memories are encoded by combining features of the environment, objects, context and emotional salience within the hippocampus and its associated cortical structures. Consolidation of episodic memories is hypothesised to occur by transfer of memories from the hippocampal formation to the neocortex. The entorhinal cortex, which is a main relay between the hippocampus and neocortex, is thought to associate pairs of features that contribute to episodic memories. The entorhinal cortex can be divided into two areas: the medial entorhinal cortex (MEC) is involved in spatial and contextual processing; and the lateral entorhinal cortex (LEC) encodes objects in their locations, olfactory information and, possibly, emotional salience. Despite their differing roles in the memory formation, their circuitry has been assumed to be similar, producing parallel pathways that interact with the hippocampus. In this thesis, I will present experiments that compare the functional organisation of the MEC and LEC in mice and how distinct circuits may be involved in how emotional signals are encoding for the formation of episodic memories.
In the first set of experiments, I compare the cytoarchitecture and connectivity of deep layers in the MEC and LEC by using immunohistochemistry, retrograde virus tracing and whole-cell electrophysiology recordings from neurons in mouse brain sections. I find that the protein expression in deep layers of the MEC and LEC are broadly similar. However, differences in interneuronal densities, output projection patterns and morphological and intrinsic properties may reflect their different functions and could in future be manipulated to investigate their distinct roles in memory encoding and consolidation.
My investigations of connectivity led me to ask whether there are differences between the afferent inputs to LEC and MEC. I discovered that projections to MEC and LEC from the baso-amygdala (BA) originate from different neuronal populations in the BA and have distinct laminar targets. The BA contains the basolateral (BLA), basomedial amygdala (BMA) and the lateral amygdala (LA) and is involved in associative memory encoding and adding emotional valence to memories. Retrograde virus tracing shows that the separate neuronal populations in the BLA project to the MEC and LEC whereas only the LEC receives strong input from the BMA and LA. Intersectional viral approaches were performed to replicate retro-virus results and found that LEC projecting cells in the BA did not embed axon collaterals in MEC but did so in other brain areas such a ventral CA1, parasubiculum and presubiculum.
Projections from the BLA to the MEC are mainly found in L1 and L5a, suggesting that axons synapse onto apical dendrites of innervated MEC cells. Optogenetic stimulation of BA axons in the MEC and LEC produced a variety of innervated cell types. The majority of cells innervated in the MEC were in L2 and L5a, whereas the cells innervated in the LEC were in L2a and L3. Final experiments were applied in order to test whether cre-driver mouse lines could give genetic access to BA subnuclei that project to the MEC or LEC. This is especially important to delineate the separate functions of the BA subnuclei for future functional studies.
In summary, I find that although intrinsic connectivity is moderately similar, afferent inputs differ substantially between the MEC and LEC. Amygdala inputs project to separate cell types in the MEC and LEC, suggesting distinct emotional signals are passed onto the EC, before projecting to dorsal hippocampus. The contrast in projections from the amygdala to the LEC compared to the MEC suggests that the LEC integrates more emotional and contextual information for the encoding of episodic memories, whereas MEC projections could be involved in precise encoding of spatially relevant contextual information.||en