Harvard University COVID-19 updates

Department News



(L to R) Sirui Zou, Bret Redwine, Andres Leschziner, Rogelio Hernandez-Lopez, and Samara Reck-Peterson, front row Locksley

Walking takes an enormous amount of coordination. As your leading foot hits the ground a signal must be sent to the brain to let it know this has happened. With this foot safely in place, the brain can tell the lagging foot take a step before the cycle is repeated. This coordination benefits both from a nervous system and the convenient fact that gravity keeps us safely attached to the ground. Molecular motors, protein machines that “walk” along cytoskeletal tracks in a loosely anthropomorphic way, accomplish a similar level of coordination. However, they must do this using entirely mechanochemical means. Furthermore, these motors operate at scales where gravity has little effect and must therefore tune their own affinity for the track as they step on or off it.
       Of the three families of molecular motors—dyneins, kinesins and myosins—dynein is by far the largest and most complex, its basic motile unit being a dimer of two identical 500 kDa subunits. Cytoplasmic dynein is responsible for moving a variety of cargo from the cell periphery towards the nucleus along microtubules and plays roles in mitosis and meiosis. Each dynein monomer has a ring-shaped, force-generating motor domain; out of it comes a long antiparallel coiled-coil termed the “stalk”, which is capped by a small alpha helical domain responsible for binding to microtubules. One of the many intriguing aspects about dynein is how ATP hydrolysis in the motor domain, which powers and controls walking, is coupled to microtubule binding and release given the 25 nm that separate these two domains. It was already known that the two alpha helices in the stalk can slide past each other to adopt different registers and that these registers result in different affinities between dynein and the microtubule. This sliding could explain the coordination between the microtubule-binding domain and the motor domain but not how microtubule binding is sensed by dynein or how it triggers the molecular changes that eventually lead to sliding in the stalk.
       In a paper recently published in Science, the groups of Andres Leschziner in MCB and Samara Reck-Peterson in the department of Cell Biology at Harvard Medical School teamed up to address this question using a combination of high-resolution cryo-electron microscopy (cryo-EM), molecular dynamics simulations and single-molecule approaches. The team, composed of Bret Redwine, Rogelio Hernandez-Lopez, Sirui Zou and Julie Huang, obtained the structure of dynein’s microtubule-binding domain bound to microtubules at a resolution of 10Å. Since alpha helices can be seen at this resolution, they were able to use molecular dynamics to generate a pseudo-atomic model of the complex by starting with the available crystallographic structure of a free microtubule-binding domain and fitting it into the cryo-EM density.
       The new structure showed a dramatic conformational change in one of the alpha helices of the microtubule-binding domain, which moved 10Å relative to the free form. The new position of this helix, stabilized by additional interactions with the microtubule, pushes against and moves one of the two alpha helices in the coiled-coil, suggesting how microtubule binding is detected by this “sensor” helix and in turn is coupled to events in the motor domain via changes in the register of the stalk.
       Unexpectedly, the molecular dynamics simulations, which provide frame-by-frame “movies” of the atoms in the structure through time, identified a couple of interactions whose purpose appeared to be to lower dynein’s affinity for microtubules. Both interactions were intramolecular salt bridges that competed with intermolecular interactions between the microtubule-binding domain and the microtubule. In fact, removal of either of these competing interactions resulted in a motor capable of walking along microtubules for distances 5 to 6 times longer than the wild type, the longest any dynein has ever been seen to walk. Interestingly, these affinity-dampening amino acids are universally conserved in cytoplasmic dynein. Why is being able to walk over long distances without falling off so deleterious that selective pressure has kept these changes from becoming established? Current work in the Leschziner and Reck-Peterson labs is aimed at answering this question.
       The coordination between the cycles of nucleotide hydrolysis and microtubule binding and release is likely to be an important nexus for regulating dynein. In a separate paper published in Cell, the Leschziner and Reck-Peterson groups looked at Lis1, a ubiquitous regulator of dynein that is mutated in type I lissencephaly, a severe brain developmental disorder. Julie Huang and Anthony Roberts showed that Lis1 acts as a “clutch”, uncoupling ATP hydrolysis from microtubule release and thus allowing the motor to remain bound to its track for long periods regardless of what the motor domain is doing. This may be important for some of dynein’s biological functions, such as “waiting” at microtubule plus ends to be loaded onto cargo and bearing high loads. Interestingly, a structural analysis of the dynein-Lis1 complex by electron microscopy showed that Lis1 binds on the motor domain next to where dynein’s stalk protrudes, the one point where conformational changes connecting the two would have to take place. The two groups are now trying to understand, mechanistically, how Lis1 accomplishes this uncoupling.

Read more in Science

Read more in Cell

Read more in Harvardgazette

Watch Video: “Structural Model for the Binding of Dynein to a Microtubule”