Stochastic Induction of Long-Term Potentiation and Long-Term Depression

Area of Science:

• Neuroscience
• Computational Biology
• Synaptic Plasticity

Background:

• Long-term depression (LTD) and long-term potentiation (LTP) are key synaptic plasticity mechanisms. LTD involves protein kinase C, phospholipase A2, and ERK pathways, reducing AMPA receptors. LTP requires protein phosphatases and increases AMPA receptors.
• Previous computational models showed LTD is probabilistic and bistable at single synapses.

Purpose of the Study:

• To expand a stochastic computational model to simulate long-term potentiation (LTP).
• To investigate the distinct dynamics of LTD and LTP at single synapses.
• To unify the understanding of macroscopic LTD and LTP properties from single-synapse dynamics.

Main Methods:

• Developed and expanded a stochastic computational model of synaptic plasticity signaling pathways.
• Simulated both long-term depression (LTD) and long-term potentiation (LTP) at the single synapse level.
• Analyzed the influence of intracellular calcium ion concentration ([Ca(2+)]) and signaling network states on synaptic modifications.

Main Results:

• In single synapses, LTD was found to be bistable, whereas LTP exhibited gradual dynamics.
• Both LTD and LTP were stochastically induced by calcium ion concentration ([Ca(2+)]) changes.
• Calcium signal magnitudes and network states determined the probability of LTD/LTP and set dynamic macroscopic thresholds, following an inverse BCM rule or sigmoidal function.

Conclusions:

• The developed model provides a unifying mechanism for LTD and LTP.
• Macroscopic synaptic plasticity properties emerge from the stochastic dynamics at the single-synapse level.
• This framework explains how calcium signals and network states collectively regulate synaptic modifications.