Aggregation Turns BODIPY Fluorophores into Photosensitizers: Reversibly Switching Intersystem Crossing On and Off for Smart Photodynamic Therapy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE7 Nov 2022Aggregation Turns BODIPY Fluorophores into Photosensitizers: Reversibly Switching Intersystem Crossing On and Off for Smart Photodynamic Therapy Yan-Fei Kang, Wen-Kai Chen, Kun-Xu Teng, Ling-Yun Wang, Xiao-Cheng Xu, Li-Ya Niu, Ganglong Cui and Qing-Zheng Yang Yan-Fei Kang College of Chemistry, Beijing Normal University, Beijing 100875 College of Laboratory Medicine, Hebei North University, Zhangjiakou, 075000 Hebei , Wen-Kai Chen College of Chemistry, Beijing Normal University, Beijing 100875 , Kun-Xu Teng College of Chemistry, Beijing Normal University, Beijing 100875 , Ling-Yun Wang College of Chemistry, Beijing Normal University, Beijing 100875 , Xiao-Cheng Xu College of Chemistry, Beijing Normal University, Beijing 100875 , Li-Ya Niu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Chemistry, Beijing Normal University, Beijing 100875 , Ganglong Cui *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Chemistry, Beijing Normal University, Beijing 100875 and Qing-Zheng Yang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Chemistry, Beijing Normal University, Beijing 100875 https://doi.org/10.31635/ccschem.021.202101600 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail We report for the first time a practical and simple supramolecular approach to turn fluorophores into photosensitizers (PSs). Using boron dipyrromethene (BODIPY) as a proof-of-concept, eight BODIPY derivatives manifest bright fluorescence and generate negligible singlet oxygen in solution. In contrast, aggregation fails to emit fluorescence and enhances singlet oxygen generation. Experimentally, these aggregates have excellent photodynamic therapy (PDT) performance, and one even exhibits much stronger photocytotoxicity than the commercialized PS Ce6 under identical conditions. Theoretical studies show that this property originated from significantly reduced energy gaps between relevant excited singlet and triplet states, leading to considerably improved intersystem-crossing efficiency. Importantly, a simple disaggregation recovers the original properties of the fluorophores. This reversible switching property between fluorophores and PSs assists the development of smart PDT systems, in which singlet oxygen generation in tumors can be controlled in an intelligent manner after PDT treatment. The present work provides a novel strategy to design heavy-atom-free PSs and may pave the way to the development of smart PDT systems. Download figure Download PowerPoint Introduction Organic fluorophores are widely used in many areas such as fluorescent probes,1–6 bioimaging,7–11 and organic light emitting diodes.12–14 Normally, exciting a fluorophore in dilute solution generates a lowest singlet excited state (S1), which rapidly relaxes back to the ground state (S0) through emitting fluorescence. For these organic compounds, there exists a notorious photophysical phenomenon called aggregation-caused quenching (ACQ). Fluorophores in molecularly dissolved solution emit bright fluorescence, yet become very weak and even disappear in aggregate states. This kind of fluorophore includes fluoresceins and boron dipyrromethenes (BODIPYs), whose strong intermolecular π–π interaction is beneficial to form detrimental aggregated species.15 This interaction prompts excitonic coupling of an electronically excited molecule with a nearby ground one. This interaction opens additional nonradiative relaxation channels that induce ACQ effects, which are usually unfavorable for fluorescent materials in practical applications. Yet the fluorescence-quenching mechanism remains elusive and under debate.16 A possible mechanism, but not one attracting much attention, is related to efficient intersystem crossing (ISC) processes to form triplet manifolds upon fluorophore aggregation.17–19 Taking a three-state model as an example, available in most cases due to the Kasha rule,20 the S1-to-T1 ISC process is significantly accelerated upon aggregation so that the S1 state has no chance to fluoresce. Obviously, this aggregation-facilitated S1-to-T1 ISC process is "a two-edged sword." On one side, it can lead to fluorescence quenching of fluorophores, which is unfavorable for luminescence materials. On the other side, it could be exploited to turn fluorophores into heavy-atom-free photosensitizers (PSs) for photodynamic therapy (PDT).21–24 PDT is a promising strategy to kill cancer cells, which involves using PSs.25 The treatment is based on photochemical reactions between PS in triplet-excited states and molecular oxygen, which generates various toxic reactive oxygen species (ROS). As one of the key elements in PDT, PSs have in recent years attracted significant experimental attention.26–35 However, most of these efforts have been dedicated to synthesizing PSs through molecular modification of fluorophores.36–46 One of the most successful strategies is through introducing heavy atoms into fluorophores to enhance spin–orbit couplings (SOCs) and thereby improve ISC efficiency.47–49 The ROS generation of traditional PSs is usually degraded in aggregation states because of efficient decay of triplet states.50 Moreover, these PSs are developed through molecular covalent synthesis, which makes it difficult to reversibly switch ISC on and off and eventually switches between fluorescence and ROS generation (Scheme 1a). On the other side, single-component organic materials with exclusive switch between singlet and triplet states are critical to develop multifunction optoelectronic materials. Even though fluorophores can use singlet excitons as fluorescent materials or use triplet excitons as PSs via chemical modification, reversible switching between singlets and triplets at the single-fluorophore level remains highly challenging.51 The aggregation-facilitated ISC provides a new supramolecular strategy to switch fluorophores into PSs without chemical tailoring. More importantly, reversible switching between electronically excited singlet and triplet states can be achieved by simple aggregation and disaggregation processes (Scheme 1b). Undoubtedly, this will pave one way for developing smart PDT in which ROS generation is switched on when PSs in tumors are aggeregated upon certain pathological stimulus and then switched off after the treatment.52–56 Scheme 1 | Schematic illustration of (a) traditional PSs based on covalent modification on the current fluorophores; (b) this work: aggregation turns fluorophores into PSs with reversibly switching ISC on and off. Download figure Download PowerPoint In this study, for the first time we report a novel supramolecular approach to turn highly fluorescent organic dyes into efficient PSs through aggregation. As a proof of concept, BODIPYs 1–8 manifested bright fluorescence and failed to generate singlet oxygen in solution. In contrast, when aggregation states formed through simple coassembly with surfactants, their fluorescence was quenched, and singlet oxygen was produced efficiently. Further theoretical calculations revealed that aggregation reduces singlet-triplet energy gaps, thus promoting the ISC process to form triplet manifolds. Furthermore, aggregates of BODIPYs 1–8 exhibit satisfactory PDT performance in tumor cells, in particular 8, which demonstrates superior photocytotoxicity with a half-maximal inhibitory concentration (IC50) of 0.70 μg/mL, which is better than the commercial PS Ce6 at identical conditions. Finally, compound 4 was employed as a proof-of-concept application of smart PDT via reduced glutathione (GSH)-deactivated ROS generation. In summary, we present a general supramolecular strategy to construct efficient and heavy-atom-free PSs based on fluorescent dyes, which could facilitate the development of smart PDT through reversible switching between fluorophores and PSs. Experimental Methods General information Unless otherwise noted, materials were purchased from commercial suppliers and used without further purification. Absorption spectra of liquid samples were measured on a Hitachi UV-3900 spectrophotometer. Fluorescence spectra of liquid samples were determined on a Hitachi F-4600 spectrophotometer. Dynamic light scattering (DLS) investigations were carried out with a DynaPro NanoStar DLS detector. Scanning electron microscopy (SEM) images were obtained using a Hitachi SU-8010 instrument. Confocal fluorescence imaging was performed with Nikon A1R microscopy. The cell viability test was performed on a Thermo Scientific Multiskan spectrophotometer. Irradiation was performed using a light-emitting diode (LED) light (PLS-LED 100, Perfect Light, Beijing, China). Singlet oxygen generation in solutions 1,3-Diphenylisobenzofuran (DPBF) was used as an indicator for the detection of singlet oxygen in solution. 1NP–8NP were dispersed in 2 mL aqueous solution [dimethylformamide (DMF)/H2O = 20/80, v/v] containing 60 μM of DPBF. The mixture was then placed in a cuvette and irradiated with monochromatic light (by a Hitachi F-4600 spectrophotometer). The absorption change of sample at 408 nm was recorded by the UV–vis absorption spectrophotometer. Singlet oxygen generation in living cells 1O2 generation in cells treated with PDT was measured using ROS indicators immediately after the photosensitization experiments. In a specific experiment, HepG2 cells were seeded and cultured with Dulbecco's Modified Eagle Medium (DMEM) in confocal dishes for 24 h at 37 °C under 5% CO2. At that time, the culture medium was exchanged, and the cells were treated with DMEM containing 8NP (5 μg mL−1) for 6 h and then washed three times with phosphate-buffered saline (PBS). The fresh culture medium containing 5 μM 2′,7′-dichlorofluorescein-diacetate (DCFH-DA) (1 mL) was added to the adhering cells, and the cells were then subjected to the photosensitization experiment with light irradiation (590 nm LED light) for 10 min. After that, the medium was discarded, and the cells were washed with PBS buffer three times. Fluorescent images of DCFH-DA and staining on the cells were promptly captured by Nikon confocal microscopy. Meanwhile, to further confirm the 1O2 generation, the singlet oxygen sensor green reagent (SOSG) staining was carried out using the same method as DCFH-DA. Assay for cytotoxic activity The cytotoxic activity was assessed by Cell Counting Kit-8 (CCK-8) assay in which a tetrazolium salt was reduced to formazan dye with high water solubility by the dehydrogenase in living cells under the action of electron carrier 1-methoxy-5-methylsulfoninium dimethyl sulfate (1-methoxy PMS). There is a linear relation between the amount of generated formazan and the number of living cells. Therefore, this property can be used for direct cell proliferation and toxicity analysis. HeLa cells, A549 cells, and HepG2 cells were seeded in 96-well flat microtiter plates for adherence for 24 h, and then the cells were incubated with assigned concentrations of 1NP–8NP for 6 h. Subsequently, the cells were irradiated by an LED light source for 10 min and incubated for another 18 h. Thereafter, 100 μL fresh culture medium with CCK-8 solution was supplemented to each well, and plates were incubated for another 30 min at 37 °C in the dark. The absorbance was recorded at 450 nm using a microplate reader. The percentage of cell viability was calculated relative to the control wells designated as 100% viable cells. DAPI staining of cells in PDT HepG2 cells were seeded on confocal dishes and incubated for 24 h. The medium was replaced with fresh culture medium containing 8NP (5 μg mL−1). After incubation for 6 h, the cells were irradiated by an LED light source (590 nm, 20 mW/cm2) for 10 min. After further incubation for 12 h, the cells were stained with 4′,6-diamidino-2-phenylindole (DAPI). The medium was removed, and the cells were washed three times with PBS buffer. Fluorescent images of stained cells were promptly captured by Nikon A1R microscopy. Flow cytometry test of cell apoptosis HepG2 cells were seeded on six-well cell culture plates and incubated for 24 h. The medium was then replaced with fresh culture medium containing different concentrations of 8NP, and plates were incubated for 6 h. Then, the cells were irradiated with an LED light (590 nm, 20 mW cm−2) for 10 min. After further incubation for 18 h, the cells were collected and treated with AnnexinV-FITC/PI cell apoptosis detection kit. The flow cytometry was used to detect cell apoptosis. Results and Discussion Photophysical properties of 1–8 and 1NP–8NP As a proof of concept, we chose different types of BODIPY derivatives to demonstrate this strategy (Figure 1). BODIPYs are well-known fluorophores that have found wide applications in diverse research fields.57–59 They show superior characteristics such as high extinction coefficients, excellent photostability, and rich synthetic functionalization,60,61 making them ideal for building PSs.62,63 Compounds 1–5 are regular BODIPYs attached with different substitutions on the BODIPY cores. Compound 6 is a typical aza-BODIPY.64 Compounds 7 and 8 are O, B, N-strapped BODIPY derivatives.65,66 Figure 1 | Chemical structures of BODIPYs 1–8. Download figure Download PowerPoint Initially, we took 1 as an example to demonstrate the detailed photophysical properties in solution and aggregation state. The absorption and fluorescence spectra of 1 in tetrahydrofuran (THF)/water with variable water fractions were recorded. As shown in Figure 2a, compound 1 displayed a sharp absorption band with an absorption maximum at 630 nm in THF. Upon gradually increasing the water fraction, a broad spectrum with a bathochromically shifted absorption band was observed, suggesting the formation of aggregates. The dilute solution of 1 in THF showed strong fluorescence at 647 nm. The fluorescence signal of 1 was decreased gradually with the increasing fraction of water, displaying typical ACQ effect (Figure 2b). BODIPYs 2–8 exhibited similar spectroscopic changes ( Supporting Information Figure S1 and Table S1). The fluorescence quenching of 1–8 in the aggregation state indicates that the excited energy dissipates through nonradiative transition, probably involving internal conversion, charge transfer, and/or ISC. If the ISC process (S1 → Tn) is enhanced, the chromophore could serve as a sensitizer to transfer its excited energy of the triplet excited state to the molecular oxygen to generate 1O2. To confirm this speculation, the singlet oxygen generation capability was assessed by the commercial 1O2 probes, DPBF as an indicator. In the presence of compound 1 in THF/water with varying water fractions, the 1O2 generation efficiency was evaluated from the degradation of DPBF (Figure 2c). The signal for ROS generation is enhanced with increasing water fractions, suggesting that the ISC efficiency of the aggregates increased with the formation of aggregates. Figure 2 | (a) Absorption and (b) fluorescence spectra of compound 1 in THF/water mixtures with various ratios (inset: image of compound 1 in THF/water mixtures upon 365 nm light irradiation. THF:H2O = 100:0, 72:28, 28:72, 2:98 from left to right). (c) Degradation rates of DPBF in the presence of 1 in THF/water mixture. A0 and A are the absorbance of DPBF at 418 nm before and after irradiation. (d) Absorption spectra of DPBF and 1 monomer in THF upon light irradiation. (e) Absorption spectra of DPBF and 1NP in aqueous dispersion upon light irradiation. (f) The EPR spectra to detect 1O2 generated by 1NP under illumination, using TEMP as spin trap. Download figure Download PowerPoint To obtain stable aggregates for the further bioapplication, the nanoparticles of 1–8, named as 1NP–8NP, were prepared by encapsulating 1–8 into hydrophobic cores of biocompatible surfactant Pluronic F-127. The SEM images showed that 8NP had a regular spherical morphology ( Supporting Information Figure S2). Moreover, the average size of 8NP, estimated by using DLS, was 120.0 nm ( Supporting Information Figure S3). The nanoparticles displayed a broader and bathochromically shifted absorption spectra and a significantly quenched emission spectra compared to the monomers, suggesting the formation of aggregates ( Supporting Information Figures S4 and S5). The absorption of 8NP remains unchanged after being incubated with cell culture medium (RPMI-1640) containing 10% fetal bovine serum in the dark for 48 h at room temperature ( Supporting Information Figure S6a), indicating that they are remarkably stable under physiological conditions. Moreover, the intensity of the absorption only showed negligible changes when exposed to 590 nm light for 30 min, suggesting that the nanoparticles have the favorable light stability ( Supporting Information Figure S6b). Compounds 1–8 in organic solution caused almost no attenuation of the DPBF absorption, suggesting that they have no 1O2-generating ability as the monomer (Figure 2d and Supporting Information Figure S7). By contrast, in the presence of 1NP–8NP, the absorption of DPBF decreases rapidly (Figure 2e and Supporting Information Figure S7), indicating efficient 1O2 generation. Electron paramagnetic resonance (EPR) was used to prove the existence of 1O2 using 2,2,6,6-tetramethylpiperidine (TEMP) as a spin trap. Upon light irradiation of the aerated aqueous solution of 1NP and TEMP, a characteristic signal of paramagnetic 2,2,6,6-Tetramethylpiperidine 1-oxyl (TEMPO), a product of the reaction from 1O2 with TEMP, was observed. No detectable EPR signals were observed in the solution of 1NP and TEMP in the dark (Figure 2f). The 1O2 signals were also detected for 2NP–8NP ( Supporting Information Figure S8). The results showed that aggregation of 1–8 greatly improved the 1O2 generation efficiency of BODIPY fluorophore and turned the fluorophore into a PS successfully. Theoretical studies To explore the in-depth aggregation effects on ISC efficiency, we have carried out theoretical studies for compound 1 ( Supporting Information Table S2).67 Unlike a unique monomer structure, different dimer and trimer structures were built and optimized ( Supporting Information Figures S9–S13). As shown in Figure 3, for the monomer, when the S1 state is populated directly upon photoirradiation or as a result of internal conversion from higher excited singlet states within the Kasha rule, it is difficult for the S1 state to transition to the T1 state because there is a large energy difference of 0.75 eV between them. Interestingly, in the dimer structure, the related S1-T3, S1-T2, and S1-T1 energy differences were significantly reduced to 0.00, 0.08, and 0.12 eV, respectively. The same situation was also observed in the trimer structure, where there were four triplet states below the S1 state with small singlet-triplet energy differences ranging from 0.00 to 0.17 eV (see Supporting Information Table S3). A similar trend was also seen in the other dimer and trimer structures (see Supporting Information). On the other hand, according to the classical Marcus theory, ISC rate constants are proportional to spin–orbit couplings H i f but inversely proportional to relevant energy differences Δ E , that is, k i → f ∝ | H i f | 2 / Δ E . Since spin–orbit couplings are usually small for heavy-atom-free organic systems, energy differences become the most dominant factor to control ISC efficiency.68 Theoretical studies for compound 8 were also performed and showed similar results as compound 1 ( Supporting Information Tables S4 and S5 and Figures S14–S18). Therefore, the aggregation-enhanced ISC efficiency, as observed in the present experiments, mainly originated from reduced energy gaps between relevant singlet and triplet states. Figure 3 | s-TD-DFT calculated energy-level diagrams of excited singlet (blue) and triplet (red) states of selected (left) monomer, (middle) dimer, and (right) trimer structures of 1. Download figure Download PowerPoint It is commonly known that aggregation-induced emission (AIE) represents an effective approach to enhance upon In recent have also employed strategy to generate ROS in aggregation However, these efforts from present these efforts additional chemical of to improve ISC efficiency via and into their molecular PSs generate ROS and fluorescence in aggregation states. of the fluorescence ROS generation is to a certain In contrast, PSs in strategy are generated via aggregation of fluorophores without further chemical Importantly, one can a reversible switching between fluorescent dyes and heavy-atom-free PSs by of simple aggregation and Photodynamic performance of 1NP–8NP in cancer cells To PDT of 1NP–8NP as photodynamic we their by the CCK-8 assay in the cell the cell and the cancer cell and Supporting Information Figure we evaluated the dark toxicity of 1NP–8NP in the cancer cells, and the results showed that they not exhibit when incubated for 24 h in the even the concentration to Upon light they significantly cell proliferation in a 8NP showed the in the test cells with a of 0.70 To the of tumor cell proliferation was superior to that of commercial Ce6 under identical conditions. results that 8NP has as an excellent PDT for cancer treatment. Figure 4 | Cell viability of and HepG2 cells, and A549 cells, and and HeLa cells subjected to a of 8NP and Ce6 and in the dark and and upon light irradiation (590 nm LED 20 Download figure Download PowerPoint To explore the mechanism at the we further evaluated the capability of 8NP by using DCFH-DA as an ROS indicator in HepG2 cells (Figure HepG2 cells incubated with only 8NP, or DCFH-DA and 8NP without light irradiation showed almost no green fluorescence, the with 8NP for 6 h and then treated with DCFH-DA before light irradiation of 10 min showed green fluorescence from the product of the In the green fluorescence not when the cells were with a 1O2 at the identical Therefore, these results the ROS generation ability of 8NP in living cells. In we further the 1O2 of 8NP in HepG2 cells using the which demonstrates that 1O2 a in tumor cells ( Supporting Information Figures and Figure 5 | Confocal fluorescence images of ROS generation in HepG2 cells after incubation with 8NP (5 μg mL−1) using DCFH-DA as an indicator. Download figure Download PowerPoint We evaluated the cell by 8NP using DAPI as an indicator ( Supporting Information Figure The cell when the cells were incubated with 8NP without By contrast, when the cells were incubated with 8NP after light the have Moreover, the of 1O2 Therefore, the results that 8NP tumor cell proliferation and 1O2 are the to tumor cell To into the mechanism of tumor cells, the effect of 8NP on cell was by flow cytometry ( Supporting Information Figure cell apoptosis was detected and a apoptosis was observed after cells were treated with 8NP for 24 h, suggesting that 8NP tumor cells by the of cell apoptosis. Smart for photodynamic therapy In PDT, are to in the dark for a of time after PDT treatment to to effective strategy to PDT action rapidly after PDT treatment is highly for research and The reversibly ISC the development of such PDT that the ISC can be switched off by disaggregation to ROS generation and to turn on the fluorescence after PDT treatment. For this compound 4 was as a smart PS for a proof-of-concept application in PDT by ROS (Figure We compound 4 into hydrophobic to form nanoparticles fluorescent dyes into PSs. The reversible switch between fluorophore and PS was achieved by using the reaction of the BODIPY 4 with a high concentration of to generate that are to into This strategy could the general of PSs with and the to the PS through fluorescent Figure 6 | (a) Schematic illustration for the of and the of into in the presence of (b) The absorption spectra of aqueous dispersion of in the of (5 in PBS buffer at 37 (c) Absorption spectra of DPBF and aqueous dispersion of upon light irradiation. (d) Absorption spectra of DPBF and the reaction mixture of aqueous dispersion of with (5 for 24 h upon light irradiation. (e) Cell viability of HepG2 cells incubated with for 4 or 12 h upon light irradiation. Download figure Download PowerPoint The absorption of in the presence of (5 was measured at 37 °C (Figure As the reaction the original absorption band of 4 at nm a new absorption at nm increased which the formation of The formation of to into to the of bright fluorescence ( Supporting Information Figure In contrast, absorbance spectra of in the presence of 5 μM of with time ( Supporting Information Figure that was stable in The singlet oxygen generation capability of before and after with was As in the presence of the absorption of DPBF decreased upon light indicating efficient 1O2 generation (Figure After with the change of the DPBF absorption was negligible (Figure indicating that they have no 1O2-generating ability when is Confocal microscopy was employed to the process in living cells ( Supporting Information Figure HepG2 cells incubated with for 4 h showed weak fluorescence, the fluorescence was observed when the incubation time was increased to 12 h. To the ROS generation in tumor cells, the of was in HepG2 cells (Figure After incubation with for 4 h, satisfactory photocytotoxicity was In contrast, when the incubation time was to 12 h, the showed negligible upon light irradiation.

Topics & Concepts