Non-alcoholic fatty liver disease (NAFLD) is one of the most prevalent chronic liver diseases, characterized by abnormal lipid accumulation in the liver, accompanied by inflammation and oxidative stress (Sahu et al., 2023). Its incidence is on the rise globally, posing a significant threat to human health (Riazi et al., 2022). Current treatment strategies primarily focus on lifestyle modifications and managing metabolic disorders such as obesity and diabetes. Pharmacological interventions, including pioglitazone, metformin, and simvastatin, have demonstrated efficacy in addressing insulin resistance, regulating lipid metabolism, and reducing liver fat accumulation. However, these treatments have limited success in mitigating inflammation and oxidative stress. Recent studies have highlighted the central roles that these two processes play in the process of NAFLD progression, underscoring the need for new therapeutic interventions.

Fat accumulation in the liver triggers local inflammatory responses, activating macrophages and inducing the release of pro-inflammatory cytokines such as IL-1β, IL-6, and TNF-α. These cytokines not only damage hepatocytes but also promote reactive oxygen species (ROS) production, leading to oxidative stress. Excessive ROS, in turn, activate pathways such as NF-κB, further amplifying inflammation (Das et al., 2024; Taru et al., 2024). In addition, inflammation and oxidative stress can exacerbate lipotoxicity, trigger ER stress (ERS), promote apoptosis, and further drive the progression of NAFLD to non-alcoholic steatohepatitis (NASH) (Arroyave-Ospina et al., 2021). This vicious cycle accelerates the progression of NAFLD from simple steatosis to non-alcoholic steatohepatitis and liver fibrosis, potentially culminating in cirrhosis or hepatocellular carcinoma. Thus, in - depth research on the roles of inflammation and oxidative stress in NAFLD development and the relevant prevention and treatment strategies, as well as exploration of ER stress’ role in NAFLD, is crucial for delaying NAFLD progression.

Polygala tenuifolia Willd. (PTW) is a traditional Chinese medicine known for its lipid-lowering, anti-inflammatory, and antioxidant properties. Our previous research has shown that the active component of PTW extract (EPT), specifically the total saponins derived from Polygala, exhibits significant lipid-lowering effects and mitigates lipid accumulation in rats (Wen et al., 2015). Our previous study showed that EPT can exert anti - inflammatory and antioxidant effects by regulating the TLR4/NF-κB signaling pathway, reducing neutrophil counts and pro - inflammatory factors like IL - 1β, thus alleviating acute lung injury in rats (Guo et al., 2024). We also confirmed that EPT significantly inhibits PGE2, a downstream product of COX2. Given the close relationship between lipid metabolic disorder, inflammation, oxidative stress and ERS in driving NAFLD progression, investigating EPT’s regulatory effects on COX2 and ERS might offer new therapeutic insights. This experiment aims to explore EPT’s dual effects on COX2 and ERS and whether they can delay NAFLD progression.

In this study, we established an NAFLD rat model using high-fat emulsion gavage. We employed network pharmacology and transcriptomics to screen the targets and pathways associated with EPT, to preliminarily explore the molecular mechanisms of early NAFLD development.

This study identified the blood-absorbed components of EPT when taken orally and investigated its mechanisms in delaying the progression of NAFLD. Using UPLC-MS/MS technology, we analyzed the components absorbed into the serum of rats treated with EPT. Through network pharmacology and transcriptomics approaches, we screened the main targets and signaling pathways associated with EPT’s therapeutic effects. Further experiments show that EPT can significantly inhibit COX2 to reduce inflammation and oxidative stress, and regulate the PERK - ELF2α - ATF4 pathway to alleviate ER stress and apoptosis. These findings help us preliminarily explore EPT’s early molecular mechanisms in delaying NAFLD progression, offering initial evidence for its potential application in NAFLD treatment.

Traditional Chinese medicine-derived compounds that enter the bloodstream play a critical role in disease treatment. Using UPLC-MS/MS, we identified 456 blood-absorbed compounds in the serum of rats treated with EPT, including saponins, flavonoids, alkaloids, terpenoids, lignans, coumarins, phenolic acids, quinones, and other substances. Representative compounds such as Fallaxsaponin A, Tenuifoliside B, and Polygalaxanthone IX exhibit notable anti-inflammatory and anti-apoptotic effects (Dong et al., 2014). Tenuifolin and Polygalaxanthone V scavenge free radicals, reduce pro-inflammatory factor production, and exhibit anti-inflammatory and antioxidant properties (Jiang et al., 2023; Liu et al., 2021). Similarly, Tenuifoliside A and Presenegenin present with strong anti-inflammatory potential. These findings suggest that the components of EPT that are ultimately absorbed into the blood possess excellent anti-inflammatory and antioxidant characteristics (Kim et al., 2013; Luyen et al., 2024). Network pharmacology analysis further indicated that EPT influences fatty acid metabolism, reduces lipid accumulation, and regulates pathways such as arachidonic acid metabolism, NF-κB signaling, TNF signaling, IL-17 signaling, AMPK signaling, and ferroptosis. These effects ultimately enhance antioxidant defenses, improve mitochondrial function, and mitigate oxidative stress. The observed enrichment in ER-associated and apoptosis-related pathways suggests that EPT might protect hepatocytes by alleviating oxidative damage. Together, these findings suggest that EPT might exert significant therapeutic effects in the treatment of NAFLD by regulating lipid metabolism, inhibiting inflammatory responses, alleviating oxidative damage, and improving stress conditions, thereby preventing protein misfolding.

In the progression of NAFLD, patients commonly exhibit hepatic fat accumulation, hepatocyte degeneration, and abnormal liver function, alongside chronic inflammation and gradual fibrosis (Han et al., 2023; Wang et al., 2024). Clinical indicators include elevated TC, TG, AST, and ALT levels, with histopathological features such as lipid vacuoles, fat deposition, and collagen accumulation (Pafili and Roden, 2021; Lu and Wang, 2025; Li et al., 2025). Consistent with this, the present analyses revealed the yellowing of the liver, hyperlipidemia, and liver injury in the model group rats. H&E, Oil Red O, and Sirius Red staining revealed significant lipid accumulation, fat deposition, and collagen accumulation in the model group’s liver tissues. However, EPT treatment significantly reversed these pathological changes, indicating that EPT effectively reduces lipid accumulation, alleviates liver injury, and mitigates NAFLD.

Transcriptomic sequencing performed based on these initial analyses provided further insights into the underlying mechanisms through which EPT exerts its beneficial effects. Significant changes were observed in the arachidonic acid metabolism pathway, inflammation-related pathways, and ER-related pathways after EPT treatment, aligning with network pharmacology predictions. Our transcriptomic analysis identified key targets like PTGS2 (COX-2) and Keap1. EPT was found to markedly downregulate PTGS2, suggesting that it might reduce prostaglandin synthesis and, consequently, alleviate inflammation. Keap1 modulation suggests EPT might activate the Keap1 - Nrf2 pathway to curb oxidative stress (Li et al., 2024; Yang et al., 2024). Additionally, some differentially expressed genes were enriched in ER - associated pathways, indicating that the pharmacological effects of EPT are linked to ER stress modulation. In this experiment. Based on targeting COX2 and modulating the Keap1 - Nrf2 and ER stress - related pathways, we preliminarily explored the underlying mechanism of EPT in alleviating early NAFLD progression.

COX2 is the rate - limiting enzyme in arachidonic acid metabolism. It generates prostaglandins and other bioactive lipid mediators that play crucial roles in inflammation and metabolic processes (Ilari et al., 2020). In NAFLD, COX2 expression is significantly upregulated and closely linked to hepatic steatosis and inflammation (Desai et al., 2018). This upregulation induces the release of pro - inflammatory cytokines such as IL - 1β, IL - 6, and TNF - α, thereby exacerbating inflammation and driving the pathological progression of NAFLD (Cheng et al., 2013). Moreover, COX2 overexpression is associated with oxidative stress and lipid metabolic disorders, which are key mechanisms in NAFLD development (Tsujimoto et al., 2016). In rats with high-fat-diet-induced NASH, celecoxib, a selective COX2 inhibitor, significantly reduces hepatic lipid accumulation and improves insulin resistance (Tian et al., 2014). Additionally, celecoxib can also mitigate high-fat-diet-induced steatosis and inflammation in NASH model rats (Chen et al., 2011). NASH and NAFLD share similar pathological features in the early stages of disease progression. Additionally, PGE2, a downstream product of COX2, acts via its receptors in the liver. It is involved in inflammatory responses and lipid metabolism (Wang et al., 2021). These clues indicate that COX2 might be a potential target for early NAFLD prevention and treatment. Our prior research shows that EPT can effectively suppress PGE2 expression, offering anti-inflammatory effects (Wen et al., 2015). Thus, in this study, we chose COX2 as the target to preliminarily explore the molecular mechanism of EPT in preventing and treating NAFLD.

In NAFLD research, Keap1, a key oxidative stress regulator, has been widely studied for its mechanism and potential as a therapeutic target. Keap1 interacts with Nrf2 to regulate cellular antioxidant responses and metabolic homeostasis. Under normal conditions, Keap1 inhibits Nrf2, but under oxidative stress, they detach, allowing Nrf2 to move to the nucleus. Studies have shown that inflammation and oxidative stress are crucial factors in the development of liver diseases, including NAFLD. Nrf2, a key regulator of cellular protection, defends the liver through its anti-inflammatory and antioxidant functions. The activation or inhibition of Nrf2 can significantly impact the progression of such diseases (Xu et al., 2018). Moreover, the Keap1-Nrf2 pathway in NAFLD exerts anti-inflammatory, antioxidant, and lipid - metabolism regulatory effects and also protects hepatocytes from lipotoxic injury by promoting autophagy (Lee et al., 2020; Gao et al., 2021). Thus, targeting the Keap1 - Nrf2 pathway might be a potential therapeutic strategy for NAFLD. HO-1, a downstream protein of the Keap1-Nrf2 pathway, offers hepatic protection through anti-inflammatory, antioxidant, and cytoprotective mechanisms. Studies have shown that HO-1 downregulation is closely related to inflammation, oxidative stress, and hepatocyte injury during NAFLD progression. Enhancing HO-1 expression might be an effective strategy for treating liver diseases (Li et al., 2023). In this study, rats in the model group showed significantly increased Keap1 expression and decreased Nrf2 and HO-1 expression, indicating weakened anti-inflammatory and antioxidant capacity in NAFLD rats. After EPT administration, we found that EPT significantly reduced Keap1 expression and increased Nrf2 and HO-1 expression. This suggests that EPT can restore the anti-inflammatory and antioxidant capacity in NAFLD rats, thereby improving NAFLD.

Inflammation and oxidative stress exacerbate ERS, ERS reciprocally intensifies inflammation and oxidative stress, this interplay complicates the progression of NAFLD (Malhotra and Kaufman, 2011). The endoplasmic reticulum (ER) is responsible for protein processing, folding, and modification. During inflammation, COX2 upregulation increases prostaglandin synthesis, inducing the release of pro - inflammatory cytokines such as IL - 1β, IL - 6, and TNF - α. This disrupts ER function, leads to increased protein translation, and results in the accumulation of misfolded proteins (Shieh et al., 2024; Yu and Kim, 2010). Simultaneously, the COX2 - catalyzed arachidonic acid metabolism generates substantial ROS. This causes oxidative stress, disrupts the ER’s redox balance, intensifies protein folding issues, and exceeds the ER’s processing capacity, thereby inducing ERS. ERS is closely related to hepatic metabolic functions. In metabolic syndrome contexts like obesity and type 2 diabetes, sustained ERS activation causes intrahepatic lipid and cholesterol accumulation. This triggers oxidative stress, further intensifying ERS (Bozaykut et al., 2016). In the pathophysiology of NAFLD, ERS is closely linked to autophagy dysfunction. Autophagy, a cellular degradation process that removes damaged organelles and proteins, is inhibited by ERS activation. This inhibition exacerbates hepatic lipid accumulation and inflammation, promoting the progression of NAFLD to NASH (Lebeaupin et al., 2018). Moreover, in NAFLD, ERS affects not only hepatocytes but also other liver cells like hepatic stellate cells (HSCs) and Kupffer cells (KCs). During NAFLD progression to NASH, activated HSCs produce excessive extracellular matrix, causing liver fibrosis and releasing cytokines to worsen inflammation. Kupffer cells exacerbate hepatic inflammation by releasing pro - inflammatory cytokines and participating in lipid metabolism regulation. Simultaneously, ERS also intensifies inflammation and oxidative stress, creating a positive feedback loop that accelerates NAFLD progression toward NASH (Flessa et al., 2022). These findings suggest that exploring the relationship between ERS and NAFLD might perhaps offer new therapeutic insights and potential targets for NAFLD treatment.

When ERS is persistently activated, the body triggers the unfolded protein response (UPR) to restore ER homeostasis. Initially, this process activates the PERK pathway. Upon dissociation from GRP78, PERK dimerizes and undergoes trans-autophosphorylation, activating its kinase domain. This activation further phosphorylates the α-subunit of eukaryotic translation initiation factor 2 (eIF2α), halting the translation of most mRNAs to alleviate ERS (Liu et al., 2015). However, it simultaneously enhances the translation of activating transcription factor 4 (ATF4), selectively upregulating the expression of C/EBP homologous protein (CHOP). The upregulation of CHOP, in coordination with Bcl-2 expression, regulates apoptotic signaling pathways and activates the apoptotic effector protein Caspase-3, further promoting the execution of apoptosis (Zinszner et al., 1998). In this study, model group rats exhibited significant increases in GRP78, PERK, p-PERK, eIF2α, p-eIF2α, ATF4, CHOP, Bcl-2, and Caspase-3 expression, indicating pronounced ERS and apoptosis. It demonstrates significant ER stress and apoptosis. After EPT treatment, the significant reduction in these indicators suggests that EPT might alleviate ER stress and inhibit apoptosis by modulating the PERK-ELF2α-ATF4 pathway.

Our experiment in this study also has certain limitations. In our previous study, we demonstrated EPT’s effect on PGE2. The primary objective of this experiment was to observe whether EPT could act as a selective COX2 inhibitor to improve NAFLD. Consequently, we used celecoxib, a selective COX2 inhibitor, as a positive control, rather than clinically established NAFLD drugs like pioglitazone. This is a limitation of our study. In this experiment, based on the transcriptomic results, we also made a preliminary exploration of the ER stress pathway. The results indicate that EPT has a good inhibitory effect on ER stress. However, the failure to select an inhibitor with blocking effects on ER stress as a positive control is also a limitation of this experiment. Additionally, NAFLD is a multifactorial disease involving numerous signaling cascades. This raises concerns that a single signaling pathway might not fully elucidate a drug’s mechanism of action in treating NAFLD. In this experiment, we only explored the COX2 and PERK-eIF2α-ATF4 pathways. We failed to show broader metabolic evidence and explore other potential therapeutic pathways for NAFLD. These are also limitations of this experiment. In follow-up experiments, we will strive to improve these shortcomings by enhancing experimental clarity, transparency, and scientific soundness.

In summary, our exploration of NAFLD’s early molecular mechanisms indicates that EPT might target COX2 to exert anti-inflammatory and antioxidant effects. It might also alleviate ER stress and inhibit apoptosis via the PERK-ELF2α-ATF4 pathway, thereby improving NAFLD. In subsequent studies, we will further explore the correlation between the COX2 and PERK-ELF2α-ATF4 pathways.