Time, Love , Memory

by Jonathan Weiner

  Kyriacou knew that repetitions in the fly genome are common, just as they are in the human genome, where it is not unusual to find repeats of as many as 1,000 nucleotides. At one hypervariable site in the human genome, the sequence that repeats is only fourteen letters long, but the repeat length is 600 letters in some people, 1,200 in others, 2,200 in still others. People all over the planet carry so many different lengths of repeats like these that to find two individuals who have several different repeats of exactly the same length, one might have to look at the DNA of many hundreds of thousands of people. The molecular geneticist Alec Jeffreys, who works in the same building as Kyriacou at the University of Leicester, has used these repeats to design the DNA fingerprint tests that are now produced as evidence and debated in courtrooms all over the world.

  Kyriacou was interested in the repeated runs of ACA GGT in the center of period because he knew that the same repeat had been discovered in a clock gene of the fungus Neurospora. This stutter or stammer of DNA is the only sequence that the two clock genes have in common. Fly and fungus last shared a common ancestor at the start of the Cambrian period, the dawn of every great animal type from the invertebrates to the vertebrates. Fungus and fly have been evolving separately for 600 million years, which means that their two branches on the tree of life are separated by a total of 1 billion, 200 million years of independent evolution. Since they still carry this same run of repeated letters in the middle of their clock genes while everything else has evolved and changed, Kyriacou assumed that the repeats must be doing something fundamental and irreplaceable for the clock. This proved to be a good assumption; in fact, investigators have since found these same threonine-glycine repeats in the clockwork genes of yeast, mice, white laboratory rats, and naked mole rats.

  To find out what the repeats might mean, one of Kyriacou’s collaborators, Tony Tamburro of the University of Potenza in Italy, tried synthesizing the repetitive portion of the period protein in his laboratory. Synthesizers were now standard equipment for molecular biologists. Almost everyone had a DNA synthesizer, for instance, and the machine was almost as simple to use as a typewriter with four keys, A, C, T, and G. The biologist typed in the letters, and the DNA synthesizer cobbled together the gene. Many molecular biologists also had a peptide synthesizer. The biologist typed, and the machine cobbled together a protein.

  So Kyriacou’s collaborator, Tamburro, typed up the repeats in a period protein, and the repeats promptly coiled into a helix. Tamburro tried it over and over: always a helix. In a series of experiments, he found that any quantity of these repeats would make a helix and that three repeats is the minimum required to make one complete turn. Kyriacou realized that this might be why his variants tended to differ by three: seventeen in Casablanca, twenty in Bordeaux, twenty-three in Bristol.

  By now Kyriacou was intrigued. He knew that with only one piece of only one gene, he could not hope to understand how the whole clock works. As the drosophilist Peter Lawrence of the MRC Laboratory in Cambridge, England, writes in his book The Making of a Fly, “Attempting to study a process with only a subset of the genes involved carries a risk: like trying to understand how a car engine works from a few pieces—say, a piston, the main gasket and the bolts that attach it to the chassis.” Nevertheless, Kyriacou began to think that the repeats might code for a sort of clock spring, and that flies far from the equator, by evolving one or two extra coils to their springs, might have become, in a sense, slightly overwound. He wondered why the colder flies would have added these coils.

  Kyriacou found by further experiments that at warm temperatures, a clock with a short spring runs at very close to a period of twenty-four hours. That is, it keeps a beat that is very close to the period of the earth’s rotation, without needing to be reset. A clock with more coils in its spring runs faster, so it needs to be reset every day. In warmer temperatures—an oasis in Egypt, an open-air market in Morocco—the short clock spring may be the favored model partly because it runs so close to twenty-four hours and a fly does not have to go through the physiological effort and trouble of resetting it. On the other hand, Kyriacou sees some evidence in laboratory tests that a clock with more springs is sturdier and keeps better time in the cold. In northern Europe, it may be worth the trouble of resetting the clock every day for the sake of having a robust clock that keeps good time in Bristol.

  The short spring and the siesta may be the aboriginal forms of the clock, since fruit flies evolved in Africa. In Africa the greatest danger in the lives of flies is desiccation. They wake each day when it is light enough to see but still early enough for Drosophila the dew lover to get an easy drink. They always take a nap in the heat of the day. A nap is for them what faith is in Hebrew Scripture: “A defense from heat, and a cover from the sun at noon,/a preservation from stumbling, and a help from falling.” The flies in the North have evolved their overwound spring since leaving home.

  Of course, Darwin’s process, which the evolutionist Richard Dawkins calls the Blind Watchmaker, is tinkering and experimenting with the watch even now. In fact, the watch seems to be a mechanism that fascinates—so to speak—the Watchmaker. Kyriacou and his colleagues have studied tens of thousands of descendants from several hundred families of flies. In one study they analyzed almost 40,000 individual period genes on 40,000 X chromosomes. They found that the mutation rate of the DNA that codes for the clock spring is tens of times higher than the average mutation rate of fly DNA. Apparently nature is busily selecting the best clocks all over the planet, selecting “daily and hourly,” in Darwin’s famous phrase from the Origin. In this way the Blind Watchmaker is maintaining exquisitely subtle differences throughout the clock genes of the flies in England compared to those of flies in Africa. The Blind Watchmaker is tuning and fine-tuning their watches every day, everywhere.

  Our own species has not had much time to adapt to the new schedules of artificial light that we have imposed on ourselves all over the planet. The clock watchers in their Fly Rooms sometimes wonder if there may be lessons here for our own longevity. If we do not treat our clocks right, will they kill us or make us sick? This is not just a question of coping with jet lag or with changes between night shifts and day shifts. It is a question of following the dictates of our own personal rhythms from day to day. “If a man does not keep pace with his companions,” says Thoreau in a quotation beloved of dreamers and nappers everywhere, “perhaps it is because he hears a different drummer. Let him step to the music which he hears, however measured or far away.”

  Benzer knows how dangerous it can be to interfere with his own eccentric circadian rhythms. Back in his gypsy days as an itinerant geneticist, when he worked at Oak Ridge National Laboratory in 1948, government rules required him to start work in the morning, like everyone else. He had two car accidents that year on his way to the lab. One fellow scientist who saw Benzer driving at that hour noticed the look in his eyes and determined to take a different route from then on.

  Many people with normal clocks notice themselves fading out at about the same hour almost every afternoon, usually at three or four o’clock. If they are doing something interesting, they can bull through that dimming hour. But often clock watchers, when they are trying to read a dense paper in a field they do not know well, wonder drowsily whether midafternoon may be a window of time when most of us are supposed to be asleep, perhaps because our own species evolved in a hot climate … in Africa, along with the little fly.

  IN 1988, after years of slogging, Rosbash and Young and their laboratories finally began to get their breakthroughs. First, they caught a glimpse of the period gene in action inside the clock. That discovery was made by Kathy Siwicki, a postdoc in Hall’s lab. Siwicki harvested flies hour by hour and checked to see how much period protein they contained. She discovered that the protein’s concentration inside a fly’s head rises and falls with a daily rhythm. In 1990, a student in the Rosbash lab, Paul Hardin, discovered that period RNA also cycles.

  In the rival Fly Room at Roc
kefeller, two of Michael Young’s postdocs, Amita Sehgal and Jeff Price, after screening seven thousand lines of flies, found a new mutant that had something wrong with its sense of time: timeless. The timeless gene is on the fly’s second chromosome. In 1995, another of Young’s postdocs, Mike Myers, working twelve hours and more a day, managed to clone the gene, which meant that the Young lab could now sequence it and read every last letter. The timeless mutant is missing sixty-four letters of code. Once he had cloned the gene, Myers could also check the expression of the gene in the fly heads. He could do just what Siwicki had done: freeze fly heads hour by hour and see if timeless switches on and off too. “It’s definitely the right gene,” he told Young, “and it’s oscillating.”

  So period makes a protein that oscillates; timeless makes a protein that oscillates; and when the clock watchers mixed the two proteins in a petri dish, they formed a strong bond. They mesh the way gears’ teeth mesh in a clock.

  The clock gene was rapidly becoming famous in molecular biology as a kind of exemplar of the work ahead. Everyone interested in genes was interested, at least casually, in the discovery of two working parts of a clock that would mesh. And the same year Sehgal and Price found timeless, Joseph S. Takahashi, a molecular biologist at Northwestern University, found a mouse with something wrong with its sense of time. He found it by Benzer’s method: injecting his subjects with a mutagen and watching their descendants for the odd mouse that seemed to march to a different drummer. Virtually all of his mice were extremely punctual about waking up and running on their exercise wheels, but this mutant’s clock had a period that was more than one hour longer than normal. Takahashi had thought he and his group would have to screen thousands of mice to find something like Konopka had found, but he struck gold with the twenty-fifth mutant mouse—another triumph for Konopka’s Law. Takahashi and his team named their mutant Clock (Circadian Locomotor Output Cycles Kaput). In 1997, Takahashi cloned Clock. Three mouse genes, mper1, mper2, and mper3 (mouse period 1, 2, and 3) turned out to contain long stretches of code that are very similar to the fly’s. Soon after Clock was discovered in the mouse, the corresponding gene was found in the fly as well, a gene now called dClock. It maps to the sixty-sixth region on the fly’s third chromosome.

  For all of the clock watchers on the planet, whether they were working on the clock in flies or mice or ferns or moths, Clock was another cog in the machine, another clue to the working of the clock, which appears to be a mechanism that has been conserved (at least in its broad schematic outlines) since the beginning of time. On blackboards and whiteboards in three or four laboratories, researchers drew the molecular equivalent of a description of the innards of a universal grandfather clock. “We’ve cracked the case,” Hall began telling the reporters who came around more and more often to knock on the old Dutch double doors and ring the little hotel bell of the Drosophila Arms and disturb his terriers. “Almost every article ever written about biological clocks starts out hypnotically, saying, ‘Biological clocks are a total mystery!’ ” Hall would cry as they sat together, knees to knees, in his small office, surrounded by escaped clock mutants, Civil War paraphernalia, the scents of ether and molasses, and the agitated terriers. “They still say that! It’s no longer true, OK? That is to say, it just simply is not true any more! We now have ways to think about how this clock is really ticking. And in fact, I would be willing to say, ‘This is the way it is ticking!’ ”

  In the current model, the genes period and timeless are both switched on inside the nucleus of a nerve cell in the brain. The finger of the angel, so to speak, is always inclined to hold that switch on. So the cell makes the period protein and the timeless protein. When concentrations of these proteins reach a critical threshold, they enter the nucleus and turn off the switch—turn off the very genes that made them. This cycle takes about twenty-four hours.

  Once the switch is off, the period and timeless proteins gradually decay within the cell, and when their levels fall low enough, the gene Hicks back on. The proteins that flick the gene back on are encoded by Clock and by another newfound gene called cycle.

  Any living clock must have three basic working parts. There has to be an input pathway, so that dawn and dusk can reset the clock. Otherwise human clocks, for instance, which have a twenty-five-hour period in caves and windowless rooms, would cycle into and out of phase with the planet. In fact, that is just what happens to many blind people. The seat of the human clock is in the suprachiasmatic nucleus of the hypothalamus, just above the optic chiasm that figured in Sperry’s studies of split-brain patients. Many blind people’s clocks run out of phase with the world, rather like split-brain patients’ when part of their thoughts run independently of the other half.

  Today the rival clockwork laboratories are continuing to look for new cogs and study how they fit together. They are studying the input pathway; the clock mechanism itself; and the output pathway, by which the clock controls the rhythmic behavior of life, from our lying down to our rising up. For molecular biologists and drosophilists, per has fulfilled its promise at last. It is now a model of the way genes work together with the outside world and among themselves to keep a body alive. Philosophers used to ask whether we are born with or invent our sense of time. Now we know that we have clocks woven into every one of our cells. Philosophers also used to ask how we know that the sun will rise tomorrow. In a sense we have that answer built into us, too. The clock is a kind of orrery in the heart of every one of our cells, revolving to help us keep time with our world, a model of the cosmos inside our heads that cycles whether we are in or out of sight of the sun. The revolution of the stars and the seasons is written in the turns of our DNA.

  The clock is also one of late-twentieth-century biology’s best demonstrations of the family likeness of genetic mechanisms, even between animals as far apart on the tree of life as flies and fungi on the one side and mice and human beings on the other. Human beings carry a number of genes that are homologues, close cousins of per’s, and at least one of these human genes comes complete with the same threonine-glycine repeats whose evolution Kyriacou is studying. We also carry a timeless gene, and our timeless protein meshes with our period protein like two gears in a clock. Since the Cambrian period, all over the face of the earth, the clock genes have intermeshed with one another and also with the sunrises and sunsets overhead. We think of our set of genes as a letter in a closed envelope that we receive from the generation before us and pass on unopened to the next. But genes are not like that at all. The essence of these messages is to be read and reread, opened and closed continually, like open letters or scrolls beneath the silver finger of a ceremonial pointer, in a nonstop give-and-take that is the essence of life.

  IN 1998, as these pieces of the clock fell into place at last, a molecular biologist in Switzerland (a country that appreciates fine watchmaking) praised the clockwork story in Nature for its heuristic value, a value that makes it an instant classic in the annals of molecular biology. The now-celebrated gene period also stands as a corrective to the hype that surrounds the gene and the Human Genome Project by showing that single genes are usually not the answer, and that even a consortium of laboratories will often have to struggle a long time to figure out how a gene complex works.

  More often than not, this is the way these detective stories will run in the future. It is the way the study of the obesity gene complex is running, for instance. There the discovery of one gene (in a lab across the hall from Michael Young’s Fly Room at Rockefeller) was followed almost immediately by the discovery of many more that intermesh with it.

  The gene that causes Huntington’s disease is another case that parallels the story of period. After a struggle as long, and often as agonized, as the struggle with period, it has now been cloned and sequenced. It, too, has a conspicuous repeat embedded in it. Most of the repeats in the human genome seem to be as harmless as the whorls on fingerprints. But certain unstable repeats of three nucleotides can produce fragile-X syndrome and myo
tonic dystrophy, both of which cause mental retardation. Certain longer repeated units can lead to cancer and diabetes. And in the mid-1990s—a dozen years after the gene was mapped—molecular biologists began to suspect that it is the repeat in the Huntington’s mutation that does the damage.

  The repetition is a single codon: CAG, CAG, CAG. Human beings carry many different lengths of this repeat, from fewer than twenty to more than forty. CAG is genetic code for the amino acid glutamine. A gene with extra repeats makes a protein with a long string of extra glutamine molecules one after the other. A normal Huntington gene makes a protein with a run of nineteen to twenty-two of these repeats. A mutant protein has from about forty to hundreds of repeats.

  Like the period gene’s protein, the Huntington gene’s protein works with a second protein that acts as its partner. Its partner is a protein called glyceraldehyde-3 phosphate dehydrogenase, which is an enzyme that does many kinds of housekeeping work in the cell. When the Huntington gene’s protein carries many extra repeats, it binds more tightly than normal to its partner, like two gears jamming, in what eventually becomes a fatal embrace. In other human genes, these same CAG repeats can cause other problems, a whole Merck Manual of nervous disorders, including various kinds of ataxia and muscular atrophy.

  In this way, by following the molecules, biologists are beginning to trace some of the first links from human molecules to human behavior, beginning with aberrant behavior, as in Benzer’s Fly Room. But even at the century’s close, molecular geneticists still knew so little about the Huntington complex that Nancy Wexler, a biologist who had helped begin the search for the gene and whose mother had died of the disease, refused to take a test to find out how long her own number of repeats might be. No one could do anything to help her if she had a long number of repeats, and she decided that she did not want to know.