Six-week-old adult male C57BL/6 mice were obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). Animal experiments were carried out in accordance with the approved guidelines involved in animal care by the Committee for Animal Research at Ningbo University. All mice were kept in a controlled temperature (21–23°C) and a constant humidity (55–60%) under a 12/12 h light/dark cycle, while food and water were provided ad libitum. The NASH model was established via feeding a methionine- and choline-deficient (MCD) diet, which can mimic the clinical characteristics of human NASH (Machado et al., 2015). The methionine and choline supplemented diet (normal diet, ND) and MCD diet (Trophic Animal Feed High-Tech Co., Ltd., Nantong, China) used in this study both contained 4.2 kJ/g energy, and the composition of the diet is shown in Supplementary Table S1. The energy in a normal diet was derived from 18% of protein, 60% of carbohydrate, and 22% of fat, while the MCD diet contained energy from 17% of protein, 61% of carbohydrate, and 22% of fat. To examine the preventive effect of PEA on MCD-induced NASH in mice, the mice were all randomly grouped as follows (n = 6–8/group): 1) Ctrl: control group fed with normal diet; 2) MCD: NASH model group fed with MCD diet and treated with the vehicle of PEA; 3) PEA: PEA group fed with MCD diet and intraperitoneal injection with PEA (10 mg/kg/day). The mice in the MCD group and PEA group were adaptively fed with a mixed diet of a normal diet and MCD diet for 2 weeks (2:1 for 4 days; 1:1 for 4 days; 1:2 for 6 days) and then received MCD feeding or MCD feeding + PEA treatment for 7 weeks. The experimental animal scheme is shown in Figure 1A.

The m-PEA was obtained from Wuxi Cima Science (Wuxi, China). Hematoxylin-Eosin (HE) staining kit was purchased from Solarbio Life Sciences (Beijing, China). Oil Red O was from Sigma-Aldrich (Shanghai, China). Tween-80 and PEG-400 were purchased from Sangon Biotech (Shanghai, China). Anti-PPAR-α antibody (#ab24509) and anti-CD68 antibody (#ab125212) were obtained from Abcam (Shanghai, China). Anti-caspase-1 antibody (clone 14F468) was purchased from Santa Cruz (Shanghai, China). Anti-GAPDH antibody (clone 1E6D9) was purchased from Proteintech (Wuhan, China). Anti-NLRP3 antibody (#A5652) was acquired from ABclonal Technology (Wuhan, China). Anti-α-SMA antibody (clone 1A4) and anti-LC3B antibody (L7543) were obtained from Sigma-Aldrich (Shanghai, China). Antibodies against p62 (#A19700), Beclin1 (#A11761), and ATG7 (#A0691) were obtained from ABconal (Wuhan, China). Horseradish peroxidase (HRP)-conjugated goat anti-mouse and goat anti-rabbit immunoglobulin G (IgG) were purchased from Proteintech (Wuhan, China).

Liver tissue samples were collected and fixed (4% paraformaldehyde solution) at 4°C overnight, embedded with paraffin, and cut into 5 μm thick sections by sliding microtome (Leica, Shanghai, China). Oil Red O staining, HE staining, and Masson trichrome staining were performed to assess lipid accumulation, liver morphological characteristics, and liver fibrosis with standard procedure. The histological features were observed and imaged under the microscope (Leica, Shanghai, China), and the measurement of NAFLD activity score was based on the degree of steatosis, hepatocellular ballooning, and inflammation as previously described (Dyson et al., 2014).

After being blocked with goat serum for antigen retrieval, the liver sections were incubated with primary antibodies against CD68 or α-SMA at 4°C overnight. The secondary antibody was applied to the sections after washing and incubated at room temperature for 1 h. The 3,3-diaminobenzidine (DAB) stained areas were observed and imaged under the microscope (Leica, Shanghai, China). The immunostained areas of liver sections were quantified with ImageJ software.

Alanine transaminase (ALT), aspartate transaminase (AST), lactate dehydrogenase (LDH), malondialdehyde (MDA), glutathione (GSH), glutathione peroxidase (GSH-px), and superoxide dismutase (SOD) levels were measured by commercial assay kits acquired from Nanjing Jiancheng Bioengineering Institute (Nanjing, China) according to the manufacturer’s instructions. Triglycerides (TG) levels were detected by the TG quantification kit (BioVision, Milpitas, United States). Hepatic fatty acids (palmitic, palmitoleic, and oleic acids) levels were analyzed as previously reported (Chen et al., 2014). The determination and quantification of PEA were performed as described by our previous study (Jin W. et al., 2020).

The levels of TNF-α, MCP-1, RANTES, IL-1β, and IL-18 in each plasma sample were assessed by commercial ELISA kits (MultiSciences Biotech, Hangzhou, China) according to the manufacturer’s instructions.

Proteins from liver tissues were extracted and detected following a routine protocol, as previously reported (Hu et al., 2020). Briefly, proteins were loaded on SDS-PAGE gel for separation and transferred onto polyvinylidene fluoride membranes. These membranes were blocked in 5% fat-free milk at room temperature for 1 h and then incubated with anti-PPAR-α (1:1,000), anti-α-SMA (1:1,000), anti-LC3B (1:1,000), anti-p62 (1:1,000), anti-Beclin1 (1:1,000), anti-ATG7 (1:1,000), anti-NLRP3 (1:1,000), anti-caspase-1 (1:500), or anti-GAPDH (1:10,000) antibody overnight at 4°C. After washing with TBST three times, the membranes were incubated with a horseradish peroxidase-conjugated secondary antibody at room temperature for 1 h, following the visualization using an enhanced chemiluminescence agent. The quantitation of target proteins was performed by the ImageJ software and normalized to the GAPDH levels.

Total RNA was extracted from liver tissues using RNA simple Total RNA Kit (Tiangen, Beijing, China). The mRNA was used to synthesize cDNA by a FastQuant cDNA kit with gDNase (Tiangen, Beijing, China). The quantitation of mRNA was conducted with a SYBR Green kit (Tiangen, Beijing, China) according to the manufacturer’s instructions and normalized to GAPDH. The primer sequences are listed in Table 1.

All statistical analyses were performed using GraphPad Prism version 8.3.0 software. The experimental results are presented as the mean ± standard error of the mean (SEM). Statistical differences between multiple groups were analyzed using one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test. Statistical significance was set as * p < 0.05, ** p < 0.01, *** p < 0.001.

To validate the effects of PEA in the development of NASH, PEA was administrated to MCD diet-induced NASH mice. The experimental scheme is shown in Figure 1A. MCD diet feeding caused a significant reduction of body weight, while PEA had no obvious impact on body weight compared with the MCD diet-fed mice during the experiment (Figure 1B). After sacrificing the mice, we first detected the role of PEA on MCD-induced NASH in mice through a histopathological evaluation, which is the gold standard for NASH assessment. HE staining showed that MCD diet-induced hepatic steatosis and ballooning compared to the mice who received a normal diet, while PEA treatment reduced both steatosis and ballooning (Figure 1C). Meanwhile, Oil Red O staining revealed that MCD diet feeding caused excessive lipid accumulation, while PEA significantly reduced the abundant lipid distribution in the liver (Figure 1D). Additionally, the high TG levels in the liver of MCD diet-treated mice were also obviously downregulated (Figure 1E). To further investigate the role of PEA on liver injury in MCD-treated mice, we determined the levels of ALT, AST, and LDH in plasma by a microplate-based method. MCD diet greatly induced the increase of plasma ALT, AST, and LDH levels, while PEA significantly decreased the levels of these biomarkers of liver function (Figures 1F–H). These results suggest that PEA significantly attenuated steatohepatitis in MCD diet-treated mice. In addition, we detected the levels of palmitic acid, palmitoleic acid, and oleic acid in the liver and found that palmitic acid levels were unchanged after being fed with a MCD diet, while hepatic palmitoleic acid and oleic acid levels were significantly decreased in mice fed MCD diet. Nevertheless, PEA treatment had no effect on all the above fatty acids compared with MCD-treated mice (Supplementary Figures S1A–C). Moreover, MCD diet feeding markedly elevated PEA levels, while PEA treatment enhanced this elevation compared to the Ctrl group (Supplementary Figure S1D). Meanwhile, we examined the expression of N-acylethanolamine-hydrolyzing acid amidase (NAAA) and fatty acid amide hydrolase (FAAH), both of which can hydrolyze endogenous PEA to palmitic acid and ethanolamine. The results showed that NAAA expression in the MCD group is higher than that in the Ctrl group and a reduced expression was observed in the PEA group, but these differences were not statistically significant (Supplementary Figure S1E). Notably, FAAH expression was obviously decreased with MCD feeding and unchanged after PEA treatment (Supplementary Figure S1F).

To explore the protective mechanism conferred by PEA on lipid accumulation in MCD-treated mice, we examined the gene expression related to lipid metabolism. As shown in Figures 2A, B, PEA treatment obviously reduced the mRNA expression of acetyl-CoA carboxylase 1 (ACC1) and CD36, which has a major effect on the biosynthesis and transporter of fatty acids, respectively. Moreover, PEA markedly upregulated the mRNA and protein expression levels of PPAR-α, which could promote fatty acid oxidation (Figures 2C–E). These results indicated that PEA could reduce lipid accumulation by decreasing ACC1 and CD36 and promoting the expression of PPAR-α. As the AMPK signaling pathway is the key metabolic pathway involved in regulating the above lipid metabolism associated genes, we further detected the protein levels of p-AMPK and AMPK, and the results suggested that the PEA could activate the AMPK signaling pathway in MCD-fed mice (Figures 2F, G).

Oxidative stress is a common characteristic relative to the development of NASH (Chen et al., 2020). To assess the anti-oxidative activity of PEA, we detected the hepatic MDA levels, which is an important biomarker of oxidative stress (Zelber-Sagi et al., 2020). The data showed that the MCD diet obviously elevated the hepatic MDA levels, while PEA treatment significantly reduced MDA levels in the liver (Figure 3A). GSH is a powerful antioxidant in the liver, which plays a key role in radical scavenging (Delli Bovi et al., 2021). The results showed that MCD diet induced decreasing in GSH levels in the liver, which were markedly increased by the PEA treatment (Figure 3B). GSH-px and SOD are both enzymes with anti-oxidation activities to protect the liver from oxidative stress (Delli Bovi et al., 2021). We further found that the mice given the MCD diet showed reduced hepatic GSH-px and SOD activities, and PEA significantly reversed these changes (Figures 3C, D). Consistently, MCD diet feeding resulted in decreased mRNA levels of SOD1 and SOD2, and PEA treatment obviously increased SOD1 and SOD2 expression (Figures 3E, F). The above data showed that PEA treatment significantly alleviated oxidative stress in the liver.

Inflammatory response is a key mechanism in the progress of NASH (Schuster et al., 2018). In order to verify the anti-inflammatory effects of PEA in MCD-induced NASH mice, we first detected the expression of CD68, a marker of Kupffer cells, via IHC staining. We found CD68 expression was greatly increased in liver tissues from MCD diet-treated mice, while PEA treatment significantly reduced this increase (Figure 4A,B). Next, we indicated that PEA dramatically attenuated MCD diet-induced production of inflammatory mediators, including MPO, iNOS, TNF-α, RANTES, and MCP-1 (Figures 4C–G). Moreover, the elevated expression of TNF-α, RANTES, and MCP-1 induced by MCD diet in plasma was also significantly decreased by PEA treatment (Figures 4H–J). These results suggested that PEA administration obviously inhibited hepatic inflammation in MCD diet-induced NASH mice.

Several recent reports have demonstrated the crucial effects of NLRP3 inflammasome in the progression of NASH (Huang et al., 2021). To confirm the impact of PEA on the activation of the NLRP3 inflammasome, we examined the expression pattern of NLRP3, pro-caspase-1, and activated caspase-1 in the liver. As a result, MCD-induced elevated protein levels of NLRP3, pro-caspase-1, and activated caspase-1 were significantly inhibited by PEA treatment (Figures 5A–D). To further determine the potential role of PEA on the NLRP3 inflammasome pathway in MCD-treated NASH mice, we tested plasma IL-1β and IL-18 levels through ELISA assays. We found that plasma IL-1β and IL-18 levels in mice fed with MCD diet increased compared to those in mice fed with standard diet and were markedly reduced after PEA treatment (Figures 5E, F). The above results indicated that PEA could substantially block the NLRP3 signaling pathway in NASH mice fed with an MCD diet.

Numerous previous studies have indicated that autophagy plays an important part in the progress of NASH (Amir and Czaja, 2011; Allaire et al., 2019). To detect whether PEA regulates autophagy in mice who received MCD diet feeding, we analyzed the hepatic expression of Beclin-1, ATG7, LC3, and p62 via western blotting. We observed that PEA treatment greatly increased the protein levels of Beclin-1, ATG7, and LC3-II/LC3-I compared to the MCD group (Figures 6A–D). In addition, the low level of p62 protein in the MCD group was significantly upregulated after PEA treatment (Figure 6E). These data suggested that PEA reactivated autophagy signaling in MCD-treated mice.

In mice fed the MCD diet, ongoing chronic liver injury ultimately results in liver fibrosis (Chen et al., 2015). To determine whether PEA affects MCD-induced nutritional fibrosis in mice, we measured the hepatic collagen deposition via Masson’s trichrome staining and found that the mice who received MCD diet feeding had much more collagen deposition than that of the control mice. However, the Masson staining area was obviously reduced in the PEA-treated mice (Figure 7A). α-Smooth muscle actin (α-SMA) is an important biomarker of hepatic stellate cells activation, which drives the progression of liver fibrosis. The protein levels of α-SMA were quantified by IHC staining and western blotting; the results revealed a marked increase of α-SMA expression in mice fed MCD diet compared to the mice given a normal diet, while PEA treatment obviously decreased the protein levels of α-SMA in comparison with the MCD group (Figures 7B, C). Moreover, PEA treatment also greatly decreased the MCD-induced mRNA expression of α-SMA (Figure 7D). Furthermore, we detected the mRNA levels of fibrotic genes, including CTGF, TGF-β, Col1a, and Col3a. The results demonstrated that the genes involved in liver fibrogenesis were significantly elevated in the MCD group compared with that in the control group, while PEA treatment obviously reversed these increases (Figures 7E–H). These results revealed the anti-fibrotic effects of PEA in MCD-induced NASH mice.