Dendritic location of synapses and possible mechanisms for the monosynaptic EPSP in motoneurons.

THE PURPOSE OF THIS PAPER is to bring together recent experimental and theoretical results which, when considered together, add to our understanding of the monosynaptic excitatory postsynaptic potential (EPSP) in moto-neurons. A general mathematical neuron model has been proposed which provides a means of computing EPSP properties that result when synapses are distributed over the soma and dendrites of a nerve cell (29). This general model contains a set of theoretical parameters corresponding to such neu-ronal properties as the membrane time constan .t, the dendritic electro tonic length, the surface area ratio and the condu .ctance ra tio of dendrites to soma, the spatial distribution of synaptic input over the soma .-dendritic surface, and the time course of synaptic current or synaptic conductance. Every specific application of this general model depends upon selecting a set of values for these parameters. We wish to explore the application of this general model to the particular case of the cat spinal motoneuron. Intracellular recordings from motoneurons have provided a description of many of the properties of the monosynaptic EPSP's of these cells (4, 7, 8, 24, 32). In particular, the shapes of the unitary, miniature EPSP's resulting from activation of single Ia afferent fibers have been described. The relationship of these unitary EPSP's to the composite EPSP's elicited by electrical stimulation of many Ia fibers has been discussed (4). We wish to determine whether the experimental data will allow us to put some restrictions on the model parameters which determine EPSP shapes and to provide a more sharply defined picture of the organizati on of synapses on the motoneuronal soma and dendritic tree. In PART I of this paper we will show that a number of restrictions do emerge from\ the analysis. For example, we have found that simple variations in the time course of synaptic currents cannot reproduce the observed EPSP shapes if the synaptic input is distributed uniformly over the motoneuron surface. The most restrictive model which successfully simulates the observed data, and the model which we will discuss in some detail, involves a single time course for the synaptic current, but with the synaptic input occurring in different combinations of somatic and dendritic locations. Thus, a