This article is mentioned in:

Introduction

Ovarian cancer continues to be the most frequent cause of cancer mortality among females in Western Europe and the United States, with the highest mortality rate of all gynecological malignancies (1). The most common histological subtype of ovarian cancer is epithelioid cancer (serous, endometrioid, mucinous and clear cell), accounting for 90% of ovarian malignancies (2,3). Although >70% of patients have increased 5-year survival rates subsequent to surgery followed by chemotherapy plus second-line therapies, a number of tumor types fail to respond to chemotherapy. This is due to the fact that in consecutive chemotherapy, these tumors appear to become less sensitive or resistant to chemotherapeutic drugs. The intolerable side-effects of the systemic treatment with chemotherapy means that the development of novel and effective therapeutic modalities and the identification of new rational targets for these novel therapies are required (4).

Organisms living under aerobic conditions are exposed to reactive oxygen species (ROS), including superoxide anions (O2−), hydrogen peroxide (H2O2) and nitric oxide (NO), which are generated by the redox metabolism mainly in the mitochondria (5). A number of studies have demonstrated that ROS in small amounts participate in numerous physiological processes, including cell proliferation, differentiation and apoptosis and the modulation of transcription factors and signal transduction pathways (6,7). Cancer cells have been demonstrated to possess higher levels of intracellular ROS due to increased cellular respiration and are believed to have the potential to oxidize macromolecules, induce DNA mutation, impair protein function and peroxidize lipids, thus leading to the development of tumors (8,9). However, to facilitate the development of tumors, numerous antioxidative systems have been developed to maintain an appropriate level of ROS, which include antioxidant enzymes such as glutathione peroxidases (GPXs), superoxide dismutases (SODs), catalases and the recently identified rapidly growing family of peroxiredoxins (Prxs) (10,11).

Prxs have been shown to be involved in diverse cellular roles, including the control of cell proliferation, differentiation and apoptosis, the protection of oxidant-sensitive proteins, the regulation of cellular H2O2 and redox regulation (12,13). Recent studies have suggested that Prxs are involved in the development, progression and drug resistance of cancers (14,15). As a member of the Prx family, PRDX3 is overexpressed in a number of diverse tumors, including hepatocellular carcinoma and lung and prostate cancer (16–18). To the best of our knowledge, there has only been one study describing the potential role of PRDX3 in ovarian cancer, in which PRDX3 was determined to be responsible for resistance to platinum-based chemotherapies (19). As an anti-apoptotic protein for tumor cell proliferation and survival, therapeutic strategies targeting PRDX3 may therefore be effective broad-spectrum anticancer agents.

In the present study, the focus was on the role and mechanism behind the downregulation of PRDX3 in cisplatin-induced ovarian cancer cell apoptosis. For the first time we reported that the overexpression of PRDX3 was observed in ovarian cancer and that the PRDX3 knockdown resulted in accelerated cisplatin-induced cell apoptosis through suppression of the NF-κB signaling pathway. This may provide a new insight into anticancer therapy for ovarian cancer.

Materials and methods

Tissue samples

In total, 104 paraffin-embedded ovarian cancer tissues and paired non-cancerous tissues were surgically obtained from the Maternal and Child Health Hospital of Hubei Province, Wuhan, Hubei, China, between 2005 and 2011. All the procedures were approved by the research ethics committee of the Maternal and Child Health Hospital of Hubei Province. All the patients agreed to the procedure and signed consent forms. None of the patients had received chemotherapy or radiation therapy prior to the surgery. The staging and grading were performed by two experienced gynecological pathologists according to the criteria of the International Federation of Gynaecologists and Obstetricians (FIGO) and the World Health Organization (WHO). The median patient age was 56 years (range, 33–72 years). Detailed patient characteristics are presented in Table I.

Table I Associations between the expression of PRDX3 and the clinicopathological parameters in ovarian cancer.

Table I

Associations between the expression of PRDX3 and the clinicopathological parameters in ovarian cancer.

		PRDX3 expression
Parameters	Cases (n)	−	+	++	+++	P-value
Age at first diagnosis
<56 years	45	2	5	22	16	0.691
≥56 years	59	6	8	27	18
Serous
I/II	20	3	4	9	4	0.045a
III/IV	41	1	2	22	16
Well-differentiated	21	2	2	15	2	0.028b
Moderately-differentiated	23	2	3	11	7
Poorly-differentiated	17	0	1	5	11
Mucinous
I/II	8	1	2	3	2	0.173a
III/IV	11	1	1	9	0
Well-differentiated	9	2	2	3	2	0.251b
Moderately-differentiated	6	0	1	5	0
Poorly-differentiated	4	0	0	4	0
Endometrioid
I/II	4	1	2	1	0	0.032a
III/IV	7	0	0	1	6
Well-differentiated	4	1	2	0	1	0.278b
Moderately-differentiated	4	0	0	1	3
Poorly-differentiated	3	0	0	1	2
Clear cell
I/II	4	1	1	2	0	0.114a
III/IV	9	0	1	2	6
Well-differentiated	5	0	1	2	2	0.71b
Moderately-differentiated	3	0	1	1	1
Poorly-differentiated	5	1	0	1	3
Lymph node metastasis
Negative	82	6	8	39	29	0.352
Positive	22	2	5	10	5

a Statistical difference between FIGO stages I/II and III/IV;

b statistical difference among the well-, moderately- and poorly-differentiated cancer types.

{ label (or @symbol) needed for fn[@id='tfn3-mmr-07-05-1688'] } −, negative; +, weakly positive; ++, moderately positive; +++, strongly positive; FIGO, International Federation of Gynaecologists and Obstetricians.

Immunohistochemistry and staining score evaluation

The paraffin slices were successively incubated for 5 min in a series of xylene and ethanol baths of decreasing concentration. The antigen was retrieved by boiling the slices in a 0.1 M Na citrate buffer (pH 6.0). The endogenous peroxidase was inhibited by 3% H2O2. The monoclonal mouse anti-PRDX3 antibodies (Abcam, Cambridge, UK), diluted 100-fold in 1% BSA, were applied and incubated overnight at 4°C. The incubation with the horseradish peroxidase (HRP)-conjugated goat anti-mouse antibodies at 37°C for 30 min was followed by the reaction with the 3,3′-diaminobenzidine substrate solution and counterstaining with Mayer’s hematoxylin. All the steps were performed in a moist chamber.

The expression of PRDX3 was scored based on the intensity of the staining and the percentage of positively stained cells. Staining intensity was scored as follows: lack of staining, 0, weak staining 1, moderate staining, 2 and strong staining, 3. The percentage of the positively stained cells was scored as 0 if no staining was observed or if it was present in <5% of the cells; 1 if positive staining was present in 5–25% of the cells; 2 if positive staining was present in 25–50% of the cells; 3 if positive staining was present in 50–75% of the cells; and 4 if positive staining was present in in >75% of cells. The score for each section was measured as the staining intensity × the percentage of the positively stained cells. The result was defined as negative (−, score of 0), weakly positive (+, score of 1–3), moderately positive (++, score of 4–7) or strongly positive (+++, score of 8–12). The slides were examined separately by two independent pathologists who had no prior knowledge of each patient’s clinical information. Any discrepancies between the two evaluators were resolved by a re-evaluation and careful discussion until agreement was reached.

RNA interference experiments

First, the expression of PRDX3 was examined in the HO-8910, SKOV3, OVCAR-3 ovarian cancer cell lines (ATCC, Manassas, VA, USA). To knock down the PRDX3 expression, a pGCsi-U6/Neo/GFP vector was used that encoded a small hairpin RNA directed against the target gene in the SKOV3 ovarian cancer cells. The target sequences for PRDX3 were 5′-ACCTTCTGAAAGTACTCTT-3′ [small interfering (si)RNA-1], 5′-CTTTAGACGAATCAATTCA-3′ (siRNA-2) and 5′-CGTACGACCCACTGTCGTC-3′ (siRNA-3). As a negative control, the shRNA vector without hairpin oligonucleotides (siRNA-mock) was used. For the transfection, the ovarian cancer cells were seeded to achieve 70–80% confluence and then transfected in serum-free medium for 6 h with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) and siRNA-PRDX3 or the siRNA-mock. Subsequent to 48 h, the cells were harvested and a limited dilution was performed in 96-well plates for the generation of individual cell clones. Three weeks later, the levels of PRDX3 expression in the cell clones that had been transfected with siRNA-PRDX3 or the siRNA-mock, were characterized by a western blot analysis for the PRDX3 protein.

Western blot analysis

The cells were lysed in a lysis buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 50 mM dithiothreitol and 0.01% bromophenol blue). The cell extract protein amounts were quantified using the Bicinchoninic Acid (BCA) Protein Assay kit (Beyotime, Shanghai, China). Equivalent amounts of protein (50 μg) were separated using 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore Corporation, Billerica, MA, USA). The blots were probed with specific antibodies for PRDX3 (1:1000, ab16751; Abcam), p-NF-κB (1:1000, sc135769; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), Bcl-2 (1:500, ab692, Abcam), Bcl-XL (1:1000, sc8392; Santa Cruz Biotechnology Inc.), Bax (1:1000, ab5714; Abcam), caspase-3 (1:3000, ab32351, Abcam), caspase-9 (1:1000, sc81650; Santa Cruz Biotechnology Inc.) and β-actin (1:1000, ab3280; Abcam) followed by a secondary detection step with goat anti-mouse IgG (Abcam). The immunoreactive proteins were revealed by an Enhanced Chemiluminescence (ECL) kit (Pierce, Rockford, IL, USA).

Cell proliferation assay

A cell proliferation assay was performed using the Cell-Counting kit (CCK)-8 (Dojindo, Kumamoto, Japan) to analyze the proliferation potential of the parental cells, empty vector and siRNA-PRDX3-transfected cells treated with/without cisplatin. The cells were harvested and plated in 96-well plates at 1×103 cells/well and maintained at 37°C in a humidified incubator. At the indicated time-points, 10 μl of the CCK-8 solution were added and incubated for 1.5 h and the number of viable cells in each well was calculated by measuring the absorbance at a wavelength of 450 nm.

Colony formation assay

The colony formation assay was performed to measure the cell growth according to the protocol described previously (20), with certain modifications. Identical numbers of parental cells, empty vector and siRNA-PRDX3-transfected cells treated with/without cisplatin were seeded in 6-well tissue-culture plates to form colonies. Subsequent to incubation for 10 days, the number of colonies (≥20 cells) within a field was counted under a light microscope at a magnification of ×200. For each test, a total of five fields were selected at random and the numbers were averaged. For the calculation of the colony forming ability, the following formula was used: colony forming efficiency = number of colonies/number of inoculated cells × 100.

Tumorigenicity assays in athymic mice

Male, 6–8-week old, BALB/c nude (nu/nu) mice were purchased from Slac Laboratory Animal Co., Ltd. (Shanghai, China) and used for the in vivo studies. The mice were randomly divided into six groups (n=8) and injected with the following: i) 5×106 SKOV3 cells in 100 μl PBS; ii) 5×106 SKOV3/Si-Mock cells in 100 μl PBS; iii) 5×106 SKOV3/Si-1 cells in 100 μl PBS; iv) 5×106 SKOV3 cells plus 0.1 mg cisplatin in 100 μl PBS; v) 5×106 SKOV3/Si-Mock cells plus 0.1 mg cisplatin in 100 μl PBS; and vi) 5×106 SKOV3/Si-1 cells plus 0.1 mg cisplatin in 100 μl PBS (5 mg/kg body weight). The 5×106 cells were subcutaneously injected into the left back flank of these animals. Cisplatin was administered by the intraperitoneal route on day 6 subsequent to the implantation of the tumor cells and again when the tumor diameter reached ~5 mm (every week, 4 times). The tumor variables were measured every three days by an electronic caliper and the tumor volume was calculated using the formula: tumor volume (mm3) = 0.52 × length (mm) × width2 (mm2). The measurements began in the first week when the tumor mass was well-established. A third operator evaluated the morphometric variables in a coded and blinded maner. The tissue samples were harvested for terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) analysis.

TUNEL assay

TUNEL analysis was performed with an in situ Cell Death Detection kit (Roche, Basel, Switzerland). The cell apoptosis index was quantified by determining the percentage of the positively stained cells for all of the nuclei in 20 randomly chosen fields at a magnification of ×200. Slides of the apoptosis studies were quantified in a blind manner by two independent reviewers at two differing times.

Statistical analysis

Data are presented as the mean ± standard deviation. Statistical analyses were performed by SPSS 13.0 software. Differences among the groups were determined by an ANOVA analysis and the comparison between two groups was analyzed by the Student’s t-test. Nominal variables were compared using a cross-tabulation test. P<0.05 was considered to indicate a statistically significant difference.

Results

Association between PRDX3 expression and the clinicopathological characteristics of ovarian cancer

To investigate the role of PRDX3 in the progression of ovarian cancer, immunohistochemistry was used to assess the PRDX3 expression levels in the paraffin-embedded tissues. The association between PRDX3 expression and the clinicopathological characteristics of the ovarian cancer tissues is shown in Table I. PRDX3 was heavily stained in the cytoplasm and nuclei of the cancer tissues (12.5% +, 47.1% ++ and 32.7% +++), but little staining was observed in the adjacent non-cancerous tissues (35.5% − and 64.5% +). The difference between the cancerous and non-cancerous tissues was significant (P<0.05). The patients with serous ovarian cancer accounted for 58.7% of the total cases. Serous ovarian cancer mostly presented at FIGO stage III or IV (FIGO I + II, 32.8% and FIGO III + IV, 67.2%) and a significant difference was observed between stages I + II and III + IV (P<0.05) in the various types of ovarian cancer. With regard to tumor differentiation, significant differences were identified in the PRDX3 expression among the well-differentiated, moderately-differentiated and poorly-differentiated serous ovarian cancer samples (P<0.05). With respect to the mucinous, endometrioid and clear cell types, there were significant differences between stages I + II and III + IV in the endometrioid ovarian cancer samples, but not in the mucinous or clear cell types. With regard to the tumor differentiation of the mucinous, endometrioid and clear cell carcinomas, the differences in the PRDX3 expression among the well-, moderately- and poorly-differentiated carcinomas were not significant. Only slight differences in the PRDX3 expression were observed with respect to the histological subtype, age at first diagnosis and lymph node metastasis (P>0.05).

Inhibition of PRDX3 expression in the ovarian cancer cells

To select the optimal cells to be transfected, the expression of PRDX3 was detected in three cell lines (HO-8910, SKOV3 and OVCAR-3), as shown in Fig. 1A. The expression level of PRDX3 in the HO-8910, SKOV3 and OVCAR-3 cell lines showed no significant difference, with the lowest expression observed in the HO-8910 cells and the highest in the SKOV3 cells. To investigate the biological significance of PRDX3 on the ovarian cancer cells, three siRNA expression vectors (Si-1, Si-2 and Si-3) specific to the PRDX3 transcripts were constructed and transfected into the ovarian cancer SKOV3 cells that endogenously expressed a high level of PRDX3 (Fig. 1A). A knockdown effect was observed by western blot analysis and SKOV3/Si-1 was observed to be the most effective when compared with the SKOV3/Si-Mock (Fig. 1B). The successful establishment of the PRDX3 gene silent ovarian cancer cell clone provided a useful tool for the investigation into the function of PRDX3 in the growth of the ovarian cancer cells.

Figure 1 Expression of PRDX3 analyzed by western blot assay. (A) Expression of PRDX3 in three ovarian cancer cell lines (HO-8910, SKOV3 and OVCAR-3) was detected. β-actin served as an internal control. Lanes 1, HO-8910 cells; 2, SKOV3 cells; 3, OVCAR-3 cells. (B) Western blot analysis of PRDX3 in SKOV3 cells transfected with control (Si-Mock) or small interfering RNA (Si-1, Si-2 and Si-3). β-actin served as an internal control. Lanes 1, SKOV3 cells; 2, SKOV3/Si-Mock; 3, SKOV3/Si-1; 4, SKOV3/Si-2; 5, SKOV3/Si-3.

Effect of PRDX3-siRNA on the growth of the ovarian cancer cells

To examine the effect of PRDX3 on the growth of the ovarian cancer cells, the dynamics of SKOV3, SKOV3/Si-Mock and SKOV3/Si-1 cell growth were determined by a cell proliferation assay. Following a 7-day period, the growth of the SKOV3/Si-1 cells was much slower compared with the SKOV3 or SKOV3/Si-Mock groups (Fig. 2A). The other pattern of inhibition for the PRDX3 expression in the ovarian cancer cells was achieved in the colony formation assay. Consequently, the average colony number of the SKOV3/Si-1 cells was decreased compared with the SKOV3 or SKOV3/Si-Mock groups. This difference was significant (P<0.05; Fig. 2B). To further explore the reason behind the decrease in cell viability, the tumorigenicity of the SKOV3/Si-1 cells was examined, as shown in Fig. 2C. The tumor size in the nude mice injected with the SKOV3/Si-1 cells was significantly smaller than that in the mice injected with the SKOV3/Si-Mock cells (P<0.05). This was associated with the enhanced induction of apoptosis, as shown by Fig. 4A.

Figure 2 Effect of PRDX3-small interfering (si)RNA on the growth of the SKOV3 ovarian cancer cells. (A) The growth of the SKOV3, SKOV3/Si-Mock and SKOV3/Si-1 cells was determined by CCK-8 assay. (B) The growth of the SKOV3, SKOV3/Si-Mock and SKOV3/Si-1 cells was examined by a colony formation assay. * vs. the SKOV3/Si-Mock or SKOV3 group, P<0.05. (C) Graphical representation of the tumor size over time in mice injected with SKOV3, SKOV3/Si-Mock and SKOV3/Si-1 cells. Data are expressed as the mean ± SEM of three independent experiments. CCK-8, Cell-Counting kit; si, small interfering.

Figure 4 PRDX3-siRNA enhances cisplatin-induced ovarian cancer cell apoptosis via the NF-κB signaling pathway. (A) Apoptotic index of the various ovarian cancer tissues. In the tumorigenicity assay, tumor tissue samples were collected for the TUNEL analysis. The apoptotic index of the groups was counted and calculated as a ratio of the apoptotic cell number to the total cell number in each field. * vs. SKOV3/Si-Mock or SKOV3 groups, P<0.05 and # vs. No cisplatin-treated group, P<0.05. (B) Western blot analysis of the expression of apoptosis-related proteins Bcl-2, Bcl-XL, NF-κB p65, Bax, Caspase-3 and Caspase-9 in the SKOV3/Si-Mock or SKOV3/Si-1 cells with/without treatment of cisplatin. Lanes 1, SKOV3/Si-Mock; 2, SKOV3/Si-Mock+Cisplatin; 3, SKOV3/Si-1; 4, SKOV3/Si-1+Cisplatin; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling; si, small interfering.

PRDX3-siRNA enhances cisplatin-induced ovarian cancer cell apoptosis

To investigate the impact of the PRDX3-siRNA on cisplatin-induced ovarian cancer cell apoptosis, the effect of cisplatin on the growth of the SKOV3 cells was firstly examined by a cell proliferation assay. The SKOV3 cells were treated with 0, 1, 2, 3 and 4 μg/ml cisplatin for 1, 2, 3, 4 and 5 days, respectively (Fig. 3). The addition of cisplatin to the culture media inhibited the proliferation of the SKOV3 cells in a dose-dependent manner. When the cisplatin concentration was 3 μg/ml, the effect was more evident.

Figure 3 PRDX3-siRNA enhanced cisplatin-induced ovarian cancer cell apoptosis. (A) Effect of cisplatin on SKOV3 ovarian cancer cells. Under the stimulation of various concentration of cisplatin (0, 1, 2, 3 and 4 μg/ml) for 1, 2, 3, 4 and 5 days, respectively, the absorbance of SKOV3 cells was observed by CCK-8 assay. (B) Growth of SKOV3/Si-Mock and SKOV3/Si-1 cells with/without treatment of 3 μg/ml cisplatin was examined by CCK-8 assay. (C) Growth of SKOV3/Si-Mock and SKOV3/Si-1 cells with/without treatment of 3 μg/ml cisplatin was examined by colony formation assay. * vs. SKOV3/Si-Mock group, P<0.05, # vs. No cisplatin-treated group, P<0.05. (D) Graphical representation of the tumor size over time in mice injected with SKOV3/Si-Mock or SKOV3/Si-1 cells with/without treatment of 3 μg/ml cisplatin (n=8). CCK-8, Cell-Counting kit; si, small interfering.

To study whether the PRDX3-siRNA was able to enhance cisplatin-induced ovarian cancer cell apoptosis, the SKOV3/Si-1 and SKOV3/Si-Mock cells were treated with or without 3 μg/ml cisplatin. The cells were then collected and analyzed by cell proliferation (Fig. 3B) and colony formation assays (Fig. 3C). These assays revealed a marked reduction in the cisplatin-treated SKOV3/Si-1 cells compared with the cisplatin-treated SKOV3/Si-Mock cells for which no knockdown effect was observed. The tumor size of the nude mice injected with the cisplatin-treated SKOV3/Si-1 cells was significantly smaller than that of the mice injected with the cisplatin-treated SKOV3/Si-Mock cells (Fig. 3D). This was associated with the enhanced induction of apoptosis, as shown by Fig. 4A.

Cisplatin-induced apoptosis may be enhanced by PRDX3- siRNA via the activation of the caspase, apoptosis- related proteins

The NF-κB-mediated expression of Bcl-2 and Bcl-XL is known to protect cancer cells from apoptosis, whereas Bax, caspase-3 and caspase-9 inhibit cancer cell growth and induce apoptosis. To validate whether PRDX3-siRNA was able to enhance cisplatin-induced apoptosis in the ovarian cancer cells via the activation of the caspase and apoptosis-related proteins, the expression of these proteins was investigated. The pro-apoptosis proteins Bax, caspase-3 and caspase-9 were analyzed in the SKOV3/Si-1 and SKOV3/Si-Mock cells treated with or without cisplatin (3 μg/ml). The caspase-3 protein was activated and the Bax expression was increased in the SKOV3/Si-1 or SKOV3/Si-1 cells treated with cisplatin, but not in the SKOV3/Si-Mock cells treated with or without cisplatin. Additionally, the activation of the caspase-9 protein was observed in the SKOV3/Si-1 or SKOV3/Si-1 cells treated with cisplatin and the partial activation of caspase-9 was observed in the cisplatin-treated SKOV3/Si-Mock cells, but not in the SKOV3/Si-Mock cells.

Elevated levels of the anti-apoptotic proteins Bcl-2 and Bcl-XL confer chemoresistance, therefore the participation of Bcl-2 and Bcl-XL was examined by a western blot analysis. As shown in Fig. 4B, the PRDX3-siRNA was able to markedly downregulate the expression of the Bcl-2 and Bcl-XL proteins in the cisplatin-treated SKOV3/Si-1 cells compared with the SKOV3/Si-Mock cells treated with/without cisplatin. The phosphorylation of NF-κB p65 was examined and the results showed a significant inhibition in the cisplatin-treated SKOV3/Si-1 cells.

Discussion

Ovarian cancer remains a leading gynecological malignancy across the world. As it possesses the highest mortality rate among the gynecological cancers, ovarian cancer has become the most lethal malignancy of the female reproductive system (2). To extend the long-term survival rate, certain improvements have been made, including the modulation of cellular chemosensitivity, the reversal of tumor resistance and the increase in the therapeutic effects of chemotherapy. However, due to the dose-dependent toxicity, eventual tumor recurrence and emergence of drug-resistance in the disease, the therapeutic efficacy remains poor (21). Mounting evidence suggests that cancer cells exhibit increased intrinsic ROS. In order to protect themselves against oxidative stress induced by the ROS, cells use several antioxidants or reductants to maintain the intracellular redox environment in a highly reduced state. The upregulation of antioxidant enzymes, particularly in the mitochondria, would provide cancer cells with certain survival advantages under intrinsically high oxidative conditions (22). As a significant cellular antioxidant, PRDX3 may regulate the physiological levels of H2O2, thus protecting cells against the apoptosis caused by high levels of H2O2. Increased mitochondrial ROS generation and the disturbance of prx production in cancer cells may lead to oxidative stress and hypoxic microenvironments, subsequently leading to the induction of apoptosis (17). Increased expression levels of PRDX3 were therefore assumed to protect tumor cells against the hypoxia microenvironment and drug-induced H2O2-dependent apoptosis.

To the best of our knowledge, several studies have shown that the PRDX3 protein is overexpressed in numerous types of cancer, including lung, breast and prostate cancer and hepatocellular carcinoma (16–18), but few studies have reported the correlation between the overexpression of PRDX3 and the progress of ovarian cancer. The present study examined the association between the expression of PRDX3 and the clinicopathological parameters in various types of ovarian cancers. The results showed that the high expression of PRDX3 in serous ovarian cancer was associated with poorly-differentiated cancer cells. Moreover, FIGO stage IV/metastatic serous ovarian cancer samples were shown to overexpress PRDX3 at the highest level, which suggested that the PRDX3 expression was significantly associated with increasing cancer progression. A similar tendency was evident in the mucinous, endometrioid and clear cell types of ovarian cancer, but the differences between stages I + II and III + IV, among well-, moderately- and poorly-differentiated ovarian cancer were not significant. This result should be further verified due to the limited number of cases. The most common type of ovarian cancer is serous, therefore, inhibiting the mitochondrial antioxidant enzyme PRDX3 in the ovarian cancer cells may act as an effective cancer therapeutic method.

Cisplatin has been shown to induce apoptosis in various types of cancer cells. Despite the great efficacy at treating these certain types of cancer, the side-effects or resistance to cisplatin undermine its curative potential (23). Based on the the previously stated factors, we assumed that downregulation of the PRDX3 protein would enhance cisplatin-induced ovarian cancer cell apoptosis. Therefore, ovarian cancer cells stably transfected with PRDX3-siRNA were first established by the RNAi approach, in which the target gene was efficiently knocked down. In the present study, the SKOV3 and OVCAR-3 cell lines were stably transfected with PRDX3-siRNA simultaneously, but the transfection efficiency of OVCAR-3 was not high. This may have been affected by the dose or toxicity of the transfection reagent or by the transfection time. The gene-silenced clones of the SKOV3 cells transfected with the PRDX3-siRNA were screened by a limited dilution method and verified by western blot analysis (Fig. 1B). Functional analysis using the siRNA of PRDX3 strongly supported the involvement of PRDX3 in the development and progression of ovarian cancer. As shown in Fig. 2, the proliferation rate of the SKOV3 cells following PRDX3 knockdown was reduced significantly, whereas the proliferation of the SKOV3/Si-Mock cells was not inhibited. To further explore whether the silencing of PRDX3 reversed the resistance of the ovarian cancer cells to cisplatin, a series of experiments were designed. The results showed that when comparing the proliferative activity of the SKOV3/Si-Mock cells treated with cisplatin and the SKOV3/Si-1 cells, the proliferation of the SKOV3/Si-1 cells treated with cisplatin significantly decreased (P<0.05). The enhanced antitumor efficacy in vivo was associated with the enhanced induction of apoptosis, as verified by the TUNEL analysis (Fig. 4). All the data suggested that the combination of the PRDX3-siRNA and cisplatin showed a synergistic anti-cancer effect.

PRDX3 is one of the most significant antioxidant enzymes that modulates intracellular signaling pathways correlated to apoptosis and cell proliferation. Understanding the anticancer mechanism of PRDX3, with the goal of enhancing its efficacy as a valuable adjunct or single agent in anticancer therapy, is extremely important. NF-κB has been implicated in carcinogenesis and the development of drug resistance in cancer cells (24). A previous study indicated that NF-κB was the most reliable target of the combination chemotherapeutic treatment to overcome drug resistance (25). It is well-known that the NF-κB-mediated expression of Bcl-2 and Bcl-XL protects cancer cells from apoptosis, whereas Bax, caspase-3, and caspase-9 inhibit cancer cell growth and induce apoptosis. Therefore, whether PRDX3-siRNA in combination with cisplatin is able to disrupt the NF-κB pathway or not is extremely significant. The present study showed that a significant inhibition of the phosphorylation of NF-κB p65 and a decrease in the levels of the anti-apoptotic proteins, Bcl-2 and Bcl-XL, were induced in the cisplatin-treated SKOV3/Si-1 cells. By contrast, an increase in the level of Bax and the activation of caspase-3 and caspase-9 were observed in the cisplatin-treated SKOV3/Si-1 cells.

In conclusion, these results suggest that silencing the expression of PRDX3 may enhance cisplatin-induced ovarian cancer cell apoptosis through the suppression of the NF-κB signaling pathway. This provides new evidence for the potential application of PRDX3-siRNA as a chemosensitizer in ovarian cancer therapy.

Acknowledgements

The authors would like to thank Junbo Hu for his pathological assistance and technical support.

References