The 143B (CRL-8303), HOS (CRL-1543), MG63 (CL-0157), U2OS (CL-0236) and K7M2-WT (CL-0371) cell lines were obtained from Procell Life Science & Technology Co., Ltd., China. The hFOB 1.19 (GNHu14) cells were obtained from the Cell Bank of Chinese Academy of Sciences Type Culture Collection (CBTCCCAS). All cell lines were identified by STR analysis. The K7M2-WT, MG63 and U2OS cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Procell Life Science & Technology). The 143B and HOS cell line was cultured in Minimum Essential Medium (MEM) (Procell Life Science & Technology). All media listed above were supplemented with 10% fetal bovine serum (FBS) (Procell Life Science & Technology) and 1% penicillin-streptomycin (Procell Life Science & Technology). The hFOB 1.19 cells are cultured in dedicated complete medium (CM-0353, Procell Life Science & Technology). All cells were maintained at 37 °C in 5% CO2 humidified atmospheres.

Allicin (M9963, AbMole, shanghai, China, >98%) was dissolved in DMSO. Solutions of allicin at concentrations of 12.5, 25, and 50 μmol/L were prepared using the DMSO solution, with the final DMSO volume fraction set at 0.1%. Add allicin solution to the culture medium of HOS and K7M2 cells. At the same time, establish untreated HOS and K7M2 cell lines as control groups under the same culture conditions.

Colivelin (STAT3 activator, MedChemExpress, New Jersey, USA) was dissolved in DMSO. A solution containing 50 μg/mL was prepared using DMSO, ensuring the DMSO volume remained below 2%. Seed HOS and K7M2 cells into 10cm cell culture dishes, after reaching 70%–80% confluence, treat the cell lines with allicin for 24 hours, then treated with Colivelin for 4 hours.

Total RNA was extracted from cells treated with allicin using the RNA-Quick Purification Kit (RN001, Yishan, China). Following the kit instructions to set reaction parameters, GAPDH was used as an internal control. PD-L1 expression levels in each cell line were quantified using the 2−ΔΔCt method. Primer sequences are listed in the Table 1 below.

Cell Counting Kit-8 (G4103) was purchased from Servicebio (Wuhan, China). Prepare 1.5× 10³ HOS and K7M2 cells in a 100 μL cell suspension and seed them into a 96-well plate. After 24 hours, once the cells have fully adhered. Both cell lines were treated with allicin at concentrations of 12.5 μmol/L, 25 μmol/L, 50 μmol/L, 100 μmol/L, and 200 μmol/L for 24 and 48 hours. Add 10 μL of CCK-8 to each well and incubate for 2 hours. Measure the optical density (OD) at 450 nm using a microplate reader (Thermo Fisher).

Each cell group was seeded at a density of 3 × 10³ cells per well in a 6-well plate and cultured for 14 days at 37 °C with 5% CO₂, with medium changes every two days. Cells were fixed with 4% paraformaldehyde and stained with crystal violet, followed by colony counting using ImageJ software.

HOS and K7M2 cells were seeded into 6-well plates at a density of 3 × 10⁵ cells per well. Once cells have confluently seeded, scrape the cell layer using a 200 μL pipette tip. After thoroughly washing away the scraped cells, apply allicin treatment. Immediately observe cells under a microscope following the scratch, then re-observe after 24 hours and record cell migration status.

Transwell chamber (3422) was supplied by Corning (New York, USA). After 24 hours of allicin treatment, resuspend 1 × 10⁵ cells in 200 μl of medium. For the migration assay, seed cells into the upper transwell chamber of serum-free medium. For invasion assays, Matrigel (Corning) was first coated onto the upper transwell chamber. Add medium containing 20% FBS to the lower transwell chamber. After 24 hours of culture, remove the upper transwell chamber and scrape off cells adhering to the surface using a cotton swab. Fix cells on the lower surface with 4% paraformaldehyde at room temperature for 30 minutes, stain with crystal violet solution (CB0331, Sangon Biotech, Shanghai, China), and observe under a microscope.

RIPA lysis buffer (Solarbio), PMSF (Solarbio), and phosphatase inhibitor (Solarbio) were mixed at a ratio of 1 mL:10 μL:10 μL to prepare the protein lysis buffer. After collecting the cleaved proteins, perform protein quantification using the BCA kit (Solarbio). Protein samples were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to PVDF membranes. The membrane was blocked with 5% milk at room temperature for 1 hour, followed by overnight incubation with primary antibodies at 4 °C. The primary antibodies used were as follows: PD-L1 (Proteintech, 66248-1-IG, China), IL-6 (Proteintech, 21865-1-AP, China), STAT3 (Affinity, AF6294, China), Phospho-STAT3 (Affinity, AF3293, China), JAK2 (Affinity, AF6022, China), Phospho-JAK2 (Affinity, AF3024, China), and GAPDH (Affinity, AF7021, China). Incubate with secondary antibody at room temperature for 1 hour. Protein bands were visualized using a highly sensitive chemiluminescent reagent and analyzed with ImageJ.

Animal experiments were approved by the Institutional Animal Care and Use Committee and Animal Ethics Committee of Affiliated Hospital of Shandong University of Traditional Chinese Medicine (approval number SDSZYYAWE20250615001). Twelve four-week-old male BALB/c mice were purchased from Jiangsu Huachuang Xinnuo Pharmaceutical Technology Co., Ltd. (Jiangsu, China). They were housed under specific pathogen-free (SPF) conditions at the Animal Experiment Center of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine. A 5×10⁶ K7M2 cell injection was administered into the left axilla of each mouse. When the neoplastic size was grown to about 0.05 cm³, mice were divided into two groups: the control group received an intraperitoneal injection of saline, while the treatment group received an intraperitoneal injection of allicin (50 mg/kg), with 6 mice per group. Body weight and tumor volume were recorded throughout the study period. After 14 days of treatment, mice were euthanized, and the allograft tumors were excised. Tumors were either paraffin-embedded or prepared as a single-cell suspension.

Paraffin-embedded tumor sections were processed using Environmental Friendly transparent dewaxing liquid (YA0031, Solarbio, China) for dewaxing. Following hydration with ethanol solutions, sections were heat-treated in sodium citrate antigen retrieval solution (Solarbio) for antigen retrieval. Subsequently, sections were fixed with 4% paraformaldehyde (G1101-500ML, Servicebio, China) for fixation, permeabilized with 0.3% Triton X-100 (IT9100, Solarbio, China), blocked with 10% normal goat serum (SL038, Solarbio, China), and incubated overnight at 4 °C with the primary antibody. The primary antibody required is consistent with Western Blot, with the addition of CD8 (1:100, Proteintech, 29896-1-AP, China) and Interferon gamma (1:50, Affinity, DF6045, China). The secondary antibody was incubated at room temperature for 1 hour. Nuclei were stained with DAPI for 10 minutes at room temperature, and samples were mounted with antifade medium.

The dewaxing, rehydration, and antigen retrieval procedures for immunohistochemical analysis were consistent with those for immunofluorescence analysis. After incubation with H₂O₂ for inactivation, samples were blocked with 5% BSA. Primary antibodies for Ki67 (Proteintech, 27309-1-AP, China) and Interferon gamma were incubated at room temperature for 1 hour, followed by secondary antibody incubation at room temperature for 30 minutes.

After digesting animal tumor tissue with tissue digestion solution and filtering it, prepare a single-cell suspension. Incubate with the corresponding antibody at 4 °C in the dark for 30 minutes. Detect stained cells using a flow cytometer (CytoFLEX, BECKMAN, USA) and analyze the results with FlowJo software. Antibodies used included CD3 (100205, Biolegend, USA), CD4 (100405, Biolegend, USA), CD8a (100711, Biolegend, USA), IFN-γ (505815, Biolegend, USA), and Granzyme B Recombinant (396413, Biolegend, USA).

GraphPad Prism v10.0 software (La Jolla, CA, USA) and SPSS v26.0 software (IBM, Chicago, IL, USA) were used for statistical analysis. All in vitro experiments were performed independently at least three times. Comparisons between two groups were analyzed by a Student’s t-test. A one-way analysis of variance (ANOVA) with Bonferroni correction was used for comparisons between multiple groups. P < 0.05 was considered as statistical significance.

The structure of allicin is shown in Figure 1A. To select osteosarcoma cell lines suitable for this study, we assessed PD-L1 expression levels in untreated HOS, 143B, MG63, and U2OS cell lines. The human osteoblast cell line hFOB 1.19 was used as a control (Figure 1B). We found that PD-L1 expression was highest in the HOS cell line. This result is broadly consistent with the findings of Shen et al (17). Notably, osteoblasts exhibit higher PD-L1 expression than most osteosarcoma cell lines, a difference likely determined by the physiological characteristics of osteogenic differentiation. Having previously demonstrated allicin’s dose- and time-dependent inhibition of osteosarcoma cell viability, we further analyzed K7M2 and hFOB 1.19 cell lines using the CCK-8 assay. Optimal concentrations and durations were determined, with no adverse effect observed on normal osteoblasts (Figure 1C). In a clonogenic assay, allicin treatment significantly suppressed the colony-forming capacity of osteosarcoma cells (Figure 1D). The wound healing assay demonstrated that allicin significantly reduced the wound-healing ability of osteosarcoma cells in a dose-dependent manner (Figure 1E). Transwell assays demonstrated that allicin inhibited the migration and invasion of osteosarcoma cells in a dose-dependent manner (Figures 2A–D). We simultaneously measured PD-L1 expression levels in two osteosarcoma cell lines and found that allicin significantly inhibited PD-L1 expression (Figure 2E).

We demonstrated the inhibitory effect of allicin on PD-L1 expression. To elucidate the mechanism underlying this effect, we conducted bioinformatics analysis. The volcano plot displayed differentially expressed genes associated with favorable prognosis in the GEO database chip GSE19276 (Figure 3A). KEGG and GO pathway analyses identified several pathways correlated with favorable prognosis, including PD-L1 expression and PD-1 checkpoint pathway in cancer, as well as the JAK-STAT signaling pathway (Figures 3B–D). Combining our findings with previous studies, we hypothesize that allicin suppresses PD-L1 expression in osteosarcoma cells via the IL-6/JAK2/STAT3 pathway (18).

To further validate our hypothesis, we employed molecular docking technology and discovered that allicin exhibits strong binding affinity to the IL-6 protein (Figure 4A). We employed Western blot analysis to assess the expression levels of pathway proteins including IL-6, phosphorylated JAK2, and phosphorylated STAT3 (Figures 4B–E). In the allicin-treated group, the expression levels of phosphorylated JAK2 and phosphorylated STAT3 in osteosarcoma cells were significantly suppressed. This demonstrates that allicin indeed inhibits the expression of the IL-6/JAK2/STAT3 pathway. To further determine whether allicin modulates PD-L1 by inhibiting this pathway, we treated both allicin-treated and untreated HOS osteosarcoma cell lines with Colivelin. Western blot analysis revealed that Colivelin-treated cells exhibited significantly elevated phosphorylated STAT3 expression compared to untreated cells. Consequently, PD-L1 protein transcription increased, leading to heightened expression. Concurrently, in cells treated with allicin, Colivelin reversed the inhibition of phosphorylated STAT3 induced by allicin, partially restoring PD-L1 expression (Figure 4F). Integrating these findings, we conclude that allicin suppresses PD-L1 expression by inhibiting the IL-6/JAK2/STAT3 pathway.

We have demonstrated that allicin suppresses the expression of the immune checkpoint PD-L1. To further validate allicin’s ability to inhibit immune evasion in osteosarcoma, we established an osteosarcoma model in BALB/c mice. The mice were divided into two groups: a control group and an allicin-treated group. Analysis of mouse body weight and tumor volume revealed that allicin exhibited favorable therapeutic effects against osteosarcoma (Figures 5A–C). We first employed immunofluorescence to detect the expression of the IL-6/JAK2/STAT3 pathway and PD-L1 protein (Figure 5D). Results demonstrated that allicin effectively suppressed activation of the JAK/STAT pathway, thereby inhibiting PD-L1 expression. To further validate suppression of immune evasion following PD-L1 blockade, immunofluorescence was used to detect CD8+ T cells and IFN-γ (Figure 6A). Immunohistochemical analysis confirmed this conclusion: the proliferative capacity of mouse tumors decreased, while their immune response capability significantly increased (Figure 6B). Following the aforementioned examination, allicin treatment resulted in a significant increase in CD8+ expression within tumor tissues. To further examine the immune microenvironment, flow cytometry analysis of tumor T lymphocyte subsets and immune function expression markers revealed that allicin treatment effectively promoted the recruitment of both CD4+ and CD8+ T cells while enhancing the expression of IFN-γ and granzyme B (Figure 6C). This indicates that allicin indeed promotes tumor immune responses, improves the tumor microenvironment, and enhances the cytotoxic activity of immune cells.