Synthesis and biological evaluation of a new class of glycoconjugated disulfides that exhibit potential anticancer properties

Article abstract of DOI:10.1016/j.bmc.2012.03.070

A synthetic strategy, based on the in situ generation of sulfenic acids and their thermolysis in the presence of thiols, was developed for obtaining a collection of polyvalent disulfides in which a benzene scaffold accommodates two or three flexible arms

Full text of DOI:10.1016/j.bmc.2012.03.070

Bioorganic & Medicinal Chemistry 20 (2012) 3186–3195
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and biological evaluation of a new class of glycoconjugated disulfides
that exhibit potential anticancer properties
Paola Bonaccorsi a, , Francesca Marino-Merlo b, Anna Barattucci a, Gianluca Battaglia a,
⇑
Emanuela Papaianni b, Teresa Papalia a, Maria C. Aversa a, , Antonio Mastino b,
a Dipartimento di Chimica Organica e Biologica, Università degli Studi di Messina, Viale Ferdinando Stagno d’Alcontres 31 (vill. S. Agata), 98166 Messina, Italy
b Dipartimento di Scienze della Vita «M. Malpighi» Università degli Studi di Messina, Viale Ferdinando Stagno d’Alcontres 31 (vill. S. Agata), 98166 Messina, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A synthetic strategy, based on the in situ generation of sulfenic acids and their thermolysis in the pres-
ence of thiols, was developed for obtaining a collection of polyvalent disulfides in which a benzene scaf-
fold accommodates two or three flexible arms connecting saccharide moieties. Targeting carbohydrate
metabolism or carbohydrate-binding proteins may constitute important approaches in the discovery pro-
cess of new therapeutic anticancer agents. Therefore, a preliminary screening to ascertain the cytostatic/
cytotoxic potential of this new class of enantiopure glycoconjugated disulfides has been conducted.
Among them, products with two disulfide arms, harbouring galactose rings, induced high levels of apop-
tosis on U937 histiocytic lymphoma cells, but lower levels of cell death on peripheral blood mononuclear
cells from healthy donors. Further experiments indicated that apoptosis induced by these glycoconjugat-
ed bis(disulfides) in U937 cells corresponds to the Bcl-2-sensitive, intrinsic form of apoptotic cell death.
The bioinvestigation was extended to a panel of human cancer cell lines with different levels of malig-
nancy and resistance to chemotherapeutic agents. Compounds under study proved to induce detectable
levels of cell death towards all the tested cancer cell lines.
Received 22 December 2011
Revised 26 March 2012
Accepted 30 March 2012
Available online 6 April 2012
Keywords:
Sulfenic acids
Unsymmetrical disulfides
Thioglycoconjugates
Cytotoxicity
Apoptosis
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
therapeutic agents and in the research of information about com-
plex biological processes, such as the recognition events.
The biological role of carbohydrates is extremely important for
a variety of cellular processes, including those involved in cell
transformation and tumorigenesis. In fact, evidence accumulated
in the last decades in a variety of cancer cell systems, as well as
in specimens from cancer patients, indicates that modifications
in carbohydrate-related structure, function or metabolism are
common changes of cancer cells and especially of those showing
resistance to traditional chemotherapy.1 Thus, there is growing
interest in the development of novel cancer therapeutics based
on glycobiology.2
In this paper we describe our synthetic strategy that involves
transient sulfenic acids as intermediates in the obtainment of gly-
coconjugated disulfides, and a preliminary screening of the effect
of the new compounds on cancer cell viability. In particular, we
have synthesized a collection of new sugar molecules whose carbo-
hydrate residues are joined by spacers of different length and flex-
ibility through disulfide bonds, taking into account the pivotal role
of apoptosis regulation in cancer progression and therapy,4 and the
potential of galactose-based small molecules as cancer therapeu-
tics. In a previous work we have reported indeed a three-step syn-
Functionalization, chirality, and structural diversity in carbohy-
drates allow the induction of several biological activities within a
small number of structural variations so that they have been de-
scribed as ‘privileged structures’.3 The principle of privileged struc-
tures matches well the idea of creating a collection of molecules in
which the privileged skeleton allows a wide investigation in both
medical and biological directions. Glycoconjugates in particular
constitute fundamental tools in the discovery process of new
thesis of bis-b-D-glucopyranosides containing thioalkane and
thioarene spacers, and two compounds of this family are endowed
with a specific cytotoxic potential attributable to induction of cell
death by apoptosis.5
Starting from these biological evidences, the control of molecu-
lar architecture and saccharide spacing playing a significant role on
the enhancement of the apoptotic activity prompted us to intro-
duce into the molecular skeleton of compounds—such as 1 and 2
(Fig. 1)—the disulfide bond, which is present in many biological
systems and is hydrolysis-resistant. Moreover, glycosyldisulfides
constitute a new class of carbohydrate derivatives with promising
biological properties.6 We have generated the disulfide bond by
condensation of suitable sulfenic acids with thiols, a reaction that
⇑
Corresponding author.
E-mail address: paola.bonaccorsi@unime.it (P. Bonaccorsi).
These authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.03.070

P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
3187
from the commercially available benzenedimethanethiols 16 and
17 that were reacted with methyl acrylate in the presence of a cat-
alytic amount of benzyltrimethylamonium hydroxide (TRITON B),
to give sulfoxides 20 and 21 after the controlled oxidation of the
corresponding sulfides 18 and 19 (Scheme 2).
b-(Methoxycarbonyl)ethanesulfoxides, such as 20 and 21, have
been used in the synthesis of all compounds 3–9 (Fig. 2) since the
2-(methoxycarbonyl)ethyl residue, involved in the b-syn-elimina-
tion, allows the thermolysis to occur at the reflux temperature of
1,2-dichloroethane (DCE, 83 °C). Our b-(methoxycarbonyl)ethane
sulfoxides,9 described in the Supplementary data, can be kept on
the bench without showing any trace of decomposition and can
be easily purified by column chromatography.
Figure 1. Glycoconjugated sulfoxide and sulfone.
In Scheme 3 the thermolysis of sulfoxides 20 and 21 is depicted.
Sulfenic acids 22 and 23 were generated in situ, respectively, and
reacted with 2,3,4,6-tetra-O-acetyl-1-thio-b-D-gluco- or -galacto-
pyranose, leading to bis(disulfides) 4, 6, and 8. Due to the high pur-
ity of precursors, the sulfenic acid/thiol condensation occurs in
is well known in the literature7 but applied by us for the first time
in the synthesis of bis(disulfides) bearing bioresidues.8 While sym-
metric glycosyl disulfides are easily accessed following various
straightforward procedures,6 the synthesis of molecules in which
each sulfur atom of the disulfide bond is linked to a different resi-
due occurs through less efficient and direct approaches.8 The col-
lection of molecules that we have obtained following the
synthetic procedure reported in this work may provide a basis
for the development of a new series of pharmacologically active
carbohydrate-based molecules.
good yields in the presence of 2,3,4,6-tetra-O-acetyl-1-thio-b-
glucopyranose.
D-
When the thermosensitive 2,3,4,6-tetra-O-acetyl-1-thio-b-
D-
galactopyranose was used as sulfenic acid partner we observed a
decrease in the yield of the corresponding bis and tris(disulfides)
5, 8 and 159 (Fig. 2 and Scheme 3) due to a minor availability of
the easily decomposable aforesaid galactothiol to react with the
transient sulfenic functions.
The choice of the starting thiol in the preparation of the sulfenic
acid precursor is crucial in the efficiency of the total process. For
instance, compound 3 can be prepared either by using 1,3-benzen-
edithiol (24) as condensation partner of the sulfenic acid 258 or by
involving thiol 27 in the condensation with sulfenic acid 26 gener-
ated in situ in three steps from dithiol 24 (Scheme 4) through
1,3-di-[(2-methoxycarbonylethyl)thio]benzene. This last way is
described in the Supplementary data and leads to the acetylated
glucoconjugate bis(disulfide) 3 in 85% yield against the 50% previ-
ously reported.8
The difference in yields between the two alternative routes of
the same synthetic approach is mainly due to the difference in
the sulfoxides used for generating sulfenic acids 25 and 26.
The two sulfoxides, one of which is similar in structure to the
sulfoxides shown in Scheme 3 and the other has been already de-
scribed,8 have two different alkane sulfinyl moieties involved in
the b-elimination process. Sulfoxide precursor of sulfenic acid 25,
allows the generation of the transient species at mild temperature
(40 °C), but undergoes readily decomposition and is difficult to
handle. The more stable and well purified sulfoxide precursor of
sulfenic acid 26 thermolyzed at higher temperature (83 °C) but,
in this case, raised its reactivity as generator of the corresponding
transient species, leading to a better total yield in the condensation
reaction.
2. Results and discussion
2.1. Chemistry
In accordance with our research interests devoted to the use of
sulfenic acids as intermediates in organic synthesis, we have devel-
oped the methodology for obtaining unsymmetrical disulfides by
sulfenic acid/thiol condensation.8 This synthetic strategy is based
on the preparation of a suitable precursor of the desired sulfenic
acid and its generation in situ by thermolysis in the presence of a
thiol. The general procedure is depicted in Scheme 1 where thiols
represent in turn the starting material for the sulfoxide precursor
of the sulfenic acid and the condensation partner of the aforesaid
sulfenic acid.
A sulfoxide possessing at least a hydrogen atom and an elec-
tron-withdrawing group b-situated with respect to the sulfur func-
tion acts as a good precursor of the corresponding sulfenic acid.
The advantages of (i) easy modulability of this methodology and
in particular (ii) possibility of choosing the suitable sulfenic acid
precursor allow an efficient entry to sugar connecting disulfides
in a concise manner. Following this procedure we have accom-
plished the synthesis of the collection of glycoconjugated bis and
tris(disulfides), some of which protected at OH groups, shown in
Figure 2. In each of them the benzene platform accommodates
two or three flexible disulfide arms connecting the saccharide moi-
eties, in analogy to the structural skeleton of compounds 1 and 2.
The synthesis of disulfides 3, 10, 11, 13, and 15 has been already
described.8,9 Bis(disulfides) 4–6 and 8 have been obtained starting
Compounds 7, 9, 12, and 14 have been obtained in almost quan-
titative yield after deacetylation of the sugar residues in com-
pounds 6, 8, 11, and 13 respectively that was performed with
aqueous ammonia (30%).10 They were purified by flash chromatog-
raphy (CHCl3/MeOH 9:1).
Bis- and tris(disulfides) 3–15 have been all obtained in enantio-
pure form, fully characterized and subjected to biological studies
described in the following section.
2.2. Biological data
In a first round of experiments we performed a preliminary
screening to ascertain the cytostatic/cytotoxic potential of enantio-
pure bis and tris(disulfides) 3–15, using the U937 histiocytic lym-
phoma cell line, that is, a tumor cell line showing moderate level of
malignancy, as an experimental model. Disulfides 3–6, 8, 10, 11,
Scheme 1. General synthetic procedure for unsymmetrical disulfides via sulfenic
acids.

3188
P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
Figure 2. Glycoconjugated disulfides under biological evaluation.
Scheme 2. The synthesis of sulfoxides 20 and 21.
Scheme 3. The generation in situ of sulfenic acids 22 and 23 and their reaction with suitable thiosugars.
Scheme 4. Two alternative applications of the same synthetic strategy towards disulfide 3.

P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
3189
13, and 15 were dissolved in DMSO at the concentration of 20 mM,
glycoconjugated disulfides 7, 9, 12, and 14 were dissolved at the
same concentration in RPMI-1640 medium. These mother solu-
tions were further diluted in RPMI-1640 medium supplemented
In a second round of experiments, we investigated the expres-
sion impact of the anti-apoptotic Bcl-2 protein on apoptosis in-
duced by bis(disulfide) 8, in order to further characterize the
form of cell death induced by galactose-bearing glycoconjugates,
such as 5 and 8. We compared apoptosis levels following 24 h
treatment with compound 8 in U937 stable transfectants over-
expressing a functional active murine bcl-2 gene (U937mBCL2)
and in control cells transfected with an empty vector (U937pMEP).
Results shown in Figure 6 demonstrate that over-expression of
Bcl-2 highly, even if not fully at the highest concentration assayed,
protected U937mBCL2 cells from apoptosis induced by bis(disul-
fide) 8. Conversely, U937pMEP cells were highly sensitive, as ex-
pected, to apoptosis induced by 8. These experiments confirmed
that galactose-bearing glycoconjugated bis(disulfides), such as 8,
have the ability to specifically induce cell death by apoptosis and
indicated that apoptosis induced in U937 cells by this new class
of compounds corresponds to the classical, Bcl-2-sensitive, mithoc-
ondrial-dependent, intrinsic form of apoptotic cell death.
We then focused our attention on compounds 5 and 8, which
showed the higher pro-apoptotic potential in U937 cells, and ex-
tended our investigation to a panel of human cancer cell lines with
different characteristics. To this purpose, the following cell lines
were utilized: THP-1 (established from the peripheral blood of a pa-
tient with acute monocytic leukaemia), MOLT-3 (established from
the peripheral blood of a patient with acute lymphoblastic leukae-
mia), HeLa (established from cancer cells of a patient with cervical
carcinoma), HT-29 (adenocarcinoma cells established from a pa-
tient with colorectal carcinoma), MCF7 (adenocarcinoma cells
established from a patient with breast cancer) and HepG2 (hepato-
cyte carcinoma cells established from a patient with hepatocellular
carcinoma). We gathered the cell lines into a first group of cells
showing moderate malignancy/resistance to chemotherapy (THP-
1, MOLT-3, and HeLa cells) and a second group with higher degree
of malignancy/resistance (HT-29, HepG2, and MCF7 cells). For the
first group of cancer cell lines, cytotoxicity was assessed as reported
for U937 cells and results obtained with bis(disulfides) 5 and 8 were
also compared with those obtained using glycoconjugated
bis(disulfides) 7, 9, and thiogal11 in some of the cell lines. For the
second group of cell lines, all growing as adherent cells, cell death
was detected by trypan blue exclusion test and a MTS assay per-
formed directly in the culture wells after 48 h of treatment, to avoid
possible bias in detection of death due to detachment procedures.
As shown in Table 1, cell lines of the first group exhibited vari-
able sensitivity to the assayed glycoconjugates, and compounds 5
and 8 proved to specifically act as pro-apoptotic agents towards
all tested cancer cell lines. In particular, MOLT-3 lymphoma cells
were highly susceptible to apoptosis induced by these molecules.
Furthermore cell number and viability were evidently reduced by
bis(disulfides) 5 and 8 even in the second group of cancer cell lines,
that is, in those generally considered as more malignant and resis-
tant to chemotherapeutic agents (Table 2).
with 5% fetal bovine serum (FBS), 2 mM
L-glutamine and penicil-
lin-streptomycin to reach a set of 1 in 10 serial dilutions ranging
from 100 lM to 1 lM. The obtained samples were added to
5 ꢀ 104 U937 cells in 100
lL total volume and incubated at 37 °C
in a CO2 incubator. The same cells were exposed to vehicles alone,
as a control experiment, in equal amounts to those employed for
dissolving compounds 3–15. After 24 h incubation, the absolute
number of total cells and the percentage number of cells showing
plasma membrane permeability were evaluated by microscopy
analysis and trypan blue exclusion test, while percentage of apop-
totic cells was evaluated by fluorescence microscopy analysis of
cells stained with acridine orange.
Results shown in Figure 3 refer to the assayed tris(disulfides)
10–15, while those shown in Figure 4 refer to glycoconjugated
bis(disulfides) 3–9. Data concerning 2,3,4,6-tetra-O-acetyl-1-thio-
b-D
-galactopyranose (thiogal)11 are also reported in Figure 2 for
comparison. Mean values standard deviations are reported.
As shown in Figure 3A, tris(disulfides) 10 and 15 remarkably re-
duced the absolute number of cells, in a dose-dependent fashion, in
comparison with vehicle treated cells. However, this reduction was
not associated with a consistent induction of cell death by part of
the same compounds, as shown in Figures 3B and C, suggesting a
cytostatic rather than a cytotoxic activity. Tris(disulfides) 11–14
exerted the higher cytotoxic activity among this group of mole-
cules, as shown in Figure 3B. Nevertheless, in neither case mean
cell death values were beyond 25% nor remarkable levels of apop-
tosis were detected for any of the tris(disulfides) 10–15.
Results shown in Figure 4 demonstrate that some of the
bis(disulfides) 3–9 exerted an evident cytotoxic effect on U937 cells.
In particular compounds 5 and 8, that is, those harbouring galactose
rings, induced very high levels of apoptosis at the higher concentra-
tion assayed (Fig. 4C; Supplementarydata). After treatment with the
same two compounds, dead cell levels, detected by trypan blue as-
say, were consistently lower than those of cells showing apoptotic
morphology (Fig. 4B). This suggests that membrane permeability in-
duced at 24 h by bis(disulfides) 5 and 8 was a late event of apoptosis
that should be considered the only form of cell death induced by
these compounds. Conversely, levels of trypan blue positive cells in-
duced by bis(disulfides) 4 and 6 were higher than that of apoptotic
cells detected for the same compounds. This suggests that forms of
cell death other than apoptosis could contribute to the appreciable
cytotoxicity exerted by compounds 4 and 6.
In addition, bis(disulfides) 4 and 5 exerted the higher cytostatic
effect, as assessed by reduction of total cell number (Fig. 4A). No
remarkable difference was observed between apoptosis values de-
tected in samples treated with thiogal11 and vehicle treated sam-
ples, this suggesting that apoptosis induced by bis(disulfides) 5
and 8 is a consequence of their whole molecular structures rather
than the effect of the single saccharide moieties.
The high, pro-apoptotic effect of compound 5 on U937 cells was
then compared with that exerted on non-malignant cells, using
peripheral blood mononuclear cells (PBMCs) from healthy donors
and flow cytometry analysis following propidium iodide staining
to detect hypodiploid, apoptotic nuclei. The results obtained by
this last technique confirmed the capability of compound 5 to in-
duce high levels of apoptosis on U937 cells, in a dose-dependent
fashion, after 24 h incubation (Fig. 5). In contrast, levels of apopto-
sis detected on PBMCs after 24 h incubation with compound 5
were very modest (data not shown). Hypodiploid nuclei from these
cells were appreciable only after 48 h incubation at the higher
concentrations (Fig. 5) and even in this case they were significantly
lower than those observed in U937 cells.
The cytotoxicity assays, adopted for HT-29, HepG2 and MCF7
cell lines, that is, a panel of cancer cells corresponding to malignan-
cies of high worldwide incidence, do not allow us to assess the spe-
cific form of cell death induced by compounds 5 and 8 in these
cells. However apoptosis remains a possible candidate on the basis
of the results obtained with other cell lines.
3. Conclusions
We have synthesized, by an easy three-step procedure, a
collection of molecules in which a benzene platform accommodates
two or three flexible disulfide arms connecting saccharide moieties.
This synthetic route allows: (i) the diversification of the carbohy-
drate residues within the organic scaffold; (ii) the introduction of

3190
P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
Figure 3. Effects of tris(disulfides) 10–15 on cytotoxicity in U937 cells after 24 h incubation. Each data point represents the mean and standard deviation from at least three
different determinations. (A) Absolute number of total cells/well assessed by microscopy analysis using a haemocytometer. Comparisons of the means between treated
samples and corresponding controls, by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test: 10 l00
10 M, p = 0.007; 15 100 M, p <0.001; all other comparisons, not significant (NS). (B) Percentage of cells showing membrane permeability by trypan blue exclusion test.
Comparisons of the means between treated samples and corresponding controls, by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test: 11 l0 M and
l00 M, p <0.001; 12 10 M, p = 0.006; 12 100 M, p <0.001; 13 l0 M and l00 M, p <0.001; 14 10 M and l00 M, p <0.001; all other comparisons, NS. (C) Percentage nuclei
showing apoptotic features at microscopy analysis following staining with acridine orange. Comparisons of the means between treated samples and corresponding controls,
by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test: 10 l00 M, p = 0.012; 13 100 M, p = 0.012; 14 100 M, p = 0.028; other comparisons, NS.
lM, p <0.001; 14 10 lM and 100 lM, p = 0.050; 15
l
l
l
l
l
l
l
l
l
l
l
l
l
the disulfide bond as biologically significant functional group; (iii)
the modulability of the overall process through the use of suitable
and easily accessible thiols (Scheme 3). Bis(disulfides), with a
thread-like core and galactosyl groups at both ends of the molecule,

P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
3191
Figure 4. Effects of bis(disulfides) 3–9 and thiogal on cytotoxicity in U937 cells after 24 h incubation. Each data point represents the mean and standard deviation from at
least three different determinations. (A) Absolute number of total cells/well assessed by microscopy analysis using a haemocytometer. Comparisons of the means between
treated samples and corresponding controls, by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test: 4 l0
100 M, p <0.001; 7 100 M, p = 0.015; 8 1 M, p = 0.002; 8 10 M, p = 0.003; all other comparisons, NS. (B) Percentage of cells showing membrane permeability by trypan
blue exclusion test. Comparisons of the means between treated samples and corresponding controls, by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test:
3 l00 M, p <0.001; 4 1 M, 10 M and 100 M, p <0.001; 5 10 M, p = 0.019; 5 100 M, p <0.001; 6 l00 M, p <0.001; 8 l00 M, p <0.001; all other comparisons, NS. (C)
Percentage nuclei showing apoptotic features at microscopy analysis following staining with acridine orange. Comparisons of the means between treated samples and
corresponding controls, by Hochberg’s GT2 post-hoc multiple comparison One-Way ANOVA test: 4 l0 M and l00 M, p <0.001; 5 10 M, p = 0.01; 5 100 M, p <0.001; 6
100 M, p <0.001; 8 l0 M and 100 M, p <0.001; 9 100 M, p = 0.043; other comparisons, NS.
lM, p = 0.005; 4 l00 lM, p <0.001; 5 10 lM and
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
have proven to be cytotoxic, even if at relatively high concentrations,
towards a panel of human cancer cells with different levels of
malignancy and resistance to chemotherapeutic agents: impor-
tantly, also cell lines rather resistant to cytotoxic agent showed a

3192
P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
Figure 5. Effect of compound 5 on apoptosis in U937 cells and in peripheral blood mononuclear cells (PBMC). U937 cells and PBMC were treated with 0, 1, 10, 50 and 100 lM
of compound 5 for 24 and 48 h, respectively, and apoptosis was evaluated by hypodiploid nuclei analysis after DNA staining with propidium iodide. Percentages of
hypodiploid nuclei (M1) are reported in the cytograms. One experiment out of two performed with similar results, is shown.
This allows us to hypothesize a highly specific effect related, but
not simply attributed, to the galactosyl residues. One possible
hypothesis is that our new galactoconjugates target cellular compo-
nents that play fundamental roles in normal galactose metabolism,
thus causing an irreversible cell damage. An attractive mechanism is
that galactoconjugates could intracellularly bind to galectin-3, thus
inhibiting the capacity of this protein to act as an anti-apoptotic,
Bcl-2 mimetic.13 Results obtained using the deprotected compound
9 seem to not support this hypothesis unless the binding to galectin-
3 could prescind from classical hydrogen bonding interactions.
Further investigation is needed to clarify the relationships be-
tween apoptotic effects of the assayed compounds on cancer cells
and related events. However, our data indicate that the new group
of galactoconjugates bis(disulfides) specifically cause cell death as-
Figure 6. Effects of bis(disulfide)
8 on apoptosis in U937 transfectants over-
cribed to apoptosis on cancer cells of different origin. This aspect
encourages us to search for improving the specific cytotoxic activ-
ity of this new family of compounds towards cancer cells. Results
of these studies will elucidate the possibility of the therapeutic
application of galactoconjugates bis(disulfides).
expressing Bcl-2 (U937mBCL2) and in U937 control transfectants (U937pMEP),
after 24 h incubation. Each data point represents the mean and standard deviation
from three different determinations of percentage nuclei showing apoptotic
features at microscopy analysis following staining with acridine orange. Compar-
isons of the means by Hochberg’s GT2 post-hoc multiple comparison One-Way
ANOVA test: l0
U937mBCL2+DMSO versus U937pMEP+8
U937pMEP+8 p <0.001; l00 M, U937pMEP+8 versus U937pMEP+DMSO p <0.001,
U937mBCL2+8 versus U937pMEP+DMSO <0.001, U937pMEP+8 versus
lM, U937pMEP+8 versus U937pMEP+DMSO p = 0.005,
p
<0.001, U937mBCL2+8 versus
l
4. Experimental section
p
U937mBCL2+DMSO p <0.001, U937mBCL2+8 versus U937mBCL2+DMSO p <0.001,
U937mBCL2+8 versus U937pMEP+8 p <0.001; other comparisons, NS.
4.1. General methods for the synthetic procedures
Solvents were purified according to standard procedures. Light
petroleum (petrol) used refers to the fraction boiling at 40–60 °C.
All reactions were monitored by TLC on commercially available
precoated plates (Aldrich Silica Gel 60 F254) eluted with ace-
tone/CHCl3, CHCl3/MeOH, or EtOAc/MeOH, and the products were
visualized with vanillin [1 g dissolved in MeOH (60 mL) and concd
H2SO4 (0.6 mL)]. Silica gel used for column chromatography was
Aldrich 60. 1H and 13C NMR spectra were recorded with a Varian
Mercury 300 spectrometer at 300 and 75 MHz, respectively, in
CDCl3 or CD3OD solutions; the assignments are supported by At-
tached Proton Test (APT) and homodecoupling experiments;
proton and carbon nuclei identified by apex pertain to glucose
or galactose residues in the bis(disulfides) 4–9 and tris(disulfides)
12 and 14. The remaining glycoconjugated disulfides 3, 10, 11, 13,
and 15 have been already described.8,9 The purities of all tested
compounds are higher than 95% by 1H/13C NMR spectroscopy
and elemental analyses, which were reported to be within 0.4%
of calculated values.
certain sensitivity to cell death induced by these compounds. More-
over, experiments using Bcl-2-overexpressing transfectants fully
confirmed the apoptotic feature of death induced by the newly syn-
thesized 5 and 8. These experiments also showed that high expres-
sion of the Bcl-2 anti-apoptotic protein, which occurs in a wide
variety of human cancers,12 did not completely protect from apopto-
sis induced by the same bis(disulfides) 5 and 8, giving an additional
therapeutic value to our results. We cannot discern at the moment
the mechanisms involved in induction of apoptosis by the galactose
derivatives, and whether the action of these pro-apoptotic com-
pounds is exerted at external membrane level or in the inner
compartments of the cancer cells. However, we can notice that: (i)
2,3,4,6-tetra-O-acetyl-b-D-galactopyranose did not induce apopto-
sis even at high concentrations, (ii) gluco- and galactoconjugates
under study exert different effects, (iii) deprotected compound 9
did not exert the same pro-apoptotic effect exerted by the corre-
sponding protected compound (Fig. 4C and Supplementary data).

P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
3193
Table 1
Effects of bis(disulfides) 5, 8, 9, and 2,3,4,6-tetra-O-acetyl-1-thio-b-D-galactopyranose (thiogal) on cytotoxicity, as assessed by trypan blue exclusion
test and detection of apoptotic cells, in MOLT-3, THP-1 and HeLa cancer cell lines
a
b
c
Cell lines
MOLT-3
Compounds
IC50
(
lM)
CC30
(l
M)
AC30 (lM)
5
8
9
>100
>100
>100
>100
54 (0.98)d
29 (0.98)
>100
15 (0.73)
24 (0.97)
>100
Thiogal
>100
>100
THP-1
HeLa
5
7
8
9
>100
>100
>100
>100
>100
>100
>100
93 (0.99)
>100
>100
92 (0.84)
>100
88 (0.95)
>100
Thiogal
>100
8
>100
>100
62 (0.98)
a
b
c
Drug concentration able to cause inhibition of the total cell count by 50% (IC50), after 24 h incubation.
Drug concentration able to cause 30% cytotoxicity by trypan blue assay (CC30), after 24 h incubation.
Drug concentration able to cause 30% apoptotic cells by fluorescence microscopy analysis of acridine orange stained cells (AC30), after 24 h
incubation.
d
Pearson’s correlation coefficient shown in brackets.
Table 2
Effects of bis(disulfides) 5 and 8 on cytotoxicity, as assessed by trypan blue exclusion test and MTS assay, in HT-29, HepG2 and MCF7 cancer cell lines
a
b
c
Cell lines
HT-29
Compounds
IC50
(
l
M)
CC30
(
l
M)
MAIC30 (lM)
d
5
8
60 (0.80)
63 (0.80)
>100
61 (0.80)
12 (0.85)
23 (0.90)
HepG2
MCF7
5
8
54 (0.89)
51 (0.90)
77 (0.90)
98 (0.90)
41 (0.92)
30 (0.92)
5
8
42 (0.80)
38 (0.80)
>100
69 (0.80)
61 (0.99)
63 (0.99)
a
b
c
Drug concentration able to cause inhibition of the total cell count by 50% (IC50), after 48 h incubation.
Drug concentration able to cause 30% cytotoxicity by trypan blue assay (CC30), after 48 h incubation.
Drug concentration able to cause reduction of the formazan product formation (MTS assay) by 30% (MAIC30), after 48 h incubation.
Pearson’s correlation coefficient shown in brackets.
d
J1 ,2 = 9.9 Hz, 2H, 2 ꢀ H-10), 4.2–4.0 (m, 10H, 2 ꢀ H-50, 2 ꢀ H2-60,
2 ꢀ ArCH2), 2.19, 2.06, 2.05, and 2.00 (four s, 24H, 8 ꢀ CH3). 13C
NMR (CDCl3): d 170.3, 170.1, 170.0, and 169.3 8 ꢀ CO, 137.2 C-
1,3, 129.9 (C-2), 128.7 and 128.5 (C-4-6), 89.3 (2 ꢀ C-10), 74.7,
71.7, 67.1, and 66.6 (2 ꢀ C-20-50), 61.5 (2 ꢀ C-60), 44.4 (2 ꢀ ArCH2),
20.73, 20.67, and 20.5 (8 ꢀ CH3). Anal. Calcd for C36H46O18S4
(895.0): C, 48.31; H, 5.18. Found: C, 48.15; H, 5.27.
0
0
4.2. Standard procedure for the preparation of bis(disulfides)
4–6, and 8
A solution of 1 mmol of bis(sulfoxide) 20 or 21 and 4 mmol of 2,3,
4,6-tetra-O-acetyl-1-thio-b-D-gluco- or -galactopyranose (2 mmol
for each sulfinyl function) in 20 mL of 1,2-dichloroethane was main-
tained under stirring, at reflux temperature (83 °C). The reaction was
monitored via TLC and 1H NMR. All the obtained disulfides 4–6 and 8
were purified by flash chromatography. The excesses of glycosyl
thiols were always recovered from the column, together with 5%
maximum of octaacetyldiglycosyl disulfides.
4.2.3. 1,4-Di-{[(2,3,4,6-tetra-O-acetyl-b-D-glucopyranosyl)
dithio]methyl}benzene 6
Yield 60%. TLC: Rf 0.74 (eluant EtOAc/MeOH 9.5:0.5). Mp 208–
212 °C. 1H NMR (CDCl3): d 7.26 (s, 4H, ArH), 5.34 (t, Jvic = 9.4 Hz,
2H) and 5.14 (t, Jvic = 9.7 Hz, 2H) (2 ꢀ H-20,40), 5.25 (t, Jvic = 9.1 Hz,
4.2.1. 1,3-Di-{[(2,3,4,6-tetra-O-acetyl-b-D-glucopyranosyl)dithio]
methyl}benzene 4
2H, 2 ꢀ H-30), 4.52 (d, J1 ,2 = 9.9 Hz, 2H, 2 ꢀ H-10), 4.28 and 4.19
0
0
0
0
0
0
0
0
Yield 60%. TLC: Rf 0.72 (eluant EtOAc/MeOH 95:5). Mp 178–
180 °C. 1H NMR (CDCl3): d 7.2 (m, 4H, ArH), 5.29 (t, Jvic = 9.4 Hz,
2H) and 5.09 (t, Jvic = 9.6 Hz, 2H) (2 ꢀ H-20,40), 5.20 (t, Jvic = 9.0 Hz,
(split AB system, J5 ,6 A = 4.7, J5 ,6 B = 2.4, J6 A,6 B = 12.5 Hz, 4H,
2 ꢀ H2-60), 4.01 (s, 4H, 2 ꢀ ArCH2), 3.75 (ddd, J4 ,5 = 10.0 Hz, 2H,
2 ꢀ H-50), 2.09, 2.05, 2.03, and 2.02 (four s, 24H, 8 ꢀ CH3). 13C
NMR (CDCl3): d 170.5, 169.4, 169.1, and 170.2 8 ꢀ CO, 136.1 C-
1,4, 129.6 (C-2,3,5,6), 87.7 (2 ꢀ C-10), 76.1, 73.8, 69.1, and 68.0
(2 ꢀ C-20-50), 62.0 (2 ꢀ C-60), 44.0 (2 ꢀ ArCH2), 20.7, 20.64, 20.59,
and 20.56 (8 ꢀ CH3). Anal. Calcd for C36H46O18S4 (895.0): C,
48.31; H, 5.18. Found: C, 48.40; H, 5.08.
0
0
2H, 2 ꢀ H-30), 4.49 (d, J1 ,2 = 10.0 Hz, 2H, 2 ꢀ H-10), 4.23 and 4.15
0
0
0
0
0
0
0
0
(split AB system, J5 ,6 A = 4.4, J5 ,6 B = 2.4, J6 A,6 B = 12.4 Hz, 4H,
2 ꢀ H2-60), 3.96 (s, 4H, 2 ꢀ ArCH2), 3.72 (ddd, J4 .5 = 9.7 Hz, 2H,
2 ꢀ H-50), 2.03, 1.99, 1.98, and 1.96 (four s, 24H, 8 ꢀ CH3). 13C
NMR (CDCl3): d 170.3, 170.0, 169.2, and 168.9 8 ꢀ CO, 136.9 C-
1,3, 130.1 (C-2), 128.6 and 128.4 (C-4-6), 87.4 (2 ꢀ C-10), 75.8,
73.6, 68.8, and 67.8 (2 ꢀ C-20-50), 61.8 (2 ꢀ C-60), 44.0 (2 ꢀ ArCH2),
20.6, 20.5, 20.44, and 20.41 (8 ꢀ CH3). Anal. Calcd for C36H46O18S4
(895.0): C, 48.31; H, 5.18. Found: C, 48.20; H, 5.22.
0
0
4.2.4. 1,4-Di-{[(2,3,4,6-tetra-O-acetyl-b-D-galactopyranosyl)
dithio]methyl}benzene 8
Yield 55%. TLC: Rf 0.67 (eluant EtOAc/MeOH 9.5:0.5). Mp 76–
84 °C. 1H NMR (CDCl3): d 7.27 (m, 4H, ArH), 5.42–5.49 (m, 4H,
0
0
0
0
4.2.2. 1,3-Di-{[(2,3,4,6-tetra-O-acetyl-b-D-galactopyranosyl)
dithio]methyl}benzene 5
2 ꢀ H-20,40), 5.06 (dd, J2 ,3 = 10, J3 ,4 = 2.9 Hz, 2H, 2 ꢀ H-30), 4.49
(d, J1 ,2 = 9.9 Hz, 2H, 2 ꢀ H-10), 3.96–4.18 (m, 10H, 2 ꢀ H-50,
2 ꢀ H2-60, 2 ꢀ ArCH2), 2.19, 2.06, 2.05, and 1.99 (four s, 24H,
8 ꢀ CH3). 13C NMR (CDCl3): d 170.3, 170.1, 170.0, and 169.3
8 ꢀ CO, 136.2 C-1,3, 129.67 (C-2) 129,5 (C-4-6), 89.2 (2 ꢀ C-10),
0
0
Yield 50%. TLC: Rf 0.51 (eluant acetone/CHCl3 1:9). Mp 78–80 °C.
1H NMR (CDCl3): d 7.3 (m, 4H, ArH), 5.5–5.4 (m, 4H, 2 ꢀ H-20,40),
5.08 (dd, J2 ,3 = 10.2, J3 ,4 = 3.3 Hz, 2H, 2 ꢀ H-30), 4.53 (d,
0
0
0
0

3194
P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
74.7, 71.7, 67.1, and 66.6 (2 ꢀ C-20-50), 61.5 (2 ꢀ C-60), 44.2
(2 ꢀ ArCH2), 20.74, 20.68, and 20.56 (8 ꢀ CH3). Anal. Calcd for
a patient with colorectal carcinoma), MCF7 (adenocarcinoma cells
established from a patient with breast cancer) and HepG2 (hepato-
cyte carcinoma cells established from a patient with hepatocellular
carcinoma). In some experiments, peripheral blood mononuclear
cells (PBMCs) from healthy donors, separated by density gradient
centrifugation according to the standard technique, were also used.
All cell lines were cultured in RPMI 1640 medium supplemented
C
36H46O18S4 (895.0): C, 48.31; H, 5.18. Found: C, 48.24; H, 5.01.
4.3. Standard procedure for the preparation of deprotected
disulfides 7, 9, 12, and 14
To a solution of 1 mmol of acetylated disulfide in MeOH/THF
(20 mL, 5:5) aqueous ammonia (5 mL) was added under stirring
and stirring was maintained for 48 h, at room temperature. The
reaction was monitored by TLC (CHCl3/MeOH 9.5:0.5) and 1H
NMR. The solvent was evaporated and the deprotected glycoconju-
gated disulfides 7, 9, 12 and 14 were purified by column chroma-
tography (CHCl3/MeOH 8:2 up to 1:9).
with 5–10% fetal bovine serum (FBS), 2 mM L-glutamine and peni-
cillin-streptomycin at 100 units/mL (GIBCO-Invitrogen, UK). For
THP-1 cells, medium was supplemented with 10 mM HEPES,
1 mM sodium pyruvate and 4.5 g/L glucose (Sigma–Aldrich, USA).
PBMCs were cultured at an initial density of 106 cells/ml, with the
addition of human recombinant interleukin-2 (IL-2), 20 U/ml (Chir-
on corporation, Emeryville, CA). U937, THP-1 and MOLT-3 cell lines
share common characteristics such as rapid growth in suspension,
easily detectable apoptosis, relatively low level of malignancy and
moderate sensitivity to cytotoxic agents. HeLa, HT-29, MCF7 and
HepG2 cell lines share the common characteristics to grow as
adherent cells and to present a relatively high level of malignancy
and resistance to cytotoxic agents.
4.3.1. 1,4-Di-{[(b-D-glucopyranosyl)dithio]methyl}benzene 7
Yield 53%. TLC: Rf 0.49 (eluant CHCl3/MeOH 6:4). Mp 78–80 °C.
1H NMR (CD3OD): d 7.30 (s, 4H, ArH), 4.32 (d, J1 ,2 = 9.4 Hz, 2H,
2 ꢀ H-10), 4.08 (s, 4H, 2 ꢀ ArCH2), 3.9–3.3 (m, 12H, 2 ꢀ H-20-50,
2 ꢀ H2-60). 13C NMR (CD3OD): d 138.0 C-1,4, 130.6 (C-2,3,5,6),
92.0 (2 ꢀ C-10), 82.5, 79.5, 72.8, and 71.4 (2 ꢀ C-20-50), 63.0
(2 ꢀ C-60), 45.3 (2 ꢀ ArCH2). Anal. Calcd for C20H30O10S4 (558.7):
C, 42.99; H, 5.41. Found: C, 43.17; H, 5.18.
0
0
U937 cells expressing the vector containing the murine bcl-2
gene (U937mBCL2) or the pMEP control vector (U937pMEP), have
been previously described.14 They were maintained in culture as
were wild type U937 cells, except for the addition of hygromycin
4.3.2. 1,4-Di-{[(b-
D-galactopyranosyl)dithio]methyl}benzene 9
B (50 lg/mL).
Yield 50%. TLC: Rf 0.47 (eluant CHCl3/MeOH 6:4). Mp 103–
For biological assays, glycoconjugated disulfides 3–15 in DMSO
or in culture medium at a concentration of 20 mM. Cells were inoc-
ulated into 96-well microtiter plates at density of 10 ꢀ 104 or
50 ꢀ 104 cells/well, depending on the doubling time of individual
110 °C. 1H NMR (CD3OD): d 7.32 (s, 4H, ArH), 4.30 (d, J1 ,2 = 9.4 Hz,
2H, 2 ꢀ H-10), 4.10 (s, 4H, 2 ꢀ ArCH2), 3.9–3.5 (m, 12H, 2 ꢀ H-20-50,
2 ꢀ H2-60). 13C NMR (CD3OD): d 138.0 C-1,4, 130.7 (C-2,3,5,6), 93.2
(2 ꢀ C-10), 81.0, 76.2, 70.5, and 70.1 (2 ꢀ C-20-50), 62.8 (2 ꢀ C-60),
45.3 (2 ꢀ ArCH2). Anal. Calcd for C20H30O10S4 (558.7): C, 42.99; H,
5.41. Found: C, 42.77; H, 5.58.
0
0
cell lines, in 100 ll of culture medium containing increasing con-
centrations of the compounds to assay. Cells were then incubated
at 37 °C and 5% CO2 for 24 or 48 h. Control cultures received equiv-
alent volumes of control medium and vehicle.
4.3.3. 1,3,5-Tri-{[(b-
trimethylbenzene 12
Yield 51%. TLC: Rf 0.16 (eluant CHCl3/MeOH 7.5:2.5). Mp 129 °C.
D-glucopyranosyl)dithio]methyl}-2,4,6-
4.4.2. Evaluation of absolute cell number and of percentage
cells showing membrane permeabilization and apoptosis
The absolute number of total cells and the percentage number
of cells showing membrane permeabilization were evaluated by
microscopy analysis in a haemocytometer chamber, using the try-
pan blue exclusion test as a standard viability assay. Apoptosis was
evaluated by morphological analysis followed by staining with
acridine orange and by flow cytometry analysis following staining
with propodium iodide, as previously described.15 For morpholog-
ical analysis, over 600 cells, including those showing typical apop-
1H NMR (CD3OD): d 4.40 (d, J1 ,2 = 9.2 Hz, 3H, 3 ꢀ H-10), 4.32 and
4.26 (AB system, Jgem = 11.4 Hz, 6H, 3 ꢀ ArCH2), 3.9–3.3 (m, 18H,
3 ꢀ H-20-50, 3 ꢀ H2-60), 2.54 (s, 9H, 3 ꢀ ArCH3). 13C NMR (CD3OD):
d 138.5 (C-1,3,5), 132.7 (C-2,4,6), 91.3 (3 ꢀ C-10), 82.6, 79.4, 72.1,
and 71.4 (3 ꢀ C-20-50), 62.9 (3 ꢀ C-60), 42.4 (3 ꢀ ArCH2), 17.1
(3 ꢀ ArCH3). Anal. Calcd for C30H48O15S6 (841.1): C, 42.84; H,
5.75. Found: C, 43.03; H, 5.69.
0
0
4.3.4. 1,3,5-Tri-{[(b-
D-glucopyranosyl)dithio]methyl}-2,4,6-
totic characteristics, were analyzed using
a
fluorescence
triethylbenzene 14
microscope. The identification of apoptotic cells was based on
the presence of uniformly stained nuclei showing chromatin con-
densation and nuclear fragmentation. For flow cytometry analysis,
isolated nuclei were analyzed by fluorescence and by forward- and
side-angle-scatter multiparameter analysis on a Becton Dickinson
FaCScalibur using the CELLQuest II software. A minimum of 5000
events were collected for each sample.
Yield 52%. TLC: Rf 0.14 (eluant CHCl3/MeOH 7:3). Mp 100 °C. 1H
NMR (CD3OD): d 4.42 (d, J1 ,2 = 9.4 Hz, 6H, 3 ꢀ H-10), 4.31 and 4.20
(AB system, Jgem = 11.4 Hz, 6H, 3 ꢀ ArCH2S), 4.0–3.3 (m, 18H,
3 ꢀ H-20-50, 3 ꢀ H2-60), 2.98 (q, Jvic = 7.5 Hz, 6H, 3 ꢀ CH2CH3), 1.28
(t, 9H, 3 ꢀ CH2CH3). 13C NMR (CD3OD): d 145.4 (C-1,3,5), 131.9
(C-2,4,6), 91.2 (3 ꢀ C-10), 82.8, 79.4, 72.3, and 71.7 (3 ꢀ C-20-50),
63.2 (3 ꢀ C-60), 41.7 (3 ꢀ ArCH2S), 24.3 (3 ꢀ ArCH2CH3), 17.1
(3 ꢀ ArCH2CH3). Anal. Calcd for C33H54O15S6 (883.2): C, 44.88; H,
6.16. Found: C, 44.53; H, 5.94.
0
0
4.4.3. Evaluation of metabolic activity
The effects on the metabolic activity were examined by the MTS
colorimetric method, using a commercial kit (MTS, Cell Titer 96
Aqueous One Solution, Promega). The assay was performed accord-
4.4. Biological methods
ing to the manufacturer’s protocol, by directly adding 20 lL of
4.4.1. Cells and treatments
‘CellTiter 96 Aqueous One Solution Reagent’ to the culture wells
at the end of incubation period. After further 4 h incubation, absor-
bance was read at 490 nm.
The biological assays were performed on the following cell lines:
U937 (established from the pleural effusion of a patient with histio-
cytic lymphoma), THP-1 (established from the peripheral blood of a
patient with acute monocytic leukaemia), MOLT-3 (established
from the peripheral blood of a patient with acute lymphoblastic
leukaemia), HeLa (established from cancer cells of a patient with
cervical carcinoma), HT-29 (adenocarcinoma cells established from
4.4.4. Calculation of dose–response indexes
Inhibitory concentration 50 (IC50, drug concentration able to
cause inhibition of the total cell count by 50%), cytotoxic concen-
tration 30 (CC30, drug concentration able to cause 30% cytotoxicity

P. Bonaccorsi et al. / Bioorg. Med. Chem. 20 (2012) 3186–3195
3195
by trypan blue assay), apoptotic concentration 30 (AC30, drug con-
References and notes
centration able to cause 30% apoptotic cells by fluorescence
microscopy analysis of acridine orange stained cells), and meta-
bolic activity inhibitory concentration 30 (MAIC30, drug concentra-
tion able to cause reduction of the formazan product formation,
MTS assay, by 30%) were calculated according to the best-fit curve,
y value versus logx, where y is the value of the examined function
and x is the drug concentration. Results from at least three differ-
ent determinations were used to calculate the dose–response
curve. The 30% level for CC30, AC30 and MAIC30 ensures that values
lie within the concentration range utilized.
1. Koppenol, W. H.; Bounds, P. L.; Dang, C. V. Nat. Rev. Cancer 2011, 11, 325;
Saeland, E.; van Kooyk, Y. Biochem. Soc. Trans. 2011, 39, 388; Lau, K. S.; Dennis, J.
W. Glycobiology 2008, 18, 750.
2. Ghazarian, H.; Idoni, B.; Oppenheimer, S. B. Acta Histochem. 2011, 113, 236;
Tennant, D. A.; Durán, R. V.; Gottlieb, E. Nat. Rev. Cancer 2010, 10, 267;
Oppenheimer, S. B.; Alvarez, M.; Nnoli, J. Acta Histochem. 2008, 110, 6; Fuster,
M. M.; Esko, J. D. Nat. Rev. Cancer 2005, 5, 526.
3. Meutermans, W.; Le, G. T.; Becker, B. ChemMedChem 2006, 1, 1164.
4. Opferman, J. T. Cell Death Differ. 2008, 15, 234; Melet, A.; Song, K.; Bucur, O.;
Jagani, Z.; Grassian, A. R.; Khosravi-Far, R. Adv. Exp. Med. Biol. 2008, 615, 47.
5. Aversa, M. C.; Barattucci, A.; Bonaccorsi, P.; Marino-Merlo, F.; Mastino, A.;
Sciortino, M. T. Bioorg. Med. Chem. 2009, 17, 1456.
6. Adinolfi, M.; Capasso, D.; Di Gaetano, S.; Iadonisi, A.; Leone, L.; Pastore, A. Org.
Biomol. Chem. 2011, 9, 6278. and references cited therein.
7. Goto, K.; Tokitoh, N.; Okazaki, R. Angew. Chem., Int. Ed. Engl. 1995, 34, 1124.
8. Aversa, M. C.; Barattucci, A.; Bonaccorsi, P. Eur. J. Org. Chem. 2009, 6355. and
references cited therein.
9. Aversa, M. C.; Barattucci, A.; Bonaccorsi, P. Synlett 2011, 254.
10. Hasegawa, T.; Numata, M.; Okumura, S.; Kimura, T.; Sakurai, K.; Shinkai, S. Org.
Biomol. Chem. 2007, 5, 2404.
11. Bonner, W. A.; Khan, J. E. J. Am. Chem. Soc. 1951, 73, 2241.
12. Plati, J.; Bucur, O.; Khosravi-Far, R. J. Cell. Biochem. 2008, 104, 1124; Cory, S.;
Huang, D. C. S.; Adams, J. M. Oncogene 2003, 22, 8590.
4.4.5. Statistical analysis
Data analysis was performed using the SPSS statistical software
system (version 17.0 for Windows, Chicago, IL). Comparison of
means among sample groups was carried out using Hochberg’s
GT2 post-hoc multiple comparison One-Way ANOVA test. Differ-
ences were considered significant at p <0.05 and highly significant
at p <0.001.
13. Fukumori, T.; Kanayama, H.; Raz, A. Drug Resistance Updates 2007, 10, 101.
14. Borner, C. J. Biol. Chem. 1996, 271, 12695.
15. Mastino, A.; Sciortino, M. T.; Medici, M. A.; Perri, D.; Ammendolia, M. G.; Grelli,
S.; Amici, C.; Pernice, A.; Guglielmino, S. Cell Death Differ. 1997, 4, 629;
Matteucci, C.; Grelli, S.; De Smaele, E.; Fontana, C.; Mastino, A. Cytometry 1999,
35, 45.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.03.070.