page 671, Khodjakov et al. uncover a new mechanism of spindle morphogenesis that involves the capture of microtubule minus ends.
In the established model of spindle formation in vertebrate cells, spindle microtubule minus ends are focused at spindle poles in association with centrosome-nucleated microtubules. Kinetochore fibers, which pull chromosomes to opposite poles, form when the plus ends of centrosome-associated microtubules capture kinetochores. A second mechanism must also form kinetochore fibers, however, because functional spindles form in the absence of centrosomes.
In their new article, the authors got a close look at this second mechanism by focusing on unobstructed kinetochores. They avoided the obtrusive mass of non-kinetochore microtubules that usually blocks imaging by inducing monopolar spindle formation and examining kinetochores that faced away from the centrosome. Fully formed bundles of microtubules occasionally appeared near the kinetochores. These bundles became attached at their plus ends to the kinetochore in the end-on arrangement typical of kinetochore fibers, but their minus ends remained oriented away from the centrosomes. The growing minus ends looped back as if captured by spindle microtubules and were incorporated into the spindle. This strategy is not limited to monopolar spindles, as looping and capture of kinetochore-anchored microtubules was also seen during bipolar spindle formation.
The loop-and-capture behavior depended on NuMA, a minus end–localized protein that associates with dynein. Thus, dynein may be the motor that pulls kinetochore fiber minus ends toward centrosomes along tracks provided by centrosome-induced microtubules. Centrosomes, therefore, may be critical to orient the axis of division by drawing in microtubule minus ends, irrespective of the direction the kinetochores face. This function is consistent with results from centrosome removal experiments, which often result in misorientation of the division plane. ▪