Because of their large size and highly extended morphology, neurons incur a substantial burden of intracellular movement and transport. These processes are crucial to neuronal viability and function, and can cause neuronal death and neurodegeneration when they are disrupted. My laboratory is interested in understanding the logic and molecular mechanisms of these neuronal transport systems, with the related goal of elucidating the likely role of transport failure in neurodegenerative diseases.
Previous work revealed that most intracellular movement in neurons is generated by motor proteins such as kinesin, dynein, or myosin. These proteins convert the chemical energy in ATP into directed transport along a cytoskeletal filament such as a microtubule or microfilament. Of critical importance is an understanding of how these motor proteins attach to appropriate neuronal cargoes, how their activities are regulated, and how appropriate intracellular destinations are found.
Our current efforts are directed at attacking five interrelated questions: (1) What role(s) do kinesin motor proteins play in axonal transport, transport of visual system components in photoreceptors, and transport of informational signaling molecules? (2) Does failure of motor-driven transport play a major role in neurodegenerative diseases such as retinitis pigmentosa, Alzheimer's disease, and Huntington's disease? (3) How are kinesin motor proteins coupled to intracellular cargoes and regulated? (4) How are appropriate destinations in the neuron found (e.g., axons versus dendrites)? (5) Do intracellular transport processes play important roles in neuronal cell polarization, signaling, growth, and pathfinding? To address these questions, we use an array of genetic and molecular strategies in Drosophila and mice to identify new genes playing important roles in transport and to understand the molecular and cellular functions of genes identified by other means.
To elucidate the transport pathways dependent on conventional kinesin, called kinesin-I, we generated a series of mutants in which genes encoding critical subunits are deleted. Kinesin-I is composed of a catalytic force-producing motor subunit (KHC, kinesin heavy chain) and an accessory subunit (KLC, kinesin light chain) thought to play roles in attachment or regulation of KHC. In mammals, KHC is encoded by three different genes, KIF5A, KIF5B, and KIF5C. KIF5A and KIF5C are only expressed in neurons, and KIF5B is used in nonneuronal cells.
To probe the functions of KIF5A and KIF5C, we made mouse mutants lacking these genes completely and conditional mutants in which the Kif5a gene is selectively ablated in defined target cells. Although Kif5c mutants are viable and appear phenotypically normal, Kif5a mutants die at birth. We have also removed Kif5a in neurons after birth and find that such animals die at 2–3 weeks of age, with prominent neuromuscular defects and substantial seizures. While fast axonal transport looks normal in these animals, there appear to be defects in neurofilament transport in some neurons, which is a process usually thought to be part of "slow" axonal transport. (A grant from the National Institutes of Health provides support for this work.)
To identify potential receptors for kinesin motor protein binding on intracellular cargoes and to find likely regulatory components for these motor proteins, we used genetic screens in Drosophila and biochemical and genetic analyses in the mouse. Previous work showed that various kinesin and dynein mutants have characteristic phenotypes, which can be used to find other mutants. In one large screen of this sort, several candidate genes were found that when mutant give phenotypes identical to mutants lacking KHC, KLC, and some dynein subunits.
One mutant, sunday driver, appears to identify an evolutionarily highly conserved protein that has many of the properties expected of a kinesin-I receptor. This protein is associated with membrane vesicles in axons. Intriguingly, sunday driver is the homolog of a signal scaffold protein called JIP-3 in mammals that can bind multiple components of a JNK-signaling pathway. Further work now suggests that kinesin motor proteins may be required to localize signaling complexes in neurons and that the signaling complexes themselves may attach kinesin motor proteins to vesicular cargoes, and perhaps regulate their activities. In addition, we observed that axonal damage activates JNK signaling in axons; we are currently searching for the retrograde signal carrier. (A grant from the National Institutes of Health provides support for this work.)
In a complementary biochemical approach, we obtained data suggesting that the amyloid precursor protein (APP), whose normal functions have been mysterious, may play a role in neurons as a vesicular receptor protein for kinesin-I motor proteins. This protein is of great interest because mutant versions cause familial Alzheimer's disease, and proteolytic processing of APP appears to give rise to the insoluble peptides that aggregate and form the characteristic plaques found in disease. Further genetic work in Drosophila and mouse provided strong evidence that APP is required in vivo to move a particular class of axonal vesicle that may contain the processing machinery for APP. Strikingly, mouse models of Alzheimer's disease in which APP is overexpressed exhibit axonal transport defects in cholinergic neurons and interact genetically with genetic reductions of kinesin function. Thus, alterations in kinesin-I–based transport of APP may play a role in neuronal death and perhaps pathogenesis of Alzheimer's disease. We are initiating tests of this hypothesis in humans. In related work, we are examining the hypothesis that proteins implicated in polyglutamine neurodegenerative disease, such as Huntington's disease, may induce disease by impairing normal neuronal transport.