Awake responses suggest inefficient dense coding in the mouse retina
3 November 2023
10.7554/eLife.78005
Read the latest Issue
New evidence on retinal function
A new study from the Asari group at EMBL Rome shows how retinal function in awake mice differs from isolated retinal samples – insights that could help to develop prosthetic devices that can act as a retina in the future
When light enters the eyes, it is focused onto the retina, a thin layer of brain tissue at the back of the eye. In a new study, scientists from the Asari Group at EMBL Rome successfully studied the visual responses of the retina in awake mice for the first time, providing novel insights and overcoming technical limitations that had previously held the field back.
The vertebrate retina consists of ∼100 cell types in total, including ∼30 different types of retinal ganglion cells (RGCs) that send the retinal output signals to the brain via the optic nerve. Each of these RGC types forms distinct neural circuits within the retina to extract specific features of the visual image coming into the eye, such as colour, contrast, and motion.
One of the best-known brain tissues
Unlike the rest of the brain, the retina can continue working even when removed from an animal, making it easier for scientists to study how this light-processing tissue works. This has helped it become one of the most accessible and hence best-understood parts of the brain. Most knowledge of retinal signal processing comes from studies of isolated retinas.
However, such ex vivo physiological approaches have certain limitations. First, one cannot perform long-term recordings of neuronal signals. Second, one cannot avoid artefacts due to retinal dissection. For example, sometimes the retinal tissue detaches from surrounding epithelial cells or the retinal ganglion cells react differently due to their axons being cut during dissection. Most importantly, despite a long history of research on the retina, it still remains unclear whether isolated retina samples behave in the same way as retinas in live animals, and whether findings in isolated retinas apply to natural visual processing in an awake state.
In vivo studies are thus indispensable for clarifying retinal function, though measurements in awake mice have never been done due to technical difficulties.
A novel approach to study retinal function in vivo
In the new study, scientists from the Asari group overcame the technical barriers and successfully compared the visual responses of the retina in awake mice, anaesthetised mice and isolated retinas. By measuring retinal electrical signals, they showed that awake mice responded to light substantially more quickly and strongly than anaesthetised mice or isolated retinas. This change in the kinetics is likely a direct consequence of the high baseline activity of retinal neurons in awake animals. The high activity results in the nerve cells reaching a state where they can be stimulated faster, allowing them to respond more quickly to a light signal.
Towards the development of visual prostheses
The researchers also performed computational analyses that suggested that the amount of information carried to the brain was largely comparable across the different conditions, but the retina in awake mice used more energy. “These findings indicate that we can no longer rely on the assumption that observations made on isolated retinas can be used to explain how the retina operates in vivo,” said Hiroki Asari, Group Leader at EMBL Rome. “Further studies are needed to better understand how the retina processes visual information in awake animals. In particular, it is critical to characterise how the retinal processing depends on an animal’s internal states, such as circadian rhythm, thirstiness, or alertness levels. Progress in this understanding could ultimately help to develop prosthetic devices that can act as a retina in the future.”
