Drug resistance is a major cause for ovarian cancer recurrence. New drug discovery requires significant resources and time. Alternatively, the concept of ‘drug repurposing’ appears promising. In the present study, we explored the antitumor potential of BT in pre-clinical ovarian cancer model. BT was tested against a panel of ovarian cancer lines exhibiting varying sensitivities to cisplatin. Our results demonstrate the cytotoxic effects of BT towards all the ovarian cancer cells lines tested with IC50 values ranging from 19 μM to 60 μM, at 72 hrs post treatment. Interestingly, BT IC50 values were almost indistinguishable between cisplatin-sensitive and cisplatin-resistant variants of isogenic ovarian cancer cell line pairs, although cisplatin IC50 values varied significantly. These results are significant when considering that clinically, all recurrent ovarian cancers will eventually be platinum-resistant. Interestingly, BT IC50 values observed for various ovarian cancer cell lines are significantly below the clinically tolerable doses of BT for humans. In several published studies, chronic BT dosing up to 50 mg/kg every other day was well tolerated with the 40 mg/kg dose level best tolerated. Fifty mg/kg in three divided alternate daily doses for 5 days will maintain serum levels of BT in the range of 140 to 550 μM in rabbits, dogs and humans [13, 18]. Based on the fact that BT exerts similar cytotoxic effects on cisplatin-sensitive and resistant ovarian cancer cell lines with clinically tolerable IC50 values, it is reasonable to speculate that BT may be useful in halting ovarian cancer cell growth irrespective of the sensitivity that cells may display to cisplatin, and this merits further exploration.
It is well known that invalid apoptosis pathway has often been one of the hallmarks of cancer cells and an important cause of resistance to cytotoxic agents . It is therefore essential to focus on type of cell death induced by therapeutic agents. Ability to induce apoptosis is a critical factor for effective treatment against cancer . Previous reports show the inhibitory effect of BT on cervical cancer cell growth via induction of caspase 3/7 activity . Our results also indicate that ovarian cancer cells undergo apoptosis upon BT treatment initially at lower concentrations. Hallmarks of apoptosis, such as nuclear condensation, DNA fragmentation, and loss of mitochondrial potential, were observed further demonstrating that BT triggers apoptosis in ovarian cancer cells. However, at higher concentrations, no caspase activity was detected while LDH was detected, indicating that cells die via necrosis at higher concentrations. The ability of BT to induce cell death via apoptosis makes this drug a good candidate for the treatment of ovarian cancer.
This study also demonstrates that BT induces apoptosis in ovarian cancer cells via activation of proteolytic effector caspases such as Caspase 3 and 7 and subsequent cleavage/inactivation of PARP-1 (involved in DNA repair). Apoptosis is known to be mediated by two pathways, the extrinsic (death receptor) and the intrinsic (mitochondrial). The majority of anticancer (cytotoxic) drugs induce apoptosis via the intrinsic (mitochondrial - cytochrome c/Apaf-1/caspase-9 pathway) [21, 22]. Mitochondria are considered to be both a source and a target of ROS. Although we did not focus on which apoptotic pathway was induced by BT, decreased mitochondrial transmembrane potential following BT treatment implicates the intrinsic (mitochondrial) pathway. Disruption of mitochondrial potential can lead to oxidation of mitochondrial pores by ROS, resulting in release of cytochrome C into the cytosol . Cytochrome C, Apaf1 (apoptotic protease activating factor-1) and dATP then form an apoptosome to which procaspase-9 is recruited and activated. Caspase-9 in turn activates downstream effector caspases −3 and −7 which execute the final steps of apoptosis.
We observed a switch from apoptosis to necrosis with increasing BT concentrations. Apoptosis is a carefully regulated, energy-dependent process that involves a complex cascade of events resulting in cell death. It is dependent on availability of ATP, which in turn depends on the correct function of mitochondria. As mentioned in our manuscript, BT causes mitochondrial transmembrane depolarization, thus affecting mitochondrial function. This disruption may cause ATP depletion to a level that is insufficient for cell survival, thus switching from apoptosis to necrosis. Additionally, reactive oxygen species (ROS) are known to cause apoptosis or necrosis, depending on the amount and type of ROS generated . We postulate that high concentrations of BT lead to increased ROS, ultimately causing severe cellular injury. High levels of ROS can inhibit apoptosis by inactivating caspases by oxidation of their thiol groups. Furthermore, ROS can affect mitochondrial energy (ATP) production causing depletion of ATP. These events would ultimately switch cells to necrosis.
Inhibition of the cell cycle is a known target for the treatment of cancer [25–28]. Anticancer agent may cause cell cycle arrest via altering the regulation of cell cycle machinery. Various regulatory proteins, including cyclin E, cyclin D1, cyclin D2, cyclin A, CDK2, CDK4 and the CDK inhibitors p27Kip1 and p21Cip1 are known to regulate cell cycle. It is well known that kinase activities of CDK-cyclin complexes are essential for progression of cell cycle at many check points [29–31]. p21Cip is regarded as universal inhibitor of cyclin-CDK complexes [29–31], thus blocking the entry of cells at the G1-S-phase transition checkpoint and induce apoptosis . Our data demonstrate that BT treatment resulted in G1-phase cycle arrest and up-regulation of the expression of p27Kip1 and p21Cip1. Increased expression of CDK inhibitors p21cip1 and p27kip1 may result in increased association with CDKs, thus inhibiting their activity. The cascade of downstream events in response to BT treatment may lead to blockage of the cell cycle at the G1-to-S phase transition, and thus halting ovarian cancer cell growth. Additionally, cell cycle arrest following BT treatment could be ROS mediated. We showed that BT enhanced ROS generation. ROS mediated inactivation of CDKs by via oxidation  and enhanced expression of p21 can cause cell cycle arrest in G1- and S-phases resulting in reduced cellular proliferation. ROS mediated DNA damage is known to cause stabilization and elevation of known tumor suppressor protein, p53, which in turn induces and enhances the synthesis of p21 . As mentioned earlier, p21 is known inhibitor of CDK activity. These observations suggest that cell cycle regulation is one of the mechanisms of action of BT in ovarian cancer cells.
Increased ROS generation can be frequently observed in cells subjected to anticancer drugs such as paclitaxel, cisplatin, doxorubicin [34, 35]. Accumulation of ROS inside the cell may result in apoptosis or terminal differentiation . Our results demonstrate significant generation of ROS in BT treated cells as compared to untreated cells in both a concentration and time dependent fashion. In order to ascertain role of ROS in BT induced cytotoxicity, we performed a cell viability assay in the presence of BT and antioxidant, ascorbic acid. Our results demonstrate a significant restoration of cell viability in the presence of 1 mM ascorbic acid in all cell lines tested. Interestingly, cisplatin-resistant variants of IGROV-1 and A2780 demonstrated greater responses to ascorbic acid pre-treatment than their cisplatin-sensitive counterparts. These observations imply a significant role of ROS in BT mediated cytotoxicity, and more so in cisplatin-resistant cell lines. This unique effect of BT on ROS generation in cisplatin-resistant cells implies that BT could have a role in the treatment of platinum-resistant ovarian cancer, either alone or in combination with other cytotoxic drugs.
Reactive oxygen species are known to modify signalling molecules important in cellular survival such as Akt1, and transcription factors including NF-kB, due to the presence of redox-sensitive cysteine or methionine groups that are susceptible to oxidation . It is widely reported that cisplatin-resistant cell lines maintain high levels of Akt and NF-kB as compared to cisplatin-sensitive cell lines . Keeping in mind the greater role of ROS generation observed in cisplatin resistant variants upon BT treatment, it may be possible that modification of pro-survival molecules such as Akt and NF-kB via oxidation may be a possible mechanism of action of BT, especially in cisplatin-resistant cell lines.
To further define key signalling responses of ovarian cancer cells to treatment with BT, we analyzed the expression and activation/phosphorylation of cellular markers involved in pro-apoptotic (p38) or pro-survival (pAkt, NF-kB) signalling. Immunoblotting of PAGE-separated cellular lysates revealed sustained activation of pP38 MAPK upon BT treatment. In order to assess the role of pP38 signalling in BT induced cytotoxicity, a cell viability assay was performed in the presence of a p38 inhibitor, SB203580. Pre-treatment with the p38 inhibitor did not restore cell viability when cells were treated with BT. These results rule out any significant role for p38 MAPK signalling in BT mediated cytotoxicity.
Activation of the PI-3 K/Akt pathway has been shown to induce resistance to apoptosis induced by a number of drugs and has been linked to cisplatin resistance in ovarian cancer cell lines [39, 40]. In view of this, we studied the expression of pAkt upon BT treatment. Significant down-regulation of pAkt expression was observed at 24 hrs post BT treatment. It has been reported that Akt inactivation is essential for drug sensitivity [41, 42]. In order to understand whether further inactivation of Akt can enhance the effectiveness of BT, we performed cell viability assays in the presence of PI3k inhibitor LY294002. LY294002 neither enhanced BT cytotoxicity nor restored the cell viability at 48 hrs post BT treatment. These results show that the Akt pathway may not mediate BT cytotoxicity in ovarian cancer cell lines.
Inhibition of the IKK/NF-κB activation pathway is considered an effective target for many anticancer drugs . NF-kB inhibition in cancer cells has been shown to enhance chemotherapeutic response [44, 45]. BT has also been reported to inhibit NF-kB signalling via inhibition of IkBα phosphorylation in vitro. Given the relevance of the NF-κB pathway in cancer, we assessed the effect of BT on phospho-NF-κB p65 and subsequent effect on NF-kB regulated proteins such as pIkBα, pbcl-2, bcl-xL, xIAP. Immunoblot analyses of whole cell lysate reveal decreased phospho-NF-κB p65 expression with increasing treatment time. BT treatment also down-regulated the expression of pIkBα. Suppression of proliferation, induction of apoptosis and G1/S cell cycle arrest can all be due to inhibition of phosphorylation of NF-kB and IkBα. BT can affect the DNA binding activity of NF-kB directly via oxidation by ROS  and/or indirectly by inhibiting phosphorylation of NF-κB  and IkBα. Phosphorylation of p65 at ser536 is essential for the DNA binding activity of NF-κB and it is known to be mediated via the PI3-kinase pathway . Because BT also decreased pAkt expression, BT appears to indirectly reduce the DNA binding activity of NF-κB and affect the expression of NF-κB regulated anti-apoptotic proteins such as pIkBα, pbcl-2, bcl-xL, xIAP. Indeed, we observed that NF-kB regulated proteins XIAP, bcl-xl, pbcl2 were down regulated upon BT treatment. XIAP is known to prevent apoptosis through up-regulation of PI3k/Akt cell survival signalling pathway . Down–regulation of XIAP induces apoptosis and increases cisplatin sensitivity . Inhibition of Bcl-xl may increase sensitivity to drugs such as carboplatin . Expression of Bcl-2 is important in protection from drug-induced apoptosis in ovarian cancer thereby contributing to chemo-resistance [51, 52]. These reports implicate NF-kB as a desirable target for anticancer agents in ovarian cancer. Our results demonstrate inhibitory effect of BT on NF-kB regulated proteins in ovarian cancer cell lines. BT treatment may promote apoptotic role for NF-κB by repressing anti-apoptotic gene expression. Our results indicate an important role for NF-kB in BT induced cytotoxicity. However, further studies are required to confirm role of NF-kB in the anti-tumor effects of BT in ovarian cancer cell lines.
Autotaxin (ATX) inhibition was considered major mechanism of action of BT. Previously BT was shown to inhibit solid tumor growth in several preclinical cancer models by targeting ATX [12, 13]. ATX plays a major role in modulation of the cellular process through its enzymatic production of lysophosphatidic acid (LPA). ATX is known to increases the aggressiveness and invasiveness of transformed cells, and directly correlates with tumor stage and grade in several human malignancies, including ovarian cancer [53, 54]. ATX was shown to delay carboplatin induced apoptosis in ovarian cancer cells . ATX inhibition was a proposed mechanism of action of BT in a melanoma model via inhibition of cell migration and invasion . Given the significance of ATX in ovarian cancer [55–58], we studied the effect of BT on ATX in a panel of ovarian cancer cell lines. Our results clearly demonstrate significant inhibition of ATX in a concentration and time dependent fashion. ATX/LPA stimulate the PI3-K, Akt, and ERK pathways and cause the activation of Rho and Rac . These pathways facilitate cell division, survival, and migration [60, 61]. BT may inhibit cell survival directly via inhibition of ATX or indirectly via inhibition of PI3-K, Akt or NF-kB pathways. Additionally, ATX is known to act as antioxidant, thus, protecting cells from oxidative stress . The fact that BT treatment reduced ATX activity would imply that treated cells are exposed to a higher oxidative stress, eventually leading to apoptosis or necrosis. In view of the significance of ATX in chemoresistance in a majority of widely used chemotherapeutic agents, ATX inhibition or the LPA pathway can be considered as a significant therapeutic target. In our studies, we also observed a significant inhibition of ATX by BT.
Based on our findings, BT affects cells by causing mitochondrial dysfunction, ROS generation, cell cycle arrest and ATX inhibition, ultimately leading to cell death (apoptosis at low concentrations and necrosis at higher concentrations). BT appears to be a viable therapeutic agent against ovarian cancer cell lines in vitro. Further exploration of its anti-tumor potential in ovarian cancer animal xenograft model is essential before proceeding to clinical trials. Additionally, it is interesting to focus on synergistic, additive or antagonistic effects of BT in combination with other standard chemo drugs. These studies are currently underway.