Proximity-based activation of AURORA A by MPS1 potentiates error correction

SUMMARY

Faithful cell division relies on mitotic chromosomes becoming bioriented with each pair of sister kinetochores bound to microtubules oriented toward opposing spindle poles. Erroneous kinetochore-microtubule attachments often form during early mitosis but are destabilized through the phosphorylation of outer kinetochore proteins by centromeric AURORA B kinase (ABK) and centrosomal AURORA A kinase (AAK), thus allowing for re-establishment of attachments until biorientation is achieved^{1,2,3,4,5,6,7,8,9}. MPS1-mediated phosphorylation of NDC80 has also been shown to directly weaken the kinetochore-microtubule interface in yeast¹⁰. In human cells, MPS1 has been proposed to transiently accumulate at end-on attached kinetochores¹¹ and phosphorylate SKA3 to promote microtubule release¹². Whether MPS1 directly targets NDC80 and/or promotes the activity of AURORA kinases in metazoans remains unclear. Here, we report a novel mechanism involving communication between kinetochores and centrosomes, wherein MPS1 acts upstream of AAK to promote error correction. MPS1 on pole-proximal kinetochores phosphorylates the C-lobe of AAK thereby increasing its activation at centrosomes. This proximity-based activation ensures the establishment of a robust AAK activity gradient that locally destabilizes mal-oriented kinetochores near spindle poles. Accordingly, MPS1 depletion from Drosophila cells causes severe chromosome misalignment and erroneous kinetochore-microtubule attachments, which can be rescued by tethering either MPS1 or constitutively active AAK mutants to centrosomes. Proximity-based activation of AAK by MPS1 also occurs in human cells to promote AAK-mediated phosphorylation of the NDC80 N-terminal tail. These findings uncover an MPS1-AAK cross-talk that is required for efficient error correction, showcasing the ability of kinetochores to modulate centrosome outputs to ensure proper chromosome segregation.

eTOC Blurb

Leça et al demonstrate that MPS1 on pole-proximal kinetochores phosphorylates AAK, enhancing its activation and potentiating a local gradient of AAK activity that destabilizes incorrect kinetochore-microtubule attachments. This finding highlights a crucial interplay between kinetochores and centrosomes in promoting error correction during mitosis.

Graphical Abstract

RESULTS

While it has long been appreciated that MPS1 contributes to error correction in human cells, defining whether it plays a direct or indirect role in the process has proven difficult. Initial evidence suggested that MPS1 could contribute to error correction indirectly via activation of centromeric AURORA B kinase (ABK)^13,14,15, however, other studies concluded that MPS1 acted downstream of ABK¹⁶ and reported that MPS1 inhibition did not reduce ABK’s ability to phosphorylate known substrates in cells^17,18. A direct role for kinetochore-associated MPS1 in error correction was later supported by the identification of kinetochore-microtubule attachment factors as MPS1 substrates in human and budding yeast. In humans, a pool of MPS1 localizes to end-on attached kinetochores prior to error correction¹¹ and MPS1-mediated phosphorylation of the SKA complex, which accumulates at end-on attachments to stabilize them, hinders its association with dynamic microtubules in vitro and is required for error correction in cells¹². Similarly, in budding yeast, phosphorylation of the attachment factor NDC80 by kinetochore-associated MPS1 weakens the binding of kinetochore particles to microtubules in vitro and contributes to error correction and accurate chromosome segregation in vivo¹⁰. Thus, efficient error correction in yeast and humans relies on direct inputs from both MPS1 and AURORA kinases via phosphorylation of kinetochore-microtubule attachment factors. However, whether MPS1 directly targets NDC80 and/or promotes the activity of AURORA kinase in metazoans are open questions.

Drosophila melanogaster offers a unique model to further dissect error correction because key players in the pathway (MPS1, AURORA kinases and the NDC80 complex) are conserved but kinetochore composition in flies is divergent relative to human and budding yeast in notable ways. For example, flies do not have a functional SKA complex but, like humans, possess centrosome-enriched AURORA A kinase (AAK) and centromere/kinetochore-localized ABK while budding yeast has a single AURORA kinase orthologue (Ipl1). AURORA kinases play a conserved role in error correction by targeting NDC80, but the localization of AAK and ABK imparts a spatial component to error correction that is likely absent in budding yeast. Indeed, AAK contributes to a polar error correction pathway in flies and mammals via phosphorylation of NDC80 at pole-proximal kinetochores^7,8. In flies, MPS1 also localizes to microtubule-attached kinetochores¹⁹, but if it contributes to error correction then it does so in the absence of the SKA complex and in the presence of spatial cues from AURORA kinases.

Polar chromosomes promote AAK activation at centrosomes in an MPS1-dependent manner

The activation status of AURORA kinases throughout the cell cycle was assessed in D. melanogaster S2 cells using antibodies against the phosphorylated T-loops of ABK (ABK^T209Ph) and AAK (AAK^T311Ph), the latter of which was distinguished from cross-reactive ABK^T209Ph by its colocalization with the Drosophila Pericentrin-Like Protein (DPLP). Active AAK^T311Ph was undetectable at interphase centrosomes but increased significantly during mitosis peaking in prophase and declining from prometaphase through metaphase and anaphase (Figures 1A and 1B). MG132-treated cells in prometaphase or co-treated with colchicine to depolymerize microtubules exhibited comparable levels of AAK^T311Ph that were significantly higher than AAK^T311Ph levels in MG132-arrested metaphase cells (Figure S1A and S1B). This suggests that AAK activation is increased by the presence of mis-aligned and unattached kinetochores, regardless of mitotic duration. Interestingly, robust activation of AAK required MPS1 activity as AAK^T311Ph levels at centrosomes remained low throughout mitosis in cells depleted of MPS1 relative to controls (Figures 1A, 1B and S1C).

FIGURE 1. Polar chromosomes promote AAK activation at centrosomes in an MPS1-dependent manner.

(A) Representative immunofluorescence images of AAK T311 phosphorylation (AAK^T311Ph) in control and MPS1-depleted Drosophila S2 cells in the indicated cell cycle stages. Insets depict magnifications of selected centrosomes.

(B) Quantification of the levels of AAK T311 phosphorylation (AAK^T311Ph) relative to DPLP at centrosomes of control and MPS1-depleted Drosophila S2 cells in the indicated cell cycle stages, as presented in (A). All values were normalized to the mean value determined for prometaphase control cells, which was set to 1. n ≥ 15 centrosomes for each condition from 2 biological replicates.

(C) Representative immunofluorescence images of AAK T311 phosphorylation (AAK^T311Ph) in control, MPS1-depleted or NDC80-depleted Drosophila S2 cells treated with 20 μM MG132 for 3 h and 30μM colchicine for 2h. Insets depict magnifications of selected centrosomes. Technical replicates were also stained for MPS1 and CID to quantify the kinetochore levels of MPS1. The quantification and the corresponding representative immunofluorescence images are presented in Figure S1D and S1E, respectively.

(D) Quantification of the levels of AAK T311 phosphorylation (AAK^T311Ph) relative to DPLP at centrosomes of control, MPS1-depleted and NDC80-depleted Drosophila S2 cells treated with 20 μM MG132 for 3 h and 30 μM colchicine for 2h in the indicated conditions, as presented in (C). All values were normalized to the mean value determined for control cells, which was set to 1. n ≥ 72 centrosomes for each condition from 2 biological replicates.

(E) Quantification of the levels of AAK T311 phosphorylation (AAK^T311Ph) relative to DPLP at each centrosome vs the 3D distance between that centrosome and the closest kinetochore (measured as indicated in the schematic representation at the left) of control Drosophila S2 cells treated with 20 μM MG132 for 3 h or cells treated with and 20 μM MG132 for 3 h and 30 μM colchicine for 2h, as presented in (Figures S1A and S1B). Centrosomes to kinetochore (C-KT) distances were organized in 3 different column ranges: less than 1μm (<1); between 1μm and 3μm (1-3) and above 3μm (>3). All values were normalized to the mean value determined for prometaphase control cells treated with 20 μM MG132 for 3 h, which was set to 1. n ≥ 17 centrosomes for each condition from 2 biological replicates.

(F) Schematic of the centrosome-targeted CM2-AURORA FRET sensor used in (G) and (H).

(G) Representative images of the FRET reporter in a control cell with one polar chromosome (white arrow) near the left centrosome. The FRET emission ratio images “FRET/mTurquoise2” are pseudo-colored, according to the color wedge in display. All images are from a single selected Z-slice in the indicated channels.

(H) Quantification of the levels of FRET emission relative to total mTurquoise2 levels at centrosomes of cells and representative FRET emission ratio images “FRET/mTurquoise2” pseudo-colored in the indicated conditions. All images are from a single selected Z-slice and depict magnifications of selected centrosomes from the indicated conditions. All values were normalized to the mean value determined for prometaphase control cells, which was set to 1. n ≥ 32 centrosomes for each condition from 3 biological replicates.

Data information: Data in (B), (D), (E) and (H) are presented as median ± interquartile range. Statistical analysis was performed using a Kruskal–Wallis test for multiple comparisons. P values: ns, not significant; *< 0.05; **< 0.01; ***< 0.001; ****< 0.0001. Scale bar: 5 μm. See also Figure S1.

The effect of MPS1 depletion on AAK activation was next examined following depolymerization of microtubules with colchicine since MPS1 becomes enriched at unattached kinetochores via direct interaction with the NDC80 complex^{20,21,22,23,24,25}. High levels of AAK^T311Ph were observed in control cells, but AAK activation was significantly compromised following depletion of either MPS1 or NDC80, which mis-localized MPS1 from kinetochores (Figures 1C, 1D and S1D). Thus, kinetochore-associated MPS1 at mis-aligned and unattached kinetochores was required for the efficient activation of centrosomal AAK in mitosis. It was noteworthy that the effect of inhibiting MPS1 was specific to AAK because ABK^T209Ph levels at centromeres remained high throughout mitosis or following colchicine treatment in MPS1- and NDC80-depleted cells relative to control RNAi-treated cells (Figures S1E–S1H). Consistent with impaired AAK activity, the phosphorylation of D-TACC on S863, an established substrate of AAK²⁶, and the accumulation of γ-Tubulin^27,28,29, were reduced at centrosomes of MPS1-depleted prometaphase cells (Figure S1I–S1K).

Centrosomal AAK^T311Ph levels were plotted as a function of distance to the nearest kinetochore to assess whether there was a spatial component to the MPS1-dependent activation of AAK. There was a clear correlation between the extent of AAK activation at centrosomes and proximity to kinetochores such that higher levels of centrosomal AAK^T311Ph were measured when kinetochores were present within a ~1 μm radius of the centrosome (Figure 1E). While the spatial effect was observed in both colchicine and untreated cells, the extent of kinetochore-proximal AAK activation was generally higher in colchicine treated cells consistent with there being elevated levels of MPS1 at unattached kinetochores.

A FRET-based biosensor for AURORA kinase activity^8,30 was next used to study if MPS1-dependent activation of centrosomal AAK affected phosphorylation of downstream substrates. The reporter, which undergoes a conformational change upon phosphorylation that results in reduced FRET efficiency³¹, was tethered to centrosomes via an N-terminal fusion to the CM2 domain of Centrosomin³² (Figure 1F). Centrosome targeting put the reporter in close proximity to AAK, but higher phosphorylation (lower FRET) of the reporter was observed at centrosomes with nearby polar chromosomes (Figure 1G). Quantification of FRET emission ratios in prometaphase revealed a significant decrease in phosphorylation (higher FRET) of the reporter following MPS1 depletion (Figure 1H and S1C). FRET emission ratios were comparable in control and MPS1 over-expressing (MPS1 OE) prometaphase cells suggesting that the centrosome-targeted sensor approaches maximal phosphorylation in prometaphase. Phosphorylation of the reporter in MPS1 over-expressing prometaphase cells required AAK activity since treatment with the inhibitor MLN8237 at a concentration previously determined to specifically inhibit AAK in Drosophila S2 cells⁸ decreased phosphorylation of the reporter to the same extent as that measured in MPS1-depleted cells (Figure 1H). Consistent with the prior observation of decreased AAK activation at centrosomes in metaphase (Figures 1A and 1B), the CM2-fused reporter exhibited reduced phosphorylation at metaphase centrosomes of control and MPS1-depleted cells (Figure 1H). However, over-expression of MPS1 increased phosphorylation of the reporter in metaphase to comparable levels as that measured in prometaphase cells (Figure 1H). Taken together these findings support the conclusion that the proximity-based activation of AAK at centrosomes by MPS1 potentiates the phosphorylation of downstream AAK substrates.

MPS1-mediated phosphorylation of AAK C-lobe potentiates AAK auto-activation

In vitro kinase assays with purified human orthologues of His-AAK and GST-MPS1 were employed to gain mechanistic insights into how MPS1 activates AAK. Co-incubation of AAK with MPS1 resulted in a ~4-fold increase in the phosphorylation of AAK activation loop (AAK^T288Ph, equivalent to Drosophila AAK^T311Ph), which was reduced by half in the presence of the MPS1 inhibitor Cpd-5³³ (Figures 2A and 2B). The AAK inhibitor MLN8237 prevented MPS1-induced AAK activation indicating that MPS1 does not directly phosphorylate the T-loop (T288) but rather stimulates its auto-phosphorylation. Mass spectrometry of in vitro phosphorylated AAK identified a putative MPS1 target site located at a conserved residue (T337/T360 in human and Drosophila AAK respectively) that is predicted by AlphaFold3 to be positioned in the C-lobe αG-helix with close proximity to the T-loop (Figure S2A – S2C). We raised a phospho-specific antibody that confirmed by immunoblotting the phosphorylation of AAK^WT at T337 (T337Ph) upon incubation with MPS1 in vitro (Figure S2D). As expected, the T337Ph signal failed to be detected when T337 was converted to alanine (AAK^T337A) or when Cpd-5 was added to the reaction (Figure S2D). To test if MPS1 phosphorylates AAK at T360 in Drosophila S2 cells, we used the phospho-specific antibody for immunofluorescence analysis. Centrosomal signal of T360 phosphorylation (T360Ph) was readily detected in prometaphase cells but became significantly reduced in metaphase or upon depletion of MPS1 (Figure 2C and 2D). Mirroring AAK T-loop activation, the levels of T360Ph at centrosomes were positively correlated with the proximity of kinetochores in an MPS1-dependent manner (Figure 2C and 2E). Depleting AAK, or replacing the endogenous protein with an unphosphorylatable mutant for T360 (T360A) also compromised T360Ph staining, which further validated the specificity of the antibody in S2 cells (Figure S2E and S2F). Collectively these results support the conclusion that centrosomal AAK is phosphorylated at T360 by kinetochore-proximal MPS1.

FIGURE 2. MPS1-mediated phosphorylation of AAK C-lobe potentiates AAK auto-activation.

(A) In vitro kinase assay with purified human orthologues of AAK and MPS1. Western blot analysis of human AAK T288 activating T-loop autophosphorylation (AAK^T288Ph) and human MPS1 T676 activating T-loop autophosphorylation (MPS1^T676Ph). Recombinant His6-AAK was incubated with 100μM ATP during 30 min at 30°C in the indicated conditions. Total protein levels were assessed by gel silver staining.

(B) Quantification of human AAK T288 activating T-loop phosphorylation (AAK^T288Ph) blot intensity relative to total AAK levels in the indicated conditions, as presented in (A). All values were normalized to the mean value determined for basal AAK T288 phosphorylation, which was set to 1. The bars indicate the mean intensities; the dots indicate the measurements from triplicates.

(C) Representative immunofluorescence images of AAK T360 phosphorylation (AAK^T360Ph) in prometaphase and metaphase control Drosophila S2 cells and prometaphase cells depleted of MPS1. Insets depict magnifications of selected centrosomes.

(D) Quantification of the levels of AAK T360 phosphorylation (AAK^T360Ph) relative to DPLP at centrosomes of prometaphase and metaphase control Drosophila S2 cells and prometaphase cells depleted of MPS1, as presented in (C). All values were normalized to the mean value determined for control prometaphase cells, which was set to 1. n ≥ 56 centrosomes for each condition from 2 biological replicates.

(E) Quantification of the levels of AAK T360 phosphorylation (AAK^T360Ph) relative to DPLP at centrosomes of control Drosophila S2 cells and cells depleted of MPS1 in the indicated conditions. Centrosomes presented in (D) were organized in 3 different columns representing ranges of distance to the closest kinetochore: less than 1μm (<1); between 1μm and 3μm (1-3) and above 3μm (>3). All values were normalized to the mean value determined for control prometaphase cells, as presented in (D), which was set to 1. n ≥ 7 centrosomes for each condition from 2 biological replicates.

(F) Representative immunofluorescence images of AAK T311 phosphorylation (AAK^T311Ph) in prometaphase or metaphase Drosophila S2 cells depleted of endogenous AAK and expressing EGFP-AAK^WT or EGFP-AAK^T360A in the indicated conditions. Insets depict magnifications of selected centrosomes. EGFP-AAK^WT cells expressing endogenous MPS1 or overexpressing mCherry-MPS1 (OE) were imaged in the same coverslip. EGFP-AAK^T360A cells expressing endogenous MPS1 or overexpressing mCherry-MPS1 (OE) were imaged in the same coverslip.

(G) Quantification of the levels of AAK T311 phosphorylation (AAK^T311Ph) relative to EGFP at centrosomes of prometaphase or metaphase Drosophila S2 cells depleted of endogenous AAK and expressing EGFP-AAK^WT or EGFP-AAK^T360A in the indicated conditions, as presented in (F). All values were normalized to the mean value determined for prometaphase cells expressing EGFP-AAK^WT and endogenous levels of MPS1, which was set to 1. n ≥ 32 centrosomes for each condition from 2 biological replicates.

(H) Representative immunofluorescence images of AAK T311 phosphorylation (AAK^T311Ph) in prometaphase or metaphase Drosophila S2 cells expressing EGFP-AAK^WT or EGFP-AAK^T360D and expressing endogenous MPS1 or depleted of endogenous MPS1 in the indicated conditions. Insets depict magnifications of selected centrosomes.

(I) Quantification of the levels of AAK T311 phosphorylation (AAK^T311Ph) relative to EGFP at centrosomes of prometaphase or metaphase Drosophila S2 cells expressing EGFP-AAK^WT or EGFP-AAK^T360D and expressing endogenous MPS1 or depleted of endogenous MPS1 in the indicated conditions, as presented in (H). All values were normalized to the mean value determined for prometaphase cells expressing EGFP-AAK^WT and endogenous levels of MPS1, which was set to 1. n ≥ 27 centrosomes for each condition from 2 biological replicates.

Data information: Data in (B) are presented as mean ± SD and in (D), (E), (G) and (I) are presented as median ± interquartile range. Statistical analysis in (D), (E), (G) and (I) was performed using a Kruskal–Wallis test for multiple comparisons. P values: ns, not significant; *< 0.05; **< 0.01; ***< 0.001; ****< 0.0001. Scale bar: 5 μm. See also Figure S2.

A series of cell-based assays were next employed to test whether MPS1 stimulates AAK auto-phosphorylation of the T-loop (AAK^T311Ph) via phosphorylation of T360 in S2 cells. Consistent with the FRET results, over-expressing (OE) MPS1 caused a significant increase in AAK^T311Ph at metaphase centrosomes relative to controls (Figures S2G and S2H). Loss of AAK activation upon MPS1 OE in AAK depleted cells (Figures S2G and S2H) could be rescued by expression of WT but not kinase-dead (K183I) AAK (Figures S2I and S2J). Importantly, expression of the AAK non-phosphorylatable T360A mutant (AAK^T360A), predicted to have a similar structure as AAK^WT (Figure S2K), also failed to rescue MPS1-mediated activation of AAK in either prometaphase or metaphase cells (Figures 2F and 2G). Conversely, the phospho-mimetic mutation of T360 to aspartic acid (AAK^T360D) was sufficient to retain elevated levels of AAK T-loop phosphorylation in metaphase and restore AAK activation in prometaphase cells depleted of MPS1 (Figure 2H and 2I). Thus, phosphorylation of the conserved AAK T360 residue by MPS1 promotes auto-phosphorylation of the AAK T-loop (T311) and hence efficient kinase activation in vitro and in cells.

Activation of AAK by MPS1 is required for efficient error correction and chromosome congression

The functional contributions of proximity-based activation of AAK by MPS1 to cell division was evaluated by live- and fixed-cell analyses of mitotic S2 cells. To isolate potential chromosome alignment and error correction functions of MPS1 from its role as a checkpoint kinase in D. melanogaster³⁴, cells were arrested in mitosis via RNAi of the APC/C subunit CDC27^35,36,37. Cells efficiently aligned their chromosomes on a metaphase plate upon depletion of CDC27, however, co-depleting MPS1 impaired timely chromosome congression and increased the abundance of persistent polar chromosomes (Figures 3A, 3C and Video S1). The congression defect in CDC27/MPS1 co-depleted cells was rescued by centrosomal targeting (via C-terminal fusions to CM2) of either wild type MPS1 (MPS1^WT-CM2) or a phospho-mimetic mutant of AAK in which the previously identified MPS1 site (T360) was mutated to aspartic acid (AAK^T360D-CM2) (Figures 3A, 3C and Video S2). The rescue of chromosome alignment required AAK kinase activity since cells expressing a kinase dead (K183I) version of AAK^T360D-CM2 (AAK^K183I/T360D-CM2) failed to effectively congress their chromosomes (Figures 3A, 3C and Video S2). As expected, co-depletion of AAK and CDC27 delayed chromosome congression (Figure 3B, 3D and Video S3). This defect was rescued by the expression of EGFP-AAK^WT and EGFP-AAK^T360D but not by the non-phosphorylatable T360A AAK mutant (EGFP-AAK^T360A) (Figure 3B, 3D and Video S3). These results support the conclusion that MPS1-mediated phosphorylation of AAK T360 is required for efficient chromosome congression

FIGURE 3. Activation of AAK by MPS1 is required for efficient error correction and chromosome congression.

(A) Selected stills of representative mitotic progression movies of CDC27-depleted, CDC27/MPS1 co-depleted or CDC27/γTubulin co-depleted Drosophila S2 cells stably expressing GFP-H2B, mCherry-αTubulin and the indicated EGFP-tagged and CM2-fused transgenes (EGFP-X-CM2).

(B) Selected stills of representative mitotic progression movies of CDC27/AAK co-depleted Drosophila S2 cells stably expressing GFP-H2B, mCherry-αTubulin and the indicated transgenes. Asterisks indicates the centrosomes in the EGFP-AAK^T360A condition.

(C) Quantifications of the congression time of individual polar chromosomes (time each chromosome was retained near the spindle pole) in the conditions indicated in (A). n ≥ 16 polar chromosomes followed through time for each condition from at least 2 biological replicates.

(D) Quantifications of the congression time of individual polar chromosomes (time each chromosome was retained near the spindle pole) in the conditions indicated in (B). n ≥ 51 polar chromosomes followed through time for each condition from at least 3 biological replicates.

(E) Representative immunofluorescence images of mitotic Drosophila S2 cells expressing the indicated EGFP-tagged and CM2-fused transgenes (EGFP-X-CM2) and depleted of the indicated proteins. Insets depict magnifications of selected sister-kinetochores in amphitelic or syntelic attachment configurations.

(F) Percentage of syntelic kinetochore-microtubule attachments per cell in the indicated conditions, as presented in (E). Data corresponding to CDC27/γTubulin co-depleted cells are presented in Figure S3H. n ≥ 36 cells for each condition from at least 2 biological replicates.

(G) Representative immunofluorescence images of CDC27/AAK co-depleted Drosophila S2 cells expressing EGFP-tagged AAK^WT, AAK^T360A and AAK^T360D. Insets depict magnifications of selected sister-kinetochores in amphitelic or syntelic attachment configurations.

(H) Percentage of syntelic kinetochore-microtubule attachments per cell in the indicated conditions, as presented in (G). n ≥ 21 cells for each condition from at least 2 biological replicates.

Data information: Data in (C), (D), (F) and (H) are presented as median ± interquartile range. Statistical analysis was performed using a Kruskal–Wallis test for multiple comparisons. P values: ns, not significant; *< 0.05; **< 0.01; ***< 0.001; ****< 0.0001. Scale bar: 5 μm. See also Figure S3 and Videos S1, S2 and S3.

The attachment status of kinetochores was next assessed by analyzing confocal Z-sections of chromosomes (DAPI), kinetochores (CENP-C), and microtubules (α-Tubulin) in calcium-treated fixed cells to assess error correction (Figures 3 E–H and S3A–D). Chromosomes were scored as (i) bioriented when each sister kinetochore was clearly associated with microtubules oriented toward opposing spindle poles; (ii) syntelic when sister kinetochore pairs were clearly attached to microtubules from the same spindle pole; and (iii) pinned when the kinetochore pair was within 1 micron of the spindle pole regardless of whether its attachment status could be clearly discerned. While a vast majority of chromosomes were bioriented in CDC27-depleted cells, co-depletion of CDC27 and MPS1 resulted in a significant increase in the frequency of syntelic and pinned chromosomes revealing that MPS1 is required for error correction in fly cells (Figures 3E, 3F, S3A and S3B). The error correction defect in CDC27/MPS1 RNAi cells could be rescued by expressing centrosome-tethered MPS1^WT-CM2 but not AAK^WT-CM2 (Figures 3E, 3F, S3A and S3B). Co-depletion of AAK hindered MPS1^WT-CM2 capacity to promote chromosome alignment and biorientation, further supporting the conclusion that MPS1 acts upstream of AAK in the pole-based error correction pathway (Figures 3E, 3F, S3A and S3B). The requirement for MPS1 to potentiate AAK-mediated error correction could be bypassed in CDC27/MPS1 co-depleted cells by centrosomal targeting of either the T-loop (T311D) or C-lobe (T360D) AAK-CM2 phospho-mimetic mutants but not the catalytic dead AAK^K183I/T360D-CM2 version (Figures 3E, 3F, S3A and S3B).

To directly test the requirement of AAK T360 phosphorylation for error correction, we co-depleted CDC27 and endogenous AAK and evaluated kinetochore-microtubule attachments upon the expression of untethered versions of AAK. Contrasting with EGFP-AAK^WT and EGFP-AAK^T360D, the non-phosphorylatable EGFP-AAK^T360A mutant failed to restore chromosome biorientation and alignment in S2 cells depleted of endogenous AAK (Figures 3G, 3H, S3C and S3D). Given that MPS1 depletion hindered the accumulation of γ-Tubulin and D-TACC^S863 phosphorylation at centrosomes (Figure S1I–K), it was possible that the mitotic defects observed in the absence of MPS1-dependent AAK activation stemmed from an attenuation of centrosome maturation and function. However, S2 cells partially depleted of γ-tubulin, which formed abnormal mitotic spindles (Figure 3A) but maintained robust AAK activation (Figure S3E–G), successfully aligned their chromosomes and established proper kinetochore-microtubule attachments as efficiently as control cells (Figure 3A, 3C, S3H–S3J and Video S1). This finding suggests that compromised microtubule nucleation and polymerization at the centrosome play a minor role in the observed accumulation of erroneous kinetochore-microtubule attachments and chromosome misalignment in S2 cells that fail to fully activate AAK.

Collectively, the data demonstrate that the AAK-mediated error correction pathway is impaired in cells depleted of MPS1 or expressing AAK harboring the T360A mutation. AAK targets a number of the same sites on the N-terminal tail of NDC80 as ABK but S69 on HEC1 (human NDC80) is preferentially phosphorylated by AAK⁹. Thus, we used a phospho-specific antibody against the equivalent phosphorylation on Drosophila NDC80 (NDC80^S67Ph) to evaluate whether it requires MPS1-mediated phosphorylation of AAK T360. Kinetochore levels of NDC80^S67Ph were elevated in prometaphase cells and decreased as chromosomes congressed to the metaphase plate (Figure S3K and S3L). Consistent with impaired error correction, depletion of MPS1 or expression of the EGFP-AAK^T360A mutant significantly compromised the phosphorylation of NDC80 at S67 regardless of kinetochore proximity to the spindle pole (Figure S3K–S3M). To test whether MPS1 directly potentiates AAK-mediated phosphorylation of NDC80/HEC1 N-terminal tail, we performed an in vitro kinase assay with recombinant bonsai NDC80 complex as a substrate³⁸ and probed for the phosphorylation of HEC1 at S55, another well-established AAK phosphorylation site⁹. In accordance with its activation status, AAK capacity to phosphorylate S55 increased dramatically when co-incubated with MPS1 (Figure S3N and S3O). These findings support the conclusion that pole-based error correction in fly cells is carried out by AAK activated via phosphorylation of its C-lobe (T360) by MPS1 and further validates results from the FRET reporter indicating that phosphorylation of downstream AAK targets is potentiated by MPS1.

Proximity-based activation of AAK by MPS1 is conserved in human cells

To examine if proximity-based activation of AAK by MPS1 is conserved outside D. melanogaster, the extent of AAK activation was evaluated under experimental conditions where the positioning of kinetochores relative to centrosomes/spindle poles could be manipulated in human RPE-1 cells³⁹. When compared to control monopolar spindles (STLC treatment), kinetochores were positioned closer to the monopole center upon inhibition of the kinetochore-associated motor CENP-E with GSK923295 (CENP-Ei) while depletion of the Dynein adaptor Spindly (siSpindly) positioned kinetochores further from the monopole center (Figures 4A–C). Consistent with there being spatial cross-talk between kinetochores and centrosomes, AAK activation (AAK^T288Ph) levels at centrosomes/spindle poles was significantly increased in the CENP-Ei cells and significantly reduced in the siSpindly cells relative to control cells (Figures 4D and 4E). Active AAK^T288Ph signal at centrosomes was significantly reduced compared to controls in nocodazole- (Figures S4A and S4B) and STLC-treated cells (Figure 4F and 4G) when MPS1 was inhibited with Cpd-5 (MPS1i). Importantly, inhibiting MPS1 prevented the increase in AAK activation in CENP-Ei monopoles (Figure 4F and 4G). Thus, the potentiation in AAK activation observed when kinetochores were positioned closer to centrosomes in human RPE-1 cells required MPS1 activity. Accordingly, the activation of AAK at centrosomes during an unperturbed mitosis peaked in prometaphase in an MPS1-depedent manner and decreased significantly in metaphase (Figure S4C and S4D). While the observations of centrosomal AAK activation in RPE-1 cells were broadly consistent with those from fly cells, one notable difference was in prophase cells where AAK activation at centrosomes was high in fly cells but low in RPE-1 cells (Figure S4C and S4D). It is unclear why this is the case or if the AAK activation status fulfills unique prophase functions in fly versus human cells.

Figure 4. Proximity-based activation of AAK by MPS1 is conserved in human cells.

(A) Schematic representation of the monopolar spindle assay used to assess AAK^T288Ph and HEC1^S69Ph changes in response to the distance between centrosomes and kinetochores.

(B) Representative immunofluorescence images of STLC-treated RPE-1 cells in control, CENP-E inhibited (CENP-Ei) and Spindly depleted (siSpindly) conditions.

(C) Quantification of centrosome (PCNT)-kinetochore (CENP-C) (C-KT) distances of STLC-treated RPE-1 cells in the indicated conditions, as presented in (B).

(D) Representative immunofluorescence images of AAK T288 phosphorylation (AAK^T288Ph) and AAK levels in STLC-treated RPE-1 cells in control, CENP-E inhibited (CENP-Ei) and Spindly depleted (siSpindly) conditions.

(E) Quantification of the levels of AAK T288 phosphorylation (AAK^T288Ph) relative to total AAK levels at centrosomes of STLC-treated RPE-1 cells in the indicated conditions, as presented in (D). All values were normalized to the median value determined for control cells, which was set to 1.

(F) Representative immunofluorescence images of AAK T288 phosphorylation (AAK^T288Ph) and AAK levels in STLC-treated RPE-1 cells in control, MPS1 inhibited (MPS1i), MPS1/CENP-E co-inhibited (MPS1i+CENP-Ei) and MPS1 inhibited in Spindly depleted background (MPS1i+siSpindly). In all conditions the proteasome was inhibited with MG132.

(G) Quantification of the levels of AAK T288 phosphorylation (AAK^T288Ph) relative to total AAK levels at centrosomes of STLC-treated RPE-1 cells in the indicated conditions, as presented in (F). All values were normalized to the median value determined for control cells, which was set to 1.

(H) Representative immunofluorescence images of HEC1 S69 phosphorylation (HEC1^S69Ph) and CENP-C levels in STLC-treated RPE-1 cells in control, CENP-E inhibited (CENP-Ei) and Spindly depleted (siSpindly) conditions.

(I) Quantification of the levels of HEC1 S69 phosphorylation (HEC1^S69Ph) relative to CENP-C levels at kinetochores of STLC-treated RPE-1 cells in the indicated conditions, as presented in (H). All values were normalized to the median value determined for control cells, which was set to 1.

(J) Representative immunofluorescence images of HEC1 S69 phosphorylation (HEC1^S69Ph) and CENP-C levels in STLC-treated RPE-1 cells in control, MPS1 inhibited (MPS1i), MPS1/CENP-E co-inhibited (MPS1i+CENP-Ei) and MPS1 inhibited in Spindly depleted background (MPS1i+siSpindly). In all conditions the proteasome was inhibited with MG132.

(K) Quantification of the levels of HEC1 S69 phosphorylation (HEC1^S69Ph) relative to CENP-C levels at kinetochores of STLC-treated RPE-1 cells the indicated conditions, as presented in (J). All values were normalized to the median value determined for control cells, which was set to 1.

(L) Proposed model for proximity-based activation of AAK by MPS1 as a strategy to potentiate error correction of polar chromosomes and ensure chromosome congression. A pool of MPS1 recruited to kinetochores of polar chromosomes indirectly contributes to pole-based error correction by phosphorylating AAK on the conserved residue T360. This potentiates AAK activation at centrosomes and thereby its capacity to phosphorylate the N-terminal tail of HEC1/NDC80 at kinetochores to destabilize end-on attachments. This mechanism corrects and prevents syntelic kinetochore-microtubule interactions in the vicinity of centrosomes and ensures efficient chromosome congression.

Data information: Data are presented in (C), (E), (G), (I) and (K) as mean ± SD. Statistical analysis was performed using a Mann-Whitney non-parametric t-test. P values: ns, not significant; *< 0.05; **< 0.01; ***< 0.001; ****< 0.0001. n ≥ 75 cells for each condition. Scale bar: 5 μm. See also Figure S4.

We then measured the extent of HEC1 phosphorylation at S69 (HEC1^S69Ph) as a function of kinetochore positioning within monopolar spindles. Consistent with the extent of AAK activation (AAK^T288Ph levels), phosphorylation of HEC1-S69 was increased in CENP-Ei cells and reduced in siSpindly cells relative to controls (Figures 4H and 4I). In nocodazole-treated cells, HEC1^S69Ph levels were reduced to the same extent following inhibition of either MPS1 or AAK and to a lesser extent upon ABK inhibition reflecting an upstream role of MPS1 in the preferential phosphorylation of HEC1-S69 by AAK (Figures S4E and S4F). Importantly, HEC1^S69Ph levels were significantly reduced when MPS1 was inhibited including in CENP-Ei cells that normally exhibited a significant increase in HEC1^S69Ph levels (Figures 4J and 4K). Altogether, the data support the conclusion that kinetochore proximity-based activation of AAK by MPS1 potentiates phosphorylation of the bona fide error correction substrate HEC1 in human cells.

DISCUSSION

We propose a model whereby a proximity-dependent crosstalk between kinetochores and centrosomes/spindle poles promotes error correction (Figure 4L). The model envisions that as erroneous kinetochore-microtubule interactions (e.g. a syntelic attachment) become positioned within the vicinity (~1 μm) of the spindle pole, kinetochore-recruited MPS1 phosphorylates AAK on a conserved site (T360 in flies) located in the αG-helix of the kinase C-lobe. Phosphorylation of the C-lobe, in turn, triggers auto-catalytic phosphorylation of the AAK T-loop to amplify a ~2 μm radius activity gradient^7,8,40 to locally promote microtubule detachment via phosphorylation of kinetochore substrates such as the NDC80 complex. AlphaFold3 predicts that phosphorylation of T360 causes an outward stretch of the αG-helix that pulls on the T-loop by interacting with D317 in manner that could render the T-loop more flexible and leave the activating T311 more exposed for phosphorylation (Figure S2L–S2N). Further investigation into the structural basis of this novel activation mechanism is warranted.

It has been proposed that the establishment of stable end-on attachments prevents recruitment of MPS1 to kinetochores because of mutually exclusive binding to either microtubules or MPS1 by the CH-domains of the NDC80 complex^20,21,22. However, a pool of MPS1 still localizes to microtubule-attached kinetochores either through association with the NDC80 complex or other kinetochore receptors^11,19. Three recent studies in budding yeast showed that MPS1 binds to regions in the CH-domains of the NDC80 complex that are on the opposite side of the microtubule binding region^23,24,25 and, accordingly, purified budding yeast NDC80 complex was shown to simultaneously bind microtubules and MPS1 in vitro²⁴. It has been suggested that this interaction interface is conserved²⁵; if so then this mechanism could explain how a population of MPS1 remains at microtubule-attached metazoan kinetochores. Regardless of how it is recruited, we hypothesize that the residual MPS1 population on erroneously attached polar kinetochores activates AAK to promote their own correction.

An interesting feature of the crosstalk pathway described here is that it contains an intrinsic positive feedback loop since the destabilization of kinetochore-microtubule interactions by AAK-mediated phosphorylation of NDC80 would recruit more MPS1. Such a spatio-chemical feedback loop would heavily disfavor production of end-on attached kinetochores in the vicinity of spindle poles. Thus, we speculate that MPS1-driven potentiation of AAK activity following nuclear envelope breakdown is important to minimize the formation of end-on attachments near the centrosomes, where syntelic attachments would otherwise be prone to accumulate due to the high density of nucleated microtubules. Instead, robust activation of AAK favors the formation of lateral attachments and facilitates motor activation of CENP-E⁴¹ to ensure timely chromosome congression towards the cell equator, where amphitelic attachments can form synchronously as a result of interactions between non-centrosomal short microtubules emanating from kinetochores and bundles of antiparallel microtubules^42,43,44.

Overall, our findings support the conclusion that a pool of MPS1 at kinetochores indirectly contributes to pole-based error correction. This model is consistent with an earlier proposal that MPS1 plays an indirect role in error correction with a notable difference being that the relevant downstream target of MPS1 is AAK rather than ABK^13,14,15. MPS1 clearly activated AAK in our cell-based assays, but consistent with prior findings^16,17,18 we did not see evidence of MPS1-mediated ABK activation in cells despite the facts that (i) the centromere/kinetochore localized ABK is closer than centrosomal AAK to kinetochore-bound MPS1, (ii) the C-lobe threonine is present in ABK αG-helix, and (iii) ABK is phosphorylated and activated by MPS1 in vitro (not shown). Since constitutive ABK activation would be detrimental to forming stable attachments, kinetochore/centromere associated protein phosphatases likely protect ABK from MPS1-mediated activation in cells. Direct and indirect roles for MPS1 in error correction need not be mutually exclusive; in fact, it now appears that diverse organisms rely on each mechanism to different extents in carrying out error correction. Budding yeast utilize MPS1 to directly target the NDC80 complex without indirectly inducing error correction through activation of AURORA kinase¹⁰. Conversely, our work shows that activation of AAK-mediated error correction by kinetochore-recruited MPS1 is the dominant pathway in flies although a direct role for MPS1 in error correction cannot be ruled out. Since we observed proximity-based activation of AAK by MPS1 in RPE-1 cells, efficient error correction in humans likely requires both indirect and direct pathways via MPS1-mediated phosphorylation of AAK and kinetochore-microtubule attachment factors respectively. Future efforts to dissect the relative contributions of direct versus indirect roles of MPS1 in error correction are warranted, but the findings presented here define a novel and conserved mechanism by which polar kinetochores modulate centrosome-based activities to promote accurate chromosome segregation.

RESOURCE AVAILABILITY

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Carlos Conde (cconde@ibmc.up.pt).

Materials availability

All unique/stable reagents generated in this study will be made available on request.

Data and code availability

The mass spectrometry proteomics data have been deposited in the ProteomeXchange Consortium via the PRIDE⁴⁵ partner repository with the dataset identifier PXD056173. This study did not generate code.

STAR METHODS

EXPERIMENTAL MODELS AND STUDY PARTICIPANT DETAILS

Drosophila Cell Lines

The experiments have been conducted in male Drosophila S2 cells (S2-DGRC, RRID: CVCL_TZ72) cultured at 25°C in Schneider’s Insect Medium (Sigma-Aldrich) supplemented with 10% FBS. This cell line has been authenticated by DGRC.

Human Cell Lines

The experiments have been conducted in female hTERT RPE-1 (ATCC, RRID: CVCL_4388) cultured in DMEM/F12 supplemented with 10% FBS and cultured at 37°C in humidified conditions with 5% CO2. This cell line has been authenticated by ATCC.

METHOD DETAILS

RNAi-mediated depletion and drug treatments

For protein depletion in Drosophila S2 cells, dsRNA synthesis was performed using TranscriptAid T7 High Yield Transcription Kit (ThermoFischer), according to the manufacturer’s instructions, at 37°C for 16h. The incubation with dsRNA was performed at 25°C for 1h in non-supplemented Schneider’s Insect Medium (Sigma-Aldrich) followed by 96h-120h of cell culture in Schneider’s Insect Medium (Sigma-Aldrich) supplemented with 10% FBS, as previously described³⁴. The PCR product used as template to produce MPS1 RNAi was amplified with the set of primers MPS1F (5′-TAATACGACTCACTATAGGGTCTTCCAAACACCTATGACG-3′) and MPS1R (5′-TAATACGACTCACTATAGGGCGTTTAGATATCCCTGCACCA-3′) (Table S1). The PCR product used as template to produce AAK 5′UTR RNAi was amplified with AAK5′_F (5′-TAATACGACTCACTATAGGGAGAACTTGCCATTCGCCTCATC-3′) and AAK5′_R (5′-TAATACGACTCACTATAGGGAGAGACGGCACGGCACAGGCACTCG-3′) (Table S1). The PCR product used as template to produce AAK 3′UTR RNAi was amplified with AAK3′_F (5′-TAATACGACTCACTATAGGGAGAACACATTCTTGTTTAATTTTC-3′) and AAK3′_R (5′-TAATACGACTCACTATAGGGAGAGAAAACACACACAAACTTT-3′) (Table S1). The PCR product used as template to produce NDC80 RNAi was amplified with NDC80F (5′-TAATACGACTCACTATAGGGTTTGATCCGGATGTTTTGTTGTAGTTC-3′) and NDC80R (5′-TAATACGACTCACTATAGGGTTTTTTAATTTGAAAACAAAGAATACTCTT-3′) (Table S1). The PCR product used as template to produce CDC27 RNAi was amplified with CDC27F (5′-TAATACGACTCACTATAGGGAACAATAGC-3′) and CDC27R (5′-TAATACGACTCACTATAGGGTCTTCATGTAGAATTGCATGGC-3′) (Table S1). The PCR product used as template to produce γTubulin RNAi was amplified with γTub RNAiF (5’-TAATACGACTCACTATAGTTTTCTCATGACAGGCTACACG-3’) and γTub RNAiR (5’-TAATACGACTCACTATAGGGTCACAGATCGACTATCCTCC-3’) (Table S1). At selected time points, cells were collected and processed for immunofluorescence or time-lapse microscopy. When required, cells were subjected to several drug treatments before being collected and processed for analysis. To promote microtubule depolymerization, cells were incubated for 2 h with 30 μM colchicine (Sigma-Aldrich). To inhibit the proteasome, cells were incubated for 3 h with 20 μM MG132 (Calbiochem).

For siRNA-mediated depletion experiments in human RPE-1 cells, cells were transfected in OptiMEM (Thermo Fisher) with Lipofectamine RNAiMAX (Thermo Fisher) with 40 nM siRNAs for 48 h – siSpindly: 5’-GAAAGGGUCUCAAACUGAA-3’, siControl (non-targeting control siRNA): 5’-UGGUUUACAUGUCGACUAA-3′ (Table S1). Where indicated, monopolar spindles were induced by inhibiting Eg5 using 5 μM S-Trityl-L-Cysteine (STLC; Santa Cruz Biotechnology) which was added to the medium 5 h before fixation. CENP-E was inhibited by adding 200 nM GSK923295 (MedChemExpress) to the medium 1 h before fixation. MPS1 was inhibited by adding 500 nM Cpd-5 (³³; a gift from René Medema, Oncode Institute, The Netherlands Cancer Institute, The Netherlands) to the medium 1 h before fixation. To avoid mitotic exit after MPS1 inhibition, the proteasome was inhibited by adding 5 μM MG132 (Sigma) to the medium 1 h before MPS1 inhibition. To depolymerize microtubules, 2 μM nocodazole (Sigma) was added 2 h before fixation. For inhibition of AAK and ABK, 250 nM MLN-8054 (Selleck Chemicals) and 4 μM ZM447439 (AstraZeneca) were respectively added to the cells 2 h before fixation.

Transgene constructs and S2 cell transfection

EGFP-MPS1^WT, H2B-GFP and mCherry-α-Tubulin constructs were previously characterized³⁴. For the generation of the expression vector mCherry-MPS1^WT, mCherry cDNA was amplified by PCR and inserted into the EGFP-MPS1 construct, by FastCloning⁴⁶, in replacement of previously removed EGFP. EGFP-AAK^WT and FLAG-AAK^WT constructs were generated by PCR amplification of AAK cDNA and insertion into modified versions of pENTR entry vector through FastCloning⁴⁶. Entry clones were then used for Gateway recombination (Thermo Fischer Scientific) together with pHGW (for EGFP-AAK constructs) or pHFMW (for FLAG-AAK constructs) destination vectors (heat-shock inducible promoter, N-terminal EGFP or FLAG tag, respectively). To generate FLAG-AAK^K183I, EGFP-AAK^T311D, EGFP-AAK^T360A, EGFP-AAK^T360D and EGFP-AAK^K183I/T360D versions, either FLAG-AAK^WT or EGFP-AAK^WT expression vectors had their codons corresponding to K183, T311, T360, T360 and K183 and T360 changed by site-directed mutagenesis with a primer harboring the desired mutation to Isoleucine (I), Aspartate (D), Alanine (A), Aspartate (D) and Isoleucine (I) and Aspartate (D), respectively. The PCR reactions were performed with Phusion polymerase (Thermo Fisher Scientific). To generate the EGFP-MPS1^WT-CM2 and EGFP-AAK-CM2 constructs, the CM2 domain of Centrosomin (CNN) cDNA was amplified by PCR and was inserted in the C-terminal region of the available EGFP-MPS1^WT and EGFP-AAK constructs, respectively, by FastCloning⁴⁶. Stable transfection of indicated vectors into S2 cells was performed using Effectene Transfection Reagent (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. To induce EGFP-MPS1^WT, mCh-MPS1^WT and EGFP-MPS1^WT-CM2 expression, S2 cells were incubated overnight with 200 μM CuSO₄ at 25°C. To induce the expression of either AAK constructs cells were heat-shocked for 30 min at 37°C and allowed to rest for at least 6 h before being processed for immunofluorescence or live cell imaging.

FRET-based AURORA kinase sensor

S2 cells stably expressing the CM2-FRET reporter, GFP-H2B and either mCherry-α-tubulin or mCherry-MPS1 were seeded onto Concanavalin A- (ConA) (Sigma-Aldrich) treated 35 mm glass bottom petri dishes (Cellvis) for 1 hour prior to bringing the volume of Schneider’s media up to 2 mL. Cells were then imaged at 24°C on a TiE inverted microscope (Nikon) equipped with an iXON EMCCD camera (Andor Technology) using a 100x 1.4 NA Plan Apo violet-corrected series differential interference contrast objective (Nikon). Metamorph software (Molecular Devices) was used to control the imaging system. Mitotic cells and the best focal plane were identified on the RFP channel (mCherry-α-tubulin or mCherry-MPS1). After acquiring the RFP image, sequential images of mTurquoise2, mVenus, and FRET (ECFP excitation, EYFP emission) were taken with equal exposure times (200 ms). Over-expression of MPS1 was induced by overnight incubation with 200 μM CuSO₄. AAK was inhibited by treating the seeded cells in the chamber for 1 hour in 100 nM MLN8237 (Selleck Chemicals) prior to FRET imaging. Background-corrected fluorescence intensities for mTurquoise2 and FRET were measured in Metamorph software by subtraction of local background using a duplicate area. The reported FRET emission ratios represent the ratios of the background corrected FRET signal over the background corrected mTurquoise2 signal and values were normalized against the mean prometaphase FRET emission ratio for each biological replicate.

Live cell imaging

For live imaging of S2 cells, cells were plated onto μ-Dish 35mm, high Glass Bottom dishes (Ibidi) previously treated with 0.2 mg/mL ConA (Sigma-Aldrich). Images were obtained at 30s intervals up to a maximum of 90min. All images were acquired at 25°C with either a spinning disc confocal system (Revolution; Andor) equipped with an electron multiplying charge-coupled device camera (iXonEM+; Andor) and a CSU-22 unit (Yokogawa) based on an inverted microscope (IX81; Olympus) driven by iQ software (Andor) or a Nikon ECLIPSE Ti2 (Nikon, Japan) equipped with Crest X-Light V3 spinning disk confocal system (Crest CrestOptics, Italy) and Kinetix 25 (Photometrics) camera driven by NIS elements 5.5 (Nikon, Japan) software. The microscopes are served with two laser lines, 488 and 561 nm, for the excitation of EGFP/GFP and mCherry, respectively. Time-lapse imaging of z stacks with 0.8 μm steps covering the entire volume of the cell was collected, and image sequence analysis and video assembly were done with Fiji software (https://fiji.sc/). The time each polar chromosome spent closer to the spindle pole than to the metaphase plate was quantified.

Immunofluorescence imaging and analysis

For immunofluorescence analysis of kinetochore-microtubule attachments in S2 cells, 10⁵ cells were seeded onto coverslips previously treated with 0.2 mg/mL ConA (Sigma-Aldrich). For detection of calcium-stable k-fibers, cells were permeabilized for 120 seconds in a buffer containing 100 mM Pipes (pH 6.8), 1 mM MgCl2, 0.1 mM CaCl₂, and 0.1% Triton X-100 and then fixed for 10 min in the same buffer supplemented with 4% formaldehyde as previously described⁴⁷ followed by immunostaining. For immunofluorescence analysis of AAK^T311Ph, AAK^T360Ph or NDC80^S67Ph levels in correlation to the distance between the centrosome and its closest kinetochore in S2 cells, 10⁵ cells were seeded onto coverslips previously treated with 0.2 mg/mL ConA (Sigma-Aldrich), processed for fixation in 4% paraformaldehyde in PBS for 12 min and further extracted for 8 min with 0.1% Triton X-100 in PBS, followed by immunostaining. For immunofluorescence analysis of AAK^T311Ph and ABK^T209Ph levels, 10⁵ cells were centrifuged onto slides for 5 min, at 239 g, and processed for fixation in 4% paraformaldehyde in PBS for 12 min and further extracted for 8 min with 0.1% Triton X-100 in PBS, followed by immunostaining. Immunostaining of S2 cells was performed as previously described³⁴. Blocking was performed in 1xPBS, 10% fetal bovine serum, 0.05% Tween 20 at room temperature for 1h. Primary antibody incubations were prepared in blocking solution and incubated at 4°C overnight, followed by three 10-min washes in 1xPBS, 0.05% Tween. Secondary antibody incubations were prepared in blocking solution and incubated at room temperature for 1h, followed by three 10-min washes in 1xFBS, 0,05% Tween. Slides were then mounted using Vectashield mounting medium for fluorescence with DAPI (Vector Laboratories, Burlingame, CA). Images were collected in a Leica Scanning Confocal SP8 inverted microscope (Leica Microsystems). For immunofluorescence analysis of human cells, cells were grown on 12 mm round cover glasses (Menzel Glaser) and fixed by ice-cold methanol at −20°C for 4 minutes. DAPI was used at 1 μg/mL (Sigma) for DNA counterstaining. For quantification of centrosome to kinetochore distance, as well as for AAK and HEC1 phosphorylation status, images were acquired using a Plan-Apochromat DIC 63x/1.4 NA oil objective mounted on an inverted Zeiss Axio Observer Z1 microscope (Marianas Imaging Workstation from 3i-Intelligent Imaging and Innovations Inc.) equipped with a Hamamatsu ORCA-Flash4.0 v2 sCMOS Camera (Hamamatsu Photonics). Representative images were acquired using LSM700 confocal microscope (Carl Zeiss Inc.) mounted on a Zeiss-Axio imager Z1 equipped with alpha-Plan-Apochromat 100x/1.46 oil DIC M27 and plan-apochromat 63×/1.40 oil DIC M27 objective (Carl Zeiss Inc.) and Zen 2010 B SP1 software (Carl Zeiss, Inc.). In S2 cells, the distance (in 3D) between the centrosome and its closest kinetochore was calculated using the measured values of the distance in the XY plane and the distance in the Z-axis (number of slices x Z-step). For immunofluorescence quantification, the mean fluorescence intensity of kinetochore or centrosome proteins was measured within a specific predefined region of interest (ROI) where individual kinetochores or centrosomes could fit. The kinetochores were identified based on their constitutive marker CID. The centrosomes were identified based on either their constitutive marker DPLP or, when stated, the transfected FLAG-AAK or EGFP-AAK versions. After background subtraction, identified as a region in the cell with no kinetochores or centrosomes, the intensities of the proteins of interest were determined relative to the respective constitutive markers, unless stated otherwise. Control values were averaged and used for normalization of values determined in the different biological conditions tested.

In human cells, kinetochore to centrosome distance and HEC1^S69Ph signal quantification on kinetochores in fixed cells were both measured using a custom routine written in MATLAB 8.2 (Mathworks). For analysis of kinetochore to centrosome distance, Pericentrin signal at centrosomes and CENP-C signal at kinetochores were used to determine the three-dimensional distance between the midpoint between the centrioles and the centroids of a kinetochore. For HEC1^S69Ph quantification on kinetochores, intensities were determined by quantification of pixel gray levels of the focused z plane within a ROI. Background fluorescence was measured outside the ROI and subtracted. For quantifications at kinetochores, the ROI encompassed a single kinetochore measuring the signal intensity of HEC1^S69Ph at the kinetochores. HEC1^S69Ph intensity was quantified relative to CENP-C fluorescence. HEC1^S69Ph/CENP-C intensity ratios were subsequently normalized to control median value. At least 200 kinetochores from 10 different cells in each of three independent experiments were measured in both quantifications.

For AAK^T288Ph quantification at centrosomes, intensities were determined by measuring the signal intensity of AAK^T288Ph and total AAK in ImageJ software. Protein signal was quantified by drawing a circle closely along the total AAK staining at centrosomes and using the same ROI for AAK^T288Ph quantification. The ROI were adjusted in every cell to accurately match the total AAK staining. AAK^T288Ph intensity was quantified relative to total AAK fluorescence. AAK^T288Ph/AAK intensity ratios were subsequently normalized to control median value. At least 25 cells in each of three independent experiments were used for the quantifications.

Antibodies

Rabbit polyclonal anti-phospho(ph)Thr360-AAK (AAK^T360Ph) was raised against a phosphorylated peptide SKNYDE[pT]YKKIL, generated by Eurogentec, Seraing, Belgium. The primary antibodies used for immunofluorescence in S2 cells were rabbit anti-phosphorylated Thr360-Aurora A (AAK^T360Ph, this study) used at 1:500; mouse anti-α-tubulin B512 (Sigma-Aldrich) used at 1:5000; rat anti-CID (Rat4) used at 1:500; chicken anti-GFP (#ab13970, Abcam, Cambridge, UK) used at 1:2000; rabbit anti-CENP-C (Rb1⁴⁸; a gift from Christian Lehner, University of Zurich, Switzerland) used at 1:3000; guinea pig anti-MPS1 (Gp15⁴⁹; RRID:AB_2567774; a gift from Scott Hawley, The Stowers Institute for Medical Research, USA) used at 1:250; rabbit anti-phosphorylated Thr232-Aurora B (ABK^T232Ph, Rockland, Lim-erik, PA) used at 1:1000; rabbit anti-phosphorylated Thr288-Aurora A (AAK^T288Ph, C39D8, mAb #3079, Cell Signaling Technology, Danvers, MA, USA) used at 1:500; chicken anti-DPLP (⁵⁰; a gift from Mónica Bettencourt-Dias, Instituto Gulbenkian de Ciencia, Portugal) used at 1:2000; mouse anti-FLAG (F1804, Sigma-Aldrich) used at 1:500; mouse anti-γTubulin (GTU-88, T5326 Sigma-Aldrich) used at 1:250; rabbit anti-phosphorylated Ser863-D-TACC (D-TACC^S863Ph, ²⁶; a gift from Jordan Raff) used at 1:500. Alexa secondary antibodies were used according to the manufacturer’s instructions.

The primary antibodies used for immunofluorescence in human cells were guinea pig anti-CENP-C (PD030, MBL) used at 1:1000, mouse anti-Pericentrin (ab28144, Abcam) used at 1:1000, mouse anti-Aurora A (ab13824, Abcam) used at 1:200, rabbit anti-phosphorylated Thr288 Aurora A (AAK^T288Ph) (C39D8 #3079, Cell Signalling) used at 1:1600 and rabbit anti-phosphorylated Ser69-HEC1 (HEC1^S69Ph, ⁹; a gift from Jennifer DeLuca, Colorado State University, USA) used at 1:3000. Goat anti-Mouse IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 568; Goat anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 488; and Goat anti-Guinea Pig IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 647 (Invitrogen) secondary antibodies were used at 1:1000.

The primary antibodies used for Western blotting analysis were rabbit anti-phosphorylated Thr360-Aurora A (AAK^T360Ph, this study) used at 1:250, rabbit anti-phosphorylated Thr676-MPS1 (MPS1^T676Ph, ⁵¹; a gift from Geert Kops, Oncode Institute, Hubrecht Institute, The Netherlands) used at 1:2000; rabbit anti-phosphorylated Thr288-Aurora A (AAK^T288Ph, C39D8, mAb #3079, Cell Signaling Technology, Danvers, MA, USA) used at 1:1000; mouse anti-αTubulin (DM1a, sc-32293, Santa Cruz Biotechnology) used at 1:10000; guinea pig anti-MPS1 (Gp15, ⁴⁹; RRID:AB_2567774; a gift from Scott Hawley, The Stowers Institute for Medical Research, USA) used at 1:5000; mouse anti-AURKA antibody (BD #610938, 4/IAK1) used at 1:1000, rabbit anti-AURKA pT288, AURKB pT232, AURKC pT198 (Cell Signalling #2914, D13A11) used at 1:1000, mouse anti-GST (Novagen/Millipore #71097-3) used at 1:1000, mouse anti-NDC80 (Gene Tex (via Biozol) # GTX 70268-100, 9G3.23), rabbit anti-phosphorylated Ser55-NDC80 (NDC80^S55Ph, Invitogen #PA5-85846) used at 1:1000.

Recombinant Protein Production

For the mass spectrometry analysis, the catalytic domain of either human or Drosophila AAK wild-type or kinase-dead (AAK^K162I in human or AAK^K183I in Drosophila) were fused with N-terminal MBP. To generate these constructs, human and Drosophila AAK cDNAs were used as template to amplify the regions corresponding to the amino acids 126-403 and 144-411, respectively. The PCR products were introduced in pMal-c2 (New England Biolabs) expression vector through FastCloning⁴⁶. Recombinant vectors were subsequently transformed into BL21-star competent cells, and expression was induced overnight at 15°C by the addition of 0.05 mM IPTG. Lysates were sonicated and clarified by centrifugation at 4°C. The recombinant proteins were then incubated with MBP amylose magnetic beads (New England Biolabs) for 1 h at 4°C and washed with 200 mM NaCl, 20 mM Tris–HCl, 1 mM EDTA, 1 mM DTT, pH 7.4 and eluted in Column Buffer supplemented with 10 mM Maltose. The same protocol was followed for expression and purification of MBP protein alone.

To produce human MBP-AAK-8xHis for in vitro kinase assays with bonsai NDC80 complex, a PCR product containing the sequence encoding full length human AAK was inserted into the BamHI/XhoI sites of a pETDuet-1 vector modified as described in⁵² by Gibson assembly⁵³. All PCR amplifications were performed with Phusion High Fidelity PCR master mix (Thermo Fisher Scientific, Waltham, US-MA). The ligation and PCR products were used to transform Omnimax competent bacteria (Invitrogen Corporation, Waltham, US-MA), and cells harboring plasmids were kept under antibiotic selection. The final DNA construct was validated by Sanger sequencing. The sequence encoding GST-MPS1-6xHis was subcloned into BamHI/HindIII sites of a pLIB vector for insect cell expression. The construct expressing bonsai NDC80 complex was generated previously³⁸.

Expression of human wild-type (WT) and non-phosphorylatable T337 mutant (T337A) MBP-AAK-6xHis was carried out in E. coli BL21CodonPlus(DE3)-RIL cells (Agilent Products, Santa Clara, US-CA) grown in 2 L Terrific Broth (TB). Protein expression was induced with 0.1 mM IPTG at 18°C for 18 hours. Cells were pelleted, snap-frozen, and stored at −80°C All purification steps were performed on ice or at 4°C. Cells expressing MBP-AAK-6xHis were resuspended in purification buffer (20 mM Tris-HCl pH 8, 500 mM NaCl, 1 mM TCEP) with added DNase (Roche, Basel, Switzerland) and protease inhibitors (Serva Electrophoresis GmbH, Heidelberg, Germany). Cells were lysed via sonication, and the lysate clarified by centrifugation (85,000×g, 30 min). The supernatant was incubated with 2 mL cOmplete^™ His-tag NiNTA beads (Roche) pre-equilibrated in purification buffer for 2 hours with agitation. Beads were washed with 60 mL wash buffer (20 mM Tris-HCl pH 8, 500 mM NaCl, 1 mM TCEP, 20 mM Imidazole) using a 10-mL gravity column (Thermo Fisher Scientific). Protein was eluted using 300 mM imidazole in purification buffer. Following concentration by centrifugation, the protein was injected into a Superdex 200 10/600 (Cytiva, Marlborough, US-MA). Eluted fractions were analyzed by SDS-PAGE and Coomassie blue staining; the purest fractions were concentrated, spun down and snap-frozen to be stored at −80°C. GST-MPS1-6xHis and mCherry-MPS1-6xHis had been previously expressed and purified as described in⁵⁴. Expression and purification of NDC80^bonsai had been previously done in³⁸.

In vitro kinase assay and mass-spectrometry

For in vitro kinase assays, recombinant human or Drosophila MBP-fused catalytic domain of wildtype or kinase-dead AAK, or 70nM of N-terminal His6-tagged, recombinant full-length human active AAK (#14-511 Merck Millipore, Germany) were incubated either in the presence or the absence of 14nM of N-terminal GST-fused active HsMPS1/TTK (SignalChem, Richmond, Canada) at 30°C for 30 min in a total volume of 30 μL kinase reaction buffer (5 mM MOPS, pH 7.2, 2.5 mM b-glycerol-phosphate, 5 mM MgCl₂, 1 mM EGTA, 0.4 mM EDTA, 0.25 mM DTT, 0.1 mM ATP. For inhibition of HsMPS1/TTK activity, Compound 5 (Cp5) (³³; gift from René Medema, Oncode Institute, The Netherlands Cancer Institute, The Netherlands) was added to the reaction at a final concentration of 100 nM. For AAK inhibition, MLN8237 (Selleck Chemicals) was added to the reaction at a final concentration of 100 nM. For Western blot analysis of AAK phosphorylation levels, reactions were stopped by addition of Laemmli sample buffer (4% SDS, 10% mercaptoetanol, 0.125 M Tris-HCl, 20% glycerol, 0.004% bromophenol blue) and heated for 5 min at 95ΰC. Peptides were resolved in a 10% SDS-PAGE. The activating autophosphorylation on T676 of human MPS1/TTK and the activating autophosphorylation on T288 of human AAK were detected by western blotting with phospho-specific antibodies. Total protein levels were detected by silver staining of the gel.

For identification of phosphorylated residues, the reaction was stopped by addition of 6 M urea and analyzed by liquid chromatography (LC) coupled to a mass spectrometer (LC/MS-MS). Samples were digested with LysC/Trypsin and prepared for LC-MS/MS analysis as previously described⁵⁵. Peptides (100 ng) were separated on an EASY-nLC 1000 HPLC system (Thermo Fisher Scientific) with a 45 min gradient from 5–60% acetonitrile with 0.1% formic acid and directly sprayed via a nano-electrospray source in an Orbitrap Exploris^™ 480 (Thermo Fisher Scientific). The MS was operated in data-dependent mode, acquiring one survey scan and subsequently 20 MS/MS scans⁵⁶. Resulting raw files were processed with MaxQuant software (v.2.0.1.0) using a database composed of the BL21-DE3 proteome (uniprot release 2024), drosophila AAK, human TTK and common contaminants. Oxidation (M) and phosphorylation (STY) were given as variable modifications and carbamidomethylation (C) as fixed modification⁵⁷. A false discovery rate cut-off of 1% was applied at the peptide, and protein levels and as well on phosphorylated peptides. Samples were run at least twice as technical replicates. Phosphosites were extracted from the phosphor(STY).txt output table. The MS/MS spectrum of pT360 was visualized and exported from MaxQuant.

In vitro AAK T337/T360 antibody validation

To test the sensitivity and specificity of the T337/T360 antibody produced in this study in detecting the AAK phosphorylation by MPS1, 200 nM of MBP-AAK (WT or T337A) was incubated either in the presence or absence of 200 nM mCherry-MPS1 (previously obtained in⁵⁴) at 30°C for 30 min in a total volume of 40 μL kinase reaction buffer (5 mM MOPS, pH 7.2, 2.5 mM b-glycerol-phosphate, 5 mM MgCl₂, 1 mM EGTA, 0.4 mM EDTA, 0.25 mM DTT, 0.1 mM ATP. For inhibition of HsMPS1/TTK activity, Cpd-5 (Cp5) (³³; gift from René Medema, Oncode Institute, The Netherlands Cancer Institute, The Netherlands) was added to the reaction at a final concentration of 100 nM. For Western blot analysis of AAK T337/T360 phosphorylation levels, reactions were stopped by addition of Laemmli sample buffer (4% SDS, 10% mercaptoetanol, 0.125 M Tris-HCl, 20% glycerol, 0.004% bromophenol blue) and heated for 5 min at 95°C. Peptides were resolved in an 8% SDS-PAGE. The T337 AAK phosphorylation by MPS1 was detected by Western blotting with the phospho-specific antibody produced in this study. Total protein levels were detected by Coomassie staining.

NDC80 Bonsai phosphorylation assay

To test in vitro MPS1-mediated potentiation of HEC1/NDC80 phosphorylation by AAK, recombinant MBP-AAK-8xHis at a concentration of 5 μM was pre-incubated for 18 hours at 10°C with 1 mM MnCl2 and 200 nM λ-PP (produced in house) to remove T-loop phosphorylation. Subsequently, MBP-AAK-8xHis was diluted at a final concentration of 70 nM and incubated with NDC80^Bonsai (500 nM), with or without GST-MPS1 (70 nM) for 30 min. Reactions were carried out at 30°C in the presence of 5 μM sodium orthovanadate, and started by adding a mix of ATP (final concentration of 2.5 mM) and MgCl2 (5 mM) to a 40 μL solution of proteins in reaction buffer (50 mM Hepes, pH 7.5, 150 mM NaCl, 1 mM TCEP). Reactions were stopped by addition of 5x Laemmli sample buffer (4% SDS, 10% β-mercaptoethanol, 0.125 M Tris-HCl, 20% glycerol, 0.004% bromophenol blue). Proteins were resolved by 15% SDS-PAGE. The autophosphorylation of T288 on MBP-AAK-8xHis and the phosphorylation of S55 on HEC1 were detected by Western blotting with phospho-specific antibodies. Total protein levels were detected by Western blotting or Coomassie blue staining of the gel.

Western blotting and gel staining

For in vitro kinase assays in Figure 2A, Coomassie staining of proteins resolved by SDS–PAGE was first conducted to determine comparable amounts of recombinant protein in solution to be used in the kinase assays. For Western blotting analysis, resolved proteins were transferred to a nitrocellulose membrane, using the iBlot Dry Blotting System (Thermo Fisher Scientific) according to the manufacturer’s instructions. Membranes were incubated for at least 1 h at room temperature in blocking solution (5% powder milk in PBS1×, 0.05% Tween 20). All primary and secondary antibodies were diluted in the blocking solution. Membranes were incubated with primary antibody solutions overnight at 4°C under constant stirring and washed three times in PBS1×, 0.05% Tween 20 for 10 min each. Then, membranes were incubated with secondary antibody solutions for 1 h at room temperature under constant stirring. Secondary antibodies were conjugated to horseradish peroxidase (Jackson Immuno Research, UK). The blots were visualized by ECL detection and exposure to X-ray films (Fuji Medical X-Ray Film). Silver staining was conducted by gel fixation (in 30% ethanol, 10% acetic acid solution) for 2 h, followed by washing in 20% ethanol solution for 20 min and washing in water for 10 min. Then, the gel was incubated for 1min in 0,2g/L sodium thiosulfate, followed by a quick rinse with water and a 30 min incubation with 2g/L silver nitrate at 4°C. After a quick rinse with water, the gel was incubated with developer solution (10mg/L sodium thiosulfate, 0,025% formaldehyde, 30g/L sodium carbonate). When the adequate degree of staining was achieved, the development was stopped by adding a 50g/L Tris-base, 2,5% acetic acid solution.

For in vitro NDC80^Bonsai phosphorylation assays in Figure S3N, proteins were transferred onto nitrocellulose membranes using the TransBlot Turbo transfer system (BioRad Laboratories, Hercules, US-CA) as per the manufacturer’s protocol. Membranes were blocked for 1 hour at room temperature in 5% milk dissolved in TBS 1x, 0.1% Tween 20. Primary antibodies, diluted in the blocking solution, were incubated overnight at 4°C with constant stirring. Membranes were washed three times (10 min each) with TBS 1×, 0.1% Tween 20, followed by a 1-hour incubation at room temperature with horseradish peroxidase-conjugated secondary antibodies (Amersham/Cytiva, catalog #NXA931-1ML, 1:10000 dilution) in TBS 1×, 0.1% Tween 20. Blots were visualized via ECL detection, and images were acquired using the ChemiDoc MP System (BioRad) with Image Lab 5.1 software.

AlphaFold 3 predictions

AlphaFold 3⁵⁸ was used to model the structure of Drosophila melanogaster AAK and the impact of T360 phosphorylation and T360A and T360D mutations. Models were obtained by inputting the kinase domain sequence (154-406 amino acids) of Drosophila AAK (Q9VGF9) as a single copy ‘Protein’ together with a single ADP ‘Ligand’ and two Mg2+ ‘Ion’ copies. Similarly, the mutants were obtained in the same conditions, but with either Alanine (A) or Aspartic acid (D) replacing Threonine (T) in amino acid position 360. For the Phospho-threonine (pT) model, the appropriate PTM was added onto the original sequence. For each of the five output models, ClashScore analysis and T-loop Confidence scores were assessed in SwissModel to determine the best model for each prediction. Confidence scores can be found in Figure S2N. All images of structures were processed on RCSB PBD - 3D view, where alignments are overall cartoon chain superpositions with ball and stick highlights on aa360, aa317, aa311 and molecular surface of the structure. The color coding was set as cyan blue for AAK^WT, yellow for AAK^T360A, red for AAK^T360D and orange for AAK^pT360.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical analysis

All statistical analysis was performed using GraphPad Prism V10.0 (GraphPad Software, Inc.). Values were considered statistically different whenever P < 0.05. P values: ns, not significant; *< 0.05; **< 0.01; ***< 0.001; ****< 0.0001. Drosophila cells and western blot data are presented as median ± interquartile range or mean ± SD. Statistical analysis were performed using a Kruskal–Wallis test for multiple comparisons or Mann-Whitney test when comparing between two conditions. Details on the specific test, the sample sizes and what is shown are indicated in the respective figure legends. Human cells data points were assessed for normality using the D’Agostino & Pearson test. Based on the results, statistical significance was evaluated either by Student’s t-test (unpaired, two-tailed for normally distributed data) or the Mann-Whitney U-test (unpaired, two-tailed for non-normally distributed data). Variances were compared using the F-test, and Welch’s correction was applied when variances were unequal. Detailed information on statistical significance for each condition can be found in the figures and their legends.

Supplementary Material

2

Video S1. Depletion of MPS1 delays chromosome congression. Related to Figures 3A and 3C. Mitotic progression of control Drosophila S2 cells depleted of CDC27 (left), co-depleted of CDC27 and MPS1 (center) or co-depleted of CDC27 and γTubulin (right). Merge colors of Histone 2B-GFP (green) and mCherry-αtubulin (red) channels are shown. Frames were acquired every 30sec. Time 0 corresponds to nuclear envelope breakdown.

3

Video S2. Centrosomal-tethered MPS1 or a AAKT360D mutant can restore chromosome congression in MPS1-depleted cells. Related to Figure 3A and 3C. Mitotic progression of Drosophila S2 cells co-depleted of CDC27 and MPS1 expressing EGFP-MPS1WT-CM2 (left), expressing EGFP-AAKT360D-CM2 (middle) and EGFPAAKK183I/ T360D-CM2 (right). Merge colors of the expressed transgenes (green), Histone 2B-GFP (green) and mCherry-αTubulin (red) channels are shown. Frames were acquired every 30sec.Time 0 corresponds to nuclear envelope breakdown.

4

Video S3. Phosphorylation of AAK T360 is required for chromosome congression. Related to Figure 3B and 3D. Mitotic progression of Drosophila S2 cells co-depleted of CDC27 and AAK expressing no transgene (left), expressing EGFP-AAKWT (center left), expressing EGFP-AAKT360A (center right) or expressing EGFP-AAKT360D (right) (related to Fig. 3B, D). Merge colors of the expressed transgenes (green), Histone 2B-GFP (green) and mCherry-αTubulin (red) channels are shown. Frames were acquired every 30sec. Time 0 corresponds to nuclear envelope breakdown.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti Aurora A T360Ph, rabbit	This paper	N/A
Anti α-Tubulin B512, mouse	Sigma-Aldrich	RRID: AB_477582
Anti CID, rat	Conde et al.34	Clone Rat4
Anti GFP, chicken	Abcam	RRID: AB_300798
Anti CENP-C, rabbit	Heeger et al.48	Clone Rb1
Anti MPS1, guinea pig	Gilliland et al.49	RRID: AB_2567774
Anti Aurora B T232Ph, rabbit	Rockland	RRID: AB_2061641
Anti Aurora A T288Ph, rabbit	Cell Signaling Technology	RRID: AB_2061481
Anti DPLP, chicken	Rodrigues-Martins et al.50	N/A
Anti FLAG, mouse	Sigma-Aldrich	RRID: AB_262044
Anti γ-Tubulin GTU-88, mouse	Sigma-Aldrich	RRID: AB_532292
Anti D-TACC S863Ph, rabbit	Barros et al.26	N/A
Anti CENP-C, guinea pig	MBL	RRID: AB_10693556
Anti Pericentrin, mouse	Abcam	RRID: AB_2160664
Anti Aurora A, mouse	Abcam	RRID: AB_300667
Anti Hec1 S69Ph, rabbit	DeLuca et al.9	N/A
Anti MPS1 T676Ph, rabbit	Jelluma et al.51	N/A
Anti α-Tubulin DM1a, mouse	Santa Cruz Biotechnology	RRID: AB_628412
Anti Aurora A, mouse	BD	RRID: AB_398252
Anti Aurora B T232Ph, rabbit	Cell Signaling Technology	RRID: AB_2061631
Anti GST, mouse	Sigma-Aldrich	RRID: AB_10807767
Anti NDC80, mouse	Gene Tex	RRID: AB_371632
Anti NDC80 S55Ph, rabbit	Invitrogen	RRID: AB_2792966
Anti chicken Alexa Fluor 488	Invitrogen	RRID: AB_142924
Anti chicken Alexa Fluor 647	Invitrogen	RRID: AB_2535866
Anti rabbit Alexa Fluor 488	Invitrogen	RRID: AB_2576217
Anti rabbit Alexa Fluor 568	Invitrogen	RRID: AB_10563566
Anti rabbit Alexa Fluor 647	Invitrogen	RRID: AB_2536183
Anti rat Alexa Fluor 647	Invitrogen	RRID: AB_141778
Anti guinea pig Alexa Fluor 488	Invitrogen	RRID: AB_2534117
Anti guinea pig Alexa Fluor 568	Invitrogen	RRID: AB_141954
Anti guinea pig Alexa Fluor 647	Invitrogen	RRID: AB_141882
Anti mouse Alexa Fluor 488	Invitrogen	RRID: AB_2534088
Anti mouse Alexa Fluor 568	Invitrogen	RRID: AB_144696
Anti mouse Alexa Fluor 647	Invitrogen	RRID: AB_162542
Anti rabbit HRP conjugated	Jackson ImmunoResearch	RRID: AB_2313567
Anti guinea pig HRP conjugated	Invitrogen	RRID: AB_2535546
Anti mouse HRP conjugated	Jackson ImmunoResearch	RRID: AB_10015289
Anti rabbit HRP conjugated	GE Healthcare	RRID: AB_772191
Anti mouse HRP conjugated	GE Healthcare	RRID: AB_772193
Bacterial and virus strains
E. coli BL21CodonPlus(DE3)-RIL	Agilent Products	CAT#200131
E. coli DH5α competent cells	ThermoFischer	CAT#EC0112
E. coli Omnimax competent bacteria	Invitrogen	CAT#C854003
E. coli BL21(DE3) competente bacteria	ThermoFischer	CAT#C601003
Chemicals, peptides, and recombinant proteins
Protease inhibitors cOmplete cocktail	Roche	CAT#11697498001
Phosphatase inhibitor cocktail 3	Sigma-Aldrich	CAT#P0044
MG132 proteasome inhibitor	Sigma-Aldrich	CAS 133407
Colchicine	Sigma-Aldrich	CAS 64-86-8
Nocodazole	Sigma-Aldrich	CAS 31430-18-9
MLN8054	Selleck Chemicals	CAT#S1100
MLN8237	Selleck Chemicals	CAT#S1133
Compound5 (Cp5)	Koch et al.33	N/A
ZM447439	AstraZeneca	CAS 331771-20-1
GSK923295	MedChemExpress	CAT#HY-10299
STLC Eg-5 inhibitor	Santa Cruz Biotechnology	CAS 2799-07-7
Aurora A Protein, active	Merck, Millipore	CAT#14-511
Recombinant Human Mps1/TTK Protein	SignalChem	CAT#T20-10G
Vectashield Antifade Mounting Medium with DAPI	Vector Laboratories	CAT#H-1200
NDC80bonsai	Cifferi et al.38	N/A
MBP-AAK WT	This paper	N/A
MBP-AAK T337A	This paper	N/A
mCherry-MPS1	Raisch et al.54	N/A
GST-MPS1-6xHis	Raisch et al.54	N/A
Concanavalin A	SigmaAldrich	CAT#C5275
Deposited data
Mass spectrometry data	This paper	Pride Archive: PXD056173
Experimental models: Cell lines
Drosophila: S2 cells	DGRC	RRID: CVCL_TZ72
Human: hTERT RPE-1	ATCC	RRID: CVCL_4388
Oligonucleotides
For oligonucleotides used in this study, see Table S1.	N/A	N/A
Recombinant DNA
pGFP-Histone2B	Conde et al.34	N/A
pAc-mCherry-αTubulin	Conde et al.34	N/A
pCENP-C-CM2-AuroraFRETsensor	This paper	N/A
pHGW-AAKWT	This paper	N/A
pMT-mCherry-MPS1	This paper	N/A
pHGW-AAKT360A	This paper	N/A
pHGW-AAKT360D	This paper	N/A
pMT-EGFP-MPS1WT-CM2	This paper	N/A
pHGW-AAKWT-CM2	This paper	N/A
pHGW-AAKT311D-CM2	This paper	N/A
pHGW-AAKT360D-CM2	This paper	N/A
pHGW-AAKK183I/T360D-CM2	This paper	N/A
pHFMW-AAKWT	This paper	N/A
pHFMW-AAKK183I	This paper	N/A
pMT-EGFP-MPS1	Conde et al.34	N/A
pMalc-AAK 126-403	This paper	N/A
pMalc-AAK 126-403 K162I	This paper	N/A
pMalc-AAK 144-411	This paper	N/A
pMalc-AAK 144-411 K183I	This paper	N/A
pETDuet-AAK WT	Pan et al.52	N/A
pETDuet-AAK T337A	This paper	N/A
pLIB-GST-MPS1	Raisch et al.54	N/A
pLIB-mCherry-MPS1	Raisch et al.54	N/A
pGEX6P-2rbs-NDC80bonsai	Ciferri et al.38	N/A
Software and algorithms
Fiji software	ImageJ	RRID:SCR_002285
AlphaFold3	Abramson et al.58	RRID:SCR_025454
GraphPad Prism 10.2.3	GraphPad Software Inc.	RRID:SCR_002798
KT-SP distance measurement custom routine	Barisic et at.39	N/A
Other
μ-Dish 35mm, high Glass Bottom	Ibidi	CAT#81158
35 mm glass bottom petri dishes	Cellvis	CAT#D35-20-0-N
12 mm round cover glasses	Menzel Glaser	CAT#MENZCB00120RA020
Superdex® 200 SEC column	GE Healthcare	CAT#GE28-9909-44
TranscriptAid T7 High Yield Transcription Kit	(ThermoFischer)	CAT#K0441
MBP amylose magnetic beads	New England Biolabs	CAT#E8035S

HIGHLIGHTS.

• Pole-proximal kinetochores activate centrosomal AAK in an MPS1-dependent manner

• MPS1 activates AAK through phosphorylation of a conserved residue at the C-lobe

• Potentiation of AAK by MPS1 ensures chromosome congression and error correction

• Proximity-based activation of AAK by MPS1 is conserved in human cells