Cisplatin or LA-12 enhance killing effects of TRAIL in prostate cancer cells through Bid-dependent stimulation of mitochondrial apoptotic pathway but not caspase-10

Searching for new strategies for effective elimination of human prostate cancer cells, we investigated the cooperative cytotoxic action of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and two platinum-based complexes, cisplatin or LA-12, and related molecular mechanisms. We demonstrated a notable ability of cisplatin or LA-12 to enhance the sensitivity of several human prostate cancer cell lines to TRAIL-induced cell death via an engagement of mitochondrial apoptotic pathway. This was accompanied by augmented Bid cleavage, Bak activation, loss of mitochondrial membrane potential, activation of caspase-8, -10, -9, and -3, and XIAP cleavage. RNAi-mediated silencing of Bid or Bak in Bax-deficient DU 145 cells suppressed the drug combination-induced cytotoxicity, further underscoring the involvement of mitochondrial signaling. The caspase-10 was dispensable for enhancement of cisplatin/LA-12 and TRAIL combination-induced cell death and stimulation of Bid cleavage. Importantly, we newly demonstrated LA-12-mediated enhancement of TRAIL-induced cell death in cancer cells derived from human patient prostate tumor specimens. Our results provide convincing evidence that employing TRAIL combined with cisplatin/LA-12 could contribute to more effective killing of prostate cancer cells compared to the individual action of the drugs, and offer new mechanistic insights into their cooperative anticancer action.


Introduction
Prostate cancer is the second most frequently diagnosed cancer and one of the leading causes of cancer deaths in men worldwide [1].Currently available treatments mainly involve surgery, radiation therapy, hormonal therapy (androgen ablation) or chemotherapy [2].As prostate cancer cells often develop the ability to grow in the absence of androgens or become resistant to chemotherapy, there is still no efficient cure for this type of disease especially in the later metastatic stages.Considerable attention has therefore been paid to novel tumor-selective anticancer agents whose cytotoxic potential may not strictly depend on cellular status of androgen receptor or frequently mutated p53.
The tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) cytokine possesses a unique capacity to kill selectively cancer cells in vitro and in vivo without causing toxicity to normal cells or tissues [3][4][5].TRAIL can trigger apoptosis by interaction with two of his five known receptors-death receptor 4 and 5 (DR4/DR5) at the cell surface.Upon its binding, DR4 and DR5 are trimerized and death-inducing signaling complex (DISC) is formed.Intracellular part of DR called death domain (DD) recruits Fas-associated death domain (FADD) protein that consequently binds initiator pro-caspase-8/-10 via the death effector domain (DED) interaction.The caspase-8 activated at the DISC further mediates effector caspase-3 activation, followed by execution of apoptotic program.Apoptotic signaling can also be enhanced by initiator caspase-mediated BH3-only protein Bid cleavage, generating truncated Bid (tBid).The tBid activates pro-apoptotic Bcl-2-family members Bak or Bax, leading to mitochondrial outer membrane permeabilization [6,7].Other mitochondria-related proapoptotic events such as release of cytochrome c, Smac/DIABLO, apoptosome formation, caspase-9 activation and effector caspases cleavage further multiply apoptotic death signaling [8].Although application of recombinant TRAIL or agonistic DR4/5 monoclonal antibodies emerged as a promising anticancer strategy [9], apparent resistance of mainly primary tumors including prostate to their killing effects poses a serious obstacle in establishing clinically efficient TRAIL-based monotherapies [10,11].This could be overcome by combining DR4/5 ligands with some chemotherapeutic drugs.
Cisplatin is a largely employed platinum(II) compound that exerts clinical activity against numerous solid tumors, and was also shown to have potential in management of metastatic castration-resistant prostate cancer [12,13].However, its application may be limited due to the undesired side effects or the resistance in various cancer cell types [14,15].These limitations evoke a need to uncover yet unknown molecular mechanisms of cisplatin action or research and development of new platinum-based complexes with improved antitumor potential.The promising anticancer effects of platinum (IV) complex LA-12 [16] have been reported in vitro and in vivo by us and others in various cancer cell types including prostate [17][18][19][20][21][22][23][24][25].In addition to their individual potential to affect cancer cell death on their own, platinum-based drugs have been suggested to sensitize prostate cancer cells to TRAIL-induced apoptosis.To date, only a limited number of studies report on the cooperative cytotoxic effects of cisplatin and TRAIL/anti-DR5 antibodies in this type of cancer [25][26][27].Previously, we were the first to demonstrate the ability of LA-12 to enhance TRAIL-induced apoptosis [25], which is so far the only publication referring to the effects of the combined action of the two drugs in prostate cancer cells.The molecular mechanisms behind the above-mentioned effects are still incompletely understood.
In this study, we showed that exposure of human prostate cancer cells to sub-lethal doses of cisplatin or LA-12 resulted in profound potentiation of TRAIL-induced cell death that involved activation of mitochondrial apoptotic pathway, and investigated the novel molecular mechanisms behind the cooperative action of the drugs.We also newly demonstrated the National Program of Sustainability II (MEYS CR) (http://www.msmt.cz),by the European Unionproject ICRC-ERAHumanBridge (No. 316345) (http://www.icrc-era-humanbridge.eu),and by Brno City Municipality (Brno Ph.D. Talent, MK) (http://www.jcmm.cz/en/phd-students.html).These funders provided support in the form of salaries for authors [OVB, JH, MK, ZK, JB, GK, MK, AK, AHV], but had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.Platinum Pharmaceuticals provided support in the form of the LA-12 supply and information about physicochemical properties of the compound, in the form of salaries for one of the co-authors [PS], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.The specific roles of these authors are articulated in the 'author contributions' section.The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Petr Sova is employed by Platinum Pharmaceuticals.Petr Sova is an employee of the company owning intellectual property rights regarding LA-12 and co-author of patents in the field of LA-12 (Polymorphic form (I) of the adamantylamino-platinum (IV) complex LA-12, WO2009100690, Polymorphic form (II) of adamantylamino-platinum (IV) complex LA-12, WO2009100689, Amourphous form of the adamantylamino-platinum (IV) complex LA-12, WO2009100689, Polymorphic form (III) of the adamantylamino-platinum (IV) complex LA-12, WO2009100691, Oral pharmaceutical composition for targeted transport of a platinum complex into the colorectal region, method for producing and use as medicament thereof, CZ20041167, Pharmaceutical composition containing platinum complex as active substance and method of manufacturing thereof, WO2005077357).This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Densitometry analysis
Following the Western blot/immunodetection, the results were subjected to densitometry analysis using ImageJ software (NIH), and represent arbitrary units relative to control (corrected by β-actin).The statistical significance (P < 0.05) was determined by one-way ANOVA followed by Fisher's Least Significant Difference (LSD) test using STATISTICA 12 (Stat Soft, Inc., Tulsa, OK, USA) software.

Annexin V/propidium iodide assay
The cells were harvested, washed with PBS, incubated with annexin V-fluorescein (1:200, P030-F100, Apronex, Prague, Czech Republic) in specific binding buffer for 30 min in 37˚C, and then stained (5 min) with propidium iodide (PI, 0.5 μg/ml, Sigma-Aldrich).Samples were measured by flow cytometry (FACSVerse TM ), and analyzed using FACSSuite TM software version 1.0.5 (both Becton Dickinson, San Jose, CA).Minimum of 10 000 cells excluding doublets and debris were subjected to analysis and total of annexin V + /PI -and annexin V + /PI + cells were assessed as percentage of dead cells.

Measurement of mitochondrial membrane potential (MMP)
The cells were incubated (20 min) in Hanks' balanced salt solution (HBSS) with 10 nM of tetramethylrhodamine ethyl ester perchlorate (TMRE, Molecular Probes, Eugene, OR, USA), then washed twice with HBSS, and analyzed by flow cytometer (FACSVerse TM ) and FACSSuite TM v1.0.5 (both Becton Dickinson).Minimum of 10 000 cells excluding doublets and debris were subjected to analysis, and the percentage of cells with decreased MMP was determined.

Detection of caspase-3 activity
The cells were harvested, washed in PBS, fixed and stained using FITC active caspase-3 apoptosis kit (#550480, BD Pharmingen) according to the manufacturer's protocol.The percentage of cells with active caspase-3 was detected by flow cytometry (FACSVerse TM , FACSuite TM v1.0.5).

Analysis of Bak activation
The cells were washed with PBS, fixed (10 min, 4˚C) in 1% paraformaldehyde (pH 7.4), and stained with monoclonal mouse anti-Bak antibody (1:250, #AM03, Calbiochem-MERCK) raised against recombinant human Bak from which the C-terminal and transmembrane domains have been deleted.The protocol was previously described in [30].The analysis was performed using flow cytometry (FACSCalibur TM ) and CellQuest software (both Becton Dickinson), and the results were expressed as percentage of cells with active Bak.

Statistical analysis
The results of at least three independent experiments were expressed as the means + S.E.M. Statistical significance (P < 0.05) was determined by t-test or one-way ANOVA followed by Fisher's Least Significant Difference (LSD) test using STATISTICA 12 (Stat Soft, Inc., Tulsa, OK, USA) software.The results from Western blotting are representatives of at least three independent experiments.

Results
Cisplatin or LA-12 enhance the sensitivity of human prostate cancer cells to TRAIL-induced cell death The decrease in XIAP protein level was also apparent in PC-3 and LNCaP cells treated with the drug combination versus single drugs (S3 Fig) .For further more mechanistically-oriented studies focused on the role of mitochondria, we took advantage of the DU 145 cell line that does not express the Bax protein [33], and may serve as a useful tool for further efficient modulation of the mitochondrial apoptotic pathway.

Caspase-10 is dispensable for enhancement of cisplatin/LA-12 and TRAIL combination-induced apoptosis and Bid cleavage
We showed that potentiation of cell death in cisplatin/LA-12 and TRAIL-treated prostate cancer cell lines was accompanied by enhanced processing of caspase-10 (Fig 6A

LA-12 sensitizes primary human prostate cancer cells to TRAIL-induced apoptosis
The combination of LA-12 with TRAIL enhanced the cytotoxic response also in primary human prostate cancer cells, which was documented by an enhanced cleavage of PARP and caspase-8 (Fig 7A

Discussion
In order to propose novel strategies for prostate cancer cell sensitization to TRAIL-induced killing, we examined the cooperative cytotoxic effects of platinum drugs and this cytokine, and related molecular mechanisms.Our previous work showed the ability of cisplatin or LA-12 to effectively enhance TRAIL-induced apoptosis in human PC3 prostate cancer cells, which was associated with modulations of some upstream events in the extrinsic apoptotic pathway such as re-localization of DR4/5 into lipid rafts or TRAIL internalization [25].The present work was aimed to investigate the involvement of mitochondria in platinum-based drug-mediated stimulation of TRAIL-induced apoptosis in prostate cancer cells.Here we demonstrate that  We showed that the killing effect of cisplatin/LA-12 and TRAIL in DU 145 cells was suppressed in the presence of z-LEHD-fmk, which indicated the functional role of mitochondria and caspase-9 in apoptotic machinery triggered by the drug combination.However, being aware of the fact that under some circumstances, z-LEHD-fmk might also inhibit other targets in addition to caspase-9 [34], we enlarged our observations using other cellular models with specifically blocked mitochondrial functions.We took advantage of the DU 145 cell line that carries a monoallelic Bax frameshift mutation and a second silent Bax allele, thus being completely devoid of Bax protein expression [33].The Bax and Bak are known as the two main effectors of the Bcl-2 family at the level of mitochondria, crucial for further propagation of apoptotic signal downstream of these organelles [35].Therefore, in Bax-deficient DU 145 cells, the silencing of Bak may serve as a useful approach to block the mitochondrial apoptotic pathway prior to the drug treatments, and further study the consequences.We showed that siRNAmediated silencing of Bak in DU 145 cells significantly suppressed caspase-9 activation and apoptosis induced by the drug combination.Similar apoptosis-suppressing effects were also obtained in the drug combination-treated DU 145 cells with silenced Bid, a crucial BH3-only protein connecting proximal DR signals to the mitochondrial apoptotic machinery via its interaction with Bcl-2 family members such as Bax or Bak [6,7].Thus, using three different approaches we unequivocally demonstrated the indispensable role of mitochondria in the efficient realization of cytotoxic potential of the combination of cisplatin/LA-12 and TRAIL in prostate cancer cells.
We further showed that pretreatment with z-LEHD-fmk or siRNA-mediated silencing of Bak resulted in significant suppression of the enhanced caspase-8 cleavage observed in cisplatin/LA-12 and TRAIL combination-treated DU 145 cells.It means that activation of caspase-8 in the drug combination-treated cells was mediated not only via its DR-DISCdependent processing, but also due to Bak-dependent mitochondrial pathway and resulting caspase-9 activation.Previous studies also demonstrated that caspase-8 activation induced by various apoptotic stimuli including TRAIL and cisplatin required positive mitochondrial feedback amplification loop [27,30,[36][37][38].In this system, caspase-9, -3 and -6 were implicated in the caspase-8 activation downstream of mitochondria [39][40][41].Based on the data obtained by us and the above-mentioned studies, we suggest that stimulation of mitochondrial pathwayderived caspase-9 and resulting effector caspase activity is required for the efficient caspase-8 activation in the cisplatin/LA-12 and TRAIL combination-treated prostate cancer cells.We also showed that in contrast to caspase-8 processing, the cleavage of Bid was not attenuated in the drug combination-treated DU 145 cells with silenced Bak.This indicates that the secondarily activated caspase-8 did not significantly contribute to Bid cleavage in the drug combination-treated DU 145 cells.Instead, we suggest the existence of complex feedback mechanism where secondarily activated caspase-8 can further cleave the effector caspases, additionally amplifying the caspase cascade in the prostate cancer cells treated with cisplatin/LA-12 and TRAIL combination.
We assumed that the specific modulations of the ratio between proapoptotic and antiapoptotic Bcl-2 family proteins in cancer cells treated with cisplatin/LA-12 and TRAIL may help to render mitochondria more susceptible to apoptotic signal transmitted from DRs via tBid.In addition to Bid, which we identified as a crucial mediator of cisplatin/LA-12-and TRAIL- induced apoptosis in prostate cancer cells, we examined the potential changes in the level of other BH3 only Bcl-2 family members.In accordance with our previous work in colon cancer models [30], we also observed that LA-12 acts as an inducer of Noxa in some prostate cancer cells, while the impact of cisplatin on this protein level was less evident.Cisplatin-mediated upregulation of Noxa was functionally important for the killing activity of this drug in other cancer cells [42], while it was not essential for cisplatin-mediated colon cancer cell sensitization to TRAIL-induced apoptosis [30].Previously, we also demonstrated the ability of LA-12 to upregulate the BH3 only Bim L level in colon cancer cells [30], which was not obvious in the present study with prostate models, further excluding its major role in the drug-mediated cytotoxicity.In addition, we could not detect any significant changes in the amount of antiapoptotic Bcl-2 family proteins Bcl-2 and Bcl-xL in the cisplatin/LA-12 and TRAIL combination-treated prostate cancer cells.Similar results were also reported in human TE12 esophageal or HCT116 colon cancer cells treated with the combination of TRAIL with cisplatin or LA-12 [30,38].In contrast, reduction of the level of these proteins was suggested to contribute to cisplatin-mediated enhancement of TRAIL-induced apoptosis in human CRL2335 and MDA-MB-468 triple negative breast or SKOV3 ovarian cancer cells [43,44].We concluded that the role of Bcl-2 or Bcl-xL in modulation of prostate cancer cell sensitivity to combined cisplatin/LA-12 and TRAIL treatment may be cell type-specific, but not crucial in prostate cancer cell lines used within our study.
In type II cells that rely on mitochondria for further amplification of DR-mediated apoptotic signal, regulation of XIAP represents an important control point in apoptosis execution [45].XIAP is the most potent and most studied inhibitor of caspases among human cIAP family, and promising therapeutic target in cancer treatment [46].Abrogation of its expression by antisense phosphorodiamidate morpholino oligomer was essential for DU 145 prostate cancer cell sensitization to the cytotoxic effects of the individually applied cisplatin or TRAIL [47].The ability of cisplatin to induce apoptosis through the inhibition of XIAP expression has also been reported in LNCaP cells [48].Interestingly, caspase-3-dependent cleavage of XIAP, followed by its proteasomal degradation, has been shown to accompany cisplatin-mediated sensitization of melanoma cells to TRAIL-induced apoptosis [49].In the present study, the enhancement of apoptosis induced by cisplatin/LA-12 and TRAIL was associated with potentiation of XIAP cleavage that coincided with caspase activity.We suggest that the loss of XIAP we observed in prostate cancer cells treated with the drug combination might contribute to their increased apoptotic response compared to the individual drug applications.
An apparent enhancement of caspase-10 cleavage in cisplatin/LA-12 and TRAIL-treated prostate cancer cells prompted us to investigate in more detail its possible involvement in the cooperative cytotoxic action of the drugs.While the function of caspase-8 in TRAIL-induced signaling and Bid cleavage is well established, the role of caspase-10 is more controversial and less understood.Whereas some studies indicated that caspase-10 is crucial for TRAIL signaling [50][51][52], other reports suggest that it is not important [53,54].The ability of caspase-10 to trigger Bid cleavage and caspase cascade activation in FasL-induced apoptosis has also been reported [55].Interestingly, caspase-10 has recently been shown to negatively regulate caspase-8-mediated cell death [56].In present study, we newly showed that caspase-10 was dispensable for realization of the cytotoxic potential of the cisplatin/LA-12 and TRAIL combination in prostate cancer cells, and also did not significantly affect the drug-induced Bid cleavage.Furthermore, calpain-mediated Bid cleavage has been shown as functionally important in cisplatin-induced apoptosis of human melanoma cells, which was blocked by calpain inhibitors [57].We demonstrated that cisplatin/LA-12 and TRAIL combination-induced enhancement of cell death was not affected by calpeptin, further excluding the possible importance of calpain-mediated Bid cleavage in drug combination treated prostate cancer cells.
A very frequent resistance to TRAIL-induced apoptosis was previously reported in a large cohort of patient prostate cancer tissue samples [10], and our observations also support these findings.Although TRAIL monotherapy seems to be inefficient for treatment of this type of cancer, our results highlight a possible applicability of platinum-based complexes in sensitization of prostate cancer cells to TRAIL-induced apoptosis.Importantly, we newly showed that cisplatin/LA-12 and TRAIL cooperate to induce cell death in primary prostate cancer cell cultures isolated from patient tumor biopsies.This drug combination-induced stimulation of the cytotoxic response was associated with activation of caspases including caspase-9, indicating the involvement of mitochondrial apoptotic pathway.As cisplatin is commonly used drug in cancer treatment, and both TRAIL and LA-12 have individually entered clinical trials, we suggest that the drug combinations examined in our study could be considered as an interesting approach for future treatment strategies.In summary, we showed that pretreatment of human prostate cancer cells with cisplatin or LA-12 resulted in significant potentiation of TRAIL-induced cell death via engagement of mitochondrial apoptotic pathway, associated with pronounced Bid cleavage, Bak activation, loss of MMP and caspase activation.RNAi-mediated silencing of Bid or Bak in Bax-deficient DU 145 cells suppressed drug combination-induced cytotoxicity, further underscoring the importance of mitochondrial signaling.The cisplatin/LA-12 and TRAIL combination-induced stimulation of Bid cleavage did not seem to essentially require caspase-10 or Bak.We also newly showed that caspase-10 was dispensable for platinum drug-mediated stimulation of TRAIL-induced apoptosis in prostate cancer cells (Fig 8).Compared to cisplatin, lower doses of LA-12 were required for similar sensitizing effects on TRAIL-induced prostate cancer cell apoptosis.Importantly, an enhancement of TRAIL-induced apoptosis mediated by LA-12 was for the first time demonstrated in cancer cells derived from prostate tumor specimens obtained from human patients.Our results indicate that anticancer strategies employing TRAIL combined with platinum-based drugs could contribute to more effective elimination of prostate cancer cells compared to the individual action of the drugs, which warrants further investigation.
Pretreatment of three different human prostate cancer cell lines (DU 145, PC-3, and LNCaP) with subtoxic doses of cisplatin or LA-12 enhanced TRAIL-induced cytotoxicity, as demonstrated by a significant increase in percentage of dead cells (annexin V/propidium iodide assay) and enhanced cleavage of caspase-3 and PARP (Fig 1A-1C) compared to the individual treatments.In all cell lines, the drug combination induced an apparent enhancement of caspase-9 cleavage, indicating the involvement of mitochondrial apoptotic pathway (Fig1B).Cisplatin/LA-12 and TRAIL cooperate to promote stimulation of mitochondrial apoptotic pathway associated with activation of Bid and Bak proteinsTo investigate the role of mitochondria in the cooperative cytotoxic action of cisplatin/LA-12 and TRAIL, we examined several factors related to mitochondrial apoptotic signaling.We observed a significant enhancement of Bid cleavage in all prostate cancer cells lines treated with the combination of cisplatin/LA-12 and TRAIL compared to the agents used alone (Fig2Aand S1 Fig).After treatment with LA-12, the levels of Noxa were upregulated especially in DU 145 and PC-3 cells (Fig 2A and S1 Fig).The treatment of prostate cancer cells with cisplatin/LA-12 and TRAIL did not seem to have any strong impact on the level of Bim (Fig 2A and S1 Fig).Cisplatin/LA-12 and TRAIL combination-induced enhancement of apoptosis was associated with a significant increase in percentage of the cells with decreased MMP (Fig 2B and S2 Fig) and active Bak protein (Fig 2C).The total Bak protein level remained unchanged following the drug treatments in the cell lines studied (Fig 2D and S3 Fig).The treatments with platinum drugs and TRAIL also did not affect the Bax protein level in PC3 and LNCaP cells (S3 Fig).No apparent modulation of the anti-apoptotic Bcl-2 or Bcl-x L protein levels was observed in all prostate cancer cell lines treated with platinum drugs and/or TRAIL (Fig 2D and S3 Fig).We observed a remarkable increase in XIAP protein cleavage in drug combination-treated DU 145 cancer cells compared to the incubation with individual agents (Fig 2D).
and 7B), and caspase-3, -9 (Fig 7A) or percentage of dead cells in annexin V/ PI assay (Fig 7B) in drug combination versus individually treated samples of three patients marked as PY001, PY002 and PY005.These cells were apparently resistant to the cytotoxic/ cytostatic effects of TRAIL alone used up to dose of 1 μg/ml, as documented by CyQuant Cell Proliferation Assay Kit (S7 Fig).

Fig 8 .
Fig 8. Cisplatin/LA-12-mediated enhancement of TRAIL-induced apoptosis in prostate cancer cells-main molecular mechanisms involved.Cisplatin or LA-12 enhance the sensitivity of human prostate cancer cells to TRAIL-induced cell death via an engagement of mitochondrial apoptotic pathway.This was accompanied by augmented Bid cleavage, Bak activation, loss of mitochondrial membrane potential, activation of caspase-8, -10, -9, and -3, and XIAP cleavage.The caspase-10 was dispensable for enhancement of cisplatin/LA-12 and TRAIL combination-induced cell death and stimulation of Bid cleavage.https://doi.org/10.1371/journal.pone.0188584.g008