A Study on How Methionine Restriction Decreases the Body’s Hepatic and Lipid Deposition in Rice Field Eel (Monopterus albus)

Methionine restriction reduces animal lipid deposition. However, the molecular mechanism underlying how the body reacts to the condition and regulates lipid metabolism remains unknown. In this study, a feeding trial was performed on rice field eel Monopterus albus with six isonitrogenous and isoenergetic feeds that included different levels of methionine (0, 2, 4, 6, 8, and 10 g/kg). Compared with M0 (0 g/kg), the crude lipid and crude protein of M. albus increased markedly in M8 (8 g/kg) (p < 0.05), serum (total cholesterol, triglyceride, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and non-esterified free fatty acids), and hepatic contents (hepatic lipase, apolipoprotein-A, fatty acid synthetase, total cholesterol, triglyceride, and lipoprteinlipase). However, in the serum, very-low-density lipoprotein and hepatic contents (hormone-sensitive triglyceride lipase, Acetyl CoA carboxylase, carnitine palmitoyltransterase, and mirosomal triglygeride transfer protein) decreased markedly in M8 (p < 0.05). The contents of hepatic C18:2n-6, C22:6n-3, and n-3PUFA in the M8 group were significantly higher than those in M0 (p < 0.05), and the contents of lipid droplets in M8 were higher than those in M0. Compared with M0, the hepatic gcn2, eif2α, hsl, mttp, ldlrap, pparα, cpt1, and cpt2 were remarkably downregulated in M8, while srebf2, lpl, moat2, dgat2, hdlbp, srebf1, fas, fads2, me1, pfae, and icdh were markedly upregulated in M8. Moreover, hepatic SREBP1 and FAS protein expression were upregulated significantly in M8 (p < 0.01). In short, methionine restriction decreased the lipid deposition of M. albus, especially for hepatic lipid deposition, and mainly downregulated hepatic fatty acid metabolism. Besides, gcn2 could be activated under methionine restriction.


Introduction
Recently, studies reported that soybean protein can be used to replace fish meal (FM) in aquatic feed [1]. However, methionine is the most limiting amino acid in soybean protein, and essential sulfur amino acids for fish [2] must be obtained from feed [3]. Methionine not only participate in the body's protein synthesis but also directly or indirectly (through transsulfuration, transamination, and transmethylation) regulates the body's metabolism as a signal molecule, mainly metabolizing into cysteine, creatine, carnitine, hydrogen sulfide, taurine, and glutathione for various metabolic purposes [4]. If an aquatic animal's methionine intake is deficient, the process of protein synthesis will be limited, metabolism will be disturbed, and the growth performance of the fish will be inhibited [2,5,6].
A previous study showed that methionine restriction enhances the clearance of glucose, promotes hepatic fat accumulation, and decreases muscular fat accumulation in rainbow trout (Oncorhynchus mykiss) [7]. In addition, methionine restriction suppresses the targets of amino acid response pathways in the primary muscular cells of turbot (Scophthalmus maximus L.), reduces cellular protein synthesis, enhances protein degradation, increases levels of intracellular free amino acid, and leads to amino acid degradation. Methionine restriction also reduces glycolysis and lipogenesis while stimulating lipolysis, decreases the intracellular lipid pool, remarkably enhances energy expenditure by stimulating the tricarboxylic acid cycle and oxidative phosphorylation, and upregulates general controlled nonderepressible 2 (gcn2, also encoded by eif2ak4) expression [8].
In the process of evolution, animals have gradually evolved the ability to adapt to a lack of essential nutrients such as essential amino acids. In vertebrates, gcn2 plays a key role in sensing essential amino acid deprivation and activates the translational derepression of specific mRNAs by inhibiting general translation initiation [9]. A study on the Cobia (Rachycentron canadum) showed that crude lipids were markedly elevated with a higher level of dietary methionine and then plateaued. Hepatic lipid synthesis genes (sterol regulatory element binding protein-1 (srebp1), fatty acid synthetase (fas), peroxisome proliferator activated receptor γ (pparγ), and stearoyl-CoA desaturase-1 (scd1)) were significantly upregulated when the animals were fed a diet with higher levels of methionine, whereas the expression of lipolytic genes (peroxisome proliferator activated receptor α (pparα), carnitine acyl transferase-1 (cpt1), and lipase lipoprotein lipase (lpl)) was elevated in fish fed a methionine-deficient diet [10]. Guo et al. (2007) found that the adipose tissue of wild-type mice lacking leucine decreased by 50% after one week and almost completely disappeared after 17 days. Further research found that when leucine was deficient in the diet, gcn2 was activated, and its downstream eif2α, the level of mRNA, and protein expression increased. However, there was no significant difference observed in the expression of srebp1a and srebp2 mRNA and protein, although srebp1c mRNA and protein expression were significantly inhibited. Moreover, the expression of srebp1c mRNA and protein was regulated by the gcn2-eif2α pathway. The expression of fat synthesis genes (srebp1c, ATP-citrate lyase (acl), fas, scd, glucose 6-phosphate 1-dehydrogenase (g6pd), and malic enzyme (me)) occurred downstream, and hepatic SREBP1 and FAS protein expression was downregulated. The authors also found that a diet lacking leucine led to an increase in lipid absorption and fatty acid oxidation in the livers of mice, suggesting that the increase of lipid absorption and decomposition in mice under the condition of leucine deficiency was an adaptive change to reduce lipid synthesis [11].
Rice field eel (Monopterus albus, M. albus) is a subtropical freshwater benthic fish that is widely raised in central and southern China in cages [12]. Our previous studies showed that M. albus needs better-quality and higher levels of protein, as well as an optimum protein/lipid ratio [13]. In the study, FM was replaced by soybean meal [14], and soy protein concentrate inhibited the growth performance of M. albus [15]. Moreover, dietary deficiency methionine feed decreased the growth performance of M. albus, induced lipid metabolism disorder, and decreased lipid content [16]. Our laboratory is focused on studying the nutrition of M. albus. We also consulted a large number of papers of M. albus and found no obvious adipose tissue in M. albus. Lipids mainly accumulated in tissues, especially in the liver, which provides a new and interesting avenue for exploring lipid metabolism. In the present study, we generated more severe methionine-deficient diets compared to our previous study [16] and explored the mechanism by which methionine regulates lipid deposition and the metabolism of M. albus.

Composition of M. Albus
There was no significant difference in the moisture and crude ash of M. Albus among all groups (p > 0.05). The crude lipid of M. albus increased markedly as methionine concentrations increased to 8 g/kg (M8) (p < 0.05) and gradually decreased as methionine concentrations increased to 10 g/kg (M10). The crude protein of M. albus increased markedly as methionine concentrations were increased (p < 0.05) ( Table 1). Values are presented as the means ± SEM (n = 3). Values in the same row with the same superscript or the absence of superscripts are not significantly different (p > 0.05) (the same below).

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2).

Discussions
Our previous study showed that dietary methionine restriction induced lipid me olism disorder, decreased the lipid content [16], and also decreased the growth per

Discussions
Our previous study showed that dietary methionine restriction induced lipid metabolism disorder, decreased the lipid content [16], and also decreased the growth performance of M. Albus [17]. In the present study, the crude lipids of M. albus increased markedly as methionine concentrations increased to 8 g/kg and gradually decreased as methionine concentrations increased to 10 g/kg. The crude protein of M. albus increased markedly as methionine concentrations were increased. Our results are similar to those of a study on juvenile yellow tail (Seriola dorsalis) [18]. We inferred that more energy was allocated to visceral organs to maintain basic metabolism while fewer nutrients were allocated to growth performance when methionine was restricted.
Amino acids are commonly involved in life activities through the synthesis of proteins. Excess amino acids are generally decomposed into ammonia and carbon skeletons, while ammonia is further metabolized into urea nitrogen [19]. The acid phosphatase (ACP) enzyme is involved in protein pinocytosis and intracellular digestion [20]. Alkaline phosphatase (AKP) is a key enzyme with a protective role in fish under stress, parasitic infection, and wound healing [21]. In this study, serum ACP increased markedly when supplemented with a suitable level of methionine (8 g/kg), the serum AKP decreased markedly as methionine concentrations increased to 8 g/kg, and hepatic AKP significantly increased when supplemented with methionine. Transaminases are produced by the liver. Aspartate aminotransferase (AST) primarily transfers the amino of aspartic acid to a-ketone glutaric acid, producing oxaloacetic acid and glutamic acid, while alanine aminotransferase (ALT) primarily transfers the amino of alanine to a keto-glutamic acid, producing pyruvate and glutamic acid; these acids are also the main indexes used to evaluate hepatic injury [22]. In this study, the serum ALT and AST decreased markedly under supplementation with methionine (8 g/kg). Meanwhile, the hepatic ALT and AST increased markedly when supplemented with methionine concentrations greater than 4 g/kg. Moreover, the serum total protein, blood urea nitrogen, and blood ammonia increased markedly when supplemented with a suitable level of methionine (8 g/kg). The contents of hepatic amino acids, total essential amino acids, total nonessential amino acids, and total amino acids also increased. This phenomenon increased the utilization efficiency of amino acid. In this study, we also observed that the proportions of vacuoles decreased under supplementation with methionine (8 g/kg). Meanwhile, the nucleus moved and blurred the boundaries of hepatic cells when methionine was restricted. We concluded that suitable methionine may be better for hepatic amino-acid metabolism and a healthy condition, as we reported in [16].
Interestingly, we also observed that the serum glucose, total cholesterol, and triglycerides increased significantly with 8 g/kg dietary methionine. Meanwhile, the hepatic total cholesterol and triglycerides increased markedly when supplemented with higher than 2 g/kg methionine in this study. High-density lipoprotein (HDL) and low-density lipoprotein (LDL) are major lipoproteins produced by the liver. LDL transports lipid molecules from the liver around the body, while HDL carries lipids from the surrounding tissue into the liver. These lipoproteins mainly carry cholesterol and are formed as HDL-C and LDL-C, respectively [23]. Very-low-density lipoprotein (VLDL) is secreted by hepatocytes of the liver; the large sizes of VLDL particles secreted by the liver result in major disturbances to lipoprotein metabolism [24]. Hormone-sensitive lipase (HSL) regulates lipolysis, especially in adipose tissue [25]. Microsomal triglyceride transfer protein (MTP) facilitates the transport of fat by assisting in the assembly and secretion of triglyceride-rich apolipoprotein [26]. Apolipoprotein A-1 (ApoA1) is considered to be an important factor in lipid transport and metabolism in various tissues [27]. Lipoprotein lipase (LPL) is a key enzyme in lipid metabolism and primarily catalyzes the hydrolysis of triglycerides in chyle particles and very-low-density lipoprotein [28]. Fatty acid synthase (FAS) is involved in fatty acid synthase [29], and Acetyl-CoA carboxylase (ACC) is the rate-limiting enzyme for fatty-acid synthesis [30]. Carnitine palmitoyltransterase (CPT) participates in the process of fatty acid β-oxidation [31]. In the present study, hepatic HL, Apo-A, FAS, and LPL increased markedly when supplemented with methionine, while hepatic HSL, ACC, CPT, and MTTP decreased markedly when methionine was added. This phenomenon indicated that methionine restriction not only inhibited amino-acid metabolism but also disturbed lipid metabolism. Our results showed that dietary methionine offers benefits for lipid metabolism. This phenomenon is similar to that observed in Cobia (Rachycentron canadum) [10].
In addition, the lipid droplets (visualized by hepatic Oil red O staining) was increased in the group supplemented with methionine (8 g/kg). This result intuitively shows the difference in the hepatic lipid deposition of M. Albus between the M0 (0 g/kg) and M8 (8 g/kg) groups. To further explain the reasons why methionine deficiency affects the lipid metabolism of M. albus, we chose the M0 (0 g/kg) and M8 (8 g/kg) groups to explore the molecular mechanisms of lipid metabolism. gcn2 and eif2a respond to essential amino acid deprivation and regulate the downstream genes related to lipid metabolism [32]. In this study, compared to M0 (0 g/kg), hepatic gcn2 and eif2α were remarkably downregulated in M8 (8 g/kg), which means that the gcn2 and eif2α genes may be regulated by different levels of methionine. Thus, we determined the genes related to lipid metabolism and explored the relationship between amino-acid sensing and lipid metabolism. Sterol regulatory element binding transcription factor (srebf ), including srebf1 (mainly regulates fatty acids biosynthesis) and srebf2 (mainly regulates cholesterol synthesis), controls cellular lipid metabolism and homeostasis and performs functions in lipid biosynthesis and uptake-gene expression [33]. scap (srebf cleavage-activating protein) is a sterol-regulated escort protein that transports srebf from its site of synthesis in the endoplasmic reticulum to its site of cleavage in the Golgi [34]. Peroxisome proliferator-activated receptor α (pparα) mainly controls the β-oxidation of fatty acids [35], while peroxisome proliferator-activated receptor γ (pparγ) regulates the adipogenic and lipogenic pathways [36]. mogat2, dgat2, me1, me2, fas, fads2, and acc are key enzymes involved in lipogenesis and fatty-acid synthesis [37][38][39][40][41][42][43], while lpl, hsl, cpt1, and cpt2 are key genes involved in lipolysis and fatty-acid β-oxidation [44,45]. icdh is one of the key enzymes involved in the production of NADPH, which is an essential cofactor for fat cholesterol biosynthesis and fat metabolism [46]. The polyunsaturated fatty acid elongase (pfae) gene encodes desaturase and elongase enzymes with all the activities required for the production of long-chain polyunsaturated fatty acid [47]. Here, compared to M0 (0 g/kg), hepatic pparα, cpt1, and cpt2 were remarkably downregulated in M8 (8 g/kg), while hepatic srebf1, srebf2, lpl, moat2, dgat2, fas, fads2, me1, pfae, and icdh were upregulated in M8 (8 g/kg). We also observed that lipid synthesis genes were upregulated under a dietary-suitable level of methionine, while genes related to lipid catabolism were downregulated. These phenomena observed in the present study are similar to those observed in a previous study on Cobia (Rachycentron canadum) [10,48] and large Yellow croaker (Larimichthys crocea) [49]. We also found that hepatic SREBP1 and FAS protein expression was upregulated significantly in M8 (8 g/kg). Interestingly, hepatic C18:2n-6, C22:6n-3, and n-3PUFA remarkably increased when supplemented with methionine (8 g/kg). Thus, we determined that M. albus dietary intake deficient in methionine mainly affected fatty-acid metabolism, specifically unsaturated fatty-acid synthesis.
Microsomal triglyceride transfer protein (mttp) facilitates the transport of fat by assisting in the assembly and secretion of triglyceride-rich lipoproteins [26]. High-density lipoprotein-binding protein (hdlbp) mainly participates in the endocrine regulation of both lipids and cholesterol [50], while low-density lipoprotein receptor adapter protein (ldlra) maintains levels of homeostatic LDL. Moreover, the ldlra pathway has emerged as a target to reduce circulating cholesterol [51]. The very-low-density lipoprotein receptor (vldlr) receptor binds triglyceride-rich lipoproteins, along with lpl [52]. In this study, hepatic mttp and ldlrap were remarkably downregulated when supplemented with methionine (8 g/kg), while hdlbp was up strongly regulated in M8 (8 g/kg). This result indicates that lipid metabolism is more active if the feed intake of M. albus features a suitable level of methionine. Moreover, hepatic eif2α, scap, hsl, mttp, ldlrap, pparα, cpt1, and cpt2 gene expression was positively correlated with gcn2, and hepatic srebf2, lpl, moat2, dgat2, hdlbp, vldlr, srebf1, fas, fads2, me1, pfae, and icdh gene expression was negatively correlated with gcn2. These results indicate that gcn2 could respond to the condition of methionine restriction in M. albus and regulate lipid metabolism genes. However, the specific mechanism by which gcn2 regulates hepatic lipid metabolism requires further study.

Ingredients and Experimental Diets
The basic diet (110 g/kg fish meal and 400 g/kg soy protein concentrate) was based on our previous data [15,16]. Different levels of methionine (0, 2, 4, 6, 8, or 10 g/kg) were supplemented in the basic diet based on the rule of equal nitrogen and our previous studies [14,16], showed in Tables 6-8. Proximate analysis (moisture, crude protein, crude lipid, ash, and gross energy) of experimental feed and M. albus was performed based on our previous papers [53]. Amino acids were analyzed by an automatic amino acid analyzer (Agilent-1100, Agilent Technologies Co., Ltd., Santa Clara, CA, USA) based on Wijerath's method [54], and fatty acids were analyzed by GC-MS (Agilent 7890B-5977A, Agilent Technologies Co., Ltd., Santa Clara, CA, USA) based on Jin's method [55], the results are shown in Tables 2 and 3.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2). 15  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2). 36  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2).  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2  MUFAs: mono-unsaturated fatty acids. 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2). 29 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2). 18 3 n-6PUFAs: n-6 poly-unsaturated fatty acids. 4 n-3PUFAs: n-3 poly-unsaturated fatty acids.

Hepatic H&E and Oil Red O-Stained Pictures
Hepatic H&E and Oil red O-stained images are shown in Figures 1 and 2. Vacuoles were observed in these two groups. As the number of vacuoles increased in the M0 group, the proportions of vacuoles decreased in the M8 (8 g/kg) group. We also observed movement of the nucleus in M0 (0 g/kg) and blurred boundaries of hepatic cells in the M0 group. Compared to M0 (0 g/kg), the number of lipid droplets was increased in the M8 (8 g/kg) group (Figures 1 and 2). essential amino acids; Trp not detected.

Fish Rearing and Management
M. albus was obtained from Changde, China. M. albus of uniform size (25.08 ± 0.31 g) was stochastically divided into 18 float cages (2.0 m × 1.5 m × 1.5 m). Each group contained triplicates with 60 fish, based on our previous study [23].

Ethics Statement
Our study was supported by the Animal Care Committee of Hunan Agricultural University (Changsha, Hunan, China) and conducted according to the Chinese guidelines for animal welfare. According to the guidelines established by the National Institutes of Health, all experimental fish were anesthetized with eugenol (1:12,000; Shanghai Reagent Corporation, Shanghai, China). Ethic code number: 2021094; date of Ethics Statement: 13 December 2021.

Sample Collection and Analyses
After the feeding trial, the caudal vein blood was heparinized from five fish in each cage. Serum (3500× g) was obtained by centrifugation for 10 min and then stored at −80 • C until use. Serum alanine aminotransferase, aspartate aminotransferase, acid phosphatase, alkaline phosphatase, glucose, lactate dehydrogenase, total cholesterol, triglyceride, total protein, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, nonesterified free fatty acids, blood urea nitrogen, and blood ammonia were determined by a kit from NanJing JianCheng Bioengineering (Nanjing, China). Serum very-low-density lipoproteins were determined using a kit from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China).
Hepatic amino acids were analyzed by an automatic amino acid analyzer (Agilent-1100, Agilent Technologies Co., Ltd., Santa Clara, CA, USA), and hepatic fatty acids were analyzed by GC-MS (Agilent 7890B-5977A, Agilent Technologies Co., Ltd., Santa Clara, CA, USA) using the same method.
The liver was taken from five fish per cage for histometric evaluation. The methods for creating slides and observing the muscular sections stained with H&E were based on those used in our previous paper [17]. The liver was sectioned (8 µm) using a cryostat microtome and stained with Oil Red O [56]. The slides were then observed using CaseViewer.
Total hepatic RNA was obtained from five fish per cage using the Monzol™ reagent (Monad, Shanghai, China). Smart cDNA was synthesized using a SMART cDNA Synthesis kit (Clontech Laboratories, Palo Alto, CA, USA). Primers were synthesized by Biosune Biotechnology, Inc. (Shanghai, China), as shown in Table 4. Quantitative real-time PCR (qPCR) was performed as described in our previous paper [57]. The amplification efficiency was between 0.95 and 1.10, as calculated by the formula E = 10*(−1/slope)−1, and 5-fold serial dilutions of cDNA (triplicate) were used to generate the standard curve. The 2 − Ct method was used to calculate the relative mRNA expression [58].
Hepatic proteins were extracted from the liver with a lysis solution. After centrifugation for 5 min at 12,000 rpm/min and 4 • C, we determined the content of protein, ensured the protein concentrations were consistent, and used the concentrations for Western blot analysis. The first antibody was as follows: GAPDH Mouse Monoclonal antibody (proteintech, catalog number: 60004-1-Ig), SREBP1 anti-Rabbit pAb (Wanleibio, WL02093), and FAS anti-Rabbit pAb (Wanleibio, WL03376). We used the ImageJ software to calculate the expression of the protein (Table 9). Data Availability Statement: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest:
There are no conflict of interest to this manuscript.