|Title||Dendritic resonance in a detailed model of pyramidal tract neuron of mouse primary motor cortex|
|Publication Type||Conference Paper|
|Year of Publication||2019|
|Authors||Kelley, C.., Dura-Bernal S.., Neymotin S.., & Lytton W.. W.|
|Conference Name||Society for Neuroscience 2019 (SFN '19)|
|Keywords||2019, sfn 2019, Society for Neuroscience|
Dendritic computation, though poorly understood, presents one of the strongest candidates for distinguishing complex biological neurons from simple integrate-and-fire analogs and still simpler units of artificial neural networks. We simulated frequency responses to subthreshold sinusoidal inputs of 0.5-20 Hz distributed across M1 apical and basilar dendritic arbors of pyramidal tract (PT) type neuron from mouse primary motor cortex (M1). Voltage recordings from the stimulated dendrites yielded three distinct groups of impedance profiles: those with broad peaks at 5-6 Hz in the basal dendrites and the apical trunk (perisomatic), those with narrow resonant peaks at 1.8-2.5 Hz for apical dendrites, and those with no discernible peaks for apical obliques. However, voltage recordings from the soma during stimulation of the saturating apical tufts showed transfer impedance profiles with resonant peaks at 1.8-2.5 Hz. At the soma, stimulation from apical dendrites showed strong positive correlation between peak impedance and frequency, while the perisomatic group had a relatively weak negative correlation between the two. The two groups both showed an inverted sigmoid relationship between peak impedance and distance from the soma, with highest impedance for proximal dendrites and a drop-off between 50-100 microns. Our findings suggest that the different populations of dendritic sections act as selectivity filters with distinct frequency response characteristics. Since synapses from long and short range inputs are spatially organized across the dendritic arbor of PT cells, we hypothesize that these differences in dendritic resonance will play an important role in the behavior of the network and intend to investigate this further in a detailed network model.