2.1 Public databases

EGR-1 expression datasets for head and neck SCC (HNSCC) patients and healthy individuals were acquired from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov), which encompass data from 522 HNSCC cases and 43 controls. Furthermore, the Cancer Cell Line Encyclopedia (CCLE) database (https://sites.broadinstitute.org/ccle/), comprising of cancer cell lines, was also utilized.

2.2 Cell culture

The primary human HNSCC cell line, which included OSCC cells, encompassed HSC2, HSC3M3, koscc33A, SCC9, HOC313, KOSCC25B, and HOC719NE cells. HSC2, HSC3M3, SCC9, HOC19, and KOSCC25B cells were derived from tongue SCC. Further, kosc33A and HOC313 cells were established from SCC of maxillary sinus and floor of mouth, respectively. HSC2 and HSC3M3 cells were sourced from the Japanese Collection of Research Bioresources (JCRB) Cell Bank, Osaka, Japan. SCC9 cells were obtained from the American Type Culture Collection (ATCC). HOC313 and HOC719NE cells were provided by Prof. Kamata (Hiroshima University, Hiroshima, Japan) and koscc33A and KOSCC25B cells were obtained from Dr SP Hong (Seoul National University, Seoul, South Korea). HSC2, HSC3M3, koscc33A, and KOSCC25B cells were cultured in Dulbecco’s modified Eagle medium (DMEM) (Wako Pure Chemical, Osaka, Japan), while others were cultured in a 1:1 mixture of DMEM and Ham’s F12 medium (Wako Pure Chemical), supplemented with 10% fetal bovine serum (Nichirei Biosciences, Tokyo, Japan). Culturing conditions comprised 5% CO2 in air at 37 °C.

2.3 The quantitative reverse transcription polymerase chain reaction (qRT-PCR)

Total RNA was isolated from each cell using the FastGene RNA extraction kit (Nippon Genetics, Tokyo, Japan). Subsequently, cDNA was synthesized using the ReverTra Ace qRT kit (Toyobo, Osaka, Japan), employing 1 μg of total RNA as a template. qRT-PCR targeting EGR-1, cyclin-dependent kinase 1 (CDK1), CDK2, CDK4, cyclin A, cyclin D1, cyclin E2, matrix metalloproteinase 2 (MMP2), MMP9, and vimentin (Table 1) was conducted using the Thermal Cycler Dice Real Time System II (Takara Bio, Kusatsu, Japan), along with THUNDERBIRD SYBR qPCR Mix (Toyobo). Normalization of gene expression was carried out using glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and comparisons were made by calculating ΔΔCt. Each experiment was performed in triplicate and repeated at least three times.

Table 1 List of primers used in this study2.4 Immunoblotting

Whole cell lysates were obtained using the M-PER mammalian protein extraction reagent (Thermo Fisher Scientific, Rockford, IL, USA). Subsequently, 50 µg of the lysates were utilized for immunoblotting, employing 10% SDS-PAGE gels, followed by electrophoresis on nitrocellulose blotting membranes (GE Healthcare Life Sciences). These membranes were then subjected to incubation with anti-EGR-1 antibody (Santa Cruz Biotechnology, CA, USA), succeeded by peroxidase-conjugated IgG (MBL, Nagoya, Japan). The immune complexes were visualized using an ECL Western Blotting Detection System (GE Healthcare, Amersham, UK). To ensure internal control, anti-GAPDH antibody (Santa Cruz Biotechnology, CA, USA) was employed. After electrophoresis, each protein band was quantified using the imaging software ImageJ distributed by the National Institutes of Health (NIH). The expression level of each molecule corrected by the value of GAPDH was determined, and those in the control group were considered as 1.

2.5 Transient transfection of siRNA and the plasmid

Human OSCC cells were plated in 6-well plates at a density of 300,000 cells per well. Following an overnight incubation period, the cells were transfected with 10 nM/well of hsRiEGR1 for EGR-1 (107310823: Integrated DNA Technologies, Coralville, USA) and siRNA Universal Negative Control (Sigma-Aldrich). For the overexpression of EGR-1, pMXs (Cell Biolabs, San Diego, CA, USA) was employed. Transfection was carried out using LipoTrust EX Oligo (Hokkaido System Science, Sapporo) and 100 ng of plasmid. After 24 h of incubation at 37 °C, the cells were harvested and utilized for subsequent analyses.

2.6 Proliferation assay

For the assessment of OSCC cell proliferation, a Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) was employed. OSCC cells were seeded in 96-well plates at a density of 1000 cells/well and maintained at 37 °C. After 24, 72, and 120 h, the absorbance at 450 nm was determined using a TriStar2 LB942 Multimode Plate Reader (Berthold Technologies, Germany).

2.7 Invasion assay

The invasion assay was performed using a Matrigel-coated chamber (BD BioCoat, Corning, USA). Approximately 20,000 cells/well were suspended in 500 μL DMEM and placed in the Matrigel-coated insert, while 500 μL of fresh DMEM was added to the bottom chamber. Following a 24-h incubation period at 37 ℃, the filters were fixed with 4% paraformaldehyde and stained with hematoxylin. Subsequently, the filters were examined under a microscope at 100× magnification, and stained infiltrating cells were counted in four randomly chosen fields of view.

2.8 Wound healing assay

For the wound healing assay, we utilized inserts placed in a 24-well plate. Transfected HSC3M3 cells were seeded into each compartment of the insert. Following a 24-h incubation period, the inserts were removed, and images were captured using a phase-contrast microscope at 4× magnification. The wound area was subsequently measured using ImageJ.

2.9 Cell cycle analysis

Cell cycle analysis was conducted using flow cytometry. Cells from the EGR-1 knockdown or overexpression cell line were fixed with 70% ethanol and stored at −20 ℃. After 24-h incubation, a solution of 50 μL of 50 μg/mL RNase (Nippon gene, WAKO, Japan) diluted with PBS was added and incubated at 37 ℃ for 30 min. Subsequently, the cells were treated with 450 μL of 50 μg/mL propidium iodide (PI) (Sigma Aldrich) and incubated at room temperature for 10 min. The resulting solution was analyzed using a flow cytometer (CyAn ADP, BECKMAN COULTER) and Cell Quest software (BD GmbH, Heidelberg, Germany).

2.10 Tumor specimens

Formalin-fixed and paraffin-embedded surgical specimens were obtained from 50 untreated primary OSCC patients who underwent surgery at the Department of Oral and Maxillofacial Surgery, Kagoshima University Hospital, Kagoshima, Japan. The tumor stage and histological grade were determined based on the Union for International Cancer Control’s TNM classification system, 8th edition (O’Sullivan 2017), and the criteria outlined by the World Health Organization (Sloan et al. 2022), respectively. Patient medical records were retrieved from the hospital’s database. This study involving human samples adhered to the principles of the Declaration of Helsinki and received approval from the Ethics Committee on Epidemiological Studies at Kagoshima University [Approval No. 213223-1].

2.11 Immunohistochemistry

Consecutive 3-µm sections were cut from each block and subjected to immunohistochemistry. Following the deparaffinization process, endogenous peroxidase activity was blocked for 30 min with methanol containing 0.3% H2O2. Antigen retrieval was achieved through microwave treatment in Tris-ethylenediaminetetraacetic acid (EDTA) buffer for 1 h. Protein blocking serum was applied to block nonspecific antibody reactions for 10 min, after which anti-EGR-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted to 0.5 mg/mL was used as the primary antibody. Specimens were incubated overnight at 4 °C, followed by treatment for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibody (Medical & Biological Laboratories, Nagoya, Japan) diluted to 0.5%. Color development was achieved using diaminobenzidine (DAB) solution (DAKO, Carpinteria, CA, USA), and counterstaining was performed with Mayer’s hematoxylin (Sigma-Aldrich, St. Louis, MO, USA).

To evaluate immunohistochemistry, three cancerous and noncancerous areas on the same section were randomly selected, and the average EGR-1 positivity rate was determined using Image J software.

2.12 Statistical analysis

Statistical analysis was performed using Student’s t-test and unpaired t-test. All statistical analyses were conducted using GraphPad Prism 9.3.1 (GraphPad, USA), and p < 0.05 was considered statistically significant.