2D Titanium-Based Nanozyme Induced a Ferroptosis-like Strategy for Triple-Negative Breast Cancer Therapeutics

Abstract

Ferroptosis, a regulated cell death pathway driven by iron-dependent lipid peroxidation, has emerged as a promising strategy for cancer therapy. Nevertheless, the direct delivery of iron can lead to systemic toxicity. In this study, a nonferrous ferroptosis-like strategy employing 2D Ti₃C₂ MXene nanosheets was developed. The Ti₃C₂ nanozyme has been shown to possess peroxidase- and catalase-like activities, capable of decomposing H₂O₂ into cytotoxic·OH and O₂, respectively. When exposed to near-infrared II laser irradiation, Ti₃C₂ induced a localized increase in temperature and amplified reactive oxygen species production, resulting in a synergistic effect that led to mitochondrial damage and glutathione peroxidase 4 inactivation. In vivo studies demonstrated tumor suppression in triple-negative breast cancer (TNBC) models, accompanied by M1 macrophage polarization and CD8⁺ T cell activation. This work provides a novel paradigm for nanocatalytic ferroptosis therapy, combining enzymatic activity, photothermal effects, and immunomodulation to overcome TNBC resistance.

Introduction

Breast cancer ranks as the most prevalent malignant tumor worldwide, with triple-negative breast cancer (TNBC) accounting for 15% to 20% of cases and exhibiting a dismal 5-year survival rate of just 12.2% in advanced stages [1,2]. Unlike other subtypes, TNBC cells display unique metabolic vulnerabilities, including dysregulated iron homeostasis and glutathione (GSH) balance, rendering them susceptible to ferroptosis, which is a nonapoptotic cell death driven by iron-dependent lipid peroxidation (LPO) [3,4]. While ferroptosis induction holds therapeutic potential, direct iron delivery risks systemic toxicity [5], and tumor antioxidant defenses often counteract reactive oxygen species (ROS)-mediated lethality [6,7].

Current strategies predominantly rely on iron-based nanomaterials to catalyze Fenton reactions, generating hydroxyl radicals (·OH) from H₂O₂ [8]. However, their efficacy is limited by tumor microenvironment (TME) constraints [9,10]. Moreover, excessive iron exacerbates oxidative stress in normal tissues [5]. Recent advances propose nonferrous metal-based nanozymes as alternatives, leveraging ROS amplification and GSH depletion to disrupt redox homeostasis [11,12]. Such approaches may concurrently stimulate antitumor immunity by modulating tumor-associated macrophage (TAM) polarization and dendritic cell maturation [13,14]. The development of nonferrous metals analogous to the concept of a ferroptosis-like effect may provide an alternative and effective strategy by integrating a nanoplatform that achieves maximum cell death through a synergistic ROS burst and GSH depletion.

Ti₃C₂ MXenes, characterized by their 2-dimensional (2D) structure, provide an optimal surface for catalytic site attachment, thereby facilitating the etching of the aluminum layer from a MAX-phase material [15]. In a manner analogous to that observed with iron-based nanomaterials, 2D Ti₃C₂ nanosheets demonstrate multiple enzymatic activities, including peroxidase (POD) activity, catalase (CAT) activity, and glutathione peroxidase (GPX) activity [16]. Moreover, previous studies have demonstrated that the introduction of external energy, such as laser irradiation, can effectively lower the activation energy barrier of metal cations, thereby potentially augmenting the catalytic efficiency of nanozymes [6]. Hence, near-infrared II (NIR-II) laser irradiation can be used as an exogenous stimulation focused on the tumor site, with the potential advantages of deep tissue penetration and precise spatiotemporal selectivity.

In this study, we develop a nonferrous ferroptosis-like strategy using 2D MXene nanozymes to synergistically integrate nanocatalytic ferroptosis, photothermal therapy, and immune activation (Fig. 1). The strategy involves depleting GSH, down-regulating GPX4, and amplifying lipid peroxidation, thereby overcoming TNBC resistance. This multimodal strategy has the potential not only to directly eradicate tumor cells but also to reshape the immunosuppressive TME, thus offering a transformative approach for the treatment of TNBC.

Fig. 1.

The schematic diagram of titanium-based nanozyme for immunomodulation-enhanced tumor therapy (created with BioRender.com).

Materials and Methods

Materials

The mouse breast cancer cells and normal human kidney cells were obtained from Jennio-bio (China). RPMI 1640 medium, fetal bovine serum, trypsin-EDTA, and penicillin-streptomycin were purchased from Gibco (USA). The live-dead cells kit (calcein-AM and propidium iodide) and the ROS indicator (dichlorofluorescin diacetate) were purchased from Beyotime Biotechnology (China).

Synthesis of Ti₃C₂ nanosheets

To prepare Ti₃C₂, a mixture of lithium fluoride and hydrochloric acid solution was stirred thoroughly. Then, the Ti₃AlC₂ powder was added to the mixture and left at 4 °C for 24 h. The resulting precipitate was collected by centrifugation, washed with water, and stored at 4 °C.

Characterization

Transmission electron microscopy (Talos F200i S) was used to obtain transmission electron microscopy and mapping images of Ti₃C₂. The crystal structure of Ti₃C₂ was captured using an x-ray diffractometer (Rigarke smartlab). Thermo 250XI spectrophotometers were used to study the x-ray photoelectron spectroscopy (XPS) spectra.

Cell culture

Mouse breast cancer cells (4T1) and normal human kidney cells (293T) were purchased from the Jennio-bio (China). The cancer cells were cultured with RPMI 1640 medium containing 10% fetal bovine serum and 1% penicillin-streptomycin, while the normal cells were cultured with Dulbecco’s modification of Eagle’s medium containing 10% fetal bovine serum and 1% penicillin-streptomycin. Both cell types were incubated at 37 °C with 5% CO₂.

The cell viability in vitro

The cytotoxicity of Ti₃C₂ to cancer cells was studied using the Cell Counting Kit-8 (CCK8) assay. 4T1 cells were seeded in 96-well plates and treated with different concentrations of Ti₃C₂ for 24 h. A laser with a power of 1.5 W/cm² was applied to the cells for 3 min. After removing the medium, fresh medium containing 10% CCK8 was added to the plates, and cell viability was measured using a multimode microplate reader (PerkinElmer EnVision 2105, USA). Live/dead cells were investigated using Calcein-AM/propidium iodide (PI) staining. Cancer cells were seeded in 24-well plates and exposed to Ti₃C₂ for 8 h before laser irradiation for 3 min. Calcein-AM/PI staining was added to the plates for 30 min and detected using confocal laser scanning microscopy (Carl Zeiss LSCM 880, Germany).

Evaluation of ROS, LPO, Fe²⁺, and GSH in vitro

ROS, LPO, Fe²⁺, and GSH generation were analyzed using 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA), Liperfluo, FerroOrange, and ThiolTracker Violet probes. 4T1 cells were seeded in 6-well plates and treated with various agents. After incubation with DCFH-DA, Liperfluo, FerroOrange, and ThiolTracker Violet probes for 20 min, the samples were imaged using confocal laser scanning microscopy.

The evaluation of mitochondrial dysfunction in vitro

The study evaluated mitochondrial dysfunction in 4T1 cells using a JC-1 probe to detect changes in mitochondrial membrane potential. The cells were seeded into a 6-well plate and confocal dishes, and Ti₃C₂ was added to the medium for 8 h. The cancer cells were detected using flow cytometry (FCM) and confocal laser scanning microscopy under laser irradiation. Mito-tracker dye was used to evaluate mitochondrial dysfunction in 4T1 cells. The cells were treated with phosphate-buffered saline (PBS), laser, Ti₃C₂, and Ti₃C₂ plus laser (1.5 W/cm², 3 min). After treatment, the cells were stained with Mito-tracker, and the nuclei with 4′,6-diamidino-2-phenylindole.

Density functional theory calculation

To investigate the activation process of H₂O₂, we utilized the Cambridge Serial Total Energy Package code based on density functional theory (DFT) to calculate the reaction between H₂O₂ and Ti₃C₂. The ultrasoft pseudopotential was used to describe the electron-ion interaction, and the generalized gradient approximation with Perdew–Burke–Ernzerhof parameters was employed to describe the exchange and correlation terms. The calculation utilized cutoff energy of 380 eV, and the appropriate κ-point mesh was selected based on the Monkhorst–Pack scheme. The self-consistent of first-principles calculations were considered converged when the maximum force on the atom, maximum displacement between cycles, maximum stress, and total energy changes were less than 0.01 eV/Å, 5.0 × 10⁻⁴ Å, 0.02 GPa, and 5.0 × 10⁻⁶ eV/atom, respectively.

Evaluation of apoptosis induced by Ti₃C₂

The study investigated the apoptosis of 4T1 cells using the Annexin V/PI apoptosis kit. The cells were seeded in a 6-well plate at a density of 10⁶ cells per well. After 24 h, Ti₃C₂ was added to the medium and incubated for 8 h. Following laser irradiation, the cells were harvested and stained with the Annexin V/PI apoptosis kit for FCM analysis.

Western blotting analysis

The cells were lysed, and total protein was extracted. Western blotting was performed as previously described. Protein concentration was determined using a BCA Protein Assay Kit (P0012S, Beyotime, Shanghai, China). Protein samples (20 μg/lane) were separated on a 12.5% sodium dodecyl sulfate polyacrylamide gel by electrophoresis and transferred to a polyvinylidene fluoride membrane. The membrane was blocked with 5% nonfat milk at room temperature for 1.5 h. Primary antibodies Anti-GPX4 (1:1000, 52455S, Cell Signaling Technology, MA, USA) and glyceraldehyde 3-phosphate dehydrogenase monoclonal antibody-horseradish peroxidase (1:20,000, MB001H, Bioworlde, MN, USA) were incubated overnight at 4 °C. After washing the membrane 3 times with tris-buffered saline with Tween 20 for 30 min, a secondary antibody (S0001, 1:5000, Affinity, Jiangsu, China) was added and incubated. Protein bands were detected using an ultrasensitive electrochemiluminescence kit (Jingxin, Guangzhou, China) and analyzed with ImageJ 6.0 software (National Institutes of Health, Bethesda, Maryland, USA) through an imaging system (iBright CL750, Invitrogen, Hercules, MA, USA).

The anticancer effect of Ti₃C₂ in vivo

The Animal Ethic Committee of Fifth Affiliated Hospital of Sun Yat-sen University approved the in vivo antitumor effect of Ti₃C₂. Balb/c mice were subcutaneously injected with 4T1 cancer cells on their back. After a week, the mice were treated with one of the following: PBS, Ti₃C₂, laser, and Ti₃C₂ plus laser (1.5 W/cm², 5 min). Tumor volume and body weight were measured every 4 d. After the fifth treatment, the mice were anesthetized and main organs and tumors were harvested and fixed in paraffin for further immunohistochemical assays. Routine blood tests were conducted to assess peripheral blood.

Statistical analysis

The experiments were conducted 3 times, and all comparisons were analyzed using either the Student t test or 2-way analysis of variance (ANOVA) with GraphPad 9.0. Statistical significance was marked as follows: ns for P > 0.05; * for P < 0.05;** for P < 0.01; *** for P < 0.001; and **** for P < 0.0001.

Results

Ti₃C₂ nanozyme preparation and characterization

The Ti₃C₂ nanozyme was produced via chemical exfoliation. A mixture of lithium fluoride and hydrochloric acid was used to remove the surface oxide layer, and ultrathin Ti₃C₂ nanosheets were obtained using selective etching techniques. The obtained Ti₃C₂ exhibits a 2D structure, as shown in Fig. 2A and B. EDS analysis revealed the presence of Ti, C, and O elements within Ti₃C₂, as illustrated in Fig. 2C. XPS measurements were performed to investigate the surface chemical compositions of Ti₃C₂ nanosheets. The high-resolution spectrum of Ti2p exhibited prominent peaks at 457.56 eV, corresponding to Ti2p3/2 (Fig. 2D). The asymmetric C1s spectrum was deconvoluted into 3 peaks at 287.36, 284.81, and 283.56 eV, which correspond to C=C, C–O, and Ti–C, respectively (Fig. 2E). Deconvolution of the O1s XPS spectrum (Fig. 2F) reveals the following chemical bonds: Ti–O (530.01 eV), Ti–(OH)_x (532.01 eV), and Ti–C–O_x (533.8 eV). The alteration in the valence state of Ti₃C₂ upon reaction with H₂O₂ was also examined using XPS, as depicted in Fig. S1. The Ti2p spectrum clearly showed the presence of Ti⁴⁺, indicating the formation of Ti⁴⁺ resulting from the reaction between Ti₃C₂ and H₂O₂. Subsequently, further interaction between Ti⁴⁺ and H₂O₂ led to the generation of an orange-colored peroxo-titanic acid [17]. Moreover, we employed an x-ray powder diffractometer to study the crystal structure of Ti₃C₂ nanosheets, revealing diffraction peaks that exhibited resemblance to those previously reported in the literature for Ti₃C₂ nanosheets (Fig. 2G) [18].

Fig. 2.

Characterization of Ti₃C₂ nanosheets. (A) The transmission electron microscopy (TEM) image of Ti₃C₂ nanosheets. The scale bar is 200 nm. (B) The atomic force microscopy (AFM) image of Ti₃C₂ nanosheets. The scale bar is 500 nm. (C) The transmission electron microscopy (TEM) image of Ti₃C₂ nanosheets and corresponding energy-dispersive x-ray spectroscopy (EDX) mapping. The scale bar is 500 nm. (D) The Ti2p spectrum of Ti₃C₂ nanosheets. (E) The C1s spectrum of Ti₃C₂ nanosheets_. (F) The O1s spectrum of Ti₃C₂ nanosheets. (G) X-ray diffraction (XRD) pattern of Ti₃C₂ nanosheets. (H) The electron spin resonance (ESR) spectrum for detecting ·OH using 5,5-dimethyl-1-pyrroline N-oxide (DMPO). (I) The ESR spectrum for detection of ¹O₂ using 2,2,6,6-tetramethylpiperidine hydrochloride (TEMP).

Multienzyme activity of Ti₃C₂ nanosheets

Nanozymes possessing enzyme-mimicking capabilities exhibit properties that are potentially advantageous for tumor therapy [10]. The catalytic activity was assessed via electron spin resonance spectroscopy by detecting the generation of ·OH and ¹O₂, serving as indicators (Fig. 2H and I). As shown in Fig. 3A, the POD-like activity catalyzes H₂O₂ decomposition into ·OH, while the CAT-like activity converted H₂O₂ into O₂, alleviating tumor hypoxia [19]. GSH, a pivotal redox buffer, plays a crucial role in tumor progression [20]. Elevated GSH levels are indispensable for effectively scavenging excessive ROS; hence, GSH depletion can enhance ROS-based antitumor therapy [21]. The GPX-like activity of Ti₃C₂ nanosheets was determined using 5,5′-dithiobis-2-nitrobenzoic acid. As illustrated in Fig. 3B, Ti₃C₂ nanosheets effectively consumed the GSH solution in a time-dependent manner, as evidenced by a decreased absorption peak at 412 nm of sulfhydryl groups. Ti₃C₂ nanosheets demonstrated dual enzyme-mimicking activities under physiological conditions (Fig. 3C and D). The catalytic ability of Ti₃C₂ in generating ¹O₂ was further investigated using the 1,3-diphenylisobenzofuran probe (Fig. 3E). A time-dependent increase in ¹O₂ generation through Ti₃C₂ catalytic activity was enhanced by laser irradiation.

Fig. 3.

The multienzyme mimicking activity of Ti₃C₂ nanosheets. (A) Ti₃C₂ nanosheets multienzyme activity scheme. (B) Glutathione peroxidase (GPX)-mimicking activity of Ti₃C₂ nanosheets using 5,5′-dithiobis-2-nitrobenzoic acid (DTNB) assay. (C) The generation of ·OH by Ti₃C₂ nanosheets reaction with H₂O₂. (D) The generation of O₂ by Ti₃C₂ nanosheets reacted with H₂O₂. (E) Time-dependent absorption spectra of 1,3-diphenylisobenzofuran (DPBF) treated by Ti₃C₂ under 1,064-nm irradiation (1.0 W/cm²). (F) The reactive oxygen species (ROS) generation of different concentrations of Ti₃C₂ nanosheets was incubated with 4T1 cells. (G) ROS generation of 4T1 cells treated by Ti₃C₂ nanosheets and laser irradiation. (H) In vitro antitumor effect of Ti₃C₂ nanosheets.

Real-time ROS generation, detected by the Incucyte S3 system using DCFH-DA probe, revealed a time-dependent increase in ROS accumulation over time in cancer cells induced by Ti₃C₂ (Fig. 3F and Fig. S2). The generation of ROS was further enhanced by NIR-II laser irradiation. Figure 3G illustrates the examination of ROS production in the Fenton-like reaction using a DCFH-DA probe in TNBC cells. The results demonstrated a marked increase in ROS levels within cancer cells upon H₂O₂ and laser irradiation, suggesting that Ti₃C₂ nanosheets effectively elevate intracellular ROS. Notably, the Ti₃C₂+H₂O₂+Laser group exhibited the highest ROS level, suggesting that laser irradiation effectively enhanced the catalytic activity of Ti₃C₂ nanosheets. The cytotoxicity of the Ti₃C₂-catalyzed reaction was evaluated via the CCK8 assay. As shown in Fig. 3H, the Ti₃C₂+Laser group exhibited higher cytotoxicity, while the Ti₃C₂+H₂O₂+Laser showed the lowest cell viability. The results demonstrated a dose-dependent decrease in cell viability induced by Ti₃C₂, with both laser irradiation and H₂O₂ potentiating its antitumor effect. The Ti₃C₂ shows negligible cytotoxicity to normal cells (Fig. S3).

The study of catalytic activity by DFT calculations

To investigate Ti₃C₂’s catalytic activity, we conducted first-principles calculations using the Cambridge Serial Total Energy Package code based on DFT. Initially, we optimized the crystal structures of the (001) surface of Ti₃C₂ employing the Broyden–Fletcher–Goldfarb–Shanno method, as depicted in Fig. S4. Band structure analysis (Fig. 4A) demonstrates Ti₃C₂’s metallic characteristics due to the absence of a significant band gap. Examination of the density of states highlights the primary contribution of d-orbital electrons of Ti to the total energy level at the Fermi surface (at 0 eV on the abscissa), with a smaller contribution from the p-orbitals of C (Fig. 4B). Then, we calculated the equilibrium crystal structures of the (001) surface of Ti₃C₂ and the adsorbed H₂O₂ surface using the Broyden–Fletcher–Goldfarb–Shanno optimization method. Subsequent calculations of adsorption energies for intermediates during H₂O₂ decomposition reveal their relative value to the adsorbed H₂O₂ structure. The free energy diagrams in Fig. 4C and E reveal that intermediates display negative adsorption energies, indicating stable adsorption during a hydroxyl radical desorption process.

Fig. 4.

The density functional theory (DFT) explorations, photothermal, and photoacoustic (PA) properties of Ti₃C₂ nanosheets. (A) Band structure of Ti₃C₂ and (B) the corresponding electron density of states projected onto the C and Ti atoms. (C and D) The free energy diagrams and corresponding reaction pathways of catalase (CAT)-like activity on the (001) surface of Ti₃C₂. (E and F) The free energy diagrams and corresponding reaction pathways of peroxidase (POD)-like activity on the (001) surface of Ti₃C₂. (G) Photothermal conversion of Ti₃C₂ nanosheets at different power densities. (H) Photothermal conversion of Ti₃C₂ nanosheets at different concentrations. (I) The concentration-dependent photothermal heating curves of Ti₃C₂ nanosheets incubated with 4T1 cells. (J) Photoacoustic images of Ti₃C₂ nanosheets at different concentrations. (K) The photoacoustic images and (L) quantities of Ti₃C₂ in vivo.

Based on the foregoing analysis, we propose the presence of CAT-like activity on the (001) surface of Ti₃C₂. The energy distribution and corresponding reaction pathways are shown in Fig. 4C and D. According to the proposed mechanism, 2 H₂O₂ molecules are initially adsorbed onto the Ti₃C₂ surface, decomposing to form OOH and OH adsorbed on the outer Ti atoms. The energy loss for the generation of H₂O molecules from the remaining H and OH is 0.117 eV, indicating that the reaction is thermodynamically feasible. Subsequently, OOH* and OH* adsorbed on Ti atoms react to form OH and H, respectively, while releasing O₂. This step is found to be the rate-determining step, as it results in an energy gain of 0.028 eV. Finally, Ti₃C₂ exhibits a relatively weak binding affinity for H₂O, resulting in an energy gain of 0.089 eV when OH* and H* combine to form the H₂O molecule. These findings corroborate the hypothesis that Ti₃C₂ mimics the CAT-like catalytic activity, which facilitates the conversion of H₂O₂ into O₂.

Moreover, the reaction pathways and energy distribution of POD-like activity are illustrated in Fig. 4E and F. The results demonstrated that H₂O₂ was captured by Ti₃C₂ and adsorbed onto Ti atoms with adsorption energy of −0.004 eV, indicating favorable energetics for H₂O₂ adsorbed onto Ti₃C₂. Subsequently, the reaction proceeds to the decomposition of the intermediate into 2 OH* ions, accompanied by a reduction in energy of 0.269 eV. Concurrently, the remaining OH* undergoes a reaction with the approaching H⁺ to form H₂O adsorbed on the surface. This reaction demonstrates the thermodynamically favorable conversion of OH* into H₂O. Overall, the multienzymatic-mimicking activity of Ti₃C₂ has been theoretically validated by a detailed investigation of the proposed reaction pathways and energy distributions.

The photothermal and photoacoustic property of Ti₃C₂ nanosheets

The multienzyme mimicking activities of Ti₃C₂ enable efficient generation of ROS upon external laser stimulation, rendering it a promising candidate for tumor treatment. Moreover, Ti₃C₂ exhibits superior NIR laser absorption capability, facilitating effective conversion of light energy to heat. With increasing laser power from 0.75 to 2.0 W/cm², the temperature of Ti₃C₂ rose from 30 to 80 °C. Additionally, a correlation between the concentration of Ti₃C₂ and the temperature of the solution was observed (Fig. 4G and H). These results suggest that Ti₃C₂ exhibits excellent photothermal performance and exceptional stability. The temperature of the Ti₃C₂ within tumor cells rapidly escalated beyond 80 °C with prolonged irradiation time. In contrast, cells incubated with PBS showed no significant temperature increase. These results highlight the exceptional photothermal conversion efficiency exhibited by Ti₃C₂ nanosheets. Photoacoustic (PA) imaging, a noninvasive technique that combines ultrasound and optical imaging modalities, provides enhanced spatial resolution and increased imaging depth compared to traditional optical imaging methods, enabling in vivo tissue penetration of up to 70 mm [22]. Due to its strong NIR absorption and excellent photothermal conversion capabilities, Ti₃C₂ demonstrates suitability for in vivo PA imaging investigation. Figure 4J illustrates the concentration-dependent increase in PA signal intensity of Ti₃C₂. Notably, Ti₃C₂ exhibits remarkable performance in PA imaging by efficiently accumulating at tumor sites. The effective accumulation of Ti₃C₂ in the tumor was confirmed through in vivo experiments (Fig. 4K and L). This unique capability enables its utilization for imaging guidance, laser amplification, and efficient enzyme-catalyzed therapeutic performance.

Ti₃C₂ nanosheets induce mitochondrial damage in TNBC cells

The ferroptosis-like therapy capability of Ti₃C₂ nanosheets was observed through live/dead cell staining using calcein AM/PI assay (Fig. 5A). The concentration of Ti₃C₂ nanosheets exhibited a positive ​correlation with the intensity of red fluorescence signal, which served as an indicator for cell death. The exceptional photothermal conversion synergized with enzymatic activity to induce cell death. Mitochondrial damage was further investigated using the Mito-tracker probe, demonstrating a direct correlation between Ti₃C₂ concentration and the extent of the mitochondrial damage inflicted on tumor cells. As shown in Fig. 5B, the Ti₃C₂-based catalytic treatments induced significant mitochondrial impairment due to the catalytic activity of nanozymes. Moreover, JC-1 staining results showed the multienzyme activity with photothermal therapy resulted in the most severe mitochondrial damage (Fig. 5C).

Fig. 5.

The calcium overload induced by Ti₃C₂’s catalytic activity. (A) Live/dead double staining of 4T1 cells with different treatments indicated by Calcein-AM (green, live cells) and propidium iodide (PI) (red, dead cells). Scale bar: 200 μm. (B) Confocal laser scanning microscopy (CLSM) images of mitochondria damage of 4T1 cells with different treatments. Scale bar: 20 μm. (C) The mitochondria damage of 4T1 cells with different treatments. Scale bar: 20 μm. (D) CLSM images of Ca²⁺ in 4T1 cells with different treatments. Scale bar: 100 μm. (E) Flow cytometry (FCM) and quantification results of Ca²⁺ in 4T1 cells with different treatments.

Excessive calcium influx can result in mitochondrial damage and subsequent of apoptosis of cancer cells. Disruption of calcium homeostasis within tumor cells can be induced by ROS [23]. Considering the robust ROS generation and the mitochondrial damage induced by the multienzyme mimicking activity of Ti₃C₂, it can be postulated that Ti₃C₂ may mediate Ca²⁺ overload. The Fluo-4 AM fluorescence probe was utilized to investigate the intracellular Ca²⁺ levels in cancer cells, indicating that Ti₃C₂ nanosheets induced an accumulation of Ca²⁺ in cancer cells, which was further augmented by laser irradiation (Fig. 5D and E).

The ferroptosis and the immunomodulatory effects induced by Ti₃C₂ nanosheets

Ferroptosis, a regulated form of cell death characterized by elevated ROS production and lipid peroxidation, leads to cellular demise [5]. As shown in Fig. 6A and B, Ti₃C₂ nanosheets induce elevated ROS production and lipid peroxidation in cancer cells, which is further exacerbated by laser irradiation. This elevation in lipid peroxidation is accompanied by a depletion of GSH, which is an important antioxidant [24]. These events are critical in the process of ferroptosis and play a crucial role in maintaining cellular redox homeostasis [25]. In Fig. 6B, the FerroOrange probe was used to detect the levels of Fe²⁺ in cancer cells following various treatments, demonstrating that Ti₃C₂ treatment induces intracellular accumulation of Fe²⁺, aligning with its role in promoting ferroptosis. The increased Fe²⁺ levels imply that Ti₃C₂ treatment triggers the ROS generation through the Fenton-like reaction, thereby contributing to lipid peroxidation and ultimately leading to ferroptosis (Fig. 6C to E). Moreover, this is supported by the significant decrease in the expression levels of GPX4, a characteristic enzyme involved in cellular ferroptosis (Fig. 6F). Together, these findings suggest that Ti₃C₂ nanosheets can initiate TME-responsive catalytic reactions, induce LPO damage, and ultimately trigger ferroptosis in cancer cells.

Fig. 6.

The ferroptosis and immunomodulation induced by the catalytic activity of Ti₃C₂. (A) Confocal laser scanning microscopy (CLSM) images of reactive oxygen species (ROS) in 4T1 cells with different treatments. Scale bar: 100 μm. (B) CLSM images of glutathione (GSH), Fe²⁺, and lipid peroxidation (LPO) in 4T1 cells with different treatments. Scale bar: 100 μm (top), 20 μm (middle), and 100 μm (bottom). (C) Flow cytometry (FCM) results of LPO in 4T1 cells with different treatments. (D) FCM results of Fe²⁺ in 4T1 cells with different treatments. (E) FCM results of GSH in 4T1 cells with different treatments. (F) The expression of glutathione peroxidase 4 (GPX4) in 4T1 cells with different treatments. (G) The inducible nitric oxide synthase (iNOS) and Arg1 expression in macrophages incubated with different cell supernatants. (H) CLSM images of live/dead M1 and M2 macrophages after different treatments. Scale bar: 100 μm. (I) FCM analysis of iNOS expression in M1 and M2 macrophages. (J) CLSM images of LPO expression in M1 and M2 macrophages. Scale bar: 20 μm.

The presence of TAMs within the TME plays a pivotal role in facilitating tumor proliferation [26]. TAMs are classified into 2 main subtypes: M1 macrophages, which exhibit anticancer properties, and M2 macrophages, which promote tumorigenesis [27,28]. The polarization is often exacerbated by tumor hypoxia, which directs TAMs toward the M2 phenotype, consequently hindering the efficacy of immunotherapy [29]. Nevertheless, the M1-type TAMs possess the potential to enhance immune activation and elevate the concentration of intertumoral H₂O₂ [30], which in turn could improve the catalytic activity of nanozymes [31]. Therefore, the integration of nanozymes with CAT-like activity can overcome these conditions and alleviate the associated limitations [32]. Moreover, combining nanozyme-based ferroptosis-like therapy with TME immunomodulation holds great promise for enhancing the overall effectiveness of cancer treatment [33]. Ti₃C₂ reshaped the immunosuppressive TME by polarizing TAMs toward antitumor M1 phenotype (iNOS⁺, Arg1⁻) (Fig. 6G). FCM revealed M1 macrophages exhibited higher inducible nitric oxide synthase (iNOS) expression than M2, correlating with resistance to ferroptosis (Fig. 6I). Conversely, M2 macrophages showed higher cell death (PI⁺) and elevated LPO levels (Fig. 6H), indicating selective ferroptosis susceptibility. In contrast, M1 macrophages demonstrated resistance to ferroptosis. The accumulation of LPO in different macrophage subtypes has been demonstrated and illustrated in Fig. 6J. The integration of ferroptosis induction and M1 macrophage polarization achieved dual therapeutic effects: (a) direct tumor cell killing via LPO-driven ferroptosis, and (b) immune activation through TAM reprogramming.

In vivo antitumor and the immune effect of Ti₃C₂ nanosheets

The in vivo antitumor effect was evaluated by creating a BALB/c mouse xenograft model of TNBC. The mice were randomly divided into 4 groups: PBS, laser, Ti₃C₂, and Ti₃C₂+laser. Ti₃C₂ (5 mg/kg) was administered via tail vein injection, followed by laser irradiation (1,064 nm, 1.5 W/cm², 5 min) after 6 h, as shown in Fig. 7A. Subsequently, the mice were euthanized, and their main organs, including the heart, liver, spleen, lung, kidney, and tumor, were collected and fixed for further experimentation. Tumor volume in the Ti₃C₂+laser group was reduced by 60.58% compared to PBS controls, while laser-only treatment showed no significant inhibition (Fig. 7B and C). Importantly, the addition of laser irradiation to Ti₃C₂ treatment elicited the most significant inhibitory effect on the tumor, suggesting a synergistic interaction between Ti₃C₂ and laser treatment in suppressing tumor growth. The body weight of the mice in the different treated groups did not exhibit any significant differences, indicating the safety profile of the treatment (Fig. 7D). To investigate the biocompatibility of the treatments, routine blood tests were conducted. The results indicated no significant differences between the white blood cells, neutrophil count, and red blood cells, as depicted in Fig. 7E. Histological analysis of major organs using hematoxylin and eosin staining under different treatment conditions is presented in Fig. 7F. Notably, Ti₃C₂ nanozyme-mediated immunomodulation-enhanced catalytic tumor therapy exhibited remarkable biocompatibility. The immunohistochemistry demonstrated that Ti₃C₂+laser treatment down-regulated Ki67 and GPX4, while up-regulating COX IV (Fig. 7G).

Fig. 7.

The antitumor and biocompatibility immunomodulation effect of Ti₃C₂ in vivo. (A) Schematic illustration of anticancer therapy in vivo. (B) The images of tumors after different treatments. (C) Tumor growth curves on tumor-bearing mice after various treatments. (D) Body weights of tumor-bearing mice after different treatments. (E) Blood routine of tumor-bearing mice after different treatments. (F) The hematoxylin and eosin (H&E) of the main organ in different treatments. Scale bar: 50 μm. (G) Expression of Ki67, COX IV, and glutathione peroxidase 4 (GPX4) were investigated by immunohistochemistry assays. Scale bar: 100 μm. n.s., not significant.

In this study, we aimed to investigate the immunomodulatory effect of Ti₃C₂ by analyzing tumor and spleen cells using FCM. Ti₃C₂ reprogrammed TAMs toward the antitumor M1 phenotype (CD80⁺), increasing their proportion from 11% (PBS group) to 40.8% (Ti₃C₂+laser group) (Fig. 8A and Fig. S5). CD8⁺ T cells recognize and eliminate infected cells or cancerous cells through the recognition of specific antigens [34]. In our study, we observed a significant increase in percentage of CD8⁺ T cells following treatment with Ti₃C₂ and laser therapy, suggesting enhanced activation and recruitment of CD8⁺ T cells to improve antitumor immune response (Fig. 8B). Treatment of the TME with Ti₃C₂ could induce polarization of the macrophages into M1-type macrophages, augmenting their phagocytic activity and synergizing with Ti₃C₂’s tumor cell-killing effect to inhibit tumor proliferation. Furthermore, Ti₃C₂-induced ferroptosis elicits the production of tumor-specific antigens, thereby activating systemic T-cell immune responses that secrete cytotoxic factors such as tumor necrosis factor-α, Granzyme B (GzmB), and interferon γ. Consequently, this amplifies the killing capacity of CD8⁺ T cells against tumors (Fig. 8C). This dual mechanism suggests that Ti₃C₂ treatment can not only directly eliminate tumor cells but also stimulate an immune response against the tumor. Overall, our findings indicate that combining Ti₃C₂ with photothermal therapy enhances CD8⁺ T cells activation, promotes macrophages polarization toward an antitumor phenotype, and generates tumor-specific antigens to facilitate an improved antitumor immune response.

Fig. 8.

The immunomodulation effect of Ti₃C₂ in vivo. (A) Flow cytometry results of tumor-associated macrophages (TAMs) from tumor tissues. (B) The CD8⁺ T cells from tumor tissues treated by different treatments. (C) The IFN-γ⁺CD8⁺ T cells from tumor tissues treated by different treatments. (D) The GzmB⁺CD8⁺ T cells from tumor tissues treated by different treatments. (E) The TNF-α⁺CD8⁺ T cells from tumor tissues treated by different treatments.

Conclusions

This study presents a novel ferroptosis-driven therapeutic strategy for TNBC utilizing 2D Ti₃C₂ nanosheets, which integrate nanocatalytic activity, photothermal effects, and immunomodulation synergistically. The Ti₃C₂-based approach has been demonstrated to exhibit dual enzyme-mimetic acidity. The decomposition of H₂O₂, a by-product of tumor activity, into highly levels of ROS is a hallmark of the Ti₃C₂-based approach. Additionally, it exhibits CAT-like activity, converting H₂O₂ into O₂, thereby alleviating tumor hypoxia. It is worth noting that the ferroptosis process, characterized by GSH depletion and GPX4 down-regulation, disrupts redox homeostasis, leading to lipid peroxidation and mitochondrial dysfunction. This Ti₃C₂-based platform overcomes the limitations of conventional ferroptosis inducers by combining spatiotemporal control, immune activation, and hypoxia alleviation, offering a transformative strategy for TNBC therapy.

Data Availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Supplementary Materials

Supplementary 1

Figs. S1 to S5