Retinal horizontal cells express a chemogenetic channel after systemic AAV infection. Cell bodies of horizontal cells are labeled in red; chemogenetic channels are labeled in yellow. Strong expression of the chemogenetic channel in the entire horizontal cell network allowed us to reversibly and efficiently perturb the activity of all horizontal cells throughout the retina, which had not been possible before. The image was obtained using a confocal microscope.
Using machine learning and artificial intelligence, eye diseases could someday be described in terms of the perturbations of computations performed by the retina.
To get to that point, we will first need to have a significant understanding of retinal circuits and how perturbations affect the computations the retina performs.
An international team of scientists has combined experiments on genetics, viral and molecular tools, high-density microelectrode arrays and computer models to get a better idea.
Vision begins in the retina where photoreceptor cells capture the light that falls on the eye and transduce it into neuronal activity. Ganglion cells—the output neurons of the retina—then send the visual signals to the brain.
The retina contains intricate neuronal circuits that are assembled from several different types of neuronal cells and process the incoming signals in a complex way. The circuits also extract important features of the visual scene, where at the output level, the computations of the retinal circuits result in about 30 different neuronal representations of the visual scene. These are then transmitted in parallel to the brain.
To better understand how the retinal output channels represent the visual world and how their different functional properties arise from the architecture of the retinal circuits, the researchers perturbed a specific retinal circuit element while studying how the perturbation changes the functional properties of the different retinal output channels.
The researchers developed a new method to control the activity of horizontal cells—a retinal circuit element that provides feedback inhibition at the first visual synapse between photoreceptors and bipolar cells.
The new approach involves a specific set of viruses, transgenic mice and engineered ligand-gated ion channels, allowing the scientists to switch the feedback at the first visual synapse on and off. The researchers used high-density microelectrode arrays to measure the effects of the perturbation in the retinal output and recorded the electrical signals of hundreds of ganglion cells simultaneously.
“We were astonished by the variety of effects that we observed due to the perturbation of a single, well-defined circuit element,” Antonia Drinnenberg, a former graduate student from Botond Roska’s group at the Friedrich Miescher Institute, and lead author of the paper, said in a statement. “At first, we suspected that technical issues might underlie this variety.”
However, after measuring the signals in thousands of ganglion cells and in defined retinal output channels, the researchers saw that the variety in the horizontal cell contributions measured must arise from the specific architecture of the retinal circuitry.
The team then built a computer model of the retinal that simulated the different pathways that the signal can take through the retina. This enabled them to investigate if the current understanding of the retinal circuitry could account for the effects the researchers saw during the experiments.
The model was able to reproduce the entire set of changes that were measured experimentally and make five further predictions about the role of horizontal cells.
“We were surprised to see that the model went further than what we had in mind at the time we built it,” Felix Franke, co-first author of the paper, said in a statement. “All additional predictions turned out to be correct when we conducted additional experiments to test them.”
If researchers gain a better understanding as to how the retina transforms images into signals to the brain, it could lead to better insights into brain computations and ultimately lead to be better classes of medicine.
The study was published in Neuron.
