Passive synaptic normalization and input synchrony-dependent amplification of cortical feedback in thalamocortical neuron dendrites.

William M Connelly, Vincenzo Crunelli & Adam C Errington                                               

(2016) The Journal of Neuroscience, 36(13) 3735-3
754.

The numerically dominant synaptic input to sensory thalamic nuclei including the dorsal lateral geniculate nucleus (dLGN), does not, as might be expected, come from sensory sources (e.g retinal ganglion cells) but, in fact, comes from cortical feedback. Up to 50% of synapses on thalamocortical (TC) neurons are corticothalamic (CT) inputs from layer VI cortical pyramidal neurons. These synapses produce very small EPSPs at the soma of TC neurons and have consequently traditionally been considered as ‘modulators’ of TC neuron activity. One aspect of our understanding of CT feedback to TC neurons has, however, remained unknown. Specifically, how TC neurons integrate these CT numerous inputs and the effects that these inputs produce locally in dendrites. We have published a paper in The Journal of Neuroscience that has addressed some of these questions and provides new insight into the role of CT feedback in TC neurons.

By using paired somatodendritic patch clamp recordings, patterned two photon glutamate uncaging and computational modelling we have demonstrated in vitro that individual quantal sized CT synapses produce large EPSPs in dendrites that are strongly attenuated as they propagate to the soma. However, because of the dendritic properties (see also Connelly et al., 2015) of TC neurons, CT EPSPs are ‘passively normalized’ such that, unlike synaptic potentials in dendrites of other neurons, the amplitude and shape of CT EPSPs is almost entirely unaffected by the location of the synapse within the dendritic tree. Thus, CT feedback seems to be a synaptic democracy with all inputs having an equal effect on TC neuron membrane potential.

Our new data also shows that TC neuron dendrites have mechanisms, involving T-type Ca2+ channels and NMDA receptors, that can amplify CT feedback in a manner dependent upon the temporal pattern of the input. We suggest that these mechanisms might significantly increase the range of responses that CT feedback can produce in TC neurons to exert their modulatory effects. In fact, we find that under certain conditions CT feedback might be sufficient to act as a ‘conditional driver’ of TC neuron firing.

Follow the link above to read the paper in full (freely available).