Abstract
Free full text
Constitutive Notch Activation Upregulates Pax7 and Promotes the Self-Renewal of Skeletal Muscle Satellite Cells
a,dAbstract
Notch signaling is a conserved cell fate regulator during development and postnatal tissue regeneration. Using skeletal muscle satellite cells as a model and through myogenic cell lineage-specific NICDOE (overexpression of constitutively activated Notch 1 intracellular domain), here we investigate how Notch signaling regulates the cell fate choice of muscle stem cells. We show that in addition to inhibiting MyoD and myogenic differentiation, NICDOE upregulates Pax7 and promotes the self-renewal of satellite cell-derived primary myoblasts in culture. Using MyoD−/− myoblasts, we further show that NICDOE upregulates Pax7 independently of MyoD inhibition. In striking contrast to previous observations, NICDOE also inhibits S-phase entry and Ki67 expression and thus reduces the proliferation of primary myoblasts. Overexpression of canonical Notch target genes mimics the inhibitory effects of NICDOE on MyoD and Ki67 but not the stimulatory effect on Pax7. Instead, NICD regulates Pax7 through interaction with RBP-Jκ, which binds to two consensus sites upstream of the Pax7 gene. Importantly, satellite cell-specific NICDOE results in impaired regeneration of skeletal muscles along with increased Pax7+ mononuclear cells. Our results establish a role of Notch signaling in actively promoting the self-renewal of muscle stem cells through direct regulation of Pax7.
INTRODUCTION
Skeletal muscles can efficiently regenerate after injury. This remarkable regenerative capacity requires the contribution of a population of muscle resident stem cells called satellite cells (27, 29, 42). Pax7 expression is crucial for the embryonic specification, survival, and maintenance of the satellite cell lineage (22, 26, 35, 39, 46). In mature resting muscles, satellite cells are predominantly quiescent. Upon muscle injury, satellite cells are activated, enter the cell cycle, and proliferate. The proliferating satellite cells, called myoblasts, subsequently withdraw from the cell cycle to either differentiate or self-renew. The quiescent, proliferating, self-renewing, and differentiating satellite cells can be readily distinguished by the expression of several transcription factors, i.e., Pax7+, Pax7+ MyoD+, Pax7+ MyoD−, and Pax7− MyoD+ myogenin+, respectively (16, 32, 55). However, the signaling pathways that actively maintain/suppress satellite cells in these different states are not well understood.
The Notch signaling pathway plays critical roles in cell-cell communication and cell fate regulation during tissue morphogenesis, homeostasis, and regeneration (20). In vertebrates, Notch signaling is initiated by extracellular ligands (Dll1, -3, and -4 and Jag1 and -2) that interact with Notch1 to -4 receptors, leading to sequential cleavage of Notch extracellular and intracellular domains. Once cleaved, the Notch intracellular domain (NICD) translocates to the nucleus, where it interacts with RBP-Jκ (also called CBF1) and activates the transcription of specific target genes, including Hes and Hey family genes. During embryonic muscle development (myogenesis), oscillatory Notch activity defines the anterior boundary of somites and progenitor cell fate within the somites (17, 18, 38). Notch signaling further inhibits the myogenic differentiation of embryonic progenitors (12). Disruption of Notch signaling through Dll1 or RBP-Jκ mutation leads to premature differentiation of myogenic precursors and depletion of satellite cells (44, 50).
Notch signaling is also critical for postnatal myogenesis (muscle regeneration). Notch signaling is activated upon muscle injury, and reduced Notch activation has been associated with defective muscle regeneration in aged muscles (8, 9). Sarcopenia, a common aging-associated muscle wasting condition, is thought to be at least partially due to reduced satellite cell activity. Intriguingly, activation of the Notch signaling pathway by intramuscular injection of a Notch1 antibody rescued the regenerative defects of aged muscles, whereas blockage of Notch signaling via a diffusible ligand for Notch receptors led to defective regeneration of young muscles (8). Furthermore, exposure of aged muscle to a younger systemic environment by parabiotic pairing restored the ability of aged muscles to activate Notch signaling and to efficiently regenerate (9). As both focal and systemic manipulation of Notch signaling can in principle target all cell types in muscle, including satellite cells, myofibers, nerve- and blood vessel-associated cells, as well as other interstitial cells, whether satellite cell-specific activation of Notch signaling enhances the regeneration of injured muscles remains unclear.
Notch signaling has been shown to play various roles in the quiescence/activation, proliferation, and differentiation of satellite cells and myogenic C2C12 cells. Notch signaling regulates the asymmetric cell fate segregation in newly divided satellite daughter cells (10, 23, 47). The committed daughter cell expresses elevated Dll1, which presumably initiates Notch signaling in the neighboring sister cell that undergoes self-renewal (23). Likewise, Numb, a Notch inhibitor, is specifically expressed in the daughter cell that differentiates (10). Consistently, activation of Notch signaling or overexpression of Notch target genes in C2C12 myoblasts markedly inhibits MyoD, Myogenin, and Mef2c expression and myogenic differentiation (4, 21, 24, 25, 54). Other studies have indicated that Notch signaling promotes satellite cell activation and proliferation (8, 10). Satellite cells isolated from aged muscles exhibited delayed activation and reduced proliferation kinetics due to deficient upregulation of Notch signaling (8). In contrast, activation of Notch signaling in cultured myoblasts enhanced their proliferation (8, 9). However, two recent studies demonstrated that Notch signaling is necessary for maintaining satellite cell quiescence, as conditional knockout of RBP-Jκ in satellite cells resulted in spontaneous activation and differentiation of satellite cells (2, 28). One way to address some of the conflicting observations awaits techniques to constitutively activate Notch signaling specifically in satellite cells.
Notch's effects on satellite cells can be modulated by both micro- and systemic environments. For example, matrix heparan sulfate proteoglycans Syndecan 3 and Syndecan 4 regulate the self-renewal, proliferation, and differentiation of satellite cells through modulating Notch signaling (11, 13, 48). Loss of Syndecan 3 leads to reduced expression of Notch target genes due to defective ADAM17/TACE-mediated S2 cleavage and γ-secretase-mediated S3 cleavage in Notch processing (37). More recently, it has been shown that Dll1 expressed by neural crest progenitors migrating en route through embryonic somites transiently activates Notch signaling in myogenic progenitors and promotes myogenic differentiation (40). Notch signaling may also be modulated by environmental oxygen levels, as HIF1a functionally interacts with NICD (15). Lastly, systemic WNT and transforming growth factor β (TGF-β) signaling and also tumor necrosis factor alpha released by injured myofibers have been shown to antagonize Notch signaling in myogenic satellite cells (1, 3, 6).
To date, whether Notch signaling actively regulates the self-renewal (rather than doing so by passively repressing differentiation) of satellite cells has not been known. In the present study, we address this question through genetic gain-of-function studies. We show that constitutive activation of Notch signaling in myogenic progenitors results in decreased proliferation and differentiation but increased self-renewing of progenitor cells. Consistently, de novo activation of Notch signaling upregulated Pax7 expression in culture and inhibited muscle regeneration in vivo. Target gene profiling and functional analysis suggested that canonical Notch targets mediate the functional output of Notch activation on proliferation and differentiation, and NICD directly regulates Pax7 to enhance self-renewal. These results provide mechanistic insights into how Notch signaling regulates skeletal muscle development and homeostasis.
MATERIALS AND METHODS
Mice and animal care.
ROSA26-N1ICD mice were obtained from Jackson Laboratories (stock number 008159; Bar Harbor, ME) (30). Pax7-CreER mice were constructed by insertion of an internal ribosome entry site-CreER fragment at the 3′ end of the endogenous Pax7 gene (31). Mouse maintenance and experimental use were performed according to protocols approved by the Purdue Animal Care and Use Committee.
RNA isolation and quantitative pCR (qPCR) analysis.
Tissue samples for RNA were excised after euthanasia and either stored in RNAlater (Ambion, Woodlands, TX) for later processing or homogenized immediately for RNA purification by using the Qiagen RNeasy fibrous tissue minikit (Qiagen, Inc., Valencia, CA). Myoblasts were lysed in RNAlater for RNA purification using the Qiagen RNeasy minikit (Qiagen, Inc., Valencia, CA). An on-column DNase I digestion set (Sigma-Aldrich, St. Louis, MO) was used to remove any trace amounts of genomic DNA. Purified RNA samples were then quantified by using a NanoDrop 1000 apparatus (Wilmington, DE). Equal amounts of RNA were reverse transcribed using random hexamer primers and Moloney murine leukemia virus reverse transcriptase (Invitrogen, Inc., Carlsbad, CA).
Quantitative PCR was performed using the Roche LightCycler 480 system with SYBR green master mix reagents (Roche Applied Science, Indianapolis, IN). Samples were assayed in duplicate with cDNA reverse transcribed from 10 ng of RNA in 10-μl PCR mixtures. The genes for ribosomal protein large protein 38 (Rplp38), glyceraldehyde-3-phosphate dehydrogenase (Gapdh), and β-actin were used as the housekeeping gene controls for ΔCT calculations [ΔCT = (CT of the target gene) − (average CT of housekeeping genes)]. Primer sequences and PCR conditions are listed in Table S2 of the supplemental material. Fold expression values were calculated using the 2−ΔΔCT method, where ΔΔCT = (ΔCT of the treatment sample) − (ΔCT of control samples) (with the control value normalized to 1). Statistical significance was determined by analysis of variance.
Western blotting.
Expressional levels of proteins were determined by Western blot analysis. Briefly, myoblasts attached to plates were rinsed with ice-cold phosphate-buffered saline (PBS), and cells were scraped on ice in radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris-Cl [pH 8.0], 200 mM NaCl, 50 mM NaF, 1 mM dithiothreitol, 1 mM Na3VO4, 0.3% IGEPAL, and protease inhibitor cocktail). Cells were collected in 1.5-ml Eppendorf tubes, lysed on ice for 30 min, and centrifuged at 15,000 rpm for 10 min at 4°C to removed cell debris. Tissue samples were homogenized in RIPA buffer, lysed, and centrifuged in similar ways. Protein concentrations were quantified with bicinchoninic acid reagent, and 30-μg protein samples were loaded in each lane onto 10-to-12% SDS-PAGE gels. Electrophoresis, electrotransfer of proteins to polyvinylidene difluoride membranes, and antibody staining were conducted following standard procedures previously described (51). The antibodies used are listed in Table S1 of the supplemental material. Antibody-labeled protein bands were revealed by using the RapidStep enhanced chemiluminescence reagent (Calbiochem) and detected with the GEL Logic 2200 imaging system (Carestream Health Inc.).
Muscle injury and regeneration.
Muscle regeneration was induced by injections of cardiotoxin (CTX; Sigma-Aldrich, St. Louis, MO) into the tibialis anterior (TA) muscle. Mice were anesthetized using a ketamine-xylazine cocktail, and then 50 μl of 10 μM CTX was injected into the left TA muscle. Muscles were then harvested at 7 and 21 days postinjection to assess the completion of regeneration and repair.
Muscle cryosectioning and staining.
Fresh or fixed TA muscles were embedded in OCT compound and frozen in isopentane chilled on a dry ice-ethanol slurry. For fixation, muscles were placed in 4% paraformaldehyde (PFA) for 1 h, washed in 100 mM glycine for 1 h, and then sunk in 30% sucrose at 4°C overnight. Frozen muscles were cut into 10-μm-thick cross-sections by using a Leica CM1850 cryostat.
Immunofluorescence staining of muscle sections was performed as previously described (22) using the antibodies listed in Table S1 of the supplemental material. Fluorescent images were captured with a CoolSnap HQ charge-coupled-device camera (Photometrics) driven by IP Lab software (Scanalytics Inc.) using a Leica DM6000 microscope with a 20× objective (numerical aperture, 0.70). Single-channel black-and-white fluorescent images were assembled into multicolor images with the Photoshop software. Hematoxylin and eosin (H&E)-stained images were captured with a Nikon D90 digital camera installed on a Nikon (Diaphot) inverted microscope. The cross-sectional area (CSA) of myofibers was analyzed with the Image Analysis Tool in Photoshop. Specifically, the CSA was calculated as the total myofiber area divided by the number of myofibers.
Culture and staining of single myofibers and primary myoblasts.
Single myofibers were isolated from the extensor digitorum longus (EDL) muscles after digestion with collagenase A (Sigma) and trituration as previously described (7, 41). Suspended fibers were cultured in 60-mm horse serum-coated plates in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT), 2% chicken embryo extract (Accurate Chemical, Westbury, NY), and 1% penicillin-streptomycin for 3 days. Freshly isolated fibers and cultured fibers were then fixed in 4% PFA and stained for Pax7, MyoD, and green fluorescent protein (GFP) as previously described (22).
Primary myoblasts were isolated from hind limb skeletal muscle. Muscles were minced and digested in type I collagenase and Dispase B mixture (Roche Applied Science). Cells were then filtered from debris, centrifuged, and cultured in growth medium (F-10 Ham's medium supplemented with 20% FBS, 4 ng/ml basic fibroblast growth factor, and 1% penicillin-streptomycin) on collagen-coated cell culture plates at 37°C, 5% CO2, as previously described (51).
Viral and electroporation-mediated gene transfer.
Notch target genes Hes5, HeyL, and Nrarpa were cloned from myoblast cDNA with restriction enzyme sites at each terminal and then were ligated to pCMV-2b plasmids. Hes1, Hes6, and Hey1 were subcloned from plasmids purchased from Open Biosystems (Thermo Fisher Scientific, Huntsville, AL) to pCMV-2b plasmids separately. The pPGK-Cre-bpA plasmid was obtained from Addgene (Cambridge, MA).
Adenovirus containing Cre-EGFP or EGFP (generously provided by Y.-X. Wang, University of Massachusetts) was used to activate N1ICD in primary myoblasts isolated from ROSA-N1ICD mice. Adenovirus infection was performed when myoblasts reached 50 to 80% confluence. Myoblasts in plates were washed in PBS 3 times, and then adenovirus diluted in Ham's F-10 was added. The Ham's F-10 medium was replaced with fresh growth medium and plates were returned to the incubator for 30 to 60 min. Virus titers were determined in myoblasts by scoring GFP-positive cells (36). Adenovirus-infected myoblasts were harvested at 24 h for RNA and protein extraction.
Lentivirus-mediated shRNA knockdown.
An shRNA set targeting RBP-Jκ was purchased from Open Biosystems (Thermo Fisher Scientific, Huntsville, AL) and packaged into lentiviral particles in 293T cells together with lentiviral envelope plasmids. Supernatant was removed and replaced with fresh medium after 24 h. In another 24 h, supernatant was collected and centrifuged to remove the cells and debris (3,000 rpm, 5 min). Equal volumes of lentiviral supernatant were mixed with fresh myoblast growth medium and added to the cultured primary myoblasts. Myoblasts were harvested at 48 h for isolation of DNA, RNA, and protein.
Electroporation-mediated transfection was performed on primary myoblasts by using a Neon transfection system (Invitrogen) as previously described (51). Briefly, 1 × 105 cells were suspended in 10 μl electroporation buffer, including 2.5 μg plasmids containing Notch target genes, and electroporated (1,300 V, 10 ms, 3 times), using a 10-μl tip. After electroporation, the cells were cultured in 6-well plates. In 24 h, the cells were harvested for RNA purification.
Cell growth rate and 5-bromodeoxyuridine (BrdU) incorporation analysis.
Primary myoblasts were seeded in 6-well plates (1 × 105 cells/well) and cultured under standard myoblast conditions (see above). The cells were harvested at days 2, 4, 6, 10, and 12 and counted by using a hemocytometer.
A BrdU incorporation assay was employed to determine the proliferation of myoblasts. Myoblasts were cultured in 4- or 8-well chamber slides. At 70% confluence, BrdU was added to the medium at a 60 μM final concentration. Cells were fixed in 4% PFA after 24 h and incubated in HCl (1 M) for 10 min on ice, followed by HCl (2 M) for 10 min at room temperature (RT). Borate buffer was added to neutralize the cells. BrdU staining was performed, followed by GFP, Pax7, and MyoD staining.
Chromotin immunoprecipitation.
Proliferating primary myoblasts and C2C12 myoblasts were cross-linked with 1% formaldehyde in Ham's F-10 medium for 10 min at RT followed by the addition of 125 mM glycine for 5 min at RT, after which cells were scraped into SDS lysis buffer. The cells were further sonicated and diluted for immunoprecipitation with the indicated antibodies. The immunoprecipitates were eluted and reverse cross-linked overnight at 65°C. DNA fragments were purified using the Cycle Pure kit (Omega Bio-Tek, Inc., Norcross, GA), and qPCR was performed for quantification.
RESULTS
Satellite cell-specific NICDOE promotes the self-renewal of myogenic progenitors.
To investigate the role of Notch signaling in postnatal satellite cells, we established Pax7-CreER/ROSA-N1ICD (called Pax7-NICDOE) mice (Fig. 1A). In this model, Cre-dependent expression of N1ICD and nuclear GFP is driven by the ubiquitously expressed endogenous Rosa26 promoter. Upon activation of CreER by Tamoxifen (TAM), NICD is specifically expressed in Pax7-expressing satellite cells. The expression of GFP in the nucleus serves as a marker to identify which cells have NICDOE. To examine how Notch signaling regulates the cell fate dynamics of satellite cells, we dissociated single EDL myofibers and cultured them in suspension to activate satellite cells. Suspended culture of satellite cells while still attached to host myofibers provides an excellent in vitro model for analyzing cell fate segregation among self-renewal, proliferation, and differentiation based on Pax7 and MyoD expression. Specifically, activated satellite cells (called myoblasts) undergoing self-renewal, proliferation, and differentiation can be distinguished as Pax7+ MyoD−, Pax7+ MyoD+, and Pax7− MyoD+, respectively (16, 32, 55).
Overexpression of N1ICD promotes the self-renewal of satellite cells. (A) Experimental design. Results are based on Pax7-CreER/Rosa-N1ICD mice, in which N1ICDOE is induced by 4-OH-TAM and N1ICDOE myoblasts coexpress nEGFP (nuclear localized GFP). Satellite cells undergo activation, proliferation, self-renewal, and differentiation while attached on their host myofibers. (B) Myoblast clusters on cultured single fibers were labeled with antibodies against Pax7, MyoD, and GFP. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Note that GFP marks the N1ICDOE myoblasts. Dotted circles indicate the lack of MyoD staining in GFP+ cells. Bar, 20 μm. (C and D) Percentage of self-renewing (Pax7+ MyoD−) and proliferating (Pax7+ MyoD+) myobalsts in control and N1ICDOE myoblast colonies (calculated from 28 cultured colonies in each group). (E) Percentage of self-renewing Pax7+ MyoD− and proliferating Pax7+ MyoD+ myobalsts for GFP− (~30%) and GFP+ (~70%) subpopulations among N1ICDOE myoblast clusters (n = 28). (F) Number of myoblasts per cluster in control and N1ICDOE myofiber cultures (n = 28).
EDL myofibers from control and Pax7-NICDOE mice pretreated with TAM were cultured for 72 h and labeled with Pax7, MyoD, and GFP (Fig. 1A). Whereas the majority of the myoblasts on the control EDL myofibers coexpressed Pax7 and MyoD (Fig. 1B, left panels), the NICDOE myoblasts (labeled with GFP) exhibited reduced MyoD expression (Fig. 1B, right panels). This observation is consistent with previous reports that Notch signaling inhibits myogenic differentiation. However, it is unclear whether the myoblasts blocked from differentiation are in the proliferation or self-renewal state. To address this, we quantified the percentage of self-renewing (Pax7+ MyoD−), proliferating (Pax7+ MyoD+), and differentiating cells (Pax7− MyoD+). Strikingly, the proportions of self-renewing cells doubled (Fig. 1C), while proliferating cells were reduced (Fig. 1D), on the NICDOE myofibers.
As nuclear GFP expression distinguishes whether a cell expresses the constitutive activated N1ICD (GFP+) or not (GFP−), we further analyzed the GFP+ and GFP− cell populations on cultured Pax7-NICDOE myofibers (Fig. 1B, right panels). Consistent with our earlier finding, over 90% of the GFP+ cells were self-renewing, while only 10% of the GFP− cells were self-renewing (Fig. 1E, left). On the other hand, less than 10% of the GFP+ cells proliferated, while nearly 80% of the GFP− cells proliferated (Fig. 1E, right). Overall, the average number of myoblasts within each colony of cells growing on the Pax7-NICDOE myofibers was reduced compared to the control myofibers at 72 h (Fig. 1F). Using the same culture paradigm, de novo activation of Notch by using adenovirus-mediated Cre transduction further confirmed the above observations (data not shown). Together, these results provide compelling evidence that NICDOE promotes self-renewal of activated satellite cells while inhibiting their proliferation.
NICDOE inhibits the proliferation of primary myoblasts.
That constitutive Notch activation inhibits myoblast proliferation was totally unexpected, as previous studies indicated that Notch signaling promoted proliferation. To confirm this novel observation, we compared the growth curves and BrdU incorporation and Ki67 expression levels of control and NICDOE primary myoblasts. ROSA-N1ICD myoblasts were transfected with a Cre plasmid to induce NICDOE, which resulted in a 3.5-fold increase in N1ICD transcripts and a 6-fold increase in Notch target Hey1 gene expression (Fig. 2A). Such a moderate upregulation of Notch signaling may represent a more “physiological” condition. Next, we compared the growth of Pax7-CreER/ROSA-N1ICD myoblasts treated with vehicle control or 0.4 uM 4-hydroxy-TAM, which activates overexpression of NICD. Strikingly, the NICDOE myoblasts exhibited much-reduced growth rates at all time points examined (Fig. 2B). These results demonstrated that NICDOE inhibits the growth of myoblasts in culture.
Overexpression of N1ICD inhibits myoblast proliferation. Results are based on Rosa-N1ICD myoblasts electrotransfected with pGK-Cre or empty plasmids. Cre+ myoblasts (N1ICDOE) are marked by nGFP, while Cre− cells (control) don't exhibit GFP staining. (A) Expression levels of N1ICD and Hey1 upon electroporation of pGK-Cre in Rosa-N1ICD myoblasts (n = 3). Results in panels C and D, below, are based on this N1ICDOE strategy. (B) Growth rates of Pax7-CreER/Rosa-N1ICD myoblasts treated with vehicle (control) or 4-OH-TAM (0.4 μM), which activates N1ICD in Pax7-expressing myoblasts (n = 3). (C) BrdU incorporation by GFP+ and GFP− Rosa-N1ICD myoblasts transfected with pGK-Cre. Arrows show that GFP+ myoblasts failed to incorporate BrdU and lacked the expression of MyoD. Asterisks indicate GFP− myoblasts that incorporated BrdU and expressed MyoD. The bar graph shows the percentage of BrdU+ cells in Rosa-N1ICD myoblasts transfected with empty (CON) or pGK-Cre (N1ICD) plasmids (n = 3). Bar, 40 μm. (D) Proliferating cell marker Ki67 expression in GFP+ and GFP− Rosa-N1ICD myoblasts transfected with pGK-Cre. Arrows show that GFP+ myoblasts were Ki67− and mostly MyoD−. Asterisks indicate GFP− myoblasts that expressed Ki67 and MyoD normally. The bar graph shows the percentage of Ki67+ cells in Rosa-N1ICD myoblasts transfected with empty (CON) or pGK-Cre (N1ICD) plasmids (n = 3). Bar, 40 μm.
We next incubated myoblasts with BrdU for 24 h and quantified the percentage of S-phase cells, based on BrdU incorporation and using a Cre plasmid to activate NICD (Fig. 2C). As Cre transfection induces coexpression of GFP and N1ICD in ROSA-N1ICD cells, GFP expression distinguishes whether a cell expresses the constitutively activated N1ICD (GFP+) or not (GFP−). Importantly, NICDOE (GFP+) cells were predominately BrdU− (Fig. 2C, arrows). Conversely, BrdU+ cells were almost all GFP− (see Fig. 2C [asterisks], below). Overall, the percentage of BrdU+ cells was significantly reduced in Cre-transfected cells compared to control empty plasmid-transfected cells (Fig. 2C, graph). This mutually exclusive expression pattern of GFP (NICDOE) and BrdU incorporation (DNA synthesis) indicated that NICDOE inhibits S-phase entry of myoblasts.
In addition, we analyzed the expression of the cell proliferation marker Ki67 in ROSA-N1ICD myoblasts transfected with Cre or empty plasmids. Consistent with the BrdU incorporation results, GFP expression, which marks NICDOE, and Ki67 expression were mutually exclusive (Fig. 2D). The overall percentage of Ki67+ cells in Cre-transfected culture was less than half that in the control group (Fig. 2D, graph). Together, our growth curve, BrdU uptake, and Ki67 expression analyses provided convincing evidence for an inhibitory role of constitutively activated Notch in the proliferation of satellite cell-derived primary myoblasts.
NICDOE upregulates Pax7 and elicits coordinated changes in target gene expression.
To identify potential target genes mediating the effect of Notch signaling in satellite cell self-renewal, we analyzed the expression of a number of genes involved in the Notch signaling pathway as well as Pax7, which is known to be critical for satellite cell self-renewal in control and NICDOE myoblasts. NICDOE was induced by adenovirus-mediated Cre transduction into ROSA-N1ICD myoblasts. NICDOE resulted in a 14-fold increase in N1ICD transcript levels compared to the control treatment, in which the same myoblasts were transduced with adenovirus-GFP (Fig. 3A). Western blot analysis confirmed the increased N1ICD and GFP expression in Cre-transduced cells, but the endogenous Notch1 receptor expression remained unchanged (Fig. 3B). As expected, MyoD protein was dramatically reduced by NICDOE, validating the effectiveness of the treatment. What interested us was that Pax7 protein was also increased in the NICDOE cells (Fig. 3B). To confirm this result, we examined the mRNA by qPCR. Again, Pax7 mRNA was increased by 3.5-fold, whereas MyoD and Myogenin were both decreased by 3- to 4-fold in NICDOE cells (Fig. 3C). As MyoD has been reported to inhibit Pax7, we further investigated if Pax7 upregulation is a secondary effect of MyoD inhibition by Notch activation. We overexpressed N1ICD in MyoD−/− myoblasts and observed robust upregulation of Pax7 (Fig. 3G). Thus, Notch activation upregulated Pax7 in the absence of MyoD. Our results indicate that NICDOE upregulates Pax7 at both mRNA and protein levels independent of MyoD inhibition, resulting in enhanced self-renewal of satellite cells.
N1ICDOE upregulates Pax7 and elicits coordinated changes in the expression of Notch-related genes. (A) Relative levels of N1ICD expression in myoblasts treated with adenovirus-GFP (normalized to 1) or adenovirus-Cre (n = 3). (B) Western blots showing changes in N1ICD and myogenic proteins in Rosa-N1ICD myoblasts treated with adenovirus-GFP or adenovirus-Cre. (C) Relative expression levels of myogenic genes Pax7, Myf5, MyoD, and Myogenin in myoblasts treated with adenovirus-GFP (normalized to 1) or adenovirus-Cre (n = 3). (D) Relative expression levels of Notch target genes Hes1, Hes5, Hes6, Hey1, HeyL, and Nrarpa in Rosa-N1ICD myoblasts treated with adenovirus-GFP (normalized to 1) or adenovirus-Cre (n = 3). (E) Relative expression levels of Notch receptor genes in myoblasts treated with adenovirus-GFP (normalized to 1) or adenovirus-Cre (n = 3). (F) Representative Western blot confirming the increased protein level of the Notch3 receptor in Rosa-N1ICD myoblasts treated with adenovirus-Cre. (G) NICDOE upregulates Pax7 independently of MyoD. Primary myoblasts were isolated from MyoD−/− mice and transfected with a RAMIC plasmid to overexpress N1ICD (+). Control (−) myoblasts were transfected with an empty plasmid.
To provide insights into how NICDOE upregulates Pax7, we examined Notch target genes that are coregulated with Pax7. Interestingly, Nrarp and Hey family genes Hey1 and HeyL were most robustly upregulated, with 140- to 640-fold increases induced by NICDOE (Fig. 3D). The Hes family genes were modestly upregulated, with 7-fold and 12-fold increases in the expression levels of Hes1 and Hes5, respectively (Fig. 3D). In contrast, Hes6 expression was reduced by 70% in NICDOE cells (Fig. 3D). Moreover, we examined the expression of Notch receptors in order to understand if there was a compensatory reduction of endogenous Notch receptors. Whereas Notch1 and Notch2 expression levels were not affected, Notch3 expression was unexpectedly increased by 50-fold in NICDOE cells (Fig. 3E). Consistently, Notch3 protein was also robustly upregulated by NICDOE (Fig. 3F). Therefore, NICDOE results in increased Pax7 expression and coordinated alterations in the expression of Notch receptor and target genes in primary myoblasts.
Notch signaling regulates satellite cell self-renewal, proliferation, and differentiation through distinct targets.
As Pax7 upregulation is correlated with corresponding increases with Hes, Hey, and Nrarp genes, we asked if Pax7 expression is regulated by these canonical Notch targets. To address this, we individually overexpressed Hes, Hey, and Nrarp genes in primary myoblasts (Fig. 4A), which led to 8- to 1,000-fold increases in their expression (except for Hes5, whose expression was increased by 100,000-fold due to extremely low Hes5 expression in control cells). To our surprise, none of these overexpressed genes was able to mimic the effect of NICDOE on Pax7 expression (Fig. 4B). Although overexpression of Hey1 resulted in a 20% increase in Pax7, this increase was not near the 3-fold increase induced by NICDOE. Overexpression of other genes either reduced or had no effect on Pax7 expression (Fig. 4B). Consistent with the NICDOE results (Fig. 3C), Myf5 expression was largely unaffected by the Notch target genes (Fig. 4C). In contrast, individual overexpression of all the Notch target genes resulted in roughly 40% suppression of MyoD (Fig. 4D), and all genes except for Hes6 suppressed Myogenin gene expression (Fig. 4E). Intriguingly, individual overexpression of Notch target genes also led to 20 to 30% reductions in Ki67 expression (Fig. 4F). These results are consistent with our previous observation that NICDOE inhibits myoblast proliferation. Collectively, these overexpression experiments suggest that Notch signaling regulates satellite cell fate choice through distinct mediators: while Notch suppresses proliferation and differentiation through canonical Hes, Hey, and Nrarp proteins, Notch regulates self-renewal (Pax7) independent of these canonical targets.
Canonical Notch target genes mimic the effects of N1ICDOE in the suppression of MyoD, Myogenin, and Ki67 but fail to upregulate Pax7. (A) Fold increases (on a log scale) of various Notch target gene expression levels upon transfection of corresponding plasmids to overexpress Hes1, Hes5, Hes6, Hey1, HeyL, and Nrarpa (n = 5). (B to F) Fold changes of Pax7 (B), Myf5 (C), MyoD (D), Myogenin (E), and Ki67 (F) gene expression in response to Hes1, Hes5, Hes6, Hey1, HeyL, and Nrarpa gene overexpression. Fold changes are relative to control myoblasts transfected with empty plasmids (normalized to 1) (n = 3).
N1ICD directly regulates Pax7 expression through RBP-Jκ.
As canonical Notch targets Hes, Hey, and Nrarpa had no effect on Pax7 expression, we hypothesized that N1ICD directly regulates Pax7. Analysis of the Pax7 upstream genomic sequence identified two RBP-Jκ consensus binding sites (GTGGGAA) (Fig. 5A). To investigate if they are associated with RBP-Jκ, we performed a chromatin immunoprecipitation (ChIP) assay. Indeed, both binding sites were enriched by more than 5-fold in the ChIP assay with RBP-Jκ antibody (Fig. 5B and C), suggesting that RBP-Jκ occupies these sequences in the Pax7 promoter region. Importantly, ChIP analysis using N1ICD antibody also led to 2- to 3-fold enrichment of these binding sites (Fig. 5B and C), indicating that N1ICD is also associated with the RBP-Jκ consensus binding sites in the Pax7 gene upstream region. In contrast, both Pax7 and RBP-Jκ failed to enrich DNA fragments 200 bp away from the consensus RBP-Jκ binding sites (Fig. 5D and E), confirming the specificity of N1ICD and RBP-Jκ binding. As the N1ICD enrichment of target DNA is lower than that by RBP-Jκ, we predicted that the association between N1ICD and Pax7 promoter regions is mediated by RBP-Jκ. Consistent with this notion, lentivirus-mediated shRNA knockdown of RBP-Jκ abolished the enrichment of the target DNA sequences by both N1ICD and RBP-Jκ antibodies (Fig. 5F and G) and resulted in decreased Pax7 expression (Fig. 5H). Together, these results provide strong biochemical and functional evidence that N1ICD directly regulates Pax7 gene expression through RBP-Jκ.
N1ICD directly regulates Pax7 through consensus RBP-Jκ binding. Results were based on ChIP analysis of enrichment of DNA fragments surrounding evolutionally conserved RBP-Jκ GTGGGAA-binding motifs upstream of the Pax7 allele. (A) DNA sequence analysis identified conserved RBP-Jκ-binding sites upstream of the transcriptional start site of Pax7. Nonbinding sites were selected as negative controls for DNA enrichment analysis. (B and C) ChIP assay combined with quantitative PCR results, showing enrichment of two RBP-Jκ-binding sites at nucleotide positions −4532 and −7405, based on both anti-Notch1 and anti-RBP-Jκ antibodies (n = 3). RNA polymerase antibody was used as a positive control, and mock IgG was used as a negative control for nonspecific binding. (D and E) Two nonbinding sites (−3001 and −6601) were selected for ChIP-PCR analysis to show the lack of enrichment by anti-Notch1 or anti-RBP-Jκ but strong enrichment by RNA polymerase antibody (n = 3). (F and G) ChIP-PCR results after shRNA knockdown of RBP-Jκ. Enrichments by Notch1 and RBP-Jκ were both abolished at the −4532- and −7405-binding sites, while the enrichment by RNA polymerase was not affected by RBP-Jκ knockdown (n = 3). (H) shRNA-mediated knockdown of RBP-Jκ inhibits Pax7 expression in primary myoblasts. Two lentiviral knockdown vectors (97288 and 97286) together with an empty lentiviral vector (control) were used.
Satellite cell-specific NICDOE impairs the regeneration of postnatal skeletal muscles.
Previous reports showed that activation of Notch signaling restores the regenerative deficiencies of aged muscles (8). However, it has not been shown if muscle satellite cell-specific Notch activation is responsible for such improvements in muscle regeneration. To address this question, we examined the regenerative response of acutely injured TA muscles of Pax7-NICDOE and wild-type mice. Both types of mice were treated with TAM for 5 days, followed by a 5-day waiting period to allow for Cre-mediated activation of N1ICD, and then treated with CTX to degenerate the TA muscles. The regenerative efficiency was evaluated at 7 and 21 days postinjection (dpi) of CTX (Fig. 6A). NICDOE in quiescent satellite cells did not alter the satellite cell number in the resting (noninjured) EDL muscles (Fig. 6B), suggesting that muscle regeneration was mediated by equal numbers of satellite cells prior to CTX-induced injury. At 7 dpi, the damaged muscles were fully replaced by newly regenerated myofibers characterized by centrally localized myonuclei in the control mice (Fig. 6C). However, the injured Pax7-NICDOE muscles were poorly repaired, with increased interstitial space and adipose infiltration (Fig. 6D). At 21 dpi, the regenerated myofibers became significantly larger and the muscle structure was restored in the control mice (Fig. 6E and G). By sharp contrast, the regenerated myofibers remained small, and large interstitial spaces were evident in the NICDOE mice (Fig. 6F and G). At 7 dpi, the number of regenerating myofibers per microscopic field (~0.15 mm2) was reduced by 50% in the Pax7-NICDOE compared to the control muscles (Fig. 6H). Interestingly, due to their extreme small size, the number of regenerating myofibers per unit area was increased by 50% in the NICDOE compared to the control mice at 21 dpi (Fig. 6G and H). In contrast to the markedly reduced muscle regeneration, the number of Pax7+ progenitors was strikingly increased in the regenerating Pax7-NICDOE muscle cells compared to the control muscles at both 7 and 21 dpi (Fig. 6K). Finally, analysis of phospho-histone 3 (PH3) expression (Fig. 6L and M) indicated that NICDOE reduced satellite cell proliferation (Fig. 6N). Thus, constitutive activation of Notch in postnatal satellite cells reduced their proliferation and differentiation but enhanced satellite cell self-renewal, leading to impaired muscle regeneration in vivo.
Satellite cell-specific N1ICDOE leads to defective muscle regeneration. (A) Experimental design. Two-month-old littermate control (ROSA-N1ICD) and N1ICDOE (Pax7-CreER/Rosa-N1ICD) mice were injected intraperitoneally with 5 doses of Tamoxifen. TA muscles are subsequently injected with CTX to induce regeneration, and samples were collected at 7 and 21 dpi of CTX. (B) N1ICDOE did not affect the number of quiescent Pax7+ satellite cells per EDL myofiber in noninjured (resting) muscles (n = 3 pairs of mice, >20 fibers per EDL muscle analyzed). (C and D) Representative H&E staining of regenerating TA muscles at 7 dpi in control (C) and N1ICDOE (D) mice. Red cells with central nuclei represent regenerating myofibers, and white round cells are adipocytes. (E and F) The same experiment as shown in panels C and D, but at 21 dpi. Bar, 100 μm (C to F). (G) Number of regenerated myofibers/mm2 in control and N1ICDOE TA muscles at 7 dpi (n = 4) and 21 dpi (n = 3 pairs of mice). (H) Average CSA of regenerated fibers in control and N1ICDOE TA muscles at 7 dpi (n = 4) and 21 dpi (n = 3 pairs of mice). (I and J) Representative Pax7 immunostaining images of regenerating TA muscles in the control (I) and N1ICDOE (J) mice at 7 dpi. Bar, 50 μm. (K) Average number of Pax7+ cells per field (~0.15 mm2) in control and N1ICDOE TA muscles at 7 dpi (n = 4) and 21 dpi (n = 3 pairs of mice). (L and M) Representative images of phospho-histone 3 (PH3) and Pax7 double labeling of regenerating TA muscles in control (L) and N1ICDOE (M) mice at 7 dpi. The dotted line in panel M indicates a degenerated myofiber. Bar, 20 μm. (N) Percentage of PH3+ metaphase satellite cells in control and N1ICDOE regenerating TA muscles at 7 dpi (n = 3 pairs of mice).
DISCUSSION
Postnatal muscle growth, maintenance, and regeneration are mediated by muscle-specific stem cells (satellite cells) that are capable of self-renewal and myogenic differentiation. Upon activation, satellite cell progenies adopt divergent cell fates: while some remain in the cell cycle and proliferate, others withdraw from the cell cycle and self-renew or differentiate (55). To investigate the potential role of Notch signaling in the cell fate specification of muscle stem cells, we specifically activated Notch1 in postnatal satellite cells and revealed several novel roles of Notch signaling. We found that NICDOE inhibited myoblast proliferation. We also discovered a role of Notch signaling in the self-renewal of satellite cells independent of inhibition of myogenic differentiation (Fig. 7). Furthermore, we elucidated the molecular mechanisms through which Notch regulates Pax7.
Model for Notch's regulation of stem cell fate in muscle. Quiescent satellite cells (stem cells) express high levels of Pax7. The activated satellite cells (myoblast) coexpress Pax7 and MyoD and proliferate. The proliferating myoblast can either downregulate Pax7 to differentiate or downregulate MyoD to self-renew. Activated Notch (NICD) binds to RBP-Jκ to form a transcriptional activation complex that upregulates the transcription of Pax7 as well as canonical Notch targets (Hes and Hey family genes). Hes/Hey proteins inhibit MyoD gene transcription. Therefore, Notch activation upregulates Pax7 to promote satellite cell self-renewal, while inhibiting MyoD to block myogenic differentiation. Pax7 upregulation and MyoD downregulation may together lead to cell cycle withdrawal. The reciprocal inhibitory action between Pax7 and MyoD further amplifies the effect of Notch signaling.
Although in vivo analysis of Notch gain- and loss-of-function mutants demonstrated an important role of Notch signaling in myogenic lineage development and postnatal muscle regeneration, it has been unknown if Notch signaling regulates self-renewal and differentiation independently or if Notch only blocks differentiation, which subsequently affects the balance between self-renewal and differentiation (Fig. 7). Our analysis of satellite cell fate choice on cultured myofibers supports a conclusion that NotchOE enhances the self-renewal of activated satellite cells through upregulation of Pax7. Previous studies established that Notch activation robustly inhibits the expression of MyoD and Myogenin, which have been shown to be able to repress the expression of Pax7 (34). These results led to the speculation that Notch-induced Pax7 upregulation is a secondary consequence due to MyoD and Myogenin inhibition (49). Our studies clearly demonstrated that whereas Hes1, Hes5, Hey1, HeyL, and Nrarp overexpression all suppressed the expression of MyoD and Myogenin, they had little effect on Pax7. Thus, NICDOE-induced Pax7 upregulation is not mediated by the canonical Notch targets and is thus independent of MyoD/Myogenin suppression. The relatively modest inhibition of MyoD/Myogenin by individual canonical Notch targets compared to the level of inhibition by NICDOE could be due to several possibilities. First, NICDOE alters the expression of a repertoire of target genes whose coordinated changes define the final level of MyoD. Thus, individually manipulating these genes cannot fully recapitulate the effect of NICDOE. Second, overexpression of Hes/Hey target genes had no effect on Pax7 expression, while NICDOE upregulated Pax7, which may further inhibit MyoD expression. It will be interesting to investigate if Notch-induced Pax7 upregulation contributes to the downregulation of MyoD, as shown in other studies (32–34).
Our current results, indicating that Notch upregulates Pax7 and promotes satellite cell self-renewal, are consistent with several recent studies. Specifically, two recent studies indicated that Notch signaling is required for maintaining the quiescence of satellite cells (2, 28). This conclusion is consistent with our results that Notch activation promotes self-renewal of activated satellite cells. It has also been shown that pharmacological inhibition of Notch signaling decreased self-renewing progeny in satellite cell cultures and Dll1-induced Notch activation increased Pax7 expression (23, 49). However, how Notch regulates Pax7 at the molecular level is unknown. Using ChIP analysis we demonstrated that N1ICD and RBP-Jκ are both associated with the Pax7 promoter region. Importantly, lentivirus-mediated shRNA knockdown of RBP-Jκ not only abolished the physical association between N1ICD and the Pax7 promoter but also functionally inhibited Pax7 gene expression. These results provide strong evidence that N1ICD directly regulates Pax7 expression through RBP-Jκ, but this is independent of inhibition of MyoD and Myogenin, which is mediated by Hes and Hey family proteins.
Our growth curve analysis, BrdU incorporation, Ki67 assays, and gene expression analysis all suggest that Notch activation inhibits the proliferation of primary myoblasts. These results were unexpected, although they are in line with a recent study showing that Notch3 mutation enhanced the proliferation of primary myoblasts (19). Indeed, we observed elevated expression of Notch3 in NICDOE myoblasts, consistent with previous results showing that Notch3 was increased by N1ICD and downregulated by the γ-secretase inhibitor in T6E cells (53). In addition, our observation is consistent with recent studies demonstrating that compound mutation of Notch targets Hey1 and HeyL, or mutation of Rbpj, leads to loss of quiescence in satellite cells and spontaneous entry into the cell cycle or differentiation (2, 14, 28). Moreover, upregulation of Pax7 itself may act as a regulator to withdraw satellite cells from the cell cycle (32). In contrast, several previous studies indicate that Notch activation promotes myoblast proliferation. Specifically, retroviral transduction of constitutively activated Notch1 into primary myoblasts increased their proliferation, whereas transfection of Numb, a Notch inhibitor, inhibited myoblast proliferation (10). At the molecular level, activation of Notch signaling blocks TGF-β-induced upregulation of cyclin-dependent kinase (CDK) inhibitors p21, p15, p16, and p27 (6). Interestingly, several studies have demonstrated that Notch inhibits proliferation through upregulation of p21 in other systems (45). Thus, whether Notch activation stimulates or inhibits proliferation is highly cell context dependent. The contradictory effects of Notch activation on myoblast proliferation observed by different groups may also be due to culture conditions, as growth factors drastically influence the outcome of Notch activation in hematopoietic progenitor cells (52). As these studies were all performed in vitro, it is imperative to clarify how Notch activation affects satellite cell proliferation in vivo.
The perinatal lethality caused by constitutive activation (Bi P, et al, unpublished data) or inactivation of Notch signaling in myogenic cell lineages during embryonic myogenesis precludes analysis of postnatal satellite cells and muscle regeneration in vivo. To circumvent the lethality problem, we used Tamoxifen-inducible Pax7-CreER mice to drive NICDOE in postnatal satellite cells and examined muscle regeneration in vivo. We found that satellite cell-specific NICDOE inhibited muscle regeneration, accompanied by increased Pax7+ cells, due to enhanced self-renewal and reduced differentiation of activated satellite cells. This observation is consistent with the developmental phenotypes associated with Notch mutation, which resulted in the ablation of satellite cells (44, 50). However, these results are in contrast to previous observations that Notch activation improves muscle regeneration in old mice (8). This discrepancy may be caused by several mechanisms. First, Conboy et al. injected recombinant proteins (Jagged1-Fc and Notch1 antibody) to injury sites to inhibit and activate Notch signaling, respectively (8). Therefore, the inhibition and activation of Notch signaling are transient (depending on protein stability), focal (at the injection site), and nonspecific (any cells near the injection site can be targeted). In contrast, our genetic activation approach specifically activates Notch in Pax7-expressing cells (satellite cells in the muscle) and results in widespread expression (no physical boundaries). Second, previous results were based on aged mice (23 to 24 months old), but our current results were based on experiments in young mice (2 to 3 months old). However, given the well-established inhibitory role of Notch in myogenic differentiation, we predict that constitutive Notch activation will similarly inhibit the regeneration of old muscles, while transiently activating Notch may facilitate the regeneration. Indeed, the age-associated decline in muscle injury repair is reported to be largely attributed to decreased activation of satellite cells by Notch signaling, and Wnt-mediated inhibition of Notch signaling is necessary for myogenic differentiation (5, 9, 43). Third, the amplitude of Notch activation may have been different in these different studies. Together, Notch signaling must be dynamically regulated at different stages (activation, proliferation, and differentiation) in order for damaged muscle to regenerate properly.
ACKNOWLEDGMENTS
Contributions of the authors were the following: Yefei Wen, conception and design of experiments, collection and assembly of data, data analysis and interpretation, manuscript writing, and final approval of the manuscript; Pengpeng Bi, collection and assembly of data, paper revision, and final approval of the manuscript; Weiyi Liu, collection and assembly of data and final approval of the manuscript; Atsushi Asakura, provision of MyoD−/− primary myoblasts and final approval of the manuscript; Charles Keller, provision of Pax7-CreER mouse and final approval of the manuscript; Shihuan Kuang, conception and design of experiments, financial support, administrative support, provision of study materials, data analysis and interpretation, manuscript writing, and final approval of the manuscript.
We thank Yong-Xu Wang (University of Massachusetts School of Medicine) for the adeno-Cre virus, Philip Soriano (MSSM) for Myf5-Cre mice, and Kazuki Kuroda for the NICDOE plasmid. We appreciate the advice of Michael Rudnicki and the technical assistance of Jian Luo.
The project was supported by funding from the MDA, NIH, and USDA to S.K.
We declare no conflicts of interest.
Footnotes
Published ahead of print 9 April 2012
Supplemental material for this article may be found at http://mcb.asm.org/.
REFERENCES
Articles from Molecular and Cellular Biology are provided here courtesy of Taylor & Francis
Full text links
Read article at publisher's site: https://doi.org/10.1128/mcb.06753-11
Read article for free, from open access legal sources, via Unpaywall:
https://mcb.asm.org/content/mcb/32/12/2300.full.pdf
Citations & impact
Impact metrics
Citations of article over time
Alternative metrics
Article citations
Review: myogenic and muscle toxicity targets of environmental methylmercury exposure.
Arch Toxicol, 28 Mar 2024
Cited by: 0 articles | PMID: 38546836
Review
Forkhead box O proteins: steering the course of stem cell fate.
Cell Regen, 13(1):7, 11 Mar 2024
Cited by: 0 articles | PMID: 38466341 | PMCID: PMC10928065
Review
Free full text in Europe PMCNeurofibromin 1 controls metabolic balance and Notch-dependent quiescence of murine juvenile myogenic progenitors.
Nat Commun, 15(1):1393, 15 Feb 2024
Cited by: 0 articles | PMID: 38360927 | PMCID: PMC10869796
Metabolic Changes during In Vivo Maturation of PSC-Derived Skeletal Myogenic Progenitors.
Cells, 13(1):76, 29 Dec 2023
Cited by: 1 article | PMID: 38201280 | PMCID: PMC10778145
METTL3 Promotes the Differentiation of Goat Skeletal Muscle Satellite Cells by Regulating MEF2C mRNA Stability in a m6A-Dependent Manner.
Int J Mol Sci, 24(18):14115, 14 Sep 2023
Cited by: 1 article | PMID: 37762418 | PMCID: PMC10531580
Go to all (161) article citations
Data
Data behind the article
This data has been text mined from the article, or deposited into data resources.
BioStudies: supplemental material and supporting data
Similar Articles
To arrive at the top five similar articles we use a word-weighted algorithm to compare words from the Title and Abstract of each citation.
Proinflammatory cytokine tumor necrosis factor (TNF)-like weak inducer of apoptosis (TWEAK) suppresses satellite cell self-renewal through inversely modulating Notch and NF-κB signaling pathways.
J Biol Chem, 288(49):35159-35169, 22 Oct 2013
Cited by: 26 articles | PMID: 24151074 | PMCID: PMC3853267
Notch Signaling Rescues Loss of Satellite Cells Lacking Pax7 and Promotes Brown Adipogenic Differentiation.
Cell Rep, 16(2):333-343, 23 Jun 2016
Cited by: 32 articles | PMID: 27346341 | PMCID: PMC4945439
Reduced Notch signalling leads to postnatal skeletal muscle hypertrophy in Pofut1cax/cax mice.
Open Biol, 6(9):160211, 01 Sep 2016
Cited by: 7 articles | PMID: 27628322 | PMCID: PMC5043585
Concise Review: Epigenetic Regulation of Myogenesis in Health and Disease.
Stem Cells Transl Med, 5(3):282-290, 21 Jan 2016
Cited by: 67 articles | PMID: 26798058 | PMCID: PMC4807671
Review
Free full text in Europe PMCMarking the tempo for myogenesis: Pax7 and the regulation of muscle stem cell fate decisions.
J Cell Mol Med, 16(5):1013-1025, 01 May 2012
Cited by: 62 articles | PMID: 21615681 | PMCID: PMC4365881
Review
Free full text in Europe PMC