https://orcid.org/0009-0009-9378-4775
Figures
Abstract
Background
Huanglian Wendan Decoction (HLWDD), a classical traditional Chinese medicine (TCM) formula, has shown therapeutic promise in treating metabolic disorders. However, its underlying mechanisms against non-alcoholic fatty liver disease (NAFLD) remain unclear.
Objective
This study aimed to elucidate the pharmacological mechanisms by which HLWDD ameliorates NAFLD, focusing on its impact on lipid metabolism, gut microbiota, and amino acid regulation.
Methods
A NAFLD rat model was established by administering a high-sugar, high-fat, high-salt diet for 20 weeks. The core components of HLWDD were identified and quantified using UPLC-Q-TOF-MS/MS and HPLC, and further validated via network pharmacology and molecular docking. Therapeutic efficacy was assessed through analysis of body weight, serum lipid profiles, inflammatory cytokines, hepatic histology, and protein expression. Gut microbiota composition and liver-intestine metabolite profiles were evaluated using metagenomic sequencing and LC-MS/MS.
Results
Seven key constituents, including quercetin and berberine, were quantified (15.11–164.37 μg/mL) and shown to interact with lipid metabolism targets such as liver kinase B1 (LKB1), AMP-activated protein kinase (AMPK), peroxisome proliferator-activated receptor alpha (PPARα), and carnitine palmitoyltransferase 1A (CPT1A). HLWDD treatment significantly reduced body weight, hepatic lipid accumulation, and serum levels of triglycerides, total cholesterol, and low-density lipoprotein cholesterol, while increasing high-density lipoprotein cholesterol. Proinflammatory cytokines (IL-6, IL-1β, TNF-α) were notably suppressed. Mechanistically, HLWDD activated the LKB1/AMPK signaling pathway and modulated aspartic acid metabolism in association with increased abundance of Akkermansia in the gut. Metabolomic analysis identified 13 differential metabolites, with aspartic acid showing strong correlations with Akkermansia and LKB1/AMPK activity.
Conclusion
HLWDD exerts its anti-NAFLD effects by enhancing Akkermansia-mediated aspartate metabolism, thereby activating the LKB1/AMPK axis and promoting lipid oxidation via CPT1A and PPARα. This study provides new mechanistic insight into the gut–liver axis in NAFLD and highlights HLWDD as a multi-targeted therapeutic approach for restoring metabolic balance.
Citation: Zhu J, Chen Y, Han Y, Li J (2025) Mechanism of Huanglian Wendan Decoction in ameliorating non-alcoholic fatty liver disease via modulating gut microbiota-mediated metabolic reprogramming and activating the LKB1/AMPK pathway. PLoS One 20(9): e0331303. https://doi.org/10.1371/journal.pone.0331303
Editor: Fahrul Nurkolis, Airlangga University Faculty of Medicine: Universitas Airlangga Fakultas Kedokteran, INDONESIA
Received: May 2, 2025; Accepted: August 11, 2025; Published: September 2, 2025
Copyright: © 2025 Zhu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by Hunan Provincial Department of Education Research Project (23B0375); Hunan Provincial Administration of Traditional Chinese Medicine Research Project (C2023021); Hunan University of Traditional Chinese Medicine University-level Research Project (2024XJZC017).
Competing interests: The authors declare no conflict of interest.
Abbreviations: AMPK, AMP-activated Protein Kinase; CPT1A, Carnitine Palmitoyltransferase 1A; HDL, High Density Lipoprotein; H&E, Hematoxylin and Eosin; HPLC, High Performance Liquid Chromatography; HLWDD, Huanglian Wendan Decoction; IL, Interleukin; LDL, Low Density Lipoprotein; LKB1, Liver Kinase B1; NAFLD, Non-Alcoholic Fatty Liver Disease; NF-κB, Nuclear Factor kappa-B; PPARα, Peroxisome Proliferator-Activated Receptor Alpha; PPARγ, Peroxisome Proliferator-Activated Receptor Gamma; SCFAs, Short-Chain Fatty Acids; TC, Total Cholesterol; TCM, Traditional Chinese Medicine; TG, Triglyceride; TNF-α, Tumor Necrosis Factor-α; UPLC-Q-TOF-MS/MS, Ultra-Performance Liquid Chromatography Quadrupole-Time-of-Flight Tandem Mass Spectrometry
Introduction
NAFLD, the most prevalent chronic liver disorder globally, affects over 25% of the population and is strongly linked to obesity, type 2 diabetes, and metabolic syndrome [1–3]. Its hallmark pathological feature—excessive lipid deposition in hepatocytes—arises from dysregulated energy metabolism, gut microbiota dysbiosis, and gut-liver axis impairment [4–6]. The LKB1/AMPK signaling axis plays a central role in hepatic lipid homeostasis. Phosphorylation of AMPK (Thr172) by LKB1 inhibits fatty acid synthesis while promoting CPT1A/PPARα-driven oxidation [7–9]. Conversely, LKB1 deficiency or AMPK inactivation exacerbates gluconeogenesis and oxidative stress, accelerating NAFLD progression [10]. Although AMPK activators show therapeutic potential, concerns regarding tissue specificity and long-term safety persist [10].
The gut microbiota critically interacts with host metabolism via the gut-liver axis. NAFLD patients exhibit characteristic dysbiosis, often marked by reduced beneficial bacteria [11–13]. Dysregulated gut-derived metabolites (e.g., LPS, TMAO) exacerbate hepatic lipid accumulation and insulin resistance [11–13]. While probiotics partially alleviate NAFLD, inter-individual variability and microbiota instability hinder clinical translation [14,15].
TCM offers promising multi-target therapeutic strategies for complex diseases like NAFLD. HLWDD, a classical heat-clearing formula, has demonstrated efficacy in improving dyslipidemia, insulin resistance, and hepatic inflammation [16,17]. Evidence suggests HLWDD reduces hepatic steatosis by suppressing inflammatory pathways (e.g., PPARγ/NF-κB) and modulates gut microbiota, increasing beneficial metabolites like short-chain fatty acids (SCFAs) [16,18]. However, despite evidence for the metabolic and microbiota-modulating effects of HLWDD constituents, the precise mechanisms through which HLWDD influences the gut-liver axis and AMPK signaling in NAFLD remain unclear.
Therefore, this study aimed to address the specific question: Does HLWDD ameliorate NAFLD through modulation of the gut-liver axis (specifically Akkermansia-dependent aspartate metabolism) and activation of the LKB1/AMPK pathway? We hypothesize that HLWDD alleviates NAFLD by enriching beneficial gut flora (e.g., Akkermansia), regulating gut-liver axis metabolites to activate LKB1/AMPK signaling, and restoring lipid oxidation (via CPT1A/PPARα) and amino acid homeostasis. This study will elucidate the synergistic effects of HLWDD on microbial metabolites and hepatic energy metabolism, providing novel mechanistic insights and a potential therapeutic avenue for NAFLD.
Materials and methods
Reagents and chemicals
TE buffer (Cat. No. R541019-0100), ELISA kits for IL-6 and IL-1β (Batch No. GR20230610), TNF-α (M130055-48T), triglyceride (TG), HDL-C, and LDL-C assay kits were purchased from commercial suppliers. RNA extraction kit (Cat. No. 5003050), SYBR qPCR SuperMix (E099-01A), and cDNA synthesis mix (E047-01B) were used for gene expression analysis. Primers for AMPK, LKB1, CPT1A, PPARα, and internal control (ACTB) were obtained from Sangon Biotech (Shanghai) Co., Ltd. Primary antibodies (p-LKB1, p-AMPK, LKB1, AMPK, PPARα, CPT1A, GAPDH) were sourced from Abcam. Chromatographic-grade solvents (methanol, acetonitrile, formic acid, water, isopropanol) were obtained from Fisher Scientific. Metformin (catalog: 20200A) was obtained from Patheon France. Reference standards of HLWDD compounds (berberine, palmatine, quercetin, etc.) were purchased from Sichuan Cuiyirun Biotechnology.
Preparation of HLWDD
Herbal ingredients with specified botanical sources and plant parts were weighed: Coptis chinensis Franch. (rhizome; Sichuan, China; Batch: CC202305) 10g, Pinellia ternata (Thunb.) Makino (tuber; Shanxi, China; Batch: PT202304) 10g, Phyllostachys edulis (Carrière) J.Houz. (caulis; Zhejiang, China; Batch: PE202306) 10g, Citrus aurantium L. (immature fruit; Jiangxi, China; Batch: CA202302) 15g, Citrus reticulata Blanco (pericarp; Guangdong, China; Batch: CR202303) 10g, Glycyrrhiza uralensis Fisch. Ex DC. (root; Inner Mongolia, China; Batch: GU202307) 5g, Zingiber officinale Roscoe (fresh rhizome; Shandong, China; Batch: ZO202305) 5g, Wolfiporia cocos (F.A.Wolf) Ryvarden & Gilb. (sclerotium; Yunnan, China; Batch: WC202308) 15g. All specimens were morphologically authenticated by Prof. Tasi Liu (School of Pharmacy, Hunan University of Chinese Medicine) and taxonomically verified using Medicinal Plant Names Service (MPNS; http://mpns.kew.org). Voucher specimens were deposited in the University Herbarium (Accession No. HLWD2023−001–008).
Decoction was performed in two steps: First decoction: add 8-fold water, boil for 1 h, filter. Second decoction: add 6-fold water, boil for 30 min, filter. Combine filtrates, concentrate to 100 mL. Mix 10 mL of decoction with ethanol to form 70% solution, refrigerate for 12 h, then filter. Dilute 2 mL filtrate with methanol, mix with internal standard, filter through 0.45 µm membrane, and centrifuge at 15,000 rpm for 5 min.
Quantification of core components
Reference compounds were weighed, dissolved in methanol, and diluted into calibration solutions. HPLC analysis was performed using Elite C18 column (4.6 × 250 mm, 5 µm), with a mobile phase of acetonitrile (A) and 0.1% ammonium acetate (B). The gradient was 10% A to 50% A over 40 min. Detection was at 280 nm, flow rate 0.8 mL/min, column temperature 30°C.
Network pharmacology and molecular docking
Candidate compounds from HLWDD were identified using UPLC-Q-TOF-MS/MS. Targets were retrieved from TCMSP and SwissTargetPrediction, standardized via UniProt, and NAFLD-related targets identified from GeneCards. Venny 2.1.0 was used to find overlapping targets. PPI networks were constructed with STRING and analyzed with Cytoscape (v3.9.0). GO and KEGG enrichment analyses were conducted using DAVID. Molecular docking was performed using AutoDock 4.2.6 and visualized with PyMol.
Animal experimentation
Animal experiments were approved by the Animal Ethics Committee of the Hunan University of Chinese Medicine (Approval No. LLBH-LL2022111704) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals. A total of 64 male SPF-grade Sprague-Dawley rats (200 ± 20 g) were housed (3/cage, 20°C ± 2°C, 12 h light/dark cycle) with ad libitum access to water and standard chow. After 1 week of adaptation, 8 rats formed the control group (normal diet), and 56 received a high-fat, high-sugar, high-salt diet for 20 weeks to induce NAFLD. Successfully modeled rats were randomized into five groups (n = 8/group): model (vehicle), metformin (90 mg/kg) [19], and HLWDD low (3.6 g/kg), medium (7.2 g/kg), and high (14.4 g/kg) dose groups. The selection of HLWDD doses was based on clinical equivalent dose conversion. According to FDA guidelines for dose translation from animal to human studies [20], the human-to-rat equivalent dose ratio is 6.25:1 based on body surface area normalization. Given the clinical dosage of HLWDD in adults is 2.271 g·kg ⁻ ¹·d ⁻ ¹ [21], the calculated rat equivalent dose is 14.196 g·kg ⁻ ¹·d ⁻ ¹ (rounded to 14.4 g/kg). Medium and low doses (7.2 and 3.6 g/kg) were set at 50% and 25% of the high dose, respectively, to evaluate dose-response relationships. All efforts were made to minimize suffering; body weight, food intake, and clinical signs of distress were monitored daily, and veterinary staff were available. No animals required early euthanasia due to distress. After 4 weeks of oral treatment, rats were fasted for 12 h, deeply anesthetized via intraperitoneal injection of sodium pentobarbital (50 mg/kg) (unresponsiveness confirmed by pedal reflex), and euthanized by exsanguination via cardiac puncture. Death was confirmed by cessation of heartbeat and respiration. Blood and liver tissues were collected for analysis. All animals completed the study without mortality or attrition.
Histological and biochemical analysis
Liver tissues were fixed in 4% paraformaldehyde and embedded in paraffin. Sections (5 µm) were stained with H&E and examined by light microscopy. Serum lipid profiles (TC, TG, HDL-C, LDL-C) and liver cytokines (TNF-α, IL-1β, IL-6) were measured by commercial kits.
Gene and protein expression
RNA was extracted via TRIzol and reverse transcribed to cDNA. qPCR was performed using SYBR Green with β-actin as reference. Expression of AMPK, LKB1, CPT1A, PPARα, etc., was calculated using the 2-ΔΔCt method. The sequences of primers are shown in Table 1. Protein expression was analyzed by Western blot. Liver proteins were lysed, separated via SDS-PAGE, transferred to PVDF membranes, and probed with primary/secondary antibodies. Detection was via ECL and quantified with Image J.
Microbiome and metabolomic analysis
Fecal DNA was extracted using PF Mag-Bind Stool DNA Kit. DNA integrity was assessed by agarose gel and quantified by NanoDrop. Metagenomic reads were assembled by MEGAHIT, ORFs predicted with Prodigal, and gene abundance analyzed using SOAPaligner.
Metabolites were analyzed using a Thermo UHPLC-Q Exactive HF-X system with ACQUITY HSS T3 column. Positive and negative ion mode separation was performed with defined gradients. Flow rate: 0.40 mL/min; column temperature: 40°C.
Statistical analysis
Data were analyzed using GraphPad Prism 9 and SPSS 24.0. Results are expressed as mean ± SD. One-way ANOVA followed by LSD-t post hoc test was applied. P < 0.05 was considered statistically significant. KEGG-based pathway enrichment was performed via Python’s scipy.stats and Fisher’s exact test.
Results
Ingredient identification analysis
UPLC-Q-TOF-MS analysis of HLWDD yielded a total ion current chromatogram (Figs 1 and 2). A total of 58 chemical components were identified based on MS/MS fragmentation patterns and database matching using ChemSpider and MassBank; detailed information, including retention times and MS/MS spectra, is provided in S1 Table. These included two amino acids and derivatives (1, 3), two nucleosides and derivatives (2, 58), seven phenolic acids (5, 6, 8, 9, 11, 18, 23), twenty-five flavonoids and glycosides (12–17, 22, 24, 26, 27, 30, 33, 34, 36–38, 40–44, 49, 53, 54), six alkaloids (7, 19, 25, 29, 32, 52), one terpenoid (35), eight coumarins and furanocoumarins (31, 39, 45, 46, 50, 51, 55, 57), two fatty acids and carboxylic acids (4, 28), and five other compounds (20, 21, 47, 48, 56).
Base peak chromatogram of HLWDD in the negative-ion MS mode.
Base peak chromatogram of HLWDD in the positive-ion MS mode.
Network pharmacology analysis
Network pharmacology identified 229 shared targets between HLWDD active ingredients and NAFLD (Fig 3). These are listed in S2 Table for transparency. Among them, 30 potential core targets were used to construct a PPI network in Cytoscape 3.9.0. The top 10 hub targets based on degree values were TP53, AKT1, HSP90AA1, STAT3, EGFR, JUN, MTOR, CASP3, HSP90AB1, and MAPK1 (Fig 4).
Diagram of the target intersection.
Interaction network diagram of core target proteins of HLWDD.
GO and KEGG enrichment analyses via DAVID 6.8 revealed 363 GO terms: 265 biological processes (BP), 45 cellular components (CC), and 72 molecular functions (MF). Top BP terms included response to cadmium ion, regulation of gene expression and nitric oxide production, miRNA transcription, and inhibition of apoptosis (Fig 5). Dominant CC terms were cytoplasm, perinuclear region, protein complex, mitochondrion, and nucleus. Key MF terms included enzyme binding, protein phosphatase binding, and ubiquitin ligase interaction.
KEGG pathway enrichment results pathway diagram.
KEGG analysis yielded 148 enriched pathways; the top 20 are shown in Fig 6, including “Lipid and atherosclerosis,” “Pathways in cancer,” and “Fluid shear stress and atherosclerosis.”
GO enrichment result bubble chart.
The compound-target network (Fig 7) highlighted the top 10 HLWDD components—Obacunone, Caffeic acid, Coptisine, Quercetin, Palmatine, Naringenin, Berberine, Isomeranzin, Tangeretin, and Epiberberine—based on pharmacological relevance and network centrality, along with corresponding targets such as EGFR, AKT1, MAPK1, F2, MAPK8, CYP1A2, CASP3, TNF, TP53, and STAT3.
Drug–ingredients–genes–pathway interaction network diagram.
We validated compound-target interactions via semi-flexible molecular docking using AutoDock. From the drug-ingredient-target-pathway network, the top 10 core compounds and 4 NAFLD-related targets (LKB1, AMPK, PPARα, CPT1A) were selected (Table 2). Binding energy < –5.0 kcal/mol indicated stable interactions; 8 of 24 pairs showed strong affinity (<–7.0 kcal/mol) (Fig 8), suggesting HLWDD compounds may modulate these targets to exert anti-NAFLD effects.
Representative docking results.
Method validation and component quantification of HLWDD
Seven core HLWDD components were quantified based on HPLC analysis (Figs 9–11). All showed excellent linearity (r² ≥ 0.9993). For instance, quercetin (11.00–110.0 μg/mL, Y = 14,647X - 55,556) and berberine (13.00–130.0 μg/mL, Y = 2,046.2X - 8,890.9) displayed strong correlations. Method validation showed precision (RSD ≤ 1.25%), repeatability (RSD ≤ 2.20%), 24 h stability (RSD ≤ 1.62%), and high recovery (99.35–101.33%, RSD ≤ 1.53%). Epiberberine had the highest concentration (164.37 μg/mL), followed by obacunone (122.12 μg/mL) and naringenin (87.57 μg/mL) (Table 3). Detailed validation data are in S3 Table.
Mixed reference substance. Peaks: 1. Quercetin, 2. Epiberberine, 3. Coptisine, 4. Palmatine, 5. Berberine, 6. Naringenin, 7. Houttuynia cordata.
Sample extract. Peaks: 1. Quercetin, 2. Epiberberine, 3. Coptisine, 4. Palmatine, 5. Berberine, 6. Naringenin, 7. Houttuynia cordata.
Blank control. Peaks: 1. Quercetin, 2. Epiberberine, 3. Coptisine, 4. Palmatine, 5. Berberine, 6. Naringenin, 7. Houttuynia cordata.
Animal experiment verification
Therapeutic effects of HLWDD on NAFLD rats.
As NAFLD progression is closely associated with abnormal lipid metabolism, serum biomarkers were assessed. Compared with the control group, the model group exhibited significantly elevated levels of TG, TC, and LDL-C (all P < 0.01), along with a significant reduction in HDL-C (P < 0.01). Treatment with high-dose HLWDD significantly decreased TG (P < 0.05), TC (P < 0.01), and LDL-C (P < 0.01), and significantly increased HDL-C levels (P < 0.01) (Figs 12–17).
Body weight changes during intervention. Data expressed as mean ± SEM (n = 3). Statistical significance: **P < 0.01, ***P < 0.001 vs. Control; #P < 0.05, ##P < 0.01, ###P < 0.001 vs. Model.
Abdominal circumference changes during intervention. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum TC levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum TG levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum LDL-C levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum HDL-C levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Histopathological evaluation using H&E staining revealed clear differences among groups (Fig 18). The control group displayed normal hepatic architecture, characterized by polygonal hepatocytes with homogeneous eosinophilic cytoplasm, centrally located nuclei, and intact lobular structure. In contrast, the model group exhibited classical features of NAFLD, including extensive cytoplasmic lipid vacuolization, eccentric nuclei, disorganized hepatic cords, and marked perivenular fat accumulation, confirming successful model establishment after 20 weeks of a high-fat/high-sugar diet.
Representative H&E staining of liver tissues (scale bar = 100 μm).
As a clinically approved first-line anti-NAFLD agent, Metformin was included as a positive control. Compared with the control group, rats in the model group had significantly elevated serum IL-6, IL-1β, and TNF-α levels (all P < 0.001) (Figs 19–21). These inflammatory markers were significantly reduced following high-dose HLWDD and Metformin treatment, indicating potential anti-inflammatory effects.
Serum IL-6 levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum IL-1β levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Serum TNF-α levels. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 12.
Effects of HLWDD on AMPK/PPARα signaling.
To investigate the regulatory effect of HLWDD on lipid metabolism signaling, we assessed both protein and gene expression of the AMPK/PPARα pathway in liver tissue (Figs 22–30).
Western blot bands of p-LKB1, LKB1, p-AMPK, AMPK, PPARα, CPT1A, and GAPDH.
Relative protein expression of p-LKB1. Data are statistically presented as means ± SEM (n = 3). Compared with the Control, **P < 0.01, ***P < 0.001; compared with the Model, #P < 0.05, ##P < 0.01, ###P < 0.001.
Relative protein expression of p-AMPK. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative protein expression of PPARα. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative protein expression of CPT1A. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative mRNA expression of LKB1. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative mRNA expression of AMPK. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative mRNA expression of PPARα. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Relative mRNA expression of CPT1A. Data expressed as mean ± SEM (n = 3). Statistical significance: same as Fig 23.
Western blot analysis showed that the expression of AMPK, CPT1A, LKB1, and PPARα proteins was significantly decreased in the model group compared to the control (P < 0.05). High-dose HLWDD treatment significantly restored protein expression of all four markers (P < 0.05) (Figs 22–26).
Quantitative PCR results similarly indicated significant downregulation of Ampk, Cpt1a, Lkb1, and Ppara mRNA levels in the model group (P < 0.05), which were significantly upregulated following HLWDD administration (P < 0.05) (Figs 27–30).
Multi-omics integration: Gut-liver axis and AMPK activation.
Metagenomics and metabolomics identified 13 differential metabolites, including fatty acid derivatives (e.g., N-palmitoylthreonine, sphingosine), amino acids (lysine, isoleucine, aspartic acid), and bile acid-related metabolites (e.g., enterolactone) (Fig 31).
Venn diagram of differential metabolites identified by gut metagenomics and liver metabolomics.
Akkermansia positively correlated with aspartic acid (r = 0.71) and pantothenic acid (r = 0.65) (Fig 32). It may alleviate NAFLD by: (1) activating AMPK and promoting PPARα-mediated β-oxidation, and (2) enhancing aspartate metabolism to reduce oxidative stress [22].
Heatmap showing correlations between Akkermansia and key metabolites (aspartic acid, pantothenic acid).Color scales indicate Pearson correlation coefficients (r). Arrows highlight HLWDD-induced restorative trends. Data expressed as mean ± SEM (n = 3). Statistical significance: P < 0.05 (red), P < 0.01 (dark red).
Model rats showed disrupted microbiota-metabolite correlations (e.g., with allysine, L-isoleucine), which HLWDD restored. HLWDD also increased N-palmitoylthreonine (+58%, P < 0.01) and enterolactone (+43%, P < 0.01) (Fig 33), supporting AMPK-mediated lipid regulation.
Gut microbiota–metabolite correlation network in model vs. HLWDD groups. Statistical significance: same as Fig 32.
Pantothenic acid levels decreased by 42% in model rats (P < 0.01), but increased by 37% after HLWDD-H treatment (P < 0.01) (Fig 34).
Hepatic metabolite correlation shifts (model vs. HLWDD-H). Statistical significance: same as Fig 32.
Heatmap analysis showed strong correlations between key proteins (LKB1, AMPK, CPT1A) and metabolites (e.g., aspartic acid: r = 0.82; pantothenic acid: r = 0.76), confirming coordinated regulation of lipid oxidation and providing mechanistic insight (Fig 35).
Integrated heatmap of metabolite–protein interactions (LKB1/AMPK/CPT1A). Statistical significance: same as Fig 32.
Discussion
NAFLD, a common chronic liver condition, is strongly associated with insulin resistance, inflammation, and hepatic lipid metabolic dysfunction [23]. Recent studies have suggested that intestinal flora contributes to NAFLD development by influencing inflammation, oxidative stress, and hepatic energy metabolism through microbial metabolites [23]. Traditional Chinese medicine has garnered interest for its potential to prevent and treat NAFLD. HLWDD, a classical formulation with reputed effects of clearing heat and phlegm, soothing the stomach, and promoting bile secretion, may exert protective effects in NAFLD. This study explored whether HLWDD may modulate intestinal microbiota and metabolic profiles, potentially activating the LKB1/AMPK pathway to attenuate NAFLD progression. Given that AMPK activation is clinically associated with attenuated hepatocellular injury (e.g., reduced ALT/AST) and enhanced insulin sensitivity [24], our molecular findings imply HLWDD may confer similar hepatoprotective and metabolic benefits via AMPK activation. Specifically, the LKB1/AMPK pathway plays a central role in maintaining hepatic energy homeostasis, inhibiting lipogenesis, and enhancing insulin sensitivity [25]. AMPK activation can suppress lipogenic genes such as SREBP-1c and FAS, reducing hepatic fat accumulation [26]. In this study, HLWDD intervention increased hepatic LKB1 and p-AMPK protein expression while reducing SREBP-1c and FAS levels. Mechanistically, AMPK activation suppresses lipogenesis and promotes GLUT4 translocation while inhibiting PGC-1α-mediated gluconeogenesis [27]—collectively suggesting improved hepatic glucose/lipid homeostasis.
Given HLWDD’s modulation of gut microbiota, we focused on SCFAs—key metabolites produced by intestinal flora—as important mediators of host energy metabolism [28]. Our results showed that HLWDD treatment increased the abundance of Akkermansia muciniphila, a mucus-degrading bacterium associated with improved gut barrier integrity and metabolic regulation [29]. Previous research indicates that Akkermansia colonization correlates with amino acid metabolism, particularly aspartate, which may contribute to hepatic energy balance through the gut-liver amino acid axis [30]. SCFAs such as butyrate are known to regulate lipid metabolism via GPR43/41 and AMPK signaling pathways [31].
This study suggests that HLWDD may modulate the gut microbiota in NAFLD rats, increase butyrate-producing bacteria, enhance SCFA levels, and help restore bile acid homeostasis. Metabolomics analysis further revealed increased serum and hepatic aspartate levels following HLWDD treatment. Literature reports support that aspartate may enhance mitochondrial function and suppress lipid synthesis by activating the LKB1/AMPK pathway [32]. Thus, HLWDD may facilitate a regulatory network involving microbiota, metabolites, and signaling pathways.
It is noteworthy that the relationship between aspartate metabolism and LKB1/AMPK activation appears to be bidirectional. AMPK activation can promote aspartate transporter expression, increasing aspartate availability in hepatocytes [33]. Conversely, aspartate, a mitochondrial intermediate, may also modulate LKB1 phosphorylation activity [34]. Correlation analysis in this study showed a strong association between hepatic aspartate levels and the p-AMPK/AMPK ratio, as well as between Akkermansia abundance and liver aspartate concentration. This suggests that HLWDD establishes a self-reinforcing cycle that sustains metabolic improvements by enhancing Akkermansia-associated aspartate flux and LKB1/AMPK activity.
Beyond these metabolic benefits, inflammation is a major contributor to NAFLD progression. Liver-resident Kupffer cells, when overactivated, release inflammatory cytokines such as TNF-α and IL-6 via the NF-κB pathway, exacerbating hepatic injury [35]. Our findings showed that HLWDD reduced the phosphorylation of NF-κB p65 and downregulated these inflammatory cytokines. Critically, combined with microbiota data indicating enhanced gut barrier integrity and SCFA production, these results suggest HLWDD attenuates hepatic inflammation at least partly by modulating gut-derived metabolites—a mechanism unaddressed by conventional NAFLD pharmacotherapies. Consequently, HLWDD demonstrates superior multi-targeted efficacy compared to existing therapies: Metformin only partially activates AMPK while failing to correct gut dysbiosis [36]; pioglitazone significantly increases cardiovascular risk [37]; whereas HLWDD concurrently resolves inflammation through gut-liver axis modulation while avoiding safety concerns associated with conventional agents.
Despite the promising findings, several limitations of this study must be acknowledged. First, only one animal model was used, which limits generalizability. Second, the study did not include a dose-response analysis or compare HLWDD with established therapies beyond metformin. Third, while associations between HLWDD, the microbiota, and metabolic pathways were observed, no functional validation—such as AMPK inhibition or germ-free animal experiments—was performed to confirm causality. Future studies should address these limitations by including multiple models, conducting pharmacodynamic comparisons, and applying targeted interventions to verify mechanisms.
In conclusion, this study suggests that HLWDD may alleviate NAFLD by modulating gut microbiota and associated metabolites and by activating the LKB1/AMPK pathway. These findings provide experimental support for HLWDD’s therapeutic potential and underscore the role of the gut-liver axis in NAFLD pathogenesis.
Supporting information
S1 Table. Chemical constituents identified in HLWDD by UPLC-Q-TOF-MS. Retention times, mass spectral data, and compound classifications.
https://doi.org/10.1371/journal.pone.0331303.s001
(DOCX)
S2 Table. Shared targets between HLWDD active compounds and NAFLD.
Full list of 229 targets identified through network pharmacology.
https://doi.org/10.1371/journal.pone.0331303.s002
(DOCX)
S3 Table. Method validation data for quantification of seven HLWDD components.
Linear regression equations, precision, repeatability, stability, and recovery results.
https://doi.org/10.1371/journal.pone.0331303.s003
(DOCX)
References
- 1. Harrison SA, Allen AM, Dubourg J, Noureddin M, Alkhouri N. Challenges and opportunities in NASH drug development. Nat Med. 2023;29(3):562–73. pmid:36894650
- 2. Leow W-Q, Chan AW-H, Mendoza PGL, Lo R, Yap K, Kim H. Non-alcoholic fatty liver disease: the pathologist’s perspective. Clin Mol Hepatol. 2023;29(Suppl):S302–18. pmid:36384146
- 3. Grander C, Grabherr F, Tilg H. Non-alcoholic fatty liver disease: pathophysiological concepts and treatment options. Cardiovasc Res. 2023;119(9):1787–98. pmid:37364164
- 4. Han SK, Baik SK, Kim MY. Non-alcoholic fatty liver disease: definition and subtypes. Clin Mol Hepatol. 2023;29(suppl):S5–16. pmid:36577427
- 5. Stern C, Castera L. Identification of high-risk subjects in nonalcoholic fatty liver disease. Clin Mol Hepatol. 2023;29(Suppl):S196–206. pmid:36472050
- 6. Tsompanaki E, Thanapirom K, Papatheodoridi M, Parikh P, Chotai de Lima Y, Tsochatzis EA. Systematic review and meta-analysis: the role of diet in the development of nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol. 2023;21(6):1462-1474.e24. pmid:34838723
- 7. Jansen M, Ten Klooster JP, Offerhaus GJ, Clevers H. LKB1 and AMPK family signaling: the intimate link between cell polarity and energy metabolism. Physiol Rev. 2009;89(3):777–98. pmid:19584313
- 8. Kottakis F, Bardeesy N. LKB1-AMPK axis revisited. Cell Res. 2012;22(12):1617–20. pmid:22801477
- 9. Luo Z, Saha AK, Xiang X, Ruderman NB. AMPK, the metabolic syndrome and cancer. Trends Pharmacol Sci. 2005;26(2):69–76. pmid:15681023
- 10. Parker SJ, Svensson RU, Divakaruni AS, Lefebvre AE, Murphy AN, Shaw RJ, et al. LKB1 promotes metabolic flexibility in response to energy stress. Metab Eng. 2017;43(Pt B):208–17. pmid:28034771
- 11. Leung C, Rivera L, Furness JB, Angus PW. The role of the gut microbiota in NAFLD. Nat Rev Gastroenterol Hepatol. 2016;13(7):412–25. pmid:27273168
- 12. Aron-Wisnewsky J, Vigliotti C, Witjes J, Le P, Holleboom AG, Verheij J, et al. Gut microbiota and human NAFLD: disentangling microbial signatures from metabolic disorders. Nat Rev Gastroenterol Hepatol. 2020;17(5):279–97. pmid:32152478
- 13. Albillos A, de Gottardi A, Rescigno M. The gut-liver axis in liver disease: Pathophysiological basis for therapy. J Hepatol. 2020;72(3):558–77. pmid:31622696
- 14. Fang J, Yu C-H, Li X-J, Yao J-M, Fang Z-Y, Yoon S-H, et al. Gut dysbiosis in nonalcoholic fatty liver disease: pathogenesis, diagnosis, and therapeutic implications. Front Cell Infect Microbiol. 2022;12:997018. pmid:36425787
- 15. Saeed H, Díaz LA, Gil-Gómez A, Burton J, Bajaj JS, Romero-Gomez M, et al. Microbiome-centered therapies for the management of metabolic dysfunction-associated steatotic liver disease. Clin Mol Hepatol. 2025;31(Suppl):S94–111. pmid:39604327
- 16. Liu Z-C, Fu H-J, Li N-C, Deng F-J, Gan Y-K, Ye Y-J, et al. Systematic pharmacology and experimental validation to elucidate the inflammation-associated mechanism of Huanglian Wendan (HLWD) decoction in the treatment of MAFLD associated with atherosclerosis. J Ethnopharmacol. 2025;337(Pt 1):118841. pmid:39299361
- 17. Li Y-B, Zhang W-H, Liu H-D, Liu Z, Ma S-P. Protective effects of Huanglian Wendan Decoction aganist cognitive deficits and neuronal damages in rats with diabetic encephalopathy by inhibiting the release of inflammatory cytokines and repairing insulin signaling pathway in hippocampus. Chin J Nat Med. 2016;14(11):813–22. pmid:27914525
- 18. Shi M, Yang J, Liu Y, Zhao H, Li M, Yang D, et al. Huanglian Wendan Decoction improves insomnia in rats by regulating BDNF/TrkB signaling pathway through gut microbiota-mediated SCFAs and affecting microglia polarization. Mol Neurobiol. 2025;62(1):1047–66. pmid:38954253
- 19. Alhamhoom Y, Ahmed SS, M RK, Salahuddin MD, D R B, Ahmed MM, et al. Synergistic antihyperglycemic and antihyperlipidemic effect of polyherbal and allopolyherbal formulation. Pharmaceuticals (Basel). 2023;16(10):1368. pmid:37895839
- 20. Reagan-Shaw S, Nihal M, Ahmad N. Dose translation from animal to human studies revisited. FASEB J. 2008;22(3):659–61. pmid:17942826
- 21. Yue Q-Q, Hao Y-F, Li T, Miao H-Y, Chen B-C, Li S-Y. Traditional Chinese medicine for premature ventricular contraction caused by obstructive sleep apnea: a case report and literature review. Medicine (Baltimore). 2025;104(5):e41206. pmid:39889189
- 22. Ecarnot F, Maggi S. The impact of the Mediterranean diet on immune function in older adults. Aging Clin Exp Res. 2024;36(1):117. pmid:38780713
- 23. Tilg H, Adolph TE, Dudek M, Knolle P. Non-alcoholic fatty liver disease: the interplay between metabolism, microbes and immunity. Nat Metab. 2021;3(12):1596–607. pmid:34931080
- 24. Cai W-F, Chen Q-C, Ni Q, Liu L, Liu Q, Yi Y-K, et al. p-Synephrine ameliorates non-alcoholic fatty liver disease by regulating liver-adipose axis via AMPK/NF-kappa B pathway. J Ethnopharmacol. 2025;348:119890. pmid:40294666
- 25. Tilg H, Adolph TE, Trauner M. Gut-liver axis: Pathophysiological concepts and clinical implications. Cell Metab. 2022;34(11):1700–18. pmid:36208625
- 26. Omidkhoda N, Mahdiani S, Hayes AW, Karimi G. Natural compounds against nonalcoholic fatty liver disease: a review on the involvement of the LKB1/AMPK signaling pathway. Phytother Res. 2023;37(12):5769–86. pmid:37748097
- 27. Su F, Koeberle A. Regulation and targeting of SREBP-1 in hepatocellular carcinoma. Cancer Metastasis Rev. 2024;43(2):673–708. pmid:38036934
- 28. Yang S, Li J, Yan L, Wu Y, Zhang L, Li B, et al. Molecular mechanisms of fucoxanthin in alleviating lipid deposition in metabolic associated fatty liver disease. J Agric Food Chem. 2024;72(18):10391–405. pmid:38669300
- 29. de Vos WM, Tilg H, Van Hul M, Cani PD. Gut microbiome and health: mechanistic insights. Gut. 2022;71(5):1020–32. pmid:35105664
- 30. Liu M-J, Yang J-Y, Yan Z-H, Hu S, Li J-Q, Xu Z-X, et al. Recent findings in Akkermansia muciniphila-regulated metabolism and its role in intestinal diseases. Clin Nutr. 2022;41(10):2333–44. pmid:36113229
- 31. Rao Y, Kuang Z, Li C, Guo S, Xu Y, Zhao D, et al. Gut Akkermansia muciniphila ameliorates metabolic dysfunction-associated fatty liver disease by regulating the metabolism of L-aspartate via gut-liver axis. Gut Microbes. 2021;13(1):1–19. pmid:34030573
- 32. Kimura I, Ozawa K, Inoue D, Imamura T, Kimura K, Maeda T, et al. The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43. Nat Commun. 2013;4:1829. pmid:23652017
- 33. Deng L, Yao P, Li L, Ji F, Zhao S, Xu C, et al. p53-mediated control of aspartate-asparagine homeostasis dictates LKB1 activity and modulates cell survival. Nat Commun. 2020;11(1):1755. pmid:32273511
- 34. Kobayashi T, Nguyen-Tien D, Sorimachi Y, Sugiura Y, Suzuki T, Karyu H, et al. SLC15A4 mediates M1-prone metabolic shifts in macrophages and guards immune cells from metabolic stress. Proc Natl Acad Sci U S A. 2021;118(33):e2100295118. pmid:34385317
- 35. Mridha AR, Wree A, Robertson AAB, Yeh MM, Johnson CD, Van Rooyen DM, et al. NLRP3 inflammasome blockade reduces liver inflammation and fibrosis in experimental NASH in mice. J Hepatol. 2017;66(5):1037–46. pmid:28167322
- 36. Jayachandran M, Qu S. Non-alcoholic fatty liver disease and gut microbial dysbiosis- underlying mechanisms and gut microbiota mediated treatment strategies. Rev Endocr Metab Disord. 2023;24(6):1189–204. pmid:37840104
- 37. Cusi K, Orsak B, Bril F, Lomonaco R, Hecht J, Ortiz-Lopez C, et al. Long-term pioglitazone treatment for patients with nonalcoholic steatohepatitis and prediabetes or type 2 diabetes mellitus: a randomized trial. Ann Intern Med. 2016;165(5):305–15. pmid:27322798