Sonosensitized Aggregation-Induced Emission Dots with Capacities of Immunogenic Cell Death Induction and Multivalent Blocking of Programmed Cell Death-Ligand 1 for Amplified Antitumor Immunotherapy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Feb 2022Sonosensitized Aggregation-Induced Emission Dots with Capacities of Immunogenic Cell Death Induction and Multivalent Blocking of Programmed Cell Death-Ligand 1 for Amplified Antitumor Immunotherapy Shaorui Jia†, Zhiyuan Gao†, Zelin Wu, Heqi Gao, He Wang, Hanlin Ou and Dan Ding Shaorui Jia† Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 †S. Jia and Z. Gao contributed equally to this work.Google Scholar More articles by this author , Zhiyuan Gao† Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 †S. Jia and Z. Gao contributed equally to this work.Google Scholar More articles by this author , Zelin Wu Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 Google Scholar More articles by this author , Heqi Gao Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 Google Scholar More articles by this author , He Wang Department of Urology, First Affiliated Hospital of Soochow University, Suzhou 215006 Google Scholar More articles by this author , Hanlin Ou Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 Google Scholar More articles by this author and Dan Ding *Corresponding author: E-mail Address: [email protected] Key Laboratory of Bioactive Materials Ministry of Education, College of Life Sciences, State Key Laboratory of Medicinal Chemical Biology, Nankai University, Tianjin 300071 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101458 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The combination of immunogenic cell death (ICD) induction and immune checkpoint blockade has emerged as a major direction of cancer immunotherapy. Among currently available ICD inducers, sonosensitizers that produce reactive oxygen species (ROS) under an external trigger to evoke ICD of tumor cells have shown great promise. However, a highly efficient sonosensitizer-based ICD inducer with an aggregation-induced emission (AIE) characteristic has yet to be developed. Herein, a novel AIE sonosensitizer with a twisted molecular structure, very small energy gap between the singlet and triplet excited states (ΔEST), and efficient ROS generation ability, which can serve as an effective ICD inducer, is reported for sonodynamic processes in cancer immunotherapy. Furthermore, an AIE sonosensitizer-based nanosystem with surface modification of anti-PD-L1 peptide is constructed for boosting antitumor immunotherapy. In this system, AIE sonosensitizer-mediated sonodynamic therapy can successfully convert a hypoimmunogenic cold tumor to a hot one and further facilitate the multivalent blocking of programed death ligand (PD-L1) by anti-PD-L1 peptides. Such an advanced nanosystem could effectively initiate the activation of antitumoral immune reactions and modulation of an immunosuppressive microenvironment, contributing to systemic antitumor effects to further inhibit the growth of distant tumors. Download figure Download PowerPoint Introduction Antitumor immunotherapy, which boosts specific cytotoxic T cells to eliminate tumor cells, is recognized as an effective cancer treatment strategy.1 Currently, inducing immunogenic cell death (ICD) of tumor cells is an effective strategy for enhancing recruitment and infiltration of specific cytotoxic T cells into solid tumors (e.g., triple negative breast cancer), converting a cold tumor, with a paucity of T cell infiltration, to hot.2–4 During ICD of tumors, tumor-associated antigens (TAAs) and damage-associated molecular patterns (DAMPs) are generated, featuring surface-exposed calreticulin (ecto-CRT), adenosine triphosphate (ATP) secretion, and release of high-mobility group protein B1 (HMGB1) and heat shock protein 70 (HSP70).5,6 DAMPs act as a vital signal and a natural adjuvant to stimulate the presentation of TAA by antigen-presenting cells (APCs) (such as dendritic cells (DCs)) to T cells, which lead to the further activation of tumor-specific T cell-mediated immune response.7,8 Because of the pivotal role of ICD in antitumor immunotherapy, the development of highly effective ICD inducers with few side effects has attracted great interest during the past decade.5,9 Among currently available ICD inducers, photosensitizers evoke ICD of tumor cells by producing reactive oxygen species (ROS) upon light irradiation, which has showed conspicuous spatiotemporal precision and excellent biocompatibility.10 Our previous work has demonstrated that an aggregation-induced emission (AIE) photosensitizer with high ROS production capacity can massively induce sufficient ICD to evoke antitumor immunity.11,12 However, the limited penetration depth of light is a major challenge for the application of photosensitizer-based photodynamic therapy (PDT) in vivo.13 Sonodynamic therapy (SDT), derived from PDT, which utilizes ultrasound (US) to activate the sonosensitizer to produce ROS, enables deep tissue penetration, provides a more effective and safer strategy for the treatment of deep tumors, and shows unique advantages and great potential in antitumor immunotherapy.14–16 The mechanism of SDT is a complex, combinational output of different mechanisms.17–19 Among them, sonoluminescence was proven important to generate ROS.15,20 Sonoluminescence refers to the light emitted by the collapse of a bubble that is generated by US irradiation, which can activate the energy-matching photoactive sonosensitizers and generate the same photochemical reaction like PDT.21,22 For traditional organic photosensitizers with planar molecular structure, strong intermolecular interactions (e.g., π–π stacking) facilitate a nonradiative pathway of excited states, and the aggregation of photosensitizers within nanoparticles (NPs) causes quenching of the emission and ROS generation.23,24 At present, many organic photosensitizers that can be used as sonosensitizers to produce ROS, kill tumor cells, and induce ICD in tumor cells have been reported in SDT.25–27 Similar to photosensitizers, the aggregation-caused quenching (ACQ) effect also occurs when they are used as sonosensitizers in the aggregate state.28,29 Fortunately, sonosensitizers with AIE properties provide a good solution for the ACQ problem.30,31 AIE sonosensitizers have peripheral intramolecular motion units (e.g., rotating benzene rings) and a three-dimensional (3D) molecular structure.32,33 The 3D molecular structure significantly reduces the intermolecular interactions in NPs and aggregates. Meanwhile, the steric hindrance restricts the molecular motion in the excited state, which allows as much absorbed excitation energy as possible to be used in fluorescence emission or ROS generation.34,35 To the best of our knowledge, however, the ICD effect of AIE in SDT has never been studied. The reason may be the lack of appropriate strategies and AIE sonosensitizers with high efficiency. The application of SDT in immune system activation also faces challenges. Therefore, the design of an efficient AIE sonosensitizer is urgently needed to explore whether AIE can build an effective platform for sonosensitivity-based ICD inducers. Additionally, after successfully eliciting antitumor immune response by the induction of ICD, tumors may express autoprotective checkpoint molecules such as programmed cell death-ligand 1 (PD-L1) to bind with programmed death-1 (PD-1) on activated T cells, escaping attack from the immune system.36–38 Thus, an immune checkpoint blockade (ICB), such as the PD-1/PD-L1 blockade, has been employed to block PD-1/PD-L1 interaction to improve the tumor cell killing effect of T cells, thereby enhancing the inhibition efficacy against ICD-induced hot tumors.39–43 Although ICB holds great promise in cancer treatment, its therapeutic effect is largely limited by the insensitive response and insufficient T cells in cold tumors.44,45 Moreover, the synergistic anticancer efficacy of the integrated ICD induction and ICB is superior to monotherapy with only one of them.37,46,47DPPA (CNYSKPTDRQYHF, D-type PD-L1 peptide antagonists) can effectively block the PD-1/PD-L1 pathway, offering salient advantages such as higher stability, lower cost, and easier modification as compared with clinical anti-PD-L1 monoclonal antibodies.48,49 Consequently, for tremendously amplifying the effect of antitumor immunotherapy, the strategy of combing anti-PD-L1 peptide and ICD inducer in a single nanosystem is valuable and highly desirable but currently rare.50,51 Herein, we report sonosensitized AIE ICD inducer-based dots modified with anti-PD-L1 peptides on the surface for considerable improvement of antitumor immunotherapy outcomes. We designed and synthesized three new AIE luminogens (AIEgens): diphenylamino (DPA)-tetraphenylethylene (TPE)-(4-styryl-cyano)pyridinium salt (SCP), triphenylamine (TPA)-Ph-SCP, and TPA-2Ph-SCP). They have D–π–A structure and abundant intramolecular movement unit. Compared with TPA-Ph-SCP and TPA-2Ph-SCP, DPA-TPE-SCP has a more twisted molecular structure and very small ΔEST, which lead to stronger AIE activity, weaker intermolecular interactions of aggregates, and more efficient ROS generation. By comparison, DPA-TPE-SCP is not only the best photosensitizer, but also the best sonosensitizer. AIE sonosensitizers under US excitation showed superior ICD induction capability when compared with the reported sonosensitizing ICD inducer hematoporphyrin monomethyl ether (HMME), and successfully transformed a cold tumor into a hot one that was sensitized to PD-L1 blocking in vivo. Subsequently, surface modification of sonosensitized AIE dots with anti-PD-L1 peptides endowed them complementary advantages. In vivo experiments demonstrate that such function-cooperative dots significantly enhanced immune responses and relieved immune suppression, triggering a systemic antitumor immune therapeutic effect by virtue of the excellent ICD induction and multivalent PD-L1 blockade. Thus, this work not only provides a molecular guideline to design advanced sonosensitizer-based ICD inducers but also introduces new insights into the combination of ICD induction and ICB for synergistic antitumor immunotherapy (Scheme 1). Scheme 1 | Synthesis of AIE-dots-DPPA and the mechanism of antitumor immune responses induced by AIE-dots-DPPA-mediated sonodynamic therapy. Download figure Download PowerPoint Experimental Methods Computational method Structures of TPA-Ph-SCP, TPA-2Ph-SCP, and DPA-TPE-SCP were optimized in the water phase with the B3LYP method and 6-311G (d, p) basis set, Gaussian 09 program. Energy levels of S1–S6 and T1–T6 were calculated by the vertical excitation of the above optimized structures, with the same method of B3LYP/6-311G (d, p). Preparation of nano-sonosensitizer Nano-sonosensitizer was formulated by a nanoprecipitation approach using an amphiphilic co-polymer, DSPE-PEG2000. In brief, sonosensitizer (1 mg) and DSPE-PEG2000 (3 mg) were dissolved in 1 mL of tetrahydrofuran (THF); the solution was added dropwise into 9 mL of water undergoing sonication (3 min) by a microtip probe sonicator (XL2000, Misonix). An air pump was subsequently used to volatilize THF in the mixture to obtain a sonosensitizer-encapsulated dot aqueous solution. Then, the sonosensitizer-encapsulated dot solution was purified by 5000 rpm ultrafiltration for 10 min and filtration using a 0.45 μm syringe driven filter. Preparation of DPPA-conjugated dots DPPA peptides were modified on the AIE dots. First, DPPA peptides were added to the dots suspension, and then the suspension was stirred for 12 h to couple the thiol group of the peptide with the maleimide group of PEG2000 via the click reaction. Peptides that were not conjugated to the dots were then removed by centrifugation. ROS detection in 4T1 cells The 4T1 cancer cells were incubated in special confocal chambers with each nano-sonosensitizer (10 μg/mL of DPA-TPE-SCP dots, 10 μg/mL of HMME NPs) for 4 h at 37 °C. Subsequently, the cells were washed with 1× phosphate-buffered saline (PBS) three times, and then incubated with 2′,7′-dichlorodihydrofluorescein diacetate (DCF-DA; 20 μM) in FBS free culture medium. The above-mentioned process was performed in the dark. Then, under US exposure (1 MHz, 50% duty cycle, 0.5 W/cm2, 3 min), the cells were imaged with confocal laser scanning microscopy (CLSM) for DCF detection (Ex: 488 nm, Em: 530 ± 20 nm). Ecto-CRT staining in 4T1 cancer cells After the 4T1 cancer cells were incubated with each ICD inducer (10 μg/mL of DPA-TPE-SCP dots, 10 μg/mL of HMME NPs) for 4 h at 37 °C, the cells were washed and irradiated by US for 3 min (1 MHz, 50% duty cycle, 0.5 W/cm2). After 12 h, the cells were washed by precooling 1× saline, fixed with 4% paraformaldehyde on ice for 20 min, and then successively incubated with anticalreticulin antibody (ab2907, 1:200 dilution with 1× PBS) for 2 h and stained with Alexa Fluor 633-conjugated secondary antibody (1:200 dilution with 1× PBS) for another 2 h. The ecto-CRT expression of each ICD inducer-treated cancer cell was visualized by the ecto-CRT immunofluorescence using CLSM with the excitation at 633 nm and signal acquisition in the range from 640 to 670 nm. Detection of extracellular ATP, HMGB1, and HSP70 The 4T1 cancer cells (1 × 105 cells mL−1) were cultured in a black 12-well plate. After adherence, the 4T1 cancers cells were incubated with 10 μg/mL of DPA-TPE-SCP dots and 10 μg/mL of HMME NPs for 4 h at 37 °C, respectively. Next, the treated cells were washed and irradiated by US for 3 min (1 MHz, 50% duty cycle, 0.5 W/cm2). After 12 h, the supernatants were collected and then subjected to centrifugation using 12,000 rpm for 10 min at 4 °C, which was followed by addition of protease and phosphatase inhibitors into the supernatants. Furthermore, the cancer cells in each group were harvested and lysed, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal control. Finally, the extracellular levels of HMGB1 and HSP70 were analyzed by western blot with the anti-HMGB1 antibody (1:500, Abcam, ab79823) and anti-HSP70 antibody (1∶1000, Abcam, ab181606), respectively. The level of secreted ATP was quantitatively determined by ATP Bioluminescent Assay Kit per manufacturer’s instruction. In vivo SDT treatment 4T1 cells (1 × 106) dispersed in 50 μL of 1× PBS were injected into the second breast fat pad on the right side of each BALB/c mouse. The mice bearing 4T1 tumors were randomly divided into four groups when their tumors reached about 50 mm3. The AIE dots were intravenously injected into the mice three times on day 2, 4 and 6, respectively (once a day for three days (2, 4, and 6)). The tumors were exposed to US irradiation (1.0 MHz, 50% duty cycle, 1.5 W/cm2, 5 min) at 6-h post each injection. The tumor volume was measured by a caliper and determined with the following formula: Volume = Width2 × Length/2. The mice with tumors reaching 1500 mm3 were euthanized due to the standard animal protocol in this work. To establish a bilateral 4T1 orthotopic tumor model, 4T1 cells (1 × 106) were injected into the second breast fat pad on the right side of each female BALB/c mouse as primary tumor. Four days later, 4T1 cells (2 × 105) were injected into the second breast fat pad on the left side of mice as a distant tumor. The mice bearing ∼50 mm3 of primary tumors were randomly divided into four groups. The AIE dots were intravenously injected into the mice three times on day 2, 4 and 6, respectively (once a day for three days (2, 4, and 6)). At 6 h after each injection, the tumors were exposed to US irradiation for 5 min (1.0 MHz, 50% duty cycle, 1.5 W/cm2). The tumor volume was measured by a caliper and determined with the following formula: Volume = Width2 × Length/2. Flow cytometry analysis At designated time points, lymph nodes and tumors were harvested and homogenized to single-cell suspensions. After that, cell suspensions of lymph nodes were co-stained with anti-CD11c-FITC, anti-CD86-APC, and anti-CD80-PE for flow cytometry analysis of DCs. Single-cell suspensions in tumors were co-stained with different antibodies for flow cytometry analysis of CD8+ T cells (anti-CD3-FITC, anti-CD8-PE, or anti-CD8-APC), PD-L1+ tumor cells (anti-CD45-PE and anti-PD-L1-APC), Treg cells (anti-CD3-FITC, anti-CD4-APC, anti-CD25-PerCP/Cyanine5.5, and anti-Foxp3-PE), and IFN-γ+ of CD8+ T cells (anti-CD3-FITC, anti-CD8-APC, and anti-IFN-γ-PE). Statistical analysis Quantitative data are shown as mean ± standard deviation (SD). All the experiments were repeated at least three times. Statistical comparisons were made by ANOVA analysis and two-sample Student’s t-test. P value < 0.05 was considered statistically significant. Results and Discussion Design, synthesis, and characterization of AIEgens Based on our design strategy, the phenylethylene core was used as a π-linker, and compared to triphenylethylene and diphenylethylene, TPE is an excellent AIE skeleton. The D–π–A structure was constructed using a DPA block as an electronic donor (D) and the pyridinium as an electronic acceptor (A). The compounds DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP were designed and synthesized according to the synthetic route shown in Figure 1a. Compound 1 was prepared through a Horner–Wadsworth–Emmons reaction. After quenching with dimethylformamide (DMF), compound 1 containing bromine was functionalized to afford the corresponding aldehyde derivatives, and DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP were obtained from the Knoevenagel condensation of compound 2 and SCP under basic conditions (for detailed synthetic routes of all compounds see Supporting Information Scheme S1). The intermediates were characterized by 1H NMR, and DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP were characterized by 1H NMR, 13C NMR, and high-resolution mass spectrometry (HRMS) ( Supporting Information Figures S1–S17). Figure 1 | (a) Synthetic route to DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP. (b) Plot of ln(A0/A) against light exposure time, where A0 and A are the ABDA absorbance (378 nm) before and after irradiation, respectively. (c) Jablonski diagrams displaying the photophysical properties of the DPA-TPE-SCP. (d) HOMO–LUMO distributions by DFT calculations of DPA-TPE-SCP. (e) Energy levels of S1–S6 and T1–T6 calculated by the vertical excitation of the optimized structures in (f). (f) The dihedral angles of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP by DFT calculations. Download figure Download PowerPoint Photophysical properties The absorption and photoluminescence (PL) spectra of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP are shown in Supporting Information Figure S18. The conjugated structure of TPA-Ph-SCP, TPA-2Ph-SCP, and DPA-TPE-SCP becomes more and more distorted, so that their absorption peaks show hypochromatic shifts at 447, 382, and 350 nm; gradual red-shifting maximum emission wavelengths are located at 628, 630, and 638 nm. The PL quantum yields (QYs) of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP in the solid state were 12.4%, 7.9%, and 2.6%, respectively. When a poor solvent (toluene) was gradually added to the good solution [dimethyl sulfoxide (DMSO)], aggregate formation occurred, and the PL intensities of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP were significantly intensified, which prove that DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP have an obvious AIE feature. Notably the DPA-TPE-SCP showed the best AIE property when comparing the PL intensity of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP in a DMSO/toluene mixture with 99% toluene fraction to that in pure DMSO ( Supporting Information Figure S18). Because high ROS generation capacity is a prerequisite for an ICD inducer, we next studied and compared the ROS production of DPA-TPE-SCP, TPA-Ph-SCP, and TPA-2Ph-SCP (aggregated in a 90% water-DMSO solution, 10 μM) under white light irradiation (10 mW cm−2) with 9,10-anthracenediyl-bis(methylene)dimalonic acid (ABDA) as the ROS indicator. As shown in Figure 1b, the ABDA decomposition rate constants of DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP were 0.01088, 0.00747, and 0.00338 s−1, respectively, manifesting that DPA-TPE-SCP is an outstanding photosensitizer. Internal mechanism Based on the Jablonski diagram, the fixed absorption energy excited by light is dissipated through three competitive pathways: (1) fluorescence through radiation, (2) intersystem crossing (ISC) followed by production of ROS or phosphorescence, and (3) nonradiative thermal deactivation (Figure 1c). Weakening one or two energy dissipation pathways while concentrating as much absorbed energy as possible on the required dissipation pathways is an effective way to boost the efficacy of an optical material. To fully understand the mechanism of the superior ROS production ability of DPA-TPE-SCP compared to TPA-2Ph-SCP and TPA-Ph-SCP, density functional theory (DFT) calculations were conducted. The study on electronic structures in the ground state (S0) indicates that DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP exhibit a charge separation characteristic, with the highest occupied molecular orbital (HOMO) electron density majorly distributed on the electron-donating segments and the lowest unoccupied molecular orbital (LUMO) electron density mainly located on the electron-withdrawing moieties (Figure 1d and Supporting Information Figure S19). As depicted in Figure 1e, TPA-Ph-SCP has a quite large ΔEST [energy gap between the singlet (S1) and triplet excited state] of 0.5742 eV. However, using the TPE unit as the π-linker, DPA-TPE-SCP has a very small ΔEST value of 0.0429 eV. Due to the highly reduced ΔEST, ISC efficiency was remarkably improved, which accounts for the superior ROS generation capability of DPA-TPE-SCP. As shown in Figure 1f, the optimized geometric structures calculated by DFT reveal that compared with TPA-2Ph-SCP and TPA-Ph-SCP, DPA-TPE-SCP has a much greater 3D-twisted molecular structure. With a crowded TPE group used as a π-linker, the whole structure becomes more twisted. For example, the same dihedral angle between phenyl ring and vinyl increases to 48.02° and 48.11° in DPA-TPE-SCP from the introduction of phenyl groups ( Supporting Information Figure S20). The more twisted molecular structure would undoubtedly decrease the intermolecular interactions such as π–π stacking and nonradiative decay in the aggregated or solid state, making the saved absorbed excitation energy flow to the other two pathways (Figure 1c). Therefore, it successfully explains why DPA-TPE-SCP showed higher fluorescence QY and ROS generation efficiency than TPA-2Ph-SCP and TPA-Ph-SCP (there is no phosphorescence for DPA-TPE-SCP, TPA-2Ph-SCP, and TPA-Ph-SCP). Such excellent performance of DPA-TPE-SCP is attributed to our molecular design (D–π–A structure, introduction of intramolecular motion units on vinyl π-linker, 3D twisted molecular structure), which greatly promoted the ISC process and elevated the ROS generation efficiency. Preparation of AIE dots and detection of ROS in dots The AIE dots were prepared by the polymer-encapsulation method (Figure 2a). The amphiphilic polymer chain (DSPE-PEG2000) was used to change the biocompatibility. Transmission electron microscopy (TEM) images confirmed that DPA-TPE-SCP dots exhibited spherical morphologies with average diameters of about 53.5 nm, respectively (Figure 2b). Notably, dot formulation has negligible interference on the ROS generation ability of DPA-TPE-SCP, as evidenced by nearly the same ABDA decline rates of DPA-TPE-SCP (10 μM) solution and DPA-TPE-SCP dots (10 μM, based on DPA-TPE-SCP) upon light irradiation. ( Supporting Information Figure S21). As shown in Figure 2c, the previously reported TPE-DPA-TCyP dots (25 μg/mL) and DPA-SCP dots (25 μg/mL) with excellent ROS production and the commercially available sonosensitized compound HMME NPs (25 μg/mL) were used to evaluate the ROS production of DPA-TPE-SCP dots (25 μg/mL), TPA-2Ph-SCP dots (25 μg/mL), and TPA-Ph-SCP dots (25 μg/mL) under US irradiation (1 MHz, 50% duty cycle, 1.5 W/cm2). Interestingly, DPA-TPE-SCP dots showed higher ROS generation efficiency than the other sonosensitizers (Figure 2d). The result of US-excited ROS production matches the light-excited ROS production, indicating that the mechanism of DPA-TPE-SCP dots producing ROS may be activated by sonoluminescence. Next, electron spin resonance (ESR) was used to detect the species of ROS. The trapping agent 2,2,6,6-tetramethylpiperidine was mixed with water and DPA-TPE-SCP dots, respectively, followed by US irradiation; DPA-TPE-SCP dots showed an obvious peak (Figure 2e). For detecting •OH, 5,5-dimethyl-1-pyrroline-N-oxide was used as the trapping agent under 1O2 detection procedures, and the result indicated that the dots could also generate •OH efficiently post US irradiation (Figure 2f). Consequently, DPA-TPE-SCP dots were then selected as AIE dots for the following sonosensitized ICD induction experiments. Figure 2 | (a) The preparation of sonosensitized AIE dots. (b) DLS profile and TEM image (inset) of DPA-TPE-SCP dots. (c) The chemical structure of TPE-DPA-TCyP, DPA-SCP, and HMME. (d) Plot of ln(A0/A) against ultrasound (US) exposure time, where A0 and A are the ABDA absorbance (378 nm) before and after irradiation, respectively. (e) 1O2 generation by DPA-TPE-SCP dots and water with US radiation using ESR. (f) •OH generation by DPA-TPE-SCP dots and water with US radiation using ESR. Download figure Download PowerPoint In vitro SDT effect and ICD induction by AIE dots It has been verified that SDT can evoke ICD of cancer cells by producing ROS.52,53 First, internalization of DPA-TPE-SCP dots by murine 4T1 breast cancer cells within 4 h was analyzed by flow cytometry ( Supporting Information Figure S22). Next, the ROS generation level in 4T1 cells was evaluated by oxidation of DCF-DA by ROS to the green fluorescent DCF. Obviously, upon US irradiation for 3 min (1 MHz, 0.5 W/cm2), the CLSM images of the 4T1 cells treated with DPA-TPE-SCP dots showed the brightest green fluorescence (Figure 3a and Supporting Information Figure S23). Quantitatively, the average fluorescence intensity of “DPA-TPE-SCP dots + US” is ∼3.8-fold higher than that of “HMME NPs + US” ( Supporting Information Figure S24a), demonstrating the excellent ROS production capacity of DPA-TPE-SCP dots. Then, MTT [(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)] assay was used to detect the therapeutic effect of SDT on 4T1 cells in vitro (Figures 3b and 3c). The cytotoxicity of DPA-TPE-SCP dots in 4T1 cells was negligible at different concentrations without US exposure, revealing the great biocompatibility of DPA-TPE-SCP dots. After exposure to US for 3 min (1 MHz, 0.5 W/cm2), the viability of 4T1 cells decreased as the concentration of DPA-TPE-SCP dots increased. Furthermore, under different US irradiation time, the AIE dots-treated 4T1 cancer cells exhibited obviously more reduced cell viability than that of HMME NPs. All these results demonstrate that DPA-TPE-SCP dots can produce highly efficient ROS in 4T1 cells, suggesting a superior SDT effect triggered by DPA-TPE-SCP dots in vitro. Figure 3 | (a) CLSM images display the intracellular ROS levels in living 4T1 cancer cells after various treatments. DCF-DA was used as the ROS indicator. Green: DCF. Scale bar: 50 μm. (b) Cell viability of 4T1 cells incubated with different concentrations of DPA-TPE-SCP dots with or without US irradiation. (c) Cell viability of 4T1 cells after incubation with DPA-TPE-SCP dots and HMME NPs, followed by US irradiation with different time. (d) CLSM images of ecto-CRT (red fluorescence) on 4T1 c