Effects of high-intensity aerobic exercise during multiple cycles of doxorubicin treatment on cardiac catabolism

Abstract

[Purpose]

Doxorubicin causes dose-dependent cardiotoxicity characterized by cardiac atrophy. Although moderate-intensity aerobic exercise may mitigate doxorubicin cardiotoxicity, the effects of high-intensity aerobic exercise during doxorubicin treatment on cardiac homeostasis remain unclear. Therefore, this study investigated the impact of high-intensity aerobic exercise (EXE) during doxorubicin (DOX) chemotherapy on cardiac autophagy, protein catabolism, and lipolysis.

[Methods]

Male C57BL/6J mice were randomized into sedentary-saline, exercise-saline, sedentary-doxorubicin, and exercise-doxorubicin groups. Doxorubicin was administered intraperitoneally (5 mg/kg, 5 times at 2-week intervals, totaling 25 mg/kg). The exercise groups performed high-intensity treadmill running (12-15 m/min, 60 min/day, for 7 weeks) during doxorubicin treatment. Body and heart weights were measured, and cardiac tissues were analyzed via Western blotting.

[Results]

Doxorubicin reduced the body and heart weights, and the combination of exercise and doxorubicin further exacerbated these effects. While exercise administered during doxorubicin treatment improved autophagic flux and enhanced lipolysis, it significantly exacerbated cardiac protein catabolism, as evidenced by decreased structural proteins (β-actin, Talin), upregulated ubiquitin-proteasome system components, and increased proteolytic cleavage in the exercise-doxorubicin group compared to the sedentary-doxorubicin group. These findings demonstrate a complex interaction between exercise during doxorubicin treatment, while simultaneously promoting the clearance of damaged components through enhanced lipid utilization and amplifying the degradation of essential cardiac structural proteins.

[Conclusion]

These results suggest that concurrent high-intensity aerobic exercise during cardiotoxic chemotherapy regimens may compromise myocardial integrity.

INTRODUCTION

Doxorubicin (DOX), a prominent anthracycline chemotherapeutic agent, is widely used due to its efficacy against a diverse range of malignancies. However, its clinical application is constrained by the development of dose-dependent cardiotoxicity [1]. Severe adverse effects can lead to irreversible cardiac damage and potentially fatal heart failure. Histopathological investigations of the myocardial tissue following DOX administration have consistently demonstrated characteristic structural abnormalities (pronounced myofibrillar loss, cytoplasmic vacuolization, increased interstitial fibrosis), and other indicators of cardiomyocyte injury and atrophy [2-5]. Furthermore, the reduction in cardiac mass induced by DOX often surpasses the general loss of body weight, suggesting a specific form of cardiac atrophy that is distinct from systemic cachexia [6]. Recognizing the substantial clinical challenge posed by DOX-induced cardiotoxicity, considerable scientific efforts have focused on developing strategies to mitigate myocardial damage while preserving the antitumor effectiveness of DOX [7].

Exercise training has emerged as a promising nonpharmacological strategy for cardioprotective interventions. Exercise is well established for its beneficial effects on cardiovascular health, including the improvement of mitochondrial function, reduction of oxidative stress, and modulation of protein turnover pathways. These mechanisms could counteract DOX-mediated cellular damage [8-10]. Accumulating preclinical and clinical evidence suggests that regular exercise, particularly at moderate intensities when initiated before/concurrently with DOX treatment, may confer protection against its cardiotoxic effects [11-15]. Nevertheless, the specific consequences of high-intensity exercise, and crucially, the optimal timing of exercise relative to chemotherapy cycles, remain unclear.

Despite growing evidence supporting the cardioprotective potential of exercise during chemotherapy, the vast majority of prior investigations have employed low- to moderate-intensity aerobic exercise protocols, which demonstrates important limitations when administered concurrently with DOX. Low-intensity aerobic exercise, which is a feasible adjuvant strategy, has been shown to exert limited effects on key cytoprotective mechanisms within the DOX-treated myocardium, including heat shock protein expression and antioxidant enzyme systems, suggesting that a low-intensity stimulus may be insufficient to engage the signaling pathways necessary for meaningful cardiac protection [16]. Accordingly, a preclinical study employing low-intensity treadmill exercise (10 m/min, 45 min/day) concurrent with DOX treatment in mice demonstrated that this regimen failed to attenuate DOX-induced cardiac wasting/reverse the cardiotoxic effects of the drug [17]. Moderate-intensity exercise has demonstrated more consistent cardioprotective benefits, including attenuation of mitochondrial dysfunction and apoptotic signaling [18], and a systematic review and meta-analysis of preclinical studies confirmed that planned, supervised moderate-intensity aerobic exercise confers superior cardioprotective effects in DOX-treated animals compared to low-intensity voluntary exercise [19]. Nevertheless, even concurrent low-to-moderate-intensity aerobic exercise training during DOX chemotherapy is insufficient to preserve the left ventricular ejection fraction (LVEF)/attenuate myocardial fibrosis [20], highlighting the limitations of lower-intensity regimens as a therapeutic strategy for such cardiotoxicity.

High-intensity exercise represents a fundamentally distinct physiological stressor compared to moderate-intensity exercise and is capable of eliciting substantially greater cellular remodeling and adaptive responses [21,22]. Of particular relevance to the DOX-treated myocardium, high-intensity exercise has been demonstrated to be a more potent activator of the AMPK-ULK1 signaling axis and autophagic flux than low- or moderate-intensity exercise. In well-trained athletes, high-intensity, but not low-intensity exercise robustly activated autophagic flux and the autophagic transcriptional program, including upregulation of LC3B, p62/SQSTM1, GabarapL1, and Cathepsin L mRNA, in a manner dependent on AMPKα activation [23]. Furthermore, high-intensity endurance exercise has been identified as a more potent stimulus than moderate-intensity continuous training for increasing autophagic flux in cardiac and skeletal muscles, and its magnitude appears to depend on the intensity and volume of the exercise stimulus [24]. Given that DOX-induced cardiotoxicity involves pronounced dysregulation of autophagic flux and impaired cellular quality control mechanisms [25], high-intensity exercise was selected in the present study as a strategy hypothesized to engage the AMPK-ULK1-autophagy axis more robustly than previously tested protocols to determine whether an amplified autophagic stimulus could offer superior restoration of cardiac homeostasis during multi-cycle DOX chemotherapy. Moreover, exercise-induced cardiac metabolic adaptations are intensity-dependent; high-intensity interval exercise (HIIE) promotes distinct patterns of cardiac metabolic remodeling, including alterations in glucose and fatty acid oxidation, which are not elicited by moderate-intensity continuous training [26]. DOX is known to profoundly disrupt cardiac-lipid metabolism and fatty-acid oxidation, leading to energy deficits and excessive production of reactive oxygen species (ROS) [27,28]. The use of high-intensity exercise in the present study aimed to examine whether the amplified metabolic stress of this exercise modality would interact with DOX-induced metabolic perturbations in a physiologically and clinically meaningful manner.

High-intensity exercise is also a significant physiological stressor capable of triggering substantial cellular remodeling and adaptive responses [29-31], which can interact uniquely with DOX-induced cellular disturbances. Given the established roles of cellular degradation pathways (autophagy-lysosome system and the ubiquitin-proteasome system [UPS]), in mediating weight loss and muscle atrophy [32-34], and the activation of these systems by DOX [35,36], understanding how high-intensity aerobic exercise influences these processes within the DOX-treated myocardium is important.

Accordingly, this study aimed to investigate the effects of a high-intensity aerobic exercise regimen administered during multiple cycles of DOX administration on indicators of cardiac catabolism and associated cellular processes, including autophagy and lipolysis. By assessing the expression levels of key proteins involved in autophagy (LC3, ATG7, p62, LAMP2, Cathepsin L), mitophagy (BNIP3), their regulatory kinases (AMPK⍺, ULK1), components of the UPS (MuRF1, MAFbx, K48-linked polyubiquitin conjugates), proteasome subunits (Proteasome 26S Subunit, ATPase 1, PSMC1; Proteasome 20S Subunit Alpha 1, PSMA1; Proteasome 20S Subunit Beta type-5, PSMB5), essential structural proteins (β-actin, Talin, α-spectrin II), and markers of lipid metabolism (Acetyl-CoA Carboxylase [ACC], fatty acid synthase [FASN], diglyceride acyltransferase [DGAT], carnitine palmitoyltransferase 1B [CPT1B], very long chain specific Acyl-CoA Dehydrogenase [ACADVL], 3-Hydroxyacyl-CoA dehydrogenase [HADHSC]), we aimed to elucidate the interplay between high-intensity aerobic exercise and DOX-induced cardiac alterations and determine whether this specific training approach provides cardioprotection/potentially exacerbates injury via heightened catabolic processes.

METHODS

Animals

Mature adult male C57BL/6J mice (DBL company, Chungcheongbuk-do, South Korea), aged 12 weeks, were maintained at the Center for Laboratory Animal Sciences in a regulated environment (temperature, 22℃; humidity, 55%; 12-h light/dark cycle), under semi-specific pathogen-free standards. They were provided with unrestricted access to standard chow (#5053, PicoLab® Rodent Diet 20, LabDiet) and distilled water. Following a 1-week adaptation period, the mice were randomly allocated into four experimental groups: sedentary group treated with saline (SED-SAL; n = 7), exercise group treated with saline (EXE-SAL; n = 7), sedentary mice treated with DOX (SED-DOX; n = 7), and exercise group treated with DOX (EXE-DOX; n = 7) (Figure 1A). Body weights were monitored weekly for all mice. All procedures were conducted in accordance with the recommendations outlined in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The Institutional Animal Care and Use Committee of Hanyang University approved all procedures (approval no. 2019-0039A).

Figure 1. High-intensity aerobic exercise during doxorubicin treatment significantly reduces body weight and absolute heart weight.

(A) Schematic representation of the experimental design, illustrating the timing of doxorubicin administration, the duration, and the protocol for exercise. (B-D) Graphs depicting changes in body weight, absolute heart weight, and the ratio of absolute heart weight-to-body weight across experimental groups following doxorubicin treatment with concurrent exercise (n = 7 per group). Asterisks (*) denote statistically significant differences between experimental groups (p < 0.05). EXE, exercise; DOX, doxorubicin.

Experimental Mice Doxorubicin Exposure

Mice in the DOX treatment groups received intraperitoneal injections of DOX hydrochloride (#1515, Sigma-Aldrich, St. Louis, MO, USA) dissolved in 0.9% saline at a dosage of 5 mg/kg. These injections were administered five times on a biweekly schedule, culminating in a total dose of 25 mg/kg [2,17]. Conversely, mice in the non-DOX treatment groups were given equivalent volumes of 0.9% saline via intraperitoneal injection, serving as vehicle controls.

Aerobic Exercise Training

Mice in the exercise training groups commenced with an adaptation phase, during which they ran on a motorized treadmill at speeds of 8-12 m/min (= VO_2max 60-75%) for 30 min daily for five consecutive days. Following this, they undertook the primary exercise phase, running at 12-15 m/min (= VO_2max 75-80%) for 60 min each day for 7 weeks. This exercise protocol was derived from a previous in vivo investigation [37,38]. The relative intensity of exercise on the treadmill was indirectly estimated based on a previous study using the same mouse species [39]. To mitigate the influence of environmental variables such as treadmill noise, mice in the non-exercise groups were positioned in the same training room as the exercise groups during their training sessions. Furthermore, to reduce animal distress, bristle brushes were installed at the end of each treadmill lane as an alternative to electrical shocks.

Experimental Tissue Collection and Storage

Whole cardiac tissues were dissected from all groups by pumping a phosphate-buffered saline solution through each sample to remove blood and then wiping it clear. The left ventricle was specifically isolated and collected. They were frozen in liquid nitrogen and maintained at -80℃ to preserve them for Western blot.

Tissue Homogenization and Western Blot Assay

Heart tissues were mechanically disrupted, homogenized in T-PER^TM Tissue Protein Extraction Reagent (#78510, ThermoFisher Scientific, Waltham, MA, USA), and augmented with a Halt^TM phosphatase and protease inhibitor cocktail (100X) (#1861281, ThermoFisher Scientific, Waltham, MA, USA). The homogenates were centrifuged (14,000 rpm for 15 minutes at 4℃), and the resulting supernatants were collected. Protein concentrations were measured using a Pierce^TM BCA Protein Assay Kit (#A55864, ThermoFisher Scientific, Waltham, MA, USA). The tissue lysates were diluted with distilled water to obtain consistent protein concentrations. These standardized lysates were mixed with 4X Sample Loading Buffer (#B0007, ThermoFisher Scientific, Waltham, MA, USA) and Reducing Reagent (#B0009, ThermoFisher Scientific, Waltham, MA, USA), heated at 70°C for 10 min for protein denaturation, and stored at -80°C until analysis. For protein separation, SDS-PAGE was conducted using Bolt^TM 10%/12% BisTris Plus WedgeWell^TM gels (#NW00107BOX and #NW00125BOX, ThermoFisher Scientific, Waltham, MA, USA), selected according to the molecular weights of the target proteins. Equal quantities of protein were loaded into each well, along with a molecular weight marker as a size reference. Electrophoresis was carried out with Bolt^TM MOPS SDS Running Buffer (#B0001, ThermoFisher Scientific, Waltham, MA, USA), initially at 100 volts for 10 min, and then at 150 volts for 50-55 min. Proteins were transferred onto a membrane using Bolt^TM Transfer Buffer (#BT00061, ThermoFisher Scientific, Waltham, MA, USA) at 30 volts for 60 min. The membrane was stained with Ponceau post-transfer stain to confirm the presence of proteins. The membranes were blocked with a 5% skim milk/bovine serum albumin (BSA) solution to reduce non-specific binding for 60 min. Diluted at 1:1000/1:500 in 5% BSA, primary antibodies were applied to the membrane and incubated overnight at 4°C. The antibodies used in this study were as follows: LC3 (#12741, 1:1000), ATG7 (#8558, 1:1000), p62 (#5114, 1:1000), BNIP3 (#3769, 1:1000), p-AMPK⍺^Thr172 (#2535, 1:1000), AMPK⍺ (#2532, 1:1000), p-ULK1^Ser555 (#5869, 1:1000), ULK1 (#8054, 1:1000), β-actin (#4970, 1:1000), K48-linked Polyubiquitin (#4289, 1:1000), ACC (#3676, 1:1000) from Cell Signaling (Danvers, MA, USA); ⍺-spectrin II (sc-48382, 1:500), MuRF1 (sc-398608, 1:500), MAFbx (sc-166806, 1:500), p-ACC (sc-271965, 1:500), ⍺-spectrin II (sc-48382, 1:500), MuRF1 (sc-398608, 1:500), MAFbx (sc-166806, 1:500), DGAT (sc-271934, 1:500), ACADVL (sc-376239, 1:500), HADHSC (sc-376525, 1:500) from Santa Cruz Biotechnology (Dallas, TX, USA); Cathepsin L (66914-1-lg, 1:1000), PSMA1 (11175-1-AP, 1:1000), PSMB5 (19178-1-AP, 1:1000), PSMC1 (11196-1-AP, 1:1000), FASN (10624-2-AP, 1:1000), CPT1B (22170-1-AP, 1:1000) from Proteintech (Rosemont, IL, USA); LAMP2 (PA1-655, 1:1000) from ThermoFisher Scientific (Waltham, MA, USA); and Talin (ab11188, 1:1000) from Abcam (Cambridge, Cambs, UK). Following this, the membrane was washed with TBS-T and then incubated for 60 min at room temperature with Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody (G21040, ThermoFisher Scientific, Waltham, MA, USA) and Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody (G21234, ThermoFisher Scientific, Waltham, MA, USA), diluted 1:3000 in 5% skim milk/5% BSA. Chemiluminescence was induced by treating the membrane with ECL Selected Western Blotting Detection Reagent (RPN2235; Cytiva, Marlborough, MA, USA) for 30 s. The emitted signals were detected using a ChemiDoc Imaging System (BioRad, Hercules, CA, USA), and protein band intensities were analyzed using Image Lab Software 6.1 (Bio-Rad, Hercules, CA, USA).

Statistical Analyses

All data are presented as the mean ± the standard error of the mean. Statistical comparisons between groups were conducted using a two-way analysis of variance to evaluate the main effects and interactions. Statistical significance was established a priori at p < 0.05. When significant main effects of interactions were detected, Tukey’s honestly significant difference post-hoc test was applied to identify specific group differences.

RESULTS

Reduced Body and Absolute Heart Weight

The DOX-treated groups showed significantly reduced body and absolute heart weights in sedentary mice compared to the SAL-treated groups. Specifically, the SED-DOX group had lower body and heart weights than the SED-SAL and EXE-SAL groups. The combination of exercise and DOX treatment resulted in more significant reductions, with the EXE-DOX group showing lower body and heart weights than the SED-DOX group (p < 0.05; Figure 1B and C). However, no significant differences were observed in the ratio of heart weight to body weight between the groups (Figure 1D).

Improved Autophagy Dynamics

To investigate the potential modulatory effects of high-intensity exercise during DOX treatment, we analyzed the expression of key proteins involved in autophagy. Markers of autophagy initiation and elongation, LC3-II and ATG7, were significantly increased in the DOX-groups compared to those in the SAL-groups (Figure 2A, C, and E; p < 0.05). Although LC3-I showed no change among all groups (Figure 2A and B), the ratio of LC3-II/LC3-I, an indicator of conversion from LC3-I to LC3-II, was increased in the EXE-SAL, SED-DOX, and EXE-DOX groups compared to that in the SED-SAL group (Figure 2A, C, and D; p < 0.05). Furthermore, the level of the autophagy receptor p62, was significantly elevated in the SED-DOX group compared to that in the SED-SAL and EXE-SAL groups (Figure 2A and F; p < 0.05). Interestingly, exercise attenuated this increase as p62 levels were significantly lower in the EXE-DOX group than in the SED-DOX group (Figure 2A and F; p < 0.05). LAMP2, a marker of lysosomal abundance, was increased in the SED-DOX group compared to the EXE-SAL group. However, LAMP2 expression was significantly lower in the EXE-DOX group than that in the SED-SAL, EXESAL, and SED-DOX groups (Figure 2A and G; p < 0.05). Finally, Cathepsin L and BNIP3 levels were upregulated in the EXE-SAL, SED-DOX, and EXE-DOX groups compared to the SED-SAL group (Figure 2A, H, and I; p < 0.05).

Figure 2. High-intensity aerobic exercise mitigates DOX-induced defects in autophagic flux.

(A) Representative Western blot images illustrating the protein expression of key components within the autophagy pathway. Ponceau S staining confirms equal protein loading. (B-I) Quantitative analysis of protein levels for LC3-II, LC3-I, the LC3-II/LC3-I ratio, ATG7, p62, LAMP2, Cathepsin L, and BNIP3. Data are presented as the mean ± SEM (n = 7 per group). Asterisks (*) denote statistically significant differences between experimental groups (p < 0.05). EXE, exercise; DOX, doxorubicin; SED, sedentary; SAL, saline; SEM, standard error of the mean.

The phosphorylation states and total protein levels of critical autophagy regulators, such as AMPKα and ULK1, were investigated to elucidate the molecular underpinnings of these autophagic changes. Phosphorylation of AMPK⍺ at Thr172 (p-AMPK⍺^Thr172), a marker of autophagy activation, was significantly increased across the EXE-SAL, SED-DOX, and EXE-DOX groups compared to the SED-SAL group (Figure 3A and B; p < 0.05). Total AMPK⍺ levels were elevated in the EXE-SAL group relative to the SED-SAL group, consistent with exercise-induced upregulation, yet were reduced in the EXE-DOX group compared to the other SED-SAL, EXE-SAL, and SED-DOX groups (Figure 3A and C; p < 0.05). Nevertheless, the ratio of p-AMPK⍺^Thr172 to total AMPK⍺ was highest in the EXE-DOX group (Figure 3A and D; p < 0.05). Phosphorylation of ULK1 at Ser555 (p-ULK1^Ser555), was significantly elevated in the EXE-DOX group compared to that in the SED-SAL, EXE-SAL, and SED-DOX groups (Figure 3A and E; p < 0.05). Additionally, levels of p-ULK1^Ser555 were higher in the SED-DOX and EXE-SAL groups than in the SED-SAL group, with the SED-DOX group showing higher levels than the EXE-SAL group. Because ULK1 was also increased in the DOX-treated groups, the ratio of p-ULK1^Ser555 to total ULK1 did not change between the groups (Figure 3A, F, and G).

Figure 3. High-intensity aerobic exercise modulates upstream regulators of autophagy.

(A) Representative images of Western blots showing key proteins of autophagy-regulating AMPK/ULK1 signaling pathways. Ponceau S staining confirmed equal protein loading. (B-G) Protein quantification of p-AMPKα^Thr172, AMPKα, p-AMPKα^Thr172/AMPKα ratio, p-ULK1^Ser555, ULK1, and p-ULK1^Ser555/ULK1 ratio. Data are presented as the mean ± SEM (n = 7 per group). Asterisks (*) indicate significant differences between ex perimental groups (p < 0.05). EXE, exercise; DOX, doxorubicin; SED, sedentary; SAL, saline; SEM, standard error of the mean.

High-Intensity Aerobic Exercise with DOX Exacerbates Protein Catabolism

The effect of high-intensity aerobic exercise on cardiac protein catabolism during DOX treatment was evaluated by examining the expression of key cytoskeletal (β-actin), adhesion-related (Talin), and proteolytic (cleaved ⍺-spectrin II) markers. β-actin was significantly reduced in the EXE-DOX group compared to the other groups (SED-SAL, EXE-SAL, and SED-DOX) (Figure 4A and B; p < 0.05). Similarly Talin was decreased in the EXE-SAL group compared to the SED-SAL group, and its levels were also lower in the EXEDOX group than in the SED-DOX and SED-SAL groups (Figure 4A and C; p < 0.05). Conversely, cleaved ⍺-spectrin IIwas significantly increased in the EXE-DOX group compared to the SED-SAL, EXE-SAL, and SED-DOX groups (Figure 4A and D; p < 0.05). The muscle-specific E3 ubiquitin ligase MuRF1 was upregulated in all experimental groups (EXE-SAL, SED-DOX, and EXE-DOX) compared to that in the SED-SAL group (Figure 4A and E; p < 0.05). Furthermore, MuRF1 expression was significantly higher in the EXE-DOX group than in the EXE-SAL and SED-DOX groups. The expression of MAFbx was higher in the EXE-DOX group compared to the EXE-SAL group (Figure 4A and F; p < 0.05). K48-linked polyubiquitin was increased in all intervention groups (EXE-SAL, SED-DOX, and EXE-DOX) compared to the SED-SAL group. Notably, the EXEDOX group exhibited significantly higher levels of K48-linked polyubiquitinated proteins than the EXE-SAL and SED-DOX groups (Figure 4A and G; p < 0.05). Finally, the expression of several proteasome subunits (PSMC1, PSMA1, and PSMB5), was higher in the DOX-treated groups (SED-DOX and EXE-DOX) than in the saline-treated groups (SED-SAL and EXE-SAL) (Figure 4A, H, I, and J; p < 0.05).

Figure 4. High-intensity aerobic exercise accelerates DOX-induced extracellular matrix and cardiac protein catabolism.

(A) Representative Western blot images displaying the protein expression of key markers related to the extracellular matrix and the ubiquitin-proteasome system (UPS) pathway. (B-D) Quantitative analysis of protein levels for β-actin, Talin, and Cleaved α-spectrin II. (E-J) Quantitative analysis of protein levels for MuRF1, MAFbx, K48-linked polyubiquitin, PSMC1, PSMA1, and PSMB5. Data are presented as the mean ± SEM (n = 7 per group). Asterisks (*) indicate statistically significant differences between experimental groups (p < 0.05). EXE, exercise; DOX, doxorubicin; SED, sedentary; SAL, saline; SEM, standard error of the mean.

High-Intensity Aerobic Exercise during DOX Treatment Improves Lipolysis

DOX is a known disruptor of lipid metabolism [27,28]. Moreover, alterations promoting increased fatty acid oxidation can affect heart size [40]. As our data demonstrated decreased body and heart weights with high-intensity aerobic exercise during DOX treatment, we analyzed the key markers of lipolysis to investigate the effects of high-intensity aerobic exercise on lipid metabolism during DOX treatment. Phosphorylated Acetyl-CoA Carboxylase (p-ACC), an indicator of reduced fatty acid synthesis and increased fatty acid oxidation, was increased in the DOX-treated groups compared to the SED-SAL groups (Figure 5A and B; p < 0.05). Total ACC and the ratio of p-ACC to total ACC showed no statistical differences between the groups (Figure 5A, C, and D). However, FASN, a key enzyme in de novo lipogenesis, showed no significant differences among the experimental groups (Figure 5A and E). DGAT, an enzyme involved in triglyceride synthesis, was decreased in the EXE-SAL, SED-DOX, and EXE-DOX groups compared to that in the SED-SAL group (Figure 5A and F; p < 0.05). CPT1B, the rate-limiting enzyme for the transport of long-chain fatty acids into the mitochondria for β-oxidation, showed no significant changes across groups (Figure 5A and G). ACADVL, a mitochondrial enzyme involved in the β-oxidation of very long-chain fatty acids, was increased in the SED-DOX group compared to saline-treated SED-SAL and EXE-SAL groups. ACADVL expression was higher in the EXE-DOX group than in all other experimental groups (SED-SAL, EXE-SAL, and SED-DOX) (Figure 5A and H; p < 0.05). HADHSC, another enzyme involved in fatty acid β-oxidation, was increased in all experimental groups (EXE-SAL, SED-DOX, and EXE-DOX) compared to the SED-SAL group (Figure 5A and I; p < 0.05).

Figure 5. High-intensity aerobic exercise between multiple cycles of DOX treatment improves lipid metabolism.

(A) Representative Western blot images showing the protein expression of key enzymes involved in lipolysis. (B-I) Quantitative analysis of protein levels for p-ACC, total ACC, the p-ACC/ACC ratio, FASN, DGAT, CPT1B, ACADVL, and HADHSC. Data are presented as the mean ± SEM (n = 7 per group). Asterisks (*) denote statistically significant differences between experimental groups (p < 0.05). EXE, exercise; DOX, doxorubicin; SED, sedentary; SAL, saline; ACC, Acetyl-CoA Carboxylase; FASN, fatty acid synthase; DGAT, diglyceride acyltransferase; CPT1B, carnitine palmitoyltransferase 1B; ACADVL, Acyl-CoA Dehydrogenase Very Long Chain; HADHSC, 3-Hydroxyacyl-CoA dehydrogenase.

DISCUSSION

This study investigated the specific impact of a high-intensity aerobic exercise regimen administered during multiple cycles of DOX treatment. Our findings revealed a complex interplay between DOX-induced cardiotoxicity and aerobic training, demonstrating that, while this exercise protocol may enhance certain aspects of cellular homeostasis, it can concurrently exacerbate protein degradation in the heart under these conditions.

Administration of DOX consistently leads to significant weight loss in the body and heart in animal models [2,17,41]. The decrease in heart weight does not solely reflect the overall body weight loss. Linear regression analyses indicated that heart weight indexed to body weight decreased in a dose-dependent manner, suggesting that DOX induced cardiac atrophy was not proportional to general cachexia [6]. Consistent with a previous study, DOX treatment with the SED-DOX group led to a significant reduction in both body and absolute heart weights compared to saline-treated controls. Interestingly, the addition of high-intensity exercise during DOX treatment (EXE-DOX group) further exacerbated this reduction, suggesting a synergistic effect of DOX and high-intensity exercise on overall and cardiac mass. The lack of change in the heart weight-to-body weight ratio across all groups indicated that the reduction in heart weight was proportional to the overall body weight loss, implying a systemic rather than solely cardiac-specific effect on tissue wasting.

The present findings are significant when considered in the context of prior studies employing lower-intensity exercise protocols in DOX-treated animals. Low-intensity exercise training administered concurrently with DOX treatment has been reported to attenuate left ventricular (LV) dysfunction and reduce apoptotic signaling, potentially through the upregulation of glutathione peroxidase activity, suggesting a modest cardioprotective role at lower exercise intensities [42]. However, these protective effects were achieved with lower physiological stimuli, and were not accompanied by the degree of autophagic flux enhancement observed in the present study. Notably, concurrent low-to-moderate intensity aerobic exercise training (40-50% maximal exercise capacity) during DOX treatment is insufficient to preserve LVEF/attenuate myocardial fibrosis, despite preserving myocardial circumferential strain and improving exercise tolerance [20]. This finding highlights a critical limitation of low-intensity regimens: while they may partially maintain certain structural/functional indices, they appear to be inadequate for restoring comprehensive cardiac homeostasis during active cardiotoxic chemotherapy. In contrast, moderate-intensity aerobic exercise administered prior to DOX exposure has demonstrated more robust cardioprotection, including attenuation of cardiac inflammation and fibrosis, and restoration of LVEF and fractional shortening (FS) [11], and corroborated by a systematic review and meta-analysis confirming that exercise training before DOX exposure yields superior FS improvement (mean difference [MD] = 8.20%; 95% confidence interval [CI], 6.27-10.13) compared to concurrent administration (MD = 4.94%; 95% CI, 3.24-6.65) [19]. Collectively, these differences underscore the nuanced dose-response relationship between exercise stimulus magnitude and cardioprotective efficacy in DOX chemotherapy. The present study extends this understanding by demonstrating that amplifying the exercise stimulus to high-intensity levels during concurrent DOX treatment does not linearly enhance cardioprotection; rather, it elicits a paradoxical outcome wherein improvements in autophagic flux and lipid metabolism are simultaneously accompanied by the exacerbation of cardiac protein catabolism via the UPS. This observation suggests that there may be a threshold for exercise-induced stress, beyond which the adaptive capacity of the DOX-treated myocardium is overwhelmed, resulting in net structural protein degradation.

The current study focused on protein-level analysis of cardiac catabolic and autophagic pathways and did not include direct assessments of cardiac function, such as echocardiographic measurements. Previous investigations have demonstrated that DOX treatment causes significant impairment in LV systolic function, as evidenced by reductions in LVEF and FS, which can persist and progress over time in the absence of cardioprotective interventions [43]. Exercise training, particularly when initiated before/during DOX exposure, attenuates these functional deficits. Lee et al. (2020) demonstrated that endurance exercise administered after DOX treatment significantly prevented DOX-induced apoptosis and enhanced autophagic flux, including increased LC3-II levels, reduced p62, and increased autophagosome-containing mitochondria, without altering canonical autophagy regulators (AMPK/mTOR), suggesting that exercise-induced autophagy can be activated through alternative signaling pathways [44]. Furthermore, a meta-analysis of preclinical studies reported that exercise training improved FS by a mean difference of 7.40% (95% CI 5.75-9.05, p < 0.001) compared to sedentary DOX-treated animals [19]. A recent study employing high-intensity interval training (HIIT) in a rat model of DOX-induced cardiotoxicity demonstrated that HIIT significantly improved LVEF and FS, reduced myocardial fibrosis, and normalized electrocardiographic parameters, including QRS and QT intervals [45]. Similarly, a pilot study investigating acute HIIE in mice showed that a single HIIE session performed 3 days after DOX injection did not exacerbate acute cardiotoxicity and was associated with the smallest reduction in LVFS compared to sedentary controls [46]. These findings suggest that the timing, duration, and format of high-intensity exercise relative to DOX administration are critical determinants of whether such an intervention is cardioprotective/detrimental. In the present study, the high-intensity continuous treadmill protocol administered concurrently across multiple DOX cycles appeared to create a sustained catabolic stress environment that might compromise structural protein integrity. Future investigations incorporating echocardiographic assessments alongside molecular analyses will be essential to determine whether the exacerbated protein catabolism observed herein translates into measurable deficits in cardiac contractile function,/whether compensatory mechanisms (enhanced autophagic flux and β-oxidation capacity demonstrated in the present study), are sufficient to preserve cardiac function during the treatment period.

DOX-mediated autophagy dysregulation has been implicated in the pathogenesis of cardiotoxicity. Exercise performed before/after DOX exposure suppresses DOX-induced autophagy dysregulation and confers cardioprotection, despite exercise being an established inducer of autophagy [44,47-49]. Consistent with prior observations indicating autophagy activation in response to DOX-induced cellular stress [25], we observed that DOX treatment stimulated autophagy, as evidenced by the elevated levels of LC3-II and ATG7 in the SED-DOX and EXE-DOX groups. Although exercise in the absence of DOX has been shown to upregulate autophagy and LC3-II levels exhibited a tendency to increase, no statistically significant difference was detected in the SED-DOX and EXE-DOX groups. This finding suggests that high-intensity exercise administered concurrently with DOX treatment enhances autophagy initiation. The abundance of p62 inversely reflects autophagic flux. Elevated flux results in the degradation of p62; consequently, impaired flux causes its accumulation [50,51]. Supporting the potentially beneficial influence of exercise on autophagic flux, a previous investigation demonstrated that exercise postconditioning administered after DOX treatment effectively restores p62 turnover [44]. Consistent with previous research, we observed a reduction in p62 in the EXE-DOX group compared to the SED-DOX group, suggesting that high-intensity aerobic exercise during DOX treatment may enhance autophagic flux, potentially aiding in the removal of damaged proteins accumulated due to DOX. However, decreased expression of the LAMP2 observed in EXE-DOX mice, even in the presence of increased Cathepsin L levels, necessitates further investigation to fully elucidate the impact of exercise on lysosomal function within this specific context. Moreover, exercise triggers the activation of Transcription Factor EB (key regulator of lysosomal biogenesis), thereby restoring lysosomal function even under DOX-induced stress [52]. Additionally, effective autophagic flux is contingent on the enzymatic activity of Cathepsin L, which is vital for hydrolyzing structural and regulatory proteins trafficked to lysosomes within autophagosomes. Studies have shown that Cathepsin L deficiency results in compromised degradation of autolysosomal components [53]. Thus, elevated Cathepsin L levels would suggest an attempt to enhance the breakdown of cellular components. Despite this, our finding of decreased LAMP2 expression in the EXE-DOX group, along with increased Cathepsin L levels, warrants further study. Lastly, in hearts subjected to DOX treatment, BNIP3-driven mitophagy appeared to be maladaptive, exacerbating mitochondrial respiratory defects and promoting cell death [54]. Moreover, another study demonstrated that exercise does not mitigate the DOX-induced elevation of BNIP3 [55]. Thus, increased BNIP3 levels across all intervention groups aligns with previous findings. This suggests that both DOX exposure and exercise influence mitochondrial turnover, potentially indicating the presence of persistent mitochondrial stress despite exercise intervention.

Therefore, we examined the upstream regulatory mechanisms governing autophagy, emphasizing the central role of the AMPK⍺/ULK1 pathway in the initiation and control of autophagy. While molecular markers indicate autophagy induction and flux, the underlying regulatory network dictates its occurrence and extent. The AMPK⍺/ULK1 pathway is highlighted as a key regulator, with AMPK⍺ and its downstream target, ULK1, being crucial for autophagosome formation [56]. The p-AMPK⍺^Thr172 increased across all intervention groups, suggesting a general activation of this energy sensor in response to DOX and exercise, consistent with prior research [57,58]. Notably, EXE-DOX exhibited the highest ratio of p-AMPK⍺^Thr172 to total AMPK⍺, even with reduced total AMPK⍺ levels, potentially indicating enhanced AMPK⍺ signaling under combined stress. Furthermore, the most pronounced elevation of p-ULK1^Ser555, a marker of ULK activation and a direct downstream target of AMPK⍺ [56,59], was observed in EXE-DOX, suggesting a robust initiation of autophagy under these specific conditions.

Despite the potential for enhanced autophagic flux observed with high-intensity aerobic exercise during DOX treatment, our findings indicate a detrimental impact on cardiac protein homeostasis. To assess the effect on structural protein integrity several cardiac proteins (β-actin, Talin, and cleaved α-spectrin II), were quantified. Interestingly, EXE-DOX exhibited a significant reduction in both β-actin and Talin. β-actin, a highly conserved protein, is integral to the actin cytoskeleton, providing cellular mechanical stability and shape [61]. Through its actin-binding sites, talin connects integrins to the actin cytoskeleton, facilitating force transmission and stabilizing focal adhesions [62]. The observed decrease in essential structural proteins strongly suggests active proteolytic degradation targeting these components. Further supporting this notion, a marked increase in cleaved α-spectrin II was observed specifically in EXE-DOX. Elevated levels of cleaved α-spectrin II are well-documented indicators of cellular injury, degradation, and cell death in various pathological conditions [63,64].

Furthermore, we investigated the potential involvement of the UPS, another major proteolytic pathway implicated in muscle wasting and cellular stress. Muscle atrophy is critically associated with the action of muscle-specific E3 ubiquitin ligases (MAFbx and MuRF1), which mediate the ubiquitination of intracellular protein substrates. Following the activity of these E3 ligases, target proteins are typically conjugated to polyubiquitin chains. These ubiquitin chains, specifically those assembled via isopeptide bonds between the lysine 48 (K48) residue of one ubiquitin molecule and the C-terminus of another, function as primary and canonical signals that direct the tagged proteins toward the cellular proteolytic system [65,66]. The 26S proteasome recognizes this characteristic ubiquitination signal, a substantial multicatalytic protease complex that facilitates the ATP-dependent breakdown of ubiquitinated proteins into constituent peptides [66]. An investigation using a mouse model demonstrated that the administration of DOX resulted in elevated mRNA expression levels of MuRF1 and MAFbx. Moreover, the findings indicated that preconditioning with short-term exercise mitigated this DOX-induced increase in the mRNA expression of these catabolism-associated proteins [55]. In our study, although MAFbx levels were not significantly altered, MuRF1 levels increased in the SED-DOX group. Furthermore, high-intensity exercise administered concurrently with DOX did not mitigate the DOX-induced increase in MuRF1 expression. To assess the overall activity of the UPS, we examined K48-linked polyubiquitinated proteins and several subunits of the 26S proteasome (PSMC1, PSMA1, and PSMB5). DOX treatment increased the levels of K48-linked polyubiquitinated proteins and proteasome subunits. High-intensity exercise administration during DOX treatment further augmented the levels of these proteasomal activity indicators. These findings collectively provide evidence that while high-intensity aerobic exercise may promote the clearance of damaged cellular components via autophagy, when combined with DOX, it simultaneously stimulates the breakdown of essential cardiac structural proteins. It is plausible that the combination of high-intensity exercise and DOX exposure created a unique cellular stress environment that exacerbated the degradation of specific protein pools.

Autophagy is recognized for its role in promoting lipolysis and suppressing adipogenic gene expression, thereby facilitating fat loss, particularly during metabolic states, such as fasting. Systemic autophagic activity influences weight loss as evidenced by the finding that autophagy deficiency impairs starvation-induced weight loss, whereas its activation enhances fat utilization [67]. Elevated systemic lipolysis results in increased flux of free fatty acids (FFAs) into systemic circulation. This augmented FFA availability can exceed the oxidative capacity of the myocardium, promoting the intracellular accumulation of toxic lipid species, including ceramides and diacylglycerols, within cardiomyocytes [68]. This lipotoxicity is implicated in the disruption of mitochondrial function, leading to the enhanced generation of ROS and induction of endoplasmic reticulum stress. Collectively, these cellular stressors initiate downstream signaling cascades that activate pathways that mediate the degradation of cellular constituents, such as proteins. Interestingly, in our study, high-intensity aerobic exercise administered concurrently with DOX treatment was associated with improved lipolysis markers and autophagy flux. DOX induces the elevation of ACC phosphorylation, often via an AMPK-dependent pathway [69] and causes a dramatic decline in DGAT levels, has been reported [70]. Consistent with these observations, we found that ACC phosphorylation increased and DGAT levels decreased in DOX-treated groups. This pattern suggested reduced fatty acid synthesis and potential disruption of triglyceride synthesis, which were attributed to DOX treatment [71]. Additionally, the expression of ACADVL and HADHSC, key enzymes essential for the mitochondrial β-oxidation of very long-chain fatty acids and short/medium-chain fatty acids, respectively, was increased in DOX-treated groups. Notably, high-intensity exercise during DOX treatment further augmented the expression levels of these β-oxidation enzymes. As discussed previously, the concurrent evidence of pronounced protein catabolism and enhanced markers of lipid metabolism, specifically increased β-oxidation enzyme expression and indicators of reduced fatty acid/triglyceride synthesis, may indicate a metabolic shift towards increased lipid utilization. This could compensate for potential energy deficits arising from structural protein degradation. Alternatively, these alterations in lipid metabolism may represent a direct consequence of the interplay between exercise stimuli and complex metabolic perturbations induced by DOX treatment.

Based on our findings, we concluded that while incorporating high-intensity exercise into DOX chemotherapy may enhance autophagic flux, it concurrently and significantly exacerbates cardiac protein catabolism by promoting lipolysis within the heart. This exacerbation, linked to increased UPS activation and degradation of essential structural proteins, suggests a potentially detrimental effect of this specific exercise regimen on the structural integrity of cardiac muscles during ongoing DOX chemotherapy. However, our analysis was conducted solely after the DOX and exercise treatments. Therefore, this study did not determine whether the observed cardiac protein catabolism in the post-treatment phase might contribute to beneficial processes, such as clearance of damaged cells and facilitation of tissue regeneration. These findings underscore the crucial need for the rigorous evaluation and planning of exercise interventions in patients undergoing cardiotoxic chemotherapy, particularly regarding temporal dynamics and long-term outcomes.

This study has certain limitations: the use of only male mice, which limits the generalizability of findings to female subjects; the absence of pharmacological autophagy inhibitors to directly confirm autophagic flux; the endpoint-only tissue analysis, which precludes assessment of temporal dynamics during and after treatment; the lack of cardiac functional measurements; and indirect VO_2max regarding exercise intensity of the treadmill protocol.

In conclusion, this study showed that the incorporation of high-intensity aerobic exercise between multiple cycles of DOX treatment in mice yields a complex outcome. While this specific exercise regimen appeared to enhance autophagic flux, it concurrently and significantly exacerbated cardiac protein catabolism. This heightened catabolism was evidenced by the degradation of essential structural proteins and increased UPS activity. These results suggest that, under the conditions tested, this high-intensity exercise protocol may have a detrimental impact on the structural integrity of the myocardium during ongoing DOX chemotherapy, counteracting potential cardioprotectiveness. However, a critical limitation is the endpoint analysis performed at the conclusion of the treatment period, which precludes an understanding of whether the observed transient catabolism contributed to beneficial post-treatment cellular clearance and tissue regeneration. This outcome underscores the need for the meticulous evaluation and personalized planning of exercise interventions for patients receiving cardiotoxic chemotherapy. Future investigations are warranted to delineate therapeutic strategies that optimize the beneficial effects of exercise-induced autophagy while simultaneously preventing deleterious levels of protein degradation. Furthermore, exploring the potential contributions of transient catabolic states to tissue remodeling and repair in the post-treatment phase is crucial. Concurrently, a comprehensive understanding of the divergent molecular signaling cascades that govern the influence of exercise on autophagic processes versus proteolytic systems within the context of DOX-induced cardiac insult is essential for the development of targeted interventions.