Self-renewal and Differentiation of Muscle Satellite Cells Are Regulated by the Fas-associated Death Domain

Background: The balance between self-renewal and differentiation in muscle stem cells is critical for tissue homeostasis.

Results: Fas-associated death domain (FADD) phosphorylation mutation regulates muscle satellite cell cycle progression, inhibits muscle stem cell differentiation, and promotes muscle stem cell self-renewal.

Conclusion: FADD regulates muscle stem cell fate decisions.

Significance: The characterization of proteins involved in fate decisions in muscle stem cells is crucial for muscle stem cell biology.

Abstract

Making the decision between self-renewal and differentiation of adult stem cells is critical for tissue repair and homeostasis. Here we show that the apoptotic adaptor Fas-associated death domain (FADD) regulates the fate decisions of muscle satellite cells (SCs). FADD phosphorylation was specifically induced in cycling SCs, which was high in metaphase and declined in later anaphase. Furthermore, phosphorylated FADD at Ser-191 accumulated in the uncommitted cycling SCs and was asymmetrically localized in the self-renewing daughter SCs. SCs containing a phosphoryl-mimicking mutation at Ser-191 of FADD (FADD-D) expressed higher levels of stem-like markers and reduced commitment-associated markers. Moreover, a phosphoryl-mimicking mutation at Ser-191 of FADD suppressed SC activation and differentiation, which promoted the cycling SCs into a reversible quiescent state. Therefore, these data indicate that FADD regulates the fate determination of cycling SCs.

Introduction

Muscle satellite cells (SCs)² are responsible for both postnatal myogenesis and adult muscle regeneration. SCs are abundant in perinatal muscle, accounting for ∼30% of the sublaminar nuclei, and decline to less than 5% of the sublaminar nuclei in adult muscle. Most of the SCs in newborn mice are primed for differentiation to support postnatal muscle development. The remaining, less committed SCs that become quiescent serve as the adult muscle stem cell pool (1, 2).

Upon muscle injury, the quiescent adult SCs are activated to repair the damage (3). Activated SCs, which are identified by the expression of the myogenic regulatory factor MyoD, proliferate to generate a sufficient number of committed myogenic progenitors (4). The progenitors that withdraw from the cell cycle enter into an irreversible G₀ state and engage in terminal differentiation. The mechanisms promoting the irreversible exit from the cell cycle and commitment to terminal differentiation have been elucidated (5). In addition to differentiation, another critical function of the cycling SCs is to replenish the stem cell pool (6). A subset of myofiber-associated cycling myogenic precursor cells exits from the cell cycle and becomes quiescent without expressing commitment and differentiation markers such as myogenin and desmin, respectively (7–9). Moreover, a subpopulation of these “reserve cells” is also observed in cultures of myogenic cells, which are capable of generating both committed and new reserve cell progenies (10, 11). These findings indicate that a subpopulation of the cycling SCs commit to a reversible quiescent fate to contribute to the self-renewal of muscle stem cells. The distinctly quiescent states of a subset of the progeny of the cycling SCs share some of the characteristics of cell cycle machinery. However, the diverse mechanisms involved in the determination of the distinctly quiescent state, which is critical for the fate decisions of the cycling SCs, remain largely unknown (4).

The Fas-associated death domain (FADD) was initially identified as an adapter protein that is critical for death receptor-mediated apoptosis (12, 13). Further studies indicated that FADD is also required for embryonic development and T-cell maturation (14, 15). Recently, FADD has been shown to play a role in regulating embryogenesis, thymus development, and chronic intestinal inflammation by suppressing receptor-interacting protein (RIP)-induced necrosis (16, 17). Interestingly, we have demonstrated previously that a single phosphorylation site in the C terminus of FADD (Ser-191) regulates postnatal growth, the size of the adult thymus, and lymphocyte maturation and that these effects are not due to necrosis, indicating distinct mechanisms for FADD phosphorylation in regulating these processes (18, 19).

In this study, we explored the role of FADD phosphorylation in the fate determination of cycling SCs. We show that a phosphoryl-mimicking mutation in FADD (FADD-D) increased the expression of stem-cell-like markers (Pax7, M-cadherin, CD34, and CXCR4) and decreased the expression of commitment and differentiation markers (desmin and myogenin) in the SCs. FADD-D induced SCs to enter a phase of reversible cell cycle arrest at the G₂/M boundary instead of terminally differentiating or becoming apoptotic. In its role as a cell cycle regulator, FADD phosphorylation is dynamically regulated during cell cycle progression. Moreover, biased distribution of phosphorylated FADD was observed in the daughter SC that retains the template DNA or numb protein. FADD regulates these processes through modulating Notch signaling. Finally, FADD phosphorylation was compromised in aged or dystrophin-deficient cycling SCs. These data reveal a novel mechanism for the fate determination of cycling muscle stem cells.

EXPERIMENTAL PROCEDURES

Mice and Animal Care

The FADD phosphorylation mutant knockin mice (FADD-D in FADD^−/− alleles) were generated as reported previously (19). Age- and gender-matched adult (3–6 months old) FADD-D mice and their littermates were used. Old C57BL/6 mice (>20 months) were housed in the specific pathogen-free (SPF) animal facility. X-linked muscular dystrophic (mdx) and control mice (C57BL/10 background) were purchased from the National Resource Center for Mutant Mice (Nanjing, China). All mice were housed in a specific pathogen-free animal facility in Nanjing Drum Tower Hospital, which is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care in Jiangsu Province, and all animal protocols were approved by the Animal Care and Use Committee of the School of Life Sciences of Nanjing University.

Isolation Culture of Satellite Cells and Myofiber Isolation

The hind limb muscles were minced into small pieces and digested in DMEM with 0.2% type II collagenase (Worthington Biochemicals) in the culture incubator for 40 min. The suspensions were filtered through a 70-μm filter (BD Biosciences). The cells were then collected by centrifugation at 300 × g for 5 min and washed. SCs were then enriched by FACS (FACSAria, BD Biosciences) as described previously (11). The SCs were sorted using the following profile: CD31⁻, CD45⁻, CD11b⁻, propidium iodide (PI), Syndecan 4⁺, and CD34⁺. The purity of the sorted SCs was confirmed by the presence of Pax7 (Fig. 1A). Anti-Pax7 and anti-MHC were purchased from R&D Systems. Anti-HistoneH3 (Ser-10), anti-cleaved Notch-1, anti-phospho-FADD (Ser-191), anti-cdc2 (Tyr-15), anti-cyclin B1, and anti-cleaved caspase 3 were purchased from Cell Signaling Technology. Anti-Lamini-A2 was purchased from Alexia Biochemicals. Anti-DLL1 and anti-Ki67 were purchased from Abcam. Anti-desmin was purchased from Neomarkers. Anti-M-cadherin was purchased from Santa Cruz Biotechnology. Anti-FADD was purchased from Epitomics. Anti-myogenin, anti-MyoD, and anti-syndecan 4 were purchased from BD Biosciences. Anti-CD34 was purchased from eBioscience.

FIGURE 1.

Identification of highly enriched myogenic cells (SC) isolated from hind limb muscle. A, representative FACS plots from SCs isolated from hind limb muscle. Cells gated on CD31⁻/CD45⁻/CD11b⁻/propidium iodide (PI) were used to analyze the expression of Syndecan 4 and CD34. To collect highly enriched SCs, ∼10% of gated cells expressing both Syndecan 4 and CD34 were sorted for further study of SCs. The sorted cells were plated and fixed immediately to examine the expression of Pax7. Immunostaining results (bottom panel) indicate that more than 90% of cells express Pax7. FSC, forward light scatter. B, Regenerating TA muscle 5 days post-injury was used for pFADD immunostaining in the presence of the blocking peptides as indicated. Scale bar = 20 μm. Sections were counterstained with DAPI.

SC-derived primary myoblasts were isolated as described previously, with minor modifications (20). Briefly, the cells were centrifuged through a Percoll gradient (Sigma). The satellite cells were collected from the 20%/60% interface of the Percoll gradient. The purity of satellite cells was confirmed by Pax7 expression. SCs or primary myoblasts were counted and then cultured in 8-well chamber slides (BD Biosciences) that were coated with Matrigel (BD Biosciences), Ham's F-10 nutrient mixture containing 20% FBS (Invitrogen), and 1% penicillin/streptomycin with 5 ng/ml of basic fibroblast growth factor (Cell Signaling Technology). For differentiation, the cells were cultured in low-serum medium (2% horse serum in DMEM). For quantification of differentiating colonies, ∼10,000 SC-derived primary myoblasts were cultured in a 35-mm dish with 3 ml of Methocult M3434 (Stem Cell Technologies) for 14 days. Single myofibers were isolated as described previously (21).

Muscle Injury

Prior to local tissue injury, the mice were anesthetized with isoflurane. The TA muscles were frozen by a 15-s application of a metallic rod precooled in liquid nitrogen. The skin was sutured, and the mice were kept on a warm plate until they recovered. To assess the content of activated SCs in regenerating muscles, the regenerating hind limb muscles were injected at 40 sites with cardiotoxin (20 μm).

EdU

To examine asymmetric cell division, EdU (Invitrogen) was injected as described, with minor modifications (22). Briefly, mice were intraperitoneally injected with EdU five times at 8-hour intervals starting at 72 h post-injury. The mice were sacrificed 24 h after the last injection. EdU detection was performed following the instructions of the Click-iT EdU cell proliferation kit (Invitrogen).

Histology and Immunofluorescence

10-μm-thick frozen sections adhered to Superfrost Plus slides were fixed with a freshly prepared solution of 4% paraformaldehyde in 0.1 m phosphate buffer (pH 7.2) for 10 min. Immunofluorescence was performed as described previously (23). Briefly, slides were washed three times with PBS for 5 min each and permeabilized with 0.2% Triton X-100/PBS for 30 min. To block the unspecific background, sections were incubated with the blocking solution in an M.O.M. (Mouse On Mouse) kit (Vector Lab) following the protocol of the manufacturer. Sections were blocked again by incubation in 15% goat serum for 20 min. Slides were then incubated overnight with primary antibodies diluted in blocking solution. The slides were then washed in PBS for 5 min. Then, the anti-IgG1 isotype-specific secondary antibodies (eBioscience) were diluted in the blocking solution from the M.O.M. kit to label the Pax7, MyoD, and myogenin monoclonal IgG1 antibodies. The fluorescent secondary antibodies (Invitrogen) directed against other species-specific IgG were diluted in blocking solution. The slides were washed and counterstained with DAPI. All of the images were obtained using an Axioplan 2 florescence microscope (Carl Zeiss) equipped with an AxioCam HR camera.

Real-time quantitative RT-PCR

Total RNA was isolated by using TRIzol reagent (Invitrogen). The first-strand cDNAs were directly synthesized using the SuperScript III CellsDirect cDNA synthesis kit (Invitrogen). Real-time quantitative RT-PCR was performed with a StepOne/StepOnePlus^TM real-time PCR system (Applied Biosystems) using SYBR Green PCR Master Mix according to the instructions of the manufacturer.

Statistical Analysis

A minimum of three and as many as six replicates were performed for all of the experiments. The data are presented as the mean ± S.D. Comparisons within groups were conducted using Student's t test with repeated measures. The p values indicated in the figures are as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

RESULTS

FADD Is Specifically Phosphorylated in Mitotic SCs

Previous studies have indicated that FADD phosphorylation is dynamically regulated during cell cycle progression, which peaks at metaphase and then declines in certain cells (24–26). We then determined the phosphorylated FADD (pFADD) levels in cycling SCs. Primary quiescent SCs have been shown to enter the cell cycle and then proliferate in cultures (21, 27). SCs were isolated by FACS (Fig. 1A) and then cultured. As shown in Fig. 2A, coexpression of pFADD and histone H3 phosphorylation at serine 10 (ph3), which is an established marker for M phase (28), was observed in SCs that were in metaphase (Fig. 2A, arrow). pFADD was present at relatively lower levels in SCs that were in late anaphase (Fig. 2A, asterisk). In contrast, pFADD was undetectable in quiescent cells (Fig. 2A). Furthermore, pFADD was specifically induced in activated SCs attached in cultured myofiber, as demonstrated by their MyoD immunostaining in isolated myofibers (Fig. 2B). To examine the pFADD levels in activated SCs in vivo, we induced SC activation by muscle injury. Nearly all of the activated SCs (96 ± 2%, n = 4) (denoted by MyoD expression) in cross-sections of injured TA muscles exhibited a high level of pFADD 2 days post-injury (Fig. 2, C and D). The specificity of the pFADD antibody was confirmed (Fig. 1B). These data indicate that FADD is also specifically phosphorylated in cycling SCs.

FIGURE 2.

FADD phosphorylation in cycling SCs. A, SCs were isolated from uninjured muscles and cultured for 48 h. Cells were then stained with antibodies as indicated. ph3, histone H3 phosphorylation at serine 10. Cells in metaphase (arrow) and anaphase (asterisk) were denoted. B, single myofibers from uninjured muscle were cultured for 36 h. C, transverse sections were collected from regenerating TA muscle 2 days post-injury. Samples were immunostained with antibodies as indicated (n = 4). D, quantification of MyoD⁺/pFADD⁺ cells in regenerating TA muscle in C. Data are presented as the mean ± S.D. Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

FADD-D Inhibits SC Proliferation

Because FADD phosphorylation was induced in the cycling SCs and FADD phosphorylation has been shown to regulate the proliferation in other cell types (24–26), we were interested in whether FADD phosphorylation also regulates SC proliferation and activation. Similarly, the proliferation rate in FADD-D SCs was decreased compared with that of control SCs (Fig. 3A). Consistent with the suppressed proliferation of FADD-D SCs, MyoD levels were also reduced in the regenerating TA muscles of FADD-D mice compared with those of the controls (Fig. 3B), indicating that the activation of FADD-D SCs was compromised. FADD phosphorylation has been shown to delay mitosis entry and suppress proliferation (24–26). Thus, we examined the phosphorylation levels of cdc2Y15, whose dephosphorylation is required for mitosis phase entry. As expected, SCs isolated from the FADD-D animals exhibited increased cdc2Y15 phosphorylation levels compared with control SCs (62 ± 9% and 24 ± 8%, respectively; n = 3) (Fig. 3, C and D). Consistent with this result, less cyclin B1 and phosphorylated ph3 was observed in cultured SCs isolated from FADD-D animals compared with SCs of the control (14 ± 6%, 62 ± 10%, 31 ± 3%, and 5 ± 1% respectively; n = 3) (Fig. 3, C and D). These data indicate that the entry into M phase is suppressed in the mutant SCs. Indeed, the regenerating areas in FADD-D TA muscles exhibited reduced Pax7⁺/ph3⁺ cells compared with that of the littermate controls (4 ± 0.5% and 21 ± 2%, respectively) (Fig. 3E, right).

FIGURE 3.

FADD regulates the proliferation of SCs. A, average number of cells per colony of 10 single-cell-derived colonies. Data points represent days 1, 2, 3, and 4 of culture. Data are mean ± S.D. (n = 3). ***, p < 0.001. B, MyoD was visualized in regenerating TA muscle of control and FADD-D animals 5 days post-injury (n = 3). C, SCs isolated from littermate controls and FADD-D mice were cultured for 72 h. Cdc2/CDK1 phosphorylated at tyrosine 15 (cdc2Y15) and cyclin B1 were stained (left and center columns). Ph3 was visualized in cultured SCs (right column). D, quantification of cdc2Y15⁺, cyclin B1⁺, or ph3⁺ cells from C. Data are mean ± S.D. (n = 3). **, p < 0.01; ***, p < 0.001. E, quantification of Pax7⁺/ph3⁺ cells in regenerating TA muscle 5 days post-injury from control and FADD-D mice. Data are mean ± S.D. (n = 3). **, p < 0.01. Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

FADD Regulates SC Commitment and Differentiation

To examine the commitment and differentiation potential of FADD-D SCs, we cultured isolated SCs in low-serum medium and examined the expression of myosin heavy chain (MHC). As shown in Fig. 4A, myotube formation was largely inhibited in cultures of FADD-D SC cultures compared with that of littermates. Quantitative analysis of the myotube clones using an established clone formation assay indicated that the FADD-D SCs formed less than half as many MHC⁺ clones than the control SCs (15 ± 2% and 39 ± 7%, respectively; n = 4) (Fig. 4B) (29). Consistently, reduced expression of myogenin and desmin in cultured FADD-D SCs relative to those of control SCs was observed (Fig. 4, C and D). Consistent with the decreased expression of desmin and myogenin in isolated FADD-D SCs, the levels of both desmin and myogenin were reduced in FADD-D-regenerating TA muscles relative to those of control animals (Fig. 4C, right panel). These data indicate that FADD phosphorylation represses the expression of commitment- and differentiation-associated markers, which then maintains the SCs in a less committed and undifferentiated state.

FIGURE 4.

FADD regulates SC differentiation and commitment. A, MHC-expressing myofibers developed in cultures of SCs from littermate controls and FADD-D mice grown in low-serum medium (2% horse serum in DMEM) for 3 days. Myofibers were visualized by immunostaining (n = 3). B, quantification of MHC-immunopositive myoblast clones in primary cultures of SCs from control and FADD-D animals that were seeded in 35-mm dishes with Methocult M3434 medium. Three or more dishes were plated for each sample. Data are mean ± S.D. (n = 4). **, p < 0.01. C, expression of myogenin and desmin in SCs isolated from control and FADD-D animals and cultured for 72 h (left panel) (n = 3). Transverse sections were collected from regenerating TA muscle 7 days post-injury and then stained with the antibodies as indicated (right panel) (n = 5). D, quantification of myogenin⁺ or desmin⁺ cells in the cells described in D. Data are mean ± S.D. (n = 5). *, p < 0.05; **, p < 0.01. Representative cells are shown (n = 3). Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

FADD-D Promotes the Quiescence of Cycling SCs

Because FADD-D SCs exhibited an impaired differentiation potential, we were interested in examining the fate of the undifferentiated cycling SCs. Cell death was examined first. Although the presence of FADD-D induced SCs to arrest at the G₂/M boundary, there was an insignificant increase in TUNEL staining in the regenerating region of FADD-D TA muscles compared with those of control muscles (Fig. 5A). This result was confirmed by the comparable amounts of cleaved caspase 3-expressing cells observed (Fig. 5, B, bottom panel, and C). These data indicate that FADD-D did not induce apoptosis in the SCs. A subset of cycling SCs has been shown to return to reversible quiescence instead of differentiation. Therefore, we asked whether FADD-D promotes the return of the cycling SCs to quiescence. An established reserve cell preparation provides an ideal in vitro model for cycling SCs returning to reversible quiescence. FADD-D SCs were cultured for proliferation and then switched to low-serum conditions for 4 days. Interestingly, the SCs from FADD-D mice exhibited ∼4 times as much quiescent myogenic cells (Pax7⁺Ki67⁻) as those of control mice (37 ± 5% and 9.7 ± 1.8%, respectively; n = 4) (Fig. 5, B and C). Moreover, FADD phosphorylation was sustained in the Pax7⁺ reserve cells as opposed to differentiated cells, whose Pax7 levels were decreased (Fig. 5D). These data strongly suggest that FADD phosphorylation controls the decision between a fate of reversible quiescence and a fate of irreversible differentiation. Because FADD-D induced a G₂/M arrest (Fig. 3, C and D), we tested whether G₂/M arrest regulates the fate of the cycling SCs switched to low-serum conditions. SCs were arrested at prophase by treatment with nocodazole, which also induces FADD phosphorylation (26), to examine the fate of the cycling SCs in low-serum cultures. Similar to the results from the FADD-D SCs, nocodazole dramatically inhibited differentiation, whereas the content of quiescent SCs was increased (Fig. 5E). The effect of G₂/M arrest on the quiescence of SCs was confirmed by treatment with CGP74514A, a selective inhibitor of cdc2 (Fig. 5, E and F). These data suggest that G₂/M arrest plays a role in determining whether cycling SCs will differentiate or enter a quiescent state.

FIGURE 5.

FADD regulates the quiescence of SCs. A, TUNEL assays were performed to assess apoptosis in transverse sections of regenerating TA muscles of control and FADD-D animals 7 days post-injury. B, SCs were cultured for 24 h in growth medium and then switched to low-serum medium for 4 days of culture. Cells were double-stained with Pax7 and Ki67 antibodies. Arrows indicate Pax7⁺/Ki67⁻ cells (top row). Apoptotic cells were stained with an antibody that specifically recognizes cleaved caspase 3 (bottom row). C, quantification of quiescent SCs (Pax7⁺/Ki67⁻) and cleaved caspase 3⁺ cells from B. Data are mean ± S.D. (n = 4). ***, p < 0.001. D, colocalization of Pax7 and pFADD in SCs cultured in low-serum medium for 4 days (n = 3). E, quantification of the fusion index to assess myofiber formation in SCs cultured in low-serum medium for 4 days in the presence of nocodazole (40 ng/ml) or CGP74514A (5 μm). Data are mean ± S.D. (n = 3). ***, p < 0.001. F, quantification of quiescent SCs (Pax7⁺/Ki67⁻) in the presence of nocodazole or CGP74514A as indicated. Data are mean ± S.D. (n = 3). ***, p < 0.001. Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

FADD-D Regulates SC Self-renewal

Because asymmetric division regulates the fate decisions of SCs (30), we were interested in whether FADD phosphorylation is involved in the asymmetric division of SCs. Template DNA has been observed to be asymmetrically localized in daughter SCs (31). Thus, we examined the localization of phosphorylated FADD in the daughter SCs of asymmetric SC divisions, which were distinguishable by the biased distribution of EdU. Template DNA with incorporated EdU was distributed asymmetrically into the mitotic daughter cells. More interestingly, ∼85% of the cells cosegregated EdU-labeled template DNA and phosphorylated FADD into one of the daughter cells (Fig. 6A, n = 14). In contrast, total FADD was distributed symmetrically in ∼90% of the daughter cells of asymmetric mitosis (Fig. 6A, n = 7). These data indicate that FADD is asymmetrically phosphorylated in the self-renewing daughter SCs and suggest that FADD phosphorylation plays a role in the regulation of SC self-renewal.

FIGURE 6.

FADD regulates SC self-renewal. A, pFADD is asymmetrically localized in self-renewing daughter cells. Among 14 asymmetric pairs denoted by EdU staining, cosegregation of pFADD (red) with EdU in one daughter cell was detected in 12 of the asymmetric pairs, whereas non-biased distribution of total FADD (tFADD, red) was detected in 7 of the 8 asymmetric pairs observed. Representative images are shown. B, quantitative RT-PCR analysis of relative gene expression in freshly isolated SCs from control and FADD-D. Data are mean ± S.D. (n = 6). *, p < 0.05; **, p < 0.01. C, quantification of M-cadherin⁺/CD34⁺ myogenic cells in the hind limb muscles of control and FADD-D animals (n = 3). PI, propidium iodide. D, quantification of Pax7⁺ cells per 100 myonuclei in TA muscle from control and FADD-D animals. Data are mean ± S.D. (n = 3). *, p < 0.05; **, p < 0.01. E, immunoblotting analysis of protein expression, as indicated, in TA muscle from control and FADD-D animals (n = 4). F, representative image of a hind limb from an adult FADD-D mouse and a littermate (Control) (n = 3). G, Lamini A2 expression in TA muscles from control and FADD-D animals. Muscle sections were immunostained for Lamini A2 (red) (n = 4). Scale bars = 20 μm in A and 50 μm in E. Sections or cells were counterstained with DAPI (blue).

To examine the effect of FADD-D on the self-renewal of SCs, we analyzed the levels of stem-like markers in quiescent FADD-D SCs. As expected, higher levels of “stemness” markers such as CD34, CXCR4, Myf-5, and Pax7 (21, 22, 32) were observed in quiescent FADD-D SCs compared with their control counterparts (Fig. 6B). Together with the decreased levels of myogenin and desmin (Fig. 4C), these alterations suggest that the FADD-D SCs exist in a more stem-like and less committed state.

To examine the effect of FADD-D on SC self-renewal in vivo, we determined the SC pool in muscles. Interestingly, FADD-D hind limb muscles exhibited ∼3 times as much SCs (characterized by the coexpression of M-cadherin and CD34), as did their control counterparts (Fig. 6C) (33). Similarly, TA muscles from 3-month-old FADD-D animals exhibited approximately twice as many Pax7⁺ SCs as the controls (Fig. 6D). The SCs maintain their pool throughout regeneration by self-renewal. To test whether the greater size of the SC pool in FADD-D muscles is transient or permanent, we performed muscle injury experiments to assess the self-renewal ability of SCs. The quiescent SCs (Pax7⁺Ki67⁻) were quantified in the muscles 45 days post-injury. The increased number of SCs in FADD-D muscles compared with control muscles was confirmed by quantifying Pax7-expressing cells. Cross-sections of the TA muscles of FADD-D mice exhibited approximately twice as many SCs as their littermate controls (Fig. 6D), indicating that the SCs in FADD-D animals are capable of self-renewal. The elevated expression of Pax7 and M-cadherin in the muscles of FADD-D mice was confirmed by immunoblotting (Fig. 6E).

Although the TA muscles of FADD-D mice exhibited a larger SC pool, the size of the TA muscles of adult FADD-D mice was smaller than normal, and this difference persisted through adulthood (Fig. 6F). Similarly, the cross-sectional areas of the myofibers in the TA muscles of 3-month-old FADD-D mice were smaller than those in control mice (Fig. 6G). The expression of MHC was also reduced in the FADD-D muscles, suggesting impaired differentiation in FADD-D (Fig. 6E). Together with the larger induced SC pool in the FADD-D muscles, these data suggest that the balance between differentiation and self-renewal during postnatal muscle development is impaired in FADD-D SCs.

FADD-D Regulates Notch Signaling

Notch signaling has been shown to promote DNA synthesis and repress the differentiation of SCs (34). Therefore, we examined Notch signaling in FADD-D SCs. TA muscles from FADD-D animals exhibited more Notch-1 levels compared with the controls (Fig. 7A). Higher levels of DLL-1 were found in the regenerating TA muscles of FADD-D mice compared with those of control mice (Fig. 7B, top row). Accordingly, more cleaved Notch-1 was observed in the FADD-D regenerating muscles (Fig. 7B, bottom row). These results revealed an interesting link between FADD and Notch signaling. To test this, we examined whether Notch signaling inhibition affects the suppressed differentiation potential of FADD-D SCs. As expected, L-685, 468 treatment restored MHC⁺ clone formation in FADD-D SCs (Fig. 7C). These data strongly indicate that the compromised differentiation property of the FADD-D SCs originates mainly from enhanced Notch signaling.

FIGURE 7.

FADD regulates Notch signaling. A, TA muscles from a FADD-D mouse and a littermate (Control) were homogenized and lysed for immunoblotting analysis to detect the expression of proteins as indicated. Representative images are shown (n = 3). B, expression of DDL-1 and cleaved Notch-1 in regenerating TA muscle at 7 days post-injury. Transverse sections were immunostained for DLL-1 (top row) and cleaved Notch-1 (bottom row) (n = 3). C, quantification of the clones immunopositive for MHC after 15 days of culture in M3434 medium. Seven days after seeding, L685 was added to the medium to a final concentration of 4 μm. Data are mean ± S.D. (n = 3). *, p < 0.05. DMSO, dimethyl sulfoxide. D, normal or regenerating TA muscles from wild-type C57 mice were homogenized and lysed for immunoblotting analysis to detect the expression of proteins as indicated. The regenerating area was injected with 10 μl of dimethyl sulfoxide 24 h post-injury, and the contralateral leg (denoted by the same mouse number) was injected with 10 μl of Go6976 (5 μm). Data are mean ± S.D. (n = 3). Representative images are shown. E, quantification of clones positive for MHC by immunostaining SCs cultured in M3434 medium for 15 days. Seven days after seeding, Go6976 was added to the medium to a final concentration of 5 μm. Data are mean ± S.D. (n = 3). **, p < 0.01. F, quantification of quiescent SCs (Pax7⁺/Ki67⁻) in low-serum cultures grown for 4 days in the presence of 5 μm of Go6976 or dimethyl sulfoxide. Data are mean ± S.D. (n = 3). ***, p < 0.001. Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

We demonstrated recently that FADD phosphorylation plays a critical role in the termination of conventional PKC (cPKC) signaling (35). Previous studies indicated that PKC and its downstream effectors promote Notch signaling (36–38). We sought to determine whether the enhanced Notch signaling was due to enhanced cPKC signaling in the FADD-D muscles. Interestingly, cPKC phosphorylation was indeed induced in regenerating muscles (Fig. 7D). Furthermore, TA muscles from FADD-D animals exhibited more phospho-PKC accumulation compared with controls (Fig. 7A). Interestingly, cPKC inhibition by the selective inhibitor Go6976 led to reduced protein levels of Notch-1 compared with that in the regenerating contralateral leg (Fig. 7D). These data suggest that cPKC plays a role in the activation of Notch signaling during myogenic lineage progression. To verify the effect of PKC on the FADD-D SCs, we performed myotube formation assays in the presence of Go6976. Strikingly, Go6976 restored the number of myotube clones to normal levels in cultures of FADD-D SCs (Fig. 7E). Consistent with this finding, the increase in the quiescent reserve cells in the cultures of FADD-D SCs was largely suppressed by Go6976 treatment (Fig. 7F). These data indicate that FADD regulates Notch signaling through inactivating cPKC, suggesting that cPKC plays a role in SC fate determination.

Role of FADD Phosphorylation in Deregulated SCs

Recently, dystrophin deficiency-induced muscle degeneration has been characterized as a stem cell disease in an mdx mouse model lacking telomerase activity (39, 40). The SCs from these mice exhibited a reduced proliferation potential and accelerated exhaustion of their capacity for renewal in vivo. We examined the FADD phosphorylation levels in these mdx mice. Interestingly, the induction of FADD phosphorylation in regenerating TA muscles was largely compromised in these mdx mice compared with those of control mice (Fig. 8A). Like those from the mdx mice, the SCs from aged muscles also exhibited abnormal fate decisions that were mediated by decreased Notch signaling and could be restored by the presence of serum from young mice (41, 42). Because FADD enhanced Notch signaling, we examined FADD phosphorylation in the aged SCs. As shown in Fig. 8, B and C, FADD phosphorylation was reduced in aged SCs exposed to aged serum. Moreover, the young serum restored FADD phosphorylation in the aged SCs. These data suggest that FADD phosphorylation plays a role in the progression of muscle stem cell deregulation.

FIGURE 8.

FADD phosphorylation in deregulated SCs. A, transverse sections of regenerating TA muscle from control and MDX mice 5 days post-injury. SCs were isolated from regenerating muscles 5 days post-injury with cardiotoxin (n = 5). B, SCs isolated from adult (3–6 months old) and old (>20 months old) mice were cultured for 48 h in F10 medium with 5% mouse serum, as indicated, without basic fibroblast growth factor (bFGF) (n = 3). C, quantification of the pFADD⁺ cells in the muscles shown in B. Data are mean ± S.D. (n = 3). **, p < 0.01. Scale bars = 20 μm. Sections or cells were counterstained with DAPI (blue).

DISCUSSION

Cell cycle progression plays critical roles in the fate determination of various stem cells. In muscle stem cells, the role of cell cycle exit in the transition from proliferation to differentiation has been well delineated. Cell cycle regulators such as p21 and Rb repress cycling SCs from reentering S phase and cause them to enter an irreversible quiescent state that facilitates differentiation (43). Interestingly, the cycling SCs that are engaged in self-renewal also withdraw from the cell cycle but enter into a reversible quiescent state. How the cell cycle machinery determines whether the exit from the cell cycle is irreversible or reversible is critical for cell fate determination. This question remains largely unanswered. In this study, we linked the G₂/M arrest of the SCs with their reversible quiescence. We demonstrated that FADD phosphorylation was specifically induced in the cycling SCs, which exhibited high expression levels in metaphase and declined in later anaphase. When FADD was mutated to be constitutively phosphorylated, mitosis entry was delayed, and the cycling SCs were prone to enter into a quiescent state under low-serum conditions instead of differentiating. FADD phosphorylation has been shown to induce G₂/M arrest in certain cells, and it sensitizes cancer cells to chemotherapy-induced apoptosis (44, 45). In this report, the G₂/M arrest induced by the presence of FADD-D did not lead directly to apoptosis. Alternatively, FADD-D leads cycling SCs to a quiescent state, suggesting that the G₂/M arrest is reversible, which then favors SCs quiescence. Previous studies have indicated that cdc2 and cdc25, which regulate the G₂/M transition, control the fate decision of neural stem cells (46, 47). We show here that nocodazole or a selective inhibitor of cdc2, both of which induced G₂/M arrest, switched the fate of the cycling SCs from irreversible differentiation to reversible quiescence in low-serum conditions. These results indicate that FADD is a cell cycle regulator that regulates the reversible quiescence of SC via its phosphorylation and suggest that FADD has a role in G₂/M arrest, affecting the fate decisions of SCs. Accumulation of the forkhead transcription factor FoxM1 is critical for entry into M phase during the cell cycle (48). FADD-D has been observed to inhibit the expression of FoxM1 (49). FoxM1 promotes the expansion and differentiation of several types of stem cells, suggesting a mechanism for how FADD phosphorylation regulates the fates of SCs (50, 51). Additional experiments are required to determine whether FoxM1 is a downstream effector of FADD phosphorylation.

Previous studies indicated that Notch signaling is required for the activation and proliferation of SCs (34). Our data demonstrate that Notch-1 levels are regulated by cPKC (Fig. 7D) and that TA muscles from FADD-D exhibited higher cPKC levels than the controls (Fig. 7A). cPKC is up-regulated immediately upon muscle injury (Fig. 7D), suggesting a mechanism for Notch-1 up-regulation during muscle repair. Furthermore, FADD phosphorylation has been demonstrated to regulate thymus development, although the mechanisms for this remain unclear (19). Our data provide a mechanistic cue for the investigation of the role of FADD phosphorylation in development. The detailed mechanisms for cPKC regulating Notch-1 expression need further investigation.

Asymmetric cell division has been shown to play critical roles in the self-renewal of SCs. Here we demonstrated that phosphorylated FADD is asymmetrically localized in the daughter cells. Phosphorylated FADD was prone to be distributed in the self-renewing daughter cell that contained numb protein or the template DNA. Moreover, the localization of phosphorylated FADD and desmin was mutually exclusive. These data identify a novel molecular signature for self-renewing daughter SCs.

The balance among stem cell self-renewal, differentiation, and programmed cell death is critical for tissue development and homeostasis (52). Previous studies have indicated that apoptotic protease caspase 3 governs the differentiation of myoblasts (53, 54). In this study, we characterized a role of FADD in governing muscle stem cell fates regulated by its phosphorylation without affecting cell death. FADD has been shown to be essential for embryonic development by mediating programed cell death (14, 17). These data suggest complicated roles of FADD during tissue development and maintenance, which is possibly determined by its phosphorylation.

These data provide new insights into the intracellular mechanisms of the cell fate decisions of cycling SCs and identify FADD as a critical regulator of the balance between self-renewal and differentiation in cycling SCs regulated by its phosphorylation.