Establishment of a non‐alcoholic fatty liver disease model by high fat diet in adult zebrafish

Abstract

Background

Non‐alcoholic fatty liver disease (NAFLD) has become the most common chronic liver disease in recent years, but the pathogenesis is not fully understood. Therefore, it is important to establish an effective animal model for studying NAFLD.

Methods

Adult zebrafish were fed a normal diet or a high‐fat diet combined with egg yolk powder for 30 days. Body mass index (BMI) was measured to determine overall obesity. Serum lipids were measured using triglyceride (TG) and total cholesterol (TC) kits. Liver lipid deposition was detected by Oil Red O staining. Liver injury was assessed by measuring glutathione aminotransferase (AST) and glutamic acid aminotransferase (ALT) levels. Reactive oxygen species (ROS) and malondialdehyde (MDA) were used to evaluate oxidative damage. The level of inflammation was assessed by qRT‐PCR for pro‐inflammatory factors. H&E staining was used for pathological histology. Caspase‐3 immunofluorescence measured apoptosis. Physiological disruption was assessed via RNA‐seq analysis of genes at the transcriptional level and validated by qRT‐PCR.

Results

The high‐fat diet led to significant obesity in zebrafish, with elevated BMI, hepatic TC, and TG. Severe lipid deposition in the liver was observed by ORO and H&E staining, accompanied by massive steatosis and ballooning. Serum AST and ALT levels were elevated, and significant liver damage was observed. The antioxidant system in the body was severely imbalanced. Hepatocytes showed massive apoptosis. RNA‐seq results indicated that several physiological processes, including endoplasmic reticulum stress, and glucolipid metabolism, were disrupted.

Conclusion

Additional feeding of egg yolk powder to adult zebrafish for 30 consecutive days can mimic the pathology of human nonalcoholic fatty liver disease.

The zebrafish NAFLD model was generated by feeding adult male zebrafish with a high‐fat diet combined with egg yolk powder to induce lipid accumulation and histological pathology. Expression of gene transcripts was then analyzed by RNA‐seq. Our assay successfully established a rapid and effective zebrafish model that can mimic the progression of human NAFLD.

1. INTRODUCTION

Non‐alcoholic fatty liver disease (NAFLD) has become the most common liver disease worldwide over the past few decades due to the harmful effects of overweight and obesity on human health. ¹ The progression of NAFLD includes a series of different degrees of liver injuries, from nonalcoholic fatty liver (NAFL) to non‐alcoholic steatohepatitis (NASH). ² NAFL has a good prognosis, while NASH may progress to advanced liver fibrosis, cirrhosis and even hepatocellular carcinoma (HCC). ³ By 2018, the number of NAFLD patients in the world had reached 25% of the total population, ⁴ and the prevalence of NAFLD in China was even as high as 29.2%. ⁵ In spite of this huge medical demand, there is still no specific drug for the treatment of NAFLD/NASH in the clinic because of the complex pathogenesis. ⁶ At present, the scientific community has proposed more widely accepted pathogenesis mechanisms including the two‐hit theory and the multiple hit theory, but the complexity of NAFLD still cannot be fully summarized. ⁷ , ⁸ Therefore, the establishment of a fast and effective animal model of NAFLD is of great significance for exploring the molecular mechanism of NAFLD.

At present, the pathogenesis of NAFLD has been explored mainly using rodent models in vivo and cell models in vitro. ⁹ After years of development, rodent models have yielded good results. However, rodent models inevitably involve high feeding costs, large individual differences, and long experimental times, and are not suitable for large‐sample screening. ¹⁰ Most of the cell models in vitro are single cell cultures, but the pathogenesis of NAFLD is closely related to multiple tissues and organs, and because cell models grow in artificial environment, their morphology or function shows different degrees of change compared with the in vivo environment, meaning they cannot fully reflect real physiological and pathological changes.

Zebrafish (Danio rerio) were first used in the laboratory 30 years ago. Due to its many advantages such as low maintenance cost, rapid reproduction and development, and ease of inheritance control and gene editing, the zebrafish has attracted more and more attention as an in vivo model in many fields such as toxicology, developmental biology and drug screening. ¹¹ The zebrafish has become a model animal for studying NAFLD in recent years because of its metabolic consistency with humans and the highly conserved molecular program of liver development. ¹² , ¹³ At present, several models of nonalcoholic fatty liver disease have been established in zebrafish, including genetic, diet‐induced, and chemo‐induced models. ³ , ¹⁴ , ¹⁵ , ¹⁶ The disadvantage of genetic models is that mutant genes rarely occur in humans. ¹⁷ In contrast, drug‐induced models show high animal mortality. Diet‐induced models are widely used in the study of molecular mechanisms because they are most similar to human NAFLD pathogenesis. The MCD diet is a common model of NASH and can rapidly cause steatosis, necrosis, inflammation, and fibrosis. ¹⁸ However, its drawback is that as a nutrient‐deficient diet, it does not induce the same pattern of NAFLD pathogenesis as humans. ¹⁹ High‐fat diet‐fed mice are widely used since they present similar histological features and metabolic properties to those observed in humans. ²⁰ , ²¹ , ²² However, monitoring of rodents fed a high‐fat diet involves high costs and time requirements, whereas the use of zebrafish allows for the rapid construction of an economical and effective model of NAFLD. An et al. induced lipid accumulation and oxidative stress by feeding a high‐cholesterol diet to zebrafish larvae. ²³ Sapp et al. showed that fructose induced endoplasmic reticulum (ER) stress and oxidative stress in zebrafish larvae and successfully established a model of nonalcoholic fatty liver disease. ²⁴ However, most zebrafish NAFLD models use larvae. Although the age profile of NAFLD is getting younger, the juvenile version of NAFLD simulated by zebrafish larvae cannot fully reflect the pathogenesis in adults. An et al. established a model of NAFLD using adult zebrafish fed a high cholesterol diet for 8 weeks. ²⁵ Although the model was successfully established, it required a long experimental time.

In this study, we used adult male zebrafish as a vertebrate model. We established NAFLD by additional feeding of egg yolk powder, and evaluated the relationship between diet‐induced lipid metabolism disorders and related pathways in the zebrafish. The study aimed to more fully understand the effects of high‐fat diet on basic physical characteristics, liver histology and lipid metabolism, thereby extending our understanding of the pathogenesis of NAFLD.

2. MATERIALS AND METHODS

2.1. Care and feeding of adult Zebrafish

Embryos of the wild‐type zebrafish AB strain were collected and placed in embryonic water, and incubated at 28.5°C in a constant temperature incubator until 5 days post fertilization (dpf). Paramecium and prawns were fed in still water from 6 dpf, and cultured in the system at 15 dpf until adulthood (3 months post fertilization, mpf). Adult male zebrafish were randomly divided into a normal diet group (ND) and a high‐fat diet group (HFD), with 15 fish in each group. The ND group was fed with 30 mg of artemia per adult fish per day, while the HFD group was fed with 30 mg of artemia (20% fat, 60% protein and polyunsaturated fatty acids; Shangjia, Shandong, China) per adult fish per day and 70 mg of egg yolk powder (containing 55.8% fat, 34.2% protein, 17.2% saturated fatty acids; purchased from Yuanye Bio‐Technology, Shanghai, China) per adult fish per day (egg yolk powder thoroughly mixed with 2 L water until no precipitation was observed). The feeding lasted for 30 days. After feeding, the animals were fasted for a day before being killed. Based on standard laboratory guidelines, all zebrafish were fed in an environment at 28.5°C with a periodic rhythm of 14 h of light and 10 h of darkness. The zebrafish were treated according to the regulations of Guangdong Province on the Administration of Experimental Animals (2010).

2.2. Body mass index (BMI)

At the time of randomization (day 0), adult zebrafish were anesthetized with embryonic water containing 0.02% Tricaine. After drying the body surface water, the body weight (mg) was measured using an analytical balance, and the length (cm) from the fish mouth to the end of the tail fin was measured with a ruler. The BMI of adult zebrafish was calculated at the beginning and end of the feeding experiment using the formula: BMI = weight/length².

2.3. Liver to body ratio (LBR)

Adult zebrafish were anesthetized and the abdomen was opened with scissors and forceps to remove the intact liver. Livers were weighed using the analytical balance and the LBR was calculated using the formula: LBR = liver weight/body weight *100%.

2.4. Histology

The whole visceral mass of adult zebrafish sacrificed after anesthesia was completely removed with scissors and forceps, embedded in Tissue‐Tek O.C.T. compound (SAKURA, USA), and frozen on dry ice. The samples were fixed on a Leica CM1860 UV cryostat for collection of 8 μm thick serial sections on adhesive slides. After sample collection, the sections were fixed in frozen methanol for 5 min. After rinsing off excess methanol and O.C.T., the sections were stained with H&E staining kit (Solarbio, Beijing, China) according to the manufacturer's instructions. Samples for ORO staining were stained with 0.5% Oil Red O dye (Aladdin, Shanghai, China) solution without methanol fixation. The staining results were analyzed under a BX53 microscope (Olympus, Germany).

2.5. Biochemical analysis

After adult fish were sacrificed, intact livers were removed with scissors and forceps, and liver total cholesterol (TC), triglycerides (TG), reactive oxygen species (ROS), and malondialdehyde (MDA) were measured using relevant kits according to the manufacturer's specifications. After centrifugation of blood taken from the tail of the fish, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured using the appropriate kit (Jiancheng, Nanjing, China). ²⁶ , ²⁷ , ²⁸

2.6. Imaging

In order to obtain the liver changes in adult zebrafish, the abdomen was exposed and imaged using a Leica M205 FA stereo microscope (Leica, Germany).

2.7. Quantitative real‐time PCR

Fresh adult zebrafish liver tissue was used to extract total RNA using RNAiso Plus (Takara Bio), and a total of 1000 ng of RNA was reverse transcribed into cDNA using a reverse transcription kit (Yeasen Biotech). The cDNA obtained was mixed with specific primer pairs, the Hieff® qPCR SYBR Green Master Mix (Yeasen Biotech) with Low Rox and then subjected to real‐time PCR using the ABI QuantStudio 6 Flex system (Applied Biosystems). The sequences of primers are listed in Table S1.

2.8. Immunofluorescence (IF)

Samples were sectioned to 8 μm thick using a freezing microtome, fixed with acetone for 10 min on ice, and then washed with PBS. The samples were incubated with 5% goat serum for 1 h at room temperature, and incubated with primary antibody (1:150, ab13847, Abcam, UK) diluted with 5% goat serum overnight at 4°C in a refrigerator. The next day, the residual primary antibody was washed with PBS. The sections were incubated with diluted secondary antibody (1:200, 111‐605‐003, Jackson, PA, USA) and DAPI (1:500, Sigma, USA) for 2 h at room temperature, washed with PBS, sealed with anti‐fluorescence quenched sealing (Beyotime Biotechnology, Shanghai, China) solution, and fixed with nail polish. Images were obtained with an Olympus FV3000 confocal microscope.

2.9. RNA‐seq

The quality of RNA was examined using a Thermo Scientific NanoPhotometer Spectrophotometer, Qubit2.0 Fluorameter and an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). High quality RNA extracts from each tissue sample were pooled together for cDNA synthesis and sequencing. Fastp was used for quality control and the RNA‐seq reads were filtered to remove low quality sequences. The processed reads were then mapped to a reference zebrafish genome using HISAT2. The mRNAs and genes with differential expression levels were screened out on the basis of log2(fold change) >1 or < −1 and at p < 0.05 level. The subsequent functional and signaling pathway analyses of the above‐screened differential genes were conducted using Gene Ontology (GO; http://current.geneontology.org/) and Gene Set Enrichment Analysis (GSEA; http://www.gsea‐msigdb.org/).

2.10. Statistical analyses

All data were statistically analyzed using Prism software (version 8.0 GraphPad, San Diego, CA). The normality of data distribution was analyzed using the Shapiro–Wilk test. The unpaired Student's t test was used to compare the differences between groups. Data were expressed in the form of mean ± standard error of mean. p < 0.05 was considered as statistically significant.

3. RESULTS

3.1. Establishment of NAFLD zebrafish model

In order to establish the zebrafish model of NAFLD, adult male zebrafish were fed with artemia mixed with or without egg yolk powder for 30 days (Figure 1A), and NAFLD was characterized by observing the accumulation of lipids in the liver. ²⁹ At the end of the feeding assay, the obesity phenotype of zebrafish in HFD group (Figure 1B, right) was investigated and was compared with that in ND group (Figure 1B, left). After exposing the zebrafish belly, the livers of the HFD group (Figure 1C, right) presented a more obvious yellow color than those of ND zebrafish (Figure 1C, left). The results of H&E staining showed that the livers of zebrafish in the HFD group had inflammatory lesions (black arrow: aggregation of large numbers of nuclei), ballooning degeneration (with blank regions in cytoplasm, or nucleus hanging in the center or squeezed to the side), and larger areas of steatosis (p < 0.0001), indicating the progression of NAFLD (Figure 1D). In the ND group fed with artemia at 50 mg/fish per day, the steatosis area reached 30%–45% of total liver area. In the HFD group fed 50 mg artemia/fish and 150 mg/fish egg yolk powder per day, the steatosis area reached 35%–50% (Figure S1). Based on the comparison of area of steatosis in the HFD group with that of the ND group (Figure 1D), the following experiments were carried out.

FIGURE 1.

Feeding protocol, zebrafish body size and liver phenotype. (A) Feeding protocol. Adult zebrafish (3 mpf) were randomly divided into two groups. The normal diet (ND) group was fed 30 mg artemia/fish/day, while the high‐fat diet (HFD) group was fed 30 mg artemia/fish/day and 70 mg egg yolk/fish/day powder. feeding lasted for 30 days. (B) After completion of feeding, the HFD group (right) showed a significantly obese body shape compared to the ND group (left). (C) After exposing the abdomen, the liver of zebrafish in the HFD group (right) appears distinctly yellow, while the liver of zebrafish in the ND group (left) is orange‐red. Scale bar: 1 mm. (D) H&E staining results demonstrating steatosis and inflammatory cell infiltration. In the ND group 10–20% steatosis was observed, while in the HFD group 20–30% steatosis with marked inflammatory cell infiltration was observed. The black arrow indicates the regions where large numbers of nuclei are aggregated and the ballooning degeneration is seen as blank regions in the cytoplasm with the nucleus hanging in the center or squeezed to the side. The solid black box indicates the magnified area, and the high magnification image is shown below. Scale bars: yellow, 100 μm; black, 20 μm. 5 fields of view per fish (ND: n = 14; HFD: n = 11). The rate of steatosis was counted by Image J. Bar graphs, mean ± SEM. ****p < 0.0001.

3.2. HFD diet causes abnormal lipid accumulation in adult zebrafish

At the beginning of the feeding experiment, there was no significant difference in BMI between the two groups. At the end of the feeding experiment, the BMI of the HFD group (25.60 ± 0.38 mg/cm²) was significantly higher than that of the ND group (23.12 ± 0.33 mg/cm², p < 0.001, Figure 2A), indicating that significant obesity occurred in the fish of the HFD group. LBR was also significantly increased in the HFD group (1.90 ± 0.08%) compared to the ND group (1.56 ± 0.08%, p < 0.01, Figure 2B), suggesting lipid accumulation in the liver. The results of Oil Red O staining showed that a large number of red lipid droplets were present in the hepatocytes of the HFD group (13.12 ± 1.89%), while only a small number of lipid droplets were present in the ND group (0.57 ± 0.12%, p < 0.0001, Figure 2C). Hepatic TC and TG levels were significantly lower in the ND group (0.06 ± 0.01 mmol/g protein and 0.18 ± 0.02 mmol/g protein, respectively) compared with the HFD group (0.16 ± 0.01 mmol/g protein and 0.26 ± 0.02 mmol/g protein, respectively, p < 0.0001 and p < 0.05, respectively; Figure 2D,E). These results indicate that the HFD diet can lead to abnormal lipid accumulation in adult zebrafish.

FIGURE 2.

High‐fat diet causes lipid accumulation in zebrafish. (A) BMI at the beginning and at the end of feeding. No significant difference in BMI between ND and HFD groups at the beginning of feeding experiment was recorded (ND: n = 14; HFD: n = 15), but at the end of feeding, BMI was significantly higher in the HFD group compared to the ND group (ND: n = 14; HFD: n = 8). (B) Liver‐to‐body ratio (LBR) was significantly higher in the HFD group (ND: n = 13; HFD: n = 8). (C) ORO staining showed large lipid accumulation in the livers of the HFD group. The solid black box indicates the magnified area, and the high magnification image is shown below. Scale bars: yellow 100 μm, black 20 μm. 5 fields of view per fish (ND: n = 8; HFD: n = 6). The ratio of red lipid droplets was counted by Image J. (D) Zebrafish liver TC content (two livers were combined into one sample, ND: n = 4; HFD: n = 5). (E) Zebrafish liver TG content (two livers pooled into one sample, ND: n = 5; HFD: n = 5). Bar graphs show mean ± SEM. n.s., not statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

3.3. HFD diet induced pathological changes in adult zebrafish liver

The serum levels of AST (Figure 3A) and ALT (Figure 3B) in the HFD group (300.90 ± 37.39% and 180.10 ± 11.05%, respectively) were significantly higher than those in ND group (100.00 ± 5.00% and 100.00 ± 13.11%, respectively, p < 0.01). In addition, we measured liver ROS levels, which were significantly higher in the HFD group (6.52 ± 0.28) than in the control group (0.94 ± 0.18, p < 0.0001) (Figure 3C). At the same time, we detected the level of MDA, the damaging oxidative product of lipid metabolism, and also found that the level of the ND group (1.91 ± 0.79 mmol/mg protein) was significantly lower than that of the HFD group (6.91 ± 1.10 mmol/mg protein, p < 0.01) (Figure 3D). qRT‐PCR results showed that the expression levels of proinflammatory factors (tnf‐α, il‐1β, il‐6, il‐8) in the HFD group were significantly higher than those in the ND group (p < 0.01, p < 0.05 and p < 0.05, respectively) (Figure 3E). Caspase‐3 ³⁰ immunofluorescent staining results showed that severe apoptosis occurred in the liver of the HFD group, while the level of apoptosis was lower in the ND group (Figure 3F). These results suggest that HFD diet can lead to pathological changes in adult zebrafish liver.

FIGURE 3.

High‐fat diet causes pathological changes in the liver of zebrafish. (A) Zebrafish serum AST levels (serum from 5 fish in one sample, ND: n = 3; HFD: n = 4 repeats). (B) Zebrafish serum ALT levels (serum from 5 fish in one sample, ND: n = 3; HFD: n = 3 repeats). (C) Zebrafish liver ROS levels (two livers pooled into one sample, ND: n = 4; HFD: n = 4 repeats). (D) Zebrafish liver MDA levels (two livers pooled one sample, ND: n = 5; HFD: n = 5 repeats) (E) Liver RNA was extracted for qRT‐PCR assay to detect the expression levels of mRNAs of inflammatory factors. (F) Hepatocyte apoptosis was observed by staining with Caspase‐3 protein. The expression level of Caspase‐3 was significantly increased in the HFD group compared with the ND group, and hepatocytes were extensively apoptotic. The solid white box indicates the magnified area, and the high magnification image is shown below. Scale bars: white 100 μm, green 30 μm. 3 fields of view per fish (ND: n = 4; HFD: n = 4 repeats). Mean grayscale values were counted by Image J. Bar graphs show mean ± SEM. n.s. means not statistically significant, *p < 0.05, **p < 0.01, ****p < 0.0001.

3.4. RNA‐seq analysis of changes in gene expression levels during NAFLD progression

Following a high‐fat diet, a volcano map showed that the expression of 293 genes was significantly increased and 452 genes were significantly decreased in the high‐fat group (Figure S2). The results of GSEA and GO analysis showed that the expression of genes related to the response to endoplasmic reticulum stress and genes related to glycolipid metabolism were upregulated in the HFD group (Figure 4A,B). This is consistent with previously published results. ²⁰ , ³¹ , ³² , ³³ The genes associated with NAFLD in the GO analysis were networked and enriched with the STRING database, and then the cytoHubba app in Cytoscape software was used for module screening, and 20 core genes were obtained in order of score (Figure 4C). The expression of ER stress molecules (hspa5, dnajc3), fatty acid synthase (fasn) and insulin substrate receptor 2 (irs2) were significantly elevated in the HFD group as verified by qRT‐PCR (Figure 4D), demonstrating the occurrence of ER stress and disorders of glucolipid metabolism in the model group.

FIGURE 4.

RNA‐seq analysis of changes in gene transcript levels during NALFD progression. (A) GSEA analysis of differentially expressed genes with significant upregulation of response pathways to endoplasmic reticulum stress. (B) Bubble plot of GO analysis restlt, lipid synthesis, insulin response and endoplasmic reticulum stress pathways were significantly upregulated. (C) Screening results of cytoHubba app module with the top 20 highest scoring genes. (D) qRT‐PCR validation of mRNA expression levels of endoplasmic reticulum stress molecules, fatty acid synthase and insulin substrate receptor (more than three samples were tested for each gene). n.s., no statistical significance, *p < 0.05, **p < 0.01, ***p < 0.001.

4. DISCUSSION

In this study, a model for studying NAFLD was constructed using egg yolk powder fed to adult zebrafish; this model is more consistent with human epidemiological characteristics than models using juvenile fish. In addition, additional intake of high‐fat diet beyond that required to meet basic physiological needs can better mimic the pathogenesis and pathological characteristics of human NAFLD. Importantly, the dose of egg yolk powder used in the study allowed the model to be completed in a shorter period of time.

First, to ensure that each zebrafish consumed the same amount of food, we selected zebrafish of similar body size as the experimental subjects, and we dispersed the egg yolk powder evenly in the water to reduce any differences in feeding. Second, to obtain suitable feeding concentrations, we performed feeding experiments using two concentrations and compared them by H&E staining. It was found that the ND group fed 30 mg/fish/day had a lower area of steatosis and did not produce inflammatory lesions, which better mimicked conditions in the normal physiological state. Therefore, 30 mg artemia /fish per day was chosen as the basal feeding level in subsequent experiments.

Under normal conditions, only a very small amount of TG is stored in the lipid droplets of hepatocytes in the liver. However, excessive storage of TG leads to the development of lipotoxicity and steatosis. ³⁴ , ³⁵ TC in vivo includes cholesteryl esters and free cholesterol (FC), and the accumulation of FC has been shown to correlate with inflammation and fibrosis, among other symptoms, in the progression of NASH. ³⁶ Elevated TG and TC were detected in our experimental results, and the results of Oil Red O staining of frozen sections also showed the presence of a large number of lipid droplets in the HFD group, confirming the excessive accumulation of TG. In addition, the results of H&E staining also showed a large extent of fatty degeneration and inflammatory lesions.

An imbalance between the bioavailability of ROS and the antioxidant system in the body can lead to the development of oxidative stress. Low levels of ROS play physiological roles such as signaling in vivo, but high levels of ROS can lead to DNA damage, which in turn can cause alterations in the expression of certain genes, ultimately leading to increased cell death. We detected a significant increase in ROS levels in the HFD group, indicating that the antioxidant balance was disrupted. In addition, high levels of ROS were able to induce protein and lipid peroxidation, leading to functional alterations of substances such as enzymes and cell membranes. We detected a significant increase in MDA levels in the HFD group. Excessive accumulation of ROS affects the mitochondrial depolarization potential by inducing lipid peroxidation reactions to produce MDA. In the mitochondrial depolarized state, the burden of mitochondrial autophagy increases and the number of mitochondria decreases, promoting further development of steatosis, inflammation and fibrosis. ³⁷ , ³⁸ , ³⁹ Hepatocyte injury usually releases AST and ALT into the blood. Therefore, biochemical assays of serum AST and ALT are widely used as markers of liver injury. ⁴⁰ In zebrafish in the HFD group, significantly elevated serum AST and ALT levels were detected, suggesting that the high‐fat diet induced the process of liver injury.

In recent years, as the pathogenesis of NAFLD/NASH has been increasingly studied, the role played by endoplasmic reticulum (ER) stress has received wide attention and ER stress has been shown to play a regulatory role in apoptosis of hepatocytes. ⁴¹ In the presence of ER stress, the unfolded protein response (UPR) is initiated and undergoes repair. However, prolonged ER stress leads to a shift from adaptive to terminal UPR, followed by increased inflammation and activation of pro‐apoptotic pathways in the liver, resulting in massive hepatocyte death. ⁴² , ⁴³ Using a immunofluorescence assay, we detected a large amount of the apoptotic marker Caspase‐3 in the HFD group, suggesting that ER stress may have occurred in zebrafish in the HFD group. The dynamic mechanisms regulating ER stress in NAFLD are not fully understood, and although extensive studies have been conducted and many relevant molecules have emerged as new therapeutic targets, more in‐depth experiments are needed to explore the complete mechanisms. ⁴⁴

Analysis of the results of RNA‐seq and qRT‐PCR assays allowed us to identify several biochemical processes that are most altered in this model. First, fatty acid synthase (FASN) is a key rate‐limiting enzyme for hepatic de novo lipogenesis (DNL), ⁴⁵ , ⁴⁶ and we found that a high‐fat diet induced a rise in FASN expression, leading to increased lipid synthesis and further leading to TG accumulation in the liver. Second, altered expression of insulin receptor substrate 2 (IRS2) regulates the AKT pathway and affects NAFLD progression. ⁴⁷ Consistent with the results of Chi et al, ⁴⁸ we detected increased mRNA expression of irs2. Finally, the endoplasmic reticulum plays an important role in lipid biosynthesis in vivo, ⁴⁹ and consistent with previous findings, ⁵⁰ , ⁵¹ we detected increased expression of the ER stress chaperone molecule heat shock protein 5 (hspa5) and the DNAJ heat shock protein family member C3 (danjc3) in the HFD group. In conclusion, a high‐fat diet induces upregulation of the DNL pathway and increased lipogenesis, leading to excessive accumulation of lipid droplets in the liver. Lipid accumulation produces lipotoxicity, leading to mitochondrial dysfunction or endoplasmic reticulum stress, which, together with oxidative stress, activation of pro‐inflammatory substances and apoptotic pathways, leads to hepatocyte death and promotes the development of NAFLD.

Currently, research on the pathogenesis of NAFLD is limited, and no drugs have been approved for treatment. ¹⁴ Previously proposed theories such as the ‘double‐hit’ theory based on triglyceride accumulation, inflammation, oxidative stress and autophagy, ⁸ and the ‘three strike’ theory ⁵² based on steatosis, lipotoxicity and inflammation, have failed to summarize the complexity of human NAFLD. In recent years, the ‘multiple‐hit’ theory, which focuses on mitochondrial dysfunction, endoplasmic reticulum stress, inflammatory response, lipotoxicity, genetic factors, nutritional excess and intestinal flora disorders, has flourished. ⁷ Here, we constructed a short, convenient, and high‐volume experimental model of NAFLD in adult zebrafish. 3mpf male zebrafish were additionally fed with egg yolk powder for 30 days, and relevant indexes were tested to verify the success of the model, which successfully mimicked the pathological characteristics of NAFLD and will hopefully provide a stable and reliable in vivo model for drug screening and the study of NAFLD pathogenesis.

AUTHOR CONTRIBUTIONS

Xiang Li: carried out the animal experiment and drafted the manuscript. Yuying Zheng: information collection and revised the manuscript. LeiZhou: design methodology and revision of manuscript. Honghui Guo, Jiangbin Li and Taiping He: supervision, extra suggestions on this study. Jingjing Zhang: supervision, conceptualization, writing ‐ review and editing.

FUNDING INFORMATION

This study was supported by the Science and Technology Planning Project of Zhanjiang, China (2022A01231) and the Discipline Construction Project of Guangdong Medical University (4SG22258G).

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflicts of interest. Jingjing Zhang is an editorial board member of Animal Models and Experimental Medicine and a coauthors of this article. To minimize bias, he was excluded from all editorial decision making related to the acceptance of this article for publication.

ETHICS STATEMENT

This study does not contain any studies with human subjects or with rodent vertebrates. Handling of zebrafish was performed in accordance with Guangdong State Regulations on Laboratory Animal Management (2010). The whole experimental procedures were examined by the Clinical Ethics Committee of the Affiliated Hospital of Guangdong Medical University, China.

Supporting information

Appendix S1.

Li X, Zhou L, Zheng Y, et al. Establishment of a non‐alcoholic fatty liver disease model by high fat diet in adult zebrafish. Anim Models Exp Med. 2024;7:904‐913. doi: 10.1002/ame2.12309