Search

TP53INP1 is a stress-induced protein, which acts as a dual positive regulator of transcription and of autophagy and whose deficiency has been linked with cancer and metabolic syndrome. Here, we addressed the unexplored role of TP53INP1 and of its Drosophila homolog dDOR in the maintenance of neuronal homeostasis under chronic stress, focusing on dopamine (DA) neurons under normal ageing- and Parkinson’s disease (PD)-related context. Trp53inp1−/− mice displayed additional loss of DA neurons in the substantia nigra compared to wild-type (WT) mice, both with ageing and in a PD model based on targeted overexpression of α-synuclein. Nigral Trp53inp1 expression of WT mice was not significantly modified with ageing but was markedly increased in the PD model. Trp53inp2 expression showed similar evolution and did not differ between WT and Trp53inp1−/− mice. In Drosophila, pan-neuronal dDOR overexpression improved survival under paraquat exposure and mitigated the progressive locomotor decline and the loss of DA neurons caused by the human α-synuclein A30P variant. dDOR overexpression in DA neurons also rescued the locomotor deficit in flies with RNAi-induced downregulation of dPINK1 or dParkin. Live imaging, confocal and electron microscopy in fat bodies, neurons, and indirect flight muscles showed that dDOR acts as a positive regulator of basal autophagy and mitophagy independently of the PINK1-mediated pathway. Analyses in a mammalian cell model confirmed that modulating TP53INP1 levels does not impact mitochondrial stress-induced PINK1/Parkin-dependent mitophagy. These data provide the first evidence for a neuroprotective role of TP53INP1/dDOR and highlight its involvement in the regulation of autophagy and mitophagy in neurons.

Complex animals build specialised muscle to match specific biomechanical and energetic needs. Hence, composition and architecture of sarcomeres as well as mitochondria are muscle type specific. However, mechanisms coordinating mitochondria with sarcomere morphogenesis are elusive. Here we useDrosophilamuscles to demonstrate that myofibril and mitochondria morphogenesis are intimately linked. In flight muscles, the muscle selectorspaltinstructs mitochondria to intercalate between myofibrils, which in turn mechanically constrain mitochondria into elongated shapes. Conversely in cross-striated muscles, mitochondria networks surround myofibril bundles, contacting myofibrils only with thin extensions. To investigate the mechanism causing these differences, we manipulated mitochondrial dynamics and found that increased mitochondrial fusion during myofibril assembly prevents mitochondrial intercalation in flight muscles. Strikingly, this coincides with the expression of cross-striated muscle specific sarcomeric proteins. Consequently, flight muscle myofibrils convert towards a partially cross-striated architecture. Together, these data suggest a biomechanical feedback mechanism downstream ofspaltsynchronizing mitochondria with myofibril morphogenesis.

Charcot-Marie-Tooth disease type 2A (CMT2A) is caused by dominant alleles of the mitochondrial pro-fusion factor Mitofusin 2 (MFN2). To address the consequences of these mutations on mitofusin activity and neuronal function, we generate Drosophila models expressing in neurons the two most frequent substitutions (R94Q and R364W, the latter never studied before) and two others localizing to similar domains (T105M and L76P). All alleles trigger locomotor deficits associated with mitochondrial depletion at neuromuscular junctions, decreased oxidative metabolism and increased mtDNA mutations, but they differently alter mitochondrial morphology and organization. Substitutions near or within the GTPase domain (R94Q, T105M) result in loss of function and provoke aggregation of unfused mitochondria. In contrast, mutations within helix bundle 1 (R364W, L76P) enhance mitochondrial fusion, as demonstrated by the rescue of mitochondrial alterations and locomotor deficits by over-expression of the fission factor DRP1. In conclusion, we show that both dominant negative and dominant active forms of mitofusin can cause CMT2A-associated defects and propose for the first time that excessive mitochondrial fusion drives CMT2A pathogenesis in a large number of patients.

International audience; L-glutamate is the major excitatory neurotransmitter in the mammalian brain. Specific proteins, the Na+/K+-dependent high affinity excitatory amino acid transporters (EAATs), are involved in the extracellular clearance and recycling of this amino acid. Type I synapses of the Drosophila neuromuscular junction (NMJ) similarly use L-glutamate as an excitatory transmitter. However, the localization and function of the only high-affinity glutamate reuptake transporter in Drosophila, dEAAT1, at the NMJ was unknown. Using a specific antibody and transgenic strains, we observed that dEAAT1 is present at the adult, but surprisingly not at embryonic and larval NMJ, suggesting a physiological maturation of the junction during metamorphosis. We found that dEAAT1 is not localized in motor neurons but in glial extensions that closely follow motor axons to the adult NMJ. Inactivation of the dEAAT1 gene by RNA interference generated viable adult flies that were able to walk but were flight-defective. Electrophysiological recordings of the thoracic dorso-lateral NMJ were performed in adult dEAAT1-deficient flies. The lack of dEAAT1 prolonged the duration of the individual responses to motor nerve stimulation and this effect was progressively increased during physiological trains of stimulations. Therefore, glutamate reuptake by glial cells is required to ensure normal activity of the Drosophila NMJ, but only in adult flies.

Huntington's disease (HD) is a late onset heritable neurodegenerative disorder caused by expansion of a polyglutamine (polyQ) sequence in the protein huntingtin (Htt). Transgenic models in mice have suggested that the motor and cognitive deficits associated to this disease are triggered by extended neuronal and possibly glial dysfunction, whereas neuronal death occurs late and selectively. Here, we provide in vivo evidence that expanded polyQ peptides antagonize epidermal growth factor receptor (EGFR) signaling in Drosophila glia. We targeted the expression of the polyQ-containing domain of Htt or an extended polyQ peptide alone in a subset of Drosophila glial cells, where the only fly glutamate transporter, dEAAT1, is detected. This resulted in formation of nuclear inclusions, progressive decrease in dEAAT1 transcription and shortened adult lifespan, but no significant glial cell death. We observed that brain expression of dEAAT1 is normally sustained by the EGFR-Ras-extracellular signal-regulated kinase (ERK) signaling pathway, suggesting that polyQ could act by antagonizing this pathway. We found that the presence of polyQ peptides indeed abolished dEAAT1 upregulation by constitutively active EGFR and potently inhibited EGFR-mediated ERK activation in fly glial cells. Long polyQ also limited the effect of activated EGFR on Drosophila eye development. Our results further indicate that the polyQ acts at an upstream step in the pathway, situated between EGFR and ERK activation. This suggests that disruption of EGFR signaling and ensuing glial cell dysfunction could play a direct role in the pathogenesis of HD and other polyQ diseases in humans.

International audience; Cristae are mitochondrial inner-membrane structures that concentrate respiratory chain complexes and hence regulate ATP production. Mechanisms controlling crista morphogenesis are poorly understood and few crista determinants have been identified. Among them are the Mitofilins that are required to establish crista junctions and ATP-synthase subunits that bend the membrane at the tips of the cristae. We report here the phenotypic consequences associated with the in vivo inactivation of the inner-membrane protein Pantagruelian Mitochondrion I (PMI) both at the scale of the whole organism, and at the level of mitochondrial ultrastructure and function. We show that flies in which PMI is genetically inactivated experience synaptic defects and have a reduced life span. Electron microscopy analysis of the inner-membrane morphology demonstrates that loss of PMI function increases the average length of mitochondrial cristae in embryonic cells. This phenotype is exacerbated in adult neurons in which cristae form a dense tangle of elongated membranes. Conversely, we show that PMI overexpression is sufficient to reduce crista length in vivo. Finally, these crista defects are associated with impaired respiratory chain activity and increases in the level of reactive oxygen species. Since PMI and its human orthologue TMEM11 are regulators of mitochondrial morphology, our data suggest that, by controlling crista length, PMI influences mitochondrial diameter and tubular shape.

L-glutamate is both the major brain excitatory neurotransmitter [1, 2] and a potent neurotoxin [3, 4] in mammals. Glutamate excitotoxicity is partly responsible for cerebral traumas evoked by ischemia [5, 6] and has been implicated in several neurodegenerative diseases including amyotrophic lateral sclerosis (ALS) [7–9]. In contrast, very little is known about the function or potential toxicity of glutamate in the insect brain. Here, we show that decreasing glutamate buffering capacity is neurotoxic in Drosophila . We found that the only Drosophila high-affinity glutamate transporter, dEAAT1 [10–13], is selectively addressed to glial extensions that project ubiquitously through the neuropil close to synaptic areas. Inactivation of dEAAT1 by RNA interference led to characteristic behavior deficits that were significantly rescued by expression of the human glutamate transporter hEAAT2 or the administration in food of riluzole, an anti-excitotoxic agent used in the clinic for human ALS patients. Signs of oxidative stress included hypersensitivity to the free radical generator paraquat and rescue by the antioxidant melatonin. Inactivation of dEAAT1 also resulted in shortened lifespan and marked brain neuropil degeneration characterized by widespread microvacuolization and swollen mitochondria. This suggests that the dEAAT1-deficient fly provides a powerful genetic model system for molecular analysis of glutamate-mediated neurodegeneration.