The skeletal muscle slow myosin heavy chain regulates mammalian metabolic homeostasis through the NRF2 pathway Authors: Jaydeep Sharma, Somnath Mondal, Aparna Rai, Mahima Kumari, Anushree Bharadwaj, Sam J. Mathew Published: 2026 ABSTRACT The mammalian skeletal muscle is central to metabolic homeostasis. Myosin heavy chains (MyHCs), key muscle contractile proteins, use energy from adenosine triphosphate hydrolysis to produce mechanical force, fundamental to muscle function. However, the link between MyHCs and metabolic regulation is unclear. Here, we demonstrate the role of Myh7, encoding the MyHC-slow protein, in regulating skeletal muscle function and metabolic homeostasis using skeletal muscle–specific knockout mice. The absence of MyHC-slow causes early postnatal skeletal muscle hypertrophy followed by atrophy, degeneration of the oxidative slow myofibers, alterations in fiber-type proportions, decreased force production, and muscle dysfunction. It also leads to impaired glucose utilization, insulin resistance, and reduced muscle GLUT4 levels, characteristic of type 2 diabetes. These are mediated through decreased levels of the antioxidant NRF2, elevated reactive oxygen species and mitochondrial dysfunction in the Myh7 knockouts, which can be rescued by activating NRF2 signaling via sulforaphane administration. Our findings link skeletal muscle contractility to metabolic homeostasis, identifying the NRF2 pathway as a key therapeutic target. INTRODUCTION In addition to its roles in locomotion, posture maintenance, and support, the mammalian skeletal muscle is crucial for metabolic homeostasis. It is the largest tissue in our body, contributing to about 40% of the total body mass in healthy individuals. Owing to its size and contractile function, the skeletal muscle is one of the key contributors to regulation of whole-body metabolism. The skeletal muscle is one of the primary sites of glucose and lipid uptake and utilization in the body. Insulin resistance in the skeletal muscle, characterized by impaired insulin-stimulated glucose uptake and utilization, is a leading cause for type 2 diabetes and associated metabolic disorders. How skeletal muscle contractility and contractile proteins contribute to metabolic homeostasis is of substantial interest in understanding metabolic regulation and dysfunction. The mature, terminally differentiated cell type of the skeletal muscle are the myofibers, which arise by differentiation of the muscle progenitors or stem cells. The muscle progenitors express proteins such as Pax3 and Pax7, followed by Myf5, MyoD, and MRF4 at early (myoblasts) and myogenin (MyoG) at later (myocytes) stages of differentiation (1). The muscle progenitors, myoblasts and myocytes are mononuclear cells, which undergo fusion to give rise to the multinucleated myofibers. Myofibers contain arrays of sarcomeres, the structural and functional unit of muscle contraction. The sarcomeres are made up of thin and thick filaments, composed primarily of actin and myosin proteins, respectively. Myosin is a heterohexamer, comprising two pairs of light chains and one pair of heavy chains. Several myosin heavy chain (MyHC) isoforms encoded by distinct genes are expressed in the mammalian skeletal muscle: Three isoforms are expressed during developmental stages, namely, MyHC-embryonic, MyHC-slow, and MyHC-perinatal, whereas five isoforms are expressed during adult life, namely, MyHC-IIa, MyHC-IIx, MyHC-IIb, MyHC-slow, and MyHC-extraocular (2–4). Most myofibers express a single MyHC, although hybrid myofibers expressing multiple MyHC isoforms can occur. Adult myofibers are broadly of the “slow” or “fast” type, based on their contractile and metabolic properties. MyHC isoforms have long been used as molecular markers to identify the myofiber type, with MyHC-slow used to label slow fibers whereas MyHC-IIa, MyHC-IIx, and MyHC-IIb label fast fiber subtypes. Slow and fast fibers differ in their speed and force of muscle contraction, fatiguability, use of primarily oxidative or glycolytic metabolism, and mitochondrial abundance. The slow myofibers principally use lipids as their energy source through mitochondrial oxidative phosphorylation, whereas the fast myofibers primarily use glucose as an energy source through anaerobic glycolysis. Skeletal muscle fiber type has been closely linked to metabolic regulation. For instance, the slow fiber proportions are reduced considerably in individuals with obesity or type 2 diabetes (5, 6). Slow fiber–rich, oxidative muscles such as the soleus have been shown to be excellent targets for specialized exercise to regulate blood glucose and triglyceride levels (7). Recent evidence indicates that MyHCs can transition between two conformations in the resting state, namely, the disordered relaxed state, where the head domain can bind adenosine triphosphate (ATP), and a super-relaxed state, where the adenosine triphosphatase (ATPase) site cannot bind ATP due to steric hindrance; MyHCs in the super-relaxed state have an ATP turnover rate that is 10 times slower than the disordered relaxed state (8). Type 2 diabetes led to increased MyHC glycation, resulting in increased proportion of MyHCs in the super-relaxed conformation, leading to reduced ATP demand (9, 10). Thus, MyHC isoforms, their modifications, conformation, and function could be crucial regulators of metabolic homeostasis. Among the mechanisms that contribute to type 2 diabetes and metabolic dysfunction, elevated levels of reactive oxygen species (ROS) and mitochondrial abnormalities have been repeatedly observed. Thus, pathways that regulate ROS levels and mitochondrial function such as antioxidant signaling and biogenesis factors, respectively, emerge as prime candidates associated with such disorders. In metabolic diseases such as obesity and type 2 diabetes, changes in the skeletal muscle including insulin resistance, elevated inflammation, myofiber atrophy, and mitochondrial dysfunction are reported to occur. Although MyHCs are ATP-hydrolyzing contractile proteins that facilitate muscle function and metabolic regulation, very few studies directly explore the molecular mechanisms and functional consequences underlying the role of MyHCs in metabolic dysfunction. In this study, we address this gap by generating a skeletal muscle–specific mouse knockout for the Myh7 gene, encoding the slow fiber–specific MyHC-slow. We find that loss of MyHC-slow function results in an early, transient skeletal muscle hypertrophy, followed by atrophy. Depending on the type of muscle, MyHC-slow function is required for maintenance of the slow myofibers, with a complete loss or severe atrophy of the slow myofibers seen in the Myh7 knockouts. Alterations in the proportion of other fiber types, myofiber size, force production, and muscle function were observed upon loss of MyHC-slow. MyHC-slow is important in regulating the myofiber mitochondria, with the Myh7 knockouts exhibiting reduced myofiber mitochondrial abundance, activity, biogenesis, and altered dynamics. At the metabolic level, the skeletal muscle–specific Myh7 knockouts exhibit impaired glucose homeostasis and insulin resistance from 6 months of age onward, which was accelerated upon administration of a high-fat diet (HFD) and observed from 4 months of age. Proteomic analysis of the Myh7 knockout skeletal muscle revealed a highly significant down-regulation of the nuclear factor erythroid 2-related factor 2 (NRF2) signaling pathway and elevated ROS, which could underlie the observed metabolic dysfunction. Administration of sulforaphane, an activator of NRF2, led to improvement of almost all the musculoskeletal, mitochondrial, and metabolic defects in the Myh7 knockouts. These results validate that NRF2 signaling downstream of MyHC-slow is crucial for metabolic homeostasis and could be a vital therapeutic target in diseases associated with metabolic dysfunction such as type 2 diabetes. RESULTS Skeletal muscle–specific deletion of Myh7 leads to postnatal muscle fiber hypertrophy MyHC-slow encoded by the Myh7 gene is the only MyHC thought to be expressed during mammalian skeletal muscle developmental and adult phases. It could thus have important and distinct functions in the skeletal muscle through various stages, which has not been explored. Therefore, we generated a conditional mouse allele where exon 26 of Myh7 is flanked by LoxP sites, referred to as Myh7fl (Fig. 1A and fig. S1A). A knockout Myh7Δ mouse allele was generated by crossing the Myh7fl to the germline CMVCre driver (fig. S1A). To delete Myh7 in the skeletal muscle at early and late embryonic stages, we used the Pax3Cre/+ and Pax7iCre/+ mouse driver lines, respectively (2, 11) (fig. S1, B and Ba). Compared to control Pax+/+, Myh7fl/+ pups, Pax3Cre/+, Myh7fl/Δ pups exhibited a complete loss of MyHC-slow expression (Fig. 1, B to Ca), whereas Pax7iCre/+, Myh7fl/Δ pups exhibited a drastic reduction in MyHC-slow expression (Fig. 1, D and Da), in cross sections through the hindlimb shank region at postnatal day 0 (P0), which, upon quantification, was found to be a reduction of ∼75% (fig. S1C). By P30, the Pax7iCre/+, Myh7fl/Δ mice exhibited a complete loss in MyHC-slow expression as seen in the gastrocnemius and soleus muscles (Fig. 1, E to Fa). These results validate the Myh7 conditional and knockout alleles and the early and late embryonic activity of Pax3Cre/+ and Pax7iCre/+ driver lines, respectively. Skeletal muscle–specific deletion of Myh7 leads to postnatal muscle fiber hypertrophy. (A) Schematic depicting the Myh7floxed (Myh7fl) allele, where LoxP sites (red arrowheads) flank exon 26 of the Myh7 gene. (B to Fa) Cross sections through the hindlimbs of P0 and P30 Pax+/+, Myh7fl/+ (B, Ba, E, and Ea), Pax3cre/+, Myh7Δ/fl (C and Ca), and Pax7icre/+, Myh7Δ/fl (D, Da, F, and Fa) mice labeled by immunofluorescence for MyHC-slow (green), laminin (red), and 4′,6-diamidino-2-phenylindole (DAPI) (blue). (Bb to Db, Ea, and Fa) Magnifications of boxed areas of the TA and soleus (P30) muscles from (Ba) to (Da), (E), and (F), respectively. Scale bars, 100 μm (Da, E, and F) and 10 μm (Db, Ea, and Fa). In (Ba) to (Da): E, EDL; S, soleus; TA, tibialis anterior; Tib, tibia; Fib, fibula. (G to I) Quantification of the average myofiber area (μm2), the number of myofibers/mm2, and the proportion of myofibers grouped according to myofiber area through the soleus of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl, and Pax7icre/+, Myh7Δ/fl mice at P0 (n = 5). (J to N) Body weight and the TA, Gas, and Quad muscle weight of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl, and Pax7icre/+, Myh7Δ/fl mice at P0 and P30, respectively (n = 12 and n = 5 to 6). (O to Pa) Cross sections through the Gas and soleus of Pax+/+, Myh7fl/+ (O and P) and Pax7icre/+, Myh7Δ/fl (Oa and Pa) mice at P30 labeled by immunofluorescence for laminin (red) and DAPI (blue) (n = 4). Scale bar, 100 μm (Pa). (Q to T) Average myofiber area (μm2) and the number of myofibers/mm2 through the Gas and soleus of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at P30 (n = 4 to 5). (U and V) Quantification of the proportion of myofibers grouped according to myofiber area through the Gas and soleus of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at P30 (n = 4). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001; n.s., not significant. An increase in average myofiber area was observed in the soleus, extensor digitorum longus (EDL), and tibialis anterior (TA) muscles of both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups at P0 (Fig. 1, Bb, Cb, Db, and G, and fig. S1, D and G). A decrease in the number of myofibers normalized to total cross-sectional area was seen in the soleus, EDL, and TA of both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups (Fig. 1H and fig. S1, E and H). This was further characterized by assembling myofibers into distinct groups based on their cross-sectional area and quantifying the percentage of myofibers falling in each group. A significant decrease in the proportion of smaller-sized myofibers (10 to 50 μm2) and an increase in the proportion of larger-sized myofibers (100 to 150 μm2, 150 to 200 μm2, 200 to 250 μm2, 250 to 300 μm2, and bigger) were observed, in most cases, in the soleus, EDL, and TA in both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups (Fig. 1I and fig. S1, F and I). There was no significant change in total body weight between the control and knockout mouse genotypes at both P0 and P30 (Fig. 1, J and K). Individual muscle weights for the TA, gastrocnemius, and quadriceps muscles did not exhibit significant changes between the control and conditional knockouts (Fig. 1, L to N). Cross sections through the gastrocnemius, soleus, and TA muscles at P30 suggested a visible increase in fiber size in the Pax7iCre/+, Myh7fl/Δ mice (Fig. 1, O to Pa, and fig. S1, J and Ja). Quantification indicated that the average myofiber area increased and the number of myofibers normalized to muscle cross-sectional area decreased in these muscles (Fig. 1, Q to T, and fig. S1K). Like P0, a decrease in the proportion of smaller myofibers and an increase in the proportion of larger myofibers were seen in these three muscles at P30 (Fig. 1, U and V, and fig. S1L). Because the fast MyHC-IIb+ and MyHC-IIx+ myofibers are relatively larger in size (12), we labeled gastrocnemius muscle cross sections from control and Pax7iCre/+, Myh7fl/Δ mice at P30 (fig. S1, M to Na). The average MyHC-IIb+ and MyHC-IIx+ myofiber area and the percentage of MyHC-IIb+ and MyHC-IIx+ myofibers normalized to total myofiber numbers were both significantly increased in the gastrocnemius of P30 Pax7iCre/+, Myh7fl/Δ mice (fig. S1, O to R). Because we observed an increase in average myofiber area of MyHC-IIb+ and MyHC-IIx+ myofibers, a decrease in smaller myofiber proportions, and an increase in larger myofiber proportions, we checked myofiber atrophy and hypertrophy protein levels. No differences were observed in the protein levels of the atrophy markers Atrogin-1, MuRF-1, and Myostatin or the hypertrophy marker IGF-1 in the gastrocnemius muscle of both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups at P0 (fig. S1, S to W). The levels of MyoD, a crucial myogenic regulatory factor that drives myogenic differentiation, was elevated in the gastrocnemius of Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups, indicating that enhanced differentiation most likely underlies the increase in myofiber size upon loss of MyHC-slow function at early postnatal stages (fig. S1, S and X). Overall, we observe an increase in the average myofiber area, a decline in the number of myofibers normalized to area, a reduction in the proportion of smaller myofibers, and an increase in the proportion of larger myofibers at P0, upon developmental loss of Myh7 using both early and late recombining Cre driver lines. Similar results were also seen at P30 with the Pax7iCre driver, where, in addition, the proportion and size of MyHC-IIb+ and MyHC-IIx+ myofibers were increased. These myofiber size alterations were independent of myofiber atrophy and hypertrophy and are most likely due to enhanced myogenic differentiation. These results clearly indicate that skeletal muscle–specific embryonic loss of Myh7 leads to myofiber hypertrophy at P0 and P30. Skeletal muscle–specific deletion of Myh7 leads to alterations in adult body weight, muscle weight, myofiber area, and number Because we observed postnatal skeletal muscle hypertrophy, next we wished to characterize the adult phenotypes upon loss of MyHC-slow function. Pax7iCre/+, Myh7fl/Δ female mice exhibited significantly elevated body weight at 2 months of age, but the body weight was decreased at all time points thereafter, from 4 to 12 months of age (fig. S2A). An identical trend was seen in Pax7iCre/+, Myh7fl/Δ male mice (fig. S3A). These changes were not related to the quantity of food consumed because food consumption by Pax+/+, Myh7fl/+ and Pax7iCre/+, Myh7fl/Δ mice, measured as hourly and average food intake, using an animal monitoring system, did not differ significantly at 2 to 12 months of age (fig. S3, D to I). The presence of either Cre driver did not significantly alter the body weight or the grip strength normalized to body weight compared to mice lacking the Cre driver, at any time point (fig. S2, T and U). The muscle weight exhibited no significant change for the TA and quadriceps, whereas the gastrocnemius exhibited an increase at 2 months of age in Pax7iCre/+, Myh7fl/Δ mice (fig. S2, B to D). However, the TA, gastrocnemius, and quadriceps all exhibited a significant reduction in weight thereafter, at 6 and 12 months of age (fig. S2, B to D). The soleus, the muscle with the highest proportion of MyHC-slow+ myofibers, exhibited reduced muscle weight consistently, from 2 to 12 months, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S2E). The number of myofibers normalized to cross-sectional area was increased in the gastrocnemius and decreased in the soleus of Pax7iCre/+, Myh7fl/Δ mice at 2 and 12 months (fig. S2, F and G). Cross sections through the gastrocnemius indicated smaller fiber size at 2 and 12 months in the Pax7iCre/+, Myh7fl/Δ mice and significantly reduced average myofiber area upon quantification (fig. S2, H to K). However, the soleus cross sections suggested a radical increase in myofiber size and significantly increased average myofiber area at 2 and 12 months in the Pax7iCre/+, Myh7fl/Δ mice (fig. S2, L to O). Contrary to P30, an increase in the proportion of smaller myofibers and a decrease in the proportion of larger myofibers were seen in the gastrocnemius of the Pax7iCre/+, Myh7fl/Δ mice at 2 and 12 months (fig. S2, P and R). The soleus presented an exceptional phenotype, with an increase in the proportion of both small and large myofibers in the Pax7iCre/+, Myh7fl/Δ mice at 2 and 12 months (fig. S2, Q and S). The 2-month TA showed a similar trend to P30, with larger myofibers, reduced proportion of small myofibers, and increased proportion of larger myofibers; however, the reverse was seen at 12 months, with reduced average myofiber area, an increase in the proportion of smaller myofibers, and a decrease in the proportion of larger myofibers, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S3, J to O). These results suggest a complex, muscle-type-specific, and age-dependent effect on myofiber characteristics upon loss of MyHC-slow function. Briefly, the hypertrophy-like effect seen during P0 to P30 in the Myh7 conditional knockouts continues to be observed at 2 months of age, with respect to body weight, and in some muscles, with respect to muscle weight, average myofiber area, a decrease in smaller fiber proportions, and an increase in larger fiber proportions. Intriguingly, a reversal of trend from hypertrophy to muscle atrophy is observed thereafter in the Myh7 conditional knockouts, with consistently reduced body weight, muscle weight, average myofiber area, an increase in smaller fiber proportions, and a decrease in larger fiber proportions seen at 6 to 12 months of age. The Myh7 conditional knockouts exhibited a persistent reduction in soleus weight at all time points from 2 months onward; however, the average myofiber area was consistently higher and the proportion of smaller and larger myofibers was significantly increased, indicating that Myh7 has unique regulatory roles in the soleus, distinct from other muscles. One unifying feature across muscles seems to be the accelerated age-associated atrophy from 2 months onward, in the Pax7iCre/+, Myh7fl/Δ mice. Loss of MyHC-slow leads to loss or atrophy of slow myofibers, reduced mitochondrial content, and altered mitochondrial dynamics MyHC-slow has been used extensively as the marker for slow myofibers, with very few other markers known. Because troponin proteins are thought to be fiber-type specific, we checked the expression of the slow skeletal muscle troponin T1 (encoded by the Tnnt1 gene). We found that slow skeletal muscle troponin T1 expression was specific to MyHC-slow+ fibers at both 2 and 12 months in the gastrocnemius and soleus (Fig. 2, A to Bb, and fig. S4, A to Bb). In Pax7iCre/+, Myh7fl/Δ mice at 2 and 12 months, we observed very few to no troponin T1+ myofibers in the gastrocnemius (Fig. 2, C to Db). In the soleus, highly atrophied troponin T1+ myofibers, which are a fraction of the size of the control slow fibers, were seen at both 2 and 12 months (fig. S4, C to Db); these atrophied myofibers were not positive for any other adult MyHCs (fig. S7, G to Nc). Because slow myofibers exhibit oxidative phosphorylation and harbor more mitochondria, we carried out succinate dehydrogenase (SDH) staining, where a significant reduction in SDH activity was seen in the gastrocnemius and soleus of Pax7iCre/+, Myh7fl/Δ mice at 2 and 12 months (Fig. 2, E to H, and fig. S4, E to H). We validated the slow skeletal muscle troponin T1 levels by Western blots, finding significantly reduced levels in the gastrocnemius and TA of Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups at P0 and Pax7iCre/+, Myh7fl/Δ mice at 2 and 6 months of age (Fig. 2, I to K, and fig. S4, I to K). Because we observed reduced mitochondrial enzyme (SDH) activity, we tested for PGC-1α, a master regulator of mitochondrial biogenesis, and detected significantly reduced levels in the gastrocnemius and TA of Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups at P0 and Pax7iCre/+, Myh7fl/Δ mice at 2 and 6 months of age (Fig. 2, I and L, and fig. S4, I and L). Similarly, the mitochondrial content was checked via the protein levels of the five mitochondrial complexes involved in oxidative phosphorylation, all of which were significantly decreased in the gastrocnemius at P0 and 2 and 6 months and in the TA at 2 and 6 months (Fig. 2, M to P, and fig. S4, M to R). FIS1 and DRP1 are crucial regulators of mitochondrial fission, whereas OPA1, Mitofusin 1, and Mitofusin 2 regulate mitochondrial fusion (13). Increased mitochondrial fission marked by elevated FIS1, DRP1, and p-DRP1S616 protein levels and reduced mitochondrial fusion by decreased levels of OPA1, Mitofusin 1, and Mitofusin 2 were observed in the Myh7 knockout gastrocnemius at P0 and 2 and 6 months and in the TA at 2 and 6 months (Fig. 2, Q to T, and fig. S4, S to U). For mitochondrial function, ATP levels were quantified in the gastrocnemius and TA at 2 and 12 months of age, where the levels were found to be less than half in the Myh7 knockouts (Fig. 2, U and V, and fig. S4, V and W). Myh7 knockout muscle exhibits loss of slow myofibers, reduced mitochondrial content, and altered mitochondrial dynamics. (A to Db) Cross sections through the Gas muscle of Pax+/+, Myh7fl/+ (A to Bb) and Pax7icre/+, Myh7Δ/fl (C to Db) mice at 2 and 12 months, respectively, labeled by immunofluorescence for MyHC-slow (green), troponin T1 (red), and DAPI (blue). Scale bars, 100 μm (Cb and Db). (E to H) SDH staining on Gas muscle sections of Pax+/+, Myh7fl/+ (E and G) and Pax7icre/+, Myh7Δ/fl (Ea and Ga) mice and quantification of the percentage SDH activity (F and H) at 2 and 12 months of age, respectively (n = 5). Scale bars, 100 μm (Ea and Ga). (I to L) Representative Western blots for troponin T1, PGC-1α, and α-tubulin (loading control) using Gas muscle protein lysates of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl (P0), and Pax7icre/+, Myh7Δ/fl mice at P0 and 2 and 6 months (I) and their densitometric quantification (J to L) (n = 4). (M to P) Representative Western blots for mitochondrial complexes and α-tubulin (loading control) using protein lysates from the Gas muscle of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl (P0), and Pax7icre/+, Myh7Δ/fl mice at P0 and 6 months (M and O) and their densitometric quantification (N to P) (n = 4 to 6). (Q to T) Representative Western blots for FIS1, pDRP1, DRP1, OPA1, Mitofusin 1, Mitofusin 2, and α-tubulin (loading control) using protein lysates from the Gas muscle of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl (P0), and Pax7icre/+, Myh7Δ/fl mice at P0 and 2 and 6 months (Q) and densitometric quantification (R to T) (n = 4). (U and V) Quantification of ATP levels in the Gas muscle of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 2 and 12 months of age (n = 5). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. These results suggest that loss of Myh7 function leads to loss of slow myofibers in the gastrocnemius and severe atrophy of the slow myofibers in the soleus, although it is intriguing that the atrophying slow myofibers persist even at 12 months of age in the soleus. We also find severe effects on the skeletal muscle mitochondria upon Myh7 loss of function, with reduced mitochondrial activity, biogenesis, content, dynamics (increased fission and reduced fusion), and ATP production in the skeletal muscle. Overall, loss of slow myofibers and altered myofiber mitochondrial properties are seen upon loss of MyHC-slow function. Loss of Myh7 in the skeletal muscle leads to mitochondrial abnormalities To further characterize the mitochondrial phenotypes in the Myh7 knockout skeletal muscle, we carried out Western blots for Tom20 (translocase of the outer membrane 20), a crucial mitochondrial outer membrane receptor, and VDAC1 (voltage-dependent anion channel 1), a key mitochondrial outer membrane channel for metabolite and ion transport. Levels of both Tom20 and VDAC1 were significantly reduced in the gastrocnemius of Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups at P0 (Fig. 3, A and B). Similar results were observed for both proteins in the gastrocnemius and TA muscles of Pax7iCre/+, Myh7fl/Δ mice at 2 and 6 months of age (Fig. 3, A and C to F). These results validate the reduction in mitochondrial content in the Myh7 knockout skeletal muscle. Loss of Myh7 in the skeletal muscle leads to mitochondrial abnormalities. (A to F) Representative Western blots for Tom20, VDAC1, and α-tubulin using protein lysates from the Gas and TA (2 and 6 months) muscle of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl (P0), and Pax7icre/+, Myh7Δ/fl mice at P0 and 2 and 6 months of age (A) and their densitometric quantification (B to F). α-Tubulin was used as a loading control (n = 4). (G to N) Quantification of the OCR measured using myofibers isolated from Gas and EDL muscles (G, I, K, and M), and their basal, uncoupled, and maximal respiration (H, J, L, and N), of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 2 and 12 months of age (n = 3–4). AA, antimycin A. (O to Pa) Representative TEM through the Gas of Pax+/+, Myh7fl/+ (O and Oa) and Pax7icre/+, Myh7Δ/fl (P and Pa) mice at 12 months of age. Scale bars, 500 nm (O to Pa). (Q) Average mitochondrial area (μm2) through the Gas of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 12 months of age (n = 4). (R) Quantification of the proportion of mitochondria grouped according to mitochondrial area through the Gas muscle of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 12 months of age (n = 4). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. To better characterize mitochondrial function in the Myh7 knockout muscle, the oxygen consumption rate (OCR) was measured using the Seahorse analyzer. A clear decrease in OCR was observed throughout, at basal, uncoupled (after oligomycin injection, indicating ATP-independent OCR), and maximal [after carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) injection, indicating maximal mitochondrial respiration] mitochondrial respiration, in Myh7 knockout myofibers from the gastrocnemius and EDL muscles at 2 and 12 months of age (Fig. 3, G to N). Thus, loss of MyHC-slow leads to reduced mitochondrial respiration and oxidative capacity. Transmission electron microscopy (TEM) was carried out through gastrocnemius muscle sections of control and Pax7iCre/+, Myh7fl/Δ 12-month-old mice, where an increase in the number of smaller mitochondria was observed in the Myh7 knockout muscle (Fig. 3, O to Pa). Upon quantification, a significant decrease in the average mitochondrial size and an increase in the proportion of smaller-sized mitochondria at the expense of larger-sized mitochondria were observed in the Myh7 knockout muscle (Fig. 3, Q and R). These results validate the increased mitochondrial fission markers and decreased fusion, which was observed by Western blots for specific markers (Fig. 2, Q to T). Overall, we find that loss of MyHC-slow function in the skeletal muscle leads to mitochondrial abnormalities including reduced mitochondrial content, diminished mitochondrial respiration, and a decrease in mitochondrial size. Myh7 knockout mice exhibit altered glucose homeostasis and insulin resistance Mitochondrial dysfunction is known to lead to altered metabolic homeostasis. For instance, the inhibition of DRP1, a mitochondrial fission factor, improved muscle insulin resistance, whereas mouse knockout of Mitofusin 2, a protein involved in mitochondrial fusion, resulted in glucose intolerance (14, 15). Other studies indicate that individuals with type 2 diabetes exhibit reduced levels of the mitochondrial fusion factors Mitofusin 1, Mitofusin 2, and OPA1 in the skeletal muscle (16). Skeletal muscle–specific knockout of the mitochondrial biogenesis regulator PGC-1α led to altered glucose homeostasis (17). Because we observed reduced mitochondrial biogenesis, activity, content, dynamics, and function in the Myh7 muscle-specific knockout mice, we tested fasting blood glucose levels in these mice. We found no significant difference in the fasting blood glucose levels between Pax+/+, Myh7fl/+ and Pax7iCre/+, Myh7fl/Δ mice at 2 months, but a consistent increase in fasting blood glucose levels was seen in the Pax7iCre/+, Myh7fl/Δ mice at 6 and 12 months (time point 0 in Fig. 4, A, C, and E). Next, we tested glucose utilization by the glucose tolerance test, where the 6- and 12-month-old Pax7iCre/+, Myh7fl/Δ mice showed significantly elevated glucose levels at all time points compared to controls and increased area under the curve; the 2-month-old mice did not exhibit significant changes in glucose utilization (Fig. 4, A to F). We also carried out insulin tolerance tests to check the response to insulin in control and Pax7iCre/+, Myh7fl/Δ mice, where we observed no significant changes in 2-month-old mice but significantly elevated glucose levels at all time points and increased area under the curve in the 6- and 12-month-old Pax7iCre/+, Myh7fl/Δ mice (Fig. 4, G to L). The effects on fasting blood glucose, glucose utilization (by the glucose tolerance test), and response to insulin (by the insulin tolerance test) did not show any significant difference between control and Pax7iCre/+, Myh7fl/Δ mice at 4 months of age (fig. S5, A to D). Similarly, no difference in fasting blood glucose levels were seen between control and Pax7iCre/+, Myh7fl/Δ mice at 5 months of age (fig. S5E). Thus, the impaired glucose homeostasis in Myh7 muscle-specific knockout mice starts by 6 months of age and does not improve with exogenous insulin administration, indicating insulin resistance. Myh7 knockout mice exhibit altered glucose homeostasis and insulin resistance. (A to F) Glucose tolerance test (GTT) on Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl 2-, 6-, and 12-month-old mice and their respective area under the curve (n = 5 to 6). (G to L) Insulin tolerance test (ITT) of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 2, 6, and 12 months and their respective area under the curve (n = 5 to 6). (M to Q) Representative Western blots of P-Akt, Akt, and α-tubulin using protein lysates from the Gas muscle of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl (Myh7KO) mice at 2 and 6 months (M) and their densitometry (N to Q). α-Tubulin was used as a loading control (n = 4). (R to V) Representative Western blots for GLUT4 and α-tubulin using protein lysates from the Gas muscle of Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl (P0), and Pax7icre/+, Myh7Δ/fl mice at P0 and 2 and 6 months (R and S) and their densitometric quantification (T to V). α-Tubulin was used as a loading control (n = 4 to 5). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. To test the responsiveness of the control and Myh7 knockout muscle to insulin, we assayed for Akt phosphorylation in the gastrocnemius, TA, and soleus at 2 and 6 months in the presence and absence of exogenous insulin. Although there was a basal decline in phsopho-AktS473 normalized to total Akt levels in the absence of exogenous insulin in the Myh7 knockout muscle, we observed highly reduced phospho-AktS473 normalized to total Akt levels following insulin administration in the Myh7 knockout muscle (Fig. 4, M to Q, and fig. S5, I to R). This confirms the insulin resistance in the Myh7 knockout muscle, as was seen in the insulin tolerance test results. GLUT4, the glucose transporter expressed in the skeletal muscle and fat cells, is thought to be regulated by Akt signaling (18–20). Furthermore, GLUT4 expression is higher in slow myofibers compared to fast ones (21). Because Akt phosphorylation and signaling is reduced and slow fiber atrophy is seen in the Myh7 knockout muscle, we analyzed GLUT4 expression in the skeletal muscle of these mice. At P0, both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ pups and at 2 and 6 months Pax7iCre/+, Myh7fl/Δ mice exhibited significantly reduced GLUT4 levels in the gastrocnemius (Fig. 4, R to V). Similar results were observed in the TA at 2 and 6 months of age (fig. S5, F to H). Thus, reduced Akt signaling resulting from insulin resistance in the Myh7 knockout muscle leads to a decrease in expression of the glucose transporter GLUT4. Because the Pax7iCre/+, Myh7fl/Δ mice exhibited elevated fasting glucose levels, impaired glucose homeostasis, and insulin resistance, we wished to test the effect of an HFD on them. Pax+/+, Myh7fl/+ and Pax7iCre/+, Myh7fl/Δ mice were shifted to an HFD at 2 months of age, which was continued until 6 months of age (a duration of 4 months on an HFD). This led to weight gain in both control and muscle-specific Myh7 knockout mice; however, the weight gain in Pax7iCre/+, Myh7fl/Δ mice was significantly decreased in both females and males, after 3 to 4 months on an HFD (Fig. 5, A and B). To test whether the reduced body weight in the muscle-specific Myh7 knockout mice after 4 months on an HFD was due to reduced muscle mass, the TA, gastrocnemius, quadriceps, and soleus muscles were weighed. We observed significantly reduced muscle weight for all these muscles in the Pax7iCre/+, Myh7fl/Δ mice, whereas there was no difference in the white adipose tissue weight (Fig. 5, C to G). This suggests that the reduced body weight of the HFD-fed Pax7iCre/+, Myh7fl/Δ mice is a direct consequence of reduced muscle weight. Myh7 knockout mice on an HFD exhibit earlier incidence of altered glucose homeostasis and lipid accumulation in myofibers. (A and B) Body weight of female and male Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice fed with an HFD starting from 2 months after birth for 4 months (n = 5 to 6). (C to G) TA, Gas, Quad, and soleus muscle and white adipose tissue (WAT) weight of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice fed an HFD for 4 months (at 6 months of age) (n = 4 to 5). (H and I) Fasting blood glucose levels of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl female and male mice fed with an HFD from 2 months after birth for 4 months (n = 5 to 6). (J to M) GTT and ITT on Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice fed with an HFD for 4 months (at 6 months of age) and their respective area under the curve (n = 5 to 6). (N and O) Quantification of the grip strength of female and male Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice fed with an HFD for 4 months (at 6 months of age) (n = 7). gf, gram-force. (P to S) Oil Red O staining on TA muscle sections of normal chow diet–fed Pax+/+, Myh7fl/+ (P) and Pax7icre/+, Myh7Δ/fl (Q) or HFD-fed (4 months on an HFD) Pax+/+, Myh7fl/+ (R) and Pax7icre/+, Myh7Δ/fl mice (S). Arrows indicate the lipid accumulation inside the myofibers in HFD-fed Pax7icre/+, Myh7Δ/fl mice (S). Scale bar, 100 μm (S). (T and U) Quantification of the percentage Oil Red O+ area/total area of normal chow diet–fed Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (T) or HFD-fed (4 months on an HFD) Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (U) at 6 months of age (n = 4). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. Fasting blood glucose levels were increased in the Pax7iCre/+, Myh7fl/Δ mice after 1 month on an HFD, which reached significantly higher levels by 2 months and thereafter, in both females and males (Fig. 5, H and I). Glucose tolerance and insulin tolerance tests revealed significantly elevated glucose levels at all time points and increased area under the curve in the Pax7iCre/+, Myh7fl/Δ mice after 4 months on an HFD (Fig. 5, J to M). Reduced grip strength was also exhibited by both females and males in the muscle-specific Myh7 knockout mice after 4 months on an HFD (Fig. 5, N and O). Thus, compared to the Pax7iCre/+, Myh7fl/Δ mice on a normal diet, which developed glucose homeostasis impairment at 6 months of age (Fig. 4, C, D, I, and J), the HFD-fed mice exhibited symptoms of diabetes by 4 months of age (2 months on an HFD) (Fig. 5, H and I). Because we observed a reduction in muscle weight in the Myh7 knockout muscle of mice on a regular diet and HFD at 6 months, we carried out Oil Red O staining for lipids on TA cross sections. We found an increase in the Oil Red O+ area in the TA of a regular diet and HFD Pax7iCre/+, Myh7fl/Δ mice (Fig. 5, P to U). A clear accumulation of Oil Red O+ lipid droplets within the myofibers was observed in the HFD Pax7iCre/+, Myh7fl/Δ mice, which was not seen in the HFD Pax+/+, Myh7fl/+ or the regular diet–fed Pax7iCre/+, Myh7fl/Δ mice (arrows in Fig. 5S). The increased lipid accumulation in the muscle could lead to the observed decrease in muscle mass in the Pax7iCre/+, Myh7fl/Δ mice. Overall, muscle-specific loss of MyHC-slow function in mice resulted in impaired glucose homeostasis, insulin resistance, decreased Akt signaling, and reduced GLUT4 levels, characteristic of diabetes, by 6 months of age, which was accelerated when fed an HFD. Myh7 knockout mice exhibit decreased activity, reduced muscle force, and compromised muscle function To assess the functional effects of loss of MyHC-slow, Pax7iCre/+, Myh7fl/Δ and control mice were kept in animal monitoring cages to measure their activity. Pax7iCre/+, Myh7fl/Δ mice exhibited markedly reduced hourly locomotor activity and average locomotor activity over the entire 48-hour duration they were monitored, at the 2-, 6-, and 12-month time points (Fig. 6, A to F). The grip strength normalized to body weight was significantly reduced in the Pax7iCre/+, Myh7fl/Δ female mice from 4 months onward, until 12 months, the last time point that measurements were made; this difference was not observed at the 2-month time point (Fig. 6G). Identical results were observed in Pax7iCre/+, Myh7fl/Δ male mice (fig. S3B). In treadmill performance tests, the Pax7iCre/+, Myh7fl/Δ mice exhibited significantly diminished running distance, running speed, and duration they ran, at the 2-, 6-, and 12-month time points (Fig. 6, H and I, and fig. S3C). The ex vivo muscle contractility and maximum force produced exhibited significant reduction in the TA and gastrocnemius of Pax7iCre/+, Myh7fl/Δ mice at the 2-, 6-, and 12-month time points (Fig. 6, J to U). These results exhibited comparable trends upon normalization to the cross-sectional area of the muscle (fig. S6, A to L). At the 2-month time point, the force-frequency relationship exhibited similar trends between the Pax7iCre/+, Myh7fl/Δ and control mice to start with in both TA and gastrocnemius muscles but started showing differences at higher frequencies (Fig. 6, J and L, and fig. S6, A and C). These results indicate that the overall activity, grip strength, running capability, and muscle contractile force are all compromised upon loss of MyHC-slow function. Myh7 knockout mice exhibit decreased activity, reduced muscle force, and compromised muscle function. (A to F) Hourly (A, C, and E) and average (B, D, and F) total locomotor activity (x, y, and z axes) of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice at 2, 6, and 12 months, respectively (n = 5 to 8). (G) Quantification of the grip strength of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl female mice normalized to body weight, measured from 2 to 12 months of age (n = 8 to 12). (H and I) Treadmill running distance and speed of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl female mice, measured at 2, 6, and 12 months of age (n = 5 to 6). (J to U) Ex vivo contractility quantifying the absolute and maximal absolute force of TA and Gas muscles from Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl female mice at 2 months (J to M), 6 months (N to Q), and 12 months (R to U) of age (n = 4 to 6). Data are presented as means ± SEM. **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. Skeletal muscle–specific deletion of Myh7 leads to adult muscle fiber–type alterations Because loss of MyHC-slow function results in loss or atrophy of type I or slow myofibers, and compromised muscle function, we characterized the effect of Myh7 knockout on muscle fiber type. In the gastrocnemius muscle, a consistent decrease in the area and proportion of MyHC-IIa+ and MyHC-IIb+ myofibers was seen at the early (2 months) and late (12 months) time points in the Pax7iCre/+, Myh7fl/Δ mice compared to controls (fig. S7, A to Dc). A similar trend was observed for MyHC-IIx+ myofibers in the gastrocnemius, except that the myofiber area was not significantly different at the 2 months (early) time point (fig. S7, E to Fc). These results were validated by Western blots for MyHC-IIa and MyHC-IIb in the gastrocnemius at 2 and 12 months, where a robust decline was seen in their protein levels in the Pax7iCre/+, Myh7fl/Δ mice compared to controls (fig. S7, O to S). In the soleus, the area and proportion of MyHC-IIa+ myofibers were significantly increased at the 2- and 12-month time points in the Pax7iCre/+, Myh7fl/Δ mice (fig. S7, G to Hc). There was a significant increase in the MyHC-IIx+ myofiber area, whereas a decrease was seen in the proportion of MyHC-IIx+ myofibers in the Pax7iCre/+, Myh7fl/Δ mice (fig. S7, I to Jc). This is likely due to the unique effect on the soleus at all time points, where there is a decline in the total number of myofibers and MyHC-IIx+ myofibers, along with an increase in the average fiber size, upon loss of MyHC-slow function. The soleus has none to very few MyHC-IIb+ myofibers normally; however, we observed an increase in the proportion of MyHC-IIb+ myofibers at both 2- and 12-month time points, with no significant difference in the MyHC-IIb+ myofiber area, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S7, K to Lc). Another isoform, MyHC-embryonic, is expressed only during embryonic development and adult muscle regeneration in the skeletal muscle and not observed in normal adult muscle; Pax7iCre/+, Myh7fl/Δ mice exhibited expression of MyHC-embryonic in the smaller myofibers of the soleus, indicating active differentiation, atrophy, or regeneration happening in the muscle upon loss of MyHC-slow function (fig. S7, M to Nc). The troponin T1+ myofibers in the soleus (fig. S4, A to Db) were found to be MyHC-embryonic+, indicating atrophy or active regeneration, whereas such MyHC-embryonic+ myofibers were not seen in the gastrocnemius in the 1-year-old Pax7iCre/+, Myh7fl/Δ mice (fig. S8, W to Zc). In the TA, unlike the gastrocnemius, the area of MyHC-IIa+ and MyHC-IIx+ myofibers was increased whereas the proportion of MyHC-IIa+ and MyHC-IIx+ myofibers was decreased at 2 months, whereas the MyHC-IIa+ and MyHC-IIx+ myofiber area and proportion both were decreased at 12 months, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S8, A to Da, G to J, and M to P). Similarly, whereas the area of MyHC-IIb+ myofibers was unchanged, the proportion of MyHC-IIb+ myofibers was increased at 2 months; both the MyHC-IIb+ myofiber area and proportion were decreased at 12 months, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S8, E to Fa, K and L, Q, and R). Western blots confirmed these results, with MyHC-IIb levels exhibiting an increase and MyHC-IIa levels exhibiting a decrease at 2 months; the levels of both MyHCs were decreased at 12 months, in the Pax7iCre/+, Myh7fl/Δ mice (fig. S8, S to V). Overall, these results indicate severe fiber-type effects upon loss of MyHC-slow, with substantial variability across muscles. Previously, we observed similar results with loss of MyHC-embryonic, indicating that this could be a conserved feature seen with developmentally expressed MyHCs, which could have major roles in fiber-type specification (2, 3). Another aspect is the myofiber hypertrophy that was observed at early time points, which persisted in the soleus and TA at 2 months; in the gastrocnemius, this was not seen and atrophy had started by 2 months. Also, the soleus not only had an abnormally large number of smaller, regenerating or atrophying myofibers, which were MyHC-embryonic+, but also exhibited overall muscle hypertrophy even at 12 months; this is likely due to the high proportion of MyHC-slow+ fibers that are normally present in the soleus, which, when absent or atrophying, could lead to such drastic effects. Loss of MyHC-slow function in the skeletal muscle results in reduced NRF2 signaling and mitochondrial proteins by proteomic analysis Because major differences were observed in muscle structure and function in the MyHC-slow knockouts, we performed a mass spectrometry analysis to characterize the muscle proteomes of control and Pax3Cre/+, Myh7fl/Δ mice. Three distinct muscles—gastrocnemius, TA, and diaphragm—at P0 were used for mass spectrometry analysis. All three muscles exhibited distinct segregation between control and Myh7 knockout proteins by principal components analysis (PCA) (Fig. 7B and fig. S9, B and D). In the gastrocnemius, 76 proteins were unique in control, 138 were unique in the Myh7 knockouts, and 729 were common between the two (Fig. 7A). Of the 729 proteins expressed in control and Myh7 knockout gastrocnemius, 330 were down-regulated and 184 were up-regulated in the Myh7 knockout muscle; 215 proteins showed no notable change in expression between the control and Myh7 knockout muscle (Fig. 7C). Loss of MyHC-slow function in the skeletal muscle results in reduced NRF2 signaling and mitochondrial proteins by proteomic analysis. (A) Venn diagram depicting the number of unique and shared proteins expressed in the Gas muscle of Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice at P0 (n = 6). KO, knockout. (B) PCA plot of the Gas muscle from Pax+/+, Myh7fl/+ (red) and Pax3cre/+, Myh7Δ/fl (green) mice at P0, where each dot represents a biological replicate (n = 6). (C) Volcano plot showing up-regulated and down-regulated proteins in the Gas muscle from Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice at P0 (n = 6). (D) Pathway analysis (Reactome database) of Gas muscle proteins from Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice at P0. (E) Heatmap depicting the NRF2 pathway target protein expression in the Gas muscle of Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice at P0 (n = 6). (F and G) Representative Western blots for SOD1, GPx1, Prdx2, Prdx1, catalase, and α-tubulin (loading control) using protein lysates from the Gas muscle of P0 Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl, and Pax7icre/+, Myh7Δ/fl mice (F) and their densitometry (G) (n = 5). (H to M) Representative Western blots for NRF2, P-NRF2, and α-tubulin (loading control) using Gas and TA muscle protein lysates of P0 Pax+/+, Myh7fl/+, Pax3cre/+, Myh7Δ/fl, and Pax7icre/+, Myh7Δ/fl mice (H and I) and their densitometry (J to M) (n = 5). (N and O) Label-free quantitation (LFQ) intensity of DRP1 and FIS1 by mass spectrometry on P0 Gas muscle protein lysates of Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice (n = 6). (P) Heatmap depicting the expression of mitochondrial proteins in the Gas muscle of P0 Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl mice (n = 6). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. In the TA, 161 proteins were unique in control, 54 were unique in the MyHC-slow knockouts, and 464 were common between the two (fig. S9A). Among these 464 proteins, 50 were down-regulated, 11 were up-regulated, and 403 showed no notable change in expression. Similarly, in the diaphragm, 111 proteins were unique in control, 105 were unique in the MyHC-slow knockouts, and 691 were common between the two (fig. S9C). Among these 691 proteins, 34 were down-regulated, 23 were up-regulated, and 634 showed no notable change in expression. Heatmaps depicting the top 50 up-regulated or down-regulated proteins in the TA and diaphragm are shown (fig. S9, E and F). To identify the pathways that are dysregulated, pathway analysis was carried out using the proteins uniquely expressed or up-regulated in the Myh7 knockout muscle, using the Reactome database. Pathways related to the nuclear factor erythroid-derived 2-like 2 (NFE2L2), encoding the NRF2 protein, were consistently affected in all the three muscles analyzed (Fig. 7D and fig. S9, G and H). NRF2 is a transcription factor master regulator of antioxidant defense response, sensing, and responding to reactive oxidants; dysregulation of the NRF2 signaling pathway leads to a wide range of diseases spanning from cancers to genetic disorders (22–26). Pathways related to oxidative stress, ROS, stress response, and glutathione conjugation were also found to be affected in the Myh7 knockout muscle in the pathway analysis (Fig. 7D and fig. S9, G and H). Numerous NRF2 pathway target proteins (27, 28) were found to be down-regulated in the Myh7 knockout muscle among the list of candidates identified by mass spectrometry (Fig. 7E). A few well-established NRF2 targets that were down-regulated by mass spectrometry, namely, superoxide dismutase 1 (SOD1), glutathione peroxidase 1 (Gpx1), peroxiredoxin 1 and 2 (Prdx1 and Prdx2), and catalase, were tested by Western blotting, which clearly confirmed their down-regulation in the P0 Myh7 knockout gastrocnemius muscle (Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ) (Fig. 7, F and G). Total NRF2 and phospho-NRF2S40 (active form which promotes NRF2 nuclear localization) levels were clearly down-regulated in the P0 Myh7 knockout gastrocnemius and TA muscles (Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ) (Fig. 7, H to M). Validating the results showing the up-regulation of the mitochondrial fission proteins (Fig. 2, Q and R), mass spectrometry analysis indicates that both DRP1 and FIS1 were significantly up-regulated in the P0 Myh7 knockout gastrocnemius (Fig. 7, N and O). Numerous proteins associated with mitochondrial function were found to be down-regulated in the P0 Myh7 knockout gastrocnemius by mass spectrometry (Fig. 7P), confirming the decline in oxidative phosphorylation complexes and mitochondrial content in the gastrocnemius in both Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ P0 mice (Fig. 2, M and N). Thus, the mass spectrometry analysis indicates a notable down-regulation of the key oxidative stress regulating NRF2 signaling pathway in the Myh7 knockout muscle; these results were confirmed by target protein expression and levels of total and phospho-NRF2S40, all of which exhibited significant down-regulation. The enhanced mitochondrial fission and reduced mitochondrial content that was observed in the Myh7 knockout muscle were validated in the mass spectrometry analysis. NRF2 signaling is elevated in slow muscles with reduced expression of NRF2 target proteins in the Myh7 knockout mice To characterize NRF2 signaling in depth in the skeletal muscle, Western blots were carried out on protein lysates from representative fast (TA) and slow (soleus) muscles of wild-type (WT) mice. Troponin T1, a marker of slow fibers, showed clearly increased expression in the soleus, whereas glyceraldehyde phosphate dehydrogenase (GAPDH), a key glycolytic enzyme, exhibited elevated expression in the TA (Fig. 8, A and B). NRF2 levels were almost sixfold higher in the soleus compared to the TA, indicating that it is expressed preferentially in the slow fibers; KEAP1 levels did not exhibit any significant changes in expression between the TA and the soleus (Fig. 8, A and B). To further validate this, Western blots for the NRF2 targets catalase, Gpx1, Prdx1, and SOD1 were carried out on the WT TA and soleus protein lysates, where all of them exhibited significantly increased expression in the soleus compared to the TA (Fig. 8, A and C). NRF2 signaling is elevated in slow muscles, and Myh7 knockout mice exhibit reduced expression of NRF2 target proteins. (A to C) Representative Western blots for troponin T1, GAPDH, NRF2, KEAP1, catalase, GPx1, Prdx1, SOD1, and α-tubulin (loading control) using TA and soleus muscle protein lysates from 6-month-old WT mice (A) and their densitometry (B and C) (n = 5). (D to H) Representative Western blots for P-NRF2, NRF2, and α-tubulin (loading control) using Gas muscle protein lysates of 2- and 6-month-old Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (D) and their densitometry (E to H) (n = 5). (I to K) Representative Western blots for total NRF2 and H3A in the nuclear fraction and total NRF2 and GAPDH in the cytoplasmic fraction (I), using protein lysates from the Gas muscle of P0 Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice and their densitometric quantification (J and K) (n = 4). (L to Q) Representative Western blots for SOD1, GPx1, Prdx2, Prdx1, catalase, carbonylated proteins, and α-tubulin (loading control) using protein lysates from the Gas muscle of 2- and 6-month-old Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (L and O) and their densitometry (M, N, P, and Q) (n = 5). (R to T) Cross sections through the Gas of Pax+/+, Myh7fl/+ (R to Rb), and Pax7icre/+, Myh7Δ/fl (S to Sb) mice at 12 months of age, labeled by immunofluorescence for γ-H2AX (red) and DAPI (blue), and quantification of γ-H2AX+ cells (T). Scale bars, 100 μm (S) and 10 μm (Sb). (U to X) Representative Western blots for γ-H2AX, cleaved caspase-3, p21, IGF-1, and α-tubulin (loading control) using protein lysates from the Gas muscle of 2- and 6-month-old Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (U) and their densitometry (V to X) (n = 5). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. As observed at the P0 time point, at the 2- and 6-month time points, both total NRF2 and phospho-NRF2S40 levels were clearly down-regulated in the gastrocnemius and TA muscle of Pax7iCre/+, Myh7fl/Δ mice (Fig. 8, D to H, and fig. S10, A to E). To further confirm this, the nuclear and cytoplasmic fractions were isolated from the gastrocnemius of P0 control and Pax7iCre/+, Myh7fl/Δ mice; Western blots indicated that levels of NRF2 were significantly reduced in both the nuclear fraction (active form of NRF2) marked by histone 3A (H3A) and the cytoplasmic fraction (inactive form of NRF2) marked by GAPDH (Fig. 8, I to K). Similar to P0, the NRF2 targets SOD1, Gpx1, Prdx1, Prdx2, and catalase were all significantly down-regulated in the gastrocnemius and TA muscle of Pax7iCre/+, Myh7fl/Δ mice at 2 and 6 months of age (Fig. 8, L to N, and fig. S10, F to H). Because a clear down-regulation of NRF2, a master regulator of antioxidant defense response, and several of its target antioxidant enzymes listed above was observed, we hypothesized that the Pax7iCre/+, Myh7fl/Δ muscle would exhibit increased oxidative stress. To test this, Western blots using an antibody against carbonylated proteins, an irreversible posttranslational modification resulting from increased oxidative stress or ROS production, was used to probe 2- and 6-month-old control and Pax7iCre/+, Myh7fl/Δ gastrocnemius protein lysates. Results indicated a clear, almost fourfold increase in carbonylated proteins in the Pax7iCre/+, Myh7fl/Δ lysates (Fig. 8, O to Q). No change in carbonylated proteins were seen in the gastrocnemius muscle lysates of Pax7iCre/+, Myh7fl/Δ mice at P0, suggesting that the ROS levels are similar at early postnatal stages (fig. S10, Q and R). Increased oxidative stress is also associated with DNA damage, which was analyzed using the DNA double-strand break marker phosphorylated H2AX (γH2AX), which was found to be highly up-regulated in the 1-year-old gastrocnemius muscle of Pax7iCre/+, Myh7fl/Δ mice (Fig. 8, R to T). This was also validated by Western blots at 2 and 6 months using gastrocnemius protein lysates, where γH2AX levels were high in the Pax7iCre/+, Myh7fl/Δ mice, as were the levels of cleaved caspase-3, marking increased apoptosis and the senescence marker p21, whereas reduced levels were seen of the growth factor IGF-1, possibly due to elevated DNA damage (Fig. 8, U to X). Although γH2AX+ cells were high in number at P0 possibly because this is related to proliferation-associated DNA replication, no difference was found between control, Pax3Cre/+, Myh7fl/Δ and Pax7iCre/+, Myh7fl/Δ mice at P0, as was the case with γH2AX, cleaved caspase-3, and p21 levels by Western blots (fig. S10, I to P). Thus, muscles rich in slow fibers exhibit increased NRF2 and target gene expression compared to fast fiber–rich muscles. Loss of MyHC-slow function results in an overall down-regulation of NRF2, including the active and inactive forms. This leads to decrease in levels of the NRF2 targets, majorly antioxidant enzymes, resulting in elevated oxidative stress and increased levels of carbonylated proteins upon loss of MyHC-slow function. Also, the increased oxidative stress, elevated ROS, senescence, and DNA damage are not seen at early postnatal stages, which become apparent from 2 months of age, which could explain the early muscle hypertrophy followed by the atrophy in the Myh7 knockouts. Activating the NRF2 pathway by sulforaphane administration rescues most of the skeletal muscle and metabolic defects in Myh7 knockout mice Because loss of MyHC-slow function led to depletion of NRF2 activation, reduction in antioxidant enzymes, and consequent increase in oxidative stress, we hypothesized that administration of an activator of the NRF2 pathway might rescue these defects. Sulforaphane is an isothiocyanate compound that can interact with KEAP1, leading to release of NRF2 and pathway activation (29, 30). We administered sulforaphane or dimethyl sulfoxide (DMSO; vehicle) to Pax7iCre/+, Myh7fl/Δ mice starting at 16 weeks of age, at 1 mg/kg body weight, three times a week for 8 weeks (until 24 weeks of age), following which the mice were tested (Fig. 9A). A significant increase was seen in the body weight of the sulforaphane-treated mice, compared to vehicle-treated mice, at 6 months of age, which was also observed at the individual muscle level for the TA, gastrocnemius, quadriceps, and soleus muscles (Fig. 9, B to F). At the myofiber level, the proportion of smaller fibers (500 to 1250 μm2) was decreased, and the proportion of larger fibers (>2250 μm2) was increased in the gastrocnemius, upon sulforaphane treatment (Fig. 9, G to Gb). A significant increase in the average fiber area was seen in the gastrocnemius (Fig. 9Gc). Similar results with respect to the decrease in smaller fiber proportion, increase in larger fiber proportion, and increase in average fiber area were observed in the TA (fig. S11, U to Uc). The ex vivo muscle contractility and maximum force produced in the TA and gastrocnemius muscles exhibited a significant increase upon sulforaphane treatment in the Pax7iCre/+, Myh7fl/Δ mice (Fig. 9, H to Ia). In treadmill performance tests, the sulforaphane-administered Pax7iCre/+, Myh7fl/Δ mice exhibited significantly increased running distance and running speed, compared to vehicle treatment (Fig. 9, J and K). Activating the NRF2 pathway by sulforaphane administration rescues most of the skeletal muscle and metabolic defects in Myh7 knockout mice. (A) Schematic depicting the sulforaphane dosing strategy starting from 16 weeks. (B to F) Body weight (grams) and muscle weight (TA, Gas, Quad, and soleus) (milligrams) of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice at 6 months of age (n = 5 to 6). (G to Gc) Gas muscle cross sections labeled by immunofluorescence for laminin (red) and DAPI (blue) of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice at 6 months of age and quantification of the proportion of myofibers grouped according to myofiber area (Gb) and average myofiber area (μm2) (Gc) through the Gas (n = 4). Scale bar, 100 μm (Ga). (H to Ia) Ex vivo contractility quantifying the absolute and maximal absolute force of TA and Gas in DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice (n = 4). (J and K) Treadmill running distance and speed of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice (n = 6). (L to V) Representative Western blots for P-NRF2, NRF2, KEAP1 (Gas and TA), SOD1, GPx1, Prdx1, catalase, MyHC-IIa, MyHC-IIb, PGC-1α, mitochondrial complex, carbonylated proteins (Gas), and α-tubulin (loading control) using protein lysates of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice (L, O, S, and T) and their densitometry (M, N, P to R, U, and V) (n = 4). (W) Grip strength quantification of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice normalized to body weight (n = 6). (X to Ya) GTT and ITT of DMSO- or sulforaphane-treated Pax7icre/+, Myh7Δ/fl mice and their area under the curve (n = 5 to 6). (Z) Quad muscle ATP level quantification of DMSO- or sulforaphane-treated Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (n = 5). Data are presented as means ± SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. Sulforaphane-administered Pax7iCre/+, Myh7fl/Δ mice exhibited elevated levels of total and phospho-NRF2S40 and decreased levels of KEAP1 in the gastrocnemius and TA muscle (Fig. 9, L to N). A previous report indicates that sulforaphane treatment leads to a decrease in KEAP1 levels (31). Accordingly, a significant increase was seen in the levels of the downstream targets of NRF2 signaling such as SOD1, Gpx1, Prdx1, and catalase in the sulforaphane-treated mice gastrocnemius and TA (Fig. 9, O and P, and fig. S11, X and Y). Levels of the MyHC-IIa and MyHC-IIb isoforms exhibited a significant increase upon sulforaphane administration in the gastrocnemius and TA (Fig. 9, O and Q, and fig. S11, X and Z). Increased expression of the mitochondrial biogenesis factor PGC-1α was also observed upon sulforaphane treatment in the gastrocnemius of Pax7iCre/+, Myh7fl/Δ mice (Fig. 9, O and R). Sulforaphane treatment resulted in a significant increase in the protein levels of the five mitochondrial complexes involved in oxidative phosphorylation and a clear reduction in the carbonylated protein levels indicating reduced oxidative stress, in the gastrocnemius (Fig. 9, S to V). Increased ATP levels were observed in the quadriceps muscle, indicating improved mitochondrial function, upon sulforaphane treatment (Fig. 9Z). At the functional level, a significant increase in grip strength normalized to body weight was observed upon sulforaphane administration in the Pax7iCre/+, Myh7fl/Δ mice (Fig. 9W). The elevated fasting glucose levels, glucose utilization capability measured by the glucose tolerance test, and the response to insulin measured by the insulin tolerance test all exhibited a significant improvement in the sulforaphane-treated Pax7iCre/+, Myh7fl/Δ mice (Fig. 9, X to Ya). The reduced activity that was observed in the Pax7iCre/+, Myh7fl/Δ mice exhibited a significant increase following sulforaphane treatment (fig. S11, V and Va). All these effects were independent of food consumption because the vehicle- and sulforaphane-treated Pax7iCre/+, Myh7fl/Δ mice did not show any difference in the quantity of food consumed (fig. S11, W and Wa). Furthermore, administration of sulforaphane to the control Pax+/+, Myh7fl/+ mice following the same treatment regime as described above for the Pax7iCre/+, Myh7fl/Δ mice did not have any effect on the body weight, individual muscle weight (TA, gastrocnemius, quadriceps, and soleus), grip strength normalized to body weight, running distance and speed in treadmill assays, activity levels, food consumption, glucose utilization capability, and response to insulin, compared to vehicle-treated controls (fig. S11, A to P). No change was observed in MyHC-IIb levels, whereas an increase in both MyHC-IIa and MyHC-slow levels was observed in the sulforaphane-treated Pax+/+, Myh7fl/+ mice (fig. S11, Q to T). Our findings indicate that sulforaphane treatment for a duration of 2 months, before onset of diabetes, increased NRF2 levels and signaling effectively, leading to elevated levels of downstream antioxidant enzymes and rescue of almost all the structural, functional, mitochondrial, and metabolic defects exhibited by the skeletal muscle–specific MyHC-slow loss of function. Intriguingly, the administration of sulforaphane failed to provide any additional functional benefits to control mice, indicating that the observed effects are specific to loss of MyHC-slow function and resultant reduction in NRF2 signaling. Administration of sulforaphane after diabetes onset results in a partial rescue of the skeletal muscle and metabolic defects in Myh7 knockout mice Because sulforaphane administration starting before diabetes onset led to improvement in most of the musculoskeletal, functional, and metabolic defects associated with loss of MyHC-slow function, we wished to test whether sulforaphane administration after the onset of diabetes would have any effect. Therefore, sulforaphane or DMSO (vehicle) was administered to Pax7iCre/+, Myh7fl/Δ mice starting at 24 weeks of age, at 1 mg/kg body weight, three times a week for 8 weeks (until 32 weeks of age), following which the mice were tested (fig. S12A). A significant increase was seen in the body weight of the sulforaphane-treated mice; however, at the individual muscle level, only the TA exhibited increased weight, and the gastrocnemius, quadriceps, and soleus muscles exhibited a trend toward an increase, which was not significant (fig. S12, B to F). The ex vivo muscle contractility and maximum force produced in the TA and gastrocnemius muscles exhibited a significant increase upon sulforaphane treatment in the Pax7iCre/+, Myh7fl/Δ mice (fig. S12, G to J). We observed a clear improvement in fasting glucose levels, glucose utilization capability measured by the glucose tolerance test and the response to insulin measured by the insulin tolerance test, in the sulforaphane-treated Pax7iCre/+, Myh7fl/Δ mice (fig. S12, K to N). A significant increase in grip strength normalized to body weight was also observed upon sulforaphane administration (fig. S12O). Administration of sulforaphane to the control Pax+/+, Myh7fl/+ mice following the same treatment regime as described above did not have any effect on the body weight, individual muscle weight (TA, gastrocnemius, quadriceps, and soleus), muscle contractility, glucose utilization capability, and response to insulin and grip strength normalized to body weight, compared to vehicle-treated controls (fig. S12, P to Y). Thus, administration of sulforaphane post–diabetes onset did improve most of the abnormalities such as body weight, muscle contractility, glucose utilization, response to insulin, and grip strength but did not improve other defects such as individual muscle weight, exhibited by the skeletal muscle–specific MyHC-slow loss of function. The administration of sulforaphane also failed to provide any additional benefits to control mice in this treatment paradigm. DISCUSSION MyHC-slow is the lone slow-type MyHC isoform expressed in the mammalian skeletal muscle, with very little known about its specific function. Here, using skeletal muscle–specific Myh7 knockout mice, we characterize the role of MyHC-slow during early postnatal and adult stages. Our results indicate that MyHC-slow is required for proper development of the skeletal muscle and metabolic homeostasis. This study provides direct evidence for the role of MyHC-slow and the slow oxidative fibers in type 2 diabetes and metabolic regulation (Fig. 10). Model. (A) MyHC-slow mediates normal oxidative phosphorylation levels and mitochondrial dynamics in the myofibers, maintaining normal blood glucose levels, insulin sensitivity, GLUT4 levels, mitochondrial function, NRF2 signaling, and MyHC proportions. (B) The absence of MyHC-slow results in reduced oxidative phosphorylation levels and increased mitochondrial fission in the myofibers, leading to increased ROS generation, muscle atrophy, altered MyHC proportions, and significant reduction in NRF2, GLUT4, and mitochondrial function, causing reduced insulin sensitivity and elevated blood glucose levels. FAPs, fibro-adipogenic progenitors. Created in BioRender. Mathew, S. (2026) https://BioRender.com/d6n81hh. MyHC-slow expression is required for slow myofiber maintenance at adult stages, in a muscle-specific manner. Although Myh7 knockout mice exhibited a complete loss of slow myofibers in the gastrocnemius, slow myofibers were severely atrophied but retained in the soleus (Fig. 2, C to Db, and fig. S4, C to Db). This points to muscle-specific effects and developmental differences with respect to MyHC expression between these muscles (32). However, it is clear that loss of MyHC-slow has effects not only on slow but also on fast myofibers and muscles. This could be because all myofibers express MyHC-slow during primary myogenesis in embryonic development, although MyHC-slow expression becomes restricted to the slow myofibers postnatally (32). The future fast myofibers down-regulate MyHC-slow or the fast myofibers form during secondary myogenesis (32, 33). The other fiber type that consistently exhibited a decline in fiber proportions (except in the soleus) across time points was MyHC-IIa, indicating that MyHC-slow regulates additional fiber types than slow fibers (figs. S7, A to Bc, and S8, A to Ba, H, and N). Although MyHC-IIa+ fibers are fast fibers, they are oxidative in nature and are therefore metabolically closer to slow fibers. The elevated oxidative stress pathways in the Myh7 knockouts could have a more severe effect on the MyHC-IIa fibers than the glycolytic MyHC-IIx+ and MyHC-IIb+ fibers. However, the Myh7 knockouts also exhibited changes in the proportion of MyHC-IIx+ and MyHC-IIb+ fibers, possibly due to direct effects of these fibers having expressed MyHC-slow during developmental stages or due to compensatory effects from loss of the oxidative fiber types. In the gastrocnemius, the proportion of all adult MyHC+ myofibers decreased at all the time points, except P30 (fig. S7, A to Fc). Similar results were observed in the TA, where, except for MyHC-IIb+ myofibers at 2 months, the proportion of all other MyHC+ myofibers decreased across time points (fig. S8, A to R). Generally, loss of one MyHC isoform results in the gain of one or more other isoforms, which seems not to be the case here. This could most likely be due to the muscle atrophy associated with loss of MyHC-slow or due to an increase in the proportion of hybrid MyHC+ myofibers expressing multiple MyHC isoforms (34). Intriguingly, loss of MyHC-slow led to opposing effects at early and late postnatal stages; whereas muscle hypertrophy was observed at early time points (P0 to ∼2 months of age) (Fig. 1 and figs. S1 to S3), atrophy was seen at later time points (4 to 12 months of age) (figs. S2 and S3). These phenotypes correlated with increased proportions of the MyHC-IIb+ and MyHC-IIx+ myofibers at early time points and their reduction at later time points. MyHC-IIb+ and MyHC-IIx+ myofibers are relatively larger in size (12, 35), and their increase at early time points suggests a compensatory effect due to the loss of MyHC-slow+ myofibers in the Myh7 knockouts. Also, at early time points until weaning, mice are mostly sedentary, with relatively limited muscle use; however, they are more active postweaning, with relatively increased muscle use, leading to enhanced muscle contraction–induced ROS production. Thus, at early postnatal stages, no changes were seen between controls and Myh7 knockouts in carbonylated proteins for ROS levels, γH2AX+ cells for DNA damage, cleaved caspase-3 for apoptosis, and p21 for senescence whereas all of these were increased consistently from ∼2 months of age onward, contributing to the atrophy (Fig. 8, O to X, and fig. S10, I to R). Reduced NRF2 signaling and antioxidant enzyme production in the Myh7 knockout mice add to this, all of which contribute to the muscle atrophy at later time points (28, 36, 37). The atrophying slow fibers in the Myh7 knockout soleus express MyHC-embryonic, a marker of atrophy or regeneration, suggesting that these fibers could be undergoing continuous degeneration or regenerative response, allowing them to survive by satellite cell-mediated fusion (fig. S8, X to Xc); MyHC-embryonic expression was not seen in the Myh7 knockout fast muscles such as gastrocnemius, suggesting that this is unique to slow fiber–rich muscles. The soleus exhibited an increase in the proportion of large fibers and small fibers, leading to a mixed atrophy-hypertrophy phenotype of the muscle at all stages; with the fast muscles, although early hypertrophy of the muscle was seen, this was quickly followed by atrophy and smaller average fiber size thereafter. Thus, distinct mechanisms regulating slow fiber maintenance, atrophy, MyHC-embryonic expression, and fiber size operate in fast versus slow muscles. A precise balance between ROS production and antioxidant systems is crucial to proper skeletal muscle formation. A switch from glycolytic to oxidative metabolism, leading to elevated ROS by fetal stages, is associated with skeletal muscle differentiation during development (38). The Pitx2 and Pitx3 transcription factors regulate the expression of nuclear respiratory factor 1 (NRF1) and antioxidant enzymes during myogenic differentiation at fetal stages to adjust the redox state and prevent DNA damage (38). Similar results are seen in skeletal muscle regeneration, where increased ROS levels promote satellite cell differentiation through the P38α mitogen-activated protein (MAP) kinase pathway (39). Our results with MyHC-slow indicate that its loss of function tilts the balance to augment ROS levels, causing a decline in antioxidant systems involving the key regulator NRF2, leading to impaired muscle development, homeostasis, and function. The elevated ROS levels result in increased protein carbonylation, DNA damage, and apoptosis in the Myh7 knockout muscle. Loss of MyHC-slow led to marked effects on the myofiber mitochondria. Although slow myofibers have elevated mitochondrial numbers and atrophy or loss of slow myofibers will lead to a decline in overall mitochondrial numbers in the whole muscle, the mitochondrial effects were also seen in muscles with a higher proportion of fast myofibers such as the TA and gastrocnemius. The myofiber mitochondria not only exhibited reduced abundance but also decreased activity, biogenesis, fission-fusion dynamics, and oxidative capacity, indicating that developmental expression of MyHC-slow could be crucial to mitochondrial characteristics in all myofibers. An interesting parallel can be drawn with the PGC-1α-β double knockouts, where similar effects were observed on mitochondrial characteristics in the myofibers (40). However, unlike the Myh7 knockouts, the PGC-1 coactivator knockouts did not show notable changes in muscle fiber type, insulin sensitivity, or glucose utilization (40). Another study reported that muscle-specific PGC-1α knockouts exhibited impaired glucose tolerance but normal insulin sensitivity (17). This indicates that MyHC-slow plays additional roles in skeletal muscle development, homeostasis, and metabolic regulation, independent of the PGC-1 coactivators. Mitochondrial dysfunction in the skeletal muscle leads to impaired glucose utilization and insulin resistance (41). Down-regulation of oxidative phosphorylation genes such as PGC-1α has been reported in the human diabetic skeletal muscle (42). Altered glycolytic to oxidative enzyme capacity, implying mitochondrial dysfunction, has been reported in individuals with type 2 diabetes (43). Reduction in the skeletal muscle mitochondrial oxidative phosphorylation capability and function contributes to human age-related insulin resistance and type 2 diabetes (44, 45). Decreased mitochondrial oxidative activity and ATP production was associated with age-related insulin resistance in the skeletal muscle (45). Mitochondrial dysfunction and increased oxidative stress in the skeletal muscle causes a switch in the cellular redox milieu, leading to insulin resistance and metabolic dysregulation (46). Increased ROS production has been shown to cause mitochondrial abnormalities in the skeletal muscle, which preceded insulin resistance, indicating that mitochondrial dysfunction in the skeletal muscle could lead to type 2 diabetes (41). The crucial ROS-generating enzyme NADPH oxidase 2 (NOX2) levels were increased in the skeletal muscle of mice given an HFD, leading to insulin resistance, which was not seen in NOX2 null mice, indicating that HFD-induced ROS levels contribute to insulin resistance in the skeletal muscle (47). Our results with Myh7 knockouts exhibiting elevated ROS and mitochondrial dysfunction agree with many of these studies. For instance, HFD administration led to elevated blood glucose levels and insulin resistance in the skeletal muscle–specific Myh7 knockout mice ∼2 months earlier than with a normal diet, possibly due to increased ROS levels. The reduced muscle GLUT4 levels seen in the Myh7 knockouts raise the possibility that the observed dysregulated glucose homeostasis could be multifactorial in nature. In a notable parallel in adipocytes, elevated ROS and mitochondrial dysfunction led to reduced GLUT4 levels (48). We also observed that the Myh7 knockouts exhibited reduced locomotor activity and were relatively more sedentary at all time points (Fig. 6, A to F). This could contribute to the insulin resistance and reduced glucose utilization. NRF2 has been proposed to be a key target pathway in type 2 diabetes. For instance, activation of NRF2 signaling in the diabetic db/db mouse model by crossing to a hypomorph of the negative regulator Keap1 or by administration of chemical inducers of the pathway led to reduced insulin resistance and blood glucose levels (49, 50). NRF2 activation by skeletal muscle–specific deletion of Keap1 resulted in enhanced force generation by the muscle, indicating that the reduced force production that we observe in the Myh7 knockouts could be mediated in part by the reduced NRF2 signaling (51). A key question that arises from this study is the mechanism underlying the down-regulation of NRF2 in the Myh7 skeletal muscle–specific knockouts. One likely reason could be that the slow fibers, known to express the highest levels of NRF2, are completely lost (gastrocnemius) or severely atrophied (soleus) in the Myh7 knockouts, leading to depletion of NRF2 levels (Figs. 2, A to Db, and 8, A to C; and fig. S4, A to Db). Another possibility is that loss of MyHC-slow leads to impaired mechanotransduction, as seen with the loss of MyHC-embryonic (3). Studies indicate that NRF2 responds to mechanical cues in various contexts such as in the skin plasmacytoid dendritic cells, running exercise–induced improvement in femur bone mass and structure, and beneficial effects of exercise on the skeletal muscle (52–54). Cross-talk between NRF2 signaling and mechanotransduction pathways and molecules such as NOX4, YAP, and integrins have been reported. NRF2 expression drastically increased following exercise in the skeletal muscle of control compared to sedentary mice; this effect was not observed in the skeletal muscle–specific Nox4 knockout mice, indicating that the exercise-induced NRF2 expression in the muscle is mediated by NOX4 (28). YAP signaling activates the nuclear matrix–associated protein SATB2, which regulates downstream targets of NRF2 in renal cell carcinoma; YAP and FOXO1 interact and activate antioxidant genes such as catalase, known to be NRF2 targets, in cardiomyocytes (55, 56). Integrin α4β1 regulates NRF2 in fibroblasts dependent on environmental cues (57). Thus, the loss of slow fibers and aberrant mechanical cues due to loss of MyHC-slow contributes to the reduced NRF2 levels in the Myh7 skeletal muscle–specific knockouts. Sulforaphane was identified as a potent inducer of phase 2 detoxification enzymes such as quinone reductase and glutathione transferase in various mouse tissues (30). A recent study suggests that sulforaphane can regulate hepatic glucose production and improve glucose intolerance in patients with type 2 diabetes by promoting NRF2 nuclear translocation and reduced gluconeogenesis in the liver (29). NOX4 required for redox balance, which, when knocked out in the muscle, led to increased oxidative stress and insulin resistance, was rescued by sulforaphane treatment (28). Thus, induction of oxidative stress and insulin resistance in metabolically active tissues mediated by attenuation of NRF2 signaling can be treated effectively with sulforaphane. In a study investigating previously unidentified human mutations in MYH7 in patients with dilated cardiomyopathy, 27 variations were identified in the MYH7 gene in 137 patients (58). Twenty-five percent of these patients with dilated cardiomyopathy were also diabetic, suggesting that there could be a correlation between mutations in MYH7 and diabetes (58). Certain case studies on individuals with cardiovascular dysfunction with MYH7 mutations also indicate that such patients exhibit diabetes (59, 60). Thus, although unexplored currently, a genetic correlation between MYH7 mutations, gene function, and type 2 diabetes seem to be emerging, which our results make a strong case for. MATERIALS AND METHODS Mice Mice were euthanized at selected time points (P0, P30, and 2, 6, and 12 months), and the TA, gastrocnemius, soleus, and quadriceps hindlimb muscles and diaphragm muscle were harvested. Mice were maintained, and all animal experiments were performed according to the protocols approved by an institutional animal ethics committee (protocol number: RCB/IAEC/2021/089). DMSO and sulforaphane were administered intraperitoneally to Pax7+/+, Myh7fl/+ (Control) and Pax7icre/+, Myh7Δ/fl (knockout) mice at 1 mg/kg body weight concentration. DMSO and sulforaphane were administered 3 days a week for 2 months starting from 4 months of age. All experiments were performed after 2 months of drug treatment (6 months of age), and different muscles were harvested. In the second set of experiments, DMSO and sulforaphane were administered 3 days a week for 2 months starting from 6 months of age; all experiments were performed after 2 months of drug treatment at 8 months of age. Mice generation Myh7 germline and conditional knockouts were generated at the Transgenic and Gene Targeting Core Facility, University of Utah (Salt Lake City, UT, USA) from the targeted embryonic stem (ES) cell line Myh7tm4a(KOMP)Wtsi obtained from the KOMP repository, University of California, Davis, Mouse Biology Program. Briefly, Myh7fl/+ mice were generated from the ES cells, where exon 26 of the Myh7 gene is flanked by LoxP sites (fig. S1A). Myh7fl/+mice were crossed to the ubiquitous Cre expressing CMV-Cre mice (JAX 006054) to generate the Myh7Δ/+ mice. To generate skeletal muscle–specific knockout of Myh7, two muscle-specific Cre driver lines, Pax3cre (61) and Pax7icre (62), were used. First, the heterozygous Myh7Δ/+ mice were crossed with Pax3cre and Pax7icre mice; from this cross, Pax3cre/+, Myh7Δ/+ and Pax7icre/+, Myh7Δ/+ mice were obtained and crossed to Myh7fl/fl mice. From this cross, the skeletal muscle–specific knockout of Myh7, namely, Pax3cre/+, Myh7Δ/fl and Pax7icre/+, Myh7Δ/fl, were obtained and used as the knockout, whereas Pax3+/+, Myh7fl/+ or Pax7+/+, Myh7fl/+ siblings were taken as controls for the experiments. Genotyping Genotyping of Myh7 WT, floxed, delta alleles, and Cre alleles were performed by polymerase chain reaction (PCR) using whole genomic DNA extract isolated from mouse ear clips. Primer sequences used for genotyping of the different alleles are listed. Briefly, genomic DNA was isolated by the HotSHOT lysis method (63). PCR master mix was prepared as follows: 7.5 μl of GoTaq G2 Hot Start Green Master Mix (Promega, M7423), 1 μl of a 5 μM solution of primers for Myh7 (one forward and one reverse, for each of Myh7-wt, Myh7-delta, and Myh7-fl), 3 μl of HotSHOT DNA, and 2.5 μl of nuclease-free water. For the WT allele, CSD F and CSD ttR primers were used; for the delta allele, LacF and CSD R primers were used; and for the floxed allele, CSD loxF and CSD R primers were used (table S1). The PCR cycling conditions were as follows: 95°C for 5 min; 34 cycles of 95°C for 30 s, 60°C for 20 s, and 72°C for 40 s; 72°C for 10 min; and 10°C for 10 min. The PCR products were separated on 2% agarose gel, and Myh7 genotypes were determined on the basis of the band size (WT: 509 bp; delta: 606 bp; floxed: 308 bp). Grip strength measurement and treadmill test A grip strength meter (GT3, Panlab, Harvard Apparatus) was used to measure the forelimb and hindlimb grip strength, which were normalized to the body weight. The treadmill assay was performed as described previously (64). Briefly, mice were acclimatized to running on a treadmill apparatus (Columbus Instruments), for 3 days at 10 m/min for 5 min. Following this, the final experiment was performed, where mice were allowed to run for 5 min at 10 m/min, after which the speed was increased by 2 m/min every 2 min until exhaustion. The running time, distance, and speed reached by the animals were quantified. The investigators were blind to the treatment given for these assays. Indirect calorimetry study To measure locomotor activity and food consumption, the Comprehensive Laboratory Animal Monitoring System (CLAMS) (Columbus Instruments) was used. Briefly, mice were kept in animal monitoring cages with free access to food and water for 72 hours, maintaining a 12-hour light and a 12-hour dark cycle. The mice were allowed to acclimatize for 24 hours, following which the locomotor activity and food consumption were measured for the next 48 hours. Metabolic assays The glucose tolerance test and insulin tolerance test were performed as described previously (17). Briefly, for the glucose tolerance test, Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice were kept in a fasting state for 16 hours, with unlimited access to water, following which the fasting baseline blood glucose levels were measured. Thereafter, 2 g of glucose per kilogram of body weight was administered by intraperitoneal injection. Blood glucose levels were monitored after this at specific time intervals (15, 30, 60, 90, and 120 min) by taking blood from the tail, using test strips for blood glucose monitoring (Dr. Morepen, BG-03). For the insulin tolerance test, Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice were kept in a fasting state for 6 hours after which the fasting baseline blood glucose levels were measured. Thereafter, 0.75 U insulin (Eli Lilly and Company, HI0240) per kilogram of body weight (for a normal diet) or 1 U insulin per kilogram of body weight (for HFD) was administered by intraperitoneal injection. Blood glucose levels were monitored after this at specific time intervals (15, 30, 60, 90, and 120 min) by taking blood from the tail, using test strips for blood glucose monitoring. In vivo insulin signaling pathway activation by p-Akt was performed as described previously (65). Briefly, Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice (at 2 and 6 months of age) were kept in a fasting state for 6 hours. Thereafter, 0.75 U insulin per kilogram of body weight was administered by intraperitoneal injection. After 15 min of insulin injection, different muscles were harvested. HFD experiments were performed using a rodent HFD (60% fat, Research Diets, D12492); an HFD was given to Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice starting from 2 months and continuing until 6 months of age, following which the glucose tolerance test, insulin tolerance test, and other experiments were performed. ATP levels in the skeletal muscle were quantified by the ATP Determination Kit from Thermo Fisher Scientific (A22066) by following the manufacturer’s instructions. Briefly, 100 mg of tissue was homogenized in radioimmunoprecipitation assay (RIPA) buffer containing 1 mM EDTA and EGTA (Sigma-Aldrich, R0278-500ml) with protease and phosphatase inhibitors. Homogenized muscle samples were centrifuged at 16,000g at 4°C for 30 min. Supernatants were collected in fresh tubes and proceeded with the assay according to the manufacturer’s instructions. Mitochondrial respiration The OCR was measured as described previously (66). Briefly, myofibers were isolated from the Pax7+/+, Myh7fl/+ (control) and Pax7icre/+, Myh7Δ/fl gastrocnemius and EDL muscles. The OCR was measured in isolated myofibers using an Agilent XF24 analyzer. One hour before the experiment, the isolated myofibers were incubated for 1 hour at 37°C in Seahorse assay medium (Seahorse XF DMEM, Seahorse, Agilent, Santa Clara, USA) with Hepes (pH 7.4) and supplemented with 10 mM glucose,1 mM pyruvate, and 2 mM l-glutamine. Post–1 hour incubation, continuous OCR measurements were recorded, starting at baseline, followed by measurements after the sequential addition of 10 μM oligomycin (Sigma-Aldrich, 75351), 10 μM FCCP (Sigma-Aldrich, C2920), and 6 μM rotenone/antimycin A (Sigma-Aldrich, R8875 and A8674). Basal respiration was measured under normal conditions, and uncoupled respiration was calculated as the difference in OCR before and after oligomycin injection. Maximal respiration was determined as the difference in OCR following FCCP injection. The data were normalized to total protein content of the myofibers. Ex vivo contractility The contractile properties of Pax+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice muscle were measured using a muscle contractility measurement apparatus (iWORX, IX-TA-220 data recorder and FT-302 force transducer) as reported previously (67). Measurements were carried out in female mice at three time points: 2, 6, and 12 months. Briefly, mice were anesthetized, and the TA and gastrocnemius muscles were excised at the distal tendon, keeping the proximal tendon intact. The TA muscle distal tendon was excised first, and its contractility was measured, followed by the gastrocnemius distal tendon excision and contractility measurement. The surgically excised muscle tendon was connected to the force transducer and recorder for contractility measurements. Electric pulses (frequency of 25 to 300 Hz) were delivered by a pair of electrodes to the muscle, and the force-frequency relationships were measured by the recorder. After force measurements, the muscles were removed and weighed. Recorded forces were plotted as absolute force and maximum force in grams. To determine the specific force of contraction of Pax7+/+, Myh7fl/+ (control) and Pax7icre/+, Myh7Δ/fl mice, the absolute force obtained was normalized to the cross-sectional area as previously described (68, 69). Immunofluorescence, histology, microscopy, and image analysis The hindlimb muscles TA, gastrocnemius, EDL, and soleus were dissected out from euthanized mice and embedded in cryomatrix (Polysciences, 19636-1), snap-frozen in liquid nitrogen–cooled 2-methylbutane for 1 to 2 min, transferred to liquid nitrogen, and stored at −80°C. The embedded muscle was cryosectioned at 10 μm in thickness using a cryomicrotome (SLEE MNT). Sections were collected on charged slides (Avantor SuperFrost Plus, 631-0108). Adjacent slides were processed for detecting different proteins. To stain adult MyHCs, slides with muscle tissue sections were thawed at 4°C for 10 min, followed by blocking in 10% goat serum in phosphate-buffered saline (PBS) containing 0.1% NP-40 for MyHC-IIa and MyHC-IIb detection, without any fixation. For MyHC-IIx detection, slides were fixed in chilled acetone for 10 min at 4°C, followed by 2× PBS wash of 5 min each and one wash of 15 min with PBS containing 0.5% Triton X-100, followed by 10 min blocking in 10% goat serum in PBS containing 0.1% NP-40. A primary antibody cocktail of the respective MyHC isoform with laminin was added overnight at 4°C, in a humid chamber. Slides were rinsed twice with ice-cold PBS for 5 min and once with PBS containing 0.5% Triton X-100 for 15 min. Slides were incubated with a biotinylated secondary antibody (for the MyHC isoform primary antibody) and incubated overnight at 4°C, followed by a 2-hour incubation with a fluorophore-tagged secondary antibody (for the laminin primary antibody) and streptavidin Cy2 (for detecting the biotinylated secondary antibody for the MyHC isoform). The sections were again rinsed in PBS, stained with Hoechst (Thermo Fisher Scientific, H3570), fixed in 4% paraformaldehyde, and mounted in Fluoromount-G (Southern Biotech, 0100-01). To stain MyHC-slow, MyHC-embryonic, and troponin T1, slides with muscle tissue sections were thawed at room temperature for 5 min and then fixed in 4% paraformaldehyde. Fixed slides were incubated in citrate buffer (1.8 mM citric acid and 8.2 mM sodium citrate in water) for 10 min at room temperature, followed by antigen retrieval in citrate buffer using an antigen retriever (2100 PickCell Retriever, Aptum Biologics). The sections were then blocked in TNB blocking buffer (PerkinElmer, FP1012) for 1 hour at room temperature and incubated with the respective primary antibodies in the appropriate concentration overnight at 4°C. Appropriate biotinylated secondary antibodies were added to the slides, and the respective signals were amplified using VECTASTAIN (Vector Laboratories, PK-6100) and TSA reagent kits (PerkinElmer, NEL741B001KT and NEL744B001KT). The sections were rinsed in PBS, stained with Hoechst, fixed in 4% paraformaldehyde, and mounted in Fluoromount-G. Muscle sections were imaged using an Olympus BX63F fluorescence microscope with a Hamamatsu ORCA-Flash 4.0 camera. Maximum intensity projections for each image were obtained, and MyHC+ fiber numbers and positive fiber area were quantified using SMASH (semiautomatic muscle analysis using segmentation of histology) (70). Cross-sectional area measurements were performed using the CellSens Dimension 1.16 software, and statistical analysis was carried out using the GraphPad Prism 9. A list of antibodies used is given in table S2. SDH and Oil Red O staining Sectioned slides were thawed at room temperature for 10 min. Slides were incubated in an SDH working solution for 10 to 15 min, depending on the type of muscle (∼10 min for the soleus and 15 min for the EDL/TA). The SDH working solution was prepared as follows: First, modified phosphate buffer (pH 7.6) was prepared containing disodium succinate (SRL, 87578), disodium EDTA (SRL, 4088), and sodium azide (SRL, 17782); 1-methoxyphenazine methosulfate (mPMS) (Sigma-Aldrich, M8640) and nitro blue tetrazolium (NBT) (SRL, 48898) were added to the modified phosphate buffer, fresh on the day of SDH staining. Enzymatic reaction for SDH staining was stopped by rinsing the slides with water. NBT forms nitro blue diformozan, a blue-colored precipitate, with more SDH activity resulting in a darker staining (71). The stained sections were imaged using a Nikon Eclipse Ti microscope equipped with a DS-Fi2 camera. SDH activity was calculated by converting the images into grayscale and measuring the gray intensity of the fibers using the ImageJ software (72). Sectioned slides were thawed at room temperature for 5 min for Oil Red O staining. Slides were fixed in 3.6% paraformaldehyde for 1 hour, followed by a 3× 30-s wash in deionized water. Slides were incubated for 30 min in Oil Red O staining solution containing Oil Red O (3 mg/ml; Sigma-Aldrich, O0625) and 36% triethyl phosphate, followed by 3× 30-s wash in deionized water and rinsed with running tap water for 5 min. Slides were dried and mounted using Fluoromount-G. Quantification was done using the ImageJ software. TEM and image analysis For TEM analysis, the gastrocnemius muscle of 12-month-old Pax7+/+, Myh7fl/+ and Pax7icre/+, Myh7Δ/fl mice were fixed overnight in 2% paraformaldehyde and 2% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4). The tissue samples were postfixed for 1.5 hours with 1% osmium tetroxide in 0.1 M phosphate buffer, stained overnight with 0.5% uranyl acetate, and dehydrated using an ethanol series. The samples were infiltrated with a mixture of propylene oxide and Epon 812 resin for viewing and imaging under the JEM1400 Flash transmission electron microscope (Japan Electron Optics Laboratory) operated at 80 kV at the RCB ATPC Electron Microscopy facility. At least five fields of view of mitochondrial populations were captured for each section. From each biological replicate, at least 100 mitochondria were measured. The CellSens Dimension 1.16 software was used to calculate the mitochondrial size. Western blotting Hindlimb muscles were harvested from mice of the required genotypes at specific time points (P0 and 2, 6, and 12 months) and flash-frozen in liquid nitrogen. Harvested muscles were homogenized in RIPA buffer (Sigma-Aldrich, R0278-500ml) containing 1x protease and phosphatase inhibitors (Cell Signaling Technology, 5872s) and centrifuged at 16,000g for 30 min at 4°C. Supernatants were collected, and total protein estimation was done using a BCA assay kit (Thermo Fisher Scientific, 23225) as per the manufacturer’s protocol. After quantification, an equal amount of protein was separated by SDS–polyacrylamide gel electrophoresis (PAGE), followed by Western blotting, where proteins were transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, iPVH00010). PVDF membranes were blocked using 5% skimmed milk or 5% bovine serum albumin (BSA) (for phosphoforms), followed by overnight incubation in primary antibody at 4°C. Blots were washed thrice in tris-buffered saline (TBS) containing 0.1% Tween 20, 10 min each. This was followed by incubation with horseradish peroxidase (HRP)–conjugated secondary antibodies at room temperature for 2 hours and HRP substrate detection (Millipore, WBLUF0100). The blots were developed using an ImageQuant LAS 4000 (GE), and densitometric quantification was performed using the ImageJ, where protein levels were normalized to α-tubulin levels. To isolate cytoplasmic and nuclear fractions, P0 hindlimb muscle samples were harvested and flash-frozen in liquid nitrogen. Samples were thoroughly washed with cold PBS and homogenized in STM buffer [250 mM sucrose, 50 mM tris-HCl (pH 7.4), and 5 mM MgCl2] containing 1x protease and phosphatase inhibitors. Homogenates were vortexed for 30 s, followed by centrifugation at 8000g for 15 min. The pellet obtained was processed for nuclear fraction, and the supernatant (S0) for the cytoplasmic fraction. Nuclear protein–containing pellets were resuspended in STM buffer and centrifuged at 5000g for 15 min. Obtained pellets were resuspended in NET buffer [20 mM Hepes (pH 7.9), 1.5 mM MgCl2, 0.2 mM EDTA, 20% glycerol, and 1% Triton X-100] containing protease and phosphatase inhibitors and incubated on ice for 30 min. To get the nuclear protein, resuspended lysates were sonicated, followed by centrifugation at 9000g for 30 min, and the supernatant was taken as the nuclear fraction. The cytoplasmic fraction supernatant (S0) was centrifuged at 800g for 10 min. Supernatants were precipitated with chilled acetone at −20°C for 1 to 2 hours and centrifuged for 5 min at 12,000g. The obtained pellet was dissolved in STM buffer and taken as the cytoplasmic fraction. Protein concentrations in the isolated nuclear and cytoplasmic fractions were estimated as above, and Western blots were performed. H3A and GAPDH were used to normalize protein levels in the nuclear and cytoplasmic fractions, respectively (3). The antibodies used for Western blotting were as follows: anti-mouse MyHC-IIa antibody (SC-71, Developmental Studies Hybridoma Bank) at 1:10 dilution; anti-mouse MyHC-IIb antibody (BFF3, Developmental Studies Hybridoma Bank) at 1:10; anti-mouse alpha-tubulin antibody (30021290, Cell Signaling Technology) at 1:10,000; anti-rabbit GAPDH antibody (PAB932Mu01, Cloud-Clone Corp) at 1:10,000; anti-rabbit Akt antibody (4641, Cell Signaling Technology) at 1:1000; anti-rabbit P-Akt S473 (4640, Cell Signaling Technology) at 1:1000; anti-rabbit Histone H3 antibody (Ab1741, Abcam) at 1:10,000; anti-rabbit troponin T1 antibody (NBP238855, Novus Biologicals) at 1:5000; anti-rabbit PGC-1α antibody (A12348, ABclonal; 66369, Proteintech; and 2178S, Cell Signaling Technology) at 1:2000, 1:2000, and 1:1000, respectively; anti-mouse total OXPHOS cocktail (ab110413, Abcam) at 1:1000; anti-rabbit DRP1 antibody (A21968, ABclonal) at 1:2000; anti-rabbit FIS1 antibody (A19666, ABclonal) at 1:2000; anti-rabbit mitofusin1 antibody (A9880, ABclonal) at 1:2000; anti-rabbit mitofusin2 antibody (A12771, ABclonal) at 1:2000; anti-rabbit GLUT4 (MA5-17176, Thermo Fisher Scientific) at 1:1000; anti-rabbit NRF2 antibody (A0674, ABclonal and 98020000, Proteintech) at 1:2000; anti-rabbit P-NRF2 antibody (AP1133, ABclonal) at 1:2000; anti-rabbit SOD1 antibody (A0274, ABclonal) at 1:2000; anti-rabbit Prdx1 antibody (A13524, ABclonal) at 1:2000; anti-rabbit Gpx1 antibody (A1110, ABclonal) at 1:1000; anti-rabbit catalase antibody (A11220, ABclonal) at 1:2000; anti-rabbit KEAP1 antibody (A1820, ABclonal) at 1:1000; anti-rabbit Tom20 antibody (HPA011582, Merck) at 1:2500, anti-rabbit VDAC1 antibody (Ab154856, Abcam) at 1:2500; anti-rabbit pDRP1 (S616) antibody (AP0849, ABclonal) at 1:2500; and anti-rabbit OPA1 antibody (A9833, ABclonal) at 1:2500. A list of antibodies used is given in table S2. Carbonylated proteins Carbonylated proteins were quantified by the OxyBlot Protein Oxidation Detection Kit (Merck Millipore, S7150) following the manufacturer’s instructions. Briefly, 10 μg of the protein lysate was denatured by adding SDS to a final concentration of 6%. Denatured proteins were derivatized by DNPH (2,4-dinitrophenylhydrazine) solution, incubated for 15 to 30 min at room temperature, followed by addition of neutralizing solution and β-mercaptoethanol. These samples were separated by 10% SDS-PAGE followed by Western blotting. Mass spectrometry Hindlimb muscle samples of the Pax+/+, Myh7fl/+ and Pax3cre/+, Myh7Δ/fl genotypes at the P0 stage were harvested and flash-frozen in liquid nitrogen. Harvested muscles were homogenized in RIPA buffer containing 1x protease and phosphatase inhibitors. Homogenized muscle lysates were centrifuged at 16,000g for 30 min, supernatants were collected, and chilled acetone was used to precipitate proteins for 1 hour at −20°C and overnight at −80°C. Precipitated proteins were centrifuged for 45 min at 6000g at 4°C. The obtained pellet was washed two times with chilled acetone, dried, and dissolved in 8 M urea followed by centrifugation at 10,000g for 10 min. The supernatant was taken and quantified by BCA assay. One hundred microgram of protein was taken in a fresh tube, dithiothreitol (DTT) (Sigma-Aldrich, 53819) was added, and the samples were incubated for 1 hour at 56°C. This was followed by the addition of the cysteine blocking agent iodoacetamide (IAA) (Sigma-Aldrich, 15161) and incubation in the dark at room temperature for 1 hour. Samples treated with DTT and IAA were then digested with trypsin (1:20) (Thermo Fisher Scientific, 90057), at 37°C for 18 hours. Subsequently, the digested peptides were dried using a speed vacuum concentrator at 25°C. The dried samples were desalted using Pierce C18 ZipTips (Thermo Fisher Scientific, 87782) and again vacuum dried and reconstituted in buffer A (2% acetonitrile and 0.1% formic acid) for further LC-MS/MS (liquid chromatography–tandem mass spectrometry) analysis. Each sample was run in triplicate on a Triple TOF 5600+ (Sciex, Concord, Canada), equipped with an Eksigent NanoLC-2DPlus system (Eksigent, Dublin, CA, USA) consisting of a ChromXP reversed-phase C18-CL trap column (350 μm by 0.5 mm, 3 μm, and 120 Å, Eksigent) and a nanoViper C18 separation column (75 μm by 250 mm, 3 μm, and 100 Å; Acclaim PepMap, Thermo Fisher Scientific). The MaxQuant software (v1.6, Max Planck Institute of Biochemistry, Germany) was used to identify and quantify peptides and proteins. Intensity data from raw files were used for additional analysis and group comparisons. Statistical analysis was carried out using the Perseus software. The ShinyGO (v7.6) tool was used to determine pathway enrichment using the Reactome database. The PCA and heatmap visualization were done using the MetaboAnalyst, SR plot, and GraphPad Prism. Other graphical representations were made using the SR plot. Statistical analysis Experimental data were analyzed using the GraphPad Prism 9 (GraphPad Software Inc., CA, USA) and are presented as means ± SEM. Statistical significance was determined using a parametric, two-tailed unpaired t test, with *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001 considered significant. All animal experiments include a minimum of four biological replicates per condition, with specific sample sizes detailed in the corresponding figures and legends.