Abstract
Perforated synapses refer to a synaptic type found in the central nervous system. They are characterized by their large size and by a discontinuity of the postsynaptic density when viewed in transverse sections, and by a doughnut or horseshoe shape when viewed in en face views. Of recent morphological studies, one approach has followed their characteristics throughout development and maturity, while others have concentrated on their probable roles in activities including kindling, long-term potentiation, spatial working memory, differential rearing, and the functioning of neuroleptics. An assessment is made of the hypotheses and models that have proved determinative in the emergence of perforated synapses as being significant in synaptic plasticity. Their distribution and frequency are summarized, with emphasis on the importance of unbiased stereological procedures in their analysis. Using three-dimensional approaches various subtypes are recognized. Of these, a complex or fragmented subtype appears of especial significance in synaptic plasticity. Ideas regarding the life-cycle of perforated synapses are examined. The view that they originate from conventional, non-perforated synapses, enlarge, and subsequently split to give rise to a new generation of non-perforated synapses, is critically assessed. According to an alternative model, perforated and non-perforated synapses constitute separate populations from early in their development, each representing complementary forms of synaptic plasticity. An attempt is also made to discover whether synaptic studies on the human brain in normal aging and in Alzheimer's disease throw light on the role of perforated synapses in synaptic plasticity. The loss of synapses in Alzheimer's disease may include a loss of perforated synapses - of particular relevance for an understanding of certain neuropathological conditions.