Introduction

Obesity is a global epidemic that poses significant challenges to public health due to its association with an array of metabolic, cardiovascular, and musculoskeletal disorders1,2. While substantial research has identified the systemic consequences of obesity, the impact on cellular homeostasis remains poorly understood. Among the critical processes implicated in obesity is ferroptosis, an iron-dependent form of regulated cell death driven by lipid peroxidation and oxidative stress, which activate pathological processes such as chronic inflammation3, oxidative stress4, and disruption of iron metabolism5. Together, these processes destabilize cellular equilibria, with mounting evidence implicating ferroptosis as a central mechanism underlying obesity-induced cellular dysfunction. An iron-dependent form of regulated cell death driven by lipid peroxidation, ferroptosis is highly responsive to metabolic disturbances and redox imbalances6,7.

In obesity, ferroptosis is exacerbated by key pathological factors, including increased oxidative stress, iron dysregulation, and systemic inflammation, creating a permissive cellular environment for this form of cell death3,5. This process impairs tissue integrity and contributes to comorbidities such as insulin resistance, cardiovascular dysfunction, and tissue-specific degeneration8,9. However, the mechanisms by which ferroptosis is regulated in specific tissues, particularly skeletal muscle, remain poorly characterized. Given its critical role in glucose and lipid metabolism, skeletal muscle dysfunction represents a pivotal contributor to the metabolic disturbances associated with obesity10,11.

Gastrocnemius exhibits heterogeneity in its metabolic and physiological properties, with oxidative (red) and glycolytic (white) muscle displaying distinct responses to metabolic stress12. Red gastrocnemius (RG), characterized by a high mitochondrial content and oxidative metabolism, may exhibit unique ferroptotic vulnerabilities compared with white gastrocnemius (WG), which relies predominantly on glycolysis12. Understanding RG and WG differences could inform the development of targeted interventions aimed at preserving muscle function in patients with obesity.

Emerging evidence suggests that lifestyle interventions, such as dietary restriction and exercise, may attenuate ferroptosis by mitigating oxidative stress and improving iron homeostasis13,14,15,16,17. Although these approaches have demonstrated benefits for metabolic health, their effects on ferroptosis within distinct skeletal muscle types under obesity remain largely unexplored. Addressing this gap is crucial for developing personalized therapeutic strategies that optimize muscle health and systemic metabolic outcomes11.

This study aims to investigate the GW and WG regulation of ferroptosis in an obese mouse model subjected to dietary restriction and exercise interventions. By revealing the differential mechanisms driving ferroptosis in red and white gastrocnemius, we seek to advance our understanding of obesity-induced cellular damage and identify potential therapeutic strategies to mitigate its impact. These findings have the potential to establish a foundation for personalized interventions that preserve skeletal muscle integrity and improve metabolic health.

Methods

Animals and experimental design

Male C57BL/6 N mice (KOATCH, Pyeongtaek, Republic of Korea, 12 months old) were randomly divided into four groups: normal diet (NDC, n = 7), high-fat diet (HF, n = 7), 20% high-fat dietary restriction (HFR, n = 7), and 20% high-fat dietary restriction with voluntary wheel running exercise (HFRV, n = 7). For the first 4 months, all mice were fed either a standard chow diet (Teklad Global 2018, Envigo Inc.) or a high-fat diet (45% kcal from fat, D12451, Research Diets Inc.) (Table 1). In the subsequent 4 months, dietary restriction (DR) and dietary restriction combined with voluntary wheel running exercise (DR + Ex) interventions were applied to the respective groups (Fig. 1). All mice were housed individually to allow precise monitoring of food intake. During the one-week adaptation period prior to intervention, food consumption was recorded weekly for each mouse. A 20% dietary restriction was then applied based on the individual’s baseline intake. Pre-weighed food portions were provided to each animal during the intervention period.

All animals completed the full 8-month protocol with no mortality observed. The mice were housed in a controlled environment, maintained at a temperature of 22 ± 2 °C with 50–60% humidity, and subjected to a 12-h light/dark cycle. All animals had ad libitum access to water and were allowed physical activity. The experimental procedures were approved by the Institutional Animal Care and Use Committee of Hanyang University (HYU 2021–0239 A) and adhered to relevant guidelines and regulations, in compliance with ARRIVE guidelines (https://arriveguidelines.org).

Fig. 1
figure 1

Schematic overview of the experimental study design.

Full size image
Table 1 Diet composition.
Full size table

To evaluate the effects of obesity and subsequent interventions in a structured manner, we adopted a two-phase analysis framework: (1) comparisons between NDC and HF groups to determine obesity-induced ferroptosis in RG and WG; and (2) comparisons among HF, HFR, and HFRV groups to assess the efficacy of interventions on reversing these effects.

Physical phenotypes test

All mice underwent tests before euthanasia. Walking speed, endurance, physical activity, grip strength, total mass, lean mass, fat mass, lean mass ratio, fat mass ratio, and fat mass in tissues were recorded18.

Rota-Rod test: Walking speed was assessed using the Rota-Rod test. The first three days constituted an adaptation period involving a pre-test exercise at 5 rpm for 1 min once each day. The official test was conducted in acceleration mode, in which the speed was increased from 5 to 50 rpm over 5 min (Model 76–0770, Harvard Apparatus Inc., Holliston, MA, USA). The latency period before falling from the device was recorded. Each mouse was tested three times with a 10 min interval between tests, and the best performance of the three trials was used as the outcome measure.

Treadmill test: Endurance capacity was evaluated using a treadmill test. Adaptation took place three days before the official test, with each mouse running at 5 cm/s for 5 min once each day, on a 0° slope. During the official test, the treadmill speed was increased by 1 cm/s every 20 s starting from 5 cm/s, with a 0° slope. The trial ended when the mouse touched the shock pad (set at 0.6 mA) three times.

Voluntary wheel test: Physical activity was measured using the voluntary wheel test. Running distance was recorded using a voluntary wheel (MAN86130, Lafayette Instrument Company, Lafayette, IN, USA), with each wheel rotation corresponding to a distance of 0.4 m. The average running distance over a 5-day period was recorded for each mouse.

Inverted-cling grip test: Grip strength was assessed using the inverted-cling grip test. Adaptation took place over the 3 days prior to the official test, with the adaptive test performed once each day. During the official test, mice were placed in the centre of a wire mesh screen, and a stopwatch was started. The screen was then rotated to an inverted position over 2 s, with the mouse’s head descending first, and held 40–50 cm above a padded surface. The time until the mouse fell was recorded. This measurement was repeated three times with 10 min intervals between tests, and the maximum reading was recorded.

Body Composition: After physical phenotype tests, mice were anaesthetized with ketamine (40 mg/kg) and medetomidine (0.8 mg/kg), and their body composition was analysed using an InAlyzer Dual-energy X-ray Absorptiometry (DEXA) system (Micro Photonics Inc., PA, USA). Measured parameters were total mass (g), lean mass ratio (%), fat mass (g), lean mass (g), fat mass in tissue (%), and fat mass ratio (%).

Preservation and Preparation of skeletal muscle

Mice were deeply anesthetized (40 mg/kg ketamine and 0.8 mg/kg medetomidine) and subsequently euthanized by cervical dislocation by cervical dislocation. After euthanasia, the gastrocnemius (GAS) muscle was carefully removed from the mice. The left GAS was fixed in 10% formalin for 24 h and then dehydrated and embedded in paraffin. Sections were prepared for haematoxylin and eosin (H&E) staining, Prussian blue staining, Masson’s trichrome staining, and immunohistochemistry (IHC). The right GAS was snap-frozen in liquid nitrogen and stored at − 80 °C for subsequent western immunoblot (WB) analyses.

Haematoxylin and Eosin staining

Rehydrated tissue Sect. (5 μm) were stained using ClearView hematoxylin (MA0101010, StatLab, McKinney, TX, USA) and ClearView eosin (MA0101015, StatLab, McKinney, TX, USA). The sections were dehydrated, cleared, and mounted with neutral resin. Images were acquired using a slide scanner (Axio Scan. Z1, Zeiss) at a magnification of 200×. For each mouse, a single section of GAS was chosen at random and photographed to produce 5 images, each covering an area of 0.271 mm219. The average area of these 5 images was used to represent the cross-sectional area (CSA) of that GAS for the mouse. The CSA was automatically analysed using “Cellpose”, an open-source deep learning-based segmentation tool20.

Prussian blue staining

Rehydrated tissue Sect. (5 μm) were stained using a Prussian blue staining kit (G1422, Beijing Solarbio Science & Technology Co., Ltd.). Images were acquired using a slide scanner (Axio Scan. Z1, Zeiss) at a magnification of 200×. For each mouse, a single section of GAS was chosen at random and photographed to produce 5 images, each covering an area of 0.271 mm219. The average of these 5 images was used to represent the iron accumulation in the mouse. ImageJ software was used for the analysis.

Masson’s trichrome staining

Rehydrated tissue Sect. (5 μm) were stained with a Masson’s trichrome staining kit (G1340, Beijing Solarbio Science & Technology Co., Ltd.). Images were captured using a slide scanner (Axio Scan.Z1, Zeiss) at a magnification of 200×. For each mouse, a single section from each GAS was chosen at random and photographed to produce 5 images, each covering an area of 0.271 mm²19. The average of these 5 images was used to represent the extracellular matrix content of the GAS for that mouse. Image analysis was performed in ImageJ.

Western Immunoblotting

The red and white GAS were dissected separately21. The medial-deep region, darker in color and enriched in oxidative fibers, was classified as RG, while the lateral-superficial region, paler and composed primarily of glycolytic fibers, was designated as WG21. This separation method has also been validated by previous studies that the white WG region is composed predominantly of fast glycolytic fibers (97% MyHC IIB, 1.5% IIX/B, and 1.5% IIX), while the RG region contains a mixture of oxidative fibers (22% MyHC IIB, 3% IIX/B, 20% IIX, 42% IIA, and 8% type I)22,23. Although RG and WG were manually separated under a stereomicroscope following established anatomical criteria, some degree of fiber-type overlap at the boundary regions may be unavoidable. This limitation could introduce minor cross-contamination between samples, and future studies using fiber-type–specific markers may help validate separation accuracy.

As previously described18, the RG and WG (Supplementary Fig. 1) was lysed using an EzRIPA Lysis Kit separately (WSE-7420, ATTO). The protein concentration was determined using a Pierce Bicinchoninic Acid Protein Assay Kit (Thermo Scientific, Waltham, MA, USA). Equal amounts of protein samples (40 µg) were separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis and electro-transferred to nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA, USA). The membranes were blocked with 5% non-fat milk dissolved in Tris-buffered saline with Tween-20 (TBST; 10 mM Tris, 150 mM NaCl, and 0.1% Tween-20; pH 7.6) for 1.5 h at room temperature and then incubated with primary antibodies (Table 2) overnight at 4 °C. The membranes were incubated with horseradish peroxidase–conjugated secondary antibodies (Table 2) for 1.5 h at room temperature. The targeted bands were quantified by densitometry in ImageJ (National Institutes of Health, Bethesda, MD, USA., Version 1.54p, URL: https://wsr.imagej.net/distros/win/ij154-win-java8.zip).

Immunohistochemistry (IHC)

Following an IHC protocol with modifications based on prior studies, paraffin-embedded tissue Sect. (5 μm) were deparaffinized and rehydrated24,25. Heat-induced antigen retrieval was performed using a 0.01 mol/L sodium citrate buffer (pH 6.0, C1010, Beijing Solarbio Science & Technology Co., Ltd.) in a microwave oven for 20 min at 100 °C. The sections were then washed twice with phosphate-buffered saline (PBS; 0.137 M NaCl, 0.0027 M KCL, 0.010mM Na2HPO4, and 0.002mM KH2PO4; pH = 7.2) for 5 min each and blocked twice with 3% hydrogen peroxide (4104–4400, DAE JUNG) for 15 min each to remove endogenous peroxidases. Following 3 additional 5 min washes with PBS, the sections were blocked with 100–200 µL of blocking buffer (5% bovine serum albumin in PBS) for 30 min at room temperature. After blocking, the sections were washed 3 times with PBS for 5 min each and incubated with the primary antibody (Table 2) overnight at 4 °C. Each section was then washed 3 times with PBS for 5 min each and incubated with the secondary antibody (Table 2) in a humidified chamber for 60 min at room temperature. Following 3 more 5 min washes with PBS, the sections were incubated with a Pierce DAB Substrate Kit (Thermo Scientific, #34002) for 6 min and washed three 3 with PBS for 5 min each. Finally, the sections were rehydrated and mounted. Images were acquired using a slide scanner (Axio Scan. Z1, Zeiss) at a magnification of 200×. For each mouse, a single section of each GAS was chosen at random and photographed to produce 5 images, each covering an area of 0.271 mm2. The average of these 5 images was used to represent the antibody expression value of that GAS for the mouse. ImageJ (National Institutes of Health, Bethesda, MD, USA., Version 1.54p, URL: https://wsr.imagej.net/distros/win/ij154-win-java8.zip) was used for analysis.

Table 2 Primary and secondary antibodies.
Full size table

Statistical analysis

Statistical analyses were conducted using GraphPad Prism software (version 9). Comparisons between two groups were performed using an unpaired Student’s t-test. For comparisons among 3 groups, a one-way analysis of variance (ANOVA) followed by a least significant difference post-hoc test was employed. To assess the effects of muscle type (white vs. red) and dietary conditions (HF vs. DR vs. DR + Ex), a 2-way ANOVA with Bonferroni’s post-hoc test was applied. Outliers were identified and removed using the Z-score approach prior to conducting statistical analyses to ensure the robustness and reliability of the results. All data are presented as mean ± standard deviation (SD), unless otherwise noted. Statistical significance was set at P < 0.05, with asterisks indicating the level of significance: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Results

Obesity modifies physical phenotypes and regulates ferroptosis in skeletal muscle

In the first phase of this study, we determine whether obesity regulates ferroptosis in skeletal muscle, induced obesity in mice by feeding them a high-fat diet. All data are shown in Fig. 2.

A DEXA analysis revealed that a high-fat diet was associated with increases in body weight (P < 0.0001), lean mass (P = 0.00212), fat mass (P < 0.0001), fat mass ratio (P < 0.0001), and fat mass in tissues (P < 0.0001) and decrease in lean mass ratio (P < 0.0001) (Fig. 2A-H). These indicators are widely used to evaluate and identify mouse models of obesity26,27,28,29. Consistent with impaired motor functions in obese mice30,31, we found that a high-fat diet significantly reduced walking speed (P = 0.0424), grip strength (P = 0.00534), and endurance in mice (P = 0.0217) (Fig. 2I-K). These findings validate the successful establishment of the mouse model.

The histological features of ferroptosis in skeletal muscle include the accumulation of inflammatory cells32 and fibrosis33. H&E staining revealed a significant increase in the CSA of skeletal muscle fibers in the HF group (P = 0.00210) (Fig. 2M and R). Moreover, histological analysis showed pronounced disorganization of the myofiber architecture and an increased infiltration of mononuclear cells, suggesting an ongoing inflammatory response. Additionally, Masson’s trichrome staining revealed a significant increase in collagen fibers in the HF group compared with the NDC group, suggesting a significant increase in skeletal muscle fibrosis (P = 0.0110) (Fig. 2N and R). Ferroptosis is closely associated with lipid peroxidation and abnormal iron accumulation7. As markers of lipid peroxidation, malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) can reflect the extent of ferroptosis in skeletal muscle8,34. IHC analyses showed, in the HF group, elevated levels of 4-HNE (P = 0.0320) and MDA (P = 0.0150) in skeletal muscle compared with the NDC group (Fig. 2O-P and R). Furthermore, Prussian blue staining revealed abnormal iron accumulation in the skeletal muscle of the HF group (P = 0.0210) (Fig. 2Q and R). These data are indicative of obesity leading to skeletal muscle dysfunction and ferroptosis.

Fig. 2
figure 2

Obesity modifies physical phenotypes and further regulates ferroptosis in skeletal muscle. (A) Representative DEXA-scanned images of NDC, HF, HFR, and HFRV mice; skeletal muscle, fat tissue, and bone are shown in blue, red, and white, respectively. (B) Weekly change in body weight (group × weeks): NDC vs. HF (*, 54–72 weeks). (C–H) Quantitative data of final total mass, lean mass, lean mass ratio, fat mass, fat mass ratio, and fat mass in tissue were acquired by DEXA scanning. (I–L) Skeletal muscle function tests. (I) Walking speed test. (J) Strength test. (K) Endurance test. (L) Physical activity test. (M) Skeletal muscle cross-sectional area in NDC and HF. (N) Skeletal muscle collagen area in NDC and HF. (O) 4-HNE expression levels in skeletal muscle. (P) MDA expression levels in skeletal muscle. (Q) Iron percentage in skeletal muscle, and (R): Representative images showing H&E staining, Masson’s trichrome staining, IHC, and Prussian blue staining of GAS. Scale bar = 50 μm. Significant differences are denoted by asterisks: P < 0.05 (*), P < 0.01 (**), P < 0.001 (***), and P < 0.0001 (****). All values are presented as mean ± SD. Normal diet control (NDC, n = 7; 20.5 months old), high-fat diet (HF, n = 7; 20.5 month sold).

Full size image

WG and RG regulation of ferroptosis by p-AMPK Thr172, GPX4 and NCOA4 in obesity

In the first phase, to further elucidate the mechanisms underlying the regulation of ferroptosis, we employed WB analysis in an investigation of the impact of obesity on the expression of representative proteins in 3 ferroptosis-related signalling pathways: amino acid (antioxidant), lipid, and iron metabolism in different muscle types. All data are shown in Figs. 3 and 5L.

In the NDC group, the expression of glutathione peroxidase 4 (GPX4) (P = 0.00734) and nuclear receptor coactivator 4 (NCOA4) (P < 0.001) was significantly higher and that of 4-HNE (P = 0.00267) was lower in WG compared with RG (Fig. 3A and I). In the HF group, the expression of NCOA4 was significantly lower in WG compared with RG (P < 0.0001) (Fig. 3I). Additionally, within the same muscle type, expression of GPX4 (P < 0.001) and NCOA4 (P = 0.0015) was significantly lower in the HF group, and 4-HNE (P < 0.0001) expression was higher in WG compared with the NDC group (Fig. 3A, I, and J). Although obesity did not alter the expression of GPX4 in RG, it significantly increased the expression of NCOA4 (P < 0.0001) and 4-HNE (P = 0.0375) (Fig. 3A, I, and J). In the HF group, in the expression of p-AMPK Thr172 in WG was significantly lower in ND group (Fig. 3E). Obesity can lead to a significant reduction in the ratio of p-AMPK Thr172 and AMPK in RG (Fig. 3F). These findings suggest that obesity may induce ferroptosis by modulating the expression of NCOA4 and p-AMPKThr172 in the RG, as well as GPX4 and p-AMPKThr172 in the WG.

Fig. 3
figure 3

Muscle type–specific regulation of ferroptosis by GPX4 and NOCA4 in high-fat diet-induced obese mice. WB was performed to determine levels of protein expression for ferroptosis signalling pathways: (A): Glutathione peroxidase 4 (GPX4), (B): P53, (C): solute-carrier family 7 member 11 (SLC7A11), (D): AMP-activated protein kinase (AMPK), (E): p-AMPK Thr172, (F): p-AMPK Thr172/AMPK ratio, (G): Ras-related protein Rab-7a (Rab7), (H): telomeric repeat-binding factor 1 (TRF1), (I): nuclear receptor coactivator 4 (NCOA4), (J): 4-hydroxynonenal (4-HNE), and (K): malonaldehyde (MDA). Significant differences are denoted by asterisks: P < 0.05 (*), P < 0.01 (**), P < 0.001 (***) and P < 0.0001 (****). All values are presented as mean ± SD. Normal diet control (NDC, n = 7; 20.5 months old), high-fat diet (HF, n = 7, 20.5 months old).

Full size image

DR and DR + Ex show improvement and regulation in physical phenotypes and ferroptosis in obesity

In the second phase of this study, we identified the regulatory effect of obesity on ferroptosis in skeletal muscle and subsequently investigated the intervention mechanisms of DR and DR + Ex on this process. All data are shown in Figs. 2R and 4.

Results showed significant differences in body weight between the HF and HFRV groups at 70 weeks (P = 0.0435) (Fig. 4A). Compared with the HF group, the HFRV group showed significantly reduced total mass (P = 0.00612), lean mass (P = 0.0171), and fat mass (P = 0.0159) in mice, but no changes in lean mass ratio, fat mass ratio, or fat mass in tissues (Fig. 4B-G). DR and DR + Ex did not appear to improve walking speed, endurance, physical activity, and grip strength (Fig. 4H-K).

Additionally, H&E staining showed that DR and DR + Ex reduced the CSA of skeletal muscle (P = 0.0296) and decreased disorganized myofiber structure and increased mononuclear cell presence, may suggest a lowered inflammatory cell infiltration (Figs. 2R and 4L). Masson’s trichrome staining and Prussian blue staining found no evidence of changes in collagen fibers and iron accumulation in skeletal muscle (Figs. 2R and 4M and P). Last, IHC analysis showed no significant differences in the expression of 4-HNE and MDA among the HF, HFR, and HFRV groups (Figs. 2R and 4N and O). These data suggest that an intervention of DR and DR + Ex has limited effects on improving and regulating physical phenotypes and ferroptosis in obese mice.

Fig. 4
figure 4

DR and DR + Ex show improvement in and regulation of physical phenotypes and ferroptosis in obese mice. (A): Weekly changes in body weight (group × weeks): NDC vs. HF (*, 54–72 weeks). (B–G): Quantitative data of final total mass, lean mass, lean mass ratio, fat mass, fat mass ratio, and fat mass in tissue were acquired by DEXA scanning. (H–K): Skeletal muscle function tests. (H): Walking speed test. (I): Strength test. (J): Endurance test. (K): Physical activity test. (L): Skeletal muscle cross-sectional area in NDC and HF. (M): Skeletal muscle collagen area in NDC and HF. (N): 4-HNE expression levels in skeletal muscle. (O): MDA expression levels in skeletal muscle. (P): Iron percentage in skeletal muscle. Significant differences are denoted by asterisks: P < 0.05 (*), P < 0.01 (**), P < 0.001 (***), and P < 0.0001 (****). All values are presented as mean ± SD. High-fat diet (HF, n = 7; 20.5 months old), 20% high-fat dietary restriction (HFR, n = 7, 20.5–months old), 20% high-fat dietary restriction with voluntary wheel running exercise (HFRV, n = 7; 20.5 months old).

Full size image

RG regulation of ferroptosis by GPX4 through DR + Ex in obese mice

In the second phase, these interventions did not significantly regulate ferroptosis-associated histological features across the entirety of skeletal muscle. However, the regulation of ferroptosis by obesity appears to be specific to muscle type. It is therefore worthwhile to explore the muscle type–specific regulation of ferroptosis by these interventions through different pathways. The results of the WB analysis are shown in Fig. 5.

In the HFRV group, expression of 4-HNE in WG was significantly higher compared with RG (P < 0.0001) (Fig. 5J). In RG, expression of GPX4 was significantly higher in the HFRV group compared with the HF group (P = 0.0471) (Fig. 5A); expression of 4-HNE decreased sequentially in the HF, HFR, and HFRV groups, with significant differences (P < 0.0001 and P = 0.0136) (Fig. 5J). Although the p-AMPK Thr172/AMPK ratio in the WG was significantly elevated in the HFRV group (Fig. 5F), the expression levels of lipid peroxidation markers 4-HNE and MDA remained largely unaltered. This observation highlights the multifaceted regulatory nature of ferroptosis, involving multiple intersecting pathways. These data suggest that the regulation of ferroptosis by our method of intervention in obese mice is muscle type–specific. Specifically, DR + Ex inhibits obesity-induced ferroptosis by increasing the expression of GPX4 in RG, but not in WG.

Fig. 5
figure 5

DR and DR + Ex muscle type–specific regulation of ferroptosis by GPX4 in high-fat diet-induced obese mice. WB was performed to determine the protein expression levels of ferroptosis signalling pathways: (A) Glutathione peroxidase 4 (GPX4), (B) P53, (C) solute-carrier family 7 member 11 (SLC7A11), (D): AMP-activated protein kinase (AMPK), (E): p-AMPK Thr172, (F): p-AMPK Thr172/AMPK ratio, (G) Ras-related protein Rab-7a (Rab7), (H) telomeric repeat-binding factor 1 (TRF1), (I) nuclear receptor coactivator 4 (NCOA4), (J) 4-hydroxynonenal (4-HNE), (K) malonaldehyde (MDA), and (L) The representative image of WB. The quantification (4-HNE and MDA) was performed across all visible bands in the entire lane. Significant differences are denoted by asterisks: P < 0.05 (*), P < 0.01 (**), P < 0.001 (***), and P < 0.0001 (****). All values are presented as mean ± SD. High-fat diet (HF, n = 7; 20.5 months old), 20% high-fat dietary restriction (HFR, n = 7; 20.5 months old), 20% high-fat dietary restriction with voluntary wheel running exercise (HFRV, n = 7; 20.5 months old).

Full size image

Discussion

In recent years, the role of ferroptosis in obesity-induced pathologies has attracted research attention5,35,36,37. This study investigated the mechanisms by which obesity regulates ferroptosis in skeletal muscle and evaluated the potential of DR and combined DR + Ex as therapeutic interventions. By focusing on muscle type–specific responses, our findings highlight the distinct roles of red and WG in ferroptosis regulation, offering novel insights into the interplay among obesity, skeletal muscle health, and cell-death pathways. Due to ethical and legal limitations on obtaining human skeletal muscle samples and applying potentially harmful interventions, such as long-term high-fat diet and dietary restriction, a mouse model was used to enable controlled, mechanistic investigation of ferroptosis-related changes in skeletal muscle.

We used a high-fat diet to induce obesity in middle-aged mice, modelling the sustained obesity typically observed in humans between the ages of 30 and 65 years38. This approach aligns with clinical trends, providing a robust preclinical model to investigate changes in obesity-related skeletal muscle. The experimental design, which employed a 16-week HFD regimen, is consistent with protocols established for inducing obesity in mice39. A DEXA analysis confirmed significant increases in body weight, and fat mass, as well as decreases in lean mass ratios, paralleling human obesity-induced changed in body composition40,41. These changes were accompanied by impaired motor function, including reduced walking speed, strength, and endurance, corroborating prior studies that identified motor decline as a hallmark of obesity42. This validated model serves as a reliable platform for examining ferroptosis and its modulation by DR and DR + Ex. Although physical performance was reduced in obese mice, this may be partly attributable to increased body weight rather than intrinsic muscle weakness. Future studies using body weight–normalized performance tests or ex vivo muscle force measurements would help disentangle the mechanical effects of obesity from direct impairments in muscle function.

Our findings demonstrate that obesity induces ferroptosis in skeletal muscle, characterized by iron accumulation, lipid peroxidation, and inflammation. Prussian blue staining revealed significant accumulation of iron in the skeletal muscle of obese mice, consistent with reports of dysregulated iron metabolism in obesity36. Obesity-associated inflammation increases hepcidin levels, reducing iron absorption and promoting intracellular iron retention43,44. Excess iron catalyses the production of reactive oxygen species via Fenton reactions, leading to lipid peroxidation and ferroptosis6,7. Markers of lipid peroxidation, including 4-HNE and MDA, were elevated in obese mice, further substantiating the role of oxidative stress in triggering ferroptosis45,46.

Inflammatory cell infiltration and fibrosis are both hallmarks of ferroptosis47 and both were observed in obese muscle tissue. The primary driver of fibrosis induced by cellular content release induced is inflammatory cell infiltration. Specifically, ferroptosis-associated release of oxidized lipids and cellular contents, such as high-mobility group protein B1, activates fibroblasts and induces fibrosis48. Elevated expression of 4-HNE and fibrosis-related markers underscores the link between ferroptosis and tissue remodelling in obesity. Together, these findings establish a mechanistic connection among obesity, ferroptosis, and skeletal muscle dysfunction. While HE staining provides morphological evidence suggestive of inflammatory cell infiltration future studies employing immune cell-specific markers (e.g., CD68, CD45) would help to definitively identify the cellular composition of the infiltrates.

Our results highlight the muscle type–specificity of ferroptosis regulation under obesity. RG, with its reliance on oxidative metabolism and higher antioxidant capacity, displayed increased NCOA4 expression, facilitating ferritinophagy and iron release. In contrast, WG exhibited reduced GPX4 expression, likely due to a higher susceptibility to lipid peroxidation and oxidative stress49,50. These findings align with the distinct metabolic profiles of red and WGs and their differential responses to high-fat diets51,52.

No significant changes in MDA levels were observed, despite elevated levels of 4-HNE in both muscle types. This discrepancy may stem from differences in the metabolic pathways and reactivity of these lipid peroxidation products53. The stable metabolism of MDA compared with the accumulation of 4-HNE in tissue could explain these divergent results54,55. Additionally, methodological differences, such as IHC’s sensitivity to local MDA accumulation versus WB’s measurement of overall protein levels, may have contributed to these observations. As anesthesia was administered prior to DEXA and tissue collection, it is possible that short-term anesthetic exposure may have affected redox-sensitive markers such as MDA56,57. While all animals were treated identically, we acknowledge this as a methodological limitation and recommend that future studies assess potential confounding effects of anesthesia when analyzing rapidly fluctuating biomarkers.

While obesity is often associated with reduced physical activity, our activity monitoring data showed no significant difference in voluntary locomotion between high-fat diet and normal diet mice. Thus, the observed ferroptosis-related changes are more likely to reflect obesity-induced metabolic stress rather than inactivity per se. Nevertheless, the indirect effects of obesity on muscle activity and associated signaling remain part of the broader obesity-related physiological landscape. Future studies could apply treadmill-based forced exercise or activity-controlled feeding protocols to further disentangle these variables.

DR and DR + Ex interventions partially mitigated obesity-induced ferroptosis, with effects observed primarily in RG. The upregulation of GPX4 in RG but not WG suggests that RG’s higher oxidative metabolism and antioxidant capacity make it more responsive to interventions58. The increased expression of GPX4 reduced 4-HNE levels in RG, highlighting the efficacy of DR + Ex in modulating ferroptosis pathways. However, the lack of significant changes in WG suggests that its glycolytic nature and lower oxidative stress may require alternative strategies for effective intervention. The levels of AMPK, a key regulator of cellular energy and oxidative stress response, remained stable in RG but decreased in WG following DR + Ex. This differential response likely reflects the distinct metabolic demands of the two muscle types and their adaptation to exercise and dietary interventions. RGs’ ability to maintain AMPK contents under low-intensity exercise may explain their greater antioxidant capacity and ferroptosis resistance compared with WG. Although our results suggest distinct ferroptosis-related molecular profiles between RG and WG, further pathway-level analyses (e.g., transcriptomic or proteomic profiling) are needed to determine whether different signaling mechanisms underlie these changes.

Although exercise combined with dietary restriction effectively reduced ferroptosis-related markers in red skeletal muscle, the lack of significant improvement in muscle function may reflect a time lag between molecular recovery and physiological adaptation. Functional restoration likely requires longer or more intense interventions, as well as contributions from additional pathways beyond ferroptosis.

This study focused exclusively on the GAS muscle, which may not fully represent the effects of DR and DR + Ex on other muscle groups. Further studies should explore a wider range of muscles to validate these findings. Additionally, 8 weeks of intervention may have been insufficient to induce pronounced changes, particularly in aged and obese mice with lower metabolic responsiveness59,60. Longer-term studies and high-intensity interventions could provide further insights into the chronic effects of DR + Ex on ferroptosis and muscle health. Moreover, while we observed significant changes in markers associated with ferroptosis (e.g., GPX4, NCOA4, lipid peroxidation, and iron accumulation), we did not directly confirm ferroptotic cell death using specific ferroptosis inhibitors or ultrastructural analysis (e.g., transmission electron microscopy). To detect ferroptosis more accurately, future research will include direct lipid peroxidation measurements, such as BODIPY staining or LPO determination, where feasible.

In conclusion, this study suggests that obesity influences ferroptosis in distinct regions of the gastrocnemius muscle through both region-specific responses—namely, NCOA4 upregulation in the RG and GPX4 downregulation in the WG—and a common response characterized by reduced AMPK activity in both regions. DR + Ex partially mitigated ferroptosis in RG by upregulating GPX4, whereas its effects on WG were limited. These findings provide novel insights into the molecular mechanisms of ferroptosis in skeletal muscle and provide a basis for targeted therapeutic strategies to address obesity-related muscle dysfunction.