Abstract
Purpose :
Amacrine cells (ACs) are the most diverse class of neurons in the mammalian retina and the least well-characterized in terms of typology. A substantial fraction of ACs have their somas displaced in the ganglion cell layer, where they are easily accessible for electrophysiology. We aim to establish a multi-modal classification of displaced amacrine cells in the mouse.
Methods :
We used a triple-transgenic mouse line (ChAT-Cre x Vglut2-Cre x Ai14) to target non-starburst displaced ACs (dACs) by their position in the ganglion cell layer and their lack of fluorescence under two-photon illumination. We recorded dACs in whole-cell current clamp. We measured the light responses of each cell using a battery of light stimuli, including spots from 30 µm – 1200 µm and gratings drifting at 12 different orientations. To measure the intrinsic electrical properties of dACs, we injected a series of hyperpolarizing and depolarizing current steps. Finally, to measure the morphology of each cell, we filled them with AlexaFluor488 and imaged their neurites along with those of starburst amacrine cells (in red) as a fiducial marker. Using all three modalities – light responses, intrinsic electrophysiological properties, and morphology – we established a unified typology of dACs.
Results :
We characterized our transgenic line and found that consistent with previous reports, ~50% of cells in the ganglion cell layer are dACs. Our current dataset includes over 100 dACs with all three modalities measured, separated into 21 types. 11 types are wide-field ACs (neuritic arbor >500 µm), and 10 types are medium-field ACs (< 500 µm).
Conclusions :
Multi-modal measurements are important for neuronal classification. Our results suggest that there are more dAC types than previously appreciated because we were able to separate types that were similar in one modality because of distinct differences in another modality. We are working toward an online resource for dAC typology similar to our previous work on ganglion cells at rgctypes.org.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.