Patients and tumor specimens

Download and analyze gene expression data and relevant clinical information from the TARGET website (https://ocg.cancer.gov/programs/target) for osteosarcoma patients. Eighty-four patients were involved and split into two groups based on the PSMD14 gene expression levels: high and low. The differences in relapse-free and overall survival between the two groups were examined using the Kaplan–Meier curve. From January to August 2021, we obtained fresh tumor samples and non-neoplastic para-tumoral bone tissue samples from 13 patients with osteosarcoma hospitalized at the Department of Bone and Soft Tissue at Henan Cancer Hospital.

The bone samples were placed in liquid nitrogen immediately, and following quick freezing, they were held in a freezer at − 80 °C for RNA and protein extraction. Additionally, we gathered paraffin-embedded samples and clinicopathological characteristics (gender, age, surgical approach, and survival period) from 76 patients with complete clinical data who were pathologically diagnosed with osteosarcoma between January 2014 and January 2017. These samples were subjected to immunohistochemistry. The Henan Cancer Hospital’s institutional research ethics committee approved this study after receiving the informed permission of all participants or their duly appointed representatives.

Immunohistochemistry (IHC) analysis

The paraffin-embedded human osteosarcoma tissue samples were sectioned at a thickness of 4 microns and, after baking at 70 ℃ for 30 min, dewaxed twice with 100% xylene each time, hydrated with gradient alcohol series, and incubated with 0.3% H2O2 in methanol to block endogenous peroxidase activity at room temperature for 15 min. Antigen recovery was done with citric acid antigen repair solution (high temperature and pressure); slices were treated with rabbit anti-PSMD14 monoclonal antibody (Abcam, 1:500, UK) at 4 °C overnight after being blocked with 10%t goat serum for 15 min. The sections were incubated with horseradish peroxide-labeled secondary antibodies (IgG-HRP, Abcam, 1:2000, UK) at room temperature for 1 h after being washed three times with PBS buffers. The samples were stained using a diaminobenzidine color development kit (Beyotime, Shanghai) following the manufacturer’s instructions, and subsequently hematoxylin was used to reverse-stain them. Last, the software Image Scope (Leica Biosystems, Germany) was used to visualize staining.

According to the percentage of stained cells (0 = negative; 1 = 6–25%; 2 = 26–50%; 3 = 51–75%; and 4 = 75–100%) and staining intensity (0 = no staining; 1 = slight staining; 2 = moderate staining; and 3 = strong staining), the specimen was graded. The score was determined using a semi-quantitative scoring method, with the low-expression group divided into products ranging from 0 to 6, and the high-expression group divided into products ranging from 7 to 12.

Cell culture and reagents

Human OS cell lines (MG63, U2OS, 143B, and Saos-2) and normal human osteoblasts (NHOst and hFOB 1.19) were obtained from the Chinese Academy of Sciences. 143B cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco); Saos-2, U2OS, MG63, hFOB 1.19, and NHOst cells in RPMI-1640 medium (Invitrogen, USA) containing with 10% fetal bovine serum (Gibco, USA) and 100 units/mL penicillin/streptomycin (Invitrogen, USA). All cell lines were grown in a humidified incubator under humanized conditions (37 °C, 5% CO2). Anlotinib (AL3818, S8726) was obtained from Selleckchem, dissolved in DMSO at a stock concentration of 10 mM, and frozen at − 20 °C for future use.

Lentiviral Infection and transfection

The negative control and lentivirus harboring PSMD14 shRNAs (sense-loop-antisense) were acquired from GeneChem (Shanghai, China). The OE-PSMD14-expressing plasmids were purchased from GeneChem (Shanghai, China). To establish a stable cell line, osteosarcoma cells were infected with lentivirus for 72 h and screened with 2 µg/mL purinomycin. The sequences for shRNAs are as followed: shPSMD14-1 (5′-CAGAAGATGTTGCTAAATT-3′), shPSMD14-2 (5′-GTACTTATGACCTCAAATA-3′), and shPSMD14-3 (5′-GTTGGATACTGTCGTATTT-3′). Transfection was carried out with Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) per the manufacturer’s protocol.

RNA isolation and quantitative RT-PCR (qRT-PCR)

Using qRT-PCR, the relative expression level of reference genes was measured. Briefly, total RNA was isolated from cells and tissues using TRIzol (Takara, Japan) according to the manufacturer’s recommendations. Then, a PrimeScript RT reagent Kit (Takara, Japan) was used to reverse transcribe into cDNA, and RT-PCR was performed using SYBR Premix Ex Taq (TaKaRa) on a 7500 Real-Time PCR system (Applied Biosystems, USA) according to the manufacturer’s instructions. The gene-specifific primers were as follows: PSMD14 (sense: 5′-GGAGGAGGTATGCCTGGACT-3′, antisense: 5′-GGTTTTCTCCATGCTGTTTCTT-3′), GAPDH (sense: 5′-AGGGCTGCTTTTA ACTCTG-3′, antisense: 5′-CTGGAAGATGGTGATGGG-3′). GAPDH was used as an endogenous control, and each sample’s relative gene expression was calculated according to the 2−ΔΔCt method.

Cell viability determination

Cells (1–2 × 104 cells/well) were seeded into a 96-well plate in a 5% CO2 incubator at 37 ℃ for 24 h. Cell viability was assessed using Cell Counting Kit-8 (CCK-8) (Dojindo, Japan) after 30 min of incubation, per the manufacturer’s instructions. The absorbance values of the water-soluble tetrazolium salt were assayed using a microplate reader (Molecular Devices, USA) at 450 nm.

Colony formation

Colony formation assays could be useful for assessing cell viability; cells at plating efficiency density (600 cells/well) were placed into a 6-well plate and cultured for 10–14 days. Following two PBS washes of each well’s cells, 100% methanol was used to fix the attached cells for 20 min. Finally, the cells were visualized by 0.5% crystal violet staining, and the colonies (> 50 cells) were manually counted with an inverted microscope (Nikon).

TUNEL assay

For apoptosis assays, the TUNEL assay was employed to evaluate cell apoptosis ratio using a cell death kit (Beyotime, Shanghai) according to the instructions, and LSM 880 laser scanning confocal microscope (Zeiss, Germany) was used to image and analyze each slide.

Western blot analysis

According to the manufacturer’s instructions, the total cell protein was extracted using the RIPA Lysis Buffer Kit (Biotechnology, CA). Equal amounts of protein from each sample were resolved on polyacrylamide gels with a sodium dodecyl sulfate content of 10% before being transferred to polyvinylidene fluoride membranes (Millipore, USA). The membranes were treated with certain primary antibodies overnight at 4 °C after being blocked with 5% skim milk for 1 h at 37 °C. The membranes were incubated with the proper secondary antibodies at 37 °C for 1 h the next day, following three PBS washes. The primary antibodies used were: anti-PSMD14 (1:500, Abcam), anti-GAPDH (1:10,000, Abcam), anti-PI3K (1:500, Affinity biosciences), anti-p-PI3K (1:10,000, Affinity biosciences) anti- AKT (1:10,000, Abcam), anti- p-AKT (1:10,000, Abcam), anti-mTOR (1:1000; CST), anti- p-mTOR (1:1000, Abcam), anti-Vimentin (1:2000, Abcam) and anti-Snail (1:1000, CST). An enhanced chemiluminescence kit (ECL, Germany) was used to identify protein bands, and GAPDH was used as an internal standard to gauge the protein’s intensity.

Wound-healing assay

Single-cell suspensions were seeded into 6-well plates and then allowed to grow at 37 °C until confluence levels greater than 90% were achieved. A line was scratched with a sterile 10-µL pipette tip across the center of the cell monolayer. Then, the cells were washed and cultivated for an additional 36 h, and plate images of the gaps were obtained under a microscope at 0 and 36 h of the experiment. The cell migration rate of each sample was analyzed using Image J software.

Transwell assay

Cells were plated in a Transwell chamber (Corning, Inc., USA) with 8 μm wells and a gel matrix (BD Biosciences, USA) for the invasion and migration assay. The technique was streamlined: plasmid-free lentivirus-infected cells were employed as negative controls, and 40 µL of diluted VitoGel 3D-hydrogel (PeproTech, USA) was added to the upper chamber in the invasion assay. Briefly, 200 µL osteosarcoma cells were added to the chamber at a concentration of 3 × 104. About 600 µL of a medium containing 20% FBS was introduced into the chamber. After 36 h of incubation at 37 °C, migratory or invasive cells were fixed with absolute methanol for 5 min, washed twice with water, and stained with 0.1% crystal violet. After closure, cells were imaged and counted using an inverted microscope in three randomly selected statistical fields.

Establishing of anlotinib-resistant osteosarcoma cell lines

To verify the effect of PSMD14 in reversing anlotinib resistance in vivo, we first established anlotinib-resistant cells (Saos-2-R and U2OS-R) with increasing concentrations of anlotinib from 0.5 to 7 µM over 6 months. The concentration of anlotinib in the medium increased gradually at a rate of 0.5 µM until the resistant sublines stabilized. During that time, the medium changed every day. Cellular sensitivity to anlotinib among resistant sublines (Saos-2-R and U2OS-R) was determined by cell viability assay. Finally, the sublines were obtained and cultured in the medium with 10 mmol/L of anlotinib.

Xenograft nude mouse model

To investigate the effect of PSMD14 on tumor growth in vivo in nude mice, two osteosarcoma cells were stably transfected with PSMD14 knocked down and then cultured and amplified for further study. Six weeks old BALB/c nude mice were implanted subcutaneously with two exponentially growing cells (100 µL 3 × 107 cells) into the flanks of their abdomen. Caliper measurements assessed Tumor sizes every other day, and tumor volume was determined according to the formula: V = [length/2] × [width2]. Mice were euthanized 5 weeks after planting of tumor cells. Then the transplanted tumor was dissected, weighed, and photographed. All animal experiments were conducted under a protocol approved by the Institutional Animal Care and Use Committee of Henan cancer hospital.

Statistical analysis

The normal distribution data were expressed in means ± standard deviation (SD). Statistical differences were analyzed between the two groups using the Student’s t-test, while one-way ANOVA was applied among multiple groups. The non-normal distribution data were represented by median and interquartile, and statistical differences were determined using non-parametric rank sum tests (Mann–Whitney U and Kruskal–Wallis H tests). Kaplan–Meier method and the log-rank test was used to assess survival analysis. The relationships between PSMD14 expression and clinicopathological parameters were evaluated by the χ2 test or Fisher’s exact test. GraphPad Prism 9.0 (GraphPad Software, CA, USA) was used for all statistical analyses. P < 0.05 was considered to be of statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.