Heterogeneous firing responses predict diverse couplings to presynaptic activity in mice Layer V pyramidal neurons.

Yann Zerlaut and Alain Destexhe

BioRxiv preprint: http://biorxiv.org/content/early/2016/12/04/091587 (2016)

Copy of the full paper (PDF)


In the present study, we present a theoretical framework combining experimental characterizations and analytical calculus to capture the firing rate input-output properties of single neurons in the fluctuation-driven regime. We use this framework to investigate the functional impact of the heterogeneity in firing responses found experimentally in young mice layer V pyramidal cells. We first design and calibrate in vitro a simplified morphological model of layer V pyramidal neurons with a dendritic tree following Rall's branching rule. Then, we propose an analytical derivation for the membrane potential fluctuations at the soma as a function of the properties of the synaptic bombardment in dendrites. This mathematical description allows us to easily emulate various forms of presynaptic activities: either balanced, unbalanced, synchronized, purely proximal or purely distal synaptic activity. We found that those different forms of activity led to various comodulations of the membrane potential fluctuation properties, thus raising the question whether individual neurons might differentially couple to specific forms of activity because of their different firing responses. We indeed found such a heterogeneous response for all types of presynaptic activity. This heterogeneity was explained by different levels of cellular excitability in the case of the balanced, unbalanced, synchronized and purely distal activity. A notable exception appeared for proximal activity: increasing activity could either promote firing response in some cells or suppress it in some other cells whatever their individual excitability. This behavior could only be explained by various sensitivities to the speed of the fluctuations, which was previously associated to heterogeneous levels of sodium channel inactivation and density. Because local network connectivity targets rather proximal region, our results suggest that this biophysical heterogeneity might be relevant to neocortical processing by controlling how individual neurons couple to local network activity.

return to publication list
return to main page