Abstract
Abstract: :
Purpose: It has been shown that starburst amacrine cell dendrites respond more strongly to centrifugal movement than to centripetal movement (Euler et al 2002). We tested if network interactions can account for this phenomenon, or if intrinsic properties of starburst dendrites are necessary. Methods: We made a computer model of interacting starburst amacrine cells. Cell bodies were arranged in a 30 µm square lattice, each cell had four 150 µm long dendrites pointing in the cardinal directions. Each dendrite was modeled as a single compartment so that no intrinsic dendritic properties could contribute to any observed asymmetries. Dendrites received inputs over their distal 110 µm, and made output synapses at their distal third. By setting the model parameters for "synaptic connection strength" to positive or negative values, overlapping starburst dendrites could either excite or inhibit each other. The model simulated the response of individual dendrites to stationary and moving light stimuli. Results: We observed a very robust inward/outward asymmetry of the starburst dendrites consistent with the reported physiological findings, i.e. the dendrites preferred outward movement. This result was obtained under three conditions: (1) Starburst dendrites pointing in the same direction excite each other (corresponding to mutual cholinergic excitation of starburst dendrites in the retina) (2) Starburst dendrites pointing in opposing directions inhibit each other (corresponding to mutual GABAergic inhibition of starburst dendrites in the retina) (3) Both of the above conditions are true at the same time. If any of these conditions is met, the observed asymmetry is robust in the sense that the asymmetry does not depend on the magnitude of the parameters representing synaptic connection strength. On the other hand, if the conditions are not met (for example if dendrites pointing in the same direction inhibit each other), the result depends on the relative strength of the synaptic parameters: individual dendrites can either respond best to inward or outward movement, or respond symmetrically. Conclusions: Our results suggest that a properly wired network of starburst cells can account for the experimentally observed asymmetry of their response to movement, and that internal biophysical or biochemical properties of starburst cell dendrites are not strictly necessary. Even if starburst dendrites behave asymmetrically because of internal properties, a properly wired network could complement this behavior, making it more robust, while an "improper" network could adversely influence the asymmetry.
Keywords: amacrine cells • computational modeling • retina: proximal (bipolar, amacrine, and ganglion cells)