Enzyme/Reactive Oxygen Species-Dually Activated Hyaluronic Acid Nanocarriers Enable Celastrol Delivery for Site-Specific Therapy of Inflammatory Bowel Diseases and Colorectal Cancer.

Ulcerative colitis (UC) is a chronic and relapse-prone nonspecific inflammatory disease associated with hemorrhagic diarrhea, mucosal ulcers, and weight loss.[1–3] The high recurrence rate of UC not only affects 7 million people worldwide but also takes up a lot of medical resources.^[4] Furthermore, long-term and repeated inflammation can trigger the occurrence and progression of colitis-associated colorectal cancer (CAC). According to reports, the lifetime risk of CAC in UC patients will be 2–4 times higher than that in the general population.^[5] Furthermore, CAC underscores the perpetual stimulation of epithelial cell proliferation within the inflammatory milieu. Consequently, CAC is often identified at advanced stages, highlighting the necessity for early detection strategies to preempt neoplastic progression.^[6] Currently, a range of therapeutic options is available for colon diseases at various stages, with mesalamine and sulfasalazine being the mainstay for mild-to-moderate UC, neurosteroids frequently administered for moderate-to-severe UC,^[7] and combined chemotherapy regimens incorporating bevacizumab or cetuximab serving as the standard approach for severe colorectal cancer (CRC).^[8] Although these drugs have obvious therapeutic effects on different stages of intestinal diseases, inconvenient dosing methods, and frequent drug alterations pose challenges for patients.^[9]

Researchers are actively engaged in the extraction of bioactive compounds from natural plants for the treatment of inflammation and cancer. Celastrol (Cel) is a pentacyclic triterpene compound extracted from Tripterygium wilfordii Hook F, which has various pharmacological activities and is considered one of the five most promising bioactive compounds in nature.^[10],^[11] Studies have demonstrated that low concentrations of Cel exerts anti-inflammatory and anti-cancer effects, enabling its application across a spectrum of intestinal diseases (including UC, CAC, and CRC).^[5],^[10] Specifically, Cel reverses the imbalance of Treg/Th17 and Treg/Th1 subsets in intestinal mucosa, while suppressing the production of pro-inflammatory cytokines (TNF-α, IL-1β) in human monocytes and macrophages.^[12] It also prevents UC progression to CRC by inhibiting inflammation, intervening in epithelial-mesenchymal transition (EMT), and downregulating mutant p53 and phosphorylated p53 (p-p53).^[13] Additionally, Cel inhibits CRC cell growth through multiple pathways: promoting β-catenin degradation via the HSF1-LKB1-AMPKα-YAP axis,^[14] and suppressing cell growth/migration by inhibiting NOS activity and angiogenesis.^[15] Despite these promising pharmacological properties, the therapeutic potential of Cel is severely constrained by its poor water solubility, rapid systemic clearance, and extremely low oral bioavailability, which together result in insufficient local drug concentrations at colonic lesions following conventional oral administration.^[16] To achieve effective therapy for intestinal diseases, it is therefore essential to not only improve the systemic oral bioavailability of Cel but also, more critically, enhance its local availability within the diseased colonic tissue through lesion-specific delivery strategies that overcome the gastrointestinal barriers and achieve targeted accumulation at inflammatory/malignant sites.

Currently, nanoparticles (NPs) have gained widespread application in oral drug delivery systems for the treatment of intestinal diseases and have shown promise in enhancing the therapeutic efficacy of Cel.^[7],^[17],^[18] Nevertheless, oral colon-targeted delivery faces several inherent challenges that have not yet been fully resolved. First, the harsh gastrointestinal environment—including gastric acidity, extensive digestive enzymes, and the mucus barrier—severely compromises the stability and mucosal penetration of conventional nanocarriers, leading to premature drug release and poor bioavailability.^[19] Second, the lack of lesion-specific targeting ability often results in nonspecific biodistribution, thereby limiting local therapeutic concentrations at the diseased sites and increasing the risk of systemic side effects.^[20] Third, the pathological heterogeneity across different stages of intestinal diseases—from acute inflammation to malignancy—demands a versatile delivery platform capable of adapting to the dynamically changing microenvironment, yet most current systems are designed for a single pathological state.^[21] For example, a flexible oral colon-targeting delivery system was developed via electrostatic layer-by-layer alternate deposition of pectin-trimethyl chitosan (TMC) onto celastrol-loaded liposomes (Cel/PT-LbL Lipo) to improve its anti-UC efficacy.^[22] However, practical issues such as the toxicity of polymer materials, insufficient responsiveness to disease-specific stimuli, and inadequate targeting precision for colonic lesions remain to be fully addressed. β-Cyclodextrin (β-CD, a cyclic heptaglucoside) and hyaluronic acid (HA, a natural anionic polysaccharide) are biocompatible colon-targeted carriers.^[23] β-CD, a cyclic heptasaccharide of glucopyranose units, offers exceptional biocompatibility and enzymatic degradability, enabling colon-specific drug release via microbial fermentation.[24–26] HA, an anionic polysaccharide resistant to gastrointestinal pH variations, binds CD44 receptors on inflamed macrophages and tumor cells, facilitating targeted cellular uptake.[27–31] Therefore, we propose the use of HA and β-CD as drug carriers for the targeted delivery of Cel, enabling site-specific release at the colonic lesion site in response to local stimuli.

Based on the above hypotheses, we developed and fabricated versatile nanoplatforms that employ HA for active targeting and incorporate β-CD and 4-(hydroxymethyl)phenylboronic acid (PAPE) for multiple (enzyme/ROS) sensitivity to encapsulate Cel (HA@Cel/NPs). Firstly, ROS/enzyme-sensitive material (PAPE-SA-CD) was formed by the covalent linkage of PAPE with β-CD, and the amphiphilic nature encapsulated Cel. Subsequently, with the host-guest interaction of β-CD and 1-adamantanecarboxylic acid (AD), the HA was uniformly wrapped on the outside of this substance, finally forming a drug delivery system (HA@Cel/NPs). Leveraging CD44 overexpression on target cells, HA@Cel/NPs are internalized by macrophages via CD44-mediated endocytosis for UC treatment and actively target colon cancer cells for CRC therapy. Elevated levels of colonic microenvironment about intracellular ROS and intestinal enzymes lead to rapid degradation of NPs. To validate this design, we developed three intestinal disease models (UC, CAC, and CRC) and conducted a comprehensive investigation into the pharmacological efficacy of HA@Cel/NPs both in vitro and in vivo. Notably, Cel induces immunogenic cell death (ICD) by promoting premature release of calreticulin (CRT) and high mobility group box 1 (HMGB1), exerting both direct anti-cancer effects and robust anti-tumor immune responses. Thus, we combined HA@Cel/NPs-mediated ICD therapy with anti-PD-L1 to reshape the tumor immunosuppressive microenvironment, inhibit colon cancer cell proliferation, and enhance therapeutic efficacy. In summary, our findings suggest that HA@Cel/NPs are promising as a novel drug delivery system for drug delivery to treat a range of colon-related diseases.

Methods

Materials

Celastrol (Cel) was acquired from Chengdu DeSiTe Biological Technology Co., Ltd. (Chengdu, China). 4-(hydroxymethyl) phenylboronic acid (PAPE), Succinic anhydride (SA), and Coumarin-6 (C6) were purchased from Aladdin Reagent Co., Ltd (Shanghai, China). 1-Adamantanecarboxylic acid (AD) was provided by Adamas (Shanghai, China). Hyaluronic acid (HA, MW 8 kDa) was bought from Freda Biochem Co., Ltd. (Shandong, China). Mono-(6-amino-6-deoxy)-beta-cyclodextrin (6-NH2-β-CD) originated from Shandong Binzhou Zhiyuan Biotechnology Co., Ltd. (Shandong, China). The near-infrared lipophilic carbocyanine dye 1,10-dioctadecyltetramethyl indotricarbocyanine iodide (DiR) was supplied by Mellon, Biological Technology Co., Ltd. (Dalian, China). Dextran sodium sulfate (DSS, colitis grade, MW 36–50 kDa) was provided by MP Biomedicals Inc. (Irvine, CA, USA). Azoxymethane (AOM) was purchased from FUJIFILM Wako Pure Chemical Co. (Osaka, Japan). Anti-mouse PD-L1 (anti-PD-L1) was bought from Bio X Cell. FITC anti-mouse CD4, APC anti-mouse CD3, PE/Cyanine7 anti-mouse CD8a and PE anti-mouse FOXP3 were purchased from BioLegend (San Diego, USA). Antibodies against cell surface markers (high mobility group box 1 protein (HMGB1) and calreticulin (CRT)) were obtained from Boster Biological Technology Co., Ltd. (Wuhan, China). CT26 cells, NCM460 cells and Raw264.7 cells were obtained from Boster Biological Technology Co., Ltd. (Wuhan, China).

ICR mice, C57BL/6 mice, and Balb/c mice were provided by SPF (Beijing) Biotechnology Co., Ltd. (Beijing, China). All animal experiments were approved by the Animal Ethics Committee of the Chengdu Medical College (Permit NO. 2023–044) and were conducted in accordance with the institutional guidelines for the care and use of laboratory animals, as well as the national standard for laboratory animal environment and facilities (GB 14925–2023). The mice were euthanized by inhalation of 40% CO2.### Preparation and Characterization of HA@Cel/NPs

The PAPE-SA-CD was synthesized following our established method.^[7] Detailed procedures can be found in the Supplementary Materials. The synthesis of HA-AD is elaborated in the Supplementary Materials.

Dissolve 2 mg of Cel and 20 mg of PAPE-SA-CD in 2 mL of DMSO with vortex sonication. Dissolve 10 mg of HA-AD in 10 mL of reverse osmosis (RO) water. Then, slowly add the DMSO mixture solution dropwise to the water solution and stir for 2 h. Transfer the solution to a dialysis bag (MWCO 3000, Millipore) and dialyze to remove DMSO. Filter it through a 0.45 μm filter to obtain the product.

We utilized dynamic light scattering (DLS) with a Nanosizer ZS90 (Malvern, UK) to determine the hydrodynamic diameters (nm), polydispersity index (PDI), and zeta potential (mV) of HA@Cel/NPs. The morphology of HA@Cel/NPs was assessed using a Transmission electron microscope (TEM, JEM 1200X, JEOL, Japan). Furthermore, we conducted X-ray powder diffraction (XRD) analysis on all polymers using an XRD diffractometer (D8 Advance, BRUKER, Germany) in the range of 5 ° to 90 °. Additionally, we monitored the stability of HA@Cel/NPs over a 14-day storage period by measuring their hydrodynamic diameters, PDI, and zeta potentials.

The HPLC analysis was conducted to assess the encapsulation efficiency (EE) and loading efficiency (LE) of HA@Cel/NPs for Cel. The HPLC conditions for Cel quantification were established following our previously reported method.^[32] The details are provided in the Supplementary Materials (Figure S1). The following formula was utilized for calculating EE and LE:### The ROS/Enzyme Dual Sensitivity of HA@Cel/NPs

The sensitivity of HA@Cel/NPs to oxidation or enzymes was investigated under various conditions.^[26] HA@Cel/NPs were exposed to 1 mM H2O2 and 10 IU/mL α-amylase solutions, followed by agitation at 100 rpm at 37°C for 24 h. Subsequently, the average particle diameters of HA@Cel/NPs were measured using DLS, and their morphologies were examined using TEM after the incubation period.### In vitro Drug Release

To evaluate the release profile of HA@Cel/NPs, we utilized the dialysis technique.^[33] Free Cel and HA@Cel/NPs (Cel concentration was 0.2 mg/mL) were dispersed sequentially in various simulated digestive fluids to mimic the complete digestion process. This process involved releasing from 0 to 2 h in simulated gastric fluid (SGF, pH 1.2), from 2 to 6 h in simulated intestinal fluid (SIF, pH 6.8), and from 6 to 48 h in simulated colonic fluid (SCF, pH 7.4) with incorporated Tween-80 (0.5% w/v) into the release medium (30 mL). Furthermore, 3 mL of each specimen was individually introduced into dialysis bags with a molecular weight cut-off (MWCO) of 3000. At specified intervals, we withdrew 1 mL of the simulated solution and replenished it with an equal volume of fresh solution. The concentration of Cel was measured via HPLC.

Additionally, 3 mL of HA@Cel/NPs were introduced into the dialysis bags with a molecular weight cut-off (MWCO) of 3000. Subsequently, they were incubated in pH 7.4 PBS (containing 0.5% (w/v) Tween-80) at 37°C for 48 h, either with or without the presence of 1 mM H2O2 and 10 IU/mL α-amylase, to investigate the release properties of Cel in HA-AD/PAPE-SA-CD-based-NPs within the colon.### In vitro Anti-Inflammatory Activity, Macrophage Polarization and Cellular Uptake Profiles

In order to investigate the in vitro anti-inflammatory and macrophage polarization effects of HA@Cel/NPs, LPS-induced Raw264.7 cells were employed as the cellular model. Briefly, Raw264.7 macrophages were seeded in 12-well plates and incubated overnight. Subsequently, they were co-cultured for 12 h with various Cel preparations (free Cel, Cel/NPs, and HA@Cel/NPs, all containing 0.5 µg/mL of Cel). Following this, the cells were rinsed with PBS and exposed to LPS (1 µg/mL) for 3 h. Finally, the levels of inflammatory cytokines (TNF-α, IL-1β) in the supernatant were assessed using corresponding enzyme-linked immunosorbent assay (ELISA) kits as per the manufacturer’s instructions.

To investigate the potential of HA@Cel/NPs in inducing a shift of macrophages towards the M2 phenotype, immunofluorescence (IF) analysis was conducted.^[34] Primary antibodies targeting CD206 (M2) and CD86 (M1) were utilized and DAPI for nuclear staining. The protein expression levels were assessed using a laser scanning confocal microscope (CLSM; Nikon A1R+SIM; Nikon, Tokyo, Japan).

The cellular uptake of drug-loaded NPs was determined using C6 encapsulated free C6, C6/NPs, and HA@C6/NPs. Raw264.7 cells were used as model cells, and performed quantitative analysis and qualitative analysis of cellular uptake of drugs. CLSM was used to analyze cellular uptake qualitatively. Raw264.7 cells were cultivated in confocal dish overnight and added with free C6, C6/NPs, HA@C6/NPs and HA@C6/NPs+HA (HA incubated 2 h in advance) for 4 h. Removed cultivation medium and rinsed three times with PBS. Afterwards, the cells were fixed with 4% paraformaldehyde for 10 min and stained with DAPI for 15 min. Finally, analyzing the cells by CLSM.

Moreover, flow cytometry (FCM; NovoCyte; ACEA, San Diego, CA) was used to analyze cellular uptake quantitatively. Raw264.7 cells were cultivated for 24 h in 6-well plates. In the time-dependent study, HA@C6/NPs (C6 concentrations of 100 ng/mL) was added to 6-well plates and incubated for different time (0.25, 0.5, 1, 2, 4 h), respectively. In the concentration-dependent study, various concentrations of HA@C6/NPs (25, 50, 100, 150, 200 ng/mL) was added and incubated for 4 h. In addition, the Raw264.7 cells were also treated with variant formulation (free C6, C6/NPs, HA@C6/NPs and HA@C6/NPs+HA (incubated 2 h in advance)) for 4 h. Lastly, fluorescence content was measured by FCM.### In vitro Anticancer Activity and Cellular Uptake Profiles

To assess the material’s biocompatibility, we performed a CCK8 assay on CT26 cells. CT26 cells were taken and inoculated into 96-well plates overnight. Different concentrations of PAPE-SA-CD or PAPE-SA-CD/HA-AD were co-cultured with cells. To evaluate the safety and selectivity of the formulations, the cytotoxicity of free Cel, Cel/NPs, and HA@Cel/NPs against normal human colonic epithelial cells (NCM460) was assessed using the CCK-8 assay. Briefly, NCM460 cells were seeded in 96-well plates and cultured overnight. Cells were then treated with various concentrations (0.125, 0.25, 0.5, 1, 2, and 4 μg/mL, based on Cel concentration) of the indicated formulations for 24 h. Furthermore, in the study of drug cytotoxicity, CT26 cells were incubated with free Cel, Cel/NPs, and HA@Cel/NPs at different Cel concentrations (0.45, 0.90, 1.80, 3.60, 5.40, 7.20 μg/mL) for 48 h, respectively. The cell viability was assessed using a standard CCK8 assay kit.

CT26 cells were inoculated uniformly in twelve-well plates at 2×105 cells in each well and were cultivated with free Cel, Cel/NPs and HA@Cel/NPs (Cel content is 1 μg/mL) for 24 h. Subsequently, the cells underwent dissociation into single-cell suspensions and were labeled with Annexin V-Alexa Fluor488/PI. Finally, the cells were harvested for FCM analysis using a NovoCyte instrument (ACEA, San Diego, CA).

CT26 cells were used as carcinoma of colon model cells and the process of cellular uptake is consistent across Raw264.7 cells.### In vivo and ex vivo Imaging

NPs loaded with DiR were formulated to enable real-time fluorescence tracking in vivo and ex vivo, encompassing free DiR, DiR/NPs, and HA@DiR/NPs.^[35] AOM/DSS-induced CAC model mice (AOM, 10 mg/kg; DSS, 2.5%, w/v) were randomly divided into three groups, including free DiR, DiR/NPs and HA@DiR/NPs (the concentration of DiR was 3 mg/kg). The mice were photographed at different times (6, 12 and 24 h) by the IVIS system (PerkinElmer, Waltham, MA, USA). The mice were sacrificed after being given drugs for 24 h and then the small intestine, cecum, colon segments and main organs were imaged. Fluorescence image quantification was performed using IVIS Living Image software, with results reported as mean radiance (p/s/cm2/sr).### In vivo Target Properties

CAC model mice were randomly divided into three groups, including free C6, C6/NPs, HA@C6/NPs (the concentration of C6 was 5 mg/kg). After 12 h, the colons were collected and soaked in 4% paraformaldehyde for 20 min and sliced into 8 μm flakes. They were rinsed with PBS for 5 min, dyed with DAPI for 15 min. Ultimately, the fluorescence in the colon was observed via CLSM.### In vivo Therapeutic Effect of HA@Cel/NPs Against UC

Male ICR mice were randomly allocated into five groups (n=6), with one group serving as the control (Saline) and the remaining four groups receiving 3% (w/v) DSS for 7 days to induce the UC mouse model. From day 3 to day 10, mice were orally administered different preparations daily, including free Cel, Cel/NPs, and HA@Cel/NPs, at a Cel dose of 2 mg/kg per day. These formulations were prepared by dispersing the lyophilized HA@Cel/NPs and Cel/NPs powder in sterile phosphate-buffered saline (PBS, pH 7.4), followed by gentle sonication to obtain a homogeneous suspension. For the free Cel control group, Cel was first dissolved in a minimal amount of DMSO and then diluted with sterile PBS containing 10% PEG400 to achieve the final injection concentration, with the final organic solvent content kept below 5% to avoid systemic toxicity. The control group was given normal drinking water, while the model group was treated with an equal amount of saline solution. The mice were weighed daily, and body mass index, stool shape, and bleeding were recorded to assess the disease activity index (DAI) (scoring criteria detailed in Table S1). At the end of the experiment (day 10), the mice were sacrificed and the colon as well as major organs (heart, liver, spleen, lung, and kidney) were collected. Furthermore, the length of the colon and the weight of the spleen were measured. Finally, we stained the colon tissue with hematoxylin/eosin (H&E) and periodic acid schiff (PAS), respectively.### In vivo Therapeutic Effect of HA@Cel/NPs Against CAC

C57BL/6 mice were divided into 5 groups: the control group, the model group, the free Cel group, the Cel/NPs group and the HA@Cel/NPs group. The formulation preparation details are described in In Vivo Therapeutic Effect Of HA@Cel/Nps Against UC. In vivo CAC mouse model was established in C57BL/6 mice using the AOM and DSS method.^[36] In detail, mice were intraperitoneally injected with AOM at a dosage of 10 mg/kg and maintained for 7 days. Subsequently, mice received three cycles of DSS treatment, each consisting of 7 days of 2.5% (w/v) DSS in drinking water followed by a 14-day recovery period with normal water. After the third DSS cycle, mice were switched to normal drinking water and received oral administration of different formulations (free Cel, Cel/NPs, or HA@Cel/NPs) at a Cel dose of 2 mg/kg every other day for 21 consecutive days (ie, on days 1, 3, 5, 21 post-DSS withdrawal). Body weights were documented weekly during the entire duration of the experiment. At the conclusion of the treatment (day 70), colons and major organs were harvested for histological analysis, and the quantities and size distributions of tumors in each experimental group were subjected to statistical examination. For histological assessment, colon sections were stained with H&E to evaluate general morphology, inflammatory cell infiltration, and tissue damage. In addition, immunohistochemical (IHC) staining for Ki-67 (a marker of cell proliferation) was performed on colon sections to assess the proliferative activity of epithelial cells. The major organs were also processed for H&E staining to evaluate the systemic biosafety of the treatments.### In vivo Combination Therapy of HA@Cel/NPs with Anti-PD-L1

Balb/c mice received a subcutaneous injection of 100 μL of cell suspension containing 1×107 CT26 cells in the right groin region. After 7–10 days following injection, the tumors reached a volume of approximately 100 mm3. Subsequently, mice were randomly divided into five groups (n = 6): (i) saline (control), (ii) free Cel, (iii) Cel/NPs, (iv) HA@Cel/NPs, and (v) HA@Cel/NPs + anti-PD-L1. For the free Cel, Cel/NPs, and HA@Cel/NPs groups, the corresponding formulations were intravenously administered every other day at a Cel dose of 3 mg/kg, for a total of seven injections (on days 1, 3, 5, 7, 9, 11, and 13). For the combination group (HA@Cel/NPs + anti-PD-L1), HA@Cel/NPs was administered intravenously at the same dose and schedule as described above, while anti-PD-L1 was given intraperitoneally at 5 mg/kg on days 3, 6, and 9 (three doses in total).^[28],^[37] The control group received an equal volume of saline. Tumor volume was monitored every two days throughout the treatment period, and all mice were euthanized on day 14 for further analysis. The formulation preparation details are described in Section 2.9. Tumor volume was assessed every 2 days during treatment using a caliper and subsequently calculated. Body weight change was also recorded. Following a 14-day period of treatment, the mice were euthanized, and the tumors as well as major organs were removed for subsequent analysis. The harvested tumors were immediately photographed and weighed. H&E staining was performed to assess general tumor morphology and necrotic areas. In addition, immunofluorescence (IF) staining against HMGB1 (high-mobility group box 1) and CRT (calreticulin) was conducted on tumor sections to evaluate the expression of immunogenic cell death (ICD)-related markers. The major organs were also processed for H&E staining to evaluate the systemic biosafety of the treatments.

To evaluate the antitumor immune response, tumor and spleen tissues were collected and processed to prepare single-cell suspensions. Stained cells were analyzed using FCM, and data were processed with FlowJo software. The percentages of CD8⁺ and CD4⁺ T cells were calculated, and the proportion of Foxp3⁺ cells within the CD4⁺ T-cell population was determined to evaluate the regulatory T-cell response.### Statistical Analysis

All data are presented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using GraphPad Prism 10. For comparisons between two groups, an unpaired two-tailed Student’s t-test was applied. For comparisons among multiple groups, one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was used for single time-point comparisons, while two-way ANOVA followed by Bonferroni’s post-hoc test was applied for repeated-measures data. A value of P < 0.05 was considered statistically significant, with significance levels denoted as *P < 0.05, P < 0.01, and *P < 0.001.

Materials

Celastrol (Cel) was acquired from Chengdu DeSiTe Biological Technology Co., Ltd. (Chengdu, China). 4-(hydroxymethyl) phenylboronic acid (PAPE), Succinic anhydride (SA), and Coumarin-6 (C6) were purchased from Aladdin Reagent Co., Ltd (Shanghai, China). 1-Adamantanecarboxylic acid (AD) was provided by Adamas (Shanghai, China). Hyaluronic acid (HA, MW 8 kDa) was bought from Freda Biochem Co., Ltd. (Shandong, China). Mono-(6-amino-6-deoxy)-beta-cyclodextrin (6-NH2-β-CD) originated from Shandong Binzhou Zhiyuan Biotechnology Co., Ltd. (Shandong, China). The near-infrared lipophilic carbocyanine dye 1,10-dioctadecyltetramethyl indotricarbocyanine iodide (DiR) was supplied by Mellon, Biological Technology Co., Ltd. (Dalian, China). Dextran sodium sulfate (DSS, colitis grade, MW 36–50 kDa) was provided by MP Biomedicals Inc. (Irvine, CA, USA). Azoxymethane (AOM) was purchased from FUJIFILM Wako Pure Chemical Co. (Osaka, Japan). Anti-mouse PD-L1 (anti-PD-L1) was bought from Bio X Cell. FITC anti-mouse CD4, APC anti-mouse CD3, PE/Cyanine7 anti-mouse CD8a and PE anti-mouse FOXP3 were purchased from BioLegend (San Diego, USA). Antibodies against cell surface markers (high mobility group box 1 protein (HMGB1) and calreticulin (CRT)) were obtained from Boster Biological Technology Co., Ltd. (Wuhan, China). CT26 cells, NCM460 cells and Raw264.7 cells were obtained from Boster Biological Technology Co., Ltd. (Wuhan, China).

ICR mice, C57BL/6 mice, and Balb/c mice were provided by SPF (Beijing) Biotechnology Co., Ltd. (Beijing, China). All animal experiments were approved by the Animal Ethics Committee of the Chengdu Medical College (Permit NO. 2023–044) and were conducted in accordance with the institutional guidelines for the care and use of laboratory animals, as well as the national standard for laboratory animal environment and facilities (GB 14925–2023). The mice were euthanized by inhalation of 40% CO2.

Preparation and Characterization of HA@Cel/NPs

The PAPE-SA-CD was synthesized following our established method.^[7] Detailed procedures can be found in the Supplementary Materials. The synthesis of HA-AD is elaborated in the Supplementary Materials.

Dissolve 2 mg of Cel and 20 mg of PAPE-SA-CD in 2 mL of DMSO with vortex sonication. Dissolve 10 mg of HA-AD in 10 mL of reverse osmosis (RO) water. Then, slowly add the DMSO mixture solution dropwise to the water solution and stir for 2 h. Transfer the solution to a dialysis bag (MWCO 3000, Millipore) and dialyze to remove DMSO. Filter it through a 0.45 μm filter to obtain the product.

We utilized dynamic light scattering (DLS) with a Nanosizer ZS90 (Malvern, UK) to determine the hydrodynamic diameters (nm), polydispersity index (PDI), and zeta potential (mV) of HA@Cel/NPs. The morphology of HA@Cel/NPs was assessed using a Transmission electron microscope (TEM, JEM 1200X, JEOL, Japan). Furthermore, we conducted X-ray powder diffraction (XRD) analysis on all polymers using an XRD diffractometer (D8 Advance, BRUKER, Germany) in the range of 5 ° to 90 °. Additionally, we monitored the stability of HA@Cel/NPs over a 14-day storage period by measuring their hydrodynamic diameters, PDI, and zeta potentials.

The HPLC analysis was conducted to assess the encapsulation efficiency (EE) and loading efficiency (LE) of HA@Cel/NPs for Cel. The HPLC conditions for Cel quantification were established following our previously reported method.^[32] The details are provided in the Supplementary Materials (Figure S1). The following formula was utilized for calculating EE and LE:

The ROS/Enzyme Dual Sensitivity of HA@Cel/NPs

The sensitivity of HA@Cel/NPs to oxidation or enzymes was investigated under various conditions.^[26] HA@Cel/NPs were exposed to 1 mM H2O2 and 10 IU/mL α-amylase solutions, followed by agitation at 100 rpm at 37°C for 24 h. Subsequently, the average particle diameters of HA@Cel/NPs were measured using DLS, and their morphologies were examined using TEM after the incubation period.

In vitro Drug Release

To evaluate the release profile of HA@Cel/NPs, we utilized the dialysis technique.^[33] Free Cel and HA@Cel/NPs (Cel concentration was 0.2 mg/mL) were dispersed sequentially in various simulated digestive fluids to mimic the complete digestion process. This process involved releasing from 0 to 2 h in simulated gastric fluid (SGF, pH 1.2), from 2 to 6 h in simulated intestinal fluid (SIF, pH 6.8), and from 6 to 48 h in simulated colonic fluid (SCF, pH 7.4) with incorporated Tween-80 (0.5% w/v) into the release medium (30 mL). Furthermore, 3 mL of each specimen was individually introduced into dialysis bags with a molecular weight cut-off (MWCO) of 3000. At specified intervals, we withdrew 1 mL of the simulated solution and replenished it with an equal volume of fresh solution. The concentration of Cel was measured via HPLC.

Additionally, 3 mL of HA@Cel/NPs were introduced into the dialysis bags with a molecular weight cut-off (MWCO) of 3000. Subsequently, they were incubated in pH 7.4 PBS (containing 0.5% (w/v) Tween-80) at 37°C for 48 h, either with or without the presence of 1 mM H2O2 and 10 IU/mL α-amylase, to investigate the release properties of Cel in HA-AD/PAPE-SA-CD-based-NPs within the colon.

In vitro Anti-Inflammatory Activity, Macrophage Polarization and Cellular Uptake Profiles

In order to investigate the in vitro anti-inflammatory and macrophage polarization effects of HA@Cel/NPs, LPS-induced Raw264.7 cells were employed as the cellular model. Briefly, Raw264.7 macrophages were seeded in 12-well plates and incubated overnight. Subsequently, they were co-cultured for 12 h with various Cel preparations (free Cel, Cel/NPs, and HA@Cel/NPs, all containing 0.5 µg/mL of Cel). Following this, the cells were rinsed with PBS and exposed to LPS (1 µg/mL) for 3 h. Finally, the levels of inflammatory cytokines (TNF-α, IL-1β) in the supernatant were assessed using corresponding enzyme-linked immunosorbent assay (ELISA) kits as per the manufacturer’s instructions.

To investigate the potential of HA@Cel/NPs in inducing a shift of macrophages towards the M2 phenotype, immunofluorescence (IF) analysis was conducted.^[34] Primary antibodies targeting CD206 (M2) and CD86 (M1) were utilized and DAPI for nuclear staining. The protein expression levels were assessed using a laser scanning confocal microscope (CLSM; Nikon A1R+SIM; Nikon, Tokyo, Japan).

The cellular uptake of drug-loaded NPs was determined using C6 encapsulated free C6, C6/NPs, and HA@C6/NPs. Raw264.7 cells were used as model cells, and performed quantitative analysis and qualitative analysis of cellular uptake of drugs. CLSM was used to analyze cellular uptake qualitatively. Raw264.7 cells were cultivated in confocal dish overnight and added with free C6, C6/NPs, HA@C6/NPs and HA@C6/NPs+HA (HA incubated 2 h in advance) for 4 h. Removed cultivation medium and rinsed three times with PBS. Afterwards, the cells were fixed with 4% paraformaldehyde for 10 min and stained with DAPI for 15 min. Finally, analyzing the cells by CLSM.

Moreover, flow cytometry (FCM; NovoCyte; ACEA, San Diego, CA) was used to analyze cellular uptake quantitatively. Raw264.7 cells were cultivated for 24 h in 6-well plates. In the time-dependent study, HA@C6/NPs (C6 concentrations of 100 ng/mL) was added to 6-well plates and incubated for different time (0.25, 0.5, 1, 2, 4 h), respectively. In the concentration-dependent study, various concentrations of HA@C6/NPs (25, 50, 100, 150, 200 ng/mL) was added and incubated for 4 h. In addition, the Raw264.7 cells were also treated with variant formulation (free C6, C6/NPs, HA@C6/NPs and HA@C6/NPs+HA (incubated 2 h in advance)) for 4 h. Lastly, fluorescence content was measured by FCM.

In vitro Anticancer Activity and Cellular Uptake Profiles

To assess the material’s biocompatibility, we performed a CCK8 assay on CT26 cells. CT26 cells were taken and inoculated into 96-well plates overnight. Different concentrations of PAPE-SA-CD or PAPE-SA-CD/HA-AD were co-cultured with cells. To evaluate the safety and selectivity of the formulations, the cytotoxicity of free Cel, Cel/NPs, and HA@Cel/NPs against normal human colonic epithelial cells (NCM460) was assessed using the CCK-8 assay. Briefly, NCM460 cells were seeded in 96-well plates and cultured overnight. Cells were then treated with various concentrations (0.125, 0.25, 0.5, 1, 2, and 4 μg/mL, based on Cel concentration) of the indicated formulations for 24 h. Furthermore, in the study of drug cytotoxicity, CT26 cells were incubated with free Cel, Cel/NPs, and HA@Cel/NPs at different Cel concentrations (0.45, 0.90, 1.80, 3.60, 5.40, 7.20 μg/mL) for 48 h, respectively. The cell viability was assessed using a standard CCK8 assay kit.

CT26 cells were inoculated uniformly in twelve-well plates at 2×105 cells in each well and were cultivated with free Cel, Cel/NPs and HA@Cel/NPs (Cel content is 1 μg/mL) for 24 h. Subsequently, the cells underwent dissociation into single-cell suspensions and were labeled with Annexin V-Alexa Fluor488/PI. Finally, the cells were harvested for FCM analysis using a NovoCyte instrument (ACEA, San Diego, CA).

CT26 cells were used as carcinoma of colon model cells and the process of cellular uptake is consistent across Raw264.7 cells.

In vivo and ex vivo Imaging

NPs loaded with DiR were formulated to enable real-time fluorescence tracking in vivo and ex vivo, encompassing free DiR, DiR/NPs, and HA@DiR/NPs.^[35] AOM/DSS-induced CAC model mice (AOM, 10 mg/kg; DSS, 2.5%, w/v) were randomly divided into three groups, including free DiR, DiR/NPs and HA@DiR/NPs (the concentration of DiR was 3 mg/kg). The mice were photographed at different times (6, 12 and 24 h) by the IVIS system (PerkinElmer, Waltham, MA, USA). The mice were sacrificed after being given drugs for 24 h and then the small intestine, cecum, colon segments and main organs were imaged. Fluorescence image quantification was performed using IVIS Living Image software, with results reported as mean radiance (p/s/cm2/sr).

In vivo Target Properties

CAC model mice were randomly divided into three groups, including free C6, C6/NPs, HA@C6/NPs (the concentration of C6 was 5 mg/kg). After 12 h, the colons were collected and soaked in 4% paraformaldehyde for 20 min and sliced into 8 μm flakes. They were rinsed with PBS for 5 min, dyed with DAPI for 15 min. Ultimately, the fluorescence in the colon was observed via CLSM.

In vivo Therapeutic Effect of HA@Cel/NPs Against UC

Male ICR mice were randomly allocated into five groups (n=6), with one group serving as the control (Saline) and the remaining four groups receiving 3% (w/v) DSS for 7 days to induce the UC mouse model. From day 3 to day 10, mice were orally administered different preparations daily, including free Cel, Cel/NPs, and HA@Cel/NPs, at a Cel dose of 2 mg/kg per day. These formulations were prepared by dispersing the lyophilized HA@Cel/NPs and Cel/NPs powder in sterile phosphate-buffered saline (PBS, pH 7.4), followed by gentle sonication to obtain a homogeneous suspension. For the free Cel control group, Cel was first dissolved in a minimal amount of DMSO and then diluted with sterile PBS containing 10% PEG400 to achieve the final injection concentration, with the final organic solvent content kept below 5% to avoid systemic toxicity. The control group was given normal drinking water, while the model group was treated with an equal amount of saline solution. The mice were weighed daily, and body mass index, stool shape, and bleeding were recorded to assess the disease activity index (DAI) (scoring criteria detailed in Table S1). At the end of the experiment (day 10), the mice were sacrificed and the colon as well as major organs (heart, liver, spleen, lung, and kidney) were collected. Furthermore, the length of the colon and the weight of the spleen were measured. Finally, we stained the colon tissue with hematoxylin/eosin (H&E) and periodic acid schiff (PAS), respectively.

In vivo Therapeutic Effect of HA@Cel/NPs Against CAC

C57BL/6 mice were divided into 5 groups: the control group, the model group, the free Cel group, the Cel/NPs group and the HA@Cel/NPs group. The formulation preparation details are described in In Vivo Therapeutic Effect Of HA@Cel/Nps Against UC. In vivo CAC mouse model was established in C57BL/6 mice using the AOM and DSS method.^[36] In detail, mice were intraperitoneally injected with AOM at a dosage of 10 mg/kg and maintained for 7 days. Subsequently, mice received three cycles of DSS treatment, each consisting of 7 days of 2.5% (w/v) DSS in drinking water followed by a 14-day recovery period with normal water. After the third DSS cycle, mice were switched to normal drinking water and received oral administration of different formulations (free Cel, Cel/NPs, or HA@Cel/NPs) at a Cel dose of 2 mg/kg every other day for 21 consecutive days (ie, on days 1, 3, 5, 21 post-DSS withdrawal). Body weights were documented weekly during the entire duration of the experiment. At the conclusion of the treatment (day 70), colons and major organs were harvested for histological analysis, and the quantities and size distributions of tumors in each experimental group were subjected to statistical examination. For histological assessment, colon sections were stained with H&E to evaluate general morphology, inflammatory cell infiltration, and tissue damage. In addition, immunohistochemical (IHC) staining for Ki-67 (a marker of cell proliferation) was performed on colon sections to assess the proliferative activity of epithelial cells. The major organs were also processed for H&E staining to evaluate the systemic biosafety of the treatments.

In vivo Combination Therapy of HA@Cel/NPs with Anti-PD-L1

Balb/c mice received a subcutaneous injection of 100 μL of cell suspension containing 1×107 CT26 cells in the right groin region. After 7–10 days following injection, the tumors reached a volume of approximately 100 mm3. Subsequently, mice were randomly divided into five groups (n = 6): (i) saline (control), (ii) free Cel, (iii) Cel/NPs, (iv) HA@Cel/NPs, and (v) HA@Cel/NPs + anti-PD-L1. For the free Cel, Cel/NPs, and HA@Cel/NPs groups, the corresponding formulations were intravenously administered every other day at a Cel dose of 3 mg/kg, for a total of seven injections (on days 1, 3, 5, 7, 9, 11, and 13). For the combination group (HA@Cel/NPs + anti-PD-L1), HA@Cel/NPs was administered intravenously at the same dose and schedule as described above, while anti-PD-L1 was given intraperitoneally at 5 mg/kg on days 3, 6, and 9 (three doses in total).^[28],^[37] The control group received an equal volume of saline. Tumor volume was monitored every two days throughout the treatment period, and all mice were euthanized on day 14 for further analysis. The formulation preparation details are described in Section 2.9. Tumor volume was assessed every 2 days during treatment using a caliper and subsequently calculated. Body weight change was also recorded. Following a 14-day period of treatment, the mice were euthanized, and the tumors as well as major organs were removed for subsequent analysis. The harvested tumors were immediately photographed and weighed. H&E staining was performed to assess general tumor morphology and necrotic areas. In addition, immunofluorescence (IF) staining against HMGB1 (high-mobility group box 1) and CRT (calreticulin) was conducted on tumor sections to evaluate the expression of immunogenic cell death (ICD)-related markers. The major organs were also processed for H&E staining to evaluate the systemic biosafety of the treatments.

To evaluate the antitumor immune response, tumor and spleen tissues were collected and processed to prepare single-cell suspensions. Stained cells were analyzed using FCM, and data were processed with FlowJo software. The percentages of CD8⁺ and CD4⁺ T cells were calculated, and the proportion of Foxp3⁺ cells within the CD4⁺ T-cell population was determined to evaluate the regulatory T-cell response.

Statistical Analysis

All data are presented as mean ± standard deviation (SD) from at least three independent experiments. Statistical analyses were performed using GraphPad Prism 10. For comparisons between two groups, an unpaired two-tailed Student’s t-test was applied. For comparisons among multiple groups, one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was used for single time-point comparisons, while two-way ANOVA followed by Bonferroni’s post-hoc test was applied for repeated-measures data. A value of P < 0.05 was considered statistically significant, with significance levels denoted as *P < 0.05, P < 0.01, and *P < 0.001.

Results

Characterization of HA@Cel/NPs

The synthetic hydrogen spectra of PAPE-SA-CD and HA-AD are presented in Figures S2 and S3, respectively. The HA@Cel/NPs were prepared according to a nanoprecipitation method.^[38] The experimental results show that the average particle size of HA@Cel/NPs is about 76.87 ± 2.65 nm, and the PDI is about 0.166 ± 0.012 (Figure 1A). TEM revealed that HA@Cel/NPs exhibited a spherical morphology with smooth surfaces (Figure S4). In addition, compared with Cel/NPs, HA@Cel/NPs exhibited a slightly higher EE (90.85 ± 3.69% vs 88.37 ± 4.36%) (Table S2), which could be attributed to the formation of the HA shell that minimized drug leakage during purification. Successful HA coating was confirmed by the increased particle size and the zeta potential shift from −15.5 mV to −24.7 mV (Figure 1B). XRD analysis was performed to study the effect of drug-carrier interaction on drug release characteristics. There are multiple peaks between 5° and 90° in free Cel, indicating that its crystallinity is high. But these spikes are not observed in HA@Cel/NPs, indicating that there is no crystalline complex between Cel and the matrix (Figure 1C). Therefore, Cel exists in the form of amorphous or disordered crystals.### The Stability of HA@Cel/NPs

The stability evaluation results after 14 days of storage at 4°C are shown in Figure 1D and E. It could be observed that the particle size, PDI and zeta potential of HA@Cel/NPs did not change much. This suggests that HA@Cel/NPs can retain stable properties for a period of 14 days.### ROS and Enzyme Sensitivity Evaluation of HA@Cel/NPs

The sensitivity of HA@Cel/NPs to ROS or α-amylase was verified by using DLS and TEM. As shown in Figure 1F, the size distribution and morphology of HA@Cel/NPs had little change in the PBS solution. However, an increase in size and the appearance of heterogeneous peaks were observed when HA@Cel/NPs were incubated with 1 mM H2O2 or 10 IU/mL α-amylase for 24 h. Further, the results of TEM showed that the presence of H2O2 or α-amylase led to the cracking of HA@Cel/NPs (Figure 1F), which may be due to ROS sensitivity caused by H2O2 destruction or enzyme sensitivity under α-amylase. Consequently, it is anticipated that HA@Cel/NPs exhibit monosensitivity to ROS or α-amylase.### Release Profiles of HA@Cel/NPs

The release of contents from free Cel or HA@Cel/NPs was assessed throughout the entire gastrointestinal tract (continuously placed in an alternative medium). Figure 1G showed that free Cel was released rapidly in SGF and SIF with a cumulative release rate of 57.73% within 6 h, so that it was heavily absorbed in the stomach and small intestine and less released in the colon tissue. In contrast, HA@Cel/NPs exhibited slower release properties, allowing the drug to reach its target location more completely. Moreover, the cumulative release rate was 39.95% after 48 h. As a result of the protection provided by PAPE-SA-CD and HA-AD, HA@Cel/NPs can effectively transit from the stomach and small intestine to reach the colon tissue.

In addition, the synergistic action of ROS and enzyme stimulators were used to simulate drug release behavior at the colon site. The release medium included pH 7.4 buffer, pH 7.4 buffer containing 1 mM H2O2, pH 7.4 buffer containing 10 IU/mL α-amylase, pH 7.4 buffer containing 1 mM H2O2 and 10 IU/mL α-amylase (Figure 1H). Compared to pH 7.4 buffer, the cumulative release rate was significantly accelerated in the presence of H2O2 or α-amylase, which reached 55.89% and 68.77% within 6 h, respectively. Additionally, the combined action of H2O2 and α-amylase resulted in a cumulative release rate of 71.87% for HA@Cel/NPs within 8 h, representing the highest level. This suggests that drug release of HA@Cel/NPs can be controlled by internal stimulation of ROS/enzymes.### In vitro Anti-Inflammatory Activity, Macrophage Polarization and Cellular Uptake Profiles

Anti-Inflammatory Activity of HA@Cel/NPs

The pharmacological study of HA@Cel/NPs in vitro was carried out using LPS-induced Raw264.7 macrophages, as depicted in Figure 2. As shown in Figure 2A and B, the secretion of pro-inflammatory cytokines TNF-α and IL-1β significantly increased in LPS-treated (positive control) cells compared to the negative control cells (P 0.05), whereas Cel/NPs (160.3 mg) and HA@Cel/NPs (152.5 mg) significantly reduced spleen weight compared with the model group (P < 0.01). H&E staining was used to evaluate the pathological changes of colon tissue after drug intervention (Figure 5G). The colonic tissue of the model group showed that the crypt structure was obviously destroyed, goblet cells were obviously reduced, and inflammatory cells were obviously infiltrated. Compared with the model group, the groups given different Cel preparations showed that the crypt structure was protected and the goblet cells were restored. In addition, according to PAS staining of colonic goblet cells in the model group, goblet cells in the crypt structure were severely damaged, and goblet cells in the free Cel group, Cel/NPs group and HA@Cel/NPs group were all varying degrees of protection (Figure 5H). What is astonishing is that the HA@Cel/NPs group has the best protection effect. In general, the findings indicate that the therapeutic effect of the HA@Cel/NPs group on UC is superior to that of both the free Cel and Cel/NPs groups. Therefore, our results demonstrated that HA@Cel/NPs had a significant alleviating effect on DSS-induced UC mice. Moreover, examination of H&E sections of the primary organs revealed no significant pathological abnormalities or injuries in the major organs (Figure S6), indicating that HA@Cel/NPs possess outstanding biocompatibility and minimal toxicity.### Therapeutic Efficacy and Safety of HA@Cel/NPs in CAC Mouse Model

The study utilized a murine model of CAC induced by AOM/DSS to assess the potential of oral HA@Cel/NPs in inhibiting the progression of CAC (Figure 6). The animal model and the course of drug administration is illustrated in Figure 6A. As shown in Figure 6B, by observing the trend of mouse body weight change, it was found that the body weight of mice in all groups decreased to some extent compared with that of control group. Mice in the model group exhibited a significant body weight loss at the end of the experiment, reaching 86% of their initial weight. After being treated with different Cel preparations, the weight loss of the mice in the HA@Cel/NPs, Cel/NPs and free Cel groups was significantly alleviated, indicating that all three groups had a certain degree of alleviation of the weight loss induced by AOM/DSS. In addition, the image in Figure 6C depicts a representative photo of the colons, and we can find that the colonic mucosa of the mice in the control group was free of congestion and oedema as well as did not show the characteristics of tumor nodules. However, the surface of the colon and its interior in the model group showed obvious diffuse congestion of the mucosa, oedema throughout the body, accompanied by the enlargement of the prominent tumor nodules. The free Cel group, the Cel/NPs group and the HA@Cel/NPs group showed different degrees of improvement compared with those in the model group. What was truly remarkable was that the colon of the HA@Cel/NPs group exhibited no discernible differences from that of the normal group, effectively thwarting inflammation’s progression into colon cancer. Additionally, since the degree of atrophy of the mouse colon is one of the important indicators to observe the degree of transformation of colitis to colon cancer, we observed and recorded the degree of atrophy of the colon of each group of mice. As shown in Figure 6D, the colon length in the model group (5.2 cm) was significantly shortened compared with the control group (8.8 cm, P < 0.001), indicating severe colonic atrophy in the CAC model. Treatment with free Cel (6.5 cm), Cel/NPs (7.3 cm), and HA@Cel/NPs (8.2 cm) significantly attenuated AOM/DSS-induced colonic shortening to varying extents. Notably, HA@Cel/NPs exhibited the strongest protective effect among all treatment groups, restoring colon length to near-normal levels, which was significantly longer than that of the free Cel and Cel/NPs groups (P < 0.05). Moreover, the spleen is the largest peripheral immune organ of the body, and the mouse spleen index can reflect the severity of inflammation as a marker of changes in the state of the immune system. As shown in Figure 6E, compared with the control group, the spleen index of the model group was significantly higher (P < 0.001). However, after being treated with free Cel, Cel/NPs and HA@Cel/NPs, the spleen index of mice showed different degrees of reduction, which indicated that our drugs could reduce inflammation, further preventing the progression towards colon cancer.

Moreover, we also observed and recorded the total number of tumor nodules in each group of mice, and the statistical results are shown in Figure 6F. The average total number of tumor nodules in the model group reached 10, whereas there was a different degree of reduction in each of the Cel-delivered groups, among which the average total number of tumor nodules in the HA@Cel/NPs group was the smallest, and was even reduced to 2, which better inhibited the occurrence of the colitis-carcinoma transformation. We further statistically analyzed the size of colon tumor nodules, and the results are shown in Figure 6G. Compared with the model group, the average number of tumors with tumor sizes of 3mm in the HA@Cel/NPs group were all reduced (P < 0.05), and the number of tumors >3mm was close to 0, indicating that the HA@Cel/NPs group had the inhibitory effect on the inflammatory-cancerous transformation of mice colon, and could inhibit the occurrence of inflammatory-cancerous transformation of colon.

Furthermore, the effectiveness of HA@Cel/NPs was assessed using H&E staining and Ki67 staining techniques. As shown in Figure 6H, we can observe that the colonic mucosa of mice in the control group was structurally intact, with well-arranged glands, crypts and goblet cells, and the submucosa was not congested and edematous, with no lesions or inflammation. Compared with the control group, the model group showed a partially defective mucosal epithelium, infiltration of a large number of tumor cells, mucosal edema, extensive necrosis of crypts. The HA@Cel/NPs group had the most obvious protective effect on the colonic epithelial cells, with almost no accumulation of tumor cells in the colonic mucosa, no obvious abnormality in the epithelial mucosa and glandular structure, a more intact crypt lumen, obvious goblet cells, a reduction in the degree of inflammation and malignancy of tumors, and a morphology similar to that of the control group. The above results showed that HA@Cel/NPs had the inhibitory effects on colitis-cancer transformation. We conducted additional assessments of the proliferation patterns in relation to CAC using Ki67 staining. The results illustrated in Figure 6H indicate that the group of mice treated with HA@Cel/NPs exhibited a lower number of Ki67-positive cells in comparison to the other groups. Additionally, the H&E results for the main organs indicated the absence of any notable pathological irregularities or damage (Figure S7), suggesting that HA@Cel/NPs exhibit excellent biocompatibility and minimal toxicity on CAC model mice.### Combination Therapy of HA@Cel/NPs with Anti-PD-L1 in the CT26 Mouse Model of Colorectal Cancer

Utilizing immune checkpoint blockades (anti-PD-L1) may enhance the ability of cytotoxic T lymphocytes to eliminate tumors, offering a potential treatment for various types of cancer. Recent research has shown that when combined with other therapies, PD-1/PD-L1 blockades can significantly boost their immunotherapeutic effects.^[42] Therefore, it is anticipated that combining anti-PD-L1 with HA@Cel/NPs could lead to positive outcomes in anti-tumor treatment. In order to verify this, we assessed the effectiveness of HA@Cel/NPs + anti-PD-L1 on a CT26 tumor model, and the experimental procedure was illustrated in Figure 7A. As depicted in Figure 7B, in the treatment group receiving free Cel, a slight decrease in mouse weight was observed within 14 days, whereas the weights of mice in the other groups remained stable. The findings indicate that the provided NPs exhibits specific biocompatibility and does not induce weight reduction in mice. As shown in Figure 7C, the tumor volume in the saline group exhibited a rapid and sustained increase, reaching 1478 mm3 by the end of the experiment. In contrast, all treatment groups inhibited tumor growth to varying degrees. The mean tumor volumes at the endpoint were 1152 mm3 for the free Cel group, 998 mm3 for the Cel/NPs group, 814 mm3 for the HA@Cel/NPs group, and 412 mm3 for the HA@Cel/NPs + anti-PD-L1 combination group. Notably, the combination therapy with HA@Cel/NPs and anti-PD-L1 exhibited the most potent antitumor efficacy. On day 14, the mice were euthanized and removed tumors (Figure 7D). As shown in Figure 7E, the trend of tumor weight was in accordance with the outcomes of tumors volume.

Meanwhile, H&E staining of tumor tissues was used to further explore the antitumor efficacy in vivo. As displayed in Figure 7F, tumor cells were densely accumulated in the saline group, while in other treatment groups, tumor cells displayed looser, destroyed structures and necrosis could be observed, especially in HA@Cel/NPs and HA@Cel/NPs + anti-PD-L1. Besides, as key factors to mediate specific antitumor immunity, HMGB1 and CRT were measured by immunofluorescence staining (Figure 7G).^[28] Compared with saline group, Cel/NPs, HA@Cel/NPs, and HA@Cel/NPs+anti-PD-L1 treatment significantly increased the HMGB1 and CRT levels in CT26 tumors, indicating that they triggered the specific anti-tumor immune responses of the body. Besides, HA@Cel/NPs+anti-PD-L1 achieved the best results, which could motivate Cel to inhibit tumors effectively and stimulate the anti-tumor immune responses of the body to undergo further tumor suppression. Moreover, in comparison to the saline group, mice treated with free Cel exhibited histopathological alterations in liver tissue, marked by the infiltration of inflammatory cells. However, the mice treated with HA@Cel/NPs+anti-PD-L1, HA@Cel/NPs and Cel/NPs exhibited no noticeable pathological changes in main organs in comparison with the saline group, indicating good biocompatibility (Figure S8).

Subsequently, we evaluated the alteration of the immunosuppressive tumor microenvironment in mice with tumors by examining the shifts in immune cells infiltrating the tumor following a 14-day treatment. As shown in Figure S9A–C, the proportion of CD8+, CD4+, and Tregs in spleen tissues was examined across various treatment groups. The proportions in the saline group were 3.04%, 3.32%, and 18.7%, respectively. In contrast, the free Cel group displayed 6.23% CD8+, 10.9% CD4+, and 10.5% Tregs. The Cel/NPs group showed increases to 8.39% CD8+, 15.2% CD4+, and decrease to 6.94% Tregs. Notably, the HA@Cel/NPs group demonstrated a marked elevation with 14.2% CD8+, 23.6% CD4+, and a reduction in Tregs to 4.8%. Furthermore, the combination of HA@Cel/NPs + anti-PD-L1 yielded the most significant immunomodulation, with 22.6% CD8+, 31.4% CD4+, and a drastic decrease in Tregs to 1.70%. These results indicate that compared to the free Cel and Cel/NPs groups, the HA@Cel/NPs and HA@Cel/NPs + anti-PD-L1 groups significantly enhanced CD8+ and CD4+ expression in spleen tissues while concurrently reduced Treg proportions, with the latter group exhibiting the most pronounced effects. Analysis of tumor tissues revealed similar trends (Figure S9D–F). In conclusion, the HA@Cel/NPs+anti-PD-L1 group was able to significantly improve the immunosuppressive state of the tumor microenvironment and exert an enhanced anti-tumor immune response.

Characterization of HA@Cel/NPs

The synthetic hydrogen spectra of PAPE-SA-CD and HA-AD are presented in Figures S2 and S3, respectively. The HA@Cel/NPs were prepared according to a nanoprecipitation method.^[38] The experimental results show that the average particle size of HA@Cel/NPs is about 76.87 ± 2.65 nm, and the PDI is about 0.166 ± 0.012 (Figure 1A). TEM revealed that HA@Cel/NPs exhibited a spherical morphology with smooth surfaces (Figure S4). In addition, compared with Cel/NPs, HA@Cel/NPs exhibited a slightly higher EE (90.85 ± 3.69% vs 88.37 ± 4.36%) (Table S2), which could be attributed to the formation of the HA shell that minimized drug leakage during purification. Successful HA coating was confirmed by the increased particle size and the zeta potential shift from −15.5 mV to −24.7 mV (Figure 1B). XRD analysis was performed to study the effect of drug-carrier interaction on drug release characteristics. There are multiple peaks between 5° and 90° in free Cel, indicating that its crystallinity is high. But these spikes are not observed in HA@Cel/NPs, indicating that there is no crystalline complex between Cel and the matrix (Figure 1C). Therefore, Cel exists in the form of amorphous or disordered crystals.

The Stability of HA@Cel/NPs

The stability evaluation results after 14 days of storage at 4°C are shown in Figure 1D and E. It could be observed that the particle size, PDI and zeta potential of HA@Cel/NPs did not change much. This suggests that HA@Cel/NPs can retain stable properties for a period of 14 days.

ROS and Enzyme Sensitivity Evaluation of HA@Cel/NPs

The sensitivity of HA@Cel/NPs to ROS or α-amylase was verified by using DLS and TEM. As shown in Figure 1F, the size distribution and morphology of HA@Cel/NPs had little change in the PBS solution. However, an increase in size and the appearance of heterogeneous peaks were observed when HA@Cel/NPs were incubated with 1 mM H2O2 or 10 IU/mL α-amylase for 24 h. Further, the results of TEM showed that the presence of H2O2 or α-amylase led to the cracking of HA@Cel/NPs (Figure 1F), which may be due to ROS sensitivity caused by H2O2 destruction or enzyme sensitivity under α-amylase. Consequently, it is anticipated that HA@Cel/NPs exhibit monosensitivity to ROS or α-amylase.

Release Profiles of HA@Cel/NPs

The release of contents from free Cel or HA@Cel/NPs was assessed throughout the entire gastrointestinal tract (continuously placed in an alternative medium). Figure 1G showed that free Cel was released rapidly in SGF and SIF with a cumulative release rate of 57.73% within 6 h, so that it was heavily absorbed in the stomach and small intestine and less released in the colon tissue. In contrast, HA@Cel/NPs exhibited slower release properties, allowing the drug to reach its target location more completely. Moreover, the cumulative release rate was 39.95% after 48 h. As a result of the protection provided by PAPE-SA-CD and HA-AD, HA@Cel/NPs can effectively transit from the stomach and small intestine to reach the colon tissue.

In addition, the synergistic action of ROS and enzyme stimulators were used to simulate drug release behavior at the colon site. The release medium included pH 7.4 buffer, pH 7.4 buffer containing 1 mM H2O2, pH 7.4 buffer containing 10 IU/mL α-amylase, pH 7.4 buffer containing 1 mM H2O2 and 10 IU/mL α-amylase (Figure 1H). Compared to pH 7.4 buffer, the cumulative release rate was significantly accelerated in the presence of H2O2 or α-amylase, which reached 55.89% and 68.77% within 6 h, respectively. Additionally, the combined action of H2O2 and α-amylase resulted in a cumulative release rate of 71.87% for HA@Cel/NPs within 8 h, representing the highest level. This suggests that drug release of HA@Cel/NPs can be controlled by internal stimulation of ROS/enzymes.

In vitro Anti-Inflammatory Activity, Macrophage Polarization and Cellular Uptake Profiles

Anti-Inflammatory Activity of HA@Cel/NPs

The pharmacological study of HA@Cel/NPs in vitro was carried out using LPS-induced Raw264.7 macrophages, as depicted in Figure 2. As shown in Figure 2A and B, the secretion of pro-inflammatory cytokines TNF-α and IL-1β significantly increased in LPS-treated (positive control) cells compared to the negative control cells (P3mm was close to 0, indicating that the HA@Cel/NPs group had the inhibitory effect on the inflammatory-cancerous transformation of mice colon, and could inhibit the occurrence of inflammatory-cancerous transformation of colon.

Furthermore, the effectiveness of HA@Cel/NPs was assessed using H&E staining and Ki67 staining techniques. As shown in Figure 6H, we can observe that the colonic mucosa of mice in the control group was structurally intact, with well-arranged glands, crypts and goblet cells, and the submucosa was not congested and edematous, with no lesions or inflammation. Compared with the control group, the model group showed a partially defective mucosal epithelium, infiltration of a large number of tumor cells, mucosal edema, extensive necrosis of crypts. The HA@Cel/NPs group had the most obvious protective effect on the colonic epithelial cells, with almost no accumulation of tumor cells in the colonic mucosa, no obvious abnormality in the epithelial mucosa and glandular structure, a more intact crypt lumen, obvious goblet cells, a reduction in the degree of inflammation and malignancy of tumors, and a morphology similar to that of the control group. The above results showed that HA@Cel/NPs had the inhibitory effects on colitis-cancer transformation. We conducted additional assessments of the proliferation patterns in relation to CAC using Ki67 staining. The results illustrated in Figure 6H indicate that the group of mice treated with HA@Cel/NPs exhibited a lower number of Ki67-positive cells in comparison to the other groups. Additionally, the H&E results for the main organs indicated the absence of any notable pathological irregularities or damage (Figure S7), suggesting that HA@Cel/NPs exhibit excellent biocompatibility and minimal toxicity on CAC model mice.

Combination Therapy of HA@Cel/NPs with Anti-PD-L1 in the CT26 Mouse Model of Colorectal Cancer

Utilizing immune checkpoint blockades (anti-PD-L1) may enhance the ability of cytotoxic T lymphocytes to eliminate tumors, offering a potential treatment for various types of cancer. Recent research has shown that when combined with other therapies, PD-1/PD-L1 blockades can significantly boost their immunotherapeutic effects.^[42] Therefore, it is anticipated that combining anti-PD-L1 with HA@Cel/NPs could lead to positive outcomes in anti-tumor treatment. In order to verify this, we assessed the effectiveness of HA@Cel/NPs + anti-PD-L1 on a CT26 tumor model, and the experimental procedure was illustrated in Figure 7A. As depicted in Figure 7B, in the treatment group receiving free Cel, a slight decrease in mouse weight was observed within 14 days, whereas the weights of mice in the other groups remained stable. The findings indicate that the provided NPs exhibits specific biocompatibility and does not induce weight reduction in mice. As shown in Figure 7C, the tumor volume in the saline group exhibited a rapid and sustained increase, reaching 1478 mm3 by the end of the experiment. In contrast, all treatment groups inhibited tumor growth to varying degrees. The mean tumor volumes at the endpoint were 1152 mm3 for the free Cel group, 998 mm3 for the Cel/NPs group, 814 mm3 for the HA@Cel/NPs group, and 412 mm3 for the HA@Cel/NPs + anti-PD-L1 combination group. Notably, the combination therapy with HA@Cel/NPs and anti-PD-L1 exhibited the most potent antitumor efficacy. On day 14, the mice were euthanized and removed tumors (Figure 7D). As shown in Figure 7E, the trend of tumor weight was in accordance with the outcomes of tumors volume.

Meanwhile, H&E staining of tumor tissues was used to further explore the antitumor efficacy in vivo. As displayed in Figure 7F, tumor cells were densely accumulated in the saline group, while in other treatment groups, tumor cells displayed looser, destroyed structures and necrosis could be observed, especially in HA@Cel/NPs and HA@Cel/NPs + anti-PD-L1. Besides, as key factors to mediate specific antitumor immunity, HMGB1 and CRT were measured by immunofluorescence staining (Figure 7G).^[28] Compared with saline group, Cel/NPs, HA@Cel/NPs, and HA@Cel/NPs+anti-PD-L1 treatment significantly increased the HMGB1 and CRT levels in CT26 tumors, indicating that they triggered the specific anti-tumor immune responses of the body. Besides, HA@Cel/NPs+anti-PD-L1 achieved the best results, which could motivate Cel to inhibit tumors effectively and stimulate the anti-tumor immune responses of the body to undergo further tumor suppression. Moreover, in comparison to the saline group, mice treated with free Cel exhibited histopathological alterations in liver tissue, marked by the infiltration of inflammatory cells. However, the mice treated with HA@Cel/NPs+anti-PD-L1, HA@Cel/NPs and Cel/NPs exhibited no noticeable pathological changes in main organs in comparison with the saline group, indicating good biocompatibility (Figure S8).

Subsequently, we evaluated the alteration of the immunosuppressive tumor microenvironment in mice with tumors by examining the shifts in immune cells infiltrating the tumor following a 14-day treatment. As shown in Figure S9A–C, the proportion of CD8+, CD4+, and Tregs in spleen tissues was examined across various treatment groups. The proportions in the saline group were 3.04%, 3.32%, and 18.7%, respectively. In contrast, the free Cel group displayed 6.23% CD8+, 10.9% CD4+, and 10.5% Tregs. The Cel/NPs group showed increases to 8.39% CD8+, 15.2% CD4+, and decrease to 6.94% Tregs. Notably, the HA@Cel/NPs group demonstrated a marked elevation with 14.2% CD8+, 23.6% CD4+, and a reduction in Tregs to 4.8%. Furthermore, the combination of HA@Cel/NPs + anti-PD-L1 yielded the most significant immunomodulation, with 22.6% CD8+, 31.4% CD4+, and a drastic decrease in Tregs to 1.70%. These results indicate that compared to the free Cel and Cel/NPs groups, the HA@Cel/NPs and HA@Cel/NPs + anti-PD-L1 groups significantly enhanced CD8+ and CD4+ expression in spleen tissues while concurrently reduced Treg proportions, with the latter group exhibiting the most pronounced effects. Analysis of tumor tissues revealed similar trends (Figure S9D–F). In conclusion, the HA@Cel/NPs+anti-PD-L1 group was able to significantly improve the immunosuppressive state of the tumor microenvironment and exert an enhanced anti-tumor immune response.

Discussion

In this study, we developed a HA-functionalized multifunctional nanoplatform (HA@Cel/NPs) featuring enzyme/ROS dual-responsive release and CD44-mediated active targeting for the site-specific treatment of UC, CAC and colon cancer. Our results demonstrated that HA@Cel/NPs achieved efficient lesion-specific accumulation in inflamed and malignant colonic tissues, as confirmed by both in vitro cellular uptake assays and in vivo fluorescence imaging. Upon reaching the colonic pathological microenvironment characterized by elevated ROS and enzyme levels, the NPs underwent rapid degradation, triggering on-demand release of the encapsulated Cel. Orally administered HA@Cel/NPs not only effectively alleviated acute intestinal inflammation in the UC model by suppressing pro-inflammatory cytokine production and promoting M1-to-M2 macrophage polarization, but also significantly retarded tumor progression in the CAC model. In the colon cancer setting, intravenous administration of HA@Cel/NPs in combination with intraperitoneal anti-PD-L1 checkpoint blockade significantly potentiated systemic antitumor immunity, as evidenced by enhanced CD8⁺ and CD4⁺ T cell infiltration, leading to effective elimination of established tumors compared with HA@Cel/NPs monotherapy. Collectively, these findings establish HA@Cel/NPs as a versatile, dual-route nanotherapeutic platform that bridges inflammation management, cancer suppression, and immunotherapy—a capability that distinguishes our system from conventional single-mechanism or single-disease nanocarriers.^[43]

In vitro and in vivo studies confirmed that HA@Cel/NPs achieved efficient lesion-specific accumulation in colonic tissues, as demonstrated by enhanced cellular uptake in CD44-positive Raw264.7 and CT26 cells, as well as prominent fluorescence signal retention at diseased sites in AOM/DSS-induced CAC model mice. Upon exposure to the elevated ROS and enzyme levels in the pathological colonic microenvironment, the NPs rapidly degraded, releasing the encapsulated Cel in an on-demand manner. This dual-responsive design ensures that the therapeutic payload is liberated precisely at the target sites, thereby maximizing local drug concentration while minimizing systemic exposure and off-target toxicity. Over the course of treatment, no significant adverse effects on body weight, or major organ histology were observed across all disease models, confirming the excellent biosafety profile of HA@Cel/NPs. The biocompatible nature of HA, combined with the biodegradable dual-responsive linkers, ensures that the nanocarrier is eventually cleared from the body without eliciting chronic toxicity.^[44],^[45]

In vitro studies confirmed that HA@Cel/NPs effectively suppressed the production of pro-inflammatory cytokines (TNF-α and IL-1β) and promoted M1-to-M2 macrophage polarization, indicating their capacity to rewire the inflammatory milieu at the cellular level. In DSS-induced acute colitis models, oral administration of HA@Cel/NPs significantly ameliorated disease severity, as evidenced by attenuated weight loss, reduced disease activity index, restored colon length, and alleviated histological damage. These findings collectively suggest that HA@Cel/NPs reprogram the intestinal inflammatory microenvironment—a feature that fundamentally distinguishes our system from conventional anti-inflammatory agents that merely alleviate symptoms without rectifying the underlying immunological imbalance. In the AOM/DSS-induced CAC model, HA@Cel/NPs dramatically retarded tumorigenesis, as reflected by decreased tumor multiplicity and burden, concomitant with reduced expression of the proliferation marker Ki-67 in colonic tissues. This full-spectrum therapeutic efficacy—spanning from acute inflammation to malignancy—underscores the translational potential of our platform in managing the entire inflammation–carcinoma continuum, a capability rarely achieved by existing nanomedicines that are typically confined to a single pathological stage. Furthermore, in a subcutaneous CT26 colon cancer model, intravenous administration of HA@Cel/NPs in combination with intraperitoneal anti-PD-L1 checkpoint blockade elicited significantly enhanced systemic antitumor immunity compared with HA@Cel/NPs monotherapy, as evidenced by increased infiltration of CD8⁺ and CD4⁺ T lymphocytes and upregulation of immunogenic cell death markers HMGB1 and CRT. It should be noted that, due to the absence of an anti-PD-L1 monotherapy group, the current data support a combination benefit rather than a formal synergistic interaction between HA@Cel/NPs and anti-PD-L1. Nevertheless, the significantly improved antitumor efficacy and enhanced T cell infiltration in the combination group compared with HA@Cel/NPs alone strongly suggest that the addition of anti-PD-L1 potentiates the therapeutic effect of our nanocarrier. This immunochemotherapeutic combination strategy not only expands the therapeutic repertoire of Cel beyond its conventional anti-inflammatory and cytotoxic roles but also positions HA@Cel/NPs as a versatile bridge linking conventional chemotherapy and immunotherapy, offering a promising paradigm for comprehensive management of intestinal malignancies.

While the present results demonstrate compelling preclinical promise, several methodological and translational constraints require transparent acknowledgment. First, although HA@Cel/NPs exhibited consistent therapeutic efficacy across three mechanistically diverse disease models, the molecular determinants governing their dual-stimuli–responsive release—particularly the spatiotemporal coordination between nanoparticle disassembly, payload liberation, and dynamic shifts in colonic pH, redox potential, and proteolytic activity—remain insufficiently characterized and demand rigorous, hypothesis-driven mechanistic validation. Second, our safety assessment was confined to acute endpoints; definitive evaluation of clinical translatability necessitates longitudinal toxicology studies—including repeated-dose regimens over clinically relevant durations—to assess cumulative organ toxicity, immunogenicity, and microbiome perturbations. Third, the observed synergy between HA@Cel/NPs and anti-PD-L1 blockade in subcutaneous tumors has not been corroborated in orthotopic colon cancer models or immunocompetent patient-derived xenograft models, which preserve tumor–stroma interactions, immune cell infiltration, and anatomical microarchitecture essential for predicting clinical immunomodulatory outcomes. Future efforts targeting mechanistic dissection, extended toxicological characterization, and validation in clinically relevant tumor models will be critical to bridging the gap between preclinical innovation and clinical application of HA@Cel/NPs.

Conclusion

In this study, we developed an oral Cel delivery nanoplatform based on the core of the obtained PAPE-SA-CD combined with the shell of HA-AD. The resulting nano-system (HA@Cel/NPs) maintained stability throughout passage through the gastrointestinal tract, exhibited excellent colon-targeting ability attributed to the properties of HA, and demonstrated precise burst release behavior of Cel due to the dual enzyme/ROS sensitivity of PAPE-SA-CD. In vitro experiments indicated that HA@Cel/NPs exhibited potent abilities to anti-inflammatory and induce apoptosis. Additionally, HA-AD/PAPE-SA-CD showed minimal cytotoxicity in vitro and indicated favorable biosafety in vivo, thus confirming the biocompatibility of the nano-system for delivery. Finally, in vivo experiments have shown that HA@Cel/NPs significantly inhibited the progression of UC, CAC, and CRC. Moreover, when combined with an immune checkpoint inhibitor (anti-PD-L1), HA@Cel/NPs demonstrated the capacity to eliminate CRC and enhance overall antitumor immunity. In conclusion, it seems that this innovative nanoplatform with versatile pharmacological properties could be utilized for effective treatment of colon-related ailments.