H-Ferritin Affects Cisplatin-Induced Cytotoxicity in Ovarian Cancer Cells through the Modulation of ROS

Reactive oxygen species (ROS) mediates cisplatin-induced cytotoxicity in tumor cells. However, when cisplatin-induced ROS do not reach cytotoxic levels, cancer cells may develop chemoresistance. This phenomenon can be attributed to the inherited high expression of antioxidant protein network. H-Ferritin is an important member of the antioxidant system due to its ability to store iron in a nontoxic form. Altered expression of H-Ferritin has been described in ovarian cancers; however, its functional role in cisplatin-based chemoresistance of this cancer type has never been explored. Here, we investigated whether the modulation of H-Ferritin might affect cisplatin-induced cytotoxicity in ovarian cancer cells. First, we characterized OVCAR3 and OVCAR8 cells for their relative ROS and H-Ferritin baseline amounts. OVCAR3 exhibited lower ROS levels compared to OVCAR8 and greater expression of H-Ferritin. In addition, OVCAR3 showed pronounced growth potential and survival accompanied by the strong activation of pERK/pAKT and overexpression of c-Myc and cyclin E1. When exposed to different concentrations of cisplatin, OVCAR3 were less sensitive than OVCAR8. At the lowest concentration of cisplatin (6 μM), OVCAR8 underwent a consistent apoptosis along with a downregulation of H-Ferritin and a consistent increase of ROS levels; on the other hand, OVCAR3 cells were totally unresponsive, H-Ferritin was almost unaffected, and ROS amounts met a slight increase. Thus, we assessed whether the modulation of H-Ferritin levels was able to affect the cisplatin-mediated cytotoxicity in both the cell lines. H-Ferritin knockdown strengthened cisplatin-mediated ROS increase and significantly restored sensitivity to 6 μM cisplatin in resistant OVCAR3 cells. Conversely, forced overexpression of H-Ferritin significantly suppressed the cisplatin-mediated elevation of intracellular ROS subsequently leading to a reduced responsiveness in OVCAR8 cells. Overall, our findings suggest that H-Ferritin might be a key protein in cisplatin-based chemoresistance and that its inhibition may represent a potential approach for enhancing cisplatin sensitivity of resistant ovarian cancer cells.


Introduction
The oxygen-containing reactive species (ROS) are unstable by-products of cellular metabolism that are essential for several biological processes including mitochondrial and plasma membrane functioning, cell signalling and immune response [1]. The rate and magnitude of ROS production are tightly controlled by an antioxidant defense system (catalases, superoxide dismutase (SOD), glutathione peroxidase (GPx), thioredoxin (Trx), and ferritin heavy subunit (FHC)) that eliminates them over time [2,3]. The imbalance in the circuitries of ROS production and removal leads to impairment of cell signalling, oxidative damage of cell components, and cytotoxicity [1][2][3][4]. On the other hand, recent evidences indicate that ROS are characterized by a dualistic property in determining cell fate [4,5]. A persistent ROS overproduction may induce cellular adaptation as it occurs in many diseases, especially in cancer. Otherwise, an excessive ROS production may give rise to fatal lesions that cause cell deaths [4,5]. Thus, many of the current chemotherapy strategies are aimed at raising ROS over the cytotoxic threshold levels in malignant cells [6][7][8][9][10]. To keep ROS in the prooncogenic zone, cancer cells are provided by an extensive supply of antioxidant molecules that reduce the efficacy of prooxidant drugs and enable tumor cells to acquire chemoresistance [11,12].
Cisplatin is a prooxidant chemotherapeutic agent widely used for the treatment of ovarian cancer [13][14][15][16]. Nevertheless, ovarian cancer cells often develop cisplatin resistance by increasing the expression of the antioxidant systems [14][15][16][17]. Consequently, the final concentration of ROS evoked by cisplatin exposure is crucial for the effectiveness of this prooxidant cancer therapy [17]. For this reason, there is a strong need to develop new therapeutic strategies able to overcome platinum resistance in recurrent and metastatic ovarian cancer.
In the antioxidant enzyme family, the heavy subunit of human ferritin (H-Ferritin, FHC) acts by sequestering iron in a bioavailable and catalytically inactive form thus preventing its accumulation in the intracellular labile pool (LIP) and its participation in ROS-generating Fenton reactions [18]. The role of FHC as an antioxidant protein is underscored by the variety of mechanisms leading to its transcriptional and post-transcriptional upregulation in response to oxidative stimuli [19]. FHC expression is usually altered in cancer cells as reported in lung [20], breast [21,22], melanoma [23], and ovarian cancer cells [24]. However, the possible relationship between FHC amounts and the aptitude of cancer cells to develop chemoresistance to cisplatin is still poorly characterized.
In the current study, we investigated whether and to which extent the modulation of H-Ferritin amounts might affect the cisplatin sensitivity in OVCAR3, a well-established in vitro experimental model of chemoresistant ovarian cancer cells [25], in comparison to OVCAR8 cells. Through a series of assays including FHC knockdown and forced overexpression, we demonstrate for the first time that ferritin heavy subunit, through its ability to modulate ROS amounts, is a key element in determining the response of ovarian cancer cells to cisplatin exposure.

Materials and Methods
2.1. Cell Lines and Cell Culture. We selected as in vitro experimental models two human ovarian cancer lines OVCAR3 and OVCAR8 representative of epithelial ovarian adenocarcinoma. Cells obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). According to ATCC, OVCAR3 cells derive from an ascitic metastatic site and are an appropriate model system to study cisplatin resistance in ovarian cancer. Both the cell lines are characterized by mutations in p53 gene. OVCAR3 and OVCAR8 cells were maintained in RPMI 1640 media (Sigma-Aldrich, St. Louis, MO, USA). Both culture media were supplemented with 10% fetal bovine serum at 37°C in a humidified atmosphere containing 5% CO 2 . Each cell line has been examined for mycoplasma contamination through LookOut® Mycoplasma PCR Detection Kit (Sigma-Aldrich, St. Louis, Missouri, USA).

Patients and Specimens.
We selected a group of 28 patients with High-Grade Serous Ovarian Cancer (HGSC) who were treated at the Unit of Gynaecologic Oncology, Magna Graecia University, Germaneto, and Pugliese-Ciaccio Hospital, Catanzaro, Italy, between April 2013 and March 2016. Tissue and serum samples of patients were retrieved from our biobank to perform analysis of FHC mRNA expression. Inclusion criteria were as follows: availability of clinical data and biological samples; stage II-III-IV HGSC surgically staged. Patients with previous or concurrent cancer located in other sites, known genetic susceptibility to gynecologic or nongynecologic cancers (BRCA1-2 carriers, associated polyposis conditions (APC), Fanconi syndrome) [14], or positive family anamnesis for ovarian and/or breast cancer were excluded. Patients' clinical data are reported in Table 1.
Procedures carried out in this study were in accordance with the guidelines of the Helsinki Declaration on human experimentation and good clinical practice (GCP). Approval by the "Pugliese-Ciaccio" institutional review board (IRB number: AOPC12404) was obtained before starting patient's enrolment. Furthermore, an informed consent was obtained from all patients before processing their data from the time of hospitalization, even if data did not include any personal identifying information. Biological samples consist in  [29]. RNA from paraffin-embedded tissue specimens were obtained by a series of incubation with xylene and subsequent ethanol washes. Absolute qPCR analysis was also used to determine the expression of FHC mRNA in the 28 tumor tissue specimens. FHC expression analysis was performed by using SYBR Green qPCR Master Mix (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Primers used to detect FHC were as follows: FW: 5′ -CATCAACCGCCAGATCAAC-3′ and REV: 5′-GATGGC TTTCACCTGCTCAT-3′. Analysis was performed on QuantStudio 3 Applied Biosystems by Thermo Fisher Scientific. Starting from a sample of known template concentration, a 5-point 10-fold serial standard curve was prepared, and the concentration of all other samples was calculated by simple interpolation of each threshold cycle (Ct) into this standard curve. FHC mRNA expression data are reported as log (quantity, ng) and represent the mean of three independent technical replicates.
2.11. Statistical Analysis. Results are expressed as mean ± SD and analyzed using the unpaired Student's t-test or two-way ANOVA as indicated in the figure legends. GraphPad Prism® version 5.01 was used to calculate EC 50 ; Sidak test was used to identify statistical significance in the EC 50 values. FHC tumor tissue levels in chemoresistant vs. chemosensitive HGSC patients were compared using the nonparametric Kruskal-Wallis test to identify statistical differences between groups. The Kruskal-Wallis test has been chosen as a nonparametric alternative to one-way ANOVA and an extension of the Mann-Whitney U test to allow the comparison of more than two independent groups. p ≤ 0:05 was considered to be significant.

OVCAR3 Cells Exhibit Lower Endogenous ROS and
Higher Endogenous FHC Levels Than OVCAR8 Cells. We selected as in vitro experimental models OVCAR3 and OVCAR8 cell lines representative of epithelial ovarian adenocarcinoma. OVCAR3 has been chosen as experimental model for studying cisplatin chemoresistance [25]. First, we performed ROS analysis in OVCAR3 and OVCAR8 cells by using two fluorogenic probes: CellROX® Green Reagent able to detect total intracellular ROS content and MitoSOX™ Red Indicator able to selectively detect superoxide radicals. As shown in Figures 1(a) and 1(b), fluorescence microscopy highlights that CellROX® Green fluorescence intensity was significantly higher in OVCAR8 cells compared to OVCAR3 cells. Conversely, MitoSOX staining showed only a slightly different intensity between the two cell lines. These results suggest that OVCAR8 cells are characterized by higher levels of total ROS content compared to OVCAR3 cells. The superoxide radical contribution to this different levels appeared inconsistent. Differences in baseline ROS amounts might reflect, in principle, diverse expression of antioxidant enzymes. Thus, we analyzed the expression of FHC protein, belonging to the antioxidant system, in both OVCAR3 and OVCAR8 cells. Representative western blot analyses and relative densitometry reported in Figure 1(c) highlight that the antioxidant protein FHC was consistently more expressed in OVCAR3 compared to OVCAR8 cells. This behaviour was mirrored by the expression of the other two antioxidant enzymes SOD1 and GPx ( Figure S1).

OVCAR3 Cells Show Enhanced Cell Cycle S-Phase and
Cell Growth. We next assessed whether the differences in intracellular baseline ROS amount and FHC protein levels were paralleled by different cancer cell growth. Figure 2(a) shows a representative plot and histograms indicating the mean ± SD of three cell cycle cytofluorimetric analyses, performed by staining cells with PI solution. Results highlight that a significant higher percentage of OVCAR3 cells were in S-phase compared to OVCAR8 cells (S%: 73:9 ± 0:3 vs. 53:9 ± 0:7, p < 0:05). Accordingly, results from the MTT analysis show that OVCAR3 exhibited an enhanced cell growth potential at 24 h, 48 h, and 72 h (Figure 2(b)). As reported in Figure 2(c), OVCAR3 cells were also characterized by consistent overexpression of the specific S-phase cyclin E1 (CCNE1) along with increased expression of the proto-oncogene c-Myc and enhanced phosphorylation of ERK1/2 and AKT compared to OVCAR8 cells. On the contrary, no phosphorylation of the S-phase cyclin-dependent kinase Chk2 (Thr68) was observed in OVCAR3 and OVCAR8 cells. Optical densitometry of each WB analysis is reported in Figure S1.

Cisplatin Treatment Induces Significant FHC Downregulation and ROS Increase Exclusively in
Chemosensitive OVCAR8 Cells. The sensitivity of OVCAR3 and OVCAR8 cells to cisplatin was determined by treating both cell lines with increasing concentrations of the drug (6 μM, 12 μM, 24 μM and 48 μM). After 24 h, we performed the MTT assay to monitor cell viability following treatment. As shown in Figure 3(a), we found that OVCAR3 cells were, overall, more resistant to treatment than OVCAR8 cells (log EC 50 OVCAR3 vs. log EC 50 OVCAR8: 1:46 ± 0:06 vs. 0:85 ± 0:05, p < 0:0001). In particular, at the lowest cisplatin concentration (6 μM) OVCAR8 cell viability was almost halved while OVCAR3 cells were totally unresponsive. DCFDA luminometric analysis highlighted that the extent of ROS accumulation induced by cisplatin treatments in OVCAR8 cells was significantly higher than that induced in OVCAR3 cells at each concentration apart from 48 μM (Figure 3(b)). Next, we observed that the exposure to 6 μM cisplatin induced a clear cleavage of caspase 3 in OVCAR8 cells and not in OVCAR3 cells (Figure 3(c)). Furthermore, FHC protein levels showed a completely different behaviour between the two cell lines; at 6 μM cisplatin, FHC was consistently downregulated in OVCAR8 cells whereas it was almost unaffected in OVCAR3 cells (Figure 3(c)). Optical densi-tometries of WB are reported in Fig. S2. The analysis of SOD1 and GPx protein levels upon 6 μM cisplatin showed, instead, a slight increase in both the cell lines  ( Figure S2). Furthermore, immunofluorescence analysis highlighted that 6 μM cisplatin induced a significant increase in both total ROS and superoxide radical levels in the drug-responsive OVCAR8 cells compared to OVCAR3 resistant cells (Figures 3(d) and 3(e)).

Modulation of Intracellular FHC Levels Affects
Sensitivity to Cisplatin in OVCAR3 and OVCAR8 Cells. Here, we asked whether a change in FHC levels might affect the EOC cell response to cisplatin. To this, we first transiently transfected OVCAR3 cells with a specific FHC siRNA (OVCAR3 siFHC ) or negative control (OVCAR3 Neg Control ) for 48 h. Annexin V/7-AAD cytofluorimetric analysis showed that 6 μM cisplatin was unable to induce a consistent apoptosis in OVCAR3 Neg Control (early apoptosis: 9:70% ± 0:57; late apoptosis: 6:80% ± 0:99). On the contrary, the same drug concentration promoted a significant increase of apoptotic cell death in OVCAR3 siFHC cells (early apoptosis: 45:65% ± 0:78; late apoptosis: 9:80% ± 1:13) compared to either untreated OVCAR3 Neg Control (p < 0:05) or OVCAR3-Neg Control treated with 6 μM cisplatin alone (p < 0:05). Apoptosis assays performed in OVCAR3 Neg Control cells treated with (i) 10 mM N-acetyl-cysteine (NAC) alone and (ii) 10 mM N-acetyl-cysteine (NAC) in combination with 6 μM cisplatin and in untreated OVCAR3 siFHC revealed no considerable changes. Results of three independent biological replicates are reported as mean ± SD in Table 2 as well as in Figure 4(a).
Accordingly, detection with CellROX® Green Reagent revealed that the ROS amounts evoked by 6 μM cisplatin treatment in combination with FHC knockdown in OVCAR3 cells (OVCAR3 siFHC cisplatin 6 μM) were consistently higher than those induced by either cisplatin treatment alone (OVCAR3 Neg Control cisplatin 6 μM) or FHC silencing alone (OVCAR3 siFHC ) (Figure 4(b)). No considerable changes have been observed in OVCAR3 cells treated with 10 mM NAC (OVCAR3 Neg Control NAC 10 mM) in comparison with OVCAR3 Neg Control untreated cells. As expected, NAC treatment reduced ROS accumulation in OVCAR3 cells treated with 6 μM of cisplatin (OVCAR3 Neg Control  (c) Representative WB of c-Myc, cyclin E1 (CCNE1), pChk2 (Thr68), pERK1/2, and pAKT in OVCAR3 and OVCAR8 cells. γ-Tubulin was used as internal control. WB analysis was performed three times and results were reproducible. cisplatin 6 μM/NAC 10 mM). In addition, western blot analysis revealed that 6 μM of cisplatin exposure led to a further downregulation of FHC protein levels in OVCAR3 siFHC cells; conversely, cisplatin left FHC levels unaltered in OVCAR3 Neg Control cells (Figure 4(c)).

FHC Tissue Levels Are Higher in Chemoresistant High-Grade Serous Ovarian Cancer (HGSC) Patients Compared to Chemosensitive
Ones. In light of the in vitro findings, we performed absolute qPCR analysis of FHC mRNA tissue levels in a small cohort of 28 patients with High-Grade Serous Ovarian Cancer (HSGC, stages II, III, and IV) treated with platinum-based chemotherapy, among which 13 were chemoresistant and 15 were chemosensitive. As shown in the box plot in Figure S3, statistical Kruskal-Wallis test suggests that patients with resistance to chemotherapy may be characterized by higher FHC levels compared to the chemosensitive ones. However, data do not reach the statistical significance.

Discussion
Despite considerable efforts for developing novel and more efficient therapeutic strategies, ovarian cancer patients often suffer from aggressive and therapy-resistant disease characterized by poor prognosis and high mortality [14,30,31]. Cisplatin is a prooxidant chemotherapeutic agent largely used as first-line therapy in ovarian cancer; however, its efficacy is quite limited since most patients ultimately die with platinum-resistant disease [30,31]. Numerous evidences indicate that altered redox balance, which is now widely considered as one of the main cancer hallmarks, can be pivotal in the resistance to antitumor agents including cisplatin [1,2,4].
As a consequence of genetic, metabolic, and microenvironment-related aberrations, cancer cells are subjected to persistent prooxidant stimuli that ultimately increase baseline ROS levels and promote tumor growth by inducing genomic instability and metabolism reprogramming [32]. However, tumor cells have developed an efficient ROS detoxification system through which they gain advantage when subjected to further prooxidant conditions. This dependency from the antioxidant systems represents a specific vulnerability so as the current used prooxidant chemotherapeutic agents act by increasing oxidative stress above the toxicity threshold [1][2][3][4][5].
In this study, by analyzing changes of intracellular ROS levels, we explored the role of FHC, an important antioxidant enzyme, in the development of resistance to cisplatin-based therapy in ovarian cancer cells.
FHC, the heavy subunit of the human ferritin, has a ferroxidase activity through which it safely stores iron in catalytically inactive Fe 3+ form thus tightly controlling the homeostasis of the labile prooxidant iron pool [33]. We and others have previously demonstrated that FHC is a downstream effector of NFkB-mediated inhibition of the oxidative stress-induced apoptosis [26,34]. In addition, FHC is transcriptionally upregulated by the antioxidant transcription factor Nrf2 to maintain iron and redox homeostasis [35,36].
As an in vitro experimental model, we selected OVCAR3 and OVCAR8 cell lines as representative of ovarian cancer cells. In particular, OVCAR3 cells have been selected as cisplatin refractory cell line established from metastatic ascites of a patient with ovarian adenocarcinoma [25].
Overall, our results strongly suggest that the antioxidant properties of FHC play a key role in determining the response of ovarian cancer cells to cisplatin treatment. First, we observed that the chemoresistant OVCAR3 cells are characterized by higher constitutive FHC levels and by lower endogenous ROS content in comparison to OVCAR8 cells. Moreover, prooxidant cisplatin treatment affects FHC levels in OVCAR8 cells by inducing its downregulation while it leaves unchanged FHC amounts in OVCAR3 cells. These effects appear to be selective for FHC since other two antioxidant enzymes, namely SOD1 and GPx, were not consistently modified upon cisplatin exposure. Recent evidences indicate that H-Ferritin may undergo degradation in cells exposed to anticancer compound and this is accompanied by intracellular iron accumulation and increase in iron-dependent ROS production [37]. Indeed, we noticed that in OVCAR8 cells the downregulation of FHC mediated by cisplatin exposure was accompanied by an enormous increase in ROS accumulation that likely exceed the cytotoxic threshold levels. Conversely, in OVCAR3 cells the cisplatin hit was insufficient to push ROS production over the cytotoxic levels.
In the past decades, three main approaches have been proposed to exploit the cancer cell killing potential of ROS: (i) enhancing the generation of ROS in tumor cells by increasing the dose of a single prooxidant chemotherapeutic drug, (ii) combination of conventional anticancer agents with natural Table 2: Data analysis of Annexin/7-AAD cytofluorimetric apoptosis assays in OVCAR3 cells.

Samples
Early apoptosis (% ± SD) Late apoptosis (% ± SD) Live cells (% ± SD)     compounds that increase ROS production, the so-called "ROS+ROS concept," and (iii) inhibition of the antioxidant defense system of tumor cells. Remarkably, the first two approaches are very challenging to translate from in vitro models to in vivo conditions because of significant side effects [10,38,39]. On the contrary, disabling key antioxidant systems in the presence of ROS inducers represents the most promising new anticancer strategy in resistant tumor cells. Indeed, impairing antioxidant capacity, such as Nrf2, SOD, and GPx, has emerged as a good strategy to target many cancer types [40]. Here, we proved, for the first time, that modulation of intracellular H-Ferritin (FHC) protein is able to condition ovarian cancer cell response to cisplatin thus adding this molecule to the targetable antioxidant protein panel. The relevance of FHC/ROS axis in modulating ovarian cancer cell response to cisplatin has been demonstrated by FHC knockdown or forced overexpression in our in vitro system. The knockdown of FHC, by using a specific siRNA, is accompanied by a significant augment in the cytotoxic effects of cisplatin in the drug-resistant OVCAR3 cells. Accordingly, an overexpression of FHC in the drug-sensitive OVCAR8 cells suppresses the cytototoxic effects of the drug at a level comparable to that obtained by scavenging ROS through NAC treatment. The fundamental mechanism through which FHC knockdown is able to restore OVCAR3 sensitivity to cisplatin appears to be strictly related to its antioxidant properties and to its capacity to lead the effective final amounts of ROS over those evoked by the cisplatin treatment alone. At last, we also analyzed FHC cancer tissue levels in 28 patients with HGSC receiving a platinum-based chemotherapy. Although suggestive of a possible association between high levels of FHC and chemoresistance, the collected data do not reach the statistical significance. This trend prompted us to increase, in future studies, the analyzed cohort of patients to provide additional arguments in favor of the importance of estimating ROS amounts and FHC status to improve the therapeutic outcomes in treatment of ovarian cancer.

Conclusions
In conclusion, our data demonstrate for the first time the association of FHC/ROS axis with cisplatin resistance in ovarian cancer cells. Furthermore, we propose that inhibition of FHC might be a potential approach for restoring cisplatin sensitivity of resistant ovarian cancer cells. The conjugation of siRNA carrier system with ligands that exhibit high affinity to specific receptors overexpressed in ovarian cancer cells could make feasible this approach also in vivo.

Data Availability
The data used to support the findings of this study are included within the article.

Samples
Early apoptosis (% ± SD) Late apoptosis (% ± SD) Live cells (% ± SD) WB analysis was performed three times and results were reproducible. Fig. S4: analysis of FHC tumor tissue levels in 28 HGSC patients. Representative images of ovarian cancer tissue specimens (left). Box plot depicting FHC mRNA levels, expressed as log (ng), as assessed by absolute qPCR analysis in cancer tissue specimens of chemoresistant HGSC patients (n = 13) and chemosensitive HGSC patients (n = 15) (right). Data are not statistically significant (N.S.). (Supplementary Materials)