The road less traveled to the spindle equator.

Mitosis is the process by which chromosomes move on the spindle apparatus to align and then segregate the duplicated genome evenly into two daughter cells. To achieve faithful segregation, bi-oriented attachments of the sister kinetochores to bundles of microtubules (K-fibers) emanating from opposite spindle poles must be established, and in most cases, chromosomes then congress to the spindle equator. Fundamental questions in biology are to understand how cells guide the directed movement of chromosomes towards the metaphase plate during the process of congression and how chromosomes establish bi-orientation. A prevailing idea is that chromosomes are pulled by the K-fibers that are attached to each of the two sister kinetochores until they reach a balanced position at the center of the spindle.1–3 This model assumes that the forces driving chromosome movement are generated by the dynamics of the microtubule polymers within the K-fiber, and that chromosome congression is actually a consequence, rather than a cause, of chromosome bi-orientation. However, enthusiasm for this model has wavered since the identification of motor proteins on the kinetochore that could serve to drive congression of the chromosomes4–7 and with the finding that mono-oriented chromosomes can also undergo congression.8 While it is clear that bi-oriented chromosomes can congress by utilizing dynamic microtubules, the question remains as to whether or not motor forces alone are sufficient to drive chromosome congression. This question has been challenging to answer because of the difficulty in creating a properly organized spindle that lacks K-fibers. In our recent paper,9 we showed that motor proteins are sufficient to drive chromosome congression in cells lacking K-fibers. We disrupted K-fibers through knockdown of Nuf2,10 a subunit of the Ndc80 complex responsible for kinetochore-microtubule attachment.11,12 Perturbation of components of the Ndc80 complex destabilizes K-fibers and causes complete chromosome misalignment.10 Surprisingly we found that elimination of the minus-end-directed Kinesin-14 motor HSET in this K-fiber deficient spindle partially restored chromosome alignment,9 showing that neither K-fibers nor chromosome bi-orientation are required for chromosome congression. This congression is dependent on the action of motor proteins. We found that perturbation of CENP-E, a kinetochore-associated plus-end directed motor, impaired the rescue of chromosome congression on the K-fiber-deficient spindle. Our data demonstrate that motors on the kinetochores are sufficient to drive chromosome movement to the spindle equator. The question remains as to how the chromosomes congress when K-fibers are absent and whether this mode of congression serves as a primary or back-up mechanism for chromosome alignment. Previous studies suggested that CENP-E could drive congression of mono-oriented chromosomes (chromosomes connected to K-fibers from one pole) by sliding along adjacent K-fiber microtubules.8 However in our studies, there are no adjacent K-fiber bundles for CENP-E to walk along, so how this motor is utilized to drive chromosome congression is not clear. The role of CENP-E in driving chromosome congression has been somewhat perplexing because CENP-E is the only identified plus-end directed motor associated with the kinetochore, and yet its disruption normally only results in the misalignment of a small number of chromosomes. From this observation one might conclude that CENP-E does not play a dominant role in congression. However, our data suggest that the action of CENP-E may have been underestimated. It appears that when K-fibers are deficient, the requirement for CENP-E becomes more prominent. Therefore, one possible explanation is that K-fiber-based chromosome congression mechanisms mask the requirements for CENP-E, and only when K-fibers are compromised does the function of CENP-E become predominant. Another possibility is that the requirement for CENP-E or other motor protein activity may be dependent on the type of spindle. For example, it has been shown that lateral-associated chromosome congression driven by motor proteins is especially important during meiosis.13,14 When chromosomes are properly bi-oriented, it is easy to envision how a chromosome finds the spindle equator because this is the point at which forces on the two sister kinetochores will be balanced. However, for congressing chromosomes that are not attached to microtubules of a K-fiber, how do they know where to form the metaphase plate? One possibility is that it is specified by microtubule organization in the spindle. In spindles assembled in Xenopus egg extracts, the average length of spindle microtubules is only about 40% of the spindle length, suggesting that microtubules are not long enough to overlap throughout the spindle.15 Meanwhile, microtubule minus ends concentrate around the spindle poles and are also dispersed in the spindle body, while the microtubule plus-ends are enriched around chromosomes.16 These data suggest there might only be a narrow region of microtubule overlap in the middle of the spindle that could specify where the metaphase plate will form. In this model, as chromosomes moved toward the spindle equator on uniformly organized microtubules, their motility would be disrupted when they encountered microtubules of opposite polarity (Fig. 1). Clearly, it will be critical to understand the detailed organization of spindles in the absence of K-fibers and in those spindles that are further disrupted by simultaneous inhibition of HSET. Figure 1 Chromosome congression mechanisms. The diagram shows how chromosome congression can be achieved by the coordination of microtubule (MT) dynamics and motor protein activity. The microtubules from two half spindles are marked with red and green colors to ... Overall our data suggest a new `pathway' by which chromosomes can congress. Previous studies have discussed multiple models to describe the congression process. Does a specific individual chromosome really follow a distinct pathway, or is it possible that different pathways cooperatively act on a single chromosome to drive its movement? For example, when a mono-oriented chromosome pair begins to congress, the leading kinetochore could capture a few microtubules coming from the opposite pole and become bi-oriented. As the leading kinetochore accumulates more microtubules it is likely that kinetochore-associated motors are still active and exerting a force for congression. Therefore it is likely that the process of congression is a hybrid pathway in which the individual pathways cooperate to enhance the efficiency of congression (Fig. 1). In summary, during chromosome congression both microtubule-based and motor-based forces act independently but are not mutually exclusive. Our work shows that motors are sufficient to drive chromosome movement to the metaphase plate even when K-fibers are lacking. It will be critical to understand the contributions of each of these forces during mitosis in unperturbed systems as well as to elucidate the mechanisms by which spindle organization contributes to chromosome motility.