THE ROLE OF MICROTUBULES IN GENE- AND TOXIN-BASED NEURODEGENERATION UNDERLYING PARKINSON'S DISEASE

Currently, there are just symptomatic treatments available for Parkinson’s disease (PD), that is the second most common neurodegenerative disease after Alzheimer’s disease and is predicted to increase in prevalence as the world population ages. Its central pathological features is the selective degeneration of dopaminergic neurons in the Substantia nigra (SN) pars compacta projecting to the Corpus striatum (CS), leading to a striatal dopamine deficiency resulting in the typical movement disorders of parkinsonism. Even though the majority of PD cases are sporadic, mutations in a number of genes have been associated with familial PD. It’s also known that parkinsonism can be induced by exposure to environmental toxins such as pesticide, chemical compounds and hydrocarbon solvents, including 2,5-hexanedione (2,5-HD), the toxic metabolite of n-hexane. 2,5-HD has been shown to induce parkinsonism in animals and humans and to affect directly the cytoskeletal proteins. In particular, microtubules (MTs) have been found to interact with some of the proteins mutated in PD, such as α-synuclein, LRRK2 and parkin, and to be affected by the action of some PD toxins like MPP+ and rotenone. Therefore, in the last years, the MT dysfunction has become an emerging hypothesis in PD pathogenesis. In this scenario, our goal was to investigate the MT dysfunction in neuronal cells, primary skin fibroblasts from PD patients and transgenic mice, taking advantage of both a gene- (using PARK2 mutations) and toxin- (using 2,5-HD) based models of PD neurodegeneration. In the first part of the project, nerve growth factor (NGF)-differentiated PC12 cell line has been used as a model of dopaminergic neurons in culture using three different concentrations of toxin (0,2 mM; 2 mM and 20 mM 2,5-HD) for 24 hours, in order to study the early events of neurodegeneration. Thus, the characterization of the effects of 2,5-HD on cytoskeleton has been carried out through both western blot analysis and immunofluorescence techniques, revealing an impact on all its components (actin, neurofilaments and MTs). Subsequently, I focused on MT system through the analysis of different post-translational modified forms of α-tubulin, showing significative MT stabilizing effects of 2,5-HD in both levels and distribution, in particular it could increase the levels of stable MTs that appeared fragmented or accumulated in the cell body. In accordance with these results, the analysis of tubulin polymerization in cell revealed a higher content of MT mass caused by 2,5-HD. On the contrary, from our in vitro data no significant effects of 2,5-HD emerged in the tested conditions and also the ultrastructure of MTs obtained in the presence of toxin resulted conventional. Interestingly, the first signs of mitochondrial damage, in our experimental conditions, seemed to be induced only at the highest concentration of 2,5-HD, while strong effects on cytoskeleton came up earlier at lower doses in our cellular model. Following, the effect of 2,5-HD has been tested on skin fibroblasts, obtained from healthy donors and PD patients carrying mutation in PARK2 gene, since it encodes for parkin, an E3 ubiquitin ligase that is supposed to bind and stabilize MTs. Cell viability was not affected by 2,5-HD whereas cell morphology appeared significantly modified just in PD patient fibroblasts. Moreover, we found that cytoskeletal organization and stability were affected, with a consequent alteration of cell morphology and behaviour, in PD patient cells already at baseline conditions without the addition of any stressor: all parkinsonian fibroblasts showed a reduced MT mass and displayed significant changes in MT stability-related signalling pathways, without any activation of autophagy or apoptosis. This shows for the first time that MT dysfunction occurs in patients and not only in experimental models of PD. The PD fibroblasts were also much more susceptible to 2,5-HD effects than healthy controls, suggesting that the genetic background may really make the difference in MT susceptibility to environmental factors. Consistent with this hypothesis, we observed the increase of fragmentation of stable MTs in PARK2 patient-derived ventral midbrain neurons. The second part of the project has been dedicated to in vivo experiments in wild type and PARK2 heterozygous (PARK2+/-) mice. In fact, although PARK2 mutations are responsible for a familial early-onset autosomal recessive form of PD, some individuals carrying heterozygous mutations, usually asymptomatic, have been found to present nigrostriatal abnormalities, making PARK2 haploinsufficiency a possible risk factor for developing the late-onset disease or other neurological disorder. Biochemical analysis of the cytoskeletal protein level in lysates from SN and CS and confocal microscopy on immunostained brain slices have revealed that the MT system is more dynamic in PARK+/- mice respect to wt ones. In addition, we evaluated the motor behaviour of these mice using a video-tracking system mounted above open field cages. Surprisingly, we found the heterozygous mice were significantly more active than wt ones. Finally, we have found no cell loss in both genetic backgrounds or terminal loss in the CS after treatment. The imbalance of post-translationally modified tubulins, that are associated with differences in MT stability, occurs also in PARK2 knockout mice and precedes the block of mitochondrial transport. Our data showed that PARK2 mutations or haploinsufficiency impacts MT system in vivo, unravelling parkin as a regulator of MT stability in neurons and suggesting a key role for MT dysfunction in the PD selectively dopaminergic neurodegeneration.