Reactive oxygen species generation and mitochondrial dysfunction in the apoptotic cell death of human myeloid leukemia HL-60 cells by a dietary compound withaferin A with concomitant protection by N-acetyl cysteine
- First Online:
- Cite this article as:
- Malik, F., Kumar, A., Bhushan, S. et al. Apoptosis (2007) 12: 2115. doi:10.1007/s10495-007-0129-x
- 683 Views
Induction of apoptosis in cancer cells has become the major focus of anti-cancer therapeutics development. WithaferinA, a major chemical constituent of Withania somnifera, reportedly shows cytotoxicity in a variety of tumor cell lines while its molecular mechanisms of action are not fully understood. We observed that withaferinA primarily induces oxidative stress in human leukemia HL-60 cells and in several other cancer cell lines. The withanolide induced early ROS generation and mitochondrial membrane potential (Δψmt) loss, which preceded release of cytochrome c, translocation of Bax to mitochondria and apoptosis inducing factor to cell nuclei. These events paralleled activation of caspases −9, −3 and PARP cleavage. WA also activated extrinsic pathway significantly as evidenced by time dependent increase in caspase-8 activity vis-à-vis TNFR-1 over expression. WA mediated decreased expression of Bid may be an important event for cross talk between intrinsic and extrinsic signaling. Furthermore, withaferinA inhibited DNA binding of NF-κB and caused nuclear cleavage of p65/Rel by activated caspase-3. N-acetyl-cysteine rescued all these events suggesting thereby a pro-oxidant effect of withaferinA. The results of our studies demonstrate that withaferinA induced early ROS generation and mitochondrial dysfunction in cancer cells trigger events responsible for mitochondrial -dependent and -independent apoptosis pathways.
Apoptosis inducing factor
Electrophoretic mobility shift assay
High Performance Liquid Chromatography
Nuclear factor κB
Reactive oxygen species
Tumor necrosis factor receptor
Dysregulation of apoptosis is the hallmark of all cancer cells and agents that activate programmed cell death could be valuable anticancer therapeutics . Most of the current anti-cancer drugs are derived from plant sources, which act through different pathways converging ultimately into activation of apoptosis in cancer cells leading to cell cytotoxicity . Anti-neoplastic agents therefore, act through several pathways in the death of cancer cells. Recent studies have amply documented that two major pathways are involved in the regulation of apoptosis . One pathway is mediated via cell surface death receptors, such as Fas/CD95 and TNFR1, which upon activation recruit cytoplasmic tail of the receptors and down stream associated signaling complex leading to the activation of caspase-8. The second pathway is mitochondrial-dependent, which is regulated by signaling cascade-involving members of Bcl-2 family. A loss of mitochondrial membrane potential (Δψm) brings about translocation of pro-apoptotic Bax to mitochondria and cytochrome c from mitochondria to cytosol resulting in caspase-9 activation . Members of Bcl-2 family therefore play a crucial role in the regulation of apoptosis. For instance, overexpression of anti-apoptotic Bcl-2 prevents the release of cytochrome c while overexpression of pro-apoptotic Bcl-2 member Bax facilitates the formation of mitochondrial pores and release of cytochrome c  after depolarisation of mitochondrial membranes . Many anti-cancer drugs would act as pro-oxidant, which target mitochondria  and may initially involve generation of free radicals such as reactive oxygen/nitrogen species [8, 9] eventually leading to the activation of apoptosis. Currently, natural plant based products are increasingly investigated for their cytotoxicity in cancer cells targeting apoptosis activation for the development of anti-cancer leads .
Withania somnifera plant has found extensive uses in the Indian traditional system of medicine and also as dietary supplement [10, 11]. It has also been reported for its tumor cell growth inhibitory activity, antitumor and radiosensitizing effect on transplantable mouse tumor . This herbal plant yields a host of steroidal lactones called withanolides, some of which have shown growth inhibition of human tumor cell lines . WithaferinA (WA) amongst these withanolides reportedly is very active in impairing metastasis and angiogenesis  while it was also shown to suppress nuclear factor-kB (NF-kB) activation and its regulated genes expression in cancer cells [15, 16]. Current studies further demonstrated that WA also acts like a proteosome inhibitor and that inhibition of proteosomal chymotrypsin-like activity may contribute to anti-tumor action in vivo . In an another recent study, WA has been reported to induce Par-4 selective apoptosis of prostrate cancer cells in both androgen responsive and androgen refractory prostrate cancer cells, and causes regression of PC-3 xenografts in nude mice . WA thus appears to exert at multi-targets proteins in the cancer cell. This is also observed from studies where WA interferes with actin cytoskeleton and disrupts F-actin organization via interaction with annexin II  and thus markedly limits the migratory and invasive capabilities of cancer cells. WA has also been proposed as a new generation molecule capable of eliciting growth inhibitory effect on cancer cells .
The aim of this study therefore is to further broaden and understand the molecular mechanisms of WA action on cancer cells cytotoxicity. A central point to alteration of plethora of proteins phenotypes in cancer cells by WA appeared to involve the role of free radicals in altering the redox balance of the cells. We first postulated that WA might be inhibiting cell proliferation by initiating oxidative stress through generation of reactive oxygen species (ROS) in cancer cells. We used human promyelocytic leukemia HL-60 cells as a model for investigating in details the analysis of oxidative stress mediated pathways involved in cancer cell killing because some anticancer therapies are known to mediate apoptosis through oxidative stress within the cancer cells . Moreover, some anticancer therapies may add to the oxidative stress within cancer cells. For instance, the chemotherapeutic agents doxorubicin, mitomycin C, etoposide and cisplatin are superoxide -generating agents . We here describe for the first time that WA alters the redox potential of cells by inducing oxidative stress and demonstrate that initial events involve generation of ROS and loss of mitochondrial membrane potential. Because WA appeared to elicit a pro-oxidant effect, we used antioxidants ascorbate, trolox and N-acetyl cysteine (NAC) as the ROS scavenger to rescue cells from oxidative stress. Amongst these only NAC was found highly effective against WA induced cytotoxicity. WA was found to induce early ROS formation, disrupt mitochondrial membrane functions, translocate cytochrome c, apoptosis inducing factor (AIF) and Bax with concomitant activation of caspases leading to cleavage of NF-kB and PARP, all of which were rescued efficiently by N-acetyl cysteine. Our studies have provided a deeper insight into one of the mechanisms of action of WA induced apoptosis in cancer cells. The observed apoptotic activity of WA is associated with ROS generation not only in HL-60 cells but also in other cancer cell lines too.
Materials and methods
Isolation and structural elucidation of withaferin A (5β, 6β-epoxy-4β, 27-dihydroxy-1-oxo-witha-2, 24 dienolide) from withania somnifera
Reagents and antibodies
RPMI-1640, 2′, 7′-dichlorofluoresceine diacetate (DCFH-DA), d, l-buthionine-S, R-sulfoximine (BSO), propidium iodide (PI), DNase-free RNase, proteinaseK, Hoechst-33258, 3-(4,5, -dimethylthiazole-2-yl)-2,5-diphenyltetra- zolium bromide (MTT), NAC, penicillin, streptomycin, l-glutamine, pyruvic acid, eukaryotic protease inhibitors cocktail and camptothecin were purchased from Sigma chemical Co. St. Louis. Fetal bovine serum was obtained from GIBCO Invitrogen Corporation (#16000-044, lot No. 1237517) USA. AnnexinV-FITC apoptosis detection kit and Cycle Test™ Plus DNA reagent Kit were from BD Biosciences while Apoalert caspases assay kits were from B.D. Clontech. Mouse anti-human antibodies to Bax (#SC20067), PARP-1 (#SC8007), Bcl-2 (SC7382), actin (#SC-8432), TNFR1 (#SC8436), Bid (#SC 6538) and goat anti-rabbit IgG-HRP (#SC2030) and goat anti-mouse IgG-HRP (#SC2031) were from Santa Cruz, USA. Rabbit anti-AIF (#PC536) was from Calbiochem, USA while mouse anti-NFkB (#554184, clone G96-337) and cytochrome c (#556433, clone 7H8.2C12) from BD, Pharmingin. Electrophoresis reagents, protein markers were from Bio-RAD, USA: Hyper film and ECL reagents from Amersham Biosciences, UK.
Cell culture, growth conditions and treatment
Human promyelocytic leukemia cells HL-60, acute lymphoblastic leukemia cell line Molt-4, prostrate carcinoma PC-3 and DU 145, T cell lymphoma HuT-78 and cervical carcinoma HeLa cells were obtained from National Cancer Institute (NCI), Bethesda, USA. The cells were grown in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), penicillin (100 units/ml), streptomycin (100 μg/ml), l-glutamine (0.3 mg/ml), pyruvic acid (0.11 mg/ml), and 0.37% NaHCO3 at 37°C in an atmosphere of 95% air and 5% CO2 with 98% humidity. WithaferinA was dissolved in dimethylsulfoxide and delivered to cell cultures in complete medium while the controls received only DMSO (<0.2%, v/v).
Assay of cell proliferation
The cells were plated in 96-well plates at a density of 2.5 × 104 cells/200 μl of medium. Cultures were incubated with different concentrations of WA for indicated time periods. The MTT assay was performed as described earlier . Cell growth was calculated by comparing the absorbance of treated versus untreated cells.
Flow cytometric analysis of apoptosis and necrosis
WA treated HL-60 cells were washed and resuspended in 100 μl of the binding buffer provided with the apoptosis detection kit (BD Pharmingin). Cells were stained with annexinV-FITC antibody and PI as per the instructions given by the manufacturer, and scanned for fluorescence intensity in FL-1 (FITC) and FL-2 (PI) channels.
TUNEL assay for detection of DNA fragmentation
DNA fragmentation is a late event because of interplay of large number of molecules involved in signaling cascades in apoptotic cell death in contrast to the annexinV binding assay of early apoptosis. DNA strand breaks were evaluated by labeling 3′-hydroxyl (OH) termini of double stranded or single stranded DNA breaks employing instructions as described in the Apo-Direct assay kit (BD Biosciences). The process involves end-labelling of DNA fragments with FITC-tagged deoxyuridine triphosphate nucleotide (FITC-dUTP). The preparations were analyzed for end-labeled DNA content using BD-LSR flow cytometer equipped with electronic doublet discrimination capability using blue (488 nm) excitation from argon laser. Data were collected in list mode on 10,000 events for FL1 fluorescence intensity where an increase in fluorescence intensity indicated apoptotic cell fraction.
Hoechst 33258 staining of cells for nuclear morphology
WA treated HL-60 cells (2 × 106 cells/3 ml) were washed twice with PBS, fixed and stained with Hoechst 33258 as described earlier . The slides were observed for any nuclear morphological alterations and apoptotic bodies under inverted fluorescence microscope.
DNA content and cell cycle phase distribution
Cells were treated with WA, collected, washed in PBS, fixed in 70% cold ethanol and placed at −20°C overnight. Cells were washed with PBS, subjected to proteinase-K and RNase digestion followed by staining of clean nuclear materials (nuclei) with propidium iodide using procedures and reagents as described in the instruction manual of the Cycle Test plus DNA reagent kit (Becton Dickinson, USA). The preparations were analyzed for DNA content using BD-LSR flow cytometer. Data were collected in list mode on 10,000 events for FL2-A versus FL2-W.
Flow cytometric analysis of reactive oxygen species (ROS)
Influence of WA on the endogenous generation of reactive oxygen species was measured with DCFH-DA probe as described earlier .
Measurement of mitochondrial membrane potential
Mitochondrial membrane potential (ΔΨm) was measured by using a Mitochondrial Membrane Sensor Kit containing JC-1, as described by the manufacturer (BD Bioscience, CA). Briefly, cells after treatment were washed twice with PBS and centrifuged at 300g at 4°C for 5 min. Each cell pellet was suspended in 1 ml of diluted BD Mito-Sensor reagent and incubated at 37°C/5% CO2 for 15 min. The cells were washed and suspended in 1 ml incubation buffer and analyzed by Flow cytometry for FL-1 versus FL-2 fluorescence.
Cells (2 × 106/2 ml/well, 6-well plate) were incubated with WA for the indicated time periods. At the end of treatment cells were washed in PBS and pellet lysed in cell lysis buffer. Activities of caspase-3, -8 and -9 in the cell lysates were determined fluorometrically using BD Apoalert caspase fluorescent assay kits. Caspase-3 and -8 employed fluorochome conjugated peptides DEVD-AFC and IETD-AFC as substrates, respectively while caspase-9 employed LEHD-AMC. Release of AFC (7-amino-4-trifluoromethyl coumarin) and AMC (7-aminomethylcoumarin) were assayed according to the instructions provided in the Manual by the supplier. Specific inhibitors were used as negative control to determine whether fluorescence intensity changes were specific for the activity of caspases. The peptide based inhibitors used were DEVD-CHO for caspase-3, IETD-fmk for caspase-8 and LEHD-CHO for caspase-9.
Measurement of GSH contents in Cells
Intracellular levels of GSH were estimated using the fluorescent reagent ortho-phthalaldehyde (OPT) . Briefly HL-60 cells (1 × 106/ml) were treated with WA (4 μM) along with and without NAC (5 Mm) for different time periods. GSH was measured fluorometrically at excitation and emission wavelengths of 350 nm and 420 nm, respectively.
Preparation of cytosolic and mitochondrial lysates of HL-60 cells
Cells were collected and washed twice with PBS. The cytosolic and mitochondrial fractions were obtained after selective plasma membrane permeabilization with digitonin . The cell lysates were transferred to fresh tubes and stored at −80°C for immunoblotting of proteins.
Preparation of total cell lysates for expression of NF-kB, PARP and Bcl2
HL-60 cells (3 × 106) after treatment with WA were harvested and resuspended in 0.2 ml of RIPA buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 5 mM EDTA, 30 mM Na2HPO4, 50 mM NaF, 0.5 mM NaVO4, 2 mM phenylmethylsulfonyl fluoride, and 10% protease cocktail inhibitor). Cells were incubated on ice for 30 min, vortexed and centrifuged at 12,000g for 15 min. Supernatants were collected and stored at −80°C .
Preparation of cytosolic and nuclear extracts
HL-60 cells (5 × 106) were washed with ice-cold phosphate-buffered saline after WA treatment and centrifuged. All steps of fractionation were carried out at 4°C . Cell pellets were homogenized in 200 μl of buffer A (10 mM Hepes, pH 7.9, 1 mM EDTA, 1 mM EGTA, 100 mM KCl, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 5 mM NaF, 1 mM NaVO4 and 10% protease cocktail inhibitor). The tubes were placed in ice for 10 min. Nonidet P-40 was added (0.5%, v/v), tubes vortexed briefly and centrifuged at 8,000g for 15 min. The cytosolic supernatants were stored at −80°C. The pellets obtained were resuspended in 50 μl of buffer A supplemented with 20% glycerol, 0.4 M KCl, kept on ice for 30 min and centrifuged at 13,000g for 15 min. The supernatants were stored at −80°C for analysis of nuclear NF-kB and AIF.
Electrophoretic mobility shift assay (EMSA) for NF-κB activation
To determine NF-κB activation, EMSA was conducted essentially as described . Briefly, nuclear extracts prepared from WA treated and untreated HL60 and HUT-78 cells were incubated with 32P-end-labeled 22-mer (5′AGTTGAGGGGACTTTCCCAGCC-3′), underlining indicates NF-κB binding site) double-stranded NF-κB oligonucleotide (Promega, USA). The incubation of the protein lysate with the oligomer (8 μg of protein with 10 fmoles DNA) was carried out for 30 min at 37°C. The DNA–protein complex formed was separated from the free oligonucleotide on 6.6% native polyacrylamide gels. The specificity of the binding was examined by competition with unlabelled oligonucleotide. The dried gels were visualized, and the radioactive bands quantitated by Phosphor Imager (Bio-RAD, USA) using Quantity One software.
Western blot analysis
Proteins from the mitochondrial, nuclear and cytosolic lysates were analyzed on SDS-PAGE. The resolved proteins were electro transferred to polyvinylidene difluoride (PVDF) membranes (Bio-RAD) over night at 30 V, 4°C. The membranes were blocked in blocking buffer (10 mM Tris–HCl, 150 mM NaCl, 0.1% Tween-20) containing 5% milk for 1 h and blotted with respective mouse anti- human primary antibodies for 2 h. Blots were washed in TBS and incubated with horseradish peroxidase-conjugated secondary antibody. Protein bands were detected using enhanced chemiluminescence’s reagent (ECL kit, Amersham Biosciences). The density of the bands was arbitrarily quantified using Quantity One software of Bio-RAD gel documentation system. The protein contents were determined using Bradford reagent (Bio-Rad protein assay kit) and aliquots normalized to equal quantities before loading.
Data are presented as mean ± S.D. of the number of experiments indicated. The comparisons were made with ‘t’ test and the difference was considered to be statistically significant if the P value was <0.05.
Purity of WA
The purity of WA used in the present study was almost 100% as confirmed by HPLC (Fig. 1).
WA inhibits cancer cell proliferation
WA induces apoptosis in HL-60 cells
HL-60 cells were incubated with different concentrations of WA for 12 h, and the percentages of cells undergoing apoptosis/necrosis were determined by staining with annexinV-FITC and PI (Fig. 2B). WA produced concentration dependent increase in apoptosis, which was 34, 49 and 70% at 2, 4, and 10 μM, respectively. The PI positive post-apoptotic/necrotic cell population however, was relatively small suggesting that WA induced cytotoxicity is predominantly through apoptotic pathways (Fig. 2B).
WA induces DNA fragmentation measured by TUNEL assay
To further verify the extent of apoptosis induced by WA in HL-60 cells, we applied a single-step staining method for labeling DNA strand breaks with FITC-dUTP to detect apoptotic cells by flow cytometry. It may be mentioned that in contrast to phosphatidylserine translocation measuring the immediate early onset of apoptosis, the tunnel assay on the contrary represents late events of apoptosis as a result of engagement of several signaling cascades leading to DNA fragmentation. The TUNEL assay was therefore, performed in samples treated with different concentrations of WA for 12 h (Fig. 2C). The highest number of apoptotic cells with DNA strand breaks were detected as a fraction of FITC-positive cell population depicting concentration dependent increase in fluorescence intensity. These results bear close correspondence to the AnnexinV/PI positive cell populations. The studies again supplement our above claims that cell death by WA is by way of activation of apoptosis signaling pathways.
WA induces early generation of ROS in HL-60 cells
ROS generation is protected by NAC against WA induced oxidative stress
Because WA treatment led to the enhancement of ROS generation, it is possible that alterations in the cellular redox state could play a role in WA induced apoptosis. To examine this, the antioxidant agent’s ascorbate, trolox and NAC were used to counter the WA-induced ROS generation and its attendant consequent events. NAC is known as a thiol antioxidant and functions as both redox buffer and reactive oxygen intermediate scavenger . Our results clearly demonstrated that WA is able to generate a strong oxidative stress not only in HL-60 cells but also in other cancer cells and pretreatment with NAC resulted in marked protection against ROS generation (Fig. 3A, B) and hence WA induced apoptosis measured by annexinV binding (Fig. 2B). However, ascorbate and trolox used as ROS scavengers in WA treated cells did not show any significant protection (data not shown).
WA disrupts mitochondrial membrane potential (Δψm)
Loss of mitochondrial membrane potential is an early event elicited by WA
We asked if ROS generation is an early event in the induction of apoptosis. In fact, ROS generation appeared to parallel Δψm loss and onset of apoptosis happened to be a late subsequent event (Fig. 4B). The relationship between these three important critical events was determined in the gated cell population exposed to WA for different time periods and measured by flow cytometery. At 30 min there was a significant loss of Δψm and an increase in ROS generation while apoptotic cells were at basal minimal levels. However after 1 h, the extent of ROS formation and Δψm loss were almost similar and increased with time of exposure. A high degree of apoptotic population appeared only after 6 h when most of the cells had lost their Δψm (Fig. 4B). The increase in apoptotic population on the contrary was gradual and lower compared to ROS formation and Δψm loss.
N-acetyl cysteine protects WA induced cell damage
NAC protects WA induced cell cytotoxicity
HL-60 cells were treated with NAC before exposure to various concentrations of WA (Fig. 5A). Cell proliferation was assayed in terms of the mitochondrial reduction of MTT by viable cells. WA alone produced concentration dependent cytotoxicity, which was rescued, completely to untreated control cultures by NAC.
WA depleted GSH level in HL-60 cells is a late event
GSH level in the untreated cells is usually very high while in WA treated cells the GSH pool observed a gradual and time-related decline (Fig. 5B). WA produced time dependent decrease in GSH content with significant decline starting after 3 h of treatment when at this period cells were already overwhelmed with ROS formation. NAC, a potent scavanger of ROS enabled to protect cellular GSH from depletion by WA.
WA induced altered nuclear morphology is rescued by NAC
We further sought to examine whether the abrogation of intracellular ROS by NAC could rescue the WA induced apoptotic death. After treatment for 24 h with WA alone and along with NAC, the Hoechst-33258 stained cells were observed under fluorescence microscopy for nuclear morphology and apoptotic bodies (Fig. 5C). Marked morphological changes were observed in WA treated cells such as nuclear condensation, formation of apoptotic and scattered apoptotic bodies. However, WA exposed cultures pretreated with NAC exhibited features that were comparable to the untreated control cells. The gross morphological changes were also observed under light microscopy in WA treated PC-3 cells, which again were rescued by NAC (Fig. 5C).
NAC protects WA induced hypo-diploid sub-G0 DNA population in HL-60 cells
Induction of apoptosis through ROS and protection by antioxidant NAC in HL-60 cells was also confirmed by FACScan analysis of cell cycle phase distribution studies. The results of the cell cycle analysis of HL-60 cells treated with WA for 24 h resulted in increase in hypo-diploid sub-G0 DNA fraction. As Sub- G0 peak is reported to be a qualitative indicator of apoptosis , increase in this fraction therefore not only supports apoptotic mode of cell death by WA, but also that apoptosis is the result of ROS formed as the cells pretreated with NAC exhibited marked decline in Sub G0 population. Similar findings were also observed in Molt-4 and HeLa cells (Fig. 5D). The increase in annexin V binding is also protected by NAC as observed earlier (Fig. 2B).
WA causes early cleavage of PARP protein
PARP (poly (ADP ribose) polymerase), an enzyme involved in DNA repair, is a preferential substrate for caspase-3. We investigated PARP protein cleavage in WA treated HL-60 cells. WA treatment caused early cleavage of PARP, 116 kDa into 89 kDa fragment beginning in less than 3 h that was almost complete after 9 h. This corresponded with the activation of caspase-3. Again pre-treatment with NAC resulted in a complete protection against WA-induced PARP cleavage (Fig. 5E).
WA is a potent activator of caspases
Activation of caspase-3 by WA is influenced by activation of both capase-9 and -8 signaling
Mitochondrial membrane disruption by WA: Release of Cyt. c, AIF and translocation of Bax across outer mitochondrial membrane without affecting the expression of Bcl-2 (Fig. 8 A–G)
WA inhibits NF-kB activation by inducing cleavage in the p65/Rel subunit, and by inhibiting binding to DNA
We also performed immunoblot analysis of NF-κB expression in cytosolic and nuclear fractions as well as total cell lysate by employing monoclonal antibody which recognizes -COOH terminus epitope of p65/Rel. In total cell lysate, we observed for the first time that WA not only suppressed the expression of NF-κB-p65 but also caused its cleavage during 9–24 h exposure time which was protected completely in cells pre-treated with NAC (Fig. 9B). To understand whether WA induced NF-κB suppression and p65 cleavage occurs in cytosol or after translocation into the nucleus, we observed that NF-kB is cleaved only after its translocation into the nucleus (Fig. 9C) as a result of WA treatment of cells. The cleavage appeared to follow only at a much later stage after it had initially failed to bind DNA. Since NF-kB cleavage is proposed to be mediated by activated caspase-3, we investigated if WA induced caspase-3 is involved in the cleavage (Fig. 9D). For this purpose, HL-60 cells were pre-incubated for 3 h with caspase-3 inhibitor Z-DEVD-CHO before further treating with WA for 18 h. The cleavage of NF-kB was completely protected demonstrating that WA mediated activation of caspase-3 is responsible for NF-KB cleavage.
One of the goals of cancer chemotherapy is to explore and develop discovery leads that can selectively induce apoptosis in cancer cells . In this study we report for the first time a novel insight in deciphering the mechanisms involved in WA induced early events leading to the activation of signaling cascades culminating in apoptotic cancer cell death. Our results demonstrate that exposure of HL-60 cells to WA enabled apoptotic cell death as evidenced by apoptotic bodies formation, increased sub-Go hypo-diploid DNA fraction and enhanced FITC-labeled dUTP incorporation into the 3′-hydroxyl-DNA ends and annexin V binding of cells. However, ROS generation was overwhelmed during the early events of WA exposure before the onset of apoptosis of comparable magnitude suggesting thereby an early pro-oxidative environment triggered by WA. This exquisitely has enabled to induce disruption of mitochondrial function, with concurrent loss of mitochondrial membrane potential (Δψm). Occurrence of all these events was time dependent and subtle changes appeared within less than 3 h of exposure to WA.
The origin of WA induced ROS formation by WA in cancer cells could not be ascertained in the present studies. The withanolide may be inhibiting respiratory electron transport chain (ETC) delivering electrons to reduce molecular oxygen to form superoxide oxidants as have been observed with mitochondrial ETC inhibitors for complex-III and complex-I  or by activation of NADPH oxidases. The ETC complexes are the major sites of ROS generation in mitochondria. A blockade of these sites by WA may enable delivery of one electron to molecular oxygen and allow release of superoxide either on the cytoplasmic side or in the mitochondrial matrix resulting in Δψm loss. Nevertheless, the ROS formation by WA was strongly blocked by the strong antioxidant NAC.
We expected a significant fall in GSH pool during the early exposure of cells to WA because of the possibility of oxidation of thiol groups by ROS. However, the pool was not affected critically during this period when free radicals generation was at its maximum. The GSH depletion may thus not be the primary cause of cell death because cells exposed to WA are able to maintain the reducing environment due to GSH at least for 3 h contrary to ROS formation. This may be the reason that optimal activation of caspase-3 and caspase-9 was observed through 9 h and thereafter the activity started declining because caspases contain an active site cysteine nucleophile, which is prone to oxidation by ROS in fatally dying cells  when GSH levels are not sufficient enough to counter overwhelming ROS accumulated. Caspase-8 however, exhibited a continuing uprising trend with optimal activation at 18 h, which may be viewed as a late complimentary mechanism to support cell death.
The activation of two apoptosis initiators caspases-9 and -8 suggested at least two signaling cascades [36, 39] involved in the apoptotic cell death by WA. Caspase-9 activation by WA suggests engagement of mitochondrial signaling cascade as evidenced by early release of cytochrome c from permeabilized mitochondria to cytosol and simultaneous translocation of Bax to mitochondria. Cytochrome c released is known to bind Apaf-1 (apoptotic protease activating factor 1) in the cytosol to form a complex called apoptosome that recruits and binds pro-caspase-9 to release active caspase-9, which up-regulates down stream pathways leading to the activation of executioner caspse-3 . Increased level of cytochrome c in the cytosol and its corresponding decrease in mitochondria suggests that release of cytochrome c in fact is a ‘point of no return’ for cells to enter apoptosis.
WithaferinA also induced early translocation of Bax from cytosol to mitochondria consequent to disruption of mitochondrial membrane functions and Δψmt loss. The translocation of Bax may also aid and abet the oxidative burst leading to the release of Cyt c from mitochondrial inner membrane . In the cytosol of non-apoptotic cells Bax exists constitutively in inactive form and its activation by ROS mediated cytosolic sensors is often required to oligomerize and stably insert it in the outer mitochondrial membrane. This would enable the onset of MOMP, which is often associated with the loss of Δψmt. Therefore, the translocation of active Bax to mitochondria is a critical event in the discharge of cyt c and other pro-apoptotic molecules from inner mitochondrial space to cytosol. The anti-apoptotic protein Bcl-2 is reported to block the release of cytochrome c and MPT opening  by preventing ROS production. We however, could not observe any change in the expression of Bcl-2 though cells were overwhelmed with ROS when exposed to WA suggesting that Bcl-2 may function differently depending upon various stimuli. For instance, continued Bcl-2 expression was observed in EBV-transformed lymphoblastoid cells undergoing spontaneous apoptosis because of the inhibition of NF-kB in these cells .
The loss of Δψm by WA is a sign of mitochondrial swelling and disruption of outer mitochondrial membrane  with subsequent release of apoptosis inducing factors, i.e., AIF, cytochrome c, from ROS damaged mitochondria of WA treated cells. It is also recognized that translocation of AIF to nucleus is associated with increase in ROS formation and induction of apoptosis ; AIF thus released produces peripheral chromatin condensation and high molecular weight DNA fragmentation in the nucleus. WA induced apoptosis consequent to AIF translocation into the nuclei amounts to a similar situation where microinjection of recombinant AIF into the isolated nuclei or cells resulted in apoptotic phenotypes, and mitochondrial membrane potential loss .
Another mode of cytotoxicity by WA happened through activation of caspase-8 suggests involvement of extrinsic pathway of apoptosis . Because activation of caspase-8 activity over 18 h period of time corresponded with continuing increased expression of TNFR-1 on cell surface. ROS generation has reportedly been acclaimed to induce over-expression of cell surface death receptors TNFR-1/Fas  and WA induced ROS generation increased the activity of caspase-8 with simultaneous expression of TNFR-1. Activation of caspase-8 is also associated with the cleavage of Bid, an important pro-apoptotic member of the Bcl-2 family of proteins. The truncated product tBid translocates to mitochondria and is believed to induce permeabilization of the outer mitochondrial membrane. It is also believed to reorganize the inner mitochondrial membrane leading to rapid release of Cyt c and other molecules involved in apoptotic response. Our studies have shown that WA caused time dependent decrease in the expression of Bid which corresponded with increased enzymatic activity of caspase-8 and over expression of TNFR-1. The decreased expression of Bid may thus be related to its truncation, which forms a central point in the cross talk between caspase-8 and caspase-9 activation. These events clearly demonstrate that WA induced apoptotic cell death is the consequences of involvement of both mitochondrial and non-mitochondrial signaling cascades. The mitochondrial dependent caspase-9 activation by WA however, appeared predominant in regulating cell death during the early hours of exposure. This is evidenced from our studies where caspase-9 inhibitor blocked almost completely the WA activated caspase-3 activity while caspase-8 inhibitor exerted significant but relatively lesser effect after 9 h exposure of HL-60 cells to WA. Thus caspase-8 signaling pathway exerts an important augmenting effect on the major WA activated executioner caspase-9 pathway. Further, the caspases upon activation cleave numerous cellular proteins  of which poly (ADP-ribose) polymerase cleavage happened to be an early target of WA induced apoptotic onslaught. All these studies suggested that ROS production functions as a positive regulator of caspases activation. This is a classical example where a dietary product WA is able to mediate ROS production leading to apoptosis, though studies are needed to find precise mechanisms of ROS formation by WA in cancer cells.
Further more, NF-kB family of transcription factor plays a central role in regulation of apoptosis, oncogenesis, inflammatory and immune responses and is activated by a wide range of stimuli. The ability of NF-kB to inhibit apoptosis appears to be stronger than its ability to promote apoptosis , and therefore, inhibition of NF-kB is suggested to be a useful strategy for cancer therapy. WA in this regard exhibited its strong ability not only to block completely the binding of the transcription factor to DNA but also caused cleavage of nuclear NF-κB. Such an effect has also been observed earlier with other agents . The failure of NF-kB binding to DNA in the nucleus subjects this protein further to nuclear cleavage by activated caspase-3 observed in our studies thereby abrogating its complete functionality facilitating obliquely the pro-apoptotic machinery of cell death. In other words inhibition of NF-kB may have profound effect on transcription of several anti-apoptotic genes, anti-oxidant enzymes and early formation of ROS. This also suggests that down regulation of the factor can be used as a suitable molecular target for development of anticancer chemotherapeutic agents such as WA and its semi-synthetic analogs.
In conclusion our studies demonstrate that WA induced early ROS formation and mitochondrial dysfunctions are directly responsible for induction of apoptotic cell death. The withanolide pursued predominantly mitochondrial intrinsic signaling pathway by release of cytochrome c, and AIF from mitochondria and translocation of Bax from cytosol to mitochondria facilitating caspase-9 activation and up regulation of down stream pathways leading to caspase activation and PARP cleavage. WA also increased the activity of caspase-8 with simultaneous expression of TNFR-1 demonstrating activation of extrinsic signal cascade too. However, the intrinsic pathway through caspase-9 signaling exhibited predominance over extrinsic pathway. Further, WA induced decreased expression of Bid may be an important link in the cross talk between caspase-8 and -9 signaling pathways of apoptosis. Another interesting feature was that WA inhibited not only NF-κB binding of DNA but also caused its nuclear cleavage. All these events were reversed by ROS scavenger NAC suggesting that WA alters the redox balance of the cells. The results of these in-depth studies provide putative mechanism of action of WA mediated oxidative stress in cancer cells. The studies therefore raise the potential usefulness of WA as an anti-cancer therapeutic candidate present in the dietary supplement Withania somnifera.
We are highly grateful to Dr. M.S.Majumdar of Institute of Microbial Technology, Chandigarh, India for providing facilities to perform EMSA. Thanks are also to the Council of Scientific and Industrial Research, India, for financial support for senior research fellowoships to Fayaz Malik and Ajay Kumar.