Molecular mechanisms underlying the ameliorative effects of crocodile head-derived bioactive peptides on DEX-induced muscle atrophy: insights from proteomics and gut microbiota analysis

Abstract

Muscle atrophy is relatively common among older adults and can markedly impair their physical function and overall quality of life. Crocodile head-derived bioactive peptides (CP), with high bioavailability and low allergenicity, show promise as a nutritional intervention. This study aimed to investigate the alleviating effects of CP on muscle atrophy and clarify their underlying action mechanisms. Results demonstrated that CP alleviated muscle atrophy-related weight loss, increasing gastrocnemius, quadriceps, and tibialis anterior muscle indices by 60%, 33%, and 28%. The lean body mass percentage increased by 3.7%, while the body fat rate decreased by 6%. Following CP administration, grip strength, motion displacement, and exhaustion time in mouse recovered to 1.71 N, 495 m, and 1130 s. Proteomic analysis revealed that CP potentially ameliorates muscle atrophy by modulating the AMPK signaling pathway to restore energy metabolism homeostasis, activating autophagy to clear dysfunctional organelles, and reprogramming lipid metabolism to suppress ectopic fat deposition in muscle tissue. Analysis of the gut microbiota further indicated that CP intervention significantly increased the abundance of beneficial bacteria, including Muribaculaceae, Allobaculum, Lactobacillus, Monoglobus, and Dubosiella. In conclusion, CP likely mitigates muscle atrophy progression via multi-target mechanisms, including modulation of energy metabolism, autophagy, lipid metabolism, and the gut microenvironment.

Introduction

Global aging has become a profound and irreversible trend in this century. It is not only a huge challenge for developed countries, but also an increasingly urgent problem for many developing countries. According to the United Nations standards, when a country or region’s population aged 65 and above accounts for 7%, it is classified as an aging society. At present, more than 100 countries around the world have reached this threshold. This trend is irreversible. As of 2015, 94 countries and regions worldwide have entered the aging society stage, and this number is expected to increase to 158 by the middle of the 21st century¹. Age-related health problems associated with population aging have become increasingly prominent, and muscle atrophy has received special attention due to its significant impact on quality of life². The muscle atrophy of the elderly is characterized by decreased muscle mass, decreased muscle strength, and decreased physical fitness. Epidemiological data show that about 33% of the world ‘s population aged 60 and over are affected by muscle atrophy, and the prevalence rate increases significantly in an age-dependent manner³. Muscle atrophy not only poses a serious threat to the health and quality of life of the elderly but also poses a huge social and economic burden. Therefore, there is an urgent need to strengthen prevention and treatment strategies for this age-related disease.

Currently, no drug has been globally approved for the specific treatment of muscle atrophy. The main clinical interventions are still non-drug therapy, mainly nutritional support and exercise therapy⁴. Existing studies have shown that higher protein intake is negatively correlated with the prevalence of muscle atrophy in the elderly. Adequate dietary protein intake may help reduce the risk of muscle atrophy⁵. Aging is accompanied by a significant decline in digestive function, especially the decrease of protease activity in the three major digestive enzymes⁶. Simply increasing protein intake is not enough to meet the daily protein needs of the elderly. Bioactive peptides provide a promising alternative to dietary supplements due to their high bioavailability and low allergenicity, which may reduce muscle atrophy associated with impaired protein absorption in the elderly.

Crocodile is a biological resource with both ecological value and medicinal and edible characteristics. Its comprehensive value covers many fields such as food, health care, medicine, and luxury goods. In traditional Chinese medicinal diet, crocodile meat is recognized as a high-quality protein source, which is characterized by low fat and cholesterol content, and is rich in essential amino acids, unsaturated fatty acids, and trace elements such as iron and zinc. In traditional food culture, crocodile is considered to nourish the heart and lung, replenish qi and blood, and strengthen bones and muscles. Crocodile skin is considered to be a high-quality material in the luxury leather industry due to its unique texture and durability. Crocodile head, a protein-rich source of potential bioactives, is typically wasted as a by-product, remaining an underexploited resource. Sixty-seven potential bioactive peptides were identified from crocodile head protein by enzymatic hydrolysis in our previous study⁷. Our preliminary experiments revealed that CP exhibits potent immunomodulatory activity⁸. However, its effect on muscle atrophy remains unclear.

The study aimed to investigate the effects of CP intervention in a muscle atrophy mouse model, and motor function, muscle mass, and strength parameters were assessed, confirming the ameliorative effect of CP on muscle atrophy. Additionally, proteomic of gastrocnemius muscle tissue and microbiome sequencing of fecal samples were performed to elucidate the underlying molecular mechanisms.

Results

CP intervention improves basic status in DEX-treated mice

During the modeling of muscle atrophy, compared with the control group, the body weight of mice in the DEX treatment group (Fig. 1a, p < 0.05) was significantly reduced, while food intake showed no significant change (Fig. 1b). Systemic catabolism and loss of appetite are typical manifestations of muscle atrophy². Compared with the DEX group, CP intervention led to a significant recovery in body weight (Fig. 1c, p < 0.05). No significant difference in food intake was observed, further indicating that CP effectively alleviated DEX-induced muscle atrophy (Fig. 1d).

Fig. 1. Effects of CP on basic parameters in sarcopenic mice.

a Body weight and b food intake before CP intervention; c body weight and d food intake after CP intervention; e grip strength, f movement distance, and g exhaustion time following CP treatment; h spleen index, i cardiac index, j renal index, k bone health index, and l hepatic index after CP administration. Data of six or more independent experiments were expressed as mean ± SEM (P < 0.05).

CP intervention alleviates motor function decline in DEX-treated mice

In the motor function test, DEX-treated mice showed significantly reduced grip strength (Fig. 1e, p < 0.05), reduced motor displacement (Fig. 1f, p < 0.05) and shorter exhaustion time (Fig. 1g, p < 0.05). These consistent results suggested that DEX administration results in a comprehensive impairment of muscle contraction function and exercise endurance, further confirming the successful establishment of an animal model of muscle atrophy. Compared with the DEX-treated group, CP intervention significantly improved motor function in mice, as evidenced by a grip strength of 1.71 N (Fig. 1e, p < 0.0001), a motion displacement of 495 m (Fig. 1f, p < 0.05), and an exhaustion time of 1130 s (Fig. 1g, p < 0.01). Although CP treatment did not fully restore motor function parameters to normal levels, the overall results showed that CP intervention effectively alleviated DEX-induced muscle atrophy-related phenotypes.

CP alleviates organ index abnormalities in DEX-treated mice

Aging is often accompanied by multiple organ dysfunction⁹. In order to systematically evaluate the systemic effects of CP intervention on multiple organs, this study measured a series of indicators, including spleen index, cardiac index, renal index, bone health index, and hepatic index. The results showed that compared with the control group, DEX treatment significantly reduced spleen index (Fig. 1h, p < 0.001) and renal index (Fig. 1j, p < 0.01), but had no significant effect on cardiac index (Fig. 1i), bone health index (Fig. 1k) and hepatic index (Fig. 1l). Further analysis showed that CP intervention completely reversed the DEX-induced abnormalities, restoring the spleen index to 0.21% and the renal index to 1.29%, which were comparable to the control group levels. This indicated that CP has a significant protective effect on DEX toxicity in multiple organs.

Effect of CP on muscle histological structure in DEX-treated mice

H&E staining showed that compared with the control group, mice in the DEX treatment group showed typical pathological features of muscle atrophy, including increased muscle fiber diameter variation, disordered arrangement, partial fiber atrophy with serrated contours, blurred or missing stripes. After CP intervention, these pathological changes were significantly alleviated, the arrangement and compactness of muscle fibers were restored, the stripes were clear, and there was no congestion, edema and inflammatory cell infiltration in the stroma (Fig. 2a). Further DEXA analysis of muscle mass showed that the indexes of gastrocnemius, quadriceps femoris, tibialis anterior and extensor digitorum longus in the DEX model group (Fig. 2b–f, p < 0.001) were significantly decreased, the percentage of lean tissue (Fig. 2h, p < 0.05) was decreased, and the percentage of adipose tissue (Fig. 2g, p < 0.01) was significantly increased. CP alleviated muscle atrophy-related weight loss, increasing gastrocnemius, quadriceps, and tibialis anterior muscle indices by 60%, 33%, and 28% (Fig. 2c-e). Concurrently, the lean body mass percentage increased by 3.7% (Fig. 2h, p < 0.05), while the body fat rate decreased by 6% (Fig. 2g, p < 0.01). Figure 2i shows the changes in the cross-sectional area (CSA) of gastrocnemius muscle fibers. Compared with the control group, DEX administration significantly reduced muscle fiber CSA (Fig. 2i, p < 0.0001), indicating successful establishment of the muscle atrophy model. Compared with the DEX group, CP treatment markedly increased CSA (Fig. 2i, p < 0.0001), suggesting that CP effectively ameliorated DEX-induced muscle atrophy. However, compared with the control group, the CSA in the CP group remained slightly lower (Fig. 2i, p < 0.05), indicating that muscle fiber size was partially, but not fully, restored. The combination of wheat oligopeptides and fish oil-derived ω-3 PUFAs significantly increased muscle mass and grip strength in rats, while also ameliorating fat infiltration, muscle atrophy, interstitial congestion, and inflammatory cell infiltration in muscle tissue¹⁰. These findings demonstrate that peptides exhibit pronounced efficacy in alleviating muscle atrophy. Recent studies have shown that muscle atrophy not only involves muscle loss, but also involves abnormal infiltration of adipose tissue into muscles, known as “intramuscular fat infiltration” or “intermuscular fat deposition”. This process is characterized by the fact that during aging and muscle mass loss, adipose tissue invades the intermuscular space and even the intracellular region of muscle cells, replacing the original contractile tissue, and impairing muscle function and strength regulation¹¹. We speculated that CP may alleviate muscle function damage by inhibiting ectopic fat deposition and reducing its erosion of muscle structure. This mechanism was consistent with the recently proposed “muscle-fat metabolism reprogramming” theory, which provided a new explanation for the efficacy of CP in improving muscle atrophy.

Fig. 2. Effects of CP on muscle histological structure and body composition.

a Pathological changes in muscle tissue; b muscle mass content; c gastrocnemius index; d quadriceps index; e tibialis anterior index; f extensor digitorum longus index; g adipose tissue percentage; h lean tissue percentage; i mean cross-sectional area of myofibers in gastrocnemius. Data of six or more independent experiments were expressed as mean ± SEM (P < 0.05).

The molecular mechanism of CP alleviating muscle atrophy: Proteomic analysis

Proteomic results showed that DEX treatment induced significant changed in the expression of many proteins in the gastrocnemius muscle of mice compared with the control group (Fig. 3c), indicating that DEX disrupted muscle protein metabolism. CP intervention significantly reversed these abnormal protein expressions (Fig. 3b), indicating that it has the potential to improve DEX-induced muscle metabolic disorders. Further quantitative analysis of differentially expressed proteins showed that CP treatment significantly up-regulated the expression levels of Mapkapk3, Atp6v1e1, DCNL2, Lama5 and Raptor, while down-regulated the expression levels of Pdk4, Cd36, Cav1, Rac1 and Lipe (Fig. 3a, d–m). These proteins are involved in multiple biological processes such as inflammatory response, energy metabolism, cell signal transduction, and muscle fiber structure maintenance, suggesting that CP may play a role through multi-target and multi-pathway mechanisms, thereby alleviating the pathophysiological changes associated with muscle atrophy.

Fig. 3. Proteomic analysis of gastrocnemius muscle tissue.

a Heatmap of differentially expressed proteins following CP intervention; b volcano plot of protein differences between CP and DEX groups; c volcano plot of protein differences between DEX and Control groups; d–m intensity comparison of key proteins in CP/DEX groups. Data of six or more independent experiments were expressed as mean ± SEM (P < 0.05).

To explore the regulatory effect of CP on the biological process of mouse gastrocnemius muscle, GO enrichment analysis was performed on its potential targets. GO enrichment analysis showed that CP intervention significantly enriched biological processes mainly related to mitochondrial function (mitochondria), proton transport ATP synthase complex activity, cell response to linoleic acid, low-density lipoprotein particle clearance, and nitric oxide synthase binding (Fig. 4a). These results suggested that CP may interfere with the pathological progression of muscle atrophy at the molecular level by regulating key biological processes such as mitochondrial energy metabolism, lipid metabolism homeostasis and nitric oxide-mediated signal transduction. In order to further study the regulatory effect of CP on the metabolic pathway of gastrocnemius muscle in mice, KEGG enrichment analysis was performed on its potential targets. The results showed that the signal pathways significantly enriched by CP intervention included phagosome, fat digestion and absorption, AMPK signaling pathway and adipocytokine signaling pathway (Fig. 4b). GSEA results showed that the AMPK signaling pathway and the mTOR signaling pathway were significantly positively enriched in the CP treatment group (NES = 1.571 and 1.6787, respectively; p < 0.05), suggesting that CP may exert its biological effects by regulating energy metabolism and nutrient-sensing pathways (Fig. 4c, d). Meanwhile, the autophagy pathway also exhibited a positive enrichment trend (NES = 1.4947), which, although close to statistical significance (p = 0.05296), indicates its potential involvement in the regulatory effects of CP (Fig. 4e). In contrast, the longevity-regulating pathway did not reach statistical significance (p > 0.05) (Fig. 4d). GSEA further verified the significant enrichment trend of these pathways in the CP treatment group, which was consistent with the conclusions of GO functional enrichment and KEGG pathway analysis. These results suggest that CP alleviates muscle metabolic disorders by regulating key biological processes such as immune-related phagocytosis, lipid metabolism and absorption, and energy sensing coupled with inflammation regulation.

Fig. 4. Proteomic analysis of gastrocnemius muscle tissue.

a GO functional enrichment analysis of CP/DEX group; b KEGG pathway enrichment analysis of CP/DEX group; c–f gene set enrichment analysis (GSEA) of CP/DEX group. Data of three or more independent experiments were expressed as mean ± SEM (P < 0.05).

Molecular docking

The above results suggest that CP may alleviate DEX-induced muscle atrophy by regulating multiple differentially expressed proteins and related signaling pathways. Therefore, two key regulatory proteins, Pdk4 and Raptor, were selected for molecular docking analysis to further explore the potential molecular mechanisms involved. In the evaluation of binding affinity with Pdk4, the top three ranked peptides were YPTL, HGSF, and DPF, with Vina scores of −9.164, −9.144, and −8.933 kcal/mol, respectively, indicating the potential formation of relatively stable binding conformations with Pdk4. Specifically, YPTL formed stable hydrogen bonds with THR358, ARG261, ASN258, GLY329, and ASN321 of Pdk4, along with a π–cation interaction with LYS257 (Fig. 5a). HGSF primarily formed hydrogen bonds with ASN258, GLY329, and ASN321, and exhibited π–π stacking interactions with PHE330 and TYR332, thereby enhancing binding stability (Fig. 5b). DPF formed multiple hydrogen bonds with TYR332, GLY331, GLY333, LEU327, ASN258, THR313, and ARG261, and additionally formed a salt bridge interaction with an ARG residue, demonstrating a relatively rich and stable non-covalent binding pattern (Fig. 5c). In the evaluation of binding affinity with Raptor, the top three ranked peptides were DPF, NY, and GGY, with Vina scores of −7.950, −7.316, and −7.306 kcal/mol, respectively, suggesting the potential formation of relatively stable binding conformations with Raptor. Specifically, DPF exhibited π–π stacking interactions with a PHE residue (Fig. 5d). NY formed stable hydrogen bonds with LEU182, LEU177, and GLY162, π–π stacking interactions with PHE148, and a π–cation interaction with TYR150 (Fig. 5e). GGY formed hydrogen bonds with GLY162 and TYR150, along with π–π stacking interactions with PHE148 (Fig. 5f), further contributing to binding stability.

Fig. 5. Predicted binding modes of selected CP with Pdk4 and Raptor.

a–c Docking poses of YPTL, HGSF, and DPF with Pdk4 (2ZKJ) are shown in panels, d–f while docking poses of DPF, NY, and GGY with Raptor (8RCN) are presented in panels. The left panels display surface representations of the proteins, the middle panels show the overall docking conformations, and the right panels provide enlarged views of the binding pockets. Key non-covalent interactions are illustrated according to the legend.

CP modulates gut microbiota dysbiosis in muscle atrophy

In this study, 16S rDNA high-throughput sequencing technology was used to systematically analyze the effect of CP intervention on the intestinal flora of DEX-induced muscle atrophy mouse model. Beta diversity analysis showed that there was a significant separation between the DEX treatment group and the control group in PCA and PCoA, indicating that DEX significantly destroyed the overall microbial community structure. After CP intervention, changes in microbial composition and sample distribution were observed compared with the DEX group, suggesting that CP has a certain impact on DEX-induced gut microbiota dysbiosis (Fig. 6a, b). Further evaluation using the dysbacteriosis index found that the DEX group was significantly increased compared with the control group, confirming the state of intestinal ecological imbalance. In contrast, CP intervention significantly reduced this indicator, demonstrating its positive regulatory effect on the structure of intestinal flora and its partial mitigation effect on DEX-induced ecological damage (Fig. 6c). The results from the two dimensions of community structure and dysbacteriosis level suggested that CP may improve the metabolic environment related to muscle atrophy by regulating the intestinal microbiota. At the phylum level, compared with the DEX group, CP intervention significantly increased the relative abundance of Bacillota and Bacteroidota. The increase of Bacillota abundance indicated that the ability of carbohydrate catabolism was enhanced, while the increase of Bacteroidota indicated that protein metabolism and the maintenance of intestinal mucosal barrier were improved. These findings suggested that CP may synergistically enhance intestinal metabolism and barrier function by regulating the structure of key microorganisms (Fig. 6d). At the genus level, CP intervention significantly increased the relative abundance of beneficial bacteria (e.g., Allobaculum and Lactobacillus) and decreased the relative abundance of opportunistic pathogenic bacteria (e.g., Alistipes). However, we observed that the levels of Allobaculum and Lactobacillus in the Control group were lower than those in the DEX and CP groups, whereas the relative abundance of Alistipes was comparatively higher in the Control group. These findings suggest that the composition of the gut microbiota is complex, and the observed changes in relative abundance may represent accompanying phenomena rather than direct mechanisms underlying the therapeutic effects (Fig. 6e). Consistent results were observed at the species level, further demonstrating the ability of CP to regulate the structure of gut microbiota by promoting beneficial microorganisms and inhibiting potentially harmful bacteria, thereby improving microbial ecological balance (Fig. 6f). Circos analysis of microbial communities showed that CP intervention significantly increased the relative abundance of beneficial bacteria, including norank_f_Muribaculaceae, Allobaculum and Lactobacillus (Fig. 6g). These microbial groups are key producers of acetic acid, butyric acid and lactic acid, respectively, and are important beneficial bacteria for maintaining intestinal health¹². The increase in their abundance directly reflects the beneficial effect of CP on improving the intestinal microecological environment. Ternary plot analysis further demonstrated that CP treatment specifically enriched Allobaculum and Faecalibaculum, of which Allobaculum showed the most obvious enrichment (Fig. 6h), indicating its potential role as a key target for CP regulation of gut microbiota. CP-targeting strains were screened by multi-group comparison, and it was found that the core flora was significantly enriched, especially Allobaculum and Lactobacillus (Fig. 7a). Pairwise comparison analysis at the species level found that CP intervention significantly increased the relative abundance of the four bacterial strains (Fig. 7b). They were uncultured _ Bacteroidales _ bacterium _ f _ _ Muribaculaceae, unidentified _ g _ _ Monoglobus, Lactobacillus _ sp. _ C30An8 and uncultured _ bacterium _ g _ _ Dubosiella. These strains play an active role in maintaining intestinal barrier function, promoting SCFA production, and regulating immune response. Therefore, it is speculated that CP may enhance the integrity of intestinal barrier by specifically enriching these functional bacteria, thereby alleviating the pathological phenotype related to muscle atrophy. In order to further study the effects of different treatments on the structure of intestinal flora, we performed LEfSe analysis. The results showed that DEX intervention significantly enriched potentially harmful bacteria including o_ Erysipelotrichales, o_Bacillales and g_Parabacteroides (Fig. 7c). In contrast, CP intervention significantly increased the abundance of beneficial bacteria such as f__Monoglobaceae, g__Monoglobus, and g__Dubosiella (Fig. 7d). These results suggest that CP may ameliorate DEX-induced gut dysbiosis by modulating microbiota composition through suppressing potentially pathogenic bacteria and promoting beneficial bacterial growth.

Fig. 6. Analysis of gut microbiota.

a PCA based on ASV level; (b) PCoA based on genus level; c MDI index representing microbial community differences among three groups; d phylum-level composition; e genus-level composition; f species-level composition; g circos diagram showing microbiota-sample relationships; h ternary plot showing taxonomic distribution among groups.

Fig. 7. Analysis of gut microbiota.

a Microbial composition bar plot based on one-way ANOVA; b species-level differential analysis bar plot based on Wilcoxon rank-sum test; c LEfSe analysis bar plot of DEX/Control groups; d LEfSe analysis bar plot of CP/DEX groups.

We further performed correlation analyses between gut microbiota (at the family and genus levels) and key differentially expressed proteins identified by proteomic screening. At the family level (Fig. 8a), Akkermansiaceae showed significant positive correlations with Atp6v1e1, DCNL2, Lama5, and Raptor, and negative correlations with Cav1 and Cd36, suggesting its potential involvement in the regulation of energy and lipid metabolism. In contrast, Streptococcaceae and Anaerovoracaceae were significantly negatively correlated with Atp6v1e1, DCNL2, and Lama5, indicating a possible association with suppression of metabolic signaling or impaired muscle function. At the genus level (Fig. 8b), Akkermansia was significantly positively correlated with energy metabolism–related proteins, including Atp6v1e1, DCNL2, Raptor, and Mapkapk3, suggesting that it may contribute to the maintenance of muscle metabolic homeostasis through regulation of energy sensing, autophagy, and related signaling pathways. Parabacteroides exhibited a similar correlation pattern to Akkermansia, implying a potential protective role in host metabolic regulation. In contrast, Muribaculum and Streptococcus were significantly negatively correlated with Atp6v1e1, DCNL2, Mapkapk3, and Raptor, suggesting that these taxa may be associated with suppressed energy metabolism or exacerbation of muscle atrophy. Overall, the correlation analysis indicates that certain potentially beneficial taxa are positively associated with proteins involved in energy metabolism and signaling regulation, whereas taxa potentially linked to metabolic dysregulation show negative correlations, further supporting a functional association between gut microbiota alterations and skeletal muscle metabolic regulation.

Fig. 8. Correlation heatmap analysis between gut microbiota and skeletal muscle–related proteins.

a Correlation analysis between gut microbiota at the family level and the expression of skeletal muscle–related proteins; b correlation analysis between gut microbiota at the genus level and the expression of skeletal muscle–related proteins.

Discussion

CP alleviated muscle atrophy in mice, improving grip strength, travel distance, exhaustion time, lean mass, and body fat composition. We hypothesize these benefits occur via inhibition of ectopic fat deposition and its structural damage to muscle.

Proteomic analysis of mouse gastrocnemius muscle tissue was performed. The results showed that CP intervention remodeled protein metabolism, which may be the molecular mechanism of CP alleviating muscle atrophy. Further analysis was conducted on the strongly enriched proteins. MAP kinase activated protein kinase 3 (Mapkapk3) is a serine/threonine protein kinase that acts downstream of the mitogen-activated protein kinase (MAPK) pathway. MAPK, also known as extracellular signal-regulated kinase (ERK), is the key integration node of a variety of biochemical signals in cells¹³. The activation of ERK1/2 signaling pathway usually promotes the repair and regeneration of muscle tissue. Studies have shown that fibroblast growth factor 6 (FGF6) enhances muscle regeneration after nerve injury through an ERK1/2-dependent signaling mechanism¹⁴. It is therefore hypothesized that the ameliorative effect of CP on muscle atrophy may partially depend on the upregulation of Mapkapk3. The V-type proton ATPase subunit E1 (Atp6v1e1) is a component of the V-ATPase, which uses the energy generated by ATP hydrolysis to pump protons (H + ) into organelles (e.g., lysosomes) or through the cell membrane, thereby acidifying the environment¹⁵. Muscle damage often induces secondary mitochondrial damage. When dysfunctional mitochondria cannot be cleared in time by quality control mechanisms such as autophagy, they produce excessive reactive oxygen species (ROS), thereby exacerbating oxidative stress. This, in turn, leads to further destruction of the structure of muscle fibers, forming a vicious cycle of self-amplification and accelerating the decline of muscle function¹⁶. As a key subunit of V-type proton ATPase, Atp6v1e1 promotes the effective degradation of damaged mitochondria by acidifying the autolysosomal cavity (pH 4.5 ~ 5.0), thereby maintaining cell energy homeostasis. This process contributes to the amelioration of muscle atrophy-associated metabolic dysfunction and muscle mass loss. DCN1-like protein 2 (DCNL2) is a positive regulator of cullin-RING ligase (CRL) complex, which can activate CRL to promote ubiquitination of specific target proteins, leading to its proteasome degradation. Based on the observed remission of muscle atrophy symptoms, we hypothesized that DCNL2 may maintain muscle mass and improve the phenotype of muscle atrophy by labeling and accelerating the degradation of specific proteins associated with muscle atrophy. Laminin α-5 subunit (Lama5) is a key component of laminin-511 (LN-511) and laminin-521 (LN-521), which mediates cell adhesion, differentiation and survival in various tissues and plays a vital role in maintaining tissue structural integrity. In the tumor microenvironment, Lama5 may promote tumor cell proliferation, migration and angiogenesis by activating signaling pathways such as Notch and STAT3, thereby driving tumor progression¹⁷. Under normal physiological conditions, especially in muscle tissue, extracellular matrix components containing Lama5 provide structural support for muscle fibers and transmit biochemical signals, which may promote muscle cell proliferation and differentiation while maintaining tissue morphology and functional stability. Although Lama5 seems to have the opposite effect in tumor and muscle tissues, its function in both physiological environments highlights its important regulatory value. Lama5 can be used not only as a target to inhibit tumor progression, but also as an intervention to promote muscle regeneration, which has potential therapeutic value. However, its exact mechanism of action needs further experimental verification. Raptor is the core component of m TOR complex 1 (m TORC1), which is the central hub of cell nutrition and energy perception, regulating protein synthesis, cell growth and metabolism. Extensive research has been conducted on mTORC1’s role in muscle tissue. Zein-derived peptides have been shown to promote C2C12 myoblast proliferation through crosstalk between mTORC1 and mTORC2 signaling pathways¹⁸. HydroxyprolyI-glycine-rich collagen peptide activates mTORC1 to promote skeletal muscle protein synthesis in mice¹⁹. The improvement of CP on muscle atrophy may be related to the up-regulation of key proteins such as Mapkapk3, Atp6v1e1, DCNL2, Lama5 and Raptor. These proteins are involved in the regulation of cellular stress response, autolysosomal acidification, protein ubiquitination and degradation, extracellular matrix integrity, and mTOR-mediated muscle anabolism pathways, and jointly promote muscle fiber function recovery and muscle mass maintenance.

The down-regulated protein after CP intervention was analyzed and it was found that the expression of pyruvate dehydrogenase kinase isoenzyme 4 (Pdk4) was inhibited. Pdk4 inhibits pyruvate dehydrogenase (PDH) activity by phosphorylation, thereby shifting glucose metabolism from an oxidative pathway to an alternative metabolic pathway²⁰. Under the condition of muscle injury, continuous Pdk4 activity will aggravate the disorder of glucose metabolism, lead to insufficient energy supply in muscle tissue, and further damage muscle function. CP intervention significantly reduced the activity of Pdk4 and alleviated its inhibition of PDH, thereby promoting the entry of normal pyruvate into the tricarboxylic acid (TCA) cycle. This restored glucose oxidation metabolism in muscle tissue and provides the necessary energy support for muscle fiber repair. Cluster of differentiation 36 (CD36) is a kind of scavenger receptor, which mainly promotes the uptake of long-chain fatty acids by cells²¹. In muscle tissue, normal CD36 expression regulates the transport and utilization of fatty acids to maintain lipid homeostasis. Abnormal up-regulation will promote ectopic lipid deposition between muscle fibers, leading to intramuscular fat infiltration. This pathological change not only destroys the structure of muscle cells, but also impairs contractile function, thereby significantly exacerbating the loss of muscle mass and functional decline associated with muscle atrophy. These mechanisms are consistent with the phenotypic changes observed in previous studies, further supporting the key role of CD36 dysregulation in the progression of muscle atrophy. Hormone-sensitive lipase (Lipe) is a key enzyme expressed mainly in adipocytes, which catalyzes the hydrolysis of triglycerides and releases stored fat as free fatty acids to produce energy²². Although the expression of Lipe is mainly located in adipose tissue, its abnormal up-regulation in muscle tissue may indicate systemic lipolysis. When co-occurs with the high expression of fatty acid transporter CD36, it can promote ectopic lipid deposition in muscle tissue and lead to intramuscular fat infiltration. This pathological change destroys the integrity of muscle fiber structure, impairs contractile function, aggravates insulin resistance, and jointly promotes the occurrence and development of muscle atrophy. CP intervention may alleviate the progression of muscle atrophy by down-regulating the expression of the above proteins. Membrane invaginin-1 (Cav1) and Ras-related C3 botulinum toxin substrate 1 (Rac1) show functional duality in a variety of physiological and pathological processes. Cav1 plays a beneficial role in maintaining cell metabolic homeostasis by regulating insulin receptor signaling and participating in cholesterol transport²³. The abnormal expression of Cav1 may aggravate insulin resistance and promote metabolic disorders²⁴. Rac1 is essential for regulating cytoskeletal reorganization, promoting cell proliferation, and promoting muscle regeneration²⁵. Its excessive activation will induce excessive ROS production²⁶, leading to oxidative stress damage and accelerating muscle protein degradation. Given the context-dependent and potentially opposing roles of these two proteins, their exact effects on muscle atrophy require further research to elucidate their specific mechanisms and therapeutic potential.

GO enrichment analysis found that CP intervention significantly enriched biological processes related to mitochondrial function, proton transport ATP synthase complex activity, and low-density lipoprotein particle clearance. These findings are consistent with our previous proteomics results. CP may alleviate muscle atrophy by promoting the removal of damaged mitochondria, improving lipid metabolism disorders, and restoring cell energy homeostasis. KEGG pathway enrichment analysis further showed that CP intervention significantly regulated AMPK signaling pathway, phagosome pathway, adipocytokine signaling pathway and fat digestion and absorption pathway, which was consistent with the biological process revealed by GO analysis. The results of GSEA also confirm this point. Meanwhile, molecular docking analysis was performed for the key proteins Pdk4 and Raptor involved in the aforementioned pathways. The results showed that the peptides YPTL, HGSF, and DPF could form stable complexes with Pdk4 through hydrogen bonds, π–cation interactions, π–π stacking, and salt bridge interactions. Similarly, the peptides DPF, NY, and GGY were predicted to form stable complexes with Raptor via hydrogen bonds, π–cation interactions, and π–π stacking interactions. These findings suggest that CP-derived peptides may potentially interact with key regulatory proteins such as Pdk4 and Raptor, thereby modulating related signaling pathways and contributing to the alleviation of muscle atrophy.

Previous studies have shown that decreased activity of the AMPK signaling pathway is one of the key features of muscle atrophy, leading to impaired energy sensing, reduced mitochondrial oxidative capacity, and disrupted metabolic homeostasis, thereby promoting muscle protein degradation²⁷. Impaired autophagy, particularly defective mitophagy, results in the accumulation of damaged mitochondria and increased oxidative stress, which further aggravates muscle atrophy²⁸. Muscle atrophy is also frequently accompanied by abnormal lipid metabolism and ectopic fat deposition within skeletal muscle, and such lipid accumulation is closely associated with insulin resistance, chronic low-grade inflammation, and deterioration of muscle quality²⁹. CP alleviates muscle atrophy through three main mechanisms: (1) regulating AMPK signaling pathway to restore energy metabolism homeostasis; (2) Activate the autophagy pathway to remove dysfunctional organelles (especially mitochondria) caused by muscle damage; (3) Remodeling lipid metabolism and inhibiting ectopic fat deposition in muscle tissue.

The research on the intervention of muscle atrophy is paying more and more attention to the intestinal microbiota. The analysis of intestinal flora in elderly patients with muscle atrophy found that microbial composition changed significantly. The fecal microbiota of elderly donors with high muscle function (HF) or low muscle function (LF) was transplanted into sterile mice, and it was found that the muscle performance of HF receptors was significantly better than that of LF receptors³⁰. We analyzed the intestinal flora of mice and found that CP intervention significantly alleviated DEX-induced dysbacteriosis. CP treatment significantly increased the abundance of phylum Bacillota. Some key genera in this phylum can produce a large amount of SCFAs, which play a vital role in maintaining intestinal barrier function. Circos and Ternary analysis showed that CP intervention significantly up-regulated the abundance of norank_f _ Muribaculaceae, Allobaculum and Lactobacillus. Studies have shown that norank_f _ Muribaculaceae can degrade proteoglycans and compound dietary fibers in the intestine and produce a large amount of SCFAs. There is a cross-feeding between this genus and Bifidobacterium and Lactobacillus³¹. Allobaculum is a well-characterized butyrate-producing genus³². Butyrate, as the preferred energy source for colonic epithelial cells, is essential for enhancing intestinal barrier function. A complete intestinal barrier can prevent the translocation of lipopolysaccharide (LPS) into the blood, thereby reducing the key drivers of systemic chronic inflammation in muscle catabolism. Lactobacillus is a recognized probiotic, mainly through the production of lactic acid. Accumulating evidence indicates that Lactobacillus species can modulate muscle tissue. Lactobacillus paragasseri SBT2055 has been shown to reduce obesity in mice through the adipose tissue-muscle-gut axis³³. Velvet antler fermented by Lactobacillus curvatus HY7602 can significantly restore the muscle mass of aged mice and improve their athletic ability³⁴. Lactobacillus also showed strong anti-inflammatory properties. Lactobacillus paracasei KW3110 has been shown to prevent mitochondrial dysfunction induced by inflammatory stress in mouse macrophages³⁵. There is a syntrophic relationship between norank _ f_Muribaculaceae and Lactobacillus, but the regulatory mechanism of CP on these two strains needs further study. LEfSe analysis found that Monoglobus and Dubosiella were significantly up-regulated. They are key dietary fiber degradation experts in the intestine and can produce a large amount of propionate, acetate and lactic acid. These microbial metabolites contribute to intestinal barrier integrity. CP intervention significantly up-regulated the abundance of beneficial bacteria such as norank _ f_Muribaculaceae, Allobaculum, Lactobacillus, Monoglobus and Dubosiella in the intestine. SCFAs produced by these bacteria can enhance the integrity of the intestinal barrier and prevent the translocation of endotoxin into the blood, thereby alleviating muscle atrophy by reducing systemic inflammation. Meanwhile, correlation analysis showed that the aforementioned beneficial taxa were significantly positively correlated with Atp6v1e1, DCNL2, Lama5, Raptor, and Mapkapk3, and negatively correlated with Cav1, Cd36, Lipe, Pdk4, and RAC1. These findings suggest that gut microbiota may serve as potential modulatory factors in alleviating muscle atrophy.

This study systematically evaluated the ameliorative effects of CP on DEX-induced muscle atrophy. Behavioral tests demonstrated that CP intervention significantly restored grip strength, improved exercise endurance, and increased muscle mass in mice. Proteomic analysis revealed that CP exerts its effects through multiple mechanisms: (1) modulating the AMPK signaling pathway to restore energy metabolism homeostasis; (2) activating autophagy to clear dysfunctional organelles (particularly mitochondria) in muscle tissue; (3) improving lipid metabolism to suppress ectopic fat deposition in muscle. Gut microbiota analysis further revealed that CP significantly increased the abundance of beneficial bacteria including Muribaculaceae, Allobaculum, Lactobacillus, Monoglobus, and Dubosiella. These microbial populations enhance intestinal barrier function through SCFA production, reducing endotoxin translocation and systemic inflammation, thereby ameliorating muscle atrophy via the gut-muscle axis. In summary, CP exerts protective effects through multi-target and multi-pathway regulation of muscle metabolism and gut microecology, demonstrating potential as a nutritional intervention strategy.

Although the DEX-induced model is widely used due to its stability and reproducibility in evaluating the effects of functional food components on muscle atrophy, it primarily reflects glucocorticoid-induced enhancement of muscle protein degradation, which represents a relatively acute catabolic condition. This model does not fully recapitulate the complex and multifactorial nature of age-related muscle atrophy, which typically involves chronic low-grade inflammation, metabolic dysregulation, mitochondrial dysfunction, and long-term imbalance in muscle protein homeostasis. Therefore, future studies will employ naturally aged mice to further validate the present findings under physiological aging conditions, thereby improving the translational relevance and providing stronger evidence for nutritional intervention and functional food development. In addition, considering that CP used in the present study is a complex mixture of peptides, future work will focus on screening and identifying key bioactive peptide sequences from CP. These peptides will be subjected to structure–function relationship analysis and mechanistic validation experiments to clarify their direct molecular targets and underlying mechanisms of action. This approach will provide a stronger theoretical basis for precision nutritional intervention and the development of functional foods.

Methods

Preparation and identification of crocodile-derived bioactive peptides

The crocodile peptides were obtained from Hainan Huayan Collagen Technology Co. (Haikou, China). The detailed preparation and identification procedures were performed as previously described⁷.

Animals and experimental design

All animal experiments were carried out according to the protocol approved by the Animal Ethics Review Committee of Guangzhou Huateng Biomedical Co., Ltd. (Approval No.: B202410-2). Male C57BL/6j mice of specific pathogen-free (SPF) grade and comparable body weight were obtained from Zhuhai Best Biotechnology Co., Ltd. (Guangdong, China). After one week of adaptive feeding under standard feeding conditions (free access to food and water), the mice were randomly divided into two groups: control group (n = 10) and model group (n = 20). The control group was intraperitoneally injected with 0.9% normal saline solution every day, and the model group was intraperitoneally injected with 25 mg/kg body weight dexamethasone (DEX) solution every day for 14 days to establish an animal model. After successful modeling, the model animals were further divided into dexamethasone group (DEX, n = 10) and crocodile head derived peptide intervention group (CP, n = 10). In the subsequent experiments, DEX group and CP group continued daily intraperitoneal injection of dexamethasone. In addition to this program, the CP group was given a daily oral gavage of crocodile head derived peptide solution at a dose of 1000 mg/kg, and the intervention period was 43 days. At the end of the experiment, the mice were sacrificed by CO₂ asphyxiation, and bilateral tibialis anterior, gastrocnemius, quadriceps femoris, extensor digitorum longus and intestinal contents were dissected and collected. All tissue samples were stored at –80°C for subsequent analysis.

Strength and exercise test

All animal behavioral tests in this study, including limb grip strength test and running endurance test, were performed on the experimental animal platform of Guangzhou Huateng Biomedical Co., Ltd. (1) Limb grip strength test (ZS-ZL): Each mouse was placed on the grid of the dynamometer, and all four limbs naturally grasped the grid. Then the experimenter gently pulled the tail of the mouse, kept the animal ‘s body parallel to the force plate, and pulled back at a constant speed until the grip strength was lost. The maximum grip strength (in Newton, N) during each pull is recorded. Each mouse was measured 5 times continuously, and the highest value was taken as the final result. (2) Running endurance test (XR-PT-10B): All mice underwent a three-day adaptation period before the formal test, and 15 minutes of running training at 0°slope every day. The speed gradually increased from 10 m/min (day 1) to 15 m/min (day 2) and 18 m/min (day 3). During the formal test, the treadmill slope was set to 20°and the fixed speed was 18 m / min. The electric shock parameter was set to 0.8 m A, and the tolerance time was 3 s. Exhaustion is defined as the inability to continue running, which is manifested as continuous falls to the oscillating grid, continuous stay in the back 1/3 of the runway, prone position, shortness of breath. The time from the start of running to exhaustion (exhaustion time) and the cumulative number of electric shocks received during this process were recorded for each mouse.

Determination of spleen, cardiac, renal, bone, and hepatic indices

At the end of the experimental period, animals were fasted overnight and euthanized under anesthesia. The spleen, heart, bilateral kidneys, and liver were carefully excised immediately after sacrifice. Surrounding adipose and connective tissues were removed. The bilateral femurs and tibias were also isolated, and residual soft tissues were completely cleared. All tissues were gently blotted dry with filter paper to remove surface blood and moisture before weighing to obtain wet weight. The final body weight of each animal was recorded at the time of sacrifice.

Organ indices were calculated using the following formula: Organ index (%) = organ wet weight (g)/body weight (g)×100%.

The bone health index was determined based on the weights of the femur and tibia using the same calculation formula. All measurements were performed under standardized conditions to minimize experimental variability.

Body composition analysis

Animals were placed in an induction chamber connected to an isoflurane vaporizer. After anesthesia induction, they were transferred to a 3D dual-energy X-ray absorptiometry (DEXA) instrument (Model KUBTEC/PARAMETER 3D) with continuous isoflurane administration to maintain anesthesia. The instrument was activated with settings of 1000 μA current and 40 kV voltage. Following automatic calibration, body composition analysis was performed.

Histological analysis of the gastrocnemius muscle

Gastrocnemius muscle tissues were fixed in 4% paraformaldehyde, embedded in paraffin, sectioned, and subjected to hematoxylin-eosin (HE) staining for subsequent histomorphological observation.

Proteomics analysis

The protein content of the resulting extracts was assessed using a bicinchonolic acid (BCA) assay following the homogenization of gastrocnemius muscle tissue in a lysis buffer. Next, an aliquot of the protein solution was reduced with 5 mM dithiothreitol (DTT) for one hour at 37 °C. Alkylation was then carried out using 15 mM iodoacetamide (IAA) for 30 min at room temperature in the dark. After diluting the sample, we added a 1:50 enzyme-to-protein ratio of sequencing-grade trypsin for an overnight digestion at 37 °C. The resulting peptide mixture was desalted and purified using a C18 reversed-phase solid-phase extraction column. Prior to analysis, the purified peptides were lyophilized and stored at –80 °C. An Orbitrap Astral Zoom mass spectrometer was used for mass spectrometric detection after peptide separation was completed using an ultra-high-performance liquid chromatography (UHPLC) system (Thermo Scientific Vanquish Neo).

Molecular docking

The three-dimensional structures of crocodile-derived peptides were generated using RDKit and energy-minimized under the MMFF94 force field. Optimized structures were saved in SDF format and converted to PDBQT format using Open Babel. Crystal structures of Pdk4 (PDB ID: 2ZKJ) and Raptor (PDB ID: 8RCN) were obtained from the RCSB PDB database. Proteins were preprocessed in Chimera by removing water molecules and adding hydrogen atoms, followed by conversion to PDBQT format. Molecular docking was performed using AutoDock Vina. The docking grid centers were set at (12.200, 9.950, 12.980) Å for Pdk4 and (172.320, 165.690, 208.050) Å for Raptor. Binding affinity was evaluated using the Vina scoring function (kcal/mol), and the lowest-energy conformation was selected for interaction analysis.

Gut microbiota analysis

The FastPure Stool DNA Isolation Kit (MJYH, Shanghai, China) was used to extract total genomic DNA from the samples. After extraction, we used barcode-tagged primers to amplify the V3–V4 hypervariable regions of the bacterial 16S rRNA gene using polymerase chain reaction (PCR), and we then purified the resulting amplicons. The NEXTFLEX Rapid DNA-Seq Kit (Bioo Scientific, Austin, Texas, USA) was used to create the sequencing libraries using these refined products as the input. The raw data was first subjected to a quality control step using fast to eliminate low-quality reads after it had been sequenced. We then assembled the high-quality paired-end reads using FLASH. We used denoising algorithms like DADA2 or Deblur in the QIIME2 pipeline to process the data and produce a final amplicon sequence variant (ASV) table³⁶. We rarefied all samples to a uniform sequencing depth in order to standardize them prior to comparative analysis. Finally, we inferred metabolic potential from the 16S rRNA gene data using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2, version 2.2.0) for functional prediction³⁷.

Statistical analysis

The data were plotted using GraphPad Prism version 8.0.2. The results were represented as means ± SEM. For comparison between two groups, Student’s t-test was used for significance analysis. For comparisons among multiple groups, significance analysis was performed using one-way or two-way analysis of variance (ANOVA) tests. *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001.

Author contributions

Hao Song: Writing—original draft, Validation, Formal analysis, Data curation. Linyuezhi Yan: Methodology, Conceptualization. Zhipeng Yu: Writing—review & editing, Supervision, Project administration, Investigation. Wenzhu Zhao: Writing—review & editing, Supervision, Funding acquisition, Conceptualization.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to due to privacy or ethical restrictions but are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.