OSCC tissues were obtained from 29 patients (including 14 males and 15 females) between October 2021 and March 2023 in Jining No. 1 People’s Hospital (Shandong, China) with written informed consent as a prospective study. The inclusion criteria were as follows: 1) All tissues were confirmed by pathological examination, and radiotherapy or chemotherapy and other forms of antineoplastic therapy were not given before surgery. 2) Patients were between 40 to 65 years old. 3) According to the Union for International Cancer Control (UICC) standards in 2002, all the cases were classified as phases I to IV. Patients with an unclear diagnosis or who had radiotherapy or chemotherapy prior to surgery were excluded. In addition, 10 cases of normal oral tissue samples (aged 35 to 65, including 6 males and 4 females) were taken from patients with oral trauma, glossoplasty surgery and some non-neoplastic oral diseases were collected as control. The human material was performed in accordance with the Declaration of Helsinki and the experimental procedures were approved by the Research Ethics Committee of Jining No. 1 People’s Hospital (2021-050).

Human oral squamous cell carcinoma cell lines HSC3 and HSC6 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and authenticated by short tandem repeat typing. DNA staining with Hoechst 33258 has been done to detect mycoplasma. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 100 μg/ml penicillin and 100 μg/ml streptomycin at 37˚C in a humidified 5% CO₂ atmosphere. The miR-221/222 inhibitor and the empty vector were chemically synthesized by Gene Pharma Co., Ltd (Shanghai, China), and cell transfection was performed by the transfection system at 100 nm of miRNA inhibitor concentrations. Scramble sequences were used as negative control at the same concentration as miR-221/222 inhibitor. Cells were collected to be used for further analysis at different time points after transfection.

Total miRNAs were extracted by means of the miRNeasy Mini Kit (Qiagen GmbH, Hilden, Germany), and mRNAs were extracted by means of the Total RNA Purification Kit (Qiagen GmbH) from HSC3 and HSC6 cells according to the manufacturer’s instructions. Real-time qRT-PCR analysis was performed to validate miRNA and mRNA expression with the one-step RT-PCR kit. GAPDH (Invitrogen; Thermo Fisher Scientific, Inc.) and U6 were used as endogenous controls.

After treatments for 48 h, HSC3 and HSC6 cells were washed three times with cold phosphate-buffered saline (PBS) and subjected to lysis in a buffer composed of RIPA (Beijing, China) and PMSF (Beijing, China) (100:1). Total proteins were heated at 95˚C for 10 min after the protein concentration was measured by means of the Enhanced BCA Protein Assay Kit (Shanghai, China). Equal amounts of heated proteins from each sample were separated by 12% SDS-PAGE gel (Beijing, China) and transferred to polyvinylidene difluoride (PVDF) membranes (Beijing, China). The blots were blocked in TBST containing 5% non-fat milk for 1 h at room temperature and incubated with specific primary antibodies against PUMA, Bcl-2, Bax, Bak and GAPDH (Cell Signaling Technology, Inc.) overnight at 4˚C. Subsequently, the blots were rinsed three times with TBST and then incubated with the secondary antibody at room temperature for 1 h. The protein bands were imaged by the AlphaView SA western blot detection system (Carl Zeiss AG, Oberkochen, Germany) and quantified after normalization with the density of GAPDH (Cell Signaling Technology, Inc.) with Image J instrument software (National Institutes of Health, Bethesda, MD, USA). Each test was repeated in triplicate.

After different treatments for 48 h, HSC3 and HSC6 cells were washed with PBS, fixed in 4% paraformaldehyde, permeabilized with 1% Triton X-100 and sequentially blocked in 1% BSA. Then the treated cells were incubated with primary antibodies (PUMA, Bcl-2, Bax and Bak) (Cell Signaling Technology, Inc.) and the corresponding secondary antibody according to the manufacturer’s protocols. DAPI was used to stain nuclei. Images were captured by confocal laser scanning microscopy. Each test was repeated in triplicate.

Cell proliferation was measured by means of cell counting kit-8 (CCK-8) (Beijing, China) according to the manufacturer’s instructions. After treatment with miR-221/222 inhibitor, HSC3 and HSC6 cells were seeded into 96-well plates at a density of 3000 cells/well and then allowed to attach overnight. After transfection, a 10-μl quantity of CCK-8 was added to each well at 0, 24, 48 and 72 h, and the cells were incubated for 1 h. Optical density (OD) was measured at a wavelength of 450 nm by spectrophotometric analysis. Each experiment was repeated in triplicate.

The wound-healing assay was used to detect HSC3 and HSC6 cell migration. Twenty-four h after transfection, linear scratch wounds were created by means of a 200-μl pipette tip when cells grew to near-100% confluence in 6-well plates. Cells were maintained in serum-free medium to prevent cell proliferation. Images were taken at 0, 24, 48 and 72 h after scratching for visualization of wound healing and migrating cells. At least three areas were selected randomly from each well, and the data were derived from quintuplicate samples of three independent experiments.

After treatment, HSC3 and HSC6 cells were transferred to 8-μm-pore inserts and then placed in companion wells containing DMEM medium and 10% FBS. The inserts were removed after a 12-hour incubation, and the non-migrating cells on the upper surface were harvested. HSC3 and HSC6 cells on the lower surface were fixed and stained and counted under a microscope. Each test was repeated in triplicate.

HSC3 and HSC6 cells with different treatments were washed, harvested and stained with propidium iodide (PI, 20 μg/ml) (Invitrogen; Thermo Fisher Scientific, Inc.) for 30 min at room temperature. Then the samples were analyzed by flow cytometry (Beckman Coulter GmbH, Krefeld, Germany), and the data were elaborated by Modfit software. Each test was repeated in triplicate.

HSC3 and HSC6 cells were washed 3 times with PBS and resuspended in buffer at a concentration of 10⁶/ml. Then the cells were mixed with 5 μl FITC-conjugated Annexin V reagent and 5 μl propidium iodide. After 15 min of incubation at room temperature in the dark, the samples were analyzed by flow cytometry (Beckman Coulter GmbH, Krefeld, Germany). Each test was repeated in triplicate.

The 293T cells were co-transfected with 0.5 μg pMIR vectors containing PUMA 3’UTR (Shanghai, China) or mutPUMA 3’UTR (Shanghai, China) and 20 μM miR-221/222 mimics (Shanghai, China) or 20 μM miR-221/222 inhibitor (Shanghai, China) by Lipofectamine^TMRNAiMAX (Invitrogen; Thermo Fisher Scientific, Inc.). Cells underwent lysis with Passive Lysis Buffer (Beijing, China) and were collected at 48 h post-transfection. The luciferase activity was detected with the dual luciferase reporter assay kit (Beijing, China) according to the manufacturer’s protocol. Renilla luciferase activity was used for normalization.

BALB/c-nu mice were selected for this study, and the experimental procedures were approved by the Research Ethics Committee of Jining No. 1 People’s Hospital (2020-053). All BALB/c-nu mice used were maintained and handled under regulatory and institutional guidelines and surgical procedures. HSC3 and HSC6 cell suspensions (5 × 10⁶) were injected subcutaneously into the interscapular area of the BALB/c-nu mice, and 30 mice in total were used in the study. The animal treatment was initiated after two weeks following HSC3 and HSC6 cell inoculation, when the xenograft tumors grew to approximately 40 mm³. The treatments were performed by intra-tumor injection with 1 mg/kg per mouse of miR-221/222 inhibitor or inhibitor NC as control every two days for a total of seven injections. For nude mice that needed to be transfected, after the successful administration of anesthesia, the transplanted tumor sites of nude mice were disinfected with 75% alcohol, and needles were injected from 3 places, a 20-μl quantity of transfection reagent was injected into the tumor and a 20-μl quantity of normal saline was injected into the control group. Tumor sizes were assessed before each treatment according to the formula: (maximum diameters × minimum diameters²)/2. An inhibition ratio was calculated by the formula [1-(treated tumor average volume/untreated tumor average volume)] × 100%. After the experiment, the BALB/c-nu mice were euthanized by CO₂ euthanasia. The BALB/c-nu mice were placed in a euthanasia cage box, and CO₂ was filled at a balanced rate according to the CO₂ replacement rate of 30% - 70% container volume/min, so that the BALB/c-nu mice could quickly lose consciousness and minimize pain. Tumors were collected and placed for further analysis.

OSCC tissues and xenograft tumors were fixed and embedded and then cut into 4-μm-thick sections. ISH analysis was performed on deparaffinized sections by means of the ISH tissue implementation kit (Beijing, China). OSCC tissues and xenograft tumor sections were incubated at 60˚C for 1 h and deparaffinized. Then the sections were rehydrated in decreasing concentrations of ethanol and washed in deionized water. The slides were incubated in 0.2 mol/l HCl at room temperature for 5 min and washed 3 times in PBS. Proteins were digested by the addition of pepsin solution at 37˚C for 15 min. Dehydration of sections was performed through 70%, 95% and 100% ethanol at each concentration for 1 min. The miR-221 and miR-222 probes labeled with digoxin were added to the hybrids. Sections were denatured at 90˚C for 4 min, followed by incubation in the ISH slide denaturation and hybridization system at 37˚C overnight. Finally, the slides were observed under an inverted microscope.

Immunohistochemistry was used to detect the expression of PUMA in OSCC tissues and PUMA, Bcl-2, Bax and Bak in xenograft tumors. The sections were deparaffinized with xylene and rehydrated by increased grades of ethanol. After antigen retrieval, primary rabbit polyclonal antibodies to PUMA, Bcl-2, Bax and Bak (Cell Signaling Technology, Inc.) at 1:200 dilution were added for treatment. Then immunohistochemical staining was performed by the recommended protocol.

All the experiments were repeated at least 3 times, and the data analysis was performed with SPSS 19.0 software. The results were presented as means ± standard deviation (SD) with Student’s t-test or one-way analysis of variance (ANOVA). The minimal level of significance was specified as P < 0.05.

The expression of miR-221/222 and PUMA in OSCC tissues and normal tissues was examined by ISH and immunohistochemical staining, respectively. As shown in Figure 1(A), miR-221/222 expression was higher in OSCC tissues than in normal tissues. By contrast, the expression of PUMA exhibited an opposite trend in OSCC tissues compared with the expression of miR-221/222. miR-221/222 and PUMA expression in HSC3 and HSC6 cells were analyzed by qRT-PCR. For knockdown of endogenous miR-221 and miR-222, miR-221 inhibitor and miR-222 inhibitor were synthesized and transfected into HSC3 and HSC6 cells. qRT-PCR analysis showed that miR-221/222 inhibitor transfection significantly reduced the miR-221 (Figure 1(B)) and miR-222 (Figure 1(C)) levels compared with control groups. In contrast, a notable upregulation of PUMA was observed in the miR-221/222 groups (Figure 1(D)). No significant differences between the empty vector transfection groups and the control groups were noted.

miR-221/222 inhibitor transfection altered the expression of apoptotic proteins (PUMA, Bcl-2, Bax and Bak), and the apoptosis-related protein expression was measured by western blot and immunofluorescence. As shown in Figure 2, a significant increase of PUMA was observed in both HSC3 and HSC6 cells in miR-221/222 inhibitor groups. In contrast, Bcl-2 expression in miR-221/222 groups was downregulated relative to that in the control and inhibitor NC

groups. The expression of Bax and Bak was increased as the result of Bcl-2 protein downregulation. Analysis of the data indicated that the miR-221/222 inhibitor could induce HSC3 and HSC6 cell apoptosis through altering the expression of PUMA, Bcl-2, Bax and Bak.

The effect of miR-221/222 inhibitor on HSC3 and HSC6 cell proliferation and invasion was determined by CCK-8 assay, wound-healing assay and transwell assay. As shown in Figure 3, both the proliferation and the invasive abilities of HSC3 cells and HSC6 cells transfected with the miR-221/222 inhibitor were significantly reduced compared with those in the control groups (P < 0.05). There were no obvious differences between the control groups and inhibitor NC groups (P > 0.05).

We assessed whether miR-221/222 inhibitor transfection modulated HSC3 and HSC6 cell apoptosis and cell cycle progress. The results showed that the apoptotic cells were significantly increased in cells treated with miR-221/222 inhibitor compared with control and inhibitor NC groups, suggesting that apoptosis was obviously induced in cells when treated with miR-221/222 inhibitor (Figure 4(A) and Figure 4(B), P < 0.01). As for cell cycle progress, analysis of the data suggested that miR-221/222 silence induced a significant increase in the percentage of G1 phase cells in HSC3 and HSC6 cells (Figure 4(C) and Figure 4(D)).

PUMA 3'UTR and mutPUMA 3’UTR luciferase constructs were transfected into 293T cells with miRNA mimics NC, miRNAs inhibitor NC, miR-221/222 mimics or miR-221/222 inhibitor to validate whether PUMA was a direct target of miR-221/222. Luciferase activity was measured by means of a dual-luciferase reporter assay system. Figure 5 showed that, compared with other groups, the luciferase activity of 293T cells transfected with PUMA 3’UTR and miR-221/222 mimics was reduced significantly (P < 0.05).

In the previous experiment, we demonstrated the effect of miR-221/222 inhibitor on OSCC cells in vitro. Thus, we further evaluated the effect of miR-221/222 silence on the growth of OSCC cell xenograft tumors in vivo. As shown in Figure 6, the xenograft tumor volume of BALB/c-nu mice in the miR-221/222 inhibitor group was significantly decreased compared with that in the control group and the inhibitor NC group (Figure 6(A), P < 0.05). The growth curve analysis showed that the rate of xenograft tumor growth in the miR-221/222 inhibitor group was slower than that in the other two groups (Figure 6(B) and Figure 6(C)). The expression of miR-221/222 and apoptotic proteins was examined by ISH

and immunohistochemical assay, respectively, in both HSC3 and HSC6 tumor xenografts (Figure 6(D) and Figure 6(E)). The results suggested that miR-221/222 expression in the miR-221/222 inhibitor group was significantly lower than that of the control group. As for the apoptotic proteins, the expression of PUMA, Bax and Bak was increased, while Bcl-2 expression was decreased in the miR-221/222 inhibitor group.