A stretched merotelic kinetochore (green) segregates toward the pole with which it has the strongest attachment (right).


Merotelic chromosome attachments present the cell with a particularly threatening scenario. With a single kinetochore attached to both spindle poles, the tension on this kinetochore ensures that no error message is sent out, as would be the case with a detached kinetochore. Most researchers presumed that the relatively lower tension on the “incorrect” attachment (the one that is relieved by the codirectional pull on the singly attached sister kinetochore) ensures detachment of that connection, followed by attachment to the correct pole.

But now Daniela Cimini, Lisa Cameron, and Ted Salmon (University of North Carolina, Chapel Hill, NC) show that this detachment process rarely makes it to completion. Instead, a tension-dependent elongation of the incorrect attachment allows the chromosome to reach the correct side during anaphase.

The group first determined that only 8 out of 54 merotelic attachments were corrected during metaphase. This may not be surprising—the likelihood of all kinetochore microtubules letting go at one time is presumably very low, and newly released sites will often still be oriented toward and thus reattach to the incorrect pole. The tension differential did, however, result in more microtubules attaching to the correct rather than incorrect pole for most merotelic situations.

Once anaphase began, this difference was crucial. With the sister kinetochore departed for its pole, the merotelic kinetochore experienced the same force at both of its attachments. But the incorrect attachments, with fewer microtubules, experienced greater tension per microtubule as the spindle poles separated. Microtubule polymerization rate is thought to be proportional to tension, which would explain why the thinner microtubule bundles got relatively much longer. This put the chromosomes on the side of the spindle where they needed to be. Thus, says Cimini, “you get correct chromosome segregation without [fully] correcting the error of microtubule attachment.”


Cimini, D., et al.
Curr. Biol.