Zn²⁺-mediated siRNA-doxorubicin self-assembled nanoparticles amplify Immunogenic cell death via aggravating redox dyshomeostasis for cancer therapy

Abstract

Immunogenic cell death (ICD), instigated by reactive oxygen species (ROS), has emerged as an efficacious strategy for augmenting the immunogenicity of tumor cells. However, the effects of ICD are severely diminished by elevated levels of glutathione (GSH) within tumor cells to maintaining intracellular redox homeostasis. To address this, the novel carrier-free self-assembled nanoparticles are designed to aggravating redox dyshomeostasis by reducing GSH and increasing ROS levels. The nanoparticles (ZDS NPs) were established by the self-assembly of zinc ions (Zn²⁺), doxorubicin (DOX) and Nrf2 siRNA. The ZDS NPs exhibited the ultra-high entrapment efficiency of DOX (99%) and siRNA (89%), and the releases of DOX and siRNA were both pH-dependent owing to the cleavage of coordinate and hydrogen bonds under acidic conditions. Following the endocytosis of ZDS NPs by tumor cells, redox homeostasis was significantly disrupted, DOX and Zn²⁺ enhanced the production of ROS via activating cGAS/STING pathway, whereas siRNA reduced GSH levels by decreasing the expression of Nrf2 protein. This further promoted a stronger ICD effect with elevated secretion of ATP, HMGB1, CRT, thereby inducing the maturation of DCs and activating a more robust anti-tumor immunity. This study presents a novel approach for the synergistic enhancement of ICD in cancer immunotherapy.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-04003-8.

Introduction

Tumor immunotherapy, particularly represented by immune checkpoint inhibitors, has achieved significant breakthrough, markedly prolonging patient survival [1]. Immunogenic cell death (ICD) is one of the important hallmarks of tumor immunotherapy. Apoptotic cells under endoplasmic reticulum stress trigger the release of damage-associated molecular patterns (DAMPs), including calreticulin (CRT), adenosine triphosphate (ATP) and high mobility group protein B1 (HMGB1) [2]. They stimulate dendritic cells (DCs) and enhance antigen presentation, which in turn elicit specific T cell responses, ultimately resulting in the enduring anti-tumor immune memory [2, 3]. However, the challenge lies in eliciting a more pronounced ICD effect [4]. Aromatic hydrocarbon-based drugs, such as doxorubicin (DOX), have been proven to trigger mild ICD through an indirect mechanism. It is necessary to emphasize that oxidative stress has significantly associated with ICD [2]. Stimulating the formation of reactive oxygen species (ROS) within cells can induce a large amount of ICD. Due to the presence of intracellular redox homeostasis, an elevation in ROS levels is accompanied by a corresponding increase in glutathione (GSH) concentrations [5, 6]. Therefore, merely augmenting the dosage of DOX does not effectively elevate the intracellular levels of ROS [7]. An innovative approach should be proposed to modulate the redox equilibrium that enhances the therapeutic efficacy of ICD [8, 9].

Cyclic GMP-AMP Synthase (cGAS) within tumor cells is capable of detecting intracellular damaged DNA and activating STING to upregulate type I interferon and proinflammatory cytokines, thereby inhibiting tumor progression [9]. The interaction between tumors and neighboring immune cells can also be mediated by the cGAS/STING pathway [10]. The metal-ion plays an instrumental role in the mechanism of modeling signal transmission involving cGAS/STING pathways. X Liang et al. reported that manganese oxide nanoparticles mitigated immune cell ferroptosis by interacting with the STING/NLRP3 signaling pathway, thereby enhancing immune responses [11]. Zinc ions (Zn²⁺) facilitated the phase separation and catalytic behavior of cGAS [12]. Furthermore, intracellular overload of ROS disrupts mitochondria and releases their DNA, thus enabling activation of the cGAS/STING signaling pathway [13–15]. Zn²⁺ within cancer cells promotes the creation of ROS, thereby leading to the death of cancer cells [15]. Consequently, we employed a novel strategy that involved the combination of DOX with Zn²⁺ to enhance the intracellular concentration of ROS for inducing a strong ICD effect.

When the intracellular ROS level in tumors is higher than that in normal tissues, in order to avoid cell death caused by excessive oxidative stress [16], tumor cells activate a series of antioxidant defense mechanisms that can reduce cytotoxicity, achieving a stable balance and survival under high oxidative stress conditions. Therefore, how to break this redox homeostasis is a problem that needs to be solved at present [17]. The maintenance of redox homeostasis inhibits the elevated levels of ROS [18]. Therefore, reducing GSH is also an important and essential strategy [19]. Nuclear factor E2-related factor 2 (Nrf2) is an important synthetase in regulating the biosynthesis of GSH [18, 19]. Through the modulation of glutathione reductase (GSR) and the utilization of nicotinamide adenine dinucleotide phosphate (NADPH) as a cofactor, Nrf2 regenerates oxidized GSH (GSSG), which is generated by the enzyme glutathione peroxidase (GPx), into GSH [22]. Small interfering RNA (siRNA) is a type of double-stranded RNA that possesses the capability to selectively degrade corresponding mRNA and facilitate the incorporation of associated proteins [20–22]. Abnormal protein expression is an key cause of cancer, and the use of siRNA is a very promising therapeutic approach [23–25]. Y Jun et al. reported the novel lipid nanocarrier (EYLN) was used to carry DOX and siRNA targeting the lipid synthesis metabolism gene LPCAT1. The carrier EYLN-DOX/siLPCAT1 was further coated with white blood cell membranes to obtain the mEYLNs-DOX/siLPCAT1 delivery system, which was used to improve the therapeutic effect of esophageal cancer [27]. However, the broader application of siRNA encounters challenges, including limited stability in serum [26, 27]. Moreover, the widely utilized carriers for siRNA are predominantly cationic liposomes, which cause the cytotoxicity effects [28–30]. Consequently, it is essential to the development of a reliable as well as safe siRNA delivering system to reduce the GSH levels.

The self-assembled nanoparticles (SNs) are governed by various important noncovalent bonds, such as hydrogen bonds, π-π stacking, van der Waals forces, charge transfer effects, dipole-dipole and hydrophobic interactions. Drug molecules self-assembles into nanoparticles at the nanometer scale [31]. The SNs have an extremely high drug loading capacity (exceeding 60% or even approaching 100%), and their self-assembly behavior can effectively provide conditions for dual-drug or multi-drug combination therapy [32]. Deng et al. constructed a folic acid-modified liposome loaded with DOX and responsive to matrix metalloproteinase 2, and used it in combination with the immunoadjuvant cytosine-phosphate-guanine for the treatment of breast cancer [33].Consequently, we hypothesized that DOX, siRNA, and Zn²⁺ collectively established the stable SNs.

In this work, we developed a new drug delivery system (ZDS NPs). It was demonstrated that the lone pair of electrons present in DOX interacted with Zn²⁺ to facilitate the formation of coordination polymer nanoparticles. In addition, siRNA had the capacity to engage with coordination polymer nanoparticles through π-π stacking interactions. Under mildly acidic conditions, the Zn²⁺, DOX, and siRNA were released. DOX induced cytotoxic effects and triggered ICD, while Zn²⁺ significantly enhanced the production of ROS. Additionally, siRNA contributed to downregulating Nrf2 protein expression. This combination of elevated ROS and decreased GSH levels effectively disrupted redox homeostasis and promoted a stronger ICD effect, thereby enhancing anti-tumor immunity (Fig. 1).

Fig. 1.

Schematic representation of SNs designed for breaking the redox hemostasis to enhancing the anti-tumor efficacy through the ICD effect. (A) Design and preparation of ZDS NPs. (B) the accumulation mechanism of ZDS NPs in the tumor tissue. (C) Zn²⁺ delivered by SNs significantly enhanced ROS production. siRNA effectively diminished Nrf2 protein expression, consequently reducing GSH levels. DOX demonstrated significant anti-tumor efficacy and prompted ICD. The synergistic effects of elevated ROS levels and DOX contributed to enhancing ICD, ultimately achieving robust anti-tumor efficacy

Materials and methods

Materials

Doxorubicn hydrochloride (DOX·HCl), polyvinylpyrrolidone (PVP, 30 kDa) and ZnCl₂ were acquired from Med Chem Express (China). siRNA (sense (5’−3’) GGUUGAGACU-ACCAUGGUUTT, antisense (5’−3’) AACCAUGGUAGUCUCAACCTT); FAM-siRNA (sense (5’−3’) GGUUGAGACU-ACCAUGGUUTT, antisense (5’−3’) AACCAUGGUAGUCUCAACCTT)); Negative control siRNA (sense (5’−3’) UUCUCCGAACGUGUCACGUTT, antisense (5’−3’) ACGUGUACACGUUCGGAGAATT)) were synthesized by Gene Pharma (China). Annexin V-FITC apoptosis/Propidium Iodide Detection Kit and Cell Counting Kit-8 (CCK-8 kit), Hoechst 33258, Daliaagle’s medium (DMEM) and fetal bovine serum (FBS) were from MeilunBio (Dalian, China). 2’,7’-dichlorofluorescin diacetate (DCFH-DA) were purchased from Aladdin Corporation (Shanghai, China). All fluorescence-labeled antibodies for immune cell analysis were purchased from BD or eBioscience.

Cells and animals

Mouse breast cancer (4T1) cells were purchased from the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China), and incubated in DMEM with 10% (v/v) FBS and 1% penicillin and streptomycin at 37℃ under 5% CO₂. To establish a mouse model bearing 4T1 tumors, a PBS solution containing approximately 10⁷ cells/mL was injected intradermally into the right axillary region of mice. Tumors were used for further analysis when tumor volume reached 100 mm³.

Preparations of ZD NPs and ZDS NPs

1.7 mg of DOX·HCl was added into a 50 mL volumetric flask, and dissolved with methanol, and then 1 mL of triethylamine was added into the flask. 2.4 mg of ZnCl₂ was dissolved in methanol. Subsequently, the solutions were mixed facilitated by ultrasonic treatment. After stirring for 12 h, 10 mg PVP was added to prevent the appearance of large agglomerated particles, then the solution was continually stirred for 36 h at 25℃ and centrifuged to obtain ZD NPs. ZD NPs were added to an equivalent volume of siRNA aqueous mixture. After gentle gassing and blending, the solution was placed at room temperature for 15 min to achieve ZDS NPs. The particular molar ratio was examined and screened by agarose gel electrophoresis.

Characterization of ZD NPs and ZDS NPs

The ZD NPs and ZDS NPs solution were examined by ultraviolet spectrophotometry and fluorescence spectrophotometry using Microplate Reader (Bio-Rad Laboratories Ltd., Hertfordshire, UK). DOX and ZD NPs, free-siRNA and ZDS NPs were respectively placed in 96-well plates for ultraviolet scanning spectrum and fluorescence scanning spectrum. DOX was drawn to investigate the changes in the ultraviolet absorption peak and fluorescence absorption peak of the solution after the coordination of DOX with Zn²⁺, ZD NPs and siRNA. The lyophilized ZD NPs were sampled by the KBr pellet method. Fourier transform infrared spectroscopy (FT-IR) analysis was carried out using an infrared spectrograph (Bruker, Switzerland).

The morphological characteristics of the nanoparticles was examined using transmission electron microscopy (TEM, H-7650, Hitachi, Tokyo, Japan). The zeta potential and particle size of ZD NPs and ZDS NPS were calculated by Zetasizer (Nano ZS, Malvern, Worcestershire, U.K.).

1 mL of ZD NP resolution was formulated to 1 mL of 0.2 mol/L HCl. The volumetric flask was filled to volume to prepare a nanoparticle solution containing 0.1 mol/L HCl, and sonicated for 20 s. Then, transferred 0.5 mL to a 10 mL flask and 0.1 mol/L HCl was added to replenish to volume. The absorbance value at a wavelength of 480 nm was measured by using a UV spectrophotometer, and the drug loading (LE%) and encapsulation efficiency (EE%) were computed.

m_e represents the total mass of DOX loaded onto nanoparticles, and m_t represents the total mass of DOX input. m_c represents the total mass of the input carrier.

The liberation profiles of ZD NPs and ZDS NPs were assayed by the dialyzing them. 1 mL of NPs was placed in a dialysis bag (with a molecular weight cutoff of 2.0 kDa). The samples were immersed in 40 mL of PBS (pH 7.4, 6.8, and 5.5, respectively) and incubated in a constant-temperature shaking incubator (100 rpm) at 37 °C. 1 mL of dialysates was taken and replaced with 1 mL of preheated PBS at 0, 1, 2, 4, 6, 8, 12, 24, and 48 h. The amount of DOX and siRNA released from ZDS NPs was determined quantitatively with a microplate reader.

In vitro serum stability assay

The in vitro serum stability of different formulations was characterized by agarose gel electrophoresis and changes in particle size in a 10% FBS. The stabilization of NPs in an in vitro mimetic physiological situation was also studied. ZD NPs and ZDS NPs were coincubated with a solution of PBS (pH 7.4) containing 10% FBS at 37 °C. Subsequently, the particle diameter of the nanoparticles was measured at 0, 1, 2, 3, 5, 10, and 15 days.

Cytotoxicity assessment in vitro

The cytotoxicity of various reagents was examined using the CCK-8 test. The procedure was as follows: 4T1 cells were inoculated at a concentration of 5 × 10³ cells per well in a 96-well plate. All marginal wells were filled with 200 µL of PBS. DOX·HCl, ZD NPs, and ZDS NPs were separately added at various concentrations (5, 10, 15 and 20 µg/mL) for 48 h. Afterwards, 10 µL of CCK-8 solution was introduced into every culture well and the incubation was carried out for 4 h at room temperature. After the reaction was completed, the absorbance value at 450 nm was determined with a microplate reader. Cell viabilities were computed from the OD values in accordance with the following formula.

where A_s represents the absorptivity of experimental wells (medium containing cells, CCK-8 solution, and drug to be tested); A_c represents the absorption of control wells (medium containing cells and CCK-8 solution); A_B represents the absorbance of blank wells (medium only containing CCK-8 solution).

Apoptosis

The induction of apoptosis by different nanoparticles was performed by using the apoptosis detection kit. 4T1 cells were inoculated into a 12-well plate at a database concentration of 5 × 10⁴ cells per well and incubated in a cell culture incubator for 24 h. DOX, ZD NPs and ZDS NPs were added separately, and incubate for 4 h. 5 µL PI and 10 µL Annexin V-FITC were added individually and incubated in the dark at 37 °C for 20 min. Subsequently, cells were suspended in 500 µL Annexin V binding buffer and passed through a 200-mesh sieve. Apoptosis was characterized by flow cytometry, and data were analyzed using Flow Jo software.

Cellular uptake

The 4T1 cells were inoculated into a 6-well plate at a population density of 1 × 10⁶ cells per well and cultivated overnight. FAM-siRNA, DOX, ZD NPs and ZDS NPs were added to each well respectively and incubated for 4 h. The original culture medium was disposed after incubation for 4 h and rinsed three times with PBS. These cells were digestion with trypsin, before being cleansed with PBS solution and resuspension in 500 µL of PBS solution, and subjected to flow cytometry after passing through a 200-mesh sieve. At minimum 10⁴ cells were collected from each specimen, and the fluorescence signals were analyzed using Flowjo software. The cells were then cleaned with PBS and processed with 4% polyformaldehyde and DAPI for fixation. The cell slides were pressed back onto the slides that had been prewrapped with the sealing agent, and the fluorescence imaging of the cells was observed with CLSM.

In vitro intracellular tracking study

CLSM was employed to measure the lysosomal escape behavior tracking of nanoparticles. The cell sheets were positioned in a 12-well plate. 4T1 cells (5 × 10⁴ cells/well) were inoculate ed, and cells were incubated culture box was kept overnight. FAM-siRNA-containing ZDS NPs dispersed in a complete medium were added into a 12-well plate and incubated for 1, 2, and 4 h. Subsequently, 300 µL lysosomal-red fluorescent probe Lyso-Tracker Red working solution was added to the cells, and cells were further incubated in for an additional 50 min. After washing with PBS 3 times, 250 µL 4% paraformaldehyde solution to each well and fix at room temperature for 15 min, then mounted in an anti-fluorescence quenching sealant, and placed under the CLSM for observation and photography.

In vitro ROS generation assay

4T1 cells were inoculated into a 6-well plate, allowed to attach, and then processed with the difference nanoparticles for 12 h. Afterwards, the culture medium was replaced with novel cell culture medium comprising DCFH-DA (10 µg/mL), and incubated for 30 min. Cells were then washed with PBS, fixated with 4% formalin, and stabilized with DAPI. Finally, the cells were observed by CLSM.

Western blotting

Western blotting was used to assess the influence of different formulations against p-TKB1, TKB1, p-STING, STING and Nrf2 protein. Briefly, 4T1 cells were inoculated into a 6-well plate at a database of 1 × 10⁴ cells per well, permitted to attach, and then treated with different siRNA-containing formulations (equivalent to a FAM siRNA concentration of 100 nM) and further cultured for 48 h. RIPA buffer was utilized to extract the complete protein and the concentration of the total protein was measured using a BCA kit. Subsequently, 20 µL of protein was obtained for SDS-PAGE, then protein transfer to a PVDF membrane and standard procedures were followed for primary and secondary antibody incubation. Protein strips were developed using an ECL and a ChemiDoc XRS+ (Bio-Rad, USA).

In vitro ICD induction

In vitro CRT expression

For CRT expression detection, cells were inoculated onto coverslips in a 6-well plate. The cells were processed with DOX, ZD NPs, and ZDS NPs for 24 h. Subsequently, the cells were cleaned twice with PBS and immobilized with 4% paraformaldehyde. The cells were then respectively cultured with anti-CRT prime antibody and Alexa 488-conjugated second antibody and visualized by CLSM. Concurrently, cells underwent PI staining for flow cytometric analysis to identify CRT⁺ PI⁻ cell subpopulations.

In vitro ATP and HMGB1 release

4T1 cells were inoculated into 6-well plates and co-cultured with different nanoparticles for 24 h. Following the manufacturer’s operating procedures, HMGB1 and ATP levels in the cell culture supernatants were measured using an ATP assay kit and an HMGB-1 detection ELISA kit, respectively.

In vitro DC maturation assay

4T1 tumor cells were cultured in a 24-well plate with the corresponding Transwell chambers. The 4T1 cells were cultured in the culture medium containing different preparations (DOX, ZD NPs, and ZDS NPs) (pH 6.8). The cells in the upper chamber of the Transwell, were embedded in the 24-well plate for DCs. The 4T1 cells and DCs cells were co-cultured for 24 h.

In vivo anti-tumor efficacy

Mice bearing 4T1 tumors were divided into four groups (n = 6) at random. Normal saline (control), DOX, ZD NPs, and ZDS NPs were injected into the tail vein (the concentrations of DOX and siRNA were 5 mg/kg and 1 mg/kg, separately, and the amount of normal saline in the control group was 10 mL/kg). When the 4T1 tumors grew to a certain volume, the tail vein injections were administered once every 3 days for a total of 5 doses. During this period, mice were weighed daily, and the length and width of the tumors were calculated using Vernier calipers to calculate the tumor size. At the end of the experimental period, mice were, and the tumors and major organs were harvested. The surface of the tumors was cleaned with normal saline, and the tumors were weighed after being blotted dry with filter paper, and photos were taken. Subsequently, tumor tissue and major parts (heart, liver, spleen, lung, and kidney) were taken for H&E (Hematoxylin-Eosin/HE Staining Kit, Solarbio Science & Technology, Beijing, China) and TUNEL (Colorimetric TUNEL Apoptosis Assay Kit, Beyotime Biotechnology, Dalian, China). ROS were visualized via immunofluorescence staining. The tumor inhibition rate (TIR) was computed using the following formula.

W_c and W_t represent the tumor weight of control and treatment group at the end of experiment, respectively.

In vivo antitumor immune effect of ZDS NPs

The 4T1 cells were implanted into the primary tumors on the − 9 day, and on the 1 st day, the 4T1 tumor cells were intravenously injected to induce lung metastasis. A lung metastasis tumor-bearing mice model was established to investigate the antitumor immune effects of different NPs. Briefly, mice were randomly divided into several groups (n = 6) and received tail vein injections of saline, DOX, ZD NPs and ZDS NPs (5 mg/kg of DOX) every 3 days. Mice from each group were euthanized on day 16, and their major organs (lung tissues) were obtained. Tumor tissues were gathered for H&E staining, TUNEL staining, and CRT staining. Pulmonary metastatic nodules were counted and collected for H&E staining.

Statistical analysis

All data are presented as mean ± standard deviation (SD). Statistically significant differences were determined using one-way ANOVA and was set at *p < 0.05, **p < 0.01 and ∗∗∗p < 0.001. All experiments were performed in triplicate.

Results and discussion

Characterization of ZD NPs and ZDS NPs

ZD NPs were prepared by metal complexation and coordination of phenolic hydroxyl groups of DOX with Zn²⁺ (Fig. 2A). The UV-vis showed the absence of characteristic peak at 480 nm of DOX in ZD NPs, indicating the successful fabrication of NPs (Fig. 2B). This phenomenon was attributed to loss of the characteristic absorption peak of DOX following complexation with Zn²⁺ [34]. The fluorescence absorption spectra showed that the fluorescence absorption peak of ZD NPs was significantly lower than that of DOX at 470 nm (Fig. 2C), which was due to the disappearance of the characteristic absorption peak of DOX after complexation with Zn²⁺. Infrared absorption spectroscopy showed that the characteristic absorption peak of hydroxyl group at 3425.7 cm^{− 1} of DOX disappeared (Fig. 2D). This was due to the fact that Zn²⁺ coordinated with the oxygen atom in the hydroxyl group by attracting negative electrons on the hydroxyl group. The abovementioned results indicated the successful preparation of ZD NPs. The EE % and LE % of ZD NPs were summarized (Tab. S1). The EE% of ZD NP was 99.42 ± 0.72%. Furthermore, the loading efficiency of siRNA onto ZD nanoparticles was 89.0% (Fig. S2). Given that DOX functions as both a carrier and a therapeutic agent, the encapsulation efficiency of the drug was notably elevated.

Fig. 2.

(A) The preparation of ZD NPs and ZDS NPs. (B) The fluorescence signal change of ZD NPs and DOX. (C) The UV absorption signal change of ZD NPs and DOX. (D) The FT-IR spectra of ZD NPs and DOX. (E) The agarose gel electrophoresis of ZDS NPs at different molar ratios (ZD NPs: siRNA). (F) The fluorescence signal change of ZDS NPs and FAM-siRNA. (G) The UV absorption signal change of ZDS NPs and FAM-siRNA. (H) The zeta potential of ZD NPs and ZDS NPs. (I) Particle size distribution of ZD NPs and TEM image. Scale bar: 100 nm. (J) Particle size distribution of ZDS NPs and TEM image. Scale bar: 100 nm. (K) Hemolysis ratio of DOX, ZD NPs, and ZDS NPs, respectively (n = 3) and images of hemolysis test. In vitro accumulative release of DOX (O) and siRNA (P) from ZDS NPs (n = 3)

Subsequently, the self-assembly behavior between ZD NPs and siRNA was investigated. Initially, the capability of ZD nanoparticles to assemble with siRNA was optimized via agarose gel electrophoresis (Fig. 2E). At a molar ratio of 3: 1 between ZD NPs and siRNA, the complete incorporation of siRNA was realized. The characteristic peaks of FAM-siRNA disappeared in the fluorescence spectrum of the ZDS NPS (Fig. 2F). Furthermore, compared with FAM-siRNA, the UV fluorescence absorption peak of ZDS NPs was significantly decreased (Fig. 2G). The reduction of ultraviolet absorption peak and the decrease of fluorescence peak indicated the successful preparation of ZDS NPs. ZD NPs displayed mean size of 170.3 ± 3.4 nm with negatively surface charge of −15.4 ± 2.1 mV(Fig. S1A). After siRNA modification, the particle size of ZDS NPs slightly increased to 160.3 ± 1.7 nm, and ζ-potential became to −13.9 ± 1.9 mV (Fig. S1B & Fig. 2H). ZDS NPs were characterized as rounded spheres of uniform size, as evidenced by TEM images (Fig. 2I-J). The particle size of ZDS NPs was marginally smaller than that of ZD NPs, primarily due to the surface loading of siRNA, which electrostatic compression contributed to the reduction in particle size [35].

Since the NPs were administered intravenously during application, in vitro hemolytic test was performed to verify the hemocompatibility of ZDS NPs. The hemolysis ratio of DOX, ZD NPs, and ZDS NPs were 7.61, 4.27, and 3.52%, respectively, as it was distinctly observed that all nanoparticles showed transparent and without hemolysis trend (Fig. 2K). The crucial safety threshold for hemolysis in biomaterials was being defined as below 5%. Therefore, ZDS nanoparticles with superior biocompatibility were employed for subsequent experiments.

The particle size was dynamically monitored to evaluate the stability of nanoparticles in various media. The particle size and PDI of ZD NPs and ZDS NPs were changed slightly within 15 days in PBS (pH = 7.4) with 10% FBS at 37 °C. After 15 days, no significant change of size and PDI were observed after being stored in PBS (pH = 7.4) with 10% FBS at 37 °C (Fig. S3-S4).

A serum stability assay was conducted to assess the stability of siRNA encapsulated within nanoparticles, with the objective of determining whether ZDS nanoparticles possessed the capability to safeguard siRNA from degradation. Free siRNA was completely degraded in serum for 3 h, whereas ZDS NPs effectively protected the siRNA for at least 24 h (Fig. S5). Through the nucleases digestion experiment. Under the protection of ZDS NPs, after incubation of RNase A for 30 min, the siNrf2 band that was released from the vector remained bright, while the naked siRNA without vector protection completely disappeared under the same conditions of incubation. This indicates that ZDS NPs can protect siNrf2 from being degraded by nucleases (Fig. S6). This may be due to the steric hindrance of siRNA on the surface of the nanoparticles, reducing direct contact of RNases with siRNA. The results showed that ZDS NPs had the ability to protect siRNA from degradation.

The accumulated release rate of DOX from ZDS NPs was quantitatively determined. Under weakly acidic temperatures of pH 5.5 and pH 6.8, the release of DOX occurred at a higher rate, compared to the condition at pH 7.4. After a duration of 48 h, the cumulative release percentages of ZDS NPs were recorded as 50.23 ± 1.13%, 45.57 ± 0.47%, and 38.23 ± 0.14%, respectively, in pH 5.5, 6.8 and 7.4. These results indicated ZDS NPs had a favorable pH-responsive behavior (Fig. 2O). The tumor microenvironment (TME) had a lower pH than normal physiology, and the pH difference was used to control drugs release in response to specific stimuli [36]. The quinone moiety of DOX, in conjunction with the phenolic hydroxyl group and Zn²⁺, facilitated the formation of a metal complex. Upon arrival at the TME, the acidic conditions triggered the release of DOX. Similarly, the pH level decreased from 7.4 to 5.5, resulting in an increase in siRNA release from 44.1 ± 0.76% to 64.7 ± 0.38%. This observation suggested a pH-dependent mechanism governing the release of siRNA. This phenomenon was mainly due to the pH-sensitive property of hydrogen bond between siRNA and DOX [37]. Agarose gel electrophoresis further proved that ZDS nanoparticles exhibited a pH-dependent siRNA release behavior (Fig. 2P).

Apoptosis and cytotoxicity

In vitro cytotoxicity assays were conducted using 4T1 tumor cells. DOX, ZD NPs, and ZDS NPs induced cell death in a concentrations-dependent fashion (Fig. 3A), and the IC₅₀ value of DOX, ZD NPs and ZDS NPs were 16.585 ± 3.68, 12.379 ± 0.570, and 10.674 ± 1.788 µg/mL, respectively (Fig. 3B). In vitro cytotoxicity assays were conducted using H9C2 myocardial cells. After incubation with ZD NPs and ZDS NPs, the survival rate of cardiomyocytes was greater than 90%. The cell survival rate in the DOX group was above 60% (Fig. S12). The reason for the lower cell survival rate in the DOX group was that DOX caused extensive damage to cardiac cells. This further demonstrated that ZDS NPs can effectively protect the heart from the cardiotoxicity caused by DOX. These findings demonstrated that ZDS NPs had significant cytotoxic effects. Furthermore, ZDS NPs demonstrated a significant capacity to induce apoptosis in cells, with an observed apoptotic rate of 51.0%, which was much higher than that of ZD NPs and DOX (Fig. 3C). In contrast, the apoptosis induced by siRNA-free nanoparticles and DOX were not obvious, suggesting a significant correlation between ROS production and apoptosis [38]. The results suggested that ZDS NPs promoted apoptosis by producing excess ROS, thereby achieving more efficient anti-tumor effects.

Fig. 3.

Cytological studies. (A) Cytotoxicity analysis of DOX, ZD NPs and ZDS NPs. (B) IC₅₀ values of 4T1 cells after co-culture with different nanoparticles. (C) Flow cytometry analysis of apoptosis induced by different nanoparticles after 4 h. The internalization of various nanoparticles delivering siRNA by 4T1 cells was analyzed (D) and quantified (E) by flow cytometry. The internalization of various nanoparticles delivering DOX by 4T1 cells was analyzed (G) and quantified (H) by flow cytometry. (F) and (I) The internalization of various nanoparticles delivering siRNA by 4T1 cell was analyzed by CLSM. Scale bar = 50 μm. (J) CLSM images of lysosomal escape status in 4T1 cells under different time points. Scale bar = 10 μm (K) Observing the in vitro ROS production in 4T1 tumor cells by different nanoparticles using CLSM. Scale bar = 50 μm (*p < 0.05, **p < 0.01, ***p < 0.001)

Cellular uptake of ZD NPs and ZDS NPs

Cell internalization studies were conducted on the 4T1 cell line using flow cytometry and CLSM. Initially, we investigated the capacity of 4T1 cells to internalize siRNA. The uptake curves had shifted significantly to the right, indicating that they were all taken up by 4T1 cells in varying degrees. Cells had a higher cellular uptake capability against ZDS NPs, compared with FAM-siRNA (Fig. 3D-E). The results were also confirmed by the CLSM images (Fig. 3F). The elevated uptake rate of ZDS NPs was attributed to the instability of free siRNA. Subsequently, we investigated the capacity of 4T1 cells to internalize DOX. The order of nanoparticle uptake by 4T1 cells was as follows (Fig. 3G-H). The results were also confirmed by CLSM images (Fig. 3I). The increased cellular uptake of ZDS NPs and ZD NPs compared to DOX was attributed to the enhanced affinity of lipophilic particles for cellular membranes relative to hydrophilic molecules.

Lysosomal escape assay

To further evaluate lysosomal escape function, the intracellular distribution of ZDS NPs was determined using CLSM. The CLSM analysis showed the colocalization of FAM-siRNA (green) and lysosomes (red) as yellow pigmented spots. The ZDS NPs agents exhibited the co-localization phenomenon with the cell line at all time points, indicating that the majority of agents were present within lysosomes. However, within 2–4 h, the yellow fluorescence gradually diminished, while the green fluorescence continued to intensify in the cell line (Fig. 3J), suggesting that the antitumor drugs were discharged from late endosomes into the cytoplasm. In summary, ZDS NPs successfully delivered siRNA to the cytoplasm.

Intracellular ROS generation

DCFH-DA was utilized as a fluorescent probing agent for the detection of ROS by ZDS NPs (Fig. 3K). Low fluorescence signal was observed in the Control group, suggesting relatively that the concentration of ROS was relatively low. However, after incubation with DOX and ZD NPs, the green fluorescence increased, with ZDS NPs exhibiting a more pronounced fluorescence signal. This phenomenon was assigned to the presence of slightly acidic relations. DOX exerted cytotoxic effects and triggered ICD effects, and Zn²⁺ effectively increased ROS production, siRNA reduced GSH levels by reducing Nrf2 protein levels. By increasing ROS and reducing GSH levels, the redox homeostasis was effectively disrupted.

Western blot assay

The phosphorylation levels of key molecules in the cGAS/STING signaling pathway (TBK1 and STING) are significantly elevated [39]. Therefore, we examined the expression of TKB1 and STING by western blotting to elucidate the inhibitory ability of the ZDS NPs on cGAS/STING expression (Fig. 4A & Fig. S8A). The analysis showed no significant variation between the Control and DOX group. These results indicated that the cGAS/STING pathway was suggested by ZD NPs, and the Zn²⁺, incorporated within the NPs effectively activated the cGAS/STING pathway (Fig. 4B-C & Fig.S8B-C).

Fig. 4.

(A) Western blot analysis of p-TBK1, TBK1, p-STING, STING, Nrf2 protein. (B) Quantitative analysis of p-TBK1 and TBK1 expression levels. (B) Quantitative analysis of the expression level of p-STING and STING. (D) Quantitative analysis of Nrf2 expression levels. (E) Quantitative ELISA of ATP release. (F) Quantitative ELISA of HMGB1 release. (G) Extracellular CRT exposure detection by flow cytometry. (H) and (I) In vitro maturation of DCs (CD80⁺ CD86⁺, the matured DCs were gated on CD11c⁺) and (G) frequency of matured DC. (J) CRT exposure of 4T1 cells. Statistical significance was determined by one-way ANOVA (***p < 0.05, ***p < 0.01, ***p < 0.001)

Expression of Nrf2 was detected to elucidate the inhibitory ability on Nrf2 expression (Fig. 4A, D) [38, 39]. No significant differences were observed between the Control group, DOX group, and ZD NP group. In contrast, Nrf2 expression in ZDS NPS group was significantly lower, indicating a substantial reduction. Furthermore, we conducted a qPCR test to measure the transfection efficiency of the siRNA targeting Nrf2. The results showed that Nrf2 was effectively reduced (Fig. S9). Therefore, these results demonstrated that ZDS NPs effectively delivered siNrf2 to the cytoplasm and inhibited Nrf2 expression.

ICD induction assay by ZDS NPs

During the ICD response cascade, various signals emitted by DAMPs include the expression of CRT, the liberation of HMGB1, and the secretion of ATP [37]. DAMPs enhance as “eat me” signals that facilitate the recruitment of APCs, including DCs. These signals enable the recognition and phagocytosis of antigens, presenting them to naïve T cells to activate T lymphocyte-mediated antitumor immune responses. Consequently, DAMPs perform a crucial role in recognizing and clearing terrorist antigens, thereby promoting the generation of anti-tumor immune reactions [40, 41].

Firstly, the capacity of various nanoparticles to induce ICD was evaluated through the quantification of ATP release [42]. In comparison to DOX group, ATP secretion in the ZDS NPs group was significantly increased, which was due to the ability of ZDS NPs to release siRNA under tumor acidic microenvironment and effectively reduce the production of Nrf2 protein (Fig. 4E). The release of HMGB1 in the ZD NPs group and the ZDS NPs group was found to be greater than that observed in the DOX group, suggesting that Zn²⁺ enhanced HMGB1 release through the augmentation of ROS production. Furthermore, a significant increase in HMGB1 levels was detected in the ZDS NPs group compared to the ZD NPs group, providing evidence that ROS resulted in the release of additional HMGB1 and thereby inducing ICD effect. The concentration of HMGB1 observed in the ZD NPs group was similar to DOX group. However, the concentration in the ZDS NPs group was markedly elevated, compared to ZD NPs group. This finding provided key evidences that ZDS nanoparticles demonstrated the ability to dissociate and release siRNA within the tumor microenvironment, leading to a reduction in Nrf2 levels and an elevation in ROS levels, which subsequently promoted the induction of ICD (Fig. 4F).

CRT functions as a critical biomarker for identifying tumor cells undergoing ICD [43]. The study examined the induction of CRT expression after various nanoparticles treatments. The fluorescence of ZDS NPs group was greater than that of ZD NPs group, demonstrating that ZDS NPs induced an increase in CRT expression (Fig. 4G-J). The CRT induced by ZDS NPs was higher than that of ZD NPs, which was due to siRNA could effectively reduce the production of Nrf2 protein, resulting in an increase in ROS and a stronger ICD effect. In the flow cytometry analysis, the CRT⁺PI⁻ subgroups exhibited similar tendencies, thus validating the pivotal role of ZDS NPs in increasing intracellular ROS levels.

Apoptosis generates “red flags” through the ICD pathway and presents tumor-associated antigens (TAAs) to immature DCs for dendritic cell maturation, which plays a crucial role in initiating systemic immune responses [5]. Transwell model was used to measure the maturation ratio of DCs following treatment with different nanoparticles. The maturation of DCs in the lower chamber was induced by tumor-associated antigens released from apoptotic tumor cells located in the upper chamber [44]. The extent of maturation of DCs in various experimental groups was subsequently assessed. Compared with PBS, ZDS NPs could effectively increase DC maturation ratio to 47.9%, thereby induced ICD effects (Fig. 4H-I). In summary, the results of HMGB1 efflux, CRT expression and ATP secretion demonstrated that ZDS NPs enhanced ICD effect of tumor cells by increasing ROS, thereby further promoting the maturation of DCs and activating anti-tumor immunity [45].

Anti-tumor efficacy of ZDS NPs

The antitumor activity of different nanoparticles was evaluated (Fig. 5A). The mice treated with DOX exhibited rapid growth curve, similar with Control group. The ZDS NPs group with a TIR_V value of 86.4 ± 0.72% (Table. S2) demonstrated the minimal tumor volume, lowest tumor weight, and maximum survival time, confirming the superior therapeutic efficacy of the combined administration of DOX, Zn²⁺ and Nrf2 siRNA (Fig. 5B-G). ZDS NPs significantly enhanced the generation of ROS, thereby more effectively suppressing tumor progression and resulting in enhanced therapeutic outcomes. A fluorescence quantitative analysis of DOX was conducted on the mice 12 h.

Fig. 5.

(A) Schematic diagram of the administration design. (B) The images of tumor tissues separated from 4T1 tumor-bearing mice after the treatment. (C) Tumor growth curve of tumor-bearing mice injected with different DOX formulations. (D) Survival curves of mice after receiving different formulations. (E) The weights of excised tumors from 4T1 tumor-bearing mice after the treatment. (F) Body weight curves of tumor-bearing mice treated with different preparations. (G) Net body weight change in 4T1 bearing mice administered different formulations. (H) The representative Western blotting images showing the expression levels of p-TBK1, TBK1, p-STING, STING, Nrf2 from different treatment groups in vivo. (I) Quantitative analysis of the expression level of p-TBK1 and TBK1 in vivo after treated with different NPs. (J) Quantitative analysis of the expression level of p-STING and STING in vivo after treated with different NPs. (K) Quantitative analysis of the expression level of Nrf2 in vivo after treated with different NPs. Statistical significance was determined by one-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001)

after administration. The results indicated that the accumulation levels of DOX in important tissues such as the liver, spleen, and tumor, as well as the tissue distribution results in tumor-bearing mice, could prove the feasibility of the targeted drug delivery strategy of ZDS NPs (Fig. S10).

Validation of relevant protein expression

Western blot was used to detect the relevant protein expression (Fig. 5H). In comparison to Control groups, the phosphorylations of STING (p-STING) in the ZD NPs and ZDS NPs groups were markedly elevated, thereby validating the effective activation of STING signaling pathway through the release of Zn²⁺. Upon activation, STING recruits and subsequent activates TANK-binding kinase 1 (TBK1) [46]. As a result, the levels of phosphorylated TBK1 (p-TBK1) were elevated in the ZD NPs and ZDS NPs, indicating effective activation of cGAS-STING signaling pathway. Quantitative western blot analysis corroborated the aforementioned findings (Fig. 5I-J). In conclusion, the findings indicated that Zn²⁺ activated cGAS/STING signaling pathway.

Nrf2 is a critical element in redox processes and serves as an upstream regulator of GSH. Western blot was used to detect the expression of Nrf2 to elucidate the inhibitory ability of Nrf2 expression. The Control group, DOX group, and ZD NPs group exhibited no statistically significant alterations (Fig. 5H-K). The ZDS NPs demonstrated a markedly reduced signal intensity, indicating a successful downregulation of Nrf2 protein expression.

Histopathological examination

To explore the antitumor effects and the immune response of various formulations, histological evaluation was taken. H&E dyeing in the Control group revealed that the nuclei of tumor cells were enlarged and deeply stained, indicating vigorous growth of tumor tissue (Fig. 6A). DOX and ZD NPs groups showed more pink tumor cell necrosis foci, a small amount of nuclear dissolution, and still some loosely arranged non necrotic tumor cells with a high nuclear to cytoplasmic ratio, indicating that the growth of tumor cells was inhibited. Tumor tissues of ZDS NPs group showed large areas of necrosis, with almost all tumor tissue necrotic. And the nuclei were mostly in a condensed and incomplete state, indicating that their anti-tumor efficacy was efficient. According to Ki67 immunofluorescence result, ZDS NPs group showed no large proliferative areas of tumor cells. In comparison to DOX and ZD NPs, the average fluorescence strength of Ki67 was significantly reduced in mice processed with ZDS NPs (Fig. 6B). TUNEL staining revealed large apoptotic areas in the ZDS NPs group. Through CRT staining sections, it was shown that the ZDS NPs effectively induced CRT eversion, indicating their capacity to initiate the ICD. Compared to DOX and ZD NPs, the mean fluorescence intensity of ROS, TUNEL, and CRT in ZDS NPs-treated mice was significant higher (Fig. 6C-E). Activation of the anti-tumor immune response in the body leads to changes in cytokine secretion within the tumor tissue. Therefore, we used ELISA to detect the changes in cytokines in the tumor tissue and spleen. Compared with the control group, the TNF-α level in the tumor tissue of mice in the ZDS NPs experimental group was significantly increased. (Fig. S7). It indicates that ZDS NPs can activate the inflammatory response. In conclusion, ZDS NPs group demonstrated a significant capacity to suppress tumor cell proliferation, promote apoptosis in tumor cells, and facilitate the exposure of CRT on the cell surface through the generation of ROS, thereby initiating ICD response.

Fig. 6.

(A) H&E, Ki67, TUNEL, ROS, and CRT staining of the tumor for all groups. Quantitative results of the fluorescence intensity of Ki67 (B), ROS (C), TUNEL (D) and CRT (E). Statistical significance was determined by one-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001)

Assessment of antitumor efficacy in pulmonary metastasis

The antitumor efficacy against pulmonary metastases was evaluated (Fig. 7A). Lung tissue was collected for H&E dyeing to visualize the number of metastatic foci (Fig. 7B-C). The findings indicated that the incidence of lung metastases in ZDS NPs group was markedly reduced compared to other groups. This reduction was attributed to the capacity of ZDS NPs to induce elevated levels of ROS, subsequently triggering ICD effect and eliciting a systemic immune response. Consequently, ZDS NPs demonstrated the most pronounced inhibitory effect on lung metastasis. Primary tumors were collected for histological analysis to assess the in vivo antitumor efficacy of ZDS NPs (Fig. 7D). ZDS NPs demonstrated a significant ability to induce apoptosis or necrosis in tumor cells and effectively suppressed tumor proliferation. This finding further underscored their promising potential for application in cancer immunotherapy. To investigate the mechanism underlying the excellent antitumor activity of ZDS NPs, CRT expression of primary tumor tissues was observed. TUNEL staining demonstrated the presence of extensive apoptotic regions in ZDS NPs group. Additionally, analysis of CRT staining sections indicated that ZDS NPs successfully facilitated CRT eversion, thereby confirming its capacity to initiate ICD cascade reactions. Compared to DOX and ZD NPs groups, the mean fluorescence intensity of TUNEL and CRT in ZDS NPs was significantly elevated (Fig. 7E-F). In conclusion, ZDS NPs demonstrated a significant efficacy in suppressing tumor cell proliferation. The underlying mechanisms involved the generation of ROS, activation of immune cells-mediated cytotoxicity, stimulation of the overall immune response, and the effective inhibition of tumor metastasis.

Fig. 7.

(A) Treatment schedule of ZDS NPs against lung metastasis in 4T1 tumor-bearing mice (B) H&E staining of lung sections and (C) number of lung metastatic lesions on the 16th day (D) H&E staining, TUNEL staining and CRT staining of the tumor for all groups. Quantitative results of the fluorescence intensity of TUNEL (E) and CRT (F). Statistical significance was determined by one-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001)

In vivo safety evaluation

The histopathological analysis was performed utilizing H&E staining to assess in vivo biocompatibility of nanoparticles (Fig. S13). After treatment, the main organs of each group were taken and paraffin sections were prepared for H&E staining. The results of cardiac tissue slices showed that DOX group exhibited significant myocardial rupture, while ZD NPs group showed a reduction in myocardial rupture. It was worth noting that the ZDS NPs group had tightly arranged myocardial fibers with clear boundaries and no obvious myocardial cell damage, indicating that ZDS NPs could significantly reduce the cardiac toxicity of DOX. The concentrations of AST, ALT, BUN and Cre in the serum of tumor-bearing mice after different treatments were measured to evaluate the liver and kidney function indicators of the mice and assess the safety of the formulation in vivo. As shown in the figure, after treatment with each group of the formulation, the levels of ALT, AST, BUN and Cre did not show significant changes, indicating that the liver and kidney functions of the mice were not impaired (Fig. S11). No pathological abnormalities or damage were found in other organs, indicating that ZDS NPs had a good biocompatibility.

Conclusion

In this work, we developed a carrier-free nanoparticle, composed of DOX, Zn²⁺ and siRNA, via multiple driving forces used for effective approach designed to enhance antitumor immune responses while suppressing tumor metastasis. ZDS NPs increased the level of ROS through Zn²⁺, and decreased the content of GSH by reducing the expression of Nrf2 through siRNA, thereby aggravating redox homeostasis. By disrupting the redox homeostasis of cells, could effectively increase the intracellular ROS levels and effectively induce the ICD effect in tumor cells, thereby triggering an autoimmune response. These findings demonstrated that the novel carrier-free self-assembled nanoparticles exhibited a potent anti-tumor efficacy against both primary and metastatic tumors, thereby offering novel approaches for cancer therapy.

Author contributions

Jingya Xiu：Methodology, Formal analysis, Data Curation, Writing-Original Draft. Lin Liu, Lingao Yin, Yihong Peng, Ziqi Li, Ziyun Lin, Yilin Song: Supervision, Investigation, Resources. Degong Yang, Jiulong Zhang, Chunrong Yang: Conceptualization, Writing- Review & Editing.

Funding

Not applicable.

Data availability

The dataset(s) supporting the conclusions of this article is(are) included within the article (and its additional file(s)).

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Experimental Animal Administrative Committee of Shenyang Pharmaceutical University (Approval No. SCXK2019-0008). Procedures for the animal experiments were implemented under protocols of the National Regulation of China for Care and Use of Laboratory Animals.

Consent for publication

All authors of this study agreed to publish.

Competing interests

The authors declare no competing interests.