Frataxin is Reduced in Friedreich Ataxia Patients and is Associated with Mitochondrial Membranes

Victoria Campuzano 1 4 Laura Montermini 1 3 4 Yves Lutz 1 4 Lidia Cova 1 2 3 4 Colette Hindelang 1 4 Sarn Jiralerspong 1 3 4 Yvon Trottier 1 4 Stephen J. Kish 0 1 4 Baptiste Faucheux 1 4 Paul Trouillas 1 4 7 Franois J. Authier 1 4 6 Alexandra Drr 1 4 Jean-Louis Mandel 1 4 Angelo Vescovi 1 2 4 Massimo Pandolfo 1 3 4 5 8 Michel Koenig 1 4 0 Clarke Institute of Psychiatry , Toronto, Canada 1 Piti-Salptrire, Paris, France 2 Laboratory of Neuropharmacology, Istituto Nazionale Neurologico 'C.Besta' , Milan, Italy 3 Centre de Recherche du CHUM , Pavillon Notre Dame, Montral, Canada 4 Institut de Gntique et de Biologie Molculaire et Cellulaire , INSERM-CNRS-ULP, Illkirch, France 5 Dpartement de Mdecine, Universit de Montral , Montral, Canada 6 Hpital Henri Mondor, Creteil, France 7 Hpital Neurologique et Neuro-Chirurgical Pierre Wertheimer, Lyon, France 8 Department of Neurology and Neurosurgery, McGill University , Montral, Canada Friedreich ataxia is a progressive neurodegenerative disorder caused by loss of function mutations in the frataxin gene. In order to unravel frataxin function we developed monoclonal antibodies raised against different regions of the protein. These antibodies detect a processed 18 kDa protein in various human and mouse tissues and cell lines that is severely reduced in Friedreich ataxia patients. By immunocytofluorescence and immunocytoelectron microscopy we show that frataxin is located in mitochondria, associated with the mitochondrial membranes and crests. Analysis of cellular localization of various truncated forms of frataxin expressed in cultured cells and evidence of removal of an N-terminal epitope during protein maturation demonstrated that the mitochondrial targetting sequence is encoded by the first 20 amino acids. Given the shared clinical features between Friedreich ataxia, vitamin E deficiency and some mitochondriopathies, our data suggest that a reduction in frataxin results in oxidative damage. - Progressive neurodegenerative diseases are conditions that unrelentingly lead to severe physical disability or death. Neurons with long myelinated axons running through the spinal cord are affected by many of these diseases. These neurons may be considered giant cells, which must bear a heavy burden for cell maintenance and the underlying energy needs. The largest neurons in the human body are those located in the dorsal root ganglia of the spinal cord. The corresponding axons, that may be more than 1 m long, run from the peripheral sensory endings through the sensory nerves and the posterior columns of the spinal cord, up to the medulla oblongata. These neurons are primarily affected in Friedreich ataxia (FRDA), an autosomal recessive progressive neurodegenerative disorder which is the most common cause of inherited ataxia in the Caucasian population (1). Other sites of neurodegeneration in Friedreich ataxia involve neurons of the spinocerebellar and corticospinal tracts (2). Non-neuronal lesions are also found in Friedreich ataxia, including hypertrophic cardiomyopathy present in most patients. Diabetes or glucose intolerance are common and possibly relate to pancreatic deficiency. All the involved tissues are highly sensitive to impaired energy production, as evidenced by their frequent involvement in mitochondrial diseases (3). Friedreich ataxia is most commonly caused by a large GAA triplet repeat expansion within the first intron of the gene encoding frataxin (4). The majority of patients (~ 95%) are homozygous for the GAA repeat expansion and the remaining 5% are compound heterozygous for the expansion and a point mutation affecting the frataxin coding sequence. In patients homozygous for the expansion there is a correlation between the number of GAA repeats on the smaller allele and age of onset, disease progression and cardiomyopathy (59), confirming that the expansion is the primary cause of the disease. The expansion causes severely reduced levels of mature frataxin mRNA (4,10). The existence of nonsense and splice mutations leading to the same clinical presentation as homozygous expansions and the reduced levels of frataxin mRNA found associated with homozygous expansions indicate that Friedreich ataxia is due to loss of frataxin function. Frataxin is a predicted 210 amino acid protein with no homology with proteins of known function. Related genes have been found in the mouse (11) and, as anonymous open reading frames (ORFs), in the nematode Caenorhabditis elegans and in yeast (4). Frataxin also shows low but significant similarity with the CyaY proteins of Gram-negative bacteria, whose function is unknown (12). The domain encoded by human exons 35 shows the highest conservation throughout evolution. In order to unravel how frataxin loss of function causes the disease, we sought to identify frataxin in cells of normal and affected individuals. We developed specific monoclonal antibodies (mAbs) raised against different regions of the protein. These mAbs detect an 18 kDa protein that is severely reduced in tissues of FRDA patients. An inverse correlation was observed between the amount of protein and expansion size on the smaller allele. Using immunofluorescence and immunoelectron microscopy we show that frataxin localizes within the mitochondria and we found that the mitochondrial target sequence is encoded by exon 1 of the frataxin gene. Antibody production and characterization In order to develop mAbs against frataxin the complete coding sequence of human frataxin was cloned in the prokaryotic expression vector pATH (13). Overexpression of pATH constructs in Escherichia coli produced fusion proteins (containing the TrpE bacterial protein at the N-terminal end) as inclusion bodies, which were isolated and injected into mice to generate antibodies. Peptide A, a potentially immunogenic peptide from the N-terminus (Fig. 1A), was also injected into mice for mAb production. Mice sera and supernatants of positive hybridoma clones were tested by Western blot of transfected COS-1 cells overexpressing the full-length frataxin protein. Five specific hybridomas were established, 1D4, 2D4, 2F10 and 1G2 raised against the fusion protein and 1H1 raised against N-terminal peptide A. Under our experimental conditions endogenous frataxin was undetectable in non-transfected COS-1 cells. On Western blots of transfected COS-1 cells all mAbs detected an ~ 30 kDa protein and smaller products that showed intensity pattern variations, depending on the antibody used (Fig. 1B). The full-length 30 kDa frataxin expressed in COS-1 cells was identical in size to the in vitro translated protein (data not shown), excluding the possibility of major post-translational modification. In order to map the region of frataxin recognized by each anti-fusion protein mAb we expressed different forms of frataxin truncated at the C-terminus in COS-1 cells. Western blot analysis of these constructs indicated that antibodies 1D4, 2F10 and 1G2 recognize epitopes e (...truncated)