Photooxidation and phototoxicity of π-extended squaraines

Article abstract of DOI:10.1021/jm901727j

This paper describes the synthesis of π-extended squaraines and their photooxidation properties and gives an in-depth characterization of these molecules as photosensitizing agents. Squaraines show a strong absorption in the tissue transparency window (600-800 nm), and upon irradiation, they undergo a photooxidation process, leading to the formation of peroxide and hydroperoxide radicals according to a type I radical chain process. Confocal laser microscopy demonstrates that the designed squaraines efficiently internalize in the cytoplasm and not in the nucleus of the cell. In the dark, they are scarcely cytotoxic, but after irradition, they promote a strong dose-dependent phototoxic effect in four different cancer cells. In HeLa and MCF-7 cells, squaraines 4 and 5, thanks to their hydrocarbon tails, associate to the membranes and induce lipid peroxidation, as indicated by a marked increase of malonyldialdehyde after photodynamic treatment, in agreement with in vitro photooxidation studies. FACS, caspase-3/7 assays and time-lapse microscopy demonstrate that the designed squaraines cause cell death primarily by necrosis.

Full text of DOI:10.1021/jm901727j

2
188 J. Med. Chem. 2010, 53, 2188–2196
DOI: 10.1021/jm901727j
Photooxidation and Phototoxicity of π-Extended Squaraines
†
‡
‡
‡
§
Valentina Rapozzi, Luca Beverina, Patrizio Salice, Giorgio A. Pagani, Monica Camerin, and Luigi E. Xodo*
,†
†
‡
Department of Biomedical Science and Technology, University of Udine, P.le Kolbe 4, I-33100 Udine, Italy, Department of Materials Science and
§
INSTM, University of Milano-Bicocca, via Cozzi 53 I-20125 Milano, Italy, and Department of Biology, Via U. Bassi 58/B,
University of Padova, 35131 Padova, Italy
Received November 23, 2009
This paper describes the synthesis of π-extended squaraines and their photooxidation properties and gives
an in-depth characterization of these molecules as photosensitizing agents. Squaraines show a strong
absorption in the tissue transparency window (600-800 nm), and upon irradiation, they undergo a
photooxidation process, leading to the formation of peroxide and hydroperoxide radicals according to a
type I radical chain process. Confocal laser microscopy demonstrates that the designed squaraines
efficiently internalize in the cytoplasm and not in the nucleus of the cell. In the dark, they are scarcely
cytotoxic, but after irradition, they promote a strong dose-dependent phototoxic effect in four different
cancer cells. In HeLa and MCF-7 cells, squaraines 4 and 5, thanks to their hydrocarbon tails, associate to
the membranes and induce lipid peroxidation, as indicated by a marked increase of malonyldialdehyde after
photodynamic treatment, in agreement with in vitro photooxidation studies. FACS, caspase-3/7 assays and
time-lapse microscopy demonstrate that the designed squaraines cause cell death primarily by necrosis.
Introduction
aines significantly reduce the clonogenicity of AS52 Chinese
hamster ovary cells after photoactivation. Gayathri Devi et al.
26
The condensation of electron-rich compounds with squaric
acid gives rise to a variety of molecules known as squaraines.
As these molecules are characterized by an intense absorption
between 600 and 800 nm;where tissues are fairly transparent
to light;they have a potential as photosensitizers in photo-
1
-4
have shown that a tetraiodo-squaraine caused a reduction of
tumor volume and a reversal of biochemical markers to near-
normal levels in mice bearing a skin tumor, treated with tetra-
iodo squaraine and exposed to a 1000 W halogen lamp. In this
study, we report on the photobleaching and photooxidation of
mono- and bis-squaraines as well as their photosensitizing
properties in four different cancer cells. We provide strong
evidence that the mechanism of cell death promoted by this new
class of photosensitizers is mediated by lipid peroxidation.
5-8
dynamic therapy (PDT).
proposed as near-infrared dyes for the determination of low
In addition, squaraines have been
9
molecular mass aminothiols in human plasma. PDT is a thera-
peutic modality used in a number of diseases including psor-
It
1
0-13
iasis, age-related macular degeneration, and cancer.
involves the systemic or topic administration of a photosensi-
tizer that, after excitation, produces singlet oxygen and/or
reactive oxygen species (ROS), which cause photodamage to
Results and Discussion
Synthesis Squaraines 4 and 5. The bis-squaraine 4 was
prepared according to Scheme 1a. In brief, commercially avai-
14,15
cancer cells.
One of the main goals in PDT research is the
0
0
00
discovery of new photosensitizers possessing minimal dark
cytotoxycity, high photodynamic properties, preferential reten-
tion in diseased rather than in healthy tissues, chemical stability,
lable 2,2 :5 ,2 -terthiophene (1) was treated with 2 equiv of
BuLi at 0 °C in tetrahydrofuran (THF). The lithiated terthio-
phene was subsequently treated with 2 equiv of commercially
available butylsquarate to give, after acidic workup, bis-semi-
squarate (2). Acidic hydrolysis in refluxing AcOH and in the
presence of aqueous HCl gave the bis-semisquaraine (3), which
was finally converted in the bis-squaraine (4) as recently repor-
1
3,16,17
and a good cellular uptake.
phosensitizers have been proposed,
the clinic is a mixture of hematoporphyrines. It is used for
Although a number of new
18-25
the most widely used in
12
10
different cancers, but its absorption above 600 nm is weak.
2
7
Thus, there is great interest to search for new and more efficient
PDT molecules. In this paper, we report the synthesis and
photodynamic properties of two symmetric squaraines, namely,
ted. The analytically pure compound was obtained as a
golden brown precipitate. Squaraine 5 was, instead, prepared
following the method described by Santos and co-workers by
refluxing a BuOH/pyridine suspension of the benzothiazolium
salt and squaric acid under azeotropic distillation of the water
4
and 5 (Scheme 1), both of which have N-alkylbenzothiazole
residues as end-capping groups. The two squaraines each
contain two C-6 hydrocarbon chains that might favor their
association to the cell membranes. Little is known about
the photosensitizing properties of squaraines in cell cultures.
5
formed (Scheme 1b). Squaraines 4 and 5 are characterized by
an intense absorption band between 550 and 750 nm, and
because of this property, they have the potential in PDT as a
photosensitizer (Figure 1). Their absorption and emission
properties in different organic solvents are reported in Table 1.
Photobleaching and Photooxidation of Squaraines. When 5
is dissolved in air-equilibrated acetonitrile and irradiated
7
Ramaiah et al. have reported that halogenated monosquar-
*
To whom correspondence should be addressed. Tel: þ39.0432.494395.
Fax: þ39.0432.494301. E-mail: luigi.xodo@uniud.it.
r
pubs.acs.org/jmc Published on Web 02/04/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 2189
Figure 1. Absorption (black line) and fluorescence (red line) spectra for squaraines 4 and 5 in different organic solvents.
Scheme 1. (a) Synthetic Route to Bis-squarine 4 and (b) Synthetic Route to Squaraine 5
a
with a laser or halogen lamp, it undergoes extensive degrada-
tion (Figure 2). This behavior does not occur when acetoni-
trile is saturated with nitrogen, and it slowly occurs when air-
equilibrated low polar solvents such as toluene are used. We
rationalized this behavior in terms of the formation of a
charge-transfer encounter complex between the squaraine
and the molecular oxygen, eventually evolving in the forma-
tion of different chemical species. To unveil the photoche-
mical mechanism involved, we analyzed a completely sun-
bleached solution of 5 in air-equilibrated acetonitrile by
GC-MS. We observed, among other species, the formation
of 3-hexylbenzo[d]thiazol-2(3H)-one (3-HBT). We specu-
late that the formation of 3-HBT is in agreement with the
chemistry of photooxygenation of enaminones and other
2
8,29
electron-rich double bonds. In Scheme 2, we propose a
mechanism for the degradation of 5, in which the radicals
a
Abbreviations: PBS, phosphate-buffered saline solution; HepG2,
hepatocellular carcinoma cells; B78-H1, amelanotic murine melanoma
cells; MCF-7, breast cancer cells; HeLa, cervical cancer cells; 3-HBT, 3-
hexylbenzo[d]thiazol-2(3H)-one; TBARS, malonyldialdehyde bound to
thiobarbituric acid; 1) PagMDA, malonyldialdehyde.

2
190 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Rapozzi et al.
Figure 2. Change in the absorption spectra of 5 in air-equilibrated and nitrogen-bubbled acetonitrile upon irradiation with a laser source at
55 nm.
3
Table 1. Spectral Properties of the Mono- and Bis-squaraines 5 and 4 in
Different Air-Equilibrated Solvents
products distribution in the photosensitized oxygenation of
2
8
(
R)-(þ)-limonene. Row 1 shows the percentage distribu-
a
b
630
c
d
f
solvent
λ
abs,max (εmax
)
ε
λ
em
,
max
Φ
tion of oxidized products expected for singlet oxygen reac-
tion. Rows 2 and 3 show the observed product distribution
for squaraine 5 and 4 photosensitization. The fact that the
preponderance products are the two carveol derivatives V
and VI and the appearance of VII brought us to conclude
that the most likely mechanism involves the ablation of the
allylic hydrogen by a radical species (either a ROS or a triplet
carbonyl compound) and the reaction of the allyl radical
with molecular oxygen to yield a peroxy radical. The latter
can ablate the hydrogen of another limonene molecule to
yield a hydroperoxide and an allyl radical according to a type
toluene, sq 4
acetonitrile, sq 4
DMSO, sq 4
toluene, sq 5
acetonitrile, sq 5
DMSO, sq 5
a
659 (88300)
628 (78200)
647 (92000)
73200
78100
90600
35600
61000
53100
680
0.14
0.01
0.015
0.85
0.21
0.92
667, 720
680, 760
705
682 (295000)
663 (280000)
678 (301000)
675
683
Absorption maximum, in nm ( 0.5 nm, and maximum extinction
-1
-1
b
molar coefficient, in L mol cm , (10%. Extinction molar coeffi-
-
1
-1
c
cient at 630 nm, in L mol cm , (10%. Emission maximum, in nm (
d
.5. Fluorescence quantum yield evaluated using zinc phthalocyanine
0
in pyridine as standard, estimated error ( 10%.
1
2,13
I radical chain process.
The photochemical data col-
that are formed can explain the phototoxicity of these
molecules (vide infra). To further elucidate the nature and
role of the photoproduced species, we performed a photo-
degradation reaction on an air-equilibrated acetonitrile so-
lution containing 5 and limonene as the photooxidation
target. Limonene is a particularly suitable substrate for such
studies, as it gives different oxidation products, depending
on the kind of ROS formed in solution (singlet oxygen,
lected are in agreement with the photophysical one. The en-
counter complex between squaraine and molecular oxygen
can evolve by either photooxygenation of the enaminic bond
in the squaraine backbone or ROS production (Supporting
Information, S1). In both cases, the photogenerated pro-
ducts can then establish radical chain events (lipid peroxida-
tion) that lead to type I cytotoxcity in cells. Interestingly, the
same mechanism seems to describe the photooxigenation
behavior of a number of cyanine-based photosensitizers (e.g.,
3
0
peroxides, or superoxides). A solution containing squar-
aine 5 (2.5 mg, 0.04 mmol) and (R)-(þ)-limonene (50 mg,
33,34
merocyanine 540, indocyanine green, and MKT-077).
0
.37 mmol) in acetonitrile (5 mL) was irradiated with a
We carried out the same photoxidation experiments on
derivative 4 and observed the same product distribution
(Figure 3b). The main difference between the two com-
pounds is the reaction kinetics. The photoxidation observed
in the case of 4 in much slower, probably due to the fact that
the derivative 4 strongly aggregates in acetonitrile. In con-
clusion, our photooxidation studies demonstrate that the
designed squaraines, when photoactivated, produce perox-
ide, allyl, and hydroperoxide radicals.
filtered white light halogen lamp (300 W) for 30 min. The
reaction mixture was treated with NaBH4 (25 mg, 0.66
mmol) to reduce any formed peroxide to the corresponding
3
1
alcohol, according to Marine and Clemons. The solution
was then analyzed by GC-MS (Figure 3a). The oxidation
products were identified by comparison with literature data,
3
2
that is, using the NIST 05 database (Table 2). The oxyge-
nation of (R)-(þ)-limonene provides a particularly sensitive
fingerprint for the reactive intermediates. This substrate
contains two oxidizable double bonds: one disubstituted
and one trisubstituted. Photooxygenation by singlet oxygen
yields products of oxidation at the trisubstituted double
bond but not at the disubstituted double bond, which is
not reactive. When we compared our results with those
Phototoxicity and Mechanism of Photokilling of Squar-
aines. As the designed squaraines emit in the red, we used
confocal microscopy to investigate whether they are taken up
by cervical cancer cells (HeLa) cells. The images shown in
Figure 4a have been obtained after the cells were treated for
4 h with 4 or 5. It can be seen that both squaraines locate
mainly in the cytoplasm: The punctuated distribution of the
fluorescence may suggest that these compounds reside in
2
8-30
obtained by Foote et al.,
the photoreaction promoted
by the squaraine is not in keeping with a singlet-oxygen
mediated oxygenation. Table 3 shows the photooxidation

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 2191
Figure 3. GC-MS chromatograms showing resolution of a solution of (R)-(þ)-limonene in acetonitrile photosensitized by 5 (a) and 4
(
(
b). Experimental details including column and GC conditions are given in the Experimental Section. Peak identification: I and II, either
1S,4R)- or (1R,4R)-1-methyl-4-(prop-1-en-2-yl)cyclohex-2-enol; III and IV, either (1R,5S)- or (1S,5S)-2-methylene-5-(prop-1-en-2-yl)cyclo-
hexanol; V, (1R,5S)-2-methyl-5-(prop-1-en-2-yl)cyclohex-2-enol; VI, (1S,5S)-2-methyl-5-(prop-1-en-2-yl)cyclohex-2-enol; and VII, (R)-[4-(prop-
-en-2-yl)cyclohex-1-enyl]methanol (see Table 2).
1
Scheme 2. Proposed Mechanism for the Photodegradation of Squaraine 5
endocytotic vescicles and/or associate to the organelle mem-
branes, through their double hydrocarbon chains. The fact
that squaraines do not accumulate in the nucleus suggests
that their potential to cause DNA damage, mutations, and
carcinogenesis should be quite low.
The photodynamic effects promoted by the designed squa-
raines were tested in four cell lines: HeLa human cervical
carcinoma, HepG2 human hepatic carcinoma, B78-H1 mur-
ine amelanotic clone derived from melanoma B16, and
MCF-7 human breast cancer cell lines. An important prop-
erty that a photosensitizer must possess is noncytotoxicity in
the dark. We therefore evaluated the cytotoxicity of 4 and 5
in the dark as well as after irradiation, and the results ob-
tained have been plotted together for comparison. Figure 4b
shows that monosquaraine 5 is not cytotoxic in the dark at
the concentrations used for PDT experiments: A limited
cytotoxicity effect is observed at concentrations as high as
5 μM. Cytotoxicity effects were not observed neither in
primary human vein endothelial cells (Huvec) nor when
squaraine 5 was injected in mouse (3 μmol/kg, typical dose
for PDT experiments) incorporated in small unilamellar
liposomes of dioleoylphosphatidyl choline (Supporting

2
192 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Rapozzi et al.
Table 2. Oxidation Products Obtained by the Photooxidation of Limonene with Squaraines 5 or 4
peak
product
characteristic ions (m/z)
retention times (min)
I
(1S,4R)- or (1R,4R)-1-methyl-4-(prop-1-en-2-yl)cyclohex-2-enol
(1S,4R)- or (1R,4R)-1-methyl-4-(prop-1-en-2-yl)cyclohex-2-enol
(1R,5S)- or (1S,5S)-2-methylene-5-(prop-1-en-2-yl)cyclohexanol
(1R,5S)- or (1S,5S)-2-methylene-5-(prop-1-en-2-yl)cyclohexanol
(1R,5S)-2-methyl-5-(prop-1-en-2-yl)cyclohex-2-enol
91, 119, 79, 134
91, 119, 134, 79
91, 109, 134, 119
91, 119, 134, 109
109, 91, 84, 119
84, 91, 134, 119
79, 91, 67, 55
9.55
10.05
11.19
11.63
11.95
12.17
13.09
II
III
IV
V
VI
VII
(1S,5S)-2-methyl-5-(prop-1-en-2-yl)cyclohex-2-enol
(R)-[4-(prop-1-en-2-yl)cyclohex-1-enyl]methanol
2
8 a
Table 3. Photooxidation Products Distribution in the Photosensitized Oxygenation of (R)-(þ)-Limonene
limonene photooxidation products distribution (percentage)
I
II
III
IV
23
V
VI
VII
2
8
singlet oxygen sensitization
photo- sensitization by 5
photo- sensitization by 4
a
10
17 þ 16
34
19
5 þ 9
9
25
29
4
0
1
2
26
33
11 þ 11
6 þ 6
Row 1 shows the percentage distribution of oxidized products expected for singlet oxygen reaction. Rows 2 and 3 show the observed product
distribution for squaraines 5 and 4 photosensitization.
Figure 4. (a) Confocal laser microscopy images of HeLa cells treated for 4 h with squaraines 4 (8 μM) and 5 (1 μM). (b) Cytotoxicity of
increasing amounts of squaraine 5 delivered to four different cancer cell lines either in the dark (light blue) or after light treatment (fluence rate,
2
5 J/cm ) (dark blue). Viable cells were measured with a resazurine assay. Histograms report in ordinate the percent of relative proliferation,
1
that is, the ratio RFU
data are the means ( SDs of three independent experiments.
T
/RFU
C
ꢀ 100, where RFU
T C
is the fluorescence of treated cells, while RFU is the fluorescence of untreated cells. The
Information, S2). In contrast, when squaraine 5 is photo-
activated, it promotes a dramatic cytototoxic effect in all
tested cancer cells (up to 80%). This experiment was
carried out by treating the cells for 3 h with increasing
amounts of 5, followed by irradiation with a metal halogen
dose-response plots were also obtained with the bis-
squaraine 4 (Supporting Information, S3). From these
dose-response plots, we estimated the IC50 values re-
ported in Table 4. The data obtained clearly indicate that
5 is a stronger photokilling agent than 4: IC50 values of 4
(8-16 μM) are, depending on cell type, from 6 to 26 times
higher than those of 5 (0.4-2 μM). The lower bioactivity of
4 is likely due to its capacity to aggregate in aqueous
solution, as indicated by the UV spectra (Supporting
2
lamp at a fluence of 15 J/cm , and 24 h after light treat-
ment, a resazurine proliferation assay was carried out. As
a control, we treated the cells with the same amount of
DMSO present in the squaraine treatment (squaraines
were dissolved in DMSO and diluted with cell medium;
the final DMSO content was lower than 0.2%). Similar
7
Information, S4). Ramaiah et al. reported that haloge-
2
nated monosquaraines, irradiated at 34 J/cm , strongly

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 2193
Table 4. IC50 Values Relative to Squaraines 4 and 5 Delivered to Cancer
Cells
a
IC50 (μM)
cells
squaraine 4
squaraine 5
HepG2
B78-H1
MCF-7
HeLa
a
10.7
16.0
11.7
8.1
0.4
2.0
1.3
1.3
IC50 obtained by treating the cells for 3 h with the squaraines,
2
followed by irradiation for 30 min (15 J/cm ) and resazurine assay (24 h
after irradiation).
Figure 6. Level of TBARS in HeLa and MCF-7 cells treated
photodynamically with squaraines 4 (9.6 μM) and 5 (1.8 μM).
Ordinate reports the level of TBARS (nmol/mg protein) relative
to that of untreated cells, which was fixed to 100.
Figure 5. Phototoxicity of squaraines 5 (a) and 4 (b) in HeLa cells
up to 72 h after light treatment (15 J/cm ). Control are cells treated
only with light.
2
investigated, by FACS, photodynamically treated HeLa
cells stained with annexin V and propidium iodide (PI). In
early apoptosis, cell membranes lose their phospholipid
asymmetry, and phosphatidyl serine (PS), normally lo-
cated in the inner leaflet, jumps into the outer leaflet. As
annexin V shows a high affinity for PS, annexin V-FITC
can mark the cells in early apoptosis. In contrast, necrotic
cells, being permeable to annexin V and PI, are stained by
inhibited the cloning efficiency of AS52 Chinese hamster
ovary cells, with an estimated IC50 between 1 and 2 μM,
rather similar to that observed with 5. We also compared
the phototoxicity of squaraine 5 with that of clinical
mTHPC (Temoporfin) and found that the latter has IC50
values ∼10-fold lower than the former, under the same
conditions (not shown). Nevertheless, squaraine 5 can be
considered as a promising PDT drug because it is photo-
toxic at relatively low concentrations (e1 μM), and it is
not expected to cause the skin side effects typical of
porphyrins.
As little is known about the photosensitizing properties
of squaraines, we focused on this. First, we tested the
antiproliferation effect of 4 and 5 in HeLa cells, up to 72 h
following irradiation (Figure 5). When 4 and 5 are used at
concentrations slightly higher than IC50 (1.5 μM for 5 and
3
6
both dyes. Figure 7a shows a FACS analysis performed
2
after irradiation (15 J/cm ) of HeLa cells treated with 4 or
5. When the squaraine concentration is lower than IC ,
5
0
the percentage of necrotic and apoptotic cells is very low
(not shown), but at concentrations close to IC50 (9.5 μM
for 4 and 1.5 μM for 5), the fraction of cells in necrosis is
nearly three times higher than the fraction of apoptotic
cells, indicating that necrosis becomes the principal photo-
killing mechanism triggered by the squaraines. In the
FACS experiment, the percentage of necrotic cells is not
high because the analysis was performed only 30 min after
light treatment, in an attempt to detect apoptotic cells.
However, by means of a time-lapse experiment conducted
over an interval of 72 h with living HeLa cells treated with
4 or 5 at a concentration close to IC50 and grown in a
medium added with Hoechst (stains nuclei of living cells)
and PI (stains nuclei of necrotic cells), we found that most
cells die primarily by necrosis (Figure 7b). It can be seen
that necrotic cells appear swelled up and stained in red
because their membrane becomes permeable to PI. How-
ever, although in a limited percentage, there are also cells
with an apoptotic morphology, characterized by shrink-
9
.6 μM for 4), they completely arrest the growth over the
entire range of time considered (72 h), whereas at lower
doses they reduce the growth rate without completely
stopping proliferation. Considering that squaraines 4
and 5 produce reactive peroxides upon photoactivation
and that these molecules are expected to associate to the
membranes through their hydrocarbon tails, we hypothe-
sized that they could cause a lipid peroxidation injury by
measuring the level of malonyldialdehyde (MDA) bound
3
5
to thiobarbituric acid (TBARS), in both untreated
and photodynamically treated HeLa cells (Figure 6). As
compared to untreated cells, which show a basal level of
TBARS fixed to 100, treated cells (squaraine dose ∼ IC ,
3
7
age, membrane blebbing, and apoptotic bodies. The
activation of a caspase-dependent cell death pathway
was confirmed by a caspase-3/7 assay performed at differ-
ent times after PDT treatment (Figure 8). As the effector
caspases 3 and 7 were found activated, the data confirmed
the presence of apoptotic cells and suggest that the squar-
aines mediate a complex cell death pathway involving
primarily necrosis but also apoptosis.
5
0
2
that is, 9.5 μM for 4 and 1.5 μM for 5; fluence of 15 J/cm )
show an increase of TBARS up to 4-fold, indicating
that the photosensitizers cause a strong lipid peroxida-
3
5
tion. A similar behavior was observed with MCF-7
cells (Figure 6). To gain insight into the mechanism of
photokilling triggered by the designed squaraines, we

2
194 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Rapozzi et al.
Figure 7. (a) FACS analysis of squaraine-treated HeLa cells photodynamically treated with 4 (9.6 μM) and 5 (1.5 μM) and stained with
annexin V-FITC and PI. (b) Time-lapse experiment performed on living HeLa cells treated with Hoechst, PI, and 4 (9.6 μM) or 5 (1.5 μM). The
images were obtained 8 and 60 h after treatment. As a control, the cells were not treated with the squaraiens (top panels). Red cells are necrotic,
and blue cells are viable. Necrotic (n) and apoptotic (a) cells are indicated by white arrows.
increase of MDA, which hints to lipid injury by peroxidation.
In the dark, the squaraines are largely nontoxic, but when they
2
are irradiated with light at a fluence of 15 J/cm , they promote
a strong photodynamic effect that causes cell death in all
tested cell lines. Monosquaraine 5 is found to be a stronger
photosensitizer than bis-squaraine 4. The results of this paper
show that the synthesized squaraines, in particular 5, repre-
sent a new class of therapeutic agent for PDT.
Experimental Section
0
0
0
00
00
Synthesis of 4,4 -(2,2 :5 ,2 -Terthiophene-5,5 -diyl)bis(3-bu-
toxycyclobut-3-ene-1,2-dione) (2). Under nitrogen atmosphere,
0
0
00
Figure 8. Caspases 3/7 activity in untreated or squaraine-treated
HeLa cells. The cells were incubated with the squaraine 5 (a) or 4 (b)
for 3 h and irradiation at a fluence of 15 J/cm . The caspase activity
a solution of 2,2 :5 ,2 -terthiophene (0.65 g, 2.60 mmol) in anhy-
drous THF (100 mL) was added dropwise with a 1.6 M solution of
butyllithium in n-hexane (2.6 mL) at 0 °C. Suddenly, a green-
yellow precipitate was formed. After 30 min at room temperature,
a secondsolutionof3,4-dibutoxycyclobut-3-ene-1,2-dione (0.81 g,
2
was measured with Apo-ONE TM caspase-3/7 assay. The data are
the means ( SDs of at least two independent experiments. (*p <
0
.05; **p < 0.01).
5
.20 mmol) in anhydrous THF (5 mL) was added to the reaction
mixture. The resulting red solution was stirred for 10 min and then
poured into a saturated NH Cl solution (100 mL). After 30 min of
stirring, the mixture was extracted with AcOEt (2 ꢀ 50 mL). The
organic layer was dried over anhydrous Na SO , and then, the
Conclusion
4
In this paper, we describe the synthesis of two π-extended
squaraines, which, being characterized by a strong absorption
between 550 and 750 nm, are potential photosensitizers in
PDT. Upon irradiation, the designed squaraines undergo a
phootooxidation process leading, through a type I phooto-
process, to the formation of peroxide and hydroperoxide
species. In the tested cells, these reactive species trigger a cell
death primarily by necrosis. Confocal microscopy shows that
these molecules efficiently internalize in HeLa cells where they
accumulate in the cytoplasm. The squaraines induce a marked
2
4
solvent was removed under reduced pressure. The residual red oil
was taken up with cyclohexane to afford the pure product as a red
precipitate that was collected by filtration (0.53 g, 0.96 mmol,
37%). H NMR (CDCl
J = 4.00 Hz), 7.26 (2H, s), 4.94 (4H, m), 1.92 (4H, m), 1.54 (4H,
1
3
): δ 7.78 (2H, d, J = 3.98 Hz), 7.32 (2H, d,
m), 1.03 (6H, m).
0
0
0
00
00
Synthesis of 4,4 -(2,2 :5 ,2 -Terthiophene-5,5 -diyl)bis(3-hy-
droxycyclobut-3-ene-1,2-dione) (3). A solution of compound 2
(0.73 g, 1.32 mmol) in AcOH (30 mL) and water (15 mL) was

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 2195
refluxed for 2 h in a mixture of 20 mL of AcOH and 1 mL of 10%
HCl. After this period, the reaction mixture was cooled at
ambient temperature. The red precipitate thus obtained was
suspension obtained was extracted with 12 mL of n-hexane and
dried over Na SO . The dried solution was filtered and injected
in the GC-MS apparatus.
2
4
filtered off and washed directly on the filter with Et
desired product as brown solid (0.53 g, 1.20 mmol, 91%). H
2
O to give the
Cell Proliferation Assay. The percentage of viable cells was
determined by the resazurine assay following the manufac-
turer’s instructions (Sigma-Aldrich, Milan, Italy). The values
were obtained by spectrofluorometer Spectra Max Gemini XS
(Molecular Devices, Sunnyvale, CA). To reach the plateau
phase growth, defined cell densities were seeded in a 96-multi-
well plate: HeLa 5000 cells/well, HepG2 7000 cells/well, B78-H1
5000 cells/well, and MCF-7 7000 cell/well.
1
NMR (DMSO-d ): δ 7.47 (2H, d, J = 3.77 Hz), 7.41 (2H, d, J =
6
4
.03 Hz), 7.38 (2H, s).
Synthesis of Squaraine (4). A suspension of N-hexyl-2-methyl-
benzothiazolium iodide (0.195 g, 0.54 mmol) and the bis-semi-
squaraine 3 (0.119 g, 0.27 mmol) was suspended in a solution of
25 mL of BuOH, 5 mL of toluene, and 0.1 mL of pyridine. The red
suspension was refluxed for 5 h under a Dean-Stark trap to
azeotropycally remove the water formed. The reaction mixture was
then cooled at 0 °C, and a golden brown precipitate was formed.
The precipitate was isolated by suction filtration and washed
directly on the filter with 20 mL of MeOH to give the pure
Apo-ONE Caspase-3/7 Homogeneous Assay. The Apo-ONE
casapase-3/7 homogeneous assay (Promega, Milano, Italy) was
used to measure the activity of caspase-3/7. Following the
manufacturer’s instructions, the cells were grown in a 96-well
plate (density of 5000 cells/well) and exposed to 4 and 5 light
treatment. The assay was performed at different times (2.5, 16,
and 22 h) following light activation. Homogeneous caspase-3/7
reagent (substrate was dilute 1:100 with provided buffer) was
added, maintaining a 1:1 ratio of reagent to sample. The
measure of fluorescence of each well was read at an excitation
wavelength of 485 ( 20 nm and an emission wavelength of 530 (
25 nm (Spectra Max Gemini XS, Molecular Devices Corp., CA).
Lipid Peroxidation (TBARS Assay). HeLa cells were photo-
dynamically treated with 4 or 5. Twenty-four hours after treat-
ment, the cells were washed with phosphate-buffered saline
title compound as a dark powder (0.160 g, 0.17 mmol, 64%); mp
1
>
8
250 °C (dec). H NMR(DMSO-d
6
):δ 8.31 (2H, d, J = 7.91 Hz),
.08 (2H, d, J = 8.47 Hz), 7.74 (2H, t, J = 7.87 Hz), 7.63 (2H, t,
J = 7.64 Hz), 7.49 (2H, d, J = 3.84 Hz), 7.47 (2H, d, J = 3.75 Hz),
.44 (2H, s), 6.64 (2H, s), 4.72 (4H, t, J = 6.26 Hz), 1.79 (4H, quint,
J = 7.51 Hz), 1.47-1.36 (4H, m), 1.36-1.20 (8H, m), 0.90-0.80
6H, m). Anal. calcd for C48 7/2H O: C, 61.71; H, 5.29;
N, 3.00. Found: C, 61.78; H, 4.69; N, 2.83.
7
(
H N O
42 2 4
S
5
2
3
Spectroscopic Experiments. Steady-state UV/vis absorption
spectra were recorded using a UV/vis/NIR Jasco V570 spectro-
meter. Steady-state emission spectra were recorded using a Jasco
FP-6200 fluorometer. Fluorescence quantum yields (Φf)
6
solution (PBS), trypsinized, and pelleted. The pellets (10 cells)
were suspended in 2 mL of 0.9% NaCl and vortexed vigorously,
and an aliquot was taken from each sample for a Lowry protein
assay. The remaining samples were added with 230 μL of trichloro-
acetic acid-saturated solution (250 g of trichloroacetic acid in
100 mL of water), vortexed, and centrifuged at 3000g (10 min) to
precipitate the proteins. The supernatants (1.6 mL) were placed in
glass test tubes, and 200 μL of 14.4 mg/mL 2-thiobarbituric acid in
0.1 N NaOH solution were added. The samples were incubated for
45 min at 75 °C. Standards were prepared from the hydrolysis of
1,1,2,2-ethoxypropane in a 40% trichloroacetic acid solution.
After the samples were cooled, the absorbances of the samples
were read at 535 nm by Spectrophotometer Ultrospec 1100 Pro
(Amersham Pharmacia Biotech, Cambridge, United Kingdom)
Thiobarbituric acid reactive substances (TBARS) values were cal-
(
Table 1) were determined by comparing the integrated emission
intensity recorded upon irradiation of a standard of a sensitizer
at a given wavelength to the integrated intensity obtained after
irradiation of a standard, with a known quantum yield of
fluorescence, at the same wavelength. The integrated intensities
were recorded at different absorbances, and the slopes obtained
were then used to obtain Φ . If the standard and the sensitizer
f
were dissolved in different solvents, we accounted for differences
in the light collection efficiency due to different refractive
indexes, theoretically derived for a 90° angle setup, such as that
in our fluorimeter:
ꢀ
ꢁꢀ
ꢁ
2
slopesens
nsens
nstd
std
Φf ¼ Φf
35
slopestd
culated as nmol/mg of total protein.
FACS Analysis. Apoptosis was assessed using annexin V, a
protein that binds to phosphatidylserine (PS) residues, which
are exposed on the cell surface of apoptotic cells. The cells were
Photobleaching and Photooxidation of Squaraines. Analysis of
limonene photooxidized derivatives was carried out by GC-MS
using a Clarus 500 instrument (PerkinElmer Instruments, Shel-
ton, CT) equipped with a 30 m ꢀ 0.25 mm capillary column
Elite-5MS (0.25 μm film thickness) (PerkinElmer). Helium was
used as the carrier gas at a flow rate of 1 mL/min. The injector
temperature was 250 °C, and the samples were injected in a
splitless mode of injection. The oven temperature was set at
5
plated in a six-well plate at density of 5 ꢀ 10 cells/well. One day
after, the cells were treated with 4 and 5 for 3 h and illuminated
for 30 min. After light activation, the cells were washed with
PBS, trypsinized, and pelleted. The pellets were suspended in
1
(
00 μL of Hepes buffer with 2 μL of Annexin V and 2 μL of PI
annexin-V FLUOS Staining kit, catalog no. 11858777001,
Roche, Penzberg, Germany) and incubated for 10 min at
5 °C in the dark. Cells were immediately analyzed by FACScan
Becton-Dickinson, San Jose, United States). A minimum of
0000 cells for sample was acquired in list mode and analyzed
8
0 °C for 5 min, raised to 100 at 5 °C/min, held for 1 min, raised
to 200 at 20 °C/min, and finally held for 3 min. The mass
spectrometer was operated in the electron ionization mode
2
(
1
(
ionization energy of 70 eV). The source and transfer line
temperatures were 160 and 180 °C, respectively. Detection was
carried out in scan mode: m/z 50 to m/z 600. The limonene
photooxidation protocol was as follows: (a) Fifty milligrams of
limonene and 2.5 mg of the squaraine dye were dissolved in 5 mL
of acetonitrile avoiding light exposure. (b) The mixture was
illuminated for 30 min with a halogen lamp filtered below 500
nm (the white light coming from the 230 V, 300 W halogen lamp
using Cell Quest software. The cell population was analyzed by
FSC light and SSC light. The signal was detected by FL1
(
annexin-V-FLUOS) and FL-2 (PI). The dual parameter dot
plots combining annexin V-FITC and PI fluorescence show the
vial cell population in the lower left quadrant (annexinV-PI-),
the early apoptotic cells in the lower right quadrant (annexin
VþPI-), and the late apoptotic or necrotic cells in the upper
right quadrant (annexinVþPIþ).
2 2 7
was filtered with a 3% by weight K Cr O solution. The light
coming out was completely absorbed for wavelengths below 500
nm). (c) The solution was dried under reduce pressure, and the
remaining residue was dissolved in 3 mL of ethanol and cooled
Confocal Microscopy. To study the uptake, HeLa cells were
5
plated (2 ꢀ 10 ) on coverslips (diameter 24 mm) and after 24 h
treated with squaraine 4 or 5 for 3 h. The cells were illuminated
with a metal halogen white lamp at a fluence of 15 J/cm . Four
2
at 0 °C. A solution of NaBH
4
(25 mg) in ethanol (1 mL) was
added to reduce the peroxides to alcohols, and the resulting
mixture was stirred in the dark for 1 h. The mixture was poured
in water, and the ethanol was removed at reduced pressure. The
hours after photoactivation, the glasses were prepared. The cells
were washed twice with PBS and fixed with 3% paraformalde-
hyde (PFA) in PBS for 20 min. After the cells were washed with

2
196 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
.1 M glycine, containing 0.02% sodium azide in PBS, to remove
Rapozzi et al.
0
(16) Castano, A. P.; Demidova, T. N.; Hamblin, M. R. Mechanisms in
photodynamic therapy: Part one-photosensitizers, photochemistry
and cellular localization. Photodiagn. Photodyn. Ther. 2004, 1, 279–
PFA and Triton X-100 (0.1% in PBS), the cells were incubated
with Hoechst to stain the nuclei. The cells were analyzed using a
Leica TCS SP1 confocal imaging system.
2
93.
(
17) Castano, A. P.; Demidova, T. N.; Hamblin, M. R. Mechanisms in
photodynamic therapy: part three-photosensitizers pharmacoki-
netics, biodistribution, tumor localization and modes of tumor
destruction. Photodiagn. Photodyn. Ther. 2005, 2, 91–106.
18) Allison, R. R.; Downie, G. H.; Cuenca, R.; Hu, X. H.; Childs, C. J.;
Sibata, C. H. Photosensitizers in clinical PDT. Photodiagn. Photo-
dyn. Ther. 2004, 1, 27–42.
19) Sessler, J. L.; Miller, R. A. Texaphyrins: New drugs eith diverse
clinical applications in radiation and photodynamic therapy. Bio-
chem. Pharmacol. 2000, 59, 733–9.
(20) Comuzzi, C.; Cogoi, S.; Overhand, M.; van der marel, G. A.;
Overkleeft, H. S.; Xodo, L. E. Synthesis and biological evaluation
of new pentaphyrin macrocycles for photodynamic therapy.
J. Med. Chem. 2006, 49, 196–204.
21) Rapozzi, V.; Lombardo, C.; Cogoi, S.; Comuzzi, C.; Xodo, L. E.
Small interfering RNA-mediated silencing of glutathione S-trans-
ferasi A1 sensitizes hepatic carcinoma cells to photodynamic
therapy with pentaphyrins. ChemMedChem 2008, 3, 565–568.
22) Rapozzi, V.; Miculan, M.; Xodo, L. E. Evidence that photoacti-
vated pheophorbide a causes in human cancer cells a photody-
namic effect involving lipid peroxidation. Cancer Biol. Ther. 2009,
Time-Lapse. The HeLa cells were plated at density of 2.5 ꢀ
5
0 cells in a six-well plate. The day after, the cells were treated
1
with squaraine 4 or 5. After photoactivation, the medium was
removed and substitued with fresh medium containing Hoechst
and PI (both at 2 μg/mL) to detect necrotic and apoptotic cells.
Samples were placed in a temperature (37 °C) and 5% CO -
controlled coverslip holder of the microscope. The time-lapse
experiment was documented by serial images obtained with a
Leica AF6000 LX (Leica Microsystems, Heidelberg, Germany).
(
(
2
Acknowledgment. This work has been carried out with the
financial support of FVG-2008, PRISMA 2007, and PRIN
2
(
(
(
(
007.
Supporting Information Available: Information about (i)
photooxidation of squaraines 4 and 5, (ii) a brief description
of the in vivo toxicity experiment, (iii) dose-response plots of
squaraine 4 in different cancer cell lines, and (iv) the absorption
spectra of squaraines 4 and 5 in water. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
4, 1–10.
23) Taquet, J. P.; Frochot, C.; Manneville, V.; Barberi-Heyob,
M. Phthalocyanines covalently bound to biomolecules for a tar-
geted photodynamic therapy. Curr. Med. Chem. 2007, 14, 1673–
1
687.
References
24) Juarranz, A.; Jaen, P.; Sanz-Rodriguez, F.; Cuevas, J.; Gonzales, S.
Photodynamic therapy of cancer. Basic principles and applica-
tions. Clin. Transl. Oncol. 2008, 10, 148–154.
(25) Josefsen, L. B.; Boyle, R.W. Photodynamic therapy: Novel third-
generation photosensitizers one step closer? Br. J. Pharmacol. 2008,
154, 1–3.
(26) Gayathri Devi, D.; Cibin, T. R.; Ramaiah, D.; Abraham, A. J. Bis
(3,5-diiodo-2,4,6-trihydroxyphenyl) squaraine: a novel candidate
in photodynamic therapy for skin cancer models in vivo. Photo-
chem. Photobiol., B 2008, 92, 153–159.
(27) Yagi, S.; Ohta, T.; Akagi, N.; Nakazumi, H. The synthesis and
optical properties of bis-squarylium dyes bearing arene and thio-
phene spacers. Dyes Pigm. 2008, 77, 525–536.
(
(
(
(
(
1) Sprenger, G. E.; Ziegenebein, W. Cyclobutenediylium Dyes. An-
gew. Chem., Int. Ed. Engl. 1968, 7, 530–535.
2) Maahas, G.; Hegenberg, P. Syntheses and derivatives of squaric
acid. Angew. Chem., Int. Ed. Engl. 1966, 5, 888–893.
3) Keil, D.; Hartmann, H. Synthesis and characterization of a new class
of unsymmetrical squaraine dyes. Dyes Pigm. 2001, 49, 161–179.
4) Schmidt, A. H. Oxocarbons; West, R., Ed.; Academic Press:
New York, 1980; p 185.
5) Santos, P. F.; Reis, L. V.; Almeida, P.; Serrano, J. P.; Oliveira,
A. S.; Vieira-Ferreira, L. F. Efficiency of single oxygen generation
of aminosquarylium cyanines. J. Photochem. Photobiol., A 2004,
163, 267–269.
(
6) Ramaiah, D.; Eckert, I.; Arun, K. T.; Weidenfeller, L.; Epe, B.
Squaraine dyes for photodynamic therapy: Study of their cytotoxi-
city and genotoxicity in bacteria and mammalin cells. Photochem.
Photobiol. 2002, 76, 672–677.
(28) Foote, C. S.; Lin, J. W.-P. Chemistry of singlet oxygen. VI.
Photooxygenation of enamines: Evidence for an intermediate.
Tetrahedron Lett. 1968, 9 (29), 3267–3270.
(29) Mazur, S.; Foote, C. S. Chemistry of singlet oxygen. IX. Stable
dioxetane from photooxygenation of tetramethoxyethylene. J. Am.
Chem. Soc. 1970, 92 (10), 3225–3226.
(30) Foote, C.; Wexler, S.; Ando, W. Chemistry of Singlet Oxygen. III.
Product Selectivity. Tetrahedron Lett. 1965, 46, 4111–4118.
(31) Marine, S.; Clemons, J. Determination of limonene oxidation
products using SPME and GC-MS. J. Chromatogr. Sci. 2003,
41 (1), 31–35.
(32) Mikaia, A.; Stein, S. E.; Zaikin, V.; Zhu, D.; Milman, B.; Babishok,
V.; Zenkevich, I.; Linstrom, P.; Mirokhin, Y.; Tchekhovskoi, D.;
Mallard, W. G.; Sparkman, O. D.; Sparkman, J. A. NIST/EPA/
NIH Mass Spectral Library;Version 2.0d; National Institute of
Standards and Technology: Gaithersburg, MD, 2005.
(
7) Ramaiah, D. A.; Joy, N.; Chandrasekar, N. V.; Eldho, S. D.;
George, M. V. Halogenated squaraine dyes as potential photo-
chemotherapeutic agents. Synthesis and study of photophysical
properties and quantum efficiencies of single oxygen generation.
Photochem. Photobiol. 1997, 65, 783–790.
(
8) Ramaiah, D.; Eckert, I.; Arun, K. T.; Weidenfeller, L.; Epe, B.
Squaraine dyes for photodynamic therapy: Mechanism of cyto-
toxicity and DNA damage induced by halogenated squaraine dyes
plus light (>600 nm). Photochem. Photobiol. 2004, 79, 99–104.
9) Sreejith, S.; Divya, K. P.; Ajayaghosh, A. A near-infrared squar-
aine dye as a latent ratiometric fluorophore for the detection of
aminothiol content in blood plasma. Angew. Chem., Int. Ed. Engl.
(
2
008, 47, 7883–7887.
(33) Kawakami, M.; Suzuki, M.; Kawai, H.; Ogawa, K.; Shishido, T. A
self-sensitized photoreaction of rhodacyanine dye, MKT 077.
Tetrahedron Lett. 1998, 39 (13), 1763–1766.
(34) Hoebeke, M.; Piette, J.; van de Vorst, A. Photosensitized produc-
tion of singlet oxygen by merocyanine 540 bound to liposomes.
J. Photochem. Photobiol., B: Biol. 1991, 9 (3-4), 281–294.
(35) Wong, S. H. Y.; Knight, J. A.; Hopfer, S. M.; Zaharia, O.; Leach,
C. N.; Sunderman, F. W. Lipoperoxides in plasma as measured by
liquid-chromatographic separation of malondialdehyde-thiobar-
bituric acid adduct. Clin. Chem. 1987, 33, 214–220.
(
(
(
10) Dolmans, D. E.; Fukumura, D.; Jain, R. K. Photodynamic therapy
for cancer. Nat. Rev. Cancer 2003, 3, 380–387.
11) Castano, A. P.; Mroz, P.; Hamblin, M.R.. Photodynamic therapy
and anti-tumour immunity. Nat. Rev. Cancer 2006, 6, 535–545.
12) Dougherty, T. J.; Gomer, C. J.; Henderson, B. W.; Jori, G.; Kessel,
D.; Korbelik, M.; Moan, P.; Peng, Q. Photodinamic Therapy.
J. Natl. Cancer Inst. 1998, 90, 889–905.
(
13) Miller, J. Chemistry behind the News;Photodynamic Therapy:
The Sensitization of Cancer Cells to Light. J. Chem. Educ. 1999, 76,
5
92–594.
(36) Koopman, G.; Reutelingsperger, C. P. M.; Kuijten, G. A. M.;
Keehnen, R. M. J.; Palsvan, S. T.; Oers, M. H. J. Annexin V for
flow cytometric detection of phosphatidylserine expression on B
cells undergoing apoptosis. Blood 1994, 84, 1415–1420.
(37) Ziegler, U.; Groscurth, P. Morphological features of cell death.
News Physiol. Sci. 2004, 19, 124–128.
(
(
14) Pervaiz, S. Reactive oxygen-dependent production of novel photo-
chemotherapeutic agents. FASEB J. 2001, 15, 612–617.
15) Josefsen, L. B.; Boyle, R. W. Photodynamic therapy and the
development of metal-based photosensitisers. Met. Based Drugs
2008, 276109–276133.