Skeletal myosin light chain kinase regulates skeletal myogenesis by phosphorylation of MEF2C

Abstract

The MEF2 factors regulate transcription during cardiac and skeletal myogenesis. MEF2 factors establish skeletal muscle commitment by amplifying and synergizing with MyoD. While phosphorylation is known to regulate MEF2 function, lineage-specific regulation is unknown. Here, we show that phosphorylation of MEF2C on T(80) by skeletal myosin light chain kinase (skMLCK) enhances skeletal and not cardiac myogenesis. A phosphorylation-deficient MEF2C mutant (MEFT80A) enhanced cardiac, but not skeletal myogenesis in P19 stem cells. Further, MEFT80A was deficient in recruitment of p300 to skeletal but not cardiac muscle promoters. In gain-of-function studies, skMLCK upregulated myogenic regulatory factor (MRF) expression, leading to enhanced skeletal myogenesis in P19 cells and more efficient myogenic conversion. In loss-of-function studies, MLCK was essential for efficient MRF expression and subsequent myogenesis in embryonic stem (ES) and P19 cells as well as for proper activation of quiescent satellite cells. Thus, skMLCK regulates MRF expression by controlling the MEF2C-dependent recruitment of histone acetyltransferases to skeletal muscle promoters. This work identifies the first kinase that regulates MyoD and Myf5 expression in ES or satellite cells.

Figure 1

skMLCK is upregulated during skeletal myogenesis in P19 cells. (A) Q–PCR was performed for the indicated genes in a time course of P19 cell differentiation with DMSO. Results were normalized to β-actin, and expressed relative to day 0. Data are shown as mean±s.e.m. (n=3). (B) A schematic outline of the cardiac and skeletal muscle differentiation programmes that occur simultaneously during P19 cell differentiation.

Figure 2

skMLCK physically interacts with and phosphorylates MEF2C in vivo and in vitro. (A) The in vivo interaction between MEF2C and skMLCK was observed in C2C12 myoblasts that were differentiated under serum starvation conditions. Co-immunoprecipitation (IP) was performed using anti-skMLCK antibodies conjugated to magnetic beads followed by western blot with antibodies against MEF2C. Anti-IgG antibodies were used in a control IP. Intervening lanes have been removed for clarity, marked by a black line. (B) Flag-tagged MEF2C and His-tagged skMLCK or its mutants were co-transfected into HEK-293 cells. Co-immunoprecipitation (IP) using anti-Flag-agarose resin was followed by western blot analysis (WB) with anti-His antibodies. WB for His-, Flag-, and β-actin show expression prior to IP. Intervening lanes have been removed for clarity, marked by a black line. (C) In vitro kinase assays were performed with recombinant His-MEF2C incubated with purified skMLCK as indicated and visualized by silver stain or autoradiography. (D) In vivo kinase assays were performed in HEK-293 cells co-transfected as indicated. After immunopurification with an anti-Flag resin, western blot analysis and autoradiography were performed. An Image J program was used to measure band intensities and the intensity of each ³²P-radiolabelled band was normalized to the corresponding level of MEF2C protein. The data are shown as the normalized average ³²P intensity±s.e.m. (n=3). (E) The MEF-domain protein sequence alignment from different species, showing conservation of T⁸⁰. ^*P<0.05.

Figure 3

MEFT80D, but not MEFT80A enhances skeletal myogenesis. (A) P19[Control], P19[MEF2C], P19[MEFT80A], and P19[MEFT80D] stable cell lines were differentiated and examined by immunofluorescence with an anti-MHC antibody, MF20, to detect skeletal muscle, and counter stained with Hoechst dye to visualize the nuclei (× 400). (B) Skeletal myocytes and total nuclei for 10 random fields from different clones were counted and shown as a percentage of total cells, ±s.e.m. (n=3). (C) Western blot analysis of total protein extracts with anti-flag antibodies, showed similar levels of exogenous wild-type and mutant MEF2C protein. (D) Q–PCR analysis from RNA harvested on day 6 or 9 was performed with the genes indicated. Results were normalized to β-actin, and expressed as normalized fold-change relative to P19[Control] cells (n=15). ^*P<0.05.

Figure 4

skMLCK is necessary and sufficient for efficient myogenesis. (A) Immunofluorescence with MF20 antibodies, showing enhanced myogenesis in day 9 differentiated P19[skMLCK] cells compared with P19[Control] cells. Cells were counter stained with Hoechst dye (× 400). (B) MHC^+ve skeletal myocytes and total nuclei were counted and the percentage of total cells was calculated (n=3). (C) Q–PCR analysis of RNA harvested on days 6 and 9 of differentiation from P19[Control] and P19[skMLCK] cells (n=10). (D) Skeletal myogenesis was inhibited in mouse ES cells differentiated in the presence of ML-7 for 15 days. Q–PCR analysis was performed and the results were normalized to GAPDH, with levels expressed as normalized fold-change over undifferentiated cells and as a percentage of the control untreated cells (n=3). Immunofluorescence was performed with MF20 antibodies and stained with Hoechst dye (× 400). ^*P<0.05.

Figure 5

Inhibition of skMLCK reduced the activation of adult satellite cells. Isolated mouse satellite cells were cultured in the presence or absence of increasing concentrations of ML-7, SB203580, separately or together, as indicated. On day 3, RNA was harvested and Q–PCR analysis was performed. Results were normalized to β-actin, and expressed as fold-change relative to non-treated cells (n=5; ^*P<0.05).

Figure 6

MEFT80A cannot synergize with MyoD on endogenous skeletal muscle-specific promoters. (A) Myogenic conversion assays were performed in C3H10T1/2 fibroblasts, transiently transfected with plasmids as indicated, with or without ML-7 treatment. Q–PCR analysis of cardiac α-actin and myogenin transcript levels were normalized to transfected GFP transcripts and expressed relative to cells transfected with MyoD alone (n=3). (B) Western blot analysis with anti-flag antibodies, showing similar protein expression of wild-type and mutant MEF2C in transfected C3H10T1/2 cells. (C) A reporter assay with the muscle reporter MCK-luciferase shows similar levels of synergy for MEF2C or its mutants with MyoD on an exogenous promoter. Luciferase activity was measured and normalized against Renilla (n=4). ^*P<0.05.

Figure 7

MEFT80A cannot recruit p300/PCAF to endogenous skeletal muscle-specific promoters. (A) ChIP was performed using the indicated antibodies and analysed by Q–PCR with primers flanking the MEF2C site in the myogenin or GATA-4 promoters. Graphs represent Q–PCR analysis from day 7 of differentiation for P19, P19[MEF2C], or P19[MEFT80A] cultures. Relative enrichment was calculated as the percent chromatin input normalized to IgG (n=4). (B) HEK-293 cells were transfected with Flag-MEF2C, -MEFT80A, or -MEFT80D. Co-immunoprecipitation (IP) using anti-Flag-agarose resin was followed by western blot analysis (WB) with antibodies against endogenous p300. Western blots were prepared from total extracts, reacted with antibodies as indicated, and quantified using the Image J program. The p300 Co-IP band intensities were normalized to the intensity of their corresponding control β-actin bands and then to total p300 for each sample (n=3). ^*P<0.05.

Figure 8

Working model for the mechanism by which skMLCK can enhance myogenesis. (A) Previous studies have shown that MEF2C is inhibited by class II HDACs (HDAC CII), synergizes with MRFs, and recruits p300 to promoters (Sartorelli et al, 1997; Lu et al, 2000; Potthoff and Olson, 2007). Here, we show that skMLCK directly phosphorylates MEF2C, leading to p300/PCAF recruitment, increased acetylation of skeletal muscle-specific genes, and enhanced skeletal myogenesis. (B) Comparison of the stages of myogenesis in embryonic stem cells and satellite cells. ML-7 reduced the efficient upregulation of MRFs in the ES-derived premyogenic mesoderm or during the activation of quiescent satellite cells. In agreement with other studies (Palacios et al, 2010), SB inhibited the loss of Pax7, required for myoblast formation and subsequent differentiation.