Identification and validation of physiological substrates and interaction partners of PINK1 and LRRK2, 2 kinases linked to familial forms of Parkinson's disease.

Parkinsons disease (PD) is a common neurodegenerative disorder affecting more than 4 million people worldwide. To date, the exact etiology of PD is poorly understood. Current therapy improves motor symptoms but does not slow disease progression. In the past 17 years, identification of highly penetrant mutations in several genes that cause PD has contributed enormously to our understanding of PD pathology. We believe that further efforts in understanding the key molecular processes that are involved in these genetic forms of PD will eventually lead to the discovery of promising therapeutic strategies. In 2004, mutations in PINK1 were identified in families with early onset PD with an autosomal recessive inheritance pattern. PINK1 is serine/threonine kinase with an N-terminal mitochondrial localisation signal and a central kinase domain. It has been shown that PINK1 is involved in mitochondrial quality control either by affecting mitochondrial dynamics or by the clearance of damaged mitochondria by mitophagy. Mutations in LRRK2 were also identified in 2004 as a cause of autosomal dominant forms of PD. LRRK2 is a complex multidomain protein consisting of two catalytic domains (a GTPase domain and a kinase domain) and several domains involved in protein-protein interactions. LRRK2 is a member of the Ras of complex (ROCO) protein family. Of the four human ROCO proteins, LRRK1 is the closest homolog of LRRK2. Despite this close homology, there is no evidence so far that mutations in LRRK1 are associated with PD. The exact biological function of LRRK2 is poorlyunderstood. Several studies have investigated the effect of LRRK2 on cellular toxicity although due to variation it has been difficult to draw a clear conclusion. In addition LRRK2 has also been implicated in the regulation of neurite outgrowth. To date however, the exact upstream regulators and downstream effectors of LRRK2 remain unclear. In order to gaininsight in the signalling cascades linked to these two kinases, we searched for cellular protein interactors and physiological substrates of PINK1 and LRRK2 by using different proteomic approaches.

First, in order to study the cellular functions of LRRK1 and LRRK2 and to investigate the possibility of overlapping functions, we searched for LRRK1 and LRRK2 interacting proteins. Therefore we used a protein microarray-basedapproach in combination with affinity purification from cell lysates coupled to mass spectrometry. We found that 14-3-3 proteins interacted specifically with LRRK2, while EGF-R specifically interacted with LRRK1. These observations were consistent with phosphosite mapping of LRRK1, revealing phosphosites outside of 14-3-3 consensus binding motifs.
In addition, we also showed that the cellular relocalisation observed for LRRK2 after treatment with LRRK2 kinase inhibitor IN1 and for LRRK1 after stimulation with EGF was specific for each LRRK and RNAi-mediated knockdown of one of the two LRRK proteins had no effect on these observations. Our results suggest that, although LRRK1 and LRRK2 have a number of common interactors, there is no evidence for signalling crosstalk between the LRRK2 proteins at the level of two LRRK-specific cellular interactions: LRRK1:EGF-R and LRRK2:14-3-3. Next, we used a chemical genetic approach to identify LRRK2 substrates from mouse brain lysates. This approach allows the direct identification of substrates of a particular kinase in the context of a complex cellular environment. We identified MAP/Microtubule Affinity-Regulating Kinase 1 (MARK1) as a substrate and showed thatin vitro and in living cells MARK1 is phosphorylated by LRRK2. MARK1 isa ser/thr kinase that regulates stability of microtubules via the phosphorylation of microtubule-associated proteins including Tau. Thereby, our results provide a possible explanation for the observed effects of LRRK2 on cytoskeleton dynamics.

To further elucidate the cellular pathways that involve PINK1 we used two different proteomic approaches. First, to gain insight in the kinase signalling pathways that involve PINK1 we performed a 2D-DIGE proteomic study on mitochondria-enriched brainextracts from PINK1 KO mice compared to a control mice. This revealed adownregulation of proteins involved in energy metabolism, cytoskeleton and synaptic transmission. Our study provides an explanation for the observed decrease in dopamine release and dopamine content in PINK1 KO mice. Second, we have identified PINK1 interacting proteins. Therefore, a strategy was developed to isolate PINK1 protein complexes from mouse striatum and from SH-SY5Y neuroblastoma cell lines. Interesting candidates have been tested for their capacity to modify the CCCP-induced translocation of parkin. We showed that knockdown of some of the candidates significantly affected the translocation of parkin. Although further experiments are needed to study the exact role of these hits in PINK1/parkin-mediated mitophagy, they are a good starting point for developing therapies to target impaired mitophagy in PD.

Keywords:Parkinson's disease

Disciplines:Biochemistry and metabolism, Medical biochemistry and metabolism, Neurosciences, Biological and physiological psychology, Cognitive science and intelligent systems, Developmental psychology and ageing

Project type:PhD project