TOPK interacts with Prx1 in vitro and ex vivo. A, tandem mass spectrum for the doubly charged ion from the peptide sequence LVQAFQFTDK (theoretical, monoisotopic neutral MW of 1195.62 Da), labeled with select experimental product ions, corresponding m/z values, and select ammonium ions (Im). The precursor mass error is 40 ppm. The experimentally determined diagnostic b- and y-type product ions with S/N >3 are shown on the peptide sequence above the spectrum (S/n = signal to noise). B, TOPK binding with Prx1 ex vivo. Lane 1: input, whole-cell lysate from RPMI7951 human melanoma cells; lane 2: negative control, whole-cell lysate from RPMI7951 cells precipitated with Ni-NTA-agarose beads; lane 3: whole-cell lysate from RPMI7951 cells precipitated with TOPK-Ni-NTA-agarose beads. TOPK and Prx1 binding was confirmed by immunoblotting with anti-Prx1. C and D, the endogenous TOPK-Prx1 complex was immunoprecipitated from RPMI7951 human melanoma cells with a TOPK (C) or a Prx1 (D) antibody, and TOPK or Prx1 were detected by Western blotting with a TOPK or Prx1 antibody. Precipitation with normal IgG served as a negative control. E, schematic diagrams for construction of Prx1 deletion mutants. N_del (D1–56), C_del (D128–199), or Δ_del (D57–127) were each introduced into the pET46 Ek/LIC His fusion vector. F, Prx1 (Wt) and deletion mutants (N_del, C_del, or Δ_del) purified from BL21 (bottom) were used for binding with GST-TOPK beads (top). Protein abundance were confirmed by Western blotting with an antibody against the His tag.

TOPK phosphorylates Prx1 (Ser-32). A, comparison of amino acid sequences of Prx1 and Prx3. Amino acids Ser-32 and Ser-126 in Prx1 correspond with Asp-32 and Ala-126 in Prx3, respectively. B, in vitro kinase assay using [γ-³²P]ATP to determine whether TOPK phosphorylates Wt or mutant (S126A or S32A) Prx1. Proteins were visualized by autoradiography. C, TOPK binds with phosphorylated Prx1. TOPK-phosphorylated or nonphosphorylated Prx1 (Wt) or Prx1 mutant (S32A) protein (in vitro kinase assay) were used for binding with GST-TOPK beads. TOPK and Prx1 binding was confirmed by immunoblotting with anti-Prx1. D and E, CD spectra in the far UV range were prepared to determine changes in the secondary structure of Prx1 induced by phosphorylation. D, Prx1-Wt (solid line, 1) or phosphorylated Prx1-Wt (dashed line, 2). E, Prx1-S32A (solid line) or phosphorylated Prx1-S32A (dashed line).

TOPK inhibits UVB-induced apoptosis in RPMI7951 melanoma cells, mediated through phosphorylation of Prx1 at Ser-32. A, analysis of apoptosis by flow cytometry in RPMI7951 melanoma cells; B, RPMI7951 melanoma cells transfected with control siRNA or TOPK siRNA. C, RPMI7951 melanoma cells stably expressing mock, Prx1 wild type (Wt), or Prx1 mutant (S32A) constructs. For A–C, cells were incubated with annexin V-conjugated FITC at 24 h after UVB (4 kJ/m²); data are shown as means ± S.D. from three independent experiments. The asterisk (*) indicates a significant increase (p < 0.0001) in UVB-induced apoptosis. D, RPMI7951 melanoma cells stably expressing mock, Prx1 Wt or Prx1 mutant (S32A) constructs were treated or not treated with UVB and harvested after 18 or 24 h. The level of cleaved caspase-3 was assessed by Western blot, and β-actin was used as an internal control to monitor equal protein loading.

TOPK phosphorylates Prx1 in vitro and UVB induces phosphorylation of TOPK in RPMI7951 melanoma cells. A, His-Prx1 and His-Prx3 fusion proteins were purified from BL21 bacteria using Ni-NTA-agarose beads and purity was confirmed by Coomassie Blue R-250 staining. B, in vitro kinase assay to determine whether TOPK phosphorylates Prx1 or Prx3. Results are visualized by autoradiography with [γ-³²P]ATP. C, different human melanoma cell lines were screened by Western blot to determine total protein abundance of endogenous TOPK, Prx1, or Prx2. β-Actin verified equal protein loading. D, time-dependent UVB-induced phosphorylation of TOPK (Thr-9). RPMI7951 cells were harvested at various times after UVB (4 kJ/m²) treatment. Phosphorylation of TOPK (Thr-9) or nonphosphorylated TOPK was detected. Data for C and D are representative of three independent experiments.

TOPK increases the peroxidase activity of Prx1 and suppresses H₂O₂ accumulation. A, in vitro peroxidase activity was determined as described under “Materials and Methods” using active TOPK and FPLC purified His-Prx1-Wt (Wt) or His-Prx1-S32A (S32A) fusion proteins. Peroxidase activities of nonphosphorylated Wt and S32A were used as controls. To convert the spectrophotometric data (A₅₇₀) to peroxidase activity (mU/ml), a peroxidase standard curve was prepared according to the manufacturer's instructions. B, peroxidase activity was determined as for (A), but using RPMI7951 cells (2 × 10⁴) transfected with control siRNA or TOPK siRNA. Samples were harvested at different times following UVB (4 kJ/m²) treatment. Data are shown as means ± S.D. of three samples from two independent experiments. The asterisk (*) indicates a significantly increased (p < 0.001) activity induced by UVB in control siRNA cells compared with untreated or UVB-treated TOPK siRNA cells. C, oxidation of Prx1 to Prx1-SO₃ was assessed by Western blotting using anti-Prx1-SO₃ (recognizes oxidized Cys-52 in Prx1) after UVB (4 kJ/m²) in RPMI7951 cells transfected with control siRNA or TOPK siRNA. Total TOPK, Prx1, or β-actin protein levels were used as internal controls to confirm TOPK deficiency and equal protein loading. D, effect of TOPK on H₂O₂ accumulation was measured as for (A) in mock, Prx1 Wt or Prx1 mutant (S32A) stably transfected cells. To convert the spectrophotometric data (A₅₇₀) to reflect the levels of H₂O₂ (μm), an H₂O₂ standard curve was prepared according to the manufacturer's instructions.