Mitochondria‐Targeting Iridium(III) Complexes Induce PANoptosis and Ferroptosis for Boosting Chemoimmunotherapy Against Immune‐Desert Colorectal Cancer

ABSTRACT

Over 80% of colorectal cancer (CRC) cases are classified as microsatellite stable (MSS), a subtype characterized by an immunosuppressive tumor microenvironment that poses a significant challenge for chemotherapy. Emerging evidence suggests that inflammatory programmed cell death pathways can effectively activate anti‐cancer immunity across various cancer types. However, studies exploring the role of these death pathways in reversing the immune desertification and providing unique therapeutic advantages in MSS‐CRC remain scarce. Herein, we present a strategy using iridium(III)‐based inducers to concurrently trigger immunogenic PANoptosis and ferroptosis, aiming to reverse the “immune desert” phenotype of MSS‐CRC and integrate chemotherapy with immunotherapy. Upon mitochondrial accumulation, these iridium(III) compounds inhibit complex I of the electron transport chain, leading to electron leakage and excessive reactive oxygen species generation, which collectively initiate the PANoptotic and ferroptotic signaling pathways in MSS‐CRC. Proteomic analysis and in vivo experiments further demonstrated that Ir2 can activate PANoptotic and ferroptotic pathways to counteract immune desertification in MSS‐CRC. Importantly, this study establishes a mechanism‐guided chemical design strategy whereby subtle ligand engineering within a cyclometalated Ir(III) platform governs mitochondrial accumulation, redox disruption, and subsequent induction of immunogenic cell death.

Mitochondria‐localized iridium(III) complexes disrupt the electron transport chain to produce mitochondrial oxidative stress, triggering PANoptosis and Ferroptosis to boost chemoimmunotherapy against immune‐desert colorectal cancer.

1. Introduction

Colorectal cancer (CRC) is a major global health burden, ranking among the top three most common malignancies and the fourth leading cause of cancer‐related mortality [1]. The Cancer Genome Atlas (TCGA) categorizes CRC into three types: microsatellite stable (MSS), microsatellite instability‐high (MSI‐H), and microsatellite instability‐low (MSI‐L) [2, 3]. Notably, MSS‐CRC represents approximately 80% of metastatic CRC cases and is commonly associated with a low tumor mutational burden and an “immune desert” phenotype, which is characterized by scarce immune cell infiltration in the tumor microenvironment [4, 5]. Conventional platinum‐based chemotherapy, exemplified by oxaliplatin (OXA), frequently fails to overcome the “immune desert” phenotype or elicit durable antitumor immune responses in MSS‐CRC, due to chemoresistance and systemic toxicities [6, 7]. While intensified strategies such as FOLFOXIRI (triplet therapy) or the combination of chemotherapeutic drugs with immune adjuvants aim to enhance immunogenicity, clinical efficacy remains limited and short‐lived, often failing to sustain T‐cell activation in an immunosuppressive tumor microenvironment [8, 9]. Consequently, there is an urgent need to develop innovative chemotherapeutic agents to reprogram the tumor microenvironment and overcome the inherent resistance of MSS‐CRC.

Emerging evidence underscores the pivotal role of immunogenic inflammatory cell death pathways, particularly PANoptosis and ferroptosis, for remodeling the tumor immune microenvironment and enhancing antitumor immunity [10, 11, 12, 13]. Ferroptosis is an iron‐dependent form of non‐apoptotic cell death, characterized by excessive lipid peroxidation, oxidative stress, mitochondrial damage, and eventual cell membrane rupture [14, 15, 16, 17]. Most recently, PANoptosis is a genetically regulated inflammatory cell death pathway orchestrated by the PANoptosome, integrating key features of pyroptosis, apoptosis, and necroptosis through the activation of caspases and receptor‐interacting protein kinases (RIPKs) [18, 19]. Notably, both PANoptosis and ferroptosis can induce immunogenic cell death (ICD), thereby promoting the release of tumor‐associated antigens (TAAs) and damage‐associated molecular patterns (DAMPs) to facilitate immune cell recruitment and activation. These features make PANoptosis an attractive strategy for reversing the “immune desert” in MSS‐CRC. However, the therapeutic significance of PANoptosis remains highly context‐dependent and cannot be defined solely by the activation of death‐pathway markers. Depending on tumor type, immune context, and the surrounding cytokine milieu, PANoptosis‐associated inflammation may either support antitumor immunity or instead reinforce potentially tumor‐supportive inflammation [11, 20]. Taken together, these features highlight targeted PANoptosis induction as a promising approach to reverse the immune‐desert phenotype of MSS‐CRC by promoting functionally effective antitumor immunity.

Mitochondria play a central role in PANoptosis and ferroptosis by regulating cellular metabolism, redox homeostasis, and cell death signaling [21]. Disruption of mitochondrial glycolysis or oxidative phosphorylation leads to severe oxidative stress and lipid peroxidation, triggering PANoptotic and ferroptotic pathways and forming an immunogenic tumor microenvironment [18]. These mechanisms are particularly relevant to MSS‐CRC, which is characterized by an immunologically “immune‐desert” phenotype and poor responsiveness to current immunotherapies. In recent years, various mitochondria‐targeting therapeutic complexes, including polypyridine‐, triphenylphosphine‐, and cyclometalated iridium(III)‐based complexes, have been developed [22, 23, 24, 25, 26, 27, 28, 29, 30, 31]. Among these, cyclometalated Ir(III) complexes have attracted significant attention due to their high physiological stability, suitable biocompatibility, structural versatility, and favourable pharmacokinetic properties [28, 31, 32, 33, 34, 35, 36, 37, 38]. Notably, recent studies have highlighted the immunotherapeutic potential of metallodrugs, particularly their ability to induce ICD‐associated danger signaling and promote anticancer immune activation, thereby providing an important foundation for this work [39, 40, 41, 42]. Consequently, the development of mitochondrial‐targeted and electron transport chain‐disrupted Ir(III) complexes, which simultaneously induce PANoptosis and ferroptosis, represents a promising approach to convert MSS‐CRC from an immune‐desert into an immune‐responsive tumor, thereby achieving effective cancer immunotherapy.

In this study, we have rationally developed a series of simple Ir(III) complexes with distinct ligands to induce PANoptosis/ferroptosis in MSS‐CRC. Building upon our previously developed cyclometalated Ir(III) ligand framework [35, 43, 44], distinct aromatic extensions (phenyl, naphthyl, and biphenyl groups) were introduced to systematically investigate how subtle ligand engineering affects physicochemical properties, mitochondrial accumulation, and the resulting cell death phenotypes. Notably, among these complexes, Ir2 bearing the naphthyl moiety exhibited the most pronounced effects, demonstrating the highest cytotoxicity, the strongest inhibition of complex I (NADH‐ubiquinone oxidoreductase) in the mitochondrial electron transport chain, and the most robust ROS generation (Scheme 1). Subsequently, the amplified oxidative stress efficiently initiates both ferroptotic and PANoptotic signaling in MSS‐CRC cells. Proteomic analysis reveals that Ir2 robustly activates key signaling pathways tightly associated with PANoptosis and ferroptosis. In vivo studies demonstrate that the Ir(III) complex promotes the release of DAMPs from dying tumor cells, increases tumor immunogenicity, and reprograms the immunosuppressive tumor microenvironment, thereby reversing immune desertification and improving anti‐tumor immunotherapeutic efficacy. Therefore, this work not only identifies the naphthyl‐functionalized Ir2 as the optimal candidate but also establishes a clear structure–property–function relationship linking this ligand modification to enhanced mitochondrial targeting, respiratory‐chain disruption, and ROS‐amplified PANoptosis/ferroptosis in MSS‐CRC.

SCHEME 1.

The mechanisms of cyclometalated iridium(III) complexes selectively target mitochondria, disrupt the electron transport chain, induce PANoptosis and ferroptosis, and reverse the immune‐desert phenotype of CRC.

2. Results and Discussion

The cyclometalated iridium(III) complexes Ir1‐Ir3 were synthesized following a modified literature protocol (Figure 1a) [45, 46, 47]. A comprehensive characterization of Ir1–Ir3 was achieved through mass spectrometry (ESI‐MS) and proton nuclear magnetic resonance (¹H NMR) spectroscopy, which provided essential structural validation (Figures S1–S6). The ultraviolet‐visible (UV‐vis) absorption spectra of these complexes exhibited strong absorption in the visible range, specifically between 350 and 430 nm, with a prominent maximum absorption wavelength centered around 400 nm (Figure S7a). Ir1–Ir3 exhibited an intense phosphorescence in the red region upon excitation at 405 nm (Figure S7b). High‐performance liquid chromatography (HPLC) analysis confirmed the high purity of the complexes, with a purity exceeding 98% (Figure S8a and Table S1). Stability assessments in DMEM containing 10% fetal bovine serum revealed no significant decomposition after 48 h of incubation (Figure S8b).

FIGURE 1.

(a) Chemical structures of the Ir(III) complexes. (b) Mitochondrial co‐localization of Ir1‐‐Ir3 in SW480 cells for 4 h by confocal laser scanning microscopy (CLSM). Scale bars: 10 µm. (c) The activity of Complex I (NADH dehydrogenase) in SW480 cells incubated with Ir1‐‐Ir3 and rotenone (ROT) for 48 h (ROT is the positive control group). n = 3, **p < 0.01. (d) CLSM of ROS generation in SW480 cells incubated with Ir1–Ir3 and oxaliplatin (OXA) at their IC₅₀ values for 4 h. Scale bars: 10 µm. (e) Fluorescence microscopy images of SW480 cells incubated with the mitochondrial membrane potential‐specific probe JC‐1 and treated with Ir1‐‐Ir3 and OXA with IC₅₀ for 4 h. Scale bars: 20 µm.

The logP_o/w value, representing the oil/water partition coefficient of metal complexes, plays a crucial role in determining their cellular uptake and subcellular localization. Previous studies showed that subtle ligand modification can strongly affect lipophilicity, subcellular targeting, and biological activity [48, 49, 50, 51]. Based on this rationale, we retained the same cyclometalated Ir(III) scaffold and subtly modified the pendant aryl substituent on the potip‐derived ligand. Specifically, phenyl, naphthyl, and biphenyl substituents were introduced to afford Ir1‐Ir3, respectively. The octanol‐water partition coefficients (logP_o/w) of Ir1‐Ir3 were measured using a UV‐vis spectrophotometer, showing that Ir2 had the highest lipophilicity with a logP_o/w value of 0.9 (Figure S9). This logP_o/w value is higher than those of Ir1 (0.7) and Ir3 (0.5) and is closer to the lipophilicity range considered favorable for mitochondrial accumulation, according to Horobin's studies [50, 51].

We next examined cellular uptake and subcellular distribution by CLSM and inductively coupled plasma mass spectrometry (ICP‐MS). The findings indicated that all complexes reached maximal cellular accumulation at 4 h, with Ir2 exhibiting the highest cellular uptake level at 55 ng per 10⁴ cells (Figure S10a,b). Based on this, a 4‐h incubation period was selected to analyze the organelle‐targeting behavior of the Ir(III) complexes. CLSM imaging and ICP‐MS analysis indicated that complexes were predominantly colocalized in mitochondria, with Ir2 showing the highest mitochondrial accumulation, up to 79% (Figure 1b and Figure S10c). Further time‐dependent colocalization analysis showed that the free iridium complexes displayed partial lysosomal localization at early incubation time points, followed by gradual mitochondrial accumulation upon prolonged incubation, eventually exhibiting predominant mitochondrial colocalization (Figure S11). These results validate that even subtle ligand modification can modulate the lipophilicity of metal complexes, thereby influencing their cellular uptake and mitochondrial targeting efficiency.

The cytotoxic effects of Ir(III) complexes were assessed on three microsatellite‐stable CRC (MSS‐CRC) cell lines (SW620, SW480, and CT26), two microsatellite instability‐high CRC (MSI‐H‐CRC) cell lines (HCT‐116 and Caco‐2), and the regular colorectal epithelial cell line (MCN460) (Table S2). Ir1–Ir3 exhibited potent cytotoxicity against MSS‐CRC cells. The relatively lower IC₅₀ values in cancer cells than in normal MCN460 cells may be attributed to their higher mitochondrial membrane potential, elevated oxidative stress, and metabolic reprogramming of tumor cells, which increase susceptibility to Ir2 [52, 53]. Notably, Ir2 demonstrated significant efficacy against the SW480 cell line, with an IC₅₀ value of 0.69 µM. Accordingly, SW480 cells were selected as the model system for subsequent experiments.

We then explored whether the difference in mitochondrial accumulation translated into distinct mitochondrial damage. Indeed, examination of the electron transport chain revealed that Ir2, compared to the control (Ctrl) and other complexes, significantly inhibited NADH dehydrogenase (Complex I) activity (Figure 1c). This inhibition triggered disruption of electron flow, substantial ROS generation, and ultimately, mitochondrial membrane potential collapse. To further investigate the ROS formation, the SW480 cells were co‐incubated with the ROS‐specific fluorescent dye 2,7‐dichlorodihydrofluorescein diacetate (DCFH‐DA) (Figure 1d and Figure S12a,b). The intense green fluorescence of the ROS‐specific probe was monitored upon treatment with Ir1‐Ir3, and Ir2 exhibited the strongest green fluorescence. To directly assess mitochondrial ROS production, SW480 cells were stained with MitoSOX Red after treatment with Ir1–Ir3. Compared with others, Ir2‐treated cells exhibited significantly enhanced MitoSOX fluorescence, confirming the accumulation of mitochondrial ROS (Figure S12c). Excessive ROS generation within mitochondria can induce mitochondrial dysfunction, triggering a cascade of cellular damage, including a drop in mitochondrial membrane potential and inhibition of the respiratory chain. Subsequently, mitochondrial membrane potential was monitored by CLSM using the specific dye JC‐1. As shown in Figure 1e, SW480 cells treated with Ir1, Ir2, and Ir3 exhibited increased green fluorescence and decreased red fluorescence. Notably, Ir2 induced a significantly stronger green fluorescence than Ir1 and Ir3. Collectively, Ir2 induced the most pronounced collapse of mitochondrial membrane potential. Taken together, these results establish a clear structure‐property‐activity correlation within this Ir(III) series: subtle modification of the pendant aryl substituent on the potip‐derived ligand tunes lipophilicity, thereby affecting cellular uptake, mitochondrial accumulation, and mitochondrial damage. Among the three complexes, Ir2 showed the superior mitochondrial accumulation and more pronounced mitochondrial damage.

To assess the impact of mitochondrial damage, particularly ROS generation, on cell death modes, cytotoxicity assays were performed using Ir1‐Ir3 in combination with specific cell death inhibitors. SW480 cells were pretreated for 1 h with the apoptosis inhibitor Z‐VAD‐fmk, the autophagy inhibitor 3‐methyladenine (3‐MA), the necroptosis inhibitor necrostatin‐1 (Nec‐1), the lysosomal protease inhibitor leupeptin, the pyroptosis inhibitor disulfiram, the ferroptosis inhibitor ferrostatin‐1 (Fer‐1), or the iron chelator deferoxamine (DFO), and were subsequently incubated with Ir1–Ir3 for an additional 48 h. In addition, two PANoptosis inhibitor combinations, consisting of Z‐VAD‐fmk, Nec‐1, and disulfiram at low or high doses, were also tested. A stronger rescue effect than that produced by concentration‐matched combination controls further ruled out a merely dose‐dependent nonspecific protection and supported the concurrent involvement of multiple cell death pathways. As shown in Figure 2a, although pretreatment with individual inhibitors (Z‐VAD‐fmk, Nec‐1, disulfiram, or Fer‐1) only modestly enhanced cell viability, the PANoptosis inhibitor cocktail (Z‐VAD‐fmk + Nec‐1 + disulfiram) conferred a substantially greater protective effect. Transmission electron microscopy (TEM) and CLSM were used to analyze the changes in cell morphology caused by the Ir(III) complexes. TEM provided a time‐dependent delineation of ultrastructural alteration induced by Ir2, with mitochondrial swelling and vacuolization evident as early as 30 min post‐treatment. After 1 h, bleb‐like protrusions began to appear at the edge of the cell membrane. At 6‐h post‐treatment with Ir2, the mitochondria underwent a transition from swelling to shrinkage, accompanied by increased membrane density and markedly diminished or absent cristae, exhibiting a characteristic ferroptotic morphology. Prominent pyroptotic bodies, along with the release of cytoplasmic content, became evident at 12 h post‐treatment with Ir2. At 24‐h post‐treatment with Ir2, there was not only a significant rise in apoptotic bodies but also concomitant loss of plasma membrane integrity, a hallmark of necroptotic cell death (Figure 2b). Similar morphological progression of cell death was observed for Ir1 and Ir3 (Figure S13). Furthermore, following 24‐h co‐incubation with Ir1–Ir3, CLSM images revealed marked cellular swelling, membrane blebbing, and rupture in SW480 cells, collectively presenting morphological hallmarks strongly suggestive of PANoptosis (Figure S14). Subsequently, western blotting analysis was conducted to verify changes in characteristic protein expression associated with these cell death mechanisms. As shown in Figure 2d and Figure S15a, PANoptosis upstream regulator PANoptosome‐associated proteins (ZBP‐1, ASC, cleaved caspase 8, p‐MLKL, and p‐RIPK3) were significantly upregulated in Ir2‐treated SW480 cells. Simultaneously, the downstream execution markers corresponding to the key apoptotic factors (cleaved caspase‐3), key necroptotic molecules (p‐RIPK3 and p‐MLKL), and key pyroptotic molecules (N‐GSDMD and cleaved IL‐1β) were upregulated in Ir2‐treated cells at 24 h (Figure 2e, Figure S15a,b). Furthermore, downregulation of GPX4 and SLC7A11 was associated with ferroptosis induction (Figure 2f and Figure S15c). CLSM imaging with the lipid peroxidation (LPO) probe BODIPY 581/591 C11 revealed significantly increased LPO in SW480 cells incubated with Ir1‐Ir3 (Figure 2c). In conclusion, the Ir(III) complexes can induce PANoptosis and ferroptosis in MSS‐CRC, with Ir2 exhibiting a superior practical therapeutic effect. Therefore, Ir2 was selected for further in‐depth research.

FIGURE 2.

Cell death mechanism of Ir(III) complexes in SW480 cells. (a) Cell viability of SW480 cells after preincubation with cell death inhibitors, including z‐VAD‐FMK (20 µM, apoptosis inhibitor), 3‐MA (100 µM, autophagy inhibitor), Nec‐1 (60 µM, necroptosis inhibitor), leupeptin (100 µM, lysosomal protease inhibitor), disulfiram (4 µM, pyroptosis inhibitor), Fer‐1 (50 µM, ferroptosis inhibitor), and DFO (100 µM, iron‐dependent ferroptosis inhibitor). Combined inhibitors I (low‐dose PANoptosis inhibitor cocktail) consisted of z‐VAD‐FMK (10 µM), Necrostatin‐1 (30 µM), and disulfiram (2 µM), whereas Combined inhibitors II (high‐dose PANoptosis inhibitor cocktail) consisted of z‐VAD‐FMK (20 µM), Necrostatin‐1 (60 µM), and disulfiram (4 µM), and incubation with Ir1–Ir3 with 2*IC₅₀ concentration, error bars = SD (n = 3), **p < 0.01, ***p < 0.001. (b) Representative TEM images in SW480 cells treated with Ir2 for different time points (0, 0.5, 1, 6, 12, 24 h) with IC₅₀ concentration. c) CLSM of BODIPY 581/591 C11 stained SW480 cells after Ir1–Ir3 and OXA treatments. Scale bars: 10 µm. (d–f) WB analysis of the (d) ZBP‐1, ASC, Pro caspase‐8, Cleaved caspase‐8, MLKL, p‐MLKL, RIPK3, p‐RIPK3, (e)GSDMD, GSDMD‐N, Pro IL‐1β, Cleaved IL‐1β, Pro caspase‐3, Cleaved caspase‐3, and (f) SLC7A11 and GPX4 expression levels of SW480 cells upon treatment with Ir1–Ir3 and OXA for 24 h.

To demonstrate that PANoptosis and ferroptosis can reverse the immunodepression in MSS‐CRC, we assessed the release of DAMPs. CLSM showed that Ir2 treatment (1 and 2 µM, 24 h) effectively promoted the nuclear‐to‐cytoplasmic translocation of high mobility group box 1 (HMGB1) protein and significantly increased the surface exposure of calreticulin (CRT) on SW480 cells. In contrast, OXA failed to elicit a comparable effect (Figure 3a, Figures S16a and S17a). The flow cytometry results also indicated that Ir2 treatment induced higher levels of HMGB1 and CRT than OXA treatment (Figures S16b,c and S17b,c). Bioluminescence assays further revealed that Ir2 robustly stimulated adenosine triphosphate (ATP) secretion into the extracellular milieu (Figure S18). Collectively, these findings convincingly demonstrate that Ir2 is substantially more potent than OXA in triggering the release of all three hallmark DAMPs of ICD. Emerging evidence suggests that the inflammatory cell death mode can facilitate the delivery of specific cargo that promotes macrophage M1 polarization, thereby orchestrating innate immune responses [20]. To further investigate this immunomodulatory effect, we employed a transwell co‐culture system of SW480 cells and RAW 264.7 macrophages to evaluate the secretion of inflammatory cytokines (Figure 3b). Macrophages (in the upper chamber) cultured in proximity to Ir2‐treated SW480 cells exhibited marked morphological changes, characterized by enhanced spreading and pseudopodia formation (Figure 3c). As depicted in Figure 3d–g, enzyme‐linked immunosorbent assay (ELISA) analysis further demonstrated that Ir2‐treated SW480 cells potently stimulated the secretion of key pro‐inflammatory cytokines associated with M1 polarization, that is, tumor necrosis factor‐α (TNF‐α), interferon‐β (IFN‐β), interleukin‐6 (IL‐6), and interleukin‐1β (IL‐1β), in RAW 264.7 macrophages. Notably, the Ir2‐treated SW480 cells cultured alone in the lower chamber showed no significant alterations in inflammatory cytokine secretion compared to the control groups. (Figure S19). Furthermore, flow cytometry analysis revealed that the supernatant from Ir2‐pretreated SW480 cells greatly upregulated CD86 expression on RAW 264.7 cells (Figure 3h), validating M1 polarization driven by secreted inflammatory cytokines. Collectively, these findings suggest that the synergistic induction of PANoptosis and ferroptosis in tumor cells may reactivate macrophages, thereby restoring their immunomodulatory functions and enhancing both innate and adaptive antitumor immune responses in the immune desert MSS‐CRC.

FIGURE 3.

(a) Fluorescence images of SW480 cells treated with Ir2 and OXA (11 µM) for 24 h and stained with the CRT and HMGB1‐specific antibodies. (b) Schematic illustration of the transwell co‐culture system experiment. (c) Representative CLSM images of RAW 264.7 macrophages stimulated with conditioned media from the PBS‐, I‐, II‐, OXA‐, and lipopolysaccharide (LPS)‐treated SW480 cells. (d–g) Quantification of the TNF‐α, IFN‐β, IL‐6, and IL‐1β secretion levels in upper‐chamber supernatants upon PBS, I, II, OXA, and LPS treatments. (h) Flow cytometry analyses the CD86 expression as a marker of M1 macrophage polarization. (i) GO categorization of down‐regulated biological processes for assembled unigenes of the transcriptome upon Ir2 (2 µM) treatment for 24 h. (j) KEGG pathway enrichment analysis of up‐regulated DEGs identified after 24 h treatment with Ir2 (2 µM). (k) Heat map of proteomic profiling among different groups. Ir2 (I: 1 µM; II: 2 µM). Scale bars: 10 µm.

To further investigate whether Ir2 induces PANoptosis and ferroptosis in MSS‐CRC tumor cells and elicits an immune response, we employed 4D label‐free quantitative proteomic analysis (4D Fast DIA) on SW480 cells treated with Ir2 (1 and 2 µM) for 24 h. A series of quality control (QC) assessments was performed on the raw mass spectrometry data, including peptide length distribution (predominantly ranging from 7 to 20 amino acids), peptide count distribution (with most proteins identified by more than two peptides), and protein coverage distribution (with most proteins having a coverage below 30%) (Figure S20). These QC results validated data reliability, rendering them suitable for subsequent analyses. Scatter plots illustrated changes in protein abundance in the Ir2‐treated groups (1 and 2 µM) compared with the control group. Specifically, the Ir2 (1 µM) treatment group exhibited 803 differentially expressed proteins (DEPs) (|fold change| > 2 compared to the control group), with 292 upregulated and 511 downregulated (Figure S21a). Following Ir2 (2 µM) treatment, 1352 DEPs were identified, comprising 468 upregulated proteins and 684 downregulated proteins, indicating a clear dose‐dependent influence of cellular protein networks (Figure S22a).

To further elucidate the functional mechanisms underlying these proteomic alterations, gene ontology (GO) enrichment analysis was performed on all DEPs. The core GO functional entries most likely influenced by Ir2 treatment (1 and 2 µM) within the biological process (BP) category are presented: mitochondrial translational termination, NADH dehydrogenase complex assembly, mitochondrial respiratory chain complex assembly, ATP synthesis coupled electron transport, oxidative phosphorylation, and regulation of acute inflammatory response (Figure 3i, Figures S21b,c and S22b). These proteomic findings indicate that Ir2‐induced significant alterations occurred predominantly in cellular responses related to mitochondrial energy metabolism and the regulation of acute inflammatory responses.

To comprehensively understand the molecular reprogramming induced by Ir2, we performed gene set enrichment analysis (GSEA) and analyzed the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. GSEA revealed a significant downregulation of mitochondrial function pathways, particularly associated with NADH dehydrogenase complex assembly and ATP synthesis coupled electron transport, following treatment with Ir2 (1 or 2 µM) (Figures S21f,g and S22d,e). This finding is further supported by the previously observed suppression in NADH dehydrogenase activity (Figure 1c), confirming a functional impairment of mitochondrial oxidative phosphorylation at the protein level. Concurrently, genes involved in inflammatory responses were consistently upregulated (Figures S21h and S22f). KEGG analysis further supported the disruption of oxidative phosphorylation, with the overall pathway and its key component ND2 showing reduced activity (Figures S21d, S22c, S23, and S24). This profound metabolic collapse induced a robust host defense response, which was driven by the upregulation of immune‐related proteins (e.g., OAS family members) and systemically reflected in the significant activation of acute inflammatory pathways. (Figure 3j, Figures S25 and S26). Heatmap analysis (Figure 3k) further revealed a significant divergence in gene expression patterns. Specifically, genes enriched in pathways associated with PANoptosis, ferroptosis, and acute inflammatory responses were markedly upregulated. These results demonstrated that Ir2 directly targets mitochondria, thereby disrupting cellular energy metabolism, leading to impaired ATP synthesis and a subsequent burst of ROS. These synergistic effects, including metabolic collapse, cell death receptor‐mediated PANoptosis and ferroptosis, and pro‐inflammatory/pro‐coagulatory responses, ultimately reverse the “immune‐desert” of MSS‐CRC.

Mechanistically, whereas previous reports primarily associated metallodrug‐induced PANoptosis with ROS elevation [17], the present study advances a more clearly defined structure–property–activity framework by identifying mitochondrial complex I as a key mechanistic entry point. Subtle ligand engineering endows Ir2 with favorable electronic and lipophilic characteristics, facilitating mitochondrial accumulation and disruption with respiratory complex I. The resulting electron leakage and excessive ROS production drive mitochondrial redox collapse, ultimately triggering both PANoptosis and ferroptosis. Notably, this mechanism‐guided strategy was validated in MSS‐CRC, an immune‐desert and clinically refractory subtype, underscoring its translational relevance beyond conventional models.

To enhance the solubility and bioavailability for intravenous administration, Ir2 was encapsulated within DSPE‐PEG₂₀₀₀ micelles, forming Ir2@PEG NPs. TEM and dynamic light scattering (DLS) analyses revealed that Ir2@PEG NPs exhibited aspherical morphology with excellent hydration characteristics, displaying an average hydrodynamic diameter of approximately 108 nm (PDI = 0.27) (Figure S27a,b). Elemental mapping confirmed the homogeneous distribution of iridium, oxygen, and nitrogen throughout these nanoparticles (Figure S27c). The zeta potential significantly shifted from +37.1 ± 2.7 mV for free Ir2 to −20.9 ± 1.6 mV for Ir2@PEG NPs (Figure S28a), indicating surface charge modification upon micelle formation. The Ir2@PEG NPs demonstrated high colloidal stability, maintaining their hydrodynamic diameter over 7 days under physiological conditions (Figure S28b).

To investigate the intracellular localization of Ir2@PEG NPs in SW480 cells, ICP‐MS suggested that Ir2@PEG NPs initially localized predominantly within lysosomes at 1‐h post‐incubation (Figure S29a). As incubation time increased, Ir2@PEG NPs gradually accumulated in the mitochondria (Figure S29b,c). These findings suggest that Ir2 encapsulation by DSPE‐PEG₂₀₀₀ preserves the intrinsic intracellular targeting behavior. Hemolysis assays, essential for evaluating the biocompatibility and safety of drug candidates, demonstrated that the direct exposure of red blood cells (RBCs) to free Ir2 resulted in significant hemolysis, highlighting the considerable adverse effects associated with the intravenous administration of unencapsulated Ir2. In sharp contrast, Ir2@PEG NPs exhibited minimal hemolysis (< 5%) even at concentrations exceeding 100 µg/mL, indicating that the DSPE‐PEG₂₀₀₀ formulation effectively reduces the hemolytic potential of Ir2 (Figure S30).

Building on these promising in vitro findings, we next evaluated the in vivo therapeutic potential of Ir2@PEG NPs in CT26 tumor‐bearing mice. CT26 was chosen because it is widely regarded as an MSS murine CRC model with a poorly immunogenic, immune‐refractory phenotype, making it suitable for evaluating therapeutic strategies against MSS‐CRC [54, 55, 56]. In contrast, MC38 is generally recognized as a more immunogenic, MSI‐like model and was therefore less suitable for this study [57]. Initial safety evaluations in healthy BALB/c mice following intravenous administration showed no signs of pain, distress, or discomfort. Comprehensive biochemical analyses of blood markers performed on Day 7, including alkaline phosphatase (ALP), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and creatinine, remained within normal physiological ranges, confirming the excellent biocompatibility of Ir2@PEG NPs (Figure S31 and Table S3). Pharmacokinetic profiling using ICP‐MS revealed a prolonged blood circulation half‐life of 10.07 h (Figure S32a), accompanied by nearly complete fecal excretion within 96 h (Figure S32b), indicating favorable clearance. Crucially, biodistribution studies in CT26 tumor‐bearing mice confirmed marked tumor accumulation of Ir2@PEG NPs, reaching a maximum at 24 h post‐intravenous injection (Figure S32c), thereby validating their promising tumor‐targeting efficacy.

To investigate the anti‐tumor efficacy of our novel treatment, BALB/c mice bearing primary CT26 tumors (∼50 mm³) were intravenously injected with PBS, OXA, or Ir2@PEG NPs at 2‐day intervals for a total of 2 administrations. Subsequently, CT26 cell suspension was injected subcutaneously into the contralateral flank to establish distal tumors (Figure 4a). Both primary and distal tumor volumes, as well as body weight, were monitored every 2 days for 14 consecutive days. The results demonstrated significant suppression of primary tumor growth (Figures S33a and S34), with distal tumor progression also significantly reduced (Figure 4b and Figure S33b). Notably, in contrast to OXA, which caused a marked reduction in mouse body weight, Ir2@PEG NP treatment did not significantly alter body weight throughout the treatment period (Figure S33c). Histopathological analysis of major organs, including the heart, liver, spleen, lungs, kidneys, and intestines, via hematoxylin and eosin (H&E) staining revealed no observable tissue damage or structural abnormalities (Figure S35), indicating a favorable safety profile for Ir2@PEG NPs. H&E staining of distal tumor tissues revealed extensive membrane rupture and cell shrinkage following Ir2@PEG NP treatment. In contrast, such morphological changes were not observed in the OXA‐treated group, indicating a potent abscopal anti‐tumor effect induced by the Ir2@PEG NPs (Figure S36).

FIGURE 4.

Therapeutic efficacy and immune modulation of Ir2@PEG NPs in primary and distant CT26 tumor‐bearing mouse model. (a) Schematic depiction of the treatment timeline. (b) Photograph of the dissected distant tumors after different treatments. (c) Representative images of ASC (red), CASPASE8 (green), and RIPK3 (yellow) co‐stained tumors collected from the CT26 tumor‐bearing mouse model after various treatments, with the nucleus counter‐stained with DAPI (blue). (d) Representative images of GPX4 (red) obtained from the CT26 tumor‐bearing mouse model subjected to various treatments. (e) Flow cytometry plots and (f) percentages of the M1 macrophages upon treatment. (g) Flow cytometry plots of and (h) Percentages of the M2 macrophages upon treatment. (i) Flow cytometry plots and (j) percentages of the maturation of CD80⁺ CD86⁺ cells upon treatment. I: Control, II: OXA 5 mg kg⁻¹, III: Ir2@PEG NPs 3 mg kg⁻¹, IV: Ir2@PEG NPs 5 mg kg⁻¹. Scale bar: 10 µm. n = 5, ***p < 0.001, ****p < 0.0001.

To demonstrate that Ir2@PEG NPs can induce both PANoptosis and ferroptosis in vivo, tumor tissues were harvested and analyzed. The results revealed that Ir2@PEG NPs markedly induced the formation of PANoptosome complexes (ASC, Caspase‐8, and RIPK3) and strongly activated the PANoptosis signaling pathway in tumor cells. Meanwhile, the expression of the ferroptosis regulator GPX4 was markedly downregulated, indicating interference with the cellular antioxidant defense against lipid peroxidation and thereby synergistically promoting ferroptosis in tumor cells. Consequently, the potent anti‐tumor effects of Ir2@PEG NPs are primarily attributed to their simultaneous activation of the complementary cell death pathways, PANoptosis and ferroptosis, which collectively promote tumor cell demise (Figure 4c,d). To shift the focus to the immune landscape and elucidate the potential mechanisms by which Ir2@PEG NPs activate innate and adaptive anti‐tumor immune responses, tumor tissues were harvested on Day 7 after the initial treatment and analyzed for immune cell infiltration. Flow cytometry revealed that Ir2@PEG NP treatment potently induced the polarization of tumor‐associated macrophages (TAMs) toward M1 phenotype (the proportion of M1 TAMs increased from 16.3% to 33.6%), accompanied by a marked decrease in M2 TAMs (from 33.9% to 14.9%) (Figure 4e–h). This macrophage repolarization effectively reprogrammed the tumor microenvironment with enhanced anti‐tumor inflammatory activity. Furthermore, we investigated the potential of Ir2@PEG NPs to activate innate immunity by analyzing serum cytokines. ELISA results demonstrated that serum levels of pro‐inflammatory cytokines (TNF‐α, IFN‐β, IL‐6, and IL‐1β) were significantly elevated in mice treated with Ir2@PEG NPs compared to the control group, confirming robust activation of the in vivo innate immune responses (Figure S37).

Concurrently, given the critical role of effector T lymphocytes in shaping adaptive anti‐tumor immune responses, we evaluated the maturation status of dendritic cells (DCs) to explore their immunogenicity. Ir2@PEG NP treatment (5 mg/kg) significantly increased the proportion of activated DCs expressing co‐stimulatory molecules CD86 and CD80 in primary tumors (from 14.6% to 36.2%) (Figure 4i,j), indicating robust DC activation and enhanced antigen‐presenting capacity. Moreover, Ir2@PEG NPs treatment significantly upregulated the proportion of CD3⁺ CD4⁺ T cells (from 25.5% to 66.0%) (Figure S38a,b), suggesting the potent activation of T cell‐mediated immune responses. Critically, the levels of immunosuppressive regulatory T cells (Tregs, CD3⁺ CD4⁺ CD25⁺ Foxp3⁺ cells) within the tumor microenvironment were markedly reduced following Ir2@PEG treatment (5 mg/kg) (from 24.9% to 7.0%) (Figure S38c,d), thereby effectively diminishing the tumor immune evasion. Finally, to explore the potential of Ir2@PEG NPs to promote immune‐memory‐related responses, we analyzed the proportion of effector memory T cells (CD3⁺ CD8⁺ CD44⁺ CD62L⁻). The results showed a significant increase in the proportion of effector memory T cells from 10.6% to 28.2% (Figure S38e,f), indicating that Ir2@PEG NPs can induce long‐lasting, effective anti‐tumor immune memory. Furthermore, a marked reduction in Treg cells and a significant increase in cytotoxic T cells and effector memory T cells were consistently observed in distal tumors (Figure S39) and in the spleens (Figure S40), further reinforcing the systemic nature of Ir2@PEG NP‐induced anti‐tumor immunity.

In summary, Ir2@PEG NPs exhibit potent therapeutic activity against immunologically “cold” MSS‐CRC through dual actions: direct tumor‐cell eradication via PANoptosis/ferroptosis and concurrent remodeling of the tumor immune microenvironment. Relative to the corresponding control groups, Ir2@PEG NPs promoted M1‐like macrophage repolarization, DC maturation, effector T‐cell activation, and suppression of distant tumors, collectively demonstrating enhanced antitumor immunity in MSS‐CRC. In this system, the downstream consequences of Ir2‐induced PANoptosis were consistently associated with immune activation rather than immune suppression. In addition to inducing canonical hallmarks of ICD, including CRT exposure and HMGB1/ATP release, Ir2 markedly reshaped the tumor microenvironment by enhancing innate immune activation and adaptive immune responses, as evidenced by enhanced M1‐like macrophage polarization, DC maturation, CD8⁺ T‐cell infiltration, and reduced Treg cell abundance. Collectively, these findings support the establishment of durable antitumor immune memory by Ir2‐triggered PANoptosis in the present MSS‐CRC model. Thus, our study not only identifies mitochondrial complex I dysfunction as a mechanistic contributor to PANoptosis induction but also provides experimental evidence that therapy‐induced PANoptosis can be immunologically beneficial in an appropriate tumor context. Nevertheless, the molecular determinants governing whether PANoptosis leads to antitumor immunity, deleterious inflammation, or mixed immune states remain to be further defined in future studies.

3. Conclusion

This study demonstrates that a cyclometalated iridium(III) complex, through rational structural modification, can effectively overcome the key limitations of oxaliplatin in MSS‐CRC by inducing concurrent PANoptosis and ferroptosis. Structure–function analysis revealed that the naphthyl‐functionalized Ir2, owing to its enhanced mitochondrial accumulation, inhibited the activity of complex I and triggered excessive ROS generation, which in turn led to PANoptosome assembly and the coordinated activation of pyroptotic, apoptotic, and necroptotic executors, as well as ferroptosis. Proteomic data further revealed that Ir2 induced an overall downregulation of the oxidative phosphorylation signaling pathway and its key component, the NADH dehydrogenase subunit ND2, which subsequently activated associated acute inflammatory pathways, thereby triggering an innate immune response. In vivo immunogenicity experiments demonstrated that this iridium(III) complex could polarize M1‐type macrophages and induce DC maturation, thereby reprogramming the immunosuppressive tumor microenvironment and eliciting potent innate and adaptive antitumor immune responses. Collectively, this strategy holds great promise for overcoming drug resistance and immune tolerance in MSS‐CRC by harnessing PANoptosis‐ and ferroptosis‐driven immune activation, thereby converting the inherently immunologically “cold” and immunodesert tumor microenvironment into an immunosensitive state.

Author Contributions

Xing Liu: investigation, data curation, validation, formal analysis, writing – original draft, writing – review and editing, resources, methodology, software. Tao Feng: methodology, software, data curation, investigation, validation, formal analysis, funding acquisition, writing – review and editing, writing – original draft. Xiaoting Yang: data curation, formal analysis, investigation. Zhuoli Chen: methodology, software, formal analysis, visualization. Qingming Pan: software, visualization, resources. Jinzhe Liang: methodology, software, supervision, visualization, validation, writing – original draft, writing – review and editing. Shuo Fang: resources, funding acquisition, visualization, methodology, supervision, data curation. Hui Chao: conceptualization, methodology, software, data curation, supervision, resources, project administration, formal analysis, validation, visualization, writing – review and editing, writing – original draft, funding acquisition, investigation. Yihang Pan: conceptualization, investigation, funding acquisition, writing – review and editing, project administration, supervision, validation, resources, data curation, methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: anie72924‐sup‐0001‐SuppMat.docx.

Data Availability Statement

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