Metal-ion-coordinated silk fibroin nanogels for enhanced oral chemotherapy of colorectal cancer

Abstract

The efficacy of oral chemotherapy for colorectal cancer (CRC) is hampered by poor drug stability and absorption in the upper gastrointestinal tract, as well as inadequate targeting efficiency at tumor sites. To address these issues, we proposed a simple and biocompatible nanogel, which was co-assembled from chemotherapeutic drug 5-fluorouracil (5-FU), regenerated silk fibroin (SF) as a natural protein carrier, and metal ions (Ni²⁺/Cu²⁺). The obtained nanogel system exploited the coordination interactions among 5-FU, amino acid residues, and metal ions to form a multifunctional oral nano-drug system with excellent biocompatibility, high delivery efficiency, and superior tumor penetration capacity. Embedding this nanogel in the chitosan/alginate hydrogel enabled it to effectively traverse the gastrointestinal (GI) tract and accumulate at colorectal tumor sites. Furthermore, the multi-stimuli-responsive properties of SF-based nanogel facilitated tumor microenvironment-responsive drug release, while metal ion-mediated chemodynamic therapy synergistically amplified the chemotherapeutic efficacy of 5-FU. This nanogel system provides a facile and translational strategy for improving the therapeutic performance of CRC chemotherapy.

Graphic abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04075-0.

Introduction

Colorectal cancer (CRC) remains one of the most prevalent and lethal malignancies worldwide [1]. Chemotherapy remains the treatment mainstay [2], and 5-fluorouracil (5-FU), a pyrimidine analogue, serves as the first-line drug for CRC therapy by inhibiting thymidylate synthase and disrupting nucleic acid synthesis [3]. However, its clinical efficacy is compromised by rapid metabolism, nonspecific distribution, and severe systemic toxicity [4]. Oral administration of 5-FU offers unique advantages, including noninvasiveness, improved patient compliance, and potential for local absorption in the gastrointestinal (GI) tract [5–7]. Nevertheless, 5-FU exhibits poor colonic bioavailability due to its rapid degradation and limited intestinal absorption [8]. Therefore, developing a safe, efficient, and tumor-selective oral chemotherapeutic system for CRC remains an urgent need.

The tumor microenvironment (TME) of CRC features high levels of hydrogen peroxide (H₂O₂), acidity, and glutathione (GSH) [9], which provide ideal conditions for chemodynamic therapy (CDT) [10]. Among the various transition metal ions capable of mediating CDT, Cu²⁺, and Ni²⁺ have received particular attention due to their superior redox activity, tumor selectivity, and ability to participate in both Fenton-like catalysis and drug coordination [11]. These ions not only promote reactive oxygen species (ROS) generation but also form stable complexes with therapeutic agents, enhancing both drug stability and antitumor efficacy [12, 13]. Furthermore, Cu²⁺ and Ni²⁺ can potentiate immunogenic cell death (ICD), bridging CDT with chemotherapy-induced immune activation [14]. Recent studies have demonstrated that metal ion-based nanoplatforms can effectively integrate CDT with other therapeutic modalities to improve antitumor efficacy through TME-responsive ROS generation. For example, carrier-free or metal-based theranostic systems combining CDT with photothermal therapy or radiotherapy have shown obvious tumor growth inhibition and metastasis suppression [15, 16]. These advances highlight the versatility of metal-ion-mediated ROS catalysis as a general strategy for synergistic cancer therapy.

However, uncontrolled release and poor stability of metal-drug complexes limit their biomedical applications, underscoring the need for a biocompatible carrier to stabilize metal coordination and enable controlled oral delivery [10, 17]. Silk fibroin (SF), a natural protein approved by the U.S. FDA, provides an ideal biopolymer platform for constructing stable and biocompatible metal-drug hybrid systems [18–20]. Our recent investigations reveal that SF nanoparticles (SFNPs) possess inherent multi-stimuli responsiveness to ROS, GSH, and hyperthermia [19, 21, 22]. Exposure to these stimuli disrupts the primary and secondary structural integrity of SF molecules, resulting in matrix loosening and consequent acceleration of drug release. These functional attributes underscore the significant promise of SFNPs as a nanogel-based platform for targeted tumor treatment.

Recent advances in nanomedicine have highlighted diverse strategies for improving the therapeutic efficacy and immunomodulatory potential of NP-based platforms. For instance, unimolecular chiral poly(amino acids) have been shown to act as effective adjuvants in nanovaccines, promoting dendritic cell activation and antitumor immune responses [23], while glycoengineering‑assisted biomineralization enables selective tumor blockade through tailored nanostructures [24]. Beyond traditional chemotherapeutics, engineered bacteria combined with nanomaterials offer novel approaches to reprogram the TME and potentiate antitumor immunity [25]. Concurrently, the development of nanoantidotes emphasizes the importance of balancing efficacy with safety, enhancing clinical applicability by reducing systemic toxicity [26]. Moreover, insights from brain‑targeted drug delivery platforms illustrate general principles for overcoming biological barriers, providing conceptual guidance for improving intestinal mucus penetration and localized delivery [27]. Collectively, these recent studies underscore the critical role of rational nanoplatform design in optimizing both therapeutic performance and immunomodulatory effect, complementing our approach of metal‑ion‑functionalized SFNPs for CRC-oriented oral drug delivery.

Herein, we used a coordination-driven desolvation strategy to fabricate Ni²⁺- and Cu²⁺-chelated 5-FU-loaded SFNPs (FU-Ni@SFNPs and FU-Cu@SFNPs) as an oral nanoplatform for CRC therapy [28, 29]. The system integrated chemotherapy and CDT within a single carrier, featuring excellent GI stability, redox-responsiveness, and TME-triggered drug release. Upon oral administration, the NPs embedded in chitosan/alginate hydrogel resisted gastric acid and enzymatic degradation, maintaining structural integrity until reaching the tumor sites [30]. In the acidic and GSH-rich colorectal TME, the hydrogel was destructed, and metal ions were released from NPs to convert endogenous H₂O₂ into hydroxyl radicals (•OH), thereby amplifying the anti-tumor and immunogenic effects of 5-FU. This multifunctional, safe, and efficient oral nanoplatform provides a promising strategy for CRC treatment and offers a translational paradigm for intelligent oral nanomedicines targeting GI malignancies.

Materials and methods

Materials

Silkworm cocoons were supplied by the State Key Laboratory of Silkworm Genome Biology (China). Nickel (II) acetate tetrahydrate (Ni(CH₃COO)₂·4H₂O), copper (II) acetate monohydrate (Cu(CH₃COO)₂·H₂O), 5-FU, and hydroxyethyl cellulose (HEC) were purchased from Aladdin (Shanghai, China). Adriamycin hydrochloride (DOX), H₂O₂, GSH, and sodium carbonate (Na₂CO₃) were supplied by Sigma Aldrich (St. Louis, USA). Fetal bovine serum (FBS) was obtained from Shanghai ExCell Bio, Inc. (Shanghai, China). 3-(4,5-Dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT), penicillin-streptomycin-gentamicin solution (100×), trypsin, high mobility group box 1 protein (HMGB1) rabbit polyclonal antibody, calreticulin (CRT) rabbit polyclonal antibody, Cy3-labeled goat anti-rabbit IgG (H + L), DAPI, live/dead assay kit, enhanced adenosine 5’-triphosphate (ATP) Assay Kit, ROS detection kit, and glutathione/oxidized glutathione (GSH/GSSG) assay kit were purchased from the Beyotime Institute of Biotechnology (Nanjing, China). Dimethyl sulfoxide (DMSO) and triton X-100 were from Adamas-beta (Shanghai, China). DMEM (high sugar) was supplied by Thermo Fisher (Massachusetts, USA). Azoxymethane (AOM) and dextran sulfate sodium salt (DSS, MW = 36–50 kDa) were obtained from MP Biomedical Inc. (Solon, OH, USA). ELISA kits for TNF-α and IFN-γ were purchased from Solarbio Technology Co., LTD (Beijing, China). The optimal cutting temperature (OCT) compound, bicinchoninic acid (BCA) assay protein kit, bovine serum albumin (BSA), paraformaldehyde (4%), hematoxylin-eosin staining (H&E), Ki67, TdT-mediated dUTP Nick-End Labeling (TUNEL), and CD4⁺/CD8⁺ staining kits were supplied by Wuhan Servicebio Technology Co., Ltd (Wuhan, China).

Preparation of various SFNPs

FU-Ni@SFNPs and FU-Cu@SFNPs were fabricated through metal-drug coordination followed by desolvation. Briefly, 5-FU was dissolved in ultrapure water (1 mg/mL), and Ni(CH₃COO)₂·4H₂O (10 mM) or Cu(CH₃COO)₂·H₂O (10 mM) was added dropwise under constant stirring to obtain a homogeneous metal-drug precursor solution. Subsequently, SF solution (10 mg/mL) was introduced at a 1:1 (w/w) ratio to stabilize the composite and promote nanostructure formation. The obtained solution was added dropwise into acetone (1:5, v/v) under magnetic stirring and allowed to evaporate acetone in a fume hood. The suspension was sonicated (120 W, 2 min) and purified by gradient centrifugation. The precipitate was redispersed in deionized water and washed 3 times to obtain stable dispersions of FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs.

In vitro anti-colorectal tumor activity of various SFNPs

CT-26 cells (1 × 10⁴) were cultured in 96-well plates and incubated overnight. Thereafter, the complete medium was replaced with basic medium containing FU@SFNPs, FU-Ni@SFNPs, or FU-Cu@SFNPs at different 5-FU concentrations (3, 6, 12, 24, and 48 µg/mL) for 24 h. After co-incubation, MTT solution (0.5 mg/mL, 50 µL) was replaced to each well and incubated for 3 h. The MTT medium was carefully removed, and 100 µL DMSO was added to dissolve the formazan crystals. The absorbance was measured at 570 nm using a microplate reader (Spark 10 M, Tecan, Austria). Untreated cells were treated as negative controls, and cells treated with 2.0% (w/v) Triton X-100 served as positive controls.

Detection of ICD after the treatment of SFNPs

CT-26 cells (2 × 10⁵) were cultured in 12-well plates and incubated overnight. Thereafter, the complete medium was replaced with basic medium containing FU@SFNPs, FU-Ni@SFNPs, or FU-Cu@SFNPs (5-FU: 12 µg/mL). After 24 h incubation, the cells were washed with PBS and treated with anti-CRT and anti-HMGB1 antibodies at 4 ℃ for 12 h to detect the CRT and HMGB1 levels, respectively. Cells used for CRT and HMGB1 detection were incubated with Cy3-conjugated secondary antibodies at 37 ℃ for 60 min. Finally, cells were stained with DAPI for 10 min and visualized using CLSM (FV3000, Olympus Corporation, Japan).

We also assessed the contents of ATP released from CT-26 cells receiving different treatments using the corresponding kit (Beyotime, China). After undergoing similar treatment, the cells were disrupted by discarding the culture medium and introducing ATP detection lysate at a ratio of 200 µL/well of the 12-well plate. After cell lysis, the resulting solution was centrifuged at 12,000 g for 5 min at 4 °C to collect the supernatant. Subsequently, the ATP concentration of the collected supernatant was determined using an ATP detection working solution by the microplate reader (SPARK 10 M, Switzerland Tecan).

GSH/GSSG ratios in SFNP-treated colon carcinoma cells

The intracellular redox status was evaluated using a GSH/GSSG assay kit according to the manufacturer’s instructions. Briefly, cells were harvested after treatment, lysed, and the levels of reduced GSH and GSSG were determined spectrophotometrically according to the manufacturer’s instructions. The GSH/GSSG ratio was calculated to assess the cellular oxidative state.

Mucus penetration capacity of SFNPs

The HEC hydrogel was chosen as a mucus simulator due to its comparable spatial structures and microrheological properties with colonic mucus [31]. The movements of NPs were analyzed within a simulated mucus environment containing 1% (w/v) HEC, utilizing a live cell imaging system (IX83, Olympus, Japan). This investigation utilized the Dox molecule with fluorescence properties as a model drug. The round dishes (34 mm in diameter) were filled with 100 µL of HEC hydrogel solution, followed by the addition of 50 µL of SFNP suspensions (Dox: 0.2 µg/mL) onto the hydrogel surface. The HEC hydrogels were visually imaged with CLSM (FV3000, Olympus Corporation, Japan).

Animal model and treatment

After successful tumor induction, mice were randomly divided into 4 groups (n = 5 per group): healthy control, AOM-DSS control, FU@SFNPs, and FU-Cu@SFNPs. To ensure oral stability and prevent premature degradation of the metal–drug nanocomposites in the gastric environment, the NPs were embedded in a chitosan/alginate hydrogel matrix prior to gavage administration. The hydrogel encapsulation effectively protected the NPs from acidic conditions and enzymatic digestion, while allowing gradual release and improved retention within the colon [32]. Oral administration of SFNPs (Fu dose: 5 mg/kg) was performed every 3 days for a total of 5 doses (For all therapeutic evaluations, the dosage was normalized to the equivalent 5-FU amount across different formulations).

The body weights were recorded daily. The 5 internal organs, colon, and blood were collected at the end of investigations. Colon tumor tissues and 5 principal organs (heart, liver, spleen, lung, and kidney) were weighed and underwent histological examination with H&E staining. Blood samples were analyzed for routine hematology and pro-inflammatory factors (TNF-α and IFN-γ).

Colorectal tumor tissues were collected from AOM/DSS-induced CRC mice at the end of the treatment period, embedded, sectioned, and subjected to immunofluorescence staining. Tissue sections were stained with antibodies against HMGB1, CRT, CD4, and CD8, followed by counterstaining with DAPI. Fluorescence images were acquired using confocal microscopy. Mean fluorescence intensity (MFI) was quantified in tumor regions using ImageJ software for comparative analysis.

Statistical analysis

All the results were expressed as the mean ± standard error of the mean (s.e.m.). Unless otherwise indicated, biological replicates were used for all the experiments. Statistical analyses were conducted using a two-tailed t-test or one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test for post hoc analysis by GraphPad Prism version 9.0.0. Statistical significance was represented by p < 0.05, p < 0.01, and p < 0.001.

Results and discussion

Preparation and physicochemical characterization of FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs

5-FU was successfully coordinated with Ni(CH₃COO)₂·4H₂O or Cu(CH₃COO)₂·H₂O to form metal-drug complexes, which were subsequently encapsulated within SF matrix via an acetone-induced desolvation process, yielding stable, water-dispersible nanocomposites after sonication and gradient centrifugation. The obtained FU-Ni@SFNPs appeared light blue to blue-green, while the resultant FU-Cu@SFNPs appeared blue-green, indicating the successful formation of metal-drug coordination (Fig. 1a). Dynamic light scattering (DLS) analysis was performed to evaluate the hydrodynamic sizes and size distribution profiles of the NPs, which reflected their colloidal stability and uniformity in aqueous media. The analysis showed that all 3 types of SFNPs exhibited narrow size distributions, with average particle sizes of 234.9 nm (FU@SFNPs), 239.1 nm (FU-Ni@SFNPs), and 217.3 nm (FU-Cu@SFNPs) (Fig. 1b and Table S1). Transmission electron microscopy (TEM) was used to observe the morphology and structural integrity of the NPs, further confirming their uniform spherical morphology and good dispersibility (Fig. 1c).

Fig. 1.

Preparation and physicochemical characterization of various SFNPs. (a) Schematic diagram of the construction process of FU-Ni@SFNPs and FU-Cu@SFNPs. (b) Hydrodynamic particle size distribution profiles of FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs. (c) Representative TEM images of FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs. Scale bar = 500/100 nm. (d) Contents of 5-FU in FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs determined by HPLC. (e) Contents of Ni/Cu (μg per 10 mg sample) in the SFNPs determined by ICP analysis. Data are expressed as mean ± s.e.m. (n = 3)

High-performance liquid chromatography (HPLC) analysis is essential to evaluate how metal coordination affects 5-FU encapsulation within the SF matrix. The 5-FU contents in FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs were approximately 7.0, 2.3, and 3.7 µg/mg, respectively, with FU-Cu@SFNPs showing a higher 5-FU content than FU-Ni@SFNPs (Fig. 1d). To verify the incorporation of Ni²⁺ and Cu²⁺ into SF matrix, the NPs were digested and analyzed by inductively coupled plasma-optical emission spectrometer (ICP-OES), enabling quantitative measurement of metal content. The analysis demonstrated the successful incorporation of metal ions, with Ni²⁺ and Cu²⁺ contents of 7.7 and 14.7 µg per 10 mg sample, respectively (Fig. 1e).

DLS analysis showed that FU-Ni@SFNPs and FU-Cu@SFNPs remained monodisperse in PBS and exhibited minimal size variations under 100 µM H₂O₂, indicating their colloidal stability under oxidative conditions. In contrast, exposure to 10 mM reduced glutathione (r-GSH) induced pronounced increases in particle sizes (FU-Ni@SFNPs: 2.5 μm; FU-Cu@SFNPs: 1 μm), consistent with reduction-triggered structural transformation. These stimuli-responsive behaviors suggest that the particles undergo GSH-mediated disintegration within the reductive microenvironment of tumor cells, facilitating on-site drug release and enhanced intracellular retention (Fig. 2a, b).

Fig. 2.

Physicochemical characterization of various SFNPs. Particle size variations of (a) FU-Ni@SFNPs and (b) FU-Cu@SFNPs under different conditions, including PBS, H₂O₂, and GSH. In vitro drug release profiles of (c) FU@SFNPs, (d) FU-Ni@SFNPs, and (e) FU-Cu@SFNPs under different conditions, including PBS, H₂O₂, and GSH. Data are expressed as mean ± s.e.m. (n = 3)

To assess the stimuli-responsive drug release, in vitro drug release studies were conducted under H₂O₂ and GSH conditions. All three formulations exhibited slow drug release in PBS, whereas exposure to H₂O₂ or GSH markedly accelerated the release of drugs from SFNPs. It was observed that after 24 h of co-incubation, FU-Cu@SFNPs achieved a release rate of 91.9% under H₂O₂ and GSH stimulation, which was 1.1- and 1.3-fold higher than FU@SFNPs (82.0%) and FU-Ni@SFNPs (72.9%), respectively. Significantly, FU-Cu@SFNPs showed the fastest drug release under the combination of H₂O₂ and GSH. These results indicate that the metal-FU nanogels possess favorable morphology, stability, and redox-responsive drug release properties, enabling the efficient and tumor-specific drug delivery (Fig. 2c-e).

In vitro antitumor effect of NPs

Efficient cellular uptake is crucial for effective drug delivery and subsequent antitumor activity [33]. Since 5-FU lacks intrinsic fluorescence, DOX was employed as a fluorescent model drug for cellular uptake assays. Flow cytometric results revealed a time-dependent uptake of Dox@SFNPs by CT-26 cells, increasing from 47.6% at 1 h to 80.2% at 5 h (Fig. 3a), indicating the efficient and time-dependent cellular uptake profiles of SFNPs.

Fig. 3.

In vitro antitumor activities of various SFNPs. (a) Flow cytometric histograms of cellular uptake profiles of Dox@SFNPs (Dox: 1 µg/mL). In vitro viabilities of CT-26 cells treated with (b) FU@SFNPs, (c) FU-Ni@SFNPs, and (d) FU-Cu@SFNPs for 24 h. (e) Live/dead staining of CT-26 cells with various treatments. Scale bar = 100 μm. (f) Intracellular ROS levels of CT-26 cells receiving various treatments. Scale bar = 100 μm. CLSM images of (g) CRT and (h) HMGB1 after various treatments. Scale bar = 100 μm. (i) Extracellular ATP released from CT-26 cells after various treatments for 24 h. (j) Intracellular ATP content of CT-26 cells after various treatments for 24 h. (k) Quantitative analysis of intracellular GSH/GSSG ratios after various treatments for 24 h. Data are expressed as mean ± s.e.m. (n = 3). For all in vitro experiments involving FU-based SFNPs (e-k), cells were treated with SFNPs corresponding to an equivalent 5-FU concentration of 12 µg/mL

Having demonstrated that SFNPs can be efficiently internalized by CT-26 cells, it was essential to assess whether this uptake translates into cytotoxic effects, and how metal coordination influences the antitumor activity of the NPs. Further MTT assays revealed that FU@SFNPs, FU-Ni@SFNPs, and FU-Cu@SFNPs exhibited dose-dependent cytotoxicity. Compared with FU@SFNPs and FU-Ni@SFNPs, FU-Cu@SFNPs displayed lower toxicity at a low concentration (5-FU: 3 µg/mL) but significantly reduced cell viability at a high concentration (5-FU: 48 µg/mL), suggesting the sustained drug release profile and synergistic antitumor effect mediated by Cu²⁺ (Fig. 3b-d). To directly visualize the cytotoxic effect at a representative concentration (5-FU: 12 µg/mL), calcein AM/PI staining was performed, allowing assessment of cell death induced by the NPs. These findings were confirmed by the subsequent calcein AM/PI staining images, with FU-Cu@SFNPs showing the highest proportion of dead cells, consistent with enhanced tumoricidal activity via metal incorporation (Fig. 3e and Fig. S1). Since intracellular ROS production is a key mechanism contributing to NP-induced tumor cell death [34], ROS generation assays were conducted to evaluate oxidative stress triggered by the treatments. It was found that FU@SFNP treatment induced a modest increase in intracellular ROS levels. In comparison, FU-Ni@SFNPs and FU-Cu@SFNPs markedly elevated ROS production, reaching 1.5- and 1.7-fold higher than FU@SFNPs, respectively. Notably, FU-Cu@SFNPs generated the strongest oxidative stress response, with ROS levels increasing by more than 5.2-fold relative to the control group (Fig. 3f and Fig. S2a). As shown in Fig. S2b, FU-Ni@SFNPs and FU-Cu@SFNPs treatments significantly increased intracellular ROS levels, as evidenced by a marked right shift of the fluorescence peak and a higher fluorescence intensity compared with the control group. This enhancement correlates with increased cell death, indicating that metal ion incorporation amplifies ROS-mediated oxidative stress, contributing to enhanced antitumor activity.

To evaluate whether the NPs could trigger ICD, key ICD markers, including CRT exposure, HMGB1 release, and extracellular ATP secretion, were analyzed, as these signals are critical for initiating antitumor immune responses [35]. CRT exposure analysis showed that FU-Ni@SFNPs and FU-Cu@SFNPs markedly enhanced CRT expression compared with FU@SFNPs. Specifically, FU-Cu@SFNPs induced the strongest CRT exposure, with an average intensity 7.4-fold higher than the control, slightly surpassing FU-Ni@SFNPs (6.8-fold) and significantly exceeding FU@SFNPs (5.7-fold) (Fig. 3g and Fig. S3a). HMGB1 release results further confirmed the ICD-triggering capability. Both SFNPs promoted strong HMGB1 release. Notably, FU-Ni@SFNPs and FU-Cu@SFNPs elevated HMGB1 secretion by approximately 3.4- and 3.2-fold relative to the control, respectively, whereas FU@SFNPs yielded only a 3.5-fold increase, indicating a significantly stronger ICD response induced by metal coordination-assisted formulations (Fig. 3h and Fig. S3b). In addition, metal coordination-assisted formulations markedly enhanced ATP release, as evidenced by both intracellular ATP depletion and extracellular ATP accumulation (Fig. 3i, j). To further understand how ROS elevation affects intracellular redox homeostasis, we measured the GSH/GSSG ratio to assess oxidative stress induced by the NPs. The intracellular redox state of GSH/GSSG ratio decreased significantly after FU@SFNP treatment and further declined in the FU-Ni@SFNP- and FU-Cu@SFNP-treated groups, with FU-Cu@SFNPs showing the lowest ratio, indicating that Cu²⁺ could mediate the amplification of ROS-induced redox imbalance and oxidative stress in CT-26 cells (Fig. 3k).

Collectively, these findings demonstrate that Cu²⁺- and Ni²⁺-incorporated FU@SFNPs are able to enhance cell death, ROS generation, and ICD hallmarks, with FU-Cu@SFNPs consistently showing the strongest effects, highlighting the superior role of Cu²⁺ in potentiating both cytotoxic and immunogenic effects.

Mucus infiltration and CRC tissue penetration of SFNPs

Effective CRC treatment requires nanotherapeutics to penetrate intestinal mucus and accumulate at the tumor sites [36]. To evaluate the penetration ability of SFNPs in the colorectal mucus, a HEC hydrogel was employed to mimic the physiological mucus layer. After 5 min of incubation, Dox-Ni@SFNPs (90.0 μm) and Dox-Cu@SFNPs (103.6 μm) displayed similar mucus penetration depths, with Dox-Cu@SFNPs exhibiting slightly superior depth, likely due to their smaller particle size and higher efficient diffusivity (Fig. 4a).

Fig. 4.

Colorectal mucus infiltration and tumor penetration of SFNPs. (a) Mucus penetration of Dox-Ni@SFNPs and Dox-Cu@SFNPs. (b) Distribution profiles of Dox@SFNP-embedded hydrogel in the GI tract after oral administration. (c) Semi-quantification of the accumulated amounts of Dox@SFNPs in the colorectal tissues. (d) Colorectal tissue penetration profiles of Dox-Ni@SFNP- and Dox-Cu@SFNP-embedded hydrogel after 6 h of oral administration. Scale bar = 200 μm. Data are expressed as mean ± s.e.m. (n = 3)

To verify whether the NPs could achieve effective delivery to the tumor tissues in vivo, we next investigated their biodistribution and tumor accumulation in the colon tissues. Ex vivo imaging and OCT-based tissue analysis further demonstrated enhanced penetration and accumulation of Dox@SFNPs in the colon tumor tissues. After oral administration, fluorescence signals of Dox@SFNPs in the colon tissues peaked at 6 h, aligning with semi-quantitative fluorescence analysis (Fig. 4b, c). To gain deeper insight into the spatial distribution of the NPs across the colonic tissue, we employed OCT imaging to evaluate their penetration depth within the mucosal layer. The results confirmed that FU-Cu@SFNPs penetrated deeper into the colonic mucosal layer than FU-Ni@SFNPs, showing slightly deeper penetration of FU-Cu@SFNPs compared with FU-Ni@SFNPs, highlighting Cu²⁺-mediated enhancement of tissue infiltration (Fig. 4d). These results demonstrate that Cu²⁺ incorporation enhances both mucus infiltration and colon tumor tissue penetration, thereby facilitating effective local drug delivery. This might be attributed to the differences in the coordination states of the metals [37].

In vivo antitumor effect of NPs

Based on the above superior in vitro performance, FU-Cu@SFNPs were selected for the subsequent in vivo anti-colon cancer evaluation based on an AOM/DSS-induced CRC mouse model (Fig. 5a). We subsequently evaluated the therapeutic outcomes of these SFNPs in CRC mouse model following oral administration. It was observed that treatment with FU-Cu@SFNPs markedly reduced tumor burden compared with the AOM-DSS control group. On average, FU-Cu@SFNPs decreased the total number of tumors by 2.1-fold, compared with a 1.4-fold reduction by FU@SFNPs. Notably, the number of large tumors (> 3 mm) was also substantially reduced, with FU-Cu@SFNPs achieving a 2.9-fold decrease relative to the AOM-DSS control group, whereas FU@SFNPs reduced large tumors by 2.5-fold. These data demonstrate that FU-Cu@SFNPs not only suppress overall tumor formation but also preferentially prevent the development of large, aggressive tumors (Fig. 5b, c). We measured serum cytokine levels to assess whether the NPs could stimulate systemic immune responses. TNF-α is a pro-inflammatory cytokine that plays a key role in activating immune cells and mediating antitumor effects, while IFN-γ is critical for promoting T-cell responses and enhancing tumor immunosurveillance [38]. Serum TNF-α and IFN-γ levels were markedly elevated relative to both the healthy control and AOM-DSS control groups (Fig. 5d, e), demonstrating the enhanced systemic immune activation induced by FU-Cu@SFNPs. Quantitatively, FU-Cu@SFNP treatment increased TNF-α to 132.5 pg/mL, achieving a 1.2- and 1.3-fold elevation compared with the healthy control and AOM-DSS control groups, respectively. Similarly, the treatment of FU-Cu@SFNPs raised IFN-γ levels to 24.8 pg/mL, corresponding to 1.4-fold and 1.3-fold increases relative to the healthy and AOM-DSS groups, respectively. Together, these data indicate that FU-Cu@SFNPs induce substantially stronger systemic cytokine activation than FU@SFNPs or untreated controls. In addition, we monitored mouse body weights as a general indicator of systemic toxicity, since significant weight loss could reflect adverse effects on overall health and organ function. Mice in the AOM-DSS control group exhibited continuous weight decline throughout the experiment, whereas both FU@SFNPs and FU-Cu@SFNPs markedly alleviated this trend. These results demonstrate that FU@SFNPs and FU-Cu@SFNPs yield therapeutic benefits without inducing systemic toxicity (Fig. S4). Histological examination revealed reduced Ki67-positive cells and improved mucosal architecture in the FU-Cu@SFNP-treated group (Fig. 5f). Consistently, TUNEL staining confirmed increased apoptosis (Fig. 5g, h). We performed immunofluorescence staining to evaluate ICD, as CRT exposure on the cell surface and HMGB1 release into the extracellular space serve as key hallmarks of ICD and indicators of antitumor immune activation [39]. Immunofluorescence revealed the enhanced CRT and HMGB1 exposure (Fig. 6a-c). To connect the induction of ICD with functional antitumor immunity, we examined whether enhanced CRT and HMGB1 exposure translated into increased recruitment of effector T cells within the TME. This analysis revealed increased infiltration of CD4⁺ helper T cells and CD8⁺ cytotoxic T cells in the colon tumor tissues (Fig. 6d, e). Collectively, FU-Cu@SFNPs effectively suppress tumor progression and reprogram the immune microenvironment of colon tumors, leading to robust antitumor efficacy.

Fig. 5.

In vivo therapeutic outcomes of FU-Cu@SFNPs against CRC. (a) Treatment protocol of FU-Cu@SFNPs against AOM/DSS-induced CRC. (b) Total tumor numbers per mouse and (c) numbers of different-sized tumors per mouse. Data are expressed as mean ± s.e.m. (n = 5). Levels of (d) TNF-α and (e) IFN-γ in the serum from various mouse groups. Data are expressed as mean ± s.e.m. (n = 3). (f) H&E and Ki67 staining of colorectal tumor tissues from various groups. Scale bar = 100 μm. (g) TUNEL staining of colorectal tumor tissues from various groups. Scale bar = 100 μm. (h) MFIs of TUNEL fluorescence signals in the colorectal tumors from mice receiving various treatments. Data are expressed as mean ± s.e.m. (n = 3)

Fig. 6.

In vivo therapeutic outcomes of FU-Cu@SFNPs against CRC. (a) HMGB1 and CRT staining of colorectal tumor tissues from various groups. Scale bar = 100 μm. MFIs of (b) CRT and (c) HMGB1 in the colorectal tumors from mice receiving various treatments. Data are expressed as mean ± s.e.m. (n = 3). (d) CD4/CD8 staining of colorectal tumor tissues from various groups. Scale bar = 100 μm. (e) MFIs of CD4⁺/CD8⁺ in the colorectal tumors from mice receiving various treatments. Data are expressed as mean ± s.e.m. (n = 3)

Beyond tumor suppression, we next assessed the systemic biosafety of the treatment. Histological analysis was used to examine the potential tissue damage or inflammation in the principal organs, while hematological analysis was applied to evaluate the blood parameters, such as red and white blood cell counts and hemoglobin levels. These measurements provide a straightforward assessment of systemic toxicity and overall physiological health, confirming the in vivo safety of the NPs. Histological and hematological analyses demonstrate the excellent in vivo biosafety of FU-Cu@SFNPs, showing minimal organ damage and near-normal blood parameters compared to the healthy group (Fig. S5 and S6).

Conclusion

In this study, we developed a coordination-driven SF nanogel (FU-Cu@SFNPs) as a biocompatible and orally administrable platform for CRC therapy. The incorporation of Cu²⁺ not only stabilized the 5-FU-protein matrix through coordination interactions but also enabled redox-mediated chemodynamic activity, resulting in enhanced ROS generation and ICD. Both in vitro and in vivo evaluations demonstrated that FU-Cu@SFNPs effectively suppressed tumor growth, restored intestinal mucosal integrity, and promoted immune reprogramming characterized by elevated CRT/HMGB1 expression, ATP release, and increased infiltration of CD4⁺ and CD8⁺ T cells. Importantly, the oral formulation exhibited excellent biosafety and systemic tolerance, indicating its suitability for long-term administration. Collectively, this work establishes FU-Cu@SFNPs as a safe, redox-responsive, and immunoactive oral nanotherapeutic platform that integrates chemotherapy and chemodynamic therapy. This design concept provides a promising framework for the clinical translation of intelligent nanomedicines targeting GI malignancies.

Abbreviations

CRC

Colorectal cancer

5-FU

5-Fluorouracil

GI

Gastrointestinal

TME

Tumor microenvironment

H₂O₂

Hydrogen peroxide

GSH

Glutathione

CDT

Chemodynamic therapy

•OH

Hydroxyl radicals

ROS

Reactive oxygen species

ICD

Immunogenic cell death

SF

Silk fibroin

NP

Nanoparticle

Ni(CH₃COO)₂·4H₂O

Nickel (II) acetate tetrahydrate

Cu(CH₃COO)₂·H₂O

Copper (II) acetate monohydrate

HEC

Hydroxyethyl cellulose

DOX

Adriamycin hydrochloride

Na₂CO₃

Sodium carbonate

FBS

Fetal bovine serum

MTT

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

HMGB1

High mobility group box 1 protein

CRT

Calreticulin

ATP

Adenosine 5’-triphosphate

GSSG

Oxidized glutathione

DMSO

Dimethyl sulfoxide

AOM

Azoxymethane

DSS

Dextran sulfate sodium salt

OCT

Optimal cutting temperature

BCA

Bicinchoninic acid assay

BSA

Bovine serum albumin

H&E

Hematoxylin-eosin staining

TUNEL

Tdt-mediated dutp nick-end labeling

s.e.m.

Standard error of the mean

DLS

Dynamic light scattering

TEM

Transmission electron microscopy

HPLC

High-performance liquid chromatography

ICP-OES

Inductively coupled plasma-optical emission spectrometer

r-GSH

Reduced glutathione

Author contributions

Baoyi Li performed the experiments and wrote the manuscript. Jinhua Liu, Xiaoyan Li, Yuting Luo, Ga Liu, Menghang Zu, and Xiaoxiao Shi performed the experiments. Yajun Wang, Zhenhua Zhu, and Bo Xiao designed experiments, supervised studies, and wrote the manuscript. Yajun Wang, Rui L. Reis, Subhas C. Kundu, Bo Xiao, and Zhenhua Zhu edited and revised the manuscript. All authors have given approval to the final version of the manuscript. The animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of Southwest University, and all methods were carried out in compliance with relevant guidelines and regulations. All authors read and agreed to submit the manuscript.

Funding

This study was supported by the Jiangxi Provincial Natural Science Foundation (20224BAB206073), the national natural Science Foundation of China (82360110 and 32401170), the Fundamental Research Funds for the central Universities (SWU-KQ22075), and the Science and technology Research Program of chongqing Municipal education commission (KJQn202300223).

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Ethics approval and consent to participate

Animal studies were approved by the Institutional Animal Care and Use Committee of Southwest University.

Consent for publication

All authors read and agreed to submit the manuscript.

Competing interests

The authors declare no competing interests.