SKOV-3, a human ovarian cancer cell line was obtained from ATCC. The SKOV3-DDP cell line, which was resistant to cisplatin was obtained from The Hospital Central Laboratory of Qingdao University (Qingdao, China). OV4 cells (meaning the OVCAR-4 cell line in the present study) were obtained from the Type Culture Collection of the Chinese Academy of Sciences. SKOV-3 and SKOV3-DDP cells were cultured in DMEM (Gibco; Thermo Fisher Scientific Inc.) containing 10% (v/v) FBS (Gibco; Thermo Fisher Scientific Inc.) at 37°C with 5% CO₂, whereas OV4 cells were cultured in RIPM 1640 (Gibco; Thermo Fisher Scientific Inc.). 100 IU/ml penicillin and 10 mg/ml streptomycin were added for all sterile cell culturing.

SKOV-3 and OV4 cells (4×10⁵/well) were plated into 6-well plates and incubated for 48 h in DMEM or RIPM 1640 containing 10% (v/v) FBS at 37°C with 5% CO₂. The cell layer was scratched with a pipette tip to create a fresh wound in the middle of the wells. Subsequently, FBS-free medium was used for the wound healing assay. Different concentrations (0, 0.3, 1 and 3 µM) of dacomitinib (cat. no. C9154; Selleck Chemicals) were added into respective wells. After culturing for 0, 24 and 48 h at 37°C with 5% CO₂, photos of each well were taken to determine cell migration.

SKOV-3 and SKOV3-DDP cells (5×10⁴/well) were plated in 96-well plates with 100 µl RPMI 1640. Cells were first treated with 1 µM dacomitinib for 24 h, then 0, 2.5, 5, 10, 20, 40, and 80 µM of cisplatin (DDP) (cat. no. 15663-27-1; Selleck Chemicals) was added into the wells and cultured at 37°C with 5% CO₂. After 48 h, 20 µl MTT reagent (5 mg/ml; pH=7.4) was added to the cell culture and left for 4 h, following which 150 µl DMSO was added for 10 min avoiding light. Optical density (OD) value of each well was determined at 490 nm using an universal microplate spectrophotometer.

SKOV3-DDP (1×10⁶) cells treated with or without dacomitinib and cisplatin were cultured for 24 h at 37°C with 5% CO₂. After washing 3 times with PBS, cells were stained with a kit containing Annexin V-PE and 7AAD (cat. no. SY0479; Beijing Biolab Technology Co., Ltd.). Subsequently, cells were resuspended with 0.2 ml PBS and analyzed using the fluorescence-activated cell sorting technique using a FACSCantoII (BD Biosciences) and Flowjo v.10 (Becton, Dickinson & Company). Both early and late stage apoptosis were assessed.

SKOV-3 and SKOV3-DDP cells treated with dacomitinib or cisplatin were lysed on ice for 30 min using mammalian cell lysis RIPA buffer (cat. no. P0013C; Beyotime Insitute of Biotechnology) containing protease and phosphatase inhibitor cocktails (1:1,000). Samples were then centrifuged at 13,000 × g for 20 min at 4°C. Protein concentration was measured by BCA kit (cat. no. P0012S; Beyotime Institute of Biotechnology). Supernatants containing 20 µg proteins/lane were mixed with loading buffer for 10 min at 100°C. Cell lysate proteins were then separated using 12% SDS-PAGE gels and were transferred to PVDF membranes. Clipped PVDFs were blocked with 5% non-fat skimmed milk (dissolved in 1X Tris-buffered saline with 0.1% Tween-20) for 1 h at room temperature. PVDF membranes were then incubated with monoclonal primary antibodies against CDH1 (cat. no. Ab76055; Abcam), SLUG (snail family transcriptional repressor 2) (cat. no. Ab180714; Abcam), EGFR (cat. no. Ab52894; Abcam), P-EGFR (cat. no. Ab40815; Abcam), P-GP (cat. no. Ab103477; Abcam), and the internal reference β-actin (cat. no. 9710; Origene Technologies Inc.) overnight at 4°C. After washing 3 times with TBST (TBS+0.05%Tween 20), PVDF membranes were incubated with an appropriate horse radish peroxidase (HRP)-conjugated secondary antibody (cat. no. 15140-122; Gibco; Thermo Fisher Scientific Inc.), or Goat Anti-Mouse IgG H&L (HRP) (cat. no. Ab6789; Abcam), or Goat Anti-Rabbit IgG H&L (HRP) (cat. no. Ab6721; Abcam) respectively. DAB kit (cat. no. ZLT-9031; Origene Technologies Inc.) was used to determine protein levels. The dilutions of all the primary antibodies (1:1,000) and secondary antibodies (1:5,000) were used in the present study. Image J software (National Institutes of Health) was used to measure the densitometry of the grey band. CDH1 and SLUG markers were used for EMT related protein (21,22). EGFR and P-GP markers were used for drug resistant protein (23,24).

GSE31432 (Illumina Human HT-12 v.3.0 expression bead chip) data (25) were retrieved from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo). The dataset comprised 4 types of human OC samples: negative group cells treated with trastuzumab, or pertuzumab, or both. These inhibitors reduce HER2 expression, hence have similar activity on OC cells as dacomitinib treatment. DEGs in the negative group and the drug-treatment group (trastuzumab and pertuzumab) were first determined (unpaired Student's t-test was used and genes with an adjusted P-value <0.0004 were defined as DEGs). Functional proteins were predicted using the String webserver (https://string-db.org/) with Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genomics and Genes (KEGG) pathway analysis.

Three or more independent experiments were performed and data were reported as means ± SD. Statistical analysis were performed using SPSS Statistical Package v.17 (SPSS, Inc.) or GraphPad5 software (GraphPad Software Inc.). Unpaired Student's t-test and one-way ANOVA followed by the post hoc test were performed to compare means of 2 or more groups, respectively. P<0.05 was considered to indicate a statistically significant difference.

Wound healing assay was performed to determine effect of dacomitinib treatment (0, 0.3, 1, 3 µM) on migration ability of SKOV3 and OV4 cells. The percentage of wound healing of OV4 cells treated with 3 µM dacomitinib was 29.30% after 48 h compared with the control group which had a percentage of wound healing of 90.33% (Fig. 1A). The percentage of wound healing of SKOV3 cells treated with 3 µM dacomitinib was 43.67% after 48 h compared with the control group which had a percentage of wound healing of 95.33% (Fig. 1B). The aforementioned findings indicate that dacomitinib inhibits migration ability of human OC cells. High dosages of dacomitinib demonstrated higher inhibition of migration ability of OC when compared with control group (without dacomitinib) (Fig. 1A and B).

EMT related proteins, CDH1 and SLUG, were chosen to assess migration and invasive ability of human ovarian cancer cell following dacomitinib treatment. Western blotting data demonstrated increased expression of CDH1 and decreased expression of SLUG in SKOV3 and OV4 cells following treatment with high dacomitinib doses (3 µM) and vice versa. (Fig. 1C and D). The aforementioned results indicated that dacomitinib inhibited migration and invasive ability of OC cells in a dose-dependent manner.

To explore the synergistic effects of dacomitinib with other drugs, cisplatin was chosen and tested on SKOV3 and SKOV3-DDP cells. Prior to the assay, cisplatin resistant cells were developed and treated with dacomitinib. Cells were treated with cisplatin (0, 2.5, 5, 10, 20, 40 and 80 µM) for 24 h. Cells demonstrated decreased viability following treatment with a higher dosage of cisplatin (80 µM) (Fig. 2A). Cisplatin had an IC₅₀ of 12.27 µM against SKOV3 cells, and IC₅₀ of 64.34 µM against SKOV3-DDP cells (drug resistance index=5.24) (data not shown).

SKOV3-DDP cells were then treated with or without dacomitinib (1 µM) plus 0, 2.5, 5, 10, 20, 40, 80 µM cisplatin for 24 h. Cells demonstrated decreased viability following treatment with a higher dosage of cisplatin (80 µM) (Fig. 2B). An IC₅₀ of 11.30 was observed against SKOV3-DDP cells (a 5.69 times decrease when compared with IC₅₀ of 64.34 µM against SKOV3-DDP cells) (data not shown). These findings implied that dacomitinib improved the antiproliferative effect of cisplatin-resistance in SKOV3-DDP cells (Fig. 2B). Apoptosis of SKOV3-DDP cells treated with dacomitinib, cisplatin, or both was tested using FACS based on Annexin V and 7AAD staining. Apoptotic assays demonstrated a higher apoptosis rate of cisplatin group (19.3%), dacomitinib group (23.4%), and dacomitinib plus cisplatin group (39.2%) compared with the apoptosis rate of the control group (2.3%). The dacomitinib plus cisplatin group demonstrated a higher percentage of apoptosis compared with the cisplatin group (DDP) (Fig. 2C and D). The aforementioned finding demonstrated that dacomitinib promotes apoptosis of cisplatin-resistant OC cells.

P-EGFR and P-GP protein levels are associated with EGFR signaling (26,27). Protein levels of p-EGFR and P-GP in SKOV3-DDP cells treated with dacomitinib and cisplatin were determined (Fig. 3A and B). Expression levels of P-EGFR and P-GP in cisplatin, dacomitinib, and dacomitinib plus cisplatin groups were significantly lower compared with that in the control group (Fig. 3A). Western blotting analysis demonstrated higher expression levels of p-EGFR and P-GP in SKOV3-DDP cells (resistant cells) compared with the levels in SKOV3 cells (Fig. 3B). These findings collectively implied that dacomitinib, an EGFR inhibitor, regulates p-EGFR and P-GP directly or indirectly to modulate growth of cisplatin-resistant OC cells.

To further explore the effect of EGFR inhibitors in regulation of signaling pathways in human ovarian cancer cells, SKOV3 samples treated with EGFR inhibitors, such as trastuzumab and pertuzumab were retrieved from the GEO database (GSE31432). Dacomitinib, trastuzumab, and pertuzumab are EGFR and HER2 inhibitors (28–30). EGFR inhibitor-treated samples were used to determine potential genes regulated by dacomitinib. A total of 6 samples without any treatment were chosen for group A, and 5 samples treated with EGFR inhibitor trastuzumab plus pertuzumab were chosen for group B. GEO analysis demonstrated that these samples had good quality of total gene expression (data not shown). The top 100 genes were identified based on P-value (Table SI).

GO enrichment and KEGG pathway analysis were performed on these candidate genes using the String webserver. A total of 66 proteins demonstrated highly related interaction (PPI enrichment P-value=4.77×10⁻⁷) from the top 100 genes (Fig. 4A). Biological process (GO) category demonstrated that only 15 proteins were enriched in regulation of cell adhesion, whereas 8 proteins were enriched in negative regulation of cell adhesion (Fig. 4B). Cellular component (GO) demonstrated extracellular region (49 proteins), extracellular space (18 proteins), and secretory vesicle/granule (29 proteins) were highly enriched (Fig. 4B). KEGG pathway analysis showed 4 proteins enriched in glutathione metabolism signaling (Fig. 4B).

There were several key nodes, such as CDH1 (E-cadherin) and matrix metalloproteinase (MMP7) (Fig. 4A). Expression of CDH1 and MMP7 was detected in all samples in the GSE31432 dataset. Following EGFR inhibition, samples of human OC demonstrated increased expression level of E-cadherin and decreased expression level of MMP7 (Fig. 4C). These findings were consistent with EGFR inhibition by dacomitinib through reducing migration and invasive ability of OC by modulation of CDH1.