Stereological analysis of synaptogenesis in the molecular layer of piriform cortex in the prenatal rat.

The development of synapses in the molecular layer of the rat piriform cortex was studied at embryonic days 15, 17, 19, and 21. The present study has sought to extend past studies of synaptogenesis by identifying not only changes in numbers of synapses, but also changes in numbers of potential precursors of synapses. A stereological method (Cruz-Orive, '80) was used to make volumetric estimations of the numbers of synapses, axonal puncta, vesicle-associated puncta, and unapposed postsynaptic specializations. This stereological method was preferred to other morphometric methods because it is not influenced by changes in the size, shape, or orientation of the structures of interest. This was considered important since such changes might be expected during development. Large numbers of unapposed axonal specializations (axonal puncta and vesicle-associated puncta) were found in all three sublaminae (lateral olfactory tract, Ia, and Ib) at all ages. The numerical density (number per unit volume of neuropil) and relative frequency of these structures changed significantly with time. In all three sublaminae, these changes were associated with changes in the number of synapses, although the numerical density and relative proportions varied between the sublaminae. These results suggested that axonal puncta could accumulate vesicles, thus becoming vesicle-associated puncta, and that vesicle-associated puncta could contact dendrites, thus forming synapses. In contrast, the numerical density of lone postsynaptic specializations remained low and no significant changes in their relative proportion in the population were found. This suggested that although lone postsynaptic sites were observed, they did not appear to play a major role in synaptogenesis in this region of the cortex. In addition to documenting developmental differences between the three sublaminae in the molecular layer, the results support a synaptogenic hypothesis in which the axon can form surface specializations that appear to be involved in synaptogenesis, independent of direct dendritic contact.