Evidence that changes in spine neck resistance are not responsible for expression of LTP

From modeling studies it is known that changes in spine neck resistance can influence the shape of the non-linear curve relating synaptic current to synaptic conductance if the resistance of the neck approaches the synaptic input resistance. Such work also indicates that the effects of resistance will be much more pronounced for fast rather than slow synaptic currents. Accordingly, a reduction in neck resistance could produce an increase in the rapid responses generated by the quisqualate/AMPA class of glutamate receptors while only minimally affecting the slower NMDA receptor-mediated responses and thus account for the pattern of changes known to be associated with long-term potentiation (LTP). This hypothesis predicts that large reductions in synaptic conductance should have disproportionate effects on potentiated versus control responses. This was tested by using field potential recordings of synaptic currents in CA1 pyramidal cells in hippocampal slices in response to stimulation of Schaffer/commissural inputs that either received LTP-inducing stimulation or did not. Two manipulations were used to systematically reduce synaptic conductances: reductions of extracellular Ca++ and partial blockade of postsynaptic receptors. Reductions of synaptic field potentials by 40-75% by either method at control synapses were accompanied by equivalent reductions at previously potentiated synapses. These results suggest that LTP expression is not due to a change in the curves relating synaptic current to synaptic conductance as would be predicted by the spine resistance hypothesis.