Patients and clinical samples

Paired samples of PTC tissues and paracancerous tissues were collected at Shenzhen People’s Hospital. The PTC patients included in the study had not received any chemotherapy or radiotherapy before the tissue collection. Patients with infectious diseases such as chronic hepatitis B virus infection, autoimmune diseases like Hashimoto’s thyroiditis, and primary hyperthyroidism were excluded from the study. The study ethics was approved by the ethics committees of the Shenzhen People’s Hospital (LL-KY-2022091). The informed consents were collected from all the participants.

RNA extraction and quantitative real-time PCR

RNA was extracted from PTC and paracancerous tissues or cells using the TranZol Up Plus RNA Kit (TranGen Biotech, ER501), following the manufacturer’s instructions. Subsequently, the RNA was reverse transcribed into cDNA using the EasyScript All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (One-Step gDNA Removal) (TranGen Biotech, AE341). Quantitative real-time PCR was then conducted according to the manufacturer’s instructions using PerfectStart Green qPCR SuperMix (TranGen Biotech, AQ601). The primers used in this study are listed in Supplemental Table 1.

Data source and description

We obtained the mRNA expression matrix file of TCGA-THCA which includes 397 samples with 712 normal, from the Genomic Data Commons Data Portal (https://www.ncbi.nlm.nih.gov/). Additionally, two mRNA expression datasets (GSE3678 and GSE129879) were retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). These datasets consist of PTC and normal samples. To investigate protein expression data, we utilized the Human Protein Atlas (HPA) (https://www.proteinatlas.org/), a proteomics database that provides information on the organization and cellular distribution of human proteins. Specifically, we searched for PUS7, DNMT3B, and CD47 protein expression data in the HPA.

Arraystar human pseudouridine (ψ) small RNA modification microarray analysis

Nine pairs of PTCs and paracancerous tissue samples were selected for RNA extraction. The Arraystar Seq-StarTM poly(A) mRNA Isolation Kit (Aksomics, AS-MB-006-01/02) was used according to the manufacturer’s instructions to obtain the small RNA. The quantity of RNA samples was determined using a NanoDrop ND-1000 spectrophotometer, and RNA integrity was assessed using a Bioanalyzer 2100 or gel electrophoresis. Each total RNA sample (1–5 µg) was immunoprecipitated with 4 µg of anti-pseudouridine antibody (Diagenode, C15200247) and 1 mg of Protein G Dynabeads (Thermo Fisher, 11203D) in 500 µL of RIP buffer. The modified RNAs, referred to as ‘IP’, were extracted from the immunoprecipitated magnetic beads. The unmodified RNAs, known as ‘Sup’, were obtained from the supernatant. Arraystar’s standard protocols were used to enzymatically label the ‘IP’ RNAs with Cy5, and the ‘Sup’ RNAs with Cy3 in separate reactions. The labeled RNAs were then combined and hybridized onto the Arraystar Human small RNA Modification Microarray (8 × 15 K). Finally, the array was scanned using an Agilent Scanner G2505C in two-color.

The acquired array images were analyzed using Agilent Feature Extraction software (version 11.0.1.1). The raw intensities of IP (immunoprecipitated, Cy5-labelled) and Sup (supernatant, Cy3-labelled) were normalized by taking the average of log2-scaled Spike-in RNA intensities. After normalization, the probe signals that had Present (P) or Marginal (M) QC flags in at least 4 out of 18 samples were retained. Multiple probes from the same small RNA (miRNA/tsRNA (tRF&tiRNA)/pre-miRNA) were combined into a single RNA level. The abundance of ‘ψ’ was analyzed based on the normalized intensities of Cy5-labelled IP (modified RNA). Differentially ψ-modified RNAs between two comparison groups were identified using fold change (FC) and statistical significance (p-value) thresholds. The ψ-modification patterns among samples were displayed using a hierarchical clustering heatmap created with R.

MeRIP-PCR

RNA was isolated from paired PTC and paracancerous tissue samples and diluted to a concentration of 2 µg/µl. The RNA was then fragmented at 65 ℃ for 5 min. A small aliquot of the fragmented RNA (5 µg) was reserved as the input sample for qPCR normalization. DynabeadsTM M-280 Sheep Anti-Mouse IgG (Invitrogen, 11201D) were used and washed twice with IP buffer (10 mM Tris pH_7.4, 250 mM NaCl, 0.1% NP-40) and then coupled with pseudouridine antibody (Diagenode C15200247) for 2 h at 4 ℃. Subsequently, the fragmented RNA and RNase inhibitor (Enzymatics, Y9240L) were added to the IP buffer. Samples were eluted with 200 µl Elution buffer (100 mM Tris pH 7.4, 1 mM EDTA, 0.05% SDS), 4 µl Proteinase K (Qiagen, 19131), and 2 µl RNase inhibitor for 1 h at 50 ℃ on a rotor. The supernatant was collected for RNA isolation.

For the RT reaction, 2 µl of RNA was used with either M-MuLV Reverse Transcriptase (Enzymatics, P7040L) for miRNA or SuperScriptTM III Reverse Transcriptase (Invitrogen) for pre-miRNA. The RT reactions were performed in the Gene Amp PCR System 9700 (Applied Biosystems). In the case of tsRNA, the RNA underwent pretreatment using the rtStar™ tRF&tiRNA Pretreatment Kit (Arraystar, AS-FS-005) to undergo acylation, 3’-cP removal, 5’-P addition, and demethylation. The rtStar™ First-Strand cDNA Synthesis Kit (3’ and 5’ adaptor) (Arraystar, AS-FS-003) instructions were followed, including ligation of the 3’ splice, reverse transcription primer hybridization, and ligation of the 5’ for cDNA synthesis for tsRNA. Each cDNA sample was configured separately with a realtime PCR reaction system. The prepared 384-PCR plate was then placed on a QuantStudio5 Real-time PCR System (Applied Biosystems) for PCR reactions. The resulting data were analyzed using the 2−△△Ct method to calculate the percentage of input (%Input) for each MeRIP fraction. The %Input value was determined using the formula %Input = 2− CtMeRIP/(2− CtMeRIP+2−CtSupernatant)*Fd*100%. The primers used are listed in Supplemental Table 1.

Cell culture and transfection

The human PTC cell lines B-CPAP and human normal thyroid cells Nthy-ori 3 − 1 were obtained from Fenghui Biotechnology Co., Ltd (Hunan, PR China). The cell lines were cultured in RPMI-1640 medium (Gibco, C11875500BT) supplemented with 10% fetal bovine serum (ExCell Bio, FSP500), 100 µg/mL penicillin, and 0.1 mg/mL streptomycin. The cells were maintained in a humidified atmosphere at 37 °C with 5% CO2.

For the transfection experiments, B-CPAP and Nthy-ori 3 − 1 cells were divided into different groups. These groups included the si-NC control group (transfected with si-NC control), the mimic NC control group (transfected with mimic NC), the si-PUS7 or si-CD47 group (transfected with si-PUS7 or si-CD47), and the si-PUS7 + miR-8082 group (transfected with both si-PUS7 and miR-8082 mimic). The si-PUS7 (5’-GGAAGAAGAGGAGGAAGAU-3’) or si-CD47 (5’-GGAUCCAGUCACCUCUGAATT-3’) and si-NC (5’-UUCUCCGAACGUGUCACGUTT-3’) control were obtained from Sangon Biotech (Shanghai, PR China). The microRNA mimic (5’- UGAUGGAGCUGGGAAUACUCUG-3’) and mimic NC (5’- UUGUACUACACAAAAGUACUG-3’) were also obtained from Sangon Biotech (Shanghai, PR China). B-CPAP and Nthy-ori 3 − 1 cells were transfected at ~ 70–80% confluency with siRNA or microRNA mimic using Lipofectamine 3000 (Invitrogen, L3000-015) following the manufacturer’s instructions.

MTT assay

Cells were seeded into 96-well plates at a density of 2 × 10^3 cells per well, 24 h after transfection with si-RNA or si-NC. The plates were then incubated at 37 °C with 5% CO2 for 48 h. Next, 10 µl of MTT solution (Sigma-Aldrich, M5655-1G) was added to each well, followed by an additional 4 h of incubation. After dissolving the intracellular formazan crystals in DMSO (150 µL per well; Sigma-Aldrich, D8418) for 20 min, the absorbance at 490 nm was measured using an Emax precision microplate reader (BioTek, PR China). The experiments were repeated at least three times, with 5–6 wells being set up each time.

Cell cycle analysis

Cell digestion was carried out using trypsin, followed by centrifugation and removal of the supernatant. The cells were then resuspended in ice-cold PBS and underwent repeated washing. To ensure proper fixation, single cell suspensions were prepared by adding ice-cold 70% ethanol solution to the cell and incubating it for more than 24 h at 4 °C. After centrifugation and washing with ice-cold PBS, the samples were stained with PI staining solution (MedChemExpress, HY-K1071-50T). Flow cytometry analysis was conducted using a BD FACSCanto II flow cytometer (BD Biosciences, US) and the results were analyzed using FlowJo X 10.0.7 (BD, US). The experiments were repeated at least three times.

Measurement of cell apoptosis

The Annexin V-FITC/PI apoptosis kit (MULTI SCIENCES, AP101-100) was utilized to evaluate apoptosis in B-CPAP and Nthy ori 3 − 1 cells. After treatment, the cells were collected and washed with ice-cold PBS twice. They were then resuspended in binding buffer and harvested into centrifuge tubes. Subsequently, the cells were incubated with Annexin V/PI in the dark for 15 min. The apoptotic cells were quantified using a BD FACSCanto II flow cytometer (BD Biosciences, US). Cells transfected with a scrambled si-RNA were used as the negative control. Early phase apoptotic cells were categorized as Annexin V+/PI- and late phase apoptotic cells as Annexin V+/PI+. The results of the experiments were detected and analyzed using FlowJo X 10.0.7 (BD, US) and the BD FACSCanto II flow cytometer. The experiments were repeated at least three times.

Cell migration assay

A Wound Healing assay was performed to investigate cancer cell migration. B-CPAP cells were transfected with si-RNA or si-NC and then seeded into 12-well plates and grew until they reached 80% confluence. To create wounds, the monolayer cells were scraped using a 200 µl pipette tip, and any non-adherent cells were removed by washing with the medium. After 48 h, the treated and control cells were examined, and migration images were captured using a 10x objective in phase-contrast microscopy. The rate at which cells moved towards the scratched area was used to determine cell migration. The size of the scratched area was quantified using ImageJ™ software.

Statistical analysis

The experiments were repeated at least three times. Statistical data were analyzed using GraphPad Prism 7.0 software and presented as the Mean ± SD of results from three independent experiments. The statistical analyses included the Student’s t-test and one-way ANOVA. A p-value < 0.05 was considered statistically significant.