FIGURE 2.1. Kathy Parker preparing a patient for a sleep study.

  Source: Age Fotostock America.

  PHYSICIAN EXTENDER

  Describing Parker as a “physician extender,” a term sometimes used for nurse practitioners, understates her accomplishments. Her father, a chemical engineer, drilled her and her sisters on the periodic table when they were little. When she was hospitalized for a kidney condition as a five-year-old, a nurse she admired inspired her to pursue nursing as a career. Parker’s specialty was treating patients with kidney disease, and she became interested in sleep while working on a dialysis unit at the Atlanta Veterans Affairs Medical Center, close to the Emory campus. “When you take care of dialysis patients from day to day, there’s not much you can do to fix them, so the focus is more on quality of life,” she recalled. “They would complain bitterly about their ability to sleep, and it seemed to be worse on dialysis days.”

  Parker’s observations, along with the experience of being sleep deprived after having her first child, nudged her to go back to school. For her PhD at Georgia State, which she finished in 1990, she analyzed her kidney disease patients’ sleep and dream patterns.1 Soon after finishing her degree, Parker walked into Donald Bliwise’s office at Emory and asked to join his efforts. Originally a PhD psychologist, Bliwise was part of a cohort of sleep researchers who entered the field when it was young. Captivated by research on dreams as an undergraduate, he had trained with leaders in the field at the University of Chicago and Stanford and then was recruited to Emory to develop the sleep medicine program in the Department of Neurology. His own research focused on sleep disturbances connected with aging and with neurodegenerative diseases such as Alzheimer’s and Parkinson’s.

  In a newspaper profile of Parker, Bliwise said he viewed their relationship as mentor-mentee at the start and later began to see her as a peer.2 Parker carved out a niche probing the interactions between dialysis and sleep, finding that her patients’ sleep could be improved by slightly lowering the temperature of dialysis fluids, whose warmth appeared to fuel their insomnia.3

  Parker also joined day-to-day operations at Emory’s sleep clinic. This involved a step back—at least, temporarily. “I asked Don and Dave whether I could come into the clinic, just to observe,” Parker said. “At the start, I would sometimes run errands for them, or take care of blood work, or help transfer patients.” Parker was used to more autonomy working at the Veterans Affairs Medical Center. Under Georgia state law, however, nurse practitioners were more dependent on a supervising physician than in other states and could not prescribe medications such as opioids or stimulants on their own.4

  In 2001, Parker passed an exam given by the American Board of Sleep Medicine. She was one of only a few nurses in the country with this credential, which meant she could evaluate and score overnight sleep tests.5 Patients with kidney disease were referred to her, and she began seeing others with a variety of conditions.

  In her clinical duties, Parker’s supervisor was David Rye, who had come to Atlanta at the same time as Bliwise. Asked about her first impressions of Rye, Anna said: “I remember this big tall guy in the doorway. He was very matter of fact, just saying: ‘Let’s try this.’ She [Parker] and Rye were the first medical professionals I met who didn’t think I was making it up, or told me I just had to live with it. They were great bookends. The two of them were very different, but he really paid a lot of attention to what I was experiencing. It seemed to consume him—he wanted the answer.”

  RESTLESS SPIRIT

  Although Rye later became known for Anna’s case, he was occupied with other research concerns that summer. He recalled later: “Kathy came in and said Anna’s back, she’s still really sleepy and she’s worried she might lose her job.” For Rye, insight into Anna’s condition required a change in thinking: away from focusing on molecules in the brain that keep us awake and attentive and toward those that clear the path to sleep. At the time, his major research interest was not hypersomnia but restless leg syndrome (RLS). From 2004 to 2007, Rye was chair of the medical advisory board for the Restless Leg Syndrome Foundation.

  Since Rye had RLS on his mind so much, it is not surprising that Anna was initially treated for it, before the flumazenil experiment. Anna called herself “an inveterate leg wiggler,” to the point of annoying people around her, and her overnight sleep test picked up the signs. She was prescribed iron supplements and pramipexole, an RLS drug. However, calming her legs did not improve her sleep situation.

  RLS is more common than IH and different in both its clinical and molecular features. Those differences can help us understand what Rye was used to thinking about, before IH took his career in a different direction. When it did, he would draw upon his RLS connections.

  RLS disrupts sleep, but it also occurs when someone is awake. It’s relatively straightforward to track one aspect of RLS, involuntary periodic limb movements, by attaching an accelerometer to someone’s ankle. RLS also has a sensory component: discomfort in the legs, which intensifies in the evening and is relieved by movement. People with RLS can describe their feelings as like worms burrowing, spiders crawling, or a tingling wave that never stops. Those sensations keep them pacing at night or drive them to kick their bedmates.

  Rye has estimated that around 10 percent of the United States population has RLS. For a fraction of that group, it bothers them enough to affect their quality of life.6 Variations in several genes contribute to the risk of developing RLS, as well as iron deficiency, explaining why menstruating individuals are more likely to have it. RLS was first described by the English anatomist Thomas Willis in the seventeenth century, although the Swedish neurologist Karl-Axel Ekbom is the one who defined it in the 1940s.

  Rye became involved with clinical studies of RLS in the late 1990s, when pharmaceutical companies began repurposing drugs that were FDA approved for Parkinson’s disease. He had a personal connection to RLS as well. In 2001, he discovered that he experienced periodic leg movements, after he spent time lying in bed with a broken ankle. While his leg was encased, he had the repeated urge to move his foot. He medicated the feeling away with one of the drugs he had been testing. He then fitted himself with an ankle bracelet and recorded his leg kicking thirty times per hour while he slept. He would point out the symptom in fellow travelers when he flew on airplanes with colleagues.7 He noticed that a few spoonfuls of ice cream tended to trigger acute attacks, for both him and his patients.8

  RLS was not well known until television advertising brought it to greater awareness, and skepticism lingered. Some news articles from this time ask: “Is RLS Real?” It was the topic of jokes on Seinfeld; Rush Limbaugh mocked RLS as fake on his radio show. As a response to trivialization by others, some in the RLS community have proposed renaming it Willis-Ekbom disease, although the term has not seen widespread adoption.9

  By disrupting sleep, RLS increases the risk of high blood pressure, and severe cases can become debilitating, interfering with daily life. However, critics have cited the disorder as an example of “disease mongering,” inflation through commercial promotion.10 Not everybody with wandering legs needs to take medication, they said, especially when the drugs can have side effects such as nausea and dizziness.

  FIGURE 2.2. David Rye in 2008.

  Source: Jessica McGowan / New York Times / Redux.

  FIGURE 2.3. Rye’s restless legs.

  Source: Jessica McGowan / New York Times / Redux.

  Rye’s status as the RLS guy meant he was often called upon to explain it. He provided cheerful counterpoint in a 2006 Washington Post article, which said that RLS was being pushed onto the public through TV advertising.11 “I don’t know of any evidence that it’s being over-diagnosed,” Rye told the Post. “I look at the positive side of it rather than the doomsday view. I think it helps to make a diagnosis.”

  In the summer of 2007, Rye was finishing work on a New England Journal of Medicine paper with geneticists in Iceland.12 His collaboration with DeCODE, a company that harvests Iceland’s well-recorded genealogy for biomedical research, had begun a decade earlier. He had developed the diagnostic tools used in the study and sent two members of his lab to Iceland to test ankle monitors. The NEJM paper identified the first genetic risk factor for periodic limb movements in sleep, present in more than 60 percent of the Icelandic and American populations. The paper attracted media coverage, and Rye’s comments suggested he viewed it as vindication for his specialty. He said: “We now have concrete evidence that RLS is an authentic disorder with recognizable features and underlying biological basis.”

  During this period, RLS wasn’t only a research interest for Rye; it had become part of his identity. Because RLS was so common, he hesitated to call it a disorder, describing it instead as a trait. In an interview with the New York Times, he riffed expansively: “This isn’t just restless leg, it’s a restless curiosity, it’s a restless mind, it’s a restless spirit.”13

  FROM DOPAMINE TO GABA

  In Anna’s case, RLS wasn’t a useful explanation. What struck Rye and Parker was how she described craving sleep constantly. It was clear that the medications Anna had been taking were not helping. Stimulants such as amphetamines are sometimes called “sympathomimetic,” in that they mimic the effects of adrenaline, making someone irritable and jumpy and increasing blood pressure and heart rate. High doses of amphetamines put stress on Anna’s body and led to periodic crashes. The rebound sleep Anna experienced is a well-known feature of amphetamines, although the duration was extreme. Parker wondered whether Anna’s diagnosis of idiopathic hypersomnia was appropriate because in the literature, stimulants were reported to be beneficial for the majority of cases.14

  In addition to stimulants and RLS drugs, Anna had been taking an antidepressant, as well as a beta-blocker to control high blood pressure brought on by amphetamines. This was a lot of different drugs. Some may have been making her sleep situation worse, which is why Parker wanted to wipe the slate clean and start over. “We had never had a patient quite like her before,” she said. “I thought to myself that she’s going to sleep her life away if we don’t do something. I felt like the answer was there, but we weren’t asking the right question.”

  Parker made a chart of the neurotransmitters affected by the drugs she and Rye had been prescribing to Anna. For stimulants such as amphetamines, dopamine and its chemical cousin norepinephrine are critical. Amphetamines cause dopamine and norepinephrine to be released from storage inside brain cells. Also, by inhibiting enzymes that would normally clear the neurotransmitters away, the drugs make dopamine and norepinephrine stick around longer at the junctions between brain cells, stimulating circuits that help keep us awake.

  Dopamine’s complex relationship with sleep and wake was one of Rye’s research specialties, since RLS is often treated with drugs that supplement or mimic dopamine. People with a casual interest in neuroscience have probably heard of dopamine because of its association in popular media with pleasure and addiction; as one writer put it, dopamine is “the molecule behind all our most sinful behaviors and secret cravings.”15 A software consulting firm (Dopamine Labs) was named after it, based on the business-friendly idea that people using mobile phones are searching for something that will trigger a squirt of it.

  However, neuroscientists say that dopamine doesn’t simply trigger a burst of pleasure or attention; it has several roles, depending on what part of the brain is involved.16 In one area called the nucleus accumbens, it promotes a feeling of reward. In Parkinson’s disease, cells that usually produce dopamine in the middle of the brain deteriorate and die. At that location, dopamine is needed for initiating and controlling movement. Dopamine-related signals can both wake someone up or make them sleepy, depending on the receptors engaged. Asking what dopamine or any other neurotransmitter does is like asking the meaning of a violin or flute in a piece of music. It depends what notes they play and in what context.

  MIDNIGHT INSPIRATION

  In the spring of 2007, Anna was exploring whatever was available, including changing her diet and alternative-medicine approaches such as acupuncture. For a while, Parker had tried giving Anna donepezil (Aricept), which boosts another neurotransmitter, acetylcholine. Anna reported no noticeable effect of donepezil, sometimes prescribed to patients with Alzheimer’s disease, on her mental fog or sleepiness.

  While Parker was searching for options for Anna, at some point she woke up in the middle of the night. She wrote the word GABA on a piece of paper and circled it. She said later: “We had tried to tinker with many of the neurotransmitters in her brain, but not GABA.” In relation to dopamine and norepinephrine, GABA (gamma-aminobutyric acid) belonged on the other side of Parker’s chart. GABA is the main inhibitory neurotransmitter in the adult brain. Its dominant effect is to make brain cells less likely to fire. Many drugs thought of as sleeping pills, such as barbiturates and benzodiazepines, enhance the action of GABA. Several injected and inhaled anesthetics function in a similar way. Alcohol also strengthens GABA signals, although it acts on other neurotransmitters too. All of these drugs make GABA signals stronger. Parker wanted to go in the opposite direction. “We had spent a lot of time trying to push Anna’s brain to wake up,” she said. “Maybe what we needed to do was make her less sleepy.”

  We can think of neurotransmitters such as dopamine like pepper. They are loaded into packets called vesicles and stored inside the cell until electrical signals release them. Outside the cell, the vesicles dump out their contents so that a neighboring neuron can take a whiff. The effects depend on a receptor on the neighboring cell’s surface, which the neurotransmitter fits into. The receptor’s altered shape triggers some action inside the cell: an ionic passageway opens or an enzymatic machine starts churning.

  GABA comes in vesicles too, but instead of acting like a spice, it’s more like an ice cube or sour cream. It soothes and calms. GABA has this effect by opening a set of gates spanning the cell membrane. The gates, named GABA-A receptors, come in barrel-shaped bundles of five subunits, which can open or close by changing their orientation. When the gates are open, chloride ions flow into the cell. The resulting accumulation of negative charge pushes the neuron away from sending an action potential, the electrical pulse that carries a signal to another neuron. (GABA also has a second set of GABA-B receptors, discussed in chapter 15.)

  Signals from GABA may have an overall “dimmer knob” effect on brain cells, as well as inhibiting regions involved in arousal, but that does not mean that flooding the brain with GABA, or otherwise mashing the brain’s GABA buttons, will produce healthy sleep. Even though several anesthetics enhance GABA’s actions, general anesthesia can’t replace sleep; the anesthetized brain does not pass through the complex oscillations thought to make sleep restorative. As with dopamine, no single neurotransmitter controls sleep and wake by itself.

  NANOSCALE CRAFT

  Given the inadequacy of what she and Rye had already tried, Parker saw counteracting GABA as an option worth investigating further. Parker had been discussing Anna’s case with her daughter Kathryn, who was working in the laboratory of the psychiatrist Kerry Ressler, who studied anxiety and post-traumatic stress disorder. The younger Parker relayed her mother’s interests, asking: “Who studies GABA at Emory?” Ressler suggested Andrew Jenkins, a young anesthesiology researcher who had been at Emory for just a few years. “I was willing to work with anyone. I was calling people across the country,” Parker said. “It was luck to be matched up with Andy.”

  Trained as a biophysicist, Jenkins’s expertise was not in clinical or even animal research, but he knew a lot about GABA receptors. His work, together with others’, helped break down an old idea concerning anesthetics. Despite their use in surgery since 1846, how anesthetics dissolve consciousness was unclear until the twenty-first century. The observation that anesthetics’ potency can be predicted by their affinity for olive oil, as opposed to water, led to the proposal that anesthetics function by seeping into the lipids in cell membranes. At Imperial College London, Jenkins worked with Nick Franks, whose lab was in the process of dispelling that greasy idea.17 Franks and his colleagues have shown that several anesthetics have interactions with specific sites on GABA receptors. Not all anesthetics act this way, but commonly used ones such as propofol and sevoflurane do.

  In 2006, Jenkins gave a cheeky interview to a graduate student newsletter.18 He described his time at Imperial, where he completed university and graduate school, as the United Kingdom’s equivalent of the Massachusetts Institute of Technology. His experience there was dominated by “beer, rugby, and an insane lecture and lab schedule.” Franks ran his lab in an understated way, sparing with praise. At some point, Jenkins was told: “Andy, you’re not the smartest guy in the lab, so you’ll have to work harder than the other guys to make up for it.”

  Jenkins encountered several dead ends during his time in graduate school. One of his projects involved experiments on pond snails, another the breeding of alcohol-insensitive guppies. Eventually, he did make headway on anesthetics, but he had to invest a lot of hours. He was the first person in his lab to learn a painstaking technique called “patch clamping.” “It requires a tremendous amount of patience,” said Adam Hall, a scientist from a neighboring lab who taught Jenkins the technique. “Andy really took to it, with a physicist’s precision.”

  Developing the patch clamp earned two German scientists the 1991 Nobel Prize in Physiology or Medicine, and the technique supports the foundations of modern neuroscience. Patch clamping involves grasping a cell with the tip of a thin glass tube, which holds a patch (a small section) of the cell membrane. The patch contains a handful of molecules—GABA receptors or others—that let charged ions through their gates. Inside the tube, an electrode allows the experimenter to monitor the current and voltage.19