From clear-cut endings to complex beginnings
The renowned 19th-century French neurologist Jean-Martin Charcot advised young scientists that they should be prepared to deal with complex origins and manifestations of disease. But even he couldn't have predicted the hidden complexity of a disease he helped to discover — Charcot-Marie-Tooth disease (CMT).
CMT — first described by Charcot and his contemporaries Pierre Marie and H.H. Tooth in 1886 — causes weakness by damaging nerves that are required for muscle control, and is the most common inherited disorder of the nervous system. Despite its remarkable history and prevalence, CMT has no treatment, and its genetic origins were completely unknown until 1992.
The search for a genetic cause of CMT began in the mid-1980s. Around that time, researchers had discovered that Duchenne/Becker muscular dystrophy (DMD/ BMD) was caused by defects in a single gene. The same is true in both Huntington's disease and cystic fibrosis. In each case, the discovery of a single disease gene provided a key for understanding disease pathology and for designing possible treatments that are currently being tested.
Those successes provided CMT researchers with the technical advances and the optimism they needed to hunt for a CMT gene, but they didn't anticipate a simple resolution. By the 1970s, several types of CMT were recognized based on characteristic features of the disease within different families. So, researchers weren't expecting to find just one CMT gene.
But they've been stunned by the number of CMT genes their search has uncovered. As is the case for Huntington's, DMD/BMD and cystic fibrosis, a single defective gene can cause CMT in one person. But there appear to be at least 18 CMT genes among the world population. So far, seven of these CMT genes have been identified, and 11 more are suspected.
Although CMT is more complex than Charcot could have imagined, modern researchers are now beginning to unravel that complexity. Studies of CMT genes and their functions are leading to a better understanding of how nerves work, how they become damaged in CMT and how to protect them from damage.
|Peripheral nerves control movement by relaying impulses from the spinal cord (not shown) to our muscles (shown in the forearm). A single peripheral nerve is composed of many long nerve cell branches — or axons — that extend from the spinal cord and connect to muscle fibers. Each axon is surrounded by myelin made from the wrappings of Schwann cells.|
Despite its diverse genetic origins, CMT is universally recognized as an inherited disorder that affects the peripheral nerves — nervous tissue that connects the spinal cord to the muscles and sensory organs. Its defining symptoms are muscle weakness and wasting, and usually some loss of sensation, in the body's extremities (the forearms, lower limbs, hands and feet). Often, these symptoms appear in adolescence.
Because of these features, CMT is sometimes called hereditary motor and sensory neuropathy or HMSN (a neuropathy is any disease of the peripheral nerves). CMT also has a more old-fashioned name — peroneal muscular atrophy — which refers to wasting of the peroneal muscle in the lower leg.
Usually, CMT isn't life-threatening. But it can cause significant disability, including difficulty with routine activities like walking and grasping objects. Sometimes, a person with CMT requires braces or a wheelchair for mobility. The muscle weakness of CMT can cause deformities of the hands and feet, and the loss of sensation can cause minor foot injuries to go unnoticed, sometimes leading to infected sores called ulcerations. In severe cases, the ulcerations might require amputation.
Understanding how CMT affects the peripheral nerve is key to identifying treatments for the disease, says geneticist Jeffery Vance, who as an MDA grantee helped to find one of the first CMT genes, and who continues to study CMT at Duke University in Durham, N.C. "The peripheral nerve is not a really well understood organ," says Vance, "and until you understand its normal biology, it's difficult to understand its disease biology."
The peripheral nerves are bundles of nerve cell tendrils — or axons — which send electrical signals to muscles, and receive signals from sensory organs. Within each peripheral nerve, each axon is surrounded by a coating called myelin, which is produced by supportive cells called Schwann cells (see the illustration). The myelin insulates the electrical signals traveling through axons, allowing rapid transmission of the signals to and from the spinal cord.
Though it's been highly instructive, the emerging genetic complexity of CMT is wrecking the orderly distinctions among different types of the disease. With only five major types of CMT, and roughly 20 CMT genes, researchers are finding that mutations in different genes can cause the same type of CMT.
Not only that: "We're seeing that the same gene, depending on where the mutation is in the gene, can cause a different [type of CMT]," says geneticist Jeffery Vance.
Even at the beginning of the CMT gene hunt, it became clear that CMT genes weren't going to fit neatly into the disease categories that physicians and scientists had constructed. The hunt first focused on CMT1 — the major demyelinating type of CMT — and has uncovered at least three different "myelin" genes. About 80 percent of CMT1 cases are caused by mutations that affect the PMP22 gene (on chromosome 17), about 5 percent are caused by mutations in the P0 gene (on chromosome 1) and very rare cases are caused by mutations in the EGR2 gene (chromosome 10).
"That was a surprise for sure," says Vance.
Another surprise is that certain mutations in each of these genes can cause Dejerine-Sottas disease (a.k.a. CMT3). Because it's more severe than other types of CMT, Dejerine-Sottas has traditionally been considered a distinct disease. "Now we know that, genetically, Dejerine-Sottas is actually a severe version of CMT1," says Vance. Another "myelin" gene — called Cx32 — has been implicated in CMTX, and two others have been implicated in the demyelinating CMTs, CMT4B and HMSN-Lom. So, at first glance, there's a straightforward relationship between defective "myelin" genes and demyelinating forms of CMT.
But that relationship doesn't always hold up. "For instance," says Vance, "Cx32 mutations can cause a neuropathy that looks very much like the axonal form of CMT — CMT2. A lot of clinicians have had families that looked like CMT2 [based on NCV tests], but turned out to be CMTX [based on genetic testing]." In addition, some families thought to have CMT2 have turned out to have mutations in the P0 gene.
Although the complex relationship between CMT genes and CMT types gets confusing, patients shouldn't worry too much about it, says neurologist Michael Shy. Genetic testing is now available for mutations in the PMP22, P0 and Cx32 genes, so a patient can sometimes determine the genetic basis of his or her disorder. Ultimately, patients will find that information more useful than being told they have CMT1, CMT2 or any other classic type of CMT, says Shy.
"What physicians really have to do is go over the implications of what [the specific mutation] means in terms of who's at risk in the patient's family, what's known about the natural history of the disease, and to work with them about how best to remain independent," he says.
Still, says Vance, the sometimes unpredictable connection between CMT symptoms and their underlying cause can make it difficult for clinicians to help CMT patients. "If you're a clinician, you want to know: What type of test should I run on somebody who has this clinical picture?" says Vance. Besides that, patients will naturally want specific information about how particular mutations are likely to affect their health.
To improve the understanding of "what particular mutations do to CMT patients," Shy is helping to develop a national inherited neuropathy database as a collaborative effort between Wayne State University and Indiana University. The database will collect statistics on CMT patients, and use those statistics to correlate mutations with CMT symptoms.
To find out more about the CMT database, go to the clinical trials section of MDA's website. For physicians, registry information will also be available with genetic test results from Athena Diagnostics.
Long before the CMT gene hunt began, clinical examinations of patients suggested that CMT could be caused by inheritable defects in the axons or the myelin of peripheral nerves. Those defects ultimately damage the nerves, and are believed to be especially harmful to longer nerves, which explains why CMT mostly causes motor and sensory problems in the body's extremities.
Based partly on whether it's caused by defects in axons or in myelin, and partly on its pattern of inheritance, CMT has been divided into five major types.
CMT1 and CMT2 are the most common types, and both are autosomal dominant, meaning that a child can inherit the disorder from one parent. CMT1 is caused by a loss of myelin (demyelination), which physicians can detect by using a nerve conduction velocity test (NCV) to show a decline in the speed of nerve signaling. In contrast, CMT2 is caused by defects in axons (axonopathy), and shows little or no slowing of the NCV.
The three other types of CMT are all caused by demyelination. CMT3 (usually called Dejerine-Sottas disease) and CMT4 are both very rare. They're autosomal recessive, meaning that a child can inherit the disease only if he gets the gene from both parents. Finally, CMTX is X-linked dominant, meaning that a girl can inherit the disease from either parent, while a boy can inherit it only from his mother.
Pretty simple, huh? Not quite. For a more in-depth look at CMT types, see A Pathological Puzzle.
Schwann cells: more than myelin
Much to the satisfaction of researchers, the identification of CMT genes has confirmed the idea that CMT can be caused by defects in either the myelin or the axons of peripheral nerves. Six of the known CMT genes appear to be required in Schwann cells (which make myelin), and one is known to be critical for axon structure. (For descriptions of the seven identified CMT genes and their functions in peripheral nerve, see The Players.)
Identification of these genes also helped establish the surprising discovery that defects in myelin or in axons ultimately produce the same type of damage in peripheral nerves — destruction of the axons.
"As more and more causes of CMT became apparent, this gave more clues to how peripheral nerve damage could occur," says neurologist Michael Shy, who's studying possible CMT treatments at Wayne State University in Detroit with MDA support. "Investigators have now begun to realize that even in the demyelinating forms of CMT, disability correlates more with degeneration of the axons inside the myelin than the demyelination itself." This means that a child with a family history of CMT might show signs of demyelination during an NCV test, but won't begin to show symptoms of CMT until the demyelination causes axonal damage, explains Shy.
|For axons and Schwann cells, communication is the key to a healthy relationship. Axons send chemical messages (red) that attract Schwann cells and encourage myelin formation, and Schwann cells appear to send messages (green) that nourish and protect axons.|
How does the loss of myelin eventually destroy axons? Researchers suspect that some essential communication must normally take place between axons and Schwann cells. "One thing that researchers are learning is that the Schwann cell isn't just hanging around making myelin," says Vance. "It's clearly involved in regulating the health of the axon."
Nonetheless, the process by which Schwann cells make myelin might reveal clues to how they nurture axons. In the developing nervous system, the Schwann cells essentially become myelin by lining up along an axon and wrapping themselves around the axon to form successive layers (see illustration). The axon sends the Schwann cells chemical messages that tell them to form myelin, and the Schwann cells appear to respond with their own messages. In the mature nervous system, continued communication between the axon and the Schwann cells appears necessary to keep the myelin intact and the axon healthy, explains Shy.
Mutations in the different "myelin" genes probably cause demyelination in different ways, Shy points out. "Figuring all these out is likely to be tricky," he says, "so the current research is beginning to focus on how Schwann cells communicate with axons and how axonal degeneration occurs." Finding a means to promote normal communication between Schwann cells and axons might provide a way to treat CMT, he says.
In the late 1980s, while the CMT gene hunt was gaining momentum, neurologist Phillip Chance was searching for the genetic cause of an apparently unrelated neuropathy — hereditary neuropathy with liability to pressure palsies (HNPP).
To his surprise, Chance's investigation revealed that HNPP and most CMT1 cases result from the same mutation event. Once that mutation occurs, the emergence of either HNPP or CMT1 in a family is sort of like flipping a coin. HNPP is an inherited disorder that causes temporary attacks of paralysis typically localized to a single limb. Those attacks take weeks to months to recover from, and are brought on by palsy — the malfunction of a peripheral nerve that serves the affected limb.
"[HNPP] is so different clinically from CMT," says Chance, "that few people suspected it had anything to do with CMT."
But in 1992, Chance published an MDA-funded study showing that HNPP is genetically related to the most common form of CMT1 — CMT1A. Earlier that year, MDA-funded researchers had established that CMT1A is caused by a duplication of the PMP22 gene on chromosome 17. (It's important to remember that, except for the X and Y chromosomes, everyone normally has two copies of each chromosome, one from dad's sperm and one from mom's egg.
So, having the duplication means that a person has three copies of the PMP22 gene.) The resulting overproduction of the PMP22 protein — a component of myelin — is somehow detrimental to Schwann cells. Though it's a rare event, the duplication usually arises while single copies of dad's chromosomes are being parceled into his sperm. As his two copies of chromosome 17 are preparing for separation into the sperm, a piece of genetic material containing PMP22 is accidentally removed from one copy of the chromosome and inserted into the other. So, one sperm receives an extra PMP22 gene, and one sperm loses its PMP22 gene.
"The existence of the duplication predicted that there would be a reciprocal deletion, but it was generally assumed that the deletion would be lethal, of no consequence, or it could have resulted in a disorder which wasn't necessarily a peripheral neuropathy," says Chance, who's now at the University of Washington in Seattle. Chance's research showed that deletion of the PMP22 gene — the flip side of the mutation underlying CMT1A — is the cause of HNPP. "We had no reason to think that studying HNPP would give us any particular insights into CMT," says Chance. But it has helped make clear that peripheral nerve is sensitive to high or low levels of PMP22.
"The relationship with CMT also has done a lot to bring HNPP into the clinical arena," says Chance.
Protective genes and potential treatments
So far, researchers haven't identified any specific chemical messages that Schwann cells produce to maintain axon health. However, they're working on other promising ways to prevent the axonal damage that underlies CMT.
With MDA support, Shy is developing a gene therapy method to supply ailing axons with glial-derived neurotrophic factor (GDNF), a naturally occurring protein that stimulates nerve cell growth and survival. Rather than preventing demyelination, the goal of gene therapy with GDNF is to protect axons from the effects of demyelination.
An advantage of Shy's approach is that it has potential for treating any type of CMT, regardless of the underlying genetic defect.
In contrast, conventional gene therapy — trying to correct a genetic defect by supplying normal copies of the defective gene — would have to be tailored to each of the roughly 20 CMT genes.
Besides that, many cases of demyelinating CMT aren't caused by loss of an essential gene, but by a toxic gain of gene function (see CMT1A and HNPP). In those cases, conventional gene therapy won't work because "giving more of the natural protein isn't going to remove the bad one," says Shy.
At this point, Shy and his Wayne State colleagues John Kamholz and Gyula Acsadi have packaged the GDNF gene into viruses that can be used to safely deliver it to muscles and nerves. They've shown that when the gene-laden viruses are injected into mouse muscle, GDNF gets produced in the muscle, and taken up by nerve cells connected to the muscle.
Their next goal, says Shy, is to test this method in mouse models of CMT and amyotrophic lateral sclerosis (ALS), a paralyzing disease brought about by the death of muscle-controlling nerve cells. "In both of these models, the basic concept is figuring out how to prevent axonal degeneration," says Shy.
While Shy and his colleagues flesh out their gene therapy approach, the hunt for CMT genes continues. But instead of just focusing on genes that can cause CMT, investigators are starting to search for genes that can modify severity of the disease, says Jeffery Vance. Evidence for such modifier genes comes from the observation that individuals within a single CMT family can have very different clinical features.
For other inheritable diseases, such as familial ALS, researchers are homing in on modifier genes by studying mice. When mice that have a dominant ALS-causing mutation are mated with inbred, healthy mice, some offspring inherit the mutation but develop the disease significantly later than expected — suggesting they've inherited beneficial modifier genes from their healthy parents.
Vance expects that researchers will soon use similar methods to track down modifier genes that affect CMT, and says that his lab has begun to look for such genes in humans. In general, modifier genes "hold real promise" for CMT treatment, he suggests.
"If you understood how to modify the effect [of a defective gene], that's a lot easier than trying to replace the gene."
The seven identified CMT genes encode proteins that perform essential (though in some cases unknown) functions in the peripheral nerves. Though malfunction of any of the proteins can lead to CMT, the normal functions of each protein appear quite different.