A neutrophil-tumor cascade-targeting Trojan horse for heterobifunctional prodrug delivery to enhance cGAS-STING cancer immunotherapy Authors: Huamiao Zhang, Yang Li, Qiuchen Bi, Heping Zhu, Huiwen Wang, Zhao Wang, Zhangqiang Ma, Huiping Liu, Xiaoling Zhu, Jiaqi Cao, Xin Wang, Zhe Tang, Xiaoyuan Chen, Longguang Tang, Kesong Peng Published: 2026 ABSTRACT AbstractBackgroundSince the discovery of the cGAS-STING pathway, attempts to utilize it as an anti-tumor immunotherapy have attracted significant research interest and investment. However, relevant clinical translation remains hindered by immune evasion and systemic toxicity.ResultsWe introduce BMSA, a first-in-class STING–PD-L1 heterobifunctional prodrug in which the PD-L1 inhibitor BMS-1 and the STING agonist MSA-2 are bridged by a tumor-cleavable linker. BMSA executes glutathione-triggered extracellular release of BMS-1 and intracellular esterase-mediated liberation of MSA-2, synchronizing dual immune signals at their respective sites of action. To confine activation to the tumor, we encapsulated BMSA into neutrophil-hitchhiking nanoparticles (T-NPs). After tail intravenous injection, T-NPs hijacked circulating neutrophils, accumulated at irradiated tumors via X-ray-induced inflammation, and exposed the fibrin-binding peptide CREKA through MMP-2/9 cleavage, producing markedly intratumoral accumulation while minimizing systemic exposure.ConclusionThis “Neutrophil–Tumor” cascade delivered heterobifunctional immunomodulation drugs with spatial and temporal precision, offering a translatable solution to the toxicity–efficacy dilemma that currently constrains STING-based cancer therapy.Graphical abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12951-026-04575-z.Keywords: Prodrug, STING agonist, PD-L1 inhibitor, Neutrophil hitchhiking, Cascade-targeting Delivery Background Since the 2006 discovery of cytosolic DNA sensing via cGAS-STING [1], this endoplasmic-reticulum-localized adaptor has emerged as the central innate immune hub [2]. Identification of STING as the direct receptor for bacterial cyclic di-GMP in 2013 ignited intensive efforts to harness its potent type-I interferon-inducing capacity for cancer immunotherapy [3, 4]. Since then, STING agonists have been viewed as the most promising immunomodulatory agents after immune-checkpoint blockers, drawing major research investment [5, 6]. First-generation agents, synthetic cyclic dinucleotides that mimic endogenous 2′3′-cGAMP [7] suffer rapid enzymatic degradation and negligible membrane permeability, restricting their use to direct intratumoral injection [8, 9]. The field advanced in 2018 with the first small-molecule STING agonist, ABZI, compatible with intravenous infusion [10]. Progress accelerated in 2020 when Pan et al. discovered MSA-2 by high-throughput screening of 2.4 million compounds; this orally bioavailable agonist permits flexible, repeated dosing and is now the benchmark for systemic STING-targeted therapy [11, 12]. Yet, despite a decade of investment, no STING agonist has progressed beyond phase II; DMXAA, once the most advanced candidate, was abandoned after it proved to be selective for murine STING [8, 13, 14]. Recognizing that STING activation can up-regulate PD-L1 and lead adaptive resistance [15, 16], ongoing trials now couple agonists with PD-1/PD-L1 antibody, but systemic toxicity remains [8, 17]. This rationale is twofold: Type I interferons elicited by STING prime T-cell infiltration and sensitize tumors to checkpoint blockade [18], yet the same pathway can transcriptionally up-regulate PD-L1, creating an adaptive immune escape circuit that must be simultaneously disrupted [19]. For example, SB-11285, currently in phase I clinical trials, demonstrated favorable tolerability and sustained antitumor immune responses when combined with the anti-PD-L1 monoclonal antibody atezolizumab in patients with advanced solid tumors [20]. Additionally, other clinical-stage STING agonists, such as SNX-281, KL340399, and BMS-986301, also adopt combination therapies with PD-1/PD-L1 inhibitors [8]. A second, equally formidable barrier is off-tumor toxicity: STING is ubiquitously expressed, so systemic exposure can ignite widespread type-I interferon release, precipitating dose-limiting cytokine-release syndrome and tissue-damaging inflammation in healthy organs [21]. For example, XMT-2056 caused a fatal case during its Phase I clinical trial, leading to the forced termination of the study. Consequently, successful STING-based immunotherapy will hinge on two innovations delivered in tandem: (i) co-opting PD-1/PD-L1 blockade to neutralize STING-driven immune escape and (ii) a tumor-selective delivery platform that concentrates agonist within the lesion while sparing normal tissues from systemic cytokine storm [22, 23]. Herein, translating the clinically validated co-dosing concept into a single molecule, we devised BMSA: a heterobifunctional prodrug that stitches the PD-L1 inhibitor BMS-1 to the STING agonist MSA-2 via a glutathione-cleavable disulfide linker (Scheme 1A). This bond, already safety-qualified in multiple clinical-stage ADCs and small-molecule prodrugs [24], remains intact in circulation yet fragments within the GSH-rich tumor microenvironment, affording conditional activation without additional systemic toxicity. This strategic design capitalizes on three key chemical and biological principles: First, the disulfide linkage provides tumor selectivity through differential glutathione (GSH) concentrations in the tumor microenvironment versus normal tissues. Second, the prodrug’s increased molecular weight and lipophilicity prevent passive diffusion into normal cells, while allowing active uptake in tumors. Third, the design enables spatiotemporal control of drug release, with extracellular GSH triggering liberation of BMS-1 for immediate PD-L1 blockade, followed by intracellular esterase-mediated release of MSA-2 for STING pathway activation. To translate the prodrug’s molecular selectivity into macroscopic tumor enrichment we enlisted neutrophils as living Trojan horses. These first-responder leukocytes: (a) swarm irradiated tumors under X-ray-induced local inflammation [25], and (b) extend circulation time, confine activation to the lesion, and shield cargo from renal/hepatic clearance—an established paradigm for safe, cell-mediated targeted therapy [26]. We therefore engineered a chimeric polypeptide, NEMCR, that concatenates: (a) the neutrophil elastase binding peptide NEBP for high-affinity hitchhiking on circulating neutrophils [27], and (b) the tumor-homing motif CREKA, linked through an MMP 2/9-cleavable spacer (PVGLIG) that becomes exposed only within the tumor microenvironment [28].The targeting polymer DSPE-PEG-NEMCR was synthesized by conjugating DSPE-PEG2K-Maleimide with NEMCR, which was subsequently co-assembled with DSPE-mPEG, DOTAP, DSPC, cholesterol, and prodrug BMSA to form neutrophil-hitchhiking and tumor-targeting lipid nanoparticles (BMSA@T-NPs) (Scheme 1B). Following intravenous injection, surface-exposed NEBP on the nanoparticles latched onto activated neutrophils that swarm the tumor under X-ray-induced inflammation. Once inside the tumor microenvironment, locally overexpressed MMP-2/9 cleaved the PVGLIG spacer, shedding the neutrophil anchor and unmasking CREKA. Liberated particles immediately bind fibrin-fibronectin complexes of the tumor vasculature, executing a two-stage cascade targeting that converts transient leukocyte hitchhiking into sustained intratumoral retention (Scheme 1C). By merging conditional prodrug chemistry with a neutrophil-driven delivery cascade, we convert the body’s innate inflammatory circuitry into a self-amplifying targeting mechanis. The tandem system simultaneously disables STING-induced immune escape (via PD-L1 blockade) and abolishes systemic cytokine storm (via tumor-confined agonist release), offering a ready-to-translate solution to the two central roadblocks that have stalled clinical deployment of STING-based cancer immunotherapy. Illustration of STING Agonist Prodrug BMSA@T-NPs for cancer immunotherapy via cascade-targeting delivery. (A) Design and mechanism of BMSA. (B) Design and mechanism of cascade-targeting delivery system. (C) Illustration of BMSA@T-NPs initiating anti-tumor immune response under X-ray irradiation Results and discussion Design and mechanism of BMSA BMSA was synthesized by covalently coupling STING agonist MSA-2 and PD-L1 inhibitor BMS-1 through 2, 2'-dithiobis-Ethanol (Fig. 1A). The detailed synthetic processes and related characterization were shown in Supporting Information, including 1H NMR, HRMS, and HPLC (Figures S1-S7). Such disulfide bonds were widely applied in the design of prodrugs and ADCs, as it can be reduced to thiol by reductive species, such as Glutathione (GSH) [29]. Disulfide bonds are stable under physiological pH but susceptible to nucleophilic attack by thiols. In plasma, the primary thiol species is the reduced form of human serum albumin; however, its reactivity is hindered due to the limited exposure of the free thiol-containing residue and the excessively low concentration. In contrast, GSH, as the most abundant small-molecule thiol anti-oxidant (1–10 mM), can effectively cleavage disulfide bonds. What’s more, compared to normal tissues, tumor-associated oxidative stress typically increases GSH levels up to 1000 times [30], which would endow BMSA additional selectivity towards to tumor tissue. As shown in Fig. 1B, when BMSA was incubated with dithiothreitol (DTT) in vitro, the reaction yielded two new products (peaks at 19.01 min and 21.97 min). Then, we purified and isolated the two products through HPLC for further studies. Unexpectedly, the subsequent HPLC and HRMS studies indicated that BMS-1 has been completely released from BMSA, which means the hydrolysis of BMS-1 ester bond doesn’t require involvement of esterase and could be accomplished solely through disulfide bond cleavage (Fig. 1C). As shown in Figure S8, thiirane formation was observed in BMSA upon DTT treatment, demonstrating that the BMS-1 moiety undergoes further decomposition after disulfide bond cleavage. This results in the elimination of the ester group, thereby directly releasing BMS-1. The differences between BMSA and previously reported disulfide linker cleavage mechanisms are illustrated in Figure S9. However, HPLC and HMRS results of another product (peak at 21.97 min) indicated that the ester bond remained intact after the cleavage, and MSA-2 existed in a thiol-containing form. Full restoration of MSA-2 would be achieved upon esterase-catalyzed hydrolysis of ester bonds (Fig. 1D-E). Interestingly incubation of BMSA with esterase alone only resulted in minimal amount hydrolytic release of MSA-2 (Fig. 1F). This may result from the pronounced steric hindrance of BMSA, which impeded substrate-enzyme binding and thereby suppressing esterase-mediated hydrolysis. Because of the stronger steric hindrance of BMS-1, BMS-1 was undetectable in HRMS spectrum compared with MSA-2. To further validate the mechanism of BMSA, we incubated cells with BMSA and DTT-pretreated BMSA, respectively, and subsequently monitored intracellular metabolites. The results showed that no degradation products were detected when treated with BMSA, but a BMSA signal (888.2908) was observed (Figure S10). In contrast, when cells were treated with DTT-pretreated BMSA, signals corresponding to both MSA-2 (293.0489) and BMS-1 (474.2289) were detected (Figure S11), while BMSA signal completely disappeared. Taken all above together, we proposed the mechanism of BMSA in Fig. 1C. BMSA must first undergo cleavage of its disulfide bonds by GSH in the tumor microenvironment. The BMS-1 moiety was then directly released via nucleophilic attack of the thiol group on the α-carbon of the ester bond, whereas MSA-2 required further transmembrane transport to enter into the cytoplasm followed by esterase-mediated hydrolysis for restoration. Although disulfide bonds have been widely used in various clinical drugs, there has been limited research on the further release of payloads after disulfide bond cleavage. To date, all reported drugs capable of directly releasing payloads upon disulfide bond cleavage have been achieved through chemically unstable groups such as carbonate or carbamate [29, 31]. To the best of our knowledge, this is the first example of a prodrug that directly releases its payload via an ester linker. Such unique characteristics of BMSA maximize the pharmacological activity of both the BMS-1 moiety and the MSA-2 component. The direct release of BMS-1 facilitated binding to membrane protein PD-L1, while the esterified MSA-2 likely exhibits enhanced membrane permeability compared to its non-esterified counterpart. By contrast, BMCC, in which BMS-1 was directly coupled with MSA-2 through 1,6-Hexanediol, exhibited no detectable interaction with DTT (Figure S12). Design and Mechanism of BMSA. (A) Synthetic route of BMSA. (B) High-Performance Liquid Chromatography (HPLC) analysis of BMSA, BMSA incubated with DTT and BMSA incubated with Esterase. (C) Proposed working mechanism of BMSA in tumor microenvironment. (D) High Resolution Mass Spectra (HRMS) of BMSA incubated with DTT. (E) HRMS of BMSA incubated with DTT and Esterase. (F) HRMS of BMSA incubated with esterase. (G) mRNA level analysis of STING-dependent downstream cytokines in THP-1 cells treated with different reagents via qRT-PCR (n = 3). (H) IFN-β and IL-6 secretion of THP-1 cells under various treatments. The concentration of IFN-β and IL-6 in culture medium was determined via enzyme-linked immunosorbent assay (ELISA). Data are presented as ± s.d. of the mean, n = 3. (I) Immunoblot analysis of STING signaling in THP-1 cells treated with MSA-2 (10uM), BMS-1 (10 uM), BMCC (10 uM), BMCC (10 uM) with DTT (50 uM), BMSA (10 uM) and BMSA (10 uM) with DTT (50 uM). Uncropped gel images are attached at the end of the supplementary material. (J) BMDCs exhibit significantly increased CD80/CD86 expression, n = 5. (K) Mean fluorescence intensity analysis in (J). *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001 STING activation and delivery system design of BMSA To further elucidate the mechanism of BMSA at cellular level, we tested the activation of STING pathways in THP-1 cells. Consistent with the proposed mechanism, western blot results showed remarkable phosphorylation of STING and the downstream IRF-3 [32] only when BMSA was pretreated with DTT (Fig. 1G-I). Although HRMS revealed trace amounts of MSA-2 released via esterase-mediated hydrolysis, such level was insufficient to activate STING. Moreover, intact BMSA possessed limited plasma membrane permeability due to its excessive molecular weight and strong lipophilicity. STING activation of BMSA was further validated by elevated expression and secretion of related cytokines IFN-β and IL-6 (Fig. 1G and H). Compared to MSA-2, the slightly stronger STING-activating effect of BMSA after DTT treatment observed in the experiment may be attributed to the enhanced cell membrane permeability of the esterified MSA-2. Finally, we conducted an in vitro DC maturation analysis through flow cytometry experiments, the upregulation of the costimulatory factors CD80/CD86 confirmed that BMSA were capable of promoting DC maturation in vitro after DTT treatment (Fig. 1J and K). Although disulfide bond could endow BMSA selectivity towards tumor tissue in some ways, molecular structure of BMSA violated the rule of five and would compromise its biophysical activity. In order to optimize the selectivity and bioavailability of BMSA, we designed MMP-response tumor-targeting lipid nanoparticles based on neutrophil-hitchhiking strategy for targeted delivery of BMSA (Fig. 2A). As shown in Fig. 2B, a targeting peptide (NEMCR) was designed and synthesized, which was consisted of three functional moieties, a NEBP (neutrophil elastase-binding peptide) fragment which has been demonstrated to mediate nanoparticle surface functionalization, enabling precise targeting and stable adhesion to activated neutrophil membranes [33]; a MMP-cleavable peptide fragment and a CREKA peptide fragment that capable of specifically binding to coagulation-associated plasma proteins within tumor vasculature [34]. The mass spectra (Figure S13) revealed successful synthesis of the designed targeting peptide containing 26 amino acids. Then the binding peptide was covalently conjugated with DSPE-PEG2000-MAL through Michael addition between cysteine and maleimide, thus the molecular weight increased to 4379.1 Da (Fig. 2C). Addition of MMP led to cleavage of the targeting peptide and then resulted in a reduction of the molecular weight of the conjugate to 2538.6 Da, accompanied by the appearance of a peptide fragment at 1824.0 Da, which aligns with the theoretical prediction. Subsequently, we formulated the complex with multiple components into liposomes through reverse phase evaporation and performed comprehensive characterization of the resulting nanoparticles. In the subsequent formulation screening, by adjusting the proportion of NEMCR, we achieved a high encapsulation efficiency of up to 93% (Figs. 2D and S14). As shown in Fig. 2E and F, the nanoparticles modified with targeting peptides (BMSA@T-NPs) have the same particle size (around 120 nm) as those without targeting peptide modification (BMSA@U-NPs), but exhibit a lower zeta potential (Fig. 2G). This phenomenon may be attributed to the positive charges carried by NEMCR under neutral pH conditions (Figure S15), which neutralized the negative charges of the phospholipids. Subsequent stability experiments confirmed that the lipid nanoparticle formulation obtained exhibited excellent stability, showing no significant aggregation within two weeks (Fig. 2H). As a neutrophil target among neutrophil-specific biomarkers, neutrophil elastase (NE) exhibits unique advantages over Ly6G. Its surface exposure is strictly confined to activated neutrophils [33], and its expression profile is exclusively restricted to neutrophils without any other leukocyte subsets.27 To verify the delivery capability and mechanism of the nanoparticles, we co-encapsulated Cy5 and BMSA into the lipid nanoparticles. The incorporation of Cy5 did not alter the size or zeta potential of the nanoparticles. Relevant characterizations were shown in Figure S16. The results in Fig. 3A demonstrated that when staining inactive neutrophils with BMSA@T-NPs, minimal Cy5 fluorescence signal was observed, showing identical staining patterns to BMSA@U-NPs. In contrast, fMLP-activated neutrophils exhibited intense fluorescence in Cy5 channel upon BMSA@T-NPs treatment, which would be abolished after MMP pretreatment. In contrast to the pronounced internalization of Ly6G staining patterns, a substantial proportion of BMSA@T-NPs retained on the neutrophil membrane surface. This membrane-anchored characteristic established a strategic foundation for subsequent MMP-responsive cleavage and tumor cell targeting through neutrophil-mediated delivery. The SEM images also demonstrated the same results (Fig. 3B), directly confirming the binding capability of the nanoparticles to activated neutrophils and their responsiveness to MMP. The findings above demonstrated that the targeting peptide modification endowed BMSA@T-NPs with neutrophil-specific binding capability on the one hand, such membrane-anchored characteristic on neutrophils facilitated MMP-mediated enzymatic cleavage, enabling following release from the neutrophil surface for subsequent tumor-targeting cascades on the other hand. At last, the PI staining (Fig. 3C) and CCK8 (Figure S17) results also demonstrated that both BMSA and BMSA@T-NPs exhibited low cytotoxicity, making it a safe STING agonist prodrug. Design and characterization of LNP delivery system for BMSA. (A) Construction of the LNP delivery system was composed of DSPE-mPEG, DSPC, DOTAP, cholesterol and DSPE-PEG2000-NEMCR. (B) Molecular structure of targeting peptide NEMCR and the resulting peptide after MMP cleavage. (C) Maldi-TOF mass spectra of conjugate of NEMCR with DSPE-PEG2000-MAL(Left); products of the conjugate after MMP cleavage (Right). (D) The entrapment efficiency of BMSA under different proportions of NEMCR. (E) Particle size of BMSA@U-NPs and BMSA@T-NPs via dynamic light scattering analysis, n = 3. (F) TEM images of BMSA@U-NPs and BMSA@T-NPs, bar = 200 nm. (G) Zeta potential of BMSA@U-NPs and BMSA@T-NPs, n = 3. (H) Stability analysis of BMSA@U-NPs and BMSA@T-NPs Delivery mechanism of LNP delivery system and safety assessment. (A) Fluorescent images of neutrophils stained with DAPI, fluorescent-labelled Ly6G antibody, BMSA@U-NPs and BMSA@T-NPs in absence or presence of 1 µM fMLP, bar = 10 µm. (B) SEM images of neutrophils under different treatment. (C) Fluorescent images of live/dead dual Staining of LL/2 cells upon various treatment Cy5@T-NPs enable precise tumor targeting with prolonged retention under X-ray irradiation Having validated the critical role of the targeting peptide in BMSA@T-NPs cellular uptake, we subsequently conducted in vivo investigations to elucidate the tumor-targeting capability of this delivery system through intravital imaging and biodistribution analyses on a xenograft tumor model. All animal experiments were performed by the Zhejiang University guidelines for the care and use of laboratory animals, and approved by the Institutional Animal Care and Use Committee of the Zhejiang University. Hepa1-6 cells were subcutaneously injected into C57BL/6 mice which were then intravenous administered with free Cy5 dye, U-NPs loaded with Cy5 and T-NPs loaded with Cy5. In the last group, the mice were exposed to X-ray 12 h before the treatment (Fig. 4A). Intravital imaging was conducted on these mice at five time points in the following 48 h to track real-time dye biodistribution. For the free Cy5 dye group, strongest fluorescence occurred at 4 h (first time point), and then declined over the subsequent 48-h monitoring period. For Cy5@U-NPs and Cy5@T-NPs, maximum fluorescence were observed at 12-h time point in tumors, followed by gradual signal attenuation over subsequent observation (Fig. 4B). Both of these two groups showed better tumor- specificity and stronger fluorescence compared with the free Cy5. The results demonstrated that lipid nanoparticle encapsulation significantly enhanced tumor-specific Cy5 accumulation and prolonged metabolic cycle due to enhanced permeability and retention (EPR) effect. Critically, in the absence of induced inflammatory priming in these murine cohorts, the inactive neutrophils rendered the targeting peptide modification strategy inoperative. Consequently, intravital imaging revealed negligible difference between the two groups. For the last group, mice received localized X-ray irradiation before drug administration, to establish an inflammatory microenvironment at tumor site. The local inflammatory microenvironment not only induced neutrophil activation enabling targeting peptide binding, but also led to radiation-induced tissue damage generating related chemokine that mediate neutrophil recruitment [35]. The resultant neutrophil swarm effect facilitated tumor-specific nanoparticle accumulation. As expected, the fluorescence of the last group was the strongest among the four groups at any time point. Consistent with previous groups, the last group exhibited peak fluorescence at 12 h post-injection (Fig. 4C). However, unlike the other group, the irradiated group showed transient peritumoral accumulation that resolved within 48 h. This spatial confinement likely arose from X-ray-induced bystander neutrophil recruitment in the irradiated peritumoral regions. In contrast, tumor sites with elevated MMP2/9 expression can proteolytically cleave the targeting peptide NEMCR, exposing the tumor-homing peptide CRKEA and thereby anchoring the drug for long-term retention. The mice were euthanized at the end of experiment, major organs and tumors were collected for fluorescence imaging. As shown in Fig. 4D, free Cy5 was primarily located in the liver, with minimal presence in tumors, the low fluorescence intensity indicated that most of Cy5 had been excreted through metabolism. For Cy5@U-NPs and Cy5@T-NPs, like most nanoparticles, they accumulated both in liver and tumor, with slightly aggregating in spleen. For irradiation group, the dye was successfully delivered into tumor with high specificity. Unlike the hepatic accumulation tendency observed in most nanoparticles, this group exhibited negligible fluorescence in the liver, with only minor off-target fluorescence detected in the lungs (Fig. 4E). The aforementioned experimental results demonstrate that this delivery system can hitchhike neutrophils to precisely target tumors via X-ray guidance, significantly reducing off-target effects while achieving prolonged retention at the tumor site. In vivo distribution of Cy5 with/without LNP encapsulation. (A) Schematic overview of the experimental design for tumor targeting ability of the delivery system. (B) Representative intravital fluorescence images of mice at indicated time points following intravenous administered with free Cy5 dye, Cy5@U-NPs, Cy5@T-NPs with/without X-ray irradiation. (C) Quantitative analysis of tumoral fluorescence intensity (yellow box in Fig. 4B) at various time points post-injection. (D) Representative fluorescence images of main organs and tumors collected at the end of the experiment. (E) Fluorescence intensity quantitative analysis of organs and tumors in Fig. 4D. Data in C and E are presented as mean ± s.d. of n = 5 mice per group. *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001 In vivo therapeutic evaluation of BMSA@T-NPs against cancer After confirming the cellular activity of BMSA and the in vivo functionality of the targeted delivery system, we further investigated the in vivo antitumor efficacy of BMSA@T-NPs on LL/2 tumor-bearing mouse model. Before the experiment, we first determined the dosage of BMSA@T-NPs (Fig. 5A). Based on the standard curve in Fig. 5B, the concentration of IFN-β in the mouse serum exhibited a dose-dependent relationship after X-ray irradiation. When the dosage reached 15 mg/kg, the IFN-β concentration tended to saturate, thus 15 mg/kg was chosen as the dosage for BMSA@T-NPs administration via tail vein injection. The therapeutic effects of Saline, BMSA@U-NPs and BMSA@T-NPs were analyzed upon tail vein administration on days 10, 13, and 16 (Fig. 5C). In the X-ray group, tumor sites were irradiated with X-ray 12 h before drug administration. Tumor volume and body weight were recorded every two days for all groups. The results showed that all treatment groups exhibited tumor-suppressive effects (Fig. 5D). Notably, BMSA@U-NPs and BMSA@T-NPs exhibited comparable tumor suppression efficacy with favorable inhibitory outcomes during the early stage of experiment; however, significant rapid tumor regrowth was observed after day 16 (Fig. 5E-H). By contrast, the irradiated group exhibited significant therapeutic efficacy throughout the entire experimental period, achieving tumor inhibition rate of 91.8%. This finding underscores the necessity of X-ray irradiation in the therapeutic regimen, as it substantially enhanced drug accumulation at the tumor site and prolonged retention time. No significant changes in body weight were observed across all experimental groups (Fig. 5I). This finding suggested the favorable safety of the therapeutic strategy, likely attribute to the introduction of disulfide bonds in BMSA that prevented drug activation in normal tissues, combining with liposomal encapsulation that promoted tumor-specific accumulation. On experimental day 20, mice were euthanized followed by collection of heart, liver, spleen, lungs, kidneys and tumor tissue for comprehensive analysis. As shown in Fig. 5J, the tumors were photographed on a A4 paper, and the tumor weight was recorded in Fig. 5K. To further investigate the therapeutic impact of BMSA decomposition in antitumor activity, we added (BMS-1 + MSA-2) @ T-NPs group in the experiment (Figure S18). The results demonstrate that completely decomposed BMSA significantly enhances tumor suppression, while also indicating that BMSA in the BMSA@T-NPs group was not fully converted to BMS-1 and MSA-2. Meanwhile, the X-ray irradiation group still exhibited the strongest antitumor activity, proving that drug delivery is a more decisive factor for BMSA-based therapy. Antitumor activity evaluation of BMSA@T-NPs under X-ray irradiation. (A) Dose escalation of BMSA@T-NPs by intravenous injection and corresponding plasma IFN-β levels in C57BL/6 mice (n = 5) 12 h after dosing. (B) Standard curve of plasma IFN-β levels. (C) Schematic representation of the therapeutic regimen for the LL/2 tumor-bearing mouse model in C57BL/6 mice. (D) Tumor growth kinetics of LL/2 tumor model under various treatment. (E–H) Individual tumor volume of LL/2 tumor model for Saline group (E), BMSA@U-NPs group (F), BMSA@T-NPs group (G) and BMSA@T-NPs group + X-ray group (H). (I) Body weight kinetics of mice under various treatment. (J) Obtained tumors shot on a A4 paper at the end of experiment. (K) Tumor weight of mice under various treatment. *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001 BMSA@T-NPs activated STING pathway and anti-tumor iImmune response Following the analysis of tumor suppression, tumor tissues were processed into paraffin-embedded sections for subsequent immunofluorescence and immunohistochemical analyses. We first performed immunofluorescence imaging of IL-6, which acts as a downstream effector of the cGAS-STING pathway to further activate STAT1 and STAT3. The results demonstrated that the irradiated group exhibited the strongest IL-6 fluorescence intensity, while the two non-irradiated groups showed only marginal signal enhancement compared to the saline group (Fig. 6A). These findings demonstrated that MSA-2 underwent substantial release at the tumor site, thereby activating the STING pathway. Such results also served as indirect evidence for the successful release of BMS-1, as the release of MSA-2 is dependent on cleavage of the disulfide bond. It is well known that the antitumor activity of STING depends on the infiltration, maturation, and triggered antigen cross-presentation of dendritic cells (DCs). Simultaneously, the enhanced fluorescence of CD11c in the irradiated group also indicates that DC infiltration was significantly increased compared to other experimental groups. The unmerged images are displayed in the Figures S19 and S20. In accordance with the immunofluorescence results, amount of CD3+ cells and CD8+ cells increased dramatically in the irradiated group (Fig. 6B). Meanwhile, CD4+ cells exhibited a certain degree of reduction. Finally, we repeated the experiment to analyze DC cell maturation in vivo (Figs. 6C and S21). In this experiment, a new experimental group was added, and the results showed that BMSA@T-NPs effectively promoted DC maturation. Compared to the combined administration of MSA-2 and BMS-1, the introduction of disulfide bonds and NEMCR modification significantly inhibited STING activation, indirectly indicating that this strategy could reduce the risk of cytokine storms triggered by STING agonists. These findings collectively demonstrated that the combined treatment of BMSA@T-NPs and X-ray irradiation significantly increased infiltration and maturation of DC cells and activated anti-tumor immune responses. Moreover, the immunofluorescence results of Ki67 in Figure S22 and TUNEL staining further confirmed that tumor cell proliferation was significantly suppressed, while apoptosis was effectively induced in tumor cells (Figure S23). Mechanisms of BMSA@T-NPs activating anti-tumor immune response under X-ray irradiation. (A) Immunofluorescence imaging of IL-6 and CD11c of formalin-fixed paraffin-embedded tumor tissue slides. (B) Immunohistochemical analysis of CD3+, CD4+ and CD8.+ cells in tumor tissues after different treatments. (C) Flow cytometry scatter plots and quantification of CD80/CD86 expression by DCs in the tumor-draining lymph nodes (TDLN) (n = 5). *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001 In Vivo biocompatibility evaluation To further evaluate the biocompatibility and potential adverse effects of above therapeutic regimen, the same grouping strategy was adopted for immunocompetent mice received the aforementioned treatment once daily for seven consecutive days, with blood samples collected for systematic analysis (Fig. 7A). As shown in Fig. 7B, the values of albumin (ALB), aspartate transaminase (AST/GOT), lactate dehydrogenase (LDH) and total protein (TP) of all groups were within the normal range, indicating that their liver function exhibited no abnormalities. Besides, the test results of indicators related to renal function, such as creatinine (CRE), uric acid (UA) and urea nitrogen (BUN), were also within the normal range and showed no significant difference compared to the untreated group. In addition to blood biochemical tests, key complete blood count (CBC) parameters, such as white blood cell (WBC), red blood cell (RBC), platelet (PLT) and neutrophils (NEU) all fallen within healthy ranges, while other routine blood test results have been statistically compiled in Figure S24. The H&E staining of the main organs demonstrated no detectable organ damage in all major organs (Fig. 7C). Although the introduction of X-ray irradiation may increase the risk of systemic inflammation, owing to the shielding of non-tumor areas by lead blocks and the limited X-ray dosage employed [36, 37], no detectable systemic toxicity or organ damage were observed. All the above results demonstrated that whether it is the BMSA prodrug itself, the delivery system developed for it, or the introduced X-ray irradiation, all exhibited excellent biocompatibility and low toxicity. Throughout the treatment period, no detectable damage was observed in various organs of mice. This strategy successfully circumvented the systemic toxicity of STING agonists commonly associated with potential cytokine storms. In Vivo biosafety evaluation of various treatment. (A) Schematic illustration of the experimental design for biosafety evaluation. (B) Blood biochemistry analysis and key complete blood count of blood samples obtained from Saline group, BMSA@U-NPs group, BMSA@T-NPs group and BMSA@T-NPs group + X-ray group. (C) H&E staining of heart, liver, spleen, lung and kidney for different groups BMSA@T-NPs achieved anti-tumor efficacy against distal tumor It is well-known that, even though STING agonists are often administered intratumorally, they tend to activate systemic immune responses, which can, in severe cases, even trigger a cytokine storm. As is illustrated in Fig. 8A, immunocompetent mice were subcutaneously injected with LL/2 cells on day 0 (right side: primary tumor) and day 4 (left side: distal tumor) to establish a bilateral tumor model, followed by intratumorally administration of BMSA@T-NPs specifically to the right flank tumors on day 9, day 12 and day 15. To verify whether BMSA@T-NPs can activate systemic immunity, the experiment adopted an intratumorally administration treatment modality, thereby eliminating the need of X-ray irradiation to induce neutrophil aggregation at tumor site. As shown in the results (Fig. 8B - G), the treatment targeting the primary tumor not only reduces its volume but also inhibits the growth of the distal tumor. The mice were sacrificed and tumors were excised at day 20. As illustrated in the results, the primary tumors in the treatment group demonstrated significantly better therapeutic efficacy compared to the distal tumors, with one mouse even exhibiting complete tumor regression in the primary lesion (Fig. 8H and I). Unlike the above experimental results from tail vein intravenous injection, intratumorally administration caused a slight decrease in mice body weight during the initial phase of the experiment, followed by a gradual recovery thereafter, demonstrating a therapeutic effect similar to that of STING agonists (Fig. 8J). Collectively, the experimental results demonstrate that BMSA@T-NPs retained the capacity to activate systemic immunity even when administered intratumorally. This finding suggested its therapeutic potential against metastatic tumors or multifocal malignancies. BMSA@T-NPs boosted systemic immunity. (A) Schematic overview of the therapeutic regimen against bilateral tumor model. (B) Representative images of mice at indicated time points following intratumorally administered with Saline or BMSA@T-NPs. (C) Bilateral tumor growth kinetic of LL/2 tumor model under various treatment. (D-G) Individual tumor volume of LL/2 tumor model for Saline-Primary group (D), BMSA@T-NPs-Primary group (E), Saline-Distal group (F) and BMSA@T-NPs-Distal group (G). (H) Bilateral tumor weight of mice in the two groups. (I) Obtained tumors shot on a A4 paper at the end of experiment. (J) Body weight kinetics of mice in the two groups. *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001 Conclusions In this research, we report BMSA, a dual-compartment prodrug that coupled the PD-L1 inhibitor BMS-1 to the STING agonist MSA-2 through a GSH-cleavable disulfide linker. Tumor glutathione triggered a self-immolative cascade reaction that instantly released BMS-1 extracellularly, whereas intracellular esterase hydrolyzed the remaining MSA-2 ester, enabling subsequent STING activation. Despite the extra hydrolysis steps, the lipophilic ester promoted membrane permeation compared with MSA-2, yielding cGAS-STING signaling that surpassed free MSA-2. This first-in-class heterobifunctional construct thus executes spatiotemporally segregated release: membrane-level checkpoint blockade and cytosolic innate immune ignition. To confine the prodrug to tumor site, we employed the Trojan horse delivery system, which has attracted considerable research interest recently [38, 39]. We utilized the inflammatory chemotactic properties of neutrophil and encapsulated BMSA in neutrophil-hitchhiking lipid nanoparticles (T-NPs) whose surface carries a designed NEMCR peptide consisting of a NEBP peptide and a caged CREKA motif separated by an MMP-2/9-sensitive spacer. Intravenously injected T-NPs bind circulating neutrophils that accumulated into irradiated tumor site via X-ray-induced inflammation; local MMP cleavage then exposed CREKA, anchoring the particles to fibrin–fibronectin complexes within the aberrant vasculature. The cascade delivery system produced remarkable higher intratumoral accumulation than free dye, abolished hepatic sequestration, and limited off-target lung accumulation. In LL/2 tumor-bearing mice, the combination of BMSA@T-NPs with local irradiation achieved 91.8% tumor suppression, dramatic increasement of intratumoral CD11c⁺ DCs and CD8⁺ T cells, and elevated IL-6 without detectable systemic cytokine storm. Intratumoral administration further elicited abscopal control of contralateral tumors, confirming systemic immune engagement. No weight loss or organ toxicity was observed over 30 days. By integrating clinically validated PD-1/PD-L1 blockade with STING agonism inside one tumor-activated prodrug and a neutrophil-driven delivery circuit, this strategy resolved the efficacy–toxicity impasse that currently constrains cGAS-STING therapeutics. Materials and methods Materials BMS-1, MSA-2, EDCI, DMAP, 2,2'-dithioiethanol and 1,6-Hexanediol were purchased from J&K Chemical Co. Ltd. The solvents used are in analytical grade. Nuclear magnetic resonance spectra were obtained on a Bruker AVIII500M spectrometer. High Resolution Mass Spectra were measured on Bruker timsTOFPro. The mass spectra of peptides were recorded through UltrafleXtreme MALDI − TOF/TOF MS (Bruker Daltonic, Germany), and data were collected using Bruker Flexcontrol 3.4 and analyzed by Bruker Flexanalysis 3.4. The High-Performance Liquid Chromatography spectra were recorded with Agilent 1260 Infinity II Liquid Chromatography, and the column used was Shim-pack VP-ODS 150 × 4.6 mm (SHIMADZU, Japan). The fluorescence imaging was carried out on Evident (Olympus) SpinSR. Synthesis of 2-((2-hydroxyethyl)disulfaneyl)ethyl 1-(2,6-dimethoxy-4-((2-methyl-[1,1'-biphenyl]-3-yl)methoxy)benzyl)piperidine-2-carboxylate BMSS 475 mg of BMS-1 (1 mmol), 960 mg of EDCI (5 mmol) and 610 mg of DMAP (5 mmol) was added into a dry round flask containing 20 ml anhydrous DCM, the mixture was stirred at room temperature for 1 h. After that, 1.18 ml 2,2'-dithioiethanol (10 mmol) was quickly added into the mixture and kept the reaction at room temperature for 48 h. At the end of reaction, as showed by TLC, the solvent was removed under reduced pressure. The residue was dissolved in 30 ml EtOAc, and washed with water (200 ml × 3) to remove excess 2,2'-dithioiethanol. The organic phase was dried with Magnesium Sulfate (20 g) and concentrated under reduced pressure. The crude product was used directly in the next step without further purification. Yield: 76%. HRMS (ESI) m/z: calcd for C33H42NO6S2+: 612.2448, found: 612.2448. Synthesis of BMSA 960 mg of EDCI (5 mmol) and 610 mg of DMAP (5 mmol) was added into a dry flask containing 294 mg of MSA-2 (1 mmol) which was dissolved in 20 ml DCM. The reaction was stirred at room temperature for 1 h. Then, the 611 mg of BMSS (1 mmol) was was added into the flask. The mixture was stirred for 48 h under room temperature to finish the reaction. After reaction completed, the solvent was evaporated under reduced pressure. The residue was further purified by column chromatography (silica, DCM/MeOH, v/v = 20:1). Yield: 55% 1H NMR (500 MHz, DMSO-d6, δ): 8.20 (s, 1H), 7.59 (s, 1H), 7.50 – 7.46 (m, 4H), 7.40 (t, J = 7.5 Hz, 1H), 7.31 (q, J = 8.3 Hz, 3H), 7.23 (d, J = 5 Hz, 1H), 6.38 (s, 2H), 5.18 (s, 2H), 4.46 (s, 1H), 4.32 – 4.28 (m, 3H), 3.87 (s, 3H), 3.84 (s, 3H), 3.75 (s, 6H), 3.33 (t, J = 5 Hz, 3H), 3.06 – 2.99 (m, 4H), 2.71 (t, J = 7.5 Hz, 2H), 2.23 (s, 3H), 2.05 – 1.98 (m, 1H), 1.68 (s, 1H), 1.48 (s, 2H), 1.36 – 1.25 (m, 6H). HRMS (ESI) m/z calcd for C47H54NO10S3+: 888.2904, found: 888.2902. Synthesis of 6-hydroxyhexyl 1-(2,6-dimethoxy-4-((2-methyl-[1,1'-biphenyl]-3-yl)methoxy)benzyl)piperidine-2-carboxylate BMDO 475 mg of BMS-1 (1 mmol), 960 mg of EDCI (5 mmol) and 610 mg of DMAP (5 mmol) was added into a dry round flask containing 20 ml anhydrous DCM, the mixture was stirred at room temperature for 1 h. Then, 1.18 g 1,6-Hexanediol (10 mmol) was quickly added into the mixture and kept the reaction at room temperature for 48 h. At the end of reaction, as monitored by TLC, the solvent was removed under reduced pressure. The residue was dissolved in 30 ml EtOAc, and washed with water (200 ml × 3) to remove excess 1,6-Hexanediol. The organic phase was dried with Magnesium Sulfate (20 g) and concentrated under reduced pressure. The crude product was used directly in the next step without further purification. Yield: 60%. HRMS (ESI) m/z calcd for C35H46NO6 + : 576.3320, found: 576.3322. Synthesis of BMCC 960 mg of EDCI (5 mmol) and 610 mg of DMAP (5 mmol) was added into a dry flask containing 294 mg of MSA-2 (1 mmol) which was dissolved in 20 ml DCM. The reaction was stirred at room temperature for 1 h. Then, the 575 mg of BMDO (1 mmol) was added into the flask. The mixture was stirred for 48 h under room temperature to finish the reaction. After reaction completed, the solvent was evaporated under reduced pressure. The residue was further purified by column chromatography (silica, DCM/MeOH, v/v = 25:1). Yield: 60% 1H NMR (500 MHz, DMSO-d6, δ): 8.17 (s, 1H), 7.56 (s, 1H), 7.49—7.43 (m, 4H), 7.37 (t, J = 7 Hz, 1H), 7.31—7.26 (m, 3H), 7.20 (d, J = 5 Hz, 1H), 6.35 (s, 2H), 5.14 (s, 2H), 4.12—4.08 (m, 1H), 4.02—3.96 (m, 3H), 3.84 (d, J = 10 Hz, 6H), 3.71 (s, 6H), 3.64 (q, J = 15 Hz, 2H), 3.29 (t, J = 7.5 Hz, 2H), 3.16 (t, J = 5 Hz, 1H), 3.05—3.00 (m, 1H), 2.67 (t, J = 7.5 Hz, 2H), 2.32—2.28 (m, 1H), 2.21 (s, 3H), 1.64 -1.61 (m, 2H), 1.58—1.51 (m, 4H), 1.43—1.22 (m, 8H). HRMS (ESI) m/z calcd for C49H58NO10S+: 852.3776, found: 852.3774. Metabolite analysis of BMSA in THP-1 cells THP-1 cells were cultured in 12-well plates and were treated with 10 uM BMSA or DTT-pretreated BMSA for 12 h. Then cells were lysed in RIPA buffer and the lysate was dialyzed using a 2000 Da dialysis cassettes (2 K MWCO Thermo Scientific Slide-A-Lyzer). The obtained filtrates were directly used in HRMS analysis. Western blotting THP-1 cells cultured in six-well plates were washed three times with PBS and subsequently lysed in RIPA buffer containing 10 µg/mL protease inhibitor at 4 °C for 30 min. The lysates were centrifuged at 15,000 g for 15 min at 4 °C, and the supernatants were collected for protein concentration determination using the BCA assay. Equal amounts of protein (20 µg per lane) were separated by SDS-PAGE (10% Bis–Tris gel) and transferred onto a nitrocellulose membrane. The membrane was blocked with 5% non-fat milk in TBS-T for 1 h at room temperature. Specific antibodies against p-IRF-3, IRF-3, STING, and p-STING were used to detect the target proteins. Finally, the membrane was imaged using a LI-COR chemiluminescence imaging system, and band intensities were analyzed using ImageJ for relative quantification. Quantitative RT–PCR analysis Total RNA was isolated using Trizol reagent (VeZol Reagent, Norway) and reverse-transcribed into cDNA using a one-step cDNA synthesis kit (HiScript II Reverse Transcriptase, Norway). Each sample was analyzed in triplicate, and the cDNA was subsequently used for qRT-PCR. Relative gene expression changes were calculated using the ΔΔCT method. Quantification of cytokines The levels of human IL-10, TNF-α, and IFN-β were measured using sandwich enzyme-linked immunosorbent assay (ELISA) kits. All assays were performed in triplicate from three independent biological replicates, following the manufacturer's instructions. CCK-8 Assay: Cell viability was assessed indirectly by CCK-8 assay (Beyotime Biotechnology, China). Briefly, Cells were seeded into 96-well plates with three replicates per group. Treatments included DTT, MSA-2, BMCC, BMCC combined with DTT, BMSA, and BMSA combined with DTT. After a 12-h incubation period, CCK-8 reagent was added to each well, followed by a 2-h incubation at 37 °C. Absorbance at 450 nm was then measured using a microplate reader (BioTek Epoch), and cell viability was calculated accordingly. Dual staining of live/dead cell LLC cells were seeded into 12-well plates at a density of 1 × 104 cells per well and cultured for 24 h prior to experimental treatment. Cell staining was performed according to the following steps: PBS washing: Cells were gently washed three times with pre-warmed phosphate-buffered saline (PBS, pH 7.4). Calcein-AM staining: Cells were incubated with PBS containing 1 μM Calcein-AM at 37 °C in the dark for 30 min. Propidium iodide (PI) counterstaining: After removing the Calcein-AM solution, cells were incubated with 100 μg/mL PI solution at room temperature in the dark for 5 min. Termination of staining: Cells were briefly rinsed with PBS to remove unbound dyes. Fluorescence imaging: Finally, 100 μL of PBS was added to each well to cover the cell layer, and dual-channel fluorescence imaging was immediately performed using a fluorescence microscope. Animals and xenograft tumors All animal experiments were performed by the Zhejiang University guidelines for the care and use of laboratory animals, and approved by the Institutional Animal Care and Use Committee of the Zhejiang University with approval number ZJU20240840 . Female C57BL/6 (4–6 weeks old) were purchased from GemPharmatech Biotechnology Co., Ltd. (Nanjing, China). To obtain xenograft tumor models in C57BL/6, 5 × 105 LL/2 cells were suspended in a mixture of PBS. Then, the suspension was injected at the right dorsal side of the upper hind limb of each 6–7-week-old mouse. When the tumor size reached approximately 40–60 mm3, different drugs were administered via intraperitoneal injection at the same dose of the prodrug every two times per week. BMSA@T-NPs (200 μg 10 mg/kg) or BMSA@U-NPs (200 μg 10 mg/kg) were administered via tail vein injection for C57BL/6 mice. The same amount of saline was also injected as the negative control. For in vivo anti-tumor efficacy of BMSA@T-NPs under the X-ray, mice were irradiated with X-rays 12 h before drug administered. Tumor growth was measured with calipers in two dimensions, and tumor volume was calculated using the formula: tumor volume = length × width2/2. The tumor size and body weight were measured every 2 days. At the end of experiment, C57BL/6 mice were sacrificed, and tumor and major normal tissues were harvested for analyses. In vivo fluorescence imaging and biodistribution To generate the tumor model, 5 × 105 Hepa1-6 cells in PBS were subcutaneously inoculated into the right flank of female nude mice and allowed to grow to a size of 100 mm2. For the biodistribution study, PBS, Cy5@U-NPs (200 μg 10 mg/kg), and Cy5@T-NPs (200 μg 10 mg/kg) were tail vein administered, and the mice were imaged at 4, 8, 12, 24 and 48 h using a Xenogen IVIS 200 system to detect Cy5 fluorescence. The major organs (hearts, livers, spleens, lungs, and kidneys) and tumors were collected for ex vivo fluorescence imaging and the ROI value was automatically calculated by V3vison software. In vivo biosafety To evaluate safety, PBS, BMSA@U-NPs, BMSA@T-NPs or BMSA@T-NPs + X-ray were tail vein administered into immunocompetent female C57BL/6 mice every 2 times per week at a drug dosage of 10 mg per kg (body weight) for a total of three administrations. At 24 h after the final injection, whole blood was collected into Microvette 100 potassium–EDTA blood collection tubes (Sarstedt) for complete cell counts (CBC) analysis. All blood analyses were performed by Hangzhou Haoke Biotechnology Co. Ltd. To evaluate serum chemistry, blood was collected without an anticoagulant and allowed to clot for 30 min at room temperature, followed by centrifuging at 3,000 g to isolate the serum. The serum samples were analyzed using Fuji DRI-CHEM SLIDE kits with the Fuji automatic dry biochemical analyzer NX500i. Tissue immunofluorescence Fixing fresh tissue in 4% paraformaldehyde (reusable) at 4 °C for 2 h. After fixation, wash once with PBS, then transfer the tissue to 30% sucrose (prepared in PBS) overnight at 4 °C. Taking the tissue out of the sucrose the next day, fixing it in the mold with OCT (embedding agent) to avoid air bubbles, and storing it at -80 °C. Pre-cool the cryostat to minus 20- 25 °C (about 2 h), and move the tissue-frozen blocks and utensils into the cryostat 30 min before sectioning to stabilize the temperature. The thickness of the tissue section is 8 μm. After slicing, let it dry. Then soak the slices in acetone (reusable) at -20 °C for 10 min and wash them with PBS. The slices were first washed once (5 min) with 200 μL of 5% BSA/TBST (TBS + 0.1% Tween20) and then blocked with 200 μL of 5% BSA/TBST at room temperature for 1 h. Next, use 200 μL of 1% BSA/TBST (TBS + 0.1% Tween20) to wash it at least 4 times, 5 min/time, and take it down to wash it 2 times. Applying 100 μL primary antibody (prepared with 1% BSA/TBST) to the slices overnight at 4 °C. The next day, slices were washed at least 4 times with 200 μL of 1% BSA/TBST (TBS + 0.1% Tween20), taken down, and washed 2 times. After incubating the secondary antibody (prepared with 1%BSA/TBST) at room temperature for 1 h, the slices used DAPI (1%BSA/TBST dilution) to stain cell nuclei at room temperature for 10 min. Wash with 200 μL of 1% BSA/TBST (TBS + 0.1% Tween20) at least 4 times, 5 min each time, and dry in a dark place at room temperature. Using 20 μL of mounting medium to prevent fluorescence quenching, cover the slips gently, and fixing with clear nail polish. For the TUNEL assay, sections were deparaffinized, and apoptotic cells were detected using the TUNEL assay kit (KeyGEN Biotech, China) according to the manufacturer’s instructions. The tissue slices were ready for IF imaging with the above-mentioned Zeiss microscope and the number of positive cells and the mean fluorescence intensity were counted by Image J software. Quantitative flow cytometry (FACS) analysis To evaluate the maturation status of DC cells after treatment, BMDC cells were seeded in 12-well plates (5 × 105 cells per well) and treated with related drugs. After 12 h of incubation, the cells were collected, washed twice with PBS, and labeled with fluorescent antibodies: APC-Cy7-CD45, PE-CD11c, APC-CD80, and PE-Cy7-CD86. The stained cells were then analyzed by flow cytometry (FACS Calibur, BD Biosciences, San Jose, CA). To evaluate the maturation status of DC cells in tumors, tumors were collected. For flow cytometry analysis, cells were passed through a 70 μm strainer to obtain a single-cell suspension. The maturation status of DC cells was analyzed using anti-CD11c, anti-CD80, and anti-CD86 antibodies. H&E staining H&E staining was performed on formalin-fixed paraffin-embedded 3 μm sections of dissected tissues. Slides with tissue sections were incubated at 58 °C overnight and the next day deparaffinized in xylene, followed by serial hydration steps in ethanol (100%, 70%) and water for 5 min each. These sections were stained with hematoxylin and eosin, and imaged under OLYMPUS VS200. Statistical analysis Statistical comparisons between groups were evaluated by Student's t-test. Data are shown as mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 were considered to be statistically significant. Supplementary Information Additional file1 Acknowledgements We thank our colleagues from the Center for Regenerative and Aging Medicine for their discussions and comments on this manuscript. Abbreviations Immune checkpoint blockade Glutathione Neutrophil elastase-binding peptide Antibody drug conjugates Matrix metalloproteinase High resolution mass spectrum High performance liquid chromatography Dithiothreitol Neutrophil elastase Enhanced permeability and retention Albumin Aspartate transaminase Lactate dehydrogenase Total protein Creatinine Uric acid Urea nitrogen Complete blood count White blood cell Red blood cell Platelet Neutrophils Authors contributions Conceptualization: Zhe Tang, Xiaoyuan Chen, Longguang Tang and Kesong Peng. Methodology: Huamiao Zhang, Yang Li, Qiuchen Bi, Heping Zhu, Huiwen Wang, Zhao Wang, Zhangqiang Ma, Huiping Liu, Xiaoling Zhu, Jiaqi Cao, and Xin Wang. Investigation: Huamiao Zhang, Yang Li, and Qiuchen Bi. Writing—original draft: Huamiao Zhang, Yang Li, and Qiuchen Bi. Funding This work was supported by the National Natural Science Foundation of China (Grant No. T2522033, 82270113) and start-up fund from Shandong Cancer Hospital (rcyj-202603). Data availability The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials. Declarations Ethics approval and consent to participate All animal experiments were performed by the Zhejiang University guidelines for the care and use of laboratory animals, and approved by the Institutional Animal Care and Use Committee of the Zhejiang University with approval number ZJU20240840. Consent for publication Not applicable. Competing interests Xiaoyuan Chen is a co-founder of and holds shares in Yantai Lannacheng Biotechnology Co., Ltd. Other authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Huamiao Zhang, Yang Li, and Qiuchen Bi have contributed equally to this article. Contributor Information Qiuchen Bi, Email: biqiuchen@zju.edu.cn. Zhe Tang, Email: 8xi@zju.edu.cn. Xiaoyuan Chen, Email: chen9647@gmail.com. Longguang Tang, Email: tanglongguang@zju.edu.cn. Kesong Peng, Email: kesongpeng@zju.edu.cn. References Associated Data Supplementary Materials Additional file1 Data Availability Statement The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.