Structure of a Blinkin-BUBR1 Complex Reveals an Interaction Crucial for Kinetochore-Mitotic Checkpoint Regulation via an Unanticipated Binding Site

Summary The maintenance of genomic stability relies on the spindle assembly checkpoint (SAC), which ensures accurate chromosome segregation by delaying the onset of anaphase until all chromosomes are properly bioriented and attached to the mitotic spindle. BUB1 and BUBR1 kinases are central for this process and by interacting with Blinkin, link the SAC with the kinetochore, the macromolecular assembly that connects microtubules with centromeric DNA. Here, we identify the Blinkin motif critical for interaction with BUBR1, define the stoichiometry and affinity of the interaction, and present a 2.2 Å resolution crystal structure of the complex. The structure defines an unanticipated BUBR1 region responsible for the interaction and reveals a novel Blinkin motif that undergoes a disorder-to-order transition upon ligand binding. We also show that substitution of several BUBR1 residues engaged in binding Blinkin leads to defects in the SAC, thus providing the first molecular details of the recognition mechanism underlying kinetochore-SAC signaling.

Suppl. Figure S2, related to Figure 3. Amino acid sequence alignments of BUB1 and BUBR1(Mad3) from Saccharomyces cerevisiae (Sc) and Homo sapiens (Hs). The secondary structure elements show above the aligned sequences are from the crystal structure of TPR BUB1 (PDB 3ESL). Figure generated with the programme ESPript (Gouet et al., 1999).
Suppl. Figure S3, related to Figure 4. Characterization of BUBR1 mutants. (A) The single (L128A) and two double mutants (L128A-L131A and Y141A-L142A) were studied by far-UV CD after purification by gel filtration chromatography. The CD analyses confirmed that in all the cases the residues substitutions did not compromise the stability of the protein domain and that the protein adopted the native fold state. (B) Nano ES MS data shows that the double mutants BUBR1 L128A-L131A and Y141A-L142A have a very low affinity for the Blinkin mimic peptide compared to the WT protein. Peptide-free BUBR1 is show for comparison. (C) 1d 1 H NMR spectra of BUBR1 wild-type and bindingsite mutants (only an expansion of the up-field methyl region is shown here). Akin to the wild-type, both double-mutants exhibit ring-current shifted methyl signals indicative of tertiary structure. Some differences in chemical shifts presumably arise from Leu-Ala mutation of L128 and L142, which resonate in this region, and potentially subtle local conformational differences/loss of the Y141 ring-current in this area of the structure. (D) yeast-two hybrid experiments showed not interaction between the double mutants BUBR1 L128A-L131A and Y141A-L142A and Blinkin. Mated cells were selected in quadruple dropout (SD/-Ade/-His/-Leu/-Trp) plates containing alpha-galactosidase using protocols described in materials and methods.
Supplementary Movies 1-4. Time-lapse microscopy of BUBR1 mutants. Movie 1. SAC function in a 3FLAG-Venus-BUBR1(WT)-expressing cell as monitored by time-lapse microscopy. The time between one frame and the next is 6 minutes. The time from the NEBD to anaphase for WT BUBR1 was 258 min.
Movie 2. SAC function in a 3FLAG-Venus-BUBR1(KEN26AAA)mutant-expressing cell as monitored by time-lapse microscopy. The time between one frame and the next is 6 minutes. The time from the NEBD to anaphase of BUBR1(KEN26AAA) was 18 min.
Movie 3. SAC function in a 3FLAG-Venus-BUBR1(L128A/L131A)mutant-expressing cell as monitored by time-lapse microscopy. The time between one frame and the next is 6 minutes. The time from the NEBD to anaphase of BUBR1(L128A/L131A) was 180 min.
Movies 4. SAC function in a 3FLAG-Venus-BUBR1(Y141A/L142A)mutant-expressing cell as monitored by time-lapse microscopy. The time between one frame and the next is 6 minutes. The time from the NEBD to anaphase of BUBR1(Y141A/L142A) was 78 min. with 0.5x YPDA and 100 μl plated on SD/-Leu/-Trp (double dropout) and SD/-Ade/-His/-Leu/-Trp (quadruple dropout) plates containing 1 mg/ml X-α-Gal (Clontech). Plates were incubated at 30 °C and growth monitored for two, three or five days. Shown plates are representative of at least three replicates performed using independent yeast transformants.

Supplemental Experimental Procedures
Mass Spectrometry-For the preparation of protein-peptide complexes, pure BUBR1 57-220 was mixed with an excess of peptide (typically 1:3 molar ratio). After 4 °C incubation of the mixture for few hours, each protein-peptide complex was purified by size exclusion chromatography in a Superdex 75 26/60 column previously equilibrated in 50 mM sodium phosphates buffer pH 6.0 containing 200 mM NaCl. Protein-peptide complex was concentrated to 9 mg/ml and flash frozen before storage at -80 °C. Samples for nano-electrospray mass spectrometry (Nano-ESI MS) were prepared by dilution of the original 220 μM stock solution to a final concentration of 45 μM and buffer exchanged to 0.2 M ammonium acetate pH 6.9 using Micro biospin chromatography columns (BioRad). High resolution mass spectroscopy measurements were recorded using a Waters LCT Premier mass spectrometer optimized for the transmission of noncovalent complexes. Typically, 3 μL solution containing BUBR1-peptides were electrosprayed from gold-coated glass capillaries. The pressures and accelerating potentials in the mass spectrometer were optimized to remove adducts while preserving non-covalent interactions. The optimum experimental conditions were obtained with a cone voltage of 87 V, capillary voltage of 1.9 kV, ion energy, 80 V; source pressure 2.7 mbar and time-of-flight analyser pressure 3.3e-7 mbar. All spectra were calibrated internally using a solution of cesium iodide (100 mg/mL). Data were processed withMassLynx 4.0 software (Waters/Micromass) with minimal smoothing and without background subtraction.
Small, single crystals were obtained after optimization of two different crystallization conditions derived from sparse matrix screenings (0.1M magnesium acetate-0.1M sodium cacodylate pH 6.0, 15% PEG 6000 and 0.1M MES pH 6.5, 25% PEG 6000) using conventional vapour diffusion methods. The crystal structure solution was obtained by molecular replacement method using structure of BUBR1 (PDB 2WVI) as the search probe. Crystallographic refinement was performed in REFMAC5 (Murshudov et al., 1997). The model was improved by several rounds of refinement with REFMAC5 and manual rebuilding using COOT (Emsley and Cowtan, 2004) resulting in the final R factor of 19.3% and Rfree = 24.7% (Table I).
Electron density maps obtained clearly showed all -helices and most of the connecting loops were readily traced. Ramachandran plot analysis using PROCHECK (Laskowski et al., 1993) shows that 98.8% residues fall in the preferred and allowed regions, and only 3 residues are present in disallowed regions. Ltd). Backbone atom assignments were completed with standard experiments (Sattler et al., 1999) for the BUBR1-Blinkin complex and extended using 33% sparse, nonuniformly sampled (NUS) experiments and a conventionally-sampled 15 N NOESY-HSQC for the chimera. Data were processed using NMRPipe (Delaglio et al., 1995), with reconstruction of NUS time-domain data using MDD1.6 (Jaravine et al., 2006), and analysed in NMRView (One Moon Scientific). Assignment was aided by an NMRView module (Marchant et al., 2008), which provided rapid input for MARS automated assignment (Jung and Zweckstetter, 2004). Titration of BUBR1 incrementally up to 2 moles of Blinkin peptide to one mole of protein was monitored using 2D 1 H-15 N SOFAST-HMQC spectra (Schanda et al., 2005). Shift changes were ranked based on the