Frataxin deficiency enhances apoptosis in cells differentiating into neuroectoderm

Manuela M. Santos 0 1 2 Keiichi Ohshima 1 2 Massimo Pandolfo 1 2 0 UnIGENe, Instituto de Biologia Molecular e Celular , 4150-180 Porto , Portugal 1 Quebec , H2L 4M1 , Canada 2 Department of Medicine, Centre Hospitalier de l'Universite de Montreal, Hopital Notre-Dame , Montreal Deficiency of the mitochondrial matrix protein frataxin causes Friedreich ataxia. Frataxin function is believed to be related to mitochondrial iron metabolism and free radical production. In Friedreich ataxia, loss of dorsal root ganglia neurons occurs early in life, suggesting a developmental process. In addition, frataxin knockout mice die during embryonic life, further suggesting that frataxin is necessary for normal development. In this study we examine the role of frataxin in neuronal differentiation by using the P19 embryonic carcinoma cell line as a model system. We produced stably transfected clones with antisense or sense frataxin constructs. During retinoic acid-induced neurogenesis of frataxin-deficient cells there was a striking rise in cell death, while cell division remained unaffected. However, frataxin deficiency does not affect cell survival in cells induced to differentiate into cardiomyocytes. Frataxin deficiency enhances apoptosis of retinoic acid-stimulated cells, and the number of neuronal-like cells expressing MAP2 was dramatically reduced in these clones. In addition, we found that antisense clones induced to differentiate into neuroectoderm with retinoic acid have increased production of reactive oxygen species, and that only cells noncommitted to the neuronal lineages could be rescued by the addition of the antioxidant N-acetyl-cysteine (NAC). However, NAC treatment had no effect in increasing the number of terminally differentiated neuronal-like cells in frataxin-deficient clones. Our results suggest that frataxin deficiency may render cells susceptible to apoptosis after exposure to appropriate stimuli. - Friedreich ataxia (FRDA) is an autosomal recessive degenerative disease affecting the nervous system and the heart (1,2). Its neuropathology is characterized by atrophy of the sensory pathways, with early loss of large neurons in the dorsal root ganglia (DRG), sensory axonal neuropathy and degeneration of the posterior columns of the spinal cord (2,3). Onset is usually in childhood or adolescence, but may be delayed to middle or later years of life (46). A considerable breakthrough towards the understanding of the molecular pathogenesis of FRDA was achieved by the discovery of its causative gene (7). The encoded protein, named frataxin, is localized in mitochondria (811). Frataxin is highly conserved in evolution, with homologs in essentially all eukaryotes and some prokaryotes (7,8). Yeast cells with a disrupted frataxin homolog gene (YFH1) accumulate 10-fold more iron in mitochondria than wild-type, lose mitochondrial DNA, and become unable to carry out oxidative phosphorylation (9,12). Loss of respiratory competence requires the presence of iron and occurs more rapidly as iron concentration in the medium is increased (13). The mechanism of mitochondrial iron accumulation is unknown, but Yfh1p has been shown to induce a flux of non-heme iron out of mitochondria (13). Iron in mitochondria amplifies the toxicity of reactive oxygen species (ROS) leaking from the respiratory chain. The free hydroxyl radical (OH), in particular, may be produced by Fenton chemistry and causes lipid peroxidation, protein and nucleic acid damage. Occurrence of the Fenton reaction in YFH1 yeast cells is suggested by their enhanced sensitivity to H2O2 (9). Many lines of evidence indicate that frataxin function is conserved in humans (14), suggesting that the mechanism by which cellular damage might occur in FRDA patients involves mitochondrial dysfunction triggered by ROSmediated damage (15,16). That frataxin plays an important role during early development is exemplified by the observation that frataxin knockout mice die in utero shortly after implantation at embryonic day (E)6.5 (17). In order to study the role of frataxin during cell differentiation and development, we turned to an embryonic carcinoma (EC) cell model system, the P19 mouse EC cells, which can be induced to differentiate into a variety of cell types (18,19). P19 cells resemble those of the inner mass of the blastocyst, and their differentiation is believed to closely mimic critical events in early embryogenesis. Under appropriate culture conditions, P19 cells display the ability to differentiate into derivatives of three germ layers; endoderm, mesoderm and ectoderm. Treatment of aggregated P19 cells with retinoic acid (RA) effectively induces the development of neurons, astroglia and microglia, cell types normally derived from the neuroectoderm (18). Aggregates of P19 cells exposed to dimethyl sulfoxide (DMSO) differentiate into endodermal and mesodermal derivatives, including cardiac and skeletal muscle (19). In this study we evaluated how frataxin levels change during differentiation and characterized the effect on morphology, growth and differentiation of stable transfectants containing either a frataxin sense or antisense vector. P19 mouse EC cells are induced to differentiate into neuronal precursors upon treatment with RA and aggregation on bacterial-grade Petri dishes for 4 days, followed by dissociation and 4 days further growth on tissue culture plates (19). Untreated P19 cells grow densely packed and display a characteristic cuboidal morphology, evident in phase contrast microscopy. Treatment with 1 M RA results in a dramatic change in morphology and differentiation to a neuron-like phenotype with neurite-like morphology. Neuronal differentiated P19 cells are prominently stained by an antibody directed against the microtubule associated protein-2 (MAP2). We investigated frataxin expression in P19 cells during differentiation at the protein level by immunoblotting. Frataxin levels were found to increase 23-fold over levels in untreated cells upon aggregation in the presence of RA between days 2 and 4, followed by a decrease after dissociation and regrowth on tissue culture plates (Fig. 2A). By comparison, -tubulin levels remained unchanged. The development of neurons was confirmed here by the appearance of the neuronspecific marker MAP2. We next examined frataxin expression during cardiomyocyte and endodermal differentiation. To induce muscle and cardiomyocyte differentiation, P19 cultures were grown as aggregates in bacterial dishes in the presence of 1% DMSO and then plated in cell culture-treated dishes on day 4 without DMSO. Rhythmically beating regions were detectable by microscopic examination from day 6. By immunoblot, frataxin levels were observed to dramatically increase from day 2 (Fig. 2B), in a very similar way as when cells were differentiated primarily into endodermal cells by aggregation in the absence of inducer (Fig. 2C). An actin antibody (clone 5C5) and the TROMA-1 antibody were used (...truncated)