Chronic disruption of body clock may speed up muscle loss in mice.

Male C57BL/6JJcl mice (10 weeks old) were purchased from CLEA Japan, Inc. (Tokyo, Japan). Mice were randomly assigned to LD and ADV groups (n = 16) and housed in individual cages. The group was housed (170 × 350 × 145 mm) with a 120-mm diameter running wheel (SANKO, Osaka, Japan). All mice were kept in light-shielded mouse housing boxes at the room temperature of 25.0 °C ± 1.5 °C with food and water available ad libitum, as described previously (Koike et al., 2024). The mice were entrained to a 12:12-h light-dark cycle with an 8:00–20:00 light period for 2 weeks, followed by 2 weeks in the constant darkness (DD). After DD exposure, mice were returned to the normal light condition (8:00–20:00) followed by the same LD condition (light–dark condition with an 8:00–20:00 light period) or the ADV condition (8-h phase advance once every 4 days) for 64 weeks. Body weight of each mouse was measured once every 2 weeks during these CJL conditions, and relative body weight was calculated by comparing each mouse's weight at the time of purchase (10 weeks old). Food intake was measured by providing a pre-weighted food pellet (CRF-1; Oriental Yeast Co., Ltd., Tokyo, Japan) in the top hopper of each home and weighing the remaining food once every 2 weeks. Criteria for humane endpoints established at the beginning of the study included: exhibiting gait abnormalities or showing weight below 20% of what is expected for the animal. Two mice in the LD group were excluded from analysis, as one exhibited gait abnormalities due to injuries was euthanized (4% isoflurane inhalation followed by cervical dislocation), and the other was found dead. Tissue sampling from LD (n = 14), ADV (n = 16) mice was performed from 9:30 to 17:30 after 64 weeks under the CJL paradigm, when ADV mice were in the same light-dark cycle as LD mice with an 8:00–20:00 light period. Young mice were evaluated separately as a control group. At 10 weeks of age, the mice (n = 30) entrained to the same 12:12-h light-dark cycle protocol as the experimental groups for 2 weeks. The sample size for the young group was calculated based on grip strength data from the LD and ADV groups using a power analysis (α = 0.05, power = 0.8). After this entrainment period, body weight, food intake, and grip strength were measured. For tissue collection, mice were anesthetized with 4% isoflurane inhalation and blood was collected from the heart. Then the mice were decapitated, and the gastrocnemius, soleus, and tibialis anterior muscles were collected. The muscle weights were measured using an electronic analytical balance (ER-180A, A&D Company, Tokyo, Japan). The plantaris muscle shares a similar fiber type composition with the gastrocnemius, making it difficult to separate the two. Therefore, it was collected together as a part of gastrocnemius muscle (Adamovich et al., 2021). Muscle weights were determined as the average of both legs and normalized to body weight at the time of sampling (Shang et al., 2020). After weighing, for RNA-seq analysis, the gastrocnemius muscle was incubated with RNAlater (Thermo Fisher Scientific) for several hours according to the manufacturer's instructions and snap-frozen in liquid nitrogen. For histological analysis, the soleus muscle was fixed in 10% neutral buffered formalin. All experiments were approved by the Experimental Animals Committee, Kyoto Prefectural University of Medicine (approval No. M2022-194-1, M2023-191, M2024-180-1), and were performed in accordance with the institutional guidelines and Guidelines for Proper Conduct of Animal Experiments by the Science Council of Japan.

The behavioral analysis of the mice exposed to the CJL conditions was performed as described in our previous reports (Inokawa et al., 2020; Koike et al., 2024). The wheel-running frequency was measured by counting the number of signals from a magnet sensor (59070-010, Littelfuse Inc., Chicago, IL, United States). Clocklab software (Actimetrics, Wilmette, IL, United States) was used to analyze the behavioral activity in wheel revolutions per 1-min bin collected using CompACT AMS (Muromachi Kikai Co. Ltd., Tokyo, Japan) and ClockLab data collection (Actimetrics, Wilmette, IL, US). The mean activities every 4 weeks were calculated using the daily activities. Relative activity was calculated based on activity levels during the 3 weeks prior to the start of the CJL condition (Koike et al., 2024). Chi-square periodogram analysis was carried out in R using the Rethomics package on 14-day activity segments (Geissmann et al., 2019).

The four-limb grip strength test was performed with the timing of light conditions aligned between the LD and ADV groups, using a digital force meter (GPM-101B; Melquest, Toyama, Japan). Mice gripped a metal net with all four limbs, and their tails were gently pulled backward in a horizontal direction. The maximum grip force exerted when the mice released the net was recorded as the grip strength. Each mouse underwent the test three times, and the average of the three measurements was recorded as the final grip strength (Shang et al., 2020).

The gastrocnemius and soleus muscles were fixed in 10% neutral buffered formalin at room temperature for 2 days. Haematoxylin and eosin (HE) staining and immunohistochemical staining were performed by KAC Co., Ltd. (Kyoto, Japan). To identify MHC co-expressing fibers, immunohistochemical staining for fast and slow MHC was performed on serial sections, and corresponding fibers were identified by positional matching across adjacent sections. Monoclonal anti-fast MHC antibody (M4276, Sigma-Aldrich) and anti-slow MHC antibody (ab234431, Abcam) were used, respectively. Images of all muscle fibers were obtained with a BZ-X710 microscope (Keyence, Osaka, Japan). The CSA of the gastrocnemius and soleus muscles was calculated from the entire HE-stained sections using the BZ-X700 Analyzer (Keyence, Osaka, Japan). In the gastrocnemius muscle, fiber size distribution and mean CSA were quantified by measuring at least 100 fibers from HE-stained sections (Wang et al., 2018). Centrally nucleated fibers were defined as fibers with nuclei not in contact with the sarcolemma (Kay et al., 2022), and their number was normalized to muscle CSA. In the soleus muscle, the number of MHC co-expressing fibers was determined from immunohistochemically stained sections. For each sample, the average of three sections was used for quantification, and values were normalized to muscle CSA. Centrally nucleated fibers were also counted on the same immunostained sections.

Frozen gastrocnemius muscle was homogenized twice for 30 s in TRIzol reagent (Thermo Fischer Scientific) with 5-mm- and 3-mm-diameter stainless beads using a desktop bead Crusher Shakeman 6 (Bio Medical Science Inc., Tokyo, Japan) at 3,500 rpm. Poly (A)-enriched stranded RNA sequencing was carried out by the NGS core facility at the Research Institute for Microbial Diseases of Osaka University on Illumina NovaSeq X with 101-bp paired-end reads. After adaptor sequences were trimmed using Trimmomatic, the sequence reads were mapped to the mouse genome (GRCm39/mm39) using STAR as described previously (Bolger et al., 2014; Dobin et al., 2013; Koike et al., 2024). To obtain reliable alignments, the reads with a mapping quality of less than 10 were removed by SAM tools (Li et al., 2009). The University of California, Santa Cruz (UCSC) known canonical gene set (57,186) were used for annotation, and the reads mapped to the exons were quantified using analyzeRNA.pl script in the Homer software with an option of–rpkm or -noadj for FPKM or raw read counts, respectively (Heinz et al., 2010). Since the Homer treat each half of the read separately and count each as 0.5 reads for paired-end reads, the raw read counts were rounded to the nearest integer before transforming rlog using DESeq2 (Love et al., 2014). To report one isoform per locus (gene symbol), the longest expressed isoform was chosen. We assumed that a gene was expressed if there were more than 20 reads mapped on average in the exons of the gene. The expression level cutoff, average FPKM >0.5, was used for the downstream data analysis. Differentially expressed genes in the RNA-seq data were determined using DESeq2 with thresholds of false discovery rate (FDR) < 0.05 and fold change >1.3. Gene ontology enrichment analysis of differentially gene expressed genes and gene set enrichment analysis was carried out using clusterprofiler in R language (Yu et al., 2012) with a threshold of false discovery rate (FDR) < 0.05. The enriched GO terms with less than 5 gene counts were excluded.

Data are presented as mean ± SD. Statistical analysis was performed by the statistical methods stated in each legend using GraphPad Prism version 10.0 or R software. The significance level was set at p < 0.05 unless otherwise noted.

Wild-type C57BL/6J male mice were exposed to either the normal LD cycles or the previously established circadian misalignment paradigm, the aforementioned ADV conditions (Inokawa et al., 2020). Given that mice aged 18 months or older are commonly used as a natural aging model exhibiting early features of age-related sarcopenia, including reduced grip strength and muscle weight, we subjected 14-week-old mice to 64 weeks under the LD and ADV conditions, when all mice reached approximately 18 months of age (Shang et al., 2020; Xie et al., 2021). Before the initiation of the CJL paradigm, all mice were first entrained to a 12-h/12-h LD cycle for 2 weeks, followed by 2 weeks in constant darkness (DD) to measure baseline free-running activities (Figure 1A). In the ADV group, the circadian period progressively diverged from their intrinsic circadian rhythm with age, indicating that ADV mice underwent chronic circadian misalignment (Supplementary Figure 1). To assess whether these mice developed age-related sarcopenia based on human clinical criteria (e.g., reduced muscle mass and strength compared to younger populations) (Cruz-Jentoft et al., 2019), we evaluated grip strength and muscle weight normalized to body weight relative to young 12-week-old mice. Longitudinal assessments revealed that, after 53-week CJL treatment, ADV mice were impaired in grip strength compared to LD mice (Figure 1B). Furthermore, ADV mice exhibited significant decreases in not only grip strength, but also the normalized weights of soleus, gastrocnemius, and tibialis anterior muscles relative to both young mice and LD mice (Figures 1B,C). On the other hand, LD mice showed significant reductions in the normalized weights of gastrocnemius and tibialis anterior muscles, whereas grip strength and the normalized weight of soleus muscle were not significantly reduced compared to young mice (Figures 1B,C). ADV mice also showed increased body weight and reduced activity levels compared with LD mice, without an increase in food intake (Supplementary Figure 2), consistent with a previous report (Koike et al., 2024). These experiments using the CJL model reveal that environmental circadian misalignment accelerates the progression of sarcopenia.

To investigate the molecular basis for the distinct effects of circadian misalignment on the grip strength and muscle weight, we performed RNA-seq analysis using gastrocnemius muscles from the mice exposed to LD and ADV conditions for 64 weeks. Analysis of differentially expressed genes (DEGs) identified differential expression of 304 genes between LD and ADV groups (Figure 2A; Supplementary Table 1). GO enrichment analysis of the DEGs showed that GO terms including circadian rhythm, rhythmic process, muscle organ development, and skeletal muscle cell differentiation were significantly enriched (Figure 2B; Supplementary Table 2). These enriched GO terms included key myogenic regulators such as Myod1, a master transcription factor involved in myoblast proliferation and differentiation. In addition, myogenesis-related genes such as Myog and Myf6, which promote myotube formation by myoblast fusion (Aziz et al., 2010; Hasty et al., 1993), and Dll1, which is activated by Myod1 and promotes the myogenic process (Kollias et al., 2006; Zhang et al., 2021), were significantly upregulated in the ADV group compared to the LD group (Figure 2C). These findings suggest that circadian misalignment may promote muscle regeneration and remodeling processes.

To further investigate the paradoxical observation of increased expression of myogenic regulators alongside reduced grip strength and muscle weight, gene set enrichment analysis (GSEA) was performed using the Hallmark gene set from the Molecular Signature Database (MSigDB) (Castanza et al., 2023). The most significantly enriched gene set was the TNFα signaling pathway via NF-κB (Figure 3A; Supplementary Table 3), which is interesting because tumor necrosis factor (TNFα) is known to be associated with sarcopenia with its ability to promote protein catabolism, muscle degeneration and muscle atrophy (Sishi and Engelbrecht, 2011; Zhou et al., 2016). Therefore, we next examined the expression of genes in the TNFα signaling pathway. Whereas Tnf expression was not significantly increased in muscle tissues from the ADV group compared to the LD, Tnfsf12, encoding TNF-like Weak Inducer of Apoptosis (TWEAK), was significantly increased in the ADV group, along with its receptor Fn14 encoded by Tnfrsf12a (Figure 3B; Supplementary Figure 3A). This is interesting because TWEAK is a pro-inflammatory cytokine of the TNF superfamily, and recognized as a key mediator of muscle wasting during aging (Dogra et al., 2007; Tajrishi et al., 2014a). In addition, the expressions of Myod1, NF-κB-related transcription factors (Rela, Nfkb1, and Nfkb2), and unfolded protein response-related genes (Atf4 and Ddit3) involved in the repression of protein synthesis were significantly higher under the ADV condition (Figure 2C; Supplementary Figure 3B). These genes are known downstream targets of the TWEAK/Fn14 signaling pathway (Tomaz da Silva et al., 2025; Zhou et al., 2016), suggesting that CJL activates the TWEAK/Fn14 signaling pathway. Collectively, the observed transcriptional changes in our CJL mouse model recapitulates key molecular features known to be associated with sarcopenia.

To investigate the effects of CJL on skeletal muscle morphology, we examined the CSA of gastrocnemius muscle fiber using HE staining. The mean CSA of the muscle fiber was significantly smaller in the ADV group compared to the LD group (Figure 4A), indicating circadian misalignment induces muscle atrophy. These results are consistent with the observed reductions in grip strength and muscle weight, as well as with RNA-seq-based predictions involving activation of the TWEAK/Fn14 signaling pathway, suggesting that our mouse model also recapitulates histopathological features of sarcopenia. To experimentally validate RNA-seq-predicted enhancement of regenerative activity, we next quantified centrally nucleated fibers, a well-established histological marker of regenerated myofibers known to increase following muscle injury or exercise (Meyer, 2018; Narita and Yorifuji, 1999; Soffe et al., 2016; Wernig et al., 1990). The number of centrally nucleated fibers was significantly elevated in the ADV group compared to the LD group (Figure 4B), suggesting an accumulation of regenerating myofibers under CJL. Notably, despite reduced wheel-running activity levels in the ADV group relative to the LD group (Supplementary Figure 2), the observed increase in centrally nucleated fibers indicates that CJL itself promotes muscle regeneration. To further assess increased remodeling process, we quantified MHC co-expressing fibers (Andersen et al., 1999; Snow et al., 2005; Yoshimura et al., 1998). Skeletal muscle is composed of fast and slow muscle fibers, which generally express either fast muscle- or slow muscle-specific MHC (Blaauw et al., 2013). However, some muscle fibers are known to co-express both MHC markers. The presence of fibers co-expressing fast and slow muscle-specific MHC isoforms has been observed in intermediate fiber-type transitions such as the regenerative process or reinnervation following denervation (Jerkovic et al., 1997; Launay et al., 2006; Marini et al., 1991; Snow et al., 2005; Yoshimura et al., 1998). Therefore, we used MHC co-expressing fibers for semiquantitative analysis as a marker of ongoing remodeling. Because soleus muscle is relatively rich in slow twitch (Type I) muscle fibers and is commonly used for histological analysis of MHC markers (Snow et al., 2005), we performed quantitative analysis of MHC co-expressing fibers in soleus muscle. Immunohistochemical staining of muscle tissue was performed using monoclonal anti-fast and anti-slow MHC antibodies, and as expected, most fibers expressed only one MHC isoform (Figure 4C). Importantly, the number of MHC co-expressing fibers, along with centrally nucleated fibers, was significantly increased under the ADV condition compared to the LD group (Figure 4D; Supplementary Figures 4A, B). These results support the notion that chronic circadian misalignment enhances regeneration and remodeling processes including fiber-type transitions in skeletal muscle. However, these processes appear insufficient to maintain muscle homeostasis, which ultimately contributes to progressive muscle atrophy.