Msi2 expression in myoblast differentiation

To examine the contribution of Msi2 in muscle development, we first analyzed its expression in mouse skeletal muscles. Western blot (WB) analysis showed that Msi2 is highly expressed in all three hindlimb muscles tested, soleus, tibialis anterior, and gastrocnemius (Fig 1A). Consistent with its high expression level in skeletal muscles, we found that two MRFs, Myog and MyoD1, bind to the Msi2 gene promoter region (Fig 1B). These data implied that Msi2 is activated during myogenesis and myocyte differentiation. Next, we examined Msi2 expression during myocyte differentiation in the murine C2C12 myoblast cell line. In an FBS-based media (Growth Media or GM), C2C12 cells expand as myogenic progenitors known as myoblasts. The progenitor cells undergo myocyte differentiation after incubation in media with horse serum (Differentiation Media or DM), and within 5 d, these myocytes form multinucleated elongated myotubes (Fig 1C). We found that Msi2 is expressed at a low level in myoblasts cultured in GM and that the expression is up-regulated within 24 h after the differentiation induction by DM (Fig 1D and E). Immunofluorescence staining of the Msi2 protein confirmed its up-regulation upon induction of differentiation (Fig 1F). Interestingly, C2C12 cells exhibited heterogeneity in the levels of Msi2 protein, and cells with higher Msi2 expression showed more differentiated morphology with more nuclear Msi2 staining at Days 3 and 6 after differentiation induction. Up-regulation of Msi2 was concomitant with that of Myog and myosin heavy chain (MHC; Fig 1D). In contrast, Msi1, the other family member of the mammalian Msi RBPs, is not expressed or up-regulated during C2C12 differentiation and no promoter binding of Myog or MyoD1 was observed (Fig S1A and B), indicating Msi2-specific roles in myogenic differentiation.

Msi2 is essential for the terminal differentiation of myocytes independent of MRF expression

Next, we investigated whether Msi2 is functionally required for myocyte differentiation using a lentiviral shRNA-mediated gene knockdown approach. 3 d after viral transfer of an shRNA against mouse Msi2 (shMsi2), the cells showed no detectable Msi2 protein expression, with knockdown (KD) continued up to 7 d (Fig 2A, Msi2). In Growth Media, shMsi2-expressing cells proliferate as much as those with control shRNA (shLuc), indicating that Msi2 is not essential for cell growth or survival of myoblasts (Fig S2A). After differentiation induction by switching from GM to DM, control shRNA-treated cells started to show elongated myotubes at as early as 3 d post-differentiation induction (Fig 2C, shLuc). The cells expressing shMsi2 also generated elongated cells, but they appeared shorter in length and their numbers were much less than those with shLuc even after 5 d of differentiation, indicating a significant impairment of myotube formation in the absence of Msi2 (Fig 2C, shMsi2). MHC protein was up-regulated in shMsi2-expressing cells, but the magnitude of expression induction was not as robust as that of shLuc cells (Fig 2A and B, MHC). In contrast, Myog expression was comparable between the control and the Msi2 knockdown cells, suggesting the lower MHC induction is not due to a defect in Myog expression (Fig 2A and B, Myog). Consistent with the immunoblots and immunostaining of MHC proteins (Fig 2A and D), mRNA analysis of mouse MHC genes Myh1 and Myh4 confirmed their attenuated expression by the Msi2 knockdown (Fig S2B). In contrast, Myod1 and Myog mRNA expression levels were comparable in both groups, suggesting the decreased Myh gene expression by Msi2 knockdown is downstream of the transcriptional regulation by the myogenic transcription factors Myod1 and Myog (Fig S2B). Moreover, the expression level of the myocyte fusion gene Myomixer was reduced by Msi2 knockdown at the late stage of differentiation (Fig 2E). In contrast, we observed no decrease in viabilities of the Msi2 knockdown cells, and the levels of the cleaved PARP and caspase-3 proteins were not affected by the Msi2 loss during the differentiation process, indicating the differentiation defect is not due to increased cell death (Fig S2C).

To address whether the differentiation is impaired at a specific stage during myotube formation, we monitored differentiation status using immunofluorescence microscopy (Fig 2D). Consistent with the phase-contrast images shown in Fig 2C, shLuc-treated control cells generated MHC-positive elongated myotubes. In contrast, cells treated for Msi2 knockdown cells failed to generate mature myocytes and most of the MHC-positive cells, if any, were shorter and smaller in size (Fig 2D). We observed some elongated cells, but interestingly, most of them were GFP-negative, indicating that the myocytes that matured were not expressing the shRNA against Msi2 (Fig 2D, arrows). Differentiation Indices, defined as the ratio of nucleus numbers in MHC-positive cells to the total nuclei of all GFP-positive cells in an analyzed field, were significantly lower in the shMsi2 group than in the control group (Fig 2F). In addition, the shMsi2 cells exhibited a significantly lower Fusion Index, defined as the number of myonuclei in one MHC⁺ cell, indicating an impact of Msi2 knockdown on the myocyte fusion step (Fig 2G), which is consistent with the decreased Myomixer expression (Fig 2E). Furthermore, electron microscopic images of the knockdown cells show distinct changes in mitochondria (Fig 2H). The mitochondria in control cells appeared elongated and contained complex crista structures with higher electron densities that are typical for mature myotubes. In contrast, those in the shMsi2 cells were round in shape and exhibited simple and scarce crista structures, implying defective mitochondrial remodeling in the Msi2 knockdown cells. Collectively, these results clearly demonstrated an essential role of Msi2 in the terminal differentiation of myocytes.

Msi2 expression rescues the differentiation defects conferred by Msi2 knockdown

To exclude the possibility of an off-target effect by the lentiviral shMsi2 and to further confirm the functional requirement of Msi2 in myocyte differentiation, we investigated whether Msi2 overexpression (Msi2 OE) can rescue the differentiation defect caused by the knockdown. The C2C12 cells expressing exogenous Msi2 with Msi2 knockdown (shMsi2 + Msi2 OE, Fig 3A-iv and B-iv) exhibited as many elongated MHC-positive myotubes as those in the control cells (shLuc + vector control, Fig 3A-i and B-i), whereas the cells with shMsi2 and an empty vector control showed smaller myotubes with low MHC expression (shMsi2 + vector control, Fig 3A-iii and B-iii). A few cells in this treatment group still showed MHC expression, but they were typically smaller in size or contained only one nucleus, suggesting no mature myocyte formation. The vector control cells with Msi2 knockdown exhibited Differentiation and Fusion Indices of 10% and 1.3, respectively, whereas the Msi2 OE with Msi2 knockdown increased these indices to 21% and 2.1 (Fig 3C). In addition, Msi2 OE also reverted the mitochondrial morphology changes observed in the Msi2 KD cells (Fig S3). Collectively, these results exclude the possibility of an off-target gene knockdown and demonstrate that Msi2 is required for differentiation and fusion into mature myocytes during the terminal differentiation step of myotube formation.

Enforced expression of Msi2 promotes myotube formation

During the above-mentioned experiment to assess on-target shRNA effects, we noticed that Msi2 overexpression alone increased the formation of MHC-positive, multinucleated elongated myotubes (shLuc + Msi2 OE, Fig 3A-ii and B-ii). In fact, Differentiation and Fusion Indices were significantly higher in cells expressing shLuc and Msi2 OE than in control cells with shLuc and vector. We repeated the Msi2 overexpression experiment without the co-introduction of any shRNA constructs and observed a similar enhancement of myocyte differentiation (Fig 3D). In addition, Msi2 overexpression increased overall mitochondrion numbers and promoted their fusion in both naïve and Msi2 knockdown cells (Figs 3E and S3). Immunoblot analysis showed that the MHC protein is further up-regulated by Msi2 OE compared with the vector control (Fig 3F, MHC), confirming the direct and functional impact of Msi2 on mature myocyte formation. Msi2 overexpression did not lead to an increase in Myog protein expression (Fig 3F, Myog), providing further evidence for a role of Msi2 downstream of the MRF Myog.

The RRM1 RNA-binding motif and the C-terminal domain of Msi2 are required for the promotion of myocyte differentiation

Because the Msi protein contains several domains for its molecular function (Okano et al, 2002; Kawahara et al, 2008), we next performed a structure–function relationship analysis. We introduced a retroviral vector encoding WT or a mutant Msi2 into C2C12 cells and observed their effects on differentiation (Fig 4A). The RNA binding–deficient mutant contains three point mutations at codons for Phe64, Phe66, and Phe69 in the RRM1 mutating each to a leucine residue, which leads to a loss of RNA-binding activity (Hattori et al, 2017). The Msi2 [1–190] mutant encodes only the first 190 aa and thus lacks the C-terminus domain. The ΔRRM1 and ΔRRM2 variants lack the first (21–101 aa) or second (110–187 aa) RRM, respectively. All the above Msi2 constructs are N-terminally fused with FLAG- and HA-tags. WB analysis using an anti-FLAG antibody showed the successful overexpression of all the Msi2 mutant proteins in the C2C12 cells (Fig 4B), except for the ΔRRM2 mutant that showed lower expression compared with the other mutants, despite its high retroviral infection (Fig 4C, GFP). On Day 3 after induction of differentiation, WT and ΔRRM2 Msi2 OE samples exhibited more MHC-positive cells, indicating enhanced differentiation (Fig 4C). Msi2 [1–190] and ΔRRM1 showed similar numbers of MHC-positive myocytes compared with a vector control. Interestingly, the expression of the RNA binding–deficient mutant resulted in attenuated differentiation compared with the vector control. Differentiation Indices were also significantly increased in WT and ΔRRM2 Msi2 OE samples but not in the others compared with the vector control (Fig 4D, Day 3 of differentiation). Because myocyte fusion mainly happens later in differentiation, Fusion Indices were comparable among all groups (Fig 4E). MHC detection by immunofluorescence correlated with MHC protein expression by WB, which also showed higher MHC expression levels in WT and ΔRRM2 than the others (Fig 4B, MHC). These data demonstrated that the RRM2 domain is dispensable for myoblast differentiation, whereas the RRM1 RNA-binding activity and C-terminus sequence of Msi2 are required for Msi2 to regulate myoblast differentiation.

Msi2 regulates myoblast differentiation through autophagy

So far, our data indicated that Msi2 is not only essential but also sufficient for myoblast differentiation, without affecting the expression of MRFs. Because previous studies have shown that autophagy and mitophagy processes are required for myoblast differentiation (McMillan & Quadrilatero, 2014; Sin et al, 2016), we analyzed autophagosome formation by examining LC3A and LC3B expressions in Msi2 knockdown cells. LC3-II, the lipid-modified form of LC3, is an indicator of an intracellular autophagosome formation level, which represents the extent of overall autophagy activity (Zois et al, 2011; Baeken et al, 2020; Klionsky et al, 2021). After differentiation induction, both non-lipidated LC3A-I and LC3A-II were elevated in control shLuc cells, whereas in Msi2 KD cells, LC3 protein levels were relatively unchanged (Fig 5A), suggesting attenuated initiation of autophagy by the loss of Msi2. The mRNA levels of LC3A were comparable between shLuc- and shMsi2-treated cells (Fig 5B), implying post-transcriptional regulation of LC3A protein levels by Msi2. LC3B mRNA levels were changed by Msi2 knockdown in the late (Day 5) but not the early (Days 1 and 2) differentiation stage. Immunofluorescence staining further confirmed that Msi2 knockdown cells showed weaker overall LC3A and LC3B staining intensities and significantly less cytoplasmic LC3 puncta per cell (Fig 5C and D), indicating attenuated autophagosome formation in the absence of Msi2. In addition, Msi2 OE increased both LC3A and LC3B autophagic fluxes in shLuc- and shMsi2-treated cells (Fig S4A and B), demonstrating an autophagy-promoting function of Msi2. The expression level of a lysosomal membrane protein LAMP2 was comparable between control and Msi2 knockdown cells during differentiation, suggesting no overt defects in lysosome biogenesis in the absence of Msi2 (Fig S4C).

To examine whether the autophagy defect in Msi2 KD is the direct cause of the myoblast differentiation arrest, we activated autophagy and analyzed its impact on C2C12 differentiation. To induce autophagy, we used a small cell-permeable peptide Tat-Beclin 1-D11 (D11), which is known to activate autophagy by binding to the autophagy repressor GAPR1 to release Beclin 1 (Shoji-Kawata et al, 2013). A scrambled peptide Tat-Beclin 1-L11S (L11S) was used as a negative control. With the D11 peptide treatment, both control and Msi2 KD cells generated more and larger myotubes on Day 5 post-differentiation induction (Fig 6A and B). Immunoblot analysis confirmed that LC3A-II and LC3B-II levels in control and Msi2 KD cells were increased by the D11 treatment, confirming activated autophagy (Fig 6C). Autophagy activation by the D11 peptide treatment significantly increased the Differentiation Index of Msi2 KD compared with the L11S scrambled control peptide from 13% to 29%, which is comparable to 34% in control cells (Fig 6D), suggesting the differentiation defect of Msi2 KD cells was fully rescued by autophagy induction. Tat-Beclin 1-D11 treatment also significantly increased the Fusion Index of Msi2 KD cells twofold, which is a partial rescue compared with 4.1 ± 0.43 of the control (Fig 6E). Surprisingly, Tat-Beclin 1-D11 alone significantly increased both Differentiation and Fusion Indices in the control cells (Fig 6D and E), implying a differentiation-promoting function of autophagy. Taken together, the Tat-Beclin 1-D11 peptide successfully rescued both autophagy and differentiation in Msi2 KD cells, demonstrating that Msi2 regulates myoblast differentiation via autophagy through the up-regulation of LC3 protein expression.

Reduced endurance capacity in Msi2 gene-trap mutant mice

To investigate the role of Msi2 in skeletal muscle in vivo, we used an Msi2 mutant mouse model generated via a gene-trap mutagenesis strategy (Ito et al, 2010). From heterozygous intercrosses, we obtained 408 WT, 700 heterozygous mutant, and 87 homozygous mutant (KO) mice in total at weaning, indicating a skew in genotype distribution because of decreased survival before weaning (chi-square test, P < 0.0001). Of mice that survived until weaning, KO mice were underweight, and exhibited kyphosis and reduced mobility, implying a defect in muscular function. Consistently, soleus muscles from KO mice were significantly smaller and presented a pale color compared with WT littermate controls (Fig 7A and B). Interestingly, we did not see a size difference in extensor digitorum longus (EDL) muscles, implying predominately slow-twitch muscles, but not fast-twitch, are affected by Msi2 deficiency. WB analysis confirmed that Msi2 expression was lost in the skeletal muscle tissues in KO mice, whereas Msi1 expression was comparable (Fig S5A). Next, we performed several locomotor activity assays. In an open-field activity paradigm, the total travel distance for KO mice was 7.64 m ± 0.52 m, whereas it was 20.63 m ± 0.81 m for the littermate controls for the duration of 30 min (Fig S5B). Vertical activity counts, which are relevant to hindlimb strength, are significantly lower in KO mice compared with WT in both males and females (Fig S5C). In contrast, mice exhibit no overt changes in stereotypic behaviors regardless of their genotypes and sexes, suggesting that lower locomotor activity is due to physical, not neurological, defects in the mutants (Fig S5D). To further analyze the locomotor defect, we tested 16-wk-old mice with a downhill exhaustion treadmill run (Fig 7C). The littermate control mice could run for more than 100 m, whereas Msi2 mutant mice met the criteria for exhaustion by 33 m ± 5.21 m regardless of their sexes, indicating that the mutants had reduced capacity for movement, consistent with defects in skeletal muscle and myocyte differentiation from Msi2 deficiency shown above.

To further investigate the impact of Msi2 loss on muscle functions, we performed histological analysis of the soleus, as well as tibialis anterior (TA) and EDL muscles (Figs 7D and S5E). Total cross-sectional area, muscle fiber number, and average fiber area were all smaller in the Msi2 KO muscles compared with WT, indicating muscular atrophy phenotypes (Fig S5F and G). In addition to the smaller muscle fiber sizes in the KO mice, we also examined whether Msi2 KO mice have a change in fiber-type distribution. We performed immunofluorescence staining of type I fibers (slow, oxidative), type IIA fibers (fast, oxidative and glycolytic mixed), and type IIB fibers (fast, glycolytic) using specific antibodies to each MHC subtype (Fig 7E). Type IIX fibers (fast, glycolytic) appear unstained. Surprisingly, we observed a significant increase in type I fibers at the expense of type IIA fibers in the soleus (Fig S5H). In soleus muscles, type I, type IIA, and type IIX fibers were 43%, 34%, and 23% in KO, whereas those in the WT soleus were 34%, 55%, and 10%, respectively (Fig 7F, top panel). A similar change was observed in fast muscles TA and EDL as well; in WT, we observed no detectable type I fibers, but nearly 5% of the fibers were type I in TA and EDL muscles from KO mice. The proportion of type IIB fibers also significantly decreased from 83% in WT to 74% in KO (Fig 7F, bottom panel). These fiber typing results indicate a shift in muscular cell fates from fast-twitch to slow-twitch fibers in the Msi2 KO skeletal muscles. Collectively, our results clearly demonstrated that Msi2 is essential for mature myocyte formation and thus indispensable for robust muscular activities in vivo.

Cell culture

The mouse C2C12 myoblast cell line was obtained from the ATCC (CRL-1772) and maintained in Growth Media (GM; DMEM with high glucose, 15% FBS [VWR], 1% non-essential amino acids, 100 IU/ml penicillin, and 100 μg/ml streptomycin) in a humidified incubator with 5% CO₂ at 37°C. For cell proliferation assays, the cells were plated in 96-well plates at a density of 1 × 10⁴ cells/ml in GM. Cells were trypsinized and counted at the indicated time points. Media were changed every 3 d. To induce differentiation of C2C12 myoblasts into myotubes, the cells were plated in a six-well plate at 4 × 10⁴ cells/ml in GM. After 24 h, the media were changed to Differentiation Media (DM; DMEM with 2% horse serum [HyClone], 1% non-essential amino acids, and 100 IU/ml penicillin and 100 μg/ml streptomycin). DM was replenished every 3 d during the differentiation period. For autophagy induction during C2C12 differentiation, C2C12 cells were plated in a six-well plate and differentiated as described above. At 24 h after the induction, cells were treated with 20 μM Tat-Beclin 1-L11S or Tat-Beclin 1-D11 for 5 h, followed by fresh DM replenishment as described above. All medium components were purchased from Corning or Nacalai Tesque unless otherwise described. For autophagy flux analysis, cells were treated with 100 nM bafilomycin or DMSO as a negative control for 3 h before harvesting for protein extracts.

Viral constructs and production

Retroviral MSCV-IRES-EGFP and lentiviral FG12-UbiC-GFP vectors were obtained from Addgene (plasmid numbers #20762 and #14884, respectively). Mouse Msi2 cDNA was cloned into the MSCV vector, and lentiviral shRNA constructs were cloned into FG12 essentially as described previously (Ito et al, 2010). The target sequences are 5′-AGTTAGATTCCAAGACGA-3′ for mouse Msi2 (shMsi2) and 5′-CTGTGCCAGAGTCCTTCGATAG-3′ for luciferase as a negative control (shLuc). Viruses were produced in 293FT cells transfected using polyethylenimine with viral constructs along with VSV-G, Gag-Pol, and Rev (in the cases of FG12) constructs as described before (Ito et al, 2010).

For viral transduction, cells were plated at a density of 2 × 10⁴ cells/ml in a 60-mm dish and cultured for 24 h before infection. At 48 h post-infection, cells were monitored for GFP fluorescence using a microscope or flow cytometry. Then, cells were collected and replated for further analysis.

Antibodies

The following antibodies were used: rabbit monoclonal anti-MSI2 (EP1305Y; Abcam); rat monoclonal anti-Msi1 (14989682; Thermo Fisher Scientific); mouse monoclonal anti-Hsp90 (F-8; Santa Cruz Biotech); rabbit monoclonal anti-LC3A (D50G8; Cell Signaling Technology); rabbit monoclonal anti-LC3B (43566; CST); rabbit polyclonal anti-PARP (9542; Cell Signaling Technology); rabbit monoclonal anti-caspase-3 (14220; Cell Signaling Technology); rat monoclonal anti-Lamp2 (ABL-93; Developmental Studies Hybridoma Bank); mouse monoclonal anti-MHC (MF20; Developmental Studies Hybridoma Bank); mouse monoclonal anti-MHC1 IgG2 (BA-F8; Developmental Studies Hybridoma Bank); mouse monoclonal anti-MHC2A IgG1 (SC-71; Developmental Studies Hybridoma Bank); mouse monoclonal anti-MHC2B IgM (BF-F3; Developmental Studies Hybridoma Bank); mouse monoclonal anti-Myog (F5D; Developmental Studies Hybridoma Bank); mouse monoclonal anti-FLAG (F1804; Sigma-Aldrich); goat anti-mouse IgG, HRP-conjugated (31342; Thermo Fisher Scientific); goat anti-rabbit IgG, HRP-conjugated (A16140; Thermo Fisher Scientific); goat anti-rabbit IgG, Alexa Fluor 594–conjugated (A11037; Thermo Fisher Scientific); goat anti-mouse IgG, Alexa Fluor 594–conjugated (A11032; Thermo Fisher Scientific); goat anti-mouse IgG1, Alexa Fluor 568–conjugated (A21124; Thermo Fisher Scientific); goat anti-mouse IgG2B, Alexa Fluor 647–conjugated (A21242; Thermo Fisher Scientific); goat anti-mouse IgM, Alexa Fluor 488–conjugated (A21042; Thermo Fisher Scientific); normal mouse IgG (I8765; Sigma-Aldrich); normal rabbit IgG (P120-101; Bethyl Laboratories).

Western blotting

Cell lysates were prepared with NP-40 lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, and 1% NP-40) with protease and phosphatase inhibitors (Nacalai Tesque). The BCA assay was used to determine protein concentrations (FUJIFILM Wako Pure Chemicals). After separation on SDS–PAGE, proteins were transferred to a polyvinylidene difluoride membrane using a semi-dry transfer system (iBlot2; Thermo Fisher Scientific) for immunoblotting. A chemiluminescence signal was detected using EzWestLumi Plus (2332638; ATTO) with ChemiDoc MP Imaging System (Bio-Rad).

Immunofluorescence staining

For immunofluorescence staining, C2C12 cells were plated in a six-well plate with a glass coverslip at the bottom of each well before differentiation induction as described above. After 5 or 6 d in DM, cells were fixed in 4% PFA in PBS for 15 min at RT, washed three times with PBS, and permeabilized in PBS with 0.5% Triton X-100 for 15 min at RT. Samples were then blocked using Blocking One (Nacalai Tesque) for 30 min, and stained at RT for 1 h with primary antibody followed by incubation with an Alexa Fluor–conjugated secondary antibody and DAPI (Invitrogen). Samples were washed three times with PBS between primary and secondary antibody incubations. Coverslips were mounted onto slides using anti-fade mounting media containing 0.2% n-propyl gallate (FUJIFILM Wako Pure Chemicals). Fluorescence images were captured on AxioImager Z1 with Apotome.2 (Zeiss). For mitochondrion staining, cells were incubated with 200 nM MitoTracker Deep Red (Thermo Fisher Scientific) in a humidified incubator with 5% CO₂ at 37°C for 25 min before the fixation step as described above. For Differentiation and Fusion Indices, images were analyzed using ImageJ software, version 1.53q. Nucleus and cell counting were performed only in GFP-positive, that is, virally transduced, cells. Quantifications of LC3 puncta were performed using ImageJ with a published macro (Dagda et al, 2008; Chu et al, 2009). For the muscle fiber typing experiment, muscles were isolated, embedded in OCT compound followed by snap freezing in isopentane under liquid nitrogen cooling, and cryosectioned. The staining procedure was the same as for C2C12 cells except for the fixation step. The primary antibodies targeting MHC I, MHC IIa, and MHC IIb are different immunoglobulin subtypes such as IgG2, IgG1, and IgM, respectively. Different Alexa Fluor–conjugated secondary antibodies targeting specific immunoglobulin subtypes were used to costain MHC I, IIa, and IIb on the same section.

Electron microscopy

C2C12 cells were plated in a 100-mm dish at a density of 4 × 10⁴ cells/ml in GM. After 24 h, the media were changed to DM to induce differentiation. After 5 d of differentiation, the cells were fixed with 2.5% glutaraldehyde (TAAB Laboratories Equipment Ltd) and postfixed with 1% osmium tetroxide (TAAB Laboratories Equipment Ltd) at 4°C. After en bloc staining with 1% uranyl acetate, the cells were dehydrated with a series of ethanol gradients, followed by propylene oxide, and embedded in Epon 812 resin (TAAB Laboratories Equipment Ltd). The ultrathin sections were stained with 2% uranyl acetate and lead citrate, and observed using Hitachi HT-7700 at 80 kV.

ChIP-seq data analysis

ChIP-seq data for Myog and MyoD binding sites in differentiated C2C12 cells were retrieved from ReMap Atlas (dataset GSE44824) (Hammal et al, 2022). The H3K4me3 ChIP-seq data of differentiated C2C12 cells were obtained from ENCODE (datasets ENCSR000AHT) (He et al, 2020). Figures were generated using Integrated Genomics Viewer, version 2.14.0 (Robinson et al, 2011).

Real-time and standard RT–PCR analysis

Total cellular RNAs were purified using a TRI Reagent (Sigma-Aldrich), and cDNAs were prepared from equal amounts of RNAs using Superscript IV Reverse Transcriptase (Thermo Fisher Scientific). Quantitative real-time PCRs were performed using EvaGreen qPCR Master Mix (Bio-Rad) on StepOnePlus (Thermo Fisher Scientific). Results were normalized to the level of β-2 microglobulin. PCR primer sequences are as follows. All primer pairs gave a specific PCR product with a single peak in melting curve analysis.

mB2m-F, 5′-ACCGGCCTGTATGCTATCCAGAA-3′

mB2m-R, 5′-AATGTGAGGCGGGTGGAACTGT-3′

mMyog-F, 5′-CTAAAGTGGAGATCCTGCGCAGC-3′

mMyog-R, 5′-GCAACAGACATATCCTCCACCGTG-3′

mMyh1-F, 5′-ATGAACAGAAGCGCAACGTG-3′

mMyh1-R, 5′-AGGCCTTGACCTTTGATTGC-3′

mMyh3-F, 5′-TGAACAGATTGCCGAGAACG-3′

mMyh3-R, 5′-GGAGAATCTTGGCTTCTTCGTG-3′

mMyod1-F, 5′-TTCTTCACCACACCTCTGACA-3′

mMyod1-R, 5′-GCCGTGAGAGTCGTCTTAACTT-3′

mMyh4-F, 5′-CACCTGGAGCGGATGAAGAAGAAC-3′

mMyh4-R, 5′-GTCCTGCAGCCTCAGCACGTT-3′

mMap1lc3a-F, 5′-GACCGCTGTAAGGAGGTGC-3′

mMap1lc3a-R, 5′-CTTGACCAACTCGCTCATGTTA-3′

mMap1lc3b-F, 5′-TTATAGAGCGATACAAGGGGGAG-3′

mMap1lc3b-R, 5′-CGCCGTCTGATTATCTTGATGAG-3′

mMyomixer-F, 5′-GTTAGAACTGGTGAGCAGGAG-3′

mMyomixer-R, 5′-CCATCGGGAGCAATGGAA-3′

Mice

The Msi2 mutant mice, B6;CB-Msi2^{Gt(pU-21T)2Imeg}, established via gene-trap mutagenesis, were bred, and genotyped as previously described (Ito et al, 2010), and maintained on a 12-h:12-h light:dark cycle in the facilities of the University Research Animal Resources at University of Georgia, and of the Institute for Life and Medical Sciences at Kyoto University. All mice were 5–20 wk old, age-matched, and randomly chosen for experimental use. No statistical methods were used for sample size estimates. All animal experiments were performed according to protocols approved by the University of Georgia and the Kyoto University Institutional Animal Care and Use Committees.

Open-field activity test

Open-field activity monitors (16 × 16 inch SuperFlex Open Field; Omnitech Electronics) with infrared beams to track horizontal and vertical mouse movements were used for the experiments. The same handler performed all experiments. Up to four mice of the same sex were placed in separate, adjacent testing chambers, initiating a 30-min test session: horizontal travel distance, horizontal activity count (horizontal beam breaks), ambulatory activity count, rest time (inactivity greater or equal to 1 s), rest episode count, movement time, movement episode count (separated by rest periods of at least 1 s), stereotypy time, stereotypy episode count (stereotypy behavior greater or equal to 1 s was counted as 1 episode), stereotypy activity count (beam breaks), vertical episode count, vertical activity count (vertical beam breaks), and vertical activity time. Stereotypy behavior was defined as repeatedly breaking the same beam, which typically happens during grooming or head bobbing. Total horizontal distance, total time in stereotypical behavior, and total vertical activity time over the 30-min session were exported and analyzed in Microsoft Excel.

Mouse treadmill exhaustion assay

A four-lane variable speed mouse treadmill (AccuScan) placed at a 15° angle was used for acclimation and exhaustion testing. The downhill running paradigm was selected over flat or uphill running as it is associated with greater exercise-induced muscle damage, in part, by forcing muscle to lengthen in an eccentric muscle contraction (reviewed by Proske & Morgan [2001]). For acclimation, up to four mice of the same sex were placed in adjacent lanes on the treadmill for 30 min at 0 m/min for 25 min and 3 m/min for 5 min, with the shock pad activated to 0.3 mA on two consecutive days before testing. On the test day with the shock pad activated, mice were placed on the treadmill for a 3-m/min warm-up for 5 min. Then, the test run was started at a speed of 10 m/min for 5 min, and the speed was increased by 5 m/min every 5 min until reaching a maximum speed of 25 m/min. Mice ran until exhaustion or 15 min at the maximum speed. Exhaustion was defined as a mouse dwell time of over 10 s on the shock pad. Total running time was recorded as soon as the mouse was exhausted. Total running distance was calculated based on the total running time and speed. The mice used for this assay were all 4 mo old.

Statistical analysis

Statistical analyses were carried out using GraphPad Prism 9 software (GraphPad Software Inc.). Data are shown as the mean ± the standard error of the mean. Two-tailed unpaired t tests or one-way ANOVA with Tukey’s post hoc tests was used unless otherwise indicated to determine statistical significance. ns, not significant (P ≥ 0.05), *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.