Sprayable asymmetric hydrogel prevents postoperative pleural adhesions via physical and biological dual therapy Authors: Ni Li, Zherui Shao, Nuo Cheng, Yongbin Wang, Bin Wu, Bingchuan Hu, Qinning Wang, Yanggang Dan, Yuwei Zhou, Kanghao Zhu, Renyuan Xiao, Caoyun Ju, Cong Wang, Dawei Zheng Published: 2026 ABSTRACT Postoperative pleural adhesions represent a prevalent complication that causes chronic pain, impaired pulmonary function, and escalated reoperation risks. Current preventions strategies remain limited by inadequate multifunctionality and an underestimated fibro-inflammatory microenvironment. Here, we report a powder-based asymmetric adhesive hydrogel integrated with dual-drug liposomes for adhesion prophylaxis. Upon contact with body fluids, the powder rapidly forms an asymmetric adhesive hydrogel within 20 s, exhibiting strong wet-tissue adhesion (∼26 kPa), effective hemostasis, sealing, antibacterial and ROS scavenging capabilities, which display superiority over the commercial absorbable adhesion barrier (INTERCEED®). Embedded liposomes co-loaded with miR-124-3p (an identified critical miRNA from adhesive tissues) and simvastatin enable sustained release, synergistically disrupting the fibro-inflammatory cascade—miR-124-3p attenuates fibrosis via targeting TGFBR1 while simvastatin suppresses pro-inflammatory signaling. In pleural injury rats, the system markedly reduces adhesion, inflammation and fibrosis, while promoting alveolar repair. This study combines an easy-to-apply hydrogel with rationally designed combination therapy, offering a promising strategy for preventing postoperative adhesions with broad potential for translation to other fibrotic disorders. Introduction Postoperative pleural adhesions, characterized by the formation of aberrant fibrous bands between the visceral and parietal pleura, constitute a frequent and serious complication following thoracic surgeries such as pulmonary resections, pleurectomies, and cardiac interventions [1,2]. These adhesions can lead to chronic pain, restrictive lung dysfunction, substantial technical challenges during re-operative surgeries, and elevated risks of iatrogenic injury [2,3]. The resultant decline in patients’ quality of life highlights an urgent clinical demand for more effective preventive strategies. Current management primarily depends on refined surgical techniques combined with physical barrier materials, most commonly oxidized regenerated cellulose meshes (e.g., INTERCEED®) or hyaluronic acid-based gels [4,5]. However, these conventional approaches are limited by poor adhesion to wet, dynamic visceral surfaces, a lack of integrated drug intervention, and insufficient multifunctionality that combines hemostatic, anti-infective and anti-inflammatory capacities, all of which are essential for directing a well-coordinated healing process [[6], [7], [8]]. Consequently, there is a critical unmet need for an integrated, bioactive system capable of actively modulating the complex postoperative pleural microenvironment. Hydrogels have emerged as a promising candidate for adhesion prevention due to their excellent biocompatibility, tunable mechanical properties, and versatility as both physical barriers and drug depots [9], with recent designs incorporating tissue adhesion, self-healing, and antimicrobial activity [10,11]. Their translation to the pleural cavity is challenged by continuous respiratory motion, which demands strong yet elastic adhesion to the visceral wound while strictly avoiding attachment to the parietal pleura. Surgical wounds in this setting are often irregular and accompanied by bleeding or air leakage; persistent hemorrhage accelerates fibrosis, and air leaks compromise respiration, while post-surgical inflammation or infection further exacerbates adhesion formation. Powder-based hydrogels that rapidly absorb wound fluids to form a cohesive gel in situ are uniquely suited to address these demands: they enable effortless spray application, conform to irregular topographies, and provide immediate hemostasis and sealing. Nevertheless, asymmetric adhesive powder-based hydrogels that simultaneously achieve visceral-surface adhesion and parietal-surface antifouling remain largely unexplored for pleural adhesion prevention. Beyond physical barriers, pharmacological inhibition represents a crucial biological intervention. Given the complex pathogenesis involving inflammation, fibroblast hyperactivation, and mesothelial-to-mesenchymal transition (MMT) [12,13], monotherapies targeting isolated pathways, such as corticosteroids, antifibrinolytic agents, or anti-proliferative drugs, have yielded suboptimal outcomes [14]. This underscores the necessity for a multi-target combinatorial strategy. Emerging evidence has identified specific microRNAs (miRNAs) as master regulators of fibrotic cascades with strong therapeutic potential [15], making the identification of a key dysregulated miRNA in adhesion pathogenesis a critical step toward rational combination therapy. In this context, statins (e.g., simvastatin), represent an ideal co-therapeutic candidate due to their pleiotropic anti-inflammatory, antioxidant, and direct anti-fibrotic actions [16,17]. The synergistic combination of this genetic regulator and small molecule promises a concerted attack on adhesion formation at multiple levels. However, the design of a stable and efficient co-delivery system for these functionally distinct agents, and their effective incorporation into a powder-based hydrogel matrix, remain a significant challenge yet to be addressed. Herein, we report a sprayable, powder-based asymmetric adhesive hydrogel that rapidly forms a Janus barrier upon contact with body fluids (Fig. 1). The adhesive layer (L@HLS) incorporates liposomes co-loaded with simvastatin and a fibrosis-suppressing miRNA (Sm-lipo) within a network of aldehyde-modified hyaluronic acid, polylysine, and 5-hydroxydopamine-modified silk fibroin, enabling robust wet-tissue adhesion, hemostasis, and sustained drug release. The non-adhesive layer (HP), composed of HAA and 4-arm PEG-SH, provides antifouling properties. Applied in two steps, the system achieves durable visceral-side adhesion while preventing parietal attachment; concurrently, the released Sm-lipo suppresses mesothelial-to-mesenchymal transition, inhibits fibroblast activation, and modulates macrophages. Evaluated in a rat thoracotomy model, this strategy combines a physical barrier with synergistic biological intervention to reprogram the wound microenvironment, utilizing fully biocompatible materials with clinical translational potential.Fig. 1Schematic diagram of preparation and mechanism of asymmetric powder-based hydrogel system (HP/L@HLS). (a) Preparation of HP/L@HLS. (b) Two-step application process of HP/L@HLS. (c) Mechanisms of HP/L@HLS against postoperative pleural adhesions, including the physical barrier function, hemostatic/ROS-scavenging/antimicrobial properties, and the biological intervention of the fibro-inflammatory microenvironment via suppression of MMT, inhibition of fibroblast hyperactivation, and bidirectional modulation of macrophage function. Schematic diagram of preparation and mechanism of asymmetric powder-based hydrogel system (HP/L@HLS). (a) Preparation of HP/L@HLS. (b) Two-step application process of HP/L@HLS. (c) Mechanisms of HP/L@HLS against postoperative pleural adhesions, including the physical barrier function, hemostatic/ROS-scavenging/antimicrobial properties, and the biological intervention of the fibro-inflammatory microenvironment via suppression of MMT, inhibition of fibroblast hyperactivation, and bidirectional modulation of macrophage function. Results and discussion Design and characterization of HP/HLS hydrogel To achieve the contact adhesion of hydrogel to the surgical wound and avoid the adhesion to the parietal pleura, a biocompatible Janus hydrogel with a potentiated adhesive layer and an antifouling coating was designed as shown in Fig. 2a. For the purpose of satisfactory adhesion to the wet surface of post-surgical lung wound and to avoid displacement during respiratory motion, a triple-adhesive approach was introduced via the catechol groups that are usually present in mussel byssus to enhance the wet adhesive properties, aldehyde groups that can react with proteins to achieve close contact with wound, and lysine residues with positive charges that confer both strong interfacial adhesion in a moist environment and a persistent antibacterial potential. More beneficially, catechol groups can be oxidized into benzoquinone groups by ROS, exhibiting ROS-scavenging ability. Based on these merits, 5-hydroxydopamine-grafted silk fibroin (5-HDP-SF) (Figure S1a and Figure S1b) and aldehyde-modified hyaluronic acid (HAA) were synthesized and characterized by 1HNMR (Figure S2a and Figure S2b). Among them, the FT-IR spectroscopy of HAA revealed a new C=O absorption band at approximately 1720 cm−1 (Fig. S2c), and the oxidation degree of HAA was determined to be 68.30% by Schiff reagent titration method, further confirming the successful synthesis. Moreover, the FT-IR spectroscopy of HP showed that the characteristic peak of the aldehyde group at 1728.3 cm−1 disappeared, while a characteristic peak of the imine bond at 1651.7 cm−1 formed (Fig. S4).Fig. 2Design and characterization of powder-based HP/HLS hydrogel. (a) Schematic diagram of preparation of HP/HLS hydrogel. (b) In vitro gelation of HLS hydrogel powder. (c) Gelation of HLS hydrogel powder on the lung surgical wound. (d) In vitro gelation of HP hydrogel powder. (e) SEM morphology of lyophilized HP/HLS hydrogel. The white dot line indicates the interface between HP hydrogel and HLS hydrogel layer. Scale bar: 100 μm. Design and characterization of powder-based HP/HLS hydrogel. (a) Schematic diagram of preparation of HP/HLS hydrogel. (b) In vitro gelation of HLS hydrogel powder. (c) Gelation of HLS hydrogel powder on the lung surgical wound. (d) In vitro gelation of HP hydrogel powder. (e) SEM morphology of lyophilized HP/HLS hydrogel. The white dot line indicates the interface between HP hydrogel and HLS hydrogel layer. Scale bar: 100 μm. Next, the hydrogel powders were prepared. In brief, the mass ratio of HAA, PL and 5-HDP-SF was optimized at 1:2:4 (HAA:5-HDP-SF: PL, w/w/w) to gain the potentiated adhesive HLS layer (Fig. S3a). Specifically, the ratio of HAA and PL was set as 1:4 (HAA: PL, w/w), obtaining a hydrogel with relatively low swelling ratio (∼120%) after 48 h to maintain the hydrogel network (Fig. S3b). Subsequently, the HLS hydrogel was lyophilized and ground into powders for application convenience (Fig. S5a and S5b). Notably, the HLS powder readily underwent re-gelation upon contact with minimal liquid within 20 s (Fig. 2b and Supplementary video 1) and it could rapidly re-gel when spraying on the lung surgical wound (Fig. 2c). This feature also endows HLS powder with sponge-like hemostatic ability due to the large specific surface area of gel powder. Then, the antifouling coating layer (HP gel) composed of HAA and 4-arm-PEG-SH at a 1:4 mass ratio exhibited the rapid crosslinking ability due to the thiol-aldehyde coupling. The lyophilized HP powder retained the rapid re-gelation ability within several seconds (Fig. 2d and Supplementary video 2), similar to HLS powder. Supplementary video related to this article can be found at https://doi.org/10.1016/j.mtbio.2026.103297 The following is the supplementary data to this article. To prevent pleural adhesions and facilitate lung tissue repair, the HLS powder was designed to be sprayed onto the surgical wound for hemostasis and wound repair, followed by the addition of HP powder for anti-adhesion. Therefore, we evaluated whether HP powder could form a stable gel on the HLS hydrogel surface to simulate the in situ administration. Results confirmed that upon adding HP powder and a small amount of PBS to the pre-formed HLS hydrogel, a bilayered asymmetric hydrogel with a hierarchically porous structure was obtained (Fig. 2e and S6). SEM imaging of the native (never-lyophilized) hydrogel revealed that both layers exhibited interconnected porous networks, but the adhesive HLS layer displayed a denser pore arrangement and thicker pore walls compared to the non-adhesive HP layer (Fig. 2e and S7), clearly indicating the asymmetric Janus structure. After lyophilization and subsequent rehydration, the hydrogel retained the same asymmetric, interconnected porous architecture, demonstrating that the microstructure is faithfully recovered upon rehydration of the powder. Additionally, the lyophilized-rehydrated hydrogel exhibited smaller and more uniform pore sizes than the native state (Fig. S8). Moreover, the adhesion work between the HLS and HP layers reached approximately 150 J/m2 (Fig. S9), implying the robust interlayer bonding of HP/HLS hydrogel as a bilayer hydrogel. Mechanical performance, asymmetric adhesion and sealing capacity of HP/HLS hydrogel Following the fabrication of hydrogel powders, we evaluated the rheological and mechanical behaviors. All HLS, HP and HP/HLS hydrogels showed a significantly higher storage modulus (G’) than the loss modulus (G”) (Fig. 3a), confirming the formation of elastic hydrogel networks. Mechanical testing indicated that the compressive fracture strength of each hydrogel group reached approximately 103 Pa (Fig. 3b), while the tensile strengths exceeded 2 × 103 Pa (Fig. 3c). These results demonstrated that all hydrogels possessed adequate mechanical properties for practical application.Fig. 3Mechanical performance, asymmetric adhesion and sealing capacity of HP/HLS hydrogel. (a) The changes in G′ and G″ of HP, HLS and HP/HLS hydrogels under frequency scanning. (b) The stress of HP, HLS and HP/HLS hydrogels under different strain. (c) Tensile strengths of HP, HLS and HP/HLS hydrogels (n = 3) (mean ± SD; ns: no significant difference). (d) Representative images of the adhesion of HLS onto different soft tissues and hard materials. (e) Representative images of the HLS layer resisting water erosion at different times. (f) The schematic diagram of the lap-shear test. (g) The lap shear strength of different group (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ****P < 0.0001). (h) The schematic diagram of the antifouling ability of HP/HLS hydrogel. (i) Quantitative analysis of the unadsorbed BSA after different hydrogels treatments (n = 3) (mean ± SD; **P < 0.01, ****P < 0.0001). (j) Representative images of the sealing capacity on the lung-injured mouse model at different times. (k) Representative images of hemostasis on liver bleeding model. (l) Quantitative analysis of hemostatic time in different groups (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). (m) Quantitative analysis of blood loss in different groups (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). Mechanical performance, asymmetric adhesion and sealing capacity of HP/HLS hydrogel. (a) The changes in G′ and G″ of HP, HLS and HP/HLS hydrogels under frequency scanning. (b) The stress of HP, HLS and HP/HLS hydrogels under different strain. (c) Tensile strengths of HP, HLS and HP/HLS hydrogels (n = 3) (mean ± SD; ns: no significant difference). (d) Representative images of the adhesion of HLS onto different soft tissues and hard materials. (e) Representative images of the HLS layer resisting water erosion at different times. (f) The schematic diagram of the lap-shear test. (g) The lap shear strength of different group (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ****P < 0.0001). (h) The schematic diagram of the antifouling ability of HP/HLS hydrogel. (i) Quantitative analysis of the unadsorbed BSA after different hydrogels treatments (n = 3) (mean ± SD; **P < 0.01, ****P < 0.0001). (j) Representative images of the sealing capacity on the lung-injured mouse model at different times. (k) Representative images of hemostasis on liver bleeding model. (l) Quantitative analysis of hemostatic time in different groups (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). (m) Quantitative analysis of blood loss in different groups (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). Since the firm adhesion of the HP/HLS hydrogel to the wound surface acts as a precondition for the prevention of postoperative pleural adhesions, we first evaluated the adhesion of HLS onto different soft tissues and hard materials. Our adhesive layer showed strong adhesion performance to all the tissues including heart, liver, spleen, lung and kidney, as well as different material interfaces capable of supporting a weight of 50 g (Fig. 3d), suggesting its broad application potential. Further, the HLS layer could resist the continuous flush of water at the flow rate of 0.77 m/s (at the water pressure of 0.1 MPa) for at least 20 s (Fig. 3e and Supplementary video 3), demonstrating the exceptional adhesive durability under the continuous shear stress. Next, we focused on evaluating the potentiated adhesion of the HLS layer on tissues. The lap-shear strength of HLS layer was then measured via a porcine skin lap-shear test (Fig. 3f). As expected, the lap-shear strength of HLS adhesive layer showed the highest adhesion compared with the 5-HDP-SF self-crosslinked hydrogel and the HAA-PL hydrogel, reaching about 26.2 ± 1.04 kPa (Fig. 3g), confirming that the potentiated adhesion was due to the combined effect of catechol groups in 5-HDP-SF, aldehyde groups in HAA and lysine residues in PL. These data verify that HLS possesses the robust adhesive properties necessary for long-term performance on the moist and dynamic lung surface. Supplementary video related to this article can be found at https://doi.org/10.1016/j.mtbio.2026.103297 The following is the supplementary data to this article. Furthermore, the asymmetric adhesion of HP/HLS hydrogel serves as the critical feature to avoid bridging the thoracic wall and lung tissue. Thus, the antifouling ability of HP layer was explored via an antifouling test (Fig. 3h). The result showed that the unadsorbed bovine serum albumin (BSA) in the HP layer group was about 89.00 ± 2.75%, similar to the control group (Fig. 3i), indicating the excellent antifouling ability of HP layer. This proves that the HP layer can serve as an antifouling coating to reduce the adhesion force on the other side of the HLS layer. To visually evaluate the asymmetric adhesion of HP/HLS hydrogel, a thoracic wall-lung tissue adhesion model was applied. The result demonstrated that the HP/HLS hydrogel obviously prevented attachment between the thoracic wall and lung tissue (Supplementary video 4), while the HLS layer bridged the thoracic wall and lung tissue, underscoring the importance of asymmetric adhesion in the hydrogel design (Supplementary video 5). The above results validated the asymmetric adhesive properties of the HP/HLS hydrogel and its potential for preventing pleural adhesions. Supplementary video related to this article can be found at https://doi.org/10.1016/j.mtbio.2026.103297 The following is the supplementary data to this article. Notably, the injury of lung as the vital organ in the respiratory system could result in air leakage. Serious cases can lead to breathing difficulties and can even be life-threatening. Based on this consideration, the sealing capacity was explored on the lung-injured mouse model (Supplementary videos 6-9). Specifically, an untreated lung-injured mouse model was used as a negative control (Supplementary video 6) and the commercial absorbable adhesion barrier (INTERCEED®) was applied as a positive control (Supplementary video 7). We found that although HLS hydrogel (Supplementary video 8), HP/HLS hydrogel (Supplementary video 9), and INTERCEED® exhibited effective sealing capability when lung went through the breath movement via a breathing machine, INTERCEED® experienced obviously displacement when the lung moved, while HP/HLS hydrogel maintained the close adhesion to the wound (Fig. 3j). Moreover, we observed that INTERCEED® might have no effect on the wound exudate, whereas the hydrogel possessed the capability of adsorbing wound exudate (Fig. 3j). Additionally, the sealing capability of HP/HLS hydrogel powder showed promise in hemostasis (Fig. 3k and S10a). The HP/HLS hydrogel powder quickly stopped the bleeding within 90 s in both the liver bleeding model and the mouse tail amputation model, whereas the control group required more than 4 min for the bleeding to stop (Fig. 3l and S10b). The cumulative blood loss in the HP/HLS hydrogel powder-treated group was about 26.37 ± 3.12 mg and 21.20 ± 3.41 mg, respectively, which was approximately one-sixth and one-half the volume lost in the control group on different models (Fig. 3m and S12c). Postoperative bleeding is known as one of the causes of postoperative adhesions [18,19]. Therefore, immediate hemostasis benefits the prevention of postoperative adhesions. In short, these data demonstrated the merits of HP/HLS hydrogel powder in the prevention of postoperative bleeding, wound exudate and air leakage over the commercial absorbable adhesion barrier, holding promise for postoperative application. Supplementary video related to this article can be found at https://doi.org/10.1016/j.mtbio.2026.103297 The following is the supplementary data to this article. Biocompatibility, biodegradability, antibacterial efficacy and ROS-scavenging activity of HP/HLS hydrogel powder Excellent biocompatibility represents an essential requirement for postoperative application. Thus, the biocompatibility of each component within the hydrogel was next evaluated by a hemolysis assay. All components exhibited a hemolysis ratio of less than 2.5% hemolysis ratio (Fig. 4a), suggesting their suitability for in vivo application. Cell counting kit-8 (CCK-8) assays revealed that no toxicity of the HP/HLS hydrogel was observed on the human pleural mesothelial cell line (MeT-5A) and the human fibroblast cell line (HFL1) for 72 h (Fig. 4b), verifying ideal biocompatibility.Fig. 4Biocompatibility, biodegradability, antibacterial efficacy and antioxidation activity of HP/HLS hydrogel powder. (a) Representative images and quantitative analysis of hemolysis test of different materials in HP/HLS hydrogel (n = 3) (mean ± SD; ns: no significant difference, ****P < 0.0001 vs negative group). (b) Quantitative analysis of MeT-5A and HFL1 cells viability cultured with HP/HLS hydrogels for 24 h, 48 h and 72 h via CCK-8 assay. (c) In vivo degradation behaviors of HP/HLS hydrogel in rat after subcutaneous injection. Bar: 2.5 mm. (d) The representative images of S.aureus and E.coli colony units after different treatment. (e) Quantitative analysis of the corresponding numbers of S.aureus and E.coli colony units after different treatment (n = 3) (mean ± SD; ****P < 0.001 vs control group). (f) Quantitative analysis of bacterial inhibition rate after different treatment. (n = 3) (mean ± SD; ****P < 0.001 vs control group). (g) The OD value of S.aureus and E.coli bacterial suspension at 600 nm after different treatment (n = 3) (mean ± SD; ****P < 0.001 vs control group). (h) The CLSM images of RAW 264.7 cells. Nucleus were labeled by Hoechst 33,342. ROS labeled by DCFH-DA. (Scale Bar: 50 μm) (i) Quantitative analysis of relative level of DCFH-DA in CLSM images (n = 3) (mean ± SD; ****P < 0.001 vs LPS group). Biocompatibility, biodegradability, antibacterial efficacy and antioxidation activity of HP/HLS hydrogel powder. (a) Representative images and quantitative analysis of hemolysis test of different materials in HP/HLS hydrogel (n = 3) (mean ± SD; ns: no significant difference, ****P < 0.0001 vs negative group). (b) Quantitative analysis of MeT-5A and HFL1 cells viability cultured with HP/HLS hydrogels for 24 h, 48 h and 72 h via CCK-8 assay. (c) In vivo degradation behaviors of HP/HLS hydrogel in rat after subcutaneous injection. Bar: 2.5 mm. (d) The representative images of S.aureus and E.coli colony units after different treatment. (e) Quantitative analysis of the corresponding numbers of S.aureus and E.coli colony units after different treatment (n = 3) (mean ± SD; ****P < 0.001 vs control group). (f) Quantitative analysis of bacterial inhibition rate after different treatment. (n = 3) (mean ± SD; ****P < 0.001 vs control group). (g) The OD value of S.aureus and E.coli bacterial suspension at 600 nm after different treatment (n = 3) (mean ± SD; ****P < 0.001 vs control group). (h) The CLSM images of RAW 264.7 cells. Nucleus were labeled by Hoechst 33,342. ROS labeled by DCFH-DA. (Scale Bar: 50 μm) (i) Quantitative analysis of relative level of DCFH-DA in CLSM images (n = 3) (mean ± SD; ****P < 0.001 vs LPS group). Concerning in vitro biodegradability, both the HP and HLS hydrogels exhibited remarkably similar degradation profiles under physiologically relevant conditions, and the bilayer HP/HLS hydrogel achieved nearly complete degradation within 21 days (Figs. S11–S12). These results indicate that the two layers degrade at comparable rates, thereby preserving the asymmetric barrier integrity throughout the degradation period, demonstrating the superior biodegradability and providing a suitable time window to support pleural separation and lung tissue repair [[20], [21], [22]]. To assess the in vivo degradation profile of the HP/HLS hydrogels, a subcutaneous implantation in the dorsal skin of rats was performed. We found that the HP/HLS hydrogel was retained for at least 15 days and underwent complete degradation (Fig. 4c). In vivo degradation experiments were conducted on HP (Blue), HLS (Red), and HP/HLS after staining. It was observed that the retention times of HP and HLS in the body were essentially the same, and both were almost completely degraded by 15 days (Fig. S13). These results collectively demonstrate that the two layers degrade at comparable rates in vivo, thereby maintaining the asymmetric barrier integrity during the critical adhesion-forming period, and further confirm the favorable biodegradability and biosafety of the hydrogel system. The functional performances of the HP/HLS hydrogel, including antibacterial efficacy and antioxidation activity, was next assessed. Firstly, we evaluated the antibacterial efficacy of HP/HLS hydrogel against Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli). We found that the colony number was significantly reduced after treatment with both HLS hydrogel and HP/HLS hydrogel (Fig. 4d and 4e) and bacterial inhibition rate on both S. aureus and E. coli exceeded 85% (Fig. 4f). The decrease in the OD value of the bacterial suspension (Fig. 4g) further indicated that HLS hydrogel and HP/HLS hydrogel inhibited the growth of both gram-negative and gram-positive bacteria. This is primarily attributed to the antibacterial effect of PL in the hydrogel. Secondly, due to the presence of the catechol group in 5-HDP-SF, we further verified whether the HP/HLS hydrogel had the ability to scavenge excess intracellular ROS, thereby protecting cells from ROS damage. We observed a significant reduction in the number of DCFH-DA-positive macrophages (RAW264.7) in 5-HDP-SF, HLS, and HP/HLS groups (Fig. 4h). Especially, we found that the relative level of DCFH-DA in the HP/HLS group was significantly reduced by 25% compared to the LPS-stimulated group (Fig. 4i). In short, the above data demonstrated that the HP/HLS hydrogel has excellent biocompatibility, and the different components have endowed it with antibacterial and antioxidant properties, further expanding its potential for postoperative application. Identifying the potential therapeutic effects of miR-124-3p and simvastatin in prevention of pleural adhesion Despite the successful fabrication of HP/HLS hydrogel as a physical barrier, the complex fibro-inflammatory microenvironment continues to promote pleural adhesion formation. Therefore, a synergistic combination of medication and a physical barrier might achieve satisfactory anti-adhesion effect. Herein, using a combination of clinical samples from patients with postoperative pleural adhesions and a pulmonary surgery rat model, we aimed to identify a potential therapeutic target and develop an associated therapeutic approach (Fig. 5a). Histological evaluation and immunofluorescence detection demonstrated that pleural adhesion tissues displayed a dense architecture, substantial collagen deposition, and upregulation of α-SMA expression (Fig. S14). Subsequent proteomic profiling of clinical samples and adhesion tissue from the rat model identified upregulation of key fibrosis-associated proteins in adhesion tissues (Fig. 5b and S15), and quantitative real-time PCR (qPCR) analysis confirmed markedly elevated expression levels of α-SMA, Collagen I, and CTGF (Fig. 5c).Fig. 5Integrated multi-omics and functional validation reveal the role of miR-124-3p/TGFBR1 in pleural adhesion and the modulatory effects of simvastatin. (a) Schematic diagram of the processing of tissue samples from patients and rats with pleural adhesions. (b) Proteomic analysis of rat pleural adhesion tissues. (c) Human pleural adhesion tissues were subjected to qPCR to determine fibrotic-related gene expression (n = 5) (mean ± SD; *P < 0.05, **P < 0.01 vs Control group). (d) MiRNA high-throughput sequencing of human pleural adhesion tissues. (e) Luciferase reporter assay showed TGFBR1 as a target of miR-124-3p (n = 3). NC, negative control; WT, wild type; MUT, mutant (mean ± SD; *P < 0.05 vs miR-NC group). (f) MeT-5A cells transfected with miR-124-3p mimics were subjected to qPCR to determine fibrosis and MMT related gene expression (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs miNC + TGF-β group). (g) CTGF mRNA abundance in MeT-5A cells treated with simvastatin for 24 h. (h) MeT-5A cells treated with simvastatin were subjected to qPCR to determine fibrosis-related gene expression (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001, ****P < 0.0001 vs TGF-β group). (i) Abundance of IL-1β, IL-6, TNF-α, and CXCL9 in M1 macrophages treated with simvastatin for 24 h (n = 3) (mean ± SD; **P < 0.01). (j) Abundance of TGF-β, PDGF, and CTGF in M2 macrophages treated with simvastatin for 24 h (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). (k) Schematic diagram of the in vitro process for target validation. Integrated multi-omics and functional validation reveal the role of miR-124-3p/TGFBR1 in pleural adhesion and the modulatory effects of simvastatin. (a) Schematic diagram of the processing of tissue samples from patients and rats with pleural adhesions. (b) Proteomic analysis of rat pleural adhesion tissues. (c) Human pleural adhesion tissues were subjected to qPCR to determine fibrotic-related gene expression (n = 5) (mean ± SD; *P < 0.05, **P < 0.01 vs Control group). (d) MiRNA high-throughput sequencing of human pleural adhesion tissues. (e) Luciferase reporter assay showed TGFBR1 as a target of miR-124-3p (n = 3). NC, negative control; WT, wild type; MUT, mutant (mean ± SD; *P < 0.05 vs miR-NC group). (f) MeT-5A cells transfected with miR-124-3p mimics were subjected to qPCR to determine fibrosis and MMT related gene expression (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs miNC + TGF-β group). (g) CTGF mRNA abundance in MeT-5A cells treated with simvastatin for 24 h. (h) MeT-5A cells treated with simvastatin were subjected to qPCR to determine fibrosis-related gene expression (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001, ****P < 0.0001 vs TGF-β group). (i) Abundance of IL-1β, IL-6, TNF-α, and CXCL9 in M1 macrophages treated with simvastatin for 24 h (n = 3) (mean ± SD; **P < 0.01). (j) Abundance of TGF-β, PDGF, and CTGF in M2 macrophages treated with simvastatin for 24 h (n = 3) (mean ± SD; **P < 0.01, ***P < 0.001). (k) Schematic diagram of the in vitro process for target validation. MiRNAs have been recognized as important regulators of organ fibrosis [23]. Next, we performed high-throughput miRNA sequencing on clinical samples to identify the key miRNAs. Results showed significant downregulation of several miRNAs, including miR-124-3p, miR-130 b-5p, and miR-92 b-3p (Fig. 5d and S16a). We further validated these findings using qPCR, which confirmed markedly reduced expression of miR-124-3p (Fig. S17a). Integrated co-expression analysis of miRNA sequencing and proteomic data revealed a negative correlation between miR-124-3p and TGFBR1, a key regulator of fibrosis (Fig. S16b). Consistent with these findings, immunofluorescence staining demonstrated elevated TGFBR1 protein expression in human pleural adhesion tissues (Fig. S17b). Similar results were demonstrated in a rat pleural adhesion model (Fig. S18). Furthermore, to determine the direct regulatory effects of miR-124-3p on TGFBR1, we constructed luciferase reporter plasmids containing either the wild-type or a mutant 3′ untranslated region (UTR) of TGFBR1. The luciferase reporter assay revealed that transfection with miR-124-3p mimics significantly reduced luciferase activity in constructs containing the wild-type TGFBR1 3’ UTR compared to those with the mutant version (Fig. 5e). To further elucidate the role of miR-124-3p in pleural adhesion, we performed in vitro experiments using MeT-5A and HFL1 cells [24,25]. We demonstrated that both cell lines after transfection with miR-124-3p mimics showed a significant decrease in TGFBR1 expression and an obvious increase in the level of miR-124-3p (Fig. S19a–b and S20a), confirming the functional activity of the mimics. We next examined the influence of miR-124-3p overexpression on fibrosis- and MMT-related gene expression. qPCR analysis indicated that miR-124-3p markedly suppressed the expression of these genes (Fig. 5f and S20b), supporting its role in inhibiting both fibrotic and MMT processes. Consistent with these findings, immunofluorescence staining of MeT-5A cells and HFL1 cells demonstrated a corresponding reduction in protein levels of relevant markers (Fig. S15c–d and S20c). The above results confirmed the critical role of miR-124-3p in directly downregulating the expression of TGFBR1 and the following inhibition on fibrotic signaling. Given the critical involvement of CTGF in pleural adhesion formation (Fig. 5d) and the previously reported inhibitory effects of statins on CTGF synthesis as well as their well-established anti-inflammatory, antioxidant, and immunomodulatory properties [[26], [27], [28]], we hypothesized that simvastatin may counteract the inflammatory and fibrotic processes implicated in pleural adhesion pathogenesis. The results showed that both qPCR and immunofluorescence staining of MeT-5A cells and HFL1 cells treated with simvastatin confirmed a significant reduction of CTGF levels by simvastatin (Fig. 5g and h) and marked suppression of fibrosis-associated markers (Fig. S20d–e). Regarding the anti-inflammatory effect of simvastatin, we next explored its role in modulating macrophages, the central effector cells in the inflammation process [29]. As is well known, M1-like macrophages release pro-inflammatory cytokines in response to stimuli [30], while M2-like macrophages are strongly associated with fibrotic environments and disease pathogenesis [31]. Through the secretion of pro-fibrotic mediators such as TGF-β1, CTGF, and MMPs, M2-like macrophages promote fibroblast migration, proliferation, and activation, thereby contributing to abnormal tissue remodeling and fibrosis [32,33]. Thus, the immunomodulatory effects of simvastatin on both M1-and M2-polarized macrophages were explored. In M1 macrophages, simvastatin significantly reduced the expression levels of inflammatory factors (Fig. 5i). ELISA assay further confirmed the reduced secretion of IL-1β and TNF-α in the culture medium of simvastatin-treated M1 macrophages (Fig. S21). While in M2 macrophages, simvastatin suppressed key pro-fibrotic mediators, notably reducing CTGF expression by 32.22% (Fig. 5j). In short, simvastatin acts as a dual-function agent that counteracts both fibrosis and inflammation (Fig. 5k). Collectively, these results suggest that the combination of miR-124-3p and simvastatin provides a promising strategy for breaking the fibro-inflammatory microenvironment, which might favor the prevention of postoperative pleural adhesions. Notably, the clinical relevance of this rat pleural adhesion model is strongly supported by the molecular congruence we observed between the rat adhesion tissues and human postoperative pleural adhesion samples. Multi-omics analyses revealed that proteomic and miRNA profiles, including the key TGFBR1/miR-124-3p axis, are similarly dysregulated in both species, indicating that the core fibro-inflammatory pathways are conserved. This concordance validates the rat model as a biologically meaningful system for investigating adhesion mechanisms and testing therapeutic interventions. Fabrication and characterization of miR-124-3p and simvastatin co-loaded liposome (Sm-lipo) encapsulated hydrogel (HP/L@HLS) Having demonstrated the effects of miR-124-3p and simvastatin, we thus fabricated a dual-drug co-loaded liposome (Sm-lipo) to minimize miRNA degradation and enhance the delivery efficiency of miR-124-3p and simvastatin to PMCs, fibroblasts and macrophages at adhesion sites. Given the distinct physicochemical properties of miR-124-3p and simvastatin, simvastatin was encapsulated within the cationic liposomes (S-lipo) composed of cationic lipid DOTAP and cholesterol via a film-dispersion method, followed by the condensation of miR-124-3p with S-lipo via electrostatic interactions to gain the Sm-lipo (Fig. 6a). Specifically, S-lipo exhibited an average particle size of 97.4 nm, a zeta potential of +60.9 mV, an encapsulation efficiency (EE%) of 60.1%, and a drug loading capacity of 12.2% (Fig. 6b and Table S2), suggesting the successful encapsulation of simvastatin. Moreover, the high positive surface charge of S-lipo enabled efficient condensation with negatively charged miR-124-3p, resulting in Sm-lipo with an average particle size of 71.5 nm, a near-neutral zeta potential of +2.7 mV, and a spherical morphology (Fig. 6b, c and Table S2). We measured the drug loading of simvastatin in both native and lyophilized-rehydrated HP/L@HLS hydrogels. The drug loading is about 56.26 ± 2.88% in native HP/L@HLS hydrogel and 52.86 ± 3.77% in lyophilized-rehydrated HP/L@HLS hydrogel, suggesting the comparable drug loading efficiency (difference <5%) (Table S3). Additionally, we collected liposomes released from lyophilized-rehydrated hydrogels and imaged them by TEM. The sample exhibited spherical morphology with smooth surfaces and sizes consistent with freshly prepared Sm-lipo, indicating that the liposomal structure is well preserved throughout the process (Fig. S22). Furthermore, agarose gel electrophoresis confirmed the successful condensation of miR-124-3p into Sm-lipo (Fig. S23). Next, Sm-lipo was encapsulated within HLS hydrogel by blending Sm-lipo with the HLS hydrogel precursor to gain the drug depot (L@HLS). After lyophilization and grinding, L@HLS also possessed the re-gelation ability (Fig. 6d).Fig. 6Fabrication and characterization of miR-124-3p and simvastatin co-loaded liposome (Sm-lipo) encapsulated hydrogel (HP/L@HLS). (a) Schematic diagram of Sm-lipo preparation. (b) Particle sizes and zeta potentials of Sm-lipo (n = 3). (c) TEM morphology of Sm-lipo. Scale bar: 100 nm. (d) In vitro gelation of L@HLS hydrogel powder. (e-f) The cellular CLSM images of Sm-lipo after released from HP/L@HLS hydrogel in MeT-5A cells (e) and HLF1 cells (f). miRNA was labeled by Cy3. Sm-lipo was labeled by rhodamine-PE. Lysosome was labeled by Lysotracker green. Nucleus were labeled by Hoechst 33,342. The right images indicate the colocalization along the white line in the CLSM images. Bar: 5 μm. Fabrication and characterization of miR-124-3p and simvastatin co-loaded liposome (Sm-lipo) encapsulated hydrogel (HP/L@HLS). (a) Schematic diagram of Sm-lipo preparation. (b) Particle sizes and zeta potentials of Sm-lipo (n = 3). (c) TEM morphology of Sm-lipo. Scale bar: 100 nm. (d) In vitro gelation of L@HLS hydrogel powder. (e-f) The cellular CLSM images of Sm-lipo after released from HP/L@HLS hydrogel in MeT-5A cells (e) and HLF1 cells (f). miRNA was labeled by Cy3. Sm-lipo was labeled by rhodamine-PE. Lysosome was labeled by Lysotracker green. Nucleus were labeled by Hoechst 33,342. The right images indicate the colocalization along the white line in the CLSM images. Bar: 5 μm. To further evaluate the application possibility of HP/L@HLS hydrogel in preventing pleural adhesions, various characteristics of the asymmetric hydrogel including rheological, mechanical, adhesive, and degradation properties were investigated. These results showed that the incorporation of Sm-lipo modestly enhanced the mechanical strength of the crosslinked network (Fig. S24a), likely due to hydrogen bonding and electrostatic interactions between liposomal phosphate groups and hydrogel macromolecules, while it had no effect on the mechanical performance (Fig. S24b and S24c), asymmetric adhesion (Fig. S9), and degradation (Fig. S11). Finally, the drug release behavior and the following uptake by cells were explored. We found that simvastatin exhibited sustained release from the HP/L@HLS hydrogel (native state) over at least 14 days and the same situation was observed in the hydrogel (lyophilized-rehydrated state), even under simulated physiological conditions containing 30 U/mL HAase and 1 mM H2O2 (Fig. S25). Additionally, we conducted release studies for simvastatin from the native hydrogel under the same conditions and observed that the release profile was similar to that of the lyophilized-rehydrated hydrogel (Fig. S26). These results suggested that HP/L@HLS hydrogel enabled long-term local anti-fibrotic action, which might help suppress pleural hyperplasia and prevent adhesion recurrence. However, the release of miR-124-3p was hard to detect due to the extremely low content. Thus, we examined the uptake of Sm-lipo released from the HP/L@HLS hydrogel by different cells (Met-5A cells, HFL1 cells, and RAW264.7 cells) to illustrate the effective delivery of miR-124-3p and simvastatin to target cells. Prior to uptake studies, the HP/L@HLS hydrogel was confirmed to exhibit no cytotoxicity toward these cells (Fig. S27). Subsequently, confocal laser scanning microscopy (CLSM) images revealed that Sm-lipo was effectively internalized by all cell types, and subsequently escaped from lysosomes (Fig. S28). Moreover, we demonstrated that miR-124-3p remained encapsulated within Sm-lipo after being released from HP/L@HLS hydrogel and underwent lysosomal escape (Fig. 6e, f and S29), supporting its potential for downstream gene regulation. The frozen section of lung tissue after treatment with HP/L@HLS hydrogel further confirmed that the encapsulated Sm-lipo could be released from the hydrogel and penetrate into the lung tissue (Fig. S30). In summary, we successfully constructed a bilayer hydrogel-liposome hybrid system with appropriate mechanical strength, excellent biodegradability, and sustained release capacity. These features collectively support the potential of HP/L@HLS hydrogel for preventing postoperative pleural adhesions. Building on recent advances in anti-adhesion hydrogels, such as a microgel-based cream hydrogel with inherent anti-oxidative and anti-inflammatory properties and a dual-layer HPPT system combining a PEI/TA adhesive inner layer with a zwitterionic outer layer for peritoneal adhesion prevention [34,35], our HP/L@HLS hydrogel is specifically engineered for the more demanding pleural environment. Anti-pleural adhesion activity of HP/L@HLS hydrogel in vitro To evaluate the anti-pleural adhesion activity of HP/L@HLS, we employed MeT-5A, HFL1, and RAW264.7 cells for assessment. Firstly, we assessed the expression of key fibrosis-related proteins in MeT-5A cells via immunofluorescence. Compared with the control group, HP/L@HLS treatment significantly downregulated the expression of fibrotic markers (TGFBR1, Collagen I, CTGF, α-SMA, and Vimentin) while upregulating the epithelial marker CK19 (Fig. 7a and b and S31). This coordinated alteration indicates that HP/L@HLS effectively inhibits TGFBR1, thereby reversing the TGF-β-induced MMT (Fig. 7c).Fig. 7The efficacy of HP/L@HLS hydrogel was validated in vitro. (a) MeT-5A cells treated with HP/L@HLS hydrogel were subjected to immunofluorescence staining to determine fibrosis- and MMT-related protein expression. (b) The corresponding relative fluorescence intensity quantifications presented in the panel below (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs TGF-β group, #P < 0.05, ##P < 0.01, ###P < 0.001 vs indicated group). (c) Schematic diagram of MeT-5A cells treated with HP/L@HLS. (d) HFL1 cells treated with HP/L@HLS hydrogel were subjected to immunofluorescence staining to determine fibrosis-related protein expression. (e) The corresponding relative fluorescence intensity quantifications presented in the panel to the right (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs TGF-β group, #P < 0.05, ###P < 0.001 vs indicated group). (f) Schematic diagram of HFL1 cells treated with HP/L@HLS. (g) M1-type macrophages treated with HP/L@HLS hydrogel were subjected to ELISA to determine IL-1β secretion levels (n = 3) (mean ± SD; ***P < 0.001 vs M1 group, ##P < 0.01, ###P < 0.001 vs indicated group). (h) M1-type macrophages treated with HP/L@HLS hydrogel were subjected to ELISA to determine TNF-α secretion levels (n = 3) (mean ± SD; *P < 0.05, **P < 0.01 vs M1 group). (i) Abundance of IL-1β, IL-6, TNF-α, and CXCL9 in M1 macrophages after 24 h of HP/L@HLS hydrogel treatment (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs M1 group, #P < 0.05, ###P < 0.001 vs indicated group). (j) Abundance of TGF-β, PDGF, and CTGF in M2 macrophages after 24 h of HP/L@HLS hydrogel treatment (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs M1 group, #P < 0.05 vs indicated group). (k) Schematic diagram of Raw264.7 cells treated with HP/L@HLS. The efficacy of HP/L@HLS hydrogel was validated in vitro. (a) MeT-5A cells treated with HP/L@HLS hydrogel were subjected to immunofluorescence staining to determine fibrosis- and MMT-related protein expression. (b) The corresponding relative fluorescence intensity quantifications presented in the panel below (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs TGF-β group, #P < 0.05, ##P < 0.01, ###P < 0.001 vs indicated group). (c) Schematic diagram of MeT-5A cells treated with HP/L@HLS. (d) HFL1 cells treated with HP/L@HLS hydrogel were subjected to immunofluorescence staining to determine fibrosis-related protein expression. (e) The corresponding relative fluorescence intensity quantifications presented in the panel to the right (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs TGF-β group, #P < 0.05, ###P < 0.001 vs indicated group). (f) Schematic diagram of HFL1 cells treated with HP/L@HLS. (g) M1-type macrophages treated with HP/L@HLS hydrogel were subjected to ELISA to determine IL-1β secretion levels (n = 3) (mean ± SD; ***P < 0.001 vs M1 group, ##P < 0.01, ###P < 0.001 vs indicated group). (h) M1-type macrophages treated with HP/L@HLS hydrogel were subjected to ELISA to determine TNF-α secretion levels (n = 3) (mean ± SD; *P < 0.05, **P < 0.01 vs M1 group). (i) Abundance of IL-1β, IL-6, TNF-α, and CXCL9 in M1 macrophages after 24 h of HP/L@HLS hydrogel treatment (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs M1 group, #P < 0.05, ###P < 0.001 vs indicated group). (j) Abundance of TGF-β, PDGF, and CTGF in M2 macrophages after 24 h of HP/L@HLS hydrogel treatment (n = 3) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs M1 group, #P < 0.05 vs indicated group). (k) Schematic diagram of Raw264.7 cells treated with HP/L@HLS. Meanwhile, we observed pronounced anti-fibrotic effects of HP/L@HLS in HFL1 fibroblasts, which represent the primary effector cells in fibrosis. Compared to the S-lipo and M-lipo groups, the Sm-lipo treatment group exhibited a significant reduction in the expression of TGFBR1, as well as other fibrosis-related markers including Collagen I and α-SMA. Furthermore, HP/L@HLS demonstrated a more pronounced effect, achieving a further significant reduction relative to the Sm-lipo group (Fig. 7d and e). These results suggest that HP/L@HLS suppresses HFL1 fibroblast activation and their differentiation into ECM-hypersecreting myofibroblasts, thereby targeting fibrosis at its cellular source. Taken together, HP/L@HLS not only inhibited the MMT of MeT-5A cells but also suppressed the activation of HFL1 cells (Fig. 7f). Pleural adhesions are frequently linked to sustained inflammatory responses, wherein macrophages exert multifaceted functions [36]. We further investigated the effect of HP/L@HLS on both M1-like and M2-like macrophage phenotypes. In M1-like macrophages, HP/L@HLS treatment significantly suppressed the secretion of key pro-inflammatory cytokines, including IL-1β and TNF-α, into the culture supernatant. Specifically, compared to the Sm-lipo group, the HP/L@HLS group showed a significantly greater reduction (Fig. 7g and h). This anti-inflammatory effect was corroborated at the transcriptional level by qPCR (Fig. 7i), confirming the potent anti-inflammatory activity of HP/L@HLS. Moreover, in M2-like macrophages, HP/L@HLS markedly reduced the secretion of several critical pro-fibrotic mediators, such as TGF-β, PDGF, and CTGF, which are known to drive fibroblast activation, proliferation, and extracellular matrix deposition. Specifically, compared to the Sm-lipo group, CTGF expression was reduced by an additional 55.92% in the HP/L@HLS group (Fig. 7j). These findings establish that HP/L@HLS abrogates key macrophage-dependent fibrotic pathways through dual inhibition of M1 macrophage-mediated inflammation and M2-like macrophage-driven fibrogenesis (Fig. 7k). In summary, compared with S-lipo, M-lipo, or even Sm-lipo, the HP/L@HLS hydrogel more potently inhibited MMT and fibroblast activation. It also modulated the immune microenvironment in pleural adhesions by regulating macrophage activity, thereby exhibiting a synergistic effect on preventing pleural adhesions. Robust anti-adhesion efficacy of HP/L@HLS hydrogel in rat lung surgery model To evaluate HP/L@HLS hydrogel efficacy against postoperative adhesions, a rat model of pleural adhesion was established via mechanical pulmonary surgery. Treatment groups received Sm-lipo, HP/HLS (without Sm-lipo), or HP/L@HLS, respectively, followed by ultrasound evaluation on day 14 before euthanasia and thoracotomy (Fig. 8a). Lung ultrasonography revealed distinct B-lines and reduced lung sliding in the model group, indicating severe adhesions. In contrast, the HP/L@HLS group showed disappearance of B-lines and restoration of normal lung sliding (Fig. 8b), demonstrating a significant reduction in adhesion severity. On postoperative day 15, the extent of adhesion between the visceral and parietal pleura was evaluated and quantified. Representative images of adhesion formation are presented in Fig. 8c. The results revealed a 90.44% reduction in adhesion area in the HP/L@HLS group compared to the model group. Moreover, compared to the Sm-lipo group, the HP/L@HLS group achieved a further reduction of 73.58%, demonstrating its superior anti-adhesion efficacy.Fig. 8HP/L@HLS hydrogel serves as the tissue adhesive and anti-postoperative adhesion physical barrier for lung wound repair. (a) Pleural adhesion in rats induced by physical injury and treatment regimens using HP/L@HLS. (b) Ultrasound imaging of rat lungs after HP/L@HLS hydrogel treatment. (c) Area of pleural adhesive tissue in rats (n = 5) (mean ± SD; ****P < 0.0001). (d) H&E and Masson's trichrome staining of rat pleural adhesion tissues treated with HP/L@HLS hydrogel. (e) Abundance of miR-24-3p in rat adhesive tissues after HP/L@HLS hydrogel treatment (n = 5) (mean ± SD; **P < 0.01, ***P < 0.001 vs Model group). (f) TGFBR1 mRNA abundance in rat adhesion tissues treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; **P < 0.01 vs Model group). (g) Fibrosis-related gene mRNA abundance in rat adhesion tissues treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs Model group). (h) Immunofluorescence staining for TGFBR1, Collagen I, α-SMA, and CTGF expression in rat pleural adhesion tissues treated with HP/L@HLS hydrogel. (i) Serum IL-1β levels in rats treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; *P < 0.05, **P < 0.01 vs Model group). (j) Serum TNF-α levels in rats treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; **P < 0.01, ***P < 0.001 vs Model group). HP/L@HLS hydrogel serves as the tissue adhesive and anti-postoperative adhesion physical barrier for lung wound repair. (a) Pleural adhesion in rats induced by physical injury and treatment regimens using HP/L@HLS. (b) Ultrasound imaging of rat lungs after HP/L@HLS hydrogel treatment. (c) Area of pleural adhesive tissue in rats (n = 5) (mean ± SD; ****P < 0.0001). (d) H&E and Masson's trichrome staining of rat pleural adhesion tissues treated with HP/L@HLS hydrogel. (e) Abundance of miR-24-3p in rat adhesive tissues after HP/L@HLS hydrogel treatment (n = 5) (mean ± SD; **P < 0.01, ***P < 0.001 vs Model group). (f) TGFBR1 mRNA abundance in rat adhesion tissues treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; **P < 0.01 vs Model group). (g) Fibrosis-related gene mRNA abundance in rat adhesion tissues treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001 vs Model group). (h) Immunofluorescence staining for TGFBR1, Collagen I, α-SMA, and CTGF expression in rat pleural adhesion tissues treated with HP/L@HLS hydrogel. (i) Serum IL-1β levels in rats treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; *P < 0.05, **P < 0.01 vs Model group). (j) Serum TNF-α levels in rats treated with HP/L@HLS hydrogel (n = 5) (mean ± SD; **P < 0.01, ***P < 0.001 vs Model group). Furthermore, adhesion-related tissues were harvested 15 days post-treatment for histological evaluation. H&E staining of the model group showed substantial inflammatory cell infiltration, obscured boundaries between the lung and chest wall, and extensive connective tissue formation, confirming robust adhesion development. In contrast, the HP/L@HLS hydrogel group demonstrated excellent anti-adhesion efficacy, with only mild inflammatory responses observed (Fig. 8d). Masson's trichrome staining revealed dense collagen deposition at adhesion sites in the model group, whereas the HP/L@HLS group exhibited markedly reduced adhesion zone thickness and more loosely organized collagen fibers, further supporting the role of the hydrogel as an effective barrier against adhesion progression. Although Sm-lipo alleviated adhesion severity compared to the model group, substantial adhesions remained 15 days post-treatment, likely due to limited retention of liposomes at the target site. By comparison, the HP/L@HLS hydrogel group exhibited markedly enhanced anti-adhesion efficacy. Our analysis indicates that the hydrogel formed a robust physical barrier on the lung surface, nearly completely preventing adhesions. In addition to its superior anti-adhesion outcome, HP/L@HLS also nearly eliminated hemorrhage points and tissue defects created during model establishment, thereby promoting repair at the injury site. Importantly, no residual HP/L@HLS hydrogel was observed in the thoracic cavity on day 15, confirming its complete biodegradation. Notably, no residual hydrogel remained on day 15, confirming complete biodegradation, and body weights remained comparable to controls (Fig. S32). Mechanistically, we next investigated whether HP/L@HLS alleviates fibrosis by modulating miR-124-3p in adhesion tissues. HP/L@HLS significantly upregulated miR-124-3p (Fig. 8e) and downregulated its target gene TGFBR1 (Fig. 8f). Compared with the model group, miR-124-3p expression increased by 2.48-fold in the Sm-lipo group and 3.44-fold in the HP/L@HLS group. Correspondingly, TGFBR1 expression was reduced by 20.61% in the HP/L@HLS group compared to the Sm-lipo group. Consistent with these findings, HP/L@HLS treatment substantially reduced the expression of fibrosis-related genes in rat adhesion tissue (Fig. 8g). Immunofluorescence analysis further confirmed that miR-124-3p delivered via HP/L@HLS downregulated TGFBR1, while simvastatin concurrently inhibited CTGF expression (Fig. 8h). Additionally, increased CK19 expression and decreased Vimentin expression in the HP/L@HLS group indicated effective suppression of MMT (Fig. S33). Meanwhile, assessment of serum cytokines revealed significantly lower levels of IL-1β and TNF-α in the HP/L@HLS group, confirming the anti-inflammatory activity of the hydrogel system (Fig. 8i and j). Finally, we assessed the biosafety of the HP/L@HLS hydrogel. Organ coefficient analysis revealed no notable differences in the heart, liver, spleen, or kidneys among the groups (Fig. S34a). The elevated lung coefficient in the model group likely reflected compensatory enlargement associated with pleural adhesion formation. H&E staining of major organs showed no significant immune cell infiltration in any group (Fig. S34b). Furthermore, serum levels of AST, ALT, CREA, CK, and BUN remained comparable across all groups (Fig. S34c). Collectively, these results indicate that the bilayer powder-hydrogel combats pleural adhesion by combining physical barrier protection with multi-faceted biological interference. While the hydrogel itself physically prevents adhesion formation, the loaded miR-124-3p biologically inhibits MMT and fibroblast activation. Simvastatin contributes synergistically by concurrently suppressing these pathways and modulating macrophage function. This integrated combination of physical and biological interventions enables the hydrogel to effectively halt fibrotic progression in pleural adhesion. However, it should be noted that the final thickness of the in-situ formed hydrogel cannot yet be precisely controlled, as it depends on the amount of wound exudate or bleeding, which varies among individuals. The low swelling ratio of the hydrogel indicates that the gel does not undergo excessive volume expansion, thereby limiting the variation in thickness when the applied powder mass and wound dimensions are fixed. For future clinical translation, the use of calibrated spraying devices or metered-dose applicators could further standardize both coverage and hydrogel thickness, enhancing the reproducibility of the procedure. Conclusion In this study, we have developed a body-fluid-triggered asymmetric hydrogel integrated with dual-drug liposomes for the prevention of postoperative pleural adhesions. Unlike conventional hydrogel formulations that suffer from poor conformity to irregular, dynamic wound beds or require complex in situ injection procedures, our powder-based platform offers a paradigm shift. The dry powder can be easily sprayed onto the surgical site and rapidly forms a cohesive barrier upon contact with body fluids, perfectly adapting to the complex topography of the pleural defect. Moreover, we address the adhesion dilemma in pleural surgery by engineering an asymmetric adhesive architecture, which achieves robust wet-tissue adhesion on the visceral side to seal air leakage and bleeding, while presenting a non-fouling, anti-adhesive interface on the parietal side to physically block fibrotic encapsulation. Moving beyond passive physical barriers, this work also presents an active immunomodulatory and anti-fibrotic strategy by co-delivering miR-124-3p (inhibiting TGFBR1 signaling) and simvastatin (suppressing CTGF) to break the self-perpetuating cycle of inflammation and fibrosis at the molecular level. Despite these promising results, a primary limitation of this study lies in the translational gap between rodent models and human clinical conditions. We recognize that the rat model cannot fully reproduce the anatomical complexity, respiratory dynamics, and wound healing patterns of the human thoracic cavity. The small size of the pleural space, differences in immune responses, and the relatively short follow-up period (15 days) may limit the direct extrapolation of these findings to clinical scenarios. Therefore, future studies should validate the long-term efficacy and safety of this hydrogel system in larger animal models (e.g., porcine or canine) with longer observation windows, which will be critical for advancing toward clinical translation. In summary, this work presents a multifunctional and strategically designed powder-based hydrogel system that not only acts as a physical barrier but also actively intervenes in the biological processes driving adhesion formation. By integrating hydrogel innovation with rational drug combination, this study offers a new paradigm for preventing of postoperative adhesions and improving outcomes in thoracic surgery. Experimental section Ethics statement This study complied with the guidelines of the Helsinki Declaration. The Ethics Committee of the Ningbo Medical Center (Lihuili Hospital Affiliated to Ningbo University) approved this study protocol (No: KYSB2024L013-01). All participants provided written informed consent, and patient anonymity was preserved. All the animal experiments were conducted in strict compliance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and were approved by the Experimentation Ethics Review Committee of China Pharmaceutical University (Ethics review number: 2024-10-045). Patient samples A total of 6 subjects were included and divided into a pleural adhesion group (n = 6) and control group (n = 6). The basic information of the patient is presented in Table S1. These individuals were patients with pleural adhesion who were undergoing thoracic surgery. Preparation of the rat pleural adhesion model Six-week-old male Sprague-Dawley rats (200 ± 10g) (Huachuang Sino, China) were randomly divided into five groups, with six rats per group, and maintained under standard laboratory conditions with ad libitum access to food and water. Animals were fasted for 12 h before the experiment but had free access to water. Rats were anesthetized with 2% isoflurane (RWD, China) and placed on a surgical table. The rats were intubated with 16G plastic catheters (RWD, China). During the procedure, mechanical ventilation was provided in volume-control mode (RWD, China) at a respiratory rate of 70 breaths per minute and a tidal volume of 15-20 mL/kg. Body temperature was maintained at 37.0 °C using a heating pad to prevent hypothermia and ensure hemodynamic stability. Animals were positioned in the right lateral decubitus position, and the surgical site was disinfected and shaved under strict aseptic conditions. A left posterolateral thoracotomy was performed through the 4th intercostal space. To ensure reproducible wound dimensions and consistent powder application, a 5 mm diameter circular area was first gently marked on the left lung surface using a sterile skin punch (without cutting into the parenchyma). A uniform injury (5 mm length × 1 mm depth) was then created within this pre-marked zone. Subsequently, a pre-weighed quantity of hydrogel powder was evenly sprayed onto the wound to guarantee that the same mass of material was delivered per animal. The thoracic cavity was subsequently closed with sutures. Following resumption of spontaneous breathing, rats were extubated and housed individually. This procedure established a reproducible pleural adhesion model for subsequent analyses. The organ index was defined as the ratio of the wet weight of an organ to the body weight of an organism (expressed in g/g × 100%). Pulmonary ultrasound The rats were anesthetized with 2% isoflurane (RWD, China) before examination. The dorsal and thoracic regions were shaved to facilitate gel application for the ultrasound probe. Each rat was positioned supine, and pulmonary ultrasonography (FUJIFILM VisualSonics Vevo LAZR, USA) was performed with bilateral scanning of the dorsal and lateral chest walls of each hemithorax, extending from the posterior axillary line to the scapular line and from the scapular line to the paravertebral line. A total of four scans were performed for each rat. Histopathology staining Harvested pleural adherent tissues were fixed with 4% paraformaldehyde (Biosharp, China) for 48 h. Selected tissues were dehydrated in 75% ethanol for 12 h, embedded in paraffin, and sectioned at 5 μm thickness. Tissues were deparaffinized using a standard xylene-based protocol, followed by hematoxylin and eosin (H&E) staining and Masson's trichrome staining. Proteome Frozen pleural adhesion tissue samples (about 100 mg) were pulverized under liquid nitrogen. An appropriate amount of the resulting powder was transferred to a 2 mL high-speed microcentrifuge tube. The Proteomic data analysis was performed by Shanghai OE Biotech Co., Ltd. (Shanghai, China). All analyses were performed using a Tims TOF Pro mass spectrometer (Thermo, Bruker) equipped with an Easyspray source (Thermo, USA). Samples were loaded by a C18 column (15 cm × 75 μm) on an EASY-nLCTM 1200 system (Thermo, USA). The flow rate was set at 600 nL/min, and the linear gradient was as follows: 0∼8 min, 8% B; 8∼9 min, 100% B; 9∼12min, 100% B. Ion mobility ranged from 0.7 to 1.3 V s/cm2, and the collision energy ranged from 20 to 59eV. The MS/MS spectra were recorded from 300 to 1500 m/z. MS/MS spectra were analyzed using Spectronaut Pulsar™ 18.4 (Biognosys, Switzerland). Database research parameters are set as follows: Fixed modifications: Carbamidomethyl (C); Variable modification: Oxidation (M) and Acetyl (Protein N-term); digestion: trypsin; Precursor Qvalue cutoff: 0.01; Protein Qvalue cutoff: 0.01; Missed cleavage: 2; Quantity MS-Level: MS2. The P-value <0.05 and thresholds of fold change ≥2.0 or ≤1/2.0 were used to identify differentially expressed proteins (DEPs). Functional annotation of all identified proteins was performed using GO (http://www.blast2go.com/b2ghome; http://geneontology.org/) and KEGG pathway (http://www.genome.jp/kegg/). DEPs were further subjected for GO and KEGG enrichment analyses. Protein-protein interaction analysis was performed using the String (https://string-db.org/). Small RNA-seq Total RNA from pleural adhesion tissue was extracted using mirVana miRNA Isolation Kit (Ambion) according to the manufacturer's protocol. The small RNA sequencing and analyses were conducted by OE Biotech Co., Ltd. (Shanghai, China). The raw reads were converted into sequence data (also called raw data/reads) through base calling. Low-quality reads were filtered, and the reads with 5′ primer contaminants and poly (A) were removed. The reads lacking a 3′ adapter or insert tag, as well as shorter than 15 nt or longer than 41 nt, were excluded to obtain clean reads. Differentially expressed miRNAs were calculated and filtered with the threshold of q value < 0.05 and FC > 2 or FC < 0.5. The q value was calculated using the DEG algorithm in the R package. The targets of differentially expressed miRNAs were predicted using miRanda software with the parameters as follows: S ≥ 150, ΔG ≤ −30 kcal/mol, and demand strict 5′ seed pairing. GO enrichment and KEGG pathway enrichment analyses of differentially expressed miRNA-target-Gene were respectively performed using R based on the hypergeometric distribution. Cell culture and treatment Human pleural mesothelial cells (MeT-5A) (RRID:CVCL_3749), Human fetal lung fibroblast cells (HFL1) (RRID:CVCL_0298), mouse leukemia monocyte-macrophage cells (Raw264.7) (RRID:CVCL_0493) cell lines were purchased from by Suzhou Haixing Biosciences Co., Ltd on October 29, 2024. All cell lines were confirmed to be free of mycoplasma and were authenticated using STR profiling. MeT-5A was cultured in a Met-5A cell-specific medium (Suzhou Haixing Biosciences Co., Ltd, China). HFL1 and Raw264.7 were cultured in high-glucose Dulbecco's modified Eagle medium (DMEM) (BDBIO, China) supplemented with 10% fetal bovine serum (BDBIO, China) at 37 °C in a humidified atmosphere containing 5% CO2. MeT-5A and HFL1 cells were treated with 5 ng/mL transforming growth factor-β1 (TGF-β1) (MedChemExpress, USA) for 24 h to induce fibrotic changes. In selected experiments, MeT-5A and HFL1 cells were transfected with miR-124-3p mimics (Chemshine Biotechnology, China) and their corresponding negative controls (NCs). Cells at 60–70% confluence were transfected using Lipofectamine 2000 (Thermo Fisher Scientific, USA) according to the manufacturer's protocol. For phenotype induction, the M1 and M2 phenotypes were established by stimulating RAW 264.7 cells with LPS (100 ng/mL) (MedChemExpress, USA) and IL-4 (20 ng/mL) (MedChemExpress, USA) for 24 h. To investigate the anti-inflammatory effect of simvastatin and its influence on the expression of CTGF, M1 and M2 type macrophages were treated with simvastatin for 24 h. Subsequently, cells and culture supernatants were collected for RT-qPCR and ELISA analyses. Reverse transcription and quantitative PCR (RT-qPCR) Total RNA was extracted from Pleural adhesive tissues or cultured cells using Trizol reagent (Vazyme, China), followed by purification and quantification using a microplate reader. cDNA was synthesized from 1 μg total RNA using a cDNA Reverse Transcription Kit (Genstar, China). The cDNA was then amplified using SYBR Green Master Mix (Genstar, China) and gene-specific primers (Genscript, China) in a real-time PCR system (Thermofisher, USA). The relative expression level of the target genes and miRNAs was analyzed using the 2−ΔΔCt method, with mRNA expression normalized to β-actin and miRNA expression normalized to U6 snRNA. Primer sequences are provided in Table S4. Immunofluorescence and confocal microscopy Frozen Pleural adhesive tissue sections were fixed in 4% paraformaldehyde, permeabilized with 0.3% Triton X-100 (Beyotime, China), and blocked with 3% BSA (Solarbio, China) for 1h. Sections were incubated with primary antibodies against specific proteins at 4 °C overnight, followed by incubation with fluorophore-conjugated secondary antibodies at 37 °C for 1 h. Sections were mounted using an antifade reagent containing DAPI (Proteintech, China) and imaged with a confocal laser scanning microscope (ZEISS, Germany). Antibodies used are listed in Table S5. Dual-luciferase reporter assay Wild-type (WT) or mutant TGFBR1 3′-UTR plasmids were co-transfected with miR-124-3p mimics or negative controls into HEK293 cells. Luciferase activity was measured by adding Luciferase Assay Reagent (Vazyme, China) to the transfected cell lysates. Firefly luciferase activity was normalized to Renilla luciferase activity. Enzyme-linked immunosorbent assay IL-1β and TNF-α levels in pleural adherent tissues were determined using commercial enzyme-linked immunosorbent assay (ELISA) kits (Sinobestbio, China) according to the manufacturer's instructions. In brief, strips required for the assay were removed from the aluminum foil pouch after equilibration at room temperature for 20 min. The standard wells received 50 μL of different standard concentrations. Sample wells were first loaded with 10 μL of test sample, followed by the addition of 40 μL of sample diluent; blank wells received no additions. Subsequently, 100 μL of horseradish peroxidase (HRP)-conjugated detection antibody was added to each well of the standard and sample wells. The plate was then sealed with adhesive film and incubated at 37 °C in a water bath or incubator for 60 min. Each well was washed five times by filling with wash buffer, allowing 1-min incubation, and then aspirating and blotting dry. Next, 50 μL of substrate A and 50 μL of substrate B were added to each well. The plate was incubated at 37 °C in the dark for 15 min. Finally, 50 μL of stop solution was added, and the optical density was measured at a wavelength of 450 nm within 15 min. Preparation and characterization of Sm-lipo The Sm-lipo was prepared by film-dispersion method. Briefly, simvastatin (Bide Medical Technology, China), DOTAP, cholesterol (Aladdin, China), and DOPE at a mol ratio of 36: 31.5: 30: 2.5 were dissolved in 2 mL of chloroform. After rotation at 37 °C with a speed of 100 rpm for 30 min to form a film, it was further dried under vacuum for 4 h to ensure that the organic solvents were completely evaporated. Then, 5 mL of ultrapure water (UPW) was added to hydrate at 37 °C with a rotation speed of 15 rpm for 1 h. The mixture was ultrasonicated for 15 min in an ice bath (with a power ratio of 30%, on for 1.0 s and off for 2.0 s) to obtain S-lipo. For miR-124 loading, miR-124 was mixed with S-lipo at an N/P ratio of 7 and incubated at room temperature for 30 min to generate Sm-lipo. The stability of loaded miR124 was assessed by agarose gel electrophoresis. Particle size, polydispersity index (PDI) and Zeta potential of Sm-lipo were measured using a particle size and zeta potential analyzer (Brookhaven, USA). The content of Simvastatin encapsulated in Sm-lipo was measured by high-performance liquid chromatography (HPLC, Shimadzu, Japan), and encapsulation efficiency (EE, %) was calculated as follows: EE = (W1/W2) × 100%, where W1 is the measured simvastatin content and W2 is the initial simvastatin input. Transmission electron microscopy (TEM, Hitachi, Japan) was performed at 100 kV to visualize the morphology of Sm-lipo after negative staining with 2% phosphotungstic acid. Synthesis and characterization of HAA HAA was synthesized as previously reported with modifications. Briefly, HA (800 kDa, Aladdin, China) was fully dissolved in ultra-pure water (UPW) at 1% (w/v) concentration. Then, NaIO4 (Bide Medical Technology, China) UPW solution was slowly added to HA solution at a molar ratio of 1:1 (HA:NaIO4). After stirring for 24 h in darkness, ethylene glycol with a final concentration of 0.1 M was added to the solution, followed by stirring at room temperature for 1 h to terminate the oxidation. Subsequently, unreacted reactants were removed by dialysis against UPW for 48 h using a dialysis bag (MWCO 8000-14000 Da). The synthesized HAA was then lyophilized and stored at −20 °C until use. The structure of HAA was elucidated by 1H NMR and FT-IR. The oxidative degree of synthesized HAA was measured by Schiff Reagent Titration Method. Synthesis and characterization of 5-HDP-SF 5-HDP-SF was synthesized as followed. Briefly, SF (6.5 kDa, EFL, China) was fully dissolved in UPW at 1% (w/v) concentration, and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDCI, Bide Medical Technology, China) and N-hydroxysuccinimide (NHS, Bide Medical Technology, China) at a molar ratio of 1:1.2:2.4 (HA:EDCI: NHS) were added to the SF solution. After stirring for 30 min in darkness, 5-hydroxydopamine hydrochloride (5-HDP·HCl, Aladdin, China) was added at a predetermined molar ratio to SF (SF:5-HDP·HCl = 1:4) and stirred overnight in darkness at pH 8-9 under an argon atmosphere. Subsequently, unreacted reactants were removed by dialysis against UPW and darkness using a dialysis bag (MWCO 8000-14000 Da). The synthesized 5-HDP-SF was then lyophilized and stored at −80 °C with minimized air exposure until use. The structure of 5-HDP-SF was elucidated by 1H NMR and FT-IR. Preparation and characterization of L@HLS hydrogel Firstly, HL hydrogel was prepared as followed. Briefly, synthetic HAA and ε-poly-L-lysine·HCl (PL, 2-5 kDa, Aladdin, China) was fully dissolved in UPW at a mass ratio of 1:2, 1:3 and 1:4 (HAA:PL). After completely dissolving, 1 M NaOH was slowly added to adjust the pH to about 7. The formation and gelation of hydrogels was verified by the tube-reversing method after pH adjustment. The swelling properties of HL hydrogels were measured by the Gravimetric Method. Specifically, the hydrogels were lyophilized. The initial weights of HL hydrogel powder were recorded as W0. Afterwards, HL hydrogel powders were redissolved, and the weights of reconstructed hydrogels were recorded as Wt at different corresponding time points. The swelling rate was calculated as follows: Swollen Rate = (Wt - W0)/W0×100%. Next, HLS hydrogel was prepared. Briefly, synthetic HAA, PL, and 5-HDP-SF was dissolved in UPW at a mass ratio of 1:4:2 (HAA:PL:5-HDP-SF). After complete dissolution, 1 M NaOH was slowly added to adjust the pH to approximately 7. Hydrogel formation and gelation were verified by the tube-inversion method following pH adjustment. After lyophilization, the re-gelation performance of HLS hydrogel powder was characterized by reconstitution in UPW. Similarly, L@HLS hydrogel was prepared according to the same protocol of HLS hydrogel except for the addition of Sm-lipo (with a final Simvastatin concentration of 25 μg/mL). To evaluate the in-situ gelation ability and hemostatic performance of L@HLS hydrogel powder, a wound with a diameter of 2 mm was created in the lung lobes adjacent to the pleura of C57BL/6 mice (8 weeks, 20 ± 2 g), and 1 mg of L@HLS hydrogel powder was filled in situ, and the gelation of the powder and lung lobes hemostasis were observed. Preparation and characterization of HP hydrogel HP hydrogel was prepared as followed. Briefly, HAA, 4-arm-PEG-SH (10 kDa, Yarebio, China) was fully dissolved in UPW at a mass ratio of 1:4 (HAA: 4-arm-PEG-SH). Hydrogel formation and gelation were verified by the tube-reversing method. Subsequently, the HP hydrogel was lyophilized, and the re-gelation performance of HP hydrogel powders was characterized by reconstitution in UPW. Preparation and characterization of HP/L@HLS hydrogel The HP/L@HLS hydrogel was prepared on the basis of L@HLS and HP hydrogel powder. Firstly, the L@HLS powder was re-dissolved to obtain L@HLS hydrogel. And HP hydrogel powder was then spread onto the upper surface of L@HLS hydrogel at a mass ratio of 1:2 (HP: L@HLS). The formation of HP/L@HLS hydrogel was verified by the tube-reversing method. For morphological investigation, HP/L@HLS hydrogel was cross-sectioned to expose the internal structure and then lyophilized. After gold sputtering, the morphological structure of different HP/L@HLS hydrogel layers was examined by scanning electron microscopy (SEM) (FEI Quanta 250 FEG, Thermo Fisher Scientific, USA). Rheological properties of hydrogels The rheological properties of hydrogels were characterized by rotary rheometer (DHR-2, Anton Paar, Austria). Hydrogels were placed on the surface of the test platform with the temperature of 37 °C, the strain of 1%, and the angular velocity (ω) of 0.1-100 rad/s and measured the change of its storage modulus (G′) and loss modulus (G″). Mechanical strengths of hydrogels The mechanical strengths of hydrogels were characterized by rotary rheometer (DHR-2, Anton Paar, Austria) and tensile testing machine. For compressive stress-strain curve testing, hydrogels were placed on the surface of the test platform of rotary rheometer with the temperature of 37 °C, the frequence of 1 Hz, and the strain of 0.1%-100%, and measured the change of its stress. For tensile strength testing, hydrogels were fixed on the fixtures of the tensile testing machine and pulled to failure with a load cell of maximum 5 N capacity and a crosshead speed of 1 mm/min. Mechanical strength was calculated by dividing the maximum tensile force by the cross-sectional area (1 cm × 1 cm). Asymmetric adhesion property of hydrogels The asymmetric adhesion property of HP/L@HLS hydrogel was characterized by tissue adhesion test, with L@HLS hydrogel as a control. Specifically, the lung and its adjacent thoracic wall tissue with pleura of C57BL/6 mice (8 weeks, 20 ± 2 g) were separated. HP/L@HLS hydrogel (500 μL) was applied to the lung tissue via the adhesive L@HLS layer, and the adjacent thoracic wall tissue was attached to the opposite side of the HP/L@HLS hydrogel. Subsequently, the thoracic wall tissue was grasped with forceps to assess its detachability and the tightness of adhesion between the thoracic wall and lung tissue. The tissue adhesive strength test of hydrogels was conducted by porcine skin lap-shear test. Briefly, the adhesive ability of the hydrogels to the host tissue was evaluated by using fresh porcine skin. The skin tissue was cut into 10 mm × 40 mm rectangle and immersed into PBS before use. Then, 100 μL of HP, HLS and L@HLS hydrogels were applied onto the surface of porcine skin, and another piece of skin was placed on top. The adhesive area (A) was 1 cm × 1 cm. Subsequently, the porcine skin was placed at room temperature for 30 min. The adhesion properties were tested on the tensile testing machine with porcine skin fixed on the fixture. The maximum load force was set at 5 N, and the crosshead speed was 1 mm/min. The force (F) versus extension (δ) curves were integrated to determine the work of adhesion (Wadh) by the equation: Wadh = ∫ F dδ/A. The adhesion stability between HP and HLS or L@HLS hydrogels was similarly conducted by lap-shear testing. Briefly, hydrogels were cut into 10 mm × 40 mm rectangles. Then, HP hydrogel was attached onto the surface of HLS or L@HLS hydrogels. The adhesive area (A) was 0.5 cm × 0.5 cm. Subsequently, the adhesion properties were tested on the tensile testing machine with hydrogels fixed on the fixture. The maximum load force was set at 5 N, and the crosshead speed was 1 mm/min. The force (F) versus extension (δ) curves were integrated to determine the work of adhesion (Wadh) by the equation: Wadh = ∫ F dδ/A. In vitro drug release behavior of HP/L@HLS hydrogel The drug release behavior of HP/L@HLS hydrogel was assayed by dialysis method. Briefly, 1 mL HP/L@HLS hydrogel was sealed in a dialysis bag, and soaked in 10 mL PBS containing 30 U mL−1 HAase (500 U mg−1, BioFROXX GmbH, Germany) and 1 mM H2O2 at 37 °C with shaking at 50 rpm. At specified time points, 1 mL of medium was collected and replaced with an equal volume of fresh medium. Simvastatin concentration was quantified by HPLC. Cellular uptake and endosomes/lysosomes escape MET, Raw264.7, or HFL1 cells were cultured with Cy5-miRNA-loaded Sm-lipo with RhoB labeling, and cellular uptake of the formulation in the presence of hydrogel was assessed qualitatively by CLSM. At each time point, the cells were washed with phosphate-buffered saline (PBS) three times and stained with LysoTracker Green (for lysosomes) and Hoechst 33,342 (for nuclei). The representative gray values along the white line were analyzed by Image J. Degradation behavior of HP/L@HLS hydrogels The in vitro degradation performance of hydrogels was tested by weight-loss method to investigate their biodegradability and stability. Hydrogels were freshly prepared and weighed to record their initial weight (Ws). Then, they were soaked in PBS buffer (pH 7.4) containing 30 U mL−1 HAase and 1 mM H2O2 at 37 °C, followed by taking out and wiping away the surface water at specific time point. The weight at each point (WT) was detected until the samples were completely degraded. The calculation of the degradation rate (DR, %) is shown in the followed equation: DR=(Ws-Wt)/Ws×100%. The in vivo degradation performance of HP/L@HLS hydrogel was evaluated by subcutaneously implanting a volume of 40 μL of the HP/L@HLS hydrogel bulk stained with red dye into the dorsal side of C57BL/6 mice (8 weeks, 20 ± 2 g). The volume of remaining hydrogel was observed and recorded at specific time point until the sample was completely degraded. Subcutaneous injection of saline mixed with red dye served as a control. Hemostatic activity test In the mouse tail amputation model, C57BL/6 mice (male, 12 weeks old, 25–30 g) were randomly assigned to three groups. A pre-weighed filter paper was positioned beneath the tail. Approximately 5 cm of the tail was amputated using surgical scissors, after which either HP or HLS (10-15 mg) was sprayed onto the bleeding wound. Hemostasis time was defined as the interval from the initiation of bleeding until the point when no further expansion of the blood stain was observed on the filter paper. The blood-saturated filter paper was then weighed to quantify blood loss. An untreated group served as the control. A commercially available anti-adhesion membrane was used as the positive control. In the mouse hepatic hemorrhage model, C57BL/6 mice (male, 12 weeks old, 25–30 g) were anesthetized and secured on a surgical corkboard tilted at approximately 45°. Following abdominal incision, the liver was exposed and surrounding serous fluid was carefully removed. A pre-weighed filter paper was placed under the liver. A standardized wound was created using an 18G needle to induce hepatic bleeding. Subsequently, HP or HLS (10-15 mg) was immediately applied to the bleeding site. Hemostasis time was recorded as the duration from injury until cessation of bleeding, indicated by no further increase in the blood stain area on the filter paper. The filter paper containing absorbed blood was then weighed to determine blood loss. An untreated group was used as the control. A commercially available anti-adhesion membrane was used as the positive control. ROS scavenging capability Cell suspension (2 × 105 cells/mL) was seeded into confocal dishes at 1 mL per well. RAW264.7 cells were pre-incubated in the dishes for approximately 3 h to allow for adequate adhesion. Subsequently, the culture medium was aspirated, and each well was treated with 1 mL of LPS solution (1 μg/mL), followed by incubation for 6 h. Following this, the supernatant was discarded, and the cells were washed twice with PBS. Then, 1 mL of hydrogel extract was added to each well, and the cells were incubated for 24 h. After discarding the supernatant and washing the cells twice with PBS, 1 mL of 5 μM DCFH-DA solution was added to each well and incubated for 15 min. After incubation, the solution was removed, and the cells were washed three times with PBS. Next, 1 mL of Hoechst 33,342 (Solarbio, China) solution (1 μg/mL) was added to each well and incubated for 10 min, followed by three washes with PBS. Imaging was performed using a confocal laser scanning microscope (ZEISS, Germany). Protein adsorption assessment The adsorption of bovine serum albumin (BSA) onto the HP and HLS were quantified using the bicinchoninic acid (BCA) assay. Briefly, 400 μL of the prepared hydrogel was placed in a 48-well plate and incubated with 400 μL of protein solution (1 mg/mL) at 37 °C for 2 h. Subsequently, the hydrogel surfaces were rinsed with PBS to remove unbound proteins. To desorb the adsorbed proteins, 400 μL of sodium dodecyl sulfate (SDS, 1 wt%) was added to each well, followed by sonication for 1 h. The absorbance of the resulting supernatant was measured at 562 nm using a microplate reader with a BCA kit. The amount of protein adsorbed of the hydrogel was determined by comparing the measured absorbance against a standard curve prepared with known protein concentrations. Evaluation of hydrogel adhesion to injured lung tissue The injured lung surface was covered with the hydrogel loaded with RhoB-containing liposomes. The tissue was then embedded in OCT compound and sectioned using a cryostat. Sections were mounted using an antifade reagent containing DAPI (Proteintech, China) and imaged with a confocal laser scanning microscope (ZEISS, Germany). Hydrogel adhesion to lung under simulated breathing C57BL/6 mice (male, 12 weeks old, 25–30 g) were euthanized via cervical dislocation and subjected to a thoracotomy. The lungs were exposed by careful dissection of extra tissue and ribs. A standardized wound was inflicted on the lung parenchyma, followed by immediate application of the HLS hydrogel. To assess hydrogel performance under dynamic conditions, the trachea was cannulated and connected to a 1 mL syringe fixed with a suture. Controlled rhythmic aspiration was then administered through the syringe to simulate respiratory motion, facilitating the observation of hydrogel adhesion and wound sealing. In vitro antibacterial activity assay Bacterial suspensions of S. aureus and E. coli in the logarithmic growth phase were adjusted to a concentration of 1 × 106 CFU/mL. A volume of 100 μL of each bacterial suspension was mixed with 100 μL of the corresponding test solution in a 96-well plate. After incubation at 37 °C for 12 h, the optical density (OD) at 600 nm was measured using a microplate reader. Subsequently, the mixtures from each well were serially diluted, plated onto agar plates, and incubated at 37 °C for 24 h. Colony-forming units (CFU) were then counted, and images of the plates were captured under clear field conditions. Hemolysis assay Fresh anticoagulated whole blood (from mouse) was centrifuged, and the supernatant was removed. The collected red blood cells (RBCs) were washed three times with normal saline and finally resuspended to prepare a 4% (v/v) RBC suspension. Each extract of hydrogel material was incubated with an equal volume of the RBC suspension. Normal saline and deionized water were used as the negative and positive controls, respectively. After incubation at 37 °C for 2 h, the mixtures were centrifuged at 3000 rpm for 10 min. The absorbance of the supernatant was measured at 545 nm using a microplate reader. Statistical analysis Data are expressed as means ± standard deviation (SD). Statistical significance was assessed using a two-tailed unpaired Student's t-test for comparisons between two sets of values or ANOVA for comparisons among multiple groups, utilizing GraphPad Prism 8. Exact P values are provided in the Figures or their legends. No exclusion criteria were applied in the experimental design for this study. CRediT authorship contribution statement Ni Li: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Zherui Shao: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Nuo Cheng: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Yongbin Wang: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft. Bin Wu: Conceptualization, Project administration, Writing – review & editing. Bingchuan Hu: Investigation, Methodology. Qinning Wang: Investigation, Methodology. Yanggang Dan: Validation, Visualization. Yuwei Zhou: Validation, Visualization. Kanghao Zhu: Methodology, Visualization. Renyuan Xiao: Methodology, Visualization. Caoyun Ju: Conceptualization, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing. Cong Wang: Conceptualization, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing. Dawei Zheng: Funding acquisition, Project administration, Supervision, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.