In vitro anti-glioblastoma activity of L-valine derived boroxazolidones

Article abstract of DOI:10.1016/j.ejphar.2019.04.020

In the present study, a series of L-valine derived boroxazolidones, previously synthesized and reported to have residual activity in a human epithelial cell line, have been evaluated in vitro for their anti-glioblastoma activity. A boroxazolidone derivative containing 2,4-difluorophenyl moieties (6) was found to have higher cytotoxicity than the standard drug, Temozolomide (TMZ). Compound 6 was found to exhibit dose-dependent growth inhibitory effects with an IC₅₀ of 49 μM and 53 μM for LN229 and SNB19 cells, respectively. Additionally, 6 was assessed for its role in apoptosis, caspase 3/7 activation and oxidative stress in SNB19 and LN229 cells. SNB19 cells treated with 6 showed 45.3% apoptosis in the population, while TMZ had 24.7%. In LN229 cells, the percentage of apoptotic cells treated with compound 6 and TMZ were the same. Both 6 and TMZ induced apoptosis through the activation of caspase 3/7 in SNB19 and LN229 cells. Interestingly, 6 exhibited a higher effectivity in promoting reactive oxygen species production in LN229, while it was 6-fold less in SNB19. Boroxazolidone-treated GBM cell lines increased reactive oxygen species production, suggesting that such species may be interlinked with the observed programmed cell death. Additionally, the treatment of both GBM cell lines with 6 led to G2/M phase arrest. The magnitude of anti-GBM effect of 6 is significantly higher than the known chemotherapeutic agent TMZ. This work further demonstrates the anticancer properties of L-valine derived boroxazolidones, adding another potential derivative to the collection of promising chemotherapeutic agents for GBM treatment.

Full text of DOI:10.1016/j.ejphar.2019.04.020

Accepted Manuscript
In vitro anti-glioblastoma activity of L-valine derived boroxazolidones
Anisha Viswanathan, Giulia Sebastianelli, Kenna Brown, Janne Raunio, Vili Sipilä, Olli
Yli-Harja, Nuno R. Candeias, Meenakshisundaram Kandhavelu
PII:
S0014-2999(19)30248-1
https://doi.org/10.1016/j.ejphar.2019.04.020
EJP 72339
DOI:
Reference:
To appear in: European Journal of Pharmacology
Received Date: 24 January 2019
Revised Date: 5 April 2019
Accepted Date: 8 April 2019
Please cite this article as: Viswanathan, A., Sebastianelli, G., Brown, K., Raunio, J., Sipilä, V.,
Yli-Harja, O., Candeias, N.R., Kandhavelu, M., In vitro anti-glioblastoma activity of L-valine
derived boroxazolidones, European Journal of Pharmacology (2019), doi: https://doi.org/10.1016/
j.ejphar.2019.04.020.
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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
In vitro anti-glioblastoma activity of L-valine derived boroxazolidones
Anisha Viswanathan1, Giulia Sebastianelli1, Kenna Brown1, Janne. Raunio2, Vili Sipilä1 Olli Yli-Harja3,4,
Nuno R. Candeias2,* and Meenakshisundaram Kandhavelu1,*
1Molecular Signaling Lab, Faculty of Medicine and Health Technology, Tampere University and
BioMediTech, P.O. Box 553, 33101 Tampere, Finland.
2Faculty of Engineering and Natural Sciences, Tampere University, Korkeakoulunkatu 8, 33101
Tampere, Finland
3Computational Systems Biology Group, Faculty of Medicine and Health Technology, Tampere
University and BioMedi Tech, P.O. Box 553, 33101 Tampere, Finland.
4Institute for Systems Biology, 1441N 34th Street, Seattle, WA 98103-8904, USA
* E-mail: nuno.rafaelcandeias@tut.fi and meenakshisundaram.kandhavelu@tut.fi
*Phone: (+358)417488772
Abstract
In the present study, a series of L-valine derived boroxazolidones, previously synthesized and
reported to have residual activity in a human epithelial cell line, have been evaluated in vitro for their
anti-glioblastoma activity. A boroxazolidone derivative containing 2,4-difluorophenyl moieties (6) was
found to have higher cytotoxicity than the standard drug, Temozolomide (TMZ). Compound 6 was
found to exhibit dose-dependent growth inhibitory effects with an IC50 of 49 µM and 53 µM for LN229
and SNB19 cells, respectively. Additionally, 6 was assessed for its role in apoptosis, caspase 3/7
activation and oxidative stress in SNB19 and LN229 cells. SNB19 cells treated with 6 showed 45.3 %
apoptosis in the population, while TMZ had 24.7 %. In LN229 cells, the percentage of apoptotic cells
treated with compound 6 and TMZ were the same. Both 6 and TMZ induced apoptosis through the
activation of caspase 3/7 in SNB19 and LN229 cells. Interestingly, 6 exhibited a higher effectivity in
promoting reactive oxygen species production in LN229, while it was 6-fold less in SNB19.
Boroxazolidone-treated GBM cell lines increased reactive oxygen species production, suggesting that
such species may be interlinked with the observed programmed cell death. Additionally, the treatment
of both GBM cell lines with 6 led to G2/M phase arrest. The magnitude of anti-GBM effect of 6 is
significantly higher than the known chemotherapeutic agent TMZ. This work further demonstrates the

ACCEPTED MANUSCRIPT
anticancer properties of L-valine derived boroxazolidones, adding another potential derivative to the
collection of promising chemotherapeutic agents for GBM treatment.
KEYWORDS: Apoptosis, boroxazolidone, cell cycle arrest, cytotoxicity, glioblastoma.
1. Introduction
Glioblastoma (GBM) is often treated with the combination of chemotherapy and surgery. However, the
application of anticancer drugs for chemotherapy is limited by deleterious side effects, limited activity
in vivo, and developed drug resistance. Thus, novel, broad-spectrum drugs with less severe side
effects are necessary to improve cancer treatment (Doan et al., 2016). Treatment of various cancers
has been demonstrated using boron derivatives (Tülüce et al., 2017). The identification of alkyl or aryl
boron-containing compounds as serine protease inhibitors(Philipp and Bender, 1971), and the role of
this enzymes in cancer progression(Poddar et al., 2017), has triggered several studies on the
antitumor properties of boron-containing compounds(Baker et al., 2009). Ultimately, several classes
of boron-containing compounds were disclosed to have the ability to reduce tumor cell proliferation
and initiate apoptosis(I. Scorei and Popa, 2010). In 2002, boron-containing Bortezomib became the
first proteasome inhibitor to be approved by the FDA for multiple myeloma(Paramore and Frantz,
2003). Increased target cell specificity resulting from functionalizing boron with other chemical groups
(Żołnierczyk et al., 2016) could decrease the deleterious side effects of anticancer drugs, as they
stem from cytotoxic effects towards healthy cells. By reducing chemotherapeutic drug side effects,
cancer treatment could be improved (Andrews et al., 1988).
Amino acid-based compounds are stable in a biological environment and catabolically resistant, an
advantageous trait for treatment because a high concentration of the compound can be accumulated
in target cells. Prodrugs based on amino acid skeletons can be used to increase cellular uptake, while
maintaining a consistent intracellular concentration of the anticancer agent, despite efflux caused by
various transporters. This strategy has been proven semi-successful with valine prodrug, antitumor
SN-38.However, efficacy decreased due to the low stability of the prodrug (Kwak et al., 2012a,
2012b). Boroxazolidones, formed by condensation of a borane with an α-amino acid, have been
determined to induce cell death in human epithelial colorectal adenocarcinoma and brain astrocytoma
cells (Raunio et al., 2017; Velasco-Bejarano et al., 2007; Velasco et al., 2007) Several methods for

ACCEPTED MANUSCRIPT
the preparation of boroxazolidones have been developed, although most of them rely on reacting an
α-amino acid with borane or borinates.
We and others have previously reported modest antitumor properties of boroxazolidones against
human epithelial colorectal adenocarcinoma, brain astrocytoma, leukaemia and murine lymphoma
cells(Raunio et al., 2017; Velasco-Bejarano et al., 2007; Velasco et al., 2007). However, the
antineoplastic effect of boroxazolidones against the growth of various types of GBM is not well
understood. Hence, this study aims to investigate the effect of selective compounds against GBM that
originates in the parieto-occipital region. The cytotoxic effects of novel derivatives were tested on two
GBM cell lines, LN229 and SNB19. Herein we present the results of the in vitro cytotoxicity evaluation
of L-valine derived boroxazolidones for potential use in GBM treatment. A normal cell (Mouse
Embryonic Fibroblast, MEF) line was tested to support that the compound targeted GBM cells. To
determine the potential broad-spectrum application of the compound, many in vitro studies were
performed, including dose-dependent GBM growth inhibition assays. Cell death mechanism was
ascertained via apoptosis screening, caspase 3/7, and reactive oxygen species assays, and cell cycle
analysis was performed to determine the inhibition of cellular division.
2. Materials and methods
2.1. Chemicals
Dulbecco’s Modified Eagle Medium (DMEM) (Sigma-Aldrich, St. Louis, MO), Fetal Bovine Serum
(FBS) (Sigma-Aldrich, St. Louis, MO, #F7524), L-glutamine (Sigma-Aldrich, St. Louis, MO #G7513),
penicillin and streptomycin (Sigma-Aldrich, St. Louis, MO, P4333), Amphotericin B (Sigma-Aldrich, St.
Louis, MO, #A9528), Trypan Blue Solution (Sigma-Aldrich, St. Louis, MO, #T8154), Propidium Iodide
(Molecular Probes by Life Technologies, Eugene, OR, #P3566), Trypsin-EDTA solution (Sigma-
Aldrich, St. Louis, MO, #59418C), Dead Cell Apoptosis Kit with Annexin V FITC and PI (Thermo
Fisher Scientific), Reactive Oxygen Species Detection Reagent (Molecular Probes by Life
Technologies, Eugene, OR, #C13293), Caspase-Glo® 3/7 Assay kit (Promega, Madison USA),
Dimethyl Sulphoxide (DMSO) Hybri-Max (Sigma-Aldrich, St. Louis, MO, #D2650), Temazolamide
(TMZ) (Sigma-Aldrich, USA).
2.2.Chemical preparation

ACCEPTED MANUSCRIPT
Boroxazolidones 1-6 compounds and TMZ, a chemotherapeutic agent for glioblastoma treatment,
were dissolved in DMSO to obtain a final concentration of 100 mM, from which, intermediate dilutions
were prepared. For the dose dependent cytotoxicity assay, the final concentrations were prepared in
the same culture medium to concentrations of 100 µM, 75 µM, 50 µM, 25 µM, and 10 µM.
2.3.Cell culture
The human glioblastoma cell lines SNB19 and LN229 (ATCC® CRL-2611™/ CRL-2611™)
and mouse embryonal fibroblast (MEF, normal brain cells) were cultured in DMEM
supplemented with 10% FBS, 0.1 mg/ml Streptomycin, 100 U/ml Penicillin, and 0.025
mg/ml Amphotericin B. Cell cultures were incubated at 37 °C supplemented with 5 % CO2
in a humidified incubator. The LN229 cell line was derived from a patient with right frontal
parieto-occipital GBM. The line exhibits mutated p53 proteins. SNB19 is a malignant GBM
cell line initiated from the surgical resection of a left parietooccipital GBM and also exhibits
mutated p53 proteins. The non-cancerous cells, MEFs, have an E10.5 genotype with Vin+/+
(Vinculin) originally obtained from Wolfgang H.Ziegler (Hannover Medical School,
Hannover, Germany)(Xu et al., 1998).
2.4.Trypan blue exclusion assay for detecting cells viability
The cytotoxic effects of boron derivatives, reported in Fig 1, were studied on multiple glioblastoma cell
lines, SNB19 and LN229. For this, cells were seeded on 12-well plates with a density of 1 X 105 cell/
well. After 48 h, the cells were treated with 100 µM concentration of the compounds of interest. TMZ
and DMSO (0.1%) were used, respectively, as a positive control (PC) and negative control. Treated
cells were collected after trypsinization, and followed by centrifugation at 153 g for 10 min. After
centrifuging, the cells were resuspended in culture medium, and stained with Trypan blue dye at a
ratio of 1:1. To quantify live cell and dead cell populations, a Bürker hemocytometer (Heinz Herenz,
Hamburg, Germany) was used. Biological and technical repeats were used to obtain the statistically
significant results. The percentage of cell viability inhibition was calculated according to the following
equation:
ꢆ
ꢇ
ꢈꢉꢊꢋ ꢌꢍ.ꢍꢎ ꢏꢋꢐꢑꢉꢊꢐꢉꢒ ꢓꢉꢔꢔꢕ ꢖꢈꢗꢘ ꢓꢍꢋꢐꢑꢍꢔ ꢙꢈꢉꢊꢋ ꢌꢍ. ꢍꢎ ꢐꢑꢉꢊꢐꢉꢒ ꢓꢉꢔꢔꢕ
ꢆ
ꢇ
ꢀꢁℎꢂꢃꢂꢄꢂꢅꢁ % =
ꢚ100 (1)
ꢆ
ꢇ
ꢈꢉꢊꢋ ꢌꢍ. ꢍꢎ ꢏꢋꢐꢑꢉꢊꢐꢉꢒ ꢓꢉꢔꢔꢕ ꢖꢈꢗꢘ ꢓꢍꢋꢐꢑꢍꢔ

ACCEPTED MANUSCRIPT
2.5.In vitro measurement of dose dependent cytotoxicity
Compound 6 was identified as the most lethal compound from the preliminary cytotoxicity assay,
where it exhibited the highest cytotoxicity effect. Dose dependent cytotoxicity was measured after the
24 h exposure to different concentrations 6; 100, 75, 50, 25, 10 µM concentrations were tested on
SNB19 and LN229 cell lines. After treatment, the cells were collected as described in the Trypan blue
exclusion assay procedure above. TMZ was used as a positive control. The percentage of mortality
was calculated using formula 1. After that, the dose-response curves were plotted using GraphPad
software and half-maximal inhibitory concentration (IC50) was calculated from the fitted curves.
2.6.Quantification of apoptosis by Annexin V-FITC/PI staining
To quantify apoptosis in the compound 6-treated cells, the Dead Cell Apoptosis Kit with Annexin V-
FITC and PI was used. The procedure for the assay followed the manufacturer’s protocol, which is as
follows: Glioblastoma cell lines, SNB19 and LN229, were cultured in 6-wellplates with the initial
density of 0.3 X 106 cells/well using the complete medium. After 24 h, the complete medium was
replaced with serum-free medium and incubated at 37 °C overnight. Post-incubation, the cells were
treated with the IC50 concentration of the compound of interest, TMZ, and DMSO (0.1%). Treated
cultures were incubated for 24 h. Cells were harvested by trypsinization followed by centrifugation at
153 g for 10 min before being washed in ice-cold PBS. The cell pellets were resuspended in ice-cold
1X annexin-binding buffer. Annexin V-FITC and PI working solutions were added to the cell
suspension as suggested by manufacturer protocol. The cells were incubated in dark conditions at
room temperature for 15 min prior to fluorescence measurement. An EVOS imaging system
(ThermoFisher Scientific) with 20X objective magnification was used to observe apoptotic cells, and
images were taken for the quantification of apoptotic and non-apoptotic cell percentages.
2.7.Caspases 3/7 activities assay
Glioblastoma cell lines, SNB19 and LN229, cells were seeded on 96-wellplates with white flat bottoms
at an initial density of 1 X 104 cells/well in complete medium. After 24 h, the medium was replaced
with 100 µl of serum-free medium and incubated overnight. Post–incubation, the cells were treated
with the IC50 concentration of 6 and TMZ. Each well had a total volume of 100 µl, and then incubated
for 5 h. Glo® 3/7 Assay kit (Promega, Madison USA) was used according to the manufacturer’s
protocol to determine the caspase activity. A 96-wellplate containing cells was removed from the

ACCEPTED MANUSCRIPT
incubator and allowed to equilibrate to room temperature for 30 min. The 100 µl of Caspase-Glo
reagent was added to the treated, untreated, and blank wells. The plate was placed on an agitator for
30 seconds at 15 – 25 g. The plate was incubated 1 h in dark conditions. After incubation, the
luminescence signal was measured using a plate-reading luminometer (Fluoroskan Ascent FL,
Thermo Labsystems). The fold change in caspase 3/7 was calculated using formula 2.
ꢢ
ꢙ ꢢ
ꢦꢧꢨꢩꢪ
ꢣꢤꢥꢣ
ꢛꢅꢜꢝ ꢞℎꢟꢁꢠꢡ = ꢢ
(2)
ꢙ ꢢ
ꢦꢧꢨꢩꢪ
ꢫꢬꢩꢣꢭꢬꢧ
Where Ftest is the luminescence readings from the treated wells, Fcontrol is the luminescence readings
from the untreated wells, and Fblank is the luminescence readings from the unstained wells.
2.8.Detection of Intracellular Reactive Oxygen Species
The glioblastoma cell lines, SNB19 and LN229, were grown in 12-well plates with the initial density of
1 X105 cells/well. After 24 h, the complete medium was replaced with serum-free medium and
incubated overnight. Cells were treated with the IC50 concentration of 6 and TMZ, then left in a
humidified incubator at 37° C supplied with 5 % CO2 for 5 h. Cells were harvested by centrifugation at
153 g for 10 min and the cell pellet was suspended in 2 µM 2',7'-dichlorodihydrofluorescein diacetate
(H2DCFDA), and then incubated for 10 min under dark conditions. The cells were centrifuged at 153
g for 10 min, then washed with warm PBS and recovered in warmed complete medium. Cells were
incubated at 37° C for 20 min prior to the fluorescence measurement. Equal volumes of cells in the
96-well plate were used for measuring fluorescence intensity using a plate reader at 485 nm and 538
nm. DMSO was used as a control. The fold change in reactive oxygen species production was
calculated from the fluorescence reading using formula 2.
2.9.Analysis of cell cycle arrest
To study the effect of compound 6 on cell cycle arrest in SNB19 and LN229 lines, cells were cultured
in 6-well plates with the initial density of 0.3 X 106 cells/well. Cells were incubated overnight at 37 ºC
with 5% CO2. After 24 h the complete medium was replaced with serum-free medium. Cultures were
incubated overnight before they were treated with the IC50 concentration of compound 6 for 48 h. After
48 h from the treatment of the cells, the cells were harvested by trypsinzation. Pellet was
resuspended in 1 ml of ice-cold PBS, centrifuged at 153 g for 10 min, and then the cells were
resuspended in PBS. 70 % ice-cold ethanol was added to the cell suspension and incubated at 4° C

ACCEPTED MANUSCRIPT
for 15 min. The suspension was centrifuged at 153 g for 10 min and the pellet was washed with ice-
cold PBS. 200 µl of PBS containing 20 µg/ml PI, 0.2 mg/ml RNase and 0.1 % Triton X-100 was
added, and samples were incubated for 15 min at 37 °C. Fluorescent signals of the cells were
captured using an EVOS imaging system with 10x objective magnification. The images were analyzed
by CellProlifer software to quantify the percentage of cell cycle phases in all treated conditions. The
untreated and DMSO controls were used to compare the cell cycle arrest profiles.
2.10.
Statistical analysis
Results are expressed as the mean ± standard error of mean (S.E.M) of biological and technical
replicates. Statistical significance was evaluated by ANOVA test, when GraphPad Software
(GraphPad Software, USA) was used, or by equal variance t-tests between samples. Results were
considered statistically significant if p < 0.05.
3. Results
3.1.Cytotoxicity effect of L-valine-derived boroxazolidones 1-6
First, L-valine derived boroxazolidones 1-6 were prepared as previously described elsewhere
(Scheme 1(Kuuloja et al., 2011; Raunio et al., 2017)). Triethylammonium tetra-arylborates were
prepared by Grignard reagent addition to NaBF4. This was followed by cation exchange and
condensation with L-valine in overnight refluxing toluene, providing pure boroxazolidones after
precipitation or recrystallization.
The biological study was performed by determining the cytotoxicity of the compounds against
glioblastoma cell line, LN229, using Trypan blue exclusion assay. Compound 6 showed high
cytotoxicity at 100 µM concentration, inhibiting more than 90% of cell growth. Compound 5 induced 44
% cell death, and boroxazolidones 1-4 displayed less cell growth inhibition. The results indicate that 6
had 13% higher cytotoxicity than the positive control, TMZ. Fig 1A demonstrates 6 as the top cytotoxic
compound compared to the cell death percentage of the negative controls of DMSO and untreated
cells. Given the substantial cytotoxic increase of tetrafluorinated derivative 6 when compared with the
other boroxazolidones, subsequent analyses on such compound were carried out. Compared to our
previous findings, the cytotoxic effect of 6 compound was 40 % higher in parieto-occipital GBM. Thus,
this result suggest that 6 has selectivity for parieto-occipital GBM cells. Boroxazolidone 6 was further

ACCEPTED MANUSCRIPT
evaluated against the growth of normal brain cells, mouse embryonal fibroblast (MEF). Different
concentrations of 6 (10 and 100 µM) were tested on MEF cells followed by a Trypan blue exclusion
assay. Fig 1B shows that at 10 µM and 100 µM treatment there was 7.8% and 14% cell death,
respectively. The comparative analysis of inhibition percentage between cancer and non-cancerous
cells treated with these concentrations suggests that compound 6 has cell line specific activity (10 µM,
non-cancer cells vs cancer cells; P = 0.0319 and 100 µM, non-cancer cells vs cancer cells; P <
0.0001). Overall, the cytotoxicity of the boroxazolidones is dependent on the phenyl rings’
substitutions.
3.2.Dynamics of inhibiting effect
LN229 and SNB19 cell lines were treated with drug concentrations of 10, 25, 50, 75, and 100 µM to
characterize the cellular effects of 6’s inhibition in glioblastoma. The identified molecule reduced cell
proliferation in all cell lines tested (Figs 2A and 2B). Phase-contrast microscopy images were
investigated to observe whether the compounds under study exert morphological changes through
cytotoxicity activity. These computer-assisted morphological observations show that the appearance
of the LN229 and SNB19 glioblastoma cells treated with the various concentrations of 6 were
markedly different. It is evident that compound 6-treated cells appeared shrunk in their morphology,
while the untreated cells appeared thin and flattened, typical of attached morphology. The observed
morphological changes were dose-dependent, which can be seen in Fig 2A for 10 µM and 100 µM
treated conditions. From Fig 2, it is apparent that compound 6 affected cell growth in a dose
dependent manner. The IC50 of compound 6 was inferred to be 49 µM and 53 µM, respectively, for
LN229 and SNB19. The same compound exhibited significant cytotoxic effects in both glioblastoma
cells over these concentrations. Effect of TMZ against the growth of Sbn19 and LN229 was also
tested and IC50 was calculated as 87.7614 ± 6.919 and 84.389 ± 2.599 for LN229 and Snb19,
respectively (data not shown). The cytotoxic effect of 6 is significantly higher than TMZ. Hence,
compound 6, which produced inhibitory activity in a dose-dependent manner, was selected for further
analysis.
3.3.Apoptosis mediated cell death induced by compound 6
To determine the apoptotic effect induced by compound 6, Annexin V/PI double staining was
performed. The percentage of apoptotic cells was calculated based on the results of cellular staining.

ACCEPTED MANUSCRIPT
Apoptotic cells presented with Annexin V-FITC positive/PI negative and both Annexin V-FITC/PI
positive. In untreated (Fig 3A, left panel) and DMSO treated (Fig 3A, middle panel) samples, the live
cells exhibited both as Annexin V/PI negative. Apoptotic cells are cells which possess high levels of
Annexin V and low levels of PI expression. From the microscopy images, it is evident that 6 induced
cell death by apoptosis when cells expressed as Annexin V/PI positive (Fig 3A, last panel). Fig 3B
shows the percentage of live and apoptotic cells of LN229 and SNB19 when treated with 6 and TMZ.
Generally, apoptosis induction was observed in both cell lines. Results were normalized against an
untreated control and compared with a positive control. The percentage of apoptotic SNB19 cells
treated with 6 was 45.3% while TMZ treated cells was 24.7% (6 vs TMZ; P < 0.0001). In the case of
LN229 cells, the percentage of apoptotic cells was approximately equal in 6 and TMZ treated
conditions (6 vs TMZ; P < 0.0001). These results demonstrate that both 6 and TMZ trigger apoptosis
mediated cell death in SNB19 and LN229 cells.
3.4.Caspase 3/7 and reactive oxygen species activation by 6
Caspase activation is one of the key signalling pathways leading to apoptosis in cancer cells.
Caspase 3, 7, 8, and 9 are the different types of caspases associated with apoptotic cell death.
However, caspases 3/7 are directly linked to apoptosis through the intrinsic and extrinsic signalling
pathway interactions (Walsh et al., 2008). To examine the molecular mechanism underlying apoptosis
upon the treatment of boroxazolidone 6, caspase 3/7 activity was measured. As shown in Fig 4A,
compound 6 induced the activation of caspases 3/7. In general, caspases 3/7 activity increased in 6
and TMZ treated samples in comparison to the untreated ones. A slight increase in caspase activation
was observed for 6-treated LN229 cells, when compared with treatment with TMZ (6 vs TMZ; P =
0.1271). Approximately 1-fold and 1.2-fold increase in caspase 3/7 was found in LN229 cells treated
with 6 and TMZ, respectively (Fig 4A). A different profile was determined for SNB19 cells, where
similar increases in caspase 3/7 activity levels were observed regardless of the cytotoxic agent used
(6 vs TMZ; p = 0.4519). It is notable that the production of reactive oxygen species in ln229 was
higher than snb19, which is similar to the trend of caspase 3/7 change. This suggests that there is a
possibility of signalling pathway interaction between reactive oxygen species and caspase 3/7. In
addition, both snb19 and ln229 cells expressed moderately higher levels of caspase 3/7 compared to

ACCEPTED MANUSCRIPT
the caspase 3/7 level of cells treated with TMZ. Overall, the data suggests that 6 and TMZ induced
the intrinsic caspase pathway in snb19 and ln229 cells.
Reactive oxygen species are produced in cancer cells and crucial to several cellular processes,
including cell proliferation, metastasis, and angiogenesis. (Clerkin et al., 2008; Giles, 2006; Ushio-
Fukai and Nakamura, 2008; Wu and Hua, 2007). Reactive oxygen species induces apoptosis and cell
cycle arrest in cancer cells during chemotherapeutic treatment (circu and aw, 2010; toler et al., 2006).
in our study, the effects of 6, H2O2 and TMZ on the levels of reactive oxygen species in snb119 and
ln229 cells were assessed using reactive oxygen species production assay (walsh et al., 2008). The
increase of reactive oxygen species in ln229 cells treated with 6 was higher than in H2O2 and TMZ
treated cultures (fig 4b). as seen in fig 4b, ln229 cells treated with 6 increased their reactive oxygen
species production by 5.2-fold and 6.4-fold in comparison with TMZ and H2O2 treatment, respectively
(6 vs TMZ; p < 0.0001 and 6 vs H2O2 p < 0.0001). Interestingly, reactive oxygen species production in
snb19 cells treated with 6 was lower than in H2O2 and TMZ treated samples (6 vs TMZ; p < 0.0001
and 6 vs H2O2 p < 0.0001). However, compared to the untreated condition, there was a 0.2-fold
increase in reactive oxygen species production in snb19. When comparing between ln229 and snb19,
6 had better effectiveness in promoting reactive oxygen species production in ln229, while it was 6-
fold less in snb19. Both cell lines produced higher reactive oxygen species levels when compared to
the untreated condition, suggesting that it could be interlinked with the observed apoptotic cell death
of ln229 and snb19 upon treatment with 6.
3.5.Cell cycle G2/M phases arrest by boroxazolidone 6
Cellular DNA content analysis following propidium iodide (PI) staining is a well-established method for
quantifying the percentage of cells in cell cycle arrest. The deconvolution of the cellular DNA content
frequency and histogram analysis allows observation of the fraction of cells in each cycle phase
(G0/G1, S and G2/M). It is established that DNA damage can induce cell cycle arrest in cancer cells
during chemotherapeutic treatment. In the present study, cell cycle progression was imaged using PI
fluorescent staining and microscopy. Different phases of the cell cycle were determined based on the
signature fluorescence signals corresponding to each cell cycle phase. In this study, similar trends in
cell cycle arrest were observed in both LN229 and SNB19 cell lines after treatment.

ACCEPTED MANUSCRIPT
In LN229 cells, the control (DMSO treated) conditions resulted in the highest number of cells in the
G2/M phases, moderate numbers of cells in the G0/G1 phases, and the least number of cells in the S
phase. When treated with compound 6, the percentage of G2/M phase cells increased, while G0/G1
and S phase cells decreased compared to the control cells, but the overall relationship between the
percentages of cells in each phase was similar to the control (Fig 5A). SNB19 had a similar trend to
LN229; however, G0/G1 and G2/M were switched. Where the percentage of G0/G1 phase cells was
the highest, followed by G2/M phases, and then the S phase. This may be explained by unequal
growth rates in a population of proliferating cells. It is well known than, the LN229 is faster growing
than SNB19(Al-Moujahed et al., 2017; Diao et al., 2019; Howard and Pelc, 1986). Therefore, the
percentage of SNB19 cells in the G0/G1phases was higher than LN229 cells. Suggesting that many
cells were in the resting phase or entering into the G1 phase in LN229 due to fast growth rate. When
SNB19 was treated with 6, the percentage of G2/M phase cells increased by one fold, while the
percentage in other phases decreased (Fig 5B). According to these results, it is concluded that
SNB19 and LN229 cells were arrested at G2/M phase when they were treated with 6.
4. Discussion
Our previous studies have reported the synthesis of several new L-valine derived boroxazolidones
and evaluated their anticancer activity against the growth of human brain astrocytoma and colorectal
adenocarcinoma cells. As a continuation of previous studies, we presently report the biological activity
of L-valine derived boroxazolidones against the growth of human glioblastoma. From a small
collection of six compounds, fluorinated derivative 6 demonstrated superior cytotoxic effects against
GBM cancer cell lines when compared with the other boroxazolidones tested. In addition, the present
study shows that this compound has a higher mortality rate in glioblastoma cells than in colon cancer
cells, therefore, suggesting selective inhibition of the brain tumor cells. This boroxazolidone induced
cell death via apoptosis and caspase 3/7 activation.
Reactive oxygen species, such as hydrogen peroxide, hydroxyl radicals, and superoxide anions are
produced in living cells, and played significant roles in many cellular functions (Pelicano et al., 2004).
increased reactive oxygen species production in cancer cells may lead to genetic mutation, genetic
instability, and result in cellular damage (fang et al., 2009; kovacic and jacintho, 2001; pelicano et al.,
2004; shi et al., 2014; tominaga et al., 2004). Cancer cells usually possess higher level of reactive

ACCEPTED MANUSCRIPT
oxygen species compared to the normal cells, and it was previously reported that boron-containing
molecules increased cellular oxidative stress, thus, inducing apoptosis in many cancer cell types. Our
current observations are in agreement with such reports as higher levels of reactive oxygen species
production were observed in glioblastoma cells, therefore being a plausible cause for apoptotic
pathway induction. Given the pronounced six-fold increase in reactive oxygen species production in
ln229 cells, it is probable that such species are responsible for the selective cytotoxicity compound 6
exhibits. Drugs targeted towards changing the reactive oxygen species levels are important for anti-
GBM drugs, and compound 6 can be considered to be promising in this regard. Although further
pharmacological and biolological studies are required to fully disclose the mechanism of action of
compound 6.
Boroxazolidones-related diaryl borinic acids have been widely demonstrated to be strong inhibitors of
serine proteases (Steiner et al., 1994), of store-operated calcium entry(Ozaki et al., 2013; Suzuki et
al., 2010) and acyl protein thioesterases 1 and 2(Zimmermann et al., 2013). Moreover,
boroxazolidones derived from amino acids other than valine were demonstrated to have an IC50
inhibitory activity on store-operated calcium entry(Ozaki et al., 2013) as low as 0.2 µM. As the
influence of the substitution pattern of the aromatic rings was not studied previously, this study
suggests that, if the antitumor activity of boroxazolidones relies solely on calcium channel inhibition,
the use of other amino acids as the synthetic precursor may provide cytotoxic species of higher
potency.
Wild-type p53 is documented to suppress cancer formation in normal cells and protects against
stress-induced DNA damage by inducing cell cycle arrest, DNA repair, and apoptosis(Van Meir et al.,
1994). However, many types of cancer cells, including glioblastoma cells, over express mutant p53
proteins. GBM cells bearing p53 mutations promote chemo-resistant phenotypes, while wild-type p53
does not (Hientz et al., 2017). p53 mutations may cause resistance to DNA-damaging agents and, as
a consequence, cause cellular resistance to multiple cytotoxic agents. Many cytotoxic drugs were
developed to modulate the p53 singling pathway and, thus, stop the progression of tumor growth
(Stegh, 2012). The present study demonstrates that compound 6 potentiates the cytotoxic effect on
LN229 and SNB19 cells, both of which exhibit mutated p53 proteins. However, the effect is lower than

ACCEPTED MANUSCRIPT
that of TMZ. The results suggest that compound 6 may induce the cytotoxic effect by affecting the p53
signaling pathway.
5. Conclusion
Overall, this study demonstrated that boroxazolidone 6, has a strong cytotoxic effect and was found to
have the highest cytotoxicity among the library of compounds tested. Further analyses of the
mechanism of cell death induction by 6 suggests that cell death occurs through apoptosis and has the
potential to activate caspase 3/7 pathways, reactive oxygen species induction, and cell cycle arrest of
GBM. The magnitude of anti-GBM effect of 6 is significantly higher than known chemotherapeutic
agent TMZ. However, testing effects of 6 in an in vivo model, patient derived xenografts, could allow
this compound to be used as a potential therapeutic agent for glioma treatment.
AUTHOR INFORMATION
Corresponding Author
*nuno.rafaelcandeias@tut.fi and meenakshisundaram.kandhavelu@tut.fi; Phone: (+358)417488772
Author Contributions
JR and NRC synthesized and characterized the compounds, AV and GS executed the experiments
and data analysis. VS executed the cell cycle image analysis. KB performed the experiment on the
normal brain cell line. MK, OY and NRC conceived and managed all studies. All the authors
contributed to writing the manuscript.
Funding Sources
MK, AV and OYH acknowledge the Academy of Finland for the project grant support (decision no.
297200) and Tampere University of Technology for Instrumental facility grant support. NRC
acknowledges the Academy of Finland (decision no. 287954) for the financial support.
REFERENCES
Al-Moujahed, A., Brodowska, K., Stryjewski, T.P., Efstathiou, N.E., Vasilikos, I., Cichy, J., Miller, J.W.,
Gragoudas, E., Vavvas, D.G., 2017. Verteporfin inhibits growth of human glioma in vitro without
light activation. Sci. Rep. 7, 7602. https://doi.org/10.1038/s41598-017-07632-8

ACCEPTED MANUSCRIPT
Andrews, P.L., Rapeport, W.G., Sanger, G.J., 1988. Neuropharmacology of emesis induced by anti-
cancer therapy. Trends Pharmacol. Sci. 9, 334–341. https://doi.org/10.1016/0165-
6147(88)90106-X
Baker, S.J., Ding, C.Z., Akama, T., Zhang, Y.-K., Hernandez, V., Xia, Y., 2009. Therapeutic potential
of
boron-containing
compounds.
Futur.
Med.
Chem.
1,
1275–1288.
https://doi.org/10.4155/fmc.09.71
Circu, M.L., Aw, T.Y., 2010. Reactive oxygen species, cellular redox systems, and apoptosis. Free
Radic. Biol. {&} Med. 48, 749–762. https://doi.org/10.1016/j.freeradbiomed.2009.12.022
Clerkin, J.S., Naughton, R., Quiney, C., Cotter, T.G., 2008. Mechanisms of ROS modulated cell
survival during carcinogenesis. Cancer Lett. https://doi.org/10.1016/j.canlet.2008.02.029
Diao, W., Tong, X., Yang, C., Zhang, F., Bao, C., Chen, H., Liu, L., Li, M., Ye, F., Fan, Q., Wang, J.,
Ou-Yang, Z.-C., 2019. Behaviors of Glioblastoma Cells in in Vitro Microenvironments. Sci. Rep.
9, 85. https://doi.org/10.1038/s41598-018-36347-7
Doan, P., Karjalainen, A., Chandraseelan, J.G., Sandberg, O., Yli-Harja, O., Rosholm, T., Franzen,
R., Candeias, N.R., Kandhavelu, M., 2016. Synthesis and Biological Screening for Cytotoxic
Activity of N-Substituted Indolines and Morpholines. Eur. J. Med. Chem. 120, 296–303.
https://doi.org/10.1016/j.ejmech.2016.05.024
Fang, J., Seki, T., Maeda, H., 2009. Therapeutic strategies by modulating oxygen stress in cancer
and inflammation. Adv. Drug Deliv. Rev. 61, 290–302. https://doi.org/10.1016/j.addr.2009.02.005
Giles, G.I., 2006. The redox regulation of thiol dependent signaling pathways in cancer. Curr. Pharm.
Des. 12, 4427–4443. https://doi.org/10.2174/138161206779010549
Hientz, K., Mohr, A., Bhakta-Guha, D., Efferth, T., 2017. The role of p53 in cancer drug resistance and
targeted chemotherapy. Oncotarget 8, 8921–8946. https://doi.org/10.18632/oncotarget.13475
Howard, A., Pelc, S.R., 1986. Synthesis of Desoxyribonucleic Acid in Normal and Irradiated Cells and
Its Relation to Chromosome Breakage. Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 49,
207–218. https://doi.org/10.1080/09553008514552501

ACCEPTED MANUSCRIPT
I. Scorei, R., Popa, R., 2010. Boron-Containing Compounds as Preventive and Chemotherapeutic
Agents for Cancer. ACAMC 10, 346–351. https://doi.org/10.2174/187152010791162289
Kovacic, P., Jacintho, J.D., 2001. Mechanisms of carcinogenesis: focus on oxidative stress and
electron transfer. Curr. Med. Chem. 8, 773–796. https://doi.org/10.2174/0929867013373084
Kuuloja, N.M., Kylmälä, T.M., Tois, J.E., Sjöholm, R.E., Franzén, R.G., 2011. Preparation of
Triethylammonium Tetra-Arylborates (TEATABs): Coupling Partners for the Suzuki Reaction.
Synth. Commun. 41, 1052–1063. https://doi.org/10.1080/00397911003718086
Kwak, E.-Y., Choi, M.-K., Yang, S.-G., Shim, C.-K., Shim, W.-S., 2012a. Investigation into the Efficacy
of Val-SN-38, a Valine-Ester Prodrug of the Anti-Cancer Agent SN-38. Biomol. Ther. 20, 326–
331. https://doi.org/10.4062/biomolther.2012.20.3.326
Kwak, E.-Y., Shim, W.-S., Chang, J.-E., Chong, S., Kim, D.-D., Chung, S.-J., Shim, C.-K., 2012b.
Enhanced intracellular accumulation of a non-nucleoside anti-cancer agent via increased uptake
of its valine ester prodrug through amino acid transporters. Xenobiotica 42, 603–613.
https://doi.org/10.3109/00498254.2011.646339
Ozaki, S., Suzuki, A.Z., Bauer, P.O., Ebisui, E., Mikoshiba, K., 2013. 2-Aminoethyl diphenylborinate
(2-APB) analogues: regulation of Ca2+ signaling. Biochem. Biophys. Res. Commun. 441, 286–
290. https://doi.org/10.1016/j.bbrc.2013.08.102
Paramore, A., Frantz, S., 2003. Bortezomib. Nat. Rev. Drug Discov. 2, 611–612.
https://doi.org/10.1038/nrd1159
Pelicano, H., Carney, D., Huang, P., 2004. ROS stress in cancer cells and therapeutic implications.
Drug Resist. Updat. https://doi.org/10.1016/j.drup.2004.01.004
Philipp, M., Bender, M.L., 1971. Inhibition of serine proteases by arylboronic acids. Proc. Natl. Acad.
Sci. U. S. A. 68, 478–480. https://doi.org/10.1073/pnas.68.2.478
Poddar, N.K., Maurya, S.K., Saxena, V., 2017. Role of Serine Proteases and Inhibitors in Cancer, in:
Chakraborti, S., Dhalla, N.S. (Eds.), Proteases in Physiology and Pathology. Springer
Singapore, Singapore, pp. 257–287. https://doi.org/10.1007/978-981-10-2513-6_12

ACCEPTED MANUSCRIPT
Raunio, J., Mannoja, J., Nguyen, T., Ahmad, N., Kemppainen, N.M., Franzén, R.G., Kandhavelu, M.,
Candeias, N.R., 2017. Base Catalysed N-Functionalisation of Boroxazolidones. RSC Adv. 7,
20620–20627. https://doi.org/10.1039/C7RA03266H
Shi, J., Zuo, H., Ni, L., Xia, L., Zhao, L., Gong, M., Nie, D., Gong, P., Cui, D., Shi, W., Chen, J., 2014.
An IDH1 mutation inhibits growth of glioma cells via GSH depletion and ROS generation. Neurol.
Sci. 35, 839–845. https://doi.org/10.1007/s10072-013-1607-2
Stegh, A.H., 2012. Targeting the p53 signaling pathway in cancer therapy - the promises, challenges
and
perils.
Expert
Opin.
Ther.
Targets
16,
67–83.
https://doi.org/10.1517/14728222.2011.643299
Steiner, S.J., Bien, J.T., Smith, B.D., 1994. Diphenylborinic acid is a strong inhibitor of serine
proteases. Bioorg. Med. Chem. Lett. 4, 2417–2420. https://doi.org/10.1016/S0960-
894X(01)80401-7
Suzuki, A.Z., Ozaki, S., Goto, J.-I., Mikoshiba, K., 2010. Synthesis of bisboron compounds and their
strong inhibitory activity on store-operated calcium entry. Bioorg. Med. Chem. Lett. 20, 1395–
1398. https://doi.org/10.1016/j.bmcl.2009.12.108
Toler, S., Noe, D., Sharma, a, 2006. Selective enhancement of cellular oxidative stress by
chloroquine: implications for the treatment of glioblastoma multiforme. Neurosurg. Focus 21,
E10. https://doi.org/10.3171/foc.2006.21.6.1
Tominaga, H., Kodama, S., Matsuda, N., Suzuki, K., Watanabe, M., 2004. Involvement of reactive
oxygen species (ROS) in the induction of genetic instability by radiation. J. Radiat. Res. 45,
181–8. https://doi.org/JST.JSTAGE/jrr/45.181 [pii]
Tülüce, Y., Lak, P.T.A., Koyuncu, İ., Kılıç, A., Durgun, M., Özkol, H., 2017. The Apoptotic, Cytotoxic
and Genotoxic Effect of Novel Binuclear Boron-Fluoride Complex on Endometrial Cancer.
Biometals 30, 933–944. https://doi.org/10.1007/s10534-017-0060-8
Ushio-Fukai, M., Nakamura, Y., 2008. Reactive oxygen species and angiogenesis: NADPH oxidase
as target for cancer therapy. Cancer Lett. https://doi.org/10.1016/j.canlet.2008.02.044
Van Meir, E.G., Kikuchi, T., Tada, M., Li, H., Diserens, A.C., Wojcik, B.E., Huang, H.J., Friedmann, T.,

ACCEPTED MANUSCRIPT
de Tribolet, N., Cavenee, W.K., 1994. Analysis of the p53 gene and its expression in human
glioblastoma cells. Cancer Res. 54, 649–652.
Velasco-Bejarano, B., Trujillo-Ferrara, J., Miranda, R., 2007. Preparation of Apoptotic Inducers, 2,2-
Diphenyl-1,3,2-oxazaborolidin-5-ones, Under Alkaline Conditions. Synlett 2007, 921–924.
https://doi.org/10.1055/s-2007-973886
Velasco, B., Trujillo-Ferrara, J.G., Castillo, L.H.F., Miranda, R., Sánchez-Torres, L.E., 2007. In vitro
apoptotic activity of 2,2-diphenyl-1,3,2-oxazaborolidin-5-ones in L5178Y cells. Life Sci. 80,
1007–1013. https://doi.org/10.1016/j.lfs.2006.11.034
Walsh, J.G., Cullen, S.P., Sheridan, C., Lüthi, A.U., Gerner, C., Martin, S.J., 2008. Executioner
caspase-3 and caspase-7 are functionally distinct proteases. Proc. Natl. Acad. Sci. U. S. A. 105,
12815–12819. https://doi.org/10.1073/pnas.0707715105
Wu, X.J., Hua, X., 2007. Targeting ROS: Selective killing of cancer cells by a cruciferous vegetable
derived pro-oxidant compound. Cancer Biol. Ther. https://doi.org/10.4161/cbt.6.5.4092
Xu, W., Baribault, H., Adamson, E.D., 1998. Vinculin knockout results in heart and brain defects
during embryonic development. Development 125, 327–337.
Zimmermann, T.J., Bürger, M., Tashiro, E., Kondoh, Y., Martinez, N.E., Görmer, K., Rosin-Steiner, S.,
Shimizu, T., Ozaki, S., Mikoshiba, K., Watanabe, N., Hall, D., Vetter, I.R., Osada, H., Hedberg,
C., Waldmann, H., 2013. Boron-based inhibitors of acyl protein thioesterases 1 and 2.
Chembiochem 14, 115–122. https://doi.org/10.1002/cbic.201200571
Żołnierczyk, J.D., Olejniczak, A.B., Mieczkowski, A., Błoński, J.Z., Kiliańska, Z.M., Robak, T.,
Leśnikowski, Z.J., 2016. In vitro antileukemic activity of novel adenosine derivatives bearing
boron
cluster
modification.
Bioorg.
Med.
Chem.
24,
5076–5087.
https://doi.org/10.1016/j.bmc.2016.08.028
Figs Legend:
Scheme 1. Synthesis of L-valine derived boroxazolidones

ACCEPTED MANUSCRIPT
Fig 1: Synthesis and in vitro cytotoxicity assays: A) Testing of boroxazolidones 1-6 against the growth
of human glioblastoma LN229 cell line at 100 µM. LN229 cell line was treated with the target
compounds for 48 h and cell inhibition was determined with Trypan blue staining. UT - Untreated,
DMSO - negative control and TMZ - positive control. B) Percentage of cell growth inhibition versus
MEF (non-cancerous cells) and LN229 (glioblastoma cells) (10 µM, non-cancer cells vs cancer cells;
P = 0.0319 and 100 µM, non-cancer cells vs cancer cells; P < 0.0001). Each data point represents
mean ± S.E.M of biological and technical replicates.
Fig 2: Compound 6 inhibits glioma cell proliferation. (A) Representative phase-contrast micrographs
of glioma cell lines left untreated (control) or treated for 48 h with 10 µM and 100 µM of
boroxazolidone 6. (B) Different concentrations including 1, 10, 25, 50, 75 and 100 µM were added in
LN229 and SNB19 cells, then incubated for 48 h. Trypan blue staining was used to determine cell
viability. The represented data is normalized against vehicle control, DMSO. Each data point
represents an average of biological and technical replicates. A R2 value of about 1 indicates that the
regression predictions fits the data.
Fig 3: Apoptosis induction determination using Annexin V and Propidium Iodide. A) Microscopic
images of the cells stained with Annexin V-FITC and PI in SNB19 and LN229 cells. Images panels:
Untreated, DMSO and 6 treated cells. B) Percentage of apoptosis cells in 6 and TMZ treated
conditions (6 vs TMZ; P < 0.0001). Each bar represents mean ± S.E.M. Baseline effect due to DMSO
has been deducted.
Fig 4. Effect of compound 6 in caspase 3/7 and intracellular reactive oxygen species: A) Induction of
caspase 3/7 in LN229 (6 vs TMZ; P = 0.1271) and SNB119 cells (6 vs TMZ; P = 0.4519). Cells were
treated with IC50 concentrations of 6 and TMZ. Activity of caspases 3/7 was measured using
luminescence reader. B) Quantification of reactive oxygen species level in LN229 and SNB119 cells
treated with 6, H2O2 a positive control, and TMZ as a drug control. Biological and technical repeats
were used for each condition (6 vs TMZ; P < 0.0001 and 6 vs H2O2 P < 0.0001). Each bar represents
a mean ± S.E.M. Each treatment was performed with biological and technical replicates.
Fig 5. Cell cycle analysis using PI staining. A. Percentage of total cell in different phases when they
were treated with boroxazolidone 6 and DMSO (control) in LN229 and B) SNB119. Each bar
represents a mean ± S.E.M. Each treatment was performed with biological and technical replicates.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Conflicts of Interest Statement
All the authors report that they do not have conflicts of interests.