2013-14 Funded Research

Enhanced α-Synuclein induced dopaminergic neurodegeneration in G2019S-BAC transgenic rats. Previously, we have reported LRRK2 knockout rats as resistant to dopaminergic neurodegeneration elicited by intracranial administration of highly purified preparations of LPS or α-synuclein over-expression (Daher et al, 2014). Such resistance to dopaminergic neurodegeneration correlated well with reduced pro-inflammatory myeloid cell recruitment to the midbrain (Daher et al, 2014). Recently, we characterized novel G2019S-BAC transgenic rats and found robust expression of LRRK2 in the substantia nigra. We aimed to test whether G2019S-LRRK2 expression in rats enhances α-synuclein induced dopaminergic neurodegeneration and inflammatory responses. We observed that G2019S-LRRK2 expression resulted in enhanced dopaminergic neurodegeneration elicited by rAAV2-α-synuclein mediated neurodegeneration. Pro-inflammatory cells recruited to the substantia nigra were enhanced in G2019S-LRRK2 rats and correlated well with neurodegeneration. These data showed that, similar to some mouse transgenic models of PD, G20129S-LRRK2 exacerbates α-synuclein-linked neurodegeneration. The relatively short timeline of the model (4 weeks) makes this model ideal to test novel neuroprotective therapies.

1. As shown by the preliminary data in the proposal, we found that Parkin protein levels are greatly up-regulated in endophilin double KO and endophilin triple KO brains. This change is not specific of brain, as we recently found that Parkin level is also increased relative to WT in primary mouse embryonic fibroblasts (MEFs) derived from endophilin triple KO mice. Overexpression of endophilin-GFP in endophilin TKO MEFs significantly rescued this up-regulation. We also examined the ubiquitination state of the total proteome in newborn endophilin TKO brains and found that it was significantly increased. In contrast, Parkin protein level is normal in the brains of other mutant mice we tested, including synaptojanin 1 KO and dynamin KO mice (two major binding partners of endophilin), Vac14 KO and Fig4 KO mice (two neurodegeneration models with abnormal PI(3,5)P2 metabolism), auxilin KO mice (a factor for clathrin uncoating) and PIPK1γ KO mice (the major PI(4,5)P2 synthesizing enzyme at synapses).

We further performed RT-PCR and found Parkin mRNA level is also strikingly increased in the endophilin double and triple KO brains and in endophilin triple KO MEF cells, indicating the increased level of Parkin is due to enhanced gene transcription. We are currently investigating which upstream transcription factors are responsible for this specific up-regulation of Parkin using promoter region analysis for PARK2 and next generation sequencing in endophilin KO brain. We have also specifically searched for a potential increase of the transcription factor ATF4, known to be responsible for the ER stress-induced Parkin up-regulation, but the result is negative.

2. To test the hypothesis that Parkin-mediated ubiquitination regulate the sequential interaction of endophilin during clathrin-mediated endocytosis, we first confirmed the interaction between Parkin and endophilin using both GST pull-down and co-immunoprecipitation assays. We also performed in vivo ubiquitination assay and found not only endophilin itself, but also two major interactors of endophilin, dynamin and synaptojanin, are mono-ubiquitinated in the presence of Parkin, indicating Parkin mediated mono-ubiquitination may play a role to regulate endophilin containing protein complex formation but not degradation. We also examined whether the Parkin-endophilin interaction affects Parkin’s ligase activity by performing auto-ubiquitination assay and the result suggests that co-expression of endophilin with Parkin has no effect on Parkin’s self-ubiquitination.

3. To investigate the functional connection between Parkin and endophilin, we generated mice harboring KO mutations in Parkin, endophilin 1 and endophilin 2. Most of these triple KO mice die perinatally, with no milk in their stomach. However, recently the similar phenotype is also observed from endophilin 1/2 double KO mice with the same genetic background. So far, it is unclear whether the further loss of Parkin contributes to the neurological phenotype in these triple KO mice.

* Current status: the manuscript for this project is submitted and under the review now.

During this fellowship period, I set up a microscope platform capable of high-throughput single-cell imaging and developed computer code to analyze the resulting images. This provided me with the technological foundation for carrying out the proposed experiments. In order to monitor how alpha-synuclein perturbs the normal cellular physiology, I constructed a yeast fluorescent reporter library that contains live-cell reporters for various organelles, biomolecules, and biochemical pathways that I expected to be affected by alpha-synuclein. I then measured with fluorescence microscopy the response of those reporters upon overexpression of alpha- synuclein. I also monitored the reporters in control cells, which do not express alpha-synuclein, to verify that the reporter responses are specific to alpha-synuclein. I found five single-cell phenotypes that were highly specific to the cells that overexpressed alpha-synuclein and were not observed in the control cells. These phenotypes are novel and will be investigated further in future work. Many reporters did not show a response (yet), most likely because the alpha-synuclein overexpression was only moderate. I expect many additional hits when I switch to a higher toxicity strain background. My initial findings clearly illustrate that the experimental platform that I developed is functional and capable of producing novel biological observations, which will eventually result in a more complete picture of the alpha-synuclein-mediated perturbations in live cells.

Future Directions: This research is still in progress and will be continued after the end of the one-year funding period of the APDA postdoctoral fellowship. Substantial progress towards the proposed goals was made. In future work, I will continue analyzing the fluorescent reporters and then use the single-cell analysis platform that I established here to study the cellular effects of several small-molecule compounds that were previously shown to rescue neurons from alpha-synuclein-mediated toxicity and investigate their mechanism of action.

Mutations in PINK1 and parkin lead to autosomal recessive forms of Parkinson’s disease (PD)/Parkinsonism. We and others have previously found that PINK1 and parkin function in a common genetic pathway to regulate mitochondrial integrity and maintenance in Drosophila. This pathway has subsequently shown to be conserved in mammals including humans. The PINK1/parkin pathway ensures tissue health by regulating mitochondrial fusion and fission, and by promoting mitophagy to remove damaged mitochondria. However, there are many unanswered questions related to how PINK1/parkin regulate mitochondrial health. One way to increase our understanding of this pathway and how dysregulation leads to Parkinson’s Disease involves identifying other pathway components.

To accomplish this, we use Drosophila to carry out genetic screens to identify genes that suppress the PINK1/parkin mutant phenotypes. We have identified a novel suppressor named MUL1, which encodes an E3 ligase. MUL1, but not the ligase-dead version of MUL1, suppresses PINK1 or parkin mutant phenotypes in Drosophila in dopaminergice neurons and muscle. MUL1 physically binds to Mitofusin and regulates ubiquitin-dependent degradation of Mitofusin, a protein that promotes mitochondrial fusion and causes PINK1/parkin mutant-like toxicity when overexpressed. We further show that removing MUL1 in PINK1 or parkin loss-of-function mutant aggravates phenotypes caused by loss of either gene alone, leading to lethality, significantly reduced ATP levels, mitochondrial morphological changes and increased Mitofusin levels in flies. The MUL1’s ability to regulate Mitofusin is conserved in mammalian cells. Removal of MUL1 in Parkin null mouse cortical neurons results in severe neurodegeneration and mitochondrial dysfunction, while removal of MUL1 or Parkin alone shows little phenotypes. Together, these observations show that MUL1 acts in parallel to the PINK1/parkin pathway on a shared target mitofusin to maintain mitochondrial integrity. The MUL1 pathway compensates for loss of PINK1/parkin in both Drosophila and mammals. MUL1 serves as an excellent therapeutic target for patients with PINK1/parkin mutations and may help Parkinson’s Disease patients with mitochondrial dysfunction in general.

Reference: Yun, J., Puri, R., Yang, H., Lizzio, M., Wu, C., Sheng, Z.H. and Guo, M. MUL1 acts in parallel to the PINK1/parkin pathway in regulating mitofusin and compensates for loss of PINK1/parkin, eLife, 2014; 3(e01958).

Other publications: Dauer, W.T. and Guo, M. Multiplying messages LRRK beneath Parkinson disease. Cell, 2014; 157: 291-293.

We have preliminary findings that LRRK2G2019S mutations interact with pathologic α -syn and that LRRK2 inhibitors may play a role in this interaction. Recent emerging evidence indicates that abnormal forms of α-syn can spread to interconnected neurons within the brain in a predictable pattern, and cause neurodegeneration in a manner that recapitulates the progression of pathology found in Parkinson’s disease (PD) brains. We will determine the impact of mutant LRRK2 on the spread of α –syn. These studies will help to determine a possible way of reducing or eliminating the threat of cell death.

We tested whether GBA (the most common gene associated with PD) mutation was neuronal toxic as the results of CMA (chaperone-mediated autophagy) blockade and impaired protein degradation in the cell. Our findings suggest that neurotoxicity of GBA mutation was a-synuclein dependent.

In addition, with the aid of APDA funding, we have obtained the frozen brain samples (all anterior cingulate cortex, that are involved in Stage 5 of PD) from the New York Brain Bank. The data we collected suggested that GBA mutation could cause DA neuronal toxicity and this process of a-synuclein dependent. GBA also perturbs CMA machinery (underlines the pathogenesis of PD), particularly LAMP2A (single membrane protein) levels. Further Studies will need to confirm that GBA mutation lead to a-synuclein accumulation, which could be due to the CMA blockade. Interventions on the CMA steps can potentially provide a novel therapy for PD.

We have evaluated the effect of 14-3-3theta overexpression in asyn-producing cells and measured the amount of asyn released by these cells. Using a virus that expresses 14-3-3theta, we introduced 14-3-3theta into asyn-producing cells and then treated our cells with doxycycline to induce asyn expression for several days and then collected the cell culture media to measure the amount of asyn released into the media. Our results show 14-3-3theta overexpression causes an increase in the total amount of asyn released by these asyn-producing cells. Conversely, when we inhibit 14-3-3s with the pan-14-3-3 inhibitor difopein in the asyn-producing cells, we find that the total amount of asyn released is significantly reduced compared to control.

We next examined the effect of 14-3-3s on the toxicity of the released asyn. aSyn normally released from the asyn-producing cells causes cell death when transferred to cultured neurons. While 14-3-3theta overexpression caused an increase in total asyn released, the released asyn caused no toxicity when transferred to separately cultured neurons compared to control. Conversely, inhibition of 14-3-3s with difopein in the asyn-producing cells dramatically increased the toxicity of the released asyn. Based on these results, we have concluded that the total amount of asyn released is not the key factor required for asyn toxicity, but that other properties of asyn are required for its toxicity.

We are now investigating what properties of asyn are altered by 14-3-3theta so that asyn is less toxic in the presence of increased 14-3-3theta levels. Previous studies have suggested that the toxic form of asyn is aggregated. Using a luciferase-based assay that can detect only aggregated forms of asyn, we have observed that 14-3-3theta overexpression reduces the level of aggregated asyn released from cells, while difopein increases the amount of aggregated asyn. We are now investigating whether 14-3-3theta may also alter other chemical modifications of asyn that have been associated with toxicity.

We received a second year of funding which allowed us to increase the length of our intervention from six months to 1-year. This is an important step and allows us to assess the effectiveness of using mobile health technology to help people with Parkinson’s Disease exercise over the long-term. We have completed our study enrollment with 51 persons with Parkinson’s Disease currently participating in the study. We expect to complete the final assessments in the spring of 2015 and submit manuscripts for publication in the fall of 2015. These results will serve as the basis for further studies to be submitted to the Patient Centered Outcomes Research Institute (PCORI) and to the National Institutes of Health (NIH).