Clinical samples
One hundred and forty formalin-fixed, paraffin-embedded blocks of OSCC samples and twenty normal samples were collected from the Oncology Department of Oral and Maxillofacial Surgery of The First Affiliated Hospital of Xinjiang Medical University (Urumqi, China) between March 2010 and March 2021. The use of OSCC samples was approved by the ethics review board (approval no. IACUC20180411‑13). The TNM and clinicopathological classification and staging of patients with OSCC were performed according to the American Joint Committee on Cancer (AJCC) guidelines.
Bioinformatics analyses
The GSE87539 [18] and GSE138206 [19] data were downloaded from Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo). The probe ID was converted into an international standard name for gene symbols using perl programming and the gene expression matrix of the two datasets were merged by perl language also. The data normalized were used by R language “sva” and “limma” packages, the differentially expressed genes (DEG) were filtered by “limma” package and using a cutoff criterium of adj. P.Val < 0.05 and |log2(fold change)|> 2. The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were used by R language “colorspace” and “string” packages, and Bioconductor packages(“DOSE”, “clusterProfiler” and “pathview”). The PPI network of DEGs using online database STRING version 11 (https://string-db.org/) and the combining score >0.9 were used for the PPI network construction. The top 30 hub genes were visualized by using Cytoscape software. Gene set enrichment analysis (GSEA) was used to evaluate the gene function between treated and control groups.
Immunohistochemistry (IHC) and immunofluorescence (IF) evaluation
OSCC samples and normal oral mucosa samples were harvested from humans and mice for HE staining and IHC; the methods for HE and IHC were described in our previous study [20]. Samples for IHC and IF were embedded in the optimal cutting temperature (OCT) compound and sectioned into 8-μm sections. IHC and IF were performed with anti‑P. gingivalis (Dia‑An, lnc, 1:200), anti‑CXCL2 (bs‑1162R, BIOSS, 1:500) and anti‑TANs (CD66b+, ab197678, Abcam, 1:500) antibodies. The immune expression of TANs in mice was assessed with an antibody against CD11b+ (bs-1014R, BIOSS, Biotech, Beijing, China, 1:500) [21]. Image-Pro Plus version 6.0 software (Media Cybernetics, Inc., Bethesda, MD, USA) was used to evaluate the IHC expression levels. The IOD/total area of each field was also calculated simultaneously. According to the IOD value, the staining proportions (PS) were divided into: 0 (–), 1 (+), 2 (++), and 3 (+ + +). Four levels: 0 (0%); 1 (1–25%); 2 (26–50%); 3 (51–75%); and 4 (76–100%) were divided by the staining proportion score (PS). The IS multiplied by PS was the final result of the staining score [22]. The expression level were categorized into a low expression group (0–6) and a high expression group (7–12) in this study.
Bacterial and cell culturing
The P. gingivalis ATCC33277 strain was purchased from The American Type Tissue Culture Collection (ATCC), and the bacteria were grown in two culture systems: (1) Columbia liquid medium (purchased from Bei-Na Biotech, Inc., Beijing, China) and (2) Columbia blood agar culture medium (purchased from Bei-Na Biotech, Inc., Beijing, China). The human OSCC cell line TSCCa was provided by the Central Laboratory of First Affiliated Hospital of Xinjiang Medical University. TSCCa cells were maintained in RPMI 1640 (Thermo Fisher Scientific, CA, USA) containing 10% FBS (Thermo Fisher Scientific, CA, USA), penicillin and streptomycin (100 µg/ml) (Thermo Fisher Scientific, CA, USA). The coculture cell model (P. gingivalis and TSCCa) was established with a multiplicity of infection (MOI) of 50. Human TANs were extracted from the peripheral blood of OSCC patients with a human peripheral blood neutrophil isolation kit (Solarbio, Biotech, Inc., Beijing, China). TANs were added to the coculture cell model at a 1:1 ratio.
Fluorescent staining of cell co-culture model
Bacterial staining: P. gingivalis was marked by the fluorochrome of SYTO9 Green Fluorescent Nucleic Acid Stain (Thermo Fisher Scientific, CA, USA). Nuclear staining was performed with DAPI (Bioss, Biotech, Inc., Beijing, China). Cellular mitochondrial staining was marked by MitoTracker red (Beyotime Biotech, Inc., Beijing, China). The coculture cell model was established at an MOI of 50 in a 37 °C incubator for 48 hours, and confocal laser photography was performed.
Neutrophil chemotaxis assay
We used Transwell chambers (Corning, Bioscience, NY, USA) with 5-µm polycarbonate membranes. Briefly, TANs were treated with the supernatant of different groups, and the supernatant was placed into the bottom chamber. TANs suspended in RPMI 1640 containing 2% FBS (1×105 cells/100 µL) were added to the upper wells and incubated for 2 hours in a 37 °C incubator. We collected TANs that migrated to the lower chamber and counted them after staining with crystal violet.
Cell Counting Kit-8 (CCK-8) assay
Cell proliferation was evaluated by CCK-8 assay (Keygen Biotech, China) following the manufacturer’s instructions. Ten microlitres of CCK-8 were added to each well. The 96-well plates were incubated for 2 hours at 37 °C, and wells with only RPMI 1640 medium were used as blank controls. The absorbance of the coloured solution was quantified at 450 nm using a microplate reader (Tecan, Untersbergstrasse, Austria).
Invasion assay
Basement Matrigel membrane (BD Biosciences, CA, USA) was used to precoat the Transwell chamber with a filter with 8‐μm pores (Corning, Bioscience, NY, USA); 2 × 105 cells from 5 different groups were suspended in FBS-free medium and seeded on the upper chamber. In the lower chamber, medium containing 10% FBS was added. Forty-eight hours later, the medium was aspirated, and the cells on the upper chamber were wiped away. The cells on the lower surface of the filter were counted after staining with crystal violet.
Enzyme-linked immunosorbent assay (ELISA)
To detect the level of CXCL2 in cell culture supernatants, we used the corresponding Quantikine Human ELISA Kit (Boster Biotech, Inc., Wuhan, China) in accordance with the manufacturer’s instructions. The absorption was measured at 450 nm on a Thermo Scientific microplate reader. The assay can determine the levels in unknown samples according to the standard curve, which is made based on the absorption values of the standard sample.
OSCC mouse model
C57BL/6 mouse OSCC models were created with SCC7 cell lines, and all animal protocols were approved by the Animal Care Committee. The mice were divided into 4 groups, the treatment groups were injected with different doses of SCH527123 in the tumour cell site at 4, 8, 12 and 16 days. The mice were sacrificed at 10 or 20 days. The tumour tissue was removed for volume calculation, IHC and IF studies.
Transfection of lentiviral vectors
Recombinant lentiviruses named sh-CXCL2 and sh-CXCR2 were used to interfere with CXCL2/CXCR2 axis expression, and two empty carrier lentiviruses were used as negative controls. The lentivirus was purchased from GeneChem Biotech, Inc. (Shanghai, China). The cells were incubated with lentivirus for 48 hours and screened with puromycin.
RNA isolation and quantitative real-time PCR
Total RNA was isolated from cells according to the TRIzol method (Invitrogen, Carlsbad, CA) in accordance with the manufacturer’s instructions. We measured mRNA expression in cells via qRT-PCR. qRT-PCR was performed using the SYBR Prime Script RT-PCR Kit (Thermo Fisher Scientific, CA, USA) in accordance with the manufacturer's instructions. We used GAPDH as an internal control. The primers used are listed in Table 4.
Western blot analysis
The same amounts of protein samples (40 μg each) were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis for 150 min at 75 V, and then the proteins were transferred to polyvinylidene difluoride (PVDF) membranes for 2 hours at 100 V; the membranes were incubated in blocking buffer (5% milk in 1× tris‐buffered saline [TBS]) for 1 hour at RT. The PVDF membranes were rocked in 1:1,000 primary antibody solution for at least one overnight at 4 °C. Then, the PVDF membranes were incubated in 1:1,000 secondary antibody solution for 1 hour at RT. The results were visualized with chemiluminescent HRP substrate (Millipore, MA). Primary antibodies against P. gingivalis (Dia‑An, lnc, it was freshly prepared, 1:1000), CXCL2 (bs‑1162R, BIOSS, Biotech, Beijing, China, 1:2000), CXCR2 (bs‑1629R, BIOSS, Biotech, Beijing, China, 1:2000), JAK1 (bs‑1439R, BIOSS, Biotech, Beijing, China, 1:2000), STAT3 (bs‑55208R, BIOSS, Biotech, Beijing, China, 1:2000), p-JAK1 (bs‑3238R, BIOSS, Biotech, Beijing, China, 1:2000), p-STAT3 (bs‑1658R, BIOSS, Biotech, Beijing, China, 1:2000), E-cadherin (bs‑10009R, BIOSS, Biotech, Beijing, China, 1:2000), N-cadherin (bs‑1172R, BIOSS, Biotech, Beijing, China, 1:2000), and GAPDH (bs‑13282R, BIOSS, Biotech, Beijing, China, 1:2000) and secondary antibodies (anti‐rabbit‐HRP, Cell Signalling Technology, Inc, Boston, USA) were used in this study.
Statistical analysis
All data were expressed as mean ± SE of three independent experiments and were analysed using Student’s t-tests or one‐way analysis of variance as indicated. For analysis of clinical data, counting data were expressed by frequency and composition ratio, chi-square test was used for comparison between groups, COX regression were used to analyze the correlation between clinical indicators and death. Statistical significance was defined as p < 0.05 (*p < 0.05, **p < 0.01, and ***p < 0.001).