Arsenical C-Glucoside Derivatives with Promising Antitumor Activity

Article abstract of DOI:10.1002/ejoc.201500529

C-Glucoside derivatives covalently linked to the arsenic atom in different oxidation states have been synthesized by avoiding protection-deprotection steps. While compound 1, bearing an As^V atom, did not show any significant biological activity, compound 2, in which the glucose moiety is linked to the arsenic atom as dithioarsenical (As^III), showed promising antiproliferative activity on a cell line of neuroblastoma (SK-N-BE). C-Glucoside derivatives covalently linked to the arsenic atom in different oxidation states have been synthesized. The C-glucoside conjugated to the phenyldithioarsolan group, showed promising antiproliferative activity on human neuroblastoma cells (SK-N-BE).

Full text of DOI:10.1002/ejoc.201500529

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201500529
Arsenical C-Glucoside Derivatives with Promising Antitumor Activity
Giuseppe D’Orazio,[a] Gelsomina Parisi,[b] Claudia Policano,[c] Rosella Mechelli,[c]
Giovanni Codacci Pisanelli,[d] Michele Pitaro,[b] Giovanni Ristori,[c] Marco Salvetti,[c]
Francesco Nicotra,[a] and Barbara La Ferla*[a]
Keywords: C-Glucosides / Arsenic / Antiproliferation / GLUT transporter / Arsenic-glyco derivatives
C-Glucoside derivatives covalently linked to the arsenic
atom in different oxidation states have been synthesized by
avoiding protection-deprotection steps. While compound 1,
bearing an AsV atom, did not show any significant biological
activity, compound 2, in which the glucose moiety is linked to
the arsenic atom as dithioarsenical (AsIII), showed promising
antiproliferative activity on a cell line of neuroblastoma (SK-
N-BE).
the preferential uptake of glucose by cancer cells, known as
Warburg effect.[4] This effect starts in the early stages of
carcinogenesis and seems to increase with tumor aggressive-
ness.[5] Despite this process being functional for tumor
growth, it is absolutely unfavourable from an energetic
point of view, thus forcing tumor cells to import a great
quantity of glucose. This results in an increase in the expres-
sion of membrane proteins devoted to the uptake of glucose
molecules, in particular the GLUT transporters.[6] The use
of 2-(18F)-fluoro-2-deoxy-d-glucose (FDG) as tracer in the
positron emission tomography (PET) confirms the in vivo
accumulation of glucose inside tumor cells, and represents
the main application of the Warburg effect. The success in
this diagnostic application has suggested the therapeutic ex-
ploitation of the preferential uptake of glucose, through its
Introduction
Since Hippocrates and then through Chinese alchemy,
Fowler’s solution, Erlich’s Salvarsan, until the recent use of
arsenic trioxide As2O3 (ATO) as a therapy for promyelo-
cytic leukemia, arsenic has represented a Janus bifrons of
medicine, being remedy and poison, healer and killer.
This double-edged sword properties stem for a large part
from arsenic physicochemical similarity with phosphorus as
metabolic pathways intended for PO43– may not distinguish
it from AsO4 .
3– [1] This means that arsenic is a serious threat
for health but also an ideal bullet to be delivered to a thera-
peutic target in a masked way, provided it is conjugated
with a highly selective carrier.
The similarity between PO4 and AsO4 is such that
some algae are able to take up arsenate from seawater and
to synthesize arsenosugars, among which are arsenic-con-
taining ribosides (Figure 1).[2] Also synthetic arsenic-glyco
derivatives are not new in the literature. Wimmer et al.[3]
proposed an arsenical glycolipid in which a glucose moiety
was conjugated with the aim of improving chemical-physi-
cal properties of the drug (Figure 1). Aiming at the design
of an arsenic-based anticancer drug, we envision to exploit
3–
3–
[a] Department of Biotechnology and Bioscience, University of
Milano-Bicocca,
P.zza della Scienza 2, 20126 Milan, Italy
E-mail: barbara.laferla@unimib.it
http://www.unimib.it
[b] Xenus S.r.l.,
Via Antonio Ciamarra 222, 00173 Rome, Italy
[c] Centre for Experimental Neurological Therapies (CENTERS),
Neurology and Department of Neurosciences, Mental
Health and Sensory Organs, University of Rome La Sapienza,
S. Andrea Hospital,
Via di Grottarossa, 00189 Rome, Italy
[d] Department of Medical and Surgical Sciences and
Biotechnologies, University of Rome La Sapienza,
Corso della Repubblica 79, 04100 Latina, Italy
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201500529.
Figure 1. Inorganic and organic arsenic derivatives, natural arseno-
furanose and a synthetic arsenical glycolipid; target C-glucoside
compounds 1 and 2.
4620
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 4620–4623

Arsenical C-Glucoside Derivatives with Promising Antitumor Activity
conjugation to cytotoxic drugs, such as cyclophosphamide.
Unfortunately, these bulky conjugates do not penetrate into
the cells through the GLUT transporters, but through the
co-transporter sodium-glucose (SAAT1),[7] thus showing a
reduced selectivity towards tumor cells.
Being devoid of significant steric hindrance, arsenic is in
principle an ideal conjugate for glucose. Moreover, Gibaud
et al.[8] have prepared a small library of dithiarsolane deriv-
atives of phenylarsonic acid (Figure 1) that have shown to
possess antileukemic properties. Therefore, by matching the
selectivity of glucose to the widespread biological impact of
arsenic, we meant to obtain compounds able to specifically
deliver to cancer cells a pharmacological agent capable of
simultaneously targeting different metabolic pathways. This
approach may be particularly useful in cancer since it may
limit the development of resistance by targeting different
key points of tumor metabolism.
In this work we generated arsenical C-glucoside deriva-
tives, in which the arsenic group is covalently linked to the
sugar moiety through the alkyl appendage at the anomeric
centre, thus maintaining the pyranosidic structure of the
glucose moiety, whilst generating a metabolically stable de-
rivative. The arsenal compound should therefore accumu-
late and exert its toxic effect inside the tumor cell.
Scheme 1. Reagents and conditions: (a) ref.[11], then 0.01 m HCl,
70 °C, 1 h; (b) MeOH/H2O, NaCNBH3 (5.6 equiv.),
(1.2 equiv.), yield 35% (1) and 44% (2).
5 or 6
at the concentrations tested (10, 30, 50, 100 μm) (Support-
ing Information), 2 and As2O3 led to a concentration-
dependent increase of cells with morphological signs of
apoptosis (irregular outline, picnotic nuclei, spherical con-
formation and loss of adherence).
The data showed that As2O3 significantly reduces the
proliferative activity of SK-N-BE cells in a dose- and time-
dependent manner, while the activity of 2 was dose-depend-
ent, but it was not influenced by time (Figure 2). The anti-
proliferative effect at different time points, expressed as
IC50, is shown in Table 1.
Results and Discussion
We designed and synthesised two C-glucoside derivatives
covalently linked to the arsenic atom in different oxidation
states as arsenic acid (AsV) and dithioarsenical (AsIII) (1
and 2, respectively; Figure 1) and carried out a preliminary
in vitro biological evaluation on a tumor cell line of human
neuroblastoma (SK-N-BE) in order to evaluate the antipro-
liferative activity.
The synthesis (Scheme 1) was designed in order to avoid
the use of protecting groups and to insert the arsenic-con-
taining moiety at the end of the synthetic route. As arsenic
moiety we chose the commercially available (4-amino-
phenyl)arsonic acid (arsanilic acid, 5), and the readily avail-
able[8,9] phenylarsine derivative 6, that can be chemoselec-
tively attached to the hemiacetal group of the glucose deriv-
ative 4 by reductive amination (Scheme 1). Thus, the key
C-glycosyl intermediate 4 was generated from the readily
available α-C-allylglucoside[10] by ozonolysis.[11] As pre-
viously described, this procedure afforded 4 as a mixture of
the diastereoisomeric bicyclic hemiacetal structures, along
with the formation of the minor methyl acetal derivative,
not reactive in the subsequent reductive amination
(Scheme 1).
In order to increase the yield, this crude product was
treated submitted to acidic conditions in order to convert
the acetal to the hemiacetal form prior to the reductive
amination reaction with the arsenic-containing moieties 5
and 6 to afford the final products 1 and 2. The effects of
the two compounds were tested on human neuroblastoma
cells (SK-N-BE) and compared with that of As2O3. While
compound 1 did not show any significant biological activity
Figure 2. Cell viability analysis of SK-N-BE cells treated with 2 (A)
or As2O3 (B).
Table 1. The in vitro antiproliferative activity IC50 values of As2O3
and 2 on an SK-N-BE cell line.
Test compound
IC50 (48 h)
IC50 (72 h)
IC50 (96 h)
As2O3
2
28.18 μm
110.5 μm
7.10 μm
91.93 μm
3.31 μm
88.43 μm
Eur. J. Org. Chem. 2015, 4620–4623
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4621

B. La Ferla et al.
SHORT COMMUNICATION
0.01 m HCl solution (7 mL) at 70 °C for 1 h. The pH of the solution
was adjusted to pH Ͼ 7 with solid Na2CO3, and compound 6
(1.16 mmol, 1.2 equiv.), suspended in CH2Cl2/MeOH (3:4) (7 mL),
was added. Then, the pH of the solution was decreased to pH = 4
with AcOH; after 20 min, NaCNBH3 (1.36 mmol, 5.6 equiv.) was
added. After 30 min, the solvent was removed under reduced pres-
sure, and the crude product was purified by flash chromatography
(AcOEt/MeOH, 95:5) to afford compound 2 as amorphous solid
(overall yield 191 mg, 44% from 4). [α]2D5 = +31.7 (c = 0.5, DMSO).
1H NMR (400 MHz, CD3OD): δ = 7.29–7.21 (m, 1 H), 7.04–6.94
(m, 1 H), 6.60–6.46 (m, 2 H), 4.00–3.90 (m, 1 H), 3.75 (br. d, J =
11.8 Hz, 1 H), 3.56–3.47 (m, 2 H), 3.44–3.35 (m, 2 H), 3.23–3.06
(m, 7 H + solvent), 1.90–1.81 (m, 2 H) ppm. 13C NMR (100 MHz,
CD3OD): δ = 150.22, 133.01, 130.01, 118.04, 114.23, 113.47, 76.04,
75.27, 74.83, 72.91, 72.51, 63.32, 42.19, 42.05, 25.03 ppm. HRMS:
calcd. for C16H24AsNO5S2 [M + H]+ 450.0385; found 450.0389.
Conclusions
We have demonstrated the antiproliferative effect of an
AsIII derivative covalently linked to glucose, that, if com-
bined with the Warburg effect, still to be demonstrated,
could open the way to new selective anticancer agents.
Interestingly, the antiproliferative effect of the arsenal C-
glucoside is not time-dependent like that of As2O3. Com-
pound 2, synthesised by avoiding protection-deprotection
steps, represents therefore an interesting potential lead com-
pound for the development of a “magic bullet” aiming at
cancer cells.[12]
Experimental Section
Biological Studies
Syntheses
Drugs and Chemicals: Arsenic trioxide (As2O3-ATO) and p-
arsanilic acid were purchased from Sigma Aldrich Company (MI).
Arsenic trioxide was dissolved in 0.1 m NaOH and kept as a stock
solution of 1 mm As2O3 dissolved in distilled water. Growth me-
dium DMEM high glucose, fetal bovine serum (FBS), antibiotics
(penicillin and streptomycin), l-glutamine, and l-phosphate buff-
ered saline (PBS) were obtained from Euroclone. 3-(4,5-Dimeth-
ylthiazol-2-yl)-2,4-diphenyltetrazolium bromide (MTT) solution
and dimethyl sulfoxide (DMSO) were purchased from Sigma
Aldrich Company (MI).
General Remarks: All solvents, when necessary, were dried with
molecular sieves for at least 24 h prior to use. p-Arsanilic acid was
purchased from Sigma Aldrich Company (MI). Thin layer
chromatography (TLC) was performed on silica gel 60 F254 plates
(Merck) with detection by using UV light when possible, or by
charring with a solution of concd. H2SO4/EtOH/H2O (5:45:45) or
a solution of (NH4)6Mo7O24 (21 g), Ce(SO4)2 (1 g), concd. H2SO4
(31 mL) in water (500 mL). Flash column chromatography was per-
formed on silica gel 230–400 mesh (Merck). 1H and 13C NMR
spectra were recorded at 25 °C, unless otherwise stated, with a Var-
ian Mercury 400 MHz instrument. Chemical shift assignments, re-
ported in ppm, are referenced to the corresponding solvent peaks.
HRMS data were recorded with an ABSciex 2000 QTRAP LC/MS/
MS system with an ESI source. Optical rotations were measured at
room temperature by using an Atago Polax-2L polarimeter and are
reported in units of 10–1 degcm3 g–1.
Cell Lines and Culture Conditions: Human neuroblastoma cell line
SK-N-BE was cultured in DMEM high glucose supplemented with
10% fetal bovine serum (FBS), 1% (w/v) penicillin/streptomycin
and 2 mm l-glutamine in a humidified incubator containing 95%
air + 5% CO2. MTT assay for cell proliferation was evaluated by
MTT metabolism as described previously.[13] Briefly, SK-N-BE
were plated in 96-well polystyrene tissue culture plates (Falcon) at
a density of 5ϫ103/well in fresh medium. After 48 h of incubation,
As2O3 or compound 2 solutions (range 10–200 μm) were added to
each well. Cells incubated in the culture medium alone served as
untreated control. Cells were cultured for 48, 72 and 96 h. After
incubation, 20 μL aliquots of MTT solution (5 mg/mL in PBS)
were added to each well and re-incubated at 37 °C for 4 h. Then,
the supernatant culture medium was carefully aspirated, and
200 μL aliquots of DMSO were added to each well to dissolve the
formazan crystals, followed by incubation for 10 min to dissolve
air bubbles. The culture plate was placed on a micro-plate reader,
and the absorbance was measured at 540 nm. The amount of color
produced is directly proportional to the number of viable cells. All
assays were performed in six replicates for each concentration and
meanϮSD values were used to estimate the cell viability. The cell
proliferation rate was calculated as the percentage of MTT absorp-
tion as follows: % survival = (mean experimental absorbance/mean
control absorbance)ϫ100. Morphological changes were monitored
by phase-contrast microscopy.
Crude Compound 4:[11] Briefly, a solution of allyl-C-glucoside 3[10]
(1.5 g, 7.3 mmol) in MeOH/H2O (100 mL) was cooled to –78 °C
and saturated with ozone. After 75 min, the pale blue solution was
purged with oxygen and then with nitrogen until colorless. Then
Me2S (73 mmol, 10 equiv.) was added, and the reaction mixture
was warmed to room temp. and stirred overnight. The solvent was
then removed under reduced pressure to afford crude compound 4
as a colorless oil.
4-[2-(α-D-Glucopyranosyl)ethylamino]phenylarsonic Acid (1): Crude
compound 4 (1.46 mmol, 1 equiv.) was treated with a 0.01 m HCl
solution (7 mL) at 70 °C for 1 h. The pH of the solution was
adjusted to pH Ͼ 7 with solid Na2CO3, and p-arsanilic acid 5
(1.75 mmol, 1.2 equiv.), suspended in MeOH/H2O (30 mL), was
added. Then, the pH of the solution was decreased to pH = 4 with
AcOH; after 20 min, NaCNBH3 (2.04 mmol, 5.6 equiv.) was added.
After 30 min, the solvent was removed under reduced pressure, and
the crude product was purified by flash chromatography (AcOEt/
MeOH/H2O/AcOH, 70:30:5:5) to afford compound
1 as an
amorphous solid (overall yield 208 mg, 35% from 4). [α]2D5 = +61.7
1
(c = 1, H2O). H NMR (400 MHz, D2O): δ = 7.56 (d, J = 8.5 Hz,
2 H), 6.86 (d, J = 8.5 Hz, 2 H), 4.18–4.03 (m, 1 H), 3.83 (dd, J =
12.1, 1.7 Hz, 1 H), 3.73–3.51 (m, 4 H), 3.37–3.20 (m, 3 H), 2.05–
1.89 (m, 2 H) ppm. 13C NMR (100 MHz, D2O): δ = 154.6, 134.1,
123.7, 116.3, 76.83, 75.93, 75.45, 73.664, 72.99, 63.74, 42.32,
25.64 ppm. HRMS: calcd. for C14H22AsNO8 [M + H]+ 408.0634;
found 408.0638.
[1] B. P. Rosen, FEBS Lett. 2002, 529, 86–92.
[2] a) P. Taar, A. Rumpler, T. Madl, G. Saischek, K. A. Fran-
cesconi, Aust. J. Chem. 2009, 62, 538–545; b) D. P. McAdam,
R. V. Stick, Tetrahedron Lett. 1986, 27, 251–254.
[3] N. Wimmer, J. A. Robinson, N. Gopisetty-Venkata, S. J. Rob-
erts-Thomson, G. R. Monteith, I. Toth, Medicinal Chemistry
2006, 2, 79–87.
4-(1,3,2-Dithiarsolan-2-yl)-N-[2-(α-
D-glucopyranosyl)ethyl]aniline
[4] O. Warburg, F. Wind, E. Negelein, J. Gen. Physiol. 1927, 8,
519–530.
(2): Crude compound 4 (0.97 mmol, 1 equiv.) was treated with a
4622
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 4620–4623

Arsenical C-Glucoside Derivatives with Promising Antitumor Activity
[5]
[6]
Whittaker, P. J. Hogg, T. P. Davis, J. Am. Chem. Soc. 2015, 137,
4215–4222.
[10] J. A. Bennek, G. A. Gray, J. Org. Chem. 1987, 52, 892–897.
[11] F. Cardona, B. La Ferla, J. Carbohydr. Chem. 2008, 27, 203–
213.
[12] F. Nicotra, M. Pitaro, B. La Ferla, M. Salvetti, G. Codacci
Pisanelli, G. Ristori, WO2013124874, PCT/IT2013/000051,
2013.
[13] T. Mosmann, J. Immunol. Methods 1983, 65, 55–63.
Received: April 26, 2015
R. A. Cairns, I. S. Harris, T. W. Mak, Nat. Rev. Cancer 2011,
11, 85–95.
F. Q. Zhao, A. F. Keating, Curr. Genomics 2007, 8, 113–128.
[7] M. Veyhl, K. Wagner, C. Volk, V. Gorboulev, K. Baumgarten,
W. M. Weber, M. Schaper, B. Bertram, M. Wiessler, H. Koep-
sell, Proc. Natl. Acad. Sci. USA 1998, 17;95, 2914–2919.
[8] S. Gibaud, R. Alfonsi, P. Mutzenhardt, I. Fries, A. Astier, J.
Organomet. Chem. 2006, 691, 1081–1084.
[9] P. Wilson, A. Anastasaki, M. R. Owen, K. Kempe, D. M.
Haddleton, S. K. Mann, A. P. R. Johnston, J. F. Quinn, M. R.
Published Online: June 22, 2015
Eur. J. Org. Chem. 2015, 4620–4623
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4623