AbstractParkinson’s Disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s and old age is the strongest risk factor for developing PD. PD has traditionally been seen as a motor disorder, but its non-motor symptoms such as dysautonomia, sensory dysfunction, sleeping problems and neuropsychiatric features equally add to the disease burden. There is no cure for PD and this is probably a reflection of our poor understanding of the disease pathogenesis. One way of tackling this is to focus on the small, but significant number of PD patients with a family history compatible with Mendelian autosomal inheritance (10-15%). Hereditary and sporadic PD share important clinical and neuropathological features, and there is reasonable hope that dissecting molecular pathways of PD gene products will have more general implications for the pathophysiology of PD associated neurodegeneration and help device new treatment strategies.
Mutations in the FBXO7 gene have recently been shown to cause an autosomal recessive early onset Parkinsonian-pyramidal syndrome and FBXO7 has been designated as PARK 15 (Di Fonzo et al., 2009). FBXO7 is a member of the F-box protein family, which functions as the variable subunit of Skp1-Cullin1-F-box protein (SCF) E3 ubiquitin ligase complexes and as such dictate substrate specificity. The canonical outcome of ubiquitylation is proteasomal degradation and my working hypothesis is that FBXO7 may be involved in protein quality control in the brain. A perturbation thereof may be a first step towards FBXO7 dependent disease.
At the time of starting with my PhD project, little was known about the molecular function of FBXO7 and how mutations in FBXO7 result in neurodegeneration. In order to learn more and dissect the signalling pathway of FBXO7 I have used tagged stable overexpression cell lines of the FBXO7 wildtype as well as human disease mutant proteins for tag-pulldowns followed by mass-spectrometry to identify interacting partners and possible substrates. With this approach I have been able to confirm the interaction between FBXO7 and its core SCF E3 ligase partners as well as some of the previously reported interacting partners. I have been able to show that not only the FBXO7 wildytpe protein, but also all of the so far reported human disease mutants are able to assemble into an SCF complex. Hence, my fist conclusion is that the human disease mutants do not exert their pathogenicity by SCF complex disruption.
Next, a knock-in (KI) mouse model of one of the pathogenic FBXO7 mutations (R378G) was generated and evaluated by molecular and biochemical approaches as well as motor and behaviour phenotyping. In particular, I have used the Fbxo7 mouse model for extensive proteomic screens to identify wildtype (wt) and KI Fbxo7 interactors: endogenous Fbxo7 immunoprecipitations from mouse brain lysates and subsequent fingerprint mass-spectrometry; differential whole proteome: ex vivo differential dimethyl labelling of wt and KI brain samples, and Fbxo7-dependent ubiquitinome analysis: quantitative di-GLY capture proteomics combining in vivo SILAC labelling with antibody-based affinity enrichment of “di-GLY remnant motifs”- containing peptides prior to proteomic profiling of the wild-type in comparison to the homozygous R379G Fbxo7 KI ubiquitinome in MEF lysates. The di-GLY remnant motif is the signature peptide of ubiquitinylated protein sites at peptide level after tryptic digestions. Some of my findings are:
• For the first time I show that endogenous Fbxo7 actually assembles into an Skp1-Cullin1-Fbxo7 complex and that the pathogenic R378G does not disrupt SCFFbxo7-KI complex formation in vivo. This is true for the Fbxo7 KI mouse model, but also for patient derived immortalized cell lines carrying the R378G FBXO7 mutation.
• Endogenous Fbxo7 interacts with the Sumo E3 ligase complex RanBP2/ RanGAP1*Sumo1/Ubc9 complex.
• In the differential enrichment of ubiquitylated protein species in SILAC labelled wild-type and homozygous R379G Fbxo7 KI MEFs, I have clearly identifies 2 highly conserved lysine residues, which are conserved amongst VDAC 1, 2, and 3 in mouse as well as human homologous, to be preferentially ubiquitinylated in a Fbxo7 wild-type background (in collaboration with Dr. Patrick Pedrioli, MRC Programme leader).
• There is a significant difference in motor performance between wildtype and homozygous R379G KI Fbxo7 mice at 10 months of age (in collaboration with Dr. Steve Martin, Neuroscience Division, Dundee).
• Furthermore, I have successfully set up an in vitro FBXO7 dependent ubiquitinylation assays.
|Date of Award||2014|
|Supervisor||Dario Alessi (Supervisor)|
- signalling pathway