2.1 Materials

Carbon tetrachloride (CCl₄, purity > 98 %) was purchased from Macklin Inc. (Shanghai, China). AB23A (HPLC ≥ 98 %, batch number: RFS- Y03601906010) was purchased from Chengdu Herbpurify Co., Ltd. (Chengdu, China). Bifendatatum (purity > 97 %, batch number: C1904020) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). E06 mouse monoclonal antibody was purchased from Avanti Polar Lipids, Inc. (Alabama, USA). Goat anti-mouse IgM / HRP antibody was purchased from Beijing Biosynthesis Biotechnology Co., Ltd. (Beijing, China). Hematoxylin staining solution and Eosin staining solution were provided by Wuhan Servicebio Technology Co., Ltd. (Wuhan, China). Assay kits for aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), glutathione (GSH) and malondialdehyde (MDA) were purchased from Nanjing Jiancheng Biological Engineering Research Institute (Nanjing, China). Enzyme-linked immunosorbent assay (ELISA) kits for glutathione peroxidase 4 (GPX4) was obtained from ZCIBIO Technology Co., Ltd (Shanghai, China).

2.2 Extraction of AB23A

Alismatis Rhizoma was provided and identified by Beijing Tongrentang Tianjin Nankai Pharmacy Co., Ltd (Tianjin, China). The samples were kept in the State Key Laboratory of Component-based Chinese Medicine. Extraction of AB23A from the powdered medicinal herb was performed using the reflux extraction method. The Alismatis Rhizoma powder (passed through a No. 5 sieve) was weighed at 1 g and reflux-extracted with 10 mL of 95 % ethanol for 1 h. After reflux extraction, the reduced weight was compensated with additional extraction solvent. Then the extract was diluted 5 times and centrifuged at 20,000 × g for 5 min. The supernatant was collected to obtain the test sample solution. AB23A reference standard was dissolved in methanol to prepare a reference solution with a concentration of 128 μg/mL. The determination method for the content of AB23A in the sample was established using the test sample solution and the reference solution.

Based on the aforementioned method for preparing the test sample solution, a single-factor experiment was employed to optimize the extraction efficiency of AB23A by varying the following parameters: ethanol concentration (50 %, 60 %, 70 %, 80 %, 95 %), solid–liquid ratio (1:6, 1:8, 1:10, 1:12, 1:15), extraction time (1 h, 1.5 h, 2 h, 2.5 h, 3 h), and extraction cycles (1, 2, 3, 4, 5). The extracted solution was then subjected to centrifugation at 20,000 × g for 5 min, and the supernatant was collected for analysis using ultra performance liquid chromatography (UPLC).

2.3 Ultra performance liquid chromatography analyses of samples

Samples were analyzed using the ACQUITY UPLC H-Class PLUS system (Waters, USA) equipped with an ACQUITY UPLC BEH C18 column (2.1 × 50 mm, 1.7 μm, Waters, USA). 3 μL of the sample was injected into the system, and then eluted through the chromatographic column at a flow rate of 0.3 mL/min while the column temperature was set at 35 °C. The chemical components in the sample were detected at a wavelength of 208 nm. The separation was achieved using a gradient elution method with water and acetonitrile as mobile phases A and B, respectively. The gradient elution program was as follows: 0–1 min, 30 %-55 % B; 1–3 min, 55 %-70 % B; 3–7 min, 70 % B; 7–8 min, 70 %-90 % B; 8–9 min, 90 %-30 % B; 9–10 min, 30 % B. The development and validation of the AB23A measurement method in the sample were conducted using the AB23A standard substance. The linearity of the method was evaluated by plotting peak areas against standard substance concentrations, and the accuracy of the method for determining AB23A content in the sample was determined by conducting examinations of precision, repeatability, stability, and recovery rate.

2.4 Response surface methodology design

Based on the preliminary single-factor experiments, a four-factor three-level Box-Behnken design (BBD) was conducted. A binary polynomial regression equation for the extraction efficiency of AB23A concerning factors A, B, C, and D was obtained by conducting 29 sets of randomized sequential experiments. Subsequently, response surface plots were generated to analyze the interactions between various factors and determine the optimal extraction process.

2.5 Separation and purification of AB23A

4 kg of Alismatis Rhizoma was pulverized, and extraction was carried out using the optimized method. Combining various chromatographic methods, AB23A was separated, purified, and enriched from the extract, resulting in 3.3 g of monomeric AB23A with a purity of 95.48 % (details are provided in the supplementary materials).

2.6 Animals

All animal maintenance and treatment procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and were approved by the Committee for Institutional Animal Care and Use at Tianjin University of Traditional Chinese Medicine. Fifty Specific Pathogen Free (SPF) SD rats, weighing 200 ± 20 g, were obtained from Beijing HFK Bioscience Co., Ltd. (Beijing, China). The animals were housed at 23 ± 3 ℃ temperature and 60 ± 10 % humidity with a 12 h light/dark photoperiod. The animals were provided with regular laboratory chow and water.

2.7 Administration

The modeling method and dosage for inducing acute liver injury in rats with CCl₄ were referenced from previous reports with slight modifications (Foroutan et al., 2020; Sheng et al., 2022; Zahira et al., 2023). Animals were allowed 1 week to accommodate and randomly divided into five groups (n = 10): the control group (C, 0.5 % CMC-Na solution), the model group (M, 0.5 % CMC-Na solution), the bifendatatum group (B, 200 mg/kg), the low-dose AB23A (L, 15 mg/kg) and the high-dose AB23A (H, 30 mg/kg). Intragastric administration was performed daily. On the seventh day, acute liver injury was induced in the M, B, L and H group rats by intraperitoneal injection of 2 mL/kg CCl₄ (CCl₄ / olive oil volume ratio of 1:1) 6 h before the last dosage. Then the animals were fasted without water restriction. The following day, animals were sacrificed under anesthesia, and blood and liver samples were collected. The experimental design is summarized in Fig. 2A.

2.8 Histopathological assessment and immunohistologic analysis

The liver tissue was rinsed with normal saline, weighed, and photographed. Subsequently, the caudate lobe of the liver was fixed in 4 % paraformaldehyde and embedded in paraffin. Followed by cutting into 5 μm-thick sections, tissue sections were stained with hematoxylin and eosin (H&E) and photographed using an optical microscope (Nikon, Japan). Immunohistochemical analysis was conducted on liver tissue sections using E06 IgM to detect oxidized phospholipids (OxPLs). Endogenous peroxidase was blocked with hydrogen peroxide, and non-specific binding sites were blocked using serum. The primary and secondary antibodies were incubated successively, and the nucleus was finally counterstained with hematoxylin.

2.9 Biochemical assays

Fresh blood samples were centrifuged at 1,400 × g for 10 min at 4 °C, and the resulting supernatant was collected to obtain serum for the analysis of ALT, AST, and ALP. The liver tissue was homogenized at 10,000 × g for 10 min at 4 °C. The supernatant was taken, and the levels of GSH, MDA and GPX4 were measured following the instructions provided by the kit manufacturer.

2.10 Lipidomic analysis

The extraction and analysis of lipids from the liver were performed as we previously reported with slightly modified (Lei et al., 2023). Briefly, 100 mg of liver tissue homogenate was mixed with 1 mL of dichloromethane/methanol (2:1, V/V). The sample was thoroughly mixed and then centrifuged. The resulting dichloromethane layer was collected and evaporated using nitrogen gas. The dried sample was reconstituted with 100 μL of isopropanol/methanol (1:1, V/V) and centrifuged for detection. Quality control (QC) sample was obtained by mixing 10 μL of each sample.

Liver lipid separation was performed using UltiMate 3000 UHPLC system (Thermo Scientific, USA) equipped with a Hypersil GOLD C18 column (2.1 × 100 mm, 1.9 μm; Thermo Scientific, USA). The column temperature was maintained at 45℃. The mobile phase A consisted of acetonitrile/water (60:40, V/V) with 5 mM ammonium formate and 0.1 % formic acid, while mobile phase B comprised acetonitrile/isopropanol (10:90, V/V) with 5 mM ammonium formate and 0.1 % formic acid. The following gradient elution was performed at a flow rate of 0.35 mL/min: 0–14.50 min, 40 % − 100 % B; 14.50–16.50 min, 100 % B; 16.50–16.51 min, 100 % − 40 % B; 16.51–20.00 min, 40 % B.

Lipid data were collected using the Q Exactive Orbitrap MS system (Thermo Scientific, USA) connected to UHPLC. HESI − II was performed in positive/negative ionization mode with + 3.5 kV for positive mode and −3.0 kV for negative mode. The MS conditions were as follows: the capillary temperature was 320℃, probe heater temperature was 350℃; sheath gas flow was 35 arb, aux gas flow was 10 arb. Data were acquired in the m/z range of 100 – 1500 Da, and product ion scan was acquired using data dependent acquisition (Top N = 5).

2.11 Metabolomic analysis

Metabolites were extracted according to our previous report with appropriate modifications (Lei et al., 2023). 100 mg of liver tissue was homogenized with 500 μL pre-cooled water, followed by the addition of 1 mL of methanol and thorough mixing. The homogenate was subsequently centrifuged at 10,000 × g for 10 min. The supernatant was collected and dried by blowing with nitrogen. The resulting product was redissolved in 100 μL of methanol/water (2:1, V/V), and the supernatant was analyzed by UHPLC-MS/MS after centrifugation. 10 μL of each sample was taken and mixed as the QC sample for the determination of instrument stability.

Same as the above analysis conditions of lipidomics, metabolomic analyses were conducted with UltiMate 3000 UHPLC system (Thermo Scientific, USA) coupled with a Q Exactive Orbitrap mass spectrometer (Thermo Scientific, USA). A Waters ACQUITY UPLC HSS T3 column (2.1 × 100 mm, 1.8 μm; Waters, USA) was equipped on the UHPLC-MS/MS system for metabolites separation, and the column was maintained at 45℃ with a flow rate of 0.4 mL/min and the injected volume of each sample was 5.0 μL. The mobile phase was composed of 0.1 % formic acid in distilled water (mobile phase A) and 0.1 % formic acid in methanol (mobile phase B). The applied elution conditions were as follows: 0–2 min, 0.1 % B; 2–6 min, 0.1 % − 25 % B; 6–10 min, 25 % − 80 % B; 10–12 min, 80 % − 90 % B; 12–21 min, 90 % − 99.9 % B; 21–23 min, 99.9 % B; 23–24 min, 99.9 % − 0.1 % B; 24–26 min, 0.1 % B.

Metabolites in samples were detected via Q Exactive Orbitrap mass spectrometer (Thermo Scientific, USA), and the MS parameters were the same as described for lipidomics.

2.12 Data processing and analysis

Lipidomic and metabolomic data were normalized using MS DIAL 4.9 software (Tsugawa et al., 2020) (RIKEN, Japan), and compounds identification was conducted using the built-in database combined with the Human Metabolome Database (HMDB) (Wishart et al., 2022) and previous report (Aoyagi et al., 2017). The multivariate statistical analysis and pathway analysis were carried out by SIMCA (Sartorius, Germany) and MetaboAnalyst (Pang et al., 2021), respectively. Differential lipids and metabolites were screened based on fold change (FC) > 1.5 and P < 0.05.

2.13 Statistical analysis

Data were analyzed using GraphPad Prism 9.4 (GraphPad Software, USA) and expressed as mean ± standard deviation. Differences between groups were evaluated by one-way ANOVA followed by the Tukey multiple comparison test. Differences were considered significant at P < 0.05.

3.1 AB23A extraction process optimization

The chromatograms of the reference standard of AB23A and the extracted liquid sample were shown in Fig. 1B. In the quantitative method for AB23A, the peak area showed a linear relationship with the concentration in the range of 2.00 to 64.0 μg/mL (y = 8713.1x − 524.72, r² = 0.9999). Additionally, the method demonstrated good precision (RSD = 0.51 %), repeatability (RSD = 0.18 %), and stability (RSD = 1.26 %). The average recovery rate was 101.20 % with an RSD of 3.91 %. Therefore, this method could be used for the quantitative analysis of AB23A in the samples.

Close

Fig. 1

Structure, chromatograms, and extraction process optimization of Alisol B 23-acetate (AB23A). (A) The chemical structure of AB23A. (B) Chromatograms of the reference standard of AB23A and sample of AB23A. (C) Effect of different solid–liquid ratio, volume fraction of ethanol, extraction time and extraction cycles on the extraction rate of AB23A. (D) 3D response surface plots of the interaction of extraction time and volume fraction of ethanol, extraction cycles and volume fraction of ethanol, solid–liquid ratio and volume fraction of ethanol, extraction cycles and extraction time, solid–liquid ratio and extraction time, solid–liquid ratio and extraction cycles on the yield of AB23A.

Export to PPT

The results of the single-factor experiments were shown in Fig. 1C. Based on these results, four factors influencing the extraction efficiency of AB23A were identified and encoded as follows: ethanol volume fraction (60 %, 70 %, 80 %) as A, extraction time (1.5 h, 2 h, 2.5 h) as B, extraction cycles (1, 2, 3) as C, and solid–liquid ratio (1:10, 1:12, 1:15) as D. Each factor was assigned three different levels coded as −1, 0, and 1, corresponding to low, medium, and high levels, respectively. These factors were then used in a BBD for optimizing the best process parameters for AB23A extraction (Table S1). And the binary polynomial regression equation for AB23A extraction efficiency was described by the following equation. $$\mathit{Extraction}\mathit{rate}=2819.01-157.97A-33.30B+356.09C+86.64D-181.00AB+83.37AC-115.26AD+98.50BC+117.25BD-42.06CD-439.64A^2-146.47B^2-212.70C^2-150.53D^2$$ where A, B, C and D are independent variables in coded form for ethanol volume fraction, extraction time, extraction cycles and solid–liquid ratio, respectively.

The results of the variance analysis for the model were presented in Table S2. The regression equation model showed an F-value of 13.14, with a P-value < 0.0001, indicating that the model was statistically significant. Additionally, the lack-of-fit term had a P-value of 0.1779, suggesting that the lack-of-fit for the model was not significant. The coefficient of determination (r²) for the model was 0.9293, and the adjusted R-squared (radj²) was 0.8586, indicating a high correlation between the predicted values and the actual experimental values. Additionally, based on the F-values and the slopes of the three-dimensional surface plots (Fig. 1D), the interaction effects among factors A, B, C, and D had different effects on the yield of AB23A extraction efficiency, the order was AB > BD > AD > BC > AC > CD, and the order of the impact levels of the single-factor was C > A > D > B.

The data was optimized using Design Expert 12.0 software (Stat-Ease Inc., USA), and the maximum extraction rate of AB23A was found to be 3005.51 μg/g when the ethanol volume fraction was set at 67.29 %, the extraction time was 2.27 h, the extraction cycles were 2.86 times, and the solid–liquid ratio was 1:13.70. Based on practical operational constraints, the optimized extraction process was modified to use a solid–liquid ratio of 1:13 with 70 % ethanol for reflux extraction for 3 cycles, with each cycle lasting 2 h. The validation experiment was conducted using the modified extraction conditions, and the extraction rate was found to be 2905.82 μg/g, with an RSD of 0.49 %. The close agreement between the experimental results and the predicted values indicated that the extraction process is reliable, and the regression model was reasonable.

3.2 AB23A alleviated acute liver injury caused by carbon tetrachloride

The appearance of the liver was shown in Fig. 2B. Compared with the control group, the liver of the rat treated with CCl₄ exhibited enlargement and whitening, accompanied by a rough and severely damaged surface of the capsule. However, in the bifendatatum and high-dose AB23A groups, the liver surface became smoother, and the color restored to a dark reddish brown, while few areas still showed slight damage to the liver capsule. According to the H&E staining, in the model group, the central vein of the liver was dilated with the surrounding cells necrosis. The hepatocytes appeared enlarged, and ballooning degenerative cells were densely arranged around the central vein. AB23A demonstrated a dose-dependent reduction in lesion damage, maintaining a normal hepatic lobular structure and significantly reducing ballooning degeneration.

Close

Fig. 2

AB23A could ameliorate acute liver injury induced by CCl₄. (A) Experimental groups and treatments. (B) Representative appearance photographs of livers and H&E staining of liver sections showed AB23A could maintain hepatocyte morphology and reduce carbon tetrachloride toxicity. (C) Liver index evaluation and AST, ALT, ALP levels in serum. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with the model group.

Export to PPT

The liver index in the model group was markedly increased compared with that in the control group, while AB23A administration could significantly reduce the liver index (Fig. 2C). Additionally, liver conditions were evaluated by analyzing the levels of AST, ALT, and ALP in the serum. As shown in Fig. 2C, CCl₄ treatment led to elevated levels of AST, ALT, and ALP in the serum compared to the control group (P < 0.05). Administration of bifendatatum and different doses of AB23A significantly decreased the levels of serum AST, ALT, and ALP (P < 0.05), indicating the beneficial effect of AB23A in alleviating CCl₄-induced liver injury.

3.3 AB23A regulated lipid metabolism profiling of the liver

Total ion chromatograms (TIC) of representative samples from various groups in lipidomics were shown in Fig. 3A. Principal component analysis (PCA) was performed to assess the overall differences in lipid profiling among the C, M, and H groups (Fig. 3B). And orthogonal projections to latent structures discriminant analysis (OPLS-DA) was used to compare the profiling between C and M groups (R²X = 0.557, R²Y = 0.961, Q² = 0.913), as well as the M and H groups (R²X = 0.426, R²Y = 0.986, Q² = 0.665) (Fig. 3C). The results showed that CCl₄-induced liver injury led to alterations in the composition of liver lipids, which were ameliorated by AB23A administration. Volcano plots (Fig. 3D) were used to screen the differential variables affected by CCl₄ and AB23A intervention. A total of 1203 upregulated and 1131 downregulated differential lipids were screened and identified in the M group compared to the C group, indicative of the occurrence of acute liver injury induced by carbon tetrachloride. Similarly, 118 upregulated and 129 downregulated differential lipids were identified in the H group compared to the M group. Venn plots (Fig. 3E) were employed to identify the common differential lipids between the C vs. M and M vs. H groups. Among the 1203 lipids upregulated by acute liver injury, AB23A downregulated 48 of them, including 12 glycerophospholipids (GPs), 15 glycerides (GLs), 8 sphingolipids (SPs), 7 fatty acyls (FAs), 5 sterol lipids (STs), and 1 prenol lipid (PR). Simultaneously, AB23A elevated the levels of 51 out of 1131 abnormally downregulated lipids, including 25 GPs, 4 GLs, 18 SPs, 3 FAs, and 1 ST, indicating the close relationship between these lipids and the efficacy of AB23A (Supplemental files). The results indicated (Fig. 3F) that AB23A played an important role in preventing acute liver injury caused by CCl₄ by affecting glycerophospholipid metabolism, sphingolipid metabolism and glycerolipid metabolism in the liver.

Close

Fig. 3

AB23A mitigated the changes of liver lipid profiling caused by CCl₄. (A) Total ion chromatograms of representative samples from various groups in lipidomics in negative and positive ion modes. (B) Principal component analysis (PCA) score plot showed the lipid profiling of control (C), model (M), and high dose AB23A administration (H) groups were altered. (C) The orthogonal partial least squares-discriminant analysis (OPLS-DA) score plots displayed a clear separation between the two groups (C vs. M, M vs. H). (D) The volcano plot analysis showed significantly regulated variables between the two groups (P < 0.05, fold change > 1.5). (E) The Venn plots demonstrated AB23A could regulate the abnormal lipid caused by CCl₄. (F) Lipid category analysis regulated by AB23A.

Export to PPT

Sphingolipids plays a significant role in the transmission of inflammatory signals. Twenty-two sphingolipid levels were improved under the regulation of AB23A (Fig. 4A). Additionally, immunohistochemical analysis revealed an increased abundance of macrophages and enhanced inflammatory response following exposure to CCl₄. However, AB23A could alleviate the inflammatory response and reduce macrophage level in the liver (Fig. 4B&C).

Close

Fig. 4

AB23A mitigated the inflammatory response induced by CCl₄ in the liver. (A) The heat map analysis revealed the regulation of AB23A on sphingolipids. (B) Representative immunohistochemical staining (macrophage) of liver sections. (C) Positive area analysis of immunohistochemical staining (macrophage). *P < 0.05, **P < 0.01, and ***P < 0.001 compared with the model group.

Export to PPT

3.4 AB23A improved the metabolic function of the liver

TIC of representative samples from various groups in metabolomics were shown in Fig. 5A. PCA analysis (Fig. 5B) displayed a clear separation of samples among C, M and H groups. Besides, the OPLS-DA score (Fig. 5C) indicated that the C vs. M (R²X = 0.682, R²Y = 0.995, Q² = 0.946) and M vs. H (R²X = 0.560, R²Y = 0.988, Q² = 0.889) groups were scattered in distinct areas. It showed that CCl₄-induced liver metabolic dysfunction could be modulated by AB23A. 496 up regulated and 271 down regulated metabolites affected by CCl₄ were screened and identified using volcano plots (Fig. 5D) and Venn plots (Fig. 5E), while AB23A could regulate 94 (including 30 lipids and lipid-like molecules, 26 organic acids and derivatives, 19 sugars, 10 peptides, 8 nucleosides, nucleotides and analogues, and 1 vitamin) and 42 (including 15 organic acids and derivatives, 5 lipids and lipid-like molecules, 10 peptides, 4 sugars, 7 nucleosides, nucleotides and analogues, and 1 organoheterocyclic compound) of them to normal levels, respectively (Supplemental files). Pathway enrichment analysis (Fig. 5F) further demonstrated that AB23A primarily exerted its effect against CCl₄-induced acute liver injury through the regulation of purine metabolism in the liver.

Close

Fig. 5

AB23A could ameliorate liver metabolic disorder caused by CCl₄. (A) Total ion chromatograms of representative samples from various groups in metabolomics in negative and positive ion modes. (B) The PCA score plot showed the metabolism profiling of C, M and H groups were altered. (C) The OPLS-DA score plots exhibited distinct separation between the two groups (C vs. M, M vs. H). (D) The volcano plot analysis displayed significantly regulated variables between the two groups (P < 0.05, fold change > 1.5). (E) The Venn plots demonstrated AB23A could regulate the abnormal metabolites caused by CCl₄. (F) Pathway analysis elucidated the metabolic pathways modulated by AB23A in the liver.

Export to PPT

Purine metabolism is an important pathway involved in cell proliferation and oxidative balance. We observed upregulation of dGMP, GMP, xanthosine, and IDP levels in the model group, while adenosine and inosine levels were downregulated (Fig. 6A). AB23A treatment led to a restoration of their levels. Besides, in the lipidomic analysis, an increase in the content of oxidized phospholipids, a product of lipid peroxidation, was observed in the model group (Fig. 6B). As shown in Fig. 6C, there were a large amount of oxidized phospholipid around the central vein in the model group, whereas the AB23A administration groups exhibited a dose-dependent reduction in their production. Furthermore, the contents of GSH and GPX4 in the liver were significantly reduced (P < 0.05), while the level of MDA was significantly increased (P < 0.01) (Fig. 6D). AB23A administration significantly reversed the levels of GSH and GPX4 and reduced the content of MDA in the liver (P < 0.01), indicating that AB23A had the ability to alleviate liver oxidative stress and protect hepatocytes by reducing lipid peroxidation, thereby reducing the liver toxicity induced by CCl₄.

Close

Fig. 6

AB23A could reduce lipid peroxidation caused by CCl₄. (A) Heat map analysis of key metabolites involved in purine metabolism. (B) The heatmap demonstrated the oxidized phospholipid were up-regulated in M group and down-regulated in H group. (C) Representative immunohistochemical staining (oxidized phospholipid) of liver sections. (D) Positive area analysis of immunohistochemical staining (oxidized phospholipid) and GSH, GPX4, MDA levels in liver. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with the model group.

Export to PPT

To clarify the changes of differential lipids and metabolites mediated by AB23A in pathways, the normalized peak areas of key metabolites and lipids were quantified (Supplementary materials) and integrated. Using a filtering criterion of a Pearson correlation coefficient greater than 0.8, we constructed a correlation network among key metabolites influenced by AB23A (Fig. 7). The results revealed that the metabolites regulated by AB23A formed a correlated network centered around adenosine and phospholipids, indicating a closely interconnected pattern of changes among these metabolites. Furthermore, the metabolic pathways are illustrated in Fig. 8. Compared with the C group, the M group animals exhibited dysregulated levels of lipids and metabolites in the liver as a result of CCl₄ exposure. In lipid metabolism, the levels of LPE, OxPS, OxPC, OxPE, OxPI, PI, TAG, DAG, and SM were significantly increased (P < 0.05) in the model group. However, high-dose AB23A administration effectively downregulated these lipids (P < 0.05). Conversely, the levels of C1P, LPC, PA, PS, and PE were significantly downregulated (P < 0.05) by CCl₄, which could be significantly upregulated (P < 0.05) in the administration group. Additionally, AB23A demonstrated significant modulation of abnormally upregulated dGMP, GMP, IDP, xanthosine, as well as abnormally downregulated adenosine, inosine, betaine.

Close

Fig. 7

Correlation network diagram among key metabolites. The size of the circles represents the number of nodes, the thickness of the lines indicates the strength of the correlation. Orange solid lines represent positive correlations between metabolites, while blue dashed lines represent negative correlations.

Export to PPT

Close

Fig. 8

Schematic representation of the metabolic regulation network illustrating the amelioration of CCl₄-induced acute liver injury by AB23A. The left squares represent the changes in the levels of metabolites affected by CCl₄, while the right squares represent the regulation of these metabolites by AB23A. Red indicates upregulation, and blue indicates downregulation. Asterisks highlight significant changes in metabolite levels (P < 0.05). Different background colors represent various metabolic pathways regulated by AB23A.

Export to PPT