A protein aggregation-dependent scaling-free synaptic facilitation rule for neural network simulations

The regulation of mRNA translation at synaptic level is believed to be fundamental in memory and learning at cellular level. A family of RNA binding proteins (RBPs) which emerged to be important during development and in adult neurons is the one of Cytoplasmic Polyadenylation Element Binding proteins (CPEBs). Drosophila Orb2 (homolog of vertebrate CPEB2 protein and of the neural isoform of Aplysia CPEB) has been found to be involved in the translation of plasticity-dependent mRNAs and has been associated to Long Term Memory (LTM). Orb2 protein presents two main isoforms, Orb2A and Orb2B, which form an activity induced amyloid-like functional aggregate, which is thought to be the translation-inducing state of the RBP. Here we present a two-states continuous differential model for Orb2A-Orb2B aggregation and we propose it, more generally, as a new synaptic facilitation rule for learning processes involving protein aggregation-dependent plasticity (PADP).