A symposium on the hereditary ataxias.

A symposium on the hereditary ataxias was held on Sunday, Sept. 30, 2001, at the Hyatt Regency Hotel in Chicago, IL. It occurred in conjunction with the annual meetings of the American Neurological Association and the Society for Experimental Neuropathology. Nine speakers from three different countries gave formal presentations on many aspects of the recessive and dominant ataxias. The following is a summary:

Dr. Arnulf H. Koeppen (Albany, NY) summarized the history of Friedreich's ataxia and emphasized that Dr. Nikolaus Friedreich (1825-1882) had considerable insight into the disease that now bears his name, "Friedreich's ataxia." It is the most common form of recessive ataxia. Friedreich recognized that the disease was, in part, a developmental problem rather than a pure atrophy. This observation implies that the mutation in the frataxin gene causes problems in the unborn child though after birth, additional illness of the central nervous system and other organs continues. Friedreich was aware of the disease of the heart in his patients though he did not consider it a direct result of the disease. The disorder of the heart moved into focus in 1980 when Dr. Jacques B. Lamarche (Sherbrooke, QC, Canada) discovered an accumulation of iron. It is now widely accepted that iron excess in the heart is the direct result of a protein deficiency. The name derives from Friedreich's ataxia. Normally, frataxin is in the service of cellular iron metabolism. When absent, iron accumulates in the power packs of many cells, the mitochondria, where it injures several enzymes that provide energy.

Dr. Massimo Pandolfo (Brussels, Belgium, and Montreal, Canada), updated the participants on the frataxin gene, the mechanism of its mutation, and the worldwide epidemiology of Friedreich's ataxia. The gene was cloned in his laboratory in 1995, and the mutation is an overly long stretch of the nucleic acid bases guanine-adenine-adenine (GAA). This triplet occurs in a chain that is known as a "repeat", e.g., GAA-GAA-GAA, until 6-10 such triplets have been connected together. Patients with Friedreich's ataxia have more than 90 repeats in both genes that they have inherited from their parents. The number of repeats may exceed 1,500. Only occasionally do patients with Friedreich's ataxia have one expanded repeat and one point mutation. Dr. Pandolfo described how the overlong repeats interfere with normal development. The chemical structure of the abnormal genes prevents the proper messaging of the genetic information to the cell that makes frataxin. Friedreich's ataxia does not occur in all peoples of the world. It does not affect persons whose ethnic background is Sub-Saharan African, Amerindian, Chinese, Japanese or Southeast Asian. These observations imply that Friedreich's ataxia evolved over centuries as relatively short GAA repeats that gradually grew to a pathological length.

Dr. Robert B. Wilson (Philadelphia, PA) focused on the properties of frataxin and its detailed function in iron metabolism. He also reviewed the experimental evidence from other laboratories. It is remarkable that during early studies of the frataxin gene, researchers were helped greatly by observations in yeast. Yeast cells contain a homologue of frataxin called "yeast frataxin homologue" (abbreviated as YFH1p) which is essential for the growth of this microorganism. As in the heart of patients with Friedreich's ataxia, the mitochondria of YFH1p-deficient yeast become overloaded with iron, and the excess of the metal causes damage to the energy-providing enzymes of the cell. The detailed mechanism of how normal frataxin manages iron inside the mitochondrial power packs is the subject of intensive study. There is little question that frataxin-deficient yeast and the tissues of patients with Friedreich ataxia are energy-deficient. Recently, a non-invasive method called magnetic resonance spectroscopy has revealed lack of high-energy phosphate in the limbs of patients with Friedreich's ataxia. This observation is of greatest interest because the disease does not specifically attack skeletal muscles. Evidence of injury to enzymes and tissue membranes can now also be detected in the blood of some patients with Friedreich's ataxia. 'Recently, abnormal frataxin genes were introduced into mice ("transgenic" mice). Mice that completely lack the frataxin gene die in the womb and hence never reach postnatal life. However, the abnormal genes in transgenic mouse could be modified in such a manner that their offspring would indeed survive. In these experiments, frataxin deficit was also targeted to heart muscle or the central nervous system. The resulting abnormalities were quite similar to those in Friedreich patients and included the iron accumulation in the heart. Treatment trials with a drug called idebenone have focused on mitochondrial damage and repair. The drug strengthens the energy production by these cellular power packs but results have been somewhat variable. More information on the effectiveness of idebenone is expected.

Drs. Yves Robitaille and Andrea Richter (Montreal, QC, Canada) described clinical, pathological, and genetic findings in an autosomal recessive form of ataxia that is relatively frequent in Quebec: autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS). Their presentation included brief videos of patients with this illness, and the clinical features are ataxia with an onset at the age of 12-18 months, spasticity of the all limbs, and slurred speech. The examiner also finds a disturbance of eye movements, hyperactive reflexes, and Babinski signs (a response of the big toes to stimulation of the sole of the feet). Many patients also have abnormal eye grounds in that some of the nerve fibers are surrounded by myelin, the white substance of the brain. The disease affects the spinal cord, the cerebellum, and much of the gray matter of the brain itself. In some ways, the pathological findings in the spinal cord resemble Friedreich's ataxia but the disorder also injures the most prominent nerve cells in the cerebellum, the Purkinje cells. In Friedreich's ataxia, this nerve cell escapes. Also, the characteristic heart disease of patients with Friedreich's ataxia is not present in ARSACS. Dr. Richter discovered that the gene for ARSACS is located on the long arm of the 13th chromosome (which is also quite different from the involvement of the ninth chromosome in Friedreich's ataxia). The normal gene codes for a protein that was termed "sacsin," and ARSACS occurs when it is absent due to mutations in the genes inherited from both parents. The normal function of sacsin is not known but Dr. Richter reported that it is a very large protein. It may have a role in the structural integrity of brain cells. In contrast to Friedreich's ataxia, homologues in microorganisms are not known, providing little help in the search for sacsin function in the human nervous system.

Turning to the dominant ataxias, Dr. Koeppen gave an overview of the history of the spinocerebellar ataxias (SCA) ranging from the first publication by Menzel in 1891 and the probing analysis of cases by the famous French neurologist Pierre Marie in 1893 to the cloning of genes for SCA-1, SCA-2, Machado-Joseph disease/SCA-3, SCA-6, and SCA-7. For many years, the term "olivopontocerebellar atrophy" (OPCA) was thought to be equivalent to autosomal dominant ataxia, but there remains little justification for this term at the present time. Some patients have only a disease of the cerebellar cortex, while others have combined atrophy of the spinal cord and a substructure of the cerebellum, the dentate nucleus (e.g., Machado-Joseph disease/ SCA-3). Blood tests are available for several SCA's but physicians may be unable to select the correct test unless they are familiar with the clinical manifestations and the appearance on computed tomography and magnetic resonance images. Due to the high cost of these blood tests, it is advisable that only one or two be correctly selected. Several clinical findings in addition to ataxia suggest which test should be selected. If the patient has slow eye movements, SCA-2 and SCA-7 should be determined with priority. If the patient has poor vision due to atrophy of the retina, the first test should be for SCA-7. If the magnetic resonance image shows only cerebellar atrophy, blood analysis for SCA-1 and SCA-6 is recommended. Despite these advances in the laboratory diagnosis of SCA, all tests may remain inconclusive in some patients.

Dr. H. Brent Clark (Minneapolis, MN) described the clinical and pathological findings in SCA-l, and his observations in transgenic mice. The gene of SCA-1 was the first to be cloned from among the various other forms of SCA. The function of the normal protein, ataxin-1, remains unknown. In SCA-1, a cytosine-adenine-guanine (CAG) repeat on the 6th chromosome is abnormally expanded, and the result is the biosynthesis of an excessively long protein due to the chaining-together of a single amino acid, glutamine (amino acids are the molecular building blocks of proteins). Typically, the mutation causes loss of the Purkinje cells in the cerebellar cortex but structures of the brain stem and spinal cord may be affected to variable degree. An important observation is the presence of condensed protein in the nuclei of nerve cells, termed nuclear inclusion bodies. They occur in areas where nerve cells are lost but also in regions of the brain that are exempt from attack by SCA-1. Accordingly, their presence is not an absolute indication that the tissue is actually diseased. To overcome the limitations imposed by donated autopsy tissues of patients with SCA-1, Dr. Clark and his collaborators Dr. Harry Orr and Dr. Huda Zoghbi have developed transgenic mouse models. Mice that carry a mutated human gene (a "transgene") with the SCA-1 expansion develop neurological illness, atrophy of Purkinje cells, and nuclear inclusion bodies. If the mutated transgenic protein was prevented from entering the nuclei of neurons by further experimental manipulation, the mice did not become ill. If the mutated transgenic protein was altered experimentally so that it does not aggregate into nuclear inclusions, the mice still became ataxic. Ataxin-1 aggregation requires the presence of other proteins, among which one is often cited in SCA-1 and most other SCA: ubiquitin. When transgenic mice were developed that lacked ubiquitin but carried the abnormal human SCA-1 gene, the animals became ill but did not develop many inclusion bodies. An important conclusion of this work is that the disease-causing mechanism may be related to protein "trafficking" in a piggyback manner with one or more other proteins than to the isolated accumulation of only one abnormal protein.

Dr. Stefan M. Pulst (Los Angeles, CA) summarized current knowledge of SCA-2 that has many similarities to SCA-1, including a CAG repeat expansion. However, a different chromosome (number 12) is involved, and the mutated protein was named ataxin-2. As in ataxin-1, abnormal ataxin-2 contains overly long stretches of glutamine. In patients with very long CAG repeats, the disease tends to begin very early in life and to be quite severe. However, the correlation is not precise, and there is great variability in the age of onset and disease duration, irrespective of the length of the CAG expansion. Dr. Pulst illustrated that the formation of inclusion bodies is not required for the disease to become apparent in a patient. He and his collaborators also have developed a transgenic mouse model of SCA-2 in which the mutated ataxin-2 causes cerebellar atrophy.

Dr. Henry Paulson (Iowa City, IA) presented a summary of Machado-Joseph disease/SCA-3 which is now known to be the most common autosomal dominant ataxia. Similar to several other SCA's (including SCA-1, SCA-2, SCA-6, SCA-7 and SCA17), Machado-Joseph disease/SCA-3 is due to an expanded CAG repeat that codes for an elongated stretch of the amino acid glutamine in the disease protein known as ataxin-3. As in the other SCA's, the disease protein in Machado-Joseph disease/SCA-3 forms inclusion bodies in the nuclei of select nerve cells. The formation of inclusion bodies by mutant ataxin-3 involves several other proteins, and other parts of the disease protein influence the type of aggregation that occurs in the nucleus. The abnormal gene has been introduced into the fruit fly, the zebra fish, and cultured cells, to create various models of the disease. Each experimental model of Machado-Joseph disease/SCA-3 has its own uses and advantages. Taken together, this collection of models should help us understand the cause of the disease and identify potential therapies.

Dr. Tetsuo Ashizawa (Houston, TX) gave a comprehensive description of the remaining SCA that have now been numbered SCA-6 to SCA-17. Some of these (SCA-6, SCA-7 and SCA-17) resemble SCA-l, SCA-2 and SCA-3 in that abnormal proteins are being made in the brain, i.e., the CAG expansions are in the "coding region" of the gene. Interestingly, the normal product in SCA-6 is known from prior research work. It is a constituent of calcium channels in the brain. The protein in SCA-7 was termed ataxin-7 in analogy to SCA-1, SCA-2, and SCA-3 but its function is not yet known. The gene product in SCA-17 is known and has been termed the "TATA-binding protein." SCA-12 is a rare form of dominant ataxia that is caused by CAG expansion in another known gene. However, there is no gene product that contains overly long glutamine stretches. Dr. Ashizawa reported his personal experience with SCA-10. Patients with this disease have a unique combination of cerebellar ataxia and epilepsy. In contrast to other forms of SCA, this form is due to a large expansion of a pentanucleotide repeat (i.e., five bases, ATTCT where C stands for cytosine). Some of the SCA are known to be linked to specific chromosomes but their genes remain unknown. In these types of ataxia, blood tests on individual patients have no practical value. No single therapy or medication will benefit all patients with ataxia. The best strategy remains the identification of the gene and its gene product, followed by a rational search for a treatment.

Financial support for the symposium was provided by the following organizations: VA Healthcare Network of Upstate New York, Albany, NY; Neurochemical Research, Inc., Glenmont, NY; National Ataxia Foundation, Minneapolis, MN; Friedreich's Ataxia Research Alliance, Arlington, VA; Connecticut Neurological Foundation, New Haven, CT; and Department of Neurology, Albany Medical College, Albany, NY.

Arnulf Koeppen, MD
Albany, NY