Tumor necrosis factor α deficiency promotes myogenesis and muscle regeneration

Abstract

Tumor necrosis factor α (TNFα) exhibits diverse biological functions; however, its regulatory roles in myogenesis are not fully understood. In the present study, we explored the function of TNFα in myoblast proliferation, differentiation, migration, and myotube fusion in primary myoblasts and C2C12 cells. To this end, we constructed TNFα muscle-conditional knockout (TNFα-CKO) mice and compared them with flox mice to assess the effects of TNFα knockout on skeletal muscles. Results indicated that TNFα-CKO mice displayed phenotypes such as accelerated muscle development, enhanced regenerative capacity, and improved exercise endurance compared to flox mice, with no significant differences observed in major visceral organs or skeletal structure. Using label-free proteomic analysis, we found that TNFα-CKO altered the distribution of several muscle development-related proteins, such as Hira, Casz1, Casp7, Arhgap10, Gas1, Diaph1, Map3k20, Cfl2, and Igf2, in the nucleus and cytoplasm. Gene set enrichment analysis (GSEA) further revealed that TNFα deficiency resulted in positive enrichment in oxidative phosphorylation and MyoD targets and negative enrichment in JAK-STAT signaling. These findings suggest that TNFα-CKO positively regulates muscle growth and development, possibly via these newly identified targets and pathways.

INTRODUCTION

Tumor necrosis factor α (TNFα) is a cytokine that exists in secreted, membrane-bound, or intracellular soluble forms, exhibiting biological activities across various cell types (Horiuchi et al., 2010; Kriegler et al., 1988). By binding to receptors on the cell membrane, it activates different cascading signaling pathways to regulate processes such as cell growth, differentiation, apoptosis, and inflammatory responses (Jang et al., 2021). Skeletal muscle formation primarily depends on complex processes such as myoblast proliferation, migration, differentiation, cytoskeletal rearrangement, and myotube fusion, which are regulated by multiple genes. Although TNFα has been extensively studied in the context of immunology, recent studies have also highlighted its significant roles in skeletal muscle development, including myogenesis, muscle repair, muscle disease, and muscle atrophy (Di Credico et al., 2023; Fang et al., 2021; Li et al., 2020; Shirakawa et al., 2021).

As an endogenous muscle factor, TNFα is expressed in cultured myoblasts and strongly activated in injured muscle tissue (Collins & Grounds, 2001; Li & Schwartz, 2001). However, its role in myoblast proliferation and muscle regeneration remains controversial. Overexpression of TNFα can substantially reduce the number of Pax7⁺ cells (Li et al., 2020). Conversely, the exogenous addition of TNFα protein can stimulate myoblast proliferation and satellite cell activation (Alvarez et al., 2020; Li, 2003; Shirakawa et al., 2021). Muscle regeneration is impaired in both TNFα knock-in mice (Li et al., 2020) and TNFα receptor double-knockout mice (p55^−/−p75^−/−) following cardiotoxin (CTX)-induced muscle injury (Chen et al., 2005). Furthermore, TNFα consistently inhibits the accumulation of muscle-specific proteins in differentiated C2C12 cells. Notably, it promotes MyoD degradation through the ubiquitin-proteasome pathway, reducing the half-life and abundance of the MyoD protein (Langen et al., 2004). TNFα also suppresses myotube formation (Meyer et al., 2015). Interestingly, low concentrations of exogenous TNFα protein can promote myogenesis, while high concentrations induce the opposite effect (Chen et al., 2007). Additionally, triptolide can prevent lipopolysaccharide-induced skeletal muscle atrophy by gradually reducing TNFα levels (Fang et al., 2021). However, the physiological role of TNFα in skeletal muscle development and associated molecular mechanisms remain unclear.

Regarding immunity, there is a well-established consensus on the role of the TNFα signaling pathway. The binding of TNFα to its receptor ultimately leads to NF-κB-dependent cellular responses, cell apoptosis, or necrosis (Hsu et al., 1995; Li & Schwartz, 2001; Pasparakis & Vandenabeele, 2015; Varfolomeev & Vucic, 2018). TNFα also activates three groups of MAP kinases, including extracellular signal-regulated, p38 MAP, and c-Jun NH2-terminal kinases (Sabio & Davis, 2014). However, these known signaling pathways do not fully explain the mechanisms by which TNFα regulates muscle development. Therefore, it is plausible that additional, muscle-specific regulatory pathways remain undiscovered and warrant further research.

In the current study, we explored the role of TNFα and its associated mechanisms in muscle development. We systematically examined the effects of TNFα on myoblast proliferation, migration, differentiation, and myotube fusion. We successfully established TNFα muscle-conditional knockout (TNFα-CKO) mice in vivo and evaluated their skeletal muscle development, muscle regenerative capacity following CTX-induced damage, and exercise endurance. Mechanistically, we identified potential new targets and pathways influenced by TNFα. Our results provide a theoretical foundation for further studies on the regulation of muscle development by TNFα and propose a novel strategy for myopathy treatment.

MATERIALS AND METHODS

C2C12 cell culture and RNA interference

The C2C12 myoblast cell line (Type Culture Collection of the Chinese Academy of Sciences, China) was cultured in Dulbecco’s modified Eagle medium (DMEM, Gibco, USA) with 10% fetal bovine serum (FBS, Gibco, USA) and 1% penicillin/streptomycin (Gibco, USA). To induce differentiation, the cells were incubated in DMEM supplemented with 2% horse serum (Gibco, USA) and 1% penicillin/streptomycin. To interfere with TNFα expression, we designed small interfering RNAs (siRNAs) based on the TNFα gene sequence and selected the top candidates with the highest scores (siRNA-403, siRNA-509, siRNA-631) from the design website. The number assigned to each siRNA corresponds to its position relative to the start site of TNFα. The TNFα siRNA sequences were: siRNA-403, sense: CUCAGAUCAUCUUCUCAAATT, antisense: UUUGAGAAGAUGAUCUGAGTT; siRNA-506, sense: CCAACGGCAUGGAUCUCAATT, antisense: UUGAGAUCCAUGCCGUUGGTT; and siRNA-631, sense: AGCCGAUUUGCUAUCUCAUTT, antisense: AUGAGAUAGCAAAUCGGCUTT. Cell transfection was performed using Lipofectamine 2000 in accordance with the manufacturer’s instructions (Invitrogen, USA).

Animals

The floxed-TNFα mouse line (C57BL/6N background) was established by Cyagen Biosciences (China) by inserting LoxP site flanking exons 1 and 4 of the TNFα gene. The TNFα-CKO were generated by crossing TNFα^flox/flox and Myf5-Cre mice obtained from Jackson Laboratory (USA). The mouse genotype was determined via polymerase chain reaction (PCR) using tail DNA with the following primers: Cre-F: 5′-AACCAGAGACTCCCCAAGGT-3′, Cre-R: 5′-ACGAAGTTATTAGGTCCCTCGAC-3′, Cre-WR: 5′-CGGCTCTTAAAGCAATGGTC-3′, F2: 5′-GAGACCTAAGCTCTATCCCTGCTC-3′, R2: 5′-TCACATTTTCAGCCCTGTGTCTTT-3′). Cre⁻ mice had an amplicon of 240 bp, while Cre⁺ mice possessed an amplicon of 120 bp. Furthermore, flox^+/+ mice showed a single band at 205 bp, while flox^−/− mice showed a single band at 138 bp. Consequently, flox^+/+Cre⁻ mice served as control TNFα^flox/flox mice, while floxflox^+/+Cre⁺ represented TNFα-CKO mice. For muscle regeneration assays, 50 μL of CTX (20 μmol/L, Sigma-Aldrich, USA) was injected into the tibialis anterior (TA) muscles of 2-month-old mice. All animal procedures were performed in accordance with the guidelines for good laboratory practice. The animals were provided with sufficient water, standard chow diet, and appropriate feeding conditions. Furthermore, the animal care and experimental procedures were conducted based on the National Research Council Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at China Agricultural University, Beijing, China (approval number: AW80203202-1-1).

Primary myoblast purification and culture

Mouse primary myoblasts were isolated from 6-week-old C57BL/6N mice. Briefly, mouse muscles were minced and digested in 0.2% type II collagenase (Sigma-Aldrich, USA) and 2.5 U/mL dispase (Roche Applied Science, USA) for 1 h. After digestion, the cells were filtered and centrifuged at 1500 ×g for 5 min at room temperature, then cultured in growth medium (F10 supplemented with 20% FBS, 1% penicillin-streptomycin, 5 ng/mL basic fibroblast growth factor, and 1% chicken embryo extract) on collagen-coated cell culture plates at 37℃ and 5% CO₂.

EdU and CCK8 assays

5-Ethynyl-2-deoxyuridine (EdU) and Cell Counting Kit-8 (CCK8) assays were conducted in accordance with the instructions provided with the EdU (Ribobio, China) and CCK8 (Solarbio, China) assay kits, respectively. In the EdU assay, cells transfected for 24 h were incubated with 50 μmol/L EdU for 2 h, then fixed with 4% paraformaldehyde (PFA), permeabilized with 0.5% Triton X-100, and incubated with Apollo reaction buffer. EdU-labeled cells and 4’,6-diamidino-2-phenylindole (DAPI)-stained (Solarbio) nuclei were photographed under a microscope (ZEISS, Germany). For the CCK8 assay, transfected cells were incubated with 100 μL of 10% CCK8 solution at 37℃ for 1 h. Absorbance was measured at 450 nm using a multimode reader (Biotek, USA).

Cell wounding and Transwell migration

For wounding, the cells (24–48 h after transfection) were scratched with a pipette tip of the same width. Floating cells were washed with phosphate-buffered saline (PBS) and the medium was then replaced with DMEM plus 2% FBS. At 24 h post-wounding, cell migration distance was recorded. For the Transwell assay, equal amounts of cells were cultured in the lower and upper Transwell chambers for 12 h, with crystal violet then used to stain the migrated cells induced by the serum gradient.

Immunofluorescence (IF)

Myoblasts or muscle tissues were fixed in 4% PFA for 20 min and permeabilized for 15 min in 0.3% Triton X-100 with PBS. The samples were then blocked in blocking buffer (Beyotime, China) for 2 h at room temperature and incubated overnight at 4℃ with the following primary antibodies: MyHC (M4276, 1:200; Sigma-Aldrich, USA), eMyHC (F1.652, 1:50; Developmental Studies Hybridoma Bank, USA), and laminin (L9393, 1:1 000; Sigma-Aldrich, USA). After incubation, the samples were washed with PBS and incubated with fluorescently labeled secondary antibodies (SA00013-2 and SA00013-3, 1:400; Proteintech, USA) for 1 h at room temperature in darkness. The samples were then washed with PBS and the cell nuclei were labeled using DAPI for 5 min. Finally, immunofluorescence images were visualized using a fluorescence microscope (Model Q500MC; Leica Image Analysis System, Germany).

Magnetic resonance imaging (MRI)

MRI was performed under anesthesia using 2% isoflurane on an MRI system equipped with a 7 T/16-cm wide-bore instrument. Thus, 20 contiguous coronal slices (thickness=1 mm), covering the region from the knee to the ankle, were acquired. The scan parameters were: echo time (TE)=7.8 ms, repetition time (TR)=500 ms, field of view=100 mm×60 mm, matrix size=256×256, and acquisition time=20 min. The images were analyzed using VNMRJ v.4.0 (Varian Associates, USA).

Hematoxylin-eosin staining (H&E)

After fixation in 4% PFA, tissues were processed for routine paraffin histology and stained with H&E. The sections were then examined under an inverted microscope.

Swimming exhaustion test

Several large beakers were filled with water and maintained at a constant temperature of 28℃. The mice were fasted overnight (15 h), and a piece of metal (7% of body weight) was attached to their tails to ensure that they would sink without swimming. The duration each mouse swam until exhaustion was recorded.

qPCR analysis

Total RNA was isolated from myoblasts and muscle tissues using a TRNzol Universal Reagent (Tiangen, China). Subsequently, 2 μg of total RNA was reverse transcribed into cDNA using a FastKing gDNA Dispelling RT SuperMix Kit (Tiangen, China) according to the manufacturer’s instructions. qPCR was then performed using the Universal SYBR Green Fast qPCR Mix system (ABclonal, China). The primer sequences used are shown in Supplementary Table S1.

Western blot analysis

The C2C12 cells and primary myoblasts were lysed in RIPA buffer (Beyotime, China) or a nuclear/cytoplasmic protein extraction kit (Beyotime, China). Proteins were then resolved using sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were then blocked in QuickBlock buffer (Beyotime, China) for 30 min at room temperature and incubated overnight at 4℃ with primary antibodies, including: mouse anti-MyoG (ab1835, 1:500; Abcam, USA), mouse anti-MyHC (M4276, 1:1 000; Sigma-Aldrich, USA), rabbit anti-TNFα (A11534, 1:1 000; ABclonal, China), mouse anti-Casz1 (sc-398303, 1:1 000; Santa Cruz Biotechnology, USA), rabbit anti-Hira (A12527, 1:1 000; ABclonal, China), rabbit anti-Diaph1 (A11596, 1:1 000; ABclonal, China), rabbit anti-Tnnt1 (A10354, 1:1 000; ABclonal, China), rabbit anti-Fgf21 (A3908, 1:2 000; ABclonal, China), rabbit anti-Map3k20 (A14513, 1:1 000; ABclonal, China), rabbit anti-Cfl2 (A8009, 1:2 000; ABclonal, China), rabbit anti-Histone H3 (A2348, 1:5 000; ABclonal, China), rabbit anti-GAPDH (2118s, 1:5 000; Cell Signaling Technology, USA), and rabbit anti-β-Tubulin (ab6046, 1:5 000; Abcam, USA). Incubation with the secondary antibodies was performed for 1 h at room temperature.

Proteomic analysis

Flox and TNFα-CKO primary myoblasts were induced to differentiate for 2 d. Thereafter, nuclear and cytoplasmic proteins were extracted using nuclear/cytoplasmic protein extraction kits (Beyotime, China). Briefly, the samples were sonicated three times with 1% protease inhibitor cocktail on ice using a high intensity ultrasonic processor (Scientz, China). The supernatant was then obtained via centrifugation at 12 000 ×g and 4℃ for 10 min. Protein concentration was measured using a BCA kit (Beyotime, China).

The protein samples were added to four volumes of pre-cooled acetone and precipitated at −20℃ for 2 h. Thereafter, they were re-dissolved in 200 mmol/L triethylammonium bicarbonate and ultrasonically dispersed. Trypsin was added at a 1:50 trypsin-to-protein mass ratio for the first digestion overnight. The samples were then reduced with 5 mmol/L dithiothreitol for 30 min at 56℃ and alkylated with 11 mmol/L iodoacetamide for 15 min at room temperature in the dark.

To dissolve and separate the tryptic peptides, an EASY-nLC 1200 UPLC system (Thermo Fisher Scientific, USA) was used at a constant flow rate of 500 nL/min. The separated peptides were then analyzed using an Orbitrap Exploris 480 mass spectrometer (Thermo Fisher Scientific, USA) with a nano-electrospray ion source. The resulting tandem mass spectrometry (MS/MS) data were processed using the Proteome Discoverer search engine (v.2.4.1.15, Thermo Fisher Scientific, USA). The MS/MS data were searched against the Mus_musculus_10090_SP_20230103.fasta database (17 132 entries) concatenated with the reverse decoy database. Trypsin (Full) was specified as the cleavage enzyme, allowing up to two missed cleavages. The false discovery rates (FDRs) for proteins, peptides, and peptide-spectrum match were adjusted to <1%. The minimum peptide length was set to 6. Label-free measurements were performed using the PTM Bio System (PTM-Biolabs, China). Data were obtained via ProteomeXchange with the identifier PXD046525.

Gene lists were analyzed using GSEA v.4.3.2 from the Broad Institute (Subramanian et al., 2005). The submitted gene list was normalized by applying the –log₁₀ transformation to the count values, assigned as positive or negative based on the direction of the fold change. All other GSEA parameters were set to default. The annotated HALLMARK, DELASERNA, and KEGG gene sets obtained from the DAVID and MSigDB databases were used as enrichment inputs.

Statistical analysis

All data were analyzed using SPSS v.22.0 (IBM, USA) and presented as mean±standard deviation (SD). P-values were calculated by performing a two-tailed Student’s t-test, and statistical significance was indicated as follows: ^*: P<0.05, ^**: P<0.01, and ^***: P<0.001.

RESULTS

TNFα knockdown promoted cell proliferation, migration, differentiation, and myotube fusion in C2C12 cells

Skeletal muscle formation primarily relies on processes such as myoblast proliferation, migration, differentiation, and myotube fusion. We successfully reduced TNFα expression in C2C12 cells using RNA interference fragments, with siRNA-631 showing the highest interference efficiency (Figure 1A). Regarding proliferation, the CCK8 assay showed higher absorbance for the siRNA-TNFα (si-TNFα) transfection group than for the siRNA negative control group (si-NC) (Figure 1B). TNFα knockdown also significantly increased the number of EdU⁺ cells (Figure 1C) and up-regulated the expression of proliferation marker genes, Ki67, Cyclin B, and CDK4 (Figure 1D). Cell wounding (Figure 1E) and Transwell (Figure 1F) assays demonstrated that TNFα knockdown enhanced myoblast migration and promoted C2C12 cell differentiation into more numerous, longer, and thicker myotubes (Figure 1G), with most myotubes exhibiting multi-nucleation (as indicated by white arrows in Figure 1H). TNFα knockdown also significantly up-regulated the expression of myogenic marker genes, MyoG and MyoD (Figure 1I), as well as fusion marker genes, β-1 integrin and Myomaker (Figure 1J). Muscle hypertrophy and degradation are crucial processes in muscle development (Bodine et al., 2001). Here, the qPCR results showed that TNFα knockdown significantly promoted the expression of muscle hypertrophy genes (Fst and Nog), while inhibiting the expression of myostatin and muscle degradation marker genes (Atrogin1, Bmp4, and Foxo3) (Figure 1K). Overall, TNFα knockdown promoted myoblast proliferation, migration, differentiation, and fusion in C2C12 cells, implying that TNFα plays a significant regulatory role in the growth and development of skeletal muscles.

Figure 1.

Promotion of myoblast proliferation, migration, differentiation, and myotube fusion in C2C12 cells through TNFα knockdown

A: Detection of interference efficiency for different siRNA-TNFα fragments. Cells were transfected for 24 h, then collected for qPCR. B: CCK8 assay of si-NC and si-TNFα C2C12 cells. C: Representative photographs of EdU-stained si-NC and si-TNFα cells. EdU⁺ staining (red) of S-phase cells and DAPI staining (blue) of cell nuclei are shown (Scale bar: 400 μm), and proliferative cells were quantified as a ratio of EdU⁺/DAPI⁺. D: Analysis of expression levels of proliferation marker genes. C2C12 cells were transfected with si-NC and si-TNFα for 24 h, then collected for qPCR. E, F: Cell wounding (E) and Transwell (F) analysis of migration ability of myoblasts. Migrated cells were stained purple with crystal violet and counted. G: Immunofluorescence staining for MyHC (red) in si-NC and si-TNFα C2C12 cells after culturing for 4 d in differentiation medium (Scale bar: 400 μm). H: Myoblast fusion analysis via immunofluorescence staining for MyHC. White arrows represent multinucleated myotubes (Scale bar: 200 µm). I–K: Analysis of expression levels of myogenic differentiation (I), fusion (J), hypertrophy, and degradation marker genes and myostatin (K) in si-NC and si-TNFα C2C12 cells using qPCR. Transfected cells were cultured in differentiation medium for 4 d. Data are presented as mean±standard deviation (SD); n=3. ^*: P<0.05; ^**: P<0.01; ^***: P<0.001 (Student’s t-test).

TNFα-CKO accelerated mouse muscle development, improved exercise endurance, and enhanced muscle regenerative capacity

To investigate the role of TNFα in muscle development, we generated conditional knockout mice by selectively deleting TNFα in skeletal muscles using Myf5-Cre/TNFα and flox/flox alleles. The TNFα-CKO mice exhibited normal birth and survival, with no apparent abnormalities in appearance (Figure 2A). The male TNFα-CKO mice demonstrated significantly higher body weights (Figure 2B) and body lengths (Figure 2C) than the flox control mice, while the female TNFα-CKO and flox mice showed no significant differences in body length (Supplementary Figure S1A). Additionally, the TNFα-CKO mice developed more robust forelimbs and hindlimbs (Figure 2D). MRI analysis revealed an increase in hindlimb muscles and subcutaneous fat content in TNFα-CKO mice, with no significant changes in skeletal volume between the two groups (Figure 2E). Furthermore, the muscle weights of quadriceps (Quad), tibialis anterior (TA), extensor digitorum longus (EDL), gastrocnemius (GAS), soleus (SOL), trapezius (Trap), and triceps were higher in 2-month-old TNFα-CKO male mice than in 2-month-old male flox mice (Figure 2F and 2G). Examination of TA and GAS weights in 3-month-old mice showed similar results to those obtained at 2 months (Figure 2H). Muscle weight analyses for female and male TNFα-CKO mice were similar (Supplementary Figure S1B). Subcutaneous fat (SAT), epididymal white adipose tissue (eWAT), and perirenal white adipose tissue (pWAT) weights were also significantly increased in 2- and 3-month-old TNFα-CKO mice, while differences in interscapular brown adipose tissue (iBAT) weight were not significant (Figure 2I, J). The change in fat tissue weight in female TNFα-CKO mice was similar to that in male mice at 3 months, although this was not the case for male and female mice at 2 months (Supplementary Figure S1C). Heart, liver, spleen, lung, and kidney weights revealed that TNFα-CKO did not affect the development of major visceral organs in either male or female mice (Supplementary Figure S1D, E). Thus, TNFα-CKO accelerated skeletal muscle development and adipose tissue formation, which may be responsible for the changes in mouse body weight observed in this study.

Figure 2.

Improved development speed, exercise endurance, and regeneration ability of TNFα-CKO mouse muscles

A: Gross body morphology of TNFα-CKO and flox mice. B, C: Body weight (B) and length (C) of TNFα-CKO and flox male mice at 2, 3, and 4 months (n≥50). D: Representative photographs of forelimbs and hindlimbs of TNFα-CKO and flox mice. E: Magnetic resonance imaging (MRI) of hindlimbs, subcutaneous fat (SAT), muscle, and bone volume. F–H: Representative photographs (F) and weight analysis (G and H) of skeletal muscles of TNFα-CKO and flox mice. BW, body weight. I, J: Representative photographs (I) and weight analysis (J) of adipose tissue in TNFα-CKO and flox mice. K: H&E staining of TA and GAS muscles in TNFα-CKO and flox mice (Scale bar: 100 µm). CSA, cross-sectional area. L: Swimming exhaustion time assay for TNFα-CKO and flox mice (n≥10). M: Analysis of expression level of Myh4 in EDL and Myh7 in SOL using qPCR. Mice were sacrificed immediately after swimming exhaustion tests to collect SOL samples. N: Analysis of expression level of TNFα after CTX-induced injury at d 0, 3, 5, 7, and 14. O: MRI results and damage volume data 5 d after CTX-induced injury. Area circled by red line indicates damaged muscle tissue (n=6). P: H&E staining of GAS 3 and 5 d after CTX-induced injury (Scale bar: 210 µm). Q: Immunofluorescence staining for MyoG (red) and laminin (green) in GAS 5 d after CTX-induced injury (Scale bar: 400 µm). Data are presented as mean±SD (n=6). *: P<0.05; **: P<0.01; ***: P<0.001; ns: Not significant (Student’s t-test).

The H&E staining results revealed larger GAS and TA muscle fiber cross-sectional areas in TNFα-CKO mice (Figure 2K), consistent with the greater muscle weight observed. A well-developed skeletal muscle system is often associated with endurance. As expected, swimming exhaustion tests showed significantly improved endurance for male TNFα-CKO mice. Interestingly, however, no significant differences in endurance were observed between female TNFα-CKO and flox mice (Figure 2L). TNFα-CKO also significantly down-regulated the expression of the fast-twitch muscle fiber marker Myh4 and up-regulated the expression of the slow-twitch muscle fiber marker Myh7 (Figure 2M). These findings suggest that TNFα-CKO accelerates the development of mouse skeletal muscles and enhances endurance.

Muscle regenerative capacity after injury is an important indicator of the muscle development process, reflecting adaptability to external stimuli and self-repair abilities. To simulate muscle damage, we injected mice with CTX. In flox mice, qPCR analysis revealed a significant increase in TNFα expression induced by CTX, accompanied by the entire regeneration process (Figure 2N). MRI analysis further revealed a smaller injury area in TNFα-CKO mice 5 d after CTX-induced muscle injury (Figure 2O). Both H&E (Figure 2P) and IF staining (Figure 2Q) demonstrated that, compared to flox mice, TNFα-CKO mice exhibited more intact muscle fiber structures 3 or 5 d after CTX-induced injury. This was accompanied by a clearer laminin outline and the up-regulation of embryonic myosin heavy chain (eMyHC), a key marker of muscle regenerative capacity. These observations suggest that TNFα-CKO mice possess a stronger muscle regenerative ability than flox mice.

Functional validation of TNFα in primary myoblasts of TNFα-CKO mice

To confirm the in vitro function of TNFα, we established primary myoblasts from flox and TNFα-CKO mice. As expected, TNFα mRNA and protein levels were extremely low in the TNFα-CKO primary myoblasts (Figure 3A), which also exhibited a higher number of EdU⁺ cells (Figure 3B) and significantly up-regulated CDK4 and Cyclin B expression (Figure 3C). Bright-field microscopy also indicated that TNFα deletion facilitated the formation of myotubes following the induction of primary myoblast differentiation for 2 d (Figure 3D). Furthermore, IF staining further revealed that TNFα-CKO led to myotube hypertrophy, as demonstrated by the formation of a greater number of longer and thicker myotubes marked by MyHC (Figure 3E–H). Concurrently, the mRNA and protein levels of myogenic marker genes MyHC, MyoG, and MyoD were significantly up-regulated (Figure 3I, J). In summary, similar to our observations for C2C12 cells, deletion of TNFα significantly promoted the proliferation and differentiation of primary myoblasts.

Figure 3.

Functional verification of TNFα expression in primary myoblasts

A: Detection of TNFα mRNA level in flox and CKO primary myoblasts using qPCR. B: Representative photographs of EdU-stained flox and TNFα-CKO primary myoblasts. EdU⁺ staining (red) of S-phase cells and DAPI staining (blue) of cell nuclei are shown (Scale bar: 400 μm), and proliferative cells were quantified as a ratio of EdU⁺/DAPI⁺. C: Analysis of expression levels of proliferation marker genes in flox and TNFα-CKO primary myoblasts using qPCR. Same number of flox and TNFα-CKO primary myoblasts were inoculated in plates and cultured in growth medium. D: Differentiation of primary myoblasts observed under a microscope 2 d after incubation in differentiation medium (Scale bar: 520 μm). E: Immunofluorescence staining of MyHC in flox and TNFα-CKO primary myoblasts 2 d after incubation in differentiation medium (Scale bar: 130 μm). Fusion index (F), myotube diameter (G), and myotube length (H) were quantified. I: Analysis of expression levels of myogenic marker genes in flox and TNFα-CKO primary myoblasts using qPCR. Myoblasts were differentiated for 2 d. J: Analysis of protein expression levels of MyHC, MyoG, and TNFα in flox and TNFα-CKO primary myoblasts. Myoblasts were differentiated for 2 d. Data are presented as mean±SD (n=3). ^*: P<0.05; ^**: P<0.01; ^***: P<0.001 (Student’s t-test).

Identification of new targets for TNFα in myoblast cells

Current understanding, such as TNFα facilitating NF-κB nuclear entry and activating certain MAP kinases, fails to fully elucidate the role of TNFα in myoblasts. To explore this further, we isolated the nucleus (N) and cytoplasm (C) of flox and TNFα-CKO primary myoblasts in the early stages of differentiation to identify new targets using label-free proteomic profiles. To illustrate changes in the distribution of differentially abundant proteins (DAPs; fold-change>1.5 or <0.67, P-value<0.05) between the cell nucleus and cytoplasm, we presented up-regulated and down-regulated proteins separately. Most DAPs exhibited changes exclusively in either the cell nucleus or cytoplasm, while some exhibited synchronous changes in both compartments (Figure 4A). Notably, a small subset of DAPs, including six nucleus-down-regulated and cytoplasm-up-regulated proteins and one nucleus-up-regulated and cytoplasm-down-regulated protein, displayed a pattern reminiscent of translocation between the nucleus and cytoplasm (Figure 4A).

Figure 4.

TNFα-CKO-induced changes in protein distribution and pathway enrichment

A: Venn diagram showing overlap between up-regulated and down-regulated differentially abundant proteins (DAPs) in nucleus (N) and cytoplasm (C) based on fold change>1.5 and P<0.05. B: Heatmap showing degree of changes in DAP levels due to TNFα-CKO. Scale size was normalized and shown by area. C: Western blots showing degree of changes in levels of certain proteins related to muscle development. Flox and TNFα-CKO primary myoblasts were induced to differentiate for 2 d. Histone 3 and GAPDH were used as reference proteins for N and C, respectively. Corresponding quantifications are shown in (D). D: Relative level indicates target protein/internal reference ratio of densitometry on western blot bands (n=3). Data are presented as mean±SD. ^*: P<0.05; ^**: P<0.01; ^***: P<0.001 (Student’s t-test). E–G: GSEA assays of proteomic data showing positive enrichment of MyoD targets and oxidative phosphorylation and negative enrichment of JAK-STAT signaling pathway following TNFα-CKO.

We used a heatmap to display changes in the abundances of selected proteins of interest, some of which may be potential targets of TNFα (Figure 4B). Compared to the flox group, Hira and Casz1 were significantly up-regulated in the nucleus of the TNFα-CKO group, with no detection in the cytoplasm, whereas Casp7, which was down-regulated, was only present in the cytoplasm. Arhgap10, up-regulated in the nucleus and down-regulated in the cytoplasm, exhibited the opposite shuttle direction to Gas1, which was down-regulated in the nucleus and up-regulated in the cytoplasm. Furthermore, Fgf21, Des, Igf2, and Igfbp5 were up-regulated in both the nucleus and cytoplasm, while fhl1, Pik3r1, and Myh4 were down-regulated in both compartments, implying corresponding changes in their total cellular levels. Map3k20, Cfl2, Tnni1, and Myom3 were significantly up-regulated, while Stat1, Stat2, Diaph1, Pdcd4, and Ripk1 were significantly down-regulated in the nucleus, with no significant changes in the cytoplasm (Figure 4B). These alterations in protein abundance in the cell nucleus suggest potential regulatory roles in myogenesis. Additionally, western blot analysis (Figure 4C) using Histone 3 and GAPDH as reference proteins for the nucleus and cytoplasm, respectively, confirmed the expression of several proteins related to muscle development. Densitometry on the western blot bands (Figure 4D) indicated that the changes in protein abundance closely matched the heatmap results, confirming the reliability of the proteomic data.

GSEA can assess the interactions between predefined gene lists and the standard hallmark collection, potentially uncovering biological pathways not limited to DAPs. As expected, TNFα deficiency resulted in negative enrichment of the TNF (normalized enrichment score (NES)=−1.89, P<0.001, FDR Q=0.004) and NF-κB (NES=−1.74, P=0.006, FDR Q=0.01) signaling pathways, accompanied by TNFα-induced apoptosis (NES=−2.12, P<0.001, FDR Q<0.001) (Supplementary Figure S2A). Furthermore, TNFα deficiency led to positive enrichment in MyoD targets (NES=2.39, P<0.001, FDR Q<0.001) (Figure 4E) and oxidative phosphorylation (NES=2.22, P<0.001, FDR Q<0.001) (Figure 4F). Conversely, it resulted in negative enrichment of the JAK-STAT signaling pathway (NES=−1.62, P=0.023, FDR Q=0.052) (Figure 4G). For instance, highly enriched and up-regulated DAPs, including Igf2, Igfbp5, Cfl2, Myh3, and Myl4, were identified in MyoD targets, while down-regulated DAPs, such as Jun, Stat1, Stat2, and Stat3, were observed in the JAK-STAT signaling pathway (Supplementary Figure S2B). These findings reveal novel TNFα-regulated targets and pathways that potentially mediate the effects of TNFα on muscle growth and development.

DISCUSSION

In this study, we explored the influence of TNFα on myoblast proliferation, migration, differentiation, and myotube fusion. Specifically, we assessed the effects of TNFα-CKO on muscle development and regeneration in mice, and identified novel TNFα-regulated targets and pathways, which may act as critical mediators in muscle development. These findings could significantly enhance our understanding of the role of TNFα in muscle development and provide a theoretical basis for molecular breeding and myopathy treatment.

Our results indicated that TNFα knockdown promoted the proliferation of C2C12 myoblasts, which was also confirmed in primary myoblasts. Previous studies have indicated that TNFα overexpression significantly reduces the number of Pax7⁺ cells (Li et al., 2020), while exogenous TNFα protein can stimulate myoblasts and satellite cell proliferation (Alvarez et al., 2020; Li, 2003; Shirakawa et al., 2021). These contrasting results may be due to the differing effects of high and low TNFα concentrations on myoblast proliferation or suggest distinct mechanisms of action between exogenously administered and endogenous TNFα. Additionally, we demonstrated the role of TNFα in myoblast migration and established TNFα-CKO mice to investigate the effects of TNFα on muscle development and regeneration, which has not been previously reported. Results indicated that the TNFα-CKO mice exhibited enhanced muscle regeneration after CTX-induced injury, consistent with the phenotype observed in TNFα knock-in mice (Li et al., 2020) but in contrast to the muscle regeneration impairment observed in TNFα receptor double-knockout mice (p55^−/−p75^−/−) (Chen et al., 2005). These findings, combined with the results of cell proliferation analysis, suggest the involvement of both autocrine TNFα and receptor-mediated TNFα in muscle development. Furthermore, we observed that TNFα-CKO accelerated fat tissue deposition in skeletal muscles, without affecting the development of bones and visceral organs. We inferred that TNFα-CKO influenced lipid deposition through paracrine or endocrine mechanisms. Therefore, the TNFα-CKO mice established in this study may serve as valuable models for further research into the roles of TNFα in muscle development or other physiological effects.

Identifying new targets and pathways of TNFα in skeletal muscle is essential for understanding its regulatory mechanisms in muscle development. In this study, we found that TNFα deletion significantly up-regulated the levels of histone chaperone Hira and zinc finger transcription factor Casz1 in the cell nucleus. Previous studies have suggested that Hira knockout can significantly reduce histone variant H3.3 deposition and H3K27ac modification in muscle gene regulatory regions, thereby inhibiting myoblast proliferation, differentiation, and skeletal muscle development and regeneration (Valenzuela et al., 2017). Casz1 promotes skeletal muscle differentiation by directly affecting regional epigenetic modifications, chromatin accessibility, and super-enhancer establishment (Liu et al., 2020). GSEA has shown positive enrichment of MyoD targets following Casz1 overexpression (Liu et al., 2020), consistent with our results (Figure 4E). These findings suggest that the enhancement of muscle development following TNFα knockout partially depends on the biological functions of Hira and Casz1. Arhgap10, identified from genome-wide selection signals in meat from goat breeds, promotes myoblast proliferation and differentiation (Yuan et al., 2023). Our results showed that TNFα knockout induced Arhgap10 translocation from the cytoplasm to the nucleus, indicating its activation and nuclear function following TNFα knockout. Conversely, Gas1 exhibited the opposite translocation direction compared to Arhgap10. Gas1 has been shown to negatively impact skeletal muscle stem cells and muscle regeneration (Li et al., 2019), consistent with the stronger muscle regeneration capacity observed in TNFα-CKO mice in this study. Cfl2 is an essential mediator for myogenic differentiation in C2C12 cells (Nguyen et al., 2020), and its loss leads to abnormal sarcomere formation (Balakrishnan et al., 2020). Map3k20 and Diaph1 positively and negatively regulate myogenic differentiation, respectively, by affecting MyoG expression (Saleh et al., 2019; Yan et al., 2022). Our results showed that TNFα knockout also resulted in the up-regulation of Cfl2 and Map3k20 and the down-regulation of Diaph1, theoretically favoring myogenic differentiation. In addition, TNFα-CKO significantly up-regulated slow-twitch muscle fiber markers (mRNA level of Myh7, protein levels of Tnnt1 and Tnni1) and down-regulated the protein and mRNA levels of the fast-twitch muscle fiber marker MYH4 (Figure 4B). These results align with the increased endurance observed in TNFα-CKO male mice. The absence of similar endurance improvements in female mice requires further investigation, as their phenotypic differences were also less pronounced compared to those in males.

Muscle fibers are highly metabolically active cells. During myogenic differentiation, myoblasts undergo autophagy-mediated mitochondrial remodeling, shifting from a highly glycolytic state to a pronounced reliance on oxidative phosphorylation (Sin et al., 2016). The significant positive enrichment in oxidative phosphorylation suggests that TNFα deficiency may affect the energy metabolism of myoblasts, in accordance with the relatively rigid structural demands imposed by muscle contraction. TNFα is known to significantly activate the JAK/STAT pathway in endothelial and microglial cells (Coelho-Santos et al., 2012; Kandhaya-Pillai et al., 2022). Typically, the JAK-STAT pathway inhibits myogenic differentiation (Jang & Baik, 2013). For instance, Stat1 interacts with MEF2, repressing its transcriptional activity and thereby suppressing myogenic differentiation (Xiao et al., 2011). Similarly, circSVIL, SOCS1, SOCS3, and PIAS1 promote myogenic differentiation by inhibiting the JAK-STAT pathway (Diao et al., 2009; Yue et al., 2022). Our results showed that TNFα deficiency led to negative enrichment in the JAK-STAT pathway, accompanied by decreased levels of Stat1, Stat2, and Stat3 in the cell nucleus, although the down-regulation of Stat1 was only 1.32-fold and did not meet differential criteria. Further research on the impact of TNFα on the phosphorylation of STAT proteins, or validation using activators and inhibitors of the JAK/STAT signaling pathway, may be beneficial. These findings provide initial insights into the relationship between TNFα and these pathways in skeletal muscle, offering new avenues for understanding the mechanisms by which TNFα regulates muscle development. These findings are integrated and presented in Figure 5.

Figure 5.

Diagram showing mechanism of action of TNFα-CKO in regulating muscle growth and development

Red and blue arrows indicate up-regulation and down-regulation of protein levels or pathway enrichment, respectively. Horizontal line indicates non-significant changes, pink arrow indicates direction of protein transfer. Figure was drawn by Figdraw.

SUPPLEMENTARY DATA

Supplementary data to this article can be found online.

Funding Statement

This work was supported by the Xizang Major Science and Technology Project (XZ202101ZD0005N), Yunnan Major Science and Technology Project (202302AE090015), National Key R&D Program of China (2023ZD04044-04), and National Natural Science Foundation of China (32060736)