Perhaps the most fundamental and remarkable feature of the mammalian central nervous system is its ability to process and store large amounts of information. For many decades it has been postulated that the brain uses Ionglasting modifications of synaptic strength in critical neural circuits to accomplish this feat. One such activitydependent modification is long-term potentiation (LTP) in the hippocampus, a sustained increase in synaptic strength that is elicited by brief high frequency stimulation of excitatory afferents. Recent excitement about the phenomenon of LTP has arisen from three major sources. First, compelling evidence from lesion studies in higher primates, including humans, shows that the hippocampus is a critical component of a neural system that is required for the initial storage of certain forms of long-term memory (Squire and Zola-Morgan, 1991). Second, several properties of LTP make it an attractive cellular mechanism for information storage or memory (Bliss and Collingridge, 1993). Like memories, LTP can be generated rapidly and is strengthened with repetition. It exhibits input specificity; that is, LTP occurs only at synapses stimulated by afferent activity but not at adjacent synapses on the same postsynaptic cell. Input specificity presumbably dramatically increases the storage capacity within a neural circuit. Most importantly, LTP is associative; temporally pairing activity in a “weak” input (incapable of generating LTP by itself) with activation of a strong input (capable of eliciting LTP at adjacent synapses on the same postsynaptic cell) results in LTP of the weak input. This associative property, which is reminiscent of classical conditioning, can be considered a cellular analog of associative learning. Third, LTP is readily elicited in in vitro preparations of the hippocampus, and this makes it amenable to rigorous experimental manipulations. There are several different forms of LTP, but the majority of experimental work has focused on the LTP observed in hippocampal CA1 pyramidal cells. This minireview will provide an update on the cellular mechanisms of LTP and distinguish those mechanisms that are firmly established from those that remain contentious. The evidence connecting LTP to real learning and memory also will be reviewed briefly. The Induction of LTP: NMDA Receptors and Cap+ It is well accepted that the induction of LTP requires activation of postsynaptic N-methyl-o-aspartic acid (NMDA) receptors (a subtype of glutamate receptor) during postsynaptic depolarization, which is normally generated by high frequency afferent activity. This results in a rise in CaH concentration ([Ca%],), a necessary trigger for LTP. Figure 1 shows that during normal low frequency synaptic transmission, the excitatory neurotransmitter glutamate is
Read full abstract