Gene Expression Omnibus (GEO) dataset analysis
We downloaded raw gene expression data from the U.S. National Center for Biotechnology Information (NCBI) GEO available online at: https://www.ncbi.nlm.nih.gov/geo/. The samples (filename GSE53072_RAW.tar) were categorized into two groups: the control group comprising data from three normal thyroid glands (GSM1281638, GSM1281639, and GSM1281640) and the thyroid cancer group comprising data from five thyroid carcinoma groups (GSM1281636, GSM1281637, GSM1281641, GSM1281642, and GSM1281643). We performed Affymetrix Human Genome U133 Plus Version 2.0 Array (GPL570) analysis using the Affymetrix Transcriptome Analysis Console (both from Affymetrix, Santa Clara, CA, USA). Differentially expressed miRNAs/mRNAs were considered those that met the following criteria: P < 0.05 and log2|fold change (FC)| > 1.5. We drew a heatmap and a volcano plot using the results of differentially expressed miRNA/mRNA analysis. We used data from TCGA-THCA to analyze the disregulated mRNAs and the TargetScan database (version 7.2; http://www.targetscan.org/vert_72/) to investigate potential downstream miRNA targets.
Pathway enrichment analysis.
We performed pathway enrichment analysis of differentially expressed mRNAs using the KEGG and the R software package cluster filter (version 3.10.1; https://guangchuangyu.github.io/software/clusterProfiler/).
Cell culture
The human thyroid cell line KAT18, human papilloma thyroid cancer cell line TPC-1, human follicular thyroid cancer cell line FTC-133 derived from lymph node metastasis of follicular thyroid cancer, human thyroid cancer papillary cell line BCPAP, and Nthy-ori 3 − 1 control cells were from ProCell (Wuhan, China), and were grown in RPMI-1640 (Gibco, MA, USA) with 10% FBS (Gibco), streptomycin (100U/mL), and penicillin sodium (100U/mL). The culture conditions were 37°C and 5% CO2. IGF-1) treatment, cells were grown in media containing 400 ng/mL IGF-1 (Sigma-Aldrich, MO, USA) at 37°C for 24 h. Following IGF-1 treatment, the milieu was substituted with fresh milieus. Cells were tested for mycoplasma contamination approximately once a month using the MycoAlert Mycoplasma Detection Kit (Cat. No. LT07-218; Lonza Cologne GmBH, Cologne, Germany).
Transfection
MiR-1246 mimics (50nM) and it’ s negative control (NC; 50nM) or overexpressed plasmid PIK3AP1 (ov-PIK3AP1; 100nM) and it’s negative control (ov-NC; 100nM) were transfected into TPC-1 and FTC-133 cells at 37˚C utilizing Lipofectamine® 2000 (Invitrogen) for 4 h following the instructions of the manufacturer. Their sequence is as follows: miR-1246: 5′-AATGGATTTTTGGAGCAGG-3′; negative controls (scramble): 5′-AGCGTGGTGTGAATGTATA-3′. Construction of PIK3AP1 overexpression vector: cDNAs of the PIK3AP1 coding region were cloned into the pcDNA3.1 vector (Invitrogen) using the EcoRI and NotI sites. All RNA and plasmids were prepared through Sangon Biotech Co., Ltd (Shanghai, China).
Cell proliferation assay
TPC-1 and FTC-133 cells were digested with trypsin, inoculated in 96-well plates (3×103/well) for 24–72 h. Absorbance at 450 nm was then assessed every 24 h with 10 µl of the CCK-8 reagent (Solarbio, Beijing, China) as per the provided directions to calculate the logarithmic growth of cells.
Cell apoptosis assay
The assay of cell apoptosis was performed by applying an Annexin V–FITC Apoptosis Detection Kit (BD Biosciences, CA, USA) based on provided directions. The cells of TPC-1 and FTC-133 were collected, washed two times in cold PBS, resuspended in 500 µL binding buffer, and stained by using 5 µL each of PI and Annexin V–FITC for 15 min at 23 ± 2°C protected from light. A flow cytometer was then used to assess apoptosis (FCM; BD Biosciences).
Transwell assay
The invasion and migration of TPC-1 and FTC-133 cells were measured by utilizing 24-well Transwell inserts (BD Biosciences). For the invasion assay, these inserts were initially coated for 6 h using a layer of Matrigel® (BD Biosciences) at 37˚C. Harvested cells (1x105) in serum-free milieu were placed in the top chamber, while RPMI-1640 with 10% FBS was added to the lower chamber. After 2 days of incubation, invasive cells were fixed using anhydrous ethanol for 20 min and stained using 0.1% crystal violet (Solarbio) for 15 min at 23 ± 2˚C. Cells were then counted using a light microscope (Olympus Corporation; magnification, x200).
RT-qPCR
We extracted total RNA from TPC-1 and FTC-133 cells with TRIzol (Invitrogen) as per provided directions, after which a Kit of PrimeScript RT Reagent (TaKaRa, Dalian, China) was employed to prepare cDNA, and RT-qPCR was conducted utilizing a SYBR Premix ExTaq II kit (Invitrogen) and an ABI 7500 system (Applied Biosystems, CA, USA). The PIK3AP1 and GAPDH primers were: PIK3AP1: forward, 5′-TCATCGTCTACAGCCCGGAT-3′, reverse, 5′-TCAGTATCTTCTGGCTGCGG-3′; GAPDH: forward, 5′-GCTCATTTGCAGGGGGGAG-3′, reverse, 5′-GTTGGTGGTGCAGGAGGCA-3′. GAPDH was used as a reference for calculating PIK3AP1 expression. The primers for miR-1246 and U6 were as follows: miR-1246: forward, 5′-ACACTCCAGCTGGGAATGGATTTTTGG-3′, reverse, 5′-CTCAACTGGTGTCGTGGA-3′; and U6: forward, 5′-CTCGCTTCGGCAGCACA-3′, reverse, 5′-AACGCTTCACGAATTTGCGT-3′. U6 was employed as an internal reference for calculating the levels of miR-1246. Relative miR-1246 and PIK3AP1 expression were determined using the 2−ΔΔCt method [26].
Western blotting
The lysis of TPC-1 and FTC-133 was carried out utilizing chilled lysis buffer (Solarbio), and the protein levels were assessed via BCA assay kit (Solarbio). We separated denatured proteins (20 µg) using 10% SDS-PAGE and transferred them onto membranes of PVDF (Millipore, Sigma-Aldrich). Following the blocking with 5% BSA Blocking Buffer (Solarbio), the incubation of the membranes was carried out at 4°C during the night hours with the following diluted primary antibodies: PIK3AP1 (1:1,000, ab237629, Abcam, Cambridge, UK), AKT (1:500; ab8805; Abcam), p-AKT (1:500, ab38449, Abcam), PI3K (1:1000, #4257, CST, MA, USA), and p-PI3K (1:1,000, #4228, CST). After washing thrice in TBST (10 min/wash; Solarbio), blots were probed with secondary HRP-goat-anti-rabbit IgG (1:10,000, ab205718; Abcam) for 2 h at 23 ± 2°C. Proteins were discovered via improved chemiluminescence (ECL; Thermo Fisher Scientific), and images were obtained with an imaging system (DNR Bio-Imaging Systems Ltd., Mahale HaHamisha, Israel). GAPDH (1:10,000, ab181602, Abcam) was used as the loading control.
Dual-luciferase reporter assay
The amplified 3′-UTR fragments of PIK3AP1 were cloned into the psi-CHECK-2 vector (Promega Corporation). Subsequently, 293T cells (2x104, CRL-11268, American type culture collection) were transfected with 50 nM miR-1246 mimics or inhibitor, 100 nM of their negative controls, 0.5 µg of psi-CHECK-2 luciferase reporter vector comprising the wild-type (WT) or mutant (Mut) 3′-UTR sequences of PIK3AP1, or empty vector, with Lipofectamine® 2000 (Invitrogen) based upon provided directions. At 48 h post-transfection, the luciferase activity was assessed utilizing the Dual-Luciferase Reporter Assay System (GloMax; Promega), with Renilla luciferase activity serving as a normalization control.
Animal experiments
Twenty-four 8-week-old female BALB/c-nu/nu specific pathogen-free (SPF) mice with a weight of 13–15 g obtained from the Guangdong Medical Laboratory Animal Center (license, No. SCXK, Guangdong, 2016-0041) were used in this study. Mice were placed in a room with a constant temperature and humidity (temperature: 23 ± 2˚C, humidity: 50 ± 10%), under a 12 h cycle of dark/light with free access to standard rat food and water in a polystyrene cage. All animal assessments were confirmed through the Animal Care and Use Committee of Sun yat-sen university and conducted conforming to the National Institutes of Health protocols.
The vital characteristics, food intake and body weight of the experimental mice were normal. Following terminal anesthesia using i.p. injection of 3% pentobarbital sodium (100 mg/kg), a subcutaneous tumorigenesis model was established by injecting 1×106 TPC-1 and FTC-133 cells transfected with miR-1246 agomir and agomir NC into the subcutaneous side of nude mice. Furthermore, a total of 1×106 TPC-1 and FTC-133 cells transfected with miR-1246 agomir and agomir NC were injected into the tail vein of nude mice. Following a 6-week period, an intraperitoneal injection of 3% pentobarbital sodium (200 mg/kg) was used to euthanize these mice. The lung and tumor tissues were quickly stripped and then fixed in 10% neutral formalin solution. Tumor volume was calculated as: Tumor volume = A (maximum diameter) × B2 (vertical diameter) × 0.5. miR-1246 agomir and miR-1246 agomir NC (scramble) were synthetic by Ribobio Co., Ltd (Guangzhou, China)
Hematoxylin and eosin (H&E) staining
Lung tissue was paraffin-embedded and sliced into 5 mm thick sections, fixed on a glass slide, and dried. As per provided directions, the sections were stained with the HE staining reagent (Solarbio, Beijing, China). The slices were soaked in xylene and a gradient concentration of ethanol, soaked in hematoxylin, and finally sealed with resin. Subsequently, the lung tissue sections were observed and imaged via light microscope (Olympus, Tokyo, Japan).
Statistical studies
Data are means ± standard deviation (SD) from triplicate evaluations, and were assessed by SPSS 22.0 (IBM Corp., NY, USA). We performed statistical analysis using one-way ANOVAs and Dunnett’s post hoc test. For independent two-group analyses, Student’s t-tests were used.