The Johns Hopkins Gazette: February 25, 2002
February 25, 2002
VOL. 31, NO. 23


Single Cell Type May Control Internal Clock and Pupil of Eye

By Joanna Downer
Johns Hopkins Medicine
Johns Hopkins Gazette Online Edition

Using genetically engineered mice, Johns Hopkins and other scientists have shown for the first time that a single kind of cell in the retina seems to detect light for the body's internal clock and for the pupil, they report in a recent issue of Science.

The research represents an important step in understanding how light resets the internal clock, or circadian rhythm, and how the pupil opens and closes in response to light, the scientists say.

To learn how light reaches those brain centers, postdoctoral fellow Samer Hattar created a mouse whose melanopsin protein was partially replaced with another, easy-to-detect protein. Melanopsin is suspected to be light-sensitive but not involved in forming visual images.

The scientists discovered that only a tiny fraction of nerve cells in the retina make melanopsin. These melanopsin-expressing nerve cells, which reach deep into the brain to areas that control the clock and the pupil, join image-producing rods and cones as the only retinal cells that can detect light, the researchers report.

The clock regulates the body's daily cycles, including sleep, hormone production, body temperature and blood pressure. While an individual's natural cycle may be more or less than 24 hours, the 24-hour cycle of day and night keeps the body's rhythm in tune with the environment. Light adjusts the cycle when it gets out of whack, as with jet lag or workers switching to the late shift.

"The melanopsin-containing cells create a light-detecting network across the retina in the mice," says King-Wai Yau, a Howard Hughes Medical Institute investigator and professor of neuroscience and ophthalmology at Johns Hopkins. "The cells seem sensitive to how much light there is and how long it lasts, unlike the cells involved in vision, which detect borders between light and dark."

The findings support the idea that there are two primary groups of light-detecting cells in the eye: one responsible for creating visual images and the other for detecting levels of light, Yau adds.

For their experiments, the researchers created a line of mice with one normal copy of the melanopsin gene and one copy that coded for a protein called tau-lacZ instead. In addition, Hopkins neuroscience graduate student Hsi-Wen Liao developed an antibody against melanopsin. The antibody flags melanopsin in cells, while tau-lacZ lights up the tentacle-like axons of nerve cells in which the melanopsin gene is turned on.

With these tools, the scientists figured out exactly which cells contained melanopsin, whether those cells were naturally sensitive to light and where they connected in the brain. While reports by others have linked melanopsin-containing cells in the retina to the clock's central controller, the new findings show that these cells also connect to other areas of the brain.

"With our technique, we can visually trace single cells and follow their axons for long distances to see where they begin and where they end," Yau says.

The experiments showed that the melanopsin-containing cells (about 1 percent or 2 percent of so-called "retinal ganglion cells") connect to the clock's central controller and to all other parts of the brain known to regulate the clock and the pupil's response to light.

Much remains unknown about melanopsin itself, including whether it is sensitive to light, and about these cells' roles in controlling the clock and pupil, the researchers stress. "The genetically engineered mice give us an important tool to learn how nonvisual functions of the eye depend on light," Hattar says.

Developing and studying a mouse that completely lacks melanopsin will help expand the scientists' understanding of the protein's role in the internal clock, they say, as the protein's absence may dramatically affect the animals' ability to adapt to different light cycles.

"If there is a cycle of light and dark, a normal mouse will adjust its natural cycle to match," Hattar says. "When and how much a mouse runs on a wheel reveals whether the animal thinks it is day or night. If melanopsin is required for resetting the internal clock, however, mice without it shouldn't be able to adjust to cycles of light and dark."

Other authors on the paper are Motoharu Takao and David Berson of Brown University. The experiments were funded by the National Eye Institute and the Howard Hughes Medical Institute.