AB-Flu was prepared according to the previous study (Ma et al., 2022). Briefly, human serum albumin (HSA, Sigma-Aldrich, United States) (10 mg) and tris (2-carboxyethyl) phosphine (TCEP) (10 mg) were dissolved separately in 1 mL of PBS buffer to prepare a 10 mg/mL solution, followed by sonication for 10 min. Fluvastatin (Flu, Sigma-Aldrich, United States) (1 mg) was dissolved in 20 µL of DMSO buffer. Subsequently, 500 µL of the HSA solution and 500 µL of the TCEP solution were added to 10 mL of PBS buffer, along with 20 µL of the Flu solution, and the mixture was sonicated for 30 min.

The study strategy is presented in Figure 1. The aims of this strategy are to investigate the individual efficacy of the AB-Flu and the exercise, as well as the synergistical effects of AB-Flu + exercise treatment on the hypoxic TME within B16F10 tumor tissues.

Mouse melanoma B16F10 cell line (sourced from the Chinese Academy of Sciences Cell Bank) was cultured in DMEM medium (Thermo Fisher Scientific, United States) supplemented with 10% fetal bovine serum (FBS). The culture was maintained in an incubator with 5% carbon dioxide at a temperature of 37 °C. B16F10 cells were digested and collected using 0.05% trypsin-EDTA (Thermo Fisher Scientific, United States) to obtain a sufficient number of cells.

Twenty female C57BL/6 mice (4–5-week-old) were purchased from SPF Biotechnology (Beijing, China) and housed under SPF conditions. Mice were housed using standardized breeding methods in a specific pathogen-free environment with a regular light-dark cycle. Mice were randomly divided into four groups, and subsequently, 5 × 10⁵ cells (in 100 μL PBS) were subcutaneously inoculated into the unilateral hindquarters of 4–5-week-old female C57BL/6 mice.

Following the inoculation of B16F10 cells, mice were used to evaluate the synergistic effects of AB-Flu combined with swimming intervention: the control group (Control, n = 5) receiving PBS injections; the monotherapy group (AB-Flu, n = 5) receiving intraperitoneal injections of 2 mg/kg AB-Flu; the moderate-intensity exercise intervention group (ME, n = 5) undergoing weighted swimming training; and the combination therapy group (ME+AB-Flu, n = 5) receiving both weighted swimming and 2 mg/kg AB-Flu injections. Correspondingly, mice in the control group were placed in the same swimming environment, ensuring that water levels did not interfere with their normal activities and respiration. When the average tumor volume of the mice reached approximately 50 mm³, PBS or AB-Flu was administered via intraperitoneal injection every 2 days. Tumor volume (V) was calculated using the formula: V = (width² × length)/2. The tumor growth inhibition (TGI) rate was calculated using the formula: TGI (%) = (1-T_Treatment/T_Control) × 100, where T_Treatment represents the mean tumor volume of the treated group, and T Control represents the mean tumor volume of the control group at the same time point. After 2 weeks, mice were euthanized by manual cervical dislocation, and tumor tissues, major organs as well as the whole blood of mice were removed and collected.

All mouse experiments conducted in this study were approved by the Animal Ethics Committee of the Medical Ethics Committee of Xi’an Jiaotong University and adhered to relevant regulations. The ethical approval code is No. XJTUAE2024-2651.

As the intensity swimming with 5% body weight loading is relatively low, and mainly provides energy through the aerobic metabolism system (Yan et al., 2023). In this study, the weighted swimming as the exercise intervention was performed. Swimming training was conducted in a transparent water tank measuring 30 cm in height and 25 cm in diameter, with a maintained water depth of 20 cm and a temperature of 31 °C ± 1 °C (Figure 2A) (Jones, 2007). Preliminary swimming experiments were performed with various loads. A load equivalent to 5% of the mouse’s body weight was identified as an appropriate and tolerable weight (Figure 2B), which was utilized in subsequent animal experiments.

Lactate concentrations were quantified using a Lactic Acid Assay Kit (Catalog No. A019-2-1, Nanjing Jiancheng Bioengineering Institute, China) following the manufacturer’s instructions. Briefly, 0.1 mL of EDTA-anticoagulated murine whole blood was mixed with 0.6 mL of ice-cold protein precipitant (pre-chilled to 4 °C), vortexed for 30 s, and then centrifuged at 4,000 × g for 10 min at 4 °C to collect the supernatant. The supernatant was incubated with a freshly prepared enzyme working solution (lactate dehydrogenase (LDH)-NAD⁺ system) and chromogenic agent (nitroblue tetrazolium/phenazine methosulfate, NBT/PMS) at 37 °C for 10 min, under light-protected conditions. The reaction was terminated by adding 50 μL of termination solution, and absorbance was immediately measured at 530 nm using a Thermo Scientific™ Multiskan™ GO microplate reader (1 cm light path). Lactate concentrations were calculated using a six-point standard curve (0–20 mmol/L, R ² > 0.99) generated with kit-provided lactate standards.

According to standards established by the American College of Sports Medicine (ACSM) (Garber et al., 2011), sustained blood lactate levels exceeding 4 mmol/L combined with a clearance rate of 0.8–1.2 mmol/L/min are typically indicative of high-intensity interval training (HIIT). In the present study, however, the lactate concentration decreased to 5.41 mmol/L at the 15-min mark, suggesting that the mice predominantly utilized aerobic metabolism—consistent with moderate-intensity exercise. The observed clearance rate of 0.87 mmol/L/min between 10 and 15 min approaches the range associated with HIIT but is not entirely congruent with it. Furthermore, the metabolic threshold for high-intensity exercise in rodents, characterized by peak lactate levels of 8–12 mmol/L (Bangsbo et al., 1994), was not fully reached in this experiment. These observations further justify the selection of the 15-min time point as representative of moderate-intensity exercise.

Following treatment, the tumors and major organs (liver, kidney, lung, spleen, heart) were excised from the mice, fixed in formalin, and subsequently embedded in paraffin before sectioning. The sections were subjected to H&E and IHC stainings. The antibodies employed in this study were as follows: TUNEL (China Biological, China), anti-CD3 (China Biological, China), anti-CD8 (China Biological), anti-CD4 (Santa Cruz, United States), and anti-CD25 (Abcam, United States). Images were captured using a Nikon Eclipse Ni-U microscope (Nikon, Japan) (Yan et al., 2022). Five random areas from each sample were selected to assess the IHC scores. The intensity of immune staining (I) was evaluated by using the numerical score (0–3): 0-unstained, 1-weak staining, 2-moderate staining, and 3-strong staining. The immunostaining area (A) was assessed using numerical score (Bray et al., 2024; Mattiuzzi and Lippi, 2019; Siegel et al., 2025; Lorusso et al., 2017) according to the proportion of positive area: 1%–0%–10%, 2%–10%–50%, 3%–50%–90%, and 4%–90%–100%. Accordingly, the IHC score of each area was calculated as the staining intensity score (I) multiplied by the staining area score (A), i.e., IHC score = I × A (Wang J. et al., 2023).

At the conclusion of the experimental intervention, tumor tissues were collected and processed using the ROS detection kit (Nanjing Jiancheng Bioengineering Institute, China) for the mouse tumor samples. Absorbance was measured using a microplate reader (Thermo Fisher Scientific, United States) with an emission wavelength set at 525 nm and a path length of 1 cm.

MDA is an important biomarker for assessing the level of oxidative stress (Demir et al., 2022). At the end of the experimental intervention, tumor tissues were collected and the oxidative stress status was measured by the MDA detection kit (Beijing Solarbio Science & Technology Co., Ltd., China) from the mouse tumor samples. Absorbance was measured at 532 nm and 600 nm using a microplate reader (Thermo Fisher Scientific, United States), and MDA concentrations for each sample were calculated using the standard curve.

One hour prior to sacrifice, approximately 100 µL of Hypoxyprobe-1 solution (60 mg/kg body weight) was injected intraperitoneally into each mouse. Following sacrifice, tumor tissues were excised and subjected to routine fixation, embedding, and sectioning. Staining analysis was conducted to detect hypoxia within the tumor tissues, and representative images were captured using a Nikon Eclipse Ni-U microscope (Nikon, Japan) along with NIS Elements imaging software.

At the end of the experimental intervention, fresh whole blood was collected from mice, processed into plasma and serum, and analyzed for complete blood counting, liver and kidney functions. Key parameters were included white blood cells (WBC), red blood cells (RBC), platelets, aspartate aminotransferase (AST), albumin (ALB), alanine aminotransferase (ALT), creatinine (Crea), and blood urea nitrogen (BUN). All blood tests were conducted in accordance with the standard clinical trial procedures established by Xi’an Jiaotong University Health Science Center.

All data were statistically analyzed using a two-tailed t-test, with a significance level set at p < 0.05. Results are presented as mean ± standard deviation (mean ± s.d.). Statistical analyses were conducted using GraphPad Prism software (version 9.0, GraphPad Software, United States).

In this study, we utilized swimming with 5% body weight loading as training model, which mainly provides energy through the aerobic metabolism system (Yan et al., 2023; Martha et al., 2024). During continuous swimming, we found three distinct phases of blood lactate dynamics (Figure 2C). In the initial Rapid accumulation phase (0–10 min), lactate level increased from 8.91 to 9.76 mmol/L (a rate: 0.17 mmol/L/min), this was followed by the compensatory clearance phase (10–15 min), during which lactate declined to 5.41 mmol/L (clearance rate: 0.87 mmol/L/min). Finally, in the Metabolic rebound phase (15–25 min), lactate level rebounded to 5.53 mmol/L. Based on these dynamics, the 15-min time point was selected as the exercise intervention period. At this time, lactate concentration fell within the moderate-intensity exercise range (4–8 mmol/L), suggesting that the mice predominantly utilized aerobic metabolism, thereby avoiding excessive fatigue and facilitating effective recovery (Ishihara and Taniguchi, 2018).

To further investigate the antitumor effects of the combination of AB-Flu and exercise intervention, we established a duration of 15 min for the daily exercise intervention in an intact B16F10 syngeneic melanoma mouse model accordingly (Yan et al., 2023). These mice underwent continuous 14-day weighted swimming training, with a load equivalent to 5% of their body weight (Figure 2D). When the tumor volume reached approximately 50 mm³, the mice received intraperitoneal injections of either PBS or AB-Flu (2 mg/kg) every 2 days (n = 5 of each group).

Compared to the control group, the weight (Figure 3A) and volume of tumors (Figure 3B) were significantly reduced in the AB-Flu group, ME group, and ME+AB-Flu group, demonstrating that both AB-Flu treatment and exercise intervention resulted in substantial antitumor effects. Furthermore, the tumor growth curves provide additional support for these findings (Figures 3C–F), with tumor growth inhibition rates (TGI) of 41% and 56% for the Moderate-intensity exercise intervention group (ME group) and the combination therapy group (ME+AB-Flu group), respectively. In contrast, the TGI for the AB-Flu monotherapy group was 30%. These results indicate that the antitumor effects in the combined therapy were superior to those of monotherapy and exercise intervention alone, demonstrating a synergistic enhancement.

The combination of exercise intervention and AB-Flu therapy significantly suppressed tumor growth, highlighting the necessity of investigating internal changes within the tumor. Notably, compared to the AB-Flu monotherapy group, the ME group, and the control group, the tumor tissues in the combination therapy group (ME+AB-Flu) exhibited extensive cell necrosis and a disorganized tissue structure (Figure 4A). Although both the AB-Flu and ME groups demonstrated some degree of tumor cell destruction, the tumor tissues in the ME+AB-Flu group displayed markedly more pronounced differences. TUNEL staining and IHC scoring results indicated a significant increase in the number of positive cells in the ME+AB-Flu group, suggesting that this group exhibited the highest levels of apoptosis (Figures 4B,D). These findings indicate that exercise intervention exerts an inhibitory effect on tumor growth, while the combined use of AB-Flu further enhances this effect.

Additionally, we evaluated the hypoxic conditions in tumor tissues across the groups. The results demonstrated that the AB-Flu group significantly improved tumor hypoxia, and ME group also showed a trend toward improving tumor hypoxia. Importantly, the combination therapy group (ME+AB-Flu) exhibited the lowest levels of hypoxia in tumor tissues, indicating a synergistic effect of AB-Flu and exercise intervention in ameliorating tumor hypoxia. This finding further reinforces the connection between hypoxia alleviation and the antitumor effects of exercise (Figures 4C,E).

Hypoxia is a prevalent phenomenon in most solid tumors. During exercise, increased muscle contraction and energy metabolism lead to increased ROS production (He et al., 2016).

To further investigate the potential mechanisms underlying the enhanced antitumor effects in the ME+AB-Flu group, we assessed ROS levels in excised tumors from the mice through immunofluorescence experiments (Figures 5A,B). As expected, ROS levels in the tumor tissues of the control group were markedly lower than those in the ME and ME+AB-Flu groups. The AB-Flu group exhibited no significant difference in ROS levels compared to the control group but differed from the ME group. These data indicate that in the ME+AB-Flu group, AB-Flu alleviated tumor hypoxia, exercise induced ROS production, which synergistically enhances the antitumor efficacy.

Furthermore, to evaluate the oxidative stress levels, we measured the MDA levels from the tumor samples of each group. The results demonstrated that MDA levels in the ME and ME+AB-Flu groups were significantly lower than those in the control and AB-Flu groups (Figure 5C). This finding indicates that exercise not only inhibits tumor growth, but also alleviates ROS-induced oxidative damage to some extent.

To investigate the biosafety of this exercise combined with AB-Flu treatment, we monitored the body weight changes, and carried out the pathological and hematological assessments of blood and major organs at the end of the study. As shown in Figure 6A, the body weights of mice in all groups exhibited no significant differences, indicating that the combination of AB-Flu and exercise intervention has minimal systemic toxicity.

Further hematological analysis (Figures 6B–I) revealed no significant changes in the routine blood parameters (such as WBC, RBC, platelets, etc.) among different groups, and liver and kidney function markers (such as ALT, AST, ALB, BUN, Crea, etc.) also showed no significant differences, suggesting that this combined treatment did not cause noticeable damage to liver and kidney functions. Although AST and Crea levels were slightly elevated in the combined therapy group, they remained within the normal range and did not cause significant toxic responses.

Additionally, the histological analysis on major organs (heart, liver, spleen, lung, and kidney) revealed no pathological changes, such as inflammation or necrosis, in the major organs of mice across all groups, and the tissue structure remained intact (Figure 7). These findings demonstrate that the combination of AB-Flu and exercise intervention exhibits good biosafety and biocompatibility in vivo.