TCGA data mining and analysis tools
RNA-sequencing data (level 3) in raw form and related clinical data regarding HNSCC were derived online from The Cancer Genome Atlas (TCGA) dataset (https://portal.gdc.cancer.gov/), which contained 502 HNSCC cases and 44 normal cases. Analysis and visualization of the genes’ expression was conducted with R software (version 4.0.3) packages ‘ggplot2’ and ‘pheatmap’. The GEPIA website tool (http://gepia2.cancer-pku.cn/#index) was used to conduct the OS analysis. Analysis and visualization of the Kaplan-Meier survival analysis was performed with the R software packages ‘survival’ and ‘survminer’. Multivariate cox regression analysis was conducted in order to determine the fitting terms, which were used for building the nomogram. The ‘forestplot’ package, found in R, was used to identify each variable’s P value, hazard ratio (HR), and 95% confidence interval (CI). To predict the 1-, 3-, and 5-year overall recurrence, we developed a nomogram in conformity with the results of the multivariate Cox proportional hazards analysis. The nomogram is a visual representation of factors, and can be used for calculating the risk of recurrence for a distinct patient through summing up the points correlated to each risk factor in the ‘rms’ package of R. The two-gene correlation map was constructed with the ‘ggstatsplot’ package in R software based on Spearman’s correlation analysis.
HNSCC patients and clinical samples
For this study, 20 paired fresh HNSCC tissues and normal adjacent tissues (NATs), and 20 fresh HNSCC tissues with or without lymphatic metastasis (consisting of 10 cases with lymphatic metastasis and 10 cases without lymphatic metastasis) were retrieved from HNSCC patients who received surgery at the Department of Otorhinolaryngology of the First Affiliated Hospital of Chongqing Medical University (Chongqing, China). Fresh tissues were washed with saline after resection, and liquid nitrogen was used to freeze them immediately after which they were kept at -80 °C for total protein and RNA extraction. Furthermore, in total 78 HNSCC tissue specimens embedded in paraffin were collected from the Department of Pathology of the First Affiliated Hospital of Chongqing Medical University. The 78 HNSCC patients underwent surgery between January 2012 and December 2019 in the Otolaryngology department of the First Affiliated Hospital of Chongqing Medical University. The clinical features of the patients are detailed in Table 1. Informed consent was acquired from all patients before surgery. The study was authorized by the Biomedical Ethics Committee of the First Affiliated Hospital of Chongqing Medical University.
Cell lines and cell culture
FaDu cells were obtained from the Chinese Academy of sciences (Shanghai, China). SCC15 cells were the kind gift of Prof. Kai Yang (Chongqing Medical University, China). FaDu and SCC15 cells were both kept in Dulbecco’s modified Eagle’s medium (DMEM, Gibico, USA), to which 10% fetal bovine serum (FBS, PAN-Biotech, Germany) and 1% penicillin-streptomycin was added, in an atmosphere of 37 °C and 5% CO2.
Immunohistochemistry (IHC) staining and scoring analyses
IHC was performed with a detection kit (SP-9000, Beijing, Zhongshan Jinqiao) in accordance with the manufacturer’s instructions. In short, fresh xylene was used to deparaffinize the paraffin sections, which were then hydrated in gradient alcohol. The citric acid buffer was used for antigen retrieval at 90°C−100°C for 30 minutes to which an appropriate amount of endogenous peroxidase blocker was added. The standard goat serum working solution was sealed at room temperature for 15 minutes, and then transferred to a refrigerator of 4°C and incubated with IGF2BP2 and LYVE-1 primary antibodies overnight. On the second day, horseradish enzyme-labeled streptavidin was dripped into the working solution, then incubated for 15 min at room temperature, and rinsed with PBS 3 times for 3 minutes each time. Diaminobenzidine (DAB) was used for color development, followed by hematoxylin staining for 30 seconds, and rinsing with tap water for 5 minutes. Finally, it was dehydrated, transparent and mounted with neutral gum. The IHC stained sections were reviewed and scored independently by two superior pathologists. A final score was then calculated by multiplying the signaling intensity score and the staining distribution score. The signal intensity scores were categorized as follows: 0 (no signal), 1 (weak), 2 (moderate), and 3 (strong). The staining distribution scores were determined according to the percentage of positive cells: 0 (0%), 1 (1–10%), 2 (10–50%), and 3 (51–100%).
Extraction of RNA and quantitative real-time PCR (qRT-PCR)
An E.Z.N.A.® Total RNA Kit I (Omega Bio-tek, USA) was utilized for isolation of the total RNA from HNSCC specimens and cells according to instructions of the manufacturer. Subsequently, an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA) was used to determine the total RNA’s quality through absorbance readings at 260 nm. Reverse transcription of total RNA into complementary DNA (cDNA) was done with a gDNA Eraser (TaKaRa, Japan) using a PrimerScriptTM RT Reagent Kit. Next, for the amplification process an ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and a SYBR Premix Ex TagTM Kit (TaKaRa, Japan) were utilized in accordance with instructions of the manufacturer. The conditions in which the PCR amplification was conducted was as follows: 40 cycles in total at 95°C for the duration of 30 s, then 95°C for 5 s, and lastly 60°C for 1 min. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) functioned as internal control. The 2-ΔΔCt method was applied for analysis of the results, which were presented as relative expression. All experiments were performed three times to independently verify the findings. The primers for PCR analysis are listed in Additional file 1: Table S1.
Protein extraction and western blotting analysis
Total proteins from HNSCC specimens and cells were derived with a protein extraction kit (KGP250, KeyGen, Jiangsu, China). A bicinchoninic acid (BCA) protein assay kit (P0010S, Beyotime, Shanghai, China) was utilized for the detection of the concentration of protein. Next, protein extracts were diluted in 5×loading buffer and boiled for 10 min. The boiled proteins were subjected to 10% SDS-PAGE, placed onto polyvinylidene fluoride (PVDF) membranes, and blocked with 5% nonfat dry milk for the duration of 2 h. Subsequently, we incubated the membranes overnight at 4 °C with primary antibodies, of which the details are shown in Additional file 2: Table S1. The next day, horseradish peroxidase (HRP)-conjugated secondary antibodies were used for incubating the membranes for 1h in total at room temperature. Finally, an enhanced chemiluminescence (ECL) kit (12043-D10, Advansta, USA) was utilized to visualize the protein blots, of which the images were taken with the ChemiDoc Touch Imaging System (Bio-Rad, USA) and analyzed by ImageJ software (version v1.8.0).
siRNA and cell transfection
To reduce the off-target effect of the siRNAs, three independent siRNAs targeting IGF2BP2 and a negative control siRNA were constructed and generated by GenePharma (Shanghai, China). In brief, HNSCC cells were seeded, at a concentration of 2×105 cells/well, in six-well plates with culture medium, and incubated till 60%−70% confluence. Then, Lipofectamine iMAX Reagent (Invitrogen, USA) was utilized for transfection of the cells with siRNAs at a ratio of 1:3 and incubated for 24h in serum-free medium. 24h later, we discarded the transfection medium, and fresh complete culture medium without siRNA and Lipofectamine iMAX was added. At the timeframe of 48 h or 72 h following transfection, we collected the cells for protein or RNA extraction to conduct RT-qPCR and western blotting to validate the transfection efficiency. The IGF2BP2 target and negative control sequences are listed in Additional file 1: Table S1.
Lentivirus vector and cell infection
For knockdown, short hairpin RNA (shRNA) of human IGF2BP2 was cloned into a hU6-MCS-Ubiquitin-firefly_Luciferase-IRES-puromycin lentiviral vector (GV344, Genechem, Shanghai, China). To achieve overexpression, human full-length cDNA of IGF2BP2 was cloned into a Ubi-MCS-3FLAG-CBh-gcGFP-IRES-puromycin lentiviral vector (GV492, Genechem, Shanghai, China). The target sequences of IGF2BP2 have been detailed in Additional file 1: Table S1. When HNSCC cells had reached the logarithmic growth phase, the cells were seeded in a six-well plate at a concentration of 5×104 cells/well and incubated until roughly 30% confluence was reached. Then, the cells were infected with shRNA or overexpression lentiviral vectors with HiTransG A infection-enhancing solution according to the multiplicity of infection (MOI, MOI=10). After incubating for 16h of at 37°C, the viral medium was discarded and substituted by fresh culture. Then, the cells were screened in culture medium containing puromycin (2 μg/ml) for one week and stably silenced- and overexpressed-IGF2BP2 HNSCC cells were generated.
Wound-healing and Transwell assays
Wound-healing and transwell assays were carried out in vitro for the detection of cell migration and invasion. The detailed procedures of these assays have been described in our previous study [44].
Popliteal lymphatic metastasis model in vivo.
Male BALB/cA nude mice aged 4−6 weeks and with a weight between 18 and 22 g were obtained from Huafukang Biotechnology Co., (Beijing, China), and accommodated in specific pathogen free (SPF) barrier facilities. To construct the metastasis model, 5×106 FaDu cells were transfected with sh-IGF2BP2-luc and sh-NC-luc, suspended in 60 μl PBS, and then injected into the footpads of the mice. Six weeks after injection, mice were subjected to bioluminescence imaging to evaluate lymphatic metastasis. For bioluminescence imaging, mice were anesthetized by inhaling 2% isoflurane for approximately 5 min, injected intraperitoneally with D-Luciferin potassium salt (200 μl, 150 μg/ml, ST196, Beyotime, Shanghai, China), and imaged with a bioluminescence system (NightOwl II LB983, Berthold Technologies, Germany). All the primary tumors and popliteal LNs were harvested and embedded in paraffin for IHC analysis. The LN volumes were calculated based on this formula: LN volume (mm3) = length × width2 × 0.5. All the experiments were authorized by the Laboratory Animal Use Management Committee of the Experimental Animal Center of Chongqing Medical University.
Immunofluorescence
FaDu and SCC15 cells seeded on sterile coverslips were fixated with 4% paraformaldehyde for the duration of 20 min, then permeabilized with 01% Triton X-100 for 15 min, and lastly blocked with goat serum at room temperature for a total of 30 min. Subsequently, primary antibodies were utilized for incubation of the cells overnight at 4°C, of which the details are shown in Additional file 2: Table S1. The following day, fluorescence-labeled secondary antibodies were utilized incubation of the cells for 1h at room temperature. DAPI (C1006, Beyotime, Shanghai, China) was used to stain nuclear DNA and the images were captured with a confocal system (Nikon).
RNA binding protein immunoprecipitation (RIP)
HNSCC cells at 90% confluence, cultured in 10-cm plates (approximately 1.2×107 cells each plate), were accumulated and then lysed with IP lysis buffer (P0013J, Beyotime, Shanghai, China) supplemented with protease inhibitor (100×) and RNase inhibitor (40U/μl) on ice for the duration of 30 min. The cell lysates, pipetted up and down several times, were stored in -80°C for 5 min, then allowed to thaw on ice, and centrifuging at 12,000g for 10 min. Then the cell lysates were divided into two parts, one was saved as input group for the whole cell extraction, and the other was used for the following immunoprecipitation (IP) treatment (IP group). For the IP group, cell lysates were incubated with 5μg anti-IGF2BP2 (ab128175, Abcam, USA) or IgG antibody (14678-1-AP, Proteintech, Wuhan, China), respectively, which was alternated continually throughout the night at 4°C. Protein A/G magnetic beads (Bimake, China) were rinsed five consecutive times with 0.1% Tween-20 in PBS, and then mixed with cell lysate-antibody complexes and rotated continuously at 4°C for 6h. Next, the RNA-protein complexes were rinsed five consecutive times with elution buffer (containing 150 mM NaCl, 50 mM Tris-HCl, 5 mM EDTA, 0.5 mM DTT, 0.5% NP-40, 10% SDS, and RNase inhibitor) and treated with proteinase K at 55°C for 1h. Bound RNAs were extracted and subjected to RT-qPCR for quantitative analysis. Relative enrichment was normalized to the input. The primers for RT-qPCR were showed in Additional file 1: Table S1.
MeRIP-qPCR
Total RNA was extracted as described above. Ten percent of the total RNA was reserved for the input control and the remaining RNAs were used for m6A-IP. Anti-m6A antibody (ab151230, abcam, USA) or mouse IgG were immobilized on magnetic beads using the DynabeadsTM Antibody Coupling Kit (14311D, Invitrogen, USA) according to the manufacturer’s instruction. Subsequently, total RNA was incubated with antibody-conjugated beads in 500μl binding buffer (containing 140 mM NaCl, 50 mM Tris-HCl, 5 mM EDTA, 0.5% NP-40, and RNase inhibitor) and rotated continually for 4 h at 4°C. M6A-modified mRNAs were eluted from the beads with elution buffer and then purification was carried out for further analysis by RT-qPCR. Relative enrichment was normalized to the input. The primers for RT-qPCR were showed in Additional file 1: Table S1.
Luciferase reporter assay
2×105 HNSCC cells were first seeded in plates with 24-wells, followed by culturing for 24h. Then plasmids carrying wild-type or mutated-type Slug CDS were transfected into the cells. After transfection for 12h, cells were re-seeded into a plate with 96-wells and then for 24h incubated. The Dual-Luciferase® Reporter Assay System (E1910, Promega, USA) was used to analyze the luciferase activities. Renilla Luciferase (R-luc) was used to achieve normalization of the firefly luciferase (F-luc) activity.
mRNA stability assay
HNSCC cells were seeded in plates with six-wells and grown to approximately 50% confluence following incubation for 24h. Then, actinomycin D (5 μg/ml, Sigma, USA) was used to treat the cells, which were harvested at 0h, 3h, and 6h. The total RNA was derived and then analyzed with qRT-PCR. The mRNA value of each group was calculated and normalized to GAPDH at the designated time. The degradation rate of mRNA was approximated based on previously published protocols [45].
Statistical Analysis
All the statistical analysis were conducted with GraphPad Prism (version 7.0, GraphPad software, USA) and SPSS 21.0 software (IBM, SPSS Statistics, USA). The correlation between IGF2BP2 expression and clinicopathologic parameters were analyzed with the Chi-Square test. Univariate and multivariate regression analyses were employed with the Cox proportional hazards model to identify independent factors affecting the survival of HNSCC patients. The Kaplan-Meier method was carried out to produce survival curves, and any significant differences in survival probability between groups were compared by log-rank statistics. We applied the two-tailed Student’s t test to compare results between two different groups, and the one-way ANOVA was conducted for multiple comparisons. All the data are presented in the form of mean ± SD and are the result of no less than 3 independently performed experiments. Statistical significance was determined as values of P<0.05. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns: not statistically significant.