Integrated Network Pharmacology, Metabolomics, And in vivo Validation of Liangxue Dihuang Decoction In DSS-Induced Ulcerative Colitis Mice

Summary

This protocol describes a reproducible workflow integrating network pharmacology, untargeted metabolomics, and in vivo validation to elucidate the multi-component mechanisms of Liangxue Dihuang Decoction in a dextran sulfate sodium–induced mouse model of ulcerative colitis.

Abstract

Ulcerative colitis (UC) is a chronic inflammatory bowel disease characterized by persistent intestinal inflammation and metabolic dysregulation. Traditional Chinese medicine (TCM) formulas are widely used in clinical practice for inflammatory disorders; however, systematic experimental approaches to elucidate their multi-component and multi-target mechanisms remain limited. This article presents a reproducible experimental protocol integrating network pharmacology, untargeted metabolomics, and in vivo validation to investigate the anti-inflammatory effects of Liangxue Dihuang Decoction (LXDHD) in a dextran sulfate sodium (DSS)–induced mouse model of UC. The workflow includes preparation of the herbal decoction, DSS-induced colitis modeling, disease activity assessment, histopathological and hematological evaluation, serum cytokine measurement, LC–MS-based metabolomic profiling, and computational network analysis. Using this integrated approach, DSS-induced colitis was successfully established, and LXDHD treatment resulted in measurable improvements in the disease activity index, colon length, histopathological damage, inflammatory cytokine levels, and metabolic profiles. Network pharmacology and metabolomics analyses further demonstrated that this protocol enables the identification of key targets, pathways, and metabolites associated with therapeutic intervention. This visualized and standardized workflow provides a practical methodological framework for studying the pharmacological mechanisms of multi-component TCM formulas in inflammatory bowel disease and can be adapted to other herbal prescriptions and disease models.

Introduction

Ulcerative colitis (UC) is a chronic inflammatory bowel disease characterized by continuous mucosal inflammation of the colon¹, leading to symptoms such as abdominal pain, diarrhea, rectal bleeding, and weight loss². The incidence of UC has been increasing worldwide, imposing a substantial burden on healthcare systems and significantly impairing patients’ quality of life³. Although the etiology of UC remains incompletely understood, accumulating evidence indicates that dysregulated immune responses, excessive production of inflammatory cytokines, oxidative stress, and metabolic disturbances play central roles in disease pathogenesis^4,5.

Current therapeutic strategies for UC mainly include aminosalicylates, corticosteroids, immunosuppressive agents, and biologics targeting inflammatory cytokines⁶. While these treatments can alleviate symptoms and induce remission, their long-term use is often associated with adverse effects, high costs, and variable therapeutic responses^7,8,9. Moreover, a considerable proportion of patients fail to achieve sustained remission, highlighting the need for alternative or complementary therapeutic approaches with improved safety and multi-target efficacy.

Traditional Chinese medicine (TCM) has been widely used for the treatment of gastrointestinal disorders for centuries and is increasingly recognized for its potential benefits in complex inflammatory diseases such as UC¹⁰. Unlike single-target therapies, TCM formulas typically exert therapeutic effects through multiple components acting on multiple targets and pathways, which may better address the multifactorial nature of UC¹¹. Clinical and experimental studies have suggested that certain TCM prescriptions can modulate immune responses, suppress inflammation, protect intestinal barrier integrity, and regulate metabolic homeostasis^12,13,14,15.

Liangxue Dihuang Decoction (LXDHD) is a classical TCM formula traditionally used to “cool the blood and clear heat” and has been clinically applied for inflammatory and hemorrhagic disorders¹⁶. The prescription consists of multiple herbs, including Scutellaria baicalensis, Phellodendron chinense, Rehmannia glutinosa, Paeonia suffruticosa, Paeonia lactiflora, Sanguisorba officinalis, Platycladus orientalis, Saposhnikovia divaricata, Citrus aurantium, Angelica sinensis, and Glycyrrhiza uralensis. Modern pharmacological studies have shown that several constituent herbs and their active compounds possess anti-inflammatory, antioxidant, and immunomodulatory properties^17,18,19. However, the overall therapeutic effects and underlying mechanisms of LXDHD in UC have not yet been systematically elucidated.

Network pharmacology has emerged as an effective approach to explore the complex interactions between multi-component herbal formulas and disease-related molecular targets²⁰. By integrating information from multiple databases, network pharmacology can predict potential active compounds, therapeutic targets, and signaling pathways, providing a systems-level understanding of TCM mechanisms²¹. Nevertheless, network pharmacology primarily relies on database predictions and does not directly reflect in vivo metabolic changes.

Metabolomics, as a powerful tool for profiling global metabolic alterations, offers complementary insights into disease pathophysiology and therapeutic responses²². In UC, metabolomic analyses have revealed significant disturbances in lipid and amino acid metabolism, as well as oxidative stress-related pathways^23,24,25. Integrating metabolomics with network pharmacology can help bridge the gap between predicted molecular targets and actual metabolic phenotypes, thereby improving the reliability of mechanistic interpretations.

In the present study, an integrated strategy combining network pharmacology, untargeted metabolomics, and experimental validation was employed to investigate the anti-inflammatory effects and potential mechanisms of LXDHD in a dextran sulfate sodium (DSS)–induced mouse model of UC (Figure 1A). This overall methodological design, integrating in silico predictions with in vivo metabolomic and phenotypic validation, is in accordance with established rigorous research frameworks²⁶. Network pharmacology analysis was used to identify active compounds, potential therapeutic targets, and signaling pathways enriched by LXDHD treatment. Untargeted metabolomics was performed to characterize metabolic alterations induced by UC and their modulation by LXDHD. Furthermore, key findings were validated through in vivo experiments, including assessments of disease activity, histopathological changes, hematological parameters, and inflammatory cytokine levels.

By integrating computational predictions, metabolic profiling, and biological validation, this study aims to provide a comprehensive understanding of the multi-component, multi-target, and multi-pathway mechanisms underlying LXDHD's therapeutic effects in UC. This work not only offers experimental evidence supporting the use of LXDHD as a potential complementary therapy for UC but also provides a methodological framework for investigating the mechanisms of other complex TCM formulas.

Protocol

All animal experimental procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the Animal Ethics Committee of the Affiliated Hospital of Nanjing University of Chinese Medicine (Approval No. #2025DW-034-01; Approval Date: March 24, 2025). The preparation of Liangxue Dihuang Decoction was performed in accordance with the procedures described in the Chinese Pharmacopoeia²⁷. The reagents, chemicals, and software used in the protocol are listed in the Table of Materials.

1. Preparation of Liangxue Dihuang Decoction (LXDHD)

1. Prepare the dried herbal materials according to the standard formulation of LXDHD. The precise herbal composition, botanical names, and exact prescribed dosages of the eleven constituent herbs are specified in Table 1.
   NOTE: Authenticate all herbal materials by a qualified pharmacognosist prior to use to ensure botanical identity and quality consistency. Retain authentication records for reproducibility.
2. Powder preparation
   1. Air-dry all herbal materials at room temperature in a clean, well-ventilated environment.
   2. Pulverize each herb individually using a mechanical grinder until a fine powder is obtained.
   3. Combine the powders according to the prescribed ratios.
   4. Pass the mixed powder through an 80-mesh sieve (inner diameter 180 µm ± 7.6 µm) to ensure uniform particle size.
   5. Thoroughly homogenize the sieved powder.
      NOTE: Uniform particle size is essential for consistent extraction efficiency. Incomplete grinding or uneven mixing may result in variability in chemical composition.
3. Preparation of LXDHD aqueous extract
   1. Accurately weigh the mixed herbal powder and add distilled water at a ratio of 1:10 (w/v).
   2. Heat the mixture to boiling and decoct for 20 min.
   3. Filter the extract and collect the filtrate.
   4. Add fresh distilled water to the residue and repeat decoction for an additional 20 min.
   5. Combine the two filtrates to obtain the final LXDHD aqueous extract.
      NOTE: Double decoction is essential to ensure efficient extraction of both water-soluble and moderately soluble components.
4. Concentration and storage
   1. Concentrate the combined decoction under reduced pressure or gentle heating until the desired concentration is achieved.
   2. Allow the extract to cool to room temperature.
   3. Store the prepared LXDHD extract at 4 °C and use within 48 h.
      NOTE: Avoid repeated freeze–thaw cycles, as they may degrade bioactive compounds and reduce experimental reproducibility.
5. Administration
   1. Bring the extract to room temperature prior to administration.
   2. Administer the concentrated aqueous decoction of LXDHD to mice by oral gavage at the designated doses according to experimental grouping.

2. Animal experiment design

1. Animals and housing conditions
   1. Obtain male C57BL/6 mice (aged 6–8 weeks, 20–25 g) from a certified laboratory animal supplier.
   2. House mice under standard laboratory conditions with a 12 h light/dark cycle, controlled temperature (22–25 °C), and relative humidity (50%–60%).
   3. Provide mice with free access to standard chow and water throughout the experiment.
   4. Allow mice to acclimatize for at least 7 days prior to experimentation.
      NOTE: Include only healthy mice with normal activity and body weight in subsequent experiments.
2. Grouping and dose calculation
   1. Calculate the mouse dosage of LXDHD according to the human–animal dose conversion principle described in Methodology of Pharmacological Experiments, based on a standard adult body weight of 70 kg and a daily clinical dose of 8.4 g²⁸.
   2. Randomly divide mice into the following groups (n = 6 per group): Normal control group, DSS-induced UC model group, low-dose LXDHD group (1.09 g/kg), high-dose LXDHD group (2.18 g/kg), and 5ASA positive control group (100 mg/kg)²⁹.
      NOTE: Perform randomization using a random number table to minimize selection bias.
   3. Induction of UC and drug administration
      1. Prepare 2.5% (w/v) DSS solution in drinking water.
      2. Administer DSS solution ad libitum to mice in the model, LXDHD-treated, and 5ASA groups for 7 consecutive days³⁰.
      3. Provide purified drinking water to mice in the normal control group.
      4. Administer LXDHD and 5ASA to the treatment groups once daily by oral gavage at the corresponding doses throughout the experimental period.
      5. Administer an equal volume of vehicle solution to the normal control and DSS model groups.
         NOTE: Prepare the DSS solution fresh and replace it every 2 days. DSS batches may vary in colitis-inducing potency; perform pilot testing when using a new batch. Successful establishment of the DSS-induced UC model is characterized by progressive body weight loss, bloody diarrhea, elevated DAI scores, and marked colon shortening. These criteria and the methodological design of the mouse experiments are consistent with established reference standards³¹.
3. Preparation of blank and drug-containing serum for exogenous metabolomics
   1. Use a separate cohort of healthy male C57BL/6 mice.
   2. Randomly divide mice into two groups: the blank group and the LXDHD group.
   3. Administer LXDHD to the LXDHD group by oral gavage at 2.18 g/kg once daily for 7 consecutive days.
   4. Administer an equal volume of distilled water to the blank group.
   5. Collect blood samples 1 h after the final administration.
   6. Process blood samples to obtain serum as described ahead in Section 3.5.

3. Assessment of disease activity

1. Monitor body weight, stool consistency, and fecal bleeding daily during the experiment.
2. Calculate the Disease Activity Index (DAI) based on body weight loss, stool consistency, and fecal bleeding. The specific scoring framework is detailed in Table 3. Calculate the
   final DAI score as the average (or sum) of these three parameters³².
3. At the end of day 8, anesthetize mice from each group with 1% sodium pentobarbital (30 mg/kg). Collect approximately 400 µL of blood from the retro-orbital venous plexus of each mouse. Subsequently, euthanize the animals by cervical dislocation to ensure death, in strict accordance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals (Figure 1B).
   NOTE: Confirm adequate anesthesia by gently pinching the toes and plantar surface of the hind paws; absence of a withdrawal response indicates sufficient depth of anesthesia.
4. Collect colon tissues and measure colon length as an indicator of disease severity.
   NOTE: Shortened colon length is considered a hallmark of DSS-induced colitis severity³³.
5. Obtaining blood serum
   1. Allow a portion of the collected blood samples to clot at room temperature for 1 h.
   2. Centrifuge samples at 1,000 × g for 15 min at 4 °C to obtain serum.
   3. Quantify serum levels of interleukin (IL)-1β, IL-6, IL-17, tumor necrosis factor-alpha (TNF-α), and IL-10.
   4. Collect colon tissue samples for histopathological analysis.
   5. Use remaining blood samples for peripheral blood analysis.
   6. Reserve additional serum samples for metabolomics analysis.

4. Hematoxylin–Eosin (H&E) staining

1. Fix colon tissues in 4% paraformaldehyde for at least 24 h.
2. Dehydrate tissues through a graded ethanol series.
3. Clear tissues with xylene and embed in paraffin.
4. Cut tissue sections (3–5 µm thick) using a microtome.
5. Deparaffinize sections with xylene.
6. Rehydrate sections through graded ethanol.
7. Stain sections with hematoxylin for nuclear visualization.
8. Counterstain sections with eosin for cytoplasmic staining.
9. Observe histopathological changes under a light microscope and capture representative images using the associated imaging software.
10. Evaluate histological scores (HCS) based on the severity of inflammation, depth of injury, and crypt damage.
11. Refer to Table 4 for detailed histological scoring criteria.
    NOTE: Capture representative images of tissue embedding, sectioning, and staining to enhance protocol visualization.

5. Peripheral blood analysis

1. Collect peripheral blood samples into EDTA-K2 anticoagulant tubes (1.5 mg EDTA per mL of blood).
2. Gently invert tubes 8–10 times to ensure proper mixing.
3. Analyze blood cell counts using a three-part veterinary hematology analyzer (Mindray BC-2600vet) with species-specific calibration.
4. Transport samples at 4  °C.
5. Analyze samples within 4 h after collection.
   NOTE: Exclude samples showing hemolysis or coagulation from analysis.

6. Enzyme-linked immunosorbent assay (ELISA)

NOTE: IL-6, IL-17, IL-1β, and TNF-α are key pro-inflammatory cytokines, whereas IL-10 is a key anti-inflammatory cytokine. These cytokines reflect the presence and severity of inflammatory responses³⁴.
Measurement of these cytokines is crucial for understanding inflammatory processes and evaluating the effectiveness of anti-inflammatory treatments.

1. Collect blood samples from the retro-orbital venous plexus.
2. Allow samples to clot at room temperature for 1 h.
3. Centrifuge samples at 1,000 × g for 15 min at 4 °C to obtain serum.
4. Measure serum levels of IL-1β, IL-6, IL-17, TNF-α, and IL-10 using commercial ELISA kits.
5. Measure absorbance using a microplate reader equipped with standard data acquisition software.
6. Calculate cytokine concentrations based on standard curves.
   NOTE: Assay all samples and standards in duplicate to reduce technical variability.

7. LC–MS analysis of Liangxue Dihuang Decoction

1. Sample preparation
   1. Accurately weigh 10 g of the LXDHD mixed raw herbal powder.
   2. Dissolve the powder in 30 mL of hot water.
   3. Perform ultrasonic extraction for 30 min.
   4. Mix 100 µL of the extract with 400 µL of methanol.
   5. Vortex for 10 min.
   6. Centrifuge at 14,000 × g for 10 min.
   7. Collect the supernatant as the test solution.
2. LC–MS conditions
   1. Perform chromatographic separation using an LC system coupled with a Q mass spectrometer.
   2. Use a C18 column (150 × 2.1 mm, 1.8 µm).
   3. Set the mobile phase as follows: solvent A, 0.1% formic acid in water; solvent B, methanol.
   4. Set the flow rate to 0.30 mL/min.
   5. Set the column temperature to 35 °C.
   6. Set the injection volume to 5 µL.
   7. Apply the following gradient-elution program:
      0–2 min, 5% B;
      2–15 min, 5–95% B;
      15–18 min, 95% B;
      18–18.1 min, 95–5% B;
      18.1–22 min, 5% B.
   8. Set mass spectrometry parameters as follows: sheath gas flow rate, 35 arb; aux gas flow rate, 10 arb; spray voltage, 3.5 kV (positive) and 2.8 kV (negative); capillary temperature, 320 °C; aux gas heater temperature, 350 °C.
   9. Acquire MS data in both positive and negative ion modes with a full MS scan range of m/z 100–1500 (resolution: 70,000).
   10. Perform data-dependent MS2 (dd-MS2) scans at a resolution of 17,500.
   11. Process raw data and identify compounds using the mzCloud database.

8. Serum untargeted metabolomics analysis

1. Serum sample preparation and quality control (QC)
   1. Thaw serum samples on ice.
   2. Transfer 50 µL of serum into a 1.5 mL centrifuge tube.
   3. Add 200 µL of cold methanol/acetonitrile (1:1, v/v).
   4. Vortex for 1 min to precipitate proteins.
   5. Incubate on ice for 15 min.
   6. Centrifuge at 14,000 × g for 15 min at 4 °C.
   7. Transfer the supernatant to a new tube.
   8. Dry the extract using a vacuum concentrator.
   9. Reconstitute in 100 µL of acetonitrile/water (1:1, v/v).
   10. Vortex thoroughly.
   11. Centrifuge again at 14,000 × g for 15 min at 4 °C.
   12. Collect the supernatant for LC-MS analysis.
   13. Prepare QC samples by pooling 10 µL aliquots from each sample.
   14. Inject one QC sample for every 8–10 experimental samples to monitor instrument stability and repeatability.
2. LC-MS acquisition settings
   1. Perform chromatographic separation using the LC system with a C18 column (150 × 2.1 mm, 1.8 µm).
   2. Set the column temperature to 35 °C.
   3. Set the flow rate to 0.30 mL/min.
   4. Set the injection volume to 2 µL (positive mode) or 5 µL (negative mode).
   5. Use solvent A (0.1% formic acid in water) and solvent B (acetonitrile with 0.1% formic acid).
   6. Apply a linear gradient elution program.
   7. Operate the mass spectrometer in positive and negative HESI modes.
   8. Set the mass scan range to m/z 100–1500.
   9. Set spray voltage to 3.5 kV (positive) and 2.8 kV (negative).
   10. Set capillary temperature to 320 °C.
3. Data processing and differential metabolite identification
   1. Import raw LC-MS data (.raw files) into the software.
   2. Perform peak alignment, peak picking, and baseline correction.
   3. Normalize extracted peak area features to total ion intensity.
   4. Identify metabolites by matching accurate mass (mass tolerance < 5 ppm) and MS/MS fragmentation patterns against mzCloud and ChemSpider databases.
   5. Perform multivariate statistical analysis (PCA and OPLS-DA/PLS-DA).
   6. Define differential metabolites based on the following thresholds: VIP > 1.0 and p < 0.05 (Student’s t-test).

9. Network pharmacology analysis

NOTE: Network pharmacology was employed to predict active components and key targets of LXDHD in UC^35,36. The methodology includes database mining, chemical information processing, bioactivity data acquisition, protein target retrieval, gene expression profiling, interaction network construction, and pathway enrichment analysis^37,38.

1. Active compound screening
   1. Retrieve active compounds from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis (TCMSP) database (accessed December 10, 2025) using OB ≥ 30% and DL ≥ 0.18.
   2. Supplement compounds through literature searches in PubMed and Web of Science (December 10–15, 2025).
   3. Standardize targets using the UniProt database (accessed December 15, 2025).
2. Disease target identification
   1. Collect UC-related targets from GeneCards, OMIM, TTD, and DrugBank databases (accessed December 15, 2025).
3. Network construction and analysis
   1. Identify intersecting targets using the VennDiagram package in R software.
   2. Construct the network using Cytoscape software.
   3. Analyze topology using the NetworkAnalyzer plugin and the CytoNCA plugin.
4. PPI network and core target identification
   1. Import intersecting targets into the STRING database.
   2. Restrict species to Homo sapiens.
   3. Remove isolated nodes.
   4. Apply a PPI score threshold > 0.9.
   5. Export data and analyze in the software.
   6. Calculate DC, BC, EC, CC, NC, and LAC.
   7. Perform median-based filtering to generate subnetworks and identify core targets³⁹.
5. GO and KEGG Enrichment Analysis
   1. Perform GO and KEGG enrichment using R and Bioconductor packages.
   2. Use R packages including org.Hs.eg.db, clusterProfiler, DOSE, enrichplot, stringi, and ggplot2.
   3. Set significance thresholds at p = 0.05 and q = 0.05.
   4. Conduct GO analysis for MF, BP, and CC categories.
   5. Perform KEGG pathway enrichment and classification⁴⁰.
6. Molecular docking
   1. Download compound structures from TCMSP in SDF format.
   2. Download protein structures from AlphaFold in PDB format.
   3. Perform docking using the CB-DOCK2 web server (CADD Lab, Sichuan University, Chengdu, China).
   4. Define receptor and ligand files accordingly.
   5. Record optimal binding pose and binding energy.
      ​NOTE: Docking was performed using a genetic algorithm⁴¹. Results were ranked by binding energy.

10. Statistical analysis

1. Perform statistical analysis using GraphPad Prism software.
2. Express quantitative data as mean ± standard deviation (SD).
3. Test for normality using the Shapiro–Wilk test.
4. Test for homogeneity of variance using the Brown–Forsythe test.
5. Perform one-way ANOVA with Tukey’s post hoc test for normally distributed data.
6. Perform Kruskal–Wallis test with Dunn’s test for non-parametric data.
7. Consider p < 0.05 as statistically significant.
   NOTE: A smaller p-value indicates a greater statistical difference between groups.

Results

Network pharmacology analysis of LXDHD in UC

Network pharmacology was applied to predict the potential therapeutic targets and molecular mechanisms of LXDHD in UC. A total of 211 compounds with favorable pharmacokinetic properties were screened from the TCMSP database, including 36 from Scutellaria baicalensis, 37 from Phellodendron chinense, 10 from Rehmannia glutinosa, 11 from Paeonia suffruticosa, 29 from Paeonia veitchii, 5 from...

Discussion

This method presents a visualized, reproducible experimental workflow integrating network pharmacology, untargeted metabolomics, and in vivo validation to investigate the anti-inflammatory effects of TCM formulas in a DSS-induced mouse model of UC. Unlike conventional pharmacological studies that primarily focus on isolated mechanistic conclusions, this protocol emphasizes the robustness of the methodological framework itself while revealing critical biological insights. A major advantage of this integrated work...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
5-Aminosalicylic Acid (5-ASA)	Sigma-Aldrich	A3537	Positive control drug for ulcerative colitis; administered at 100 mg/kg by oral gavage.
Acetonitrile (LC-MS grade)	Merck	100030	Used for metabolomics sample preparation and LC-MS mobile phase.
AlphaFold Protein Structure Database	EMBL–European Bioinformatics Institute	https://alphafold.ebi.ac.uk/	Database used to obtain predicted three-dimensional structures of target proteins.
Angelica sinensis (Dang Gui)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 1.6 g).
CB-Dock2	CADD Lab	https://cadd.labshare.cn/cb-dock2/	Web-based platform used for molecular docking and prediction of ligand–protein binding modes.
Citrus aurantium (Zhi Ke)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.5 g).
Compound Discoverer Software	Thermo Fisher Scientific	Version 3.3	Used for LC–MS data processing and compound identification.
CTAn Software	Bruker	Version 1.17	Used for Micro-CT 3D image reconstruction and bone erosion analysis.
Cytoscape	Cytoscape Consortium	Version 3.8.0	Software used for network construction and visualization.
Dextran Sulfate Sodium (DSS)	MP Biomedicals	160110	Molecular weight 36,000–50,000 Da; used to induce ulcerative colitis in mice.
EDTA-K2 Anticoagulant Tubes	Sinopharm Chemical Reagent Co., Ltd.	N/A	Used for peripheral blood collection.
Eosin Staining Solution	Biosharp	N/A	Used for cytoplasmic counterstaining in H&E staining.
Ethanol	Chengdu Cologne Chemicals Co., LTD	64-17-5	Used for tissue dehydration.
Formic acid (LC-MS grade)	Merck	5.33002	Used for LC-MS mobile phase preparation.
Glycyrrhiza uralensis (Gan Cao)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.5 g).
GraphPad Prism	GraphPad Software	Version 9.0	Used for statistical analysis and graphical presentation.
Hematology Analyzer	Mindray	BC-2600vet	Three-part veterinary hematology analyzer for blood routine analysis.
Hematoxylin Staining Solution	Biosharp	BL700B	Used for H&E staining of colon tissue sections.
High-speed Centrifuge	Hitachi	CT15E/CT15RE	Used for serum and sample preparation.
IL-10 ELISA Kit (Mouse)	MEIMIAN	MM-0176M2	Used to assess serum IL-10 levels.
IL-17 ELISA Kit (Mouse)	Abclonal	RK00039	Used to measure serum IL-17 levels.
IL-1β ELISA Kit (Mouse)	Abclonal	RK00006	Used for detection of serum IL-1β levels.
IL-6 ELISA Kit (Mouse)	Abclonal	RK00008	Used to quantify serum IL-6 levels.
ImageJ (Fiji) Software	National Institutes of Health (NIH)	Version 1.53c	Open-source software used for image processing and quantification.
Leica Application Suite X (LAS X)	Leica Microsystems	Version 3.7	Imaging software used for capturing histopathological slides.
Methanol (LC-MS grade)	Merck	106035	Used for metabolomics sample preparation and LC-MS mobile phase.
Mice (Male C57BL/6)	Anokang Biotechnology Co., Ltd.	N/A	Experimental animals (aged 6–8 weeks, 20–25 g) used for DSS-induced UC model.
mzCloud Database	Thermo Fisher Scientific	https://www.mzcloud.org/	Database used for MS/MS spectral matching.
Optical Microscope	Leica	DM500	Used for histopathological observation.
Paeonia lactiflora (Chi Shao)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.6 g).
Paeonia suffruticosa cortex (Mu Dan Pi)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.5 g).
Paraformaldehyde (4%)	Sinopharm Chemical Reagent Co., Ltd.	N/A	Used for fixation of colon tissues.
Pathological Microtome	Leica	RM2016	Used for preparation of paraffin tissue sections.
Phellodendron chinense (Huang Bo)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.5 g).
Platycladus orientalis (Ce Bai Ye Tan)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.7 g).
Q Exactive High-Resolution Mass Spectrometer	Thermo Fisher Scientific	Q Exactive	Orbitrap-based mass spectrometer for LC–MS analysis.
R Software	R Foundation for Statistical Computing	Version 4.2.1 (https://www.r-project.org/)	Software used for GO and KEGG enrichment analyses.
Rehmannia glutinosa (Sheng Di Huang)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 2.1 g).
Sanguisorba officinalis (Di Yu Tan)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.6 g).
Saposhnikovia divaricata (Fang Feng)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.3 g).
Scutellaria baicalensis (Huang Qin)	Jiangsu Province Hospital of Chinese Medicine	Batch: 2409001	Constituent herb of LXDHD (Dose: 0.5 g).
Sodium pentobarbital (1%)	Sigma-Aldrich	P3761	Used for animal anesthesia prior to sample collection.
SoftMax Pro Software	Molecular Devices	Version 7.1	Data acquisition and analysis software used with the microplate reader for ELISA.
STRING Database	STRING Consortium	https://string-db.org/	Database used for protein–protein interaction analysis.
TCMSP Database	TCMSP Platform	https://tcmsp-e.com/	Database used to screen active compounds and predict drug targets.
TNF-α ELISA Kit (Mouse)	Abclonal	RK00027	Used to determine serum TNF-α concentrations.
Ultimate 3000 RS LC System	Thermo Fisher Scientific	Ultimate 3000 RS	UHPLC system coupled with mass spectrometry.
Ultrasonic Cleaner	Kunshan Ultrasonic Instruments Co., Ltd	KQ3200E	Used for ultrasonic extraction of LXDHD.
UniProt Database	UniProt Consortium	https://www.uniprot.org/	Database used for gene and protein name standardization.
Vortex Mixer	Beijing PowerStar Technology Co., Ltd	LC-Vortex-P1	Used for sample mixing.
Welch AQ-C18 Column (150 × 2.1 mm, 1.8 μm)	Welch Materials	AQ-18-15021	Column used for separation of metabolites and herbal components.
Xylene	Chengdu Cologne Chemicals Co., LTD	1330-20-7	Used for tissue clearing during histological processing.