Myogenic differentiation triggers PML nuclear body loss and DAXX relocalization to chromocentres

Abstract

The promyelocytic leukemia protein (PML) is expressed in most normal human tissues and forms nuclear bodies (NBs) that have roles in gene regulation and cellular processes such as DNA repair, cell cycle control, and cell fate decisions. Using murine C2C12 myoblasts, we demonstrate that activation of skeletal muscle differentiation results in loss of PML and PML NBs prior to myotube fusion. Myotube formation was associated with marked chromatin reorganization and the relocalization of DAXX from PML NBs to chromocentres. MyoD expression was sufficient to cause PML NB loss, and silencing of PML induced DAXX relocalization. Fusion of C2C12 cells using the reptilian reovirus p14 fusogenic protein failed to disrupt PML NBs yet still promoted DAXX redistribution and loss; whereas ectopic expression of PML in differentiated cells only partially restored PML NB formation and DAXX localization at NBs. Finally, we determined that the C-terminal SUMO-interacting motif of DAXX is required for its colocalization with ATRX in heterochromatin domains during myotube formation. These data support a model in which activation of myogenic differentiation results in PML NB loss, chromatin reorganization and DAXX relocalization, and provides a paradigm for understanding the consequence of PML loss in other cellular contexts, such as during cancer development and progression.

Figure 1

PML nuclear body loss is associated with C2C12 differentiation. (a) C2C12 myoblast cells were cultured in growth media (control) or differentiation media for 4 days (differentiated). Formaldehyde-fixed cells were immunostained for PML, actin was visualized with fluorophore-conjugated phalloidin and DNA was visualized with DAPI. Syncytial nuclei associated with myotubes are indicated with arrowheads. Scale bar=20 μm. (b) Quantification of PML NBs in C2C12 myoblasts and syncytial nuclei in myotubes. Data are presented as the mean±standard error, n=3, *P<0.01

Figure 2

DAXX localizes to PML NBs in myoblasts and chromocentres in myotubes. (a) C2C12 myoblasts and myotubes were immunostained for DAXX, PML, and myogenin as indicated. DNA was visualized with DAPI. Examples of colocalization between PML and DAXX are indicated with arrowheads (>). Examples of colocalization between DNA and DAXX are indicated with arrows (→). (b) C2C12 myoblasts and myotubes were immunostained for DAXX and HP1α. DNA was visualized with DAPI. Examples of colocalization between DNA, HP1α, and DAXX are indicated with arrows. (c) Quantitative RT-PCR for Pml, Pax7 (myoblast expressed), and Myogenin (myotube expressed) mRNA expression in C2C12 cells at day 0 (undifferentiated) and day 5 post differentiation. Data are presented as the mean±standard error,n=3, *P<0.05, **P<0.01. (d) Detection of PML and DAXX protein expression in C2C12 myoblasts and myotubes as assessed by western blot analysis. (e) C2C12 cells were infected with lentivirus encoding either control shRNA (shControl) or shRNA-targeting PML (shPML). Cells were immunostained for PML and DAXX as indicated. DNA was visualized with DAPI. Arrowheads (>) indicate colocalization between PML and DAXX. Arrows (→) indicate colocalization between DAXX and chromocentres. Asterisks indicate the location of nuclei. Scale bars=10 μm

Figure 3

Fusion is not sufficient for PML loss. C2C12 cells were transfected with a plasmid encoding the fusogenic FLAG-tagged RRV-p14 protein and prepared at 36 h post transfection for immunofluorescence microscopy for RRV-p14 (anti-FLAG), PML, DAXX or myosin heavy chain. DNA was visualized with DAPI. Scale bar =10 μm. (a) Examples of PML expression within syncytial nuclei is indicated with arrowheads. (b) Examples of syncytial nuclei with DAXX relocalization (arrowheads) or DAXX loss (asterisks) are indicate. (c) RRV-p14-transfected cells were assessed for expression of the differentiation marker myosin heavy chain. Syncytial (a) and individual (b) RRV-p14-positive cells did not express myosin heavy chain. Asterisks indicate untransfected cells that express myosin heavy chain

Figure 4

DAXX is retained at PML NBs under low-calcium conditions. (a) Immunofluorescence microscopy of C2C12 cells differentiated for 5 days under control (1.8 mM) or myotube-inhibiting low-calcium (0.05 mM) conditions. Under low-calcium conditions, PML NBs (green) were increased relative to control cells and a fraction of DAXX (red) was retained there. Scale bar=10 μm. (b) Quantification of the number of PML NBs and DAXX foci at PML NBs. Data are presented as the mean±standard error, n=3, *P<0.01. A minimum of 100 cells were analyzed for each treatment. (c) Western blot analysis of DAXX and PML protein levels following differentiation under the two conditions. B, myoblasts; T, myotubes

Figure 5

PML loss precedes myotube fusion. (a) Immunofluorescence and differential interference contrast (DiC) microscopy of C2C12 cells differentiated for 4 days. Arrows indicate individual unfused cells with loss of PML NBs (green). (b) Immunofluorescence microscopy of C2C12 cells differentiated for 3 days and immunostained for PML and myosin heavy chain (M-HC) as indicated. Arrow indicates an unfused but differentiated cell with PML NB loss. DNA was visualized with DAPI. Scale bars=10 μm

Figure 6

MyoD expression is sufficient for PML NB loss. (a) Undifferentiated C2C12 cells were immunostained for PML and MyoD as indicated. Arrows indicate cells with higher MyoD expression and lower PML NBs. (b) Quantification of the average number of PML NBs in C2C12 cells based on high (HI), medium (MED), or low (LOW) MyoD expression. n=4, **P<0.01, *P<0.05, n.s.=not significant. (c) C2C12 cells were transfected with plasmid expressing myc-tagged MyoD and prepared for immunofluorescence microscopy 48 h post transfection. DNA was visualized with DAPI. Scale bars=10 μm. (d) Quantification of the average number of PML NBs in C2C12 cells±expression of myc-tagged MyoD or transfection control plasmid (GFP). n=4, **P<0.01

Figure 7

DAXX can associate with PML NBs in myotubes in the presence of interferon, but not ectopically expressed PML cDNA. (a) Fully differentiated C2C12 myotubes were treated with the indicated concentration of interferon alpha for 24 h and then prepared for immunofluorescence microscopy. Arrowheads indicate association of DAXX with reformed PML NBs in interferon-treated cells. Scale bar=10 μm. (b) Quantification of the number of PML NBs and DAXX foci at PML NBs in C2C12 myotubes the presence or absence of interferon. Data are presented as the mean±standard error, n=3, *P<0.01. A minimum of 120 cells were analyzed for each treatment. (c) Western blot analysis of DAXX and PML protein levels in differentiated C2C12 cells treated with the indicated concentration of interferon. (d) Immunofluoresence microscopy of C2C12 myoblasts transfected with FLAG-PML and differentiated into myotubes according to standard conditions. In myoblasts, DAXX (red) is enriched in ectopic PML (FLAG, green) structures (arrowheads). The presence of PML-containing structures in differentiated myotubes failed to recruit DAXX. Under these conditions, DAXX remained at chromocentres. Scale bar=10 μm

Figure 8

DAXX associates with ATRX and SUMO at heterochromatin in myotubes and this interaction requires the DAXX SUMO-interacting motif (SIM). (a) Immunofluoresence microscopy of ATRX and DAXX localization in C2C12 myoblasts and myotubes. Arrowheads indicate colocalization between ATRX and DAXX. (b) Imunofluoresence microscopy of SUMO and DAXX localization in C2C12 myoblasts and myotubes. Arrowheads indicate colocalization between SUMO and DAXX. (c) C2C12 myoblasts were transfected with FLAG-tagged full-length DAXX or DAXX ΔSIM expression plasmids prior to differentiation into myotubes. Scale bars=10 μm