UC Berkeley

Molecular motors hydrolyze ATP to produce mechanical work by stepping along the cytoskeleton network and carrying cargos. Dynein has many cellular roles which require its minus end-directed motility and force generation along the microtubules (MTs). All dynein activity needs to be tightly regulated by its many associated factors. Lis1 is the only associated factor that directly binds to dynein’s ATP hydrolyzing AAA ring, and it is involved in most, if not all, cellular processes that require dynein activity. In my thesis work, working with both mammalian and yeast proteins, I showed how dynein motility and force generation is regulated by Lis1 and its yeast homolog Pac1 (both Lis1 from here on). Mammalian dynein is mostly autoinhibited, that is, it cannot take many steps before detaching from the microtubules, a property that is essential for dynein-mediated cargo transportation. For processive motility, dynein needs to be relieved from this autoinhibition and needs to bind to dynactin and a cargo adapter. Using protein engineering, single molecule motility, optical trapping, and biophysical characterization assays, we have shown that Lis1 relieves dynein from autoinhibition, thus allowing the formation of dynein dynactin cargo adapter complex (DDX). Through the same mechanism, Lis1 increases the copy number of dynein in DDX complexes which enables faster motility and higher force generation. However, even after the formation of these complexes Lis1 can remain bound to dynein. In that case, we see an inhibitory effect of Lis1. In my thesis, I have shown the mechanism by which Lis1 binding affects dynein motility I switched my research to S. cerevisiae cytoplasmic dynein which has inherently processive motility without needing any cofactors, unlike mammalian dynein. I showed that Lis1 binding to the motor domain slows down dynein motility thus confirming previous studies done on yeast dynein and Lis1. Through multicolor TIRF colocalization assays, I have demonstrated that binding of individual Lis1 molecules causes dynein to pause or stop, and its unbinding restores dynein velocity. I have made three discoveries: 1. Lis1 binding to dynein has been proposed to inhibit or slow dynein motility by tethering dynein to the microtubule. I ruled out this model by showing that Lis1 only weakly interacts with the microtubule lattice, and this interaction does not slow dynein motility. 2. Lis1 binding has been proposed to block the force-generating conformational changes of the dynein linker domain. Using optical trapping, we ruled out this model by showing that Lis1 does not reduce the dynein stall force. 3. I observed that Lis1 binding decreases the asymmetry in detachment kinetics of force-induced detachment of dynein from the microtubule. Mutations that disrupt Lis1’s interactions with dynein’s stalk (an anti-parallel coiled-coil that leads to dynein’s microtubule-binding domain) partially restore the asymmetry. Because dynein’s stalk “slides” or changes its coiled-coil registry in a nucleotide-dependent manner, my data suggest that Lis1’s interaction with the dynein stalk interferes with the stalk sliding mechanism. I propose that this is what leads to slowing the detachment of dynein from the microtubule under force. These results are compatible with studies of Lis1 in live cells and provide a mechanistic explanation for why Lis1 needs to dissociate from dynein for efficient minus-end-directed motility. They also suggest an additional regulatory role for Lis1, such as anchoring dynein to the microtubule in order to facilitate the proper assembly of dynein with dynactin. I believe that the studies presented in this thesis will be broadly interesting to biophysicists studying the mechanics of motor proteins in vitro, cell biologists interested in the mechanism and regulation of intracellular transport, and neurobiologists who study the molecular basis of neurodevelopmental disorders.