Human osteoblasts hFOB1.19, OS Saos2 and U2OS cells were procured from American Type Culture Collection (VA, USA). The cells were cultivated in DMEM (Thermo Fisher Scientific, Massachusetts, USA) supplemented with 10%FBS and 100U/mL penicillin streptomycin. The hFOB1.19 cells were maintained at 33 °C, while Saos2 and U2OS were cultured at 37 °C in a 5%CO2 incubator.

Fifteen gram of sieved (20 mesh) CS powder was mixed with water at a ratio of 1:10 (weight to volume) as the extraction solvent and was subjected to two rounds of hot reflux extraction for 1 h each. After filtration, the filtrates were combined. Subsequently, the CS extracts were concentrated using a rotary evaporator and a vacuum oven until dry. The resulting weighted CS extracts were dissolved in DMSO to obtain a solution with a concentration of 2 mg/mL.

To knock out CXCR4, Saos2 and U2OS were treated with DMEM and a 10-µM CXCR4-specific antagonist AMD3100 (Sigma-Aldrich) or a synthetic small interfering RNA targeting CXCR4 (si-CXCR4) for 24 h. The si-CXCR4 was synthesized by Gene Tech (Shanghai, China), with non-targeting small interfering RNA (siRNA) serving as a negative control (si-NC). The human CXCR4 sequence was then inserted into the pcDNA3.1 plasmid (GenePharma) to create oe-CXCR4. The empty pcDNA3-1 plasmid was used as a negative control (oe-NC). Cell transfection was carried out by Lipofectamine 3000 following manufacturer's instructions. The transfection efficiency of CXCR4 plasmid and siRNA was determined 48 h later using RT-qPCR and western blotting.

Saos2 and U2OS were seeded in 96-well plates at a density of 4 × 103 cells/well and treated with various concentrations of CS (50, 100, 200, 400 and 800 μg/mL) for 24, 48 and 72 h. The MTT assay was performed to assess OS cell viability after drug treatment. Following a 4-h incubation with MTT solution (Sigma-Aldrich, MO, USA),absorbance at 570 nm was measured using a microplate reader from Molecular Devices (CA, USA). The half-maximum inhibitory concentration (IC50) of CS at each time point was calculated using GraphPad Prism software (v8.0; GraphPad software, Inc., CA, USA).

Saos2 and U2OS were seeded in 6-well plates at a density of 200 cells/well and incubated overnight for cell adhesion. Subsequently, they were exposed to various concentrations of CS (50, 100, and 200 μg/mL) and cultured for 14 days. The colonies were fixed with 4% paraformaldehyde for 30 min, stained with 0.1% crystal violet for 10 min, then photographed and counted under a microscope (Olympus, Tokyo, Japan).

Total RNA from Saos2 and U2OS cells was isolated using TRIzol reagents (Thermo Fisher Scientific) and subjected to first-strand cDNA synthesis using a PrimeScript™ RT kit (Beijing Tiangen, China). RT-qPCR analysis was performed on a real-time PCR system (Bio-Rad, California, USA) with a SYBR®Green Master Mix (TaKaRA, Japan) to detect the mRNA expression of CXCR4, PI3K, Akt, and other target genes, using β-actin as an internal reference. The relative expression of the target gene was quantified by calculating the ratio of the Ct value of the target gene to that of the internal reference gene.

Total proteins were extracted from Saos2 and U2OS cells using RIPA lysis buffer (Solarbio, Beijing, China) containing a protease inhibitor mixture (Sigma-Aldrich). After quantification using a BCA kit (Beyotime, Shanghai, China), the proteins were separated by SDS-PAGE and transferred to a PVDF membrane. Subsequently, the membrane was incubated with primary and secondary antibodies, and the protein signals were captured using a chemiluminescence device (Tanon, Shanghai, China) and quantified by Image J software.

For cell migration evaluation, Saos2 and U2OS cells (2 × 105 cells/well) were collected and resuspended in 200 μL FBS-free medium and placed in the apical chamber of a 24-well Transwell culture chamber (Corning, New York, USA) after designated treatment or/and transfection. The basolateral chamber was filled with 750 μL of 10% FBS-supplemented DMEM medium. After 24 h of incubation, migrated cells were fixed in the basolateral chamber with 4% paraformaldehyde for 30 min, and stained with 0.1% crystal violet for 20 min. Finally, the migrated cells were photographed by microscope (Olympus), and the stained cells were counted by Image J software. For cell invasiveness assessment, the Transwell chamber was precoated with Matrigel (Sigma-Aldrich), with subsequent procedures similar to those for migration detection.

All the experiments were repeated at least three times. Statistical analysis was performed using GraphPad Prism v5.0 (GraphPad Software, La Jolla, CA, USA), and SPSS Statistics 21.0. Data were presented as mean ± standard deviation (SD). Statistical significance was determined by Student's t-test or one-way analysis of variance followed by LSD post hoc test with a significance level set at P < 0.05.