Isolating the circuit components behind contrast adaptation within the Rod-Bipolar pathway

Abstract

Purpose : Visual stimuli with significant temporal variability are encoded by a relatively small range of ganglion cell (GC) firing rates. Here, we assess stimulus coding by identified inner retinal circuits presynaptic to GCs by optogenetic and computational methods. We focus on a single computation, the gain control responsible for contrast adaptation.

Methods : Retinas were extracted from two strains of mice with Cre-mediated ChR2 expression in either rod bipolar (RB) cells or CRH-1 amacrine cells (ACs). Voltage-clamp recordings were made from ON alpha GCs, which receive input from RBs and CRH-1 ACs. RBs communicate with ON-alpha GCs as follows: RB->ON cone bipolar (CB)->GC (excitation) and RB->AII->ON cone bipolar (CB)->AC->GC (inhibition); CRH-1 ACs make direct inhibitory synapses onto ON-alpha GCs. With photoreceptor-mediated signaling blocked, temporally modulated white noise stimuli evoked ChR2-driven responses. Linear-Nonlinear (LN) and more complex nonlinear cascade models were used to analyze signal transfer and contrast adaptation.

Results : Excitatory synaptic inputs to ON-alpha GCs show contrast adaptation. This mechanism was sensitive to TPMPA (a GABA_CR antagonist), suggesting a role for presynaptic inhibition of transmission from bipolar cells. The transformation from stimulus to synaptic current was captured by conventional LN models, but adapted forms of Nonlinear Input Models (NIMs) performed better (LN r² = 0.61 [+/- 0.06], NIM r² = 0.80 [+/- 0.05], with a gain change of 30% [+/-9%]). In the case of the CRH-1-GC circuit, no clear contrast adaptation was apparent, and LN models appeared to perform as well or better than NIM models.

Conclusions : These results suggest that the mechanism for gain control occurs primarily along the excitatory pathway onto ON-alpha RGCs, as contrast adaptation is not seen in the inhibitory input onto ON-alpha GCs from CRH-1 synapses. Finally, the success of NIMs relative to other models demonstrates that the mechanisms behind contrast adaptation are likely divisive (such as suppression), rather than subtractive.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.