Synthesis of a photostable near-infrared-absorbing photosensitizer for selective photodamage to cancer cells

Article abstract of DOI:10.1002/chem.201402285

A new class of near-infrared (NIR)-absorptive (>900 nm) photosensitizer based on a phenothiazinium scaffold is reported. The stable solid compound, o-DAP, the oxidative form of 3,7-bis(4-methylaminophenyl)-10H-phenothiazine, can generate reactive oxygen species (ROS, singlet oxygen and superoxide) under appropriate irradiation conditions. After biologically evaluating the intracellular uptake, localization, and phototoxicity of this compound, it was concluded that o-DAP is photostable and a potential selective photodynamic therapy (PDT) agent under either NIR or white light irradiation because its photodamage is more efficient in cancer cells than in normal cells and is without significant dark toxicity. This is very rare for photosensitizers in PDT applications. Selective PDT agent: A photostable near-infrared (NIR)-absorptive phenothiazinium derivative, o-DAP, has been successfully synthesized and demonstrated to be a potential tumor-specific NIR-absorptive (>900 nm) photosensitizer for photodynamic therapy (PDT; see figure). This compound is expected to become a PDT reagent in a solar environment.

Full text of DOI:10.1002/chem.201402285

DOI: 10.1002/chem.201402285
Full Paper
&
Photosensitizers
Synthesis of a Photostable Near-Infrared-Absorbing
Photosensitizer for Selective Photodamage to Cancer Cells
[
a]
Tung-Sheng Hsieh, Jhen-Yi Wu, and Cheng-Chung Chang*
Abstract: A new class of near-infrared (NIR)-absorptive
>900 nm) photosensitizer based on a phenothiazinium scaf-
localization, and phototoxicity of this compound, it was con-
cluded that o-DAP is photostable and a potential selective
photodynamic therapy (PDT) agent under either NIR or
white light irradiation because its photodamage is more effi-
cient in cancer cells than in normal cells and is without sig-
nificant dark toxicity. This is very rare for photosensitizers in
PDT applications.
(
fold is reported. The stable solid compound, o-DAP, the oxi-
dative form of 3,7-bis(4-methylaminophenyl)-10H-phenothia-
zine, can generate reactive oxygen species (ROS, singlet
oxygen and superoxide) under appropriate irradiation condi-
tions. After biologically evaluating the intracellular uptake,
Introduction
of a photosensitizer to cancer cells and its subcellular distribu-
tion not only depend on cellular processes but also on the
physicochemical properties and structural characteristics of the
Photodynamic therapy (PDT) presents several advantages over
surgery and radiation therapy because it is relatively non-inva-
sive and a localized form of therapy. It is known that PDT in-
volves the light activation of a photosensitizer with subse-
quent in situ production of singlet oxygen and other reactive
oxygen species (ROS) that destroy photosensitizer-accumulat-
[7]
molecule itself. Therefore a method to modify the photosen-
[8]
sitizer with conjugating recognizers (antibodies, epidermal
[9]
[10,11]
growth factors etc.
) has been developed to improve the
selectivity of sensitizers. However, these high-molecular-
weight, branched peptide conjugates are difficult to synthe-
size, purify, and characterize. Although enhanced cellular
uptake and selectivity for tumor tissues has been achieved,
most of these examples are usually found in cytoplasmic mem-
branes rather than in sensitive intracellular sites.
[
1,2]
ed cells by necrosis and/or apoptosis.
That is, a photosensi-
tizer that predominantly accumulates in cancer tissues com-
bined with accurate and highly penetrative light delivery is
very important requirement for a good photosensitizer. To de-
velop a photosensitizer with an absorption maximum at
a longer wavelength, organic near-infrared (NIR) photonic
dyes, which have been widely utilized in the fields of life and
materials sciences, were considered. Because of the advantag-
es in biomedical applications of minimizing photon interfer-
ence from absorption, auto-fluorescence, and scattering from
Among the many reported or commercially available photo-
sensitizers, the investigation of non-porphyrin photosensitizers
has primarily focused on phenothiazinium-based photosensitiz-
[
12–15]
ers such as Methylene Blue (MB),
a Nile Blue ana-
[16–18]
logue,
and the chalcogenopyrylium class of sensitiz-
Significant drawbacks of these photosensitizers appear
[19,20]
ers.
[
3]
solvents, biomolecules, and the biological matrix, the utiliza-
tion of dyes that absorb and emit in the NIR range match the
tissue-transparent window and allow for deeper light penetra-
tion. Hence, NIR dyes that are capable of generating cytotoxic
singlet oxygen would find promising applications as photosen-
sitizers for the non-invasive treatment of deep tumors by
to be their low light-to-dark toxicity ratios and their relatively
low absorption maxima (<700 nm) in aqueous solutions,
which, with the exception of MB, make them unavailable for in
[21,22]
vivo applications.
In fact, most photosensitizers have some
application limitations and therefore it is important to develop
a new generation of photosensitizers with high efficiency and
photostability that are widely applicable under various
conditions.
[
4,5]
PDT.
In addition, selective photodamage is the most important
criterion for a superior PDT reagent. The lack of a targeting
mechanism causes poor tumor selectivity, leading to a low
Based on the above discussion, and with appropriate chemi-
cal modification, we have designed and produced a new NIR-
absorptive phenothiazinium-based photosensitizer that over-
comes the issues of dark toxicity and nonselective photodam-
age. This new photosensitizer was easily achieved by attaching
aniline groups directly onto the phenothiazine core scaffold,
followed by oxidation to give a stable phenothiazinium prod-
uct with an absorption maxima above 900 nm. Preliminary bio-
logical studies on this compound indicate that it is highly
membrane permeable and can selectively localize within spe-
[
6]
tumor/normal tissue accumulation ratio. The efficient delivery
[
a] T.-S. Hsieh, J.-Y. Wu, Prof. Dr. C.-C. Chang
Graduate Institute of Biomedical Engineering
National Chung Hsing University
2
50 Kuo Kuang Road, Taichung 402, Taiwan (R.O.C.)
E-mail: ccchang555@dragon.nchu.edu.tw
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201402285.
Chem. Eur. J. 2014, 20, 9709 – 9715
9709
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cific subcellular organelles, which suggests that it has potential
as a PDT agent.
Results and Discussion
Molecular preparation and basic spectroscopic properties
Because commercial phenothiazinium photosensitizers based
on MB exhibit much higher singlet oxygen quantum yields
[
14]
than the benzologues or benzo[a]phenothiazinium, it was
logical to pursue the rational design and synthesis of ana-
logues based on MB and its derivatives as lead compounds.
Furthermore, it is known that substitution of the phenothiazi-
nium framework with an electron-donating group can lead to
more active photosensitizers, whereas an electron-accepting
[
14,23,24]
group decreases the photosensitizing effect.
However,
both in cancer PDT and in the emerging antimicrobial regi-
mens, the commercial phenothiazinium dyes based on amino-
phenothiazines potentially exhibit metabolic deactivation
through chromophore reduction to the leuco dye, which
[
25,26]
Scheme 1. Synthetic route to phenothiazinium-based photosensitizers.
would result in a lower therapeutic efficacy.
In this investi-
Reagents and conditions: 1) pinacolborane, [Pd(PPh)
) N-bromosuccinimide (NBS), THF, ice bath; 3) K CO
DME/H O (5:1), N , reflux; 4) I , chloroform, RT.
3
Cl
, (o-tol)
2
], dioxane/Et
3 2
N, N ;
gation, the phenothiazinium moiety, the oxidation product of
phenothiazine, was selected as the electron acceptor and ani-
line was chosen as the electron-donating unit. The goal of the
molecular design, shown in Scheme 1, was to synthesize a 3,7-
bis(4-aminophenyl)phenothiazine (2) and not 3,7-bis(phenyl-
2
2
3
3
P, Pd(OAc) ,
2
2
2
2
[
26]
amino)phenothiazine. Then, the parent compound was con-
[
27]
veniently oxidized with iodine to furnish 3,7-bis(4-aminophe-
nyl)phenothiazinium tetraiodide (o-DAP). We expected that
this type of phenothiazinium derivative would retain the PDT
superiority of MB but diminish the possibility of intracellular re-
duction and dark toxicity. More importantly, a lower-energy
NIR absorption was also expected.
Figure 1a shows the progress of the oxidation of compound
2
by I in DMSO solution. Evaluation of the titration process re-
2
vealed that the oxidation of the phenothiazine occurs immedi-
ately upon titration. In the first stage (<1 equiv of I ), the
2
peaks at 296 and 355 nm of compound 2 disappeared and
peaks at 538 and 740 nm of an intermediate species appeared.
Then, peaks at 293, 370, and 957 nm grew with an isosbestic
point at 545 nm during the second stage of the titration. On
the basis of Figure 1b, which shows the effect of solvent on
the synthesized product powder, the second species obtained
during the titration can be ascribed to the formation of the ox-
idative product, o-DAP. On the other hand, the intermediate
should be the product of deprotonation at NÀH of the pheno-
thiazine moiety (Figure S1 in the Supporting Information). We
assumed that these visible and NIR peaks were produced by
either electron delocalization between phenothiazinate and
[
28,29]
aniline or the hybridization of the sulfur atom.
In addition,
the apparent solvent effect of o-DAP revealed in Figure 1b in-
dicates that there is considerable scope for altering the chemi-
cal or physical properties of the delocalized phenothiazinium
Figure 1. Variation of the absorption spectra of: a) compound 2 in DMSO
solution in a titration against iodine, and b) the final powder compound
o-DAP dissolved in various solubilizing solvents.
[
30]
ring system, as for MB, and this should be considered when
cells are incubated with this compound.
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org
9710
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Singlet oxygen generation and photostability
The production of singlet oxygen was studied by monitoring
the reaction of the known singlet oxygen acceptor 1,3-diphe-
nylisobenzofuran (DPBF) with photosensitizer-generated sin-
glet oxygen. This reaction was monitored by following the dis-
[31]
appearance of the absorbance band of DPBF at 410 nm.
Figure 2 shows the oxidation of DPBF following the excitation
of the absorption band of o-DAP in DMSO as solvent under an
NIR light source (700–1000 nm) over a period of time. It was in-
teresting to observe that the decrease in DPBF was accompa-
nied by a decrease in the intensity of the absorption maxima
of o-DAP (Figure 2a). The mechanism of photobleaching in
dyes has been shown to be complex; it may be sensitive to
the environment, excitation conditions or may be specific to
[
32,33]
each dye.
As seen in Figure 2b, o-DAP itself is photostable
under similar irradiation conditions in the absence of DPBF.
Moreover, it was found that the photobleaching rate is depen-
dent on the concentration of DPBF (Figure S2 in the Support-
ing Information). Thus, we can infer that the decay of the dye
absorption is likely to result from its reaction with the singlet
oxygen produced upon irradiation, which becomes more prob-
able as the concentrations of the photosensitizer and DPBF
[
34]
become comparable.
Thus, the signal of singlet oxygen
(1270 nm) was directly measured by NIR spectroscopy (see
inset of Figure 2 b). It is notable that the singlet oxygen peak
can be observed in deuterium solvent and on bubbling with
oxygen. This effect is expected because the lifetime of the trip-
[35]
let state is strongly enhanced in deuterated solvents. This
finding demonstrates that o-DAP has the ability to produce
singlet oxygen and is a potential photosensitizer candidate.
Figure 2. a) Variation of the absorption spectra of 10 mm DPBF in the pres-
ence of 5 mm o-DAP in DMSO after irradiation with an average 3 mWcm
À2
light source (700–1000 nm; the arrows indicate adsorption changes from 0
to 36 min). Inset: spectrum of DPBF in the absence of o-DAP. b) Variation of
the absorption spectra (12 measurements during 36 min) of 5 mm o-DAP
under similar irradiation conditions in the absence of DPBF. Inset: NIR lumi-
nescence of singlet oxygen (ca. 1270 nm) sensitized by irradiation at 950 nm
Real-time photodamage and photostability in cells
Having demonstrated that o-DAP has the ability to generate
singlet oxygen as well as showing photostability, it was inter-
esting to examine whether o-DAP could be used as a tool for
NIR cell photosensitization. We found that gradational irradia-
tion periods were optimal for observing different photoin-
duced cytotoxicities between cancer and normal cells. Here,
each phototoxicity protocol period was divided up into
bubbling, c), and D
2
O (with (*) and with-
in DMSO (a), [D
6
]DMSO (O
2
out (ꢁ) O bubbling).
2
contrast, MRC-5 normal cells remained healthy under the same
experimental conditions (Figure 3b). However, o-DAP killed
both cancer and normal cells without gradational irradiation
cycles, that is, under continuous irradiation, o-DAP showed no
selective photodamage towards cancer cells. The results of the
light-induced cytotoxicity assays are presented in Table 1.
a period of irradiation of 60 min, with as average light intensity
À2
(
700–1000 nm) of 3 mWcm on the dish surface, and a 30 min
period of incubation. The results in Figure 3a illustrate the
photodamage process through
real-time phase-contrast images
recorded under a microscope. It
is clear that HeLa cancer cells
become condensed, and acute
cell death caused by irradiation
with NIR light of cells incubated
with o-DAP could be observed.
We also observed that every irra-
diation cycle caused gradational
Figure 3. Selected bright-field images from the real-time photodamage assays of: a) HeLa cancer cells, and
b) MRC-5 normal cells incubated with 15 mm o-DAP for 8 h. Four irradiation cycles were performed; each cycle is
divided up into a 60 min period of irradiation (700–1000 nm light source, 3 mWcm ) and a 30 min period for
cell death and, eventually, acute
total cell death was observed
À2
after four cycles of irradiation. In incubation. The same dish was then observed after incubation, overnight.
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9711

Full Paper
light illumination with increasing concentrations of o-
DAP. Figure 4 reveals that the cell lines displayed no
measurable dark toxicity with o-DAP up to a concen-
tration of 30 mm. In contrast, gradational irradiation
with NIR light or irradiation only once with white
light showed significant light-induced toxicity with
EC50 values of around 8 and 6.3 mm, respectively, de-
termined for the HeLa cells. Here, the very important
conclusion is that o-DAP causes apparent photodam-
age to HeLa cancer cells but not to MRC-5 normal
cells under appropriate irradiation conditions. This
suggests that o-DAP is a potentially useful reagent
for cell photosensitization due to its selective PDT, re-
gardless of its use in NIR light or sunlight. In fact, we
have studied the phototoxicities of o-DAP in several
cancer and normal cell lines. Distinct PDT effects
were also observed in the human lung adenocarcino-
ma A549, but smaller effects were observed in the
human lung adenocarcinoma cell lines CL1-0 and
human breast cancer cells MCF7 under our optimum
NIR irradiation conditions (gradational irradiation
with NIR light). Meanwhile, we also found that the
Detroit 551 human skin fibroblast normal cells dis-
Table 1. Irradiation conditions for light-induced cytotoxicity assays of compound
o-DAP.
[
a]
[
b]
Light
source
Light power
Irradiation
period
Course
Cell death
[%]
À2
[c]
[mWcm
]
[
d]
NIR
NIR
NIR
NIR
3
irradiate 60 min
cancer, <10
normal, <10
cancer, >80
normal, <10
cancer, >95
3
3
irradiate 60 min
incubate 30 min
irradiate 60 min
incubate 30 min
irradiate 60 min
repeat
three times
repeat
[
e]
four times
normal, <20
6
cancer, >60
normal, >40
cancer, >40
normal, <10
cancer, >80
normal, >30
cancer, >95
[f]
white light
white light
white light
3
irradiate 60 min
irradiate 120 min
irradiate 30 min
6
12
[
g]
normal, <20
[a] The MRC-5 normal cells and HeLa cancer cells were incubated with the compound
in the dark for 8 h. No more compounds were added in the irradiation period.
b] Average light intensity. [c] Observed overnight. [d] 700–1000 nm. [e] See Figures 3
[
and 4. [f] Xenon lamp, no filter. [g] See Figure 4 and Figure S2 in the Supporting
Information.
More excitingly, the selective cancer cell phototoxicity of o-
DAP became more efficient when the system was irradiated
under a white light source (without a 700–1000 nm mirror
module). Under these conditions (Figure S3 in the Supporting
Information), we also observed selective photodamage of HeLa
cancer cells but not of normal MRC-5 cells when the dish was
irradiated for 30 min and incubated, overnight. This indicates
that the UV absorption of o-DAP at approximately 400 nm con-
tributes some phototoxicity. A similar result was observed
when this phototoxicity protocol was performed under a solar
generator light source. Based on the results above, we then as-
sessed the quantitative cellular toxicity of different forms of
played no measurable phototoxicity with o-DAP under the
same irradiation conditions (Figure S4 in the Supporting Infor-
mation).
Intracellular accumulation and subcellular localization
The fluorescent emission of o-DAP in the NIR region is very
low, but a blue fluorescent emission can be observed when ex-
citing the compound with UV light. Thus, we aimed to deter-
mine the quantitative accumulation and subcellular localization
of o-DAP by using a blue light cube, which passed light
through a 370(Æ10) nm band pass filter and the emission was
collected through a 450 nm long pass filter and fluorescent mi-
croscopy. The inset of Figure 5 shows the intracellular accumu-
lation of o-DAP under these conditions, and indicates that the
uptake of o-DAP in cancer cells is barely higher than in normal
cells. In addition, the kinetic quantitative uptake of o-DAP was
[36]
determined by the extraction of o-DAP from cells. The in-
crease in the concentration of o-DAP in the cells was quanti-
fied by monitoring the absorption of o-DAP in the ethyl ace-
tate layer after extracting it from the cells, as described in the
Experimental Section (Figure 5). Our results reveal that the ac-
cumulation of o-DAP in the cancer cells is slightly higher than
in the normal cells. This observation leads us to propose that
the difference in uptake of o-DAP in the cancer and normal
cells should be the one factor determining selective photo-
damage, even though the uptake difference is not readily
apparent.
Figure 4. Dark toxicity and phototoxicity of HeLa cancer cells and MRC-5
normal cells when treated with variable concentrations of o-DAP for 8 h
prior to irradiation. Cell death was tested, overnight; irradiation with NIR
light involved gradational illumination, whereas with the white light source
only one illumination was performed. The experimental assays were per-
formed as described in Figure 3 and Figure S2 in the Supporting Informa-
tion, respectively. Each data point represents the average of four separate
experiments.
Meanwhile, because the emission of o-DAP is barely ob-
served by confocal microscopy, fluorescence microscopy was
also used to determine the cellular localization of o-DAP. As de-
tailed in the Experimental Section, the cells that had been in-
cubated with o-DAP were double-stained with cellular Tracker
Red, which merges with the blue-light emission of o-DAP. Fig-
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org
9712
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 6a reveals that o-DAP may localize predominantly in the
endoplasmic reticulum (ER) and lysosome of cancer cells, as
seen by the bright spots in the images. In the double-stained
normal cells, the compound was not localized in the mitochon-
dria or lysosome, and was only minimally localized in the ER
(
Figure 6b). It is known that the intracellular localization of
photosensitizers strongly affects the mechanism of cell death.
The biological efficacy of PDT sensitizers of tumors depends on
their efficient translocation across cellular membranes and
their delivery into specific organelles within cancer cells. In ad-
dition to mitochondria, the ER and lysosome are very attractive
sites for the localization of photosensitizers to improve the
[
37–39]
efficiency of PDT,
and we have also reported some ER- or
[40,41]
lysosome-targeting photosensitizer candidates.
In summary, the localization and uptake differences between
cancer and normal cells are the likely factors that enable o-
DAP to induce selective photodamage. Moreover, an interest-
ing result was attained when cell death was determined fol-
Figure 5. Intracellular accumulation of o-DAP in normal and cancer cells,
which was determined from the fluorescence calibration curves of the ethyl
acetate layer after extraction from the cells. Each data point represents the
average of three separate experiments. Insets: photographs showing the flu-
orescent image of 15 mm o-DAP in cancer and normal cells. Excitation light
filter: 370(Æ10) nm band pass; emission light filter: 450 nm long pass.
lowing double-staining with Hoechst 33342/propidium iodide
[42]
(
PI).
Fluorescence microscopy revealed an apparent en-
hancement of blue emission in the nucleus at an early stage
double-stained immediately after illumination, see Figure S5 in
(
the Supporting Information), which is due to the interaction
between Hoechst and condensed DNA, whereas the red fluo-
rescence from PI became dominant in the nucleus when the ir-
radiated cell line was incubated, overnight. It is likely that the
nuclear membrane is damaged during the later period of irra-
diation. The cell death pathway very likely involves disruption
of an anti-apoptosis protein of the mitochondria or ER, as we
discussed above. However, this is the first report of a NIR PDT
agent selective towards cancer cells but not normal cells.
Generation of new reactive oxygen species
It is interesting that the PDT effect of o-DAP under white light
illumination is better than that under NIR light, and its efficien-
cy is greater than we expected. Hence, we tried to verify
whether PDT with o-DAP occurs by another mechanism, such
as type I PDT. This type of mechanism involves hydrogen-atom
abstraction or electron transfer between the excited sensitizer
and a substrate to yield free radicals. The free-radical probe
TEMPO-9-ac, which can capture radicals and show fluorescence
(
lex/em =358/440 nm), can be an indicator of a type I PDT pho-
[43,44]
tosensitizer,
and shows no apparent response to singlet
[45]
oxygen. When we studied the free-radical generation of o-
DAP under aqueous conditions under white light illumination,
an apparent increase in emission from TEMPO-9-ac was ob-
served (Figure 7), and the emission enhancement was slightly
quenched in D O (Figure S6 in the Supporting Information).
2
This indicates that singlet oxygen is unlikely to be the cause of
the fluorogenic effect of TEMPO-9-ac. A recent report revealed
that a type I PDT photosensitizer (with superoxide property)
Figure 6. Subcellular localization of o-DAP in: a) HeLa cancer cells, and
b) MRC-5 normal cells. The images of Tracker Red were excited by a green
light cube, which passed light through a 530(Æ20) nm band pass filter and
emission was collected through a 590 nm long pass filter. The blue images
of the compound were excited by a UV light cube, which passed light
through a 370(Æ10) nm band pass filter and emission was collected through
a 440 nm long pass filter.
[46]
kills cancer cells in H O more efficiently than in D O. Howev-
2
2
er, there still exists an ongoing discussion as to whether
TEMPO-9-ac is a predictor of specific ROS type, such as super-
oxide or hydroxyl radicals, or simply an indicator of ROS exis-
[47,48]
tence and their relative abundance.
Nevertheless, we do
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org
9713 ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Apparatus: The absorption spectra were recorded with a Thermo
Genesys 6 UV-visible spectrophotometer, and the fluorescence
spectra were recorded using a Horiba Jobin-Yvon Fluoromas-4
spectrofluorimeter with a 1 nm band pass and a 1 cm cell path
length at room temperature. The cellular and dual-staining fluores-
cence images were obtained with a Leica AF6000 fluorescence mi-
croscope with a DFC310 FX Digital color camera. Fluorescence
photographs were taken across the wavelength ranges using pho-
tomultiplier tubes (PMT). The light source used to measure the sin-
glet oxygen yield and PDT cell death was the Xenon Light Source
LAX-Cute (Asahi Spectra).
Cell culture conditions and compound incubation: The MRC-5
human embryonal lung normal fibroblast cells and HeLa human
cervical cancer cells were maintained in Modified Eagle’s Medium
(
MEM) containing non-essential amino acids, Earle’s salts, l-glut-
amine, 1 mm sodium bicarbonate, 1 mm sodium pyruvate, 1%
penicillin+streptomycin), and 10% fetal bovine serum (FBS). The
(
human lung adenocarcinoma cell lines CL1-0 and A549 cancer
lines were grown in RPMI medium containing 10% fetal bovine
serum (FBS). MCF7 human breast cancer cells and Detroit 551
human skin fibroblast normal cell lines were grown in Dulbecco’s
modified Eagle’s medium (DMEM) with non-essential amino acids
supplemented with 10% fetal calf serum (FCS). Cells were cultured
Figure 7. Variation of the absorption spectra of 10 mm TEMPO-9-ac upon
irradiation in the presence of 10 mm of o-DAP (13 measurements during
80 min). Inset: fluorescent spectra of TEMPO-9-ac (lex =360 nm). Experimen-
tal conditions are given in Figure S2 in the Supporting Information.
at 378C in a humid atmosphere of 95% air and 5% CO . Before ob-
2
not exclude the possibility that the absorption of o-DAP in
a subcellular environment is similar to that in aqueous condi-
tions. We propose that it is very possible that o-DAP can un-
dergo type II PDT under NIR illumination but undergo type I
and II PDT simultaneously under white light illumination. Most
reports generally infer that singlet oxygen formation in type II
PDT is primarily responsible for the biological PDT effect. How-
ever, several recent studies have indicated that radical species
from the type I mechanism may lead to an amplified PDT re-
serving cellular localization, cells were seeded onto coverslips and
incubated for 24 h. The next day, cells were incubated with differ-
ent concentrations of compounds for 8 h, for which the DMSO
stock solutions were diluted in serum-free medium before use
(1:100, v/v).
Light-induced cytotoxicity assay: The MRC-5 normal cells and
HeLa cancer cells were examined in the assay. Varying concentra-
tions of compounds were incubated with the cells in the dark for
8
h. Subsequently, the culture medium was removed and fresh cul-
[
49]
ture medium was added to each well. The plates were irradiated
by a light source and then incubated, overnight, at 378C. Details of
the light-induced cytotoxicity assays and the results are presented
in Table 1. All assays were carried out in triplicate and the averages
of three to five individual runs are presented.
sponse, particularly under conditions of low oxygen. Further-
more, with respect to the illumination assays in vitro, an im-
portant issue is that o-DAP is photostable, regardless of its sus-
pension in DMSO or in aqueous conditions.
Extraction and quantification of compounds in cells: HeLa cells
were seeded into 60 mmꢁ15 mm cell culture dishes. o-DAP was
added to one dish when the cells reached approximately 90% con-
fluence and was then incubated for 12 h. After incubation, all
dishes were washed with PBS and the cells were removed from
the substrate with trypsin and then pellets were prepared. Ethyl
acetate (1.8 mL) was added to extract o-DAP after the collected
Conclusion
Phenothiazine and phenothiazinium derivatives show many su-
perior optical properties. In this study we have successfully
synthesized an NIR-stable phenothiazinium derivative and
demonstrated that o-DAP may be a potential tumor-specific
NIR-absorptive photosensitizer for PDT. Following optical and
biological studies, the PDT characteristics of o-DAP have been
found to include 1) targeted selectivity for the PDT of cancer
cells, 2) photostability in DMSO and aqueous conditions (under
NIR as well as white light illumination), and 3) an ability to sup-
port extra phototoxicity through superoxide generation. This
compound is expected to become a PDT reagent in a solar
environment.
6
cells (2ꢁ10 ) had been sonicated for 1 h. Both absorbance and
fluorescence spectra of the organic phase were recorded. A similar
protocol was also applied to extract o-DAP and record the spectra
of MRC-5 normal cells; the cell number in the collected suspension
6
was consistent with that of the cancer cell (2ꢁ10 ).
Measurements: An NIR light source (20 W power-tunable Xenon
lamp that passes through a 700–1000 nm mirror module) was used
for the measurement of ROS and PDT cell death. The light from the
optional light guide and collimator lens was an average of
À2
3 mWcm on the dish surface. The singlet oxygen quantum yields
from o-DAP dissolved in DMSO were determined by a photo-
steady-state method using 1,3-diphenylisobenzofuran (DPBF) as
the scavenger. The generation of superoxide from o-DAP in an
aqueous solution was determined by a photo-steady-state method
using 4-[(9-acridinylcarbonyl)amino]-2,2,6,6-tetramethylpiperidin-1-
oxyl (TEMPO-9-ac, Invitrogen, Carlsbad, CA) as the radical trap.
Experimental Section
Materials: The chemicals employed in this study were the highest
grade available and were obtained from Acros Organic, Merck, or
Aldrich Chemical, and used without further purification. All of the
solvents were of spectrometric grade.
Dual staining of the tracker and o-DAP: The cells were first incu-
bated with o-DAP for 8 h, which was followed by incubation with
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org
9714
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
5
3
3
mm organelle probes (MitoTracker Red CMXRos for 20 min at
78C, ER-Tracker Red or LysoTracker Red DND99 for 30 min at
78C). Cells were then washed with PBS before observation. To ob-
[20] N. K. Brennan, J. P. Hall, S. R. Davies, S. O. Gollnick, A. R. Oseroff, S. L.
Gibson, R. Hilf, M. R. Detty, J. Med. Chem. 2002, 45, 5123–5135.
[
[
[
21] L. M. Williamson, R. Cardigan, C. V. Prowse, Transfusion 2003, 43, 1322–
329.
22] S. Jockusch, D. Lee, N. J. Turro, E. F. Leonard, Proc. Natl. Acad. Sci. USA
996, 93, 7446–7451.
23] G. J. Fowler, R. C. Rees, R. Devonshire, Photochem. Photobiol. 1990, 52,
89–494.
1
serve red emission of the Tracker dyes, a green light cube was
used as the excitation source, in which the light passed through
a 530(Æ20) nm band pass (bp) filter and the emission was collect-
ed through a 590 nm long pass (lp) filter. To observe blue emission
from the Tracker dyes, a UV light cube was used, in which light
passed through a 370(Æ10) nm bp filter and the emission was col-
lected through a 440 nm lp filter. Having obtained the definite
pink color image, we concluded that o-DAP was co-localized with
the probes. PDT-induced phototoxicity was determined by using
Hoechst 33342 (H-1399, Invitrogen, USA) and propidium iodide (PI,
P3566, Invitrogen, USA) under a fluorescence microscope.
1
4
[24] M. Wainwright, D. A. Phoenix, L. Marland, D. R. Wareing, F. J. Bolton,
FEMS. Immunol. Med. Microbiol. 1997, 19, 75–80.
[25] K. Buchholz, R. H. Schirmer, J. K. Eubel, M. B. Akoachere, T. Dandekar, K.
Becker, S. Gromer, Antimicrob. Agents Chemother. 2008, 52, 183–191.
[
26] M. Wainwright, N. J. Grice, E. C. Pye Lynnette, Dyes Pigm. 1999, 42, 45–
1.
5
[
27] F. Andreani, P. C. Bizzari, C. D. Casa, M. Fiarini, E. Salatelli, J. Heterocycl.
Chem. 1991, 28, 295–299.
[
[
28] S. Y. Su, H. H. Lin, C. C. Chang, J. Mater. Chem. 2010, 20, 8653–8658.
29] H.-H. Lin, C.-C. Chang, Dyes Pigm. 2009, 83, 230–236.
Acknowledgements
[30] M. Wainwright, R. M. Giddens, Dyes Pigm. 2003, 57, 245–257.
[31] K. Gollnick, A. Griesbeck, Tetrahedron 1985, 41, 2057–2068.
[32] T. L. Andrew, T. M. Swager, Macromolecules 2008, 41, 8306–8308.
[33] M. Talhavini, T. D. Z. Atvars, J. Photochem. Photobiol. A 1999, 120, 141–
This work was supported financially by the National Science
Council (NSC 101-2113M-005-016-MY3) of Taiwan.
1
49.
34] Y. Prostota, O. D. Kachkovsky, L. V. Reis, P. F. Santos, Dyes Pigm. 2013, 96,
54–562.
35] T. A. Jenny, N. J. Turro, Tetrahedron Lett. 1982, 23, 2923–2926.
[
5
Keywords: near infrared · photodamage · photosensitizers ·
phenothiazinium · reactive oxygen species
[
[36] Y. Cheng, A. C. Samia, J. Li, M. E. Kenney, A. Resnick, C. Burda, Langmuir
010, 26, 2248–2255.
2
[
[
[
[
[
[
[
37] M. H. Teiten, L. Bezdetnaya, P. Morliꢂre, R. Santus, F. Guillemin, Br. J.
Cancer 2003, 88, 146–152.
38] A. P. Castano, T. N. Demidova, M. R. Hamblin, Photodiagn. Photodyn.
Ther. 2004, 1, 279–293.
[
1] T. J. Dougherty, C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M, Kor-
belik, J. Moan, Q. Peng, J. Natl. Cancer Inst. 1998, 90, 889–805.
[
[
2] R. K. Pandey, J. Porphyrins Phthalocyanines 2000, 4, 368–373.
3] P. N. Prasad, Introduction to Biophotonics, Wiley, Hoboken, New Jersey,
39] N. L. Oleinick, R. L. Morris, I. Belichenko, Photochem. Photobiol. Sci. 2002,
2003.
1
, 1–21.
40] C. C. Chang, M. C. Hsieh, J. C. Lin, T. C. Chang, Biomaterials 2012, 33,
97–906.
41] Y. C. Lai, S. Y. Su, C. C. Chang, ACS Appl. Mater. Interfaces 2013, 5,
2935–12943.
[4] R. Weissleder, V. Ntziachristos, Nat. Med. 2003, 9, 123–128.
[5] D. E. Dolmans, D. Fukumura, R. K. Jain, Nat. Rev. Cancer 2003, 3, 380–
8
387.
[6] A. Orenstein, G. Kostenich, L. Roitman, Y. Shechtman, Y. Kopolovic, B.
1
Ehrenberg, Z. Malik, Br. J. Cancer 1996, 73, 937–944.
42] F. Belloc, P. Dumain, M. R. Boisseau, C. Jalloustre, J. Reiffers, P. Bernard, F.
Lacombe, Cytometry 1994, 17, 59–65.
43] M. E. Bulina, D. M. Chudakov, O. V. Britanova, Y. G. Yanushevich, D. B.
Staroverov, T. V. Chepurnykh, E. M. Merzlyak, M. A. Shkrob, S. Lukyanov,
K. A. Lukyanov, Nat. Biotechnol. 2006, 24, 95–99.
[
[
7] D. Kessel, J. Porphyrins Phthalocyanines 2008, 12, 877–880.
8] R. Hudson, M. Carcenac, K. Smith, L. Madden, O. J. Clarke, A. Pelegrin, J.
Greenman, R. W. Boyle, Br. J. Cancer 2005, 92, 1442–1449.
[9] M. Tijerina, P. Kopeckova, J. Kopecek, Pharm. Res. 2003, 20, 728–737.
[
10] M. R. Hamblin, E. L. Newman, J. Photochem. Photobiol. B 1994, 26, 147–
57.
11] B. Mestre, M. Pitie, C. Loup, C. Claparols, G. Pratviel, B. Meunier, Nucleic
Acids Res. 1997, 25, 1022–1027.
12] M. Wainwright, K. B. Crossley, J. Chemother. 2002, 14, 431–443.
13] K. J. Mellish, R. D. Cox, D. I. Vernon, J. Griffiths, S. B. Brown, Photochem.
Photobiol. 2002, 75, 392–397.
[
44] R. B. Vegh, K. M. Solntsev, M. K. Kuimova, S. Cho, Y. Liang, B. L. W. Loo,
L. M. Tolbert, A. S. Bommarius, Chem. Commun. 2011, 47, 4887–4889.
45] ꢃ. Hideg, Cent. Eur. J. Biol. 2008, 3, 273–284.
1
[
[
[
46] S. Pletnev, N. G. Gurskaya, N. V. Pletneva, K. A. Lukyanov, D. M. Chuda-
kov, V. I. Martynov, V. O. Popov, M. V. Kovalchuk, A. Wlodawer, Z. Dauter,
V. Pletnev, J. Biol. Chem. 2009, 284, 32028–32039.
[
[
[
[
47] D. Pagoria, A. Lee, W. Geurtsen, Biomaterials 2005, 26, 4091–4099.
48] V. J. Thannickal, B. L. Fanburg, Am. J. Physiol. Lung Cell Mol. Physiol.
[
14] M. Wainwright, D. A. Phoenix, L. Rice, S. M. Burrow, J. Waring, J. Photo-
chem. Photobiol. B 1997, 40, 233–239.
2
000, 279, L1005–L1028.
[
[
[
15] K. Orth, A. Ruck, A. Stanescu, H. G. Berger, Lancet 1995, 345, 519–520.
16] Z. Tong, G. Singh, A. J. Rainbow, Photochem. Photobiol. 2001, 74, 707.
17] A. E. Frimberger, A. S. Moore, L. Cincotta, S. M. Cotter, J. R. Foley, Clin.
Cancer Res. 1998, 4, 2207–2218.
[
49] E. F. Silva, C. Serpa, J. M. Dabrowski, C. J. Monteiro, S. J. Formosinho, G.
Stochel, K. Urbanska, S. Simoes, M. M. Pereira, L. G. Arnaut, Chem. Eur. J.
2
010, 16, 9273–9286.
[18] L. Cincotta, J. W. Foley, T. Maceachern, E. Lampros, A. H. Cincotta, Cancer
Res. 1994, 54, 1249–1258.
Received: February 21, 2014
[19] M. R. Detty, P. B. Merkel, J. Am. Chem. Soc. 1990, 112, 3845–3855.
Published online on July 2, 2014
Chem. Eur. J. 2014, 20, 9709 – 9715
www.chemeurj.org
9715
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim