CENP-T provides a structural platform for outer kinetochore assembly

Abstract

The kinetochore forms a dynamic interface with microtubules from the mitotic spindle during mitosis. The Ndc80 complex acts as the key microtubule-binding complex at kinetochores. However, it is unclear how the Ndc80 complex associates with the inner kinetochore proteins that assemble upon centromeric chromatin. Here, based on a high-resolution structural analysis, we demonstrate that the N-terminal region of vertebrate CENP-T interacts with the 'RWD' domain in the Spc24/25 portion of the Ndc80 complex. Phosphorylation of CENP-T strengthens a cryptic hydrophobic interaction between CENP-T and Spc25 resulting in a phospho-regulated interaction that occurs without direct recognition of the phosphorylated residue. The Ndc80 complex interacts with both CENP-T and the Mis12 complex, but we find that these interactions are mutually exclusive, supporting a model in which two distinct pathways target the Ndc80 complex to kinetochores. Our results provide a model for how the multiple protein complexes at kinetochores associate in a phospho-regulated manner.

Figure 1

The CENP-T N-terminal region is required for kinetochore localization of the Ndc80 complex. (A) Diagram showing the chicken CENP-T sequence and the tested deletion mutants in the CENP-T N-terminal region. ‘+’ or ‘−’ indicates whether the given deletion mutant can complement growth when endogenous CENP-T is depleted. (B) Graph showing the growth curve of cells expressing CENP-TΔ69–90 in the presence (red rectangle) or absence (blue diamond) of tetracycline to repress the expression of wild-type CENP-T. (C) Immunofluorescence analysis of cells expressing CENP-T Δ69–90 after 72 h in the presence (lower panel) or absence (upper panel) of tetracycline to repress the expression of wild-type CENP-T. Cells were probed for either CENP-T or Ndc80 and the kinetochore signal intensities of each protein were measured relative to an adjacent background signal. Bar, 10 μm. (D) Immunofluorescence analysis of cells in which expression of CENP-T is replaced with a Spc25-Δ1–90-CENP-T fusion protein. Ndc80 localizes to kinetochores in these cells, unlike the CENP-T Δ69–90 mutant alone in (C). (E) Cell viability analysis for CENP-T conditional knockout cells in the presence (CENP-T OFF) or absence (CENP-T ON) of tetracycline, or expressing a Spc25-Δ1–90 CENP-T fusion protein in the presence of tetracycline (CENP-T OFF+Spc25-Δ1–90 CENP-T).

Figure 2

Phospho-mimetic CENP-T binds directly to the Spc24/25 portion of the Ndc80 complex. (A) Top, traces (OD₂₁₄) from the gel filtration column showing the co-migration of chicken Spc24^125–195/Spc25^132–234 with phospho-mimetic chicken CENP-T (63–98; T72D and S88D). The phospho-mimetic CENP-T fused with MBP and the globular domains of the Spc24/25 complex were tested individually or mixed and incubated for 15 min at room temperature prior to separation by gel filtration using a Superdex 75 column. Bottom, peak fractions were analysed by SDS–PAGE and stained with Coomassie. (B) Non-phosphomimetic CENP-T^63–98 does not strongly interact with Spc24/25. Wild-type MBP-CENP-T^63–98 and the Spc24^125–195/Spc25^132–234 complex were analysed as in (A). (C) The phospho-mimetic chicken CENP-T N-terminal region binds to Spc24/25 with high affinity. Composition gradient multi-angle light scattering (CG-MALS) analysis of phospho-mimetic MBP-CENP-T^63–98 together with the globular domain of the Spc24^125–195/Spc25^132–234 complex. Nine different composition gradients were analysed by multi-angle light scattering to measure the molar mass. Composition of various forms of proteins was calculated by fitting the CG-MALS data and the concentration distribution graph is shown. The two components interacted with 1:1 stoichiometry with a K_D of 381 nM. (D) ITC binding curve for the interaction of phospho-mimetic chicken CENP-T peptide^63–98 (T72D and S88D) with the Spc24^125–195/Spc25^132–234 complex. The measured K_D is 481 nM. (E) ITC binding curve for the interaction of singly phospho-mimetic chicken CENP-T peptide^63–98 (T72D) with the Spc24^125–195/Spc25^132–234 complex. The measured K_D is 610 nM. (F) ITC binding curve for the interaction of wild-type chicken CENP-T peptide^63–98 with the Spc24^125–195/Spc25^132–234 complex. The measured K_D is 6.67 μM. (G) ITC binding curve for the interaction of a synthetic phosphorylated chicken CENP-T peptide^63–98 (T72p and S88p) with the Spc24^125–195/Spc25^132–234 complex. The measured K_D is 518 nM. (H) ITC binding curve for the interaction of a synthetic phosphorylated chicken CENP-T peptide^63–98 (T72p) with the Spc24^125–195/Spc25^132–234 complex. The measured K_D is 1.29 μM.

Figure 3

Crystal structure of the phospho-mimetic CENP-T–Spc24/25 complex. (A) Schematic diagram of Spc24, Spc25, CENP-T, and CENP-W showing the presence of α-helices and β-sheets. The expression constructs used for the structural analyses are indicated by the box. (B) Structural model showing the side view of the Spc24/25 complex superimposed with the CENP-T–Spc24/25 complex. For the Spc24/25 complex, Spc24 is coloured in light cyan and Spc25 is in light green. For the CENP-T–Spc24/25 complex, CENP-T is coloured in magenta, Spc24 is cyan, and Spc25 is green. (C) Structural model showing the top view of the superimposed structures of the Spc24/25 complex and the CENP-T–Spc24/25 complex as in (B). (D) Structural model showing the surface charge of the Spc24/25 complex interacting with phospho-mimetic CENP-T peptide. Electrostatic surface charges of the Spc24/25 complex were calculated by APBS and are contoured from −8.0 (red) to 8.0 (blue). The complex is viewed from the same angle as in (C). Side chains of CENP-T are shown as stick models. (E) Structural model showing the phospho-mimetic CENP-T peptide from the CENP-T–Spc24/25 complex structure in (D) on its own.

Figure 4

A conserved salt bridge is essential for the interaction of CENP-T with the Spc24/25 complex. (A) Sequence alignment of CENP-T from chicken, human, mouse, frog, sea bass, fission yeast, and filamentous fungi. Residues involved in the interaction of CENP-T with Spc24 and Spc25 are denoted by blue and orange dots, respectively. (B) Structural model showing a close-up view of the phospho-mimetic CENP-T peptide. The salt bridge between T72D and R74 is highlighted. (C) Mutation of CENP-T R74 (R74E or R74A) disrupts complex formation even in the presence of phospho-mimetic T72D and S88D residues. Stoichiometric amounts of MBP-CENP-T^63–98 mutant (T72D/R74E/S88D or T72D/R74A/S88D) and the Spc24^125-195/Spc25^132-234 complex were mixed and analysed by gel filtration as in Figure 2A. Upper panel: phospho-mimetic CENP-T (T72D/S88D) and the Spc24/25 complex interaction as in Figure 2A. Middle panel: CENP-T mutant (T72D/R74E/S88D) with the Spc24^125–195/Spc25^132–234 complex. Lower panel: CENP-T mutant (T72D/R74A/S88D) with the Spc24^125–195/Spc25^132–234 complex. Schematic diagrams of CENP-T and the Spc24/25 complex are shown to the right (blue: Spc24, green: Spc25). (D) CENP-T mutation at L68 disrupts complex formation with Spc24/25. Gel filtration analysis of the MBP-CENP-T^63–98 mutant (L68R7/T72D/S88D) with the Spc24^125–195/Spc25^132–234 complex as in (C).

Figure 5

Binding of CENP-T and the Mis12 complex to the Ndc80 complex is mutually exclusive. (A) L161R mutations in Spc25 disrupt the interaction with CENP-T. Phospho-mimetic chicken MBP-CENP-T^63–98 (T72D/S88D) and the mutant Spc24^125–195/Spc25^132–234 (L161R) complex were separated by gel filtration using a Superdex 75, analysed by SDS–PAGE, and stained with Coomassie. (B) I156R mutations in Spc25 disrupt the interaction with CENP-T. Phospho-mimetic MBP-CENP-T^63–98 (T72D/S88D) and the mutant Spc24^125–195/Spc25^132–234 (I156R) complex were analysed by gel filtration as in (A). (C) I149A L154A double mutants in human Spc25 disrupt binding to human CENP-T. Human phospho-mimetic CENP-T (MBP-hsCENP-T^76–106 T85D) and the wild-type Ndc80^Bonsai complex or the Ndc80^Bonsai I149A L154A mutant complex were mixed and analysed by gel filtration. Protein mixtures were incubated on ice for 30 min before conducting the chromatography using a Superose 6 column. Fractions were collected, analysed by SDS–PAGE, and stained with Coomassie. Elution profiles from the size-exclusion chromatography for the experiment are shown (top). Elution of proteins was monitored at A_{280 nm}. (D) The Ndc80^Bonsai complex or the Ndc80^Bonsai I149A L154A mutant complex and the human Mis12–KNL1^2106–2316 complex were mixed and analysed by gel filtration chromatography using a Superose 6 column as in (C). Elution profiles from the size-exclusion chromatography are shown (top). (E) CENP-T and the Mis12/KNL1^CT complex show mutually exclusive binding to the Ndc80^Bonsai complex. Human phospho-mimetic MBP-CENP-T^76–106(T85D), wild-type Ndc80^Bonsai complex, and the human Mis12-KNL1^2106–2316 complex were mixed in the indicated combinations and analysed by gel filtration. A large complex containing all components was not detected, although both CENP-T and KNL1/Mis12 bound individually to the Ndc80 complex based on altered migration.

Figure 6

The Ndc80 complex is targeted to kinetochores by two parallel pathways. (A) Kinetochore localization of Spc25 I156R mutants was reduced to ∼60% in DT40 cells. Immunofluorescence images showing the co-localization of chicken wild type or I156R mutant Spc25-GFP with CENP-T. Signal intensities of each protein were measured relative to an adjacent background signal. Bar, 10 μm. (B) Graph showing growth curves of DT40 cells in which expression of Spc25 is replaced with Spc25 I156R mutant. The doubling time of these cells was 14.2 h compared to 13.1 h for control cells. Tetracycline was added at time 0 to repress transcription of wild-type Spc25. (C) Spc25 mutants defective for CENP-T interactions require the Mis12 complex to localize to kinetochores. Images showing localization of Spc25 (I156R) in Dsn1- or CENP-T-degron cells. Dsn1 or CENP-T was degraded within 1 h after the addition of auxin (see Supplementary Figure S6).

Figure 7

Parallel pathways for outer kinetochore assembly. Model for the molecular architecture of the kinetochore based on the structural, biochemical, and cell biological work presented in this paper. The CENP-T N-terminal region is unphosphorylated during interphase and phosphorylated by CDK during mitosis. The phosphorylated residue (T72 in chicken CENP-T) forms a salt bridge with R residues to allow adjacent downstream hydrophobic residues to interact with Spc24/25. The CENP-T pathway serves an important role to recruit the Ndc80 complex to kinetochores. In addition, a second parallel pathway for Ndc80 complex localization is mediated by the Mis12 complex.