2.1 Human specimens and cell culture
We obtained human thyroid cancer (TC) specimens from a cohort of 50 patients underwent TC resection at the Second Affiliated Hospital of Nanchang University during March 2018 to June 2022. The acquisition of samples received appropriate informed consent from the patients, and all experimental protocols were approved by the research ethics committee of the Second Affiliated Hospital of Nanchang University.
2.2 Cell lines and cell culture
The thyroid epithelial cell line Nthy was acquired from the European Collection of Animal Cell Cultures, while the human thyroid carcinoma cell lines (Bcpap, TPC-1 and 8505c) were procured from the Shanghai Institute of Cell Biology (Shanghai, China). The K1 cell line was obtained from the American Type Culture Collection. All cells were cultured in Dulbecco’s Modified Eagle’s media (DMEM) supplemented by 10% fetal bovine serum (FBS) at 37°C in a humidified atmosphere with 5% CO2.
2.3 Data Collection
Our digital data was derived solely from The Cancer Genome Atlas (TCGA) (https://cancergenome.nih.gov) database and the Genotype-Tissue Expression (GTEx) (https://www.gtexportal.org/) database. The TCGA-THCA dataset includes expression data from 571 samples, consisting of 59 normal samples and 512 tumor samples. For our research, we obtained GTEx-thyroid dataset which includes 276 samples from GTEx.
2.4 Weighted Gene Co-expression Network Analysis
The WGCNA algorithm, which is widely employed for high-throughput gene coexpression profiling, was utilized in this study to identify gene coexpression networks associated with different diseases [17]. For constructing a scale-free co-expression network encompassing all genes, a soft threshold of 10 was applied. Furthermore, a dynamic tree cutting method was employed to identify modules within the network. To determine the modules significantly correlated with thyroid cancer and ubiquitination, the correlations between Eigengenes and THCA and GOBP_PROTEIN_POLYUBIQUITINATION, respectively, were analyzed. The gene connectivity within modules was assessed using the absolute value of Pearson correlation, allowing for the identification of hub genes exhibiting high within-module connectivity.
2.5 Protein-Protein Interaction, Molecular Docking and Gene set enrichment Analysis
As previously described [18], we utilized the BioGRID (https://thebiogrid.org/) interactome dataset to identify the interacting genes in TC patients. Subsequently, the construction of the protein-protein interaction (PPI) network was conducted using the Cytoscape software, followed by its visualization. The identification of protein nodes was accomplished using the Cytoscape plugin.
The HDOCK (http://hdock.phys.hust.edu.cn/), a commonly employed computational resource, employs a hybrid docking algorithm that combines template-based modeling and free docking for the automated utilization of binding data from the PDB [19]. To investigate the interaction between USP1 and KRT17, we employed HDOCK as a predictive tool to determine their docking mode. Subsequently, we utilized PyMOL software to visualize the obtained results.
GSEA (http://bioinfo.life.hust.edu.cn/GSEA), a software package equipped with an extensive collection of 1,325 gene sets, offers a valuable tool for the interpretation of gene expression data [20]. The samples were classified into two groups based on the expression levels of THCA, and this approach was utilized to examine the enrichment of gene sets containing THCA. The analysis conducted 1,000 permutations and employed the h.All.V7.4 Symbols.gmt (Hallmarks) gene set database for comprehensive investigation. Significance was determined at P value < 0.05 and FDR < 0.05.
2.6 Quantitative real-time PCR (qrt-PCR)
We used Trizol reagent (catalog number: 15596026, Invitrogen, USA) to extract total RNA. mRNA expression levels were assessed using SYBR Green assays with RT primers and SYBR Green from Takara Biotechnology (Catalog: DRR041A, TAKARA, Dalian, China). During the simultaneous process, we used human microtubule-associated protein as an internal control for amplification. The primer sequences used in this study are provided in Supplementary Table S1.
2.7 Western blotting
We prepared the total protein extract according to the method described earlier. We used RIPA buffer containing a mixture of protease inhibitors (manufactured by Shanghai Berry Times Company) to extract the protein. The antibodies employed in this study included anti-USP1 monoclonal antibody (1:1000 dilution; CST, 8033), anti-KRT17 monoclonal antibody (1:1000 dilution; CST, 12509), anti-β-catenin monoclonal antibody (1:1000 dilution; abcam, ab32572), anti-c-Myc monoclonal antibody (1:1000 dilution; abcam, ab32072), and anti-Tubulin monoclonal antibody (1:1000 dilution; Proteintech, 11224-1-AP). Following a 2-hour incubation with secondary antibodies at room temperature, the Quantity One software (Bio-Rad Laboratories Inc., Hercules CA, USA) was employed for quantitative analysis of protein band intensity. [21].
2.8 Immunohistochemistry (IHC)
The study involved the collection of TC samples and paired normal thyroid tissues, which underwent an antigen retrieval process. Heated the antigen retrieval solution (EDTA, pH 8.0) in a microwave for 40 minutes, then incubated with goat serum for 30 min. Subsequently, tissue slices were incubated with anti-USP1 monoclonal antibody (1:1000, CST, 8033) overnight at 4°C, followed by incubation with an HRP-conjugated secondary antibody (Boster) at room temperature for two hours. before immunostaining with the DAB Detection Kit (Maxim) for two minutes. Ultimately, the positive area proportion was semi-quantitatively evaluated by three pathologists blinded to the clinical parameters.
2.9 Plasmids and reagents
In all diagrams, the shRNA structures of USP1 and KRT17 are provided in supplemental table S1. These shrnas were integrated into the lentivirus pLKO vector obtained from a gene pharmaceutical company based in Shanghai. The overexpressed plasmid was sourced from another genetics company also located in Shanghai. Specifically, USP1, HIS-labeled KRT17, and HIS-labeled beta-catenin were inserted into the P-CMV vector. After 48 hour of transfection, the viral supernatant was collected and filtered as per the manufacturer's instructions for selecting lentiviral transduction cells using purinomycin.
2.10 Cell colony formation assay
In colony formation experiments, the cells were seeded onto a 6-well plate. Once the colonies reached the desired size, they were fixed with 4% paraformaldehyde for 30 minutes and stained with 1.0% crystal violet for an additional 30 minutes until the clones became visible. Colonies comprising more than 50 cells were enumerated and subjected to analysis in order to investigate their correlation with the initial number of seed cells.
2.11 Cell counting kit-8 (CCK-8) assay
The CCK-8 assay was conducted in accordance with the manufacturer's instructions using the CCK-8 kit. Transfected cells were incubated for 48 hours and then seeded into each well of a 96-well plate at a density of 5×103 cells per well. Following incubation for 24, 48, 72, 96, and 120 hours, each well was supplemented with 10 µL of CCK8 reagent and further incubated at a temperature of 37℃ for a duration of two hours. Absorbance at the wavelength of 450 nm was measured using an enzyme-labeled apparatus (ELX-800; Bio-Tek, Winooski, VT, USA). Moreover, the inhibitory effects of PTL were evaluated utilizing CCK-8 assays. Parthenolide (with a purity exceeding 98%) was procured from Absin Bioscience Inc (#20554-84-1), while Dimethyl sulfoxide (DMSO) was sourced from Sigma-Aldrich (#276855) [22]. Following a 24-hour incubation period, cells were exposed to PTL at different concentrations or DMSO for 24, 48, 72, 96, and 120 hours. The specific operation is as described above.
2.12 Luciferase reporter assay
Cells were co-transfected with plasmids containing firefly luciferase reporters and USP1 plasmids in the TOP/FOP-Flash reporter assay. Following a 48-hour transfection period, cells were harvested and subjected to analysis using the Dual-Luciferase Reporter Assay System from Promega (Madison, WI, USA). Luciferase activity was assessed using the PerkinElmer EnSpire Multilabel Reader 2300 (PerkinElmer Inc., Waltham, MA, USA), with normalization to Renilla luciferase activity to ensure uniform transfection efficiency.
2.13 Co-immunoprecipitation experiment
Cell lysates were subjected to overnight incubation at 4°C with 50 µl of protein G beads and 1 µg of the designated antibody. Subsequent centrifugation effectively separated the protein G beads from the solution. Following this, loading buffer was then introduced to the resulting mixture and subjected to thermal treatment at 100°C for a quarter. The immunoprecipitated proteins were then subjected to analysis via SDS-PAGE and immunoblotting. The intensity of the protein bands was quantitatively examined and assessed using the specialized software.
2.14 Mouse xenograft assay
The animal experiments conducted in this study adhered to the ethical guidelines prescribed by the Animal Ethics Committee of the Second Affiliated Hospital of Nanchang University. Bcpap cells were suspended in a medium devoid of fetal bovine serum and subsequently subcutaneously injected into 6-week-old nude mice at a volume of 100 µl. Each experimental group, consisting of six mice, received daily intraperitoneal injections of either a vehicle solution (comprising 2% DMSO, 40% PEG400, and 2% Tween 80 in normal saline) or PTL (20 mg/kg) for a duration of 42 days. Regular assessments were made to monitor changes in body weight on a daily basis. Additionally, tumor measurements and dimensions were taken at intervals of 3 days employing a digital caliper. Meanwhlie, the corresponding tumor volumes were computed according to the dedicated formula [23].
2.15 Statistical analysis
The results are presented as mean ± SD and were evaluated through GraphPad Prism 5 (GraphPad Software, USA) based on data obtained from a minimum of three distinct experiments. Statistical analyses involved pairwise comparisons utilizing the Statistical comparisons among multiple groups were conducted using Student's t-test or one-way ANOVA. Furthermore, logistic regression models were utilized for both univariate and multivariate analyses. All P values were computed as two-tailed tests, with statistical significance defined at a threshold of < 0.05.