Investigation of cytotoxic and apoptotic effects of disodium pentaborate decahydrate on ovarian cancer cells and assessment of gene profiling

After revealing the anti-cancer properties of boron, which is included in the category of essential elements for human health by the World Health Organization, the therapeutic potential of boron compounds has been begun to be evaluated, and its molecular effect mechanisms have still been among the research subjects. In ovarian cancer, mutations or amplifications frequently occur in the PI3K/Akt/mTOR pathway components, and dysregulation of this pathway is shown among the causes of treatment failure. In the present study, it was aimed to investigate the anti-cancer properties of boron-containing DPD in SKOV3 cells, which is an epithelial ovarian cancer model, through PI3K/AKT/mTOR pathway. The cytotoxic activity of DPD in SKOV3 cells was evaluated by WST-1 test, apoptotic effect by Annexin V and JC-1 test. The gene expressions associated with PI3K/AKT/mTOR pathway were determined by real-time qRT-PCR. In SKOV3 cells, the IC50 value of DPD was found to be 6.7 mM, 5.6 mM, and 5.2 mM at 24th, 48th and 72nd hour, respectively. Compared with the untreated control group, DPD treatment was found to induce apoptosis 2.6-fold and increase mitochondrial membrane depolarization 4.5-fold. DPD treatment was found to downregulate PIK3CA, PIK3CG, AKT2, IGF1, IRS1, MAPK3, HIF-1, VEGFC, CAB39, CAB39L, STRADB, PRKAB2, PRKAG3, TELO2, RICTOR, MLST8, and EIF4B genes and upregulate TP53, GSK3B, FKBP8, TSC2, ULK1, and ULK2 genes. These results draw attention to the therapeutic potential of DPD, which is frequently exposed in daily life, in epithelial ovarian cancer and show that it can be a candidate compound in combination with chemotherapeutics.


Introduction
Ovarian cancer (OC), which is among the malignant gynecological tumors, ranks first in deaths due to cancer [1]. Epithelial ovarian cancer (EOC) constitutes 90% of OC cases [2]. Since 37% of EOC patients are diagnosed at grade III and 28% at grade IV, 5-year survival rates are limited to 41 and 20%, respectively [3]. In the treatment of EOCs, diagnosed at a late stage, the treatment modalities consisting of cytoreductive surgery and platinum or taxane-derived chemotherapeutics have been used as standard. However, after treatment with taxane and platinum derivatives, remission occurs in 60-80% of patients, and in the next stage, recurrence simultaneously with chemotherapy resistance occurs in 80% of these patients [4]. Therefore, the need for specific new therapeutic agents that can delay relapse, prolong survival and provide chemotherapeutic sensitivity in incurable advanced ovarian cancer is increasing day to day.

Cytotoxicity test
The cytotoxic effect of DPD on the SKOV3 cell line was evaluated using Water Soluble Tetrazolium Salt 1 (WST1) [4(3(4iodophenyl) 2(4nitrophenyl)2H5tetrazolio)1,3 benzene disulfonate; Cat. No. 11644807001; Roche Applied Science, Indianapolis, IN]. SKOV3 cells were seeded as 1 × 10 4 cells in triplicate into each well of the 96-well plate and the cells were allowed to adhere to the plate surface for 24 h. Twenty-four hours after seeding the cells, DPD was given to the cells at seven different concentrations in the range of 1.5-20 mM in triplicate for 24, 48, and 72 h incubation times. The untreated group was taken as a control. At the end of the specified periods (24,48, and 72 h), 10 μl of WST-1 solution was added to each well, and the cells were incubated at 37 °C for 3 h.

Annexin V test
The externalization of phosphatidylserine, which is one of the early stages of apoptosis, was detected using the FITC Annexin V Apoptosis Detection Kit (Cat. No: 556547; BD Pharmingen; USA). The cells (3.5 × 10 5 ) for the DPD treatment and untreated control group were seeded in each well of the 6-well plate and incubated for 24 h for the cells to adhere to the plates. SKOV3 cells were treated with the determined IC 50 dose of DPD for 48 h, and the cells were removed with trypsin and washed 2 times with PBS. To detect apoptotic and necrotic cells, Annexin V/PI dye was added to both groups and incubated for 15 min at room temperature. Measurements were performed using Flow cytometry (BD Accuri C6 flow cytometry; Becton Dickinson, Franklin Lakes, NJ) using FL1-A and FL2-A filters. The groups to which the DPD was not administered were taken as the control group.

JC-1 test
Following treatment with DPD for 48 h, the cells were lifted with trypsin and washed twice with PBS. After administration of JC-1 dye to each group, all groups were incubated for 30 min at 37 °C in the incubator. Washing was carried out twice by adding 1 × Assay Buffer into the tube containing each sample, and the cells were centrifuged at 400 × g for 5 min. The samples diluted in 1 × Assay Buffer (100 µl) were analyzed by Flow cytometry using FL1-A and FL2-A channels.

Real-time qRT-PCR analysis
Changes in the expression of the target genes in the PI3K/ Akt/mTOR pathway were analyzed after SKOV3 cells were treated with the IC 50 value of DPD. After treatment with DPD, total RNA isolation from the cells was performed using the RNeasy Mini Kit (Cat. No: 74104; QIAGEN) according to the protocol provided by the manufacturer. The concentration and purity of RNA, isolated from SKOV3 cells, were assessed by measuring the absorbance at wavelengths of 260/280 and 230/260 nm in a spectrophotometer. Reverse transcription into cDNA was carried out using RT2 First Strand Kit (Cat. No: 330401; Qiagen, USA) by taking 800 ng of total RNA for each sample. Forward and reverse primers of genes whose expression was examined are given in Table 1. 2 × RT2 SYBR Green Master Mix (Cat. No. 330501; Qiagen, USA), cDNA, forward primer, reverse primer, RNase free water was added into each well. The qRT-PCR reaction was performed in three-step, which are the denaturation (at 95 °C for 30 s), and the annealing/the extension (at 60 °C for 30 s and 72 °C for 30 s) in the Light-Cycler 480 PCR System. In addition to them, the activation of the enzyme (1 cycle at 95 °C for 15 min) was realized.
The housekeeping genes consisting of ACTB and GAPDH were used as array normalization and internal control. Data analysis was carried out using the 2 −ΔΔCt method. The fold regulation of the genes whose expression levels were evaluated is given in Fig. 4.

Statistical analysis
All experiments were performed in triplicate. The data show the mean of the values with the standard deviation. Statistical differences between control and experimental groups were analyzed using one-way ANOVA in the Graphpad v.5.0

DPD causes programmed cell death in SKOV3 cells
Following SKOV3 cells were treated with DPD, apoptotic activity was evaluated by the Annexin V method. After DPD treatment, early and late apoptosis was found to be 0.8 and 3.5% in the untreated control group, respectively. Early apoptosis increased by 10.6% in the DPD-treated group, while the rate of late apoptosis was 0.6%. In line with these results, DPD application increased apoptosis by 11.6% compared to the untreated control group. Thus, we found an apoptosis increase of 2.6-fold in DPD-treated SKOV3 cells compared to the control (Fig. 2).

DPD induces apoptosis in association with mitochondrial membrane depolarization
After treatment of SKOV3 cells with DPD, it was determined that fluorescence increased in the FL-1 channel with accumulation in the cytoplasm as a result of depolarization of the JC-1 dye in the mitochondrial membrane. While an apoptosis rate of 7.2% was observed in the untreated group incubated with fresh medium, DPD application increased the cell population in the FL-1 channel and the ratio of apoptotic cells reached 32.8%. Therefore, DPD increased apoptosis 4.5-fold due to the change in mitochondrial membrane potential in ovarian cancer cells (Fig. 3).

Discussion
The relapse and chemoresistance that occurred after treatment with taxane and platinum derivatives, used as chemotherapeutic in ovarian cancer, orientated researchers to search for new therapeutic targets. However, targeting a single oncogene or a single mutation with a single drug in ovarian cancer has been an approach that limits the chance of treatment success. Because ovarian cancer has been defined as a heterogeneous disease with different molecular profiles. Especially epithelial ovarian cancer consists of four subtypes with different backgrounds such as Endometrioid (PTEN 40%, PIK3CA 20%, and CTNNB1 40% mutations), Clear cell (PIK3CA 35% mutation and MET 25% amplification), Lowgrade serous (KRAS 40%, BRAF 5%, HER2 15% mutation) and Mucinous (KRAS 50%, BRAF 5%, HER2 15% mutation). Even though this heterogeneity indicates that a single target is a limiting factor in ovarian cancer, there are pathways in which aberrations are concentrated in each subtype.  -CAT GCC CTA TGC GAC CTG AT-3'  ACTB  Forward 5′-GTT GCT ATC CAG GCT GTG -3′  Reverse 5′-TGA TCT TGA TCT TCA TTG TG-3′  GAPDH Forward 5'-CTG ACT TCA ACA GCG ACA CC-3'  Reverse 5'-TAG CCA AAT TCG TTG TCA TACC-3' SKOV3 cells represent low-grade endometrioid and clear cell subtypes, and in these two subtypes, activating/inactivating mutations and amplifications are concentrated in the PI3K/AKT/mTOR pathway, often causing dysregulation of this pathway [18]. Besides, in patients with ovarian cancer, copy number variations of the genes encoding both p110a (PIK3CA) and p110β (PIK3CB) subunits of PI3K indicate a poor prognosis, and an increase in the number of copies in the AKT gene is associated with decreased survival. AKT and mTOR phosphorylation levels were determined to be negative prognostic factors independent of the course of the disease. Considering all these, the drug candidates to be used in the treatment of ovarian cancer exhibit activity through this pathway, which may provide an advantage in terms of prevention of both pro-survival signaling and relapse, utilizing the control of multiple signal transductions [19]. To activate the PI3K/Akt/mTOR pathway, the autophosphorylation of the tyrosine residues of the receptor, and the recruitment of multiple adapter proteins such as insulin receptor substrates (IRS) and Shc to the internal site of the receptor occur by binding of growth factor system such as IGF-1 primarily to the tyrosine kinase receptor. Subsequently, Shc-mediated Ras/MAPK signaling is activated by phosphorylation of adapter proteins; Phosphatidyl-inositol-3-kinase (PI3K) is taken up to the membrane via IRS1/2, and then phosphatidylinositol-3,4,5-bisphosphate (PIP2) is phosphorylated to phosphatidylinositol-3,4,5-triphosphate (PIP3) [20]. Following the uptake of AKT to the cell membrane by PIP3, phosphorylation of AKT by PDK2, DNAPK, ATM, ATR, and mTORC2 on Ser473 residue and by PDK1 on Tyr308 residue mediates the initiation of a series of cellular processes. Activated AKT allows Rheb GTPase to activate mTORC1 to inactivate the TSC1-TSC2 complex. Afterward, mTORC1 inhibits autophagy by suppressing ULK1, while inactivation of 4E-BP1 supports protein synthesis with activation of S6K [21]. On the other hand, while  AKT provides the regulation of GSK3B, FOXO, p27, and MDM2, which are responsible for cell cycle and proliferation, all these proteins enable AKT to be activated again with the feedback mechanism, allowing the process to repeat itself [22]. Also, while IRS1 is suppressed by active mTOR/ S6K under normal physiological conditions, IRS1, whose basal inhibition disappears in case of mTORC1 inhibition, supports PI3K activation via IGFR with a feedback mechanism.
In the present study, the cytotoxic activity of boron compound DPD in SKOV3 cells was investigated and it was determined that the half-maximal inhibitory concentration decreased depending on the dose and time. The boron content of DPD produced by the boron institute was prepared as 19% (3.32 mg). The daily consumption of boron was determined to be 1-13 mg/day (92.49 µg-1.2 mM) in adults according to the World Health Organization. Based on the Dietary reference intake upper limit, daily consumption of boron has been shown to be 20 mg (1.85 mM) for adults, 17 mg (1.57 mM) for the 14-18 age group, and 11 mg (1.01 µM) for the 9-13 age group [23]. Therefore, this result shows that the boron content in the DPD compound has a cytotoxic effect in SKOV3 cells even at a safe dose within the daily consumption range. Thus, DPD may have a protective effect against ovarian cancer.
Many studies have shown that high IGF1 protein levels in the serum of patients with ovarian cancer are associated with cancer cell proliferation and progression [24]. Based on this, when we examined the expression levels of IGF-1, the ligand of the tyrosine kinase receptor, and IRS1 providing activation of the receptor with a negative feedback mechanism, we found that DPD downregulates two leading genes in the activation of the PI3K/AKT/mTOR pathway. It has been supported by studies that the AKT2 gene is increased in mRNA level and gene amplification in 36.3% of ovarian cancers [25]. In a study performed by Noske et al., it was stated that proliferation was reduced in OVCAR-3 cells when AKT2 was suppressed by siRNA [26]. In the current study, it was determined that the expression of AKT2 was reduced after the treatment of SKOV3 cells with DPD, in correlation with the downregulations in the upstream region. GSK3B, which is known to play a role in the progression of epithelial ovarian cancer, is located in the downstream region of AKT and is frequently inactivated by AKT by being phosphorylated on its N-terminal region [27]. Thus, GSK3B phosphorylation causes the accumulation of betacatenin, which acts as a transcription factor for the induction of transcription of genes responsible for proliferation and cell growth, such as MYC, and Cyclin D [25]. Based on this, in this study, expression of GSK3B was decreased following DPD administration. Thus, there may be a relation between the downregulation of AKT2 and GSK3B. Downregulation of genes that are effective in the initiation of PI3K signaling may be among the reasons why DPD decreases cell viability in SKOV3 cells in a dose-and time-dependent manner.
While healthy cells respond to the apoptotic stimulus created by stimulators under physiological conditions, the resistance of cancer cells to this stimulus supports their survival. Inhibition of key components of the PI3K/AKT/ mTOR pathway, which is known to play a role in the development of ovarian cancer and whose abnormal activation induces malignant transformation and chemoresistance, including PI3K, AKT, and mTORC1, supports the inhibition of transcription of anti-apoptotic genes by breaking this survival resistance [28,29]. In this regard, Shayesteh et al. has shown that inhibition of apoptosis caused by the increase in PIK3CA copy number is one of the reasons for survival in SKOV3 cells with PIK3CA mutation, and the increased PI3 kinase activity transforms the phenotype to a more malignant profile in the following processes [30]. Huang et al. has determined that when AKT was targeted with siRNA in SKOV3 cells, apoptosis was induced tenfold, in subsequent studies, it was observed that AKT2 mediated the induction of apoptosis through oxidative stress [31,32]. MDM2, one of the downstream targets of AKT, ensures the ubiquitination and degradation of p53. Gao et al. showed that apoptosis stimulation occurs as a result of suppression of AKT in ovarian cancer cell lines, which activates p53 [33]. In this study, we found that after the treatment of SKOV3 cells with DPD, genes encoding PIK3CA and PIK3CG, which are key components that cause exacerbation of PI3K/AKT signaling, are downregulated, and TP53, one of the important tumor suppressor genes in inducing apoptosis, is upregulated. These results suggest that DPD exposure may cause downregulation of key genes encoding proteins located in the upstream region of the PI3K/AKT pathway, thus reducing the requirement for increased expression of the downstream components. eIF4B, the target of mTORC1, mediates the initiation of translation by providing ribosome recruitment to mRNA with structured 5′UTRs in the initiation step, thus forming auxiliary links between the mRNA and the 40S ribosome. Shahbazian et al. has indicated that silencing eIF4B inhibited the translation of mRNAs of anti-apoptotic genes (BCL-2, XIAP) and induced apoptosis in a caspasedependent manner [34]. In the present study, DPD caused a dramatic reduction in the expression of the eIF4B gene in an ovarian cancer model. Based on these results, which showed the changes in the expressions of genes regulating apoptosis in the PI3K/AKT/mTOR pathway after treatment of SKOV3 cells with boron compound, it was determined that apoptosis was induced significantly. As a result, DPD may have induced apoptosis in ovarian cancer cells by targeting components of the PI3K/AKT/mTOR pathway.
The PI3K/AKT/mTOR signaling pathway controls the translation of some important proteins that regulate mitochondrial functions such as protein synthesis and mitochondrial electron transport via mTORC1. Therefore, targeting mTORC1 and its components causes changes in mitochondrial membrane potential [35]. Hu et al. has determined that dual inhibition of PI3K and mTOR in SKOV3 cells resulted in a decrease in membrane potential with decreased JC-1 aggregate form and increased monomer form [36]. In this study, SKOV3 cells exposed to DPD were found to cause downregulation of the IRS1 gene, upregulation of the gene encoding the mTOR inactivating protein TSC2, and a decrease in the expression of the genes encoding mTORC1 components MLST8 and TELO2 and downregulation of the mTORC1 antagonist FKBP8 gene [37].
The boron compound can induce mitochondrial membrane depolarization, as the lack of proteins to form complexes will hinder the functioning of the mTORC1 complex, even if the expression of mTOR is increased. To confirm this, it was shown the membrane potential to decrease compared to the untreated control group in accordance with the study of Hu et al., when the potential effect of the boron compound on mitochondrial membrane potential in ovarian cancer cells was evaluated. Thus, the apoptosis that occurred may be arisen from the intrinsic pathway by inducing mTORC1/eIF4B. The PI3K/AKT/mTOR pathway simultaneously regulates both autophagy and apoptosis [38]. In physiological conditions, the active mTORC1 complex suppresses autophagy, another death pathway, by phosphorylation of ULK1. As a result of the inactivation of mTORC1, the ULK1/ATG13/ FIP200 complex was formed due to the inability of the complex members to collect, which are compressed by TSC1/2, or the AMPK activation under starvation stress conditions. The autophagic process begins with the autophosphorylation of ULK1 or its phosphorylation by AMPK [39,40]. While mTORC1 has a pivotal role in the inhibition of autophagy, mTORC2 is known to hinder autophagosome formation indirectly via AKT or AKT/mTORC1 after complexing with RICTOR. In addition, GSK3B can prevent the assembly of the mTORC2 complex by phosphorylating RICTOR [41]. In ovarian cancer cells treated with DPD, the RICTOR gene, which is involved in the assembly of the mTORC1 and mTORC2 complex, was downregulated, while GSK3B, which was involved in decomposing the complex, was upregulated. In addition, the expression of ULK1 and ULK2 genes involved in the autophagosome membrane formation was increased. These results may signify that the autophagic machinery of DPD may also be activated in correlation with apoptosis.
Especially metabolic plasticity represents one of the important phenomena in the malign progression of behavior in cancer cells [42]. In this process, metabolic conditions become suitable for malignancy development due to crosstalk between PI3K/AKT/mTOR, HIF-1 and AMPK. AMPK becomes active to stabilize the ATP by increasing the ATP level by anaerobic glycolysis or mitochondrial oxidative metabolism under stress conditions that cause ATP depletion such as hypoxia or food starvation. Cells adapt to starvation and low energy conditions, reprogramming cells with transcriptional changes so that they can grow and differentiate [43]. Therefore, AMPK, which consists of three subunits, PRKAA, PRKAB, and PRKAG, is phosphorylated by the CAB39, LKB1 and STRADB complex [44,45]. HIF-1 degraded in normoxic conditions becomes stable by increasing transcription in hypoxic conditions and plays a role as a transcription factor in inducing the transcription of many genes responsible for the proliferation, resistance development, and metastasis of tumor cells to adapt to hypoxic conditions [46,47]. While PI3K/AKT increases the expression of HIF-1 at the transcriptional level, the mTOR complex plays a role in the translation of HIF-1 mRNA [48]. AMPK regulates the posttranslational modification of the HIF-1 protein [49]. HIF-1 activated after all these processes induces the expression of several genes, including DDIT4 and VEGF [50,51]. In the present study, it was determined that the expression of HIF-1, which is located downstream of PIK3CA, PIK3CG, and AKT2 decreased after treatment with the boron compound. Besides, the downregulation of MAPK3, which regulates the expression of HIF-1, was found to be correlated with the decrease in the expression of HIF-1 in present study. Considering the downregulation of mTORC1 components, as well as the decreased expression of CAB39, CAB39L, PRKAB2, PRKAG3 (components of the AMPK complex), DDIT4 and VEGFC (target genes of HIF-1), DPD treatment in SKOV3 cells may indicate that transcriptional, translational and posttranslational activity of HIF-1 can be targeted.
In conclusion, in this study, we evaluated for the first time, the effect of DPD, which is a compound of boron mineral, in an epithelial ovarian cancer model. DPD exhibited anti-cancer activity through the PI3K/AKT/mTOR pathway, which is frequently altered in epithelial ovarian cancer. Considering all these results, it has been shown that DPD is a candidate compound to be evaluated as a drug that can be used in co-treatment with chemotherapeutics frequently used in ovarian cancer treatment or as a priming agent for chemotherapeutics.