Standardized Extract of Flavonoids from Smilax glabra Alleviates Gouty Arthritis by Multi-Target Inhibition of Neutrophil Extracellular Traps via the Raf/ERK and Histone Citrullination Pathways

Introduction

Gouty arthritis (GA), a metabolic disorder driven by monosodium urate (MSU) crystal deposition and subsequent inflammation, has seen a global rise in prevalence due to dietary shifts and metabolic disease epidemics.^1,2 While GA is manageable with diverse agents (eg, colchicine, NSAIDs, glucocorticoids, ACTH, IL-1β antagonists), key limitations persist: contraindications in those with comorbidities (eg, renal/cardiovascular disease), dose-dependent side effects restricting long-term use in susceptible patients, and pervasive issues of cost and access.³ Thus, there remains an unmet need for safer, more accessible, and broadly effective therapies.

Neutrophils are the most abundant immune cells in peripheral blood and play dual roles in host defense and tissue damage during GA.^4,5 Upon MSU crystal stimulation, a key pathogenic event is the release of neutrophil extracellular traps (NETs). These web-like structures, composed of DNA, histones, myeloperoxidase (MPO), and neutrophil elastase (NE), amplify inflammation through cytokine storms and collateral tissue injury.^6–9 Critically, MSU crystals activate the Raf/ERK signaling pathway, which triggers reactive oxygen species (ROS)-dependent NETosis—a process characterized by histone citrullination (particularly of histone H3) and chromatin decondensation.^10–13 While inhibitors targeting NETosis (eg, Cl-amidine, DNase I)^14,15 or antioxidants (eg, quercetin)^16,17 have shown efficacy in preclinical models, no existing drugs directly modulate Raf/ERK signaling in GA, leaving a critical therapeutic gap. To address this, our previous work successfully established a robust in vitro NETs formation model, validated for screening NETosis-targeting agents.¹³ This model provides a foundational tool for the current study to explore novel multi-target interventions.

Flavonoids, a class of polyphenolic compounds, are promising candidates for GA treatment due to their dual capacity to inhibit uric acid production and suppress inflammation.^18,19 Smilax glabra Roxb. (tufuling), a traditional Chinese medicine rich in flavonoids, has been historically used to treat GA and immune disorders.^20,21 Our previous studies identified six bioactive flavonoids in Smilax glabra Roxb. (neoastilbin, astilbin, neoisoastilbin, isoastilbin, (-)-epicatechin, and engeletin), collectively termed total standardized extract of flavonoids from Smilax glabra (SFSG), and established its anti-hyperuricemic and anti-inflammatory properties.^22–26 Notably, while studies on single constituents such as astilbin have revealed its ability to inhibit NETs via the P2Y6 receptor pathway²⁷ the potential synergistic effect of SFSG — which comprises multiple bioactive components — on NETosis through the distinct Raf/ERK signaling axis remains unexplored. This represents a critical research gap, as the therapeutic superiority of botanical extracts often lies in their multi-target, synergistic actions rather than in isolated compounds.

Therefore, we hypothesize that SFSG alleviates GA by multi-target inhibition of NETosis, simultaneously targeting both the Raf/ERK signaling pathway and histone citrullination. Using an MSU-induced GA mouse model and in vitro NETs assays, we demonstrate that SFSG: (1) suppresses Raf/ERK phosphorylation and downstream ROS production, (2) reduces histone H3 citrullination (CitH3) and NETs formation, and (3) attenuates neutrophil infiltration and cytokine release. These findings not only bridge the traditional use of Smilax glabra Roxb. with modern mechanistic insights but also position SFSG as a novel multi-target agent for GA therapy.

Materials and Methods

Drugs and Reagents

The rhizome of Smilax glabra Roxb. (Lot: 130101) was sourced from a traditional Chinese medicine factory (Hangzhou, China), and authenticated by associate Prof. Kongrong Chen from Zhejiang Chinese Medical University. Colchicine (Lot: 16IS) was obtained from Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China) for cell experiment. Colchicine tablets (Lot: 21XP) were obtained from KPC Pharmaceuticals, Inc. (Kunming, China) for animal experiments. PolymorphPrep^TM (Lot: 00121) was purchased from Axis-Shield (Serumwerk Bernburg AG, Germany). Phorbol 12-myristate 13-acetate (PMA, Lot: SLBX8889), monosodium urate (MSU, Lot: BC8R7559) and 4′,6-dia-midino-2-phenylindole (DAPI, Lot: 059M4025V) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Cl-amidine (Lot: 146055) was from MedChemExpress (Monmouth Junction, NJ, USA). Sytox Green (Lot: 2284662) was from Thermo Fisher Scientific (Waltham, MA, USA). Hoechst 33342 (Lot: 60321210713), Triton X-100, ROS assay kit (Lot: 101121220118) and lactate dehydrogenase (LDH) assay kit (Lot: 42821211105) were all from Beyotime Institute of Biotechnology (Shanghai, China). ELISA kits for human IL-8 (Lot: 1341762820), human TNF-α (Lot: 2391772820), human IL-6 (Lot: 1311771820), human IL-1β (Lot: 1141779820), mouse TNF-α (Lot: 2411832406), mouse IL-6 (Lot: 1321837406) and mouse IL-1β (Lot: 1161825406) were from Boster Biological Technology Co., Ltd. (Wuhan, China). ELISA kit for mouse IL-8 (Lot: M220507-104a) was from NeoBioscience Technology Co., Ltd. (Shenzhen, China).

Antibodies for CD16 (Lot: 1018518) and CD11b (Lot: 9301774) was from BD Bioscience (San Jose, CA, USA). Antibody for CD66b (Lot: B336289) was from Biolegend (San Diego, CA, USA). Antibodies for Histone H3 (citrulline R2+R8+R17) (Lot: GR3402072-1), donkey polyclonal secondary antibody to rabbit IgG H&L (Alexa Fluor^® 488) (Lot: GR3377618-2), donkey polyclonal secondary antibody to goat IgG H&L (Alexa Fluor^® 555) (Lot: GR3374229-3), donkey polyclonal secondary antibody to mouse IgG H&L (Alexa Fluor^® 647) pre-adsorbed (Lot: GR3369249-3) were all from Abcam (Cambridge, UK). MPO antibody (Lot: YBZ0420072) was from R&D Systems (Minneapolis, MN, USA). Neutrophil Elastase (NE) antibody (Lot: D0221) was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Antibodies for Raf (Lot: 1), p-Raf (Lot: 10), ERK1/2 (Lot: 28), p-ERK1/2 (Lot: 24) and Ly6G (Lot: 4) were from Cell Signaling Technology (Danvers, MA, USA). Anti-GAPDH (Lot: 10020246) was from Proteintech (Rosemount, IL, USA).

Human Neutrophils Isolation and Culture

Peripheral blood was collected from healthy volunteers (n = 12, aged 20–30 years). This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Zhejiang Chinese Medical University (Approval number: 20220111–4). Informed consent was obtained from all participants prior to blood donation.

Neutrophils were isolated from the anticoagulant blood using PolymorphPrep™.¹³ Briefly, the neutrophil layer (second white cell layer) was harvested and red blood cells were completely removed. An appropriate amount of RPMI-1640 solution was added to fully mix the cells, and then these cells were cultured at 37°C and 5% CO₂ in an incubator and used for NETosis studies.

Mouse Bone Marrow Neutrophils (BMNs) Isolation and Purity Analysis

BMNs were isolated using a previously described protocol²⁸ that results in the acquisition of neutrophils of high purity. Briefly, the femurs and tibiae of adult male C57BL/6 mice were collected and freed of soft tissues attachments. Then, RPMI-1640 solution supplemented with 1% antimycotic was forced through the bone with a 1 mL syringe. Red blood cells from the bone marrow were removed and the suspension was filtered on 70 μm mesh filters. BMNs were isolated according to the protocol of “Neutrophil isolation kit, mouse” by Miltenyi (NO: 130–097-658). Subsequently, BMNs were identified by APC/Cyanine7 anti-mouse Ly6G antibody (Biolegend) and BB515 anti-mouse CD11b antibody (BD Pharmingen), and the purity was >90%.

HL-60 Cells Treatments

HL-60 cells were provided from ATCC (Lot: 36686) and were induced to the differentiated HL-60 (dHL-60) cells after being treated with 1.3% DMSO for 5 days. dHL-60 cells were cultured with RPMI-1640 medium supplemented with 10% FBS and 1% antimycotic.

Cell Grouping and Treatments

To model the primed state of neutrophils in an inflammatory environment, NETosis was induced using a combination of MSU and PMA, following our established protocol¹³ Human neutrophils were divided into six groups: Control group, MSU + PMA group (stimulated with 500 μM MSU + 50 nM PMA for 4 h), MSU + PMA + colchicine group (50 nM), MSU + PMA + SFSG group (50, 100, 200 μg/mL). Except Control group, groups were pretreated with colchicine and SFSG for 1 h before stimulation.

Neutrophils Viability Assay

Human neutrophils cell suspension (5×10⁵ cells/mL) was seeded into a 96-well plate (100 µL/well) and then the cells were treated with SFSG of various concentrations (0, 10, 20, 50, 100, 200, 400, 800, 1000 μg/mL) for 4 h to evaluate the cytotoxicity of SFSG. Subsequently, CCK-8 reagent was introduced into each well, followed by a 3-hour incubation period. The optical density readings at 450 nm were then recorded using a microplate reader (BioTek, USA, Synergy H1).

Immunofluorescence (IF) Staining

The purified neutrophils (2×10⁵ cells/well) were seeded in 12-well plates containing poly-l-lysine-coated coverslips and were allowed to stand at 37°C under 5% CO₂ for 0.5 h to adhere to the slide. After different treatments, neutrophils were fixed with pre-cooled 4% paraformaldehyde for 20 min. To visualize NETs, they were stained with Sytox Green (a dye specifically designed to label NETs) and Hoechst 33342 (a nuclear dye) respectively, and imaged using a fluorescence microscope (AXIO SCOPE.A1) with DAPI and GFP filter sets.

To observe the distribution of major components of NETs, the fixed samples were punctured with 0.3% Triton X-100 and sealed with 5% BSA. Incubation overnight using rabbit Histone H3 (citrulline R2 + R8 + R17) antibody (1:400), goat MPO antibody (1:200), mouse NE antibody (1:200), respectively. DAPI stains the nucleus. Antifade mounting medium was utilized to affix the coverslips onto glass microscope slides. Images were acquired using a laser scanning confocal microscope (Zeiss LSM 880) with a 20× objective. Four laser lines (405, 488, 543, and 633 nm) were used to excite DAPI, Alexa Fluor 488, Alexa Fluor 555, and Alexa Fluor 647, respectively, corresponding to the applied fluorophores.

Detection of LDH Release

5×10⁴ cells/well neutrophils were inoculated into 96-well plates, and cell-free culture medium wells and a cell-containing sample maximum enzyme activity Control well were set. At the same time, the drug Control group was set to get out the influence caused by the color of the drug itself. Follow the instructions for detailed operation. Using dual wavelengths (490/600 nm) to measure and calculate LDH release rates (cell death rates).

Transmission Electron Microscopy (TEM)

Differently treated neutrophils were fixed with 2.5% glutaraldehyde for 2 h at room temperature. The following procedures were performed in sequence: fixation with a mixture of 1% osmic acid + 1.5% K₃[Fe(CN)₆] and 1% osmic acid, staining with 2% uranyl acetate solution, ethanol dehydration, embedding, polymerization, and sectionalization. Then, observed the microstructure of NETs by TEM (Hitachi Limited, Japan, Hitachi H - 7650).

GA Mouse Model Establishment and Experimental Design

Male C57BL/6J mice (6–8 weeks, 18–20 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (SCXK 2017–0011), and housed under standard conditions (23 ± 1°C temperature, 12 h light/dark cycle). They had free access to food and water, unless otherwise specified. All animal experiments were approved by the Animal Care and Use Committee of Zhejiang Chinese Medical University (Hangzhou, China. Permission number: SYXK 2021–0012), and all animal procedures (Ethical Approval Number: 20220214–16) adhered to the guidelines for the Care and Use of Laboratory Animals (National Institutes of Health, Bethesda, MD, United States).

Male C57BL/6J mice (n = 120) were randomly divided into eight groups (n=15 per group) using a random number table, with group allocation balanced for body weight. The groups were as follows: Control group, SFSG group (400 mg/kg, oral gavage, 7 days), MSU group (MSU-induced GA), MSU + Cl-amidine group (10 mg/kg, intraperitoneal injection 1 h pre-MSU), MSU + colchicine group (1 mg/kg, oral gavage, 7 days), MSU + SFSG group (100 mg/kg, 200 mg/kg, 400 mg/kg, oral gavage, 7 days). On day 6, 1.25 mg MSU in 25 μL sterile PBS was injected into the ankle joint cavity to induce acute GA. The mice were euthanized by CO₂ inhalation after gavage on the 7th day. Some ankle joints cavities were exposed and rinsed with 200 μL of PBS for NETs quantification. The excised ankle joint tissue was promptly divided, with a portion being fixed for histopathological examination and the remainder being preserved in the freezer for subsequent analyses.

Neutrophils Depletion

18 male C57BL/6J mice were randomly allocated into three groups (6 mice per group): Control group (Control), acute GA group (MSU), neutrophil-depleted group (Ly6G + MSU). Neutrophil depletion was achieved by daily intraperitoneal administration of 100 μg anti-Ly6G mAb (1A8, Bio X Cell) for 7 days. On day 6, MSU suspension (50 mg/mL, 0.03 mL) was intra-articularly injected into the ankle joint of the Ly6G + MSU and MSU groups. To sustain depletion, the Ly6G + MSU group received additional intraperitoneal injections of 50 μg mAb 1A8 at 4 and 10 h post-MSU challenge. Depletion efficacy was validated by flow cytometry (CD11b⁺Ly6G⁺ cells in peripheral blood on days 4 and 7). Post-sacrifice, ankle tissues were processed for histopathology and molecular analyses.

Evaluation of Ankle Joint Edema

Relevant operation methods and detection methods are referred to Wu et al.²⁴ In short, the toe volume of mice was measured before and at different time points after MSU injection, and swelling index was calculated.

Histology Analysis

As previously described,²⁴ paraffin sections containing joint tissue were stained with hematoxylin and eosin (HE). For immunohistochemistry (IHC), tissue sections were incubated overnight at 4°C in a humidified environment with Ly6G or MPO. Primary labeling was identified through the application of a biotinylated horse anti-rabbit IgG secondary antibody, followed by staining with DAB and hematoxylin. Images were acquired using a digital pathological section scanner (Hamamatsu, Japan). All subsequent histopathological evaluations and analyses of IHC staining were conducted by researchers blinded to the experimental groups.

Assessment of MPO Activity

Joint tissue lapping fluid was taken and tested MPO activity. The MPO activity in the supernatant was measured by MPO determination kit. The experimental procedure was in accordance with the manufacturer’s instructions.

Quantification of NETs

NETs were quantified using Sytox Green.¹³ 5×10⁴ cells/well neutrophils were seeded in 96-well transparent black plates (Corning). After different treatments, the cells samples and the collected synovial fluid were immediately incubated with Sytox Green (final concentration 1 μM) at 37°C, 5% CO₂ for 10 min. The fluorescence intensity was detected at Ex/Em: 502/525 nm using a fluorescence microplate reader (BioTek, USA, Synergy H1).

Cytokine Analysis

As previously described,²⁴ the cell supernatants were collected and the tissue samples were prepared. The contents of inflammatory factors (IL-1β, IL-6, IL-8 and TNF-α) were determined according to the instructions of ELISA kits.

Flow Cytometry Analysis

Human peripheral blood neutrophils were identified by CD16 and CD66b.²⁹,100 μL cell suspension (1 × 10⁶ cells) was incubated with 5 μL PE anti-human CD16 and/or FITC anti-human CD66b or PBS in the dark for 30 min on ice, and then washed twice with PBS. Neutrophils were gated based on forward and side scatter (FSC/SSC) to exclude debris and aggregates, followed by fluorescence analysis for purity, as per our standard protocol.¹³ Neutrophils with a purity greater than 93% were analyzed by Beckman flow cytometry (Beckman, USA, Cytoflex) for subsequent experiments. Neutrophils with a purity greater than 93% were analyzed by Beckman flow cytometry (Beckman, USA, CytoFlex) for subsequent experiments.

DCFH-DA (1:1000) prepared with serum-free RPIM-1640 medium was added to all cells and incubated at 37°C and 5% CO₂ for 20 min, and then washed twice with RPIM-1640 medium. The cells were then grouped and treated, and after a given time, the cells were collected and then resuspended in PBS. The intracellular ROS fluorescence intensity was measured by Beckman flow cytometry.

Submaxillary blood from mice was collected, and after the removal of red blood cells, APC/Cyanine7 anti-mouse Ly6G antibody (2:100) (Biolegend) and BB515 anti-mouse CD11b antibody (2:100) (BD Pharmingen) were added. The mixture was incubated at 25°C for 30min, and then the cells were re-suspended in PBS for flow cytometry.

Real-Time Quantitative PCR (qPCR)

Mice ankle joint tissues were used to detect PADI4 and NOX2 gene expression. RNA was extracted from the joint tissue using Qiagen kit (Qiagen, Hilden, Germany). Reverse transcription into cDNA was carried out with the Evo M-MLV reverse transcription premix kit. Then, the production from reverse transcription was amplified with SYBR Green Pro Taq HS premixed qPCR kit, and the PCR products were detected using ABI 7500 Real-Time PCR System. Relative gene expression was determined by the 2^−ΔΔCt. The murine primers sequences were listed Table 1. The primers sequences for mouse GAPDH and PADI4 are referenced from Schneider AH.³⁰

Table 1 Primer Sequences Used for qPCR in Mice

Western Blotting (WB)

Based on the previous description,²⁴ briefly: neutrophils and tissue samples are gathered, lysis is carried out to obtain protein samples, and then electrophoresis, membrane transfer, sealing, and antibody incubation are conducted. These target proteins were visualized by enhanced chemiluminescence system (Guangzhou Boluteng, China). The optical density analysis of the protein bands was performed with Image J analysis program.

Statistical Analysis

Statistical analyses were performed using GraphPad Prism software (version 8.0). For comparisons among three or more groups, one-way ANOVA or two-way ANOVA was first applied to determine overall significance, followed by an appropriate post-hoc test if the ANOVA result was significant. Data are presented as mean ± SEM. P < 0.05 was considered statistically significant.

Results

Neutrophil Depletion Attenuates MSU-Induced Joint Swelling and Inflammation in GA Mice

To validate the pivotal role of neutrophils in GA pathogenesis, we selectively depleted neutrophils in mice via intraperitoneal injection of the anti-Ly6G mAb 1A8 antibody. Flow cytometry confirmed >99% reduction in circulating neutrophils (CD11b⁺Ly6G⁺ cells) after 4 days of treatment (Figure 1A and C). Following MSU injection, compared to the MSU group, neutrophil-depleted mice (Ly6G + MSU group) exhibited lower neutrophil counts (Figure 1B and D) and significantly attenuated ankle swelling (Figure 1E and F). Histopathology showed diminished leukocyte infiltration and restored joint architecture in neutrophil-depleted mice, accompanied by decreased pro-inflammatory cytokines (IL-1β, IL-8) in joint tissues (Figure 1G–J). These data confirm neutrophils as central drivers of GA progression.

Figure 1 Neutrophils depletion alleviated joint swelling and inflammation in MSU - induced GA mice. The mice were intraperitoneally injected with mAb 1A8 antibody daily for 7 consecutive days. On day 6, 3 h after the final intraperitoneal injection of mAb 1A8, a MSU suspension was injected into the ankle joint. (A–D) Number of neutrophil cells (labeled with Ly6G+ and CD11b+ antibodies) in the blood of mice before and after MSU injection at the ankle analyzed by flow cytometry. n=3. (E) The tendency of the mice ankle swelling degree over time. (F) Representative images of ankle joint swelling in mice 24 h after MSU injection. (G and H) The concentrations of IL-1β and IL-8 were detected by ELISA, n=4. (I) Representative morphological photographs of the mice ankle joint; HE images with × 100 and × 400. Scale bar, 100 μm and 20 μm. The red arrows represent leukocyte infiltration. (J) A representative IHC image of the mice ankle joint in each group is shown (× 200, scale bar: 50 μm). The red arrows represent neutrophil aggregation. The data were presented as mean ± SEM with *P < 0.05, **P < 0.01 vs Control group, #P < 0.05, ##P < 0.01 vs MSU group. Statistical analyses were performed with one-way ANOVA, two-way ANOVA or post-hoc Tukey’s tests.

SFSG Inhibits MSU and PMA-Induced NETs Formation in vitro

Using a standardized extract of SFSG, prepared as previously described,¹³ we investigated its effects on NETs formation in a model induced by MSU and PMA. Neutrophils isolated from human peripheral blood using PolymorphPrep™ showed >94% purity (Figure 2A). Cell viability assessed by CCK8 assay indicated no cytotoxicity at SFSG concentrations up to 200 μg/mL, whereas 400 μg/mL significantly suppressed viability by nearly 50% (Figure 2B). Consequently, 50, 100, and 200 μg/mL were selected as low, medium, and high experimental doses.

Figure 2 SFSG inhibits MSU- and PMA-induced NETs formation in vitro. (A) Neutrophils were isolated from human peripheral blood using Polymorphprep™ following the manufacturer’s protocol (red arrows indicate neutrophil layer). Purity was assessed by CD16/CD66b staining via flow cytometry. (B) CCK8 reagent detected the viability of cells pretreated with SFSG in varying doses (10 to 1000 μg/mL), n=3. (C) Representative fluorescent images of NETs formation in 500 μM MSU + 50 nM PMA-induced human neutrophils, pretreated with the indicated concentrations of SFSG (50, 100, 200 μg/mL) or colchicine (50 nM). Cells were stained with Hoechst 33342 (blue, nuclei) and Sytox Green (green, NETs/DNA). Scale bars, 50 or 20 μm. (D) Quantification of NETs release, as measured by Sytox Green fluorescence intensity (Ex/Em 502/525 nm). n=4. (E) The femurs and tibiae of adult male C57BL/6 mice were collected and freed of soft tissues attachments. Then, BMNs were isolated according to the protocol of “Neutrophil isolation kit, mouse” by Miltenyi. BMNs were identified by Ly6G antibody and CD11b antibody, and the purity was >90%. IF staining (F) and quantitative analysis (G) of NETs in MSU + PMA induced BMNs with pretreatment of SFSG. n=5-6. The data were presented as mean ± SEM with **P < 0.01 vs Control group, #P < 0.05, ##P < 0.01 vs MSU + PMA group. Statistical analyses were performed with one-way ANOVA, post-hoc Tukey’s tests or Dunnett’s tests.

NETs formation, visualized by Sytox Green staining, was markedly increased upon MSU and PMA stimulation, as evidenced by extensive neutrophil aggregation and net-like structures. SFSG treatment appeared to attenuate this effect, with the reduction in net-like structures being more pronounced at higher doses (Figure 2C). Quantitative analysis showed that SFSG at 50, 100, and 200 µg/mL inhibited NETs-associated fluorescence intensity by approximately 29%, 52%, and 64% respectively (Figure 2D). Colchicine exhibited a comparable inhibitory effect (~41% inhibition). These results were corroborated in BMNs, where SFSG also suppressed NETs formation (Figure 2E–G).

SFSG Alleviates Neutrophil Damage and Cytokine Release

To explore the mechanisms underlying SFSG’s inhibitory effect on NETs formation, we assessed the localization of CitH3, MPO, and NE using confocal microscopy. In unstimulated neutrophils, CitH3 was nuclear, and MPO/NE were cytoplasmic. Upon MSU and PMA stimulation, all three components were released extracellularly, forming NETs structures. SFSG treatment attenuated the extracellular distribution of CitH3, MPO, and NE, and this effect increased with higher doses (Figure 3A).

Figure 3 SFSG alleviates neutrophil damage and cytokine release. The purified neutrophils were pretreated with colchicine (50 nM) and SFSG (50, 100, 200 μg/mL) for 1 h before stimulation with 500 μM MSU + 50 nM PMA. (A) IF was used to detect pretreatment of SFSG on the expressions of CitH3 (green), MPO (red), NE (pink) and DAPI (blue) stimulated by MSU + PMA. Scale bars, 20 μm. (B) TEM was used to observe pretreatment of SFSG on the internal structure of NETs stimulated by MSU + PMA. Scale bars, 1 μm. (C) The LDH release (cell death rate) was detected pretreatment of SFSG stimulated by MSU + PMA, n=6. (D–F) The concentration of IL-8, TNF-α and IL-1β in the supernatant of MSU + PMA-induced human neutrophils were detected by ELISA, n=3. The data were presented as mean ± SEM with **P < 0.01 vs Control group, #P < 0.05, ##P < 0.01 vs MSU + PMA group. Statistical analyses were performed with one-way ANOVA, post-hoc Tukey’s tests.

Transmission electron microscopy further revealed that MSU and PMA induced nuclear membrane rupture and cytoplasmic disintegration. SFSG partially preserved cellular ultrastructure, with observable nuclear integrity and reduced organelle damage (Figure 3B). LDH release assays demonstrated that SFSG reduced membrane damage and cell death (by approximately 82% at 200 µg/mL, Figure 3C), indicating cytoprotective effects. ELISA analysis showed SFSG significantly reduced the elevated levels of pro-inflammatory cytokines in stimulated neutrophils, suppressing IL-1β, TNF-α, and IL-8 by approximately 68%, 61%, and 53% at 200 µg/mL, respectively (Figure 3D–F), confirming its potent anti-inflammatory activity.

SFSG Suppresses NETs Formation by Inhibiting ROS Production and MAPK Signaling

ROS play a pivotal role in NETosis. Flow cytometry revealed that SFSG significantly reduced intracellular ROS levels induced by MSU and PMA, with the effect becoming more pronounced at higher concentrations (Figure 4A and B). WB analysis showed SFSG reduced CitH3 protein levels (Figure 4C and D) and downregulated phosphorylated Raf and ERK (Figure 4E–G), indicating inhibition of the Raf/ERK signaling cascade and NETosis.

Figure 4 SFSG suppresses NETs formation by inhibiting ROS production and MAPK signaling. (A) Intracellular ROS was detected by flow cytometry. (B) Quantitative analysis of intracellular ROS, n=3. (C–J) Representative WB and densitometry analyses were shown of protein expression of p-Raf, Raf, p-ERK, ERK and CitH3. GAPDH was used as a loading control. (H) HL-60 cells were induced to the differentiated HL-60 (dHL-60) cells after being treated with 1.3% DMSO for 5 days. (I–L) SFSG for 1 h before stimulation with 500 μM MSU + 400 nM PMA. Representative WB and densitometry analyses were shown of protein expression of p-Raf, Raf, p-ERK, ERK and CitH3. n=3. GAPDH or β-actin was used as a loading control. The data were presented as mean ± SEM with *P < 0.05, **P < 0.01 vs Control group, #P < 0.05, ##P < 0.01 vs MSU + PMA group. Statistical analyses were performed with one-way ANOVA, post-hoc Tukey’s tests.

In dHL-60 cells, SFSG also suppressed CitH3 expression and inhibited the Raf/ERK pathway, further confirming its role in modulating NETosis via the MAPK pathway and CitH3 inhibition (Figure 4H–J).

SFSG Reduces Joint Swelling and Inflammation in a Murine GA Model

Based on the in vitro finding that SFSG blocks NETosis by inhibiting Raf/ERK and CitH3, we therefore hypothesized that it would alleviate GA in vivo. To test this, we employed an MSU-induced mouse model (Figure 5A). SFSG treatment reduced MSU-induced ankle swelling (Table 2). The 400 mg/kg dose showed efficacy comparable to colchicine, with a 24-hour swelling index of 19.75 ± 5.64% versus 19.75 ± 14.46% in the colchicine group (Figure 5B and C). Similar anti-inflammatory effects were observed in the Cl-amidine (PADI4 inhibitor) group.

Table 2 Measurement of Swelling Degree of C57BL/6J Mice Ankle Joint (%) (, n = 15)

Figure 5 SFSG reduces joint swelling and inflammation in a murine GA model. (A) The GA mice were pretreated with SFSG (100, 200, 400 mg/kg, i.g.) and colchicine (1 mg/kg, i.g.) for 7 days, and Cl-amidine (10 mg/kg, i.p.) was 1 h before MSU injection on the 6th day. (B) Representative images of ankle joint swelling in mice 24 h after MSU injection. (C) The tendency of the swelling index of ankle joints in GA mice over time. (D) Representative morphological photographs of the mice ankle joint; HE images with × 100 and × 400. Scale bar, 100 μm and 20 μm. The red arrows represent leukocyte infiltration. (E–H) The levels of inflammatory factors (TNF-α, IL-1β, IL-6) and chemokine (IL-8) in the mice ankle joint tissues. n=5-6. The data were presented as mean ± SEM with **P < 0.01 vs Control group, ##P < 0.01 vs MSU group. Statistical analyses were performed with one-way ANOVA, post-hoc Tukey’s tests.

Histological examination (HE staining) showed severe inflammatory infiltration and joint damage in MSU-injected mice, while SFSG treatment preserved tissue integrity, particularly in the high-dose group. Cl-amidine and colchicine also reduced inflammatory infiltration (Figure 5D). ELISA analysis demonstrated elevated levels of IL-1β, IL-6, TNF-α, and IL-8 in MSU-treated joints, all significantly decreased upon SFSG treatment, highlighting its potent anti-inflammatory effects (Figure 5E–H).

SFSG Inhibits Joint Neutrophil Infiltration and NETs Formation via the Raf/ERK Pathway

MPO is normally located within the cytoplasm of neutrophils. During NETosis, it binds to chromatin and facilitates chromatin condensation. Therefore, detecting MPO in joint tissue can indicate both the presence of neutrophils and their potential activation toward NETosis. IHC (MPO) revealed pronounced neutrophil infiltration in MSU-injected joints, which was significantly attenuated by SFSG (Figure 6A). Sytox Green staining confirmed reduced intra-articular NETs after SFSG treatment (Figure 6B). Quantitative analysis of MPO levels supported this observation (Figure 6C).

Figure 6 SFSG inhibits joint neutrophil infiltration and NETs formation via the Raf/ERK pathway. (A) IHC image of MPO staining (× 400, scale bar: 100 μm). The red arrows represent MPO. (B) Quantification of NETs in GA mice synovial fluid with Sytox Green. n=5. (C) MPO content in GA mice joint tissues. n=6. (D) and (E) NOX2 and PADI4 mRNA expression. n=3. (F) WB analysis of Raf and ERK expression in GA mice joints tissues. (G) Quantification of p-Raf to Raf ratios. n=3. (H) Quantification of p-ERK to ERK ratios. n=3. (I) WB detection of CitH3 expression in GA mice joints tissues. (J) Quantification of CitH3 to GAPDH ratios. n=3.The data were presented as mean ± SEM with **P < 0.01 vs Control group, #P < 0.05, ##P < 0.01 vs MSU group. Statistical analyses were performed with one-way ANOVA, post-hoc Tukey’s tests.

Figure 7 Schematic illustration of the mechanism by which SFSG improves GA by inhibition of NETs formation via Raf/ERK pathway.

RT-PCR analysis showed that MSU markedly increased mRNA levels of PADI4 and NOX2 in joint tissue (by 7.46-fold and 35.71-fold, respectively). These increases were dose-dependently suppressed by SFSG, with the SFSG-H decreasing expression to near baseline levels (only 0.46-fold and 0.85-fold over control, respectively) (Figure 6D and E). Protein levels of p-Raf, p-ERK, and CitH3 were also upregulated by MSU, and SFSG treatment was associated with reduced levels of these phosphorylated and citrullinated proteins (Figure 6F–J). These results indicate that SFSG reduces NETs formation and inflammation by modulating the Raf/ERK signaling pathway and inhibiting histone H3 citrullination.

Discussion

The pathophysiology of GA involves a complex interplay between innate immune responses and inflammatory cascades, with neutrophils emerging as central orchestrators of joint inflammation. Synovial fluid neutrophils from GA patients exhibit heightened metabolic activity and a propensity for spontaneous NETosis, thereby creating a self-amplifying inflammatory feedback loop that drives disease progression.^31–34 Our findings, in alignment with emerging evidence,^31,32,35 underscore the pivotal role of neutrophil infiltration and NETosis in perpetuating GA pathogenesis. While systemic neutrophil depletion via anti-Ly6G antibody attenuated MSU-induced inflammation (Figure 1), consistent with reports that neutrophil reduction alleviates GA-related tissue damage,³⁶ this blunt therapeutic strategy disregards neutrophils’ indispensable roles in host defense. Given that NETosis is a key effector mechanism by which neutrophils drive GA pathology, targeting this process—rather than depleting the cells themselves—may offer a more precise therapeutic strategy. Therefore, selectively inhibiting the pathogenic effector functions of neutrophils represents a more therapeutically viable approach. Our study strategically focuses on disrupting neutrophil effector functions - particularly NETosis - through a multi-targeted phytochemical intervention, thereby preserving physiological immunity while mitigating pathological inflammation.

Building upon our previous characterization of the Ras-Raf-ERK/ROS axis in MSU+PMA-induced NETosis,¹³ we systematically investigated how SFSG, a multi-component flavonoid preparation, intervenes at critical nodes of this inflammatory cascade. SFSG dose-dependently reduced ankle swelling, inflammatory cytokines (IL-1β, IL-6, TNF-α), and neutrophil infiltration in GA mice, with high-dose efficacy comparable to colchicine (Figure 5). This is consistent with the results of previous study.³⁷ Prior studies^38–42 have identified various flavonoids (eg, (-)-epicatechin, quercetin, gallate) capable of inhibiting NETs formation by targeting ROS or proteolytic enzymes (MPO/NE). While previous studies have highlighted the role of individual flavonoids in NETs inhibition, our findings uniquely demonstrate that SFSG, a multi-compound flavonoid mixture, achieves broader antioxidant and cytoprotection effects by simultaneously modulating ROS production and diminishing the permeability of the cell membrane in neutrophils (Figures 3 and 4). The synergy among its constituent flavonoids likely underlies this multi-target capacity, enabling concurrent reduction of oxidative stress, membrane damage, and, as downstream consequences, CitH3 and NETosis. This multimodal mechanism distinguishes SFSG from current mainstay therapies⁴³ such as colchicine and IL-1 inhibitors, potentially offering a broader therapeutic window against neutrophil-driven inflammation. While SFSG exhibited comparable acute anti-inflammatory efficacy to colchicine in our model, its ability to simultaneously suppress oxidative stress, NETosis, and Raf/ERK signaling suggests a distinct pharmacological profile that may be advantageous in recurrent or refractory GA. Further studies in chronic models are warranted to evaluate its long-term benefits and safety.

MPO serves as an integral component of NETs and is a characteristic marker of neutrophils.⁴⁴ Patients with GA exhibit spontaneous NETs formation and increased release of MPO and NE, with enhanced NETosis observed in synovial fluid and NETs released by synovial tissue.^33,34 Similarly, MPO content in joint tissues of MSU-induced GA mice was significantly increased, with enhanced expression (Figure 6C). Surprisingly, the level of MPO was significantly reduced after the intervention of SFSG. This reduction could be attributed to either diminished neutrophil infiltration into the joint or impaired degranulation/activation of neutrophils already present.⁴⁵ However, the concomitant decrease in MPO-positive cells observed via IHC (Figure 6A) support that attenuated recruitment is the predominant mechanism. Therefore, our data imply that SFSG’ anti-GA effects may be attributed to its inhibition of neutrophilic activity and NETs formation.

In many cases, the activation of NOX2 and production of ROS are indispensable for the formation of NETs. The pathogenesis of GA involves Raf signaling pathway activation, which leads to the activation of ERK/MAPK and ultimately initiates the inflammatory response.⁴⁶ The formation of NETs, essential for GA, also requires the activation of Raf-MEK-ERK pathway.⁴⁷ Our data reveal that SFSG treatment is associated with a significant reduction in Raf phosphorylation, coinciding with disrupted downstream ERK activation and reduced NETs formation (Figures 4 and 6). The observed downregulation of NOX2 and PADI4—established effectors downstream of ERK—further supports that SFSG exerts its effects through pathway-level inhibition of the Raf/ERK axis. The findings agree with a recent report indicating that GPR105 is widely in neutrophils and is significantly expressed in peripheral blood neutrophils of GA patients. MSU activates GPR105, which subsequently activates the Raf-MEK1/2-ERK1/2 pathway, leading to NOX activation and contributing to NETs formation.⁴⁸ In addition, the use of Cl-amidine,⁴⁹ an inhibitor of PADI4 (an important enzyme involved in NETosis), significantly attenuated joint swelling and inflammatory factor infiltration in MSU-induced GA mice, reducing NETs production. Unlike single-target inhibitors (eg, Cl-amidine targeting PADI4), SFSG exerts multi-faceted effects by downregulating both NOX2-driven ROS production and Raf/ERK signaling, offering a synergistic approach to attenuate GA progression. These data indicated that the preventive effects of SFSG on GA are related to its inhibition of NOX2 and PADI4 activation, reduction in ROS production, blockade of the Raf/ERK signaling pathway, and suppression of histone H3 citrullination (Figure 6).

In this study, we utilized a well-established in vitro NETosis model in which neutrophils were stimulated with a combination of MSU and PMA. While PMA is a potent pharmacological activator of protein kinase C and may not fully mimic the physiological initiation of NETosis in vivo, it is widely employed in neutrophil research to provide a robust and reproducible signal that engages downstream NETosis mechanism, including Raf/ERK activation and histone citrullination.⁵⁰ Indeed, as we have previously reported,¹³ MSU alone at physiological/low concentrations may induce only modest NETosis, whereas the addition of PMA enhances the responsiveness and allows clearer dissection of execution-phase mechanisms—a strategy also adopted in other NETs-focused studies. Importantly, the relevance of the pathways examined here (eg, Raf/ERK/ROS axis) is strongly supported by our parallel in vivo GA model, in which SFSG similarly inhibited neutrophil infiltration, NETosis, and inflammatory cytokine production. Thus, while we acknowledge that PMA may intensify NETosis beyond purely MSU-driven levels, this model remains a methodologically valid and informative system for evaluating pharmacological inhibitors like SFSG within a controlled mechanistic framework.

Despite the compelling evidence, several limitations of this study should be acknowledged. While we demonstrate the potent multi-target effect of the SFSG, the relative contribution and potential synergy of its individual components (eg, astilbin, neoastilbin) to NETosis inhibition remain to be dissected. Furthermore, although the Raf/ERK-CitH3 axis is strongly implicated, contributions from other pathways cannot be excluded. Moreover, the specific upstream receptors or membrane targets through which SFSG initially engages to modulate neutrophil signaling remain to be identified, representing a key area for future mechanistic investigation. Future studies should therefore employ compound-depletion approaches or gene-editing tools to deconvolute the roles of specific flavonoids, investigate the initial binding events, and validate the therapeutic potential in clinically relevant models, such as using neutrophils from GA patients.

Conclusions

In summary, this study demonstrates that SFSG, a flavonoid-rich extract from Smilax glabra Roxb., alleviates GA in experimental models by targeting neutrophil-driven inflammation. Our findings indicate that SFSG treatment is associated with reduced activation of the Raf/ERK pathway, diminished histone H3 citrullination and NETosis, and attenuated joint inflammation (Figure 7). The downregulation of NOX2 and PADI4 further supports the involvement of this axis. This multi-target profile distinguishes SFSG from current therapies such as colchicine and IL-1 inhibitors, highlighting its potential as a novel, phytochemical-based candidate for gout management. These results support the translational potential of SFSG as a basis for developing safer, multi-target phytotherapeutics, meriting further preclinical optimization and clinical investigation.