Abstract
Purpose: :
Starburst amacrine cells (SACs) exhibit directionally–selective (DS) light responses (Gavrikov et al., 2003, PNAS) that depend on different types of cation–Cl cotransporter in different dendritic compartments so that GABA depolarizes one compartment and hyperpolarizes the other (Mangel et al., 2005, ARVO). We therefore modeled interactions of PSPs within the SAC dendritic tree to examine whether a combination of depolarizing glutamate–mediated and both hyper– and depolarizing (depending on the Cl– electrochemical gradient) GABA–mediated synaptic inputs could produce the DS responses that we have observed in SACs.
Methods: :
Voltage changes at the starburst cell body and at two dendritic tips on opposite sides of the body, which were produced by a moving light stimulus, were modeled with a custom–made computer program. The receptive field of SACs measured in our experiments was approximately 5 times larger than the actual SAC dendritic tree. Thus, glutamate from vertically oriented bipolar cells provided input only from the central 20% of the SAC receptive field, although these synapses were distributed throughout the entire SAC dendritic tree. GABA synapses were also distributed throughout the entire SAC dendritic tree, but they provided input from the entire receptive field because GABA synapses derived from laterally oriented amacrine cells. In addition, GABA–mediated inputs were either excitatory or inhibitory depending on the local Cl– electrochemical gradient.
Results: :
Glutamate–mediated input is probably responsible for the large and relatively fast depolarizing response of SACs to moving light bars when they cross the receptive field center. The Cl– equilibrium potential in the distal portion of SAC dendrites has to be more negative than the local membrane potential to generate the hyperpolarizing responses to stationary and moving stimuli in the receptive field periphery. Also, GABA–induced PSPs have to last longer than glutamate–gated PSPs to produce the DS light responses that we have observed. Moreover, the DS responses of SACs are enhanced if the Cl– equilibrium potential in the proximal portion of SAC dendrites is more positive than the local membrane potential.
Conclusions: :
Our modeling results indicate that SAC DS responses similar to those we have observed can be generated if there is a chloride gradient along SAC dendrites and if the GABA–evoked increase in the chloride conductance is relatively long–lasting. Our findings therefore support the idea that the differential distribution of the Na–K–Cl and K–Cl cotransporters on SAC dendrites underlies their DS responses.
Keywords: amacrine cells • computational modeling • ion transporters