Reciprocal Interaction between TRAF6 and Notch Signaling Regulates Adult Myofiber Regeneration upon Injury

Abstract

Skeletal muscle is a postmitotic tissue that repairs and regenerates through activation of a population of stem-cell-like satellite cells. However, signaling mechanisms governing adult skeletal muscle regeneration remain less understood. In the present study, we have investigated the role of tumor necrosis factor (TNF) receptor-associated factor 6 (TRAF6), an adaptor protein involved in receptor-mediated activation of multiple signaling pathways in regeneration of adult myofibers. Skeletal muscle-specific depletion of TRAF6 in mice (TRAF6^mko) improved regeneration of myofibers upon injury with a concomitant increase in the number of satellite cells and activation of the Notch signaling pathway. Ex vivo cultures of TRAF6^mko myofiber explants demonstrated an increase in the proliferative capacity of myofiber-associated satellite cells accompanied by an upregulation of Notch ligands. Deletion of TRAF6 also inhibited the activity of transcription factor NF-κB and the expression of inflammatory cytokines and augmented the M2c macrophage phenotype in injured muscle tissues. Collectively, our study demonstrates that specific inhibition of TRAF6 improves satellite cell activation and skeletal muscle regeneration through upregulation of Notch signaling and reducing the inflammatory repertoire.

INTRODUCTION

Skeletal muscle regeneration following injury is facilitated by a population of undifferentiated muscle precursor cells, commonly referred to as satellite cells (32). Satellite cells reside between the plasma membrane and basal lamina in a relatively quiescent state and with diminished metabolic activity (11, 18). Quiescent satellite cells express cell surface markers, such as CD34, M-cadherin, and Pax7 (6). When activated by stimuli, such as muscle injury or exercise, satellite cells begin to proliferate and commit to a myoblast cell fate, characterized by the expression of certain myogenic regulatory factors (MRFs) and lineage markers, such as Myf5, MyoD, and α7 integrin, and in resolution exit the cell cycle to either terminally differentiate and fuse to form nascent myotubes or self-renew and return back to quiescence to replenish the satellite cell pool to participate in the next rounds of regeneration (11).

Myofiber regeneration is dynamically regulated by signals released from both the damaged/regenerating muscle as well as other cell types either resident in the muscle or recruited to assist in clearing the damaged myofibers (32, 57, 58). Although considerable progress has now been made to understanding the mechanisms of muscle regeneration, the proximal signaling events leading to the activation of various signal pathways in injured myofibers remain poorly understood. Tumor necrosis factor (TNF) receptor-associated factors (TRAFs) are a family of conserved adaptor proteins which act as signaling intermediates for TNF receptor superfamily members and several other receptor-mediated events leading to context-dependent activation of nuclear factor kappa B (NF-κB), phosphatidylinositol 3-kinase (PI3K)/Akt, and mitogen-activated protein kinase (MAPK) (12, 13). Distinct from other TRAFs, TRAF2 and TRAF6 are also E3 ubiquitin ligases which promote Lys63-linked polyubiquitination of target proteins (35). TRAF6 is unique because it is the only TRAF that mediates Toll-like receptor (TLR)/interleukin-1 receptor (IL-1R) superfamily signaling (12). Interestingly, the expression of TRAF6 (but not other TRAFs) is highly regulated in C2C12 myoblasts. Proliferating myoblasts and the skeletal muscle of neonatal mice express high levels of TRAF6. However, the expression of TRAF6 is considerably reduced upon differentiation of myoblasts into myotubes (49). Importantly, the levels of TRAF6 are increased in differentiated myofibers in response to catabolic stimuli, and TRAF6 mediates skeletal muscle wasting in multiple catabolic conditions (48, 49). However, the role of TRAF6 in satellite cell activation and adult myofiber regeneration remains completely unknown.

Notch is a key signaling pathway involved in embryonic myogenesis and in regulating events that lead to regeneration of adult skeletal muscle (8, 16). Notch signaling is initiated when a Notch ligand, such as Jagged1, Jagged2, Delta-like 1 (DLL1), DLL3, or DLL4, binds to a transmembrane cell surface Notch receptor (Notch1 to Notch4) on the neighboring cell (19). These ligand-receptor interactions lead to proteolytic cleavage of the Notch receptors via the γ-secretase complex, releasing the Notch intracellular domain (NICD), which translocates to the nucleus and binds the transcriptional repressor RBP-Jκ, converting it into an activator and inducing the expression of downstream target genes (19, 29). Some of the most well-defined RBP-Jκ-dependent, Notch target genes include specific members of the Hes/Hey family of basic helix-loop-helix (bHLH) transcription factor Hes1, Hes5, Hes7, Hey1, Hey2, and HeyL genes, which encode bHLH transcriptional repressors that specifically bind to E-box (CANNTG) DNA sequences (4, 46) and mediate much of the Notch function (25). In addition, Notch-regulated ankyrin repeat protein (Nrarp) has also been shown to be a target of Notch signaling (34).

Notch regulates proliferation and commitment of activated satellite cells to myogenic lineage. Activation of Notch signaling is a prerequisite for the expansion of postnatal satellite cells and to prevent the premature differentiation of myogenic precursors in injured myofibers (14, 16, 31, 32, 52, 54, 56, 60). Age-associated decline in satellite cell proliferative capacity is attributed, at least in part, to the insufficient upregulation of Notch ligand DLL and hence reduced activation of Notch leading to impaired muscle regeneration (14). The critical role of Notch in muscle regeneration has been validated by the findings that the forced activation of Notch restored the regenerative potential of aged skeletal muscle (14), an outcome recapitulated by exposure of aged regenerating muscle to serum from young animals (15). Notch signaling is also essential for asymmetric satellite cell division and for progression of cultured myoblasts through cell cycle (10, 31, 56). Recently, it has been shown that the basal level of Notch activity is required for maintenance of satellite cells in an undifferentiated state (7, 21, 42). Furthermore, Notch3 is highly expressed in a subpopulation of quiescent satellite cells (27, 31), indicating that Notch signaling may underlie the heterogeneity of satellite cells.

In contrast to Notch, the activation of the NF-κB signaling pathway has been found to inhibit the differentiation of cultured myoblasts (23, 37) and attenuates the regeneration of adult myofibers upon injury (43). Activation of NF-κB causes the expression of a number of proinflammatory molecules, such as TNF-α and IL-1β, which function by directly inhibiting differentiation of muscle progenitor cells (37). Cross talk between NF-κB and Notch pathways has been implicated in regulating multiple cellular events, such as proliferation, differentiation, and apoptosis. Previous studies have shown that NICD can modulate NF-κB-regulated promoters both positively, by sequestering RBP-Jκ (22, 47), or negatively, by interacting with the p50 subunit of NF-κB (22). It has been suggested that NICD functions as an IκB-like molecule and regulates NF-κB-mediated gene expression through a direct interaction with the p50 subunit of NF-κB (22). This interaction prevents binding of NF-κB to DNA to regulate NF-κB-dependent gene expression (22). However, such interplay between Notch and NF-κB has not yet been characterized in the settings of skeletal muscle regeneration. It is also unclear whether there is a common denominator that controls the activation of these two signaling pathways in regenerating myofibers.

Using skeletal muscle-specific TRAF6-knockout mice, in the present study, we have investigated the role and the mechanisms by which TRAF6 regulates regeneration of adult skeletal muscle. Our study provides initial evidence that the ablation of TRAF6 dramatically improves myofiber regeneration upon injury. Inhibition of TRAF6 augments the expression of Notch ligands in injured myofibers, leading to the activation of satellite cells in a Notch-dependent manner. Moreover, inhibition of TRAF6 attenuates activation of NF-κB and expression of inflammatory cytokines in injured skeletal muscle.

MATERIALS AND METHODS

Animals.

Generation of transgenic floxed TRAF6 (TRAF6^f/f) mice and muscle-specific knockout for TRAF6 (TRAF6^mko) mice has been described previously (28, 49). All mice were in the C57BL/6 background, and their genotypes were determined by PCR from tail DNA. At the age of 8 weeks, 100 μl of 10 μM cardiotoxin (CalBiochem) dissolved in phosphate-buffered saline (PBS) was injected into the tibial anterior (TA) muscle to induce necrotic injury. At various time points, TA muscle was collected from euthanized mice for biochemical and histology studies. All experimental protocols with mice were approved by the Institutional Animal Care and Use Committee at the University of Louisville.

Histology and morphometric analysis.

Hind limb muscles from mice were isolated and frozen in isopentane cooled in liquid nitrogen and sectioned in a microtome cryostat. For the assessment of tissue morphology or visualization of fibrosis, 10-μm-thick transverse sections of muscle were stained with hematoxylin and eosin (H&E) and examined under a Nikon Eclipse TE 2000-U microscope (Nikon). The fiber cross-sectional area (CSA) was analyzed in H&E-stained TA muscle sections using Nikon NIS Elements BR 3.00 software (Nikon). For each muscle, the distribution of the fiber CSA was calculated by analyzing 200 to 250 myofibers as described previously (49). The extent of fibrosis in transverse cryosections of TA muscle was determined by using Masson's trichrome staining kit by following a protocol suggested by the manufacturer (Richard-Allan Scientific).

Electroporation of plasmid DNA in TA muscle.

The injection of plasmid DNA into TA muscle of mice and electroporation were performed according to a protocol as described previously (40, 49). In brief, pcDNA3 and pcDNA3-TRAF6 plasmids were amplified using an endotoxin-free kit (Qiagen) and suspended in sterile saline solution. Mice were anesthetized, and a small portion of TA muscle of both hind limbs was surgically exposed and injected with 30 μl of 0.5 U/μl hyaluronidase (EMD Biosciences). After 2 h, plasmid DNA (50 μg in 25 μl saline) was injected in TA muscle using a 26-gauge needle, and 1 min later a pair of platinum plate electrodes was placed against the closely shaved skin on either side of the small surgical incision, and electric pulses were delivered. Four 20-ms square-wave pulses of 1-Hz frequency at 75 V/cm were generated using a stimulator (model S88; Grass Technologies) and delivered to the muscle. The polarity was then reversed, and an additional four pulses were delivered to the muscle. After electroporation, the wound was closed with surgical clips, and mice were returned to their cages and fed a standard diet.

Indirect immunofluorescence.

For the immunohistochemistry study, TA muscle sections were blocked in 1% bovine serum albumin in phosphate-buffered saline (PBS) for 1 h and incubated with anti-Pax7 (1:10; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA), anti-E-MyHC (1:150; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA), or anti-Jagged2 (1:100; SantaCruz Biotechnology) antibody in blocking solution at 4°C overnight under humidified conditions. The sections were washed briefly with PBS before incubation with Alexa Fluor 488- or 594-conjugated secondary antibody (1:3,000; Invitrogen) for 1 h at room temperature and then washed 3 times for 5 min with PBS. The slides were mounted using fluorescence medium (Vector Laboratories) and visualized at room temperature on Nikon Eclipse TE 2000-U microscope (Nikon), a digital camera (Nikon Digital Sight DS-Fi1), and Nikon NIS Elements BR 3.00 software (Nikon). Image levels were equally adjusted using Abode Photoshop CS2 software (Adobe).

Isolation, culture, and staining of single myofibers.

Single myofibers were isolated from the extensor digitorum longus (EDL) muscles after digestion with collagenase A (Sigma) and trituration as previously described (17). 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; Invitrogen), 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% paraformaldehyde (PFA) and stained for Pax7, MyoD, Ki67, or Jagged2 as described previously (30). To study the role of Notch signaling, myofibers were treated with 10 μM N-[2S-(3, 5-difluorophenyl)acetyl]-l-alanyl-2-phenyl-1,1-dimethylethyl ester-glycine (DAPT) after 24 h of establishing cultures.

Western blot analysis.

Quantitative estimation of specific protein was performed by Western blot analysis using a method described previously (40, 49). TA muscle was washed with PBS and homogenized in lysis buffer (50 mM Tris-Cl [pH 8.0], 200 mM NaCl, 50 mM NaF, 1 mM dithiothreitol [DTT], 1 mM sodium orthovanadate, 0.3% Igepal, and protease inhibitors). Approximately 100 μg protein was resolved on each lane on an 8 to 10% SDS-PAGE gel, electrotransferred onto a nitrocellulose membrane, and probed using anti-MF20 or anti-E-MyHC (1:100; Developmental Studies Hybridoma Bank), anti-TRAF6 (1:1,000; Millipore), anti-phospho-Akt (1:1,000; Cell Signaling, Inc.), anti-total Akt (1:1,000; Cell Signaling, Inc.), anti-phospho-p38 (1:1,000; Cell Signaling, Inc.), anti-total p38 (1:1,000; Cell Signaling, Inc.), and anti-α-tubulin (1:2,000; Cell Signaling, Inc.) antibodies and detected by chemiluminescence.

FACS.

Activated satellite cells and M1 and M2c macrophages were analyzed by fluorescence-activated cell sorting (FACS) as described previously (17, 31). Approximately 2 × 10⁶ cells were incubated in Dulbecco's modified Eagle's medium (DMEM) [supplemented with 2% FBS and 25 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid], and dead cells (positive for propidium iodide staining) that were ∼1% and were excluded from all FACS analysis. For satellite cell quantification from a heterogeneous cell population, cells were immunostained with antibodies against CD45, CD31, CD56/Sca-1, and Ter-119 for negative selection (all phycoerythrin [PE] conjugated; eBiosciences) and with α7β1-integrin (MBL International) for positive selection. A tandem conjugate of R-PE (Alexa 647; Molecular Probes) was used as a secondary antibody against α7β1-integrin. Macrophages were quantified from heterogeneous cell population by selection of F4/80⁺ (PerCP Cy5.5 conjugated; eBiosciences) cells against negative selection by CD56/Sca-1 and Ter-119 (all PE conjugated; eBiosciences). From F4/80⁺ cells, CD11c⁺ (allophycocyanin [APC] conjugated; eBiosciences) M1 and CD206⁺ (fluorescein isothiocyanate [FITC] conjugated; Biolegend) M2c macrophages were isolated. FACS analysis was performed on a C6 Accuri cytometer equipped with three lasers. The output data were processed, and plots were prepared using FCS Express 4 RUO software (De Novo Software).

EMSA.

DNA binding of NF-κB was measured by performing electrophoretic mobility shift assay (EMSA) as previously detailed (49). In brief, 20 μg of nuclear extracts prepared from control or cardiotoxin (CTX)-injected TA muscle was incubated with 16 fmol of ³²P end-labeled NF-κB consensus oligonucleotide (Promega) at 37°C for 30 min, and the DNA-protein complex was resolved on a 7.5% native polyacrylamide gel. The radioactive bands from the dried gel were visualized and quantified by PhosphorImager (GE Health Care) using ImageQuant TL software.

qRT-PCR.

RNA isolation and quantitative real-time PCR (qRT-PCR) were performed using a method as previously described (49). The sequence of the primers is described in Table S1 in the supplemental material.

Statistical analyses.

Results are expressed as means ± standard deviations (SD). Statistical analyses used Student's t test to compare quantitative data populations with normal distribution and equal variance. A P value of <0.05 was considered statistically significant unless otherwise specified.

RESULTS

TRAF6 negatively regulates regeneration of myofibers upon injury.

An acute injury to skeletal muscle is followed by a well-orchestrated series of events which facilitate rapid repair and regeneration of injured muscle (11, 36). We first investigated how the expression of TRAF6 is affected in skeletal muscle in response to injury. Wild-type mice were given intramuscular injection of saline alone or cardiotoxin (CTX) in the tibial anterior (TA) muscle. Western blot analysis showed that the levels of TRAF6 protein dramatically induced in CTX-injected TA muscle at day 5 postinjury (Fig. 1A). The increased levels of TRAF6 protein was potentially due to its increased transcription, because mRNA levels of TRAF6 (normalized with Myh4) were also found to be significantly increased in CTX-injected TA muscle compared to those in contralateral muscle injected with saline alone (Fig. 1B).

Fig 1.

Ablation of TRAF6 improves skeletal muscle regeneration in mice. TA muscle of 8-week-old WT mice was injected with 100 μl of saline alone or containing 10 μM cardiotoxin (CTX) and analyzed at day 5. (A) Representative immunoblots of TRAF6 and an unrelated protein tubulin in saline- and CTX-injected tibial anterior (TA) muscle. (B) Transcript levels of TRAF6 at day 5 after CTX injection in TA muscle measured by qRT-PCR. Myh4 was used as an endogenous normalizing gene. (C) TRAF6^f/f and TRAF6^mko mice were injected with saline or CTX in TA muscle followed by their isolation and analyses at different time points. Representative photomicrographs of H&E-stained transverse sections of saline- or CTX-injected TA muscle at different time points. Scale bar, 20 μm. (D) Frequency distribution histograms representing the fiber cross-sectional area (CSA) in TA muscle of TRAF6^f/f and TRAF6^mko mice at day 5 after CTX injection. (E) Average CSA of myofibers in TA muscle of TRAF6^f/f and TRAF6^mko mice at day 5 after CTX injection. (F) Quantification of myofiber containing more than one centrally nucleated fiber (CNF) in TA muscle sections at 5 day after CTX injection of TRAF6^f/f and TRAF6^mko mice. Error bars represent standard deviation (SD). n = 4 in each group. *, P < 0.01, values significantly different from CTX-injected TA muscle of TRAF6^f/f mice. (G) TA muscle of WT mice was electroporated with pcDNA3 or pcDNA3-TRAF6 plasmids. After 7 days, the muscle was injected with 100 μl of 10 μM CTX solution followed by their isolation and performing H&E staining. Representative photomicrographs of H&E-stained transverse sections at days 5 and 7 after CTX injection are presented here. n = 3 at each time point. Scale bar, 20 μm.

We next sought to determine the role of TRAF6 in skeletal muscle regeneration. To specifically delete TRAF6 in differentiated myofibers, floxed TRAF6 (TRAF6^f/f) mice were crossed with muscle creatine kinase (MCK)-Cre mice to obtain muscle-specific TRAF6-knockout (henceforth TRAF6^mko) mice as previously detailed (49). The levels of TRAF6 protein (but not other TRAFs) are considerably reduced, specifically in the skeletal muscle of TRAF6^mko mice compared to that in littermate TRAF6^f/f mice (49). Satellite cells prepared from TRAF6^f/f and TRAF6^mko mice showed no difference in the protein levels of TRAF6 (see Fig. S1A in the supplemental material), suggesting that TRAF6 protein is reduced in differentiated myofibers but not in muscle progenitor cells of TRAF6^mko mice. TA muscles of 8-week-old TRAF6^mko mice and their littermate TRAF6^f/f mice were given intramuscular injection of saline alone or with CTX, followed by isolation of the TA muscle at different time points and processing for hematoxylin and eosin (H&E) staining. Intramuscular injection of CTX caused equal necrosis in the TA muscle of both TRAF6^f/f and TRAF6^mko mice examined at day 2 after CTX injection (data not shown). Interestingly, regeneration of TA muscle was dramatically improved in TRAF6^mko mice compared to that in TRAF6^f/f littermates (Fig. 1C). TA muscle of TRAF6^mko mice contained the majority of newly formed centronucleated fibers (CNF) and reduced cellular infiltrate at day 5 after CTX injection. Improved regeneration in TRAF6^mko mice was also evident at 10 days and 21 days after CTX injection (Fig. 1C). Morphometric analyses of CTX-injected TA muscle sections showed about 36% improvement in average fiber cross-sectional area (CSA) in TRAF6^mko mice compared to that in TRAF6^f/f littermates (Fig. 1D and E). Moreover, the number of fibers containing two or more centrally located nuclei was significantly higher in TRAF6^mko mice than in TRAF6^f/f mice after 5 days of CTX injection (Fig. 1F), further suggesting accelerated regeneration of injured myofibers in TRAF6^mko mice.

We next studied the effects of overexpression of TRAF6 protein on adult myofiber regeneration. The left-side TA muscle of C57BL/6 mice was electroporated with vector (pcDNA3) alone, whereas the right side was electroporated with plasmid encoding wild-type TRAF6 cDNA. Protein levels of TRAF6 were higher in TA muscle transfected with TRAF6 cDNA than in that transfected with pcDNA3 alone (see Fig. S1B in the supplemental material). However, there was no overt phenotype in TRAF6-transfected TA muscle compared to that transfected with vector alone in unchallenged conditions, studied 7 days postelectroporation (see Fig. S1C in the supplemental material). Next, TA muscle was injected with CTX followed by analysis of muscle regeneration at 5 days and 7 days by H&E staining. Interestingly, overexpression of TRAF6 inhibited the regeneration of TA muscle (Fig. 1G). The TRAF6 cDNA-transfected TA muscle contained a considerably increased amount of cellular infiltrate (darkly stained nuclei of inflammatory cells) both at 5 days and 7 days after CTX injection (Fig. 1G). In addition, the number of regenerating myofibers and the fiber cross-sectional area were noticeably reduced in TRAF6-transfected myofibers (Fig. 1G). These results indicate that TRAF6 inhibits the regeneration of adult myofiber upon injury.

Depletion of TRAF6 leads to early restoration of muscle architecture in response to injury.

A shift from the degenerative to the regenerative stage is followed by transition of myogenic cells through expression of specific transcription factors and related genes. This pattern also mimics the embryonic development of skeletal muscle (11). CTX injection in mouse skeletal muscle stimulates the expression of MyoD in satellite cells by 2 days. Thereafter, a decline in MyoD expression and an increase in myogenin expression occur by day 3 postinjury, followed by a consequential and persistent elevation of an embryonic form of myosin heavy chain (eMyHC) (64, 65). Furthermore, as a regenerating muscle progresses toward normal architecture, embryonic isoform of MyHC is replaced by adult isoform. To investigate whether depletion of TRAF6 causes any change in the temporal expression pattern of these markers, we further examined CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice. Immunostaining revealed more uniform and abundant expression of eMyHC in TA muscle of TRAF6^mko compared to TRAF6^f/f mice at day 5 after CTX injection (Fig. 2A). Western blot analyses also showed increased levels of eMyHC in the TA muscle of TRAF6^mko mice compared to those in TRAF6^f/f mice 5 days after CTX injection (Fig. 2B). By performing immunoblotting with MF20 antibody, we measured an adult fast-type isoform of MyHC (i.e., MyHCf) protein. A noticeable improvement in the level of MyHCf was observed in TA muscle of TRAF6^mko mice compared to that in TRAF6^f/f mice (Fig. 2B). Western blot analysis showed that TRAF6 levels were reduced in uninjured TA muscle of TRAF6^mko mice compared to levels in TRAF6^f/f mice. Furthermore, TRAF6 levels in CTX-injected TA muscle of TRAF6^mko mice was ∼50% less compared to levels in corresponding TRAF6^f/f mice (Fig. 2B), suggesting that infiltrating cells and myofibers contribute almost equally to the increased levels of TRAF6 in CTX-injected TA muscle.

Fig 2.

Ablation of TRAF6 accelerates restoration of muscle architecture after injury. TRAF6^f/f and TRAF6^mko mice were injected with saline or CTX in the TA muscle followed by their isolation and analyses at day 5. (A) Transverse sections of CTX-injured TA muscle from TRAF6^f/f and TRAF6^mko mice stained with anti-embryonic myosin heavy chain (anti-eMyHC) or isotype control (mouse IgG). (B) Western blot analysis of expression levels of eMyHC, adult MyHC fast type (MyHCf), TRAF6, and tubulin in saline- or CTX-injected TA muscle from TRAF6^f/f and TRAF6^mko mice. (C) Transcript levels of IGF-1 and myogenin in TA muscle of TRAF6^f/f and TRAF6^mko mice measured at day 6 after CTX injection. Error bars represent SD. n = 4 in each time point. *, P < 0.01, values significantly different from CTX-injected TA muscle of TRAF6^f/f mice.

Insulin growth factor 1 (IGF-1) is a major growth factor which induces skeletal muscle regeneration through augmenting the proliferation and differentiation of myogenic cells. We next investigated whether signaling through TRAF6 also affects the expression of IGF-1 in injured skeletal muscle. As shown in Fig. 2C, transcript levels of IGF-1 were significantly higher in CTX-injected TA muscle of TRAF6^mko mice than in that of TRAF6^f/f mice (Fig. 2C). Moreover, increased mRNA levels of myogenin in CTX-injected TA muscle of TRAF6^mko mice affirmed enhanced muscle regeneration in these mice compared to that in TRAF6^f/f mice at 6 days after CTX-mediated injury (Fig. 2C). Collectively, these results suggest that depletion of TRAF6 in adult myofibers accelerates the muscle regenerative program in response to injury.

Inhibition of TRAF6 promotes satellite cell activation in injured myofibers in vivo.

Muscle injury is followed by the activation of satellite cells, which is a prerequisite for induction of an efficient regeneration program in injured muscle (11, 36). We next investigated whether TRAF6-mediated signaling affects the activation of satellite cells in injured myofibers. Pax7 is a marker of both quiescent and activated satellite cells (31, 55). We first performed immunostaining for Pax7 to evaluate the number of satellite cells in TA muscle of TRAF6^f/f and TRAF6^mko mice. There was no noticeable difference in the number of Pax7⁺ cells between uninjured TA muscle of TRAF6^f/f and TRAF6^mko mice. However, the number of Pax7⁺ cells per unit area was considerably higher in TA muscle of TRAF6^mko than in that of TRAF6^f/f mice at 5 days after CTX injection (Fig. 3A and B). By performing qRT-PCR, we also measured transcript levels of Pax7 in TA muscle of TRAF6^f/f and TRAF6^mko mice. Consistent with immunostaining results, we found there was no significant difference in the mRNA levels of Pax7 in uninjured TA muscle of TRAF6^f/f and TRAF6^mko mice (Fig. 3C). However, the transcript levels of Pax7 were significantly increased in CTX-injected TA muscle of TRAF6^mko mice (∼4.5-fold) compared to those of TRAF6^f/f mice (∼2-fold) (Fig. 3C). A comparable basal level of expression of Pax7 in uninjured mice suggests that TRAF6 depletion does not influence satellite cell formation. In contrast, the higher expression of Pax7 in injured TA muscle of TRAF6^mko mice could be attributed to the increased proliferation of activated satellite cells from which a subpopulation reenters into quiescence (32).

Fig 3.

Ablation of TRAF6 promotes the activation of satellite cells during muscle regeneration. Three-month-old TRAF6^f/f and TRAF6^mko mice were injected with saline or CTX in TA muscle and analyzed after 5 days. (A) Representative photomicrographs after immunostaining of transverse muscle sections with Pax7 antibody. Nuclei were identified by costaining with DAPI. Arrows point to Pax7⁺ cells; (B) average number of Pax7⁺ cells per field (∼0.15 mm²) in TA muscle of TRAF6^f/f and TRAF6^mko mice injected with saline and CTX after 5 days; (C) transcript levels of Pax7 in TA muscle injected with saline or CTX after 5 days from TRAF6^f/f and TRAF6^mko mice, measured by qRT-PCR; (D) FACS analysis of saline- or CTX-injected TA muscle for α7β1-integrin-positive activated satellite cells in TRAF6^f/f and TRAF6^mko mice. Representative dot plots are shown. Negative selection antibodies (CD45, CD31, Ter119, Sca-1) are boxed in red, whereas positive selection antibody (α7β1-integrin) is boxed in blue. (E) Quantification of activated satellite cells in saline- or CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice by FACS. Error bars represent SD. n = 6 in each group. *, P < 0.01, values significantly different from contralateral saline-injected TA muscle; #, P < 0.01, values significantly different from CTX-injected TA muscle of TRAF6^f/f mice.

A unique combination of cell surface markers (CD45⁻, CD31⁻, Ter119⁻, Sca-1⁻, and α7-β1 integrin⁺) identify satellite cells in adult mouse skeletal muscle and allow their direct quantification by a fluorescence-activated cell sorting (FACS) technique (9). To further evaluate whether signaling through TRAF6 affects the activation of satellite cells in injured muscles, we also performed FACS analysis. The gating strategy for quantification of satellite cells by FACS is depicted in Fig. S2A in the supplemental material. There was no difference in the numbers of satellite cells in uninjured TA muscle of TRAF6^f/f and TRAF6^mko mice. However, intramuscular injection of CTX significantly increased the number of satellite cells in the TA muscle of both TRAF6^f/f and TRAF6^mko mice measured at day 5 (Fig. 3D and E). Furthermore, the number of satellite cells was significantly higher in CTX-injected TA muscle of TRAF6^mko mice than in littermate TRAF6^f/f mice (∼8.5% in TRAF6^f/f versus ∼13% in TRAF6^mko) (Fig. 3D and E).

Ablation of TRAF6 promotes activation and self-renewal of satellite cells in single myofiber cultures.

A suspension culture of myofiber explants represents an ex vivo model that mimics muscle injury in vivo with respect to satellite cell activation, proliferation, and differentiation (16, 31, 55). Upon isolation, each myofiber is associated with a fixed number (Pax7⁺/MyoD⁻) of satellite cells resting in quiescence. At around 24 h in culture, satellite cells undergo their first round of cell division, through upregulating MyoD (Pax7⁺/MyoD⁺) and proliferating to form cell aggregates. Cells then either terminally differentiate (Pax7⁻/MyoD⁺) or self-renew (Pax7⁺/MyoD⁻) (31). Consistent with our in vivo results, immunostaining of freshly isolated myofibers from EDL muscle of TRAF6^mko and their TRAF6^f/f littermates revealed comparable numbers of (Pax7⁺/MyoD⁻) cells and negligible levels of MyoD expression (data not shown). After 72 h in suspension culture, a dramatic increase was observed in the number of Pax7⁺/MyoD⁻ as well as Pax7⁺/MyoD⁺ cells in TRAF6^mko compared to TRAF6^f/f mice (Fig. 4A to E), accompanied by an upregulation of cells expressing the proliferation marker Ki67. Moreover, the number of cells per cellular aggregate was also increased, characterized by an upregulation of Pax7⁺/MyoD⁻ cells (Fig. 4D to F). Further analysis of clusters on myofibers in suspension cultures showed that while there was a significant increase in the proportion of Pax7⁺/MyoD⁻ cells (Fig. 4F), there was no significant difference in the distribution of Pax7⁺/MyoD⁺ cells (Fig. 4G) in cellular aggregates of TRAF6^f/f and TRAF6^mko myofibers, suggesting that while ablation of TRAF6 increases the proliferation and self-renewal, it also favors restoration of the satellite cell pool by a significant margin.

Fig 4.

Ablation of TRAF6 promotes proliferation and self-renewal of satellite cells. Single myofiber cultures were established from EDL muscle of TRAF6^f/f and TRAF6^mko mice. (A and B) After 72 h, myoblast clusters on the single myofibers were labeled with antibodies against Pax7 and MyoD. Nuclei were counterstained with DAPI. Representative merged images of Pax7, MyoD, and DAPI staining are presented here. (C) Average number of clusters (containing >4 cells per cluster) per myofiber of TRAF6^f/f and TRAF6^mko calculated from 35 myofibers in each group. (D) Number of myoblasts per cluster in TRAF6^f/f and TRAF6^mko mice (n = 22). (E) In a separate experiment, myoblasts were also stained with proliferation marker Ki67 and DAPI, and the number of Ki67⁺ cells per cluster was enumerated. (F and G) Percentage of self-renewing (Pax7⁺ MyoD⁻) and proliferating (Pax7⁺/MyoD⁺) myoblasts in TRAF6^f/f and TRAF6^mko myoblast colonies (calculated from 22 cultured colonies in each group). *, P < 0.01, values significantly different from TRAF6^f/f mice.

TRAF6 inhibits Notch signaling in regenerating myofibers.

Notch signaling is an important regulator of cell proliferation, cell fate determination, and asymmetric cell division during embryogenesis. Moreover, the role of Notch signaling in orchestrating satellite cell activation in regenerative adult muscle is well documented (8). To determine whether the accelerated regenerative phenomenon observed in TRAF6-ablated skeletal muscle is brought about in a Notch-dependent manner, transcript levels of a set of Notch target genes were analyzed using a qRT-PCR technique. A significant increase in the mRNA level of Hes1, Hes6, Hey1, HeyL, and Nrarp was observed in the CTX-injected TA muscle of TRAF6^mko mice compared to that in TRAF6^f/f mice (Fig. 5A). Additionally, transcript levels of Notch3 receptor (Fig. 5B) and Notch ligands DLL1, DLL2, Jagged1, and Jagged2 (Fig. 5C) were upregulated in CTX-injected TA muscle of TRAF6^mko mice compared to those in TRAF6^f/f mice. Western blot analysis also showed that the protein levels of Jagged2 were ∼2.2-fold and those of DLL1 were ∼1.7-fold higher in regenerating TA muscle of TRAF6^mko mice than in TRAF6^f/f mice (Fig. 5D). Notch signaling involves the interaction between two neighbor cells, one expressing Notch ligand and another expressing Notch receptors (8, 16). Since in our model we depleted TRAF6 specifically in differentiated myofibers, by performing immunostaining, we tested the hypothesis that inhibition of TRAF6 increases the expression of Notch ligands in regenerating myofibers. As shown in Fig. 5E, Jagged2 protein colocalized with eMyHC in CTX-injected TA muscle of both TRAF6^f/f and TRAF6^mko mice. Furthermore, the expression of Jagged2 was higher around injured/regenerating myofibers of TRAF6^mko than in TRAF6^f/f mice (Fig. 5E). Similarly, immunostaining analysis revealed that Jagged2 was expressed in myofibers in suspension cultures, and the level of expression was increased in cultured myofibers from TRAF6^mko mice compared to that in TRAF6^f/f mice (Fig. 5F).

Fig 5.

Muscle-specific inhibition of TRAF6 activates Notch signaling upon injury. TA muscle of TRAF6^f/f and TRAF6^mko mice were injected with cardiotoxin (CTX), and 5 days later the muscles were isolated and processed for qRT-PCR, Western blot analysis, or immunostaining. Relative mRNA levels of Hes1, Hes6, Hey1, HeyL, and Nrarp Notch target genes (A); Notch receptors Notch1, Notch2, and Notch3 (B); and Notch ligands Jagged1, Jagged2, DLL1, and DLL2 in CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice (C). n = 6 in each group. (D) Western blot analysis of Jagged2 and DLL1 protein in saline- or CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice. (E) CTX-injected TA muscle transverse frozen sections were stained for embryonic myosin heavy chain (eMyHC) and Jagged2. Representative photomicrographs presented here demonstrate increased immunostaining for Jagged2 in myofibers of TRAF6^mko mice compared to that of TRAF6^f/f mice. (F) Single myofiber cultures were prepared from TRAF6^f/f and TRAF6^mko mice. After 24 h in cultures, myofibers (n = 12) were stained for Jagged2. (G) Relative mRNA levels of Hes6, HeyL, and Nrarp Notch target genes in satellite cells isolated by the FACS method from CTX-injected TA muscle of TRAF6^f/f (n = 3) and TRAF6^mko (n = 3) mice after 5 days. Bars represent SD. *, P < 0.05, values significantly different from those of corresponding TRAF6^f/f mice.

We also studied the activation of the Notch signaling pathway in satellite cells of TRAF6^f/f and TRAF6^mko mice. TA muscle of TRAF6^f/f and TRAF6^mko mice were given intramuscular injection of CTX for 5 days, and satellite cells were isolated using the FACS method followed by qRT-PCR assay to study the expression levels of Notch target genes. Interestingly, the mRNA levels of Hes6, HeyL, and Nrarp were found to be significantly higher in satellite cells from TRAF6^mko mice than from TRAF6^f/f mice (Fig. 5G), indicating higher activation of the Notch signaling pathway in satellite cells of injured myofibers of TRAF6^mko mice.

Inhibition of Notch signaling blunts the proliferation and self-renewal of satellite cells in myofiber cultures of TRAF6^f/f and TRAF6^mko mice.

Although the role of Notch signaling in satellite cell activation and self-renewal and muscle regeneration has been established using both genetic mouse models and pharmacological inhibitors (14, 16, 31, 32, 52, 54, 56, 60), we further investigated whether the higher levels of activation of Notch pathway are responsible for the increased proliferation of satellite cells in myofibers of TRAF6^mko mice. Previous studies have shown that γ-secretase inhibitor N-[2S-(3, 5-difluorophenyl)acetyl]-l-alanyl-2-phenyl-1,1-dimethylethyl ester-glycine (DAPT) efficiently inhibits the activation of Notch signaling in various cell types, including satellite cells (31, 52). DAPT functions by inhibiting the cleavage of the Notch intracellular domain (NICD) from the transmembrane domain of Notch receptor and hence blocks the downstream Notch signaling. Single myofibers were prepared from TA muscle of TRAF6^f/f and TRAF6^mko mice and treated with DAPT followed by immunostaining for Pax7, MyoD, and/or Ki67. Nuclei were identified by costaining with DAPI (4′,6-diamidino-2-phenylindole). Consistent with published reports, DAPT reduced the number of clusters, the average number of satellite cells per cluster, and the number of Ki67⁺ cells in myofibers of both TRAF6^f/f and TRAF6^mko mice (Fig. 6). Interestingly, treatment with DAPT completely blunted the increased proliferative response of satellite cells observed in cultured myofibers of TRAF6^mko mice. Similarly, the proportion of Pax7⁺/MyoD⁻ and Pax7⁺/MyoD⁺ cells was also dramatically reduced upon treatment with DAPT (Fig. 6). Collectively, these results demonstrate that depletion of TRAF6 in differentiated myofibers induces satellite cell proliferation and self-renewal through the Notch signaling pathway.

Fig 6.

Activation of Notch pathway causes activation and self-renewal of satellite cells on cultured myofibers of TRAF6^mko mice. Single myofiber cultures were established from EDL muscle of TRAF6^f/f and TRAF6^mko mice. After 24 h, myofibers were treated with vehicle alone or along with 10 μm for DAPT. Myoblasts on single myofibers were stained with antibodies against Pax7, MyoD, and/or Ki67, and nuclei were counterstained with DAPI. (A) Average number of clusters (containing >4 cells per cluster) per myofiber of DAPT-treated TRAF6^f/f and TRAF6^mko myofiber cultures (calculated from 18 myofibers in each group). (B) Number of myoblasts per cluster in DAPT-treated myofiber cultures from TRAF6^f/f and TRAF6^mko mice (n = 16). (C) Number of Ki67⁺ cells per cluster. (D and E) Percentage of self-renewing (Pax7⁺ MyoD⁻) and proliferating (Pax7⁺/MyoD⁺) myoblasts in TRAF6^f/f and TRAF6^mko myoblast colonies (calculated from 17 cultured colonies in each group). Bars represent SD. *, P < 0.05, values significantly different from corresponding TA muscle of TRAF6^f/f mice.

Ablation of TRAF6 inhibits the activation of NF-κB transcription factor in regenerating myofibers.

NF-κB is a proinflammatory transcription factor which negatively regulates the regeneration of adult myofibers (37, 43) and is also known to be regulated by Notch signaling (22, 47). We first investigated whether TRAF6 affects the activation of NF-κB in regenerating myofibers. Consistent with published reports (41, 43), DNA-binding activity of NF-κB was significantly increased in CTX-injected TA muscle compared to that in contralateral TA muscle injected with saline alone (Fig. 7A). However, the level of activation of NF-κB was significantly reduced in CTX-injected TA muscle of TRAF6^mko mice compared to that of TRAF6^f/f mice (Fig. 7A and B).

Fig 7.

Depletion of TRAF6 inhibits NF-κB activity and promotes M2c macrophage phenotype in regenerating myofibers. TA muscle of TRAF6^f/f and TRAF6^mko mice were injected with saline or CTX and analyzed at day 5. (A) Representative EMSA gel shows reduced DNA-binding activity of NF-κB transcription factor in CTX-injected TA muscle of TRAF6^mko compared to littermate TRAF6^f/f mice. S, saline; C, cardiotoxin. (B) Quantification of fold change in NF-κB activity in TA muscle of saline- or CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice. n = 6 in each group. (C) FACS analysis for M1 (CD11c⁺) and M2 (CD206⁺) macrophage phenotypes in CTX-injected TA muscle of TRAF6^f/f and TRAF6^mko mice. M1 macrophages (CD11c⁺) are in red, while M2 macrophages (CD206⁺) are in blue. Representative dot plots are shown. Quantification of M1 and M2c macrophages in TA muscle of TRAF6^f/f and TRAF6^mko mice at day 3 (D) and day 6 (E) after CTX injection. (F) Relative mRNA levels of TNF-α, IL-1β, IL-4, and IL-10 in TA muscle of TRAF6^f/f and TRAF6^mko mice measured at day 6 after CTX injection. Error bars represent SD. n = 6 in each group. *, P < 0.01, values significantly different from CTX-injected TA muscle of TRAF6^f/f mice.

We also studied the role of TRAF6 in the activation of Akt and p38 MAPK, which positively regulate skeletal muscle regeneration and growth (26, 51, 53). While the phosphorylation levels of both p38 MAPK and Akt kinase were considerably increased in CTX-injected myofibers, we found no significant difference in the level of phosphorylation of Akt or p38 MAPK between TRAF6^f/f and TRAF6^mko mice (see Fig. S3 in the supplemental material). These results indicate that TRAF6 specifically affects the activation of Notch and NF-κB pathways in injured skeletal muscle.

The innate immune response that starts within hours of muscle injury is critical for efficient skeletal muscle regeneration (59). It has been suggested that the initial inflammatory response in an injured muscle is driven by Th1 cell cytokines, such as tumor necrosis factor alpha (TNF-α) and interleukin 1β (IL-1β), which stimulate the activation of the phagocytic macrophage phenotype (M1), that invade the injured tissue first and stay at an elevated concentration from 24 h to 48 h postinjury, after which their levels start to decline (59). The shift from the phagocytic to regenerative phase in injured skeletal muscle coincides with macrophage phenotype transition from M1 to M2c. M2c macrophages promote muscle regeneration by deactivating the M1 phenotype (38, 59, 61, 62). We investigated whether the improved muscle regeneration in TRAF6^mko mice is also a consequence of a bias in macrophage activation. After 3 days of CTX injection, a time point at which transition in macrophage phenotype occurs (59), we examined percentage of M1 and M2c macrophages in the F4/80⁺ population by the FACS method. The gating scheme for quantification of the M1 and M2c macrophage population is shown in Fig. S2B in the supplemental material. As shown in Fig. 6C and D, the proportion of M1 macrophages (CD11c⁺) was significantly higher in CTX-injected TA muscle of TRAF6^f/f (∼10%) than in that of TRAF6^mko (∼7.7%). In contrast, the proportion of CD206⁺ M2c macrophages was significantly elevated in TA muscle of TRAF6^mko mice (∼4.6%) compared to that in TRAF6^f/f (∼2%) littermates (Fig. 7C and D). We also quantified M1 and M2c macrophage populations at day 6 after CTX injection, a time point at which muscle regeneration is at peak. Although we did not find any noticeable difference in M1 macrophages, the proportion of M2c macrophages was significantly higher in CTX-injected TA muscle of TRAF6^mko mice than in that of TRAF6^f/f mice (Fig. 7E).

Inflammatory cytokines such as TNF-α and IL-1β favor M1 macrophages, whereas anti-inflammatory cytokines IL-4 and IL-10 promote the M2 phenotype in macrophages (38, 59). By performing qRT-PCR, we investigated whether the depletion of TRAF6 affects the mRNA levels of these cytokines in injured skeletal muscle. As shown in Fig. 7F, the transcript levels of TNF-α and IL-1β were significantly reduced in CTX-injected TA muscle of TRAF6^mko mice compared to that in TRAF6^f/f mice 6 days after CTX injection. In contrast, the mRNA levels of IL-4 were increased, whereas IL-10 was not affected in CTX-injected TA muscle of TRAF6^mko (Fig. 7F). Consistent with reduced expression of inflammatory cytokines, we also found that the level of fibrosis was considerably reduced in CTX-injected TA muscle of TRAF6^mko mice compared to that in TRAF6^f/f mice assayed by performing Masson's trichrome staining (see Fig. S4 in the supplemental material).

DISCUSSION

Skeletal muscle regeneration involves activation of a complex array of signaling proteins not only in satellite cells and myoblasts but also in regenerating myofibers. However, except NF-κB, where its targeted inhibition in adult/differentiated myofibers improved regeneration (43), the majority of the studies have been performed employing myoblast- or satellite cell-specific knockout mice or global knockout mouse models which made no distinction between signals originating in muscle progenitor cells and injured myofibers. Moreover, the initial events which govern the activation of downstream signaling pathways in injured myofibers remain poorly understood. In the present study, we provide genetic evidence that signaling through TRAF6 negatively regulates adult myofiber regeneration. We have also uncovered a previously unrecognized link between TRAF6, Notch signaling, and skeletal muscle regeneration. Our results demonstrate that the inhibition of TRAF6 upregulates the expression of Notch ligands, leading to enhanced Notch-driven activation of satellite cells. Our study also shows that blocking TRAF6 in differentiated myofibers inhibits the activation of NF-κB and increases levels of promyogenic M2c macrophages in regenerating skeletal muscle potentially through cross talk with the Notch pathway.

Following muscle injury, quiescent satellite cells present in basal lamina get activated, which proliferate and finally fuse with injured myofibers leading to regeneration and repair (11, 18). Notch signaling is critical not only for the activation of satellite cells but also for their terminal differentiation into myofibers and to maintaining the satellite cell pool (14, 16, 31, 32, 52, 54, 56, 60). We found that the depletion of TRAF6 specifically in differentiated myofibers significantly increases the number of satellite cells upon injury, leading to rapid and faster restoration of muscle architecture (Fig. 1, 2, and 3). Similarly, satellite cells associated with cultured TRAF6^mko myofibers displayed increased proliferative potential accompanied by upregulation of Pax7⁺/MyoD⁻ and Pax7⁺/MyoD⁺ satellite cells (Fig. 4). Transcriptional analysis of injured myofibers from TRAF6^mko compared to their TRAF6^f/f littermates denoted an augmentation of a Notch signaling cascade evident by enhanced levels of Notch ligands (DLL1, DLL2, Jagged1, and Jagged2), Notch3 receptor, and a subset of Notch target genes (Hes1, Hes6, Hey1, HeyL, and Nrarp genes). This suggests that the inhibition of TRAF6 leads to the production of certain factors from regenerating myofibers and/or creates a muscle microenvironment which augments Notch signaling, resulting in the activation of resident satellite cells. Alternatively, the increased expression of Notch ligands on injured myofibers itself could be sufficient to activate Notch signaling in satellite cells residing on these myofibers. The latter possibility is strongly supported by our results demonstrating that the expression of Jagged2 was noticeably higher in the periphery of regenerating myofibers of TRAF6^mko mice (Fig. 5C, D, and E). Furthermore, freshly isolated satellite cells from CTX-injected TA muscle of TRAF6^mko mice showed increased mRNA levels of Notch target genes, suggesting increased Notch signaling in satellite cells (Fig. 5G). Although the current study identifies TRAF6 signaling as being a negative regulator for the expression of Notch ligands on injured myofibers, one can next inquire as to the mechanism by which TRAF6 suppresses the expression of Notch ligands. This is an area of interest for future investigation.

Several molecules, such as IGF-1, fibroblast growth factor, and hepatocyte growth factor, have now been identified which affect the proliferation and/or differentiation of muscle progenitor cells (11, 18). Among them, IGF-1 is a well-known growth factor which stimulates both the proliferation and differentiation of muscle progenitor cells in vivo and in vitro (44). It has also been reported that muscle-specific overexpression of IGF-1 augments regeneration of myofibers in response to injury (50). Interestingly, the expression of IGF-1 was significantly higher in injured myofibers of TRAF6^mko mice than in those of TRAF6^f/f mice (Fig. 2C), suggesting that the improved muscle regeneration in TRAF6^mko mice could also be a result of increased production of IGF-1.

NF-κB is one of the important signaling pathways activated through TRAF6-dependent mechanisms in response to various cytokines, growth factors, and recruitment of Toll-like receptors (3, 12). NF-κB activation can occur via either the canonical or alternative pathway (24). In the absence of activating stimuli, NF-κB dimers are retained in the cytoplasm by binding to specific inhibitors—the inhibitors of NF-κB (IκBs). The classical pathway is IKKβ and IKKγ dependent, and NF-κB activation occurs through the degradation of IκB proteins (24, 37). Activated IKK phosphorylates NF-κB-bound IκB proteins and targets them for polyubiquitination and rapid degradation. Proinflammatory cytokines such as TNF-α activate NF-κB through IKKβ-mediated site-specific phosphorylation and subsequent ubiquitination and degradation of inhibitory protein IκBα by the 26S proteasome. NF-κB complexes liberated from IκB inhibitory proteins then translocate to the nucleus, leading to transcriptional activation of several target genes (24). In addition to this classical activation mechanism involving IκB degradation, posttranslational modifications of p65 by phosphorylation, acetylation, and ubiquitination have been shown to modulate the trans-activation potential of NF-κB (24). Inhibition of NF-κB in differentiated muscle improves its regeneration in response to CTX-mediated injury (43) and in the mdx model of Duchenne muscular dystrophy (2). Coincidently, we found that the DNA-binding activity of NF-κB (Fig. 7A and B) and the transcript levels of TNF-α and IL-1β (Fig. 7F) were reduced in injured skeletal muscle of TRAF6^mko mice. These results are also in agreement with our recently published report demonstrating that TRAF6 mediates the activation of NF-κB in skeletal muscle in response to catabolic stimuli, such as denervation and cancer cachexia (49).

Cross talk between NF-κB and Notch signaling pathways has been implicated in different cellular contexts (22, 47). Cytoplasmic sequestration of p65 by IκBα was shown to both translocate nuclear corepressors SMRT/N-CoR (silence mediator for retinoic acid and thyroid receptors/nuclear receptor corepressor) to the cytoplasm and upregulate transcription of Notch-dependent genes (20). Moreover, p65 and IκBα are able to directly bind SMRT, and this interaction can be inhibited in a dose-dependent manner by the CREB binding protein (CBP) coactivator and after TNF-α treatment, suggesting that stimuli that promote IκBα degradation, p65 acetylation, and NF-κB activation, such as TNF-α, inhibit Notch-dependent transcriptional activity (20). More recently, it has been reported that TNF-α is capable of inhibiting Notch1 in satellite cells and C2C12 myoblasts in an NF-κB-dependent manner (1). Thus, the reduced expression of TNF-α and suppression of NF-κB activity may be another mechanism for the increased Notch signaling in regenerating myofibers of TRAF6^mko mice (Fig. 7A, B, and F).

While the role of TRAF6 in innate immune response has been extensively studied, there is a dearth of information on its mediation in activation of different phenotypes of macrophages. Accumulating evidence suggests that macrophage phenotype transition is critical for the regeneration of skeletal muscle upon injury, because proinflammatory (M1) and anti-inflammatory (M2c) macrophages exert antagonistic effects on myogenesis (59). Initial activation of proinflammatory macrophages is mediated by inflammatory cytokines, which is not influenced by the factors produced by muscle cells. However, as the regeneration progresses, cytokines produced by muscle cells may also contribute to prolonged activation of macrophages (59). TNF-α and IL-1β are two inflammatory cytokines produced both by M1 macrophages and skeletal muscle cells (33, 41, 59). Increased levels of these cytokines facilitate the activation of M1 macrophages and inhibit transition from the M1 to the M2 phenotype. Our results demonstrate that the expression of both IL-1β and TNF-α are decreased in injured myofibers of TRAF6^mko mice, and this may be responsible for the reduced activation of M1 macrophages (Fig. 7F). In contrast, IL-4 and IL-10 are predominately anti-inflammatory cytokines which promote the M2c phenotype of macrophages and induce the proliferation of satellite cells (38, 61, 62). While we did not find any major difference in mRNA levels of IL-10, the expression of IL-4 was increased in injured myofibers of TRAF6^mko compared to that in TRAF6^f/f mice (Fig. 7F). This suggests that the inhibition of TRAF6 limits the levels of inflammatory cytokines, which hastens the appearance of M2c macrophages, resulting in increased proliferation of myogenic cells and rapid restoration of myofiber architecture.

It is also of interest to note that some of the pathways purported to be involved in skeletal muscle regeneration or synthesis of new myofibers were not affected by depletion of TRAF6. Ablation of TRAF6 did not have any major effect on the phosphorylation of Akt kinase and p38 MAPK (see Fig. S3 in the supplemental material). While these data are in contrast to previous reports of TRAF6-dependent activation of Akt (66) and p38 MAPK (39), they highlight TRAF6-mediated differential activation of downstream signaling pathways (5). This context-dependent role of TRAF6 supports our inference that in injury-induced regeneration of skeletal muscle, TRAF6 acts as a negative regulator of myogenesis.

While the TRAF6-mediated signaling in differentiated myofibers inhibits their regeneration upon injury, it is noteworthy that the role of TRAF6 could be quite different in muscle progenitor cells. A published report suggests that siRNA-mediated knockdown of TRAF6 in cultured C2C12 myoblasts inhibits their proliferation as well as differentiation into multinucleated myotubes (45). Using the siRNA electroporation approach, Xiao et al. (63) have recently investigated the in vivo role of TRAF6 in skeletal muscle regeneration. In contrast to our study, they reported that TRAF6 is essential for skeletal muscle regeneration, because its depletion through the siRNA-mediated technique inhibited regeneration of injured myofibers (63). However, in their study, they performed electroporation of TRAF6 siRNA in TA muscle 1 day after CTX injection and continually performed the same procedure every day before studying muscle regeneration. This implies that they depleted TRAF6 in all cell types, including satellite cells and immune and other cell types which infiltrate myofibers after injury (63). While the specific in vivo role of TRAF6 in satellite cell proliferation and differentiation requires further investigation using genetic mouse models, in the present study, we have shown that TRAF6 signaling that specifically originates in differentiated myofibers attenuates the muscle regeneration program in response to injury.

In summary, TRAF6-mediated regulation of muscle regeneration unveiled by this study provides an unanticipated link between Notch signaling, satellite cell activation, and muscle formation in an injured tissue microenvironment. Considering the importance and limited availability of therapeutic interventions that can influence the balance between inflammation and myogenesis in pathological conditions, such as muscular dystrophy, we believe that interventions targeting TRAF6-mediated signaling will enhance the ability to improve pathological conditions in inflammatory muscle disorders.