function (Figure 1). The greater degree of redundancy Washington University School of Medicine is perhaps the most salient difference between the large St. Louis, Missouri 63110 nervous systems of typical vertebrates and the much smaller nervous systems of invertebrates. Bailey, C.H., Montarolo, P., Chen, M., Kandel, E.R., and Schacher, 1980), in humans monocular deprivation can affect vi-S. (1992). Inhibitors of protein and RNA synthesis block structural sual acuity even in children 6-7 years old (Daw, 1998). changes that accompany long-term heterosynaptic plasticity in A second argument against synapse elimination being Aplysia. Neuron 9, 749-758. a mechanism for memory is that the total number of Balice-Gordon, R.J., and Lichtman, J.W. (1990). In vivo visualization synapses in the brain is increasing during early postnatal of the growth of pre-and postsynaptic elements of neuromuscular life when learning is certainly taking place (Huttenlocher junctions in the mouse. J. Neurosci. 10, 894-908. and Dabholkar, 1997). Moreover, animals raised in inter-Balice-Gordon, R.J., and Lichtman, J.W. (1993). In vivo observations esting environments or in situations where learning is of pre-and postsynaptic changes during the transition from multiple to single innervation at developing neuromuscular junctions. J. Neuoccurring show an increase in synapse number (Klintrosci. 13, 834-855. sova and Greenough, 1999). Recent demonstrations of Balice-Gordon, R.J., and Lichtman, J.W. (1994). Long-term synapse dendritic extensions associated with LTP also make the loss induced by focal blockade of postsynaptic receptors. Nature case that synaptic additions may be associated with 372, 519-524.

Permanent removal of axonal input to postsynaptic cells helps shape the pattern of neuronal connections in response to experience, but the process is poorly understood. Intracellular recording from newborn and adult mouse muscle fibers temporarily innervated by two axons showed an increasing disparity in the synaptic strengths of the two inputs before one was eliminated. The connection that survived gained strength by increasing the amount of neurotransmitter released (quantal content), whereas the input that was subsequently removed became progressively weaker, because of a reduction in quantal content and a reduction in quantal efficacy associated with reduced postsynaptic receptor density. Once the synaptic strengths of two inputs began to diverge, complete axonal withdrawal of the weaker input occurred within 1 to 2 days. These experiments provide a link between experience-driven changes in synaptic strength and long-term changes in connectivity in the mammalian nervous system.

No abstract

The lateral geniculate nucleus relays visual information from the retina to cortex. One well-known anatomical consequence of monocular deprivation during early postnatal development is a shrinkage of neurons in the lamina of the lateral geniculate nucleus that receive input from the deprived eye. This is thought to reflect the competition of afferents subserving the two eyes, possibly at the level of the visual cortex. We frnd that blockade of N-methyl-D-aspartate receptors in kitten visual cortex disrupts this process of binocular competition. These data provide direct evidence that postsynaptic activation of cortical neurons is required for competitive changes in lateral geniculate cell size and suggest a role forN-methyl-D-aspartate receptors in anatomical as well as physiological plasticity in the mammalian visual system. Monocular deprivation (MD) In the present paper we address these important unanswered questions. We find that the local infusion of the NMDA receptor antagonist 2-amino-5-phosphonovaleric acid (AP5) into the striate cortex of monocularly deprived kittens creates a critical segment of LGN that is resistant to the effects of MD. Our results provide direct evidence that the shrinkage of LGN neurons after MD depends upon postsynaptic cortical activity and strengthen the hypothesis that the site of competition is the visual cortex. Further, this work provides additional support for the hypothesis that visual cortical NMDA-receptor activation plays a central role in the mechanisms, both anatomical and physiological, that subserve binocular competition. MATERIALS AND METHODSKittens were reared normally until =5 weeks of age (Table 1) at which time they were fitted bilaterally with osmotic minipumps (Alzet 2001) that were attached via polyethylene tubing to 30-gauge needles implanted directly into the visual cortex. These minipumps deliver their contents at a rate of 1 Al per hr for up to 10 days. In the experimental group the pumps contained 50 mM D,L-AP5; in an age-matched control group the pumps contained vehicle solution only (sterile Ringer's solution). At the same time as the pump implant, the animals were monocularly deprived by lid suture.The extracellular concentration of AP5 declines exponentially with increasing distance from the infusion cannula; at 6 mm the steady-state concentration is estimated to be -'150 ,uM and is sufficient to block the ocular-dominance shift measured electrophysiologically (7). To identify the geniculate segment that projects to this region of striate cortex, 0.2-0.3 ul of a 20-30% horseradish peroxidase (HRP) solution was injected into the striate cortex 6 mm anterior to the infusion site on the eighth day of the infusion (Fig. la). After allowing 2 days for retrograde transport of the HRP, the animal was deeply anesthetized with sodium pentobarbital and perfused through the ascending aorta with saline followed by 1.25% glutaraldehyde/1% paraformaldehyde/0.1 M sodium phosphate buffer, pH 7.4. The brains were removed and cryoprotected for several days ...