Anatomy and electrophysiology of fast central synapses lead to a structural model for long-term potentiation.

Detailed knowledge of the anatomy of central synapses is essential to the interpretation of the vast quantity of electrophysiological findings that have been published in recent years. When their function is considered, it is not surprising that, in both anatomy and electrophysiology, fast central synapses show important differences to the neuromuscular junction. This review concentrates on the detailed anatomy of the common excitatory synapses that impinge on dendritic spines, but also refers to other glutamatergic and GABAergic synapses. This information is brought together with present knowledge of the electrophysiology of fast neurotransmission in the brain. Various types of evidence are outlined, explaining why it is now widely accepted that release of transmitter from a single vesicle virtually saturates the small number of receptors available on the postsynaptic membrane of central synapses. Finally, the anatomic literature suggests that a particular type of spine synapse, which electron microscopy reveals to have a perforated active zone, may represent a synapse with high efficacy. This suggestion is shown to be completely compatible with the electrophysiological data, and a model is presented that shows that all the apparently conflicting data in the field of long-term potentiation could be compatible. This stresses the need for cooperative collaboration between laboratories that have apparently conflicting findings.