Chemico-Biological Interactions 258 (2016) 142e152

Contents lists available at ScienceDirect

Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint

DNA binding and apoptotic induction ability of harmalol in HepG2:

Biophysical and biochemical approaches
Sarita Sarkar 1, Paromita Bhattacharjee 1, Kakali Bhadra*
Department of Zoology, University of Kalyani, Nadia, West Bengal, 741235, India

a r t i c l e i n f o a b s t r a c t

Article history: Harmalol administration caused remarkable reduction in proliferation of HepG2 cells with GI50 of
Received 26 March 2016 14.2 mM, without showing much cytotoxicity in embryonic liver cell line, WRL-68. Data from circular
Received in revised form dichroism (CD) and differential scanning calorimetric (DSC) analysis of harmalol-CT DNA complex shows
6 August 2016
conformational changes with prominent CD perturbation and stabilization of CT DNA by 8
C. Binding
Accepted 29 August 2016
constant and stoichiometry was calculated using the above biophysical techniques. The Scatchard plot
Available online 31 August 2016
constructed from CD data showed cooperative binding, from which the cooperative binding affinity (K’u)
of 4.65 ± 0.7  105 M1, and n value of 4.16 were deduced. The binding parameter obtained from DSC
Keywords:
Harmalol-DNA interaction
melting data was in good agreement with the above CD data. Furthermore, dose dependent apoptotic
Apoptosis induction ability of harmalol was studied in HepG2 cells using different biochemical assays. Generation of
DSC ROS, DNA damage, changes in cellular external and ultramorphology, alteration of membrane, formation
CD of comet tail, decreased mitochondrial membrane potential and a significant increase in Sub Go/G1
Comet assay population made the cancer cell, HepG2, prone to apoptosis. Up regulation of p53 and caspase 3 further
ROS indicated the apoptotic role of harmalol.
© 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction neurophysiological and biochemical activities [9e15] as well as

in vivo and in vitro chemopreventive action against different cancer
Deoxyribonucleic acid that serves as the repository of genetic cell line [16e18], but its detail apoptotic effect on human cancer cell
information of the cell is the potential therapeutic targets in cancer. line has not been studied yet.
Again, plant alkaloids, a natural product as chemotherapeutic Inspired by these results, in the present investigation we
agents isolated so far, have been reported to have remarkable anti- elucidate, harmalol inducing apoptotic effects in HepG2 cells and
cancer applications [1e5] and consequently there is growing in- consequently study its stabilization, binding and conformational
terest in the search for better anti-cancer drugs with high efficacy, changes with CT DNA using different biochemical and biophysical
low toxicity and minimum side effects. But most of the chemo- techniques. Further, event like ROS generation and cell cycle arrests
therapeutic agents due to their rather non-selective nature and were also accompanied in the harmalol induced apoptosis. A
dose limiting toxicity, use is often restricted, necessitating search complete understanding of these therapeutic aspects of beta car-
for newer drugs having greater potential and suitability for use. b boline alkaloid will enable the researchers for the future drug
carbonil is one such large group of natural and synthetic alkaloids design for the betterment of mankind.
(ligand) of indole derivatives. These alkaloids were originally iso-
lated from plants like Peganum harmala L and Banisteriopsis caapi.
2. Materials and methods
Our previous investigation showed that beta carboline plant alka-
loid, harmalol, is an excellent intercalator preferring hetero G/C
2.1. Biochemicals
sequence [6,7]. Recently, we further showed its RNA binding ability
[8]. Harmalol, has been reported to have several pharmacological,
Harmalol was obtained from Sigma-Aldrich (St. Louis, MO, USA).
The purity of the sample was confirmed as done earlier [6,7]. Its
* Corresponding author.
concentration was determined using molar extinction coefficient
E-mail address: [email protected] (K. Bhadra). value of 19,000 M1 cm1 at 371 nm [7]. Double stranded calf
1
Department of Zoology, University of Kalyani, Nadia, Kalyani, 741235. thymus DNA (CT DNA) was also obtained from SigmadAldrich.

http://dx.doi.org/10.1016/j.cbi.2016.08.024
0009-2797/© 2016 Elsevier Ireland Ltd. All rights reserved.
 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152 143

Concentration was determined using the molar extinction coeffi- cells were placed on coverslips at 2  104/well in a 6 well plate for
cient (ε) of 6600 M1 cm1 at 260 nm. Harmalol and CT DNA was 48 h before treatment. Cells were then stained for 15 min with DAPI
dissolved in 15 mM CitrateePhosphate (CP) buffer of pH 6.8. (5 mM) at room temperature. Cells on cover slips were mounted for
fluorescence microscopic observation.
2.2. Cell lines and culture conditions
2.7. Scanning electron microscopy (SEM)
Five different human cell lines viz. HeLa (cervical carcinoma),
MDA-MB-231 (breast carcinoma), A549 (lung carcinoma), HepG2 To study the external morphology of cell, they were first fixed in
(liver epitheloid carcinoma) and WRL-68 (normal hepatic cells) glutaraldehyde and dehydrated through a graded series of ethanol.
were chosen, obtained from National Centre for Cell Science, Pune. Subsequently, they were cleaned in tetrachloromethane, air-dried
Cells were grown in DMEM supplemented with 10% heat inacti- and coated with gold in IB2 ION COATER and observed using
vated fetal bovine serum (FBS) and 1% antibioticeantimycotic, in a S530 Hitachi scanning electron microscope.
humidified atmosphere at 37
C with 5% CO2.
2.8. Transmission electron microscopy (TEM)
2.3. Cell viability test: MTT assay
Cultured materials were centrifuged at 3000 rpm for 4e5 min to
We tested the percentage of cell viability and calculated the GI50 discard the supernatant. The pellet was suspended in 0.1 M phos-
(50% growth inhibition) values for the above mentioned cell lines phate buffer (pH 7.4), dispersed and centrifuged again. Resus-
by MTT (1 mg/ml of the tetrazolium dye and 3-(4,5- pended the pellet and fixed in a mixture of 2% paraformaldehyde
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide dissolved and 2.5% glutaraldehyde in buffer for 3e4 h at 4
C. Centrifuged for
in phosphate buffer saline, pH 7.4) assay as reported earlier [19]. 10 min at 4
C and the supernatant were discarded. The samples
GI50 was calculated by using the equation were finally observed using TECNAI 200 KV TEM (Fei Electron
Optics), 35 mm Photography System from All India Institute of
GI50 ¼ ðT­TO Þ100=ðC  TO Þ (1) Medical Science, N. Delhi.

where, T is the optical density of the test well after 48 h drug 2.9. Estimation of DNA damage (comet assay and DNA gel
exposure, To is the optical density at time zero & C is optical density electrophoresis)
of control. The experiment was repeated three times and average
was taken. Comet assay was performed on isolated HepG2 cells by single
cell gel electrophoresis as reported by Liao et al. [22]. It consists of
2.4. LDH assay mainly the following steps- Slide preparation and cell lysis to
liberate the DNA, DNA unwinding, electrophoresis, neutralization
LDH activity was measured quantitatively by counting the of the alkali, DNA staining and scoring. Reagents required were- 1%
number of floating dead cells and adherent viable cells for both normal agarose; 1% low melting agarose; electrophoresis buffer
control and drug treated cells according to the protocol of Biswas (pH 10); neutralization buffer (pH 7.2) and lysis solution (pH 13 to
et al., [20]. The percent apoptotic and percent necrotic cell deaths 14). Cell suspensions of control and three different concentrations
were determined as follows: (7, 14 and 21 mM) of harmalol treated sets of cells were mixed with


1% low melting agarose at 37
C in 1:3 ratio and 100 ml of this
%Apoptosis ¼ LDHp  100% LDHp þ LDHi þ LDHe (2) mixture was applied for three times quickly on top of the fully
frosted four different glass-slides which were coated previously


%necrosis ¼ ðLDHe  100%Þ LDHp þ LDHi þ LDHe (3) with the 1% normal agarose. After solidification of the gel, these
glass-slides of control and three harmalol treated samples, con-
The floating cells were collected from culture media by centri- taining about 1000e2000 cells, were conveniently immersed in
fugation (3000 r.p.m) at 4
C for 4 min, and the LDH content from ice-cold lysis solution at 4
C for 16 h. After washing with
the pellets was marked as an index of apoptotic cell death (LDHp). neutralization buffer these slides were placed on the platform of
The released extracellular LDH in the culture supernatant was the electrophoresis apparatus and the electrophoresis was carried
marked as an index of necrotic cell death (LDHe), and the LDHi out at 0.8 V/cm for 15 min. Ethidium bromide solution of 0.5 mg/ml
present in the adherent viable count was used as intracellular LDH. was added to the gel and the slides were covered with cover slips.
HepG2 cells were seeded in 6-well plates at a density of The stained DNA in the cells was examined at 200 magnification
2  104 cells/well and treated with harmalol of 7, 14 and 21 mM with the help of OLYMPUS Fluorescence Microscope. We deter-
concentrations at 37
C for 48 h. mined the extent of DNA damage by measuring the % DNA in comet
tail by using software CASP. One hundred of each class of visually
2.5. FITC-Annexin V/PI FCM double staining classified comets from undamaged or no discernible tail to the
category having almost all DNAs in tail was selected for image
Double staining for FITC (fluorescein isothiocyanate) - Annexin analysis using CASP. We compared 100 comets of different cate-
V binding and for cellular DNA using PI (propidium iodide) FCM gories using 100 magnifications as the use of 100  magnification
(flow cytometry) was performed as described by Mallick et al. [21]. allows a faster analysis of gels.
For performing the DNA gel electrophoresis in 1% agarose gel,
2.6. Fluorescence microscopy we extracted DNA of HepG2 cells by standardized phenol-
chloroform method.
For the identification of nuclear changes such as chromatin
condensation, nuclear fragmentation and MMP (mitochondrial 2.10. Cell cycle inhibition analysis
membrane potential) alteration by JC-1, cells were visualized using
a fluorescence microscope (Olympus). DAPI, a blue fluorescent HepG2 cells were grown at a density of 2  105 cells/ml in
stain, that preferentially stains nuclei (apoptotic or viable). Briefly, 70 mm culture plate and treated with different concentrations of
144 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152

harmalol (7, 14, 21 mM) for 48 h. The other experimental protocol DSC thermogram was analyzed using Origin 7.0 software.
was followed as reported earlier [19].
2.16. UV optical melting study
2.11. RT-PCR analysis
Melting curves were recorded on a Jasco V-630 unit equipped
Equal amounts of total RNA, extracted with RNA express reagent with the peltier controlled Jasco PAC-743 model accessory (Jasco
(Himedia, Mumbai, India) were reverse transcribed and then sub- International Co. Ltd. Tokyo, Japan).
jected to PCR with enzymes and reagents of the reverse tran-
scription system (Chromous) using Techine PCR system
3. Results and discussions
(Staffordshire, UK). Sequences of primers used in the study are
given in Table 1.
3.1. Determination of GI50 by MTT

2.12. Immunofluorescence To evaluate the efficacy of harmalol (Fig. 1A) on different human
cancer cell line viz. HeLa, MDA-MB-231, A549, and HepG2, the
Anti active caspase 3 monoclonal primary antibody (17 Kd,
Santa Cruz Biotechnology), anti p53 monoclonal primary antibody
(Santa Cruz Biotechnology) and anti-mouse secondary antibody
(FITC tagged, Sigma) were used. HepG2 Cells were first grown on
cover slips and after getting 80% cell confluency the cells were
treated with harmalol with concentration of 7, 14 and 21 mM,
respectively. The cells were then washed with cold PBS, fixed with
4% paraformaldehyde, washed again and incubated with 0.1%
Triton X-100. Cells were then blocked with 5% BSA in TBS and
incubated with appropriate primary antibodies (activated caspase 3
and p53) in 1% bovine serum albumin at 4
C. After that cells were
again washed with TBS and incubated with an appropriate
fluorescence-conjugated secondary antibody at room temperature.
Finally, the cells were washed, counter stained with DAPI, mounted
on a glass slide and observed under a confocal microscope and was
subsequently analyzed with image J.

2.13. Generation of ROS and its quantification by FACS

Cells (2  105) were treated with harmalol at GI50 concentration

of 14.2 mM for 12, 24 and 48 h and the levels of intracellular ROS
were assessed by flow cytometry after incubating with DCFH-DA
(20 mM) (dichloro fluorescence diacetate) for 30 min at 37
C. For
inhibition of ROS generation, cells were pretreated with NAC (N-
acetyl cysteine) (10 mM) for 30 min before harmalol treatment.

2.14. Circular dichroism studies

Circular dichroism (CD) spectra were recorded on a Jasco J815

spectropolarimeter (Jasco International Co. Ltd.) attached with a
temperature controller and programmer (model PFD 425 L/15) and
the experimental parameters were selected as reported earlier
[6,7].

2.15. Differential scanning calorimetry (DSC)

DSC experiments were done on a Microcal VP-differential

scanning calorimeter (MicroCal, Inc., Northampton, MA, USA) and
was used to calculate the binding constant as reported earlier [23].

Table 1
Oligo-primer sequences for RT-PCR. Oligos shipped in lyophilized form (Company
name: Xcelris).

Name of the gene Sequence of primer

Fig. 1. (A) Chemical structure of harmalol. (B) Percentage growth inhibition of HeLa,
Casepase3 Forward-50 -AGGGGTCATTTATGGGACA-30
MDA-MB-231, A549, HepG2 and WRL-68 cells following treatment with harmalol of
Reverse-50 -TACACGGGATCTGTTTCTTTG-30
varying concentrations (5, 10, 20, 40 and 55 mM) for 48 h. Data are statistically analyzed
P53 Forward-50 -GGAAATTTGTATCCCGAGTATCTG-30
with ANOVA test and significant values are calculated against both untreated as well as
Reverse-50 -GTCTTCCAGTGTGATGATGGTAA-30
solvent control (*P < 0.05 vs. solvent control). (C) LDH assay, performed after harmalol
G3PDH Forward-50 -ATGGGGAAGGTGAAGGTCGG-30
treatment. Along with cytotoxicity, percentage of apoptotic and necrotic cell death was
Reverse-50 -GGATGCTAAGCAGTTGGT-30
also calculated. Data are representative of three independent experiments.
 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152 145

alkaloid was tested for its cytotoxicity using the MTT assay. After are anticipated to be far more specific and effective in cancer
48 h treatment, it was found that among the above four cell lines, therapeutics. Our results too supported these views in CT DNA.
harmalol showed GI50 values of 42.5 mM in HeLa cells, 23.7 mM in Though the conformational changes associated with the binding of
MDA-MB-231 cells, 14.2 mM in HepG2 cell and 45.2 mM in harmalol with CT-DNA was probed earlier by the authors through
A549 cells. Hence, the highest degree of concentration-dependent circular dichroic (CD) studies [6], but no quantitative data was re-
increase in growth inhibition was reported in HepG2 cell line ported from it. Both intrinsic (Fig. S2) as well as extrinsic (inset of
(Fig. 1B). However, for WRL-68, we could proceed only up to con- Fig. S2) CD measurements of the DNA on complexation with har-
centration of 55 mM for measuring its GI50 value through MTT assay. malol at pH 6.8 showed a regular and very prominent perturbation
Beyond this concentration, the alkaloid does not obey the Beer- in molar ellipticity. The data from CD was used here to evaluate the
Lambert's law [6], hence we could not proceed beyond this con- binding phenomena as described by Zhong et al. and Maidul et al.
centration. Therefore it can be predicted that WRL-68 might have [27,28]. The different spectra showed maximum at 278 nm and the
GI50 value < 55 mM, or in other words harmalol exhibited minimal change in the ellipticity of the different spectrum at 278 nm (q278)
effect of cytotoxicity (8.5 ± 0.5% apoptosis), by observing the versus harmalol concentration was plotted (Fig. 2A). The distance
morphology of the cells under microscope, on WRL-68 at concen- from the ordinate to the tangent at a given Dq was equivalent to the
tration of 55 mM. To further characterize harmalol-induced HepG2 concentration of the bound alkaloid (CB) while the distance from
cell death, the LDH assay was performed. LDH assay quantitatively the tangent to the curve gives the concentration of the free alkaloid
measure lactate dehydrogenase (LDH) released into the media from (Cf). A Scatchard plot constructed from this data showed coopera-
damaged cells as a biomarker for cellular cytotoxicity. It emphasizes tive binding (inset of Fig. 2A) from which the cooperative binding
on apoptotic index parameters in both treated and controlled affinity (K’u) of 4.65 ± 0.7  105 M1, an n value of 4.16 and a
(untreated) cells and observed to show 3.7 ± 1.1% necrotic, cooperativity factor (u) of 31 were deduced. These values are pre-
58.5 ± 1.2% apoptotic and 37.8 ± 1.0% viable in treated cell line, sented in Table 2, which are in good agreement to the values ob-
where as in controlled cell line, 4.8 ± 1.2%, 11.2 ± 0.9% and tained earlier from absorption and fluorescence studies [6]. The
84.0 ± 2.1% necrotic, apoptotic and viability, respectively, was increased magnitudes of the CT DNA CD bands are speculative of
observed (Fig. 1C). This was calculated on the basis of the intercalative ligand-DNA complexation and it also signifies that
morphological study of the cells under confocal microscope harmalol induces conformational changes in the DNA as shown by
(Fig. S1) and accordingly the percent apoptotic and percent necrotic the changes of the CD spectra of DNA. The extrinsic CD also con-
cell deaths were determined using the formula given in Section 2.4. stitutes strong evidence of intercalation as opposed to outside
Previously also harmalol has been reported to be very effective on stacking, though the confirmatory test for intercalative mode of
dioxin mediated induction of CYP1A1 in human hepatoma HepG2 binding was supported earlier by gel electrophoresis and visco-
cells [24] and reported to have antimutagenic and antigenotoxic metric technique by the author [6,7]. The study indicated the strong
effects in mammalian cells because of the antioxidant properties chiral environment of the intercalated harmalol in the stranded
[25,26]. organization of CT DNA and is due to the effective coupling of the
transition moments of the bound harmalol with the transition
moments of the chirally arranged base pairs of CT DNA.
3.2. Binding, conformational changes and stabilization of CT DNA Apart from getting information about the stabilization of the
polymer, thermal melting of DNAedrug complexes is an important
Apoptosis is specifically known to be intimately associated with tool to elucidate the binding parameters with enthalpy data from
the alteration of nucleic acid structure, function or stability. A new DSC. DSC melting temperature of CT-DNA (Fig. 2B) under similar
generation of agents that target nucleic acids associated processes,

Fig. 2. (A) Plot of molar ellipticity (q) change of CT DNA on harmalol titration versus harmalol concentration in the same buffer. Inset: Scatchard plot for CT DNAeharmalol
complexation. The points represent data obtained from CD analysis and the solid line is the best-fit curve to the cooperative equation of McGheeevon Hippel. (B) DSC thermogram
of CT DNA (curve 1) and its complexation with harmalol (curve 2). Inset: Optical thermal melting profile of the same polymer.
146 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152

Table 2
Calculation of binding constant and melting temperature of CT DNA and its complexation with harmalol at saturation D/P ratio of 0.7 in15 mM CitrateePhosphate (CP) buffer of
pH 6.8.

circular dichroic (CD) method K’ ¼ 1.5 ± 0.2  104; u ¼ 31; K0 u ¼ 4.65 ± 0.7  105; n ¼ 4.16

DSC Tm
(
C) Tm (
C) DTm (
C) DHm (kcal/mol) Ka (  105 M1)

69 77 8 8.19 ± 0.9 5.10 ± 0.6  105

Melting temperature of free CT DNA (Tm
) and in presence of saturating amount of harmalol (Tm). KTm is the binding constant at the melting temperature and K is the drug-
binding constant at 25
C calculated using equations (4) and (5) described in the text. DHm values have been obtained from the DSC data.

conditions were in complete agreement with the optical melting methodology is in reasonable concordance with that evaluated
data with a stabilization of DTm 8
C (inset of Fig. 2B) and the values directly from ITC and other spectroscopic techniques reported
of DHm evaluated from the DSC thermograms are presented in earlier [6]. The binding parameter obtained from DSC melting data
Table 2. It may be noted that in the present study, the melting was also in good agreement with the CD data (vide supra).
experiment was performed at saturating D/P of 0.7 and the data
was used to calculate the binding constant of harmalol to CT-DNA 3.3. FITC-Annexin V/PI flow cytometry of HepG2 cells
using the equation derived by McGhee [29].
To further study the efficacy of harmalol in HepG2 cells various
 o
1 Tm  1=Tm ¼ ðR= DHm Þ ln ð1 þ KTm LÞ1=n (4) apoptotic parameters were measured. Treatment with harmalol,
induced phosphatidylserine (PS) externalization in HepG2 cells,
where Tm
is the optical melting temperature of DNA alone, Tm is which is again another remarkable characteristic of early stage of
the melting temperature in presence of saturating amounts of the apoptosis. Fig. 3 shows the quantitative results of bivariate FITC-
drug, DHm is the enthalpy of DNA melting obtained by DSC exper- Annexin V/PI FCM of HepG2 cells after treatment with harmalol
iment, R is the gas constant, KTm is the drug binding constant at Tm, for different concentrations (at GI25, GI50 and GI75 concentrations of
L is the free drug concentration, and n is the size of the drug- 7, 14 and 21 mM respectively). The lower left quadrant of the cyto-
binding site. The calculated apparent binding constant at the grams shows the viable cells, which exclude PI and are negative for
melting temperature can be extrapolated to a reference tempera- FITC-Annexin V binding because of intact cytoplasmic membrane.
ture using the standard relationship. The upper right and left quadrant together represents the non-
viable, necrotic cells, positive for FITC-Annexin V binding and
lnðK=KTm Þ ¼ ðDHb =RÞð1=T  1=Tm Þ (5) showing PI uptake. Further, the lower right quadrant represents the
apoptotic cells, FITC-Annexin V positive and PI negative, demon-
where K is the drug binding constant at the reference temperature strating Annexin V binding to PS and cytoplasmic membrane
T (in Kelvin) and DHb, the binding enthalpy that was directly integrity. The FITCþ/PI apoptotic cell population increased grad-
determined from the isothermal titration calorimetry experiment ually from 4.3 ± 0.5% of control to 12.8 ± 1.3% after treatment with
done earlier [6]. The binding constant value obtained by this 7 mM, 29.3 ± 1.6% after treatment with 14 mM and 47.3 ± 2.1% after

Fig. 3. Contour diagram of FITC-Annexin V/PI flow cytometry of HepG2 cells after 48 h of incubation at (A) untreated, (B) GI25, (C) GI50 and (D) GI75 concentrations of 0, 7, 14 and
21 mM respectively. Data are representative of three independent experiments.
 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152 147

Fig. 4. (A) Changes in cell morphology and apoptotic index parameters as observed using S530 Hitachi scanning electron microscope (SEM), (B) Chromatin condensation and
nuclear fragmentation of HepG2 cells (marked with arrows) induced by harmalol at varying concentrations. Untreated HepG2 cells show homogenous staining of their nuclei. The
nuclei were stained with DAPI and observed under a fluorescence microscope (200).

treatment with 21 mM of harmalol respectively. On the other hand 3.4. Morphological and ultramorphological changes of apoptosis
FITCþ/PI þ necrotic cell population was increased from 1.1 ± 0.4% to
4.4 ± 1.4% after 48 h of treatment. Thus, FITC-Annexin V/PI FCM and The morphology of treated and untreated cells as observed by
LDH assay (vide supra) both supported higher percentage of SEM, have been shown in Fig. 4A. The morphological changes
apoptosis comparative to negligible percentage of necrosis in har- observed were cell rounding, cytoplasmic blebbing, shrinkage as
malol treated HepG2 cells. well as irregularities in cell contour and size in harmalol treated,

Fig. 5. Transmission electron micrograph (TEM) of (A) untreated HepG2 cell showing enlarge nucleolous, numerous intact mitochondria and lipid droplets, (B) HepG2 cell after
treatment with harmalol at GI25 showing sign of nucleolous fragmentation, (C) HepG2 cell after treatment with harmalol at GI50 showing sign of blebs at the cell surface with further
progress of nucleolous fragmentation (inset) and (D) HepG2 cell after treatment with harmalol at GI75 showing sign of phago lysosome formation with disorganized cytoplasmic
organelles. All the positive changes are marked with arrows.
148 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152

whereas, in untreated cell line, the cells remain spread out without fragmentation, a process which results from the activation of en-
any sign of irregularities in cell contour and size. In addition HepG2 donucleases during the apoptotic programme [30]. Hence, we
cells with DAPI staining revealed a significant increase in oligonu- further characterized the apoptotic HepG2 cells by the formation of
cleosomal fragmentation and nuclear condensation in harmalol comet tails (Fig. 6A). The results showed more DNA strand breaks
treated cells with increasing doses (Fig. 4B). Furthermore, ultra- constituting of 59 ± 4.3% DNA (longest tail length) with the treat-
morphological study by TEM (Fig. 5AeD), which is considered the ment of harmalol at 21 mM of concentration and least amount of
gold standard to confirm apoptosis, further revealed some early and DNA (18 ± 2.2%) with lowest concentration of 7 mM (Fig. 6B), while
late apoptotic features like nucleolous fragmentation, cytoplasmic at GI50 of 14.2 mM, 31 ± 2.8% DNA was observed in comet tail as
blebbing and formation of phago lysosomal complex. calculated using the software Casp. Furthermore, as evident in DNA
gel electrophoresis of harmalol treated cell (Fig. 6C), apoptotic
hallmark of chromosomal DNA laddering was clearly evident in
3.5. Comet tail and DNA ladder formation: indicative of DNA HepG2 treated with different doses of the alkaloid as compared to
damage untreated cells. Thus both these two assays indicated damage in
DNA, taking place during the apoptotic programme.
One of the later steps in apoptosis is ultimately the DNA

Fig. 6. (A) Single cell gel electrophoresis (commet assay) showing detectable comet tails (marked with arrows) with increasing concentration of harmalol, indicative of DNA
damage. (B) Bar graph representation of effect of harmalol treatment on % DNA of HepG2 cell in the comet tail using the software Casp. The graph displays the mean ± SD (standard
deviation) of three independent experiments. (C) Degradation of chromosomal DNA as DNA laddering formation during harmalol treatment at various concentrations.

Fig. 7. Dose-dependent effect of harmalol upon changes of mitochondrial membrane potential. HepG2 cell were treated with and then subjected to JC-1 staining to evaluate the
changes in mitochondrial membrane potential viewed under fluorescence microscope (200). In JC-1 stained cells, red fluorescence is visible in cell areas with high mitochondrial
membrane potential, while green fluorescence of JC-1 monomer is present in cell areas with low mitochondrial potential. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152 149

Fig. 8. Cell cycle analysis of harmalol treated HepG2 cells at 7 mM, 14 mM and 21 mM for 48 h of incubation. Data are representative of three independent experiments, and the means
are significant compared with their control (P < 0.05).

3.6. Loss of mitochondrial membrane potential

To further validate the ability of harmalol on inducing apoptosis

in HepG2 cells, we performed a cellular functional assay. JC-1 probe,
a voltage sensitive fluorescent cationic dye, exhibits membrane
potential dependent accumulation in mitochondria, indicated by a
fluorescence emission shift from red to green. The dye detects
normal polarized mitochondria as red color (due to aggregation)
and depolarized mitochondrial membranes as green color (due to
monomer formation). The exposure of harmalol at GI50 to HepG2
cells caused remarkable loss of mitochondrial membrane potential,
hence the fluorescence shifted from red to green as the membrane
potential (Jm) decreased (Fig. 7AeD).

3.7. Cell cycle arrest

Cell cycle analyses of harmalol-treated HepG2 cells were further

analyzed by using PI staining, and the cell distributions were Fig. 9. Harmalol (14.2 mM) induced ROS generation in HepG2 cells. HepG2 cells were
expressed among sub Go-/G1, Go-/G1, S and G2/M phases. Time ki- cultured in absence (control, green) and presence of harmalol for the specified time
netics study of harmalol-treated cells indicated increasing accu- period. DCFDH-DA (25 mM) fluorescence intensity was detected by using flow
cytometry. (inset) Percentage of growth inhibition in HepG2 cells in presence of NAC
mulation of cells at sub Go-/G1-phase. The percentages of sub Go-/
(10 mM). Data are representative of three independent experiments, and the means
G1arrest were 3.33 ± 0.6 and 10.42 ± 1.0 and 17.14 ± 1.2% after 7, 14 are significant compared with the control (*P < 0.05). (For interpretation of the ref-
and 21 mM treatment for 48 h in HepG2 cells (Fig. 8AeD). erences to colour in this figure legend, the reader is referred to the web version of this
article.)

3.8. ROS generation and its inhibition by NAC

Our result further complimented with ROS dependent cytotox-

icity in the above cell line. ROS generation, induced by various anti-
cancer agents [19,31e34], plays key role in apoptosis. HepG2 Cells
were exposed to GI50 concentration of harmalol (14.2 mM) for 6, 12,
24 and 48 h and analyzed for the presence of ROS using ROS sen-
sitive probe DCFH-DA (Fig. 9). ROS levels were found to be elevated
with time during the course of treatment in the cell lines. The mean
fluorescent intensity changed from 20.93 (control) to 22.68, 37.85,
74.05 and 89.49 after 6, 12, 24 and 48 h treatment, respectively
(inset of Fig. 9). ROS intensity increased nearly about 4 folds relative Fig. 10. RT-PCR analysis was conducted with cytosolic fraction of caspase 3 and p53
to control till 48 h treatment of harmalol. The cell-permeant 20 ,7'- activation. GAPDH was used as internal control for normalization of equal loading.
150 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152

dichlorodihydrofluorescein diacetate, (H2DCFDA), is very shown to protect against oxidative stress injury and inflammation in
commonly used to detect the generation of reactive oxygen species various hypoxia induced insult models [35]. Both NAC and catalase,
(ROS) intermediates like hydrogen peroxide, hydroxyl radical commonly known scavengers of ROS, inhibiting ROS levels and
(HO), nitric oxide (NO), singlet oxygen (1O2), autooxidation etc. apoptosis [36]. From this study it may further inferred that harmalol
Apoptosis is induced as a result of ROS and the change in fluores- induced death in HepG2 cell precede via ROS mediated pathway.
cence is measured using excitation/emission wavelengths of 500/
520 nm for DCF. DCF is obtained by hydrolysis of H2DCFDA, dihy- 3.9. Expression of apoptotic biomarkers: caspase 3 and p53
drofluorescein diacetates as colorless and nonfluorescent until both
of the acetate groups are hydrolyzed and the products are subse- Detection of harmalol induced apoptosis in HepG2 cells were
quently oxidized to fluorescein derivatives. Upon cleavage of the further confirmed by RT-PCR data revealing that unlike in un-
acetate groups by intracellular esterases and oxidation, the treated cells, harmalol treatment resulted in activation of Caspase 3
nonfluorescent H2DCFDA is converted to the highly fluorescent and increased expression of p53 in a dose dependent manner
20 ,7'-dichlorofluorescein (DCF) [19,32e34]. The fluorescence (Fig. 10). In addition to RT-PCR, we also performed the immuno
change after adding H2DCFDA (25 mM) that we have observed in the fluorescence with anti active caspase 3 (Fig. 11) and p53 (Fig. 12) to
control system i.e. HepG2 cells in absence of harmalol, might be see the analysis at the protein level. Simultaneously we also
because of some redox environment, but the same system after showed the nuclear accumulation by DAPI of excess concentration
adding the alkaloid (harmalol at concentration of GI50) showed of active caspase 3 and p53 of apoptotic HepG2 cells. The expression
further change/shift in fluorescence intensity with time (shown in of p53 and caspase-3 that favored apoptosis, were up-regulated in a
the inset bar graph) which is because of apoptosis induced ROS and dose dependent manner. Evidences suggested that p53 tumor
we have even calculated the level of significance by using one way suppressor gene is involved in cell cycle regulation, DNA repair, and
ANOVA in Origin that revealed that the values were statistically, programmed cell death, and it further lead to caspase activation
significantly different (increased) from the control (*p values after [30,37]. It is through mitochondrial cytochrome c release that p53
6 h of incubation- 0.017843, after 12 h of incubation- 0.028057, induces apoptosis by caspase activation [38].
after 24 h s of incubation- 0.032464 and after 48 h of incubation- Ability to induce apoptosis in cancer cells is considered as one of
0.036374, with respect to the control). The flow cytometric analysis the hallmark features of any drug with a good anti-cancer potential.
of the control cells (cells cultured in the absence of harmalol, It is a complex set of multi branched, multistep pathways with
Fig. S3) was measured at every interval (i.e. after 6, 12, 24 and 48 h) many checks and balances [39]. As the activators, effectors and
along with the tested cells and there was no change in fluorescence regulators of this cascade continue to be elucidated, a large number
intensity because of the absence of apoptosis induced by ROS. of apoptosis assays are devised to detect and count apoptotic cells.
Moreover, preincubation with NAC (N-acetyl-L-cysteine) also However, many features of apoptosis and necrosis can overlap, and
initiated the growth inhibitory role of harmalol (inset of Fig. 9). NAC therefore it is important to employ two or more specific assays to
is commonly used to identify and test ROS inducers, and to inhibit confirm that cell death is occurring via apoptosis only [39].
ROS and this is probably due to the formation of H2S that has been Apoptosis triggered by various stimuli is characterized by a series of

Fig. 11. Immunocytochemical localization of active caspase 3 in HepG2 cells by confocal microscope after exposure to 7, 14 and 21 mM of Harmalol for 48 h.
 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152 151

Fig. 12. Immunocytochemical localization of p53 in HepG2 cells by confocal microscope after exposure to 7, 14 and 21 mM of Harmalol for 48 h.

distinct biochemical and morphological changes [40]. HepG2 cells dx.doi.org/10.1016/j.cbi.2016.08.024.

exposed to different doses of harmalol at GI25, GI50 and GI75 con-
centrations of 7, 14 and 21 mM, respectively, also showed several Transparency document
such apoptotic characteristic like DNA ladder formation, chromatin
condensation, nucloeolus fragmentation, cytoplasmic blebbing, Transparency document related to this article can be found
formation of comet tails, alteration of membrane structure, loss of online at http://dx.doi.org/10.1016/j.cbi.2016.08.024.
mitochondrial membrane potential and formation of phago lyso-
some, all of which confirmed the progression of the cells towards References
apoptosis.
In conclusion, our results demonstrated harmalol as a plausible [1] M.F. Roberts, M. Wink, Biochemistry, in: M.F. Roberts, M. Wink (Eds.), Ecology
and Medicinal Applications, Alkaloids vol. 1, Plenum Press, , New York, Lon-
effective alkaloid for inducing ROS dependent; caspase activated don, 1998.
apoptosis in the human liver cancer cell line, HepG2 and further- [2] K. Bhadra, G. Suresh Kumar, Therapeutic potential of nucleic acid binding
more, shows binding, stabilization and conformational changes in isoquinoline alkaloids: binding aspects and implications for drug design, Med.
Res. Rev. 31 (2011) 821e862.
DNA. Hence, this research work will enhance harmalol as a po-
[3] R. Ali, Z. Mirza, G.M. Ashraf, M.A. Kamal, S.A. Ansari, G.A. Damanhouri,
tential chemo preventive agent against liver cancer and will enable A.M. Abuzenadah, A.G. Chaudhary, I.A. Sheikh, New anticancer agents: recent
the researchers in future to develop a better, potent beta carboline- developments in tumor therapy, Anticancer Res. 32 (2012) 2999e3005.
[4] M. Moudi, R. Go, C.Y.S. Yien, M. Nazre, Vinca alkaloids, Int. J. Prev. Med. 4
indole derivative for future biomedical applications.
(2013) 1231e1235.
[5] G. Suresh Kumar, G. Brahmachari (Ed.) Wiley-VCH Verlag GmbH & Co. KGaA
pp (2015) 241e277, DOI: 10.1002/9783527684403.ch9.
Acknowledgements [6] S. Sarkar, K. . Bhadra, Binding of alkaloid harmalol to DNA: photophysical and
calorimetric approaches, J. Photochem. Photobiol. B 130 (2014) 272e280.
KB and SS are indebted to the Council of Scientific and Industrial [7] S. Sarkar, P. Pandya, K. Bhadra, Sequence specific binding of beta carboline
alkaloid harmalol with deoxyribonucleotides: binding heterogeneity, confor-
Research (CSIR), Government of India for the financial support mational, thermodynamic and cytotoxic aspects, PLos One 9 (2014) 1e14,
(Ref. No. 37 (1538)/12/EMR-II). PB is supported by grant from Uni- http://dx.doi.org/10.1371/journal.pone.0108022 e108022.
versity of Kalyani. Authors are thankful to DST-PURSE and PRG- [8] P. Bhattacharjee, S. Sarkar, P. Pandya, K. Bhadra, Targeting different RNA
motifs by beta carboline alkaloid, harmalol: a comparative photophysical,
2015-16, University of Kalyani and also grateful to the Depart- calorimetric and molecular docking approach, J. Biomol. Struct. Dyn. (2015),
ment of Environmental Science, University of Kalyani for providing http://dx.doi.org/10.1080/07391102.2015.1126694.
fluorescence microscope facility. [9] D.H. Aarons, G.V. Rossi, R.F. Orzechowski, Cardiovascular actions of three
harmala alkaloids: harmine, harmaline and harmalol, J. Pharm. Sci. 66 (1977)
1244e1248.
Appendix A. Supplementary data [10] A.F.M. Abdel Fattah, K. Matsumoto, Y. Murakami, K.A.W. El-Hady,
M.F. Mohamed, H. Watanabe, Facilitatory and inhibitory effects of harmaline
on the tryptophan-induced 5-hydroxytryptamine syndrome and body tem-
Supplementary data related to this article can be found at http:// perature changes in pargyline-pretreated rats, Jpn. J. Pharmacol. 72 (1996)
152 S. Sarkar et al. / Chemico-Biological Interactions 258 (2016) 142e152

39e47. [26] D.J. Moura, M.F. Richter, J.M. Boeira, J.A.P. Henriques, J. Saffi, Antioxidant
[11] M. Mahmoudian, H. Jalilpour, P. Salehian, Toxicity of Peganum harmala: re- properties of b-carboline alkaloids are related to their antimutagenic and
view and a case report, Iran. J. Pharmacol. Ther. 1 (2002) 1e4. antigenotoxic activities, Mutagenesis 22 (2007) 293e302.
[12] R. Cao, W. Peng, Z. Wang, A. Xu, b-Carboline alkaloids: biochemical and [27] W. Zhong, J.S. Yu, Y. Liang, K. Fan, L. Lai, Chlorobenzylidine-calf thymus DNA
pharmacological functions, Curr. Med. Chem. 14 (2007) 479e500. interaction II: circular dichroism and nuclear magnetic resonance studies,
[13] T.J. Herraiz, Identification and occurrence of beta-carboline alkaloids in raisins Spectrochim. Acta A 60 (2004) 2985e2992.
and inhibition of monoamine oxidase (MAO), Agric. Food Chem. 55 (2007) [28] M.M. Islam, P. Pandya, S. Kumar, G. Suresh Kumar, RNA targeting through
8534e8540. binding of small molecules: studies on t-RNA binding by the cytotoxic pro-
[14] K. Patel, M. Gadewar, R. Tripathi, S.K. Prasad, D.K. Patel, A review on medicinal toberberine alkaloid coralyne, Mol. Biosyst. 5 (2009) 244e254.
importance, pharmacological activity and bioanalytical aspects of beta- [29] J.D. McGhee, Theoretical calculations of helix-coil transition of DNA in the
carboline alkaloid Harmine, Asian Pac. J. Trop. Biomed. 2 (2012) 660e664. presence of large, cooperatively binding ligands, Biopolymers 15 (1976)
[15] A.M. Lucyna, R. Hans (Eds.), Current Topics in Neurotoxicity, Springer, 2012. 1345e1375.
[16] T. Uezono, W. Maruyama, K. Matsubara, M. Naoi, K. Shimizu, O. Saito, [30] J. Das, S. Das, A. Samadder, K. Bhadra, A.R. Khuda-Bukhsh, Poly (lactide-co-
Norharman an indoleamine-derived betacarboline, but not Trp-P-2, a gamma- glycolide) encapsulated extract of Phytolacca decandra demonstrates better
carboline, induces apoptotic cell death in human neuroblantoma SH-SY5Y intervention against induced lung adenocarcinoma in mice and on A549 cells,
cells, J. Neural Transm. 108 (2001) 943e953. Eur. J. Pharma. Sci. 47 (2012) 313e324.
[17] F. Lamchouri, M. Zemzami, A. Jossang, A. Settaf, Z.H. Israili, B. Lyoussi, Cyto- [31] S.K. Jaganathan, A. Mazumdar, D. Mondhe, M. Mandal, Apoptotic effect of
toxicity of alkaloids isolated from Peganum harmala seeds, Pak. J. Pharm. Sci. eugenol in human colon cancer cell lines, Cell Biol. Int. 35 (2011) 607e615.
26 (2013) 699e706. [32] R.H. Bo €hmer, L.S. Trinkle, J.L. Staneck, Dose effects of LPS on neutrophils in a
[18] D.L. Bemis, J.L. Capodice, P. Gorroochurn, A.E. Katz, R. Buttyan, Anti-prostate whole blood flow cytometric assay of phagocytosis and oxidative burst,
cancer activity of a beta-carboline alkaloid enriched extract from Rauwolfia Cytometry 13 (1992) 525e531.
vomitoria, Int. J. Oncol. 29 (2006) 1065e1073. [33] O. Myhre, J.M. Andersen, H. Aarnes, F. Fonnum, Evaluation of the probes 2',7'-
[19] S. Chatterjee, S. Mallick, F. Buzzetti, G. Fiorillo, T.M. Syeda, P. Lombardi, K. Das dichlorofluorescin diacetate, luminol, and lucigenin as indicators of reactive
Saha, G. Suresh Kumar, New 13-pyridinealkyl berberine analoguesintercalate species formation, Biochem. Pharmacol. 65 (2003) 1575e1582.
to DNA and induce apoptosis in HepG2 and MCF-7 cells through ROS mediated [34] A. Paul, S. Das, J. Das, A. Samadder, A.R. Khuda-Bukhsh, Cytotoxicity and
p53 dependent pathway: biophysical, biochemical and molecular modeling apoptotic signaling cascade induced by chelidonine-loaded PLGA nano-
studies, RSC Adv. 5 (2015) 90632e90644. particles in HepG2 cells in vitro and bioavailability of nano-chelidonine in mice
[20] R. Biswas, S.K. Mandal, S. Dutta, S.S. Bhattacharyya, N. Boujedaini, A.R. Khuda- in vivo, Toxicol. Lett. 222 (2013) 10e22.
Bukhsh, Thujone-rich fraction of Thuja occidentalis demonstrates major anti- [35] C. Yang, Z. Yang, M. Zhang, Q. Dong, X. Wang, A. Lan, F. Zeng, P. Chen, C. Wang,
cancer potentials:evidences from in vitro studies on A375 cells, Evid.e Based J. Feng, Hydrogen sulfide protects against chemical hypoxia-induced cyto-
Complement. Altern. Med. 2011 (2011) 1e16, http://dx.doi.org/10.1093/ toxicity and Inflammation in HaCaT cells through inhibition of ROS/NF-kB/
ecam/neq042. COX-2 pathway, PLoS One 6 (2011) e21971.
[21] S. Mallick, P. Ghosh, S.K. Samanta, S. Kinra, B.C. Pal, A. Gomes, [36] M. Halsi, M. Wang, T.S. Chavan, V. Gaponenko, N. Hay, A.L. Gartel, ROS in-
J.R. Vedasiromoni, Corchorusin-D, a saikosaponin-like compound isolated hibitor N-acetyl-L-cysteine antagonizes the activity of proteasome inhibitors,
from Corchorus acutangulus Lam., targets mitochondrial apoptotic pathways Biochem. J. 454 (2013) 201e208.
in leukemic cell lines (HL-60 and U937), Cancer Chemother. Pharmacol. 66 [37] S. Sahaa, S. Sikdar, A. Mukherjee, K. Bhadra, N. Boujedainic, A.R. Khuda-
(2010) 709e719. Bukhsh, Ethanolic extract of the Goldenseal, Hydrastis canadensis, hasde-
[22] W. Liao, M.A. McNutt, W.-G. Zhu, The comet assay: a sensitive method for monstrable chemopreventive effects on HeLa cells in vitro: drugeDNA inter-
detecting DNA damage in individual cells, Methods 48 (2009) 46e53. action with calf thymus DNA as target, Environ. Toxicol. Pharm. 36 (2013)
[23] K. Bhadra, M. Maiti, G. Suresh Kumar, Thermodynamics of the binding of 204e214.
cytotoxic protoberberine molecule coralyne to deoxyribonucleic acids, Bio- [38] M. Schuler, E.B. Wetzel, J.C. Goldstein, P. Fitzgerald, D.R. Green, p53 induces
chim. Biophys. Acta 1780 (2008) 298e306. apoptosis by caspase activation through mitochondrial cytochrome c release,
[24] A.M. El Gendy, A.A. Soshilov, M.S. Denison, O.S. El-Kadi, Transcriptional and J. Biol. Chem. 275 (2000) 7337e7342.
posttranslational inhibition of dioxin-mediated induction of CYP1A1 by har- [39] S. Elmore, Apoptosis: a review of programmed cell death, Toxicol. Pathol. 35
mine and harmol, Toxicol. Lett. 208 (2012) 51e61. (2007) 495e516.
[25] J.N. Picada, K.V.C.L. da Silva, E.D. Erdtmann, A.T. Henriques, J.A.P. Henriques, [40] A.H. Wyllie, The genetic regulation of apoptosis, Curr. Opin. Genet. Dev. 5
Genotoxic effects of structurally related b-carboline alkaloids, Mutat. Res. 379 (1995) 97e104.
(1997) 135e149.