backp97

Abstract

Inhibitory control of somatic and dendritic sodium spikes in neocortical pyramidal neurons in vivo: an intracellular and computational study.
Denis Paré, Erik Lang and Alain Destexhe

Neuroscience 84: 377-402, 1998.

Copy of the full paper (PDF)
Abstract
The effect of synaptic inputs on somatodendritic interactions during action potentials was investigated using in vivo intracellular recording and computational models of neocortical pyramidal cells. An array of 10 microelectrodes, each ending at a different cortical depth, was used to preferentially evoke synaptic inputs to different somatodendritic regions. Relative to action potentials evoked by current injection, spikes elicited by cortical microstimuli were reduced in amplitude and duration, with stimuli delivered at proximal (somatic) and distal (dendritic) levels evoking the largest and smallest decrements, respectively. When the IPSP reversal was shifted to around -50 mV by recording with KCl pipettes, synaptically evoked spikes were significantly less reduced than with potassium acetate or cesium acetate pipettes, suggesting that spike decrements are not only due to a shunt, but also to voltage-dependent effects.

Computational models of neocortical pyramidal cells were built based on available data on the distribution of active currents and synaptic inputs in the soma and dendrites. The distribution of synapses activated by extracellular stimulation was estimated by matching the model to experimental recordings of EPSP/IPSP sequences evoked at different depths. The model successfully reproduced the progressive spike amplitude reduction as a function of stimulation depth, as well as the effects of chloride and cesium. The model revealed that somatic spikes contain an important contribution from proximal dendritic sodium currents up to about 100 um and 300 um from the soma under control and cesium conditions, respectively. Proximal IPSPs can prevent this dendritic participation thus reducing the spike amplitude at the soma. The model suggests that the somatic spike amplitude and shape can be used as a “window” to infer the electrical participation of proximal dendrites. Thus, our results suggest that IPSPs can control the participation of proximal dendrites in somatic sodium spikes.


Movies

Several movie files illustrate the dynamics of membrane potential in soma and dendrites in a simulated neocortical layer V pyramidal neuron. They are an excellent complement to the figures of the paper. The somatodendritic distribution of membrane potential is shown by colors in three cases of action potential generation:

backp_control.mpg Back-propagating action potential following current injection in the soma

backp_stim_dist.mpg Forward-propagating action potential following “distal” stimulation

backp_stim_prox.mpg Action potential following stimulation of “proximal” synapses. In this case, there is no action potential invasion in dendrites and the amplitude of the somatic spike is reduced (see paper)


See also the book chapter Destexhe A, Lang E and Paré D. Somato-dendritic interactions underlying action potential generation in neocortical pyramidal cells in vivo. In: Computational Neuroscience. Trends in Research (edited by J. Bower), Plenum Press, New York, pp. 167-172, 1998.