The MIS12 complex is a protein interaction hub for outer kinetochore assembly - PubMed (original) (raw)
. 2010 Sep 6;190(5):835-52.
doi: 10.1083/jcb.201002070.
Sebastiano Pasqualato, Prakash Dube, Veronica Krenn, Stefano Santaguida, Davide Cittaro, Silvia Monzani, Lucia Massimiliano, Jenny Keller, Aldo Tarricone, Alessio Maiolica, Holger Stark, Andrea Musacchio
Affiliations
- PMID: 20819937
- PMCID: PMC2935574
- DOI: 10.1083/jcb.201002070
The MIS12 complex is a protein interaction hub for outer kinetochore assembly
Arsen Petrovic et al. J Cell Biol. 2010.
Abstract
Kinetochores are nucleoprotein assemblies responsible for the attachment of chromosomes to spindle microtubules during mitosis. The KMN network, a crucial constituent of the outer kinetochore, creates an interface that connects microtubules to centromeric chromatin. The NDC80, MIS12, and KNL1 complexes form the core of the KMN network. We recently reported the structural organization of the human NDC80 complex. In this study, we extend our analysis to the human MIS12 complex and show that it has an elongated structure with a long axis of approximately 22 nm. Through biochemical analysis, cross-linking-based methods, and negative-stain electron microscopy, we investigated the reciprocal organization of the subunits of the MIS12 complex and their contacts with the rest of the KMN network. A highlight of our findings is the identification of the NSL1 subunit as a scaffold supporting interactions of the MIS12 complex with the NDC80 and KNL1 complexes. Our analysis has important implications for understanding kinetochore organization in different organisms.
Figures
Figure 1.
Reconstitution and structural analysis of human MIS12C. (A) Schematic representation of the components of the MIS12C, NDC80C, and KNL1C. Coiled-coil (CC) predictions calculated with program COILS (Lupas et al., 1991) are shown exclusively for subunits with partial or complete coiled-coil content. Alternate names in humans are indicated. Hs, Homo sapiens. (B) Size-exclusion chromatography run of recombinant MIS12CNSL1-258 with corresponding SDS-PAGE separation stained with Coomassie brilliant blue. The molecular mass of the recombinant complex is ∼120 kD, but the protein elutes earlier than expected for a globular protein of equivalent molecular mass, suggesting that it is an oligomer or that it is elongated. The dashed gray line and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively. a.u., arbitrary unit. (C) Negative-stain EM was performed on the recombinant MIS12CNSL1-258. The maximum length of the complex varies between ∼21 and 23 nm depending on the curvature. The maximum thickness of the rodlike structures is ∼3 nm. (D) The class averages represent the characteristic views of the MIS12C and reveal a varying amount of curvature. Bars: (C) 10 nm; (D) 5 nm.
Figure 2.
The C-terminal domain of KNL1 interacts directly with MIS12C. (A) Size-exclusion chromatography analysis demonstrates that MIS12C and KNL12106–2316 form a stoichiometric complex. (B) Cross-linking analysis. In lanes 1–3, 5 nmol, 11 nmol, or 22 nmol BS2G was added, respectively. (C) Summary of identified intersubunit cross-links. The full list of cross-links is listed in
Table S2
. (D) Multiple sequence alignment of the C-terminal regions of NSL1 from different species. The two red dots indicate residues that were mutated into glutamic acid (E) in the NSL1EE mutant. (E) Size-exclusion chromatography analysis on the interaction of KNL12106–2316 with a construct encompassing residues 227–281 of the NSL1 subunit of the MIS12C. Elution profile of KNL12106–2316 (top), NSL1227–281 (middle), and of a stoichiometric combination of KNL12106–2316 and NSL1227–281 (bottom). (A and E) Dashed gray lines and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively. (F) ITC on the KNL12106–2316–NSL1227–281 interaction reveals a _K_d of ∼130 nM and a stoichiometry of 1:1.
Figure 3.
Interaction of MIS12C with NDC80C. (A) Schematic representation of the NDC80C. (B) Binding curves were measured by ITC by titrating NDC80C (left) or SPC2457–197–SPC2570–224 (right) in a cell containing MIS12CNSL1-258. (C) Size-exclusion chromatography elution profiles and SDS-PAGE analysis of recombinant NDC80C expressed in, and purified from, insect cells (top), MIS12CNSL1-258 (middle), and their stoichiometric combination (bottom). (D) Elution profiles and SDS-PAGE analysis of NDC80C (top), MIS12CNSL1-227 (middle), and their stoichiometric combination (bottom). The MIS12CNSL1-227 construct binds NDC80C. (E) As in C, but with the kinetochore-binding portion of NDC80C, the SPC2457–197–SPC2570–224 construct. (C–E) Dashed gray lines and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively.
Figure 4.
Interaction of MIS12C with HP1. (A) Size-exclusion chromatography elution profiles and SDS-PAGE analysis of MIS12CNSL1-258 (top), HP1-α (middle), and their stoichiometric combination (bottom). (B) The MIS12CNSL1-258EE mutant does not bind HP1-α (top). SPC2457–197–SPC2570–224 is also unable to bind MIS12CNSL1-258EE, indicating that the binding sites for HP1-α and SPC2457–197–SPC2570–224 overlap, at least in part (middle). When HP1-α and SPC2457–197–SPC2570–224 were combined in stoichiometric amounts with MIS12CNSL1-258, HP1-α did not coelute with MIS12CNSL1-258, whereas SPC2457–197–SPC2570–224 was incorporated in a complex with MIS12CNSL1-258, suggesting that SPC2457–197–SPC2570–224 binds MIS12CNSL1-258 with higher affinity (bottom). (A and B) Dashed gray lines and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively. (C) ITC binding curve for the interaction of MIS12CNSL1-258 with HP1-α. (D) Elution profile from a size-exclusion chromatography Superdex 75 PC 3.2/30 column of a fluorescein-labeled synthetic peptide encompassing residues 205–217 of NSL1 (fluorescein-NSL1205–217). The red trace reports absorbance (Abs) at 525 nm. Panels D through H were run under the same experimental conditions. (E) Elution profile of the chromoshadow domain of HP1-α. (F) Elution profile of a mixture of the chromoshadow domain of HP1-α and fluorescein-NSL1205–217 demonstrating a shift in the elution profile of the fluorescein-NSL1205–217 peptide. (G) Elution profile of SPC2457–197–SPC2570–224. (H) The SPC2457–197–SPC2570–224 construct does not coelute with the fluorescein-NSL1205–217 peptide, suggesting that this region of NSL1 is insufficient for high-affinity binding to NDC80C. (I) Summary of interactions presented in the figure. The position of the second binding site for SPC24–SPC24, indicated by a black curved segment, is actually on MIS12C, but not necessarily on the NSL1 subunit as shown. CD, chromodomain; CSD, chromoshadow domain.
Figure 5.
Role of the NSL1 C-terminal tail. (A) The DSN1–NSL1 subcomplex is sufficient to bind KNL1 and NDC80C. GST-NSL1–DSN1 complex was purified by glutathione Sepharose affinity purification and used as an affinity bait for pull-down assays. Copurification of DSN1 with GST-tagged NSL1 was assessed by immunoblotting. GST-NSL1–DSN1 specifically binds KNL12106–2316 (lane 4) and NDC80C (as judged by immunoblotting on the NDC80 subunit; lane 5), and the binding is not mutually exclusive (lane 6). The levels of unspecific binding to GST are shown in lanes 1 and 2. (B) HeLa cells were transiently transfected with plasmids containing GFP alone, GFP-tagged NSL1, and GFP-tagged NSL1EE. After 48 h, cells were washed once and treated with 5 µM S-trityl-
l
-cysteine for 16 h. Mitotic cells were harvested by vigorous shake-off. Immunoprecipitates (IPs) with an anti-GFP antibody were examined by immunoblotting (IB) with the indicated antibodies. Only GFP-NSL1 was able to bind endogenous NDC80. The band indicated by the asterisk may represent a degradation product of the GFP fusion proteins. wt, wild type. (C) HeLa cells were transiently transfected with plasmids containing GFP alone, GFP-tagged NSL1, GFP-tagged NSL1EE, and GFP-tagged NSL1258. After 48 h, cells were washed once, treated with 3.3 µM nocodazole for 14 h, and analyzed by immunofluorescence. Bar, 5 µm. (D) The means of KNL1 or NDC80 kinetochore (KT) intensities normalized to CREST signal in cells transiently expressing GFP alone, GFP-tagged NSL1, GFP-tagged NSL1EE, and GFP-tagged NSL1258 from the experiment in C were calculated and plotted. Error bars represent the SEM. For each condition, at least four different cells were used in the quantification.
Figure 6.
Interaction of MIS12C with HP1. (A) Elution profile and SDS-PAGE analysis of a stoichiometric mixture of MIS12C–KNL12106–2316 and NDC80C. The elution position of individual complexes in indicated. (B) ITC analysis of the interaction of MIS12C–KNL12106–2316 with NDC80C. (C) SPC24–SPC25 binds the C-terminal tail of NSL1 and an additional site on MIS12 (see legend to Fig. 4 I). If the PVIHL motif is mutated, SPC24–SPC25 does not bind. If SPC24–SPC24 bound KNL1 (the red arrow indicates this interaction is hypothetical), additional binding energy may be recovered, and SPC24–SPC25 would be expected to bind even with a mutated PVIHL motif. (D) Addition of KNL12106–2316 to MIS12C does not rescue the lack of interaction of NDC80C to the PVIHL mutant, suggesting that SPC24–SPC25 and KNL12106–2316 do not interact. (A and D) Dashed gray lines and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively.
Figure 7.
Interactions of ZWINT. (A) Elution profile and SDS-PAGE analysis of NDC80 and recombinant ZWINT. No complex with NDC80C was observed. (B) As in A, but with MIS12C. (A and B) Dashed gray lines and numbers indicate elution markers in the size-exclusion chromatography experiments and their molecular masses (in kilodaltons), respectively. (C, left) Only background binding to GST bound to glutathione Sepharose beads was observed. (right) A GST-KNL11904–2316 fusion protein was used as an affinity bait on glutathione Sepharose beads. This construct was then incubated with the indicated proteins. Asterisks mark two bands that copurify with GST-KNL11904–2316 on the glutathione Sepharose beads. Additional controls and input proteins are shown in
Fig. S3
.
Figure 8.
EM analysis of MIS12C–SPC24–SPC25. (A) Negative-stain EM was performed on the recombinant MIS12CNSL1-258–SPC2457–197–SPC2570–224 complex. (B) Class averages from negative-stain EM images shown in A. The arrowheads point to an element of density that is not present in the class averages of the MIS12CNSL1-258 complex shown in Fig. 1 D and that likely represents the SPC24–SPC25 complex. Bars: (A) 10 nm; (B) 5 nm.
Figure 9.
Organization of the KMN network. (A) Schematic illustration of the organization of the human MIS12C complex. (B) Summary of interactions identified in this study. The black dots connected by a line represent cross-links. Yellow dots represent defined binding sites. Yellow stars represent undefined binding sites. The orange cylinder represents a predicted C-terminal helix in the MIS12 subunit. (C) Summary of interactions in KMN complexes in different species. Black arrows represent established interactions. Blue arrows represent established binding requirements that have not been mapped at the molecular level. Purple arrows represent putative interactions. In C. elegans, KBP-4 and KBP-3 (SPC24 and SPC25 homologues, respectively) are 97- and 134-residue proteins lacking the globular domains of SPC24 and SPC25 in other species. A globular domain at the C-terminal end of KNL1 is also missing in this organism (see alignment in
Fig. S4
). In S. cerevisiae, Mtw1 (MIS12 homologue) does not have a PVIHL motif or a positively charged C-terminal domain, and the binding site for Spc24p–Spc24p and Spc105p (KNL1 homologue) is therefore unknown. The four-color scheme for MIS12C in C. elegans and S. cerevisiae conveys that the binding sites have not been mapped and could therefore be anywhere on these structures. (D) An extension of the two-hand model (Cheeseman et al., 2008) for kinetochore recruitment of the KMN network based on our experiments. MT, microtubule.
References
- Bieniossek C., Richmond T.J., Berger I. 2008. MultiBac: multigene baculovirus-based eukaryotic protein complex production. Curr. Protoc. Protein Sci. Chapter 5:Unit 5.20. - PubMed
- Brasher S.V., Smith B.O., Fogh R.H., Nietlispach D., Thiru A., Nielsen P.R., Broadhurst R.W., Ball L.J., Murzina N.V., Laue E.D. 2000. The structure of mouse HP1 suggests a unique mode of single peptide recognition by the shadow chromo domain dimer. EMBO J. 19:1587–1597 10.1093/emboj/19.7.1587 - DOI - PMC - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Research Materials