Effects of NMDA antagonists on developmental plasticity in kitten visual cortex.

The existence of Hebb synapses in the visual cortex of young kittens has long been postulated. A mechanism for the correlation of activity in simultaneously active pre- and postsynaptic neurons could be provided by the properties of the N-methyl-D-aspartate (NMDA) receptor and its associated Ca2+ channel, which opens in a transmitter- and voltage-dependent manner. We have studied the effects on cortical plasticity of blocking NMDA receptors in different ways with competitive and non-competitive NMDA antagonists. In our first approach, the non-competitive NMDA antagonist ketamine, a short-acting dissociative anaesthetic, was injected systemically after each of a series of brief monocular exposures. This procedure prevented the development of an ocular dominance shift towards the experienced eye in the visual cortex. Other short-acting anaesthetics, such as xylazine or methohexital, while providing the same depth of anaesthesia, did not have the same effect on ocular dominance plasticity. We conclude, therefore, that ketamine quite specifically interferes with synaptic consolidation in the visual cortex. In order to establish a role of NMDA receptors for cortical plasticity directly in the visual cortex, we performed another series of experiments: 2-amino-5-phosphono-valerate (APV), a competitive NMDA antagonist, was infused intracortically by means of implanted osmotic minipumps in kittens, which were monocularly deprived for 1-2 weeks. Within a radius of 4-5 mm, the expected ocular dominance shift was prevented or reduced. In addition, however, physiologically determined cell density and responsiveness to visual stimuli were grossly abnormal around the infusion site, and histological cell density was also reduced. Similar effects were found when MK801 (a non-competitive NMDA antagonist) was used in the same type of experiment. The outcome of both experimental approaches makes it very likely that NMDA antagonists somehow interfere with cortical plasticity. Their mode of action, however, remains ambiguous. Although it is quite possible that blocking of the NMDA channel prevents the Hebbian correlative process necessary for synaptic consolidation, more complex effects, such as an interference with a neurotrophic action normally exerted via the NMDA receptor, may have to be taken into account as well.