In situ Gel–Nano transformable hydrogel for therapeutic intervention in refractory inflammation-related trauma Authors: Jingxuan Xu, Yifeng Zhu, Guangjian Fan, Yaoyao Li, Yuxiang Zhang, Bowen Li, Wenwei Cai, Chenyue Zhan Published: 2026 ABSTRACT AbstractIn this study, we introduce a novel injectable hydrogel, termed hDAGel, which exhibits a unique gel-to-nano (Gel–Nano) configuration inversion capability in response to local pathological microenvironments. The hydrogel is formulated with a high loading density of dopamine, serving the dual function: an injectable dual-component drug-loading gel and triggering the in situ formation of drug-loaded polydopamine-based nanoparticles upon dopamine release and oxidative polymerization. This Gel–Nano transformation enables sustained and targeted drug delivery precisely at inflamed or injured sites. Moreover, hDAGel effectively scavenges excess reactive oxygen species (ROS) within inflamed tissues, alleviating oxidative stress and modulating immune responses. Upon local administration, such as intracerebral injection for cerebral ischemia or topical/intradermal application for allogeneic skin transplantation, hDAGel persists at the target site and dynamically responds to microenvironmental stimuli, releasing therapeutic nanoparticles that suppress inflammation and immune overactivation. Collectively, hDAGel integrates the advantageous features of injectable, adhesive hydrogels with the precise and controlled release of nanoparticle-delivered therapeutics via Gel–Nano transformation. This multifunctional platform holds great promise for treating complex inflammation-associated conditions, including neuroinflammation and transplant rejection, by providing localized, sustained, and responsive therapy with enhanced antioxidant and immunomodulatory effects.Keywords: Injectable hydrogel, Gel–Nano transformation, Dopamine-based materials, Inflammation-responsive drug delivery, Reactive oxygen species scavenging, Immunomodulatory Introduction In situ drug administration has emerged as an effective and safer therapeutic strategy for treating localized inflammatory diseases by enabling direct delivery of therapeutic agents to lesion sites at high local concentrations [[1], [2], [3]]. Compared to systemic delivery, in situ approaches such as topical spraying onto surgical wounds or direct intracerebral injection allow for enhanced targeting of inflamed tissues, reduced systemic exposure, and improved therapeutic efficacy [4,5]. However, traditional local formulations suffer from several limitations. Many require sustained retention at the target site, which can be difficult in dynamic physiological environments such as bleeding wounds or cerebrospinal fluid circulation [6]. In addition, frequent re-application or high local drug concentrations may lead to reduced patient compliance or local toxicity [7]. To address these challenges, a variety of advanced drug delivery systems have been developed for inflammatory disease management, particularly in complex models like allogeneic skin transplantation, where graft rejection triggers local immune activation, or cerebral ischemia, where inflammation contributes to secondary brain injury [[8], [9], [10], [11], [12]]. Among these systems, inflammation-responsive nanoparticles and hydrogels are gaining attention [[13], [14], [15], [16], [17], [18], [19]]. These smart materials are designed to remain at the application site, respond to the inflammatory microenvironment, and release therapeutic agents in a controlled manner. Specifically, injectable or sprayable hydrogels designed for in situ administration have demonstrated significant potential [20,21]. They improve drug biocompatibility, minimize systemic toxicity, and allow for better patient compliance or procedural convenience. Moreover, inflammation-targeted nanoparticle systems enhance drug accumulation at the lesion site, prolong release duration, and stabilize sensitive drugs under inflammatory conditions [[22], [23], [24], [25]]. Substantial efforts have been devoted to integrating the complementary advantages of injectable hydrogels and targeted nanoparticles [26,27]. Yet, simple mixing or physical fusion of these components frequently introduces critical limitations, such as diminished drug-loading efficiency, increased formulation complexity, and disruption of the intrinsic hydrogel network. These challenges underscore the need for a fundamentally redesigned hydro-nano platform—one that unites structural fidelity with programmable therapeutic functionality, rather than relying on conventional hybridization strategies. Dopamine is a potent natural antioxidant, and its oxidative polymerization products, polydopamine nanoparticles (one kind of melanoidin-like particles), exhibits significant anti-inflammatory and neuroprotective properties [[28], [29], [30]]. Recent evidence highlights dopamine's immunomodulatory roles in various tissues, including the central nervous system and transplanted skin. In cerebral ischemia, excessive reactive oxygen species (ROS) and neuroinflammation contribute to secondary brain injury [31]; dopamine has been shown to attenuate microglial activation and promote neuronal survival through modulation of inflammatory signaling and oxidative stress. Similarly, in skin transplantation, dopamine-mediated immunoregulation may help suppress graft rejection by enhancing regulatory T cell (Treg) induction and inhibiting dendritic cell (DC) activation within the local immune microenvironment [32,33]. Therefore, incorporating dopamine into a newly engineered hydrogel system offers a compelling strategy for treating refractory inflammation. Dopamine provides a unique chemical versatility—undergoing oxidation-mediated polymerization, enabling dynamic covalent interactions, and conferring strong adhesive and tissue-interactive properties. When integrated into a rationally designed hydrogel matrix, these attributes can endow the material with programmable responsiveness, enhanced drug retention, and improved localization within inflamed microenvironments. Herein, we have developed an intelligent gel-to-nano (Gel–Nano) transformative antioxidative injectable hydrogel with an exceptionally high dopamine content (Fig. 1). This hydrogel, named hDAGel, exhibits a unique Gel–Nano transformation ability under pathological microenvironmental triggers such as pH changes or oxidative stress, converting from a bulk hydrogel into uniform polydopamine-based nanoparticles. This Gel–Nano transition not only facilitates targeted drug delivery but also enhances the hydrogel responsiveness and therapeutic efficacy. Compared with previously reported injectable hydrogels, hDAGel introduces an oxidation-triggered Gel–Nano transformation mechanism that enables in situ formation of drug-loaded nanoparticles during hydrogel degradation. This mechanism integrates ROS-responsive drug release with ROS scavenging in a single injectable platform. Composed of PVA and a microenvironment-sensitive linker (hDAlinker), hDAGel enables controlled release of dopamine, a potent natural antioxidant, which upon oxidation forms polydopamine nanoparticles with strong anti-inflammatory and antioxidative properties. This transformation mechanism allows hDAGel to dynamically adapt to disease-specific microenvironments, such as those found in cerebral ischemia and allogeneic skin transplantation. In cerebral ischemia models, intracerebrally administered hDAGel undergoes Gel–Nano transformation triggered by the oxidative microenvironment, releasing antioxidative nanoparticles that mitigate neuroinflammation and protect neuronal tissue from ischemia-reperfusion injury. Likewise, in skin transplantation, topical or intradermal application of hDAGel leads to adhesion on inflamed graft tissues and local Gel–Nano conversion, resulting in sustained dopamine and drug release that modulates immune responses and promotes graft acceptance. Furthermore, the Gel–Nano transition improves hydrogel retention and penetration at inflamed sites, while the negatively charged hydrogel network selectively adheres to inflamed tissues, maximizing localized therapeutic effects. Overall, the Gel–Nano transformative property of hDAGel represents a promising strategy for smart drug delivery, enabling site-specific, responsive therapy tailored for oxidative stress and inflammation in both neurological injury and transplantation contexts. Schematic diagram. (A) Schematic overview of the preparation of hDAGel and its gel-to-nano transformation (Gel–Nano Trans). (B) Schematic illustration of the therapeutic applications of hDAGel in allogeneic skin transplantation (left) and ischemic stroke (right). The hydrogel spray protects skin grafts by providing antioxidation, controlled drug release, and immunomodulation, reducing pro-inflammatory cytokines (IFN-γ) and promoting graft survival (left); upon injection, the Gel–Nano transformation enables controlled release of growth factors (VEGF and BDNF), facilitating cellular regeneration, vascular repair, and neuroprotection. This schematic diagram highlights the dual functionality of the hDAGel in oxidative stress modulation and therapeutic delivery. To demonstrate the therapeutic efficacy of hDAGel, we loaded it with tacrolimus for the allogeneic skin transplantation model and with brain-derived neurotrophic factor (BDNF) and vascular endothelial growth factor (VEGF) for the cerebral ischemia model. The drug-loaded hDAGel was applied via topical or intradermal administration on transplanted skin and via intracerebral injection in the ischemic brain, respectively. Compared to free drugs, hDAGel exhibited significantly prolonged residence time at the target sites, enabling sustained and localized drug release that enhanced therapeutic effects. Thanks to its negative surface charge, hDAGel preferentially adhered to inflamed graft tissues and ischemic brain regions, improving drug retention and bioavailability. Therapeutic evaluations demonstrated that in situ administration of these drug-loaded hDAGel formulations achieved superior outcomes, including immunosuppression for graft acceptance and neuroprotection with enhanced angiogenesis and neuronal survival. Our findings establish a novel proof of concept for hDAGel as an adaptable injectable hydrogel platform for targeted treatment of inflammatory and ischemic conditions in both allogeneic skin transplantation and cerebral ischemia. Results and discussion Synthesis of hDAlinker and preparation of hDAGel The hDAlinker was synthesized in four sequential reactions. The synthetic route is shown in Fig. S1. In the first step, the amine group of dopamine was protected with a Boc group to prevent undesired side reactions. In the second step, the phenolic hydroxyl groups of dopamine were conjugated with phenylboronic acid groups that respond to reactive oxygen species (ROS). After that, the Boc protecting group was removed to restore the free amine. To enhance the hydrophilicity of the compound, the amine group of dopamine was ionized into trifluoroacetate in the final step of the synthesis process. This modification improves the solubility of the compound in aqueous environments. The structures of the related products were confirmed by 1H NMR spectroscopy, and the results are presented in Fig. S2–S5. The final compound was named hDAlinker. As shown in Fig. 2A and Table S1, different PVA/hDAlinker formulations were evaluated to optimize the hydrogel composition based on their gelation efficiency. Increasing the hDAlinker content significantly improved gelation efficiency due to the increased density of dopamine-mediated dynamic interactions. The formulation containing 40 wt% hDAlinker and 10 wt% PVA (molecular weight ≈ 47,000 Da) exhibited the highest gelation efficiency (>80%) and formed a stable hydrogel. Therefore, this formulation was selected as the optimized hDAGel composition for subsequent experiments. Considering that the high density of dopamine moieties is critical for enabling the dual functions of antioxidant protection and catechol-mediated dynamic interactions, the theoretical dopamine content within hDAGel was quantified. The hDAlinker molecule possesses a molecular weight of 422.19 g mol−1, containing one dopamine unit (154.09 g mol−1), corresponding to a dopamine contribution of 36.5 wt% within the linker structure. Since hDAlinker accounts for 40 wt% of the hDAGel network, the calculated dopamine unit content in hDAGel reaches approximately 14.6 wt%, equivalent to 0.947 mmol g−1 (approximately 0.95 M assuming a hydrogel density close to 1 g mL−1). This high density of dopamine motifs provides abundant catechol groups for ROS scavenging and reversible dynamic interactions, thereby contributing to the adaptive properties of hDAGel. Following this, the morphology of the dried scaffold network was observed by scanning electron microscopy (SEM), revealing a homogeneous porous structure (Fig. 2B). We then performed lap shear tests to assess the tissue adhesive properties of hDAGel. As shown in Fig. S6, fresh porcine skin was utilized as a test surface for the simulation of human skin tissue as previously reported [34]. Characterization, Gel–Nano transformation, and antioxidant activity of hDAGel. (A) Images of PVA solution and hDAGel formed after hDAlinker addition, showing the formation of hDAGel at the optimized formulation (40 wt% hDAlinker and 10 wt% PVA). (B) Morphological observation of hDAGel by scanning electron microscopy (SEM). Scale bars: 10 μm (left), 250 nm (right). (C) Quantification of the remaining percentage of hDAGel during degradation in 1 mM H2O2 or PBS over 5 days (n = 3). (D) Representative SEM images of hDAGel showing Gel–Nano transformation during the degradation process in 1 mM H2O2. Scale bars = 5 μm. (E) Quantification of nanoparticle density in SEM images at different degradation time points. Nanoparticle density was quantified from multiple randomly selected SEM fields (n = 3) and is expressed as particles/μm2. (F) Representative transmission electron microscopy (TEM) image of nanoparticles formed by Gel–Nano transformation of the hDAGel (scale bar = 0.2 μm). (G) Hydrodynamic diameter distribution of nanoparticles formed from the Gel–Nano transformation of hDAGel, measured by dynamic light scattering (DLS). (H) Statistical analysis of the hydrodynamic diameter of nanoparticles during the 3-day continuous degradation process (n = 3). (I) H2O2 clearance efficiency of hDAlinker (2 μM and 5 μM) and hDAGel, compared with PBS as control (n = 3). (J) PTIO levels after treatment with PBS, hDAlinker (2 μM and 5 μM), or hDAGel, indicating radical scavenging ability (n = 3). Data are presented as mean ± SD. Notable differences between groups are denoted by *P < 0.05, **P < 0.01, ***P < 0.001, and ns indicates no significant difference. Gel–nano transformation and ROS-scavenging activity of hDAGel during oxidative degradation To investigate the degradation performance of hDAGel under oxidative conditions, the responsive degradation test of hDAGel was conducted in hydrogen peroxide (H2O2) solution (200 μM and 1 mM), compared with phosphate-buffered saline (PBS, pH 7.4). As shown in Fig. 2C, hDAGel exhibited slight degradation in PBS over a period of 5 days; however, in the presence of 1 mM H2O2, the gel underwent gradual and significant degradation. This result indicated the rapid degradation of hDAGel under oxidative conditions, likely due to breaking borate bonds and the release of dopamine. These findings confirm that the designed hDAGel achieved the intended controlled degradation performance. Interestingly, we observed that the hDAGel degradation process was accompanied by a dark color change (Fig. S7), indicating that melanin-like byproducts were generated during the degradation process. Based on this, we hypothesized that under oxidative conditions, the dopamine molecules released from gel degradation undergo further oxidation, polymerization, and self-assembly into melanin-like polymer nanoparticles. To validate the above hypothesis, the morphology of the degraded gel samples was observed using SEM. As shown in Fig. 2D, we successfully captured and observed the morphological changes during the gel degradation process. With prolonged degradation time, the gel gradually transitioned from a cross-linked network state to a nanoparticle state. We refer to this gel-to-nano structural transformation as the “Gel–Nano Trans”. The collapsed grid structure and the presence of particles in the images indicate the degradation of hDAGel and its morphological transformation. Quantitative analysis further showed a progressive increase in nanoparticle density during oxidative degradation (Fig. 2E). Simultaneously, transmission electron microscopy (TEM) imaging captured the self-assembled melanin-like nanoparticles, which exhibited a uniform spherical morphology (Fig. 2F). Notably, similar observations have not been reported in earlier studies investigating the degradation of medical injectable hydrogels. Furthermore, the occurrence of this structural transformation was confirmed by measuring the hydrodynamic size of the nanoparticles in the degradation solution, which showed significant changes (Fig. 2G and H). A pronounced Tyndall effect was also observed. (Fig. S8). To quantitatively evaluate the Gel–Nano transformation, dopamine recovery in the nanoparticle fraction was measured at different time points during incubation with 1 mM H2O2. As shown in Fig. S9, dopamine recovery progressively increased over time. This time-dependent increase was consistent with the gradual degradation of the hydrogel network and the progressive formation of melanin-like nanoparticles observed by SEM, TEM, and DLS. These results provide quantitative support for the oxidation-triggered Gel–Nano transformation of hDAGel. In addition to size and morphology characterization, the surface charge of the transformed nanoparticles was evaluated by zeta potential measurement. As shown in Fig. S10, the nanoparticles exhibited a negative zeta potential of approximately −15.6 mV, indicating moderate colloidal stability in aqueous suspension. Catechol and phenylboronic acid have emerged as promising antioxidants, supported by numerous clinical and preclinical studies [35]. The morphological transformation test reveals that hDAGel can effectively interact with reactive oxygen species (ROS). Thus, we conducted an evaluation of antioxidant properties of hDAGel against various types of ROS. For H2O2 clearance, hDAlinker exhibited anti-oxidation properties. Remarkably, hDAGel demonstrated stronger H2O2 clearance than the individual monomer, achieving a high clearance efficiency (Fig. 2I). To further demonstrate the broad applicability of hDAGel in ROS clearance, we conducted radical scavenging tests using PTIO as a coloration radical indicator. As depicted in Fig. 2J, while hDAlinker displayed limited anti-radical properties, hDAGel showcased a notably more robust scavenging effect on PTIO radical. To further examine whether the Gel–Nano transformation occurs under PTIO treatment, dynamic light scattering (DLS) measurements were performed at different time points. As shown in Fig. S11, DLS revealed the presence of nanoparticles generated from hDAGel under PTIO treatment. Additionally, we compared the anti-oxidation properties of hDAGel with those of the strong natural antioxidant Trolox using ABTS as a free radical source. The blue-green color of ABTS radical solution faded immediately after the addition of hDAGel, suggesting its effective radical scavenging activity; however, Trolox showed a low radical clearance ability, illustrating the advanced antioxidant performance of hDAGel (Fig. S12). Cytotoxicity and cellular antioxidant effects of hDAGel Low cytotoxicity and reliable safety are crucial for the clinical use of an injectable hydrogel. Therefore, we performed assessments on various related cell lines, including human umbilical vein endothelial cells (HUVECs) and murine fibroblast L929 cells. The viability of these cells was analyzed using the CCK-8 assay. Due to the controlled release mechanism, 10 μL of hDAGel in a 100 μL culture system exhibited negligible cytotoxicity in L929 cells compared with the PBS control (Fig. 3A). Moreover, different concentrations of PVA and hDAlinker also exhibited minimal cytotoxicity (Fig. 3B). As illustrated in Fig. S13, hDAGel demonstrated low cytotoxicity toward HUVECs. These findings were crucial for subsequent cellular and in vivo experiments. In addition, we observed that the application of 1 mM H2O2 had a significant detrimental impact on the aforementioned cell lines. It resulted in a decrease in cell viability of over 70%. However, when hDAGel was co-cultured with the cells before H2O2 addition, it effectively protected the cells from the harmful effects of H2O2, leading to cell rescue (Fig. 3C). Apoptosis was further quantified by flow cytometry, as presented in Fig. 3D and E. Treatment with hDAlinker or hDAGel significantly mitigated H2O2-induced apoptosis in L929 cells, providing compelling evidence for the antioxidant protective effect of hDAGel. Additionally, we performed a live/dead assay using calcein/propidium iodide (calcein/PI) staining. The results, depicted in Fig. 3F and S14, show that the presence of 1 mM H2O2 had a significant impact on cell viability, resulting in a decrease in viable cells. However, this decrease was attenuated by hDAGel treatment, and hDAGel demonstrated a higher cell survival rate than hDAlinker treatment, indicating its superior protective effect on cell viability. Cytotoxicity, cellular antioxidant, and anti-inflammatory effects of hDAGel. (A) Cell viability of L929 cells incubated with 10 μL of hDAGel or PBS (control) in a 100 μL culture system, as determined by CCK-8 assay (n = 8). (B) Cell viability of L929 cells incubated with different concentrations of PVA or hDAlinker, measured by CCK-8 assay (n = 8). (C) Cell viability rescue profile of L929 cells pretreated with hDAlinker or hDAGel, followed by H2O2 (1 mM) treatment as follows: Control (PBS treatment); H2O2 (1 mM); H2O2 + hDAlinker (5 μM); H2O2 + hDAlinker (10 μM); and H2O2 + hDAGel (n = 5). (D–E) Flow cytometry analysis of apoptosis in L929 cells under different treatment conditions using Annexin V-FITC/PI staining, and the apoptosis ratio was quantified (n = 3). (F) Live/dead (Calcein-AM/PI) staining of L929 cells under different treatment conditions (scale bar = 20 μm). (G) DCF-DA fluorescent staining and fluorescence intensity quantification representing ROS levels in L929 cells in different treatment groups (scale bar = 20 μm). (H–I) Flow cytometry analysis and quantification of the percentage of CD86+ cells (M1 macrophages) in bone marrow–derived macrophages (BMDMs) from C57BL/6 mice following LPS stimulation and co-culture with hDAlinker or hDAGel for 24 h, with control (M0), IL-4 (M2), and LPS (M1) as reference groups (n = 3). (J) Quantification of IFN-γ secretion by BMDMs co-cultured with different treatments using ELISA (n = 3). Data are presented as mean ± SD. Notable differences between groups are denoted by *P < 0.05, **P < 0.01, ***P < 0.001, and ns indicates no significant difference. To examine whether the cellular protection provided by hDAGel is due to its ROS clearance capability, we analyzed the intracellular ROS levels in L929 cells after various treatments using the DCF-DA fluorescence probe. Fig. 3G and S15 demonstrate the results of this analysis. The addition of H2O2 substantially elevated the intracellular ROS concentration, as indicated by the enhanced green fluorescence. However, when the cells were co-cultured with hDAGel, the intracellular ROS levels significantly decreased. hDAlinker (10 μM) also reduced intracellular ROS levels. In vitro anti-inflammatory properties of hDAGel In order to investigate the modulatory properties of hDAGel in inflammation reactions, we conducted lipopolysaccharide (LPS) stimulation experiments using bone marrow-derived macrophages (BMDMs) derived from C57BL/6 mice. Flow cytometry was employed to analyze the differentiation of BMDMs upon stimulation. As shown in Fig. 3H and I, LPS stimulation induced the differentiation of the majority of unpolarized BMDMs into M1 macrophages (CD86+) within 24 h. However, co-culture with hDAGel notably decreased the polarization of BMDMs into M1 macrophages, indicating its strong ability to inhibit M1 macrophage polarization. Furthermore, ELISA was performed to confirm that hDAGel treatment decreased the secretion of the pro-inflammatory cytokine interferon-γ (IFN-γ) in LPS-stimulated BMDMs (Fig. 3J). This further supports the anti-inflammatory effect of hDAGel in reducing cytokine production. In vivo subcutaneous retention and oxidation-responsive degradation of hDAGel To evaluate the suitability of hDAGel for the inhibition of allogeneic skin graft rejection, we first analyzed its in vivo subcutaneous retention and response to ROS in the microenvironment. hDAGel or PVA (control group) was subcutaneously injected into C57BL/6 mice on day 0. Then, 10 μL of PBS (ROS- group) or 100 μM of H2O2 (ROS+ group) was injected on day 1. Three days post-procedure, the skin of mice was dissected for photographic documentation, and the residual gel was retrieved for subsequent analysis. As shown in Fig. 4A–C, under normal physiological conditions of the subcutaneous environment, hDAGel rapidly formed a gel and exhibited sustained morphological stability. Conversely, when exposed to the ROS, the gel underwent swift degradation, accompanied by a decrease in volume and weight over 3 days. To further corroborate the oxidation-triggered degradation and subsequent cargo release kinetics of hDAGel, the hydrophobic near-infrared fluorophore Cy5.5 was encapsulated within the hydrogel matrix (referred to as Cy5.5@hDAGel). Free Cy5.5 and Cy5.5@hDAGel were administered via subcutaneous implantation into C57BL/6 mice, respectively. The spatiotemporal distribution and diffusion kinetics of the released fluorophore were then longitudinally monitored and quantified using a calibrated in vivo imaging system (IVIS). As shown in Fig. S16, the unencapsulated free Cy5.5 rapidly diffused across most of the dorsal region within 24 h after subcutaneous injection. In contrast, dye encapsulated within hDAGel (Cy5.5@hDAGel) exhibited significantly reduced diffusion. Concurrently, in skin tissue pre-treated under oxidative conditions, the responsive degradation of hDAGel facilitated gradual dye release, leading to a recovery in diffusion rate. These findings robustly confirm the efficient encapsulation capacity and stimuli-responsive release functionality of hDAGel, thereby providing a solid foundation for subsequent experiments. In vivo subcutaneous retention, ROS-responsive degradation, and therapeutic application of hDAGel in allogeneic skin transplantation. (A–C) In vivo subcutaneous retention of hDAGel in ROS+ (H2O2), ROS− (PBS) and PVA (control) groups of C57BL/6 mice. Representative images of the treated site at 3 days are presented in (A). Residual gel volumes are shown in (B), and the gel weight is quantified in (C) (n = 3 per group). (D) Representative images of skin allografts at day 7 post-transplantation in different treatment groups: (i) Control (PBS treatment), (ii) Tac (tacrolimus, intraperitoneal injection, 0.6 mg/kg), (iii) Tac (tacrolimus, subcutaneous injection, 2.1 mg/kg), (iv) hDAGel (sprayed onto the recipient wound, 100 μL), and (v) Tac@hDAGel (tacrolimus-loaded hDAGel, sprayed onto the recipient wound, tacrolimus 2.1 mg/kg, 100 μL). Three representative images are presented for each group. (E) Graft rejection scores of skin allografts in different treatment groups (n = 10). Rejection was graded on a scale of 0 to 5, with higher scores indicating greater rejection. (F) Kaplan–Meier survival curves of skin allografts in different treatment groups (n = 10). Statistical significance was assessed by log-rank test (p < 0.05, Control vs. Tac@hDAGel). (G) Representative H&E staining of skin graft tissues from different treatment groups. Scale bars = 100 μm. Notable differences between groups are denoted by *P < 0.05, **P < 0.01, ***P < 0.001, and ns indicates no significant difference. Tacrolimus-loaded hDAGel preparation and its anti-inflammatory and anti-immune rejection effects in allogeneic skin transplantation In clinical practice, topical formulations such as tacrolimus ointment and corticosteroid creams (e.g., Protopic, clobetasol) are currently available [36]; however, their therapeutic efficacy in allogeneic skin transplantation remains limited [37]. Frequent reapplication is often required, and due to their highly lipophilic properties and the poor permeability of the stratum corneum, drug penetration into the grafted skin is inconsistent, leading to variable local tissue concentrations and suboptimal immunomodulatory effects. The utilization of hDAGel offers a promising strategy to overcome these limitations by enabling localized and sustained topical immunosuppression. Tacrolimus (Tac) can be incorporated into hDAGel through a hydrophobic affinity-driven encapsulation process, thereby enhancing its skin penetration and enabling controlled release. Furthermore, an additional Gel–Nano transformation step was introduced in tacrolimus-loaded hDAGel (referred to as Tac@hDAGel), which further improved the precise targeting of hydrophobic drug domains and optimized the release kinetics, distinguishing it from conventional sustained-release hydrogels. To fabricate Tac@hDAGel, tacrolimus was first dissolved in a PVA solution using DMSO (<10%)-assisted hydrotropy to enhance drug solubility. The resulting Tac@PVA solution was then mixed with the hDAlinker solution, leading to rapid formation of Tac@hDAGel within a short period of time. A representative SEM image of Tac@hDAGel is shown in Fig. S17. To further investigate the oxidation-responsive release behavior of Tac@hDAGel, cumulative tacrolimus release was evaluated under physiological (PBS, pH 7.4) and oxidative (1 mM H2O2) conditions. As shown in Fig. S27A, tacrolimus was released more rapidly in the presence of H2O2 than in PBS, indicating oxidation-triggered drug release. This release behavior was accompanied by a progressive increase in nanoparticle formation (Fig. S27D). Furthermore, cumulative tacrolimus release was strongly correlated with nanoparticle formation under both PBS and oxidative conditions (Fig. S27F–G), suggesting that oxidation-induced Gel–Nano transformation was closely associated with drug release. Tac@hDAGel was subsequently applied in an allogeneic skin transplantation model of acute graft rejection. Briefly, a circular full-thickness wound was created on the dorsum of recipient C57BL/6 mice, followed by spraying the Tac@PVA and hDAlinker solutions directly onto the wound site. The hydrogel formed in situ within seconds, creating a stable and adhesive layer on the skin surface. Finally, donor skin allografts from BALB/c mice were carefully placed and sutured onto the Tac@hDAGel adhesive interface to complete transplantation. The appearance of the skin allografts was compared among five groups: (i) control group (PBS treatment), (ii) Tac (intraperitoneal, 0.6 mg/kg, administered every other day until day 7), (iii) Tac (subcutaneous, 2.1 mg/kg, administered every other day until day 7), (iv) hDAGel (100 μL, sprayed onto the recipient wound), and (v) Tac@hDAGel (tacrolimus 2.1 mg/kg, 100 μL, sprayed onto the recipient wound). As shown in Fig. 4D, at 7 days post-transplantation, most skin allografts in the control group exhibited severe rejection characterized by marked shrinkage and detachment from the wound bed. In contrast, skin allografts treated with Tac@hDAGel remained largely intact and firmly attached to the wound area. Compared with the free drug groups (Tac i.p. and Tac s.c.), the Tac@hDAGel group displayed healthier skin appearance, indicating a significantly enhanced effect in alleviating rejection. Notably, owing to the strong antioxidant capacity of hDAGel, the hDAGel-treated group also demonstrated partially improved graft appearance compared with the control group. However, Tac@hDAGel achieved significantly greater therapeutic efficacy than hDAGel alone, indicating an additional anti-rejection effect provided by the sustained local delivery of tacrolimus. Consistently, graft rejection scores (Fig. 4E) and survival curves (Fig. 4F) revealed that Tac@hDAGel treatment substantially reduced graft rejection and significantly prolonged skin allograft survival. Furthermore, histological examination demonstrated that skin grafts treated with Tac@hDAGel maintained a well-organized tissue structure with significantly reduced inflammatory infiltration and graft injury compared with the control group. Quantitative analysis confirmed a significant reduction in the inflammatory cell infiltration area (Fig. 4G and Fig. S18). ROS levels in skin graft tissues were quantified after treatment (Fig. S28A). Both hDAGel and Tac@hDAGel reduced ROS levels, and ROS levels were lower in the Tac@hDAGel group. To further exclude potential confounding caused by differences in administration route, dose, and dosing frequency, an additional skin transplantation experiment was performed under route-, dose-, and dosing frequency-matched conditions. In this experiment, free Tac and Tac@hDAGel were both administered locally at the graft site as a single treatment at an equivalent tacrolimus dose of 2.1 mg/kg. As shown in Fig. S19, Tac@hDAGel more effectively preserved graft integrity, reduced rejection scores, and prolonged graft survival compared with a single local dose of free Tac. In addition, Tac@hDAGel significantly alleviated enlargement of the draining lymph nodes and spleen. These findings further confirmed that the improved therapeutic efficacy of Tac@hDAGel was maintained even when the administration route, dose, and dosing frequency were matched. To evaluate the systemic biosafety of local hDAGel and Tac@hDAGel treatments, H&E staining of major organs (lung, kidney, heart, and spleen) was performed. No obvious histopathological abnormalities were observed in the examined organs (Fig. S20). Tac@hDAGel inhibits DC maturation and T-cell activation Persistent inflammation is known to induce excessive extracellular matrix accumulation within lymph nodes (LNs). In this study, the ipsilateral axillary lymph nodes and brachial lymph nodes were defined as the draining lymph nodes (DLNs). As shown in Fig. 5A and B, the DLNs harvested from the Tac@hDAGel group were markedly smaller and lighter compared to the control group, indicating that local delivery of Tac@hDAGel beneath the skin allograft effectively suppressed immune activation and mitigated graft rejection. A similar trend was observed in the spleen, a major systemic immune organ, where Tac@hDAGel treatment significantly reduced splenic enlargement (Fig. 5A and C). Tac@hDAGel suppresses dendritic cell (DC) maturation and T-cell activation. (A) Representative images of draining lymph nodes (DLNs, including ipsilateral axillary and brachial LNs) and spleens from C57BL/6 mice in five different treatment groups (i) control (PBS treatment), (ii) Tac (i.p.), (iii)Tac (s.c.), (iv) hDAGel, and (v) Tac@hDAGel. (B) Weights of DLNs from different groups of mice. (C) Weights of spleens from different groups of mice. (D) Representative flow cytometry plots of mature DCs (CD80+CD86+ cell population) gated on CD45+CD3−CD11c+ cells in DLNs from different groups. (E) Representative flow cytometry plots of MHC II expression on DCs gated on CD45+CD3−CD11c+ cells in DLNs from different groups. (F) Representative flow cytometry plots of mature DCs (CD80+CD86+ cell population) gated on CD45+CD3−CD11c+ cells in spleens from different groups. (G-H) Representative flow cytometry plots of CD3+CD4+ T cells and CD3+CD8+ T cells gated on CD45+ cells in spleens from different groups. Data are presented as mean ± SD (n = 5). Notable differences between groups are denoted by *P < 0.05, **P < 0.01, ***P < 0.001, and ns indicates no significant difference. To investigate whether topical Tac@hDAGel treatment could alleviate systemic immune rejection, we analyzed immune responses in draining lymph nodes (DLNs), spleens (Sp), and peripheral blood mononuclear cells (PBMCs) using flow cytometry (Fig. S21). As shown in Fig. 5D and S22, Tac@hDAGel treatment significantly suppressed dendritic cell (DC) maturation in DLNs, with the proportion of mature DCs reduced by 19.5% compared to the PBS-treated control group. Consistent results were observed in splenocytes (Fig. 5F and S23). Moreover, MHC II expression on DCs in DLNs was markedly downregulated in the Tac@hDAGel group (Fig. 5E and S22), indicating attenuated antigen presentation. Flow cytometry analysis of splenocyte-derived T cells revealed a significant reduction in CD4+ and CD8+ T cell populations in the Tac@hDAGel-treated group compared to other groups (Fig. 5G and H, and S23). A similar suppressive trend was observed in PBMCs (Fig. S24). Collectively, these findings demonstrate that localized delivery of tacrolimus via hDAGel effectively inhibited systemic T-cell activation and dampened immune rejection following allogeneic skin transplantation. These findings further support a coordinated spatiotemporal therapeutic mechanism in which oxidative stress-triggered hydrogel degradation induces Gel–Nano transformation, leading to both ROS scavenging and sustained local tacrolimus release. Together, these processes contribute to the suppression of dendritic cell maturation and T-cell activation, ultimately improving graft survival. VEGF- and BDNF-loaded hDAGel preparation and its anti-inflammatory and neural repair effects in the MCAO model Ischemic stroke, caused by the obstruction of cerebral blood flow, results in extensive neuronal death, neuroinflammation, blood–brain barrier (BBB) disruption, and impaired functional recovery [38,39]. A major challenge in post-stroke therapy is to promote angiogenesis, neurogenesis, and synaptic repair while simultaneously suppressing excessive inflammatory responses. VEGF and BDNF are two critical bioactive molecules that synergistically facilitate stroke recovery by enhancing neurovascular regeneration and neuronal survival [[40], [41], [42]]. However, systemic administration of VEGF or BDNF is limited by rapid clearance, poor BBB penetration, and significant off-target side effects [43,44]. To overcome these challenges, we developed an hDAGel-based delivery platform for the controlled and localized release of VEGF and BDNF at the ischemic site. The in vivo release behavior was evaluated by co-loading a hydrophilic near-infrared fluorescent dye, indocyanine green (ICG), into hDAGel together with VEGF and BDNF to generate ICG@hDAGel. Using a stereotaxic apparatus, ICG@hDAGel was administered by intracerebroventricular (ICV) injection in Sprague–Dawley (SD) rats, and its retention and release kinetics were continuously monitored via in vivo fluorescence imaging over a period of 5 days. As shown in Fig. 6A and B, compared to free ICG, the ICG@hDAGel group exhibited a more prolonged and stronger fluorescence signal within the lateral ventricle throughout the 5-day observation period, indicating that hDAGel significantly enhanced hydrophilic drug retention and provided a better-controlled release profile. At 5 days post-injection, SD rats from both groups were sacrificed, and the brains were dissected and sectioned for ex vivo NIR fluorescence imaging. As shown in Fig. 6C and D, a strong and localized NIR signal was detected in the lateral ventricular region of the ICG@hDAGel group, whereas only a weak and diffuse fluorescence was observed in the free ICG group, further confirming the superior local retention capability of hDAGel. VEGF- and BDNF-loaded hDAGel exhibits anti-inflammatory, neuroprotective, and neural repair properties in the MCAO model. (A–B) Longitudinal in vivo fluorescence imaging of free indocyanine green (ICG) and ICG@hDAGel (ICG co-loaded into hDAGel together with VEGF and BDNF) after intracerebroventricular (ICV) injection in Sprague–Dawley (SD) rats over 5 days, and fluorescence intensity was quantified as average radiant efficiency (n = 3 per group). (C–D) Ex vivo NIR fluorescence imaging of cerebrums dissected from SD rats at day 5 after ICV injection, with average radiant efficiency used to quantify fluorescence intensity. Data are presented as mean ± SD (n = 3 per group). (E) Schematic illustration of middle cerebral artery occlusion (MCAO) model construction, treatment, and monitoring process in SD rats. (F) The representative SEM image of BV@hDAGel (hDAGel co-loaded with VEGF and BDNF). Scale bar is 5 μm. (G) Neurological scores of SD rats measured weekly over 4 weeks (day 1, 7, 14, 21, and 28) in five groups: (i) Sham surgery, (ii) MCAO model with no treatment, (iii) MCAO + free VEGF + free BDNF, (iv) MCAO + hDAGel, and (v) MCAO + BV@hDAGel (n = 5 per group). (H) Representative MRI images of SD rat brains in five different treatment groups at two time points (day 1 and day 30) after ICV injection. The white dotted line highlights the ischemic area. (I–J) Immunofluorescence staining of myelin integrity (MBP), neurofilament 200 (NF200), brain-derived neurotrophic factor (BDNF), and neuronal nuclei (NeuN) in brain sections of SD rats from five different treatment groups at day 14 (scale bars = 300 μm). (K) Evaluation of white blood cell (WBC) counts of SD rats from different treatment groups. Data are presented as mean ± SD (n = 3). (L) Assessment of alanine aminotransferase (ALT) in SD rats from different treatment groups. Data are presented as mean ± SD (n = 3). Notable differences between groups are denoted by *P < 0.05, **P < 0.01, ***P < 0.001, and ns indicates no significant difference. We next established a middle cerebral artery occlusion (MCAO) model in SD rats to mimic ischemic stroke and investigate the anti-inflammatory and neural tissue repair efficacy of hDAGel co-loaded with VEGF and BDNF (referred to as BV@hDAGel). The schematic illustration of the MCAO model construction, treatment, and monitoring process is shown in Fig. 6E. To prepare BV@hDAGel, VEGF and BDNF were first dissolved in a PVA solution, followed by mixing with hDAlinker to form the hydrogel. A representative SEM image of BV@hDAGel is shown in Fig. 6F. The oxidation-responsive release of BDNF and VEGF from BV@hDAGel was further evaluated under physiological (PBS, pH 7.4) and oxidative (1 mM H2O2) conditions. As shown in Fig. S27B–C, the release of BDNF and VEGF was accelerated in the presence of H2O2. Meanwhile, nanoparticle formation from BV@hDAGel increased over time and was more pronounced under oxidative conditions (Fig. S27E). Pearson correlation analysis revealed positive correlations between cumulative BDNF or VEGF release and nanoparticle formation under both PBS and H2O2 conditions (Fig. S27H–K), indicating a close relationship between BDNF/VEGF release and the Gel–Nano transformation of hDAGel. The resulting BV@hDAGel formulation was then intracerebroventricularly (ICV) injected into the ischemic region. Five experimental groups were included: (i) Sham surgery, (ii) MCAO model without treatment, (iii) MCAO + free VEGF + free BDNF, (iv) MCAO + hDAGel, and (v) MCAO + BV@hDAGel. After successful establishment of the MCAO model, a single ICV injection of either free drugs, hDAGel, or BV@hDAGel was administered (VEGF: 1 μg; BDNF: 20 μg; hDAGel: 20 μL). Neurological function was assessed weekly using standardized neurological scoring. As shown in Fig. 6G, BV@hDAGel treatment led to a significant improvement in neurological scores during the first two weeks compared with the control group, whereas free VEGF + BDNF administration exhibited only limited therapeutic benefit. At day 1 and day 30 after ICV injection, MRI scans were performed to evaluate therapeutic outcomes. As shown in Fig. 6H, the ischemic area is highlighted by the white dotted line. Compared with the MCAO group and other groups, BV@hDAGel treatment markedly reduced the infarct size and significantly alleviated ischemic brain injury, demonstrating its superior therapeutic efficacy. ROS levels in ischemic brain tissues were quantified after treatment (Fig. S28B). Reduced ROS levels were observed in the hDAGel and BV@hDAGel groups. To further evaluate neuroinflammatory responses after MCAO, immunofluorescence staining for Iba-1, GFAP, iNOS, and Arg1, together with immunohistochemical staining for IL-1β and TNF-α, was performed. As shown in Fig. S29, BV@hDAGel reduced microglial activation, astrocyte activation, and the expression of IL-1β, TNF-α, and iNOS, whereas Arg1 expression was increased. Additionally, SD rats in each group were sacrificed at day 14, and cerebral tissues were sectioned for immunofluorescence staining. The results showed that BV@hDAGel treatment significantly upregulated both myelin basic protein (MBP) and neurofilament 200 (NF200) expression levels (Fig. 6I). MBP, the major structural protein of the central nervous system myelin sheath, serves as a marker of remyelination, white matter repair, and restored neuronal conductivity, whereas NF200 is essential for maintaining axonal architecture, stability, and conduction velocity. Furthermore, rats in the MCAO model exhibited pronounced neuronal loss, as indicated by decreased neuronal nuclei (NeuN) expression, along with reduced BDNF levels. In contrast, treatment with BV@hDAGel markedly improved neuronal survival, promoted robust neuronal regeneration, and enhanced BDNF expression within the ischemic region (Fig. 6J). hDAGel alone also exhibited moderate therapeutic effects in the MCAO model. Compared with hDAGel alone, BV@hDAGel achieved superior therapeutic outcomes, indicating that sustained delivery of VEGF and BDNF further enhanced the therapeutic efficacy. Collectively, these results demonstrated the superior efficacy of BV@hDAGel in preserving axonal integrity and promoting neuroregeneration following ischemic stroke. To comprehensively evaluate the systemic biosafety of hDAGel and its drug-loaded formulations, detailed hematological and hepatorenal function analyses were conducted in SD rats. Hematological parameters included red blood cell–related indices (RBC, HGB, HCT, MCV, MCH, MCHC, RDW-SD, RDW-CV), platelet-related indices (PLT, MPV, PDW, PCT, P-LCR, P-LCC), and white blood cell profiles (WBC and differential counts of LYM, MON, NEU, EOS, and BAS), collectively reflecting hematopoietic integrity, immune status, oxygen-carrying capacity, and coagulation balance. As shown in Fig. 6K and S25, no significant differences were observed between the sham surgery control group and either the hDAGel or BV@hDAGel treatment group, indicating that administration of hDAGel did not induce hematological toxicity, immune dysregulation, or coagulation abnormalities. In parallel, hepatic and renal function was assessed by measuring serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), total bile acids (TBA), total bilirubin (TBIL), direct bilirubin (DBIL), γ-glutamyl transferase (GGT), and albumin (ALB), together with renal markers including blood urea nitrogen (BUN), creatinine (CR), and uric acid (UA). The results showed that all biochemical parameters remained within normal physiological ranges, with no statistically significant differences between the sham surgery group and the hDAGel and BV@hDAGel groups (Fig. 6L and S26). These results indicate preserved hepatic synthetic, metabolic, and biliary functions, as well as intact renal glomerular and tubular function following hDAGel administration. Collectively, the absence of detectable alterations in hematological profiles and hepatorenal biomarkers demonstrates that hDAGel and its drug-loaded formulations exhibit negligible systemic toxicity and maintain in vivo biosafety. Conclusion In summary, we developed a novel injectable hydrogel, termed hDAGel, which incorporates a unique Gel–Nano transformation mechanism for the treatment of refractory inflammation-associated trauma, as demonstrated in models of allogeneic skin transplantation and ischemic stroke. This study demonstrates the integration of Gel–Nano transformation into an injectable hydrogel platform for topical and localized therapies, synergistically combining strong bioadhesive properties with the controlled drug delivery capabilities of nanoparticles. Importantly, the Gel–Nano transformation enables versatile drug encapsulation, accommodating both hydrophobic molecules (e.g., Cy5.5, tacrolimus) and hydrophilic macromolecules (e.g., VEGF, BDNF). In inflamed microenvironments characterized by elevated reactive oxygen species (ROS), hDAGel undergoes a distinct process: the gel matrix releases dopamine, which spontaneously self-assembles into nanoparticles. These nanoparticles are then efficiently transported from the gel to the surrounding tissue, thereby enabling precise and sustained delivery of therapeutic agents at the inflammation site. This delivery strategy results in significant in vivo therapeutic efficacy without detectable systemic toxicity, highlighting hDAGel as a promising platform for managing inflammation-related disorders. Collectively, our findings establish hDAGel as a versatile therapeutic system, with broad potential for the targeted treatment of inflammatory diseases through its distinctive Gel–Nano transformation behavior. Materials and methods hDAlinker synthesis and hDAGel preparation Four sequential reactions were used to synthesize the hDAlinker, with detailed procedures provided in the Supplementary Material. The hDAGel was then prepared as follows. First, PVA and hDAlinker were separately dissolved into deionized water to form 10 wt% PVA and 40 wt% hDAlinker. Equal volumes of PVA and hDAlinker solutions were thoroughly mixed to form hydrogels within a few seconds. All the compounds and solvents were purchased from Aladdin Reagent Co., Ltd. (Shanghai, China) and were used without further purification. Preparation of drug-loaded hDAGel Firstly, dye (Cy5.5 or ICG) or drug (tacrolimus, VEGF, and BDNF) stock solutions were diluted in a 10 wt% PVA solution, followed by mixing with the 40 wt% hDAlinker solution to form drug-loaded hydrogels within a few seconds. Tacrolimus (FK506), Cy5.5, and VEGF164 (Rat) were purchased from MedChem Express (USA), and BDNF (Rat) was purchased from Cloud-Clone Corp (USA). Morphology observation hDAGel was freeze-dried overnight and then fixed on a circular stub of 10 mm diameter using carbon tape. Gold was sprayed on the surface of samples for 30 s to form a layer. The morphology of hDAGel was then observed using SEM (FEI Nova Nano 450). Porcine skin test Porcine skin was cut into small pieces (5 cm × 1 cm). Gels (100 μL) were sprayed over a 1 cm2 area (the outer surface) at the end of one piece, to which the end of another piece was stuck. After incubation, the maximum separation tension was measured using a digital force gage (SC-50, Shence). In vitro hydrogel degradation test 20 μL of hDAGel was added into a 96-well plate, followed by the addition of H2O2 (200 μM and 1 mM) or PBS (pH 7.4). The shape and size of hDAGel from different groups were observed and recorded daily using an optical microscope for 5 days. The hydrogel size was quantitated using Image J. Each experiment was repeated three times. Morphological changes observation 20 μL of hDAGel was incubated with H2O2 (1 mM) for 3 days. The samples were freeze-dried overnight and then fixed on a circular stub of 10 mm diameter using carbon tape. Gold was sprayed on the surface of samples for 30 s to form a layer. The morphology of samples was then observed using SEM (FEI Nova Nano 450). Dynamic light scattering (DLS) analysis The hydrodynamic diameter and size distribution of nanoparticles were determined by Malvern Nano-ZS90 (Malvern, UK) at room temperature. Quantification of dopamine recovery during Gel–Nano transformation hDAGel samples containing equal initial amounts of dopamine moieties were incubated in PBS or PBS containing 1 mM H2O2 at 37 °C. At predetermined time points, the samples were centrifuged at 1000 × g for 5 min at 4 °C to remove the remaining bulk gel fragments. A calibration curve was established using dopamine hydrochloride solutions of known concentrations. The dopamine-equivalent content in the nanoparticle fraction was quantified by measuring the absorbance at 280 nm using a UV–vis spectrophotometer. Dopamine recovery was expressed as the percentage of the dopamine-equivalent amount recovered in the nanoparticle fraction relative to the initial amount of dopamine moieties incorporated into hDAGel. In vitro drug-release studies Tac@hDAGel and BV@hDAGel were incubated in PBS (pH 7.4) or PBS containing 1 mM H2O2 at 37 °C. At predetermined time points, the release medium was collected and replaced with an equal volume of fresh medium. Tacrolimus concentrations were quantified using HPLC, whereas BDNF and VEGF concentrations were measured using the corresponding ELISA kits. Cumulative release was expressed as the percentage of the initially loaded amount. Cell culture Human umbilical vein endothelial cells (HUVECs) and mouse fibroblast cell line (L929 cells) were obtained via commercial approaches and used for in vitro experiments. Cells were cultured in Dulbecco's modified Eagle's medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and kept in an incubator with a constant temperature of 37 °C with 5% CO2. Live/dead cell viability assay Cell viability and cytotoxicity were evaluated using a Calcein-AM/Propidium Iodide (PI) Cell Viability/Cytotoxicity Assay Kit. Briefly, cells were seeded into 24-well plates and cultured under standard conditions until reaching the desired confluence. After treatment, the culture medium was removed, and the cells were incubated with the working solution containing Calcein-AM (to stain viable cells) and PI (to stain dead cells with compromised membranes) according to the manufacturer's instructions. Following incubation at 37 °C for 20 min, the cells were gently washed with PBS to remove excess dye. Fluorescence images were acquired using a fluorescence microscope, where live cells emitted green fluorescence and dead cells exhibited red fluorescence. Cellular imaging for ROS analysis The Cellular ROS Assay Kit was utilized to measure ROS generation in L929 cells. Briefly, L929 cells were initially seeded in a 48-well plate and cultured overnight to reach 80% cell density. The cells were then washed with PBS and incubated with various formulations for 4 h. Following incubation, the cells were incubated with the medium containing DCFH-DA (10 μM) from the ROS assay kit for an additional 30 min. After washing three times with PBS, the cells were imaged using a fluorescence microscope. Animal experiments All animal experiments were performed in accordance with the relevant guidelines and regulations and approved by the Institutional Animal Care and Use Committee (IACUC) of Zhejiang Center of Laboratory Animals (Approval No. ZJCLA-IACUC-20011032). Flow cytometry for macrophage phenotyping Bone marrow-derived macrophages (BMDMs) were obtained as follows. Briefly, bone marrow cells were flushed from the femurs of C57BL/6 mice and cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 1% penicillin–streptomycin, and 20 ng/mL recombinant macrophage colony-stimulating factor (M-CSF) for 6 days. On day 7, adherent BMDMs were harvested and subsequently polarized into distinct macrophage subtypes. Cells were stimulated with 20 ng/mL interleukin-4 (IL-4) for 24 h to induce M2 macrophages or with 10 ng/mL LPS for 24 h to induce M1 macrophages. Flow cytometry was employed to characterize macrophage polarization. BMDMs were harvested, washed with cold PBS containing 1% BSA, and incubated with anti-mouse CD16/CD32 Fc block (clone 93, BioLegend, 1:100) for 10 min at 4 °C to minimize nonspecific binding. Cells were then stained with the following fluorochrome-conjugated antibodies for 30 min at 4 °C in the dark: anti-F4/80, anti-CD11b, anti-CD80, and anti-CD86. Zombie Aqua was used to exclude dead cells. Corresponding isotype controls were included to ensure specificity. After staining, cells were washed twice and resuspended in PBS for flow cytometry analysis. Cytokine secretion assay BMDMs were obtained as described above and seeded into 12-well plates at a density of 1 × 105 cells per well. Cells were then treated with LPS, IL-4, LPS + hDAGel, LPS + hDAlinker (2 μM), or LPS + hDAlinker (5 μM) for 12 h under standard culture conditions. Following stimulation, the cell culture supernatants were collected for cytokine secretion analysis. The concentration of interferon-γ (IFN-γ) was quantified using an ELISA kit according to the manufacturer's instructions. Allogeneic skin transplantation model for acute graft rejection An acute skin graft rejection model was established using a fully mismatched donor–recipient pair, with BALB/c mice serving as donors and C57BL/6 mice as recipients. A full-thickness trunk skin graft was excised from BALB/c donors, while a 1 cm2 section of skin was removed from the left upper back of recipient mice. For the hDAGel and Tac@hDAGel groups, 50 μL of PVA or Tac@PVA was combined with 50 μL of hDAlinker and applied to the wound bed of the recipient mice. The mixture was allowed to crosslink in situ for 30 s before graft placement. The harvested donor skin was trimmed to fit and positioned over the crosslinked hydrogel. The graft was secured using eight interrupted sutures placed at the corners and along the margins. In the Tac (i.p.) group, tacrolimus was administered intraperitoneally at 0.6 mg/kg every other day until day 7. In the Tac (s.c.) group, tacrolimus was administered by local subcutaneous injection into the tissue adjacent to the graft bed at 2.1 mg/kg every other day until day 7. Following transplantation, the grafts were covered with a protective bandage for 7 days. Graft rejection was independently scored by two investigators blinded to the treatment groups using predefined graft rejection grading criteria. To further compare free Tac and Tac@hDAGel under matched administration conditions, an additional skin transplantation experiment was performed using four groups: Control, free Tac, hDAGel, and Tac@hDAGel. The skin transplantation procedure was performed as described above. All treatments were administered locally at the graft site as a single treatment at the time of transplantation. In the free Tac group, tacrolimus was administered by local subcutaneous injection into the tissue adjacent to the graft bed at a dose of 2.1 mg/kg. In the Tac@hDAGel group, Tac-loaded PVA solution containing an equivalent tacrolimus dose of 2.1 mg/kg was mixed with hDAlinker solution and applied directly to the graft bed. Specifically, 50 μL of Tac-loaded PVA solution was mixed with 50 μL of hDAlinker solution, allowing rapid in situ gelation within approximately 30 s. In the hDAGel group, 50 μL of unloaded PVA solution was mixed with 50 μL of hDAlinker solution and applied using the same procedure. The Control group received an equivalent volume of vehicle locally at the graft site. Graft appearance and rejection scores were recorded after transplantation. Graft survival was analyzed using Kaplan–Meier analysis. Draining lymph nodes and spleens were collected and weighed at the predetermined endpoint. Analysis of immune cells in the acute graft rejection model At 7 days post-transplantation, the mice were sacrificed, and spleens, lymph nodes (LNs), and peripheral blood were harvested and processed into single-cell suspensions as follows. Spleens were gently ground and filtered through a 70-μm cell strainer with ice-cold PBS, then centrifuged at 650 g for 10 min. The pellet was incubated with Red Blood Cell Lysis Buffer (R1010, Solarbio) to remove red blood cells. LNs were gently ground and filtered through a 70-μm cell strainer with ice-cold PBS to obtain a single-cell suspension. peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque™ PLUS (GE17-1440-03, Cytiva) and washed with PBS to obtain a single-cell suspension. The proportions of mature dendritic cells (DCs), CD4+ T cells, and CD8+ T cells were analyzed by flow cytometry. Middle cerebral artery occlusion (MCAO) model The MCAO model was established in Sprague-Dawley (SD) rats to induce cerebral ischemia. Rats were anesthetized with sodium pentobarbital (50 mg/kg, intraperitoneally). A midline cervical incision was made to expose the left common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA). The CCA and ECA were carefully isolated and ligated, while a temporary microvascular clip was placed on the ICA. A 4-0 nylon filament with a rounded tip was introduced into the ICA via a small incision in the ECA and advanced approximately 18 mm from the carotid bifurcation to occlude the middle cerebral artery. After 90 min of ischemia, the filament was withdrawn to allow reperfusion. Sham-operated controls underwent the identical surgical steps without filament insertion. Throughout surgery, rectal temperature was maintained at 36–37 °C using a feedback-regulated heating pad. On the following day, cerebral infarction was confirmed by nuclear magnetic resonance imaging (MRI). MRI images were analyzed by a blinded investigator using standardized anatomical criteria. The infarct regions were manually delineated slice-by-slice, and the infarct area was calculated using identical segmentation criteria across all groups. For the hDAGel and BV@hDAGel treatment groups, 10 μL of PVA or BV@PVA was mixed with 10 μL of hDAlinker and stereotactically injected into the lateral ventricle of SD rats, where rapid in situ crosslinking occurred within 30 s. The surgical incision was subsequently sutured. Neurological function was assessed at 7-day intervals post-surgery. On day 14, rat brains were harvested, and frozen sections were prepared for immunofluorescence staining. The expression of brain-derived neurotrophic factor (BDNF), neuronal nuclei (NeuN), myelin basic protein (MBP), and neurofilament 200 (NF200) was examined by fluorescence imaging. In vivo fluorescence imaging An in vivo imaging system (IVIS) was utilized to monitor hydrogel distribution and fluorescence intensity after administration. Fluorescence images and data were collected for quantitative evaluation. Statistical analysis Data are presented as the mean ± SD. One-way ANOVA with Tukey's test was used to analyze differences among multiple groups, and two-way ANOVA with Tukey's test was used to analyze differences among multiple groups at multiple time points. Log-rank (Mantel–Cox) testing was used for survival curve analysis. Statistically significant differences are shown as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not statistically significant. All data were analyzed using GraphPad Prism 8.0. CRediT authorship contribution statement Jingxuan Xu: Writing – original draft, Visualization, Investigation, Conceptualization. Yifeng Zhu: Software, Investigation, Formal analysis, Data curation. Guangjian Fan: Writing – review & editing, Supervision, Formal analysis. Yaoyao Li: Validation. Yuxiang Zhang: Validation. Bowen Li: Writing – review & editing, Supervision, Conceptualization. Wenwei Cai: Supervision, Methodology, Conceptualization. Chenyue Zhan: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Declaration of competing interests 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. Acknowledgement This work was supported by grants from National Natural Science Foundation of China (32101063), Pioneer R&D Program of Zhejiang (2025C02056) and Hangzhou Institute of Medicine, Chinese Academy of Sciences (2024ZZBS19). We acknowledge Hangzhou AiTing Biotechnology Co., Ltd. for its assistance with the analysis of flow cytometry data. Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.mtbio.2026.103605. Contributor Information Jingxuan Xu, Email: Jingxuan.Xu@medma.uni-heidelberg.de. Yifeng Zhu, Email: yifeng.zhu@tum.de. Guangjian Fan, Email: gjfan@shsmu.edu.cn. Yaoyao Li, Email: 17984644630@qq.com. Yuxiang Zhang, Email: 2023112028109@stu.hznu.edu.cn. Bowen Li, Email: bowen_li@ahmu.edu.cn. Wenwei Cai, Email: wwcai@139.com. Chenyue Zhan, Email: bruce.zcy.bz@gmail.com. Supplementary data The supplementary data include additional experimental methods and Fig. S1–29, including the synthesis of hDAlinker, characterization of compounds, and other supporting results. Supplementary data The following is the Supplementary data to this article. Multimedia component 1 Data availability Data will be made available on request. References Associated Data Supplementary Materials Multimedia component 1 Data Availability Statement Data will be made available on request.