pH-Dependent Cancer-Directed Photodynamic Therapy by a Water-Soluble Graphitic-Phase Carbon Nitride–Porphyrin Nanoprobe

Article abstract of DOI:10.1002/cplu.201600085

Theranostic photodynamic nanomaterials suffer from poor water solubility and nontargeted toxicity. A water-soluble graphitic-phase carbon nitride-based material (g-C₃N₄) conjugated to a positively charged porphyrin P2 (conjugating concentration: 60 μm mg^?1 mL^?1) is shown to be a new concept of photodynamic therapeutic agent (g-C₃N₄–P2). The pH-sensitive emission of g-C₃N₄ is the driving force for the generation of ¹O₂ from g-C₃N₄–P2. The amount of ¹O₂/light generated from a photosensitizer porphyrin can be controlled by the pH-sensitive emission of g-C₃N₄ at acidic pH (pH 4–6), especially under acidic conditions mimicking those of tumor tissue, and thus has the potential to be utilized as a cancer-selective PDT agent.

Full text of DOI:10.1002/cplu.201600085

DOI: 10.1002/cplu.201600085
Full Papers
pH-Dependent Cancer-Directed Photodynamic Therapy by
a Water-Soluble Graphitic-Phase Carbon Nitride–Porphyrin
Nanoprobe
Chi-Fai Chan,[b] Yan Zhou,[b] Hongyu Guo,[a] Junying Zhang,*[a] Lijun Jiang,[b] Wei Chen,*[c]
Kwok-Keung Shiu,[b] Wai-Kwok Wong,*[b] and Ka-Leung Wong*[b]
Theranostic photodynamic nanomaterials suffer from poor
water solubility and nontargeted toxicity. A water-soluble
graphitic-phase carbon nitride-based material (g-C3N4) conju-
gated to a positively charged porphyrin P2 (conjugating con-
centration: 60 mmmgÀ1 mLÀ1) is shown to be a new concept of
photodynamic therapeutic agent (g-C3N4–P2). The pH-sensitive
emission of g-C3N4 is the driving force for the generation of
1
1O2 from g-C3N4–P2. The amount of O2/light generated from
a photosensitizer porphyrin can be controlled by the pH-sensi-
tive emission of g-C3N4 at acidic pH (pH 4–6), especially under
acidic conditions mimicking those of tumor tissue, and thus
has the potential to be utilized as a cancer-selective PDT agent.
Introduction
1
Most photodynamic therapy (PDT) agents are not as effective
as anticipated because of poor solubility, nonspecific toxicity
and lack of tumor selectivity.[1] To overcome these limitations,
several new nanomaterial-based PDT agents have recently
been reported, such as UV-emitting upconversion-based TiO2
nanoparticles,[2] lanthanide-doped upconverting nanocrystal-
phyrin for O2 generation/light emission, conjugation of g-C3N4
and porphyrin represents a salient strategy for the develop-
ment of next-generation PDT agents with improved cancer se-
lectivity.
In this study, we synthesized a new water-soluble g-C3N4-
based material conjugated with porphyrin (g-C3N4–P2) as
a new proof-of-concept PDT model agent (Scheme 1; R1 corre-
line NaYF4:Yb3+,Er3+ [3]
and self-assembled nanoscale coordina-
,
tion polymers for PDT of cancers of the head and neck.[4] Nev-
ertheless, the water solubility and cancer selectivity issues
remain the long-standing design challenges for developing
practical PDT agents. We reasoned that the combined effect of
the pH-sensitive emission properties of water-soluble graphit-
ic-phase carbon nitride (g-C3N4) nanomaterials[5] on the well-
known hydrophilic singlet oxygen (1O2)-generator porphyrin
can be strategically utilized as a possible and sound solution.[6]
There are many reports in the literature showing that tumor
tissue is more acidic than healthy tissue and this difference in
pH can be exploited as a cancer-specific marker.[7] Given that
the pH-dependent emission maximum of g-C3N4 is perfectly lo-
cated in the excitation/absorption region (Soret band) of por-
sponds to P1, R2 corresponds to P2). The amount of O2/light
1
emission from a photosensitizer porphyrin can be controlled
by the pH-sensitive emission of g-C3N4 at pH 4–6, which might
be useful in the acidic conditions of tumorous tissue. The pho-
tophysical studies of g-C3N4–Pn (n=1 or 2) were performed in
buffered aqueous solutions. The effect of varying the linker
length between g-C3N4 and the porphyrin on the quantum
1
yield of the induced O2 production/light emission of g-C3N4–
P1 (shorter linker) and g-C3N4–P2 (longer linker) was investigat-
1
ed. Only g-C3N4–P2 exhibited O2 production and light emis-
sion from the porphyrin upon excitation of the g-C3N4 moiety
1
in acidic conditions. The pH-controlled generation of O2 and
light by g-C3N4–P2 was also applied in vitro and shown to be
1
a promising cancer-specific (by pH-responsive O2 generation)
PDT agent with tremendous potential as an eco-friendly mate-
rial for bioimaging and biomedical applications.
[a] H. Guo, Dr. J. Zhang
Department of Physics, Beihang University
Beijing 100191 (P. R. China)
E-mail: zjy@buaa.edu.cn
g-C3N4 can be characterized as a stacked N-substituted
graphite structure and has various morphologies from micron-
sized bulk material to nanometer-sized quantum dots.[8] Never-
theless, low water solubility makes bulk g-C3N4 a poor candi-
date for biomedical applications in which excellent water solu-
bility is usually a principal requirement of the design brief. Re-
cently, water-soluble nanosheets have been developed by
Zhang et al. to function as imaging probes and have exhibited
another interesting feature of g-C3N4 in which the emission
signal is enhanced twofold in an acidic environment.[9] There
have been several studies on the acidity of human solid
tumors and adjacent normal tissues, which show that hyper-
[b] C.-F. Chan, Y. Zhou, L. Jiang, Dr. K.-K. Shiu, Prof. W.-K. Wong, Dr. K.-L. Wong
Department of Chemistry, Hong Kong Baptist University
Kowloon Tong, Hong Kong SAR (Hong Kong)
E-mail: wkwong@hkbu.edu.hk
klwong@hkbu.edu.hk
[c] Prof. W. Chen
Department of Physics,The University of Texas at Arlington
Arlington, TX 76019 (USA)
E-mail: weichen@uta.edu
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cplu.201600085.
ChemPlusChem 2016, 81, 535 – 540
535
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. Synthesis of g-C3N4–Pn nanoparticles as potential pH-responsive photosensitizers for PDT.
glycotic tumor cells are either induced or selected by hypoxia,
thus generating local acidosis.[10] The pH value is substantially
and consistently lower in tumors than in healthy tissue.[10] Also
recently, Lin et al. used this pH-responsive feature to demon-
strate the g-C3N4 nanosheet as a carrier for the anticancer drug
doxorubicin. The doxorubicin could be released in acidic con-
ditions and killed the cancer cells.[11]
of the porphyrin and g-C3N4 is also favorable for energy trans-
fer. The emission intensity of g-C3N4–P1 and g-C3N4–P2 nano-
particles at 671 nm increased by approximately 6.8% and
25.3%, respectively, from pH 8 to 4 under the same excitation
at 350 nm for g-C3N4. Fçrster resonance energy transfer (FRET)
occurs if the acceptor molecule (P1 or P2) is located nanome-
ters away from the donor (g-C3N4),[12] so the distance between
g-C3N4 and P1 is too short for the FRET phenomenon to occur.
Hence, the emission enhancement for g-C3N4–P1 is much
lower than for g-C3N4–P2. In the control experiment, upon exci-
tation at 430 nm, no pH-sensitized emission change (even
quenching) could be observed. pH-sensitized emission can
help in molecular imaging to distinguish cancer cells from
normal cells. g-C3N4–P1 and g-C3N4–P2 can also utilize the pH-
Results and Discussion
The g-C3N4 nanoparticles were synthesized by acid digestion of
bulk g-C3N4 and further conjugated with porphyrin molecules
in a reaction mediated by N-hydroxysuccinimide (NHS) and 1-
ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC). The con-
centrations of P1 and P2 coating the surface of g-C3N4 were
quantified as 68.4Æ3 and 60.5Æ3 mm in 1 mgmLÀ1 g-C3N4 so-
lution, respectively; the extinction coefficients of P1 and P2
were substituted into the Beer–Lambert equation to obtain
the coating concentration. The g-C3N4-coupled porphyrin
system was characterized by transmission electronic microsco-
py, which showed the average size to be approximately 5 nm
(Figure S1). In order to demonstrate the success of the conju-
gation, zeta potential measurements were performed. Because
the porphyrin moiety is positively charged, zeta potential
values of g-C3N4–P1 and g-C3N4–P2 were, as expected, higher
than that of g-C3N4 nanoparticles (Figure S1).
1
sensitized properties for the generation of O2, in keeping with
PDT agents.
Furthermore, the pH-sensitized emission of g-C3N4–P2 could
be observed with the naked eye. g-C3N4, g-C3N4–P1 and g-
C3N4–P2 are highly water soluble and dispersed (Figure S2).
The clear solution can be observed at a high concentration of
g-C3N4–P1 and g-C3N4–P2 in aqueous solution. The yellow solu-
tion of g-C3N4–P1 and g-C3N4–P2 is observed as the conjugated
porphyrin with strong absorption (Figure 1c). At pH 4, upon
excitation at 350 nm (a g-C3N4-specific wavelength, porphyrin
does not absorb in this region), g-C3N4 shows blue emission;
however, porphyrin-based yellowish red emission can be ob-
served by the naked eye (Figure 1c). This emission of g-C3N4–
P2 with g-C3N4 excitation can be explained by FRET from g-
C3N4 to the porphyrin P2.
After the characterization of our nanoprobes, detailed pho-
tophysical measurements were performed. g-C3N4 showed an
absorption band at approximately 300 nm with no significant
bathochromic shift after conjugation with either of the two
porphyrin moieties. The absorption band shape and position
of g-C3N4, g-C3N4–P1 and g-C3N4–P2 materials were all unaffect-
ed by the change of pH, and the Soret absorption band of the
charged porphyrin at approximately 430 nm after conjugation
with two different linkers (P1 and P2) was also unaffected.
Nonetheless, the emission intensity of g-C3N4, g-C3N4–P1 and
g-C3N4–P2 are dependent on the pH. When the pH was
changed from basic to the acidic, the emission intensity at
430 nm of g-C3N4, g-C3N4–P1 and g-C3N4–P2 were enhanced up
to 40%, which is similar to the pH-responsive emission proper-
ties of g-C3N4 nanosheets reported by Zhang and co-workers.[9]
Furthermore, the emission band of g-C3N4 is exactly the same
as the absorption band of the porphyrin moieties. For g-C3N4–
P2, upon the pH change from 8 to 4, enhancement of the
Soret emission band can be observed (Figure 1c). Elongation
As 1O2 generation is the central process in PDT, we com-
1
pared the abilities of g-C3N4–P1 and g-C3N4–P2 to produce O2
upon irradiation. However, our nanoprobes are insoluble in
chloroform so that the conventional standard (tetraphenylpor-
phyrin in chloroform) cannot be utilized for the comparison of
singlet oxygen quantum yield. Therefore, in this study, 1O2 gen-
eration from g-C3N4–P1 and g-C3N4–P2 were evaluated by
using dihydrorhodamine 123 (DHR). Non-emissive DHR is oxi-
dized by reactive oxygen species (ROS) such as 1O2 to form
rhodamine 123 and emits photon at 524 nm.[13,14] Time-depen-
dent emission spectroscopy of g-C3N4, g-C3N4–P1 and g-C3N4–
P2 was performed at 20 mgmLÀ1 in water (Figure 1d). Mixtures
of DHR and g-C3N4–P1 and g-C3N4–P2 produced about two-
and sevenfold higher emission signals, respectively, at 524 nm
than that of g-C3N4 after irradiation for 4 min at 365 nm at
1
pH 7.4. In fact, it is notable that g-C3N4–P2 can produce O2 at
ChemPlusChem 2016, 81, 535 – 540
www.chempluschem.org
536
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 1. The photographs show the emission of a) g-C3N4 and c) g-C3N4–P2 in an acidic environment (lex =350 nm). Absorption and emission spectra demon-
strating a pH-dependent signal of a) g-C3N4, b) g-C3N4-P1 and c) g-C3N4–P2. d) Time-dependent fluorescence intensity of DHR mixed with g-C3N4, g-C3N4–P1
and g-C3N4–P2 nanoparticles at various pH with irradiation to demonstrate the efficiency of 1O2 generation.
an even higher rate in acidic medium. Due to the high linearity
of the curve, it can be claimed that g-C3N4–P2 can generate
1O2 more efficiently than g-C3N4 and g-C3N4–P1, thereby ach-
ieving a better photodynamic effect.
P1 (Table S1). However, there is no severe mortality in normal
cells under the same concentration of nanoparticles. This is be-
cause the low lysosomal pH of normal cells can protect normal
tissues from anticancer agents.[15] The pH-sensitive emission of
g-C3N4 can be the driving force for 1O2 generation from g-
C3N4–P2. Furthermore, the dark cytotoxicity of g-C3N4, g-C3N4–
P1 and g-C3N4–P2 are similar in cancer and normal cell lines
with IC50 values higher than 1700 mgmLÀ1. The data suggests
that our porphyrin-coupled g-C3N4 nanoparticles are nontoxic
in the absence of light.
To assess the PDT efficacy of our nanomaterials, the 3-(4,5-di-
methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
assay was performed on both cancer and normal cells. Figure 2
shows a significant PDT effect occurs with g-C3N4-P2, especially
on the cancer cells under irradiation at a relatively low power
(6 mWcmÀ2). The photo-cytotoxicity of g-C3N4-P2 is also ap-
proximately four times higher than that of g-C3N4 and g-C3N4–
Apart from the photodynamic efficiency, real-time in vitro
visualization is also an essential feature of nanoprobes in bio-
medical applications. By utilizing the excellent biocompatibility
of the water-soluble g-C3N4 nanoparticles, in vitro imaging was
performed by confocal laser scanning microscopy with excita-
tion wavelength l=380 nm (Figure S3). The T24, HeLa and
MRC-5 cells were incubated with g-C3N4, g-C3N4–P1 and g-
C3N4–P2 for 24 h. The HeLa cells were stained by our nano-
probes (indicated by green circles in Figure S3), but little signal
was detected from MRC-5 cells. Furthermore, the subcellular
localization of the probes in HeLa and T24 cells was verified by
co-staining with the commercial probe LysoTracker Green
(Figure 3 and Figure S5, respectively; the negative control with
the mitochondrial marker MitoTracker is shown in Figure S4).
The data in Figure S6 shows that our nanoprobes can be taken
up by HeLa and MRC-5 cells; the reason they show similar
uptake profiles is that the 488 nm laser of the flow cytometer
can only excite the pH-independent fluorophore porphyrin
and not the pH-responsive g-C3N4 nanoparticles.
Figure 2. Photo-cytotoxicity of g-C3N4, g-C3N4–P1 and g-C3N4–P2 in cancer
cells (HeLa and T24) and normal cells (MRC-5).
ChemPlusChem 2016, 81, 535 – 540
www.chempluschem.org
537
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 3. In vitro fluorescence imaging of a) g-C3N4–P1 and b) g-C3N4–P2 at
a concentration of 50 mgmLÀ1 for 24 h (band pass=650–750 nm), co-local-
ized with c,d) 50 nm LysoTracker Green in HeLa cells (band pass=500–
540 nm), as well as their corresponding e,f) overlaid images (lex =380 nm).
The percentage of CO2 in the incubator was reduced from
5% to 4% and finally to 3% in order to demonstrate that the
higher CO2 environment does lead to a higher emission signal
generated by g-C3N4–P1 and g-C3N4–P2. The emission intensity
from the g-C3N4–P1 and g-C3N4–P2 increased gradually with
higher CO2 content environment (Figure 4a). This is because
the cells are exposed to more acidic conditions if the CO2 con-
tent is high. The cells were subsequently irradiated at 380 nm
to visualize the in vitro PDT effect. The morphology and activity
of HeLa cells were not affected initially but later the regular
cell morphology was lost substantially and eventually cell
death was observed after irradiation for 10 min. The cells incu-
bated with g-C3N4–P2 were killed at a faster rate than those in-
cubated with g-C3N4–P1 (Figure 4b). Conversely, no obvious
cell death occurred in the MRC-5 cell culture with or without ir-
radiation. This is a strong proof of concept to show that g-
C3N4–P2 can be a potential cancer (acidic)-selective PDT agent.
Although this conjugate is far from being realized as a practical
cancer treatment, this proof-of-concept study shows significant
promise and the possibility of using the combination of g-C3N4
conjugated with a porphyrin as a cancer-specific PDT agent.
Figure 4. a) Confocal microscopy of HeLa and MRC-5 cells showing lyso-
somes stained by g-C3N4–P1 and g-C3N4–P2 under 3%, 4% and 5% CO2
(24 h incubation, 50 mgmLÀ1 g-C3N4–Pn, band pass=650–750 nm, 30 min
equilibrium time between images). b) Photodynamic effect of g-C3N4–P1 and
g-C3N4–P2 in HeLa cells over time; lex =380 nm.
Experimental Section
Chemicals and materials
All chemicals used were of reagent grade and were purchased
from Sigma–Aldrich and used without further purification. All ana-
lytical-grade solvents were dried by standard procedures, distilled
and deaerated before use. NMR spectra were recorded on a Bruker
Conclusion
1
Ultrashield 400 Plus NMR spectrometer. The H NMR chemical shifts
In conclusion, we synthesized a water-soluble, tumor-pH-acti-
vated PDT nanomaterial containing a conjugated porphyrin.
This nanomaterial consists of hydrophilic, pH-controlled emis-
sive g-C3N4 conjugated to a hydrophobic porphyrin via a five-
carbon linker (g-C3N4–P2). This biocompatible nanomaterial is
nontoxic in dark conditions but highly toxic once exposed to
were referenced to tetramethylsilane (d=0.00 ppm). High-resolu-
tion mass spectra, reported as m/z, were obtained on a Bruker Au-
toflex MALDI-TOF mass spectrometer. The synthetic scheme (in-
cluding the compound numbers mentioned here) and procedures
are described in the Supporting Information.
1
suitable linear/two-photon excitation to generate O2 and light
Synthesis of g-C3N4 nanoparticles
emission through FRET between the g-C3N4 and porphyrin
moieties, especially in the more acidic environment of cancer
cells. This model represents a new direction for the use of
a combination of nanomaterials and porphyrins and a new
generation of PDT agents.
Water-dispersible g-C3N4 nanoparticles were prepared as follows.
Firstly, melamine 5 g was heated at 5008C in air for 2 h to obtain
g-C3N4 as a yellow powder. The g-C3N4 powder was then ground in
ethanol using a planetary ball mill for 200 min at a speed of
ChemPlusChem 2016, 81, 535 – 540
www.chempluschem.org
538
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
580 rpm, then dried in air at 708C to obtain a fine g-C3N4 powder.
A portion of this g-C3N4 powder (1.5 g) was dispersed at a concen-
tration of 60.7 wt% in nitric acid solution (10m, 25 mL) in a sealed
beaker, and heated at 908C with stirring for 2 h. The acid-treated
powder was washed using deionized water several times and then
air dried at 908C. The dried powder (0.3 g) was added to a mixture
of ethylene glycol (16 mL) and deionized water (4 mL) and heated
at 908C for 2 h in a sealed beaker. After washing and drying, the
powder (0.3 g) was added to deionized water (30 mL) and agitated
with sonication for 20 h. The supernatant was dialyzed, exchanging
with water every 6 h until its pH was close to neutral to obtain the
final water-dispersed C3N4 nanoparticles.
methanol (4:1 v/v). Washing was continued until the filtrate
became colorless and odorless. The product was dried to give
a dark-violet solid (1.7 g, 20%), which was a mixture of porphyrins.
The mixture was separated by column chromatography (silica gel:
60–230 mesh; CHCl3/MeOH 9:1 v/v). After evaporation of the sol-
vent, compound 3 was obtained as a dark-violet solid (300 mg,
4%).
A large excess of methyl iodide (93 mL) was added to a stirred solu-
tion (or suspension) of porphyrin 3 in anhydrous DMF (5 mL per
25 mg of porphyrin) and the reaction mixture was heated at 408C
in a one-necked flask equipped with a condenser for 3–4 h. The re-
action progress was monitored by TLC (acetic acid/methanol/water
5:2:1). Once the reaction was complete, the cationic porphyrin 4
was precipitated with diethyl ether, filtered, and washed several
times with the same solvent. The solids were dissolved in acetone/
water (1:1) and reprecipitated with acetone. The product was dried
under reduced pressure to give compound 4 as a dark-purple solid
Preparation of compound 1
4-Pyridinecarboxaldehyde (2.40 mL, 3 equiv) and 4-nitrobenzalde-
hyde (1.28 g, 1 equiv) were added to propionic acid (200 mL). The
mixture was heated at reflux with vigorous stirring for 1 h, and
then freshly distilled pyrrole (2.35 mL, 4 equiv) was added. After
1 h, the mixture was cooled, the solvent was evaporated to dry-
ness and the crude product was purified by column chromatogra-
phy (100% CHCl3 to CHCl3/EtOH 90:10 v/v) to give compound 1
(110 mg, 7%). Rf =0.61 (CHCl3/EtOH 90:10); 1H NMR (400 MHz,
CDCl3,) d=9.02 (d, J=5.1 Hz, 6H; 3,5-pyridyl), 8.82 (m, 6H; b-pyr-
rolic), 8.77 (d, J=4.8 Hz, 2H; b-pyrrolic), 8.63 (d, J=8.7 Hz, 2H; 3,5-
nitrophenyl), 8.10 (d, J=5.1 Hz, 6H; 2,6-pyridyl), 8.37 (d, J=8.7 Hz,
2H; 2,6-nitrophenyl), À2.99 ppm (brs, 2H; pyrrole NH); MS
(MALDI): m/z: calcd for C41H26N8O2: 662.68; found: 663.2214 [M+
H]+.
1
(75%). H NMR (400 MHz, DMSO): d=9.45 (d, J=5.64 Hz, 6H; 2,6-
pyridyl), 9.12 (m, 8H, pyrrole), 9.00 (d, J=5.62 Hz, 6H; 3,5-pyridyl),
8.80–8.30 (m, 4H; Ph), 4.68 (s, 9H; CH3), À3.10 ppm (s, 2H; pyrrole
NH).
Preparation of compound 5
Benzotriazol-l-yl-oxytripyrrolidinophosphonium
hexafluorophos-
phate (PyBOP; 15 equiv) and N,N’-diisopropylethylamine (DIPEA;
10 equiv) were added to a stirred solution of 4 (53 mg, 5 equiv) in
anhydrous DMF (5 mL). After it had been stirred at room tempera-
ture for 5 min to activate the carboxylate, the reaction mixture was
added into a solution of tert-butyl N-(5-aminopentyl)carbamate
(30 mg in anhydrous DMF (5 mL)). The mixture was allowed to
react at room temperature overnight, and then the resin was fil-
tered and washed with DMF (3”2 mL) and MeOH (3”2 mL). Then
the solvent was removed under vacuum to obtain the dry com-
pound. The residue was recrystallized from diethyl ether and dried
to give compound 5 (93% yield). 1H NMR (400 MHz, DMSO), d=
9.45 (d, 6H, J=5.64 Hz; 2,6-pyridyl), 9.12 (m, 8H; pyrrole), 9.00 (d,
6H, J=5.62 Hz; 3,5-pyridyl), 8.80–8.30 (m, 4H; Ph), 4.68 (s, 9H;
CH3), 2.99 (s, 2H), 1.41 (s, 2H), 1.37 (s, 13H), À3.10 ppm (s, 2H, pyr-
role NH).
Preparation of compound P1
An excess of iodomethane (62 mL, 1 mmol, 10 equiv) was added to
a solution of 1 (76 mg, 0.1 mmol, 1 equiv) in anhydrous DMF
(10 mL) under an argon atmosphere. The mixture was kept at
room temperature and magnetically stirred for 24 h. After precipi-
tation with diethyl ether and filtration, the corresponding cationic
nitroporphyrin 2 was obtained with 94% yield (87 mg). To a solu-
tion of 2 (71 mg, 80 mmol, 1 equiv) in H2O (10 mL), a solution of
SnCl2 (54 mg, 240 mmol, 3 equiv) in 37% HCl (10 mL) was added.
Acetic acid (10 mL) was added to make a homogeneous solution
and the resulting mixture was heated at 70–808C overnight with
stirring. The reaction was quenched by neutralization with 1m
NaOH (100 mL). The mixture was washed with water (2”100 mL),
dried over MgSO4 and filtered to give P1 as a purple solid (54 mg,
90%). Rf =0.12 (CHCl3/MeOH/H2O 5:4:1); 1H NMR (400 MHz,
[D6]DMSO): d=9.51 (brs, 6H; 3,5-pyridyl), 9.10 (m, 8H; b-pyrrolic),
9.00 (d, J=4.4 Hz, 6H; 2,6-pyridyl), 7.90 (d, J=8.3 Hz, 2H; 2,6-
phenyl), 7.04 (d, J=8.3 Hz, 2H; 3,5-phenyl), 4.74 (s, 9H; NCH3),
À3.06 ppm (br s, 2H; pyrrole NH); MS (MALDI): m/z: calcd for
C44H40N8O3: 677.2210 [M+H]+; found: 677.3173.
Preparation of compound P2
Compound 5 and HCl (5m, 5 mL) were added to a 25 mL round-
bottom flask and stirred at room temperature for 2 h. Then, the re-
action mixture was neutralized with NaOH (5m). The solvent was
removed under vacuum to obtain the dry compound. The residue
was recrystallized from diethyl ether and dried to give P2 (94%
yield). MS (MALDI): m/z: calcd: 791.9944; found: 830.5395 [M+K]+.
Synthesis of g-C3N4–Pn conjugates
g-C3N4 nanoparticles (10 mg), EDC (23.6 mg) and NHS (34.5 mg)
were dissolved in anhydrous DMF (10 mL) and stirred at room tem-
perature for 30 min. Then, Pn (n=1 or 2; 20 mg) was added to the
reaction mixture, which was stirred at room temperature for 48 h.
Excess EDC, NHS and porphyrin were washed out with DMF until
no luminescence from the porphyrin was detected by irradiation
under a UV lamp. The g-C3N4–P1 and g-C3N4–P2 nanoprobes were
centrifuged at 15000 rpm (25200 g) at room temperature for
10 min. Finally, the product was dispersed in water and stored at
48C for future use.
Preparation of compound 4
Propionic acid (600 mL), 4-pyridinecarboxaldehyde (3.2 g, 30 mmol)
and 4-carboxybenzaldehyde (1.5 g, 10 mmol) were placed consecu-
tively in a 1 L round-bottom flask equipped with a mechanical stir-
rer. The mixture was heated to 1408C and pyrrole (6.7 g, 40 mmol)
was added dropwise over 30 min with stirring. Heating was contin-
ued for additional 30 min and the mixture was then cooled and al-
lowed to recrystallize. The recrystallized product was filtered off,
washed several times with water and with a mixture of water and
ChemPlusChem 2016, 81, 535 – 540
www.chempluschem.org
539
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Characterization
Flow cytometry measurements of cellular uptake
The morphology of the g-C3N4-coupled porphyrin system samples
was characterized by using a Tecnai G2 20 S-TWIN transmission
electron microscope operated at 200 kV, and the zeta potential
was measured by using a Malvern Zetasizer Nano ZS and an aque-
ous sample solution.
HeLa and MRC-5 cells (105 per sample) were seeded onto 35 mm
Petri dishes and incubated overnight. Then the cells were incubat-
ed with nanoparticles (20 mgmLÀ1) for 3 and 6 h. Cells were har-
vested with trypsin and washed twice with PBS. The uptake of the
nanoparticles by the HeLa and MRC-5 cells was analyzed by flow
cytometry. The cells were excited with a 488 nm argon laser and
emission was collected in the FL-3 channel (with a 650 nm long-
pass filter); 10000 events were analyzed.
Photophysical measurements
The emission and absorption spectroscopy of g-C3N4–Pn in aqueous
solution at various pH values was performed with a Fluorolog
spectrometer equipped with a xenon lamp and an Agilent 8453
UV/Vis spectrometer, respectively.
Acknowledgements
This work was supported by a grant from the Research Grant
Council of the Hong Kong Special Administrative Region (HKBU
203112), Hong Kong Baptist University (FRG2/14-15/013) and the
National Science Foundation of China (Grant No. 91222110 and
51472013). W.C. acknowledges supportfrom the U.S. Army Medi-
cal Research Acquisition Activity (USAMRAA) under contracts
W81XWH-10-1-0279 and W81XWH-10-1-0234.
Cell culture
Human bladder carcinoma (T24) cells were cultured in RPMI 1640
medium (Gibco) supplemented with 10% fetal bovine serum (FBS,
Gibco) and antibiotics (penicillin, 50 mgmLÀ1
;
streptomycin,
50 mgmLÀ1). Human cervical carcinoma (HeLa) cells were cultured
in DMEM (Gibco) supplemented with 10% FBS (Gibco) and antibi-
otics (penicillin, 50 mgmLÀ1; streptomycin, 50 mgmLÀ1). Human
normal lung fibroblast (MRC-5) cells were maintained in minimum
essential medium (MEM) supplemented with 10% FBS and 1%
50 mgmLÀ1 penicillin/50 mgmLÀ1 streptomycin. All cells were incu-
bated at 378C in a humidified environment with 5% CO2.
Keywords: cancer
·
graphitic-phase carbon nitride
·
nanoparticles
therapy
·
pH-responsive materials photodynamic
·
Dark cytotoxicity
[1] a) J. F. Lovell, M. W. Chan, Q. Qi, J. Chen, G. Zheng, J. Am. Chem. Soc.
2011, 133, 18580–18582; b) J. Xu, S. Shanmugam, C. Boyer, ACS Macro
Lett. 2015, 4, 926–932.
[2] Z. Hou, Y. Zhang, K. Deng, Y. Chen, X. Li, X. Deng, Z. Cheng, H. Lian, C.
Li, J. Lin, ACS Nano 2015, 9, 2584–2599.
[3] R. Naccache, E. M. Rodríguez, N. Bogdan, F. Sanz-Rodríguez, M. C. I.
Cruz, . J. Fuente, F. Vetrone, D. Jaque, J. G. SolØ, J. A. Capobianco, Can-
cers 2012, 4, 1067–1105.
[4] C. He, D. Liu, W. Lin, ACS Nano 2015, 9, 991–1003.
[5] S. Barman, M. Sadhukhan, J. Mater. Chem. 2012, 22, 21832–21837.
[6] T. Zhang, R. Lan, C.-F. Chan, G.-L. Law, W.-K. Wong, K.-L. Wong, Proc.
Natl. Acad. Sci. USA 2014, 111, E5492–E5497.
T24, HeLa and MRC-5 cells (105) were treated with g-C3N4–Pn nano-
particles for 24 h at six concentrations (1, 20, 50, 100, 200, and
500 mgmLÀ1). The cell monolayers were rinsed once with phos-
phate-buffered saline (PBS) and incubated with 500 mgmLÀ1 MTT
solution. The cellular inhibitory potency of the nanoparticles was
examined by treating the cells with MTT for 3 h to allow formazan
production during cell metabolism. After that time, the formazan
crystals were fully dissolved in DMSO with oscillation. Finally, the
absorbance of solution was measured with Biotek PowerWave XS
microplate reader at the wavelengths of 570 and 690 nm.
[7] R. A. Gatenby, R. J. Gillies, Nat. Rev. Cancer 2004, 4, 891–899.
[8] W. Wang, J. C. Yu, Z. Shen, D. K. L. Chan, T. Gu, Chem. Commun. 2014,
50, 10148–10150.
Photo-cytotoxicity
T24, HeLa and MRC-5 cells (105) were treated with g-C3N4–Pn nano-
particles for 24 h at four concentrations (1, 5, 10, and 50 mgmLÀ1).
Then, the cells were irradiated (6 mWcmÀ2) and incubated for a fur-
ther 24 h. The cells were then treated according to the same proto-
col as the previous MTT assay.
[9] X. Zhang, X. Xie, H. Wang, J. Zhang, B. Pan, Y. Xie, J. Am. Chem. Soc.
2013, 135, 18–21.
[10] N. Raghunand, R. A. Gatenby, R. J. Gillies, Br. J. Radiol. 2003, 76, S11 –22.
[11] L.-S. Lin, Z.-X. Cong, J. Li, K.-M. Ke, S.-S. Guo, H.-H. Yang, G.-N. Chen, J.
Mater. Chem. B 2014, 2, 1031–1037.
[12] A. J. W. Visser, O. J. Rolinski, Photobiological Sciences Online (Eds. J. Lee,
K. C. Smith) American Society for Photobiology, 2010, http://photobio-
logy.info/.
In vitro confocal microscopy
[13] K. M. Hanson, R. M. Clegg, Photochem. Photobiol. 2002, 76, 57–63.
[14] I. L. Christensen, Y. P. Sun, P. Juzenas, J. Biomed. Nano-tech. 2011, 7,
667–676.
[15] R. A. Ndolo, M. L. Forrest, J. P. Krise, J. Pharmacol. Exp. Ther. 2010, 333,
120–128.
To investigate the suitability of the obtained porphyrin-coupled g-
C3N4 nanoparticles as bioprobes, T24, HeLa and MRC-5 cells (105)
were imaged. After incubation with g-C3N4–Pn for 24 h, the cells
were washed with PBS three times before imaging. Images were
acquired on a Leica TCS SPE confocal laser-scanning microscope
equipped with a Ti:sapphire laser (Libra II, coherent). The samples
were excited at a wavelength of 380 nm.
Manuscript received: March 14, 2016
Accepted Article published: April 28, 2016
Final Article published: May 9, 2016
ChemPlusChem 2016, 81, 535 – 540
www.chempluschem.org
540
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim