Replicated chromosomes bind to the mitotic spindle through specialized structures — kinetochores — that assemble on centromeres. Kinetochores must be captured by spindle microtubules and transported to spindle poles for accurate chromosome segregation. By visualizing individual kinetochore–microtubule interactions in budding yeast, Tomoyuki Tanaka and colleagues have started to unravel the molecular mechanisms underlying kinetochore capture.

In this study, centromere CEN3, on chromosome III, and microtubules were marked with green and yellow fluorescent protein, respectively. Metaphase arrest was induced and, simultaneously, CEN3 was inactivated, allowing its detection at a distance from the spindle pole. Once cells had arrested, CEN3 was reactivated and its movement followed in metaphase. The authors observed that centromeres were captured laterally by microtubules that extended from the spindle pole, and moved poleward along the microtubules. After CEN3 reached the spindle pole, the green fluorescent signal split, indicating that sister centromeres bi-orient on the spindle pole.

Using mutational analysis, the authors dissected the functional roles of the different kinetochore components. Mutants of the microtubule plus-end tracking proteins Bim1, Bik1 and Stu2 and the Ran GDP/GTP exchange factor showed reduced numbers of nuclear microtubules, indicating that these factors are necessary for nuclear microtubule extension from spindle poles. By studying the kinetics of kinetochore capture in different mutants, Tanaka and co-workers identified the CBF3, Ndc80, Mtw1 and Ctf19 complexes, but not the Dam1 or Ipl1 complexes, as being required for kinetochore capture by microtubules.

As the rate of microtubule shrinkage is faster than that of poleward kinetochore transport, the plus ends of the shrinking microtubules are expected to meet the kinetochores, causing them to slide off. However, this never happens — the reason being, as Tanaka and colleagues observed, that microtubules associated with CEN3 can be 'rescued' (that is, microtubule shrinkage is converted to growth). Stu2 that is located at the plus ends of microtubules, and originates from CEN3, is believed to have a mediating role in this process.

The authors postulate that kinetochore transport along the microtubule might be mediated by an ATP-driven motor protein. Indeed, a mutant of the kinesin-14 family member Kar3 showed defective kinetochore transport. But, kinetochore transport continued in most mutant cells, indicating that other regulators are probably involved.

Finally, Tanaka and colleagues note that, even though the current study highlights some important differences between yeast and vertebrate cells, kinetochore capture is such a crucial event that the mechanisms are probably conserved.