Fluorescent labeling of ursolic acid with FITC for investigation of its cytotoxic activity using confocal microscopy

Article abstract of DOI:10.1016/j.bioorg.2018.11.052

Fluorescent labeling is a widely-used approach in the study of intracellular processes. This method is becoming increasingly popular for studying small bioactive molecules of natural origin; it allows us to estimate the vital intracellular changes which occur under their influence. We propose a new approach for visualization of the intracellular distribution of triterpene acids, based on fluorescent labeling by fluoresceine isothiocyanate. As a model compound we took the most widely-used and best-studied acid in the ursane series – ursolic acid, as this enabled us to compare the results obtained during our research with the available data, in order to evaluate the validity of the proposed method. Experimental tracing of the dynamics of penetration and distribution of the labeled ursolic acid has shown that when the acid enters the cell, it initially localizes on the inner membranes where the predicted target Akt1/protein kinase B – a protein that inhibits apoptosis – is located.

Full text of DOI:10.1016/j.bioorg.2018.11.052

Accepted Manuscript
Fluorescent labeling of ursolic acid with FITC for investigation of its cytotoxic
activity using confocal microscopy
Tatiana S. Frolova, Alla V. Lipeeva, Dmitry S. Baev, Sergey I. Baiborodin,
Кonstantin E. Orishchenko, Alexey V. Kochetov, Olga I. Sinitsyna
PII:
S0045-2068(18)30794-6
DOI:
https://doi.org/10.1016/j.bioorg.2018.11.052
YBIOO 2660
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 August 2018
26 November 2018
29 November 2018
Please cite this article as: T.S. Frolova, A.V. Lipeeva, D.S. Baev, S.I. Baiborodin, K.E. Orishchenko, A.V. Kochetov,
O.I. Sinitsyna, Fluorescent labeling of ursolic acid with FITC for investigation of its cytotoxic activity using confocal
microscopy, Bioorganic Chemistry (2018), doi: https://doi.org/10.1016/j.bioorg.2018.11.052
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Fluorescent labeling of ursolic acid with FITC for investigation of its cytotoxic activity using confocal microscopy
Tatiana S. Frolovaa,b,c,d*, Аlla V. Lipeevab, Dmitry S. Baevb,c, Sergey I. Baiborodina, Кonstantin Е. Orishchenkoa,
Alexey V. Kochetova,c, Оlga I. Sinitsynaa,c
aFederal Research Center Institute of Cytology and Genetics Siberian Branch of the Russian Academy of Sciences, 10,
Lavrentyev ave., 630090, Novosibirsk, Russia
bNovosibirsk Institute of Organic Chemistry Siberian Branch of the Russian Academy of Sciences, 9, Lavrentyev ave.,
630090, Novosibirsk, Russia
cNovosibirsk State University, 2, Pirogov Street, 630090, Novosibirsk, Russia
dFederal Research Center of Fundamental and Translational Medicine SB RAS, 2, Timakov Street, 630117,
Novosibirsk, Russia
*Corresponding author, e-mail: frolova@bionet.nsc.ru
ABSTRACT
Fluorescent labeling is a widely-used approach in the study of intracellular processes. This method is becoming
increasingly popular for studying small bioactive molecules of natural origin; it allows us to estimate the vital
intracellular changes which occur under their influence. We propose a new approach for visualization of the
intracellular distribution of triterpene acids, based on fluorescent labeling by fluoresceine isothiocyanate. As a model
compound we took the most widely-used and best-studied acid in the ursane series – ursolic acid, as this enabled us to
compare the results obtained during our research with the available data, in order to evaluate the validity of the
proposed method. Experimental tracing of the dynamics of penetration and distribution of the labeled ursolic acid has
shown that when the acid enters the cell, it initially localizes on the inner membranes where the predicted target
Akt1/protein kinase B – a protein that inhibits apoptosis – is located.
Abbreviations:
DMSO – dimethyl sulfoxide; FITC – fluoresceine isothiocyanate; UA – ursolic acid, UA* – beta-epimer of fluorescent
ladeled ursolic acid, MD – molecular dynamics.
1

Keywords:
Triterpene acids, ursolic acid, fluorescence, cytotoxicity, Akt1/protein kinase B, MDM2/E3 ubiquitin-protein ligase.
2

1. Introduction
Triterpenoids comprise a large class of natural compounds, widespread among plants [1]. One of the best studied
triterpenoid compounds is 3β-hydroxy-urs-12(13)-en-28-oic (ursolic) acid (Fig.1), which can be found in significant
amounts in plant material [2].
Fig. 1. Ursolic acid
Particular attention has been drawn to ursolic acid’s property of having selective cytotoxicity in respect of cancer cells
[3]. The acid is capable of inducing apoptosis via a mitochondrial pathway; by activating caspase-3 in M4Beu
melanoma cells; by inhibiting the proliferation of cells of the B16 line by arresting the cell cycle [4,5]. It has been noted
that a concentration of ursolic acid that is non-toxic for normal cells can induce apoptosis in В16F-10 melanoma cells
by inhibiting the transcription factor NF-κB – a universal transcription factor that is involved in the control of gene
expression during immune responses, apoptosis and the cell cycle [6,7]. Ursolic acid also shows significant anti-
angiogenic activity in models in vivo and in vitro [8–10]. Studies of the activity of various ursolic acid derivatives show
that the modifications at C-3 and C-28 positions enhance the cytotoxic properties, especially introduction of an amino
group [11–13]
Study of the mechanisms of anti-cancer action involves the use of a large number of molecular-biological methods. In
healthcare, pharmacology and cellular biology, fluorescence methods are frequently used, as they generally represent
the fastest and safest approaches [14,15].
Fluorophore attachment is applicable not only for proteins, but also for small bioactive molecules. Thus, the labeled 7-
nitro-1,2,3-benzadiazole (NBD) ligand for the opiate receptor σ2 has been used in the study of the mechanism of its
operation in the pancreatic gland oncotransformed BxPC3 line of cells [16]. It has been established that large amounts
of this receptor are also present in some other lines of cancer cells where they participate in triggering the apoptosis
[17–19], therefore, this is a potential target for therapy of oncological diseases. Adding NBD and Cy5.5 fluorophores to
fructose is an approach that has been developed to enable the detection of cancer cells of the lacteal gland. Such cells
differ from normal ones in their high levels of the Glut5 transporter and, therefore, an increased fructose uptake [20].
The mechanisms of action of some other anti-proliferative compounds – wortmannin [21] and azathilone [22] – have
also been investigated with the help of NBD. The mechanism of the antimalarial effect of 1,2,3-trioxolanes has been
3

examined with the use of the fluorescent dye dansyl chloride [23]. Fluorescence is often used when investigating the
intracellular effects of anticancer compounds that cause changes in the cytoskeleton: rhodamine and fluorescein-marked
taxoids have turned out to be very convenient for examining the processes of microtubule assembly [24,25]; while
phalloidin is used to visualize the reconstruction of actin filaments [26].
In biological research fluorescein isothiocyanate (FITC) has often been used for visualizing the attachment of NH - or
2
SH-groups [27–29], because it is non-toxic and allows us to observe such intracellular processes in vivo. In this paper
we propose a new approach for visualization of the intracellular distribution of triterpene acids, based on fluorescent
labeling. As a model compound we took the most widely-used and best-studied acid in the ursane series – ursolic acid,
as this enabled us to compare the results obtained during our research with the available data, in order to evaluate the
validity of the proposed method.
To synthesize the necessary derivatives, we used a well-known method for the preparation of 3α- and 3β-amino-urs-
12(13)-en-28-oic acid 2, 3 from 3β-hydroxy-urs-12(13)-en-28-oic acid 1 [30] followed by further reaction of the
derivatives with fluorescein isothiocyanate 4 (Fig.2).
Fig. 2. The scheme for obtaining ursolic acid labeled with FITC
Thus, the reaction of 3β-hydroxy-urs-12(13)-en-28-oic acid with the Jones reagent [31] leads smoothly to the formation
of 3-oxo-urs-12(13)-en-28-oic acid, which further reacts with ammonium acetate in the presence of sodium
cyanoborohydride to form a mixture of the necessary epimers of 3-amino-urs-12(13)-en-28-oic acid 2 and 3 with yields
of 20 and 40% respectively for the α and β-forms (after column chromatography and re-crystallization according to the
operating procedure [30]).
The key stage in the synthesis of fluorescence labeled 3β-hydroxy-urs-12(13)-en-28-oic acid is the reaction of the
epimers 2 and 3 with fluorescein isothiocyanate 4, which proceeds under mild conditions at room temperature, resulting
in products 5 and 6 with yields of 88 and 82% respectively. It should be noted that, despite the starting materials being
widely-known, the synthesized products have not previously been described.
2. Results and discussion
2.1. Molecular modeling and molecular dynamics simulations for ursolic acid and its FITC labeled epimers
4

Previously, we have reported on a proposed mode of action of pentacyclic triterpene acids, where we suggested that
Akt1/protein kinase B and MDM2/E3 ubiquitin-protein ligase were their binding targets [32]. In order to test this
hypothesis we suggested a new approach, based on fluorescent labeling of the compounds under test, and chose ursolic
acid as the modeling compound.
We performed a molecular docking, which allowed us to estimate the binding energy between the protein
macromolecule and the potential ligand and to draw conclusions on whether an interaction was likely, in order
theoretically, to substantiate the idea that binding would still continue after the addition of the label.
Docking was performed for ursolic acid and its fluorescent labeled epimers. The binding energies are shown in Table 1.
The results are visualized in Fig. 3 and 4.
Table 1. Vina scoring function results: molecular docking of ursolic acid and its fluorescent labeled epimers into Akt1
and MDM2
Fig. 3. Interaction between ligands and Akt1: A) referent ligand XM1; В) ursolic acid; С) α-epimer; D) β-epimer. Non-
covalent interactions are shown as dashed lines: hydrogen bonds - green, hydrophobic interactions - purple, ionic
interactions - orange. In ligands, ursolic acid is shown in yellow, FITC part of molecule is gray.
Fig 4. Interaction between ligands and MDM2: A) referent ligand Nutlin 3a; В) ursolic acid; С) α-epimer; D) β-epimer.
Non-covalent interactions are shown as dashed lines: hydrogen bonds - green, hydrophobic interactions - purple. In
ligands, ursolic acid is shown in yellow, FITC part of molecule is gray.
These results indicate that the interaction with Akt1 and MDM2 is preserved even after addition of the fluorescent label.
To determine the stability of the predicted conformations six protein-ligand complexes of ursolic acid and its FITC
labeled epimers with Akt1 and MDM2 proteins were exposed to MD simulation.
The RMSD of the backbone atoms of each residue in complexes of UA, α-epimer and β-epimer with Akt1 and MDM2
was analyzed to observe the stability of the ligand/protein structure. The comparisons of the RMSD values of UA/Akt1,
α-epimer/Akt1 and β-epimer/Akt1 complexes are shown in Fig. 5.
5

Fig. 5. RMSD of the backbone atoms of Akt1 docked complexes at 300K. Complexes are indicated: blue – UA/Akt1,
red - α-epimer/Akt1, green - β-epimer/Akt1.
During the first 6.500 ps, the values for β-epimer/Akt1 complex are more variable and on average exceed the values for
UA/Akt1 and α-epimer/Akt1 complexes. Then, up to 10.000 ps, the RMSD values and its variability for β-epimer/Akt1
complex are significantly reduced. The α-epimer/Akt1 complex is generally more stable than UA/Akt1, but after 8000
ps of MD simulation the values of its RMSD converge. Thus, it should be assumed that α-epimer/Akt1 complex is more
stable than β-epimer/Akt1. The presence of β-epimer leads to significant fluctuations in the backbone structure of the
Akt1 protein.
The RMSD values of UA/MDM2, α-epimer/MDM2 and β-epimer/MDM2 complexes are shown in Fig. 6.
Fig. 6. RMSD of the backbone atoms of MDM2 docked complexes at 300K. Complexes are indicated: blue –
UA/MDM2, red - α-epimer/MDM2, green - β-epimer/MDM2.
The UA/MDM2 complex is the most unstable - high variability of RMSD value is observed. During the first 3.000 ps,
FITC labeled epimers complexes exhibit a similar stability pattern. However, then the average value of RMSD for β-
epimer /MDM2 complex increases, while the variability and average value of RMSD for α-epimer/MDM2 complex
becomes the lowest. The most stable complex is α-epimer/MDM2.
2.2. Synthesis of fluorescent labeled epimers
After molecular modeling a synthesis to obtain labeled epimers was performed. The following show the structures of
the obtained ursolic acid epimers and of FITC (Fig. 7 and 8).
Fig. 7. α-epimer of ursolic acid labeled with FITC
Fig. 8. β-epimer of ursolic acid labeled with FITC
The products at each stage of synthesis were characterized by NMR-analysis. An elemental analysis was conducted,
confirming the appropriate ratio of the elements in the above structural formulae.
6

2.3. Investigation of cytotoxic activity
The next stage was to check the changes in properties under in vitro conditions. For this, an MTT-assay was carried out
for ursolic acid. Data are summarized in Table. 2.
Table. 2. Half-maximal inhibitory concentration (IC50) for ursolic acid after 72 h of incubation.
Then, using the same technique, the cytotoxic properties of the resulting conjugate was checked. The ratio of label was
varied from 0 to 100%, the other fraction being native ursolic acid. The total concentration of the compounds was 50
μM in all cases (Fig. 9).
Fig. 9. Investigation of the cytotoxic properties of ursolic acid labeled with FITC on the MCF-7 cell line. The ratio of
label was varied from 0 to 100%, the other fraction being native ursolic acid. The total concentration of the compounds
was 50 μM in all cases.
During the in vitro experiments we demonstrated that adding the label intensifies the cytotoxic properties of ursolic
acid, but there was a more pronounced effect with the β-epimer (UA*) and so this was then examined in greater detail
with cells from the MCF-7 line. As this line has been thoroughly studied, including the effects of treatment with
triterpene compounds, this allowed us to make a full comparison of our results with the available data.
2.4. Detection of apoptosis
Apoptosis was detected with acridine orange/ethidium bromide double staining, 24 and 48 h after the treatment of the
MCF-7 cells with the labeled ursolic acid at a concentration of 25 μM. The results are shown in Table 3, Fig. 10.
Table 3. Apoptosis stages (%) of MCF-7 cell line treated with ursolic acid and UA* at 25 μМ after 48 and 72 h of
incubation.
7

Fig. 10. Investigation of proapoptotic activity of UA* on MCF-7 cell line. Normal cells are colored green with acridine
orange, at apoptosis the cell nucleus is coloured red with ethidium bromide, and the cytoplasm is coloured green with
acridine orange. Dead cells are stained only with ethidium bromide.
2.5. Investigation of cell morphology using the Cell IQ system
As indicated above, the addition of the label intensifies the apoptotic properties of the ursolic acid. A detailed
examination of the dynamics of the cytotoxic effect of UA* was made visually using the Cell IQ system. Cell death was
detected 3 h after the start of incubation (the cells became rounded, detached from the surface and fragmented). At the
end of the experiment almost all the cells were dead. The results are shown in Table 4.
Table 4. Investigation of the morphology of MCF-7 cells treated with ursolic acid and UA* at a concentration of 25 μM
using the Cell IQ system during 72 h of incubation
Also using Cell IQ, we chose a duration (the first 24 hours of incubation) that would enable a more detailed
examination of cell morphology, using confocal microscopy.
2.6. Confocal microscopy
We carried out microscopical analysis of MCF-7 line cells treated with the β-epimer. The cells were examined over the
first 24 hours after treatment, at intervals of 3 hours, and by 24 hours there was a significant, visible, apoptotic effect.
To define the changes in cell morphology we used additional stains: DAPI, CellMask Orange and MitoTracker Red.
Such a combination allowed us to identify the changes in structures of the plasmalemma, the inner membranes, the
mitochondria and the nuclei. The results of confocal fluorescence microscopy are shown in Table 5.
Table 5. Analysis of distribution of UA* in MCF-7 cells carried out with confocal fluorescence microscopy
The appearance of a labeled ursolic acid signal in the MCF-7 line cells was registered 6 hours after the treatment; the
signal intensified as time went on. After 12 hours of incubation with labeled ursolic acid its signal distribution visually
overlapped with the signal from the CellMask Orange stain, leading us to infer that the label was located on the inner
8

membranes (EPR, Golgi apparatus, and endosomes). After 18 hours the labeled ursolic acid signal localization started to
coincide with the MitoTracker Red, suggesting that the preparation was binding with mitochondrial receptors. Finally,
after 24 hours the signal could be identified not only associated with the inner membranes and mitochondria, but also
within the nucleus. Image processing with Zen (Carl Zeiss) software visualized overlapping of the signals from the
labeled ursolic acid, CellMask Orange and MitoTracker Red, which confirmed their colocalization (Fig. 11, 12).
Fig. 11. Signal colocalization of UA* and CellMask in MCF-7 cells after 24 h of incubation: A) 3D visualization of
CellMask distribution; B) 3D visualization of UA* distribution; C) merged 2D visualization, white color marks
overlapping signals from UA* and CellMask Orange. The nuclei are colored with DAPI.
Fig. 12. Signal colocalization of UA* and MitoTracker Red in MCF-7 cells after 24 h of incubation: A) 3D
visualization of MitoTracker Red distribution; B) 3D visualization of UA* distribution; C) merged 2D visualization,
white color marks overlapping signals from UA* and MitoTracker Red. The nuclei are colored with DAPI.
We have performed a model experiment to obtain labeled ursolic acid and to study its biological activity. The aim of
this was to test the effectiveness of a new approach for studying the mechanism of the cytotoxic effect of the ursane
series of pentacyclic triterpenic acids.
Ursolic acid has been shown to be capable of causing apoptosis both externally in its association with the cellular
membranes, and by action on the mitochondria [33–36]. It is also capable of suppressing the growth of cancerous cells
through mobilization of AMP-activated protein kinase [37]. We found out that ursolic acid can also initiate another type
of programmed cell death – autophagy, and thus it can inhibit the growth of cancer cells [38,39]. However, there is
evidence from experiments on PC3 cancer cells of the prostate gland that suggests autophagy could be considered to be
a cell defense reaction to the induction of apoptosis by ursolic acid [40]. Adding autophagy inhibitors intensified the
apoptosis. As a result, of this work the authors conclude that if ursolic acid is to be used as an anti-neoplastic agent, it is
necessary to add autophagy inhibitors.
With the help of bioinformatic approaches we predicted a number of targets, in particular, Akt1 and MDM2, for the
ursolic acid [7,41–43]. There is other experimental data that has demonstrated the impact of ursolic acid on these
proteins [34,44]. In the cell, Akt1 is localized mostly on the inner membranes [45], while MDM2 is located in the
9

nucleoli [46,47]. Experimental tracing of the dynamics of penetration and distribution of the labeled ursolic acid has
shown that when the acid enters the cell, it initially localizes on the inner membranes where the predicted target Akt1 –
a protein that inhibits apoptosis – is located. Its inhibition by ursolic acid may induce apoptosis in the cancerous cells,
followed by breaching of the mitochondrial membrane and the integrity of the cell nucleus. The signal present in the
mitochondria 18 hours after treatment may witness that they are vulnerable to the labeled ursolic acid because of the
developing apoptotic changes in the cell. However, such an assumption is not exclusive of the possibility that these
membranes may have their own targets for the compound being tested, but these were not included in the scope of this
research. The penetration of the label into the nucleus after 24 hours may be interpreted as an integrity breach of the
nuclear membrane of the type that is typical of apoptosis. We found no localization of labeled ursolic acid typical of
binding with nucleoli: the signal was equidistributed across the nucleus, which indirectly evidences a lack of interaction
between the labeled ursolic acid and MDM2. The results obtained are in agreement with, and complement, the available
data.
3. Experimental section
3.1. Molecular Docking and Dynamics
The macromolecules and ligands were prepared with the use of the UCSF Chimera 1.10.2 software program [48], the
structure of the ursolic acid was obtained from the PubChem database (CID 64945). We used the MMFF94 force field
to obtain their three-dimensional structures [49]. The molecular docking was performed in AutoDock Vina [50] using
the PyRx 0.8 graphical interface. In order to visualize the results, Accelrys Discovery Studio Visualizer 4.0 software
was used.
As potential targets for the triterpenic acids we chose Akt1/protein kinase B (PDB ID: 3OCB) [51] and MDM2/ E3
ubiquitin-protein ligase (PDB ID: 4HG7) [52] (referent ligands were (2S)-2-(4-chlorobenzyl)-3-oxo-3-[4-(7H-
pyrrolo[2,3-d]pyrimidin-4-yl)piperazin-1-yl]propan-1-amine (XM1) and 4-({(4S,5R)-4,5-bis(4-chlorophenyl)-2-[4-
methoxy-2-(propan-2-yloxy)phenyl]-4,5-dihydro-1H-imidazol-1-yl}carbonyl)piperazin-2-one (nutlin 3a) respectively).
The docking procedure was performed with the use of standard AutoDock Vina parameters using a strict conformation
of the macromolecule but a flexible conformation of the ligand. The conformations with the lowest values of minimal
binding energy from the point of view of evaluating the docking function have been chosen for the visualization.
10

Molecular dynamics simulations were performed using the NAMD 2.13 package [53]. Ligand-enzyme complexes with
the lowest binding energy were selected for MD. The ligand parameters were analyzed on the SwissParam online server
[54] using MMFF [55] and CHARMM22 [56] forcefields. Complexes preparation, solvation, neutralization and
analysis were carried out using VMD program [57]. Energy minimization was performed for 1000 steps using the
steepest descent method [58]. The complexes were heated to 300K and equilibrated at the NVT ensemble for 100 ps at
300K. Finally, 10 ns MD simulations were performed at the NPT canonical ensemble under periodic boundary
conditions. The trajectory of each complex was analyzed using VMD RMSD Trajectory Tool.
3.2. Synthesis of Ursolic Acid Labeled with FITC
The ursolic acid used was manufactured by Aldrich, and the FITC, by Fluka. All other reagents were purchased from
ReaChem, class CP (Chemically Pure).
The NMR-spectra were obtained with a Bruker AV-400 (1H: 400.13 MHz, 13C: 100.78 MHz) or Bruker AV-600
spectrometer (1H: 600.30 MHz, 13C: 150.95 MHz) (Bruker BioSpin GmbH, Rheinstetten, Germany), using tetramethyl
silane as the external standard. The melting points were determined on a Stuart SMF-38 (Bibby Scientific,
Staffordshire, UK). The elemental analysis was carried out with the use of a Carlo-Erba analyzer (Carlo-Erba, Milan,
Italy). The reaction progress was monitored using TLC on Siluf l UV-254 plates (Kavalier, Czech Republic), the
chromatographic separation being carried out on silica gel 60 (0.063–0.200 mm, Merck KGaA, Darmstadt, Germany).
The epimers of 3α- and 3β-amino-urs-12(13)-en-28-oic acid 2 and 3 were prepared according to established technique;
their spectral data completely corresponding to the data stated in the paper [30].
Compounds 5 and 6 were obtained by reacting the corresponding epimers: (20 mg (0.04 mM)) to 10 ml of methanol we
added 17 mg of fluorescein isothiocyanate 4 and stirred the solution for 12 h at room temperature in the dark. Then the
solution was evaporated under vacuum, and the residue further dried. We obtained the products 5 and 6 with yields of
88 and 82% respectively.
α
-10-(3-(3-carboxy-4-(6-hydroxy-3-oxo-3H-xanthen-9-yl)phenyl)thioureido)-1,2,6a,6b,9,9,12a-heptamethyl-
1,2,3,4,4a,5,6,6a,6b,7,8,8a,9,10,11,12,12a,12b,13,14b-icosahydropicene-4a-carboxylic acid 5.. Yield 88%. white
powder. Rf 0.33 (hexan: methyl tert butyl ether 5:1) M.p. 259-2600C (diethylic ether). IR, ν, см-1 : 3400, 3234, 3068,
2927, 2871, 2618, 2113, 2028, 1731, 1695, 1635, 1606, 1525, 1506, 1457, 1378, 1311, 1295, 1255, 1209, 1178, 1112,
1081, 1035, 995, 852. 1H NMR (CDCl3+CD3OD, 400 MHz) δ 0.65 (3H, s, Me-25), 0.78 (3H, s, Me-26), 0.80 (3H, s,
Me-27), 0.90 (3H, s, Me-24), 1.07 (3H, s, Me-23), 1.56 (3H, s, Me-30), 1.89 (1H, m, H33), 2.35 (1H, m, H19), 5.44
11

(4H, m, H31, H29), 5.65 (s, 1H, H18’), 6.90 (d, 1H, H12’), 7.44, 7.47 (s, 2H, H11’,H5’), 7.49 (d, 1H, ,H21’), 7.72 (d,
1H, H4’), 7.95 (s, 1H, H14’), 8.13 (s, 1H, H8’), 8.87 (d, 1H, H20’). 13C NMR (CDCl3+CD3OD, 100 MHz) δ 15.04,
16.42, 17.83, 20.48, 22.94, 23.72, 26.19 (7*CH3 23,24,25,26,27,29,30), 19.19 (C6), 22.78 (C11), 23.62 (C16), 26.21
(C15), 27.48 (C2), 28.19 (C21), 30.16 (C7), 32.59 (C22), 35.89 (C19), 36.36 (C1), 38.20 (C10), 38.50 (C8), 38.66
(C20), 39.00 (C14), 46.64 (C17), 47.11 (C4), 48.93 (C9), 52.40 (C5), 54.85 (C18), 102.01 (C8’), 112.29 (C6’), 116.25
(C10’), 121.55 (C2’’), 124.61 (C4a’), 125.07 (C12), 127.31 (C3’’), 127.38 (C4’’), 127.72 (C1’’), 128.22 (C3’), 128.69
(C2’), 129.07 (C5’’), 130.64 (C5’), 137.74 (C6’’), 140.29 (C4’), 143.37 (C13), 145.37 (C3a’), 159.67 (C7’), 159.84
(C9a’), 160.20 (C3), 160.36 (C8a’), 171.28 (C7’’), 179.19 (C28), 180.16 (C1’), 189.15 (NH=(C=S)=NH). Anal. calcd.
for C51H60N2O7S C(72.48%) H(7.16%) N(3.31%) S(3.79%) found C(72.16%) H(6.93%) N(3.22%) S(3.60%). MS
m/z calcd. 845.0963 found 845.1144
β
-10-(3-(3-carboxy-4-(6-hydroxy-3-oxo-3H-xanthen-9-yl)phenyl)thioureido)-1,2,6a,6b,9,9,12a-heptamethyl-
1,2,3,4,4a,5,6,6a,6b,7,8,8a,9,10,11,12,12a,12b,13,14b-icosahydropicene-4a-carboxylic acid 5. Yield 82%. white
powder. Rf 0.31 (hexan: methyl tert butyl ether 5:1) M.p. 229-2300C (diethylic ether). IR, ν, см-1 : 3413, 3403, 3111,
3025, 2917, 2870, 2628, 2110, 2008, 1731, 1695, 1635, 1604, 1524, 1506, 1437, 1375, 1311, 1295, 1245, 1207, 1178,
1112, 1081, 1035, 995, 852. 1H NMR (CDCl3+CD3OD, 400 MHz) δ 0.68 (3H, s, Me-25), 0.74 (3H, s, Me-26), 0.80
(3H, s, Me-27), 0.89 (3H, s, Me-24), 1.01 (3H, s, Me-23), 1.54 (3H, s, Me-30), 1.85 (1H, m, H33), 2.34 (1H, m, H19),
5.41 (4H, m, H31, H29), 5.68 (s, 1H, H18’), 6.90 (d, 1H, H12’), 7.42, 7.47 (s, 2H, H11’,H5’), 7.48 (d, 1H, ,H21’), 7.65
(d, 1H, H4’), 8.01 (s, 1H, H14’), 8.18 (s, 1H, H8’), 8.82 (d, 1H, H20’). 13C NMR (CDCl3+CD3OD, 100 MHz) δ 15.06,
16.44, 17.35, 20.39, 22.92, 23.72, 26.19 (7*CH3 23,24,25,26,27,29,30), 19.20 (C6), 22.75 (C11), 23.60 (C16), 26.21
(C15), 27.48 (C2), 28.19 (C21), 30.14 (C7), 32.55 (C22), 35.89 (C19), 36.32 (C1), 38.46 (C10), 38.51 (C8), 38.60
(C20), 39.00 (C14), 46.62 (C17), 47.11 (C4), 49.03 (C9), 52.41 (C5), 54.80 (C18), 102.03 (C8’), 112.29 (C6’), 116.25
(C10’), 121.46 (C2’’), 123.61 (C4a’), 125.06 (C12), 127.13 (C3’’), 127.38 (C4’’), 127.65 (C1’’), 128.12 (C3’), 128.69
(C2’), 129.07 (C5’’), 130.63 (C5’), 137.74 (C6’’), 140.26 (C4’), 143.17 (C13), 145.35 (C3a’), 159.67 (C7’), 159.84
(C9a’), 160.21 (C3), 160.34 (C8a’), 171.25 (C7’’), 179.19 (C28), 180.16 (C1’), 188.97 (NH=(C=S)=NH). Anal. calcd.
for C51H60N2O7S C(72.48%) H(7.16%) N(3.31%) S(3.79%) found C(72.36%) H(7.13%) N(3.46%) S(3.30%). MS
m/z calcd. 845.0963 found 845.1085.
3.3. MTT-assay
12

MTT-assay was performed according to standard procedures [59]. The cell cultures were grown on IMDM (with 3.02g
NaHCO3 per liter of medium) containing 10% bovine serum. To each well of a 96-well plate we added 100 μl of the
cell suspension (density 106 cells per ml); the plate was placed in an incubator containing 5% СО2 at a temperature of
37 ⁰С. After 24 h the test compounds were added at 1 μl per well. The plate was then returned to the incubator for a
further 72 hours. Then, 10 μl of MTT (5 mg of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (МТТ)
per 1 ml of PBS) was added to each well. 3 hours later the supernatant fluid was removed from the wells, and the
precipitate was dissolved in 100 μl of isopropanol. The amount of precipitate was determined on a Multiskan RC plate
spectrophotometer, using the inbuilt Genesis software (LabSystems). Percent inhibition was calculated by the formula:
Inhibition(%) = 100 – [(T/M)×100],
T is the optical density of the precipitate with the test compound, M is the optical density of the precipitate with DMSO
(used as a solvent, negative control). The experiment was repeated independently three times, with 3 replications in
each. Average values were reported for all repeated experiments; the data were analyzed for significance using
Wilcoxon rank test.
3.4. Detection of Apoptosis
Detection of Apoptosis was carried out using double staining with acridine orange/ethidium bromide. Into a 12-well
plate we added 1 ml of cells in culture medium (density 106 cells per ml). This was incubated for 24 h to allow
adhesion, and then treated with the test compounds. Assays were carried out at 24 and 48 hours using standard
procedures [60] with subsequent analysis on an Axio Observer (Carl Zeiss) microscope. The experiment was repeated
independently three times, with 3 replications in each. Average values were reported for all repeated experiments; the
data were analyzed for significance using Wilcoxon rank test.
3.5. Cell IQ System
A Cell IQ System (CM Technologies, Finland) was used for in vivo observation of the cells over a 72 hour period with a
recording frequency of 6 images per hour. The cultures were grown in IMDM (with 3.02 g of NaHCO3 per liter of
medium) containing 10% bovine serum. 100 μl of the cell suspension (density 106 cell per ml) was placed into each
well of a 96-well plate and this was then incubated at 37 ⁰С in 5% СО2. 24 h later the test compounds were added at 1
13

μl per well. The plate was then placed into the Cell-IQ system for 72 h, with cell imaging being recorded every 10
minutes, using a Nikon CFI Plan Fluorescence DL 10x objective lens.
3.6. Confocal Microscopy
The samples for confocal microscopy were prepared as follows: a cover slip was placed into the bottom of each well of
a 12-well plate, 1 ml of cell culture medium and 100 μl of cell suspension (density 106 cell per ml) were then added to
each well. The cells were incubated for 24 hours, and then washed with PBS. After that, to each well was added a
sample of one of the compounds under test, dissolved in cell culture medium at the desired concentration. After a
predetermined time the cover slips were removed from the plate, carefully washed with PBS and put into
paraformaldehyde overnight. Subsequently, each coverslip was placed on a slide with antifade containing DAPI. The
sample was dried with filter paper and the edges of the coverslips were sealed to the slides with lacquer. Microscopic
examination was carried out with the help of an LSM 510 META (Carl Zeiss) confocal scanning microscope at the
Center for Microscopic Analysis of Biological Objects of the Institute of Cytology and Genetics SB RAS. used standard
filters were used for detection of FITC labeled epimeres (490 nm for excitation and 525 nm for emission). The results
were processed using Zen (Carl Zeiss) software.
Conflict of interest
The authors confirm that this article content has no conflict of interest.
Acknowledgements
The authors thank Dr. Alexander Baev for help with the MD similations, and the Center for Collective Use of
Microscopic Analysis of Biological Objects (Institute of Cytology and Genetics, Siberian Branch of Russian Academy
of Science, Novosibirsk, Russia) for providing the equipment.
The research was supported by the Program of fundamental research of the Siberian Branch of the Russian Academy of
Sciences № 0324-2018-0037.
References
[1]
W.N. Setzer, M.C. Setzer, Plant-derived triterpenoids as potential antineoplastic agents., Mini Rev. Med. Chem.
3 (2003) 540–556.
14

[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
N. Sultana, Clinically useful anticancer, antitumor, and antiwrinkle agent, ursolic acid and related derivatives as
medicinally important natural product., J. Enzyme Inhib. Med. Chem. 26 (2011) 616–642.
B.B. Aggarwal, Y. Takada, O. V Oommen, From chemoprevention to chemotherapy: common targets and
common goals., Expert Opin. Investig. Drugs. 13 (2004) 1327–1338.
D. Es-saady, A. Simon, M. Ollier, J.C. Maurizis, A.J. Chulia, C. Delage, Inhibitory effect of ursolic acid on B
16 proliferation through cell cycle arrest, Cancer Lett. 106 (1996) 193–197.
P.O. Harmand, R. Duval, C. Delage, A. Simon, Ursolic acid induces apoptosis through mitochondrial intrinsic
pathway and caspase-3 activation in M4Beu melanoma cells, Int. J. Cancer. 114 (2005) 1–11.
K. a. Manu, G. Kuttan, Ursolic acid induces apoptosis by activating p53 and caspase-3 gene expressions and
suppressing NF-kB mediated activation of bcl-2 in B16F-10 melanoma cells, Int. Immunopharmacol. 8 (2008)
W. Zhang, X. Men, P. Lei, Review on anti-tumor effect of triterpene acid compounds, J. Cancer Res. Ther. 10
(n.d.).
K.-H. Sohn, H.-Y. Lee, H.-Y. Chung, H.-S. Young, S.-Y. Yi, K.-W. Kim, Anti-angiogenic activity of triterpene
acids, Cancer Lett. 94 (1995) 213–218.
C. Cárdenas, A.R. Quesada, M.Á. Medina, Effects of ursolic acid on different steps of the angiogenic process,
Biochem. Biophys. Res. Commun. 320 (2004) 402–408.
P. Skopinski, J. Szaflik, B. Duda-Król, J. Nartowska, E. Sommer, J. Chorostowska-Wynimko, U. Demkow, E.
Skopinska-Rózewska, Suppression of angiogenic activity of sera from diabetic patients with non-proliferative
retinopathy by compounds of herbal origin and sulindac sulfone., Int. J. Mol. Med. 14 (2004) 707–11.
[11] C.-M. Ma, S.-Q. Cai, J.-R. Cui, R.-Q. Wang, P.-F. Tu, M. Hattori, M. Daneshtalab, The cytotoxic activity of
ursolic acid derivatives, Eur. J. Med. Chem. 40 (2005) 582–589.
[12] H. Chen, Y. Gao, A. Wang, X. Zhou, Y. Zheng, J. Zhou, Evolution in medicinal chemistry of ursolic acid
derivatives as anticancer agents, Eur. J. Med. Chem. 92 (2015) 648–655.
[13] Y.-Q. Meng, D. Liu, L.-L. Cai, H. Chen, B. Cao, Y.-Z. Wang, The synthesis of ursolic acid derivatives with
cytotoxic activity and the investigation of their preliminary mechanism of action, Bioorg. Med. Chem. 17
(2009) 848–854.
[14] R. Weissleder, V. Ntziachristos, Shedding light onto live molecular targets., Nat. Med. 9 (2003) 123–128.
[15] M. Sameiro, T. Gonçalves, Fluorescent labeling of biomolecules with organic probes, Chem. Rev. 109 (2009)
15

190–212.
[16] C. Abate, J.R. Hornick, D. Spitzer, W.G. Hawkins, M. Niso, R. Perrone, F. Berardi, Fluorescenct Derivatives of
σ Receptor Ligand 1-Cyclohexyl-4-[3-(5-methoxy-1,2,3,4-tetrahydronaphtalen-1-yl)propyl]piperazine (PB28)
as a Tool for Uptake and Cellular Localization Studies in Pancreatic Tumoe Cells, J Med Chem. 54 (2011)
5858–5867.
[17] K.W. Crawford, W.D. Bowen, Sigma-2 receptor agonists activate a novel apoptotic pathway and potentiate
antineoplastic drugs in breast tumor cell lines., Cancer Res. 62 (2002) 313–322.
[18] M.S. Ostenfeld, N. Fehrenbacher, M. Hoyer-Hansen, C. Thomsen, T. Farkas, M. Jaattela, Effective Tumor Cell
Death by σ-2 Receptor Ligand Siramesine Involves Lysosomal Leakage and Oxidative Stress, Cancer Res. 65
(2005) 8975–8983.
[19] H. Kashiwagi, J.E. McDunn, P.O. Simon, P.S. Goedegebuure, S. Vangveravong, K. Chang, R.S. Hotchkiss,
R.H. Mach, W.G. Hawkins, Sigma-2 receptor ligands potentiate conventional chemotherapies and improve
survival in models of pancreatic adenocarcinoma., J. Transl. Med. 7 (2009) 24.
[20] J. Levi, Z. Cheng, O. Gheysens, M. Patel, C.T. Chan, Y. Wang, M. Namavari, S.S. Gambhir, Fluorescent
Fructose Derivatives for Imaging Breast Cancer Cells, Bioconjug Chem. 18 (2007) 628–634.
[21] K.R. Barnes, J. Blois, A. Smith, H. Yuan, F. Reynolds, R. Weissleder, L.C. Cantley, L. Josephson, Fate of a
bioactive fluorescent wortmannin derivative in cells., Bioconjug. Chem. 19 (2008) 130–7.
[22] J. Gertsch, F. Feyen, A. Bützberger, B. Gerber, B. Pfeiffer, K.-H. Altmann, Making Epothilones Fluoresce:
Design, Synthesis, and Biological Characterization of a Fluorescent N12-Aza-Epothilone (Azathilone).,
Chembiochem. 10 (2009) 2513–21.
[23] C.L. Hartwig, E.M.W. Lauterwasser, S.S. Manajan, J.M. Hoke, R.A. Cooper, A.R. Renslo, Investigating the
Antimalarial Action of 1,2,4-Trioxolanes with Fluorescent Chemical Probes, J Med Chem. 54 (2011) 8207–
8213.
[24] J. a. Evangelio, M. Abal, I. Barasoain, A. a. Souto, M. Pilar Lillo, a. Ulises Acuña, F. Amat-Guerri, J.M.
Andreu, Fluorescent taxoids as probes of the microtubule cytoskeleton, Cell Motil. Cytoskeleton. 39 (1998) 73–
90.
[25] J.F. Díaz, I. Barasoain, A. a Souto, F. Amat-Guerri, J.M. Andreu, Macromolecular accessibility of fluorescent
taxoids bound at a paclitaxel binding site in the microtubule surface., J. Biol. Chem. 280 (2005) 3928–37.
16

[26] M. Izdebska, A. Klimaszewska-Wiśniewska, D. Lewandowski, J. Marcin Nowak, M. Gagat, A. Grzanka,
Arsenic trioxide preferentially induces nonapoptotic cell deaths as well as actin cytoskeleton rearrangement in
the CHO AA8 cell line, Postep. Hig Med Dosw. 68 (2014) 1492–1500.
[27] S. Wu, N.J. Dovichi, High-sensitivity fluorescence detector fluorescein isothiocyanate derivatives of amino
acids separated by capillary zone electrophoresis., J. Chromatogr. 480 (1989) 141–155.
[28] L. Fülöp, B. Penke, M. Zarándi, Synthesis and fluorescent labeling of beta-amyloid peptides., J. Pept. Sci. 7
(2001) 397–401.
[29] C. Hoffmann, J. Leroy-Dudal, S. Patel, O. Gallet, E. Pauthe, Fluorescein isothiocyanate-labeled human plasma
fibronectin in extracellular matrix remodeling, Anal. Biochem. 372 (2008) 62–71.
[30] C.H. Brieskorn, H. Eschelbach, Glykamine von Ursol- und 18P-Glycyrrhetinsaure, Arch. Pharm. Res. 312
(1979) 752–762.
[31] K. Bowden, M. Heilbron, E.R.H. Jones, B.C.L. Weedon, Acetylenic Ketones by Oxidation of Acetylenic
Carbinols and Glycols, (1946).
[32] T.S. Frolova, A. V. Lipeeva, D.S. Baev, Y.A. Tsepilov, O.I. Sinitsyna, Apoptosis as the main mechanism of the
cytotoxic action of ursolic and grind acids in glioma cells, 51 (2017) 61–68.
[33] K.H. Kim, H.S. Seo, H.S. Choi, I.H. Choi, Y.C. Shin, S.G. Ko, Induction of apoptotic cell death by ursolic acid
through mitochondrial death pathway and extrinsic death receptor pathway in MDA-MB-231 cells, Arch.
Pharm. Res. 34 (2011) 1363–1372.
[34] J. Li, X. Liang, X. Yang, Ursolic acid inhibits growth and induces apoptosis in gemcitabine-resistant human
pancreatic cancer via the JNK and PI3K / Akt / NF- κ B pathways, Oncol. Rep. 28 (2012) 501–510.
[35] S. Saraswati, S.S. Agrawal, A.A. Alhaider, Chemico-Biological Interactions Ursolic acid inhibits tumor
angiogenesis and induces apoptosis through mitochondrial-dependent pathway in Ehrlich ascites carcinoma
tumor, Chem. Biol. Interact. 206 (2013) 153–165.
[36] Y. Li, X. Lu, H. Qi, X. Li, X. Xiao, J. Gao, Ursolic Acid Induces Apoptosis Through Mitochondrial Intrinsic
Pathway and Suppression of ERK1 / 2 MAPK in HeLa Cells, 210 (2014) 202–210.
[37] H. Son, H.Y. Kwon, E.J. Sohn, J. Lee, H. Woo, M. Yun, S. Kim, Y. Kim, Activation of AMP-activated Protein
Kinase and Phosphorylation of Glycogen Synthase Kinase3 b Mediate Ursolic Acid Induced Apoptosis in
HepG2 Liver Cancer Cells, (2013).
17

[38] S. Shen, Y. Zhang, R. Zhang, X. Tu, X. Gong, Ursolic acid induces autophagy in U87MG cells via ROS-
dependent endoplasmic reticulum stress, Chem. Biol. Interact. (2014).
[39] S. Leng, Y. Hao, D. Du, S. Xie, L. Hong, H. Gu, X. Zhu, J. Zhang, D. Fan, H.K. Kung, Ursolic acid promotes
cancer cell death by inducing Atg5-dependent autophagy, Int J Cancer. 133 (2013) 2781–2790.
[40] S.W. Shin, S.Y. Kim, J. Park, Biochimica et Biophysica Acta Autophagy inhibition enhances ursolic acid-
induced apoptosis in PC3 cells, BBA - Mol. Cell Res. 1823 (2012) 451–457.
[41] W. He, F. Shi, Z.-W. Zhou, B. Li, K. Zhang, X. Zhang, C. Ouyang, S.-F. Zhou, X. Zhu, A bioinformatic and
mechanistic study elicits the antifibrotic effect of ursolic acid through the attenuation of oxidative stress with
the involvement of ERK, PI3K/Akt, and p38 MAPK signaling pathways in human hepatic stellate cells and rat
liver., Drug Des. Devel. Ther. 9 (2015) 3989–4104.
[42] D. Kashyap, H.S. Tuli, A.K. Sharma, Ursolic acid (UA): A metabolite with promising therapeutic potential,
Life Sci. 146 (2016) 201–213.
[43] J. Iqbal, B.A. Abbasi, R. Ahmad, T. Mahmood, S. Kanwal, B. Ali, A.T. Khalil, S.A. Shah, M.M. Alam, H.
Badshah, Ursolic acid a promising candidate in the therapeutics of breast cancer: Current status and future
implications, Biomed. Pharmacother. 108 (2018) 752–756.
[44] X. Zhang, X. Song, S. Yin, C. Zhao, L. Fan, H. Hu, p21 induction plays a dual role in anti-cancer activity of
ursolic acid., Exp. Biol. Med. (Maywood). 241 (2016) 501–8.
[45] A. Dufner, M. Andjelkovic, B.M. Burgering, B.A. Hemmings, G. Thomas, Protein kinase B localization and
activation differentially affect S6 kinase 1 activity and eukaryotic translation initiation factor 4E-binding protein
1 phosphorylation., Mol. Cell. Biol. 19 (1999) 4525–34.
[46] M.A. Lohrum, M. Ashcroft, M.H. Kubbutat, K.H. Vousden, Identification of a cryptic nucleolar-localization
signal in MDM2., Nat. Cell Biol. 2 (2000) 179–81.
[47] Q. Yang, L. Liao, X. Deng, R. Chen, N.S. Gray, J.R. Yates, J.D. Lee, BMK1 is involved in the regulation of
p53 through disrupting the PML-MDM2 interaction., Oncogene. 32 (2013) 3156–64.
[48] E.F.. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt, E.C. Meng, T.E. Ferrin, UCSF
Chimera - a visualization system for exploratory research and analysis, J Comput Chem. 25 (2004) 1605–1612.
[49] T.A. Halgren, Merck molecular force field. I. Basis, form, scope, parameterization, and performance of
MMFF94, J. Comput. Chem. 17 (1996) 490–519.
18

[50] O. Trott, A.J. Olson, AutoDock Vina: Improving the Speed and Accuracy of docking with a New Scoring
Function, Efficient Optimization, and Multithreading, J Comput Chem. 31 (2010) 455–461.
[51] J.F. Blake, N.C. Kallan, D. Xiao, R. Xu, J.R. Bencsik, N.J. Skelton, K.L. Spencer, I.S. Mitchell, R.D.
Woessner, S.L. Gloor, T. Risom, S.D. Gross, M. Martinson, T.H. Morales, G.P. Vigers, B.J. Brandhuber,
Discovery of pyrrolopyrimidine inhibitors of Akt, Bioorg Med Chem Lett. 20 (2010) 5607–5612.
[52] B. Anil, C. Riedinger, J.A. Endicott, M.E. Noble, The structure of an MDM2-Nutlin-3a complex solved by the
use of a validated MDM2 surface-entropy reduction mutant, Acta Crystallogr. 69 (2013) 1358–1366.
[53] J.C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot, R.D. Skeel, L. Kalé, K.
Schulten, Scalable molecular dynamics with NAMD, J. Comput. Chem. 26 (2005) 1781–1802.
[54] V. Zoete, M.A. Cuendet, A. Grosdidier, O. Michielin, SwissParam: a fast force field generation tool for small
organic molecules., J. Comput. Chem. 32 (2011) 2359–68.
[55] T.A. Halgren, Merck molecular force field. I. Basis, form, scope, parameterization, and performance of
MMFF94, J. Comput. Chem. 17 (1996) 490–519.
[56] A.D. MacKerell, D. Bashford, M. Bellott, R.L. Dunbrack, J.D. Evanseck, M.J. Field, S. Fischer, J. Gao, H. Guo,
S. Ha, D. Joseph-McCarthy, L. Kuchnir, K. Kuczera, F.T. Lau, C. Mattos, S. Michnick, T. Ngo, D.T. Nguyen,
B. Prodhom, W.E. Reiher, B. Roux, M. Schlenkrich, J.C. Smith, R. Stote, J. Straub, M. Watanabe, J.
Wiórkiewicz-Kuczera, D. Yin, M. Karplus, All-atom empirical potential for molecular modeling and dynamics
studies of proteins., J. Phys. Chem. B. 102 (1998) 3586–616.
[57] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics., J. Mol. Graph. 14 (1996) 33–8, 27–8.
[58] H. Attouch, R. Cominetti, A Dynamical Approach to Convex Minimization Coupling Approximation with the
Steepest Descent Method, J. Differ. Equ. 128 (1996) 519–540.
[59] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and
cytotoxicity assays., J. Immunol. Methods. 65 (1983) 55–63.
[60] S. Kasibhatla, D. Amarante-Mendes, G. P. Finucane, T. Brunner, E. Bossy-Wetzel, D.R. Green, Acridine
Orange/Ethidium Bromide (AO/EB) Staining to Detect Apoptosis, Cold Spring Harb. Protoc. (n.d.).
19

20

21

22

23

24

25

26

27

28

29