Silylation improves the photodynamic activity of tetraphenylporphyrin derivatives in vitro and in vivo

Article abstract of DOI:10.1002/chem.201303120

The effects of silyl and hydrophilic groups on the photodynamic properties of tetraphenylporphyrin (TPP) derivatives have been studied in vitro and in vivo. Silylation led to an improvement in the quantum yield of singlet oxygen sensitization for both sulfo and carboxy derivatives, although the silylation did not affect other photophysical properties. Silylation also improved the cellular uptake efficiency for both sulfo and carboxy derivatives, enhancing the in vitro photodynamic activity of the photosensitizer in U251 human glioma cells. The carboxy derivative (SiTPPC₄) was found to show higher cellular uptake efficiency and in vitro photodynamic activity than the corresponding sulfo derivative (SiTPPS₄), which indicates that the carboxy group is a more promising hydrophilic group than the sulfo group in the silylated porphyrin. SiTPPC₄ was found to show high selective accumulation efficiency in tumors, although almost no tumor selectivity was observed for the nonsilylated porphyrin. The concentration of SiTPPC₄ in tumors was 13 times higher than that in muscle 12 h after drug administration. We also studied tumor response after treatment and found that silylation enhanced in vivo photodynamic activity significantly. SiTPPC ₄ shows higher photodynamic activity than NPe6 with white light irradiation. Improved photosensitizers: Silylation improves the quantum yield of singlet oxygen sensitization, cellular uptake efficiency, and selective accumulation efficiency in tumors. As a result of these improvements, silylation significantly enhances photodynamic activity (see figure). The results of this work suggest that silylation is a promising strategy for improving photosensitizers for photodynamic therapy.

Full text of DOI:10.1002/chem.201303120

DOI: 10.1002/chem.201303120
Full Paper
&
Photodynamic Therapy
Silylation Improves the Photodynamic Activity of
Tetraphenylporphyrin Derivatives In Vitro and In Vivo
Hiroaki Horiuchi,*[a] Masahiro Hosaka,*[b] Hiroyuki Mashio,[a] Motoki Terata,[b]
Shintaro Ishida,[c] Soichiro Kyushin,[a] Tetsuo Okutsu,[a] Toshiyuki Takeuchi,[d] and
Hiroshi Hiratsuka[a]
Abstract: The effects of silyl and hydrophilic groups on the
photodynamic properties of tetraphenylporphyrin (TPP) de-
rivatives have been studied in vitro and in vivo. Silylation led
to an improvement in the quantum yield of singlet oxygen
sensitization for both sulfo and carboxy derivatives, although
the silylation did not affect other photophysical properties.
Silylation also improved the cellular uptake efficiency for
both sulfo and carboxy derivatives, enhancing the in vitro
photodynamic activity of the photosensitizer in U251 human
glioma cells. The carboxy derivative (SiTPPC4) was found to
show higher cellular uptake efficiency and in vitro photody-
namic activity than the corresponding sulfo derivative
(SiTPPS4), which indicates that the carboxy group is a more
promising hydrophilic group than the sulfo group in the sily-
lated porphyrin. SiTPPC4 was found to show high selective
accumulation efficiency in tumors, although almost no
tumor selectivity was observed for the nonsilylated porphy-
rin. The concentration of SiTPPC4 in tumors was 13 times
higher than that in muscle 12 h after drug administration.
We also studied tumor response after treatment and found
that silylation enhanced in vivo photodynamic activity signif-
icantly. SiTPPC4 shows higher photodynamic activity than
NPe6 with white light irradiation.
Introduction
as radiotherapy, surgery, or chemotherapy, and can improve
the quality of life of the patient. However, PDT is a minor
cancer therapy at present because of the low efficiencies of
photosensitizers, and consequently the photosensitizers need
to be improved.
Photodynamic therapy (PDT) is a promising treatment for a va-
riety of cancers, including bladder, cervical, lung, head, neck,
esophageal, and gastric cancers. In PDT, three individually non-
toxic components (photosensitizer, light, and molecular
oxygen) are combined only in tumor tissue to produce a reac-
tive oxygen species (ROS), in particular, singlet oxygen. Thus,
PDT is very different from conventional cancer therapies, such
To increase the effectiveness of photosensitizers, the follow-
ing pharmacological and photochemical efficiencies should be
improved: 1) The selective accumulation efficiency in tumors,
which is based on the preferential uptake efficiency by the
tumor, cellular uptake efficiency, and clearance efficiency from
normal tissue, 2) the light absorption efficiency from the red to
near-IR region, and 3) the quantum yield of the singlet oxygen
sensitization FD. To improve tumor selectivity, photosensitizers
have been conjugated to antibodies,[1] peptides,[2] sugar
units,[3] and nanoparticles,[4] and to improve cellular uptake effi-
ciency, folic acid,[5] sugar units,[3] and nanoparticles[6] have also
been introduced. To enhance light absorption efficiency, phtha-
locyanines,[2,7,8] chlorins,[9] and N-confused porphyrins[10] have
been studied. Finally, metal complexes of porphyrinoids[11]
have been studied to improve FD.
[a] Prof. H. Horiuchi, H. Mashio, Prof. S. Kyushin, Prof. T. Okutsu,
Prof. H. Hiratsuka
Division of Molecular Science
and International Education and Research Center for Silicon Science
Faculty of Science and Technology, Gunma University, Kiryu (Japan)
Fax: (+81)277-30-1244
E-mail: hhoriuchi@gunma-u.ac.jp
[b] Prof. M. Hosaka, M. Terata
Division of Molecular and Cellular Biology
Department of Biotechnology
Akita Prefectural University, Akita (Japan)
Fax: (+81)18-872-1676
Silicon is considered to have potential for improving phar-
macological and/or photochemical efficiencies. Bains and Tacke
have widely studied the effect of carbon-to-silicon switching in
previously known drugs and reported an enhancement in drug
activity,[12] and West and co-worker reported that the anticanc-
er activity of indomethacin is improved by introducing a silyl
group.[13] The silicon atom is also reported to improve the pho-
tophysical and -chemical properties of aromatic molecules. The
introduction of silyl group(s) into aromatic molecules induces
E-mail: mhosaka@akita-pu.ac.jp
[c] Prof. S. Ishida
Department of Chemistry, Graduate School of Science
Tohoku University, Sendai (Japan)
[d] Prof. T. Takeuchi
Department of Molecular Medicine
Institute for Molecular and Cellular Regulation
Gunma University, Maebashi (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201303120.
Chem. Eur. J. 2014, 20, 6054 – 6060
6054
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a redshift of their absorption bands and enhancement of their
molar absorption coefficients.[14,15] Silylation also leads to an in-
crease in the quantum yields of fluorescence[14] and phosphor-
escence.[15,16]
the in vivo photodynamic activity of tetraphenylporphyrin de-
rivatives by silylation.
In the case of photosensitizing drugs for PDT, silicon has
been introduced as a metal center in porphyrin or phthalocya-
nine to allow the introduction of substituents perpendicular to
the aromatic ring.[8,17] We introduced silyl groups on to the
outside of porphyrin (the phenyl moiety of tetraphenylpor-
phyrin) and found that the silylation enhanced FD.[18] We have
also studied the effect of the silylation of water-soluble por-
phyrins bearing sulfo groups (SiTPPS4 and TPPS4, Figure 1) on
Results and Discussion
Photophysical and photochemical properties
First, we synthesized the 5,10,15,20-tetrakis(3-carboxyphenyl)-
porphyrin sodium salt (TPPC4) and its silylated derivative
(SiTPPC4), the chemical structures of which are shown in
Figure 1. Figure 2a shows the UV/Vis absorption and fluores-
Figure 2. UV/Vis absorption (Abs; full lines) and fluorescence spectra (Flu;
broken lines) of SiTPPC4 (red), TPPC4 (blue), SiTPPS4 (green), and TPPS4
(black) in ethanol.[19]
cence spectra of SiTPPC4 and TPPC4. The spectra of SiTPPC4 are
essentially the same as those of TPPC4. In our previous paper,
we reported the photophysical and -chemical properties of the
corresponding sulfo derivatives, SiTPPS4 and TPPS4
(Figure 1),[19] and their absorption and fluorescence spectra are
essentially the same as those of the carboxy derivatives
(Figure 2). The fluorescence quantum yields and lifetimes of
SiTPPC4 and TPPC4 were determined and are presented in
Table 1 together with those of SiTPPS4 and TPPS4.[19] The fluo-
Figure 1. Molecular structures of photosensitizers.
pharmacological properties using human cancer cell lines and
found that silylation enhances not only FD but also cellular
uptake efficiency.[19] As a result of these enhancements, we suc-
ceeded in improving significantly the in vitro photodynamic
activity. The enhancement of cellular uptake efficiency by sily-
lation was explained by the increase in lipophilicity. The car-
boxy group is known to be a weaker hydrophilic group than
the sulfo group, and it has been used in many kinds of photo-
sensitizers for PDT, including clinically applied photosensitizers
(Photofrin and NPe6). We expected that the lipophilic property
of a photosensitizer could be improved by replacing the sulfo
group with a carboxy group to enhance cellular uptake effi-
ciency and photodynamic activity. In this paper we describe
the synthesis of carboxylated tetraphenylporphyrin derivatives
with and without silyl groups (SiTPPC4 and TPPC4, Figure 1)
and a study of the effect of both the silyl and hydrophilic
groups on the photochemical and pharmacological efficiencies.
We succeeded in improving FD, cellular uptake efficiency, and
in vitro photodynamic activity by replacing the sulfo group
with the carboxy group. We also carried out in vivo studies on
tumor-bearing nude mice and found that the silylation im-
proved the selective accumulating efficiency in tumors. Thus,
on the basis of these results, we have succeeded in enhancing
Table 1. Photophysical parameters of the photosensitizers in ethanol.
Sample
Ff
tf [ns]
FD
SiTPPC4
TPPC4
0.068
0.059
0.059
0.057
10.6
10.5
12.6
12.3
0.72
0.56
0.66
0.58
[a]
SiTPPS4
[a]
TPPS4
[a] From ref. [19].
rescence quantum yields (Ff) and fluorescence lifetimes (tf) of
the four photosensitizers are also essentially the same. These
results indicate that the excited-state properties of these pho-
tosensitizers are mostly independent of both the silyl and hy-
drophilic substituents.
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6055
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Singlet oxygen is a key species for inducing cell death. We
previously reported that silylation enhances the quantum yield
of singlet oxygen sensitization (FD) of tetraphenylporphyrin
derivatives.[18,19] The values of FD for the carboxy derivatives
were also determined by comparing the near-IR phosphores-
cence intensity of singlet oxygen and were found to be 0.72
and 0.56 for SiTPPC4 and TPPC4, respectively. This indicates
that the value of FD for TPPC4 improved as a result of silyla-
tion, as was also the case for tetraphenylporphyrin (TPP)[18] and
TPPS4.[19] In the case of the silylation of TPP and TPPS4, the in-
crease in FD was not explained by the change in the quantum
yield of intersystem crossing, but by an improvement in the
fraction of triplet sensitizer quenched by oxygen to yield sin-
tosensitizers for 12 h. The fluorescence intensity observed for
SiTPPC4 is stronger than that of TPPC4, as is the case for the
sulfo derivatives. We have previously reported for the sulfo de-
rivatives that silylation improves the cellular uptake efficien-
cy.[19] Because the fluorescence quantum yields of SiTPPC4 and
TPPC4 are similar, the intense fluorescence of SiTPPC4 relative
to that of TPPC4 would appear to reflect the higher cellular
uptake of the silylated derivative, as in the case of the sulfo de-
rivatives. Thus, we evaluated the cellular uptake efficiency for
the carboxy derivatives. The U251 cells were cultured with
15 mm photosensitizers for 12 h, and then the photosensitizers
were extracted from the cells. The relative cellular uptake effi-
ciencies (UErel) were estimated by comparing the fluorescence
intensities of the extracted solutions. Note that the fluores-
cence intensity was corrected for the total amount of proteins
in the cells determined by the Bradford method to consider
the difference in the number of U251 cells. These UE values
are normalized relative to that of TPPS4 (Figure 3). As in the
case of the sulfo derivatives, silylation improves the cellular
uptake efficiency of the carboxy derivative. The partition coeffi-
cients (logP) for the photosensitizers in 1-octanol/Tris HCl
buffer (pH 7.6) were estimated to be 2.34, 0.79, 1.95, and 0.30
for SiTPPC4, TPPC4, SiTPPS4, and TPPS4, respectively. The values
of UErel decrease in the same order as the values of logP. This
indicates that the increase in lipophilicity as a result of silyla-
tion is one of the reasons for the increase in the cellular
uptake efficiency.[19,20]
T
glet oxygen (fD ). In the case of the carboxy derivative, silyla-
tion did not change the excited-state properties, as in the case
of TPP and TPPS4. Thus, the same mechanism can be proposed
for the carboxy derivative, that is, the increase in FD by the si-
T
lylation of TPPC4 can be explained by the improvement in fD .
It can be noted that the value of FD for SiTPPC4 is greater
than that of SiTPPS4, although the values of FD for the nonsily-
lated derivatives TPPS4 and TPPC4 are similar.
Photodynamic properties of photosensitizers for cancer cells
To evaluate the photodynamic properties, cell culture studies
were carried out by using the U251 human glioma cell line
(U251). First, the cytotoxicity was evaluated. It has been report-
ed that the silylation of TPPS4 leads to an increase in cytotoxic-
ity due to an increase in cellular uptake efficiency.[19] As in the
case of TPPS4, the silylation of TPPC4 also leads to an increase
in cytotoxicity. This is also due to an increase in the cellular
uptake efficiency, as described below. Below 20 mm, no photo-
sensitizer showed any cytotoxicity. Figure 3a–d show the fluo-
rescence images of U251 cells cultured with 15 mm of the pho-
The photocytotoxicity was also examined. Again, the U251
cells were cultured with 15 mm photosensitizers for 12 h. After
washing the cells with phosphate-buffered saline (PBS), we
then irradiated the cells with visible light. It has been reported
that the silylation of TPPS4 enhances the photocytotoxicity at
a concentration of 25 mm.[19] SiTPPS4 also shows higher photo-
cytotoxicity than TPPS4 at 15 mm (Figure 4). The change in cell
survival with increasing radiation dose is negligibly small for
TPPC4, but a significant decrease in cell survival was observed
for SiTPPC4. Thus, an increase in photocytotoxicity caused by
silylation was observed for the carboxy derivative, as in the
case of the sulfo derivative. This is due to the improvements in
FD and cellular uptake efficiency. Note that the photocytotox-
Figure 3. Fluorescence images of U251 cells cultured with 15 mm of
a) SiTPPC4 and b) TPPC4 for 12 h in comparison with those of c) SiTPPS4 and
d) TPPS4.[19] e) Relative cellular uptake efficiency (UErel) of the photosensitizers
in U251 cells. Scale bars: 15 mm.
Figure 4. Survival of U251 cells cultured with 15 mm of SiTPPC4, TPPC4,
SiTPPS4, and TPPS4 for 12 h as a function of the irradiation light dose.
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6056
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
icity of SiTPPC4 is greater than that of SiTPPS4, and therefore
we have succeeded in improving the silyl photosensitizer by
changing the hydrophilic groups.
that of TPPC4. This indicates that silylation also improves the
selective accumulation efficiency. The tissue concentration of
SiTPPC4 was studied to evaluate the tissue distribution quanti-
tatively. SiTPPC4 was extracted from each tissue and analyzed
by fluorescence measurements (Table 2). The SiTPPC4 concen-
tration in blood decreased quickly and was almost zero at 48 h
after administration. However, with the decrease in the con-
centration in blood, the concentration of SiTPPC4 in other tis-
Photodynamic properties for tumor-bearing mice
Because the silylation of tetraphenylporphyrin improves the
photodynamic activity in cells, we addressed the photodynam-
ic activity of silylated tetraphenylporphyrin in
vivo by using tumor-bearing mice. Tumor
Table 2. Tissue concentration after the injection of 50 nmol SiTPPC4 [nmolgÀ1 of tissue].
transplants were established in female nude
mice by injection of mouse oral squamous
Tissue
Time after drug administration [h]
12 24
carcinoma cells (SCC-7). SiTPPC4 is easily dis-
1
4
6
48
96
solved in pure water as the monomer, but it
blood 35.7Æ8.5
tumor
11.2Æ2.4
4.72Æ0.81 2.76Æ0.36 0.23Æ0.10 0.03Æ0.05 0.00Æ0.00
forms aggregates in buffer solution (pHꢀ7)
0.75Æ0.13 1.51Æ0.13 1.61Æ0.27 2.56Æ0.39 1.29Æ0.33 1.49Æ0.89 1.53Æ0.05
muscle 0.11Æ0.05 0.17Æ0.09 0.24Æ0.02 0.20Æ0.03 0.21Æ0.05 0.15Æ0.08 0.17Æ0.03
liver 11.3Æ0.7 11.2Æ1.1 8.11Æ3.44 6.82Æ3.43
6.57Æ0.81 8.84Æ0.37 11.5Æ0.9
kidney 4.23Æ0.38 2.86Æ0.69 1.98Æ0.25 2.21Æ0.94 1.58Æ0.57 1.47Æ0.18 2.03Æ0.65
due to the low acidity of the carboxy group.
However, SiTPPC4 is readily taken up as the
monomer by albumin, and thus bovine
serum albumin (BSA) was added to the solu-
tions of photosensitizers to avoid aggrega-
sues increased, with the highest concentration observed at 6–
12 h after administration, except in the kidney. The concentra-
tion in the liver was higher than in the kidney and tumor.
These results show a similar trend to that found with NPe6.[21]
The concentrations in the tumor tissue were always higher
than that in muscle, and the highest ratio (13 times) was ob-
tained at 12 h after administration.
tion. The photosensitizers were administered intravenously
through the tail of tumor-bearing nude mice. Unexpectedly,
the mice died within 1 day after the administration of 14 nmol
of SiTTPS4. Although the cytotoxicity of SiTPPS4 is sufficiently
low (50 mm) for U251 cells, it is clear that it is not applicable
for drug use. On the other hand, mice were still alive 6 days
after the administration of 100 nmol of SiTPPC4, although its
cytotoxicity is higher than that of SiTPPS4. Thus we studied
SiTPPC4 and TPPC4 in vivo.
In the case of PDT treatment using NPe6, light irradiation is
generally performed 4–6 h after drug administration. Thus, we
evaluated tissue distribution at 4 h after administration. The
concentration of SiTPPC4 in the tumor was 1.51 nmolgÀ1 and
was nine times higher than that in muscle. In the case of
NPe6, the ratio of the concentration in the tumor to that in
muscle was reported to be 7.4 at 4 h after administration,[22]
which indicates that the tumor selectivity of SiTPPC4 is as
strong as that of NPe6. We administered 50 nmol of SiTPPC4
and the averaged tissue weight at 4 h was 0.96Æ0.21 g, thus
3% of the injected SiTPPC4 is estimated to have been retained
in the tumor at 4 h. To improve tumor selectivity and accumu-
lation efficiency, many kinds of drug delivery systems (DDS)
have been used. Mitsunaga et al. studied the tumor selectivity
of the conjugate of the antibody panitumumab and photosen-
sitizer by a fluorescence imaging technique, and the ratio of
the fluorescence intensity of the tumor with respect to the
background was about 7 at 4 days after the injection.[1] Pep-
tides are also expected to improve tumor selectivity. Ke et al.
studied the tumor retention properties of the conjugate of the
photosensitizer and peptide, and reported that the ratio of the
fluorescence intensity of the photosensitizer in the tumor with
respect to skin was about 10.[2] It is difficult to compare the
tumor selectivities of SiTPPC4 and the reported systems, but
they seem to be similar.
The localization of the photosensitizers was studied by an in
vivo fluorescence imaging technique. To compare the fluores-
cence intensities, the fluorescence images of mice were mea-
sured with microtubes containing 5 mm of photosensitizer solu-
tions as references (Figure 5). In the case of TPPC4, almost no
fluorescence was observed at 4 h after drug administration. On
the other hand, strong fluorescence was observed from the
tumor tissue in the case of SiTPPC4, although the fluorescence
intensity of SiTPPC4 from normal tissue was also stronger than
Figure 5. In vivo fluorescence imaging with a) SiTPPC4 and b) TPPC4 ob-
served at 4 h after the injection of photosensitizers. Red: fluorescence of
photosensitizers; cyan: autofluorescence from mice. Tumor-bearing athymic
nude mice were injected with 50 nmol photosensitizers+3 mg BSA/200 mL
saline through the tail vein. Images were obtained with the reference micro-
tubes (5 mm photosensitizers + 3 gLÀ1 BSA in saline).
Next, PDT-induced changes in tumor volume were studied.
To address the effect of the interval between drug administra-
tion and light irradiation for SiTPPC4, we performed PDT on the
tumor at different times and periodically measured tumor size
(see Figure S1 in the Supporting Information). Although the
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6057
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
concentration of SiTTPC4 in the tumor was highest at 12 h
after the injection (see Table 2), the tumor responses to
SiTTPC4 did not depend on the time of irradiation after drug
administration (1–24 h). This finding is unusual and it is difficult
to fully explain these results at present. A photosensitizer in
blood is also known to show a PDT effect,[22] which indicates
that the PDT effect in the tumor is induced by photosensitizers
in both the tumor and blood in the tumor. At 1 h, the concen-
tration of SiTPPC4 is low in the tumor but high in the blood,
and, with the lapse of time, the concentration increased in the
tumor and decreased in the blood (Table 2). Thus, the inde-
pendency of the PDT effect with respect to time may be ex-
plained by the PDT effect induced by SiTPPC4 in blood. As de-
scribed above, light irradiation is performed 4–6 h after drug
administration in the case of the PDT treatment using NPe6.
Thus, we compared the PDT activity at the interval of 4 h
(Figure 6). After PDT treatment using TPPC4, the tumor volume
singlet oxygen sensitization, cellular uptake efficiency, and se-
lective accumulation efficiency in tumors. Carboxy and sulfo
derivatives of tetraphenylporphyrins have been studied to clar-
ify the effect of the hydrophilic group on photodynamic prop-
erties, with the carboxy group showing higher cellular uptake
efficiency and in vitro photodynamic activity. Tumor selectivity
of the silylated porphyrin (SiTPPC4) is a factor of about 13
higher than in muscle at 12 h after drug administration. By
using tumor-bearing mice, we demonstrated that silylation en-
hances in vivo photodynamic activity significantly. For white
light irradiation, the photodynamic efficiency of SiTPPC4 is
much higher than that of NPe6, which is one of the most effi-
cient photosensitizers reported to date. Thus, we propose that
silylation is a promising strategy for improving photosensitizers
for photodynamic therapy.
Experimental Section
Synthesis
The reactions were carried out under an argon atmosphere. Diethyl
ether was distilled from sodium. Tetrahydrofuran (dehydrated) and
toluene were purchased from Wako Pure Industries Chemical Co.
and used as received. The 5,10,15,20-tetrakis(3’-trimethylsilyl-5’-car-
boxyphenyl)porphyrin tetrasodium (SiTPPC4) and 5,10,15,20-tetra-
kis(3’-carboxyphenyl)porphyrin tetrasodium (TPPC4) salts were pre-
pared by following the procedures described in the Supporting In-
formation. The 1H NMR spectra were recorded by using a JEOL
JNM-AL300 spectrometer.
Photophysical and photochemical measurements
Absorption spectra were recorded on a Hitachi U3310 spectropho-
tometer and the fluorescence emission and excitation spectra were
measured by using a Hitachi F4500 fluorescence spectrometer.
Fluorescence quantum yields were determined by using an Abso-
lute PL Quantum Yield Measurement System (Hamamatsu C9920-
02). Fluorescence lifetimes were measured by using an Edinburgh
Analytical Instruments FL900CDT Spectrometer (H2pulser, pulse
width 0.8 ns, 108 photonspulseÀ1, repetition rate 40 kHz). Nanosec-
ond transient absorption spectra were recorded by using a Unisoku
TSP601H nanosecond laser photolysis system with a Nd3+:YAG
laser (Tokyo Instruments Lotis II, 355 nm, 2.0 mJpulseÀ1, pulse
width 8 ns, 10 Hz). For the phosphorescence measurements of the
singlet oxygen, 355 nm light from a Nd3+:YAG laser (Tokyo Instru-
ments Lotis II, 1.0 mJpulseÀ1, pulse width 8 ns, 10 Hz) was used as
the excitation light source. Phosphorescence was detected with
a photomultiplier tube for the NIR region (Hamamatsu R5509-42)
and cooled at À808C after dispersion with a monochromator
(Ritsu MC-10N, blaze wavelength 1250 nm and slit width 0.7 mm).
Signals from the photomultiplier tube were amplified by a factor
of five with a DC-300 MHz amplifier (Stanford Research Systems
SR445) and processed with a gated photon counter (Stanford Re-
search Systems SR400). The gate width and delay of the photon
counter were set at 0.5 and 0.5 ms, respectively, and data acquisi-
tion was ꢁ300.
Figure 6. Tumor volume after PDT treatment with SiTPPC4 (filled circles),
TPPC4 (filled squares), and NPe6 (filled triangles). Tumor-bearing athymic
nude mice were injected with 100 nmol photosensitizers + 3 mg BSA/
200 mL saline through the tail vein. The tumor was irradiated with visible
light 4 h after drug administration. As a control, the tumor was irradiated
with visible light without the administration of photosensitizer (open
circles).
increased as well as that of the control, which indicates that
the PDT activity of TPPC4 is negligibly small. In the case of
SiTPPC4, the tumor volume decreased with elapsed time and
the tumor disappeared 6 days after the treatment. Thus, we
can conclude that silylation enhances the in vivo PDT activity
significantly and that it is a promising strategy for improving
photosensitizers effectively for photodynamic therapy. At pres-
ent, NPe6 is regarded as a highly promising photosensitizer for
PDT, and thus we also compared the PDT activity of SiTPPC4
and NPe6. The same PDT treatment was carried out by using
100 nmol of NPe6. Under this condition, the tumor volume did
not change, which suggests that SiTPPC4 shows higher PDT ac-
tivity than NPe6.
Cell culture conditions
Conclusion
Human glioma cells (U251) were grown in a 64 cm2 plastic tissue
culture dish in Dulbecco’s Modified Eagle medium (DMEM) supple-
mented with 10% fetal bovine serum (FBS) solution and a 1% so-
Although silylation is one of the simplest modifications in or-
ganic chemistry, it additionally improves the quantum yield of
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6058
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
lution of 2 mmLÀ1 glutamine, 100 unitsmLÀ1 penicillin, and
100 mgmLÀ1 streptomycin in a 5% CO2 atmosphere at 378C.
Mouse oral squamous carcinoma cells (SCC-7) were cultured in
DMEM with 10% FBS. Cells were subcultured by dispersal with
aqueous layer was recorded. The maximum absorbance of this so-
lution at around 415 nm (Soret band) was defined as Aw. The parti-
tion coefficient (log P) was estimated by using the equation
log P=log co/cw =log [(A0ÀAw)/Aw], in which co and cw are the con-
centrations of the photosensitizer in 1-octanol and water,
respectively.
0.25% trypsin and seeded at 5ꢁ104 cellsmLÀ1
.
Cell survival measurements
Tumor transplantation to nude mice
Cell survival was measured by using the 3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium (MTT) assay. Before performing growth
inhibition assays, we examined the linearity of the MTT assay by in-
We conducted our animal experiments in accordance with the
guidelines for the Care and Use of Laboratory Animals of the Medi-
cal Research Council of Gunma University. Tumor transplants were
established at the lower thigh in female athymic nude mice by in-
jection of 5ꢁ106 cells of SCC-7. Experiments with tumor-bearing
mice were performed 2 weeks after the injection of tumor cells.
creasing the number of U251 cells plated from
0 to 2ꢁ
105 cellsmLÀ1; a good linear relationship was obtained. U251 cells
were seeded at an initial density of 1ꢁ106 cellsmLÀ1 in 96-well mi-
crotitration plates. Forty-eight hours after plating, the cells were
exposed to photosensitizers in Roswell Park Memorial Institute
(RPMI) medium supplemented with 9% bovine serum albumin
(BSA). After 12 h incubation at 378C, the medium was removed,
the cells were washed with phosphate-buffered saline (PBS), and
fresh supplemented DMEM was added. In the case of the photocy-
totoxicity experiments, samples were irradiated with visible light.
After 24 h incubation at 378C, the MTT assay was carried out. The
absorbance was measured by using a microplate reader (Corona
Electric Co., Ltd. MTP500). As a light source, output from a 500 W
Xe short-arc lamp was used after passing through a long-pass filter
(390 nm. Sigma Koki Co. Ltd., SCF-50S-39 L). The light intensity was
estimated to be 28.1 mWcmÀ2 in the wavelength region of 380–
800 nm by using an Ushio Spectro-Reflectance Meter (USR-45V/D).
Intravenous injection of photosensitizers
An equimolar amount of BSA was added to the saline solution of
a photosensitizer. This mixture of photosensitizer and BSA (200 mL)
was injected through the tail vein.
In vivo fluorescence imaging
Luminescence close-up images of mice were obtained by using
the Maestro FL500 in vivo imaging system (CRi, Inc.). A band-pass
filter from 503 to 548 nm and a long-pass filter over 560 nm were
used for excitation and emission light, respectively. The tunable
filter was automatically stepped in increments of 10 nm from 630
to 800 nm, and the camera captured images at each wavelength
interval with a 120 ms exposure. Spectral fluorescence images con-
sisting of autofluorescence and fluorescence spectra of the photo-
sensitizers were obtained and then unmixed, based on their spec-
tral patterns using commercial software (Maestro software, CRi,
Inc.). The spectra for unmixed images are shown in Figure 5 (sky-
blue represents autofluorescence images; red represents the fluo-
rescence of photosensitizers).
Fluorescence microscopy
U251 cells (3ꢁ104 cellsmLÀ1) were cultured in a plastic chamber
slide (0.6 cm2/chamber) and cultured for 48 h for proper attach-
ment to the substratum. The U251 cells were then exposed to
photosensitizers (25 mm) in RPMI supplemented with 9% BSA for
1 to 12 h. For the dual staining experiments, specific organelle
markers in RPMI were added at 11.5 h of the exposure time and
cultured for a further 0.5 h. After incubation with the photosensi-
tizers, the cells were washed with PBS. Fluorescence images were
observed by using a fluorescence microscope (Olympus BX50). Ex-
citation wavelengths of around 535 and 475 nm were selected to
excite photosensitizers and specific organelle markers, respectively.
Tissue concentration
To address the tissue distribution of photosensitizers, tumor-bear-
ing mice were anesthetized with sodium pentobarbital (100 mggÀ1
body weight) after injection and the blood was washed out with
PBS (10 mL) by cardiac perfusion. Isolated organs were homogen-
ized with lysis buffer (50 mm Hepes/NaOH, 2 mm EDTA, 100 mm
NaCl, 0.1% Triton X-100). The homogenized tissues were centri-
fuged at 11000g for 15 min at 48C and then divided into superna-
tant and pellet. The supernatant fraction was diluted with ethanol,
and then the fluorescence intensity of each fraction was measured
by using F-4500 fluorescence spectrometer (Hitachi). The excitation
wavelength was 417 nm and the emission wavelength was
650 nm.
Cellular uptake efficiency
U251 cells (3ꢁ104 cellsmLÀ1) were inoculated in a 6-well plate and
cultured for 48 h for proper attachment to the substratum. The
U251 cells were then exposed to photosensitizers (25 mm) in RPMI
supplemented with 9% BSA for 1 to 12 h. After exposure, the cells
were washed with PBS, and then PBS (1 mL) and 5m KOH (20 mL)
were added to extract the photosensitizer from the cells. The fluo-
rescence spectrum of the extracted solution was recorded by
using a Hitachi F-2500 fluorescence spectrophotometer, and the
relative cellular uptake efficiency was estimated from the fluores-
cence intensity.
In vivo photodynamic activity
Tumor-bearing mice were anesthetized 4 h after the injection of
photosensitizers. The tumor tissue was irradiated for 24 min by
using a 500 W Xe lamp (USHIO UXL-500D) with a 390 nm cut-off
filter and water filter (1.65 JcmÀ2). The sizes of the tumors on the
mice were measured with a micrometer caliper every 24 h. The di-
ameters of the tumors were obtained by subtracting the thickness
of skin. The volumes of the tumors were calculated as a half ellip-
Partition coefficient
A solution of the photosensitizer (10À5 m) in Tris-HCl buffer (pH 7.6)
was prepared and its absorption spectrum was recorded. The maxi-
mum absorbance of this solution at around 415 nm (Soret band)
was defined as A0. An equal amount of 1-octanol was added to
this aqueous solution and mixed. After phase separation between
the aqueous and 1-octanol layers, the absorption spectrum of the
soid by using the equation V=4= abc/2 in which a is the long
3
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6059
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[8] X. J. Jiang, P. C. Lo, Y. M. Tsang, S. L. Yeung, W. P. Fong, D. K. P. Ng, Chem.
Eur. J. 2010, 16, 4777–4783.
radius of a tumor, b is the short radius of a tumor, and c is the
height of a tumor.
[9] a) S. Banfi, E. Caruso, S. Caprioli, L. Mazzagatti, G. Canti, R. Ravizza, M.
Gariboldi, E. Monti, Bioorg. Med. Chem. 2004, 12, 4853–4860; b) J. Gravi-
er, R. Schneider, C. Frochot, T. Bastogne, F. Schmitt, J. Didelon, F. Guille-
min, M. Barberi-Heyob, J. Med. Chem. 2008, 51, 3867–3877; c) A. M. G.
Silva, A. C. Tome, M. Neves, A. M. S. Silva, J. A. S. Cavaleiro, J. Org. Chem.
2005, 70, 2306–2314.
[10] J. P. Belair, C. J. Ziegler, C. S. Rajesh, D. A. Modarelli, J. Phys. Chem. A
2002, 106, 6445–6451.
[11] I. Scalise, E. N. Durantini, J. Photochem. Photobiol. A 2004, 162, 105–113.
[12] W. Bains, R. Tacke, Curr. Opin. Drug Discov. Dev. 2003, 6, 526–543.
[13] G. Bikzhanova, I. Toulokhonova, S. Gately, R. West, Silicon Chem. 2007, 3,
209–217.
Acknowledgements
This work was partly supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture Sports, Sci-
ence, and Technology (MEXT) of the Japanese government and
by the Japan Science and Technology Agency, the Element In-
novation Project of Gunma University, and Akita Prefectural
University President’s Research Project Fund.
[14] a) S. Kyushin, M. Ikarugi, M. Goto, H. Hiratsuka, H. Matsumoto, Organo-
metallics 1996, 15, 1067–1070; b) S. Kyushin, Y. Ishikita, H. Matsumoto,
H. Horiuchi, H. Hiratsuka, Chem. Lett. 2006, 35, 64–65.
[15] S. Kyushin, N. Takemasa, H. Matsumoto, H. Horiuchi, H. Hiratsuka, Chem.
Lett. 2003, 32, 1048–1049.
Keywords: biological activity · fluorescence · photodynamic
therapy · porphyrinoids · silylation
[16] S. Kyushin, T. Kitahara, H. Matsumoto, Chem. Lett. 1998, 471–472.
[17] a) P. C. Lo, J. D. Huang, D. Y. Y. Cheng, E. Y. M. Chan, W. P. Fong, W. H. Ko,
D. K. P. Ng, Chem. Eur. J. 2004, 10, 4831–4838; M. Nishida, H. Horiuchi,
A. Momotake, Y. Nishimura, H. Hiratsuka, T. Arai, J. Porphyrins Phthalo-
cyanines 2011, 15, 47–53.
[18] H. Horiuchi, T. Tanaka, K. Yoshimura, K. Sato, S. Kyushin, H. Matsumoto,
H. Hiratsuka, Chem. Lett. 2006, 35, 662–663.
[19] H. Horiuchi, T. Kameya, M. Hosaka, K. Yoshimura, S. Kyushin, H. Matsu-
moto, T. Okutsu, T. Takeuchi, H. Hiratsuka, J. Photochem. Photobiol. A
2011, 221, 98–104.
[1] M. Mitsunaga, M. Ogawa, N. Kosaka, L. T. Rosenblum, P. L. Choyke, H.
Kobayashi, Nat. Med. 2011, 17, 1685–U1210.
[2] M. R. Ke, S. L. Yeung, W. P. Fong, D. K. P. Ng, P. C. Lo, Chem. Eur. J. 2012,
18, 4225–4233.
[3] B. Di Stasio, C. Frochot, D. Dumas, P. Even, J. Zwier, A. Mꢂller, J. Didelon,
F. Guillemin, M. L. Viriot, M. Barberi-Heyob, Eur. J. Med. Chem. 2005, 40,
1111–1122; S. Ballut, A. Makky, B. Chauvin, J.-P. Michel, A. Kasselouri, P.
Maillard, V. Rosilio, Org. Biomol. Chem. 2012, 10, 4485–4495.
[4] M. R. Saboktakin, R. M. Tabatabaie, F. S. Amini, A. Maharramov, M. A.
Ramazanov, Med. Chem. 2013, 9, 112–117; E. Secret, M. Maynadier, A.
Gallud, M. Gary-Bobo, A. Chaix, E. Belamie, P. Maillard, M. J. Sailor, M.
Garcia, J.-O. Durandad, F. Cunin, Chem. Commun. 2013, 49, 4202–4204.
[5] R. L. Schneider, F. Schmitt, C. Frochot, Y. Fort, N. Lourette, F. Guillemin,
J. F. Muller, M. Barberi-Heyob, Bioorg. Med. Chem. 2005, 13, 2799–2808.
[6] W. Punfa, S. Yodkeeree, P. Pitchakarn, C. Ampasavate, P. Limtrakul, Acta
Pharmacol. Sin. 2012, 33, 823–831.
[20] G. Oenbrink, D. Gabel, Strahlenther. Onkol. 1989, 165, 130–131.
[21] J. D. Spikes, J. C. Bommer, J. Photochem. Photobiol. B 1993, 17, 135–
143.
[22] C. J. Gomer, A. Ferrario, Cancer Res. 1990, 50, 3985–3990.
[7] a) S. G. Kimani, T. A. Shmigol, S. Hammond, J. B. Phillips, J. I. Bruce, A. J.
MacRobert, M. V. Malakhov, J. P. Golding, Photochem. Photobiol. 2013,
89, 139–149; b) A. R. Karimi, A. Khodadadi, Tetrahedron Lett. 2012, 53,
5223–5226.
Received: August 7, 2013
Revised: January 29, 2014
Published online on April 7, 2014
Chem. Eur. J. 2014, 20, 6054 – 6060
www.chemeurj.org
6060
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim