---

Epilepsy

Genes may build the road in treatment

BY CHRISTOPER A. WALSH M.D., PH.D.

Epilepsy affects almost one percent of the population of the United States. It is a brain disorder causing unpredictable, uncontrolled seizures that can occur at any time or place.

Seizures result when the normal, tightly controlled electrical activity of the brain becomes excessive and disordered, interrupting normal awareness and normal activities. In the most dramatic and severe seizures (called "grand mal"), the patient loses consciousness and has wild movements of the limbs, followed by temporary suppression of conscious brain activity. The brain slowly regains its normal activity as the patient wakes up, unaware of what happened.

Epilepsy is the most common neurological disorder among young people. And because it produces problems with school, driving, and keeping a job, it has a huge economic cost (estimated by the government at $3.5 billion). It takes a social toll impossible to measure in disrupted lives for both epileptics and their families.

That cost is all the more frustrating because it seems as if it should be greatly reduced if only we had the right drugs to control the seizures. Many useful drugs exist for epilepsy, and several new ones have become available recently. But, for about one third of patients, seizures cannot be controlled by presently available medical therapy. Some epileptics can be helped by brain surgery, but thousands remain whose epilepsy cannot be controlled by any present- day treatments. So, the hunt continues for improved antiepileptic medications -- a search now leading onto genetic ground.

The genetic picture
Epilepsy can be caused by many factors that have nothing to do with genetics; head trauma, stroke, infections, tumors, and drug or alcohol abuse can all induce epilepsy. However, up to one half of all epilepsy has no other obvious causes, and there is increasing consensus that most of these cases have some relationship to inherited genes.

Recent rapid advances in the understanding of the human genome have begun to allow the identification of genes that predispose to epilepsy. Each one of these new genes seems to cause epilepsy by interrupting key processes in the normal function of the brain's neurons.

Some rare types of epilepsy seem to be almost completely caused by the action of epilepsy genes. In families where these genes are found, anyone who inherits the epilepsy gene (which usually means half of the children of an affected parent) will eventually develop epilepsy no matter what.

However, we are coming to understand that epilepsy may be genetically more "complex," like diabetes or cancer. In these complex genetic diseases, anyone theoretically can develop the disorder, as it can be caused by the interactions of many different genes. Depending upon which combination of genes people inherit, some people will be more susceptible to epilepsy than others, but the genes do not make epilepsy inevitable.

Inherited epilepsies usually require at least four complicated, unpronounceable medical words to capture their characteristic features.

Human Epliepsy Genes

Disorder	Chromosome	Gene Identified?
Benign familial neonatal convulsions (BFNC1)	20	no
Benign familial neonatal convulsions (BFNC2)	8	no
Juvenile myoclonic epilepsy (JME)	6	no
Idiopathic generalized epilepsy (IGE)	8	no
Partial epilepsy with auditory features (EPT)	10	no
Nocturnal frontal lobe epilepsy (ADNFLE)	20	yes
Unverricht-Lundborg disease (EPM1)	21	yes
Myoclonic epilepsy with ragged red fibers (MERRF)	mitochondrial	yes
Northern epilepsy syndrome (EPMR)	8	no
Ceroid lipofuscinosis, juvenile type (CLN3)	16	yes
Myonclonic epilepsy of Lafora (MELF)	6	no
Periventricular heterotopia (PH)	X	no
Double cortex (DC)	X	no

For example, some families have a remarkable epilepsy: Seizures occur in their newborn babies just during the first week of life, and then magically go away and usually never come back. The disorder is called benign (because they go away) familial neonatal (because they occur in newborns) convulsions -- BFNC for short -- and two different genes (BFNC1, BFCN2) can cause it. Other families have a disorder called autosomal dominant (transmitted directly from parent to child) nocturnal frontal lobe epilepsy, ADNFLE. This epilepsy causes bizarre types of movements because the seizures involve the frontal lobe; the seizures occur mostly or only just after the onset of sleep.

Researchers usually find epilepsy genes by studying families in which it is very clear that epilepsy is a genetic trait passed on from parent to child. Often such families have very rare and distinctive types of epilepsy. This makes it easier for researchers to be sure that different family members have the same kind of epilepsy and that it is caused by the action of a gene, rather than some other cause (such as drug abuse or head trauma).

For individuals who think epilepsy may run in their family, the Epilepsy Foundation of America has set up a website (www.efa.org/index/htm) that describes genetic studies in epilepsy and allows interested individuals to contact epilepsy researchers.

Another way that epilepsy genes can be found is by finding families with epilepsy in which slight abnormalities occur in the way the brain develops, since abnormalities in brain development have long been associated with seizures. The human brain can be imaged very precisely using Magnetic Resonance Imaging (MRI), and the inheritance of certain subtle malformations of the brain can then be analyzed by taking MRI pictures of everyone in a family.

Our group has used this method to localize two epilepsy genes, each one associated with subtle abnormalities of brain development that would never have been suspected before the advent of MRI imaging. The genes for the developmental epileptic disorders appear to be involved in communication between cells in the developing brain as well as in the adult brain; therefore, they may be especially central to normal brain function and thus among the most desirable targets for drug design.

Matched magnetic resonance images of normal (right, upper and lower panels) and two inherited malformation of the brain associated with seizures.
The "periventricular heterotopias" (left) are blobs of neurons that are in the wrong place (hetero-topia). They belong in the cortex, but instead are located along the venticles, the two slit-like dark structures deep in the brain (hence, peri-ventricular). The affected patient is a young woman with normal intelligence who is a college graduate, but who has intractable epilepsy.
In the double cortex disorder (left), there is a normal cerebral cortex, and then a "second cortex" below the first (seen as a gray band halfway between the outer surface of the brain and the dark ventricle). (Images courtesy of C. Walsh.)

Sorting through DNA
The 50,000 to 100,000 total genes present in our DNA are like individual entries in a vast encyclopedia; they collectively tell our cells what to do and when. Our 46 chromosomes are like the volumes of that encyclopedia: large DNA molecules each binding together hundreds to thousands of genes. Once a large family with inherited epilepsy is found, the tools of the international Human Genome Project can usually allow the "mapping" or localization of that gene to a place within one of the 46 chromosomes. Therefore, mapping an epilepsy gene basically means narrowing the search for the epilepsy gene down to about 500 to 1,000 genes in a smaller area of one individual chromosome.

Often, however, taking the last step to find that one disease-causing gene in a thousand takes longer than the initial mapping: It often requires painstakingly sifting through the DNA sequence of many genes to find the one that contains an error, or mutation. Finding a mutation can be like trying to find a single typographical error among a thousand encyclopedia entries, all of them written in a language that we do not yet understand! That often requires years of work and/or a stroke of good fortune.

The road from genes to drug design
Most epilepsy drugs currently on the market were discovered to have activity against seizures first, and only later did we come to understand a little bit about how or why they were effective. However, the ideal way to design epilepsy drugs would be first to understand the critical workings of the brain and the key steps that give rise to the seizures, and then make new drugs that specifically attack the key steps. Such drugs should be more specific, effective, and have fewer side effects.

The road that will lead to such drug development is beginning to be built, thanks to the rapid advances in identifying epilepsy genes. The power of genetics in opening up new avenues of investigation is unique. For example, some recently identified genes turn out to be involved in aspects of neuronal function that were previously not thought to be abnormal in epilepsy. Drug design can now be aimed at the neuronal functions mediated by these genes, without upsetting other systems. Therefore, each new epilepsy gene provides a potential new target for developing new drug therapies that may be more effective and have fewer side effects.

It is likely to take several more years to go from genes to successful epilepsy drugs: Only in 1994 were the first drugs released onto the market that were designed "rationally" -- that is, by exploiting our knowledge of neuronal function to target aspects of it critical to epilepsy. However, newly released drugs (protease inhibitors) that treat AIDS interfere with processes that were not even understood a few years ago, illustrating how far and how fast "rational" drug design has come and showing its potential for new treatments given adequate understanding of epileptic mechanisms. The further study of inherited human epilepsies should provide critical information for future drug design. *

Dr. Walsh is Senior Associate in Neurology, Beth Israel Deaconess Medical Center and Assistant Professor of Neurology at Harvard Medical School. (Cover photo courtesy of Uniphoto, Washington, DC)