Courtesy Samer Hattar

At top left, an X-gal stained retina from a mouse heterozygous for a LacZ knockin at the melanopsin locus reveals axons coursing toward the optic disc. At top right, a melanopsin antibody labels cell bodies, dendrites and initial axon segments of roughly 1% of ganglion cells in the rat retina. At bottom left, a coronal section from the brain of a heterozygous animal shows bilateral innervation of the suprachiasmatic nucleus (SCN). At bottom right, a stained whole brain cleared with a Benzene-based solution allows visualization of the optic nerve fibers and the optic chiasm with concomitant view of the two SCN nuclei visible in the background.

"I see," said the blind man. Outlandish as it seems, people lacking functional rods and cones can receive and process visual information. At least two decades ago, scientists began speculating that there may be...


A vitamin A-based opsin photopigment, related to cone opsin and the rods' rhodopsin, is the most likely candidate for setting and resetting the biological clock. But while animals without eyes can't photoentrain, many blind people, and rodents lacking rods and cones, still synchronize hormonal, sleep, and other rhythms to a 24-hour light/dark cycle. This implies the existence of a novel opsin in the eye responsible for circadian rhythm.

RNA from a homolog of an opsin photoreceptor, first discovered in frog melanocytes (and thus named melanopsin), was later found to be expressed in rodent and primate RGCs. Some of these cells project to the suprachiasmatic nucleus (SCN) where the master clock is located, so melanopsin became a prime candidate for the circadian photoreceptor.

Building on these results, at least five papers were published in 2001 and 2002, including this issue's Hot Papers.2 David Berson's lab "showed that cells that innervate the SCN could function as autonomous photoreceptors," the Brown University neuroanatomist recalls.3 King-Wai Yau's Johns Hopkins lab, in collaboration with Berson's, created a mouse strain with a LacZ reporter construct replacing the melanopsin gene. Using this and antibodies against melanopsin, they demonstrated that melanopsin is present in the cell bodies, dendrites, and proximal parts of RGC axons that end up in the SCN and other brain nuclei involved with circadian rhythm and the pupillary light reflex.4 That paper also showed that all light-sensitive RGCs contained melanopsin.


But issues remained, including whether melanopsin-containing cells were necessary for nonimage-forming photoreception, and whether melanopsin was itself the photoreceptor or an accessory protein. Several labs raced to find an answer to the first question.

Three groups – Yau's; Iggy Provencio and researchers from the Uniformed Services University in Bethesda, Md., and the Novartis Research Foundation in San Diego; and a third group consisting of Norman Ruby and colleagues at Stanford University – each created melanopsin knockout strains.567 In what Provencio calls "the first functional characterization of what's going on here," the researchers collectively found that mice "can restrict their activity to the night portion of the light/dark cycle, which is what nocturnal rodents typically do, [but] when you look for more subtle effects of the circadian system, such as the effect of single light pulses, we saw real deficits in these animals," including a deficiency in the pupil's response to bright light and in the ability of light to moderate activity, a process known as masking.

Clearly the nonimage-forming pathway is wholly dependent neither on melanopsin nor on rods and cones. "Was there a possibility that there was even something else, [perhaps] more photopigments?" asks Ruby. To answer this question, the other two groups crossed mice lacking melanopsin to mice without functional rods or cones.89 They found that although the triple-knockout animals have an intact retina, they fail to show any significant pupil reflex; the animals also failed to entrain to light/dark cycles, and to mask.

"We finally showed that the melanopsin gene, the rods, and the cones, can fully account for the circadian light responses and pupillary responses," notes chronobiologist Russell Foster, a coauthor with Yau.


Meanwhile, researchers have been extending the anatomical bases for these physiological results. Using Yau's LacZ mice, Berson and colleagues have shown that melanopsin-expressing RGCs project to areas of the brain implicated in sleep regulation, masking behavior, and gaze control. Cliff Saper's Harvard University lab reached similar conclusions using in situ hybridization.10

Dennis Dacey of the University of Washington, Seattle, and collaborators set out to replicate the findings in a primate model. Dacey says that in work yet to be published, they have "confirmed that these neurons that give rise to this specialized circadian pathway exist anatomically in human retina and macaque."

But there may be important differences. In the primates, for example, melanopsin-expressing RGCs receive significant input from rods and cones, so that the melanopsin system is not even noticeable in intact animals. The two visual pathways are "merged into one," Dacey says. "This single pathway conveys rod, cone, and this novel circadian signal all to the SCN and the other places that are involved in using this signal."

Dacey's work also implicates a role for the main relay for visual perception. But that raises another question for Dacey: "Does this signal actually get into the primary visual pathway, and how is it utilized by the primary visual pathway?"

Foster, at Imperial College London, has evidence that the intrinsically light-sensitive, melanopsin-containing cells (only 1% of all RGCs) also may not be acting alone in rodents. Gap junctions connect these cells into a "syncytium of coupled ganglion cells," he says. "Why that should be, we don't know." He wonders whether different classes of ganglion cells project to different areas of the brain, or if the information is integrated and synthesized to extract different bits of brightness information such as amplitude, duration, or rate of change.


"We still don't know what melanopsin does," points out Yau. The photopigment likely absorbs light and then signals downstream, he adds, but "there are still open questions, the answers to which are still unclear."

For example, a University of Maryland group expressed melanopsin in heterologous tissue-culture cells, and showed the cells to be photosensitive.11 Yet the system's absorption spectrum does not seem to match those of the RGCs. Researchers say they have reservations about the data, primarily because the pigments in this system behave peculiarly in several ways. Provencio, Yau, and others say those experiments need to be independently confirmed.

That issue may be resolved by one of the many labs currently probing the biochemistry of the nonvisual phototransduction system. Provencio, for example, uses frog melanophores as a model system to discover the immediate downstream components of the melanopsin-signaling cascade. Several investigators use microarray and other technologies to tease out which genes are affected by light, and which oscillate with the circadian clock.

Those seeking the roots of melanopsin's influence on circadian rhythms say the research could profoundly affect the understanding of human health. "Much of our physiology is being modified by environmental brightness," says Foster. The circadian clock influences not just sleep cycles, but hormonal release, mood, alertness, cognitive performance, and activity level. Even common phenomena such as the "winter blues" (seasonal affective disorder, or SAD) and jet lag are related to the intensity, duration, spectrum, and timing of light that we experience. Provencio speculates that an understanding of how the clock is set may allow for the development of light and pharmacological therapies to reset it as needed.

Josh P. Roberts tcwriter@msn.com is a freelance writer in Minneapolis, Minn.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!