Antiproliferative Effect of Β-escin — an in Vitro Study

This study examined the antiproliferative effects of β-escin (E) in cancer cells. The study showed that E inhibited cancer cells growth in a dose-dependent manner. The flow cytometric analysis revealed an escin-induced increase in the sub-G1 DNA content, which is considered to be a marker of apoptosis. Apoptosis was also confirmed by annexin V staining and DNA fragmentation assay. These effects were associated with increased generation of reactive oxygen species (ROS), caspase-3 activation and decreased mitochondrial membrane potential (MMP). Moreover, escin decreased mitochondrial protein content and mitochondrial fluorescence intensity as well as caused depletion of glutathione (GSH). However, activity of glutathione peroxidase (GPx) and glutathione reductase (GR) was not significantly changed in escin-treated cells. In conclusion, our results demonstrated that E has apoptotic effects in human cancer cells through the mechanisms involving mitochondrial perturbation. Although the exact mechanism needs to be investigated further, it can be concluded that E may be a useful candidate agent for cancer treatment.


INTRODUCTION
Horse chestnut (Aesculus hippocastanum) is one of the oldest and widely used medicinal plant since ancient times.For example, the extracts from the horse chestnut seeds have been traditionally used in China as a carminative, stomachic, and anti-inflammatory agent in treating varicosis, in general, and hemorrhoids, in particular (Sirtori, 2001;Pittler & Ernst, 2012;Suter et al., 2006;Wang et al., 2013).The primary active constituent found in horse chestnut seed extract is a triterpene saponin β-escin (Sirtori, 2001).A number of other compounds such as bioflavonoids (quercetin, kaemferol and their diglycosyl derivatives), and coumarins (aesculin and fraxin) have been isolated from the chestnut seeds (Sirtori, 2001;Bombardelli & Morazzoni, 1996;Patlolla et al., 2006).Escin was found to be effective as anti-inflammatory, antiviral, antiallergic, antioxidant and vasorelaxant agent (Küçükkurt et al., 2010;Lindner et al., 2010;Ćalić-Dragosavac et al., 2011).Moreover, ability of β-escin to suppress growth of cancer cells has also been documented by Guo et al. (2003) found that β-escin can inhibit the growth of various human and mice tumor cell lines and the tumors developed after their transplantation.Later, Niu and co-workers demonstrated that β-escin is a po-tent natural inhibitor of cell proliferation and inducer of apoptosis in K 562 chronic myeloid leukemia cells (Niu et al., 2008).Escin was also found to exhibit significant antitumor effects in human hepatocellular carcinoma both in vitro and in vivo (Zhou et al., 2009).Recently, Ji and coworkers (2011) found that the antiproliferative effect of escin may be related to inhibition of the JAK/ STAT signaling pathway.Furthermore, other signaling pathways may also be involved in anticancer effect of escin.It is accepted that nuclear factor-κB (NF-κB) signalling pathway plays an important role in physiological as well as pathological processes, including tumorigenesis (Shishodia & Aggarwal, 2004).In the study of Harikumar et al. (2010), Wang et al. (2012) and Rimmon et al. (2013) escin was found to downregulate NF-κB signaling pathway with subsequent induction of apoptosis or sensitization of cancer cells to standard chemotherapy.
Recently, we also demonstrated that extract of horse chestnut (HCE) possesses antiproliferative and antiangiogenic properties.We found that HCE inhibited growth and colony formation, and induced apoptosis in cancer cells.Moreover, HCE inhibited migration of endothelial cells as well as decreased the secretion of matrix metalloproteinase and vascular endothelial growth factor (Mojžišová et al., 2013).Because E is the main active component of horse chestnut, the aim of this study was to evaluate possible mechanism of its antiproliferative effect.
Cell lines and culture.The following human cancer cell lines were used for this study: Jurkat (human acute T-lymphoblastic leukaemia), CEM (acute T-lymphoblastic leukaemia), HeLa (cervical carcinoma cells), and MCF-7 (breast cancer cells).All cell lines used were purchased from American Type Culture Collection (ATCC).Jurkat, CEM and HeLa cells were cultured in RPMI 1640 medium.MCF-7 cells were cultured in growth medium consisting of high glucose Dulbecco's Modified Eagle Medium.Both media were supplemented with Glutamax and 10% fetal calf serum, penicillin (100 IU/mL), and streptomycin (100 μg/mL).Cells were kept at 37°C in a humidified atmosphere with 5% CO 2 .Cell viability, estimated by trypan blue exclusion, was greater than 95% before each experiment.

Growth inhibition assay. The antiproliferative
effect of E was studied by MTT assay (Mosman, 1983).The method is based on the conversion of tetrazolium dye MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide inside the cells to insoluble formazan.The stock solution of E was prepared in DMSO and diluted to a different concentration.The final concentration of DMSO in culture medium was 0.2%.
Briefly, cells were seeded on 96-well plates (Sarstedt, Nümbrecht, Germany) at density of 8 × 10 3 cells/well in 80 μL of the medium.After 24 h of adhesion, 20 μL aliquots of E (7.8-500.0μg/mL) were added.After 72 h of culturing, 10 μl of the MTT solution were added to each well and the cells were incubated for another 4 h.Then, the formazan crystals were dissolved with 100 μL of 10% SDS.The absorbance was measured at 540 nm using the automated uQuant™ Universal Microplate Spectrophotometer (Biotek, Winooski, VT, USA).The absorbance of the control wells was taken as 100% and the results were expressed as the percentage of the control (untreated cells).
xCELLigence cell analysis system.xCELLigence system is a unique, impedance-based system for cellbased assays, allowing for label-free and real-time monitoring of cellular processes such as cell growth, proliferation, cytotoxicity, adhesion, morphological dynamics and modulation of barrier function.It measures impedance changes in a meshwork of interdigitated gold microelectrodes located at the well bottom (E-plate) or at the bottom side of a microporous membrane (CIM16-plate).These changes are caused by the gradual increase of electrode surface occupation by (proliferated/ migrated/ invaded) cells during the course of time and thus can provide an index of cell viability, migration and invasion.This method of quantitation is directly proportional to cellular morphology, spreading, ruffling and adhesion quality as well as cell number (Ke et al., 2011).
The xCELLigence RTCA system was initialized, as per manufacturer's instructions, prior to commencement of the experiment by filling all 16 wells of the E-plate (ACEA Biosciences) with the growth medium (100 μL), and equilibrated at room temperature for 30 min.The plate was placed into the single plate (SP) station cradle (housed in a humidified incubator at 37°C with 5% CO 2 atmosphere) to establish the background reading.Then, HeLa cells were seeded in E-plates at density of 2 × 10 3 cells per well.After 24 h, escin was added at concentrations of 31.2 and 62.5 μg/ml and cells were allowed to grow for additional 72 h under label-free conditions.The electrical impedance was measured by the RTCAintegrated software of the xCELLigence system (ACEA Biosciences) as a dimensionless parameter termed CI.
Colony formation analysis.For colony formation assay, HeLa cells were seeded in six-well plates at density of 1 000 cells per well and allowed to adhere for 10 h before treatment.RPMI 1640 medium containing various concentrations of E was added to the cells and the plates were incubated for 14 days.The cells were then fixed in buffered formalin (pH 7.2), stained with 0.01% crystal violet and washed to remove excessive dye.Quantitative changes in clonogenicity were determined by extracting the colonies with 10% acetic acid for 60 minutes and measuring the absorbance of the extracted dye at 540 nm.Cell survival at each drug concentration was expressed as a percentage of survival in comparison to the controls (untreated cells).
Cell cycle analysis.Cell cycle distribution in cells treated with the tested agents was analysed by propidium iodide (PI) DNA staining.Briefly, 5 × 10 5 Jurkat cells were treated with tested compounds for 24, 48 and 72 h.After treatment, cells were harvested, washed twice in PBS and fixed overnight in 70% ethanol at 4ºC.Then, cells were washed with PBS, incubated for 1 h with 1 mg/mL RNase and 10 μg/mL PI at room temperature in the dark.After staining, samples were immediately analysed in FACS Canto flow cytometer using FACS Diva software (Becton Dickinson, USA).Ten thousand cells were acquired per sample.Data were analysed using Win MDI software and represented in the form of histograms.Percentages of cells corresponding to G 0 /G 1 , S and G 2 /M phases of the cell cycle were calculated.The cells with DNA content lower than that of G 1 -phase cells (hypoploid population) were considered to be apoptotic (sub-G 1 ).
Annexin V-FITC labelling.The plasma membrane changes characteristic of apoptosis were analysed by double staining with Annexin V-FITC and PI according to the manufacturer's instructions.Jurkat cells (1 × 10 5 ) were harvested 24, 48 and 72 h after treatment and stained with Annexin V-FITC (BD Biosciences Pharmingen, San Diego, CA, USA) in binding buffer for 15 min, washed, stained with PI for 5 min and thereafter analysed using BD FACS Calibur flow cytometer.Three populations of the cells were observed: viable cells: Annexin V-FITC negative and PI negative; apoptotic cells: Annexin V-FITC positive and PI negative; late apoptotic/necrotic cells: Annexin V-FITC positive and PI positive or Annexin V-FITC negative and PI positive.
DNA fragmentation assay.Treated and untreated Jurkat cells (for 24, 48 and 72 h; 1 × 10 6 ) were washed twice with phosphate-buffered saline w/o calcium and magnesium.Cells were lysed in a lysis buffer containing 10 mmol/L tris (hydroxymethyl)aminomethane, 10 mmol/L EDTA, 0.5% Triton X-100.Proteinase K (1 mg/mL) was added and cells were incubated at 37ºC for 1 h.Subsequently, 10 min incubation at 70ºC was performed.Then, RNase (200 mg/mL) was added and cells were incubated for another 1 h at 37ºC.Samples were transferred to 2% agarose gel and run at 40 V for 3 h.DNA fragments were visualized with UV illuminator after ethidium bromide staining.
Measurement of ROS.The intracellular production of ROS was detected with FCM analysis using dihy-Antiproliferative effect of β-escin drorhodamine-123 (DHR-123, Fluka, Buchs, Schwitzerland), which reacts with intracellular hydrogen peroxide.Jurkat cells treated with an appropriate agent were harvested, washed twice with PBS, and resuspended in PBS.DHR-123 was added at a final concentration of 0.2 μM.The samples were then incubated for 15 min in the dark and placed on ice after the incubation.Fluorescence was detected with 530/30 (FL-1) optical filter.Forward and side scatters were used to gate the viable populations of cells.
Detection of MMP.The changes in MMP were analysed with FCM using tetramethylrhodamine ethyl ester perchlorate (TMRE, Molecular Probes, Eugene, OR, USA).Jurkat cells were washed with PBS, resuspended in 0.1 μM of TMRE in PBS, and incubated for 30 min at room temperature in the dark.The cells were then washed twice with PBS, resuspended in 500 μL of the total volume, and analysed (1 × 10 4 cells per sample).Fluorescence was detected with 585/42 (FL-2) optical filter.
Detection of active caspase 3. The changes in caspase 3 activation were analysed with FCM using BD Pharmingen Active Caspase-3 PE MAb Apoptosis kit (BD Bioscience, San Diego, CA, USA).Jurkat cells were prepared according to manufacturer's guidelines, stained with PE conjugated antibody and incubated for 30 min at room temperature in the dark.The cells were then washed twice with PBS, resuspended in 500 μL of the total volume and analysed (1 × 10 4 cells per sample).Fluorescence was detected with 585/42 (FL-2) optical filter.
Mitochondria isolation, determination of proteins and enzymes activities.Mitochondria were isolated from Jurkat cells according the method of Mela & Seitz (1979).Cells were incubated for 72 hours with E (31.2 μg/mL).
The medium for mitochondria isolation pH = 7.2 consisted (in mol/L) of sucrose 0.25, EDTA 0.002, TRIS HCl 0.010.The content of proteins in isolated mitochondria was estimated with the method of Bradford (1976).
Fluorescence spectra measurements.The fluorescence spectra were analysed with Perkin Elmer Model LS 55 luminescence spectrometer using 1 cm pathlength quartz cuvettes at ambient temperature.The wavelength scan speed of both monochromators was 1200 nm/min.The setting for instrument's excitation slit was 10 nm and the one for the emission slit was 15 nm.
Immediately before use, the isolated mitochondria have been diluted 1:100 in respiration media containing 1 mM sodium succinate.The mitochondrial characteristic pattern in respiration medium was investigated by synchronous fluorescence fingerprint (SFF).
The synchronous spectra have been measured at the constant setting of difference between emission and excitation monochromators Dl = 10-200.The setting for excitation wavelengths was 200-400 nm.The contour lines of 3-D spectra were constituted from 20 (number of scans) simple spectral records of mitochondria and placed in space with increment (distance) of 10.Data processing was managed by the FL Winlab (Perkin-Elmer) software package.
Statistical analysis.Statistical data was expressed as mean ± standard deviation (S.D.).Student's t-test and analysis of variance were employed to determine statistical significance.Values of p < 0.05, p < 0.01 and p < 0.001 were considered to be statistically significant.

Effect of escin on cell proliferation
In this study we used a microculture assay based on metabolic reduction of MTT to evaluate the antiproliferative effect of E on four cancer cell lines.Survival of CEM, HeLa, Jurkat as well as MCF-7 cells exposed to various E concentrations is shown in Fig 1a .At the higher concentrations of E (62.5-500.0μg/mL), cell viability was supressed almost completely.At the concentration of 31.25 μg/mL, E caused 44.7%, 42.7%, 86.4%, and 41.7% reduction in cell survival (p < 0.001; p < 0.01).Moreover, E at the concentration of 15.6 mg/ mL significantly decreased Jurkat and MCF cells survival (p < 0.05; p < 0.01).IC 50 ranged from 21.2 to 42.9 μg/mL (Fig. 1b).
The antiproliferative effect of compound β-escin on HeLa cells was also evaluated by xCELLigence system.As shown in Fig. 2 cell index of HeLa cells decreased significantly after treatment with E in a concentrationdependent manner.

Escin inhibits colony formation
To analyze further the antiproliferative effects of E on anchorage-independent cell growth, we tested the ability of this compound to inhibit colony formation in a clonogenic assay.In accordance with MTT assay, E inhibited the clonogenic survival of HeLa cells during colony formation assay in dose depended manner (Fig. 3) (p < 0.001; p < 0.01; p < 0.05).

Effect of escin on cell cycle
To examine the potential mechanism through which E suppresses cell proliferation, we examined the cell cycle profiles of Jurkat cells treated with E (31.2 μg/ mL).
The results of the cell cycle analysis revealed that E significantly increased the percentage of cells with sub-G 1 DNA content.As shown in Table 1, the cell population undergoing apoptosis (the sub-G 1 content) increased to 80.1% after 72 hours of incubation compared to 3.3% in untreated cells (p < 0.001).This ef-fect was accompanied by a proportional decrease in the percentage of cells in G0/G1, S and G2/M phase.

Escin induces apoptotic cell death
It has been generally accepted that the increase of cells having sub-G1 DNA content is a marker of apoptotic cell death.To clarify whether or not escininduced cell death involves apoptosis, flow cytometry analysis was employed.We found that β-escin significantly increased percentage of cells in early apoptosis in a time-dependent manner after 24, 48 and 72 h of treatment (10.44-31.88% vs. 3.52% in untreated cells).Simultaneously, percentage of annexin V/PI positive cells increased from 3.38% (control) to 28.15% (E-treated cells after 72 h of incubation).These experimental results demonstrate that E induced apoptosis in Jurkat cells (Table 2).

DNA fragmentation assay
Analysis of DNA fragmentation by agarose gel electrophoresis is one of the most widely used biochemical markers for cell death.The detection of internucleosomal DNA cleavage (DNA laddering) is considered to be an indicator of apoptosis (Nagata, 2000).As shown in Fig. 4, treatment of Jurkat cells with E (31.2 μg/mL) resulted in the formation of definite fragments that were visible during the electrophoresis as the characteristic ladder pattern (for 24, 48 and 72 h).

Escin stimulates ROS production
Reactive oxygen species (ROS) have been documented to activate either extrinsic or intrinsic apoptotic pathway (Circu & Aw, 2010).Because the production of ROS can be associated with escin-induced apoptosis, we decided to verify whether escin was able to stimulate ROS generation with flow cytometry and using the fluorescent probe rhodamine 123.
As shown in Fig. 5, E induced ROS increase in a time dependent manner in Jurkat cells.There was a significant increase in the ROS levels starting from 6 h and continuing up to 72 h (p < 0.05 and p < 0.01, respectively).

Activation of caspase-3 by escin
Caspase-3 is a crucial mediator of programmed cell death.To confirm whether E-induced cell death is associated with the activation of caspase-3, we analysed its activity with flow cytometry.We observed time-dependent increase in caspase-3 activity in E-treated Jurkat cells (Fig. 6).

Escin decreases mitochondrial membrane potential
There is accumulating evidence suggesting that loss of MMP can be associated with apoptosis.To evaluate mi-tochondrial membrane integrity, a highly specific probe for detection of changes in MMP (cationic dye TMRE) was used.Treatment of Jurkat cells with E (31.2 μg/mL) led to dissipation of MMP in a time-dependent manner (Fig. 7).

Effect of escin on mitochondrial protein content
Pre-treatment of Jurkat cells with E significantly decreased protein content in mitochondria in comparison to untreated cells (Table 3) (p < 0.01).

Effect on mitochondrial fluorescence spectra
The synchronous fluorescence fingerprint (SFF) of mitochondrial autofluorescence and horizontal cut of SFF at ∆λ = 50 nm are methods which in a simplistic form describe the fluorophores of outer mitochondrial membrane (Fig. 8A, 8B).Topographic map of SFF shows a specific pattern of spatial distribution of fluorescence for a given mitochondrial suspension after isolation.This mixture is the final result of total fluorescence of all fluorophores which are present in the mitochondria and reflects the quality of mitochondria.
Horizontal cut of SFF at selected ∆λ = 50 nm (Fig. 8B) of control and experimental mitochondria revealed two main characteristic fluorescence zones of mitochondrial mixture.The first zone (zone 1), a large peak of fluorescence located at 280/340 nm (excitation/emission) is consistent with protein fluorescence and most of the emission is due to excitation of tryptophan residues.The second zone (zone 2), a smaller peak at 350/400 nm is consistent mainly with NADH fluorescence (Kirkpatrick et al., 2005;Tomečková et al., 2011).
Our results showed that E influenced fluorescence of mitochondria isolated from Jurkat cells.We found decrease of fluorescence signal in the zone 1 as well as in the zone 2 in comparison with control mitochondria (Table 3, Fig. 9).

Enzyme activities and GSH content
Dismutation of superoxide radicals results in the formation of hydrogen peroxides by superoxide dismutase   (SOD).SOD activity in mitochondria from E-treated cells significantly increased in comparison with control (p < 0.001, Fig. 10A).Increase in SOD activity was associated with increased activity of GPx and GR.However, changes in metabolic activity of GPx and GR did not reach statistical significance (Fig. 10B, 10C).
GSH plays an important role in protection against oxidative stress-induced injury and intracellular GSH depletion results in apoptosis.We therefore analysed changes in the mitochondrial GSH level in E-treated Jurkat cells.As shown in Fig. 10D, a significant decrease in GSH content was observed in E-treated cells as compared to untreated cells (p < 0.001).

DISCUSSION
Cancer is one of the most important health problems in the world.In spite of the progress in cancer therapy, it is still necessary to develop new therapeutic strategies.Natural compounds, as highlighted also in this study, are becoming an important research area for novel and bioactive molecules for drug discovery (Vidinský et al., 2012;Bak et al., 2013;Tepe et al., 2013).
In the present study we aimed to analyze the effect of β-escin on the viability and proliferation of cancer cells.First, we evaluated the antiproliferative effect of this compound by MTT assay and impedance-based real-time cellular assay.In our in vitro studies, the treatment of cancer cells with E inhibited the growth of cells in a dose-dependent manner.Consistently with these results, it also suppressed clonogenic cell growth of HeLa cells and caused a significant decrease of cell viability at the concentration of 3.9 µg/mL.However, the exact mechanisms by which β-escin exerts its growth-inhibitory effect in cancer cells remain unclear.
One of the hallmarks of cancer is the ability of cancer cells to avoid apoptosis (Hanahan & Weinberg, 2011).Therefore, the apoptotic pathways are an interesting target of cancer treatment and many of the anticancer agents in current use have been shown to induce apoptosis in cancer cells.In an attempt to determine the mechanism responsible for the antiproliferative effect of the compound tested, we assessed apoptosis using DNA content analysis, annexin V/PI staining and DNA fragmentation assay.In our study we observed that the decrease in cell viability resulting from E treatment was associated with a reduction in the proportion of Jurkat cells in the S, G0/G1 and G2/M phases with concomitant increase in the fraction of cells with sub-G1 DNA content which is considered to be a marker of apoptotic cell death.In comparison with untreated cells, in E-treated cells the number of cells with sub-G1 DNA content was increased 12-, 22.2-and 24.3-fold after 24, 48 and 72 hours of treatment, respectively.Increase in the proportion of cells with sub-G1 DNA content after exposure of Jurkat cells to E was also documented by Zhang et al., 2011.To confirm whether E could induce  apoptosis in Jurkat cell line, we performed annexin V/PI staining, which detects both the early stage of apoptosis as well as the late stage of apoptosis/necrosis.Our results showed time-dependent decrease of vital cells number (An -/PI -) and a significant increase in annexin V positive cells (An + /PI -or An + /PI + ).The presence of apoptosis induced by E was further evidenced by nuclear DNA fragmentation which is a classical feature of apoptotic cell death.Our experimental results demonstrated that E induces apoptosis of cancer cells.These findings also supported caspase-3 activation observed in E-treated cells that could be one of the crucial steps to execution of apoptotic features.Pro-apoptotic effect of escin in different cancer cell lines has also been documented by other authors (Niu et al., 2008;Çiftçi et al., 2015;Güney et al., 2013).
It is well known that mitochondria play essential roles in the cellular metabolism (Nemoto et al., 2000).Mitochondria were also shown to be crucial for the regulation of intracellular Ca 2+ homeostasis, redox signalling, and regulation of cell proliferation.In addition to their role in cellular energy metabolism, mitochondria are now recognised as central players in cell death (Olszewska & Szewczyk, 2013).Furthermore, it is well known that changes in the structure of mitochondrial membrane may lead to the induction of apoptosis (Gogvadze & Orrenius, 2006;Parsons & Green, 2010) associated with changes in MMP, intracellular ROS production and caspase activation (Lemasters 1999;Wu et al., 2014).It was also stated that loss of MMP may increase permeability of the mitochondrial membrane with subsequent leakage of apoptogenic molecules from intermembrane space into the cytoplasm (Ly et al., 2003).Hence, in our study, assessment of ROS species, mitochondrial function and activity of antioxidant system were carried out.
In our work, MMP was measured in order to examine whether apoptosis was accompanied by the loss of mitochondrial membrane potential.We found significant decrease of MMP in cancer cells 6 h after E treatment.Our results indicate that decrease of MMP may play a role in E-induced cell death.Moreover, according to our best knowledge, the effect of β-escin on MMP has not been documented previously.
Furthermore, mitochondria are also important source of ROS, which are produced there in a consequence of aerobic respiration and oxidative phosphorylation (Księżakowska-Łakoma et al., 2014).Under physiological conditions ROS are involved in specific functions, such as signalling and protein regulation (Okon & Zhou, 2015).On the other hand, ROS can oxidize proteins and induce lipid peroxidation, compromising the barrier properties of biological membranes (Gogvadze & Orrenius, 2006).An increase in ROS production can  result in the impairment of the mitochondrial function that triggers the apoptosis pathway (Fleury et al., 2002).To investigate if the mitochondrial dysfunction found in E-treated Jurkat cells is initiated by ROS production, we measured ROS levels using the cell-permeable dye DHR-123.The result showed that E causes a rapid increase of intracellular ROS levels in Jurkat cells after 6 h of incubation.Other authors have also reported ability of E to generate ROS either in malignant (Shen et al., 2011) or non-malignant cells (Zhang et al., 2010).
Because there is a close relation between increased levels of ROS and MMP disruption (Mates et al., 2012), we suggest that E-induced cell apoptosis in Jurkat cells might be mitochondria dependent.
Moreover, in the present study we also demonstrated the changes of outer mitochondrial membrane in Etreated cells.As it was mentioned above, fluorescence at zone 1 (280/340 nm; excitation/emission) is consistent with protein/tryptophan-related fluorescence.We found that E decreased tryptophan-related signal which is in correlation with decreased protein content of mitochondria isolated from E-treated cells.Similar decrease in tryptophan fluorescence related to protein content was also documented by Kirkpatrick and co-workers (2005).
Furthermore, the NADH-related signal (zone 2) was strongly decreased by E treatment.Because NADH plays a role not only in a wide range of cellular reactions (e.g.energy metabolism, calcium homeostasis) (Cortassa et al., 2003), but may act as a natural biomarker for mitochondrial dysfunction (Heikal, 2010), our results indicated that mitochondria may be the target structures for E. This hypothesis is supported by the work of the Poque and co-workers (2001), who found that a decrease in NADH fluorescence is associated with mitochondrial membrane disruption.Furthermore, the study of Petit and co-workers (2001) indicated a close temporal relationship between decreased NADH fluorescence and the dissipation of the MMP.The similar decrease in the NADH fluorescence accompanied by a marked decrease of MMP and elevated intracellular ROS levels was also observed by Lemar et al., (2007).
As it was mentioned previously, ROS are mediators involved in different physiological as well as pathological processes.To prevent tissue injury, ROS are eliminated either by enzymatic (e.g.SOD, CAT, GPx) or non-enzymatic (e.g.glutathione) protective mechanisms.
In the present study a significant increase in SOD activity and mild increase in GPx as well as GR activity was observed.Because increased activity of antioxidant enzymes may be an adaptive cellular response to increased oxidative stress (Kale et al., 1999;Devi et al., 2000), we suggest the same mechanism is present also in E-treated cells.
Furthermore, some anticancer agents have demonstrated that GSH depletion can increase ROS generation and subsequently induces apoptosis in different cancer cells (Khan et al., 2012;Ong et al., 2008).To understand whether GSH is related to the β-escin-induced increase in ROS levels in cancer cells, changes in GSH levels were assessed.We found significant decrease in GSH in cells exposed to E. Because GSH is the most abundant intracellular thiol and the GSH-dependent antioxidant system plays the central role in cellular protection against oxidant injury (Armstrong et al., 2002), depletion of mitochondrial GSH increases susceptibility to oxidant injury (Marí et al., 2013).These results suggest that the oxidative stress plays an important role in the antiproliferative effect of escin on Jurkat cells.
In conclusion, our results showed that E exerts apoptotic effects on human cancer cells through the mechanism involving ROS generation, loss of mitochondrial GSH content, decrease of MMP and caspase-3 activation.Changes of mitochondrial autofluorescence also indicate the mitochondrial dysfunction.Although the exact mechanisms need to be investigated further, the growth inhibitory properties of E on cancer cell proliferation indicated the potential use of this compound as a chemotherapeutic agent for the treatment of some carcinomas.

Figure 9 .
Figure 9.Comparison of the maximum fluorescence signal of the first and the second fluorescence zones of synchronous spectra Δλ = 50 nm of control mitochondria isolated from Jurkat cells and experimental mitochondria isolated from Jurkat cells treated with β-escin (31.2 μg/mL).The first and the second zone show the maximal total fluorescence intensities of all fluorophores present in control and experimental mitochondria.The proteins and their interaction with escin are shown in the first zone of fluorescence which is strongly affected by light scattering (should be taken into evaluation).The interaction of mitochondrial coenzyme NADH + H + with β-escin was studied in the second zone of fluorescence as marker of the presence of oxygen and respiration metabolism in mitochondria.

Table 2 . Induction of apoptosis after β-escin treatment measured with Annexin V/PI staining.
Jurkat cells were treated with E (31.2 μg/mL) for 24, 48 and 72 h, stained with fluoresceinated Annexin V and propidium iodide (PI), and analysed using flow cytometry.