Molecular details and phosphoregulation of the CENP-T-Mis12 complex interaction during mitosis in DT40 cells Authors: Yusuke Takenoshita, Masatoshi Hara, Reiko Nakagawa, Mariko Ariyoshi, Tatsuo Fukagawa Published: 2024 ABSTRACT SummaryTo establish bipolar attachments of microtubules to sister chromatids, an inner kinetochore subcomplex, the constitutive centromere-associated network (CCAN), is assembled on centromeric chromatin and recruits the microtubule-binding subcomplex called the KMN network. Since CCAN proteins CENP-C and CENP-T independently bind to the Mis12 complex (Mis12C) of KMN, it is difficult to evaluate the significance of each interaction in cells. Here, we demonstrate the molecular details of the CENP-T-Mis12C interaction using chicken DT40 cells lacking the CENP-C-Mis12C interaction. Using AlphaFold predictions combined with cell biological and biochemical analyses, we identified three binding surfaces of the CENP-T-Mis12C interaction, demonstrating that each interface is important for recruiting Mis12C to CENP-T in cells. This interaction, via three interaction surfaces, is cooperatively regulated by dual phosphorylation of Dsn1 (a Mis12C component) and CENP-T, ensuring a robust CENP-T-Mis12C interaction and proper mitotic progression. These findings deepen our understanding of kinetochore assembly in cells. Introduction Chromosome segregation during mitosis is critical for transmitting genetic information to the progeny. This is achieved by attaching sister chromatids to bipolar mitotic spindles, leading to chromosome segregation into daughter cells. Spindle microtubules bind to a large protein complex called the kinetochore, which is formed on the centromere of each sister chromatid to ensure accurate chromosome segregation.1,2,3,4 To understand the mechanisms underlying chromosome segregation, it is crucial to clarify how the kinetochore is assembled on the centromere. Kinetochores comprise two major complexes. One is the constitutive centromere-associated network (CCAN), containing 16 protein components that constitutively localize to the centromere throughout the cell cycle, forming a base for kinetochore assembly.2,3,5,6,7,8,9,10 Another large complex is the KMN network, comprising the Knl1 complex (Knl1C), Mis12 complex (Mis12C), and Ndc80 complex (Ndc80C), which are associated with CCAN from G2 phase to mitosis to establish a functional kinetochore structure for spindle microtubule binding.2,6,7,8 Ndc80C in the KMN network directly associates with spindle microtubules,11,12,13 and also associates with CCAN directly or via Mis12C.2,7,8 Since CCAN associates with centromeric chromatin, the linkage between CCAN and the KMN network is critical for bridging centromeric chromatin with spindle microtubules, ultimately ensuring accurate chromosome segregation. In vertebrate cells, the KMN network is recruited to the CCAN through two parallel pathways: the CENP-C and CENP-T pathways.1,3,4,6,14,15,16,17,18,19,20 Mis12C binds to CENP-C and CENP-T independently in each pathway, acting as a hub for KMN network assembly by recruiting Ndc80C and Knl1C.21 In addition, CENP-T directly recruits Ndc80C via the Ndc80C binding domain(s) at its N-terminus (human CENP-T has two binding domains and chicken CENP-T has one domain).15,17,22 Although both pathways contribute to Mis12C recruitment for KMN network assembly at an appropriate time,16,17,18,23,24 the CENP-T-Mis12C interaction has a dominant function in chromosome segregation compared to the CENP-C-Mis12C interaction in chicken DT40 cells.18,24 Previous studies have proposed a molecular basis for the human CENP-T-Mis12C interaction based on in vitro binding assays and structural prediction analyses.22,25,26 However, it remains unclear how the CENP-T-Mis12C interaction is regulated and how this regulatory mechanism is significant in cells, as it is challenging to distinguish CENP-C- and CENP-T-binding Mis12C in native kinetochores. Therefore, clarifying the regulatory mechanisms and biological significance of CENP-T-Mis12C interaction in cells is not trivial. In this study, we examined the molecular details of the CENP-T-Mis12C interaction at the native kinetochore using chicken DT40 mutant cells lacking the Mis12C binding region of CENP-C. Combined with structure prediction and cell biology analyses, we identified two binding sites in CENP-T for Mis12C interaction, which are essential for DT40 cell viability. In addition, a third region of CENP-T seems to contribute to the binding to Mis12C when CENP-T is phosphorylated. We also propose that the dual phosphorylation of Dsn1 and CENP-T cooperatively facilitates the CENP-T-Mis12C interaction, ensuring mitotic progression in DT40 cells. The molecular details of the dual phosphoregulation of the CENP-T-Mis12C interaction provide insights into the robust assembly mechanism of kinetochores for accurate chromosome segregation. Results Localization profile of CENP-C- and CENP-T-binding Mis12C in chicken DT40 cells Both CENP-C and CENP-T contain a binding region for Mis12C14,15,18,22,24,25,27,28; however, it is unclear when and how Mis12C is recruited to each binding region. To distinguish the cell cycle timing of Mis12C recruitment to CENP-C and CENP-T, we examined Mis12C localization at kinetochores in CENP-CWT cells (wild-type chicken DT40 cells) and CENP-CΔ73 cells lacking the Mis12C-binding region of CENP-C (aa 1–73; Figures 1A and S1A). Although depletion of CENP-C leads to chromosome segregation defect due to loss of interactions with various kinetochore proteins,29,30 the Mis12C-binding region of CENP-C is dispensable for chromosome segregation in chicken DT40 cells.18 Therefore, CENP-CΔ73 cells enable the observation of CENP-T-bound Mis12C at the native kinetochore. To visualize Mis12C, we introduced a GFP-Dsn1 construct (a Mis12C component) into the endogenous Dsn1 locus (Figures S1B–S1E). Consistent with our previous findings showing that Mis12C starts to localize to centromeres in interphase and continues to localize during mitosis in chicken DT40 cells,18 GFP-Dsn1 localized to centromeres in both interphase and M phase in CENP-CWT cells (Figure 1B). However, GFP-Dsn1 localized to centromeres only during mitosis and not during interphase in CENP-CΔ73 cells (Figure 1B), suggesting that Mis12C associates only with CENP-C and does not bind to CENP-T in interphase cells.Figure 1CENP-T interacts with Mis12C during mitosis in chicken DT40 cells(A) Schematic representation of chicken CENP-C (864 aa). The Mis12C-binding domain was deleted in CENP-CΔ73.(B) Localization of GFP-Dsn1 in CENP-CWT and CENP-CΔ73 cells. Cells were fixed and stained with an anti-CENP-T antibody. DNA was stained with DAPI. Scale bar, 10 μm.(C) Schematic representation of chicken CENP-T (639 aa). The Mis12C-binding domain was deleted in CENP-TΔ161–216.(D) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells in which wild-type CENP-T protein was degraded by IAA addition. Cells were treated with IAA for 12 h, fixed, and stained with an anti-CENP-A antibody. DNA was stained with DAPI. Scale bar, 10 μm.(E) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (D). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 672.4; p < 0.0001) followed by Tukey’s test.(F) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands.(G) Cell cycle distribution profiles of AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 in the presence of IAA for 12 h.(H) Chromosome alignment values in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 in the presence of IAA for 12 h and MG132 for the last 2 h. Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,57) = 14.78; p < 0.0001) followed by Tukey’s test. CENP-T interacts with Mis12C during mitosis in chicken DT40 cells (A) Schematic representation of chicken CENP-C (864 aa). The Mis12C-binding domain was deleted in CENP-CΔ73. (B) Localization of GFP-Dsn1 in CENP-CWT and CENP-CΔ73 cells. Cells were fixed and stained with an anti-CENP-T antibody. DNA was stained with DAPI. Scale bar, 10 μm. (C) Schematic representation of chicken CENP-T (639 aa). The Mis12C-binding domain was deleted in CENP-TΔ161–216. (D) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells in which wild-type CENP-T protein was degraded by IAA addition. Cells were treated with IAA for 12 h, fixed, and stained with an anti-CENP-A antibody. DNA was stained with DAPI. Scale bar, 10 μm. (E) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (D). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 672.4; p < 0.0001) followed by Tukey’s test. (F) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. (G) Cell cycle distribution profiles of AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 in the presence of IAA for 12 h. (H) Chromosome alignment values in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT or 3xFLAG-CENP-TΔ161–216 in the presence of IAA for 12 h and MG132 for the last 2 h. Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,57) = 14.78; p < 0.0001) followed by Tukey’s test. Previous studies have suggested that the Dsn1 basic motif stabilizes a Mis12C auto-inhibited state,26,31 which prevents the CENP-C-Mis12C interaction during interphase. The Mis12C auto-inhibited state is relieved by phosphorylation of the Dsn1 basic motif by the mitotic kinase Aurora B during mitosis.18,26,28,31 However, Mis12C was recruited to CENP-C in the interphase of DT40 cells (Figure 1B), suggesting that CENP-C binds to Mis12C in the auto-inhibited state. To examine the molecular basis of the chicken CENP-C-Mis12C interaction, we utilized AlphaFold2 to predict the structure of Mis12C, including four full-length proteins (Mis12, Nsl1, Dsn1, and Pmf1) in complex with CENP-C aa 1–73 (the Mis12C-binding region). The predicted structure of chicken Mis12C comprised two head domains (Head1: composed of Mis12 and Pmf1; and Head2: composed of Dsn1 and Nsl1) connected to a coiled-coil stalk domain, which is composed of Mis12, Pmf1, Dsn1, and Nsl1. This structure is similar to the crystal structures of human and yeast Mis12C.28,32 The structure suggests that the Dsn1 basic motif binds to both Head1 and Head2 of Mis12C to form the auto-inhibited Mis12C state, consistent with the cryo-EM structure of the human KMN complex (Figures S1G and S1H).26 The Predicted Aligned Error (PAE) suggests that the aa 1–21 and aa 22–42 regions of CENP-C bind to Mis12C in the auto-inhibited state (Figures S1I and S1J). The aa 1–21 region of CENP-C may bind to Mis12C through Nsl1 and Mis12 (pLDDT <50; Figures S1J and S1K). However, the comparison between the crystal structure of human Mis12C in complex with CENP-C28 and the cryo-EM structure of the human KMN complex26,31 suggests that the N-terminus of human CENP-C, which corresponds to the aa 1–21 of chicken CENP-C, causes a steric clash with the basic motif of human Dsn1. Similar to the human case, it is possible that the aa 1–21 of chicken CENP-C might also cause a steric clash with the basic motif of Dsn1. The aa 22–42 region of CENP-C also binds to Mis12C through Pmf1 and Mis12 (90 > pLDDT >50; Figures S1J and S1K). Even if the aa 1–21 region of CENP-C causes a steric clash with the Dsn1 basic motif, the Mis12C-CENP-C interaction should be maintained using the aa 22–42 region of CENP-C. This finding is consistent with previous in vitro binding assay showing that human CENP-C binds to human Mis12C in the auto-inhibited state.26 We previously demonstrated that Tet-off-based CENP-T conditional knockout CENP-CΔ73 cells expressing CENP-TΔ161–200 or CENP-TΔ201–216 showed a significant reduction in Mis12C localization at kinetochores compared to CENP-CΔ73 cells expressing CENP-TWT.24 This suggests that both the aa 161–200 and aa 201–216 regions of CENP-T are important for interaction with Mis12C. To efficiently examine the molecular mechanisms of interaction between CENP-T and Mis12C at centromeres using CENP-CΔ73 cells, we used Auxin-Inducible Degron (AID)-based knockdown cells, instead of Tet-off-based knockout cells. We expressed AID-tagged CENP-T and introduced an FLAG-fused CENP-T mutant construct lacking the aa 161–216 region into the endogenous CENP-T locus in CENP-CΔ73 cells (CENP-CΔ73/CENP-TΔ161–216 cells; Figures 1C and S2A–S2H). In this cell line, wild-type CENP-T was degraded after 3-indole acetic acid (IAA) addition, and instead of wild-type CENP-T, only CENP-TΔ161–216 was expressed (Figure S2H). As a control, the FLAG-fused wild-type CENP-T construct was introduced into the CENP-T locus (CENP-CΔ73/CENP-TWT cells; Figures S2F–S2H). Next, we examined the mitotic localization of Mis12C in these cells. The mitotic GFP-Dsn1 localization observed in CENP-CΔ73/CENP-TWT cells was almost undetectable in CENP-CΔ73/CENP-TΔ161–216 cells (Figures 1D and 1E). Although AID-fused CENP-T is not completely degraded after IAA addition (Figure S2H), the Mis12C levels at the kinetochore reduced to ∼15% in CENP-CΔ73/CENP-TΔ161–216 cells (Figure 1E), suggesting the AID-based CENP-T knockdown cells are appropriate for the following experiments. Consistent with this result, Mis12C co-immunoprecipitated with CENP-TWT but not with CENP-TΔ161–216 (Figure 1F). In addition to the CENP-T-Mis12C interaction data, we examined the cell cycle profile of CENP-CΔ73/CENP-TΔ161–216 cells by flow cytometry. The G2/M, SubG1, and Others fractions were increased in CENP-CΔ73/CENP-TΔ161–216 cells, similar to AID-based CENP-T knockdown in CENP-CΔ73 cells (Figures 1G and S2I), suggesting that mitotic (G2/M) arrest and subsequent cell death (SubG1 and Others) occurred in CENP-CΔ73/CENP-TΔ161–216 cells. To further analyze the mitotic phenotype of CENP-CΔ73/CENP-TΔ161–216 cells, we examined the chromosome alignment of mitotic cells. As shown in Figure 1H, chromosomes were well aligned in the metaphase plate in CENP-CΔ73/CENP-TWT cells, whereas chromosomes were not aligned in CENP-CΔ73/CENP-TΔ161–216 cells and AID-based CENP-T knockdown CENP-CΔ73 cells. Taken together, we propose that the aa 161–216 region of CENP-T is required for Mis12C mitotic kinetochore localization via CENP-T-Mis12C interaction in CENP-CΔ73 cells and is critical for proper mitotic progression due to its binding to Mis12C. Structural prediction of interfaces for the chicken CENP-T-Mis12C interaction Next, to examine the molecular basis of the chicken CENP-T-Mis12C interaction, we utilized AlphaFold2 to predict the structure of Mis12C, including four full-length proteins (Mis12, Nsl1, Dsn1, and Pmf1) in complex with the aa 161–216 region of CENP-T (Figures 2A and S3A–S3D). The predicted structure of Mis12C were similar to the structure of Mis12C complexed with CENP-C. The model of CENP-T161–216 contains two α-helices connected via an extended linker region.Figure 2Molecular basis for the chicken CENP-T-Mis12C interaction based on a complex structure predicted by AlphaFold2(A) AlphaFold2 predicted structure of chicken Mis12C in complex with the Mis12C-binding domain of CENP-T (rank_1 model; rank_2, 3, 4, and 5 models are shown in Figure S3). The CENP-T aa 161–216 region and the full lengths of Mis12C components (Mis12, Nsl1, Dsn1, and Pmf1) were used for the prediction, displayed as a cartoon, and color-coded as indicated in the figure.(B) Overview of the CENP-T-Mis12C interface. Mis12C is shown as a surface model, and the CENP-T aa 161–216 region is shown as a cartoon. The side chains of key residues of CENP-T for binding to Mis12C are shown with labels.(C) Schematic representation of the CENP-T aa 161–216 structure. There are two binding regions for Mis12C (binding regions 1 and 2). Binding region 1 interacts with the acidic surface of Nsl1, and binding region 2 binds to the Head1 region of Mis12C via hydrophobic interactions. Molecular basis for the chicken CENP-T-Mis12C interaction based on a complex structure predicted by AlphaFold2 (A) AlphaFold2 predicted structure of chicken Mis12C in complex with the Mis12C-binding domain of CENP-T (rank_1 model; rank_2, 3, 4, and 5 models are shown in Figure S3). The CENP-T aa 161–216 region and the full lengths of Mis12C components (Mis12, Nsl1, Dsn1, and Pmf1) were used for the prediction, displayed as a cartoon, and color-coded as indicated in the figure. (B) Overview of the CENP-T-Mis12C interface. Mis12C is shown as a surface model, and the CENP-T aa 161–216 region is shown as a cartoon. The side chains of key residues of CENP-T for binding to Mis12C are shown with labels. (C) Schematic representation of the CENP-T aa 161–216 structure. There are two binding regions for Mis12C (binding regions 1 and 2). Binding region 1 interacts with the acidic surface of Nsl1, and binding region 2 binds to the Head1 region of Mis12C via hydrophobic interactions. AlphaFold2 predicted that CENP-T161–216 binds across the stalk domain to the Head1 domain of Mis12C (Figures 2A, 2B, and S3C–S3E). The pLDDT scores indicate that the aa 161–186 region of CENP-T has lower confidence (pLDDT <70), whereas the aa 187–216 region has higher confidence (pLDDT >70) and adopts a more reliable structure (Figure S3E). The PAE also shows that the aa 187–216 region of CENP-T binds to Mis12C with high reliability (Figure S3C; rank_1). The predicted structure suggested two main Mis12C-binding regions of CENP-T (binding region 1 and 2; Figures 2B and 2C). In the binding region 1, R192 and R193 of CENP-T (90 > pLDDT >70) in the extended region is supposed to interact with the Nsl1 acidic surface in the stalk region of Mis12C (Figures 2B and 2C). In the binding region 2, the C-terminal α-helix of CENP-T161–216 harboring I202, F206, and V209 (100 > pLDDT >70) is predicted to associate with the hydrophobic surface of the Head1 domain of Mis12C (Figures 2B and 2C). The AlphaFold2 predicted structure of human Mis12C in complex with the aa 176–232 region of human CENP-T is similar to the predicted structure of chicken Mis12C in complex with chicken CENP-T161–216 (Figures S3F–S3I). As these two Mis12C-binding regions of CENP-T seem important for the CENP-T-Mis12C interaction, we further analyzed the molecular details of the CENP-T-Mis12C interaction and its significance in CENP-CΔ73 DT40 cells. Electrostatic interaction of Mis12C with the binding region 1 of CENP-T As described above, the AlphaFold2 model suggested that the two arginine residues of CENP-T (R192 and R193) in the binding region 1 form electrostatic interactions with the acidic surface of Nsl1 in Mis12C (Figure 3A). These arginine residues of CENP-T and the acidic region of Nsl1 are highly conserved (Figures S4A and S4B). Human CENP-T residues R209 and R210, corresponding to R192 and R193 of chicken CENP-T, are predicted to bind to the acidic surface of human Nsl1 (Figure S3H). To evaluate the importance of this interaction surface in chicken DT40 cells, we replaced R192 and R193 with alanine in CENP-T (CENP-TRR-AA) and introduced a mutant construct into the CENP-T locus in CENP-CΔ73 cells expressing AID-CENP-T (CENP-CΔ73/CENP-TRR-AA cells; Figure S4C). We examined the GFP-Dsn1 levels at the mitotic kinetochores in these cells and found that these levels were ∼35% of those in CENP-CΔ73/CENP-TWT cells, which were slightly higher than that in CENP-CΔ73/CENP-TΔ161–216 cells (Figures 3B and 3C). We also confirmed that Mis12C did not co-precipitate with CENP-TRR-AA like CENP-TΔ161–216 (Figure 3D). Consistent with these results, we observed a clear growth delay in CENP-CΔ73/CENP-TRR-AA cells following IAA addition (Figure S4D). Although we cannot completely rule out the possibility that CENP-TRR-AA indirectly reduced Mis12C binding, we propose that CENP-T R192 and R193 form electrostatic interactions with the Nsl1 acidic region, which is important for the CENP-T-Mis12C binding and cell proliferation.Figure 3Two binding regions of CENP-T are required for the chicken CENP-T-Mis12C interaction(A) The magnified view of the Nsl1 acidic surface associated with the binding region 1 of CENP-T. Left: The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. CENP-T R192 and R193 and Nsl1 acidic residues (E121, D122, D125, and E126) are labeled. Right: Electrostatic potential of Mis12C mapped on the surface representation. CENP-T is shown as a cartoon.(B) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TRR-AA was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells in which wild-type CENP-T protein was degraded by IAA addition. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm.(C) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (B). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 262.6; p < 0.0001) followed by Tukey’s test.(D) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TRR-AA. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands.(E) Magnified view of the Mis12C-Head1 hydrophobic surface interacting with the binding region 2 of CENP-T. Left: The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. CENP-T I202, F205, and V209 are labeled. Right: Hydrophobic potential of Mis12C mapped on the surface representation. CENP-T I202, F205, and V209 are shown as sticks.(F) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TIFV-RRR was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm.(G) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (F). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 284.8; p < 0.0001) followed by Tukey’s test.(H) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TIFV-RRR. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. Two binding regions of CENP-T are required for the chicken CENP-T-Mis12C interaction (A) The magnified view of the Nsl1 acidic surface associated with the binding region 1 of CENP-T. Left: The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. CENP-T R192 and R193 and Nsl1 acidic residues (E121, D122, D125, and E126) are labeled. Right: Electrostatic potential of Mis12C mapped on the surface representation. CENP-T is shown as a cartoon. (B) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TRR-AA was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells in which wild-type CENP-T protein was degraded by IAA addition. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm. (C) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (B). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 262.6; p < 0.0001) followed by Tukey’s test. (D) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TRR-AA. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. (E) Magnified view of the Mis12C-Head1 hydrophobic surface interacting with the binding region 2 of CENP-T. Left: The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. CENP-T I202, F205, and V209 are labeled. Right: Hydrophobic potential of Mis12C mapped on the surface representation. CENP-T I202, F205, and V209 are shown as sticks. (F) Localization of GFP-Dsn1. 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TIFV-RRR was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm. (G) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (F). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 284.8; p < 0.0001) followed by Tukey’s test. (H) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/GFP-Dsn1 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-TIFV-RRR. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. Hydrophobic interaction of Mis12C with the binding region 2 of CENP-T Structure prediction also suggested that the hydrophobic residues of CENP-T (I202, F205, and V209) aligning on the α-helix in the binding region 2 were associated with the hydrophobic surface of the Mis12C Head1 domain (Figure 3E). Although the hydrophobic residues of CENP-T are not thoroughly conserved in other species, the corresponding regions in human and mouse CENP-T contain a cluster of hydrophobic residues (Figure S4A). The human CENP-T residues V219, F222, and L226, corresponding to I202, F205, and V209 of chicken CENP-T, are predicted to bind to the hydrophobic region of human Mis12C Head1 domain (Figure S3I). To evaluate the importance of this hydrophobic interface, we replaced I202, F205, and V209 of chicken CENP-T with arginine (CENP-TIFV-RRR) and introduced a mutant construct into the CENP-T locus of CENP-CΔ73 cells expressing AID-fused CENP-T (CENP-CΔ73/CENP-TIFV-RRR cells; Figure S4E). We observed a significant reduction in GFP-Dsn1 levels in the mitotic kinetochores of CENP-CΔ73/CENP-TIFV-RRR cells compared to those of CENP-CΔ73/CENP-TWT cells (Figures 3F and 3G), and Mis12C did not co-precipitate with CENP-TIFV-RRR (Figure 3H). As in the case of CENP-CΔ73/CENP-TΔ161–216 cells, after IAA addition, CENP-CΔ73/CENP-TIFV-RRR cells stopped growing and ultimately died (Figure S4D). As the case of CENP-TRR-AA mutant, we cannot rule out the possibility that CENP-TIFV-RRR may indirectly cause reduction of Mis12C-binding. However, our results strongly suggest that the hydrophobic interaction of the binding region 2 of CENP-T with Mis12C is critical for the CENP-T-Mis12C binding and cell viability. In summary, CENP-T has at least two binding regions for Mis12C, each important for its binding to Mis12C. This result is consistent with our previous finding that either the CENP-T aa 161–200 (containing binding region 1) or the aa 201–216 (containing binding region 2) is required for CENP-T function via binding to Mis12C.24 Phosphorylation of Dsn1 in Mis12C facilitates stable CENP-T-Mis12C interaction Although CENP-C interacts with Mis12C in both interphase and M phase, CENP-T interacts with Mis12C during mitosis but not in interphase (Figure 1). This implies that there are regulatory mechanisms by which CENP-T binds to Mis12C only during mitosis. The structural model of the human KMN complex suggests that the Dsn1 basic motif binds to the Head1 and Head2 of Mis12C, which forms the closed conformation of the Head domains and masks the Nsl1 acidic surface (auto-inhibited state).26,31 The corresponding region of chicken Nsl1 is recognized by the CENP-T binding region 1 (Figure 3A). Our AlphaFold2 prediction model for chicken Mis12C also suggests that the Dsn1 basic motif binds to the Head2 of Mis12C via K110 in the RKSL motif (R109, K110, S111, and L112)26 and the hydrophobic pocket of Head1 (Mis12 and Pmf1) and acidic surface of Nsl1 via the WRR linchpin (W98, R99, and R100)26 and additional lysine residue (K104; Figure 4A). These interactions might result in Mis12C adopting an auto-inhibited state. The Dsn1 basic motif possesses Aurora B phosphorylation sites,16,33 and phosphorylation by Aurora B at these sites inhibits the binding of the Dsn1 basic motif to the Head domains of Mis12C.26,28,31,32 Thus, the interaction between the binding region 1 of CENP-T and the Nsl1 acidic surface might be inhibited by the Dsn1 basic motif during interphase; this inhibition is released during mitosis by Aurora B phosphorylation, similarly to the CENP-C-Mis12C interaction.18,28,32 Although this possibility was discussed based on an in vitro binding assay and structural model of human Mis12C,25,26 it is unclear whether this mechanism regulates the CENP-T-Mis12C interaction in cells. To test this hypothesis, we generated a Dsn1 construct lacking the basic motif (aa 93–114; Dsn1Δ93–114) and introduced it into the endogenous Dsn1 locus of CENP-CΔ73 cells (Figures 4B, S1D, and S5A). While Dsn1WT localized to kinetochores only during mitosis, Dsn1Δ93–114 localized to both interphase and mitotic kinetochores in CENP-CΔ73 cells (Figure 4C), suggesting that the Dsn1 basic motif inhibits the CENP-T-Mis12C interaction during interphase. While other mechanisms that inhibit the CENP-T-Mis12C interaction during interphase cannot be ruled out, the inhibitory effect of the Dsn1 basic motif is significant. Since both binding region 1 and 2 of CENP-T are important for the Mis12C binding (Figure 3), CENP-T seems to interact with Mis12C via binding region 1 and 2 in CENP-CΔ73/Dsn1Δ93–114 interphase cells. Combined, we propose that the Dsn1 basic motif masks the binding surface of Mis12C for binding region 1 of CENP-T, and the mask is released upon phosphorylation, leading to the interaction between binding region 1 of CENP-T and Mis12C during mitosis. In turn, the release of the mask also facilitates the interaction between binding region 2 of CENP-T and Mis12C (Figure S5B). These results suggest a linkage between the regulatory mechanisms governing the interactions of binding region 1 and 2 of CENP-T with Mis12C.Figure 4The Dsn1 basic motif regulates the chicken CENP-T-Mis12C interaction(A) Overview of the Mis12C auto-inhibited state in the AlphaFold2 predicted structure of chicken Mis12C. Left: Mis12C is shown as a cartoon and color-coded, as indicated in the figure. Middle: The transparent surface model of Mis12C, except for the basic motif of Dsn1, is overlaid on a cartoon model. The basic motif of Dsn1 is shown as a cartoon. The key residues of Dsn1 important for maintaining the auto-inhibited state are labeled. Right: Electrostatic potential of Mis12C, except for the basic motif of Dsn1, mapped on the surface representation. The basic motif of Dsn1 is shown as a cartoon.(B) Schematic representation of chicken Dsn1 (349 aa). The basic motif was deleted in Dsn1Δ93–114. Amino acid sequences are shown in Figure S5A.(C) Localization of GFP-Dsn1WT or GFP-Dsn1Δ93–114 in CENP-CΔ73 cells. Cells were fixed and stained with an anti-CENP-T antibody. DNA was stained with DAPI. Scale bar, 10 μm.(D) Localization of GFP-Dsn1. GFP-Dsn1WT or GFP-Dsn14A, in which Aurora B phosphorylation sites are substituted with alanine (4A: S97A, S101A, S102A, and S111A), was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/CENP-TWT cells or CENP-CΔ73/CENP-TΔ161–216 cells. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm.(E) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (D). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 245.1; p < 0.0001) followed by Tukey’s test.(F) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/CENP-TWT cells or CENP-CΔ73/CENP-TΔ161–216 cells expressing GFP-Dsn1WT or GFP-Dsn14A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. The Dsn1 basic motif regulates the chicken CENP-T-Mis12C interaction (A) Overview of the Mis12C auto-inhibited state in the AlphaFold2 predicted structure of chicken Mis12C. Left: Mis12C is shown as a cartoon and color-coded, as indicated in the figure. Middle: The transparent surface model of Mis12C, except for the basic motif of Dsn1, is overlaid on a cartoon model. The basic motif of Dsn1 is shown as a cartoon. The key residues of Dsn1 important for maintaining the auto-inhibited state are labeled. Right: Electrostatic potential of Mis12C, except for the basic motif of Dsn1, mapped on the surface representation. The basic motif of Dsn1 is shown as a cartoon. (B) Schematic representation of chicken Dsn1 (349 aa). The basic motif was deleted in Dsn1Δ93–114. Amino acid sequences are shown in Figure S5A. (C) Localization of GFP-Dsn1WT or GFP-Dsn1Δ93–114 in CENP-CΔ73 cells. Cells were fixed and stained with an anti-CENP-T antibody. DNA was stained with DAPI. Scale bar, 10 μm. (D) Localization of GFP-Dsn1. GFP-Dsn1WT or GFP-Dsn14A, in which Aurora B phosphorylation sites are substituted with alanine (4A: S97A, S101A, S102A, and S111A), was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/CENP-TWT cells or CENP-CΔ73/CENP-TΔ161–216 cells. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm. (E) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (D). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 245.1; p < 0.0001) followed by Tukey’s test. (F) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/CENP-TWT cells or CENP-CΔ73/CENP-TΔ161–216 cells expressing GFP-Dsn1WT or GFP-Dsn14A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. Next, we generated Dsn14A, in which the four Aurora B phosphorylation sites of Dsn1 (Figure S5A) were substituted with alanine and introduced into the endogenous Dsn1 locus in CENP-CΔ73 cells expressing AID-fused CENP-T (CENP-CΔ73/Dsn14A cells; Figures S2D, S2G, and S5C). We quantified the GFP-Dsn1 levels in these cells and found that they were reduced to approximately 55% relative to CENP-CΔ73/Dsn1WT cells (Figures 4D and 4E). We also demonstrated that Dsn14A did not co-precipitate well with CENP-T compared with Dsn1WT (Figure 4F). The growth of CENP-CΔ73/Dsn14A cells was comparable to that of CENP-CΔ73/Dsn1WT cells (Figure S5D), suggesting that half of the Dsn1 was sufficient for the viability of CENP-CΔ73 cells. The expression of Dsn14A did not drastically suppress Mis12C localization during mitosis in CENP-CΔ73 cells unlike the expression of CENP-TRR-AA (with mutations in binding region 1 of CENP-T). Since the presence of the Dsn1 basic motif completely suppresses the Mis12C localization to kinetochores during interphase in CENP-CΔ73 cells (Figure 4C), in addition to phosphorylation of the Dsn1 basic motif by Aurora B, there should be a regulatory mechanism for the CENP-T-Mis12C interaction. CENP-T phosphorylation is involved in the regulation of the CENP-T-Mis12C interaction As the additional regulatory mechanism for the CENP-T-Mis12C interaction, CENP-T-mediated phosphorylation is one of the candidates because chicken CENP-T is highly phosphorylated during mitosis.24 First, we examined the phosphorylation sites in the CENP-T aa 161–216 region using mitotic DT40 cell extracts by mass spectrometry and found that S175 and T184 were phosphorylated during mitosis (Figures 5A and S6A). T184 in chicken CENP-T might correspond to S201 of human CENP-T, which is phosphorylated by CDK122,25; however, surrounding sequences of S175 in chicken CENP-T are not clearly conserved in other species (Figure S4A), and it is unclear which mitotic kinase phosphorylates this site. As these two residues were not included in the binding regions 1 and 2, we re-evaluated the AlphaFold 2 prediction model of Mis12C in complex with the aa 161–216 region of CENP-T. As shown in Figure 5B, S175 and T184 are close to the basic regions of Nsl1 and Pmf1, suggesting that phosphorylation of these CENP-T residues might provide the additional Mis12C binding sites mediated via electrostatic interactions and stabilize the CENP-T-Mis12C interface. On the other hand, the AlphaFold3 prediction with phosphorylated S175 and T184 suggested that Mis12C might not directly interact with phosphorylated S175 and T184 of CENP-T (Figures S6B and S6C), but the spatial proximity of the charges might still have a significant effect on the affinity. We generated a CENP-T mutant construct in which S175 and T184 were substituted with alanine and introduced into the CENP-T locus in CENP-CΔ73 cells expressing AID-fused CENP-T (CENP-CΔ73/CENP-T2A cells; Figures S2G and S6D). We examined GFP-Dsn1 levels at kinetochores in CENP-CΔ73/CENP-T2A cells and found that GFP-Dsn1 levels were significantly reduced compared to those in CENP-CΔ73/CENP-TWT cells (∼70% of those in CENP-CΔ73/CENP-TWT cells; Figures 5C and 5D). We also found that the levels of Mis12C that co-precipitated with CENP-T2A were lower than those with CENP-TWT (Figure 5E). These results are consistent with previous studies that showed phosphorylation of human CENP-T facilitates its interaction with Mis12C in vitro.22,25 Based on these results, we propose that CENP-T phosphorylation contributed to facilitating the CENP-T-Mis12 interaction (Figure S6E; see discussion). However, as in the case of CENP-CΔ73/Dsn14A cells, substantial growth defects were not observed in CENP-CΔ73/CENP-T2A cells (Figure S6F), indicating that CENP-CΔ73 cells are still viable even when the interaction between CENP-T phospho-sites and Mis12C is impaired.Figure 5Phosphorylation of CENP-T regulates the chicken CENP-T-Mis12C interaction(A) S175 and T184 of CENP-T are phosphorylated in mitotic DT40 cells based on phospho-mass spectrometry analysis.(B) AlphaFold2 predicted structure of chicken Mis12C in complex with the Mis12C-binding domain of CENP-T. The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. S175 and T184 of CENP-T are labeled. The dashed box shows the electrostatic potential of Mis12C mapped on the surface representation. CENP-T is also shown as a cartoon.(C) Localization of GFP-Dsn1. 3xFLAG-CENP-T WT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells. CENP-CΔ73/CENP-TWT/Dsn1WT cells and CENP-CΔ73/CENP-TΔ161–216/Dsn1WT cells were also used in previous experiments. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm.(D) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (C). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 102.4; p < 0.0001) followed by Tukey’s test.(E) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown GFP-Dsn1/CENP-CΔ73 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. Phosphorylation of CENP-T regulates the chicken CENP-T-Mis12C interaction (A) S175 and T184 of CENP-T are phosphorylated in mitotic DT40 cells based on phospho-mass spectrometry analysis. (B) AlphaFold2 predicted structure of chicken Mis12C in complex with the Mis12C-binding domain of CENP-T. The transparent surface model of Mis12C is overlaid on a cartoon model. CENP-T is shown as a cartoon and models are color-coded as indicated in the figure. S175 and T184 of CENP-T are labeled. The dashed box shows the electrostatic potential of Mis12C mapped on the surface representation. CENP-T is also shown as a cartoon. (C) Localization of GFP-Dsn1. 3xFLAG-CENP-T WT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73 cells. CENP-CΔ73/CENP-TWT/Dsn1WT cells and CENP-CΔ73/CENP-TΔ161–216/Dsn1WT cells were also used in previous experiments. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm. (D) Quantification data of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (C). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (2,27) = 102.4; p < 0.0001) followed by Tukey’s test. (E) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown GFP-Dsn1/CENP-CΔ73 cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. Double mutant cells expressing CENP-T2A and Dsn14A cause a more substantial reduction of Mis12C levels at kinetochores than single mutant cells Although phosphorylation of Dsn1 and CENP-T facilitates the CENP-T-Mis12C interaction at different binding interfaces, inhibition of either phosphorylation did not impair CENP-CΔ73 cell proliferation. This prompted us to examine the phenotype of double mutant cells expressing CENP-T2A and Dsn14A. We introduced a construct containing either GFP-Dsn1WT or GFP-Dsn14A into the endogenous Dsn1 locus of CENP-CΔ73/CENP-T2A cells (Figures S2G and S7A). We then analyzed the effect of CENP-T and Dsn1 double mutations on Mis12C localization at the kinetochores. GFP-Dsn1 levels at kinetochores in CENP-CΔ73/CENP-TWT/Dsn14A and CENP-CΔ73/CENP-T2A/Dsn1WT cells were reduced by ∼65% and ∼70%, respectively, compared to those in CENP-CΔ73/CENP-TWT/Dsn1WT cells (Figures 6A and 6B). However, Dsn1 levels at kinetochores in CENP-CΔ73/CENP-T2A/Dsn14A cells were reduced to approximately 30% of those in CENP-CΔ73/CENP-TWT/Dsn1WT cells (Figures 6A and 6B and Table S1), indicating an additive reduction effect of Dsn14A and CENP-T2A on Mis12C levels at kinetochores. This additive effect was confirmed using co-immunoprecipitation. GFP-Dsn1 levels were further decreased in the CENP-T-immunoprecipitated fraction of CENP-CΔ73/CENP-T2A/Dsn14A cells, compared to those in the CENP-CΔ73/CENP-TWT/Dsn14A or CENP-CΔ73/CENP-T2A/Dsn1WT cells (Figure 6C). Consistent with these results, CENP-CΔ73/CENP-T2A/Dsn14A cells showed a growth delay (Figure S7B). Then, we found that the M phase proportion in CENP-CΔ73/CENP-T2A/Dsn14A cells was significantly increased compared to that in CENP-CΔ73/CENP-TWT/Dsn1WT, CENP-CΔ73/CENP-TWT/Dsn14A, or CENP-CΔ73/CENP-T2A/Dsn1WT cells (Figure 6D). We also assessed the sensitivity of these cells to low doses of nocodazole, inhibitor of microtubule polymerization. While the 5 ng/mL nocodazole treatment did not affect the cell viability of CENP-CΔ73/CENP-TWT/Dsn1WT, CENP-CΔ73/CENP-TWT/Dsn14A, or CENP-CΔ73/CENP-T2A/Dsn1WT cells, the cell viability of CENP-CΔ73/CENP-T2A/Dsn14A cells was decreased by the treatment with the same concentration of nocodazole (Figures 6E and 6F), suggesting that CENP-CΔ73/CENP-T2A/Dsn14A cells are sensitive to the low-dose nocodazole treatment possibly due to the reduction in the kinetochore-localized Mis12C levels. Based on these results, phosphorylation of both Dsn1 and CENP-T is important for timely and stable CENP-T-Mis12C interaction during mitosis, and this regulatory mechanism is critical for mitotic progression and proliferation of CENP-CΔ73 cells.Figure 6Dual phosphorylation of Dsn1 and CENP-T ensures the timely and robust chicken CENP-T-Mis12C interaction(A) Localization of GFP-Dsn1: 3xFLAG-CENP-T WT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells. CENP-CΔ73/CENP-TWT/Dsn1WT cells, CENP-CΔ73/CENP-TΔ161–216/Dsn1WT cells, CENP-CΔ73/CENP-TWT/Dsn14A cells, and CENP-CΔ73/CENP-T2A/Dsn1WT cells were also used in previous experiments. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm.(B) Quantification of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (A). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (4,45) = 89.33; p < 0.0001) followed by Tukey’s test.(C) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands.(D) The ratio of M phase population in indicated cell lines. Cells were treated with IAA for 12 h and fixed. DNA was stained with DAPI. Populations for M phase and interphase were counted in each cell line, and the ratio of M phase cells was calculated. The experiments were repeated three times, and error bars show the mean of the three replicates ±SD; p-values were calculated using one-way ANOVA (F (3,8) = 67.87; p < 0.0001) followed by Tukey’s test.(E) Viability of cells treated with various nocodazole concentrations in indicated cell lines. 3xFLAG-CENP-T WT or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells, treated with IAA for 12 h, and added nocodazole. Cell viability was measured 24 h after nocodazole addition. The experiments were repeated three times, and error bars show the mean of the three replicates ±SD.(F) Half-maximal inhibitory concentration (IC50) of nocodazole calculated using the data shown in (E). The experiments were repeated three times, and error bars show the mean of the three replicates ±SD; p-values were calculated by one-way ANOVA (F (3,8) = 17.76; p = 0.0007) followed by Tukey’s test. Dual phosphorylation of Dsn1 and CENP-T ensures the timely and robust chicken CENP-T-Mis12C interaction (A) Localization of GFP-Dsn1: 3xFLAG-CENP-T WT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells. CENP-CΔ73/CENP-TWT/Dsn1WT cells, CENP-CΔ73/CENP-TΔ161–216/Dsn1WT cells, CENP-CΔ73/CENP-TWT/Dsn14A cells, and CENP-CΔ73/CENP-T2A/Dsn1WT cells were also used in previous experiments. Cells were treated with IAA for 12 h, fixed, and stained with an anti-FLAG antibody. DNA was stained with DAPI. Scale bar, 10 μm. (B) Quantification of GFP-Dsn1 signals at kinetochores in mitotic cells shown in (A). Error bars show the mean ± SD; p-values were calculated by one-way ANOVA (F (4,45) = 89.33; p < 0.0001) followed by Tukey’s test. (C) Immunoprecipitation using anti-FLAG antibody in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells expressing 3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, or 3xFLAG-CENP-T2A. Cells were cultured with IAA for 12 h and nocodazole was added for last 10 h before immunoprecipitation experiments. Immunoprecipitated samples were subjected to immunoblot analyses using anti-CENP-T, anti-FLAG, and anti-Dsn1 antibodies. Asterisk indicates nonspecific bands. (D) The ratio of M phase population in indicated cell lines. Cells were treated with IAA for 12 h and fixed. DNA was stained with DAPI. Populations for M phase and interphase were counted in each cell line, and the ratio of M phase cells was calculated. The experiments were repeated three times, and error bars show the mean of the three replicates ±SD; p-values were calculated using one-way ANOVA (F (3,8) = 67.87; p < 0.0001) followed by Tukey’s test. (E) Viability of cells treated with various nocodazole concentrations in indicated cell lines. 3xFLAG-CENP-T WT or 3xFLAG-CENP-T2A was expressed in AID-based CENP-T conditional knockdown CENP-CΔ73/Dsn1WT cells or CENP-CΔ73/Dsn14A cells, treated with IAA for 12 h, and added nocodazole. Cell viability was measured 24 h after nocodazole addition. The experiments were repeated three times, and error bars show the mean of the three replicates ±SD. (F) Half-maximal inhibitory concentration (IC50) of nocodazole calculated using the data shown in (E). The experiments were repeated three times, and error bars show the mean of the three replicates ±SD; p-values were calculated by one-way ANOVA (F (3,8) = 17.76; p = 0.0007) followed by Tukey’s test. Discussion In this study, we demonstrated that CENP-T binds to Mis12C mainly via two interaction regions (binding regions 1 and 2), and each of them is important for Mis12C binding. In addition, a third region of CENP-T containing residues S175 and T184 seems to contribute to the binding to Mis12C when CENP-T is phosphorylated. These interactions are cooperatively regulated by dual phosphorylation of Dsn1 and CENP-T, ensuring timely and robust CENP-T-Mis12C interactions during mitosis (Figure 7; see below).Figure 7A model of the regulatory mechanism of the CENP-T-Mis12C interaction during interphase and M phaseA model of the regulatory mechanism of the CENP-T-Mis12C interaction. In interphase, the Dsn1 basic motif masks the Nsl1 acidic surface, inhibiting the interaction between binding region 1 of CENP-T and the Nsl1 acidic surface of Mis12C. This inhibition leads to the inhibition of the interaction between binding region 2 of CENP-T and the Head1 hydrophobic surface of Mis12C. Since CENP-T is not phosphorylated in interphase, S175 and T184 of CENP-T do not interact with the stalk basic surface of Mis12C. In M phase, the Dsn1 basic motif is precluded from the Nsl1 acidic surface by the CENP-T and Dsn1 dual phosphoregulation. Phosphorylation of the Dsn1 basic motif by Aurora B precludes the mask by inhibiting the basic motif binding to the Nsl1 acidic surface. Phosphorylated residues of CENP-T may interact with Mis12C basic surface, helping to preclude the mask by increasing the overall affinity of the CENP-T-Mis12C interaction. By coordination of CENP-T and Dsn1 dual phosphorylation, the Nsl1 acidic surface associates with binding region 1 of CENP-T. A model of the regulatory mechanism of the CENP-T-Mis12C interaction during interphase and M phase A model of the regulatory mechanism of the CENP-T-Mis12C interaction. In interphase, the Dsn1 basic motif masks the Nsl1 acidic surface, inhibiting the interaction between binding region 1 of CENP-T and the Nsl1 acidic surface of Mis12C. This inhibition leads to the inhibition of the interaction between binding region 2 of CENP-T and the Head1 hydrophobic surface of Mis12C. Since CENP-T is not phosphorylated in interphase, S175 and T184 of CENP-T do not interact with the stalk basic surface of Mis12C. In M phase, the Dsn1 basic motif is precluded from the Nsl1 acidic surface by the CENP-T and Dsn1 dual phosphoregulation. Phosphorylation of the Dsn1 basic motif by Aurora B precludes the mask by inhibiting the basic motif binding to the Nsl1 acidic surface. Phosphorylated residues of CENP-T may interact with Mis12C basic surface, helping to preclude the mask by increasing the overall affinity of the CENP-T-Mis12C interaction. By coordination of CENP-T and Dsn1 dual phosphorylation, the Nsl1 acidic surface associates with binding region 1 of CENP-T. The two Mis12C-binding regions of CENP-T are important for the CENP-T-Mis12C interaction. The binding region 2 alone does not seem to be sufficient for binding as evidenced by the ability of the Dsn1 basic motif that competes only with the binding region 1 to completely abolish the interaction with CENP-T in interphase. This mechanism explains complete inhibition of the CENP-T-Mis12C interaction in interphase (Interphase; Figure 7). In addition, Mis12C containing the Dsn1 mutant lacking the basic motif (Dsn1Δ93–114) binds to CENP-T in interphase cells, suggesting that the deletion of the Dsn1 basic motif facilitates the interactions via two interaction surfaces. Therefore, we propose that phosphorylation of the Dsn1 basic motif by Aurora B can facilitate the interactions via two interaction surfaces simultaneously for timely CENP-T-Mis12C interaction during mitosis (M phase; Figure 7). A highlight of this paper is CENP-T and Dsn1 dual phosphoregulation for the CENP-T-Mis12C interaction during mitosis. Although each phosphorylation can recruit sufficient levels of Mis12C at kinetochores for cell proliferation, both mutations for CENP-T and Dsn1 showed additional reduction of Mis12C and caused mitotic defects. How are these dual phosphoregulation related each other? Mis12C levels at kinetochores were reduced in CENP-CΔ73/Dsn14A cells, but these are still higher than those in CENP-CΔ73/CENP-TRR-AA cells that have mutations in binding region 1 of CENP-T, which is important for Mis12C-binding. This result suggests that there is a state of Mis12C in which the Dsn1 basic motif is precluded, even if Aurora B phosphorylation sites of Dsn1 are mutated in CENP-CΔ73/Dsn14A cells. The CENP-T phosphorylation sites may directly bind to Mis12C, as AlphaFold3 predicted that phosphorylated S175 and T184 of CENP-T are close to the basic region of Mis12C. We propose that the CENP-T phosphorylation contributes to the exclusion of the Dsn1 basic motif mask by increasing the overall affinity of the CENP-T-Mis12C interaction (Figures 7 and S6F). Our results suggest that phosphorylation of CENP-T and Dsn1 can coordinately preclude the Dsn1 basic motif mask. Thus, in our model, both CENP-T and Dsn1 phosphorylation may contribute to the CENP-T-Mis12C interaction in early mitosis (Figure S7D). The Dsn1 mask is released by co-ordination of Dsn1 and CENP-T phosphorylation allowing CENP-T-binding region 1 to stably bind to Nsl1 of Mis12C, which leads to the hydrophobic interaction via CENP-T-binding region 2 with Mis12C (Figures 7 and S7D). Interestingly, Mis12C localization to interphase centromeres depends on CENP-C (Figure 1). For the CENP-C-Mis12C interaction during mitosis, the involvement of phosphoregulation of Dsn1 has been well studied.18,26,28,31,32 Our AlphaFold2 prediction suggests that CENP-C is likely to bind to Mis12C in the auto-inhibited state (Figures S1G–S1K) during interphase, independent of Dsn1 phosphoregulation. This finding supports the results of a previous in vitro binding assay showing that CENP-C binds to Mis12C in the auto-inhibited state.26 However, the significance of interphase Mis12C localization at kinetochores via CENP-C remains unclear. A recent study demonstrated that Mis12C constitutively localizes to the centromeres in mouse germ cells expressing the Dsn1 isoform, which lacks the basic motif.34 They proposed that the continuous kinetochore localization of Mis12C might be important for stable KMN network assembly because germ cells immediately enter meiosis II after meiosis I without DNA replication. Similarly, interphase Mis12C localization at kinetochores via CENP-C may be important for timely KMN network assembly for the subsequent mitosis in cells. In both CENP-C- and CENP-T-pathways, phosphorylation by Aurora B facilitates the CENP-C- or CENP-T-Mis12C interaction. Aurora B also phosphorylates Ndc80C on kinetochores on unaligned chromosomes to correct erroneous microtubule attachments and phosphorylation levels by Aurora B is decreased at kinetochores on aligned chromosomes.35 However, even at Aurora B less-kinetochores on aligned chromosomes, Mis12C localizes at kinetochores. Therefore, we propose that the CENP-C- or CENP-T-Mis12C interaction independent of Aurora B-mediated regulations is important to maintain Mis12 C at kinetochores on aligned chromosomes for proper mitotic progression. Recent structural analyses of the KMN network or CCAN using in vitro reconstitution have provided valuable insights into the mechanism underlying kinetochore assembly.9,10,26,31 However, it is still largely unknown how kinetochore components play a role in mitotic progression. Therefore, it is critical to evaluate the significance of the binding surfaces between kinetochore components, which have been clarified by structural analyses. Some kinetochore components have similar functions and simultaneously recruit the same proteins,14,15,17,36 making it difficult to investigate the significance of each interaction in cells. In this study, using chicken DT40 cells lacking the CENP-C-Mis12C interaction, we clearly elucidated the significance of the interaction surfaces and phosphoregulation for the CENP-T-Mis12C interaction in cells. This knowledge provides important insights into how functional kinetochores are established in cells. Limitations of the study We propose the interaction surfaces and phosphoregulation for the CENP-T-Mis12C interaction, based on structure prediction and cell biological and biochemical experiments. However, we did not obtain experimental structure data for the CENP-T-Mis12C interaction. We found that the CENP-T-Mis12C interaction is dynamic in cells, and it is possible that there are additional regulations for the interaction. Further structural studies would uncover unsolved mechanisms for the CENP-T-Mis12C interaction. Resource availability Lead contact Further information and any requests should be directed to and will be fulfilled by the lead contact, Tatsuo Fukagawa (fukagawa.tatsuo.fbs@osaka-u.ac.jp). Materials availability This study did not generate new unique reagents. Plasmids and cell lines generated in this study are available upon request. Any additional analysis information for this work is available by request to the lead contact. Data and code availability •All data reported in this paper will be shared by the lead contact upon request.•This article does not report original code.•Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. •All data reported in this paper will be shared by the lead contact upon request. All data reported in this paper will be shared by the lead contact upon request. •This article does not report original code. This article does not report original code. •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Acknowledgments The authors are very grateful to Y. Kubota and R. Fukuoka for their technical assistance. This work was supported by 10.13039/501100003382CREST of JST (21460153), 10.13039/501100001691JSPS KAKENHI Grant Numbers 20H05389, 21H05752, 22H00408, 22H04692, and 24H02281 to T.F., 10.13039/501100001691JSPS KAKENHI Grant Numbers 22K20630 and 23K14180 to Y.T., 10.13039/501100001691JSPS KAKENHI Grant Numbers 21H02461, 22H04672, and 24K02005, 10.13039/100007449Takeda Science Foundation and 10.13039/501100005927Daiichi Sankyo Foundation of Life Science to M.H., 10.13039/501100001691JSPS KAKENHI Grant Number 24K09340 to M.A.. Author contributions Conceptualization: Y.T. and T.F.; Investigation: Y.T., M.H., R.N., and M.A.; Formal analysis: Y.T. and R.N.; Resources: M.H.; Data curation: Y.T.; Writing – original draft: T.F.; Writing – review and editing: T.F. and Y.T.; Project administration: T.F.; Funding acquisition: T.F., Y.T., M.H., and M.A. Declaration of interests The authors declare no competing interests. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesMouse monoclonal anti-α-tubulinSigmaCat#T9026; RRID: AB_477593Mouse monoclonal anti-FLAGSigmaCat#F1804;RRID: AB_262044Rabbit anti-chicken CENP-CTakenoshita et al.24RRID: AB_3665351Rabbit anti-chicken CENP-THori et al.5RRID: AB_2665551Rabbit anti-chicken CENP-ARegnier et al.37RRID: AB_2665547Rabbit anti-chicken Dsn1Hara et al.18RRID: AB_2756344Rat monoclonal anti-RFPChromotekCat#5F8;RRID: AB_2336064HRP-conjugated anti-Rabbit IgGJackson ImmunoResearchCat#: 111-035-144;RRID: AB_2307391HRP-conjugated anti-mouse IgGJackson ImmunoResearchCat#: 115-035-003;RRID: AB_10015289HRP-conjugated anti-Rat IgGJackson ImmunoResearchCat#112-035-003; RRID: AB_2338128Cy3-conjugated mouse anti-rabbit IgGJackson ImmunoResearchCat#: 211-165-109;RRID: AB_233915Cy3-conjugated goat anti-mouse IgGJackson ImmunoResearchCat#: 115-165-146;RRID: AB_2338690Bacterial and virus strainsE.coli. DH5αTOYOBOCat#: DNA-903Chemicals, peptides, and recombinant proteinsPenicillin-StreptomycinThermo FisherCat#15140-1223-Indole acetic acid (IAA)WakoCat#090-07123In-Fusion Snap Assembly Master MixTakara BioCat#639648puromycinTakaraCat#Z1305NG418Santa Cruz BiotechnologyCat#SC-29065BBlasticidin S hydrochlorideKaken PharmaceuticalCat#KK-400L-Histidinol dihydrochlorideSigmaCat#H6647ZeocinInvitrogenCat#45-0430mycophenolic acidTokyo Chemical IndustryCat#M2216xanthineSigmaCat#X4002ECL Prime Western Blotting Detection ReagentCytivaCat#RPN2232NocodazoleSigmaCat#M1404Signal Enhancer HikariNacalai tesqueCat#02270-81DAPIRocheCat#10236276001VECTASHIELD Mounting MediumVector LaboratoriesCat#H-1000cOmplete™, EDTA-free Protease Inhibitor CocktailRocheCat#5056489001PhosSTOP™RocheCat#4906845001TurboNucleaseAccelagenCat# N0103PMG132SigmaCat#C2211Critical commercial assaysRealTime-Glo MT Cell Viability AssayPromegaCat#G9711Experimental models: Cell linesCL18 (DT40 WT) cellsBuerstedde et al.38N/ACL18 / GFP-Dsn1WT cellsTakenoshita et al.24N/ACL18 / CENP-CΔ73 / GFP-Dsn1WT cellsTakenoshita et al.24N/ACL18 / CENP-CΔ73 / GFP-Dsn1Δ93–114 cellsThis paperN/ACL18 / CENP-CΔ73 / 3xFLAG-CENP-TWT / spot-Dsn1WT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TWT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TΔ161–216 cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TIFV-RRR cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-T RR-AA cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-T2A cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn14A / 3xFLAG-CENP-TWT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn14A / 3xFLAG-CENP-T2A cellsThis paperN/AOligonucleotidesF1: 5’-ATTCCGCACGCTTCAAAAGC-3’Takenoshita et al.24N/AR1: 5’-CTCTCCAGGGTCAGGTTCTGTG-3’Takenoshita et al.24N/AF2: 5’-CTGAATGGATGTTTCTGACCTGCCTTG-3’This studyN/AR2: 5’-CGGCGACGATGAAAATAAAACTAAGGCTG-3’This studyN/AF3: 5’-GAAAGTCCCTATTGGCGTTAC-3’Takenoshita et al.24N/AR3: 5’-CTGTGATGGCACCAGACACTG-3’This studyN/AF4: 5’-GGCGTAGGCCAGAGGTGTCTCC-3’Takenoshita et al.24N/AR4: 5’-AGGCTCTGGTCGAGCATCTTCTG-3’Takenoshita et al.24N/AF5: 5’-GCCGATCATAATCAGCCATACC-3’Takenoshita et al.24N/AR5: 5’-GAACTGTCTGAAGTGCTAGAGG-3’Takenoshita et al.24N/AR6: 5’-TAAACGCCCTCAAGGAGCAAG-3’This studyN/ARecombinant DNApX330Cong et al.39Addgene plasmid #42230pX330_ggDsn1Takenoshita et al.24N/ApX335_ggPGKCao et al.40N/ApX330_ggCENP-TThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1WT_PuroRTakenoshita et al.24N/ApBS_Dsn1-N KI_GFP-Dsn1WT_EcoGPTTakenoshita et al.24N/ApBS_Dsn1-N KI_GFP-Dsn14A_PuroRThis paperN/ApBS_Dsn1-N KI_GFP-Dsn14A_EcoGPTThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1Δ93–114_PuroRThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1Δ93–114_EcoGPTThis paperN/ApAID1.2-CMV-NmScarlet-mAIDNishimura et al.41Addgene plasmid #140618pBS_PGK N-KI_ pCMV-OsTIR1-T2A-BSR-IRES-mScarlet-mAID-CENP-TThis paperN/ApBS_CENP-T N-KI_NeoRTakenoshita et al.24N/ApBS_CENP-T N-KI_HisDThis paperN/ApBS_CENP-T N-KI_3xFLAG-CENP-TWT_NeoRTakenoshita et al.24N/ApBS_CENP-T N-KI_3xFLAG-CENP-TWT_HisDThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-TΔ161–216_NeoRThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-TΔ161–216_HisDThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-T2A_NeoRThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-T2A_HisDThis paperN/ASoftware and algorithmsGraphPad Prism 9.5.1GraphPadhttps://www.graphpad.com/; RRID:SCR_002798FijiSchindelin et al.42https://imagej.net/ij/; RRID:SCR_002285Image Lab 6.0.1Bio-Radhttps://www.bio-rad.com/en-jp/product/image-lab-software?ID=KRE6P5E8Z; RRID:SCR_014210Imaris 9.1.2BitplaneRRID:SCR_007370Xcalibur4.6Thermo FisherRRID:SCR_014593Orbitrap eclipse tune application v.4.0.4091Thermo FisherN/A REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesMouse monoclonal anti-α-tubulinSigmaCat#T9026; RRID: AB_477593Mouse monoclonal anti-FLAGSigmaCat#F1804;RRID: AB_262044Rabbit anti-chicken CENP-CTakenoshita et al.24RRID: AB_3665351Rabbit anti-chicken CENP-THori et al.5RRID: AB_2665551Rabbit anti-chicken CENP-ARegnier et al.37RRID: AB_2665547Rabbit anti-chicken Dsn1Hara et al.18RRID: AB_2756344Rat monoclonal anti-RFPChromotekCat#5F8;RRID: AB_2336064HRP-conjugated anti-Rabbit IgGJackson ImmunoResearchCat#: 111-035-144;RRID: AB_2307391HRP-conjugated anti-mouse IgGJackson ImmunoResearchCat#: 115-035-003;RRID: AB_10015289HRP-conjugated anti-Rat IgGJackson ImmunoResearchCat#112-035-003; RRID: AB_2338128Cy3-conjugated mouse anti-rabbit IgGJackson ImmunoResearchCat#: 211-165-109;RRID: AB_233915Cy3-conjugated goat anti-mouse IgGJackson ImmunoResearchCat#: 115-165-146;RRID: AB_2338690Bacterial and virus strainsE.coli. DH5αTOYOBOCat#: DNA-903Chemicals, peptides, and recombinant proteinsPenicillin-StreptomycinThermo FisherCat#15140-1223-Indole acetic acid (IAA)WakoCat#090-07123In-Fusion Snap Assembly Master MixTakara BioCat#639648puromycinTakaraCat#Z1305NG418Santa Cruz BiotechnologyCat#SC-29065BBlasticidin S hydrochlorideKaken PharmaceuticalCat#KK-400L-Histidinol dihydrochlorideSigmaCat#H6647ZeocinInvitrogenCat#45-0430mycophenolic acidTokyo Chemical IndustryCat#M2216xanthineSigmaCat#X4002ECL Prime Western Blotting Detection ReagentCytivaCat#RPN2232NocodazoleSigmaCat#M1404Signal Enhancer HikariNacalai tesqueCat#02270-81DAPIRocheCat#10236276001VECTASHIELD Mounting MediumVector LaboratoriesCat#H-1000cOmplete™, EDTA-free Protease Inhibitor CocktailRocheCat#5056489001PhosSTOP™RocheCat#4906845001TurboNucleaseAccelagenCat# N0103PMG132SigmaCat#C2211Critical commercial assaysRealTime-Glo MT Cell Viability AssayPromegaCat#G9711Experimental models: Cell linesCL18 (DT40 WT) cellsBuerstedde et al.38N/ACL18 / GFP-Dsn1WT cellsTakenoshita et al.24N/ACL18 / CENP-CΔ73 / GFP-Dsn1WT cellsTakenoshita et al.24N/ACL18 / CENP-CΔ73 / GFP-Dsn1Δ93–114 cellsThis paperN/ACL18 / CENP-CΔ73 / 3xFLAG-CENP-TWT / spot-Dsn1WT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TWT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TΔ161–216 cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-TIFV-RRR cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-T RR-AA cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn1WT / 3xFLAG-CENP-T2A cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn14A / 3xFLAG-CENP-TWT cellsThis paperN/ACL18 / CENP-CΔ73 / AID-CENP-T / GFP-Dsn14A / 3xFLAG-CENP-T2A cellsThis paperN/AOligonucleotidesF1: 5’-ATTCCGCACGCTTCAAAAGC-3’Takenoshita et al.24N/AR1: 5’-CTCTCCAGGGTCAGGTTCTGTG-3’Takenoshita et al.24N/AF2: 5’-CTGAATGGATGTTTCTGACCTGCCTTG-3’This studyN/AR2: 5’-CGGCGACGATGAAAATAAAACTAAGGCTG-3’This studyN/AF3: 5’-GAAAGTCCCTATTGGCGTTAC-3’Takenoshita et al.24N/AR3: 5’-CTGTGATGGCACCAGACACTG-3’This studyN/AF4: 5’-GGCGTAGGCCAGAGGTGTCTCC-3’Takenoshita et al.24N/AR4: 5’-AGGCTCTGGTCGAGCATCTTCTG-3’Takenoshita et al.24N/AF5: 5’-GCCGATCATAATCAGCCATACC-3’Takenoshita et al.24N/AR5: 5’-GAACTGTCTGAAGTGCTAGAGG-3’Takenoshita et al.24N/AR6: 5’-TAAACGCCCTCAAGGAGCAAG-3’This studyN/ARecombinant DNApX330Cong et al.39Addgene plasmid #42230pX330_ggDsn1Takenoshita et al.24N/ApX335_ggPGKCao et al.40N/ApX330_ggCENP-TThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1WT_PuroRTakenoshita et al.24N/ApBS_Dsn1-N KI_GFP-Dsn1WT_EcoGPTTakenoshita et al.24N/ApBS_Dsn1-N KI_GFP-Dsn14A_PuroRThis paperN/ApBS_Dsn1-N KI_GFP-Dsn14A_EcoGPTThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1Δ93–114_PuroRThis paperN/ApBS_Dsn1-N KI_GFP-Dsn1Δ93–114_EcoGPTThis paperN/ApAID1.2-CMV-NmScarlet-mAIDNishimura et al.41Addgene plasmid #140618pBS_PGK N-KI_ pCMV-OsTIR1-T2A-BSR-IRES-mScarlet-mAID-CENP-TThis paperN/ApBS_CENP-T N-KI_NeoRTakenoshita et al.24N/ApBS_CENP-T N-KI_HisDThis paperN/ApBS_CENP-T N-KI_3xFLAG-CENP-TWT_NeoRTakenoshita et al.24N/ApBS_CENP-T N-KI_3xFLAG-CENP-TWT_HisDThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-TΔ161–216_NeoRThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-TΔ161–216_HisDThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-T2A_NeoRThis paperN/ApBS_CENP-T N-KI_ 3xFLAG-CENP-T2A_HisDThis paperN/ASoftware and algorithmsGraphPad Prism 9.5.1GraphPadhttps://www.graphpad.com/; RRID:SCR_002798FijiSchindelin et al.42https://imagej.net/ij/; RRID:SCR_002285Image Lab 6.0.1Bio-Radhttps://www.bio-rad.com/en-jp/product/image-lab-software?ID=KRE6P5E8Z; RRID:SCR_014210Imaris 9.1.2BitplaneRRID:SCR_007370Xcalibur4.6Thermo FisherRRID:SCR_014593Orbitrap eclipse tune application v.4.0.4091Thermo FisherN/A Experimental model and study participant details Cell culture Chicken DT40 cells38 were cultured in Dulbecco’s modified Eagle’s medium (Nacalai Tesque) containing 10% fetal bovine serum (Sigma), 1% chicken serum (Thermo Fisher Scientific), 10 μM 2-mercaptoethanol (Sigma), and penicillin (100 units/mL)-streptomycin (100 μg/mL) (Thermo Fisher Scientific) at 38.5°C with 5% CO2. For degradation of mScarlet-mAID-CENP-T in AID based-CENP-T conditional knockdown cells, 500 μM 3-indole acetic acid (IAA; Wako) was added. All cell lines utilized in this study are listed in the key resources table. DT40 cells were tested for mycoplasma contamination. Method details Plasmid constructions GFP-Dsn1WT, GFP-Dsn14A (S97A, S101A, S102A, and S111A),18 GFP-Dsn1Δ93–114, and Dsn1WT-spot were cloned into the plasmid containing approximately 2 kb homology arm around the start codon of Dsn1 with the puromycin resistance gene cassette and EcoGPT gene cassette by In-Fusion® HD (TAKARA) (pBS_Dsn1-N KI_GFP-Dsn1WT, _GFP-Dsn14A, _GFP-Dsn1Δ93–114, or _Dsn1WT-spot_PuroR or _EcoGPT). cDNA for CENP-T full-length sequence was cloned into pAID1.2-CMV-NmScarlet-mAID (Addgene; #140618). To express the OsTIR1-T2A-BSR and mScarlet-mAID-CENP-T under control of the CMV promoter from the PGK locus, CMV promoter_OsTIR1-T2A-BSR_IRES_mScarlet-mAIDCENP-T fragment was cloned into the plasmid containing approximately 2 kb genome homology arm around the start codon of PGK by In-Fusion® HD (TAKARA) (pBS_PGK N-KI_pCMV-OsTIR1-T2A-BSR-IRES-mScarlet-mAID-CENP-T). cDNA for CENP-T full-length sequence was cloned into p3xFLAG-CMV-10 (SIGMA). Several CENP-T mutants (CENP-TΔ161–216, CENP-TRR-AA, CENP-TIFV-RRR, and CENP-T2A) were generated by PCR based mutagenesis. These 3xFLAG-CENP-T mutants were then cloned into the plasmid containing approximately 2 kb homology arm around the start codon of CENP-T with the neomycin resistance gene cassette and hisD gene cassette by In-Fusion® HD (TAKARA) (pBS_CENP-T N-KI_3xFLAG-CENP-TWT, _3xFLAG-CENP-TΔ161–216, _3xFLAG-CENP-TRR-AA, _3xFLAG-CENP-TIFV-RRR, or _3xFLAG-CENP-T2A _NeoR or _HisD). Generation of cell lines To express the GFP-Dsn1 constructs under control of the endogenous Dsn1 promoter, pBS_Dsn1-N KI_GFP-Dsn1WT, _GFP-Dsn14A, or _GFP-Dsn1Δ93–114_PuroR and pBS_Dsn1-N KI_GFP-Dsn1WT, _GFP-Dsn14A, or _GFP-Dsn1Δ93–114_EcoGPT were transfected with pX330_ggDsn124 using Neon Transfection System (Thermo Fisher) with 6 times of pulse (1400 V, 5 msec) in CL18 (wild-type) cells, CENP-CΔ73cells,24 and CENP-CΔ73 cells expressing mScarlet-mAID-CENP-T. To express the Dsn1-spot constructs under control of the endogenous Dsn1 promoter, pBS_Dsn1-N KI_Dsn1WT-spot_PuroR and pBS_Dsn1-N KI_Dsn1WT-spot_EcoGPT were transfected with pX330_ggDsn1-intron1 (GTAAGCGACGTTGTGGGCCG) using Neon Transfection System (Thermo Fisher) with 6 times of pulse (1400 V, 5 msec) in CENP-CΔ73 cells expressing 3xFLAG-CENP-T. To express OsTIR1-T2A-BSR_IRES_mScarlet-mAID-CENP-T construct under control of the CMV promoter from the endogenous PGK locus, pBS_PGK N-KI_pCMV-OsTIR1-T2A-BSR-IRES-mScarlet-mAID-CENP-T was transfected with pX335_ggPGK40 using Neon Transfection System (Thermo Fisher) with 6 times of pulse (1400 V, 5 msec) in CENP-CΔ73 cells.24 To express 3xFLAG-CENP-T constructs under control of the endogenous CENP-T promoter, pBS_CENP-T N-KI_3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, 3xFLAG-CENP-TRR-AA, 3xFLAG-CENP-TIFV-RRR, or 3xFLAG-CENP-T2A_NeoR and pBS_CENP-T N-KI_3xFLAG-CENP-TWT, 3xFLAG-CENP-TΔ161–216, 3xFLAG-CENP-TRR-AA, 3xFLAG-CENP-TIFV-RRR, or 3xFLAG-CENP-T2A_HisD were transfected with pX330_ggCENP-T (TCCGAGAGGTCCGTCATGGA) using Neon Transfection System (Thermo Fisher) with 6 times of pulse (1400 V, 5 msec) in CENP-CΔ73/GFP-Dsn1WT cells expressing mScarlet-mAID-CENP-T, CENP-CΔ73/GFP-Dsn14A cells expressing mScarlet-mAID-CENP-T. The transfected cells were selected in the medium containing appropriate selection drugs (0.5 μg/mL puromycin (TAKARA) for the selection of PuroR expression, 25 μg/mL mycophenolic acid (Wako) and 125 μg/mL xanthine (SIGMA) for the selection of EcoGPT expression, 25 μg/mL Blasticidin S hydrochloride (Wako) for the selection of BSR expression, and L-histidinol for the selection of HisD expression) in 96 well plates to isolate single colonies. Structure prediction Structures were predicted with AlphaFold2-Multimer through ColabFold using MMseqs2 and default settings.43 The structural figures were prepared using UCSF ChimeraX. Immunoblot DT40 cells were collected, washed with cold PBS, and suspended in 1xLaemmli Sample Buffer (LSB; 62.5 mM Tris-HCl (pH 6.8), 10 % Glycerol, 2% SDS, 5% 2-mercaptoethanol, bromophenol blue) (final concentration 1x10ˆ4 cells/μL). Following the sonication, the lysate was heated at 96°C for 5 min. The collected samples (1x10ˆ5 cells) were separated by SDS-PAGE (SuperSep Ace, 5–20%; Wako) and transferred onto PVDF membranes (Immobilon®-P; Merck). The membrane was probed with primary antibody diluted with Signal Enhancer Hikari (Nacalai Tesque) at 4°C for overnight. After washing with 0.1% TBST (TBS, 0.1% Tween 20) for 15 min, the membrane was probed with secondary antibody diluted with Signal Enhancer Hikari (Nacalai Tesque) at room temperature for 1 h. After washing with 0.1% TBST for 15 min, the signals were detected using ECL Prime (GE Healthcare) and visualized by ChemiDoc Touch imaging system (Bio-Rad). The Image processing was performed using Image Lab 5.2.1 (Bio-Rad) and Photoshop CC (Adobe). Primary antibodies used in immunoblot analyses were rabbit polyclonal anti-chicken CENP-T,5 rabbit polyclonal anti-chicken CENP-C,30 rabbit polyclonal anti-chicken Dsn1,18 mouse monoclonal anti-FLAG M2 (Sigma), rat polyclonal anti-RFP (Chromotek), and mouse monoclonal anti-α-tubulin (Sigma). Secondary antibodies used in immunoblot analysis were horseradish peroxidase-conjugated (HRP)-conjugated goat polyclonal anti-rabbit IgG, HRP-conjugated goat polyclonal anti-Rat IgG, and HRP-conjugated rabbit polyclonal anti-mouse IgG (Jackson ImmunoResearch). Immunoprecipitation For 3xFLAG-CENP-T immunoprecipitation, cells expressing mScarlet-mAID-CENP-T and 3xFLAG-CENP-T were cultured with 500 μM IAA for 12 h and with 100 ng/mL nocodazole for the last 10h. These cells were collected, washed with cold PBS twice, suspended in Lysis buffer (20 mM Hepes-NaOH (pH 7.4), 150 mM NaCl, 0.1% NP40, 5 mM 2-mercaptoethanol, 1xcomplete EDTA-free proteinase inhibitor (Roche), and 1xPhosSTOP (SIGMA)) (final: 2x108 cells/mL) and sonicated. The lysate was treated with TURBO nuclease (50 unit/mL) for 30 min, clarified by centrifugation, and the supernatant was incubated with Protein-G Dynabeads (Thermo Fisher Scientific) conjugated to anti FLAG-M2 antibody at 4°C for 1 h. Proteins precipitated with anti-FLAG-M2 monoclonal antibody bound beads were washed with Lysis buffer three times, eluted with 2×LSB. The eluted samples were heated at 96°C for 5 min and subjected to immunoblot analysis. Immunofluorescence analysis The cells were cytospan onto slide glasses or coverslips by the Cytospin3 centrifuge (Shandon). For the CENP-T staining, cells were fixed with 3% paraformaldehyde (PFA) in 250 mM HEPES-NaOH (pH 7.4) at RT for 10 min and permeabilized in 0.5% NP-40 in PBS at RT for 10 min. For the CENP-A staining, cells were fixed with 3% paraformaldehyde (PFA) in 250 mM HEPES-NaOH (pH 7.4) at RT for 30 sec and fixed with methanol at -30°C for 20 min. After blocking with 0.5% BSA in PBS for 5 min, the cells were incubated with primary antibodies diluted in 0.5% BSA in PBS at 37°C for 1 h. The cells were washed three times with 0.5% BSA in PBS, incubated with secondary antibodies diluted in 0.5% BSA in PBS at 37°C for 1 h, and washed three times with 0.5% BSA in PBS. The cells were post-fixed for 10 min. DNA was stained with 100 ng/mL DAPI in PBS for 20 min. The stained cells were mounted with VECTASHIELD Mounting Medium (Vector Laboratories). Primary antibodies used in immunofluorescence analysis were rabbit polyclonal anti-chicken CENP-T,5 rabbit polyclonal anti-chicken CENP-A,37 and mouse monoclonal anti-FLAG-M2 (Sigma). Secondary antibodies used in immunofluorescence analysis were Cy3-conjugated mouse polyclonal anti-rabbit IgG and Cy3-conjugated goat polyclonal anti-mouse IgG (Jackson ImmunoResearch). Image acquisition Immunofluorescence images were acquired every 0.2 μm intervals of z-slice using a Zyla 4.2 sCMOS camera (ANDOR) mounted on a Nikon Eclipse Ti inverted microscope with an objective lens (Nikon; Plan Apo lambda 100x/1.45 NA) with a spinning disk confocal scanner unit (CSU-W1; Yokogawa) controlled with NIS-elements (Nikon). The images in the figures are the maximum intensity projection of the Z-stack generated with Fiji.42 Chromosome alignment assay The cells were treated with IAA (500 μM) for 12 h and incubated with MG132 (10 μM) for the last 2 h and harvested. The harvested cells were fixed and immunostained with anti-CENP-A antibody as above and mounted with VECTASHIELD Mounting Medium. The cells images were acquired by the confocal microscopy with spinning-disc. The chromosome alignment was analyzed according to a previous report.44 Briefly, the XY positions of CENP-A signals in mitotic cells were acquired using the Imaris software and plotted on a two-dimensional plane. Scatter plot analysis was performed with confidence ellipses using Real Statistics in Excel (www.real-statistics.com). Chromosome alignment was assessed by calculating the ratio of the semi-major axis to the semi-minor axis of the ellipse and subtracting 1 to obtain the alignment value. As DT40 cells are maintained in a suspension culture, the data includes values from various angles, including non-side view chromosomes. However, since the experimental conditions are the same for all samples, comparing the control and experimental data appropriately provides us reliable data to evaluate the chromosome alignment. The experiments were repeated twice. Representative data of technical replicates are presented. Flow cytometry analysis The cell cycle distribution was analyzed according to a previous report.45 The cells were treated with IAA (500 μM) for 12 h and incubated with 20 μM BrdU (5-bromo-2-deoxyuridine) for 20 min and harvested. The harvested cells were washed with the ice-cold PBS, fixed with 70% ethanol at –30°C. The fixed cells were washed with 1% BSA/PBS, incubated in 4 N HCl with 0.5% Triton X-100 at RT for 30 min, and washed with 1% BSA/PBS three times. The cells were incubated with an anti-BrdU antibody (BD, 347580) at RT for 1 h and washed with 1% BSA/PBS two times. Then, the cells were incubated with FITC-conjugated anti-mouse IgG (Jackson ImmunoResearch, 115-095-003; at 1:20 dilution in 1% BSA/PBS) at RT for 30 min, washed with 1% BSA/PBS and stained DNA with propidium iodide (10 μg/mL) in 1% BSA/PBS at 4°C overnight. The stained cells were applied to a Flow cytometry (BD; FACS Canto II). The cell cycle gates were manually adjusted based on plots of CENP-CΔ73/CENP-TWT cells, clearly distinguishing between subG1, G1, S, and G2/M populations, which were then applied to other cell lines. The percentages of each cell cycle stage (subG1, G1, S, G2/M phase or others cell) were calculated. The experiments were repeated twice. Representative data of technical replicates are presented. LC-MS/MS analysis For 3xFLAG-CENP-T immunoprecipitation, cells expressing 3xFLAG fused CENP-T, spot-fused Dsn1, and CENP-CΔ73 were cultured with 100 ng/mL nocodazole for 12 h. These cells were collected, washed with cold PBS twice, and suspended in 300 μL Lysis buffer (20 mM Hepes-NaOH (pH 7.4), 150 mM NaCl, 0.1% NP40, 5 mM 2-mercaptoethanol, 1xcomplete EDTA-free proteinase inhibitor (Roche), and 1xPhosSTOP (SIGMA) (final: 2x108 cells/mL)). The suspension was then sonicated. The lysate was treated with TURBO nuclease (50 units/mL) for 30 min, clarified by centrifugation, and the supernatant was incubated with Protein-G Dynabeads (Thermo Fisher Scientific) conjugated to anti FLAG-M2 antibody at 4°C for 2 h. Proteins precipitated with anti FLAG-M2 antibody-bound beads were washed with 30 μL Lysis buffer five times and eluted with 25 μL Elution buffer (150 μg/mL 3xFLAG-peptide in Lysis buffer) four times. For LC-MS/MS analysis, eluted samples underwent reduction with 10 mM Tris(2-carboxyethyl)phosphine Hydrochloride (TCEP) and alkylation via incubation with 55 mM iodoacetamide (IAA) for 30 min at room temperature in the dark. Trypsin digestion and sample cleanup were performed using the SP3 method.46 Mass spectra were acquired with an Orbitrap Eclipse (Thermo Fisher Scientific) coupled to a nanoflow UHPLC system (Vanquish; Thermo Fisher Scientific). The peptide mixtures were loaded onto a C18 trap column (PepMap Neo Trap Cartridge, ID 0.3 mm x 5 mm, particle size 5 μm, Thermo Fisher Scientific) and fractionated through the C18 analytical column (Aurora, ID 0.075 × 250 mm, particle size 1.7 μm, IonOpticks). The peptides were eluted at a flow rate of 300 nL/min using the following gradient: 0% to 2% solvent B over 1 minute, 2% to 5% over 2 minutes, 5% to 16% over 19.5 minutes, 16% to 25% over 10 minutes, 25% to 35% over 4.5 minutes, a sharp increase to 95% over 4 minutes, hold at 95% for 5 minutes, and finally re-equilibration at 5%. Solvent A and B compositions were 100% H2O, 0.1% Formic acid and 100% Acetonitrile, 0.1% Formic acid, respectively. The Orbitrap operated in a data-dependent mode with a 3-second cycle time. The MS1 scan was collected at 60,000 resolution, the mass range 375 -1500 m/z, using a standard AGC and maximum injection time of 50 ms. MS/MS was triggered from precursors with intensity above 20,000 and charge states 2-7. Quadrupole isolation width was 1.6 m/z, with normalized HCD energy of 30%, and resulting fragment ions recorded in Orbitrap. MS1 scan was collected at 15,000 resolution with standard AGC target and maximum injection time of 22 ms. Dynamic exclusion was set to 20 seconds. The raw data files were searched against the Gallus gallus dataset (Uniprot Proteome UP000000539, downloaded on 20230703) with the common Repository of Adventitious Proteins (cRAP, ftp://ftp.thegpm.org/fasta/cRAP) and 3xFlag sequence for contaminant protein identification, using Proteome Discoverer 2.5 software (Thermo Fisher Scientific) with MASCOT ver.2.8 search engine, with a false discovery rate (FDR) set at 0.01. The number of missed cleavages sites was set as 2. Carbamidomethylation of cysteine was set as a fixed modification. Oxidation of methionine, phosphorylation of serine, threonine and tyrosine, and acetylation of protein N-termini were set as variable modifications. Nocodazole sensitivity assay The cells were treated with IAA (500 μM) for 12 h. The IAA treated-cells (1x10ˆ5 cells/mL) were then treated with nocodazole at indicated concentrations (5, 10, 15, or 20 ng/mL) for 24 h. After nocodazole treatment, 100 μL of cells were seeded onto opaque-walled tissue culture 96 well plates with a clear bottom. 20 μL of RealTime-Glo reagents premix (1/10000 substrate and 1/10000 enzyme in DMEM; RealTime-Glo MT Cell Viability Assay; Promega) were added to the cells and incubated for 1 h at 37°C. Luminescence signal was measured using a GloMax Discover System Microplate Reader (Promega). IC50 was calculated using GraphPad Prism by fitting the data to a sigmoidal dose-response curve with a 4 parameter logistic (4PL) model, where the X-axis represents the logarithm of the concentration. The experiments were repeated three times, and the mean and standard deviation (SD) are shown. Quantification and statistical analysis The fluorescence signal intensities of GFP-Dsn1 at more than 80 kinetochores in each of 10 cells and background signals in non-kinetochore region for each sample were quantified using the Imaris software (Bitplane). The mean fluorescence signal intensities at kinetochores in each cell were subtracted by the mean background signal intensity in the non-kinetochore region. Each experiment was repeated two (Figure 1E) or three times (Figures 3C, 3G, 4E, 5D, and 6B). Representative data of technical replicates are presented. Statistical analyses were performed using GraphPad Prism (GraphPad Software).