CABLE THEORY MODELING OF THE EFFECTIVENESS OF SYNAPTIC INPUTS IN CORTICAL PYRAMIDAL CELLS.

摘要

The effectiveness of synaptic inputs in cortical pyramidal neurons is explored in a variety of circumstances with a passive cable model. This model gives the transient voltage response in the dendritic tree due to current, voltage, or synaptic input, or distributions of ionic conductances. It has many advantages over compartmental models, other continuous cable models, and numerical integration approaches including SPICE.;The activation-distribution (distribution of activated synapses) can affect the effectiveness of individual inputs by changing the local resting potential and the electrotonic distance from the soma to the input location. Fourfold variations in the peak transient soma potential may occur with certain activation-distributions. Removal of afferents such as may occur in in vitro preparations can affect the activation-distribution and thus change the effectiveness of individual inputs and the electrophysiological properties of the cell.;Inputs on dendritic spines are less effective than inputs on dendritic shafts, especially for large synaptic conductance changes. An expression is derived relating changes in spine head and stem length and diameter to the potential generated at the dendritic shaft. Using this expression it was found that changes in spine stem dimensions, but not in spine head dimensions, could change the effectiveness of an input. However the changes were small. It appears likely that voltage-dependent conductances in spines could be very important for synaptic plasticity and effectiveness.;The model was used to explore morphological determinants of the effectiveness of synaptic input. Synaptic inputs at the same physical or electrotonic distance from the soma may produce quite different changes in soma potential because of end effects or load effects. For a single input there is a set of process diameters which will maximize the effectiveness of that input. R(,m) or R(,i) changes will affect the location of synapses operating at optimal effectiveness. Anatomical data limitations such as shrinkage or incomplete staining will affect the modeled effectiveness of different inputs in different ways. Excluding spines in a model will make synapses appear to be much more effective than they are. Tapering processes can be modeled by two or more segments of different diameters.