The BUB1B gene encodes the protein budding uninhibited by benzimidazole-related 1 (BUBR1), a vital mitotic pseudokinase of the Spindle Assembly Checkpoint (SAC). This signaling pathway is responsible for delaying anaphase onset until all chromosome are properly attached to microtubules originating from opposing poles of the mitotic spindle, and prevents errors in chromosome segregation, which can lead to aneuploidy and chromosome instability, a pathogenic state with the potential to drive oncogenesis (Holland and Cleveland 2009; Kops et al. 2005).
Human BUBR1 was discovered through sequence-database searches in a study that aimed to identify the relationship between chromosomal instability, the SAC, and neoplasia (Cahill et al. 1998). BUBR1 is considered to be the human homolog of yeast Mitotic Arrest Deficient (MAD) 3 protein, although it was first thought to be the second human homolog of a related kinase from budding yeast, budding uninhibited by benzimidazole 1 (BUB1), which resulted in some confusion in the nomenclature. Spindle checkpoint genes were indeed initially identified in the budding yeast Saccharomyces cerevisiae. MAD3 was discovered in a screen of essential genes necessary for SAC activity and proper cell cycle progression at the end of mitosis (Li and Murray 1991). Mutant yeasts with compromised SAC activity did not arrest in the presence of the microtubule poison benomyl, and died due to chromosomal instability followed by apoptosis, allowing the identification of three mad genes required for SAC activation, mad1–3. Other essential SAC genes are the Bub genes, which were discovered in a similar screen and around the same time as the mad genes. In this case, the microtubule disrupter benzimidazole was used in the screen to block the final stages of cell division in budding yeast; the mutants that continued to divide and form progeny buds were dubbed budding uninhibited by benzimidazole mutants (Hoyt et al. 1991). Subsequent to their identification in budding yeast, efforts were initiated to identify the human homologs of the SAC genes. Similarities at the amino acid level between hBUBR1 and budding yeast BUB1 led to its initial classification as a BUB1 homologue (Cahill et al. 1998). However, there are critical differences between the pseudokinase domain of hBUBR1 and the bona fide kinase domain of BUB1 from yeasts as well as higher eukaryotes, and these two domains have only 20% identity. Moreover, whereas the N-terminal region of hBUBR1 and S. cerevisiae BUB1 display 26% similarity, the N-terminal region of hBUBR1 is 35% similar to budding yeast MAD3 (Taylor et al. 1998). Consequently, it is now well established that BUBR1 is the human homologue of yeast Mad3.
In order to explain the similarity among hBUBR1 and budding yeast BUB1 and MAD3, it was suggested that human BUB1 and BUBR1 originated from a common gene – MADBUB, present in the last eukaryotic common ancestor (LECA). This protein presents two essential SAC domains with distinct functions required for the checkpoint arrest: a sequence containing a lysine(K)-glutamate(E)-asparagine(N) (KEN) box and a kinase domain. The MADBUB gene undertook different paths through evolution: it either suffered a gene duplication event on nine different occasions or remained relatively intact. It has been suggested that around 20–50 million years ago MADBUB suffered a duplication where its domains – and function – were passed on to two distinct genes (Murray 2012; Suijkerbuijk et al. 2012a; Vleugel et al. 2012). In an example of subfunctionalization, in both the fission yeast Schizosaccharomyces pombe and the very distantly related S. cerevisiae, the BUB1 protein retained the kinase domain, while MAD3 lost the kinase domain but preserved the KEN box domains.
On the other hand, in the majority of vertebrates, BUBR1 homologs retained both the KEN box domains and a degenerate kinase-like domain. Substitutions and posttranslational modifications of this domain gave rise to a lively debate in the literature as to whether BUBR1 in higher eukaryotes can function as a true kinase. Although it is now generally accepted that hBUBR1 is indeed a pseudokinase and cannot perform a phosphotransfer reaction, it is noteworthy that at least in one model organism, Drosophila melanogaster, BUBR1 has retained catalytic function, although no substrates have been identified to date (Buffin et al. 2007; Rahmani et al. 2009; Suijkerbuijk et al. 2012a). Despite the controversy surrounding the pseudokinase domain, in all organisms where it has been tested, BUBR1 is a crucial player in SAC signaling (Chan et al. 1999; Li and Murray 1991; Musacchio and Salmon 2007; Rahmani et al. 2009; Wong and Fang 2007). Both human and mouse BUBR1 have also been shown to play a role in chromosome congression to the metaphase plate and in the stabilization of kinetochore-microtubule attachments (Elowe et al. 2010; Lampson and Kapoor 2005; Park et al. 2013; Suijkerbuijk et al. 2012b; Touati et al. 2015; Wei et al. 2010; Xu et al. 2013).
Structural Diversity of MAD3/BUBR1 Orthologs
A MADBUB-like protein has been identified in a number of species, including M. brevicollis, B. dendrobatidis, C. neoformans, N. crassa, D. discoideum, and P. infestans (Vleugel et al. 2012). This protein presents five conserved domains: two KEN box domains; a tetratricopeptide repeat (TPR) domain; a BUB3-binding domain (B3BD, also known as the Gle2-binding-sequence (GLEBS) domain); and a kinase domain. After gene duplication, the BUBR1 homolog in budding and fission yeast, nematodes, A. thaliana and N. gruberi lost the C-terminal kinase domain, yielding MAD3-like proteins. Insects and vertebrates, however, maintained the kinase domain giving rise to BUBR1-like proteins, which as noted above degenerated into a catalytically inactive pseudokinase domain in most higher eukaryotes (Murray 2012; Suijkerbuijk et al. 2012a; Vleugel et al. 2012). Thus, the distinguishing feature of BUBR1 orthologs throughout evolution is the presence of the two N-terminal KEN boxes.
Functions of the Distinct Domains of hBUBR1
The first KEN box motif of BUBR1 and MAD3 proteins is crucial for the interaction with CDC20, which is the target of the SAC, and the formation of the Mitotic Checkpoint Complex (MCC), the effector of the SAC (Burton and Solomon 2007; King et al. 2007). The second KEN box motif is required for direct inhibition of APC/C, a large E3 ubiquitin ligase that when activated by CDC20 (APC/CCDC20) is the major target of the SAC. This KEN box most likely functions by blocking the recruitment of substrates to the APC/CCDC20 (Lara-Gonzalez et al. 2011). Recently, it was demonstrated that the loop region of the BUBR1 BUB3-binding domain may also be required for optimal formation of the MCC and APC/C binding (Overlack et al. 2015).
The TPR and B3BD contribute to BUBR1 kinetochore localization, which is required for SAC activation. This recruitment is complex and appears to require multiple interaction interfaces on BUBR1, as well as multiple docking partners, including the obligate BUBR1 binding partner BUB3, and the BUB3-BUB1 dimer, which directly binds to and recruits BUB3-BUBR1. These collectively dock on KNL1 (Kinetochore null protein 1), which has recently emerged as the major SAC signaling platform at kinetochores (Kiyomitsu et al. 2007; Krenn et al. 2014; Primorac et al. 2013; Vleugel et al. 2013). Binding of BUBR1 to kinetochores absolutely requires the direct interaction between BUBR1 and BUB1 mediated through residues 271–409 of hBUB1 and 362–571 of hBUBR1 (Primorac et al. 2013). Both of these regions lie downstream of the BUB3 binding domain in their respective proteins. This pseudosymmetric complex between BUB1 and BUBR1 is further stabilized by a number of additional interactions. The TPR domain of BUBR1 binds to the second lys-Ile (KI2) motif of KNL1 (Kiyomitsu et al. 2011; Krenn et al. 2014; Krenn et al. 2012), although in vivo, this interaction is not necessary for BUBR1 kinetochore docking (Krenn et al. 2012). On the other hand, the pool of BUB3 associated with BUBR1 is essential for BUBR1 kinetochore recruitment; and although its function is not entirely clear, it may play a role in direct recruitment to BUB1 (D’Arcy et al. 2010; Overlack et al. 2015), or may mediate further interaction with KNL1 phosphorylated Met-Glu-Leu-pThr (MELT) motifs (Primorac et al. 2013).
In the central region of BUBR1 is another broad motif that promotes BUBR1 interaction with CDC20, and that was independently identified by several groups, resulting in multiple names for it in the literature. These include the Phe-box (residues 528–531) (Diaz-Martinez et al. 2015); IC20BD (residues 490–560, which has a core from residues 530 to 535) (Lischetti et al. 2014); and the ABBA motif (residues 528–533), named after its presence in human A-type cyclins, BUBR1, BUB1, and in budding yeast Acm1 (Di Fiore et al. 2015). Although this region mediates CDC20 binding and APC/C inhibition, its role in the spindle checkpoint-mediated arrest remains unclear (Diaz-Martinez et al. 2015; Di Fiore et al. 2015; Lischetti et al. 2014). It has been implicated in recruiting CDC20 to kinetochores (Di Fiore et al. 2015; Han et al. 2014) and in SAC silencing, presumably by competing with the KEN-boxes for CDC20 binding (Lischetti et al. 2014). In addition, this region may compete with cyclin A for its binding to CDC20, thus explaining how this cyclin is promptly degraded at mitotic exit (Di Fiore et al. 2015).
In the C-terminal region of hBUBR1 lies the kinetochore Attachment Regulatory Domain (KARD) followed by the pseudokinase domain. At the core of the KARD is a Leu-Xaa-Xaa-Ile-Xaa-Glu motif, which binds to the concave side of the B56 pseudo-HEAT repeats of the phosphatase PP2A-B56 (Wang et al. 2016). The association between B56 and BUBR1 KARD has been suggested to recruit a subpopulation of the PP2A phosphatase to the outer kinetochore to counteract the activity of the AURORA B kinase (Nijenhuis et al. 2014). This interaction explains, at least in part, the role of BUBR1 in the regulation of many mitotic functions including chromosome congression, kinetochore-microtubule attachment stability, and the termination of SAC signaling (Espert et al. 2014; Kruse et al. 2013; Suijkerbuijk et al. 2012b; Xu et al. 2013).
The function of the BUBR1 kinase domain has garnered considerable controversy. The presence of a conventional catalytic triad led to its initial classification as an active kinase in humans, mice, and in Xenopus laevis (Cahill et al. 1998). Early studies suggested that BUBR1 kinase activity is required to fully activate the SAC (Kops et al. 2004; Malureanu et al. 2009; Mao et al. 2003), although others argued that catalytic activity of BUBR1 was dispensable for the SAC (Chen 2002; Elowe et al. 2007). Recent results however suggest that phenotypes observed in BUBR1 kinase inactivating mutants are probably due to destabilization of BUBR1, and not the mutation of the kinase domain per se, and catalytic activity previously attributed to BUBR1 has been suggested to be a result of a contaminating kinase (Suijkerbuijk et al. 2012a).
Recent work based on the structural homology with hBUB1 and other active kinases provides mechanistic insight into the inactivation of BUBR1 and demonstrates the presence of residues or potential posttranslational modifications that would render it inactive (Suijkerbuijk et al. 2012a). More specifically, BUBR1 presents two conserved deviations from conventional kinases, a substitution in the Gly-rich loop for large and negatively charged residues (Asp at the equivalent of Gly52 and Leu at the equivalent Gly55 of PKA), and two substitutions at the catalytic loop (Ser and Cys at the equivalent of Lys168 and Asn171 of PKA, respectively). In addition, in two vertebrates, A. carolinensis and D. rerio, substitutions in the catalytic triad (Asp882BUBR1) render BUBR1 in these species a conventional pseudokinase. In the fruit fly Drosophila melanogaster, BUBR1 resembles the fly BUB1 more closely and, consequently, is thought to present a catalytically active BUBR1 protein (Suijkerbuijk et al. 2012a). Despite the lack of catalytic activity, the BUBR1 kinase domain is important for protein stability (Suijkerbuijk et al. 2010). Remarkably, mutations and truncations associated with this domain have been found in patients suffering from Mosaic Variegated Aneuploidy (MVA) (Hanks et al. 2004; Matsuura et al. 2006). This disorder is characterized by high tumor formation and shorter life span due to low BUBR1 levels resulting from its instability (Suijkerbuijk et al. 2010). Furthermore, the substitution of residues at the catalytic triad, used to produce an inactive kinase, is known to destabilize the protein, as in the MVA syndrome (Suijkerbuijk et al. 2012a). Whether the BUBR1 pseudokinase domain has functions unrelated to the protein homeostasis remains to be seen.
PLK1 Is a Major Regulatory Kinase of BUBR1
BUBR1 is phosphorylated at a number of sites that collectively play an important role in its function and regulation, with polo-like kinase 1 (PLK1) being arguably the most important BUBR1 kinase identified to date. Human BUBR1 is phosphorylated at Thr620 and Ser670 by Cdk1 (Elowe et al. 2007, 2010; Huang et al. 2008). Phosphorylation of Thr620 forms a polo-box domain docking site, which promotes PLK1 binding to BUBR1. Subsequently, PLK1 phosphorylates BUBR1 at S676 (Elowe et al. 2007), T680 (Suijkerbuijk et al. 2012b), and Thr792 and Thr1008 (Matsumura et al. 2007). Phosphorylation of BUBR1 at Ser670, Ser676, and Thr680 in the KARD promotes the binding of PP2A-B56 (Kruse et al. 2013; Suijkerbuijk et al. 2012b; Xu et al. 2013). Phosphorylation of Thr1008 and Thr792 by PLK1 was thought to activate BUBR1 kinase activity to promote proper chromosome alignment (Matsumura et al. 2007). However, in light of recent evidence that BUBR1 is a pseudokinase, a re-examination of the function of these phosphosites is required. In a similar fashion, the budding yeast AURORA B kinase IPL1 and the PLK1 ortholog CDC5 is believed to phosphorylate MAD3 on unattached kinetochores in this organism (Rancati et al. 2005). In Xenopus, BUBR1 phosphorylation by PLK1 homologue, PLX1, creates the 3F3/2 epitope (Wong and Fang 2007), long considered a marker for lack of kinetochore tension that establishes a link between the mechanical regulation of chromosome segregation and a biochemically controlled sensor such as the SAC (Ahonen et al. 2005; Nicklas et al. 1995). In human cells however, the 3F3/2 epitope appears to be distinct from BUBR1 (Elowe et al. 2007).
BUBR1 also interacts with the BRCA2 (BReast CAncer susceptibility gene 2) and PCAF (P300/CBP-associated factor) complex, an acetyltransferase responsible for acetylating BUBR1 at Lys250. As part of the MCC, BUBR1 is potentially prone to ubiquitination by the APC/CCDC20. Acetylation of BUBR1 is believed to protect it from APC/CCDC20-mediated ubiquitination and destruction by the proteasome (Choi et al. 2009, 2012; Yekezare and Pines 2009). Upon chromosome biorientation and SAC silencing, BUBR1 is deacetylated which allows for sumoylation of BUBR1 at the previously acetylated Lys250 residue. Sumoylated BUBR1 promotes its release from kinetochores and thus potentially mediates SAC silencing (Yang et al. 2012a, b).
Functions of BUBR1 during Mitosis
BUBR1 contributes to several functions during mitosis: mitotic checkpoint complex (MCC) formation and spindle checkpoint activation, chromosome alignment, formation of proper kinetochore-microtubule attachments, and most recently spindle assembly checkpoint termination.
Spindle Assembly Checkpoint Activation and MCC Formation
Regulation of Kinetochore-Microtubule Attachments and SAC Silencing
The minimum time from nuclear envelope breakdown until anaphase onset is also regulated by BUBR1 and MAD2 (Meraldi et al. 2004). The MCC is also found in interphase, and this interphase MCC is thought to be responsible for the basal inhibition of the APC/CCDC20 in early mitosis before kinetochores fully mature (Sudakin et al. 2001). This function of BUBR1 appears to be dependent on the first KEN box, as flies expressing a mutant KEN box and lacking MAD2 exhibit an accelerated undisturbed mitosis (Rahmani et al. 2009).
Nonmitotic Functions of BUBR1
BUBR1 also plays an important function in meiosis; in meiosis I, BUBR1 functions much like it does in mitosis (Homer et al. 2009); it promotes kinetochore-microtubule attachments and SAC activation (Wei et al. 2010). Additionally, Drosophila BUBR1 is responsible for the integrity of the synaptonemal complex (SC) (Malmanche et al. 2007), which is important for maintaining homologous chromosomes attached. Interestingly, older oocytes present decreased levels of BUBR1 protein compared to younger ones. Consequently, BUBR1 is thought to be an important factor explaining why older women present increased rate of anomalous chromosome segregation, which is associated with miscarriages and a higher probability of whole chromosome aneuploidy (Touati et al. 2015).
BUBR1 is an essential protein for the Spindle Assembly Checkpoint signaling, where it acts as an effector in the Mitotic Checkpoint Complex to inhibit APC/CCDC20 and prevent mitotic exit. It has also an important role in the congression of chromosomes to the metaphase plate and in stabilizing microtubule attachments. Recent studies have also highlighted a potential role for BUBR1 in SAC silencing. An important novel study indicates that BUBR1 in higher eukaryotes possesses no catalytic activity but has nevertheless retained a pseudokinase domain that is apparently required for maintaining the stability of the protein. In the future, it will be interesting to determine whether this domain plays any additional roles during mitosis.
- King EMJ, van der Sar SJA, KG H. Mad3 KEN boxes mediate both Cdc20 and Mad3 turnover, and are critical for the spindle checkpoint. PLoS ONE. 2007;2(4):e342. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1829190/
- Murray AW. Don’t make me mad, Bub! Dev. Cell. 2012;22(6):1123–5.Google Scholar
- Xu P, Raetz EA, Kitagawa M, Virshup DM, Lee SH. BUBR1 recruits PP2A via the B56 family of targeting subunits to promote chromosome congression. Biol Open 2013;2(5):479–486.Google Scholar