150 years of Friedreich Ataxia: from its discovery to therapy

Abstract

In 1863, Nikolaus Friedreich (1825-1882), a German pathologist from Heidelberg, described a new spinal disease for the first time (Friedreich 1863a,b,c). However, it was only in 1876 that he had articulated the hereditary nature of the disorder (Koeppen 2013). It took a staggering 120 years to discover the genetic defect underlying Friedreich Ataxia (FRDA) (Campuzano et al. 1996). The identification of mutations in the gene encoding frataxin (FXN) initiated the rapid growth of a scientific field in which FRDA became a model disorder. Frataxin has been found to locate to mitochondria, control mitochondrial iron homoeostasis and, if deficient, lead to oxidative stress. Genetically, it is a recessively inherited disorder caused by an expansion mutation of GAA repeats in the first intron within the FXN gene, leading to a largely reduced expression of the frataxin protein. The discovery that frataxin expression can be modulated by epigenetic mechanisms has opened new research avenues and treatment strategies. Large networks have been established in Europe (European Friedreich Ataxia Consortium of Translational Studies, EFACTS), America (Collaborative Clinical Research Network, CCRN), and Australia to study cohorts of FRDA patients. The development of sensitive clinical scales and biomarkers to detect disease progression and effects of therapeutical interventions is of uttermost importance. In this issue, researchers, many of whom participate in the ‘European Friedreich Ataxia Consortium of Translational Studies (EFACTS)’, summarize the current status of FRDA research. Pierre Vankan (Vankan 2013) reports that the FRDA prevalence in Europe shows a southwest to northeast gradient, with a high prevalence of around 1 : 20 000 in northern Spain, southern and central France, and Ireland, and a low prevalence of 1 : 250 000 in Scandinavia, eastern Germany, Austria, and Russia. This observed distribution coincides with the gradient of the chromosomal R1b marker within Western Europe. Although FRDA and the R1b marker are not genetically linked, they may have co-existed in a hypothetical founder population. Because of this correlation, Pierre Vankan hypothesizes that the FRDA distribution in Europe is derived from Palaeolithic migrations out of the Franco-Cantabrian ice age refuge (Vankan 2013). The majority of FRDA cases are caused by an expansion mutation of (GAA)n trinucleotide repeats within the first intron of the FXN gene which encodes the mitochondrial protein frataxin. This leads to a silencing of the FXN gene, the subsequent lower generation of frataxin protein causing the clinical phenotype of FRDA (Cnop et al. 2013; Parkinson et al. 2013; Weidemann et al. 2013). To explain the silencing of FXN in FRDA, two models have been proposed. The first model implicates a transcriptional blockage created by the expanded (GAA)n repeats, which induce sticky triplex DNA structures. The second model implicates heterochromatin effects caused by expanded repeats. Over the recent years it has become clear that the Frataxin gene is heterochromatinized and is a target of epigenetic regulation (Yandim et al. 2013). However, a role of triplex DNA in triggering heterochromatinization is possible, reconciling the two models. From these data two lines of research on the pathophysiology of FRDA and potential therapeutic opportunities have evolved. The first line of research studies the physiological function of frataxin, the consequences of its loss and therapeutic strategies to complement frataxin in patients. The second line of research investigates the epigenetic modification of FXN, therapeutically antagonizing the aberrant silencing of the FXN gene. Human frataxin is imported into mitochondria and undergoes maturation by the mitochondrial processing peptidase. The crystal structure of frataxin has been studied extensively. Frataxin is a component of a multiprotein complex for iron–sulfur cluster biogenesis, which it allosterically activates (Pastore and Puccio 2013). A deficiency of frataxin leads to a deficiency in the generation of iron–sulfur cluster-containing proteins (including complex I, II, and III of the mitochondrial electron transport chain and aconitase), an increase of mitochondrial iron load, decrease of mitochondrial ATP generation, increase of the generation of reactive oxygen species, decrease of mitochondrial biogenesis, and induction of autophagy (González-Cabo and Palau 2013). To study the underlying pathophysiology, to identify therapeutic targets and test therapeutic interventions, cellular and animal models of FRDA were established. Perdomini and colleagues (Perdomini et al. 2013) describe in detail the advantages and disadvantages of the different models: (i) full Fxn deletion models (complete and tissue-specific deletion), (ii) models with residual frataxin expression, and (iii) FRDA patient-derived cells: fibroblasts, lymphoblasts, induced pluripotent stem cells (iPSC), and iPSC-derived cells, (iv) antisense, ribozyme, siRNA, and shRNA-based cell lines, and (v) humanized cell models with point mutations. In this review, it becomes clear that the perfect model does not exist and that the choice of the ‘right’ model depends on the scientific question that is raised. In studies focusing on the downstream consequences of frataxin deficiency, like testing drug treatment against oxidative stress and mitochondrial dysfunction (idebenone) or mitochondrial iron accumulation (deferiprone), and in those evaluating the effects of ectopic frataxin expression (e.g. by viral vectors), frataxin-deficient models were used. In contrast, studies testing for the epigenetic modulation of FXN to increase frataxin expression either used models carrying an expanded GAA repeat inserted into the mouse Fxn gene, or humanized models carrying a human mutant FXN gene. After extensive preclinical studies, several clinical trials with pharmaceutical interventions were conducted. Initial results with the antioxidants and mitochondrial enhancers idebenone and Coenzyme Q10 were promising, although larger phase III randomized, placebo-controlled trials failed to show disease modifying efficacy of idebenone treatment (Parkinson et al. 2013). To address the iron misdistribution with enhanced iron concentrations in mitochondria and relative cytosolic depletion, the use of membrane-permeable iron chelators, for example, deferiprone, has been considered. Deferiprone was tested in a pilot, open-label study and in a randomized, placebo-controlled phase II clinical trial. Although in the pilot study deferiprone reduced an MRI signal indicating iron overload, the phase II study failed to demonstrate an improvement of ataxia, which was even worsened by high doses of the drug. However, deferiprone was able to reduce heart hypertrophy at all tested doses (Pandolfo and Hausmann 2013). The failure of clinical trials to prove the efficacy of pharmaceutical interventions in patients with FRDA has raised a debate on whether the clinical rating scales are sufficiently sensitive and robust. So far, The International Cooperative Ataxia Rating Scale (ICARS) and the Friedreich Ataxia Rating Scale (FARS) have been used. The newly developed Scale for the Assessment and Rating of Ataxia (SARA) contains considerably fewer items, is faster to apply and probably more robust (Bürk et al. 2013). The natural history studies show that this scale may be superior to ICARS and FARS in monitoring disease progression. Further biomarkers of the disease and fully quantified timed measures need to be established and used as outcome markers for clinical trials. Over the next few years, we will hopefully see clinical trials that directly address the pathophysiological hallmark of FRDA and Frataxin deficiency. Pre-clinical and clinical trials with erythropoietin have demonstrated an increase in the concentration of the frataxin protein in patients with FRDA, although these results are not undisputed (Mariotti et al. 2013). In pre-clinical studies, a family of histone deacetylase (HDAC) inhibitors has been shown to increase frataxin mRNA and protein levels in white blood cells from FRDA patients and in GAA expansion-carrying mouse models. One of these compounds is being tested in a Phase I clinical trial (Gottesfeld et al. 2013). If applied early enough, a strategy that helps increase frataxin concentration in patients with FRDA to 50% of the levels of healthy controls (which would reflect the situation in the heterozygous, healthy carriers of the disease) can potentially prevent the clinical manifestation of the disorder or stop disease progression. This idea of a possible ‘cure’ for FRDA by means of an epigenetic approach 150 years after its first description is exciting, reflecting the rapid growth of knowledge in molecular pathophysiology over the last few years. Hopefully, non-toxic substances capable of stably increasing FXN expression over long periods of time will be identified and successfully tested in the not-so-distant future. The authors declare no conflict of interest.