Identification of a new tamoxifen-xanthene hybrid as pro-apoptotic anticancer agent

Article abstract of DOI:10.1016/j.bioorg.2019.02.017

Breast cancer is the most diagnosed type of cancer among women for which an exhaustive cure has not been discovered yet. Nowadays, tamoxifen still represents the gold standard for breast cancer therapy; it acts on both estrogen receptor-positive and estrogen receptor-negative breast cancers. Unfortunately, its toxicity and the related chemoresistance undermine its antitumor potential. In this paper, new tamoxifen-based derivatives with a rigid structural motif in their structure were designed, synthesized, and evaluated to assess their antitumor behavior. All the tested compounds affected estrogen receptor-positive tumor (MCF-7) cell growth, even with different extents, among which, the most active ones proved also to induce mitochondria-mediated apoptosis through activation of PARP cleavage, decrease in Bax/Bcl-2 ratio and increase in Bim gene expression levels. Here we found that the compound 1, carrying a rigid xanthene core, turned out to be the most promising of the set showing an activity profile comparable to that of tamoxifen. Furthermore, a more favorable genotoxic profile than tamoxifen made compound 1 a promising candidate for further studies.

Full text of DOI:10.1016/j.bioorg.2019.02.017

Accepted Manuscript
Identification of a new tamoxifen-xanthene hybrid as pro-apoptotic anticancer
agent
Elena Catanzaro, Francesca Seghetti, Cinzia Calcabrini, Angela Rampa, Silvia
Gobbi, Piero Sestili, Eleonora Turrini, Francesca Maffei, Patrizia Hrelia,
Alessandra Bisi, Federica Belluti, Carmela Fimognari
PII:
S0045-2068(18)31065-4
DOI:
https://doi.org/10.1016/j.bioorg.2019.02.017
YBIOO 2792
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
24 September 2018
18 January 2019
6 February 2019
Please cite this article as: E. Catanzaro, F. Seghetti, C. Calcabrini, A. Rampa, S. Gobbi, P. Sestili, E. Turrini, F.
Maffei, P. Hrelia, A. Bisi, F. Belluti, C. Fimognari, Identification of a new tamoxifen-xanthene hybrid as pro-
apoptotic anticancer agent, Bioorganic Chemistry (2019), doi: https://doi.org/10.1016/j.bioorg.2019.02.017
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.

Identification of a new tamoxifen-xanthene hybrid as pro-apoptotic anticancer
agent
Elena Catanzaroa, Francesca Seghettib Cinzia Calcabrinia, Angela Rampab, Silvia Gobbib,
Piero Sestilic, Eleonora Turrinia, Francesca Maffeia, Patrizia Hreliad, Alessandra Bisib,
Federica Bellutib*^, Carmela Fimognaria*^.
aDepartment for Life Quality Studies, Alma Mater Studiorum-University of Bologna,
Corso d’Augusto 237, 47921 Rimini, Italy.
bDepartment of Pharmacy and Biotechnology, Alma Mater Studiorum-University of
Bologna, Via Belmeloro 6, 40126 Bologna, Italy;
cDepartment of Biomolecular Sciences, University of Urbino Carlo Bo, Via I Maggetti
26, 61029 Urbino (PU), Italy
dDepartment of Pharmacy and Biotechnology, Alma Mater Studiorum-University of
Bologna, Via Irnerio 48, 40126 Bologna, Italy
*Correspondence: federica.belluti@unibo.it; Tel.: +390512099732;
carmela.fimognari@unibo.it; Tel.: +390541-434-658
^These authors contributed equally to this work.
Abbreviations
ER, estrogen receptor; FCC, flash column chromatography; i.m., isomers mixture;
PARP, poly (ADP-ribose) polymerase-1; SERM, selective estrogen receptor modulator;
TAM, tamoxifen

Abstract
Breast cancer is the most diagnosed type of cancer among women for which an
exhaustive cure has not been discovered yet. Nowadays, tamoxifen still represents the gold
standard for breast cancer therapy; it acts on both estrogen receptor-positive and estrogen
receptor-negative breast cancers, unfortunately, its toxicity and the related chemoresistance
undermine its antitumor potential. In this paper, new tamoxifen-based derivatives with a
rigid structural motif in their structure were designed, synthesized, and evaluated to assess
their antitumor behaviour. All the tested compounds affected estrogen receptor-positive
tumor (MCF-7) cell growth, even with different extents, among which, the most active ones
proved also to induce mitochondria-mediated apoptosis through activation of PARP
cleavage, decrease in Bax/Bcl-2 ratio and increase in Bim gene expression levels. Here we
found that the compound 1, carrying a rigid xanthene core, turned out to be the most
promising of the set showing an activity profile comparable to that of tamoxifen.
Furthermore, a more favorable genotoxic profile than tamoxifen made compound 1 a
promising candidate for further studies.
Keywords:
Apoptosis; Bax, Bcl-2, breast cancer; genotoxicity; γ-H2A.X phosphorylation, PARP
cleavage, tamoxifen; triphenylethylene; xanthene scaffold
2

1. Introduction
Cancer, among the non-communicable diseases, is the second leading cause of death, it
is responsible for an estimated 18.1 million new cases and 9.6 million cancer deaths in
2018 [1]. Moreover, among the various cancer subtypes, breast cancer is the second most
reported malignancy and the first within women with 2.09 million cases, 1 out of 4 women
was diagnosed with breast cancer, with a survival rate greater than 85% [2, 3]. Indeed,
improvements in prevention, early detection and innovative treatment approaches have
significantly increased the number of positive outcomes, even if the heterogeneity of the
disease and the limitations of current treatments still make breast cancer a significant
concern for public health.
Antitumor therapeutics often exert their effects by inducing apoptosis in cancer cells [4].
Apoptosis, or programmed cell death, is an essential, evolutionarily conserved process that
exerts an essential influence in development, homeostasis, but also in carcinogenesis, of
mammary gland cells. It is carried out by caspases, which are known to be activated mainly
by the extrinsic and/or the intrinsic pathway. The first of these involves the interaction of a
death receptor, a cell surface transmembrane receptor with specific protein ligands, while
the second is triggered in response to DNA damage and depends on the involvement of
mitochondria. Recent outcomes pointed out to a link between the extrinsic and intrinsic
pathways, providing support to the concept of converging instead of distinct routes [5].
Bcl-2 family of proteins controls the intrinsic apoptotic pathway and is composed by the
anti-apoptotic ‘guardian’ proteins, among which BCL-2 is responsible for programmed cell
death evasion, the pro-apoptotic ‘effectors’ BAX and BAK, and the BH3-only ‘sensor’
3

proteins. Approximately 75% of primary breast cancers express high levels of BCL-2, with
a prevalence in ER-positive tumors (85% overexpression). Thus, downregulation of BCL-2
expression can be regarded as a novel and promising anticancer strategy. Tamoxifen
((Z)-1-{4-[2-(dimethylamino)ethoxy]phenyl}-1,2-diphenyl-1-butene, TAM), is a
first-generation selective estrogen receptor (ER) modulator widely used in the treatment of
both advanced and early-stage breast cancer [6]. Since its approval in 1977, it has remained
the gold standard for hormonal therapy [7], and in 1998 its use has been also approved to
lower the incidence of breast cancer for both pre- and post-menopausal women at “elevated
risk” [8]. Noteworthy, in post-menopausal women its use is controversial, as a number of
side effects, including an increased occurrence of endometrial carcinomas, have been
reported [7, 9], mainly due to its metabolic conversion to reactive intermediates such as
4-hydroxyTAM [10] or α-hydroxy-TAM [11]. Besides, intrinsic or acquired drug resistance
can significantly reduce TAM therapeutic efficacy, leading to recurrences [12].
Interestingly, TAM efficacy in ER-negative malignancies has been also reported [8, 13],
leading to speculate on an ER-independent anticancer mechanism, that clearly widens the
therapeutic potential of this drug. A number of studies confirmed the modulation of
different signaling proteins and the induction of distinctive cellular morphological
modifications consistent with an induction of apoptosis. In this respect, both the extrinsic
and intrinsic pathways are involved, with the activation of caspases -8, -9, -3 and PARP
[14] and the modulation of oncoproteins belonging to the BCL-2 family [15], respectively.
Extensive medicinal chemistry efforts have been devoted at developing novel
TAM-based analogues, with reduced side effects, and with increased effectiveness, by
modifying the trans-triphenylethylene (TPE) framework. [16] A number of derivatives
4

endowed with expanded biological activities have been developed, by means of the
hybridization approach [17, 18].
In this scenario, aimed at identifying new lead-like molecules that would serve as
chemical platform to be further modified in order to elucidate the mechanism underlying
TAM biological effect, some established medicinal chemistry design strategies were applied
to create a small set of structurally different TPE-based analogues. In particular, 1)
conformational restriction and 2) scaffold decoration approaches were pursued. As depicted
in Figure 1, to reduce the flexibility: 1-a) the TPE core was merged with a xanthene moiety
to obtain the compound 1; 1-b) an aminobutinyloxy side chain, instead of the aminoethoxy,
was introduced (2-3); in this set, the importance of the trans-TPE configuration with
respect to the cis was also evaluated (2 vs 2a). Moreover, properly selected substituents
were incorporated on the TPE scaffold: 2-a) to mimic the active metabolite 4-OH TAM, the
4-hydroxylated analogue of 3 was also synthesized (4); 2-b) the TPE nucleus was decorated
with fluorine atoms (compound 5), since fluorinated compounds often have more favorable
pKa and lipophylicity that may be useful in enhancing therapeutic efficacy [19].
As regards the amino functions, cyclic amines were preferentially introduced, as they are
known to maintain the activity of the dimethylamino moiety, while being subjected to a
more favorable metabolic fate [20, 21].
The newly synthesized compounds 1-5 were tested to evaluate their antitumor profile in
terms of cytotoxicity and pro-apoptotic potential.
5

Figure 1. Structure of TAM and design overview of the newly synthetized analogues (1-5
and 2a).
2. Results and Discussion
2.1. Compounds synthesis
As traditionally reported for TAM-related compounds, in this paper trans and cis are
used to designate the relative positions of the ethyl group on the olefinic bond and the aryl
ring bearing the 4-alkoxyamino function (A2B2). Whether a compound of given
configuration by this nomenclature is Z or E depends on the nature of the 4-substituent [21].
6

Our attention was mainly focused on obtaining the trans isomers, due to the structural
analogy with TAM.
The desired isomers (intermediates 7, 13, 13a, 14, and 15) were obtained in a good grade
of purity by following different strategies: fractioned crystallization by suitable solvent (14)
or flash column chromatography (FCC), upon previous transformation into the
corresponding perfluorotolyl ether derivative (7, 13, 13a, and 15), as already reported [22].
The cis-isomers, as mentioned above, were labelled with the “a” letter after the compound
number.
As depicted in Scheme 1, 3-hydroxy-9H-xanthen-9-one was reacted with octafluotoluene,
using a phase transfer catalyst, to obtain intermediate 6, that was then coupled with
propiophenone under the McMurry reaction conditions [23] affording 7 i.m. (isomer
mixture) of compounds 7 (isomer Z, or trans) and 7a (isomer E, or cis). This was then
subjected to FCC to obtain the desired trans isomer 7. The perfluorotolyl protecting group
was removed by treatment with NaOMe to give the phenol 8, that was then alkylated with
1,2-dibromoethane, in acetone and in the presence of K2CO3, affording 9. Finally, it was
reacted with pyrrolidine, at room temperature and in the dark (to avoid cis-trans
isomerization), to give the desired compound 1. Scheme 2 reports the synthetic strategy to
obtain the TPE-based phenols (13, 13a, 14, and 15) as pure isomers. In detail,
4-fluorophenyl-4-hydroxyphenylmethanone 11 and 4-fluoropropiophenone were reacted
under the McMurry reaction conditions to give 14 i.m.; fractionated crystallization from
isopropanol allowed isolating the pure trans isomer 14. Conversely, reaction of
(4-hydroxyphenyl)(phenyl)methanone
10
and
(4-hydroxyphenyl)(4-methoxyphenyl)methanone 12 with propiophenone gave 13 i.m. and
15 i.m., respectively. Then, 13 i.m. and 15 i.m. were converted into the corresponding
7

perfluorotolyl ethers 16 i.m. and 17 i.m., respectively, that were subjected to FCC to obtain
the desired pure isomers: 16 (Z), 16a (E), and 17 (Z). The perfluorotolyl function was then
removed to achieve the desired hydroxy derivatives 13, 13a and 15, that were reacted with
propargyl bomide, under the classic Williamson reaction conditions (Scheme 3) to obtain
18 (Z), 18a (E), 19 and 20. Finally, the Mannich reaction, in the presence of formaldehyde,
the selected amine, and CuSO4 as catalyst, afforded the desired compounds 2 (Z), 2a (E), 3,
5, and 21, whose methoxy group was then cleaved by treatment with BBr3 affording
compound 4.
1
The configuration of the synthetized molecules was readily assigned by H NMR
spectroscopy following relative positions of the aryl proton signals arising from the A2B2
system of the 4-substituted phenyl ring and/or the OCH2 signal arising from the side chain.
In particular, in the trans isomer the above signals are considerably upfield in comparison
to those of the corresponding cis isomer [24]. Similarly, the proton signals of the basic side
chain resonate at a lower frequency in the trans-isomer, with respect to the corresponding
cis-isomer [24-26]. 1D-NOESY experiments were also performed on 1 and 18 to further
confirm the configuration assignment (See supporting information Figures S5 and S7,
respectively).
Scheme 1. Synthetic strategy for compound 1.
8

-
Reagents and conditions: i) C6F5CF3, NaOH (aqueous), (nBu)4N+HSO4 , DCM, 1M
NaOH; ii) propiophenone, TiCl4, Zn, THF, -15 °C-reflux; iii) NaOCH3, DMF, r.t.; iv)
1,2-dibromoethane, K2CO3, acetone, r.t.; v) pirrolidine, Et3N, THF, r.t.
Scheme 2. Synthetic strategy for TPE-based synthons (12, 12a, 13, 14).
9

Reagents and conditions: i) selected propiophenone, TiCl4, Zn, THF, -15 °C-reflux; ii)
-
fractioned crystallization; iii) C6F5CF3, (nBu)4N+HSO4 , DCM, 1M NaOH; iv), NaOCH3,
DMF r.t.
Scheme 3. Synthetic strategy for tested compounds 2, 2a, 3-5.
10

Reagents and conditions: i) 3-bromoprop-1-yne, K2CO3, acetone, reflux; ii) HCHO,
CuSO4, selected amine; iii) BBr3, DCM, -78 °C-r.t.
2.2. Citotoxic Activity
Firstly, the newly synthesized derivatives 1-5 were evaluated for their antitumor potential
on ER-positive brest cancer cell line (MCF-7) in comparison to TAM (Table 1). Cell
viability after 24 h treatment, determined by MTT assay, was effectively decreased by
analogues 1, 2, and 3 in a dose-dependent manner. In particular, at the highest tested
concentration of 40 µM compound 1, like TAM, killed almost all cells. Derivatives 2, 2a
and 3 appreciably modulated cell viability at the concentrations of 20 and 40 µM reaching a
plateau. On the contrary, derivative 4 and 5 significantly decreased MCF-7 cell viability
only at the highest concentration of 40 µM (Figure S1). Definitively, the xanthene-based
hybrid, 1 turned out to be the most active compound in this set, showing an IC50 value
comparable to that of TAM (12.4 µM and 10.4 µM, respectively). A slightly decreased
potency was observed for analogues 2 and 3, bearing a rigid butinyl side chain, that
maintained a two-digit micromolar potency (IC50s = 20.7 and 19.3 µM, respectively). As
11

expected, for the corresponding cis-isomer 2a, a further drop in potency was recorded.
Finally, the decoration of the TPE scaffold with an OH group (4) or fluorine atoms (5)
proved to be detrimental (IC50 values around 40 µM).
To assess whether the cytotoxic activity was dependent on ERs, the most promising
derivative 1 was selected to be evaluated on the MDA-MB-231 ER-negative cell line. In
particular, at 40 µM, the percentages of viable cells were 18 and 25 (Figure S2) for 1 and
TAM, respectively. A remarkable antiproliferative effect, comparable to that of TAM, was
thus observed, with IC50 values of 25.4 and 28.9 µM, respectively (Table 1). The low
efficacy of TAM on ER-negative tumors [27] was confirmed in our experimental settings:
in this cell line TAM’s IC50 was about three times higher than that observed in MCF-7 cells.
A similar but less marked trend was observed for 1, whose IC50 on MDA-MB-231 cells was
about twice as high as that observed on ER-positive cells. The comparable behavior of
TAM and 1 on both ER-positive and -negative cell lines might likely suggest their
capability to modulate some ER-independent pathways.
From these data some considerations could be drawn: a) the introduction of a rigid motif
on the TAM side chain proved to slightly decrease the antiproliferative effect on
ER-positive cells, as in compounds 2 and 3; b) as already known for TAM-related
compounds, the nature of the amino function did not play a pivotal role for cytotoxicity; c)
the insertion of a hydroxy group on the main scaffold of 3, as in analogue 4, induced a
further drop in potency; d) the di-fluorinated TPE substitution pattern proved to negatively
affect the cytotoxic profile (5); e) the locked TPE scaffold obtained by its hybridization
with a xanthene core, allowed to retain an appreciable anticancer activity (compound 1).
12

Table 1. Cytotoxic activities of TAM-based derivatives (1-5) and tamoxifen (TAM)
ER-positive
MCF-7
ER-negative
MDA-MB-231
Structure
Comp
1
IC50 (µM)a
12.4 ± 0.54
20.7 ± 4.05
31.3 ± 2.90
19.3 ± 2.22
39.70 ± 3.59
25.4 ± 0.40
n.d.
2
2a
3
n.d.
n.d.
n.d.
4
13

40.81 ± 3.02
n.d.
5
10.5 ± 0.76
28.9 ± 1.35
TAM
aIC50 values were obtained after 24 h treatment of MCF-7 or MDA-MB-231 cell lines; n.d.:
not determined.
2.3. Apoptosis Studies
To get insight into the molecular basis of the cytotoxic activity of the most interesting
TAM-based derivatives (1, 2 and 3) apoptotisis studies, on MCF-7 cell line, were
performed. In particular, the underlying mechanisms of the apoptotic event, by evaluating
the mRNA levels and/or protein expression of PARP, Bax, Bcl-2, p53, and Bim, key
mediators involved in the execution of this process, were explored.
PARP is an indicator of DNA single strand break and one of the most important
reporters of caspase-3 activation. During the early stages of apoptosis, activated-caspase-3
cleaves PARP in two fragments, leading to DNA binding and interfering with DNA repair
enzymes [28]. All the analyzed derivatives induced PARP cleavage to a different extent
(data not shown), being compound 1 the most effective, showing a behavior comparable to
TAM. At the highest tested concentration (10 µM), TAM and 1 induced a 1.85 and 1.35
fold increase in cleaved PARP, respectively, versus non-treated cells (Figure 2). As also
14

observed in the cytotoxicity studies, the xanthene derivative 1 again proved to be slightly
less active with respect to TAM.
Figure 2. Fold-increase of cleaved PARP compared to untreated cells after 24 h treatment
of MCF-7 cells with TAM or 1. Cleavage of PARP was determined by flow cytometry
using an antibody specific for the 85 kDa PARP cleaved portion. The bars represent the
mean values of four experiments ± SEM; the significance level compared to the control was
specified as p < 0.01 (**), p < 0.001 (***) using one way ANOVA and the Dunnett’s
multiple comparison test.
Bcl-2 and Bax are homologous proteins that play a central role in inducing mitochondrial
apoptosis: while Bcl-2 is an ubiquitous inhibitor of apoptosis, Bax is a promoter of cell
death by inhibiting the pro-survival role of Bcl-2. Treatment of MCF-7 cells with the
selected compounds did not affect Bax gene expression as shown in Figure 3.
15

Figure 3. Effect of TAM, 1, 2 and 3 on Bax gene expression compared to untreated
cells after 24 h treatment of MCF-7. Bax gene expression was determined by real time PCR.
Differences in mRNA expression are reported as value ± SEM and represent the relative
expression calculated through the 2-ΔΔCt method. The bars represent the mean values of four
experiments ± SEM.
In the same experimental conditions, an overall decrease in Bcl-2 protein and mRNA
levels was observed for all tested derivatives. Also in this context, 1 proved to be the most
potent inhibitor of the series as, at 10 µM concentration, it elicited effects comparable to
those observed for TAM treatment. To obtain the same effect, a double concentration was
required for derivatives 2 and 3 (Figure 4).
16

Figure 4. Effect of TAM, 1, 2 and 3 on Bcl-2 expression (A. gene or B. protein) compared
to untreated cells after 24 h treatment of MCF-7. A. Bcl-2 gene expression was determined
by real time PCR. Differences in mRNA expression are reported as value ± SEM and
represent the relative expression calculated through the 2-ΔΔCt method. B. Bcl-2 protein
expression was determined by flow cytometry using a specific antibody. The bars represent
the mean values of four experiments ± SEM; the significance level compared to the control
was specified as p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) using one way ANOVA and
the Dunnett’s multiple comparison test.
Due to the opposite effects of Bax and Bcl-2, the Bax/Bcl-2 ratio correlates with the
propensity of a cell to undergo apoptosis and may be helpful in predicting clinical outcome
[29]. In this respect, for all tested derivatived an increase of Bax/Bcl-2 ratio was noticed
suggesting their feasibility to circumvent cancer cells apoptosis deficiency by affecting
tumor progression and aggressiveness.
P53 is a tumor suppressor gene that plays a key role in cell cycle progression and
apoptosis and, among other mechanisms, has been shown to act through Bcl-2-silencing
[30]. It is mutated in more than 50 % of tumors [31] but, despite this high susceptibility and
frequency, only 20-30 % of breast cancers (and most ER-negative tumors) are characterized
by this mutation [32]. Several studies have been performed to elucidate the mechanism
17

underlying Bcl-2 expression in ER-positive cell lines and whether its downregulation is
associated with p53 intervention [33]. Different results were found in literature: in 1994,
Haldar and colleagues [30] reported a possible role of wild-type p53 in TAM-induced
apoptosis and an association with Bcl-2 expression, whereas in 1998 Fattman and
colleagues [29] and Zhang et al. in 1999 [33] proposed a p53-independent pathway.
The results obtained with the new derivatives agreed with this latter hypothesis, since
no variation in p53 gene or protein expression has been registered (data not shown), leading
to speculate that Bcl-2 downregulation is independent of a modulation at the transcriptional
level of p53. However, considering that some recent studies showed that p53 interacts with
ER and that such functional interaction results in the suppression of the p53-mediated
apoptotic response [34], the employment of an ER-positive cell line in this study could be
responsible for the lack of p53 overexpression observed after TAM and TAM-analogues
treatments.
To deeply investigate the mechanism responsible for the antitumor effect of derivatives
1-3, the involvement of Bim was also studied. This ‘BH3-only’ subclass of pro-apoptotic
proteins, intrinsically responsible for the initiation of apoptosis, is also a cytoskeletal
integrity sensor. It represents a key factor in mitochondria-mediated apoptosis since it
interacts with Bax and causes the release of apoptogenic factors that in turn activate
caspase-3 [35]. We therefore evaluated whether Bim expression was upregulated in
response to treatment with the selected compounds. As depicted in Figure 5, the tested
compounds determine an increase in Bim gene expression levels, leading to speculate on a
involvement of ‘BH3-only’ in regulating mitochondria-mediated apoptosis of the newly
synthetized TAM-based analogues 1-3. However, TAM shows a concentration-dependent
trend on BIM modulation, while compounds 1, 2 and 3 promoted it in a different course.
18

The effect of compound 2 at 10 µM was not statistically different to the effect of the same
compound at 20 µM. The same applies for compound 3. The low modulation of BIM could
reflect the poor cytotoxic potential of compounds 2 and 3 compared to that of TAM, while
a different mechanism of action could characterize compound 1. Indeed, compound 1
showed another different trend. At 10 µM it upregulated BIM at least doubling its
physiological levels, while at the highest tested concentration it didn’t modulate it. BIM has
been targeted as a critical protein for TAM mechanism of action. As an example, the
downregulation of this protein is one of the causes of TAM resistance [36]. To understand
compound 1 odd mechanism of action, further studies will be needed. For instance, it
would be worth evaluating the modulation of Forkhead box, class O 3a (foxO3a) by
compound 1. In fact, FoxO3a is accountable for the transcriptional up-regulation of BIM
that promotes apoptosis of breast cancer cells mediated by TAM [37]
Figure 5. Effect of TAM, 1, 2 and 3 on Bim gene expression compared to untreated
cells after 24 h treatment of MCF-7 cells. A. Bim gene expression was determined by real
time PCR. Differences in mRNA expression are reported as value ± SEM and represent the
relative expression calculated through the 2-ΔΔCt method. B. Bim protein expression was
19

determined by flow cytometry using a specific antibody. The bars represent the mean
values of four experiments ± SEM; the significance level compared to the control was
specified as p < 0.05 (*), p < 0.001 (***) using one way ANOVA and the Dunnett’s
multiple comparison test.
In summary, the results suggest that compounds 1-3, similarly to TAM, trigger
apoptosis via an appropriate balance between pro-apoptotic and pro-survival proteins.
Among the tested compounds, the xanthene-TAM hybrid 1, showed the most
promising anticancer effects.
2.4. Genotoxicity
In order to elaborate a preliminary risk/benefit profile, the genotoxicity of 1, the most
promising compound of this series, was investigated in Jurkat cells in comparison to TAM.
DNA damage was assessed through the quantification of histone H2A.X (γ-H2A.X)
phosphorylation, a biomarker of double-strand DNA breaks. Histone phosphorylation
provides a good indication of the ability of a xenobiotic to interact with DNA and cause
pre-mutational lesions, i.e. repairable per se. After 6 h, at 10 or 20 µM treatment with 1,
Jurkat cells showed a 1.4- and 1.2-fold increase in the phosphorylation of γ-H2A.X,
respectively, compared to non-treated cells. Under the same conditions, TAM induced an
increase of 1.7 and 1.5, respectively (Figure 6).
20

Figure 6. Phosphorylation of histone (γ-H2A.X) induced by TAM or 1 in MCF-7 cells
after 6 h treatment. Histone phosphorylation was analyzed by flow cytometry using an
anti-γ-H2A.X-Alexa Fluor® antibody. The bars represent the mean values of four
experiments ± SEM; the significance level compared to the control was specified as p <
0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****); compared to TAM 10 µM as p
< 0.01 (°°), p < 0.001 (°°°); and compared to TAM 20 µM as p < 0.05 (•).
The quantitative differences in DNA damage may lie in different metabolic processes
that the two molecules are subjected to [38]. Indeed, TAM bio-activation involves the
production of metabolites, namely a carbocation [39], ortho-quinone [40] and/or quinone
methide [41] able to induce DNA damage, both directly through the formation of adducts,
and/or indirectly, by increasing ROS levels. Several studies reported the mutagenicity of
TAM and its ability to induce micronuclei formation in metabolically competent human
cells [42, 43]. The peculiar structural features of 1, in particular the presence of the
oxygenated scaffold and the cyclic amino function in the side chain, could favorably affect
its metabolic fate, reducing this well-known TAM’s drawback.
3. Conclusions
21

A small series of TPE-based derivatives was designed, synthesized and tested to
evaluate their anticancer potential. Some structural modifications were performed in order
to reduce the flexibility of different peculiar TAM fragments i.e. the aminoethoxy side
chain and the TPE core. All the newly synthesized compounds were able to inhibit the
proliferation of the ER-positive breast cancer cell line MCF-7, even if to a different extent.
The TPE-based analogs bearing a rigid butinyl side chain (2, 2a, 3 and 4) turned out to be
less active than TAM, while the TPE-constrained compound 1 proved to maintain the same
range of activity and emerged as the most potent of the series, maintaining an appreciable
activity in ER-negative MDA-MB-231 cell line as well. In particular, while TAM was
significantly more active on ER-positive with respect to ER-negative cells (IC50 values of
10.5 and 28.9, respectively), only a two-fold drop in potency shifting from ER-positive to
ER-negative cells was observed for compound 1 (IC50 values of 12.4 and 25.4,
respectively). This trend suggests that 1 could exert its cytotoxic effect involving some
ER-independent pathways. Particular attention was given at elucidating the molecular basis
of the pro-apoptotic behavior (1, 2 and 3) on MCF7 cells. The recruitment of caspase-3,
confirmed by the activation of PARP cleavage, was induced by all derivatives, being 1 the
most active. Moreover, a downregulation of the antiapoptotic guardian Bcl-2 gene
expression was also observed for all compounds, again particularly pronounced for 1. This
analogue also proved to modulate Bim gene expression levels, providing evidence that the
induction of apoptosis is a pivotal mechanism by which compound 1 exert its anticancer
effect. Moreover, a more favorable genotoxic profile was observed for this compound with
respect to TAM, pinpointing this new molecule as a promising lead compound, worth to be
further studied and optimized. In particular, further carefully designed studies on normal
22

non-transformed cells are necessary to evaluate whether compound 1 has a selective
activity for cancer cells and to better understand its toxicological profile.
4. Experimental section
4.1. Synthesis
Starting materials, unless otherwise specified in the Experimental Section, were used
as high-grade commercial products. Solvents were of analytical grade. Melting points
were determined in open glass capillaries, using a Büchi apparatus and are uncorrected.
NMR spectra were recorded on a Varian Gemini spectrometer at 400 MHz (1H), 100
MHz (13C), and 376 MHz (19F). Unless otherwise indicated CDCl3 was used as a solvent
and tetramethylsilane (TMS) as an internal standard. Chemical shifts (δ) were reported as
parts per million (ppm value) and standard abbreviations, indicating spin multiplicities,
were given as follow: s (singlet), d (doublet), t (triplet), br (broad), q (quartet) or m
(multiplet). Mass spectra were recorded on a Waters ZQ 4000 apparatus operating in
electrospray mode (ES). Reactions were followed by thin layer chromatography (TLC)
on precoated silica gel plates (Merck Silica Gel 60 F254) and then visualized with a UV
lamp. Purification of synthesized compounds was carried out by flash column
chromatography (FCC) using 100-200 mesh (Merck) silica gel as stationary phase. The
purity of the tested compounds was determined by HPLC analysis. The analyses were
performed under direct phase conditions on a Phenomenex Lux 5u cellulose column (250
× 4.60 mm), by using a ternary mixture of n-hexane/i-PrOH/DEA 90:9.99:0.1 as mobile
23

phase, flow rate: 0.8-1.0 mL/min, the injection volume was 5 μL and peaks were detected
at 254 nm. All tested compounds were found to have >95 % purity. Compounds were
named using Chem-BioDraw Ultra 14.0 IUPAC name algorithm developed by
CambridgeSoft Corporation.
4.1.1. General procedure for the synthesis of the perfluorotolyl ethers (6, 16 i.m., 17
i.m.).
A mixture of phenol derivative (1.0 eq), octafluorotoluene (1.0 eq), tetrabutylammonium
hydrogen sulfate (TBAHS, 0.5 eq) in DCM (10 mL/mmol) and 1.0 M NaOH (10
mL/mmol) was stirred at room temperature for 1 h. The organic phase was separated, dried
over Na2SO4, and the solvent was removed under reduced pressure to give a crude product,
which was subjected to FCC puri cation using a suitable eluent.
4.1.1.1. 3-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)-9H-xanthen-9-one (6).
Starting from 3-hydroxy-9H-xanthen-9-one (0.42 g, 2.0 mmol), octafluorotoluene (0.28
mL, 2.0 mmol), TBAHS (0.34 g, 1.0 mmol) and following the previously described general
procedure a crude was obtained. FCC puri cation using PE/EA 9: 1 as eluent afforded 6 as white
solid, 0.74 g, yield 85 %, mp 88-90 °C. 1H NMR δ 7.01 (d, J = 2.0 Hz, 1H, H-4), 7.07 (dd, J =
2.4 and 8.4 Hz, 1H, H-2), 7.40 (t, J = 8.0 Hz, 1H, Ar), 7.44 (d, J = 8.0 Hz, 1H, H-5), 7.74 (t,
24

J = 8.0 Hz, 1H, Ar), 8.33 (d, J = 8.0 Hz, 1H, Ar), 8.38 (d, J = 8.0 Hz, 1H, Ar). 19F NMR δ
-151.0 (m, 2F), -139.1 (m, 2F), -55.9 (t, J = 21.8 Hz, 3F).
4.1.1.2.
(E,Z)(1-(4-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)phenyl)but-1-ene-1,2-diyl)dib
enzene (16 i.m.).
Starting from 13 i.m. (0.60 g, 1.2 mmol), octafluorotoluene (0.17 mL, 1.2 mmol),
TBAHS (0.20 g, 0.60 mmol) and following the previously described general procedure a
crude was obtained. FCC puri cation using PE/EA 9.85: 0.15 as eluent allowed separation of the
isomers 16 and 16a.
(Z)-(1-(4-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)phenyl)but-1-ene-1,2-diyl)dibenzene
16, Rf = 0.22, white solid, 0.6 g, yield 38 %, mp 142-144 °C. 1H NMR δ 0.97 (t, J = 7.2 Hz, 3H,
CH3), 2.51 (q, J = 7.2 Hz, 2H, CH2), 6.49 (d, J = 8.0 Hz, 2H, H-3 and H-5), 6.75 (d, J = 8.0
Hz, 2H, H-2 and H-6), 7.10-7.22 (m, 4H, Ar), 7.25-7.31 (m, 4H, Ar), 7.35 (d, J = 8.0 Hz,
2H, Ar). 19F NMR δ -152.0 (m, 2F), -139.4 (m, 2F), -55.5 (t, J = 21.8 Hz, 3F).
(E)-(1-(4-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)phenyl)but-1-ene-1,2-diyl)dibe
nzene 16a, Rf = 0.16, white solid, 0.25 g, yield 15 %, mp 126-128 °C. 1H NMR δ 0.97 (t, J = 7.2
Hz, 3H, CH3), 2.47 (q, J = 7.2 Hz, 2H, CH2), 6.78-6.82 (m, 4H, Ar), 6.98-7.28 (m, 4H, Ar),
7.25-7.31 (m, 4H, Ar), 7.38 (d, J = 8.0 Hz, 2H, Ar).
25

4.1.1.3. 1,2,4,5-tetrafluoro-3-(4-(1-(4-methoxyphenyl)-2-phenylbut-1-en-1-yl)phenoxy)-
6-(trifluoromethyl)benzene (17 i.m.).
Starting from 15 i.m. (0.50 g, 1.5 mmol), octafluorotoluene (0.22 mL, 1.5 mmol),
TBAHS (0.21 g, 0.75 mmol) and following the previously described general procedure 17
i.m. was obtained. that was then subjected to FCC (PE/DCM 9:1) to afford pure 17 and 17a.
(Z)-1,2,4,5-tetrafluoro-3-(4-(1-(4-methoxyphenyl)-2-phenylbut-1-en-1-yl)phenoxy)-6-(trif
luoromethyl)benzene 17, Rf = 0.27, white solid, 0.52 g, yield 45 %, mp 78-80 °C. 1H NMR δ
0.98 (t, J = 7.2 Hz, 3H, CH3), 2.52 (q, J = 7.2 Hz, 2H, CH2), 3.84 (s, 3H, OCH3), 6.64 (d, J
= 8.0 Hz, 2H, H-3 and H-5), 6.85 (d, J = 8.0 Hz, 2H, H-2 and H-6), 7.03 (d, J = 8.8 Hz, 2H,
H-3’ and H-5’), 7.10 (d, J = 8.8 Hz, 2H, H-2’ and H-6’), 7.16-7.34 (m, 5H, Ar). 19F NMR δ
-152.0 (m, 2F), -140.6 (m, 2F), -55.8 (t, J = 21.8 Hz, 3F).
(E)-1,2,4,5-tetrafluoro-3-(4-(1-(4-methoxyphenyl)-2-phenylbut-1-en-1-yl)phenoxy)-6-(trif
luoromethyl)benzene 17a, Rf = 0.22, white solid, 0.35 g, yield 22 %, mp 55-58 °C. 1H NMR δ
0.91 (t, J = 7.6 Hz, 3H, CH3), 2.45 (q, J = 7.2 Hz, 2H, CH2), 3.68 (s, 3H, OCH3), 6.54 (dd, J
= 1.6 and 8.4 Hz, 2H, Ar), 6.75 (dd, J = 2.0 and 8.4 Hz, 2H, Ar), 6.96 (d, J = 8.4 Hz, 2H,
Ar), 7.09-719. (m, 5H, Ar), 7.21 (d, J = 8.4 Hz, 2H, Ar).
4.1.2. McMurry Reaction. General Procedure for the synthesis of TPE-based
compounds (7 i.m, 13-15 i.m).
26

TiCl4 (3.0 eq) was added to a stirred suspension of Zn powder (6.0 eq) in dry THF (1.5
mL/mmol), under N2 atmosphere, at -15 °C. The mixture was heated under reflux for 1h.
The suspension was again cooled to -5 °C and the selected carbonyl compounds (1 eq)
dissolved in THF (5 mL/mmol) were slowly added, and the mixture was refluxed for 2h.
After cooling, the reaction mixture was poured into 10 % aqueous K2CO3 and extracted
with Et2O. The organic phase was washed with brine, dried over Na2SO4, and the solvent
was removed under reduced pressure. The crude material was purified by FCC or
fractionated crystallization from suitable solvent.
4.1.2.1. (E,Z)-9-(1-phenylpropylidene)-9H-xanthen-3-ol (7 i.m.).
TiCl4 (0.85 g, 4.5 mmol), Zn (0.60 g, 9.0 mmol), 6 (0.64 g, 1.5 mmol), and
propiophenone (0.20 g, 1.5 mmol) were reacted according to the McMurray general
procedure. The crude of reaction was purified by FCC (n-hexane/EtOAc 9.95:0.05) to
obtain 7 and 7a.
(Z)-9-(1-phenylpropylidene)-3-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)-9H-xant
hene 7, Rf 0.18, transparent oil, 0.27 g, 34 % yield. 1H NMR  0.97 (t, J = 7.6 Hz, 3H,
CH3), 2.81 (q, J = 7.6 Hz, 2H, CH2), 6.26 (dd, J = 2.8 and 9.2 Hz, 1H, H-2), 6.49 (d, J = 9.2
Hz, 1H, H-1), 6.71 (d, J = 2.8 Hz, 1H, H-4), 7.08 (d, J = 6.8 Hz, 2H, Ar), 7.17-7.31 (m, 6H,
Ar), 7.58 (d, J = 7.6 Hz, 1H, Ar).
27

(E)-9-(1-phenylpropylidene)-3-(2,3,5,6-tetrafluoro-4-(trifluoromethyl)phenoxy)-9H-xant
hene 7a, Rf 0.21, transparent oil, 0.21 g, 26 % yield. 1H NMR  0.97 (t, J = 7.6 Hz, 3H,
CH3), 2.77 (q, J = 7.6 Hz, 2H, CH2), 6.52 (dd, J = 2.8 and 8.4 Hz, 1H, H-2), 6.60-6.64 (m,
1H, Ar), 6.84-6.86 (m, 2H, Ar), 7.03-7.09 (m, 4H, Ar), 7.24-7.31 (m, 3H, Ar), 7.55 (dd, J =
1.6 and 7.6 Hz, 1H, Ar).
4.1.2.2. (E,Z)-4-(1,2-diphenylbut-1-en-1-yl)phenol (13 i.m.).
TiCl4 (1.14 g, 6.0 mmol), Zn (0.8 g, 12.0 mmol), 4-hydroxyphenylphenylmethanone 10
(0.39 g, 2.0 mmol), and propiophenone (0.33 g, 2.0 mmol) were reacted according to the
McMurray general procedure. The crude of reaction was purified by FCC (PE/EA 8:2) to
obtain 13 i.m. as 2:1 mixture of Z/E isomers (0.59 g, 95% yield). 1H NMR  0.9-1.03 (m,
3H, CH3), 2.45-2.60 (m, 2H, CH2), 4.55 (s, 0.66H, OH), 4.79 (s, 0.33H, OH), 6.50 (d, J =
8.4 Hz, 1.33H, Ar), 6.72 (d, J = 8.4 Hz, 1.33H, Ar), 6.82 (d, J = 8.8 Hz, 0.66H, Ar), 6.91 (d,
J = 8.8 Hz, 0.66H, Ar), 7.00-7.45 (m, 10H, Ar).
4.1.2.3. (E,Z)-4-(1,2-bis-4-fluorophenylbut-1-en-1-yl)phenol (14 i.m. and 14).
TiCl4 (2.28 g, 12.0 mmol), Zn (1.6 g, 24.0 mmol),
4-fluorophenyl-4-hydroxyphenylmethanone 11 (0.83 g, 4.0 mmol), and 4-F-propiophenone
(0.94 g, 4.0 mmol) were reacted according to the McMurray general procedure. The crude
28

of reaction was purified by fractioned crystallization from isopropyl alcohol to afford the
desired trans isomer (E)-4-(1,2-bis-4-fluorophenylbut-1-en-1-yl)phenol 14 (0.66 g, 83%
yield), mp 95-96 °C. 1HNMR  0.94 (t, J = 7.2 Hz, 3H, CH3), 2.44 (q, J = 7.2 Hz, 2H,
CH2), 3.8 (br, 1H, OH), 6.56 (d, J = 8.8 Hz, 2H, Ar), 6.75 (d, J = 8.4 Hz, 2H, Ar), 6.82-6.94
(m, 2H, Ar), 7.01-7.08 (m, 4H, Ar), 7.10-7.14 (m, 2H, Ar).
4.1.2.4. (E,Z)-4-(1-(4-methoxyphenyl)-2-phenylbut-1-en-1-yl)phenol (15 i.m.).
TiCl4 (1.14 g, 6.0 mmol), Zn (0.8 g, 12.0 mmol),
4-hydroxyphenyl-4-methoxyphenylmethanone 12 (0.46 g, 2.0 mmol), and propiophenone
(0.33 g, 2.0 mmol) were reacted according to the McMurray general procedure. The crude
of reaction was purified by FCC (PE/EA 7:3) to obtain 15 i.m. as 1.5:1 mixture of cis/trans
isomers (0.5 g, 75 % yield). 1H NMR  2.52 (t, J = 7.2 Hz, 3H, CH3), 2.54 (q, J = 7.2 Hz,
2H, CH2), 3.68 (s, 1.2H, OCH3), 3.89 (s, 1.8H, OCH3), 6.42 (d, J = 8.8 Hz, 1.2H, Ar), 6.48
(d, J = 8.8 Hz, 0.8 H, Ar), 6.78 (d, J = 8.8 Hz, 1.2H, Ar), 6.80-6.83 (m, 0.8H, Ar), 6.89 (d, J
= 8.8 Hz, 1.2 H, Ar), 7.02-7.43 (m, 7.8H, Ar).
4.1.3. Removal of the perfluorotolyl function. General Procedure for the synthesis of 8,
13, 13a, 15
29