Thirty-five 8-week-old male ApoE⁻/⁻ mice were procured from the Beijing Weitong Lihua Experimental Animal Company Beijing Biotechnology Co., Ltd. (Beijing, China) and randomly allocated into four groups. The groups were defined as follows: Control (n = 7), which received a standard diet (10% calories from fat); HFD (n = 9), which received a high-fat diet (35% fat calories, 1.5% cholesterol); HFD+MICT (n = 10), which received a high-fat diet along with moderate-intensity continuous exercise (60% maximal speed); HFD+HIIT (n = 9), which received a high-fat diet along with high-intensity intermittent exercise (9 $\times$ 1.5-min sprints at 85% maximal speed with 1-min rest intervals). The HFD contained 83.5% basal diet, 15% lard, and 1.5% cholesterol (Beijing Biotechnology Co., Ltd. (Beijing, China)), yielding approximately 3.94 kcal/g of energy with 35% calories from fat. This formulation mimics the lean hypercholesterolemic phenotype but not diet-induced obesity. The mice were housed in a controlled environment with a temperature maintained between 24–26 °C, humidity at 40–60%, and exposure to a 12-hour light/dark cycle. Each mouse was provided free access to both food and tap water. At the termination of the study, the mice were euthanized by exsanguination under deep anesthesia. Anesthesia was induced and maintained with 1.5% isoflurane (delivered via nose-cone, flow adjusted by weight). Adequate anesthetic depth was confirmed by monitoring respiratory rate, heart rate, and pedal reflexes before proceeding with carotid artery exsanguination. Blood samples were collected from the carotid artery, centrifuged (3000 rpm, 15 min, 4 °C), and stored in serum tubes at –80 °C until analysis. Cardiac tissues were fixed in 10% formalin for subsequent histological evaluation and embedded in paraffin. The remaining cardiac tissue was rapidly frozen in liquid nitrogen for subsequent western blot analysis. All animal procedures adhered to the guidelines stipulated in the “Guide for the Care and Use of Laboratory Animals” and were approved by the Animal Ethics Committee of Affiliated Central Hospital of Dalian University of Technology (approval No. [2023-057-55]). This approval covered a series of related studies investigating exercise interventions in hyperlipidemic ApoE⁻/⁻ mice, including the present work and subsequent mechanistic studies [15].

To determine the maximum running speed, ApoE⁻/⁻ mice in the exercise group underwent a running test on an XR-PT-10B treadmill (Shanghai XinRuan Information Technology Co., Ltd., Shanghai, China). The test commenced at a speed of 10 m/min with an incline of 0° for 20 min, and the speed was gradually increased by 4 m/min until exhaustion was reached. The mice were considered exhausted if they remained stationary on the grid for 3 s or exhibited no movement in response to gentle nudges with a soft brush. The maximum velocity achieved during exercise was defined as maximum running speed (Vmax). All exercise groups underwent a 5-minute warm-up at 40% Vmax before formal training. For the HFD+HIIT group, the exercise regimen involved nine sets of higher-intensity treadmill exercises at 85% Vmax, each lasting 1.5 min, with a 1-min recovery interval at 40% Vmax between each sprint set, totaling 21.5 min per session. No adverse events (injury or death) occurred during the training, confirming the safety of the intensity protocol for this genotype. Conversely, the HFD+MICT group participated in continuous endurance training, maintaining a speed equivalent to 60% Vmax until their running distance matched that of the HFD+HIIT group. This design ensured that any observed differences were attributable to the intensity modality rather than the total work volume. Following formal training, both groups underwent a 5-min recovery period at 40% Vmax. This exercise routine was repeated five times per week for 12 weeks.

Serum lipid (TC and triglyceride [TG]) and B-type natriuretic peptide (BNP) levels were quantified using a commercial assay kit, following the protocols outlined by the manufacturer (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

The procured hearts were fixed in 10% neutral-buffered formalin for 24 h, then dehydrated in graded ethanol (75%, 85%, 90%, and 100%, each for 5 min each), and embedded in paraffin. Sections (4 µm) were cut and mounted on glass slides coated with poly-L-lysine. To prepare the sections for analysis, deparaffinization was achieved by immersion in xylene, with three changes of xylene (5 min each). Rehydration was then performed using reduced concentrations of alcohol (100 to 75% for 5 min each). Hematoxylin and eosin (HE) staining was performed according to standard protocols. Masson’s trichrome staining was used to visualize the collagen deposition (blue) and myocardial architecture (red) to assess interstitial fibrosis. Imaging was conducted using an upright light microscope (Murzider, Beijing, China) at 200$\times$ magnification (objective 20$\times$, numerical aperture 0.45). Scale bars = 20 µm.

Heart tissue sections were fixed in 10% formalin, dehydrated using an ethanol series, and embedded in paraffin for a histological evaluation. After deparaffinization and antigen retrieval (citrate buffer, pH 6.0, 95 °C, 20 min), the sections were blocked with 3% H₂O₂ (15 min) and 5% bovine serum albumin (BSA), then incubated overnight with primary antibodies at 4 °C (all from Proteintech, Wuhan, China). The dilutions were as follows: rabbit anti-CD36 (1:600, Cat# 32371-1-AP), rabbit anti-CD68 (1:500, Cat# 30929-1-AP), rabbit anti-PPAR-$\gamma$ (1:400, Cat# 16643-1-AP), rabbit anti-LOX-1 (1:200, Cat# 11837-1-AP), rabbit anti-interleukin (IL)-1$\beta$ (1:100, Cat# 26048-1-AP), rabbit anti-IL-6 (1:100, Cat# 21865-1-AP), rabbit anti-IL-10 (1:100, Cat# 15102-1-AP), rabbit anti-IL-18 (1:100, Cat# 10663-1-AP), rabbit anti-NRF2 (1:200, Cat# 16396-1-AP), rabbit anti-SIRT1 (1:200, Cat# 13161-1-AP), rabbit anti-SIRT3 (1:200, Cat# 10099-1-AP), rabbit anti-SOD2 (1:200, Cat# 24127-1-AP), rabbit anti-TGF-$\beta$ (1:500, Cat# 26155-1-AP), rabbit anti-SMAD3 (1:200, Cat# 30130-1-AP), rabbit anti-Collagen I (1:500, Cat# 14695-1-AP), and rabbit anti-Collagen III (1:1000, Cat# 22734-1-AP). Subsequently, the sections were incubated for 30 min at room temperature with goat anti-rabbit Horseradish Peroxidase (HRP) secondary antibody (Anti-rabbit Universal Immunohistochemical Detection Kit; Cat# PK10006; Proteintech Group, Inc., Wuhan, China). All sections were examined meticulously using an upright light microscope (Nikon ECLIPSE Ti-U, Nikon Corporation, Tokyo, Japan) for a precise and detailed analysis. Positive cells were quantified as the percentage of total nucleated cells in five random high-power fields (200$\times$ magnification) per section.

Total protein was extracted using Radio Immunoprecipitation Assay (RIPA) buffer (P0013B; Beyotime Biotechnology, Shanghai, China) supplemented with protease/phosphatase inhibitors. The protein concentration was determined using a bicinchoninic assay (Beyotime Biotechnology, Shanghai, China) with bovine serum albumin (BSA) standards and transferred to polyvinylidene fluoride membranes (Immobilon, MilliporeSigma, Billerica, MA, USA). After being blocked with 5% milk/Tris-buffered saline containing 0.1% Tween-20 at room temperature for 1 hour, the membranes were incubated overnight at 4 °C with primary antibodies (all from Proteintech, Wuhan, China) at the following dilutions: rabbit anti-CD36 (Cat# 32371-1-AP), anti-CD68 (Cat# 30329-1-AP), anti-PPAR-$\gamma$ (Cat# 16643-1-AP), anti-LOX-1 (Cat# 11837-1-AP), anti-SIRT1 (Cat# 13161-1-AP) and anti-SOD2 (Cat# 24127-1-AP) at 1:1000; rabbit anti-SIRT3 (Cat# 10099-1-AP) and anti-NRF2 (Cat# 16396-1-AP) at 1:2000; rabbit anti-$\beta$-actin (Cat# 20536-1-AP) at 1:10,000. After being thoroughly washed, the membranes were incubated with the appropriate secondary antibody (anti-rabbit IgG, 1:5000) for 1 h. Bands were quantified using ImageJ software (Version 1.43u, National Institutes of Health, Bethesda, MD, USA), with $\beta$-actin serving as the internal control. The protein levels were expressed as protein/$\beta$-actin ratios to ensure an accurate and comparative analysis.

All data are presented as the mean $\pm$ standard error of the mean (mean $\pm$ SEM). The statistical analyses were performed using GraphPad Prism v9.0 (GraphPad Software, Inc., San Diego, CA, USA). Intergroup differences were assessed using one-way analysis of variance followed by Tukey’s post hoc test (*p 0.05, **p 0.01). All histological results were quantified using Image J software by a color threshold analysis of three random areas per section.

To investigate the effects of different exercise intensities on hyperlipidemia, we established a hyperlipidemia model using a high-cholesterol diet that exhibited a lean hyperlipidemic phenotype. Serological findings indicated notably higher TC and TG in the HFD group, but no significant increase in total body weight or heart weight (Fig. 1), confirming the successful establishment of the familial hypercholesterolemic mouse model in which cardiac damage precedes systemic obesity. This allowed us to isolate direct myocardial lipotoxicity from the confounding hyperlipidemic effects. In contrast, the HIIT+HFD group exhibited significantly reduced TC and TG levels (p 0.05, Fig. 1), suggesting the efficacy of HIIT at effectively lowering these lipid parameters in the mice. Owing to the limitations of the experimental conditions, echocardiography was not performed to evaluate cardiac function, and BNP was selected as a surrogate index to evaluate cardiac wall stress and hemodynamic load. We further assessed the BNP levels to ascertain their potential protective effects against hyperlipidemia-induced myocardial stress at different exercise intensities. As demonstrated in Fig. 1, HFD increased serum BNP levels 2.4-fold, indicating an increased cardiac load. BNP level was reduced by exercise in both groups, but the HIIT group intervention was more effective (p 0.01). These results suggest that HIIT exerts a significant protective effect against hyperlipidemia-induced myocardial stress.

Fig. 1.

Effects of high-intensity interval training (HIIT) on body weight, blood lipids levels and B-type natriuretic peptide (BNP) levels in mice. (A) Body weight. (B) Heart weight. (C) Total cholesterol (TC). (D) Triglyceride (TG). (E) BNP. *p 0.05 versus high-fat diet (HFD); **p 0.01 versus HFD. The data are expressed as mean $\pm$ SEM, n = 5. MICT, moderate-intensity continuous training.

The HE results confirmed HFD-induced inflammatory infiltration and microvesicular steatosis in the myocardial tissue. MICT improved these lesions moderately, whereas HIIT significantly reduced them (Fig. 2). Masson’s trichrome staining revealed no histological abnormalities in the myocardial structures in the control group. However, the mice in the HFD group showed significant myocardial interstitial fibrosis (blue). The exercise interventions reduced fibrosis, with a modest effect of MICT and a more dramatic effect of HIIT.

Fig. 2.

Histological analysis (hematoxylin and eosin [HE] and Masson’s trichrome staining) of the hearts of HFD-fed apolipoprotein E knockout (ApoE⁻/⁻) mice subjected to HIIT. (A) HE-stained tissue showing inflammatory infiltration and microvesicular steatosis in the HFD group. Masson-stained tissue reflecting collagen deposition (blue) and myocardial architecture. 200$\times$ magnification, Scale bar = 20 µm. The arrows indicate areas with positively stained cells. (B) Quantitative analysis of areas with positively stained cells. n = 3 per group. *p 0.05 versus HFD.

These lipid deposition markers collectively capture the full spectrum of lipid handling ranging, from cellular uptake (CD36) and oxidized LDL recognition (LOX-1) to storage regulation (PPAR-$\gamma$) and inflammatory responses (CD68), and their dysregulation leads to myocardial lipotoxicity. Our observations indicated substantially elevated CD36, CD68, PPAR-$\gamma$, and LOX-1 levels in HFD versus control group mice, but HIIT reversed these elevations (all p 0.05 vs. HFD), while MICT showed partial effects (Fig. 3A,B). Changes in protein expression corroborated the immunohistochemical (IHC) data (Fig. 3C,D).

Fig. 3.

HIIT reduced lipid metabolism-related protein expression in HFD-fed ApoE⁻/⁻ mice. (A) Immunohistochemical staining of CD36, CD68 (Cluster of Differentiation 36/38), lectin-type oxidized low-density lipoprotein receptor 1 (LOX-1), and peroxisome proliferator-activated receptor-gamma (PPAR-$\gamma$). The arrows indicate positively stained cells (brown 3,3^′-Diaminobenzidine (DAB) precipitate indicates positivity). 200$\times$ magnification; Scale bar = 20 µm. (B) Bar graph showing quantification of immunohistochemical (IHC) positive cell percentage. (C) Western blotting analysis of CD36, CD68, LOX-1 and PPAR-$\gamma$ protein levels in mice. (D) Bar graph quantifing the lipid metabolism protein. Data are shown as the mean $\pm$ SEM; n = 3 per group; *p 0.05 versus HFD, **p 0.01 versus HFD.

Hyperlipidemia is associated with increased reactive oxygen species (ROS) production; therefore, here we examined the effects of different exercise on oxidative stress markers and antioxidants in mice. These markers collectively reflect the mitochondrial antioxidant defense axis, spanning upstream regulators (SIRT1/3), transcriptional activators (NRF2), and the key effector enzyme (SOD2). Our findings revealed a notable reduction in SIRT1, SIRT3, NRF2, and SOD2 levels in HFD mice versus control mice. However, HIIT attenuated the reduction in SIRT1, SIRT3, NRF2, and SOD2 levels observed in the HFD group, as shown in the HIIT+HFD group (Fig. 4).

Fig. 4.

HIIT reduces antioxidant protein expression in HFD-fed ApoE⁻/⁻ mice. (A) Immunohistochemical staining of sirtuin 1 (SIRT1), sirtuin 3 (SIRT3), superoxide dismutase 2 (SOD2), and nuclear factor erythroid 2-related factor 2 (NRF2). The arrows indicate positively stained cells. Scale bar = 20 µm at 200$\times$magnification. (B) Bar graph quantifying IHC positive cell percentage (SIRT1, SIRT3, SOD2, and NRF2). Data are shown as the mean $\pm$ SEM; n = 3 per group; *p 0.05 versus HFD. (C) Western blot analysis of SIRT1, SIRT3, SOD2, and NRF2 in mice. (D) Bar graph showing the quantification of antioxidant protein expression. Data are shown as the mean $\pm$ SEM; n = 3 per group; *p 0.05 versus HFD.

IHC was performed to detect the expressions of the HFD-induced inflammation-related factors. We observed increases in IL-1$\beta$, IL-6, and IL-18 levels in HFD-induced myocardial stress, but lower IL-10 levels in HFD mice versus control mice (Fig. 5A). While some studies demonstrated that exercise increases IL-10 levels, our finding of reduced IL-10 levels in the exercised groups may reflect intensity-dependent effects or negative feedback from reduced inflammatory stimuli. However, this finding warrants further investigation in future studies. HIIT effectively reduced this elevation. These findings provide compelling evidence of the protective role of HIIT against HFD-induced inflammatory damage (Fig. 5B).

Fig. 5.

HIIT reduces inflammatory marker expression in HFD-fed ApoE⁻/⁻ mice. (A) Immunohistochemical staining for interleukin (IL)-1$\beta$, IL-6, IL-10, and IL-18. The arrows indicate positively stained cells. Scale bar = 20 µm at 200$\times$ magnification. (B) Bar graph quantifying IHC-positive cell percentages. Data are shown as the mean $\pm$ SEM; n = 3 per group; *p 0.05 versus HFD group. **p 0.01 versus HFD.

Markers such as TGF-$\beta$, an upstream cytokine; SMAD3, an intracellular transduction factor; and collagen I/III, the major structural proteins deposited during fibrosis, represent the core of the myocardial fibrosis pathway. The expression levels of SMAD3 and TGF-$\beta$ were increased in HFD mice. HIIT reversed both effects (p 0.01), whereas MICT partially reduced TGF-$\beta$ (p 0.05) but not SMAD3 levels (Fig. 6A). Collagen I (fibrotic) was more responsive to HIIT than collagen III (developmental), suggesting that HIIT preferentially reduces pathological fibrosis (Fig. 6B).

Fig. 6.

HIIT reduces the expression of fibrotic markers in HFD-fed ApoE⁻/⁻ mice. (A) Immunohistochemical staining of collagen I, collagen III, mothers against decapentaplegic homolog 3 (SMAD3), and transforming growth factor-beta (TGF-$\beta$) in the cardiac tissues. The arrows indicate positively stained cells. (B) Bar graph quantifying the IHC positive cell percentage (collagen I, collagen III, SMAD3, and TGF-$\beta$); Scale bar = 20 µm at 200$\times$ magnification. The data are shown as the mean $\pm$ SEM; n = 3 per group; *p 0.05 versus HFD; **p 0.01 versus HFD.

Although BNP can be used as a valid surrogate for cardiac load [34], the lack of echocardiographic assessments is a major limitation. Therefore, our conclusions are limited to molecular and histological correlates of myocardial stress rather than functional outcomes. Due to limited sample resources, comprehensive transcriptional validation was not performed. All mechanistic conclusions in this study are hypothesis-generating rather than definitive. This study only compared different exercise intensities at the molecular level, and in-depth mechanistic studies are required in the future. Our study cannot explain the general superiority of HIIT in myocardial damage in patients with hypercholesterolemia, especially in those with obesity and metabolic syndrome or in the elderly population.