Hybrid Nanospheres to Overcome Hypoxia and Intrinsic Oxidative Resistance for Enhanced Photodynamic Therapy

Photodynamic therapy (PDT) has been a well-accepted clinical treatment for malignant tumors and other diseases owing to its noninvasiveness and high spatiotemporal selectivity.1 PDT via systemic administration of a photosensitizer (PS) utilizes the localized reactive oxygen species (ROS) produced upon laser irradiation in the lesion area.2 The oxidative damage initiates cell apoptosis and shutdown of blood vessels and ultimately removes the nidus.3 However, the efficiency of current clinical PDT applications is hindered by multiple factors including an inherent aggregation-caused quenching (ACQ) effect of traditional PSs, in vivo tumor hypoxia microenvironment, and
intrinsic oxidative resistance.

A majority of the widely used PSs, such as porphyrin derivatives, tend to aggregate in aqueous media due to their rigid planar structures and hydrophobic nature. As a result of aggregation, quenched fluorescence and reduced ROS generation negatively affect the PDT efficacy.6 A new type of PS with bright fluorescence and high phototoxicity in an aggregate state is highly desirable. Apart from the inherent problems of PSs, intracellular oxygen consumption during PDT and distorted tumor blood vessels could induce a tumor hypoxia microenvironment, resulting in low ROS production and impaired PDT efficiency.7,8 Several agents such as catalases, MnO2, nanoscale metal−organic frameworks (MOFs), etc. were developed for increasing O2 concentration with the hope to enhance the PDT effect.9−15 Nevertheless, solely increasing O2 concentration is insufficient because antioxidant response could be induced by activating intra- cellular redox-sensitive transcription factors, causing upregula- tion of antiapoptotic pathways.16 Such survival mechanisms could cause cancer to be insusceptible to PDT and promote tumor microenvironment (TME) to contribute to tumor survival. Specifically, antiapoptotic proteins would be upregu- lated by tumors as an adaptive response to oxidative stress.17 Among various PDT resistance-related proteins, B-cell lymphoma 2 (Bcl-2) is the most typical one.18,19 Upon PDT treatment, the cancer cells respond to PDT by overexpressing Bcl-2, which yields higher cellular concentration of glutathione (GSH), shifting the intracellular redox potential to a more reduced state. As GSH reacts with ROS, the PDT resistance is produced. Therefore, applying Bcl-2 inhibitors will consume intracellular GSH to facilitate cancer cell apoptosis.

Figure 1. Chemical structure of TPEDCC, sabutoclax, and the formation of hybrid nanospheres. Schematic representation of the hybrid nanospheres taken up by tumor cells, Fe3+-activated Fenton reaction to increase intracellular O2 concentration. Upon laser irradiation at 410 nm, TPEDCC produces ROS at low intracellular PDT resistance mitigated by sabutoclax.

Figure 2. Synthetic route to AIE photosensitizer TPEDC−COOH (TPEDCC).

In contrast to porphyrin derivatives, such as clinically approved Hemoporfin, which exhibit quenched fluorescence and decreased singlet oxygen production in the aggregate state, PSs with aggregation-induced emission (AIE) show increased fluorescence and photosensitization ability in the nanoparticles, which are ideal for both tumor imaging and PDT.24−27 Herein, a type of carrier-free hybrid nanosphere containing AIE PSs was constructed to simultaneously overcome ACQ, hypoxia, and intrinsic oxidative resistance. The hybrid nanosphere was
fabricated through coordination-driven self-assembly of ferric ions (Fe3+), TPEDCC, an AIE PS with symmetric dicarboxylic acid group, and sabutoclax, a Bcl-2 inhibitor with a symmetric polyphenol group (Figure 1). Fe3+ ions were introduced to overcome tumor hypoxia by inducing ROS and O2 generation through a Fenton-like reaction.28−30 Meanwhile, to tackle the intrinsic oxidative resistance, sabutoclax was introduced with the hypothesis that inhibiting Bcl-2 activity can reduce the intrinsic oxidative resistance by decreasing the GSH level, triggering the release of cytochrome c, eventually boosting the efficiency of PDT.


Synthesis and Characterization of AIE Photosensi- tizer (TPEDCC). The AIE PS (TPEDCC) was synthesized from compound 2-((4′-(2,2-bis(4-methoxyphenyl)-1-phenyl- vinyl)-[1,1′-biphenyl]-4-yl)(phenyl)methylene)malononitrile (TPEDC) according to our previous work.31 First, TPEDC was demethylated in the presence of BBr3 to give compound 1 with a yield of 88%. Then, compound 1 was alkylated with tert- butyl 2-bromoacetate in the presence of cesium carbonate to give compound 2 with a yield of 81%. After ester hydrolysis in the presence of trifluoroacetic acid, the final product of TPEDCC was obtained with a yield of 78.6% (Figure 2). The chemical structure of TPEDCC was confirmed by 1H NMR, 13C NMR, and high-resolution mass spectroscopy (HR-MS) (Supporting Information). The absorption and emission spectra of TPEDCC in dimethyl sulfoxide/water (DMSO/ H2O) = 1/99 v/v were measured with peaks centered at 420 and 650 nm, respectively (Figures S1 and S2). Subsequently, the AIE property of TPEDCC was characterized in DMSO/ H2O mixtures with different ratios. Notably, the fluorescence intensity of TPEDCC was significantly enhanced upon aggregation formation, exhibiting deep red emission in an AIE-active manner (Figure S2). Furthermore, TPEDCC exhibits good photosensitizing capability, which was confirmed by the evaluation of its ability to bleach the absorption of 9,10- anthracenediylbis(methylene)dimalonic acid (ABDA) at 378 nm (Figure S3A,B).

Figure 3. (A) TEM image and size distribution analysis, (B) fluorescence spectrum of TPEDCC and a nanosphere, (C) Fe 2p XPS spectrum of a nanosphere, and (D) HAADF-STEM image of hybrid nanospheres, along with the corresponding EDS element maps of C, N, O, and Fe.

Figure 4. (A) Changes in the absorption of ABDA induced by ROS generation from nanospheres in water after laser irradiation (410 nm, 60 mW cm−2). (B) Accumulative drug (TPEDCC and sabutoclax) release from nanospheres in different buffers. (C) Confocal images of MDA- MB-231 tumor cells being incubated with 10 μM nanospheres (based on concentration of TPEDCC) for different time intervals. The blue channel indicates the Hoechst. The red channel is the emission from nanospheres. The scale bar is 50 μm. (D) MDA-MB-231 cell viability study after incubation with different formulations. (E) Confocal imaging of MDA-MB-231 cells upon incubation with different formulations with or without laser irradiation, which was further incubated with 20 μM of DCFDA for 30 min or a glutathione cell-based detection kit.

Preparation and Characterization of Hybrid Nanospheres. After studying the optical and photosensitizing properties, TPEDCC (5 mg mL−1 in DMSO) and sabutoclax (5 mg mL−1 in DMSO) were added dropwise into an aqueous solution of Fe3+ ions and stirred at 85 °C overnight. As shown
in the Fourier transform infrared (FTIR) spectrum (Figure S4), compared to free TPEDCC and sabutoclax, both carboxyl and phenolic hydroxyl groups were shifted, indicating the formation of a metal complex. After ultrafiltration, the hybrid nanospheres were obtained with a diameter of 50 ± 2.6 nm, as confirmed by both dynamic light scattering (DLS) and transmission electron microscopy (TEM) (Figure 3A). Then, the stability of the hybrid nanosphere was investigated by DLS in aqueous solutions at different time points. The result showed that the hybrid nanospheres are stable in aqueous solutions, as identified by the negligible increase of the particle size in stock solutions for 7 days (Figure S5). The optical property of as-synthesized nanospheres did not undergo obvious change as compared to that of TPEDCC (Figure 3B). To further characterize the hybrid nanospheres, X-ray photoelectron spectroscopy (XPS) measurements were con- ducted to determine the chemical state of Fe3+ ions in the nanospheres. As shown in the high-resolution XPS spectrum, the characteristic signal at 711.0 eV is attributed to Fe3+ (2p3/2, Figure 3C). Moreover, high-angle annular dark-field scanning TEM energy-dispersive X-ray spectroscopy (HAADF-STEM- EDS) elemental mapping confirmed the homogeneous distribution of carbon, nitrogen, oxygen, and iron in the nanospheres (Figure 3D). The 1O2 generation efficiency of hybrid nanospheres in aqueous media was also evaluated using ABDA as an indicator. As shown in Figure 4A and Figure S3B, the efficiency of nanospheres to decompose the ABDA signal was comparable to that of free TPEDCC. The efficient 1O2 generation and bright fluorescence make the hybrid nano- spheres an ideal candidate for tumor image-guided PDT.

To estimate the responsiveness of the hybrid nanospheres, the in vitro drug release behaviors were investigated in phosphate-buffered saline (PBS) at pH 7.4 and in acetate buffer at pH 5.0 and 6.0 at 37 °C to simulate different intracellular environments. The amount of dissociated TPEDCC and sabutoclax was calculated by high-performance liquid chromatography (HPLC) at scheduled time intervals. The result demonstrated that, in the neutral environment, a slower drug release was sustained over a prolonged duration. In contrast, the release rate in acidic conditions was much faster than that in neutral conditions, and more than 85% of the drug was released within 48 h (Figure 4B). It was concluded that accelerated drug release from nanospheres could be achieved by cleaving the weak supramolecular interactions (H-bond and π−π stacking) and coordination bond in an acidic tumor microenvironment, which is advantageous for antitumor applications.

In Vitro Enhanced PDT Performance of Hybrid Nanospheres. Before we investigated the PDT performance inside the cells, bioinformatics analysis was conducted through published profiles in the ONCOMINE database. Bcl-2 expression was significantly upregulated in breast cancer samples, which was further confirmed by immunohistochem- istry (IHC) staining of clinical breast cancer samples and breast paracancerous tissues (Figure S6). As a result, the breast cancer model was selected for the subsequent hybrid nanosphere PDT study both in vitro and in vivo. First, cellular uptake and in vitro PDT performance of the nanospheres were studied using MDA-MB-231 breast cancer cells as an example. As shown in Figure 4C, after incubating MDA-MB-231 cells with nanospheres (10 μM), the red fluorescence increases over time, indicating successful internalization of nanospheres into the cells (endocytosis). An in vitro cell viability test of MDA- MB-231 and 3T3 cells after being treated by different formulations was further conducted. Both nanospheres and TPEDCC performed negligible toxicity to both cell lines after 48 h incubation in the dark, presenting good biocompatibility of TPEDCC (Figure 4D). When cells were irradiated with a 410 nm laser, the group incubated with TPEDCC in combination with sabutoclax exhibited cell viability to MDA- MB-231 cells lower than that of the groups treated with TPEDCC only (Figure 4D). More importantly, nanospheres demonstrated the most superior cytotoxicity with an IC50 value of 4.67 ± 0.02 μM under laser irradiation when compared to that of other groups (Figure 4D). However, owing to the low expression level of Bcl-2 in normal 3T3 cell lines (Figure S7), such an enhanced PDT effect could not be observed in normal 3T3 cell lines (Figure S8). In addition, a cell apoptosis assay showed that nanospheres cause the highest apoptosis rate (61.6%) in MDA-MB-231 cells as compared to other groups, which is in good agreement with the cytotoxicity results (Figure S9). Next, a cellular ROS detection probe (2′,7′-dichlorodihydrofluorescein diacetate (DCFDA)) was used for staining MDA-MB-231 cells. DCFDA can detect intracellular ROS and then generate green fluorescence. Confocal image showed that bright green fluorescence was detected for the groups of MDA-MB-231 cells treated with nanospheres and TPEDCC under laser irradiation (Figure 4E). Quantitative flow cytometry analysis indicated that cells treated with hybrid nanospheres emitted the strongest green fluorescence. However, fairly weak green fluorescence was observed for the groups without laser irradiation (Figure 4E and Figure S10). Hence, the cytotoxicity, cell apoptosis, and intracellular ROS evaluation studies suggest that inhibiting antiapoptosis protein Bcl-2 may reduce the intracellular PDT resistance and improve the PDT efficiency.

To understand the enhanced PDT efficiency of nanospheres, the downstream protein of Bcl-2, cytochrome c expression level, was evaluated by Western blotting after MDA-MB-231 cells were treated with different agents for 24 h with or without laser irradiation. As shown in Figure S11, the expression of cytochrome c could be slightly upregulated by sabutoclax. When cells were treated with TPECC under laser irradiation, PDT itself could prompt the release of cytochrome c. The concentration of cytochrome c was significantly increased when TPEDCC was combined with sabutoclax under laser irradiation. For the nanosphere group under dark condition, a small amount of cytochrome c could be detected, which indicates that partial release of sabutoclax could not induce the oxidative stress sufficiently. Meanwhile, the hybrid nano- spheres contain ferric ions (Fe3+), and we inferred that the enhancement of PDT might be related to the synergistic effect of ferroptosis. To identify our hypothesis, Western blot analysis was conducted. The results demonstrated that one of the ferroptosis-related biomarkers, glutathione peroxidase (GPX4), was obviously downregulated after incubation of MDA-MB-231 cells with hybrid nanospheres, especially under laser irradiation (Figure S12). As intracellular GPX4 catalyzes the reduction of hydrogen peroxide, organic hydroperoxides, and lipid peroxides in the protection of cells against oxidative stress. Therefore, the downregulation of GPX4 could enhance the cellular oxidative stress, decrease the cellular reduced GSH, and promote the PDT effect. After that, the intracellular GSH was determined to monitor the oxidative stress upon incubation of the cells with different agents. For cells without any treatment, monochlorobimane inside the kit could react with the excess GSH to emit bright blue fluorescence. After incubation with nanospheres (10 μM) under laser irradiation, the blue fluorescence was obviously decreased (Figures 4E and S13). To evaluate the mechanism of GSH changes caused by our hybrid nanospheres, herein, MDA-MB-231 cells were treated with ferroptosis inhibitor (Ferrostatin-1) to do the rescue test. It was found that when MDA-MB-231 cells were treated by nanospheres under laser irradiation, intracellular GSH was obviously depleted. In contrast, when cell ferroptosis was inhibited by ferrostatin-1, the concentration of GSH could be enhanced but was still lower than that of the negative control (Figure S14). Therefore, it was concluded that nanospheres could inhibit Bcl-2 and initiate ferroptosis and then deplete the intracellular GSH. Moreover, to verify the O2- evolving ability of nanospheres, intracellularly dissolved oxygen was measured using the O2 consumption assay kit. As O2 could quench the phosphorescent signal of the probe inside the kit, the lower the phosphorescent signal, the higher the intra- cellular O2 concentration. As shown in Figure 5, the group incubated with nanospheres emitted the weakest red phosphorescence due to the O2 produced via the Fenton-like reaction inside the cells. In contrast, a bright red phosphor- escent signal was observed in other groups, indicating that O2 concentration is low in the tumor cells (Figure 5). The positive in vitro results demonstrated that our hybrid nanospheres have enhanced the PDT efficiency, making the tumor microenviron- ment susceptible to external oxidative stimuli.

Figure 5. Confocal images of MDA-MB-231 cells stained by O2 detection kit after incubation with TPEDCC, sabutoclax, TPEDCC/ sabutoclax drug mixture, and a nanosphere. The scale bar is 50 μm.

Figure 6. (A) In vivo fluorescence images of MDA-MB-231 breast-tumor-bearing mice after i.v. injection of the nanosphere (200 μL of 10 mg kg−1 based on TPEDCC). (B) Tissue distribution of TPEDCC after intravenous injection of nanospheres (10 mg kg−1), and data are shown as the average value ± standard error (n = 3).

In Vivo Enhanced PDT Performance of Hybrid Nanospheres. In vivo studies were further conducted with MDA-MB-231 tumor-cell-bearing nude mice to evaluate the effect of image-guided PDT. As is known, nanoparticles with an appropriate size usually perform passive targeting capacity and longer circulation time.32,33 After intravenous injection with the nanosphere (10 mg kg−1, 200 μL) into the tumor- bearing mice (Figure 6A), in vivo imaging showed that the signal of nanospheres accumulated at the tumor site reached the maximum at 4 h, and the signal remained strong at 6 h. Biodistribution profiles are also consistent with in vivo imaging results (Figure 6B). The in vivo imaging results demonstrate that the hybrid nanospheres exhibit decent tumor passive targeting ability and possess great potentials for image-guided PDT.
Subsequently, in vivo tumor growth inhibition was conducted to evaluate the PDT effect. After treatment with TPEDCC nanoparticles (Figure S15) (10 mg kg−1) and the nanosphere (10 mg kg−1) with or without laser irradiation (410 nm, 500 mW cm−2) every 3 days, tumor volume in both PBS and TPEDCC groups without laser irradiation increased from ∼100 to 800 mm3 and the body weight did not show any significant change, demonstrating that TPEDCC has negligible dark toxicity (Figure 7A,B). As indicated by hematoxylin/eosin (H&E) staining, the tumor tissues in both PBS- and TPEDCC-treated groups under dark were composed of extensive tumor cells without significant damage (Figure 7C). In contrast, after treatment with nanospheres in dark conditions, tumor growth could be slightly suppressed because the released sabutoclax could induce inhibition of tumor cell growth through inhibition of Bcl-2 (Figure 7A and Figure S16). For the TPEDCC group under laser irradiation, tumor growth was greatly inhibited (Figure 7A and Figure S16). The best treatment effect is observed with hybrid nanospheres under laser irradiation, where in vivo Bcl-2 expression was obviously decreased and tumor growth was almost entirely inhibited with little body weight change (Figure 7A,B and Figures S16 and S17). Meanwhile, a large quantity of dead cells with the loss of nuclei were observed in H&E staining, and the nanosphere group showed the highest apoptosis rate, as evidenced by the TUNEL assay (Figure 7C). To further verify the ability of hybrid nanospheres to overcome tumor hypoxia, fluorescent-dye-labeled antibody was used to stain tumor slices to detect the expression level of hypoxia-inducible factor-1α (HIF-1α). Compared to the control and TPEDCC groups, hybrid nanosphere-treated groups presented significantly decreased green fluorescence, demonstrating the reduced expression level of HIF-1α (Figure 7C), which is attributed to the intracellular O2 generation. Eventually, no significant difference and no damage in normal organs could be found in their blood biochemistry analysis after mice were treated with hybrid nanospheres (Figures S18 and 19). In addition, the concentration of aminotransferase and phosphatase underwent no obvious change after mice were treated with hybrid nanospheres (Figure S20), indicating that hybrid nanospheres did not harm the liver function. Overall, after hybrid nanosphere treatment under laser irradiation, in vivo enhanced PDT could be achieved with good biocompatibility.

Figure 7. (A) Tumor volume change after PBS, TPEDCC (10 mg kg−1, 200 μL), and nanosphere (10 mg kg−1 based on TPEDCC, 200 μL) treatment with or without laser irradiation (410 nm) in nude mice bearing MDA-MB-231, ***P < 0.001, demonstrated the statistical significance level. (B) Body weight changes of nude mice after treatment. Data are shown as average value ± standard error (n = 5). (C) H&E assay of tumor tissues, TUNEL assay of the tumor tissues. Green: apoptosis cell stained by TUNEL detection kit. Blue: cell nuclei stained by DAPI. Immunofluorescence staining for HIF-1α expression level assays. Green: HIF-1α. Blue: DAPI-stained cell nuclei. The scale bar is 100 μm. CONCLUSION In summary, we fabricated carrier-free hybrid nanospheres which integrated Fe3+, AIE PS (TPEDCC), and the Bcl-2 inhibitor (sabutoclax) into a single nanoplatform through coordination-driven self-assembly. The hybrid nanospheres exhibit bright fluorescence with high ROS efficiency under laser irradiation, making them the ideal candidates for image- guided PDT. When nanospheres were taken up by tumor cells, intracellular O2 concentration was increased via Fe3+ driving the Fenton reaction. In addition, intracellular PDT resistance of TPEDCC could be mitigated by sabutoclax and a ferroptosis signaling pathway. The positive in vitro and in vivo results indicate that the multifunctional hybrid nanospheres are promising nanoplatforms for image-guided enhanced PDT. METHODS Preparation of Hybrid Nanospheres. TPEDCC (5 mg mL−1 in DMSO) and sabutoclax (5 mg mL−1 in DMSO) were added dropwise into an FeCl3 (0.2 mM) aqueous solution, and the mixture was stirred at 85 °C for overnight. After that, the mixture was purified by ultrafiltration (Mw: 10.0 kDa) at 5000 rpm for 10 min to remove the organic solvents and unbound compounds. The stock solution was kept at room temperature in the dark. The drug ratio inside the nanospheres was determined by ICP-mass, ultraviolet−visible spectrophotometry, and fluorescence spectrophotometry. The size distribution was monitored by dynamic light scattering (90 Plus, Brookhaven Instruments Co., USA). The morphologies of the samples were studied by transmission electron microscopy (JEM- 2010F, JEOL, Japan).

In Vitro Cell Uptake and Cytotoxicity Study. MDA-MB-231 cells were cultured in confocal dishes for 24 h in DMEM culture medium. Hybrid nanospheres (10 μM) and Hoechst were added to each dish and incubated for various times. The fluorescence of Hoechst was excited with a 405 nm laser with the emission wavelength at 430−490 nm. The fluorescence of hybrid nanospheres was excited with a 405 nm laser with emission wavelength at 660−750 nm ((Leica)/TCS SP8 STED 3X).

For cytotoxicity study, MDA-MB-231 and 3T3 cells were cultured in 96-well plates overnight. Then, both cell lines were incubated with different groups at various concentrations and incubated for 72 h. Next, 20 μL MTT solutions (5 mg mL−1) were added to the 96-well plates. After 4 h incubation at 37 °C, the medium was replaced with 200 μL of DMSO. The obtained solution was measured in a Bio Tech
Synergy H4 at a wavelength of 570 nm.

Intracellular O2 Determination. Intracellular O2 was detected via an O2 consumption assay kit (Abcam). Briefly, cells were treated with various agents and then added with substrates in the kit according to the manufacturer’s protocol.In Vivo Pharmacodynamic Study. The tumor-bearing female nude mice were randomly divided into five groups (n = 5). Mice in different groups were intravenously injected via the tail vein with PBS (200 μL), TPEDCC nanoparticles (10 mg kg−1 based on AIEgen PS), and nanospheres (10 mg kg−1 based on AIEgen PS) with laser
irradiation (410 nm) for 10 min or without laser irradiation once every 3 days for 18 days. The length and width of the tumor and the body weight of mice were recorded before every injection during the therapy. The tumor volume was calculated using the following formula:
V (mm3) = 1/2 × length (mm) × width (mm)2Statistics Study. The data were demonstrated as the mean standard deviation (SD). The differences were identified by a Student’s t test. The difference was evaluated as statistical significance (*P < 0.05) and very significant (**P < 0.01, ***P < 0.001).