GSH-responsive self-assembled nanoplatform synergistically enhances cuproptosis through metabolic reprogramming and oxidative stress amplification
Authors: Bo Huang, Weijin Zhang, Yunjie Wang, Xian Luo, Wenda Wu, Rong Shen, Xiangquan Liu, Zhibo Zhang, Yi Gao, Yingying Wu, Fangwei Zeng, Yuan Huang, Jianyun Yu, Suxiao Wang, Ting Wu, Shengyu Wang, Shuitu Feng, Fanghong Luo
Published: 2026

ABSTRACT
Cuproptosis, an emerging copper-dependent regulated cell death pathway, demonstrates significant potential for overcoming therapeutic resistance in oncology. However, its clinical translation remains constrained by the poor bioavailability of copper ionophores and intrinsic resistance mechanisms in tumor cells. Here, we developed a tumor microenvironment-responsive nanoparticle platform (PEMA) co-loading an MPC1 (mitochondrial pyruvate carrier 1) overexpression plasmid and the copper ionophore Elesclomol to establish a synergistic “metabolic reprogramming-oxidative stress amplification” strategy. The PEMA nanoparticle design incorporated disulfide bonds to deplete intracellular glutathione (GSH), while Elesclomol-mediated copper transport induced mitochondrial dysfunction and reactive oxygen species (ROS) generation. In vitro, PEMA achieved >99% tumor cell eradication in ACHN renal carcinoma models, accompanied by characteristic DLAT oligomerization, FDX1 downregulation, and disruption of mitochondrial ultrastructure. In vivo, PEMA treatment induced substantial tumor regression in xenograft models without detectable systemic toxicity. This study establishes a novel therapeutic paradigm that integrates metabolic targeting with oxidative stress potentiation to overcome therapeutic resistance in solid tumors.

Introduction

As an emerging form of regulated cell death, cuproptosis is characterized by mitochondrial dysfunction triggered by intracellular copper ion (Cu2+) accumulation [[1], [2], [3]]. Mechanistically, Cu2+ directly binds to lipoylated components of the tricarboxylic acid (TCA) [4], leading to the aggregation of lipoylated mitochondrial proteins and downregulation of iron-sulfur cluster proteins, which ultimately induces proteotoxic stress and cell death [5]. Despite the promising potential of cuproptosis-based tumor therapies in achieving high efficacy and overcoming drug resistance, inherent limitations have hindered their clinical translation. (see Scheme 1)Scheme 1GSH-responsive self-assembled Nanoplatform synergistically enhances Cuproptosis through metabolic reprogramming and oxidative stress amplification.

GSH-responsive self-assembled Nanoplatform synergistically enhances Cuproptosis through metabolic reprogramming and oxidative stress amplification.

First, cuproptosis induction is highly dependent on copper ionophores [6,7]. Representative agents such as Elesclomol efficiently mediate Cu2+ translocation into cells [8], however, Elesclomol is rapidly cleared and metabolized in the systemic circulation, resulting in extremely low bioavailability within tumor tissues [9,10]. Second, tumor cells exhibit intrinsic resistance to cuproptosis, primarily attributed to two key factors: (i) the strict dependence of cuproptosis on mitochondrial Cu2+ concentration, where endogenous Cu2+ levels are typically insufficient to trigger effective cell death.and (ii) the overexpression of glutathione (GSH) in tumor cells, which chelates free Cu2+ and impairs Cu2+-protein interactions [11,12].

Notably, accumulating evidence has established a close link between cuproptosis and cellular energy metabolism [13]. Cells relying on mitochondrial oxidative phosphorylation (OXPHOS) display ∼1000-fold higher sensitivity to copper ionophores compared to glycolysis-dependent cells [14]. This phenomenon is attributed to the elevated levels of lipoylated TCA cycle enzymes (especially the pyruvate dehydrogenase complex) in OXPHOS-dependent cells, where lipoyl groups serve as direct Cu2+-binding sites to promote lipoylated protein aggregation [13,15]. Consequently, OXPHOS-dependent cells including numerous normal tissues-are more susceptible to copper toxicity than glycolysis-dominant tumor cells [16]. Thus, reprogramming tumor cell energy metabolism represents a potential strategy to enhance their sensitivity to cuproptosis.

Metabolic reprogramming is a hallmark of cancer and a key determinant of cuproptosis sensitivity. Mitochondrial pyruvate carrier 1 (MPC1) is the essential subunit of the mitochondrial pyruvate carrier complex, which mediates the influx of cytosolic pyruvate into mitochondria and fuels the tricarboxylic acid (TCA) cycle and oxidative phosphorylation (OXPHOS) [[17], [18], [19], [20]]. Tumor cells typically suppress MPC1 expression to limit mitochondrial metabolism and favor glycolysis, thereby reducing their vulnerability to cuproptosis. In contrast, restoring MPC1 expression enhances mitochondrial pyruvate utilization, increases TCA cycle flux, and elevates the abundance of lipoylated proteins (e.g., DLAT)—the direct targets of copper ions—thereby significantly sensitizing tumor cells to cuproptosis [[21], [22], [23], [24]]. Therefore, MPC1 overexpression represents a promising metabolic strategy to overcome The issue of insensitivity to cuproptosis [25,26].

To address these challenges, there is an urgent need to develop a novel (efficient Cu2+ transport-synergistic sensitization) strategy that facilitates mitochondrial Cu2+ overload, intracellular GSH depletion, and effective cuproptosis induction at low Cu2+ doses [27,28]. While increasing the dosage of copper ionophores may enhance intratumoral Cu2+ concentration, the potential systemic toxicity of excessive Cu2+ necessitates the development of safe and precise copper delivery systems [29]. Nanoparticle-based delivery strategies have emerged as promising tools for targeted drug delivery, offering improved bioavailability, controlled release, and reduced off-target effects [30,31].

Moreover, advanced nanocarrier design provides innovative solutions to overcome these limitations [32,33]. Ideal nanocarriers for cuproptosis therapy should possess three key characteristics: (i) tumor-specific targeting capacity to minimize off-target effects [34], (ii) controlled release capability responsive to tumor microenvironmental cues [35], and (iii) simultaneous delivery of multiple therapeutic agents [36]. Among these, GSH-responsive nanoplatforms have garnered significant attention due to their ability to deplete intracellular GSH (a major contributor to cuproptosis resistance) [29,37]. The disulfide bond-containing nanostructures not only serve as structural scaffolds but also act as GSH scavengers to disrupt redox homeostasis [38,39]. Additionally, the incorporation of targeting moieties (e.g., aptamers, antibodies) and gene delivery components enables multidimensional attack on tumor vulnerabilities [40,41]. Our PEMA nanoplatform precisely integrates these functions through rational molecular design.

This study developed a tumor microenvironment-responsive self-assembled nanoplatform (PEMA) that integrates three synergistic components: the nucleolin-targeting aptamer AS1411 [42,43], the cuproptosis-inducing agent Elesclomol, and an MPC1 overexpression plasmid. This design establishes a dual synergistic mechanism of “metabolic reprogramming-oxidative stress amplification” through three innovative features: (1) The disulfide bond structure enables GSH-responsive ROS burst in tumor tissues; (2) The AS1411 aptamer confers active targeting to nucleolin-overexpressing tumor cells; (3) The MPC1 plasmid reverses tumor metabolic reprogramming by enhancing mitochondrial oxidative phosphorylation efficiency. In vitro experiments demonstrated that PEMA achieved >99% cell cytotoxicity in ACHN renal carcinoma cells, accompanied by characteristic cuproptosis markers including DLAT oligomerization and FDX1 downregulation [44]. In ACHN xenograft models, PEMA exhibited remarkable tumor-targeting capability with 24-h peak accumulation and prolonged retention. The treatment group showed 78.6% tumor volume reduction compared to controls, with no detectable systemic toxicity observed. This interdisciplinary study synergizes nanotechnology, metabolic regulation, and oxidative stress mechanisms to overcome clinical translation barriers of cuproptosis therapy. The PEMA platform not only enables efficient delivery of copper ionophores and MPC1 gene regulation but also creates a “metabolic-oxidative” synergistic lethal effect through GSH depletion and ROS amplification. This strategy opens new avenues for treating resistant tumors by simultaneously targeting multiple tumor-specific vulnerabilities.

Herein, we rationally designed a GSH-responsive self-assembled nanoplatform (PEMA) that integrates AS1411-mediated targeted delivery, disulfide bond-based GSH depletion, MPC1-driven metabolic reprogramming, and Elesclomol-induced oxidative stress amplification into a single system. These four modules function synergistically rather than additively: targeted delivery ensures tumor-specific accumulation; GSH depletion reduces copper homeostasis defense; MPC1 overexpression enhances mitochondrial metabolism and cuproptosis sensitivity; and oxidative stress amplification further triggers robust cuproptosis. Collectively, PEMA addresses the critical limitations of current cuproptosis therapy, providing a promising synergistic strategy for solid tumor treatment.

Results and discussion

MPC1 enhanced cuproptosis by metabolic reprogramming

To investigate the relationship between tumor cell metabolic gene expression and cuproptosis, we performed CCK8 copper ion stress assays on four well-characterized cell lines: ACHN, A549, OVISE, and MDA-MB-231. The results (Fig. 1a–d) revealed that cells with higher glycolytic activity (ACHN, A549) [18] exhibited greater resistance to copper ion toxicity compared to cells with lower glycolytic activity (OVISE, MDA-MB-231) [19], suggesting a positive correlation between aerobic metabolism and copper ion sensitivity.Fig. 1| MPC1 Enhanced Cuproptosis by Metabolic Reprogramming. (a-d) Quantitation of cell viability of ACHN,A549,OVISE and MDA-MB-231 cells after 24 h of treatment with indicated concentrations of CuCl2. Data were presented as the means ± SD (n = 3 independent experiments). (e) PDH activity of different cells. Data were presented as the means ± SD (n = 3 independent experiments). (f) Bioinformatics analysis of MPC1 and SLC31A1 in ACHN, A549, OVISE and MDA-MB-231 cell line. (g) PPI network of MPC1 and metabolic proteins. (h) Schematic illustration of MPC1 enhanced cuproptosis by Metabolic Reprogramming. (i) ATP levels and (j) PDH activity of WT-ACHN cells and MPC1-OE ACHN cells. (k-o) Analysis of targeted metabolomics. (p) Quantitation of cell viability of WT-ACHN cells and MPC1-OE ACHN cells after 24 h of treatment with indicated concentrations of CuCl2. (q) Quantitation of cell viability of ACHN cells with/without galloflavin after 24 h incubation of indicated concentrations of CuCl2. (r) WT-ACHN cells and MPC1-OE ACHN cells analyzed for DLAT oligomerization 24 h after treatment with PBS and 50 μm CuCl2.

| MPC1 Enhanced Cuproptosis by Metabolic Reprogramming. (a-d) Quantitation of cell viability of ACHN,A549,OVISE and MDA-MB-231 cells after 24 h of treatment with indicated concentrations of CuCl2. Data were presented as the means ± SD (n = 3 independent experiments). (e) PDH activity of different cells. Data were presented as the means ± SD (n = 3 independent experiments). (f) Bioinformatics analysis of MPC1 and SLC31A1 in ACHN, A549, OVISE and MDA-MB-231 cell line. (g) PPI network of MPC1 and metabolic proteins. (h) Schematic illustration of MPC1 enhanced cuproptosis by Metabolic Reprogramming. (i) ATP levels and (j) PDH activity of WT-ACHN cells and MPC1-OE ACHN cells. (k-o) Analysis of targeted metabolomics. (p) Quantitation of cell viability of WT-ACHN cells and MPC1-OE ACHN cells after 24 h of treatment with indicated concentrations of CuCl2. (q) Quantitation of cell viability of ACHN cells with/without galloflavin after 24 h incubation of indicated concentrations of CuCl2. (r) WT-ACHN cells and MPC1-OE ACHN cells analyzed for DLAT oligomerization 24 h after treatment with PBS and 50 μm CuCl2.

Further analysis of TCGA database showed that PDH (pyruvate dehydrogenase) expression positively correlated with aerobic metabolism (Fig. 1e), while SLC31A1 (a copper transporter) expression remained unchanged, indicating a direct link between metabolic gene expression and cuproptosis [20]. Protein-protein interaction analysis of metabolic genes identified MPC1 as a central regulator (Fig. 1g). MPC1 facilitates pyruvate transport into mitochondria, driving TCA cycle and OXPHOS [[21], [22], [23], [24]], Moreover, the expression of MPC1 gene is highly correlated with the overall survival of patients with renal cell carcinoma (Fig. S1). In tumor cells, MPC1 inhibition leads to cytoplasmic pyruvate accumulation, enhanced glycolysis, and reduced mitochondrial metabolism, thereby decreasing cuproptosis sensitivity (Fig. 1h).

To validate MPC1's role, we overexpressed MPC1 in ACHN cells (low PDH activity). Overexpression increased PDH enzyme activity and ATP levels (Fig. 1i and j), confirming its role in metabolic reprogramming. Targeted metabolomics (MRM) revealed elevated TCA cycle intermediates citrate, α-KG and NADH, indicating enhanced OXPHOS but potential ETC (electron transport chain) limitations (Fig. 1k–o). Notably, PDH activation upregulates FDX1 (a cuproptosis-critical gene), forming a “metabolic-toxic” feedback loop that amplifies copper ion toxicity [[45], [46], [47], [48]].

Functional assays demonstrated that MPC1-overexpressing ACHN cells exhibited heightened sensitivity to copper ions (5 μM) compared to wild-type cells (Fig. 1p). Additionally, treatment with Galloflavin (an LDH inhibitor) forced metabolic switch to OXPHOS, further increasing copper ion sensitivity (Fig. 1q). Western blot analysis confirmed increased DLAT oligomerization (a hallmark of cuproptosis) in MPC1-overexpressing cells (Fig. 1r).

To further validate the generalizability of MPC1-mediated cuproptosis sensitization, we extended the study to three representative solid tumor cell lines: A549 (lung cancer), MDA-MB-231 (triple-negative breast cancer), and OVISE (Human ovarian cancer). As shown in Supplementary Fig. S2, MPC1 overexpression significantly reduced cell viability in a dose-dependent manner following CuCl2 treatment in all three cell lines. At 5 μM CuCl2, MPC1-OE groups exhibited viability of only 35–45%, compared to 72–78% in wild-type controls. These results confirm that MPC1-driven metabolic reprogramming is a universal mechanism enhancing cuproptosis across diverse solid tumors.

Mechanistically, MPC1 overexpression increases mitochondrial pyruvate influx, which significantly boosts TCA cycle flux and OXPHOS activity. Elevated TCA metabolism directly enriches mitochondrial lipoylated proteins (e.g., DLAT)-the primary targets of Cu2+ in cuproptosis. More lipoylated substrates provide additional binding sites for Cu2+, accelerating DLAT oligomerization and subsequent mitochondrial dysfunction. Meanwhile, enhanced OXPHOS elevates NADH levels, which further potentiate Cu2+-mediated redox stress and FDX1 downregulation. Thus, MPC1-driven metabolic rewiring specifically amplifies the molecular prerequisites for cuproptosis, rather than merely increasing general mitochondrial activity [49,50].

Synthesis and Characterization of PSP and PEMA

To achieve efficient copper ion transport and MPC1 gene transfection for synergistic cuproptosis therapy, we developed a novel tumor microenvironment-responsive nanoparticle system, designated as ‌PEMA‌. The synthesis involved a stepwise functionalization and co-assembly strategy, as illustrated in Fig. 2a and detailed in Supplementary Fig. S3–S4.Fig. 2| Synthesis and Characterization of PEMA. (a) Schematic illustration of synthetic processes of PEMA. (b) Hydrodynamic diameters of PSP and PEMA. (c) Zeta potential of PSP and PEMA. Data were presented as the means ± SD (n = 3 independent experiments). (d) SEM image of PEMA. (e) Representative SEM images of PEMA in the presence of 10 mM GSH. (f) FTIR spectra of PLGA, PLGA-NH2, PSP and PEMA (g) Peak assignment from FTIR spectra. (h) Fluorescence spectra of RhB-PEMA in response to different concentrations of PEMA. (i) Agarose gel electrophoresis of PEMA treated with different concentrations of GSH(0-15 mM). (j) GSH-responsive of PLGA and PSP (l) Fluorescence spectra of PSP-ICG in response to different concentrations of GSH.

| Synthesis and Characterization of PEMA. (a) Schematic illustration of synthetic processes of PEMA. (b) Hydrodynamic diameters of PSP and PEMA. (c) Zeta potential of PSP and PEMA. Data were presented as the means ± SD (n = 3 independent experiments). (d) SEM image of PEMA. (e) Representative SEM images of PEMA in the presence of 10 mM GSH. (f) FTIR spectra of PLGA, PLGA-NH2, PSP and PEMA (g) Peak assignment from FTIR spectra. (h) Fluorescence spectra of RhB-PEMA in response to different concentrations of PEMA. (i) Agarose gel electrophoresis of PEMA treated with different concentrations of GSH(0-15 mM). (j) GSH-responsive of PLGA and PSP (l) Fluorescence spectra of PSP-ICG in response to different concentrations of GSH.

Material Construction and Synthesis. The nanoparticle scaffold was constructed using poly(lactic-co-glycolic acid) (PLGA). To confer responsiveness to the reductive tumor microenvironment, PLGA was first modified with terminal amino groups to obtain PLGA-NH2, followed by cross-linking with dithiobis(succinimidyl propionate) (DSP) to introduce cleavable disulfide bonds. The resulting polymer, denoted as PSP, was designed to undergo GSH-triggered degradation for controlled drug release. The final PEMA nanoparticles were formulated by co-loading three key therapeutic components—the copper ionophore Elesclomol, an MPC1 overexpression plasmid, and a tumor-targeting aptamer—onto the PSP scaffold via combined hydrophobic interactions and electrostatic adsorption.

Physicochemical Characterization. The size, morphology, and surface properties of the nanoparticles were systematically characterized. Dynamic light scattering (DLS) analysis revealed that PSP nanoparticles had an average diameter of ∼80 nm, which increased to ∼100 nm after drug loading (PEMA), with low polydispersity indices (PDI) of 0.08 and 0.11, respectively (Fig. 2b). Zeta potential measurements tracked the surface charge evolution: from −21 mV for pristine PLGA to +17 mV after amination, +9 mV after DSP cross-linking, and finally +7 mV for PEMA, indicating good colloidal stability (Fig. 2c). Scanning electron microscopy (SEM) confirmed the spherical morphology and uniform size distribution of PEMA (Fig. 2d). Fourier-transform infrared (FTIR) spectroscopy verified the successful introduction of functional groups at each synthetic step (Fig. 2g).

Functional Payload Verification and GSH-Responsive Behavior. The successful loading of therapeutic cargos was confirmed through multiple assays. Using rhodamine B (RhB) as a fluorescent probe, a dose-dependent increase in fluorescence intensity confirmed copper ion loading (Fig. 2h). Agarose gel electrophoresis demonstrated the efficient encapsulation and GSH concentration-dependent release of the MPC1 plasmid (Fig. 2i, Fig. S4). The critical GSH-responsive degradation of PEMA was comprehensively validated: SEM imaging showed nanoparticle morphology transitioning from smooth spheres to irregular aggregates upon GSH treatment (Fig. 2e). DLS indicated a rapid size increase from ∼100 nm to micron-scale aggregates due to polymer precipitation (Fig. 2j and k); and fluorescence spectroscopy using ICG-loaded PSP confirmed GSH-triggered payload release in a concentration-dependent manner (Fig. 2l, Fig. S6).

In summary, these results confirm the successful fabrication of PEMA nanoparticles with well-defined physicochemical properties, high payload capacity, and specific responsiveness to the reductive tumor microenvironment, establishing a solid foundation for subsequent biological evaluation.

The concentration-dependent responsiveness of PEMA precisely matches the in vivo GSH distribution, laying a solid foundation for its selective tumor activation and biosafety.In normal blood and most tissues, GSH level (<1 mM) is far below the response threshold of PEMA, so the nanoparticles remain stable during circulation with negligible premature drug leakage. Even in GSH-rich liver and kidney (5-10 mM), PEMA only undergoes slow degradation without burst drug release, which is consistent with our in vivo results that liver and renal function markers all stay within normal ranges.

In contrast, the high intracellular GSH in tumors (10-20 mM) can trigger rapid PEMA disassembly, releasing payloads to induce potent cuproptosis for selective tumor killing. The current relatively high GSH response threshold of PEMA avoids non-specific activation in normal tissues, and we will further optimize the disulfide bond density in follow-up studies to improve its biocompatibility and therapeutic window.

Although PSP nanoparticles exhibit GSH-responsive release in the tumor microenvironment, passive accumulation via the EPR effect is limited and inefficient. The AS1411 aptamer was introduced for active tumor targeting, as nucleolin is highly overexpressed on the surface of tumor cells but low in normal cells, ensuring tumor-specific enrichment and reduced systemic toxicity. While mitochondrial targeting may enhance local copper delivery, it carries high risks of off-target toxicity to normal cells, which also contain mitochondria. Thus, cell-surface targeting (AS1411) is more clinically feasible and safer than mitochondrial targeting, with our in vivo imaging confirming significantly enhanced tumor accumulation of PEMA.The relevant targeted research data is shown in Fig. 6.

Nanoparticle-mediated intracellular copper ion transport and MPC1 gene transfection

To evaluate the copper ion transport capacity and targeting ability of PEMA, RhB was employed as a copper ion probe to quantitatively assess the copper ion content in different experimental treatment groups (Fig. 3a). After the same treatment duration, the cell group treated solely with CuCl2 (1 μM) exhibited relatively low RhB red fluorescence. Even when co-administered with Elesclomol (5 nM) and CuCl2 (1 μM), the red fluorescence intensity remained low. However, when nanoparticles PSP were utilized for copper ion delivery, the red fluorescence intensity significantly increased. Notably, the cell group treated with PEMA nanoparticles modified with nucleic acid aptamers demonstrated the strongest RhB fluorescence intensity, indicating that PEMA nanoparticles possess superior copper ion cellular transport and targeting capabilities. The copper ion content of all groups was determined by fluorescence spectroscopy, and the drug loading rate of PEMA was 8.7% (Fig. S7).Fig. 3| Nanoparticle-mediated intracellular copper ion transport and MPC1 gene transfection. (a) CLSM images of RhB in ACHN cells after different treatments. Scale bars: 100 μm. The concentrations of the probes were 10 μM. (b) The Cu concentration in ACHN cells after ingestion of Elesclomol-Cu, PSP (Loading Elesclomol-Cu) and PEMA for 12 h. Data were presented as the means ± SD (n = 3 independent experiments). (c-d) MPC1, LDHA, LDHB and TOM 20 protein levels after treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel.

| Nanoparticle-mediated intracellular copper ion transport and MPC1 gene transfection. (a) CLSM images of RhB in ACHN cells after different treatments. Scale bars: 100 μm. The concentrations of the probes were 10 μM. (b) The Cu concentration in ACHN cells after ingestion of Elesclomol-Cu, PSP (Loading Elesclomol-Cu) and PEMA for 12 h. Data were presented as the means ± SD (n = 3 independent experiments). (c-d) MPC1, LDHA, LDHB and TOM 20 protein levels after treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel.

For quantitative analysis of intracellular copper ion levels, inductively coupled plasma (ICP) spectrometry was employed. The experimental results, as depicted in Fig. 3b, revealed that compared to the PBS control group, the intracellular copper ion content in cells treated with a low concentration of Elesclomol/CuCl2 (100 nM) was 11.6 ng per million cells. When PSP was used for copper ion delivery, the copper ion content reached 91 ng per million cells, whereas PEMA delivery resulted in a copper ion level of 132 ng per million cells. The ICP results were consistent with those obtained from confocal laser scanning microscopy [51].

Furthermore, to assess the plasmid transfection capability of PEMA, WB was performed to detect the protein expression levels of MPC1, LDHA, LDHB, and TOM 20 in ACHN cells before and after PEMA transfection. The experimental results, as shown in Fig. 3c and d, indicated that the protein expression of MPC1 increased in ACHN cells transfected with an MPC1 overexpression plasmid using PEMA, while the expression of LDHA and LDHB decreased. This demonstrated successful transfection of the MPC1 plasmid. The reduction in LDHA and LDHB expression also indirectly suggested that upon MPC1 overexpression, the cellular lactate metabolism level decreased.

PSP nanoparticles Induce Oxidative Stress in Tumor Cells to Enhance Cuproptosis

Based on our preliminary findings, we observed that intracellular GSH levels play a crucial role in modulating cuproptosis. Elevated GSH in tumor cells chelates intracellular copper ions, thereby attenuating cuproptosis. To address this, we incorporated disulfide bonds into our nanoparticle design to deplete GSH and selected Elesclomol as a copper ion transporter. Elesclomol specifically binds to and inhibits ferredoxin 1 (FDX1), disrupting its-mediated iron-sulfur cluster (Fe-S cluster) biosynthesis. This leads to mitochondrial dysfunction, increased ROS production, and exacerbated oxidative stress. Concurrently, the disulfide bonds in the nanoparticles cause GSH depletion, further disrupting redox homeostasis and amplifying cellular oxidative stress. Consequently, our copper ion delivery system, PSP nanoparticles, effectively induces oxidative stress in tumor cells, enhancing their sensitivity to copper ions and promoting cuproptosis.

To evaluate the oxidative stress-inducing effects of PSP nanoparticles and PSP-loaded Elesclomol on ACHN cells, we employed confocal laser scanning microscopy to measure intracellular GSH levels and ROS production (Fig. 4a). Both standalone PSP nanoparticles and Elesclomol significantly depleted GSH in ACHN cells, comparable to high-concentration hydrogen peroxide (H2O2). The group treated with PSP-loaded Elesclomol exhibited the most pronounced GSH depletion. Similarly, ROS detection (Fig. 4b) revealed that both PSP and Elesclomol markedly increased intracellular ROS levels, with the PSP-loaded Elesclomol group showing the most significant elevation. These findings demonstrate that the disulfide bonds in PSP nanoparticles and the loaded Elesclomol effectively deplete GSH and elevate ROS levels, thereby inducing oxidative stress in tumor cells.Fig. 4| PSP nanoparticles Induce Oxidative Stress in Tumor Cells to Enhance Cuproptosis. (a) CLSM images of GSH in ACHN cells after different treatments. Scale bars: 100 μm. The concentrations of the probes were 10 μM. (b) CLSM images of ACHN cells that were stained with DCFH-DA following different treatments. (c) Quantitation of relative viability of ACHN cells 24 h after treatment with indicated concentrations of H2O2 in the presence of indicated concentrations of CuCl2 in the media. Mean of at least three replicates is color-coded, Color scale represents relative cell viability (%) (blue:high viability, white:low viability). (d) Quantitation of relative viability of ACHN cells 24 h after treatment with/without 500 μM H2O2 in the presence of indicated concentrations of CuCl2 in the media. Data were presented as the means ± SD (n = 3 independent experiments). (e) FDX1 and DLAT oligomerization protein levels after treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel. (i) PBS, (ii) 5 μM CuCl2, (iii) 500 μM H2O2 + 5 μM CuCl2, (iv) 5 nM Elesclomol + 5 μM CuCl2, (Ⅴ) PSP NPs Loading 5 nM Elesclomol + 5 μM CuCl2. (f) Flow cytometry analysis of ACHN cells treated with different samples. (g) Immunofluorescence analysis of the expression of DLAT and DLAT oligomerization in ACHN cells after different treatments. Scale bars: 10 μM.Corresponding statistical analysis results are summarized in the Supporting Information. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

| PSP nanoparticles Induce Oxidative Stress in Tumor Cells to Enhance Cuproptosis. (a) CLSM images of GSH in ACHN cells after different treatments. Scale bars: 100 μm. The concentrations of the probes were 10 μM. (b) CLSM images of ACHN cells that were stained with DCFH-DA following different treatments. (c) Quantitation of relative viability of ACHN cells 24 h after treatment with indicated concentrations of H2O2 in the presence of indicated concentrations of CuCl2 in the media. Mean of at least three replicates is color-coded, Color scale represents relative cell viability (%) (blue:high viability, white:low viability). (d) Quantitation of relative viability of ACHN cells 24 h after treatment with/without 500 μM H2O2 in the presence of indicated concentrations of CuCl2 in the media. Data were presented as the means ± SD (n = 3 independent experiments). (e) FDX1 and DLAT oligomerization protein levels after treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel. (i) PBS, (ii) 5 μM CuCl2, (iii) 500 μM H2O2 + 5 μM CuCl2, (iv) 5 nM Elesclomol + 5 μM CuCl2, (Ⅴ) PSP NPs Loading 5 nM Elesclomol + 5 μM CuCl2. (f) Flow cytometry analysis of ACHN cells treated with different samples. (g) Immunofluorescence analysis of the expression of DLAT and DLAT oligomerization in ACHN cells after different treatments. Scale bars: 10 μM.Corresponding statistical analysis results are summarized in the Supporting Information. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

To separate the GSH-depleting function from the delivery effect of nanoparticles, we constructed a control nanoparticle PP with identical composition but without disulfide bonds. Laser confocal detection of intracellular GSH revealed that PP treatment caused negligible GSH consumption, confirming that the prominent GSH depletion triggered by PSP is specifically attributed to disulfide bonds.

To further investigate the impact of oxidative stress on cuproptosis, we designed an orthogonal model using H2O2 and copper ions. As shown in Fig. 4c, increasing H2O2 concentrations in the presence of fixed copper ion doses (e.g., 20 μM) led to a sharp decline in cell viability. Notably, at H2O2 concentrations below 500 μM (which minimally affect cell viability), copper ion treatment significantly reduced cell survival. This suggests that oxidative stress enhances cuproptosis sensitivity. Using Elesclomol as a copper ion transporter, we achieved comparable cell killing effects with low-dose CuCl2 and 500 μM H2O2. Flow cytometry analysis (Fig. 4f) confirmed that 500 μM H2O2 and low-dose CuCl2 (5 μM) alone had negligible effects on cell viability, whereas co-treatment with H2O2 and Elesclomol-loaded PSP nanoparticles reduced cell viability by 73%.

To clarify whether oxidative stress enhances cuproptosis rather than simply exerting additive cytotoxicity, we examined key cuproptosis-related proteins (Fig. 4e). Copper ion treatment following H2O2 exposure resulted in significant downregulation of FDX1 and upregulation of DLAT oligomers, indicating that oxidative stress sensitizes tumor cells to cuproptosis. Immunofluorescence analysis of DLAT in ACHN cells (Fig. 4g) revealed that in the PBS and low-dose CuCl2 groups, DLAT protein exhibited dispersed and punctate mitochondrial distribution. In contrast, Elesclomol and PSP-loaded Elesclomol treatment induced DLAT aggregation, with fluorescence appearing as dense clusters. This confirms that PSP-loaded Elesclomol effectively promotes cuproptosis by enhancing oxidative stress.

To further dissect the specific cell death modality triggered by PEMA, we performed rescue assays using ferroptosis inhibitor Ferrostatin-1, apoptosis inhibitor Z-VAD-FMK and necroptosis inhibitor. CCK-8 results showed that Ferrostatin-1 significantly restored cell viability after PEMA treatment, while Z-VAD-FMK and necroptosis inhibitor barely reversed PEMA-induced cell loss. In line with this observation, DLAT oligomer accumulation induced by PEMA was markedly reduced by Ferrostatin-1 rather than the other two inhibitors. These data suggest that ferroptosis partially contributes to PEMA-mediated cytotoxicity, whereas apoptosis and necroptosis play minor roles. Importantly, incomplete rescue of DLAT oligomerization by Ferrostatin-1 verified that cuproptosis is the dominant cell death pathway initiated by PEMA, with ferroptosis as a secondary consequence of cuproptosis-elicited mitochondrial injury and oxidative stress (Fig. S8).

Mechanistically, PEMA-induced ROS plays a dual role in synergizing cuproptosis. First, ROS depletes intracellular GSH, relieving GSH-mediated chelation of Cu2+ and de-repressing cuproptosis. Second, ROS (predominantly mitochondrial superoxide and H2O2) directly promotes lipoylated protein oxidation and shifts the Cu2+/Cu+ redox equilibrium toward Cu+, which accelerates DLAT oligomerization and FDX1 downregulation. Thus, ROS amplifies cuproptosis through both GSH depletion and direct modulation of cuproptosis-related molecular events, supporting the synergistic mechanism of our platform [[52], [53], [54], [55]].

The therapeutic effects of PEMA on tumor cells

To evaluate the in vitro antitumor efficacy of PEMA nanoparticles, we employed live/dead cell staining and flow cytometry analyses. As depicted in ‌Fig. 5a‌, treatment with the MPC1-overexpressing plasmid alone had minimal impact on cell viability. In contrast, PSP nanoparticles exhibited limited cytotoxicity, while Elesclomol/CuCl2 treatment alone induced significant tumor cell damage, evidenced by enhanced red fluorescence. Notably, co-treatment with PSP@Elesclomol@CuCl2 or PEMA nanoparticles further augmented tumor cell killing, demonstrating the potent antitumor activity of PEMA nanoparticles. Flow cytometry and CCK8 (Fig. S9) quantified these effects (‌Fig. 5b‌), revealing a tumor cell killing rate of 69% for PSP@Elesclomol@CuCl2 and 99% for PEMA nanoparticles. This enhancement was attributed to two synergistic mechanisms: (1) MPC1 overexpression increased tumor cell sensitivity to copper ions, amplifying copper toxicity; and (2) the targeted design of the nanoparticles facilitated their efficient uptake, further augmenting cuproptosis.Fig. 5| Therapeutic effects of the PEMA at the cellular level. (a) Live/dead staining images, and (b) Flow cytometry analysis of ACHN cells treated with different samples. (c) Immunofluorescence analysis of the expression of DLAT and DLAT oligomerization in ACHN cells after different treatments. Scale bars: 10 μM. (d) Representative Bio-TEM images of ACHN cells before and after treatment with PEMA. (e) After treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel.Corresponding statistical analysis results are summarized in the Supporting Information.Fig. 6| In vivo imaging and anti-tumor effect of PEMA. (a) Schematic depiction of the treatment plan in the ACHN tumor model mediated by PEMA. (b-c) In vivo intravital imaging of tumor-bearing nude mice after intravenous administration of various formulations, as well as ex vivo imaging of important organs and tumors. (d) At various time points, quantitative examination of in vivo fluorescence pictures and ex vivo fluorescence images. (e) Relative tumor volume (v/v0) versus treatment time. (f) Photographs of tumors (n = 5). (g) Mouse body weight versus treatment time. (h) Images demonstrate Ki-67 and H&E staining of tumors collected of different therapies.

| Therapeutic effects of the PEMA at the cellular level. (a) Live/dead staining images, and (b) Flow cytometry analysis of ACHN cells treated with different samples. (c) Immunofluorescence analysis of the expression of DLAT and DLAT oligomerization in ACHN cells after different treatments. Scale bars: 10 μM. (d) Representative Bio-TEM images of ACHN cells before and after treatment with PEMA. (e) After treatment of ACHN cells with different samples. Images are representative of three independent experimental replicates. The samples of the same group were derived from the same experiment, and the gels/blots were processed in parallel.Corresponding statistical analysis results are summarized in the Supporting Information.

| In vivo imaging and anti-tumor effect of PEMA. (a) Schematic depiction of the treatment plan in the ACHN tumor model mediated by PEMA. (b-c) In vivo intravital imaging of tumor-bearing nude mice after intravenous administration of various formulations, as well as ex vivo imaging of important organs and tumors. (d) At various time points, quantitative examination of in vivo fluorescence pictures and ex vivo fluorescence images. (e) Relative tumor volume (v/v0) versus treatment time. (f) Photographs of tumors (n = 5). (g) Mouse body weight versus treatment time. (h) Images demonstrate Ki-67 and H&E staining of tumors collected of different therapies.

Subsequently, immunofluorescence staining was performed to assess cuproptosis. ‌Fig. 5c‌ shows that MPC1 overexpression in ACHN cells increased DLAT expression, predominantly in punctate and diffuse patterns, without aggregation. In contrast, PSP@Elesclomol@CuCl2 treatment induced DLAT aggregation, confirming cuproptosis initiation. PEMA nanoparticles further enhanced DLAT fluorescence, with prominent block-like and aggregate formations, indicating intensified cuproptosis.

TEM analysis of ACHN cells treated with PEMA nanoparticles revealed profound mitochondrial structural alterations (‌Fig. 5d‌). Untreated control cells exhibited intact mitochondrial cristae and double-membrane structures. In PEMA-treated cells, however, mitochondrial cristae were disrupted, and the double-membrane integrity was compromised, confirming cuproptosis induction [56,57].

WB experiments validated PEMA-induced cuproptosis in ACHN cells (‌Fig. 5e‌). HSP70 expression, a biomarker of both oxidative stress and cuproptosis, was significantly elevated in response to PEMA treatment. This upregulation occurred via two pathways: (1) direct oxidative stress from nanoparticle exposure, which triggered HSP70 expression to repair damaged proteins; and (2) indirect activation due to mitochondrial metabolic disruption and iron dysregulation during cuproptosis. Notably, PSP and Elesclomol co-treatment also induced HSP70 upregulation, albeit to a lesser extent than PEMA.

Further analysis of FDX1 and DLAT protein expression revealed mechanistic insights into cuproptosis. FDX1, a core regulator of cuproptosis, facilitates Cu2+ reduction to Cu+, thereby mitigating mitochondrial Fe-S cluster protein loss and lipoylated protein aggregation [3]. ‌Fig. 5e‌ demonstrates that PEMA treatment significantly reduced FDX1 expression, exacerbating cuproptosis. Concurrently, Cu+ binding to DLAT promoted its oligomerization, increasing insoluble DLAT aggregates while reducing soluble DLAT monomers. This process disrupted the TCA cycle and mitochondrial function. PEMA-treated ACHN cells exhibited the most pronounced DLAT oligomerization and the lowest FDX1 levels, confirming the exacerbation of cuproptosis.

Collectively, these findings demonstrate that PEMA nanoparticles enhance ACHN cell sensitivity to copper ions, thereby intensifying copper-mediated cuproptosis. This dual action—targeted nanoparticle delivery and copper ion amplification—offers a promising strategy for antitumor therapy.

In vivo imaging and anti-tumor effect of PEMA

To evaluate the active targeting capacity of nanoparticles in vivo, ACHN cells were subcutaneously injected into BALB/c nude mice. When the tumors reached a volume of 100 mm3, the animals were randomly allocated into three groups: Cy5.5, PSP-Cy5.5, and PSP-Cy5.5-Apt (with equivalent Cy5.5 doses across all groups). As illustrated in Fig. 6b–d, the small molecular weight and non-targeting nature of Cy5.5, coupled with its rapid metabolism, resulted in a short retention time in circulation. Following tail vein administration, Cy5.5 predominantly accumulated in the liver and kidneys, with significant elimination occurring within 4 h, leading to a marked decline in systemic fluorescence intensity. In contrast, the Cy5.5-loaded PSP nanoparticles demonstrated prolonged circulation due to their enhanced permeability and retention (EPR) effect, enabling passive tumor accumulation. Consequently, PSP-Cy5.5 exhibited moderate fluorescence signals in tumor tissues even after 48 h. The PSP-Cy5.5-Apt group, further functionalized with tumor-specific aptamers, exhibited superior tumor targeting and retention properties, with fluorescence intensity peaking at 24 h post-injection.

Although Cy5.5-labeled PEMA nanoparticles exhibited robust tumor-targeting accumulation in vivo, the current study did not quantitatively analyze the temporal accumulation of active payloads (Elesclomol-Cu complex and MPC1 plasmid) in tumor tissues. Further studies using high-performance liquid chromatography (HPLC) for Elesclomol-Cu quantification and quantitative polymerase chain reaction (qPCR) for MPC1 plasmid detection are warranted to directly validate the delivery efficiency of the active components. Additionally, long-term in vivo efficacy and immunocompatibility will be evaluated in future work to facilitate clinical translation.

Subsequent ex vivo imaging of tumors and major organs corroborated these findings. The Cy5.5 group showed strong fluorescence in the liver and kidneys but negligible signals in tumors or other organs. Both PSP-Cy5.5 and PSP-Cy5.5-Apt groups displayed detectable fluorescence in tumors in addition to hepatic and renal accumulation, with the latter group demonstrating significantly higher tumor fluorescence due to its active targeting mechanism. These ex vivo results were consistent with in vivo imaging data, validating the efficacy of the PSP-Cy5.5-Apt targeting strategy. The exceptional tumor-targeting and accumulation properties of PSP-Cy5.5-Apt prompted further investigation into its therapeutic potential.

As outlined in Fig. 6a, ACHN tumor-bearing mice were randomly assigned to five treatment groups: saline control, PSP@MPC1, PSP@Elesclomol, PSP@Elesclomol@MPC1, and PEMA. Antitumor efficacy was assessed by monitoring relative tumor volume. The saline control, PSP@MPC1, and PSP@Elesclomol groups exhibited significant tumor progression, with negligible tumor suppression observed. Although Elesclomol is a potent inducer of cuproptosis, its clinical application is limited by poor solubility, systemic toxicity, and lack of targeting specificity. While PSP encapsulation improved drug delivery, the PSP@Elesclomol group showed minimal tumor accumulation, resulting in modest therapeutic effects. The PSP@Elesclomol@MPC1 group demonstrated partial tumor suppression, but the absence of active targeting compromised its efficacy. In stark contrast, the PEMA group exhibited profound tumor regression (Fig. 6e), with corresponding in vivo imaging confirming these findings (Fig. 6f). Importantly, no significant body weight loss was observed in any treatment group during the study period, indicating negligible systemic toxicity of PEMA (Fig. 6g), hotographs of mice bearing tumors and Statistics of tumor are included in Fig. S10–S11.

Histopathological analysis revealed characteristic nuclear condensation and cell shrinkage in PEMA-treated tumors (Fig. 6h), while Ki-67 staining confirmed the group's lowest proliferative activity among all treatment cohorts. Collectively, these findings demonstrate that the PEMA platform, combining MPC1 overexpression with PSP-encapsulated Elesclomol for cuproptosis induction, represents a paradigm-shifting therapeutic strategy with clinically relevant tumor suppression capabilities. This work establishes a foundation for advancing cuproptosis-based synergistic therapies in oncology.

Compared with recently reported cuproptosis-inducing nanoplatforms, our PEMA exhibits distinct advantages in both efficacy and mechanism. Most existing platforms rely solely on copper delivery or GSH depletion, achieving tumor inhibition rates of 50–70% with relatively high copper doses (≥10 μM) doi.org/10.1002/advs.202402178 doi.org/10.1002/adfm.202403651. In contrast, PEMA integrates metabolic reprogramming (MPC1 overexpression) and oxidative stress amplification (GSH depletion + ROS burst), achieving ∼78.6% tumor regression in vivo with a low copper dose (∼5 μM). This dual synergistic strategy not only reduces systemic copper toxicity but also overcomes tumor intrinsic resistance, representing a novel paradigm for cuproptosis-based therapy.

For clinical translation, PEMA faces two foreseeable challenges: (1) GMP-scale production: The multi-component assembly (PLGA scaffold, disulfide bonds, aptamer, plasmid, Elesclomol) requires optimized synthesis protocols to ensure batch-to-batch consistency and stability; (2) regulatory approval: As a gene-drug co-delivery system, additional safety assessments (e.g., plasmid integration risk, long-term immunogenicity) are needed. Despite these hurdles, our biosafety data (no hemolysis, normal blood parameters, no organ toxicity) provides a solid foundation. Future work will focus on simplifying the formulation and evaluating efficacy in patient-derived xenograft (PDX) models to accelerate clinical translation.

In addition, this study still has several inherent limitations worthy of further improvement. Although the optimized PEMA nanoplatform achieves a satisfactory tumor inhibition rate of 78.6% in vivo and performs potent anti-tumor therapeutic efficacy, we fail to monitor the long-term tumor recurrence rate after treatment in current animal experiments. Accumulating evidence has demonstrated that tumor recurrence is tightly correlated with the immunosuppressive tumor microenvironment and insufficient systemic anti-tumor immune memory triggered by monotherapy. Given the inherent immunogenicity of cuproptosis, combining cuproptosis-based tumor therapy with immunomodulatory strategies to inhibit tumor progression and prevent postoperative recurrence is considered a promising research direction. Therefore, our follow-up research will further focus on exploring the crosstalk between cuproptosis and anti-tumor immunity, as well as evaluating the long-term tumor recurrence inhibition effect of our nanotherapeutic system, so as to further optimize the overall anti-tumor performance of PEMA nanoplatforms [17,58].

Biosafety evaluation

To further assess the biosafety of the PEMA nanomaterial, hemocompatibility evaluation was conducted via hemolysis assays. No hemolytic activity was observed at all tested concentrations, confirming the excellent blood compatibility of PEMA (Fig. 7a and b). Concurrently, the long-term storage stability of PEMA nanoparticles was investigated over a 30-day period. Dynamic light scattering analysis revealed that PEMA nanoparticles maintained consistent particle size in phosphate-buffered saline (PBS), whereas those suspended in ultrapure water exhibited significant aggregation and size increase due to charge instability (Fig. 7c). These findings demonstrate superior stability of PEMA in PBS solutions during extended storage.Fig. 7| Biosafety Evaluation. (a-b) Hemolysis assay with different concentrations of PEMA NPs. (c) The 30-day storage stability evaluation of PEMA NPs. (d) Mouse hematological index on the following various treatments. (e) Images of H&E staining of major tissues after intravenously injected with different therapies. Data were presented as the means ± SD (n = 3 independent experiments).

| Biosafety Evaluation. (a-b) Hemolysis assay with different concentrations of PEMA NPs. (c) The 30-day storage stability evaluation of PEMA NPs. (d) Mouse hematological index on the following various treatments. (e) Images of H&E staining of major tissues after intravenously injected with different therapies. Data were presented as the means ± SD (n = 3 independent experiments).

To comprehensively evaluate the potential systemic toxicity of PEMA, key hepatic (alanine aminotransferase, ALT; aspartate aminotransferase, AST) and renal (blood urea nitrogen, BUN; creatinine, CRE) function markers were analyzed after treatment (Fig. 7d). The results showed that all biochemical parameters in the PEMA-treated groups remained within normal physiological ranges and showed no statistically significant difference compared to the PBS control group. Specifically, the levels of ALT, AST, BUN, and CRE were comparable to those of healthy mice, indicating that PEMA administration at the therapeutic dose did not induce detectable liver or kidney injury.

Comprehensive in vivo biocompatibility assessment was performed through analysis of blood biochemical parameters and hematological profiles. All measured indices remained within normal physiological ranges in PEMA-treated mice (Fig. 7d), indicating negligible systemic toxicity. Histopathological examination of major organs (heart, liver, spleen, lung, and kidney) using hematoxylin and eosin (H&E) staining revealed no morphological abnormalities (Fig. 7e), further corroborating the excellent biocompatibility of PEMA. Collectively, these results establish the favorable safety profile of PEMA for biomedical applications.

Conclusion

In this study, we integrated bioinformatic analysis with experimental validation to elucidate the pivotal role of the ‌MPC1 gene‌ in enhancing tumor cell sensitivity to cuproptosis by orchestrating energy metabolic reprogramming. Building upon this mechanistic insight, we rationally engineered a ‌targeted nanocarrier‌ co-loaded with an MPC1-overexpressing plasmid and the cuproptosis inducing drug ‌Elesclomol‌. This dual-functional design synergistically amplified cuproptosis efficacy through two complementary pathways: (1) ‌metabolic rewiring‌ via MPC1-mediated mitochondrial pyruvate uptake suppression, and (2) ‌oxidative stress amplification‌ through Elesclomol/Cu2+-induced mitochondrial dysfunction. To achieve tumor-specific delivery, we incorporated the ‌AS1411 aptamer, a nucleic acid ligand with high affinity for nucleolin-overexpressing tumor cells, thereby significantly enhancing intratumoral accumulation of the nanocarrier. Comprehensive in vitro and in vivo experiments demonstrated the exceptional antitumor potency‌ of PEMA nanoparticles. In vitro assays revealed a ‌>99% tumor cell killing rate, accompanied by robust indicators of cuproptosis, including ‌DLAT oligomerization‌, ‌FDX1 downregulation, and mitochondrial structural disintegration. In vivo studies further corroborated these findings, showing tumor regression in xenograft models without detectable systemic toxicity. Importantly, biosafety evaluations confirmed the biocompatibility and biodegradability‌ of PEMA. Collectively, this study establishes a ‌therapeutic paradigm that synergizes metabolic targeting with cuproptosis induction, offering a ‌translational nanoplatform for potent and safe antitumor therapy.

Materials and methods

Materials

Poly(D,L-lactide-co-glycolide) (PLGA, Mw 7000-17000, 18000-28000, 24000-38000, 38000-54000), Di(N-succinimidyl) 3,3′-Dithiodipropionate (DSP, 97%), Ethylenediamine Monohydrate (98%), N-(3-Dimethylaminopropyl)- N′-ethylcarbodiimide hydrochloride (EDC, 98%), N-fydroxysuccinimide (NHS, 98%), Acetone (96%),RIPA lysis buffer, 100 mM PMSF, 50×cocktail protease inhibitors and phosphatase inhibitors were purchased from servicebio Bio Technology Co., Ltd (Wuhan, China). 2,7-dichlorofluorescein diacetate (DCFH-DA) was purchased from Abcam. Dulbecco's modified Eagle's medium (DMEM) medium was purchased from Thermo Fisher Scientific. The Cell Counting Kit-8 (CCK-8), Cell Tracking Kit, GSH-Tracker NIR and Hieff Trans Liposomal Transfection Reagent were purchased from Yeasen, China. Hoechst 33342 and liperfluo were purchased from DOJINDO.

Bioinformatics analysis

Bioinformatics analysis was performed using the TCGA database. The expression levels of PDH and SLC31A1 in ACHN, A549, OVISE, and MDA-MB-231 cell lines were extracted. The correlation between gene expression and aerobic metabolism was analyzed and visualized using R software (version 4.2.1) and the ggplot2 package.

Preparation of MPC1 overexpression plasmid

MPC1 overexpression (OE) plasmid were designed and constructed by MiaoLingPlasmid (Wuhan) Co.,Ltd (https://www.miaolingbio.com). Transient transfection was performed using Lipofectamine 3000.The MPC1 overexpression sequence was derived from the NCBI reference transcript NM_016098 (GenBank accession: NM_016098).

Synthesis of PLGA-NH2

PLGA were dissolved in acetone, to functionalize PLGA with amino groups, ethylenediamine (EDA) was introduced as the grafting agent in the presence of the crosslinking catalyst EDC-NHS. The reaction was carried out at room temperature for 2 h, with EDA added at 2% (w/w) relative to the mass of PLGA. The resulting mixture was then slowly dripped into ultrapure water under gentle stirring to facilitate self-assembly of the PLGA-NH2 nanoparticles. Excess EDA and the catalyst were removed by centrifugation at room temperature. The purified PLGA-NH2 nanoparticles were redissolved in acetone to obtain a homogeneous solution for subsequent experiments.

Synthesis of PSP NPs

To prepare PSP nanoparticles, the crosslinking agent DSP was added to the previously obtained PLGA-NH2/acetone solution at a concentration of 1% (w/w) relative to the mass of PLGA-NH2. The mixture was stirred at room temperature for 4 h to facilitate the crosslinking reaction. Subsequently, the resulting solution was slowly dripped into ultrapure water under gentle stirring to induce self-assembly of the nanoparticles. The formed PSP nanoparticles were then purified by centrifugation (3 cycles) to remove unreacted reagents and byproducts, yielding a stable colloidal suspension for further applications.

Synthesis of PEMA NPs

To prepare PEMA nanoparticles, the previously obtained PSP nanoparticles were redissolved in DMSO. Subsequently, Elesclomol was incorporated into the solution. The resulting mixture was then slowly dripped into PBS solution containing the MPC1 overexpression plasmid and Aptamer. The DMSO concentration in the final mixture was carefully controlled to ≤10% (v/v) to prevent degradation of the MPC1 plasmid and Aptamer. The assembled nanoparticles were purified by centrifugation (3 cycles) to remove unreacted components, yielding a stable suspension of PEMA nanoparticles dispersed in PBS buffer.

Characterization

The morphology, size, and microstructure of the samples were analyzed using SEM and transmission electron microscopy (TEM, Leica EM TP, Germany) with an electron emission gun operating voltage of 300 kV. An infrared spectrometer was used to produce the Fourier transform infrared spectra (Thermo Fisher Scientific, Inc., Nicolet iN10 MX spectrometer). Dynamic light scattering was used to identify the polydispersity index and average hydrodynamic particle size (DLS, Brookhaven Instruments-Omni). The ultraviolet absorption visible light spectrum (UV–vis) of the materials was measured with a Cary 5000 UV–vis-NIR spectrophotometer in order to obtain the light absorption peaks in the NIR region. Zeta potential detection using (Malvern Zetasizer Nano ZS90, UK).

Cell culture

ACHN, A549, OVISE, MDA-MB-231 and HUVEC cells were sourced from the Shanghai Instituteof Cell Science, China. Both cell lines were cultured in Dulbecco'smodified Eagle's medium (Gibco) supplemented with 10% fetalbovine serum (Gibco) and 1% penicillin-streptomycin (Invitrogen)in a humidified incubator at 37°C with 5% CO2.

In vitro cytotoxicity evaluation

ACHN, A549, OVISE and MDA-MB-231 cells were grown in high glucose DMEM with 1% penicillin–streptomycin and 10% FBS. The CCK-8 test was used to assess cell viability in accordance with the manufacturer's instructions (Yeason Biotechnology, China). Basically, All types of cells were seeded overnight at a density of 5 × 103 cells per well in a 96-well plate. Afterwards, various medications were administered to the various groups. Using SpectraMax Microplate Reader (model 680, Bio-Rad Laboratories, Inc., Tokyo, Japan), the optical density (OD) values of each well were measured at 450 nm, and the cell viability in each group was estimated based on these values.

ACHN cells were seeded into confocal dishes and incubated overnight. The cells were subsequently exposed to different treatments, including PBS, MPC1 Plasmid, PSP NPs, Elesclomol (100 nM), PSP NPs (loading 100 nM Elesclomol) and PEMA NPs (loading 100 nM Elesclomol). Live cells were stained with Calcein-AM [excitation/emission (Ex/Em) = 488/530 nm], while dead cells were labeled with PI (Ex/Em = 535/617 nm). After a 30-min co-incubation, the cells were washed twice with PBS and visualized using CLSM.

For quantitative assessment of the apoptosis rate, cells were seeded into six-well plates. After various treatments, the cells were stained with Annexin-FITC/PI for 20 min, detached using 0.05% trypsin-EDTA solution, resuspended in PBS, and immediately analyzed via flow cytometry.

Western blotting analysis

Using SDS–PAGE, cell lysates were separated and electrophoretically deposited onto polyvinylidene difluoride membranes (Bio-Rad). After blocking with 5% skim milk in Tris-buffered saline (TBS), primary antibodies were treated overnight at 4°C with membranes. The following day, membranes were cleaned three times in TBS solution, and incubated the secondary antibody for 45 min in a temperature of the room. The membranes were rinsed five times for 5 min in TBS buffer with 0.1% Tween-20, and then the ChemDoc Imaging equipment was used to acquire images (Bio-Rad).

Immunofluorescence staining

On 12well culture plates, ACHN cells were cultivated until about 80% fusion. Thereafter, pharmaceuticals were given to various groups. Then, apply 4% paraformaldehyde over night as a fixative. The cells were washed 3 times in 0.01 M phosphate buffer saline for 10 min each time, and then the membrane was penetrated on ice containing 0.2% Triton X-100 for 20 min. Then it was sealed with 2% bovine serum albumin at room temperature for 1 h. Then wash with PBS for 3 times, 10 min each time. Then the rabbit anti-DLAT (1:200; Abcam) was incubated at 4°C overnight. The next day, wash with 0.01 M PBS three times, and incubate with cy3 donkey anti-rabbit antibody (1:1000; Abcam) at room temperature for 1 h after 10 min each time. Then clean it again with PBS for 3 times. Finally, the cells were fixed in the medium containing 4,6-diamino-2-phenylindole (DAPI) (Abcam), and three-dimensional reconstruction observation was performed with CLSM.

Immunohistochemical staining

Tumor tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sliced into 5 μm sections. After deparaffinization and antigen retrieval, sections were blocked with 5% BSA for 1 h, then incubated with anti-Ki-67 primary antibody (1:200, Abcam) overnight at 4°C. After washing, sections were incubated with HRP-conjugated secondary antibody for 1 h at room temperature, stained with DAB, and counterstained with hematoxylin. Images were captured using a digital pathology scanner.

GSH depleting and oxidative stress with PSP NPs

On 35 mm confocal dishes, ACHN cells were grown to approximately 80% fusion. Thereafter, medicines were introduced to various groups. GSH Tracer was employed according to the instructions to detect the GSH content in cells, resulting in blue fluorescence. TRITC-phalloidin was employed to label the actin cytoskeleton for cell tracing, with the results exhibiting red fluorescence.

ACHN cells were seeded into confocal dishes and incubated over-night. The cells were then subjected to various treatments, including PBS, PSP NPs, Elesclomol (100 nM) and PEMA NPs. After washing with PBS, the cells were incubated with DCFH-DA/Hoechst 33342 staining solution in the dark for 30 min at 37°C, then fluorescence intensity was monitored using CLSM.

In vivo targeting and imaging

Six-to-eight-week-old BALB/c nude female mice were obtained from the Experimental Animal Center of Xiamen University. The model of subcutaneous tumor was produced by injecting 100 L (1 × 106 cells) of ACHN cells into nude mouse. After the tumor volume reached 100 mm3, 100 μL of fluorescently tagged (Cy5.5) PEMA NPs (20 mg kg−1) was injected into the tail vein of tumor-bearing mice. After 48 h, the mice were euthanized and the tumors and main organs were taken for fluorescence imaging ex vivo.

Targeted metabolomics analysis

Samples were thawed at 4°C. For each sample, 1000 μL of pre-chilled 80% methanol aqueous solution was added, followed by vortex mixing and ice-bath sonication for 20 min. The mixture was then incubated at −20°C for 1 h and centrifuged at 16,000 × g for 20 min at 4°C to obtain the supernatant. The protein precipitate was air-dried in a fume hood, resuspended in 600 μL of SDT buffer, and diluted 2-fold for BCA protein quantification. Based on the protein content, the corresponding cell number was calculated, and equal cell equivalents of supernatant from each sample were collected. The supernatants were lyophilized in a high-speed vacuum centrifuge. Prior to mass spectrometry analysis, the dried samples were reconstituted in 80 μL of pre-chilled 50% methanol aqueous solution, centrifuged at 20,000 × g for 15 min at 4°C, and the resulting supernatant was used for LC-MS/MS analysis.

In vivo therapeutic effects

A ACHN tumor-bearing mouse model was developed by subcutaneously injecting 1 × 106 ACHN cells suspended in PBS into the right hind leg of healthy female Balb/c nude mice (4 to 6 weeks old). When tumor volumes approached approximately 100 mm3, the mice were randomly assigned into five groups (n = 5): (i) saline, (ii) MPC1 Plasmid, (iii) Elesclomol-Cu, (iv)PSP Loading Elesclomol-Cu, (Ⅴ) PEMA NPs. PEMA NPs was administered at a dose of 10 mg/kg (Other groups maintain the same drug content), with treatments applied three times acrofss a 12-day interval. During the treatment regimen, body weight and tumor volume (V = length × width2/2) were monitored bi-daily. At the conclusion of the treatment period, the mice were euthanized, and tumors along with major organs were collected. Tumors were measured, and both tumor tissues and major organs underwent H&E staining for histo pathological examination. Furthermore, Ki-67 immunos taining were conducted to evaluate cellular apoptosis and proliferation across different treatments.

Hemocompatibility assay

Fresh mouse blood was collected and centrifuged at 3000 rpm for 10 min to isolate erythrocytes. Erythrocytes were washed three times with PBS and diluted to a 2% (v/v) suspension. The suspension was incubated with different concentrations of PEMA NPs (0, 10, 50, 100 μg/mL) at 37°C for 3 h. PBS and 0.1% Triton X-100 were used as negative and positive controls, respectively. After centrifugation, the absorbance of the supernatant at 540 nm was measured, and the hemolysis rate was calculated.

In vivo biosafety assessment

The obtained organs were fixed, and HE was applied to the tissue slices. Blood was drawn from each group, and standard blood analysis was performed. All animal experiments were approved by the Xiamen University Ethics Committees (XMULAC20230200).

Statistical analysis

All values are expressed as mean ± SD, and the significance of the data is based on Student's t-test, one-way analysis of variance (ANOVA), and two-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001).

CRediT authorship contribution statement

Bo Huang: Conceptualization, Data curation, Methodology, Writing – original draft. Weijin Zhang: Data curation, Methodology, Writing – original draft. Yunjie Wang: Data curation, Validation, Writing – review & editing. Xian Luo: Formal analysis, Methodology, Validation, Writing – review & editing. Wenda Wu: Formal analysis, Methodology, Validation, Writing – review & editing. Rong Shen: Methodology, Writing – review & editing. Xiangquan Liu: Methodology, Validation, Writing – review & editing. Zhibo Zhang: Data curation, Formal analysis, Writing – review & editing. Yi Gao: Formal analysis, Supervision, Writing – review & editing. Yingying Wu: Data curation, Validation, Writing – review & editing. Fangwei Zeng: Data curation, Validation, Writing – review & editing. Yuan Huang: Resources, Supervision, Writing – review & editing. Jianyun Yu: Formal analysis, Validation, Writing – review & editing. Suxiao Wang: Funding acquisition, Methodology, Validation, Writing – review & editing. Ting Wu: Formal analysis, Funding acquisition, Validation, Writing – review & editing. Shengyu Wang: Conceptualization, Project administration, Resources, Supervision, Writing – review & editing. Shuitu Feng: Conceptualization, Project administration, Resources, Supervision, Writing – review & editing. Fanghong Luo: Conceptualization, Project administration, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Shengyu Wang reports financial support was provided by Natural Science Foundation of Xiamen City. Ting Wu reports financial support was provided by Natural Science Foundation of Xiamen City. Fanghong Luo reports financial support was provided by Provincial Public Welfare Project of Fujian Province. Rong Shen reports financial support was provided by Provincial Public Welfare Project of Fujian Province. Fangwei Zeng reports financial support was provided by Xiamen Broad Creation Biotechology Co.,Ltd. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.