Survivin enhances Aurora-B kinase activity and localizes Aurora-B in human cells

These findings suggest that human survivin is required for targeting of Aurora-B in mitosis.


INTRODUCTION
Survivin was first identified as a novel transcript in the opposite orientation to the effector cell protease receptor-1 (EPR-1) gene. The functions of these two genes are unrelated and survivin has quickly attracted much attention because it is overexpressed in most cancer cells but not in adjacent normal tissues (1). Sequence analysis suggests that survivin contains a domain shared by members of the inhibitor of apoptosis protein (IAP) gene family. The IAP family proteins was first identified in baculovirus and all of them contain variable number of baculoviral IAP repeat (BIR) domains (2). Several human and Drosophila IAPs inhibit apoptosis by directly binding and potently inhibiting caspases (2). However, the evidence for survivin directly inhibiting caspases remains controversial (3,4). Nevertheless, overexpression of survivin provides a survival advantage for cells undergoing IL-3 withdrawal, treatment with Fas, or treatment with chemotherapeutics (1,4).
In normal cells, the expression of survivin is limited to the mitotic stage of the cell cycle. Recent studies suggest that survivin plays an essential role in cytokinesis (5).
Down-regulation of survivin in HeLa cells by antisense oligonucleotides leads to cytokinesis defects (6,7). Both immunostaining of endogenous survivin and ectopic expression of GFP-tagged survivin have shown that survivin is sequentially colocalized with Aurora-B and the inner centromere protein, INCENP, to the centromeres, the spindle midzone, and then to the cleavage furrow during mitosis, a typical behavior of chromosomal passenger proteins (7)(8)(9). 5 The genetic evidence for the functional interaction among survivin, Aurora-B kinase and INCENP comes from gene knockout studies (10,11). Survivin knockout embryos are polyploid and show disrupted microtubule formation resulting in lethality at embryonic day 4.5. This phenotype was remarkably similar to that seen in INCENP knockout mice(11), consistent with the idea that survivin and INCENP act together in the same pathway. In budding yeast, mutants of the orthologs of Survivin (Bir1p), Aurora-B (Ipl1p), and INCENP (Sli15) shown similar defects in chromosome segregation (12,13).
Moreover, overexpression of Ark1p, a fission yeast ortholog of Aurora-B, can rescue the defects associated with a Bir1p mutant (14), indicating that yeast ortholog of Aurora-B and survivin can functionally complement each other. Similar phenomena have also been observed in C. elegans. Ablation of C. elegans ortholog of Aurora-B, AIR-2 by RNAi also results in cytokinesis defects, which are indistinguishable from the down-regulation of the ortholog of survivin, BIR-1 by RNAi (15) Biochemically, yeast and Xenopus laevis Aurora-B and INCENP interact with each other and their interaction is required for the completion of cytokinesis (13,16). A recent study has also shown that human survivin interacts directly with Aurora-B and 6 Among chromosome passenger proteins, only Aurora-B is a mitotic serine/threonine kinase. It has been implicated to play key roles in chromosome segregation, cytokinesis and cancer development (8,19,20). C. elegans Aurora-B is required for recruiting ZEN-4/CeMKLP, a key kinesin related protein essential for completion of cytokinesis (21).
Aurora-B phosphorylates histone H3, a requirement for premitotic chromosome condensation (8,22). The substrates of Aurora-B kinase also include the kinesin-related protein Eg5(23), other chromosomal passenger proteins CENP-A and INCENP (24,25), and the myosin II regulatory light chain, a key component of the actomyosin ring (26).
Finally, two recent studies have shown that Aurora-B kinase activity is required for kinetochore-microtubule interactions and microtubule dynamics during mitosis (27,28).
Taken together, the data to date suggest that survivin interacts with Aurora-B physically, genetically, and functionally. Current evidence suggested that these proteins together with INCENP might form a complex (8), however, the role that survivin plays in the complex is not fully understood. Although survivin is overexpressed in most cancer cells, the mechanism which links survivin to tumor development is poorly understood. In this paper we have focused on the role survivin plays with Aurora-B in this complex. We describe that survivin binds to the catalytic domain of Aurora-B, enhancing its kinase activity both in vitro and in vivo, and targets Aurora-B to its substrates. Human survivin (residues 1-120) was cloned into a modified pET14b (Novagen) using primers: 5'-acgtgtccatgggtgccccgacgttgc-3'and 5'-acgtgtctcgagcttattgttggtttcctttg-3'.
GST-survivin constructs were generated as previously described (3,29). All clones were subsequently confirmed by sequencing analysis. Pharmaceuticals and used as previously described (7).

Protein Expression, Purification and Biotinylation -
Immunoblotting -50 µg of cell lysate or 1X10 5 cells lysed in Laemmli sample buffer (Sigma) were resolved on SDS gels. Immunoblotting were performed as described 9 previously (7). The rabbit polyclonal antibody against phosphorylated Ser-10 of histone H3 was purchased from Santa Cruz Biotechnology Inc. The rabbit polyclonal antibody against survivin was purchased from R&D System. objective.

Survivin interacts directly with the catalytic domain of Aurora-B.
To characterize the interaction between survivin and Aurora-B, we conducted pulldown experiments using biotin-labeled survivin 1-120 and [ 35 S]-methionine-labeled Aurora-B. As demonstrated by a previous study (9), we observed the same interaction between survivin and Aurora-B (Fig 1A). Similar results were obtained using GST-survivin

instead of biotin labeled survivin (data not show). The interaction between survivin and
Aurora-B was also confirmed by anti-Aurora-B antibody immunoprecipitaion followed by Western blot analysis with anti-survivin antibody (data not shown). Although a previous study reported that the localization of Aurora-A kinase protein depends on the N-terminal non-catalytic domain of the kinase in Xenopus laevis (30), we found that the catalytic domain of Aurora-B was sufficient to bind to survivin-coated streptavidin beads, but not streptavidin beads alone (Fig 1B). A non-related [ 35 S]-methionine labeled protein, p27 kip1 , was used to confirm that the interaction between survivin and Aurora-B was specific. Under the same conditions, p27 kip1 did not bind to survivin-coated streptavidin beads, nor did it bind to streptavidin beads alone ( Fig 1C). These results indicate that survivin specifically binds to the catalytic domain of Aurora-B.

Survivin enhances Aurora-B kinase activity in vitro.
When a protein binding to the catalytic domain of a kinase, it usually acts as a substrate, a regulator, or an adaptor. Since a recent study claimed that Aurora-B does not phosphorylate survivin(9), we hypothesized that survivin might regulate Aurora-B activity. Indeed, we found that in the presence of survivin, Aurora-B phosphorylated one 12 of its physiological substrates histone H3 much more strongly than in the absence of survivin (Fig 2, lane 1, 2). As survivin and histone H3 have similar molecular weights, we included a control without histone H3 to rule out the possibility that the increase of 33 P labeling was due to the phosphorylation of survivin. Under the same assay conditions we did not observe the phosphorylation of survivin by Aurora-B (Fig 2, lane 3). As a negative control, we constructed a kinase-inactive mutant Aurora-B K106R (31). Using Aurora-B K106R , we did not observe any detectable kinase activity under the same conditions (Fig 2, lane 4, 5, 6 and 7).

Overexpressing survivin enhances Aurora-B kinase activity in vivo.
To determine whether survivin stimulates Aurora-B kinase activity in vivo, we first generated HEK 293 cells stably transfected with N-terminal His-tagged survivin ( Fig   3A). Subsequently, we overexpressed Aurora-B in both HEK293 cells that overexpress survivin and vector control HEK 293 cells that do not overexpress survivin. Although similar expression of Aurora-B in both transfected cells was confirmed by immunoblot ( Fig 3B), we found that immunoprecipitated Aurora-B kinase activity was higher in cells that overexpress survivin using histone H3 as a substrate (Fig 3C).

Down-regulation of survivin decreases Aurora-B kinase activity in vivo.
Next, we used survivin antisense oligonucleotides to down-regulate endogenous survivin in HeLa cells (Fig 4A). We discovered that the immunoprecipitated endogenous Aurora-B activity in survivin antisense oligonucleotide-treated cells was 9 fold lower than that in missense oligonucleotide-treated cells (Fig 4B), even though Aurora-B protein level was unaffected (Fig 4C). This result indicates that survivin regulates Aurora-B kinase activity rather than its protein stability. We reasoned that regulating phosphorylation, since a previous study showed that RNAi mediated AIR-2 reduction led to a defect in histone H3 phosphorylation by immunostaining (17). Using an antibody against phosphorylated Ser-10 of histone H3, we found that the phosphorylated histone H3 in survivin antisense oligonucleotide-treated cells is much lower than that in survivin missense oligonucleotide-treated cells (Fig 4D). These observations further confirmed that survivin enhances Aurora-B kinase activity in vivo.

Survivin is required for targeting of Aurora-B.
To test if human survivin, like C. elegans BIR1, targets Aurora-B to chromosomes, we used a commercially available monoclonal anti-Aurora-B antibody to detect the localization of Aurora-B in HeLa cells that had been treated with survivin antisense or missense oligonucleotides. In survivin missense oligonucleotide-treated HeLa cells we observed that survivin and Aurora-B colocalized throughout mitosis as previously described (9). For example, both survivin (Fig 5A and B) and Aurora-B (Fig 5C and D) colocalized to the spindle midzone at metaphase. As reported previously by us and others, we observed an obvious increase in the number of multinucleated cells in survivin antisense oligonucleotide-treated cells (7). Survivin staining in the multinucleated cells was below the level of detection and was not observed in the mitotic apparatus (Fig 5E and F, cell shown at metaphase). In addition, Aurora-B staining was not localized to the mitotic apparatus in multinucleated cells (Fig 5G and H, cell shown in metaphase).
However, there was some diffuse staining of Aurora-B throughout the cytoplasm ( Fig   5G). This diffuse staining is consistent with the result that Aurora-B protein level remains the same in both survivin antisense and missense oligonucleotide-treated cells (Fig 4C).

DISCUSSION
In summary, we have found that survivin localizes Aurora-B to its substrates and enhances its kinase activity in vitro and in vivo. These findings may provide a possible mechanism of action of survivin in cells.
First, we speculate that the cytokinesis defects caused by depletion of survivin are through mislocalization of Aurora-B and a decrease in its kinase activity since Aurora-B is known to recruit ZEN-4 and phosphorylates the myosin II regulatory light chain, two essential proteins for the completion of cytokinesis (21,26). Results from overexpression of the Aurora-B K106R mutant also indicate that Aurora-B kinase activity is essential for normal cytokinesis (32). Our evidence that ablation of survivin in HeLa cells causes mislocalization of Aurora-B, a decrease of Aurora-B kinase activity, and increased cytokinesis defects further supports our speculation.
Second, survivin has been implicated in the regulation of microtubule stability (33).
We, along with others, could not find evidence that survivin directly binds to microtubules (34). Two recent studies have shown that Aurora-B kinase activity is required for regulation of microtubule dynamics (27,28). Therefore, survivin is a positive regulator of Aurora-B may explain how survivin regulates microtubule stability. In conclusion, our observations in this study show that survivin directly stimulates the kinase activity of Aurora-B and properly localize Aurora-B, thus providing a mechanism as to how survivin exerts its function in cells.