Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/23784809

TheCrystalStructureoftheN-TerminalRegionofBUB1ProvidesInsightintotheMechanismofBUB1RecruitmenttoKinetochores

ARTICLEinSTRUCTURE·FEBRUARY2009

ImpactFactor:5.62·DOI:10.1016/j.str.2008.10.015·Source:PubMed

CITATIONS

30

READS

26

10AUTHORS,INCLUDING:

VictorMBolanos-Garcia

OxfordBrookesUniversity

79PUBLICATIONS1,237CITATIONS

SEEPROFILE

TomomiKiyomitsu

WhiteheadInstituteforBiomedicalResearch

11PUBLICATIONS1,060CITATIONS

SEEPROFILE

JörgGünterGrossmann

UniversityofLiverpool

144PUBLICATIONS4,212CITATIONS

SEEPROFILE

TomLBlundell

UniversityofCambridge

647PUBLICATIONS39,316CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:VictorMBolanos-Garcia

Retrievedon:03February2016

Structure

Article

The Crystal Structure of the N-Terminal Regionof BUB1 Provides Insight into the Mechanismof BUB1 Recruitment to KinetochoresVictor M. Bolanos-Garcia,1,* Tomomi Kiyomitsu,2 Sheena D’Arcy,1 Dimitri Y. Chirgadze,1 J. Gunter Grossmann,3

Dijana Matak-Vinkovic,4 Ashok R. Venkitaraman,5 Mitsuhiro Yanagida,2 Carol V. Robinson,4 and Tom L. Blundell11Department of Biochemistry, University of Cambridge, Cambridge CB2 1GA, UK2CREST Research Program JST, Graduate School of Biostudies, Kyoto University, Kyoto 606-8501, Japan3Molecular Biophysics Group, School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK4Department of Chemistry, University of Cambridge, Cambridge, UK5Hutchison/MRC Research Centre, Cambridge CB2 0XZ, UK

*Correspondence: victor@cryst.bioc.cam.ac.ukDOI 10.1016/j.str.2008.10.015

SUMMARY

The interaction of the central mitotic checkpointcomponent BUB1 with the mitotic kinetochoreprotein Blinkin is required for the kinetochore locali-zation and function of BUB1 in the mitotic spindleassembly checkpoint, the regulatory mechanism ofthe cell cycle that ensures the even distribution ofchromosomes during the transition from metaphaseto anaphase. Here, we report the 1.74 A resolutioncrystal structure of the N-terminal region of BUB1.The structure is organized as a tandem arrangementof three divergent units of the tetratricopeptide motif.Functional assays in vivo of native and site-specificmutants identify the residues of human BUB1important for the interaction with Blinkin and defineone region of potential therapeutic interest. Thestructure provides insight into the molecular basisof Blinkin-specific recognition by BUB1 and, ona broader perspective, of the mechanism that medi-ates kinetochore localization of BUB1 in checkpoint-activated cells.

INTRODUCTION

The missegregation of sister chromatids during mitosis results in

the loss or gain of chromosomes in daughter cells (aneuploidy).

This disastrous outcome is avoided by the mitotic checkpoint for

spindle assembly (SAC), which monitors the proper assembly of

the mitotic spindle and blocks the onset of anaphase until the

kinetochores of all chromosomes receive a bipolar attachment

to spindle microtubules. BUB1 and BUBR1 (BUB1-related

kinase, known as MAD3 in yeast) are multidomain proteins that

play central roles in this process, working together with other

kinetochore-bound components including MAD2, BUB3,

CDC20, and MPS1.

Checkpoint proteins are not recruited simultaneously to kinet-

ochores. Instead, they obey a temporal order of assembly where

the recruitment of the later proteins is dependent on the prior

recruitment of early ones (Sharp-Baker and Chen, 2001; Vanoos-

Structure 17, 10

thuyse and Hardwick, 2005; Wong and Fang, 2006). BUB1 is re-

cruited very early in prophase (Wong and Fang, 2006), promotes

the assembly on centromeres of components of the outer kinet-

ochore (such as BUBR1 and CENP-F), and is essential for

assembly of the functional inner centromere (Taylor et al.,

1998; Boyarchuk et al., 2007). It accumulates at the kinetochore

in SAC-activated cells and assures the correct kinetochore

formation (revised in Logarinho and Bousbaa, 2008). BUBR1,

by contrast, functions as a downstream component that is re-

cruited to the kinetochore at a later stage. Kinetochores that

are not correctly attached to microtubules recruit components

of the mitotic checkpoint, initiating a signaling cascade that

results in CDC20-dependent inhibition of the anaphase-

promoting complex or cyclosome (APC/C) (Figure 1). In

Drosophila, loss of BUB1 causes chromosome missegregation

and lethality (Basu et al., 1999), whereas in mice BUBR1 insuffi-

ciency causes infertility and early aging (Baker et al., 2004). In

human cells, defects in the mitotic checkpoint proteins BUB1

and BUBR1 have been associated with various classes of cancer

(Cahill et al., 1998; Hanks et al., 2004; Kops et al., 2005).

Sequence comparison of BUB1 and BUBR1 shows that these

proteins share a common architecture: a conserved N-terminal

region, a central nonconserved region that contains the binding

region for other mitotic checkpoint components such as BUB3,

and a C-terminal serine/threonine kinase domain. The N-terminal

region mediates the binding of Hs-BUB1 to the mitotic kineto-

chore protein Blinkin (a protein also commonly referred to as

AF15q14); the interaction is essential for the kinetochore locali-

zation of this protein and its function in cell cycle arrest induced

by SAC activation (Kiyomitsu et al., 2007). Residues 1–179 of

fission yeast BUB1 are necessary for targeting the protein Shu-

goshin 1 (SGO1) to centromeres (Vaur et al., 2005). Other reports

have shown that deletion of the N-terminal residues 28–160

of Sp-BUB1 results in a truncated protein unable to recruit

BUB3 and MAD3 to kinetochores (Vanoosthuyse et al., 2004).

Furthermore, it has been shown that deletion of residues 1–47

of Hs-BUBR1 severely impairs mitotic checkpoint activity (Harris

et al., 2005), and the N-terminal region of yeast MAD3 mediates

the binding of CDC20 (King et al., 2007).

Using an approach that combines bioinformatics with

biochemical and biophysical methods, we have mapped the

boundaries of the conserved N-terminal region of BUB1 and

5–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 105

Structure

Crystal Structure of the N-Terminal Domain of BUB1

BUBR1 (MAD3) proteins (Bolanos-Garcia et al., 2005; Beaufils

et al., 2008). In human BUB1, it encompasses residues 1–200,

whereas that of budding yeast spans the first 230 residues. We

now describe the 1.74 A resolution crystal structure of the

conserved N-terminal region of Sc-BUB1 and relate the struc-

ture to the binding of Hs-BUB1 to the mitotic checkpoint factor

Blinkin, an interaction that is essential for the recruitment of

Hs-BUB1 and Hs-BUBR1 (MAD3) to the kinetochore.

RESULTS

Deletion of Residues 1–28 Does Not DisruptDomain StabilityAlthough the N-terminal region of BUB1 from Saccharomyces

cerevisiae comprising residues 1–230 (Sc-BUB1[1-230]) was

successfully overexpressed in Escherichia coli as a soluble

Figure 1. Model Showing the Functions of

BUB1 and BUBR1 in the Mitotic Spindle

Checkpoint during Mitotic Progression

(A) In prometaphase, the nuclear envelope breaks

down and microtubules emanating from opposite

poles attach to the kinetochores of individual sister

chromatids. Core checkpoint components BUB1,

BUBR1, BUB3, MAD1, and MAD2 are recruited to

unattached kinetochores in SAC-activated cells.

Kinetochore localization of BUB1 and BUBR1 is

mediated by the mitotic kinetochore protein Blin-

kin. The latter also establishes physical interaction

with the MIS12 and NDC80 complexes. Cytosolic

BUBR1, BUB3, MAD2, and CDC20 associate to

form mitotic checkpoint complexes, which

interact with the anaphase promoter complex/

cyclosome (APC/C) to render it inactive.

(B) In metaphase, the bipolar attachment and

alignment of all chromosomes at the center of

the cell is reached and APC/C-CDC20 inhibition

released by silencing of the SAC. This is followed

by the onset of anaphase, in which sister chroma-

tids separate and are pulled toward opposite

poles of the cell. When the checkpoint is silenced,

securing can be ubiquitinated by APC/C and

degraded. This results in the release and activa-

tion of separase, which leads to the cleavage of

mitotic cohesions at centromeres and chromo-

some arms to cause chromosome separation

and mitotic progression from M-phase to inter-

phase.

protein and purified to homogeneity,

numerous attempts to crystallize it were

unsuccessful. To define a stable domain

suitable for structural studies, we first

investigated the role of residues at the N

terminus. Secondary structure prediction

programs consistently predicted a flexible

region (residues 23–28) between the first

putative a helix of the N-terminal region

of BUB1 (predicted to encompass resi-

dues 9–22) and the second (H1 in our

crystal structure, residues 32–49). To test this hypothesis,

Sc-BUB1(1-230) was subjected to limited proteolysis coupled to

mass spectrometry and Edman degradation analysis. The larger

stable species had their N termini at residues 24, 26, 27, or 28

(Figure 2A).

Nano-electrospray (nano-ES) mass spectra, analytical gel

filtration, and solution X-ray scattering (SAXS) (Figure 2B; see

Supplemental Data available online) show that this region forms

stable dimers in aqueous solutions. We used SAXS to deter-

mine the compactness and the maximum extension of

Sc-BUB1(1-230) and the shorter construct Sc-BUB1(29-230) in

aqueous solutions. A scattering profile simulation using the

atomic structure of the dimer and its comparison with the

experimental scattering data produces a goodness-of-fit value

of chi of 8.8 for Sc-BUB1(1-230) and 5.6 for Sc-BUB1(29-230)

(Figure 2C). This confirms the good agreement between the

106 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Figure 2. Sequence Conservation of the N-Terminal Domain of BUB1 and BUBR1

(A) The ten a helices seen in the structure of Sc-BUB1(29-230) are indicated above the alignment by solid ribbons. Fully conserved residues in BUB1 and BUBR1

that are located in loop regions are highlighted in dark gray. Sc, Saccharomyces cerevisiae; Hs, Homo sapiens; Sp, Schizosaccharomyces pombe; Mm, Mus

musculus; Dm, Drosophila melanogaster; Xl, Xenopus laevis; Gg, Gallus gallus. Six representative BUB1 sequences and five of BUBR1 were aligned using

ClustalW (Thompson et al., 1994). The inverted triangles above the alignment (;) show the trypsin cleavage sites identified after limited proteolysis and proteo-

mics analysis.

(B) Nano-ES mass spectrum of 45 mM sample (+15 to +12), in the m/z range 3000 and 4500. A single charge state distribution was observed centered on charge

+14 at m/z 3510. These values correspond to a molecular weight of 49,313 Da, consistent with a dimer (Mw of monomeric and dimeric Sc-BUB1[29-230] calculated

from the amino acid sequence is 24,226 Da and 48,452 Da, respectively).

(C) Comparison of the distance distribution [p(r)] function of Sc-BUB1(29-230) (blue) against that of Sc-BUB1(1-230) (red) provides evidence for conformational

heterogeneity due to structural flexibility of the region encompassing residues 1–28.

(D) The SAXS scattering profile simulation (continuous lines) based on the dimer observed in the crystal structure demonstrates that the overall conformation of

Sc-BUB1(1-230) (red) and Sc-BUB1(29-230) (blue) is essentially conserved in solution.

structure of (Sc-BUB1[29-230]) in solution and the crystalline

state. Moreover, the comparison of the SAXS scattering profile

of Sc-BUB1(1-230) and Sc-BUB1(29-230) shows that the radius of

gyration of Sc-BUB1(1-230) decreases from 30.4 A to 28.4 A and

the maximum particle size narrows from 100 A to 90 A when

the N-terminal 28 residues are removed (Figure 2D). Hence,

both hydrodynamic parameters are consistent with an

N terminus extended away from the main part of the structure

of Sc-BUB1(1-230). Based on these analyses, the BUB1 construct

encompassing residues 29–230 (Sc-BUB1[29-230]) was cloned

and overexpressed in E. coli.

Structure 17, 10

Sc-BUB1(29-230) Has a Tetratricopeptide-like FoldThe crystal structure of the N-terminal region of Sc-BUB1(29-230)

from budding yeast was solved at 1.74 A resolution (Table 1 and

Figures 3A–3C), and contains two molecules of BUB1 in the

asymmetric unit, related by a noncrystallographic two-fold

axis, forming a homodimer. The two protomers are associated

primarily by hydrogen bonds and van der Waals interactions

and involve clusters of conserved residues E52-P57, M98-

K109, and R175-R182. Salt bridges between the oxygen atom

of residue D53 of one protomer and the nitrogen atoms (NH1,

NH2) of R179 and R182 of the other, as well as between D55

5–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 107

Structure

Crystal Structure of the N-Terminal Domain of BUB1

of one protomer and K109, R175, and R179 of the other,

contribute further to the stability of the homodimer. The nature

of these interactions gives confidence that this represents the

dimer observed in solution. The average surface area of the ho-

modimer interface measured as the change in the solvent acces-

sible surface area, DASA, is 965 A2, which accounts for 8.6% of

the total surface area of one Sc-BUB1(29-230) protomer. Although

small, this DASA value is within the range observed in many ob-

liged dimers (i.e., DASA R 800 A2; Jones and Thornton, 1996).

The structure of the Sc-BUB1(29-230) protomer consists of ten

a helices (H1–H10) that contribute three major features: a single

a helix (H1), a tandem repeat of three units of the tetratricopep-

tide (TPR) motif (H2-H7), and a C-terminal region encompassing

three a helices (H8-H10) (Figure 3D). Each TPR unit (TPR1,

TPR2, and TPR3, numbered from the N terminal) comprises

two antiparallel a helices, arranged as a helix-turn-helix hairpin.

Superposition of TPR units of Sc-BUB1(29-230) shows that they

adopt similar conformations despite the low sequence identity

(Figures 3E and 3F). TPR1 and TPR3 show low sequence iden-

tity (4%) but have a 1.81 A root-mean-square deviation (rmsd) of

Ca atoms. TPR2 and TPR3 have sequence identity of 8.5% and

2.7 A rmsd of Ca, and TPR1 and TPR2 share a sequence iden-

tity of 5.7% and 2.8 A rmsd of Ca. The linker regions between

Table 1. Data Collection and Refinement Statistics

Native Se-Met Crystal

Data Collection

Space group C2 C2

Cell dimensions

a, b, c (A) a = 130.30,

b = 59.78,

c = 71.29

a = 130.86,

b = 59.95,

c = 71.05

a, b, g (o) 90, 97.96, 90 90, 97.68, 90

Peak

Wavelength 0.9792 0.9792

Resolution (A) 50-1.74

(1.78�1.74)

50-1.80

(1.84�1.80)

Rmerge 3.9 (2.49) 7.2 (18.7)

I / sI 18.6 10.6

Completeness (%) 98.2 (92.8) 99.9 (100)

Redundancy 4.1 (3.5) 7.0 (6.9)

Refinement

Resolution (A) 34.92-1.74

No. reflections 52,138

Rwork / Rfree 18.9/21.6

No. atoms

Protein 3465

CHES molecules 26

Water 234

Average B-factors (A2) 30.4

Rms deviations

Bond lengths (A) 0.009

Bond angles (�) 1.075

One crystal was used for each data set. The statistics shown in paren-

theses are for the highest-resolution shell.

108 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All

TPR units differ in length and local structure: TPR1 and TPR2

are connected by a 310 helix (residues E99-R102) linking helices

H3 and H4, whereas TPR2 and TPR3 are connected by a short

loop (residues I138-K141) between helices H5 and H6.

Although there is no position characterized by an evolutionary

invariant residue, a sequence pattern of small and large

hydrophobic residues in the TPR units can be identified in

Sc-BUB1(29-230). Small hydrophobic residues are mainly at the

positions of closest contact between the a helices that define

a TPR unit, whereas large hydrophobic residues (predominantly

phenylalanine and isoleucine) form the interfaces between adja-

cent TPRs. Hydrophobic residues are essential for the structural

integrity of many TPRs (Main et al., 2003), so it is expected that

those located at equivalent positions in Sc-BUB1(29-230) play

a similar structural role. The fact that residues identified as well

conserved in many TPRs (D’Andrea and Regan, 2003) are poorly

conserved in Sc-BUB1(29-230) (and also in BUB1/BUBR1 from

different species) indicates that important deviations from the

canonical 34-residue TPR motif are tolerated in these proteins.

However, some features typical of other TPR motifs can be

recognized in Sc-BUB1(29-230). For example, the three TPR units

assemble into a relatively extended structure to form a regular

series of antiparallel a helices rotated relative to one another by

a constant 24�. The uniform arrangement of neighboring a helices

gives rise to the formation of a right-handed superhelical struc-

ture with a continuous concave surface on one side and a con-

trasting convex surface on the other. Many of the features that

appear to stabilize the fold are conserved in the close homologs.

For example, the fully conserved residues D56 and R106 form

a salt bridge, which favors the interaction between helices H2

and H4 and stabilizes the end points of the intervening, extended

loop. R106 also contacts the fully conserved residue D104 (H4),

thus allowing a close packing of helices H3 and H4.

The amphipathic helix 8, which encompasses residues Y177

to M194, has the features of a ‘‘capping helix.’’ Such capping

helices have been identified in the TPR motifs of PP5 and HOP

(Das et al., 1998; Scheufler et al., 2000). In structures that consist

of more than one TPR motif, the C-terminal non-TPR component

can adopt different kinds of conformations: it can be unstruc-

tured, it can bind to the concave face as an extended polypep-

tide, or it can assume a completely independent domain organi-

zation. Beyond H8, the C-terminal 30 residues of Sc-BUB1(29-230)

are extended and include two short, a helices (H9 residues S206

to L214 and H10 residues F221 to T228), linked by a short

irregular region (residues I215 to P220). Helices H9 and H10 lie

parallel to the long axis of the TPR core, establishing extensive

polar contacts with TPR helices H3, H4, H5, H6, and H7. These

interactions involve ion pairs E127-R213, E161-R209, K163-

E198, and E171-R207.

Both short-range and long-range interactions are essential

for the stability of tandem arrays of the TPR motif (Main et al.,

2005). The extent of these classes of interactions between

a helices of the TPR motifs of BUB1 might explain the docu-

mented instability of deletion mutants of the N-terminal regions

of Hs-BUBR1 (Bolanos-Garcia et al., 2005 and this report) and

Sc-MAD3 (Larsen et al., 2007). Of interest in this context is the

observation that thermal unfolding of Sc-BUB1(29-230) measured

by far-UV circular dichroism (CD) is highly cooperative

(Figure S1A and Supplemental Data). It follows a two-state

rights reserved

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Figure 3. Overall Structure of the N-Terminal Domain of BUB1

(A) The two molecules observed in the crystal asymmetric unit associate to form a dimer with noncrystallographic two-fold symmetry.

(B) The structure viewed 90� rotated along the minor axis.

(C) Electron density for a helices H5, H6, H7, and H9 and the connecting loop (after density modification) contoured at 1.4 s. The final, refined model is shown

using a ball-and-stick representation. The a helices, loops, and side chains are clearly visible in the initial map. Water molecules are shown as red spheres.

(D) Ribbon diagram showing that this domain consists of ten a helices with a core arrangement of a triple repeat of the TPR motif (TPR1 orange; TPR2 magenta;

TPR3 cyan).

(E,F) Superposition of the three TPRs of Sc-BUB1(29-230). Each molecule representation was generated with Pymol (DeLano, 2002).

transition with a Tm of 63�C and DH of 42 Kcal/mol, a value that

is within the range commonly observed in other TPR domains

of similar size (i.e., 40–65 Kcal/mol). Although dimeric

Sc-BUB1(29-230) retains the folded state in the pH range of 7 to

10 as shown by far-UV CD, it is predominantly disordered at

pH % 5 (Figure S1B and Supplemental Data). Interestingly, the

pH stability profile of Sc-BUB1(29-230) is comparable to that of

the equivalent region in Hs-BUBR1, residues 1–204 (Bolanos-

Garcia et al., 2005).

The pattern of sequence conservation of Sc-BUB1(29-230)

(Figure 4A) shows that loop residues D56, Y101, D104, G137,

I138, G139, and P176 are fully conserved in BUB1 and BUBR1

across species. These residues define a continuous surface

area that is extended by the conservation of residues D55

and N103. Charged residues are clustered in two regions

(Figure 4B), the most prominent of which is acidic, defined by

residues D50, E52, D53, D55, D56, D59, D63, D97, E99, and

D104 and involving helices H2 and H4. A second cluster is

defined by the basic residues R106, K109, K141, R175, R179,

and R182 (H4 and H8). As shown in Figures 3A and 3B, some

of the conserved charged residues of this region are involved

in dimer formation.

Structural Basis for the Interaction of BUB1 with BlinkinHs-Blinkin (BUB-linking kinetochore protein) is a large (265 kDa),

predominantly disordered protein that constitutes the kineto-

Structure 17, 10

chore target of BUB1 and BUBR1 in human cells. Although the

boundaries of the N-terminal domain of Hs-BUB1 that physically

interacts with Blinkin were ill-defined due to the lack of direct

structure information at the time, the study shows that the

N-terminal regions of Hs-BUB1 (residues 1–150) and Hs-Blinkin

(1–728) are involved in the interaction (Kiyomitsu et al., 2007).

The high conservation of residues (z60% similarity between

Sc-BUB1[29-230] and its human counterpart) indicates they are

likely to have similar structures. Indeed, the combined use of

the structure information with functional analyses in vivo allows

us to define the Blinkin binding region of Hs-BUB1. In the later,

the substitution of A106 by D or W, the insertion of G between

A104 and W105 and the substitution of L122 by G all disrupted

the interaction of this protein with Blinkin (Kiyomitsu et al.,

2007). The analysis of the crystal structure of Sc-BUB1(29-230)

strongly suggests that the impaired interaction between these

Hs-BUB1 mutants and Blinkin is due to the disruption of stabi-

lizing interactions among a helices of the TPR units. It also allows

the identification of two highly conserved motifs (GN/DD and

GIG) that, as explained below, play a key role in the binding of

Hs-BUB1 to Blinkin. Superposition of the crystal structure of

Sc-BUB1 with those of proteins containing similar TPRs

(i.e., Hs-PEX TPR, Hs-HOP TPR2, and Hs-PP5 TPR) shows the

GN/DD and GIG motifs of Sc-BUB1 exhibit unique conforma-

tions (Figure 4C). Mutation of residues in each motif (i.e., G20-

D22 and G93-G95) to alanines totally abolished the interaction

5–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 109

Structure

Crystal Structure of the N-Terminal Domain of BUB1

of Hs-BUB1 with Blinkin (Figure 5A). In contrast, mutation of

other highly conserved loop residues (F38A-P39A; K42A-E43A)

had no effect on the Hs-BUB1-Blinkin interaction (Figure 5A).

The observation that the mutation D73K in Hs-BUBR1 (which

was wrongly numbered as D67 in Harris et al., 2005 and is equiv-

Figure 4. Projection of Conserved Regions onto the Protein Surface

(A) Residue conservation is shown according to the sequence-conservation

score of ProSkin (Deprez et al., 2005). Blue and white denote conserved and

variable regions, respectively. Conserved residues of the GN/DD and GIG

motifs are shown in blue, two of the residues that define the hydrophobic

pocket are shown in green, and the CHES molecule is shown using a ball-stick

representation.

(B) Electrostatic surface potential computed with GRASP (Nicholls et al.,

1991). The electrostatic potential is contoured at the 10 kT/e level, with red de-

noting negative potential and blue denoting positive potential. Note the prom-

inent acidic loop defined by the conserved residues E52 to D56 and that

defined by the less conserved residues E167, E171, E188, E193, E198,

D205, and E208. The structure is view rotated 90� along the vertical axis

from that in (A).

(C) The loops encompassing residues D53-D56 and G137-I-G139 (red) show

conformations unique among TPRs of high local-structure similarity (Hs-PEX

TPR is gray, Hs-HOP TPR2 is cyan, and Hs-PP5 TPR is yellow).

110 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All

alent to D56 in Sc-BUB1) results in a defective mitotic check-

point protein confirms the important role of the GN/DD motif

for the interaction between Hs-BUB1 and Blinkin.

The GIG Motif Is Essential for SAC FunctionThe GIG motif (G137I138G139) of budding yeast BUB1 is fully

conserved in the BUB protein family and across species. It has

been suggested that replacement of residues that define the

GIG motif by alanines (also wrongly numbered as residues

G140-I141-G142 in Harris et al., 2005) in Hs-BUBR1 results in

more than 50% reduction in the SAC function by the expression

of these mutants. Further experiments conducted in a strain con-

taining only the mutant form of Sc-MAD3, G156A-S159A, show

that the mutant protein was expressed at wild-type levels but

abolished the interaction of Sc-MAD3 with Sc-CDC20 (Hardwick

et al., 2000). Consistent with these observations, circular

dichroism and analytical gel filtration experiments showed

the double Sc-BUB1(29-230) mutant G137A-G139A is predomi-

nantly dimeric, and exhibits comparable thermal stability

and a native-like fold (Figure S1C). The crystal structure of

Sc-BUB1(29-230) shows that the second and third residues of

the GIG motif are buried. It also shows that the two glycine resi-

dues exhibit a positive phi torsion angle. Interestingly, we

observed that single mutations of the GIG motif in Hs-BUB1

(i.e., G93A and G95A) and of residues flanking the GIG motif

(H92A and T96A) did not affect the interaction of Hs-BUB1 with

Blinkin, whereas the double mutant G93A-G95A and the triple

mutant G93A-I94G-G95A both abolished Blinkin binding

(Figure 5A). Given the well-defined structural roles of the resi-

dues of the GIG motif, it is probable that structural changes

become increasingly delocalized, the greater the difference in

size of the substituted residues from the wild-type and the

greater the number of substituted residues. Structural changes

that result from increased delocalization might provide an expla-

nation of the impaired interaction of the Sc-MAD3 G156A-S159A

double mutant with Sc-CDC20 (Hardwick et al., 2000). Interest-

ingly, the GIG motif of Hs-BUBR1 seems to play a similar impor-

tant role in binding Blinkin because several mutants of this motif

disrupt the interaction (S.D., V.M.B-G., and T.L.B., unpublished

data). When mapped onto the surface, the Blinkin binding region

of Hs-BUB1 defines a discontinues area involving the loop

connecting helices H1-H2, residues in H3 that form part of one

small hydrophobic pocket as well as the loops linking the short

helix 310-H4 and H5-H6 (Figure 5B).

Cancer-Associated Mutations Do Not Disruptthe Interaction with BlinkinWhen the mutations that have been identified and associated

with aneuploidy and cancer progression (Table 2) are mapped

onto the surface, their impact on the structure can be assessed.

The Hs-BUB1 mutations E36D, A130S, and H151D lie in regions

that connect the a helices of the TPR units (Figure 5C), whereas

deletion of helices H5 to H7 and a half of H8 in the mutant

D76-141 could not be accommodated without considerable

disturbance of the entire domain structure (Figure 5C). These

qualitative predictions are supported by SDM (Topham et al.,

1997), I Mutant (Capriotti et al., 2005), and other computer

programs that predict stability changes caused by punctual

mutations (Table 2). Interestingly, none of these mutations

rights reserved

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Figure 5. Analysis of the Hs-BUB1-Blinkin Interaction

(A) Yeast two-hybrid analysis of diverse Hs-BUB1 mutants. The effect of the mutant P23G could not be established as it showed self-activation.

(B) Projection of residues important for binding Blinkin onto the protein surface (salmon color). Residues whose mutation did not compromise the binding with

Blinkin are shown in blue.

(C) Mapping of Hs-BUB1 mutations associated with cancer. Residues absent in the deletion mutant D76–141 are shown in green and residues E36D, A130S, and

H151D in brown.

were predicted to lead to great reductions in stability, in good

agreement with the retention of a SAC response of these

mutants in vivo (Tighe et al., 2001).

APC is a conserved 1.5–1.7 MDa asymmetrical complex that

ubiquitinates a multitude of proteins. APC is composed of at

least 11 subunits, most of which are evolutionarily conserved

including the cell cycle proteins CDC16 and CDC23. Mutations

within TPRs of CDC23 (Sikorski et al., 1990) and the CDC16

homolog CUT9 (Samejima and Yanagida, 1994) cause mitotic

arrest at the metaphase-to-anaphase transition, probably due

to the incorrect packing of neighboring a helices (Sikorski

et al., 1990). In a similar fashion, mapping of Hs-BUB1 mutations

onto Sc-BUB1(29-230) shows some residues connecting a helices

Structure 17, 1

of the TPR units constitute a preferred location of mutations

associated with genetic instability. It also suggests that the

Hs-BUB1 mutations E36D, A130S, and H151D are likely to

disrupt the packing and stability of TPR-forming a helices. Our

functional studies in vivo suggest the Hs-BUB1 mutations

E36D, A130S, and H151D, which contribute to chromosome

instability in cancer cells, impaired the SAC through a mechanism

that is not dependent of the direct interaction between Hs-BUB1

and Blinkin. One possibility is that cancer-associated mutations

of the N-terminal domain of BUB1 impair stabilizing interactions

with BUB1 residues C-terminal to this domain that are not

present in the crystallized protein (for instance, with those

located in the adjacent BUB3 binding region).

05–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 111

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Table 2. Mutations in the N-Terminal Region of Hs-BUB1 Associated with Cancer

Residue

Hs-BUB1

Amino Acid

Substitution Clinical Condition Reference

Localization in

the Structure

Predicted Effect

(I-Mutant method,

pseudo-DDG, Kcal/mol)

36 E/D Colorectal cancer (Cahill et al., 1998) H2 (TPR1) Destabilizing (�0.04)

76–141 Frameshift Colorectal cancer (Cahill et al., 1998) Deletion of H5 (TPR2),

H6 and H7 (TPR3)

Destabilizing

130 A/S Lymph node

metastasis

(Shichiri et al., 2002) Short-loop region that

connects helix H7 (TPR3)

with ‘‘the capping

helix’’ (H8)

Destabilizing (�0.94)

140 Transition of the

splicing donor site

Colorectal cancer (Cahill et al., 1998) Unknown

151 H/D Lung cancer (Gemma et al., 2000) Loop region immediately

downstream the ‘‘capping

helix’’ (H8)

Destabilizing (�0.52)

Analysis of Protein-Protein Interaction SitesThe analysis of the crystal structure combined with the use of

bioinformatics tools suggests other residues might be involved

in protein-protein interactions, supporting the notion that BUB1

acts as a molecular scaffold for the recruitment of other compo-

nents of the SAC. In the crystal lattice, several conserved resi-

dues that are localized in a groove formed after dimerization

show extensive interactions with the C-terminal residues I215-

S230 of another Sc-BUB1(29-230) protomer (Figure 6A). These

interactions, which engage helices H1, H2 and H4 of one dimer

and H10 of another, involve salt bridge formation between the

Nz atom of K225 and the oxygen atoms of residues D59 and

D63 as well as hydrogen bonding between residue pairs D55-

S230, D56-S230, and D59-S229 (Figure 6A).

These interactions in Sc-BUB1(29-230) resemble those seen in

several TPR-peptide complexes such as the N-terminal TPR

domains 1 and 2 of Hs-HOP with the C-terminal region of

Hsc70 and Hsp90 respectively (Scheufler et al., 2000) and

Hs-PEX5 with the peroxisomal targeting signal-1 (PTS1) (Gatto

et al., 2000). A Dali search (http://www.ebi.ac.uk/dali/) for struc-

tural homologs returned parts of the TPR motif of Hs-protein

phosphatase 5 (PP5) (Das et al., 1998) longer than 70 residues

as the structure of highest local similarity with parts of

Sc-BUB1(29-230) (3.6 A rmsd of Ca). The other hits were the

TPR domains of Hs-Hsp70/Hs-Hsp90 organizing protein (HOP)

(Scheufler et al., 2000) (3.6 A rmsd of Ca) and Hs- receptor

PEX5 (Gatto et al., 2000) (3.7 A rmsd of Ca). Superposition of

Sc-BUB1(29-230) with the TPR-containing domain of Hs-PP5-

Hsp90(MEEVD), Hs-HOP-Hsp90(MEEVD), and Hs-PEX5-PTS1

complexes shows that residues involved in peptide binding are

located in equivalent a helices (Figure 6B), even though there

is little sequence conservation of residues in equivalent positions

in Sc-BUB1(29-230).

Although the size of the Sc-BUB1(29-230)-I215-S230 interface

(420 A2) is rather small, it is comparable to those of the

protein-peptide complexes Hs-HOP TPR1-Hsc70 and Hs-HOP

TPR2-Hsp90 (430 A2, 380 A2, respectively) and higher than

that of the Hs-PP5 TPR-Hsp90 complex (260 A2). Even though

the small sizes of the area buried would unlikely give tight

binding, it can be anticipated that additional residues cooperate

112 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All

in BUB1-ligand interactions. We note that for instance the inter-

action of Hs-Hsp90 with full-length HOP does not rely on the

Hs-HOP TPR2 binding site alone but also on other polypeptide

segments of HOP. Yet the binding of Hsp90(MEEVD) to Hs-HOP

TPR2 seems to induce a conformational change that most likely

generates further interaction areas within full-length HOP

(Onuoha et al., 2008).

In all of the above TPR-peptide complexes, electrostatic in-

teractions play an important stabilizing role. Use of the ODA

method, a bioinformatics approach that facilitates the identifica-

tion of potential interaction sites (Fernandez-Recio et al., 2004;

Bolanos-Garcia et al., 2006), further supports this notion: the

ODA method predicts that the Sc-BUB1(29-230) region that

extends toward the a helices H2, H4, and H6 is the most favor-

able for protein-protein interactions (Figure S2 and Supple-

mental Data). Furthermore, the residues of Sc-BUB1 shown to

be engaged in protein dimerization in the crystal lattice and those

involved in the interaction with C-terminal residues of another

Sc-BUB1 molecule are located in positions equivalent to resi-

dues N21, D22 (GN/DD motif) and G93, I94, G95 (GIG motif)

of Hs-BUB1, mutation of all of which affects binding Blinkin

(Figure 5B).

Structure superposition of Hs-PP5, Hs-HOP, and Sc-BUB1

shows that residues N36, F39, and N67 of Hs-PP5 involved in

binding Hsp90(MEEVD) and N233, Y236, and N264 of Hs-HOP

TPR2 involved in binding the same peptide Hsp90(MEEVD), map

onto the Sc-BUB1 2-(n-cyclohexylamino)ethane sulfonic acid

(CHES)-binding residues L62, M65, and I110, respectively.

Mutation of residues L45G and L49G in Hs-BUB1, which are

equivalent to residues L84 and M88 of Sc-BUB1(29-230), form

part of the corresponding hydrophobic pocket of the Blinkin

binding region of Hs-BUB1 and abolish the interaction with

Blinkin (Figure 5A), further supporting the idea that this region

is important for BUB1 function.

DISCUSSION

Until now, functions of the various regions of BUB1 have been

assigned using deletion mutagenesis, cell localization, immuno-

precipitation, and other techniques (Taylor and McKeon, 1997;

rights reserved

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Vigneron et al., 2004; Larsen et al., 2007). The structure of

Sc-BUB1(29-230) provides perspective for examining the func-

tions of BUB1 and its homologs. The structure shows that

a divergent triple tandem arrangement of the TPR motif consti-

tutes a suitable structural framework for the physical linkage of

BUB1 to interacting partners. The high conservation of the

N-terminal region among BUB1 from different species and the

use of functional analysis in vivo allow us to define the discon-

tinue Blinkin binding region of Hs-BUB1 and identify two motifs

of unique structure features, GN/DD and GIG. These motifs are

of particular importance for the interaction of Hs-BUB1 with

the N-terminal region of the human mitotic checkpoint factor

Blinkin. It would be interesting to explore whether the Hs-

BUB1-Blinkin interaction has a role, if any, in directional kineto-

chore assembly mediated by Hs-Blinkin. This is relevant

because Hs-Blinkin also interacts with the kinetochore proteins

Zwint-1 and the two subunits hMis13 and hMis14 of the

hMis12 complex (Kiyomitsu et al., 2007).

Hs-BUB1 and Hs-BUBR1 both bind Hs-Blinkin through the

conserved TPR core of the N-terminal domain (this work; S.D.,

V.M.B-G., and T.L.B., unpublished data). Hs-Blinkin binds tighter

to Hs-BUB1 than Hs-BUBR1, suggesting that additional resi-

dues in Hs-Blinkin and/or Hs-BUB1/Hs-BUBR1 participate in

the interaction. Alternatively, concerted regulation of Hs-BUB1

and Hs-BUBR1 during mitosis might occur through the Blinkin

binding region and/or be modulated depending upon the phos-

phorylation state of the N-terminal domain of Blinkin. Also,

certain residues in Blinkin might dictate the specific interaction

with Hs-BUB1 and different ones the interaction with Hs-

BUBR1. The stronger and broader affinity of Hs-BUB1 for Blinkin

suggests that Hs-BUB1 binds to this protein first, followed by

Hs-BUBR1. In any case, the relative affinity of Blinkin for binding

Hs-BUB1 and Hs-BUBR1 is likely to be important not only for

kinetochore-microtubule attachment, but also for the checkpoint

activation functions mediated by these proteins (Figure 6C). The

analysis of contacts in the crystal lattice suggests that clusters of

Q32

W67 D55 D56

D59

D63

K109

K225

S229

S230

Y132N135

Y72S71

Y216

BUB1 HOPPP5

H4

H6Hsp90

Hsp90

BUB1

CHES

H1

H8

H3

H2

PEX5

PTS1

I66

W113

S69L62

L84M88

M65

L117

CHES

D

C

A B Figure 6. Mechanistic Implications

(A) The two Sc-BUB1(29-230) protomers (show in

yellow and magenta, respectively) form a groove

that binds the C-terminal residues I215-S230 of

a symmetry-related molecule (green). Several of

the conserved residues that define an acidic

cluster (orange color) are shown in the yellow pro-

tomer. Residues of the other protomer (magenta)

that interact with residues I215-S230 of a

symmetry-related molecule are shown in blue.

(B) Superposition of TPR-containing domains of

Sc-BUB1, Hs-PP5, Hs-HOP, and Hs-PEX5 shows

that they form a similar concave face. Hs-PP5 and

Hs-HOP bind the same pentapeptide sequence

MEEVD (Hsp90, blue and red, respectively) in

a different mode. Peroxisomal targeting signal-1

peptide (PTS-1, brown) binds Hs-PEX5 in a manner

that resembles that of Hsp90-HOP. In Hs-BUB1,

residues of the Blinkin-binding region map onto

a helices equivalent to those of Hs-PP5, Hs-

HOP, and Hs-PEX5 engage in peptide binding,

suggesting a similar binding mode. In

Sc-BUB1(29-230) various residues of these helices

(H2 and H4) are involved in the interaction with

one CHES molecule (salmon) and others with

the C-terminal region of a symmetry-related

Sc-BUB1 molecule (green), suggesting that other

residues participate in additional protein-protein

interactions.

(C) The TPR-containing domain of Hs-BUB1 and

Hs-BUBR1 bind to the N-terminal region of Blinkin,

whereas the physical interaction of BUB3 with

BUB1 and BUBR1 involves the GLEBS motif.

Putative MAD1 and MAD2 binding sites can be

identified in the region preceding the kinase

domain. The different affinity of Blinkin for

Hs-BUB1 and Hs-BUBR1 might permit the simul-

taneous control of the recruitment of these

proteins to the kinetochore.

(D) Close-up view of the interaction between CHES

and the hydrophobic pocket of Sc-BUB1(29-230).

One CHES molecule is located inside the small

hydrophobic pocket constituted by residues L62,

M65, I66, S69, L84, M88, I110, W113, and L117

(brown area).

Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 113

Structure

Crystal Structure of the N-Terminal Domain of BUB1

highly conserved residues are engaged in Sc-BUB1 dimer

formation. However, only the monomeric form of the corre-

sponding N-terminal region of Hs-BUB1 and Hs-BUBR1 have

been observed under similar conditions, suggesting that homo-

dimer formation is not a prerequisite for the function of these

proteins. Mass spectrometry analysis of cell extracts showed

that stringent buffer conditions/washing steps were needed to

isolate full-length Hs-BUB1 free from contaminant proteins,

including Blinkin and BUBR1 (data not shown). This requirement

hampered the determination of whether dimeric full-length Hs-

BUB1 was present in cell extracts. Yeast-two hybrid assays

have shown the physical interaction between the N-terminal

domains of Hs-BUB1 and Hs-BUBR1 is weak (Kiyomitsu et al.,

2007). However, the formation of Hs-BUB1-BUBR1 hetero-

dimers could not be detected by analytical gel-filtration chroma-

tography, analytical ultracentrifugation, chemical cross-linking,

and native gel electrophoresis, suggesting that the Hs-BUB1-

BUBR1 interaction is of transient nature (data not shown).

It is possible that the N-terminal region of BUB1 functions as

a docking site for the interaction with other, yet unidentified,

BUB1 interacting partners. In this respect, it will be interesting

to establish whether this BUB1 domain physically interacts

with SGO1, because it has been shown that residues 1–179 of

BUB1 from fission yeast are important for targeting SGO1 to

centromeres (Vaur et al., 2005). If this turns out to be the case,

the interaction would shed light into the role of BUB1 in chromo-

some alignment and Hs-BUB1-dependent centromere localiza-

tion of the complex comprising SGO1-MEI-S332 and PP2A

phosphatase (Kitajima et al., 2005; Watanabe, 2005).

Two CDC20-binding regions have been identified in Hs-

BUBR1 (Davenport et al., 2006; Murray and Marks, 2001). One

of these regions (residues 1–477) is highly specific for CDC20

already bound to MAD2 and seems to involve large segments

of BUBR1 (Davenport et al., 2006). In budding yeast, the

N-terminal region of BUBR1 (MAD3) plays a critical role in

CDC20 turnover during mitosis (King et al., 2007). In a similar

fashion, the roles of CDC20 and CDH1 in activation are mediated

by binding of their extended C-terminal region to the TPR motifs

of the APC/C components APC3, APC7, and APC10 (Voderma-

ier et al., 2003). Interestingly, recent evidence suggests that

MAD3 is also a substrate of the APC/C (Burton and Solomon,

2007). Whether the TPR units of MAD3 play a role in the physical

interaction with APC/C remains to be established. A further inter-

esting clue about binding sites in BUB1 comes from the observa-

tion that there is electron density for a CHES molecule bound to

each protomer of the dimer, albeit that bound to chain A in

Protein Data Bank 3ESL has weaker density. Each CHES mole-

cule is located in a small hydrophobic pocket formed by residues

L62, M65, I66, S69, L84, M88, I110, W113, and L117 involving

helices H2, H3, and H4 (Figure 6D). Of these I110, W113 and

L117 contribute approximately 50% of the buried area in the

CHES-Sc-BUB1(29-230) interface. Examination of the CHES

binding site of Sc-BUB1(29-230) shows that L84 is fully conserved

across species and M65 and M88 are conservatively varied in

BUB1 and BUBR1 from vertebrates (Ile and Leu, respectively).

However, if this is a functional site, ligands must vary. The find-

ings that L62, which is conserved in BUB1 and MAD3 from

fission and budding yeast, is substituted by a glutamic/aspartic

acid residue in higher organisms, that S69 is substituted by

114 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All

a glutamic acid residue, and that W113 is substituted by lysine

in BUB1 and BUBR1 from vertebrates, support this notion.

Some of the substitution mutants targeting the TPR core

display different phenotypes in Hs-BUB1 and Hs-BUBR1

(Kiyomitsu et al., 2007). Given the association of Hs-BUB1 and

Hs-BUBR1 in oncogenesis and aneuploidy (Kops et al., 2005)

and the oncogenic property of the fusion of Hs-Blinkin

(AF15q14) with the MLL gene in acute myeloid leukemia and

lung cancer (Hayette et al., 2000), it would be worthwhile to

explore further whether the N-terminal region of Hs-BUB1 (and

also that of Hs-BUBR1) constitutes a suitable target for drug

discovery. This feature, along with the observation of specific

binding of CHES to the hydrophobic pocket of Sc-BUB1(29-230),

open up the possibility of an attractive binding site for chemical

entities that might be useful tools for characterizing its role and

possibly—in the longer term—a target site for drug design. In

conclusion, the crystal structure of the conserved N-terminal

region of BUB1 provides insight into the recognition mechanism

by which the spindle checkpoint protein BUB1 is recruited to

kinetochores, and suggests that disease-causing mutations

located in the Blinkin binding region operate under a mechanism

that is independent of the binding of Hs-BUB1 to Blinkin.

EXPERIMENTAL PROCEDURES

Experimental procedures are described in the Supplemental Data.

ACCESSION NUMBERS

The coordinates of the N-terminal domain of Sc-BUB1 have been deposited in

the Protein Data Bank under the accession code 3ESL.

SUPPLEMENTAL DATA

Supplemental Data include two figures, Supplemental Experimental Proce-

dures, and Supplemental References and are available with this article online

at http://www.cell.com/structure/supplemental/S0969-2126(08)00420-6.

ACKNOWLEDGMENTS

We thank P. Pernot at the ESRF 14-4 beamline station for assistance in data

collection, and N. Harmer for advice on the use of PHENIX. This work was sup-

ported by Cancer Research UK grant C506/A3846 (to V.M.B.G., A.R.V., and

T.L.B.) and Wellcome Trust Programme Grant RG44650 ‘‘The Structural

Biology of Cell Signalling and Regulation: Multiprotein Systems and the

Achievement of High Signal-to-Noise Ratios’’ (to T.L.B.).

Received: August 7, 2008

Revised: October 13, 2008

Accepted: October 17, 2008

Published: January 13, 2009

REFERENCES

Baker, D.J., Jeganathan, K.B., Cameron, J.D., Thompson, M., Juneja, S.,

Kopecka, A., Kumar, R., Jenkins, R.B., de Groen, P.C., Roche, P., and van

Deursen, J.M. (2004). BubR1 insufficiency causes early onset of aging-

associated phenotypes and infertility in mice. Nat. Genet. 36, 744–749.

Basu, J., Bousbaa, H., Logarinho, E., Li, Z., Williams, B.C., Lopes, C., Sunkel,

C.E., and Goldberg, M.L. (1999). Mutations in the essential spindle checkpoint

gene bub1 cause chromosome missegregation and fail to block apoptosis in

Drosophila. J. Cell Biol. 146, 13–28.

rights reserved

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Beaufils, S., Grossmann, J.G., Renault, A., and Bolanos-Garcia, V.M. (2008).

Characterization of the tetratricopeptide-containing domain of BUB1,

BUBR1, and PP5 proves that domain amphiphilicity over amino acid sequence

specificity governs protein adsorption and interfacial activity. J. Phys. Chem. B

112, 7984–7991.

Bolanos-Garcia, V.M., Beaufils, S., Renault, A., Grossmann, J.G., Brewerton,

S., Lee, M., Venkitaraman, A., and Blundell, T.L. (2005). The conserved

N-terminal region of the mitotic checkpoint protein BUBR1: a putative TPR

motif of high surface activity. Biophys. J. 89, 2640–2649.

Bolanos-Garcia, V.M., Fernandez-Recio, J., Allende, J.E., and Blundell, T.L.

(2006). Identifying interaction motifs in CK2beta–a ubiquitous kinase

regulatory subunit. Trends Biochem. Sci. 31, 654–661.

Boyarchuk, Y., Salic, A., Dasso, M., and Arnaoutov, A. (2007). Bub1 is essential

for assembly of the functional inner centromere. J. Cell Biol. 176, 919–928.

Burton, J.L., and Solomon, M.J. (2007). Mad3p, a pseudosubstrate inhibitor of

APCCdc20 in the spindle assembly checkpoint. Genes Dev. 21, 655–667.

Cahill, D.P., Lengauer, C., Yu, J., Riggins, G.J., Willson, J.K., Markowitz, S.D.,

Kinzler, K.W., and Vogelstein, B. (1998). Mutations of mitotic checkpoint genes

in human cancers. Nature 392, 300–303.

Capriotti, E., Fariselli, P., and Casadio, R. (2005). I-Mutant2.0: predicting

stability changes upon mutation from the protein sequence or structure.

Nucleic Acids Res. 33, W306–W310.

D’Andrea, L.D., and Regan, L. (2003). TPR proteins: the versatile helix. Trends

Biochem. Sci. 28, 655–662.

Das, A.K., Cohen, P.W., and Barford, D. (1998). The structure of the tetratrico-

peptide repeats of protein phosphatase 5: implications for TPR-mediated

protein-protein interactions. EMBO J. 17, 1192–1199.

Davenport, J., Harris, L.D., and Goorha, R. (2006). Spindle checkpoint function

requires Mad2-dependent CDC20 binding to the Mad3 homology domain of

BubR1. Exp. Cell Res. 312, 1831–1842.

DeLano, W.L. (2002). The PyMOL Molecular Graphics System (San Carlos, CA:

DeLano Scientific).

Deprez, C., Lloubes, R., Gavioli, M., Marion, D., Guerlesquin, F., and

Blanchard, L. (2005). Solution structure of the E. coli TolA C-terminal domain

reveals conformational changes upon binding to the phage g3p N-terminal

domain. J. Mol. Biol. 346, 1047–1057.

Fernandez-Recio, J., Totrov, M., and Abagyan, R. (2004). Identification of

protein-protein interaction sites from docking energy landscapes. J. Mol.

Biol. 335, 843–865.

Gatto, G.J., Jr., Geisbrecht, B.V., Gould, S.J., and Berg, J.M. (2000). Peroxi-

somal targeting signal-1 recognition by the TPR domains of human PEX5.

Nat. Struct. Biol. 7, 1091–1095.

Gemma, A., Seike, M., Seike, Y., Uematsu, K., Hibino, S., Kurimoto, F.,

Yoshimura, A., Shibuya, M., Harris, C.C., and Kudoh, S. (2000). Somatic muta-

tion of the hBUB1 mitotic checkpoint gene in primary lung cancer. Genes

Chromosomes Cancer 29, 213–218.

Hanks, S., Coleman, K., Reid, S., Plaja, A., Firth, H., Fitzpatrick, D., Kidd, A.,

Mehes, K., Nash, R., Robin, N., et al. (2004). Constitutional aneuploidy and

cancer predisposition caused by biallelic mutations in BUB1B. Nat. Genet.

36, 1159–1161.

Hardwick, K.G., Johnston, R.C., Smith, D.L., and Murray, A.W. (2000). MAD3

encodes a novel component of the spindle checkpoint which interacts with

Bub3p, Cdc20p, and Mad2p. J. Cell Biol. 148, 871–882.

Harris, L., Davenport, J., Neale, G., and Goorha, R. (2005). The mitotic check-

point gene BUBR1 has two distinct functions in mitosis. Exp. Cell Res. 308,

85–100.

Hayette, S., Tigaud, I., Vanier, A., Martel, S., Corbo, L., Charrin, C., Beillard, E.,

Deleage, G., Magaud, J.P., and Rimokh, R. (2000). AF15q14, a novel partner

gene fused to the MLL gene in an acute myeloid leukaemia with a

t(11;15)(q23;q14). Oncogene 19, 4446–4450.

Jones, S., and Thornton, J.M. (1996). Principles of protein-protein interactions.

Proc. Natl. Acad. Sci. USA 93, 13–20.

Structure 17, 10

King, E.M., van der Sar, S.J., and Hardwick, K.G. (2007). Mad3 KEN boxes

mediate both Cdc20 and Mad3 turnover, and are critical for the spindle check-

point. PLoS ONE 2, e342.

Kitajima, T.S., Hauf, S., Ohsugi, M., Yamamoto, T., and Watanabe, Y. (2005).

Human Bub1 defines the persistent cohesion site along the mitotic chromo-

some by affecting Shugoshin localization. Curr. Biol. 15, 353–359.

Kiyomitsu, T., Obuse, C., and Yanagida, M. (2007). Human Blinkin/AF15q14 is

required for chromosome alignment and the mitotic checkpoint through direct

interaction with Bub1 and BubR1. Dev. Cell 13, 663–676.

Kops, G.J., Weaver, B.A., and Cleveland, D.W. (2005). On the road to cancer:

aneuploidy and the mitotic checkpoint. Nat. Rev. Cancer 5, 773–785.

Larsen, N.A., Al-Bassam, J., Wei, R.R., and Harrison, S.C. (2007). Structural

analysis of Bub3 interactions in the mitotic spindle checkpoint. Proc. Natl.

Acad. Sci. USA 104, 1201–1206.

Logarinho, E., and Bousbaa, H. (2008). Kinetochore-microtubule interactions

‘‘in check’’ by Bub1, Bub3 and BubR1: The dual task of attaching and signal-

ling. Cell Cycle 7, 1763–1768.

Main, E.R., Xiong, Y., Cocco, M.J., D’Andrea, L., and Regan, L. (2003). Design

of stable a-helical arrays from an idealized TPR motif. Structure 11, 497–508.

Main, E.R., Stott, K., Jackson, S.E., and Regan, L. (2005). Local and long-

range stability in tandemly arrayed tetratricopeptide repeats. Proc. Natl.

Acad. Sci. USA 102, 5721–5726.

Murray, A.W., and Marks, D. (2001). Can sequencing shed light on cell cycling?

Nature 409, 844–846.

Nicholls, A., Sharp, K.A., and Honig, B. (1991). Protein folding and association:

insights from the interfacial and thermodynamic properties of hydrocarbons.

Proteins 11, 281–296.

Onuoha, S.C., Coulstock, E.T., Grossmann, J.G., and Jackson, S.E. (2008).

Structural studies on the co-chaperone Hop and its complexes with Hsp90.

J. Mol. Biol. 379, 732–744.

Samejima, I., and Yanagida, M. (1994). Bypassing anaphase by fission yeast

cut9 mutation: requirement of cut9+ to initiate anaphase. J. Cell Biol. 127,

1655–1670.

Scheufler, C., Brinker, A., Bourenkov, G., Pegoraro, S., Moroder, L., Bartunik,

H., Hartl, F.U., and Moarefi, I. (2000). Structure of TPR domain-peptide

complexes: critical elements in the assembly of the Hsp70-Hsp90

multichaperone machine. Cell 101, 199–210.

Sharp-Baker, H., and Chen, R.H. (2001). Spindle checkpoint protein Bub1 is

required for kinetochore localization of Mad1, Mad2, Bub3, and CENP-E,

independently of its kinase activity. J. Cell Biol. 153, 1239–1250.

Shichiri, M., Yoshinaga, K., Hisatomi, H., Sugihara, K., and Hirata, Y. (2002).

Genetic and epigenetic inactivation of mitotic checkpoint genes hBUB1 and

hBUBR1 and their relationship to survival. Cancer Res. 62, 13–17.

Sikorski, R.S., Boguski, M.S., Goebl, M., and Hieter, P. (1990). A repeating

amino acid motif in CDC23 defines a family of proteins and a new relationship

among genes required for mitosis and RNA synthesis. Cell 60, 307–317.

Taylor, S.S., and McKeon, F. (1997). Kinetochore localization of murine Bub1 is

required for normal mitotic timing and checkpoint response to spindle

damage. Cell 89, 727–735.

Taylor, S.S., Ha, E., and McKeon, F. (1998). The human homologue of Bub3 is

required for kinetochore localization of Bub1 and a Mad3/Bub1-related protein

kinase. J. Cell Biol. 142, 1–11.

Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994). CLUSTAL W:

improving the sensitivity of progressive multiple sequence alignment through

sequence weighting, position-specific gap penalties and weight matrix choice.

Nucleic Acids Res. 22, 4673–4680.

Tighe, A., Johnson, V.L., Albertella, M., and Taylor, S.S. (2001). Aneuploid

colon cancer cells have a robust spindle checkpoint. EMBO Rep. 2, 609–614.

Topham, C.M., Srinivasan, N., and Blundell, T.L. (1997). Prediction of the

stability of protein mutants based on structural environment-dependent amino

acid substitution and propensity tables. Protein Eng. 10, 7–21.

Vanoosthuyse, V., and Hardwick, K.G. (2005). Bub1 and the multilayered

inhibition of CDC20-APC/C in mitosis. Trends Cell Biol. 15, 231–233.

5–116, January 14, 2009 ª2009 Elsevier Ltd All rights reserved 115

Structure

Crystal Structure of the N-Terminal Domain of BUB1

Vanoosthuyse, V., Valsdottir, R., Javerzat, J.P., and Hardwick, K.G. (2004).

Kinetochore targeting of fission yeast Mad and Bub proteins is essential for

spindle checkpoint function but not for all chromosome segregation roles of

Bub1p. Mol. Cell. Biol. 24, 9786–9801.

Vaur, S., Cubizolles, F., Plane, G., Genier, S., Rabitsch, P.K., Gregan, J., Nas-

myth, K., Vanoosthuyse, V., Hardwick, K.G., and Javerzat, J.P. (2005). Control

of Shugoshin function during fission-yeast meiosis. Curr. Biol. 15, 2263–2270.

Vigneron, S., Prieto, S., Bernis, C., Labbe, J.C., Castro, A., and Lorca, T.

(2004). Kinetochore localization of spindle checkpoint proteins: who controls

whom? Mol. Biol. Cell 15, 4584–4596.

116 Structure 17, 105–116, January 14, 2009 ª2009 Elsevier Ltd All

Vodermaier, H.C., Gieffers, C., Maurer-Stroh, S., Eisenhaber, F., and Peters,

J.M. (2003). TPR subunits of the anaphase-promoting complex mediate

binding to the activator protein CDH1. Curr. Biol. 13, 1459–1468.

Watanabe, Y. (2005). Shugoshin: guardian spirit at the centromere. Curr. Opin.

Cell Biol. 17, 590–595.

Wong, O.K., and Fang, G. (2006). Loading of the 3F3/2 antigen onto kineto-

chores is dependent on the ordered assembly of the spindle checkpoint

proteins. Mol. Biol. Cell 17, 4390–4399.

rights reserved