Red-light-triggered microenvironment-responsive sustained carbon monoxide release for enhanced tumor therapy

Abstract

Carbon monoxide (CO), a gasotransmitter, has gained attention as a potential therapeutic agent in cancer treatment. The precise and sustained release of CO is crucial for minimizing its toxicity and enhancing its therapeutic efficacy. We have developed a light-gated, microenvironment-responsive CO release platform for precise and sustained CO delivery. FeCO was used as the CO donor and integrated into the micelle core containing tertiary amine (TA) residues and a Pd-based photosensitizer (PdTPTBP). In the first stage, the H₂O₂ generated during light irradiation, in combination with GSH, triggers the release of CO and Fe²⁺ from FeCO. Subsequently, Fe²⁺ reacts with H₂O₂via a Fenton reaction, further promoting sustained CO release under dark conditions in the second stage. Light irradiation acts as a gating mechanism to achieve precise and sustained CO release. This CO delivery platform can be efficiently internalized by 4T1 tumor cells and, upon 630 nm light irradiation, releases CO intracellularly to induce ferroptosis. By synergistically disrupting mitochondrial function, it exhibits effective antitumor activity in 4T1 tumor-bearing mice.

---

Jian Cheng

Jian Cheng received his PhD under the supervision of Professor Jinming Hu. He subsequently completed postdoctoral training with Professor Chen Zhu and Professor Jinming Hu at the First Affiliated Hospital of the University of Science and Technology of China. He is currently an Associate Researcher at the School of Biomedical Engineering, University of Science and Technology of China. His research primarily focuses on gaseous signaling molecule-based functional materials.

---

1. Introduction

Several endogenous gaseous signaling molecules—including nitric oxide (NO), hydrogen sulfide (H₂S), and carbon monoxide (CO)—have shown distinct tumor-inhibiting activities via mechanisms such as mitochondrial dysfunction, DNA damage, anti-angiogenesis, and immune modulation.^1–6 Recent studies have focused on stimuli-responsive nanocarriers to precisely control the release of these gases in the tumor microenvironment, improving both efficacy and biosafety.^7,8 Carbon monoxide (CO), once considered solely a toxic gas, has been recognized as a potent gasotransmitter with anti-inflammatory, anti-apoptotic, and cytoprotective properties.^3,9 In recent years, its potential in cancer therapy has garnered significant attention.¹⁰ Notably, the biological effects of CO are highly concentration-dependent. At low concentrations, CO can enhance mitochondrial function in tumor cells, promote ATP synthesis, and increase cellular activity, thereby facilitating tumor cell proliferation and migration.¹¹ In contrast, high concentrations of CO exhibit cytotoxic effects by disrupting mitochondrial respiration, inducing oxidative stress, and activating apoptotic signaling pathways, ultimately inhibiting tumor growth and triggering programmed cell death.^12,13 To address the challenges associated with CO delivery, carbon monoxide-releasing molecules (CORMs) have been developed, particularly those that respond to disease-specific stimuli.^14–16 Among these, endogenous microenvironmental cues—such as elevated glutathione (GSH) levels and excessive hydrogen peroxide (H₂O₂) production in pathological tissues, especially in tumors or inflamed areas—present promising opportunities for site-specific CO release.^17–21

Stimuli-responsive release strategies that leverage disease microenvironments provide enhanced spatiotemporal control of therapeutic gas release, minimizing off-target effects and systemic toxicity.^22–24 Nevertheless, microenvironment-triggered systems often face the risk of premature release during circulation or in non-targeted tissues with mildly elevated GSH or H₂O₂ levels.^25–27 In addition, although H₂O₂ can naturally accumulate in the tumor microenvironment, its concentration is often insufficient to achieve significant therapeutic effects.^5,28–30 To overcome this, external stimuli such as light irradiation have been introduced as a control mechanism. Red light (600–700 nm) is especially appealing for its deeper tissue penetration and low phototoxicity.^31,32 Light-triggered systems enable precise, non-invasive CO release but face challenges such as limited light access in deep tissues^33,34 and tissue optical heterogeneity, which affects release uniformity and efficiency.^35–37 Consequently, the integration of both endogenous and exogenous triggers represents a promising synergistic strategy for achieving controlled, site-specific gas therapy, although it necessitates careful material design to balance responsiveness, stability, and therapeutic efficacy.

In this study, we successfully encapsulate a diiron hexacarbonyl complex (FeCO) into a diblock copolymer bearing tertiary amine (TA) residues and a Pd-based photocatalyst (PdTPTBP). Upon red light irradiation, molecular oxygen is converted into H₂O₂. In the presence of GSH, this process induces the release of CO and ferrous ions (Fe²⁺), the latter of which further participate in a Fenton reaction to generate hydroxyl radicals (˙OH) that, in turn, trigger additional CO release. Under red light irradiation, the micelles effectively induce intracellular generation of hydroxyl radicals, leading to mitochondrial membrane potential collapse and enhanced lipid peroxidation, ultimately resulting in tumor cell death. In tumor-bearing mice, this system demonstrates excellent therapeutic efficacy (Scheme 1).

Scheme 1 (a) Schematic illustration of the red light-induced accumulation of H₂O₂, followed by GSH-responsive release of CO and Fe²⁺ in the tumor microenvironment. The released Fe²⁺ catalyzes a Fenton reaction with H₂O₂ to generate ROS, while CO release persists even in the absence of light. (b) The preparation of M1 and M2 micelles via supramolecular self-assembly. (c) Red light-triggered activation of tumor microenvironment-responsive intracellular CO release for cancer therapy.

2. Experimental procedure

2.1. Sample preparation

2.1.1. Synthesis of diiron hexacarbonyl complex (FeCO). Fe₃(CO)₁₂ (2.19 g, 4.35 mmol, 1 equiv.) was dissolved in anhydrous THF (20 mL) under a nitrogen atmosphere. Propanethiol (0.54 g, 8.70 mmol, 2 equiv.) was added, and the mixture was heated at 70 °C for 2 hours, during which the color gradually changed from dark green to reddish-brown. After completion, the reaction mixture was filtered through diatomaceous earth, and the solvent was removed under reduced pressure. The crude product was purified by column chromatography (petroleum ether/ethyl acetate = 20 : 1) to give a reddish-brown oil. Recrystallization from DCM/n-hexane at −40 °C afforded a red solid (1.0 g, 2.5 mmol, 58%). Due to the presence of cis/trans isomers and unpaired electrons, the product exhibits paramagnetism, leading to complex NMR spectra with challenges in field locking and shimming. ¹H NMR (400 MHz, CDCl₃, Scheme S1a and Fig. S1) δ 2.40 (td, J = 7.5, 5.1 Hz, 2H), 2.08 (t, J = 7.5 Hz, 1H), 1.70 (dh, J = 29.4, 7.4 Hz, 2H), 1.55 (s, 1H), 1.44 (h, J = 7.5 Hz, 2H), 1.27 (d, J = 10.6 Hz, 3H), 1.04 (dt, J = 15.1, 7.3 Hz, 1H), 0.97–0.81 (m, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 209.70, 208.89, 41.15, 40.00, 27.10, 26.28, 26.12, 25.72, 13.50, 13.31.

2.1.2. Synthesis of the PEG-b-P(C7A-co-Pd) (PC7APd) diblock copolymer through RAFT polymerization. C7AM (52.8 mg, 0.25 mmol, 50 equiv.), PdM (29.5 mg, 0.025 mmol, 5 equiv.), PEG-based macro-chain transfer agent (26.5 mg, 0.005 mmol, 1.0 equiv.), and AIBN (0.16 mg, 0.001 mmol, 0.2 equiv.) were dissolved in 150 μL of DMSO in a reaction tube equipped with a magnetic stir bar. The solution was degassed by three freeze–pump–thaw cycles and sealed under vacuum. The reaction was carried out at 70 °C in an oil bath for 8 hours. After completion, the reaction mixture was quenched in liquid nitrogen and the tube was opened. The mixture was diluted with THF (2 mL), and the resulting residue was dialyzed against deionized water and lyophilized to yield a green powder. The resulting diblock copolymer was denoted as PEG₁₁₃-b-P(C7A_0.93-co-Pd_0.07)₄₃ by ¹H NMR analysis (Fig. S3a). The molecular weight distribution (M_w/M_n) was determined to be 1.11 (Fig. S3b).

2.1.3. Fabrication of M1 micelles. 1 mL of THF containing 200 μL stock solution of PC7APd (5 mg mL⁻¹) and 100 μL stock solution of FeCO (5 mg mL⁻¹), were quickly injected into 4 mL of DI water in one shot. After that, THF was removed via dialysis (MWCO = 14 kDa) against DI water. The resultant micellar nanoparticles were used for further experiments. Nile Red or IR820-loaded micelles were prepared using a similar method as described above, with the substitution of FeCO by Nile Red or IR820 during the formulation process.

2.2. Detection of CO release from M1 micelles with a commercial CO detector

CO release from M1 micelles (0.2 g L⁻¹) in the presence of GSH (10 mM) was monitored using a portable CO detector (Dräger Pac 8500). The detector and a sample vial containing M1 (0.2 g L⁻¹, 10 mL) were placed in a sealed transparent container (gas volume: 350 mL). The change in detector reading was recorded. The amount of released CO (N_CO) was calculated using the following equation:
where p is the partial pressure of CO, V_g is the gas phase volume (350 mL), V_l is the liquid phase volume (5 mL), R is the ideal gas constant (0.0821 atm L mol⁻¹ K⁻¹), T is the temperature (298.15 K), and k is Henry's law constant (1052.63 atm L mol⁻¹ under the experimental conditions).

2.3. Cytotoxicity assessment of M1 and M2 micelles

In the cell cytotoxicity assays, 4T1 cells was initially seeded into individual wells of 96-well plates. Each well contained 100 μL of DMEM complete medium, and the cells were seeded at a density of 1 × 10⁴ cells per well. These plates were subsequently placed in a 5% CO₂/95% air incubator for 24 hours to allow for proper cell adherence and growth. Following this incubation period, the cells were exposed to treatments involving varying concentrations of M1 or M2 micelles (0.08, 0.12, 0.16 g L⁻¹) to assess their cytotoxic effects. After a co-incubation duration of 6 hours in a 5% CO₂/95% air incubator, the cells were exposed to light irradiation at 630 nm with an intensity of 38 mW cm⁻² for 10 minutes, although some cells were not subjected to this irradiation treatment. After irradiation, the cells were allowed to continue incubating for an additional 24 hours. Following this period, MTT reagent (10 μL, 5 mg mL⁻¹) was added to each well, and the cells were incubated for 4 hours at 37 °C. Finally, the absorbance at 490 nm was recorded on a microplate reader (Thermo Fisher Scientific).

2.4. Animals and ethics

All animal experiments involved in this work were performed following a protocol approved by the Institutional Animal Care and Use Committee (University of Science and Technology of China, USTCACUC25120124095). BALB/c (female, 6–8 weeks) were purchased from the Experimental Animal Center of Anhui Medical University. All animals were maintained on 12 h/12 h light/dark cycles at 23–25 °C and 48–52% humidity.

2.5. Tumor-bearing BALB/c mice model

BALB/c female mice (6–8 weeks) were used for in vivo tumor inhibition analysis. The 4T1 tumor model was established at first using 4T1 cancer cells. In brief, to establish the subcutaneous mouse model of triple-negative breast cancer, 4T1 cells (3 × 10⁶) suspended in PBS were inoculated into the subcutaneous of the mice BALB/c mice, after 10 days, when the tumor reached ∼150 mm³.

2.6. Antitumor study

When the tumor volumes were around 150 mm³, the 4T1 tumor-bearing mice were intratumorally injected with various formulations. The healthy tumor-bearing mice were randomly divided into five groups (n = 4) and exposed to different treatments. Control: only administration with 25 μL PBS; M1 − hv: administration with 25 μL aqueous dispersion (0.2 g L⁻¹) of M1 micelles without light; M1 + hv: administration with 25 μL aqueous dispersion (0.2 g L⁻¹) of M1 micelles; M2 − hv: administration with 25 μL aqueous dispersion (0.2 g L⁻¹) of M2 micelles without light; M2 + hv: administration with 25 μL aqueous dispersion (0.2 g L⁻¹) of M2 micelles. 0.5 hours later, the + hv group tumor site was irradiated with a 630 nm laser at a power density of 0.2 W cm⁻² for 15 min. The tumor growth was monitored by measuring the perpendicular diameter of the tumors (i.e., length and width, respectively) using calipers every two days. The estimated volume was calculated according to the formula: Tumor volume (mm³) = 0.5 × length × width². The weight of each mouse was also measured every two days. On the third day after different treatments, one tumor in each group was collected and stained by hematoxylin and eosin (H&E) staining, terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL), and antigen Ki67.

2.7. Statistical analysis

Data are presented as the mean ± standard deviations and were analyzed using Prism 8.0 software (GraphPad, San Diego, California, USA) and student's t-test. p values lower than 0.05 were considered to be statistically significant.

3. Results and discussion

3.1. CO release from diiron hexacarbonyl complex (FeCO)

A diiron hexacarbonyl complex was employed as a CO donor, capable of releasing CO upon activation by GSH.³⁸ However, its CO release profile is highly dependent on the molecular structure.^39,40 In this study, the FeCO complex was synthesized via the reaction of Fe₃(CO)₁₂ with propanethiol in THF. UV-vis spectroscopy (λ = 330 nm) revealed that the complex showed only partial degradation (17.8%) following 30-minute incubation with 1 mM GSH. Interestingly, the addition of H₂O₂ (100 μM) significantly enhanced the degradation efficiency to 60% under the same conditions, whereas H₂O₂ alone induced negligible changes (Fig. 1b–d and Fig. S4). Direct evidence of CO release from the FeCO complex was obtained via headspace analysis using a portable CO detector (Dräger Pac8500). Notably, the GSH/H₂O₂ system triggered rapid CO release (1.72 μmol within 60 minutes; efficiency: 28.6%), which was significantly higher than that observed in control experiments.

Fig. 1 (a) Proposed mechanism of CO release from FeCO in the presence of H₂O₂ and GSH. (b–d) Evolution of UV-vis absorbance spectra of DMF/PBS (v/v= 1 : 9) solutions of FeCO (100 μM) in the presence of (b) H₂O₂ (0.1 mM) and GSH (1 mM), (c) GSH, and (d) H₂O₂. (e) Absorption intensity changes at 330 nm of FeCO at different conditions. (f) CO release profiles of DMF/PBS (v/v = 1 : 9) solutions of FeCO (100 μM) under different conditions.

Concurrent analysis of the 330 nm UV-vis absorption decay and CO release profiles identified a distinct acceleration phase (pink region) in the complex's degradation kinetics (Fig. 1e and f). The CO release mechanism of the FeCO complex was investigated using infrared spectroscopy and mass spectrometry. IR analysis revealed a significant decrease in Fe–CO stretching bands (∼2100–1900 cm⁻¹) upon treatment with GSH and H₂O₂, indicating Fe–CO bond cleavage (Fig. S5a). MS results further confirmed the formation of [Fe–GSH] related intermediate species, supporting a synergistic degradation pathway in which GSH induces ligand substitution while H₂O₂ promotes oxidative breakdown of the complex (Fig. S5b). The released Fe²⁺ initiates a Fenton reaction, generating additional ROS that further contribute to the release process. To investigate the role of highly reactive oxygen species produced by the Fenton reaction, hydroxyl radicals (˙OH) were generated using Fe²⁺ and H₂O₂. When FeCO was treated with ˙OH alone, no significant change in absorbance at 330 nm was observed, indicating limited degradation under these conditions. However, upon the addition of GSH, a rapid decrease in the 330 nm peak occurred, suggesting that GSH plays a critical role in activating FeCO under oxidative environments. These results indicate that while ˙OH provides strong oxidative stress, the presence of GSH enables effective FeCO activation, and their combination synergistically accelerates its decomposition (Fig. S6). This process establishes a positive feedback loop in which GSH and oxidative intermediates cooperatively promote further degradation of FeCO and sustained CO release (Fig. 1a).⁴¹

3.2. Photo-mediated CO release in aqueous media

Building upon the CO release properties of the iron carbonyl system, we propose the development of micellar nanomaterials capable of precisely triggering CO release under physiological conditions. To mitigate potential leakage associated with purely chemical activation via GSH and H₂O₂, we aim to incorporate photo-triggered mechanisms to achieve precise control over CO delivery.

Firstly, we synthesized a diblock copolymer (PC7APd) incorporating both tertiary amine (C7A) and photosensitizer (PdTPTBP) (Scheme S1c). Through supramolecular self-assembly, the hydrophobic FeCO complex was encapsulated within the micelles (M1). The resulting M1 micellar nanoparticles exhibited a hydrodynamic diameter (〈D_h〉) of approximately 135 nm. Upon exposure to red light (630 nm, 38 mW cm⁻²), no significant change in 〈D_h〉 was observed (Fig. 2a–c and Fig. S7). Both M1 and M2 exhibited excellent colloidal stability in aqueous solution, with negligible changes in particle size over 7 days (Fig. S8). The loading content and loading efficiency of FeCO were determined to be 3.4% and 8.4%, respectively, as measured by ICP-OES (Fig. S9).

Fig. 2 (a) and (b) TEM images and (c) intensity-average hydrodynamic diameter distributions of M1 micelles with or without 630 nm irradiation for 60 min. (d) UV/vis spectra of M1 micelles under 630 nm irradiation and GSH. (e) UV-vis spectral changes of M1 micelles in the presence of GSH after 60 s irradiation with 630 nm light, followed by continuous monitoring in the dark. (f) Absorption intensity changes at 330 nm of M1 micelles under various conditions. (g) CO release profiles of M1 micelles under various conditions. (h) Changes in oxygen concentrations of aqueous dispersions of M1 micelles under 630 nm light irradiation. Inset: Detection of H₂O₂ using a test strip. (i) EPR spectra of M1 micelles with or without 630 nm irradiation for 30 min. DMPO was used as a spin-trapping agent (10 mM). In all cases the following conditions were used: M1 micelles (0.2 g L⁻¹), GSH (10 mM), and the irradiation intensity was 39 mW cm⁻².

The M1 micelles exhibited triggered CO release upon simultaneous red-light irradiation and GSH exposure. UV–vis spectroscopic analysis showed a rapid decrease in absorbance at 330 nm under 630 nm light irradiation. Notably, even after the light was turned off following 1 minute of irradiation, the absorbance at 330 nm continued to decline during subsequent monitoring in the dark, indicating sustained CO release (Fig. 2d–f and Fig. S10). Quantitative CO detection confirmed rapid CO release from the M1 system under concurrent light irradiation (630 nm) and GSH stimulation (Fig. 2g). Intriguingly, a sustained release profile was observed even after cessation of light exposure (2 minutes of irradiation followed by incubation in the dark). Control experiments showed only minimal CO generation—validated by both UV–vis spectroscopy and a CO detector—when either light or GSH was applied alone, underscoring the requirement for dual activation to achieve efficient CO release. These results demonstrate that light irradiation serves as the pivotal trigger for CO release from M1 micelles. Within the tumor microenvironment, light functions as a gating mechanism that precisely controls CO liberation, achieving spatiotemporal regulation of therapeutic release.

Subsequently, we systematically investigated the mechanism underlying light-triggered CO release. A significant decrease in dissolved oxygen concentration was observed for M1 micelles under 630 nm light irradiation, dropping from approximately 7.25 to 0.11 mg L⁻¹ within 2 minutes. The formation of O₂˙⁻ and H₂O₂ were confirmed by the EPR tests and H₂O₂ test strips (Fig. 2h and Fig. S11).

We propose that during CO release, the liberated Fe²⁺ reacts with the generated H₂O₂via the Fenton reaction, thereby promoting further decomposition of the iron–carbonyl complexes. A chromogenic reaction with 2,2′-bipyridine (bipy) was employed to detect free iron ions in the mixture of M1 micelles and GSH under red light irradiation. The [Fe(bipy)₃]²⁺ complex is formed, confirming the release of ferrous ions (Fig. S12a and b). EPR spectroscopy successfully confirmed the generation of hydroxyl radicals (˙OH) during the process.⁴² In addition to the characteristic ˙OH signals, adjacent spectral peaks were observed, which may correspond to intermediate species derived from the decomposition of iron–carbonyl complexes. However, the exact identity of these secondary species requires further verification through additional analytical characterization (Fig. 2i).⁴³ Methylene blue scavenging assays demonstrated a significant decrease in the Abs.@664 nm under concurrent light irradiation and GSH treatment (Fig. S12c–f), directly evidencing ˙OH generation. The above results demonstrate that, under the presence of GSH and light irradiation, CO release can be directly triggered. Furthermore, the subsequent Fenton reaction involving the generated ferrous ions also facilitates CO release and provides a driving force for its sustained release in the dark (Scheme 1a).

3.3. Intracellular delivery of CO in 4T1 cells

Intracellular CO release is critical for achieving therapeutic effects. To monitor CO release within cells, we further employed the CO-responsive fluorescent probe 1-Ac to confirm the generation of CO. The fluorescence signal of 1-Ac at 660 nm can be specifically activated by CO, whereas treatment with GSH or light alone did not produce any significant fluorescence signal (Fig. 3a–c and Fig. S13). Subsequently, we evaluated the internalization efficiency of M1 micelles in 4T1 murine breast cancer cells. Confocal laser scanning microscopy (CLSM) images revealed that the intracellular fluorescence signal of Nile Red gradually increased with prolonged co-incubation of the cells and the micellar carrier (0.2 g L⁻¹). We observed that the fluorescence signal tended to plateau after 6 and 8 hours of co-incubation; therefore, 6 hours was selected as the incubation time for subsequent experiments (Fig. S14). Confocal imaging revealed that the micelles colocalized with both mitochondria and lysosomes, and were also partially distributed throughout the cytoplasm (Fig. 3d and Fig. S15). After cellular uptake of M1 micelles, the 1-Ac probe (0.1 mM) was markedly activated under 630 nm light irradiation (38 mW cm⁻²) for 10 minutes in the intracellular environment. Red fluorescence was observed in the M1 + hv group, whereas no significant fluorescence signal was detected in the control group (Fig. 3e and f). In summary, the micelles can be efficiently internalized by cells and enable intracellular CO release. This light-initiation and microenvironment-responsive system for sustained CO delivery has promising potential for biomedical applications.

Fig. 3 (a) Proposed working mechanism for selective CO detection using the 1-Ac probe. (b) Irradiation time-dependent fluorescence emission spectra of micellar dispersion of M1 (0.2 g L⁻¹) in the coexistence of 1-Ac (10 μM) (λ_ex = 580 nm). (c) Fluorescence intensity changes at 660 nm of 1-Ac at various concentrations. (d) Confocal fluorescence images of cells stained with DAPI (blue, nucleus), LysoTracker (green, lysosomes), MitoTracker (yellow, mitochondria), and M-IR820-labeled micelles (red). (e) CLSM images of 4T1 cells incubated with M1 (0.2 g L⁻¹) and 1-Ac under varying conditions. (f) Quantitative analysis of intracellular fluorescence intensity of 1-Ac under different treatment conditions. Data are shown as the mean ± s.d. (n = 3). ***p < 0.001.

3.4. In vitro anticancer effects of M1 micelles

The CO release effect of M1 micelles within tumor cells encourages further exploration of their potential biological applications for CO delivery. To evaluate the cytotoxicity of M1 and M2 micelles, 4T1 cells were co-incubated with various concentrations of the two formulations. Under dark conditions, M2 exhibited negligible cytotoxicity, even at a concentration of 0.16 g L⁻¹. In contrast, upon light irradiation, M2 generated reactive oxygen species (ROS), resulting in a mild tumoricidal effect (∼54.4%). Notably, M1 at 0.16 g L⁻¹ exhibited measurable cytotoxicity in the absence of light (∼38.4%), which was further enhanced under light exposure, achieving approximately 76.2% cell killing (Fig. 4a and b). Under dark conditions, endogenous ROS and GSH within tumor cells may induce partial CO release from FeCO complexes, resulting in a modest antitumor effect. Upon light irradiation, the release of both CO and ferrous ions was significantly enhanced, subsequently triggering Fenton reactions that amplified the therapeutic outcome. Notably, under identical irradiation conditions, M1 exhibited superior therapeutic efficacy compared to M2, indicating that CO release plays a critical role in the antitumor activity (Fig. 4c). Cytotoxicity assays on RAW 264.7 macrophages and L929 fibroblasts showed that both M1 and M2 exhibited low toxicity across a concentration range up to 0.2 g L⁻¹. Cell viability remained above 85% in both cell types, confirming the good biocompatibility of the materials for further therapeutic applications (Fig. S16).

Fig. 4 (a and b) Cell viability of 4T1 cells in the presence of (a) M1 or (b) M2 micelles (0.2 g L⁻¹) with or without 630 nm irradiation. (c) Comparison of cytotoxicity between M1 and M2 micelles toward 4T1 cells under 630 nm light irradiation. (d) CLSM images of 4T1 cells stained with JC-1 dyes after incubation with M1 micelles (0.2 g L⁻¹), with or without subsequent 630 nm irradiation (38 mW cm⁻², 10 min). (e) CLSM images of C11-BIODIPY (red) and oxC11-BIODIPY (green). Data are shown as the mean ± s.d. (n = 6). **p < 0.01, ****p < 0.0001.

To further elucidate the mechanism underlying the antitumor effect of M1, we evaluated mitochondrial membrane potential (Δψm) and lipid peroxidation levels in cells following M1 treatment.^44,45 We observed that intracellular CO release, in combination with Fenton reactions, resulted in a loss of Δψm, as evidenced by a decrease in the JC-1 aggregate-to-monomer fluorescence ratio. Under normal conditions, JC-1 accumulates in healthy mitochondria as red-fluorescent aggregates, whereas a reduction in Δψm causes it to remain in the green-fluorescent monomeric form. We observed a marked increase in green fluorescence intensity in cells treated with M1 under light irradiation, whereas no significant change was detected in either the non-irradiated group or the PBS control (Fig. 4d). Additionally, a substantial amount of hydroxyl radicals was detected in tumor cells treated with M1 under light irradiation, as indicated by the fluorescence signal of the hydroxyl radical probe HPF (Fig. S17). These results indicate that mitochondrial damage was substantially induced by intracellular CO release and subsequent Fenton reactions.

CO promotes ferroptosis by disrupting mitochondrial function, elevating intracellular ROS levels, and inducing lipid peroxidation.^46,47 The oxidizing species generated within cells, together with Fe²⁺ released from M1 under light irradiation, may further trigger ferroptosis, thereby enhancing the antitumor efficacy of CO. The changes in these indicators are closely related to ferroptosis.^48,49 Finally, lipid peroxidation (LPO), a well-established marker of ferroptosis, was monitored using the C11-BODIPY probe. Under oxidative conditions, C11-BODIPY is oxidized to form oxC11-BODIPY, which emits green fluorescence, contrasting with the red fluorescence emitted by its reduced form, thus serving as an effective indicator of intracellular lipid peroxidation. Upon treatment with M1 under light irradiation, a significant increase in green fluorescence was observed, indicating a strong induction of lipid peroxidation within the cells (Fig. 4e).

These results collectively suggest that CO, Fe²⁺, and excess reactive oxygen species may contribute to the antitumor effect of M1, with their synergistic interaction driving the optimal therapeutic outcome. In addition to inducing ferroptosis and lipid peroxidation-mediated tumor cell death, CO may further contribute to the antitumor efficacy by modulating the tumor microenvironment, including the reprogramming of immune and stromal cells.^50,51 A more comprehensive mechanistic understanding of these immunometabolic effects will be pursued in future studies.

3.5. Antitumor activity of M1 micelles in a mouse model

To evaluate the antitumor potential of delivered CO in vivo, we assessed its therapeutic efficacy using an established 4T1 murine tumor model (Fig. 5a). In mice bearing 4T1 tumors (∼150 mm³), treatment with M1 micelles followed by 630 nm laser irradiation (0.2 W cm⁻², 10 min) led to rapid and sustained tumor inhibition over a period of approximately 13 days, with an inhibition rate of 86.6%.

Fig. 5 (a) Schematic illustration of the therapeutic schedule used to evaluate the in vivo antitumor efficacy. (b) Representative images of tumors excised from mice following different treatments. (c) Tumor growth curves of mice subjected to various treatments. (d) Average body weight of tumor-bearing mice during the treatment period. (e) Representative histological images of tumor sections stained with H&E, Ki67, and TUNEL. Tumor tissues were collected from mice on the third day after different treatments. Data are shown as the mean ± s.d. (n = 4). ***p < 0.001, ****p < 0.0001.

The ROS generated by the M2 group under light exposure also exerted a photodynamic therapeutic effect, leading to a moderate tumor inhibition rate of 63.7%. In contrast, the light-only group showed no significant changes in tumor volume compared to the PBS control group (Fig. 5b and c).

Additionally, H&E staining, Ki67 staining, and TUNEL staining revealed that tumors treated with M1 micelles in combination with laser irradiation exhibited significant apoptosis and reduced cell proliferation (Fig. 5e). We also assessed the potential toxicity of M1/M2 micelles in vivo. The micelles showed negligible cytotoxicity under the tested concentrations. No significant differences in body weight were observed between the treated and control groups throughout the treatment period (Fig. 5d). Histological analysis with H&E staining revealed no detectable damage to major organs in mice treated with M1/M2 micelles, similar to the control group (Fig. S13). In contrast, M1 micelles demonstrated excellent tumor eradication efficacy and good biocompatibility. These results suggest that M1 micelles have significant potential for tumor therapeutic applications.

It should be noted that while 630 nm red light enables precise activation of CO release and effective treatment of superficial tumors, its limited tissue penetration may restrict efficacy in deeper tumors. This limitation could be addressed in future studies by introducing NIR-responsive components.

4. Conclusions

In summary, we have developed a light-gated, microenvironment-responsive CO release platform by incorporating the diiron hexacarbonyl complex, C7A, and PdTPTBP into micelles. Under red light irradiation and in the presence of GSH, the micelles rapidly release CO, while the ROS accumulated during light exposure further activate FeCO, resulting in sustained CO release. This red light-initiation tumor microenvironmental responsive CO release platform not only enhances the precision of CO release but also improves its long-lasting efficacy. The current CO release nanoplatform demonstrates good in vivo antitumor activity, indicating its potential as a therapeutic agent for tumor intervention.

Author contributions

Jian Cheng, Lun-lan Li and Yao Lu supervised the project. Jian Cheng, Jiahui Sheng and Guihai Gan conceived the experiments; Jiahui Sheng, Yao Wang, and Fei Li carried out the experiments; Jian Cheng, Yao Lu and Jiahui Sheng wrote the manuscript. All authors discussed the results and commented on the paper.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the SI. The Supplementary Information file contains detailed experimental procedures, additional characterization data, spectra, control experiments, and supporting figures that complement the findings presented in the main text. See DOI: https://doi.org/10.1039/d5tb01166c

Acknowledgements

This work was supported by the National Key R&D Program of China (2024YFA1509203), the National Natural Science Foundation of China (NNSFC) projects (22305239 and 52425306), the USTC Research Funds of the Double First-Class Initiative (YD2060006006), the Key Scientific Research Foundation of the Education Department of Province Anhui (2023AH053320), and the Key Research Project of Anhui Provincial Health Commission (AHWJ2022a007).

References

1. A. J. Burke, F. J. Sullivan, F. J. Giles and S. A. Glynn, Carcinogenesis, 2013, 34, 503–512.
2. H. J. Zhao, Y. T. Zhang, X. D. Fu, C. R. Chen, S. Khattak and H. G. Wang, Front. Pharmacol., 2023, 14, 1280308.
3. R. Motterlini and L. E. Otterbein, Nat. Rev. Drug Discovery, 2010, 9, 728–743.
4. Y. B. Shi, L. Cheng, Y. Lyu and Z. J. Shi, Nitric Oxide-Biol. Chem., 2025, 156, 1–8.
5. W. B. Dirersa, G. Getachew, A. Wibrianto, A. S. Rasal, V. S. Gurav, M. Z. Fahmi and J. Y. Chang, J. Colloid Interface Sci., 2023, 647, 528–545.
6. G. Getachew, Y. C. Tien, T. C. Kan, W. B. Dirersa, A. Wibrianto, S. Orchirbat, J. S. Chang, A. S. Rasal, V. Gurav, S. Kizhepat and J. Y. Chang, Chem. Eng. J., 2023, 467, 143560.
7. Y. S. Wang, T. Yang and Q. J. He, Natl. Sci. Rev., 2020, 7, 1485–1512.
8. R. C. Liu, Y. J. Peng, L. G. Lu, S. J. Peng, T. F. Chen and M. X. Zhan, J. Nanobiotechnol., 2021, 19, 443.
9. J. Bi, E. Witt, M. K. McGovern, A. B. Cafi, L. L. Rosenstock, A. B. Pearson, T. J. Brown, T. B. Karasic, L. C. Absler, S. Machkanti, H. Boyce, D. Gallo, S. L. Becker, K. Ishida, J. Jenkins, A. Hayward, A. Scheiflinger, K. L. Bodeker, R. Kumar, S. K. Shaw, S. K. Jabbour, V. A. Lira, M. D. Henry, M. S. Tift, L. E. Otterbein, G. Traverso and J. D. Byrne, Adv. Sci., 2024, 11, 2308346.
10. J. Chai, J. Zhu, Y. Tian, K. Yang, J. Luan and Y. Wang, J. Mater. Chem. B, 2023, 11, 1849–1865.
11. P. Nuhn, B. M. Künzli, R. Hennig, T. Mitkus, T. Ramanauskas, R. Nobiling, S. C. Meuer, H. Friess and P. O. Berberat, Mol. Cancer, 2009, 8, 37–47.
12. Y. Zhou, W. Yu, J. Cao and H. Gao, Biomaterials, 2020, 255, 120193.
13. Q. Wang, C. Zhang, Y. Zhao, Y. Jin, S. Zhou, J. Qin, W. Zhang, Y. Hu, X. Chen and K. Yang, Biomaterials, 2024, 309, 122606.
14. J. Cheng, G. Gan, Z. Shen, L. Gao, G. Zhang and J. Hu, Angew. Chem., Int. Ed., 2021, 60, 13513–13520.
15. L. S. Lazarus, H. J. Esquer, A. D. Benninghoff and L. M. Berreau, J. Am. Chem. Soc., 2017, 139, 9435–9438.
16. S. Romanski, B. Kraus, U. Schatzschneider, J.-M. Neudörfl, S. Amslinger and H.-G. Schmalz, Angew. Chem., Int. Ed., 2011, 50, 2392–2396.
17. A. Bansal and M. C. Simon, J. Cell Biol., 2018, 217, 2291–2298.
18. Q. Liu, W. Ge, S. Martínez-Jarquín, Y. He, R. Wu, M. Stoffel and R. Zenobi, Angew. Chem., Int. Ed., 2023, 62, e202213703.
19. C. Song, M. Chen, J. Tan, J. Xu, Y. Zhang, G. Zhang, X. Hu and S. Liu, J. Am. Chem. Soc., 2023, 145, 17755–17766.
20. D. Chen, G. Zhang, R. Li, M. Guan, X. Wang, T. Zou, Y. Zhang, C. Wang, C. Shu, H. Hong and L.-J. Wan, J. Am. Chem. Soc., 2018, 140, 7373–7376.
21. C. Korupalli, C. C. Kuo, G. Getachew, W. B. Dirersa, A. Wibrianto, A. S. Rasal and J. Y. Chang, J. Colloid Interface Sci., 2023, 643, 373–384.
22. K. Chang, X. Sun, M. Fu, B. Han, X. Jiang, Q. Qi, Y. Zhang, T. Ni, C. Ge and Z. Yang, J. Mater. Chem. B, 2024, 12, 2737–2745.
23. J. Yang, S. Pan, S. Gao, T. Li and H. Xu, Biomaterials, 2021, 271, 120721.
24. Y. Opoku-Damoah, Z. P. Xu, H. T. Ta and R. Zhang, ACS Appl. Bio Mater., 2024, 7, 7585–7594.
25. F. Zhou, B. Feng, T. Wang, D. Wang, Q. Meng, J. Zeng, Z. Zhang, S. Wang, H. Yu and Y. Li, Adv. Funct. Mater., 2017, 27, 1606530.
26. X. Xu, P. E. Saw, W. Tao, Y. Li, X. Ji, S. Bhasin, Y. Liu, D. Ayyash, J. Rasmussen, M. Huo, J. Shi and O. C. Farokhzad, Adv. Mater., 2017, 29, 1700141.
27. J. Liu, Y. Pang, Z. Zhu, D. Wang, C. Li, W. Huang, X. Zhu and D. Yan, Biomacromolecules, 2013, 14, 1627–1636.
28. G. Ma, Z. Liu, C. Zhu, H. Chen, R. T. K. Kwok, P. Zhang, B. Z. Tang, L. Cai and P. Gong, Angew. Chem., Int. Ed., 2022, 61, e202207213.
29. X. Shi, L. Gui, L. Zhou and S. Wei, Mater. Adv., 2021, 2, 7681–7690.
30. T. Ali, D. Li, T. N. F. Ponnamperumage, A. K. Peterson, J. Pandey, K. Fatima, J. Brzezinski, J. A. R. Jakusz, H. Gao, G. E. Koelsch, D. S. Murugan and X. Peng, Cancers, 2024, 16, 2171–2204.
31. J. Karges, Angew. Chem., Int. Ed., 2022, 61, e202112236.
32. M. Li, K. H. Gebremedhin, D. Ma, Z. Pu, T. Xiong, Y. Xu, J. S. Kim and X. Peng, J. Am. Chem. Soc., 2022, 144, 163–173.
33. Y. Li, M. Jiang, Z. Deng, S. Zeng and J. Hao, Adv. Sci., 2021, 8, 2004391.
34. C. Zhu, M. Yu, J. Lv, F. Sun, A. Qin, Z. Chen, X. Hu, Z. Yang and Z. Fang, J. Nanobiotechnol., 2024, 22, 708.
35. Y. Yang, K. Long, Y. Chu, H. Lu, W. Wang and C. Zhan, Adv. Funct. Mater., 2024, 34, 2402975.
36. T. Zhou, L. Wu, N. Ma, F. Tang, J. Chen, Z. Jiang, Y. Li, T. Ma, N. Yang and Z. Zong, Bioeng. Transl. Med., 2023, 8, e10368.
37. B. Sana, A. Finne-Wistrand and D. Pappalardo, Mater. Today Chem., 2022, 25, 100963.
38. C. Gao, X. Liang, Z. Guo, B.-P. Jiang, X. Liu and X.-C. Shen, ACS Omega, 2018, 3, 2683–2689.
39. X. Jiang, L. Long, H. Wang, L. Chen and X. Liu, Dalton Trans., 2014, 43, 9968–9975.
40. X. Jiang, L. Chen, X. Wang, L. Long, Z. Xiao and X. Liu, Chem. – Eur. J., 2015, 21, 13065–13072.
41. L. Long, X. Jiang, X. Wang, Z. Xiao and X. Liu, Dalton Trans., 2013, 42, 15663–15669.
42. J. Ye, W. Lv, C. Li, S. Liu, X. Yang, J. Zhang, C. Wang, J. Xu, G. Jin, B. Li, Y. Fu and X. Liang, Adv. Funct. Mater., 2022, 32, 2206157.
43. Y. A. Belousov and T. A. Belousova, Polyhedron, 1999, 18, 2605–2608.
44. C. S. F. Queiroga, A. S. Almeida and H. L. A. Vieira, Biochem. Res. Int., 2012, 2012, 749845.
45. J.-R. Alonso, F. Cardellach, S. López, J. Casademont and Ò. Miró, Pharmacol. Toxicol., 2003, 93, 142–146.
46. W. Cao, M. Sun, K. N. Yu, L. Zhao, Y. Feng, C. Tan, M. Yang, Y. Wang, F. Zhu, L. Chen, L. Nie, Y. Zhao, G. Chen and W. Han, Cell Death Discovery, 2024, 10, 42.
47. L. Yu, Y. Xu and J. S. Kim, CCS Chem., 2023, 5, 1718–1736.
48. S. Feng, D. Tang, Y. Wang, X. Li, H. Bao, C. Tang, X. Dong, X. Li, Q. Yang, Y. Yan, Z. Yin, T. Shang, K. Zheng, X. Huang, Z. Wei, K. Wang and S. Qi, Mol. Biomed., 2023, 4, 33.
49. M. Wang, W. Zhang, B. Liu, B. Ding, K. Li, P. A. Ma and J. Lin, J. Am. Chem. Soc., 2025, 147(20), 17372–17383.
50. Z. Nemeth, E. Csizmadia, L. Vikstrom, M. Li, K. Bisht, A. Feizi, S. Otterbein, B. Zuckerbraun, D. B. Costa, P. P. Pandolfi, J. Fillinger, B. Döme, L. E. Otterbein and B. Wegiel, Oncotarget, 2016, 7, 23919–23932.
51. P. P. Oza and K. Kash, Pharmacol. Ther., 2023, 249, 108502.

---

Footnote

† Jiahui Sheng, Yao Wang and Fei Li contributed equally to this article.

---