Internal Time: Chronotypes, Social Jet Lag, and Why You’re So Tired

by Roenneberg, Till

  The first graph shows the results of an experiment that lasted thirteen days.5 Each strip, one above the next, represents the animal’s wheel-running activity during one day. Since each centimeter on each strip is equivalent to one hour of the day, one can easily see at what times the animal was running in the wheel. The animal started to run at approximately the same time every day, so that the onsets of the broad ink bands fall onto a straight vertical line. In this case, the body clock produces days of exactly twenty-four hours, and you can probably guess by now that in this example the animal was kept in a light–dark cycle. If the activity of the bunker subject, who lived through twenty-five-hour days, were recorded in a similar way (there were contacts in the bunker’s floor), the data would look quite different.

  On average, the subject’s activity starts an hour later every day, falling on an invisible oblique line tilted to the right. The disadvantage of plotting the data this way is that the regular bouts of activity are split into two components once they run into the edge of the cardboard. This is why clock researchers use “double plots.” In the early days, researchers would photograph the original cardboard and develop two prints that were glued next to each other on a new piece of cardboard, the right half being displaced upward by one day. In this way they would be able to see all activity bouts as continuous broad bands.

  If we would plot the activity of a bunker subject whose internal days were shorter than twenty-four hours, the activity onsets would fall on a line tilted to the left since he or she would get up earlier every day (in reference to local time). This is actually what mice do when they are kept in a dark, time-free environment. The final graph shows an experiment recording the activity of a mouse that was kept under a light–dark cycle during the first seven days (indicated by the black and white bar at the top of the graph) and was then released to constant darkness, whereupon its activity rhythm began to run free.

  A rodent’s wheel-running activity. In this case, the rodent ran on the wheel at about the same time every twenty-four-hour day.

  The bunker subject’s daily activity level for his twenty-five-hour internal days.

  A double plot shows bouts of activity as continuous bands without interruption by midnight.

  You can see that the activity on the first day in constant darkness makes a little “jump to the left.” This is not because the subject’s body clock suddenly jumps in time but because a mouse simply doesn’t like to run in the bright laboratory light, even if its body clock “tells” it to be active. As soon as the lights go off, however, the mouse starts to run.6 To estimate where the activity would have begun if it hadn’t been suppressed by light, clock researchers look at the rhythm when the clock was released to constant conditions and follow its path back to when the animal was still in a light–dark cycle (see the dashed lines). In this mouse’s case, this method shows that the clock’s signal to start running came approximately one hour before lights went off.

  This rodent was released from a light–dark cycle to constant darkness, whereupon its activity rhythm began to run free (at the intersection of the dashed lines).

  At last, you are equipped with enough information to readily understand the results and conclusions of Russell Foster, whose experiments approached the problem of how light synchronizes the body clock by a very straightforward question: which receptor type translates the information about light and darkness to the clock? He used mouse strains that have no rods and submitted them to experimental conditions much like those shown in the last figure. The results were easy to interpret. These mice could be synchronized just as well by light–dark cycles as mice with a perfectly intact retina. So rods were apparently not the receptors informing the clock about night and day. He then took mouse strains that had no cones and again found that these animals synchronized as well as intact or rodless mice. Once more, the results were easy to interpret. The clock apparently used both rods and cones to “know” about light and darkness. Without rods, it used cones; without cones, it used rods. So all that Foster had to show was that the clock of mice lacking both rods and cones could not synchronize to light–dark cycles. But when he tested mice lacking both receptor types, he was in for a big surprise: the activity recordings were essentially the same as in the previous experiments.

  But maybe the clock doesn’t even need the eyes to “see” the light. Perhaps it uses some other light-sensitive tissue—for example, the skin.7 Around the same time Foster performed his experiments, another clock researcher showed that one could set the human clock by shining light onto the back of the knee. But for several years, other clock researchers failed to replicate these results, clearly demonstrating that the clock only got light information through the eyes. This was already well known from numerous animal experiments—the clock in eyeless animals didn’t synchronize to light–dark cycles. Thus, the only conclusion from Foster’s experiments was that there must be another, hitherto unknown light receptor in the eye.

  This hypothesis raised a big red flag in the community of eye researchers, who thought that everything one could possibly know about the eye of mammals was already known. Many ophthalmologists refused to believe that an unknown light receptor had been discovered by a clock researcher. A couple of years later the enigmatic receptor was eventually identified. When the Foster experiments were eventually repeated with mice lacking all receptor types—rods, the three cones, and a new receptor—their clocks finally stopped synchronizing to light–dark cycles.

  The newly discovered light receptor is related to a receptor that allows amphibians to rapidly change their skin color. Rods and cones are actual cells embedded into the retina. The conversion from light to cellular signals is a biochemical process involving proteins that can catch photons (light “particles”) with the help of pigments, such as carotenoids in red carrots or chlorophyll in green leaves. Carrots look reddish because their carotenoids absorb blue light, so that the remaining, reflected light appears red, whereas leaves absorb both blue and red light and therefore look green. Both rods and cones use the same type of protein, called opsin, but the opsins in rods and in each of the cone types are slightly different, absorbing light of different wavelengths, and thus enable us to identify colors. The new photoreceptor serving the mammalian body clock is also an opsin. Because it is part of those amphibian cells (melanophores) that enable the skin to turn dark rapidly, the new opsin receptor was called melanopsin.8 Light reception by melanopsin does not use specialized cells such as rods and cones, which collect information about the exact where and when of a light stimulus. Melanopsin is spread widely across the retina in those nerve cells that give rise to the long extensions that bundle together to form the optic nerve.

  When scientists want to fathom a biological phenomenon, they of course want to know where the phenomenon (or its control) is localized. The question about the localization of the body clock was pursued extensively in the 1970s, and the outcome depended very much on the species investigated. In plants, a distinct localization was never found. The clock appears to be everywhere—in leaves, stems, and roots. In animals, the clock was found in the brain. In insects, it resides in a small number of highly specialized neurons; in slugs, in neurons at the base of the eye. In reptiles and many birds the clock resides in an annex of the brain called the pineal.9 The pineal is a gland that produces the hormone melatonin, which serves as a signal for darkness.10 In birds, amphibians, and reptiles, the pineal gland is sensitive to light, which it receives directly through the skull. In mammals, the pineal has lost its ability to perceive light but still produces the hormone melatonin at night (notably, in both night- and day-active animals). Melatonin makes us (and probably all other day-active animals) sleepy and a bit colder, while the same hormone has the opposite effects in nocturnal animals. But melatonin does not act just like a normal sleeping pill; it can also affect the body clock so that a regular prescription can synchronize its rhythm. You may have wondered how we can help people like Harriet in her existence as a p
eriodic shift worker. Melatonin is certainly an option and has been used successfully in synchronizing the body clock of totally blind people to the twenty-four-hour day.

  In mammals, the clock was found in a small group of neurons above the optic chiasm. A small brain area dedicated to a specific function is called a nucleus, and since this clock-nucleus is located directly above the optic chiasm, it is called the suprachiasmatic nucleus (SCN).11 The SCN is quite a remarkable little piece of tissue. With no more than approximately twenty thousand cells, it appears to contain all that is needed to make the body clock tick—to keep internal time.12 If the SCN is removed from a rat or a hamster, its wheel-running activity immediately loses its daily regularity. If an SCN is implanted into the brain of an animal that had its own SCN removed, its wheel-running activity becomes rhythmic again with the expected circa-twenty-four-hour period.13 The SCN is the main center of the body clock not only in hamsters, rats, and mice but also in humans. Patients with lesions in this brain area have great difficulty in keeping to regular sleep–wake and activity–rest schedules.

  I have covered a lot of ground in this chapter, beginning with the sleeping difficulties of a blind woman, then discussing the discovery of a new light receptor, specialized not for visual tasks but for the nonvisual task of generally sensing night and day, and finishing with the location of the clock’s centers. But there is still an open question—the one that Harriet asked herself. Why does the body clock of some blind people remain synchronized to light and dark while that of other blind subjects runs free? We still don’t know the full answer to this question, but we do know that the body clock of people who have no eyes almost always runs free. If eyes lack rods and cones but can still unconsciously sense light by an intact melanopsin system, the clock can still be synchronized. An unconscious light perception may seem strange, but it does exist. Russell Foster investigated blind subjects who had no conscious light experience but whose body clock was still perfectly well synchronized. To identify this unconscious light perception, he presented them with light stimuli accompanied by a short sound. Sometimes (randomly), he also presented the sound alone, without an accompanying light stimulus. He encouraged his subjects to simply guess, even if they had no conscious experience of light, whether or not they sensed the light when they heard the sound. The results of these experiments showed that the subjects “guessed” significantly better than random. Their brain must have registered the light even though they did not consciously experience a sensation.

  The clock of individuals whose eyes lack rods, cones, and melanopsin tends to run free, but some apparently remain synchronized. The reason for this may lie in the fact that the internal days of these blind people are already very close to twenty-four hours, so that other signals—activity or meals that are normally too weak to set the clock—might be strong enough to “nudge” it into synch with the outside world. Although Harriet in this chapter’s case still has eyes, she cannot perceive light via any of the receptor types. Her free-running body clock is about one hour slower than the rotation of earth, so her internal time is not synchronized to her social schedules, and she has to live her life as a perpetual periodic shift worker.

  7

  The Fast Hamster

  A new shipment of hamsters from the Charles River Breeding Laboratories had arrived. Christopher was in the process of cataloguing the new animals. When he had entered them all into his files, he placed each one of them into a separate cage equipped with a running wheel. His first task with newcomers was to record their circadian activity rhythms—a routine to check whether they were good or bad runners. A couple of weeks later, he sorted through the pile of activity recordings and made his usual list: “good runner” or “bad runner.” A typical activity recording for hamsters living in constant darkness, one which Christopher had seen hundreds of times before, shows a free-running rhythm with a period close to twenty-four hours.

  Halfway through the pile of double plots, he caught his breath. Hamster #31M18, a male, showed a very unusual activity pattern—different from every other hamster activity recording he had seen. The period of this hamster’s clock was only twenty-two hours! Christopher immediately went to the constant condition chamber and opened the box of four hamster cages that contained the label #31M18. He wondered whether the animal was sick and was even prepared to find a dead hamster in the cage. Hamster #31M18 appeared to be sleeping, so he took it out of its cage and gently stroked its fur. It was a pretty sleepy hamster, but it did not appear to be sick. After placing it back in its cage and closing the lid of the box, he went to check the timer of channel 361, which was responsible for recording the wheel-running activity of the unusual specimen. He ran some tests, but the timing of the recording device appeared to be normal.

  A typical activity recording for hamsters living in constant darkness—a free-running rhythm with a period close to twenty-four hours.

  Hamster #31M18’s double plot shows a period of only twenty-two hours.

  Christopher went back to his office and configured his computer to display the activity of channel 361 online. The double plots he had looked at earlier were already a couple of days old, so he checked the recording of the last couple of days and found that the unusually fast activity rhythm of hamster #31M18 had persisted up to now. The animal’s tiredness was to be expected because it had stopped using the wheel only a couple of hours earlier, which meant that it must have been in deep sleep when he had taken it out of its cage. For the rest of the day, Christopher couldn’t concentrate on his routine lab work, pondering the short circadian period of the newcomer. In the late afternoon, at the daily group meeting with the head of the lab, he waited until all of his colleagues had presented their reports before pulling out the double plot of hamster #31M18. As he had anticipated, all participants were clearly astonished, and he thoroughly enjoyed being the center of attention. The afternoon meeting took much longer than usual and produced several important decisions. The behavior of the unusual newcomer was to be tested under a light–dark cycle (the group decided to simulate a summer day of fourteen hours of light and ten hours of darkness). After three weeks, the hamster would be transferred to the breeding facility to mate with three females that showed the usual hamster rhythms, very close to twenty-four hours in constant darkness.

  The outcome of the recordings in a light–dark cycle showed that hamster #31M18 synchronized perfectly. There was, however, one striking difference: the newcomer started his days about four hours earlier than “normal” hamsters. Several weeks later (fortunately, the pregnancy period of a hamster is only about seventeen days), the first offspring could be tested both for the speed of its clock in constant darkness and for its ability to entrain to a light–dark cycle. It turned out that half of the offspring had inherited the clock of their father and the other half that of their mother. Of those that resembled their father, only a few were able to synchronize properly. For several days, they started their activity in synchrony with lights turning off (hamsters are nocturnal, active at night) but then suddenly got up earlier every day by several hours, as if their body clock were ignoring the light–dark cycle, thereby running free with an abnormally fast rhythm.1 Once their activity again came close to the dark portion of each cycle, it seemed as if they were able to synchronize to the light–dark cycle for a couple of days, but then they broke away again, displaying their very short rhythms. Few of the offspring that had such short activity rhythms ever synchronized and—like their father—they began their days about four hours earlier than “normal” hamsters. The team proceeded to do more breeding with the offspring. When they eventually crossed two animals with a fast clock, they discovered that all of the resulting offspring had an even faster clock than their unusual ancestors.

  As much as biologists like to know the anatomical location of the phenomenon they investigate, they also want to know which genes are involved in its function. Scientists have only recently dared to look for genes that are responsible for biological qu
alities (traits) that have more complex genetics than skin or hair color. The biological clock is certainly considered such a complex trait, and for exactly that reason, one of the pioneers of clock research, Colin Pittendrigh, was initially quite skeptical about whether scientists could identify genes that were involved in the biological clockwork. Seymour Benzer was one of the first scientists who dared to look for genes responsible for complex traits, and he met with a lot of skepticism when he approached these questions relatively late in his already successful career.2 Together with his student Ron Konopka, he set out to look for “clock genes.” They found that mutations in a single gene—which they called the period gene—drastically changed the free-running rhythm of the fruit fly’s clock.3 Mutations at different locations within the same gene produced flies that lived either very short or very long days in constant conditions. Other mutations even rendered the flies arrhythmic. Shortly after the discovery of the first clock gene in the fruit fly, genes with similar properties were identified in other organisms. The next in line was a bread mold.4 The identification of clock genes in mammals took another two decades, however.