LRRK2 is a kinase mutated in Parkinson's disease, but how the protein affects synaptic function remains enigmatic. We identified LRRK2 as a critical regulator of EndophilinA. Using genetic and biochemical studies involving Lrrk loss-of-function mutants and Parkinson-related LRRK2(G2019S) gain-of-kinase function, we show that LRRK2 affects synaptic endocytosis by phosphorylating EndoA at S75, a residue in the BAR domain. We show that LRRK2-mediated EndoA phosphorylation has profound effects on EndoA-dependent membrane tubulation and membrane association in vitro and in vivo and on synaptic vesicle endocytosis at Drosophila neuromuscular junctions in vivo. Our work uncovers a regulatory mechanism that indicates that reduced LRRK2 kinase activity facilitates EndoA membrane association, while increased kinase activity inhibits membrane association. Consequently, both too much and too little LRRK2-dependent EndoA phosphorylation impedes synaptic endocytosis, and we propose a model in which LRRK2 kinase activity is part of an EndoA phosphorylation cycle that facilitates efficient vesicle formation at synapses.

Mutations in leucine-rich repeat kinase 2 (LRRK2) are associated with Parkinson's disease, but the precise physiological function of the protein remains ill-defined. Recently, our group proposed a model in which LRRK2 kinase activity is part of an EndoA phosphorylation cycle that facilitates efficient vesicle formation at synapses in the Drosophila melanogaster neuromuscular junctions. Flies harbor only one Lrrk gene, which might encompass the functions of both mammalian LRRK1 and LRRK2. We therefore studied the role of LRRK2 in mammalian synaptic function and provide evidence that knockout or pharmacological inhibition of LRRK2 results in defects in synaptic vesicle endocytosis, altered synaptic morphology and impairments in neurotransmission. In addition, our data indicate that mammalian endophilin A1 (EndoA1, also known as SH3GL2) is phosphorylated by LRRK2 in vitro at T73 and S75, two residues in the BAR domain. Hence, our results indicate that LRRK2 kinase activity has an important role in the regulation of clathrin-mediated endocytosis of synaptic vesicles and subsequent neurotransmission at the synapse.

The prevalence of intellectual disability is around 3%; however, the etiology of the disease remains unclear in most cases. We identified a series of patients with X-linked intellectual disability presenting mutations in the Rad6a (Ube2a) gene, which encodes for an E2 ubiquitin-conjugating enzyme. Drosophila deficient for dRad6 display defective synaptic function as a consequence of mitochondrial failure. Similarly, mouse mRad6a (Ube2a) knockout and patient-derived hRad6a (Ube2a) mutant cells show defective mitochondria. Using in vitro and in vivo ubiquitination assays, we show that RAD6A acts as an E2 ubiquitin-conjugating enzyme that, in combination with an E3 ubiquitin ligase such as Parkin, ubiquitinates mitochondrial proteins to facilitate the clearance of dysfunctional mitochondria in cells. Hence, we identify RAD6A as a regulator of Parkin-dependent mitophagy and establish a critical role for RAD6A in maintaining neuronal function.