Controllable nickel ions release from deferoxamine mesylate-triggered nickel-iron layered double hydroxide for eliciting apoptotic cell death in prostate cancer

Abstract

Despite their unique advantages and vast potential, nanomaterials employed in cancer therapy still encounter challenges such as uneven biodistribution, unintended drug leakage, and especially potential tissue damage caused by off-target toxicity. Bioinert nanomaterials, known for their excellent chemical stability, and minimal biological reactivity, can exert localized tumoricidal effects in response to specific external stimuli. However, the lack of precise control or poor penetration depth largely limits the therapeutic efficacy, necessitating the development of innovative stimuli-responsive therapeutic strategies. This study presents an alternative drug-responsive cancer therapeutic approach based on nickel-iron layered double hydroxide (NiFe-LDH), which exhibited negligible toxicity to both normal cells and cancer cells. By conjugating a platelet-derived growth factor receptor (PDGFR)-β-targeting cyclic peptide, NiFe-LDH achieved high specificity for prostate cancer cells, significantly enhancing tumor targeting and accumulation. Upon administration of deferoxamine mesylate (DFOM), an FDA-approved iron chelator, NiFe-LDH transitioned from a “bioinert” state to a “bioactive” nanotherapeutic through structural disassembly and robust release of nickel ions (Ni²⁺). The released ions disrupted mitochondrial function, upregulated insulin-like growth factor binding protein 3 (IGFBP3), and further inhibited the PI3K/AKT/mTOR signaling pathway, consequently leading to potent and selective induction of apoptosis in prostate cancer cells. Unlike conventional therapies, which often cause varying degrees of toxicity in non-target organs, this stimuli-responsive nanoplatform could minimize off-target effects and systemic toxicity by combining the non-toxic LDH with the clinically used DFOM. Our findings demonstrate that DFOM-responsive NiFe-LDH can effectively inhibit tumor growth in both cultured cells and tumor xenografts, suggesting a rational and clinically translatable platform for precision cancer therapy.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03489-6.

Introduction

Malignant tumors are among the most life-threatening diseases and a persistent and critical public health challenge. In addition to their considerable health risks, they impose substantial psychological and economic burdens on families globally [1]. Despite notable advancements in medical science and technology, the cancer-associated mortality remains unavoidable and even high [2]. Therefore, the pursuit of effective therapeutic strategies for cancer therapy remains of great importance. In recent years, the rapid advancement of nanotechnology has enabled the development of a diverse array of nanomaterials, which, owing to their distinctive properties including adjustable particle size, large surface area, and tunable surface characteristics, have attracted widespread and close attention. Compared with conventional therapeutics, nanomaterials can offer significant advantages, including prolonged half-life in circulation, improved biodistribution, and enhanced tumor permeability and retention, increasing the exploration of their applications in biosensing, diagnostic imaging, tissue engineering and drug delivery [3–5]. One of the most prevalent applications of nanomaterials in cancer treatment is as drug carriers, owing to their high loading capacities, precise tumor targeting capabilities, and the flexibility with which their surface properties can be modified to enable responsive drug release [6]. However, the drug-loaded nanocarriers are associated with several limitations, including acquired resistance caused by the drugs themselves, uneven biodistribution, and unintended leakage, all of which may compromise their therapeutic efficacy and cause systemic toxicity [7–9]. Extensive efforts have been devoted to the exploration of stimuli-responsive nanocarriers to enhance the therapeutic efficacy in targeted tumor tissues while minimizing the adverse effects on normal tissues [10–12]. Among them, redox- and acid-responsive nanocarriers have demonstrated superior capabilities in realizing controlled drug release through their response to the redox and pH gradients between the extracellular and intracellular environments. However, the redox potential differences are ubiquitous in vivo, meaning that both tumor and even normal tissues exhibit redox gradients [11]. Moreover, despite the tumor microenvironment (TME) being characterized by a lower pH compared to normal tissues, the lower pH values of pathological or inflammatory sites have also been well-documented in various diseases [12]. As a result, stimuli-responsive nanocarriers may inevitably lead to premature structural collapse and drug release prior to reaching tumor tissues, causing off-target toxicity to some extent [13].

In recent years, bioactive nanomaterials that possess unique properties and functions, have played important roles in cancer treatment. Owing to their biological functions and special enzyme-like activities, bioactive nanomaterials have the potential to induce specific protein degradation in cancer cells, regulate the TME, promote catalytic therapy, and enhance ferroptotic and immunogenic cell death [14, 15]. Although strategies aimed at improving nanomaterial targeting and accumulation efficiency at the intended tumor tissues, such as surface modifications for enhanced active or passive targeting, have been extensively developed, bioactive nanomaterials are inevitably distributed in non-target organs, such as the liver and spleen, ultimately resulting in potential tissue damage [16, 17].

In contrast, bioinert nanomaterials, which exhibit minimal or no intrinsic biological reactivity, can achieve local tumoricidal capabilities by means of externally applied physical stimuli such as heat [18], light [19], ultrasound [20] and magnetic fields [15]. The aforementioned approaches have held promise in cancer therapy. Nevertheless, the complexity of achieving precise control, irreversible damage to tumor-adjacent normal tissues, poor penetration depth, low energy transfer efficiency, insufficient field strength and reliance on complex medical devices as well as the associated inconveniences, have limited their clinical translation and application to some extent [21–24]. Therefore, exploring the alternative stimuli, such as the clinically used drugs, would be promising for the effective activation of the bioinert nanomaterials and consequent tumoricidal effects.

Layered double hydroxides (LDHs) are a class of two-dimensional nanomaterials composed of metal hydroxide layers, that incorporate both divalent and trivalent metal cations, with interlayer anions providing charge compensation. The divalent metal cations typically include magnesium (Mg²⁺), zinc (Zn²⁺), or nickel (Ni²⁺), and the trivalent metal ions mainly consist of aluminum (Al³⁺) or iron (Fe³⁺) [25]. LDHs have emerged as versatile materials that hold great potential in biomedicine. Owing to their unique layered structure, anion-exchange properties and especially excellent biocompatibility, the materials are promising candidates for several therapeutic applications, such as tissue engineering, drug delivery, gene therapy as well as immunotherapy [26, 27]. Among the numerous metal cations, Ni²⁺- and Ni²⁺-based nanomaterials have garnered interest in recent years as they can interact with specific molecular targets and disrupt cellular functions, leading to apoptosis [28, 29]. Meanwhile, Fe³⁺ not only play a significant role in ferroptosis induction, they can be bound by the iron chelators and form stable complexes. For instance, deferoxamine mesylate (DFOM), an FDA-approved and clinically used iron chelator, can effectively bind to Fe³⁺ and is widely used in clinical practice to reduce iron overload and treats acute iron poisoning [30]. Thus, nickel-iron layered double hydroxide (NiFe-LDH), composed of Ni²⁺ and Fe³⁺, may be considered “bioinert” due to its structural stability under physiological conditions. However, when DFOM is employed as a chelator, it may disrupt the stable structure, thereby converting the “bioinert” NiFe-LDH into “bioactive” nanotherapeutics, which can exhibit the desired tumoricidal effects.

In this work, prostate cancer was utilized as the representative cancer type to investigate the therapeutic paradigm by combination of NiFe-LDH and DFOM. As shown in Scheme 1, NiFe-LDH was synthesized by a co-precipitation method. To further increase the enrichment of nanoconstructs at the tumor site, a crucial regulator of prostate cancer proliferation and metastasis, named platelet-derived growth factor receptor-β (PDGFR-β), was employed as a tumor-targeting motif. Through conjugating its ligand PDGFR-β-binding cyclopeptide (designated as PDGFB) to the surface of NiFe-LDH (the product was designated as pNiFe-LDH), the specific prostate cancer targeting could be achieved [31, 32]. The as-prepared LDH possessed excellent stability and biocompatibility, as evidenced by they did not cause obvious cytotoxicity or damage to the non-target major organs. Unlike the stability observed in the non-stimulated state, administration of DFOM partially disrupted the structural integrity of LDH by chelating Fe³⁺, leading to the release of Ni²⁺. Subsequently, the released Ni²⁺ upregulated insulin like growth factor binding protein 3 (IGFBP3), and further inhibited the downstream PI3K/AKT/mTOR signaling pathway, thereby eliciting apoptosis both in vitro and in vivo. Notably, with the co-administration of DFOM either through intravenous (i.v.) or intratumoral (i.t.) injection, LDH could exert high tumoricidal efficacy. Our study presents a clinically applicable drug-responsive LDH that can switch from a “bioinert” state to a “bioactive” form, offering a feasible and promising strategy for precision cancer therapy.

Scheme 1.

Schematic illustration of the preparation of pNiFe-LDH, DFOM-triggered Ni²⁺ release from pNiFe-LDH, and pNiFe-LDH combined with DFOM eliciting apoptosis to promote prostate cancer cell death. (a) pNiFe-LDH was synthesized by conjugating NiFe-LDH with PDGFB-PEG-DSPE; (b) DFOM chelation triggered the release of Ni²⁺ from pNiFe-LDH; (c) For therapeutic application, the PC3 tumor-bearing BALB/c nude mice were initially intravenously (i.v.) injected with pNiFe-LDH. Three hours later, DFOM was administered via intravenous (i.v.) or intratumoral (i.t.) injection. After cellular uptake, DFOM triggered the decomposition of pNiFe-LDH and gradually increased the release of Ni²⁺ to inhibit the PI3K/AKT/mTOR pathway by upregulating IGFBP3, ultimately eliciting apoptotic cell death in prostate cancer cells

Materials and methods

Materials

A comprehensive list of the cell culture media, chemical and biological reagents utilized in this study was provided in Table S1.

Synthesis of NiFe-LDH and pNiFe-LDH

NiFe-LDH was synthesized via the co-precipitation method described previously [33]. In brief, 20 mL of a metal salt solution containing Ni(NO₃)₂·6H₂O (0.8 M) and Fe(NO₃)₃·9H₂O (0.2 M) was added dropwise to 100 mL of vigorously stirred ammonia solution (25–28%, 0.5 M), followed by mechanical stirring at 65℃ for 18 h. After centrifugation and extensive washing with ddH₂O, the product was vacuum-dried at 60℃ overnight. For pNiFe-LDH synthesis, the PDGFB cyclopeptide (50 mg) was dissolved in dimethyl sulfoxide (DMSO, 30 mL), then an equal amount (500 mg) of EDC and NHS was added, and the reaction mixture was continuously stirred for 2 h. Later on, DSPE-PEG-NH₂ (100 mg) was dissolved in the aforementioned mixture and stirred at room temperature (RT) for another 2 h. After that, unreacted PDGFB and DSPE-PEG-NH₂ were removed via the dialysis method (MWCO = 2 kDa) and the white powder (PDGFB-PEG-DSPE) was obtained by lyophilization. Finally, surface modification was conducted by dispersing NiFe-LDH (1 g) in dimethyl sulfoxide (DMSO, 100 mL), and adding a PDGFB-PEG-DSPE solution, followed by magnetic stirring for 12 h. pNiFe-LDH was isolated via high-speed centrifugation at 25,000 rpm for 10 min.

Characterization

Transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDS) mapping were conducted with an electron microscope (JEOL-2100 F, Tokyo, Japan). Atomic force microscopy (AFM) was performed with an atomic force microscope (Dimension Icon system, Bruker, Berlin, Germany). X-ray powder diffraction (XRD) pattern was determined using Cu-Kα radiation on a diffractometer (Rigaku SmartLab, Tokyo, Japan). Dynamic light scattering (DLS) and surface zeta potentials were recorded using a Zetasizer NanoZS 90 instrument (Malvern Instruments, Southborough, MA). Fourier transform infrared (FT-IR) spectra were obtained using an FT-IR spectrometer (Nicolet iS20, Thermo Fisher Scientific, MA, USA). The released nickel and iron ions were determined using an inductively coupled plasma mass spectrometer (ICP-MS) (7800 ICP-MS, Agilent Technologies, CA, USA).

Cell culture

Human prostate cancer (PC3) cells, alpha mouse liver 12 (AML12) cells, human umbilical vein endothelial cells (HUVECs) and 293T cells were grown at a stable temperature of 37℃ in a humidified atmosphere supplied with 5% carbon dioxide (CO₂). RPMI-1640 medium was used to culture PC3 cells, and Dulbecco’s modified Eagle medium (DMEM) was used to culture HUVECs and 293T cells, the culture media were supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. AML12 cells were cultured in a customized complete medium.

Cell viability assay

The cells were seeded and cultured for 24 h in 96-well plates, followed by various treatments and further incubation for 24 h unless the duration was specified. Next, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide (MTT) was added to achieve final working concentration at 0.5 g/L. After co-culture for 4 h, the media were removed carefully and 150 µL of DMSO was added to dissolve the formazan product. Ultimately, by performing photometric analysis, the absorbance of each solution at 490 nm was quantified with a microplate reader (Infinite M1000 Pro, Tecan, Männedorf, Switzerland).

Lactate dehydrogenase assay

PC3 cells were seeded in 96-well plates at a density of 1 × 10⁴ cells per well and cultured for adherence. After treatment with respective conditions for 24 h, the supernatants were carefully collected and lactate dehydrogenase activity was measured with a commercial kit. The reaction product’s absorbance was quantified at 490 nm using a microplate reader (Infinite M1000 Pro, Tecan, Männedorf, Switzerland). Lactate dehydrogenase release was calculated as a percentage relative to the maximum release from total cell lysis.

Western blot assay

After treatment with various agents, PC3 cells were harvested and subjected to the cell lysis buffer supplemented with a protease/phosphatase inhibitor cocktail to obtain the total cell lysates. Through high-speed centrifugation at 4℃, each concentration of total protein in the supernatant was quantified and normalized. The supernatant was mixed with the corresponding amount of sample loading buffer and boiled at 100 °C for 10 min, and separated by SDS-PAGE. Afterward, the protein bands were transferred to the nitrocellulose (NC) membranes and either incubated with 5% bovine serum albumin or non-fat milk in Tris-buffered saline with 0.1% tween (TBST) at RT for 1.5 h. The blocked membranes were then probed with the primary antibodies at appropriate dilutions (1:3000-1:1000) at 4 °C overnight. After extensive washing with TBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies at a dilution of 1:5000 at RT for 1 h. Finally, the protein bands on the membranes, after extensive washing with TBST, were visualized by enhanced chemiluminescence with an imaging system (5200 Multi, Tanon, China).

Cell apoptosis assay

PC3 cells were seeded in a 12-well plate at a density of 1 × 10⁵ cells per well. After culture for 24 h and following treatment with various agents for another 24 h, the cells were harvested and probed with Annexin V-APC/7-AAD by following the protocol from the manufacturer, and the fluorescence signals were detected using a flow cytometer (CytoFlex, Beckman Coulter, CA, USA).

Mitochondrial membrane potential (MMP) assessment

MMP was assessed with a JC-1 fluorescent probe. In brief, PC3 cells were seeded in a 12-well plate at a density of 1 × 10⁵ cells per well. After various treatments, the cells were then probed with JC-1 (5 µg/mL) in the dark at 37℃ for 20 min. Afterward, the cells were carefully washed with phosphate-buffered saline (PBS) and cultured in fresh media. JC-1 aggregates emitted red fluorescence, while JC-1 monomers emitted green fluorescence. The fluorescence intensities were measured using a fluorescence microscope (IX51, Olympus, Tokyo, Japan).

Lipid peroxide (LPO) assessment

LPO was assessed using the fluorescence probe C11-BODIPY581/591. In brief, PC3 cells were seeded in a 12-well plate at a density of 1 × 10⁵ cells per well and cultured for adherence. After various treatments, the cells were carefully washed with PBS and stained with C11-BODIPY581/591 (2 µM) in the dark at 37℃ for 30 min. The fluorescence signals were measured using a fluorescence microscope (IX51, Olympus, Tokyo, Japan).

Calcein-acetoxymethyl ester (AM)/propidium iodide (PI) dual staining assay

PC3 cells were seeded in a 12-well plate at a density of 1 × 10⁵ cells per well. After culture for 24 h and following treatment with various agents for another 12–24 h, cells were co-stained with calcein-AM (1 µM) and PI (7.5 µM) solution for 20 min at 37 °C in the dark. Afterward, cells were carefully washed with phosphate-buffered saline (PBS) and cultured in fresh media. The fluorescence signals were captured using a fluorescence microscope (IX51, Olympus, Tokyo, Japan).

Lysosomes disruption assay

Lysosomes disruption was assessed using acridine orange (AO) staining. In brief, PC3 cells were seeded in a 12-well plate at a density of 1 × 10⁵ cells per well. After culture for 24 h and following treatment with various agents for another 24 h, the cells were stained with AO (5 µM) in 2 mL of culture medium and incubated in the dark at 37 °C for 30 min. Afterward, cells were carefully washed with phosphate-buffered saline (PBS) and cultured in fresh media. The green and red fluorescence signals were captured using a fluorescence microscope (IX51, Olympus, Tokyo, Japan).

ICP-MS analysis of Ni²⁺ release from pNiFe-LDH with cultured PC3 cells

PC3 cells were seeded in 12-well plates at a density of 1 × 10⁵ cells per well. After incubation for 24 h, cells were treated with various agents. At each time point, the supernatant was transferred to the individual tube. Then, the cells were lysed with RIPA lysis buffer on ice for 20 min followed by collection to the corresponding tube. After centrifugation at 15,000 rpm for 10 min, the concentration of metal ions in the supernatant was determined using ICP-MS (7800 ICP-MS, Agilent Technologies, CA, USA).

Proteomic data analyses

The raw proteomic data, obtained from a mass spectrometer (timsTOF Pro, Bruker, Bremen, Germany), were processed and analyzed using the MaxQuant software (version 1.6.14.0). The global false discovery rate (FDR) was set to 0.01, and protein quantification was performed using both Unique and Razor peptides. The changes in protein expression were evaluated by determining the fold change, and statistical significance was assessed through Student’s t-test to calculate p-values. Proteins with different expression patterns were recognized on the basis of a fold change (FC) greater than 2 or less than 0.5, and an FDR threshold of less than 0.05. Data analyses, including principal component analysis, volcano plot visualization, and hierarchical clustering, were performed via R software. Functional annotations were conducted using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases. In addition, the predicted protein-protein interaction networks were constructed using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database.

TCGA dataset

The profiles of mRNA expression and associated clinical data of prostate cancer patients were collected from the Cancer Genome Atlas Prostate Adenocarcinoma (TCGA-PRAD) dataset, which contains 52 paired tumor and adjacent normal tissues. The level 3 HTSeq-FPKM format data were presented using normalized TPM as transcripts per million reads, and finally used log₂ (TPM + 1) format.

Hemolysis assay

The hemolytic activities of pNiFe-LDH were evaluated by monitoring the hemoglobin release from mouse red blood cells (RBCs). The isolated RBCs were diluted to a suspension with PBS, and various doses of pNiFe-LDH were added to achieve the desired treatment concentrations. Following a 6-hour incubation at RT, each suspension was centrifuged at 12,000 × g for 5 min. PBS or ddH₂O were used as negative and positive controls, respectively. The optical density of each supernatant was subsequently detected at 545 nm using a microplate reader (Infinite M1000 Pro, Tecan, Männedorf, Switzerland).

In vivo half-life and long-term biodistribution

The pNiFe-LDH was intravenously injected at the dose of 30 mg/kg mouse body weight (BW). For half-life determination, blood samples were collected at 5 min, 0.5, 1, 2, 6, 12, and 24 h post injection. For long-term biodistribution assessment, the mice were sacrificed at 1, 7, and 14 d post injection, followed by collection of the liver, spleen, and kidneys. After the acid digestion, Ni²⁺ contents in the blood samples and organs were quantitatively determined by using ICP-MS (7800 ICP-MS, Agilent Technologies, CA, USA).

In vivo distribution and ex vivo tumor accumulation

The fluorescent dye Cy7 and the Cy7-conjugated LDHs (NiFe-LDH@Cy7 and pNiFe-LDH@Cy7) were intravenously injected at the dose of 10 mg/kg mouse BW. The fluorescence signals of the tumor xenografts were captured at 1, 3, 6, 12 and 24 h post injection. After the mice were sacrificed, their major organs (heart, liver, spleen, lung, and kidney) and the excised tumors were collected to assess the tissue distribution. All the fluorescence signals were visualized using an in vivo imaging system (IVIS, ABL X5 pro, Tanon, Shanghai, China).

In vivo tumoricidal effects evaluation

To establish a male BALB/c nude mouse xenograft model of prostate cancer, PC3 cells were suspended in PBS plus Matrigel (5 × 10⁶ cells in a total volume of 150 µL), followed by their subcutaneous injection into the right flank. The mice were then randomly divided into nine groups when the tumor volumes (TVs) reached approximately 200 mm³, as follows: normal saline (NS) injection (group I), DFOM (30 mg/kg) intravenous injection (i.v.) (group II), DFOM intratumoral injection (i.t.) (group III), NiFe-LDH (30 mg/kg) (group IV), pNiFe-LDH (group V) (30 mg/kg), NiFe-LDH plus DFOM (i.v.) (group VI), NiFe-LDH plus DFOM (i.t.) (group VII), pNiFe-LDH plus DFOM (i.v.) (group VIII), and pNiFe-LDH plus DFOM (i.t.) (group IX). Then, the mice were subsequently injected for a total of four times over 12 days. For each time of therapeutic process, the mice were initially intravenously injected with either NiFe-LDH or pNiFe-LDH. Three hours later, DFOM was administered via intravenous or intratumoral injection, and the corresponding BW and TVs were measured every 3 days. TVs were calculated as (length × width²) × 0.52. Following the last administration, the xenograft model mice were euthanized and the tumors were carefully removed, weighed and photographed. In addition, the excised tumors were fixed with 4% formaldehyde overnight, and then the corresponding tumor slides were subjected to immunohistochemical (IHC) and TUNEL staining.

In vivo biosafety evaluation

The serum biochemical parameters of liver and kidney function were recorded using a biochemical analyzer (7080, Hitachi, Tokyo, Japan). The major organs collected from the sacrificed mice were fixed in 4% formaldehyde overnight, followed by hematoxylin-eosin (H&E) staining and capture on a slide scanner (VS200, Olympus, Tokyo, Japan) for histopathological evaluation.

Data analyses

Data analyses were performed using GraphPad Prism (version 9.0.0.121), and the results were expressed as mean ± SD. Statistical significances were determined using one-way ANOVA, with the exception of the method specifically indicated. P-values of *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 were considered statistically significant.

Results and discussion

Synthesis and characterization of NiFe-LDH and pNiFe-LDH

NiFe-LDH was synthesized by co-precipitation using Ni(NO₃)₂·6H₂O and Fe(NO₃)₃·9H₂O, followed by reaction with PDGFB-PEG-DSPE to obtain pNiFe-LDH (Fig. 1a). As observed by TEM and high resolution TEM (HRTEM), the as-synthesized NiFe-LDH displayed round thin flakes with an average size of ~ 30 nm (Fig. 1b), AFM image and corresponding height profile showed that NiFe-LDH presented a thickness of 5.37–6.36 nm (Fig. 1c and Fig. S1), and EDS mapping revealed that Ni, Fe, O elements were uniformly distributed in the NiFe-LDH (Fig. 1d). The XRD pattern of NiFe-LDH demonstrated the characteristic diffraction peaks of (003), (006), and (012) planes at 2θ values of 11.4°, 22.9°, and 34.4°, respectively (Fig. 1e), which exhibited a pattern consistent with the standard diffraction pattern for NiFe-LDH (JCPDS card No. 40–0215) [34]. After conjugation with PDGFB, pNiFe-LDH displayed a similar size, morphology and thickness with NiFe-LDH (Fig. 1f-g and Fig. S2). The hydrodynamic sizes of NiFe-LDH and pNiFe-LDH were 84.8 ± 7.0 nm and 115.5 ± 15.3 nm, larger than the sizes observed by TEM, which was attributed to the existence of hydration layer as well as the partial aggregation of NiFe-LDH or pNiFe-LDH in the aqueous solution (Fig. 1h). Of note, the hydrodynamic sizes exhibited negligible fluctuation within 7 days, indicating their long-term dispersion stability (Fig. S3). The surface charges were then evaluated in ultrapure water and shown in Fig. 1i, NiFe-LDH carried positive charges, whereas PDGFB-PEG-DSPE carried negative charges, and the as-prepared pNiFe-LDH carried positive charges, with zeta potentials measured at 39.13 ± 0.29 mV, -16.24 ± 1.35 mV and 30.97 ± 0.55 mV, respectively. These values indicated that PDGFB-PEG-DSPE could be modified onto the surface of NiFe-LDH via electrostatic interaction, which was similar to the previous reports regarding the modification of positively charged LDHs [35, 36]. In addition, the positive values of LDHs indicated a moderate potential for efficient internalization by cell membranes, which typically carry a negative charge [3]. Later on, FT-IR spectra were recorded to identify the characteristic absorption peaks of pNiFe-LDH. In the FT-IR spectra of PDGFB-PEG-DSPE, characteristic peaks were observed at 2887, 1653, and 1111 cm^− 1 (Fig. 1j), corresponding to the C–H stretching vibration in CH₂, C = O stretching vibration, and C–O stretching vibration, respectively [37]. This result indicated the successful conjugation of PDGFB with DSPE-PEG. Furthermore, the spectrum of pNiFe-LDH exhibited the characteristic peaks of PDGFB-PEG-DSPE, further confirming that PDGFB-PEG-DSPE has been successfully conjugated to NiFe-LDH.

Fig. 1.

Characterization of NiFe-LDH and pNiFe-LDH. (a) Schematic illustration of the preparation procedure for NiFe-LDH and pNiFe-LDH. (b, f) TEM and HRTEM images of NiFe-LDH (b) and pNiFe-LDH (f). (c, g) AFM images of NiFe-LDH (c) and pNiFe-LDH (g). (d) EDS mapping images of NiFe-LDH. Scale bar, 20 nm. (e) XRD pattern of NiFe-LDH. (h) Hydrodynamic sizes of NiFe-LDH and pNiFe-LDH. (i) Surface zeta potentials and (j) FT-IR spectra of NiFe-LDH, pNiFe-LDH, and PDGFB-PEG-DSPE

DFOM-triggered Ni²⁺ release from NiFe-LDH elicited prostate cancer cell death

Inspired by the treatment regime of metal ions poisoning reported in previous studies [29, 38], we speculated that the tumor accumulated Fe³⁺-based LDH could be controllably disassembled by the iron chelator DFOM, thus triggering structural collapse of NiFe-LDH and the in situ release of cytotoxic Ni²⁺. Prior to verifying our hypothesis, we first evaluated the individual impact of NiFe-LDH or DFOM on the viability of human PC3 prostate cancer cells. As a result, exposure to NiFe-LDH at concentrations up to 200 µg/mL for 24 h demonstrated neither statistically significant reduced cell viabilities nor increased lactate dehydrogenase release compared with each control group, whereas NiFe-LDH caused statistically significant cell viabilities reduction when the duration of administration reached no less than 36 h (Fig. 2a-b and Fig. S4). In addition, the calcein-AM/PI dual staining assay [39] was adopted to further assess the potential cytotoxicity of NiFe-LDH. As shown in Fig. S5, PC3 cells after NiFe-LDH treatment exhibited exclusive green fluorescence, a signature indicative of viable cell labeling. Meanwhile, DFOM did not show any obvious cytotoxicity at concentrations up to 200 µM for 24 h (Fig. 2c), demonstrating that the administration of NiFe-LDH or DFOM alone was non-toxic to the cells within specific administrated concentration and duration. Next, we have examined the tumoricidal effect of NiFe-LDH and DFOM on PC3 cells through combined administration of the two “inert” agents with different concentrations for 24 h. As shown in Fig. 2d, co-administrations of NiFe-LDH and DFOM at different combined concentrations reduced the viabilities of PC3 cells, and the most significant effect occurred with 200 µg/mL NiFe-LDH plus 200 µM DFOM. By employing a Hoechst 33,342/propidium iodide (HO/PI) staining assay, which is commonly used to distinguish live and dead cells, quite a few PI-positive stained cells after combined administration of NiFe-LDH and DFOM were observed, demonstrating potent tumoricidal effect on PC3 cells, whereas the control or monotherapy group caused negligible cell death (Fig. 2e). To verify the release of Ni²⁺ from NiFe-LDH, ICP-MS was subsequently applied to quantitatively assess the concentration of Ni²⁺ in the supernatants of NiFe-LDH and NiFe-LDH + DFOM incubated under various pH conditions, including physiological (~ pH 7.4), tumorous (~ pH 6.5), and endosomal/lysosomal (~ pH 5.5) microenvironments [40]. As illustrated in Fig. 2f and Fig. S6, the Ni²⁺ concentration in the supernatants of NiFe-LDH + DFOM increased remarkably with prolonged incubation, whereas NiFe-LDH alone exhibited only slight release (less than 25% within 24 h) [41] even under the acidic condition at pH 5.5, highlighting its relatively excellent chemical stability. These findings indicated that the robust Ni²⁺ release of NiFe-LDH was largely induced by the iron chelator DFOM. Notably, following the co-administration of DFOM, Ni²⁺ release from NiFe-LDH under acidic condition at pH 5.5 was significantly enhanced, reaching 70% within 24 h, which was in accordance with the design expectations. Furthermore, we have compared the impact of NiFe-LDH plus DFOM on PC3 cells with or without pre-reaction and centrifugation. As shown in Fig. 2g, direct administration of NiFe-LDH and DFOM to cultured PC3 cells caused a 43.19 ± 2.31% reduction in cell viability reduction. Intriguingly, after pre-reaction of NiFe-LDH and DFOM in the cell-free tube for 24 h and subsequent high-speed centrifugation, the supernatant exerted approximately the same effect as that of direct administration, while the separated precipitate failed to cause a noticeable reduction in cell viability. These results suggested again that Ni²⁺ released from DFOM-chelated NiFe-LDH was probably responsible for the tumoricidal effect on PC3 cells. To prove the hypothesis, we further measured the viabilities of PC3 cells after nickel chloride and/or ferric chloride administrations (equal amounts of Ni/Fe in NiFe-LDH), and confirmed that Ni²⁺ rather than Fe³⁺ was the main cause of the decrease in cell viabilities (Fig. 2h). Then, PC3 cells administrated with NiFe-LDH and DFOM were co-treated with ethylene diamine tetraacetic acid (EDTA), which is frequently applied for divalent metal ions chelation [42]. As a result, the viabilities of PC3 cells reduced after treatment with NiFe-LDH plus DFOM, while EDTA could reverse the reduction in cell viabilities in a dose-dependent manner (Fig. 2i).

Fig. 2.

DFOM-triggered Ni²⁺ release from NiFe-LDH elicited prostate cancer cell death in vitro. (a–d) Viabilities of PC3 cells treated with NiFe-LDH for 24 h (a), NiFe-LDH (200 µg/mL) for 24, 36, and 48 h (b), DFOM for 24 h (c), and NiFe-LDH plus DFOM for 24 h (d). (e) HO/PI double staining images of PC3 cells after co-treatment with NiFe-LDH and DFOM for 24 h. Scale bar, 50 μm. (f) Profile of Ni²⁺ release from NiFe-LDH triggered by DFOM under different pH conditions, measured at the specified time points via ICP-MS. (g) Viabilities of PC3 cells after various treatments for 24 h (I: direct treatment with NiFe-LDH plus DFOM, II: treatment with the precipitate centrifuged after 12 h-mixture of NiFe-LDH plus DFOM, III: treatment with the precipitate centrifuged after 24 h-mixture of NiFe-LDH plus DFOM, IV: treatment with the supernatant centrifuged after 12 h-mixture of NiFe-LDH plus DFOM, V: treatment with the supernatant centrifuged after 24 h-mixture of NiFe-LDH plus DFOM, VI: treatment with NiFe-LDH plus DFOM after 12 h-mixture, VII: treatment with NiFe-LDH plus DFOM after 24 h-mixture). Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM. (h) Viabilities of PC3 cells after various treatments for 24 h. Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; NiCl₂·6H₂O, 243 µg/mL with nickel equivalent; FeCl₃·6H₂O, 75 µg/mL with iron equivalent. (i) Viabilities of PC3 cells treated with EDTA either in the absence or presence of NiFe-LDH plus DFOM for 24 h. Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; EDTA, 0.25 or 0.5 mM. Data were presented as mean ± SD (n = 6), ns, no significance; **p < 0.01; ***p < 0.001; ****p < 0.0001, compared with the control group or the indicated groups

NiFe-LDH combined with DFOM elicited apoptosis in prostate cancer cells

Nickel-based nanomaterials can elicit apoptosis in several types of cancer cells [43, 44]. To validate whether DFOM-triggered Ni²⁺ release from NiFe-LDH elicited apoptosis in PC3 cells, we employed the TUNEL staining assay. TUNEL can identify apoptotic cells by labeling the free 3’-OH ends of fragmented DNA, which is a hallmark of apoptosis [45]. As shown in Fig. 3a, after co-administration of NiFe-LDH and DFOM, the fluorescence intensity of TUNEL-positive staining significantly elevated. Next, the MMP of PC3 cells was assessed by using the JC-1 fluorescence probe, which often accumulates in the mitochondria of the intact cells, where it forms aggregates that emit red fluorescence, whereas a decrease in the MMP will lead to the emission of green fluorescence in apoptotic cells [46]. The remarkably increased green/red fluorescence intensity ratio indicated significant MMP reduction and mitochondrial injury, which were observed after NiFe-LDH and DFOM co-administration (Fig. 3b-c). In addition, acridine orange (AO) staining assay [47] was performed to evaluate the lysosomes destruction. As shown in Fig. S7a-b, the control, DFOM, and NiFe-LDH groups presented red-staining spots in the cytoplasm, whereas in NiFe-LDH and DFOM co-administration group, the red fluorescence of AO decreased sharply, indicating that the integrity of lysosomes was destroyed. Furthermore, we have employed a specific inhibitor Z-VAD-FMK [48], which can irreversibly bind to the active site of caspases, preventing them from cleaving substrates and thus inhibiting apoptosis. Poly (ADP-ribose) polymerase (PARP) generally participates in DNA repair as well as the cellular stress response. During apoptosis, caspases will cleave PARP into smaller fragments as cleaved PARP, which generally serves as reliable indicators for apoptotic cell death [49]. As a result, the combination of NiFe-LDH and DFOM elicited apoptosis in PC3 cells, characterized by the significantly elevated expression level of cleaved PARP, whereas Z-VAD-FMK could significantly abrogated PARP cleavage (Fig. 3d-e). As shown in Fig. 3f, compared with the administration of NiFe-LDH and DFOM, co-administration of NiFe-LDH and DFOM with Z-VAD-FMK significantly reversed the reduction in the number of viable cells in a dose-dependent pattern. To quantify apoptotic cells after the combined administration of NiFe-LDH and DFOM, the Annexin V-APC/7-AAD staining assay was subsequently conducted. More apoptotic cells emerged after NiFe-LDH and DFOM co-administration for 24 h in PC3 cells, and Z-VAD-FMK partially abrogated the tumoricidal effect (Fig. 3g). Of note, unlike the ferroptosis inducer RSL3, NiFe-LDH and DFOM co-administration did not induce the accumulation of lipid peroxides (LPO), as the fluorescence emission of the fluorescent probe used to detect LPO did not shift (Fig. S8). Collectively, the combined administration of NiFe-LDH and DFOM elicited apoptosis in PC3 prostate cancer cells, and mitochondrial dysfunction may participate in the process.

Fig. 3.

NiFe-LDH and DFOM co-treatment elicited apoptosis in prostate cancer cells. (a) TUNEL staining images of PC3 cells after various treatments for 24 h. Scale bar, 50 μm. (b-c) Fluorescence microscopy images (b) and the relative fluorescence intensity ratio (c) of PC3 cells stained with JC-1 probe after various treatments for 24 h. Data were presented as mean ± SD (n = 3). Scale bar, 20 μm. (d-e) Western blot images (d) and semi-quantitative analyses (e) of pro-PARP and cleaved PARP in PC3 cells after the various treatments for 24 h (n = 3). Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; Z-VAD-FMK, 40 µM. (f) Viabilities of PC3 cells treated with Z-VAD-FMK either in the absence or presence of NiFe-LDH and DFOM for 24 h (n = 6). Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; Z-VAD-FMK, 20 or 40 µM. (g) After various treatments for 24 h, PC3 cells were stained with Annexin V-APC/7-AAD and analyzed by flow cytometry. Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; Z-VAD-FMK, 40 or 80 µM. Data were presented as mean ± SD, ns, no significance; ****p < 0.0001, compared with the control group or the indicated groups

NiFe-LDH combined with DFOM upregulated IGFBP3 to inhibit PI3K/AKT/mTOR pathway for eliciting apoptosis

To elucidate the underlying mechanism involved in apoptosis caused by co-administration of NiFe-LDH and DFOM, we have conducted proteomics analysis. Among the differentially expressed proteins, IGFBP3 was identified as a notably upregulated protein after NiFe-LDH plus DFOM administration compared with the control group (Fig. 4a and Table S2). To examine the expression pattern of IGFBP3 in the clinical samples, the mRNA expression profiles and associated clinical data of prostate cancer patients were collected from the TCGA-PRAD database, which contained 495 tumorous tissues and 52 adjacent normal tissues. As shown in Fig. 4b, the normalized IGFBP3 expression was downregulated significantly in prostate cancer tissues than that in adjacent normal tissues, suggesting the tumor suppressor role of IGFBP3. The crucial impact of the PI3K/AKT/mTOR pathway on promoting tumor progression and acquired resistance to anticancer therapies, has been well elucidated [50, 51]. In addition, previous reports revealed that IGFBP3 might suppress PI3K pathway, leading to decreased AKT activation, lower mTOR phosphorylation and increased cell death [52, 53]. Thus, IGFBP3 expression and the total or phosphorylated levels of PI3K, AKT and mTOR, were examined in PC3 cells subjected to different treatments. As shown in Fig. 4c-d, compared with that in the control group, IGFBP3 expression significantly increased after combined administration of NiFe-LDH and DFOM, whereas decreased abundance of p-PI3K, p-AKT and p-mTOR were observed. Moreover, the expression levels of apoptosis-related proteins, characterized by the cleaved forms of caspase 3 and PARP, have significantly increased after the combined administration of NiFe-LDH and DFOM (Fig. 4e-f). Furthermore, to explore the modulatory interactions between IGFBP3 and the PI3K/AKT/mTOR signaling pathway, we have knocked down IGFBP3 in PC3 cells by using specific siRNA transfection (si-IGFBP3). Compared with that in non-sense control siRNA (si-NC)-transfected cells, the IGFBP3 protein level was significantly lower in PC3 cells transfected with si-IGFBP3, whereas the p-PI3K level was significantly elevated, verifying the upstream regulatory role of IGFBP3 (Fig. 4g-h). Importantly, in IGFBP3-knockdown cells, the combined administration of NiFe-LDH and DFOM exhibited significantly lower cytotoxicity than the control group (Fig. 4i). All the aforementioned findings collectively demonstrated that DFOM-triggered Ni²⁺ release from NiFe-LDH upregulated IGFBP3 expression level, which inhibited PI3K activation, thereby suppressing downstream AKT kinase activity and a series of phosphorylation cascade reactions. Suppression of the PI3K/AKT/mTOR signaling pathway led to the upregulation of cleaved caspase-3 and cleaved PARP, ultimately eliciting apoptosis.

Fig. 4.

NiFe-LDH and DFOM co-treatment elicited apoptosis via IGFBP3 upregulation-mediated PI3K/AKT/mTOR pathway inhibition. (a) Volcano plot of differential protein expression between the groups treated with vehicle control or NiFe-LDH + DFOM. Red and blue dots represented proteins with FDR < 0.05 and|log₂ FC| >1, and statistical significance was calculated using Student’s t-test. (b) mRNA expression of IGFBP3 in tumorous and adjacent normal tissues was determined on the basis of data obtained from the TCGA and GTEx databases (n = 52), and statistical significance was calculated using Student’s t-test. (c-d) Western blot images (c) and semi-quantitative analyses (d) of IGFBP3, total and phosphorylated PI3K, AKT, and mTOR expression in PC3 cells after various treatments for 24 h (n = 3). (e-f) Western blot images (e) and semi-quantitative analyses (f) of pro-PARP/caspase-3 and cleaved PARP/caspase-3 expression in PC3 cells after various treatments for 24 h (n = 3). (g-h) Western blot images (g) and semi-quantitative analyses (h) of IGFBP3, total and phosphorylated PI3K in PC3 cells treated with NiFe-LDH plus DFOM in the presence of si-NC or si-IGFBP3 (n = 3). Dosage: NiFe-LDH, 200 µg/mL; DFOM, 200 µM; si-NC or si-IGFBP3, 50 nM. (i) Viabilities of PC3 cells after various treatments for 24 h (n = 6), ns, no significance; ****p < 0.0001, compared with the control group or the indicated groups

DFOM-triggered Ni²⁺ release from pNiFe-LDH and biodistribution of pNiFe-LDH in vivo

ICP-MS was applied to quantify the release of Ni²⁺ from pNiFe-LDH. As illustrated in Fig. S9a, administration of DFOM significantly increased Ni²⁺ release from pNiFe-LDH, reaching approximately 75% over 24 h. Moreover, when evaluated under the cell culture condition (RPMI-1640 + 10% FBS) with PC3 cells, DFOM-triggered Ni²⁺ release from pNiFe-LDH showed a further amplification, exceeding 90% within 24 h (Fig. S10a). Intriguingly, due to DFOM can bind Fe³⁺ via its three hydroxamate groups and formed a water-soluble ferrioxamine complex [54], the concentration of Fe³⁺ in the supernatant or cell culture media in pNiFe-LDH + DFOM samples increased with the nearly consistent trend as that of Ni²⁺ (Fig. S9b and S10b). In addition, although TEM images indicated no substantial variation in overall sizes of pNiFe-LDH after DFOM chelation, the emergence of a few smaller-sized particles partially verified the structural disintegration of pNiFe-LDH following DFOM chelation (Fig. S11a-b). These findings confirmed that the addition of DFOM led to a significant release of Ni^2 + from pNiFe-LDH. Notably, pNiFe-LDH alone without DFOM administration exhibited extremely low Ni²⁺ and Fe³⁺ release (less than 2% within 7 days), verifying the excellent stability of pNiFe-LDH under physiological pH value (Fig. S12a-b).

To evaluate the tumor-targeting capability of NiFe-LDH and pNiFe-LDH, Cy7-labeled NiFe-LDH (NiFe-LDH@Cy7) and pNiFe-LDH (pNiFe-LDH@Cy7) were intravenously injected into the mice. The fluorescence signals were subsequently captured from the tumor xenografts at the indicated time points post injection. Free Cy7 did not possess any tumor-targeting ability, and at 24 h post injection, NiFe-LDH@Cy7 displayed a diminished fluorescence intensity in the tumor xenografts, while pNiFe-LDH@Cy7 still presented a relatively high fluorescence signal. Notably, after intravenous injection of pNiFe-LDH@Cy7, the fluorescence intensity in the tumor xenografts reached a the relatively maximum value at 3 h post injection compared with its systemic distribution (Fig. S13a). To evaluate the biodistribution of NiFe-LDH and pNiFe-LDH in the major organs, the xenograft model mice were euthanized at 24 h post injection, followed by collection of the main organs and excised tumors for fluorescence imaging. Consistent with the in vivo imaging pattern, no fluorescence signal in tumors after free Cy7 injection can be observed. Compared with the weak fluorescence displayed after NiFe-LDH@Cy7 injection, pNiFe-LDH@Cy7 exhibited a higher fluorescence intensity, suggesting that PDGFB conjugation endowed NiFe-LDH with specific tumor-targeting ability. Furthermore, NiFe-LDH@Cy7 and pNiFe-LDH@Cy7 were mainly distributed in the lungs, kidneys, spleens and especially the livers (Fig. S13b-c). Notably, as mentioned above, several types of nanomaterials preferentially accumulate in the liver and cause potential toxicological effects [15, 16]. However, in theory, the administration of NiFe-LDH or pNiFe-LDH alone would not cause obvious tissue injury, as its good biocompatibility has been revealed in Fig. 2a and Figs. S4-5, 14–15, neither statistically significant reduced cell viabilities nor increased cytotoxicities were observed in PC3 prostate cancer cells or non-malignant AML12, HUVEC and 293T cells after administration with NiFe-LDH or pNiFe-LDH at gradually increased concentrations up to 200 µg/mL for 24 h.

Furthermore, the blood circulation time of pNiFe-LDH was investigated by detecting the Ni content via ICP-MS. As shown in Fig. S16, the half-life of pNiFe-LDH was calculated to be 1.33 h. Eventually, long-term biodistribution assessment revealed that although pNiFe-LDH initially tended to be accumulated in liver and spleen compared with kidneys, nearly all the accumulated Ni content in liver, spleen and kidneys were eliminated within 2 weeks (Fig. S17). In general, nanoconstructs with hydrodynamic size no less than 100 nm quickly accumulate in the liver and spleen through macrophage phagocytosis and entrapment in the sinusoids [55], which was in consistence with the biodistribution of pNiFe-LDH. Meanwhile, the nanoconstructs can also possibly undergo renal excretion if they dissociate into small fragments [56]. Of note, the as-synthesized LDHs, exhibiting a controlled size distribution of 60–150 nm, demonstrated efficient clearance from the body through both fecal and urinary excretion pathways [35, 39].

Antitumor efficacy of pNiFe-LDH combined with DFOM in vivo

To demonstrate the antitumor effects of NiFe-LDH and pNiFe-LDH and optimize the administration route, when the volume of PC3 tumor xenografts in BALB/c nude mice reached approximately 200 mm³, nine groups were randomly established: normal saline (NS) injection (I), DFOM intravenous injection (i.v.) (II), DFOM intratumoral injection (i.t.) (III), NiFe-LDH (IV), pNiFe-LDH (V), NiFe-LDH plus DFOM (i.v.) (VI), NiFe-LDH plus DFOM (i.t.) (VII), pNiFe-LDH plus DFOM (i.v.) (VIII), and pNiFe-LDH plus DFOM (i.t.) (IX), respectively. The mice were subsequently injected every 3 days for a total of four times. For each time of injection in the therapeutic groups, the PC3 tumor-bearing mice were initially injected (i.v.) with either NiFe-LDH or pNiFe-LDH. Three hours later, DFOM was administered via intravenous (i.v.) or intratumoral (i.t.) injection, and the corresponding BWs and TVs were measured simultaneously (Fig. 5a). The BWs were recorded for preliminary evaluation of the systemic toxicity and no obvious changes were observed among the groups (Fig. 5b). For the tumoricidal efficacy assessment, NiFe-LDH or pNiFe-LDH alone did not exert significant antitumor efficacy, however, when combined with DFOM, either via intravenous (i.v.) or intratumoral (i.t.) injection, NiFe-LDH or pNiFe-LDH could exhibit significantly greater antitumor effects. In addition, compared with NiFe-LDH, pNiFe-LDH presented more potent tumoricidal effects when combined with DFOM administration (Fig. 5c-d), which was attributed to the active tumor targeting facilitated by PDGFB conjugation. Three days after the last therapeutic injection, the tumors were surgically excised from the sacrificed mice and photographed (Fig. 5e). Consistent with the antitumor efficacy reflected by the tumor volumes, the tumor weights in the groups co-treated with pNiFe-LDH and DFOM were most significantly decreased (Fig. 5f, Fig. S18). The tumor tissue slides were then subjected to H&E staining, which demonstrated tumor cell death accompanied by cavitation lesions in the groups treated with NiFe-LDH/pNiFe-LDH and DFOM (Fig. 5g). As revealed in vitro, NiFe-LDH combined with DFOM elicited IGFBP3-regulated apoptosis in cultured PC3 cells, thus we evaluated the apoptosis-inducing effects in the xenografts via immunofluorescence (IF) and IHC staining, and found that NiFe-LDH or pNiFe-LDH combined with DFOM resulted in increased TUNEL fluorescence intensity (Fig. 6a, Fig. S19), indicating the higher apoptotic activity [57]. Moreover, elevated expression level of IGFBP3 and cleaved caspase-3 accompanied with decreased expression level of phosphorylated mTOR were documented (Fig. 6b, Fig. S20). These in vivo experimental data demonstrated that LDHs mediated the upregulation of IGFBP3, thereby inhibiting the PI3K/AKT/mTOR signaling axis, which subsequently triggered the activation of apoptotic effectors, as evidenced by increased expression level of cleaved caspase-3 [58], ultimately resulting in apoptotic cell death. In addition, Ki67, a commonly used proliferative marker [59], was expressed at lower levels in the groups treated with NiFe-LDH/pNiFe-LDH and DFOM (Fig. 6b, Fig. S20), which was consistent with the results of H&E and TUNEL staining. Collectively, these results demonstrated that NiFe-LDH, especially pNiFe-LDH, could exhibit excellent tumoricidal effects when co-treated with DFOM in vivo, raising an ideal “cascade” strategy for cancer therapy.

Fig. 5.

Therapeutic efficacy of NiFe-LDH combined with DFOM. (a) Schedule diagram of the animal experiment procedure. BALB/c nude mice, with xenograft tumor volumes (TVs) reached approximately 200 mm³, were randomly divided into nine groups: I: normal saline (NS, intravenous injection, i.v.); II: DFOM (i.v.); III: DFOM (intratumoral injection, i.t.); IV: NiFe-LDH (i.v.); V: pNiFe-LDH (i.v.); VI: NiFe-LDH (i.v.) + DFOM (i.v.); VII: NiFe-LDH (i.v.) + DFOM (i.t.); VIII: pNiFe-LDH (i.v.) + DFOM (i.v.); IX: pNiFe-LDH (i.v.) + DFOM (i.t.). Dosage: NiFe-LDH, 30 mg/kg BW; pNiFe-LDH, 30 mg/kg BW; DFOM, 30 mg/kg BW. The above therapeutic injections were conducted every three days for a total of four times. (b–d) Body weight curves (b) and tumor volume growth curves (c-d) after various treatments. (e-f) Digital photographs (e) and weights of the excised tumors (f) at the endpoint of the various treatments. (g) H&E staining images of tumor tissue slides after various treatments. Scale bar, 100 μm. Data were presented as mean ± SD (n = 5), ns, no significance; *p < 0.05; **p < 0.01; ****p < 0.0001, compared with the control group or the indicated groups

Fig. 6.

Biocompatibility of NiFe-LDH or pNiFe-LDH combined with DFOM in vivo. (a) TUNEL staining images of tumor tissue slides obtained after various treatments. Scale bar, 100 μm. (b) IHC staining images of IGFBP3, p-mTOR, cleaved caspase-3 and Ki67 in tumor tissue slides after various treatments. Scale bar, 100 μm. (c-d) Serum ALT (c), and AST (d) levels in PC3 tumor-bearing BALB/c nude mice at the endpoint of various treatments. (e) H&E and IHC staining images of cleaved caspase-3 in liver tissues at the endpoint of various treatments, red arrows indicated cholestasis; black arrows indicated the formation of lipid droplets, white scale bar, 25 μm; black scale bar, 100 μm. Data were presented as mean ± SD (n = 5), ns, no significance; **p < 0.01; ***p < 0.001; ****p < 0.0001, compared with the control group

Biocompatibility of NiFe-LDH or pNiFe-LDH combined with DFOM in vivo

To perform the antitumor efficacy assessment, pNiFe-LDH was intravenously (i.v.) injected, thus we first evaluated whether the nanoconstruct would cause hemolysis. As shown in Fig. S21a, unlike ddH₂O, which served as a positive control, pNiFe-LDH did not cause obvious color changes at the tested concentrations. The hemolysis rate was calculated as 2.76% even at the highest concentration of 200 µg/mL, and the rate was below the commonly accepted threshold of 5% for hemolysis (Fig. S21b). Next, we evaluated the potential injuries to the major organs after distinct therapeutic interventions, which were represented by the kidney and liver function tests. As a result, the serum urea nitrogen (BUN) and creatinine (CREA) levels, which are the classical and widely used biomarkers for assessing kidney function, were all within the normal ranges and no significant differences were revealed between the control and experimental groups (Fig. S22). To perform the liver function assessment, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were examined. Compared with the normal saline control group, although the ALT and AST levels elevated significantly and exceeded the normal ranges in the mice co-treated with NiFe-LDH (i.v.) plus DFOM (i.v), the elevated levels of ALT and AST in the mice co-treated with pNiFe-LDH (i.v.) plus DFOM (i.v) were mostly within the normal ranges. Moreover, the levels of ALT and AST did not significantly elevate in the mice co-treated with pNiFe-LDH (i.v.) plus DFOM (i.t), suggesting the negligible off-target toxicity with this pattern of intervention (Fig. 6c-d). Furthermore, H&E staining was performed to examine the potential histological changes of the major organs. As a result, no evident pathological abnormalities or damages were observed in the heart, spleen, lung or kidney following the distinct treatments (Fig. S23). However, consistent with the elevated ALT and AST values which reflected the liver dysfunction, H&E staining of the liver in the mice that received co-administration of NiFe-LDH and DFOM (i.v.) showed mild congestion, partial loosening and edema of hepatocytes, a small amount of fatty degeneration (formation of lipid droplets), and visible cholestasis. Later on, the liver tissues were harvested for IHC staining of cleaved caspase-3, which verified that co-administration of NiFe-LDH and DFOM (i.v.) has caused slight injury to the liver rather than the other major organs, as there were a few apoptotic hepatocytes positively labeled with cleaved caspase-3 (Fig. 6e and Fig. S24). Collectively, the aforementioned results indicated that the liver injury actually occurred when the therapeutic agents can accumulate in the non-target organ. In our current study, non-toxic pNiFe-LDH alone did not cause obvious organ damage, moreover, when pNiFe-LDH was co-administrated with DFOM (i.v. or i.t. injection), there were still no significant liver injuries were detected. In particular, when pNiFe-LDH was co-administrated with DFOM (i.t.), they had no chance to react in the non-target organs or even cause any potential damage. Notably, the anti-tumor efficacy of the co-administration of pNiFe-LDH and DFOM (i.t.) was comparable to that of the co-administration of pNiFe-LDH and DFOM (i.v.) (Fig. 5c-g), verifying the promising and safe therapeutic intervention.

Conclusion

In conclusion, a drug-responsive NiFe-LDH nanoplatform was successfully designed and synthesized for promising precision cancer therapy. After sequential injection of the “bioinert” NiFe-LDH and the FDA-approved drug iron chelator DFOM, the Fe³⁺ were chelated, followed by the decomposition of NiFe-LDH to a “bioactive” form, which was characterized by the robust release of Ni²⁺. The released Ni²⁺ activated the IGFBP3/PI3K/AKT/mTOR signaling pathway, ultimately inducing potent apoptosis in prostate cancer cells. Moreover, pNiFe-LDH, which was synthesized by conjugation of the PDGFR-β-targeting cyclic peptide with NiFe-LDH, significantly enhanced the accumulation in the tumor microenvironment, which amplified the tumoricidal effect and minimized the potential systemic toxicity caused by off-target effects. However, there are still several limitations that require further consideration. Firstly, although cultured cancer cell line exhibited a typical phenotype after treatment with the as-prepared LDH in vitro, primary tumor cells may possess distinct characteristics. Meanwhile, evaluation of the tumoricidal effects relied on the mouse subcutaneous tumor model in vivo. Given the unique vascular structure, dense extracellular matrix barriers in the tumor microenvironment, subcutaneous xenograft model will inevitably lead to discrepancies in predicting clinical therapeutic efficacy. Secondly, while the blood circulation time, biodistribution, stability and preliminary biocompatibility assessments of pNiFe-LDH have been conducted in the current study, the metabolic and excretory processes over a longer time have not been monitored. Last but not least, whether pNiFe-LDH will cause the toxicities such as chronic inflammation, fibrosis, and pathological immune responses in the major organs, remain to be elucidated. In the future, more relevant studies are needed to further evaluate the therapeutic efficacy of pNiFe-LDH by using the orthotopic patient-derived xenograft model, moreover, systematic investigation into the metabolic and excretion processes of pNiFe-LDH, coupled with a comprehensive evaluation of its long-term biosafety, is imperative to establish an evidence-based foundation for the potential clinical translation. Overall, this study demonstrates an approach to activating the “bioinert” nanomaterials to “bioactive” therapeutic agents by using clinically used drugs, providing a promising and clinically translatable strategy for precise and targeted cancer therapy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Z.W.: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Writing - review and editing. H.W.: Methodology, Project administration, Software, Writing - review and editing. L.Y.: Methodology, Project administration, Validation, Writing - review and editing. R.T.: Investigation, Project administration, Writing - review and editing. W.G.: Investigation, Project administration, Writing - review and editing. S.C.: Methodology, Writing - review and editing. G.J.: Project administration, Writing - review and editing. W.L.: Project administration, Writing - review and editing. P.W.: Writing - review and editing. X.H.: Writing - review and editing. C.L.: Conceptualization, Resources. Y.Z.: Funding acquisition, Resources, Writing - review and editing. G.Z.: Funding acquisition, Resources, Supervision, Writing - review and editing. L.Z.: Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Writing - original draft, Writing - review and editing.

Funding

This work was financially supported by National Natural Science Foundation of China (82272149, 82072055, T2222014, 22007006, 82400905), Science Foundation for Outstanding Young Scholar of Anhui Colleges (2022AH020073), Key Project of Provincial Natural Science Research Project of Anhui Colleges (2023AH053288), Distinguished Young Scholar of Anhui Colleges (2021-108-10), Taishan Scholars Construction Engineering (tsqn201909144), Innovative Leading Talents of Anhui Province (T000529), Key R&D Program of Shandong Province (2023CXPT012), Science Research Project for postdoctoral of Anhui Province (2023b711), Young Scholars Cultivation Fund of the First Affiliated Hospital of Anhui Medical University (2021KJ04).

Data availability

Data will be made available on reasonable request.

Declarations

Ethics approval and consent to participate

All the animal experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committees on Animal Care of Anhui Medical University (Approval No. LLSC20241776).

Consent for publication

All authors agree with the publication.

Competing interests

The authors declare no competing interests.