Abstract
The development of sensory maps is thought to require an activity- dependent structural rearrangement of afferent terminal arbors within the CNS which recreates the topographic relations of sensory somata present in the periphery. In the frog retinotectal projection, activation of the NMDA receptor plays a role in this structural plasticity. Exposure of the optic tectum of tadpoles to NMDA receptor antagonists results in a rearrangement of retinal ganglion cell arbors so that their organization into a topographic projection and eye- specific stripes is disrupted (Cline et al., 1987; Cline and Constantine-Paton, 1989). Exposure of the optic tectum to the receptor agonist, NMDA, increases the eye-specific segregation of these arbors (Cline et al., 1987). We examined the projection of the supernumerary retina and the morphology of individual retinal afferent arbors of untreated, NMDA-treated, APV-treated, MK801-treated, and MK801/NMDA- treated 3-eyed tadpoles and young postmetamorphic frogs in an effort to understand how NMDA receptor activation is involved in the growth and ordering of retinal arbors. Treatments with MK801 in combination with NMDA resulted in a desegregation of eye-specific stripes, whereas treatments with MK801 or NMDA alone did not. As reported previously, APV treatment resulted in stripe desegregation without increasing the tangential area (measured from 2-dimensional drawings) of the terminal arbors. However, a detailed analysis revealed that the APV-treated tadpole arbors have 35% reduction in branch density (branch tips/area) compared to untreated 3-eyed tadpole arbors. We treated the optic tectum with a range of concentrations of NMDA prepared in the slow- release plastic Elvax. NMDA at 10(-4) M in Elvax was the optimal concentration to produce the sharpening of stripe borders. Exposure of the tectum to NMDA at 10(-6) M in Elvax produced no change in the stripe pattern, while 10(-2) M NMDA in Elvax resulted in beading of the arbors. At the optimal concentration NMDA treatment results in a 75% reduction in the number of axons crossing from a stripe to an interstripe zone. Drawings of individual HRP-labeled, NMDA-treated arbors demonstrate that they have fewer branch points and fewer branch tips. NMDA treatment reduced arbor density by approximately 50%. Arbors drawn from untreated postmetamorphic frogs have twice the branch density of arbors from untreated tadpoles. NMDA treatment in these animals reduced the branch density by 55%, comparable to the reduction seen in tadpole branch density. Our data support a specific hypothesis for NMDA receptor involvement in the activity-dependent structural refinement process within the developing retinotectal projection.(ABSTRACT TRUNCATED AT 400 WORDS)
