Aldehyde Oxidase 1 Deficiency Enhances Aerobic Exercise Performance by Promoting Skeletal Muscle Adaptation and Improving Mitochondrial Function

Abstract

Aerobic exercise has significant health benefits, including preventing chronic diseases like sarcopenia. It strongly depends on muscle fiber types, with higher oxidative fiber ratios enhancing endurance. However, the molecular mechanisms underlying aerobic exercise capacity remain incompletely understood. In this study, we identified 395 genes associated with muscle fiber types, among which 39 were linked to metabolic pathways. Notably, we focused on aldehyde oxidase 1 (AOX1), a molybdenum flavin enzyme, due to its unique non-mitochondrial localization, suggesting a potential causal role in regulating muscle metabolism. We further revealed a significant downregulation of Aox1 mRNA expression in the skeletal muscle of mice after two weeks of exercise training, indicating its involvement in exercise adaptation. To further explore this link, we generated Aox1 knockout (KO) mice and subjected them to endurance capacity tests. Aox1 KO mice exhibited significantly enhanced exercise endurance compared to wild-type (WT) controls, accompanied by a shift toward a more oxidative muscle phenotype, as indicated by an increased proportion of oxidative fibers. Mechanistically, Aox1 KO mice exhibit increased expression of PGC-1α, enhanced mitochondrial function, and increased capillary density in skeletal muscle, facilitating improved oxygen delivery and utilization during exercise. Additionally, in vitro experiments using C2C12 myotubes revealed that Aox1 knockdown alleviated starvation- and TNF-α-induced muscle atrophy, which partially mimics sarcopenia, highlighting its protective role against aging- and stress-induced muscle damage. These findings identify AOX1 as a negative regulator of aerobic exercise capacity and stress resilience, advancing our understanding of skeletal muscle adaptation and highlighting AOX1 as a potential target for improving exercise performance and mitigating sarcopenia.

Keywords: AOX1; aerobic exercise; capillary density; mitochondria; sarcopenia; skeletal muscle.

FIGURE 1

Gene expression and correlation analysis with muscle fiber types. (A) Workflow of transcriptomic data analysis using GEO dataset GSE100505, which includes multiple skeletal muscle types with varying oxidative fiber content (from high to low: Sol, PL, GA, Q, TA, and EDL). Gene expression was normalized by FPKM. Pearson correlations were calculated between each gene and muscle fiber markers Myh4 (glycolytic) and Myh7 (oxidative). Genes with high expression (FPKM > 10), strong correlation (|R| > 0.95), and significant p‐values (p < 0.05) were selected, resulting in 395 genes. Among these, 39 genes involved in metabolic pathways were further analyzed by GO cellular components enrichment and STRING network to identify hub genes related to muscle fiber‐type switching. (B) Heatmap of genes with high expression correlation to glycolytic and oxidative muscle fiber markers Myh4 and Myh7, respectively. Genes were selected based on criteria: FPKM > 10, |R| > 0.95, and p < 0.05, showing robust correlations across multiple muscle types (from high to low oxidative content: Sol, PL, GA, Q, TA, and EDL) (Table S1). (C) KEGG enrichment of the 395 genes identified the top 20 pathways (Table S2). Red highlights indicate metabolic pathways, including 39 metabolism‐related genes listed in Table S2. (D) STRING network analysis identified hub genes with connections greater than four (blue), including Prdx6, Agpat2, Ldhb, and Hmgcr. Aox1 is highlighted in red. (E) Gene Ontology (GO) Cellular Component enrichment analysis was performed on the 39 metabolism‐related genes. Components related to mitochondria are highlighted in red.

FIGURE 2

Exercise training downregulates Aox1 in mice skeletal muscle. (A) Correlation analysis of Aox1 expression across different muscle types, showing its relationship with glycolytic (Myh4) and oxidative (Myh7) muscle markers. (B) AOX1 protein expression levels in tibialis anterior (TA), soleus (Sol), and gastrocnemius (GA) muscles of mice measured by western blot. (C) Schematic representation of the mouse exercise training regimen (top) and specific training protocol (bottom). (D–F) Aox1 mRNA expression levels in GA (D), Sol (E), and TA (F) muscles after two weeks of aerobic exercise training, measured by qPCR. Sample size is n = 5–6 per group. (G) Western blot analysis of AOX1 protein levels in the soleus muscle (Sol) of mice after exercise training. (H) Quantification of AOX1 in (G), (n = 3). (I) Western blot analysis of AOX1 protein levels in the anterior tibial muscle (TA) of mice after exercise training. (J) Quantification of AOX1 in (I), (n = 3). Data are presented as mean ± SD, and statistical significance was assessed using an unpaired two‐tailed t‐test for comparison between groups.

FIGURE 3

Construction and characterization of Aox1 Knockout (KO) mice. (A) Genotyping strategy of Aox1 KO mice. (B) WT and Aox1 ^−/− mice genotyping was detected by PCR. (C) Aox1 mRNA level was quantified by real‐time RT‐PCR. Aox1 mRNA level was normalized to Gapdh mRNA level. (D) Body weight of littermate wild‐type (WT) and Aox1 ^−/− mice at the age of 3 months. n = 9 in each group. (E–G) GA (E), TA (F) and Sol (G) muscles weight in WT and Aox1 ^−/− mice. n = 9 in each group. (H) Glucose tolerance test (GTT), overnight fasted mice were subject to an intraperitoneal injection of glucose (1 g/kg). Data are presented as mean ± SD. Statistical analysis was performed using an unpaired two‐tailed t‐test to compare between two groups.

FIGURE 4

Aox1 knockout enhances aerobic exercise capacity and increases the proportion of oxidized fibers. (A) Schematic representation of the mouse running protocol. (B and C) Running time (B) and distance (C) of Aox1 ^−/− mice (n = 8) and wild‐type controls (n = 7) on a motorized treadmill. (D) Latency to fall in the rotarod test of Aox1 ^−/− mice (n = 8) and wild‐type controls (n = 7). (E) Grip strength of Aox1 ^−/− mice (n = 8) and wild‐type controls (n = 7). (F) Representative images of transverse sections of soleus muscle after staining for MyHC I, IIa, and IIb protein. Scale bar: 50 μm. (G) Quantification of percentage of Type, I and IIa myofibers in soleus muscle (n = 5 for WT, n = 6 for Aox1 ^−/− ). Data are presented as mean ± SD. Statistical analysis was performed using an unpaired two‐tailed t‐test to compare between two groups.

FIGURE 5

Aox1 ablation can enhance mitochondrial function in mouse skeletal muscle by increasing the expression of PGC‐1α. (A) Representative images of Sol muscle sections after staining with SDH. Scale bar: 20 μm. (B) Quantification of SDH staining in (A). (n = 6 for WT, n = 5 for Aox1 ^−/− ). (C) Representative images of GA muscle sections after staining with SDH. Scale bar: 20 μm. (D) Quantification of SDH staining in (C). (n = 7 for WT, n = 6 for Aox1 ^−/− ). (E) Representative Western blot analysis of OxPhos complex using GA muscle lysate. (F) Quantification of OxPhos complex in (E). (n = 5 for WT, n = 6 for Aox1 ^−/− ). (G) The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of GA‐isolated myofibers in WT and Aox1 ^−/− mice were measured by Seahorse XF analyzer. Principal component analysis (PCA) of the two groups data after myofibers normalization. n = 5 mice per group. (H, I) Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) was measured using Seahorse XF analyzer in myofibers isolated from GA. n = 5 mice per group. (J) Representative Western blot analysis of PGC‐1α was performed using GA muscle lysate. (K) Quantification of PGC‐1α in (J). (n = 5 for WT, n = 6 for Aox1 ^−/− ). Data are presented as mean ± SD. Statistical analysis was performed using an unpaired two‐tailed t‐test to compare between two groups.

FIGURE 6

Aox1 knockout enhances capillary density in skeletal muscle. (A) Transverse sections of GA muscle from WT and Aox1 ^−/− mice immune‐stained for CD31 (green) and Dystrophin (red). Scale bar: 50 μm. (B) Quantification of CD31‐positive capillaries per myofiber in GA muscle. n = 5 mice per group. Data are presented as mean ± SD. Statistical analysis was performed using an unpaired two‐tailed t‐test to compare between the two groups.

FIGURE 7

AOX1 knockdown mitigates starvation‐ and TNF‐α‐induced muscle atrophy. (A) Schematic diagram of the experimental design to assess the effects of AOX1 knockdown on muscle atrophy under starvation and TNF‐α treatment. (B) Efficiency of AOX1 knockdown at the mRNA and protein levels in cells transfected with si‐Control, si‐AOX1‐1, and si‐AOX1‐2, as evaluated by qPCR and Western blotting. GAPDH was used as a loading control. (C) Immunofluorescence staining of MYHC (muscle fiber marker) and DAPI (nuclear marker) in C2C12 cells after starvation, showing preserved MYHC levels upon AOX1 knockdown. Scale bar: 50 μm. (D) Quantitative analysis of MYHC‐positive areas in (C), confirming reduced muscle atrophy with AOX1 knockdown under starvation. (E) Immunofluorescence staining of MYHC and DAPI in cells treated with TNF‐α, showing reduced MYHC loss following AOX1 knockdown. Scale bar: 50 μm. (F) Quantitative analysis of MYHC‐positive areas in (E), demonstrating the protective effects of AOX1 knockdown against TNF‐α‐induced muscle atrophy. Data are presented as mean ± SD. Conducting a One‐way ANOVA for statistical analysis of three groups.

FIGURE 8

A model demonstrating enhanced aerobic exercise capacity in Aox1‐deficient mice and reduced muscle atrophy in C2C12 cells. This schematic illustrates the role of Aox1 deficiency in improving aerobic exercise capacity and mitigating muscle atrophy. A two‐week running experiment in wild‐type and Aox1 ^−/− knockout mice showed that Aox1 deficiency led to increased PGC‐1α expression, enhanced oxidative phosphorylation (OXPHOS), improved angiogenesis, a higher proportion of oxidative muscle fibers, and ultimately, better aerobic exercise capacity. In vitro, starvation and TNF‐α treatment induce muscle atrophy in C2C12 cells, characterized by reduced myotube size. Knockdown of Aox1 attenuates these effects, preserving muscle structure and function under atrophic conditions. This model highlights the potential of Aox1 as a regulatory factor in muscle metabolism and exercise adaptation. The figure was created using Figdraw.