Mesenchymal stem cell-derived apoptotic extracellular vesicles loaded with Prussian blue nanoparticles attenuate severe acute pancreatitis via neutrophil extracellular traps resolution and acinar-ductal metaplasia promotion Authors: Wenjie Xu, Wenwu Cai, Junyong Wu, Xuyang Hou, Xiaoyan Qi, Zuxing Wei, Yimiao Cheng, Yanwen Zheng, Beibei Cui, Jun He Published: 2025 ABSTRACT AbstractNowadays, severe acute pancreatitis (SAP) remains a critical clinical disease with a high mortality rate. Current treatments are mainly supportive, lacking specific therapies targeting the core pathological mechanisms including neutrophil extracellular traps (NETs)-mediated systemic inflammation and progressive acinar cell necrosis, which may lead to many complications. Therefore, in this study, we developed a novel dual-targeting nanodrug delivery system named PBs@CCR2-ApopEVs, combining Prussian blue nanoparticles (PBs) and CCR2-overexpressing mesenchymal stem cell-derived apoptotic extracellular vesicles (ApopEVs). Through the specific binding of CCR2 with numerous CCL2 released by neutrophils in SAP, the PBs@CCR2-ApopEVs can be actively recruited to the pancreatic inflammation site. Then PBs effectively inhibit neutrophil extracellular trap (NETs) formation by suppressing myeloperoxidase and elastase expression, as well as Gasdermin D pathway activation, thereby reducing oxidative stress and inflammatory responses. Simultaneously, we confirmed that MSCs-ApopEVs could activate the p-STAT3/SOX4 pathway to promote acinar-to-ductal metaplasia (ADM) process, enhancing the self-protection ability of acinar cells and reducing necrosis. This dual-approach strategy—targeting anti-inflammatory effects externally while reinforcing cytoprotective mechanisms internally—stands out as a novel therapeutic avenue for SAP management, providing a new approach for SAP treatment and significantly improving therapeutic efficacy. Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12951-025-03958-y.Keywords: Severe acute pancreatitis, Neutrophil extracellular traps, Acinar-ductal metaplasia, Mesenchymal stem cell, Prussian blue nanoparticle Introduction Severe acute pancreatitis (SAP) remains a significant global health challenge. Despite advances in modern intensive care technology, SAP still has extremely high morbidity and mortality rates, placing a heavy burden on patients and society [1]. SAP is a critical disease characterized by acinar cell necrosis, systemic inflammatory storm, and multiple organ failure, with a mortality rate as high as 20%−30% [2]. Currently, clinical treatment mainly relies on supportive therapy, including fasting, intravenous fluid therapy, anti-microbial therapy, anti-inflammation therapy and usage of painkillers, etc. However, current strategy against SAP lacks specific treatments targeting the core pathological mechanisms, thereby cannot block the two major vicious cycles: (1) Over-activated neutrophils and macrophages releases pro-inflammatory mediators, which may cause a cytokine storm [3]. In particular, neutrophil extracellular traps (NETs) can promote pancreatic tissue oxidative stress disorder, drive pancreatic and distal organ damage, and continuously recruit more inflammation-related cells to participate in the SAP process [3, 4]; (2) Acinar cell homeostasis collapse leads to persistent necrosis, releasing damage-associated molecular patterns (DAMPs), further amplifying the inflammatory cascade [5]. These pathological processes collectively lead to severe pancreatic tissue damage and systemic complications. Therefore, developing a dual strategy that simultaneously inhibiting the inflammatory storm and protecting acinar cell homeostasis is crucial to breaking through the treatment bottleneck of SAP [6]. Notably, using a single drug can no longer meet the treatment needs [6, 7]. In recent years, the combination of nanomedicines and cell-free carriers has provided new ideas for SAP treatment, but existing strategies still have significant limitations. For example, although synthetic nanocarriers (such as liposomes) can deliver antioxidants (such as Prussian blue nanoparticles, PBs), their targeting relies on the passive EPR effect, making it difficult to effectively accumulate in the early stage of SAP when vascular leakage has not yet formed [8, 9]. On the other hand, natural carriers such as exosomes have small size, low drug loading efficiency, and inability to be mass-produced, limiting their clinical application [10, 11]. Herein, there is an urgent need to develop new biomimetic nanomedicine delivery carriers to solve this problem [12]. As the active ingredient in Radiogardase (approved by the FDA in 2003), Prussian blue (PB) has been a highly sought-after candidate drug in many nanomedicine formulations, with multiple functions such as nanozyme-like activity, antioxidant, photothermal and drug delivery characteristics, and good biocompatibility [13, 14]. However, there are few studies on whether this multifunctional PBs can be used in SAP to exert anti-inflammatory and antioxidant effects. In this study, we confirmed that PBs could not only alleviate oxidative stress disorder of SAP, but also effectively reduced the expression of key components in NETs, myeloperoxidase (MPO) and elastase during the SAP process. Meanwhile, PBs significantly inhibited the activation of the Gasdermin D (GSDMD) pathway, which is a key execution molecule in the NETs formation process. These evidences further confirmed that PBs could inhibit NETs formation and release through multiple pathways, alleviating the strong inflammatory response. However, this “external” anti-inflammatory approach alone fails to maintain acinar cell homeostasis, which is insufficient for SAP treatment [15]. Considering the strong immunomodulatory and tissue repair capabilities of mesenchymal stem cells (MSCs), they have shown great potential in the treatment of various inflammatory diseases such as rheumatoid arthritis (RA) and inflammatory bowel disease (IBD) [16, 17]. However, traditional MSC cell therapy faces challenges such as low cell survival rate, insufficient targeting, and potential safety risks [18]. According to multiple studies, one of the core factors in MSCs applications is their apoptosis in vivo. The released apoptotic extracellular vesicles (ApopEVs) have significant anti-inflammatory and antioxidant capabilities, which enlightened us that ApopEVs may have great application potential in SAP [19–21]. Subsequently, our further innovative research found that MSCs-derived ApopEVs can activate the p-STAT3/SOX4 pathway to promote acinar-to-ductal metaplasia (ADM), promote its self-protection, reduce acinar cell necrosis, enhance its resistance to oxidative stress, and avoid further deterioration of SAP. This protective mechanism essentially creating an “internal” safeguard against environmental stresses. Besides, ApopEVs have good biocompatibility and are easy to prepare, making them a highly potential cell-free carrier [21]. In order to further improve the targeting of ApopEVs, we innovatively engineered MSCs to highly express the C-C chemokine receptor type 2 (CCR2). Its ligand CCL2 is highly expressed in the SAP inflammatory microenvironment, providing a natural target for ApopEVs to target neutrophils in the inflammatory site [22]. Based on these theories, in this study we combined PBs with CCR2-overexpressing MSC-derived ApopEVs to construct a novel SAP synergistic treatment strategy. We hypothesized that CCR2-targeted ApopEVs loaded with PBs (PBs@CCR2-ApopEVs) could actively target the inflammatory site of SAP (especially neutrophils) through the CCR2/CCL2 axis. Then ApopEVs activate the “internal” defense mechanism of acinar cells, enhancing the self-protection ability of pancreatic acinar tissue. The delivered PBs simultaneously inhibited neutrophil NETs formation, synergistically inhibiting the pathological progression of SAP. This strategy uniquely integrates two critical therapeutic approaches: pharmacologically inhibiting neutrophil extracellular traps (NETs)-associated cascade inflammation and leveraging extracellular vesicles to promote acinar-to-ductal metaplasia (ADM) for tissue repair. Crucially, their targeted delivery mechanism significantly enhances both therapeutic efficiency and overall safety, offering a compelling new avenue for managing SAP. The illustration is shown in Fig. 1. Illustration of the preparation of PBs@CCR2-ApopEVs and its anti-SAP mechanism. (A) Preparation of PBs@CCR2-ApopEVs. (B) Mechanisms of PBs@CCR2-ApopEVs against SAP. PBs@CCR2-ApopEVs was recruited by the numerous CCL2 released by neutrophils in SAP site, then being uptake by relevant cells. The PBs engulfed by neutrophils could inhibited the major component of NETs (MPO and ELANE) and downregulated N-GSDMD signal pathway to suppress NETs formation. This process blocked the NETs-mediated inflammation storm from the perspective of external factors. Meanwhile, ApopEVs promoted ADM occurrence to protect acinar cells by activating p-STAT3/SOX4 signal pathway. And this process blocked the persistent necrosis of acinar cells from the perspective of internal factors. Therefore, this comongbination strategy finally resulted in the alleviation of SAP Materials and methods Materials Caerulein was purchased from Selleck, Japan (FL6934). Phorbol myristate acetate (PMA) was purchased from Selleck, Japan (S7791). LDC7559 was purchased from Selleck, Japan (S9622). FeCl2, K3Fe(CN)6, PVP were purchased from Beijing Solarbio Science & Technology Co., LTD (China). Cell culture media, including RPMI 1640 and Dulbecco’s modified Eagle’s minimum essential medium (DMEM), were sourced from BDBIO (Hangzhou, China). Fetal bovine serum (cat.164210) was purchased from Wuhan Pricella Biotechnology Co., Ltd. (China). For essential supplements and antibiotics, we used penicillin-streptomycin solution (cat. C100C5) and cellsaving medium (cat. C40100), both supplied by NCM biotech (Suzhou, China). Mounting Medium, antifading (with DAPI, cat. S2110) came from Beijing Solarbio Science & Technology Co., LTD (China). Yeasen Biotechnology Co., Ltd. (Shanghai, China) provided DiR dye (cat. 40757ES25), Dil dye (cat. 40718ES60), iFluor™ 488 phalloidin (cat. 40736ES75), iFluor™ 647 phalloidin (cat. 40762ES75), the Annexin V-FITC/PI Assay Kit (cat. 40302ES60), and the Reactive Oxygen Species Assay Kit (cat. 50101ES01). Laboratory consumables and plasticware were sourced from various vendors. NEST Biotechnology (Wuxi, China) supplied all cell culture plates and dishes, ultrafiltration tubes, centrifuge tubes, transwell plates and Matrigel (cat. 211212). We also utilized 0.22 μm filters (cat. BS-PES-22) and 70 μm cell strainers (cat. BS-70-CS) from Biosharp Life Sciences (Hefei, China). Finally, confocal culture dishes (cat. 1050001) and cell cryopreservation tubes (cat. 4021000) were purchased from Zhejiang Saining Biotechnology Co., Ltd (China). Cell culture and animals The cells used in the manuscript, neutrophils and acinar cells were both isolated from C57 BL/6J mice as indicated in the following relative sections. C57BL/6J mice were purchased from Charles River (Beijing, China). This study followed the ethical standards for animal experiments. All experiments were performed after obtaining approval (No.20231282) from the Ethics Committee of the Second Xiangya Hospital of Central South University, with minimal animal suffering. Adipose-derived mesenchymal stem cells isolation Adipose-derived mesenchymal stem cells (ADSCs), one of the MSCs were isolated according to the reference [23]. Briefly, ADSCs were harvested from adipose tissue located in the lateral epididymis of male mice. The initial processing was carried out in DMEM culture medium. Following harvest, the tissue underwent mechanical separation to remove subjacent adipose tissue. The resulting dermal adipose sheets were then collected and immediately immersed in phosphate buffered saline (PBS) supplemented with 5% penicillin/streptomycin (P/S). Next, the collected tissue was carefully minced and transferred to a PBS solution containing 0.075% collagenase Type I and 2% P/S for enzymatic digestion. After a 30-minute digestion period, 10 mL of DMEM medium supplemented with 20% FBS and 1% P/S was added to neutralize the collagenase Type I activity. The sample was then centrifuged at 2000 rpm for 5 min. The resulting cell pellet was re-suspended in 1.0 mL of ice-cold red blood cell lysis buffer (Sigma–Aldrich, USA) and incubated for 10 min to lyse red blood cells. The lysis reaction was subsequently neutralized by adding 20 mL of DMEM medium containing 20% FBS and 1% P/S. After neutralization, the mixture was centrifuged again. The resulting pellet was gently re-dispersed in 3 mL of culture medium before being filtered through a 70 μm pore size filter. The filtered cells were collected and transferred to a 60 × 15 mm cell culture dish, where they were incubated at 37 °C for 72 h for initial attachment and expansion. Crucially, prior to their use in subsequent in vitro and in vivo studies, the isolated ADSCs were characterized by confirming the expression levels of key surface markers: CD73, CD90, and CD105. PBs@CCR2-ApopEVs Preparation and characterization First, mesenchymal stem cells (MSCs) were genetically modified via lentiviral transfection to achieve stable overexpression of CCR2 or CCR2-GFP. Transfected cells underwent selection using puromycin at a concentration of 7 µg/ml. Once the transfection efficiency reached an optimal 70–80%, confirmed by both Western blot and flow cytometry, cells were prepared for conditioned medium collection. To ensure minimal contamination from exogenous extracellular vesicles (EVs), both the basal culture medium and fetal bovine serum (FBS) were rigorously depleted of EVs through ultracentrifugation as previously described. Following this preparation, the culture medium was replaced with the EV-depleted medium, and the conditioned medium was collected after a 48-hour incubation period. To obtain the CCR2-ApopEVs, CCR2-overexpressed MSCs were treated with Staurosporine (STS, 250 nM) for 24 h to induce apoptosis. Then the cell supernatant was centrifuged at 500 g for 15 min to remove dead cells and debris, then the supernatant was centrifuged at 10,000 g for 30 min (4 ℃) to collect the precipitate and resuspend in PBS. Flow cytometry was used to assess the apoptotic rate of ADSCs. To construct PBs@CCR2-ApopEVs, Prussian blue nanoparticles (PBs) was synthesized according to the reference [24]. Then PBs were co-extruded with CCR2-ApopEVs to prepare PBs@CCR2-ApopEVs. Transmission electron microscope (TEM) was applied to observe the morphology, while Nanoparticle tracking analysis (NTA) and Dynamic light scatter (DLS) was used to evaluate the particle size and zeta potential of PBs@CCR2-ApopEVs. Immunohistochemistry and immunofluorescent assay For tissue immunohistochemistry, collected fresh animal tissues were fixed in 4% paraformaldehyde, paraffin-embedded, and cut into 4 μm white slices. The paraffin sections were baked in an oven at 65 °C for 60 min (during this time, 200 ml of xylene I, xylene II, anhydrous ethanol I, anhydrous ethanol II, 95% ethanol, and 75% ethanol were added to the immunohistochemistry bath in the above order and set aside). The paraffin sections were then quickly removed and placed into the corresponding immunohistochemical baths in the above order for 10 min, 10 min, 3 min, 3 min, 3 min, 3 min, 3 min, and the deparaffinized sections were rinsed with water for 3 min and then placed into boiling EDTA-Tris repair solution for 20 min for zymogen repair. After the repair was completed, the sections were naturally cooled. Next, add peroxidase blocker from rabbit specific HRP/DAB Detection IHC kit (Abcam, USA) dropwise on the sections for 10 min, and then rinse the sections with PBS for 3 times, each time for 3 min, then add primary antibody dropwise on the sections, put them into a wet box, and refrigerate them at 4℃ overnight. The next day, the wet box was taken out, the sections were washed 3 times with PBS, then rabbit HRP secondary antibody from the immunohistochemistry kit mentioned above was added dropwise, and incubated for 15 min. the sections were then washed 3 times with PBS, stained dropwise with DAB from the immunohistochemistry kit, and then rinsed under running water for 10 min. hematoxylin was then added dropwise, and incubated for 10 min, and then rinsed under running water for 10 min. the sections were set into the immunohistochemistry tank, and were dehydrated according to the The sections were placed in the immunohistochemistry tanks and dehydrated in the reverse order of dewaxing, 2 min per tank. After final dehydration, a drop of neutral resin was added to the slides, and the sections were sealed using a coverslip. Images were captured using a Zeiss Axioscope 5 microscope. Primary antibodies involved in immunohistochemistry were: cleaved casp3 (Absin, abs132005,1:100), and Ki-67 (CST, 34330, 1:200). For tissue immunofluorescence, the steps prior to zymogen repair were the same as for immunohistochemistry. After zymogen repair, the samples were blocked with sheep or donkey serum for more than 30 min, followed by the addition of the corresponding primary antibody, and then placed in the cold room at 4℃ overnight. The next day, the sections were washed with PBS, the corresponding secondary antibody was added to the sections, and incubated for 60 min, protected from light, and the sections were washed again and sealed using the Vector® TrueVIEW® Autofluorescence Quenching Kit with DAPI (Vectorlabs, SP-8500-15), i.e., TrueVIEW reagents A, B, and C were mixed in a 1:1 ratio, and the samples were then incubated for 30 min, B and C were mixed 1:1:1 and dropped on the sections and incubated for 4 min, then washed with PBS and finally sealed with VECTASHIELD Vibrance® Antifade Mounting Medium with DAPI. Images were captured using a Zeiss Axioscope 5 microscope. Primary antibodies involved in immunofluorescence were: CitHistone3 (Abcam, ab281584,1: 200), MPO (Proteintech, 66177-1-Ig, 1:400), Elane (Abclonal, A13015,1:100), N-GSDMD (Santa, sc-393656, 1:100), Krt19 (Abclonal, A25546, 1:200), Amylase (Santa, sc-46657,1:100), SOX4 (Absin, abs120837,1:100), Ly6G(BioLegand,127625,1:50), and CD68 (CST,97778s,1:200). The second resistance involved is there: Goat anti-rabbit lgG 594 (Abcam, ab150080,1:1000), Goat anti-rabbit IgG 488(Abcam, ab150077,1:1000), Goat anti-mouse IgG 488(Abcam, ab150157,1:1000) and Goat anti-mouse IgG 594(Abcam, ab150113,1:1000). Scanning electron microscopy The culture medium was aspirated, and the cells were immediately covered with a pre-cooled primary fixative solution. The primary fixative was removed, and the cells were gently rinsed with PBS to remove excess fixative. The cells were treated with a 1% osmium tetroxide solution for 1–2 h at 4 °C. They were then processed through drying, dehydration, and conductivity before being observed under Zeiss CSEM. Determination of DsDNA in supernatant Follow the instructions for the dsDNA test kit (MKBio, MF0781-1-ML), take 100 µl of cell supernatant and add it to a 96-well plate, then add 100ul of 1×PicoGreen, incubate for 5 min away from light, and then detect the concentration value in a fluorometer. ROS levels evaluation by DCFH-DA DCFH-DA (MedChemExpress, D6883, USA) was diluted in PBS buffer (Servicebio, G4250, China) at a ratio of 1:1000 to a final concentration of 10 µmol/L. For suspension cells, the treated cells were collected by centrifugation and resuspended in the prepared DCFH-DA working solution. The cells were then incubated at 37 °C for 20 min, with suspension cells being gently inverted every 5 min to ensure thorough contact between the probe and cells. After incubation, the cells were washed three times with PBS to completely remove uninternalized probes. The treated cells could be measured using a SpectraMax i3x microplate reader (Molecular Devices, USA) analysis. The relative ROS level was calculated as the fold change by normalizing the mean fluorescence intensity (MFI) of each sample to the mean MFI of the control group in the first independent experiment. Statistical analysis was performed using GraphPad Prism software. Differences among multiple groups were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test for multiple comparisons. Data are presented as mean ± standard deviation (SD). Cellular RNA sequencing RNA sequencing (RNA-seq) was performed by Novogene Co., Ltd. (China). Total RNA was extracted from each treatment group following standard protocols, and RNA concentration and purity were assessed using an Agilent 2100 Bioanalyzer. The cDNA fragments were purified using the AM Pure XP system, and the first and second cDNA strands were synthesized using M-MuLV Reverse Transcriptase and Phusion High-Fidelity DNA Polymerase, respectively. The constructed libraries were sequenced on the Illumina Nova Seq 6000 platform. After quality control (removing low-quality and adapter-contaminated reads), the high-quality clean reads were aligned to the reference genome. Feature Counts was employed to quantify reads mapped to each gene, and gene expression levels were normalized using FPKM (Fragments Per Kilobase of transcript per Million mapped reads). Differential expression analysis was conducted using DESeq2, with adjusted p-value < 0.05 considered statistically significant. qRT-PCR Total RNA was extracted from tissues and cells following our previously published protocol. RNA concentration (A260/A280 ratio) and purity (A260/A230 ratio) were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher, USA). cDNA was synthesized with a High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher, 4374966, USA). The qPCR reaction was performed using 2× Universal SYBR Green Fast qPCR Mix (Abclonal, RK21203, China) on a LightCycler 96 Real-Time PCR System (Roche, USA). All experiments included technical replicates (n ≥ 3) to ensure data reliability. The PCR primers were custom-synthesized by Sangon Biotech Co., Ltd (Shanghai, China) and the sequences were as follows: MPO, forward: CGTGTCAAGTGGCTGTGCCTAT. MPO, reverse: AACCAGCGTACAAAGGCACGGT. ELANE, forward: CAGGAACTTCGTCATGTCAGCAG. ELANE reverse: AGCCATTCTCGAAGATCCGCTG. ACTB forward: CATTGCTGACAGGATGCAGAAGG. ACTB reverse: TGCTGGAAGGTGGACAGTGAGG. Isolation of primary acinar cells and culture of ADM C57BL/6J mice were euthanized by CO₂ asphyxiation. Pancreas was aseptically dissected and transferred to pre-chilled sterile tubes. Tissues were minced into 2 mm³ fragments using sterile scissors and digested with 1 mL mouse pancreas dissociation solution (Absin, abs50096-10T, China) at 37 °C with 180 rpm shaking for 15 min. The homogenate was filtered through a 70 μm cell strainer, and the strainer was rinsed with complete medium. The filtrate was collected in a 50 mL tube, followed by centrifugation at 400 g for 3 min. Finally, the pellet was washed three times with PBS. Purified acinar cells were resuspended in Waymouth’s medium (Gibco™, 11220035, USA) supplemented with 2.5% FBS (Vivacell, C04001-500, China), 1% penicillin-streptomycin (Gibco™, 15140-122, USA), 0.25 mg/mL trypsin inhibitor (Solarbio, T8031, China), and 25 ng/mL recombinant human EGF (MedChemExpress, HY-P7067, USA). Cell suspension was mixed with Matrigel (Corning, 354234, USA) (1:2 ratio), incubated at room temperature for 10 min, then 40 µL aliquots were plated in 24-well plates. After 20 min polymerization at 37 °C, 500 µL Waymouth’s medium was gently added along the well edge. The experimental groups included a control group and a TGF-α group with medium containing 50 ng/mL TGF-α (MedChemExpress, HY-P77857, USA). Cultures were maintained for 5 days in a 37 °C, 5% CO₂ incubator. Morphological changes during acinar-to-ductal metaplasia (ADM) progression were documented by phase-contrast microscopy (Nikon Eclipse Ti2, Japan) on days 1, 3, and 5. SAP mice model establishment As stated in our previous article, mice were given 12 injections of Caerulein (Selleck, FL6934) (50 µg/kg) intraperitoneally at 1-hour intervals after 12 h of fasting, and mice were executed 24 h after the first injection, and blood, pancreas, and lungs were collected. All animals were purchased from Charles River (Beijing, China). This study followed the ethical standards for animal experiments. All experiments were performed after obtaining approval from the Ethics Committee of the Second Xiangya Hospital of Central South University, with minimal animal suffering. Isolation of mouse bone Marrow-Derived neutrophils and induction of NETs C57BL/6J mice were euthanized by CO₂ asphyxiation. The femurs and tibias were excised after hair removal and skin disinfection. Surrounding muscle tissues were carefully removed using sterile surgical scissors, and the bones were immediately placed in ice-cold PBS. The knee joints were dissected to expose the bone marrow cavities. Marrow cells were flushed out using a sterile syringe filled with PBS, followed by gentle pipetting to dissociate cell clusters. The resulting cell suspension was filtered through a 40 μm cell strainer and collected into a 50 mL conical tube. The filtrate was centrifuged at 1600 rpm for 5 min at 4 °C. The pellet was resuspended in 2 mL ACK lysis buffer (Gibco™, A1049201, USA) for 2 min at room temperature to lyse red blood cells. The reaction was terminated by adding 5 mL PBS, followed by centrifugation at 1600 rpm for 5 min. The cell pellet was resuspended in 1 mL PBS and carefully layered onto a discontinuous Percoll (Sigma, P4937, USA) gradient (1.5 mL each of 75%, 65%, and 55% Percoll solutions in a 15 mL conical tube). After centrifugation at 1600 rpm for 30 min at 25 °C, the neutrophil-enriched fraction at the 65%−75% Percoll interface was collected. The harvested neutrophils were washed with 6 mL PBS and centrifuged at 1600 rpm for 5 min. The purified neutrophils were resuspended in complete RPMI 1640 medium (supplemented with 10% FBS (Vivacell, C04001-500, China), 1% penicillin-streptomycin (Gibco™, 15140-122, USA). Cells were seeded onto culture plates and stimulated with 100 nM phorbol 12-myristate 13-acetate (PMA) (Selleck, S7791, Japan) for 12 h at 37 °C in a humidified 5% CO₂ incubator to induce NETs formation. Establishment of a co-culture system of acinar cells and neutrophils Primary acinar cells and neutrophils were isolated using the protocols. The acinar cells were first stimulated with TGF-α (MedChemExpress, HY-P77857, USA) for 5 days, then treated by PBs or the formulations for 12 h, followed by co-cultured with neutrophils stimulated by PMA in a Transwell system (separated by a sterile nylon membrane) at 37 °C for 6 h. Following co-culture, both the treated cells and supernatant could be harvested for subsequent analysis of NETs and ADM formation. Neutrophil SYTOX green staining Neutrophils were isolated using the aforementioned method and seeded onto coverslips in 24-well plates. After drug treatment, the cells were gently washed once with PBS, fixed with 4% paraformaldehyde (Beyotime, P0099, China) at room temperature for 10 min, and then washed again with PBS to remove the fixative. Next, they were stained with 1 µmol/L SYTOX Green nucleic acid stain (Invitrogen, S7020, USA) and incubated in the dark at room temperature for 30 min. After an additional PBS wash to remove excess stain, the cells were air-dried and mounted with DAPI (Abcam, ab104139, UK). Image acquisition was performed using either a ZEN fluorescence microscope (ZEN, Germany) or a Leica TCS SP8 confocal microscope system (Leica Microsystems, Germany), followed by quantitative analysis of the acquired images. H&E staining of pancreas and lung tissues Lung and pancreatic tissues from mice were collected and immediately fixed in 4% paraformaldehyde solution, and after fixation, they were paraffin-embedded and cut into 4 μm sections. After dewaxing and dehydration of the paraffin sections, H&E staining was performed, and two pathologists analyzed each sample for pancreatitis severity and score. Analysis of amylase and lipase activity in plasma Before taking blood from mice, 20 µl of 0.5 M EDTA was added to the centrifuge tube, and after adding blood, the plasma was shaken well and centrifuged at 1000 g for 10 min, and transferred to a new centrifuge tube. Lipase and amylase activities in plasma were measured by a clinical analysis system as described in our previous studies. TUNEL staining of pancreas Paraffin sections cut into 4 μm were deparaffinised by passing through xylene and anhydrous ethanol, followed by following the instructions of the TUNEL staining kit (Beyotime, China), proteinase K was first diluted with PBS in the ratio of 1:1000, and then dripped on the sections and incubated for 20 min, followed by washing the sections well with PBS. Then tdt enzyme and fluorescence chromogenic solution were prepared in the ratio of 1:50, dropped on the sections and incubated at 37 °C for 60 min under light protection, and then the sections were washed well with PBS, and then the sections were blocked using the Vector® TrueVIEW® Autofluorescence Quenching Kit with DAPI (Vectorlabs, SP-8500-15), which means that firstly the TrueVIEW Reagents A, B, and C were mixed 1:1:1 and dropped on the sections and incubated for 4 min, then washed with PBS and finally sealed with VECTASHIELD Vibrance® Antifade Mounting Medium with DAPI. Images were captured using a Zeiss Axioscope 5 microscope. Pancreas targeting ability evaluation of PBs@CCR2-ApopEVs Six to eight weeks old C57BL/6J male mice were fasted for 12 h and randomly divided into three groups: the Free DiR group, the DiR-PBs@ApopEVs group, and the DiR-PBs@CCR2-ApopEVs group. Acute pancreatitis was induced by intraperitoneal injection of caerulein (cat. S9690, Selleck, Japan) at a dose of 50 µg/kg every 1 h for a total of 12 injections. Following the 6th caerulein injection, mice in each group received an intravenous injection of their respective formulation. Eight hours after the intravenous drug injection (upon completion of all caerulein administrations), mice were monitored using an In Vivo Imaging System (IVIS) to observe fluorescent intensity. Subsequently, the pancreas and other major organs (including heart, liver, spleen, lung, kidney, and gut) were excised and imaged via IVIS to determine the biodistribution of the formulations throughout the body. Statistical analysis Data are presented as mean ± standard deviation (SD). Statistical analysis and visualization were performed using GraphPad Prism software (version 9.5.0). For comparisons between two groups, a standard t-test was applied. When comparing three or more groups, we utilized one-way or two-way analysis of variance (ANOVA) followed by Tukey’s multiple-comparison test. Importantly, the number of experimental replicates was clearly indicated in each figure caption, which also included a detailed description of the specific statistical methods employed for that dataset. Results and discussions Acute pancreatitis (AP) represents an acute inflammatory disorder of significant global health burden, frequently leading to severe complications and considerable mortality. Despite advancements in understanding its complex pathophysiology, therapeutic interventions for SAP currently remain largely supportive. Management predominantly relies on comprehensive conservative measures such as fluid resuscitation, pain management, and nutritional support, aiming to mitigate symptoms and prevent secondary organ damage. Crucially, a critical gap persists in the clinical practice: the absence of mechanism-targeted pharmaceutical agents that can fundamentally alter the progression of AP, rather than merely addressing its symptomatic manifestations. This underscores the urgent need for novel therapeutic strategies rooted in a deeper comprehension of disease-driving molecular pathways. PBs inhibit neutrophil NETs formation In pursuit of such mechanism-targeted interventions, various nanotherapeutics have garnered significant attention, among which Prussian blue nanoparticles (PBs) stand out. As one of the promising anti-inflammatory agents nowadays, PBs were comprehensively studied in inflammation-related diseases in recent years, and FDA has approved Radiogardase for the treatment of individuals contaminated with radioactive cesium or non-radioactive thallium [25]. Recognized for their remarkable intrinsic enzyme-like activities, particularly their broad-spectrum reactive oxygen species (ROS) scavenging capabilities (mimicking catalase, peroxidase, and superoxide dismutase), PBs have emerged as highly potent antioxidant and anti-inflammatory agents. Consequently, their therapeutic potential has been extensively explored and demonstrated in preclinical models across a wide array of inflammatory and oxidative stress-related diseases, including neurodegenerative disorders, cardiovascular diseases, kidney injury, and various cancers [26]. Despite this broad success and their promising anti-inflammatory properties, investigations specifically addressing the therapeutic application of PBs in the context of severe acute pancreatitis (SAP) remain surprisingly scarce, leaving a critical knowledge gap in the field. However, whether PBs could alleviate SAP in clinic remained unknown. According to reports, neutrophils play an important role in SAP inflammation [3]. Neutrophils were activated after being stimulated by inflammatory factors during SAP stages. The activation signal was transmitted through a series of intracellular signaling pathways, including the production of ROS. Then the chromatin disassembled, with DNA transitioning from a compact state to a loose state, followed by cell membrane rupture, thereby releasing the nuclear and cytoplasmic contents (such as MPO, ELANE) into the extracellular space, forming a reticular structure. Finally, NETs recruited more neutrophils by releasing pro-inflammatory factors, forming an inflammatory positive feedback loop, and exacerbating systemic inflammation [27]. In our preliminary experiments, we surprisingly found that PBs could inhibit the inflammation progress in SAP milieu via suppressing neutrophil extracellular traps (NETs) formation. To further explore behind mechanisms, we conducted various experiments to determine the changes of classic biomarkers in NETs formation. Based on the main components of NETs, we first applied phorbol myristate acetate (PMA) to simulate neutrophils activation in SAP progress. PMA is a potent neutrophil activator that induces neutrophils to produce ROS, release NETs, and alter metabolic status by activating the protein kinase C (PKC) signaling pathway. The immunofluorescent images in Fig. 2A, PMA significantly increased the Citrullinated histone H3 (CitH3) signal in neutrophils. However, after intervention with PBs at different concentrations (50 µg/mL and 150 µg/mL), the decreased signal of extracellular CitH3 was observed which exhibited concentration-dependent manner. To further confirm this phenomenon, we applied SEM for visualization. As shown in Fig. 2B, neutrophils forming NETs displayed a fibrous reticular structure covering the cell surface after PMA treatment. But the PBs alleviated this process. As the concentration of PBs increased, this phenomenon also gradually diminished, even demonstrating no significant difference compared to the control group. Considering the main framework of NETs is composed of DNA which originates from the nuclei of neutrophils, we used SYTOX Green to label DNA in NETs. Also, PMA could induce extracellular DNA accumulation while PBs inhibited this process with concentration-dependent pattern according to the IF images in Fig. 2C, demonstrating that PBs can effectively inhibit the release of extracellular DNA by neutrophils induced by PMA (white arrows), which is one of the direct evidences for inhibiting the formation of NETs. The dsDNA detected by PicoGreen further verified this conclusion (Fig. 2D). The oxidative press induced by NETs in SAP was assessed after various interventions, whose results demonstrated a promising protection effect of PBs in ROS alleviation as shown in Fig. 2E. The decreased ROS level in PBs groups indicated that less damage and inflammation activation would occur in SAP under PBs protection. In order to explore the genome signal pathway activation, we treated neutrophils with PBs 3 h before PMA administration to analyze the transcriptome change (Fig. 2F and G). The heatmap shows the differentially expressed genes between the PMA and PMA + PBs treatment groups. And GSEA was performed to compare the key signal pathway related to NETs formation. The results showed that after PMA treatment, pathways related to NETs formation including inflammatory response, cytokine signaling, ROS metabolism, and specific NETs formation regulatory pathways, etc. were significantly enriched, indicating their activity was inhibited by PBs significantly. Moreover, the other main components of NETs (MPO and ELANE) were detected via IF assay. Both biomarkers were observed to increase after PMA intervention while PBs significantly suppressed their expression (Fig. 2H). Therefore, the decrease level of both dsDNA and major components like MPO and ELANE suggested that NETs was interrupted by PBs. Crucially, we sought to pinpoint the molecular target of PBs. Gasdermin D (GSDMD) is identified as the key executioner pore-forming protein in NETosis. We determined the GSDMD in neutrophils treated with different drugs. The results demonstrated PBs could effectively suppress GSDMD in activated neutrophils (Fig. 2I), and this efficacy also exhibited concentration-dependent manner. Our previous findings established that PBs effectively clear excessive cellular ROS. Initially, we speculated that this ROS clearance might paradoxically enhance GSDMD cleavage by boosting caspase-1 activity. To investigate the true impact of PBs on caspase 1 activity, we conducted cellular experiments. Interestingly, while PMA stimulation robustly activated caspase-1 in neutrophils, PBs strikingly suppressed this activation (Fig. 2J). To firmly establish ROS as the critical link between PBs and GSDMD regulation, we designed a specific intervention. We co-treated PB-pretreated neutrophils with 200 µM H₂O₂. Remarkably, this exogenous H₂O₂ addition almost entirely reversed the PB-mediated suppression of GSDMD-NT expression, underscoring the pivotal role of ROS in this pathway (Fig. 2K). Mechanism exploration of NETs inhibition by PBs. (A) Immunofluorescence assay of CitH3 in neutrophils received different treatments. (B) SEM images of NETs formation of neutrophils after PBs intervention. (C) Immunofluorescence assay of SYTOX in neutrophils after treatments. (D) Free dsDNA concentration analysis in cell supernatant. (E) ROS level evaluation of neutrophils after PBs interventions. (F-G) Transcriptome data analysis of PBs-treated neutrophils. (G) Immunofluorescence images of major components of NETs (including MPO and ELANE). (H) Detection of MPO and ELANE in neutrophils after PBs treatments. (I) Immunofluorescence images of N-GSDMD in neutrophils after PBs treatments. (J) Cleaved caspase 1 expression in neutrophils after various treatments. (K) GSDMD expression in neutrophils under different treatments after the addition of H2O2 and its statistical analysis. Data = Mean ± SD (n = 4). Statistical analysis was performed by GraphPad Prism (9.5.0). One-way ANOVA followed by the Tukey’s multiple-comparisons test was applied for comparison among multiple groups. *P < 0.05, **P < 0.01, ***P < 0.001 Collectively, our findings suggest that PBs inhibit NETs formation through a comprehensive “ROS-dependent cascade blockade.” Acting as nanozymes, PBs efficiently scavenge intracellular ROS, thereby cutting off the inflammatory signal at its source. Crucially, our data showed that PBs significantly downregulated the activation of GSDMD, acted as the essential pore-forming executioner in NETosis and its cleavage is driven by ROS. The suppression of GSDMD confirmed that PBs function by blocking the specific ROS-GSDMD signaling axis. This upstream blockade ultimately led to the observed reduction in downstream NETs components, including MPO and ELANE, resulting in the potent suppression of the inflammatory storm. Further, we set positive and negative control groups to verify the dual inhibition effect of PBs in NETs formation (Figure S1). Nigericin was used to promote NETs formation while LDC7559 exhibited contrast effect. Single nigericin boosted the dsDNA release but did not change CitH3 level, while LDC7559 mainly affect the MPO expression in NETs due to the dsDNA and CitH3 level in NETs could not further be decreased by LDC7559 though combined with PBs. Additionally, when PBs was administrated, the dsDNA, CitH3 and MPO levels were all inhibited, indicating the multiple suppression ability of PBs in NETs formation. However, due to the lack of neutrophils targeting ability of PBs, we found that modifying PBs with lipid bilayer membranes (such as liposomes) can significantly improve the uptake efficiency of PBs in neutrophils, thereby more effectively exerting their inhibitory effect on NETs. However, its single application can only suppress the inflammatory response from “the outside,” and cannot maintain the internal homeostasis of acini. MSCs-ApopEVs promote ADM formation through p-STAT3/SOX4 pathway activation Mesenchymal stem cells (MSCs) have emerged as highly promising therapeutic agents, particularly in regenerative medicine and inflammatory diseases, largely owing to their remarkable immunomodulatory and tissue-repairing capacities [28]. However, relying on traditional MSC cell therapy presents significant challenges, including issues with low cell survival rates, insufficient targeting to disease sites, and potential safety concerns. Interestingly, evidences now point to MSC-derived ApopEVs as surprisingly potent and perhaps even crucial mediators of these therapeutic actions [29, 30]. These nanoscale vesicles, naturally formed during programmed cell death, function as sophisticated delivery systems. They efficiently transport a diverse array of cargos including proteins, lipids, and various nucleic acids, from the parent MSCs directly into recipient cells, thereby eliciting downstream cellular responses. Indeed, recent research is increasingly demonstrating that MSC-ApopEVs actively participate in key processes such as suppressing inflammation, facilitating the clearance of cellular debris, and critically, promoting tissue repair. Motivated by this evolving understanding of their powerful role, we focused our efforts on these specific effectors, isolating apoptotic extracellular vesicles from mice primary adipose-derived mesenchymal stem cells (ADSCs) for our study. First, the differentiation potential of extracted ADSCs was evaluated. The osteogenic/adipogenic/chondrogenic ability of ADSCs was displayed in Figure S2. Then we induced cell death to isolate ApopEVs. The MSCs-derived ApopEVs were identified with multiple approaches including TEM, DLS and SDS-PAGE as shown in Figure S2. The particle size of CCR2-ApopEVs was around 209 nm, and TEM images also confirmed similar sizes. And ApopEVs consistently exhibited a characteristic spherical morphology. Western blot was performed to further verify the characteristic protein in ApopEVs to validate the existence of CCR2 protein. Next, we extracted primary mouse pancreatic cells and induced acinar cell damage by treating them with TGF-α (Fig. 3A). Two days later, we applied different concentration of ApopEVs on acinar cells and recorded morphological images at Day3 and Day5. The results showed that after TGF-α treatment, ADM occurred (green arrows). Also, with time passing, ApopEVs could effectively induce ADM in a concentration-dependent manner (Fig. 3A). It could be observed that ADM formed faster and more broadly after ApopEVs addition. And the statistical analysis also confirmed this phenomenon. Mechanisms exploration of MSCs-derived ApopEVs in enhancing ADM of acinar cells. (A-B) Photographs of primary acinar cells after different treatments in 5 days and ADM number analysis. (C-D) Immunohistochemistry images of KRT19 and amylase expression in primary acinar cells after different treatments and their analysis. (E) Immunohistochemistry images of SOX4 and p-STAT3 expression in acinar cells after ApopEVs interventions. (F-G) Relative expression level of p-STAT3 and SOX4 in acinar cell received different treatments. (H-I) Immunofluorescent images and quantitative analysis of KRT19, SOX4, Amy in pancreas of SAP mice models after various interventions. (J) ADM numbers analysis of pancreatic tissues in SAP mice models after treatments. Data = Mean ± SD (n = 4). Statistical analysis was performed by GraphPad Prism (9.5.0). T-test was used for comparison between the two groups. One-way ANOVA using the Tukey’s multiple-comparison test was applied for comparison among three groups. *P < 0.05, **P < 0.01, ***P < 0.001 KRT19 was commonly used to label pancreatic ductal cells while amylase was often used as acinar cell biomarkers. It is noted that during ADM induced by TGF-α, amylase expression decreased and KRT19 expression increased in the primary acinar cells. We determined their expression via immunohistochemistry assay in Day5 to assess the ADM indirectly (Fig. 3C and D). If ApopEVs promoted ADM, the ApopEVs-treatment group should show earlier and more extensive morphological changes, with higher KRT19 expression and lower Amylase expression differing from the control group. As expected, same results were observed in Fig. 3D, indicating an important protection effect of ApopEVs. According to reports, p-STAT3/SOX4 signal pathway activation was related to ADM and pancreas repairment. To explore the behind mechanism of ApopEVs-induced ADM, we performed IHC to detect SOX4 protein expression and phosphorylation activation level of STAT3 (p-STAT3) in the ADM model by IHC in vitro. As shown in Fig. 3E, TGF-α itself can induce increased expression of SOX4 and p-STAT3, and ApopEVs treatment can further significantly increase the expression levels of both. And high concentration of ApopEVs exhibited better effects (Fig. 3F and G). This suggested the ApopEVs promoted ADM via p-STAT3/SOX4 signal pathway. We then performed in vivo experiments to verify the promising ADM-induction effect of ApopEVs. MSC-derived ApopEVs were injected intravenously. After SAP model establishment, mice were sacrificed 24 h and 48 h later. H&E images showed ApopEVs intervention could significantly decrease pancreatic necrosis and promote ADM formation (Fig. 3H and I). Also, the signal of KRT19 increased by ApopEVs treatment. The observation that amylase expression does not show obvious changes in SAP despite significant inflammation may be attributed to multiple interrelated factors involving regulation, secretion, and auto-activation dynamics. Importantly, the fluorescent signal of SOX4 obviously elevated under ApopEVs intervention (Fig. 3J). All above efficacies exhibited time-dependent manner, implying the continuous repairing effect of ApopEVs. To further verify whether the p-STAT3/SOX4 pathway is causally required for ADM formation, we conducted an inhibition experiment using cryptotanshinone, a specific inhibitor of STAT3 phosphorylation. As shown in Figure S3, compared to the TGF-α group, the addition of the inhibitor significantly suppressed ADM formation and reduced the number of surviving cells, leading to increased cell death. This result directly confirmed that the activation of p-STAT3 is an indispensable driver for ADM-mediated cell protection. Combined with previous reports stating that STAT3 positively regulates SOX4 expression to support tissue homeostasis [31–33], these data robustly demonstrated that ApopEVs alleviate injury by activating the p-STAT3/SOX4 signaling axis. Collectively, ApopEVs treatment alleviated the development of SAP from within the organ by promoting ADM by activating p-STAT3/SOX4 signal pathway, reducing acinar cell necrosis, and enhancing their resistance to oxidative stress, thereby preventing further deterioration of the SAP condition. Herein, these results enlightened us to use ApopEVs to protect pancreas itself, and PBs inhibited extracellular inflammation could synergistically exert anti-SAP effect. CCL2-overexpressed MSCs construction and derived apopevs extraction Natural apoptotic body derived from MSCs did not own the specific targeting ability towards pancreas, herein, single ApopEVs could not effectively accumulated in pancreatic tissues whether or not PBs were encapsulated, this enlightened us to determine a proper targeting protein for combination. During SAP stages, multiple chemokines were released by neutrophils for inflammatory immunocytes recruitment and later cascade immunity. We conducted transcriptomics examination on neutrophils after PMA stimulation and performed GSEA (Gene Set Enrichment Analysis) analysis of the transcriptomic data. As shown in Figure S4A, the heatmap displayed the trends of 60 differential chemokines, including 26 that were significantly highly expressed. To identify candidate genes with greater metabolic relevance, we performed qPCR on neutrophils with and without LPS stimulation, analyzing the expression of commonly upregulated chemokines, including CXCL1, CXCL2, CXCL3, CCL2, and CXCL10. Among these, CCL2 and CXCL10 showed the highest differential expression (Figure S4B). Further, qPCR detection in primary neutrophils isolated from SAP and ctrl mice revealed a greater difference in CCL2 expression (Figure S4C). Therefore, we determined the CCL2 expression in SAP tissues by immunohistochemistry, whose results demonstrated the significant increase level of CCL2 in SAP-related neutrophils (Figure S4D and S4E). Therefore, this result provided a biological basis for utilizing CCR2/CCL2-mediated targeting strategies, indicating the presence of targeting signals in the inflammatory pancreatic microenvironment. Preparation and characterization of PBs@CCR2-ApopEVs As mentioned above, based on the exciting efficacy of PBs on NETs inhibition and MSC-derived ApopEVs on ADM process, we adopted a dual combination strategy of “internal” mitigation and “external” suppression. According to researches, nanoparticle loaded in phospholipid bilayer especially natural extracellular vesicles could achieve better cell uptake efficiency [34, 35], herein we designed using ApopEVs to wrap PBs (PBs@ApopEVs). However, the pancreatic targeting ability of ApopEVs was not sufficient due to lack of specificity. And it was determined that multiple inflammatory chemokines continuously released in SAP stages, particularly CCL2 released by neutrophils to recruit diverse inflammation cells as described in the previous chapter. Therefore, we designed a CCR2-overexpressed MSC strain, to express CCR2 highly on the cell membrane for CCL2 targeting and neutralizing. The successful CCR2 expression on MSCs was labeled by GFP and determined by flow cytometry and western blot assay as shown in Fig. 4A and B. Then CCR2-overexpressed MSC-derived ApopEVs were collected to encapsulate PBs for final formulation PBs@CCR2-ApopEV preparation. Cell apoptosis was evaluated via flow cytometry (Fig. 4C and D). And TEM images confirmed the successful preparation of MSCs-derived apoptotic bodies (Fig. 4E). According to the NTA result, the particle size of CCR2-ApopEVs was around 209 nm (Fig. 4F). Further, we used CCR2-ApopEVs to encapsulate PBs to prepare PBs@CCR2-ApopEVs via extrusion. Notably, the size of PBs@CCR2-ApopEVs was smaller than that of CCR2-ApopEVs according to the DLS and NTA results. We speculated that the observed size change resulted from the extrusion process, specifically its effect on enhancing the formulation’s homogeneity. DLS results showed that size of CCR2-ApopEVs did not significantly change after PBs co-incubation, suggesting the successful encapsulation of PBs (Fig. 4G and H). Notably, PBs@CCR2-ApopEVs exhibited a promising stability within 14 days. Also, the zeta potential of PBs@CCR2-ApopEVs was − 42.3 ± 1.43 mV (Fig. 4J and K). It may be attributed to the negative charge carried by CCR2-ApopEVs themselves. Subsequently, we performed TEM to observe the morphology of PBs@CCR2-ApopEVs, which exhibiting a homogenous sphere shape (Fig. 4L). A typical membrane structure could be observed outside the PBs. Therefore, we comprehensively evaluated the basic properties of PBs@CCR2-ApopEVs as a drug delivery system, laying the foundation for subsequent pharmacodynamic studies. Characterization of PBs@CCR2-ApopEVs. (A) Flow cytometry analysis of CCR2-GFP overexpressed ADSCs. (B) Western blot images of overexpressed CCR2 in engineered ADSCs. (C) Cell apoptosis evaluation via flow cytometry. (D) Rate of CCR2-GFP labeled apoptotic bodies in ADSCs after induction through flow cytometry. (F) TEM image of apoptotic bodies from CCR2-overexpressed ADSCs. (F) NTA evaluation of CCR2-ApopEVs. (G) Size distribution of PBs@CCR2-ApopEVs. (H) Statistical analysis of size of PBs@CCR2-ApopEVs. (I) Stability evaluation of PBs@CCR2-ApopEVs. (J) Zeta potential of PBs@CCR2-ApopEVs. (K) Statistical analysis of zeta potential of PBs@CCR2-ApopEVs. (L) TEM images of PBs@CCR2-ApopEVs. Data = Mean ± SD (n = 3) Effect of PBs@CCR2-ApopEVs on neutrophil-acinar cell co-culture system The followed step was to evaluate the efficacy of PBs@CCR2-ApopEVs on co-culture system of neutrophils and acinar cell in vitro for in vivo situation simulation. Transwell assay was conducted using both cells. As shown in Fig. 5A and B, primary acinar cells were isolated and cultured in Matrigel for 5 days stimulated with TGF-α in the bottom chamber of transwell dish. Then, neutrophils were extracted and cultured in the top chamber of transwell plate for the simulation of co-existence of both cells simulating of the SAP process. Then PBs@CCR2-ApopEVs were added into the co-culture system to examined whether they could adjust NETs formation of neutrophil and ADM occurrence of acinar cells. As mentioned above, mCherry and GFP were utilized to label ApopEVs and CCR2 ligand. We first evaluated the effect of PBs@CCR2-ApopEVs upon TGF-α pre-treated acinar cells. As depicted in Fig. 5C and D, while co-localized fluorescence signal of mCherry with acinar cells could be observed, the signal intensity in PBs@ApopEVs group was notably stronger. Importantly, ADM occurrence was also observed after diverse interventions with time passing (Fig. 5E and F). Green arrows indicated ADM structures. Under natural condition, little ADM happened in control group. However, TGF-α promoted ADM progression. And ApopEVs further enhanced ADM in injured acinar cells. Interestingly, despite PBs@CCR2-ApopEVs may exhibited better ADM-promoting effects theoretically, the results still demonstrate no significance of ADM numbers between normal PBs@ApopEVs and PBs@CCR2-ApopEVs. We speculated that this may be attributed to the interactions between formulations and acinar cells was not enough in vitro even if the given concentration was equal, or CCR2 is not suitable for assisting targeting acinar cells. Further, we performed immunohistochemistry and semi-quantified the level of KRT19 and Amy to assess ADM indirectly (Fig. 5G and I), which demonstrated similar results. Ultimately, we investigated the expression of p-STAT3 and SOX4 levels in Fig. 5J and K. The red dashed line indicated positive signals. It was easy to tell that ApopEVs groups exhibited the strongest capacity to promote ADM of acinar cells. Efficacy investigation of PBs@CCR2-ApopEVs on neutrophil-acinar cell co-culture system. (A–B) Illustration and schedule of the co-culture system development. (C–D) Cell uptake images and statistical analysis of PBs@CCR2-ApopEVs with mCherry tag in primary acinar cells. (E–F) Images of acinar cells in different days after various treatments and its statistical analysis. (G–K) Immunohistochemistry images of different factors in acinar cells ADM process with various treatments and relevant quantification results. Data = Mean ± SD (n = 4). Statistical analysis was performed by GraphPad Prism (9.5.0). One-way ANOVA using the Tukey’s multiple-comparison test was applied for comparison among three groups. *P < 0.05, **P < 0.01, ***P < 0.001. N.s. = Not significant On the other hand, in SAP stages pancreas with high concentration of CCL2 could attracted PBs@CCR2-ApopEVs. Therefore, we investigated the neutrophil targeting ability and relevant NETs inhibition effect of PBs@CCR2-ApopEVs. As illustrated in Fig. 6A and B, the intensity of each fluorescence was co-localized more with neutrophils in PBs@CCR2-ApopEVs when compared to other groups, indicating the successful construction of CCR2-ApopEVs and PBs wrapping. This specific targeting strategy leveraged the binding of prominent CCR2 on these ApopEVs and its cognate ligand CCL2 in neutrophils. The accumulated PBs@CCR2-ApopEVs then showed excellent NETs inhibition effect in Fig. 6C and D. SYTOX and CitH3 fluorescence signals were obviously detected in PMA-treated neutrophils, whereas formulations involved PBs all demonstrated sufficient inhibition efficacy especially PBs@CCR2-ApopEVs. The intuitive phenomenon was that reticular structure in extracellular space significantly decreased. At the same time, dsDNA of neutrophils was determined to quantify the NETs formation in Fig. 6E, and PBs@CCR2-ApopEVs exhibited best suppressive effect. The major components of NETs such as MPO and ELANE were further found to be downregulated after PBs@CCR2-ApopEVs intervention (Fig. 6F and G). And the level of N-GSDMD protein in neutrophils was determined after treatments, which exhibiting the best sufficient inhibiting effect of PBs@CCR2-ApopEVs (Fig. 6H). Evaluation of NETs in neutrophil-acinar cell co-culture systems after treatments. (A) Cellular uptake of PBs@CCR2-ApopEVs in neutrophils. (B) Relative signal intensity of GFP and mCherry in neutrophils after treatments. (C) Immunofluorescent images of SYTOX Green and semi-quantitative analysis of NETs in neutrophils after treamtents. (D) Free dsDNA in supernatant of co-culture system after various treatments. (E) Immunofluorescent images of CitH3 in neutrophils after interventions and its semi-quantitative analysis. (F-G) Expression level of ELANE and MPO in neutrophils in different groups. (H) Quantitative analysis of N-GSDMD protein in neutrophils after diverse interventions. Data = Mean ± SD (n = 4). Statistical analysis was performed by GraphPad Prism (9.5.0). One-way ANOVA using the Tukey’s multiple-comparison test was applied for comparison among three groups. *P < 0.05, **P < 0.01, ***P < 0.001. N.s. = Not significant Accordingly, the evaluation upon neutrophil-acinar cell co-culture system demonstrated the effectiveness of PBs@CCR2-ApopEVs against inflammation process and acini injury, revealing the potential to alleviate SAP. Pancreas targeting ability of PBs@CCR2-ApopEVs The specific targeting ability of PBs@CCR2-ApopEVs in SAP process decided the efficacy. As described above, we developed the CCR2-overexpressed ApopEVs to wrap PBs. In order to evaluate the enhanced pancreas targeting, we performed In Vivo Imaging System (IVIS) to visualize the biodistribution of our formulation. The results demonstrated that after intravenous injection of different formulations, PBs@CCR2-ApopEVs exhibited significant pancreas accumulation in comparison with other groups, indicating that using CCR2/CCL2 axis was feasible (Figure S5A). The statistical analysis of fluorescence intensity also demonstrated the exciting accumulation of PBs@CCR2-ApopEVs (Figure S5B). The fluorescence in excised pancreas strongly confirmed this conclusion, though livers were still the major distribution sites of external substances (Figure S5C). Intriguingly, fluorescent signals weren’t confined to the pancreas, they were also observed in the gut (Figure S5D). This unexpected detection potentially indicates a spread of inflammation from the pancreas into the intestinal tissue. Quantitative analysis further supported this picture, demonstrating that the CCR2 incorporated into our formulation was indeed effective at directing the therapeutic vehicle towards areas exhibiting high levels of CCL2 (Figure S5E). In summary, CCL2 released by neutrophils in SAP efficiently recruited PBs@CCR2-ApopEVs towards targeted sites for later therapeutic effects. Anti-SAP efficacy in vivo of PBs@CCR2-ApopEVs The final purpose of our study was to apply the novel PBs@CCR2-ApopEVs in SAP patients. Thus, we first evaluated the efficacy of it in SAP mice models. Mice were randomly divided into six groups including normal (negative control) group, positive control group, PBs, ApopEVs, PBs@ApopEVs and PBs@CCR2-ApopEVs. SAP often showed obvious organ damage in pancreas and lungs due to the amplified inflammatory responses. As described in our previous article, mice were intraperitoneally injected with 12 doses of Caerulein (Selleck, FL6934, Japan) (50 µg/kg) at 1-hour intervals after fasting for 12 h. Different formulations were administrated at the first dose and the 12nd dose via tail vein to simulate the actual situation in clinical rescue. H&E images in Fig. 7A illustrated pathological changes of pancreatic injury. The extent of tissue structure damage in pancreatic tissue was observed and scored including edema, inflammatory cell infiltration, etc. Single PBs or ApopEVs demonstrated limited alleviation effect against SAP, while PBs@ApopEVs showed better efficacy according to the histological score analysis. Due to the better targeting ability and accumulation, PBs@CCR2-ApopEVs exhibited the best efficacy to inhibit pancreatic lesion. Meanwhile, systemic inflammation was neutralized across the body, resulting in less lung injury as shown in Fig. 7B, whereas other formulations could not significantly decrease the inflammatory responses elicited in lung tissues. These results indicated that the targeted delivery system can not only effectively treat primary pancreatic lesions, but also significantly alleviate SAP-related secondary lung injury, demonstrating systemic protective potential. Efficacy evaluation of PBs@CCR2-ApopEVs in vivo. (A-B) H&E staining images of pancreas and lungs in different treatment groups and corresponding semi-quantitative analysis. (C-D) Activity of amylase and lipase in blood after various treatments. (E) TUNEL staining images and semi-quantitative analysis of pancreatic tissues for cell death rate evaluation after treatments. (F) Immunofluorescent images of CitH3 in pancreas after interventions and its corresponding statistical analysis. (H) Immunofluorescent images of N-GSDMD in pancreatic tissues after different treatments and relevant analysis. (I) Immunofluorescent images of ADM in pancreatic tissues after various treatments. Data = Mean ± SD (n = 8). Statistical analysis was performed by GraphPad Prism (9.5.0). One-way ANOVA using the Tukey’s multiple-comparison test was applied for comparison among three groups. *P < 0.05, **P < 0.01, ***P < 0.001. N.s. = Not significant In clinical diagnosis, determining the levels of certain enzymes in blood circulation is of vital importance, including lipase and amylase. Elevating their levels reflected damage to pancreatic acinar cells and the release of enzymes into the blood. A significant increase in enzyme levels is one of the core criteria for diagnosing SAP. Therefore, after histopathological section staining, we detected the level of both enzymes in blood to assess the situation of SAP. It is gratifying that results in Fig. 7C and D demonstrated the effectiveness of PBs@CCR2-ApopEVs in inhibiting the two-enzyme leakage from pancreas. The self-digestion of pancreas by these enzymes led to the death of acinar cells, guiding us to using TUNEL to detect cell death. As show in Fig. 7E, green fluorescence in TUNEL staining indicated dead cells in pancreatic tissues. After SAP establishment, positive control exhibited the strongest fluorescent signal while other treatment groups all demonstrated decreased signals to varying degrees, among which PBs@CCR2-ApopEVs exhibited best cellular death inhibition efficacy which may stem from the synergy between the anti-inflammatory/anti-NETs effects of PBs and the pro-repair/regenerative effects of ApopEVs. The neutrophils would be recruited and activated during SAP stages. In order to assess the NETs inhibition capacity of PBs@CCR2-ApopEVs, we then determined the CitH3 expression in pancreas-associated neutrophils by immunofluorescence. As illustrated in Fig. 7F, negative control groups displayed normal CitH3 expression. In contrast, positive group exhibited a significant positive signal, implying a widespread neutrophil activation and NETs formation. However, image of PBs-treated group evidently downregulated the fluorescent signals of CitH3 in pancreas, indicating a strong mitigation. After wrapped by CCR2-ApopEVs, PBs@CCR2-ApopEVs demonstrated the strongest alleviation effect, even close to that of the negative group. To further investigated the NETs inhibition, the level of blood dsDNA was examined (Fig. 7G). Notably, there were no significant differences between positive group and mono-ApopEVs. We speculated that the major efficacy of NETs formation was due to the existence of PBs rather than ApopEVs. On the other hand, despite PBs exhibited evident dsDNA-release suppression effect, the final formulation PBs@CCR2-ApopEVs were undoubtedly the most powerful combination for systemic NETs inhibition. Next, N-GSDMD expression was determined to further prove the best efficacy of PBs@CCR2-ApopEVs (Fig. 7H). Accordingly, we verified the excellent efficacy of PBs@CCR2-ApopEVs in suppressing neutrophil activity and NETs formation during SAP. In other word, the external elements of SAP inflammation would be significantly inhibited by our formulation. This validation supported the mechanism that PBs may reduce NETs and inflammatory damage especially being delivered by CCR2-ApopEVs. Next, we verified the ADM occurrence in SAP after various treatment. We first evaluated the levels of KRT19 to indirectly reflect ADM numbers in SAP. The red fluorescence represented amylase. The stronger green fluorescence intensity (KRT19) in acinar cells meant more ADM occurred. In negative group (normal group), no KRT19 signals were detected due to the non-existence of cell damage in Fig. 7I. But the TGF-α treatment boosted ADM of acinar cells in positive control group. With various interventions, the highest fluorescence signals were observed in PBs@CCR2-ApopEVs group, and correspondingly red signals of amylase was reduced. This superior effect of drug-loaded vesicles may be due to different mechanisms: PBs inhibiting inflammation, creating a better milieu for ApopEVs-mediated repair and CCR2 protein guided the delivery system to pancreas for better therapeutic effectiveness. Then we next conducted immunohistochemistry staining of cleaved caspase 3. In Figure S6A, we observed the best apoptosis-inhibition ability of PBs@CCR2-ApopEVs in acinar cells among all interventions. Complementary to the TUNEL results (detecting DNA fragmentation), this validated that the treatment reduced cell death by inhibiting the classical caspase-dependent apoptosis pathway. Further, we reconfirmed whether the components of ApopEVs could upregulate the expression of core factor SOX4 in vivo, as evidence of its mechanism for promoting ADM and pancreas repair. The fluorescent images demonstrated a consistent trend of efficacy in Figure S6B. Moreover, the inflammatory immunocytes infiltration in pancreas also reflected the severity of SAP. In Figure S6C and S6D, we investigated the major inflammatory cells in SAP mice models including neutrophils and macrophages. All images demonstrated the strongest inhibition effect of PBs@CCR2-ApopEVs in inflammation infiltration, directly confirming that the targeted delivery system could effectively controlling the inflammatory response. And this was consistent with the potential immunomodulatory effects of PBs in inhibiting NETs (neutrophil-related) and ApopEVs as reported. It was noted that another critical consideration for clinical efficacy of PBs@CCR2-ApopEVs is the dual nature of ADM. While ADM functions as a temporary protective mechanism to preserve the acinar cell pool during acute injury, persistent ADM is recognized as a potential precursor to pancreatic cancer, particularly when driven by oncogenic Kras mutations. To address this concern, we conducted a comparison between the ADM observed in our SAP model (wild-type C57BL/6J mice) and that in Kras-mutant mice (Figure S7). The results revealed fundamental distinctions between the two phenotypes. Unlike the irreversible and atypical lesions driven by Kras mutations, the ADM promoted by ApopEVs exhibited distinct characteristics in terms of morphology, size, duration, and outcome, suggesting it was a transient, regenerative process essential for maintaining homeostasis rather than a neoplastic transformation. However, we acknowledge that in a clinical setting, we cannot fully rule out the risk if this therapy is applied to patients with pre-existing Kras mutations. Interestingly, patients with Kras mutations often present with a history of recurrent pancreatitis, and their lesions might already be detectable through medical imaging. Therefore, to ensure safety in future clinical applications, we propose a strict screening protocol to exclude these high-risk patients before administering the treatment. Additionally, it is worth noting that the specific delivery mechanism of PBs@CCR2-ApopEVs relies on the pathological characteristics of the SAP microenvironment. We proposed that upon CCR2-mediated accumulation in the pancreas, the lipid membrane of ApopEVs is partially compromised by the high concentration of inflammatory enzymes and ROS. This triggered the extracellular release of PBs to scavenge ROS and inhibit neutrophil NETosis. Meanwhile, intact ApopEVs interacted with acinar cells to initiate the ADM protective pathway. Additionally, we did not exclude the possibility that acinar cells may internalize PBs for intracellular antioxidant protection, and neutrophils may engulf ApopEVs for immunomodulation. These processes likely work synergistically to enhance the overall therapeutic efficacy. Collectively, in vivo results underscored the significant potential of PBs@CCR2-ApopEVs in alleviating SAP, presenting a promising strategy to address the critical shortcomings of current symptomatic treatments in clinical SAP therapy. Conclusion and prospects SAP is a significant clinical challenge, characterized by both high incidence and alarmingly high mortality rates. The disease’s pathology is marked by a constellation of devastating features, including extensive tissue and cellular necrosis, an amplified inflammatory cascade and oxidative stress. Consequently, effectively controlling the initial inflammatory burst especially the key neutrophils and preserving acinar cell function as early as possible are critical for preventing the systemic complications. Herein, we aimed to resolve the clinical challenge of SAP that current strategy could not effectively alleviate the core pathological stages involving cascade inflammation and acinar cell damage. We developed a delivery vehicle named PBs@CCR2-ApopEVs regarding this problem. The overexpressed CCR2 on MSCs-derived ApopEVs could guide the whole system into injured pancreatic tissues due to the specific binding between CCR2 and CCL2 (released by the numerous neutrophils in SAP milieu). Then cells took up the drugs. On the one hand, the crucial neutrophil activity was limited via NETs formation inhibition by PBs. The expression of major component in NETs, including MPO and ELANE was significantly inhibited while N-GSDMD signal pathway was also downregulated. On the other hand, ApopEVs could effectively boost the occurrence of ADM through p-STAT3/SOX4 signal pathway activation, a self-protective process of acinar cell when being stimulated. Therefore, these effects synergistically protected pancreatic tissue from injury through “external” harmful factors inhibition and “internal” beneficial factor promotion. Despite these advances, it’s important to acknowledge certain limitations inherent in the present study. Notably, our current work did not explore the potential synergistic effects of PBs@CCR2-ApopEVs when combined with established anti-SAP therapeutic approaches. In addition, while we have confirmed the role of the ROS-GSDMD axis, future studies could further investigate whether PBs also interact with other NETs-related pathways to provide a holistic map of their molecular targets. And a thorough understanding of the long-term prognosis for SAP patients receiving PBs@CCR2-ApopEVs treatment will need further investigations. Additionally, considering the clinical translation, three key aspects warrant future attention. First, regarding the optimal intervention window, we propose that early treatment is most effective. Since the immune profile of SAP shifts from neutrophil-dominated to macrophage-dominated over time, applying our system in the early phase can effectively block the initial cytokine storm. Second, given the systemic nature of SAP, our drug may also accumulate in downstream organs like the lungs. We consider this a potential advantage, as it allows for the simultaneous management of both the primary pancreatic injury and systemic complications, such as lung injury. Finally, regarding the translational potential, this study utilized mouse-derived ADSCs to ensure histocompatibility with the murine SAP model. However, clinical application inevitably requires the use of human-derived cells. Therefore, a critical future direction will be to isolate ApopEVs from human ADSCs and evaluate their therapeutic efficacy and cross-species immunocompatibility. Verifying whether human-derived vesicles can exert similar protective effects without inducing rejection will be an essential step toward bridging the gap between these preclinical findings and potential clinical trials. In summary, our PBs@CCR2-ApopEVs demonstrated efficient pancreatic targeting, favorable biocompatibility, and potent therapeutic efficacy through multiple mechanisms. These combined properties resulted in rapid relief of clinical symptoms in SAP. This novel nanodrug delivery formulation, employing PBs and CCR2-overexpressed MSCs-derived ApopEVs, thus offers a promising strategy for the acute management of SAP and provides a valuable platform for enhancing the therapeutic profile of novel agents. Supplementary Information Supplementary Material 1. Acknowledgements None. Abbreviations Acinar-ductal metaplasia Adipose-derived mesenchymal stem cells Apoptotic extracellular vesicles Citrullinated histone H3 Damage-associated molecular patterns Gene Set Enrichment Analysis Inflammatory bowel disease Mesenchymal stem cells Myeloperoxidase Neutrophil extracellular traps Prussian blue nanoparticles Phorbol myristate acetate Protein kinase C Reactive oxygen species Rheumatoid arthritis Severe acute pancreatitis Author contributions Wenjie Xu: Conceptualization, Methodology, Software, Data curation, Validation, Visualization, Writing – original draft, editing& revising.Wenwu Cai: Conceptualization, Methodology, Software, Data curation, Validation, Visualization, Writing – original draft, editing& revising.Junyong Wu: Investigation, Experiment, Data curation, Visualization.Xuyang Hou: Experiment, Data analysis.Xiaoyan Qi: Data analysis, Writing -review & editing.Zuxing Wei: Investigation, Data curation, Visualization.Yimiao Cheng: Data analysis, Writing editing.Yanwen Zheng: Writing editingBeibei Cui: Writing reviewJun He: Methodology, Funding acquisition, Resources, Supervision, Methodology, Resources, Writing – review & editing. Funding This research was supported by Hunan Provincial Natural Science Foundation (Grant No. 2023JJ40821), Hunan Provincial Health Commission Health Research Project (Grant No. W20243029), Changsha Natural Science Foundation (Grant No. kq2208308), and National Natural Science Foundation (Grant No. 82300739). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate This study did not involve any human participants, human data or human tissues. And all animal experiments were performed after obtaining approval (No.20231282) from the Ethics Committee of the Second Xiangya Hospital of Central South University, with minimal animal suffering. Consent for publication All authors agree to the publication of the article. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Wenjie Xu and Wenwu Cai contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement No datasets were generated or analysed during the current study.