Targeted Bmal1 restoration in muscle prolongs lifespan with systemic health effects in aging model

Abstract

Disruption of the circadian clock in skeletal muscle worsens local and systemic health, leading to decreased muscle strength, metabolic dysfunction, and aging-like phenotypes. Whole-body knockout mice that lack Bmal1, a key component of the molecular clock, display premature aging. Here, by using adeno-associated viruses, we rescued Bmal1 expression specifically in the skeletal muscle fibers of Bmal1-KO mice and found that this engaged the circadian clock and clock output gene expression, contributing to extended lifespan. Time course phenotypic analyses found that muscle strength, mobility, and glucose tolerance were improved with no effects on muscle mass or fiber size or type. A multiomics approach at 2 ages further determined that restored muscle Bmal1 improved glucose handling pathways while concomitantly reducing lipid and protein metabolic pathways. The improved glucose tolerance and metabolic flexibility resulted in the systemic reduction of inflammatory signatures across peripheral tissues, including liver, lung, and white adipose fat. Together, these findings highlight the critical role of muscle Bmal1 and downstream target genes for skeletal muscle homeostasis with considerable implications for systemic health.

Keywords: Muscle biology; Skeletal muscle.

Figure 1. AAV-mediated Bmal1 expression engages the muscle clock.

(A) Schematic of the mouse generation workflow. Bmal1-KO^+/– mice were used as breeders. Bmal1-KO mice were genotyped at postnatal day 2–3 (P2–P3) and injected with AAV9 at P5. (B) Western blot for anti-HA and anti-BMAL1 detection in gastrocnemius (GAS), heart, and liver tissues from Bmal1-KO+AAV mice, demonstrating the muscle-specific rescue model at 10 weeks of age. (C) Representative images of BMAL1 immunostaining of tibialis anterior (TA) sections costained with dystrophin and DAPI from WT, Bmal1-KO, and Bmal1-KO+AAV mice. Scale bar = 50 μm. (D) Expression of core clock genes and (E) secondary clock and muscle output genes in gastrocnemius muscle (n = 5–6 mice/group). Data are shown as mean ± SD. One-way ANOVA with *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 2. Muscle-specific rescue of Bmal1 increases the survival rate of Bmal1-KO mice.

(A) Kaplan-Meier survival curves of WT (n = 45), KO (n = 43), and KO+AAV (n = 36) mice. Log-rank test, *P < 0.05, ***P < 0.001, ****P < 0.0001. (B) Body weight measurements (n = 16–23 mice/group). Two-way ANOVA, ****P < 0.0001. (C) Body composition analysis (n = 15–23 mice/group). Two-way ANOVA, ****P < 0.0001. (D) Representative locomotor activity traces for individual mice (double plotted for visualization), showing 14 days in LD followed by 15 days in DD. (E) Voluntary wheel activity (n = 12–15 mice/group). One-way ANOVA, ***P < 0.001, ****P < 0.0001. (F) Latency to fall and running speed are increased in KO+AAV mice (n = 6–9 mice/group). One-way ANOVA, *P < 0.05, ***P < 0.001, ****P < 0.0001. (G) Glucose tolerance test and area under the curve (AUC) (n = 6–7 mice/group). One-way ANOVA, *P < 0.05, **P < 0.01. Box plots show the interquartile range, median (line), and minimum and maximum (whiskers).

Figure 3. Functional and histological analysis from muscle-specific rescue of Bmal1 at 10 weeks of age.

(A) Quadriceps weight normalized by body weight across groups (n = 7–11 mice/group). One-way ANOVA, *P < 0.05, **P < 0.01. (B) Representative images of hematoxylin and eosin staining from TA cross sections in WT, KO, and KO+AAV mice. Scale bar = 100 μm. (C) Muscle fiber cross-sectional area (CSA) in TA sections. (D) Muscle fiber type composition in TA cross sections (n = 5–8 mice/group). Two-way ANOVA, ****P < 0.0001. (E) Distribution of muscle fiber CSA (n = 5–8 mice/group). One-way ANOVA, **P < 0.01, ***P < 0.001. (F) Total protein concentrations (μg/mg muscle wet weight) in gastrocnemius muscle samples (n = 8 mice/group). One-way ANOVA, ***P < 0.001. (G) Extensor digitorum longus (EDL) specific force production in WT, Bmal1-KO, and KO+AAV mice (n = 4–6 mice/group). Measurements normalized to WT values. One-way ANOVA, *P < 0.05. Box plots show the interquartile range, median (line), and minimum and maximum (whiskers).

Figure 4. Multiomics analysis of gastrocnemius samples from 10-week-old rescue mice.

(A) Number of differentially expressed genes (DEGs) in each comparison group. (B) Volcano plot showing P value and log₂ fold-change of DEGs in RNA-Seq. Downregulated genes in red, upregulated in blue. (C) Biological processes enriched in DEGs. (D) Expression of genes associated with cell adhesion, amino acid transport, carbohydrate metabolism, and fatty acid metabolism from C (n = 7 samples/group). Expression is in counts per million, normalized to WT levels (dashed line). (E) Overlap analysis of DEGs from A. (F) Correlation of expression changes in 273 overlapping genes between WT versus KO and KO+AAV versus KO. (G and H) GO enrichment analysis for biological processes (G) and cellular compartments (H) enriched in overlapping genes. (I) Number of differentially expressed proteins (DEPs) in each comparison (n = 3 samples/group). (J) Volcano plot of DEPs in KO+AAV vs. KO. Downregulated proteins in red, upregulated in blue. (K) GO enrichment analysis for biological processes and cellular compartments of DEPs in KO+AAV vs. KO. (L) Heatmap analysis of proteins enriched for “Mitochondrion.” Each row represents ion intensity of a protein, with data Z-scaled. Blue indicates high abundance and red low abundance of proteins.

Figure 5. RNA-Seq analysis of gastrocnemius samples from 40-week-old rescue mice.

(A) Number of DEGs in each comparison group (n = 5 samples/group). (B) Volcano plot showing P values and log₂ fold-changes of DEGs in RNA-Seq. Downregulated genes are shown in red, upregulated in blue. (C) Biological processes enriched in DEGs. (D) Heatmap analysis of DEGs enriched for tricarboxylic acid, electron transport chain, and ATP synthesis from C. (E) Overlap analysis of DEGs from A. (F) Correlation of expression changes in 263 overlapping genes between WT vs. KO and KO+AAV vs. KO. (G and H) GO enrichment analysis for biological processes (G) and cellular compartments (H) enriched in overlapping genes. (I) Number of DEGs changing over time (40 vs. 10 weeks) in each group. (J) GO enrichment analysis for biological processes of DEGs down- or upregulated in KO+AAV. The top 10 overlapping processes are plotted based on –log₁₀(P value). (K) GO enrichment analysis for biological processes of DEGs down- or upregulated in KO. The top 10 overlapping processes are plotted based on –log₁₀(P value).

Figure 6. Systemic effects in the rescue model are associated with inflammatory status at 40 weeks of age.

(A) Number of DEGs in liver, lung, heart, and white fat tissues comparing Bmal1-KO vs. KO+AAV at 40 weeks of age (n = 4 samples/group). (B) No overlapping DEGs were found across all tissues. (C) Examples of biological processes related to inflammation or metabolism in each tissue. (D) Partial least squares discriminant analysis of 4,129 metabolomic features. (E) Manhattan plot. The dotted line represents P value < 0.05 comparing Bmal1-KO with KO+AAV plasma samples (n = 5/group). Blue and red dots above the line indicate significantly increased or decreased features, respectively. (F) KEGG pathway analysis of untargeted plasma metabolomics results. (G) Main classes of plasma metabolites. (H) Metabolites related to alanine, aspartate, and glutamate metabolism. (I) Metabolites associated with inflammatory responses (n = 5 samples/group). Two-tailed Student’s, *P < 0.05, **P < 0.01. Box plots show the interquartile range, median (line), and minimum and maximum (whiskers).