Fig 1: Aurora A–mediated phosphorylation of PKM2 at T45 inhibits the enzymatic activity of PKM2 by reducing its tetramerization.A, His-PKM2 WT and His-PKM2 T45A mutant were expressed in E. coli. The purified proteins were subjected to measurement of PK activities (n = 5). B, the purified proteins used in (A) were subjected to measurement of the binding affinity toward PEP (n = 3). C, His-PKM2 was incubated with ATP, TPX2 (1–25 aa) and GST, GST-Aurora A, or GST-KD-Aurora A (kinase dead Aurora A, D274A mutant). The activities of PK were measured (n = 3). D, Aurora A was overexpressed in H1299 cells that had endogenous PKM2 knocked down and expressed shRNA-resistant Flag-PKM2 WT or Flag-PKM2 T45A. Flag-tagged proteins were purified by IP and subjected to measurement of the PK activities (n = 2). E, extracts of H1299 cells that had endogenous PKM2 knocked down and expressed shRNA-resistant Flag-PKM2 WT or Flag-PKM2 T45A were incubated with Flag beads. Flag-tagged proteins were purified, subjected to DSS crosslinking, and immunoblotted with the indicated antibodies (left panel). Quantified data were shown (right panel) (n = 3). F, size-exclusion chromatography coupled with static laser light scattering (SEC-MALS) revealed the tetramer/monomer ratio of PKM2 WT and PKM2 T45A mutant. G, molecular dynamics simulation revealed the conformational changes in PKM2 T45 or pT45 and E385 in 10 ns. The first and the last conformations were shown. DSS, disuccinimidyl suberate; IP, immunoprecipitation; MALS, multiangle light scattering; PEP, phosphoenolpyruvate.
Fig 2: Aurora A phosphorylates PKM2 at threonine 45.A, recombinant His-Aurora A and His-PKM2 were incubated with 32P-labeled ATP followed by SDS-PAGE. The gel was subjected to autoradiography and Coomassie blue staining. B, mass spectrometry (MS) data showed threonine 45 of PKM2 was phosphorylated. C, H1299 cells were treated with or without Aurora A kinase inhibitor, MLN8237 (100 nM). PKM2 was isolated and subjected to MS analyses. The number of PSMs (the peptide-spectrum matches) containing PKM2 T45 or PKM2 pT45 were counted (upper panel). The extracts of these cells were immunoblotted with the indicated antibodies (lower panel). D, Flag-Aurora A was coexpressed with HA-PKM2 or HA-PKM2 K433E mutant in 293T cells. Co-IP was conducted with Flag beads. The elutes were subjected to immunoblotting with the indicated antibodies (upper panel). Quantified data were shown (lower panel) (n = 3). E, GST pull-down assay was performed with GST-Aurora A and His-PKM2 or His-PKM2 K433E mutant. The elutes were subjected to immunoblotting with the indicated antibodies (upper panel). Quantified data were shown (lower panel) (n = 2). Co-IP, coimmunoprecipitation.
Fig 3: Dimethylation of K191 on PLK1 attenuates its kinase activity. (A) GST-PLK1 was incubated with 6× His-tagged Aurora A plus Bora in the presence of 1 mM SAM for in vitro methylation assay. The methylation and phosphorylation levels of PLK1 were analyzed by PLK1-K191me2 and PLK1-pT210 antibodies, respectively. (B) Kinetics of PLK1 (purified as in A) kinase activity in the presence of Aurora A and SET7/9 with increased concentration of ATP. Data represent mean ± SEM from three independent experiments. (C) FLAG-PLK1WT and FLAG-PLK1K191R were purified from HEK293T cells co-transfected with GFP-SET7/9 and FLAG-PLK1WT or FLAG-PLK1K191R. ATP agarose beads were used as affinity matrices to absorb purified PLK1 with increased amount (0.25–1 μg). The methylation level of PLK1 was analyzed with the anti-PLK1-K191me2 antibody. (D) Cartoon representation of ADP docked onto PLK1 or dimethylated PLK1 (PLK1-K191me2). Residues for ADP binding are shown as sticks. (E) Representative mitotic phenotypes in PLK1-depleted HeLa cells expressing GFP-PLK1WT or GFP-PLK1K191R shown by time-lapse microscopy and visualized with mCherry-H2B. Scale bar, 10 μm. (F) Quantification of chromosome misalignment of HeLa cells expressing GFP-PLK1WT (n = 108) or GFP-PLK1K191R (n = 101). Data represent mean ± SEM from three independent experiments. (G) Quantification of mitotic duration of PLK1-depleted HeLa cells expressing PLK1WT, PLK1T210D, and PLK1K191R with or without MAD2 depletion. An aliquot of SET7/9 siRNA-treated cells were used as a control. NEBD, nuclear envelope breakdown. At least 103 cells per group were examined from three independent experiments. Data represent mean ± SEM. Statistical significance was tested by two-sided t-test. **P < 0.01; NS indicates non-significant, P > 0.05.
Fig 4: Phosphorylation of PKM2 at T45 promotes glycolysis and serine synthesis pathway.A, endogenous PKM2 was knocked down in H1299 cells and shRNA resistant Flag-PKM2 WT, Flag-PKM2 T45A, or Flag-PKM2 K433E mutant were expressed. Seahorse assays were performed to evaluate the glycolytic flux. ECAR over time (left panel) and ECAR in different stages of the measurement (right panel) were shown (n = 3). B, H1299 cells were treated with or without Aurora A kinase inhibitor, MLN8237 (100 nM). Seahorse assays were performed to evaluate the glycolytic flux. ECAR over time (left panel) and ECAR in different stages of the measurement (right panel) were shown (n = 3). C, the glycolytic tracing assay was performed with 13C-labeled glucose in cells used in (A) (left panel). The relative abundance of the 13C-labeled glycolytic metabolites: G6P, PEP, serine, and IMP were shown (right panel). D, the glycolytic tracing assay was performed with 13C-labeled glucose in cells with or without Aurora A overexpression. The relative abundance of serine was shown. E, Flag-Aurora A was coexpressed with HA-PKM2 in H1299 cells. Extracts of H1299 cells with or without the treatment of serine/glycine starvation for 24 h were incubated with Flag beads. The elutes were immunoblotted with the indicated antibodies (left panel). Quantified data were shown (right panel) (n = 3). F, proximity ligation assay (PLA) was performed in H1299 cells with Aurora A and PKM2 antibodies upon the treatment of serine/glycine starvation for 24 h. DAPI and PCNA were cell cycle indicators (left panel). The scale bar represents 10 μm. The PLA signals per cell were quantified (right panel) (0 h, n = 29; 24 h, n = 27). DAPI, 4′,6-diamidino-2-phenylindole; ECAR, extracellular acidification; G6P, glucose 6-phosphate; IMP, inosine monophosphate; PEP, phosphoenolpyruvate.
Fig 5: Structural rationale for inhibitory activity of VER-150548.(A) X-ray structure of VER-150548 bound to Chk1 highlighting key molecule-protein interactions. Only the protein binding site in the direct vicinity of VER-150548 is shown and the glycine loop has been hidden for clarity. (B) X-ray crystal structure of VER-150548 (ball and stick with green carbon atoms) binding to the ATP-binding site of Chk1 (left) or docking binding mode of Aurora A (right). Selected residues are highlighted: the structurally conserved kinase hinge residues in magenta, two conserved residues which form apolar contacts above (alanine) and below (leucine) the aromatic core of VER-150548 in light blue and the polar residues (orange) lining the solvent exposed “sugar-phosphate” pocket.
Supplier Page from Thermo Fisher Scientific for Recombinant Human AURKA (Aurora A), His Tag Protein