by Dan Stimson

In troubled times, John and Monica English of Orem, Utah, have always believed in hoping for the best. Its a philosophy that has served them well.

When their daughter, Taleah, was 9 weeks old, the Englishes noticed she couldnt lift her head. After a month filled with doctors visits and tests, she was found to have spinal muscular atrophy (SMA), a paralyzing genetic disease that affects about one in 10,000 people.

Doctors said Taleah would never sit up by herself or walk, and that she probably wouldnt live to see her second birthday. The first two predictions have become a matter of debate, and the third has proved to be flat-out wrong.

Taleah is now 3 years old, and although shes unable to move most of her body, in other respects, shes typical for a girl her age.

When Monica became pregnant with son Colin in 2002, the Englishes knew there was a possibility he, too, would have SMA. Prenatal genetic testing confirmed their fears, but also revealed that Colin was likely to develop a less severe form of the disease than Taleah has.

Colin is now 9 months old and shows no signs of SMA. As always, the Englishes are sustained by hope theyre hoping that Colin will never develop symptoms, and that Taleahs health will at least remain stable.

Fortunately, they have more than just hope on their side. Through clinical trials under way near their home, Taleah and Colin are receiving experimental drug treatments that may slow the course of SMA for them and thousands of other children.

The trials, based at the University of Utah in Salt Lake City, began after laboratory studies elsewhere showed that the drugs might compensate for the flawed gene that causes SMA. Meanwhile, ongoing experiments are expected to reveal dozens more drugs with the same effect.

A Crash Course in SMA

When they found out Taleah had SMA, the Englishes knew nothing about the disease. No one in their family had ever officially had it, although they now believe SMA might explain why two of Monicas great-aunts mysteriously died in infancy.

"I immediately started reading everything I could get my hands on," Monica says.

SMA, she learned, kills nerve cells in the spinal cord called motor neurons, causing the muscles connected to them to waste away. There are several clinically distinct forms of the disease, but about 98 percent of cases fit one of three types.

SMA type 1, or Werdnig-Hoffmann disease, manifests within the first six months of life. By definition, children with SMA1 are never able to sit unaided. They have problems with suckling and swallowing, and their respiratory (breathing) muscles may become so weak that their chests sink in. Despite a growing number of exceptions, many children with the disease die from respiratory failure within the first few years of life.

SMA type 2, or intermediate SMA, usually manifests in the first 18 months of life. Unlike children with SMA1, those with SMA2 eventually are able to sit unaided, but most dont stand or walk. Respiratory weakness is a serious danger, but life expectancy usually extends well into adulthood.

In SMA type 3, or Kugelberg-Welander disease, symptoms usually appear after 18 months. People with SMA3 can walk, but often lose the ability by the time they reach their 30s or 40s. With proper respiratory care, many people with SMA3 live into their golden years.

Beyond the Diagnosis

Taleah, if you havent already guessed, has SMA type 1. She beat her dire prognosis largely thanks to an innovative respiratory care plan. Taleah uses an air pressure ventilator with a mask to help her breathe, and another machine to help her cough.

Taleah cant walk, but she gets around fine in a standing power wheelchair. She cant lift her elbows, but she can use her hands. The muscles in her throat arent strong enough for her to swallow, so she uses a feeding tube. But she can speak and has "quite a sense of humor and quite an understanding of whats going on," her mom says.

"Shes not defined by her disease. She plays with tea sets and coloring books and paints pictures; she just has to do it a little differently."

Colin, since he hasnt yet developed symptoms, is expected to develop SMA2 or SMA3. (He has a nasogastric feeding tube because its the only way hell take the bitter-tasting medication hes getting through the Utah trial.)

It was difficult to accept Colins test results, Monica says. "Yet it was equally joyful when we found out how much less the disease would affect him. Its a world of difference between type 1 and type 2."

Genetic Secrets Revealed

Despite the observation that different types of SMA can occur within the same family, scientists once debated whether they were manifestations of a single disease, or three diseases, each linked to a unique gene.

In the late 1980s, a team of French scientists and a team of American scientists funded by MDA set out to resolve the dispute by collecting DNA from dozens of families with SMA.

By 1995, they showed that SMA types 1, 2 and 3 but not other, rarer versions of SMA could be traced to defects in a single gene located near the tip, or telomere, of chromosome 5. A nearly identical gene the result of an imperfect duplication some time during human history was found near the chromosomes center, or centromere. The French scientists named the two sister genes survival motor neuron-t and survival motor neuron-c, but theyre now usually called SMN1 and SMN2.

The discovery of SMN1 opened up many possibilities, including genetic testing for SMA. It turns out that in 95 percent of people with SMA, the SMN1 gene is deleted from both copies of chromosome 5. (One copy is inherited from each parent.)

In another 3 percent, one copy of SMN1 is deleted and the other is interrupted by a small mutation. The remaining 2 percent to 3 percent have defects in genes other than SMN1, some of which arent yet identified.

Scientists also quickly identified the SMN protein thats normally made from the SMN1 gene, and several MDA-funded groups began to examine this protein in detail.

Studies showed that SMN1 mutations reduce the amount of the protein in nearly all the bodys cells a discovery that remains puzzling given the diseases specific effect on motor neurons. Many scientists began to focus their efforts on using gene therapy to give people with the disease an intact copy of SMN1.

A Backup Gene

Meanwhile, the discovery of SMN2 gave scientists an unexpected alternative to gene therapy. It might be possible, they realized, to use SMN2 as a backup gene for SMN1.

"The obvious question was: Can you treat SMA by stimulating the SMN2 gene to make more SMN [protein]?" recalls Arthur Burghes, who was part of the MDA-funded team that helped find the SMN genes.

Burghes was one of the first scientists to tackle this question, by studying SMN in mice. Mice, it turns out, have only a single SMN gene.

Burghes, a geneticist at Ohio State University in Columbus, deleted both copies of the gene, and found that the mice died during embryonic development. But when he gave the SMN-deficient mice one or more copies of SMN2, they survived embryonic life and developed SMA type 1, 2, or 3, depending on how many copies of SMN2 they had.

Studies of people with SMA confirmed the results of Burghes mouse experiments. People with SMA type 1 tend to have one or two copies of the SMN2 gene, but people with SMA type 2 or 3 often have three or four copies of the gene. (Taleah, for example, has two copies of SMN2, while Colin has four.)

The bottom line: the more copies of SMN2, the better.

With this critical insight, scientists began to envision using drugs to stimulate production from the SMN2 gene. The problem was finding a drug with the right properties.

Some researchers took a "fishing expedition" approach, screening for candidate drugs by testing hundreds of thousands of chemicals for their effects on SMN2. Others focused experiments on existing drugs known to affect gene activity, and many have used a combination of both approaches.

So far, the two most promising candidate drugs to emerge from this search are valproic acid and phenylbutyrate. (For more about how these drugs might work against SMA, see "The Hunt for SMA Drugs.")

Drug Testing Begins

Kathryn Swoboda, a neurologist at the University of Utah who has support from MDA, is testing each drug in separate clinical trials.

Swoboda was trained as an adult neurologist, and spent much of her residency using electrophysiology (recording activity from muscles and nerves by means of electrodes) to study adult movement disorders like Parkinsons disease. But when the SMN genes were discovered, she saw an opportunity to use her skills to help children with SMA.

Troy Lino, 6, and his sister Jamie, 9, (below) travel to Salt Lake City from Denver to participate in clinical trials of drug treatments for SMA. Here, Kathryn Swoboda measures Troys motor units.

"There was a lot of excitement" about potential treatments for SMA, she says. But there was no good way to measure the effects of such treatments on young SMA patients.

"You cant do reliable strength testing in very young children. You can ask toddlers to do things, and they may just say no," she explains.

Swoboda thought that an electrophysiological technique called motor unit number estimation (MUNE) might be a good alternative to strength testing. A motor unit is a single motor neuron and all of the muscle fibers connected to it. In previous work, Swoboda has used MUNE to show that children with SMA begin to lose motor neurons around the time that symptoms appear. In the trials, shes using MUNE to see if valproic acid or phenylbutyrate can slow this loss of motor neurons.

Both trials, which were open to children with any type of SMA, are now nearing completion. Because investigators suspect that younger children have the most potential to benefit from therapy, the phenylbutyrate trial was limited to children under 2. The valproic acid trial was limited to children 2 years of age or older, because of concerns about toxicity.

"There are rare deaths associated with valproic acid, typically among children less than 2 [who have taken the drug for epilepsy]," Swoboda explains.

"Its a safe medication in most people, but SMA kids arent most people. Even kids who are older may have a higher risk of complications," she warns.

Physical therapist Janine Wood tests Jamies strength. Photos by David Ricketts

Swoboda expects to have results from the valproic acid trial by this fall. The phenylbutyrate trial may take longer: The high cost of the drug has been prohibitive, and finding infants with SMA who are well enough to travel to Utah has been difficult.

(For information about the phenylbutyrate trial, which is open to children under 2, including unborn babies predicted to develop SMA, contact Mark Wride at mwride@genetics.utah.edu, or call Swoboda at (801) 585-9717.)

Both trials are primarily meant to address safety, and Swoboda hopes that at least one of the drugs will be worth moving into a larger trial to address efficacy.

"For many disorders where theres a protein deficiency, you dont have to fully correct the problem to see a benefit, so were hoping that a modest increase in SMN2 will make a difference," she says.

Guarded Hopes

The Englishes are eagerly watching to see if the drugs will help Taleah, whos enrolled in the valproic acid trial, and Colin, whos in the phenylbutyrate trial.

"When I hear the hope in doctors voices, that tells me these drugs might do good things," Monica says. "And weve seen little things with Taleah that have made us hopeful.

Taleah English "suctions" her bunny, with which she shares her BiPAP noninvasive ventilation device.

"She can turn her head and her speaking voice has gotten louder. That may not sound like something big, but for her, it means she can see behind her, which she hasnt been able to do since she was very little."

At the same time, theyre trying to be realistic about their expectations. More than anything else, they want Taleah and Colin to have the simple abilities to cough and swallow.

"Even if these medications do everything we hope, I dont think theyre going to cure Taleah," says Monica. "Nobody expects that theyre going to have her hop up and walk away."

 

Colin English, Taleahs 9-month-old brother, receives medication through a nasogastric tube.

Meanwhile, Swoboda and others emphasize that children with SMA shouldnt be given valproic acid or phenylbutyrate outside of a clinical trial.

Arthur Burghes goes a step further, cautioning that because of safety concerns and dosage problems, valproic acid and phenylbutyrate might not be "the best" drugs to use against SMA. Still, hes optimistic that ongoing screens will reveal drugs with more potent effects against SMA, and that, at the very least, Swobodas work will provide guidance for testing these drugs.

"Valproic acid and phenylbutyrate are just the tip of the iceberg," he says. "I think theres a whole generation of compounds coming behind them."

 

The Hunt for SMA Drugs The existence of a backup gene for SMN1 the gene defective in SMA has given scientists a unique opportunity to treat the disease. Instead of using gene therapy to replace SMN1, many are working to develop drugs that can increase needed SMN protein production from the backup gene, SMN2. "Were trying to play on the chance that nature gave us. Its a much easier strategy than trying to put the gene back into the spinal cord," says Kathryn Swoboda, whos conducting clinical trials of two candidate SMA drugs valproic acid and phenylbutyrate. Still, the road to finding these drugs has been a long, winding one. The first step was to determine what the SMN protein does, and why very little of it comes from SMN2. Next, scientists had to design methods to find chemicals capable of boosting SMN2. And finally, they had to identify chemicals with the most potent effects on SMN2, a process thats still ongoing. A Strategy Takes Shape Scientists now know that the SMN protein helps edit RNA, strands of chemical letters that serve as the intermediate between DNA and proteins. Like a long ticker tape, a strand of RNA is filled with important bits of information called exons, interrupted by information of less certain meaning called introns. In a process called splicing, SMN and other proteins remove the introns and paste the exons together. When the process works, a single RNA can "code" for multiple proteins with different functions. But a mistake can lead to the production of incomplete, nonfunctional proteins. Ironically, this same process accounts for the functional difference between SMN1 and SMN2. The two genes and their RNAs are nearly identical, except in an essential region called exon 7. SMN1 contains a signal that tells splicing proteins to leave exon 7 intact, but SMN2 lacks the signal. As a result, SMN2 mostly gives rise to a short version of the SMN protein that lacks exon 7, along with essential SMN functions. Armed with these insights, scientists began to think about finding a drug that would stimulate SMN2 to make more of the full-length SMN protein normally made by the SMN1 gene. In 2000, Elliot Androphy at New England Medical Center and Tufts University School of Medicine in Boston, and Christian Lorson, then at Arizona State University in Tempe, developed one of the first methods of screening for drugs with this activity. (See "Fast-Track Pharmacy," August 2001.) Both these scientists received MDA support. Using a similar approach, the biotech company Aurora Biosciences began screening over half a million candidate SMA drugs in 2001. The screen has since been turned over to Medichem, a subsidiary of the Icelandic gene-hunting company deCODE. 1. Most people have SMN1 and SMN2 genes. Full-length SMN protein is the main product of the SMN1 gene, while short SMN protein is the main product of the SMN2 gene, but both types of protein can be made from either gene. 2. In SMA, the SNM1 gene is missing or nonfunctional, vastly decreasing the amount of full-length SMN protein. However, the remaining SMN2 genes still produce some full-length protein. 3. The goal of drug treatment is to increase the amount of full-length SMN protein produced from the SMN2 gene's instructions. The First Hits Medichems screen already has revealed that a class of drugs called histone deacetylase (HDAC) inhibitors might be useful against SMA. These drugs are able to uncoil tightly woven strands of DNA, allowing dormant genes to become active. One drug in this class is valproic acid, which is approved by the Food and Drug Administration (FDA) for treating mood disorders and epilepsy. Given its HDAC activity and its history of medical use, Kenneth Fischbeck and colleagues at the National Institutes of Health (NIH) in Bethesda, Md., thought valproic acid might be a good candidate drug for SMA. When they tested the drug on skin cells derived from people with SMA, it increased the amount of full-length RNA and protein derived from SMN2, with each one reaching normal levels after three days of treatment at the highest dose tested. A group in Germany achieved similar results. "Its exciting, of course, that this class of drugs would have an effect," says Charlotte Sumner, the lead researcher on the project. Still, shes cautious about the results since its not clear that the high doses she used on the cells could be achieved in people. Meanwhile, Taiwanese scientists have reported promising results with butyrates, a class of compounds with a history of use against the blood disease beta-thalassemia. That disease is caused by a deficiency of hemoglobin, a protein that enables red blood cells to carry oxygen. At birth, blood cells make one version of the protein, called fetal hemoglobin, but some time after the first year of life, they switch to making adult hemoglobin. The goal of therapy in beta-thalassemia is to return to patients cells the capability of making fetal hemoglobin, since they cant make the adult form. In pilot trials, butyrates have been shown to reactivate fetal hemoglobin production and reduce the need for blood transfusions. Like valproic acid, butyrate compounds increased the level of full-length SMN RNA and protein when tested on cells from people with SMA. "That evidence led us to do the trials [of valproic acid and phenylbutyrate]," Swoboda says. "If we can increase the levels of SMN proteins in patients, we might increase the number of surviving motor neurons or make the remaining ones healthier." Meanwhile, Medichem and other investigators continue the search for other SMA drugs.