Kennedy's Disease

Also called SBMA, it was unrecognized just 40 years ago. Now it's on a fast track toward treatment.

by Dan Stimson

Photo by Stuart Zolotorow Kenneth Fischbeck, whose lab at NINDS is focused on KD research, talks with lab member Federica Piccioni about the results of her latest experiment — part of an NINDS-sponsored effort to identify drug treatments for KD.

After years of frustration and unanswered questions, Len Janicki submitted to a final lab test — a blood test that he hoped would explain why his 48-year-old body felt weak and exhausted.

In high school, he'd been an athlete and an avid participant in almost every team sport imaginable. In college, he joined the Army Reserve, and later became company commander of a basic training unit.

He moved up through the Army's ranks and settled in Buffalo, N.Y., and over the years, his work gradually became less physical and more sedentary. His body seemed to respond by getting out of shape — but in his early 40s, he realized he had a more serious problem.

Slight muscle cramps and tremors, an ever-present annoyance since his youth, became painful. He began to have trouble climbing stairs and lifting heavy objects. In 1993, after a day spent walking through parking lots and up and down stadium steps to attend a sporting event, his friends nearly had to carry him back to his car.

Tests and More Tests

---

Doctors suggested that Janicki was "just getting old," or that perhaps he had a drinking problem. Finally in 1996, he was sent to neurologist Carolyn Warner, director of the Erie County MDA clinic in Buffalo.

After an EMG, an MRI, a spinal tap and other tests, she narrowed his symptoms down to one of two diseases: amyotrophic lateral sclerosis (ALS), a paralyzing disease that's usually fatal within three to five years of diagnosis, or Kennedy's disease (KD), a genetic disorder that can look like ALS, but often has no effect on life span. (See "Is It KD or ALS?")

His blood test, a genetic test for KD, came back positive.

It was "a bad news or worse news scenario," Janicki says. "I was very relieved that I didn't have ALS, but by the same token, I did have KD."

Like ALS, KD has no cure. And although it usually isn't fatal, it causes slowly progressive weakness that interferes with mobility and other basic functions like chewing, swallowing and speaking (see "Symptoms and Complications of KD").

Janicki's weakness has progressed to the point that he walks with a cane over short distances and uses a scooter for longer distances. Because of weakness in his throat muscles, he has trouble breathing during sleep and uses a ventilator at night. He's had to retire early and move into a one-story home.

But Janicki and others with KD may receive some good news in the near future. Researchers have gained key insights into the mechanisms of the disease and are on the trail of several drugs that have the potential to stop its debilitating course.

"We're not there yet ... but we've made good headway," says Kenneth Fischbeck, a leader in KD research and chief of the Neurogenetics Branch of the National Institute of Neurological Disorders and Stroke (NINDS) in Bethesda, Md. In fact, KD researchers have made quite impressive progress, considering that the disease was completely unknown to the medical community just 38 years ago.

'Maybe This Isn't ALS'

---

In 1964, fresh out of a neurology residency, William R. Kennedy joined the Neurology Department at the University of Minnesota in Minneapolis, where he was put in charge of teaching residents techniques for diagnosing neuromuscular disorders, including electromyography (EMG) and nerve and muscle biopsies. That same year, he received a referral to perform a neurological exam and EMG on George B., a 57-year-old man from the St. Paul area who'd been given a possible diagnosis of ALS.

ALS kills muscle-controlling nerve cells called lower and upper motor neurons. The lower motor neurons, located in the brainstem and spinal cord, connect directly to muscles; their activity is fine-tuned by the upper motor neurons, located in the forebrain.

George B. had signs and symptoms of lower motor neuron damage, including progressive weakness and muscle twitches known as fasciculations, but no evidence of upper motor neuron damage. What was even more peculiar, he told Kennedy that one of his brothers and two male cousins had a similar disease.

"ALS usually is not hereditary. So when you find two people in the same family, you start to say wait a minute — maybe this isn't ALS," says Kennedy, who's now 75 and still practicing neurology at the University of Minnesota.

Family Clues

---

William R. Kennedy practices neurology at the University of Minnesota.

It turned out that George had a large family with a long history in Minnesota, especially on Grey Cloud Island — a bank on the Mississippi River, cut off from the mainland by a swamp — where his ancestors had settled in the mid-1800s. By taking repeated trips to the island and surrounding areas, Kennedy and a team of doctors examined other family members for signs of the disease, and combed through birth, marriage and death certificates, carefully tracing the disease through the family's past.

They discovered that in every generation, the affected family members were male, middle-aged to elderly, and, with one possible exception, had inherited the disease from their mothers.

Around the same time, Kennedy received a visit from Robert G., a 68-year-old Iowan who seemed to have the same strange disease. Investigation of the G. family revealed male relatives with the disease, passed down from their mothers.

A few years later, Robert died of pneumonia and his wife asked Kennedy to examine tissue taken at autopsy. Consistent with the symptoms in both families, Kennedy found extensive damage of motor neurons in the spinal cord and bulbar (bulblike) part of the brainstem, and a variable degree of muscle atrophy (wasting).

In 1968, Kennedy published his findings on the B. and G. families, and called their disease spinal-bulbar muscular atrophy (SBMA), a designation that's still widely used in scientific journals. In lectures and in conversation, though, his colleagues began to use the shorter, less technical name Kennedy's disease, and it stuck.

"I wondered what they were talking about at first," Kennedy says. He's equally humble regarding the fact he spotted a disease other neurologists had long overlooked.

"All of us have blinders on to certain things.When I was training, Alzheimer's was an extremely rare disease, and we called it presenile dementia. Sometimes, we just don't see [the disease] or we think it's something else."

The Hunt for the KD Gene

---

Kennedy's work — recognized in 1998 by the American Academy of Neurology (AAN) as one of the 15 most influential studies in the academy's then 50-year history — not only established KD as a distinct disease; it showed that KD is sex-linked or "X-linked," caused by a defective gene somewhere on the X chromosome.

Women, who have two X chromosomes, can be silent "carriers" of the defective KD gene, although KD carriers do occasionally develop mild symptoms. Men, who have a Y chromosome inherited from their fathers, and only one X inherited from their mothers, get the full-blown disease.

But what was the culprit gene underlying KD? The answer would have to wait for the advent of gene mapping technology in the 1980s and an experienced geneticist to take on the project.

In the early 1980s, Kenneth Fischbeck went to an AAN conference and attended a seminar on KD and other unusual motor neuron diseases given by Lewis Rowland, then co-director of the MDA clinic at Columbia University in New York. At the time, Fischbeck was an assistant professor of neurology at the University of Pennsylvania in Philadelphia, and was collaborating with Louis Kunkel of Boston Children's Hospital on MDA-funded work that would eventually lead to the discovery of the X-linked gene underlying Duchenne muscular dystrophy.

"When [Rowland] mentioned that there was an X-linked form of motor neuron disease, I kind of perked up," Fischbeck recalls. "I went back through the literature and got in touch with William Kennedy and other people who had reported families [with KD]."

The Androgen Connection

---

Ultimately, he found several large families in the Philadelphia area, and by collecting blood samples and comparing genetic material from affected and unaffected family members, he narrowed the KD gene down to a small region of the X chromosome containing just a few genes. One of those, the gene encoding the androgen receptor protein, particularly caught his eye.

In his 1968 study, Kennedy had noted a high incidence of gynecomastia, or breast development, among men with KD. Later studies confirmed that finding, and also pointed to a high incidence of testicular atrophy and infertility, suggesting that men with KD might have defects in the way their bodies handle testosterone and other masculinizing hormones, collectively known as androgens.

The androgen receptor, found in neurons and other cell types in the body, enables cells to respond to androgens. So, Fischbeck had hit an obvious candidate for the KD gene.

In 1989, with MDA support, Fischbeck and a team of researchers began probing for mutations in the androgen receptor gene in men with KD. One of Fischbeck's graduate students, Al La Spada, discovered that each man had the same type of mutation — one that had never been seen in any genetic disorder.

Spelled out in the chemical letters that make up DNA, the mutation looked like a run-on sentence. In men without KD, the androgen receptor gene contained about 21 repeats of the three-letter phrase "CAG," but in men with KD, it had between 40 and 52 CAG repeats.

"I was quite flabbergasted because back then, the view of genetic mutation was that it came in two flavors: point mutations, where a single letter of the DNA alphabet gets changed, or gross rearrangements of DNA, like a large duplication or deletion," recalls La Spada, who's now studying KD in his own lab at the University of Washington in Seattle.

In fact, La Spada's finding was so unusual that it was hard to get published. But today it's an example of trail-blazing genetic research, as this type of mutation — now known as a trinucleotide repeat expansion — has since been implicated in more than a dozen other hereditary diseases.

"MDA really deserves some credit for that," says Fischbeck.

A Test and Clues to Treatment

---

For men with KD, La Spada and Fischbeck's discovery led to a genetic test for the disease — the same one that established Len Janicki's diagnosis.

But, more important, it pinpointed the trigger for the disease.

Since a single CAG repeat is the genetic code for the protein building block glutamine, the expanded CAG repeats underlying KD create an abnormal structure within the androgen receptor called expanded polyglutamine. It's not clear why expanded polyglutamine is so harmful, but KD researchers are closing in on an answer, MDA grantee Diane Merry says.

"I see the last seven or eight years as an accumulation of information about how the polyglutamine expansion affects the properties of the androgen receptor," says Merry, a former postdoctoral fellow with Fischbeck, now running a KD research lab at Thomas Jefferson University in Philadelphia.

Normally, the androgen receptor regulates transcription — the process of turning on genes — in a manner that's dependent on androgens. Clearly, expanded polyglutamine interferes with this function, since at least some men with KD have a reduced sensitivity to androgens (manifested as gynecomastia, infertility and/or testicular atrophy).

But that's not the whole story. Genetically male (XY) fetuses that have a complete loss of androgen receptor function develop as females (except they lack ovaries or uteruses), but they don't develop KD or any other neurological deficits. So, researchers believe that expanded polyglutamine gives the androgen receptor a new function that makes it toxic to lower motor neurons.

At least eight of the other trinucleotide repeat diseases, including Huntington's disease and several forms of spinocerebellar ataxia (SCA), are also "polyglutamine diseases," caused by expanded CAG repeats in distinct genes. In these diseases, the affected protein also appears to take on a new toxic function.

Understanding that common characteristic has helped to accelerate KD research, Merry says.

"Partly because of work done in other polyglutamine diseases, the pace has really picked up," she says. "In my own lab and others, there's a lot of work directed toward the development of therapies."

The Path to Drug Discovery

---

In hopes of finding ways to correct it, Merry and other KD researchers are examining the toxic effect of expanded polyglutamine in laboratory models of KD. Fischbeck and Merry have created cell models of KD by introducing the mutant androgen receptor gene into motor neuron cell lines.

In MDA-funded work, Merry also created a mouse model of KD, which proved to be a bit more challenging. Early attempts resulted in perfectly healthy-looking mice, and later attempts led to mice with a widespread neurological disease reminiscent of Huntington's.

Last year, by tweaking the mutant androgen receptor's toxicity to just the right level, she got mice with a disease strongly resembling KD. Fruit fly and nematode (worm) models of polyglutamine disease are also being used in KD research.

Examination of these models and of tissue from men with KD has revealed that KD fits a common theme for polyglutamine diseases: Expanded polyglutamine, whether it's in the androgen receptor or another protein, tends to accumulate in little cellular garbage heaps called aggregates. KD researchers believe that accumulation of expanded polyglutamine plays an important role in KD, most likely by gumming up the androgen receptor and other proteins that control gene transcription.

"Polyglutamine tends to stick to itself, so normally it may serve as a kind of Velcro that allows proteins to stick to each other," Fischbeck says. "When it gets too long, it might stick too much."

He adds, "The main focus of research now is to use the various model systems to screen for drugs" that could target expanded polyglutamine or other factors in KD.

Reasons for Optimism

---

NINDS is sponsoring some 30 labs across the country to test every drug approved by the Food and Drug Administration (there are about 1,100) for its effects on models of different neurodegenerative diseases, including those caused by expanded polyglutamine.

KD researchers are paying close attention to drugs capable of blocking the formation of aggregates, as well as those that might block other steps in the KD pathway, including "antiandrogens" and chemicals that act at the level of gene transcription (see "Sizing Up the Enemy").

"A treatment that's developed for any one of the nine polyglutamine diseases is likely to be effective against the others," Fischbeck says, and, in an ironic, but encouraging twist, drugs that work against KD might also work against ALS. Moreover, since the drugs being screened are FDA-approved, they could move rapidly into clinical trials.

"I'm optimistic that some treatment for people [with KD] is going to come out of this," Fischbeck says. "In some ways, I would have liked research to be further along than it is now, but I can see a light at the end of the tunnel. By the time I'm Dr. Kennedy's age, I'd like to be able to look back and say this disease is a problem that we licked."

Studies of the mutant androgen receptor that causes Kennedy's disease have led to key insights into how it damages nerve cells in the brainstem and spinal cord.

The normal androgen receptor is located in the main compartment of the cell — the cytoplasm — where it waits for androgens, fat-based steroid hormones that easily pass through the fatty membrane surrounding the cell. Normally, attachment of androgen to the receptor stimulates the receptor to move into the nucleus, the cell compartment that contains DNA. Once inside the nucleus, the receptor acts as a transcription factor, a type of protein that latches onto DNA to turn genes on or shut them down.

In KD the androgen receptor contains an expanded polyglutamine tract that interferes with its normal function and apparently that of other transcription factors, ultimately damaging the cell. Researchers don't understand all the steps between the mutant receptor's activation and degeneration of the cell, but they've identified four events in the disease process that are prime targets for therapy:

		1. tongue 2. arm muscle 3. spinal lower motor neurons 4. leg muscle 5. brainstem (bulbar) lower motor neurons
Kennedy's disease (KD), also called spinal-bulbar muscular atrophy, causes the degeneration of lower motor neurons in the brainstem and spinal cord. It arises from a triplet repeat expansion — a type of mutation — in the gene encoding the androgen receptor. Normally, in motor neurons and other cell types, the androgen receptor regulates gene expression. In motor neurons affected by KD, the androgen receptor gains new, toxic properties, including a tendency to form garbage heaps called aggregates.

1 - Binding of androgen to the mutant receptor

Androgen itself has long been a focus of potential treatment for KD. Since KD is partly caused by a loss of androgen receptor function, some clinicians have tried giving men with KD supplemental androgen to rev up the receptor. In those men, short-term androgen treatment appeared to improve strength, but since it's now clear that the mutant receptor has a toxic gain-of-function, researchers suspect that long-term androgen treatment could accelerate weakness.

In fact, recent studies on a mouse model of KD suggest that a natural deficit of androgen might be what really protects women from the disease. Diane Merry created the model by giving mice a copy of the mutant androgen receptor gene. The gene was randomly inserted into an autosome (a chromosome other than the X or Y), but that genetic quirk didn't change the sex-specific effects of the disease: Male mice get the disease and females don't.

"In our model, we've taken out the effects of the X chromosome ... so we think the females are protected because they have low levels of circulating testosterone," Merry says. "This suggests that testosterone for patients is not necessarily a good thing." Merry and others are now using the mice to test the possible benefits of antiandrogens — drugs that block interaction of androgen and its receptor.

2 - Protease cleavage of the mutant receptor

During its normal processing, the androgen receptor sometimes gets cleaved into several fragments by protein-cutting enzymes called proteases. Experiments on mouse and cell models of KD show that fragments of the mutant receptor, rather than the entire protein, may be the real culprit.

Cleavage of the mutant receptor seems to be critical to its toxicity, says MDA grantee Lisa Ellerby, a biochemist at the Buck Institute for Research in Aging in Novato, Calif.

In cell models of KD, she says, "If the androgen receptor is cleaved, you get enhanced cell death. We think the fragments interact with the full-length protein and change it functionally." Picture breaking off a key in a lock. Not only is the broken key useless, but it's made the lock useless, too.

Drugs that inhibit proteases thus hold potential for treating KD, Ellerby says. "Our hope is to find a site of interaction between the androgen receptor and the proteases, and then find [a drug] that literally knocks the protease off the receptor," she says.

3 - Aggregation of the mutant receptor

Expanded polyglutamine — which Kenneth Fischbeck compares to Velcro that's too sticky — causes the mutant androgen receptor to clump in garbage heaps called aggregates, often inside the nucleus. While it's not clear whether the aggregates themselves are toxic, their presence suggests that the cell isn't efficient at breaking down the mutant protein, says Merry.

Usually, abnormal proteins are repaired by proteins called chaperones. If they're beyond repair, the chaperones escort them to a sort of cellular garbage disposal called the proteasome.

"So what's happening when you've got this accumulation of mutant protein? One idea is that it's clogging up the proteasome, sort of acting like a trash strike," Merry says. The proteasome might become overburdened with cellular garbage, unable to clear away damaged or useless proteins, which could then clog up other essential cell pathways.

Merry recently showed that genetic enhancement with extra chaperones improved the breakdown of the mutant androgen receptor in KD. Gene therapy or certain drugs could be used to increase chaperone levels in men with KD, she speculates.

4 - Interaction between the mutant receptor and other transcription factors

Besides sticking to itself, the mutant androgen receptor appears to stick to other proteins, preventing them from carrying out their functions in the cell.

Using a cell model of KD, Fischbeck found that the mutant androgen receptor tends to trap and disable CBP, another transcription factor that contains polyglutamine. As expected, giving the cells extra CBP genes protected them from the mutant androgen receptor.

But there may be an easier way to make up for lost CBP, Fischbeck says.

CBP has acetylase activity, which means that one of its functions is to loosen up DNA for access by other transcription factors. Without CBP, deacetylases take over, keeping the DNA in tight coils.

So, Fischbeck has begun using his cell model of KD to test chemicals that block the deacetylases, called histone deacetylase inhibitors (HDIs). Other labs are testing the effects of HDIs in models of Huntington's disease.

"They seem to work in several models of polyglutamine disease," he says. "Would they work in humans? That's an interesting lead that we're pursuing."