Gut microenvironment-responsive copper selenide/disulfiram/eudragit^® L100-55 nanoflowers alleviate chronic pancreatitis via the inhibition of pyroptosis and pancreatic stellate cell activation

Abstract

Pyroptosis is involved in various inflammatory and fibrotic processes, but its specific role in the progression of chronic pancreatitis (CP) remains insufficiently explored. This study first investigated the dynamic expression changes of pyroptosis-related proteins during CP development and verified the regulatory effects of disulfiram (DSF) on key effector cells in CP, namely pancreatic stellate cells (PSCs) and macrophages. Building on these findings, a gut microenvironment-responsive nanoplatform, CuSe/DSF/EL nanoflowers, was designed and synthesized by integrating CuSe (as a carrier and selenium donor), DSF (as the core therapeutic agent), and Eudragit^® L100-55 (EL, a pH-sensitive polymer for intestinal targeting). Results demonstrated that the CuSe/DSF/EL nanoflowers exhibited excellent biocompatibility, robust antioxidant capacity, and efficient electrostatic targeting to inflamed pancreatic tissues via charge interactions. In vivo experiments using a mouse CP model confirmed that the nanoflowers significantly alleviated pancreatic inflammation, fibrosis, and pyroptosis, while effectively repairing pancreatic structure and restoring its function. Further RNA sequencing (RNA-seq) analysis revealed that the nanoflowers reversed CP-associated transcriptomic dysregulation, with a notable focus on correcting impaired pancreatic secretion pathways. This study clarifies the critical role of pyroptosis in CP pathogenesis and provides a novel, multi-mechanistic nanotherapeutic strategy for CP treatment by enhancing the bioavailability and targeting efficiency of DSF.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04669-8.

Introduction

Chronic pancreatitis (CP) is a progressive inflammatory disease characterized by acinar cell damage, macrophage infiltration, and pancreatic fibrosis. Its incidence has been increasing annually, yet no effective therapeutic approach is currently available [1–4]. Pancreatic stellate cells (PSCs) are recognized as the principal effector cells in pancreatic fibrosis, secreting excessive extracellular matrix (ECM) upon activation [5, 6]. During the repair and regeneration phase, M2-type macrophages predominate and release profibrotic factors such as transforming growth factor-β (TGF-β) and platelet-derived growth factor-BB, which activate PSCs and stimulate the secretion of matrix metalloproteinases. This process facilitates the degradation and remodeling of ECM components, including fibronectin (FN) [7–9]. Therefore, a complex crosstalk operates among immune cells, including PSCs, macrophages, and T lymphocytes, during the progression of CP.

Pyroptosis has been implicated in a variety of inflammatory diseases, including acute pancreatitis (AP) [10–16]. Specifically, activation of the non-canonical inflammasome pathway mediated by caspase-4/5/11 promotes the development of AP, while activation of NLRP3 inflammasome and Gasdermin D (GSDMD) collectively contribute to systemic inflammatory responses. Consistently, intervention via small interfering RNA (siRNA) or pharmacological agents can downregulate pyroptosis-related proteins and thereby alleviate AP severity [17–19]. Beyond acute inflammation, growing evidence supports a role for pyroptosis in the progression of fibrotic processes. For example, inflammasomes such as NLRP3 and absent in melanoma 2 (AIM2) have been reported to promote chronic inflammation and liver fibrosis through pyroptosis-related signaling pathways [20–22]. In renal pathology, Gasdermin E (GSDME)-mediated pyroptosis of renal parenchymal cells exacerbates tubular inflammation and fibrosis in obstructive kidney disease via the tumor necrosis factor-α (TNF-α)/caspase-3/GSDME axis [23]. Similarly, studies have demonstrated that NLRP3 or AIM2 inflammasomes drive macrophage pyroptosis and facilitate pulmonary fibrosis [24–26]. Collectively, these findings confirm that pyroptosis is closely associated with the progression of both inflammatory and fibrotic disorders. However, the specific roles of pyroptosis and Gasdermin family members in CP remain largely unexplored.

Disulfiram (DSF) and its bioactive metabolites exert well-documented anti-inflammatory effects by covalently modifying GSDMD, suppressing NLRP3 inflammasome activation, and inhibiting monocyte/macrophage chemotaxis and activation [27–29]. For instance, DSF ameliorates ulcerative colitis (UC) by inhibiting oxidative stress-induced pyroptosis of colonic cell via the glycogen synthase kinase 3β (GSK3β)/nuclear factor-erythroid 2 related factor 2 (Nrf2)/NLRP3 pathway [30]. Meanwhile, DSF exhibits anti-fibrotic activity by inhibiting the TGF-β/SMAD pathway, and its derivatives have been shown to reduce monocyte/macrophage infiltration, ultimately alleviating fibrosis in experimental glomerulonephritis [29, 31]. Based on these findings, we hypothesize that DSF may mitigate CP progression by targeting pancreatic inflammation, pyroptosis, and fibrosis. Despite these promising biological activities, the clinical translation of DSF is hindered by its unfavorable pharmaceutical properties: poor aqueous solubility, a short blood half-life, rapid degradation, and low oral bioavailability [32–36]. Furthermore, its instability during systemic absorption and distribution leads to unpredictable pharmacokinetics and suboptimal therapeutic efficacy [37]. Additionally, the co-delivery of DSF with copper ions (Cu²⁺) remains a major challenge, as its primary metabolite, copper diethyldithiocarbamate (CuET), is highly hydrophobic [38, 39]. Collectively, these limitations highlight the urgent need to explore and develop efficient delivery strategies for DSF-based therapy.

Advances in biomaterials, such as microspheres, polymers, and liposomes, have enhanced drug bioavailablity and targeting capability, opening up new avenes for DSF delivery [40–42]. For example, lactoferrin nanoparticles loaded with DSF effectively inhibit pyroptosis and inflammatory cytokine release in macrophages, ameliorating conditions such as sepsis and UC [40]. In another study, a dual-drug delivery system based on lipid nanoparticles co-loaded with MCC 950 and DSF significantly improves survival in LPS-induced septic peritonitis and reduces active caspase-1 expression [41]. As an essential cofactor for numerous enzymes, Cu²⁺ participates in various physiological processes and exhibits antioxidant and anti-inflammatory properties [43, 44]. The therapeutic effect of DSF is potentiated in the presence of Cu²⁺, through the formation of highly active CuET [45, 46]. Our previous study demonstrated that co-delivery of DSF and Cu²⁺ more effectively reduced colonic damage and inflammatory cytokine levels in inflammatory bowel disease (IBD) [47]. Selenium, another essential trace element, scavenges reactive oxygen species (ROS), modulates macrophage polarization, and mitigates oxidative stress-induced inflammation [48, 49]. Given these attributes, selenium has been widely applicated in bioactive material synthesis. For example, selenium nanosheets with high biocompatibility and anti-inflammatory effects were reported to alleviate tissue damage and reduce inflammatory cytokines via inhibiting apoptosis and cytokine-cytokine receptor interactions in UC [50]. These significant advances in biomaterials and delivery strategies have inspired our investigation into an intestine-responsive co-delivery system with multifaceted biological effects.

In this study, we first found an increase in pyroptosis during CP and the inhibitory effects of DSF on pancreatic acinar cell pyroptosis and PSC activation. Capitalizing on the regulatory effects of DSF, as well as the modulatory role of selenium in macrophage function, we successfully fabricated CuSe/DSF/Eudragit^® L100-55 (EL) nanoflowers. EL, a pH-sensitive copolymer, was integrated into the nanoflower system due to its unique property: it remains insoluble in acidic gastric environments while dissolving at pH ≥ 5.5 [51]. This characteristic of EL ensures the CuSe/DSF/EL nanoflowers evade gastric degradation and achieve site-specific release in the intestine, thereby improving the bioavailability of DSF and the targeting efficiency of the nanosystem (Scheme 1). The CP alleviation of these CuSe/DSF/EL nanoflowers involves the suppression of pyroptosis, a pivotal mediator of inflammation, and the inhibition of PSC activation, the primary driver of fibrosis, with the additional benefit of excellent antioxidant capacity to mitigate ROS-induced pancreatic damage. Collectively, this work presents a novel nanomaterial-based therapeutic approach for CP and provides robust experimental and theoretical support for the clinical translation of DSF in treating fibrotic diseases, particularly pancreatic fibrosis associated with CP.

Scheme 1.

Schematic diagram of the synthesis and application of the CuSe/DSF/EL nanoflowers. a) The protective effect of GSDMD knockout on CP, and the interactions among acinar cells, macrophages, and PSCs. b) The synthesis process of the CuSe/DSF/EL nanoflowers. c) The therapeutic effect of the CuSe/DSF/EL nanoflowers in the progression of CP

Materials and methods

Materials

Caerulein was purchased from APExBIO Technology LLC (USA). 4% paraformaldehyde was obtained from Beijing Labgic Technology Co., Ltd. (China). Dulbecco’s Modified Eagle’s Medium (DMEM), fetal bovine serum (FBS), penicillin/streptomycin (10,000 U/mL +10,000 µg/mL), and trypsin were purchased from Gibco Co., Ltd. (USA). De-plasma was obtained from TOKU-E company (Japan). LPS, Trizol, bromochloropropane, and polyvinylidene fluoride (PVDF) membrane were purchased from Sigma-Aldrich, Inc (USA). Interleukin-4 (IL-4) and interleukin-13 (IL-13) were obtained from PeproTech, Inc (USA). TGF-β was purchased from R&D Systems, Inc (USA). Phosphate buffer saline (PBS) was procured from Corning Co., Ltd. (USA). Cell counting kit-8 (CCK-8) kit, Live/Dead kit, and protease inhibitor were obtained from Thermo Fisher Scientific, Inc (USA). Isopropanol and ethanol were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Reverse transcription reagents and Hieff UNICON^® QPCR Blue SYBR Green Master Mix were purchased from Yeasen Biotechnology Co., Ltd. (Shanghai, China). Lactate dehydrogenase (LDH) assay kit was obtained from JianCheng Bioengineering Institute (Nanjing, China). Bicinchoninic acid (BCA) protein assay kit was bought from Beyotime Biotech, Inc (Shanghai, China). Radio immunoprecipitation assay (RIPA) lysate, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) protein loading buffer (5×, reducing), Tris-Glycine-SDS Running Buffer (10×), Transfer Buffer (10×), Tris Buffered Saline with Tween (10×), and Universal Antibody Diluent Buffer were purchased from Epizyme Biomedical Technology Co., Ltd. (Shanghai, China). Phosphatase inhibitor was obtained from Roche Pharma Ltd. (Switzerland). PAGE gels (4–20%) were purchased from GenScript Biotechnology Co., Ltd. (Nanjing, China). Color-Mixed Protein Marker 180 (10–180 kDa) was obtained from ABclonal Technology Co., Ltd. (Wuhan, China). Nonfat dry milk was bought from Cell Signaling Technology, Inc (USA). Mouse TNF-α enzyme-linked immunosorbent assay (ELISA) kit and Mouse/Rat insulin-like growth factor 1 (IGF-1) ELISA kit were purchased from Absin Bioscience, Inc and Multi Sciences (Lianke) Biotech Co., Ltd., respectively. Fecal elastase levels were measured by mouse FE-1 ELISA kit (ZCIBIO Technology Co.,Ltd). Copper chloride dihydrate (CuCl₂·2H₂O), trisodium citrate, sodium hydroxide (NaOH), ascorbic acid (AA), selenium powder, sodium borohydride (NaBH₄), 2, 2’-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid (ABTS·), 2,2-Diphenyl-1-picrylhydrazyl (DPPH·), 3,3’,5,5’-tetramethylbenzidine (TMB), and potassium persulfate (K₂S₂O₈) were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Hydrogen peroxide (H₂O₂) and ferric chloride (FeCl₃) were bought from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). EL (95%) was obtained from Shanghai Changwei Pharmaceutical Accessories Technology Co., Ltd. DSF and Cy7 were purchased from MedChemExPress. All chemicals were analytical-grade and do not require further purification before use. Kunming mice, C57BL/6 mice, and GSDMD-KO mice (C57BL/6JGpt-Gsdmd^em16Cd723/Gpt, GSDMD^−/−, Strain NO. T010437) were purchased from GemPharmatech Co., Ltd (Nanjing, China). Mouse pancreatic acinar carcinoma cells (266-6 cells) were obtained from QuiCell Biotechnology Co., Ltd. (Shanghai, China). Detailed information for primer sequences and antibodies used in this study were listed in Table S1 and Table S2.

Establishment of CP model in mice

Animal studies were approved by the Animal Ethics Committee of Ganzhou People’s Hospital (No. PJD2025-007-01). C57BL/6 mice were randomly divided into Control group (n = 12) and CP group (n = 12). The mouse model of CP was established by repeated intraperitoneal injections of caerulein solution (50 µg/kg in PBS): 6 injections per day (1-h interval between doses), 3 days a week (e.g., Monday, Wednesday, Friday), for a total of 4 weeks. Mice in the Control group were given an equal volume of normal saline. During the four-week modeling period, three mice were sacrificed in each of the two groups each week, and the pancreatic tissue samples were collected for subsequent immunofluorescence (IF) staining of pyroptosis-related proteins (NLRP3, GSDMD, and caspase-1). To validate the role of pyroptosis in CP, C57BL/6 wild-type (WT) mice and GSDMD-KO mice (C57BL/6JGpt-Gsdmd^em16Cd723/Gpt, GSDMD^⁻/⁻) were randomly divided into four groups: WT, KO, WT-CP, and KO-CP (n = 6 per group). Mice in the WT-CP and KO-CP groups were subjected to CP induction via repeated caerulein injections, while those in the WT and KO groups received equal volumes of normal saline. After modeling, mice were anesthetized with 10% chloral hydrate and euthanized by spinal cord dislocation method. The pancreatic tissues were collected for further Hematoxylin and Eosin (H&E), IF, and immunohistochemistry (IHC) staining to evaluate pancreatic pathological changes, functional indicators, and molecular markers.

Synthesis of CuSe nanoflowers

For the synthesis of CuSe nanoflowers, 20 mL of 1.25 M NaOH was added to 80 mL of an aqueous mixture containing 1.5 mM CuCl₂·2H₂O and 0.9 mM trisodium citrate, followed by thorough stirring at room temperature. Next, 50 mL of 0.25 M AA was added to the mixture, stirred evenly and aged for 2 h to obtain the pre-aged solution. Separately, 0.16 g of selenium powder, 0.4 g of NaOH, and 0.4 g of NaBH₄ were first dissolved in 10 mL of deionized water, and the resulting solution was then added to 200 mL of pre-heated water (90 °C) and stirred evenly. The pre-aged solution was added to the above heated aqueous solution over 5 min, and the reaction mixture was stirred overnight at room temperature. Finally, CuSe nanoflowers were collected by high-speed centrifugation (20,000 rpm, 20 min), and washed three times with deionized water.

Synthesis of CuSe/DSF/EL nanoflowers

To synthesize CuSe/DSF/EL nanoflowers, a 100 mg of the as-synthesized CuSe nanoflowers were dispersed in 20 mL of ethanol, and 2 mL of DSF ethanol solution (5 mg/mL) was added, followed by stirring at 37℃ for 12 h. Centrifugation (20,000 rpm, 10 min) was performed to obtain the CuSe/DSF nanoflowers. Then, 50 mg CuSe/DSF nanoflowers were weighed and dissolved in 10 mL ethanol, and 1 mL EL ethanol solution (10 mg/mL) was added. Subsequently, the solvent was evaporated using a rotary evaporator at 40℃ to obtain the CuSe/DSF/EL nanoflowers. The obtained CuSe/DSF/EL nanoflowers were dissolved in PBS for further applications.

Characterization of CuSe/DSF/EL nanoflowers

The morphology of CuSe/DSF/EL nanoflowers was observed using transmission electron microscopy (TEM [FEI Tecnai F20]) and scanning electron microscopy (SEM [Tescan mira4]). The elements in the nanoflowers (C, N, O, Cu, and Se) were analyzed by the X-ray energy spectrometer attached to TEM. The Malvern laser particle size analyzer (NanoBrook 90Plus) was used to analyze the size and zeta potentials of nanoparticles. The successful loading of DSF and EL was analyzed via Fourier transform infrared spectroscopy (FTIR [Nicolet 7000-C]). The loading contents of DSF and EL in CuSe/DSF/EL nanoflowers were determined using a thermogravimetric (TG) analyzer (NetzchSTA449C), with the testing temperature ranging from 30℃ to 800℃. The ion valence states of copper and selenium elements in CuSe/DSF/EL nanoflowers were analyzed via X-ray photoelectron spectroscopy (XPS [Thermo KalPha]). The crystallinity of CuSe/DSF/EL nanoflowers was analyzed via X-ray diffraction (XRD [X’Pert PRO MPD]). Ultraviolet-Visible-Near Infrared (UV-vis-NIR) spectrometer (Shimadzu, Japan) was used to record the spectral changes of samples and analyze the loading of DSF and EL.

Evaluation of the total antioxidant capacity of CuSe/DSF/EL nanoflowers

ABTS· was used to evaluate the total antioxidant capacity of CuSe/DSF/EL nanoflowers. In brief, 6 mg of ABTS⁺ powder was dissolved in 1.47 mL of deionized water and 2 mg of K₂S₂O₈ was dissolved in 2.86 mL of deionized water. Then, 0.8 mL of ABTS⁺ solution was mixed with 0.8 mL of K₂S₂O₈ solution, reacted in the dark for 12 h. After the reaction, the mixed solution was diluted with ethanol to 50 mL to obtain the ABTS⁺ solution, stored in the dark for later use. For the evaluation, 200 µL of CuSe/DSF/EL nanoflowers of different concentrations (0, 6, 12, 18, 24, and 30 µg/mL) were added to 1 mL of ABTS⁺ solution and incubated at 37℃ for 30 min. Subsequently, the mixed solution was centrifuged (20,000 rpm, 20 min) and the absorbance changes of ABTS⁺ solution were recorded using UV-vis-NIR spectrometer and each experiment was repeated three times. Finally, the clearance ratio of ABTS⁺ was calculated based on the absorbance value of the solution at 734 nm, and the color changes of the solution was recorded by photos. The clearance ratio was calculated as Eq. 1:

1

A₀ referred to the absorbance of the working solution, and A referred to the absorbance of the solution after incubation with different concentrations of CuSe/DSF/EL nanoflowers.

Evaluation of the reactive nitrogen species (RNS) scavenging ability of CuSe/DSF/EL nanoflowers

DPPH· was used to investigate the scavenging ability of CuSe/DSF/EL nanoflowers on RNS. Specifically, 1 mg of DPPH· was dissolved in 50 mL of ethanol, stirred thoroughly and stored in the dark for later use. 200 µL of CuSe/DSF/EL nanoflowers of different concentrations (0, 6, 12, 18, 24, and 30 µg/mL) were added to 1 mL of DPPH· solution and incubated at 37℃ for 30 min. Subsequently, the mixed solution was centrifuged (20,000 rpm, 20 min) and the absorbance changes of DPPH· solution were recorded using UV-vis-NIR spectrometer, and each experiment was repeated three times. Finally, the clearance ratio of DPPH· was calculated based on the absorbance value of the solution at 519 nm, and the color changes of the solution were recorded by photos. The clearance ratio was calculated as Eq. 1:

1

A₀ referred to the absorbance of the working solution, and A referred to the absorbance of the solution after incubation with different concentrations of CuSe/DSF/EL nanoflowers.

Evaluation of the ROS scavenging ability of CuSe/DSF/EL nanoflowers

TMB was used as a chromogenic substrate to evaluate the ROS scavenging ability of CuSe/DSF/EL nanoflowers. Under the presence of H₂O₂ and Fe³⁺, TMB can be transferred to a blue oxidation product TMBox. Breifly, TMB was dissolved in ethanol to prepare a solution (10 mM) and stored in the dark. Then, 100 µL of TMB solution and 30 µL of H₂O₂ (10 mM) were added to 870 µL of PBS. For blank control, 10 µL of FeCl₃ solution (1 mM) was added. For experiment groups, 200 µL CuSe/DSF/EL nanoflowers of different concentrations (0, 6, 12, 18, 24, and 30 µg/mL) were added, followed by the addition of 10 µL of FeCl₃ solution (1 mM). The mixed solutions were incubated at 37 ℃ for 20 min. After centrifugation (20,000 rpm, 20 min), the absorbance changes of TMB solution were recorded using UV-vis-NIR spectrometer, and each experiment was repeated three times. Finally, the clearance ratio of ROS was calculated based on the absorbance value of the solution at 652 nm, and the color changes of the solution were recorded by photos. The clearance ratio was calculated as Eq. 1.

Blood safety evaluation of CuSe/DSF/EL nanoflowers

To study the blood safety, 1 mL of whole blood was taken from Kunming mice and centrifuged at 3,000 rpm for 5 min to obtain mouse red blood cells (RBCs). These RBCs were washed with PBS for three times and diluted to 50 mL with PBS. Then, 400 µL of RBCs and 100 µL of different concentrations (100, 150, 200, and 250 µg/mL) of CuSe/DSF/EL nanoflower solutions were added to 1 mL PBS, followed by incubation at 37℃ for 2 h. The mixed solution of 1.1 mL of PBS and 400 µL of RBCs was regarded as negative control, and the mixed solution of 1.1 mL of deionized water and 400 µL of RBCs was regarded as positive control. Subsequently, the mixed solutions were centrifuged at 3,000 rpm for 5 min, and the supernatant was collected. UV-vis-NIR spectrometer was used to record the absorbance value of the supernatant at 541 nm, and the hemolysis ratio was calculated. Each experiment was repeated three times. The hemolysis ratio was calculated as Eq. 2:

2

A_a referred to the absorbance of the solution after incubation with different concentrations of CuSe/DSF/EL nanoflowers, A_b referred to the absorbance of negative control, and A_c referred to the absorbance of positive control.

Cell safety evaluation of CuSe/DSF/EL nanoflowers

The safety of CuSe/DSF/EL nanoflowers at the cellular level was evaluated via CCK-8 assay. Specifically, 266-6 cells were seeded in a 96-well plate, added with fresh DMEM and incubated in an incubator with 5% CO₂ at 37℃ for 24 h, at a density of 8,000 cells per well. Then, CuSe/DSF/EL nanoflowers of different concentrations (100, 150, 200, and 250 µg/mL) were co-cultured with 266-6 cells for 24 h. Subsequently, the DMEM containing CuSe/DSF/EL nanoflowers was replaced with fresh DMEM containing CCK-8 working solution, incubated in an incubator for 2 h. Microplate reader was used to measure the absorbance of the samples at 450 nm and cell viability was calculated. Besides, 266-6 cells were stained with Calcein AM/PI cytotoxicity assay kit to further confirm the cell viability after incubation. The living cells were stained green with Calcein AM, while the dead cells were stained red with PI. Cell viability was observed using a fluorescence microscope (Olympus, Japan).

Histocompatibility evaluation of CuSe/DSF/EL nanoflowers

C57BL/6 mice were randomly divided into control group and experimental groups (1 d, 3d, 7 d, and 14 d) (n = 3). After 12 h of fasting, mice in the experimental groups were orally administered with 200 µL of CuSe/DSF/EL nanoflower solution (1,000 µg/mL), while mice in the control group were orally administered with 200 µL PBS. After anesthesia, blood samples were collected from the eyeballs of mice at different time points for routine and biochemical analysis. RBC, white blood cell (WBC), and platelet (PLT) were detected via a fully automated blood cell analyzer. For biochemical analysis, blood was left at room temperature for 30 min, and then centrifuged at 3,000 rpm for 10 min to obtain the serum. Serum biochemical markers including total bilirubin (TB), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (CREA), and urea (UA) were detected using a fully automated biochemical analyzer. Besides, important organs including heart, liver, spleen, lungs, and kidneys were collected and fixed in 4% paraformaldehyde for subsequent tissue staining.

Therapeutic efficacy of CuSe/DSF/EL nanoflowers

C57BL/6 mice were randomly divided into Control, CP, CP + DSF, CP + CuSe, and CP + CuSe/DSF/EL groups (n = 6). CP model was established by repeated intraperitoneal injections of caerulein for 4 weeks as previously described, and mice in different treatment groups (i.e., CP + DSF, CP + CuSe, and CP + CuSe/DSF/EL groups) were daily orally administered with DSF, CuSe, or CuSe/DSF/EL nanoflower solutions, respectively. Body weights of mice in different groups were monitored regularly during the modeling process. After the modeling process, all the mice were sacrificed and tissue samples were collected for subsequent tissue staining.

Fluorescence imaging

The CuSe/DSF/EL nanoflowers were labeled with Cy7 during the preparation process for fluorescence labeling. Specifically, 5 mg of Cy7 was dissolved in 1 mL of DMSO to prepare the working solution, and the Cy7 working solution was added to 10 mL of CuSe/DSF/EL nanoflower solution (1,000 µg/mL), followed by a series of procedures including stirring, washing, and redispersion. Mice in the Control and CP groups were orally administered with 200 µL of Cy7-labeled CuSe/DSF/EL nanoflowers (1,000 µg/mL), and their pancreas and intestines were harvested at 24 h after the gavage and placed in the bioluminescence imaging equipment (Tanon ABL X6 PRO) to observe the discrepancy in enrichment level of the Cy7-labeled nanoflowers between the Control and CP groups.

RNA sequencing (RNA-seq) analysis

In order to investigate the differences in gene expression of different groups, pancreatic tissue samples were collected from mice and analyzed via RNA-seq. RNA was extracted using CTAB method, and RNA quality was tested. Afterwards, the extracted mRNA was enriched using mRNA Capture Beads magnetic beads. After magnetic bead purification, mRNA was fragmented under high temperature. The fragmented mRNA was used as a template for synthesizing the first strand of cDNA in a reverse transcriptase-mixed system. Meanwhile, end repair was performed and A tail was added. Next, the connector was connected, and target fragments were selected using Hieff NGS^® DNA Selection Beads magnetic beads. Subsequently, polymerase chain reaction (PCR) library amplification was performed, followed by quality inspection of sequencing library and sequencing on the machine. Agarose gel electrophoresis was used to analyze the RNA integrity and rule out DNA contamination; NanoPhotometer spectrophotometer was applicated to detect RNA purity (OD_260/280 and OD_260/230); Qubit2.0 Fluorometer was used to accurately quantify RNA concentration; Agilent 2100 bioanalyzer was used to detect RNA integrity. After sequencing, the obtained genes were subjected to functional enrichment analysis.

16S rDNA sequencing analysis

Feces samples of mice from different groups were collected and analyzed by 16S rDNA sequencing. DNA from fecal samples was extracted using the CTAB, and nuclear-free water was regarded as blank. The DNA was eluted in 50 µL elution buffer. Afterwards, PCR amplification was performed in a 25 µL-reaction mixture containing template DNA (25 ng), PCR Premix (12.5 µL), primer (2.5 µL for each), and PCR-grade water. PCR experiment procedures consisted of an initial denaturation at 98 °C for 30 s, 32 cycles of denaturation (98 °C) for 10 s, annealing (54 °C) for 30 s, extension (72 °C) for 45 s, and a final extension (72 °C) for 10 min. The products were purified by AMPure XP beads (Beckman Coulter Genomics, USA) and then quantified using Qubit fluorometers (Invitrogen, USA), followed by library preparation and quality inspection. Finally, all samples were sequenced on an Illumina NovaSeq platform according to the manufacturer’s recommendations. Feature table and sequence were obtained after dereplication using DADA2. Alpha diversity and beta diversity were calculated with QIIME2 by normalizing to the identical sequences randomly. Relevant diagrams were implemented by the R package (v3.5.2).

Statistical analysis

At least three parallel repeated experiments were conducted to obtain data in each group (n ≥ 3). Graphpad Prism software (version 8.0.2) was used for statistical analysis and mapping. The differences between two groups were compared by Student’s t test. For differences among three or more groups, one-way analysis of variance (ANOVA) was employed. All the experimental results were presented as mean ± standard deviation (mean ± SD). A p value of less than 0.05 was considered statistically significant (e.g., for the comparisons between control and model groups, #p < 0.05, and ##p < 0.01; for the comparisons between model and treatment groups, *p < 0.05, and **p < 0.01).

Results and discussions

Expression changes of pyroptosis-related proteins during CP modeling process

Pyroptosis, a pro-inflammatory programmed cell death pathway, has been implicated in the pathogenesis of various inflammatory and fibrotic diseases, yet its dynamic variation and functional relevance in CP remain largely uncharacterized [52, 53]. To address this knowledge gap and lay a foundation for exploring targeted therapeutic strategies, this study first analyzed the expression patterns of key pyroptosis-related proteins in pancreatic tissues during the progression of CP, based on the established mouse CP model (Fig. 1). The results showed that during the four-week CP modeling process, the expressions of three key pyroptosis-related proteins, including NLRP3 (a core component of the inflammasome), GSDMD (a pyroptosis executor), and caspase-1 (a pro-pyroptotic protease), all exhibited time-dependent dynamic changes in pancreatic tissues, with distinct specificities in their expression profiles. Among these proteins, GSDMD showed the most prominent expression alteration: compared with the Control group (normal pancreatic tissues without CP induction), the expression level of GSDMD in the pancreatic tissues of mice in the CP group increased sharply at the 2nd and 3rd weeks of modeling. Both the intensity of GSDMD-positive signals (reflecting protein expression abundance) and the range of positive signal distribution (reflecting the scope of pyroptotic cells) were significantly higher than those at other modeling time points. This made GSDMD the most sensitive molecular marker responding to the pathological progression of CP among the detected pyroptosis-related proteins. This result directly confirmed for the first time that pyroptosis was significantly upregulated during the occurrence and development of CP, and further highlighted GSDMD as a potential core molecule mediating pyroptosis-related pancreatic tissue damage in CP. These findings not only fill the gap in understanding the role of pyroptosis in CP pathogenesis but also provide key experimental evidence for the subsequent exploration of CP treatment strategies targeting the pyroptosis pathway, especially GSDMD-related signaling.

Fig. 1.

Expression changes of pyroptosis-related proteins (NLRP3, GSDMD, and caspase-1) during the four-week modeling process of CP. a-d) IF staining images during week 1 to 4 (Scale bar: 100 μm). e) Quantitative data corresponding to IF staining results in panels a-d (n = 6), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

GSDMD-knockout (KO) mice validating the crucial role of pyroptosis in CP

Given the significant upregulation of GSDMD (a key executor of pyroptosis) observed during CP modeling, we further employed GSDMD-KO mice to experimentally verify whether pyroptosis mediated by GSDMD contributed to CP pathogenesis (Fig. 2a). Results first revealed the protective effect of GSDMD deletion against CP through changes in body weight and pancreatic weight: both the WT-CP and KO-CP groups exhibited reduced body weight and pancreatic weight, consistent with the typical manifestations of CP, such as weight loss and pancreatic atrophy (Fig. 2b-c). However, the KO-CP group showed significantly higher body weight and pancreatic weight/body weight ratio (relative pancreas weight) than the WT-CP group. This indicated that GSDMD knockout alleviated systemic and pancreatic local damage caused by CP, indirectly confirming that pyroptosis was closely associated with the severity of CP.

Fig. 2.

GSDMD-KO mice verified the crucial role of pyroptosis during CP process. a) Pattern diagram of CP modeling. b) Body weights of mice in different groups (n = 6). c) Relative pancreas weight in different groups (n = 6). d) H&E staining and IHC staining (GSDMD, FN, F4/80, iNOS, and CD206) of pancreatic tissue samples from different groups (n = 6) (Scale bar: 100 μm). e-j) Quantitative data corresponding to tissue staining results in d) (n = 6), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

IHC analyses further supported the above conclusion: IHC staining for GSDMD confirmed successful knockout (GSDMD expression was detected in the pancreas of WT mice but absent in GSDMD⁻/⁻ mice) (Fig. 2d, e); H&E staining showed that both the WT-CP and KO-CP groups presented typical CP pathological features, including pancreatic acinar atrophy, acinar-to-ductal metaplasia (ADM), and inflammatory cell infiltration. Notably, the KO-CP group exhibited more significant improvements in pancreatic tissue structure, with tighter acinar arrangement and markedly reduced tissue destruction when compared to the WT-CP group (Fig. 2d, f). In addition, staining results for FN (a fibrosis marker) (Fig. 2d, g) and F4/80 (a macrophage marker) (Fig. 2d, h) showed that GSDMD knockout significantly reduced FN deposition and F4/80-positive macrophage infiltration in pancreatic tissues, suggesting alleviated fibrosis and inflammatory responses. Staining for iNOS (an M1 macrophage marker) (Fig. 2d, i) and CD206 (an M2 macrophage marker) (Fig. 2d, j) further indicated that GSDMD deletion downregulated the M1 and M2 polarization of macrophages, implying that GSDMD-mediated pyroptosis exacerbated the inflammatory and fibrotic processes of CP by regulating macrophage polarization. These results confirm that GSDMD-mediated pyroptosis is a key pathogenic link in CP. By alleviating pancreatic tissue damage, inflammatory responses, and fibrosis, GSDMD knockout highlights the potential of targeting pyroptosis (especially GSDMD) as a CP therapeutic strategy, laying an important experimental foundation for the subsequent therapeutic effect investigations.

DSF Inhibits pancreatic acinar cell pyroptosis by suppressing STAT3 phosphorylation in the JAK/STAT signaling pathway

Following the confirmation that pyroptosis was elevated during CP progression and that GSDMD-mediated pyroptosis drived CP pathogenesis, we next focused on exploring potential therapeutic agents targeting pyroptosis. DSF, a compound previously reported to exert anti-inflammatory and anti-pyroptotic effects in other inflammatory diseases, was selected to investigate its regulatory role in pancreatic acinar cell pyroptosis, an initiating event of CP-associated pancreatic damage [54]. First, the safety of DSF on 266-6 cells was evaluated via the CCK-8 assay. Results showed no significant toxicity at DSF concentrations ranging from 0.01 to 1 µM; even at 1 µM, cell viability remained at approximately 74.96%, confirming the compound’s biocompatibility (Fig. 3a). When DSF concentration was 0.5 µM or 1 µM, the cell viability of 266-6 cells was roughly 89.15% or 74.96%, respectively. Based on this safety profile, DSF concentrations of 0.05 µM and 0.1 µM were chosen for follow-up assays.

Fig. 3.

The effect of DSF on the pyroptosis of pancreatic acinar cells. a) Cell viability (%) of 266-6 cells under different concentrations of DSF (0, 0.01, 0.05, 0.1, 0.5, 1, 2.5 µM) (n = 3). b) The mRNA expression levels of pyroptosis-related markers (NLRP3, GSDMD, and caspase-1) under different concentrations of DSF after caerulein (100 nM) stimulation (n = 3). c) LDH activity under different concentrations of DSF after caerulein (100 nM) stimulation (n = 3). d) The expression levels of pyroptosis-related proteins (NLRP3, GSDMD, and caspase-1) and p-STAT3 (Y705)/STAT3 under different concentrations of DSF after caerulein (100 nM) stimulation e) Quantitative data corresponding to WB results in d) (n = 3). β-actin was used as internal reference protein. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

At the transcriptional level, PCR assay was used to assess the effect of DSF on pyroptosis-related genes in 266-6 cells stimulated with caerulein (100 nM, a trigger of pancreatic acinar cell injury and pyroptosis) (Fig. 3b). Caerulein stimulation significantly upregulated the mRNA expression of NLRP3, GSDMD, and caspase-1, consistent with the in vivo observation of increased pyroptosis during CP. In contrast, DSF treatment dose-dependently reduced the mRNA levels of these three pyroptosis markers: at 0.1 µM, DSF downregulated NLRP3, GSDMD, and caspase-1 by approximately 46.30%, 61.14%, and 65.94%, respectively, compared to the caerulein-stimulated group. The co-treatment with Cu²⁺ further enhanced this inhibitory effect, resulting in a 65%−80% reduction in mRNA expression, confirming the synergistic anti-pyroptotic activity of DSF and Cu²⁺. The LDH release assay, an indicator of cell membrane damage during pyroptosis, provided additional evidence of the efficacy of DSF (Fig. 3c). Caerulein stimulation increased LDH activity by approximately 7.6-fold, while 0.1 µM DSF reduced this activity by ~63.2%; co-treatment with DSF and Cu²⁺ further lowered LDH activity by ~67.4%, aligning with the qPCR results and confirming that DSF could mitigate acinar cell pyroptosis-induced membrane damage.

To explore the underlying molecular mechanism, Western Blot (WB) analysis was performed to detect pyroptosis-related proteins and key signaling molecules (Fig. 3d, e). Caerulein stimulation significantly upregulated GSDMD protein expression (by ~1.4-fold compared to the Control group), whereas 0.05 µM DSF reduced GSDMD levels by ~60%. Notably, DSF also suppressed the phosphorylation of STAT3 at tyrosine 705 (Y705): the p-STAT3/STAT3 ratio in the caerulein-stimulated group was ~1.3-fold higher than in the control, but this ratio decreased by ~22% after DSF treatment. Since the JAK/STAT signaling pathway (with STAT3 as a core effector) is known to regulate inflammatory responses and pyroptosis in various cell types, these results indicate that DSF inhibits pancreatic acinar cell pyroptosis by targeting the JAK/STAT3 signaling axis [55]. These findings demonstrate that DSF safely and effectively inhibits caerulein-induced pyroptosis in pancreatic acinar cells, with enhanced efficacy when combined with Cu²⁺. The suppression of STAT3 phosphorylation in the JAK/STAT pathway is identified as a key molecular mechanism underlying this effect, providing direct experimental evidence for DSF as a potential therapeutic agent targeting pyroptosis in CP, and laying the groundwork for the subsequent development of CuSe/DSF/EL nanoflowers.

DSF inhibits PSC activation via the MAPK signaling pathway

Pancreatic fibrosis, a hallmark pathological feature of CP, is primarily driven by the activation of PSCs, which can secrete ECM components like FN upon stimulation by profibrotic factors (e.g., TGF-β) [56]. Given the critical role of PSC activation in CP progression, and our previous findings that DSF inhibits pancreatic acinar cell pyroptosis, this section focuses on investigating whether DSF can further regulate PSC activation to alleviate pancreatic fibrosis, as well as the underlying molecular mechanism.

CCK-8 assay was used to determine the safe concentration range of DSF on PSCs (Fig. 4a). It was shown that relatively low concentrations (0.01–5 µM) of DSF had no significant effect on cell viability. When the concentration of DSF was 5 µM, the cell viability was approximately 87.49%, while the cell viability decreased to 77.56% under higher DSF concentration (10 µM). Based on this safety profile and preliminary experimental screening, DSF concentrations of 1 µM and 5 µM were selected for subsequent in vitro assays to assess its effects on PSC activation, proliferation, and apoptosis. To induce PSC activation, PSCs were first stimulated with TGF-β (2 ng/mL) for 1 h, a well-established in vitro model of PSC activation in fibrosis, followed by treatment with DSF (1 µM or 5 µM) for 24 h (for transcriptional analysis) or 48 h (for protein analysis). At the transcriptional level, qPCR revealed that TGF-β stimulation significantly upregulated the mRNA expression of fibrosis-related markers, including FN, α-smooth muscle actin (α-SMA [a marker of myofibroblast-like activation of PSCs]), cysteine rich protein 61 (CYR61), and connective tissue growth factor (CTGF). Specifically, FN and α-SMA mRNA levels increased by ~3.1-fold and ~3.5-fold, respectively, compared to the Control group. In contrast, DSF treatment dose-dependently downregulated these markers: at 5 µM, DSF reduced FN, α-SMA, CYR61, and CTGF mRNA expression by ~88%, ~ 79%, ~ 54%, and ~64%, respectively, confirming its inhibitory effect on PSC activation at the transcriptional level (Fig. 4b). Further analysis of proliferation-related markers (CCND1 and p21) via qPCR showed that DSF significantly upregulated their mRNA expression: 5 µM DSF increased CCND1 and p21 levels by ~17% and ~33%, respectively, compared to the TGF-β-stimulated group, indicating that DSF might promote PSC proliferation without inducing cytotoxicity (Fig. 4c). Notably, qPCR analysis of apoptosis-related markers (Bcl-xL, Bax, and cleaved caspase-3) revealed no significant changes between DSF-treated and untreated groups, suggesting that DSF did not affect PSC apoptosis and exerted a specific regulatory effect on PSC activation and proliferation (Fig. 4d).

Fig. 4.

The effect of DSF on the activation, proliferation, and apoptosis of PSCs. a) Cell viability (%) of hPSCs under different concentrations of DSF (0, 0.01, 0.1, 0.2, 0.4, 0.6, 0.8, 1, 2.5, 5, 10 µM) (n = 3). b) The mRNA expression levels of fibrosis-related markers (FN, α-SMA, CYR61, and CTGF) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation (n = 3). c) The mRNA expression levels of proliferation-related markers (CCND1 and p21) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation (n = 3). d) The mRNA expression levels of apoptosis-related markers (Bcl-xL, Bax, and cleaved caspase-3) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation (n = 3). e) The expression levels of fibrosis-related proteins (FN, α-SMA, CYR61, and CTGF) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation and its corresponding quantitative data (n = 3). f) The expression levels of proliferation-related proteins (CCND1 and p21) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation and its corresponding quantitative data (n = 3). g) The expression levels of key proteins in the MAPK signaling pathway (p-ERK1/2, ERK1/2, p-JNK, and JNK) under different concentrations of DSF after TGF-β (2 ng/mL) stimulation and its corresponding quantitative data (n = 3). GAPDH and β-actin were used as internal reference proteins. *p < 0.05, **p < 0.01, ***p < 0.001

WB analysis was then performed to validate these findings at the protein level and explore the underlying signaling pathway. Consistent with qPCR results, TGF-β stimulation upregulated the protein expression of FN, α-SMA, CYR61, and CTGF, while DSF treatment (5 µM) reduced their levels by ~21%−67% (Fig. 4e). For proliferation markers, DSF upregulated CCND1 protein expression by ~1.4-fold (Fig. 4f), with no significant changes in apoptosis-related proteins (Fig. S1). Given that the mitogen-activated protein kinase (MAPK) signaling pathway—particularly the extracellular signal-regulated kinase 1/2 (ERK1/2) and c-Jun N-terminal kinase (JNK) branches—was known to mediate hepatic stellate cell activation in liver fibrosis (a process analogous to PSC activation in pancreatic fibrosis), we further detected the phosphorylation status of ERK1/2 and JNK in hPSCs [57]. Results showed that TGF-β stimulation increased the phosphorylation of ERK1/2 and JNK by ~1.90-fold and ~1.87-fold, respectively, compared to the Control group (Fig. 4g). In contrast, DSF treatment (5 µM) dose-dependently reduced the p-ERK1/2/ERK1/2 and p-JNK/JNK ratios by ~34% and ~48%, respectively—confirming that DSF inhibited PSC activation by suppressing the MAPK signaling pathway. Collectively, these results demonstrate that DSF specifically inhibits TGF-β-induced PSC activation (without affecting apoptosis) by suppressing the MAPK signaling pathway, while also promoting PSC proliferation, suggesting a dual regulatory role of DSF in PSC function. This finding, combined with the anti-pyroptotic effect of DSF on pancreatic acinar cells, highlights the potential of DSF as a multi-target therapeutic agent for CP, addressing both the pyroptosis and fibrosis hallmarks of the disease.

Selenium inhibits macrophage M1 polarization via the MAPK signaling pathway

In terms of the synergistic therapeutic effect of DSF and Cu²⁺, we planned to use copper-based materials to load DSF for the treatment of CP, and CuSe was chosen due to the excellent peroxidase-like activity of CuSe nanoparticles [58]. First, the safety of selenium (administered as sodium selenite [Na₂SeO₃]) on RAW 264.7 macrophages was evaluated using the CCK-8 assay (Fig. 5a). Results showed that selenium exhibited no significant cytotoxicity at concentrations ranging from 25 to 500 nM; even at the highest concentration (500 nM), macrophage viability remained at approximately 99.24%, confirming its excellent biocompatibility. Based on this safety profile and preliminary experimental screening, selenium concentrations of 50 nM and 100 nM were selected for subsequent assays to assess its effects on macrophage M1/M2 polarization. At the transcriptional level, qPCR was used to detect the expression of M1 and M2 polarization markers in macrophages stimulated with pro-inflammatory or pro-fibrotic cues. For M1 polarization (induced by lipopolysaccharide, LPS, 10 ng/mL), LPS stimulation significantly upregulated the mRNA expression of M1 markers, including inducible nitric oxide synthase (iNOS), interleukin-6 (IL-6), and C-C motif chemokine ligand 2 (CCL2), with iNOS, IL-6, and CCL2 mRNA levels increasing by ~42.8-fold, ~ 155.6-fold and ~203.8-fold, respectively, compared to the Control group (Fig. 5b). In contrast, selenium treatment dose-dependently downregulated these M1 markers: at 100 nM, selenium reduced iNOS, IL-6, and CCL2 mRNA expression by ~43%, ~ 52%, and ~22%, respectively. For M2 polarization (induced by IL-4 + IL-13, 50 ng/mL for each), IL-4 + IL-13 stimulation upregulated the mRNA expression of M2 markers, including arginase 1 (ARG1), cluster of differentiation 206 (CD206), C-C motif chemokine ligand 17 (CCL17), and insulin-like growth factor 1 (IGF-1), with ARG1 and CD206 mRNA levels increasing by ~43.4-fold and ~12.0-fold, respectively (Fig. 5c). Selenium treatment further upregulated these M2 markers in a dose-dependent manner: at 100 nM, selenium increased ARG1, CD206, CCL17, and IGF-1 mRNA expression by ~26%, ~ 49%, ~ 248%, and ~362%, respectively, indicating that selenium not only inhibited M1 polarization but also promoted M2 polarization at the transcriptional level.

Fig. 5.

The effect of selenium on the M1 and M2 polarization of macrophages. a) Cell viability (%) of RAW 264.7 cells under different concentrations of Na₂SeO₃ (0, 25, 50, 100, 250, 500 nM) (n = 3). b) The mRNA expression levels of M1-polarization markers (iNOS, IL-6, and CCL2) under different concentrations of Na₂SeO₃ after LPS (10 ng/mL) stimulation (n = 3). c) The mRNA expression levels of M2-polarization markers (ARG1, CD206, CCL17, and IGF-1) under different concentrations of Na₂SeO₃ after IL-4 + IL-13 (50 ng/mL +50 ng/mL) stimulation (n = 3). d) The secretion level of TNF-α (marker of M1 polarization) (ng/mL) in cell culture medium under different concentrations of Na₂SeO₃ after LPS (10 ng/mL) stimulation (n = 3). e) The secretion level of IGF-1 (marker of M2 polarization) (pg/mL) in cell culture medium under different concentrations of Na₂SeO₃ after IL-4 + IL-13 (50 ng/mL +50 ng/mL) stimulation (n = 3). f) Flow cytometry verifying the effect of selenium on M1/M2 polarization of macrophages. (a)-(d) referred to M1-polarization state, and (e)-(h) referred to M2-polarization state. g) The expression levels of key proteins in the MAPK signaling pathway (p-JNK, JNK, p-ERK1/2, ERK1/2, p-p38 MAPK, and p38 MAPK) under different concentrations of Na₂SeO₃ after LPS (10 ng/mL) stimulation. h) Quantitative data corresponding to the WB results in g) (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

ELISA was then used to validate these findings at the protein secretion level. For M1 polarization, LPS-stimulated macrophages showed a ~ 44.3-fold increase in TNF-α secretion when compared to the Control group; 100 nM selenium reduced TNF-α secretion by ~15% (Fig. 5d). For M2 polarization, IL-4 + IL-13-stimulated macrophages showed a ~ 1.1-fold increase in IGF-1 (a key M2-secreted factor) secretion; 100 nM selenium further increased IGF-1 secretion by ~6%, consistent with the qPCR results and confirming the regulatory effect of selenium on macrophage cytokine secretion (Fig. 5e). Flow cytometry analysis provided direct phenotypic evidence for the effect of selenium on macrophage polarization (Fig. 5f). Staining for surface markers (F4/80, CD11b, CD86) and intracellular marker (CD206) showed that LPS stimulation increased the proportion of CD86⁺ (M1) macrophages, while selenium treatment (100 nM) markedly reduced this proportion. No significant adverse effect on macrophage viability was observed, further supporting the specific regulatory role of selenium in polarization rather than cytotoxicity.

Researches have shown that the MAPK signaling pathway can regulate macrophage polarization and inflammatory response in RAW 264.7 cells [59]. Therefore, to explore the underlying molecular mechanism, WB analysis was performed to detect the expressions of key proteins in the MAPK signaling pathway (Fig. 5g, h). LPS stimulation significantly upregulated the phosphorylation of JNK (T183/Y185) and ERK1/2 (T202/Y204) by ~19.1-fold and ~14.1-fold, respectively, compared to the Control group—indicating activation of the MAPK pathway during M1 polarization. In contrast, selenium treatment (100 nM) dose-dependently reduced the phosphorylation levels of JNK and ERK1/2: the p-JNK/JNK and p-ERK1/2/ERK1/2 ratios decreased by ~56% and ~55%, respectively. No significant changes in p38 MAPK phosphorylation were observed, suggesting that selenium specifically targeted the JNK and ERK1/2 branches of the MAPK pathway to inhibit M1 polarization. These results demonstrate that selenium effectively inhibits macrophage M1 polarization (alleviating pro-inflammatory responses) via suppressing the JNK/ERK1/2 branches of the MAPK signaling pathway. This finding not only clarifies the anti-inflammatory role of selenium (a core component of CuSe) but also provides a rationale for the synergistic design of the CuSe/DSF/EL nanoflower system, where CuSe contributes selenium-mediated immunomodulation, DSF targets pyroptosis and PSC activation, and EL enables intestinal targeting.

Design and characterization of CuSe/DSF/EL nanoflowers

Guided by the preceding mechanistic findings including the dual inhibitory effects of DSF on pancreatic acinar cell pyroptosis and PSC activation, and the regulatory role of selenium in macrophage polarization, the CuSe/DSF/EL nanoflower system was rationally designed. This system integrates three core functional components to address key challenges in CP treatment: CuSe acts as both a nanocarrier for targeted DSF delivery and a sustained selenium release source; DSF serves as the core therapeutic agent to target inflammatory (pyroptosis) and fibrotic (PSC activation) pathological processes during CP; and the enteric coating (EL) enables intestinal targeting, avoiding gastric acid-induced DSF degradation and improving the bioavailability of therapeutic components.

SEM (Fig. 6a, b) and TEM (Fig. 6c, d) images showed that the synthesized CuSe/DSF/EL nanoflowers exhibited a uniform nanoflower morphology with a particle size of approximately 280 nm. Element mapping further demonstrated that Cu and Se elements were evenly distributed in the CuSe/DSF/EL nanoflowers (Fig. 6e). The XRD pattern further confirmed the successful synthesis of CuSe nanoflowers, with diffraction peaks consistent with the (101), (107), and (201) crystal planes of the standard hexagonal CuSe phase (JCPDS No. 86–1240), (Fig. 6f). Additionally, XPS analysis showed that the binding energies of Cu 2p in the CuSe/DSF/EL nanoflowers were located at 952.45 eV and 932.55 eV, respectively, which corresponded to Cu²⁺; the binding energies of Se 3d were located at 54.85 eV and 53.95 eV, which corresponded to Se²⁻ (Fig. 6g, h). These data further confirmed the successful preparation of CuSe/DSF/EL nanoflowers.

Fig. 6.

Electron microscope images of the CuSe/DSF/EL nanoflowers. a-b) SEM images of CuSe/DSF/EL nanoflowers at different magnification. c-d) TEM images of CuSe/DSF/EL nanoflowers at different magnificationm). e) Element mapping images of different elements in the CuSe/DSF/EL nanoflowers (Scale bar: 100 nm). f) The XRD pattern of CuSe nanoflowers. g) The XPS spectra of Cu 2p in the CuSe/DSF/EL nanoflowers. h) The XPS spectra of Se 3d in the CuSe/DSF/EL nanoflowers

Excellent stability under physiological conditions is one of the important indicators for biomedical materials [60]. DLS analysis showed that CuSe nanoflowers without encapsulation of other components were unstable in PBS, with significant size changes within 3 days (Fig. S2a). However, CuSe/DSF/EL nanoflowers exhibited good colloidal stability in PBS (Fig. S2b). DLS analysis showed that the hydrated particle size of CuSe/DSF/EL nanoflowers in PBS was 318.2 nm, with a polydispersity index of 0.2, indicating the nanoflowers possessed good dispersion stability. By monitoring the hydration particle size of CuSe/DSF/EL nanoflowers over time, it was found that there was no significant aggregation phenomenon within 3 days, which supported the good physiological stability of CuSe/DSF/EL nanoflowers. In addition, the measured polydispersity index remained below 0.2, suggesting that the system had a high degree of dispersion, which met the requirements of biomedical applications for the uniformity of nanomaterials.

The zeta potential of the samples was measured in deionized water (Fig. 7a). The CuSe nanoflowers displayed a surface potential of −16.67 mV. After loading with DSF, the zeta potential of CuSe/DSF nanoparticles increased to −6.50 mV. Following further modification with negatively charged EL, the potential of CuSe/DSF/EL nanoflowers decreased to −19.5 mV, confirming the successful coating of EL. Given the positively charged microenvironment typically found in pancreatitis lesions due to inflammatory cell infiltration, the strong negative surface charge imparted by EL modification is expected to promote electrostatic accumulation of the nanoflowers at the inflammatory site. This charge-targeting feature provides a foundation for subsequent therapeutic applications. UV-vis-NIR absorption spectra revealed that CuSe nanoflowers exhibited a characteristic absorption peak near 525 nm. After DSF loading, an additional peak appeared around 350 nm, corresponding to DSF, indicating successful drug incorporation. Subsequent EL modification introduced a distinct absorption peak at 425 nm, further verifying surface functionalization (Fig. 7b). FTIR analysis showed a characteristic band at 1498 cm^–1 assigned to the C–N group of DSF, as well as peaks at 1741 cm^–1 and 1386 cm^–1 corresponding to the C = O and C–O groups of EL, respectively (Fig. 7c). These spectral features corroborate the successful loading of DSF and coating with EL. Furthermore, TG analysis indicated effective loading of DSF and EL onto the CuSe nanoflowers, with calculated loading contents of 46.30% and 15.84%, respectively (Fig. 7d).

Fig. 7.

Characterization and antioxidant capacity of CuSe/DSF/EL nanoflowers. a) Zeta potentials of different components in deionized water (mV). b) The UV-vis-NIR spectrum of different components (nm). c) The FTIR spectrum of different components (nm^− 1). d) TG analysis of CuSe, CuSe/DSF, and CuSe/DSF/EL nanoflowers. e) Color changes and absorbance of ABTS· solution after the incubation with different concentrations of CuSe/DSF/EL nanoflowers (0, 6, 12, 18, 24, and 30 µg/mL). f) Color changes and absorbance of DPPH· solution after the incubation with different concentrations of CuSe/DSF/EL nanoflowers (0, 6, 12, 18, 24, and 30 µg/mL). g) Color changes and absorbance of TMBox solution after the incubation with different concentrations of CuSe/DSF/EL nanoflowers (0, 6, 12, 18, 24, and 30 µg/mL). h) The UV-visible spectrum under simulated intestinal environment (pH = 6.8) for several days. i) The in vitro release of Cu²⁺ from the nanoflowers under different physiological environments (simulated gastric fluid (SGF) and simulated intestinal fluid (SIF))

Evaluation of in vitro antioxidant capacity of CuSe/DSF/EL nanoflowers

Given that oxidative stress response is a key driver of pancreatic tissue oxidative stress and pathological progression in CP, this section focuses on evaluating the nanoflowers’ in vitro antioxidant capacity, aiming to clarify whether they can alleviate CP-associated oxidative damage by scavenging ROS, thereby complementing their anti-pyroptotic, anti-fibrotic, and immunomodulatory effects (Fig. S3a) [61]. First, the total antioxidant capacity of CuSe/DSF/EL nanoflowers was studied using ABTS radical cation scavenging assay. ABTS solution appeared green with a characteristic absorption peak at 752 nm. It was shown that as the concentration of CuSe/DSF/EL nanoflower solution continuously increased, the color of ABTS· solution gradually become lighter, with a decrease in absorbance at 752 nm (Fig. 7e). This indicated that the clearance of the nanoflowers on ABTS· was concentration-dependent. When the concentration of CuSe/DSF/EL nanoflowers was 30 µg/mL, the clearance rate of ABTS· was 65.78 ± 2.01%, which confirmed the CuSe/DSF/EL nanoflowers had excellent total antioxidant capacity. Then, the scavenging ability of CuSe/DSF/EL nanoflowers on RNS was detected by DPPH method. As shown in Fig. 7f, as the concentration of CuSe/DSF/EL nanoflower solution increased, the color of DPPH· solution gradually faded, accompanied by a decrease in absorbance at 519 nm, indicating that the clearance of CuSe/DSF/EL nanoflowers on DPPH· was also dose-dependent. When the concentration of CuSe/DSF/EL nanoflowers reached 30 µg/mL, the scavenging rate on DPPH· could reach as high as 46.75 ± 7.82%, which suggested CuSe/DSF/EL nanoflowers possessed good RNS scavenging ability. Finally, the scavenging ability of CuSe/DSF/EL nanoflowers on ROS was investigated. TMBox solution appears blue and has a characteristic absorption peak at 652 nm. As shown in Fig. 7g, with the increase of CuSe/DSF/EL nanoflower solution concentration, the color of TMBox solution gradually become lighter, accompanied by a decrease in absorbance at 652 nm, which indicated the clearance of CuSe/DSF/EL nanoflowers on ·OH was concentration-dependent. When the concentration of CuSe/DSF/EL nanoflowers reached 30 µg/mL, the clearance rate of ·OH was approximately 62.67 ± 3.30%, indicating that CuSe/DSF/EL nanoflowers had excellent ROS scavenging ability. Vitamin C (L-ascorbic acid) is one of the representative antioxidant agents commonly investigated for CP[62]. We further compared the antioxidant capacity of our CuSe/DSF/EL nanoflowers with that of Vitamin C (Fig. S4). It was found that the elimination ratios of Vitamin C on DPPH· and ABTS· were ideally high. Though the scavenging ratios of the CuSe/DSF/EL nanoflowers on these free radicals were not as high as those of Vitamin C, these results indicated that the nanoflowers possessed a good free radical scavenging and antioxidant capacity. Collectively, the in vitro antioxidant evaluation results confirm that CuSe/DSF/EL nanoflowers possess strong and comprehensive antioxidant capacity, which is derived from the synergistic effect of CuSe and DSF in the integrated system. This antioxidant property enables the nanoflowers to alleviate ROS-mediated oxidative stress in CP, complementing their previously verified anti-pyroptotic, anti-fibrotic, and immunomodulatory effects.

Biodegradability and intestinal microenvironment responsiveness of CuSe/DSF/EL nanoflowers

The biodegradability and environmental responsiveness of biomaterials under physiological conditions are crucial for ensuring reliable drug delivery efficiency and therapeutic safety, particularly in oral formulations designed for CP [63]. Therefore, the biodegradability, which is essential to prevent long-term tissue accumulation, and the responsiveness to the intestinal microenvironment, which ensures the targeted release of therapeutic components, were assessed under simulated physiological conditions to model the oral administration of CuSe/DSF/EL nanoflowers. To investigate the biodegradability of CuSe/DSF/EL nanoflowers, UV-vis-NIR technology was applicated to record the changes of CuSe/DSF/EL nanoflowers’ UV-visible spectra under simulated intestinal environment (pH = 6.8). As shown in Fig. 7h, after 5 days, the color of CuSe/DSF/EL nanoflower solution changed from brown to red, and the characteristic absorption peak near 525 nm disappeared, which demonstrated the degradability of CuSe/DSF/EL nanoflowers under simulated intestinal environment. Subsequently, XPS was used to analyze the changes in the ion valence states of copper and selenium elements after the degradation (Fig. S3b, c). After degradation, the binding energy of Cu 2p showed new characteristic peaks of Cu⁰ at 943.62 eV and 963.14 eV, while Se 3d showed new characteristic peaks of SeO₃²⁻ at 59.25 eV, verifying the degradation of CuSe/DSF/EL nanoflowers.

Additionally, the intestinal microenvironment responsiveness of CuSe/DSF/EL nanoflowers was evaluated by detecting and comparing the in vitro release of Cu²⁺ (a key component derived from CuSe degradation, which enhances DSF efficacy) under two simulated physiological environments: simulated gastric fluid (SGF [pH 1.2, mimicking the stomach]) and simulated intestinal fluid (SIF [pH 6.8, mimicking the small intestine], Fig. 7i). The results indicated that Cu²⁺ release was highly environment-dependent: under SIF conditions, Cu²⁺ was released in significant quantities, while under SGF conditions, Cu²⁺ release was negligible. This finding demonstrated that CuSe/DSF/EL nanoflowers possess specific responsiveness to the intestinal microenvironment, a property that avoids premature release of components in the stomach (preventing gastric acid-induced degradation of DSF and reducing potential gastric irritation) and ensures concentrated release in the intestine. This targeted release ability can significantly improve the efficiency of DSF transport to the systemic circulation and subsequent pancreatic lesion sites, thereby enhancing its therapeutic efficacy for CP.

Biosafety evaluation of CuSe/DSF/EL nanoflowers

Hemolysis assay was performed to evaluate the blood safety of CuSe/DSF/EL nanoflowers. Different concentrations of CuSe/DSF/EL nanoflowers were incubated with RBCs at 37℃ for 2 h, followed by centrifugation and recording of supernatant absorbance at 541 nm to calculate the hemolysis rate. As shown in Fig. S3d, when the CuSe/DSF/EL nanoflowers was at higher concentrations (250 µg/mL), the hemolysis rate remained below 5%. In addition, the RBCs in the positive control group underwent complete hemolysis and rupture, while the RBCs in the CuSe/DSF/EL nanoflower group were similar to that of the negative control group, which could be completely precipitated via centrifugation. These results indicated that the CuSe/DSF/EL nanoflowers showed no significant toxic effect on RBCs. In addition, the cell safety of CuSe/DSF/EL nanoflowers was evaluated using 266-6 cells. The results of Live/Dead cell staining confirmed the excellent cell safety of CuSe/DSF/EL nanoflowers: most 266-6 cells treated with the nanoflowers appeared as living cells (stained green), without significant cytotoxicity to 266-6 cells within the tested concentration range, further verifying the nanoflowers’ good biocompatibility (Fig. S5).

Afterwards, the safety of CuSe/DSF/EL nanoflowers at the tissue level were investigated. On the 14th day, there was no significant difference in the levels of major markers such as WBC, lymphocytes (LYM), monocytes (MONO), neutrophils (GRAN), RBC, and PLT in the blood samples of different groups (Fig. S6a). Moreover, the results of blood biochemistry test showed no significant difference in important liver and kidney function indicators such as ALT, AST, CREA, and UA between control and experimental groups (Fig. S6b). Major organs (e.g., heart, liver, spleen, lungs, and kidneys) of mice were then collected for H&E staining (Fig. S6c). Obviously, CuSe/DSF/EL nanoflowers had no significant adverse effect on major organs during the 14-day period, which also supported the in vivo safety of nanoflowers.

Therapeutic efficacy of CuSe/DSF/EL nanoflowers in a mouse model of CP

Attenuation of systemic wasting and pancreatic atrophy

Progressive body weight loss and pancreatic atrophy are hallmark systemic manifestations of CP, resulting from chronic inflammation-induced catabolism and acinar cell loss [64]. Mouse CP model was established as mentioned before (Fig. 8a). As presented in Fig. 8b, mice in the CP group exhibited a significant 20.2% reduction in body weight relative to the Control group by the end of the four-week modeling period (p < 0.0001). In contrast, treatment with CuSe/DSF/EL nanoflowers markedly mitigated this weight loss, with a final body weight only 9.0% lower than that of the Control group. This protective effect was superior to that of free DSF (16.7% weight loss vs. Control) or CuSe nanoflowers (11.9% weight loss vs. Control), which showed only moderate improvements (p < 0.0001 and p < 0.01, respectively vs. CP group).

Fig. 8.

In vivo experiments verifying the therapeutic effect of CuSe/DSF/EL nanoflowers on pancreatic exocrine function and tissue integrity. a) Pattern diagram of CP modeling. b) Body weights of mice in different groups (n = 6). c) Relative pancreas weight in different groups (n = 6). d) The levels of lipase (U/L), AMY (U/dL), TNF-α (pg/mL), and IL-1β (pg/mL) in different groups (n = 6). e) Quantitative data corresponding to the H&E staining results in f) (n = 6). f) H&E staining images of pancreatic tissue samples from different groups (n = 6) (Scale bar: 100 μm). **p < 0.01, ****p < 0.0001

As a critical indicator of pancreatic atrophy, relative pancreas weight was significantly reduced in the CP group (1.4 ± 0.3) compared to the Control group (8.5 ± 1.0, p < 0.0001; Fig. 8c), which was consistent with body weight changes. The CuSe/DSF/EL group displayed a relative pancreas weight of 3.3 ± 0.7, representing an approximate 133% recovery relative to the CP group (p < 0.0001). In contrast, the CP + DSF and CP + CuSe groups showed more modest recoveries (2.6 ± 0.2 and 2.8 ± 0.4, respectively; both p < 0.0001 vs. CP group). These findings collectively demonstrate that CuSe/DSF/EL nanoflowers exert a superior protective effect against CP-induced systemic wasting and pancreatic structural deterioration.

Restoration of pancreatic exocrine function and suppression of systemic inflammation

Serum pancreatic enzymes (e.g., amylase [AMY] and lipase) and pro-inflammatory cytokines (e.g., TNF-α and interleukin-1β [IL-1β]) were measured to assess pancreatic exocrine function and systemic inflammatory status [65, 66]. As shown in Fig. 8d, no significant differences in serum AMY levels were observed among groups, which might be attributed to the progressive loss of acinar cells (the primary source of AMY) in advanced CP. In contrast, serum lipase level, a marker of pancreatic exocrine dysfunction with strong specificity, was reduced by 82.4% in the CP group compared to the Control group (p < 0.0001). Treatment with CuSe/DSF/EL nanoflowers restored serum lipase to 34.3% of the Control group level (p < 0.01 vs. CP group), whereas free DSF and CuSe nanoflowers only restored lipase to 20.9% and 22.6% of normal values, respectively (p < 0.05 and p < 0.01, respectively vs. CP group). ELISA was also used to compare the differences in fecal elastase level in different groups. As shown in Fig. S7, compared with the Control group, fecal elastase levels were extremely significantly decreased in the CP group, indicating severe exocrine pancreatic dysfunction in the progression of CP. While single-agent treatment with DSF or CuSe could slightly elevate elastase levels, the CuSe/DSF/EL nanoflower treatment remarkably restored fecal elastase levels to near-normal values, which demonstrated that the CuSe/DSF/EL nanoflower effectively ameliorated CP-induced pancreatic exocrine dysfunction. Furthermore, it was observed from Fig. S8 that the fluorescence expression levels were higher in the CP group when compared to the Control group, indicating that the pancreatic uptake of the nanoflowers was higher in the CP group than that of the Control group. These results validated the specific targeting of the nanoflowers to the pancreatitis site.

For inflammatory cytokines (Fig. 8d), the CP group exhibited a 33.6-fold increase in TNF-α (24.5 ± 3.3 pg/mL vs. 0.73 ± 0.6 pg/mL in Control group) and a 48.3-fold increase in IL-1β (19.3 ± 3.1 pg/mL vs. 0.4 ± 0.3 pg/mL in Control group; both p < 0.0001), reflecting severe systemic inflammation. All treatment groups reduced cytokine levels, but the CuSe/DSF/EL group showed the most dramatic suppression: TNF-α and IL-1β were decreased by 93.0% and 94.4%, respectively, relative to the CP group (p < 0.0001). This synergistic anti-inflammatory effect likely stems from the combined actions of DSF (inhibiting GSDMD-mediated pyroptosis), selenium (modulating macrophage polarization), and Cu²⁺ (enhancing DSF bioactivity via CuET formation).

Amelioration of pancreatic histopathological damage

H&E staining was performed to quantify CP-associated histological alterations, including acinar atrophy, ADM, inflammatory infiltration, and fibrosis. As illustrated in Fig. 8e, f, the CP group displayed extensive pancreatic parenchymal damage: normal acinar architecture was replaced by disorganized ductal structures, accompanied by massive inflammatory cell infiltration and fibrosis formation. The histopathological score of the CP group (9.0 ± 1.0) was significantly higher than that of the Control group (1.7 ± 0.6, p < 0.001). Treatment with CuSe/DSF/EL nanoflowers reduced the score to 3.0 ± 1.0 (p < 0.01 vs. CP group), with pancreatic tissues showing preserved acinar arrangement, minimal inflammatory cell accumulation, and less obvious fibrosis. In comparison, the CP + DSF and CP + CuSe groups showed less pronounced improvements, with scores of 5.7 ± 0.6 (p < 0.01 vs. CP group) and 7.3 ± 0.6, respectively (p > 0.05 vs. CP group).

Inhibition of pancreatic fibrosis and macrophage infiltration

IHC staining for FN, a major ECM component and key fibrosis marker—and F4/80, a specific marker for macrophages—was conducted to evaluate fibrosis and inflammatory cell recruitment. As shown in Fig. 9a, c, the CP group exhibited a 96.5-fold increase in FN-positive ECM deposition (area percentage: 43.8 ± 5.3% vs. 0.5 ± 0.1% in Control group) and a 7.5-fold increase in F4/80-positive macrophage infiltration (area percentage: 19.6 ± 2.6% vs. 2.6 ± 1.8% in Control group; both p < 0.001). Treatment with CuSe/DSF/EL nanoflowers significantly downregulated FN deposition by 69.8% and reduced macrophage infiltration by 94.3% (both p < 0.001 vs. CP group), whereas free DSF and CuSe nanoflowers only inhibited these parameters by 34–58% (all p < 0.01 vs. CP group). These results confirm that the CuSe/DSF/EL nanoplatform effectively suppresses CP-associated fibrosis and inflammatory cell infiltration—two key drivers of disease progression.

Fig. 9.

In vivo experiments verifying the therapeutic effect of CuSe/DSF/EL nanoflowers on pancreatic inflammation, fibrosis, and pyroptosis. a) IHC staining (FN and F4/80) of pancreatic tissue samples from different groups (n = 6) (Scale bar: 100 μm). b) IF staining (NLRP3, GSDMD, and caspase-1) of pancreatic tissue samples from different groups (n = 6) (Scale bar: 100 μm). c) Quantitative data corresponding to the tissue staining results in (a) and (b) (n = 6). **p < 0.01, ****p < 0.0001

Suppression of pancreatic pyroptosis

IF staining was used to assess the expression of pyroptosis-related proteins, including NLRP3 (the core component of the canonical inflammasome), GSDMD (the key executor of pyroptotic cell death), and caspase-1 (the pro-pyroptotic protease that cleaves GSDMD) [67]. As presented in Fig. 9b, c, the CP group displayed robust upregulation of NLRP3 and GSDMD compared to the Control group, with area percentage increased by 3.9-fold and 2.3-fold, respectively (both p < 0.001). Quantitative analysis revealed that CuSe/DSF/EL nanoflower treatment reduced the area percentages of NLRP3, GSDMD, and caspase-1 by 16.8%, 47.7% (p < 0.001 vs. CP group), and 13.3%, respectively. Notably, GSDMD showed the most dramatic reduction, which aligned with our earlier finding that GSDMD was a sensitive and critical mediator of CP-associated pyroptosis. In contrast, free DSF had no inhibitory effect on caspase-1, and CuSe nanoflowers showed no inhibitory effect on NLRP3, both of which were not as effective as nanoflowers with intact components in inhibiting pyroptosis-related proteins as a whole, further highlighting the superior therapeutic efficacy of the integrated nanoplatform. Collectively, these in vivo data demonstrate that CuSe/DSF/EL nanoflowers alleviate CP by preserving pancreatic exocrine function and tissue integrity, as well as suppressing systemic and local inflammatory responses. They also exert a therapeutic effect through inhibiting pancreatic fibrosis and macrophage recruitment, and downregulating GSDMD-mediated pyroptosis. Based on the fluorescence expression levels of pyroptosis-related proteins (e.g., NLRP3, GSDMD, and caspase-1) in pancreatic tissue sections in different groups, it was obvious that both DSF and nanoflower treatment significantly reduced GSDMD levels, and the expression of fibrosis and inflammation was also markedly reduced compared to the CP group. Combined with our previous CP model confirming that there were significant changes in pyroptosis and fibrosis during the process of CP, and the subsequent WB molecular experiments in different cells verifying the inhibitory effect of DSF on GSDMD and that of selenium on inflammation, it could be inferred that the therapeutic effect of CuSe/DSF/EL nanoflowers on CP was mainly related to the inhibitory effect of DSF on GSDMD-mediated pyroptosis. The enhanced efficacy of CuSe/DSF/EL compared to single-component treatments underscores the advantages of the nanoplatform, including improved DSF bioavailability via pH-responsive intestinal release, synergistic bioactivity of Cu²⁺ and selenium, and targeted accumulation at inflammatory sites.

RNA-seq analysis of CuSe/DSF/EL nanoflowers’ therapeutic effects on CP

Global transcriptomic profiling and differential gene expression

To systematically study the molecular mechanisms underlying the therapeutic effects of CuSe/DSF/EL nanoflowers on CP, RNA-seq was performed. Principal Component Analysis (PCA) was first conducted to evaluate the overall similarity and separation of transcriptomic profiles across groups [68]. As shown in Fig. 10a, samples were distinctly clustered into three independent groups according to their experimental conditions, with no overlap between the Control, CP, and CP + CuSe/DSF/EL groups. This clear clustering confirmed that CP induction and CuSe/DSF/EL treatment exerted profound and distinguishable effects on the pancreatic transcriptome. Venn diagram analysis further delineated the overlap of differentially expressed genes (DEGs) between comparison pairs (Fig. 10b). A total of 15,347 DEGs were identified between the Control and CP groups (15,259 upregulated and 88 downregulated), reflecting the extensive transcriptional reprogramming associated with CP pathogenesis. In contrast, comparison of the CP and CP + CuSe/DSF/EL groups revealed 82 DEGs, including 49 upregulated and 33 downregulated genes (Fig. 10c, d). The reduction in DEG number following CuSe/DSF/EL treatment suggests that the nanoflowers partially reverse the CP-associated transcriptomic dysregulation, consistent with the in vivo therapeutic effects observed in histopathology and functional assays.

Fig. 10.

RNA-seq analyzing the gene expressions of different groups. a) PCA analysis analysis of three sample clusters (n = 4). b) Venn diagram. c) Volcano plots of DEGs between the Control group and the CP group. d) Volcano Plots of DEGs between the CP group and the CP + CuSe/DSF/EL group. e) GO enrichment analysis of DEGs between different groups

Functional enrichment analysis of DEGs

To elucidate the biological significance of the identified DEGs, Gene Ontology (GO) enrichment analysis was performed, encompassing three domains: Biological Process (BP), Molecular Function (MF), and Cellular Component (CC) (Fig. 10e). In the BP domain, DEGs between the CP and CP + CuSe/DSF/EL groups were significantly enriched in pathways related to cellular process, biological regulation, metabolic process, response to stimulus, and multicellular organismal process. For the MF domain, the CP + CuSe/DSF/EL group was enriched in terms including binding (e.g., protein binding, ion binding), catalytic activity (e.g., enzyme activity), and molecular function regulator activity, suggesting that the nanoflowers modulate CP by restoring the function of key regulatory molecules (e.g., enzymes, signaling proteins) disrupted during CP progression. In the CC domain, DEGs were primarily localized to cellular anatomical entities and protein-containing complexes, indicating that the nanoflowers target multiple subcellular compartments to exert their therapeutic effects. Notably, the CP group exhibited robust enrichment of GO terms associated with pathological processes, including inflammatory response, metabolic disorder, and cellular homeostasis imbalance. In contrast, the CP + CuSe/DSF/EL group showed enhanced enrichment of terms linked to tissue protection and repair, antioxidant response, and molecular transport—aligning with the nanoflowers’ demonstrated antioxidant and anti-inflammatory properties.

KEGG pathway analysis reveals regulation of pancreatic secretion

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was further performed to identify signaling cascades modulated by CuSe/DSF/EL nanoflowers. As shown in Fig. S9, DEGs between the CP and CP + CuSe/DSF/EL groups were significantly enriched in the pancreatic secretion pathway, a critical pathway dysregulated in CP due to acinar cell damage and exocrine dysfunction. Key genes involved in pancreatic secretion that are differentially regulated include serine proteases (which mediate trypsin activation and are central to pancreatic inflammatory cascades), chymotrypsin-like proteases (essential for protein digestion and pancreatic exocrine function), carboxypeptidases (involved in peptide hydrolysis), pancreatic lipases (critical for fat digestion), and phospholipases (which regulate lipid metabolism and inflammatory mediator release). Dysregulation of these genes in the CP group is consistent with the loss of pancreatic exocrine function observed in clinical CP. The restoration of their expression following CuSe/DSF/EL treatment suggests that the nanoflowers alleviate CP, in part, by reversing pancreatic secretion dysfunction, an effect that may contribute to reduced inflammation and improved tissue homeostasis. Collectively, the RNA-seq results demonstrate that CuSe/DSF/EL nanoflowers modulate CP pathogenesis through a multi-targeted transcriptional program. This includes suppression of genes driving inflammatory signaling (consistent with reduced inflammatory cell infiltration observed in tissue staining), restoration of genes involved in pancreatic exocrine function (aligning with improved serum lipase levels and preserved acinar architecture), and upregulation of genes associated with antioxidant defense and cellular repair (supporting the nanoflowers’ in vitro antioxidant capacity). This transcriptomic profile further validates the nanoflowers as a promising multi-mechanistic therapeutic strategy for CP.

16S rDNA sequencing analysis verifies the effect of CuSe/DSF/EL nanoflowers on gut microbiota during CP

As a technique for identification and classification of bacteria in microbial communities, 16S rDNA sequencing analysis is widely used in researches related to various diseases such as IBD and cancer [69–72]. To figure out the effect of the CuSe/DSF/EL nanoflowers on the diversity and composition of gut microbiota, feces from mice in different groups were collected, and 16S rDNA sequencing analysis was then carried out. It could be observed from the rarefaction curves of Chao1, Observed species, and Shannon indices that the richness and diversity of samples in the CP group were significantly reduced when compared with the Control group, which confirmed gut microbiota imbalance caused by CP modeling (Fig. 11a). Furthermore, the treatment of DSF, CuSe, and CuSe/DSF/EL nanoflowers could restore microbiota balance to some extent.

Fig. 11.

Alpha diversity, beta diversity, and species abundance analysis of feces samples from different groups of mice (n = 5). a) Rarefaction curves of main indices in terms of alpha diversity including Chao1, Observed species, and Shannon indices in different groups. b) PCA plot. Points with the same color represented different samples within the same group. In contrast, points with different colors represented different groups. The same group was presented in the form of a circle (n = 5) with a 95% CI. c) The top 30 most abundant genera shown by stacked barplot at the genus level in each group, with different colors representing different genera and the height of each block corresponding to the relative abundance of the genus in the group. d, e) Significance difference analysis through relative abundance between different groups at the genus level. f) LEfSe analysis between CP group and CP + CuSe/DSF/EL group. Different circle layers represented different classification levels (domain, phylum, class, order, family, genus, and species). Each node represented a species, and a larger node represents a higher species abundance

Furthermore, we used PCA and non-parametric multivariate analysis of variance (Adonis) to analyze the differences in microbial composition among different groups. It was observed that the distance among samples within each group was relatively close (Fig. 11b), indicating the microbial composition and structure within the same group were similar. However, there was an obvious discrepancy in the distance among samples in the Control and CP group, similarly verified microbiota imbalance caused by CP intervention. The treatment of DSF or CuSe alone could partly restore the balance in micro-flora’s composition and structure in the progression of CP, while the administration of the CuSe/DSF/EL nanoflowers (CP + CuSe/DSF/EL group) rendered the gut microbiota of mice closest to the normal physiological state, confirming the nanoflowers’ microbial balance regulation effect. Adonis analysis showed a statistically significant difference among different groups (p < 0.01).

As shown in the Fig. 11c, the Control group presented a balanced and diverse microbiota structure, with Lactobacillus and Akkermansia as secondary dominant bacteria, without excessive proliferation of a single bacterial genus, possessing typical characteristics of a healthy gut microbiota. However, the CP group showed a severe restructuring of the bacterial community structure, with excessive proliferation of Lactobacillus and a significant decrease in bacterial diversity. The originally low-abundance bacteria were greatly inhibited, which was a typical manifestation of microbial imbalance and selective proliferation. As for the impact of different treatments on gut microbiota, it was observed that compared with the CP group, the excessive proliferation of Lactobacillus was alleviated in the CP + DSF group, and the abundance of probiotics such as Akkermansia and Bifidobacterium was significantly increased, indicating a healthier microbiota structure. For the CP + CuSe group, the abundance of Akkermansia reached to a significantly higher level than the CP group, which was a characteristic of selective proliferation of Akkermansia. Meanwhile, the diversity of bacterial community was higher than that in the CP group. Notably, in the CP + CuSe/DSF/EL group, Akkermansia became the absolute dominant bacteria, while the abundance of Lactobacillus further decreased. The proportion of other bacteria increased compared to the CP group, and its diversity of microbiota was the best in the intervention groups, with a structure dominated by probiotics (Akkermansia + Lactobacillus), indicating that the CuSe/DSF/EL nanoflowers were beneficial for the repair and reconstruction of gut microbiota in CP, which was further confirmed by the microbiota abundance differential analysis and LEfSe analysis at the genus-level (Fig. 11d-f).

Conclusion

This study reveals that pyroptosis is upregulated in CP, mediated by GSDMD, and that DSF concurrently suppresses acinar cell pyroptosis via JAK/STAT3 signaling axis and PSC activation via MAPK signaling. Selenium further alleviates inflammation by modulating macrophage polarization. Based on this rationale, we designed gut microenvironment-responsive CuSe/DSF/EL nanoflowers. This platform incorporates CuSe as a selenium donor and DSF carrier, DSF as the therapeutic agent, and EL for acid protection and intestinal pH-responsive release. The nanoflowers exhibit robust antioxidant capacity, good biocompatibility (no significant hemolysis, cytotoxicity, or organ injury), and electrostatic targeting of inflamed pancreatic sites via EL-conferred negative charge. In vivo experiments in CP mice demonstrate that the nanoflowers effectively mitigate pancreatic inflammation (reduced TNF-α/IL-1β), fibrosis (decreased FN deposition), and pyroptosis (downregulated NLRP3/GSDMD/caspase-1), while reversing systemic wasting and pancreatic exocrine dysfunction. RNA-seq confirms correction of CP-associated transcriptomic dysregulation, particularly in pancreatic secretion pathways. Further 16S rDNA sequencing analysis verifies the regulating effect of the nanoflowers on gut microbiota mainly focusing on increasing abundance of beneficial bacteria such as Akkermansia and restoring microbiota balance. This work not only clarifies pyroptosis’s role in CP but also provides a translational nanoplatform that enhances the efficacy and targeting of DSF, offering a novel strategy for treating fibro-inflammatory pancreatic diseases.

Abbreviations

CP

Chronic pancreatitis

DSF

Disulfiram

PSCs

pancreatic stellate cells

EL

Eudragit^® L100-55

RNA-seq

RNA sequencing

ECM

Extracellular matrix

TGF-β

Transforming growth factor-β

FN

Fibronectin

AP

Acute pancreatitis

GSDMD

Gasdermin D

siRNA

Small interfering RNA

AIM2

Absent in melanoma 2

GSDME

Gasdermin E

TNF-α

Tumor necrosis factor-α

UC

Ulcerative colitis

GSK3β

Glycogen synthase kinase 3β

Nrf2

Nuclear factor-erythroid 2 related factor 2

Cu²⁺

Copper ions

CuET

Copper diethyldithiocarbamate

ROS

Reactive oxygen species

DMEM

Dulbecco’s Modified Eagle’s Medium

FBS

Fetal bovine serum

PVDF

Polyvinylidene fluoride

IL-4

Interleukin-4

IL-13

Interleukin-13

PBS

Phosphate buffer saline

CCK-8

Cell counting kit-8

LDH

Lactate dehydrogenase

BCA

Bicinchoninic acid

RIPA

Radio immunoprecipitation assay

SDS-PAGE

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

ELISA

Enzyme-linked immunosorbent assay

IGF-1

Insulin-like growth factor 1

CuCl₂·2H₂O

Copper chloride dihydrate

NaOH

Sodium hydroxide

AA

Ascorbic acid

NaBH₄

Sodium borohydride

ABTS

2-2’-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid

DPPH

2-2-Diphenyl-1-picrylhydrazyl

TMB

3-3’-5-5’-tetramethylbenzidine

K₂S₂O₈

Potassium persulfate

H₂O₂

Hydrogen peroxide

FeCl₃

Ferric chloride

IF

Immunofluorescence

WT

Wild-type

H&E

Hematoxylin and Eosin

IHC

Immunohistochemistry

TEM

Transmission electron microscopy

SEM

Scanning electron microscopy

FTIR

Fourier transform infrared spectroscopy

TG

Thermogravimetric

XPS

X-ray photoelectron spectroscopy

XRD

X-ray diffraction

UV-vis-NIR

Ultraviolet-Visible-Near Infrared

RNS

Reactive nitrogen species

RBCs

Red blood cells

WBC

White blood cell

PLT

Platelet

TB

Total bilirubin

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

CREA

Creatinine

UA

Urea

PCR

Polymerase chain reaction

ANOVA

One-way analysis of variance

WB

Western Blot

α-SMA

α-smooth muscle actin

CYR61

Cysteine rich protein 61

CTGF

Connective tissue growth factor

MAPK

Mitogen-activated protein kinase

Na₂SeO₃

Sodium selenite

iNOS

Inducible nitric oxide synthase

IL-6

Interleukin-6

CCL2

C-C motif chemokine ligand 2

ARG1

Arginase 1

CD206

Cluster of differentiation 206

CCL17

C-C motif chemokine ligand 17

IGF-1

Insulin-like growth factor 1

JNK

c-Jun N-terminal kinase

ERK1/2

Extracellular signal-regulated kinase 1/2

SGF

Simulated gastric fluid

SIF

Simulated intestinal fluid

LYM

Lymphocytes

MONO

Monocytes

GRAN

Neutrophils

AMY

Amylase

IL-1β

Interleukin-1β

PCA

Principal Component Analysis

DEGs

Differentially expressed genes

GO

Gene Ontology

BP

Biological Process

MF

Molecular Function

CC

Cellular Component

KEGG

Kyoto Encyclopedia of Genes and Genomes

Adonis

Non-parametric multivariate analysis of variance

LEfSe

Linear discriminant analysis Effect Size

Author contributions

Y.L., M.J., and Q.J. performed the experiment and wrote the manuscript. X.Z., Y.X., and Y.Y. analyzed the data. S.W., L.H., and J.Z. designed and supervised the research. All authors have reviewed and approved the manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (8227033745 and 8257035413) and Shanghai Rising-Star Program (23QA1412000).

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.