The discovery of GluA3-dependent synaptic plasticity

AMPA receptors (AMPARs) are responsible for fast excitatory synaptic transmission. GluA1-containing AMPARs have been extensively studied and play a key role in several forms of synaptic plasticity and memory. In contrast, GluA3-containing AMPARs have historically been ignored because they have remained invisible at the cell physiological level. This thesis aims to "put GluA3 on the map" by studying its role in synaptic plasticity and behavior in the hippocampus and cerebellum, and its relationship with amyloid β (Aβ). Firstly, we discover that GluA3-containing AMPARs are present at synapses in large amounts, but invisible because they are closed/inactive/dormant under basal conditions, hence being electrically silent. However, when intracellular cAMP rises, GluA3 becomes functional, leading to a powerful synaptic potentiation. Fear, through an increase in norepinephrine, is a trigger for this novel synaptic plasticity. We further show that this GluA3-plasticity plays no role in memory formation. Secondly, we find that GluA3-dependent synaptic plasticity also occurs in the cerebellum. Here, GluA3-containing AMPARs are responsible for long-term potentiation at the parallel fiber to Purkinje cell synapse and adaptation of the vestibulo-ocular reflex. Thirdly, we return to the hippocampus, focusing in Alzheimer's disease (AD). Experiments in mouse models of AD have shown that soluble oligomeric Aβ triggers synaptic weakening, loss of synapses and reduced synaptic plasticity. We find that all these Aβ-driven effects depend of AMPAR subunit GluA3 and that memory deficits in our AD mouse model are absent in a GluA3-deficient background, suggesting that the expression of Aβ-mediated synaptic and cognitive deficits require GluA3.