Canonical NF-κB signaling regulates satellite stem cell homeostasis and function during regenerative myogenesis

Abstract

Skeletal muscle regeneration in adults is attributed to the presence of satellite stem cells that proliferate, differentiate, and eventually fuse with injured myofibers. However, the signaling mechanisms that regulate satellite cell homeostasis and function remain less understood. While IKKβ-mediated canonical NF-κB signaling has been implicated in the regulation of myogenesis and skeletal muscle mass, its role in the regulation of satellite cell function during muscle regeneration has not been fully elucidated. Here, we report that canonical NF-κB signaling is induced in skeletal muscle upon injury. Satellite cell-specific inducible ablation of IKKβ attenuates skeletal muscle regeneration in adult mice. Targeted ablation of IKKβ also reduces the number of satellite cells in injured skeletal muscle of adult mice, potentially through inhibiting their proliferation and survival. We also demonstrate that the inhibition of specific components of the canonical NF-κB pathway causes precocious differentiation of cultured satellite cells both ex vivo and in vitro. Finally, our results highlight that the constitutive activation of canonical NF-κB signaling in satellite cells also attenuates skeletal muscle regeneration following injury in adult mice. Collectively, our study demonstrates that the proper regulation of canonical NF-κB signaling is important for the regeneration of adult skeletal muscle.

myogenesis, satellite cells, NF-κB, proliferation, apoptosis, inflammation

Introduction

Skeletal muscle regeneration is mediated by a subset of adult stem cells, termed satellite cells, which are located between the basal lamina and the sarcolemma in a relatively dormant state (Bentzinger et al., 2012; Relaix and Zammit, 2012). Upon muscle injury, satellite cells enter the cell cycle, undergo several rounds of cell division, and then differentiate into myoblasts, which ultimately fuse with each other or with injured myofibers to complete the repair process. While a vast majority of satellite cells commit to the myogenic lineage, a fraction of them self-renews and returns to quiescence to replenish the satellite cell pool (Yin et al., 2013). Satellite cells express the transcription factor Paired box 7 (Pax7), which is essential for their self-renewal, proliferation, and maintenance of myogenic potential in adult skeletal muscle (Seale et al., 2000; von Maltzahn et al., 2013). However, the signaling mechanisms that regulate satellite cell function during regenerative myogenesis remain poorly understood.

Nuclear factor-κB (NF-κB) is a family of transcription factors that regulate a large number of genes involved in the survival, proliferation, and differentiation of both immune and non-immune cell types (Hayden and Ghosh, 2012). The NF-κB family contains five members: RelA (also known as p65), RelB, c-Rel, p105/p50, and p100/p52, which make homo- and heterodimers. In unstimulated cells, NF-κB proteins are normally sequestered in the cytoplasm by the related family member inhibitor of κB (IκB) and the IκB kinase (IKK) complex (Kumar et al., 2004). NF-κB can be activated through a canonical or non-canonical pathway (Hayden and Ghosh, 2012). The canonical NF-κB pathway involves the upstream activation of IKKβ and subsequent phosphorylation and degradation of IκB proteins, resulting in rapid and transient nuclear translocation of canonical NF-κB members, predominantly the p50/RelA and p50/c-Rel dimers. In the non-canonical pathway, a central signaling molecule is NF-κB-inducing kinase (NIK), which activates and cooperates with IKKα to mediate p100 phosphorylation, which in turn leads to p100 ubiquitination and degradation of its C-terminal IκB-like structure, resulting in the generation of mature NF-κB2 (i.e. p52) and nuclear translocation of the non-canonical NF-κB complex, p52/RelB (Kumar et al., 2004). Published reports suggest that the activation of NF-κB reduces skeletal muscle mass, metabolic function, and regeneration, especially in disuse conditions and chronic disease states (Mourkioti et al., 2006; Li et al., 2008; Shintaku and Guttridge, 2013). Intriguingly, a recent study has demonstrated that the sustained inhibition of NF-κB in myofibers increases the age-associated loss of skeletal muscle mass, suggesting that physiological levels of NF-κB are essential for maintaining skeletal muscle health (Zhang et al., 2017).

Studies using cultured myoblasts have demonstrated that the canonical and non-canonical NF-κB pathways play distinct roles in the regulation of myogenesis (Bakkar and Guttridge, 2010). The canonical NF-κB pathway is activated in proliferating myoblasts but is inhibited during the differentiation of myoblasts into multinucleated myotubes (Bakkar and Guttridge, 2010). Indeed, several proinflammatory cytokines inhibit myogenic differentiation through the activation of canonical NF-κB signaling (Langen et al., 2001; Dogra et al., 2006). In contrast, the non-canonical NF-κB pathway becomes activated during myogenic differentiation and promotes mitochondrial biogenesis (Bakkar et al., 2008, 2012). Moreover, activation of the non-canonical NF-κB signaling pathway promotes myoblast fusion during myogenesis (Enwere et al., 2012; Hindi et al., 2017).

While the role of IKKβ-mediated canonical NF-κB signaling in the regulation of myogenesis is evidenced, it remains enigmatic whether this pathway has any role in the regulation of satellite stem cell function. It has been reported that the activation of IKKβ in satellite cells is a reason for their reduced differentiation into the myogenic lineage in mouse models of cancer cachexia (He et al., 2013). In contrast, we found that the proinflammatory cytokine, TWEAK, activates NF-κB in satellite cells. Interestingly, inhibition of canonical NF-κB signaling improves the number of Pax7⁺ cells in TWEAK-treated cultures suggesting that depending on the stimuli, the activation of NF-κB can differentially regulate the fate of satellite cells (Ogura et al., 2013). Recent studies from our group have shown that TAK1 and TRAF6, the upstream regulators of the canonical NF-κB signaling pathway, are essential for the survival, self-renewal, and proliferation of satellite cells in skeletal muscle of adult mice. Targeted ablation of either TAK1 or TRAF6 in satellite cells severely impairs skeletal muscle regeneration in adult mice (Ogura et al., 2015; Hindi and Kumar, 2016). However, the role of canonical NF-κB signaling in the regulation of satellite cell function during regenerative myogenesis has not yet been investigated.

In this study, through the generation of an inducible satellite cell-specific IKKβ knockout (KO) mouse, we demonstrate that the canonical NF-κB signaling pathway in satellite cells mediates regeneration of skeletal muscle after injury. Our results also suggest that the canonical NF-κB pathway promotes the survival and proliferation of satellite cells. Moreover, canonical NF-κB signaling is essential to prevent the precocious differentiation of activated satellite cells. Finally, we demonstrate that while physiological levels support satellite cell function, supra-physiological activation of canonical NF-κB signaling delays skeletal muscle regeneration in adult mice.

Results

Canonical NF-κB signaling is activated in satellite cells after skeletal muscle injury

Tibialis anterior (TA) muscle of 12-week-old wild-type (WT) mice was injured unilaterally by intramuscular injection of 1.2% BaCl₂ solution, whereas the contralateral muscle was injected with saline alone and served as an uninjured control. At Day 5 after injury, the muscle was collected and analyzed by performing immunoblotting or immunohistochemical analysis. There was a noticeable increase in the levels of IKKβ in 5-day (5d)-injured TA muscle compared to contralateral uninjured muscle (Figure 1A and B). Moreover, the levels of phosphorylated, as well as total, IκBα and p65 proteins were considerably increased in injured muscle compared to uninjured controls (Figure 1A and B). To understand whether muscle injury leads to the activation of canonical NF-κB signaling in satellite cells, we performed double immunostaining for phospho-p65 (p-p65) and Pax7 (a marker for satellite cells) proteins. Consistent with a published report (Riuzzi et al., 2012), p-p65 protein was undetectable in satellite cells of uninjured muscle (data not shown). However, the expression of p-p65 protein was clearly visible in Pax7⁺ cells in injured muscle (Figure 1C).

Figure 1

Canonical NF-κB signaling is activated in satellite cells during regenerative myogenesis. (A and B) Representative immunoblots and densitometry analysis of bands in immunoblots demonstrating the relative levels of IKKβ, phospho- and total IκBα and p65, and unrelated protein GAPDH in uninjured and 5d-injured TA muscle of 12-week-old WT mice. (C) Representative photomicrographs of Pax7- and phospho-p65 (p-p65)-stained muscle sections from injured TA muscle of WT mice. Nuclei were identified by DAPI staining. Arrows point to Pax7⁺/p-p65⁺ cells. Scale bar, 20 μm. (D) Representative photomicrographs of EDL single myofiber suspension cultures at 0, 24, and 72 h of incubation in growth medium after immunostaining for Pax7 and phospho-p65. Nuclei were identified by DAPI staining. Scale bar, 10 μm. *P < 0.05, values significantly different from corresponding uninjured TA muscle of mice by unpaired two-tailed t-test.

To further understand whether canonical NF-κB signaling is activated in satellite cells upon injury, we employed an ex vivo model of muscle injury (Hindi et al., 2012; Hindi and Kumar, 2016). We established single myofiber cultures from extensor digitorum longus (EDL) muscle of WT mice and myofiber-associated satellite cells were examined for the expression of p-p65 protein at different time points. We could not detect p-p65 protein in satellite cells (i.e. Pax7⁺) just after isolation of myofibers (Figure 1D). Interestingly, a gradual increase in p-p65 levels in Pax7⁺ cells was evidenced at 24 and 72 h after establishing the cultures (Figure 1D). These results suggest that muscle injury leads to the activation of canonical NF-κB signaling in satellite cells of adult mice.

Satellite cell-specific deletion of IKKβ delays myofiber regeneration in adult mice

To understand the role of canonical NF-κB signaling in the regulation of satellite cell homeostasis and function, we crossed floxed IKKβ (IKKβ^f/f) mice that are homozygous for loxP sites flanking exon 3 with Pax7-CreER mice (a tamoxifen-inducible satellite cell-specific Cre line (Lepper et al., 2009)) to generate satellite cell-specific inducible IKKβ-knockout (IKKβ^f/f;Pax7-CreER) mice (Figure 2A). Since Pax7-CreER mice are knock-in mice in which the expression of Pax7 is regulated by the endogenous Pax7 promoter (Mourkioti et al., 2006), we used adult IKKβ^f/f;Pax7-CreER mice and treated them with tamoxifen or vehicle (corn oil) alone to generate satellite cell-specific IKKβ knockout (henceforth P7:IKKβ-KO) or control (Ctrl) mice, respectively. The mice were subjected to five intraperitoneal injections of tamoxifen or vehicle alone (Figure 2B). One week after the first injection of tamoxifen, TA muscle of Ctrl and P7:IKKβ-KO mice was injected with 100 μl of 1.2% BaCl₂ solution unilaterally to induce necrotic muscle injury, whereas the contralateral muscle was injected with saline and served as control. In our initial experiments, we found no difference in the extent of injury in TA muscle of Ctrl and P7:IKKβ-KO mice measured at Day 3 post-BaCl₂ injection (data not shown). We performed most of the analysis at Day 5 post-injury because necrotic area in injured muscle start filling with newly formed myofibers at this point. In addition, we also analyzed muscle regeneration at Day 14 post-injury because the necrotic area in injured TA muscle of wild-type mice is completely filled with newly formed centronucleated myofibers by 14 days of injury (Hindi et al., 2012; Ogura et al., 2015; Hindi and Kumar, 2016). Under naïve conditions, we found no difference in the average cross-sectional area (CSA) or minimal Feret’s diameter between P7:IKKβ-KO and littermate Ctrl mice (data not shown). Interestingly, we found that the regeneration of TA muscle was considerably diminished in P7:IKKβ-KO mice compared to littermate Ctrl mice at Day 5 post-BaCl₂-mediated injury (Figure 2C). Morphometric analysis of 5d-injured TA muscle sections showed a significant reduction in average CSA and minimal Feret’s diameter of regenerating myofibers in P7:IKKβ-KO mice compared to littermate Ctrl mice (Figure 2D and E). There was also a marked decrease in centrally multinucleated myofibers in TA muscle of P7:IKKβ-KO mice compared to littermate Ctrl mice, indicating further signs of an impaired regeneration (Figure 2F). Although muscle structure appeared comparable, a significant reduction in average myofiber CSA and a trend towards reduction in minimal Feret’s diameter (P = 0.08) was also noticeable in TA muscle of P7:IKKβ-KO mice compared to littermate Ctrl mice even after 14 days of injury (Figure 2G and H).

Figure 2

Satellite cell-specific deletion of IKKβ impairs muscle regeneration in adult mice. (A) Schematic representation of the breeding strategy used for the generation of IKKβ^f/f;Pax7-CreER. (B) Treatment protocol for tamoxifen-induced Cre-mediated recombination and subsequent muscle collection in P7:IKKβ-KO and Ctrl mice. (C) Representative photomicrographs of H&E-stained sections illustrating a regeneration defect in injured TA muscle of P7:IKKβ-KO compared with littermate Ctrl mice at indicated time points after BaCl₂ injection. Scale bar, 20 μm. (D–F) Average myofiber CSA (D), average minimal Feret’s diameter (E), and percentage of myofibers containing two or more centrally located nuclei per field (F) at Day 5 post-injury. N = 5. (G and H) Average CSA (G) and minimal Feret’s diameter (H) at Day 14 post-injury. N = 3 (Ctrl) or 4 (P7:IKKβ-KO). *P < 0.05, **P < 0.01, ***P < 0.001, values significantly different from corresponding TA muscle of Ctrl mice by unpaired two-tailed t-test.

Figure 3

Inactivation of IKKβ in satellite cells impedes the formation of new myofibers and diminishes the satellite cell pool. (A) Representative photomicrograph of 5d-injured TA muscle section of Ctrl and P7:IKKβ-KO 12-week-old mice after immunostaining for eMyHC and Laminin. Nuclei were identified by DAPI staining. Scale bar, 20 μm. (B–D) Percentage of eMyHC⁺ myofibers per Laminin⁺ myofiber (B), average CSA of eMyHC⁺ myofiber (C), and average minimal Feret’s diameter of eMyHC⁺ myofiber (D) in 5d-injured P7:IKKβ-KO and Ctrl mice. (E) Representative photomicrograph of 5d-injured transverse TA muscle section of Ctrl and P7:IKKβ-KO mice after immunostaining for Pax7 and Laminin. Nuclei were identified by DAPI staining. Arrowhead points to Pax7⁺ cells. Scale bar, 20 μm. (F) Quantification of number of Pax7⁺ cells per Laminin⁺ myofiber. N = 5. (G) Relative mRNA levels of Pax7 in 5d-injured TA muscle of P7:IKKβ-KO and Ctrl mice. (H) Relative mRNA levels of select inflammatory markers in 5d-injured TA muscle of P7:IKKβ-KO and Ctrl mice. N = 4. *P < 0.05, **P < 0.01, ***P < 0.001, values significantly different from corresponding TA muscle of Ctrl mice by unpaired two-tailed t-test.

Targeted inactivation of IKKβ diminishes abundance of satellite cells in regenerating muscle of adult mice

To further confirm that the deletion of IKKβ in satellite cells reduces myofiber regeneration, we also examined the number of myofibers expressing the embryonic isoform of myosin heavy chain (eMyHC). Results showed that the frequency of eMyHC⁺ myofibers within Laminin staining (Figure 3A and B) and the size of eMyHC⁺ myofibers (Figure 3C and D) were significantly reduced in 5d-injured TA muscle of P7:IKKβ-KO mice compared with corresponding Ctrl mice.

We next investigated the effect of the deletion of IKKβ on satellite cell number in naïve and injured TA muscle by immunostaining for Pax7 and Laminin proteins (Figure 3E). There was no significant difference in the number of Pax7⁺ positive cells in uninjured TA muscle of Ctrl and P7:IKKβ-KO mice (data not shown). Since peak activation of the satellite cells (i.e. Pax7⁺ cells) in skeletal muscle occurs around Day 5 post-injury (Hindi et al., 2012; Hindi and Kumar, 2016), we measured satellite cell content in TA muscle at this time point. Moreover, it is not feasible to accurately distinguish satellite cells on muscle sections at an earlier time point because we rely on specific expression of Pax7, as well as their sublaminar localization, which is disrupted before Day 5 post-injury. Intriguingly, the number of Pax7⁺ cells was significantly reduced in 5d-injured TA muscle of P7:IKKβ-KO mice compared to littermate Ctrl mice (Figure 3E and F). Furthermore, relative mRNA levels of Pax7 were significantly reduced in 5d-injured TA muscle of P7:IKKβ-KO mice compared to littermate Ctrl mice (Figure 3G). These results suggest that the deletion of IKKβ reduces satellite cell number following injury, which may be responsible for the delayed myofiber regeneration.

Activation of canonical NF-κB signaling leads to the expression of a number of inflammatory cytokines (Kumar et al., 2004; Hayden and Ghosh, 2012). However, it remains unknown whether satellite cells are also the source of inflammatory molecules in injured skeletal muscle. Nevertheless, by performing qRT-PCR, we measured relative mRNA levels of proinflammatory cytokines: interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α (TNF-α). There was no significant difference in the relative mRNA levels of these cytokines in 5d-injured TA muscle of Ctrl and P7:IKKβ-KO mice (Figure 3H), suggesting that the deletion of IKKβ in satellite cells does not affect the expression of inflammatory cytokines.

Inhibition of canonical NF-κB signaling reduces the proliferation of satellite cells

To understand the cellular mechanisms for the reduced number of satellite cells in regenerating skeletal muscle of P7:IKKβ-KO mice, we investigated the role of canonical NF-κB signaling in the proliferation of satellite cells in vivo, ex vivo, and in vitro. TA muscle of 12-week-old Ctrl and P7:IKKβ-KO mice was injured by intramuscular injection of 1.2% BaCl₂ solution. After 48 h, the mice were given an intraperitoneal injection of EdU and the number of EdU⁺ nuclei in the TA muscle was determined 12 days later. Interestingly, a significant reduction in both intramyofibrillar and interstitial EdU⁺ nuclei was observed in TA muscle of P7:IKKβ-KO mice compared with corresponding Ctrl mice (Figure 4A and B). To further understand whether IKKβ regulates the proliferation of satellite cells, we established myofiber explant cultures and the proliferation of myofiber-associated cells was studied by measuring EdU incorporation after 72 h of culturing. Consistent with our in vivo results, we found that the proportion of EdU⁺ nuclei was significantly reduced in myofiber explants prepared from P7:IKKβ-KO mice compared to Ctrl mice (Figure 4C and D).

Figure 4

Inactivation of IKKβ in satellite cells diminishes satellite cell proliferation. (A and B) TA muscle of P7:IKKβ-KO and Ctrl mice was injured by intramuscular injection of 100 μl of 1.2% BaCl₂ solution. After 2 days, the mice were given an intraperitoneal injection of EdU, and 12 days later, TA muscles were collected and muscle sections were stained to detect EdU, Laminin, and nuclei. (A) Representative photomicrograph of 14d-injured transverse TA muscle section of Ctrl and P7:IKKβ-KO mice after processing for EdU and Laminin detection. Nuclei were identified by DAPI staining. Arrowhead points to interstitial EdU⁺ nuclei. Arrow points to intramyofibrillar EdU⁺ nuclei. Scale bar, 20 μm. (B) Quantification of interstitial EdU⁺ and intramyofibrillar EdU⁺ nuclei per Laminin⁺ myofiber in TA muscle of Ctrl and P7:IKKβ-KO mice. N = 4 (Ctrl) or 3 (P7:IKKβ-KO). (C and D) Single myofibers cultures were established from EDL muscle of Ctrl and P7:IKKβ-KO mice and cultured for 72 h. For last 90 min, the cells were pulse labeled with EdU by addition of EdU in culture medium. The myofibers cultures were then fixed with paraformaldehyde and stained for EdU. Nuclei were labeled by incubation in DAPI. (C) Representative merged images of EdU and DAPI staining are presented here. Scale bar, 10 μm. (D) Quantification of the percentage of EdU⁺ cells per cluster. Cluster was defined as ≥4 cells. N = 3 mice, 9–11 fibers per group. *P < 0.05, **P < 0.01, ***P < 0.001, values significantly different from corresponding TA muscle of Ctrl mice by unpaired two-tailed t-test. (E and F) Primary WT myoblasts were transfected with scrambled siRNA or siRNA targeting IKKb or p65 for 72 h. (E) Representative photomicrograph of siRNA-transfected cells after processing for EdU detection. Nuclei were identified by DAPI staining. Scale bar, 10 μm. (F) Quantification of the percent EdU⁺ cells. N = 4. (G and H) In a parallel experiment, cells were collected after 48 h of siRNA transfection and processed for biochemical analysis. Immunoblots (G) and densitometric analysis (fold change) (H) demonstrate the protein levels of cyclin D1, cyclin A, IKKβ, p65, and unrelated protein GAPDH as a loading control. N = 3. ^$P < 0.05, values significantly different from scrambled siRNA control by unpaired two-tailed t-test. ^@P < 0.05, values significantly different from IKKβ siRNA-transfected group by unpaired two-tailed t-test.

We also investigated the role of canonical NF-κB signaling in the proliferation of satellite cell-derived myogenic cultures in a myofiber-free system. For this experiment, primary myogenic cultures established from WT mice were transfected with scrambled (control), IKKβ, or p65 siRNA. After 72 h, cellular proliferation was studied by performing an EdU incorporation assay. Results showed that the knockdown of IKKβ and p65 significantly reduced the proportion of EdU⁺ nuclei in myogenic cultures (Figure 4E and F). Our western blot analysis also showed that the levels of cyclin D1 and cyclin A were significantly reduced in cultures transfected with IKKβ siRNA or p65 siRNA compared to control cultures (Figure 4G and H). Western blot analysis confirmed a drastic reduction in IKKβ and p65 protein upon transfection with their corresponding siRNAs (Figure 4G and H).

Canonical NF-κB signaling promotes the survival of satellite cells

One of the important functions of the canonical arm of NF-κB signaling is to promote cell survival (Kumar et al., 2004; Hayden and Ghosh, 2012). Our attempts to perform double immunostaining for TUNEL and Pax7 were futile, possibly because the levels of Pax7 are reduced in satellite cells undergoing apoptosis. Therefore, we employed ex vivo myofiber explants and cultured myogenic cells to understand the role of the canonical arm of NF-κB signaling in the survival of satellite cells. We first established myofiber explant cultures from EDL muscle of Ctrl and P7:IKKβ-KO mice. After 72 h of culturing, the percentage of apoptotic cells was measured by performing TUNEL staining. Interestingly, the proportion of myofiber-associated TUNEL⁺ cells was significantly higher in P7:IKKβ-KO cultures compared to Ctrl cultures (Figure 5A and B).

Figure 5

Satellite cell-specific ablation of IKKβ promotes apoptotic cell death. (A) Single myofiber suspension cultures were established from EDL muscle of 6-week-old P7:IKKβ-KO and Ctrl mice, cultured for 72 h, and then processed for detection of TUNEL. Nuclei were identified by DAPI staining. Scale bar, 10 μm. (B) Quantification of the percentage of TUNEL⁺ cells per cluster. Cluster was defined as ≥4 cells. N = 3 mice, 9–11 fibers per group. *P < 0.05, values significantly different from Ctrl cultures by unpaired two-tailed t-test. (C–I) Primary WT myoblasts were transfected with scrambled siRNA or siRNA targeting IKKβ or p65 for 48 h. (C) Relative LDH activity (fold change) following siRNA transfection. (D) Following siRNA transfection, cells were collected and stained for annexin V and propidium iodide (PI), and analyzed by FACS for early (lower right quadrant) and late (upper right quadrant) apoptotic cells. (E and F) Early (E) and late (F) apoptotic cells following siRNA transfection were quantified. (G and H) Immunoblots (G) and densitometry analysis of bands in immunoblots (H) demonstrate the levels of BAX, Bcl2, cleaved caspase-3, cleaved PARP, and unrelated protein GAPDH. (I) Quantification of BAX/Bcl2 ratio. N = 3. ^$P < 0.05, ^$$P < 0.01, ^$$$P < 0.001, values significantly different from scrambled siRNA control by unpaired two-tailed t-test.

We next investigated the effect of siRNA-mediated knockdown of IKKβ or p65 on survival of cultured primary myogenic cells. Lactate dehydrogenase (LDH) is a stable enzyme that accumulates in culture supernatants after cell death (Ogura et al., 2015). Interestingly, we found that the levels of LDH in culture supernatants of cells transfected with IKKβ and p65 siRNA were significantly higher compared to those transfected with control siRNA (Figure 5C). We also performed annexin V/propidium iodide (PI) staining on these cells followed by analysis using FACS. Similar to our LDH results, we found that the percentages of early and late apoptotic cells were significantly increased in cultures transfected with IKKβ and p65 siRNA compared to those transfected with scrambled siRNA (Figure 5D–F). We also performed western blotting to measure the levels of cleaved caspase-3 and cleaved PARP, two established markers of apoptosis. The levels of cleaved PARP were significantly increased in cultures transfected with IKKβ or p65 siRNA compared to those transfected with scrambled siRNA (Figure 5G and H). Although we found that the knockdown of IKKβ or p65 led to a reduction in levels of BAX and Bcl2 protein in cultured myogenic cells, the proportion of BAX to Bcl2 was higher in IKKβ or p65 knocked down cultures compared with cultures, suggesting a pro-apoptotic environment (Figure 5G and I). Collectively, these results suggest that canonical NF-κB signaling promotes the survival of myogenic cells.

Canonical NF-κB signaling promotes the self-renewal and prevents the precocious differentiation of activated satellite cells

Pax7 determines satellite cell fate in cooperation with other factors, such as MyoD (Hindi et al., 2012; Ogura et al., 2015; Hindi and Kumar, 2016). Indeed, the expression pattern of Pax7 and MyoD specifies the myogenic status of satellite cells as quiescent (Pax7⁺/MyoD⁻), activated (Pax7⁺/MyoD⁺), or differentiated (Pax7⁻/MyoD⁺). We investigated whether IKKβ has any role in the regulation of satellite cell fate. To model the satellite cell dynamics of muscle injury, we generated ex vivo suspension cultures of EDL myofibers from P7:IKKβ-KO and Ctrl mice. The ex vivo suspension cultures were analyzed by staining with anti-Pax7 and anti-MyoD at 0 or 72 h of culturing (Figure 6A and B). There was no significant difference in the number of myofiber-associated Pax7⁺ cells between Ctrl and P7:IKKβ-KO cultures at 0 h (Figure 6A and C). There were a negligible number of myofiber-associated MyoD⁺ cells and there was no significant difference in their number between Ctrl and P7:IKKβ-KO cultures at 0 h (Figure 6D). Myofiber-associated satellite cells formed clusters by 72 h of culturing. Although there was no significant difference in the average number of clusters per myofiber (data not shown), the average number of cells per cluster was significantly reduced in P7:IKKβ-KO cultures compared to Ctrl cultures (Figure 6E). Our analysis also showed that the proportion of Pax7⁺/MyoD⁻, as well as Pax7⁺/MyoD⁺, cells was significantly reduced in P7:IKKβ-KO cultures compared to Ctrl cultures (Figure 6F and G). Conversely, the percentage of Pax7⁻/MyoD⁺ was dramatically increased in P7:IKKβ-KO compared with Ctrl cultures (Figure 6H), collectively suggesting that the inactivation of IKKβ causes a precocious differentiation of activated satellite cells and a loss in the capability to self-renew.

Figure 6

Satellite cell-specific deletion of IKKβ diminishes the self-renewal of satellite cells and results in their premature differentiation. Single myofiber suspension cultures were established from EDL muscle of 6-week-old P7:IKKβ-KO and Ctrl mice, cultured for 0 or 72 h, and then collected and immunostained for Pax7 and MyoD. (A and B) Representative individual and merged images of freshly isolated (A) and 72 h cultured (B) myofibers from Ctrl and P7:IKKβ-KO mice stained with PAX7, MyoD, and DAPI. Pax7⁺ cells per myofiber (C) and MyoD⁺ cells per myofiber (D) in Ctrl and P7:IKKβ-KO 0-h cultures, as well as average number of cells per cluster (E), percentage of Pax7⁺/MyoD⁻ cells per cluster (F), percentage of Pax7⁺/MyoD⁺ cells per cluster (G), and percentage of Pax7⁻/MyoD⁺ cells per cluster (H) in Ctrl and P7:IKKβ-KO 72-h cultures were quantified. Cluster was defined as ≥4 cells. N = 3 mice, 9–11 fibers per group. Data are presented as box-and-whisker plot. *P < 0.05, **P < 0.01, ***P < 0.001, values significantly different from myofiber cultures of Ctrl mice by unpaired two-tailed t-test.

We also investigated the role of canonical NF-κB signaling in the regulation of satellite cell fate in a myofiber-free environment. Early passage myogenic cells from WT mice were transfected with scrambled (control), IKKβ, or p65 siRNA. After 72 h, the cultures were fixed followed by staining with anti-Pax7 and anti-MyoD. Nuclei were labeled using DAPI (Figure 7A). Consistent with published reports (Ogura et al., 2015; Hindi and Kumar, 2016), we found that the majority of cells were Pax7⁺/MyoD⁺, whereas a small proportion was Pax7⁺/MyoD⁻ in cultures transfected with control siRNA. Although there was a trend towards reduction in the percentage of Pax7⁺/MyoD⁻ cells in cultures transfected with IKKβ or p65 siRNA, it was not significantly different from those transfected with control siRNA (Figure 7B). However, the proportion of Pax7⁺/MyoD⁺ cells was significantly reduced, whereas the proportion of Pax7⁻/MyoD⁺ cells was significantly increased in IKKβ or p65 knockdown cultures compared to control cultures (Figure 7C and D). By performing immunostaining for MyHC protein, we also investigated whether the knockdown of IKKβ or p65 can induce the terminal differentiation of satellite cell-derived myoblasts under growth conditions. There was a negligible number of MyHC⁺ cells in myogenic cultures transfected with scrambled or IKKβ siRNA (Figure 7E). Intriguingly, we found that the knockdown of p65 increased the number of MyHC⁺ cells. Consistent with immunocytochemistry results, immunoblotting showed that the levels of Pax7 were significantly reduced, whereas the levels of myogenin and MyHC were increased, especially in cultures transfected with p65 siRNA (Figure 7F and G).

Figure 7

Knockdown of IKKβ or p65 causes precocious differentiation of satellite cells in vitro. (A–E) Primary WT myoblasts were transfected with scrambled siRNA or siRNA targeting IKKβ or p65 for 72 h. (A) Representative photomicrograph of siRNA-transfected cells after immunostaining for Pax7 and MyoD. Nuclei were identified by DAPI staining. Arrows point to Pax7⁻/MyoD⁺ cells. Percentage of Pax7⁺/MyoD⁻ cells (B), percentage of Pax7⁺/MyoD⁺ cells (C), and percentage of Pax7⁻/MyoD⁺ cells (D) post-siRNA transfection were quantified. N = 3. (E) Representative photomicrograph of siRNA-transfected cells after immunostaining for MyHC. Nuclei were identified by DAPI staining. (F and G) In a similar experiment, WT myoblasts were transfected with scrambled siRNA or siRNA targeting IKKβ or p65 for 48 h. Representative immunoblots (F) and densitometry analysis of bands in immunoblots (G) show protein levels of Pax7, MyoD, myogenin, MyHC, and unrelated protein tubulin. ^$P < 0.05, values significantly different from scrambled siRNA control by unpaired two-tailed t-test. ^@P < 0.05, values significantly different from IKKβ siRNA-transfected group by unpaired two-tailed t-test.

Constitutive activation of canonical NF-κB pathway in satellite cells attenuates skeletal muscle regeneration

Our preceding results showed that the inhibition of canonical NF-κB signaling diminishes the regeneration of skeletal muscle in adult mice. However, there are also reports suggesting that the activation of canonical NF-κB signaling, especially in disease conditions, reduces skeletal muscle regeneration (Li et al., 2008; He et al., 2013). To specifically investigate the effect of heightened canonical NF-κB signaling in satellite cell function during muscle regeneration, we employed R26Stop^FLIKK2ca (henceforth Rosa26-IKKβca) mice that allow for the inducible expression of a constitutively active form of IKKβ (IKKβca) and subsequent activation of the canonical NF-κB pathway (Sasaki et al., 2006). Rosa26-IKKβca mice were crossed with Pax7-CreER mice to generate Rosa26-IKKβca;Pax7-CreER mice. To induce the expression of IKKβca in satellite cells, Rosa26-IKKβca;Pax7-CreER mice were treated with tamoxifen or vehicle (corn oil) alone to generate satellite cell-specific IKKβ overexpression (henceforth P7:IKKβca) or control (Ctrl) mice, respectively. Finally, TA muscle of Ctrl and P7:IKKβca mice was injured unilaterally by intramuscular injection of 1.2% BaCl₂ solution and muscle regeneration was evaluated at Day 5 or Day 14 by performing H&E staining (Figure 8A). There was no significant difference in myofiber size in uninjured TA muscle of Ctrl and P7:IKKβca mice (data not shown). Interestingly, we found that the average myofiber CSA, as well as minimal Feret’s diameter, were significantly reduced in TA muscle of P7:IKKβca mice compared with corresponding Ctrl mice at Day 5 post-injury (Figure 8B and C). Moreover, the proportion of myofibers containing two or more centrally nucleated myofiber was also significantly reduced in 5d-injured TA muscle of P7:IKKβca mice compared with Ctrl mice (Figure 8D). However, after 14 days of injury, muscle structure was restored in both phenotypes, as the average myofiber CSA was comparable between Ctrl and P7:IKKβca mice (Figure 8A and E). We also performed anti-eMyHC staining on 5d-injured TA muscle of Ctrl and P7:IKKβca (Figure 8F). This analysis showed that the proportion, as well as average CSA and minimal Feret’s diameter of eMyHC⁺ myofibers, was significantly reduced in TA muscle of P7:IKKβca mice compared with Ctrl mice (Figure 8G–I). Collectively, these results suggest that constitutive activation of canonical NF-κB signaling in satellite cells also diminishes their function during regenerative myogenesis.

Figure 8

Satellite cell-specific overexpression of IKKβ also impairs muscle regeneration in adult mice. (A) Representative photomicrographs of H&E-stained sections illustrating a regeneration defect in injured TA muscle of P7:IKKβca compared with littermate Ctrl mice at indicated time points after BaCl₂ injection. Scale bar, 20 μm. (B–D) Average myofiber CSA (B), average minimal Feret’s diameter (C), and percentage of myofibers containing two or more centrally located nuclei per field (D) at Day 5 post-injury. N = 6. (E) Average myofiber CSA at Day 14 post-injury. N = 4. (F) Representative photomicrograph of 5d-injured transverse TA muscle section of Ctrl and P7:IKKβca 12-week-old mice after immunostaining for eMyHC and Laminin. Nuclei were identified by DAPI staining. Scale bar, 20 μm. (G–I) Percentage of eMyHC⁺ myofibers per Laminin⁺ myofiber (G), average CSA of eMyHC⁺ myofiber (H), and average minimal Feret’s diameter of eMyHC⁺ myofiber (I) in 5d-injured P7:IKKβca and Ctrl mice. N = 6. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, values significantly different from corresponding TA muscle of Ctrl mice by unpaired two-tailed t-test.

Discussion

Skeletal muscle regeneration is a complex process that is regulated by signals released from both damaged myofibers, as well as several other cell types that are either resident in the muscle or recruited to assist in clearing cellular debris (Bentzinger et al., 2012; Relaix and Zammit, 2012). Activation of satellite cells upon muscle injury requires transcriptional and translational reprogramming that is essential for their exit from quiescence, proliferation, and survival in a metabolically activated state (Yin et al., 2013; Dumont et al., 2015; Zismanov et al., 2016; Xiong et al., 2017). In this study, we demonstrate that the activation of the canonical arm of the NF-κB is increased in the satellite cells upon skeletal muscle injury (Figure 1). This activation of NF-κB appears to be a mechanism to improve the proliferation and survival of satellite cells (Figures 2 and 3). NF-κB is known to induce cellular proliferation through augmenting the expression of a number of growth regulatory molecules and cell cycle regulators (Li et al., 2008; Hayden and Ghosh, 2012). Indeed, previous studies have shown that NF-κB promotes myogenic cell proliferation through increasing the expression of cyclin D1 (Guttridge et al., 1999). Similarly, NF-κB is known to induce the expression of several anti-apoptotic molecules, such as Bcl2 (Kumar et al., 2004; Hayden and Ghosh, 2012). Our results demonstrate that the levels of both cyclin D1 and Bcl2 were diminished upon knockdown of either IKKβ or p65 in cultured satellite cells (Figures 4G and 5G). Reduced proliferation and/or increased cell death through apoptosis appear to be important mechanisms for the reduced number of satellite cells observed in regenerating skeletal muscle of P7:IKKβ-KO mice.

For efficient regeneration of skeletal muscle, satellite cells have to undergo extensive proliferation before their differentiation and fusion with injured myofibers. Although the molecular mechanisms remain unknown, recent studies have suggested that NF-κB may play a role in the regulation of satellite cell self-renewal and differentiation. It has been reported that in the ‘settings’ of cancer cachexia, the ability of satellite cells to differentiate into the myogenic lineage is lost, which may be a mechanism for cancer-induced muscle wasting (He et al., 2013). Indeed, the activation of the canonical arm of NF-κB signaling has been found to be one of the reasons for the deregulation of Pax7 expression and satellite cell dysfunction in tumor-bearing mice (He et al., 2013).

We found no significant difference in the number of quiescent/self-renewing (Pax7⁺/MyoD⁻) satellite cells in skeletal muscle of Ctrl and IKKβ-KO mice in uninjured muscle, suggesting that the inhibition of canonical NF-κB signaling does not disrupt satellite cell homeostasis in naïve conditions. By contrast, inhibition of canonical NF-κB signaling reduces the proliferation of activated satellite cells and causes their premature differentiation (Figures 6 and 7). While previous studies have shown that the constitutive activation of canonical NF-κB signaling inhibits myogenic differentiation (Li et al., 2008; Bakkar and Guttridge, 2010), the results of the present study suggest that NF-κB signaling is essential to support the proliferation of activated satellite cells. It is important to note that while the knockdown of IKKβ reduced the number of proliferating (Pax7⁺/MyoD⁺) and increased the number of differentiating (Pax7⁻/MyoD⁺) cells, it was not sufficient to induce terminal differentiation of satellite cells. By contrast, knockdown of p65 also resulted in the expression of MyHC, a marker of terminal differentiation of myogenic cells. This could be attributed to the fact that in addition to IKKβ, the p65 subunit of NF-κB can also be activated through signaling cross-talk and post-translational modifications (Hayden and Ghosh, 2012). Moreover, since p65 is the main subunit of the canonical NF-κB complex with transactivation domain, its knockdown may lead to a complete blockade of canonical NF-κB signaling. Indeed, our experiments demonstrate that compared to IKKβ, knockdown of p65 has more pronounced effects on proliferation, differentiation, and survival of cultured satellite cells (Figures 4, 5, and 7). Consistent with our results, a previous study has also demonstrated that muscle-derived stem cells (MDSCs) from p65^+/– mice show increased differentiation potential compared to those isolated from littermate WT mice (Lu et al., 2012).

The role of NF-κB in skeletal muscle regeneration has been previously investigated using genetic mouse models and molecular and pharmacological approaches (Li et al., 2008; Tang et al., 2010). NF-κB is highly activated in dystrophic muscle of mdx (a mouse model of DMD) mice (Kumar and Boriek, 2003; Acharyya et al., 2007). Interestingly, the genetic ablation of IKKβ in myofibers or in macrophages improved the myopathy observed in mdx mice (Acharyya et al., 2007). Similarly, the pharmacological inhibition of NF-κB using NEMO binding domain peptide was effective in improving muscle pathogenesis in mdx mice (Acharyya et al., 2007; Reay et al., 2011). Another study showed that the genetic ablation of IKKβ in myofibers improves skeletal muscle strength and regeneration and inhibits the accumulation of fibrosis in otherwise normal mice (Mourkioti et al., 2006). On similar lines, we have demonstrated that the inhibition of TRAF6 in myofibers improves skeletal muscle regeneration potentially through inhibition of NF-κB (Hindi et al., 2012). However, all of these studies were performed using mice in which the components of canonical NF-κB signaling were inhibited in myofibers and the improvement in muscle regeneration was attributed to a reduction in the inflammatory milieu in the injured muscle microenvironment. Our results demonstrate that canonical NF-κB signaling is important for satellite cell homeostasis and function during skeletal muscle regeneration. These findings are also supported by our recently published articles demonstrating that the targeted ablation of TRAF6 or TAK1, the upstream activator of IKKβ, also inhibits satellite stem cell homeostasis and function (Ogura et al., 2015; Hindi and Kumar, 2016). A more dramatic effect of satellite cell-specific ablation of TAK1 or TRAF6 compared to IKKβ on regenerative myogenesis could be attributed to the fact that, being upstream signaling modules, TAK1 and TRAF6 can regulate other pathways, including the MAPKs, which are also implicated in the regulation of satellite cell fate and function (Shi et al., 2013; Ogura et al., 2015; Hindi and Kumar, 2016).

Although physiological levels of canonical NF-κB signaling in satellite cells promote regeneration, our results further demonstrate that constitutive activation of canonical NF-κB in satellite cells also impairs their function during regenerative myogenesis (Figure 8). These results are consistent with a recently published report also demonstrating that the overexpression of a constitutively active mutant of IKKβ in satellite cells diminishes muscle repair after cryoinjury (Oh et al., 2016). Indeed, we have previously reported that the overexpression of a constitutively active mutant of IKKβ induces oxidative stress and reduces survival of cultured satellite cells (Ogura et al., 2015). Moreover, sustained activation of the canonical arm of NF-κB signaling can also reduce skeletal muscle regeneration by preventing the differentiation of satellite cells into the myogenic lineage, similar to that reported during cancer cachexia (He et al., 2013). Moreover, in the settings of degenerative muscle disorders, the activation of the canonical arm of NF-κB signaling within satellite cells may also reduce their myogenic potential. Indeed, partial inhibition of NF-κB has been found to improve the engraftment of MDSCs in dystrophic muscle of mdx mice (Lu et al., 2012; Proto et al., 2015). It is notable that while we observed a deficit in muscle regeneration at early time points (i.e. Day 5), skeletal muscle of both P7:IKKβ-KO and P7:IKKβca mice eventually regenerate, such that there is no major difference in muscle structure at Day 14 post-injury, suggesting that the activation or inhibition of canonical NF-κB pathway is compensated by other factors/pathways that are also activated in satellite cells upon muscle injury.

In summary, our study provides initial evidence that physiological levels of canonical NF-κB signaling promote the proliferation and survival of satellite cells. Furthermore, we provide evidence that satellite cell expression of IKKβ is important for the successful regeneration of adult skeletal muscle. An exhaustion of satellite cells is common in several muscular disorders and conditions, both in actual number, as well as functional satellite cells (Shin et al., 2013). However, it remains to be elucidated whether a reduction in NF-κB is responsible for satellite cell dysfunction in disease conditions. It will be interesting to investigate whether there is differential regulation of canonical NF-κB signaling in satellite cells and myofibers in different physiological and pathological conditions. Indeed, other myogenic pathologies, such as rhabdomyosarcoma, demonstrate aberrations in the canonical NF-κB signaling pathway (Cleary et al., 2017). Further studies into differential expressions of the canonical NF-κB signaling pathway may pave the road for novel therapies for pathological conditions involving skeletal muscle.

Materials and methods

Animals

Satellite cell-specific inducible IKKβ-knockout mice (i.e. P7:IKKβ-KO) were generated by crossing Pax7-CreER (Jax Strain: B6.Cg-Pax7^{tm1(cre/ERT2)Gaka}/J) with floxed IKKβ (i.e. IKKβ^f/f) mice as described (Li et al., 2003). Satellite cell-specific inducible constitutive active (ca) IKKβ (i.e. P7:IKKβca) were generated by crossing Pax7-CreER with R26-Stop floxed IKK2ca mice (Jax Strain:B6(Cg)-Gt(ROSA)26Sor^{tm4(Ikbkb)Rsky}/J). All mice were in the C57BL/6J background and their genotypes were determined by PCR from tail DNA. For Cre-mediated inducible deletion of IKKβ or overexpression of IKKβca, 12-week-old mice were injected intraperitoneally (i.p.) with tamoxifen (10 mg per kg body weight) for 5 consecutive days. Control mice were injected with corn oil only. The Institutional Animal Care and Use Committee (IACUC) and Institutional Biosafety Committee (IBC) of the University of Louisville approved all experimental protocols with mice in advance.

Skeletal muscle injury

One week after the first injection of tamoxifen, 100 μl of 1.2% barium chloride (BaCl₂, Sigma Chemical Co.) in saline was injected into the TA muscle of mice to induce necrotic muscle injury. In one experiment, 2 days after intramuscular injection of BaCl₂, the mice were given an i.p. injection of 5-ethynyl-2′-deoxyuridine (EdU; 4 μg per gm body weight). Twelve days post-EdU injection, the TA was isolated and transverse muscle sections were made. The sections were subsequently immunostained with anti-Laminin, DAPI for the detection of nuclei, and processed for the detection of EdU similar to as described (Hindi et al., 2017). The number of interstitial and intramyofibrillar EdU⁺ nuclei per lammin⁺ myofiber was quantified using NIH ImageJ software.

Histology and morphometric analysis

For skeletal muscle morphology and assessment of regeneration, 10 μm-thick transverse sections of the TA were stained with hematoxylin and eosin (H&E). For quantitative analysis, average myofiber CSA and minimal Feret’s diameter, and number of myofibers containing two more centrally located nuclei were analyzed in H&E-stained TA muscle sections.

Primary myoblast cultures

Primary were isolated from the hind limb muscles of 6- to 8-week-old C57BL/6J mice as described (Hindi and Kumar, 2016). For siRNA experiments, cells were transfected using the RNAiMAX Lipofectamine system using a protocol suggested by the manufacturer (Invitrogen). Scrambled (control) siRNA-A (Cat# sc-37007), mouse NFκB p65 siRNA (Cat# sc-29411), and mouse IKKβ siRNA (Cat# sc-35645) were purchased from Santa Cruz Biotechnology.

LDH assay

The amount of LDH in culture supernatants was measured using a commercially available LDH Cytotoxicity Assay kit following the protocol suggested by the manufacturer (Thermo Scientific Life Sciences).

Isolation and culturing of myofibers

Single myofiber cultures were established from EDL muscle after digestion with collagenase II (Worthington Biochemical Corporation) and trituration as previously described (Hindi and Kumar, 2016). Following incubation in growth medium, satellite cell clusters were detected by immunostaining for specific proteins and labeling of nuclei with DAPI. Satellite cell clusters containing four or more cells were taken in consideration for our analysis. Multiple two-dimensional images at different focal planes were taken to allow for inspection of the three-dimensional shape of the satellite cells in clusters on myofibers. Single and multichannel images of the different focal planes were analyzed to determine the number of cells within a cluster and for other experimental parameters, such as the myogenic status of satellite cells.

Immunofluorescence

For IHC and ICC studies, frozen TA muscle sections (9 or 10 μm-thick sections) or myoblast/myofiber cultures, respectively, were fixed in 4% paraformaldehyde (PFA) in phosphate buffered saline (PBS), blocked in 2% bovine serum albumin (BSA) in PBS for 1 h, and subsequently incubated with anti-Pax7 (1:5-1:10, DSHB, Cat# pax7), anti-eMyHC (1:200, DSHB, Cat# F1.652), anti-MyoD (1:200, Santa Cruz Biotechnology, Cat# sc-377460), anti-Laminin (1:150, Cell Signaling Technology, Cat# L9393), or anti-p-p65 (1:100, Cell Signaling Technology, Cat# 3033) in blocking solution at 4°C overnight under humidified conditions. The sections were washed briefly with PBS before incubation with Alexa Fluor® 488 (1:2000, Thermo Fisher Scientific, Cat# A-11034) and Alexa Fluor® 594 (1:2000, Thermo Fisher Scientific, Cat# A-11037) secondary antibody for 1 h at room temperature and then washed three times for 5 min with PBS. Nuclei were visualized by counterstaining with DAPI for 5 min.

EdU and TUNEL staining

To determine the proliferation of satellite cells, EdU staining was performed using a commercially available kit and following the protocol from the manufacturer (Click-iT EdU Cell Proliferation Assay kit, Invitrogen). Briefly, during the last 90 min of incubation, 10 μM EdU (Invitrogen) was added in culture medium. The cells were fixed in 4% PFA. EdU visualization occurs through a click reaction, whereby the EdU is covalently bonded to a fluorescent label (Alexa Fluor® 488 Azide). Nuclei were visualized by counterstaining with DAPI for 5 min. TUNEL staining was performed following a protocol from the manufacturer (in situ Cell Death Detection Kit, Sigma Aldrich). Briefly, myofiber cultures were fixed in 4% PFA and permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate, and then incubated in TUNEL reaction mixture for 60 min at 37°C.

Flow cytometry

Apoptosis was assessed by annexin V/propidium iodide (PI) staining followed by FACS according to the manufacturer’s instructions (BD Biosciences).

Western blot

Relative levels of various proteins were quantified by performing western blot. Briefly, skeletal muscle of mice or cultured myoblasts was washed with PBS, homogenized in lysis buffer (50 mM Tris-Cl (pH 8.0), 200 mM NaCl, 50 mM NaF, 1 mM dithiothreitol, 1 mM sodium orthovanadate, 0.3% IGEPAL, and protease inhibitor cocktail). Approximately, 100 μg of protein was resolved on each lane on a 10% or 12% SDS-PAGE, electro-transferred onto a nitrocellulose membrane and probed using various primary antibodies: anti-total IKKβ (1:2000, Cell Signaling Technology, Cat# 2684), anti-phospho-IκBα (1:2000, Cell Signaling Technology, Cat# 2859), anti-total IκBα (1:2000, Cell Signaling Technology, Cat# 4812), anti-phospho-p65 (1:2000, Cell Signaling Technology, Cat# 3033), anti-total p65 (1:2000, Cell Signaling Technology, Cat# 8242), anti-cyclin D1 (1:1000, Santa Cruz Biotechnology, Cat# sc-717, RRID), anti-cyclin A (1:1000, Santa Cruz Biotechnology, Cat# sc-596), anti-BAX (1:1000, Cell Signaling Technology, Cat# 2772), anti-Bcl2 (1:1000, BD Biosciences, Cat# 554279), anti-cleaved caspase-3 (1:1000, Cell Signaling Technology, Cat# 9664), anti-cleaved PARP (1:1000, Cell Signaling Technology, Cat# 9544), anti-Pax7 (1:1000, DSHB, Cat# pax7), anti-MyoD (G1) (1:1000, Santa Cruz Biotechnology, Cat# sc-377460), anti-myogenin (1:1000, Santa Cruz Biotechnology, Cat# sc-576), and anti-MyHC (1:2000, DSHB, Cat# MF20). As a loading control, membranes were stripped and re-probed with anti-GAPDH (1:2000, Cell Signaling Technology, Cat# 2218) or anti-tubulin (1:2000, Cell Signaling Technology, Cat# 2125). Detection of proteins was enhanced by chemiluminescence.

RNA isolation and quantitative qRT-PCR assay

RNA isolation and qRT-PCR were performed as previously described (Hindi and Kumar, 2016).

Statistical analyses

For the sake of transparency, results were expressed as box-and-whisker plots with the box comprised of the first, second, and third quartiles, and the lower and upper whiskers corresponding to the minimum and maximum values, respectively, to display the entire range of data. Statistical analyses between two groups used unpaired two-tailed Student’s t-test to compare quantitative data populations with normal distribution and equal variance. A value of P < 0.05 was considered statistically significant unless specified otherwise for comparisons made between two groups.

Acknowledgements

We thank Prof. Michael Karin (University of California, San Diego) for providing the IKKβ^f/f mice.

Funding

This work was supported by NIH grants AR068313 and AR059810 to A.K. and AR069985 to S.M.H.

Conflict of interest

none declared.

Author contributions

A.K. conceived and designed the work. A.R.S. and A.K. wrote the manuscript and all authors edited the manuscript. A.R.S., S.M.H., and G.X. performed the experiments.