BubR1 has been implicated in fundamental biological processes such as cell division, cancer, aging and age-related diseases, but whether this mitotic checkpoint protein acts as a kinase has remained a matter of intense debate. With a multi-pronged approach involving the use of structural biology, substrate identification, pharmacological inhibition of kinase activity, and functional assays, Huang et al. now provide compelling evidence that BubR1 has enzymatic activity.

Rare autosomal recessive disorders characterized by cancer and progeroid phenotypes such as mosaic-variegated aneuploidy (MVA) syndrome are increasingly drawing the attention of basic scientists, clinicians, and the pharmaceutical industry because they provide unique opportunities for advancing our understanding of the fundamental molecular, cellular and organismal mechanisms that define the hallmarks of cancer and aging. Most MVA cases are linked to mutations in BUBR1, with patients having biallelic mutations (typically one allele harboring a missense mutation and the other harboring a nonsense mutation) or mono-allelic mutations combined with allelic variants that yield low amounts of wild-type BubR1 protein.1,2 These patients all missegregate chromosomes and accumulate near-aneuploid cells in a broad spectrum of tissues and organs, but are clinically heterogeneous with regards to congenital, progeroid and cancer phenotypes,3,4 implying that BubR1 is a functionally complex protein (Fig. 1). Indeed, BubR1 is a modular protein that contains multiple protein-binding domains that allow for interaction with a wide variety of proteins, underscoring its multifunctionality; its partners include not only classical mitotic regulators such as Mad2, Bub3, Cdc20, Cenp-E, Bub1, Plk1, Aurora B, and PP2A, but also proteins such as Hdac1, AP2B1, Parp1, Casc5 and Brca2.

Fig. 1
figure 1

Physiological functions of BubR1 that potentially involve the protein kinase activity. The demonstration that BubR1 is a functional kinase that phosphorylates Cenp-E Ser2639 to prevent chromosome missegregation in cultured cells raises the important question as to what extent BubR1′s enzymatic activity drives its biological functions in suppressing cancer, senescence, and aging, in limiting insulin receptor signaling, and in regulating ciliogenesis. These efforts will require a genetic approach at the organismal level that inactivates BubR1 kinase activity without affecting protein stability

Full size image

Studies into the biological relevance of BubR1 and its vast interactome have gained further significance with the discovery that BubR1 levels markedly decline with aging and that sustaining high BubR1 levels throughout life via a transgenic approach extends lifespan and attenuates age-related deterioration of multiple tissues, including heart, skeletal muscle and kidney, in addition to providing protection against tumorigenesis.5 Furthermore, BubR1 insufficiency has been instrumental in causally linking accumulation of senescent cells to age-related tissue deterioration, but how low BubR1 activates the senescence program remains unknown. The same holds for its newly discovered roles in insulin receptor internalization and ciliogenesis.6,7

BubR1 is highly homologous to Bub1, indicating that they originated from a duplicated common ancestor gene and subsequently diverged both in amino acid composition and function. Bub1 is a kinase that phosphorylates H2A at Thr120 to target Aurora B to centromeres for attachment error correction.8 In contrast, whether BubR1 is catalytically active has remained a topic of intense debate. Initially, the Cleveland lab documented that the BubR1 kinase domain is necessary for error-free chromosome segregation in Xenopus egg extracts and provided evidence that, in the presence of unattached kinetochores, the motor protein Cenp-E binds to BubR1 to activate its kinase activity for sustained spindle assembly checkpoint signaling and inhibition of anaphase onset (Fig. 1).9 They also showed that recombinant BubR1 can phosphorylate itself or histone H1 in vitro. The Kops laboratory was unable to demonstrate BubR1 kinase activity, but instead provided evidence that the protein is an unusual pseudokinase that depends on its conserved catalytic triad residues K795, D882 and D911 to maintain the structural integrity necessary for binding to Bub1, Bub3, PP2A and Knl, rather than for catalytic activity.10

Settling these discrepancies is particularly important now that BubR1 has been implicated in key biological processes such as maintenance of genomic integrity, cancer, senescence, and aging. In a recent paper published in Cell Research, Huang et al.11 provide compelling evidence that BubR1 is indeed an active kinase. Their first piece of evidence came from resolving the crystal structure of the BubR1 kinase domain. Attempts to crystallize the human BubR1 (hBubR1) kinase domain failed, but the crystallization of the highly homologous Drosophila BubR1 (dmBubR1) kinase domain was successful. The structure revealed that the BubR1 kinase domain adopts a canonical kinase fold with an unusual N-terminal extension that wraps around its N-lobe and stabilizes the activation segment through hydrophobic interactions. The presence of a salt bridge between K1204 and E1213 and the linear stacking of four hydrophobic residues (I1236, C1217, F1327 and H1299) that stabilize the kinase core were additional hallmarks of functional kinase domains. Mutation of key catalytic lysine and aspartate residues of dmBubR1 (K1204 and D1326) or hBubR1 (K795, D882 and D911; Fig. 1) precluded substrate phosphorylation, confirming that the BubR1 kinase domain is capable of transferring phosphates.

Huang and colleagues then identified bubristatin (BRT-1) as a small-molecule inhibitor of BubR1 kinase activity and used it to demonstrate that BubR1 enzymatic activity is important for chromosome alignment. They established that BubR1 phosphorylates S2639 of Cenp-E (Fig. 1), with mutation of BubR1 D882 and D911 residues preventing Cenp-E modification. Importantly, phosphorylation at S2639 allowed kinetochore-associated Cenp-E to switch from a lateral association with microtubules to an end-on stable configuration. Also, BRT-1 interfered with the assembly of the central spindle, which takes place in anaphase after completion of chromosome segregation. It did so by preventing recruitment of the microtubule cross-linking protein PRC1 to the central spindle. These surprising findings indicate that BubR1′s enzymatic activity has mitotic functions beyond the end-on capture of spindle microtubules. Recruitment of PRC1 to the central spindle required BubR1-mediated phosphorylation of Cenp-E at S2639. In sum, Huang et al. show that BubR1 is a functional kinase that phosphorylates Cenp-E at S2639 and provide evidence that this modification plays a central role in kinetochore-microtubule capture and central spindle assembly.

The important question that emerges from these finding is to what extent BubR1′s kinase activity drives BubR1-mediated protection against tumor formation, cellular senescence, aging and age-related diseases, insulin signaling, and ciliogenesis (Fig. 1). Answering it would require the generation of a BubR1 mutant mouse that lacks catalytic activity in the absence of other functional deficiencies. Because mutations in the BubR1 kinase domain tend to destabilize the protein, a critical step in the successful generation of such a model would be to identify a suitable residue that annihilates kinase activity without negatively impacting protein stability.