A New Chapter for Sodium Channels

The PRF caught up with Stephen Waxman, Yale University School of Medicine, New Haven, and the Veterans Affairs Medical Center, West Haven, Connecticut, for a Q&A about his work on voltage-gated sodium channels, mutations therein, and the latest findings in small fiber neuropathy (see PRF related news story Na_v1.7 Mutations Move Into the Mainstream). Waxman filled us in on some history and discussed what’s exciting about his recent, and ongoing, work.

Questions by Megan Talkington

PRF: How did you get started on sodium channels?

I’ve been interested in pain, particularly pain after injury to the nervous system, for a number of years. When I was in school, we talked about the sodium channel, as if it were a unitary, singular entity. Now we know that there are nine different genes, encoding nine distinct sodium channels. They share the same overall structure, but they have slightly different amino acid sequences, they have different physiological and pharmacological properties, and they are distributed in different ways in the nervous system. Some of them are preferentially present in particular types of neurons, and contribute to endowing different types of neurons with different patterns of activity.

One of these channels is called Na_v1.7. It’s preferentially present in two types of neurons: sympathetic ganglia neurons and pain-signaling DRG neurons.

We had done the work on profiling the wild-type [Na_v]1.7 channel, and we published that back in 1998. And then, in 2002 or 2003, after having done more work in vitro and in animal models on sodium channels and pain than anybody would want to hear about, I said to my team: "Let’s launch a worldwide search for humans with hereditary pain syndromes, and let’s see what we find." We knew that these patients, if they existed, would be very rare. I’m a neurologist; any neurologist will tell you: You can go through an entire career, and you never see families with inherited neuropathic pain. But despite that, we launched the search. And we didn’t find them.

And then, in March of 2004, we opened up one of the genetics journals and there was a paper out of a Beijing lab by Yong Yang. He’s a very smart academic dermatologist, and he and a bunch of geneticists described two families with inherited erythromelalgia! I hadn’t heard of it—no one had heard of it at that point. These patients have searing, burning, scalding pain in response to mild warmth, and it’s autosomal dominant with essentially 100 percent penetrance. Yang had studied the two families and found point mutations in 1.7. I said to my team: "We were looking, and we didn’t find them. This is the worst day of my life."

Then I read the paper and realized that they identified the mutation, but once you’ve identified a channel mutation, that’s just the beginning of the story. The industry standard is that you have to make the mutation, and express it and study its behavior. And that takes a lot of work, but we had done all the work on the wild-type channel. We literally had the DNA sitting in my freezer. So what was going to take anybody else a couple of years to do, we knew we could do in a couple of months. And I said: "No, it’s not the worst day of our life; this is the best day of our life!"

So, long story short, we profiled the mutations, and published them, and then we were contacted by the Erythromelalgia Association and they gave us their largest grant ever, but they gave us something much more important. They have become wonderful partners, giving us their families’ DNA and sharing their stories with us. The syndrome is remarkable. I’m aware of several patients who’ve asked to have their limbs amputated [because the pain is so bad], and essentially nothing works of the existing therapies. The patients have gain-of-function mutations that shift activation in a hyperpolarizing direction—that make it easier to activate the channel—and once the channel has been activated, it stays activated longer. At this point, we’ve published maybe a dozen families from around the world, and we have another half a dozen or so in the pipeline.

The next thing to happen was the discovery of patients with channelopathy-associated congenital insensitivity to pain. They have null mutations. They basically don’t make functional 1.7 channels. These families are littered with painless fractures, painless burns, painless bites to the lips and tongue. So you have EM, which is a striking gain-of-function mutation, and congenital insensitivity to pain, channelopathy-associated, which is loss of function.

Then, a few years later, still another syndrome came to light. It hadbeen called familial rectal pain syndrome, and it’s been renamed PEPD, paroxysmal extreme pain disorder. It also has gain-of-function mutations in 1.7, but they’re different mutations. There’s a process called fast-inactivation—you activate a channel and it rapidly becomes inactivated, so it can’t be activated again. It’s sort of like after you flush a toilet, there’s a period when you can’t flush it again until it’s ready. In PEPD, the process of fast-inactivation is impaired—they don’t go through this period of not being able to be activated.

PRF: What is the story with peripheral neuropathy?

I’ve also been interested in peripheral neuropathy. Karin [Catharina] Faber and Ingemar Merkies and we started to collaborate. It really is to me a very special story, in terms of how a collaboration can really work very well, because [my lab has] unique capabilities in terms of studying sodium channels—very few labs in the world are able to do what we can do. In turn, they made a momentous contribution. Because they were interested in neuropathies, we started off studying idiopathic small fiber neuropathy. We will be studying other neuropathies.

Small fiber neuropathies cause autonomic dysfunction and profound pain. One sees degeneration of small nerve fibers—the longest ones being affected first—so toes and feet, and then hands. Over a period of three years, they collected, I believe, 266 patients who were referred to them with a clinical diagnosis of small fiber neuropathy. In most academic medical centers, you find a cause in around half of the patients, and, indeed, they found causes—diabetes, sarcoidosis, other underlying diseases—in three-quarters of these patients. They were left with 63 patients with idiopathic small fiber neuropathy, and 44 of them agreed to be in the study.

To do the functional profiling—to make the mutation, to express it, to do the electrophysiology and biophysics—if everything goes right, it takes a full-time electrophysiologist, and a couple of cell biologists and molecular biologists producing cultures and constructs, three months, sometimes four or five months. I said, look, I would love to study these, but only if we make the diagnosis by the most strict criteria—and that is, we want biopsy confirmation. And so the team in Maastricht got 28 of the patients biopsy-confirmed. So we had 28 gold-standard patients with exquisitely detailed clinical characteristics. And of them, eight, which is 28 percent, turned out to have mutations of 1.7. So it’s been one of these stories that has been really highly informative.

Then we looked at the biophysics of these channels. The way most labs do it is they put the channel in a heterologous system—a Chinese hamster ovary cell, or a human embryonic kidney cell. And we do that, and we often can find at least a first-pass description of what the mutation does. But it turns out that if you take the same channel and you put it into a motor neuron or a Purkinje neuron from the cerebellum or a sensory neuron (DRG neuron), it has different functional properties, as a result of the different cell backgrounds. We have the unique capability to put the 1.7 channel back into DRG neurons—the neurons where it’s normally expressed—and we can silence all the other sodium channels and assess in isolation the properties of the 1.7 channel, or the mutated 1.7 channel, in its native cell background.

Every one of the eight mutations caused biophysical changes, and they all were gain-of-function changes, but they all were different. In terms of channel biophysics, we’re seeing three phenotypes, all of which are new, and any one of them is quite different from the changes that are characteristic of EM mutations or PEPD mutations.

As expected, all of them made DRG neurons grossly hyperexcitable. They decreased the threshold to produce a single action potential—it’s easier for the cells to fire. And they increase the frequency at which they fire when you stimulate them—so with a small stimulus that might cause a DRG neuron with wild-type channels to go dot…dot, if you put in the mutant channels, you get dotdotdotdotdot—higher frequency firing. And the mutant channels caused these neurons to become spontaneously active.

We believe that the spontaneous activity is a mechanism that produces spontaneous baseline pain. We believe that the increased frequency of firing and the lower threshold produce hyperpathia and allodynia.

And, we don’t know for sure, but we think that what’s happening with these mutations—that they are eventually leading to axonal degeneration. I’m interested in how axons die in multiple sclerosis, and some years ago we showed that sodium influx via sodium channels can overload axons. Under situations where their homeostatic mechanism is exceeded, you outstrip their capability to remove intracellular sodium via the ATPase, and they have a higher-than-normal intracellular sodium concentration. There’s another molecule called the sodium-calcium exchanger, an antiporter molecule that normally carries sodium into cells from outside, and in return, extrudes calcium. But if there’s too much sodium inside of cells, the sodium-calcium exchanger can act in a reverse mode, and can bring calcium into axons, rather than take it out, and that can cause death of small axons. We know that free nerve endings, tiny nociceptive axons in the skin, contain not only the 1.7 sodium channel, but also 1.8, 1.9, and 1.6, and the sodium-calcium exchanger is also there. So our hypothesis, and I strongly suspect it’s going to turn out to be true, is that the gain-of-function mutations of 1.7 cause these tiny axons—and remember tiny axons have a very high surface-to-volume ratio—to become overloaded with sodium, and that causes the sodium-calcium exchanger to bring in calcium, which eventually leads to their degeneration.

PRF: And you said that’s something you all are working on now….

My colleagues here in my lab at New Haven want to kill me, because I’m pushing them so hard. I stick my head into their room every 20 minutes, saying, "What have you found?"

PRF: Does this give new impetus for Na_v1.7-selective channel blockers—because it looks like there are going to be whole new classes of diseases that these mutations are involved in?

There already was a search for blockers. And that is going on; my lab is part of it. It’s public knowledge that Yale has announced a collaboration between Yale and Pfizer and a company called Icagen. They have a compound that we’re very excited about, and we’re studying it, with them, and we interact with a number of other places. So the idea of going for selective blockers is paramount for us. But that was on the table before small fiber neuropathy.

I guess to me the exciting thing is, it’s a very important model disease, but EM in its classic form is a very, very rare disorder. In contrast to that, small fiber neuropathy is anything but rare. So it’s really exciting to be able to contribute to understanding a common disorder.

PRF: You started by talking about looking for families. Have you tested family members of the SFN patients with mutations?

We have tested some family members. We didn’t have a lot of family members who consented. The small number of family members that consented did fit the pattern—if they were affected, they had the mutation; if they didn’t, they didn’t. But we don’t have strong family trees. With erythromelalgia, we had one family with I think 46 people in it, in six generations. We have nothing that strong [for SFN]. You know, with genetic testing, some people volunteer to be studied; others don’t want to know.

PRF: What other things are you following up from this study?

We really want to understand the molecular basis for pain, and so in the case of erythromelalgia, having looked at the garden variety cases, we’re beginning to look at the outlier cases. You’d expect that somebody with a channelopathy that causes pain would have pain from early childhood, and most of them do, but we have some cases that have had adult onset, and we’re looking at the basis for that. Developmental splice isoform switching is a contributor to that, although there are other things going on. We’re interested in the same questions in terms of small fiber neuropathy.

[There are no] drugs for erythromelalgia. But we have one family [with EM] in which everybody who has tried it is responsive to carbamazepine. We’ve profiled their channels and done the pharmacology, and their mutation makes their channel anomalously sensitive to carbamazepine. That at least provides some support for the idea that personalized, genomically based therapies may not be unrealistic. So we’re going in that direction.

Why do erythromelalgia mutations, which have their unique physiological signature, cause pain in hands and feet, why do PEPD mutations have their different pattern of pain, and why do small fiber neuropathies have a different set of presentations? We certainly want to understand that. We’re working on the issue of why different patients in our small fiber neuropathy cohort have different degrees of pain, have different patterns of pain. We know, for example, there are some patients with small fiber neuropathy who begin by having pain in their face; we described one patient who began with pain around his jaw that was so severe he had teeth extracted, without relief. We’re right now looking at the effect of that mutation in trigeminal neurons, which innervate the jaw.

PRF: By exogenously expressing the mutant channels in those cells?

Yes. We also know that some patients had more profound autonomic symptoms than others, and on our drawing board is to put these mutations in sympathetic ganglia neurons and see if the mutations have more profound effects on sympathetic ganglia neurons that correlate with the clinical differences.

There are a bunch of scientific questions, and I’ve got a wonderful team of close to 30 people looking at all this. But Karin and Ingemar are neurologists, as I am, and ultimately our big goal is: What can we do for these people? So we’re very interested in therapy. Society invests in us, and our job is to return it. So pharmacology and rational development of mechanism-based therapies are real drivers for us.

PRF: Other issues you’d like to mention?

A significant number of patients on chemotherapy develop a painful neuropathy, and in some cases the pain is significant enough that the patients cannot continue chemotherapy. So the neuropathy is very functionally important, and next to nothing is known about it. One hypothesis we’re putting on the table is that chemotherapy doesn’t cause neuropathy; it triggers neuropathy in people who are genetically predisposed because they have mutations.

PRF: Either in the Na_v1.7 gene SCN9A or maybe something else?

It could be anywhere. My magnifying glass works best looking at SCN9A, but we’re prepared to look at all sodium channels, and it could be anywhere. With Ingemar and Karin we have already begun to collect patients beginning chemotherapy, and we’re collecting their DNA.