SUPPLEMENTARY DATA

Methods-Supplement

Cell culture, transfection, immunoprecipitation, and sumoylation assay

HEK293T, U2OS, MEF and MEF-PML-/- cells were maintained in DMEM supplemented with 10% fetal bovine serum, 100 IU penicillin, 100 g/ml streptomycin. Transient transfections were carried out using Lipofectamine 2000 (Invitrogen). Approximately 24 h after transfection, cells were harvested for immunoprecipitation assay. Briefly, the cells were harvested in 0.5 ml of the lysis buffer for immunoprecipitation, followed by western blotting as described previously (1). For sumoylation assays, cells were transfected with HA-PML-1, HA-PML-1 3S/A, Myc-SUMO-1, Axin, pSUPER-HIPK2 and control siRNA in different combinations as indicated, together with EGFP-C3 as internal control, followed by treatment with or without 2.5 g of doxorubicin for 12 h. After treatment, cells were lysed in RIPA buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% NP-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM-glycerophosphate, 1 mM Na3VO4, 1 μg/ml leupeptin) and sonicated, followed by immunoprecipitation with anti-Myc for SUMO-1 and western blotting with anti-HA for PML.

Cell apoptosis

MEF and MEF-PML-/- cells were grown on 60 mm tissue culture plate. At 50% confluence, transfection was performed with a total of 1.5g each of plasmids including pCMV5-AXIN or/and pcDNA6-PML-1 in different combinations as indicated, together with 0.5 g of green fluorescent protein (GFP)-expressing vector pEGFPC3 (Clontech) as internal control for transfection efficiency, followed by treatment with or without 2.5 M of Dox. After treatment cells were digested and divided into two parts,with one part of cells fixed with 70% ethanol, stained with 20g/ml of propidium iodide and determined for the percentage of apoptotic cells as judged by Sub-G1 population using a FACScan flow cytometer (EPICS XL; Beckman Coulter, Fullerton, CA), andthe other partmeasured for protein levels by western blot. Results are presented as means±s.d. of three independent experiments.

Immunofluorescent staining

U2OS cells growing on glass coverslips in 6-well plates were left untreated or treated with UV (80 J/m2) following culture for additional 8 h, and were fixed with 3.7% formaldehyde-PBS for 10 min at room temperature. Fixed cells were blocked by incubating with PBS, 0.02% Triton X-100, 3%BSA and 1 mM NaN3 for 30 min. Cells were then incubated with 2 μg/mlprimary antibodies or control IgG in blockingsolution for 30 min. Next, cells were washed three times with washing buffer (PBS, 0.02% TritonX-100, 1.5% BSA and 1 mM NaN3); incubated with 4 μg/mlof second antibodies conjugated with Texas Red (Santa Cruz) or Alexa Fluor 488 (Invitrogen) in blockingsolution for 30 min. Cells were then washed three times with washing buffer,followed by incubating with DAPI (1 μg/ml)in washing solution for 2 min. Specimens were mounted in 90% glycerol, sealed with nail polishand visualized under a confocal laser scanning microscope (TCS SP2; LEICA).

1. Rui Y, et al. (2004) Axin stimulates p53 functions by activation of HIPK2 kinase through multimeric complex formation.EMBO J 23: 4583-4594.

Legends-Supplement

Fig. 1AXIN interacts with PML. (a) The up-shifted bands of PML in the immunoprecipitation result of Figure 1a are sumoylated forms of PML. The same samples used in Figure 1a were probed with anti-SUMO-1 to indicate that the upper bands are indeed sumoylated PML species.(b) The up-shifted bands of PML in the immunoprecipitation result of Figure 1b are sumoylated forms of PML. The same samples as shown in Figure 1b were detected with anti-SUMO-1.(c) AXIN directly interacts with PML. GST-pulldown assays were performed by incubating the purified His-AXIN with GST-PML-1, followed by detection of AXIN with anti-His antibody and PML with anti-GST antibody.

Fig. 2 Determination of domains of AXIN and PML-1 for their interaction. (a) Determination of AXIN domain for PML-1 interaction. Myc-PML-1 was cotransfected with HA-tagged wildtype AXIN and various deletion mutants as outlined in different combinations into 293T cells. At 30 h posttransfection, cells were harvested for immunoprecipitation with mouse monoclonal anti-Myc antibody and the immunoprecipitates were analyzed by western blotting using rabbit anti-HA antibody followed by secondary anti-rabbit HRP-conjugated IgG.. (b) Determination of the region of PML-1for AXIN interaction. Myc-AXIN was cotransfected with HA-tagged wildtype PML-1and various deletion mutants as depicted in the upper schematic diagram in different combinations into 293T cells. At 30 h posttransfection, cells were harvested for immunoprecipitation and western blotting with antibodies indicated. (c)PML-1 and its isoform 2 have different affinities for AXIN. Myc-AXIN, HA-PML-1and HA-PML isoform 2 were transfected into 293T cells in different combinations as indicated. At 30 h posttransfection, cells were harvested for immunoprecipitation with anti-HA antibody for PML. The precipitates were detected with HA antibody for PML and Myc antibody for Axin.

Fig. 3Dominant negative PML mutants abrogate p53 Ser-46 phosphorylation induced by AXIN. The plasmids that express AXIN, HA-PML-1, HA-PML-1 3SM (HA-tagged mouse PML with their three sumo-modified sites K70, K165 and K454 mutated to arginine residues), and HA-PML-1 ΔPRB were transfected alone or in different combinations, together with Myc-p53 into H1299 cells. At 30 h posttransfection, p53 was immunoprecipitated, followed by western blotting to detect total p53 and phospho-Ser-46 p53.

Fig. 4 PML and AXIN are not mutually required for interaction with p53 transactivational activity; AXIN does not affect PML activation of RARE luciferase reporter.(a) U2OS cells transfected with control siRNA or pSUPER-AXIN were subjected to immunoprecipitation with mouse anti-PML antibody (ab50637), followed by detection of PML with rabbit antibody (ab72137) and p53 with rabbit FL393 in the immunoprecipitates. Total cell lysates were detected for AXIN, PML, p53 and -actin with antibodies indicated. (b) U2OS cells were transfected with control siRNA or pSUPER-PML. At 30h posttransfection, immunoprecipitation was performed with AXIN C2b antibody. Immunoprecipitates were then detected with C2b antibody for AXIN and DO-1 antibody for p53.(c)AXIN does not affect PML activation of RARE luciferase reporter. 293T cells were transfected with AXIN and PML-1 alone or in combination, together with 0.2 g of RARE luciferase reporter and 0.5 g of LacZ and EGFP-C3. At 24h posttransfection, luciferase activities were measured and normalized to the LacZ activity, and expression levels of PML-1 and Axin were detected by western blot as shown in the insets. Results are presented as means±s.d. of three independent experiments.

Fig. 5AXIN and PML play essential roles in genotoxin-induced apoptosis. (a) Knockout of PML reduces apoptosis rates induced by lethal dosage of Dox or by overexpressed AXIN in MEF cells. Both MEF wild type and MEF-PML-/- cells transfected with or without of Axin and GFP were treated with or without of Dox. 24 h posttreatment,Cells were divided into two parts, with one part determined for the percentage of apoptotic cells by using follow cytometry, and the other for measuring protein levels by western blot as shown in the insets. Results presented are means±s.d. of three independent experiments.. (b) PML is required for AXIN to induce apoptosis. PML-/- MEF cells were transfected with AXIN or together with PML. Some 24 h later, cells were treated with a lethal dose of Dox (2.5 M) for 12 h, and apoptosis rates were obtained as in (a).Results are presentedasmeans±s.d.of three independent assays.

Fig. 6 HIPK2 is involved in PML sumoylation.(a) PML-1 was cotransfected with control siRNA, pSUPER-HIPK2, SUMO-1 and Axin in different combinations as indicated, with each transfection containing EGFP-C3 as an internal control for transfectional efficiency. At 12 h posttransfection, cells were untreated or treated with 2.5 M Dox for additional 12 h. Cells were then harvested in RIPA buffer, followed by immunoprecipitation with anti-Myc for SUMO-1 and western blotting with anti-HA for PML. Total cell lysates were probed with other antibodies as indicated. (b) U2OS cells were cotransfected with HA-PML-1, HA-PML-13S/A (a PML mutant with its three HIPK2 phosphorylation sites Ser17, Ser, 45 and Ser 47 mutated to Ala), SUMO-1 and Axin in different combinations as indicated, together with EGFP-C3 as internal control, followed by treatment with Dox, immunoprecipitation,and western blotting as described in (a).

Fig. 7Scaffolding roles of AXIN and PML for p53-activating supramolecular complex. (a, b). AXIN, PML-1, p53,and HIPK2 were cotransfected into U2OS cells, and were subjected to immunoprecipitations separately with different antibodies corresponding to the individual proteins. The immunoprecipitates were then analyzed by the antibodies. Immunoprecipitation specificity was evidenced by a lack of the other proteins precipitated in cells that did not contain the target protein of the antibody used for IP (see the leftmost lanes on each panel of western blotting results after IP).

Fig. 8 UV stimulates colocalization of PML with p53 and HIPK2 on PML-NBs. (a) Lethal DNA damage leads to increasing colocalization of PML with p53 and HIPK2. Top, U2OS cells were untreated or treated with lethal dose of UV. At 8 h posttreatment cells were co-stained with rabbit anti-PML (ab72137) and Alexa Fluor 488 donkey anti-rabbit IgG for PML (green) and mouse anti-p53 (DO-1) and Texas Red goat anti-mouse IgG for p53 (red). Bottom, U2OS cells that stably express HA-HIPK2 were untreated or treated with lethal dose UV, followed by co-staining with rabbit anti-PML (ab72137) and Alexa Fluor 488 donkey anti-rabbit IgG for endogenous PML (green), and mouse anti-HA (12CA5) and Texas Red goat anti-mouse IgG for HA-HIPK2 (red). (b) Staining of U2OS cells with combination of control IgG. U2OS cells were stained with combination of rabbit IgG and mouse IgG as primary antibody and combination of Alexa Fluo 488 donkey anti-rabbit IgG and Texas red goat anti-mouse IgG as secondary antibody.

Fig. 9 PML-1 3SM, a PML mutant that is defective for sumoylation, fails to recruit AXIN onto PML-NBs. U2OS cells cotransfected with Myc-Axin and HA-PML-1 3SM were untreated or treated with 80 J/m2 of UV irradiation. At 8 h posttreatment, cells were fixed and stained with mouse anti-HA antibody (12CA5) and Alexa Fluo 488 donkey anti-mouse IgG for PML (green) and rabbit anti-Myc antibody and Texas Red goat anti-rabbit IgG for AXIN (red)

1