TNF/p38α/polycomb signaling to Pax7 locus in satellite cells links inflammation to the epigenetic control of muscle regeneration

Abstract

How regeneration cues are converted into the epigenetic information that controls gene expression in adult stem cells is currently unknown. We identified an inflammation-activated signaling in muscle stem (satellite) cells, by which the polycomb repressive complex 2 (PRC2) represses Pax7 expression during muscle regeneration. TNF-activated p38α kinase promotes the interaction between YY1 and PRC2, via threonine 372 phosphorylation of EZH2, the enzymatic subunit of the complex, leading to the formation of repressive chromatin on Pax7 promoter. TNF-α antibodies stimulate satellite cell proliferation in regenerating muscles of dystrophic or normal mice. Genetic knockdown or pharmacological inhibition of the enzymatic components of the p38/PRC2 signaling--p38α and EZH2--invariably promote Pax7 expression and expansion of satellite cells that retain their differentiation potential upon signaling resumption. Genetic knockdown of Pax7 impaired satellite cell proliferation in response to p38 inhibition, thereby establishing the biological link between p38/PRC2 signaling to Pax7 and satellite cell decision to proliferate or differentiate.

Fig. 1. In vivo treatment of mdx mice with anti TNF and ex-vivo blockade of the p38 pathway expand a population of activated muscle satellite cells and increase Pax7 expression

Top: Schematic representation of the experimental design. A) Immunofluorescence staining using antibodies against laminin (green), Pax 7 (red) and counterstained for DAPI (blue), on transversal quadriceps sections from six weeks old mdx mice treated for five days with 20mg/kg of control (upper panels) or anti TNF antibodies (lower panels), B) The reported data represent the number of Pax7 positive cells per cent fibres in the same conditions as above. For each quadriceps over four hundred fibres were counted and the graph represents the average of three mice per experimental group. Error bars indicate the standard deviation (** Student test: p<0,01). C) Levels of activated p38 in muscles from control and anti TNF treated mice were measured by western blot using an antibody that recognises the phosphorylated form of p38. Total p38 was used as a loading control. D) Real time RT-PCR analysis of the expression of Pax7, muscle creatine kinase (MCK) and MyoD in satellite cells derived from control and treated mice. Error bars indicate the standard deviation (** Student test: p<0,01). E) Co-immunostaining of myofibers isolated from the gastrocnemius of C57/Bl6 wild type mice using antibodies against MyoD (green), Pax 7 (red) and counterstained for DAPI (blue). The single myofibers were maintained in culture for 72 h either in growth medium alone (GM) or in the presence of the p38 specific inhibitor (SB). F) The graph represents the quantification of the Pax7+/MyoD− and Pax7+/MyoD+ and Pax7−/MyoD+ cells per cluster, in the experimental conditions shown in E. Data are represented as the mean percentage of positive cells per clone. 60 clones from 2 different experiments were analyzed for each experimental point for a total 200 cell counted for each experimental point. P-value<0,01 for differences between Pax7+/MyoD− and Pax7−/MyoD+ cells in GM vs GM/SB. (See also Suppl. Fig. 1 and 2)

Fig. 2. p38 signalling represses Pax7 expression and proliferation of satellite cells

A) Immunofluorescence staining against Pax7 (red) and MyoD (green) on myofiber-derived satellite cells that were allowed to delaminate in culture for three days in GM1 (10%HS+0,5%CEE) and then induced to proliferate in GM2 (20%FBS+10%HS+1%CEE) in the presence or absence of SB. DAPI (blue) was used to counter stain nuclei. B) Quantification of the mean percentage of Pax7+/MyoD+ and Pax7−/MyoD+ nuclei in myoblasts and myotubes in the experiments reported in A. C) The expression of Pax7, Pax3 and Cyclin A2 in satellite cells cultured in the presence or absence of the p38 inhibitor SB were analyzed by real time RT-PCR. D) Pax7 protein levels (upper panel) on satellite cells treated as in C were measured by western blot. Tubulin (lower panel) was used as loading control. E) Satellite cells cultured in the same conditions as above were tested for their differentiation and proliferation potential. After a BrdU pulse (4 h), cells were harvested and stained using antibodies against MyHC (green) and BrdU (red). Nuclei were counterstained with DAPI (blue). F) Pax7, Pax3, Cyclin A2 and MCK RNA levels were measured by real time RT-PCR in satellite cells cultured as described in E. G) Protein levels of Pax7, Cyclin A2 and Myogenin were quantified by Western Blot. Tubulin was used as a loading control. H) Co-immunostaining using antibodies against Pax7 (red) and MyoD (green) on satellite cells incubated in the presence (lower panel) or absence (upper panel) of SB and induced to differentiate after drug withdraw. Nuclei were counterstained with DAPI (blue). (See also Suppl. Fig. 3 and 7)

Fig. 3. MKK6EE-dependent interactions between Ezh2, YY1 and p38 alpha on the chromatin of Pax7 promoter

Top: Schematic representation of the PRE containing region and the proximal promoter of Pax7 gene. A) Chromatin immunoprecipitation analysis (ChIP) of the Pax7 PRE and promoter, MHCIIb promoter and IgH enhancer were performed in C2C12 myoblasts cultured in growth medium and infected with control (C) or adeno-MKK6EE in the absence (MK) or presence (MK/SB) of the p38 inhibitor SB. ChIP was performed with antibodies against p38, Ezh2 YY1 and control IgG. Graph shows real time PCR values normalised against the input DNA. Error bars show standard deviation from three independent experiments. p values showing statistical significance by the student t-test between control and MKK6EE are indicated (* indicates p<0,05). B) Co-immunoprecipitation from nuclear extracts of C2C12 cells infected with control (C) or adeno-MKK6 and cultured in the absence (MK) or presence (MK/SB) of the p38 inhibitor SB. Nuclear extracts were immunoprecipitated with an anti-p38, anti-Ezh2 and anti-YY1 antibodies or control IgG and bound proteins were revealed by western blot using antibodies against Ezh2, p38, YY1 and HA. The images are representative of 3 independent experiments reproducing the same result.

Fig. 4. p38 phosphorylation of Ezh2 on Threonine 372 is necessary for Ezh2 and YY1 interaction and for Pax7 repression in response to MKK6EE

A) In vitro kinase assay using exogenously expressed flag-tagged Ezh2 that was immunoprecipitated from HEK 293 cells and then incubated with recombinant active p38 alpha (upper panel), in the absence or presence of the p38 alpha/beta inhibitor SB or the Pi3K inhibitor LY. Middle and lower panel show control western blots for the expression of Flag-Ezh2 and p38. B) In vitro kinase assay using deletion constructs of Ezh2 were performed as in A (left panel). Western blot using antibodies against the flag tag is shown as a loading control (right panel). C) In vitro kinase assay with a point mutant of Ezh2 in which threonine 372 was replaced with an alanine was performed as in A and B (left panel). Western blot using antibodies against the flag tag is shown as a loading control (right panel). D) Co-immunoprecipitation from nuclear extracts of C2C12 cells stably expressing control (pBabe) or myc-tagged Ezh2 wt and Ezh2T372A mutant. Nuclear extracts were immunoprecipitaed with anti-myc and bound proteins were revealed by western blot using antibodies against myc tag, Suz12 and EED (left panel). Input nuclear extracts before immunoprecipitation are shown in the left panel (E) Co-immunoprecipitation from nuclear extracts of the same cells as in (D), infected with control (C) or adeno-MKK6 and cultured in the absence (MK) or presence (MK/SB) of the p38 inhibitor SB. Nuclear extracts were immunoprecipitated with an anti-myc and bound proteins were revealed by western blot using antibodies against myc tag and YY1. (Left panel) Western blot of input nuclear extracts before immunoprecipitation showing Pax7, p38, YY1 and myc-ezh2 expression levels (right panel). Arrow indicates super-shifted (active, phosphorylated) endogenous p38 alpha. (See also Suppl. Fig. 4)

Fig. 5. p38-dependent modulation of PRC2-mediated histone modifications at the Pax7 promoter in satellite cells

Top: Schematic representation of the PRE containing region (−3140/−3060), the proximal promoter (−830−730) and distal elements (−11840/−11700) of the Pax7 gene. A) ChIP analysis of the PRE containing region, the proximal promoter and −12 kb region of Pax7 gene was performed using antibodies against AcH3, H3K27me3, H3K4me3 and control IgG in satellite myoblasts (MB) cultured for 5 days in growth medium in the presence (GM/SB) or absence (GM) of the p38 alpha/beta inhibitor SB. Note that SB was replaced every 24 hours in the GM for the 5 days of culture. The same analysis was performed in satellite cell-derived myotubes (MT), after incubation in differentiation medium (DM). Graph shows real time PCR values normalised against the input DNA. Error bars indicate the standard deviation (Student test: * indicates p<0,05; ** p<0,01).

Fig. 6. Stage-specific upregulation of Pax7 by EzH2 knockdown in satellite cells

A) Real time RT-PCR analysis of the expression of Ezh2 (blue line), Pax7 (pink line), myogenin (yellow line) and MCK (turquoise line) transcripts in myofibre-derived satellite cells, during 4 days of culture in GM and further incubation in DM (Fig. 5A). B) satellite cells derived from wild type C57/BL6 mice were transfected with control scramble (scr) or EzH2 siRNA at day three and harvested at day 4, and the expression levels of Ezh2, Pax7, MCK, myogenin and MyoD RNA were detected by RT-PCR. Error bars show standard deviation from two independent experiments. (See also Suppl. Fig. 5)

Fig. 7. Pharmacological blockade of PRC2 increases Pax7 expression in satellite cells both in vivo and ex vivo

A) Schematic representation of the experimental design. Briefly, five weeks old mdx mice were treated intra-peritoneally with dZNep (1 mg/kg). After five days mice were sacrificed and satellite cells were isolated and cultured in growing conditions (GM) for seven days. B) Protein levels of Ezh2 in satellite cells derived from control or dZNep treated mdx mice were evaluated by western blot. Tubulin is shown as a loading control C) Real time RT-PCR analysis of Pax7, MCK and cyclinA2 on satellite cells obtained as in A. Error bars show standard deviation from four independent experiments. p values showing statistical significance by the studend t-test are indicated (* indicates p<0,05). D) Satellite cells derived from wild type C57/BL6 mice were isolated and incubated in growth medium in the absence (C) or presence of SB and the Ezh2 inhibitors MC1947 (47) and MC1948 (48) and the expression of Pax7 and MyoD was detected by co-immunostaining. Nuclei were counterstained with DAPI. E) Western blot showing the levels of Pax7 in the same conditions as in D. Gapdh was used as a loading control. F) Real time RT-PCR analysis of Pax7, cyclinA2, MCK and Ezh2 on satellite cells treated as in D. (See also Suppl. Fig. 6).