One key step in visual processing is the transmission of signals
from photoreceptors to retinal ganglion cells by bipolar cells.
There are about 10 types of bipolar cells in a vertebrate retina,
and they form parallel channels where each bipolar cell type
carries a distinct type of visual information from the outer to
inner retina.
The signals from these bipolar channels are then integrated by
ganglion cells through intricate interactions with about 30
types of amacrine cells. Specific combinations of bipolar and
amacrine inputs generate about 10 functional types of ganglion
cells, and thus such interactions in the inner plexiform layer
are the most interesting---but least understood---processings
in the retinal circuitry.
By simultaneously recording from multiple ganglion cells while
manipulating the bipolar cell activity intracellularly, here we
explored how the signal from an individual bipolar cell is
distributed to the various types of ganglion cells.
We found that injecting current into an individual bipolar cell
elicited significant effects on the visual responses of many
ganglion cells.
(1) The contribution of a single bipolar cell was generally
excitatory at short distances (<0.3 mm) and inhibitory at longer
distances (~0.5 mm). This is consistent with the presumed
role of amacrine cells as inhibitory interneurons.
(2) Within the excitatory region, different ganglion cells showed
distinct response patterns. A sustained depolarization of the
bipolar cell produced a transient burst of spikes in some ganglion
cells, but a sustained firing in others.
Furthermore, some ganglion cells responded to the bipolar cell
stimulation in a linear fashion, whereas others showed a highly
rectifying nonlinearity.
These results emphasize the diversity of neural circuits that
distribute signals from the same bipolar cell to various ganglion
cells. Specifically, the distinction between transient and
sustained response dynamics in ganglion cells is not simply
determined by what bipolar channels they receive, but in large
part by differential circuitry in the inner plexiform layer.
Poster: Computational and Systems Neuroscience (COSYNE) 2009, Salt Lake City, Utah.
Poster (343KB, PDF):
Sloan-Swartz Annual Meeting on Computational Neuroscience,
2009, Cambridge, Massachusetts.
Preprint (in preparation).