Caffeic acid phenethyl ester ameliorates high-fat diet-induced muscle textural deterioration in grass carp (Ctenopharyngodon idellus) by modulating adipose-muscle crosstalk via myostatin-taz signaling

Abstract

The rapid development of aquaculture faces significant challenges, particularly escalating fishmeal costs. Although increasing dietary fat content can partially offset this issue, long-term high-fat diet (HFD) feeding compromises aquatic product quality by deteriorating muscle texture. Previous research demonstrated that shifting adipose tissue expansion toward hyperplasia (rather than hypertrophy) enhances fish tolerance to HFD, but its impact on muscle quality remained unclear. Here, this study investigated how caffeic acid phenethyl ester (CAPE) ameliorates HFD-induced muscle quality deterioration in grass carp (Ctenopharyngodon idellus). A total of 270 fish (initial weight: 360.23 ± 2.07 g) were distributed into 15 cages. Five diets were assigned to triplicate cages for 8 weeks: control (5% lipid, Con), HFD (10% lipid), and HFD + CAPE (400, 800, or 1200 mg/kg) (54 fish per treatment, three cages of 18 fish). The results revealed that CAPE promoted adipocyte hyperplasia by upregulating the relative mRNA expression level of pparγ (P < 0.001) and reducing adipocyte area (P < 0.001), suppressed visceral fat deposition by decreasing the intraperitoneal fat index (P < 0.001), and alleviated adipocyte hypertrophy. Consequently, CAPE significantly improved growth performance by increasing the final body weight (P < 0.001), weight gain rate (P < 0.001), and specific growth rate (P < 0.001), as well as decreasing the feed conversion ratio (P < 0.001). Notably, CAPE did not directly affect myocyte proliferation or differentiation, but significantly reduced the relative mRNA expression of the mstn in adipose tissue (P = 0.001) by modulating adipose expansion patterns through promoted hyperplasia and suppressed hypertrophy. This alleviated Mstn-mediated inhibition of muscular Taz/Hippo signaling pathway, enhancing myocyte differentiation and improving texture parameters (chewiness, hardness, and springiness) in raw/cooked meat (P < 0.05). Present study identified Mstn in fish adipose tissue as a key adipokine mediating adipose-muscle crosstalk: excessive adipocyte hypertrophy impairs muscle quality through Mstn-mediated suppression of muscular Taz/Hippo signaling pathway. These findings elucidate a novel mechanism whereby HFD impairs fish muscle quality by altering adipose tissue expansion patterns, which subsequently modulates its secretory function, thereby confirming the existence of an “adipose-muscle axis”, and provide a theoretical foundation for developing functional feed additives targeting adipose expansion modes to mitigate HFD-induced muscle quality deterioration in aquaculture.

1. Introduction

Aquatic species constitute an essential source of nutritionally complete proteins within global human nutrition systems. As consumer reliance on aquaculture-derived food resources continues to escalate, the aquaculture industry is currently undergoing rapid development globally. However, the price of fishmeal, a primary protein source in aquafeeds, is rising due to shortages (Serra et al., 2024). This has driven the use of high-fat diets (HFD), which leveraging the “protein-sparing effect” to reduce production costs (Peng et al., 2019). Nevertheless, the tolerance of fish to HFD is limited. Research have shown that prolonged exposure to HFD can lead to muscle textural deterioration, characterized by altered hardness and springiness (Regost et al., 2001; Zhang et al., 2022), ultimately diminishing consumer acceptance (Xu et al., 2018).

Adipose tissue in fish acts as a central regulatory system governing energy homeostasis through secure lipid sequestration, effectively mitigating ectopic fat deposition and associated metabolic stress (Kruppa et al., 2023). This adaptive mechanism operates via dual morphological strategies: volumetric augmentation of mature adipocytes (hypertrophy) and the formation of new adipocytes (hyperplasia) (Hofwimmer et al., 2024; Rydén et al., 2014). Hypertrophic adipocytes cause lipid homeostasis dysfunction, while adipocytes hyperplasia plays an important role in preventing hypertrophic development and further maintaining the normal function of adipose tissue (Wu et al., 2022). Previous research has found that grass carp (Ctenopharyngodon idellus) have better tolerance and utilization capacity for HFD when hyperplastic adipogenesis predominantly mediates piscine adipose expansion (Wei et al., 2024). However, it is still unclear whether this nutritional regulation strategy that enhances fish tolerance to HFD can improve the negative impact of HFD on flesh quality.

Importantly, adipose tissue also acts as a significant secretory and endocrine organ, influencing various physiological processes (Trayhurn and Beattie, 2001). Numerous studies have demonstrated that adipokines could affect the development of muscle fibers (Trayhurn et al., 2011). Leptin is one such adipokine and is generally understood to indirectly regulate skeletal muscle development (O'Leary et al., 2018; Yan et al., 2013; Zhang and Chua, 2017). Globular adiponectin induces the differentiation and fusion of skeletal muscle cells (Fiaschi et al., 2009). Previous study found that the promotion of grass carp adipocyte hyperplasia reduces the relative expression levels of genes such as interleukin-1β (il-1β) and interleukin-8 (il-8) under HFD in adipose tissue (Wei et al., 2024), suggesting that the activation of adipocyte hyperplasia may induce changes in adipokine profiles. Therefore, it is speculated that stimulating adipocyte hyperplasia may improve flesh quality through adipokine-mediated muscle fiber development.

Caffeic acid phenethyl ester (CAPE) is a bioactive phenolic constituent derived from propolis (Lv et al., 2021). Experimental evidence indicates that dietary supplementation with CAPE in HFD-fed murine models ameliorates hepatic steatosis, suppresses systemic inflammation, and restores insulin sensitivity, underscoring its prophylactic and therapeutic potential against obesity-associated metabolic dysregulation during both post-obesity and early adipogenic phases (Kim et al., 2018). Caffeic acid phenethyl ester has been demonstrated to enhance the tolerance of juvenile grass carp to HFD, thereby attenuating pathological sequelae associated with lipotoxicity (Li et al., 2025). Building on these findings, this study investigates the effect of CAPE on improving muscle texture in HFD-fed grass carp by modulating adipose-muscle crosstalk. Specifically, this study aims to characterize CAPE's effects on adipocyte hyperplasia in adult grass carp, identify the adipokine associated with improved myofiber development, and explore the underlying mechanism.

2. Materials and methods

2.1. Animal ethics statement

All animal experiments complied with the ethical protocols approved by the Animal Care and Use Committee of Northwest A&F University (Yangling, Shaanxi, China) (NWAFU-DK-20220930).

2.2. Experimental set up and feeding management

All grass carp used in this study were obtained from Ankang Fisheries Experimental and Demonstration Station of Northwest A&F University (Ankang, Shaanxi, China). The experimental protocols involved the formulation of aquafeeds with graded lipid levels, namely a control diet containing 5% lipid and a HFD containing 10% lipid. These diets were supplemented with CAPE (98% purity, T90879, Shanghai Yuanye Bio-Technology Co., Ltd., Shanghai, China), a peroxisome proliferator-activated receptor γ (pparγ) agonistic compound (Li et al., 2025). The CAPE was incorporated into the HFD at four concentrations (0, 400, 800, and 1200 mg/kg), yielding five distinct dietary treatments: Con, HFD, HFD + C400, HFD + C800, and HFD + C1200, respectively. The ingredients and nutrient levels of experimental diets are showed in Table 1. All ingredients were finely ground through an 80-mesh sieve firstly. Secondly, according to the formulation, all ingredients were weighted and fully mixed by hand. Finally, the pellet machine (JL00SL01, Beijing Jinglai Construction Machinery Co., Ltd., Beijing, China) was used to produce the pellets with a diameter of 4 mm. The oven was used to dry all experimental diets at 40 °C until the moisture fell below 10%. All experimental diets were stored in a refrigerator at −20 °C until use.

Table 1.

Ingredients and nutrient levels of experimental diets (dry matter basis).

Items	Diets1
	Con	HFD	HFD + C400	HFD + C800	HFD + C1200
Ingredients, g/kg
Soybean meal	200.00	200.00	200.00	200.00	200.00
Rapeseed meal	140.00	140.00	140.00	140.00	140.00
Cotton meal	145.00	145.00	145.00	145.00	145.00
Flour	200.00	200.00	200.00	200.00	200.00
Distillers dried grains with soluble	50.00	50.00	50.00	50.00	50.00
Rice bran	100.00	100.00	100.00	100.00	100.00
Soybean oil	31.00	81.00	81.00	81.00	81.00
Bentonite	20.00	20.00	20.00	20.00	20.00
Choline chloride	5.00	5.00	5.00	5.00	5.00
Calcium phosphate primary	20.00	20.00	20.00	20.00	20.00
Premix2	10.00	10.00	10.00	10.00	10.00
Antioxidant	1.00	1.00	1.00	1.00	1.00
Rice husk powder	78.00	28.00	27.60	27.20	26.80
CAPE	0.00	0.00	0.40	0.80	1.20
Total	1000.00	1000.00	1000.00	1000.00	1000.00
Nutrient levels, %
Crude protein3	29.00	28.94	29.06	28.96	29.11
Crude lipid3	4.99	9.96	10.10	9.94	10.11
Crude fiber3	9.68	6.70	6.64	6.80	6.75
Gross energy3, kJ/g	17.22	18.26	18.20	18.47	18.50
Organic matter	80.21	80.94	80.37	80.70	81.10

CAPE = caffeic acid phenethyl ester.

¹Con = control diet containing 5% lipid; HFD, high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE.

²Each kilogram of premix contains: vitamin D 16.2 IU; vitamin A 67 IU; vitamin K₃ 340 mg; vitamin E 7.4 g; vitamin B₁ 670 mg; vitamin B₆ 800 mg; vitamin B₂ 1000 mg; vitamin B₁₂ 1.4 mg; D-pantothenic acid 2.65 g; vitamin C 10 g; nicotinamide 5.35 g; folic acid 330 mg; choline chloride 35 g; biotin 34 mg; inositol 8 g; Cu 350 mg; Fe 14 g; Zn 4 g; Mn 1.4 mg; Mg 10 g; I 40 mg; Co 30 mg; Se 35 mg.

³Nutrient levels were measured values.

This study utilized a total of 270 healthy fish with an average initial weight of 360.23 ± 2.07 g, which were uniformly distributed across 15 cages (triplicate cages per experimental diet) and housed in net pens with 18 grass carp per net pen (dimensions: 1.5 m × 1.5 m × 1.8 m) in a pond to allow for adaptation to the experimental conditions. The diets were randomly allocated to three replicate cages per treatment group, and the fish were fed at a daily ration of 2% of their body weight, administered three times a day (08:30, 13:00, and 17:00) for an 8-week period. During the feeding trial, water temperature was maintained at 24 to 28 °C, dissolved oxygen levels ranged from 7 to 11 mg/L, pH was between 7.5 and 8.0, ammonia nitrogen concentrations were 0.02 to 0.10 mg/L, and nitrite concentrations were 0.005 to 0.010 mg/L.

2.3. Sample collection

At the end of the feeding experiment, stop feeding for 24 h, all fish were euthanized through immersion in MS-222 at a concentration of 100 mg/L. The left muscle between the head and tail fin above the lateral line of six fish from each cage were sampled quickly using a sharp scalpel for texture measurement. And the right muscle between the head and dorsal fin above the lateral line and intraperitoneal adipose tissue of the same six fish was sampled, and tissue aliquots were either fixed in 4% paraformaldehyde for histology or flash-frozen in cryovials for storage at −80 °C.

Blood samples from each cage were pooled to yield one composite sample per cage. Three composite samples per treatment group (n = 3) were processed as follows: clotting at room temperature for 60 min, centrifugation at 1900 × g for 20 min at 4 °C to separate serum, and storage in three aliquots at −80 °C until analysis.

2.4. Growth performance

At the end of feeding trial, growth performance including final body weight (FBW), carcass weight (CW), specific growth rate (SGR), weight gain rate (WGR), feed conversion ratio (FCR), intraperitoneal fat index (IPFI), viscerosomatic index (VSI), and hepatosomatic index (HSI) were computed and statistically evaluated. Growth performance was calculated with the formulas as follows:

$$\text{CW}\left(\mathrm{g}\right)=\text{FBW}-\text{Viscera}\text{weight;}$$

$$\text{SGR}\left(\%/\mathrm{d}\right)=100\times \left[\ln \text{FBW}-\ln \text{Initialbody}\text{weight}\left(\text{IBW}\right)\right]/\text{Days;}$$

$$\text{WGR}(\%)=100\times \left[\left(\text{FBW}-\text{IBW}\right)/\text{IBW}\right]\text{;}$$

$$\text{FCR}=100\times \left[\text{Feed}\text{intake}\left(\text{FI}\right)\text{in}\text{dry}\text{matter}\right]/\left(\text{FBW}-\text{IBW}\right)\text{;}$$

$$\text{Condition}\text{factor}\left(\text{CF},\mathrm{g}/\text{cm}\right)=100\times \text{FBW}/\text{Final}\text{body}{\text{length}}^3\text{;}$$

$$\text{VSI}(\%)=100\times \text{Viscera}\text{weight}/\text{FBW;}$$

$$\text{IPFI}(\%)=100\times \text{Intraperitoneal}\text{fat}\text{weight}/\text{FBW;}$$

$$\text{HSI}(\%)=100\times \text{Hepatopancreas}\text{weight}/\text{FBW;}$$

$$\text{FI}\left(\mathrm{g}/\text{fish}\right)=\text{FI}\text{in}\text{dry}\text{matter}/\text{Number}\text{of}\text{fish.}$$

2.5. Proximate composition and histological analysis

Measurement of proximate composition were performed in accordance with AOAC (2023) protocols. Moisture content was measured by oven-drying to constant mass (method 934.01). Crude protein (N × 6.25) and crude lipid were quantified via Kjeldahl digestion (method 954.01) and ethyl ether extraction (method 2003.05), respectively. A muffle furnace (TMF-3100, EYELA Co., Tokyo, Japan) at 550 °C for 12 h was used to analyze ash content (method 942.05). Organic matter was calculated as 100% minus ash content in dry matter. Gross energy was determined using a Parr 1341 oxygen bomb calorimeter (Parr Instrument Co., Moline, IL, USA) following ISO 9831: 1998 (ISO, 1998). The method (method 962. 09) described in AOAC (2023) was used to determine the crude fiber.

In short, three fish were randomly sampled per cage for histological processing. A single paraffin section per fish was stained with hematoxylin-eosin (H&E). Nine sections per treatment group were examined, with whole-slide imaging performed for each. Histological characteristics were observed and photographed under a microscope (Imager A1m, Carl Zeiss AG, Oberkochen, BW, Germany). Adipocyte area, myofiber diameter, and density were determined. The detailed method was listed in previous study (Li et al., 2023; Wei et al., 2024).

2.6. Muscle texture feature detection

For texture analysis, three fish per cage were randomly sampled. Muscle tissue (1 cm³ cubes) was collected from the right dorso-cranial region above the lateral line using a sharp scalpel, texture parameters including hardness, cohesiveness, springiness, chewiness, gumminess, and adhesiveness were quantified using a texture analyzer (TMS-PRO, Food Technology Corporation, Sterling, VA, USA). The cooked muscle texture parameters were measured by cooking muscle tissue (1 cm³ cubes) in the boiling water for 15 s.

2.7. The hydroxyproline content in muscle tissues

The hydroxyproline content in fish muscle was quantified using a commercial assay kit (Ab222941, Abcam Trading (Shanghai) Co., Ltd., Shanghai, China). Muscle samples (100 mg) were homogenized in 100 μL distilled water, and 100 μL of the homogenate was mixed with 100 μL of 10 mol/L NaOH in a screw-cap tube. The mixture was hydrolyzed at 120 °C for 1 h. After cooling on ice, 100 μL of HCl was added, followed by centrifugation at 10,000 × g for 5 min. The supernatant was analyzed for hydroxyproline following the manufacturer's instructions.

2.8. Cytokine levels

Serum and culture medium myostatin (Mstn) concentrations were quantified using a fish-specific Mstn enzyme-linked immunosorbent assay (ELISA) kit (F0043-FB, Shanghai Fankew Biotechnology Co., Ltd., Shanghai, China) following the manufacturer's instructions. Briefly, serum samples were diluted 1:5 in sample dilution buffer. Standards (5–320 ng/L) and diluted samples (50 μL/well) were added to antibody-coated microplates and incubated at 37 °C for 30 min. After five washes with 1 × wash buffer (prepared by 30-fold dilution of 30 × concentrate), 50 μL of horseradish peroxidase (HRP)-conjugated detection antibody was added to each well (excluding blanks) and incubated at 37 °C for 30 min. Plates were washed five times, followed by addition of 50 μL each of chromogen substrates A and B. After 10 min of 37 °C incubation in darkness, the reaction was terminated with 50 μL stop solution. Absorbance was measured at 450 nm within 15 min using a microplate reader (HBS-ScanY, Nanjing DeTie Experimental Equipment Co., Ltd., Nanjing, Jiangsu, China). Myostatin concentrations were interpolated from the standard curve and multiplied by the dilution factor (5 × ).

2.9. Isolation of RNA, synthesis of cDNA and quantitative real-time PCR (qRT-PCR)

After collecting samples, the specific operations of RNA isolation, cDNA synthesis, and qRT-PCR were learned from previous literature in the laboratory (Wei et al., 2024). AG RNAex Pro Reagent (AG21102, Hunan Accurate Biotechnology Engineering Co., Ltd., Changsha, Hunan, China) was used to isolate tissue RNA. Approximately 100 mg of sample was homogenized with RNAex and added with a 1/5 volume of RNAex in chloroform, mixed, and left to stand for 5 min. The supernatant was obtained after centrifugation at 12,000 × g at 4 °C for 15 min. Subsequently, the sample was supplemented with a 1/2 volume of RNAex in isopropanol, fully mixed, and left to stand for 10 min. Sediment was obtained by centrifugation at 12,000 × g and 4 °C for 10 min, followed by the addition of pre-cooled (−20 °C) 80% ethanol. The sample was then centrifuged at 12,000 × g and 4 °C for 5 min; the supernatant was discarded, and then Rnase-free water (AG11012, Hunan Accurate Biotechnology Engineering Co., Ltd., Changsha, Hunan, China) was added to dissolve the RNA. Next, the total RNA was reverse transcribed into cDNA using a PrimeScript RT reagent kit (TaKaRa, Bio Biotechnology Engineering (Dalian) Co., Ltd., Dalian, Liaoning, China). The CFX96 Real-Time PCR Detection System (CFX96, Bio-Rad Laboratories, Heracles, CA, USA) was used for qRT-PCR. The qRT-PCR comprised an activation step at 95 °C for 30 s, and then 40 cycles at 95 °C for 15 s, followed by 60 °C for 15 s. The primer sequences are listed in Table 2. The expression levels of genes were determined by the comparative Ct method (2^−ΔΔCt) (Livak and Schmittgen, 2001). β-Actin was used as the internal reference and exhibited stable expression throughout the study. The full names of the abbreviations of all genes can be found in Table S1.

Table 2.

Specific primers used for quantitative real-time PCR (qRT-PCR).

Genes	Primer sequences (5′–3′)	GenBank No.	Product length, bp
myhc	F: GACGCTCATCACCACCAACC	EU414733.1	216
	R: TGCTCCTCACGCTGCTTCT
myog	F: TTACGAAGGCGGCGATAACTT	JQ793897.1	183
	R: TGGTGAGGAGACATGGACAGA
myod	F: TCCGATGCCTCCAGTCCG	XM_051884136.1	132
	R: CCGTGCGTCAGCATTTGG
myf5	F: GGAGAGCCGCCACTATGA	XM_051891124.1	257
	R: GCAGTCAACCATGCTTTCAG
mstn	F: CTGACGCCAAGTTCCACATACA	KM874826.1	151
	R: CGACTCTGCTTCAAGTTCTTCTCT
pparγ	F: CGCTCATCTCCTACGGTCAG	EU847421.1	136
	R: ATGTCGCTGTCGTCCAACTC
cebpα	F: ACCCACATACCACCACTCTCAACA	XM_051898884.1	193
	R: TTTCCCTCGATCGCCCATCTTCAT
dgat1b	F: TGGAGGAGCGTTTGCCTTAG	XM_051865801.1	180
	R: GACTGTATACCGCTGGACGC
fas	F: GTTGCAGAAGATGGGTCTACAG	XM_051913957.1	121
	R: TCTCCAGTAAGCGGCTAGAA
taz	F: GGTTGGGAGATGGCCTACAC	XM_051908858.1	106
	R: TCACTGAGGGTGCCATGTTC
last1	F: CGGACGATCCCAGACAACAA	XM_051875069.1	92
	R: AGGCATGAGTGATTTGCGGA
β-Actin	F: TCCACCTTCCAGCAGATGTGGATT	XM_051886219.1	114
	R: AGTTTGAGTCGGCGTGAAGTGGTA

2.10. Transcriptomic analysis

Comparative transcriptomic profiling was conducted on piscine lipid depots and myotomal tissues from HFD and HFD + CAPE experimental groups. Total RNA extraction, library preparation, and RNA sequencing were performed by Wuhan Metware Biotechnology Co., Ltd. (Wuhan, Hubei, China). Subsequent bioinformatics analysis was conducted using the Metware Cloud platform (https://cloud.metware.cn/). Differentially expressed genes (DEGs) were identified based on thresholds of |fold change| ≥ 1.5 and adjusted P < 0.05. Functional annotation of unigenes was achieved through alignment against Gene Ontology (GO) databases using Diamond Blastx software.

2.11. Cell culture and treatment

2.11.1. Adipocytes culture and treatment

The method of cultivating grass carp preadipocytes, the preparation steps of preadipocytes growth and induction differentiation culture medium refer to previous articles (Wei et al., 2024). The culture method of mature adicopytes of grass carp was referred to the previous article and slightly modified on its basis (Liu et al., 2014). After euthanasia, fish gills were removed to drain blood completely. Mesenteric fat tissue was aseptically harvested post-euthanasia and immersion-stabilized in sterile phosphate buffer solution (PBS) solution (pH 7.4) supplemented with 1% antibiotics. Sequential PBS rinses (3 × 5 min), then finely chopped and treated with 0.1% collagenase I (dissolved in PBS with 2% BSA) at 27 °C for 30 min to break down connective structures. The digested mixture was strained through a 200-μm filter to remove undissolved tissue. After centrifugation at 6500 × g for 10 min, the floating mature adipocytes (top layer) were pipetted into fresh tubes. To ensure purity, these adipocytes were washed three times by re-suspending in PBS and repeating the centrifugation step.

The differentiation medium was changed every 2 days. Cell samples were collected on differentiation days 0, 2, 4, 6, and 8 (divided into D0, D2, D4, D6, and D8 groups) and stored at −80 °C for subsequent analysis.

On d 0 of adipocyte differentiation, using CAPE with a concentration of 20 μmol/L or/and GW9662 (HY-16578, pparγ inhibitor, 99.87%, MedChemExpress, Shanghai, China) with a concentration of 10 μmol/L to treat adipocytes, changed the culture medium every 2 days, and collected cell samples on d 8 and stored at −80 °C for subsequent analysis, and the other part was stained with bodipy.

On the d 8 of adipocyte differentiation, mature adipocytes were treated with growth medium supplemented with 150 μmol/L palmitic acid (PA) and/or 10 μmol/L Liensinine (HY–N0484, Mstn inhibitor, 99.97%, MedChemExpress, Shanghai, China). After 48 h, they were replaced with growth medium after PBS lubrication and washing, and cell samples and culture medium were collected after 48 h and stored at −80 °C for subsequent analysis.

2.11.2. Myoblast culture and treatment

The operation steps of cultivating grass carp myoblast, the preparation steps of myoblast growth and induction differentiation culture medium refer to previous articles (Ji et al., 2024).

On d 0 of myoblast differentiation, cells were treated with differentiation medium supplemented with CAPE, changed the culture medium every 2 days, and collected cell samples on d 4 and stored at −80 °C for subsequent analysis, and the other part was stained with immunofluorescence.

On d 0 of myoblast differentiation, cells were treated with differentiation medium supplemented with 10 ng/μL Mstn and/or 15 μmol/L IBS008738 (HY-112821, Taz activator, 98.22%, MedChemExpress, Shanghai, China) (Yang et al., 2014) compounds for 48 h. Cell samples were collected and stored at −80 °C for subsequent analysis, with the remaining portion subjected to immunofluorescence staining.

The conditioned medium of adipocytes was mixed with fresh 4% equine serum M199 differentiation medium to obtain 2% equine serum M199 differentiation medium for treating myoblast (e.g., 50% adipocyte-conditioned medium + 50% fresh differentiation medium). After 48 h, one part of the cell samples was collected and stored at −80 °C for subsequent analysis, and the other part was stained with immunofluorescence.

2.12. Myoblast viability assays

Cytocompatibility profiling of CAPE on myoblast viability assays was conducted via mitochondrial dehydrogenase activity quantification using a commercial CCK-8 assay system (Beyotime Biotech, Shanghai, China). Briefly, methodological workflow involved administering 2 to 40 μmol/L CAPE concentration gradients in triplicate experimental replicates to myoblast cultured in 96-multiwell platforms. Post-48-h exposure, following 120-min chromogenic substrate conversion, colorimetric absorbance from the mitochondrial enzyme substrate reaction was measured at 450 nm.

2.13. 5-Ethynyl-2′-deoxyuridine (EdU) assays

Quantitative assessment of proliferative activity in grass carp myoblast was conducted using a thymidine analog incorporation assay (C0075L, BeyoClick EdU Kit, Shanghai Beyotime Biotechnology Co., Ltd., Shanghai, China). Briefly, experimental procedures included with 10 μmol/L EdU under conditions (27 °C, 2 h) in multiwell culture systems. Post-labeling protocols involved cryofixation (4% paraformaldehyde, 10 min), sequential PBS rinses, and membrane permeabilization (0.3% Triton X-100, 15 min). The reaction solution was applied in the dark (30 min, room temperature) to fluorescently label EdU-positive nuclei. Finally, nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Automated quantification of mitotically active nuclei was performed using validated image analysis algorithms (ImageJ v1.53) to determine proliferation indices across experiment.

2.14. Triglyceride (TG) content assays

Triglyceride content in adipocyte samples was quantified using the protocol adapted from previously published methodology (Wei et al., 2024). The TG levels in adipocyte were measured by using an intracellular TG assay kit (E1013, Beijing Pulilai Gene Technology Co., Ltd., Beijing, China). In the assay, 25 μL of lysis buffer was added to each well to prepare cell lysates on ice, and then lysing for 15 min at room temperature. The sample was heated for 10 min at 70 °C, and then centrifuged at 380 × g for 10 min. Finally, the supernatant was collected, and the absorbance was measured at 550 nm.

2.15. Immunofluorescence

The immunofluorescence staining procedure for myosin heavy chain (Myhc) in myoblasts referred to previous articles in the laboratory (Ji et al., 2024). Briefly, cells were washed, fixed, and incubated overnight with anti-Myhc antibody (MAB4470, R&D Systems RD, Shanghai Unimproved Biological Technology Co., Ltd., Shanghai, China). After washing, cells were incubated with a secondary antibody, washed again, and counterstained for nuclei. Samples were mounted and observed/imaged using an inverted fluorescence microscope (Model CKX53, Olympus, Tokyo, Japan).

2.16. Bodipy staining

Bodipy (505/515, Thermo Fisher Scientific, Waltham, MA, USA) was dissolved in dimethylsulfoxide (DMSO) to prepare a 5 mmol/L stock solution (aliquoted and stored at 4 °C). For cell staining, the stock was diluted in PBS to 5 μmol/L immediately before use. Cultured cells in 6-well plates were fixed with 4% paraformaldehyde (15 min, room temperature), then incubated with 500 μL Bodipy working solution (30 min, dark). After aspirating the solution, cells were washed three times with PBS (5 min per wash). Nuclei were counterstained with DAPI (5 μg/mL, 10 min), followed by three additional PBS washes. Lipid droplets (Bodipy, green) and nuclei (DAPI, blue) were visualized using a fluorescence microscope.

2.17. Data statistics and analysis

Statistical analyses considered each cage as an experimental unit. For all in vivo measurements, cage means were calculated and treated as independent experimental replicates (n = 3 biological replicates). All data in this study are expressed as mean and standard error of mean (SEM). Statistical analyses were performed using one-way ANOVA (Version 21.0, SPSS Inc., Chicago, IL, USA) after verifying normality and homogeneity of variance. Post-hoc Duncan's multiple range test was applied to assess group differences, with significance defined at P < 0.05. The mathematical representation for one-way ANOVA is given by:

$$X_{ij}=\mu +a_i+e_{ij}\text{,}$$

where μ is the grand mean, a_i is the treatment effect for group i, and e_ij is the random error term. The errors e_ij are mutually independent and distributed with mean 0 and variance σ².

3. Results

3.1. Caffeic acid phenethyl ester activates adipocyte hyperplasia in adult grass carp

In order to explore whether CAPE regulate adipogenesis in adult grass carp, this study treated preadipocytes isolated from adult grass carp with 20 μmol/L CAPE. Through bodipy staining, CAPE was intuitively observed to significantly enlarge lipid droplets in differentiated adipocytes, and it was found to activate adipocyte differentiation-related genes. This phenomenon would be reversed by pparγ inhibitor GW9662 (Fig. 1A–C; P < 0.05), indicating that CAPE activates adipocyte hyperplasia though up-regulation pparγ expression. Moreover, adipose tissue was performed H&E staining and measured the size of adipocytes using ImageJ software. Morphometric analysis revealed markedly reduced cellular dimensions and more evenly distributed within the HFD + C group compared to HFD (Fig. 1D and E; P < 0.001). Furthermore, the expression of adipocyte differentiation-related genes (cebpα and pparγ) increased significantly after adding CAPE (Fig. 1 F and G; P < 0.05). Combined with previous research in (Wei et al., 2024), concluding that CAPE and promotes adipocyte hyperplasia via activating pparγ expression in adult grass carp.

Fig. 1.

On d 8 of adipocyte differentiation, relevant indicators showed that caffeic acid phenethyl ester (CAPE) promotes adipocyte hyperplasia, either alone or in combination with the pparγ antagonist GW9662. (A) Bodipy staining. Scale bar: 50 μm. (B) The relative mRNA expression of adipocyte differentiation-related genes (pparγ, cebpα, dgat1b, and fas). (C) Triglyceride content was determined after 8 weeks of feeding. (D) Representative hematoxylin-eosin (H&E) stains of adipose tissue. Scale bar: 100 μm. (E) Area of adipocytes were measured using ImageJ software. (F) The relative mRNA expression of pparγ and cebpα of adipose tissue. (A–C): Control = adipocytes were cultured and treated in differentiation medium supplemented with an equal volume of dimethylsulfoxide (DMSO) (vehicle control); CAPE = adipocytes were cultured and treated in differentiation medium supplemented with 20 μmol/L CAPE; CAPE + GW9662 = adipocytes were cultured and treated in differentiation medium supplemented with 20 μmol/L CAPE and 10 μmol/L GW9662 (a pparγ antagonist). (D–F) Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE. Results are shown as means and standard error of mean (SEM) (n = 3). Different superscripts on the top of the bars indicate significant difference (P < 0.05).

3.2. Activating adipocyte hyperplasia improves the growth performance and muscle texture

After 8 weeks of cultivation experiments, while there was no significant difference in FI (P > 0.05), and found that the addition of CAPE activated adipocytes hyperplasia, FBW, CW, SGR, and WGR were significantly enhanced compared to the HFD group (P < 0.05), and FCR was significantly reduced compared to the HFD group (Table 3; P < 0.001). At the same time, CF, VSI, and IPFI were significantly reduced compared to the HFD group (Table 3; P < 0.05). Collectively, these interventions reduced the FCR while enhancing the tolerance of adult grass carp to HFD.

Table 3.

Effects of caffeic acid phenethyl ester (CAPE) on growth performance, feed utilization and body indices of grass carp.

Items	Groups1					SEM	P-value
	Con	HFD	HFD + C400	HFD + C800	HFD + C1200
FBW, g	874.18ab	837.82c	875.47ab	868.38b	880.77a	4.286	<0.001
CW, g	813.00a	777.17b	809.98a	808.44a	811.21a	4.053	<0.001
SGR, %/d	1.48a	1.40b	1.48a	1.47a	1.49a	0.009	<0.001
WGR, %	142.91a	132.11b	143.27a	140.91a	144.67a	1.284	<0.001
FCR	1.60ab	1.73a	1.60ab	1.62b	1.58c	0.015	<0.001
CF, g/cm3	1.90b	2.01a	1.94b	1.94b	1.94b	0.011	0.002
VSI, %	6.81b	9.76a	7.48b	6.90b	6.81b	0.328	<0.001
IPFI, %	2.39c	4.31a	3.54b	3.35b	3.27b	0.179	<0.001
HSI, %	2.89	2.77	2.81	2.76	2.50	0.053	0.307
FI, g/fish	915.59	916.75	914.99	916.46	915.95	0.430	0.776

FBW = final body weight; CW = carcass weight; SGR = specific growth rate; WGR = weight gain rate; FCR = feed conversion ratio; CF = condition factor; VSI = viscerosomatic index; IPFI = intraperitoneal fat index; HSI = hepatosomatic index; FI = feed intake; SEM = standard error of the mean.

Values in a row with different superscripts are significantly different (P < 0.05). Values are expressed as mean and SEM, n = 3.

¹Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE.

Through muscle texture testing, it was found that except for the gumminess and cohesiveness of raw meat and chewiness of cooked meat, all other texture indicators in the HFD group showed significant differences compared to the control group (Table 4; P < 0.05). In addition, the raw meat quality parameters of the CAPE-supplemented groups exhibited significant differences compared with the HFD group (Table 4; P < 0.05). Notably, no significant difference was detected in the cohesiveness of raw meat for the HFD + C400 group relative to the HFD group (P > 0.05), and found that the difference was more significant in the HFD + C1200 group. These indicators (spring, shear force, chewiness and hardness of raw meat, and spring, shear force, gumminess, cohesiveness and hardness of cooked meat) showed that the HFD + C groups improved the negative effects of HFD significantly (Table 4; P < 0.05). Transverse sections of muscle analysis revealed that HFD induced a significant increase in muscle fiber diameter, while CAPE supplementation eliminated this effect (Fig. 2A and Table 5; P < 0.001). Furthermore, the frequency distribution of fiber diameter in the CAPE group showed significant changes compared to the HFD group: CAPE effectively mitigated the HFD-induced increase in muscle fiber size (Table 5; P < 0.05).

Table 4.

Dietary caffeic acid phenethyl ester (CAPE) improved the muscle texture characteristics of grass carp.

Items	Groups1					SEM	P-value
	Con	HFD	HFD + C400	HFD + C800	HFD + C1200
Raw meat
Spring, mm	0.55c	0.51d	0.61b	0.64ab	0.66a	0.158	<0.001
Shear force, gf	468.29b	401.83c	439.78b	454.89b	524.00a	11.470	<0.001
Gumminess, gf	354.73d	358.49d	466.79c	530.63b	586.20a	25.110	<0.001
Chewiness, gf	293.42d	187.35e	322.52c	364.11b	397.73a	19.440	<0.001
Cohesiveness	0.30b	0.28b	0.30b	0.34a	0.34a	0.007	0.004
Hardness, gf	1305.73b	1161.33c	1496.20a	1530.71a	1534.03a	40.684	<0.001
Cooked meat
Spring, mm	0.62b	0.56d	0.67a	0.61bc	0.59c	0.010	<0.001
Shear force, gf	335.12b	278.43c	323.84b	383.02a	407.27a	12.774	<0.001
Gumminess, gf	220.47d	166.37e	256.56c	307.58b	423.84a	23.590	<0.001
Chewiness, gf	135.54d	116.16d	169.96c	207.87b	261.88a	14.223	<0.001
Cohesiveness	0.38a	0.34b	0.39a	0.39a	0.40a	0.007	<0.001
Hardness, gf	624.27d	527.32e	718.91c	821.37b	1134.26a	56.579	<0.001

SEM = standard error of the mean.

Values in a row with different superscripts are significantly different (P < 0.05). Values are expressed as mean and SEM, n = 18.

¹Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE.

Fig. 2.

Caffeic acid phenethyl ester (CAPE) enhances muscle fiber characteristics in high-fat diets (HFD) fed grass carp muscle tissue. (A) Transverse sections of the muscle of grass carp. Scale bar: 100 μm. (B) Hydroxyproline content in muscle. Results are shown as means and standard error of mean (SEM) (n = 3). Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE.

Table 5.

The myofiber diameter distribution and frequency distribution of muscle fibers diameter of grass carp.

Items	Groups1					SEM	P-value
	Con	HFD	HFD + C400	HFD + C800	HFD + C1200
Myofiber diameter, μm	70.96b	85.15a	58.85c	58.02cd	55.11d	3.014	<0.001
Frequency distribution of muscle fibers diameter2, %
Class 1	6.69c	2.14d	7.93bc	12.23a	9.42b	0.937	<0.001
Class 2	14.73b	10.82c	23.23a	20.33a	22.92a	1.370	<0.001
Class 3	17.72d	14.61e	26.79b	20.74c	32.88a	1.771	<0.001
Class 4	18.88b	16.92b	20.35ab	23.57a	16.15b	0.871	0.020
Class 5	19.02a	20.70a	14.10b	10.91b	12.74b	1.091	<0.001
Class 6	10.92b	19.70a	6.49c	10.63b	5.29c	1.417	<0.001
Class 7	9.71b	16.60a	1.84c	1.97c	1.00c	1.644	<0.001

CAPE = caffeic acid phenethyl ester; SEM = standard error of the mean.

Values in a row with different superscripts are significantly different (P < 0.05). Values are expressed as mean and SEM, n = 3.

¹Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE.

²Class 1: 0 μm < diameter ≤20 μm, class 2: 20 μm < diameter ≤40 μm, class 3: 40 μm < diameter ≤60 μm, class 4: 60 μm < diameter ≤80 μm, class 5: 80 μm < diameter ≤100 μm, class 6: 100 μm < diameter ≤120 μm, class 7: 120 μm < diameter.

3.3. Taz/Hippo signaling pathway in muscle was activated by dietary supplementation of CAPE

Due to the change in muscle quality, content of hydroxyproline (an indicator that can affect muscle quality) in the muscle was first detected. Interestingly, the result shown that no significant difference on its hydroxyproline content (Fig. 2B; P > 0.05), indicating that CAPE improved the muscle texture by promoting muscle fiber development.

To investigate the potential mechanism for improving muscle texture by addition of CAPE to HFD, this study transcriptome sequencing was performed. GO pathway profiling demonstrated Hippo signaling pathway dysregulation in HFD groups compared to HFD + C groups (Fig. 3A and B). Specifically, the HFD group displayed substantial upregulation of lats1 gene expression with downregulation of taz gene expression (Fig. 3C; P < 0.05). However, these changes were rescued following CAPE supplementation.

Fig. 3.

Caffeic acid phenethyl ester (CAPE) induces Taz/Hippo pathway under high-fat diets (HFD). (A) Heatmaps of the expression of related genes from Gene Ontology (GO) between HFD and HFD + C1200. (B) Top 20 GO enrichment analysis of the differential genes. (C) The relative mRNA expression of taz and last1 of adipose tissue. Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE. Results are shown as means and standard error of mean (SEM) (n = 3). Different superscripts on the top of the bars indicate significant difference (P < 0.05).

3.4. Caffeic acid phenethyl ester have no direct effect on myoblast proliferation and differentiation

Based on existing evidence that myoblast proliferation and differentiation impact muscle quality, present study conducted in vitro experiments to determine whether CAPE directly modulates these processes. Primary myoblasts from grass carp were treated with CAPE at varying concentrations. Initial results indicated cytotoxicity at CAPE concentrations exceeding 2 μmol/L (Fig. 4A; P < 0.001). Consequently, subsequent experiments utilized CAPE at sub-2 μmol/L doses. Proliferation assays (EdU labeling) revealed no significant effect of CAPE on myoblast proliferation (P > 0.05). Similarly, immunofluorescence staining of Myhc and quantification of differentiation-related genes (myf5, myog, and myhc) demonstrated no alterations in myoblast differentiation upon CAPE treatment (Fig. 4B–E; P > 0.05). These results indicated that CAPE promoted muscle fiber development through others pathway.

Fig. 4.

Caffeic acid phenethyl ester (CAPE) has no effect on the proliferation or differentiation of myoblasts in vitro. (A) Cytotoxicity of CAPE in myoblast. (B) Percentage of 5 - ethynyl-2′ - deoxyuridine (EdU) - positive cells/total cells. Scale bar: 100 μm. (C) The mRNA relative expression levels of differentiation marker genes myhc, myog, and myod in myoblasts. (D) 5 - Ethynyl-2′ - deoxyuridine (red) and 4′,6-diamidino-2-phenylindole (DAPI, nuclei, blue) staining. (E) Immunostaining of Myhc, Myhc (red) and DAPI (nuclei, blue). Scale bar: 100 μm. Results are shown as means and standard error of mean (SEM) (n = 3). Different superscripts on the top of the bars indicate significant difference (P < 0.05).

3.5. Adipocyte hypertrophy leads to an increase in the content of Mstn under HFD-fed condition

Given that CAPE changed the expansion pattern of adipose tissue, it is speculated that CAPE may ameliorate HFD-induced muscle textural deterioration by modulating adipose-muscle crosstalk. To screen the involved adipokine, adipose tissues from the HFD and HFD + C1200 group were selected for transcriptome sequencing. Gene enrichment analysis revealed a substantial association between differentially expressed genes and myoblasts differentiation (Fig. 5A and B), especially mstn. Therefore, the relative mRNA expression of mstn in adipose tissue and the content of Mstn in serum were detected. The results showed that adipocyte hypertrophy elevated serum Mstn levels (P = 0.001); concurrently, HFD feeding in grass carp induced significant upregulation of mstn relative mRNA expression (P < 0.001), whereas dietary supplementation with CAPE to activate adipocyte hyperplasia ameliorated this effect. In addition, the relative mRNA expression of mstn was undetectable during adipocyte differentiation (date not show), indicating adipocyte hypertrophy contributed to Mstn content in serum (Fig. 5C and D). In vitro experiments revealed that PA-induced adipocyte hypertrophy significantly increased both secreted Mstn levels in the conditioned medium and intracellular mstn relative mRNA expression (P < 0.05). Notably, pharmacological intervention using Mstn-specific inhibitors completely abolished these PA-induced effects (Fig. 6A–C). Hence, Mstn is an adipokine secreted from hypertrophic adipocyte.

Fig. 5.

Caffeic acid phenethyl ester (CAPE) reduces serum Mstn levels and the relative mRNA expression of mstn under HFD. (A) Volcano map of differential expression mRNAs. (B) Gene Ontology (GO) analyzes (biological processes) the number of genes associated. (C) Content of Mstn in serum in culture experiment. (D) The relative mRNA expression of mstn in mature adipocytes in vivo. Con = control diet containing 5% lipid; HFD = high-fat diet containing 10% lipid; HFD + C400 = the high-fat diet containing 400 mg/kg CAPE; HFD + C800 = the high-fat diet containing 800 mg/kg CAPE; HFD + C1200 = the high-fat diet containing 1200 mg/kg CAPE. Results are shown as means and standard error of mean (SEM) (n = 3). Different superscripts on the top of the bars indicate significant difference (P < 0.05). FDR = false discovery rate.

Fig. 6.

The Mstn secreted from hypertrophic adipocyte reduces myoblast differentiation through inhibiting Hippo pathway effector Taz. (A) Schematic drawing of the experimental procedures. (B) The mRNA expression levels of mstn in adipocytes cultured with different compound treatment conditions. (C) Content of Mstn in adipocytes cultured medium with different compound treatment conditions. (D) Immunostaining of Myhc under different compound treatment conditions, Myhc (red) and 4′,6-diamidino-2-phenylindole (DAPI, nuclei, blue). Scale bar: 100 μm. (E) The mRNA expression levels of differentiation marker genes myhc, myog, and myod in myoblasts cultured with different compound treatment conditions. (F) Immunostaining of Myhc under different medium and compound treatment conditions, Myhc (red) and DAPI (nuclei, blue). Scale bar: 100 μm. (G) The mRNA expression levels of differentiation marker genes myhc, myog, and myod in myoblasts cultured with different medium and compound treatment. (B and C) Con = adipocytes were cultured and treated in growth medium supplemented with an equal volume of dimethylsulfoxide (DMSO) as the vehicle control; PA = adipocytes were cultured and treated in growth medium supplemented with 150 μmol/L palmitic acid (PA); PA + Lien = adipocytes were cultured and treated in growth medium supplemented with 150 μmol/L PA and 10 μmol/L Liensinine (Lien). (D and E) Control = myoblasts were cultured and treated in differentiation medium supplemented with an equal volume of DMSO (vehicle control); Mstn = myoblasts were cultured and treated in differentiation medium supplemented with 10 ng/μL Mstn; Mstn + IBS008738 = myoblasts were cultured and treated in differentiation medium supplemented with 10 ng/μL Mstn and 15 μmol/L IBS008738. (F and G) Con - CM = myoblasts were cultured in a medium consisting of 50% conditioned medium from control adipocytes and 50% fresh myoblast differentiation medium; PA - CM = myoblasts were cultured in a medium consisting of 50% conditioned medium from PA-treated adipocytes and 50% fresh myoblast differentiation medium; PA + Lien - CM = myoblasts were cultured in a medium consisting of 50% conditioned medium from PA + Lien-treated adipocytes and 50% fresh myoblast differentiation medium; PA - CM + IBS008738 = myoblasts were cultured in a medium consisting of 50% PA adipocyte-conditioned medium, 50% fresh myoblast differentiation medium, and 15 μmol/L IBS008738. Values are means and standard error of mean (SEM) (n = 3). Different superscripts mean significant differences among groups (P < 0.05).

3.6. Hypertrophic adipocyte-derived Mstn inhibits myoblast differentiation through Taz/Hippo signaling pathway

To investigate whether Mstn secreted from hypertrophic adipocyte leaded to muscle textural deterioration via the Taz/Hippo signaling pathway, myoblasts was treated with recombinant protein Mstn and/or a Taz activator. Immunofluorescence staining and the mRNA levels changes of myoblast differentiation-related genes including myod, myog, and myhc (P < 0.05) demonstrated that recombinant protein Mstn suppressed myoblast differentiation, whereas Taz activator treatment rescued recombinant protein Mstn-induced suppression of myoblast differentiation (Fig. 6D and E).

Previous results have demonstrated that CAPE inhibited adipocyte hypertrophy by promoting adipocyte hyperplasia under HFD condition (Fig. 1). In order to investigate whether CAPE improved muscle texture via reducing the secretion of Mstn from adipose tissue, myoblasts was treated with hypertrophic adipocyte-conditioned medium (Fig. 6A). The results revealed that the medium containing high content of Mstn inhibited myoblast differentiation, whereas lowered the Mstn concentrations in the medium enhanced differentiation capacity of myoblast. Notably, the suppressed effect under high Mstn conditions could be rescued by Taz activators (Fig. 6F). Consistent results were obtained through detection of myoblast differentiation-related genes, namely that high content of Mstn inhibited the relative mRNA expression levels of myod, myog, and myhc (P < 0.05), but the relative mRNA expression levels of myod, myog, and myhc (P < 0.05) significantly increased after inhibiting Mstn or activating Taz (Fig. 6G).

4. Discussion

Along with socioeconomic development, there has been a marked escalation in consumer demand regarding the fish flesh quality. This trend contrasts sharply with the emerging challenge of HFD-induced muscle textural deterioration in farmed fish species. Adipose tissue, recognized as a primary energy-responsive organ under HFD (Wei et al., 2024), serves as a critical reservoir for triglycerides through two distinct expansion mechanisms: adipocyte hypertrophy and adipocyte hyperplasia (Jia et al., 2024). Previous investigations have shown that adipocyte hyperplasia-dominant adipose expansion enhances HFD tolerance in juvenile grass carp (Li et al., 2025; Wei et al., 2024). Building on these findings, the present study investigated the effects of CAPE on HFD tolerance and muscle textural properties by promoting adipocyte hyperplasia in adult grass carp. This study findings establish modulating adipose expansion patterns as a viable approach to counteract HFD-induced muscle quality decline, providing new insights into developing intervention strategy for improving fish flesh quality.

At present, the effect of CAPE on adipocyte differentiation is controversial. Studies have shown that CAPE blocks adipogenesis of 3T3-L1 preadipocytes (Shin et al., 2014), while some studies have demonstrated that dietary supplementation with CAPE activates pparγ in mice, driving adipose tissue remodeling and ameliorating metabolic disorders through this pivotal regulatory pathway (Albakri et al., 2014; Kim et al., 2018; Sakers et al., 2022). In this study, the results showed that CAPE promotes adipogenesis of grass carp preadipocytes in vivo and in vitro, supporting the conclusion of the latter. In addition, CAPE has an improvement effect on the growth performance of grass carp fed with HFD, indicating that CAPE increased the tolerance of adult grass carp to HFD by promoting adipocyte hyperplasia. The newly generated adipocytes exhibited enhanced capacity to uptake circulating free fatty acid (FFA), consequently reducing ectopic lipid deposition (Liu et al., 2023). These findings elucidate how hyperplastic adipogenesis induction enhances lipid homeostasis capacity in adult grass carp under HFD. Comparative analysis across developmental stages in juvenile grass carp demonstrated that hyperplastic adipogenesis induction enhances lipid homeostasis capacity under HFD (Li et al., 2025; Wei et al., 2024), with this regulatory mechanism operating independently of ontogenetic phases in grass carp.

Muscle texture serves as an indicator for evaluating the fish flesh quality (Wang et al., 2021; Zhang et al., 2022). According to the literature, muscle texture is primarily associated with two factors. The first is the collagen content, which is estimated through hydroxyproline quantification in muscle (Hofman et al., 2011). In addition, the muscle texture is negatively correlated with the size of muscle fibers in mammals, and similar studies in fish have also proved this viewpoint (Periago et al., 2005). The diameter and size of muscle fibers are influenced by myoblast differentiation (Huang et al., 2021). Hence, the second factor is the degree of myoblast differentiation (Hatae et al., 1990). In this study, the hydroxyproline content showed no change in muscle tissue following supplementation CAPE, indicating that CAPE enhanced muscle texture via increasing myoblast differentiation instead of the hydroxyproline content. Interestingly, CAPE has no effect on myoblast differentiation. These results suggest that the promoting effect of CAPE on myoblast differentiation is attributed to the change of adipose tissue expansion pattern. Previous studies have shown that adipose tissue in mice can regulate skeletal muscle function through the secretion on insulin like growth factor 2 (Igf2) (Clark et al., 2015), while interleukin-6 secreted by skeletal muscle can modulate adipose tissue in pigs (Xuan et al., 2022), highlighting the dynamic interaction between adipose tissue and muscle (Shan et al., 2013). In fact, research in rat and fish has shown that adipocyte hypertrophy caused by impaired adipocyte hyperplasia is a pathological characteristic of obesity (Sakers et al., 2022; Wei et al., 2024; Zhao et al., 2021) and diet-induced obesity decreased muscle fiber diameter (Ji et al., 2025; Sishi et al., 2011). Hence, the pattern of adipose tissue expansion is an important factor that affects the muscle texture of fish fed with HFD.

How does adipocyte hypertrophy affect myoblast differentiation in grass carp fed with HFD? Adipose tissue is not only recognized as a primary energy reservoir, but also an active endocrine organ in vertebrates (Ahima and Flier, 2004). Excessive energy intake, such as HFD, induces adipose expansion predominantly through hypertrophy. This pathological expansion pattern is typically accompanied by alteration of adipokines (Skurk et al., 2007). In this study, Mstn was identified as a protein secreted from hypertrophic adipocyte induced by HFD. Myostatin, a myokine predominantly secreted by muscle tissue, plays a pivotal role in muscle growth regulation. Studies have demonstrated it is an inhibitor of myogenic differentiation (Ríos et al., 2002). In this study, Mstn also inhibits myoblast differentiation of grass carp, suggesting that the role of Mstn in regulating muscle development is very conserved in vertebrates. Furthermore, recent evidence has redefined Mstn as an adipokine (Wang et al., 2024). Experimental data in murine models revealed that myostatin secreted by brown adipose tissue modulates skeletal muscle function through interorgan crosstalk (Kong et al., 2018). The result of the current study aligned with these findings, highlighting the idea that Mstn is an adipokine. Studies have revealed an inverse correlation between taz expression and Mstn levels (Sun et al., 2019). Previous research reported that Taz is a Hippo signaling effector (Wu et al., 2003), and governs the differentiation of myoblast (Wang et al., 2020). This study demonstrated that the inhibitory effect of Mstn released during adipocyte hypertrophy on myoblast differentiation can be ameliorated through activation of the taz, which further proved the relationship between Taz and Mstn. Taken together, hypertrophic adipocyte inhibited myoblast differentiation by secreting Mstn in grass carp fed with HFD. And CAPE altered the pattern of adipose tissue expansion by promoting adipocyte hyperplasia, which reduces Mstn secretion from adipose tissue, then promotes muscle fiber development through activating Taz/Hippo signaling pathway and ultimately improved HFD-induced muscle textural deterioration in grass carp.

5. Conclusion

In general, HFD induced the expansion of adipose tissue in the manner of adipocyte hypertrophy, and this increased secretion of adipokine Mstn. Then, Mstn decreased myoblast differentiation via inhibiting Hippo signaling effector Taz, leading to muscle textural deterioration in adult grass carp fed with HFD. On the contrary, CAPE increased the tolerance of adult grass carp to HFD by promoting adipocyte hyperplasia and ameliorated HFD-induced muscle textural deterioration by decreasing the secretion of adipokine Mstn. Experimental findings demonstrate unprecedented level of inter-organ crosstalk between adipose tissue and muscle, involving the secreted protein Mstn and Taz/Hippo signaling pathway, and showed that the unhealthy expansion of adipose tissue is one of the reasons for the negative effects of HFD on muscle texture in grass carp fed with HFD. Altogether, altering adipose expansion pattern by promoting adipocyte hyperplasia is a viable approach to counteract HFD-induced muscle quality decline in fish.

Credit Author Statement

Handong Li: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Formal analysis, Data curation, Conceptualization. Zhiqi Tian: Visualization, Project administration, Investigation, Formal analysis, Data curation. Mingkui Wei: Software, Investigation, Data curation. Li Tang: Software, Formal analysis. Hong Ji: Writing – review & editing. Gen He: Writing – review & editing, Funding acquisition, Conceptualization. Jian Sun: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Declaration of competing interest

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, and there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the content of this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: