Acetylcholine-sensitive control of long-term synaptic potentiation in hippocampal CA3 neurons.

Abstract

A Hebbian form of long-term potentiation (LTP) is believed to be the basis of memory storage at CA3 recurrent synapses. Abnormalities in CA3 intrinsic connectivity have been related to memory deficits in a variety of neurological disorders. Despite the promise of computational modeling for illuminating the pathogenic implication of connectivity changes, common Hebbian-based models with preset structural topologies fall short in this regard. Here, I introduce a structure-independent approach to modeling CA3 network focusing on how LTP shapes CA3 functional connectivity. Network simulations demonstrate that only a small fraction of the active synapses should be potentiated onto engram-bearing cells for reaching CA3 optimal performance, and that this fraction should be actively regulated at the single-cell level to maintain precise control over excitatory inputs to and from overlapping engram cells. In light of these findings, I develop a theory suggesting that synaptic potentiation is regulated through extrinsic and intrinsic cellular mechanisms involving the cholinergic modulation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. The theory posits that the progressive increase of acetylcholine release during learning is commensurate with the rate of AMPA receptor trafficking during LTP development and that this dynamics is intended to conceal AMPA receptor potentiation and thereby to propel LTP progression until a target level of potentiation is attained at a restricted number of the active synapses. Functionally, this form of regulation allows encoding and retrieval dynamics to be dictated at the single-cell level and thereby acts at the cell-population level as a secondary mechanism complementing dentate gyrus-mediated pattern separation or compensating for possible deficiencies. Conversely, when this regulation fails, the strength of AMPA receptors and their variability across synapses change over time and lead to pathological states.