Search

Models of synaptic plasticity in the nervous system have conventionally assumed a mechanism in which spike activity of a postsynaptic cell enhances the efficacy of recently active presynaptic inputs. Making use of the prompt and dramatic response of the visual cortex to occlusion of vision in one eye during the critical period, we tested the role of postsynaptic activity in ocular dominance plasticity. To do so, we selectively blocked cortical cell discharges with a continuous intracortical infusion of the inhibitory neurotransmitter agonist muscimol during a period of monocular deprivation. This drug inhibits cortical cell discharges with no apparent effect on the activity of their presynaptic geniculocortical inputs. Recording from single cortical cells after they had recovered from the muscimol-induced blockade, we found a consistent shift in the responsiveness of the visual cortex in favor of the less-active, closed eye, while the normal shift in favor of the more-active, open eye was evident in regions not affected by the treatment. Such an inhibition-coupled expression of plasticity in favor of the less-active, closed eye cannot be explained by a nonspecific disruption of cortical function. We interpret these results to indicate (i) that the postsynaptic cell is crucially involved in plasticity of the visual cortex, (ii) that the direction of cortical plasticity depends on postsynaptic membrane conductance or polarization, and (iii) that plasticity can occur in the absence of postsynaptic spike activity.

Amblyopia can be induced by opacity of the ocular media (e.g., cataract), misalignment of the ocular axes (strabismus), or unequal refractive error in the eyes (anisometropia). Experiments in monkeys have shown that early monocular eyelid suture, a model of amblyopia caused by cataract, results in shrinkage of the eye's ocular dominance columns in striate cortex. This reduction of the geniculocortical projection from the deprived eye has been thought to explain in part the mechanism of amblyopia. We labeled the ocular dominance columns in monkeys with amblyopia by using cytochrome oxidase histochemistry. In animals rendered amblyopic by early unilateral eyelid suture, no pattern of cytochrome oxidase activity appeared in layer IVc. Outside layer IVc, alternating rows of light and dark patches were present; the pale patches fit in register with the shrunken ocular dominance columns of the deprived eye, which were labeled by autoradiography. Subsequent removal of one eye caused a striking cytochrome oxidase pattern to emerge in layer IVc that correlated precisely with the shrunken (deprived eye) and expanded (normal eye) ocular dominance columns. This correlation was shown by injecting one eye with [3H]proline. It has remained unsettled whether other forms of amblyopia are accompanied by shrinkage of ocular dominance columns. To address this issue, in an analogous clinical case, we examined the pattern of cytochrome oxidase activity in a human subject with a history of anisometropic amblyopia who suffered a lesion of one optic nerve shortly before death. The ocular dominance columns were normal in width, indicating that some forms of amblyopia occur without shrinkage of ocular dominance columns.