An EB1-kinesin complex is sufficient to steer microtubule growth in vitro.

Proper microtubule polarity underlies overall neuronal polarity, but mechanisms for maintaining microtubule polarity are not well understood. Previous live imaging in Drosophila dendritic arborization neurons showed that while microtubules are uniformly plus-end out in axons, dendrites possess uniformly minus-end-out microtubules [1]. Thus, maintaining uniform microtubule polarity in dendrites requires that growing microtubule plus ends entering branch points be actively directed toward the cell body. A model was proposed in which EB1 tracks the plus ends of microtubules growing into a branch and an associated kinesin-2 motor walks along a static microtubule to steer the plus end toward the cell body. However, the fast plus-end binding dynamics of EB1 [2-5] appear to be at odds with this proposed mechanical function. To test this model in vitro, we reconstituted the system by artificially dimerizing EB1 to kinesin, growing microtubules from immobilized seeds, and imaging encounters between growing microtubule plus ends and static microtubules. Consistent with in vivo observations, the EB1-kinesin complex actively steered growing microtubules. Thus, EB1 kinetics and mechanics are sufficient to bend microtubules for several seconds. Other kinesins also demonstrated this activity, suggesting this is a general mechanism for organizing and maintaining proper microtubule polarity in cells.