SummaryDysregulated mitophagy has been linked to Parkinson's disease (PD) due to the role of PTEN-induced kinase 1 (PINK1) in mediating depolarization-induced mitophagy in vitro. Elegant mouse reporters have revealed the pervasive nature of basal mitophagy in vivo, yet the role of PINK1 and tissue metabolic context remains unknown. Using mito-QC, we investigated the contribution of PINK1 to mitophagy in metabolically active tissues. We observed a high degree of mitophagy in neural cells, including PD-relevant mesencephalic dopaminergic neurons and microglia. In all tissues apart from pancreatic islets, loss of Pink1 did not influence basal mitophagy, despite disrupting depolarization-induced Parkin activation. Our findings provide the first in vivo evidence that PINK1 is detectable at basal levels and that basal mammalian mitophagy occurs independently of PINK1. This suggests multiple, yet-to-be-discovered pathways orchestrating mammalian mitochondrial integrity in a context-dependent fashion, and this has profound implications for our molecular understanding of vertebrate mitophagy.

Mutations in PINK1 and Parkin result in autosomal recessive Parkinson's disease (PD). Cell culture and in vitro studies have elaborated the PINK1-dependent regulation of Parkin and defined how this dyad orchestrates the elimination of damaged mitochondria via mitophagy. PINK1 phosphorylates ubiquitin at serine 65 (Ser65) and Parkin at an equivalent Ser65 residue located within its N-terminal ubiquitin-like domain, resulting in activation; however, the physiological significance of Parkin Ser65 phosphorylation in vivo in mammals remains unknown. To address this, we generated a Parkin Ser65Ala (S65A) knock-in mouse model. We observe endogenous Parkin Ser65 phosphorylation and activation in mature primary neurons following mitochondrial depolarization and reveal this is disrupted in ParkinS65A/S65A neurons. Phenotypically, ParkinS65A/S65A mice exhibit selective motor dysfunction in the absence of any overt neurodegeneration or alterations in nigrostriatal mitophagy. The clinical relevance of our findings is substantiated by the discovery of homozygous PARKIN (PARK2) p.S65N mutations in two unrelated patients with PD. Moreover, biochemical and structural analysis demonstrates that the ParkinS65N/S65N mutant is pathogenic and cannot be activated by PINK1. Our findings highlight the central role of Parkin Ser65 phosphorylation in health and disease.

BackgroundAlzheimer disease (AD) and other tauopathies develop cerebral intracellular inclusions of hyperphosphorylated tau. Epidemiological and experimental evidence suggests a clear link between type 2 diabetes mellitus and AD. In AD animal models, tau pathology is exacerbated by metabolic comorbidities, such as insulin resistance and diabetes. Within this context, anitidiabetic drugs, including the widely-prescribed insulin-sensitizing drug metformin, are currently being investigated for AD therapy. However, their efficacy for tauopathy in vivo has not been tested.ResultsHere, we report that in the P301S mutant human tau (P301S) transgenic mouse model of tauopathy, chronic administration of metformin exerts paradoxical effects on tau pathology. Despite reducing tau phosphorylation in the cortex and hippocampus via AMPK/mTOR and PP2A, metformin increases insoluble tau species (including tau oligomers) and the number of inclusions with β-sheet aggregates in the brain of P301S mice. In addition, metformin exacerbates hindlimb atrophy, increases P301S hyperactive behavior, induces tau cleavage by caspase 3 and disrupts synaptic structures.ConclusionsThese findings indicate that metformin pro-aggregation effects mitigate the potential benefits arising from its dephosphorylating action, possibly leading to an overall increase of the risk of tauopathy in elderly diabetic patients.Electronic supplementary materialThe online version of this article (doi:10.1186/s13024-016-0082-7) contains supplementary material, which is available to authorized users.