2Sherrington’s Enchanted Loom and Huxley’s Science Fiction
When Hodgkin and I finished writing the 1952 papers, each of us moved to other lines of work. … Any idea of analyzing the channels by molecular genetics would have seemed to us to be … science fiction.
—Andrew Huxley, in The Axon, 1995
The human nervous system—our brain, spinal cord, and nerves—is the world’s most complex computer. There are more than 100 billion nerve cells in the human brain and spinal cord, greater than the number of stars within the Milky Way.
These nerve cells, called neurons by scientists, act as tiny transistors, or in some cases as integrated circuits. They send electrical impulses to and fro along nerve fibers, termed “axons” by neuroscientists, as the nervous system makes countless computations each second. In 1942, the pioneering British neuroscientist Charles S. Sherrington referred in his book Man on His Nature to the active brain as “an enchanted loom where millions of flashing shuttles weave a dissolving pattern, always a meaningful pattern … a shifting harmony of subpatterns” (Sherrington 1942).
As a student, I was fascinated by the brain, and as I contemplated a career in biomedical research, I wanted to understand the ways in which the activities of the billions of nerve cells within the brain lead to human thought—consciousness, reasoning, planning, understanding, and emotion. I was fascinated by the work of MIT scientists Warren McCulloch and Walter Pitts, who, in a seminal article, “A Logical Calculus of the Ideas Immanent in Nervous Activity” (McCulloch and Pitts 1943), observed in 1943 that, at any given moment, each neuron is either firing or not, and suggested that neurons could serve as “threshold logic units,” a conclusion that led them to postulate that it might be possible to mimic brain activity by building a large electrical device consisting of a multitude of on-or-off switches. This suggestion, well before the word “neuroscience” had been coined, provided a basis for neural network theory and, in the opinion of some, contributed to the thinking that led to the development of modern computers.
My first forays into research as a student at Harvard focused on attempts to explain, at the level of single nerve cells, how the human brain categorizes complex external stimuli. I wanted, as my overall goal, to solve the “mind–brain” problem. But, while my interest in higher nervous function and in questions about the brain and behavior allowed me to publish my first papers, I came to the conclusion that full answers to these philosophically grand questions would not be forthcoming during my professional lifetime.
My interest in the nervous system deepened over the next few years, but refocused on the simpler and more tractable problem of how single neurons, or well-defined circuits of neurons, function in health and disease. Questions about the pathophysiology of neurological disease were to drive me for the rest of my career. What fundamental changes in the nervous system cause neurological disease? How do these changes cause the neurological signs and symptoms that bring patients to the clinic? And, as clinicians, what can we do about it? Might it be possible—using fundamental information about the cells and molecules responsible for diseases of the brain, spinal cord, and peripheral nerves—to develop new and more effective treatments for disorders of the nervous system? Three themes echoed as I thought about these questions: Axons, sodium channels, and pain. These themes converged in the search for a pain gene.
Axons
My early mentors, J. D. Robertson (professor of cell biology at Harvard and the discoverer of the molecular structure of the myelin insulation that surrounds nerve fibers) and J. Z. Young (professor at University College London and the discoverer of the squid giant axon, a model that subsequently yielded crucial lessons about sodium channels), provided encouragement that fueled my earliest research on axons. Discussions with J. D. Robertson while waterskiing at his lakeside retreat in Wayland, Massachusetts, and interchanges with J. Z. Young at teatime at University College London convinced me that axons were not just passive wires—they were elegantly architected biological machines that function with millisecond accuracy. I was fascinated by the ingenious design principles of nerve fibers, which optimized their performance and matched their architecture to the functional needs of each specialized part of the nervous system. Some axons were built to conduct impulses as quickly as possible (Waxman and Bennett 1972). Other axons acted as precisely timed “delay lines,” getting the message to their recipient neurons, one synapse down the line, not as quickly as possible but in just the right amount of time (Waxman 1970), and still other axons had evolved so as to process information in highly complex ways, in some cases generating external electrical fields that could, for example, be used by fish, like sonar, to navigate (Waxman, Pappas, and Bennett 1972). I did not imagine, as I did these early studies, that they would propel me toward research on the human malady of chronic pain.
A Morse Code in the Brain
Communication between neurons can be thought of as using a form of Morse code, which, in the era of the telegraph, enabled people to communicate by sending a series of dots and dashes. In a broad-brush sense, Morse code applies to pain signaling as well as other aspects of coding within the nervous system. Neurons within the brain and spinal cord use nerve impulses, called action potentials by neuroscientists, to communicate. The action potential is about 100 millivolts or one-tenth of a volt in size, and it lasts only about 1 millisecond (one one-thousandth of a second); a representative action potential from a pain-signaling spinal neuron is shown in figure 2.1. For a given nerve cell the action potentials are always the same, just like the dots in Morse code. It is the rate and pattern of the action potentials, sent on to downstream neurons, by which that nerve cell delivers its message.
Figure 2.1 Nerve impulse (action potential) from a dorsal root ganglion (DRG) neuron, shown in green. Until it is stimulated, the neuron is quiescent and sits at resting membrane potential (RMP) with the inside of the cell negative by about –60 millivolts with respect to the outside. When the cell is depolarized by a sufficient amount, it reaches threshold, and, at that point, there is an explosive, nearly simultaneous activation of many sodium channels, producing a pulse-like depolarization of the cell membrane which actually crosses 0 millivolts, so that the inside of the cell is briefly positive before the cell repolarizes and returns to resting potential. The action potential, which always has the same configuration and time course in any given cell, lasts about 1 millisecond. NaV1.7 sodium channels play a particularly important role in DRG neurons. They act within the subthreshold domain, below threshold, to amplify small depolarizing stimuli (blue). Acting in this way, NaV1.7 channels determine the sensitivity, or “set the gain,” on DRG neurons. Modified from Rush et al. (2007).
Some neurons are excitatory and stimulate downstream neurons that receive their message. Other neurons are inhibitory and have a calming effect on downstream neurons. McCulloch and Pitts argued that each individual neuron integrates its multiple excitatory and inhibitory inputs into a message which it conveys to other neurons via a series of nerve impulses or action potentials. How might a series of action potentials—the dots of Morse code but not the dashes—carry meaning? Codes based on the frequency of action potentials produced by a given neuron, the pattern of action potentials over time, or the particular neuron involved (the “labeled line” theory) have all been proposed, and each applies at some sites in the nervous system. Irrespective of which coding mechanism is involved, it would be expected that underactivity of neurons due to disease would interfere with their computational function, and that, conversely, overactivity of neurons (termed hyperexcitability) would also perturb their output. Extreme hyperactivity of neurons in some parts of the brain can cause epileptic seizures which can be likened to tornados of nervous system activity.
Sodium Channels: Molecular Batteries in Our Nerve Cells
As a beginning undergraduate I became aware of the seminal work of the British scientists Alan Hodgkin and Andrew Huxley. At a remarkably young age these intellectual giants, who had been research fellows at Trinity College, Cambridge, had discovered the crucial role of sodium channels in nerve impulse conduction (Huxley 1995). Their experiments capitalized on the large size—a diameter of about a millimeter—of specialized nerve fibers called “giant” axons of the squid, which allowed Hodgkin and Huxley to insert electrodes within them. This enabled them to measure the actual electrical currents underlying nerve impulses, a feat that had not been previously possible. An incisive analysis of the generation of action potentials in the squid’s giant axon allowed them to formulate a set of equations—still widely referred to as the Hodgkin–Huxley equations—that explained how sodium channels within nerve cells open and close to generate nerve impulses. Working before the advent of the finely honed microelectrode that neuroscientists now use, without modern computers, and prior to the maturation of molecular biology, these pioneers presciently demonstrated the presence of sodium channels within nerve cell membranes. Acting as tiny batteries, the sodium channels open and close rapidly in response to depolarizations of the nerve cell membrane to allow small flows of sodium ions that produce the electrical current underlying nerve impulses. It was not until the 1980s that, as a result of molecular cloning, it became possible to understand the conformation of sodium channels and the ways in which they open and close within milliseconds to produce nerve impulses—an attribute that led to their being described as “some of the most conformationally versatile structures in nature” (Pascual 2016). Although Hodgkin and Huxley could not see sodium channels and had no idea of their molecular structure, they accurately predicted many of their properties (Hodgkin and Huxley 1952). Their work was honored with the 1963 Nobel Prize and still is used for understanding ion channels.
It was also during my time as an undergraduate that I was fortunate to meet Patrick Wall, considered by many as the father of modern pain research. Known for his wit as well as his incisive thinking, he worked in the Department of Biology at MIT, a mile down Commonwealth Avenue from Harvard College. As a junior, and then as a senior at Harvard, I visited Pat Wall, watched his experiments, and offered my guesses as to what to do next. In the late 1960s he moved to University College London to head up a new research center and invited me to move with him as a PhD student. London was too far from home for me, and I did not take him up on the invitation. However, several years later, as an MD–PhD student at Albert Einstein College of Medicine, I applied for a fellowship from The Epilepsy Foundation, and I worked for four months with Wall in his University College laboratory on Gower Street. At that time an investigator could achieve preeminence with a small laboratory, and I had the privilege of working one-on-one together with Wall, who, between cigarettes he rolled himself, coached me on the minutiae of electrophysiological recording. The resulting paper (Wall, Waxman, and Basbaum 1974) described the barrage of impulses generated by axons within peripheral nerves in the first few minutes immediately following traumatic injury. I didn’t realize it at the time, but my experience working with one of the giants of pain research set the stage for the second half of my career. Twenty years later I returned to pain research, combining my interests in sodium channels and neuropathic pain.
The themes of axons, sodium channels, and neurological disease began to come together for me in 1975. A decade after Hodgkin and Huxley’s Nobel Prize, as a new assistant professor at Harvard and MIT, I turned my attention to nerve fibers, how they work, and why they don’t work properly in some disease states. Much of the previous work on which I based my studies had been carried out in lower species such as the squid or other invertebrates, where nerve fibers are larger and easier to study. I reasoned that if the axon of a squid could be interesting, the axon of a human being—especially a human being with diseased nerve fibers—could be an even more interesting topic for study.
My first research as a faculty member at Harvard and MIT focused on how the molecular architecture of axons determines their functional properties. In one of my projects I was exploring the ways in which sodium channels contribute to the pathophysiology of diseases such as multiple sclerosis. I was interested in multiple sclerosis for two reasons: First, it is the most frequent neurologic crippler of young adults in industrialized societies, usually rearing its head and producing symptoms in the third decade of life, just as people are establishing their adult trajectories. Second, multiple sclerosis was, for me, a “model disease,” a disease that might hold general lessons about how the nervous system adapts to injury.
My classmates and I had been taught in medical school that following any type of injury to the brain or spinal cord, there was little if any functional recovery. We watched well-meaning professors declare that following injury to the nervous system as occurs in spinal cord injury or stroke, the outlook was hopeless. We saw our mentors make the diagnosis of these disorders, turn around, and walk away from the bedside because they had no effective treatments to offer, and we were disappointed. But multiple sclerosis did not follow this rule. People with multiple sclerosis often experience remissions in which they spontaneously regain previously lost functions. A person with multiple sclerosis loses nearly all vision in one eye and, four weeks later, is able to read the newspaper. Another person with multiple sclerosis develops paralysis of the legs and then, without any treatment, recovers the ability to walk. This exception to the rule suggested to me that multiple sclerosis might hold more general lessons about recovery of function after injury to the nervous system.
It was well established at this time that the generation and transmission of electrical impulses along axons required the activity of sodium channels. We knew from earlier work that, in model systems such as the squid giant axon, sodium channels were sprinkled in a low but relatively uniform density along the entire length of the fiber. But what about axons in higher species such as humans? Many of those axons are surrounded by myelin, a lipid-containing material that acts as an insulator, like the covering of an electrical wire. The myelin sheath is periodically punctuated by small areas devoid of myelin, called nodes of Ranvier, and physiologists had known for some time that, in myelinated fibers, the impulse did not move continuously along the axon as in the squid giant fiber, but jumped in a discontinuous or “saltatory” manner from node of Ranvier to node of Ranvier, and so on, progressing node by node along the length of the nerve fiber.
My first major observation relevant to multiple sclerosis showed that, in myelinated fibers, the sodium channels are not distributed uniformly along the length of the axons, but are rather focused in a highly nonuniform way at the nodes of Ranvier where they are highly concentrated. My studies also showed that under the myelin, where they are not needed, there were very few sodium channels (Waxman 1977, 1982). As I made these observations at Harvard and MIT, J. Murdoch Ritchie, a pharmacologist working at Yale who was to become a friend and colleague, came to a similar conclusion. It was hard not to be enamored with the elegance of the axon even as reflected by the placement of the sodium channels. Every molecule in its place, precisely where needed for optimal function, a beautiful example of purpose-driven biological architecture.
But what happens when the myelin insulation is injured? Demyelination had classically been known to be a hallmark of multiple sclerosis. Traditional dogma posited that damage to the myelin caused neurological deficits such as blindness, weakness, or incoordination because the conduction of action potentials fails along axons within the brain and spinal cord as a result of current leakage through the damaged myelin insulation: a “short circuit.” This, for me, posed an enigma: Following loss of the myelin within the brain and spinal cord in multiple sclerosis, there is little remyelination. The damage to the myelin insulation was permanent. Yet remissions were common in multiple sclerosis, suggesting that some demyelinated axons had recovered the capability to conduct action potentials. How were my patients recovering the ability to see? Or to walk? For me, this raised the following more general question: How do remissions occur?
One of the challenges faced by a nerve impulse trying to invade a demyelinated part of an axon arises from the increased surface area of the denuded axon membrane. This type of problem has been termed “impedance mismatch” by electrical engineers. Working at MIT and aided immeasurably by new methods for computer simulation, developed by biophysicist John Moore and based, in part, on the Hodgkin–Huxley equations, in 1978 we showed that changes in the geometry of demyelinated nerve fibers, together with the production of new sodium channels, could provide a basis for resumption of impulse conduction along demyelinated axons (Waxman and Brill 1978). In 1980, in studies on the nerves of rats in which we induced demyelination, postdoctoral fellow Robert Foster and I showed that there is remarkable molecular plasticity in some demyelinated axons, which synthesize new sodium channels and plug them into the demyelinated, previously sodium-channel-poor membrane. The newly deployed sodium channels function like the sodium channels sprinkled along the length of the giant axon of the squid, to support restoration of impulse conduction (Foster, Whalen, and Waxman 1980). Here we had an explanation for recovery of impulse conduction along demyelinated axons. In 2004, together with colleagues Matthew Craner and Joel Black, we showed the same molecular plasticity along demyelinated axons within the human nervous system, within the brains of people with multiple sclerosis; in that study we refined our analysis so that we could precisely identify the types of sodium channels involved (Craner et al. 2004). In 2006, I was invited to give the J. Z. Young Memorial Lecture at University College London. I entitled it “From Squid to Clinic: Sodium Channels in Neurological Disease,” and in it, I described this research. It is not the topic of this book, but the goal of being able to induce remissions in multiple sclerosis continues to drive research in some of my laboratories.
Dorsal Root Ganglion (DRG) Neurons as Generators of Pain
In the mid-1990s I turned my attention to sodium channels and neuropathic pain—chronic pain rising as a result of damage to, or dysfunction of, the nervous system. My first experiments on pain-signaling neurons were an outgrowth of my earlier discovery that neurons produce new sodium channels after injury to the myelin surrounding their axons. In these new experiments I wanted to answer the question, “Might some sodium channel genes in a neuron turn on, while others turn off, following injury to its axon?” (Waxman, Kocsis, and Black 1994).
Figure 2.2 Dorsal root ganglion (DRG) neurons, with cell bodies within the dorsal root ganglia, extend an axon from the body surface and organs, all the way into the spinal cord. Sodium channels within the cell membrane of DRG neurons enable them to produce action potentials (APs). Pain-signaling DRG neurons are excited by dangerous levels of pressure, heat, cold, acidity (pH), or irritating chemicals and, in response, send action potentials to the spinal cord, which relays them to the brain. Multiple types of sodium channels, shown in orange, red, and green, participate in this signaling. NaV1.7 channels (green) play a particularly crucial role, amplifying small stimuli in the periphery and thereby setting the gain on DRG neurons, and facilitating impulse transmission close to the spinal cord. Modified from Waxman and Zamponi (2014).
Serving as sentinels or an early-warning system, pain-signaling dorsal root ganglion (DRG) neurons innervate our body surface, teeth, cornea, gut, bladder, and many of our organs. Trigeminal ganglion neurons serve the same function for the face. The cell bodies of DRG neurons are located within clusters called dorsal root ganglia located just outside the spinal cord; since DRG neurons are not located within the central nervous system (the brain or spinal cord), these cells are sometimes referred to as “peripheral” neurons. From the cell body of each DRG neuron, a peripheral nerve fiber or axon extends to the body surface, and a central axon extends into the spinal cord. Altogether, the DRG neuron provides a pathway for signaling via nerve impulses that originate in the periphery and propagate into the spinal cord (figure 2.2). Pain-signaling DRG neurons, sometimes called nociceptors, are sensitive to the presence of threatening mechanical stimuli such as a pinprick or a blow from a hammer, injurious thermal stimuli such as damaging levels of heat or dangerous levels of cold, and noxious chemical irritants such as acids. These “first-order” pain-sensing neurons signal the presence of threats to the body by generating nerve impulses that they send, via our peripheral nerves, to the spinal cord. Within the spinal cord, these nerve impulses excite “second-order” pain-signaling neurons, which relay the signal upward toward the brain. When the message reaches the brain, it is processed by still other circuits of neurons which elicit the experience of pain. The process of pain signaling begins in the periphery. The sites of origin of nerve impulses that encode pain—DRG neurons and trigeminal neurons—are major players in pain.
Pain can serve a protective purpose when it elicits an adaptive response such as pulling a hand away from a hot stove. It can have an instructive role during development, teaching a person, early in life, what is safe and what is not. Or pain can be inflammatory, signaling the presence of tissue damage. Inflammatory pain can also be protective, warning, for example, against overuse of an injured and healing joint. Alternatively, pain can be neuropathic. Neuropathic pain reflects dysfunction of the nervous system and can occur when DRG neurons take on a life of their own and generate pain signals even in the absence of a noxious stimulus or inflammation.
Like the unbridled flashing of shuttles within Sherrington’s loom, neuropathic pain is the result of inappropriate firing—in the absence of a noxious stimulus or out of proportion to a noxious stimulus—by an injured or diseased nerve cell along the pain-signaling pathway. An example of this abnormal firing is shown in figure 2.3, from a paper that Jeffery Kocsis and I published in Nature in 1983. Here, using a tiny microelectrode carefully placed within a single axon inside the nerve of a rat that had been subjected to a nerve injury, we can see the abnormal generation of multiple repetitive nerve impulses in an injured nerve fiber—machine gun–like, staccato—in response to a small stimulus that should elicit only a single nerve impulse (Kocsis and Waxman 1983). Recordings such as this from an axon less than 10 μm (1/100th of a millimeter, much smaller than a wisp of thin hair) in diameter were not easy to achieve and are testimony to Kocsis’s prowess with the microelectrode. Four years later Kocsis and I had the opportunity to record from the axons in nerves from human subjects with painful neuropathy that had been removed for diagnostic purposes (figure 2.4), and again we observed abnormal repetitive impulse activity (Kocsis and Waxman 1987). In both of the experiments the abnormal repetitive nerve impulses were generated by a sustained depolarization of the axon membrane. Here we could see neuropathic pain in the making. Something was producing abnormal depolarizations in DRG neurons and their axons after nerves were injured. These recordings suggested to us that, if we could identify the molecules responsible for generating this depolarization, we might be able to pinpoint the drivers of neuropathic pain. Although it was not yet known that there were nine different types of sodium channels, and peripheral sodium channels had not yet been discovered, recordings of this type led us to think that sodium channels might act as generators of pain.
Figure 2.3 Inappropriate repetitive firing of action potentials, recorded with a microelectrode from a single axon within the sciatic nerve of a rat that had received a nerve injury one year previously. The aberrant repetitive action potentials sit upon an abnormal depolarization of the axon membrane which suggests abnormal sodium channel activity. From Kocsis and Waxman (1983).
Figure 2.4 Microelectrode recording from a single axon within the sural nerve of a patient with a painful peripheral neuropathy. The nerve was biopsied for diagnostic evaluation. Aberrant repetitive action potentials can be seen, arising from an abnormal depolarization suggesting abnormal sodium channel activity within the axon membrane. From Kocsis and Waxman (1987).
Peripheral Sodium Channels—A Holy Grail
Anyone who has gone to a dentist knows that nerve impulses within pain-signaling nerves can be silenced with certain medications—a nerve can be put to sleep, and pain within its territory will not be felt while it is anesthetized. In the case of dental anesthesia the nerve is infiltrated by injecting it with the local anesthetic Novocaine or a related drug that blocks sodium channels, thereby preventing the generation and transmission of electrical impulses in nerve fibers that innervate the teeth and oral cavity.
Given the remarkable efficacy of sodium channel blockers when they are injected to locally prevent pain during dental procedures, it might have been hoped that sodium channel blockers could be used more broadly, as medications taken by mouth, to alleviate chronic pain. Indeed, a number of sodium channel blocking drugs exist and some can be taken orally. However, these drugs have limited effectiveness for the treatment of pain because they block sodium channels throughout the nervous system. The unwanted block of sodium channels outside of pain-signaling neurons, particularly in neurons throughout the brain, produces dose-limiting side effects that include confusion, loss of balance, double vision, and sleepiness. Thus a major question in pain research focused on whether it might be possible to develop highly specific medications that selectively block sodium channels in pain-signaling peripheral neurons so as to put these cells to sleep while having no effect on the sodium channels within other types of neurons. This targeted approach would avoid unwanted side effects.
A sodium channel is a protein molecule, consisting of a string of around 1,800 amino acids, strung together like the beads in a necklace which then folds into a barrel-like structure. Beginning in the mid-1980s, it was becoming clear that the sodium channel was not a singular entity. Studies in laboratories around the world were beginning to show that there was not a single, unitary type of sodium channel. We knew, by the early 1990s, that multiple genes encoded multiple sodium channels, all sharing a similar overall molecular structure, but with slightly different amino acid sequences, and different physiological and pharmacological properties. The question of whether there might be sodium channels that play preferentially important roles in peripheral nerve cells, particularly pain-signaling DRG neurons and their axons, emerged as a major challenge in pain research. The logic was that, if these “peripheral” sodium channels existed, it might be possible to develop appropriately focused medications that would mute the activity of peripheral pain-signaling DRG neurons without having a significant effect on neurons within the brain. If this could be achieved, it would permit pain relief without side effects such as double vision, confusion, or sleepiness, and with little potential for abuse or addiction. But first, it had to be shown that peripheral sodium channels existed. Peripheral sodium channels became a “holy grail” of pain research.
Between 1996 and 1999 three different subtypes of peripheral sodium channels meeting this specification were identified by gene cloning in the DRG neurons of rodents, rats, and mice (table 2.1).
Boosts small stimuli to initiate firing of pain-signaling peripheral neurons; facilitates neurotransmitter release at first synapse within the spinal cord; sets gain in pain-signaling DRG neurons to control their firing
NaV1.8 (gene SCN10A)
Produces the electrical current needed for high-frequency firing of action potentials in pain-signaling DRG neurons
NaV1.9 (gene SCN11A)
Depolarizes resting potential of pain-signaling neurons; amplifies response to small stimuli
Note. DRG, dorsal root ganglion.
They are called NaV1.7, NaV1.8, and NaV1.9 (Catterall, Goldin, and Waxman 2005). The NaV1.8 sodium channel, initially called SNS (Sensory Neuron Specific), was discovered and characterized in 1996 by John Wood and his colleagues at University College London (Akopian, Sivilotti, and Wood 1996). The NaV1.9 sodium, initially called NaN (Na-Nociceptive), was cloned and characterized in my laboratory in 1998 by Sulayman Dib-Hajj (Dib-Hajj et al. 1998). NaV1.9 was subsequently described by Simon Tate and his research group at Glaxo (Tate et al. 1998), who called it SNS2. Gail Mandel and her colleagues at Stony Brook University reported that a third sodium channel, initially called PN1 and hNE—now called NaV1.7—was not detectable in the brain but was present at high levels in peripheral nerve cells (Toledo-Aral et al. 1997). Our experiments also showed that this channel, not detectable within the brain, was present within DRG neurons (Felts et al. 1997). As with NaV1.8 and NaV1.9, it was not possible to rule out the presence of some NaV1.7 channels within the brain—expression at very low levels throughout the brain, or within a small subgroup of neurons within the brain might escape detection. But the high level of expression within DRG neurons, in the context of a low level of expression, if any, in the brain, suggested a much more important role of NaV1.7 within peripheral pain-signaling neurons. Even before the search for a pain gene, NaV1.7 became a major focus for me and my colleagues.
Figure 2.5 Atomic-level model of the NaV1.7 sodium channel. The green, salmon, purple, and blue spirals show the course of the channel protein as it weaves in and out of the cell membrane within four different parts (domains) of the NaV1.7 channel. Single amino acids can be seen in red and gold. The top diagram shows a side view of the channel, as seen by an observer within the membrane. The bottom diagram shows the channel as seen from within the cell, looking out. Just above the yellow amino acid, the pore in the center of the channel can be seen. From Yang et al. (2012).
Sodium channels are beautiful and complex molecules. Figure 2.5 displays the three-dimensional configuration of the folded human NaV1.7 polypeptide as determined by computer modeling at a resolution of 2.7Å (one Ångstrom, Å, equals 1.0 × 10–10 or one ten-billionth of a meter—about one millionth of a diameter of a human hair). This powerful methodology allows us to infer the locations of some of the crucial atoms within the channel and permits us to make potentially important predictions about the actions of specific drugs on the channel.
The pivotal functional role of NaV1.7 in controlling the firing of peripheral pain-signaling neurons began to emerge in 1997 when we examined the electrophysiological properties of this channel, which at that time was called PN1 or hNE. Our work was carried out at the Center for Neuroscience and Regeneration Research, a Yale University research center housed in a purpose-designed building erected with funds provided by the Paralyzed Veterans of America at the Veterans Affairs Medical Center in West Haven. The goal of the Center was to capitalize on the “molecular revolution” to bring a better understanding of, and ultimately new and more effective treatments for, the pain and paralysis that result from injury or disease of the nervous system.
Inscribed at the Medical Center’s entrance were the words “Here you can see the price of freedom.” There was no shortage of poignant illustrations of these words. Among them were men and women seeking relief from chronic pain that was an accompaniment of nerve injury, burn injury, or traumatic limb amputation, an injury which severs not only arms and legs but also the nerve fibers within them. Here one could see, loud and clear, the price of freedom. And that reminded me, daily, of the importance of unraveling pain’s mysteries.
Knowing that NaV1.7 channels were highly expressed in pain-signaling peripheral neurons, physiologist Ted Cummins and I used patch-clamp electrodes to study them. We found that NaV1.7 channels respond to, and amplify, stimuli that are too small to activate other sodium channels (Cummins, Howe, and Waxman 1998). In response to stimulation, NaV1.7 channels bring the neuron closer to the potential needed to turn on other types of sodium channels such as NaV1.8, which then produce most of the electrical current underlying the nerve impulses used by pain-signaling DRG nerve cells to signal the presence of painful stimuli (Renganathan, Cummins, and Waxman 2001; Rush, Cummins, and Waxman 2007). We learned, in our early studies between 1997 and 2001, that NaV1.7 plays a powerful role in setting the gain on peripheral nerve cells isolated from laboratory animals such as rats. We did not yet know, when we did these initial studies, that they would set the stage for subsequent demonstration, in humans, that NaV1.7 is a gatekeeper for pain.
References
Akopian AN, Sivilotti L, Wood JN. 1996. A tetrodotoxin-resistant voltage-gated sodium channel expressed by sensory neurons. Nature 379(6562): 257–262.
Catterall WA, Goldin AL, Waxman SG. 2005. International Union of Pharmacology. XLVII. Nomenclature and structure-function relationships of voltage-gated sodium channels. Pharmacol Rev 57(4): 397–409.
Craner MJ, Newcombe J, Black JA, Hartle C, Cuzner ML, Waxman SG. 2004. Molecular changes in neurons in multiple sclerosis: Altered axonal expression of Nav1.2 and Nav1.6 sodium channels and Na+/Ca2+ exchanger. Proc Natl Acad Sci USA 101(21): 8168–8173.
Cummins TR, Howe JR, Waxman SG. 1998. Slow closed-state inactivation: A novel mechanism underlying ramp currents in cells expressing the hNE/PN1 sodium channel. J Neurosci 18(23): 9607–9619.
Dib-Hajj SD, Tyrrell L, Black JA, Waxman SG. 1998. NaN, a novel voltage-gated Na channel, is expressed preferentially in peripheral sensory neurons and down-regulated after axotomy. Proc Natl Acad Sci USA 95(15): 8963–8968.
Felts PA, Yokoyama S, Dib-Hajj S, Black JA, Waxman SG. 1997. Sodium channel alpha-subunit mRNAs I, II, III, NaG, Na6 and hNE (PN1): Different expression patterns in developing rat nervous system. Brain Res Mol Brain Res 45(1): 71–82.
Foster RE, Whalen CC, Waxman SG. 1980. Reorganization of the axon membrane in demyelinated peripheral nerve fibers: Morphological evidence. Science 210(4470): 661–663.
Hodgkin AL, Huxley AF. 1952. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol 117(4): 500–544.
Huxley A. (1995). Electrical activity in nerve: The background up to 1952. In S. G. Waxman, J. D. Kocsis, & P. K. Stys (Eds.), The axon: Structure, function, and pathophysiology. New York: Oxford University Press.
Kocsis JD, Waxman SG. 1987. Ionic channel organization of normal and regenerating mammalian axons. Prog Brain Res 71: 89–101.
McCulloch W, Pitts W. 1943. A logical calculus of the ideas immanent in nervous activity. Bull Math Biol 7: 115–133.
Pascual JM. 2016. Understanding atomic interactions to achieve well-being. JAMA Neurol 73(6): 626–627.
Renganathan M, Cummins TR, Waxman SG. 2001. Contribution of Na(v)1.8 sodium channels to action potential electrogenesis in DRG neurons. J Neurophysiol 86(2): 629–640.
Rush AM, Cummins TR, Waxman SG. 2007. Multiple sodium channels and their roles in electrogenesis within dorsal root ganglion neurons. J Physiol 579(Pt 1): 1–14.
Sherrington CS. (1942). Man on his nature. Cambridge: Cambridge University Press.
Tate S, Benn S, Hick C, Trezise D, John V, Mannion RJ, et al. 1998. Two sodium channels contribute to the TTX-R sodium current in primary sensory neurons. Nat Neurosci 1(8): 653–655, doi:10.1038/3652.
Toledo-Aral JJ, Moss BL, He ZJ, Koszowski AG, Whisenand T, Levinson SR, et al. 1997. Identification of PN1, a predominant voltage-dependent sodium channel expressed principally in peripheral neurons. Proc Natl Acad Sci USA 94(4): 1527–1532.
Waxman SG. 1970. Closely spaced nodes of Ranvier in the teleost brain. Nature 227(5255): 283–284.
Waxman SG. 1977. Conduction in myelinated, unmyelinated, and demyelinated fibers. Arch Neurol 34(10): 585–589.
Waxman SG. 1982. Membranes, myelin, and the pathophysiology of multiple sclerosis. N Engl J Med 306(25): 1529–1533.
Waxman SG, Bennett MV. 1972. Relative conduction velocities of small myelinated and non-myelinated fibres in the central nervous system. Nat New Biol 238(85): 217–219.
Waxman SG, Brill MH. 1978. Conduction through demyelinated plaques in multiple sclerosis: Computer simulations of facilitation by short internodes. J Neurol Neurosurg Psychiatry 41(5): 408–416.
Waxman SG, Kocsis JD, Black JA. 1994. Type III sodium channel mRNA is expressed in embryonic but not adult spinal sensory neurons, and is reexpressed following axotomy. J Neurophysiol 72(1): 466–470.
Waxman SG, Pappas GD, Bennett MV. 1972. Morphological correlates of functional differentiation of nodes of Ranvier along single fibers in the neurogenic electric organ of the knife fish Sternarchus. J Cell Biol 53(1): 210–224.
Waxman SG, Zamponi GW. 2014. Regulating excitability of peripheral afferents: Emerging ion channel targets. Nat Neurosci 17(2): 153–163.
Yang Y, Dib-Hajj SD, Zhang J, Zhang Y, Tyrrell L, Estacion M, et al. 2012. Structural modelling and mutant cycle analysis predict pharmacoresponsiveness of a Na(v)1.7 mutant channel. Nat Commun 3: 1186.
