Structural basis of Stu2 recruitment to yeast kinetochores

Essential revisions

The reviewers and editors discussed the following experimental revision that the authors should address before publication.

For the experiment in Figure 4A, a spindle pole marker should be included. The lower percentage of split signals could either be generated by monopolar/syntelic attachments or the lack of pulling forces on bioriented chromosomes. A spindle pole marker will be useful to determine if the signal lies between poles or at one of the two poles to distinguish between these models, and also to ensure that metaphase spindle length is not affected by the Stu2 mutant.

Thank you to the reviewers for suggesting this experiment. We completely agree that distinguishing monopolar/syntelic attachments versus a lack of pulling forces on bioriented chromosomes is an important point, as well as ensuring that the stu2 mutant does not affect mitotic spindle length. To address these points, we conducted two further experiments:

1) Figure 3E. We examined spindle length in metaphase arrested cells for wild-type and stu2^EEE mutant cells and found them to be indistinguishable. In this experiment, we also examined the localization of GFP-fusion wild-type and mutant variants (Figure 3C-D). Consistent with the biochemical experiments in our original manuscript, we observed defects in kinetochore-proximal GFP signal in the mutant cells.

2) Figure 4—figure supplement 5. We monitored the axial position of the unseparated sister centromeres along the length of the metaphase spindle in stu2^EEE expressing cells as suggested by the reviewers. We found that non-bioriented sister centromeres were predominantly found proximal to one of the two spindle poles in both STU2^WT and stu2^EEE cells, suggesting that biorientation defects are caused by the persistence of monopolar/syntelic attachments and not a lack of pulling forces. (see Figure 4—figure supplement 5 and accompanying in text)

The two additional essential revisions listed below should be addressable with added discussion in the text:

1) The authors should discuss the ramifications of addition of the peptide into the cryo-protectant solution. Reduction of peptide binding after the addition of the cryoprotectant could also explain "the weak peptide features on the map" if the peptide diffused away upon bringing the crystal into cryo-protectant solution.

This is a very good point, please see discussion below.

2) The authors should add some discussion regarding the finding that the C-terminal extension of the CTS domain has density in the structure even though it does not have secondary structure. Does this region contribute to a crystal contact, and if so, are the authors confident of this structure in the protein?

We thank reviewer #3 for noticing an inconsistency, due in part to a misunderstanding on our part about occupancy when drafting the manuscript initially. Our previous statement of lower occupancy was incorrect, as we realized when examining carefully the refinement results, but we failed to eliminate it in the submitted version. We have measured the Kd of the CTS peptide for Ndc80c^dwarf by competition fluorescence polarization as 48 μM, consistent with the requirement for a dimer interaction in vitro and with the likely requirement for additional, cooperative interactions in vivo. The adventitious interaction of the C-terminal segment with a neighboring molecule in the crystal lattice probably accounts not only for retention of the peptide curing the ca. 30 minute cryo-soak, but also for the non-isomorphism (but same space group) of the present crystals with those of the unliganded Ndc80c^dwarf and for the somewhat different crystallization conditions. We have added a panel to Figure 1 (Figure 1E) showing the Kd measurement and a section to Materials and methods describing it. We have also added the following two sentences to the text when describing the structure:

“This lattice interaction may account for the somewhat different unit-cell dimensions in our crystals from those of the unliganded Ndc80c^dwarf in the same space group (Valverde et al., 2016), as well as for the nearly full occupancy but slightly higher local thermal parameters, as determined by refinement, for the low affinity peptide, which did not diffuse out during the short soak in cryopreservative (see Materials and methods).”