The molecular pathogenesis of Friedreich ataxia

Clinical Neuroscience Research 1 (2001) 127±133

www.elsevier.nl/locate/clires

The molecular pathogenesis of Friedreich ataxia HeÂleÁne Puccio*, Michel Kúnig Institut de GeÂneÂtique et de Biologie MoleÂculaire et Cellulaire (CNRS/INSERM/ULP), 1 Rue Laurent Fries BP163, 67404 Illkirch, C.U. de Strasbourg, France

Abstract Friedreich ataxia, the most frequent cause of recessive ataxia is due in most cases to a homozygous intronic expansion resulting in the loss of function of frataxin. Frataxin is a mitochondrial protein conserved through evolution. Yeast knock-out models and histological data from patients heart autopsies have shown that frataxin defect causes mitochondrial iron accumulation. Biochemical data from patients heart biopsies or autopsies have revealed a speci®c de®ciency in the activities of aconitases and of mitochondrial iron±sulfur proteins. These results suggest that frataxin may play a role either in mitochondrial iron transport or in iron±sulfur cluster assembly or transport. Iron abnormalities suggest a pathogenic mechanism involving free radicals production and oxidative stress, a process that might be sensitive to anti-oxidant therapies. q 2001 Association for Research in Nervous and Mental Disease. All rights reserved. Keywords: Friedreich ataxia; Molecular pathogenesis; Iron±sulfur proteins

1. Introduction

2. Reported ®ndings

Friedreich ataxia (FRDA) is the most frequent hereditary ataxia with an estimated incidence, in Caucasians, of 1 in 30 000 [1,2]. FRDA is an autosomal recessive neurodegenerative disease characterized by progressive gait and limb ataxia, dysarthria, lower limbs are¯exia, decreased vibration sense, muscle weakness of the legs and positive extensor plantar response [3,4]. Non-neurological signs include hypertrophic cardiomyopathy [5,6] and increased incidence of diabetes mellitus [7]. The onset of symptoms usually occurs before the age of 25, and typically around puberty. The ®rst pathologic changes occur in the dorsal root ganglia (DRG), with the loss of large sensory neurons, followed by neuron degeneration in Clarke's and posterior columns, pyramidal and spinocerebellar tracts of the spinal cord [8]. Mild degenerative changes are also observed in the medulla, cerebellum, and pons. The present review describes how the discovery of the Friedreich ataxia gene and of its major mutation, a large GAA trinucleotide expansion, lead to the understanding of the Friedreich ataxia pathology as a disturbance of mitochondrial iron homeostasis and to new therapeutic prospects.

2.1. FRDA gene

* Corresponding author. Tel.: 133-3-8865-3399; fax: 133-3-8865-3246. E-mail address: [email protected] (H. Puccio).

The human FRDA gene, X25, positionally cloned in 1996, is localized on 9q13 and is composed of seven exons spanning 80 kb of genomic DNA [9]. Northern blot analysis in human and mouse, RNAse protection assay and cDNA cloning indicate that the 1.3 kb major transcript is made from ®ve exons (1±5a). This transcript encodes a protein of 210 amino acids called frataxin. A very minor alternative transcript contains exon 5b instead of 5a, followed by the non-coding exon 6. The functional signi®cance of this transcript, if any, is still uncertain. The FRDA gene demonstrates tissue-speci®c and developmentally controlled expression by Northern blot (in human and mouse tissues) and in situ hybridization (in mouse tissues only) [9±11]. The expression partially correlates with the main sites of pathology of the disease. DRGs, where the sensory cell bodies are located, are the major sites of expression in the nervous system, from embryonic day 12 until adult life. Deep sensory neuropathy and degeneration of the posterior columns therefore appear as a direct consequence of reduced frataxin level in these structures. Expression in the spinal cord is comparatively much lower, suggesting that degeneration of the spinocerebellar tracts and of the Clarke's columns (containing, respectively, the axons and the cell bodies of secondary neurons projecting to the cerebellum) might be secondary to degeneration of the DRG neurons. Signi®cant frataxin expression is also

1566-2772/01/$ - see front matter q 2001 Association for Research in Nervous and Mental Disease. All rights reserved. PII: S 1566-277 2(00)00012-8

128

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

observed in the granular layer of the cerebellum. Degeneration of the motor corticospinal (pyramidal) tracts might correlate with frataxin expression in mature cells of the developing forebrain [10], though it was not detected in mouse adult cerebral cortex [11]. Expression in mouse brain is mostly restricted to the periventricular zone in embryos and to the corresponding ependymal layer in adults [11]. The frataxin gene is also expressed in non-neuronal tissues, such as heart and pancreas, which may account for hypertrophic cardiomyopathy and the increased incidence of diabetes observed in FRDA patients [9,11]. Frataxin is also prominently expressed in tissues apparently not affected by the disease, such as liver, muscle, thymus and brown fat. All tissues highly expressing frataxin are rich in mitochondria, with brown fat, present in new-borns, being particularly rich [11]. The difference between nonaffected and affected tissues may lie in the non-dividing nature of the latter (neurons, cardiocytes and b cells of the pancreas), implying that cells are not replaced when they die. Also, neurons and cardiocytes have an exclusive aerobic metabolism, making them more sensitive to mitochondrial defect. However, further factors must come into play to explain the speci®c vulnerability of speci®c cell types, (e.g. the affected sensory system vs. the spared motor system). 2.2. FRDA mutations FRDA is most commonly caused by a large GAA triplet repeat expansion within the ®rst intron of the gene encoding frataxin [9]. Ninety-six percent of patients are homozygous for GAA trinucleotide repeat expansions while the remaining 4% of cases are compound heterozygotes for a GAA expansion and a point mutation within the coding region of the gene [9,12]. Truncating and missense mutations are equally represented, although missense mutations have only been found in the 2nd half of the protein, suggesting it is an important functional domain. In most cases, point mutations are found in clinically typical FRDA patients with few exceptions. The G130V missense mutation always appears associated with an atypical presentation (retained knee re¯exes, often brisk, absence of dysarthria, moderate ataxia, spastic gait, and slow progression) [12±14]. The G130V mutation is one of the most frequent frataxin point mutation and arose from a common founder event [15]. Other mis-sense mutations each represented by a single case (L106S [16], D122Y [12], R16SC and L182F [14]) also seem to be associated with mild presentation. The clinical equivalence between the GAA intronic expansion and the truncating mutations suggests that the expansion acts by loss of function on frataxin. Indeed, reverse transcription polymerase chain reaction (RT-PCR) and RNAse protection experiments revealed that frataxin mRNA levels are markedly decreased in comparison to controls or unrelated ataxias [9,17,18]. RNAse protection and in vitro transcription experiments suggest that the

expansion acts at the transcriptional level, rather than interfere with the splicing of intron 1 [18,19]. Experiments using in vitro and in vivo expression systems revealed that the GAA repeat interferes with transcription in an orientation and length-dependent manner [18,19]. The molecular basis for these results was proposed to be the formation of a nonB DNA conformation, probably a triple helical structure. Indeed, Sakamoto et al. [20], recently described a novel DNA structure, `sticky DNA', for long tracts of GAA from the frataxin gene. Sticky DNA is formed by the association of two purine-purine-pyrimidime (R-R-Y) triple helical structure under the in¯uence of negative supercoiling. Therefore, the formation of sticky DNA in a FRDA patient would be the mechanism by which the expansion suppresses gene expression, resulting in reduced levels of frataxin. The direct involvement of the GAA expansion as the cause of FRDA is demonstrated by the very signi®cant inverse correlation between the size of the smaller of the two expansions and the age of onset, the severity of the disease and the risk occurrence of other signs such as cardiomyopathy, scoliosis and diabetes [21±26]. These observations suggest that some frataxin is produced from alleles carrying smaller expansions, in a length dependent manner. This was con®rmed using a monoclonal antibody directed against frataxin [27] and is in agreement with the transcript analyses. The smallest pathological expansions are in the range of 90±110 repeats, and are found in patients with late onset and atypical presentation of the disease [2,22] (one exception described [1] is most likely due to the presence of an unstable premutation in one of the parents). As a consequence, the molecular de®nition of Friedreich ataxia based on the presence of the GAA expansion mutation, is broader than the previous, clinically based de®nition. Patients with small expansions (,400 repeats) often show onset of the symptoms after age 25 or have retained tendon re¯exes [21±25], features previously considered as exclusion criteria [4]. Detection of the expansion mutation thus provides a most useful diagnostic test. The length of expansion has however little value for individual prognosis, given the large scattering of points along the correlation curve. 2.3. Frataxin mitochondrial localisation Frataxin is a 210 amino acid protein showing no similarity with protein domains of known function. Therefore, the function of frataxin cannot be inferred from its amino acid sequence. The protein however shows a striking degree of evolutionary conservation among eukaryotes, particularly in a stretch of 27 amino acids encoded by exons 4 and 5a. A ®rst suggestion that frataxin could be a mitochondrial protein came from phylogenetic studies. Sequence comparisons showed the presence of more distant homologues in gram negative (g purple), but not in gram positive bacteria [28]. This suggests that the frataxin gene might be derived

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

from the bacterial precursor of the mitochondrial genome which shares phylogenetic ancestry with gram negative bacteria, and then underwent transfer to the nuclear genome. Furthermore, computer analysis predicted a mitochondrial targeting signal at the N-terminus of yeast and mouse frataxin, a domain that is absent in the bacterial homologues [11]. Human and yeast frataxin were directly demonstrated to be mitochondrial proteins by epitope tagging experiments and colocalization with well established mitochondrial markers [11,29±31]. Mitochondrial localisation of endogenous frataxin was demonstrated using speci®c monoclonal antibodies. Immunoelectron microscopy results indicate that frataxin, that has no hydrophobic transmembrane segment, is nevertheless associated with mitochondrial membranes and crests [27]. The N-terminal mitochondrial targeting sequence of both mammalian and yeast frataxin is proteolytically removed in a two-step cleavage resulting in an 18 kDa mature protein [27,32,33]. Both cleavage steps involve the mitochondrial

129

processing peptidase (MPP), a dimeric protease whose bsubunit binds to the N-terminus of frataxin [32,33]. A con¯icting report suggesting that maturation of mammalian frataxin is a single step processing event by MPP [34] failed to demonstrate that the single step product is indeed mature frataxin and might well correspond to the intermediate product, since the second step cleavage is very inef®cient in vitro [33]. Koutnikova et al. [32], suggested that frataxin binding and cleavage by MPP appear partially affected by diseasecausing missense mutations in its C-terminal moiety (G130V and I154F), and that maturation of the I154F mutant was reduced compared to wild type frataxin in an in vivo overexpression system, while Gordon et al. [34], clearly found no evidence of reduced ®rst step cleavage for the I154F mutant. Further investigations are needed to determine whether the disease causing missense mutations affect the maturation, and if so, which step and whether this relates to the pathological process in FRDA patients.

Fig. 1. Molecular pathology in Friedreich ataxia. Frataxin, associated with mitochondrial membranes regulates iron homeostasis in this organelle. Frataxin disruption results in inactivation of mitochondrial iron±sulfur proteins (contained in complexes I±III of the oxidative phosphorylation pathway and mitochondrial aconitase) and in mitochondrial iron accumulation which leads to production of free radicals (OH z and O2 z). Frataxin disruption also results in cytosolic iron±sulfur inactivation, as evidenced by cytosolic aconitase transformation into iron-responsible element binding protein (IRE-BP) by loss of the iron±sulfur cluster. ABC7, the human homologue of the yeast ATM1 is thought to be involved in export of iron-sulfur clusters.

130

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

2.4. Mitochondrial iron accumulation Yeast as a model organism proved to be an invaluable system for unraveling frataxin function within the mitochondria. Three independent groups observed that deletion of the yeast frataxin gene, yeast frataxin homolog 1 (YFH1), results in impaired growth on glycerol, a non-fermentable source of carbon, accumulation of mitochondria de®cient rho-clones (deletion of mitochondrial DNA), and reduced respiration [11,29,30,35]. Moreover the mutant yeast showed a higher sensitivity to oxidative agents such as hydrogen peroxide, iron, and copper than the wild type strains [29,35]. The yeast frataxin gene was independently isolated as a multicopy suppressor able to rescue a yeast mutant strain unable to grow on iron-limited medium [29]. Measurements of iron content revealed that mitochondrial iron is 10-fold higher in the DYFH1 mutants than in wild type yeast, while total cellular iron concentration is doubled [29,35]. The fact that the high af®nity iron import system is constitutively turned on in the DYFH1 mutants suggests that cytosolic iron is low [29]. Iron is a well known catalyst of free radicals, and excess iron in mitochondria most likely explains the increase sensitivity to hydrogen peroxide and the mitochondrial dysfunction through irreversible oxidative damages (Fig. 1). If the function of human frataxin is similar to that of the yeast protein, this would suggest that iron accumulates in mitochondria of Friedreich ataxia patients, and could result in hypersensitivity to oxidative stress, as a consequence of the Fenton reaction (Fe 21 catalyzed production of hydroxyl radical). Indeed, iron deposits are consistently observed on autopsy in some heart myo®brils of FRDA patients and sometimes also observed in liver and spleen [36,37]. Cardiomyopathy in FRDA patients could thus be a result of iron overload (as in thalassemia or in hemochromatosis) or might re¯ect a selective sensitivity of heart mitochondria to frataxin de®ciency. Recent magnetic resonance imaging data indicate that iron also accumulates in the dentate nucleus, an affected cerebellar structure [38]. Total iron content of FRDA ®broblasts, an unaffected tissue in this disease, is within normal range with a minimal mitochondrial iron increase just at the limit of signi®cance [39]. 2.5. Iron±sulfur cluster protein de®ciency RoÈtig et al. [40], found selective de®ciencies of the respiratory chain complexes I±III and of both mitochondrial and cytosolic aconitases activities in the heart biopsy of two patients. Other respiratory chain complexes and Krebs cycle activities were normal. Aconitase activity de®ciency was profound ( < 10-fold reduction) while de®ciencies in the activities of complexes I±III were signi®cant only when their ratios to other mitochondrial activities, such as complex IV, were calculated. No de®ciency was found in the muscle, ®broblasts, nor lymphocytes of the same patients. All the de®cient enzymes and complexes contain

iron±sulfur (Fe±S) clusters in their active sites. Fe±S proteins are remarkably sensitive to free radicals, and their inactivation further suggests oxidative stress in FRDA affected tissues. De®ciency restricted to the Fe±S proteins has not been found in 60 biopsies of patients with cardiomyopathy, several of which had mitochondrial DNA mutation [40], indicating that this inactivation is speci®cally associated with the depletion of frataxin. These results have recently been con®rmed on autopsy material of nine patients, although the de®ciencies of complexes I±III were more pronounced [36], possibly due to a longer disease duration. Fe±S protein de®ciency was not found in autopsy DRGs of two patients analyzed, presumably because the large sensory neurons, the primary target cells in FRDA, are entirely replaced by gliosis at the time of death. The yeast YFH1 mutants have a generalized mitochondrial dysfunction, including Fe-S protein de®ciency as well as complex IV, and to a lesser extent complex V, de®ciencies [35,40], presumably because iron accumulation is higher in yeast mutants than in patient hearts. Is the Fe±S protein de®ciency a cause or a consequence of mitochondrial iron accumulation? Recent studies in the yeast suggest that frataxin mediates mitochondrial iron ef¯ux [41], and demonstrate that the presence of iron-chelator in the culture media restores normal intramitochondrial iron levels and normal oxidative respiration in DYFH1 yeast cells [42]. However, in the latter experimental conditions the activity of aconitase is not fully restored [42] suggesting that the reduction in the activity of the respiratory chain complexes is a consequence of mitochondrial iron accumulation, while the reduced aconitase activity is directly linked in part to frataxin de®ciency. Mitochondrial aconitase de®ciency therefore does not seem to be a mere consequence of mitochondrial dysfunction. In addition, the inactivation of cytosolic aconitase in FRDA patients (yeast do not have cytosolic aconitase), suggests that a general iron±sulfur defect is the underlying mechanism. The loss of cytosolic aconitase activity observed in FRDA might also re¯ect a decrease of cytosolic iron content, since cytosolic aconitase is converted to iron-responsive element binding protein (IRE-BP1) in response to low iron concentration [43] (Fig. 1). Two other yeast mutants that speci®cally accumulate iron in the mitochondria are worth a mention here. The ATM1 gene encodes a seven transmembrane ABC adenosine triphosphate (ATP)ase (a putative ABC transporter), that when mutated causes mitochondrial iron accumulation at 20-fold the normal level [44]. ATM1 seems to be involved in export of iron±sulfur clusters, which are assembled in the mitochondria, since cytosolic iron±sulfur enzymes are inactivated before the mitochondrial ones and before mitochondrial iron accumulation [45] (Fig. 1). Attempts to identify a direct functional relation between frataxin and ATM1 have failed so far. A conservative missense mutation in ABC7, the human homologue of the ATM1 gene, was found to cause a rare X-linked recessive disease, associating spinocerebellar ataxia and anaemia [46].

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

Mutants of the Ssq1 gene, encoding a low abundance mitochondrial heat shock 70 protein (mtHsp70), accumulates mitochondrial iron at concentration up to 40-fold the normal level, depending on extracellular iron concentration [47]. MtHsp70 proteins act as chaperones that pull imported peptides into the mitochondrial matrix, prior to their cleavage by MPP [48]. Ssq1 mutants show partially altered frataxin second step maturation cleavage. The dramatic iron accumulation of Ssq1 mutants suggest that Ssq1 acts on the import of a cascade of proteins involved in mitochondrial iron homeostasis, a pathway that may be altered in YFH1 mutants as well. Studies of the biochemical defect in the yeast mutant should reveal the role of the frataxin homolog in iron homeostasis and iron±sulfur cluster biogenesis. Very recent in vitro biochemical studies suggest that yeast frataxin might instead act as a mitochondrial iron-storage protein similar to cytoplasmic ferritin [49]. Further biochemical analysis of appropriate target tissues may be dif®cult in man, and a mouse model of FA (by knockout of the frataxin gene) should be extremely valuable to understand the pathological consequences of frataxin de®ciency in mammalian cells and tissues. 2.6. Therapeutic advances Based on the recent discoveries on the potential function of frataxin, and more speci®cally on the consequences of frataxin reduction, therapeutic advances can be envisioned. Whether the mitochondrial iron accumulation is a primary or a secondary effect of frataxin de®ciency, all data suggests that intracellular iron imbalance leading to oxidative stress is involved in the pathogenesis of FRDA. This lead to initial enthusiasm for iron chelators, such as desferrioxamine, as therapeutic reagent for treating the disease. However, desferrioxamine has a signi®cant side effect pro®le and its effect on individuals who do not have a generalized iron overload is not well studied. FRDA patients have normal serum iron and ferritin levels [50]. Rustin et al. [51], have recently shown, using an in vitro experimental system which mimics the abnormalities observed in heart biopsies from patients, that ferrous iron (Fe 21), but not ferric iron (Fe 31), causes peroxidation of lipids and inactivation of complex II without affecting aconitase activity. The addition of an iron chelator such as desferrioxamine in the extracts protects complex II and lipids from Fe 21 oxidation but causes a marked reduction in total aconitase activity [51], presumably by displacing the ferrous iron from membranes to the soluble fraction. These results suggest that caution should be taken when using iron chelators as a therapeutic agent since it could displace rather than protect against Fe 21-mediated toxicity. Since it has been proposed that reducing the load of free radicals will slow the progression of the disease, antioxidants which are usually devoid of side-effects can be considered as potential therapeutic reagents. Rustin et al. [51,52]

131

have shown by in vitro experiments that ascorbic acid, a water-soluble antioxidant, is not protective since it reduces ferric iron to its ferrous toxic form. On the other hand, the antioxidant idebenone, a short chain analogue of coenzyme Q10, protects both membranes and respiratory chain enzymes against iron-induced injury without causing a reduction in aconitase activity. In addition, the short chain of idebenone allows it to cross membranes readily (including the blood±brain barrier). Preliminary results on the cardiomyopathy of three FRDA patients are promising and warrants to establish a large clinical trial using idebenone. An ef®cient method for the evaluation of potential treatment ef®cacy is a major prerequisite for such clinical trial. Lodi et al. [53], have recently reported in vivo evidence of impaired mitochondrial respiration in skeletal muscle of FRDA patients using phosphorus magnetic resonance spectroscopy ( 31P-magnetic resonance spectroscopy (MRS)). Moreover, the observed decreased oxidative activity shows some correlation with the size of the GAA expansion. These ®nding contribute to the growing body of evidence that treatments aimed to enhance mitochondrial function and reduce toxic radical production are rational and may be a realistic hope for the patients and their families. 2.7. Generation of animal models Yeast knock-out models, and histological and biochemical data from patient heart biopsies or autopsies indicate that the frataxin defect causes a speci®c iron±sulfur protein de®ciency and mitochondrial iron accumulation leading to the pathological changes. Affected human tissues are rarely available to further examine this hypothesis. To study the mechanism of the disease, CosseÂe et al. [54], generated a mouse model by deletion of exon 4 leading to inactivation of the FRDA gene product. Homozygous deletions of frataxin cause embryonic lethality a few days after implantation, demonstrating an important role for frataxin during early development. These results suggest that the milder phenotype in humans is due to residual frataxin expression associated with the expansion mutations. Surprisingly, in the frataxin knock-out mouse, no iron accumulation was observed during embryonic resorption, suggesting that cell death could be due to a mechanism independent of iron accumulation [54]. To circumvent embryonic lethality, conditional knock-out models are being generated which will allow to induce nullisomy for frataxin in a tissue controlled or inducible manner and to establish a cell model for further analysis of the mechanism of the disease. 3. Conclusion Friedreich ataxia is a remarkable disease. The major mutation, a large intronic GAA repeat expansion is the most frequent trinucleotide repeat expansion presently known in the human population. It arises, probably recur-

132

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

rently, from alleles with pure GAA repeat at the upper limit of the normal range, and has spread unselected in healthy carriers, up to a carrier frequency of 1 in 85 individuals in the Caucasian population. Initial studies of frataxin and of its yeast homolog have revealed that Friedreich ataxia is a mitochondrial disease due to mutations in a gene from the nuclear genome. The fact that frataxin de®ciency results in alteration of mitochondrial iron homeostasis strongly suggests that oxidative stress is underlying the pathology of the disease, since iron is a well established catalyst of free radical production from hydrogen peroxide, a byproduct of the mitochondrial oxidative phosphorylation pathway. Friedreich ataxia may therefore serve as a paradigm for the growing list of neurodegenerative diseases caused by free radicals toxicity, such as ataxias due to vitamin E de®ciency [55] and familial amyotrophic lateral sclerosis due to SOD mutations [56]. Many questions remain unsolved and particularly the cause for the relatively late onset and slow progression of the disease, as well as the speci®city of neurodegeneration, while disruption of the corresponding gene in yeast results in a dramatic and rapid phenotype. We have found that an homozygous frame-shift deletion in the mouse, which would be the equivalent of the yeast knock-outs, causes early embryonic lethality [54]. This suggests that even the largest GAA expansions in FRDA patients do not extinguish totally frataxin expression, allowing the production of residual amount of frataxin compatible with cellular survival. It is also possible that cell types requiring frataxin during embryonic development select precursors with a shorter expansion mutation, allowing for a higher residual frataxin level. Lastly, mammalian cells which cannot survive in the absence of oxidative phosphorylation, unlike yeast cells, may have developed numerous protective pathways against free radicals toxicity, vitamin E being one of these, that can compensate in part for frataxin loss of function. The next challenge is to understand frataxin function and how its loss results in iron accumulation and iron±sulfur protein de®ciency in mitochondria of some, but probably not all, cell types or tissues. This should shed light on a previously unsuspected regulatory process involved in the ®ne tuning of intracellular iron homeostasis, and should provide solid bases for designing eagerly awaited therapeutic approaches for Friedreich ataxia. Acknowledgements We wish to thank our colleagues and collaborators V. Campuzano, M. CosseÂe, H. Koutnikova, J.-L. Mandel, L. Montermini, M. Pandolfo, F. Foury, A. RoÈtig, P. Rustin, A. Brice and A. Durr for their important contribution to the molecular unraveling of Friedreich ataxia. Research in our laboratory is supported by the Centre National de la Recherche Scienti®que, the Institut National de la Sante et de la Recherche MeÂdicale, and the HoÃpitaux Universitaires

de Strasbourg, as well as by grants and fellowships form the Association FrancËaise contre les Myopathies and the Human Frontier Science Program. References [1] Epplen C, Epplen JT, Frank G, Miterski B, Santos EJ, Schols L. Differential stability of the (GAA)n tract in the Friedreich ataxia (STM7) gene. Hum Genet 1997;99:834±836. [2] Cossee M, Schmitt M, Campuzano V, et al. Evolution of the Friedreich's ataxia trinucleotide repeat expansion: founder effect and premutations. Proc Natl Acad Sci USA 1997;94:7452±7457. [3] Geoffroy G, Barbeau A, Breton G, et al. Clinical description and roentgenologic evaluation of patients with Friedreich's ataxia. Can J Neurol Sci 1976;3:279±286. [4] Harding AE. Friedreich's ataxia: a clinical and genetic study of 90 families with an analysis of early diagnostic criteria and intrafamilial clustering of clinical features. Brain 1981;104:589±620. [5] Harding AE, Hewer RL. The heart disease of Friedreich's ataxia: a clinical and electrocardiographic study of 115 patients, with an analysis of serial electrocardiographic changes in 30 cases. Q J Med 1983;52:489±502. [6] Pentland B, Fox KA. The heart in Friedreich's ataxia. J Neurol Neurosurg Psychol 1983;46:1138±1142. [7] Finocchiaro G, Baio G, Micossi P, Pozza G, di Donato S. Glucose metabolism alterations in Friedreich's ataxia. Neurology 1988;38:1292±1296. [8] Oppenheimer DR. Brain lesions in Friedreich's ataxia. Can J Neurol Sci 1979;6:173±176. [9] Campuzano V, Montermini L, Molto MD, et al. Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion (see comments). Science 1996;271:1423±1427. [10] Jiralerspong S, Liu Y, Montermini L, Stifani S, Pandolfo M. Frataxin shows developmentally regulated tissue-speci®c expression in the mouse embryo. Neurobiol Dis 1997;4:103±113. [11] Koutnikova H, Campuzano V, Foury F, Dolle P, Cazzalini O, Kúnig M. Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin. Nat Genet 1997;16:345±351. [12] Cossee M, Durr A, Schmitt M, et al. Friedreich's ataxia: point mutations and clinical presentation of compound heterozygotes. Ann Neurol 1999;45:200±206. [13] Bidichandani SI, Ashizawa T, Patel PI. Atypical Friedreich ataxia caused by compound heterozygosity for a novel missense mutation and the GAA triplet-repeat expansion. Am J Hum Genet 1997;60:1251±1256. [14] Forrest SM, Knight M, Delatycki MB, et al. The correlation of clinical phenotype in Friedreich ataxia with the site of point mutations in the FRDA gene. Neurogenetics 1998;1:253. [15] Delatycki MB, Knight M, Kúnig M, Cossee M, Williamson R, Forrest SM. G13OV, a common FRDA point mutation, appears to have arisen from a common founder. Hum Genet 1999;105:343±346. [16] Bartolo C, Mendell JR, Prior TW. Identi®cation of a mis-sense mutation in a Friedreich's ataxia patient: implications for diagnosis and carrier studies. Am J Med Genet 1998;79:396±399. [17] Cossee M, Campuzano V, Koutnikova H, et al. Frataxin fracas (letter; comment). Nat Genet 1997;15:337±338. [18] Bidichandani SI, Ashizawa T, Patel PI, The GAA. triplet-repeat expansion in Friedreich ataxia interferes with transcription and may be associated with an unusual DNA structure. Am J Hum Genet 1998;62:111±121. [19] Ohshima K, Montermini L, Wells RD, Pandolfo M. Inhibitory effects of expanded GAA.TFC triplet repeats from intron I of the Friedreich ataxia gene on transcription and replication in vivo. J Biol Chem 1998;273:14588±14595. [20] Sakamoto N, Chastain PD, Pamiewski P, et al. Sticky DNA: self-

H. Puccio, M. Kúnig / Clinical Neuroscience Research 1 (2001) 127±133

[21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

[34] [35] [36] [37] [38]

association properties of long GAA.TTC repeats in R.R.Y triplex structures from Friedreich's ataxia. Mol Cell 1999;3:465±475. Montermini L, Richter A, Morgan K, et al. Phenotypic variability in Friedreich ataxia: role of the associated GAA triplet repeat expansion. Ann Neurol 1997;41:675±682. Dun A, Cossee M, Agid Y, et al. Clinical and genetic abnormalities in patients with Friedreich's ataxia. N Engl J Med 1996;335:1169±1175. Filla A, De Michele G, Cavalcanti F, et al. The relationship between trinucleotide (GAA) repeat length and clinical features in Friedreich ataxia. Am J Hum Genet 1996;59:554±560. Monros E, Molto MD, Martinez F, et al. Phenotype correlation and intergenerational dynamics of the Friedreich ataxia GAA trinucleotide repeat. Am J Hum Genet 1997;61:101±110. Lamont PJ, Davis MB, Wood NW. Identi®cation and sizing of the GAA trinucleotide repeat expansion of Friedreich's ataxia in 56 patients. Clinical and genetic correlates. Brain 1997;120:673±680. Isnard R, Kalotka H, Dun A, et al. Correlation between left ventricular hypertrophy and GAA trinucleotide repeat length in Friedreich's ataxia. Circulation 1997;95:2247±2249. Campuzano V, Montermini L, Lutz Y, et al. Frataxin is reduced in Friedreich ataxia patients and is associated with mitochondrial membranes. Hum Mol Genet 1997;6:1771±1780. Gibson TJ, Koonin EV, Musco G, Pastore A, Bork P. Friedreich's ataxia protein: phylogenetic evidence for mitochondrial dysfunction. Trends Neurosci 1996;19:465±468. Babcock M, de Silva D, Oaks R, et al. Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin. Science 1997;276:1709±1712. Wilson RB, Roof DM. Respiratory de®ciency due to loss of mitochondrial DNA in yeast lacking the frataxin homologue. Nat Genet 1997;16:352±357. Priller J, Scherzer CR, Faber PW, MacDonald ME, Young AB. Frataxin gene of Friedreich's ataxia is targeted to mitochondria. Ann Neurol 1997;42:265±269. Koutnikova H, Campuzano V, Kúnig M. Maturation of wild-type and mutated frataxin by the mitochondrial processing peptidase. Hum Mol Genet 1998;7:1485±1489. Branda SS, Cavadini P, Adamec J, Kalousek F, Taroni F, Isaya G. Yeast and human frataxin are processed to mature form in two sequential steps by the mitochondrial processing peptidase. J Biol Chem 1999;274:22763±22769. Gordon DM, Shi Q, Dancis A, Pain D. Maturation of frataxin within mammalian and yeast mitochondria: one-step processing by matrix processing peptidase. Hum Mol Genet 1999;8:2255±2262. Foury F, Cazzalini O. Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria. FEBS Lett 1997;411:373±377. Bradley JL, Blake JC, Chamberlain S, Thomas PK, Cooper JM, Schapira AH. Clinical, biochemical and molecular genetic correlations in Friedreich's ataxia. Hum Mol Genet 2000;9:275±282. Lamarche JB, Shapcott D, Cote M, Lemieux B. In: Lechtenberg R, editor. Handbook of cerebellar diseases, New York: Marcel Dekker, 1993. pp. 453±456. Waidvogel D, van Gelderen P, Hallett M. Increased iron in the dentate

[39]

[40] [41] [42] [43] [44] [45] [46]

[47] [48] [49] [50] [51] [52] [53]

[54] [55] [56]

133

nucleus of patients with Friedrich's ataxia. Ann Neurol 1999;46:123± 125. Delatyclu MB, Lamakans J, Brooks H, Pianese L, Monticelli A, Campanella G, Cocozza S Direct evidence that mitochondrial iron accumulation occurs in Friedreich ataxia. Ann Neurol 1999;45:673± 675. Rotig A, de Lonlay P, Chretien D, et al. Aconitase and mitochondrial iron-sulphur protein de®ciency in Friedreich ataxia. Nat Genet 1997;17:215±217. Radisky DC, Babcock MC, Kaplan J. The yeast frataxin homologue mediates mitochondrial iron ef¯ux. Evidence for a mitochondrial iron cycle. J Biol Chem 1999;274:4497±4499. Foury F. Low iron concentration and aconitase de®ciency in a yeast frataxin homologue de®cient strain. FEBS Lett 1999;456:281±284. Kaptain S, Downey WE, Tang C, et al. A regulated RNA binding protein also possesses aconitase activity. Proc Natl Acad Sci USA 1991;88:10109±10113. Kispal G, Csere P, Guiard B, Lill R. The ABC transporter Atm1p is required for mitochondrial iron homeostasis. FEBS Lett 1997;418:346±350. Kispal G, Csere P, Prohl C, Lill R. The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins. EMBO J 1999;18:3981±3989. Allikmets R, Raskind WH, Hutchinson A, Schueck ND, Dean M, Koeller DM. Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSAIA). Hum Mol Genet 1999;8:743±749. Knight SA, Sepuri NB, Pain D, Dancis A. Mt-Hsp70 homolog, Ssc2p, required for maturation of yeast frataxin and mitochondrial iron homeostasis. J Biol Chem 1998;273:18389±18393. Pfanner N. Mitochondrial import: crossing the aqueous intermembrane space. Curr Biol 1998;8:R262±R265. Adamec J, Rusnak F, Owen W, Naylor S, Benson L, Isaya G. Frataxin is an iron-storage protein. 1999 Wilson RB, Lynch UK, Fischbeck KH. Normal serum iron and ferritin concentrations in patients with Friedreich's ataxia. Ann Neurol 1999;44:132±134. Rustin P, von Kleist-Retzow JC, Chantrel-Groussard K, Sidi D, Munnich A, Rotig A. Effect of idebenone on cardiomyopathy in Friedreich's ataxia: a preliminary study. Lancet 1999;354:477±479. Rustin P, Munnich A, Rotig A. Quinone analogs prevent enzymes targeted in Friedreich ataxia from iron-induced injury in vitro. Biofactors 1999;9:247±251. Lodi R, Cooper JM, Bradley JL, Evans-Whipp T, Thorburn UK, Williamson R, Forrest SM De®cit of in vivo mitochondrial ATP production in patients with Friedreich ataxia (see comments). Proc Natl Acad Sci USA 1999;96:11492±11495. CosseÂe M, Puccio H, Gansmuller A, et al. Inactivation of the Friedreich ataxia mouse gene leads to early embryonic lethality without iron accumulation. Hum Mol Genet 2000:9(8)1219±9(8)1226. Muller D, Lloyd J, Wolff O. Vitamin E and neurological function. Lancet 1983;1:225±228. Sendtner M, Thoenen H. Oxidative stress and motorneuron disease. Curr Biol 1994;4:1036±1039.