The Role of Activity in the Development of Long-Range Horizontal Connections in Area 17 of the Ferret

Abstract

Horizontal connections in area 17 of adult cats and ferrets link cells with similar preferred orientations by a patchy network of projections extending several millimeters across the cortex. The maturation of orientation selectivity in ferret area 17 has been demonstrated previously by quantitative single-unit recording and optical imaging to begin at approximately postnatal days (P) 32–P36. We therefore made restricted injections of cholera toxin B-subunit (CTB) or CTB-gold into ferret area 17 at a series of developmental ages and statistically quantified the degree of clustering in plots of retrogradely labeled cells in tangential sections through layer III for comparison to the published values for orientation tuning at each age. At P21, horizontal connections within area 17 lacked patchiness entirely, although clear patches of labeled cells were present in extrastriate areas. By P27, significant clustering of horizontal connections within area 17 was present. A second phase of cluster refinement was observed to occur at approximately P34–P36, coinciding with the emergence of mature orientation tuning and maps. Continuous silencing of cortical action potentials by chronic tetrodotoxin infusion from P21 resulted in a spatially random distribution of retrogradely labeled cells at P34. In contrast, bilateral enucleation from P21 did not prevent the initial development of clustered horizontal connections. We conclude, based on our findings and those of others, that the anatomical specificity of long-range horizontal connections results from an activity-dependent process that initially can use spontaneous activity in the cortical and thalamic networks to establish crude periodic connections and later uses visual cues to refine these connections.