Abstract
Purpose: :
A retinal ganglion cell (GC) responds differently each time the same stimulus is presented, a variation that ultimately limits visual detection and discrimination. Here we investigate how the reproducibility of a GC’s response is set by the combination of its inhibitory and excitatory inputs.
Methods: :
We repeatedly presented a dark flash (100 ms, -4% Weber contrast) across the entire receptive field of an OFF alpha GC in an in vitro preparation of the guinea pig retina. We voltage- and current-clamped the GC in the whole cell mode. To control the driving forces for inhibitory and excitatory conductances, we adjusted the membrane potential by adjusting holding potential or holding current. To quantify the reproducibility of responses, we measured the signal-to-noise ratio of currents or voltages as S/N where S is the square of the average amplitude of the response and N is the variance of the response across repeats.
Results: :
The response to a dark flash was an increase in excitation and a decrease in inhibition; both conductance changes produced an inward current that depolarized the cell. As the driving force for inhibition was increased by depolarizing the cell, S/N reached a maximum of 3.5 ± 0.9 when the ratio of driving forces for inhibition and excitation (Vm - Einhibition) / (Vm - Eexcitation) was -0.32 ± 0.03 (7 cells, results from current and voltage clamp combined). Thus the optimal driving force for inhibition is about 1/3 that of excitation. We also derived excitatory and inhibitory conductances from the voltage clamp experiments, combining them to simulate currents. For the simulated currents, the optimal ratio of inhibitory to excitatory driving forces was similar to the experimental results: -0.29 ± 0.05 (2 cells).
Conclusions: :
The normal Vm of an OFF GC (-60mV to -65 mV) is more depolarized than the reversal potential for the inhibitory conductance (ECl ≈ -80 mV) but more hyperpolarized than the reversal potential for the excitatory conductance (Eglu ≈ 0 mV). Thus driving force for inhibition is about 1/4 to 1/3 that of excitation, consistent with the value we found for maximal S/N of responses. This suggests that the normal resting potential of the OFF alpha cell, as set by tonic levels of cross inhibition from the ON pathway and its own voltage gated channels, maximizes the reproducibility of its responses.
Keywords: contrast sensitivity • ganglion cells • retinal connections, networks, circuitry