Fig 2: Bora binds AURKA through two Tpx2-like motifs that position a phospho-motif in the kinase active site.a Zoom-in view of the binding interface between Tpx2 and AURKA (PDB: 1OL5). The T-loop phospho-residue Thr288, sulfate ion (yellow), and the phospho-T-loop and sulfate ion coordinating residues of AURKA (H176, R180, R255, and R286) are highlighted in stick representation. b Binding of fluorescein-labeled Tpx21–43 polypeptide to AURKA mutants assessed by monitoring the fluorescence polarization signal in the presence of increasing concentrations of AURKA mutants. Kd values represent mean value with standard deviations of the mean as error bars (n = 3 independent experiment samples). c Western blot analysis of 2-step kinase reactions carried out with Bora18-120 [S27A T52A] phosphorylated (+) or not (-) by the ERK kinase (step 1) in the presence of Plk1K82R KD (substrate) and AURKAWT or mutated versions as indicated (step 2). Blots were probed with antibodies to Bora, AURKA, and phosphoT210 Plk1 or pan Plk1 as indicated (from top to bottom). d Competitive displacement of fluorescein-labeled Tpx21–43 from the indicated AURKA mutants by increasing concentrations of pBora1–224 as monitored by fluorescence polarization. Displayed data points and the half maximal inhibitory concentration (IC50) value represent the average normalized fluorescence polarization signal for each reaction conditions (starting binding signal = 100%, end titration binding signal =0%) (n = 3 independent experiment samples). e Theoretical/illustrative model of pBora binding to the kinase domain of AURKA. AURKA is shown in gray surface with helix aC highlighted in yellow. Bora motifs M1 (blue) and M2 (purple) were modeled by two corresponding motifs from Tpx2 and the phospho-motif M3 (pink) was modeled by the TSS motif from phospho-INCENP.

Fig 3: Bora and Tpx21–43 compete for AURKA binding.a Schematics of a competitive displacement binding assay using fluorescein-labeled Tpx21–43 (FITC-Tpx21–43) as a probe. Complex formation between FITC-labeled Tpx21–43 polypeptide and AURKA followed by the disassembly of the FITC-Tpx21–43/AURKA complex by increasing amount of competitor (cold Tpx21–43, Bora, or pBora) was monitored by fluorescence polarization. b Binding of fluorescein-labeled Tpx21–43 polypeptide to AURKAWT assessed by monitoring the fluorescence polarization signal in the presence of increasing concentrations of AURKAWT. Kd value represents mean value with standard deviations of the mean as error bars (n = 3 independent experiment samples). c Competitive binding assay where fluorescein-labeled Tpx21–43 polypeptide in complex with AURKAWT is displaced by increasing amount of competitor (cold Tpx21–43, Bora1–224, or pBora1–224) and monitored by fluorescence polarization signal. Displayed data points and the half maximal inhibitory concentration (IC50) value represent the average fluorescence polarization for each reaction conditions with standard deviations of the mean as error bars (n = 3 independent experiment samples). ND: not determined. d Binding of fluorescein-labeled Tpx21–43 polypeptide to AURKAT288V assessed by monitoring the fluorescence polarization signal in the presence of increasing concentrations of AURKAT288V. Kd value represents mean value with standard deviations of the mean as error bars (n = 3 independent experiment samples). e, f Competitive binding assay where fluorescein-labeled Tpx21–43 polypeptide in complex with AURKAT288V is displaced by increasing amount of cold Tpx2, Bora1–224, or pBora1–224 competitor (e), or by increasing amounts of cold pBora1–224, pBora18–120, or the indicated pBora18–120 mutants in the M1 and M2 motifs. Data presented as in c. g Activation of AURKAWT ATPase activity by Bora1–224, pBora1–224, and Tpx21–43 as assessed using the ADP-Glo assay with Kemptide substrate. Displayed data points and EC50 values represent the average luminescence for each reaction condition with standard deviations of the mean as error bars (n = 3 independent experiment samples). RLU: relative light unit. h Activation of AURKAT288V ATPase activity by Bora1–224, pBora1–224, and Tpx21–43 as assessed using the ADP-Glo assay with Kemptide substrate. Data presented as in g. i Activity of the AURKAT288V/pBora1–224 complex in the presence of increasing concentration of competitor Tpx21–43 or Bora1–224 assessed using the ADP-Glo assay in the presence of Kemptide substrate. Displayed data points and the half maximal inhibitory concentration (IC50) value represent the average fluorescence luminescence for each reaction conditions with standard deviations of the mean as error bars (n = 3 independent experiment samples). ND: not determined.

Fig 4: In vitro reconstitution of a minimal system recapitulating T-loop phosphorylation of Plk1 by Bora and AURKA.a Domain architecture of Homo sapiens AURKA, Bora, and Plk1. The positions of the T-loop phospho-regulatory sites of AURKA (T288) and Plk1 (T210) are indicated. Cy motifs in Bora correspond to Cyclin-binding motifs that mediate substrate targeting. b Two-step in vitro reconstitution of T-loop phosphorylation on T210 of Plk1 (pT210) by AURKA and Bora1–224. In step 1, Bora is phosphorylated by CyclinA2-Cdk2 kinase (yielding pBora). In step 2, phosphorylated Bora from the reaction mix in step 1 is incubated with AURKA and the catalytically dead mutant Plk1K82R in full-length (FL), polo-box domain deleted (?PBD), or kinase domain-only (KD) forms. During step 2, AURKA phosphorylates Plk1 (noted pT210 Plk1K82R) and Bora or pBora (noted ppBora). c Western blot analysis of kinase reactions carried out with Bora1–224 phosphorylated (+) or not (-) by CyclinA2-Cdk2 (step 1) in the presence of Plk1K82R (FL or ?PBD or KD) and AURKA (step 2). Schematics of the Plk1 constructs used as substrates are shown at bottom. Blots were probed with antibodies to Bora, AURKA, and phosphoT210 Plk1 or pan Plk1 as indicated (from top to bottom). Note that pBora phosphorylated by AURKA displays slower migration on SDS-PAGE compared to pBora phosphorylated only by CyclinA2-Cdk2. In the blot performed with the anti-pT210 Plk1 antibody, the asterisk denotes cross reactivity with the pT288 residue of AURKA.

Fig 5: Tpx2-like motifs and Serine 112 phosphorylation are essential for Bora function in mitotic entry in Xenopus egg extracts.a Schematic for the structure–function analysis of Bora in mitotic entry in Xenopus egg extracts. Interphase extracts (red) were depleted with Bora antibodies and 30 min later supplemented with Bora fragments (pre-phosphorylated by the ERK kinase) and with the recombinant GwlK72M hyperactive kinase to force mitotic entry. Samples were collected at different time-points and analyzed by Western blot (0, 20, 40, 60 min) using mitotic markers to determine whether the extracts enter into mitosis (green) or not. The Western blot shows endogenous Bora levels before (Input) and after (Sn: supernatant) immunodepletion (Ip) from Xenopus egg extracts. b Interphase extracts were immunodepleted using anti-Bora antibodies and 30 min later supplemented with a recombinant hyperactive GwlK72M and Bora1–224 (lanes 2–5), Bora35–157 (lanes 6–9), Bora1–224 (lanes 11–14), Bora18–157 [S27A T52A] (lanes 15–18), Bora18–120 [S27A T52A] (lanes 19–22). A fraction of the extracts was collected at the indicated time-points (0, 20, 40, 60 min) and analyzed by Western blot using specific antibodies to monitor the levels of huGwl, ERK, Plk1, huBora and phosphorylation of Plx1 on T201 (pPlk1), PP1 phosphatase on threonine 320 (pPP1), Cdk on Tyr15 (pTyr), and ERK on Thr202 and Tyr204. The green and red areas at bottom represent high and low CyclinB-Cdk1 (MPF) activity periods. c Interphase extracts were immunodepleted using anti-Bora antibodies and 30 min later supplemented with recombinant hyperactive Gwl (GwlK72M) and Bora18-120 [S27A T52A] (lanes 2–5), Bora18–120 [S27A T52A F25D Y31D] (lanes 6–9), Bora18-120 [S27A T52A F103D F104D] (lanes 10–13). A fraction of the extracts was collected at the indicated time-points (0, 20, 40, 60 min) and analyzed by Western blot using specific antibodies to monitor the levels of Gwl, ERK, Plk1, and Bora as well as phosphorylation of Plx1 on T201 (pPlk1), PP1 phosphatase on threonine 320 (pPP1), and Cdk on Tyr15 (pTyr). The green and red areas at bottom represent high and low CyclinB-Cdk1 (MPF) activity periods.