Ultrastructural organization of normal and transplanted rat fascia dentata: II. A quantitative analysis of the synaptic organization of intracerebral and intraocular grafts.

As part of an ultrastructural analysis of the normal rat fascia dentata and intracerebral and intraocular dentate transplants the synapses in the dentate molecular layer were quantified. Hippocampal and dentate tissue from 21-day-old rat embryos were grafted into the brain of developing and adult rats and to the anterior eye chamber of adult rats. After 100 or 200 days of survival the recipient rat brains and the recipient eyes were processed for electron microscopy, and the graft dentate molecular layer with the adjacent granule cell layer selected for ultrastructural analysis. Tissue from the dentate molecular layer of normal adult rats served as controls. The dentate synapses were classified as asymmetric (Gray's type 1) or symmetric (Gray's type 2), and according to the postsynaptic element (cell body, dendritic shaft, dendritic spine). The spine synapses were further classified into simple and complex types according to the spine-terminal configuration. Also, the length of synaptic contacts of the individual synaptic types was measured in some grafts, just as the percentage of the cross sectional area of the neuropil covered by blood vessels. The results showed that the synaptic density, expressed as number per unit area of neuropil, to a large extent was the same within the different parts of the normal dentate molecular layer. Compared with this the synaptic density was reduced with 16.4% in dentate molecular layer of the intracerebral graft, primarily because of a 17.6% reduction of simple synapses on dendritic spines and almost halving of the symmetric synapses on dendritic shafts. The synaptic density was independent of the age of the recipient, the intracerebral location of the graft, and the survival time. Although the synaptic length of some of the individual synaptic types increased, this did not compensate for the loss of synapses. In the intraocular grafts the synaptic density was lower than in the intracerebral grafts. Despite the reduced synaptic density, which mainly involved two synaptic types, we conclude that grafted dentate granule cells can develop a remarkably normal, ultrastructural synaptic organization even in the absence of major afferent inputs. This outcome must accordingly be achieved by reorganization of the available intrinsic afferents.