Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 May 21;116(21):10382-10391.
doi: 10.1073/pnas.1804013116. Epub 2019 May 9.

Activation of PASK by mTORC1 Is Required for the Onset of the Terminal Differentiation Program

Affiliations
Free PMC article

Activation of PASK by mTORC1 Is Required for the Onset of the Terminal Differentiation Program

Chintan K Kikani et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

During skeletal muscle regeneration, muscle stem cells (MuSCs) respond to multiple signaling inputs that converge onto mammalian target of rapamycin complex 1 (mTORC1) signaling pathways. mTOR function is essential for establishment of the differentiation-committed progenitors (early stage of differentiation, marked by the induction of myogenin expression), myotube fusion, and, ultimately, hypertrophy (later stage of differentiation). While a major mTORC1 substrate, p70S6K, is required for myotube fusion and hypertrophy, an mTORC1 effector for the induction of myogenin expression remains unclear. Here, we identified Per-Arnt-Sim domain kinase (PASK) as a downstream phosphorylation target of mTORC1 in MuSCs during differentiation. We have recently shown that the PASK phosphorylates Wdr5 to stimulate MuSC differentiation by epigenetically activating the myogenin promoter. We show that phosphorylation of PASK by mTORC1 is required for the activation of myogenin transcription, exit from self-renewal, and induction of the myogenesis program. Our studies reveal that mTORC1-PASK signaling is required for the rise of myogenin-positive committed myoblasts (early stage of myogenesis), whereas mTORC1-S6K signaling is required for myoblast fusion (later stage of myogenesis). Thus, our discoveries allow molecular dissection of mTOR functions during different stages of the myogenesis program driven by two different substrates.

Keywords: PASK; Pax7; mTOR; muscle stem cell; myogenin.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
Nutrient and insulin signaling activates PASK via mTORC1. (A) PASK is activated during skeletal muscle regeneration. TA muscles were isolated from control or BaCl2-injured Rosa26hPASK-V5 mice 3 d postinjury (DPI). V5-tagged PASK was immunoprecipitated from tissue extract to assay PASK activation by an in vitro autophosphorylation assay, which indicates the incorporation of 32P into PASK as a function of kinase activity. An immunoblot (IB) of MyoG marks myogenic regeneration. IP, immunoprecipitation. (B) CHO-K1 cells expressing V5-tagged hPASK were stimulated with 100 nM insulin for the indicated times. PASK was immunoprecipitated using anti-V5 antibody, and an in vitro kinase assay was performed as in A. Activation of PI3K and mTORC1 signaling was demonstrated by the appearance of phospho-AKT and phospho-S6K. (C) HEK293E cells were starved of amino acids and glucose for 8 h, followed by stimulation with either 25 mM glucose or 800 μM l-leucine for 1 h. Endogenous PASK was purified using anti-PASK antibody from cell extracts, and in vitro kinase activity assay was performed as in A. (D) PASK from HEK293E cells was assayed as in C. Cells were stimulated with 100 nM insulin for 1 h after pretreatment with DMSO, 100 nM rapamycin, or 25 μM 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR). Endogenous PASK was purified using anti-PASK antibody from cell extracts, and an in vitro kinase activity assay was performed as in A. (E) Quantification of D. Phospho-PASK (32P-PASK) and total PASK from three independent experiments; the ratio was expressed as the fold change in PASK activity under the indicated stimuli. Error bars are ± SD. *P < 0.05; **P < 0.01. HG, high glucose; LG, low glucose. (F) V5-PASK was expressed in Tsc2−/− cells with or without complementation with WT hTsc2. Cells were serum-starved overnight and then stimulated with 100 nM insulin for 1 h. PASK was immunoprecipitated using anti-V5 antibody, and an in vitro kinase assay was performed. The yeast Ugp1 protein, which is robust in an in vitro substrate of PASK, was used as an exogenous substrate. (G) PASK was purified from HEK293T cells with expressing vector control (−) or RhebQ64L and subjected to kinase activity assay as in F. (H) HEK293E cells were transfected with control or mTOR-targeting siRNA for 24 h. A vector expressing V5-PASK was then transfected; after 24 h, cells were serum-starved overnight and then stimulated with 100 nM insulin for 1 h. PASK was purified and subjected to a kinase activity assay as in F. (I) Primary myoblasts isolated from Rosa26hPASK-V5 mice were transfected with control or mouse Tsc2-targeting siRNA. Twenty-four hours after transfection, cells were switched to 5% serum-containing medium overnight, followed by 4 h of total serum starvation. Cells were then treated with vehicle or 100 nM insulin for 1 h. PASK was then purified and subjected to kinase activity assay as in F. (J) Quantification of PASK kinase activity measurements from three experiments as in I. *P < 0.05; ***P < 0.005.
Fig. 2.
Fig. 2.
PASK is a direct phosphorylation target of mTORC1. (A) Endogenous mouse PASK was immunoprecipitated from Tsc2−/− or hTsc2-complemented Tsc2−/− MEFs after labeling with 32P-phosphate for 4 h. Anti-PASK immunoprecipitates were separated by SDS/PAGE and subjected to autoradiography and an immunoblot (IB). (B) HEK293T cells were transfected with vectors expressing V5-tagged WT or K1028R (KD) PASK or Myc-tagged PDK1, as well as either empty vector or a vector expressing RhebQ64L. At 24 h posttransfection, in-cell 32P labeling was conducted and the indicated immunoprecipitates were analyzed as in A. IP, immunoprecipitation. (C) Schematic indicating the domain structure of full-length PASK and the domain truncation mutants used in D. (D) WT PASK or the truncation mutants from C were coexpressed with empty vector or a vector expressing RhebQ64L and were treated with or without 40 nM rapamycin. They were assessed for in-cell phosphorylation as in A. (E) Schematic of the mTORC1-dependent phosphorylation sites on PASK. (F) WT or TT→AA (T640A T642A), S3A (S949A S953A S956A) or TS[5]A (T640, T642, S949, S953, S956 to Ala) mutants of PASK were expressed in HEK293T cells with or without expression of RhebQ64L and analyzed for cell phosphorylation as in A. (G) WT or the TS[5]A mutant of PASK was expressed in HEK293T cells with or without coexpression of RhebQ64L, and kinase activity was measured by autophosphorylation (32P-PASK) and Ugp1 phosphorylation as in Fig. 1F. (H) RhebQ64L-induced PASK in vivo phosphorylation was measured with or without 50 μM AT13148, 50 μM PF4708671, or 100 nM torin pretreatment in HEK293T cells. For Western blot analysis using indicated phosphospecific antibodies, an identical parallel experiment was performed to obtain nonradioactive cell extracts to analyze efficacy of the inhibitor treatment. (I) In vitro kinase assay was performed using purified mTORC1 and KD (K1028R mutant) PASK as described in Materials and Methods. The asterisk indicates the band corresponding to phosphorylated form of raptor in the kinase reaction mixture.
Fig. 3.
Fig. 3.
PASK associates with mTORC1 in a nutrient-sensitive manner. (A) Vector control or WT or KD (K1028R) PASK was coexpressed with either Myc-tagged WT or D2357E (KD) mTOR in HEK293T cells. Twenty-four hours after transfection, V5-tagged proteins were purified and the presence of Myc-tagged mTOR was detected by Western blotting. Immunoprecipitates were also probed with anti-AKT substrate antibody (RXRXXpS/T), as described in Materials and Methods, to detect PASK-T307 phosphorylation. IB, immunoblot; IP, immunoprecipitation. (B) mTOR protein was purified from HEK293T cells, and the presence of its associated proteins was detected in the immunoprecipitates by Western blotting. (C) mTOR silencing and V5-hPASK expression and IP were performed as described in Fig. 1H. The presence of mTOR and its complex members was detected by Western blotting of the immunoprecipitates. (D) Indicated V5-tagged PASK truncation mutants were expressed and immunoprecipitated from HEK293T cells. The co-IP of mTORC1 was determined by Western blotting of the immunoprecipitates. (E) Vector or V5-PASK was expressed with RhebQ64L as indicated. For amino acid stimulation, cells were starved of the amino acids l-leucine and l-arginine overnight. On the next day, 800 μM l-leucine and 100 μM l-arginine were added for 1 h. Cells were lysed, and V5-tagged PASK was purified from HEK293T cells. The relative abundance of mTORC1 was detected by Western blotting of the immunoprecipitates.
Fig. 4.
Fig. 4.
PASK and p70S6K are required for distinct phases of myogenesis downstream of mTORC1. (A) Indicated components of mTORC1 and mTORC2 were silenced in C2C12 myoblasts. Forty-eight hours after silencing, cells were stimulated to differentiate using 100 nM insulin. Myogenin protein expression was used as an indication of differentiation for each cell population. (B and C) C2C12 cells were pretreated with 100 nM rapamycin, 50 μM BioE-1197, or 40 μM PF4708671 (S6Ki) at day −1. Cells were allowed to attain confluency (24 h) and induced for differentiation at day 0 in the continued presence of inhibitors. Twenty-four hours following differentiation, at day +1, cells were fixed with 4% paraformaldehyde and the induction of the myogenesis program (by antimyogenin staining) and myoblast fusion (by anti-MHC staining) was quantified (in D). ***P < 0.005 (control vs. inhibitors). NS, not significant vs. control. (D and E) C2C12 cells were allowed to attain confluency and induced for differentiation in the absence of inhibitors. Twenty-four hours following differentiation, at day +1, cells were treated with 100 nM rapamycin, 50 μM BioE-1197, or 40 μM S6Ki in differentiation media. Cells were fixed at day 3, and induction of the myogenesis program (by antimyogenin staining) and myoblast fusion (by anti-MHC staining) was measured and quantified (in E). *P < 0.05; **P < 0.005 (control vs. inhibitors).
Fig. 5.
Fig. 5.
PASK phosphorylation at mTORC1 sites is required for efficient myogenesis. (A) PASK+/+ or PASK−/− MuSCs were isolated from TA muscles of mice with the respective genotype. Twenty-four hours after isolation, PASK−/− cells were infected with retroviruses expressing Flag-tagged WT, K1028R, or TS[5]A PASK. Forty-eight hours after infection, PASK+/+ and PASK−/− cells were stimulated to differentiate with 100 nM insulin. Protein extracts were prepared from all cells, and myogenesis was measured by Western blotting using the indicated antibodies. IB, immunoblot. (B and C) C2C12 myoblasts with CRISPR/Cas9-deleted PASK were infected with the retroviruses containing indicated cDNAs. Forty-eight hours after retroviral infection, C2C12 cells were induced to differentiate using 100 nM Insulin. Forty-eight hours after induction of differentiation, cells were fixed and stained with anti-MyoG antibody to determine MyoG induction efficiency (in C) as described in Fig. 4B. *P < 0.05 (between TS[5]A and vector or KD PASK, significantly better rescue); #P < 0.005 (between TS[5]A and WT hPASK, significantly worse rescue). (D) C2C12 myoblasts with CRISPR/Cas9-deleted PASK were infected with the retroviruses containing indicated cDNAs. Forty-eight hours after retroviral infection, C2C12 cells were induced to differentiate using 100 nM insulin. Myogenesis was determined by immunofluorescence microscopy using antibodies against MHC. (E) Fusion index was calculated as in Fig. 4B. *P < 0.05 between TS[5]A and vector or KD PASK; #P < 0.005 between TS[5]A and WT hPASK.
Fig. 6.
Fig. 6.
mTOR-PASK-Wdr5 signaling cascade regulates the myogenin expression. (A) V5-LacZ as a control or WT or KD PASK was coexpressed with Flag-tagged WT-Wdr5 in the presence or absence of RhebQ64L. Twenty-four hours after transfection, V5-tagged proteins were immunoprecipitated and the abundance of Flag-Wdr5 was determined by Western blotting. Activation status of PASK by RhebQ64L was measured by Western blotting of the immunoprecipitates with anti-AKT substrate antibody (Materials and Methods). IB, immunoblot; IP, immunoprecipitation. (B) Domain arrangement of hPASK indicating relative positions of mTORC1-dependent phosphorylation sites and the Wdr5-interacting region on PASK. (C) Alignment of a region of PASK encompassing the Wdr5 binding region and mTOR phosphorylation sites from different species. Conserved residues are marked by black boxes. (D) V5-tagged LacZ or WT or C924A/W926A PASK was coexpressed with Flag-Wdr5 in HEK293T cells. The association between Wdr5 and various PASK proteins was measured by probing immunoprecipitate using the indicated antibodies. (E) V5-tagged LacZ or WT or TS[5]A mutant PASK was coexpressed with Flag-Wdr5 in the presence or absence of RhebQ64L. Western blotting to detect relative enrichment of Flag-Wdr5 was performed as in A. (F) C2C12 myoblasts were infected with retroviruses expressing control or WT, S49A, or S49E Wdr5. Twenty-four hours after infection, cells were treated with DMSO or 40 nM rapamycin for 24 h, followed by induction of differentiation by 100 nM insulin in the presence or absence of rapamycin as indicated. Normalized levels of mRNA for Myog and Mylpf (Mhc) were determined using quantitative RT-PCR with 18s rRNA used as a normalizer. Blue bars indicate the extent of normal myogenesis in the absence of rapamycin inhibition for comparison. (G) Western blot analysis from an experiment as in F. (H and I) Immunofluorescence microscopic examination of myogenesis of WT and the indicated Wdr5 mutants in DMSO or rapamycin. The fusion index was calculated as described in Materials and Methods. (Scale bars, 40 μM.)
Fig. 7.
Fig. 7.
Partitioning of mTORC1 functions during myogenesis by PASK and p70S6K activation. mTORC1 controls both the early (establishment of committed myoblasts) and later (myonuclei fusion and remodeling) stages of myogenesis. Our findings show that distinct mTORC1 substrates are critical in these two stages. Stem cell-enriched PASK is a necessary downstream effector of mTORC1 function in MuSCs that is required for the transition from Pax7+ stem cells to MyoG+ committed progenitors. PASK expression declines once stable MyoG expression is induced and the nascent myogenesis program is underway. Subsequently, p70S6K1-driven translational up-regulation results in myotube maturation, hypertrophy, and the metabolic adaptation necessary for muscle function.

Similar articles

See all similar articles

Cited by 3 articles

MeSH terms

Substances

Feedback