Sonodynamic therapy (SDT), which leverages the synergistic effects of ultrasound (US) and sonosensitizers (SNSs), effectively overcomes the limitations associated with traditional photodynamic therapy (PDT) [1,2]. US accurately targets the tumor region, activating the SNS to kill tumor cells [1,3]. Due to its non-invasiveness and better therapeutic effect on deep tumors, SDT may become a promising potential cancer treatment method. Many studies have shown that SDT not only directly kills cancer cells, but also regulates the tumor microenvironment (TME) to enhance efficacy [4–6]. SDT can efficiently induce immunogenic cell death (ICD) to promote inflammatory cell infiltration [7–9]. Tumor cells undergoing ICD can release a variety of damage-associated molecular patterns (DAMPs), such as high-mobility group box 1 (HMGB1) and calreticulin (CRT) [10,11]. These DAMPs effectively promote dendritic cells (DCs) maturation, and subsequently facilitating the infiltration of T cells into the tumor and activating anti-tumor immunity [12]. In addition, SDT can also inhibit tumor angiogenesis to a certain extent [13]. However, SDT generally consumes oxygen during the induction of cytotoxicity, exacerbating hypoxia within solid tumors, which in turn further activates the hypoxia-inducible factor-1 (HIF-1) pathway in tumor cells [14]. It is known that the HIF-1 pathway is activated under hypoxic conditions and then maintained stable, endowing tumor cells with a survival advantage, accelerating their migration and invasion, as well as suppressing immunity and promoting evasion [15,16]. These factors make the application of SDT still several limitations. Importantly, the sonodynamic properties of the SNS also play a key role in the treatment process. Thus, inhibiting HIF-1 pathway and developing high-performance SNSs are very important for synergistic SDT [17].

The latest studies have elucidated that endoplasmic reticulum (ER) stress can promote the proteasomal degradation of HIF-1α [18]. It is important to note that the potential for mitigating challenges faced by SDT through endoplasmic reticulum stress-induced HIF-1α degradation has not been fully explored, which would be very meaningful research. In-depth investigation of the interplay between ER stress and the HIF-1 pathway may provide novel insights and strategies for more effective SDT, warranting further research. Moreover, ER stress is closely associated with ICD, triggering the unfolded protein response (UPR) and the effective release of DAMPs [19]. Besides, ER stress has long been implicated in the activation of the cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway, but whether ER stress activates the cGAS-STING pathway remains to be definitively explored [20].

The cGAS-STING pathway is a key mediator in triggering the innate immune response [21]. The activation of STING stimulates the release of type Ⅰ interferons (IFN) [22], recruits T cells for tumor infiltration, promotes the maturation of DCs [23], and polarizes tumor-associated macrophages towards an anti-tumor M1 macrophage phenotype [24]. Recently, the therapeutic strategy employing STING agonists has been demonstrated to be effective in eradicating residual cancer cells and preventing metastatic relapse [25,26]. These exciting advancements substantiate the efficacy of activating the cGAS-STING signaling pathway as a potential oncological treatment modality. However, the performance of monotherapy with STING agonists in clinical trials has not met the anticipated outcomes. It suggests that activation of the STING pathway may require a combination with existing cancer treatment modalities to enhance anti-tumor efficacy [27].

Metal-organic complexes, such as iridium, platinum and rhenium, have shown immense potential as effective SNSs due to their superior reactive oxygen species (ROS) generation efficiency and potent tumoricidal properties [28–30]. Iridium(Ⅲ) complexes possess a long-lived triplet excited state that facilitates energy transfer with molecular oxygen to produce singlet oxygen (¹O₂) [31]. Therefore, iridium(Ⅲ) complexes have an advantage over other metal elements as SNSs or photosensitizers. Given the potential of ER stress in degrading HIF-1α and its implications for SDT, we have developed an ER-targeting SNS (C₆IrAC) [32]. The chlorambucil structure endows C₆IrAC specific ER targeting capability, thereby inducing ER stress and subsequent apoptosis in tumor cells. Under US treatment, the in situ burst of ROS in the ER further aggravates oxidative stress and cell apoptosis. A large number of research evidence confirmed that C₆IrAC significantly amplifies the ICD effect and releases a substantial amount of DAMPs. C₆IrAC-induced ER stress activates the cGAS-STING pathway, releasing abundant type Ⅰ IFN, thereby activating innate immunity. Concurrently, C₆IrAC promotes the degradation of HIF-1α, effectively inhibiting the HIF-1 pathway, and thus enhanced SDT to suppress the survival, migration and immune evasion of tumor cells. Taken together, the main focus of our current study is illustrated in Scheme 1: (1) A specific SNS targeting the ER, C₆IrAC, was designed based on theoretical calculations. Synthesized C₆IrAC was utilized for fluorescence imaging-guided SDT in vivo. (2) C₆IrAC induces ER stress to directly kill cancer cells and triggers ICD, reshaping the tumor immune microenvironment. (3) We propose for the first time that inducing ER stress through an ER-targeting SNS can activate the cGAS-STING pathway and suppress the HIF-1 pathway. The multifunctional ER-targeting iridium(Ⅲ) SNS provides a compelling alternative strategy for treating cancer by enhancing ER stress and offers a feasible approach to overcoming the limitations of SDT, paving a new avenue for the design of SNSs.

The synthesis route of the ER-targeting organometallic SNS C₆IrAC was depicted in Scheme S1 (Supporting information). The successful synthesis of the chromophore C₆IrAC has been corroborated by ¹H nuclear magnetic resonance spectroscopy (NMR), ¹³C NMR and high-resolution mass spectrometry (HRMS) analysis (Figs. S1 and S2a in Supporting information). To further investigate the physicochemical properties of C₆IrAC, theoretical calculations of its energy levels were performed using the density functional theory (DFT) method with the PBE0 (D3BJ)/6–31 (d) basis set as implemented in Gaussian 09 (version E01). The highest occupied molecular orbital (HOMO) of C₆IrAC is localized on the phenanthroline ligand, whereas the lowest unoccupied molecular orbital (LUMO) is delocalized across the iridium atom and the coumarin moiety, suggesting efficient intramolecular charge transfer within the molecular framework (Fig. 1a, Figs. S2b and c in Supporting information). Electrostatic potential (ESP) analysis was conducted to elucidate the distribution of electron density within the C₆IrAC molecule (Fig. 1a). The negative electrostatic potential is predominantly localized on the oxygen atoms of the coumarin moiety, while the positive electrostatic potential exhibits a more dispersed distribution. The HOMO-LUMO and electron density for coumarin and chlorambucil groups were shown in Figs. S2b and c. The energy gaps for C₆IrAC, coumarin, and chlorambucil group were calculated to be 1.294, 3.788, and 5.529 eV, respectively. Notably, the energy gap of C₆IrAC is significantly lower compared to the other two compounds, suggesting that C₆IrAC can be easily excited by stimuli at lower energy levels. The ultraviolet-visible (UV–vis) spectrum of C₆IrAC in phosphate-buffered saline (PBS) exhibits notable absorption within the 400–525 nm wavelength range, with a pronounced absorption peak centered at approximately 480 nm (Fig. 1b). Upon excitation at 480 nm, the fluorescence emission spectrum of C₆IrAC is characterized by a peak around 590 nm (Fig. 1c).

To evaluate the sonodynamic performance of C₆IrAC, 1,7-dichlorodihydrofluorescein diacetate (DCFH-DA) was utilized as a probe to assess the ROS generation capability of C₆IrAC upon US activation. The characteristic peak of DCF at 525 nm exhibited a progressive increase with time throughout US stimulation, which robustly indicated that C₆IrAC produced ROS under US irradiation (Fig. 1d). Furthermore, the production of specific ROS species was confirmed using electron spin resonance (ESR) with 2,2,6,6-tetramethylpiperidine (TEMP) spin trap as a singlet oxygen (¹O₂) scavenger. The ESR spectra revealed a distinctive 1:1:1 peak ratio upon US exposure, which is indicative of the generation of ¹O₂ by C₆IrAC (Fig. 1e and Figs. S2d and e in Supporting information). These findings substantiated the superior sonodynamic efficacy of C₆IrAC.

Accumulation of drugs within tumor cells is essential for cancer therapy. To evaluate the cellular uptake characteristics of C₆IrAC, C₆IrAC was incubated with 4T1 cells for various durations, and the changes in fluorescence intensity were assessed using flow cytometry and confocal laser scanning microscopy (CLSM) (Fig. S3 in Supporting information). As shown in Fig. 2a and Fig. S3a, the results indicate the concentration-dependent and time-dependent of C₆IrAC uptake by 4T1 cells. After 9 h co-incubation, the cellular uptake rate exceeded 80%. CLSM imaging results corroborated these findings, suggesting that 4T1 cells exhibit a strong time-dependent cellular uptake of C₆IrAC (Fig. S3b).

Next, the DCFH-DA probe was utilized to stain 4T1 cells to investigate the production of ROS intracellularly. Flow cytometry results revealed that C₆IrAC induced ROS generation in 4T1 cells, which was further augmented following US exposure (Fig. S3c). Under fluorescence microscopy, cells co-incubated with C₆IrAC emerged green fluorescence, which was significantly enhanced post-US stimulation (Fig. 2b). It suggests that C₆IrAC possesses not only favorable sonodynamic properties but may also exhibit chemo-sensitivity to tumor cells. Consequently, we assessed the chemical toxicity of C₆IrAC against various tumor cell lines using a 72 h cell counting kit-8 (CCK8) assay (Fig. 2c). Notably, as shown in Table S2 (Supporting information), the IC₅₀ values of C₆IrAC in multiple cancer cell lines were all < 10 µg/mL. The cytotoxicity observed in tumor cells raises concerns regarding the biosafety profile of C₆IrAC. Gratifyingly, C₆IrAC demonstrated a higher safety profile in normal cells, such as 3T3 and HK2, compared to tumor cells (Fig. S3d). It may be attributed to the fact that the antioxidant defense systems in tumor cells are less effective than in normal cells, rendering them more vulnerable to increased ROS levels [33]. It suggests that C₆IrAC not only possesses favorable sonodynamic properties but also exhibits toxicity specifically towards tumor cells.

Subsequently, we investigated the SDT activity of C₆IrAC using the CCK8 assay in vitro. As depicted in Fig. 2d, C₆IrAC significantly inhibited cell viability upon co-culture for 24 h, with even more pronounced cytotoxicity observed under US stimulation. Coumarin 6 did not exhibit significant SDT activity at the same concentration and under the same ultrasonic conditions, demonstrating that C₆IrAC possesses superior sonosensitizing performance compared to coumarin 6 (Fig. S3e). Propidium iodide (PI) staining confirmed extensive cell death in 4T1 cells incubated with C₆IrAC following US exposure (Fig. S3f). To better understand the therapeutic mechanism of action of C₆IrAC, the protein levels of cleaved caspase-3 were investigated via Western blot (WB) analysis, indicating apoptosis in 4T1 cells following treatment with C₆IrAC, with US stimulation further enhancing the level of apoptosis (Fig. 2e). To more visually demonstrate the anti-tumor effects of C₆IrAC, 4T1 cells were stained with the Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) kit post-treatment, and the apoptotic levels were assessed using flow cytometry, yielding results consistent with those from the WB (Fig. S3g). Additionally, wound-healing assays confirmed that C₆IrAC significantly inhibited tumor cell migration, with even greater inhibition observed under US stimulation (Figs. S4a and b in Supporting information). Thus, these findings demonstrated that C₆IrAC induces apoptosis in tumor cells and inhibits their metastasis, with the therapeutic efficacy of C₆IrAC being enhanced following US exposure.

DCs are considered a crucial link between innate and adaptive immunity. The differentiation and maturation of DCs can significantly augment anti-tumor immunity. Therefore, the in vitro effects of C₆IrAC on ICD were assessed by detecting the translocation of CRT and the secretion of HMGB1. In comparison to the control group cells where red fluorescence was nearly undetectable, there was a significant upregulation of CRT expression in cells treated with C₆IrAC. Concurrently, we observed the release of HMGB1 from the nucleus relative to the control group, confirming the induction of ICD in 4T1 cells by C₆IrAC (Figs. S4c and d in Supporting information). Furthermore, a higher level of ICD was detected in the C₆IrAC+US group compared to the C₆IrAC group. Inspired by its potent ability to induce ICD, we co-cultured 4T1 cells with immature DC2.4 cells to verify whether ICD induced by C₆IrAC could promote the maturation of DCs. CD80, CD86 and Ⅱ are well-recognized markers of mature DCs [34]. We detected these surface markers using flow cytometry. As shown in Fig. 2f, the expression of CD80, CD86, and MHC Ⅱ on DCs from the C₆IrAC and C₆IrAC+US groups was significantly upregulated compared to the control group, indicating maturation of the DCs. These results suggested that C₆IrAC can induce ICD in tumor cells and promote the maturation of DCs, while also acting as a SNS under US irradiation to further enhance DCs maturation. It provides evidence supporting the hypothesis that C₆IrAC promotes anti-tumor immunity.

The subcellular localization of the SNS within cells is a critical determinant in SDT, and we verified the subcellular localization of C₆IrAC by co-staining with the commercial dyes ER-Tracker Blue and Mito-Tracker Red. As shown in Fig. 2g, C₆IrAC specifically localized to ER (R = 0.86), as opposed to mitochondria (R = 0.47). The ER-localized C₆IrAC is anticipated to induce ER stress. Under US irradiation, the in situ generated ROS by C₆IrAC severely disrupt the ER, leading to a prominent oxidative stress. The phosphorylation of eukaryotic translation initiation factor 2α (elf2α) and the expression of C/EBP homology protein (CHOP) are markers of ER stress. WB analysis detected the changes of these two proteins in 4T1 cells following various treatments to further confirm this phenomenon. As expected, significant phosphorylation of elf2α and expression of CHOP were observed in both the C₆IrAC and C₆IrAC+US groups, indicating that C₆IrAC effectively initiated ER stress to eliminate tumor cells. The highest levels of elf2α phosphorylation and CHOP expression were noted in the C₆IrAC+US group, confirming that C₆IrAC-mediated SDT induced the further pronounced ER stress (Fig. 2h).

ER stress causes the massive release of Ca²⁺ from the ER, so intracellular Ca²⁺ levels were assessed by Ca²⁺ fluorescent probe (Fura-2 AM). Unlike other groups, the C₆IrAC and C₆IrAC+US groups exhibited significant blue fluorescence (Fig. 2i). The most intense fluorescence signal was observed in the C₆IrAC+US group, indicating the strongest ER stress induced by C₆IrAC post-US irradiation. The release of Ca²⁺ promotes the ER-mitochondria Ca²⁺ flux and leads to mitochondrial Ca²⁺ overload, inducing the opening of the mitochondrial permeability transition pore (mPTP), disrupting the electron transport chain, and ultimately generating a large amount of ROS, the scheme of mechanism was presented in Fig. S5a (Supporting information) [35]. Therefore, we examined the changes in mitochondrial membrane potential in 4T1 cells. The characteristic green-to-red (aggregates/monomers) images of JC-1 indicated a significant loss of membrane potential in the C₆IrAC and C₆IrAC+US groups (Fig. S5b in Supporting information). These findings elucidated that the underlying mechanism by which co-incubation with C₆IrAC leads to the generation of ROS within the cells.

To elucidate the underlining anti-tumor mechanisms of C₆IrAC, RNA-seq analysis was performed on 4T1 cells subjected to various treatments, encompassing a total of 24,355 genes (Fig. S6 in Supporting information). Fig. 3a depicts the results of a principal component analysis (PCA), where the data points from identical groups are observed to be closely clustered. This pattern of clustering is indicative of the robustness and consistency of the sequencing data. Differential expression analysis (DEA) was conducted between the PBS and C₆IrAC groups, revealing 3888 differentially expressed genes (DEGs), with 57.6% of these genes were downregulated in the C₆IrAC group (Fig. S6a). A similar analysis comparing the C₆IrAC and C₆IrAC+US groups identified 1343 DEGs, with 45.6% exhibiting downregulation in the C₆IrAC+US group (Fig. S6b). Heatmaps were utilized to provide a more intuitive representation of the gene expression differences (Figs. S6c and d). The Kyoto encyclopedia of genes and genomes (KEGG) was employed for a detailed analysis of the DEGs, with results presented in Fig. S7 (Supporting information). Subsequently, a comprehensive expression trend analysis (mfuzz) was conducted on the transcriptome data from the three groups (Fig. 3b). Based on gene ontology (GO), functional enrichment was performed to describe genes that were significantly downregulated in the C₆IrAC group and further downregulated in the C₆IrAC+US group (cluster 3), or those that were significantly upregulated in the C₆IrAC group and further upregulated in the C₆IrAC+US group (cluster 9).

Genes within cluster 3 were predominantly enriched in pathways related to hypoxia and mitochondrial function (Fig. 3c). Notably, in contrast to the traditional SDT-induced cellular hypoxia, C₆IrAC significantly inhibited the hypoxia pathway, with further inhibition observed post-US treatment. Gene set enrichment analysis (GSEA) corroborated the GO analysis results by assessing the expression of HIF-1 target genes and cellular hypoxia hallmarks (Figs. S8a and b in Supporting information). To validate these findings, we detect the protein levels of HIF-1α in 4T1 cells following different treatment modalities. The results indicated a significant reduction in HIF-1α levels in both the C₆IrAC and C₆IrAC+US groups compared to the PBS group, with the lowest levels observed in the C₆IrAC+US group (Fig. 3d). However, we did not detect catalase-like activity in C₆IrAC, suggesting that the decrease in HIF-1α protein levels was not attributable to the catalytic decomposition of H₂O₂ into O₂ by C₆IrAC (Fig. S8c in Supporting information). We hypothesize that heightened ER stress may lead to the proteasomal degradation of HIF-1α [18]. So, we used the proteasome inhibitor bortezomib (BTZ) to block the proteasomal pathway and reassessed HIF-1α levels. WB analysis confirmed that BTZ prevented HIF-1α degradation, indicating that C₆IrAC promotes the degradation of HIF-1α by activating its proteasomal pathway (Fig. 3e).

Genes in Cluster 9 were primarily enriched in pathways associated with oxidative stress and ER stress, which are integral to the anti-tumor mechanisms of C₆IrAC (Fig. 3f). Interestingly, these genes were found to be enriched in pathways involved in type Ⅰ IFN production, a direct outcome of cGAS-STING pathway activation. Ifnb1 is a canonical downstream gene of the cGAS-STING pathway. Real time quantitative PCR (RT-qPCR) and enzyme linked immunosorbent assay (ELISA) indicated that C₆IrAC enhanced Ifnb1 mRNA and IFN- β protein levels (Fig. 3g and h). However, no significant difference was observed between the C₆IrAC and C₆IrAC+US groups. Expression of inflammatory cytokines controlled by the cGAS-STING pathway was also examined, showing a similar trend (Fig. 3i). WB analysis further confirmed a significant increase in the phosphorylation of STING and the expression of cGAS in both the C₆IrAC and C₆IrAC+US groups, substantiating the activation of the cGAS-STING pathway (Fig. 3j). We co-incubated C₆IrAC with 4T1 cells, and after with/without US treatment, the culture supernatant was added to the medium of RAW264.7 cells pre-treated with IL-4. CLSM imaging showed that C₆IrAC significantly enhanced the polarization of M2-type RAW264.7 cells to M1-type, and US treatment further promoted the level of M1 polarization. This indicates that C₆IrAC can promote the polarization of macrophages from M2 to M1, thereby activating innate antitumor immunity (Fig. S8d in Supporting information).

Cytoplasmic DNA is recognized by cGAS, and upon binding to DNA, cGAS activates the cGAS-STING pathway [36]. Inspired by the disruption of mitochondrial membrane potential by C₆IrAC, qPCR analysis was conducted to determine the cytoplasmic content of mtDNA. The results demonstrated a significant increase in cytoplasmic mtDNA levels in both the C₆IrAC and C₆IrAC+US groups compared to the PBS group, with the highest levels observed in the C₆IrAC+US group (Fig. 3k). These findings suggested that ER stress induced by C₆IrAC disrupts mitochondrial membrane potential, leading to the translocation of mtDNA into the cytoplasm and subsequent activation of the cGAS-STING pathway. While the cytoplasmic mtDNA levels in the C₆IrAC+US group were significantly higher than in the C₆IrAC group, no corresponding increase in cGAS-STING pathway activation was observed. It could be due to the fact that the level of STING pathway activation is not entirely linearly correlated with the concentration of cytosolic mtDNA. Furthermore, an upregulation in the expression levels of genes associated with antigen presentation was detected, providing additional evidence for the hypothesis that C₆IrAC enhances anti-tumor immunity (Fig. S8e in Supporting information).

Therefore, we began to consider the possibility of C₆IrAC treatment for solid tumors in vivo. Considering the significant advantages of SDT in treating deep-seated tumors, we constructed the 3D tumor spheroids model to simulate solid tumors and investigated the penetration of C₆IrAC into solid tumor tissue (Fig. S9 in Supporting information). Images of the 3D tumor model were acquired using CLZM Z-stack, revealing that C₆IrAC exhibited good penetrative ability with a penetration depth exceeding 60 µm (Fig. S9a). Due to the excitation wavelength of C₆IrAC at 488 nm, which is unable to penetrate the interior of the tumor spheroid, imaging of deeper regions was not feasible. The AM/PI staining assay of the tumor spheroid confirmed the cytotoxicity of C₆IrAC towards deep tumor cells and demonstrated enhanced cytotoxicity under US stimulation (Fig. S9b).

To enhance the precision and efficiency of tumor therapy, we utilized fluorescence imaging to investigate the biodistribution of C₆IrAC and guide optimal treatment timing. All procedures involving the animals were conducted in compliance with the ethical standards and protocols established by the Xiamen University Animal Care and Use Committee (No. XMULAC20190146). C₆IrAC (5 mg/kg) was administered intravenously to 4T1 tumor-bearing mice. At 0, 4, 8, 12, 24, and 36 h post-injection, mice were euthanized at each time point for ex vivo fluorescence imaging of their major organs and tumor tissues (Fig. S9c). The results revealed a time-sensitive increase in the fluorescence intensity within the tumor area, reaching an optimal value at 24 h (Figs. S9d and e). At 36 h, a decrease in the fluorescence signal was observed in both major organs and tumor tissues, demonstrating the clearance of the drug from the body. In summary, C₆IrAC effectively penetrated and accumulated in the tumor and was metabolized and excreted after a certain period. It suggested that C₆IrAC holds exciting potential and biosafety for cancer therapy.

Next, we further evaluated potential of C₆IrAC for anti-tumor therapy in vivo. Seven-week-old BALB/c mice were randomly assigned to four experimental groups (PBS group, PBS+US group, C₆IrAC group and C₆IrAC+US group) to establish the 4T1 tumor-bearing model. The administered dose of C₆IrAC was 5 mg/kg, delivered via tail vein injection. Ex vivo fluorescence imaging results indicate that C₆IrAC exhibits the highest enrichment in the tumor region 24 h after tail vein injection. Therefore, the ultrasound wave (5 min, power density 1 W/cm², frequency 40 kHz) was performed 24 h after intravenous injection (Fig. 4a). Body weight and tumor size were monitored every 48 h. As depicted in Fig. 4b, tumor growth was significantly attenuated in the C₆IrAC and C₆IrAC+US groups compared to the PBS and PBS+US groups, with the most pronounced reduction observed in the C₆IrAC+US group. The weight of the tumors extracted on day 14 aligned with the volumetric growth patterns, demonstrating the favorable therapeutic outcome of C₆IrAC (Fig. 4c). Throughout the treatment period, there was no discernible impact on the body weight of mice treated with US and C₆IrAC, indicating no disruption to mice normal growth (Fig. S10a in Supporting information). After the treatment regimen, major organs from the mice, including the heart, liver, spleen, lungs and kidneys, were harvested and subjected to hematoxylin and eosin (H&E) staining to evaluate histological integrity. No overt organ damage was detected in the H&E-stained sections across all treatment groups (Fig. S10b in Supporting information). Hemolysis assays demonstrated that high concentrations of C₆IrAC did not induce erythrocyte lysis (Figs. S11a and b in Supporting information). Furthermore, a panel of blood markers, including alanine aminotransferase (ALT), creatine kinase MB (CK-MB), aspartate aminotransferase (AST) and creatinine (CREA), were comparable among the groups. These results substantiated that C₆IrAC, either alone or in conjunction with US, not only exerts robust anti-tumor efficacy but also maintains a favorable biosafety profile in murine models (Figs. S11c–f in Supporting information).

Inspired by the results of transcriptomic analysis, we conducted WB analysis on tumor tissues. The results revealed that HIF-1α was significantly downregulated in tumors from the C₆IrAC and C₆IrAC+US groups, with levels nearly undetectable in the C₆IrAC+US group (Fig. 4d). Concurrently, in alignment with in vitro experimental outcomes, the levels of p-STING were markedly elevated in the C₆IrAC and C₆IrAC+US groups compared to the PBS and PBS+US groups. IF staining of p-STING on tumor sections was performed. Consistent with the WB results, the STING pathway was significantly activated in the tumor cells of the C₆IrAC and C₆IrAC+US groups (Fig. 4e). Additionally, ELISA results indicated that IFN-β were significantly increased in the C₆IrAC and C₆IrAC+US groups compared to the other two groups (Fig. 4f). These results collectively demonstrate the activation of the cGAS-STING pathway in tumor cells in vivo by C₆IrAC. It has been reported that the activation of the cGAS-STING pathway can polarize M2-type macrophages (F4/80⁺ CD206⁺) towards M1-type macrophages (F4/80⁺ CD86⁺). Therefore, we analyzed the infiltrating macrophages in the tumor tissue by flow cytometry. The results indicated a higher M1/M2 ratio in the C₆IrAC and C₆IrAC+US groups, with the C₆IrAC+US group showing the highest ratio (Fig. 4g and Fig. S12 in Supporting information). These findings suggest that C₆IrAC promotes the polarization of infiltrating macrophages towards the M1 phenotype. However, the activation of the cGAS-STING pathway induced by C₆IrAC may not be the sole mechanism involved.

Subsequently, the main immune organs (lymph nodes and spleen) of the mice were collected and the in vivo immune activation potential of C₆IrAC was monitored by flow cytometry. The results demonstrated that a significant increase in the ratios of CD3⁺ CD8⁺ T cells, CD3⁺ CD4⁺ T cells and CD11c⁺ CD86⁺ DCs in the lymph nodes of the C₆IrAC and C₆IrAC+US groups, with the highest ratios observed in the C₆IrAC+US group (Figs. 4h–j and Fig. S13 in Supporting information). Concurrently, the trends in the spleen were consistent with those in the lymph nodes, indicating that C₆IrAC and its mediated SDT can enhance the anti-tumor immune response (Fig. S14 in Supporting information).

Beyond immune cells, cytokines such as tumor necrosis factor-alpha (TNF-α) and IFN-γ also play a significant role in tumor immunomodulation. Therefore, measuring the levels of TNF-α and IFN-γ secretion allows us to evaluate the immune response mediated by C₆IrAC (Fig. S15 in Supporting information). Serum levels of IFN-γ and TNF-α in mice from the C₆IrAC and C₆IrAC+US groups are significantly higher than those in the other two groups. The highest levels of cytokines were detected in the C₆IrAC+US group, suggesting that C₆IrAC could induce a robust cytokine-mediated immune response (Figs. S15a and b).

To further elucidate the anti-tumor mechanisms of C₆IrAC in vivo, we performed Ki67 and terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) staining on tumor sections. In Fig. S15c, the Ki67 staining results indicated the significant suppression of tumor proliferation in the C₆IrAC and C₆IrAC+US groups. Furthermore, the C₆IrAC+US group exhibited the most robust TUNEL fluorescence signal in tumor cells, indicating the most significant level of cell death. TUNEL signals were also detected in the C₆IrAC group, but no distinct signals were observed in the other two groups (Fig. S15d). Moreover, immunofluorescence analysis of tumor tissue sections was similarly employed to assess the immune response within the tumor microenvironment. CRT was found to be markedly expressed in the C₆IrAC and C₆IrAC+US groups, and the signal for HMGB1 was significantly diminished in these groups, indicating the enhanced ICD process (Fig. S15c). Compared to the PBS and PBS+US groups, the tumor tissues in the C₆IrAC and C₆IrAC+US groups displayed increased infiltration of CD4⁺ T cells and CD8⁺ T cells, with the highest infiltration observed in the C₆IrAC+US group (Fig. S15d). This observation suggests that a T-cell-mediated immune response was effectively triggered. These results suggested that C₆IrAC and its mediated SDT not only exhibit direct tumor-suppressive effects but also trigger a robust anti-tumor immune response through various pathways (ICD, activation of the STING pathway, etc.).

In this work, we synthesized a novel organometallic SNS (C₆IrAC), which specifically targets the ER. Besides, C₆IrAC exhibits tumor cell-specific cytotoxicity and excellent sonosensitivity. C₆IrAC induces ER stress in tumor cells, which is further amplified by US-mediated SDT through the in situ generation of a substantial amount of ROS. The strong oxidative stress induced significant apoptosis and prominent ICD. In-depth investigation into the underlying mechanisms of C₆IrAC revealed that it induces HIF-1α degradation, activates the cGAS-STING pathway, secretes IFN-β, and upregulates the antigen presentation pathway. These multifaceted effects not only suppress the proliferation and migration of tumor cells but also enhance both innate and adaptive immunity. Furthermore, we evaluated the therapeutic efficacy of C₆IrAC in tumor models, demonstrating that it can induce activation of anti-tumor immune activation and reshape the tumor immune microenvironment. In summary, our research elucidates the excellent biomedical applications of C₆IrAC and showcases the prospects of ER-targeted SDT. Additionally, we deeply explored the therapeutic mechanisms of ER-targeted SDT through transcriptomics, clarifying the biological properties of this organometallic SNS molecule. Therefore, we believe that our current research provides an innovative strategy for the design and development of SNS, with significant clinical translational potential.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Yubo Tan: Writing – original draft, Project administration, Methodology, Investigation, Conceptualization. Ziying Wang: Software, Resources, Project administration, Formal analysis. Yibo An: Resources, Project administration, Investigation. Man Li: Methodology, Investigation, Formal analysis. Dazhuang Xu: Validation, Supervision, Software. Renyuan Liu: Methodology, Investigation, Data curation. Xinyu Tan: Software, Resources, Investigation. Yaohui He: Writing – original draft, Software, Funding acquisition, Data curation. Zhixiang Lu: Writing – review & editing, Project administration, Funding acquisition. Gang Liu: Writing – review & editing, Project administration, Funding acquisition, Conceptualization.

Acknowledgments

This work was supported by the National Key Research and Development Program of China (Nos. 2023YFB3810000, 2023YFC241570), the National Natural Science Foundation of China (NSFC, Nos. 32101113, 81925019, 82203476, U22A20333), the Natural Science Foundation of Sichuan Province (No. 2024NSFSC1741) and the China Postdoctoral Science Foundation (No. 2022M712661).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111214.