A Long‐Term Plasma Membrane‐Anchored Photosensitizer for Pyroptosis‐Induced Anticancer Therapy

ABSTRACT

The development of precise and stable membrane‐anchoring photosensitizers is important but challenging for pyroptosis‐mediated antitumor therapy. Here, we report a tumor cell membrane‐anchored BODIPY photosensitizer (TPMS) capable of inducing plasma membrane rupture via the pyroptosis pathway. Its modular structure integrates a BODIPY core and phenothiazine unit to boost reactive oxygen species (ROS) generation, a quaternary ammonium salt for membrane anchoring, and a disulfide bridge for covalent immobilization. In aqueous solution, TPMS self‐assembles into nanoaggregates with intense absorption at 658 nm and efficient type‐I ROS production. Notably, TPMS was covalently immobilized on plasma membranes with a prolonged retention time of at least 2 h and effectively induced pyroptosis of cancer cells upon light irradiation. Furthermore, in vivo studies confirmed its potent tumor eradication efficacy with minimal systemic toxicity. This work not only provides a rational molecular strategy for constructing long‐term membrane‐localized photosensitizers but also highlights the potential of pyroptosis‐mediated anticancer treatment.

A BODIPY‐based photosensitizer (TPMS) that integrates a phenothiazine unit for enhanced ROS generation, a quaternary ammonium salt for electrostatic membrane affinity, and a disulfide‐bridged alkyl chain for covalent immobilization on tumor cell membranes has been developed, which enables pyroptosis‐mediated antitumor therapy in vitro and in vivo.

1. Introduction

Pyroptosis, a gasdermin‐mediated programmed cell death defined by plasma membrane rupture and pro‐inflammatory cytokine release, represents a promising avenue for antitumor therapy [1, 2]. Compared to the non‐inflammatory nature of apoptosis, pyroptosis is a more advantageous pathway for cancer therapy due to its ability to activate antitumor immune responses by excreting inflammatory cytokines and tumor antigens, thereby creating a potent “hot” tumor microenvironment with a high density of lymphocyte infiltration. Furthermore, pyroptosis offers a potential bypass to overcome apoptotic resistance [3, 4, 5, 6]. However, conventional pyroptosis activators, often chemotherapeutic drugs, lack selectivity for cancer cells and face significant challenges such as drug resistance and severe side effects [7, 8, 9]. Therefore, developing noninvasive approaches that can selectively trigger tumor pyroptosis is highly desirable for advancing cancer therapy.

Photodynamic therapy (PDT) has emerged as a promising antitumor modality due to its non‐invasiveness, spatiotemporal controllability, and capacity to circumvent drug resistance [10, 11, 12, 13, 14]. Its ability to generate reactive oxygen species (ROS) can also potentially activate the inflammasome, thereby inducing pyroptosis for tumor eradication [15]. Owing to the short lifetime and limited diffusion distance of ROS within cells, the development of organelle‐targeted photosensitizers is crucial to enhance the efficacy of pyroptosis‐mediated anticancer therapy [16, 17, 18]. Among various cellular organelles, the plasma membrane that serves as a critical barrier for maintaining cell integrity and mediating intercellular communication is a promising target for inducing pyroptosis [19, 20, 21, 22, 23, 24, 25]. Recent studies have demonstrated that membrane‐targeted photosensitizers trigger pyroptosis by generating ROS, which subsequently damages the membrane and activates N‐terminal gasdermin D (N‐GSDMD) pore formation. Compared with organelle‐targeting photosensitizers, the cell membrane‐anchored PDT is far superior, as it not only bypasses the internalization step but also leverages the tumor‐specific membrane landscape for reducing off‐target effects [26, 27, 28, 29, 30, 31, 32]. In addition, the direct generation of ROS at the plasma membrane evades scavenging by intracellular antioxidants such as glutathione, thereby increasing killing efficiency. Despite being a good target for initiating pyroptosis, the key challenge still lies in achieving precise and stable membrane anchoring of photosensitizers. To date, various plasma membrane‐targeted photosensitizers have been developed [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44]. They typically harness electrostatic interactions to bind the negatively charged cell membrane and have demonstrated considerable anticancer efficacy. Nevertheless, their short residence time on the plasma membrane leads to a narrow therapeutic window. On the other hand, most existing agents are type‐II photosensitizers, which are often ineffective against the hypoxic tumor microenvironment [45]. The long‐time membrane‐anchored type‐I photosensitizers that can induce pyroptosis have rarely been reported [46].

In this study, we report the rational design of a long‐term membrane‐anchored type‐I photosensitizer (TPMS) for pyroptosis‐induced anticancer therapy (Figure 1). TPMS is composed of four key moieties, each contributing distinct functions: a phenothiazine unit and a BODIPY core work in concert to enhance intramolecular charge transfer for boosting ROS generation [47, 48, 49, 50]; a quaternary ammonium salt promotes membrane anchoring; and a disulfide‐bridged alkyl chain enables covalent immobilization on cell membranes. In aqueous solution, the amphiphilic TPMS self‐assembled into uniform nanoaggregates with a diameter of ∼156 nm, which displayed intense absorption/emission peaks at 658/665 nm and efficiently generated type‐I ROS. Because of the more negative membrane potential and higher thiol‐disulfide exchange (TDE) reactivity of tumor cell membranes [51], TPMS selectively targeted and covalently immobilized on plasma membranes, resulting in a prolonged retention time of at least 2 h without internalization. Notably, it was not only able to illuminate cancer cell membranes but also, upon light irradiation, effectively eliminated them by inducing pyroptosis. In vivo therapeutic studies further verified its potent ability to eradicate tumors with minimal systemic toxicity. This work provides a new strategy for constructing membrane‐anchored pyroptosis‐inducing agents for tumor ablation.

FIGURE 1.

Rational design of the membrane‐anchored photosensitizer (TPMS) for pyroptosis‐induced anticancer therapy.

2. Results and Discussion

2.1. Design, Synthesis, and Characterization of Membrane‐Anchoring BODIPY‐Based Photosensitizers

Aiming to target cell membranes, we first synthesized two amphiphilic BODIPY dyes, namely BPMC₁₀ and TPMC, by installing a hydrophilic quaternary ammonium salt and a hydrophobic alkyl chain on the skeleton of the BODIPY core (Figure 2A,B). Although both compounds exhibited a similar absorption peak at ∼640 nm, TPMC with a meso phenothiazine group showed significantly quenched fluorescence at 665 nm (Figure S1). We next examined their ability to generate ROS using 1,3‐diphenylisobenzofuran (DPBF). Upon light irradiation, the absorbance DPBF at 410 nm remained unchanged in the presence of BPMC₁₀ , while a dramatic decrease of the DPBF absorption over a time period of 0–180 s was observed for the TPMC group (Figure 2C,D). To understand this phenomenon, time‐dependent density functional theory (TD‐DFT) calculations were used to analyze the frontier molecular orbitals of BPMC₁₀ and TPMC, including the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). In BPMC₁₀ , the delocalized electron density of the HOMO and the LUMO was both distributed on the BODIPY skeleton, indicating that no charge transfer occurs. In TPMC, the HOMO is localized on the electron‐rich phenothiazine moiety, while the LUMO is distributed on the electron‐deficient BODIPY core, indicating strong intramolecular charge transfer (Figure 2E,F). This distribution enables photoinduced electron transfer (PET) in the excited state, thereby quenching the fluorescence of TPMC (Figure 2F). Compared to BPMC₁₀ , the theoretical calculations show that TPMC has a significantly reduced energy gap between the S₁ and T₂ states (ΔE_S1T2), more easily triggering the intersystem crossing (ISC) (Figure 2G,H). These results demonstrate that the introduction of the electron‐rich phenothiazine unit at the meso position of BODIPY evidently promotes the ISC process and boosts the ability to produce ROS.

FIGURE 2.

Chemical structure of BPMC₁₀ (A) and TPMC (B). Time‐dependent absorption spectra of 50 µM DPBF in DCM solution in the presence of 5 µm BPMC₁₀ (C) or TPMC (D) upon irradiation at 650 nm (300 mW/cm²). The frontier molecular orbital energy level diagram of BPMC₁₀ (E) and TPMC (F) optimized at the TD‐B3LYP/6‐31G(d) level. Singlet and triplet energy levels of BPMC₁₀ (G) and TPMC (H) in DCM calculated by Gaussian 09 at the TD‐B3LYP/6‐31G(d) level.

Typically, photosensitizers with optimal hydrophilicity‐lipophilicity balance can effectively target and immobilize on cell membranes. To further investigate the role of alkyl chains in precise membrane localization, we synthesized two additional BODIPY derivatives: TPMH (without the hydrophobic C10 alkyl chain) and TPMS (containing a disulfide bond) (Figure 3A). The photophysical properties of TPMH, TPMC, and TPMS were examined in THF/H₂O mixture solution using UV–vis absorption and fluorescence spectroscopy. All three compounds displayed nearly identical absorption maxima at approximately 640 nm and emission peaks at around 665 nm in THF (Figure 3B,C). With the gradual increase in water fraction, a slight red shift of the absorption band to ∼654 nm was observed (Figure S2A–C). Unlike TPMH, the fluorescence intensity of TPMC and TPMS decreased dramatically with increasing water content. This phenomenon revealed obvious aggregation‐induced quenching (AIQ) behavior originating from molecular self‐assembly (Figure 3D; Figure S2D–F). The self‐assembly behaviors were further demonstrated by dynamic light scattering (DLS) and transmission electron microscopy (TEM) analyses. TPMC and TPMS formed spherical nanoaggregates with average diameters of approximately 76 nm and 156 nm, respectively, whereas TPMH did not (Figure 3E–G). These nanoaggregates remained stable over 15 days, showing negligible changes in size (Figure S3). To further evaluate their ROS generation in aqueous environments, we employed singlet oxygen sensor green (SOSG) and dihydroergotamine 123 (DHR123) to detect singlet oxygen (¹O₂) and superoxide (O₂ ^•−), respectively. In comparison with TPMH, the nanoaggregates of TPMC and TPMS could more efficiently produce both type‐I and type‐II ROS (Figure 3H,I; Figure S4). Furthermore, TPMS retained high type‐I ROS‐generating capacity after binding with bovine serum albumin (BSA, as a model thiol‐containing protein) (Figure S5). Therefore, the superior nanoaggregate‐forming ability of TPMS endows it with favorable PDT performance under hypoxic conditions.

FIGURE 3.

(A) Chemical structures of TPMH, TPMC, and TPMS. (B) Absorption spectra of TPMH, TPMC, and TPMS (10 µm) in THF. (C) Fluorescence spectra of TPMH, TPMC, and TPMS (10 µM) in THF. (D) Normalized fluorescence intensity changes of TPMH, TPMC, and TPMS in THF/water mixed solutions as a function of water fraction (vol%). I₀ is the fluorescence intensity of each probe in THF. TEM image and DLS size distribution of TPMH (E), TPMC (F), and TPMS (G). (H) ¹O₂ generation of 10 µm TPMH, TPMC, and TPMS in a THF/H₂O solution (f_water = 90%) under continuous 650 nm laser (300 mW/cm²) using SOSG (5 µM) as an indicator. (I) O₂ ^•– generation of 10 µm TPMH, TPMC, and TPMS in a THF/H₂O solution (f_water = 90%) under continuous 650 nm laser (300 mW/cm²) using DHR123 (10 µm) as an indicator.

2.2. TPMS Achieved Long‐Term Membrane‐Anchored Imaging

Prior to confocal fluorescence imaging, the cytotoxicity of three photosensitizers was assessed using the CCK‐8 assay. All of them exhibited negligible cytotoxicity even at a concentration of 50 µm (Figure S6), demonstrating good biocompatibility. Besides, the TPMS nanoaggregates exhibited excellent photostability and remained stable under different pH conditions and in the presence of proteins (Figure S7). To investigate their efficacy in targeting cancer cell membranes, we incubated TPMH, TPMC, and TPMS with the commercial membrane dye DiO in HeLa cells (Figure 4A). After 10 min of incubation, we found the red fluorescence of BODIPYs overlapped well with the green emission of the membrane dye DiO, with a Pearson correlation coefficient of 0.82, 0.89, and 0.89, respectively, indicating that all three photosensitizers could sensitively target the plasma membrane (Figure 4B). However, TPMH and TPMC began to internalize after 20 min and were largely internalized with lost membrane boundaries by 60 min. By contrast, TPMS remained predominantly localized to the membrane throughout the time course (Figure 4C–F).

FIGURE 4.

(A) Schematic diagram of the workflow for probe‐cell membrane colocalization analysis. (B) Colocalization images of TPMH, TPMC, and TPMS (20 µm) with DiO (10 µm) in HeLa cells. Scale bar = 10 µm. (C) Fluorescence images of HeLa cells stained with TPMH, TPMC, and TPMS (20 µm) for 20 min. (D) Intensity profile of the white line across HeLa cells in (C). (E) Fluorescence images of HeLa cells stained with TPMH, TPMC, and TPMS (20 µm) for 60 min. (F) Intensity profile of the white line across HeLa cells in (E). Red channel: λ_ex = 638 nm, λ_em = 663–738 nm. Green channel: λ_ex = 487 nm, λ_em = 500–550 nm. Scale bar = 10 µm.

Oxidoreductases responsible for reducing disulfide bonds are overexpressed on the plasma membranes of various cancer cell lines [52]. To examine the covalent binding of disulfide bonds with membrane thiol proteins, HeLa cells were pretreated with tris(2‐carboxyethyl)phosphine (TCEP), a thiol‐reducing agent, to expose additional membrane thiol groups (Figure 5A) [53]. The TCEP treatment at 0.5–5 mm showed no influence on the survival of HeLa cells (Figure S8). Following 1 mm TCEP pretreatment, TPMH and TPMC were internalized significantly (nearly complete within 90‐min), whereas TPMS maintained robust plasma membrane labeling for up to 120 min with no observable internalization (Figure 5B–G). For instance, at 90 min post‐incubation, the Pearson correlation coefficient with the membrane dye DiO was only 0.39 for TPMH (largely internalized), while it remained as high as 0.89 for TPMS (still predominantly membrane‐localized) (Figure S9). Agarose gel electrophoresis further confirmed that TPMS readily formed a covalent conjugate with BSA within 1 h, while TPMH and TPMC showed no discernible binding even after 2.5 h (Figure S10). In addition, TPMS could more effectively label cancer cell membranes, showing 3.4‐fold higher fluorescence than that in RAW264.7 cells (Figure S11). Upon irradiation with a 650 nm laser (300 mW/cm², 3 min), TPMS induced potent and dose‐dependent cytotoxicity in HeLa cells, with only ∼10% cell viability remaining at 15 µm. In contrast, the same treatment only caused mild cytotoxicity in HEK‐293T normal cells (Figure S12). These results demonstrate that disulfide functionalization enables covalent binding to membrane thiol proteins, thereby prolonging the membrane residence time of the photosensitizer and conferring phototoxic selectivity toward cancer cells.

FIGURE 5.

(A) Schematic diagram of the workflow for cell imaging after TCEP treatment. (B, D, F) Confocal images of HeLa cells incubated with TPMH, TPMC, and TPMS (20 µm) for 60 min (B), 90 min (D), and 120 min (F). (C, E, G) Intensity profile of the white line across HeLa cells in (B), (D), and (F). Red channel: λ_ex = 638 nm, λ_em = 663–738 nm. Scale bar = 10 µm.

2.3. TPMS Could Efficiently Kill Tumor Cells via Pyroptosis‐Induced Cancer Cell Death

Given its excellent membrane‐anchoring and ROS‐generating capability, we next applied TPMS to cancer cells for photodynamic therapy. Benefiting from its excellent Type‐I ROS generation capacity in the nanoaggregated state, TPMS holds great potential to directly disrupt the cell membrane and efficiently eliminate cancer cells even under hypoxic conditions. To test this, cytotoxicity was assessed using the CCK‐8 assay. TPMS alone showed negligible toxicity toward HeLa cells in the absence of light. Upon irradiation with a 650 nm laser (0.3 W/cm², 3 min), TPMS potently inhibited HeLa cell proliferation in a dose‐dependent manner. Notably, even under hypoxic conditions, 20 µm of TPMS achieved over 85% cell killing (Figure 6A). To verify the production of ROS in HeLa cells by TPMS, 2',7'‐dichlorodihydrofluorescein diacetate (DCFH‐DA) and dihydroethidium (DHE) were used to detect total intracellular ROS and O₂ ^•−, respectively. As shown in Figure 6B, TPMS‐treated cells exhibited strong green fluorescence under both normoxic and hypoxic conditions, indicating significant ROS generation. Red fluorescence from DHE further confirmed the production of type‐I ROS. Furthermore, live/dead cell staining with calcein‐AM and propidium iodide revealed that distinct red fluorescence was exclusively observed in the TPMS‐treated group following light irradiation, consistent with the tumor cell viability assay (Figure 6C).

FIGURE 6.

(A) Under normoxic or hypoxic conditions, the viability of HeLa cells was assessed in the presence of TPMS at various increased concentrations, both with and without exposure to a 650 nm laser for 3 min at a power density of 300 mW/cm² (mean ± SD, n = 3). (B) Confocal fluorescence microscopy was used to image the overall production of ROS and O₂ ^•–generated by TPMS (30 µm) in HeLa cells under various conditions. For staining, DCFH‐DA (10 µm) was used to detect total ROS, while DHE (10 µm) was employed to detect O₂ ^•–. Scale bar = 50 µm. (C) Confocal fluorescence images of HeLa cells stained with calcein‐AM and propidium iodide after various treatments. Red channel: λ_ex = 547 nm, λ_em = 600–650 nm. Green channel: λ_ex = 487 nm, λ_em = 500–550 nm. Scale bar = 50 µm. (D) Morphological features of HeLa cells by scanning electron microscopy after various treatments. The dying cells were incubated with TPMS (30 µm) for 30 min and then irradiated at a power density of 300 mW/cm² for 3 min. (E) Cleavage of GSDMD was observed by Western blot assay. (F) Schematic diagram of TPMS‐induced pyroptosis under light irradiation. Statistical analysis was performed by one‐way ANOVA: ^*** p < 0.001.

To directly visualize the morphological alterations in cancer cells induced by TPMS, we employed both optical microscopy and scanning electron microscopy (SEM). Optical microscopy revealed characteristic pyroptotic morphology, such as membrane swelling and numerous bubble‐like protrusions, which were absent in controls (Figure S13). SEM critically identified membrane rupture and pore formation (Figure 6D). The induction of pyroptosis was further validated by western blot analysis, showing a significant upregulation of GSDMD exclusively in TPMS‐treated cells upon light irradiation (Figure 6E). These results demonstrate that TPMS is promising as a robust agent for pyroptosis‐induced cancer cell therapy (Figure 6F).

2.4. TPMS Significantly Inhibited Tumor Growth In Vivo

Inspired by its excellent performance in vitro, TPMS was subsequently applied to near‐infrared (NIR) fluorescence imaging‐guided in vivo antitumor therapy. As shown in Figure 7A, we first established a 4T1 breast tumor‐bearing mouse model and then intratumorally injected TPMS for imaging and therapeutic evaluation. TPMS rapidly lit up the xenograft 4T1 tumors, showing intense NIR fluorescence at 5 min post‐injection and reaching a plateau at 40 min post‐injection (Figure 7B; Figure S14). To investigate whether TPMS anchors to cancer cell membranes in vivo, tumor tissue sections were analyzed using confocal fluorescence microscopy. Compared to TPMC, TPMS showed superior membrane‐anchoring capability in tumors (Figure 7C; Figure S15). Next, the 4T1 tumor‐bearing mice were randomly divided into three groups: PBS with laser irradiation (PBS+L), TPMS alone, and TPMS with laser irradiation (TPMS+L). Following a 40‐min post‐injection interval for TPMS, the tumors were exposed to a 650 nm laser at 0.3 W/cm² for six minutes. Tumor progression was then monitored over 12 days by serial digital photography and volume measurements to evaluate growth and morphological alterations. In contrast to the rapid tumor progression observed in both the PBS+L and TPMS‐alone groups, the tumor growth was significantly suppressed in the TPMS+L group (Figure 7D,E). Excised tumors at the end of treatment further confirmed the potent phototherapeutic effect of TPMS (Figure 7F). Additionally, mice remained stable body weight throughout the treatment (Figure 7G). Histological analysis revealed extensive tumor cell necrosis exclusively in the TPMS+L group, with no observable abnormalities in major organs (Figure S16). Collectively, these findings indicate TPMS has excellent in vivo biocompatibility.

FIGURE 7.

(A) Schematic diagram of the schedule for in vivo treatment in 4T1 tumor‐bearing BALB/c mice. (B) Time‐dependent in vivo NIR fluorescence imaging of 4T1 tumor‐bearing mice after intratumor injection of TPMS (∼30 µL, 50 µm). Excitation: 640 nm, emission: Cy5.5 filter. (C) Confocal imaging of tumor tissue sections harvested from 4T1 tumor‐bearing mice. λ_ex = 638 nm, λ_em = 663–738 nm, scale bar = 10 µm. (D) Representative photographs of 4T1 tumor‐bearing mice after different treatments. (E) Relative tumor volume changes, (F) digital photographs of harvested tumors, (G) relative body weight changes after different treatments (mean ± SD, n = 5). Statistical analysis was performed by one‐way ANOVA: ^*** p < 0.001; NS, p > 0.05.

3. Conclusion

In summary, we have successfully developed a tumor cell membrane‐anchored photosensitizer (TPMS) for pyroptosis‐mediated photodynamic therapy. The molecular design, which integrates the BODIPY core with an electron‐rich phenothiazine group, a disulfide‐bridged tail, and a quaternary ammonium moiety, enables TPMS to self‐assemble into stable nanoaggregates in aqueous solution with superior type‐I ROS generation. The disulfide bond allows covalent attachment to the plasma membrane proteins with prolonged membrane retention. Upon irradiation, TPMS induced membrane pore formation, GSDMD upregulation, and classic pyroptotic cell death, even under hypoxia. In vivo therapeutic studies further verified that TPMS has potent antitumor efficacy with minimal systemic toxicity. Overall, this study provides a new paradigm for designing membrane‐anchored photosensitizers with long‐lasting localization and type‐I photodynamic features, offering a promising platform for future pyroptosis‐mediated photoimmunotherapy.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File: smll73892‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.