Fig 2: The YWHAH phosphorylation at S25 is associated with breast-cancer metastasis.a Representative microscopy photographs of immunohistochemical staining in breast-cancer tissue arrays. b The nuclear levels of KHK-A and YWHAH-pS25 were compared between histologic grade 2 and 3 groups (the means ± S.D. from 36 different breast-cancer tissues). c Nuclear KHH-A and YWHAH-pS25 levels were compared between the non-metastatic group (N.M) and the metastatic groups (M) including lymphatic and distant metastases (the means ± S.D. from 36 different breast-cancer tissues). In (b), (c), significance was calculated by two-sided Mann–Whitney U test. d Spearman’s correlation analysis between nuclear KHK-A and YWHAH-pS25 levels. ‘r’ is the correlation coefficient.

Fig 3: KHK-A phosphorylates YWHAH at Ser25.a In vitro binding assay. Purified recombinant GST-YWHAH (or GST) were incubated with purified KHK-A in a test tube for 1 h and pulled down using nickel-affinity beads. Interaction between KHK-A and YWHAH was evaluated by immunoblotting. b KHK-A-dependent phosphorylation of YWHAH. FLAG-YWHAH was co-transfected with Myc-KHK-A or Myc-KHK-A-dATP. After 8-h incubation with 5 mM fructose, MDA-MB-231 cells were subjected to immunoprecipitation and immunoblotting. c In vitro kinase assay. Recombinant GST-YWHAH and His-KHK-A were co-incubated without or with 1 mM fructose in a kinase buffer. The Ser/Thr-phosphorylation of YWHAH was evaluated by immunoblotting, and the protein loading was verified by Coomassie staining. d LC-MS/MS analysis. Recombinant GST-YWHAH was reacted with or without GST-KHK-A in a kinase buffer, electrophoresed on SDS-PAGE, and digested in gel. The phosphorylation of Ser25 residue in YWHAH was identified based on LC-MS/MS spectra. e MDA-MB-231 stable cells expressing KHK-A, which had been co-transfected with Flag-YWHAH or S25A plasmid, were incubated with 5 mM fructose for 8 h. The cell lysates were immunoprecipitated with anti-FLAG, and the phosphorylation of YWHAH was evaluated by immunoblotting. f MDA-MB-231 cells, which had been co-transfected with Myc-KHK-A and si-YWHAH, were incubated with 5 mM fructose for 48 h. F-actin and nucleus were stained with Alexa Flouor 633 Phalloidine (red) and DAPI (blue), and visualized under a fluorescence microscope (scale bar = 10 μm). White arrows indicate filopodia. g Endogenous YWHAH was silenced in MDA-MB-231 cells using an siRNA targeting the 3′-UTR region of YWHAH mRNA, and YWHAH or S25A mutant proteins were restored in the cells. After incubated with 5 mM fructose for 48 h, cells were subjected to Matrigel invasion assay. The numbers (means ± S.D. from four independent experiments) of invaded cells are shown as bar graphs. Significance was calculated by unpaired, two-sided Student’s t-test.

Fig 4: Molecular dynamics analysis to support the KHK-A phosphorylation of YWHAH.a Potential of mean force (PMF) profiles as a function of the distance (d) between the oxygen of S25-OH and the phosphorus of ATP-?PO3 for YWHAH/KHK-A/ATP, PRPS1-11/KHK-A/ATP, YWHAH/KHK-C/ATP, and YWHAH-11_D21A/KHK-A/ATP. Amino acid sequences of YWHAH-11 and PRPS1-11 are shown above panel. The conserved amino acids are highlighted and the phosphorylated residues are red. For data reproducibility, five different PMF profiles for each case were obtained from fully independent umbrella simulations, which are marked with five different colors. The error bars were obtained from bootstrap analysis with 200 bootstraps. b 3D structure of YWHAH-11 bound to KHK-A. YWHAH-11 and ATP are shown by sticks with exceptions for the hydroxyl oxygen group of S25 and the ?-phosphorous of ATP represented by enlarged spheres between which distance is denoted as d and KHK-A are represented by cartoons where aa. 100–114 are green. c The residue–residue contact map for ternary complex of YWHAH-11/KHK-A/ATP at d ? 3.5 nm (top) and of PRPS1-11/KHK-A/ATP at d ? 3.6 nm (bottom). The horizontal residues indicate aa. 3–298 of KHK-A and the right vertical residues indicate aa. 1–11 of YWHAH-11 or PRPS1-11. d The average number of atoms contacting the KHK-A 100–114 for YWHAH-11 residues at d ? 3.5 nm (top) and PRPS1-11 residues at d ? 3.6 nm (bottom). The contact is counted in case that the interatomic distance <5 nm. The averages were taken over n = 1000 configurations obtained from MD trajectories for 10 ns with 10 ps interval under the restraint of d = 3.5 nm (YWHAH-11) and d ? 3.6 nm (PRPS1-11). Data are presented as means ± S.D. e MDA-MB-231 cells were co-transfected with His-KHK-A and FLAG-YWHAH WT (or D21A) plasmids. The protein interaction was cross-checked by co-precipitation using anti-FLAG and Nickel-NTA affinity gels. The Ser/Thr-phosphorylation of YWHAH was identified by immunoblotting the immunoprecipitated YWHAH with anti-phospho-S/T antibody. f In vitro binding assay. Purified FLAG-YWHAH WT (or D21A) and KHK-A were co-incubated for 1 h. KHK-A was pulled down using Nickel-NTA and the KHK-A-bound YWHAH was immunoblotted.

Fig 5: Phosphorylated YWHAH promotes breast-cancer metastasis in fructose-fed mice.a Schematic diagram for the breast-cancer xenograft study. The MTV-TM-011 stable cell line co-expressing luciferase and YWHAH or YWHAH-S25A were implanted into the mammary pads of mice. Tumor-bearing mice were fed with water or 15% fructose. After primary tumors were removed, metastatic tumors were checked weekly by luminescence imaging. The condition of each experimental group (7 per each group) is described in the right panel. b Body weights of mice (the means ± S.D. from 7 mice per group). c The volumes of breast tumors (the means ± S.D. from 7 mice per group). *Denotes P < 0.05 between group ‘d’ versus ‘b’; # denotes P < 0.05 between group ‘d’ versus ‘c’; § denotes P < 0.05 between group ‘d’ versus ‘f’. d Bioluminescence images of tumor-bearing mice. e On the 5th week after cancer graft, bioluminescence images were captured in the excised lungs (left). Bioluminescence intensity (photons/s/cm2/sr) was quantitatively analyzed (right). Data are represented as means ± S.D. from the lungs from 7 mice per group. f Representative pictures of H&E-stained lungs. g In other organs, representative bioluminescence images (left) and quantitative analyses of bioluminescence emission (right). Dots represent the bioluminescence intensities from individual samples, and horizontal bars show the means ± S.D. from all vital organs obtained from 7 mice per group. In (b), (c), (e), a two-sided Mann–Whitney U test was used to calculate significance. P-values are represented in each panel.