Selenium‐Containing Type‐I Organic Photosensitizers with Dual Reactive Oxygen Species of Superoxide and Hydroxyl Radicals as Switch‐Hitter for Photodynamic Therapy

Abstract Organic type‐I photosensitizers (PSs) which produce aggressive reactive oxygen species (ROS) with less oxygen‐dependent exhibit attractive curative effect for photodynamic therapy (PDT), as they adapt better to hypoxia microenvironment in tumors. However, the reported type‐I PSs are limited and its exacted mechanism of oxygen dependence is still unclear. Herein, new selenium‐containing type‐I PSs of Se6 and Se5 with benzoselenadiazole acceptor has been designed and possessed aggregation‐induced emission characteristic. Benefited from double heavy‐atom‐effect of selenium and bromine, Se6 shows a smaller energy gap (ΔE ST) of 0.03 eV and improves ROS efficiency. Interestingly, type‐I radicals of both long‐lived superoxide anion (O2 •‾) and short‐lived hydroxyl (•OH) are generated from them upon irradiation. This may provide a switch‐hitter of dual‐radical with complementary lifetimes for PDT. More importantly, simultaneous processes to produce •OH are revealed, including disproportionation of O2 •‾ and reaction between excited PS and water. Actually, Se6 displays superior in–vitro PDT performance to commercial chlorin e6 (Ce6), under normoxia or hypoxia. After intravenous injection, a significantly in–vivo PDT performance is demonstrated on Se6, where tumor growth inhibition rates of 99% is higher than Ce6. These findings offer new insights about both molecular design and mechanism study of type‐I PSs.


Preparation of Se6-NPs and Se5-NPs:
1 mg Se6 or Se5 and 3 mg DSPE-PEG2000 were dissolve in 1.0 ml THF according to the weight ratio of 1:3.Slowly added the THF solution into 10 ml water under ultrasonic conditions.After the dripping was completed, the mixture was sonicated for 5 min.The excess THF was removed by dialysis using a dialysis bag (MWCO: 10, 000 Da) for 48 h in water.

The overall ROS detection by DCFH:
The overall ROS generation measurement was conducted using DCFH as a fluorescence sensor.Se6-NPs/Se5-NPs (30 μM) and DCFH (25 μM) were dispersed in PBS solution.White light (20 mW/cm 2 ) was used to irradiate the solution, and the change in fluorescence of these solutions were detected by a fluorescence spectrophotometer at an interval of 20 s.

The superoxide anion (O2 •− ) detection by DHR123:
The superoxide anion generation measurement was conducted using DHR123 as a fluorescence sensor.Se6-NPs/Se5-NPs (30 μM) and DHR123 (10 μM) were dispersed in PBS solution.White light (20 mW/cm 2 ) was used to irradiate the solution, and the change in fluorescence of these solutions were detected by a fluorescence spectrophotometer at an interval of 20 s.

The hydroxyl radical ( • OH) detection by HPF:
The hydroxyl radical generation measurement was conducted using HPF as a fluorescence sensor.Se6-NPs/Se5-NPs (30 μM) and HPF (20 μM) were dispersed in PBS solution.White light (20 mW/cm 2 ) was used to irradiate the solution, and the change in fluorescence of these solutions were detected by a fluorescence spectrophotometer at an interval of 1 min.
For hypoxic condition, the mixture solution of HPF with Se6-NPs/Se5-NPs was deoxygenized through three cycles of a sequential procedure involving cryofreezing, vacuuming, and defrosting.Then, high purity of nitrogen gas was used to aerate the sample.The irradiation by white light and subsequent measurement was carried out under the nitrogen atmosphere.

The superoxide anion (O2 •− ) detection by Electron ParaMagnetic Resonance (EPR):
The generation of O2 • − was qualitatively characterized using an EPR spectrometer.For the hypoxic condition, Se6/Se5 (100 μM) and DMPO were dissolved in DMSO.The mixed solution was exposed to white light (0.1 W/cm 2 ) for 5 min.The product of DMPO-O2 • − was immediately tracked with EPR spectrometer.
For the normoxia condition, Se6/Se5 (100 μM) and DMPO were also dissolved in DMSO.When exposed to white light (0.1 W/cm 2 ) for 5 min, air was continuously bubbling into the mixed solution at the same tiem, in order to increase the oxygen content of the solution.The product of DMPO-O2 • − was immediately tracked with EPR spectrometer.

The hydroxyl radical ( • OH) detection by Electron ParaMagnetic Resonance (EPR):
For the normoxia condition, Se6-NPs/Se5-NPs (100 μM) and DMPO were dissolved in PBS solution.The mixed solution was exposed to white light (0.1 W/cm 2 ) for 5 min.The product of DMPO-• OH was immediately tracked with EPR spectrometer.
For the hypoxic condition, Se6-NPs/Se5-NPs (100 μM) and DMPO were also dissolved in PBS solution.the mixture solution was deoxygenized through three cycles of a sequential procedure involving cryofreezing, vacuuming, and defrosting.High purity of nitrogen gas was subsequently used to aerate the sample.Then, the mixed solution was exposed to white light (0.1 W/cm 2 ) for 5 min.The product of DMPO-• OH was immediately tracked with EPR spectrometer.

Cell localization experiment:
The 4T1 cells were seeded in a confocal dish.The nanoparticles and the cells were incubated for different times, stained with lysosome probe (LysoGreen), and then fixed with paraformaldehyde.The position distribution of nanoparticles in the cell were observed by using a laser confocal microscope.

Cytotoxicity test:
The 4T1 cells were inoculated in a 96-well plate (1×10 4 per well).After the cells adhered to the wall, the cells were co-cultured with different concentrations of Se6-NPs, Se5-NPs and Ce6-NPs for 4 h, respectively.Then they were irradiated with white light (20 mW/cm 2 ) for 30 min, and were continued culturing for hypoxic cytotoxicity test, the adherent cells were put in a hypoxic incubator for 6 h.
Then different concentrations of Se6-NPs, Se5-NPs and Ce6-NPs were added into and co-cultured with the cells for 4 h, respectively.These samples were irradiated with white light (20 mW/cm 2 ) for 30 min, and continued culturing for 24 h.The cell viability was also detected by CCK8.Regarded to the RM-1 cells, similar procedures to 4T1 cells were performed with different concentrations of Se6-NPs and Se5-NPs.All nanoparticles were performed six samples at the same time.

Cell apoptosis analysis:
The 4T1 cells were inoculated in the culture plate.After the cells adhered to the wall, the cells were co-cultured with Se6-NPs or Se5-NPs for 4 h.Then they were irradiated with white light (20 mW/cm 2 ) for 30 min, and continued culturing for 24 h.The apoptosis kit (Annexin V-FITC/PI) was utilized to stain and the cell apoptosis was observed by using a flow cytometry or a confocal laser scanning microscope (CLSM).

Intracellular ROS detection:
The cells were inoculated in a confocal culture dish.
After the cells adhered to the wall, the cells were co-cultured with Se6-NPs for 4 h.DCFH-DA, HPF, and DHE fluorescence probes were co-cultured with the cells for 30 min, and then these cells were irradiated with white light (20 mW/cm 2 ) for 10 min.The ROS production in the cells was observed by a confocal microscope.The cells in corresponding hypoxia group were under the hypoxia condition for 6 h in advance.The cells were also kept in a hypoxia environment when they were irradiated, and then the observations of cells were performed subsequently on a confocal microscopy.

Mitochondrial membrane potential (MMP) measurement:
The MMP of cells was detected by JC-1 mitochondrial membrane potential probe in accordance with the manufacturer's protocol.Briefly, after different treatments, cells were incubated with 2 μg/mL of JC-1 probe for 20 min at 37 °C.After washing twice with PBS, the cells were observed using a confocal laser scanning microscope (CLSM).

Confocal co-localization:
The 4T1 cells were inoculated in a confocal culture dish.After the cells adhered to the wall, the cells were co-cultured with Se6-NPs or Se5-NPs for 4 h.and Then cells were incubated with 1 mL medium containing commercial dyes at 37 ℃ for 30 min, including Lyso-Tracker Green (100 nM, stock solution: 1 mM in DMSO), ER-Tracker Green (1 μM, stock solution: 1 mM in DMSO) or Mito-Tracker Red (100 nM, stock solution: 100 μM in DMSO), respectively.The medium was then removed.After rinsed with PBS for three times, the cells were observed using CLSM.For Se6-NPs and Se5-NPs, the excitation was 488 nm and the emission filter was 650690 nm; For probes, ER-Tracker Green, Mito-Tracker green and Lyso-Tracker green, the excitation was 488 nm, and the emission filter was 510540 nm.

Experimental animal models:
The animal experimental program was approved by the animal ethics committee of Sun Yat-sen University.The female BALB/c mice (6 weeks old) were purchased from Guangdong medical laboratory animal center.4T1 cells (5 × 10 5 ) suspended in PBS solution were subcutaneously injected into the mouse.
The tumor treatment was started when the tumor volume was close to 70 mm 3 after 7 days of inoculation.

In-vivo cancer treatment assessment:
The tumor-bearing mice were randomly divided into five groups (four mice in each group, n = 4): PBS, PBS + Light, Se6-NPs, Se6-NPs + Light and Ce6-NPs + Light.They were intravenously injected with nanodrug (100 μM, 100 μL) or PBS (100 μL) on tail every two days (48 h).The body weight and the tumor volume were recorded every two days during the experimental period.
After 21 days, the tumor-bearing mice were killed and their tumor tissues and viscera were examined by immunohistochemical staining.

In-vivo fluorescent assessment:
In brief, the tumor-bearing mice were injected with 100 μL Se6-NPs (Se6 100 μM) in caudal vein, and then observed the distribution of in-vivo fluorescence at different time points using an In Vivo Imaging System (IVIS).
After the mice dissected 48 h later, its major organs were removed to observe fluorescence distribution.Wavelengths for excitation and emission: 520 and 670 nm.

Pharmacokinetic study:
To evaluate the pharmacokinetics of Se6-NPs, three healthy female BALB/c mice were intravenously injected with Se6-NPs (100 µL, 100 μM).Then their blood were collected at different time points and the relative blood serum was provided by a centrifuge.Then 10 µL of blood serum was taken from each sample for the assay.The fluorescence of blood serum fluid was examined using a multifunctional microplate reader.After 48 h injection of Se6-NPs, the distribution of Se6-NPs in major organs was then observed using an In Vivo Imaging System (IVIS).
Wavelengths for excitation and emission: 520 and 670 nm.
Photoluminescence (PL) spectra were measured using a Shimadzu RF-5301PC spectrometer or a Horiba fluoromax4 spectrometer.UV-vis absorption spectra were detected using a Hitachi U-3900 spectrophotometer.Electron ParaMagnetic Resonance (EPR) signals were collected from a Bruker A300 Electron ParaMagnetic Resonance.Scanning electron microscope (SEM) imaging of nanoparticles were observed on a Hitachi S-4800 scanning electron microscope.Confocal imagings of cells were observed on an Olympus FV3000 confocal microscope.Flow cytometry analysis was performed using an Attune Nxt flow cytometer.Nanoparticle sizes were measured on a Malvern Panalytical Zetasizer nano.CCK8 UV absorptions were collected from a Synergy H1 multifunctional microplate reader.In vivo imagings were collected from PerkinElmer IVIS Lumina III Series.The quantum chemistry calculationd were performed on the Gaussian 09 software at B3LYP/6-31G(d) level of theory using the density functional theory (DFT) method.The energy levels of excited states were caculated from the corresponding ground state geometry using the combination of TD-B3LYP/6-31G(d).

Figure
Figure S1.a) The UV-vis absorption spectra of Se5 and Se6 in THF solution (10 μM).The UV-vis absorption and PL spectra of b) SeH and c) S5 in THF solution (10 μM).

Figure S3 .
Figure S3.The phosphorescence spectra of Se5 and Se6 in solid state at 77 K.

Figure
Figure S5.a) UV and PL spectra (10 μM), b) Particle size distribution, and c) SEM image of Se5-NPs.Scale bar in SEM image: 200 nm.The PDI of nanoparticles was also inset in b).

Figure S8 .
Figure S8.Absorption spectra of ABDA in PBS in the presence of a) Se6-NPs, b) Se5-NPs, c) ABDA alone, and d) MB after exposure to white light irradiation of 20 mW cm 2 with different time.Concentrations: 30 μM for MB, Se5-NPs and Se6-NPs, 100 μM for ABDA.

Figure S9 .
Figure S9.The singlet oxygen generation detection of Ce6 in various solvents.Absorption spectra of ABDA in a) water, and in the presence of Ce6 in b) water, c) DMSO, and d) water/DMSO = 1/1 (V/V) after exposure to white light irradiation of 20 mW cm 2 with different time.e) UV absorption of Ce6 in DMSO.f) The plots of decomposition rates of ABDA in solution with various ratio of water and DMSO.Concentration: 30 μM for Ce6, 100 μM for ABDA.

Figure S10 .
Figure S10.PL spectra of DHR123 in PBS in the present of a) Se6-NPs, b) Se5-NPs, and c) DHR123 alone after exposure to white light irradiation of 20 mW cm 2 with different time.d) Plot of relative PL intensity of O2 • − probe DHR123 versus irradiation time.Concentrations: 30 μM for Se5-NPs and Se6-NPs, 10 μM for DHR123.

Figure S11 .
Figure S11.PL spectra of HPF in PBS in the presence of PSs under the white light irradiation of 20 mW cm 2 with different time: a) Se6-NPs, c) Se5-NPs, and e) HPF alone under normoxia condition; b) Se6-NPs, d) Se5-NPs, and f) HPF alone under hypoxia condition.Concentrations: 30 μM for Se5-NPs and Se6-NPs, 20 μM for HPF.

Figure S17 .
Figure S17.a) Particle size distribution and b) SEM image of Ce6-NPs.Scale bar in SEM image: 200 nm.c) UV spectra of free Ce6 in DMSO and Ce6-NPs in water, concentration: 10 μM.The PDI of nanoparticles was also inset in a).

Figure S18 .
Figure S18.Apoptosis analysis using flow cytometry toward 4T1 cells after different treatments in various groups.Percentages of cell apoptosis for different groups calculated after PDT treatments were inset.

Figure S19 .
Figure S19.In vitro characterization about the damage of mitochondria on 4T1 cells in various groups by JC-1 fluorescent sensor with or without treatments upon irradiation of 20 mW cm 2 for 30 min.Scale bar: 20 μm; concentrations of Se6-and Se5-NPs: 3 μM; excited wavelength for both JC-1 monomer and aggregates: 488 nm; collected range of fluorescence for JC-1 monomer and aggregates: 510-540 nm and 580-610 nm, respectively.As shown in FigureS19, the bright red fluorescence of JC-1 aggregates on surfaces of mitochondria became weak and almost disappeared in groups of Se5-NPs + light (L) and Se6-NPs + L, respectively.These differences on fluorescence compared to the other groups revealed the damage of mitochondria by ROS generated from Se5-and Se6-NPs.

Figure S20 .
Figure S20.Cell viability of RM-1 cells in the dark incubated with a) Se6-NPs and b) Se5-NPs at various concentrations under normoxia or hypoxia conditions.Cell viability of RM-1 cells incubated with c) Se5-NPs and d) Se6-NPs at various concentrations upon irradiation of 20 mW cm 2 for 30 min under normoxia or hypoxia conditions.(n = 6, mean ± SD).

Figure S21 .
Figure S21.Pharmacokinetic investigation of Se6-NPs.a) Fluorescence intensity of Se6-NPs in blood of the mice injected of Se6-NPs.b) Ex vivo fluorescence images of major organs and e) their relative fluorescence intensity of Se6-NPs inside after 48 hours of tail vein injection.Wavelengths for excitation and emission: 520 and 670 nm, respectively.Concentration of Se6-NPs: 100 μM in 100 μL.(n = 3, mean ± SD).

Figure S22 .
Figure S22.In vivo PDT treatment characteristic of Ce6-NPs on 4T1-tumor-bearing mice.a) Photos of tumor in vitro at day 21 after various treatments.b) Images of HE, Ki67 and TUNEL staining analysis of tumor slices.Scale bars: 100 µm; treatments: PBS + Ce6-NPs + L (Group 5).

Figure S23 .
Figure S23.HE images for slices of main organs of these mice in various groups: heart, liver, spleen, lung, and kidney after treatments.Various treatments: PBS for Group 1, PBS + Light (L) for Group 2, PBS + Se6-NPs for Group 3, and PBS + Se6-NPs + L for Group 4, PBS + Ce6-NPs + L for Group 5; scale bar: 100 μm.