Synapse formation starts during development as axons establish contacts with dendrites and neuronal cell bodies. Having a single elongated axon and multiple dendrites derived from the cell body attribute neurons an exceptional polarity, resulting in synapses being located distant to the cell body where most protein synthesis takes place. Thus, to ensure stoichiometric assembly of functional presynaptic units, components of the presynaptic compartment, synaptic vesicle (SV) and active zone (AZ) proteins as well as synaptic cell adhesion molecules (sCAMs), need to be delivered to nascent synapses in a coordinated manner. Mechanisms regulating the efficient delivery of presynaptic proteins to axon terminals as well as the cellular identity of presynaptic transport organelles are incompletely understood. This thesis therefore focused on elucidating the axonal transport machinery of presynaptic protein carrying vesicles, referred to as precursor vesicles (PVs) in developing neurons. During the course of this thesis I established, that PVs transport not only SV proteins but also AZ proteins and the sCAM protein Neurexin 1β collectively to the nascent presynapse in developing human iPSC-derived neurons. Anterograde trafficking of PVs is driven by the kinesin motor protein KIF1A and is regulated by the lysosomal small GTPase Arl8. Double loss of Arl8a and Arl8b in human neurons results in a significantly decreased number of anterogradely moving PVs in developing neurons and, consequently, a drastic loss in presynaptic protein levels at mature synapses. Interestingly, despite being regulated by Arl8, PVs are distinct from mature degradative lysosomes: They are non-acidic, do not harbor cathepsin activity, and use a transport machinery that is distinct from the classical KIF5B-dependent lysosomal transport machinery. Live imaging and biochemical experiments indicate that PVs presumably originate from the endolysosomal system, as they have a specific PI(3,5)P2 lipid identity which is recognized by the PH domain of KIF1A. Arl8 and PI(3,5)P2 provide a coincidence detection mechanism for KIF1A to drive PV transport by interacting directly with Arl8 via its CC3 domain and with PI(3,5)P2 through its lipid binding PH domain. PV transport is further regulated by a multi-subunit complex named BORC, which regulates lysosomal motility upstream of Arl8. BORC activity can mediate PV transport independent of Arl8 by negatively regulating PI(3,5)P2 levels. Taken together, my data unravels the lipid identity of presynaptic transport organelles and sheds light on their biogenesis and the mechanisms that underlying the regulation of the machinery for PV transport.
