Introduction

Non-alcoholic fatty liver disease (NAFLD) is defined as lipid accumulation (more than 5%) in hepatocytes without alcohol consumption, viral infection, or other known pathogenic factors. NAFLD includes simple lipid steatosis, non-alcoholic steatohepatitis (NASH), and advanced fibrosis (Vernon et al. 2011) and is the leading cause of liver disease with a prevalence of 25.24% worldwide (Younossi et al. 2016) and 29.2% in China (Zhou et al. 2019). However, the pharmaceutical therapies available for NAFLD clinical treatment are limited due to complex and diverse etiologies and extended disease development. Therefore, treating and preventing NAFLD and chronic liver disease depend on the availability of safe, effective, and diverse therapeutic agents, the development of which is crucial.

Chitosan oligosaccharides (COS) are chitosan degradation products that can regulate body weight and lipid metabolism by modifying the dysfunctional gut microenvironment (He et al. 2020; Wang et al. 2020). Previous studies have shown that COS reduces the intracellular triglyceride (TG) levels in oleic acid-induced HepG2 cells (Cao et al. 2016), displays anti-obesity activity, and improves serum and liver lipid profile abnormalities in high-fat diet-induced C57BL/6N mice (Choi et al. 2012; Li et al. 2022a, b), highlighting the potential of COS in preventing NAFLD. Studies have increasingly focused on the molecular COS mechanism involved in the hypolipidemic effect, including lipid uptake regulation, de novo synthesis, and free fatty acid (FFAs) β-oxidation (Liu et al. 2021; Tao et al. Western blot

The liver tissues and cells were lysed and homogenized at 4 °C using RIPA lysis buffer, phosphatase inhibitors, and protease inhibitors. The homogenate was centrifuged at 15,000 g for 15 min at 4 °C. The protein concentration was determined using a BCA protein assay kit. The protein samples were separated via 10% SDS-PAGE and then transferred to a polyvinylidene difluoride (PVDF) membrane (Merck Millipore). The membrane was blotted in Tris-buffered saline with tween 20 (TBST) containing 5% skim milk for 1 h at room temperature. Then, the polyclonal antibodies were diluted to 1:1000 with primary antibody dilution buffer, and the membranes were incubated overnight at 4 °C, followed by incubation with horseradish peroxidase-conjugated secondary antibodies (1:3000) for 1 h. The target bands were scanned using a Tannon automatic chemiluminescence image analysis system and examined via Image J software.

Molecular docking simulation

The PPARα protein (PDB code: 2ZNN) was downloaded on Protein Data Bank (RCSB PDB: Homepage). The original ligand was removed by PyMOL (Version 2.1.1_0) to prepare PPARα crystal structure. The molecule structure of COS2 and butyric acid were mapped using ChemBio3D Ultra 14.0. The ligand docking center and the size of the grid box were determined on the native ligand and obtained by AutoDockTools 1.5.6. The docking studies were performed by AutoDock Vina 1.1.2 (Trott et al. 2010). The results were analyzed by PyMOL.

Statistical analysis

The experimental data were generated with GraphPad Prism 8.0 (GraphPad Software, San Diego, USA) and expressed as mean ± SD. The Oil Red O staining images were analyzed using Image J statistical software via one-way ANOVA with multiple comparisons to show the differences between groups. P  < 0.05 were considered significant.

Results

COS 2 decreased lipid deposition in the serum and liver of ob/ob −/− mice

Ten weeks of COS2 intervention distinctly reduced the body weight gain of the ob/ob−/− mice (Fig. 1A, p < 0.05) and decreased the liver index compared with the model group (p < 0.05) (Fig. 1B). The TG, TC, and LDL-C levels in the serum were improved after COS2 intervention (Fig. 1C, D, E), while the HDL-c levels increased (Fig. 1F, p < 0.05). Moreover, the lipid droplets in the liver were smaller in the COS2 treatment groups than in the model group (Fig. 2A). The lipid profiles in the liver, including TG, TC, and LDL-C, decreased while the HDL-C level increased after COS2 administration (Fig. 2B, C, D, E, p < 0.05). These results indicated that COS2 ameliorated lipid deposition in the serum and liver of ob/ob−/− mice.

Fig. 1
figure 1

The improvement of lipid accumulation induced by COS2 in ob/ob−/− mice. A Body weight. B Food intake. C Liver index. D Serum TG. E Serum TC. F Serum LDL-C. G Serum HDL-C

Full size image
Fig. 2
figure 2

The effect of COS2 on lipid accumulation amelioration in the liver of the ob/ob−/− mice. A Oil Red O staining (400\(\times\)magnification, scale bar = 200 µm). The arrows indicate the lipid droplet size in the liver. B Hepatic TG. C Hepatic TC. D Hepatic LDL-C. E Hepatic HDL-C

Full size image

COS 2 regulated lipid metabolism to promote the lipid β-oxidation levels of ob/ob −/− mice

Farnesoid X receptor (FXR) is crucial for lipid homogenesis and protects the liver from lipid accumulation and hepatic steatosis (Schmitt et al. 2015). This study showed a distinct increase in the hepatic FXR gene expression level in the ob/ob−/− mice at high COS2 gavage concentrations (Fig. 3A, p < 0.05), accounting for the inhibition of lipid uptake and synthesis shown by the previous results (Shen et al. 2021), and contributing to FFA β-oxidation upregulation (** et al. 2020). Although the β-oxidation pathways helped eliminate excessive FFAs, the role of COS2 in FFAs β-oxidation regulation remains unclear. Acyl-CoA synthetase long-chain family member 1 (ACSL1) is an enzyme located in the outer mitochondrial membranes, responsible for transferring fatty acids to acyl-CoA (Coleman et al. 2000; Huh et al. 2020). High-dose COS2 treatment significantly increased the ACSL1 mRNA levels in the livers of the ob/ob−/− mice (Fig. 3B, p < 0.05). PPARα, induced by FXR, is a transcription factor that regulates FFAs β-oxidation (Fuchs et al. 2016; Sinal et al. 2001), the gene expression of which was increased by COS2 treatment (Fig. 3C, p < 0.05). As the downstream target of PPARα, CPT1A and CPT2 represent essential regulators for FFA translocation into the mitochondria, denoting the rate-limiting steps of mitochondrial β-oxidation (** et al. 2020). COS2 significantly increased the mRNA levels and CPT1A and CPT2 protein content in a dose-dependent manner (Fig. 3D, E, G, p < 0.05). Furthermore, ACOX1 represents the first enzyme in the FFA β-oxidation pathway. High-dose COS2 treatment upregulated the ACOX1 gene level, facilitating FFA oxidation in the liver (Fig. 3F). Therefore, COS2 accelerated FFA conversion and translocation in the liver and increased the mitochondrial β-oxidation rate.

Fig. 3
figure 3

The mRNA levels of A FXR, B ACSL1, C PPARα, D CPT1A, E CPT2, and F ACOX1, and the protein levels of G CPT1A and CPT2, induced by COS2 in targets related to lipid oxidation in the livers of the ob/ob−/− mice

Full size image

COS 2 regulated the synthesis of SCFAs, leading to a significant increase in the production of butyrate in ob/ob −/− mice

Butyric acid is an SCFAs fermented from dietary fibers by gut microbiota in the colon. Differential analysis was performed on the content of SCFAs in the feces of mice from different treatment groups (Fig. 4). Acetic acid and propionic acid serve as substrates for hepatic gluconeogenesis, and enter the hepatic portal system through intestinal absorption, promoting hepatic gluconeogenesis metabolism and providing energy for liver metabolism. Compared with the model group (Fig. 4A), the acetic acid content in the COS2 high and low dose groups increased significantly (p < 0.05). The propionic acid content in the COS2 high and low dose groups increased significantly (Fig. 4B, p < 0.05). Butyrate is the energy source for intestinal epithelial cells and can have a preventive and therapeutic effect on NAFLD and T2DM. Compared with the model group, the butyric acid content in the COS2 high-dose group increased significantly (Fig. 4C), which is consistent with the results of previous reports on the changes in gut microbiota induced by COS2 (Ji et al. 2021, 2022). Analysis of the total SCFAs content in the intestine of COS-intervened model mice revealed a significant increase in the COS2 treated group.

Fig. 4
figure 4

Changes in SCFAs in the colon of ob/ob−/− mice treated with different doses of COS2. A Acetic acid. B Propionic acid. C Butyric acid. D Total SCFAs

Full size image

Recent studies have found that butyric acid can exhibit functions and impact other tissues and organs beyond the gut via the enterohepatic circulation (van der Hee et al. 2021). Our previous research showed that COS2 enhanced butyric acid accumulation by promoting the abundance of Clostridium_sensu_stricto_1, Clostridium_sensu_stricto_13, and Fusobacterium (Ji et al. 2021, 2022). Moreover, butyric acid decreased lipogenesis to alleviate PPARγ in the liver and reversed lipid accumulation via the liver–gut axis (den Besten et al. 2015). Therefore, it can be considered a potential hypolipidemic biomarker during COS2 metabolism in the colon. The role of COS2 during FFAs β-oxidation was correlated with butyric acid to determine its potential lipid-lowering mechanism in conjunction with COS2.

The hypolipidemic effect of COS 2 and the typical metabolite modulated by COS 2 in sodium oleate-induced HepG2 cells

HepG2 cells were incubated with 0.01 mM sodium oleate to obtain the NAFLD cell model. Lipid accumulation was evident in the model group, suggesting the successful establishment of the NAFLD model (Figs. 5, 6, p < 0.05). To examine the lipid-lowering effect of COS2 and butyric acid, the NAFLD cell model was subjected to different COS2 and NaB doses and incubated for 24 h. And the cell availability was detected (the data are shown in Additional file 1: Fig. S2). The Oil Red O staining results indicated that COS2 and NaB intervention significantly alleviated lipid accumulation (Figs. 5D and 6D), while only high COS2 doses (Fig. 5A, B, C, 0.2 mM, 0.5 mM, and 1 mM) reversed the TG, TC, and LDL-C levels in the NAFLD cell model. However, a less distinct lipid-lowering effect was evident in the model group at the lowest COS2 concentration (Fig. 5A, B, C, 0.02 mM). However, after NaB treatment, the TG, TC, and LDL-C levels decreased significantly, induced by sodium oleate (Fig. 6, p < 0.05).

Fig. 5
figure 5

The effect of different COS2 concentrations on the lipid levels in oleic acid-induced HepG2 cells. A TG levels. B TC levels. C LDL-C levels. D Lipid accumulation after Oil Red O staining (scale bar = 500 nm)

Full size image
Fig. 6
figure 6

The effect of different NaB concentrations on the lipid levels in sodium oleate-induced HepG2 cells. A TG levels. B TC levels. C LDL-C levels. D Lipid accumulation after Oil Red O staining (scale bar = 500 nm)

Full size image

COS 2 and NaB restored lipolysis in sodium oleate-induced HepG2 cells

The mRNA levels of the lipid β-oxidation pathway involving PPARα and its downstream targets, CPT1A and ACOX1, were inhibited in the NAFLD model compared to the control group (Fig. 7A, B, C, p < 0.05). All NaB doses and high COS2 doses elevated the mRNA levels of PPARα, indicating potential lipolysis stimulation, while low COS2 doses showed less effects to activate PPARα gene levels compared to the model group (Fig. 7A, p < 0.05). Therefore, low COS2 doses were less successful than NaB in reversing the PPARα protein levels (Fig. 7D). Similarly, all NaB doses and high COS2 doses were more effective in reversing CPT1A gene and protein levels, while low COS2 concentrations were less successful in increasing CPT1A (Fig. 7B, E). Furthermore, all COS2 and NaB doses restored CPT2 protein expression (Fig. 7F). Moreover, ACOX1, a rate-limiting enzyme involved in FFAs β-oxidation, had been more facilitated by NaB than COS2 in mRNA levels (Fig. 7C). Consequently, NaB was superior to COS2 at low concentrations in promoting oxidative lipid metabolism. Therefore, the butyric acid in the liver–gut axis may exert a more positive effect, warranting further attention.

Fig. 7
figure 7

The mRNA and protein expression levels of the lipid β-oxidative metabolism pathway in sodium oleate-induced HepG2 cells, regulated by COS2 and NaB

Full size image

Structures of COS 2 and butyric acid bound to PPARα

PPARα involves in fatty acid metabolism and can be activated by ligands binding to its LBDs (Oyama et al. 2009). The LBDs of PPARα including hydrophilic, hydrophobic and amphiphilic pockets forming Y-shape domains (Han et al. 2020). The optimal docking modes of COS2 suggested that the ligand mainly contact with H3, H2’ helix and β3 strand at the hydrophobic pocket of PPARα, which is the entrance of the LBDs (Fig. 8A). While due to the size and hydrophily of butyric acid, the hydrophilic pocket between Helix 3 and Helix 12 including the activation factor-2 (AF-2) domain was occupied and hydrogen bonds were formed with amino acid residues Tyr464 on Helix 12, Try314 on Helix 5, Ser280 on Helix 3 and His440 on helix 11 (Fig. 8B, C). These interactions stabilized AF2 helix to recruit co-activators of receptor and facilitated the transcriptional activity of PPARα (Capelli et al. 2016; Han et al. 2020; Xu et al. 2001). Thus, these results well explained that butyric acid could be more effective than COS2 at low concentrations in promoting PPARα expression and stimulating oxidative lipid metabolism.

Fig. 8
figure 8

Docking COS2 and NaB in LBDs of PPARα. A COS2 and B butyric acid binding to pocket within 4 Å of residue of PPARα. C Hydrogen bond interaction of butyric acid and PPARα

Full size image

Discussion

This study proposed that the lipid-lowering mechanism of COS2 was related to a specific metabolite, butyric acid. COS2 effectively alleviated lipid dysfunction by regulating the mitochondrial β-oxidation pathways in ob/ob−/− mice. Additionally, COS2 exhibited a prebiotic effect by facilitating the accumulation of butyric acid, a specific metabolite with the potential to relieve hepatic lipid abnormality more effectively. Therefore, this study was extended to examine the lipid-lowering mechanism of COS2.

NAFLD diagnosis typically occurs when lipid accumulation accounts for 5% of the weight of the liver. Lipid metabolism dysbiosis represents the main pathogenic factor of NAFLD and can manifest in various ways, such as lipid uptake, de novo synthesis, and oxidative metabolism. It is been proved that COS2 can be absorbed and circulated through blood and exhibited its effects in liver (Chen et al. 2022). Our previous studies showed that COS2 downregulated hepatic lipogenesis-related targets and reduced lipid uptake (Shen et al. 2021). In this study, we found that COS2 enhanced the FFA oxidation process in the liver to reverse NAFLD lipid accumulation in the ob/ob−/− mice. Mitochondrial FFA β-oxidation represents the main lipid oxidation pathway. The upstream regulator of lipid metabolism is FXR, a ligand-activated receptor belonging to the nuclear receptor superfamily and essential for regulating FFA β-oxidation by controlling PPARα (Proctor et al. 2006; Kast et al. 2001; Li et al. 2021; ** et al. 2020). COS2 improved FXR gene expression, activating liver lipolysis (Fig. 3A). PPARα, the downstream target of FXR, represents the key enzyme that controls the FFA β-oxidative system. As shown in Fig. 3C, high COS2 doses significantly upregulated the PPARα mRNA levels. Moreover, the mitochondrial entry process represented the FFA β-oxidation rate-limiting step and mainly occurred via ACSL1, CPT1A, and CPT2 catalysis. The FFAs were activated to form acyl-CoA via ACSL1 (Huh et al. 2020), which was transferred successfully into the mitochondria with the help of CPT1A and CPT2, finally initiating the β-oxidation process (** et al. 2020). COS2 promoted the expression of gene and protein levels of ACSL1, CPT1A, and CPT2 to facilitate the rate-limited process of lipolysis, demonstrating the internal mechanism of COS2 in improving hepatic lipid metabolism of NAFLD in ob/ob−/− mice.

In addition, COS2 has been found to significantly increase the production of butyric acid by enriching the microbial communities in the intestinal microenvironment. In an in vitro anaerobic fermentation study using fecal samples from healthy individuals, COS2 was shown to promote the growth of functional microbial communities, such as Clostridium_butyricum, Clostridium, and Parabacteroides, which are known to produce substantial amounts of butyric acid (Ji et al. 2022). Similarly, in the fermentation of fecal samples obtained from patients with NAFLD, COS2 enriched the core functional microbial communities responsible for butyric acid production, including Clostridium sensu stricto 13, Parabacteroides, Romboutsia, Holdemanella, Bacteroides, Bacterium NLAE zl_G201, Erysipelatoclostridium, and Lactococcus (Ji et al. 2021). Moreover, COS2 and COS3 were found to promote the growth of intestinal probiotics such as Bifidobacterium and Lactobacillus in ob/ob model mice and significantly increase the abundance of functional microbial communities, such as Akkermansia, Clostridiales, Faecalibaculum, Roseburia, Ruminiclostridium, and Alistipes, which are also known to produce butyric acid. Previous study reported that butyrate had been reported to induce the reduction of lipid accumulation (Zong et al. 2023). For the intervention mechanism investigation, NaB-induced PPARα activation stimulates fatty acid β oxidation, thus contributing to amelioration of high-fat diet-induced NAFLD in adult rats (Sun et al. 2018). Thus, enterohepatic circulation is possibly responsible for the COS2 anti-NAFLD effect.

Due to the complexity of NAFLD pathogenesis, the following hypotheses could explain the anti-NAFLD impact of COS2: 1. COS2 alleviated hepatic lipid accumulation via direct liver lipid homeostasis. 2. COS2 increased the butyrate levels, producing a distinct hepatic anti-hyperlipidemic effect via enterohepatic circulation. To confirm this, HepG2 cell experiments were established using sodium oleate to induce lipid accumulation and create a NAFLD cell model. The metabolic kinetics study results of COS2 revealed a physiological concentration of 0.02 mM in the serum of the rats after intragastric administration of 500 mg kg−1. The physiological butyric acid concentration was around 3 μM in human serum (Behary et al. 2021) and between 26 μM to 48 μM in the portal serum of mice (Jakobsdottir et al. 2013). Therefore, 0.02 mM COS2 and 0.01 mM NaB were selected as the low dose interventions in the NAFLD cell model, showing that both treatments displayed anti-hyperlipidemia activity. However, at the physiological concentration, NaB was more successful in restoring the lipid levels, as well as the gene and protein expression of PPARα, CPT1A, ACOX1, and CPT2 than COS2. Moreover, in molecular docking analysis, the results showed that COS2 interacted with PPARα in hydrophobic pocket and stuck in the entrance of LBDs of PPARα protein. While butyric acid could crush into the hydrophilic pocket and form hydrogen bonds with AF2 domain, which was important for the receptor to activate the transcriptional activity (Xu et al. 2001). Therefore, butyric acid was superior in stimulating FFA β-oxidation and preventing lipid accumulation. The intestinal microenvironment and enterohepatic circulation status are crucial for NAFLD remission. COS2 accelerated the FFA β-oxidation pathway by promoting butyric acid production in the intestinal tract, reaching the liver via enterohepatic circulation, and binding to the FFA β-oxidation targets PPARα to improve lipid metabolism abnormalities.

Conclusion

COS2 was found to induce the FFA β-oxidation pathway, which mitigated NAFLD by regulating gut microenvironment, particularly butyric acid metabolism, and its interaction with hepatic lipid metabolism. The results provide new insights into the mechanism of COS2 in lipid-lowering effects.