AIMP2 is critical for the control of cell death. (A) AIMP2^+/+ and AIMP2^−/− MEFs were irradiated by UV (50 J/m²) and measured for the presence of apoptotic cells by flow cytometry after 6 h. The bars and numbers indicate the subG1 phase cells and their percentages, respectively. (B) After transfection of AIMP2 or empty vector (EV) into AIMP2^−/− MEFs for 24 h, the cells were treated with UV, and the subG1 phase cells were measured as above. (C) AIMP2^+/+ and AIMP2^−/− MEFs were subjected to UV irradiation as above, and after 3 h, the generation of active caspase-3 from pro-caspase-3 was monitored by Western blotting with anti-pro-caspase-3, -active caspase-3, -AIMP2, and tubulin antibodies. (D) The effect of AIMP2 suppression with AIMP2 siRNA on UV-induced U2OS cell death was monitored by flow cytometry using FITC-annexin V and PI. Viable, early apoptotic, and late apoptotic cells are FITC⁻ PI⁻, FITC⁺ PI⁻, and FITC⁺ PI⁺, respectively. The numbers represent percentages of the corresponding cells. (E) U2OS cells were transfected with EV or AIMP2 and treated with UV (50J/m²). After 12 h, the effect of AIMP2 on UV-induced cell death was measured by colorimetric thiazoyl blue assay. The bars represent the mean ± SD. from triplicate samples.

UV-dependent phosphorylation, complex dissociation, and nuclear localization of AIMP2. (A) U2OS cells were UV irradiated (50 J/m²), and the cytoplasmic and nuclear fractions were separated by using a subcellular proteome extraction kit (Calbiochem). The proteins extracted from each fraction were immunoblotted with the antibodies against AIMP2, YY1 (nuclear marker), and Hsp70 (cytoplasmic marker). (B) U2OS cells were UV irradiated (50 J/m²). After 5 min, cells were harvested and proteins extracted from the untreated (Top) and UV-treated cells (Middle) were separated by 2D gel electrophoresis and subjected to Western blotting with an anti-AIMP2 antibody. To determine the UV-dependent phosphorylation of AIMP2, the proteins extracted from the UV-treated cells were reacted with +AP (Bottom). (C) Lysates from U2OS cells, irradiated with UV for 5 min, were immunoprecipitated with anti-AIMP2 antibody and immunoblotted against pSer, pThr, pTyr. (D) To determine the effect of UV irradiation on the dissociation of AIMP2, the proteins prepared as above were separated by gel filtration, and the eluted fractions from 23 to 40 were subjected to SDS/PAGE. AIMP2 and EPRS were detected by Western blotting with their respective antibodies. (E) The proteins in fractions 24 and 36 were precipitated with a 2D clean-up kit (GE Healthcare) and then subjected to 2D gel electrophoresis and Western blotting with anti-AIMP2 antibody. (F) U2OS cells were UV irradiated, harvested at 5 min, and the cytoplasmic and nuclear fractions were separated by using a subcellular proteome extraction kit (Calbiochem). The proteins extracted from both fractions were precipitated with a 2D clean-up kit (GE Healthcare), treated with AP, and then subjected to 2D gel electrophoresis. The migration of AIMP2 was determined by immunoblotting with an anti-AIMP2 antibody. (G) U2OS cell, treated or untreated with JNK inhibitor II (20 μM) or staurosporin (0.5 μM), were UV irradiated. The proteins extracted from the cells were subjected to gel filtration, SDS/PAGE, and Western blot analysis for EPRS and AIMP2 as above. (H) U2OS cells (control, 10 min after UV, 10 min after UV, and JNK inhibitor II treated) were stained with anti-AIMP2 antibody (green) and counterstained with PI (red) to monitor the cellular localization of AIMP2. Data are representative of three independent experiments.

Functional significance of AIMP2 activation of p53 in response to DNA damage. (A) The effect of AIMP2 on p53 levels was monitored by using Western blotting after transfection of AIMP2 into U2OS cells that were p14^Arf inactive, to exclude the possible involvement of p14^Arf in the induction of p53. (B) AIMP2 was transfected into U2OS cells at the indicated amounts and its effect on the expression of PUMA, a known target of p53, was monitored by RT-PCR. GAPDH was used as the control. (C) After cotransfection with the indicated amounts of AIMP2 and GADD45-luciferase (0.5 μg/ml), U2OS cells were incubated for 24 h and dissolved in lysis buffer (Promega). After collection by centrifugation, cells were lyzed, mixed with the luciferase reaction substrate, and the reaction was quantified by using a luminometer. (D) AIMP2^+/+ and AIMP2^−/− MEFs were irradiated by UV at 50 J/m²and the extracted proteins were subjected to immunoblotting with anti-p53, -AIMP2, and -tubulin antibodies. (E) AIMP2-deficient MEFs were transfected with EV or AIMP2. They were then treated with UV and checked whether p53 activation was restored by the exogenous introduction of AIMP2. (F) To determine whether AIMP2 is specifically required for the activation of p53, we treated AIMP2^+/+ and AIMP2^−/− MEFs with adriamycin (1 μg/ml, 4 h), TSA (100 ng/ml, 12 h), and EGF (10 ng/ml, 4 h), then compared the cellular levels of p53 by Western blotting. Actin was used as a loading control. Data are representative of three independent experiments.

Interaction of AIMP2 with p53. (A) The nuclear fractions were prepared from U2OS cells at the indicated times after UV irradiation and immunoprecipitated with anti-p53 antibody, and the coprecipitation of AIMP2 was determined by Western blotting. (B) p53 was expressed as a GST fusion protein and incubated with recombinant His-tagged AIMP2. GST or GST-p53 proteins were precipitated with glutathione-Sepharose, and the coprecipitation of AIMP2 was determined by Western blotting with an anti-AIMP2 antibody. (C) The interaction of p53 with AIMP2 and two other components of the multi-ARS complex, AIMP1 and KRS, were determined by yeast two-hybrid assay. p53 and the testing partners were expressed as LexA and B42 fusion proteins, respectively, and positive interactions were determined by cell growth on yeast minimal medium. (D) The different peptide regions of p53 were expressed as LexA fusion proteins, and their interactions with B42-AIMP2 were also tested as above. The numbers of each lane indicate the amino acid positions of p53. (E) The WT and different point mutants of p53 at positions 22/23, 175, and 248 (35–37) were radioactively synthesized by in vitro translation, mixed with GST-AIMP2, and affinity precipitated with glutathione-Sepharose beads. The p53 mutants coprecipitated with GST-AIMP2 were determined by autoradiography. (F) The radioactively labeled deletion fragments of AIMP2 were prepared by in vitro translation, mixed with GST-p53, and precipitated with glutathione-Sepharose beads. The AIMP2 fragments precipitated with GST-p53 were detected by autoradiography. Data are representative of three independent experiments.

AIMP2 blocks MDM2-mediated ubiquitination and degradation of p53. (A) The full-length and 1 to 83-aa peptide region of AIMP2 were transfected, and their effect on MDM2 binding to p53 was determined by coimmunoprecipitation (Left). Expression of MDM2, Myc-AIMP2 full-length and N-terminal 83-aa fragment, and p53 were determined by Western blotting of whole-cell lysates. Actin was used as a loading control (Right). (B) HA-tagged ubiquitin and Myc-AIMP2 were introduced into U2OS cells, which were then treated with ALLN (20 μM) for 3 h. The proteins were extracted and immunoprecipitated with anti-HA antibody, and the precipitated proteins were separated by SDS/PAGE for immunoblotting with an anti-p53 antibody (Upper Left). The ubiquitination of the total precipitated proteins was compared by immunoblotting with an anti-HA antibody (Lower Left). Expression of p53 and AIMP2 was determined by immunoblotting with anti-p53 and -Myc antibodies, respectively (Right). Actin was used as a loading control. (C) AIMP2^+/+ and AIMP2^−/− MEFs were treated with ALLN (20 μM) for 3 h. The culture medium was replaced with fresh medium, and CHX was added to a final concentration of 20 μg/ml. At the indicated times after the addition of CHX, the cell lysates were prepared and examined for p53 levels by Western blot analysis. (D) U2OS cells were cotransfected with the GADD45-luciferase reporter plasmid (0.2 μg) with the indicated amounts of MDM2, HDAC3, and AIMP2. The EV was used to adjust the total amount of DNA, and 0.2 μg of pCMV-galactosidase expression vector was used as an internal control. (E) AIMP2^+/+ and AIMP2^−/− MEFs were irradiated by UV (50 J/m²) or treated with nutlin-3 (5 μM) for 6 h. Cells were harvested, and the extracted proteins were subjected to immunoblotting with anti-p53, anti-AIMP2, and anti–tubulin antibodies. (F) Schematic representation of the nuclear translocation of AIMP2 and its interaction with p53 in response to genotoxic stresses. AIMP2 is normally harbored in the multi-ARS complex. Upon DNA damage, AIMP2 is phosphorylated via the JNK pathway, dissociated from the complex, translocated to the nucleus, and bound to p53, which then prevents the MDM2-mediated ubiquitination and degradation of p53.