Genetic analysis of emotional memory in AMPA and NMDA receptor mutant mice

Abstract

My goal was to study the role of AMPA and NMDA receptors in fear memory using genetic tools. In particular, I aimed to unravel the function of these glutamate receptors in the long-term retrieval of passive avoidance and fear conditioning, both in the brain in general and in the Basolateral Amygdala (BLA). Through the analysis of GluA3 knockout mice, I could show that the AMPA receptor subunit GluA3 is not necessary for the initial phases of cued fear learning, but is required for the normal attenuation of fear to a remote negative event, both in the passive-avoidance and fear-conditioning paradigms. In contrast, GluA1 is essential for the acquisition and short-term retrieval of cue- and context-induced fear. Two gene-targeted mouse lines were used, one with a global GluA1 knockout, and a ”loss of function“ mutant GluA1 (Q586R). Both lines showed impaired acquisition and reduced cue- and context-induced retrieval 24 h and 48 h after fear conditioning, supporting the hypothesis of involvement of GluA1-containing AMPA receptors in learning of emotional associations. Moreover, in a similar way as for GluA3, GluA1 is also required for the normal decrease of fear to remote events in the passive avoidance test, suggesting that both subunits play an important role in the normal destabilization of older memories such as those that no longer provide an advantage in survival. To study AMPA and NMDA receptor function specifically in the retrieval phase of fear conditioning, I used inducible recombinant adeno-associated virus (rAAV)-mediated gene manipulations in the amygdala. The efforts to generate a BLA-specific promoter for rAAV failed. Therefore a doxycycline-inducible neuron-specific system was used for synaptic silencing of neurons in the amygdala, and to inactivate NMDA and GluA1-containing AMPA receptors in the BLA. Altogether, the analysis of rAAV-injected mice provided strong evidence that retrieval of cued fear memory is dependent on the chronic expression of NMDA receptors in the BLA, whereas the contribution of the GluA1-containing AMPA receptors remains to be confirmed. In the second part of this thesis, I generated and analyzed a new rAAV vector for tissue-specific gene delivery. A new endogenous promoter was cloned from the murine lynx2 gene, a member of the Ly-6/neurotoxin superfamily, for overexpression of genes in dentate gyrus granule cells. This virus can be used to further restrict rAAV-mediated targeting to certain groups of cells.