Fructus xanthii Extract Alleviates Osteoarthritis by Preserving Cartilage Integrity

Summary

This study protocol was designed to systematically investigate the potential chondroprotective effects of Fructus Xanthii extract in osteoarthritis through integrated multi-omics analyses and in vivo experiments.

Abstract

Osteoarthritis (OA) is a progressive degenerative joint disorder for which current pharmacological and surgical interventions mainly relieve symptoms rather than reverse the underlying pathological changes, highlighting the urgent need for novel therapeutic agents. This protocol aimed to investigate the potential chondroprotective effects of Fructus Xanthii extract against OA. First, non-targeted metabolomics profiling was conducted using ultra-high-performance liquid chromatography coupled with high-resolution tandem mass spectrometry (UHPLC-HRMS/MS) to characterize the chemical constituents of the extract. An integrated strategy combining network pharmacology and transcriptomic mining was then applied to identify bioactive components and putative targets of the extract, followed by cross-analysis with OA-related dysregulated genes to obtain core target genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed to annotate the main biological processes and signaling pathways modulated by the extract. For in vivo validation, zebrafish cartilage injury models and mouse OA models were used to evaluate the chondroprotective effects of the extract. RNA sequencing (RNA-seq) was further adopted to analyze transcriptomic alterations in chondrocytes after treatment, focusing on OA-associated pathways and gene expression profiles; enzyme-linked immunosorbent assay (ELISA) was subsequently performed to verify the expression levels of key proteins/receptors screened by RNA-seq, so as to further confirm the regulatory effects of Fructus Xanthii extract on core OA targets. This integrated multi-omics and experimental protocol provides a systematic approach to explore the potential of Fructus Xanthii extract as a promising botanical candidate for OA intervention.

Introduction

Osteoarthritis (OA) is a degenerative joint disorder characterized by the progressive degradation of articular cartilage, subchondral bone sclerosis, synovial inflammation, and osteophyte formation¹. According to the Global Burden of Disease (GBD) study, the prevalence of OA has steadily increased across 204 countries and territories from 1990–2020², with this trend expected to accelerate alongside global demographic ageing³. Current estimates indicate that over 250 million individuals worldwide are affected, placing significant strain on individual well-being, healthcare systems, and socioeconomic structures⁴. Current treatments are predominantly aimed at symptomatic relief, with non-steroidal anti-inflammatory drugs (NSAIDs) and intra-articular viscosupplementation with hyaluronic acid (HA) being the most common interventions. However, these approaches offer limited clinical benefit and are associated with considerable adverse effects^5,6. This highlights the critical need for effective disease-modifying osteoarthritis drugs (DMOADs).

Traditional Chinese Medicine (TCM) has long been used in the management of OA, grounded in the concept of "bone bi," a subtype of "bi syndrome"⁷. In this framework, the pathogenesis of OA is attributed primarily to a deficiency in liver and kidney essence, insufficiency of qi and blood, and subsequent invasion by wind, cold, and dampness⁸. Accordingly, therapeutic strategies focus on dispelling wind-dampness, dispersing cold, and unblocking the collaterals⁹. TCM treatments are noted for their multi-target effects, favorable side-effect profiles, and clinically significant efficacy¹⁰. Noteworthy formulations such as Buqi Tongluo Capsule (BQTL) and Fuzi Decoction (FZD) have shown clear anti-OA effects in clinical practice, while Du Huo Ji Sheng Tang (DHJST) alleviates OA symptoms by suppressing the NLRP3 inflammasome pathway, supported by centuries of empirical use. Similarly, Shenjinhuoxue Mixture (SHM) attenuates pain and cartilage degeneration through down-regulation of pro-inflammatory mediators such as IL-1β and TNF-α^11,12,13. At the monomer level, compounds such as icariin (Cyanoside A, CyA) and erianin, derived from medicinal plants, have been reported to promote chondrocyte proliferation, inhibit inflammatory cytokine release, and maintain joint homeostasis^14,15. Collectively, TCM offers a comprehensive therapeutic approach with anti-inflammatory, chondroprotective, and symptom-alleviating properties. However, the complexity of its mechanisms and issues related to standardization necessitate ongoing, rigorous research.

The dried ripe fruits of Fructus Xanthii (Compositae), commonly known as “Cang-Er-Zi”, have been documented to exhibit a wide range of pharmacological activities, including antimicrobial, anti-inflammatory, antioxidant, analgesic, antineoplastic, and immunomodulatory effects¹⁶. Phenolic acids isolated from Fructus Xanthii effectively reduce synovial hyperplasia in rheumatoid arthritis (RA)¹⁷. Among these, chlorogenic acid (CGA), a representative caffeoylquinic acid derivative, has been found to stimulate osteoblast proliferation and differentiation while inhibiting RANKL-mediated osteoclastogenesis, thus maintaining skeletal homeostasis¹⁸. Other phenolic acids identified within the same extract, such as neochlorogenic acid (5-CQA), cryptochlorogenic acid (CCA), and protocatechuic acid, also exhibit antimicrobial, anti-inflammatory, antitumor, and antioxidant properties^19,20,21. In addition to phenolic acids, a variety of secondary metabolites present in Cang-Er-Zi contribute to these bioactivities^{16,20,22,23,24,25}. However, the precise role and underlying mechanisms of Fructus Xanthii extract in OA remain largely unexplored.

Non-Targeted Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) Metabolomics is characterized by hypothesis-free, high-throughput, and global coverage capabilities. It enables the simultaneous detection of thousands of metabolite features, providing a panoramic view of metabolic networks, and allows for the systematic discovery of novel biomarkers and aberrant metabolic pathways without prior target prespecification^26,27. Combined with high-resolution mass spectrometry technology, this approach is suitable for the comprehensive analysis of various types of biological samples (e.g., serum, synovial fluid, urine, cartilage tissue, etc.) and has demonstrated significant advantages in osteoarthritis (OA) research. On the one hand, it can comprehensively screen for differential metabolites closely associated with the pathological progression of OA (such as various classes of metabolites identified in synovial fluid and synovial tissue) and map them to key pathways, including energy metabolism, cartilage repair, osteogenesis, and lipid metabolism, thereby deepening the understanding of OA pathogenesis. On the other hand, integrating multi-omics data such as transcriptomics and proteomics helps to reveal the role of metabolic reprogramming in cartilage degeneration, inflammatory responses, and joint structure damage, providing critical evidence for identifying early diagnostic biomarkers, evaluating disease progression, and developing targeted metabolic intervention strategies (e.g., regulating bile acid, tryptophan, or fatty acid metabolism), highlighting its translational potential in research on OA precision typing and personalized treatment.

Zebrafish serves as an excellent model organism due to multiple advantages: its genome is highly conserved with that of humans, sharing homology with approximately 70% of human protein-coding genes^28,29. It exhibits key characteristics including rapid development, in vitro fertilization, transparent embryos enabling convenient in vivo imaging, high fecundity, and low maintenance costs^30,31,32,33. In skeletal research, zebrafish allow for clear visualization of bone formation, mineralization processes, and regeneration capacity, and can be utilized to simulate pathological conditions like osteoporosis and cartilage damage. However, in the study of osteoarthritis (OA), a complex whole-joint disorder characterized by articular cartilage degeneration, synovial inflammation, and subchondral bone alterations, the application of the zebrafish model remains relatively limited.

This article presents a protocol that integrates UHPLC-HRMS/MS-based untargeted metabolomics, network pharmacology, transcriptomic mining, and in vivo experimental validation to explore the potential of Xanthium sibiricum extract as a viable plant candidate for OA intervention. The aim is not only to identify the bioactive components, potential targets, and core regulatory pathways of Xanthium sibiricum extract related to OA through multi-omics integration and computational analysis, but also to verify its chondroprotective effect on OA through in vivo models and further clarify its regulatory mechanism on chondrocyte transcriptome.

Protocol

All experiments involving clinical samples and mice were conducted in accordance with the protocol approved by the Ethics Review Committee of Shandong Provincial Medical Biotechnology Research Center. The reagents and the equipment used are listed in the Table of Materials.

1. Isolation and culture of human articular chondrocytes

Human articular cartilage specimens were obtained from patients undergoing arthroscopic surgery or total joint arthroplasty at the First Affiliated Hospital of Shandong First Medical University. Cartilage slices were transported to the laboratory on ice within 30 min of surgical excision. The tissue was washed thoroughly with ice-cold Dulbecco’s phosphate-buffered saline (DPBS; 3–5 washes), then minced finely into approximately 1 mm³ fragments using sterile surgical scissors. Sequential enzymatic digestion was performed, beginning with pre-digestion using 0.25% trypsin-EDTA for 30 min at 37 °C. The digestion process was monitored until the cartilage fragments became thinned, reduced in size, and translucent, after which the cartilage fragments were collected by centrifugation (200 × g, 5 min) at room temperature. This was followed by overnight digestion (8 h, 37 °C) with 0.2% type II collagenase under continuous gentle agitation. The digestion process was monitored until all macroscopically visible cartilage fragments were completely dissolved. The released cells were collected by centrifugation (200 × g, 5 min) at room temperature, resuspended in complete DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin, and seeded into 75 cm² culture flasks. Cells were maintained at 37 °C in a humidified atmosphere containing 5% CO₂. Adherent chondrocytes were detached using 0.25% trypsin-EDTA, and the cells were subcultured at a 1:3 ratio when the adherent cells reached 80% confluence of the culture flask surface.

2. Preparation of Fructus Xanthii extract

Dried ripe fruits of Xanthium strumarium L. (Fructus Xanthii) were collected from naturally occurring populations in North China and authenticated by a senior pharmacognosist. The specimen was deposited in the herbarium of Shandong First Medical University. 20 g of finely powdered Fructus Xanthii were placed into an extraction thimble, and exhaustive Soxhlet extraction was performed using 95% ethanol (approximately 250 mL) until seven to eight siphon cycles were completed. The ethanolic extract was filtered, and the filtrate was concentrated to dryness under reduced pressure at 45 °C using a rotary evaporator. The resulting residue was resuspended in 60 mL of double-distilled water and partitioned successively with 60 mL of ethyl acetate in a separatory funnel. The mixture was shaken vigorously and allowed to undergo phase separation at 4 °C for 24 h. The aqueous layer was then collected, lyophilized, and stored at −80 °C until further use.

3. Standardized workflow for non-targeted LC-MS/MS metabolomics

1. UHPLC-HRMS/MS-based Non-targeted metabolomics analysis
   Non-targeted metabolomic profiling was performed on an ultra-high-performance liquid chromatography-high-resolution tandem mass spectrometry (UHPLC-HRMS/MS) platform. Metabolites were rapidly quenched and extracted using a chilled methanol–water system. Chromatographic separation was achieved on a UHPLC system equipped with a C18/T3 chromatographic column (100 × 2.1 mm, 1.8 µm) maintained at 40 °C. A binary gradient (0–28 min) consisting of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B) was applied at a flow rate of 0.3 mL/min. Mass spectrometric detection was performed on a hybrid quadrupole-Orbitrap instrument operated in polarity-switching full MS/data-dependent MS2 (Full MS/dd-MS2) mode, with a full scan range of m/z 70–1,050 and a resolution of 120,000 (FWHM at m/z 200). Higher-energy collisional dissociation (HCD) fragmentation was carried out at 20, 40, and 60 eV, with dynamic exclusion set to 10 s.
2. Raw data were converted to mzXML format using ProteoWizard. Peak picking, retention-time alignment, and de-isotoping were performed on XCMS Online using HRMS-optimized parameters (ppm = 5, bw = 5, snthresh = 6). Features containing more than 50% missing values across samples were excluded, and the remaining peak areas were normalized to the total ion current (TIC). Metabolites were annotated by matching accurate mass, MS2 spectra, and retention time against an in-house reference library using a mass error tolerance of ≤ 5 ppm and a dot-product score threshold of ≥80.
   NOTE: Strict standard operating procedures were followed throughout the workflow. Samples were snap-frozen in liquid nitrogen (−80 °C) immediately after collection, transported using a dry-ice chain, and processed using randomized extraction, injection, and batching procedures. Pooled quality control (QC) samples were inserted after every 10 study samples to monitor instrumental drift.

4. Network pharmacology analysis

1. Network pharmacology-based screening of active compounds and target prediction
   Active ingredients of Fructus Xanthii were retrieved from the TCMID database (https://www.tcmsp-e.com/tcmspsearch.php) and the HERB database (http://herb.ac.cn/). The retrieved compounds were filtered according to the recommended ADME thresholds of oral bioavailability ≥30% and drug-likeness ≥0.18. Predicted human targets of the qualified compounds were obtained from the two databases. The screening results were collected and collated, resulting in the identification of 15 active compounds and 151 unique protein targets associated with Fructus Xanthii.
2. Integration of transcriptomic data and identification of OA-associated core targets
   Differentially expressed genes (DEGs) in cartilage tissues between normal and osteoarthritis (OA) samples were extracted from the GEO database datasets GSE51588 (n = 5 normal, 6 OA) and GSE169077 (n = 10 normal, 40 OA) using the GEO2R web tool (https://www.ncbi.nlm.nih.gov/geo/geo2r). OA-upregulated genes were defined using the criteria of log2 fold-change >0.5 and adjusted p-value <0.05. The intersection between OA-upregulated genes and the 151 Fructus Xanthii-related protein targets was visualized using the R package ggvenn. Protein–protein interaction (PPI) networks of the intersection targets were constructed using the STRING database (https://string-db.org/). The PPI network data were imported into Cytoscape, and the top 10 hub genes were screened for subsequent functional enrichment analysis using the CytoHubba plug-in with the MCC algorithm.

5. Determination of Maximum Tolerated Concentration (MTC) in zebrafish

Transgenic zebrafish larvae (Tg col2a1a:EGFP) at 2 days post-fertilization (2 dpf) were randomly distributed into 6-well plates at a density of 30 larvae per well. Each well was supplemented with 3 mL of exposure medium containing serially diluted Fructus Xanthii extract. Moribund or deceased larvae were removed daily. The larvae were continuously exposed at 28 °C for 72 h, after which the maximum tolerated concentration (MTC) was determined as the highest concentration at which larval mortality did not exceed 10%.

6. Establishment of the zebrafish bone-injury model

Tg col2a1a:EGFP zebrafish larvae at 2 days post-fertilization (2 dpf) were assigned to four groups: control, model, chondroitin sulfate (CS) positive control, and Fructus Xanthii extract treatment. All groups except the control group were exposed to Staphylococcus aureus (OD600 = 0.8, 1 × 10⁸ CFU/mL) for 24 h to induce bone injury. Successful model establishment was confirmed by verifying a reduction of at least 50% in cartilage-specific green fluorescence intensity in the craniofacial region relative to the control group under an epifluorescence microscope.

7. Quantitative assessment of cartilage-specific fluorescence in zebrafish larvae

The larvae were exposed continuously to graded concentrations of Fructus Xanthii extract or 1,000 µg/mL sodium chondroitin sulfate A (positive control) at 28 °C for 72 h. Subsequently, 10 larvae were randomly selected from each 6-well replicate (n = 30 per group). The selected larvae were anesthetized using 0.016% tricaine and positioned laterally on 1% low-melting agarose pads. Fluorescence images were acquired at 2× magnification (excitation 488 nm, emission 525/50 nm) using a stereomicroscope equipped with a fluorescence camera. Cartilage-specific EGFP intensity in the craniofacial region, including the ceratohyal and Meckel’s cartilage, was quantified using image analysis software. Background correction was performed on the integrated density values, and the corrected values were normalized to the control group for statistical analysis of chondroprotective efficacy.

8. Induction of OA in C57BL/6J mice

Male C57BL/6J mice (6–8 weeks old, 20–22 g) were housed under specific pathogen-free (SPF) conditions at 22 °C ± 2 °C with a 12 h light/dark cycle. The mice were allowed to acclimatize for one week and were then randomly assigned into three groups (n = 8 per group) using a random number table: (1) sham-operated control, (2) ACLT-induced OA (ACLT), and (3) ACLT plus Fructus Xanthii extract intervention (ACLT + Fructus Xanthii extract). All surgical procedures were performed under sterile conditions in accordance with a previously validated protocol.

The mice were anesthetized with intraperitoneal sodium pentobarbital (50 mg/kg) (following institutionally approved protocols), and a medial parapatellar incision was made to expose the right knee joint capsule. The anterior cruciate ligament was transected under an operating microscope to induce joint instability, after which the joint capsule and skin were closed in layers using 6-0 absorbable sutures. Sham-operated mice underwent identical arthrotomy procedures without ligament transection. Joint stability was evaluated, and successful model establishment was confirmed by performing the drawer test or by assessing the degree of joint injury in histological sections from the model group mice.

Postoperative analgesia with buprenorphine (0.1 mg/kg, subcutaneous injection) was administered for three consecutive days. The treatment agents were administered once daily by gavage. The ACLT + Fructus Xanthii extract group received 100 µL of 1,000 µg/mL Fructus Xanthii extract diluted in normal saline, whereas the other two groups received 100 µL of normal saline as the vehicle control. At 8 weeks post-surgery, the mice were euthanized by CO₂ asphyxiation followed by cervical dislocation, and the knee joints were harvested for histological analyses.

9. Histopathological staining

1. Knee joint tissue histological preparation and staining
   The operated knee joints were excised en bloc at postoperative week 8 and fixed in 4% paraformaldehyde (PFA) at 4 °C for 48 h. The fixed joints were decalcified in 10% ethylenediaminetetraacetic acid (EDTA; pH 7.4) for 14 days, dehydrated through a graded ethanol series, and embedded in paraffin. Serial sagittal sections (5 µm) were prepared from the embedded joints, and the following histological staining procedures were performed.
   1. Hematoxylin and Eosin (H&E) staining
      Hematoxylin and eosin (H&E) staining was performed to assess general tissue architecture, cartilage integrity, and synovial hyperplasia.
   2. Safranin-O/Fast Green staining
      Safranin-O/Fast Green staining was performed to detect proteoglycans within the cartilage tissue. The stained sections were examined using a light microscope equipped with a digital camera, and images were captured at 10× and 20× magnifications.
2. OARSI-based semi-quantitative scoring of cartilage degeneration
   For semi-quantitative assessment of cartilage degeneration, the stained sections were retrospectively re-evaluated and scored according to the Osteoarthritis Research Society International osteoarthritis cartilage histopathology assessment system, which evaluates cartilage damage severity based on grade (0–6; extent of cartilage loss) and stage (0–4; spatial distribution of damage) of articular cartilage degeneration.
   A post-hoc blinding strategy was adopted for all evaluations. All sections were de-identified by removing the original group labels and were re-coded using anonymous serial numbers. Two experienced pathologists independently scored the sections without knowledge of group allocation. Cross-validation was performed between the two sets of scoring results. When the score difference was ≥1, the discrepant sections were jointly re-evaluated by the two pathologists to establish a consensus score. The final consensus scores were used for subsequent statistical analysis to ensure the objectivity, robustness, and reproducibility of the histopathological results.

10. RNA-sequencing (RNA-seq)

Primary human articular chondrocytes (passages 2–3) were seeded into 6-well plates at a density of 5 × 10⁵ cells per well. The cells were serum-starved for 24 h and then allocated into two groups (n = 3 per group): (1) Control group: cells were treated with vehicle only (0.1% DMSO); (2) Fructus Xanthii extract group: cells were treated with 1,000 µg/mL Fructus Xanthii extract dissolved in DMSO.
The cells were treated for 24 h, after which total RNA was extracted using TRIzol Reagent according to the manufacturer’s protocol. RNA integrity (RIN ≥7.0) and concentration were assessed using an RNA analyzer. Strand-specific cDNA libraries were prepared using an RNA sequencing library preparation kit and sequenced on a high-throughput next-generation sequencing (NGS) platform in paired-end 150 bp (PE150) mode. Quality control of the raw FASTQ files was performed using FastQC, followed by adapter and quality trimming using Trimmomatic prior to downstream analysis.

11. RNA-seq data processing

Differentially expressed genes downregulated by Fructus Xanthii extract (vs. control group; p < 0.05 and logFC < −0.5) were extracted from the RNA-seq dataset and intersected with the top 10 hub targets. The resulting core targets associated with Fructus Xanthii extract and osteoarthritis were visualized using the ggvenn package.

12. ELISA

Primary human articular chondrocytes at passages 2–3 were seeded in 6-well plates at a density of 5 × 10⁵ cells per well. Cells were incubated in serum-free medium for 24 h to induce quiescence. The cells were subsequently divided into the following three groups (n = 3 per group):
(1) Blank control: cells were treated with vehicle (0.1% DMSO) only; (2) Positive control: cellular inflammation was induced with IL-1β, followed by treatment with vehicle (0.1% DMSO); (3) Fructus Xanthii extract group: cellular inflammation was induced with IL-1β, followed by treatment with 1000 µg/mL Fructus Xanthii extract dissolved in 0.1% DMSO. Cell culture supernatants were collected after 24 h of incubation and stored at −80 °C until use. Levels of MMP1 and PGR in the culture supernatants were determined using commercially available enzyme-linked immunosorbent assay (ELISA) kits specific for human MMP1 and PGR, respectively. The assays were performed according to the manufacturers’ instructions.

13. Statistical analysis

All statistical analyses were performed using statistical analysis and graphing software, and data were presented as mean ± standard deviation (SD). Data normality was assessed using the Shapiro–Wilk test, and homoscedasticity was evaluated using the Levene test. Inter-group comparisons were performed using one-way or two-way analysis of variance (ANOVA), followed by Tukey’s or Šidák post-hoc tests, as appropriate. Pairwise comparisons were conducted using a two-tailed Student’s t-test. Non-parametric data were analyzed using the Mann–Whitney U test. A two-sided P-value < 0.05 was considered statistically significant.

Results

Quality assessment and metabolic landscape of non-targeted LC-MS/MS data

QC samples demonstrated excellent reproducibility, as indicated by highly overlapping TIC chromatograms in both positive and negative ion modes, with retention time drift remaining below 0.05 min across the entire analytical batch. Raw data underwent XCMS-based preprocessing, which included peak detection, integration, and retention-time alignment. Putative metabolites were annotated by matching accurate ...

Discussion

OA is a prevalent degenerative joint disorder and a leading cause of disability worldwide³⁶. Its pathogenesis is complex and multifactorial, involving biochemical, cellular, and molecular changes across joint tissues. Once considered a simple "wear-and-tear" phenomenon of articular cartilage, OA is now widely recognized as a chronic, whole-joint disease characterized by cartilage degradation, synovial inflammation, osteophyte formation, and subchondral bone remodelling^37<...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
4% PFA	Biosharp, China	BL539A	pH 7.4, sterile
C57BL/6 Mice	Vital River Laboratory Animal Technology Co.,Ltd, China	N/A	Male, 8-10 weeks old, 20-25g
Chondroitin sulfate	MCE, America	HY-B2162	Purity ≥98%
ClusterProfiler (R Package)	Bioconductor	Free access	Integrated in R v4.2.2
Collagenase, Type II, powder	Gibco, America	17101015	≥125 U/mg
Cytoscape	Cytoscape Consortium	Free access (https://cytoscape.org/)	v3.9.1 (with CytoHubba)
Cytoscape	Cytoscape Consortium	Free access	Version 3.9.1
DESeq2 (R Package)	Bioconductor	Free access	Integrated in R v4.2.2
DMEM, high glucose	Gibco, America	11965092	Sterile medium
Ethanol	Sinopharm Chemical Reagent Co.,Ltd, China	10009218	Analytical grade, 95%
Ethyl acetate	Sinopharm Chemical Reagent Co.,Ltd, China	10009418	Analytical grade
Fetal bovine serum (FBS)	Cellmax, China	SA102.02	Heat-inactivated
Fluorescence Microscope	Nikon, Japan	Eclipse Ti2-U	Equipped with GFP filter (488nm excitation) & digital camera
GEO Database	NCBI	https://www.ncbi.nlm.nih.gov/geo/geo2r	GEO2R analysis tool
GraphPad Prism	GraphPad Software, LLC, America	Commercial	Version 9.5.1
Hematoxylin-Eosin(HE) Stain Kit	Solarbio, China	G1120	Ready-to-use kit
HERB Database	Chinese Academy of Sciences	http://herb.ac.cn/	V2.0
Human MMP-1 ELISA Kit	Yamei Biotechnology, China	HJ088	For human, 96-well plate
Human Progesterone Receptor (PR) ELISA Kit	Shanghai Enzyme-Linked Biotechnology Co., Ltd., China	ml05997	For human, 96-well plate
ImageJ	National Institutes of Health (NIH)	Free access	Version 1.53t
Light Microscope	Olympus, Japan	BX53	10×/20× objectives & digital camera
Lyophilizer	Christ, Germany	Alpha 1-4 LDplus	Freeze-drying system
Modified Saffron-O And Fast Green Stain Kit	Solarbio, China	G1371	For bone/cartilage staining
Neutral Balsam	Solarbio, China	G8590	Mounting medium
Penicillin-Streptomycin	Gibco, America	15070063	100×, sterile
Phosphate-buffered saline (PBS)	Sparkjade, China	CR0013-500ML	pH 7.2-7.4, sterile
R software	R Foundation for Statistical Computing	Free access (https://www.r-project.org/)	v4.4.1 (with ggvenn)
Serum-free Cell Freezing Medium	Biosharp, China	BL203B	Sterile
Staphylococcus aureus	American Type Culture Collection (ATCC)	ATCC 25923	Standard strain
STRING Database	STRING Consortium	https://string-db.org/	Protein interaction analysis
TCMID Database	Shaanxi Qinling Qiyao Collaborative Innovation Center	https://www.tcmsp-e.com/tcmspsearch.php	V3.0
Tg(col2a1a:EGFP) Zebrafish	Zebrafish International Resource Center (ZIRC)	N/A	Larvae (2 dpf), AB strain
TRIzol Reagent	Thermo Fisher, America	15596018CN	For total RNA extraction
Trypsin-EDTA (0.25%), phenol red	Gibco, America	25200056	Sterile solution
Ultra-high Performance Liquid Chromatography (UHPLC)	Thermo Fisher, America	Q Exactive Focus	Coupled with HRMS/MS
XCMS-online	Scripps Research Institute	Free access	Web-based platform