Lignan enriched fraction (LRF) of Phyllanthus amarus promotes apoptotic cell death in human cervical cancer cells in vitro

Phyllanthus amarus is widely grown in this sub-continent and used traditionally to treat many common ailments. In the present study, lignan rich fraction of P. amarus extract was used on cervical cancer cell lines (HeLa, SiHa and C33A) to study it’s mechanism of cell death induction. As the cells were treated with IC50 doses of LRF, characteristic apoptotic features were observed. Increased sub G0 population were observed both in Hela and C33 cells, while G1/S arrest was observed in SiHa cells than their untreated counterparts. Increased production of ROS and change in MMP were also detected in the treated cells. Presence of γH2AX, was observed by immunofluorescence. Reduced expression of HPV (16/18) as well as ET-1, an autocrine growth substance, were observed in the treated cells. Immunoblotting as well as ICFC studies showed enhanced expressions of BAX, Caspase 3 and PARP (cleaved) in the treated cells. A major lignan, phyllanthin was isolated from the chloroform fraction and showed strong irreversible affinities for viral E6 and MDM2 in in silico analysis. The study conclusively indicates that LRF has the potential to induce apoptotic cell death in cervical cancer cells by activation of p53 and p21 against DNA damage.


Results elucidation of chemicals constituents. Several compounds were identified by LC-MS
and  [11][12][13][14] . Several phenolic compounds, terpenoids, specially, squalene (a precursor of steroid biosynthesis) were detected in the LRF by GC-MS analysis (Table S1). Several sub-fractions had also been isolated from LRF by silica gel chromatography. GC-MS analysis of the sub fractions detected methyl esters of several usual phytochemicals such as Palmitic acid, Stearic acid and α-Linolenic acid (as identified with NIST library) after derivatization with TMS. Single crystals isolated by column chromatography were identified as phyllanthin by XRD analysis (CCDC submission ID: 1820905) (Table S2). 1H-NMR and ESI-MS analysis also confirmed that (Fig. 1).

Evaluation of ROS mediated DNA damage by LRF. Flow cytometric determination of ROS induction.
All the treated cells showed enhanced ROS accumulation after three hours. Maximum ROS was generated in the treated SiHa cells, whereas, Hela cell line showed minimum accumulation of ROS,with control sets showing 7.79% and LRF treated sets with 20.82% ROS positive cell population. A 2.67 fold increase in ROS positive cells were detected in HeLa cells. While in untreated SiHa cells, 6.96% cells were detected with ROS, but, quite a higher percentage of (57.45%) ROS positive cells were detected in the treated sets. Thus, there is an increase of 8.25 fold in SiHa cells. A 3.76 fold increase in cell population was found in C33A cells with 8.19% cells in control sets and 30.83 cells in treated sets (Fig. 2a).
DNA damage detection by γ-H2AX incorporation. Reactive oxygen Species (ROS) can induce DNA damage, resulting either single /double stranded nicks. As the nicks are generated, the H2AX (a variant of H2A) becomes phosphorylated and spreads over the nicks, which are known as foci. Immunofluorescence micrographs clearly indicated the incorporation of phosphorylated H2AX(γ-H2AX) in nuclei of LRF treated cell. Incorporation of γ-H2AX is a clear marker of single or double stranded DNA damage (Fig. 2b) and apoptosis.
Study of DNA damage response by immunoblotting. Expressions of proteins as induced by DNA damage, were also studied by western blotting to validate this. In all the treated cells phospho-ATM and phospho-Chk2 could not be detected, but presence of phospho-ATR and phospho-Chk1were detected indicating activation of checkpoint-kinase 1 and ATR. In the treated HeLa cells, a significant increase in the expression of p-ATR and p-Chk1 (1.8 fold and 1.7 fold respectively) were observed. But, in treated SiHa cells, the increase was comparatively less (0.6 fold and 0.4 fold) for p-ATR and p-Chk1 respectively. In case of treated C33A cells, however, p-ATR and p-Chk1 expressions were increased by 0.7 fold and 0.6 fold respectively (Fig. 2c).This clearly indicate that, LRF induced ROS can cause DNA damage in these cancer cells.
Determination of Apoptosis inducing potential of LRF. Apoptosis induction. At 18 h treatment time, most of the cells showed signs of later stage of apoptosis. These observations clearly indicate towards induction of apoptosis in the treated cells (Fig. 3a). In HeLa cells, 45.54% cells were found to be apoptotic, with (7.19%) dead cells. In SiHa cells, 22% apoptotic cells along with 3.32% dead cells were found. In C33A cells, 44.54% cells were found to be apoptotic with 7.52% dead cells. At 24 h treatment time, most of the cells were in the 4 th quadrate, with www.nature.com/scientificreports www.nature.com/scientificreports/ lesser number of early and late apoptotic cells. In HeLa cells, 23.99% cells were found to be apoptotic, with many (41.9%) dead cells. In SiHa cells, most of the cells were dead (82.65%) with 9.02% apoptotic cells. In C33A cells, 77.88% cells were found to be apoptotic with 8.6% dead cells.
Fluorescence micrographs of 18 h and 24 h also support the flow cytometric data (Fig. S3).

Change in morphology.
Treated cells showed significant changes in cellular and nuclear morphology as seen by confocal microscopy. In HeLa cells, cytoplasmic blebbings, shrinkage of membrane, chromatin condensation and nuclear fragment, all the signs of typical apoptotic cells were observed (Fig. 4a). In SiHa cells also, cell shrinkage along with chromatin condensation and fragmentation were observed (Fig. 4a). In C33A cells, more cellular shrinkage along with nuclear condensation was found in the treated sets (Fig. 4a). Cells were counted and expressed as number of fragmented/ condensed nuclei present per 100 cells in control and LRF treated sets (Fig. S4). It was observed that, in LRF treated Hela, SiHa and C33A cells, maximum numbers of fragmented/ condensed nuclei were 54.74%, 46.87% and 34.92% respectively.

Loss of mitochondrial membrane potential (MMp).
Mitochondria play an important role in inducing cell death; loss of membrane potential (ΔΨm) is a key event in intrinsic cell death. Reduced MMP was quantified in treated cells (Fig. 4b). In HeLa control sets, 3.27% depolarized cells were observed than the treated cells which showed 55.42% depolarized cells, which is significantly (16.95 fold higher) more than the control sets. In SiHa cells, 3.71% depolarized cells were there in control sets while treated sets had 29.97% depolarized cells (8.08 fold increased). In C33A cells, 5.12% and 29.5% cells were found in control and treated sets respectively, showing a 5.77 fold increase in depolarized cell population.
change in cell cycle. With LRF treatment, varied responses were observed. In HeLa and C33A cell lines, sub G0 population has increased suggesting cytotoxicity, whereas in SiHa cells, cells were blocked at G1/S phase. A decrease of 8.79% in G0G1 populations along with an increase of 12.03%in sub G0 population was observed www.nature.com/scientificreports www.nature.com/scientificreports/ in HeLa cells. In C33A cells, increase of 8.68% in sub G0 population along with 9.48% decrease in G0G1, 3.9% decrease in S and 4.36% increase in G2/M population were observed in treated sets. In SiHa cells, S and G2/M populations decreased by 3.11% and 4.87% respectively while sub G0 population increased by 1.58% in treated sets (Fig. 4c).

interplay of cell survival -cell death genes at transcriptional and translational level. Semi-Q
RT-PCR. In all the treated cells, elevated levels of pro-apoptotic gene and reduced levels of anti-apoptotic gene expressions were observed. Viral HPV 16/18 E6 gene expression was found to be decreased in SiHa and HeLa cells along with the decrease in expression of Endothelin 1(ET-1).
In treated HeLa cells (Fig. 5a), Bcl-2 expression level was decreased (by 0.78 fold) while BAX expression was elevated (by 0.39 fold). p53 expression increased to 0.97 fold higher but decreased expression of p21 was observed (0.04 fold) in the treated sets. Huge decrease in expressions of ET-1(0.97 fold) and HPV 18E6 (0.79 fold) were observed in the treated sets. Treated SiHa cells (Fig. 5a) showed 0.61fold decrease in Bcl-2 expression along with 0.7 fold increase in BAX expression. 0.22 fold increase in p53 expression along with mild increase of 0.1 fold for p21 were observed. Expression of ET-1 and HPV 16 E6 were found to be decreased by 0.55 fold and 0.21 fold respectively. No significant change in expressions of Bcl-2 and BAX was found in treated sets of C33A cells (Fig. 5a). For both the p53 and p21, there was an increase in expressions (0.07 fold and 0.36 fold). ET-1 expression was not detected as expected in this cell line, which didn't have HPV viral load.
Immunoblotting. Balance between the pro-apoptotic and anti-apoptotic proteins decide the cell fate, if proapoptotic proteins expressed more than the antiapoptotic proteins, then the cell dies and vice-versa. In the present work, expressions of anti-apoptotic, pro-apoptotic and some relevant proteins were studied by immunoblotting to understand the cell death mechanism. Expression of caspase 8 and 9 were studied at 18 h, while other proteins were studied at 24 h post treatment. All the treated sets showed elevated levels of initiator caspase 8 and 9, effecter caspase3, pro-apoptotic BAX, p53 and reduced levels of anti-apoptotic Bcl-2 proteins. In treated HeLa cells (Fig. 5b), expression of cleaved caspase 3, caspase 9 and caspase 8 were found to be increased by 0.95, 0.92

Intracellular flow cytometry (ICFC). Cleaved Caspase 3. Increase in cells having cleaved caspase 3 was
found in all the treated sets (Fig. 5c). In control HeLa cells, 11.27% cells expressed cleaved caspase 3 while 22.88% cells were found positive for cleaved caspase 3 in treated sets. An increase of 2.03 fold of cells expressing cleaved caspase 3 was detected. In case of SiHa cells, 13.91% cells expressed cleaved caspase 3 in the control set and52.86% cellsexpressed cleaved caspase 3 in the treated sets with an increase of 3.8 fold. 24.77% and 45.36% cells expressed cleaved caspase 3 in control and treated sets respectively in C33A cells with an increase of 1.83 fold.
Cleaved PARP. Cells having cleaved PARP increased in all the treated sets ( Fig. 5c), highest in HeLa followed by C33A and SiHa cells. In HeLa control sets, 12.02% cells expressed cleaved PARP while 39.09% cells of treated sets expressed cleaved PARP, with an increase of 3.3 folds. In SiHa cells, cells expressing cleaved PARP were 8.34% and 22.63% in control and treated sets respectively with an increase of 2.72 fold. Highest fold increase in cleaved PARP positive cells (3.84 fold) was found in C33A cells with 9.02% cells in control sets and 34.68% in treated sets.
BAX. In the treated sets, mean fluorescence intensity of BAX positive cells increased in all the cell lines (Fig. 5c). Maximum expression was observed mostly in C33A cells followed by HeLa and SiHa cells. In HeLa cells, BAX expression increased by 1.75 fold while in SiHa it increased by 1.64 fold. It increases 2.43 fold in C33A. www.nature.com/scientificreports www.nature.com/scientificreports/ Bcl-2. In the treated sets, mean fluorescence intensity of Bcl-2 positive cells decreased in all the cell lines ( Fig. 5c), mostly in SiHa followed by HeLa and C33A cells. In HeLa cells, Bcl-2 decreased by 0.16 fold,but in SiHa and C33A it decreased by 0.19 fold and 0.14 fold respectively.
p53. In the treated sets, mean fluorescence intensity of p53 positive cells increased in all cell lines, most in SiHa cells followed by C33A and HeLa cells (Fig. 5c). In HeLa cells, p53 increased by 1.44 fold while in SiHa, it increased by 3.12 fold. 2.29 fold increase of p53 was there in C33A.
In silico analysis of phyllanthin binding with p53 and HPV 16/18 E6. Different types of interactions between phyllanthin and p53, MDM (known inhibitor of p53) and E6 (viral protein which affects p53 stability) were depicted in Fig. 6 and tabulated in Table S3. It was quite clear that phyllanthin had the most rigid binding with E6 protein, having more negative free energy of binding than that of with natural p53 inhibitor MDM2. Interaction between p53 and phyllanthin was not favorable. Table 1 summarizes the hydrogen and pie bond formation by different amino acid residues of the proteins with phyllanthin. Phyllanthin was found to interact with amino acid residue Glu154 and form hydrogen bonds with carbon 7 & 8. It also formed π-Interactions with purine bases and Phe157.

Discussion
Traditionally, decoction of Phyllanthus was used to treat intestinal and liver disorders, but in recent times, anti-diabetic, cardioprotective, anti-inflammatory and anti-cancer properties of this plant (tested on different cell lines) were reported 15 . Along with this, this plant extract was also found to have anti-fungal, anti-bacterial and anti-viral activity 15 .The anti-cancer activity was attributed mainly due to inhibition of cell cycle regulations and DNA repair mechanism 16 . Previous workers have reported that, hexane extract (100 µg/ml) or lignan rich fraction (100 µg/ml) of this plant extract or isolated pure compounds such as niranthin (  www.nature.com/scientificreports www.nature.com/scientificreports/ death respectively) but much less cytotoxicity was observed in vincristine resistant Lucena-1 cells (16.3,40.4,29.4, 30.2 and 24.8% respectively). It is interesting to note that K-562 cells are more sensitive (LC50 4.95 µM) to doxorubicin than the resistant Lucena-1 cells, as the LC50 dose was more (LC50 50 µM). But Lucena-1 cells treated with 5 µM Doxorubicin along with P. amarus derivatives; greater cytotoxicity was observed 17 . Quercetin was reported to inhibit tumor growth and can trigger apoptosis in several types of cancers (cervical cancer, prostate cancer, osteosarcoma and oral cancer) [18][19][20] . Methanolic extracts of hairy root culture of P. amarus increased time dependent cytotoxicity in MCF7 cells. That cytotoxicity was found to be associated with elevated levels of ROS as well as decrease in MMP 21 .
DNA damage is very often involved with in-vitro cytotoxicity. DNA damage stabilizes p53 concentration, inducing p21 transcription, which is a CDK inhibitor. This p21 then act to arrest the cell cycle at G1 or G2 22 . ROS mediated apoptosis was induced by methanolic extracts of hairy root culture of P. amarus in MCF7 cells 21 . In our study, it was observed that, short duration treatments (3 h) with IC50 doses generate ROS in the treated cells. ROS induces reduction of mitochondrial membrane potential. ROS generation led to intensive DNA damage and induces DNA damage dependent pathways. DNA damage beyond repair triggers cell death. Apoptosis triggered by various DNA lesions have already been reported by earlier works 23 . LRF was found to induce apoptotic DNA fragmentation, which was evident, as extensive DNA damage was observed both in DNA fragmentation and comet assay (reported in our earlier publication) 24 . So, it can be concluded that LRF treatment induces DNA damage. In the present work, enhanced expressions of proteins related to DNA damage responsive pathways were found. As a result of DNA damage several kinases become active. Out of them, the most important are ATM, ATR and DNA-Pkcs. ATR is activated in single stranded DNA breaks whereas, ATM and DNA-Pkcs are associated with double stranded DNA breaks 25 . As initial response to DNA damage, the histone variant of H2A, H2AX become phosphorylated by ATM/ATR, at the damaged part of the DNA 26,27 . This is essential for accumulation of MDC1 in response to damaged chromatin [28][29][30][31][32][33][34] . In our study, western blot analysis showed enhanced expression of phospho-ATR but no expression of phospho-ATM was observed. At 6 h time point expression of phospho-Chk1 increased in treated sets suggesting induction of DNA replication stress 35 . Detection of γ-H2AX also validates the presence of nicks on DNA strands. Expression of γ-H2AX was earlier considered to be associated with double stranded DNA breaks, but recent report showed that, it can also get phosphorylated by ATR dependent manner in response to DNA replicative stress and does not need ATM activation 36 . p21 mediated arrest of cell cycle is also correlated with the phosphorylation of H2AX 37 . Extensive nuclear DNA damage acts as an intrinsic signal to mitochondria, triggering release of cytochrome c to the cytosol and tripping the balance in favour of cell death rather than repair and survival. Loss of membrane potential had led to caspase activation and apoptosis. Quantitative data for ROS generation, loss in MMP and expression of apoptotic proteins with LRF treatment supports this hypothesis.
Hoechst stained cells studied under Confocal microscope, showed clear indication of condensed chromatin, presence of fragmentations was also clear in all the three treated sets indicated apoptotic cell death. Methanolic extract of P. amarus was reported to induce caspase-3 mediated apoptosis in DLA cells at a concentration of 75-200 µg/ml 38 . LRF induced cell cycle shifts in the treated cell lines and the degree of response varied among the cell lines. Flow cytometric analysis of cell cycle showed LRF treatment induced G1/S arrest only in SiHa cells, but not in other cell lines (HeLa and C33A cell lines, where increase in sub G0 population was observed). As the cell cycle progress from G1 to S phase, it is evident that cell cycle proteins, that is CDKs and cyclin Ds form an active G1phase MPF that leads the cells from G1 to S. But CKIs (mainly Kip family/ Wnk family inhibitors) may check the progression and arrest the cells in G1/S border, if the cell is not prepared to move into the next phase due to some abnormal conditions (as for example, presence of any DNA damage, etc.). In the present study, it is noted, that p21 have a role to block the progression of cells in G1/S. As p21 blocked the cells at G1/S, lesser number of cells progressed to G2/M. Caspase 3,7 mediated apoptosis with alternation in MAPK, AKT and NF-kB pathways was reported in PC3 cells by the methanolic extract of P. amarus 39 . Several signalling pathways are induced simultaneously in the treated sets and p53 acted as the prime regulator that might have controlled these pathways. In general, HPV infected cells; viral E6 binds p53 and disrupts its transcription, thereby impairing its tumor suppressor activity. But, in this study, E6 expression (both transcriptionally and translationally) was found to be reduced significantly. Down regulation of E6 in the treated counter parts of HPV(16/18) positive cells helped in p53 stabilization and subsequent p21 transcription.It was observed, that LRF has induced DNA damages in the cells (as it was evident by incorporation of γ-H2AX in the treated nuclei). Semi qRT-PCR showing the status of ET-1 in treated sets, clearly indicated that, the expression of ET-1 was absent in HeLa cells and down regulated inSiHa cells, both treated with LRF. Endothelin-1 (ET-1) is an autocrine signalling growth factor, released from HPV transfected keratinocytes. It is a vasoactive peptide and induces increased growth response and cell proliferation either directly or in synergy with other growth factors. So, reduced ET-1 expression clearly vindicated that LRF was not only able to induce cell death but also inhibited the cell proliferation of the cancer cell lines. Immunoblotting analysis also showed the enhanced expression of the pro apoptotic proteins over antiapoptotic proteins. Expression of cleaved caspase 8 and 9 (studied at 18 h) increased considerably in the treated cells. As caspase 8 is reported to inhibit autophagic or necrotic cell death 40,41 , it's activation indicates apoptotic cell death. LRF induced apoptosis was mediated by pro-apoptotic proteins, such as BAX, cleaved PARP and cleaved caspase 3, as, increased expression was detected by intra cellular flow cytometry analysis. At the same time, ROS mediated DNA damage and loss of mitochondrial membrane potential, clearly indicated p53 mediated intrinsic pathway of apoptosis.
The docking results clearly suggest that phyllanthin has more affinity towards E6 than p53. Being a natural inhibitor of p53, MDM2 served as the positive control in this experiment. Figure 6 depicts the interaction of different amino acids of the proteins with phyllanthin. Strong binding capability as well as interactions through hydrogen bonds with different amino acids of E6 made phyllanthin a good lead molecule for E6 inhibition. www.nature.com/scientificreports www.nature.com/scientificreports/ From the study, it can be concluded that LRF has the potential to induce cell death in cervical cancer cells in vitro. LRF may had induced cell death through ROS, which lead to DNA damage and loss of MMP. Presence of nicked DNA and subsequent ATR activation pathway also validates the hypothesis. Induction of ATR activation pathway has vindicated the hypothesis. Down regulation of viral E6 is also instrumental in inducing up regulation of p53. p53 played the central role in inducing apoptosis in the cell lines. The presence of viral factors in HeLa and SiHa and their subsequent down regulation by LRF treatment actually made them more susceptible towards this treatment that the C33A cell line which is devoid of any viral load. The probable mode of action of LRF fraction is summarized in Fig. 7. The anti-proliferative activity of LRF might be due to synergistic actions of the components (phyllanthin, fatty acids, niranthin, corrilagin etc.) present in them. Activation of caspase 8, 9 and 3 indicates towards apoptotic cell death.

Materials and Methods
chemicals and reagents. Chemicals needed for chemical elucidation (methanol, n-Hexane, ethyl acetate, chloroform, Silica gel) were purchased from Merck; BSTFA was from Sigma. Cell culture media (EMEM, DMEM), FBS, FCS, antibiotic-antimycotic solutions were procured from HiMedia Labs. DCFDA (ABCAM 113851), JC1 (BD Biosciences551302), AnnexinV-FITC/PI (BD-Biosciences556547) and Propidium iodide (Sigma) were used in the flow cytometric assays. Primers for RT-PCR were from Sigma. RNA extracted by TRI reagent (Sigma) and 1 st strand cDNA systhesis kit (Thermo Scientific K1621) was used. Antibodies for WB, IF and ICFC were purchased from Cell signaling technology, Sigma and Santa Cruz biotechnology (Table S4)  www.nature.com/scientificreports www.nature.com/scientificreports/ at 20-50 eV. In MSE mode, for collision-induced dissociation (CID), Argon was used as the collision gas. The mass spectrometer was calibrated over a range of 50-1500 Da using a solution of sodium formate to ensure mass accuracy and reproducibility.
GC-MS. LRF was TMS-derivatized with BSTFA and injected in HP5-MS capillary column (30 m X 0.25 mm × 0.25 µm) followed by running through a temperature gradient of 70-260 °C with 5 °C per min ramping rate (Agilent). Fragmentation pattern of the peaks were tallied with NIST library.
Column chromatography. LRF was further separated in a silica gel column (60-120 mesh) to get several sub-fractions. Concentrated LRF was mixed with silica gel (60-120) to prepare slurry which was loaded in a glass column packed with silica gel (60-120).Elution was done with hexane, ethyl acetate and methanol according to increasing order of polarity. Sub-fractions were collected for ½ of the column bed volume. They were then concentrated and separated by thin layer chromatography followed by poolingwhen necessary. Pooled fractions were tested for their cytotoxicity and analysed by GC-MS.
Characterization of single crystal. The crystal was mounted (Bruker AXS D8 Quest ECO) with the help of nylon looponthe diffractometer. The diffractometer contained a Mo-target rotating-anode X-ray source and a graphite monochromator (Mo Kα, λ = 0.71072 Ǻ). The crystallographic data have been listed in Table S2. SHELXS-97 (Sheldrick 2008) and SHELXL-2014/6 (Sheldrick 2014) were used for data solution and data refinement respectively. The crystal data was submitted to CCDC. A Bruker AM 300 L 500 MHz superconducting FT NMR spectrophotometer was used to record the 1 H NMR spectra for the isolated crystal resolved in deuterated chloroform (CDCl3) at 293 K. MicromassQtof YA 263 mass spectrometer was used to measure the ESI mass spectra in positive ion mode.
ROS induction by flow cytometry. 1 × 10 6 cells were stained with DCFDA following manufacturer's protocol and analysed by flow cytometry (BD FACSVerse). In presence of ROS, DCFDA, a cell permeant fluorogenic dye readily oxidizes to 2,7-Dichlorofluorescin which is highly fluorescent.

Determination of loss of mitochondrial membrane potential (MMP) by flow cytometry.
After treatment, 1 × 10 6 cells were stained with JC1 following manufacturer's protocol and analysed by flow cytometry (BD FACSVerse) to study change in the MMP. JC1 is a lipophilic cationic dye that can selectively enter mitochondria. With intact MMP, it exists as aggregates. It'sfluorescence changes from red (aggregates) to green (monomers) with decrease in MMP.

Determination of apoptosis by annexin V-FITC/ PI double staining. FITC conjugated Annexin V/
Propidium Iodide double staining method was used to quantifyamount of apoptotic cells. Externalization of phosphatidylserine (PS) is a well known marker of apoptosis. Annexins, a family of calcium-dependent phospholipid binding proteins, bind to PS and helps to mark apoptotic cells. After treatment, 1 × 10 6 cells were harvested, washed with PBS, stained with FITC conjugated Annexin V/Propidium Iodide for 15 min at dark according to manufacturer's protocol (BD 556547). Later they were subjected to flow cytometry (BD Accuri C6 plus/ BD FACSVerse) and analysed by FCS-express software. Also cells were grown on glass cover slips and treated with LRF for 18 and 24 h, stained with Annexin V-FITC/ PI and visualized under epi-fluroscence microscope (Leica DMi8). cellular and nuclear morphology study. . Hoechst 33258 stained cells were studied under fluorescent microscope. Then the treated cells were washed with PBS and fixed with 4% PFA followed by washing again. The fixed cells were then stained using Hoechst 33258 (2 µg/ml) for 2 min at room temperature. They were mounted in mounting medium (PBS:Glycerol = 1:9) containing propyl gallate after washing with PBS. Photographs were taken in an confocal microscope (Leica DMi8).
Cell cycle analysis Treated cells were harvested and washed with PBS. 70% ethanol was used to fix the cells which were then treated with RNaseA for 2 h at 37 °C and stained with PI. After 15 min incubation in dark, flowcytometry (BD FACSVerse) was done to study cell cycle.
Semi quantitative reverse transcription pcR. Tri regent was used to isolate total RNA from the cells as depicted from the manufacturer's protocol. RNA was quantified by Nanodrop (Eppendorf) and bleach gels were used to check the integrity of the RNA samples 43 . First strand cDNA synthesis kit was employed to prepare cDNA using 1 µg RNA as starting material as per manufacturer's protocol. With the help of Primer 3 software,primersrequired for the PCR reactions were designed, using FASTA sequences of target genes from NCBI. The primer sequences and PCR profiles were listed in (Table S5).GAPDH was used as internal control. PCR products were separated in 2% Agarose gels and Gel documentation system (UVP MultiDoc-It) was used to photograph them. Densitometric analysis was done by Image J software.