Report

Bora and the Kinase Aurora A Cooperatively Activate the Kinase Plk1 and Control Mitotic Entry

See allHide authors and affiliations

Science  20 Jun 2008:
Vol. 320, Issue 5883, pp. 1655-1658
DOI: 10.1126/science.1157425

Abstract

A central question in the study of cell proliferation is, what controls cell-cycle transitions? Although the accumulation of mitotic cyclins drives the transition from the G2 phase to the M phase in embryonic cells, the trigger for mitotic entry in somatic cells remains unknown. We report that the synergistic action of Bora and the kinase Aurora A (Aur-A) controls the G2-M transition. Bora accumulates in the G2 phase and promotes Aur-A–mediated activation of Polo-like kinase 1 (Plk1), leading to the activation of cyclin-dependent kinase 1 and mitotic entry. Mechanistically, Bora interacts with Plk1 and controls the accessibility of its activation loop for phosphorylation and activation by Aur-A. Thus, Bora and Aur-A control mitotic entry, which provides a mechanism for one of the most important yet ill-defined events in the cell cycle.

Entry into mitosis is controlled by the activation of cyclin-dependent kinase 1 (Cdk1), whose activity is regulated directly by mitotic cyclins and phosphatase Cdc25 (1) and indirectly by Polo-like kinase 1 (Plk1) and Aurora A (Aur-A) (24). Plk1, Cdc25, and Cdk1 form a feedback loop and positively regulate each other's activity (1). A fundamental question in the study of cell-cycle regulation is, what activates this feedback loop? Activation of Plk1 is probably an initiating event (5, 6), because Plk1, through phosphorylation, activates Cdc25 (5, 7) and down-regulates Wee1 (8), a kinase inhibitory to Cdk1, and because Plk1 can be activated in the absence of the feedback loop (9, 10).

The Plk1 activity is tightly controlled in the cell cycle (11). During mitosis, Plk1 is phosphorylated by an unknown kinase on a threonine residue (Thr210 in Plk1 and Thr201 in Plx1, the Xenopus homolog of Plk1) in its activation loop (T loop), and this phosphorylation is required for its activity (5, 1114). Plk1 consists of an N-terminal kinase domain and a C-terminal Polo-box domain (PBD). The PBD binds to a protein motif that contains a phosphorylated serine or threonine residue, and this binding promotes phosphorylation of the bound protein by Plk1 (15, 16). The activity of Plk1 is also regulated by its own conformation, because the PBD interacts with the kinase domain and suppresses its activity when Thr210 is not phosphorylated in interphase cells (14, 17). This inhibitory interaction is absent in mitotic Plk1 that has been phosphorylated on Thr210 (17). We report that a Plk1-interacting protein, Bora, accumulates in the G2 phase and relieves the autoinhibition by the PBD in Plk1. Furthermore, Aur-A is a physiological Plk1 kinase, which phosphorylates Thr210 in the Plk1-Bora complex and activates Plk1 at the G2-M transition.

We determined the kinetics of Plk1 accumulation and activation in HeLa S3 cells that were synchronously released from a cell-cycle arrest at the G1-S boundary (Fig. 1) (18). Although Plk1 accumulated early in the G2 phase [6 hours after release (TT6)], active Plk1 phosphorylated on Thr210 (Plk1-T210-P) was only detectable late in G2 (TT9), right before mitosis, which is consistent with the measurement of the Plk1 activity (Fig. 1, A and B). Thus, accumulation of Plk1 is not sufficient for mitotic entry, and Plk1 activity is under active regulation. Active Aur-A phosphorylated on Thr288 was detected before Plk1-T210-P (Fig. 1A), suggesting that Aur-A may act upstream of Plk1.

Fig. 1.

Regulated activation of Plk1 at the G2-M transition. (A and B) HeLa S3 cells were synchronized at the G1-S boundary by a double-thymidine arrest (TT) (18), released into fresh media, and harvested at the indicated times (18). Cell-cycle stages were determined by means of fluorescence-activated cell sorting (FACS), and levels of indicated proteins were analyzed by Western blotting (A). Cdk1-Y15-P represents the inactive Cdk1 phosphorylated on Tyr15, and p38MAPK serves as a loading control. Plk1 was immunopurified (IP) from cell lysates and blotted for Plk1 and Bora or assayed for its kinase activity with the use of casein as a substrate (B).

To identify the trigger for the activation of Plk1 at the G2-M transition, we focused on human genes whose transcription is increased in G2 (19). We analyzed the function of these genes during mitotic entry by deceasing their expression in cells with small interfering RNAs (siRNAs) (18) and identified a regulator for the G2-M transition, FLJ22624, whose Drosophila homolog, dBora, participates in asymmetric cell division (20).

Depletion of Bora in HeLa cells synchronously released from the G1-S boundary did not affect the progression from G1 to S and then to G2, but delayed entry into mitosis (Fig. 2, A to C, and fig. S1), a phenotype similar to that of cells depleted of Plk1 (fig. S2) (3, 4, 10). Depletion of Bora did not affect the accumulation of cyclin B, Plk1, or Aur-A at the G2-M transition, but delayed the activation of Cdk1 and the degradation of Wee1 (Fig. 2D). Bora appears to control the activation of Plk1, because depletion of Bora delayed the phosphorylation of Plk1-T210 and the activation of its kinase activity (Fig. 2, D and E). This effect was specific to Plk1, as depletion of Bora only marginally affected the activation of Aur-A phosphorylated on Thr288 (Fig. 2D).

Fig. 2.

Control of mitotic entry by Bora. (A) HeLa cells were control-transfected or transfected with two siRNAs against Bora (siBora-A and -B). Depletion efficiency was determined by Western blotting. (B to E) HeLa cells were synchronized to the G1-S boundary by a double-thymidine treatment and transfected with siRNAs during the second thymidine arrest (18). Cells were released and Taxol was added at 9 hours after release to prevent mitotic exit. The cell-cycle profile was analyzed by FACS [(B) and (C)] and by Western blotting (D). Plk1 was immunopurified and assayed for its kinase activity (E). Solid symbols, siControl; open symbols, siBora-A; triangles, G1 phase; squares, G2 phase and M phase; and circles, M phase. (F and G) HeLa cells were transfected, synchronized by a single-thymidine arrest, and harvested at 8 hours after release (T8) (F), or were synchronized by a thymidine-nocodazole arrest and harvested as prometaphase cells (G). The cell-cycle profile was determined by FACS, and the indicated proteins were assayed by Western blotting. Partial synchronization in (F) enriched G2 cells and generated cells of similar cell-cycle profile between siControl and siBora.

Lack of phosphorylation on Plk1-T210 appears to be the cause of the cell-cycle delay, not a consequence of a change in the cell-cycle profile, because depletion of Bora in cells with cell-cycle profiles similar to those of control cells also resulted in lower levels of Plk1-T210-P (Fig. 2, F and G). On the other hand, ectopic expression of green fluorescent protein (GFP)–Bora increased the amounts of Plk1-T210-P, both in asynchronous cells and in prometaphase cells (fig. S3, A and B). Thus, Bora controls the activation of Plk1 and entry into mitosis by promoting phosphorylation of Thr210.

To understand the mechanism of Bora function, we purified Bora-associated proteins from G2 cells. Mass spectrometric analysis identified Plk1 as a major Bora-interacting protein (fig. S4A). Next, we examined Bora expression and its interaction with Plk1 in the cell cycle. Amounts of Bora peaked in G2 (TT8 to TT9), but gradually decreased once cells entered mitosis (TT10 to TT12) (Fig. 1A) (21). Plk1 and Bora immunoprecipitated with each other, and the Bora-Plk1 complex peaked in G2 (TT8 to TT9) (Fig. 1B and fig. S4B). Although Plk1-T210-P and its kinase activity peaked in mitosis (TT10 to TT11) (Fig. 1, A and B), Bora had already associated with active Plk1-T210-P in G2 cells (TT8 to TT9) (fig. S4B). Bora interacts with Plk1 in a phosphorylation-independent manner, and recombinant Bora interacted with Plk1 and Plx1 in vitro (fig. S4, C to E).

The highly conserved sequence around Thr210 in the Plk1 T loop fits the consensus site that is phosphorylated by Aur-A (fig. S5A) (22), a kinase required for mitotic entry (2). Aur-A directly phosphorylated Plx1 in vitro, and this phosphorylation was enhanced by Bora (Fig. 3A and fig. S5, B to E). This enhancement of Plx1 phosphorylation appears not to be the result of a general stimulation of Aur-A by Bora, as measured by the activity of Aur-A toward histone H3.

Fig. 3.

Activation of Plk1 by Bora and Aur-A. (A) Plx1 and histone H3 were mixed together and then incubated with Aur-A, maltose-binding protein (MBP)–Bora/MBP, and radioactive adenosine triphosphate for 60 min (18). 32P incorporation into Plx1 and histone H3 was plotted after being normalized to their respective samples in lane 3. (B and G) Myc-Plk1 and Myc-Plk1-K were expressed in 293 cells, immunopurified, and incubated with or without recombinant Aur-A and MBP-Bora/MBP (18). Myc-Plk1/Myc-Plk1-K beads were then washed to remove Aur-A and analyzed by Western blotting (WB) or assayed for the Plk1 activity, with casein as a substrate. (C to E) Recombinant Plx1 and Plx1-T201D were sequentially incubated with or without λPPase, with EGTA (to stop the λPPase reaction), and then with or without recombinant Aur-A and MBP-Bora/MBP. Plx1/Plx1-T201D were then analyzed by Western blotting or immunopurified to assay for the Plx1 activity (18). Plx1 (inactive) (C) and Plx1-T201D (E) were purified from asynchronous Sf9 cells, whereas active Plx1 (D) was purified from okadaic-acid–treated mitotic Sf9 cells (7). In a separate experiment in (C), the kinetics of the Plx1 activity were analyzed and plotted after normalizing all the samples to that of –Bora/–Aur-A at the 15-min time point. Diamonds, +Bora/+Aur-A; squares, –Bora/+Aur-A; circles, +Bora/–Aur-A; triangles, –Bora/–Aur-A. (F) Phosphorylation of recombinant Plx1-T201D by Aur-A, as described in (A).

We analyzed the effect of Aur-A and Bora on Plk1 activity in vitro. Myc-Plk1 was expressed and affinity-purified from 293 cells and recombinant glutathione S-transferase (GST)–Plk1 from asynchronous Sf9 cells. Although Bora alone had no effect on Plk1 activation and Aur-A alone enhanced Plk1 activity to some extent, Bora and Aur-A together synergistically stimulated the Plk1 activity seven- to ninefold (Fig. 3B and fig. S5, B and F). Bora greatly stimulated the phosphorylation of the T loop in Plk1, suggesting that binding of Bora may control the accessibility of Plk1-Thr210 to Aur-A.

Aur-A– and Bora-mediated phosphorylation and activation of Plk1 are conserved evolutionarily, because inactive Xenopus Plx1 purified from asynchronous Sf9 cells was efficiently phosphorylated on Thr201 by purified Aur-A and became activated, but only in the presence of Bora (Fig. 3C and fig. S5, B, E, and G to J). However, active Plx1 that had been purified from mitotic Sf9 cells already had Thr201 phosphorylated, and the presence of Bora and Aur-A neither enhanced Plx1-T201-P nor stimulated its kinase activity (Fig. 3D). Dephosphorylation of active Plx1 by λ-phosphatase (λPPase) allowed efficient activation of Plx1 by Bora and Aur-A.

This activation of Plx1 is mediated through phosphorylation of Plx1-Thr201, because Bora and Aur-A failed to stimulate the kinase activity of the Plx1-T201D mutant, even after its dephosphorylation (Fig. 3E). Consistent with this, Bora did not enhance the phosphorylation of the Plx1-T201D protein by Aur-A (Fig. 3F), indicating that the enhancement in the wild-type Plx1 resulted from specific phosphorylation on Thr201 (compare Fig. 3F with Fig. 3A).

To understand the structural basis of the Bora-mediated activation of Plk1, we analyzed interactions between Plk1 and Bora. Both the Plk1 kinase domain (amino acids 13 to 345) (Plk1-K) and the PBD domain (amino acids 352 to 603) (Plk1-PBD) directly associated with Bora (fig. S4, F to I). The PBD interacted with both the N- and C-terminal domains of Bora, and the association between the PBD and Bora was phosphorylation-independent. Consistent with this, the phosphopeptide-binding site in the PBD is not required for the activation of Plx1 by Bora and Aur-A (fig. S5K).

We analyzed the effect of Bora and Aur-A on the kinase domain of Plk1 (Myc-Plk1-K). Myc-Plk1-K was phosphorylated on Thr210 by Aur-A, independent of Bora (Fig. 3G). This was consistent with the fact that, in the absence of the PBD, Thr210 is solvent-accessible in the Plk1-K atomic structure (23). Thus, Bora apparently controls the accessibility of Thr210 in the full-length Plk1, in which the PBD interacts with and inhibits the kinase domain (17). Bora still contributed to the activation of Myc-Plk1-K to some extent (Fig. 3G), probably through its direct interaction with Plk1-K. Therefore, Bora regulates the activation of Plk1 through both T210-P–dependent and –independent mechanisms.

To analyze the role of Aur-A in the activation of Plk1 in vivo, we depleted Aur-A in cells synchronously progressing through the cell cycle (Fig. 4, A and B). Depletion of Aur-A prevented phosphorylation of Plk1-T210 and delayed mitotic entry (Fig. 4, A and B, and fig. S6), even though both Plk1 and Bora were expressed in depleted cells. On the other hand, ectopic expression of Aur-A stimulated Plk1-T210-P, and this effect was dependent on the Aur-A activity and the presence of Bora (Fig. 4C and fig. S3C). Similarly, ectopic expression of Bora also promoted Plk1-T210-P in G2 cells in an Aur-A–dependent manner (Fig. 4D). We conclude that Aur-A appears to control the activation of Plk1 through phosphorylation of Thr210 in vivo.

Fig. 4.

Control of activation of Plk1 by Aur-A. (A and B) HeLa cells were synchronized at the G1-S boundary by a double-thymidine treatment and transfected with siControl (solid symbols) or siAur-A (open symbols) during the second thymidine arrest. Cells were released and Taxol was added at 9 hours after release. The cell-cycle profile was determined by FACS (A) and by Western blotting (B). (C and D) HeLa cells were transfected with Myc-Aur-A/Myc (C) or with GFP-Bora/GFP (D) and synchronized by a single-thymidine treatment. At 7 (D) or 8.5 (C) hours after release, cells were treated with 1 μM VX-680, a small molecule inhibitor of Aur-A, and harvested 1 hour later. The cell-cycle profile was determined by FACS, and the indicated proteins were analyzed by Western blotting of either total cell lysates or Plk1 immunoprecipitates (IP). (E) A summary of the regulation of Plk1 by Bora and Aur-A in the cell cycle.

dBora and Drosophila Aur-A (dAur-A) have been characterized to act in a genetic pathway required for asymmetric division of sensory cells (20). Although it has been reported that dBora interacts with dAur-A when both proteins are overexpressed in culture cells (20), human Bora did not coimmunoprecipitate with Aur-A when they were expressed at their physiological levels (fig. S4B). We find that Plk1 binds to Bora and is the target of regulation by Bora in mammalian cells (Fig. 1 and fig. S4). When Plx1 and histone H3 were incubated together, Bora specifically stimulated Aur-A–mediated phosphorylation of Plx1-T201, not of histone H3 or Plx1-T201D (Fig. 3 and fig. S5). Furthermore, depletion of Bora in vivo reduced the activity of Plk1, not Aur-A (Fig. 2, D, E, and G). These observations point to a difference in function of Bora in human versus Drosophila cells.

Phosphorylation of Plk1-T210 and Plx1-T201 is required for mitotic entry (6, 24). Although a Xenopus Plx1 kinase, xPlkk1, phosphorylates and activates Plx1 in vitro (25), xPlkk1 acts downstream of Plx1 and is not the trigger for mitotic entry (6, 26). We found that Bora and Aur-A function cooperatively to activate Plk1 for mitotic entry (Fig. 4E). The accessibility to Thr210 in its unphosphorylated state is blocked by the PBD in interphase cells (17). Induction of Bora in G2 and its subsequent binding to Plk1 increases the accessibility of Plk1-T210, which overrides the autoinhibition by the PBD and promotes the phosphorylation of Plk1-T210 by Aur-A. Active Plk1 then initiates the Plk1-Cdc25-Cdk1 positive feedback loop for mitotic entry. Consistent with Aur-A and Plk1 acting upstream of Cdk1, inhibition of Cdk1 did not affect or only weakly reduced the activating phosphorylation of Aur-A and Plk1 in G2 and in mitosis (fig. S7). On the other hand, activation of Plk1 and Aur-A mutually depend on each other in both G2 and mitosis (Fig. 4B and figs. S2C and S7), indicating that these two kinases form a positive feedback loop for mitotic entry. Our discovery of the Aur-A– and Bora-mediated activation of Plk1 is part of this loop (Fig. 4E).

In mitosis, active Plk1 with Thr210-P binds to its substrates that have been phosphorylated on a PBD-recognition motif, and this binding further activates Plk1 (15). We found that a phosphorylated PBD-recognition peptide (P peptide) (15) only enhanced the kinase activity of active Plx1-T201-P, but not inactive Plx1-T201 (fig. S8, A to F). Furthermore, incubation of active Plx1, first with P peptide and then with Bora, prevented the enhancement of the Plx1 activity by P peptide (fig. S8, G and H). Thus, Bora competes against P peptide in the regulation of Plk1.

Bora appears to have a dual role in regulating Plk1 activity in the cell cycle. In G2, Bora acts as an activator to allow access of Plk1-Thr210 by Aur-A. In mitosis, Bora, through direct binding to Plk1, interferes with the further activation of Plk1 by phosphorylated substrates. This inhibitory effect of Bora is relieved by its mitotic destruction mediated through a ubiquitin-proteasome pathway ubiquitin ligase in a Plk1-dependent manner (21). Thus, Bora activates Plk1 at the G2-M transition, and active Plk1 then phosphorylates Bora and promotes its degradation in mitosis (Fig. 4E) (21). Phosphorylation of mitotic Plk1 on Thr210 appears to be maintained by Aur-A, primarily through a Bora-independent mechanism (Fig. 2G and fig. S7A), because active Plk1 is no longer in an autoinhibitory conformation in mitosis (17). In summary, the identification of the Aur-A–Bora-Plk1 regulatory circuit in mammalian cells elucidates a key mechanism in cell-cycle regulation.

Supporting Online Material

www.sciencemag.org/cgi/content/full/320/5883/1655/DC1

Materials and Methods

Figs. S1 to S8

References

References and Notes

View Abstract

Navigate This Article