A H2O2-Responsive Boron Dipyrromethene-Based Photosensitizer for Imaging-Guided Photodynamic Therapy

In this study, we demonstrate a novel H2O2 activatable photosensitizer (compound 7) which contains a diiodo distyryl boron dipyrromethene (BODIPY) core and an arylboronate group that quenches the excited state of the BODIPY dye by photoinduced electron transfer (PET). The BODIPY-based photosensitizer is highly soluble and remains nonaggregated in dimethyl sulfoxide (DMSO) as shown by the intense and sharp Q-band absorption (707 nm). As expected, compound 7 exhibits negligible fluorescence emission and singlet oxygen generation efficiency. However, upon interaction with H2O2, both the fluorescence emission and singlet oxygen production of the photosensitizer can be restored in phosphate buffered saline (PBS) solution and PBS buffer solution containing 20% DMSO as a result of the cleavage of the arylboronate group. Due to the higher concentration of H2O2 in cancer cells, compound 7 even with low concentration is particularly sensitive to human cervical carcinoma (HeLa) cells (IC50 = 0.95 μM) but hardly damage human embryonic lung fibroblast (HELF) cells. The results above suggest that this novel BODIPY derivative is a promising candidate for fluorescence imaging-guided photodynamic cancer therapy.


Introduction
Photodynamic therapy (PDT) has received great attention for their advantages such as minimal invasiveness, high selectivity and low immunogenicity [1,2]. The method involves three elements: near-infrared light, molecular oxygen and photosensitizers, which can generate cytotoxic reactive oxygen species (ROS) in tumors region to induce cancer cells apoptosis and necrosis [3]. However, a drawback of the typical PDT approach is that the therapeutic effect subjects to the depth and range of light exposure [4]. Otherwise, some patients must remain in dark for long periods of time avoiding photo-allergic reactions [5,6], for instance, rashes or blisters. Also, the selectivity of typical photodynamic therapy is still not as high as expected, and the therapeutic effect is not ideal yet [7,8].
On the other hand, considering the higher concentration of H 2 O 2 in cancer cells, a large variety of H 2 O 2 -responsive fluorescent dyes have been widely employed in previous research [28], mainly based on fluorescence resonance energy transfer (FRET) [29], intramolecular charge transfer (ICT) [30] and photoinduced electron transfer (PET) [31] mechanisms. And the probes are constructed with boronate esters [32], aromatic thioethers [33] or oxalate esters [34], all of which react to H 2 O 2 . However, to the best of our knowledge, H 2 O 2 activatable photosensitizers remain very rare [35]. Indeed, a large body of evidence indicates that H 2 O 2 and other ROS play important roles in health and physiological signaling pathways [36]. Herein, we synthesize a H 2 O 2 -responsive photosensitizer for near infrared (NIR) fluorescence imaging-guided photodynamic therapy. As depicted in Scheme 1, in the presence of H 2 O 2 , the arylboronate group is broken, the meso-ester-substituted dye converts to highly labile intermediates, followed by the linker self-immolation to form meso-carboxylate-substituted BODIPY with strong fluorescence emission [37]. And when exposed to the near-infrared light, it would generate singlet oxygen to kill the cancer cells. These ideal features of tumor microenvironment-activated ability, effective therapeutic effect and NIR fluorescence imaging make this kind of PS a promising anticancer agent for selective photodynamic therapy. Scheme 1. Schematic illustration of proposed activation mechanism.

Molecular Design and Chemical Synthesis
Here, a new type of H 2 O 2 -activatable photosensitizer, derived from the design of molecular fluorescent probes in previous investigation [30], have been designed for fluorescence imaging-guided photodynamic therapy. It is well known that the core structure, a BODIPY dye, without any modifications is known to be very poor photosensitizer. Therefore, we introduce iodine atoms incorporated into the 2 and 6 positions of the BODIPY core, which can facilitate intersystem crossing to promote the generation of singlet oxygen [38]. Also, in order to improve the amphiphilicity and biocompatibility, four oligoethyleneglycol moieties are added into the BODIPY core [39]. Furthermore, the BODIPY dye bearing an electron withdrawing arylboronate group which quenches the excited state of the photosensitizer (PS) by photoinduced electron transfer (PET) (Scheme 1), can remain in a quenched state in normal tissue so as not to damage them. Figure 1 shows the detailed synthetic route for the H 2 O 2 -activatable photosensitizer 7. Firstly, compound 3 was prepared by condensation of 2,4-dimethyl pyrrole with methyl chlorooxoacetate, followed by complexation with BF 3 ·Et 2 O in dichloromethane. This compound then underwent electrophilic substitution with I 2 and HIO 3 to give iodo-BODIPY 4, which was treated with aldehyde 2 to afford 5. Compound 5 was hydrolyzed to 6 in the presence of LiI. Finally, compound 7 was obtained by treatment of compound 6 with 4-(Bromomethyl)benzeneboronic acid pinacol ester.

Photochemical and Photophysical Properties
The electronic absorption spectrum in DMSO shows that compound 7 is a typical non-aggregated BODIPY derivative with a strong Q-band absorption at 707 nm, whose position shows a heavy red-shift (75 nm) compared to compound 6 ( Figure 2, Table 1). Upon excitation at 610 nm, compound 7 shows a very weak fluorescence emission at 755 nm with a fluorescence quantum yield (Φ F ) of 0.067, which is significantly weaker than that of the compound 6 (Φ F = 0.17) ( Table 1). The results clearly indicate that the fluorescence of compound 7 is largely quenched by the arylboronate group by the photoinduced electron transfer (PET).  For the purpose of studying the effects of H 2 O 2 on the spectral properties of compound 7, we performed a series of tests with different H 2 O 2 concentrations in PBS solution ( Figure 3) and PBS buffer solution containing 20% DMSO (Figure 4), respectively. As shown in Figure 3a, there is significant blue-shift in the absorption spectra of 7 upon addition of H 2 O 2 (incubated for 8 h). The blue spectral shifts is due to "ester-to-carboxylate" conversion, as depicted in Scheme 1. As a result of conversion, the intramolecular PET is inhibited, the fluorescence of BODIPY dye is recovered ( Figure 3b). The same phenomenon is observed in Figure 4. The effect of incubation time on the absorption and fluorescence spectra of compound 7 has also been explored (the concentration of H 2 O 2 is fixed at 250 µM). As shown in Figures 5 and 6, an obvious blue-shifted absorption and enhanced fluorescence emission are observed. The fluorescence intensity of compound 7 at 678 nm increases more than 40 times after interacted with 250 µM H 2 O 2 in a mixture of PBS and DMSO (4:1, v/v) for several hours. All the results suggest that the arylboronate bond of compound 7 is susceptible to hydrogen peroxide mediated cleavage and releases meso-carboxylate-substituted BODIPY that has strong fluorescence, which depends on the H 2 O 2 concentrations and incubation time.     (Figure 8a,e). Besides, we use NaN 3 as a scavenger for 1 O 2 to get that the degradation rate of 9,10-dimerthylanthracence (DMA) is slowed down (Figure 8c,e).
It is believed that the arylboronate group of the compound 7 quenches most of their singlet excited state, reducing the fluorescence quantum yield and singlet oxygen generation efficiency, and the H 2 O 2 can activate them.

Cellular Fluorescence Imaging
To study the activation of compound 7 by H 2 O 2 at cellular level, we use the confocal laser scanning microscopy (CLSM) to observe the intracellular fluorescence imaging by incubating HeLa and HELF cells with compound 7 (5 µM) for different time. As illustrated in Figure 9a, after treatment with compound 7 in HeLa cancer cells, the intracellular fluorescence intensity increases gradually with the incubation time. This indicates that the intracellular uptake of compound 7 is time-dependent and compound 7 is efficiently activated by endogenous H 2 O 2 in the cancer cells. By contrast, the fluorescence intensity in HELF normal cells is much lower than that in HeLa cancer cells (Figure 9a-c), namely, the amount of intracellular H 2 O 2 determines the extent and rate at which the arylboronate group of compound 7 is cleaved. Those results suggesting the compound 7 is a promising photodynamic agent with good ability to target cancer cells. In addition, we incubated those two kinds of cells with exogenous H 2 O 2 (50 µM) for 0.5 h initially, followed by compound 7 (5 µM) for another 0.5 h. Obviously, the intracellular fluorescence intensity significantly increases in both two types of cells.

Measurements of Intracellular ROS
Here, the 2 ,7 -dichlorofluorescein diacetate (DCFH-DA) was employed as an indicator to detect the generation of intracellular ROS. It is well known that DCFH-DA itself is nonfluorecent, but it can be oxidized by ROS into 2 ,7 -dichlorofluorescein (DCF) with intense green fluorescence [40]. Both HeLa and HELF cells were firstly treated with compound 7 for 0.5 or 2 h, and then illuminated by red light using PBS as a control. As shown in Figure 10, there is a significantly weak fluorescence in HELF cells, whereas a remarkable green fluorescence is observed in HeLa cancer cells, indicating the prescence of high concentrations of ROS in HeLa cells. In addition, the fluorescence intensity increases with incubation time. However, after treated with compound 7 (5 µM) and sodium vitamin C (NaVC) as a ROS scavenger (100 µM) for 2 h, the CLSM images of the HeLa only show a weak green fluorescence. The above results indicate that compound 7 can be activated and have the high ROS generation ability in HeLa cells.

Cytotoxicity Studies
The cytotoxicity of compound 7 was evaluated by MTT assay using HeLa and HELF cells. As shown in Figure 11b, compound 7 has almost negligible toxicity against HELF cells in the presence of or in absence of light, that is, there is neither phototoxicity nor dark toxicity to HELF cells for compound 7. This is because that compound 7 has not been activated in the normal cells with a relatively low concentration of H 2 O 2 , which leads to significantly low phototoxicity. Also, compound 7 hardly damaged HeLa cells in absence of light (Figure 11b). However, when there is light, the cell viability of HeLa cells is significantly reduced as a result of photodynamic activity of compound 7 having been restored in HeLa cells (Figure 11b). The corresponding IC 50 value, defined as the dye concentration required killing 50% of the cells, is 0.95 µM. All above demonstrate that the photodynamic activity of compound 7 is related to the concentration of H 2 O 2 , and has good phototoxicity against cancer cells with high H 2 O 2 concentration. Compound 7 does no damage to normal cells even in the presence of light, indicating that the photosensitizer we designed has good specificity for cancer cells against normal cells.

Subcellular Localization Studies
The subcellular localization of compound 7 in HeLa cells was also studied. From the previous literatures, we get that nucleus, mitochondria and lysosomes play important roles in photodynamic therapy. Therefore, we stained the cells with compound 7 together with DAPI, Mito-Tracker Green and Lyso-Tracker which are specific fluorescence probes for nucleus, mitochondria and lysosomes, respectively. As depicted in Figure 12a, the fluorescence of Mito-Tracker Green and compound 7 are well superimposed. And the line traces in the Figure 11d also illustrate this conclusion. In addition, the experimental results in lysosomes are similar to those of mitochondria (Figure 12b,e). By contrast, the fluorescence images of compound 7 and the DAPI cannot be superimposed (Figure 12c,f). As the previous studies reported [41], intracellular hydrogen peroxide is mainly produced in the mitochondria and then spread into lysosomes. And so, H 2 O 2 -mediated cleavage of compound 7 can occur in mitochondria and lysosomes.

General Information
Most of the chemical reagents and solvents were purchased from Energy Chemical Technology Co., Ltd. (Shanghai, China) except for otherwise stated. Chromatographic purifications were performed on silica gel (Qingdao Ocean, Qingdao, Shandong, China, 200-300 mesh) columns with the indicated eluents. The two kinds of cells, human cervical carcinoma (HeLa) and human embryonic lung fibroblast (HELF) cells were obtained from the cell bank of Shanghai Institutes for Biologic Sciences. And 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was purchased from Gen View Co., Ltd., Tallahassee, FL, USA. Compounds 1, 2, 3 [37] were prepared as described.
NMR spectra were recorded on a Bruker Avance III 400 ( 1 H: 400 MHz, 13 C: 100.6 MHz) instrument (Bruker, Karlstuhe, Germany) and referenced to tetramethylsilane (TMS) as the internal standard. High-resolution mass spectra (HRMS) analysis was carried out on an Agilent 6520 ACURATE-Mass Q-TOF Mass Spectra (Agilent Technologies, Santa Clara, CA, USA). Electronic absorption spectra were measured on a Lambda 365 UV-Visible absorption spectrometer (Perking Elmer, Waltham, MA, USA) and fluorescence spectra were obtained on a VARIAN Carye Elipse Fluorescence spectrometer (Agilent Technologies, Santa Clara, CA, USA). Intracellular fluorescence imaging and subcellular localization was carried on an Olympus FV1000 Confocal Laser Scanning Microscope (Olympus Instrument Co., Ltd., Shinjuku-ku, and Tokyo, Japan).

Synthesis of Compound 4
Compound 3 (0.060 g, 0.20 mmol) was dissolved in absolute ethanol (200 mL), and then added the iodine (0.17 g, 0.67 mmol) and iodic acid (0.10 g, 0.55 mmol). After magnetic stirring under an atmosphere of nitrogen at 60 • C for 4 h, ethanol was removed under reduced pressure. The residue was chromatographed on silica gel with petroleum ether and dichloromethane (1/1, v/v) as eluent to give the red solid compound 4. 1

Synthesis of Compound 5
Compound 4 (0.20 g, 0.36 mmol) was dissolved in a two-neck round bottom flask with 100 mL toluene, and followed by the compound 2 (0.65 g, 1.52 mmol), piperidine (1.2 mL), glacial acetic acid (1.0 mL) and an appropriate amount of anhydrous magnesium perchlorate. Under nitrogen protection, it was condensed and refluxed for 2.5 h with a water separator. After that, the toluene was evaporated under reduced pressure. The residue was mixed with water, and extracted three times by dichloromethane. The combined organic fractions were collected and dried under reduced pressure. The residue was chromatographed on silica gel with dichloromethane and methanol (50:1, v/v) as eluent to afford green viscous compound 5 (0.19 g, 39%). 1

Synthesis of Compound 6
To a solution of compound 5 (0.26 g, 0.19 mmol) in ethyl acetate (100 mL) was added lithium iodide (0.13 g, 0.95 mmol). The resulting mixture was refluxed overnight under an atmosphere of nitrogen. Then the mixture was washed three times with water. The solvent was removed under reduced pressure. The residue was further purified by a silica gel column chromatography using dichloromethane and methanol (25:1, v/v) as eluent to give green viscous compound 6 (0.20 g, 78%).

Synthesis of Compound 7
Compound 6 (0.048 g, 0.035 mmol) was dissolved in 10 mL of anhydrous tetrahydrofuran in a round flask, followed by 4-bromomethylphenylboronic acid pinacol ester (0.021 g, 0.073 mmol), 1,8-diaza Heterobicyclo [5.4.0] undec-7-ene (0.017 g, 0.11 mmol) and an appropriate amount of potassium iodide. The mixture was stirring overnight under nitrogen. Then the mixture was mixed with water and then extracted three times with dichloromethane. The combined organic fractions were collected and dried under reduced pressure. The residue was further purified by a silica gel column chromatography using dichloromethane and methanol (50:1, v/v) as eluent to afford green viscous compound 7 (0.022 g, 39%). 1

Absorption and Fluorescence Studies
The electronic absorption spectra were measured on a Lambda 365 UV-Visible absorption spectrometer (Perking Elmer, Shanghai, China) and fluorescence spectra were obtained on a VARIAN Carye Elipse Fluorescence spectrometer (Agilent Technologies, Santa Clara, CA, USA) by using the compounds dissolving in DMSO as described. And we calculated the fluorescence quantum yield using the following formula: where F and F std are the measured fluorescence area (λ ex = 610 nm) of the compounds and standard; A and A std are the absorbances at the excitation wavelength (610 nm) of the compounds and standard; η and η std are the refractive index of the compounds and standard. Here, the unsubstituted zinc (II) phthalocyanine (ZnPc) in DMF was used as the standard [φ std F = 0.28]. To minimize re-absorption of the radiation by the ground-state species, the emission spectra were obtained by controlling the absorbances of compounds and ZnPc at 610 nm were within the range from 0.03 to 0.05.

Cell Culture and Conditions
Human embryonic lung fibroblast (HELF) and human cervical carcinoma (HeLa) cells were maintained in Dulbecco's modified Eagle's medium (DMEM) (Gibco) supplemented with Fetal Calf Serum (10%, v/v), penicillin streptomycin (1%, v/v) at 37 • C in a humidified atmosphere with 5% CO 2 . Approximately 7 × 10 3 (for HeLa) cells per well in these media were inoculated in 96-multiwell plates and incubated overnight at 37 • C in a humidified 5% CO 2 atmosphere. Then, we collected selectively those cells for following use.

Photocytotoxicity Studies
Compound 7 was first dissolved in DMSO to obtain 1 mM solutions, which then were diluted with the culture medium to appropriate concentrations (0, 0.01, 0.05, 0.1, 0.5, 1, 3, 6, 9 µM) Wherein the DMSO content in the final diluted drug solution does not exceed 1% (v/v). The cells were incubated with medium containing 1% DMSO in either the light or non-lighted cell control group to deduct the toxic effects of DMSO on the cells. And then select suitable HeLa and HELF cells and seeded them in 96-well plates. After incubation for 24 h, different concentrations of compound 7 were added, followed by incubating with compound 7 for 24 h, and rinsing twice with PBS, and finally fresh medium was added. The fluence rate of the light source (λ = 670 nm) was 20 mW·cm −2 .
The toxicity of compound 7 against HeLa and HELF cells was determined by MTT assay. After illumination for 2 min, the experimental group continued to incubate for 24 h. And the control group, after being rinsed twice with PBS, were incubated for 24 h. After that, a prepared 10 µL MTT (5 mg/mL) solution was added to each well. The cells then were incubated under 5% CO 2 at 37 • C for 4 h. The solution in the 96-well plate was removed, and then 100 µL/well of DMSO was added. In addition, 100 µL of DMSO was also added as a blank group in the wells without cells, and the light at λ = 490 nm was read with a microplate reader.
According to the previous work [42], the cell viability was then determined by the following equation: % viability=[∑(A i /A control × 100)]/n, where Ai is the absorbance of the ith data (i = 1, 2,..., n), A control is the average absorbance of the control wells in which the BODIPY was absent, and n (= 6) is the number of the data points. Six replicates were used for each concentration in each experiment and each experiment was repeated three times. The procedures for investigation of dark-cytotoxicity were almost the same except that there was no irradiation. The dose-dependent survival curves were prepared with GraphPad Prism 5.0 Software (GraphPad Software Inc., La Jalla, CA, USA) and IC 50 was calculated.

Measurements of Intracellular ROS
The detection of ROS is based on the indirect verification by DCFH-DA, which is non-fluorescent while can be oxidized to form the fluorescent compound DCF.
First of all, HELF and HeLa cells were seed in a CLSM-special cell dishes at 1.0 × 10 5 cells per dish. After incubation for 24 h, compound 7 (5 µM was added, and the cells in different groups were ingested for compound 7 (λ ex = 635 nm, λ em = 640-800 nm) for 0.5 h, 1 h, 2 h, respectively. Afterwards, DCFH-DA (λ ex = 488 nm, λ em = 500-550 nm), a probe for capturing ROS, was added and co-cultured with the cells. After half an hour, the cells were rinsed with PBS for 5 times, and irradiated for 2 min; then confocal fluorescence imaging was performed to give the level of intracellular ROS with the excitation wavelength of 488 nm and emission wavelength from 500 to 550 nm.

Conclusions
In conclusion, we have prepared a novel H 2 O 2 -responsive BODIPY-based photosensitizer which is caged by an arylboronate group via PET mechanism. This arylboronate-linked BODIPY shows enhanced fluorescence emission and ROS generation in the presence of H 2 O 2 , both in PBS and inside tumor cells, due to cleavage of the arylboronate group linker. With these advantages, this compound is worthy of further study as a potential theranostic agent for cancer imaging and therapy.