Molecular Mechanisms of Mitochondrial Transport in Neurons

Abstract

Dynamic mitochondrial transport into axons and dendrites of neuronal cells is critical for sustaining neuronal excitability, synaptic transmission, and cell survival. Failure of mitochondrial transport is the direct cause of some neurodegenerative diseases, and an aggravating factor for many others. Mitochondrial transport regulation involves many proteins; factoring prominently among them are the atypical mitochondrial GTPase Miro and the Milton/TRAK adaptor proteins, which link microtubule (MT) motors to mitochondria. Motors of the kinesin family mediate mitochondrial transport towards the plus ends of microtubules, while motors of the dynein family mediate mitochondrial transport towards the minus ends. Selective use of these motors determines the ultimate subcellular distribution of mitochondria, but the underlying control mechanisms remain poorly understood. Drosophila Miro (dMiro) is required for kinesin-driven transport of mitochondria, but its role in dynein-driven transport remains controversial. In Chapter 2 of this study, I show that dMiro is also required for the dynein-driven transport of mitochondria. In addition, we used the loss-of-function mutations dMiroT25N and dMiroT460N to analyze the function of dMiro's N- and C-terminal GTPase domains, respectively. We show that dMiroT25N causes lethality and impairs mitochondrial distribution and transport in a manner indistinguishable from dmiro null mutants. Our analysis suggests that both kinesin- and dynein-driven mitochondrial transport require the activity of Miro's N-terminal GTPase domain, which likely controls the transition from a stationary to a motile state irrespective of the transport direction. dMiroT460N reduced only dynein motility during retrograde axonal transport but had no effect on distribution of mitochondria in neurons, indicating that the C-terminal GTPase domain of Miro most likely has only a small modulatory role on transport. Furthermore, we show that commonly used substitutions in Miro's GTPase domains, based on the constitutively active Ras-G12V mutation, appear to cause neomorphic phenotypic effects which are probably unrelated to the normal function of the protein. In mammalian neurons, kinesin and dynein motors are linked to mitochondria via a Miro complex with the adapter proteins TRAK1 and TRAK2, respectively. Differential linkage of TRAK-motor complexes provides a mechanism for determining the direction of transport and controlling mitochondrial distributions within the cell. Drosophila has only one TRAK gene homolog, Milton, which expresses several protein isoform. Milton has been previously been shown to facilitate mitochondrial transport by binding to kinesin and dMiro, a role analogous to TRAK1. However, the question whether Milton might be able mediate dynein-based transport in a manner similar to TRAK2 has remained unknown. In Chapter 3 of this study, I show that protein isoforms A and B of Milton, generated through alternative mRNA splicing, facilitate differential motor activities analogous to mammalian TRAKs. Specifically, overexpression (OE) of Milton-A caused a mitochondrial redistribution and accumulation at axon terminals, which requires kinesin-driven MT plus end directed transport; while OE of Milton-B caused a redistribution of axonal mitochondria into the soma, which requires dynein-driven MT minus end directed transport. I further show that Milton-motor complex binding to mitochondria requires Miro exclusively, and that transport with either of the motor complexes absolutely requires the activity of Miro's N-terminal GTPase domain. Together, these results suggest that Miro controls the transition of mitochondria from a stationary to a motile phase. Thereafter the direction of transport is likely determined by an alternative binding of opposing Milton/TRAK-motor complexes to Miro, a process which appears to be regulated by a Miro-independent mechanism.