18β-Glycyrrhetinic acid mitigates adverse effects of high-fat diets on liver fibrosis in largemouth bass (Micropterus salmoides) by reducing mitochondrial Ca²⁺ level

Abstract

Appropriately increasing the level of fat in feed exhibits potential protein-saving effects, but high levels of fat in the feed can cause liver injury in fish of largemouth bass. 18β-Glycyrrhetinic acid (GA), a natural extract from Glycyrrhiza, has been shown to enhance growth performance and improve intestinal health in animals. Thus, the present study was to investigate the effects of dietary GA on high-fat-induced liver fibrosis and the potential mechanisms in juvenile largemouth bass. The experiment comprised five different diets, including normal control (NC) diet (NC group), high-fat (HF) diet (HF group), HF diet supplemented with 0.5 mg/kg GA (HFL group), HF diet supplemented with 1.0 mg/kg GA (HFM group), and HF diet supplemented with 1.5 mg/kg GA (HFH group). A total of 750 healthy largemouth bass with an initial average weight of 17.39 ± 0.09 g were selected and randomly divided into five treatment groups, with three replicates per group and 50 fish per replicate. The feeding experiment was conducted for 11 weeks. In the liver, compared to NC, HF diet led to a marked reduction of the mitochondria-associated membranes (MAMs) gap, resulting in mitochondrial Ca²⁺ overload, augmented liver Ca²⁺ transport proteins expressions and reactive oxygen species (ROS) content (P < 0.05), decreased mitochondrial membrane potential (MMP) and adenosine 5′-triphosphate (ATP) content in the liver (P < 0.05). Thus, the activation of transforming growth factorβ1 (TGFβ1)-SMAD family member 2 (Smad2/3) signaling pathway resulted in the upregulation of pro-fibrogenic markers (α-Sma, collagen I, fibrontein, and Mmp9) mRNA expressions (P < 0.05). However, GA could alleviate this phenomenon. Through virtual docking analysis showed that the docking energy between GA and Sirtuin 3 (SIRT3) was −9.0 kcal/mol, and the binding effect between the two was stable. Such proposed mechanism was further corroborated by the SIRT3 siRNA transfection cell experiment, indicating that GA may potentially influence the mitochondrial calcium uniporter (MCU) expression by targeting SIRT3, thereby reducing mitochondrial Ca²⁺ content. Together, this study elucidates the essential role of GA in ameliorating liver lipid accumulation and improving liver health, highlighting that GA improves liver fibrosis and mitigates mitochondrial Ca²⁺ overload-mediated liver injury. Consequently, GA holds promise as a potential health-functional additive or therapeutic agent for liver fibrosis.

1. Introduction

The largemouth bass (Micropterus salmoides) has become a cornerstone species in China's freshwater aquaculture industry due to its rapid growth rate, superior meat quality, absence of intermuscular bones, and significant nutritional benefits (Fang et al., 2024; Yang et al., 2024). Its annual production has surpassed 800,000 metric tons, with widespread farming across the country (Affairs, 2024). As a strict carnivorous fish, the largemouth bass exhibits low metabolic efficiency for carbohydrates, necessitating reliance on lipid utilization to achieve a protein-sparing effect. Commercial feeds for this species currently maintain a lipid level of approximately 15%. However, prolonged high-fat (HF) diet feeding poses substantial health risks, for example, excessive liver lipid deposition triggers oxidative stress and organ damage (Li et al., 2020; Ling et al., 2019), further leading to metabolic disorders and suppressed growth performance (Cao et al., 2024). The liver, a central regulatory organ for lipid metabolism in fish (Ando and Mori, 1993), demonstrates a strong positive correlation between its lipid content and dietary fat intake, as evidenced in hybrid striped bass (Morone chrysops × Morone saxatilis) (Gaylord and Gatlin, 2000). Under intensive farming conditions, chronic HF diet may predispose fish to fatty liver disease, with pathological progression potentially escalating to liver fibrosis a severe threat to both fish health and aquaculture profitability. Consequently, elucidating the molecular mechanisms underlying HF diet-induced liver fibrosis holds critical scientific and practical significance for optimizing aquaculture strategies.

Largemouth bass is indigenous to North America and has emerged as a significant economic commodity in Asia, particularly in China (Fang et al., 2024; Yang et al., 2024). Being a carnivorous species, which uses sugar less efficiently, it utilizes fat to save protein. Nevertheless, prolonged feeding of a HF diet led to excessive lipid deposition and induced oxidative stress and organ injury, resulting in metabolic disorders and negatively impacting fish growth (Li et al., 2020; Ling et al., 2019). Previous research has shown that HF diet can suppress the growth performance of largemouth bass (Table S1) (Cao et al., 2024). The liver is the main regulatory organ for lipid accumulation in fish (Ando and Mori, 1993). Studies have shown that there is a positive correlation between liver fat content and dietary fat content of hybrid striped bass (Gaylord and Gatlin, 2000). The fatty liver caused by long-term feeding of HF diet in farmed fish, and further deterioration leads to liver fibrosis. Following 90 days of continuous HF diet feeding, significant collagen fiber deposition accompanied by fibrotic lesions was observed in the liver tissues of zebrafish (Li et al., 2017). Therefore, in order to improve the effect of HF diet on liver fibrosis of fish in aquaculture, it is necessary to conduct in-depth research.

Liver fibrosis is a compensatory response seen in wound-healing, characterized by the abnormal accumulation of extracellular matrix (ECM) and the formation of fibrous scars (Roehlen et al., 2020). Extracellular matrix and fibrous scars disrupt the normal architecture of the liver, resulting in hepatocyte loss, ultimately leading to liver dysfunction and eventual failure (Aydin and Akcali, 2018). The hallmark features of liver fibrosis include excessive collagen deposition, heightened inflammation, and hepatocyte injury (Aydin and Akcali, 2018). Both animal experimental models and clinical samples have demonstrated that liver fibrosis is reversible, and the removal of fibrosis-inducing agents can mitigate fibrosis, provided the condition has not progressed to cirrhosis (Troeger et al., 2012). Earlier studies have highlighted that HF diet could trigger liver metabolic disruptions and liver inflammation (Liu et al., 2021) and mitochondrial Ca²⁺ overload (Chen et al., 2022), which are primary drivers of liver fibrosis (Acharya et al., 2021). The activation of stellate cells, key fibrogenic cells, plays a crucial role in liver fibrogenesis (Hernandez-Gea and Friedman, 2011; Puche et al., 2013). Stimulated by factors like transforming growth factor β1 (TGFβ1), stellate cells activation leads to the development of a myofibroblast-like phenotype characterized by a reduction in lipid droplets and an increase in the expression of α-smooth muscle actin (α-Sma), ultimately contributing to the enhanced accumulation and production of ECM and collagen (Balta et al., 2015). The process of liver fibrosis can be reversible (Friedman and Bansal, 2006), underscoring the need for potent anti-fibrotic agents for effective liver fibrosis treatment.

Under the context of healthy aquaculture practices for fish, the pursuit of sustainable and efficient aquaculture methods is increasingly crucial. With the expansion of farming scales, fish liver diseases, especially liver fibrosis, seriously threaten fish health and production capacity. These diseases not only increase mortality but also degrade fish product quality, posing risks to human health via the food chain. Currently, traditional therapeutic agents in fish health management often have limitations. For example, some drugs may have side effects on fish growth and development or cause environmental pollution due to residues. So, there is an urgent need to develop safe and effective alternative treatments. 18β-Glycyrrhetinic acid (GA) is an active metabolite derived from glycyrrhizic acid, sourced from Chinese herbal licorice root (Akao et al., 1991). The GA has demonstrated various beneficial pharmacological activities, including immunomodulatory, antiulcerative, hepatoprotective, antiviral, and antitumor effects (Asl and Hosseinzadeh, 2010; Rossum et al., 1998; Shiota et al., 1999). Notably, GA is commonly used as a hepatoprotective agent in clinical practice (Gao et al., 2015). Its effects encompass reduction in liver cell necrosis, lowered serum alanine aminotransferase (ALT) activity, fibrosis inhibition, and promotion of liver cell regeneration (van Rossum et al., 1998). Previous studies on blunt snout bream (Megalobrama amblycephala) and channel catfish (Ictalurus punctatus) found that GA had a positive effect on reducing liver lipid deposition (Abasubong et al., 2021; Desouky et al., 2020; Jiang et al., 2018), which was beneficial for the prevention of liver fibrosis. Previous studies have shown that GA mitigated the adverse effects caused by the HF diet via boosting anti-inflammatory properties and fortifying intestinal barrier protection functions in largemouth bass (Cao et al., 2024). However, there are still many deficiencies in the existing research. Currently, most of the research focuses on the effects of GA on the specific physiological functions of fish, and there is limited understanding of its mechanism of action and regulatory pathways, especially in the treatment of liver fibrosis. This has restricted the full application of GA in the healthy aquaculture of fish. This study aims to comprehensively explore the mechanisms and regulatory pathways of GA in treating liver fibrosis in fish, to provide a more in-depth theoretical basis for the application of GA in the healthy aquaculture of fish.

Mitochondria are not only the "power plants" of cells but also important Ca²⁺ reservoirs within cells. In fact, mitochondria are involved in the dynamic regulation of intracellular Ca²⁺, and Ca²⁺ are crucial to the function of mitochondria. The efficient uptake of Ca²⁺ by mitochondria depends on the coupling of mitochondrial-endoplasmic reticulum structures, referred to as mitochondria-associated membranes (MAMs) (Bravo-Sagua et al., 2017; Means and Katz, 2022). Mitochondria-associated membranes, a subcellular structure discovered in recent years located between the mitochondria and the endoplasmic reticulum, play a crucial role in the exchange of materials and the transmission of information between the two organelles (Herrera-Cruz and Simmen, 2017). The endoplasmic reticulum serves as the largest calcium storage within the cell and is also a significant source of Ca²⁺ for the mitochondria (Marchi et al., 2018). The Ca²⁺ released from the endoplasmic reticulum can be absorbed into the mitochondria through MAMs. Apart from yeast, in the vast majority of vertebrates and eukaryotes, the uptake of mitochondrial Ca²⁺ is regulated by mitochondrial calcium uniporter (MCU) which is a Ca²⁺ gated channel located on the mitochondrial inner membrane, and its channel activity is a prerequisite for the uptake of mitochondrial Ca²⁺ (Kamer and Mootha, 2015). Abnormalities in the structure and function of MAMs can also lead to an imbalance in mitochondrial Ca²⁺ homeostasis (Szabadkai et al., 2006). Research has found that mitochondria can produce more reactive oxygen species (ROS) when overloaded with Ca²⁺, which can cause severe damage to the mitochondria when it exceeds their metabolic capacity (Feissner Rf, 2009). This leads to the opening of the mitochondrial permeability transition pore, promoting apoptosis and necrosis, resulting in liver fibrosis (Thomas et al., 2014).

Adenosine monophosphate-activated protein kinase (AMPK)/peroxisome proliferator-activated receptor gamma coactivator 1α (PGC1α) is an intracellular energy sensor and regulator, activated during energy depletion and high demand for energy supply. Sirtuin 3 (SIRT3), as a downstream target of the AMPK/PGC1α signaling pathway, plays a crucial role in the regulation of mitochondrial biogenesis and oxidative stress (Park et al., 2011). Here, a model of liver fibrosis induced by a HF diet was established in largemouth bass (Micropterus salmoides). The results demonstrated that GA effectively inhibited liver fibrosis. Furthermore, molecular docking and in vitro experiments revealed that GA may regulate SIRT3 to influence mitochondrial function, thereby exerting an inhibitory effect on HF diet-induced liver fibrosis.

2. Materials and methods

2.1. Animal ethics statement

Experimental design and procedures in this study were subjected to review and approval by the Animal Ethical Committee of Sichuan Agricultural University (approval number: DKY-2021202027).

2.2. Materials and reagents

The GA, with a purity of ≥97.0%, was procured from Sigma–Aldrich (Shanghai) Trading Co., Ltd. (G10105, Shanghai, China). Linoleic acid (LA) was also obtained from Sigma–Aldrich (Shanghai) Trading Co., Ltd. (L1012, Shanghai, China).

2.3. Experimental design and diets

Table 1 presents the feed formulation and nutritional composition of the experimental diets. The diets encompassed four protein sources (fermented soybean meal, chicken powder, gluten meal, and fish meal), one lipid source (soybean oil) and two carbohydrate sources (whole meal flour and cassava starch). All fish were divided into five groups as follows: NC group (normal control, 9.05% crude lipid), HF group (HF diet, 16.29% crude lipid), HFL group (HF diet supplemented with 0.5 mg/kg GA, 16.25% crude lipid), HFM group (HF diet supplemented with 1.0 mg/kg GA, 16.44% crude lipid), and HFH group (HF diet supplemented with 1.5 mg/kg GA, 16.44% crude lipid). All dry ingredients were processed through a 60-mesh screen. The diets were prepared by mixing the dry ingredients with the oil using a mixer. Subsequently, each diet was extruded using a twin-screw extruder (MY-165, Nanjing Jieya Extrusion Equipment Co., Ltd., Nanjing, Jiangsu, China) with a 2-mm die. The resulting floating extruded pellets were air-dried and stored at 4 °C in plastic bags until use.

Table 1.

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

Items	Diets1
	NC	HF	HFL	HFM	HFH
Ingredients
Soybean oil	6.00	12.00	12.00	12.00	12.00
Bentonite	7.00	1.00	0.00	0.00	0.00
Fish meal	43.00	43.00	43.00	43.00	43.00
Chicken powder	16.00	16.00	16.00	16.00	16.00
Gluten meal	7.00	7.00	7.00	7.00	7.00
Fermented soybean meal	5.00	5.00	5.00	5.00	5.00
Whole meal flour	6.00	6.00	6.00	6.00	6.00
Cassava starch	5.70	5.70	5.70	5.70	5.70
Lysine	0.20	0.20	0.20	0.20	0.20
methionine	0.20	0.20	0.20	0.20	0.20
Calcium biphosphate	1.00	1.00	1.00	1.00	1.00
GA premix2	0.00	0.00	1.00	1.00	1.00
Choline chloride	0.40	0.40	0.40	0.40	0.40
Premix3	2.50	2.50	2.50	2.50	2.50
Total	100.00	100.00	100.00	100.00	100.00
Nutrients levels4
Crude protein	48.23	46.50	47.48	47.48	47.55
Crude lipid	9.05	16.29	16.25	16.44	16.44
Organic matter	94.55	94.43	94.53	94.60	94.66
Moisture	5.61	5.52	5.60	5.49	5.50
Gross energy, kJ/kg	16.68	18.16	18.18	18.25	18.27

GA = 18β-glycyrrhetinic acid.

¹NC: normal control diet; HF: high-fat (HF) diet; HFL: HF diet supplemented with 0.5 mg/kg GA; HFM: HF diet supplemented with 1.0 mg/kg GA; HFH: HF diet supplemented with 1.5 mg/kg GA.

²GA (Sigma–Aldrich, Shanghai Trading Co., Ltd., G10105, Shanghai, China) was mixed with bentonite.

³One kg of premix supplied the following minerals and vitamins: FeSO₄·H₂O, 100.00 mg; MgSO₄·H₂O, 625.00 mg; CuSO₄·5H₂O, 20 mg; ZnSO₄·H₂O, 115.94 mg; MnSO₄·H₂O, 37.74 mg; CoCl₂·6H₂O, 81.97 mg; Ca(IO₃)₂, 61.35 mg; Na₂SeO₃, 200 mg; KCl, 95.60 mg; NaCl, 76.26 mg; vitamin A, 133.33 mg; vitamin E, 220.00 mg; vitamin B₁₂, 100.00 mg; vitamin D₃, 2.40 IU; folic acid, 2.00 mg; biotin, 50.00 mg; inositol, 408.16 mg; vitamin B₁, 21.00 mg; vitamin B₂, 43.75 mg; vitamin B₆, 22.00 mg; vitamin K₃, 23.23 mg; acid regurgitation, 37.76 mg; vitamin C, 157.89 mg.

⁴Crude protein, crude lipid, moisture, and gross energy were measured values.

2.4. Fish management and feeding

The feeding trial was conducted at Experiment Station of Ya'an, Sichuan Agricultural University, China. The largemouth bass used in the trial were obtained from the New Hope Feed Research and Development Base in Chengdu, Sichuan, China.. After two weeks of acclimatization, a total of 750 fish (average initial weight: 17.39 ± 0.09 g) were randomly allocated into 5 treatment groups with 3 replicates each. Each replicate, housed in a concrete tank (200 cm × 100 cm × 105 cm), contained 50 fish, resulting in 15 tanks in total. Continuous aeration was provided through air stones, and daily, one-third of the total tank water volume was replaced with aerated tap water. Each of the diets was fed to three replicate tanks for a span of 11 weeks. Throughout the trial, the fish were fed twice daily to apparent satiation at 07:30 and 18:00, adhering to the natural photoperiod. Uneaten feed was removed, dried, and weighed 40 min after each feeding to determine feed intake. Water parameters were monitored daily using YSI Professional Plus Multiparameter Instrument (YSI Incorporated, Yellow Springs, OH, USA). The water temperature and pH were maintained at 25.0 ± 3.0 °C and 7.0 ± 0.5, respectively. Dissolved oxygen (above 5.0 mg/L) was recorded throughout the experiment.

2.5. Samples collection

At the culmination of the feeding trial, fish were fasted for 24 h, subsequently captured, counted, and bulk weighed from each replicate. Subsequently, 12 fish from each replicate were anesthetized with benzocaine solution (50 mg/L). Blood from three fish in each tank were collected by tail vein bleeding, which were left them at 4 °C and then centrifuged at 3000 × g for 10 min to extract serum. Serum samples were stored in a −80 °C ultra-low temperature freezer until subsequent analysis. From each replicate, the livers of six fish were rapidly sampled and frozen in liquid nitrogen. The samples were then stored at −80 °C for detecting gene expression and protein levels. Another set of six fish from each replicate was randomly selected for liver sampling, with these samples also stored at −80 °C for subsequent biochemical analysis.

2.6. Proximate composition measurement

The dry matter content of both the diets and the fish was analyzed following method 930.15 (AOAC, 2006). Specifically, the samples were desiccated at 104 °C until they reached a constant weight for the dry matter analysis. After acid digestion, the total nitrogen (N) was determined according to method 984.13 (AOAC, 2006). Subsequently, the crude protein was calculated by multiplying the total N content by 6.25. The moisture content was measured according to method 930.15 (AOAC, 2006). Moreover, the crude lipid was analyzed using the Soxhlet extraction method 920.39 (AOAC, 2006). Finally, the ash contents were examined in a muffle furnace at 550 °C for 4 h as per method 942.05 (AOAC, 2006). The organic matter content can be obtained by subtracting the crude ash from the raw weight. Table S2 displays the results of whole-body composition of fish. The gross energy in the diet was determined using a Parr 1271 Oxygen Bomb Calorimeter (Parr Instrument Company, Moline, IL, USA), following the standard method 991.43 specified by the AOAC (2005).

2.7. Isolation and culture of primary hepatocytes

For the experiment, healthy largemouth bass with an initial body weight of approximately 50 g were selected, followed by anesthesia with MS-222 (1:10,000; Sigma–Aldrich, St. Louis, MO, USA ). After anesthetizing, the gills were bled and the fish placed on sterile gauze and the body surface wiped with 75% alcohol three times. After dissection, liver tissue was placed in Leibovitz's L-15 (L-15) medium. Liver tissue was washed twice in 75% alcohol for 15 s each. Tissue was then transferred to phosphate-buffered saline (PBS) solution (Biological Industries, Beit-Haemek, Israel) containing penicillin-streptomycin solution (Solarbio, Beijing, China) and washed twice. Then, tissue was rinsed twice with washing solution. Liver tissue was placed in L-15 medium containing 1% penicillin-streptomycin solution and cut with scissors for about 10 min to obtain small (1 mm³) uniform tissue pieces. Tissue suspension was then filtered through a cell strainer using L-15 medium. Cells on the strainer were rinsed with 1% penicillin-streptomycin solution and centrifuged at 1000 × g for 10 min. The supernatant was discarded and cells resuspended with L-15 medium containing 15% foetal bovine serum (FBS). Cells were counted using an automatic cell counter (Millipore Corp., Billerica, MA, USA) to achieve a cell density of 1 × 10⁶ cells/mL. Cells were then seeded into a 6-well plate (Corning Incorporated, Corning, NY, USA), 2 mL per well, and placed in a 28 °C incubator with 5% CO_2.

2.8. Cell viability assay

Primary hepatocytes were seeded in 96-well plates at a density of 3 × 10⁴ cells/well and incubated in a sterile incubator for 24 h. LA coupled with fatty acid-free bovine serum albumin (BSA) L-15. The GA was dissolved in dimethylsulfoxide (DMSO). Following this, the cells were assigned to five groups and treated for 24 h as follows: a blank control, and groups receiving 1.5 mmol/L LA supplemented with 0, 10, 20, or 30 μmol/L GA. After discarding the medium, cells were washed thrice with PBS. A mixture of 100 μL of cell culture medium and 10 μL of CCK8 solution (C0038, Beyotime Biotechnology, Shanghai, China) was then added to each well, incubated for 1 h at 37 °C and 5% CO₂, and subsequently, the OD 600 was measured to using a full-wavelength microplate reader (ReadMax 1200, Shanghai Flash Spectrum Biological Technology Co., Ltd., Shanghai, China) determine relative cell viability.

2.9. Biochemical analysis

Serum samples were collected by centrifuging blood at 3000 × g for 15 min at 4 °C. Liver samples (100 mg) were homogenized in ice-cold 0.9% sterile saline (w:v, 1:10). The homogenate was centrifuged at 6000 × g at 4 °C for 20 min, and the resulting supernatant was employed for laboratory analyses. Commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China) were used to determine serum or liver levels of aspartate aminotransferase (AST; C010-2-1), ALT (C009-2-1), alkaline phosphatase (AKP; A059-2-2), triglyceride (TG; A110-1-1), total cholesterol (TC; A111-1-1), hydroxyproline (HYP; A030-2-1), malondialdehyde (MDA; A003-1-2), glutathione s-transferase (GST; A004-1-1), adenosine 5′-triphosphate (ATP; A095-1-1), and total protein (TP; A045-4-2). Similarly, AST, ALT, TG, TC, and HYP levels of primary hepatocytes were measured according to the instructions.

2.10. Histopathological analysis

These samples were immersed in 10% buffered formalin (pH = 7.4) for 24 h, Subsequently, standard procedures were followed for dehydration, embedding in paraffin, and sectioning to 5-μm slices. Hematoxylin and eosin (H&E) staining and collagen fiber-specific Masson staining were performed on the liver sections as per established protocols (Cao et al., 2024). Light microscopy (DM2500, Leica Microsystems, Wetzlar, Germany) was used to examine the stained sections. Relative quantification analysis was conducted using Image J (version: 1.53, National Institutes of Health, Bethesda, MD, USA), selecting three Masson staining photographs at the same magnification from each group.

For frozen liver tissue sections, fixation in 4% paraformaldehyde was carried out for 15 min after rehydration and drying. Oil red O was applied to the slides and allowed to stain for 8 to 10 min, followed by rinsing with tap water, differentiation using 75% ethanol, and re-staining with hematoxylin. Stained sections were observed under an optical microscope (CX 33, Olympus Corporation, Tokyo, Japan). Relative quantification analysis was performed using Image-Pro Plus 6.0.

2.11. Oil red O and bodipy 493/503 staining of primary hepatocytes

Cells were washed with cold PBS, fixed in 4% paraformaldehyde for 30 min at room temperature, and subsequently washed with PBS. For lipid droplet staining, cells were exposed to Oil red O solution (G1262, Solarbio, Beijing, China) for 20 min and then washed with 60% isopropanol. Images were captured for each group. Additionally, primary hepatocytes were stained using BODIPY493/503 (HY-W090090, MedChemExpress, Monmouth Junction, NJ, USA) for 20 min and 4′,6-diamidino-2-phenylindole (DAPI) (C1006, Beyotime Biotechnology, Shanghai, China) for 5 min. Fluorescent images were obtained using a fluorescence microscope (BX53, Olympus Corporation, Tokyo, Japan).

2.12. The mitochondrial Ca²⁺ content in primary hepatocytes

Rhod-2AM, a high-affinity Ca²⁺ indicator excitable by visible light, was employed. Rhod-2AM (40776ES72, Yeasen Biotechnology [Shanghai] Co., Ltd., Shanghai, China) was used for cell fluorescence, with excitation/emission wavelengths at 549 nm/578 nm. The measurement of mitochondrial Ca²⁺ content was conducted using a fluorescence microscope (IX73, Olympus Corporation, Tokyo, Japan).

2.13. Immunofluorescence examination

Immunocytochemistry was employed to assess SIRT3 expression in primary hepatocytes. Mitotracker (C1035, Beyotime Biotechnology, Shanghai, China) was utilized to label mitochondria. Following fixation in 4% paraformaldehyde for 15 min at room temperature, primary hepatocytes were permeabilized using 0.2% Triton X-100. Primary antibody (anti-SIRT3, 1:100; A7307, ABclonal Technology Co., Ltd., Wuhan, Hubei, China) was then applied and incubated overnight at 4 °C. Subsequently, primary hepatocytes were subjected to Alexa-488-conjugated mouse-anti-rabbit secondary antibody (1:500; A-11008, Invitrogen, Carlsbad, CA, USA) for 2 h at 37 °C. Fluorescence scanner Pannoramic MIDI (3DHistech Ltd., Budapest, Hungary) was employed to capture the images.

2.14. siRNA transfection into hepatocytes

Upon reaching 80% confluence, primary hepatocytes were transfected with either SIRT3 siRNA (5'-CGGUGCACUUCUUCAAGUA-3') or negative control siRNA (RiboBio Co., Ltd., Guangzhou, Guangdong, China) using lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions.

2.15. Electron microscopy

Fresh liver tissue samples were obtained at a thickness of 2 mm³, fixed in a fixative (2.5% glutaraldehyde in pH 7.4 cacodylate buffer) for 2 h at room temperature. The liver tissues underwent three PBS washes and were subsequently post-fixed in 1% osmium tetroxide. After a series of alcohol concentration-based dehydration steps, the samples were post-embedded in Araldite. Ultra-thin cross-oriented sections were produced and stained with uranyl acetate and lead citrate. Images were captured using a HITACHI HT7700 transmission electron microscope (Hitachi High-Technologies Corporation, Tokyo, Japan).

2.16. Real-time quantitative PCR (RT-qPCR)

RNA isolation, reverse transcription, and RT-qPCR procedures were executed in a manner similar to a prior study (Zhao et al., 2022). Total RNA was extracted from liver tissue specimens using RNAiso Plus (Takara Bio, Shiga, Japan). RNA integrity and quality were assessed through 1.5% agar-gel electrophoresis and spectrophotometry, respectively. The PrimeScript RT Reagent Kit with gDNA Eraser (Takara Bio, Shiga, Japan) was employed to synthesize complementary DNA (cDNA) from 1 μg of RNA. Real-time quantitative PCR was conducted using TB Green Premix Ex Taq (Tli RNaseH Plus, Takara Bio, Shiga, Japan). Corresponding primers were provided by Qingke Biotech Co., Ltd. (Chengdu, Sichuan, China), with detailed primer sequences provided in Table 2. The changes of gene expression were detected in the CFX Connect Real-Time System (Bio-Rad Laboratories, Hercules, CA, USA). Changes in gene expression were evaluated using the 2^−ΔΔCt method according to Livak et al. (Livak Kj, 2001), using β-actin and 18S rRNA as the internal control genes (Vandesompele et al., 2002).

Table 2.

Primer sequences and optimal annealing temperatures (OAT) of genes selected for analysis by real-time quantitative PCR.

Names	Sequences (5′-3′)	OAT, ˚C	Accession number
Tgfβ1a	F: GATGCCGATGGATGGATAG	63.3	XM_038693206.1
	R: GGAATGACAACAGTAGCAGGAG
Smad2	F: AACTTCCCTGCTGGCATAGAC	63.3	XM_038733539.1
	R:TTCTGCTGGTGAGCCTGTTTC
Smad3a	F: GCAGGGTTTAACTCCGGGAT	63.3	XM_038701121.1
	R:CCAGTGTCGAGAAGTCCGTC
α-Sma	F: GCGGGACATCAAGGAGAAGC	63.3	XM_038735874.1
	R: CCTGCCGACTCCATACCAAT
Collagen Ⅰ	F: AAAGCACGTCTGGTTCGGAG	65.0	XM_038724497.1
	R: CAGTGGTAGGTGATGTTCTGGGTA
Fibronectin	F: GTGGAGTTGAGAGCAAGCCT	59.0	XM_038701274.1
	R: AGGAAGAGACGGTAGCCACT
Mmp9	F: GTCGTGACACCGCTGTAATCG	61.4	XM_038722740
	R: ATCTCCTCGTCCCTCGCTAG
Αmpkα1	F: CCTGAAGGAGGTATGTGACAAG	60.7	MW465410
	R: CAATGATGAGATGGTAGGCAAC
Pgc1α	F: TGAACTTGAGGGACGATGGG	63.3	XM_038731662.1
	R: TCCGAATGGGAGTCCAAGTC
Sirt3	F: TGAAGAGGTACGAGGGAGACGA	63.3	XM_038701377.1
	R: AGTGAGGTGCCCATGACGAT
Ip3r1	F: TGATGTTCCTGTAGTCCCGTGA	61.4	XM_038695734.1
	R: GCAGAGCGATAGCCATACCC
Sig1r1	F: AATCTCAGCCCAGTAACGACCT	55.9	XM_038733205.1
	R: TCCTACACGCCTCACTCACAG
Casr	F: CTCCATCGTAAGTGGTGGGG	63.3	XM_038707565.1
	R: GGAAGGAGGGGCGTTGTAAA
Grp75	F: GGCGATGTTACCGATGTGCT	61.4	XM_038699058.1
	R: ACCTGGCTCTTCTTTGTGGG
Vdac1	F: GAGAAGTGGAACACCGACAACA	64.5	XM_038708837.1
	R: TTGGCGAGAAGGAGGAATCA
Mcu	F: AGCGACTGTGTGGGAGTTTT	63.3	XM_038737220.1
	R: CACGAGGTGTGTGGTGTCAT
β-Actin	F: CCCCATCCACCATGAAGA	55.7	XM_038695351.1
	R: CCTGCTTGCTGATCCACAT
18S rRNA	F: TGAATACCGCAGCTAGGAATAATG	59.0	MH018569.1
	R: CCTCCGACTTTCGTTCTTGATT

2.17. Western blot

The Western blot procedures were executed as described in the previously published study (Zhao et al., 2022). Post-separation was completed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (P0012AC, Beyotime Biotechnology, Shanghai, China) and transfer onto a polyvinylidene fluoride (PVDF) membrane, blocking with 5% skim milk followed by sequential incubation with primary and secondary antibodies took place. The detailed information for all antibodies used in this study is provided in Table S3. Expression of targeted proteins was detected using an ECL kit (P0018FS, Beyotime Biotechnology, Shanghai, China) and documented using the ChemiDoc Imaging Systems/Image Lab software (Bio-Rad Laboratories, Hercules, CA, USA). β-Actin served as the internal reference for sample loading equality.

2.18. Statistical analysis

The data were analyzed using the PROC MIXED procedure of SAS 9.3 (SAS Inst. Inc., Cary, NC, USA) using the model:

$$Y_{ijkl}=\mu +T_i+P_i+S_k+C{(K)}_l+T\times S_{ik}+e_{ijkl}\text{,}$$

where Y_ijkl, the dependent variable; μ, the overall mean; T_i, the fixed treatment effect; P_j, the random period effect; S_k, the random square effect; C(K)_l, the random effect of the lth steer with in the k-th square; T × S_ik, the interaction between the i-th treatment and the k-th square; and e_ijkl, the error residual. The Kenward-Roger option was used to calculate degrees of freedom. Differences among the means of different treatments were tested using Duncan's test. Effects were considered significant at P < 0.05.

3. Results

3.1. GA relieves HF diet-induced liver injury

3.1.1. Liver histopathology

To assess GA's potential impact on liver injury, liver organ index measurements were conducted on largemouth bass liver tissues. Images of the liver in the NC group exhibited a smooth and ruddy surface. In contrast, the liver in the HF group of largemouth bass displayed a pale surface, indicative of injury. Notably, GA administration visibly improved the surface condition, transitioning from pale to ruddy (Fig. 1A). Furthermore, H&E staining revealed prominent liver macrovesicular steatosis and cellular nuclei displacement in the HF group, with instances of nucleus disappearance and injury. The GA treatment mitigated these effects, reducing cavity presence and promoting increased cell nuclei count (Fig. 1B). Compared with the NC group, the lipid vacuoles relative area was significantly increased under HF conditions (Table 3; P < 0.001); in contrast, the GA groups exhibited a significant decrease in lipid vacuole relative area compared with the HF group (Table 3; P < 0.001). Oil red O staining demonstrated abundant lipid droplets in the liver of the HF group compared to the NC group, while GA treatment gradually reduced these droplets (Fig. 1C). Oil red O-positive areas significantly expanded in the HF group compared with NC group (P < 0.001), but were reduced following GA treatment (Table 3; P < 0.001). Based on the quadratic regression analysis, the dietary optimal GA level of largemouth bass (17.31–111.42 g) was recommended to be 0.8 mg/kg (Fig. 2).

Fig. 1.

18β-Glycyrrhetinic acid (GA) protects against high-fat (HF) diet-induced liver injury in largemouth bass. (A) Fish body and liver images. (B and C) Hematoxylin and eosin (H&E; scale bar = 100 and 20 μm) and Oil red O staining (scale bar = 20 μm). NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA.

Table 3.

Histopathological statistical analysis in liver of largemouth bass (%).

Items	Groups1					SEM	P-value
	NC	HF	HFL	HFM	HFH
Lipid vacuoles relative area	14.22c	23.77a	18.46b	15.76bc	16.47bc	0.717	<0.001
Oil red O positive area	4.39c	29.09a	19.06b	8.54c	5.47c	1.898	<0.001
Relative blue area	0.38c	0.62a	0.55b	0.38c	0.33c	0.032	<0.001

SEM = standard error of the mean.

Mean values with different superscripts in the same row are significantly different (P < 0.05).

¹NC group: normal control diet; HF group: high-fat (HF) diet; HFL group: HF diet supplemented with 0.5 mg/kg 18β-glycyrrhetinic acid (GA); HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA.

Fig. 2.

Quadratic regression analysis of percent weight gain (PWG) for largemouth bass fed diets with different levels of 18β-glycyrrhetinic acid (GA) for 77 day.

3.1.2. Hematological and liver functions parameters

Hematological and liver function parameters of largemouth bass are detailed in Table 4. Serum AST, ALT, and AKP activities in the HF group significantly exceeded those in the NC group (P < 0.05), while GA treatment resulted in decreased values (P < 0.05). Serum and liver TG and TC contents, as well as liver MDA content, were notably elevated in the HF group compared to the NC group (P < 0.05). However, administration of GA led to a significant reduction in serum TG and TC contents, as well as liver MDA content (P < 0.05). The GST activity in liver was the lowest in the HF group. Additionally, liver TP content in the HF group significantly surpassed that of the NC group (P < 0.001), with GA treatment further increasing TP content in comparison to the HF group (P < 0.001).

Table 4.

Effects of 18 β-glycyrrhetinic acid (GA) on serum and liver biochemical parameters in largemouth bass induced by high-fat (HF) diet.

Items	Groups1					SEM	P-value
	NC	HF	HFL	HFM	HFH
Serum
AST, U/L	40.83c	83.78a	61.19b	40.10c	42.38c	2.270	<0.001
ALT, U/L	4.15d	23.85a	18.42b	8.16c	5.40d	0.986	<0.001
AKP, King/100 mL	6.91d	13.36a	11.15b	9.59c	7.43d	0.394	<0.001
TG, mmol/L	10.46c	21.52a	15.59b	11.84c	10.13c	0.720	<0.001
TC, mmol/L	10.33c	19.51a	16.91b	12.15c	12.49c	0.599	<0.001
HYP, μg/mL	9.42c	24.36a	16.35b	8.78c	9.13c	0.970	<0.001
Liver
MDA, nmol/mg prot	2.43c	5.91a	3.79b	2.58c	2.74c	0.235	<0.001
GST, U/mg prot	80.97a	46.48c	55.23bc	69.29ab	81.23a	3.312	<0.001
TP, μg/mL	2.67c	2.90b	3.04a	3.00ab	3.03a	0.028	<0.001
ATP, μmol/g prot	872.54b	420.44c	571.80c	776.60b	1097.24a	44.021	<0.001
TG, mmol/g prot	2.49d	5.28a	4.52b	3.11c	2.10d	0.201	<0.001
TC, mmol/g prot	0.88cd	1.54a	1.25b	1.12c	0.68d	0.062	<0.001
HYP, μg/mg	0.93d	2.36a	1.94b	1.40c	1.08d	0.085	<0.001

AST = aspartate aminotransferase; ALT = alanine aminotransferase; AKP = alkaline phosphatase; TG = triglyceride; TC = total cholesterol; HYP = hydroxyproline; MDA = malondialdehyde; GST = glutathione S-transferase; TP = total protein; ATP = adenosine 5′-triphosphate; SEM = standard error of the mean.

Mean values with different superscripts in the same row are significantly different (P < 0.05).

¹NC group: normal control diet; HF group: high-fat (HF) diet; HFL group: HF diet supplemented with 0.5 mg/kg 18β-glycyrrhetinic acid (GA); HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA.

3.2. GA prevents HF diet-induced liver fibrosis in the liver of largemouth bass

3.2.1. The expression of pro-fibrogenic markers

Masson's trichrome staining, designed for the specific detection of collagen fibers, revealed abundant accumulation of collagen fibers in the HF group. In contrast, treatment with GA led to a reduction in fibrogenic changes, approaching levels similar to the NC group (Fig. 3A). Moreover, analysis of the relative blue area-corresponding to Masson's trichrome staining-revealed that GA-treated fish showed a noticeable improvement in liver architecture (Table 3; P < 0.001), along with reduced collagen fiber accumulation. To delve deeper into the effects of GA on liver fibrosis, the assessed HYP content and the expressions of pro-fibrotic markers α-SMA, collagen I, and fibrontein. In Table 4, HYP content, reflecting collagen fiber alterations, was notably elevated in the serum of the HF group in comparison to the NC group (P < 0.001). The GA administration successfully mitigated serum and liver HYP content, demonstrating a significant decrease compared to the HF group (P < 0.05). The relative mRNA expression levels of α-Sma, collagen I, fibrontein, and Mmp9 were significantly increased in the HF group when compared with the NC group (Fig. 3B; P < 0.05). Nonetheless, GA treatment effectively curbed these elevations in protein levels (Fig. 3C–F; P < 0.05). Collectively, these results underscore GA's potential in attenuating liver fibrosis in HF diet-induced largemouth bass.

Fig. 3.

18β-Glycyrrhetinic acid (GA) repress liver fibrosis in high-fat (HF) diet-induced largemouth bass. (A) Masson's trichrome staining (scale bar = 100 and/or 20 μm). (B) The relative mRNA expression levels of α-Sma, collagen I, fibrontein, and Mmp9. (C-F) α-SMA, collagen I, and fibrontein protein levels and protein densitometric values were normalized against β-actin. NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA. The data was presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05).

3.2.2. GA regulates TGFβ1-Smad2/3 signaling pathway in HF diet-induced liver fibrosis

The TGFβ1, a potent fibrogenic cytokine in the liver, was assessed to uncover GA's inhibitory mechanisms against HF diet-induced liver fibrosis. The relative mRNA expression and protein levels of TGFβ1 and its target proteins Smad2 and Smad3 were evaluated. Compared with the NC group, TGFβ1α relative mRNA expression level was significantly higher in the HF group (P < 0.05), while it decreased to varying degrees in the HFL, HFM, and HFH groups. Smad2 relative mRNA expression level was significantly higher in the HF group than in the NC group (P < 0.001), and was lower in the HFL and HFM groups and further reduced in the HFH group, with no significant differences among these groups (P > 0.05). Smad3a relative mRNA expression level was higher in the HF group than in the NC group, but the difference was not statistically significant (P = 0.119), and decreased successively in the HFL, HFM, and HFH groups, with no significant differences among them (P > 0.05; Fig. 4A). Compared with the NC group, the relative protein levels of TGFβ1, Smad2, and Smad3 in the HF group were significantly increased (P < 0.05). Treatment with GA successfully reduced the expressions of TGFβ1, Smad2, and Smad3 protein levels (Fig. 4B–E; P < 0.05). In liver tissues, it was noticed that α-SMA and TGFβ1 expressions were significantly increased in HF diet-treated fish compared with those in NC group (P ＜ 0.05), which GA treatment restore their expression (Fig. 4F). This underscores GA's role in ameliorating liver fibrosis by downregulating TGFβ1-mediated Smad2/3 signaling and downstream regulators.

Fig. 4.

Effects of 18β-glycyrrhetinic acid (GA) on the TGFβ1-Smad2/3 pathways in high-fat (HF) diet-induced liver of largemouth bass. (A)The relative mRNA expression levels of Tgfβ1α, Smad2, and Smad3a. (B-E) TGFβ1, p-Smad2, and p-Smad3 protein levels and protein densitometric values were normalized against β-actin. (F) Immunohistofluorescence for co-staining of α-SMA (green) and TGFβ1 (red) in the liver lobule of fish livers. Nuclei were counterstained with 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI). Scale bar: 100 and 20 μm (boxes). NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05).

3.3. Liver mitochondrial function

To explore whether GA's mitigation of HF diet-induced liver fibrosis is mediated through mitochondrial Ca²⁺ level. The SIRT3, a major deacetylase in mitochondria, regulates almost every aspect of mitochondrial health and functionality, it was conducted a virtual docking analysis of GA with SIRT3, binding energy is −9.0 kcal/mol. The analysis revealed stable hydrogen bond formations and hydrophobic interactions, suggesting the GA can bind to the SIRT3 protein spontaneously (Fig. 5A). Mitochondria are the primary sites of ATP formation. The results showed that HF group exhibited reduced liver ATP content compared to the NC group, while HFM and HFH group significantly increased ATP content (Table 4; P < 0.001). Mitochondrial dysfunction is associated with ROS generation, contributing to liver fibrosis development. Reactive oxygen species serves as a pivotal indicator for assessing mitochondrial function. 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride was used to detect the nucleus (blue fluorescence), while DCFH-DA was employed for ROS measurement (green fluorescence). A noticeable increase in green fluorescence was observed in the HF group in comparison to the NC group. However, GA treatment effectively reduced green fluorescence, indicating decreased ROS content (Fig. 5B). Meanwhile, quantitative assessment of ROS content confirmed an elevated ROS level in the HF group, which was significantly reduced with GA treatment (Fig. 5D; P < 0.001). JC-1, a cationic dye, revealed changes in MMP. The HF group showed increased JC-1 monomers and reduced aggregates, while GA-treated group displayed more aggregates and fewer monomers, indicating restored MMP (Fig. 5C). Quantification of MMP ratios (red/green fluorescence intensity) supported this observation, with GA-treated group showing an elevated ratio compared to HF group (Fig. 5E; P < 0.001).

Fig. 5.

Effects of 18 β-glycyrrhetinic acid (GA) on liver mitochondrial damage in largemouth bass induced by a high-fat (HF) diet. (A) Molecular docking of GA with Sirt3. (B) Double staining of liver cell suspension with 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) and 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA), captured under an inverted fluorescence microscope (scale bar = 100 μm). (C) Mitochondrial membrane potential (MMP) indicated by JC-1 staining under a fluorescence microscope, with a scale bar of 100 μm, and corresponding fluorescence levels. (D) Relative reactive oxygen species (ROS) content. (E) Fluorescence enzyme-linked immunosorbent assay. JC-1mom is a (JC-1monome) monomer; JC-1agg is a polymer (JC-1aggregates). NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05).

3.4. GA relieves HF diet-induced over-transfer of Ca²⁺ from endoplasmic reticulum to mitochondria

3.4.1. Transmission electron microscope

As shown in Fig. 6A, microscopic analysis revealed that the liver in the NC group exhibited abundant mitochondria, well-defined ER structures, neatly arranged mitochondrial cristae, intact mitochondrial membranes, and appropriate gaps between ER and mitochondria. Conversely, HF group displayed mitochondrial membrane rupture, disordered mitochondrial cristae arrangement, abnormal ER structures, and reduced ER-mitochondria gaps. Treatment with GA restored these features, closely resembling the NC group.

Fig. 6.

Effect of 18 β-glycyrrhetinic acid (GA) on high-fat (HF) diet-induced over-transfer of Ca²⁺ from endoplasmic reticulum to mitochondria. (A) Representative transmission electron microscopy (TEM) images of liver tissue. N: nucleus; L: lipid droplet; M: mitochondria; ER: endoplasmic reticulum; red arrow: mitochondrial ridge, blue arrow: mitochondrial membrane; black arrow: endoplasmic reticulum. The right-illustration of the measure of distance between ER and mitochondria (scale = 2 or 0.5 μm). (B) The relative mRNA expression levels of Casr, Sig1r, and Ip3r1. (C-D) IP3R1 protein level and protein densitometric values were normalized against β-actin. (E) The relative mRNA expression levels of Grp75, Vdac1, and Mcu. (F-I) Protein levels and densitometric values of GRP75, VDAC1, and MCU, normalized against β-actin. NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05).

3.4.2. Endoplasmic reticulum and mitochondria Ca²⁺ transport proteins expression

Figure 6B presented that relative mRNA expression levels of Casr, Sig1r, and Ip3r1 was elevated in the HF group, whereas GA administration significantly decreased Casr, Sig1r, and Ip3r1 relative mRNA expression levels in liver tissues (P < 0.05). HF group exhibited elevated protein level of IP3R1 compared to the NC group, which were lowered with GA treatment (Fig. 6C and D; P < 0.001). To delve into GA's mechanism against mitochondrial Ca²⁺ overload, the relative mRNA expression levels of Grp75, Vdac1, and Mcu was evaluated. HF group displayed increased relative mRNA expression levels of Grp75, Vdac1, and Mcu, while GA administration significantly reduced the expression (Fig. 6E; P < 0.05). Further insights were gained through the examination of liver VDAC1 and MCU protein levels, which exhibited a noticeable increase in the HF group (P < 0.05). Remarkably, treatment with GA led to a reduction in these levels in HFM group (Fig. 6F, H, and I; P < 0.05). The protein level of GRP75, however, remained relatively unaffected in both the HF group and the GA treatment group (Fig. 6G; P > 0.05).

3.5. GA attenuates HF diet-induced mitochondrial Ca²⁺ transport proteins via regulating SIRT3 expression

Comparative analysis of relative mRNA expressions revealed decreased levels of Ampkα1, Pgc1α, and Sirt3 in the HF group, contrasting with the NC group (P < 0.05). Notably, in the HFM group, GA treatment restored Ampkα1 relative mRNA expression level relative to the HF group (P = 0.019). Although GA treatment also upregulated the relative mRNA expression levels of Pgc1α and Sirt3, no statistically significant differences were observed among the groups (; Fig. 7A; P > 0.05). A similar trend was observed in liver p-AMPK, PGC1α, and SIRT3 protein levels. The levels of these proteins were notably diminished in the HF group compared with the NC group (P < 0.05). After GA treatment, the protein levels of p-AMPK, PGC1α, and SIRT3 in the HFL, HFM, and HFH groups showed varying degrees of increase relative to the HF group (Fig. 7B–E; P < 0.05). Consistent with the reduction of mRNA and protein expressions, it was found that GA strongly reduced the HF diet-induced expression of MCU in the liver by immunohistofluorescence analysis. In addition, it was also found the HF diet-reduced liver SIRT3 level was elevated by GA (Fig. 7F). Collectively, these data suggested that GA treatment lowers liver MCU expression via regulating SIRT3 expression.

Fig. 7.

Effect of 18 β-glycyrrhetinic acid (GA) on AMPK-PGC1α-SIRT3 pathway in high-fat (HF) diet-induced liver of largemouth bass. (A) The relative mRNA expression levels of Ampkα1, Pgc1α, and Sirt3. (B-E) Protein levels and densitometric values of AMPK, PGC1α, and SIRT3, normalized against β-actin. (F) Immunohistofluorescence for co-staining of MCU (green) and SIRT3 (red) in the liver lobule of fish livers. Nuclei were counterstained with 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI). The scale bars in rows 1 to 4 are 100 μm, and the scale bar in the 5th row is 20 μm. NC group: normal control diet; HF group: HF diet; HFL group: HF diet supplemented with 0.5 mg/kg GA; HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05).

3.6. GA inhibited LA diet-induced lipid accumulation in primary hepatocytes

Assessment of cellular toxicity of GA involved treating primary hepatocytes with varying concentrations of GA (10, 20, and 30 mmol/L) in the presence or absence of LA (1.5 mmol/L) for 24 h. Cell viability was evaluated using the CCK8 assay. As shown in Fig. 8A, LA treatments ranging from 0 to 1.5 mmol/L exhibited negligible impact on cell viability. However, a relatively higher concentration of 3.0 mmol/L LA displayed significant cytotoxicity to primary hepatocytes (P < 0.001). Conversely, GA treatments within the range of 0 to 30.0 mmol/L exerted no discernible effect on cell viability (Fig. 8B; P = 0.649). As a result, the subsequent experiments were carried out using 1.5 mmol/L LA and GA concentrations ranging from 0 to 30.0 mmol/L. Subsequent BODIPY493/503 staining reaffirmed these findings, with GA treatment effectively reducing the accumulation of lipid droplets in LA-treated primary hepatocytes (Fig. 8C). Corresponding test kit measurements unveiled that GA successfully inhibited LA-induced AST and ALT activities, TG, TC and HYP contents in Table 5 (P < 0.05).

Fig. 8.

Effects of 18 β-glycyrrhetinic acid (GA) on linoleic acid (LA)-induced lipid accumulation in primary hepatocyte. (A-B) Treatment of primary hepatocytes with LA and GA for 24 h, with hepatocyte viability evaluated using a CCK8 Assay Kit after LA or GA treatment. (C) Representative confocal microscopy images of hepatocytes with bodipy 493/503 staining (scale bar = 5 μm). The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05). "-": not added.

Table 5.

Effects of 18β-glycyrrhetinic acid (GA) on biochemical parameters in primary hepatocytes from largemouth bass induced by linoleic acid (LA).

Items	Groups1					SEM	P-value
	NC	HF	HFL	HFM	HFH
AST, U/g prot	13.19d	39.84a	29.25b	20.72c	12.45d	1.658	<0.001
ALT, U/g prot	3.04d	14.45a	9.81b	7.78c	3.30d	0.673	<0.001
TG, mmol/g prot	1.37d	2.40a	1.86b	1.43c	1.28cd	0.067	<0.001
TC, mmol/g prot	0.95d	2.12a	1.58b	1.22c	1.08cd	0.066	<0.001
HYP, μg/mL	9.70d	18.85a	16.57b	12.42c	10.65d	0.564	<0.001

AST = aspartate aminotransferase; ALT = alanine aminotransferase; TG = triglyceride; TC = total cholesterol; HYP = hydroxyproline; SEM = standard error of the mean.

Mean values with different superscripts in the same row are significantly different (P < 0.05).

¹NC group: normal control diet; HF group: high-fat (HF) diet; HFL group: HF diet supplemented with 0.5 mg/kg 18β-glycyrrhetinic acid (GA); HFM group: HF diet supplemented with 1.0 mg/kg GA; HFH group: HF diet supplemented with 1.5 mg/kg GA.

3.7. GA relieves LA-induced over-transfer of Ca²⁺ from endoplasmic reticulum to mitochondria

To further elucidate the mechanism underlying the favorable influence of GA on the expression of Ca²⁺ transporters in primary hepatocytes induced by LA, it was measured relative mRNA expression levelsof Ip3r1, Sig1r, Grp75, Vdac1, and Mcu, along with protein levels of GRP75, VDAC1, and MCU. Notably, LA induced a significant upsurge in the relative mRNA expression and protein levels of IP3R1, GRP75, VDAC1, and MCU (P < 0.05). Following GA treatment, expression of Grp75 and Vdac1 mRNA, and MCU protein were substantially reduced in LA-treated primary hepatocytes (Fig. 9A–I; P < 0.05). Utilizing the Rhod-2AM probe, commonly employed for mitochondrial Ca²⁺ labeling, a marked elevation in mitochondrial Ca²⁺ content was observed in response to LA. However, GA treatment effectively counteracted this rise, significantly reducing mitochondrial Ca²⁺ content in primary hepatocytes (Fig. 9J).

Fig. 9.

Effects of 18 β-glycyrrhetinic acid (GA) on linoleic acid (LA)-induced over-transfer of Ca²⁺ from endoplasmic reticulum to mitochondria in primary hepatocyte. (A-E) The relative mRNA expression levels of Ip3r1, Sig1r, Grp75, Vdac1, and Mcu. (F-I) GRP75, VDAC1, and MCU protein levels and protein densitometric values were normalized against β-actin. (J) Mitochondrial Ca²⁺ labeled by Rhod-2AM probe (scale bar in the first row is 100 μm, and the scale bar in the second row is 20 μm). The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05). "-": not added.

3.8. GA attenuates LA-induced over-transfer of Ca²⁺ from endoplasmic reticulum to mitochondria via SIRT3

Exploring the underlying mechanism of GA's effects on the over-transfer of Ca²⁺ from the endoplasmic reticulum to mitochondria via the regulation of the SIRT3 expression in primary hepatocytes, it was assessed relative mRNA expression and protein levels of AMPK, PGC1α, and SIRT3. Compared to the NC group, the relative mRNA expression levels of Ampkα1, Pgc1α, and Sirt3 were significantly decreased in the LA group (Fig. 10A–C; P < 0.05). GA supplementation significantly upregulated the relative mRNA expression levels of Ampkα1 and Pgc1α (P < 0.05). Although GA treatment increased Sirt3 relative mRNA expression level compared to the LA group, the difference was not statistically significant (P = 0.355). Western blot analysis revealed that the protein levels of AMPK, PGC1α, and SIRT3 were significantly reduced in the LA group compared to the NC group (P < 0.05). At a GA concentration of 30 μmol/L, the protein levels of AMPK, PGC1α, and SIRT3 were significantly increased compared to the LA group (Fig. 10D–G; P < 0.05). Immunostaining results indicated mitochondrial localization of SIRT3, revealing that exposure of primary hepatocytes to LA resulted in a reduction in SIRT3 expression. Impressively, GA treatment successfully restored SIRT3 expression in LA-treated primary hepatocytes (Fig. 10H).

Fig. 10.

Effects of 18 β-glycyrrhetinic acid (GA) on Ampk-Pgc1α-Sirt3 pathway in linoleic acid (LA)-induced in primary hepatocyte of largemouth bass. (A-C) The relative mRNA expression levels of Ampkα1, Pgc1α, and Sirt3. (D-G) AMPK, PGC1α, and SIRT3 protein levels and protein densitometric values were normalized against β-actin. (H) Expression and distribution of SIRT3 detected by immunofluorescence staining for SIRT3 (green), merge (yellow), mitochondria (red), and 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) (blue), with a scale bar of 2 μm. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05). "-": not added.

3.9. Knockdown of Sirt3 alleviates mitochondrial Ca²⁺ overload induced by HF diet

To understand the mechanism behind SIRT3-mediated mitochondrial Ca²⁺ overload induced by HF diet. Subsequent SIRT3 knockdown was executed via siRNA transfection in primary hepatocytes, and the CCK8 assay demonstrated no significant effect on cell viability in Fig. 11A (P = 0.611). Knockdown of Sirt3 relative mRNA expression was validated (Fig. 11B; P = 0.002). As shown in Fig. 11C and D, protein level analysis confirmed substantial down-regulation of SIRT3 protein levels in the SIRT3 knockdown group compared to siCtrl group (Fig. 11E and F; P < 0.001). Importantly, GA treatment effectively reversed the reduction in SIRT3 protein levels caused by LA (P < 0.001). Of note, SIRT3 knockdown led to a significant reversal in the reduction of MCU protein levels triggered by GA (Fig. 11G; P < 0.001).

Fig. 11.

Effects of 18 β-glycyrrhetinic acid (GA) on linoleic acid (LA)-induced MCU expression. (A) The impact of siSIRT3 transfection for 48 h on the viability of primary hepatocyte. (B) The relative mRNA expression level of Sirt3 in primary hepatocyte transfected with siSIRT3. (C and D) SIRT3 protein levels and protein densitometric values in primary hepatocyte transfected with siSIRT3. (E and G) GA supplementation on SIRT3 and MCU protein levels and protein densitometric values in primary hepatocyte transfected with siSIRT3, values were normalized against β-actin. The data is presented as means and standard error of the mean (SEM). Different letters indicate significant differences (P < 0.05). ns: no significant difference; ∗: significant difference, P < 0.05; "-": not added; "+": added.

4. Discussion

The present study builds on a growth experiment reported previously, which revealed that GA supplementation in an HF diet enhances the growth performance of largemouth bass (Cao et al., 2024). Liver fibrosis is a ubiquitous liver response to acute or chronic injury (Trefts et al., 2017). GA has garnered attention for its multiple biological functions, safety, efficiency, and economic value, positioning it as a promising candidate for medicinal applications (Kowalska and Kalinowska Lis, 2019; Yichuan et al., 2010). GA has been reported to protect the liver (Pan et al., 2022; Yichuan et al., 2010), but the mechanism behind it is unclear. The exact way GA shields the liver from HF diet-induced injury is also unknown. In an effort to further elucidate the liver protective attributes of GA and unravel its potential molecular mechanisms, an investigation was conducted.

Optimal fat content in fish feed is crucial for growth and efficiency, yet excessive fat addition can lead to liver fat accumulation and injury. The liver, pivotal for fat deposition and metabolism, can suffer damage due to excessive fat intake (Greene, 1987; Qiang et al., 2018). This present study provides substantial evidence that GA alleviates liver injury by HF diet-induced. The present findings indicated that GA treatment alleviated the HF diet-induced swelling and a pale liver surface, and histopathological injury. Transaminase, including ALT and AST, is indicative of liver injury when present in elevated serum levels (Lu et al., 2017; Nyblom et al., 2004). This study unveils that feeding largemouth bass with HF diet leads to increased ALT and AST activities in serum, which GA treatment effectively counteracts. This outcome may result from excessive fat accumulation in liver tissue after HF diet consumption. The subsequent surge in ROS content provokes liver injury, aligning with findings from a study on blunt snout bream Megalobrama amblycephala (Lu et al., 2017). Alkaline phosphatase serves as a biomarker for obstructive jaundice and bile stasis, and it also plays a significant role in inducing liver fibrosis (Wang et al., 2019). Remarkably, serum AKP activity significantly rises in the HF group compared to the NC group, and this elevation is appreciably mitigated upon GA administration.

In fish, serum TC predominantly originates from liver tissue. Liver cell injury can contribute to an elevated TC content in the bloodstream. Moreover, serum TG and TC levels offer insights into lipid synthesis and metabolism within the body (Ding et al., 2021; He et al., 2013). Typically, an escalation in dietary fat content correlates with heightened TG and TC contents in serum (Kj R et al., 2009). This study showed that GA treatment effectively curtails TG and TC contents in the serum, thereby alleviating the detrimental effects of HF diet on largemouth bass liver cells. Malondialdehyde, a small molecule produced in the final stages of the lipid peroxidation chain reaction, indirectly signifies oxidative stress-induced injury in fish tissues (Jiang et al., 2016). The liver MDA content was considerably decreased by GA treatment. This trend is in line with the findings from a study on blunt snout bream (Lu et al., 2017). In summary, HF diet-induced liver injury finds alleviation through GA administration.

Prolonged exposure of liver cells to a caloric surplus, such as in the HF environment, can result in triglyceride and derived toxic metabolite accumulation, triggering pro-fibrogenic markers (Liu et al., 2021). To elucidate the underlying mechanism of GA prevents liver fibrosis induced by HF diet, RT-qPCR and Western blot analysis was conducted to examine the differential expression of pro-fibrogenic genes in fish. Results analysis hinted that GA administration effectively downregulates the expression of pro-fibrotic markers α-SMA, collagen I, and fibronectin and MMP9, which are pivotal in liver fibrosis (Yuan et al., 2022). Masson's trichrome staining, a reliable method for highlighting collagen disposition, confirmed that GA effectively reduces collagen fiber accumulation induced by HF treatment (Huang et al., 2013). The results of Masson's trichrome staining further demonstrated that GA reduced liver disposition of collagens induced by HF-treated fish. Hydroxyproline, a major component of collagen, elevates during liver fibrosis (Bolarin and Azinge, 2007; Pihlajaniemi et al., 1991). Notably, GA successfully reduces serum HYP content heightened by HF treatment. These results show GA can reduce HF diet-induced liver fibrosis by regulating pro-fibrotic markers and collagen aggregation.

Transforming growth factor β1, a principal regulator of ECM accumulation, prompts fibroblast-to-myofibroblast transformation and influences the expression of key fibrotic markers like α-SMA and collagen I (Allawzi et al., 2019; Bi et al., 2012). Intracellular effectors Smad2 and Smad3 mediate TGFβ1 signaling (Xu et al., 2016). The TGFβ1-Smad pathway is pivotal in liver fibrosis pathogenesis (Uemura et al., 2005). Particularly, Smad3 plays a crucial role in regulating transduction pathways during liver fibrosis (Xu et al., 2016). Conversely, Smad2 counterbalances Smad3 signaling (Xu et al., 2016). A prior study demonstrated that GA effectively reduced the protein content of TGFβ1 and p-Smad2/3 in rat livers (Pan et al., 2022). GA intervention also inhibited Smad3 nuclear accumulation in activated liver stellate cells (Moro et al., 2008). It was consistently found that GA treatment has similar effect in liver of fish. Immunohistochemistry results further substantiated the down-regulation of TGFβ1 expressions by GA. These findings highlight GA's ability to combat liver fibrosis by targeting the TGFβ1-Smad2/3 pathway, suppressing pro-fibrogenic genes, and reducing collagen buildup.

Mitochondria, recognized as the primary ROS generation sites within cells, play a pivotal role in regulating oxidative stress and cell death (Mansouri et al., 2018). Studies on mouse liver have observed a reduction in MMP due to HF treatment, and the fluctuation in ATP content parallels this alteration (Chen et al., 2022). In this study, GA-treated group that exhibited enhanced mitochondrial MMP and ATP levels. As the main targets of ROS, mitochondria face attacks from these molecules (Yang et al., 2016). Elevated ROS content was evident in the liver of mice subjected to HF diets (Jiang et al., 2022). These findings corroborate this by showcasing increased ROS content following HF treatment, which GA intervention effectively mitigated. The TEM images portrayed significant mitochondrial membrane rupture, disordered mitochondrial cristae arrangement, anomalous endoplasmic reticulum structure, and reduced endoplasmic reticulum-mitochondria gap in the HF group. Impressively, GA treatment succeeded in restoring these structures, consistent with the NC group and analogous to research involving obese mice, which revealed minimized endoplasmic reticulum-mitochondria gap in HF groups (Chen et al., 2022). These findings confirm GA reduces HF-related mitochondrial damage by boosting MMP and ATP levels, lowering ROS, and restoring mitochondria-ER distance.

Mitochondrial Ca²⁺ overload resulting from HF is intricately tied to the endoplasmic reticulum. Essential proteins governing Ca²⁺ influx/outflux in/from the ER membrane include the Ca²⁺ release channel IP3R, primarily transporting Ca²⁺ from the ER to the cytoplasm, thereby modulating ER Ca²⁺ levels to a certain extent. The molecular partner of IP3R, Sig1R, enhances IP3R function when expressed prominently (Tian, 2021). CASR, an important receptor regulating cellular calcium homeostasis, showed heightened mRNA expression in the HF group, along with Sig1r and Ip3r1, consistently supported by their protein levels. Western blot analysis unveiled a higher IP3R1 protein level in the HF group compared to the NC group, while GA treatment brought about a reduction in IP3R1 protein levels both in vivo and in vitro. Similar research on mice reported significantly elevated protein levels of IP3R1 and Sig1R post HF treatment (Chen et al., 2022).

GRP75, an essential connecting protein in the MAMs region, demonstrated increased protein levels in HF-treated mice (Thoudam et al., 2019). In this study, HF treatment heightened both relative mRNA expression and protein levels of GRP75, while GA intervention brought about a decrease in its expression. VDAC, a prominent Ca²⁺ transporter on the outer mitochondrial membrane, was abundant (Sander et al., 2021; Tian, 2021). Szabadkai et al. (2016) demonstrated that GRP75 relocates from mitochondria to MAMs, binding ER IP3R and mitochondrial VDAC1 to form a stable ER-mitochondria channel regulating calcium uptake. The MCU is a Ca²⁺-gated channel situated on the inner mitochondrial membrane, serving as a fundamental element for mitochondrial Ca²⁺ uptake (Tian, 2021). MCU typically collaborates with IP3R-GRP75-Vdac1 to establish a comprehensive Ca²⁺ regulatory axis connecting the endoplasmic reticulum and mitochondria within the MAMs. The Ca²⁺ released by ER IP3R at the MAMs traverses through Vdac1 into the mitochondrial matrix via the mitochondrial MCU (Han et al., 2018; Tian, 2021). This investigation found that the relative mRNA expression and protein levels of VDAC1 and MCU significantly increased post HF feeding, while GA treatment effectively attenuated this elevation. Consistently, GA intervention reduced relative mRNA expressions and protein levels of GRP75, VDAC1, and MCU in LA-treated primary hepatocytes. Furthermore, GA treatment led to reduced mitochondrial Ca²⁺ content in primary hepatocytes. This intervention potentially hampers excessive Ca²⁺ transfer from the endoplasmic reticulum to mitochondria. However, this precise regulatory mechanism necessitates further exploration.

AMPK-PGC1α operates as a cellular energy sensor and modulator, activated in response to elevated energy consumption and demand. SIRT3, a downstream target of AMPK-PGC1α signaling, plays a pivotal role in governing mitochondrial biogenesis and oxidative stress regulation (Samant et al., 2014). Unlike SIRT1, and SIRT3 predominantly localizes to mitochondria and is instrumental in regulating mitochondrial activity (Sheng et al., 2019). In this study, both relative mRNA and protein levels of AMPK, PGC1α, and SIRT3 were notably diminished in the HF group or LA group, whereas GA treatment significantly elevated their mRNA and protein expression levels. Prior investigations highlighted SIRT3's capability to inhibit the mitochondrial permeability transition pore (mPTP), thereby stabilizing mitochondrial membranes (Hafner et al., 2010), as well as promoting mitochondrial fusion to uphold mitochondrial quantity (Samant et al., 2014). Mitochondrial dysfunction constitutes a pivotal hallmark of liver fibrosis, often linked with calcium overload as a prominent instigator of mitochondrial dysfunction (Shi et al., 2005). This study revealed that HF diet-induced feeding led to a pronounced increase in mitochondrial Ca²⁺ uptake within the liver of largemouth bass, with GA treatment effectively counteracting this augmentation. The pivotal role of the MCU, situated on the inner mitochondrial membrane, in regulating mitochondrial Ca²⁺ uptake is well-established. Research indicates that interference or knockout of SIRT3 significantly amplifies MCU expression through acetylation of lysine 27 on the MCU promoter (Gao et al., 2020). GA's ability to activate SIRT3 was further underscored by its potential in reducing MCU expression, thereby ameliorating the excessive Ca²⁺ transfer from the endoplasmic reticulum to mitochondria. However, the intricate mechanism of this action necessitates further comprehensive exploration. The current study has focused solely on largemouth bass, which may result in certain species-specific limitations.

5. Conclusion

This study establishes that GA exhibits distinct anti-steatotic and anti-fibrotic effects, while simultaneously maintaining mitochondrial Ca²⁺ levels by regulating SIRT3 expression. This results suggest that GA's beneficial effects may be linked with its regulation of TGFβ1-Smad2/3 pathways, pro-fibrogenic marker genes α-SMA, collagen I, and fibronectin. Quadratic regression analysis revealed that the optimal dietary GA supplementation level for juvenile largemouth bass (body weight: 17.31–111.42 g) was 0.8 mg/kg. In conclusion, this study demonstrates that GA holds potential for preventing liver fibrosis in fish, providing a theoretical basis for its application in aquafeeds.

Credit Author Statement

Ju Zhao: Writing – original draft, Investigation, Methodology, Data curation. Haifeng Liu: Writing – original draft, Investigation, Methodology, Data curation. Zongjie Yang: Writing – original draft, Investigation, Methodology, Data curation. Lingjie He: Investigation, Methodology, Data curation. Hongying Shan: Investigation, Methodology. Caterina Faggio: Writing – review & editing. Quanquan Cao: Funding acquisition, Supervision, Conceptualization. Jun Jiang: Supervision, Conceptualization, Writing – review & editing, Funding acquisition.

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: