This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Multi-Omics Identification of Fos as a Central Regulator in Skeletal Muscle Adaptation to Long-Term Aerobic Exercise
1
Department of Sport Biochemistry, Sport Science School, Beijing Sport University, Beijing 100084, China
2
China Institute of Sport Science, Beijing 100061, China
3
Laboratory of Sports Stress and Adaptation of General Administration of Sport, Beijing Sport University, Beijing 100084, China
*
Author to whom correspondence should be addressed.
Biology 2025, 14(6), 596; https://doi.org/10.3390/biology14060596 (registering DOI)
Submission received: 17 April 2025
/
Revised: 19 May 2025
/
Accepted: 19 May 2025
/
Published: 24 May 2025
(This article belongs to the Special Issue New Insights into Skeletal Muscle Metabolism in Pathological and Physiological Conditions)
Simple Summary
Long-term aerobic exercise improves skeletal muscle health by promoting metabolism and gene regulation. In this study, we analyzed transcriptomic and single-cell data from multiple long-term aerobic exercise models to uncover key genes involved in muscle adaptation. Through machine learning and experimental validation, we identified Fos as a central regulator. Fos was downregulated after long-term exercise, which may promote muscle cell differentiation and reduce inflammation. These findings reveal new insights into how regular aerobic exercise reshapes skeletal muscle at the molecular level.
Abstract
Skeletal muscle health and function are closely linked to long-term aerobic exercise, particularly in enhancing muscle metabolism and regulating gene expression. Regular endurance training can significantly ameliorate metabolic dysfunction and prevent chronic diseases. However, the precise molecular mechanisms underlying skeletal muscle adaptations to long-term aerobic exercise require further clarification. To address this, we integrated transcriptomic and single-cell omics datasets from multiple long-term aerobic exercise models retrieved from the GEO database. After merging and batch correction, differential expression analysis identified 204 DEGs, including 110 upregulated and 94 downregulated genes. Key feature genes were screened using Lasso regression, SVM-RFE, and Random Forest machine learning algorithms, validated by RT-qPCR, and refined through PPI network analysis. Among them, Fos and Tnfrsf12a were significantly downregulated following long-term aerobic exercise. Notably, Fos exhibited a more pronounced decrease than Tnfrsf12a, and was strongly associated with inflammation and muscle regeneration. PPI network analysis indicated that Fos interacted with genes such as Casp3, Egr1, Aft3, Hspa5, Src, and Igf2. GO, KEGG, and GSEA enrichment analyses revealed that Fos is involved in skeletal muscle differentiation, tissue remodeling, and the NF-κB inflammatory pathway. ssGSEA analysis further showed that samples with low Fos expression had significantly elevated Th1/Th2 and Treg cell infiltration. Single-cell analysis confirmed preferential Fos expression in muscle fiber/adipocyte progenitors, satellite cells, and tenocytes, all critical for myogenesis. In summary, our findings suggest that long-term aerobic exercise downregulates Fos, potentially alleviating inflammation and enhancing satellite cell-mediated muscle regeneration. Fos may serve as a central regulator of skeletal muscle remodeling during long-term aerobic exercise.
Keywords:
long-term aerobic exercise; skeletal muscle; Fos; Tnfrsf12a; RNA-seq; scRNA-seq; machine learning
