Synthesis of new hetero-arylidene-9(10H)-anthrone derivatives and their biological evaluation

Article abstract of DOI:10.1016/j.bioorg.2020.103849

New hetero-arylidene-9(10H)-anthrone derivatives (1) were synthesized from reaction of 1,2-dimethyl-3-alkyl imidazolium salts (2) and 9-anthracenecarboxaldehyde. Ion exchange of the anion with dioctyl sulfosuccinate and lithium bis(trifluoromethanesulfonyl)imide led to the preparation of other derivatives. The antiproliferative effect of the compounds was evaluated in human ovarian (A2780) and colorectal (HCT116) carcinoma cell lines and in normal primary human fibroblasts. Compound 1 presented an antiproliferative effect related to the imidazolium pattern of substitution with compounds having a decyl group at the R-position (1c and 3c) showing the highest cytotoxic activities in all cell lines independently of the counter ion. Compounds 1b and 1c internalize A2780 cancer cells via a passive or an active transport, respectively, inducing A2780 cell death via an extrinsic apoptosis (1b) or intrinsic apoptosis and oncosis (1c). The localization of both compounds in the cytoplasm coupled to the absence of reactive oxygen species (ROS) induction suggest that the mechanisms of toxicity might be different than those of other anthracyclines currently used in chemotherapy.

Full text of DOI:10.1016/j.bioorg.2020.103849

Journal Pre-proofs
Synthesis of New Hetero-Arylidene-9(10H)-Anthrone Derivatives and Their
Biological Evaluation
Catarina Roma-Rodrigues, Gabriela Malta, Daniela Peixoto, Luísa M.
Ferreira, Pedro V. Baptista, Alexandra R. Fernandes, Paula S. Branco
PII:
S0045-2068(19)32209-6
DOI:
Reference:
https://doi.org/10.1016/j.bioorg.2020.103849
YBIOO 103849
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
26 December 2019
9 April 2020
10 April 2020
Please cite this article as: C. Roma-Rodrigues, G. Malta, D. Peixoto, L.M. Ferreira, P.V. Baptista, A.R.
Fernandes, P.S. Branco, Synthesis of New Hetero-Arylidene-9(10H)-Anthrone Derivatives and Their Biological
Evaluation, Bioorganic Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.2020.103849
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
©
2020 Published by Elsevier Inc.

Synthesis of New Hetero-Arylidene-9(10H)-
Anthrone Derivatives and Their Biological
Evaluation
[
b]
[a]
[a]
Catarina Roma-Rodrigues, Gabriela Malta, Daniela Peixoto, Luísa M.
[a]
[b]
[b]
[a]
Ferreira, Pedro V. Baptista, Alexandra R. Fernandes,* and Paula S. Branco*
aLAQV-REQUIMTE, DQ, NOVA School of Science and Technology, 2829-516,
Caparica, Portugal.
bUCIBIO, DCV, NOVA School of Science and Technology, 2829-516, Caparica,
Portugal.
To whom correspondence should be addressed: Paula S. Branco, Chemistry
Department, LAQV-REQUIMTE, Faculdade de Ciências e Tecnologia, Universidade
Nova de Lisboa, 2829-516 Caparica, Portugal. Tel: +351 21 2948300; e-mail:
paula.branco@fct.unl.pt; Alexandra R. Fernandes UCIBIO, DCV, Faculdade de
Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal.
Tel: +351 21 2948300; e-mail: ma.fernandes@fct.unl.pt.
1

ABSTRACT
New hetero-arylidene-9(10H)-anthrone derivatives (1) were synthesized from reaction
of 1,2-dimethyl-3-alkyl imidazolium salts (2) and 9-anthracenecarboxaldehyde. Ion
exchange of the anion with dioctyl sulfosuccinate and lithium
bis(trifluoromethanesulfonyl)imide led to the preparation of other derivatives. The
antiproliferative effect of the compounds was evaluated in human ovarian (A2780) and
colorectal (HCT116) carcinoma cell lines and in normal primary human fibroblasts.
Compound 1 presented an antiproliferative effect related to the imidazolium pattern of
substitution with compounds having a decyl group at the R-position (1c and 3c)
showing the highest cytotoxic activities in all cell lines independently of the counter
ion. Compounds 1b and 1c internalize A2780 cancer cells via a passive or an active
transport, respectively, inducing A2780 cell death via an extrinsic apoptosis (1b) or
intrinsic apoptosis and oncosis (1c). The localization of both compounds in the
cytoplasm coupled to the absence of reactive oxygen species (ROS) induction suggest
that the mechanisms of toxicity might be different than those of other anthracyclines
currently used in chemotherapy.
KEYWORDS. : Arylidene-anthrone derivatives • imidazolium chemistry • ovarian
cancer • proliferation inhibition • cell apoptosis.
2

1
. Introduction
Anthraquinones are privileged structures from natural sources, such as pigments,
vitamins, and enzymes. Besides their utilization as dyes, anthraquinone derivatives have
long been used in medical applications, [1, 2] with several anthraquinone derivatives
showing interesting medicinal uses, such as anticancer, antibacterial, anti-inflammatory,
antioxidant, antidiabetic and antiviral [2-4]. The anticancer action of anthraquinone
derivatives has been associated to DNA intercalation due to the flat aromatic
anthraquinone core [5]. Some examples include doxorubicin,[3] daunorubicin and
carminomycin [4]. Despite the widespread use of these compounds in the clinics, there
is a high correlation between anthracycline based chemotherapy and cardiotoxicity, [6]
which has prompted for the search of novel drugs with lower side effects. The anthrone
analogs, 10-substituted benzylidene anthrones are compounds similar to anthraquinone
where a carbonyl group is replaced by a substituted alkene. Some anthrone derivatives
have shown potent in vitro antitumor activity, [7, 8] with some compounds acting as
inhibitors of tubulin polymerization [9, 10]. Syntheses of 10-substituted benzylidene
anthrones has been reported by a straightforward aldol-type condensation reaction of
anthrone and substituted benzaldehydes under acidic or basic conditions [8, 9] or under
microwave irradiation [11]. Recently, we described for the first time a hetero arylidene-
9
(10H)-anthrone (1a) formed by reaction of 9-anthracenecarboxaldehyde with 1,2-
dimethyl-3-ethylimidazolium iodide (2a) under basic conditions (Scheme 1) [12]. The
detailed synthetic mechanism was presented demonstrating that, N-heterocyclic olefins
(NHO) derived from 2 were postulated as reactive species in the pathway for the
formation of 1. As shown for other anthrones, compound 1a intercalates with CT-DNA
5
-1
(
K = 2.0(± 0.20) x 10 M ), being able to displace GelRed from the DNA double-
b
strand, resulting in the quenching of fluorescence of the GelRed-CT-DNA system (KSV
3

3
-1
=
3.3 ±0.3 x 10 M ).[5, 12] Still, compound 1a showed neither cytotoxicity in ovarian
(
A2780) and colorectal (HCT116) carcinoma cell lines nor in human healthy fibroblasts
12]. Since this compound is capable to intercalate within the DNA structure but it does
[
not present any cytotoxicity might be related to its inability to cross the cellular
membrane. We hypothesized that suitably designed derivatives might surpass this
limitation and, herein, we report the preparation and characterization of analogues of 1a
that could show increased antiproliferative effects on tumor cells with low or no effect
in normal cells. From the resulting selective library of compounds, the most interesting
were further studied concerning their biological effects in ovarian carcinoma cells.
2
. Results and Discussion
Synthesis. Two main approaches were followed towards preparation of new hetero
arylidene-9(10H)-anthrones (1 and 3), using the imidazolium ring substituents and the
counter anion as targets for structural diversity. For this purpose, several imidazolium
salts (2) were prepared by alkylation of 1,2-dimethylimidazole using the appropriated
alkyl halide in dry THF, under reflux according to a modified literature procedure of
Blümel [13]. The imidazolium salts precipitated from the reaction medium and were
filtered to provide 2 in good to high yield (65%-95%) (Scheme 1). Spectroscopic data of
2
a-d are in accordance with literature [14, 15]. For the synthesis of 2e, it was necessary
the preparation of the 3-{2-[2-(2-methoxyethoxy)ethoxy]ethyl} iodide from triethylene
glycol monomethyl ether following a reported procedure [16]. The imidazolium salts
2
a-e thus prepared were subsequent used for the preparation of 1a-e following our
previously described procedure (Scheme 1) [12]. The yields obtained were very low to
high (14 to 70%, Table 1).
4

X
R
N
N
R
N
N
N
a
X
b
N
(2)
(
1a-e)
O
2
2
2
2
2
a, R = Et, X = I
1
a, R = Et, X = I
b, R = Me, X = I
c, R = C H X = I
1b, R = Me, X = I
10
21,
1c, R = C H X = I
10 21,
1d, R = CH Ph, X = Br
1e, R =
d, R = CH Ph, X = Br
2
2
e, R =
O
, X = I
O
O
O
CH2
, X = I
CH2
O
O
Scheme 1. Synthesis of hetero arylidene-9(10H)-anthrone (1) derivatives: a) R-X (1
to 1.2 equiv), THF, reflux. b) 9-anthracenecarboxaldehyde, Cs CO , THF, room
2
3
temperature, eight days.
Table 1. Scope of substrates of the synthetized hetero arylidene-9(10H)-anthrone (1).
I
I
Br
I
I
O
C10H21
N
N
N
N
N
N
N
N
N
N
O
O
O
O
O
O
O
1d
, 38%
1a, 70%
1b, 14%
1c, 61%
1e, 22%
O
S
O
F3C
CF3
F3C O
S
O
N
O
O S
O
O
O
N
N S CF3
O
N
O
C10H21
N
N
N
O
O
N
S
O-
O
O
O
O
3
a, 50% (over two steps)
step 1, 70%; step 2, 72%
3
b, 18% (over two steps)
step 1, 70%; step 2, 25%
3
c, 54% (over two steps)
step 1, 61%; step 2, 89%
To proceed for the anion exchange it was followed the procedure of Jordão [17]
involving precipitation of the dioctyl sulfosuccinate (docusate) or the bistriflimide
(
NTf ) imidazolium salts from an aqueous solution (Scheme 2). The procedure allowed
2
the preparation of hetero arylidene-9(10H)-anthrones (3a-c) in low to high yield (25%
to 89%, Table 1).
5

X
Y
R
R
N
N
N
N
a
(
1a-e)
O
O
( )
3a-c
3
3
3
a, R = Et, Y = N(Tf)2
b, R = Et, Y = docusate
1
a, R = Et, X = I
1c, R = C H X = I
10 21,
c, R = C H Y = N(Tf)2
10 21,
Scheme 2. Synthesis of hetero arylidene-9(10H)-anthrone (3) derivatives: a) 1a or 1c
(
1 equiv), Sodium docusate (1.1 equiv) or Li(NTf ) (1.1 equiv), H O.
2
2
These counter anions were selected to correlate the antiproliferative effect with the
structural modifications. Docusate is a nontoxic medication used to treat occasional
constipation and its effect may not necessarily be entirely due to its surfactant
properties, [18] and we expected its long and ramified alkyl chain to penetrate the
cellular membrane. Also, the same propriety will be assessed by the substitution pattern
at the imidazolium ring. The effect of imidazole ionic liquids (IL) alkyl chains on the
cytotoxicity against human lung carcinoma A549 cell line using different classes of
cations and anions was evaluated by Chen et al., [19] and showed that compounds with
longer non-polar alkyl chains presented higher cytotoxicity when combined with the
same anion. As for the anion bistriflimide, it has been widely used in ionic liquids (IL)
since it is a less toxic and more stable non-coordinating ion than "traditional"
counterions, such as tetrafluoroborate.
Cell viability. To evaluate the importance of the R- substitution and the counter ion in
the cytotoxicity of the hetero arylidene-9(10H)-anthrone derivatives, the
antiproliferative effect of compounds 1b-e and 3a-c was assessed by the MTS assay in
human ovarian carcinoma cells (A2780), colorectal carcinoma cells (HCT116), and in
normal primary human fibroblasts. The MTS assay is based on the formation of a
6

colored product – formazan, by the reduction of 3-(4,5-dimethylthiazol-2-yl)-5-(3-
carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt (MTS) catalyzed
by dehydrogenases enzymes [20]. Hence, the quantity of produced formazan is
proportional to the number and metabolic rate of cells in culture. From the MTS data, it
was possible to calculate the IC for each compound. Regardless of the counter ion,
5
0
compounds with an ethyl group (1a, 3a and 3b, Scheme 1 and Scheme 2) showed the
lowest antiproliferative activity, with an IC higher than 50 μM in HCT116 colorectal
5
0
carcinoma cells and fibroblasts and higher than 50 μM for 1a, 22.9 ± 1.7 μM for 3a and
3.6 ± 0.6 μM for 3b in A2780 (Table 2, Figure 1, Figure S1-S3) [12]. Compounds 1c
and 3c, both with a decyl group (C H ) at the R-substitution location and different
4
1
0
21
counter-ions (Scheme 1), showed the highest antiproliferative effect in A2780 tumor
cell line (IC of 1.1 ± 0.1 μM for 1c and 1.3 ± 0.1 μM for 3c; Table 2, Figure 1)
5
0
suggesting that the R-substitution on the hetero arylidene-9(10H)-anthrone has an
important role in the antiproliferative activity of the compounds.
Table 2. IC values after 48 h incubation of compounds 1a-e and 3a-c in human
5
0
cancer cell lines A2780 and HCT116 and normal human dermal fibroblasts.
Compound
R-Group
Counter
IC (μM)
50
ion
-
-
X or Y
A2780
> 50
HCT116
> 50
Fibroblasts
> 50
1
1
1
1
1
3
3
3
a[12]
Et
I
b
c
Me
C H
I
12.5 ± 0.5 > 50
> 50
I
1.1 ± 0.1
5.0 ± 0.7
8.1 ± 1.2
> 50
1
0
21
d
e
CH Ph
Br
15.0 ± 0.3 > 50
2
CH (OC H )
4 3
I
> 50
> 50
> 50
3
2
a
b
c
Et
Et
NTf2
22.9 ± 1.7 > 50
> 50
Docusate 43.6 ± 0.6 > 50
NTf2 1.3 ± 0.2 4.8 ± 1.2
0.1 ± 0.04 0.5 ± 0.1
> 50
C H
21
7.0 ± 2.4
n.d
1
0
Doxorubicin
n.d – not determined
7

Figure 1. Anti-proliferative activity of 1b, 1c and 3c in A2780 cell line after 48 h of
incubation with increasing concentrations of each compound. Data normalized against
the control (0.1 % (v/v) DMSO) and expressed as the mean ± SEM of three independent
assays. * p-value < 0.05 relative to control.
-
In fact, compound 1b, with a methyl substitution at the R-position and iodide (I ) as
counter ion, showed an IC in A2780 cells of 12.5 ± 0.5 μM (Table 2, Figure 1), and
5
0
IC >50 μM in both HCT116 and fibroblasts cell lines (Table 2, Figure S2). What is
5
0
more, the length and structure of the aryl chain at the R-position seemed to have a
relevant impact in the compound’s antiproliferative activity. Indeed, 1c and 3c showed
the lowest IC of the series for all cell lines (Table 2, Figure 1, Figures S1-S3), and
5
0
were the only compounds with an IC < 50 μM in HCT116 and fibroblasts (Table 2,
5
0
Figure 1 and Figures S1-S3 ). Interestingly, the antiproliferative effect of compound 1d,
with Ch Ph at the R-position, was similar to that of compound 1b (Table 2, Figure 1 and
2
Figure S1), with IC in A2780 below that of compounds 1a, 3a and 3b, with an ethyl
5
0
group at the R-substitution. The polar coordinating group CH (OC H ) at R-
3
2
4 3
substitution of compound 1e, rendered no loss of cell viability for this compound
towards the analyzed cell lines (Table 2, Figure S1, Figure S2). Despite the compounds
show a higher IC (at 48h) in A2780 and HCT116 cells when compared with the
5
0
common chemotherapeutic agent doxorubicin (0.1 ± 0.04 M and 0.5 ± 0.1 M,
8

respectively), compound 1c shows the most similar value (10x higher) (Table 2, Figure
S4).
As described above, the relevance of the R-substitution is highlighted when comparing
compounds with the same counter-ion, e.g. 3a and 3c. Despite similar observations are
-
detected for compounds bearing I as counter-ion, 1b stands-out as outlier as the methyl
substitution in 1b showed higher antiproliferative activity than the ethyl substitution in
1
a (Table 2). Since compounds 1b and 1c showed a considerable antiproliferative effect
towards the A2780 ovarian cancer cell line (compound 1c showing a selectivity index of
.36) while 1b presented no cytotoxicity against normal cells, the two compounds were
7
selected for further biological analysis.
Mechanisms of cell death. To provide further insights towards the cell death
mechanism induced by compounds 1b and 1c in A2780 cells, six different biological
analysis were performed: i) apoptosis analysis by DNA staining with Hoechst 33258; ii)
quantification of apoptosis based on double staining with annexin V conjugated with
AlexaFluor488 and with propidium iodide (PI); iii) Trypan blue exclusion method
(trypan blue stain is only internalized by cells with a compromised membrane - blue
cells - non-viable [21]); iv) quantification of the apoptotic index through BAX/BCL-2
expression ratio for analysis of intrinsic apoptosis [22]; v) quantification of caspase 8
activity for the analysis of extrinsic apoptosis [23]; and vi) analysis of autophagy
induced by compounds.
The staining of DNA with Hoechst 33258 allows the detection of nuclear
modifications characteristic of apoptosis, such as chromatin condensation and nuclear
fragmentation, that result in cell death [24]. After 48 h, nuclei of A2780 cells exposed to
9

IC of compounds 1b and 1c revealed an increase in the number of nuclear
5
0
modifications consistent with augmented apoptosis relative to control (Figure 2).
Figure 2. Apoptotic morphological changes in A2780 cells treated with compounds 1b
and 1c. A) Representative images of A2780 cells incubated for 48 h with A1) 0.1 %
(
v/v) DMSO (control), A2) the IC value at 48 h of compound 1b and A3) the IC
50 50
value at 48 h of compound 1c, and stained with Hoechst 33258 for observation of
apoptotic events, pointed by white arrows, including chromatin condensation and
nuclear fragmentation. Scale bar with 20 μm. B) Percentage of apoptotic cells after
incubation of A2780 cells for 48 h with compounds 1b and 1c and respective vehicle
control. Values are represented as average ± SD of at least three independent assays. *
p<0.05 relative to control.
For a more reliable quantification of cell death and to confirm the results in Figure 2,
A2780 cells were incubated with IC of compounds 1b and 1c and then double stained
5
0
with Annexin V - AlexaFluor488 and PI (Figure 3). With this methodology, it is
possible to distinguish between live cells (unstained), cells in early apoptosis (stained
with annexin V - AlexaFluor488), cells in late apoptosis (stained with both
fluorophores) and necrotic cells (only stained with PI). As observed in Figure 3, only a
small increase in the number of apoptotic cells compared to control is observed when
1
0

A2780 cells were incubated for 48 h with the IC of compound 1c and no difference
5
0
compared to control when cells were incubated with the IC of compound 1b. As
5
0
expected DOXO was able to induce necrosis and apoptosis of A2780 cells (Figure 3).
Considering the results for the positive control (DOXO) (Figure 3) and data from Figure
2
, we were expecting to observe a higher level of cell death via apoptosis with Annexin
V - AlexaFluor488 and PI double staining (Figure 3).
Figure 3: Quantification of apoptotic and necrotic A2780 cells exposed to compounds
1
b and 1c. A2780 cells were incubated for 48 h with 0.1 % (v/v) DMSO (control), the
IC concentration of doxorubicin (DOX), positive control of necrosis and apoptosis, or
5
0
the IC concentration of compounds 1b (12.5 M) and 1c (1.1 M). Bars represent the
5
0
average ± SEM of at least three independent experiments. *p<0.05 relative to control.
To clarify these results, we decided to confirm A2780 cell death induced by both
compounds using another method - the Trypan blue exclusion method. For that, A2780
cells were exposed to the IC of each compound and to a concentration 10x higher and
5
0
1
0x lower the IC (Figure 4). Analysis of cell death was performed in cells that
50
remained adherent (Figure 4) and in their respective supernatants (Figure 5).
1
1

Figure 4: Viability of A2780 cells exposed to compounds 1b (A) and 1c (B). A2780
cells were incubated for 48 h with 0.1 % (v/v) DMSO (control), to the IC50
concentration of compounds 1b (12.5 M) and 1c (1.1 M), or to concentrations lower
and higher than IC . Viability was normalized to the control and represented as the
5
0
average ± SD of three independent assays. * p<0.05, *** p<0.001 relative to control
sample (0.1 % (v/v) DMSO).
Live cells
Dead cells
3
3
2
2
5
0
5
0
**
1
0
**
9
8
7
6
5
4
3
2
1
0
*
*
DOXO
1b
1c
Figure 5. Concentration of A2780 cells in the supernatant after exposure to compounds
b and 1c. A2780 cells were incubated for 6 h with 0.1 % (v/v) DMSO (Control), 4 μM
10 x IC ) of Doxorubicin (DOXO), 10 x IC of compound 1b, and 10 x IC of
1
(
5
0
50
50
compound 1c. Cell concentration (cells/mL) of each sample was normalized to the cell
concentration of the control. The value of the control is 1.0 ± 0.1 for live cells and 1.0 ±
0
.6 for dead cells, represented as a black dotted line. Bars represent the average ± SEM
of three independent experiments. . * p-value < 0.05 ** p-value < 0.001 comparing each
concentration with respective control.
1
2

Considering the adherent cells, data showed a direct correlation to the results obtained
with MTS analysis (Figure 1 and Figure 4), with one exception - when A2780 cells
were exposed to 12.5 μM of compound 1b (Figure 4A), that accordingly to the MTS
analysis should result in a decrease of 50% of viability and not 80% as observed in
Figure 4A. However, since each method relies on different cellular properties
(compromised membrane for Trypan blue exclusion assay and mitochondrial viability
for the MTS), some discrepancies might occur. Still, both assays confirm the decrease
of cell viability induced by the compounds.
Trypan blue analysis was also performed in cells in suspension - not adhered after
exposure to the compounds. As observed in Figure 5, the concentration of live cells in
the supernatant of A2780 culture treated with 1c (black bars) increased 2.5 ± 0.2 times
compared to the control. Interestingly, the concentration of cells stained with Trypan
blue and considered as dead cells in the supernatants is very high (7 ± 2.4 higher than
the control for cells treated with 1b, and 25.5 ±3.9 higher than control for cells treated
with 1c (Figure 5, gray bars)). The higher cell concentration in the supernatant of A2780
culture treated with 1b and 1c compared to the value obtained for cells treated with
DOXO, might indicate that both compounds induce the loss of an adherent phenotype
and cell death via detachment (Figure 5). These results indicate that apoptosis analysis
via double staining with Annexin V - AlexaFluor488 and PI (Figure 3) might be under -
estimated due to cell detachment.
The apoptotic pathway is a highly regulated process that will ultimately lead to cell
death [25-28]. Depending on the initial triggering signal, the apoptotic process can occur
throughout two different pathways, the intrinsic or mitochondrial pathway is triggered
by signals such as DNA damage or growth factors deprivation, and the extrinsic or
death receptor pathway is mainly triggered by death-inducing signals [25-28]. Central to
1
3

the intrinsic apoptosis pathway are the anti-apoptotic BCL-2 family proteins [25-28]. In
healthy cells, BCL-2 inhibit apoptosis through interaction with pro-apoptotic BAX
protein which is located at the cytoplasm as monomer [25-28]. After an apoptotic
signal, BAX insert in the mitochondrial outer membrane as homo-oligomer resulting in
the permeabilization of the mitochondrial outer membrane and consequent dissipation
of mitochondrial membrane potential (m) [25-28]. The apoptotic index, that is the
ratio of pro-apoptotic (BAX) and anti-apoptotic (BCL-2) proteins, is indicative of the
cell susceptibility to the intrinsic apoptotic pathway [25-28].
Considering the results so far, the m of A2780 cells exposed to the IC of both
5
0
compounds was analyzed (Figure 6). For this, a lipophilic cationic dye JC-1 with green
fluorescence when in monomeric form (JC-1 monomer) and red fluorescence when in
aggregated form (JC-1 aggregate) was used. The quantification of JC-1
monomer/aggregate fluorescence ratio (JC-1 ratio) give an indication of the m of the
cell. In healthy cells, when the m is high, JC-1 accumulates in mitochondria and
green / red fluorescence ratio is lower than 1 [29]. During both extrinsic and intrinsic
apoptosis pathways, the mitochondrial membrane is compromised and a m
dissipation occurs [25-28]. Figure 6 shows that a red to green shift is observed when
cells are treated with both compounds, suggesting a depolarization of the m. The

m results suggest that the 80 % cell viability observed by the Trypan blue exclusion
method (Figure 4) can be related to the presence of cells with depolarized mitochondrial
membrane (Figure 6), but integral plasmatic membrane, which impacts the MTS assay.
Indeed, after 48 h, the percentage of live A2780 cells incubated with 1b was similar to
the percentage of live cells in the control (Figure 3). Hence, at IC of compound 1b, the
5
0
cellular viability using the MTS assay should be lower than that determined via the
Trypan Blue exclusion method.
1
4

Figure 6. Mitochondrial membrane potential of A2780 cells treated with compounds
b and 1c. A) A2780 cells were incubated for 48 h with A1) 0.1 % (v/v) DMSO, A2)
1
IC at 48 h compound 1b and A3) IC at 48 h compound 1c and stained with JC-1. B)
5
0
50
JC-1 monomer/aggregate ratio obtained after 48 h incubation of A2780 cells and
quantification of JC-1 monomer and JC-1 aggregate fluorescence. Values were
normalized to the JC-1 ratio of control cells (JC-1 ratio of control is 1). Bars represent
the average ± SEM of three independent experiments. *p<0.05 relative to control
sample.
The analysis of the apoptotic index through quantification of BAX and BCL-2
expression by western blot of total protein extract of A2780 exposed for 48 h to
compounds 1b and 1c, followed by calculation of BAX/BCL-2 ratio, confirmed the
results obtained in double staining apoptosis analysis (Figure 3), with compound 1c
1
5

presenting BAX/BCL-2 ratios slightly higher, but not statistically significant, than
control sample (Figure 7, Figure S5A and S5B with the entire western blot membranes).
Interestingly, when we measure the apoptotic index of A2780 incubated for 18 h with
compound 1c, it was similar to the value obtained when cells were incubated with
apoptosis inducer doxorubicin (Figure 8, Figure S6A and S6B with the entire western
blot membranes), suggesting that the antiproliferative action of 1c might be trigger
earlier on time via the induction of intrinsic apoptosis.
Figure 7. Expression of BAX and BCL-2 proteins in A2780 cells treated 48 h with
compounds 1b and 1c. A) Western blot for quantification of A1) BAX and A2) BCL-2.
Represented western blots correspond to 10 μg total protein of A2780 incubated for 48
h with 0.1 % (v/v) DMSO (Control), 0.12 μM doxorubicin (DOXO), the IC at 48 h of
5
0
compound 1b (12.5 μM), or the IC at 48 h of compound 1c (1.1 μM). C) Protein
5
0
expression of B1) BAX and B2) BCL-2 after normalization against β-actin and
respective control (DMSO). The value of the control is 100%, represented as a grey
dotted line. C) Apoptotic index of A2780 cells, calculated by BAX/BCL-2 ratio. The
value of the control is 1, represented as a grey dotted line. Bars represent the average ±
SEM of three independent experiments. * p-value < 0.05 relative to control sample.
1
6

Figure 8. Expression of BAX and BCL-2 proteins in A2780 cells treated 18 h with
compounds 1b and 1c. A) Western blot for quantification of A1) BAX and A2) BCL-2.
Represented western blots correspond to 10 μg total protein of A2780 incubated for 18
h with 0.1 % (v/v) DMSO (Control), 0.12 μM doxorubicin (DOXO), the IC at 48 h of
5
0
compound 1b (12.5 μM), or the IC at 48 h of compound 1c (1.1 μM). C) Protein
5
0
expression of B1) BAX and B2) BCL-2 after normalization against β-actin and
respective control (DMSO). The value of the control is 100%, represented as a grey
dotted line. C) Apoptotic index of A2780 cells, calculated by BAX/BCL-2 ratio. The
value of the control is 1, represented as a grey dotted line. Bars represent the average ±
SEM of three independent experiments. * p-value < 0.05 relative to control sample.
Caspase 8 is an initiator caspase central to the extrinsic apoptotic pathway, being
activated by apoptotic stimuli from the plasmatic membrane [23]. After activation,
caspase 8 cleave several molecules, such as downstream caspases, nuclear proteins, and
plasma membrane and mitochondrial proteins [23]. To quantify the caspase 8 activity,
the chromogenic substrate IEDT-pNA consisting in the IETD (Ile-Glu-Thr-Asp) peptide
conjugated with the chromophore p-nitroanilide (pNA) is added to the total protein
extract. As the substrate is cleaved by caspase 8, the pNA released is followed at 400
1
7

nm [30]. The caspase 8 activity of A2780 cells incubated for 18 h and 48 h with
compounds 1b and 1c was measured. Similarly, at 48 h no difference compared to the
control was observed (Figure S7), but after 18 h of exposition to compound 1b the
absorbance of pNA was 1.9 ± 0.4 times higher than control sample, suggesting the
induction of apoptosis via the extrinsic pathway (Figure 9).
2.5
2.0
1.5
1.0
0.5
0.0
*
*
l
xo
1b
1c
ntro
Do
Co
Figure 9. Caspase 8 activity in A2780 cells treated 18 h with compounds 1b and 1c.
A2780 cells were incubated for 18 h with 0.1 % (v/v) DMSO (Control), 0.12μM
Doxorubicin (DOXO), IC at 48h of compound 1b (12.5 μM) or the IC at 48 h of
5
0
50
compound 1c (1.1 μM). Caspase 8 activity was quantified using the caspase 8 assay kit
Abcam). Absorbance at 400 nm values of each sample was normalized to the control.
(
Bars represent the average ± SEM of three independent experiments. * p-value < 0.05
relative to control.
To infer if 1b and 1c induce autophagy in A2780 cells, the culture was exposed to the
IC concentration of each compound and stained with an autophagic fluorescent dye
5
0
(Abcam). According to the manufacturer, this dye is a cationic amphilic tracer that
rapidly partition into cells and becomes highly fluorescent when incorporated in pre-
autophagosomes, autophagosomes and autophagolysosomes. The fluorescence
1
8

quantification by flow cytometry revealed no differences between control sample (cells
treated with 0.1 % (v/v) DMSO) and cells treated with each compound (Figure S8).
To further explore the mechanism triggered by compounds 1b and 1c that result in
apoptosis induction and to correlate with data obtained so far, four other biological
studies were performed, i) expression of E-cadherin, ii) incubation with 2’,7’-bis(2-
carboxyethyl)-5,6-carboxyfluorescein (BCECF-AM) to infer intracellular pH (pHi) and
the state of plasma membrane [31-34]; and iii) mechanism of internalization of
compounds and subcellular localization; iv) induction of reactive oxygen species
(ROS).
The increased number of live cells in suspension when A2780 culture was exposed to
compound 1c (Figure 5, black bars) lead us to hypothesize the possibility of a decrease
of E-cadherin expression and consequent cell detachment from the substrate. E-cadherin
is the protein that connect epithelial cells at adherents’ junctions, being vital for tissue
homeostasis [35]. The western blot analysis of E-cadherin expression in A2780 treated
for 48 h with IC of compounds 1b and 1c revealed a decrease of E-cadherin protein
5
0
expression (Figure 10 and Figure S9, with the entire western blot membranes) that
correlates with the loss of cell adhesion (Figure 5) and with the trigger of an apoptotic
process (Figures 2, 6 and 8) [36].
On the other side, the decreased expression of E-cadherin is a hallmark of epithelial-to-
mesenchymal transition in cancer cells and is accompanied by extrinsic apoptosis
attenuation (Figure S7) [37]. When looking to A2780 cells exposed to IC of the
5
0
common antitumor drug doxorubicin a very high decrease of E-cadherin expression is
observed [38] correlating with the observed intrinsic apoptosis (Figure 8).
1
9

Figure 10. Expression of E-cadherin protein in A2780 cells treated 18 h with
compounds 1b and 1c. A) Western blot for quantification of expression for E-cadherin.
Represented Western-blots correspond to 50 g total protein for CDH1 of A2780 cells
incubated 18 h with 0.1 % (v/v) DMSO (vector control of compounds), 0.4 M
doxorubicin (DOXO) or the IC at 48 h of compound 1b (12.5 M) or compound 1c
5
0
(1.1 M). B) Percentage of protein expression of E-cadherin after normalization against
β-actin and respective control. The value of control is 100 % represented as a dotted line
for comparison. Bars represent average ± SEM of three independent experiments. * p-
value < 0.05. relative to control.
During apoptosis, intracellular pH (pHi) might increase [33, 34]. In order to analyse
changes in pHi after exposure to both compounds, cells were incubated with BCECF-
AM, a non-charged molecule that enter cells, and once inside, the intracellular
estereases cleave the ester bond releasing BCECF whose fluorescence increase as the
pHi increases [39]. It should be noted that due to the passive diffusion of BCECF-AM
into the cells, the loss of fluorescence intensity can also occur in cells with
compromised membranes [31, 32]. The fluorescence of A2780 cells treated with
compounds 1b and 1c was measured after 3 h, 6 h, 18 h and 48 h incubation with
BCECF-AM. An increased fluorescence was detected when A2780 cells were incubated
for more than 18 h with doxorubicin and both compounds 1b and 1c (Figure S10),
2
0

suggesting an alcalinization of the pHi, possibly due to apoptosis [33, 34]. A transient
fluorescence increase is also observed when A2780 cells were incubated for 3 h with
compound 1b, suggesting a pHi alteration that might be promoted by the compound
entry into cells (Figure S10), but not observed with compound 1c. This might indicate
that both compounds enter cells via distinct mechanisms. In order to confirm this
observation and taking the fact that compounds 1b and 1c exhibited a blue fluorescence
upon excitation at 340-380 nm in an inverted fluorescence microscope, it was possible
to gather information concerning their internalization and accumulation in cells over
time.
The presence of compounds 1b and 1c in cells after 1 hour of incubation was perceived
by an increased fluorescence in treated relative to untreated cells (Figure S11).
Remarkably, the fluorescence in cells treated with compound 1b kept increasing with
time (3 h > 1 h) (Figure S11A, S11D), whereas for compound 1c the fluorescence levels
at 1 h and 3 h were equivalent (Figure S11B, S11D). This higher concentration of
compound (10x IC ) was needed to ensure the visualization of the compound inside the
5
0
cell by fluorescence microscopy. After 3 h incubation in the presence of 10x the IC of
5
0
compound 1b exhibited a fluorescence signal along the cytoskeleton stain (Figure
S12A) with higher fluorescence intensity co-localizing with vesicles near the plasma
membrane (Figure S12B). Interestingly, morphological alterations can be observed for
A2780 cells incubated with 10x the IC of compound 1c for 3 h (Figure S13 and Figure
5
0
S14). Indeed, cells treated with compound 1c seem to be rounder with a smaller
cytoplasm (Figure S13 and Figure S14). Counterstaining of both the cytoskeleton and
nucleus allowed to visualize the distribution of the compound all over the cell with
blebbing at spots (Figure S13B, white arrow). What is more, one can clearly visualize
cells with blue fluorescence but no cytoplasmic staining, suggesting compound
2
1

internalization by the cell and cytoskeleton damage (Figure S13B, red arrow). A closer
analysis to the cell morphology when A2780 cells were treated with compound 1c
(Figure S13) reveals several events in line with oncosis. In this programmed cell death,
a membrane injury causes the leak of ions and water, resulting in cell swelling without
increased cell membrane permeability, altering cell shape and volume, which can be
observed by the formation of cytoplasmic blebs [40, 41]. As the oncotic process
continues, the membrane permeability increases continuously, other alterations occur,
such as chromatin clumping, dilatation of ER and Golgi, mitochondrial condensation
and significant alterations of the cytoskeleton [41]. At later stages, cells will continue to
round up and detach from the substrate [40, 41]. All the results relating to the cell
effects of compound 1c agree with early oncotic events: the round up and blebbing of
cells in monolayer is easily observed when the ovarian cancer cells were incubated with
compound 1c and stained with Phalloidin (Figure S12B), where one unstained cell may
be observed (Figure S12B, red arrow) suggesting cytoskeleton alterations in agreement
with the literature [42].
In resume, compound 1b shows an increased internalization over time (Figure S11A),
a small but significant decrease of internalization at 4 ºC (Figure S15A), and co-
localization with vesicles located near the cell surface (Figure S12). Altogether, these
data suggest that compound uptake occurs by passive diffusion and possibly by low
level endocytosis. These observations are in line with results obtained for intracellular
pH of A2780 cells incubated with 1b, where a transient pH increase was observed after
3
h incubation (Figure S10).
Next, we evaluated whether the uptake of these compounds was due to passive or
active transport, by assessing internalization at 37 ºC and 4 ºC - at 4 ºC the energy
balance of the cell is severely diminished and active transport decreases. Incubation at 4
2
2

ºC revealed a slight decrease of compound 1b internalization when compared to 37 ºC
Figure S15A, Figure S15C1) but an accentuated decreased of compound 1c
(
internalization (Figure S15B, Figure S15C2). These results suggest that internalization
of compound 1b relies on passive diffusion, while compound 1c might also permeate
the cell membrane via an active uptake mechanism [43, 44]. This difference might be
attributed to the different size of the R-substitution.
Due to the cell damages induced by ROS, such as DNA strand breaks or membrane
blebbing, the induction of oxidative stress is intrinsically related to apoptosis [45]. The
induction of ROS by both compounds was evaluated using 2,7-
dichlorodihydrofluorescein diacetate (H DCF-DA) which, after de-acetylation and
2
oxidation by peroxides present in cells, forms the highly fluorescent
2
’7’dichlorofluorescein (DCF). A2780 cells exposed to IC of compounds 1b and 1c
50
present levels of ROS similar to the control (Figure S16). These data suggest that both
compounds are not able to induce the production of ROS and, thus, oxidative stress.
The accumulation of the compound 1c in the cytoplasm and not in the nuclei (Figure
S12), coupled with the absence of oxidative stress (Figure S16), suggest a different
mechanism of cytotoxicity for compound 1b to those of anthracyclines commonly used
in chemotherapy, such as doxorubicin [46]. The IC of compound 1b in colorectal
5
0
cancer cells and normal fibroblasts is greater than 50 μM (Table 2). However, the high
fluorescence in HCT116 cells incubated with the 1b (Figure S11A) suggests
internalization of the compound. The cytotoxicity of compound 1b in A2780 cells, with
low impact to healthy fibroblasts, might be associated to intrinsic characteristics of the
ovarian cancer cells, making this compound suitable for ovarian cancer therapy.
2
3

3
. Conclusion
The antiproliferative activity of hetero-arylidene-9(10H)-anthrone derivatives seemed
to be mainly reliant on the R-substitution of the cation counterpart and in a low scale in
the counter anion. Moreover, the nature of the R-substitution showed to be relevant for
cytotoxicity with compounds with a long alkyl chain as C H showing the highest
1
0
21
cytotoxicity in all cell lines, including normal primary dermal fibroblasts. Based on the
index selectivity towards ovarian cancer cells, two compounds containing the same
-
counter anion (I ) and distinct R-substitution were selected for further cytotoxicity
analysis. Interestingly, the cellular uptake and mechanisms of cytotoxicity of
compounds with methyl- or C H substitutions at the R-position, compounds 1b and
1
0
21
1
c, respectively, were different and distinct between each other and from those
described for anthracyclines used in chemotherapy, such as doxorubicin. Our data
provide interesting clues on the cytotoxic mechanisms affecting A2780 cells induced by
compounds 1b and 1c. As observed in the analysis of anti-proliferative activity of the
compounds, the R- substitution is preponderant for the compound’s internalization and
mechanism of cell death induced by the hetero-arylidene-9(10H)-anthrone derivatives.
The evaluation of the presence of apoptotic events in the nuclei, and decreased
monomer/aggregate JC-1 ratio after 48 h, and increased caspase-8 activity after 18 h
suggest that compound 1b induces extrinsic apoptosis.
Despite the only difference resides in the R-substitution from a methyl group in 1b to
an aliphatic hydrophobic chain, C H , in 1c, the mechanism of cell uptake and
1
0
21
cytotoxicity of compound 1c showed to be different to those of compound 1b. In fact,
following 1 h and 3 h incubation with compound 1c, A2780 cells show similar
fluorescence intensity but a dramatic fluorescence decrease was observed when A2780
was incubated at 4 ºC, which strongly indicates passive transport of the compound. As
2
4

for compound 1b, the accumulation of compound 1c in the cytoplasm and lack of
oxidative stress suggest a toxicity mechanism different to that of doxorubicin.[46] The
presence of nuclei modifications characteristic of apoptosis, such as chromatin
condensation, alteration of the mitochondrial membrane potential after 48 h, and
increased apoptotic index after 18 h is suggestive of induced intrinsic apoptosis.
Nevertheless, the increased concentration of live and dead cells at the supernatant when
A2780 cell culture was exposed to 1c, the roundup and blebbing of cells and phalloidin
unstained cells does not exclude the possibility of oncosis. In fact, it is reported that
apoptosis and oncosis are closely related mechanisms, apoptotic cells may undergo
oncosis due to energetic requirements [23].
Despite compound 1b and 1c show a higher IC in A2780 cell line when compared to
5
0
common chemotherapeutic drug - doxorubicin, the high selectivity index, and apoptosis
induction of both compounds, compared with doxorubicin that also induces necrosis to
the cells,[38] might be of interest for further studies due to the distinct mechanism of
action. Indeed, the induction of ROS is one of the most deleterious mechanism of action
attributed to doxorubicin, resulting in cardiotoxic side effects and limiting the dose and
length of treatment in cancer patients.
4
. Experimental Section
MATERIALS AND METHODS. All the reagents and solvents were obtained
commercially, and these were used without further purification. The solvents used
were dried using current laboratory techniques. Thin layer chromatography (TLC)
was carried out on aluminium backed Kieselgel 60 F254 silica gel plates (Merck).
Plates were visualized by UV light (254 and/or 366nm). Preparative layer
chromatography (PLC) was performed on Merck Kieselgel GF 254 silica gel plates
2
5

with a thickness of 0.5 mm or 1 mm. Column chromatography were carried out on
silica gel Kieselgel 60 (Merck), 70 - 230 mesh particle size as stationary phase, in
normal phase chromatography. Ultraviolet spectroscopy (UV) was recorded on a
Thermo Corporation spectrophotometer, Helius γ, on quartz cell support. Absorption
1
13
spectrum measurements were made in the range of 190 to 320 nm. H and C NMR
spectra were recorded on a Brucker ARX400 at 400 and 100 MHz, respectively.
Chemical shifts are reported in parts per million (ppm, δ units). The following NMR
abbreviations are used: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet,
brs = broad singlet. HPLC-MS-ESI was acquired on a Agilent 1200 Series LC with
quaternary pump, ALS, TCC, DAD, and Agilent G6130B LC/MSD with API-ES
source. High Resolution Mass spectra (MS) using the EI-TOF technique were
obtained from Unidade de Masas e Proteómica, the University of Santiago de
Compostela, Spain.
General procedure for the preparation of imidazolium salts (2a-e): In a round
bottom flask equipped with magnetic stir bar was added 1,2-dimethylimidazol in dry
THF to make a 0.55 M solution and the corresponding alkyl halide (1 to 1.2 equiv).
The reaction mixture was stirred at reflux under N atmosphere until the total
2
consumption of the 1,2-dimethylimidazol was verified. This monitoring was carried
out by TLC, in an eluent of MeOH/dichloromethane (5:95). The product as formed
began to precipitate from the reaction medium and after terminus of the reaction the
product was filtered and used as so.
Procedure for the preparation of 3-{2-[2-(2-methoxyethoxy)ethoxy]ethyl}-1,2-
dimethyl-1H-imidazol-3-ium iodide (2e): To a solution of triethylene glycol
-3
monomethyl ether (3.0x10 mol) in THF (20 mL) were added triphenylphosphine
-3
-3
(
0.97 g, 3.7x10 mol, 1.2 equiv) and imidazole (0.61 g, 9.0x10 mol, 3.0 equiv) The
2
6

mixture was stirred under nitrogen atmosphere at room temperature for 3 minutes.
-3
Then the reaction mixture was cooled to -20 °C and I (1.01 g, 3.98x10 mol, 1.2
2
equiv) was added. After stirring at this temperature for 15 minutes the cooling bath
was removed, and the mixture stirred at room temperature for 2 h until the total
consumption of triethylene glycol monomethyl ether was verified. This monitoring
was carried out by TLC, in an eluent of hexane and ethyl acetate (8:2) (spray reagent:
phosphomolybdic acid solution). The reaction was quenched at 0 °C with an aqueous
saturated solution of NaHCO (20 mL). The reaction mixture was filtered through out
3
celite several times until all white precipitate formed was removed. The aqueous
phase was extracted with diethyl ether and the combined organic phases were washed
with a saturated solution of Na S O (20 mL) and brine, dried over Na SO
2
2
5
2
4
anhydrous, filtered and evaporated. The residue was purified by chromatography on
silica gel using as eluent a mixture of hexane and ethyl acetate (8:2), affording 3-{2-
[2-(2-methoxyethoxy)ethoxy]ethyl} iodide (0.58g, 77%) as a yellow oil. In a round
bottom flask equipped with magnetic stir bar was added 3-{2-[2-(2-
-3
methoxyethoxy)ethoxy]ethyl} iodide (0.57g, 2.1x10 mol, 1.4 equiv) and the 1,2-
-3
dimethylimidazol (0.14g, 1.5x10 mol) in dry THF. The reaction mixture was stirred
and refluxed for 122 h under nitrogen atmosphere until the total consumption of the
1
,2-dimethylimidazol was verified. This monitoring was carried out by TLC, in an
eluent of dichloromethane/MeOH (95:5) (spray reagent: Dragendorff-Munier
solution). The reaction mixture was evaporated and washed with diethyl ether,
affording 3-{2-[2-(2-methoxyethoxy)ethoxy]ethyl}-1,2-dimethyl-1H-imidazol-3-ium
1
iodide (2e) (0,40g, 75%) as a yellow oil; H NMR (CDCl , 400 MHz) δ: 7.72 (d, J =
3
2
1
.0 Hz, 1H, H-4), 7.49 (d, J = 2.4 Hz, 1H, H-5), 4.48 – 4.43 (t, 2H, J = 4.8 Hz, H-
’’a), 3.94 (s, 3H, H-6), 3.89 – 3.86 (t, 2H, 4.8 Hz, H-2’’a), 3.61 – 3.59 (m, 2H, CH2),
2
7

3
3
.56 – 3.55 (m, 4H, 2xCH ), 3.53 – 3.48 (m, 2H, 2xCH ), 3.35 (s, 3H, CH ), 2.80 (s,
2 2 3
1
3
H, CH ) ppm. C NMR (CDCl , 100 MHz) δ: 144.86 (C-2), 122.56 (CH), 121.94
3
3
(
CH), 71.81 (CH ), 70.38 (CH ), 70.26 (2xCH ), 69.30 (CH ), 58.98 (OCH ), 49.18
2 2 2 2 3
-1
(
CH ), 36.44 (CH -6), 11.62 (CH -7) ppm. IR ν (cm ): 2897 (νs C-H aliphatic),
2 3 3 max
+
1
095 (ν C-O-C), 1586 (w, C=N). HRMS (ESI-TOF) m/z: [M + H] Calc for
as
C H N O 243.1706, found 243.1703.
1
2
23
2
3
General procedure for the preparation hetero arylidene-9(10H)-anthrone
1a-e): In a round bottom flask equipped with magnetic stir bar was added the
imidazolium salt 2 (0.6 mmol) and Cs CO (1.2 equiv) in dry THF (5 mL). The
(
2
3
reaction mixture was stirred at room temperature for 1 h and then 9-
anthracenecarboxaldehyde (1 equiv) was added. The reaction mixture was stirred at
room temperature until the total consumption of the aldehyde was verified (7 to 8
days). This monitoring was carried out by TLC, in an eluent of dichloromethane/
MeOH (85:15). The reaction mixture was evaporated under reduced pressure, the
crude product was washed with diethyl ether to remove less polar compounds and
then washed with water to remove any remaining base or imidazolium salt. The title
compound was obtained by column chromatography using dichloromethane/MeOH
(85:15) as eluent.
1
,3-Dimethyl-2-{[10-oxoanthracen-9(10H)-ylidene]methyl}-1H-imidazol-3-
1
ium iodide (1b). Yellow / Brown solid (14%); mp: >280 ᵒC. H NMR (CDCl ,
3
4
7
00MHz) δ: 8.31 (t, J = 7.6 Hz, 2H, H-4’ and H-5’), 8.25 (d, J = 8.0 Hz, 1H, H-1’),
.76 (t, J = 7.6 Hz, 1H, H-7’), 7.73 (br s, 2H, H-4 and H-5), 7.67 – 7.58 (m, 2H, H-6’
and H-3’), 7.58 (s, 1H, H-10’’), 7.49 (t, J = 7.6 Hz, 1H, H-2’), 6.68 (d, J = 7.6 Hz,
1
3
1
1
H, H-8’), 3.68 (s, 6H, 2x CH ) ppm. C NMR (CDCl 100MHz) δ: 183.20 (C-10’),
3 3,
45.97 (C-1’a), 142.92 (C-2), 136.35 (C-8’a), 134.30 (CH-7’), 134.15 (C-4’a), 133.70
2
8

(CH-2’), 131.83 (C-9’), 131.35 (CH-6’), 130.99 (CH-3’), 130.56 (C-5’a), 128.64
(CH-5’), 127.62 (CH-1’), 125.33 (CH-4’), 125,22 (CH-8’), 124.35 (CH-4 and CH-5),
+
1
09.32 (CH-10’’), 36.94 (2x CH ) ppm. ESI-MS (m/z) (+): 301.1 [M] ; ESI-MS (m/z)
3
-
-1
(
-): 126.9 [I] . IR ν (cm ): 2925 (m, ν C-H), 2849 (w, ν C-H), 1666 (m, C=O),
max as s
+
1
590 (w, C=N), 1313 (w, C-N). HRMS (ESI-TOF) m/z: [M + H] Calc for
C H N O 301.1338, found 301.1335.
2
0
17
2
3
-Decyl-1-methyl-2-{[10-oxoanthracen-9(10H)-ylidene]methyl}-1H-imidazol-
1
3
-ium iodide (1c). Red solid (61%); mp: 169-176 ᵒC. H NMR (CDCl , 400MHz) δ:
3
8
.36 (d, J = 8.0 Hz, 1H, H-4’), 8.31 (d, J = 8.0 Hz, 1H, H-1’), 8.26 (d, J = 8.0 Hz, 1H,
H-5’), 7.92 (d, J = 2.0 Hz, 1H, H-4), 7.76 (t, J = 7.2 Hz, 1H, H-7’), 7.68 (s, 1H,
H10’’), 7.67 (d, J = 2.0 Hz, 1H, H-5), 7.66 – 7.57 (m, 2H, H-3’ and H-6’), 7.47 (t, J =
8
.0 Hz, 1H, H-2’), 6.63 (d, J = 8.0 Hz, 1H, H-8’), 3.96 (dd, J = 6.4 and 15.2 Hz, 2H,
H-1’’’), 3.67 (s, 3H, CH ), 1.84 – 1.69 (m, 1H, H-2’’’), 1.59-1.49 (m, 1H, H-2’’’),
3
1
3
1
.19 – 1.30 (m, 14H, 7 X CH ), 0.81 (t, J = 6.8 Hz, 3H, H-10’’’) ppm. C NMR
2
(
CDCl 100MHz) δ: 183.13 (C-10’), 145.30 (C-1’a), 142.26 (C-2), 136.38 (C-8’a),
3,
1
34.26 (CH-7’), 134.09 (C-4’a), 133.51 (C-2’), 131.87 (C-9), 131.22 (CH-6’), 130.89
(
(
(
(
[
CH-3’), 130.49 (C-5’a), 128.64 (CH-5’), 127.56 (CH-1’), 125.39 (CH-4’), 125.11
CH-8’), 124.72 (CH-4), 122.52 (CH-5), 109.79 (CH-10’’), 49.75 (CH -1’’’), 36.82
2
CH ), 31.82 (CH ), 29.41 (CH ), 29.37 (CH ), 29.33 (CH ), 29.21 (CH ), 28.92
3
2
2
2
2
2
CH ), 26.42 (CH ), 22.65 (CH ), 14.12 (CH -10’’’) ppm. ESI-MS (m/z) (+): 427.3
2
2
2
3
+
-
-1
M] ; ESI-MS (m/z) (-): 126.9 [I] . IR ν (cm ): 3050 (w, C-H Ar.), 2925 (m, ν C-
max
as
H) 2853 (w, ν C-H), 1662 (s, C=O), 1591 (m, C=N), 1314 (m, C-N). HRMS (ESI-
s
+
TOF) m/z: [M + H] Calc for C H N O 427.2740, found 427.2744.
2
9
35
2
3
-Benzyl-1-methyl-2-{[10-oxoanthracen-9(10H)-ylidene]methyl}-1H-
1
imidazol-3-ium iodide (1d). Yellow/Orange solid (η=38%); mp: 129-137 ᵒC. H
2
9