The Sandman's Secrets: Genetics and Gene Expression in Sleep Regulation

Web Sites

Brain Basics: Understanding Sleep
The National Institute of Neurological Disorders and Stroke put together this primer on sleep and the brain.

Journal of Sleep Research
The Journal of Sleep Research, owned by the European Sleep Research Society, is an international journal that encourages important research papers presenting new findings in the field of sleep and wakefulness (including biological rhythms and dreaming). The Journal reflects the progress in this rapidly expanding field, promoting the exchange of ideas between scientists at a global level.

National Center on Sleep Disorders Research
Contains information on research, professional education, patient and public information, and communications.

National Sleep Foundation
The National Sleep foundation is an independent, nonprofit organization that aims to raise awareness of the importance of sleep and awareness. The organization is dedicated to improving the quality of life for Americans who suffer from sleep problems and disorders.

Journal Articles

Ravi Allada

Allada R, Siegel JM. 2008. Unearthing the phylogenetic roots of sleep. Curr. Biol. 18: R670-679.

Lin J-M, Kilman VL, Keegan K, et al. 2002. A role for casein kinase 2α in the Drosophila circadian clock. Nature 420: 816-820. (PDF, 312 KB) Full Text

Pitman JL, McGill JJ, Keegan KP, Allada R. 2006. A dynamic role for the mushroom bodies in promoting sleep in Drosophila. Nature 441: 753-756.

Steven McKnight

Chen Z, McKnight SL. 2007. A conserved DNA damage response pathway responsible for coupling the cell division cycle to the circadian and metabolic cycles. Cell Cycle 6: 2906-2912.

Tu BP, McKnight SL. 2007. The yeast metabolic cycle: insights into the life of a eukaryotic cell. Cold Spring Harb. Symp. Quant. Biol. 72: 339-343.

Tu BP, Mohler RE, Liu JC, et al. 2007. Cyclic changes in metabolic state during the life of a yeast cell. Proc. Natl. Acad. Sci. USA 104: 16886-16891. Full Text

Philippe Mourrain

Appelbaum L, Skariah G, Mourrain P, Mignot E. 2007. Comparative expression of p2x receptors and ecto-nucleoside triphosphate diphosphohydrolase 3 in hypocretin and sensory neurons in zebrafish. Brain Res. 1174: 66-75.

Faraco JH, Appelbaum L, Marin W, et al. 2006. Regulation of hypocretin (orexin) expression in embryonic zebrafish. J. Biol. Chem. 281: 29753-29761. Full Text

Yokogawa T, Marin W, Faraco J, et al. 2007. Characterization of sleep in zebrafish and insomnia in hypocretin receptor mutants. PLoS Biol. 5: 2379-2397. Full Text

David Raizen

Raizen DM, Zimmerman JE, Maycock MH, et al. 2008. Lethargus is a Caenorhabditis elegans sleep-like state. Nature 451: 569-572.

You YJ, Kim J, Raizen DM, Avery L. 2008. Insulin, cGMP, and TGF-β signals regulate food intake and quiescence in C. elegans: a model for satiety. Cell Metab. 7: 249-257.

Zimmerman JE, Rizzo W, Shockley KR, et al. 2006. Multiple mechanisms limit the duration of wakefulness in Drosophila brain. Physiol. Genomics 27: 337-350. Full Text

Ying-Hui Fu

Toh KL, Jones CR, He Y, et al. 2001. An hPer2 phosphorylation site mutation in familial advanced sleep phase syndrome. Science 291: 1040-1043.

Xu Y, Toh KL, Jones CR, et al. 2007. Modeling of a human circadian mutation yields insights into clock regulation by PER2. Cell 128: 59-70. Full Text

Xu Y, Padiath QS, Shapiro RE, et al. 2005. Functional consequences of a CKIdelta mutation causing familial advanced sleep phase syndrome. Nature 434: 640-644.

Paul Franken

Franken P, Dudley CA, Estill SJ, et al. 2006. NPAS2 as a transcriptional regulator of non-rapid eye movement sleep: genotype and sex interactions. Proc. Natl. Acad. Sci. USA 103: 7118-7123. Full Text

Maret S, Dorsaz S, Gurcel L, et al. 2007. Homer1a is a core brain molecular correlate of sleep loss. Proc. Natl. Acad. Sci. USA 104: 20090-20095. Full Text

Wisor JP, Pasumarthi RK, Gerashchenko D, et al. 2008. Sleep deprivation effects on circadian clock gene expression in the cerebral cortex parallel electroencephalographic differences among mouse strains. J. Neurosci. 28: 7193-7201.

David Rye

Bliwise DL, Rye DB. 2008. Elevated PEM (phasic electromyographic metric) rates identify rapid eye movement behavior disorder patients on nights without behavioral abnormalities. Sleep 31: 853-857.

Stefansson H, Rye DB, Hicks A, et al. 2007. A genetic risk factor for periodic limb movements in sleep. N. Engl. J. Med. 357: 639-647. Full Text

Trotti LM, Bhadriraju S, Rye DB. 2008. An update on the pathophysiology and genetics of restless legs syndrome. Curr. Neurol. Neurosci. Rep. 8: 281-287.

David Dinges

Basner M, Fomberstein KM, Razavi FM, et al. 2007. American time use survey: sleep time and its relationship to waking activities. Sleep 30: 1085-1095. Full Text

Chee MW, Tan JC, Zheng H, et al. 2008. Lapsing during sleep deprivation is associated with distributed changes in brain activation. J. Neurosci. 28: 5519-5528.

Kim H, Dinges DF, Young T. 2007. Sleep-disordered breathing and psychomotor vigilance in a community-based sample. Sleep 30: 1309-1316. Full Text

William Hrushesky

Kanabrocki EL, Hermida RC, Haseman MB, et al. 2006. Chronotherapy of ovarian cancer: effect on blood variables and serum cytokines. A case report. Clin. Ter. 157: 349-354.

Oh EY, Wood PA, Du-Quiton J, Hrushesky WJ. 2008. Seasonal modulation of post-resection breast cancer metastasis. Breast Cancer Res Treat. 111: 219-228.

Yang X, Wood PA, Oh EY, et al. 2008. Down regulation of circadian clock gene Period 2 accelerates breast cancer growth by altering its daily growth rhythm. Breast Cancer Res. Treat. Jul 24. [Epub ahead of print]

Fred Turek

Kim Y, Laposky AD, Bergmann BM, Turek FW. 2007. Repeated sleep restriction in rats leads to homeostatic and allostatic responses during recovery sleep. Proc. Natl. Acad. Sci. USA 104: 10697-10702. Full Text

Kohsaka A, Laposky AD, Ramsey KM, et al. 2007. High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab. 6: 414-421.

Turek FW. 2007. From circadian rhythms to clock genes in depression. Int. Clin. Psychopharmacol. 22 Suppl 2: S1-S8.

Speakers:
Steven McKnight, UT Southwestern Medical Center
Ravi Allada, Northwestern University
David Raizen, University of Pennsylvania School of Medicine
Philippe Mourrain, Stanford University

Highlights

• Sleep may be universally conserved among eukaryotes.
• Worms, flies, fish, mice, and humans use some of the same molecules to regulate their internal clocks.
• Experiments with easily cultured organisms, such as roundworms and yeast, provide easy access to the basic machinery of circadian rhythms.
• Studying more complex organisms, such as flies and fish, provides researchers with new hypotheses to test in mammals.

Everybody does it

One of the meeting's recurring themes was the extreme evolutionary conservation of sleep, which seems to be as ancient as it is scientifically opaque. While there is no clear evidence of bacteria snoozing, a primitive version of the phenomenon may be as old as membrane-bound organelles.

In a popular model, the first eukaryotes arose when a cell with a reductive metabolism engulfed a cell with an oxidative metabolism. Instead of digesting its prey, the reductive cell incorporated it as the first organelle, an ancestor of all mitochondria. "Our idea is that the resulting cell that had both of these choices decided not to do both of them at the same time, but instead decided to go back and forth between using reductive chemistry to oxidative chemistry," says Steven McKnight, of UT Southwestern Medical Center.

From that theoretical starting point, McKnight and his colleagues decided to look for metabolic phases in wild-type strains of the yeast Saccharomyces cerevisiae. Culturing this primitive eukaryote with a steady but restricted supply of food, the team found that the cells quickly synchronize their growth, oscillating together between reductive and oxidative metabolism on a regular cycle of four to five hours. The reductive cycle appears to provide a rest, where the cells rebuild the components they'll need for the next replicative and oxidative cycle.

Wild-type yeast cells may have a sleep cycle.

To see whether this "sleep cycle" was really regulated, and not merely an artifact of the nutrient conditions, the investigators looked for genes whose expression levels cycle with metabolism. They found that several clusters of metabolic and replication genes seem to use specific regulatory systems to time their expression.

"Perhaps the expression of all of these genes at a particular phase of the yeast metabolic cycle would cue the cells to be able to divide at a particular gate or particular window of this metabolic cycle," says McKnight. And, using multiple measures of DNA replication, that's exactly what the researchers found. The results suggest that the cycling yeast experience a period of oxidative metabolism, followed by a truly sleep-like phase of reductive metabolism and rebuilding, during which replication occurs. That sets the stage for the next cycle of oxidation.

Getting some compound shut-eye

Purists might argue that the yeast metabolic cycle isn't really sleep, classical definitions of which often include increased arousal threshold, immobility, and specific neural signals—measures that are meaningless for a single-celled fungus. However, many other primitive organisms do meet these criteria.

Probably the simplest animal that demonstrates unambiguous sleep-like behavior is the fruit fly, Drosophila melanogaster. Flies even use many of the same molecular mechanisms as mammals to regulate their sleep and circadian cycles, suggesting that McKnight's sleeping yeast might not be so far-fetched. "I think this conservation that we see in flies and mammals really suggests that sleep was present in a primordial form ... at a genetic level in a common ancestor of both flies and mammals," says Ravi Allada of Northwestern University.

In flies, sleep cycles seem to be controlled by just a few neuronal clusters in the brain. One cluster of "morning" neurons is responsible for triggering a peak of activity early in the day, while a cluster of "evening" neurons triggers a later activity peak. A protein called peptide dispersing factor (PDF) seems to serve as a pacemaker in this system, modulating output from the morning neurons. But how do these neurons coordinate their activity and synchronize it with the circadian clock?

Flies control their sleep cycles with two clusters of neurons.

Using an inducible gene expression system, Allada and his colleagues found that extra PDF gives the flies insomnia. "These flies seem to fall asleep, they just don't stay asleep," says Allada. Conversely, deleting either the PDF or PDF receptor gene makes the flies sleep more during the day.

While that work was underway, another group discovered that the fly version of the GABAa receptor regulates flies' sleep patterns; mutations in this receptor reduce sleep latency and increase total sleep time. "[Those] studies really established that the GABA receptor is playing a similar role in fruit flies as it's playing in mammals and humans," says Allada.

Combining those results with his own, Allada favors a model in which morning cells release PDF, which signals the evening cells to start their own daily cycle. The evening cells may then use the GABA signaling system to turn off the morning cells later in the day. The fly's circadian clock keeps the morning cells off during the night, and activates them in the morning to begin the cycle again.

A nap before molting

The chief allure of studying sleep in flies is that simpler animals are much easier and cheaper to work with than more complex ones. Taking that logic a step further, some researchers are now looking at sleep—or a state very similar to it—in the roundworm Caenorhabditis elegans.

With only 302 neurons, a rigorously defined developmental program, and a life cycle that lasts less than three days, the worm has become one of geneticists' favorite organisms. In the lab, the adult form is continuously active, but developing worms go through four rounds of a quiescent period called lethargus. After each lethargus period, they molt their exoskeletons.

Unfortunately, few researchers have studied lethargus, so when David Raizen of the University of Pennsylvania School of Medicine and his colleagues set out to see if this process had any similarities to sleep, they had to build much of their own equipment. For example, to see if the worms met one of the primary criteria for sleep, increased arousal threshold, the researchers used their own computer imaging algorithm and a robot they'd built out of Lego toys. The robot can tap a petri dish with precisely the same force each time, and the algorithm quantifies the worms' responses.

Roundworm lethargus is indistinguishable from sleep.

While the apparatus looked whimsical, the results were clear: worms in lethargus are much less responsive than other worms. Furthermore, when the researchers continuously stimulated worms that were about to go into lethargus, they could "sleep deprive" the creatures, delaying the onset of lethargus for up to an hour. Worms who enter lethargus after this deprivation are even less responsive than in a regular lethargus. "This shows two features here of homeostasis, the first is that the latency to return to sleep is shorter, and the other one is that the arousal threshold, reflecting the depth of sleep, is further increased from undeprived animals," says Raizen.

Skeptics might still argue that lethargus is simply a byproduct of molting. To eliminate that possibility, the researchers tested two strains of worms with mutations in cyclic GMP-dependent protein kinase (PKG), which is believed to be active in sleep and wakefulness. A more active version of PKG causes more periods of quiescence in adult worms, while a less active version of the enzyme reduces quiescence. "This shows that the behavior of lethargus ... can be uncoupled from the molting cycle," says Raizen.

A visit with Luca Brasi

To round out the menagerie, Philippe Mourrain of Stanford University presented his recent work on Danio rerio, or zebrafish, an inch-long organism that has attracted a growing following in the past decade as a model system. Besides developing from an egg to a fish-like larva in a single day, these animals have transparent larvae, allowing researchers to visualize and analyze neuronal networks using fluorescent microscopy.

Knocking out the hypocretin receptor gives fish insomnia.

In keeping with the general trend for eukaryotes, zebrafish also sleep, sinking to the bottom of the tank for a series of naps during the night. To see whether this process uses the same mechanisms as mammalian sleep, Mourrain and his colleagues looked at the hypocretin system, which includes two ligands and two receptors in mammals. Null mutations that turn off hypocretin signaling cause narcolepsy in dogs, so the system is clearly an important sleep regulator.

Fish have only one hypocretin receptor, and the researchers found that knocking it out causes the mutant fish to sleep 30% less than wild-type animals—hypocretin seems to regulate fish sleep, too. The investigators are now looking at other components of the sleep machinery, including the fish homologs of the GABA signaling components. "We are quite convinced that the sleep factors and the sleep molecular actors are present in fish," says Mourrain.

While we may never know whether a fish, worm, fly, or yeast can dream, all of these systems promise to provide powerful new ways to test theories about sleep. Those results can then inform more complicated studies on mammals.

Speakers:
Ying-Hui Fu, University of California, San Francisco
Paul Franken, University of Lausanne
David Rye, Emory University

Highlights

• Single-base mutations can cause complex inherited sleep disorders.
• The circadian clock and the homeostatic sleep drive have overlapping molecular controls.
• Well-designed gene expression profiling experiments can distinguish purely circadian genes from purely sleep-regulating genes.
• Restless legs syndrome has a genetic component.

Dreams of our fathers

After the opening session's focus on non-mammalian systems, Ying-Hui Fu of the University of California, San Francisco brought the discussion directly to the topic that most sleep researchers ultimately want to understand: human sleep patterns. People are notoriously difficult research subjects, but Fu and her colleagues have been fortunate enough to find several families with well-defined, inherited sleep disorders. Analyzing these groups, they've begun to discover the genetic basis of human sleep and circadian cycles.

One of the researchers' test groups is a family with familial advanced sleep phase syndrome, a disorder that shifts sufferers' sleep cycles by several hours. Some family members fall asleep very early in the evening and wake up in the middle of the night, while others can't get to sleep until the middle of the night, and sleep until early afternoon—they are the extreme versions of "morning people" and "night people."

A few years ago, Fu and her colleagues traced the syndrome to a single amino acid change in a protein called Per2. Homologs of the Per2 gene are responsible for circadian and sleep cycles throughout the metazoan family tree, and the homologs seem largely interchangeable, at least between mammals. A mouse carrying the mutant human hPER2 gene has a shortened circadian cycle and a shifted sleep phase. "So this mouse model really recapitulates the human four-hour phase advance very well," says Fu.

Affected members of kindred 5231 have an A-to-G mutation in CKId that results in a four-hour phase advance.

Per2 is normally phosphorylated by the enzyme casein kinase 1 delta (CK1d), but the sleep-advancing mutation disrupts one of Per2's phosphorylation sites. The researchers hypothesize that reduced phosphorylation translates to reduced protein stability, causing the mutant protein to break down faster and accelerate the sleep clock.

In another project, the team analyzed a family with a different mutation that also causes a four-hour sleep phase advance. In these people, a single point mutation in CK1d reduces the enzyme's activity. Transferring the mutant enzyme into mice creates animals with shorter sleep periods. That dovetails nicely with the Per2 data, suggesting that CK1d phosphorylation of Per2 is indeed a key regulator of sleep.

The same machinery seems to have been conserved across vast stretches of evolutionary time, but its precise effects have changed slightly over the years. For example, putting the mutant CK1d gene into flies produces a longer, rather than shorter, circadian period. "Despite the highly conserved nature of individual components for the circadian clock ... there may be some differences between flies and mammals," says Fu.

The homeostat's got rhythm

Besides discovering the genes that regulate mammalian sleep, researchers are also trying to understand how those genes relate to the two overall processes thought to drive the system. The circadian cycle, or "process C," triggers sleepiness and wakefulness in a regular 24-hour pattern. Meanwhile, a homeostatic sleep drive, or "process S," builds up increasing sleep pressure based on how long it's been since the last rest. Ignoring process C to pull an all-nighter, for example, causes process S to accumulate more sleep pressure, making it harder to stay awake the next day.

Many genes thought to be circadian are actually sleep-driven.

To try to distinguish these two processes at the molecular level, Paul Franken of the Center for Integrative Genomics at the University of Lausanne performed a gene expression screen in mice, and identified about 2000 genes whose levels oscillate with the circadian cycle. However, when he deprived the animals of sleep and then repeated the screen, the amplitudes of most of those oscillations decreased dramatically, and only about 400 genes remained rhythmic. The result sends a cautionary message about overinterpreting gene expression data. "Many of these genes which have been attributed to circadian control are in fact driven by the sleep/wake cycle," says Franken.

The genes that remained rhythmic, indicating true circadian involvement, include Per2 and other well-known clock regulators. However, the overall levels of many of these genes also increase after sleep deprivation, suggesting that the clock machinery and the homeostatic sleep drive machinery may overlap.

That idea fits well with other data, which show that disruptions in clock genes often cause both circadian and homeostatic sleep phenotypes in animals. "It might be that somehow these clock genes are there to sense ... the metabolic need for sleep, and are used in sleep homeostasis as well as in circadian rhythm generation," says Franken.

Legs that don't quit

While the circadian clock and sleep homeostasis may share a relatively small, highly conserved operating system, that doesn't make it much easier to troubleshoot, especially in the clinic. That's because insomnia patients, unlike sleep-deprived laboratory animals, commonly show up with a constellation of other symptoms that may or may not be related to their sleep problems.

That's a situation physicians like David Rye of Emory University face regularly. As an example, Rye cites a particular patient he sees, an 18 year-old female with attention deficit hyperactivity disorder, depression, mood disorders, and insomnia. Her sleep troubles may have a genetic component; several members of her family have also experienced insomnia, but in slightly different ways. "One of the challenges is sort of grabbing a complex disease and being able to decide how you're going to phenotype it, how you're going to measure it," says Rye.

Eventually, Rye found a likely culprit: periodic limb movements of sleep, a key component of restless legs syndrome (RLS). Rye, who also suffers from this condition, says it affects about 10% of the U.S. population, and epidemiological data reveal that it correlates strongly with diseases such as depression, stroke, hypertension, and cardiovascular disease.

Is restless legs syndrome a real condition?

Skeptics have argued that RLS is a figment of drug marketers' imaginations, and Rye concedes that some of the data on this complex syndrome are tough to interpret. "The frequency or prevalence of this disorder varies markedly between different parts of the globe—to some this would be interpreted to be a consequence of an omnivorous pharmaceutical industry in the Western hemisphere," says Rye. He argues that the same data could just as easily indicate that certain populations have a genetic predisposition to a real disease and that the human genome points more to this explanation.

To sort it out, Rye and his colleagues collaborated with researchers at DeCode Genetics, a company that maintains a massive genealogical and genomic database on the population of Iceland. Using ankle-mounted monitors to collect data on limb movements, and gene sequences called single nucleotide polymorphisms to track genetic differences, the researchers identified several alleles that correlate well with RLS. The results have been corroborated in five different populations, providing evidence that RLS is a heritable condition driven by specific biochemical pathways.

Together, the three talks in this session—and the extensive discussion they generated afterward—reveal that researchers have made a good start on uncovering the genetic wiring of sleep and circadian cycles. That's important, as the final session of the meeting revealed, because even minor glitches in these systems can have surprisingly serious consequences.

Speakers:
David Dinges, University of Pennsylvania
William Hrushesky, University of South Carolina
Fred Turek, Northwestern University

Highlights

• Circadian disruptions increase the risk for numerous chronic diseases.
• Individuals vary enormously in their tolerances for sleep deprivation.
• Shift workers suffer significantly higher rates of many serious diseases than their nine-to-five counterparts.
• Sleep schedules may have a dramatic impact on the progression of cancer.

Gotta have it

Many laypeople would consider the key question in sleep science to be: how much sleep do people really need, and exactly what happens when they don't get enough? But as David Dinges of the University of Pennsylvania School of Medicine explained, even grasping the outlines of an answer has been exceedingly difficult.

Even the field's main concept of sleep patterns, the two-process model described by Paul Franken, seems inadequate to explain seemingly obvious observations. For example, someone who runs a marathon or writes a dissertation all day usually feels sleepier than someone who lounged on a beach, but the two-process model makes no allowance for that. According to the model, Dinges says, "what you do when you're awake doesn't really matter, because the circadian system determines your wakefulness, [but] is that really true?"

Indeed, almost all of the speakers at the meeting were careful to present the two-process model as a preliminary idea, which clearly needs some modifications to account for the data. A more serious problem is that sleep studies tend to look at results across groups of people, when anecdotal evidence has long suggested that individuals vary significantly in their responses to sleep restriction.

Individual variations in sleep needs are enormous.

Dinges's studies have now backed up those anecdotes. By depriving healthy volunteers of sleep for varying lengths of time, and testing their response times in a standardized way, he and his colleagues have found that sleep needs vary enormously. "The individual differences are astonishing," says Dinges. In one experiment, for example, one individual's responses were almost unchanged during more than three days without sleep, while other subjects were catastrophically impaired. "We can all fret about sleep debt and everything else, but what [the result] says is that the differences between us matter more than what we do to ourselves," says Dinges.

Those differences get bigger with greater sleep restriction. "People are more and more alike if I give them plenty of sleep. [As] soon as I take the sleep away, they get more and more different," says Dinges. Surprisingly, the differences don't seem to correlate with any other measurement the researchers have been able to apply—a person's response to sleep restriction isn't linked to sex, age, race, IQ, or personality type.

Those night shifts are a killer

Disrupted sleep doesn't just affect one's ability to pass an arbitrary reaction time test, either. According to William Hrushesky of the University of South Carolina, epidemiological data show that shift workers suffer substantially higher rates of breast, prostate, colorectal, and endometrial cancer than regular nine-to-five workers. "These represent, in aggregate, 12 million new cancers a year," says Hrushesky.

Conversely, cancers can also disrupt circadian cycles, and the effect correlates directly with the disease's severity. "The more screwed up your circadian organization is, and the more screwed up your sleep and activity cycles are, the sooner you're going to die from cancer," says Hrushesky.

To explore the connections between these two problems, Hrushesky and his colleagues studied mice living in a constantly lit room, a setting that disrupts the animals' circadian rhythms. In these mice, implanted breast tumors grow twice as fast as in mice with a standard day/night lighting schedule. A dose of melatonin at night diminishes the differences between the two groups, and tumors that lack the melatonin receptor grow at the same rate in both sets of mice.

Constant light at night causes implanted breast cancer tumors to grow faster in mice.

Melatonin ordinarily cycles with circadian rhythms, and researchers are increasingly focusing on it as a key connection between the biological clock and tumor growth. "We avoided melatonin for a couple of decades, but in the last decade we've become enmeshed in melatonin biology, because it seems to be so very important to the host-cancer balance," says Hrushesky. He adds that the results easily justify funding a placebo-controlled trial of melatonin in cancer patients. Bright light or exercise in the morning, and optimally timed meals, might also help boost the effectiveness of other cancer therapies.

In the question-and-answer period, though, Hrushesky conceded that it can be difficult to get regular physicians to take chronobiology seriously. For example, researchers have known for decades that there are optimal times of day to give specific cytotoxic and targeted receptor-mediated anticancer agents. Despite knowing when in the day these agents are less toxic and more effective, anticancer therapy is still given when it is convenient, he said.

A heavy sleep

A different disease is starting to make doctors more aware of the importance of circadian cycles, though: obesity. In recent years, more than two dozen epidemiological studies have shown unambiguous links between sleep disruption and obesity, and both conditions are reaching epidemic proportions in many countries.

"Even the lipid people, who haven't even thought of anything but lipids and biochemistry, are beginning to say 'whoah, maybe there's some connection here that we should be paying attention to,'" says Fred Turek of Northwestern University. Since cloning the first mammalian clock gene in 1997, Turek, Joe Takahashi and their colleagues have begun to focus on exactly that, revealing some surprising new links between sleep, circadian cycles, and metabolism.

Sleep loss and obesity are linked epidemics.

One obvious question was causality: does less sleep make us fatter, or does being fatter makes us sleep less? Mice with a mutant form of the Clock gene sleep two hours less than wild-type mice each day, and the animals also get fat, suggesting that sleep loss and/or circadian disruption causes obesity. However, the reverse also seems to be true. In either genetically obese ob/ob mice or normal mice who eat an obesity-inducing diet, the team saw longer periods of activity and less sleep. The latter model bears an uncanny resemblance to many humans' lifestyles. "You're just taking in different diet, high fat, and you're altering the central circadian clock," says Turek.

Separating the effects of circadian disruption from those of sleep disruption is a trickier problem. To address it, the researchers shifted the sleep schedules of a group of mice every 90 days. The animals could still adapt to their schedules and get the same amount of sleep, but their circadian rhythms were constantly being disrupted. These phase-shifted mice become significantly more susceptible to an intestinal poison than controls kept on a constant schedule. "It's when you've challenged the system that you're going to see more effects," says Turek.

That result seems to fit a general pattern in sleep and circadian cycles, where disruptions often have very subtle but potentially far-reaching consequences. As the meeting's barrage of interesting new results shows, researchers are finally starting to glimpse the biological mechanisms underlying these complex, ancient processes. Translating those results into clinically useful therapies, though, will take a lot of additional work—which, hopefully, will take place on a regular schedule.

Over the past few million years, hominids evolved as diurnal primates, spending most of that history near the Earth's equator. Even as seasons changed, our ancestors' daily schedules remained remarkably constant. They got up at sunrise, hunted, gathered, socialized, then went to sleep at sunset.

For most people, that pattern persisted, essentially unchanged, until the dawn of the 20th century. Then all hell broke loose. Electricity, rapid transit, light-speed communication, and revolutionary new agricultural techniques transformed the human landscape radically. Soon afterward, scientists studying sleep and circadian cycles started trying to figure out what we'd done to ourselves, and how we could function best in the strange new world we'd made.

Can our ancient sleep schedules adapt to the modern world?

Those questions have been tough to answer. Sleep science and chronobiology—the study of biological clocks—have made remarkable strides in the past few decades. But at the second New York Academy of Sciences meeting on sleep disorders, held June 20, 2008, leading sleep and chronobiology researchers conceded that there was a lot more to do.

"The reality is both of these fields have actually not fundamentally helped society deal with the way it wants to live—shiftwork, jet lags, lifestyles of excessive amounts of food, et cetera," says David Dinges of the University of Pennsylvania School of Medicine. During a lively panel discussion at the meeting, Dinges cited an example to drive the point home: "There is not a single dose-response study to answer a question that all governments want to know: how many days off and how much sleep do you need before I can recycle you back to work again, whether you're a truck driver or a pilot or a maritime operator or a [medical] resident?"

While that question remains open, researchers have made remarkable progress on some of the field's other longstanding problems, producing an enthusiastic and wide-ranging discussion both during and after the talks. The presentations fell into three closely related topic areas.

In the Lessons from Model Organisms session:

• Ravi Allada of Northwestern University discussed his work on fruit flies, in which he has identified some components of an oscillating sleep/wake clock in the brain.
• David Raizen of the University of Pennsylvania School of Medicine has demonstrated through his work on roundworms that even these primitive metazoans display sleep-like behavior.
• Philippe Mourrain of Stanford University discussed his work on zebrafish, exploring some of the parallels in the evolution of sleep.
• Steven McKnight of UT Southwestern Medical Center, the meeting's keynote speaker, presented persuasive evidence that even single-celled yeast may engage in a distinctly sleep-like process.

In a session on the Genetics of Sleep:

• Ying-Hui Fu of the University of California, San Francisco, discussed her work studying inherited sleep disorders. Her studies have begun to reveal the human genes responsible for circadian cycles and sleep.
• Paul Franken of the Center for Integrative Genomics at the University of Lausanne discussed his use of gene expression screens in sleep-deprived mice. He has shown that sleep homeostasis and circadian cycles share overlapping molecular machinery.
• David Rye of Emory University has discovered a genetic basis for restless legs syndrome, a widespread condition that often disturbs patients' sleep.

In the Consequences of Abnormalities in Sleep/Circadian Rhythm session:

• David Dinges of the University of Pennsylvania School of Medicine discussed work showing that individual variations can be much more significant than general trends in human sleep-deprivation studies.
• William Hrushesky of the University of South Carolina discussed the links between circadian disruption and cancer.
• Fred Turek of Northwestern University argued that sleep deprivation and obesity are mutually reinforcing epidemics in the modern world, driven by strong connections between the sleep and metabolic systems.

• Is metabolic cycling a core purpose of sleep?
• Why do all types of animal sleep entail reduced responsiveness?
• How much sleep does a person really need to be fully functional?
• To what extent is sleep loss driving the obesity epidemic?
• Can restoring circadian rhythms help treat cancer?