We develop a method for selecting meaningful learning strategies based solely on the behavioral data of a single individual in a learning experiment. We use simple Activity-Credit Assignment algorithms to model the different strategies and couple them with a novel hold-out statistical selection method. Application on rat behavioral data in a continuous T-maze task reveals a particular learning strategy that consists in chunking the paths used by the animal. Neuronal data collected in the dorsomedial striatum confirm this strategy.

SCN1A (NaV1.1 sodium channel) mutations cause Dravet syndrome (DS) and GEFS+ (which is in general milder), and are risk factors in other epilepsies. Phenotypic variability limits precision medicine in epilepsy, and it is important to identify factors that set phenotype severity and their mechanisms. It is not yet clear whether SCN1A mutations are necessary for the development of severe phenotypes or just for promoting seizures. A relevant example is the pleiotropic R1648H mutation that can cause either mild GEFS+ or severe DS. We used a R1648H knock-in mouse model (Scn1aRH/+) with mild/asymptomatic phenotype to dissociate the effects of seizures and of the mutation per se. The induction of short repeated seizures, at the age of disease onset for Scn1a mouse models (P21), had no effect in WT mice, but transformed the mild/asymptomatic phenotype of Scn1aRH/+ mice into a severe DS-like phenotype, including frequent spontaneous seizures and cognitive/behavioral deficits. In these mice, we found no major modifications in cytoarchitecture or neuronal death, but increased excitability of hippocampal granule cells, consistent with a pathological remodeling. Therefore, we demonstrate for our model that an SCN1A mutation is a prerequisite for a long term deleterious effect of seizures on the brain, indicating a clear interaction between seizures and the mutation for the development of a severe phenotype generated by pathological remodeling. Applied to humans, this result suggests that genetic alterations, even if mild per se, may increase the risk of second hits to develop severe phenotypes.

The amyloid precursor protein (APP) harbors physiological roles at synapses and is central to Alzheimer's disease (AD) pathogenesis. Evidence suggests that APP intracellular domain (AICD) could regulate synapse function, but the underlying molecular mechanisms remain unknown. We addressed AICD actions at synapses, per se, combining in vivo AICD expression, ex vivo AICD delivery or APP knock-down by in utero electroporation of shRNAs with whole-cell electrophysiology. We report a critical physiological role of AICD in controlling GluN2B-containing NMDA receptors (NMDARs) at immature excitatory synapses, via a transcription-dependent mechanism. We further show that AICD increase in mature neurons, as reported in AD, alters synaptic NMDAR composition to an immature-like GluN2B-rich profile. This disrupts synaptic signal integration, via over-activation of SK channels, and synapse plasticity, phenotypes rescued by GluN2B antagonism. We provide a new physiological role for AICD, which becomes pathological upon AICD increase in mature neurons. Thus, AICD could contribute to AD synaptic failure.

There is a growing consensus that Alzheimer's disease (AD) involves failure of the homeostatic machinery, which underlies the firing stability of neural circuits. What are the culprits leading to neuron firing instability? The amyloid precursor protein (APP) is central to AD pathogenesis, and we recently showed that its intracellular domain (AICD) could modify synaptic signal integration. We now hypothesize that AICD modifies neuron firing activity, thus contributing to the disruption of memory processes. Using cellular, electrophysiological, and behavioral techniques, we show that pathological AICD levels weaken CA1 neuron firing activity through a gene-transcription-dependent mechanism. Furthermore, increased AICD production in hippocampal neurons modifies oscillatory activity, specifically in the γ-frequency range, and disrupts spatial memory task. Collectively, our data suggest that AICD pathological levels, observed in AD mouse models and in human patients, might contribute to progressive neuron homeostatic failure, driving the shift from normal aging to AD.