Development of corticogeniculate synapses in the cat.

The development of corticogeniculate synapses was studied in 16 cats ranging in age from newborn to adult. Tritiated proline was injected into areas 17 and 18 of the visual cortex in order to label corticogeniculate terminals in lamina A of the dorsal lateral geniculate nucleus. The labeled terminals were then characterized ultrastructurally using electron microscopic autoradiography. Labeled synaptic profiles were found in newborn kittens, indicating that corticogeniculate connections are present in the cat at birth. Morphologically, however, many corticogeniculate endings in newborn and 1-week-old kittens are different from those in older animals in that they do not form well-defined terminal boutons, and their synaptic vesicles are often loosely packed. In kittens 2 weeks of age and older, corticogeniculate axons end as RSD terminals exclusively; i.e., they are relatively small in size and contain round, densely packed synaptic vesicles, and occasionally an electron-dense mitochondrion (Guillery: Z. Zellforsch. 99: 1-38, '69). However, not all RSD terminals in the LGN represent input from visual cortex. Injections of 3H-proline into the mesencephalic reticular formation also label RSD terminals selectively in the lateral geniculate nucleus. At all ages corticogeniculate axons make synaptic contacts with dendrites exclusively, and they are always presynaptic. This suggests that the essential pattern of corticogeniculate synapses is formed early and is not altered during subsequent development. Quantitatively, there is no significant change in the size of corticogeniculate terminals or their synaptic vesicles in kittens 2 weeks of age (the youngest measured) and older. In contrast, the synaptic contact lengths of these terminals decreases about 28% between 2 and 12 weeks. During this same period there is approximately a twofold increase in the density of corticogeniculate terminals in the neuropil of lamina A. Since the volume of neuropil in lamina A increases almost fourfold between 2 and 12 weeks, the doubling of corticogeniculate terminal density represents about an eightfold increase in terminal number. After 12 weeks there is little change in the length, density, or number of corticogeniculate synaptic contacts, which suggests that the morphological development of the corticogeniculate pathway is essentially complete by this age.