American Scientist

At one point or another, we’ve all hit the wall of sleep: the point when you are so exhausted that you cannot stay awake, no matter how bright it is outside or how much caffeine you have consumed. Until recently, though, no one understood the exact molecular mechanisms responsible for this ubiquitous experience. Circadian biologists have long been aware that exposure to light can shift a mammal’s sleep-wake cycle, but they haven’t been able to work out why light exposure becomes ineffectual when an animal is sleep-deprived. Sleep researchers have also known that caffeine delays sleep by blocking the action of a molecule called adenosine. As an animal uses energy while it is awake, adenosine builds up in the body—for example, through the breakdown of a key metabolic molecule, adenosine triphosphate (ATP)—and induces a sense of sleepiness. But until recently, no one had determined how adenosine made that happen.

A study published in Nature Communications in April, led by neuroscientist Aarti Jagannath, circadian biologist Russell Foster, and pharmacologist Sridhar Vasudevan, all at Oxford University, marks a major advance in untangling the roles of adenosine and light in sleep. In the process, the researchers discovered a drug that may mimic the effects of light exposure, with potential applications to health problems ranging from sleep dysfunction after eye injuries to schizophrenia. This interdisciplinary realm of research spanning the team members’ careers has led them to the realization that the sleep-wake system draws from all the key neurotransmitter systems, and that the sleep-wake cycle is a global brain event.

We all have a built-in clock system that keeps our sense of time and determines when we are active and when we are sleepy. Light entrains this clock system when it is detected by specialized nerves in the eye, called photosensitive retinal ganglion cells, which were first identified in 2002, setting off a flurry of research on light entrainment over the following two decades. These nerve cells send signals to turn on key genes called Per1 and Per2 in the brain’s “master clock,” the suprachiasmatic nuclei. This master clock in turn regulates through molecular signaling many aspects of the clock system in various tissues throughout the body.

Animals are particularly sensitive to light at dawn and dusk, when exposure to light can delay or advance our sleep-wake cycle—a trait that we can use to help alleviate jetlag (see “Adapting Your Body Clock to a 24-Hour Society” November–December 2017). Diurnal species experience big delays in our sleep-wake cycles when exposed to light at dawn, and small advances at dusk. Nocturnal species such as mice are the opposite: Light exposure at dusk delays wakefulness substantially, whereas at dawn it brings on sleep a bit sooner (see the phase response curve in the image below, top panel). “A mechanistic explanation for that wasn’t appreciated before,” Foster says. “But we can explain that phenomenon almost exclusively within the context of adenosine.”

What the research team found was that as adenosine builds up during the time one is awake, it triggers a signaling pathway that results in feeling sleepy—the same signaling pathway, indeed, that light acts upon. Light turns on this pathway (though it has a mechanism for turning itself off, so that the clock’s sensitivity remains heightened for a short period); adenosine inhibits this pathway. Adenosine receptor antagonists such as caffeine or the drug Jagannath’s team tested disrupt this pathway, so that the feeling of sleepiness is delayed.

“One of the interesting outcomes of this paper is that they show how the chemical adenosine is able to code for how tired you are, and that can then feed back to the circadian clock to influence how the clock responds,” says Michael Antle, a psychologist at the University of Calgary who has also studied circadian clocks and adenosine and who was not involved with the study. “You respond one way when you’re tired, and in a different way when you’re not tired.”

Jagannath and Foster first teamed up as they tried to figure out the molecular pathways by which light influences circadian rhythms. They later brought Vasudevan into the collaboration to explore drugs that could act on the same pathways. The team started off by looking at compounds that shift the clock in cells in culture. They noticed that drugs that disrupt the binding of adenosine to its receptors had big effects on the cells’ expression of biological clock genes. “This was around the time when these studies on caffeine were coming out, showing that caffeine could affect circadian rhythms in humans, and also in mice,” Jagannath says. Because caffeine is an adenosine receptor antagonist (that is, it inhibits adenosine from binding to its receptors), she and her team wanted to figure out how adenosine and the clock system were talking to each other. They found that adenosine signaling is a fundamental part of the machinery that regulates the body’s sense of time.

“This space of adenosine receptor antagonists is fairly well explored in terms of the pharmacology,” Jagannath explains. “There are drugs out there that are in the clinic, or at least in clinical trials, in this space. We said, ‘Can we try any of them, and see if they have an effect on shifting rhythms in mice?’”

What Jagannath’s team worked out is that light and adenosine both act on a two-pronged signaling pathway affecting the expression of Per1 and Per2: Adenosine inhibits the pathway, and light or adenosine receptor antagonists activate it (see infographic above). These two prongs (cAMP and AP-1) come together to result in the full expression of clock genes Per1 and Per2, which are responsible for shifting the clock. Once the team worked out the pathway in cell culture, they moved to testing how an adenosine antagonist drug affected the sleep-wake cycle of mice.

Just as would be expected with light exposure at different times of day, giving the drug to mice at early dusk caused their clock to shift later, whereas giving it in the middle of the day (when mice are usually asleep) caused their clock to shift earlier. “Importantly, if you antagonize the A1 and A2A receptors [two adenosine receptors] in combination with the light pulse, the light pulse will have a much bigger effect on the clock,” Jagannath says. (See graph on the bottom right of the infographic, above.)

It took about five years to work out the molecular pathway by which adenosine and the biological clock interact. “The signaling pathways that we decoded turned out to be not as simple as we thought,” Jagannath says. “For example, we needed to do a transcription factor binding assay, and that involved assaying 3,000 different transcription factors with a high-throughput screen, which Ueli Schibler at the University of Geneva had developed. The actual delineation of that pathway was a monumental amount of work. Part of what’s been holding this field back is that people would have tried smaller experiments on different sides and seen conflicting things that they couldn’t quite put together. We needed to hammer at it for a long time before we managed to understand what was going on at the full scale.” The work also brings together two disciplines that, as Foster puts it, “simply didn’t go to the same meetings.” But finally bringing together sleep research, which has a medical history, and circadian biology, which has focused on ecology and evolution, has yielded exciting returns. “This paper shows how you can consider advances from both fields, put them together, and make the bigger picture much clearer,” Jagannath says.

The circadian rhythm affects not just sleep but such things as metabolism, mood, and alertness.

• Facebook
• Twitter

The team is now working with Blind Veterans UK to find clinical trial participants who have lost the ability to maintain a consistent internal clock. “The consequences of losing the eyes completely is getting up a bit later and later and later, about 10 or 15 minutes each day,” Foster explains. “Can we give the drug to those drifting individuals and fool their clock into thinking it’s seen light?” Although many people who are born blind have photosensitive retinal ganglion cells and can entrain their biological clocks even though they cannot see, people who sustain eye injuries, such as blinded veterans, often have these nerve cells damaged as well.

Antle notes that differences between mice and humans need to be considered when interpreting this study’s results. “Adenosine levels in a mouse will be highest in the later part of the night, when they are exposed to dawn light, whereas for us, our adenosine levels will probably be highest around dusk,” he says. “So the responses they’re seeing [in mice] can be different than what you might expect in a person.” He also points out that “adenosine pharmacology is really complex,” and that more work is needed to fully understand how this system works in mice and people, and where in the body the drug is acting to result in the outcomes the Oxford team has observed.

The team is planning to begin clinical trials in blind veterans in the fall through their spinout company, Circadian Therapeutics, because the drug they tested in mice has already been shown to be safe in humans when tested for another purpose. “The adenosine A1 and A2A antagonist has a rich history of evaluation for Parkinson’s disease,” Vasudevan says. “However, for those disorders, the drug failed in phase 3. But what we knew was that the drug was safe for human use. That’s basically saved us 5 to 10 years in the process and a lot of money as well.”

Although caffeine is also an adenosine antagonist, the authors say it doesn’t work well for treating dysfunction in the regularity of the sleep-wake cycle. “Caffeine has some other effects than just engaging with adenosine receptors,” Vasudevan says. “Its half-life is between four and six hours. With our drug, you hit the receptor, make the change you want, and then it’s out of the system.”

The authors plan to explore treatments for not only those who have lost their eyes, but also other health problems that include issues with sleep regularity as a symptom. Foster’s lab and colleagues have made strides in the past 10 years showing that disrupted circadian rhythms are an aspect of many mental health disorders, such as schizophrenia and bipolar disorder. “There’s a genuine mechanistic overlap between the pathways that generate normal sleep and the pathways that generate normal mental health,” Foster says. “Daniel Freeman [a psychiatrist at University of Oxford] was able to partially stabilize sleep-wake in schizophrenia [patients] and reduce levels of delusional paranoia by 50 percent. That suggests that the sleep-wake systems represent a new therapeutic target.”

What these researchers have come to appreciate is that if you have a neurotransmitter defect that predisposes you to mental illness, it almost certainly affects sleep. Exactly how the underlying causes of such disorders are related to the sleep systems is not yet known, but is an exciting area of research. “Whenever there’s a circadian rhythm disorder, often it manifests as sleep disturbance,” Vasudevan says. “When you go to your GP and say, this is a problem, you get a hypnotic or a sedative that puts you to sleep. The circadian rhythm extends far beyond sleep. It controls metabolism, your mood, how sharp you are at what time of the day, your alertness, all kinds of things. But your existing medication doesn’t fix any of that. The fact that we can make a fundamental drug that can treat the underlying cause of the disorder is exciting, because it’s a whole new treatment paradigm.”

Listen to podcast (transcript below) with circadian biologist Russell Foster: