Targeting CAD with a tumor microenvironment-responsive nano-heterojunction for synergistic induction of cuproptosis and inhibition of colorectal cancer progression

Abstract

Carbamoyl-phosphate synthetase II/Aspartate transcarbamylase/Dihydroorotase (CAD) is a multifunctional enzyme with demonstrated oncogenic potential across various cancers, emerging as a promising therapeutic target. However, its functional role and therapeutic relevance in colorectal cancer (CRC) remain unclear. Here, we demonstrate for the first time that elevated CAD expression promotes CRC malignancy by enhancing proliferation, motility, and apoptosis resistance. Clinically, high CAD expression correlates with poor survival in CRC patients. Leveraging these findings, we developed a tumor microenvironment (TME)-responsive nickel‑copper nano‑heterojunction (NCSH@siCAD&AVT‑18A) that selectively releases hydrogen sulfide and copper ions in the acidic, reducing TME. This platform induces intracellular acidification and redox stress, leading to synergistic tumor suppression through a trimodal strategy that integrates CAD gene silencing, cuproptosis induction, and immune modulation. In vivo studies confirmed potent antitumor efficacy across subcutaneous, orthotopic, and metastatic CRC models, without systemic toxicity. Our work identifies CAD as a key prognostic marker and therapeutic target in CRC, and presents a novel combinatory regimen that integrates gene silencing with metal ion-mediated therapy, offering a mechanistically innovative and clinically viable strategy for CRC treatment with superior efficacy and minimal toxicity.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03858-1.

Background

CRC ranks as the third most commonly diagnosed malignancy and the second leading cause of cancer-related mortality worldwide [1]. Its growing incidence, particularly among younger individuals, underscores its significance as a global health concern. In 2023, an estimated 2.08 million new CRC cases and 1.04 million associated deaths were reported globally [2]. Although early-stage CRC is often amenable to surgical resection, with five-year survival rates exceeding 90%, more than half of patients are diagnosed at an advanced stage, where treatment becomes substantially more challenging [3]. The management of advanced CRC relies heavily on systemic therapy, which has substantial limitations [4]. Standard chemotherapies such as FOLFOX/FOLFIRI are associated with significant off-target toxicity, while targeted biologics (e.g., bevacizumab, cetuximab) are often limited by inherent tumor heterogeneity and the development of acquired resistance [5–9]. Immune checkpoint inhibitors benefit only approximately 5% of patients with MSI-H tumors, leaving most microsatellite-stable (MSS) cases without effective immunotherapy options [10].

These limitations underscore the urgent need for the development of next-generation targeted therapies [11]. Nanotechnology offers a promising solution. Nanoparticles (NPs), with their unique physicochemical properties and tunable architectures, are ideal for precision oncology [12]. Their size (1–1000 nm) enables selective tumor accumulation through the enhanced permeability and retention (EPR) effect [13]. Passive targeting can be augmented by active targeting strategies through surface modifications, For instance, functionalization with hyaluronic acid (HA) enables binding to CD44 receptors overexpressed on CRC cells, which facilitates receptor-mediated endocytosis and enhances tumor specificity [14–17]. The integration of passive and active strategies can boost both intratumoral accumulation and cellular uptake significantly. Furthermore, the use of nanocarriers enables the simultaneous co-delivery of chemotherapeutics and gene therapies, overcoming the pharmacokinetic limitations of traditional combinations and promoting synergistic efficacy [18]. Crucially, the clinical translation of nanotherapeutics relies on the identification and validation of suitable molecular targets [19, 20]. Current molecularly targeted therapies for CRC face persistent efficacy barriers. KRAS is mutated in 30–40% of cases and encodes a structurally compact, conformationally stable protein with limited binding pockets, rendering it refractory to conventional inhibition [21–23]. Although KRAS G12C inhibitors demonstrate efficacy in a narrow subset, their applicability remains mutation-specific [24, 25]. Similarly, targeting EGFR is restricted by compensatory feedback—including HER2 amplification, MET dysregulation, and downstream PI3K/AKT/mTOR pathway mutations—that drives the development of resistance [26]. For BRAF V600E-mutant CRC, inhibitor responses are typically transient due to reactivation of the MAPK pathway [27, 28]. These collective limitations highlight the inherent constraints of targeting single nodes within signaling pathways and underscore the need for novel targets that operate at critical, multifunctional junctures of cancer cell regulation. Given these limitations, CAD emerges as a promising therapeutic target. Unlike conventional targets that modulate discrete signaling pathways, CAD functions as a multifunctional enzyme complex, orchestrating de novo pyrimidine biosynthesis while influencing tumor biology through non-canonical metabolic mechanisms [29]. Prior studies have implicated CAD in sustaining stemness and tumorigenesis in glioblastoma and promoting hepatocarcinogenesis via the β-catenin–AKT2–CAD axis [30–33]. Despite its recognized oncogenic functions in other malignancies, the role of CAD in CRC remains unexplored [34, 35]. Our comprehensive TCGA bioinformatics analysis identified significant CAD overexpression in CRC tissues, with elevated mRNA levels associated with poorer patient survival [36]. Given its dual metabolic and non-metabolic roles in tumor progression, CAD represents a tissue-specific biomarker with strong diagnostic/therapeutic potential. The identification of CAD as a pivotal oncogenic driver in CRC paves the way for the development of novel therapeutic strategies. However, effective modulation of CAD expression remains challenging. While the use of siRNA offers high-specificity gene silencing, its clinical translation is limited by poor stability in the circulation and insufficient tumor targeting [37]. Copper-based nanomaterials present a promising solution due to their tunable physicochemical properties and intrinsic bioactivity, enabling both the delivery of therapeutics and direct cytotoxicity [38]. Notably, one key mechanism of copper-mediated cell death is cuproptosis, a regulated pathway triggered by mitochondrial Cu⁺ accumulation. This process is initiated by the binding of Cu⁺ to lipoylated components of the TCA cycle, such as dihydrolipoamide S-acetyltransferase (DLAT), which is facilitated by ferredoxin 1 (FDX1), leading to aggregation of lipoylated proteins, loss of iron-sulfur cluster proteins, and proteotoxic stress [39, 40]. Nevertheless, monometallic copper therapies are constrained by TME heterogeneity [41]. To address this limitation, bimetallic nanosystems, especially those integrating nickel and copper, are gaining increasing attention due to their capacity for synergistic metal-mediated cytotoxicity [42, 43].

We developed a dual-responsive nanoplatform using hyaluronic acid-functionalized nickel-copper sulfide heterojunction nanoparticles (NCSH NPs) for CD44-mediated tumor targeting. This system enables a TME-responsive (acidic pH, high ROS) release of three components, namely, CAD-targeting siRNA (siCAD), the H₂S donor AVT-18A, and bioactive Cu⁺ ions. Mechanistically, AVT-18A-derived H₂S lowers the local pH, thereby enhancing Cu⁺ solubility and accumulation. H₂S also induces irreversible depletion of reduced glutathione (GSH), impairing detoxification and amplifying copper-mediated oxidative damage. Under these conditions, Cu⁺ can bind effectively to mitochondrial FDX1/DLAT, exceeding its homeostatic thresholds and resulting in significant cuproptosis. Concurrently, siCAD silences CAD, disrupting pathways promoting proliferation and metastasis. Furthermore, H₂S modulates immunity by inhibiting M2-like tumor-associated macrophages and promoting M1 polarization. This trimodal strategy—combining CAD silencing, H₂S-potentiated cuproptosis, and immune reprogramming—demonstrates the viability of CAD as a therapeutic target in CRC and introduces a transformative platform for a precise, multi-mechanistic treatment of solid tumors.

Materials and methods

Cell lines and cell culture

The normal human colorectal epithelial cell lin eNCM460(RRID:CVCL_0460; Cat.No.:YC-B143) was purchased from Yuanjing Biology (Guangzhou, China). The human Colorectal cancer cell lines RKO(RRID: CVCL_0504; Cat.No.: STCC10804P), SW480(RRID: CVCL_0546; Cat.No.: STCC10806P), SW620(RRID: CVCL_0547; Cat.No.: STCC10812P), HT29(RRID: CVCL_A8EZ; Cat.No.: STCC10801P), HCT116(RRID: CVCL_D1ZN; Cat.No.: STCC10803P), HCT8(RRID: CVCL_2478; Cat.No.: STCC10811P) and DLD-1(RRID: CVCL_0248; Cat.No.: STCC10810P) were all purchased from the Zishanbio (Wuhan, China). All cell lines were authenticated by short tandem repeat (STR) profiling and confirmed to be mycoplasma-free before experimentation. All cells were maintained and stored following the instructions obtained from their providers. Briefly, NCM460, RKO, SW480, SW620, HT29, HCT116, HCT8 and DLD-1 were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium (Servicebio, China), supplemented with all culture mediums contained 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin and streptomycin (Gibco, USA). All cell lines were cultured in a humidified incubator containing 5% CO₂ at 37 °C.

Total RNA extraction and qRT‐PCR

Total RNA was extracted using TRIzol Reagent (Vazyme, China), and the concentration and purity of total RNA were determined by an ultraviolet spectrophotometer (Eppendorf, Hamburg, Germany). Reverse transcriptions was performed using the PrimeScript RT Master Mix (Vazyme, China) with random primers. The cDNA amplification was performed using SYBR Green SuperMix (Vazyme, China) and ABI 7900HT Fast Real-Time PCR system (Applied Biosystems, CA, USA). GAPDH and U6 were used as internal reference genes. The results were analyzed by the 2-ΔΔCt method. All primer sequences are listed in Supplementary Table S3.

Plasmids, shRNAs and siRNAs

The expressing plasmids containing CAD genes were constructed in previous work. The shRNA lentiviral vector plasmid (pLKO-U6-PGK-copGFP-2A-PURO) was purchased from Changsha Qinke Biological Company (Changsha, China). The sgRNA lentiviral vector plasmid for FDX1 knockdown (pLenti-U6-FDX1(human)-sgRNA1-Cas9-Puro) was purchased from MIAOLING Biological Company (Wuhan, China). This plasmid features ampicillin resistance for Escherichia coli selection, includes puromycin resistance for eukaryotic selection, and also contains the green fluorescent protein gene. All siRNAs were synthesized by RIBOBIO (Guangzhou, China). The sense strand sequences are shown below.

shCAD#1:GACTTACAAGATGAGATATGG

shCAD#2:CAGCAAATTCTCTTGAACAGG

shCAD#3:CGAAAGATGGGATATAAGACC

siCAD#1:GTATGAGGGTCTCTTCTTA

siCAD#2:GCTCTAGCGTTGAATTTGA

siCAD#3:GTGATCGACTCTACTTTGA

sgFDX1:ATCCAGCGCGGGACCCAGCG

Lentiviral preparation and stable cell line construction

Lentivirus production was performed in HEK293T cells. Briefly, the lentiviral packaging plasmids and plasmid (pLKO-shCAD or pLenti-sgFDX1) were co-transfected into cells at a ratio of 5:7:10 (psPAX2: pMD2.G: plasmid). After 48 h of transfection, the culture supernatant was collected, centrifuged, and filtered through a 0.45 µm filter M to obtain purified viral particles. The purified lentivirus was used to infect HCT116, HCT8, HT29 and DLD-1 cells in the presence of polybrene (5 µg mL⁻1) at 37 °C by centrifugation at 1800 rpm for 40 min. After 48 h of infection, puromycin (1–2 µg mL⁻1) was added to the medium to select cells infected with the lentivirus. Immunoblotting with specific antibodies was performed to assess the efficiency of target protein knockdown or overexpression.

Cell Counting Kit-8 (CCK-8) and colony formation assays

For the cell proliferation assay, 1×10³ cells were seeded in 100 μL of complete culture media in 96-well plates for various time points. CCK-8 assay (Beyotime, Shanghai, China) was performed to measure cell viability according to manufacturer’s instructions. The cells were subsequently incubated at 37 °C for 2 hours, and the absorbance was measured at 450 nm by a microplate reader (Eppendorf, Hamburg, Germany). For the colony formation assay, 300 cells were inoculated into 6-well plates and cultured at 37 °C for 14 days. The cells were then fixed with 4% paraformaldehyde (PFA) and stained with 0.1% crystal violet (Beyotime, Shanghai, China).

In vitro migration and invasion assay

Cell migration and invasion abilities were assessed using Transwell chambers (Corning, USA). Briefly, the CRC cells were seeded in the upper chamber with serum-free (RPMI) 1640 and the lower chamber was filled with culture medium with 10% FBS as an attractant. After 48 h, the cells in the Matrigel (Sigma-Aldrich, USA) were fixed by 4% PFA and stained with 0.1% crystal violet (Beyotime, Shanghai, China). The cells were observed and photographed under light microscope, and 10 fields were selected to count the cells to reflect cell mobility. Triplicate independent experiments were performed.

Flow cytometry (FCM) assay for cell apoptosis

The Annexin V-FITC/Propidium iodide (PI) double staining kit (Vazyme, Nanjing, China) was employed to examine cell apoptosis in keeping with the protocol provided by the manufacturer. as previously described, briefly, the cells were harvested and stained with Annexin V-FITC and PI respectively, the FCM (BD FORTESSA, USA) was employed to measure cell apoptosis ratio. The FITC was detected at 530 nm, and PI was detected at 575 nm, respectively.

In vitro macrophage polarization and evaluation

The ability of the NCSH NP platform to induce macrophage polarization was assessed using the murine macrophage cell line RAW264.7. Cells were seeded at a density of 1 × 10⁵ cells per mL and treated with 200 µg/mL of NCSH NPs for 24 hours; lipopolysaccharide (LPS) was used as a positive control for M1 polarization. For flow cytometry, the cells were washed with PBS and detached by treating 1 mL of 0.25% trypsin-EDTA for 3 min. 1% BSA was added to the cell suspension for blocking for 30 min at room temperature. Then the antibodies were added (TruStain FcX™ (anti-mouse CD16/32) Antibody, Biolegend, 101320; APC/Cyanine7 anti-mouse F4/80 Antibody, Biolegend, #123118; FITC anti-mouse/human CD11b Antibody, Biolegend, #101206; APC anti-mouse CD86 Antibody, Biolegend, #105012; PE anti-mouse CD206 (MMR) Antibody, Biolegend, #141706) and maintained for an hour at room temperature. After centrifugation at 1550 rpm for 5 min, the cell pellets were resuspended in 1 mL PBS, and the samples were analyzed by a flow cytometer (AttuneNxT Flow Cytometer, Thermo Fisher Scientific, USA).

Protein extraction and western blot

Protein extraction and western blot analysis were performed as previously described. Antibodies were as follows: CAD (16617-1-AP; Proteintech), Bcl-2 (381702; Zenbio), BAX (R380709; Zenbio), Tubulin (TA503129;Origene), DLAT (HA721267; HUABIO), FDX1 (HA721329; HUABIO).

Synthesis of NCSH NPs

The Hyaluronic Acid-Functionalized Nickel-Copper Sulfide Heterojunction Nanoparticles (NCSH NPs) were synthesized via a one-pot hydrothermal method. Briefly, 12.5 mL of aqueous solution containing 2.0 mM CuCl₂ and 0.5 mM NiCl₂ was mixed with 100 mg sodium hyaluronate (MW~50,000) under stirring. Following the addition of 100 μL of 1 M N₂H₄, the solution was stirred for 20 min. Next, Na₂S·9H₂O solution (320 mg/mL) was introduced, and the mixture was heated at 60 °C for 2.5 h. The resulting NCSH NPs were washed three times with ethanol via centrifugation at 10,000 rpm and stored.

Synthesis of NCSH@siCAD&AVT-18A

After mixing 5.0 mL of NCSH nanoparticle dispersion (5.28 mg/mL) with an equal amount of AVT-18A cryopreservation solution (1.0 mg/mL) via sonication (100 W, 40 kHz), mechanical agitation was performed for 24 h at ambient temperature. After the reaction was complete, samples were centrifuged for 15 min at 20,000 ×g at 4℃, and precipitates were washed three times in deionized water to yield NCSH@AVT-18A nanocomposites. The NCSH@AVT-18A nanoparticles were then filtered through 0.22 μm membranes and ultrasonically dispersed (100 W, 40 kHz) to 2.5 mg/mL in DEPC-treated water. An siRNA working solution (0.25 μM) was then prepared using DEPC water. Equal nanoparticle suspension and siRNA solution volumes were then mixed together, ultrasonicated (100 W, 40 kHz, 30 min), and incubated for 12 h at 37℃ with orbital shaking (200 rpm). Complexation occurred via capillary adsorption and van der Waals interactions. Unbound siRNA was removed via centrifugation, and NCSH@siCAD&AVT-18A was resuspended in DEPC water and stored at −20℃.

Characterization of NCSH NP complex

The morphology of NCSH NP was detected with a transmission electron microscope (TEM, Tecnai G2 F20, USA). The particle size and zeta potential values of NCSH NP were respectively determined with a Zetasizer Nano ZS (Malvern Nano series, Malvern, UK).

Cellular uptake of NCSH@AVT-18A&siCAD

The cellular internalization of the NCSH@AVT-18A&siCAD nanocomplex was evaluated in HCT116 cells. For visualization, NCSH@AVT-18A&siCAD was labelled with Cy7 dye. Subsequently, NCSH@AVT-18A&siCAD was incubated with HCT116 cells in serum-containing medium for 6 h. Following incubation, cells were washed with phosphate-buffered saline (PBS), fixed in 4% paraformaldehyde (PFA) solution for 15 min, counterstained with 4',6-diamidino-2-phenylindole (DAPI), and imaged using confocal microscopy with a 633 nm laser excitation (Carl Zeiss LSM900).

In vitro cytotoxicity assays

HCT116 cells were seeded into 96-well plates at a density of 5×10³ cells/well. Upon reaching 50% confluency, distinct treatment groups were administered. Following a 6-hour incubation, the medium was replaced with 100 μL of serum-containing complete medium, and the cells were cultured for an additional 72 hours at 37 °C. Subsequently, every 24 hours post-treatment initiation, 10 μL of CCK-8 solution was added to each well, followed by a further 2-hour incubation period. Absorbance was then measured at 450 nm using a microplate reader (Eppendorf, Germany).

In vivo studies

Animals

BALB/c nude mice (4–6 weeks, female) were obtained from Hunan SJA Laboratory Animal Co., Ltd. (Changsha, China) and raised in a specific pathogen-free environment. All animal studies were approved by the Department of Animal Research, Central South University. The animal studies were not blinded. All animals were included in the analysis.

Biodistribution of NCSH NP complex

We used Cy7 to label the NCSH NP complex to minimize the interference from autofluorescence in the in vivo imaging studies. To study the tumor-targeting ability of NCSH NP complex, BALB/c nude mice with subcutaneous tumors were injected with free-CY7, NSCH@siNC and NSCH@AVT-18A&siCAD (40 µg Cy7-siRNA per mouse equivalent, 150 µL of 20 µM stock) through tail vein, respectively. At 0, 2, 6, 12, 24 and 48 h post injection, the fluorescent images of the mice were imaged with IVIS Spectrum in vivo imaging system (PerkinElmer). After 48 h, mice were killed, the major organs and tumors were dissected, and the ex vivo fluorescent images were acquired using the same system.

Antitumor effects in vivo

We conducted two colorectal tumor models, including subcutaneous tumor model, pulmonary metastatic tumor model and orthotopic tumor model to investigate antitumor effects of the NCSH NP complex.

In vivo therapeutic efficacy in subcutaneous tumor models

We implanted 1×10⁶ HCT116 shNC, shCAD, vector, or OE-CAD cells into the flank of BALB/c nude mice to establish subcutaneous tumors. Tumor growth was tracked every other day, with volume calculated using the formula V (volume) = (length × width²)/2. Upon study completion, the mice were euthanized; tumors were then excised and weighed.

We injected 1×10⁶ HCT116 cells into the flank of Balb/c nude mice to form subcutaneous tumors. When the tumor sizes reached about 5 mm in diameter, mice were sorted to give nearly identical mean tumor sizes, and different NCSH NP complex (4 µg siRNA per mouse equivalent, 15 µL of 20 µM stock) were injected peritumoral or intratumoral injections (n = 4 mice per group). Saline was injected as the control group. The injections were administered every two days for a total of six times. Tumor volume was calculated as follows: V (volume) = (length × width²)/2. Upon completion of the treatment regimen, the mice were euthanized, the tumor xenografts were excised, weighed and imaged.

In vivo pulmonary metastatic tumor model

We intravenously injected 1×10⁶ HCT116 shNC, shCAD, vector, or OE-CAD cells into BALB/c nude mice via tail vein to establish a pulmonary metastatic model. Throughout the experimental period, we monitored the body weights of the mice. On day 28, the mice were euthanized, and lungs were collected for ex vivo photography.

Orthotopic tumor model

We created orthotopic tumors in Balb/c nude mice to mimic the natural colorectal cancer microenvironment. Briefly, mice were anaesthetized and a midline incision was made to expose the caecum. Furthermore, HCT116 cells (5×10⁵) were orthotopically injected into the caecum wall of mice and treated with NCSH NP complex every 3 days for 15days after 9 days. Those mice were sacrificed after 16 days with injection.

In vivo toxicity

For histology analysis, 48 h after the last intravenous injection of drugs, major organs (heart, liver, spleen, lung and kidney) and tumor tissues of the mice were fixed, sectioned and stained with hematoxylin and eosin (H&E). The slices were observed by microscope (Leica DMI 6000B). At 48 h after the last intravenous injection of drugs, blood was drawn from the venous plexus of the eyes of the mice. Blood samples were immediately centrifuged at 3,000 g for 5 mins at 4 °C, and the supernatant was collected for hematological analysis. For blood biochemical analysis, ALT, AST, blood urea nitrogen and creatinine values were measured using Modular analytics (Roche, Germany), as indicators of hepatic (ALT and AST) and renal (blood urea nitrogen and creatinine) functions.

TUNEL, Ki-67 and Immunofluorescence (IF) staining

The analysis of apoptosis in vivo was performed by TUNEL staining based on the standard scheme. The paraffin-embedded tissue samples were separated and the antigens were recovered for Ki-67 detection according to the manufacturer’s test kit instructions. As for IF staining, the paraffin-embedded tumor tissue was deparaffinized for antigen repair. IF staining was done with anti-CAD, FDX1 and cleaved caspase-3 according to the standard protocol. The nucleus was stained with DAPI, followed by the use of fluorescent microscope to obtain images.

Statistical analysis

All data are presented as the mean±standard deviation. Data analysis was conducted using GraphPad Prism 8. Two-tailed, unpaired Student's t-tests and one-way ANOVA were performed to calculate the statistical significance between the different groups, and statistical significance was considered within the groups at a significance threshold of p<0.05.

Results

CAD is overexpressed in CRC and exhibits clinical significance

To investigate the pathological significance of CAD in CRC, we performed a comprehensive multi-database bioinformatics analysis. RNA-seq data from The Cancer Genome Atlas (TCGA, https://www.cancer.gov/ccg/research/genome-sequencing/tcga/using-tcga-data/citing) revealed significantly elevated CAD transcript levels in CRC tumor tissues compared to normal colonic tissues (p < 0.01, Fig 1A). This finding was corroborated by pan-cancer expression analysis using the GEPIA platform, which demonstrated pronounced CAD upregulation in both colon adenocarcinoma (COAD) and rectum adenocarcinoma (READ) cohorts (Fig 1B). Kaplan-Meier survival analysis via the KM Plotter database indicated that higher CAD expression correlates with reduced overall survival (HR = 1.37; 95% CI: 1.12–1.68; p = 0.0013; Fig 1C). To validate these bioinformatics findings, we conducted qPCR analysis on 34 paired CRC and adjacent non-tumor tissues from the Xiangya Hospital biobank. CAD mRNA levels were significantly higher in tumor samples (p < 0.001, Fig 1D, Table S1). Immunohistochemical (IHC) staining of a tissue microarray (n = 90 pairs) followed by quantitative analysis using ImageJ software confirmed elevated CAD protein expression in CRC tissues (Fig 1E-F). Multivariate Cox regression analysis integrating protein expression and clinical outcome data identified high CAD expression as an independent prognostic factor (HR = 1.57, p < 0.02; Fig. 1G, Table S2). Western blotting of 34 matched CRC and adjacent normal tissues also revealed markedly increased CAD protein levels in tumors (p < 0.01, Fig 1H). In vitro analysis further supported these results, with CRC cell lines displaying substantially higher CAD expression compared to normal colon epithelial cells (Fig 1I). Together, these findings underscore CAD’s oncogenic potential in CRC, linking its upregulation to enhanced malignancy and poor prognosis, thereby supporting its candidacy as a therapeutic target.

Fig. 1.

CAD is Overexpressed in CRC and Exhibits Clinical Significance. A Differential Expression Levels of CAD in Tumor versus Normal Tissue of COAD and READ Using RNA-seq Data from the TCGA (Wilcoxon rank sum test, The asterisk (*) numbers from 1 to 4 indicate p-values less than 0.05, 0.01, 0.001, and 0.0001, respectively.). B Differential Expression Levels of CAD in Tumor versus Normal Tissue of COAD and READ shown in TCGA(One-Way ANOVA, The "*" symbol indicates that the p-value is less than 0.05.). C Kaplan-Meier analysis revealed the prognostic values of CAD. D Relative mRNA levels of CAD in 34 paired CRC samples. E-F The representative expression (E) and IHC analysis (F) of CAD in human CRC tissue compared to normal intestinal tissue of a tissue microarray (n=90 pairs). G Multivariate Cox regression analysis integrating protein expression and clinical outcome data. H Relative protein levels of CAD in four paired CRC samples. pa: para-carcinoma tissues; ca: cancer tissues. I Relative protein levels of CAD in normal intestinal cell and CRC cell line samples

Verification of CAD as a target for nanoplatform-based treatment of CRC

To assess the functional role of CAD in tumor progression, we selected HCT116 and HCT8 cells with relatively high CAD expression, as well as DLD-1 and HT29 cells with relatively low CAD expression. Both CAD knockdown and overexpression cell lines were constructed in these cells using lentivirus and plasmid transfection, respectively. The efficiency of CAD knockdown and overexpression was verified (Figures S1A and S1B), and sh-CAD #1 was selected for further studies based on its superior efficacy. Functional assays revealed that CAD knockdown significantly reduced cellular proliferation, as evidenced by colony formation (Fig 2A and Figure S1C) and 5-ethynyl-2′-deoxyuridine (EdU) incorporation assays (Fig 2B and Figure S1D), whereas CAD overexpression had the opposite effect on DLD-1 and HT29 cells. Transwell migration and Matrigel invasion assays showed a marked decrease in the number of migratory and invasive cells upon CAD silencing. Conversely, overexpression of CAD significantly increased both cell invasion (Fig 2C and Figure S1E) and migration (Fig 2D and Figure S1F). Analysis of apoptosis revealed a complex role for CAD. Knockdown of CAD in HCT116 and HCT8 cells resulted in a significant increase in apoptosis, whereas overexpression of CAD in HT29 and DLD-1 cells conferred resistance to apoptosis (Fig 2E and Figure S1G). The results were further verified by Western blotting, which showed decreased expression of the anti-apoptotic marker Bcl-2 and increased expression of the pro-apoptotic protein Bax following CAD knockdown (Fig 2F). These results demonstrate that CAD functions as a multifunctional oncogene in CRC by promoting cell proliferation, enhancing motility, and suppressing apoptosis.

Fig. 2.

Verification of CAD as a Target for Nanoplatform-Based Treatment of CRC. A Representative images of the colony formation assay in CRC cells (n=3). B Representative images of the EdU assay in CRC cells (n=3). C Representative images of the migration assay in CRC cells (n=3). D Representative images of the invasion assay in CRC cells (n=3). E Representative images of the apoptosis assay in CRC cells (n=3). F Expression changes of CAD, Bcl-2 and Bax in CRC cells transfected with shNC or shCAD (n=3). Data are presented as mean ± SD n.s. for nonsignificances, *P < 0.05, **P < 0.01, ***P < 0.001

To evaluate the tumor-suppressive effects of CAD silencing in vivo, we utilized the stable CAD-knockdown and overexpression cell lines described previously. In the subcutaneous xenograft model, tumors derived from CAD-silenced cells exhibited significantly reduced volumes and weights, with growth inhibition of up to 70% compared to controls (Fig 3A). IHC staining of harvested tumors revealed lower Ki67 proliferation indices and elevated cleaved caspase-3 levels, indicative of enhanced apoptotic activity (Figure S2A), while CAD overexpression elicited the opposite effect (Fig 3B and Figure S2A). In a pulmonary metastasis model, mice injected with CAD-silenced CRC cells developed 57% fewer lung nodules (p < 0.001 for both HCT116 and HCT8 groups), alongside reduced expression of Ki67 and CAD proteins (Fig 3C), while CAD overexpression resulted in the opposite effect (Fig 3D). These findings reinforce the pivotal role of CAD in promoting CRC growth and metastasis in vivo.

Fig. 3.

CAD promotes tumor growth and metastasis in vivo. A Tumor growth, tumor weight, tumor volume in mice injected with shNC or shCAD CRC cells (n = 4). B Tumor growth, tumor weight, tumor volume in mice injected with Vector or OE-CAD CRC cells (n = 4). Stable CRC cell lines were injected into nude mice through tail vein (n = 4 per group). C–D The tumors in lung were collected immediately after euthanized, and the number of tumor nodules was counted. Representative images, HE images indicates the tumor nodule and statistical quantifications. Data are presented as mean ± SD n.s. for nonsignificances, *P < 0.05, **P < 0.01, ***P < 0.001

Synthesis and characterization of the targeted nanoformulation

In light of the therapeutic limitations associated with monotherapies, we designed a multifunctional nanoplatform based on nickel-copper sulfide to co-deliver siCAD and AVT-18A. This system is tailored to exploit the acidic and glutathione (GSH)-rich tumor microenvironment (TME), facilitating site-specific gene silencing and hydrogen sulfide (H₂S) release for combinatorial metabolic disruption in CRC cells. Surface functionalization with hyaluronic acid enables selective uptake via CD44-mediated endocytosis, thereby improving tumor-targeting accuracy and minimizing off-target toxicity.

NCSH NPs were synthesized using CuCl₂, NiCl₂, and sodium hyaluronate (NaHA) via a Kirkendall diffusion-based approach. Transmission electron microscopy (TEM) revealed that nanoparticle size increased with higher Na₂S concentrations, with the NCSH3 variant exhibiting a well-defined hollow core and optimal size (<100 nm) for tumor penetration (Figures S3A). Inductively coupled plasma-optical emission spectrometry (ICP-OES) analysis indicated that NCSH3 contained 36.7% copper and 13.2% nickel by mass. Scanning electron microscopy (SEM) confirmed a uniformly porous surface, conducive to drug loading (Figures S3B).

X-ray diffraction (XRD) validated the formation of heterojunctions composed of CuS and NiS, while X-ray photoelectron spectroscopy (XPS) detected characteristic peaks corresponding to Cu, Ni, S, F, and O, confirming successful hyaluronic acid modification (Figures S3C-D). Fourier-transform infrared (FTIR) spectroscopy further supported HA anchoring through coordination bonding (Figures S3E). Among the three siCAD sequences screened, siCAD-2 achieved the highest silencing efficiency (>50%) and was used for formulation (Figures S3F). AVT-18A and siCAD-2 were loaded into NCSH3 via ultrasonication and magnetic stirring. UV-Vis spectroscopy determined an AVT-18A loading efficiency of 12.8% and encapsulation efficiency of 67.5%, with 100% siRNA incorporation (Figures S3G). Post-loading, TEM images showed the nanoparticles retained spherical morphology with denser cores (Fig 4A-B). Energy-dispersive X-ray spectroscopy (EDX) detected elemental signatures of the nanocarrier (Cu, Ni, S), AVT-18A (F), and siRNA (N, O) (Fig 4C). Dynamic light scattering (DLS) demonstrated increased hydrodynamic diameter after each loading step, aligning with TEM observations (Fig 4D-E). Progressive reductions in zeta potential confirmed sequential cargo attachment, while nitrogen adsorption-desorption analysis revealed decreased surface area and pore volume due to mesoporous channel occupation (Fig 4F).

Fig. 4.

Evaluation of Nanoplatform TME Responsiveness and Drug Release. A-B TEM images and zeta analysis of NCSH@AVT-18A (A) and NCSH@AVT-18A&siCAD (B). C EDX analysis of NCSH@AVT-18A&siCAD.D-E DLS analysis of NCSH, NCSH@AVT-18A and NCSH@AVT-18A&siCAD.F Nitrogen adsorption-desorption analysis of NCSH and NCSH@AVT-18A&siCAD.G Stability of NCSH@AVT-18A&siCAD in DMEM(10% FBS). H-I Hemolysis analysis (H) and representative images (I) of NCSH@AVT-18A&siCAD. J-K In vitro release analysis of Ni (J) and Cu (K) of NCSH@AVT-18A&siCAD. L-M TMB reagent chromogenic analysis of NCSH@AVT-18A&siCAD.N DTNB reagent decolorization analysis of NCSH@AVT-18A&siCAD

Evaluation of nanoplatform TME responsiveness and drug release

The engineered inorganic nanocomposite offers robust protection to AVT-18A and siCAD against enzymatic degradation, while enabling precise release within the TME. Following composition into NCSH@AVT-18A&siRNA, the colloidal stability was sustained in DMEM with 10% FBS at 37 °C for 7 days, attributed to strong electrostatic repulsion from its highly negative surface charge (Fig 4G). Hemolysis rates remained below 2.0% across concentrations from 25 to 200 μg/mL, confirming excellent blood compatibility (Fig 4H-I). TEM revealed the nanoparticle core remained structurally intact at pH 7.4, began degrading at pH 6.7, and fully collapsed at pH 6.0, demonstrating reliable pH-triggered degradation tailored for the acidic TME (Figures S3H).

The nanoplatform displayed dual-stimulus sensitivity to both acidic pH and high GSH levels. In a simulated CRC microenvironment (pH 6.7), cumulative release of Ni²⁺ and Cu²⁺ ions reached 62.3 ± 3.8% and 65.1 ± 4.2% within 24 hours, significantly higher than at physiological pH (4.7 ± 0.9% and 3.2 ± 0.6%) (Fig 4J-K). To examine synergistic functionality, seven experimental groups were assessed: PBS (I), free AVT-18A (II), naked siCAD (III), NCSH@siNC (IV), NCSH@AVT-18A (V), NCSH@siCAD (VI), and NCSH@siCAD&AVT-18A (VII). A WSP 1 probe confirmed H₂S production in groups II and IV–VII (Figure S3I). In acidic buffers with H₂O₂, robust TMB absorption at 652 nm indicated Fenton-like catalytic activity (Fig 4L). Kinetic assays demonstrated increasing absorbance at 650 nm over time and with increasing nanodrug dosage, verifying that metal ion release drives ROS generation (Fig 4M). 5,5'-Dithiobis-(2-nitrobenzoic acid) (DTNB) assays further showed efficient GSH depletion by NCSH@AVT-18A&siRNA at pH 6.7, indicated by significant absorbance reduction at 412 nm (p < 0.01) (Fig 4N). Together, these results confirm that pH-triggered ion release, and ROS/GSH pathway disruption are tightly coordinated in a spatiotemporally controlled manner. These data demonstrate the high loading capacity of the system, as well as its selective targeting and controlled release features under TME conditions.

Examination of cellular uptake and in vitro antitumor mechanisms

CY7-labeled NCSH@AVT-18A&siRNA demonstrated a time-dependent fluorescence increase in treated cells, confirming effective cytoplasmic delivery. Comparative fluorescence imaging between CD44-deficient 293 T cells and CD44-overexpressing HCT116 cells revealed selective accumulation in HCT116—validating HA-CD44-mediated uptake (Fig 5B). The optimal working concentration is determined by CCK-8 (Supplemental Figure I-K). Functional assays showed that group VII exhibited the greatest reduction in proliferation (CCK 8) (Fig 5C), the highest PI-positive cell count (Calcein AM/PI) (Fig 5D), significant mitochondrial membrane depolarization (JC-1) (Fig 5E), elevated ROS (DCFH DA) (Fig 5F), severe mitochondrial damage (TEM) (Fig 5G), maximal ATP depletion (Fig 5H) and H₂S(Supplemental Figure 3 N). Among treated groups, GSH depletion was most pronounced in group VII (Fig 5I). Western blot analysis revealed decreased levels of cuproptosis markers DLAT and FDX1 (Fig 5J). Immunofluorescence (IF) on treated cells, which revealed prominent aggregation of lipoylated DLAT (Fig 5K). By establishing FDX1 knockdown cell lines, we demonstrated that cuproptosis constitutes the primary mechanism underlying the drug's effect (Figure S5J, L). RAW264.7 macrophages were co-cultured with NCSH@siNC and NCSH@AVT-18A&siRNA for one day, after which flow cytometry showed that the percentage of cells expressing CD86 was significantly higher in the NCSH@siNC (40.35%) and NCSH@AVT-18A&siCAD (33.9%) groups compared to the PBS group (1.954%). Meanwhile, the proportion of CD206-positive cells was markedly higher in the NCSH@siNC (1.33%) and NCSH@AVT-18A&siCAD (15.8%) groups compared to the PBS group (0.65%) (Supplemental Figure 3M). These observations confirm that the nanomaterial induces intracellular acidification and copper-dependent cell death through H₂S and copper ion release, while concurrently suppressing CAD expression and polarizing macrophages toward a more anti-inflammatory phenotype.

Fig. 5.

Examination of Cellular Uptake and In Vitro Antitumor Mechanisms. A Representative CLSM images of HCT116 cells stained with Cy7-labeled NCSH@AVT-18A&siCAD. B Comparative fluorescence images of 293 T cells and HCT116 cells stained with Cy7-labeled NCSH@AVT-18A&siCAD. C Cell proliferation assay for HCT116 cells with various treatments. D Live/dead staining of HCT116 cells after various treatments. Green and red fluorescence represented live and dead cells, respectively. E Mitochondrial membrane potentials of HCT116 cells were determined by JC-1 assay after various treatments. F Intracellular ROS level after treatment visualized by DCFH-DA. G Scanning electron microscope images of mitochondria in HCT116 cells treated with PBS or NCSH@AVT-18A and siCAD. H The ATP secretion of HCT116 cells with various treatments (n=3). I Intracellular GSH contents of HCT116 cells with various treatments (n=3). J Expression changes of DLAT and FDX1 in HCT116 cells with various treatments. K Representative CLSM images of HCT116 cells stained with various treatments. Data are presented as mean ± SD n.s. for nonsignificances, *P < 0.05, **P < 0.01, ***P < 0.001

Assessment of in vivo antitumor activity and biodistribution

In subcutaneous xenograft tumor-bearing mice, CY7-labeled NCSH@siNC and NCSH@siCAD&AVT-18A exhibited peak tumor fluorescence at 12–24 h post-injection, exceeding that achieved by free CY7 (Fig 6A-D). Ex vivo fluorescence analyses confirmed that nanodrug groups presented with higher levels of tumor fluorescence (Figure 6E). The strongest antitumor efficacy was evident in the NCSH@siCAD&AVT-18A group revealed the strongest antitumor efficacy (Figure 6F). Similar findings were evident with respect to tumor weight and volume (Figure 6G, Figure S4A). IHC analyses also indicated more extensive tumor necrosis, reduced Ki-67 expression, and higher levels of cleaved caspase-3 levels in the NCSH@siCAD&AVT-18A group (Figure S4B). Furthermore, the in vivo safety profile of the compound was thoroughly assessed by collecting blood and organ samples from healthy mice at 1, 4, and 7 days post-injection (Figure S6-7). Quantitative analysis of Cu and Ni levels in the aforementioned critical organs via ICP-MS at days 1, 7, and 28 revealed a time-dependent clearance of both metals from the organs over the 28-day period. These ICP-MS data provide critical insights into the nanodrug's biodistribution and elimination kinetics (Figure S8).

Fig. 6.

Assessment Antitumor Activity and Biodistribution in subcutaneous tumor model. A Schematic illustration of subcutaneous tumor model establishment and treatment. B-D Representative whole-body bioluminescence images and ex vivo images of mice with various treatments were measured via IVIS imaging system (B) and quantification of was shown as mean ± SD (n = 4 independent experiments) (C-D). E The quantification of representative ex vivo images of tumors were captured using the IVIS imaging system was shown as mean ± SD (n = 4 independent experiments). F-G Representative images of the subcutaneous tumors (F) and tumor weight (G). Data are presented as mean ± SD n.s. for nonsignificances, *P < 0.05, **P < 0.01, ***P < 0.001

An orthotopic CRC model further validated treatment efficacy: drug-treated groups II–VII showed greater tumor suppression than PBS control, with group VII exhibiting the most significant response (Figures 7A-B). Histological evaluation (H&E) staining confirmed increased tumor cell necrosis in treated groups. Safety studies, including evaluation of major organs and serum biochemistry, revealed no significant tissue damage or biochemical abnormalities, confirming excellent in vivo biocompatibility (Figure 7C, Figures S5A-B).

Fig. 7.

Assessment Antitumor Activity and Biodistribution in orthotopic model. A Schematic illustration of orthotopic model establishment and treatment. B Representative images of the orthotopic tumors.C Representative images of H&E staining (top) and immunofluorescence (bottom) of CAD

Discussion

The study findings demonstrate the pivotal role of CAD as an oncogenic driver and prognostic biomarker in colorectal cancer. The results provide compelling clinical and functional evidence that CAD overexpression leads to CRC progression, shown by its promotion of tumor cell proliferation, invasion, and resistance to apoptosis, phenotypes correlated directly with poor patient survival. The findings indicate the potential of CAD as a therapeutic target, particularly given its dual role in both metabolic (de novo pyrimidine synthesis) and non-metabolic oncogenic pathways. While the involvement of CAD in other malignancies has been reported, the present study represents the first comprehensive verification of its tumor-promoting functions in CRC pathogenesis.

To translate these findings into clinical utility, a tumor microenvironment-responsive nanoplatform (NCSH@siCAD&AVT-18A) was engineered. This nanoplatform simultaneously addressed three key therapeutic challenges, namely, target specificity, tumor-selective delivery, and mitigation of resistance development. The nickel-copper sulfide heterojunction architecture exploits the acidic and reductive properties of the TME to orchestrate the spatially controlled release of three synergistic components: CAD-silencing siRNA, the hydrogen sulfide donor AVT-18A, and bioactive copper ions. This tripartite strategy induces cell killing in three complementary ways, specifically, siRNA-mediated suppression of CAD disrupts nucleotide metabolism and tumorigenic signaling, while H₂S-triggered acidification and glutathione depletion amplify the solubility and mitochondrial accumulation of copper ions. The resulting cuproptotic cascade, mediated by FDX1/DLAT aggregation and proteotoxic stress, synergizes with CAD silencing to overcome compensatory survival pathways that frequently undermine conventional targeted therapies. However, a limitation of this immunophenotyping is the lack of functional validation, such as cytokine profiling or T-cell recruitment assays, to confirm the functional state of the repolarized macrophages. Furthermore, the mechanism underlying antigen-presenting cell activation remains unclear and warrants further investigation.Nevertheless, a comprehensive long-term toxicity assessment of the nanodrug was not conducted in this study, representing a limitation that should be addressed in future preclinical development to fully evaluate its safety profile.

The treatment strategy for colorectal cancer proposed in this study offers significant advantages. Unlike the application of KRAS or EGFR inhibitors, which have limitations and can develop adaptive resistance due to specific mutations, this platform relies on cuproptosis, traditionally induced by exogenous copper ions [44]. The inclusion of AVT-18A ensures dissociation of the materials triggered by the acidification of the tumor microenvironment, driving a cascade of copper death reactions. This process activates multiple synergistic regulatory pathways that together induce the death of tumor cells. Notably, the use of H₂S as a promoter of cuproptosis in colorectal cancer is a novel approach. Additionally, this method successfully overcomes the inability of immune checkpoint inhibitors to treat microsatellite-stable colorectal cancer. Mechanistic studies showed that the bimetallic nanosystem significantly enhanced the Fenton reaction, and immunomodulatory effects mediated by H₂S through the synergistic actions of nickel and copper. The latter was confirmed by the polarization of M2 macrophages to the M1 type (confirmed by flow cytometry). However, this immunophenotypic shift requires further functional validation through cytokine profiling and T-cell recruitment assays to fully confirm the functional reprogramming of the macrophages and its contribution to the antitumor immune response. Despite these immunomodulatory effects and the resulting effective inhibition of tumor growth in both subcutaneous and orthotopic models, the comprehensive long-term biodistribution and toxicity profile of the nanodrug remain to be fully established, which is a crucial next step for translational development. Nonetheless, the strategy, leveraging the CD44-targeting characteristics of the hyaluronic acid coating, exhibited negligible systemic toxicity in our studied window, underscoring its promising therapeutic index.

The innovative value of this study requires contextualization within translational considerations. While the NCSH platform demonstrated superior preclinical biocompatibility, its clinical application would require rigorous assessment of long-term metal ion biodistribution and potential cumulative toxicity [45, 46]. Furthermore, the durability of CAD silencing and copper-mediated cytotoxicity in heterogeneous tumor populations warrants investigation. It is important to note, however, that our current study has certain limitations that point to valuable future research directions. The release kinetics and therapeutic efficacy were characterized under normoxic conditions; the potential influence of hypoxic tension, a key feature of solid tumors, on drug release behavior and treatment outcomes remains to be systematically investigated. Future studies should investigate the ability of this strategy to overcome resistance in aggressive CRC subtypes, such as BRAF-mutant or consensus molecular subtype 4 (CMS4) tumors, and whether it synergizes with standard chemotherapies. Additionally, while we have experimentally confirmed that the nanodrug-induced cell death is primarily attributed to FDX1/DLAT disruption and cuproptosis, comprehensive long-term toxicity studies are necessary to fully evaluate the safety profile of this nanoplatform. Beyond the observed macrophage polarization from M2 to M1 phenotype, functional validation through cytokine profiling and T-cell recruitment assays, as well as elucidation of the precise mechanism of antigen-presenting cell activation, would strengthen our understanding of the immunomodulatory effects. From a translational perspective, challenges such as the scalability of nanoparticle synthesis and the potential immunogenicity associated with the hyaluronic acid coating need to be carefully addressed. The modular design of the nanocarrier also permits adaptation to metabolic targets other than CAD, offering a versatile blueprint for the treatment of solid tumors.

In summary, this study elucidated the central regulatory role of CAD in the malignant progression of colorectal cancer and introduced a novel nanotechnology-based synergistic therapeutic strategy that enabled precise intervention by targeting nucleotide metabolism and disruption of metal ion homeostasis in tumor cells. The development of a tumor microenvironment-responsive nanoplatform enabled the integration of CAD gene silencing with hydrogen sulfide (H₂S)-enhanced cuproptosis, achieving unique and spatially precise tumor elimination, specifically, through blocking the metabolic basis of tumor proliferation through specific gene editing and triggering spatially selective programmed cell death by the disruption of metal ion homeostasis, resulting in significant anti-tumor synergies through dual pathways. This "gene-metal drug" model offers a revolutionary approach to treating colorectal cancer and other malignant tumors resistant to traditional therapies. Its significance lies in overcoming the limitations of single therapies through synergistic epigenetic regulation and metal-associated effects, providing an innovative solution to clinical resistance. The controllable spatiotemporal release achieved by the TME-responsive platform can significantly reduce off-target toxicity, advancing precision medicine. This is the first approach involving integration of metal-ion regulation with metabolic reprogramming, deepening the understanding of cuproptosis and establishing a theoretical framework for "metabolism-metal" dual-targeted therapy. The targeted pathways are common to a variety of malignant tumors, and the platform technology can be extended to treat hard-to-treat cancers, such as liver and pancreatic cancers, indicating significant potential for clinical translation.

Conclusion

This study elucidates CAD’s oncogenic role in CRC and demonstrates that CAD overexpression is associated with poor clinical outcomes. Functional assays establish that CAD inhibition suppresses tumor growth, invasion, and metastasis, while inducing apoptosis. We design a nickel-copper sulfide nanocarrier (NCSH NPs) co-delivering siCAD and AVT-18A, triggering cuproptosis and apoptosis in a TME-responsive manner. The combined therapy shows superior antitumor efficacy in both subcutaneous and orthotopic CRC models. Our findings position CAD as a promising target for CRC treatment and, by extension, showcase NCSH NPs as a versatile nanoplatform for precision oncology.

Abbreviations

CAD

Carbamoyl-phosphate synthetase II/Aspartate transcarbamylase/Dihydroorotase

CMS4

Consensus Molecular Subtype 4

COAD

Colon Adenocarcinoma

CRC

Colorectal Cancer

DLAT

Dihydrolipoamide S-acetyltransferase

DLS

Dynamic Light Scattering

DTNB

5,5'-Dithiobis-(2-nitrobenzoic acid)

EdU

5-ethynyl-2′-deoxyuridine

EPR

Enhanced Permeability and Retention

EDX

Energy-Dispersive X-ray Spectroscopy

FDX1

Ferredoxin 1

FTIR

Fourier-Transform Infrared

GSH

Glutathione

H&E

Hematoxylin and Eosin Staining

HA

Hyaluronic Acid

H₂S

Hydrogen Sulfide

IHC

Immunohistochemistry

ICP-OES

Inductively Coupled Plasma-Optical Emission Spectrometry

MSS

Microsatellite-Stable

NaHA

Sodium Hyaluronate

NCSH

Nickel-Copper nano-Heterojunction architecture

NCSH NPs

Hyaluronic Acid-Functionalized Nickel-Copper Sulfide Heterojunction Nanoparticles

NPs

Nanoparticles

READ

Rectum Adenocarcinoma

SEM

Scanning Electron Microscopy

siCAD

CAD-targeting siRNA

TCGA

The Cancer Genome Atlas

TEM

Transmission Electron Microscopy

TME

Tumor Microenvironment

XPS

X-ray Photoelectron Spectroscopy

XRD

X-ray Diffraction

Author contributions

Y.X. and Y.L. contributed equally to this work. Y.X., Y.L., M.Y., Z.Z. and S.Z. performed laboratory experiments. Y.X. and W.Z. analyzed bioinformatics data. Y.X. and Y.L. prepared the figures. Y.X., Y.L. and W.Z. made the statistical analysis. J.P., Z.X., W.L., Y.X. and Y.L. supervised the research. Y.X. and Y.L. wrote the manuscript. J.P., Z.X. and W.L. conceived the project. All authors have read and approved the article.

Funding

This work was supported by the Natural Science Foundation of Hunan Province (2024JK2109), Natural Science Foundation of Hunan Province (2023JJ30869), National Natural Science Foundation of China (U21A20384), Open Project Program of the State Key Laboratory of Proteomics (SKLP‐O201805), Science and Technology Innovation Plan of Hunan Province (2024WZ9001), Beijing Life Science Academy (BLSA, No: 2024100CB0170 and 2024200CD0020) and Central South University Graduate Innovation Program (Independent Exploration, 2023ZZTS0570).

Data availability

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

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Ethics Committee for Animal Experiments at the Department of laboratory Animals, Central South University (CSU-2023-0294). The minimum number of animals was used, and every attempt was made to reduce the suffering of the animals. All human tumor tissues used in this experiment were approved by the Ethics Committee at the Medical Ethics Committee of Xiangya Hospital, Central South University (202005397).

Consent for publication

Written informed consent for publication was obtained from all participants.

Large language model (LLM) declaration

No LLM (e.g., ChatGPT) was used in manuscript preparation.

Competing interests

The authors declare no competing interests.