Spike timing dependent plasticity of rat hippocampal and cortical synapses and control by muscarinic transmission

Abstract

Synaptic long-term potentiation (LTP) was suggested in the late 1940s by Donald Hebb as a candidate mechanism for learning and memory in neuronal networks like those found in the central nervous system of mammals, but has been demonstrated experimentally only in the early 1970s (Bliss and L0m0, 1973). LTP of excitatory postsynaptic potentials (EPSPs) in hippocampal dentate gyms granule cells was observed after high frequency stimulation (12-20 Hz) of perforant path fibers in anaesthetized rabbits. It was recognized quite early that coincident pre- and postsynaptic activity was crucial for successful induction of homosynaptic LTP (Gustafsson, et al., 1987). That LTP induction generally required activation of NMDA receptors became particularly intriguing after the characterization of the NMDA receptor channel as a coincidence detector of the transmitter glutamate and of postsynaptic depolarization that is required to remove a blocking Mg2+ ion from the outer channel mouth. This coincidence detection allows association of different synaptic inputs by charge integration on the postsynaptic membrane after rapid spread of charge. It is now generally accepted that Ca2+ influx through the NMDA receptor channel is crucial for induction of most forms of LTP, presumably because of a close functional coupling of NMDA receptor channels on dendritic spines to Ca2+-dependent processes that are required for the expression of LTP (Regehr and Tank, 1990, Miiller and Connor, 1991, Perkel, et al., 1993, Yuste and Denk, 1995). Ca2+-influx through voltage activated Ca2+-channels appears to contribute only to a much lesser extent (Koester and Sakmann, 1998, Sabatini, et al., 2002), but can be sufficient to induce LTP by high frequency stimulation (Grover and Teyler, 1990).