Molecular determinants of the Ska-Ndc80 interaction and their influence on microtubule tracking and force-coupling

Abstract

Errorless chromosome segregation requires load-bearing attachments of the plus ends of spindle microtubules to chromosome structures named kinetochores. How these end-on kinetochore attachments are established following initial lateral contacts with the microtubule lattice is poorly understood. Two microtubule-binding complexes, the Ndc80 and Ska complexes, are important for efficient end-on coupling and may function as a unit in this process, but precise conditions for their interaction are unknown. Here, we report that the Ska-Ndc80 interaction is phosphorylation-dependent and does not require microtubules, applied force, or several previously identified functional determinants including the Ndc80-loop and the Ndc80-tail. Both the Ndc80-tail, which we reveal to be essential for microtubule end-tracking, and Ndc80-bound Ska stabilize microtubule ends in a stalled conformation. Modulation of force-coupling efficiency demonstrates that the duration of stalled microtubule disassembly predicts whether a microtubule is stabilized and rescued by the kinetochore, likely reflecting a structural transition of the microtubule end.

CENP-F stabilizes kinetochore-microtubule attachments and limits dynein stripping of corona cargoes

“…Nevertheless, our data are consistent with in vitro biophysical experiments that demonstrate how CENP-F can function as a force transducer (Kanfer et al, 2017;Volkov et al, 2015). This is, however, distinct from the Ska complex, which has a well-established forcetransducing role in vitro and during congression in vivo (Auckland et al, 2017;Helgeson et al, 2018;Huis In 't Veld et al, 2019b). We did not find a substantial role for CENP-F in congression and also found no additive defects when the Ska complex was depleted in CENP-F mutant cells (not depicted), thus ruling out any redundancy.…”

Hec1/Ndc80 Tail Domain Function at the Kinetochore-Microtubule Interface

“…For instance, Helgeson et al (2018) used optical trapping assays to show that Ska complexes can impart almost wildtype end-tracking activity to tail-less NDC80 complex-coated beads on depolymerizing microtubules, even under applied force. Similarly, Huis in't Veld et al (2019) demonstrated that Ska complexes can restore end-tracking activity to oligomerized tailless complexes in the absence or presence of applied force. These results raise the possibility that the Ska complex may be able to, at least in part, compensate for the Hec1 tail domain in human cells.…”

“…Furthermore, a recent study by Huis in't Veld et al (2019) investigated how the phosphorylation state of the Hec1 tail domain impacted the ability of human NDC80 complexes to maintain attachments to depolymerizing microtubules in vitro. The authors reported that while the phosphorylation state of the tail did not affect the ability of trimerized, bead-bound NDC80 complexes to track with depolymerizing microtubules in the absence of tension, when a resisting force was applied with an optical trap, phosphorylated NDC80 complexes detached from depolymerizing microtubules with significantly higher frequency than non-phosphorylated complexes (Huis in't Veld et al, 2019). These results suggest that, at least in vitro, Hec1 tail phosphorylation affects the ability of human NDC80 complexes under tension to transduce forces from depolymerizing microtubules.…”

“…A recent study using engineered scaffolds to multimerize human NDC80 complexes has provided insight into how the tail domain might influence NDC80 complex-microtubule binding (Huis in't Veld et al, 2019). The authors of this study found that bead-bound NDC80 complexes lacking the Hec1 tail exhibited almost wild-type microtubule residence times, in situations in which the tail-less NDC80 complexes were oligomerized on the bead surface (Huis in't Veld et al, 2019). However, unlike wild-type complexes, these oligomerized tail-less NDC80 complexes were unable to track depolymerizing microtubule plus-ends.…”

“…For instance, multiple studies have reported that phosphomimetic substitutions in the Hec1 tail decrease microtubulebinding affinity of single NDC80 complexes independently of their oligomerization (Umbreit et al, 2012;Zaytsev et al, 2015). Furthermore, neither tail deletion nor phospho-mimetic mutants of Hec1 affect cooperative microtubule binding (Umbreit et al, 2012;Zaytsev et al, 2015), and tail-less human NDC80 complexes can still assemble into oligomers that bind microtubules with high affinity (Huis in't Veld et al, 2019). Thus, while it remains possible that the Hec1 tail domain -and the phosphorylation state thereof -contributes to NDC80 complex oligomerization, it does not appear to be a critical effector for assembly or activity of NDC80 complex oligomers.…”