English

• 1. Introduction

  Skin functions as the body's primary defense to sustain the body's homeostasis by maintaining fluid equilibrium, sensing the external environment, and regulating body temperature [1]. In fact, skin is not only the largest organ of the human body but also belongs to the most vulnerable one to tissue damage. The injured skin structure and integrity caused by distinct internal or external stimuli (e.g., diabetes, trauma, surgical incisions, and burns) will lead to the formation of wounds, among which the acute wounds can be recovered naturally within several weeks or one month owing to the self-repairing capability of body skin by initiating four well-orchestrated and overlapping stages (hemostasis, inflammation, proliferation, and remodeling) [2]. However, once getting stuck in a prolonged and dysregulated inflammatory state resulting from various reasons, such as accumulated reactive oxygen species (ROS), infection, and vascular/nervous dysfunction, the wounds will transit from acute into chronic, thus delaying the tissue repair process and bringing up potential complications. Chronic wounds, including diabetic ulcers, vascular ulcers, and pressure ulcers [3], are a nonnegligible global health concern, particularly affecting elderly individuals, diabetics, and those with compromised immune systems. Chronic wounds not only result in significant physical and emotional suffering but also impose a considerable economic burden on healthcare systems worldwide [4]. Chronic wound healing is a multifactorial process influenced by several key factors, including dysfunctional local microenvironment [5], impaired vascularization [6], excessive inflammation [1], oxidative stress [7], and infection [8], etc.

  Traditional methods of chronic wound care, such as routine dressing changes, antibiotics, and debridement, often fall short of addressing the complex factors that contribute to delayed healing [9]. While these methods help with infection control and basic wound protection, they do not handle the balance of the underlying issues such as oxidative stress, poor blood flow, chronic inflammation, and bacterial biofilms, thus failing to achieve a dynamic and comprehensive treatment against the complexed disease microenvironment in chronic wounds, but causing repeated hospital visits and increased healthcare costs [10]. Given these challenges, developing effective therapeutic strategies that address the underlying factors of chronic wounds is crucial for improving patient outcomes. In this review, in addition to presenting the underlying pathological factors contributing to chronic wounds, and providing insights into the current state of chronic wound care, we will mainly aim to introduce the recent advanced therapeutic modalities against chronic wounds with examples of personalized, targeted, and adaptive strategies that not only treat symptoms but also address the underlying pathological processes hindering wound healing.

  2. Chronic wounds and the critical factors postponing wound healing

  Wounds are common injuries throughout the body and can be categorized into acute and chronic wounds according to their clinical presentation [4]. The typical healing of acute wounds involves four interconnected phases: Hemostasis, inflammation, proliferation, and remodeling [11]. Each of these phases plays a crucial role in tissue repair and skin integrity restoration. Any disruption that prevents wounds from proceeding through these normal phases could result in chronic wounds, including pressure ulcers, arterial ulcers, diabetic wounds, burn wounds [1]. See Supporting information for details.

  2.1 Local microenvironment

  The local wound microenvironment including oxygenation degree, pH levels, and glycemia content plays a nonnegligible role in influencing cellular activity and wound healing outcomes [12]. See Supporting information for details.

  2.2 Dysfunctional vascularization

  Angiogenesis is a pivotal biological process in the proliferation phase by supplying nutrients and oxygen to the wound site. The impeded angiogenesis in chronic wounds is resulted by multiple mechanisms. For example, as a hallmark of diabetic wounds, hyperglycemia drives the irreversible formation of advanced glycation end products (AGEs). These AGEs can cross-link with vascular basement membrane protein to induce vascular stiffness, and trigger intracellular ROS generation by interacting with receptor for AGEs (RAGE) on the surface of vascular endothelial cells (ECs), etc., ultimately impairing angiogenesis [13]. In addition, hyperglycemia could also downregulate the stable expression and activity of hypoxia inducible factor-1α (HIF-1α), an angiogenic factor that initiates angiogenesis under hypoxic conditions, thus inhibiting the activation of endothelial progenitor cells [13,14]. See the Supporting information for details.

  2.3 Inflammation and oxidative stress

  The inflammatory response plays a dual role in wound healing. The primitive immune responses initiate defensive reactions against necrotic debris and pathogens in the injured site. With the propelling progress of tissue repair, the inflammation phase turns to subside accompanied by the shifting of pro-inflammatory M1 macrophages to anti-inflammatory M2 macrophages, and the programmed apoptosis of neutrophils [15,16]. However, the increased chemokines and pro-inflammatory cytokines (e.g., tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6)) prolonged the leukocyte infiltration in chronic wounds and disrupted the normal apoptosis procedure for wound closure [17], thus contributing to delayed termination of the inflammation phase. See Supporting information for details.

  2.4 Infection and biofilm formation

  In chronic wounds, the long-lasting destruction state of the skin physical barrier and the resulting denaturation of wound-existed proteins, lipids, and glycans creates a natural fit microenvironment for bacterial infection [18], which in turn arrests the wound healing process in an enduring inflammation phase. According to the bioburden levels, there exist four basic conditions including bacterial contamination, colonization, critical colonization, and infection, among which the latter two states possess the potential to hinder the normal tissue repair sequence and promote the development of chronic wounds [19]. Severe microbial colonization might bring up wound deterioration and necrosis, which further induces lower extremity amputation or systemic sepsis, etc. [20]. It is demonstrated that the minimal colonization of bacteria that can hamper wound healing is discovered to be approximately 10⁵ colony-forming units (CFU) per gram of tissue [21]. With the increase of bacterial colonization density, the host immunity would be overcome for higher bacterial proliferation efficiency, thus forming infection and damaging wound tissue [18]. According to the report, 76% of the ulcers presented polymicrobial colonization (existing two or more bacterial species), thus enhancing the complexity of anti-infection treatments [22]. See Supporting information for details.

  3. Recent advanced therapeutic strategies against chronic wound healing

  Chronic wounds represent a significant challenge in medical practice, often resulting from a combination of underlying pathophysiological factors, including impaired angiogenesis, bacterial infection, dysregulated inflammatory responses. Traditional treatment modalities may prove insufficient in addressing the complexities associated with chronic wound healing, necessitating the exploration and implementation of advanced therapeutic strategies [23,24]. This section provides an overview of recent innovations in therapeutic strategies that have shown promise in promoting chronic healing and will delve into elucidating their mechanisms (Fig. S2 in Supporting information). The integration of these strategies into clinical practice may represent a paradigm shift in the approach to chronic wound management, highlighting the importance of a multifaceted and tailored therapeutic strategy.

  3.1 ROS scavenging

  ROS can be classified into four types: Hydrogen peroxide (H₂O₂), singlet oxygen (¹O₂), hydroxyl radical (^•OH), and superoxide anion (O₂ ^•−), all of which are mainly produced in the substrate end of the respiratory chain within mitochondria [25]. ROS plays a vital role in numerous cellular activities by functioning as an indispensable signaling molecule. Nevertheless, excessive ROS accumulation could disrupt the cellular redox equilibrium state and cause oxidative stress, including protein denaturation, DNA damage and lipid peroxidation, thus inducing ROS-related diseases such as accelerating skin injury to chronic wounds [25]. In order to reduce the ROS levels in chronic wounds, some wound dressings have been designed to load antioxidant agents, for example, nanozymes that scavenge ROS, anti-oxidant gaseous molecules, and ROS cleavable polymers composed of the wound bandages.

  3.1.1   Nanozyme-based ROS scavenger

  Until now, a good deal of nanozymes (or nano-catalysts) has been widely explored to eliminate ROS by mimicking the activities of three typical antioxidant enzymes, such as catalase (CAT), glutathione peroxidase (GPx), and superoxide dismutase (SOD) [25,26]. Current major nanozyme-based ROS scavengers function by facilitating the process of electron transfer process via acting as electron donors or acceptors and could be summarized into four categories, including metal-based nanozymes, carbon-based nanozymes, metal-organic complex nanozymes, and coated natural enzymes [26,27]. The most popular nanozymes belong to metal oxide NPs, including manganese (Mn)- [28], cerium (Ce)- [29], cobalt (Co)- [30], palladium (Pd)- and platinum (Pt)-based NPs [31,32]. In general, the nanozyme with single enzyme-like activity is inefficient in scavenging excessive ROS species, which inspires further investigations in nanozymes with multiple enzyme-like activities. Besides, there also exist several nonnegligible challenges that alter the sustained catalytic ability of nanozymes except the basic catalytic components of nanozymes [26], such as the nonhomogeneous size or morphology that may hinder electron transfer process, the lack of surface-modified functional groups that may cause destructions by the complexed wound micro-environment, and the rapid degradation in specific pH environments [33]. Considering the adverse impacts on the degradation of nanozymes and the bacteria-infected condition in diabetic wound healing, Li et al. fabricated a cobalt manganese-based nanozyme coated with poly(tannic acid) and ε-polylysine (MnCoO@ε-PL@PTA) [33]. This nanozyme demonstrated the ability to persistently and efficiently convert ROS into O₂, while also shielding the encapsulated nanozymes from rapid degradation and providing antibacterial properties, which is beneficial from the intrinsic anti-bacterial characteristic of ε-polylysine (Fig. S3A in Supporting information) [33].

  The mixed metal ion oxidation states equipped these metal oxide NPs with proficient ROS scavenging capabilities. Take copper, a fundamental element of natural existing enzymes (such as superoxide dismutase) and possess multiple valence states as an example, natural enzymes-like copper-based NPs have been engineered with metal-ligand cooperative catalysis [34]. For example, Chen et al. reported a ligand-metal charge transfer (LMCT)-induced Cu nanozyme utilizing polyphenols as coordinating ligands (Figs. S3B and C in Supporting information), where the substrates copper phosphate (Cu₃(PO₄)₂) nanosheets (Cu NS) were coordinated by tannic acid (TA) onto the surface by a simple ligand-exchange process to obtain the chelated nanozyme (TA-Cu NS) [35]. The incorporation of TA not only increases the antioxidant activity of Cu NS by the intrinsic ROS decomposition capability of polyphenols but also alleviates the cytotoxic effect of directly using Cu NS [35]. Apart from introducing ligands to obtain Cu-based nanozymes, incorporating other metallic elements such as Pt is also a well-liked adoption. For example, Guo et al. engineered a type of PtCuTe nanosheets with remarkable ROS-scavenging properties (Fig. S3D in Supporting information), including ^•OH scavenging capacity, as well as CAT-like, peroxidase (POD)-like, and SOD-like activity, thus exhibiting remarkable therapeutic efficacy in diabetic wound healing [36]. It was noteworthy that the Te element conferred ROS-independent antibacterial properties to the nanozymes, mediated through bacterial membrane damage and flagellar movement suppression, thus effectively solving the contradictory points between ROS-mediated anti-bacteria and ROS-increased wound healing delay [36]. The superior antioxidant function of nanozymes to tackle complex conditions in chronic wounds could be achieved by further modification in the proportion of metal oxidation states [37]. For example, by adjusting the proportion of Ce³⁺ and Ce⁴⁺, the ROS catalytic activities of cerium oxide would be converted from the predominant SOD-like activity with a higher ratio of Ce³⁺/Ce⁴⁺ to predominant CAT-like activity with a higher ratio of Ce⁴⁺/Ce³⁺ [38]. In another example, compared to nanozyme ZnMn₂O₄ with single SOD-like activity, Li doped nanozyme (LiMn₂O₄) with increased valency of Mn from +3 to +4 obtained additional antioxidant activity, including enhanced CAT-like and GPx-like activities (Fig. S3E in Supporting information) [37].

  3.1.2   Gas-based ROS scavenger

  Gas-based ROS scavengers represent an innovative approach to the management of oxidative stress associated with chronic wounds. These scavengers utilize antioxidative gaseous molecules, hydrogen (H₂) [38,39] and hydrogen sulfide (H₂S), to neutralize ROS and promote a favorable environment for chronic wound healing. See Supporting information for details.

  3.1.3   ROS-depleted vehicles

  Aside from developing nanozymes and antioxidative gaseous molecules, ROS-responsive materials are also preferred to be incorporated in the therapeutic agents-contained vehicles for ROS consumption in chronic wounds. Common ROS-responsive materials usually contain thioether, selenium (Se)/tellurium (Te), thioketal, boronic ester, peroxalate ester, polyproline, polysaccharide, aminoacrylate, and so on [40,41]. See Supporting information for details.

  3.2 Oxygen (O₂) delivery

  O₂ is a vital component in the wound healing process, playing a critical role in cellular proliferation, defense against bacteria, angiogenesis, re-epithelialization and collagen synthesis [42]. Insufficient O₂ levels caused by the impaired blood flow or metabolic dysfunctions can significantly hinder the healing process by impairing cellular functions, promoting inflammation, and exacerbating tissue damage. As a consequence, improving the hypoxic microenvironment by O₂ delivery and in situ O₂ generation strategies has emerged as a promising therapeutic strategy in managing chronic wounds.

  3.2.1   Exogenously delivering O₂

  Exploiting O₂-carrying materials such as hemoglobin (Hb), perfluorocarbons, and oxygen nanobubbles (ONBs) to exogenously delivering O₂ into wound site offers a solution to alleviate tissue hypoxia. Aside from considering the regular attention on the biocompatibility, flexibility, high O₂-loading capability of O₂-carrying materials, the tissues penetration and controllable O₂ release profile should also be regarded to enhance their practical performances. In a recent report, Zhang et al. developed a type of separable responsive microneedles (MNs), which loaded black phosphorus dots (BP QDs) and O₂-carried Hb to achieve near-infrared ray (NIR)-responsive O₂ release by utilizing the thermal effect-mediated by BP QDs that could accelerate the separation of O₂ from Hb [43]. Such a NIR-responsive O₂ delivering platform not only promoted deeper O₂ penetration owing to the intrinsic transdermal drug delivery advantage, but also precisely regulated the oxygenation in hypoxic wound site according to the dynamical healing stages [43]. See Supporting information for details.

  3.2.2   Producing O₂ in situ

  Aside from O₂ delivery strategy, in situ production of O₂ at the wound site has also become a promising method to address the hypoxia problem in chronic wounds. Metal peroxide that can be converted into O₂ when encountering moisture are commonly used for O₂ generation, for example, calcium peroxide (CaO₂) [44], magnesium peroxide (MgO₂) [45] and zinc peroxide (ZnO₂) [46]. In general, the incorporation of metal peroxide could not only contribute to hypoxia alleviation, but also produce H₂O₂ substrate which can be further utilized by ROS-generation nanozyme for synergistic antimicrobial effects. However, the inherent shortcomings of inorganic peroxides determine explosive O₂ generation profile, which is unable to achieve stable and durable supply of O₂ [47]. See Supporting information for details.

  3.2.3   Catalyzing ROS into O₂

  Exogenously delivering O₂ or producing O₂ in situ could alleviate hypoxia in chronic wound sites to some degree, however, a potential threat for such therapeutic modality also exists on the other hand, for instance, O₂ could serve as a substrate to induce over-produced ROS that in turn harms the healing process. Under such circumstances, catalyzing ROS into O₂ has become a preferred strategy that could increase the oxygenation of hypoxic tissues but also decrease the ROS accumulation in the wound site. Delivering CAT that could selectively catalyze H₂O₂ into O₂ and H₂O is a preferred adoption in such conditions [48]. For example, Han et al. reported a cascade enzymatic reaction-driven O₂-delivering nano chamber, which was incorporated into an alginate hydrogel for catalyzing chronic wound-existed ROS into O₂ as well (Fig. S8A in Supporting information) [49]. Constructed via ultrasonic cavitation approach, the nano chamber integrated a O₂-generating catalyst (OGC) nanobubble and superoxide scavenging (SOS) O₂ nanobubble by encapsulating nanoscale O₂ bubbles in a glycosylated protein conjugate of dextran-conjugated SOD and CAT [49]. The sequential enzymatic reaction could be achieved by SOS-induced dismutation of O₂ ^•− into H₂O₂, which was subsequently transformed into O₂ and water through the CAT in OGC, thus improving the oxygenation degree of the poorly healed wound and alleviating the ROS accumulation problem as well [49]. See Supporting information for details.

  3.3 Anti-infection

  Infection is a major impediment to wound healing, particularly in chronic wounds, where microbial colonization and biofilm formation are common challenges. The persistent presence of pathogenic microorganisms at the wound site triggers prolonged inflammation, tissue damage, and delayed healing [8]. Furthermore, biofilms, the structured communities of microorganisms encased in an extracellular polymeric substance (EPS), create a protective barrier that impedes immune responses and reduces the efficacy of conventional antimicrobial agents [23]. Addressing infection is therefore a critical component of managing chronic wounds and restoring the healing process.

  3.3.1   Thermotherapy

  Antibiotics are powerful weapons to deal with bacterial infections in the past time. However, considering that the overuse and misuse of antibiotics might result in rapid evolution of drug-resistant bacteria, a bunch of antibiotic-independent measures have been developed to avoid such dilemma. Thermotherapy has gained attention as an effective approach for combating bacterial infections due to its minimally invasive nature, broad-spectrum antibacterial properties, and capability of eliminating drug-resistant bacteria [50], which is beneficial from the high temperature-mediated disruption of the intact structure and bioactive substrates of bacteria [51]. Among the thermotherapy strategies, photothermal therapy (PTT) that exploits photothermal agents (PTAs) to convert NIR into heat have been widely investigated. Ideal PTAs are usually characterized with good biocompatibility, excellent photothermal stability and biodegradability, and most importantly high photothermal conversion efficiency. PTAs could be categorized into four major types: Noble metal materials (e.g., AuNPs [52], PtNPs [53]), metal compound materials (e.g., CuS [54], MoS_2-CeO₂ [55]), carbon materials (e.g., carbon nanotube [56], carbon dots [57], graphene [58]), and organic materials (e.g., polyaniline [59], polydopamine [60]).

  It is noteworthy that excessive high temperature could not only cause damage to bacteria, but also bring thermal injuries to healthy tissues. To alleviate such side effect, mild PTT at a temperature under 50 ℃ has been developed by fine-tuning the usage of photothermal materials, the laser power density and the irradiation time [50,61]. In addition to the above-mentioned strategies, developing thermostatic PTT system that stays in stationary temperature could further improve the safety. For example, Fu et al. reported a thermostatic hydrogel system by adjusting the temperature into a therapeutic window (i.e., 45–50 ℃) using the rapid phase transition of poly(N-isopropylacrylamide) (PNIPAM) to block NIR penetration (Fig. S9A in Supporting information), which hampering the further temperature increase [50]. Studies pointed out that temperature below 50 ℃ is a protective range to prevent normal tissue from injury, while effective PTT antibacterial therapy often requires high temperature (>60 ℃) [58]. The mild photothermal treatment usually exhibits unsatisfied antibacterial effects owing to the reversion by the intracellular heat shock protein [62]. Therefore, in order to enhance the antibacterial efficiency of mild PTT, strategies by incorporating heat shock protein inhibitors [58] and other strategies that make bacteria more susceptible to PTT (such as disrupting the bacterial membrane with quaternary ammonium salt [62]) could also benefit the handling with infected wound healing.

  The attenuation of NIR laser intensity in tissues severely limits its efficiency in managing deep bacterial infections. Compared to NIR-inspired PTT, magnetothermal therapy could handle deeper infectious condition owing to the penetration of magnetic fields through skin or other organs without intensity attenuation. For example, He et al. has fabricated a type of magneto-thermal responsive bilayer MNs (Fig. S9B in Supporting information), where the needle tips could penetrate the wound and producing heat in the alternating magnetic field to exert antibacterial effect, while the basal layer of the MN patch stayed at a lower temperature to ensure biosafety [63].

  3.3.2   ROS-mediated therapies

  Conventional antimicrobial antibiotics usually target one type or several kinds of microorganisms by functioning as inhibitors of specific pathways or structures (e.g., bacterial cell walls and protein synthesis), and the therapeutic efficiency is challenged by drug resistance problems and correlation with metabolic activity, which determines that they are inactive in dormant microorganisms. Distinct from antibiotics, ROS-based agents could indiscriminately eliminate a wide range of microorganisms, including Gram-positive and Gram-negative bacteria, parasites, fungi, and viruses by inducing oxidative damage across multiple biomolecules such as nucleic acids, lipids, and proteins, enabling ROS-based agents with broad-spectrum antimicrobial applications. In addition, such killing effects against microorganisms are independent on microbial metabolic activity, thus enabling ROS-based agents with capability to eliminate non-replicating or dormant cells, including persisters.

  There are several approaches to induce ROS generation, including photosensitizer-based photodynamic therapy (PDT), sonosensitizer-based sonodynamic therapy (SDT), as well as Fenton and Fenton-like reactions-based chemodynamic therapy (CDT) (Table S1 in Supporting information). Both PDT and SDT belong to non-invasive treatments that producing ROS (particularly singlet oxygen) via photosensitizer and sonosensitizer upon the exposure to specific wavelengths or ultrasound waves under the presence of O₂. Traditional photosensitizers include porphyrins and porphyrin derivatives, phthalocyanines, chlorins and chlorin derivatives, and synthetic organic dyes. In actual fact, most of the mentioned above organic photosensitizers also possess the ability of arousing SDT effect when receiving ultrasound waves, such as chlorin e6 (Ce6) and hematoporphyrin monomethyl ether (HMME). However, the unstable chemical properties and nonnegligible phototoxic side effects in superficial skin limited the further application of these photosensitizers. Inorganic sonosensitizers such as black phosphorus and TiO₂ NPs possess the potential to avoid the side effect of organic sonosensitizers, but unfortunately exhibits unsatisfying behaviors owing to their low quantum yield for ROS production due to fast electron (e^-)-hole (h⁺) recombination in pure materials [64]. In order to improve the ROS yield of these inorganic sonosensitizers, multiple improvements have been explored by hampering the rapid electron-hole pairs, including lowering band gap by introducing oxygen-deficiency-induced defect states in TiO₂ NPs [64], and incorporating Au NPs in black phosphorus nanosheets to trap the excited electrons utilizing the surface plasmon resonance [65]. As another extensively used method in producing ROS against microorganisms, the generation of ROS in H₂O₂-dependent CDT is often induced by iron-mediated Fenton reaction or other metal ions (such as copper or manganese)-mediated Fenton-like reaction to perform POD-like activities [66]. Take Fenton reaction as an instance, Fe²⁺ catalyzes H₂O₂ into highly reactive ^•OH, which are capable of damaging essential cellular components, such as proteins, lipids, and DNA. The reaction can be described as follows:

  $ \mathrm{Fe}^{2+}+\mathrm{H}_2 \mathrm{O}_2 \rightarrow \mathrm{Fe}^{3+}+\cdot \mathrm{OH}+\mathrm{OH}^{-} $

  In addition to iron, other metal ions such as manganese and copper can also participate in similar Fenton-like reactions by catalyzing the conversion of H₂O₂ into ^•OH. Several strategies have been investigated to amplify ROS production efficiency, such as increasing the redox cycling of metal ions by introducing reducing polyphenols (e.g., TA and EGCG), and increasing H₂O₂ substrate by importing H₂O₂ precursor (e.g., metal peroxide).

  3.3.3   Microbial regulation

  A newly developed route for avoiding antibiotic resistance in wound healing is making use of probiotic bacteria, which have been reported to exert multiple biological effects in combating pathogen, including competing with pathogenic bacteria for nutrients or binding sites, generating antimicrobial secreta, and producing organic acids that is beneficial for strengthening host immune response [67]. However, there exists several challenges to resolve for successful applying probiotic bacteria in curing wound infection, such as prolonging their survival time and preventing them from escaping from wound site [67]. For example, in order to perform long-acting antibacterial effects of the delivered probiotic bacteria (Lactobacillus reuteri), Jin et al. incorporated glycerol, a metabolizing substrate of Lactobacillus reuteri to generate reuterin (a non-proteinaceous broad-spectrum antibacterial substance), into the probiotic-based MN patch, which significantly facilitated the healing process of S. aureus-infected wounds than that of MNs group without glycerol [68]. In a recent report, Xu et al. fabricated a probiotic hydrogel that encapsulating Lactobacillus paracasei TYM202 (Fig. S10A in Supporting information), whose proliferation and metabolism activity could be promoted by the hydrogel matrix EPS-M76 (extracted from the extracellular polysaccharides of B. velezensis M76T11B) for producing lactic acid and acetic acid, thus further resisting the infection of pathogenic bacteria but preserving skin microbiota stability by maintaining Firmicutes and Proteobacteria balance [69].

  Generally, severe pathogen infection is often accompanied with the appearance of biofilms, which are composed of compact three-dimensional structure of EPS, blocking the entrance of antibacterial agents and immune cells into biofilms to perform functions. Incorporation of proper probiotic bacteria could effectively destroy the dense structure of biofilms to allow subsequent antibacterial effects. In an example, Qin et al. constructed a probiotic bio-heterojunction (P-bioHJ) by introducing Lactobacillus rhamnosus (LG) into MXene quantum dots/FeS heterojunction (Fig. S10B in Supporting information) [70]. The unique metabolic behaviors enable LG to utilize the saccharides (e.g., glucose, rhamnose, and maltose) of the EPS as metabolizing substrate to generate lactic acid. Such biological process could be further promoted with the assistance of LG’s inherent β-galactosidase activity which is beneficial for degrading the structure of EPS, thus contributing to the complete destruction of biofilms [70].

  Although the exploitation of probiotic bacteria showed inspiring antimicrobial effects against pathogenic bacteria in would healing, the safety and applicability concerns have been put forward regarding the utilization of live and proliferative bacteria as therapeutic agents in patients [71]. An alternative approach is adopting probiotic bacterial membrane vesicles (MVs) which are capable of exerting biological effects but exclude the potential concern in bringing up such safety problems. As a type of biogenic NPs (ranged from 50 nm to 150 nm) shed by bacteria, the bacterial MVs carried a variety of parent cells-derived cargoes including nucleic acids, lipids and proteins, which endow bacterial MVs with capability of fulfilling similar biological roles as live bacteria, such as cell-cell communication and antibacterial effects [72].

  3.4 Tissue repair promotion

  As one of the critical focuses in the treatment of chronic wound healing, promoting tissue repair addressed the core challenge of regenerating damaged tissues and restoring skin integrity. The causes that chronic wounds often fail to progress through the normal stages of healing are ascribed to the factors such as impaired cellular activity, insufficient extracellular matrix (ECM) deposition, and a lack of angiogenesis, which are potential targets for advanced therapeutic strategies to overcome the tissue repair challenge in the chronic wound bed.

  3.4.1   Fibroblast targeting therapy

  Fibroblasts play an essential role in the proliferative phase of wound healing by synthesizing ECM components, such as collagen and fibronectin, which provide structural support for the generation of granulation. In addition, fibroblasts also contribute to wound closure and facilitate cross-talk with other cell types, including keratinocytes and ECs, for coordinating repair processes. However, in chronic wounds, fibroblast dysfunction often appeared and are characterized with senescence, reduced proliferation, and impaired ECM production, thus significantly delaying tissue regeneration. See Supporting information for details.

  3.4.2   Angiogenesis promotion

  Adequate vascularization is the prerequisite to support tissue repair owing to the critical role of blood vessels in delivering oxygen, nutrients, and immune cells to the wound site. Impaired angiogenesis might be resulted from multiple causes in chronic wounds, including elevated ROS levels which damage ECs and ECM components, long-term inflammation-mediated by pro-inflammatory cytokines (e.g., TNF-α and IL-1β), insufficient secretion of angiogenic factors or signals, the dysfunction of ECs under high glucose concentrations in diabetic wound [15,73], and pathogen infection. See Supporting information for details.

  3.4.3   Bionic ECM construction

  As a complex 3D scaffold composed of structural proteins (e.g., collagen, elastin), glycosaminoglycans (e.g., hyaluronic acid, chondroitin sulfate), glycoproteins (e.g., laminin, fibronectin), and proteoglycans, the ECM plays a pivotal role in supporting and regulating cell adhesion, migration, proliferation, and cell-matrix interactions. However, the disruption and delayed reconstruction of the ECM resulted in suspended chronic wound healing. As a consequence, providing a bionic ECM that closely resembles the natural matrix is a promising strategy for restoring normal tissue architecture and regeneration. See Supporting information for details.

  3.4.4   Re-epithelization acceleration

  Successful re-epithelization of chronic wound is important for restoring the normal functions of skin, owing to the irreplaceable roles of epidermis by withstanding mechanical stress, preventing pathogen entry, maintaining fluid balance, protecting the body from extreme temperature and ultraviolet radiation [74]. In general, re-epithelialization starts with the migration of keratinocytes from the wound edges, hair follicles, and sweat glands, which proliferate and form new layers of epidermal tissue.

  Accelerating the proliferation and migration of keratinocyte using natural peptides and proteins (e.g., fibrinogen [75], laminin-derived amino acid sequence related to cell adhesion [76]) is a common method to achieve re-epithelization [77]. An interesting attempt in a recent work driven by the inspiration of the natural chiral structure in skin and proteins proved that the dextral chiral hydrogel containing D-phenylalanine derived molecules promoted much higher proliferation and migration efficiency of keratinocyte than that of the levorotatory chiral hydrogel containing L-phenylalanine derived molecules, thus resulting in faster re-epithelialization of diabetic wounds [77]. Excepting exploiting biochemical functions to equip wound dressings, actively constricting the wound via mechanical stretching could also benefit re-epithelialization process by promoting the proliferation of keratinocytes through multiple signaling pathways [78]. For an instance, Sun et al. developed a kind of antioxidative and active-shrinkage hydrogel to facilitate the wound closure and re-epithelization by offering mechanical force to allow skin shrinkage in both acute wounds and diabetic wounds (Fig. S14A in Supporting information) [78].

  Incomplete or delayed epithelial coverage would allow prolonged inflammation, abnormal fibroblast activity and abnormal matrix remodeling, thus resulting in hypertrophic scars or keloids, which can impair normal functionality and aesthetics of skin. In order to avoid such dilemma, numerous efforts have been made in scarless wound repair with timely re-epithelization. Zhao et al. fabricated a core-shell structured MN array patch to promote scarless wound repair by delivering verteporfin, a drug that not only exerted PDT efficacy to eliminate bacterial infection, but also inhibited scar formation by intercepting engrailed-1 (En1) activation in fibroblasts. Besides, restoring endogenous bioelectricity using a self-powered enzyme-linked MN that could generate stable current by consuming glucose in diabetic wound site also exhibited efficient scarless wound repair effect (Fig. S14B in Supporting information) [79].

  3.5 Immune regulation

  Acute inflammatory responses are usually activated by damage associated molecular patterns (DAMPs), elevated ROS levels, and chemokines after skin injury, recruiting leukocytes to clear pathogens and necrotic debris. Abundant immune cells have been engaged in this inflammatory period to construct a protective environment, including first responders (neutrophils and basophils), macrophages, Langerhans cells, mast cells, T cells and B cells [80]. However, failure to resolve the inflammatory phase would lead to the emergence of chronic wounds, which are characterized by impaired immune cell activity, prolonged inflammation, excessive matrix metalloproteinase (MMP) activity that degrades essential ECM components and growth factors [81], thus hindering the transition to the proliferative phase. Immune regulation strategies aim to strike a balance by restoring immune homeostasis from pro-inflammatory to anti-inflammatory status to supporting the tissue repair.

  3.5.1   Macrophages targeting therapy

  Macrophages can be classified into two primary types based on their distinct functions in wound healing: M1 macrophages, which are pro-inflammatory and appear in the early repair phase to carry out phagocytosis and secret inflammatory cytokines, and M2 macrophages, which are anti-inflammatory and evolve from M1 macrophages to promote inflammation resolution and tissue regeneration [82]. Numerous studies have investigated different strategies to rebalance this phenotypic shift by either suppressing excessive M1 activation or enhancing M2 polarization, including applying cytokines (e.g., IL-4 [83] and IL-10 [84]), designing functional NPs or vehicles that modulate macrophage activity (e.g., inhibiting nod-like receptor thermal protein domain associated protein 3 (NLRP3) inflammasome activation [85]) [40], and engineering hydrogels and scaffolds whose materials possess inherent biological effects (e.g., mannan [86] and polyphenol [87] modification) [73]. See Supporting information for details.

  3.5.2   Neutrophil targeting therapy

  Neutrophils play a dual role in the wound healing process. As first responders to injury, neutrophils contribute to pathogen clearance and debris removal via phagocytosis, releasing ROS, and the forming neutrophil extracellular traps (NETs) which are composed of DNA network and proteins. However, in chronic wounds, dysregulated or prolonged neutrophil activity leads to excessive inflammation, oxidative damage, and tissue destruction, exacerbating the impaired healing environment. Therapeutic strategies targeting neutrophils include reducing recruitment, downregulating activation, promoting apoptosis, and eliminating NETs. See Supporting information for details.

  3.5.3   T cell targeting therapy

  The population and immune responses of T cells within the injured tissue, including CD4⁺ T helper (Th) and CD8⁺ cytotoxic T cells (CTL), also affect the inflammation degrees [88]. As one of the subset of CD4⁺ T helper cells, regulatory T cells (Tregs) are crucial immune brake in tissue regeneration owing to their role in supporting the wound transition from the inflammatory to the proliferative phase by restricting accumulation of pro-inflammatory immune cells, downregulating the interferon-γ (IFN-γ) production, activating transforming growth factor-β (TGF-β) in neighboring keratinocytes [89], and secreting anti-inflammatory cytokines (e.g., IL4, IL-10) [90], et al. In addition, Tregs also support hair follicles regeneration to restore the normalization of damaged skin [91]. Therapeutic strategies targeting Tregs for wound healing include Treg adoptive transfer [92] and cytokines delivery [93], as well as strengthening the cellular biofunction [94].

  In order to avoid the adverse effects brought up by excessive ROS on recruited Tregs, Qu et al. fabricated a kind of ROS-scavenging hydrogel (RSG-CCL22) to achieve sequential release of C—C motif chemokine 22 (CCL22) for subsequent Tregs recruitment (Fig. S17A in Supporting information) [90]. Such a hydrogel was constructed with an ROS cleavable linker (NHS-TK-NHS) that formed amide bond with the amino groups of gelatin matrix and CCL22, thus enabling on-demand Tregs recruitment according to the dynamic changes of ROS levels in chronic wound [90]. In another study, Alshoubaki et al. adopted local administration of Tregs in fibrin hydrogels to achieve wound healing [92]. In general, endogenous Tregs accumulate in the late phase of tissue regeneration, while this research discovered that the rapid delivered exogenous Tregs upon skin injury would elect an “injury-specific” profile which is similar to endogenous later-accumulated Tregs to accelerate wound healing [92]. In fact, in order to improve the immunosuppression functionality of delivered Tregs, additional supplements of nutritive substance (e.g., propionate) could be supplemented in the delivery vectors [94]. See Supporting information for details.

  3.6 Smart dressings

  Smart dressings represent an innovative approach in wound management, combining advanced materials with real-time feedback mechanisms to improve the healing process. These dressings are designed to respond dynamically to the changing conditions of the wound environment, offering enhanced therapeutic effects over traditional dressings by achieving controlled drug release, enabling real-time responsiveness to wound conditions, reducing treatment times, and improving the quality of life for patients with chronic wounds.

  The key advantage of smart dressings lies in their ability to monitor and adjust to parameters such as pH, temperature, moisture levels, oxygen, and infection status through integrated bioelectronics, e.g., microelectrodes, wireless sensors), providing a more tailored and effective treatment for chronic wounds. For example, in a recent report, Ge et al. developed an intelligent wound dressing that incorporating a flexible circuit equipped with a near-field communication (NFC) antenna into the wound patch (WP) to achieve wound exudate management, infection monitoring and on-demand drug treatment (Fig. S18 in Supporting information) [95]. Such a smart dressing that adapts to different wound sizes could not only keep continuous monitoring on the humidity and temperature of patients, but also support on-demand drug delivery feedback to ensure non-excessive medical treatment [95]. Through tracking the pH changes in wounds, early warnings could be achieved to avoid infection risks. For example, Deng et al. incorporated phenol red as the pH indicator in a double cross-linked network hydrogel, which could be further integrated with smartphones to visualize the wound pH assessment using machine learning algorithms [96-98].

  Real-time monitoring, accurate diagnosis, and on-demand therapy are crucial to minimizing complications and saving lives [99-105]. See Supporting information for details.

  4. Conclusion and future outlook

  Multifarious pathological factors postponing wound healing (e.g., local wound microenvironment, impaired vascularization, prolonged inflammation, oxidative stress and infection) increase the complexity and therapeutic burden on managing chronic wounds. The interrelation and exacerbation of these factors implies the requirement of more accurate coping strategies for chronic wound therapy apart from basic infection control and wound protection. In this review, the key mechanisms hampering wound healing have been extensively introduced, which made up for the lack of a systematic summary of chronic wound development and the related focus on therapies designing. Considering the different wound conditions, strategies could be devised form the perspectives including ROS scavenging, oxygen delivery, anti-infection, tissue repair promotion, immune regulation, and smart dressings. Among these strategies, we introduced the fundamental designing principle and emerging advanced examples utilizing various functional nanomaterials and wound dressings, including hydrogels and MNs. Moreover, the development of smart dressings, with controlled-release mechanisms and real-time monitoring also represents a major innovation to enable more personalized and adaptive wound care modality.

  While these advanced therapies offer promising new directions for chronic wound management, there still exists multiple challenges to overcome. For example, utilizing ROS-based agents is beneficial for eliminating bacteria, but over-accumulated ROS in wound site is harmful to cellular activities and hindering tissue regeneration. Besides, delivering oxygen could alleviate hypoxia in chronic wounds, but also increase the possibility of causing oxidative damage by promoting ROS generation. Such contradictory effects should be carefully avoided by adjusting the strategies adapting to the dynamic changes of wound healing phases. In addition, with the rapid development of nanomaterials and biological vehicles such as probiotics, the safety issue should also be under considered to prevent secondary injury on wounds. Another point worth noting is that the heterogeneity of chronic wounds determines distinct disease microenvironment, which might require different additional treatments, and could be utilized to design specific responsive strategies. For example, as for diabetic wounds, the hyperglycemic environment could be utilized to design glucose-responsive materials and support electrical stimulation using glucose as power source.

  In summary, the remarkable achievements of nanomaterials, biological vehicles, and wound dressings endows wound care with tremendous potential in handling and balancing different pathological factors that contribute to chronic wounds. Artificial intelligence (AI) methods are increasingly being used to diagnose and treat chronic wounds. With the incorporation of computer science and AI, smart dressings could be further developed to achieve controlled drug release and real-time feedback, thus realizing a real sense of personalized wound management for enhancing the therapeutic efficacy of chronic wound healing.

  CRediT authorship contribution statement

  Weilang Zhang: Writing – original draft, Software, Conceptualization. Qixing Dong: Writing – original draft, Resources. Zhangwen Peng: Writing – original draft. Lina Su: Writing – review & editing, Funding acquisition. Hongmei Liu: Supervision, Resources. Yunfei Yi: Writing – review & editing, Project administration, Conceptualization. Lin Mei: Writing – review & editing, Supervision, Funding acquisition.

  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.

  Acknowledgments

  The authors gratefully acknowledge the support of this research by Tianjin Science Fund for Distinguished Young Scholars (No. 22JCJQJC00120), National Natural Science Foundation of China (No. 32460247), State Key Laboratory of Advanced Medical Materials and Devices Grant (No. 24ZXZSSS00200), and Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (Nos. 2024-I2M-TS-028, 2025-I2M-XHJC-048).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.112264.

  
  
• 1. [1]

     T. Liu, Y. Lu, R. Zhan, W. Qian, G. Luo, Adv. Drug Del. Rev. 193 (2023) 114670. doi: 10.1016/j.addr.2022.114670

  2. [2]

     S. Guo, L.A. Dipietro, J. Dent. Res. 89 (2010) 219–229. doi: 10.1177/0022034509359125

  3. [3]

     R.G. Frykberg, J. Banks, Adv. Wound Care 4 (2015) 560. doi: 10.1089/wound.2015.0635

  4. [4]

     B.R. Freedman, C. Hwang, S. Talbot, et al., Sci. Adv. 9 (2023) eade7007. doi: 10.1126/sciadv.ade7007

  5. [5]

     H. Li, B. Li, D. Lv, et al., Adv. Drug Deliv. Rev. 196 (2023) 114778. doi: 10.1016/j.addr.2023.114778

  6. [6]

     H.R. Moreira, A.P. Marques, Curr. Opin. Biotechnol. 73 (2022) 253–262. doi: 10.1016/j.copbio.2021.08.019

  7. [7]

     M. Hunt, M. Torres, E. Bachar-Wikstrom, J.D. Wikstrom, Commun. Biol. 7 (2024) 1534. doi: 10.1038/s42003-024-07219-w

  8. [8]

     A. Uberoi, A. McCready-Vangi, E.A. Grice, Nat. Rev. Microbiol. 22 (2024) 507–521. doi: 10.1038/s41579-024-01035-z

  9. [9]

     F. Feng, Z. Zhao, J. Li, Y. Huang, W. Chen, Prog. Mater. Sci. 146 (2024) 101328. doi: 10.1016/j.pmatsci.2024.101328

  10. [10]

      M. Farahani, A. Shafiee, Adv. Healthc. Mater. 10 (2021) e2100477. doi: 10.1002/adhm.202100477

  11. [11]

      T. Cui, J. Yu, C.F. Wang, et al., Adv. Sci. 9 (2022) 2201254. doi: 10.1002/advs.202201254

  12. [12]

      Y. Qian, B. Zhang, Y. Xu, et al., Adv. Funct. Mater. 35 (2025) 2423012. doi: 10.1002/adfm.202423012

  13. [13]

      Y. Yang, S. Zhong, F. Meng, X. Cui, Chem. Eng. J. 497 (2024) 154855. doi: 10.1016/j.cej.2024.154855

  14. [14]

      G.L. Wang, B.H. Jiang, E.A. Rue, G.L. Semenza, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 5510–5514. doi: 10.1073/pnas.92.12.5510

  15. [15]

      C. Lin, Y. Hu, Z. Lin, et al., Bioact. Mater. 43 (2025) 240–254.

  16. [16]

      O.A. Peña, P. Martin, Nat. Rev. Mol. Cell Biol. 25 (2024) 599–616. doi: 10.1038/s41580-024-00715-1

  17. [17]

      Y. Wu, Y. Quan, Y. Liu, et al., Nat. Commun. 7 (2016) 13393. doi: 10.1038/ncomms13393

  18. [18]

      Z. Lan, L. Guo, A. Fletcher, et al., Bioact. Mater. 42 (2024) 433–448.

  19. [19]

      A. Kingsley, Ostomy Wound Manag. 49 (2003) 1–7.

  20. [20]

      M. Falcone, B. De Angelis, F. Pea, et al., J. Glob. Antimicrob. Resist. 26 (2021) 140–147. doi: 10.1016/j.jgar.2021.05.010

  21. [21]

      D. Leaper, O. Assadian, C.E. Edmiston, Br. J. Dermatol. 173 (2015) 351–358. doi: 10.1111/bjd.13677

  22. [22]

      K. Gjødsbøl, J.J. Christensen, T. Karlsmark, et al., Int. Wound J. 3 (2006) 225–231. doi: 10.1111/j.1742-481X.2006.00159.x

  23. [23]

      J. Wang, X. Chen, Y. Zhao, et al., ACS Nano 13 (2019) 11686–11697. doi: 10.1021/acsnano.9b05608

  24. [24]

      F. Zhang, H. Zhang, S. Wang, et al., Nat. Commun. 16 (2025) 3738. doi: 10.1038/s41467-025-58987-w

  25. [25]

      S. Zhou, H. Cai, X. He, Z. Tang, S. Lu, Coord. Chem. Rev. 500 (2024) 215536. doi: 10.1016/j.ccr.2023.215536

  26. [26]

      F. Cheng, S. Kotha, M. Fu, et al., Nano Today 54 (2024) 102142. doi: 10.1016/j.nantod.2023.102142

  27. [27]

      C. Liu, W. Fan, W. Cheng, et al., Adv. Funct. Mater. 33 (2023) 2213856. doi: 10.1002/adfm.202213856

  28. [28]

      N. Guo, Q. Wu, H. Gan, et al., Chem. Eng. J. 499 (2024) 156172. doi: 10.1016/j.cej.2024.156172

  29. [29]

      Q. Wang, J. Du, J. Meng, et al., ACS Nano 18 (2024) 12341–12354. doi: 10.1021/acsnano.4c01370

  30. [30]

      X. Yang, J. Xiang, W. Su, et al., Nano Today 49 (2023) 101809. doi: 10.1016/j.nantod.2023.101809

  31. [31]

      J. Cheng, Y. Zhang, L. Ma, et al., Adv. Sci. 10 (2023) 2304002. doi: 10.1002/advs.202304002

  32. [32]

      B. Xue, M. Ge, K. Fan, et al., J. Control Release 350 (2022) 271–283. doi: 10.1016/j.jconrel.2022.08.026

  33. [33]

      H. Li, Y. Wang, X. Che, et al., Chem. Eng. J. 496 (2024) 153370. doi: 10.1016/j.cej.2024.153370

  34. [34]

      M.D. Wodrich, X. Hu, Nat. Rev. Chem. 2 (2017) 0099. doi: 10.1038/s41570-017-0099

  35. [35]

      Y. Chen, X. Yang, K. Li, et al., ACS Nano 18 (2024) 7024–7036. doi: 10.1021/acsnano.3c10376

  36. [36]

      Y. Guo, S. Ding, C. Shang, et al., Adv. Mater. 36 (2024) 2306292. doi: 10.1002/adma.202306292

  37. [37]

      Q. Wang, C. Cheng, S. Zhao, et al., Angew. Chem. Int. Ed. 61 (2022) e202201101. doi: 10.1002/anie.202201101

  38. [38]

      Z. Wang, X. Shen, X. Gao, Y. Zhao, Nanoscale 11 (2019) 13289–13299. doi: 10.1039/c9nr03473k

  39. [39]

      F. Gong, N. Yang, J. Xu, et al., Adv. Healthc. Mater. 12 (2023) 2201771. doi: 10.1002/adhm.202201771

  40. [40]

      P. Wang, J. Wu, H. Yang, et al., Bioact. Mater. 24 (2023) 463–467.

  41. [41]

      J. Liang, B. Liu, Bioeng. Transl. Med. 1 (2016) 239–251. doi: 10.1002/btm2.10014

  42. [42]

      Z. Yang, K. Ren, Y. Chen, et al., Adv. Healthc. Mater. 13 (2024) 2302391. doi: 10.1002/adhm.202302391

  43. [43]

      X. Zhang, G. Chen, Y. Liu, et al., ACS Nano 14 (2020) 5901–5908. doi: 10.1021/acsnano.0c01059

  44. [44]

      J. Wu, P. Shen, X. Qin, et al., Chem. Eng. J. 459 (2023) 141507. doi: 10.1016/j.cej.2023.141507

  45. [45]

      X. Han, L.S. Ju, J. Irudayaraj, Mol. Pharm. 20 (2023) 3338–3355. doi: 10.1021/acs.molpharmaceut.3c00352

  46. [46]

      J. Hu, J. Zhu, K. Wang, et al., Adv. Funct. Mater. 34 (2024) 2405383. doi: 10.1002/adfm.202405383

  47. [47]

      S. Gao, Y. Rao, X. Wang, et al., Adv. Mater. 36 (2024) 2307585. doi: 10.1002/adma.202307585

  48. [48]

      M. Li, Z. Feng, B. Wang, et al., Nano Today 57 (2024) 102324. doi: 10.1016/j.nantod.2024.102324

  49. [49]

      X. Han, L. Ju, C. Saengow, et al., Bioact. Mater. 35 (2024) 67–81. doi: 10.1016/j.bioactmat.2024.02.010

  50. [50]

      H. Fu, K. Xue, Y. Zhang, et al., Adv. Sci. 10 (2023) 2206865. doi: 10.1002/advs.202206865

  51. [51]

      L. He, D. Di, X. Chu, et al., J. Control. Release 363 (2023) 180–200. doi: 10.1016/j.jconrel.2023.09.035

  52. [52]

      H. Ma, Y. Tang, F. Rong, et al., Bioact. Mater. 27 (2023) 154–167.

  53. [53]

      Y. Luo, B. Li, X. Liu, et al., Bioact. Mater. 18 (2022) 421–432.

  54. [54]

      H. Shen, C. Zhang, Y. Meng, et al., Adv. Healthc. Mater. 13 (2024) 2303000. doi: 10.1002/adhm.202303000

  55. [55]

      T. Ma, X. Zhai, Y. Huang, et al., Adv. Healthc. Mater. 10 (2021) 2100033. doi: 10.1002/adhm.202100033

  56. [56]

      Y. Yang, L. Ma, C. Cheng, et al., Adv. Funct. Mater. 28 (2018) 1705708. doi: 10.1002/adfm.201705708

  57. [57]

      Y. Chen, Y. Gao, Y. Chen, et al., J. Control. Release 328 (2020) 251–262. doi: 10.1016/j.jconrel.2020.08.055

  58. [58]

      P. Zhao, Y. Zhang, X. Chen, et al., Adv. Sci. 10 (2023) 2206585. doi: 10.1002/advs.202206585

  59. [59]

      J. Xie, G. Liu, R. Chen, et al., J. Nanobiotechnol. 22 (2024) 232. doi: 10.1186/s12951-024-02474-9

  60. [60]

      X. Xu, M. Fan, Z. Yu, et al., Chem. Eng. J. 429 (2022) 132491. doi: 10.1016/j.cej.2021.132491

  61. [61]

      X. Ma, A. Wang, X. Zhang, et al., Mater. Today Bio 29 (2024) 101265. doi: 10.1016/j.mtbio.2024.101265

  62. [62]

      C. Liu, S. Li, R. Ma, et al., Nano Today 42 (2022) 101363. doi: 10.1016/j.nantod.2021.101363

  63. [63]

      D. He, X. Liu, J. Jia, et al., Adv. Funct. Mater. 34 (2024) 2306357. doi: 10.1002/adfm.202306357

  64. [64]

      X. Wang, X. Zhong, L. Bai, et al., J. Am. Chem. Soc. 142 (2020) 6527–6537. doi: 10.1021/jacs.9b10228

  65. [65]

      T. Chen, W. Zeng, C. Tie, et al., Bioact. Mater. 10 (2022) 515–525.

  66. [66]

      W. Wang, W. Xu, J. Zhang, et al., ACS Nano 18 (2024) 4089–4103. doi: 10.1021/acsnano.3c08054

  67. [67]

      C. Hua, F. Yang, X. Jia, et al., Chem. Eng. J. 483 (2024) 148432. doi: 10.1016/j.cej.2023.148432

  68. [68]

      Y. Jin, Y. Lu, X. Jiang, et al., Bioact. Mater. 38 (2024) 292–304.

  69. [69]

      H. Xu, Y. Li, J. Song, et al., Bioact. Mater. 35 (2024) 31–44.

  70. [70]

      M. Qin, X. Zhang, H. Ding, et al., Adv. Mater. 36 (2024) 2402530. doi: 10.1002/adma.202402530

  71. [71]

      T. Kuhn, A. Aljohmani, N. Frank, et al., J. Control. Release 365 (2024) 969–980. doi: 10.1016/j.jconrel.2023.12.011

  72. [72]

      L.P. Jahromi, G. Fuhrmann, Adv. Drug Deliv. Rev. 173 (2021) 125–140. doi: 10.1016/j.addr.2021.03.012

  73. [73]

      W. Liu, R. Gao, C. Yang, et al., Sci. Adv. 8 (2022) eabn7006. doi: 10.1126/sciadv.abn7006

  74. [74]

      M. Rodrigues, N. Kosaric, C.A. Bonham, G.C. Gurtner, Physiol. Rev. 99 (2019) 665–706. doi: 10.1152/physrev.00067.2017

  75. [75]

      T. Cui, J. Yu, Q. Li, et al., Adv. Mater. 32 (2020) 2000982. doi: 10.1002/adma.202000982

  76. [76]

      Y. Zhu, Z. Cankova, M. Iwanaszko, et al., Proc. Natl. Acad. Sci. U. S. A. 115 (2018) 6816–6821. doi: 10.1073/pnas.1804262115

  77. [77]

      H. Zhu, C. Xing, X. Dou, et al., Adv. Healthc. Mater. 11 (2022) 2201032. doi: 10.1002/adhm.202201032

  78. [78]

      J. Sun, W. Jia, H. Qi, et al., Adv. Mater. 36 (2024) 2312440. doi: 10.1002/adma.202312440

  79. [79]

      X. Zhang, Z. Wang, H. Jiang, et al., Sci. Adv. 9 (2023) eadh1415. doi: 10.1126/sciadv.adh1415

  80. [80]

      K. Raziyeva, Y. Kim, Z. Zharkinbekov, et al., Biomolecules 11 (2021) 700. doi: 10.3390/biom11050700

  81. [81]

      S. Knoedler, L. Knoedler, M. KaukeNavarro, et al., Mil. Med. Res. 10 (2023) 49.

  82. [82]

      J. Mao, L. Chen, Z. Cai, et al., Adv. Funct. Mater. 32 (2022) 2111003. doi: 10.1002/adfm.202111003

  83. [83]

      Z. Le, M.C. Ramos, Y. Shou, et al., Biomaterials 311 (2024) 122700. doi: 10.1016/j.biomaterials.2024.122700

  84. [84]

      C. Kuan, L. Chang, C. Ho, et al., Biomaterials 314 (2025) 122848. doi: 10.1016/j.biomaterials.2024.122848

  85. [85]

      D. Lv, X. Cao, L. Zhong, et al., Cell Rep. Med. 4 (2023) 101129. doi: 10.1016/j.xcrm.2023.101129

  86. [86]

      W. Zhang, Z. Ge, Y. Xiao, D. Liu, J. Du, Nano Lett. 24 (2024) 9494–9504. doi: 10.1021/acs.nanolett.4c01869

  87. [87]

      Y. Wang, C. Zhou, Z. Li, et al., Bioact. Mater. 41 (2024) 193–206. doi: 10.1016/j.bioactmat.2024.07.020

  88. [88]

      L.C. Shanley, O.R. Mahon, D.J. Kelly, A. Dunne, Acta Biomater. 133 (2021) 208–221. doi: 10.1016/j.actbio.2021.02.023

  89. [89]

      J.M. Moreau, M.O. Dhariwala, V. Gouirand, et al., Sci. Immunol. 6 (2021) eabg2329. doi: 10.1126/sciimmunol.abg2329

  90. [90]

      M. Qu, W. Xu, X. Zhou, et al., Adv. Funct. Mater. 34 (2024) 2314500. doi: 10.1002/adfm.202314500

  91. [91]

      M. Maryanovich, P.S. Frenette, Immunity. 46 (2017) 979–981. doi: 10.1016/j.immuni.2017.06.011

  92. [92]

      B. Nayer, J.L. Tan, Y.K. Alshoubaki, et al., Nat. Commun. 15 (2024) 7863. doi: 10.1038/s41467-024-51353-2

  93. [93]

      Z. Wang, W. Li, L. Gou, et al., Adv. Healthc. Mater. 11 (2022) 2200782. doi: 10.1002/adhm.202200782

  94. [94]

      W. Zhang, Y. Chen, Z. Zhao, et al., Sci. Adv. 9 (2023) eadg6007. doi: 10.1126/sciadv.adg6007

  95. [95]

      Z. Ge, W. Guo, Y. Tao, et al., Adv. Mater. 35 (2023) 2304005. doi: 10.1002/adma.202304005

  96. [96]

      D. Deng, L. Liang, K. Su, et al., Nano Today 60 (2025) 102559. doi: 10.1016/j.nantod.2024.102559

  97. [97]

      J. Wang, J. Zhang, X. Wang, et al., Chin. Chem. Lett. 35 (2024) 109697. doi: 10.1016/j.cclet.2024.109697

  98. [98]

      Z. Shi, Q. Li, L. Mei, Chin. Chem. Lett. 31 (2020) 1345–1356. doi: 10.1016/j.cclet.2020.03.001

  99. [99]

      L. Feng, Q. Peng, L. Miao, et al., Bioact. Mater. 48 (2025) 118–134.

  100. [100]

       J. Wang, X. Ge, Y. Xiang, et al., Chin. Chem. Lett. 36 (2025) 109819. doi: 10.1016/j.cclet.2024.109819

  101. [101]

       S.H. Hamed, E.A. Azooz, E.A.J. Al-Mulla, Nano Biomed. Eng. 15 (2023) 425–435. doi: 10.26599/nbe.2023.9290039

  102. [102]

       Y. Xu, L. Ma, Y. Wang, Chin. Chem. Lett. 36 (2025) 109766. doi: 10.1016/j.cclet.2024.109766

  103. [103]

       S. Cheng, M. Pan, D. Hu, et al., Chin. Chem. Lett. 34 (2023) 108276. doi: 10.1016/j.cclet.2023.108276

  104. [104]

       Y. Yi, Z. Peng, Y. Liu, et al., Acta Pharm. Sin. B 15 (2025) 2237–2252. doi: 10.1016/j.apsb.2025.02.020

  105. [105]

       J. Deng, G. Shi, Z. Ye, et al., Chin. Chem. Lett. 36 (2025) 110496. doi: 10.1016/j.cclet.2024.110496

•