- Email: [email protected]
Mitochondrial targeted curcumin exhibits anticancer effects through disruption of mitochondrial redox and modulation of TrxR2 activity
Free Radical Biology and Medicine 113 (2017) 530–538
Contents lists available at ScienceDirect
Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed
Original article
Mitochondrial targeted curcumin exhibits anticancer effects through disruption of mitochondrial redox and modulation of TrxR2 activity
MARK
Sundarraj Jayakumara,1, Raghavendra S. Patwardhana,1, Debojyoti Pala, Babita Singha, ⁎ ⁎ Deepak Sharmaa,b, , Vijay Kumar Kutalac, Santosh Kumar Sandura,b, a b c
Radiation Biology & Health Sciences Division, Modular Laboratories, Bhabha Atomic Research Centre, Trombay, Mumbai, India Homi Bhabha National Institute, Anushakti Nagar, Mumbai, India Department of Clinical Pharmacology & Therapeutics, Nizam's Institute of Medical Sciences, Hyderabad, India
A R T I C L E I N F O
A B S T R A C T
Keywords: Mitocurcumin Lung cancer Cancer stem cells Thioredoxin reductase 2 Mitochondrial DNA damage Curcumin
Mitocurcumin is a derivative of curcumin, which has been shown to selectively enter mitochondria. Here we describe the anti-tumor efficacy of mitocurcumin in lung cancer cells and its mechanism of action. Mitocurcumin, showed 25–50 fold higher efficacy in killing lung cancer cells as compared to curcumin as demonstrated by clonogenic assay, flow cytometry and high throughput screening assay. Treatment of lung cancer cells with mitocurcumin significantly decreased the frequency of cancer stem cells. Mitocurcumin increased the mitochondrial reactive oxygen species (ROS), decreased the mitochondrial glutathione levels and induced strand breaks in the mitochondrial DNA. As a result, we observed increased BAX to BCL-2 ratio, cytochrome C release into the cytosol, loss of mitochondrial membrane potential and increased caspase-3 activity suggesting that mitocurcumin activates the intrinsic apoptotic pathway. Docking studies using mitocurcumin revealed that it binds to the active site of the mitochondrial thioredoxin reductase (TrxR2) with high affinity. In corroboration with the above finding, mitocurcumin decreased TrxR activity in cell free as well as the cellular system. The anticancer activity of mitocurcumin measured in terms of apoptotic cell death and the decrease in cancer stem cell frequency was accentuated by TrxR2 overexpression. This was due to modulation of TrxR2 activity to NADPH oxidase like activity by mitocurcumin, resulting in higher ROS accumulation and cell death. Thus, our findings reveal mitocurcumin as a potent anticancer agent with better efficacy than curcumin. This study also demonstrates the role of TrxR2 and mitochondrial DNA damage in mitocurcumin mediated killing of cancer cells.
1. Introduction Lung cancer is the most prevalent cancer in the world and causes excessive morbidity [1]. Since more than seventy-five percent of the lung cancers are presented at advanced stages, the currently available treatment modalities are not very effective. Thus, there is a need to develop newer chemotherapeutic modalities for treatment of lung cancer [2]. Mitochondrion plays a vital role in the cell survival and growth as it is involved in the energy metabolism. Mitochondrion is also a major source of ROS generation and immoderation of ROS production can trigger apoptotic cell death. Hence targeting mitochondrion can be a good strategy for killing cancer cells [3]. One of the major antioxidant defense mechanisms by which mitochondria neutralize the excess ROS is through a dedicated thioredoxin system comprising of thioredoxin 2 (Trx2), thioredoxin reductase 2 (TrxR2) and NADPH [4,5]. TrxR2 is a selenocysteine containing homodimeric enzyme, ⁎
1
which is a member of the nucleotide-disulfide oxidoreductase protein family. Its primary function is to keep Trx2 in the reduced form which in turn reduces peroxiredoxins. Peroxiredoxins detoxify hydrogen peroxide and lipid hydroperoxides to water or alcohol [6]. Since cancer cells thrive under oxidative microenvironment, they are over dependent on thioredoxin system to counter this oxidative stress. There are several reports showing overexpression of thioredoxin system in different cancers [7]. TrxR2 has also been shown to be overexpressed in several cancers including lung cancer [8,9] and it is also involved in tumor progression by promoting vascularization through stabilization of HIF1α [10]. Overexpression of TrxR2 is associated with poor prognosis and therapeutic resistance of cancer and hence targeting TrxR2 can be an attractive strategy in treating cancers [8,11,12]. A subset of cancer cells which are resistant to chemotherapeutic drugs and contribute to relapse as well as metastases are identified as cancer stem cells in solid cancers [13]. Treatment outcome of cancer depends on effective elimination of
Corresponding authors at: Radiation Biology & Health Sciences Division, Modular Laboratories, Bhabha Atomic Research Centre, Trombay, Mumbai, India E-mail addresses: [email protected] (D. Sharma), [email protected] (S. Kumar Sandur). Both the authors have contributed equally.
http://dx.doi.org/10.1016/j.freeradbiomed.2017.10.378 Received 20 July 2017; Received in revised form 6 October 2017; Accepted 23 October 2017 0891-5849/ © 2017 Elsevier Inc. All rights reserved.
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
of nuclear DNA content. These hypodiploid cells can be enumerated by staining with DNA binding dye propidium iodide (PI) [22]. Exponentially growing cells were treated with mitocurcumin, cultured for 48 h and stained with PI solution (1X PBS containing sodium citrate, RNase, triton × 100 and 50 μg/ml PI) overnight at 4 °C [21]. The cells were acquired on high throughput cell analyzer, (Acumen Cellista, TTP Labtech, UK) and cells showing sub G1 DNA content (hypodiploid) were enumerated using Cellista software.
cancer stem cells. Hence, agents that can target these cells are also good candidates for cancer treatment. Curcumin is a major constituent present in an Indian herb Curcuma longa, and has been shown to possess many pharmacological properties [14]. Several human clinical trials are underway using curcumin as an adjuvant for various disease conditions [15]. Inhibition of cytosolic thioredoxin reductase (TrxR1) is one of the major mechanisms through which curcumin exert its anticancer effects [16,17]. However, curcumin has very poor stability and low bioavailability [18,19]. To circumvent this problem, a modified form of curcumin i.e., mitocurcumin which targets mitochondria has been synthesized [20]. Mitocurcumin has been shown to accumulate in mitochondria, and it could be a potent inhibitor of mitochondrial thioredoxin reductase (TrxR2). Using a combination of cell biology assays, high content screening, flow cytometry and in silico docking, we have shown the anti-tumor activity of mitocurcumin in lung cancer cells and its underlying molecular mechanism of action.
2.5. Live and dead assay Intracellular esterase activity and intact plasma membrane are characteristics of viable cells. Live and dead assay differentiates between viable and dead cells. Active esterases present in live cells convert calcein-AM into calcein which exhibits green fluorescence, whereas, dead cells accumulate ethidium homodimer-1 owing to their compromised membrane integrity and exhibit red fluorescence. Cells were grown on coverslip, treated with mitocurcumin, cultured for 24 h and live and dead assay was performed using a commercial kit following the manufacturer's protocol (ThermoFisher Scientific, MA, USA).
2. Materials and methods 2.1. Materials
2.6. MTT assay
Mitocurcumin was synthesized and characterized in the lab as mentioned previously [20]. Curcumin, verapamil, crystal violet, methanol, formaldehyde, paraformaldehyde, sodium citrate, RNase, JC-1, propidium iodide (PI), insulin, NADPH, NADPH oxidase, 5,5′-dithiobis 2-nitrobenzoic acid (DTNB), purified rat TrxR, purified human Trx, caspase-3 activity kit, annexin-V-FITC kit, fluorochrome conjugated secondary antibody, MTT, dimethyl sulfoxide (DMSO), triton X-100, DMEM, penicillin, streptomycin and trypsin-EDTA were purchased from Sigma Chemical Co. (MO, USA). Live and dead assay kit, DAPI with antifade, MitoSOX Red, 5-chloromethylfluorescein diacetate (CMFDA) dye, MitoTracker Red and Hoechst 33342 were purchased from ThermoFisher Scientific (MA, USA). Fluorochrome conjugated monoclonal antibodies against CD44 and CD24 were procured from BD Pharmingen (CA, USA). Monoclonal antibodies against BAX, BCL2 and cytochrome C were purchased from Cell Signaling Technologies (MA, USA). TrxR2 overexpression plasmids and knockout plasmids were purchased from Santacruz (MD, USA). Fetal bovine serum (FBS) was obtained from HiMedia (Mumbai, India).
Five thousand cells were seeded in 96-well plates overnight, treated with mitocurcumin, cultured for 48 h and were treated with MTT reagent for 4 h. Formazan crystals were dissolved in DMSO and absorbance was taken at 570 nm. Percent cell viability was calculated based on the absorbance value. 2.7. Enumeration of cancer stem cells Cancer stem cells can be enumerated by side population assay which is based on their ability to efflux Hoechst 33342 dye efficiently due to their active ABCG2 transporter as compared to other cancer cells. Cells were treated with vehicle or mitocurcumin and cultured in a 96-well plate for 24 h. Cells were washed and stained with Hoechst 33342 (10 μg/ml) in phenol red free DMEM for 90 min at 37 °C [23]. Hoechst 33342 red (620 nm) and blue (450 nm) fluorescence was recorded using high throughput cell analyzer, (Acumen, TTP Labtech UK). The live cells showing low Hoechst fluorescence in the red channel (620 nm) and blue channel (450 nm) were gated and plotted as side population cells. Verapamil which is a well-known inhibitor of ABCG2 transporter protein was used as a control in this experiment [24]. Cancer stem cells can also be distinguished based on characteristic surface markers like the presence of CD44 and the absence of CD24. The frequency of cancer stem cells was also enumerated by surface staining with CD44-PE and CD24-PE-Cy7 conjugated monoclonal antibodies followed by acquiring on a flow cytometer [25].
2.2. Cell culture A549 lung cancer cells, NIH 3T3 mouse embryonic fibroblast cells and PC-3 prostate cancer cells were procured form National Centre for Cell Sciences, Pune, India. Cells were grown as monolayer cultures in DMEM medium with 10% FBS along with penicillin and streptomycin in 5% CO2 and 95% air at 37 °C. 2.3. Clonogenic assay
2.8. Estimation of apoptosis by annexin-V-FITC staining Exponentially growing cells were harvested, seeded at 1000 cells per well in six well plates and allowed to adhere overnight. Cells were treated with various concentrations of mitocurcumin or curcumin and cultured for 14 days for colony formation [21]. Colonies were fixed using methanol, stained with crystal violet and counted using GelCount (Oxford Optronix, UK) instrument. Survival fraction was calculated using following formula: Survival fraction = number of colonies observed x 100 / number of cells plated x PE, where PE (plating efficiency) = number of colonies developed in control/ number of cells plated in control.
Cells undergoing apoptosis display phosphatidylserine on the outer leaflet of plasma membrane which can be detected using fluorochrome conjugated annexin-V. Cells were treated with mitocurcumin, cultured for 24 h and stained with annexin-V-FITC and PI as per manufacturer's protocol [9]. Briefly, cells were washed with 1X PBS, resuspended in 1X binding buffer containing annexin-V-FITC and PI for 10 min at room temperature. Cells were again washed and resuspended in 1X PBS and acquired using flow cytometer. Annexin-V positive cells were gated using FlowJo software.
2.4. Quantification of cell death by PI assay using high throughput cell analyzer
2.9. Intracellular antibody staining of BAX and BCL2 proteins Intracellular antibody staining was performed using BAX and BCL2 antibodies followed by staining with labeled secondary antibody as
Apoptotic cell death is characterized by DNA fragmentation and loss 531
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
2.14. TrxR activity in cell free system
described earlier [26]. Briefly, cells were treated with either DMSO (vehicle) or mitocurcumin (2.5 µM) for 24 h washed with 1X PBS, fixed with formaldehyde, permeabilized with 1X PBS containing 0.1% tritonX-100 and blocked in the presence of 5% FBS. Cells were resuspended in 20 μl antibody staining solution containing 0.3 μg each of BAX and BCL2 antibodies. Cells were incubated for 1 h at room temperature and cells were stained with fluorochrome conjugated secondary antibody. Cells were washed, resuspended in PBS and stained with Hoechst 33342 (10 μg/ml) and acquired on a flow cytometer.
Purified rat TrxR (110 nmol/L) was incubated with indicated concentrations of mitocurcumin (1, 2.5, 5 & 10 µM) or vehicle at room temperature for 1 h. TrxR activity was determined by DTNB reduction assay as mentioned earlier [23]. 2.15. Estimation of TrxR and NADPH oxidase activities Cells were treated with either DMSO (vehicle) or mitocurcumin (2.5 µM), for 24 h. TrxR activity was determined by the micro-method of insulin reduction as described elsewhere [27,28]. All assay tubes contained 0.26 M HEPES, pH 7.6, 10 mM EDTA, 2 mM NADPH, 1 mM insulin and 100 nM thioredoxin and cell extract in a final volume of 100 μl. After incubating at 37 °C for 20 min reaction was stopped by adding 500 μl stopping solution containing 0.2 M Tris-HCl, 6 M guanidine-HCl, 1 mM DTNB and absorbance was measured at 412 nm. To determine the NADPH oxidase activity, NADPH-reduced TrxR was incubated with different concentrations of mitocurcumin at room temperature for 2 h. Equal volume of DMSO was added to the control. In reaction mixture, modified TrxR was added to TE buffer containing 200 μM NADPH. NADPH oxidase enzyme was kept as positive control. Oxidation of NADPH was monitored by following the changes in absorbance at 340 nm.
2.10. Caspase-3 activity measurement Caspase-3 is a protease which is an important mediator in executing apoptosis. Caspase-3 activity is proportional to the release of fluorescent molecule (7-amino 4-methylcoumarin) which is attached to peptide substrate of caspase-3. Cells treated with mitocurcumin were harvested and whole cell extracts were prepared for determination of caspase-3 activity using a kit as per manufacturer's protocol. 2.11. Estimation of cytochrome C A549 cells were cultured for 24 h in presence of mitocurcumin, harvested, fixed with paraformaldehyde, permeabilized (using selective permeabilization buffer from InnoCyte™ which permeabilizes plasma membrane but not the mitochondrial membrane), blocked and stained with monoclonal antibody against cytochrome C followed by fluorochrome conjugated secondary antibody. Cells were acquired on a flow cytometer.
2.16. Estimation of reactive oxygen species in mitochondria For measuring mitochondrial ROS, exponentially growing cells were treated with mitocurcumin and cultured for 4 h. MitoSOX Red dye was added to the cells 30 min prior to harvesting and fluorescence was measured at 580 nm using a spectrofluorimeter (Synergy H1 Hybrid, Biotek, VT, USA) and cells were also visualized under confocal laser scanning microscope (Fluoview FV10i, Olympus, Japan).
2.12. Measurement of mitochondrial membrane potential (MMP) JC-1 exists as an oligomer that emits red fluorescence at normal MMP and gets converted to monomer upon loss of MMP and emits green fluorescence. Ratio of red to green fluorescence is an indicator of MMP. A549 cells were treated with mitocurcumin, cultured for 24 h and change in MMP was calculated by staining with JC-1 dye as described previously using a spectrofluorimeter [23]. The cells were also visualized under confocal laser scanning microscope to monitor change in MMP.
2.17. Quantification of mitochondrial GSH levels A549 cells were grown on a cover glass, treated with mitocurcumin (2.5 µM), cultured for 24 h and stained with CMFDA and MitoTracker Red for 30 min at 37 °C prior to imaging on a confocal microscope. CMFDA is a thiol reactive cell permeable fluorescent dye (green) that can detect glutathione specifically (> 95%) [29]. Mitotracker Red is used to track mitochondria in cells (red fluorescence). Co-localization of green and red fluorescence was used to quantify mitochondrial GSH levels.
2.13. In Silico docking studies In silico docking was performed to study possible interaction of mitocurcumin and curcumin with the active site of thioredoxin reductase (TrxR). Crystal structure of mammalian thioredoxin reductase (Rattus norvegicus) was obtained from protein data bank (PDB ID: 1H6V). Theoretical structure of human mitochondrial thioredoxin reductase in complex with thioredoxin was obtained from protein data bank (PDB ID: 1W1E). Crystal structure of human thioredoxin reductase 1 in complex with thioredoxin was obtained from protein data bank (PDB ID: 3QFA). 2D structure of curcumin and mitocurcumin were drawn using ChemSketch free version. DG-AMMOS was employed to generate the corresponding 3D structures and select the best energy minimized conformer. The ligand and receptor molecules were preprocessed using Autodock tools. Docking was carried out at the active site of TrxR using Autodock Vina. Rat TrxR has almost two identical active sites: one centered at C59-C64 of E chain and the other one at C59-C64 of F chain. They are referred to as E-chain active site and Fchain active site respectively. Human mitochondrial TrxR also has similar active sites: centered at C61-C66 of A chain and C61-C66 of B chain respectively. Human cytosolic TrxR has two active sites: one centered at C59-C64 of A chain and the other one at C59-C64 of B chain. They are referred to as A-chain active site and B-chain active site respectively. Interactions between TrxR and the docked ligands were studied using Ligplot plus software.
2.18. DNA damage quantification using real time qPCR Quantification of DNA damage using real time qPCR is based on the simultaneous amplification of a long region (3000–4000 bp) and a small region (~ 100 bp) of mitochondrial DNA and nuclear DNA. If the frequency of lesions is high in the DNA template, the PCR amplification of that region will be hindered, and they will have high threshold cycle value as compared to undamaged template in real time qPCR. Hence, DNA damage was quantified in mitochondria and nucleus based on the threshold cycle values. Total DNA was isolated from the cultured A549 cells treated with vehicle or mitocurcumin. The DNA quantity was measured using nano-spectrophotometer. DNA damage was quantified by qPCR as described by Stephen et al. [30]. 2.19. Overexpression and knockout studies CRISPR-Cas9 based system was used for knocking out or overexpression of TrxR2 in A549 cells (Santacruz, MD, USA). Transfection was performed using Neon® Transfection System (ThermoFisher Scientific, MA, USA) following manufacturer's protocol. The expression of TrxR2 level was confirmed by performing intracellular antibody 532
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
staining with TrxR2 antibody followed by fluorochrome conjugated secondary antibody.
3.3. Mitocurcumin inhibited mitochondrial thioredoxin reductase activity To understand the molecular mechanism underlying mitocurcumin induced apoptosis in lung cancer cells, we have studied the effect of mitocurcumin on thioredoxin reductase. The free energy values obtained by in silico docking studies revealed a high affinity binding of mitocurcumin at the active site of TrxR2 (Fig. 3A). Strong interactions were observed at E-chain, F-chain, A-chain and B-chain active sites (Table 1). Further, mitocurcumin inhibited the TrxR activity in a dose dependent manner in cell free system, and its IC50 value was around 10 µM (Fig. 3B). Corroborating with these findings, mitocurcumin at 2.5 µM inhibited cellular TrxR activity by more than 50% (Fig. 3C). Mitocurcumin induced a significant and dose dependent increase in the ROS levels in mitochondria as evinced from increased MitoSOX red fluorescence (Fig. 3D). Apart from increase in ROS, mitochondrial GSH levels were also decreased by mitocurcumin as seen from co-localization of CMF fluorescence and MitoTracker Red fluorescence (Fig. 3E). These results suggest that treatment of lung cancer cells with mitocurcumin disrupted mitochondrial redox balance leading to increased oxidative stress. Since, increased oxidative stress is known to induce DNA damage, we have examined the extent of mitochondrial DNA damage after mitocurcumin treatment. Mitocurcumin treatment (2.5 µM) induced mitochondrial DNA damage to the extent of 5 breaks per 10 Kb of DNA (Fig. 3F), whereas no significant nuclear DNA damage was observed. These results indicate that mitocurcumin inhibited mitochondrial TrxR leading to oxidative stress which is evinced by increased mitochondrial DNA damage.
2.20. Statistical analysis GraphPad Prism software was used to perform statistical analysis (La Jolla, USA). To calculate the significance between the mean of two groups t-test was used. If there were more than two groups for comparison, one-way ANOVA was performed followed by Tukey-Kramer multiple comparisons post-test. Difference between means was considered significant if p < 0.05. 3. Results 3.1. Mitocurcumin induces cell death in lung cancer cells and reduces the frequency of cancer stem cells The efficacy of mitocurcumin in inducing cell death in A549 cells was evaluated using clonogenic survival assay as compared to its parent molecule, curcumin. Mitocurcumin treatment resulted in significant decrease in survival fraction at 0.1 µM and complete abrogation of clonogenic potential was observed at 1 μM (Fig. 1A and B). On the contrary, curcumin showed a significant decrease in survival fraction at 25 µM. Curcumin at 50 µM completely suppressed clonogenic survival indicating that mitocurcumin is 50-fold more effective than curcumin (Fig. 1A and B) in inhibiting colony forming ability. Further, cell death was estimated by enumeration of cells with sub-G1 DNA content using PI assay. Mitocurcumin at 0.5 µM induced significant apoptosis and 80% of the cells were dead at 5 µM. On the other hand, curcumin at 25 μM induced significant apoptosis and 35% of the cells were dead at 50 µM suggesting the better efficacy of mitocurcumin in killing lung cancer cells (Fig. 1C and D). Similar results were obtained using MTT assay (Fig. S1A). These results were further confirmed by live and dead assay (Fig. 1E). Apart from A549 cells, effect of mitocurcumin was also evaluated in prostate cancer cells (PC-3 cells). Mitocurcumin at 2.5 µM, inhibited the colony forming ability of these cells and induced significant apoptosis indicating that it can act like an anticancer agent in different tumor types (Fig. S1 B–E). Further, the effect of mitocurcumin on normal cells was studied in primary mouse embryonic fibroblast cells (NIH 3T3 cells) and mouse lymphocytes. As compared to curcumin, mitocurcumin showed little toxicity to NIH 3T3 and lymphocytes as indicated by clonogenic survival, live and dead assay and apoptosis (Fig. S2 A–D; Fig. S3 A and B). Since mitocurcumin induced cytotoxicity in cancer cells, its effect on cancer stem cells was also investigated. Interestingly, treatment of lung cancer cells with 2.5 μM mitocurcumin led to significant decrease in the frequency of cancer stem cells as compared to vehicle treated group as demonstrated by a decrease in the side population and percent CD44+CD24- cells (Fig. 1F). These results infer that mitocurcumin induces cell death in lung cancer cells and also decreases the frequency of cancer stem cells.
3.4. TrxR2 overexpression accentuated mitocurcumin mediated apoptosis in cancer cells To confirm the role of TrxR2 in mitocurcumin mediated cell death, we have overexpressed TrxR2 in A549 cells using a CRISPR-Cas9 based activation plasmid (Fig. S3C). Surprisingly, TrxR2 overexpressing cells treated with mitocurcumin showed higher cell death as compared to cells transfected with scramble plasmid and treated with mitocurcumin (Fig. 4A). This was further confirmed by increase in sub-G1 population (Fig. 4B). Mitocurcumin mediated cytochrome-C release was also enhanced in the cells overexpressing TrxR2, as compared to cells transfected with scramble plasmid and treated with mitocurcumin (Fig. 4C). TrxR2 overexpression also accentuated the mitocurcumin mediated mitochondrial ROS accumulation (Fig. 4D). TrxR2 overexpression resulted in further decrease in the frequency of cancer stem cells following mitocurcumin treatment (Fig. 4E). In a cell free system when mitocurcumin was added to TrxR, mitocurcumin modulated its activity to NADPH oxidase like activity (Fig. 4 F). Further, in TrxR2 knock out A549 cells (Fig. S3 D), the ability of mitocurcumin to induce apoptosis was significantly reduced as compared to scramble control cells treated with mitocurcumin (Fig. 4G and H). Pretreatment with anti-oxidants glutathione or N-acetyl-cysteine also resulted in significant reduction in mitocurcumin mediated apoptosis in A549 cells. However, PEGylated catalase failed to abrogate mitocurcumin mediated apoptosis in A549 cells (Fig. S4A and B).
3.2. Mitocurcumin activated intrinsic apoptotic pathway in A549 cells Mitocurcumin treatment resulted in significant increase in the frequency of annexin-V-FITC positive cells in a dose dependent manner (Fig. 2A). The ratio between BAX (pro-apoptotic) to BCL2 (anti-apoptotic) proteins is an important indicator of mitochondria mediated apoptosis. Treatment of A549 cells with mitocurcumin led to significant increase in BAX levels and a concomitant decrease in BCL2 levels as estimated by immunostaining (Fig. 2B). Further, mitocurcumin treatment significantly enhanced the cytosolic cytochrome C levels and caspase 3 activity in lung cancer cells (Figs. 2C and 2D). Mitocurcumin treatment induced a significant loss of MMP as estimated by red to green fluorescence ratio after JC-1 staining (Figs. 2E and 2F). Together these results imply that mitocurcumin activated the mitochondrial apoptotic pathway.
4. Discussion In the search for new cancer treatment options, discovery of compounds which specifically target mitochondria may provide unique chemotherapeutic strategies. Mitocurcumin, which has been shown to accumulate in mitochondria, is one such agent. However, its anticancer effects and mechanisms of action are not well studied. Mitocurcumin showed 25–50 fold better efficacy than curcumin in killing lung cancer cells. Cancer stem cells pose a major challenge in achieving higher therapeutic efficacy and hence the effect of mitocurcumin on cancer stem cells was evaluated. Mitocurcumin treatment significantly reduced the frequency of cancer stem cells as evinced by side population assay 533
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Fig. 1. Mitocurcumin exhibits better anti-cancer efficacy than curcumin. Anticancer effect of mitocurcumin (MC) and curcumin (Cur) on A549 cells was assessed by clonogenic assay. The graph shows survival fraction (A), and representative images of the dishes containing colonies are shown in (B). The cells were treated with mitocurcumin, cultured for 48 h, stained with PI solution and acquired using a high throughput cell analyzer. The graph shows percent apoptotic (Sub G1) cells (C), and representative images of wells show live population in yellow and dead population in red (D). Live and dead assay was performed to study the anti-tumor effect of mitocurcumin. The cells were grown on a coverslip and were treated with mitocurcumin, cultured for 24 h, and stained with calcein-AM and ethidium homodimer-1. Green cells represent the live population, and red cells represent dead population (E). Effect of mitocurcumin on cancer stem cells was studied by side population assay and antibody staining (F). For enumerating the frequency of cancer stem cells, side population assay (left panel) and CD24, CD44 antibody staining (right panel) were performed 24 h after mitocurcumin treatment. Graphed are the frequency of cancer stem cells and overlaid dot plots are shown in inset (F). Each data point in the graphs represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to respective vehicle control (DMSO).
534
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Fig. 2. Mitocurcumin induces cell death through induction of apoptosis in lung cancer cells. Treatment of A549 cells with mitocurcumin (24 h) led to the induction of apoptosis as shown by annexin-V-FITC staining (A). Overlaid representative flow cytometric histograms are shown in inset (A). Mitocurcumin (MC) treatment led to change in levels of BAX and BCL2 as estimated by intracellular antibody staining and flow cytometry (B), cytosolic cytochrome-C levels (representative flow cytometric histograms are shown in inset, C), caspase-3 activity (D) and mitochondrial membrane potential as studied by JC-1 red / green fluorescence ratio (E). Red and green fluorescence of JC-1 probe was also visualized under confocal laser scanning microscope and representative mages are shown in (F). Each data point in the graphs represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to respective DMSO control group.
Fig. 3. Mitocurcumin treatment inhibits TrxR2 activity and alters mitochondrial redox. Binding affinity between TrxR2 and mitocurcumin (MC) was studied using in silico docking analysis. The site of docking is shown in (A). For in silico docking studies, binding affinity between TrxR2 and mitocurcumin was studied using Autodock Vina tool. Effect of mitocurcumin on the TrxR activity in cell free system was determined using DTNB reduction method after pre-incubating purified rat TrxR with mitocurcumin (B). Effect of mitocurcumin on TrxR activity in the cellular system was studied using micro-method of insulin reduction (C). Effect of mitocurcumin on mitochondrial ROS was studied by MitoSOX Red staining followed by spectrofluorimetry (D) and was also visualized by confocal laser scanning microscopy. Inset: Blue color represents nucleus and red color indicates mitochondrial ROS (D). Effect of mitocurcumin on mitochondrial GSH was studied by confocal laser scanning microscopy. Mitochondria was visualized using MitoTracker Red dye (Red fluorescence), the nucleus was visualized using DAPI dye (blue fluorescence) and mitochondrial GSH level was estimated from CMFDA green fluorescence. Representative images of cells are shown in (E). Effect of mitocurcumin treatment on mitochondrial and nuclear DNA damage was studied using real time qPCR technique and is graphed in (F). Each data point in the figures represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to DMSO control group.
chemotherapeutic drugs disrupt mitochondrial redox thereby inducing apoptosis in cancer cells. However, overexpression of TrxR2 results in effective neutralization of excess mitochondrial ROS induced by drugs and hence prevent ROS induced cell death. It is well known that overexpression of TrxR2 in cancer cells can adversely affect the
as well as surface marker analysis. TrxR2 is overexpressed in many cancers including lung cancer [11]. Overexpression of thioredoxin system contributes to uncontrolled proliferation, angiogenesis, resistance to cell death and protects the cancer cells from stress dysregulated redox signaling [31,32]. Many 535
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
treatment outcome and inhibition of TrxR2 can be a good strategy to kill cancer cells [32]. In silico studies revealed a very high affinity binding of mitocurcumin with TrxR2. Hydrophobic interactions played a major role in stabilizing mitocurcumin in the cleft at the active site of human mitochondrial TrxR2. Tyr406, Lys31, Lys407, Lys481 stabilized the triphenyl rings at one end of the ligand while the other end was stabilized by Val62, Thr60, Gly23, Gly25 and Gly59. The critical Cys497 and SelenoCys498 residues were located close to the backbone of mitocurcumin in the docked position (Fig. S4C). Experiments in cell free system showed that mitocurcumin indeed inhibited TrxR activity by physical interaction. In agreement with these findings, mitocurcumin also inhibited cellular TrxR activity. TrxR2 inhibition can lead to higher accumulation of ROS and thereby induce oxidative stress in cells. Our experiments confirmed that treatment of
Table 1 Docking energy values of binding of mitocurcumin at TrxR active sites. Docking energy values for binding at active sites of rat TrxR (Kcal/mol) Molecule ΔG at E-chain active site ΔG at F-chain active site Curcumin -7.3 -7.5 Mitocurcumin -8.2 -8.8 Docking energy values for binding at active sites of human mitochondrial TrxR (Kcal/ mol) Molecule ΔG at A-chain active site ΔG at B-chain active site Curcumin -7.6 -7.1 Mitocurcumin -8.4 -7.1 Docking energy values for binding at active sites of human cytosolic TrxR (Kcal/mol) Molecule ΔG at A-chain active site ΔG at B-chain active site Curcumin -6.7 -6.8 Mitocurcumin -7.8 -8.4
Fig. 4. Effect of mitocurcumin on TrxR2 overexpressing and TrxR2 knockdown cells. A549 cells overexpressing TrxR2 were treated with mitocurcumin (MC) for 24 h and apoptosis was studied by annexin-V-FITC staining followed by flow cytometry. Graphed is the percent apoptotic cells and representative flow cytometric histograms are shown in inset (A). Percent apoptotic cells were also studied by analyzing the cells with sub-G1 DNA content using a high throughput cell analyzer and are shown in (B). Cytosolic cytochrome C levels were estimated by intracellular antibody staining and flow cytometry. Graphed are the percent cells containing cytosolic cytochrome C and overlaid flow cytometric histograms are shown in inset (C). The effect of TrxR2 overexpression and mitocurcumin treatment on mitochondrial ROS was estimated by MitoSOX red staining and is graphed in (D). The frequency of cancer stem cells in these treatment groups was assessed by side population assay using high throughput cell analyzer and is shown in (E). Oxidation of NADPH by TrxR in the presence of mitocurcumin was studied in cell free system. The relative enzyme activity was calculated as the change in NADPH absorbance at 340 nm using a spectrophotometer and is shown in (F). The effect of mitocurcumin on TrxR2 knockout cells: TrxR2 knockout cells were generated using CRISPR-Cas9 plasmid and treated with mitocurcumin. Cells were cultured for 48 h and percent apoptosis was estimated by staining cells with annexin-V-FITC followed by acquisition on flow cytometer (G & H). Each data point in the figure represents mean ± S.E.M. from three replicates and two such independent experiments were carried out. *p < 0.05 as compared to DMSO control group. $p < 0.05 as compared to TrxR+NADPH+MC1 group. #p < 0.05 as compared to scramble + mitocurcumin group.
536
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Scheme 1. Mechanism of action of anti-cancer activity of mitocurcumin.
lead to higher accumulation of ROS in the presence of mitocurcumin resulting in further increase in the magnitude of apoptosis. We propose that targeting TrxR2 by mitocurcumin can be very effective in eliminating cancer cells as well as cancer stem cells. In conclusion, mitocurcumin has been found to be a potent anticancer agent against lung cancer cells as compared to curcumin. Mitocurcumin acts by modifying TrxR2 activity, disturbing mitochondrial redox, inducing a loss in MMP and thereby triggering apoptosis. Hence, mitocurcumin can be a potent chemotherapeutic agent and may also find application as an adjuvant along with the existing chemotherapeutic drugs in cancers which overexpress TrxR2.
cells with mitocurcumin significantly enhanced mitochondrial ROS. Interestingly, pretreatment with GSH or NAC significantly inhibited mitocurcumin mediated apoptosis suggesting that ROS may be playing an important role in the anticancer activity of mitocurcumin. However, PEGylated catalase did not alter the anti-cancer effect of mitocurcumin. This could be due to poor localization of PEGylated catalase in the mitochondria to scavenge ROS generated by mitocurcumin. The level of total cellular thiols (protein and non-protein) and their reduced or oxidized state is an important determinant of cellular redox status. Our experimental findings revealed that mitocurcumin treatment significantly decreased mitochondrial GSH levels thereby disturbing mitochondrial redox balance. Increased oxidative stress can damage DNA bases and also induce strand breaks. It was observed that mitocurcumin treatment induced significant mitochondrial DNA damage. Excess mitochondrial ROS can trigger apoptosis through release of cytochrome C. These pro-apoptotic proteins are released through the transmembrane mitochondrial pore formed by oligomerization of BAX recruited on the mitochondria. An acute increase in the mitochondrial ROS is essential for oligomerization and recruitment of BAX and formation of pores [33]. In our study, when TrxR2 was modulated by mitocurcumin, it led to the accumulation of ROS, increase in BAX/BCL2 ratio, cytochrome C release and loss of MMP followed by apoptosis (Scheme 1). TrxR2 knockout cells were refractory to mitocurcumin induced apoptosis indicating that this molecule modulates Trx system in exhibiting anti-cancer activity. On the contrary, when TrxR2 was overexpressed using CRISPR-Cas9 plasmid in lung cancer cells, it led to a significant increase in the apoptosis induced by mitocurcumin. Previously, Fang et al. have reported that curcumin converted thioredoxin reductase to NADPH oxidase like activity [17]. We hypothesized that mitocurcumin might also modulate the TrxR2 activity to NADPH oxidase like activity by preventing the transfer of an electron to its substrate and thereby shunting the NADPH derived electron to oxygen leading to the production of superoxide. In agreement with this hypothesis, it was observed that mitocurcumin modulated TrxR2 activity to NADPH oxidase like activity. TrxR2 overexpression also augmented mitocurcumin induced apoptosis in cancer stem cells which could be due to further increase in ROS generation. Many studies have shown that Nrf-2 mediated antioxidant machinery is highly active in cancer stem cells [34,35]. Since TrxR2 expression is governed by Nrf-2, the de novo levels of TrxR2 may be higher in cancer stem cells. We found that overexpression of TrxR2
Acknowledgements The authors would like to thank Ms. Binita Kumar for assistance in flow cytometry. We also acknowledge the technical help provided by Mr. Deepak V. Kathole, Mr. B. A. Naidu, and Mr. K.S. Munankar. This work is funded by Department of Atomic Energy, Government of India (XII-N-R & D-17). Conflict of interest None Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed.2017.10. 378. References [1] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (2015) E359–E386. [2] H.J. Mehta, V. Patel, R.T. Sadikot, Curcumin and lung cancer–a review, Target. Oncol. 9 (2014) 295–310. [3] V. Gogvadze, Targeting mitochondria in fighting cancer, Curr. Pharm. Des. 17 (2011) 4034–4046. [4] E.S. Arner, A. Holmgren, Physiological functions of thioredoxin and thioredoxin reductase, Eur. J. Biochem. 267 (2000) 6102–6109. [5] A. Miranda-Vizuete, A.E. Damdimopoulos, G. Spyrou, The mitochondrial thioredoxin system, Antioxid. Redox Signal. 2 (2000) 801–810. [6] L.E. Netto, F. Antunes, The roles of peroxiredoxin and thioredoxin in hydrogen peroxide sensing and in signal transduction, Mol. Cells 39 (2016) 65–71.
537
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al. [7] J.E. Biaglow, R.A. Miller, The thioredoxin reductase/thioredoxin system: novel redox targets for cancer therapy, Cancer Biol. Ther. 4 (2005) 6–13. [8] J.H. Choi, T.N. Kim, S. Kim, S.H. Baek, J.H. Kim, S.R. Lee, J.R. Kim, Overexpression of mitochondrial thioredoxin reductase and peroxiredoxin III in hepatocellular carcinomas, Anticancer Res. 22 (2002) 3331–3335. [9] A. Kunwar, S. Jayakumar, A.K. Srivastava, K.I. Priyadarsini, Dimethoxycurcumininduced cell death in human breast carcinoma MCF7 cells: evidence for pro-oxidant activity, mitochondrial dysfunction, and apoptosis, Arch. Toxicol. 86 (2012) 603–614. [10] J. Hellfritsch, J. Kirsch, M. Schneider, T. Fluege, M. Wortmann, J. Frijhoff, M. Dagnell, T. Fey, I. Esposito, P. Kölle, K. Pogoda, J.P.F. Angeli, I. Ingold, P. Kuhlencordt, A. Östman, U. Pohl, M. Conrad, H. Beck, Knockout of mitochondrial thioredoxin reductase stabilizes prolyl hydroxylase 2 and inhibits tumor growth and tumor-derived angiogenesis, Antioxid. Redox Signal. 22 (2015) 938–950. [11] L. Bu, W. Li, Z. Ming, J. Shi, P. Fang, S. Yang, Inhibition of TrxR2 suppressed NSCLC cell proliferation, metabolism and induced cell apoptosis through decreasing antioxidant activity, Life Sci. (2017). [12] M.R. Kim, H.S. Chang, B.H. Kim, S. Kim, S.H. Baek, J.H. Kim, S.R. Lee, J.R. Kim, Involvements of mitochondrial thioredoxin reductase (TrxR2) in cell proliferation, Biochem. Biophys. Res. Commun. 304 (2003) 119–124. [13] S. Vinogradov, X. Wei, Cancer stem cells and drug resistance: the potential of nanomedicine, Nanomedicine (London England) 7 (2012) 597–615. [14] S. Prasad, S.C. Gupta, A.K. Tyagi, B.B. Aggarwal, Curcumin, a component of golden spice: from bedside to bench and back, Biotechnol. Adv. 32 (2014) 1053–1064. [15] S.C. Gupta, S. Patchva, B.B. Aggarwal, Therapeutic roles of curcumin: lessons learned from clinical trials, AAPS J. 15 (2013) 195–218. [16] W. Cai, B. Zhang, D. Duan, J. Wu, J. Fang, Curcumin targeting the thioredoxin system elevates oxidative stress in HeLa cells, Toxicol. Appl. Pharmacol. 262 (2012) 341–348. [17] J. Fang, J. Lu, A. Holmgren, Thioredoxin reductase is irreversibly modified by curcumin: a novel molecular mechanism for its anticancer activity, J. Biol. Chem. 280 (2005) 25284–25290. [18] S. Prasad, A.K. Tyagi, B.B. Aggarwal, Recent developments in delivery, bioavailability, absorption and metabolism of curcumin: the golden pigment from golden spice, Cancer Res. Treat.: Off. J. Korean Cancer Assoc. 46 (2014) 2–18. [19] G. Shoba, D. Joy, T. Joseph, M. Majeed, R. Rajendran, P.S. Srinivas, Influence of piperine on the pharmacokinetics of curcumin in animals and human volunteers, Planta Med. 64 (1998) 353–356. [20] C.A. Reddy, V. Somepalli, T. Golakoti, A.K. Kanugula, S. Karnewar, K. Rajendiran, N. Vasagiri, S. Prabhakar, P. Kuppusamy, S. Kotamraju, V.K. Kutala, Mitochondrialtargeted curcuminoids: a strategy to enhance bioavailability and anticancer efficacy of curcumin, PLoS One 9 (2014) e89351. [21] S. Jayakumar, A. Kunwar, S.K. Sandur, B.N. Pandey, R.C. Chaubey, Differential response of DU145 and PC3 prostate cancer cells to ionizing radiation: role of
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29] [30]
[31] [32]
[33] [34]
[35]
538
reactive oxygen species, GSH and Nrf2 in radiosensitivity, Biochim. Et. Biophys. Acta 1840 (2014) 485–494. C. Riccardi, I. Nicoletti, Analysis of apoptosis by propidium iodide staining and flow cytometry, Nat. Protoc. 1 (2006) 1458–1461. S. Jayakumar, R.S. Patwardhan, D. Pal, D. Sharma, S.K. Sandur, Dimethoxycurcumin, a metabolically stable analogue of curcumin enhances the radiosensitivity of cancer cells: possible involvement of ROS and thioredoxin reductase, Biochem. Biophys. Res. Commun. 478 (2016) 446–454. H. Komuro, R. Saihara, M. Shinya, J. Takita, S. Kaneko, M. Kaneko, Y. Hayashi, Identification of side population cells (stem-like cell population) in pediatric solid tumor cell lines, J. Pediatr. Surg. 42 (2007) 2040–2045. D. Sharma, M.R. Eichelberg, J.D. Haag, A.L. Meilahn, M.J. Muelbl, K. Schell, B.M. Smits, M.N. Gould, Effective flow cytometric phenotyping of cells using minimal amounts of antibody, Biotechniques 53 (2012) 57–60. R.S. Patwardhan, R. Checker, D. Sharma, S.K. Sandur, K.B. Sainis, Involvement of ERK-Nrf-2 signaling in ionizing radiation induced cell death in normal and tumor cells, PLoS One 8 (2013) e65929. R.S. Patwardhan, D. Sharma, R. Checker, M. Thoh, S.K. Sandur, Spatio-temporal changes in glutathione and thioredoxin redox couples during ionizing radiationinduced oxidative stress regulate tumor radio-resistance, Free Radic. Res. (2015) 1–15. S. Kumar, A. Holmgren, Induction of thioredoxin, thioredoxin reductase and glutaredoxin activity in mouse skin by TPA, a calcium ionophore and other tumor promoters, Carcinogenesis 20 (1999) 1761–1767. D.W. Hedley, S. Chow, Evaluation of methods for measuring cellular glutathione content using flow cytometry, Cytometry 15 (1994) 349–358. S.O. Evans, M.B. Jameson, R.T. Cursons, L.M. Peters, S. Bird, G.M. Jacobson, Development of a qPCR method to measure mitochondrial and genomic DNA damage with application to chemotherapy-Induced DNA damage and cryopreserved cells, Biology 5 (2016). G. Powis, D.L. Kirkpatrick, Thioredoxin signaling as a target for cancer therapy, Curr. Opin. Pharmacol. 7 (2007) 392–397. E.E. Fink, S. Mannava, A. Bagati, A. Bianchi-Smiraglia, J.R. Nair, K. Moparthy, B.C. Lipchick, M. Drokov, A. Utley, J. Ross, L.P. Mendeleeva, V.G. Savchenko, K.P. Lee, M.A. Nikiforov, Mitochondrial thioredoxin reductase regulates major cytotoxicity pathways of proteasome inhibitors in multiple myeloma cells, Leukemia 30 (2016) 104–111. C. Fleury, B. Mignotte, J.L. Vayssiere, Mitochondrial reactive oxygen species in cell death signaling, Biochimie 84 (2002) 131–141. S. Bao, Q. Wu, R.E. McLendon, Y. Hao, Q. Shi, A.B. Hjelmeland, M.W. Dewhirst, D.D. Bigner, J.N. Rich, Glioma stem cells promote radioresistance by preferential activation of the DNA damage response, Nature 444 (2006) 756–760. S. Ding, C. Li, N. Cheng, X. Cui, X. Xu, G. Zhou, Redox regulation in cancer stem cells, Oxid. Med. Cell. Longev. 2015 (2015) 750798.
Contents lists available at ScienceDirect
Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed
Original article
Mitochondrial targeted curcumin exhibits anticancer effects through disruption of mitochondrial redox and modulation of TrxR2 activity
MARK
Sundarraj Jayakumara,1, Raghavendra S. Patwardhana,1, Debojyoti Pala, Babita Singha, ⁎ ⁎ Deepak Sharmaa,b, , Vijay Kumar Kutalac, Santosh Kumar Sandura,b, a b c
Radiation Biology & Health Sciences Division, Modular Laboratories, Bhabha Atomic Research Centre, Trombay, Mumbai, India Homi Bhabha National Institute, Anushakti Nagar, Mumbai, India Department of Clinical Pharmacology & Therapeutics, Nizam's Institute of Medical Sciences, Hyderabad, India
A R T I C L E I N F O
A B S T R A C T
Keywords: Mitocurcumin Lung cancer Cancer stem cells Thioredoxin reductase 2 Mitochondrial DNA damage Curcumin
Mitocurcumin is a derivative of curcumin, which has been shown to selectively enter mitochondria. Here we describe the anti-tumor efficacy of mitocurcumin in lung cancer cells and its mechanism of action. Mitocurcumin, showed 25–50 fold higher efficacy in killing lung cancer cells as compared to curcumin as demonstrated by clonogenic assay, flow cytometry and high throughput screening assay. Treatment of lung cancer cells with mitocurcumin significantly decreased the frequency of cancer stem cells. Mitocurcumin increased the mitochondrial reactive oxygen species (ROS), decreased the mitochondrial glutathione levels and induced strand breaks in the mitochondrial DNA. As a result, we observed increased BAX to BCL-2 ratio, cytochrome C release into the cytosol, loss of mitochondrial membrane potential and increased caspase-3 activity suggesting that mitocurcumin activates the intrinsic apoptotic pathway. Docking studies using mitocurcumin revealed that it binds to the active site of the mitochondrial thioredoxin reductase (TrxR2) with high affinity. In corroboration with the above finding, mitocurcumin decreased TrxR activity in cell free as well as the cellular system. The anticancer activity of mitocurcumin measured in terms of apoptotic cell death and the decrease in cancer stem cell frequency was accentuated by TrxR2 overexpression. This was due to modulation of TrxR2 activity to NADPH oxidase like activity by mitocurcumin, resulting in higher ROS accumulation and cell death. Thus, our findings reveal mitocurcumin as a potent anticancer agent with better efficacy than curcumin. This study also demonstrates the role of TrxR2 and mitochondrial DNA damage in mitocurcumin mediated killing of cancer cells.
1. Introduction Lung cancer is the most prevalent cancer in the world and causes excessive morbidity [1]. Since more than seventy-five percent of the lung cancers are presented at advanced stages, the currently available treatment modalities are not very effective. Thus, there is a need to develop newer chemotherapeutic modalities for treatment of lung cancer [2]. Mitochondrion plays a vital role in the cell survival and growth as it is involved in the energy metabolism. Mitochondrion is also a major source of ROS generation and immoderation of ROS production can trigger apoptotic cell death. Hence targeting mitochondrion can be a good strategy for killing cancer cells [3]. One of the major antioxidant defense mechanisms by which mitochondria neutralize the excess ROS is through a dedicated thioredoxin system comprising of thioredoxin 2 (Trx2), thioredoxin reductase 2 (TrxR2) and NADPH [4,5]. TrxR2 is a selenocysteine containing homodimeric enzyme, ⁎
1
which is a member of the nucleotide-disulfide oxidoreductase protein family. Its primary function is to keep Trx2 in the reduced form which in turn reduces peroxiredoxins. Peroxiredoxins detoxify hydrogen peroxide and lipid hydroperoxides to water or alcohol [6]. Since cancer cells thrive under oxidative microenvironment, they are over dependent on thioredoxin system to counter this oxidative stress. There are several reports showing overexpression of thioredoxin system in different cancers [7]. TrxR2 has also been shown to be overexpressed in several cancers including lung cancer [8,9] and it is also involved in tumor progression by promoting vascularization through stabilization of HIF1α [10]. Overexpression of TrxR2 is associated with poor prognosis and therapeutic resistance of cancer and hence targeting TrxR2 can be an attractive strategy in treating cancers [8,11,12]. A subset of cancer cells which are resistant to chemotherapeutic drugs and contribute to relapse as well as metastases are identified as cancer stem cells in solid cancers [13]. Treatment outcome of cancer depends on effective elimination of
Corresponding authors at: Radiation Biology & Health Sciences Division, Modular Laboratories, Bhabha Atomic Research Centre, Trombay, Mumbai, India E-mail addresses: [email protected] (D. Sharma), [email protected] (S. Kumar Sandur). Both the authors have contributed equally.
http://dx.doi.org/10.1016/j.freeradbiomed.2017.10.378 Received 20 July 2017; Received in revised form 6 October 2017; Accepted 23 October 2017 0891-5849/ © 2017 Elsevier Inc. All rights reserved.
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
of nuclear DNA content. These hypodiploid cells can be enumerated by staining with DNA binding dye propidium iodide (PI) [22]. Exponentially growing cells were treated with mitocurcumin, cultured for 48 h and stained with PI solution (1X PBS containing sodium citrate, RNase, triton × 100 and 50 μg/ml PI) overnight at 4 °C [21]. The cells were acquired on high throughput cell analyzer, (Acumen Cellista, TTP Labtech, UK) and cells showing sub G1 DNA content (hypodiploid) were enumerated using Cellista software.
cancer stem cells. Hence, agents that can target these cells are also good candidates for cancer treatment. Curcumin is a major constituent present in an Indian herb Curcuma longa, and has been shown to possess many pharmacological properties [14]. Several human clinical trials are underway using curcumin as an adjuvant for various disease conditions [15]. Inhibition of cytosolic thioredoxin reductase (TrxR1) is one of the major mechanisms through which curcumin exert its anticancer effects [16,17]. However, curcumin has very poor stability and low bioavailability [18,19]. To circumvent this problem, a modified form of curcumin i.e., mitocurcumin which targets mitochondria has been synthesized [20]. Mitocurcumin has been shown to accumulate in mitochondria, and it could be a potent inhibitor of mitochondrial thioredoxin reductase (TrxR2). Using a combination of cell biology assays, high content screening, flow cytometry and in silico docking, we have shown the anti-tumor activity of mitocurcumin in lung cancer cells and its underlying molecular mechanism of action.
2.5. Live and dead assay Intracellular esterase activity and intact plasma membrane are characteristics of viable cells. Live and dead assay differentiates between viable and dead cells. Active esterases present in live cells convert calcein-AM into calcein which exhibits green fluorescence, whereas, dead cells accumulate ethidium homodimer-1 owing to their compromised membrane integrity and exhibit red fluorescence. Cells were grown on coverslip, treated with mitocurcumin, cultured for 24 h and live and dead assay was performed using a commercial kit following the manufacturer's protocol (ThermoFisher Scientific, MA, USA).
2. Materials and methods 2.1. Materials
2.6. MTT assay
Mitocurcumin was synthesized and characterized in the lab as mentioned previously [20]. Curcumin, verapamil, crystal violet, methanol, formaldehyde, paraformaldehyde, sodium citrate, RNase, JC-1, propidium iodide (PI), insulin, NADPH, NADPH oxidase, 5,5′-dithiobis 2-nitrobenzoic acid (DTNB), purified rat TrxR, purified human Trx, caspase-3 activity kit, annexin-V-FITC kit, fluorochrome conjugated secondary antibody, MTT, dimethyl sulfoxide (DMSO), triton X-100, DMEM, penicillin, streptomycin and trypsin-EDTA were purchased from Sigma Chemical Co. (MO, USA). Live and dead assay kit, DAPI with antifade, MitoSOX Red, 5-chloromethylfluorescein diacetate (CMFDA) dye, MitoTracker Red and Hoechst 33342 were purchased from ThermoFisher Scientific (MA, USA). Fluorochrome conjugated monoclonal antibodies against CD44 and CD24 were procured from BD Pharmingen (CA, USA). Monoclonal antibodies against BAX, BCL2 and cytochrome C were purchased from Cell Signaling Technologies (MA, USA). TrxR2 overexpression plasmids and knockout plasmids were purchased from Santacruz (MD, USA). Fetal bovine serum (FBS) was obtained from HiMedia (Mumbai, India).
Five thousand cells were seeded in 96-well plates overnight, treated with mitocurcumin, cultured for 48 h and were treated with MTT reagent for 4 h. Formazan crystals were dissolved in DMSO and absorbance was taken at 570 nm. Percent cell viability was calculated based on the absorbance value. 2.7. Enumeration of cancer stem cells Cancer stem cells can be enumerated by side population assay which is based on their ability to efflux Hoechst 33342 dye efficiently due to their active ABCG2 transporter as compared to other cancer cells. Cells were treated with vehicle or mitocurcumin and cultured in a 96-well plate for 24 h. Cells were washed and stained with Hoechst 33342 (10 μg/ml) in phenol red free DMEM for 90 min at 37 °C [23]. Hoechst 33342 red (620 nm) and blue (450 nm) fluorescence was recorded using high throughput cell analyzer, (Acumen, TTP Labtech UK). The live cells showing low Hoechst fluorescence in the red channel (620 nm) and blue channel (450 nm) were gated and plotted as side population cells. Verapamil which is a well-known inhibitor of ABCG2 transporter protein was used as a control in this experiment [24]. Cancer stem cells can also be distinguished based on characteristic surface markers like the presence of CD44 and the absence of CD24. The frequency of cancer stem cells was also enumerated by surface staining with CD44-PE and CD24-PE-Cy7 conjugated monoclonal antibodies followed by acquiring on a flow cytometer [25].
2.2. Cell culture A549 lung cancer cells, NIH 3T3 mouse embryonic fibroblast cells and PC-3 prostate cancer cells were procured form National Centre for Cell Sciences, Pune, India. Cells were grown as monolayer cultures in DMEM medium with 10% FBS along with penicillin and streptomycin in 5% CO2 and 95% air at 37 °C. 2.3. Clonogenic assay
2.8. Estimation of apoptosis by annexin-V-FITC staining Exponentially growing cells were harvested, seeded at 1000 cells per well in six well plates and allowed to adhere overnight. Cells were treated with various concentrations of mitocurcumin or curcumin and cultured for 14 days for colony formation [21]. Colonies were fixed using methanol, stained with crystal violet and counted using GelCount (Oxford Optronix, UK) instrument. Survival fraction was calculated using following formula: Survival fraction = number of colonies observed x 100 / number of cells plated x PE, where PE (plating efficiency) = number of colonies developed in control/ number of cells plated in control.
Cells undergoing apoptosis display phosphatidylserine on the outer leaflet of plasma membrane which can be detected using fluorochrome conjugated annexin-V. Cells were treated with mitocurcumin, cultured for 24 h and stained with annexin-V-FITC and PI as per manufacturer's protocol [9]. Briefly, cells were washed with 1X PBS, resuspended in 1X binding buffer containing annexin-V-FITC and PI for 10 min at room temperature. Cells were again washed and resuspended in 1X PBS and acquired using flow cytometer. Annexin-V positive cells were gated using FlowJo software.
2.4. Quantification of cell death by PI assay using high throughput cell analyzer
2.9. Intracellular antibody staining of BAX and BCL2 proteins Intracellular antibody staining was performed using BAX and BCL2 antibodies followed by staining with labeled secondary antibody as
Apoptotic cell death is characterized by DNA fragmentation and loss 531
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
2.14. TrxR activity in cell free system
described earlier [26]. Briefly, cells were treated with either DMSO (vehicle) or mitocurcumin (2.5 µM) for 24 h washed with 1X PBS, fixed with formaldehyde, permeabilized with 1X PBS containing 0.1% tritonX-100 and blocked in the presence of 5% FBS. Cells were resuspended in 20 μl antibody staining solution containing 0.3 μg each of BAX and BCL2 antibodies. Cells were incubated for 1 h at room temperature and cells were stained with fluorochrome conjugated secondary antibody. Cells were washed, resuspended in PBS and stained with Hoechst 33342 (10 μg/ml) and acquired on a flow cytometer.
Purified rat TrxR (110 nmol/L) was incubated with indicated concentrations of mitocurcumin (1, 2.5, 5 & 10 µM) or vehicle at room temperature for 1 h. TrxR activity was determined by DTNB reduction assay as mentioned earlier [23]. 2.15. Estimation of TrxR and NADPH oxidase activities Cells were treated with either DMSO (vehicle) or mitocurcumin (2.5 µM), for 24 h. TrxR activity was determined by the micro-method of insulin reduction as described elsewhere [27,28]. All assay tubes contained 0.26 M HEPES, pH 7.6, 10 mM EDTA, 2 mM NADPH, 1 mM insulin and 100 nM thioredoxin and cell extract in a final volume of 100 μl. After incubating at 37 °C for 20 min reaction was stopped by adding 500 μl stopping solution containing 0.2 M Tris-HCl, 6 M guanidine-HCl, 1 mM DTNB and absorbance was measured at 412 nm. To determine the NADPH oxidase activity, NADPH-reduced TrxR was incubated with different concentrations of mitocurcumin at room temperature for 2 h. Equal volume of DMSO was added to the control. In reaction mixture, modified TrxR was added to TE buffer containing 200 μM NADPH. NADPH oxidase enzyme was kept as positive control. Oxidation of NADPH was monitored by following the changes in absorbance at 340 nm.
2.10. Caspase-3 activity measurement Caspase-3 is a protease which is an important mediator in executing apoptosis. Caspase-3 activity is proportional to the release of fluorescent molecule (7-amino 4-methylcoumarin) which is attached to peptide substrate of caspase-3. Cells treated with mitocurcumin were harvested and whole cell extracts were prepared for determination of caspase-3 activity using a kit as per manufacturer's protocol. 2.11. Estimation of cytochrome C A549 cells were cultured for 24 h in presence of mitocurcumin, harvested, fixed with paraformaldehyde, permeabilized (using selective permeabilization buffer from InnoCyte™ which permeabilizes plasma membrane but not the mitochondrial membrane), blocked and stained with monoclonal antibody against cytochrome C followed by fluorochrome conjugated secondary antibody. Cells were acquired on a flow cytometer.
2.16. Estimation of reactive oxygen species in mitochondria For measuring mitochondrial ROS, exponentially growing cells were treated with mitocurcumin and cultured for 4 h. MitoSOX Red dye was added to the cells 30 min prior to harvesting and fluorescence was measured at 580 nm using a spectrofluorimeter (Synergy H1 Hybrid, Biotek, VT, USA) and cells were also visualized under confocal laser scanning microscope (Fluoview FV10i, Olympus, Japan).
2.12. Measurement of mitochondrial membrane potential (MMP) JC-1 exists as an oligomer that emits red fluorescence at normal MMP and gets converted to monomer upon loss of MMP and emits green fluorescence. Ratio of red to green fluorescence is an indicator of MMP. A549 cells were treated with mitocurcumin, cultured for 24 h and change in MMP was calculated by staining with JC-1 dye as described previously using a spectrofluorimeter [23]. The cells were also visualized under confocal laser scanning microscope to monitor change in MMP.
2.17. Quantification of mitochondrial GSH levels A549 cells were grown on a cover glass, treated with mitocurcumin (2.5 µM), cultured for 24 h and stained with CMFDA and MitoTracker Red for 30 min at 37 °C prior to imaging on a confocal microscope. CMFDA is a thiol reactive cell permeable fluorescent dye (green) that can detect glutathione specifically (> 95%) [29]. Mitotracker Red is used to track mitochondria in cells (red fluorescence). Co-localization of green and red fluorescence was used to quantify mitochondrial GSH levels.
2.13. In Silico docking studies In silico docking was performed to study possible interaction of mitocurcumin and curcumin with the active site of thioredoxin reductase (TrxR). Crystal structure of mammalian thioredoxin reductase (Rattus norvegicus) was obtained from protein data bank (PDB ID: 1H6V). Theoretical structure of human mitochondrial thioredoxin reductase in complex with thioredoxin was obtained from protein data bank (PDB ID: 1W1E). Crystal structure of human thioredoxin reductase 1 in complex with thioredoxin was obtained from protein data bank (PDB ID: 3QFA). 2D structure of curcumin and mitocurcumin were drawn using ChemSketch free version. DG-AMMOS was employed to generate the corresponding 3D structures and select the best energy minimized conformer. The ligand and receptor molecules were preprocessed using Autodock tools. Docking was carried out at the active site of TrxR using Autodock Vina. Rat TrxR has almost two identical active sites: one centered at C59-C64 of E chain and the other one at C59-C64 of F chain. They are referred to as E-chain active site and Fchain active site respectively. Human mitochondrial TrxR also has similar active sites: centered at C61-C66 of A chain and C61-C66 of B chain respectively. Human cytosolic TrxR has two active sites: one centered at C59-C64 of A chain and the other one at C59-C64 of B chain. They are referred to as A-chain active site and B-chain active site respectively. Interactions between TrxR and the docked ligands were studied using Ligplot plus software.
2.18. DNA damage quantification using real time qPCR Quantification of DNA damage using real time qPCR is based on the simultaneous amplification of a long region (3000–4000 bp) and a small region (~ 100 bp) of mitochondrial DNA and nuclear DNA. If the frequency of lesions is high in the DNA template, the PCR amplification of that region will be hindered, and they will have high threshold cycle value as compared to undamaged template in real time qPCR. Hence, DNA damage was quantified in mitochondria and nucleus based on the threshold cycle values. Total DNA was isolated from the cultured A549 cells treated with vehicle or mitocurcumin. The DNA quantity was measured using nano-spectrophotometer. DNA damage was quantified by qPCR as described by Stephen et al. [30]. 2.19. Overexpression and knockout studies CRISPR-Cas9 based system was used for knocking out or overexpression of TrxR2 in A549 cells (Santacruz, MD, USA). Transfection was performed using Neon® Transfection System (ThermoFisher Scientific, MA, USA) following manufacturer's protocol. The expression of TrxR2 level was confirmed by performing intracellular antibody 532
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
staining with TrxR2 antibody followed by fluorochrome conjugated secondary antibody.
3.3. Mitocurcumin inhibited mitochondrial thioredoxin reductase activity To understand the molecular mechanism underlying mitocurcumin induced apoptosis in lung cancer cells, we have studied the effect of mitocurcumin on thioredoxin reductase. The free energy values obtained by in silico docking studies revealed a high affinity binding of mitocurcumin at the active site of TrxR2 (Fig. 3A). Strong interactions were observed at E-chain, F-chain, A-chain and B-chain active sites (Table 1). Further, mitocurcumin inhibited the TrxR activity in a dose dependent manner in cell free system, and its IC50 value was around 10 µM (Fig. 3B). Corroborating with these findings, mitocurcumin at 2.5 µM inhibited cellular TrxR activity by more than 50% (Fig. 3C). Mitocurcumin induced a significant and dose dependent increase in the ROS levels in mitochondria as evinced from increased MitoSOX red fluorescence (Fig. 3D). Apart from increase in ROS, mitochondrial GSH levels were also decreased by mitocurcumin as seen from co-localization of CMF fluorescence and MitoTracker Red fluorescence (Fig. 3E). These results suggest that treatment of lung cancer cells with mitocurcumin disrupted mitochondrial redox balance leading to increased oxidative stress. Since, increased oxidative stress is known to induce DNA damage, we have examined the extent of mitochondrial DNA damage after mitocurcumin treatment. Mitocurcumin treatment (2.5 µM) induced mitochondrial DNA damage to the extent of 5 breaks per 10 Kb of DNA (Fig. 3F), whereas no significant nuclear DNA damage was observed. These results indicate that mitocurcumin inhibited mitochondrial TrxR leading to oxidative stress which is evinced by increased mitochondrial DNA damage.
2.20. Statistical analysis GraphPad Prism software was used to perform statistical analysis (La Jolla, USA). To calculate the significance between the mean of two groups t-test was used. If there were more than two groups for comparison, one-way ANOVA was performed followed by Tukey-Kramer multiple comparisons post-test. Difference between means was considered significant if p < 0.05. 3. Results 3.1. Mitocurcumin induces cell death in lung cancer cells and reduces the frequency of cancer stem cells The efficacy of mitocurcumin in inducing cell death in A549 cells was evaluated using clonogenic survival assay as compared to its parent molecule, curcumin. Mitocurcumin treatment resulted in significant decrease in survival fraction at 0.1 µM and complete abrogation of clonogenic potential was observed at 1 μM (Fig. 1A and B). On the contrary, curcumin showed a significant decrease in survival fraction at 25 µM. Curcumin at 50 µM completely suppressed clonogenic survival indicating that mitocurcumin is 50-fold more effective than curcumin (Fig. 1A and B) in inhibiting colony forming ability. Further, cell death was estimated by enumeration of cells with sub-G1 DNA content using PI assay. Mitocurcumin at 0.5 µM induced significant apoptosis and 80% of the cells were dead at 5 µM. On the other hand, curcumin at 25 μM induced significant apoptosis and 35% of the cells were dead at 50 µM suggesting the better efficacy of mitocurcumin in killing lung cancer cells (Fig. 1C and D). Similar results were obtained using MTT assay (Fig. S1A). These results were further confirmed by live and dead assay (Fig. 1E). Apart from A549 cells, effect of mitocurcumin was also evaluated in prostate cancer cells (PC-3 cells). Mitocurcumin at 2.5 µM, inhibited the colony forming ability of these cells and induced significant apoptosis indicating that it can act like an anticancer agent in different tumor types (Fig. S1 B–E). Further, the effect of mitocurcumin on normal cells was studied in primary mouse embryonic fibroblast cells (NIH 3T3 cells) and mouse lymphocytes. As compared to curcumin, mitocurcumin showed little toxicity to NIH 3T3 and lymphocytes as indicated by clonogenic survival, live and dead assay and apoptosis (Fig. S2 A–D; Fig. S3 A and B). Since mitocurcumin induced cytotoxicity in cancer cells, its effect on cancer stem cells was also investigated. Interestingly, treatment of lung cancer cells with 2.5 μM mitocurcumin led to significant decrease in the frequency of cancer stem cells as compared to vehicle treated group as demonstrated by a decrease in the side population and percent CD44+CD24- cells (Fig. 1F). These results infer that mitocurcumin induces cell death in lung cancer cells and also decreases the frequency of cancer stem cells.
3.4. TrxR2 overexpression accentuated mitocurcumin mediated apoptosis in cancer cells To confirm the role of TrxR2 in mitocurcumin mediated cell death, we have overexpressed TrxR2 in A549 cells using a CRISPR-Cas9 based activation plasmid (Fig. S3C). Surprisingly, TrxR2 overexpressing cells treated with mitocurcumin showed higher cell death as compared to cells transfected with scramble plasmid and treated with mitocurcumin (Fig. 4A). This was further confirmed by increase in sub-G1 population (Fig. 4B). Mitocurcumin mediated cytochrome-C release was also enhanced in the cells overexpressing TrxR2, as compared to cells transfected with scramble plasmid and treated with mitocurcumin (Fig. 4C). TrxR2 overexpression also accentuated the mitocurcumin mediated mitochondrial ROS accumulation (Fig. 4D). TrxR2 overexpression resulted in further decrease in the frequency of cancer stem cells following mitocurcumin treatment (Fig. 4E). In a cell free system when mitocurcumin was added to TrxR, mitocurcumin modulated its activity to NADPH oxidase like activity (Fig. 4 F). Further, in TrxR2 knock out A549 cells (Fig. S3 D), the ability of mitocurcumin to induce apoptosis was significantly reduced as compared to scramble control cells treated with mitocurcumin (Fig. 4G and H). Pretreatment with anti-oxidants glutathione or N-acetyl-cysteine also resulted in significant reduction in mitocurcumin mediated apoptosis in A549 cells. However, PEGylated catalase failed to abrogate mitocurcumin mediated apoptosis in A549 cells (Fig. S4A and B).
3.2. Mitocurcumin activated intrinsic apoptotic pathway in A549 cells Mitocurcumin treatment resulted in significant increase in the frequency of annexin-V-FITC positive cells in a dose dependent manner (Fig. 2A). The ratio between BAX (pro-apoptotic) to BCL2 (anti-apoptotic) proteins is an important indicator of mitochondria mediated apoptosis. Treatment of A549 cells with mitocurcumin led to significant increase in BAX levels and a concomitant decrease in BCL2 levels as estimated by immunostaining (Fig. 2B). Further, mitocurcumin treatment significantly enhanced the cytosolic cytochrome C levels and caspase 3 activity in lung cancer cells (Figs. 2C and 2D). Mitocurcumin treatment induced a significant loss of MMP as estimated by red to green fluorescence ratio after JC-1 staining (Figs. 2E and 2F). Together these results imply that mitocurcumin activated the mitochondrial apoptotic pathway.
4. Discussion In the search for new cancer treatment options, discovery of compounds which specifically target mitochondria may provide unique chemotherapeutic strategies. Mitocurcumin, which has been shown to accumulate in mitochondria, is one such agent. However, its anticancer effects and mechanisms of action are not well studied. Mitocurcumin showed 25–50 fold better efficacy than curcumin in killing lung cancer cells. Cancer stem cells pose a major challenge in achieving higher therapeutic efficacy and hence the effect of mitocurcumin on cancer stem cells was evaluated. Mitocurcumin treatment significantly reduced the frequency of cancer stem cells as evinced by side population assay 533
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Fig. 1. Mitocurcumin exhibits better anti-cancer efficacy than curcumin. Anticancer effect of mitocurcumin (MC) and curcumin (Cur) on A549 cells was assessed by clonogenic assay. The graph shows survival fraction (A), and representative images of the dishes containing colonies are shown in (B). The cells were treated with mitocurcumin, cultured for 48 h, stained with PI solution and acquired using a high throughput cell analyzer. The graph shows percent apoptotic (Sub G1) cells (C), and representative images of wells show live population in yellow and dead population in red (D). Live and dead assay was performed to study the anti-tumor effect of mitocurcumin. The cells were grown on a coverslip and were treated with mitocurcumin, cultured for 24 h, and stained with calcein-AM and ethidium homodimer-1. Green cells represent the live population, and red cells represent dead population (E). Effect of mitocurcumin on cancer stem cells was studied by side population assay and antibody staining (F). For enumerating the frequency of cancer stem cells, side population assay (left panel) and CD24, CD44 antibody staining (right panel) were performed 24 h after mitocurcumin treatment. Graphed are the frequency of cancer stem cells and overlaid dot plots are shown in inset (F). Each data point in the graphs represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to respective vehicle control (DMSO).
534
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Fig. 2. Mitocurcumin induces cell death through induction of apoptosis in lung cancer cells. Treatment of A549 cells with mitocurcumin (24 h) led to the induction of apoptosis as shown by annexin-V-FITC staining (A). Overlaid representative flow cytometric histograms are shown in inset (A). Mitocurcumin (MC) treatment led to change in levels of BAX and BCL2 as estimated by intracellular antibody staining and flow cytometry (B), cytosolic cytochrome-C levels (representative flow cytometric histograms are shown in inset, C), caspase-3 activity (D) and mitochondrial membrane potential as studied by JC-1 red / green fluorescence ratio (E). Red and green fluorescence of JC-1 probe was also visualized under confocal laser scanning microscope and representative mages are shown in (F). Each data point in the graphs represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to respective DMSO control group.
Fig. 3. Mitocurcumin treatment inhibits TrxR2 activity and alters mitochondrial redox. Binding affinity between TrxR2 and mitocurcumin (MC) was studied using in silico docking analysis. The site of docking is shown in (A). For in silico docking studies, binding affinity between TrxR2 and mitocurcumin was studied using Autodock Vina tool. Effect of mitocurcumin on the TrxR activity in cell free system was determined using DTNB reduction method after pre-incubating purified rat TrxR with mitocurcumin (B). Effect of mitocurcumin on TrxR activity in the cellular system was studied using micro-method of insulin reduction (C). Effect of mitocurcumin on mitochondrial ROS was studied by MitoSOX Red staining followed by spectrofluorimetry (D) and was also visualized by confocal laser scanning microscopy. Inset: Blue color represents nucleus and red color indicates mitochondrial ROS (D). Effect of mitocurcumin on mitochondrial GSH was studied by confocal laser scanning microscopy. Mitochondria was visualized using MitoTracker Red dye (Red fluorescence), the nucleus was visualized using DAPI dye (blue fluorescence) and mitochondrial GSH level was estimated from CMFDA green fluorescence. Representative images of cells are shown in (E). Effect of mitocurcumin treatment on mitochondrial and nuclear DNA damage was studied using real time qPCR technique and is graphed in (F). Each data point in the figures represents mean ± S.E.M. from three replicates and three such independent experiments were carried out. *p < 0.05 in comparison to DMSO control group.
chemotherapeutic drugs disrupt mitochondrial redox thereby inducing apoptosis in cancer cells. However, overexpression of TrxR2 results in effective neutralization of excess mitochondrial ROS induced by drugs and hence prevent ROS induced cell death. It is well known that overexpression of TrxR2 in cancer cells can adversely affect the
as well as surface marker analysis. TrxR2 is overexpressed in many cancers including lung cancer [11]. Overexpression of thioredoxin system contributes to uncontrolled proliferation, angiogenesis, resistance to cell death and protects the cancer cells from stress dysregulated redox signaling [31,32]. Many 535
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
treatment outcome and inhibition of TrxR2 can be a good strategy to kill cancer cells [32]. In silico studies revealed a very high affinity binding of mitocurcumin with TrxR2. Hydrophobic interactions played a major role in stabilizing mitocurcumin in the cleft at the active site of human mitochondrial TrxR2. Tyr406, Lys31, Lys407, Lys481 stabilized the triphenyl rings at one end of the ligand while the other end was stabilized by Val62, Thr60, Gly23, Gly25 and Gly59. The critical Cys497 and SelenoCys498 residues were located close to the backbone of mitocurcumin in the docked position (Fig. S4C). Experiments in cell free system showed that mitocurcumin indeed inhibited TrxR activity by physical interaction. In agreement with these findings, mitocurcumin also inhibited cellular TrxR activity. TrxR2 inhibition can lead to higher accumulation of ROS and thereby induce oxidative stress in cells. Our experiments confirmed that treatment of
Table 1 Docking energy values of binding of mitocurcumin at TrxR active sites. Docking energy values for binding at active sites of rat TrxR (Kcal/mol) Molecule ΔG at E-chain active site ΔG at F-chain active site Curcumin -7.3 -7.5 Mitocurcumin -8.2 -8.8 Docking energy values for binding at active sites of human mitochondrial TrxR (Kcal/ mol) Molecule ΔG at A-chain active site ΔG at B-chain active site Curcumin -7.6 -7.1 Mitocurcumin -8.4 -7.1 Docking energy values for binding at active sites of human cytosolic TrxR (Kcal/mol) Molecule ΔG at A-chain active site ΔG at B-chain active site Curcumin -6.7 -6.8 Mitocurcumin -7.8 -8.4
Fig. 4. Effect of mitocurcumin on TrxR2 overexpressing and TrxR2 knockdown cells. A549 cells overexpressing TrxR2 were treated with mitocurcumin (MC) for 24 h and apoptosis was studied by annexin-V-FITC staining followed by flow cytometry. Graphed is the percent apoptotic cells and representative flow cytometric histograms are shown in inset (A). Percent apoptotic cells were also studied by analyzing the cells with sub-G1 DNA content using a high throughput cell analyzer and are shown in (B). Cytosolic cytochrome C levels were estimated by intracellular antibody staining and flow cytometry. Graphed are the percent cells containing cytosolic cytochrome C and overlaid flow cytometric histograms are shown in inset (C). The effect of TrxR2 overexpression and mitocurcumin treatment on mitochondrial ROS was estimated by MitoSOX red staining and is graphed in (D). The frequency of cancer stem cells in these treatment groups was assessed by side population assay using high throughput cell analyzer and is shown in (E). Oxidation of NADPH by TrxR in the presence of mitocurcumin was studied in cell free system. The relative enzyme activity was calculated as the change in NADPH absorbance at 340 nm using a spectrophotometer and is shown in (F). The effect of mitocurcumin on TrxR2 knockout cells: TrxR2 knockout cells were generated using CRISPR-Cas9 plasmid and treated with mitocurcumin. Cells were cultured for 48 h and percent apoptosis was estimated by staining cells with annexin-V-FITC followed by acquisition on flow cytometer (G & H). Each data point in the figure represents mean ± S.E.M. from three replicates and two such independent experiments were carried out. *p < 0.05 as compared to DMSO control group. $p < 0.05 as compared to TrxR+NADPH+MC1 group. #p < 0.05 as compared to scramble + mitocurcumin group.
536
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al.
Scheme 1. Mechanism of action of anti-cancer activity of mitocurcumin.
lead to higher accumulation of ROS in the presence of mitocurcumin resulting in further increase in the magnitude of apoptosis. We propose that targeting TrxR2 by mitocurcumin can be very effective in eliminating cancer cells as well as cancer stem cells. In conclusion, mitocurcumin has been found to be a potent anticancer agent against lung cancer cells as compared to curcumin. Mitocurcumin acts by modifying TrxR2 activity, disturbing mitochondrial redox, inducing a loss in MMP and thereby triggering apoptosis. Hence, mitocurcumin can be a potent chemotherapeutic agent and may also find application as an adjuvant along with the existing chemotherapeutic drugs in cancers which overexpress TrxR2.
cells with mitocurcumin significantly enhanced mitochondrial ROS. Interestingly, pretreatment with GSH or NAC significantly inhibited mitocurcumin mediated apoptosis suggesting that ROS may be playing an important role in the anticancer activity of mitocurcumin. However, PEGylated catalase did not alter the anti-cancer effect of mitocurcumin. This could be due to poor localization of PEGylated catalase in the mitochondria to scavenge ROS generated by mitocurcumin. The level of total cellular thiols (protein and non-protein) and their reduced or oxidized state is an important determinant of cellular redox status. Our experimental findings revealed that mitocurcumin treatment significantly decreased mitochondrial GSH levels thereby disturbing mitochondrial redox balance. Increased oxidative stress can damage DNA bases and also induce strand breaks. It was observed that mitocurcumin treatment induced significant mitochondrial DNA damage. Excess mitochondrial ROS can trigger apoptosis through release of cytochrome C. These pro-apoptotic proteins are released through the transmembrane mitochondrial pore formed by oligomerization of BAX recruited on the mitochondria. An acute increase in the mitochondrial ROS is essential for oligomerization and recruitment of BAX and formation of pores [33]. In our study, when TrxR2 was modulated by mitocurcumin, it led to the accumulation of ROS, increase in BAX/BCL2 ratio, cytochrome C release and loss of MMP followed by apoptosis (Scheme 1). TrxR2 knockout cells were refractory to mitocurcumin induced apoptosis indicating that this molecule modulates Trx system in exhibiting anti-cancer activity. On the contrary, when TrxR2 was overexpressed using CRISPR-Cas9 plasmid in lung cancer cells, it led to a significant increase in the apoptosis induced by mitocurcumin. Previously, Fang et al. have reported that curcumin converted thioredoxin reductase to NADPH oxidase like activity [17]. We hypothesized that mitocurcumin might also modulate the TrxR2 activity to NADPH oxidase like activity by preventing the transfer of an electron to its substrate and thereby shunting the NADPH derived electron to oxygen leading to the production of superoxide. In agreement with this hypothesis, it was observed that mitocurcumin modulated TrxR2 activity to NADPH oxidase like activity. TrxR2 overexpression also augmented mitocurcumin induced apoptosis in cancer stem cells which could be due to further increase in ROS generation. Many studies have shown that Nrf-2 mediated antioxidant machinery is highly active in cancer stem cells [34,35]. Since TrxR2 expression is governed by Nrf-2, the de novo levels of TrxR2 may be higher in cancer stem cells. We found that overexpression of TrxR2
Acknowledgements The authors would like to thank Ms. Binita Kumar for assistance in flow cytometry. We also acknowledge the technical help provided by Mr. Deepak V. Kathole, Mr. B. A. Naidu, and Mr. K.S. Munankar. This work is funded by Department of Atomic Energy, Government of India (XII-N-R & D-17). Conflict of interest None Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed.2017.10. 378. References [1] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (2015) E359–E386. [2] H.J. Mehta, V. Patel, R.T. Sadikot, Curcumin and lung cancer–a review, Target. Oncol. 9 (2014) 295–310. [3] V. Gogvadze, Targeting mitochondria in fighting cancer, Curr. Pharm. Des. 17 (2011) 4034–4046. [4] E.S. Arner, A. Holmgren, Physiological functions of thioredoxin and thioredoxin reductase, Eur. J. Biochem. 267 (2000) 6102–6109. [5] A. Miranda-Vizuete, A.E. Damdimopoulos, G. Spyrou, The mitochondrial thioredoxin system, Antioxid. Redox Signal. 2 (2000) 801–810. [6] L.E. Netto, F. Antunes, The roles of peroxiredoxin and thioredoxin in hydrogen peroxide sensing and in signal transduction, Mol. Cells 39 (2016) 65–71.
537
Free Radical Biology and Medicine 113 (2017) 530–538
S. Jayakumar et al. [7] J.E. Biaglow, R.A. Miller, The thioredoxin reductase/thioredoxin system: novel redox targets for cancer therapy, Cancer Biol. Ther. 4 (2005) 6–13. [8] J.H. Choi, T.N. Kim, S. Kim, S.H. Baek, J.H. Kim, S.R. Lee, J.R. Kim, Overexpression of mitochondrial thioredoxin reductase and peroxiredoxin III in hepatocellular carcinomas, Anticancer Res. 22 (2002) 3331–3335. [9] A. Kunwar, S. Jayakumar, A.K. Srivastava, K.I. Priyadarsini, Dimethoxycurcumininduced cell death in human breast carcinoma MCF7 cells: evidence for pro-oxidant activity, mitochondrial dysfunction, and apoptosis, Arch. Toxicol. 86 (2012) 603–614. [10] J. Hellfritsch, J. Kirsch, M. Schneider, T. Fluege, M. Wortmann, J. Frijhoff, M. Dagnell, T. Fey, I. Esposito, P. Kölle, K. Pogoda, J.P.F. Angeli, I. Ingold, P. Kuhlencordt, A. Östman, U. Pohl, M. Conrad, H. Beck, Knockout of mitochondrial thioredoxin reductase stabilizes prolyl hydroxylase 2 and inhibits tumor growth and tumor-derived angiogenesis, Antioxid. Redox Signal. 22 (2015) 938–950. [11] L. Bu, W. Li, Z. Ming, J. Shi, P. Fang, S. Yang, Inhibition of TrxR2 suppressed NSCLC cell proliferation, metabolism and induced cell apoptosis through decreasing antioxidant activity, Life Sci. (2017). [12] M.R. Kim, H.S. Chang, B.H. Kim, S. Kim, S.H. Baek, J.H. Kim, S.R. Lee, J.R. Kim, Involvements of mitochondrial thioredoxin reductase (TrxR2) in cell proliferation, Biochem. Biophys. Res. Commun. 304 (2003) 119–124. [13] S. Vinogradov, X. Wei, Cancer stem cells and drug resistance: the potential of nanomedicine, Nanomedicine (London England) 7 (2012) 597–615. [14] S. Prasad, S.C. Gupta, A.K. Tyagi, B.B. Aggarwal, Curcumin, a component of golden spice: from bedside to bench and back, Biotechnol. Adv. 32 (2014) 1053–1064. [15] S.C. Gupta, S. Patchva, B.B. Aggarwal, Therapeutic roles of curcumin: lessons learned from clinical trials, AAPS J. 15 (2013) 195–218. [16] W. Cai, B. Zhang, D. Duan, J. Wu, J. Fang, Curcumin targeting the thioredoxin system elevates oxidative stress in HeLa cells, Toxicol. Appl. Pharmacol. 262 (2012) 341–348. [17] J. Fang, J. Lu, A. Holmgren, Thioredoxin reductase is irreversibly modified by curcumin: a novel molecular mechanism for its anticancer activity, J. Biol. Chem. 280 (2005) 25284–25290. [18] S. Prasad, A.K. Tyagi, B.B. Aggarwal, Recent developments in delivery, bioavailability, absorption and metabolism of curcumin: the golden pigment from golden spice, Cancer Res. Treat.: Off. J. Korean Cancer Assoc. 46 (2014) 2–18. [19] G. Shoba, D. Joy, T. Joseph, M. Majeed, R. Rajendran, P.S. Srinivas, Influence of piperine on the pharmacokinetics of curcumin in animals and human volunteers, Planta Med. 64 (1998) 353–356. [20] C.A. Reddy, V. Somepalli, T. Golakoti, A.K. Kanugula, S. Karnewar, K. Rajendiran, N. Vasagiri, S. Prabhakar, P. Kuppusamy, S. Kotamraju, V.K. Kutala, Mitochondrialtargeted curcuminoids: a strategy to enhance bioavailability and anticancer efficacy of curcumin, PLoS One 9 (2014) e89351. [21] S. Jayakumar, A. Kunwar, S.K. Sandur, B.N. Pandey, R.C. Chaubey, Differential response of DU145 and PC3 prostate cancer cells to ionizing radiation: role of
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29] [30]
[31] [32]
[33] [34]
[35]
538
reactive oxygen species, GSH and Nrf2 in radiosensitivity, Biochim. Et. Biophys. Acta 1840 (2014) 485–494. C. Riccardi, I. Nicoletti, Analysis of apoptosis by propidium iodide staining and flow cytometry, Nat. Protoc. 1 (2006) 1458–1461. S. Jayakumar, R.S. Patwardhan, D. Pal, D. Sharma, S.K. Sandur, Dimethoxycurcumin, a metabolically stable analogue of curcumin enhances the radiosensitivity of cancer cells: possible involvement of ROS and thioredoxin reductase, Biochem. Biophys. Res. Commun. 478 (2016) 446–454. H. Komuro, R. Saihara, M. Shinya, J. Takita, S. Kaneko, M. Kaneko, Y. Hayashi, Identification of side population cells (stem-like cell population) in pediatric solid tumor cell lines, J. Pediatr. Surg. 42 (2007) 2040–2045. D. Sharma, M.R. Eichelberg, J.D. Haag, A.L. Meilahn, M.J. Muelbl, K. Schell, B.M. Smits, M.N. Gould, Effective flow cytometric phenotyping of cells using minimal amounts of antibody, Biotechniques 53 (2012) 57–60. R.S. Patwardhan, R. Checker, D. Sharma, S.K. Sandur, K.B. Sainis, Involvement of ERK-Nrf-2 signaling in ionizing radiation induced cell death in normal and tumor cells, PLoS One 8 (2013) e65929. R.S. Patwardhan, D. Sharma, R. Checker, M. Thoh, S.K. Sandur, Spatio-temporal changes in glutathione and thioredoxin redox couples during ionizing radiationinduced oxidative stress regulate tumor radio-resistance, Free Radic. Res. (2015) 1–15. S. Kumar, A. Holmgren, Induction of thioredoxin, thioredoxin reductase and glutaredoxin activity in mouse skin by TPA, a calcium ionophore and other tumor promoters, Carcinogenesis 20 (1999) 1761–1767. D.W. Hedley, S. Chow, Evaluation of methods for measuring cellular glutathione content using flow cytometry, Cytometry 15 (1994) 349–358. S.O. Evans, M.B. Jameson, R.T. Cursons, L.M. Peters, S. Bird, G.M. Jacobson, Development of a qPCR method to measure mitochondrial and genomic DNA damage with application to chemotherapy-Induced DNA damage and cryopreserved cells, Biology 5 (2016). G. Powis, D.L. Kirkpatrick, Thioredoxin signaling as a target for cancer therapy, Curr. Opin. Pharmacol. 7 (2007) 392–397. E.E. Fink, S. Mannava, A. Bagati, A. Bianchi-Smiraglia, J.R. Nair, K. Moparthy, B.C. Lipchick, M. Drokov, A. Utley, J. Ross, L.P. Mendeleeva, V.G. Savchenko, K.P. Lee, M.A. Nikiforov, Mitochondrial thioredoxin reductase regulates major cytotoxicity pathways of proteasome inhibitors in multiple myeloma cells, Leukemia 30 (2016) 104–111. C. Fleury, B. Mignotte, J.L. Vayssiere, Mitochondrial reactive oxygen species in cell death signaling, Biochimie 84 (2002) 131–141. S. Bao, Q. Wu, R.E. McLendon, Y. Hao, Q. Shi, A.B. Hjelmeland, M.W. Dewhirst, D.D. Bigner, J.N. Rich, Glioma stem cells promote radioresistance by preferential activation of the DNA damage response, Nature 444 (2006) 756–760. S. Ding, C. Li, N. Cheng, X. Cui, X. Xu, G. Zhou, Redox regulation in cancer stem cells, Oxid. Med. Cell. Longev. 2015 (2015) 750798.