Purchase this article with an account.
J. S. Diamond, W. N. Grimes, J. Zhang, C. Graydon, B. Kachar; Parallel Processing Within Single A17 Amacrine Cells. Invest. Ophthalmol. Vis. Sci. 2010;51(13):1205.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Most neurons possess branched dendrites that receive and integrate synaptic inputs and extensive axons that distribute action potential output. By contrast, amacrine cells collect input and distribute output within the same neurite arbor. We sought to determine whether input-output microcircuits operate independently within A17 amacrine cell processes, enabling the cell to function as a parallel processor within the circuitry of the inner retina.
The ultrastructure of A17 processes and synaptic feedback varicosities in rat retina was reconstructed and studied using 3-D serial electron microscopy. Feedback inhibition was recorded with whole-cell recordings from presynaptic rod bipolar cells in acute retina slices. Voltage-gated currents and excitatory feedforward synaptic responses were recorded in A17s using similar techniques. Morphological and physiological data were combined to construct an electrotonic model of A17 amacrine cells. Synaptic calcium signals were imaged in A17 synaptic processes using calcium indicator dyes and two-photon laser-scanning microscopy.
EM reconstructions showed that synaptic varicosities are separated by thin (130 nm) dendrites and that most varicosities make two distinct feedback synapses onto each presynaptic rod bipolar cell. Whole-cell recordings indicated that rat A17s do not fire action potentials and that voltage-gated sodium channels do not boost synaptic signals in A17 processes. Electrotonic modeling suggested that A17 processes prevent long-range propagation of synaptic depolarizations. Synaptic stimulation of single varicosities elicited no calcium signal in neighboring varicosities that were located more than 20 microns away, which it the average distance between neighboring varicosities.
Our results indicate that a single A17 amacrine cell contains several hundred independent input/output microcircuits operating in parallel. The A17, therefore, constitutes a novel example of how morphological and biophysical properties can combine to maximize a cell’s capacity for independent processing. We suggest that the A17 has evolved to maximize circuit computational capacity while minimizing the associated tissue volume and metabolic cost.
This PDF is available to Subscribers Only