| Summary: | <p>The regulation of synaptic strength is thought to underlie the complex emergent dynamics of neural networks. The strength of a synapse is determined by its pre- and postsynaptic properties, both of which are under tight regulatory control orchestrated by ongoing neuronal activity. We now have a good understanding of the plasticity rules underlying the regulation of postsynaptic strength. The same set of rules has been imposed onto the regulation of presynaptic strength. However, the operation of pre- and postsynaptic terminals is fundamentally different, both on a mechanistic and functional level. I have therefore systematically investigated the mechanisms underlying presynaptic plasticity at the Schaffer collateral-CA1 synapse under varying conditions of synaptic activity. I have examined three general modes of regulation: (1) synaptic changes dependent on the concerted activity of pre- and postsynaptic neurone (homosynaptic plasticity); (2) changes dependent only on presynaptic activity; (3) changes dependent on postsynaptic activity alone (heterosynaptic plasticity). Using a combination of electrophysiological and optical techniques, I have monitored and manipulated the strengths of single synapses both pre- and postsynaptically, which allowed me to impose certain activity patterns and investigate resulting synaptic modifications. I found that heterosynaptic plasticity along local segments of dendrites is expressed at both synaptic loci and depends on the spatial arrangement of synapses. Pre- and postsynaptic strength changes were weakly correlated and pharmacologically dissociable. Next, I found that glutamate release suppresses both short- and long-term presynaptic function, which required the activation of presynaptic NMDA receptors. Lastly, I found that presynaptic long-term potentiation (LTP) is spike timing-dependent but does not rely on coincidence detection via postsynaptic NMDA receptors. My findings suggest that the presynaptic terminal is functionally distinct, which is reflected in parallel regulatory pathways. I suggest the synapse to be viewed as a two- compartment model, consisting of a presynaptic non-linear transformation followed by postsynaptic linear weighting.</p>
|
|---|