Modeling the integrative properties of dendrites: application to the striatal spiny neuron.

The role of the striatum in the control of movements and in the processing of cortical information has received much attention in the recent years. We set out a simple biophysical model for the medium-spiny neuron (msn), the most abundant cell in striatum. This neuron receives two main kinds of inputs, namely, cortical excitatory inputs and dopaminergic inputs coming from the substantia nigra pars compacta. The msn axon impinges directly onto the globus pallidus and onto the substantia nigra pars reticulata neurons and onto striatal neurons through recurrent branches of the axon. The msn is characterized by spiny dendritic trees with a high density of spines (1 to 4 spines/microns) and the probable existence of dendritic spikes. The model predicts that the neuron can integrate excitable inputs in a linear or a nonlinear mode. In the nonlinear mode, the neuron allows the detection of simultaneous (or almost simultaneous) synaptic inputs; it facilitates either a slowing down or a speeding up of the information transfer between the synaptic input location and the soma and is sensitive to inhibiton-excitation pairing. Conversely, in the linear integrative mode, the somatic voltage is determined by a weighted summation of the synaptic inputs. Several geometrical, electrical, or temporal factors can control the switch between these behaviors: the density of excitable dendritic elements, the dendritic radius, the resistance of the spine stem, the membrane resistance, the time between excitations, and the distance between synaptic sites. Finally, the signification of this behavior is discussed in connection with the putative role of dopamine and with the striatal net organization.