Inhibition of p22phox Suppresses Epithelial Ovarian Cancer Cell Proliferation and Tumorigenesis.

Epithelial ovarian cancer is the leading cause of female cancer deaths, mainly attributable to the absence of specific early symptoms and effective tools for early detection. As a result, less than one-half of women with epithelial ovarian cancer survive beyond 5 years after diagnosis 1 . Epithelial ovarian cancer is a highly heterogeneous tumor and classified many histopathologic types, such as high-grade serous (HGSC), mucinous (MUC), clear cell (CCC), endometrioid (ENOC), and undifferentiated carcinomas 2 , 3 . Among those carcinoma types, HGSC has been established as of the fallopian tube origin, CCC and ENOC were endometriosis-associated ovarian cancer 4 , 5 . In general, the prognosis of epithelial ovarian cancer is significantly related to histopathologic types. Therefore, it is critical to illustrate the tumorigenesis mechanism of the occurrence and development of each type of epithelial ovarian cancer to aid in individualized treatment. Unfortunately, it remains unsettled as to the molecular mechanisms of epithelial ovarian cancer. P22 phox is a critical component of the NOX complex . Studies have demonstrated that downregulation of p22 phox results in decreased activity of several NOX enzymes 6 , 7 . As a regulatory protein that binds NOX for an effective enzymatic activity, p22 phox is associated with several diseases, such as cardiovascular disease and chronic granulomatous disease 8 , 9 . Moreover, p22 phox has been recognized as an oncogene. The study has shown elevated levels of p22 phox in primary pancreatic cancer tissues; stable knockdown of p22 phox inhibited pancreatic tumor growth 10 . In addition, downregulation of p22 phox inhibited Akt-dependent phosphorylation of tuberin and stabilized tuberin protein levels in VHL-deficient renal carcinoma cells 11 . Furthermore, p22 phox has been reported to function as a cisplatin-resistant factor that suppresses DNA adduct-induced apoptosis by blocking cisplatin uptake into the nucleus and activating the phosphatidylinositol 3-kinase (PI3K)/Akt pathway in oral squamous cell carcinoma 12 , 13 . However, whether the expression of p22 phox occurs in the different epithelial ovarian cancer types and the role of p22 phox in epithelial ovarian cancer progression has not been established. In this study, we performed a comprehensive analysis of p22 phox expression in the clinical specimens of HGSC, MUC, CCC, ENOC, and undifferentiated carcinomas by IHC. We demonstrate that inhibition of p22 phox suppressed the proliferation and tumorigenesis of A2780 and SKOV3 cells both in vitro and in vivo . Mechanism study reveals that silencing p22 phox suppressed proteasome ubiquitination and subsequent proteasome-dependent degradation of p53 in A2780 and U87 cells. Furthermore, p22 phox knockdown A2780 cells more sensitive to cisplatin or paclitaxel treatment. In summary, our findings delineated the clinical significance, biological function, and molecular mechanisms of p22 phox in epithelial ovarian cancer progression.

Human epithelial ovarian cancer samples (63), normal fallopian epithelial tissues (10), normal ovarian tissues (10), and normal endometrial epithelial tissues (10) were obtained from the Department of Pathology, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, China. This study was approved by the hospital institutional review board, and written informed consent was obtained from all patients. The study complied with the principles of the 2008 update of the Declaration of Helsinki. Human ovarian cancer cell lines (A2780, SKOV3, HO8910) and normal human ovarian epithelial immortalized cell line (IOSE386) were maintained in RPMI-1640 medium (HyClone, Logan, UT, USA), containing 10% fetal bovine serum (FBS; Gibco, USA) with 100 units/ml penicillin and 100 mg/ml streptomycin. Human OVCAR-3 cells were cultured in RPMI-1640 containing 20% fetal bovine serum with 100 units/ml penicillin and 100 mg/ml streptomycin. Human U87 glioblastoma cells were cultured in DMEM supplemented with 10% FBS, 100 units/ml penicillin, and 100 ng/ml streptomycin. Cells were incubated in an atmosphere with 5% carbon dioxide at 37 °C. Cycloheximide (CHX) was purchased from Beyotime Institute of Biotechnology, Jiangsu, China. Representative formalin-fixed paraffin-embedded tissue sections of 63 ovarian tumors in five epithelial ovarian cancer histopathologic types (HGSC, MUC, CCC, ENOC and undifferentiated carcinomas) and 30 normal tissues (ovarian, fallopian tube and endometrioid tissue) were prepared for evaluation of p22 phox protein expression by IHC, using an anti-p22 phox antibody with the following routine procedure. 5-μm sections were prepared. After dewaxing and hydration, sections were heated in citrate buffer in a microwave oven for 20 min for antigen retrieval. Anti-p22 phox rabbit polyclonal antibody (1:100, ab75941, Abcam Company, Cambridge, U.K.) was incubated with the sections overnight at 4 °C. The sections were then washed thrice in PBS for 3 min/wash and subsequently treated with a secondary antibody (ZSGB-BIO, Beijing, China) for one hour at 37 °C, followed by three washes in PBS for 3 min/wash. The p22 phox protein expression was visualized with diaminobenzidine (DAB) staining and visualized by light microscopy. A histological score (H-score) was constructed that measures the intensity and distribution of the signal using the formula: [3 (strong signal) × (percentage of cells with strong signal)] + [2 (moderate signal) × (percentage of cells with moderate signal)] + [1 (weak signal) × (percentage of cells with weak signal)]. The histological score ranges from 0 to 300. The shp22 phox plasmid was obtained from a lentivirus-based human shRNA library (GeneChem, Shanghai, China). Briefly, shp22 phox plasmid DNA and the transfection complex DNAs were transfected into human embryonic kidney 293T cells, using the trans-lentiviral packaging kit (Open Biosystems, Shanghai, China) to generate the lentivirus stock. The validated scrambled control plasmid (shSCR) was also transfected into 293T cells as vector control. Lentiviruses in the supernatant were collected and used to transduce A2780 and SKOV3 cells, followed by puromycin selection. The following antibodies were used in this study: P53 (Cell Signaling Technology, Danvers, MA, USA); GAPDH, Myc-tag, and HA-tag (Bioworld Technology, Atlanta, Georgia, USA). Western immunoblot analysis was performed with the whole-cell lysates in 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The resolved proteins were transferred into polyvinylidene fluoride (PVDF) membrane (Roche, Switzerland), followed by incubation with specific antibodies against different indicated protein factors and treated with secondary antibody (Bioworld Technology, Atlanta, Georgia, USA). Immunoreactivity was visualized by enhanced chemiluminescence (Thermo Fisher Scientific Cat. Rockford, IL, USA). Total RNA was isolated from harvested cells with the Trizol reagent (Invitrogen, CA, USA), according to the manufacturer's instructions. The mRNA levels of p22 phox and p53 were determined by reverse transcription of total RNAs using oligo(dT) primer and the PrimeScript RT Reagent Kit (Takara, Dalian, China), according to the manufacturer's instructions. The cDNAs were amplified by qRT-PCR using the SYBR Premix DimerEraser kit (Takara, Dalian, China) on the 7900HT system, according to the manufacturer's instructions. Fold changes were calculated by relative quantification (2 -ΔΔCt ). The expression of GAPDH was used as an endogenous control. The following p53 primer sequences were used for real-time PCR: p53F: 5′-GCC TGA GGT TGG CTC TGA-3′; p53R: 5′-GTG GTG AGG CTC CCC TTT-3′. The p22 phox and GAPDH primers were used as previously described 14 . A2780 cells were transfected with pCMV6-MYC-p22 phox (OriGene Technologies, Rockville, USA) and the ubiquitin recombinant vector, HA-Ub, as indicated in the Lipofectamine 2000 (Invitrogen, Carlsbad, USA) protocol, according to the manufacturer's instructions. At 48 h after transfection, the cells were treated with the proteasome inhibitor, MG-132 (10 μM) (Beyotime Institute of Biotechnology, Jiangsu, China), for 6 h. Whole-cell lysates were prepared by cell extraction and centrifugation (12,000 rpm for 20 min), and protein concentration was determined with the bicinchoninic acid (BCA) assay (Beyotime Institute of Biotechnology, Jiangsu, China). The precleared soluble supernatant was mixed with a polyclonal anti-p53 antibody and incubated overnight at 4 °C. Protein A/G beads were added to the reaction mixture. After washing with PBS, the immune complexes were resuspended in the SDS sample buffer, fractionated by SDS-PAGE, and analyzed by western blotting with a monoclonal antibody against HA. Cell viability was assessed with the cell counting kit-8 (CCK-8; Dojindo Laboratories, Kumamoto, Japan), according to the manufacturer's instructions. The cell proliferation curves were plotted with the absorbance values against different time points, using the average data from three separate experiments. EdU immunofluorescence staining was performed for fluorescence detection of replicating DNA, using the EdU kit (Ruibo Biotechnology, Guangzhou, China). Briefly, cells were grown on coverslips until they reached about 50% confluency. The EdU labeling medium was then added, and the cells were incubated for about 60 min. The cells were then fixed with 4% paraformaldehyde and stained with fresh Apollo reaction cocktail. Subsequently, cells were mounted by adding DAPI (Beyotime Institute of Biotechnology, Jiangsu, China). Images were obtained by confocal microscopy Zeiss LSM710 (ZEISS, Germany). SKOV3 and A2780 cells, which stably expressed p22 phox shRNA or shSCR, were added to growth medium with 0.6% agar and layered onto 1.2% agar beds in six-well plates. The cell culture media was exchanged every 3 days. The colonies were stained with 0.01% crystal violet and counted in 3 weeks. The results were observed under an optical microscope. The experiment was performed in triplicate. Genomic DNA was obtained from IOSE386, OVCAR3, A2780, HO8910, and SKOV3 cell lines using the Genomic DNA Extraction kit (Takara, Dalian, China) according to the manufacturer's instructions. The concentration of extracted DNA was measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific Cat. Rockford, IL, USA). The targeted fragment was amplified using the PCR-kit according to the instructions of the manufacturer (Takara, Dalian, China). Amplified samples were subjected to agarose electrophoresis with ethidium bromide as the fluorescent dye. A2780 cells were seeded at the density of 4,000 cells per well in a 96-well plate overnight. Freshly prepared cisplatin or paclitaxel (Selleck Chemicals, Houston, TX, USA) was added at varying final concentrations ranging from 0 to 80 μM or 0 to 320 nM. At 72 h after treatment, cell viability was assayed using the CCK8 kit. Flow cytometry was employed to evaluate cellular apoptosis. Cultured cells were harvested, centrifuged, and then resuspended in 100 μl solution containing 2 μl Annexin V-FITC reagent in propidium iodide solution, and then incubated for 15 min at room temperature in the dark. The samples were analyzed by flow cytometry (FACS Canto II, BD Biosciences, USA) within 1 h. Three experiments were performed in triplicate. The data were analyzed with the FlowJo software. Female BALB/c nude mice (6 weeks old) were obtained from the Shanghai Laboratory Animal Center (Chinese Academy of Sciences, Shanghai, China) and maintained in the special pathogen-free environment for one week. Animal handling and experimental procedures followed those described in the Guide for the Care and Use of Laboratory Animals and approved by the Animal Experimental Ethics Committee of Nanjing University. A2780 shp22 phox cells or A2780 shSCR cells (5 × 10 6 cells in 100 μl) were subcutaneously injected into the left flank of nude mice. Tumor sizes were measured with a vernier caliper every 2 days until the tumors were apparent grossly. The tumor volume was calculated with the following formula: volume = 0.5 × Length × Width 2 . After 25 days of implantation, mice were sacrificed and tumors were dissected, formalin-fixed, paraffin-embedded, and sectioned at 5 μm for Ki67 (ZSGB-BIO, Beijing, China) IHC staining were performed as previously described. All experiments were repeated three times and the data were averaged with GraphPad Prism 5 (La Jolla, CA, USA) as the mean ±standard deviation (SD). Statistical Analysis Significant difference of means between two groups was examined using the t-test. The difference with P < 0.05 was considered statistically significant.

As shown in Fig. 1 A-E, immunohistochemical staining showed that p22 phox expression could be detected in epithelial ovarian cancer tissues with a variety among patients. In the cohort, the average value of p22 phox expression in the HGSC is 232.3 ± 30.97, which is the highest among the five types of ovarian carcinomas (HGSC, MUC,CCC, ENOC and undifferentiated ovarian carcinoma) (Table 1 ). Mounting clinicopathologic and molecular genetic data have demonstrated various origins of ovarian carcinomas among the different histologic types. For instance, HGSC has been established as of the fallopian tube origin, but both CCC and ENOC develop from the endometrioid tissue 15 . As shown in Fig. 2 A-C, p22 phox immunostaining was intense and diffuse in normal fallopian epithelial cells, but much weaker in normal ovarian and endometrial epithelial cells. In normal ovarian, fallopian epithelial and endometrium tissues, the average value of p22 phox expression is 28.3±12.39, 262.3±21.41, 109.0±16.63, respectively (Table 1 ). We next investigated the biological consequences of p22 phox expression in four human ovarian cancer cell lines, OVCAR3, A2780, HO8910, and SKOV3, and the immortalized cell line, IOSE386. Compared with IOSE386 cells, A2780, HO8910, and SKOV3 cell lines expressed significantly higher levels of p22 phox mRNA and protein (Fig. 2 D, E) under the condition of the absence of significant differences in the DNA level among all cells tested (Figure 2 F). We then generated a stable knockdown of p22 phox in A2780 and SKOV3 cells and confirmed the inhibition efficiency by western immunoblot analysis (Fig. 3 A). Undoubtedly, shp22 phox-2 was most efficient in suppressing p22 phox expression; thus, we used this stable cell line in subsequent experiments. Likewise, the p22 phox knockdown A2780 and SKOV3 cell lines exhibited a strong inhibitory effect on cell proliferation, as indicated by the CCK-8 proliferation assay, compared to the shSCR cells (Fig. 3 B). Proliferation was also investigated with EdU immunofluorescence staining. As shown in Fig. 3 C, the p22 phox knockdown cell lines showed a significantly lower rate of proliferation than their shSCR cell line counterparts. A reduced soft agar colony formation capability was also observed in p22 phox knockdown cell lines (Fig. 3 D). Together, these results demonstrate that p22 phox knockdown arrested A2780 and SKOV3 cell growth in vitro . We further clarified the effect of p22 phox on tumor growth in vivo by subcutaneous injection of the A2780-shp22 phox or A2780-shSCR cells into the left flank of nude mice. After 25 days of implantation, we found that the tumor growth rate was significantly slower in the mice injected with A2780-shp22 phox cells than in the mice injected with A2780-shSCR cells (Fig. 4 A), as revealed by their smaller gross appearance and tumor size (Fig. 4 B). We observed that the weight of the A2780-shp22 phox cell group was significantly smaller than that of the A2780-shSCR group (Fig. 4 C). Likewise, increased ki67 immunoreactivity, which is a measure of rapid cell division, was significantly lower in the tumors from the A2780-shp22 phox group than in those from the control group (Fig. 4 D). The results suggest that the inhibition of p22 phox indeed suppressed A2780 tumor growth in vivo . It is known that the wild type p53 exists in the epithelial ovarian cancer subtypes of ENOC and CCC 16 . Immunohistochemistry analysis showed a wild type p53 pattern of expression for p53 in our ENOC and CCC tissues ( Figure S1 ). The A2780 cell line, which is wild type p53, demonstrated a very flat gene copy-number profile with mutations frequently found in ENOC and CCC, such as ARID1A, BRAF, PIK3CA, and PTEN, as reported previously 17 . Thus, we investigated p53 expression levels in A2780-shp22 phox and A2780-shSCR cells. Interestingly, compared to the A2780-shSCR cell, the A2780-shp22 phox cell line showed significant upregulation of p53 protein levels, whereas no significant differences in p53 mRNA levels (Fig. 5 A, B). Moreover, with exogenously overexpressed p22 phox , the p53 protein levels were significantly downregulated compared with the control counterparts (Fig. 5 C). Nonetheless, p53 mRNA levels did not change (Fig. 5 D). Similarly, in the U87 (wild type p53) cell line with overexpressed p22 phox , the protein amount, not mRNA levels of p53, was decreased when compared with its control cell line (Fig. 5 C, D). Therefore, with those results, we hypothesized that p53 is modulated by p22 phox through post-transcriptional or post-translational modifications. To further investigate the findings that p22 phox affected p53 stability, we treated A2780-shp22 phox and A2780-shSCR cell lines with cycloheximide (CHX) and examined p53 expression levels. The results demonstrated that p53 stability was enhanced in A2780-shp22 phox cells, as compared to A2780-shSCR cells (Fig. 5 E). Moreover, overexpression of p22 phox increased the amounts of p53 in the presence of the proteasome inhibitor MG132, suggesting the involvement of the ubiquitin chains (Fig. 5 F). Accordingly, we verified the existence of p22 phox -linked p53 ubiquitination in the A2780 cell line by transfecting this cell line with HA-Ub in the absence or presence of p22 phox . The cell lysates were immunoprecipitated with anti-p53 antibodies, followed by immunoblotting (IB) with an anti-HA antibody. The result demonstrated that overexpression of p22 phox markedly promoted the ubiquitination of endogenous p53 (Fig. 5 G). Together, these results indicate that p22 phox either ubiquitinates p53 or promotes p53 ubiquitination in A2780 cells. To explore the potential effects of p22 phox on chemosensitivity to cisplatin or paclitaxel, we treated A2780-shp22 phox and A2780-shSCR cells with different concentrations of cisplatin or paclitaxel. As shown in Fig. 6 A and 6 B, A2780-shp22 phox cells significantly decreased chemosensitivity to cisplatin or paclitaxel in vitro ; the cell viability was also considerably suppressed, as concentrations of cisplatin (range = 0-80 μΜ), or paclitaxel (range = 0-320 nΜ) were increased, compared to A2780-shSCR cells. This enhanced chemosensitivity to cisplatin or paclitaxel treatment was subsequently investigated by flow cytometry to assess the rate of cellular apoptosis. We found that the combination of p22 phox knockdown and cisplatin or paclitaxel treatment significantly increased cell apoptosis rates in the A2780 cell line (Fig. 6 C). These results suggest that the downregulation of p22 phox expression rendered the ovarian A2780 cells more sensitive to cisplatin or paclitaxel treatment.

In the present study, immunohistochemical staining showed that p22 phox expression could be detected in HGSC, MUC, CCC and ENOC tissues with a variety among patients. The p22 phox expression value in the HGSC is the highest among the five types of ovarian carcinomas. This finding is consistent with that reported previously, high grade serous carcinoma accounts for majority of the cases and most of the lethality 15 . Epithelial ovarian cancer is hypoxic by nature and could grow under hypoxic conditions, in which several cellular signaling pathways are known to be activated to enhance tumor cell proliferation and metastatic capacity 18 , 19 . Until recently, it was reported that p22 phox is ubiquitinylated under normoxic conditions and that this modification is decreased under hypoxia, indicating that p22 phox is stabilized and accumulates under hypoxia 20 . It is, thus, plausible that p22 phox overexpression could result from the hypoxic microenvironment in ovarian tumors. Further experiments are needed to prove this relevant speculation. Mounting clinicopathologic and molecular genetic data have demonstrated HGSC has been established as of the fallopian tube origin, but both CCC and ENOC develop from the endometrioid tissue 15 . The H-score of p22 phox in normal fallopian and HGSC tissues was 262.3±21.41, 232.3±30.97, respectively. The high expression of p22 phox in the HGSC and fallopian epithelial indirectly suggests that the lesion originates from the tubal epithelium. In normal endometrium tissue, CCC and ENOC, the average value of p22 phox expression was 109.0±16.63, 207.3±29.75 and 180.7±61.43, respectively. Higher expression of p22 phox in CCC and ENOC tissues than in the normal endometrial epithelium suggests that p22 phox might promote the occurrence and development of those two ovarian carcinomas. Additionally, our results confirmed that p22 phox knockdown decreased A2780 and SKOV3 cell proliferation and soft agar colony formation capacity. Furthermore, tumor growths in mouse xenograft models were inhibited by p22 phox knockdown, further confirming that p22 phox acts as oncogenes in epithelial ovarian cancer cells. P22 phox is a critical component of the NOX complex, which is involved in the regulation of endogenous ROS level 6 , 7 . P53 protein is the core of a network of pathways in which ROS play critical roles, and is an essential transcriptional factor regulating cellular homeostasis 21 , 22 . It is well known that the CCC and ENOC possess the wild type p53, we found that inhibition p22 phox appeared to upregulate p53 protein levels, while no significant reduction of p53 mRNA levels in A2780 cells. These results motivated us to hypothesize that p22 phox was involved in p53 protein polyubiquitination and proteasome-dependent degradation. Further, we verified the hypothesis by performing in vitro ubiquitination studies in A2780 and U87 cells. MDM2, a ubiquitin ligase for p53, plays a central role in the regulation of the stability of p53 23 . Moreover, Pirh2 and LIF could also play roles in regulating this process 24 , 25 . Additionally, the deubiquitinating enzyme Otubain and the ubiquitin specific protease-7 (USP7) can remove the ubiquitin from P53 26 , 27 . Furthermore, p53 is involved in apoptosis, cell cycle regulation, and DNA replication and repair, and regulates the expression of many genes, including PUMA, NOXA, FAS, and p21 genes. The exact pathways of p53 involving p22 phox in A2780 and U87 cells require further verification to gain more insight into the mechanism. Cis-diamminedichloroplatinum (II) (cisplatin II) and paclitaxel are widely used for treating various solid tumors, including cervical, ovarian, and breast cancers. It is inevitable to develop chemotherapeutic drug resistance after a prolonged period of usage. To date, cisplatin- or paclitaxel-based chemotherapy, combined with other chemotherapeutic agents, remains the first-line treatment for ovarian cancers 28 . Recent reports have shown that p22 phox is a cisplatin-resistant gene that suppresses DNA adduct-induced apoptosis by blocking cisplatin uptake into the nucleus and activating the PI3K/Akt pathway 13 . Our study found that interfering with p22 phox function could enhance the drug sensitivity of ovarian cancer cells to cisplatin or paclitaxel and thus provides a new concept for the treatment of ovarian cancer. Herein, we provide the evidence that p22 phox expression could be detected in epithelial ovarian cancer tissues. Inhibition of p22 phox showed the profound effects in retarding ovarian cancer cell proliferation and tumorigenesis in vitro and in vivo , complementing its biological function via influencing the proteasome ubiquitination and subsequent proteasome-dependent degradation of p53. Moreover, downregulation of excessive expression of p22 phox significantly increased the chemosensitivity of A2780 cells to cisplatin or paclitaxel treatment. Overall, our findings define a useful treatment target for combating epithelial ovarian cancer growth.

Supplementary figure S1. Click here for additional data file.