Fig 2: The Ndc80 complex coordinates with dynein for efficient kMT capture.A, immunofluorescence staining of MG132 + ZM treated mitotic HeLa cells depleted of the indicated proteins and stained for α-tubulin (green), a kinetochore marker (Zwint1 or CENP-A, red) with the chromosomes counterstained using DAPI. Bars, 5 μm. B, quantification of mitotic cells with null kinetochores from A. Error bars represent S.D. from three independent experiments. For each experiment, 200 mitotic cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). C, control (Ig) and anti-dynein IC immunoprecipitations (IPs) were performed using whole cell extracts prepared from nocodazole-treated prometaphase cells and the western blots of the IPs were probed with anti-dynein and anti-Ndc80 antibodies. D, analyses of dynein localization to kinetochores in Ndc80-depleted cells after Aurora B inhibition. Prometaphase HeLa cells were either treated with Ndc80 siRNA alone (middle panel) or in combination with ZM (bottom panel) as compared to untreated controls (top panel), followed by fixation and immunostaining for the dynein intermediate chain (DIC, green), a kinetochore marker (CREST, red) with the chromosomes counterstained using DAPI. Bars, 5 μm. E, quantification of dynein intensity for data in D for Ndc80-depleted cells after Aurora B inhibition as compared to the corresponding controls. A total of 50 kinetochores were analyzed from at least five different prometaphase cells. F, immunofluorescence staining of mitotic HeLa cells depleted of either CENP-E (top panel), dynein (second panel) or Spindly (third panel) alone or codepleted of either Ndc80 + Spindly (fourth panel) or Ndc80 + CENP-E (bottom panel) and stained for α-tubulin (green), a kinetochore marker (Zwint1, red) with the chromosomes counterstained using DAPI. Bars, 5 μm. G, quantification of mitotic cells with null kinetochores from E. Error bars represent S.D. from three independent experiments. For each experiment, 200 mitotic cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). CENP-E, centrosome-associated protein E; kMT, kinetochore-microtubule; Ndc80, nuclear division cycle 80.

Fig 3: A cartoon depicting the coordinated action of Ndc80 and dynein in the formation of initial kMT attachments during early mitosis. During the kMT capture process, Ndc80 at the kinetochore is involved in attaching to the microtubules, not only via dynamic lateral kMT contacts but also via coordination with dynein to form dynamic end-on contacts (A). The kinetochore motors, dynein (A, B) and CENP-E (not depicted), assist Ndc80 by making lateral kMT contacts and driving kinetochore motility. In the absence of the Ndc80 complex, only about ∼30% of lateral kMT contacts are made, mediated by dynein and/or CENP-E (B). Once the initial attachment is accomplished, all kinetochores depend on dynein for spindle poleward motility, an event that is also needs to be coordinated with the Ndc80 complex to constitute a productive kMT capture event (A). Loss of function of both the Ndc80 complex as well as dynein leads to the vast majority of kinetochores being not captured (C) and contributing to the observed kinetochore null phenotype. CENP-E, centrosome-associated protein E; kMT, kinetochore-microtubule; Ndc80, nuclear division cycle 80.

Fig 4: Characterization of the kinetochore-null phenotype induced by Ndc80 inhibition.A, cells were subjected to the indicated perturbations, treated with the indicated drugs/chemicals, and fixed according to the scheme. B and C, mitotic HeLa cells depleted of Ndc80 (B, panel) or knocked out of Nuf2 (C bottom panel) as compared to controlRNAi (B, top panel) or parental control cells (B, bottom panel), and followed by the indicated drug treatments, were immunostained for α-tubulin (green), a kinetochore marker Zwint1 or CENPA (red) with the chromosomes counterstained using DAPI. Bars, 5 μm. D, quantification of the frequency of mitotic cells with null kinetochores (see main text for more details) in samples from B and D. Error bars represent S.D. from three independent experiments. For each experiment, 200 mitotic cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). E, immunofluorescence staining of mitotic RPE1 cells depleted of Nuf2 in comparison to control cells (top panel) and stained for α-tubulin (green), a kinetochore marker Zwint1 (red) with the chromosomes counterstained using DAPI. Bars, 5 μm. F, quantification of the status of kMT attachments in cells from E. Error bars represent S.D. from three independent experiments. For each experiment, 200 mitotic cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). G, selected frames from live imaging of HeLa cells stably expressing mCherry-Histone H2B to visualize the chromosomes in addition to GFP-α-tubulin and GFP-CENPA to visualize kMT attachment in controlRNAi (top two panels) and Ndc80RNAi cells (bottom two panels). Images were captured at every 1 min interval starting from the point of nuclear envelope breakdown for around 30 min until a proper mitotic spindle was formed during prometaphase in controlRNAi cells. Bars, 5 μm. H, quantification of mitotic cells with null kinetochores from G. Error bars represent S.D. from three independent experiments. For each experiment, 50 mitotic cells were examined. I, immunofluorescence staining of MG132 + ZM treated mitotic HeLa cells depleted of endogenous Ndc80 and transfected with different Hec1-GFP constructs as indicated and stained for GFP (green), α-tubulin (red) with the chromosomes counterstained using DAPI. Bars, 5 μm. J, quantification of mitotic cells with null kinetochores from A. Error bars represent S.D. from three independent experiments. For each experiment, 200 mitotic cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). kMT, kinetochore-microtubule; Ndc80, nuclear division cycle 80; ZM, ZM447439.

Fig 5: Characterization of different microtubule-binding domains within the Ndc80 complex required for the initial kMT attachments.A, cells were subjected to the indicated perturbations, treated with the indicated drugs/chemicals, and fixed according to the scheme. B, a cartoon showing the Ndc80 complex with different domains of the Hec1 subunit that have been implicated in kinetochore-microtubule (kMT) attachments. C, immunofluorescence staining of STLC-treated mitotic HeLa cells depleted of endogenous Ndc80 and transfected with different Hec1-GFP constructs as indicated and stained for GFP (green), α-tubulin (red) with the chromosomes counterstained using DAPI. Bars, 5 μm. Inset shows the kMT attachment status of individual kinetochore pairs in the conditions indicated. Bar (applies to all insets), 1 μm. D, quantification of the status of kMT attachments in cells from C. Error bars represent S.D. from three independent experiments. For each experiment, on average ∼20 kinetochore pairs from ten monopolar cells were examined. ∗∗∗∗p < 0.0001 (Student’s t test). Ndc80, nuclear division cycle 80; STLC, S-Trityl-l-Cysteine.