A link between PARK and mitochondrial biology was

A link between PARK2 and mitochondrial biology was first established in Drosophila, which displayed impairment in mitochondrial function and neuronal loss in an age-dependent manner when rendered deficient for PARK2 (Greene et al., 2003). Likewise, similar mitochondrial defects exhibited in Park2 knockout (KO) mouse models, although only mice with conditional KO of Park2 recapitulate parkinsonian phenotype and striatonigral degeneration (Dawson et al., 2010; Goldberg et al., 2003). Analysis of single and double mutants in mice and flies also suggests that Pink1 is upstream of Park2 and that overexpression of PARK2 alone or directing PARK2 to mitochondria is sufficient to introduce mitochondrial fragmentation (Akundi et al., 2013; Clark et al., 2006; Kim et al., 2008; Shiba-Fukushima et al., 2012). Thus, both gain or loss of function can affect mitochondrial dynamics. More recently, post-mortem dexamethasone acetate tissues of PD patients also confirmed the involvement of altered mitochondrial pathologies in disease process (Henchcliffe and Beal, 2008; Schapira et al., 1989; Vila et al., 2008). The emerging hypothesis is that in normal cells PARK2 is cytoplasmic and PINK1 levels are low. However, when mitochondrial potential is lost, PINK1 accumulates on depolarized membranes and recruits PARK2 to mitochondria and are then targeted for degradation via mitophagy. Loss or damaged mitochondria stimulate mitochondrial fission and/or inhibit fusion by negatively regulating MFN and OPA1 function and/or positively regulating DRP1 (van der Bliek et al., 2013). Despite these advances, differences between species in displaying neurodegenerative phenotypes have made it difficult to extrapolate the results obtained from animal models to human. The discovery of induced pluripotent stem cells (iPSCs) has for the first time enabled us to reproduce dopaminergic neurons from individuals who suffer from familial or sporadic PD. Indeed, a recent iPSC-based study showed that PARK2 controlled dopamine utilization in iPSC-derived dopaminergic neurons (Jiang et al., 2012). Likewise, advances in gene targeting (Cathomen and Joung, 2008; Urnov et al., 2010; Zeng et al., 2014) allow us to develop the corresponding models in an isogenic background. To enable us to study the role of PARK2 in human PD, we made integration-free iPSC lines from four PD patients carrying different PARK2 mutations (NINDS collection; Table S1). We showed a deficiency in dopaminergic differentiation and a reduction in mitochondrial volume fraction in all four PARK2 lines compared with an age-matched control subject. To confirm the results from the patient-specific disease model and to overcome the genetic variation among patient lines that could mask the PARK2 phenotype, we generated PARK2 isogenic controls using a KO strategy in a well-characterized integration-free iPSC line. We found similar phenotypes in the PARK2 KO isogenic line as seen from the familial PARK2 lines. We showed that loss-of-function mutations in PARK2 impaired dopaminergic development by reducing the percentage of Tyrosine hydroxylase-positive (TH+) neurons and accumulation of α-synuclein (SNCA) in dopaminergic neurons. These results were supported by whole genome expression profiling in which alterations in expression of mitochondria and cell death-related genes were observed in the dopaminergic neuron stage but not in earlier stages of differentiation. In addition, we showed that similar changes were detected in a pure population of forebrain neurons derived from the isogenic model. Our results suggest that PARK2 is involved in mitochondrial regulation in neurons.
Results
Discussion Mutations in the PARK2 gene are associated with PD, although the exact mechanism by which PARK2 contributes to the selective neuronal degeneration in PD is unknown. Different lines of evidence indicate that alterations in many aspects of mitochondrial biology such as complex I activity, fission and fusion, mitophagy, transport of mitochondria in neurons, and alterations mitochondrial membrane potential may contribute to PD (Dauer and Przedborski, 2003; Exner et al., 2012). Consistent with the mitochondrial hypothesis, it has been postulated that the role of PARK2 and PINK1 in mitochondrial quality control underlies the basis of PARK2-related PD. Our results showing an alteration in mitochondrial volume in PARK2 mutants in a primary human dopaminergic cell model is consistent with this hypothesis. The deficits in mitochondrial volume were accompanied by a reduction in dopaminergic neurons in PARK2 patient lines, and these phenotypes were recapitulated in our isogenic PARK2 lines. Whole genome expression profiling confirmed the phenotype and identified mitochondrial-associated cell death as a cause for the reduction in cell number.