Antioxidant nanozymes for periodontal bone regeneration: multifunctional mechanisms and therapeutic applications

Abstract

The regeneration of periodontitis-related bone defects remains a significant clinical challenge due to the complex and dynamic pathological microenvironment. The primary barrier stems from a self-perpetuating cycle driven by plaque biofilm-induced chronic inflammation, hypoxia, and the consequent overproduction of reactive oxygen species (ROS). Conventional therapeutic approaches are often inadequate in simultaneously targeting these interconnected pathological factors, leading to suboptimal tissue regeneration. In recent years, as an emerging nanobiomaterial, antioxidant nanozymes have provided a promising solution for overcoming the aforementioned therapeutic bottlenecks, owing to their tunable catalytic activity, high stability, and excellent biocompatibility. This review systematically examines the multifaceted roles of ROS in the pathogenesis of periodontitis, with particular emphasis on their suppressive effects on the osteogenic niche. We provide an in-depth analysis of the catalytic mechanisms and design strategies of various antioxidant nanozymes exhibiting superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx)-like activities, and highlight their multifunctional applications in periodontal therapy. These include direct antibacterial and anti-biofilm actions, modulation of the immune-inflammatory milieu to promote macrophage M2 polarization, and facilitation of both osteogenesis and angiogenesis. Notably, the field has advanced from early single-function antioxidants toward the development of intelligent, stimuli-responsive nanoplatforms that integrate multiple enzymatic activities and environmental responsiveness, enabling precise sensing and adaptive intervention within the intricate periodontal microenvironment. Finally, we discuss key challenges facing future research and the translational potential of nanozyme-based therapies, aiming to establish a solid theoretical framework and guide the development of next-generation strategies for periodontitis treatment.

---

1. Introduction

Periodontitis is a chronic inflammatory disease initiated by dental plaque biofilm and driven by dysregulated host immune responses, wherein progressive inflammation results in the irreversible loss of periodontal supporting tissues.¹ In contrast to bone defects at other anatomical sites, periodontal regeneration presents distinct challenges due to the persistent presence of subgingival biofilm, a hypoxic microenvironment, and the consequent overproduction of reactive oxygen species (ROS), collectively establishing a complex and self-sustaining pathological network.² Excessive ROS not only induce direct oxidative damage to cellular macromolecules but also exacerbate the inflammatory cascade, suppress osteogenic differentiation, enhance osteoclast activation, and disrupt angiogenesis. Moreover, ROS interact synergistically with local perturbations such as pH imbalance and immune cell dysfunction within the periodontal pocket, collectively accelerating tissue destruction. As a result, conventional bone graft materials frequently fail to achieve sufficient integration and functional regeneration.^3,4 Consequently, mitigating oxidative stress and reconstructing a conducive periodontal microenvironment have emerged as pivotal therapeutic strategies for overcoming the current limitations in periodontitis-related bone regeneration.

Endogenous enzymes constitute the primary defense system against oxidative stress in biological systems, precisely catalyzing the clearance reactions of ROS. However, the complex synthesis and purification, high production and storage costs, and poor stability of natural enzymes have limited their further application.⁵ Therefore, the development of alternative materials with similar catalytic functions but higher stability and lower cost has emerged as a crucial research direction in this field. In recent years, nanozymes have gradually been regarded as potential substitutes for natural antioxidant enzymes.⁶ They function through a delicate balance between antioxidant and pro-oxidant activities.⁷ On one hand, nano-materials that mimic the activities of superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) can effectively alleviate oxidative stress and prevent damage to cells by catalyzing the conversion of ROS into less reactive or harmless substances.^8,9 On the other hand, pro-oxidant nanozymes that mimic the activities of oxidase (OXD) and peroxidase (POD) generate ROS and selectively induce oxidative stress in pathogenic bacteria or diseased cells.^10–13 Among them, antioxidant nanozymes with SOD, CAT, and GPx-like enzyme activities, with their controllable composition characteristics, high stability, and low cost advantages, have shown unique potential in the treatment of oxidative stress-related diseases.^14–18

Given that the core obstacle to periodontal bone regeneration lies in the persistently elevated oxidative stress microenvironment, this review focuses on antioxidant nanozymes that counteract the pathological process by scavenging ROS, and systematically elaborates their mechanisms of action, therapeutic strategies, and research progress in the context of periodontitis-mediated bone regeneration. The article sequentially analyzes the detrimental effects of oxidative stress in the periodontal pathological microenvironment, outlines the developmental history of antioxidant nanozymes, and classifies them according to catalytic mechanisms such as SOD-like, CAT-like, GPx-like, and multi-enzyme synergistic activities. It further discusses their multifunctional applications in periodontal therapy, including antibacterial/anti-biofilm effects, immunomodulation, and promotion of osteogenesis and angiogenesis. Finally, it summarizes recent advances in intelligent nano-platforms responsive to pH, ROS, and specific enzymes. This review aims to provide a theoretical basis and directional guidance for developing antioxidant nanozyme-based therapeutic strategies for periodontal bone regeneration.

2. The regulatory role of ROS in the pathological process of periodontitis

ROS represent a broad class of highly reactive oxygen-containing molecules, encompassing oxygen-derived free radicals such as superoxide anion (O₂˙⁻), hydroxyl radical (˙OH), and nitric oxide radical (NO˙), as well as non-radical derivatives including hydrogen peroxide (H₂O₂) and hypochlorous acid (HOCl).¹⁹ These molecules exert distinct biological effects depending on the physiological or pathological context. Under physiological conditions, ROS in periodontal tissues are primarily generated through enzymatic systems such as the NADPH oxidase (NOX)/Dual oxidase (DUOX) family and the mitochondrial electron transport chain. At low concentrations, ROS are not only involved in host defense by eliminating invading pathogens but also play key signaling roles in critical cellular processes such as proliferation, differentiation, and immune regulation, thereby serving as essential signaling molecules for maintaining periodontal tissue homeostasis.²⁰ However, under the chronic inflammatory conditions of periodontitis, pathogenic bacteria within the plaque biofilm, such as Porphyromonas gingivalis (P.g.), activate the host innate immune response. This triggers massive neutrophil recruitment and hyperactivation, leading to a respiratory burst that generates excessive ROS. The overproduction of ROS overwhelms the clearance capacity of endogenous antioxidant systems, including SOD, CAT, and GPx, resulting in oxidative stress and a disruption of redox homeostasis. Excessive ROS not only directly damage key cellular macromolecules such as proteins, lipids, and DNA but also indirectly exacerbate periodontal bone destruction and resorption by activating several critical signaling pathways. First, ROS act as potent activators of the nuclear factor-κB (NF-κB) pathway. They induce oxidative inhibition of the regulatory subunit of IκB kinase (IKK), promoting IκBα degradation and facilitating nuclear translocation of the NF-κB p65 subunit, which initiates the transcription of pro-inflammatory genes. This process creates a “cytokine storm”, upregulating the expression of inflammatory mediators such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β).²¹ Second, ROS disrupt the Receptor Activator of Nuclear Factor Kappa-B Ligand (RANKL)/Receptor Activator of Nuclear Factor Kappa-B (RANK)/Osteoprotegerin (OPG) axis, significantly increasing the RANKL/OPG ratio, which strongly promotes osteoclastogenesis and leads to alveolar bone loss.²² Concurrently, ROS activate the Mitogen-Activated Protein Kinase (MAPK) pathways, including c-Jun N-terminal kinase (JNK), p38 MAPK (p38), and Extracellular Signal-Regulated Kinase 1/2 (ERK1/2), which participate in regulating inflammatory factor expression, apoptosis, and osteoclast differentiation.^23–25 Furthermore, sustained oxidative stress suppresses endogenous antioxidant pathways such as nuclear factor erythroid 2-related factor 2/heme oxygenase-1 (Nrf2/HO-1), thereby impairing the cellular self-defense capacity.²⁶

Based on the aforementioned pathological basis, ROS in the periodontitis microenvironment further interact with factors such as hypoxia, pH imbalance, and immune dysregulation, forming a complex positive-feedback network that collectively drives disease progression. The persistent hypoxic environment within periodontal pockets induces mitochondrial dysfunction, leading to increased generation of mitochondrial superoxide (mtROS). This in turn suppresses prolyl hydroxylase activity, resulting in abnormal accumulation of hypoxia-inducible factor-1α (HIF-1α) protein. HIF-1α subsequently upregulates the expression of target genes such as NOX4, elevating intracellular ROS levels and amplifying oxidative stress signals. Concurrently, hypoxia and sustained oxidative stress further exacerbate mitochondrial dysfunction. Excessive mtROS not only aggravates oxidative damage to the mitochondria itself but also leaks into the cytoplasm, combining with ROS from other sources to intensify the overall oxidative stress level.^27–29 The accumulation of ROS can also lead to microvascular endothelial cell dysfunction and oxidative damage through direct or indirect pathways, such as activating matrix metalloproteinases.^30,31 This results in inadequate perfusion of the periodontal local tissue, reduced efficiency of oxygen and nutrient delivery, accumulation of metabolic waste, and worsened tissue hypoxia, forming a self-amplifying vicious cycle. Meanwhile, the hypoxic environment significantly enhances the metabolic activity of anaerobic microbiota represented by P.g. and Fusobacterium nucleatum (F.n.). Bacterial protein decomposition produces alkaline substances such as ammonia, while host immune cells, such as neutrophils and macrophages, undergo vigorous anaerobic glycolysis under hypoxic conditions, producing and accumulating large amounts of organic acids like lactic acid. Together with bacterial metabolites, this leads to a decrease in local pH. This weakly acidic environment not only favors the adaptation and proliferation of some periodontal pathogens but also inhibits normal host cell functions, exacerbating tissue damage.³² In this process, excessive ROS accumulation promotes macrophage polarization towards the M1 phenotype via the HIF-1α signaling pathway, thereby exacerbating inflammatory responses in bone tissue. Pro-inflammatory factors such as IL-6, IL-1β, and TNF-α released by M1 macrophages can activate osteoclasts, accelerating bone loss. Therefore, inducing a shift towards M2 polarization by reducing ROS levels may help alleviate inflammation and promote bone tissue repair.³³ Furthermore, pathogen-associated molecular patterns, such as lipopolysaccharide (LPS) derived from P. g., synergize with accumulated ROS under hypoxic and acidic conditions to activate the NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome. This promotes the maturation and release of pro-inflammatory cytokines like IL-1β via caspase-1, persistently amplifying the immune response,^34,35 thereby further reinforcing the cycle of oxidative stress and inflammatory damage.

In summary, the pathways directly activated by ROS, such as NF-κB and MAPK, do not act in isolation from key processes like the hypoxia-HIF-1α axis and NLRP3 inflammasome activation described herein. Instead, they intertwine within the periodontal microenvironment to form a self-amplifying vicious cycle. This network systematically drives disease progression, collectively leading to immune microenvironment dysregulation, uncoupled osteogenesis-osteoclastogenesis, and impaired tissue regeneration (Fig. 1). Understanding this multidimensional and complex regulatory network provides critical guidance for designing periodontitis treatment strategies. An ideal therapeutic approach should not only precisely regulate ROS to physiological levels but also concurrently target hypoxia-induced HIF signaling, modulate local pH homeostasis, improve immune cell function, and restore microbial ecological balance. This offers a clear direction for the future design of nanozyme-based synergistic therapy systems. Specifically, constructing multifunctional nanoplatforms capable of ROS scavenging, oxygen supply, pH regulation, and immunomodulation holds promise for systematically reshaping the periodontitis microenvironment and creating favorable conditions for periodontal tissue regeneration. In particular, integrating the catalytic properties of nanozymes with microenvironment-responsive release characteristics could enable spatiotemporally matched and specific modulation of the pathological microenvironment. This approach would overcome the lack of specificity associated with conventional antioxidant treatments, offering a novel strategy for repairing periodontitis-related bone defects.

Fig. 1 Schematic illustration of ROS-mediated pathological mechanisms in periodontitis, including oxidative stress, inflammatory response, and alveolar bone loss. Created with MedPeer (medpeer.cn).

3. Impact of oxidative stress on the periodontal osteogenic microenvironment

Oxidative stress interacts with various factors in the periodontitis microenvironment, such as hypoxia, pH imbalance, and immune dysregulation, forming a self-amplifying pathological cycle. The ultimate outcome of this cycle is the disruption of homeostasis in the periodontal osteogenic microenvironment.^36–38 Periodontal osteogenic homeostasis fundamentally relies on a dynamic balance among osteoblast-mediated bone formation, osteoclast-driven bone resorption, and the vascularization process.^39,40 Through multi-level and multi-target mechanisms, oxidative stress systematically disrupts this balance. It not only significantly inhibits bone regeneration capacity but also continuously exacerbates the progression of alveolar bone defects. This process involves several interconnected pathological aspects, including impaired cell differentiation, dysregulated immune inflammation, metabolic reprogramming, and microcirculatory failure, collectively leading to irreversible destruction of the periodontal supporting tissues, particularly the bone tissue.

3.1. Inhibition of mesenchymal stem cell differentiation and osteoblast activation

Oxidative stress profoundly impairs the osteogenic differentiation capacity of mesenchymal stem cells (MSCs) and the function of osteoblasts through multi-level, interrelated pathological mechanisms. Its impact extends from intracellular signaling and metabolic disturbances to fundamental alterations in cell fate, and is further modulated negatively by the immune system at the microenvironmental level, collectively leading to a failure of bone formation.

3.1.1. Disruption of key signaling pathways and cellular metabolism. The core effect of oxidative stress lies in directly disturbing the molecular basis that maintains the osteogenic fate in MSCs. Studies show that excessive accumulation of ROS not only directly impairs the viability and proliferation of bone marrow-derived mesenchymal stem cells (BMSCs), but also initiates a cascade of functional impairment by suppressing key pro-survival and osteogenic signaling pathways. For instance, treatment of BMSCs with H₂O₂ significantly reduces cell activity and disrupts intracellular redox balance, manifested by increased levels of ROS and malondialdehyde, alongside decreased activity of antioxidant enzymes such as SOD and GPx. This oxidative microenvironment further inhibits the protein kinase B (Akt)/forkhead box protein O1 (FoxO1) signaling pathway and enhances the expression of Cleaved caspase-3. This leads to the downregulation of key osteogenic genes, such as alkaline phosphatase (ALP), collagen type I (Collagen I), osteopontin (OPN), and Runt-related transcription factor 2 (Runx2), ultimately resulting in weakened ALP activity and impaired mineralization nodule formation.⁴¹ Moreover, when Akt signaling is suppressed, nuclear translocation of FoxO3 is enhanced, which subsequently inhibits the Wnt/β-catenin signaling pathway that is crucial for osteogenic differentiation, thereby interfering with BMSC proliferation and osteogenic differentiation.⁴² This process is closely intertwined with mitochondrial dysfunction and metabolic disturbance. ROS can cause abnormal mitochondrial morphology and decreased membrane potential, triggering an energy metabolism crisis.⁴³ For example, in glucocorticoid-induced injury models, oxidative stress inhibits ERK1/2 and cAMP response element-binding protein (CREB) signaling, while it promotes adipogenic differentiation of MSCs via upregulation of peroxisome proliferator-activated receptor gamma (PPARγ). This disrupts the osteogenic–adipogenic balance, effectively substituting bone-forming capacity with adipogenesis.⁴⁴ Thus, oxidative stress cooperatively suppresses the osteogenic potential of MSCs at both the signal transduction and energy metabolism levels.

3.1.2. Induction of epigenetic alterations and cellular senescence. Beyond interfering with immediate function, oxidative stress can cause deeper and longer-lasting damage to MSCs by inducing abnormal epigenetic modifications, even locking them into a senescent phenotype. For example, transforming growth factor-beta 1 (TGF-β1) can induce excessive ROS accumulation in human periodontal ligament stem cells (PDLSCs), which silences the PRKAG2 gene through mechanisms such as DNA methylation and activates the ataxia telangiectasia mutated (ATM) signaling pathway, ultimately leading to DNA damage and cellular senescence.⁴⁵ Additionally, histone modifications, particularly trimethylation of histone H3 at lysine 27 (H3K27me3), represent another key epigenetic mechanism linking oxidative stress to cellular senescence.^46–49 Studies show that in senescent BMSCs, expression of the histone methyltransferase enhancer of zeste homolog 2 (EZH2) is significantly upregulated, leading to a genome-wide increase in H3K27me3 modification. This modification acts as a repressive mark that directly acts on the promoter regions of specific genes. Mechanistically, EZH2 enhances H3K27me3 modification at the FoxO1 gene promoter, thereby inhibiting FoxO1 transcription. Since FoxO1 is a core transcription factor regulating the antioxidant defense system, its downregulation impairs the expression of key antioxidant enzymes such as CAT and SOD, exacerbating intracellular ROS accumulation and forming a vicious cycle.⁵⁰

The above epigenetically driven changes mediated by DNA methylation and histone methylation collectively lead to sustained activation of the DNA damage response and irreversible changes in cell fate. Senescent cells exhibit G2 phase cell-cycle arrest, upregulation of the senescence-associated secretory phenotype (SASP), and a significant decline in proliferation and osteogenic potential. Notably, this epigenetically driven senescent phenotype is partially reversible: studies have confirmed that treatment of senescent BMSCs with the EZH2 inhibitor DZNep effectively reduces H3K27me3 levels, reactivates FoxO1 and its downstream antioxidant pathway, thereby alleviating oxidative stress and restoring osteogenic differentiation capacity.⁵⁰ Similarly, treatment with the antioxidant N-acetylcysteine can reverse the senescent phenotype induced by TGF-β1.⁴⁵ The success of these interventions jointly supports the central role of oxidative stress and its mediated epigenetic dysregulation in driving stem cell senescence. This oxidative stress-driven premature cellular aging implies depletion of the stem cell pool and permanent decline in regenerative capacity.

3.1.3. Disruption of the immune-osteogenic regulatory microenvironment. In the complex pathological environment of periodontitis, the damaging effects of oxidative stress do not occur in isolation; they amplify and interact with the immune-inflammatory network to shape an osteoinhibitory microenvironment. Activated immune cells serve as key mediators in this process. For instance, in active periodontitis tissues, extensively infiltrating CD19⁺ B cells can highly express TGF-β1, which activates phosphorylated (p-) mothers against decapentaplegic homolog 2/3 (p-Smad2/3) signaling and inhibits Runx2 transcription, thereby suppressing the osteogenic differentiation capacity of MSCs.⁵¹ Moreover, B lymphocytes and plasma cells, as dominant populations in the inflammatory infiltrate,⁵² are not only antibody producers but also pro-inflammatory effector cells that highly express RANKL, directly participating in osteoclast activation and exacerbating bone resorption.⁵³ This dysregulation of the immune-osteogenic axis, together with oxidative stress, forms a self-amplifying pathological cycle that continuously aggravates alveolar bone resorption and destruction of periodontal supporting tissues.

Oxidative stress further worsens the microenvironment by modulating the balance of T-cell subsets. For example, Th17 cells expand significantly in lesions and produce cytokines such as IL-17 and IL-22, which not only enhance neutrophil recruitment and inflammatory responses but also directly upregulate the expression of matrix metalloproteinases (MMPs) and RANKL, synergistically promoting connective tissue degradation and bone destruction.⁵⁴ Meanwhile, macrophages polarize to the M1 phenotype under inflammatory drive, secreting pro-inflammatory factors such as TNF-α and IL-1β, further inhibiting osteoblast function and activating osteoclasts. In contrast, the function of anti-inflammatory M2 macrophages and regulatory T cells (Tregs) is impaired, failing to effectively suppress inflammation and resulting in loss of reparative balance in the microenvironment.³⁶

In summary, oxidative stress systematically inhibits the osteogenic differentiation of MSCs through three interconnected levels. These encompass the disruption of intracellular signaling and metabolism, the induction of epigenetic senescence, and the perturbation of immune microenvironment homeostasis. These mechanisms are not isolated; rather, they are interwoven and mutually reinforcing. Ultimately, they act in concert to drive the loss of regenerative capacity and the irreversible progression of bone defects within the periodontal osteogenic microenvironment.

3.2. Promotion of inflammatory response and osteoclast activity

Oxidative stress participates in the regulation of inflammatory response and the activation of osteoclasts in periodontitis-related bone defects through both direct and indirect pathways.

3.2.1. Direct activation of osteoclastogenesis and the inflammasome. At the level of direct action, excessively produced ROS can themselves act as potent intracellular signaling molecules. They can directly activate NOX2 and inducible nitric oxide synthase (iNOS) in osteoclast precursors, generating more reactive nitrogen species (RNS) and creating nitrosative stress, which synergistically amplifies oxidative damage.⁵⁵ This further initiates and amplifies the cascade of NF-Kb (involving IKKβ/IκBα) and MAPK (involving p38, JNK, ERK) signaling pathways. Activated NF-κB translocates into the nucleus and, in concert with transcription factors downstream of the MAPK pathway, significantly upregulates the expression of pro-inflammatory cytokines such as TNF-α, IL-6, and IL-1β.⁵⁶ These cytokines not only directly exacerbate local inflammation but are also potent inducers of osteoclast differentiation.

The regulation of osteoclast differentiation by oxidative stress is a central link. In models deficient for the energy sensor AMP-activated protein kinase (AMPK), activation of the nuclear factor erythroid 2-related factor 2/antioxidant response element (NRF2/ARE) pathway is impaired. Elevated ROS oxidatively modify the transcription factor FoxO1, which negatively regulates osteoclast differentiation, thereby weakening its transcriptional activity. Simultaneously, AMPK deficiency enhances RANKL-induced phosphorylation of p38 and JNK, further upregulating the expression of the key osteoclastogenic transcription factor nuclear factor of activated T cells 1 (NFATc1). This leads to significantly increased expression of its target genes (e.g., TRAP, Ctsk) and ultimately a marked increase in TRAP-positive multinucleated osteoclasts.⁵⁷ Furthermore, ROS can directly stimulate osteoblasts (OBs) and osteocytes (OCYs) to highly express RANKL and downregulate the expression of its decoy receptor OPG by activating the ERK1/2 and JNK pathways, resulting in a substantially increased local RANKL/OPG ratio and providing a potent initiating signal for osteoclast differentiation.⁵⁸ Further mechanistic studies indicate that oxidative stress can also directly stimulate RANKL expression in osteoblasts via activation of ERKs and heat shock factor 2 (HSF2). Additionally, oxidative stress can induce osteocyte apoptosis via activation of pathways including ERK1/2, JNK, and p38. The apoptotic osteocytes themselves can then act as a signal, stimulating osteoblasts and bone-lining cells to produce more RANKL, thereby forming a positive feedback loop that amplifies the osteoclastogenic signal.⁵⁹

Concurrently, ROS also play a crucial role in inflammasome activation, serving as a key trigger for NLRP3 inflammasome assembly. Specifically, mtROS can induce oligomerization of NLRP3, promoting recruitment of the adaptor protein apoptosis-associated speck-like protein containing a CARD (ASC) and subsequent activation of caspase-1. Activated caspase-1 then catalyzes the conversion of pro-IL-1β and pro-IL-18 into their bioactive mature forms. These inflammatory cytokines not only exacerbate the local inflammatory response but can also induce mitochondrial dysfunction, creating a positive feedback loop that continuously amplifies inflammatory signaling.^60,61 It is noteworthy that in the specific pathological context of periodontal disease, the pattern of inflammasome activation is more complex. Beyond the presence of specific NLRP3 ligands/activators, the chronic inflammatory microenvironment of periodontitis is characterized by high levels of ROS, hypoxia, and accumulation of damage-associated molecular patterns (DAMPs) accompanying tissue degradation. ROS and DAMPs (the latter may also induce ROS production) can activate not only NLRP3 but also various other inflammasomes, including absent in melanoma 2 (AIM2), NLRP1, NLRC4, and NLRP6.⁶²

3.2.2. Immune cell polarization and metabolic reprogramming. At the more complex microenvironmental level, oxidative stress exerts indirect yet powerful pro-inflammatory and pro-osteoclastogenic effects by modulating immune cell function, with macrophage polarization imbalance being a key hub. During the early phase of normal healing, controlled inflammation dominated by M1-type macrophages is beneficial for pathogen clearance. However, a persistent oxidative stress environment disrupts this precise balance. When pro-inflammatory signals trigger glycolytic activation via pathways such as Janus kinase-signal transducer and activator of transcription (JAK-STAT), NF-κB, and MAPK, the large amounts of ROS produced are not merely byproducts of metabolic reprogramming but core mediators driving the amplification of inflammatory signals.⁶³ These ROS directly enhance the transcriptional activity of NF-κB by oxidizing cysteine residues in its regulatory proteins,²¹ while also activating MAPK pathways to promote pro-inflammatory gene expression. Sustained glycolytic activation leads to a break in the tricarboxylic acid (TCA) cycle and succinate accumulation. Succinate stabilizes HIF-1α by inhibiting prolyl hydroxylase domain proteins (PHDs),⁶⁴ further amplifying the production of inflammatory factors like IL-1β and establishing a positive feedback loop.^65–67

The synergistic effect of metabolic disturbance and epigenetic modifications is also a key link in oxidative stress-driven polarization. Glycolysis consumes NAD⁺, leading to decreased activity of sirtuin (SIRT) deacetylases.^68,69 Concurrently, ROS affect DNA methylation patterns by influencing Ten-eleven translocation 2/DNA methyltransferase (TET2/DNMT) activity.^70,71 Together, these changes contribute to histone hyperacetylation and an open chromatin state at the promoter regions of pro-inflammatory genes such as TNF-α and IL-6. Furthermore, glycolytic intermediates like succinate and acetyl-CoA not only act as metabolic messengers for inflammation but also serve as substrates for histone modifications, helping to maintain the M1 polarization phenotype via epigenetic marks like histone H3 lysine 27 acetylation (H3K27ac).^65,72

Notably, the self-amplifying nature of ROS perpetuates this process. NADPH oxidase activation and mitochondrial dysfunction further increase ROS production, and ROS in turn promote IL-1β maturation via NLRP3 inflammasome activation, forming a vicious cycle between inflammatory signaling and oxidative stress. This multi-tiered regulatory network, initiated by oxidative stress, amplified by metabolic reprogramming, and locked in place by epigenetic modifications, ultimately establishes a stable M1 macrophage polarization state and effectively suppresses the activation of M2-associated pathways such as STAT6/peroxisome proliferator-activated receptor gamma (PPARγ).⁷³

3.3. Impairment of angiogenesis

The inhibitory effect of oxidative stress on angiogenesis is particularly critical during the repair of periodontal tissues. Impaired blood vessel formation directly leads to local ischemia and hypoxia, which subsequently compromises nutrient supply and cellular metabolic activity in bone defect areas, ultimately hindering effective bone regeneration.

3.3.1. Direct damage to endothelial cell structure and function. ROS can be generated by all vascular cell types, including endothelial cells, smooth muscle cells, adventitial fibroblasts, and perivascular adipocytes.⁷⁴ The vasculature has two primary endogenous sources of ROS: the mitochondrial electron transport chain and NADPH oxidases.^74–76 In the periodontal inflammatory microenvironment, xanthine oxidase (XO) is activated under inflammatory and hypoxic conditions, generating O₂˙⁻ and H₂O₂ through the oxidation of xanthine/hypoxanthine. Under hypoxic conditions (1% O₂), H₂O₂ is the primary product of XO activity.⁷⁷ Studies have demonstrated that H₂O₂ increases endothelial permeability by activating protein kinase C (PKC) and tyrosine kinase pathways, and may further contribute to vascular dysfunction by limiting NO synthesis via PKC/RhoA pathway activation.^78–80 Furthermore, research by Csiszar et al. confirmed that H₂O₂ induces NF-κB activation in endothelial cells in a concentration dependent manner.⁸¹ Shono et al. have shown that H₂O₂ (0.1–0.5 mM, for 15 minutes) induces tubulogenesis in human microvascular endothelial cells by stimulating IL-8 expression via NF-κB activation,⁸² whereas a high concentration of H₂O₂ (125 mM) causes vascular endothelial cell damage.⁸³

3.3.2. Dysregulation of key pro angiogenic signaling pathways. Oxidative stress functionally suppresses angiogenesis by interfering with core pro angiogenic signaling pathways. The Kelch like ECH associated protein 1 (Keap1)-Nrf2-ARE pathway is central to cellular defense against oxidative stress, and Nrf2 activation is crucial for angiogenesis. Research indicates that Nrf2 deficiency impairs the survival, migration, and homing capacity of bone marrow derived PACs. Concurrently, loss of Nrf2 function in endothelial cells leads to insufficient expression of protective proteins like HO-1, making them more susceptible to ROS induced apoptosis and significantly reducing their vascular endothelial growth factor (VEGF) induced in vitro lumen formation capacity.⁸⁴ Normally, Nrf2 is retained in the cytoplasm under the regulation of Keap1 and remains in a state of low activity under physiological conditions. Under oxidative stress, modification of sensitive cysteine residues on Keap1 leads to Nrf2 dissociation, nuclear import, and heterodimerization with musculoaponeurotic fibrosarcoma (MAF) proteins. This dimer binds to the ARE, initiating the transcription of Nrf2 downstream target genes, including antioxidant enzymes and phase II detoxifying enzymes such as HO-1, NAD(P)H quinone oxidoreductase 1 (NQO1), and SOD.^85,86

VEGFs and their receptor VEGFR2 play a central role in regulating angiogenesis.⁸⁷ VEGF primarily activates VEGFR2, mediating multiple downstream signaling pathways including phospholipase C gamma (PLCγ)-ERK1/2, phosphatidylinositol 3 kinase (PI3K)-AKT-mammalian target of rapamycin (mTOR), and SRC kinase, thereby regulating endothelial cell proliferation, migration, and vessel formation.⁸⁸ Given the critical role of VEGFR2 signaling in angiogenesis, modulating VEGFR2 activity/activation represents an important mechanism for controlling blood vessel formation. In the context of comorbid diabetes and periodontitis, oxidative stress induced by hyperglycemia is considered a key pathological link between the two conditions.⁸⁹ Studies show that ROS induced by hyperglycemia can directly phosphorylate VEGFR2 via oxidative stress, activating downstream signaling independently of VEGF ligand binding,⁹⁰ thereby contributing to the development of diabetes related vasculopathies.^91,92 In diabetic retinopathy, abnormal VEGFR2 activation may lead to a feedback driven increase in VEGF levels, promoting pathological angiogenesis.⁹²

The regulation of VEGF expression exhibits tissue specificity in periodontal tissues. A systematic review by Nardi et al. points out that VEGF expression is generally upregulated in the gingival tissues and gingival crevicular fluid of patients with diabetic periodontitis, particularly in gingival epithelium and vascular endothelial cells, suggesting that chronic inflammation and insulin resistance may drive VEGF overexpression, exacerbating pathological angiogenesis in periodontitis.⁹³ However, Hoshino et al. found that VEGF expression was downregulated in the periodontal tissues of diabetic mice, a mechanism potentially related to the suppression of the PI3K/Akt/SP1 signaling axis by hyperglycemia.⁹⁴ Although inconsistent results exist at the mRNA level, possibly due to detection methods or sample heterogeneity, most evidence still supports an overall upregulating trend of VEGF expression in periodontal tissues under diabetic conditions.

It is important to emphasize that, regardless of VEGF expression levels, the abnormal activation of VEGFR2 under diabetes associated oxidative stress often leads to dysregulation of vascular signaling pathways, promoting the formation of incomplete, dysfunctional pathological blood vessels rather than the ordered vascular network essential for bone regeneration. Therefore, whether by impairing protective pathways like Nrf2 or by disrupting angiogenic regulation through aberrant VEGFR2 phosphorylation, oxidative stress ultimately compromises functional neovascularization, fundamentally hindering the vascularization process upon which alveolar bone repair depends.

3.3.3. Exacerbation of angiogenic impairment by the hypoxic microenvironment. In areas of periodontitis and bone defects, hypoxia is another key pathological feature, which forms a vicious cycle with oxidative stress, jointly exacerbating angiogenic impairment. The hypoxic microenvironment activates enzymes like XO, promoting massive ROS generation. ROS impairs vasodilation by damaging the bioavailability of NO, an effective vasodilator and a key regulator of endothelial function.^95,96 Simultaneously, ROS trigger the autoxidation of hemoglobin, accelerating the conversion of ferrous iron (Fe²⁺) to ferric iron (Fe³⁺) within the hemoglobin molecule, leading to excessive methemoglobin (Met Hb) formation (>1%). Methemoglobin cannot reversibly bind oxygen molecules, causing impaired tissue oxygen delivery, a leftward shift of the oxygen dissociation curve, and further aggravating tissue hypoxia and endothelial dysfunction.⁹⁷ Furthermore, methemoglobin possesses pro oxidant properties that can exacerbate oxidative stress, causing structural damage and functional abnormalities in endothelial cells, including increased oxidative damage and reduced NO bioavailability,⁹⁸ leading to decreased vasodilation capacity, reduced local blood flow, and accompanied by upregulation of inflammatory markers, IL-1β release, neutrophil activation, and Weibel Palade body degranulation.^99,100

The interplay between hypoxia and oxidative stress disrupts pro angiogenic signaling pathways, resulting in insufficient neovascularization. Recent interventional studies provide converse validation of this mechanism. For example, a certain bone tissue engineering material loaded with catalase can catalyze the decomposition of hydrogen peroxide into water and oxygen, effectively improving the hypoxic state in bone defect areas. The oxygen rich microenvironment can promote new blood vessel formation, providing necessary nutrient and oxygen support for tissue repair. Meanwhile, the poly(propylene sulfide) (PPS) on the surface of oxygen loaded NPs in this material can also undergo hydrophilic changes in response to excess ROS, further releasing O₂ on demand to dynamically regulate local oxygen concentration, thereby optimizing the angiogenic microenvironment.¹⁰¹ Similarly, a biological material encapsulating ferrihydrite (Fht) nanozymes, by virtue of its CAT like activity, can sustainably catalyze endogenous H₂O₂ to generate O₂, alleviating the local hypoxic microenvironment. The improvement of hypoxia can activate HIFs, subsequently promoting the transcription of key genes such as VEGF and erythropoietin (EPO), thereby enhancing angiogenesis.¹⁰²

In summary, the inhibitory effect of oxidative stress on angiogenesis is intertwined with the two pathological processes mentioned above, namely the impeded osteogenic differentiation of mesenchymal stem cells and the persistent inflammation and enhanced osteoclastic activity. These factors collectively impede periodontitis-related bone defects. Oxidative stress not only directly impairs endothelial cell function and the formation of new vascular networks, leading to inadequate tissue perfusion and exacerbated hypoxia but also amplifies microenvironmental imbalance through extensive crosstalk with inflammatory signaling and cellular differentiation processes. This synergistic disruption of osteoblast function, the immune-inflammatory milieu, and the nutritional support system underscores that the core challenge in repairing periodontitis-related bone defects lies in the concurrent dysregulation of multiple homeostatic mechanisms. Understanding this complex network is crucial for developing intervention strategies targeting oxidative stress, inflammation, and angiogenesis.

4. Historical evolution of antioxidant nanozymes

Nanozymes are a class of nanomaterials endowed with intrinsic enzyme-like properties.^103,104 Since the 1990s, such materials have attracted scholarly attention, and the term “nanozyme” was formally proposed by Scrimin et al. in 2004.¹⁰⁵ In 2007, Yan et al. discovered the intrinsic horseradish peroxidase-like activity of ferromagnetic nanoparticles (Fe₃O₄ NPs), establishing a research paradigm for investigating the biocatalytic properties of nanomaterials using enzymatic methodologies.¹⁰⁶ Subsequently, approximately 1200 types of nanomaterials with biocatalytic properties have been discovered, including metal oxides, noble metals, carbon-based materials, and their hybrids, with research and application primarily focused on fields such as medicine, healthcare, diagnostics, and analytics.¹⁰⁷ In 2016, Gao et al. first reported that Fe₃O₄ NPs could generate ROS by exhibiting peroxidase-like activity to control dental plaque,¹⁰⁸ which garnered widespread attention for nanozymes in the field of oral health. Research at this stage primarily leveraged their ROS-generating catalytic property to achieve antibacterial and anti-biofilm effects within the oral cavity.^109–111

However, as research progressed, it became increasingly recognized that although this “pro-oxidant” strategy reliant on catalytic ROS generation offers efficient and broad-spectrum antibacterial advantages, it may exacerbate oxidative stress in local tissues within complex pathological environments like periodontitis due to excessive and uncontrolled ROS production. This, in turn, can potentially disrupt normal host cell function and impede tissue healing. Consequently, to achieve more precise and reparative therapies for complex oral diseases, a significant shift in research focus occurred.¹¹² The concept of “antioxidant nanozymes” was frequently mentioned, with researchers generally categorizing nanozymes that scavenge ROS or inhibit oxidative stress as the antioxidant class.^11,113,114 It was not until 2024 that Yan et al., based on the overall catalytic effect of nanozymes, explicitly classified them into two categories: antioxidant nanozymes and pro-oxidant nanozymes. The core distinguishing criteria established were the direction of their regulatory effect on ROS (scavenging vs. generating) and their application goal (protection vs. killing).¹¹⁵

In recent years, nanozyme research has gradually evolved from utilizing a single “pro-oxidant” catalytic function for efficient antibacterial purposes towards the development of intelligent systems that integrate “antioxidant”, “anti-inflammatory” and “osteogenic” multifunctional synergy.^116–118 Therefore, in-depth investigation into the catalytic mechanisms of antioxidant nanozymes is crucial for achieving the precise coordination of multifunctional activities such as antioxidation and anti-inflammation. This forms a vital prerequisite for the rational design of nanozymes tailored to the pathological characteristics of periodontal bone defects.

5. Classification and catalytic mechanisms of antioxidant nanozymes

The excessive oxidative stress microenvironment present in periodontitis-related bone defect directly mediates the amplification of inflammation, inhibition of osteogenesis, and activation of bone resorption. Therefore, the targeted elimination of excess ROS and restoration of redox balance represent a key strategy for interrupting this vicious cycle and promoting bone regeneration. Based on the type of enzymatic activity, antioxidant nanozymes can be classified into two main categories: those with single-enzyme activity and those with multi-enzyme activity. Single-activity nanozymes can specifically scavenge particular types of ROS. For instance, SOD, CAT, and GPx are specialized in eliminating O₂˙⁻, decomposing H₂O₂, and reducing peroxides and organic hydroperoxides, respectively. In contrast, multi-enzyme nanozymes, through carefully designed active sites or composite structures, can remove multiple ROS simultaneously or in a cascade manner (such as clearing O₂˙⁻, H₂O₂, and ˙OH concurrently), and may even achieve synergistic effects via cascade reactions.^11,119 Understanding the specific catalytic mechanisms of these nanomaterials, including active site structure, reaction kinetics, and substrate specificity, forms the basis for the rational design and optimization of their biological applications. This section will systematically elaborate on and analyze the classification of antioxidant nanozymes and their catalytic principles, thereby establishing a theoretical foundation for subsequent discussion of their application in the regeneration of periodontitis-related bone defects.

5.1. Single-enzyme activity nanozymes

In the study of antioxidant nanozymes, the design of single-activity nanozymes aimed at the precise scavenging of specific types of ROS has become an important direction.¹²⁰ Unlike multi-enzyme activity materials designed for the synergistic regulation of multiple ROS, these nanozymes primarily rely on atomic-level structural control to construct highly uniform active sites, enabling them to catalyze the conversion of a specific ROS in a highly specific and efficient manner. This high degree of catalytic specificity not only avoids potential reductions in catalytic efficiency caused by multiple substrates competing for the same active site but also significantly enhances the catalytic stability and behavioral predictability of the material within complex pathological environments.^121,122 In complex pathological settings such as periodontitis, where multiple ROS interact, this characteristic allows single-activity nanozymes to continuously and stably eliminate their specific targets unaffected by the presence of other ROS, thereby providing a more reliable therapeutic strategy for precisely interrupting the vicious cycle of oxidative stress. Therefore, the following section will focus on representative single-activity nanozymes, such as SOD-like, CAT-like, and GPx-like nanozymes, to elaborate on how precise design of their active centers achieves highly efficient and specific catalytic functions, and to discuss their application potential and design strategies in different pathological environments.

5.1.1. SOD-like activity. As a key member of the ROS family, O₂˙⁻ is primarily generated by the NOX family and the mitochondrial electron transport chain, playing a central role in the pathological progression of periodontitis.^123,124 Excessive ROS production triggers the body's antioxidant defense mechanisms. As the first line of defense, SOD specifically catalyzes the dismutation of the highly toxic O₂˙⁻ into the less toxic H₂O₂ and O₂. If the resulting H₂O₂ is not promptly cleared by subsequent antioxidant enzymes such as CAT and GPx, it can be converted into the more reactive ˙OH via transition metal ion-catalyzed Fenton reactions.¹²⁵ This perpetuates the oxidative stress microenvironment, directly leading to oxidative damage of biological macromolecules including proteins, lipids, and nucleic acids in periodontal tissues.

The catalytic function of SOD is highly dependent on specific transition metal ions at its active center. Natural SODs are primarily classified into four subtypes based on their bound metal cofactors: Cu/Zn-SOD, Mn-SOD, Fe-SOD, and Ni-SOD.¹²⁶ Although these isoforms are encoded by different genes, they all feature specific transition metals with variable valence states (e.g., Cu, Mn, Fe, Ni) as their active centers and share a similar catalytic mechanism. The core mechanism relies on the valence-variable nature of the metal ions (e.g., Cu²⁺/Cu⁺ or Mn³⁺/Mn²⁺), facilitating electron transfer through successive redox reactions to catalyze the dismutation of O₂˙⁻ in two steps.¹²⁷ A typical catalytic cycle involves two half-reactions: the oxidized metal ion first reduces one O₂˙⁻ molecule to O₂, and then the reduced metal ion oxidizes another O₂˙⁻ molecule to H₂O₂, thereby regenerating its original oxidation state.¹²⁸ Overall, this cycle converts two O₂˙⁻ molecules into one H₂O₂ molecule and one O₂ molecule, effectively scavenging O₂˙⁻.

Single-atom nanozymes (SAzymes), featuring well-defined active sites and tunable structures, hold significant potential for achieving high-selectivity catalysis. Recently, researchers have developed a novel Cu-SAzyme by mimicking the electronic and structural characteristics of natural copper-only superoxide dismutase (SOD5). This material employs an innovative “post-adsorption strategy”, which separates carbonization from metal coordination, enabling the precise construction of an unsaturated Cu–N₄ coordination configuration (Fig. 2a). This design not only replicates the histidine-coordinated geometry of Cu²⁺ in native Cu-SOD5 but also achieves dual electronic structure simulation by modulating the electron spin state of Cu²⁺.¹²⁹ As shown in Fig. 2b, this material can efficiently remove reactive oxygen species through catalyzing the disproportionation reaction of O₂˙⁻, generating less toxic H₂O₂ and O₂. Experimental results demonstrate a positive correlation between the SOD-like activity of the material and the Cu²⁺ loading content (Fig. 2c), with a specific activity reaching 448.72 U mg⁻¹ and O₂˙⁻ scavenging efficiency comparable to that of natural SOD5. Moreover, the Cu-SAzyme maintains stable catalytic activity over broad temperature (4–90 °C) and pH (1–13) ranges, whereas natural SOD is prone to inactivation under extreme conditions (Fig. 2d and 2e). The combination of exceptional environmental stability and high catalytic activity endows the Cu-SAzyme with broader application prospects in complex physiological environments compared to natural enzymes.

Fig. 2 SOD-like activity of Cu-SAzyme. (a) Schematic of the post-adsorption strategy for anchoring Cu²⁺ on an N-doped carbon (N–C) support to mimic the microenvironment of natural Cu-only superoxide dismutase (SOD5). (b) Schematic presentation of the Cu-SOD that scavenges O₂˙⁻ by catalyzing its dismutation reaction. (c) Relationship between the SOD-like activity and the Cu²⁺ loading content in the materials. (d and e) Comparison of the relative activities of Cu-SAzyme and natural SOD under different temperature and pH conditions. Adapted with permission from ref. 129. Copyright [2022] Ji Yang, Ruofei Zhang, Hanqing Zhao, Haifeng Qi, Jingyun Li, Jian-Feng Li, Xinyao Zhou, Aiqin Wang, Kelong Fan, Xiyun Yan, Tao Zhang.

Metalloporphyrins, known for their unique structures and excellent photoelectric properties, play roles in key physiological processes in nature and demonstrate broad application potential in the field of biomimetic catalysis.¹³⁰ Due to the structural similarity between iron porphyrins and the active center of natural POD, early studies on metalloporphyrins and their catalytic activities primarily focused on their POD-mimicking activity.^131,132 However, subsequent research has shown that by precisely tuning the type of central metal ion and the coordination microenvironment, metalloporphyrin-based materials can exhibit multiple enzyme-like activities, such as SOD-like activity. Tin (Sn) porphyrins represent a novel design strategy in this regard, demonstrating SOD-like activity but lacking POD-like activity. Using TCPP as the porphyrin ligand, Li et al. synthesized Sn-TCPP, which was then coordinated with Zr⁴⁺ to form Sn-PCN222. This material exhibits remarkable SOD-like activity originating from the Sn(IV)/Sn(II) valence transition at the Sn-TCPP center. The SOD activity of this nanozyme is 2–3 times higher than that of the control material Mn-PCN222 and two orders of magnitude higher than that of cerium oxide (CeO₂) NPs. It also maintains stable catalytic activity across a broad temperature range (4–90 °C) and a wide pH range (3–13).¹³³ These superior properties make Sn porphyrin-based nanozymes highly promising for regulating the oxidative stress microenvironment in periodontitis and promoting bone regeneration, warranting further in-depth research and development.

5.1.2. CAT-like activity. CAT is a crucial antioxidant enzyme in biological systems. Its active site contains an iron porphyrin (iron protoporphyrin IX) as a cofactor, where Fe³⁺ efficiently catalyzes the decomposition of H₂O₂ into H₂O and O₂.¹³⁴ This catalytic process occurs via a two-step reaction. The Fe³⁺ first binds with H₂O₂ to form a high-valent iron intermediate (Fe⁴⁺=O˙⁺). This intermediate then reacts with a second H₂O₂ molecule, ultimately producing two molecules of H₂O and one molecule of O₂ while regenerating the Fe³⁺ state.¹³⁵ This mechanism allows CAT to effectively eliminate H₂O₂ and suppress the production of harmful ˙OH via the Fenton reaction, making it a vital defense against oxidative damage.¹³⁶

Informed by the deep understanding of the natural CAT catalytic mechanism, researchers have focused on developing CAT-like nanozymes with similar functions. Based on the mode of H₂O₂ chemical bond cleavage, the catalytic mechanisms are primarily classified into homolytic and heterolytic pathways.¹³⁷ In homolytic cleavage, the O–O bond of H₂O₂ breaks symmetrically, generating two highly reactive ˙OH radicals. This pathway is characteristic of the classic Fenton reaction, but its non-specific reactivity limits its applicability in biotherapy. In contrast, heterolytic cleavage follows an asymmetric path, directly producing H₂O and O₂, and offers higher catalytic selectivity and biocompatibility.¹³⁸ Consequently, CAT-like nanozymes operating via a heterolytic mechanism have become an important direction for designing nanomaterials that target ROS scavenging, particularly for disease treatments requiring high selectivity.

Two-line ferrihydrite (2L-Fht) is a typical example of a CAT-like nanozyme.¹⁰² Among all iron oxides, 2L-Fht possesses the highest surface hydroxyl density (up to 33.84 nm⁻²). The iron-associated hydroxyl groups (Fe–OH) are central to its catalytic activity, which correlates exponentially with the surface –OH content (Fig. 3a). Density functional theory calculations reveal that H₂O₂ adsorbs asymmetrically on the Fe–OH active sites of 2L-Fht. This is followed by a proton-coupled electron transfer process that induces heterolytic cleavage of the O–O bond, directly generating H₂O and O₂ instead of highly reactive ˙OH radicals (Fig. 3b). This mechanism, elucidated at the atomic level, underpins the structural basis for the highly selective catalysis of 2L-Fht In terms of catalytic performance, 2L-Fht exhibits a Michaelis constant (K_m) of 40.60 mM and a maximum reaction rate (V_max) of 3.04 × 10⁻³ mM s⁻¹ for H₂O₂ decomposition. This indicates that 2L-Fht combines a high affinity for H₂O₂ with efficient catalytic turnover, with performance comparable to other reported CAT mimics. Furthermore, 2L-Fht maintains stable CAT-like activity from physiological to acidic conditions (pH 4.0–8.7), and its optimal catalytic temperature is precisely 37 °C, making it highly suitable for biological applications.

Fig. 3 Specific CAT-like activity and proposed mechanism of Fht. (a) Schematic of the enzyme-mimicking activity profile of Fht, demonstrating its exclusive CAT-like function and the critical role of surface hydroxyl groups. (b) Proposed molecular mechanism for H₂O₂ decomposition into O₂ catalyzed by surface Fe–OH species on Fht. Adapted with permission from ref. 102. Copyright 2021 Ruofei Zhang, Lei Chen, Qian Liang, Juqun Xi, Hanqing Zhao, Yiliang Jin, Xingfa Gao, Xiyun Yan, Lizeng Gao, Kelong Fan.

Beyond the aforementioned strategy of modulating surface-OH structures to enhance the selectivity of CAT-like activity, SAzymes demonstrate unique advantages in mimicking CAT activity and achieving highly selective catalysis, owing to their well-defined atomic-scale active centers and precisely tunable coordination environments. Research shows that a Co SAzyme with an asymmetric Co-N₃PS coordination configuration can significantly reduce the energy barrier for H₂O₂ dissociation through fine-tuning of the electronic structure of its active center, thereby optimizing the adsorption behavior of oxygen-containing reactive intermediates (such as ˙OH, OOH, and O). The phosphorus atom acts as an electron donor to enhance substrate binding, while the sulfur atom acts as an electron acceptor to facilitate the efficient desorption of ˙OH.¹³⁹ This synergistic effect not only stabilizes the tetrahedral transition state but also mimics the multi-site cooperative catalytic mechanism of natural CAT at the atomic scale. This mechanism enables the nanozyme to exhibit outstanding performance at pH 7.4 and 37 °C: its specific activity reaches 7046 U μmol⁻¹; the catalytic constant (K_cat) is 520 s⁻¹; the K_m for H₂O₂ is 6.10 mM; and the catalytic efficiency (K_cat/K_m) reaches 8.52 × 10⁴ M s⁻¹. Importantly, it maintains stable activity over a broad pH range (4.0–8.7) and demonstrates catalytic activity across a temperature range of 4–90 °C. Similarly, a ruthenium-based SAzyme (OxgeMCC-r SAE) exhibits a heterolytic catalytic mechanism analogous to natural CAT.¹⁴⁰ Its active center consists of atomically dispersed ruthenium sites, which selectively cleave the O–H bond rather than the O–O bond of H₂O₂. This process, involving an efficient electron transfer pathway, directly produces H₂O and O₂, thereby avoiding free radical by-reaction pathways. Kinetic studies indicate that the catalytic rate constant (k) for this material is 0.041 min⁻¹, which is significantly higher than that of conventional MnO₂ nanozymes (0.029 min⁻¹) at the same active component concentration, reflecting superior intrinsic catalytic activity. More importantly, OxgeMCC-r SAE demonstrates exceptional operational stability in a simulated acidic tumor microenvironment (pH ≈ 6.5). Its catalytic activity shows no significant decay over multiple reaction cycles with repeated H₂O₂ addition, whereas the MnO₂ nanozyme control suffers a sharp activity decrease after the second cycle. This characteristic enables OxgeMCC-r SAE to provide sustained and efficient oxygen supply in complex physiological environments, offering a key advantage for biomedical applications such as periodontitis treatment.

5.1.3. GPx-like activity. GPx as a crucial intracellular antioxidant enzyme, plays an indispensable role in scavenging ROS, particularly peroxides, by virtue of its unique catalytic mechanism.¹⁴¹ Similar to CAT, GPx efficiently catalyzes the decomposition of H₂O₂ to H₂O, but it possesses a broader substrate scope, capable of reducing not only inorganic peroxides but also organic peroxides such as lipid peroxides (ROOH) to their corresponding alcohols (ROH).^142,143 This broad-spectrum antioxidant property places it at the core of maintaining cellular redox homeostasis. The catalytic activity of GPx depends on the selenocysteine residue at its active site. The selenol group (–SeH) on this residue acts as a nucleophile, attacking the peroxide substrate (e.g., H₂O₂) to form a selenenic acid (–SeOH) intermediate.¹⁴⁴ In the reduction phase, this intermediate utilizes reduced glutathione (GSH) as an essential electron donor, undergoing sequential thiol-disulfide exchange reactions with two molecules of GSH. This process ultimately regenerates the active enzyme and produces oxidized glutathione disulfide (GSSG).¹⁴⁵ The reduction and regeneration of GSSG are vital for sustaining the continuous antioxidant capacity of the GPx system. Glutathione reductase (GR) maintains a high reduced intracellular GSH/GSSG ratio by catalyzing the NADPH-dependent reduction of GSSG to GSH; concurrently, the pentose phosphate pathway is responsible for regenerating NADPH.¹⁴² Thereby, a complete pathway from peroxide clearance to energy metabolism is established. This cascade effectively ensures the GPx system's capacity to handle complex oxidative stress. Furthermore, different GPx isoforms exhibit distinct substrate preferences due to variations in their active site microenvironments. For instance, GPx1 and GPx3 primarily act on soluble peroxides, whereas GPx4, with its hydrophobic active pocket, is the unique isoform capable of directly reducing membrane-bound phospholipid hydroperoxides.¹⁴³ This functional differentiation allows the GPx family to establish a multi-layered antioxidant defense network across different cellular compartments and physiological conditions.

Since selenium (Se) is proven essential for the activity of natural GPx, Se NPs and metal selenides are widely used to mimic GPx activity.^146,147 For example, Qu et al. prepared GO-Se nanocomposites by adding ascorbic acid to a mixture of graphene oxide (GO) and selenium dioxide, resulting in Se NPs uniformly decorated on the GO surface.¹⁴⁸ Using the classic GR-coupled assay, the study demonstrated that the Se NPs possess intrinsic GPx-like activity, which is further enhanced by the large specific surface area and rapid electron transfer capability of the GO support.

Since the initial report in 2014 on the GPx-like activity and cytoprotective effects of V₂O₅ nanowires, vanadium-based nanomaterials have been frequently employed in the design of GPx mimics.^149–151 By mimicking the nucleophilicity of the selenium atom in the natural GPx active center, researchers have applied ligand engineering to the MIL-47(V) metal-organic framework (MOF) system. The active centers of this MOF nanozyme are the vanadium metal nodes, with catalytic activity originating from the coordination between the metal nodes and the ligands, rather than from the ligands themselves. Modifying the ligand substituents (e.g., –NH₂, –F) allows for tuning the electronic properties of the vanadium nodes, thereby influencing their ability to form peroxovanadium intermediates with H₂O₂, which is a key step in the GPx-mimicking reaction (Fig. 4a). Kinetic analysis further confirmed that MIL-47(V)-NH₂ exhibited the highest GPx-mimicking activity among the series of materials under simulated physiological conditions (pH 7.4), showing the most pronounced real-time NADPH consumption rate (Fig. 4b and c). The apparent Michaelis–Menten constant (K_m) and maximum reaction rate (V_max) for GSH were approximately 2.85 mM and 0.0035 mM s⁻¹, respectively, while those for H₂O₂ were about 0.003 mM and 0.0019 mM s⁻¹. Thermogravimetric analysis (TGA) indicated that MIL-47(V)-NH₂ maintained structural stability below 300 °C. These results demonstrate that the ligand engineering strategy can effectively modulate the catalytic performance of MOF nanozymes, enabling the rational design of high-performance GPx-mimicking activity.¹⁵²

Fig. 4 The ability of ligand engineering to regulate the GPx-like activity of the MIL-47(V)-X type MOF material. (a) Schematic illustration of the construction of MIL-47(V)-X (left) and the corresponding catalytic rates of different substituents X (right). (b) Comparison of GPx-mimicking activities of MIL-47(V)-X MOFs. Control means in the absence of MOF. (c) Real-time monitoring of the GPx catalytic activity kinetics of MIL-47(V)-X MOFs. Adapted with permission from ref. 152. Copyright 2020 Wiley-VCH GmbH.

In contrast to the above strategy focused on achieving high activity, another study utilizing the MIL-47(V)-F (MVF) nanozyme system concentrated on mimicking the selenol-selenenic acid transformation process of natural GPx. This system employs a single ligand and is designed to specifically scavenge H₂O₂ without interfering with the physiological functions of other ROS.¹⁵³ Furthermore, Mugesh et al. developed a novel copper vanadate (CuV₂O₆, denoted CuV) nanowire material, which was the first reported to simultaneously mimic the antioxidant activity of GPx and mediate the release of nitric oxide (NO) from S-nitrosothiols (RSNOs).¹⁵⁴ The GPx catalytic activity originates from the redox cycling of vanadium nodes (V⁵⁺/V⁴⁺). Kinetic analysis shows that under standard conditions (pH 7.4, 25 °C), CuV exhibits high affinity for H₂O₂ (K_m ≈ 23.83 μM), with activity comparable to that of natural GPx (K_m ≈ 25 μM). The copper nodes (Cu²⁺/Cu⁺) contribute denitrosylation activity, utilizing endogenous RSNOs as natural NO reservoirs to release NO in situ in the presence of GSH, thereby directly supplementing NO bioavailability. This dual-functional synergy, achieved by clearing harmful ROS like H₂O₂ to protect existing NO while locally replenishing beneficial NO via RSNO decomposition, works cooperatively to counteract vascular endothelial dysfunction, offering a new strategic approach for the treatment of periodontitis-related bone regeneration.

5.2. Multi-enzyme activity nanozymes

Multi-enzyme activity nanozymes refer to nanomaterials possessing more than one type of enzyme-like activity. They have attracted significant attention in recent years due to their considerable potential in biomedical applications.¹⁵⁵ Multienzyme nanozymes belong to the class of antioxidant nanozymes. They typically combine two or even three enzymes such as SOD, CAT, or GPx to simulate the cascade reaction of the natural enzyme system, thereby significantly enhancing the catalytic efficiency. Compared to single-activity nanozymes, the core advantages of multi-enzyme nanozymes lie in their ability to achieve synergistic effects, cascade reactions, and intelligent responses to the microenvironment. However, beyond these advantages, the catalytic mechanisms and rational design of multi-enzyme nanozymes are considerably more complex than those of their single-activity counterparts.¹⁵⁶

Regarding catalytic mechanisms, the high efficiency of multi-enzyme nanozymes stems from their cascade reaction capability and synergistic effects. An ideal multi-enzyme nanozyme can first dismutate O₂˙⁻ into H₂O₂via its SOD-like activity; subsequently, its CAT-like or GPx-like activity rapidly decomposes the generated H₂O₂ into H₂O and O₂. This process prevents the cytotoxic accumulation of H₂O₂ and enhances overall antioxidant efficiency. A composite material developed by Huang's team, V₂O₅@pDA@MnO₂, is based on this concept.¹⁵⁷ It uses polydopamine (pDA) as a linker to assemble V₂O₅ nanowires (with GPx-like activity) with MnO₂ NPs (possessing both SOD-like and CAT-like activities), forming a multi-enzyme synergistic system. At a concentration of 15 μg mL⁻¹, this composite achieved approximately 70% scavenging of superoxide anion, while at 80 μg mL⁻¹, it scavenged about 85% of ˙OH, demonstrating exceptional broad-spectrum ROS scavenging capacity. Studies indicate this system maintains high activity at physiological pH and is relatively insensitive to pH changes. Both in vitro and in vivo experiments showed excellent intracellular ROS scavenging ability, effectively protecting cells from oxidative stress damage.

Similarly, Ai et al. developed a Se@Me@MnO₂ nanozyme exhibiting concurrent CAT-like, GPx-like, and SOD-like activities.¹⁵⁸ A significant enhancement in ROS scavenging efficiency was achieved through synergy between its components. Experiments showed that in a pH 7.4 buffer, this nanozyme could scavenge 90% of superoxide anion at 50 μg mL⁻¹ and up to 97% of ˙OH at 100 μg mL⁻¹, indicating superior antioxidant capacity compared to the V₂O₅@pDA@MnO₂ system. EPR spectroscopy confirmed its efficient radical scavenging mechanism, and recycling tests demonstrated retention of over 95% activity after six uses, indicating good operational stability. In both cellular and animal models, this nanozyme exhibited excellent antioxidant and anti-inflammatory effects, offering new insights for treating ROS-related diseases.

Furthermore, Chen et al. developed manganese-doped Prussian blue nanozymes (MPBZs) based on an inorganic biomimetic concept.¹⁵⁹ By incorporating manganese ions (Mn²⁺/Mn³⁺) with variable valence states into the Prussian blue framework, a stable catalytic interface was constructed. In this design, the manganese ions serve as electron transfer centers participating in the redox cycling of ROS, while the cyanide-bridged rigid framework provides an ordered coordination environment. Their synergy enables efficient CAT-like and SOD-like catalytic functions. This material exhibited notable multi-antioxidant activity: it effectively scavenged O₂˙⁻ at 150 μg mL⁻¹; catalyzed the decomposition of 20 mM H₂O₂ to produce O₂ at 100 μg mL⁻¹; and achieved over 89% scavenging of ABTS radicals at 20 μg mL⁻¹. This design strategy provides a new approach for developing efficient and stable multifunctional nanozymes.

The multi-enzyme mimicking activities exhibited by the aforementioned Mn-based nanomaterials are not unique among transition metal-based nanozymes. Many transition metal ions (e.g., Mn, Ce, Fe) possess variable valence states and flexible electronic structures, enabling them to exhibit various enzyme-like activities, such as SOD, CAT, or GPx, in response to microenvironmental changes.¹⁶⁰ Among these, CeO NPs are a typical representative. Their multi-enzyme activities (e.g., SOD and CAT) originate from the abundant oxygen vacancies in the fluorite crystal structure and the accompanying reversible Ce³⁺/Ce⁴⁺ redox pair. The concentration of oxygen vacancies, which act as active sites for catalytic reactions, directly determines their oxygen storage capacity. The dynamic valence switching between Ce³⁺ and Ce⁴⁺ constitutes the electron transfer foundation for their catalytic ROS scavenging reactions. Specifically, Ce³⁺ sites are responsible for capturing and dismutating O₂˙⁻, exhibiting SOD-like activity with a catalytic rate constant reaching up to 3.6 × 10⁹ M s⁻¹, a value that even exceeds that of the native Cu/Zn-SOD enzyme. In contrast, Ce⁴⁺ sites primarily participate in decomposing H₂O₂, demonstrating CAT-like functionality with a catalytic rate of approximately 2.71 nmol min⁻¹ under typical experimental conditions.^161–163

Previous studies depositing CeO NPs onto pure titanium substrates, simulating clinical implant materials, demonstrated their ability to enhance the osteogenic potential of BMSCs and promote M2 polarization of macrophages. Effective cytoprotection and pro-osteogenic effects were observed even at a low dose of 1 μg mL⁻¹ in both acute and chronic inflammation models, fully showcasing their application potential in bone repair.^164,165 However, the ROS scavenging capacity of CeO NPs can convert to a pro-oxidant activity in highly acidic environments, such as the tumor microenvironment. Excess H⁺ ions can block the redox cycling between Ce³⁺ and Ce⁴⁺, leading to inhibited antioxidant catalytic activity and accumulation of H₂O₂ in the cellular microenvironment, ultimately causing cell damage and dysfunction.^{163,166–169} This seemingly adverse activity conversion could, if precisely controlled, be developed into an intelligent therapeutic strategy for specifically eliminating tumor cells or bacteria. Precisely controlling the catalytic activity “switch” of CeO nanozymes through material design and leveraging the specific characteristics of the periodontitis microenvironment, such as mild acidity and high H₂O₂ levels, to trigger their functional transition holds significant potential.

Beyond intrinsic valence state regulation, precise design of the active center's electronic structure through heterointerfacial engineering has emerged as a key frontier for enhancing performance. For example, cobalt oxide supported with iridium nanoclusters (CoO-Ir) exhibits remarkable SOD-CAT cascade activity due to its unique electronic coupling effect (Fig. 5a).¹⁷⁰ The degradation of O₂˙⁻ by this material is dose-dependent, achieving approximately 50% removal at a concentration as low as 10 μg ml⁻¹ (Fig. 5b). Under the same conditions, the degradation efficiency of H₂O₂ reaches 84.8%, which is significantly higher than that of unmodified CoO (∼5.0%) (Fig. 5c). Meanwhile, CoO-Ir demonstrates a notably enhanced V_max of 76.249 mg L⁻¹ min⁻¹ and a K_m of 453.537 mM during H₂O₂ degradation (Fig. 5d). Moreover, its turnover number (TON) reaches 2.736 s⁻¹, substantially exceeding that of pure CoO (V_max = 3.644 mg L⁻¹ min⁻¹) and other reported artificial enzymes (Fig. 5e), highlighting the prominent advantage of heterointerface engineering in enhancing catalytic activity and reaction efficiency. Similarly, iron-nitrogen-doped carbon-based nanozymes (e.g., featuring Fe–N₄ structures) also exhibit multi-enzyme activity. Their structure resembles that of metalloporphyrin cofactors, enabling efficient mimicry of various enzymes (SOD/CAT). The SOD-like specific activity reaches 1257.13 U mg⁻¹, and the CAT-like specific activity is 41.65 U mg⁻¹, with catalytic potency comparable to that of natural enzymes.¹⁷¹ Furthermore, a recently developed core–shell manganese-based/gallium-indium liquid metal nanozyme (MnO_x@EGaIn) exemplifies the intelligent switching of OXD, CAT, and SOD activities at different pH levels achieved through heterostructure design.¹⁷²

Fig. 5 The SOD-CAT cascade catalysis of the CoO-Ir artificial antioxidant enzyme. (a) Schematic illustration of the cascade SOD-CAT catalytic activity of CoO-Ir. (b) SOD like property of CoO and CoO-Ir for scavenging of O₂˙⁻ and (c) The dynamic H₂O₂ elimination activities at different times. (d) V_max and K_m values of CoO and CoO-Ir with H₂O₂. (e) Comparison of the V_max and TON values for CoO-Ir and other recently reported ROS-elimination nanobiocatalysts. Adapted with permission from ref. 170. Copyright 2023 American Chemical Society.

Addressing the contradiction arising from the coexistence of oxidase and antioxidant enzyme activities, current strategies have evolved from simply inhibiting non-target activities to utilizing microenvironmental signals (such as pH, specific enzymes, or light) for spatiotemporally precise activation and switching of activities. For example, Liu et al. developed manganese oxide nanocluster-decorated graphdiyne nanosheets (MnO_x/GDY).¹⁷³ In the acidic microenvironment of a biofilm infection, the OXD/POD-like activities of MnO_x/GDY are preferentially activated to efficiently kill pathogens and disrupt the biofilm. Subsequently, in the tissue inflammation area after infection control (neutral/alkaline microenvironment), it intelligently switches to an antioxidant mode dominated by SOD and CAT activities. Notably, its CAT-like activity exhibits exceptional performance under alkaline conditions (optimal pH 8.4). Kinetic parameters indicate a K_m of 1.10 mM and a V_max of 6.37 mg L⁻¹ min⁻¹ for the substrate H₂O₂. Under optimal conditions (500 mM H₂O₂, pH 8.4), its specific activity reaches 115.5 U mg⁻¹. This efficient H₂O₂ decomposition capacity, synergizing with the SOD-like activity, forms a cascade reaction that continuously scavenges excess ROS, thereby effectively alleviating oxidative stress and exerting anti-inflammatory and tissue repair-promoting effects. This precise activity regulation based on microenvironmental signals demonstrates the design intelligence of a new generation of smart nanozymes. For a more intuitive comparison of the performance of the various antioxidant nanozymes described above, their key catalytic parameters are summarized in Table 1. In conclusion, a thorough understanding of the catalytic principles and structure-activity relationships of the various categories of nanozymes provides the essential knowledge base for designing intelligent-responsive nanozymes and optimizing their application strategies tailored to the complex microenvironment of periodontitis-related bone defects. The following sections will elaborate on the underlying mechanisms of action, biological functions, and related research progress based on this theoretical foundation.

Table 1 Classification of active centers and functional characteristics of antioxidant nanozymes

Classification	Nanomaterial	Active Center	Intrinsic activity	Catalytic efficiency and kinetics	Stability	Ref.
SOD-like	Cu-SAzyme	Atomically dispersed, coordinatively unsaturated Cu–N4 sites	Specific activity: 448.72 U mg^−1		pH 1–13; Temp. 4–90 °C	129
	Sn-PCN222	A tin-porphyrin structure constructed with TCPP as the organic ligand	100 times higher than CeO2 NPs		pH 1–13; Temp. 4–90 °C	133
CAT-like	2L-Fht	Iron-associated hydroxyl groups		K m (H2O2) = 40.60 mM;	pH 4.0–8.7; Temp. 37 °C.	102
				V max = 3.04 × 10^−3 mM s^−1
	Co-N3PS SAzyme	Co-N3PS coordination environment	Specific activity: 7046 U μmol^−1.	K m (H2O2): 6.10 mM;	pH 4.0–8.7; Temp. 4–90 °C;	139
				K cat: 520 s^−1;
				K cat/Km: 8.52 × 10^4 M s^−1	Optimum: pH 10.8; 37 °C.
GPx-like	MIL-47(V)-NH2	Vanadium metal nodes (V^3+) with their electronic properties tuned by ligand engineering		K m (GSH): 2.85 mM;	pH 7.4; Temp. ≤ 300 °C.	152
				V max (GSH): 0.0035 mM s^−1;
				K m (H2O2): 0.003 mM;
				V max (H2O2): 0.0019 mM s^−1
	CuV2O6	Vanadium metal nodes (V^5+/V^4+)		K m (H2O2): 23.83 μM;	pH 7.4; Temp. 25 °C.	154
				V max: 156.25 μM min^−1
Multiple enzyme-like	V2O5@Pda@MnO2	V2O5 (GPx), MnO2 (SOD/CAT)	Radical scavenging:			157
			70% (O2˙^−, 15 μg mL^−1);
			85% (˙OH, 80 μg mL^−1).
	Se@Me@MnO2	Se (GPx), Me (SOD/CAT), MnO2 (SOD/CAT)	Radical scavenging:		pH 7.4; Temp. 25 °C.	158
			90% (O2˙^−, 50 μg mL^−1);
			97% (˙OH, 100 μg mL^−1)
	MPBZ	SOD/CAT: Mn^2+/Mn^3+ switchable valence states	O2˙^− scavenging:			159
			Effective at 150 μg mL^−1;
			ABTS^+ scavenging:
			>89% at 20 μg mL^−1.
	CeO2 NPs	SOD/CAT: Ce^3+/Ce^4+ reversible redox couple and oxygen vacancies		K cat (SOD): 3.6 × 10^9 M s^−1; Kcat (CAT): 2.71 nmol min^−1		161–163
	CoO-Ir	SOD/CAT:		V max (H2O2): 76.249 mg L^−1 min^−1;	TON: 2.736 s^−1.	170
		Ir nanoclusters
		Loaded on CoO		K m: 453.537 mM.
	Fe–N4 SAzyme	SOD/CAT: Fe–N4 coordination structure and Fe atomic clusters	Specific activity (SOD): 1257.13 U mg^−1;			171
			Specific activity (CAT): 41.65 U mg^−1

---

6. Antioxidant nanozymes for periodontitis-related bone defects regeneration

In research on periodontitis-related bone defects repair, antioxidant nanozymes, with their unique enzyme-mimicking activity and intelligent response characteristics, have shown promise beyond traditional treatment strategies. These nanomaterials not only directly eliminate excessive ROS, effectively alleviating local oxidative stress in periodontal tissues, but also exert synergistic effects in multiple biological processes such as antibacterial, anti-inflammatory, immune regulation, and promotion of angiogenesis and bone regeneration through their intrinsic activity or released active ions (such as Cu²⁺, Ce⁴⁺/Ce³⁺). In recent years, with the advancement of material design, their functions have evolved from simple antioxidant properties to dynamic sensing and precise regulation of the complex pathological microenvironment of periodontitis, opening new avenues for the functional regeneration of periodontal tissues. The application characteristics of various representative antioxidant nanozymes in periodontitis bone regeneration are summarized in Table 2.

Table 2 Application of antioxidant nanozymes in periodontitis bone regeneration

Nanozyme system	Catalytic activity	Primary function(s)	Mechanism of action	Ref.
MIL-47(V)-F	GPx	Anti-inflammatory, osteogenesis, immunomodulatory	Downregulation of TNF-α and IL-1β;	153
			Induce M2 polarization of macrophages;
			Enhanced PDLSCs proliferation and differentiation via PI3K/Akt
ZIF-8:Ce	CAT/SOD	Antibacterial, anti-inflammatory, immunomodulatory	Zn^2+ induces membrane disruption via ROS;	174
			Ce^3+/Ce^4+ redox cycling scavenges ROS;
			Inhibits NF-κB pathway;
			Promotes M1-to-M2 macrophage polarization.
CuCeOx	CAT/SOD	Antibacterial, anti-inflammatory, osteogenesis	NIR-activated PDT/PTT antibacterial therapy;	175
			SOD/CAT-mimetic ROS scavenging;
			Nrf2/HO-1 activation to downregulate TNF-α & IL-1β;
			Promotion of osteogenic differentiation.
TM/BHT/CuTA	CAT/SOD	Microenvironmental response release, antibacterial, anti-inflammatory, immunomodulatory, osteogenesis	Enhances bacterial membrane permeability;	176
			Balances oxidative stress/inflammation via Nrf2/NF-κB modulation;
			Promotes M1-to-M2 shift & osteogenesis;
			TA-Cu^2+ network enables MMP-responsive drug delivery.
Pt@PCN-222	CAT	Antibacterial, anti-inflammatory, microenvironment remodeling	Catalyzes H2O2 decomposition to produce O2, alleviating hypoxia;	177
			Generates singlet oxygen (^1O2) via PDT for antibacterial action.
			Reduces oxidative stress for anti-inflammatory effect.
mPDA@CeO2	CAT/SOD	Anti-inflammatory	The enzymatic activity of CeO2 and the antioxidant effect of mPDA work together to reduce oxidative stress.	178
Cu2O@RuO2	CAT/SOD	Angiogenesis, osteogenesis, anti-inflammatory, immunomodulatory	Cu^2+-mediated HIF-α activation for angiogenesis;	179
			Responsive Cu^2+ release enhancing biomineralization/osteogenesis;
			p-MAPK pathways upregulation promoting cell migration/differentiation;
			ROS scavenging, down-regulating TNF-α and up-regulating VEGF-A.
MnO2@hPM	CAT/SOD	Inflammatory targeted delivery, mitochondrial function repair, immunomodulatory	Targeting inflammation via CXCR4-mediated chemotaxis;	180
			Repairing mitochondrial function via Mn^2+ uniporter enrichment;
			Modulating immunity by inhibiting M1 activation via TLR/NF-κB.
HA-PMs@Ce	CAT/SOD	Antibacterial, anti-inflammatory, mitochondrial function repair, immunomodulatory, osteogenesis, angiogenesis	ROS-triggered release of CeOx/TA (antibacterial);	181
			CeOx scavenges ROS/generates O2, reversing mtROS damage;
			Upregulates osteogenic/angiogenic genes to synergistically promote bone and vessel formation.

---

6.1. Antibacterial and anti-biofilm applications

The pathogenesis of periodontitis is closely associated with the formation of dental plaque biofilms. The periodontal microenvironment (e.g., subgingival hypoxia, pH fluctuations) provides favorable conditions for periodontal pathogens such as P.g. and F.n.^182,183 The structural properties of biofilms significantly enhance their resistance to antibiotics and host immune responses, making the development of antibacterial strategies capable of effectively penetrating and disrupting biofilms crucial.

In the context of bone defects repair in periodontitis, antioxidative nanozymes demonstrate multifaceted biological effects in antibacterial and anti-biofilm applications. These nanozymes not only directly catalyze the decomposition of ROS, disrupting the bacterial redox balance, but also indirectly eliminate the inflammatory microenvironment that bacteria rely on by alleviating host oxidative stress. Cerium-incorporated zeolitic imidazolate framework NPs (ZIF-8: Ce NPs) combine the dual functions of Zn²⁺ release-mediated bactericidal activity and Ce³⁺/Ce⁴⁺-driven ROS scavenging. The Zn²⁺ ions exert broad-spectrum antibacterial effects (primarily against P.g. and F.n.) by interfering with bacterial membrane potential and disrupting protein folding, while the Ce ions, by mimicking CAT and SOD activity, dynamically clear H₂O₂ and O₂˙⁻ in the inflammatory microenvironment, breaking the oxidative stress protection mechanism utilized by bacteria.¹⁷⁴

Building on this, antibacterial strategies have evolved towards synergistic dual functionalities. CuCeO_x bimetallic oxide nanozymes exemplify this trend by simultaneously exerting photodynamic and photothermal effects under near-infrared (NIR) light excitation, directly killing P.g. (inhibition rate of 98.69 ± 0.23%); under light-free conditions, they scavenge excess ROS via Ce³⁺/Ce⁴⁺ valence transition mimicking SOD and CAT activity, and modulate the Nrf2/HO-1 pathway to alleviate oxidative stress, achieving a dynamic balance between direct bactericidal activity and tissue repair (Fig. 6).¹⁷⁵ In addition, the PPAg core–satellite nanocomposite system developed by Wang et al. was constructed via polydopamine (PDA)-mediated integration of Prussian blue (PB) nanozymes with silver satellite NPs.¹⁸⁴ The multi-enzyme mimetic activities (CAT/SOD/POD) of PB cooperate with the sustained antibacterial effect of silver, facilitated by the linker and spacer function of the PDA coating, enabling synergistic and non-interfering operation as well as enhanced stability. PB responsively scavenges excess ROS in a neutral microenvironment, while concurrently achieving auxiliary mild photothermal antibacterial action (55 °C) under 808 nm NIR irradiation. Experiments confirmed that bacterial colonies of P.g. were reduced to zero with this treatment. Moreover, this system demonstrated excellent anti-inflammatory performance in a rat periodontitis model.

Fig. 6 The Schematic illustration of the synthesis process and mechanism of CuCeO_x for periodontitis therapy. Reproduced with permission from ref. 175. Copyright 2025 Elsevier B.V.

With advancements in nanotechnology, more intelligent antibacterial materials responsive to the unique periodontal microenvironment are emerging. A TM/BHT/CuTA hydrogel system developed by Wang et al. achieves highly efficient local antibacterial activity by releasing CuTA nanozymes upon hydrolysis in the MMP-rich environment. It inhibited P.g. by 98.51 ± 0.47% and Actinomyces viscosus by 97.17 ± 3.19% at 800 μg mL⁻¹, effectively disrupted their biofilms (with biofilm clearance increasing with concentration), and prolonged antibacterial duration via electrostatic adsorption. The CuTA nanozymes clear ROS by mimicking the SOD and CAT cascade, indirectly reducing pro-inflammatory factors and suppressing bacterial proliferation.¹⁷⁶ Dong et al. recently reported a platinum nanozyme-functionalized zirconium-porphyrin MOF (Pt@PCN-222), which ingeniously utilizes the endogenous H₂O₂ excessively accumulated at the lesion site as a fuel source to achieve integrated microenvironment remodeling and synergistic bactericidal effects.¹⁷⁷ The core of this design lies in the CAT-like activity of the Pt nanozyme, which efficiently decomposes H₂O₂ to generate O₂, instantly alleviating local hypoxia. Simultaneously, the PCN-222 framework, acting as a built-in photosensitizer, maximizes its photodynamic effect under the resulting O₂-sufficient conditions, generating ample singlet oxygen (¹O₂) to efficiently kill anaerobic bacteria like P.g. This work signifies a new direction in nanozyme antibacterial strategies, shifting from direct killing to a combination of microenvironment modulation and precise intervention. By leveraging endogenous ROS for self-supplied oxygen, it not only significantly enhances the efficacy of traditional photodynamic therapy but also offers a universal nanotechnology solution for reversing the pathological progression of hypoxia-related diseases.

More advanced designs further integrate multiple functions including targeting, imaging, and therapy. A microenvironment-responsive NIR-IIb multifunctional nanozyme system (DMUP), developed by Lin's team, is assembled from lanthanide-doped down-conversion NPs (DCNPs) and multi-enzyme active (CAT/SOD) manganese nanozymes, with its surface modified with the bacterial targeting peptide UBI29-41. In the inflammatory microenvironment, DMUP responds to low pH conditions to achieve bacteria-specific fluorescence imaging, in situ oxygen generation to ameliorate hypoxia, and ROS scavenging, thereby achieving highly efficient antibacterial effects. Studies showed that DMUP exhibited an inhibition rate against P.g. as high as 99.9%, significantly outperforming the conventional drug minocycline. Its antibacterial mechanisms also include disrupting bacterial biofilm structure and suppressing the expression of bacterial virulence genes encoding fimbrilin A (fimA) and gingipains (kgp, rgpA, rgpB), consequently weakening bacterial adhesion and invasion capabilities.¹⁸⁵

Compared to the indirect antibacterial effects of antioxidant nanozymes, nanozymes possessing oxidase-like activity that can directly kill bacteria generally demonstrate more significant antimicrobial efficacy. For instance, Zhu et al. developed a silk fibroin/copper sulfide nanozyme hybrid microneedle (MN-SF/CuS), which synergistically disrupts periodontal biofilms formed by P.g. and F.n. through the combined mechanical penetration of the microneedles and photothermal effects. Simultaneously, the copper ions (Cu⁺/Cu²⁺) in CuS undergo a Fenton-like reaction, catalyzing H₂O₂ to generate ˙OH and enabling controlled deep release of ROS.¹⁸⁶ Similarly, the near-infrared-driven hydrogel (AgZ@Au/PLEL) developed by Qian et al. utilizes the glucose oxidase-like activity of Au NPs to decompose high concentrations of glucose within the periodontal pocket, producing acid and H₂O₂. NIR irradiation significantly enhances this catalytic activity, thereby increasing H₂O₂ production. The generated H₂O₂ acts synergistically with the photothermal therapy effect to achieve efficient inhibition of P.g. and F.n. Meanwhile, the acidic environment promotes the degradation of the ZIF-8 framework, releasing zinc ions to further assist the antibacterial process and stimulate alveolar bone regeneration.¹⁸⁷ Therefore, multifunctional nanozymes integrating both antioxidant and oxidase-like activities have emerged as a novel design strategy.^188,189

Certain advanced materials are now capable of intelligently switching between antioxidant enzyme and oxidase-like activities. This dynamic switching characteristic significantly enhances the functional flexibility of nanozymes, offering new possibilities for the precise regulation of antibacterial therapy and inflammatory microenvironments. For example, the GNRs@CeO₂@PDS nanozyme with a Janus semi-coated structure design can achieve a pH-responsive balance between ROS scavenging and bactericidal activity by modulating the exposure of enzymatic active sites on CeO₂.¹⁹⁰ Under neutral conditions, this material exhibits significant CAT/SOD-like activities with a remarkable ROS scavenging capacity (O₂˙⁻ scavenging rate reaches 900 U mg⁻¹), while its POD-like activity is markedly enhanced in acidic microenvironments. Simultaneously, 808 nm NIR irradiation excites the plasmonic thermal effect of the gold nanorods (GNRs). The localized hot electrons injected into CeO₂ not only promote the efficient conversion of Ce⁴⁺ to Ce³⁺, enhancing the POD-like activity, but also specifically activate persulfate (PDS) to generate highly oxidizing sulfate radicals (SO₄˙⁻) and ˙OH. This photothermal-catalytic synergy enables the nanomaterial to penetrate deeply and efficiently eradicate biofilms formed by P.g. and F.n.

Furthermore, a single-atom nanozyme (Fe–B/N–C SAzyme) has been developed through precise spatial modulation of the planar Fe–N₄ active center via axial boron (B) ligand coordination.¹⁹¹ This system achieves efficient and dynamic switching between ROS-scavenging and ROS-generating modes through the synergy of local coordination environment modulation and photothermal effects. It concurrently exhibits OXD-, POD-, and CAT-like activities, promoting the heterolytic decomposition of H₂O₂ and the desorption of O₂, thereby accelerating the generation of ROS (e.g., ˙OH). The POD-like activity is reflected in the efficient catalytic production of highly toxic ˙OH in the presence of H₂O₂, directly disrupting the cell membranes and biomolecules of P.g. and F.n. In contrast, the CAT-like activity decomposes H₂O₂ to alleviate the hypoxic microenvironment at the lesion site while preventing damage to healthy tissues from excess H₂O₂. Additionally, under NIR irradiation, the nanozyme generates localized heat, which not only directly kills bacteria but also significantly enhances the above three enzyme-mimetic activities, demonstrating considerable potential in antibacterial applications and periodontitis treatment. However, a key challenge remains in precisely regulating the antioxidant and oxidase-like activities of such nanozymes and clearly defining the dominant activity boundaries under different microenvironments. Recent studies have revealed that the formation of neutrophil extracellular traps (NETs) is a key mechanism driving periodontal tissue destruction, yet their degradation has been underemphasized in treatment.¹⁹² Wu et al. designed a core-shell structured microneedle patch, Pd/Ce MNs, which integrates two functionally complementary nanozymes to achieve synergistic antibacterial and NETs-degrading functions. The palladium nanocube (Pd) core exhibits POD-like activity, generating ˙OH in the presence of H₂O₂ to rapidly kill P.g. (inhibition rate close to 100%). The cerium-modified dendritic mesoporous silica nanozyme (DMSN-Ce) shell possesses DNAzyme (DNase)-like activity, enabling sustained degradation of the DNA backbone of NETs and alleviating inflammation. This design combines dual strategies of “antibacterial” and “immune regulation (degrading NETs)” into one system, directly addressing the complex pathology of periodontitis.¹⁹³

In summary, research on the repair of periodontitis-related bone defects has demonstrated a significant evolution in the application of antioxidant nanozymes for antibacterial and anti-biofilm purposes, shifting from a sole antibacterial function towards microenvironment-responsive intelligent systems. The current research frontier focuses on material innovations that enable dynamic switching between antioxidant and pro-oxidant activities of nanozymes, allowing for precise intervention in the complex pathological microenvironment of the periodontal pocket. For instance, designs such as single-atom nanozymes regulated by axial ligand engineering (e.g., Fe–B/N–C SAzymes) utilize external stimuli (e.g., NIR) or local chemical conditions (e.g., pH, MMP concentration) to trigger intelligent switching of catalytic behavior. These systems not only efficiently eliminate pathogens but also begin to actively modulate host cell behavior, such as regulating macrophage polarization. This integration of antibacterial strategies with cellular fate regulation signifies a deepening of the therapeutic concept: successful tissue regeneration relies not only on effective infection control but, more critically, on actively remodeling an immune microenvironment conducive to repair. Consequently, the research focus on nanozymes is naturally extending from direct pathogen killing to the precise modulation of host immune responses, providing a solid foundation for exploring their mechanisms and prospects in anti-inflammatory and immunomodulatory applications.

6.2. Anti-inflammatory and Immunomodulatory effects

In the field of periodontitis treatment, the anti-inflammatory and immunomodulatory effects of antioxidant nanozymes have shown a significant evolution, shifting from direct antioxidant activity towards systemic remodeling of the immune microenvironment. Early research primarily focused on utilizing the intrinsic catalytic activity of nanozymes to neutralize excessive ROS. For instance, CeO₂ NPs scavenge ROS via their SOD/CAT-like activities and inhibit the activation of NF-κB and MAPK signaling pathways, thereby downregulating the expression of pro-inflammatory cytokines such as IL-1β and TNF-α, demonstrating for the first time that modulating redox balance can indirectly alleviate inflammation.¹⁹⁴ However, this single-mechanism strategy has limitations in its ability to modulate the complex immune network.

To further enhance immunomodulatory efficacy, research has shifted towards multifunctional synergistic designs, a progression clearly reflected in the evolution of material strategies. Studies have shown that an increased Ce⁴⁺/Ce³⁺ ratio also enhances the polarization of RAW264.7 murine macrophages towards the M2 phenotype, particularly increasing the percentage of healing-associated M2 cells and the secretion of anti-inflammatory cytokines, suggesting that manipulating the valence state of CeO₂ NPs provides effective regulation of macrophage M2 polarization.¹⁹⁵ Zhang et al. developed erythrocyte-like mesoporous polydopamine-coated ceria nanobowls (mPDA@CeO₂ NBs), which combine the non-enzymatic antioxidant properties of mesoporous polydopamine (mPDA) with the SOD-like and CAT-like activities of CeO₂ NPs for synergistic ROS elimination. The erythrocyte-like structure of these mesoporous nanobowls facilitates easy cellular uptake, and their mesopores can be loaded with both hydrophobic and hydrophilic drugs for combined anti-inflammatory therapy. In vitro and in vivo experiments demonstrated that the combination of CeO₂ NPs and mPDA synergistically achieves multi-modal complementary ROS scavenging and inhibits ROS-induced inflammation.¹⁷⁸ Wang et al. developed a GelMA photocrosslinked hydrogel loaded with CeO₂ NPs and BMSCs. Experiments confirmed that CeO₂ NPs significantly inhibit the polarization of pro-inflammatory M1 macrophages (reduced expression of CD86 and iNOS), lower the expression of pro-inflammatory factors (TNF-α, IL-6), thereby improving the local inflammatory microenvironment, while simultaneously enhancing the mRNA expression of the M2 macrophage marker CD206 and the anti-inflammatory factor IL-10, creating favorable conditions for bone regeneration.¹⁹⁶ Furthermore, the Miao team designed a multifunctional copper-ruthenium oxide-based yolk–shell nanozyme (Cu₂O@RuO₂, CRNC). In this nanozyme, RuO₂ possesses SOD-like and CAT-like activities, enabling efficient clearance of excessive ROS, such as H₂O₂ and O₂˙⁻, in the periodontitis microenvironment, thus restoring the oxidative-antioxidative balance. CRNC can reduce the expression of M1 markers like IL-1β, TNF-α, and IL-6, while increasing the expression of M2 markers such as IL-10, TGF-β, and Arg-1, indicating its significant capacity for immunophenotype regulation.¹⁷⁹ Recently, a yolk–shell structured nanozyme with an Au core and a CeO₂ shell (Au@CeO₂-DMF) further expanded the dimensions of synergy through a more sophisticated design: its Au core generates a photothermal effect under NIR irradiation to control drug release, the CeO₂ shell provides multi-enzyme catalytic activity, and the loaded dimethyl fumarate (DMF) activates the NRF2 pathway to enhance endogenous antioxidant capacity. This photothermal-triggered release mechanism not only achieves spatiotemporal synchronization of ROS scavenging and immune regulation but also demonstrates protective effects on mitochondrial function and reduced bone loss in a periodontitis model by inhibiting NLRP3 inflammasome activation and modulating macrophage polarization.¹⁹⁷ This progression demonstrates that the research focus has transitioned from mere antioxidation to active immune regulation, gradually incorporating external stimulus-responsive strategies, such as pH, ROS, and NIR.

The current research frontier focuses on the precise sensing and active remodeling of the immune microenvironment. From this cross-disciplinary perspective, research on the plasticity and reprogramming of tumor-associated macrophages (TAMs) provides an important reference for immunomodulation in periodontitis. In the tumor microenvironment (TME), the dynamic changes of TAMs involve not only ROS scavenging but also multicellular interactions, such as phagocytic escape mediated by the CD47-SIRPα axis, and metabolic reprogramming, such as M2 polarization driven by lactate accumulation. Their regulation has shifted from targeting single molecules to systemic remodeling.¹⁹⁸

Similarly, the immunomodulation of nanozymes in periodontitis also needs to break through the limitation of simply scavenging ROS, drawing on the multi-dimensional approach of TAM reprogramming. For example, targeting specific genes like regulating the CCL2/CCR2 axis to inhibit macrophage recruitment, or blocking immune checkpoints such as mimicking the mechanism of CD47 antibodies, can synergize with catalytic activity to achieve a more thorough M1-to-M2 phenotypic polarization.^199,200 Recently developed biomimetic nanozymes, such as MnO₂@hPM, have already reflected this approach. This design not only utilizes the SOD-like and CAT-like catalytic activities of MnO₂ but also ingeniously employs hypoxia-pretreated PDLSCs membranes (typically 1%–5% O₂) to enrich them with various functional receptors including CXCR4, IL-1R1, TNFR1, and TLR2/4.¹⁸⁰ This enables the NPs to not only actively target inflammatory sites via the CXCR4/SDF-1 axis but also act as a “multivalent decoy”, broadly neutralizing various inflammatory mediators such as IL-1β, TNF-α, and LPS, thereby blocking the transmission of classic inflammatory signals like NF-κB at the source. Its mechanism of action transcends mere antioxidation, achieving multifunctional synergy involving immunomodulation, metabolic regulation, and tissue regeneration. Specifically, while scavenging ROS and protecting mitochondrial function, it effectively reverses the inflammatory suppression of the osteogenic potential of PDLSCs and promotes macrophage polarization towards the reparative M2 phenotype, thereby systemically reprogramming the destructive inflammatory microenvironment into an immunometabolic environment conducive to tissue regeneration (Fig. 7).

Fig. 7 Schematic illustration of a functionalized biomimetic nanozyme (MnO₂@hPM) for inflammation targeting and microenvironment reprogramming against periodontitis. (a) A sketch showing the preparation process of MnO₂@hPM using hypoxia-educated PDLSCs and sonication methods. (b) After caudal vein administration, MnO₂@hPM targets the inflammatory periodontal sites and neutralizes local proinflammatory factors. (c) When placed in a periodontitis defect, MnO₂@hPM accumulates in PDLSCs under inflammatory environment, scavenges intracellular ROS, mitigates cellular mitochondrial dysfunction, and restores the osteogenic potential of PDLSCs. Reproduced with permission from ref. 180. Copyright 2025 Elsevier B.V.

In summary, the immunomodulatory mechanisms of nanozymes have evolved from initial antioxidant protection, through multifunctional synergistic immunophenotype regulation, to the current stage of active targeting and systemic reprogramming of the immune microenvironment. Future research is expected to extend further into emerging frontiers such as immunometabolism focusing on mitochondria and cell death modes like pyroptosis, exploring how nanozymes can more precisely guide immune responses by regulating the metabolic states of immune cells and cell death pathways, thereby providing more powerful strategies for treating chronic inflammatory diseases like periodontitis.

6.3. Promoting osteogenesis and angiogenesis

During the repair of periodontitis-related bone defects, the tight coupling of angiogenesis and osteogenesis is the core biological basis for achieving functional regeneration of periodontal tissues. The aforementioned strategies involving antibacterial, anti-inflammatory, and immunomodulatory effects are not isolated but collectively serve to create a microenvironment conducive to bone regeneration. Within this context, promoting angiogenesis is inseparable from the osteogenic process, as newly formed vascular networks not only deliver essential O₂ and nutrients to the bone repair site but also facilitate the transport of various growth factors and signaling molecules, playing a crucial role in supporting the recruitment, differentiation, and matrix mineralization of osteoblasts.²⁰¹ In recent years, research on antioxidant nanozymes in the field of promoting osteogenesis and angiogenesis has shown significant evolution from single-function applications towards systemic regulation, with continuously deepening design concepts and mechanistic exploration.

Early research primarily focused on utilizing the intrinsic antioxidant properties of nanozymes to indirectly promote osteogenesis. For instance, CeO₂ NPs, through their variable Ce⁴⁺/Ce³⁺ valence state switching, mimic SOD and CAT activities, effectively scavenging excess ROS and mitigating oxidative stress damage to stem cells. CeO₂ NPs can not only promote the adhesion, spreading, and proliferation of human MSCs but also upregulate the expression of osteogenesis-related genes, accelerating new bone formation.^202,203 Furthermore, a higher proportion of Ce⁴⁺ sites appears more beneficial for cell spreading and osseointegration,^164,195 indicating that finely tuning the electronic structure of nanozymes can optimize their biological functions. Building on this, Zhang et al. incorporated CeO₂ NPs into engineered dental pulp stem cells (E-DPSCs) via endocytosis. They found that under normal conditions, the modified E-DPSCs exhibited enhanced abilities in proliferation, migration, and osteogenic differentiation. Under oxidative stress, CeO₂ NPs also alleviated cell apoptosis by upregulating the expression of the anti-apoptotic factor Bcl-2 and potentially further enhanced cellular resistance by modulating antioxidant signaling pathways such as Nrf2/HO-1, preliminarily demonstrating the potential of nanozymes in protecting stem cell function.²⁰⁴ Furthermore, the MVF nanozyme developed by Zhu et al. represents a step forward in the pursuit of “precision antioxidant” therapy and “bidirectional regulation of bone metabolism”.¹⁵³ By specifically mimicking GPx-like activity, MVF enables selective scavenging of H₂O₂, thereby avoiding potential interference with physiological ROS signaling. Its action primarily relies on activating the Nrf2/HO-1 antioxidant pathway while synergistically regulating osteogenesis-related pathways such as PI3K/Akt. These mechanisms collectively contribute to immunomodulation (e.g., promoting macrophage polarization toward the M2 phenotype) and directly enhance the osteogenic differentiation capacity of periodontal stem cells. Additionally, MVF exhibits a concentration-dependent bidirectional effect on osteoclastogenesis-promoting osteoclast formation at low concentrations (2.5 μg mL⁻¹) while inhibiting osteoclast differentiation at higher concentrations (≥5 μg mL⁻¹). This unique property offers a novel strategy for the dynamic regulation of bone metabolic homeostasis. With deepening research, mere antioxidation has proven insufficient to address the complex pathological microenvironment of periodontitis. Consequently, current studies are increasingly focusing on the construction of intelligent responsive nanoplatforms capable of precise microenvironment sensing and on-demand therapeutic intervention. The copper-ruthenium oxide-based yolk-shell nanozyme (CRNC) developed by the Miao team is a typical representative of this direction.¹⁷⁹ This design utilizes the Cu₂O core as a responsive copper ion reservoir, releasing Cu²⁺ ions within the acidic microenvironment of periodontitis, thereby directly activating the TGF-β/phosphatidylinositol 3-kinase (PI3K) and HIF-1α signaling pathways, thereby promoting angiogenesis in human umbilical vein endothelial cells (HUVECs) and osteogenic differentiation in PDLSCs, respectively. In an animal model of periodontitis, CRNC successfully alleviated inflammation and promoted alveolar bone regeneration, demonstrating the effectiveness of a strategy combining antioxidation, immunomodulation, pro-angiogenesis, and osteogenesis.

Hu et al. developed a Cu/Ce-TA@HA-SF biomimetic hierarchical microsphere.²⁰⁵ Its outer layer consists of a copper/cerium–tannic acid network, which undergoes responsive degradation in the high-ROS microenvironment of periodontitis lesions, efficiently releasing Cu²⁺/Ce³⁺ ions. These ions mimic SOD/CAT enzyme activities to directly decompose H₂O₂/O₂˙⁻, thereby reducing BMSC apoptosis and reversing ROS-induced mitochondrial membrane potential collapse. Simultaneously, Cu²⁺ exerts dual functions by stimulating pro-angiogenic factors and enhancing the osteogenic activity of mesenchymal stem cells via pathways such as MAPK and Wnt signaling. Similarly, Zhu et al. designed a MnO₂@UiO-66(Ce) nanocomposite that integrates the carrier advantages of MOFs with the catalytic activities of MnO₂/Ce through an inorganic–organic hybrid design.²⁰⁶ Triggered by high ROS levels, the valence transitions of Ce/Mn dynamically regulate ROS scavenging and generation, thereby upregulating the expression of antioxidant enzyme genes (e.g., SOD1, CAT, GPX) in PDLSCs. Moreover, via mitochondrial targeting, it specifically scavenges mtROS and activates the SIRT1-FOXO3-BNIP3 pathway, ultimately contributing to long-term homeostasis restoration in periodontitis.

Furthermore, Jiang et al. constructed a multifunctional micelle system (HA-PMs@Ce) based on PLLA-b-PLys/PBA self-assembled micelles and tannic acid (TA), with CeO_x nanozymes encapsulated in the core and a hyaluronic acid coating on the surface.¹⁸¹ The innovation of this system lies in the use of catechol groups in TA to form ROS-sensitive borate ester bonds with phenylboronic acid (PBA), enabling on-demand release of CeO_x and TA under the high oxidative stress conditions of periodontitis. The antibacterial effect primarily relies on the released TA component, while CeO_x clears excess ROS via SOD-mimetic activity, enhances tube formation and migration of HUVECs, and upregulates angiogenesis-related genes (e.g., ANG-1, CD31, VEGF). Concurrently, the system promotes macrophage polarization toward the M2 phenotype, restores mitochondrial function, and increases ATP levels, leading to comprehensive amelioration of oxidative stress. This multi-functional design integrating ROS response and release, antioxidant, antibacterial, anti-inflammatory, immunomodulatory functions, as well as the promotion of angiogenesis and bone repair, represents the development of nanozyme therapy towards comprehensiveness and intelligence (Fig. 8).

Fig. 8 Schematic illustration of the ceria oxide nanozyme-based ROS-responsive micelles system (HA-PMs@Ce) promoting periodontal regeneration in periodontitis. Reproduced with permission from ref. 181. Copyright 2025 Wiley-VCH GmbH.

Beyond classical features such as excessive ROS and decreased pH, the overexpression of matrix metalloproteinases (e.g., MMP-9) and emerging pathological clues like ferroptosis in the periodontitis microenvironment have become critical targets guiding the design of intelligent antioxidant nanozymes. Wang et al. proposed a copper-tannic acid nanozyme hydrogel system (TM/BHT/CuTA).¹⁷⁶ The material is retained at positively charged inflammatory sites of periodontitis via electrostatic adsorption, and the CuTA nanozymes are released on demand through MMP-9-responsive cleavage of ester bonds. The CuTA nanozymes mimic the SOD and CAT cascade process, efficiently scavenging ROS while exerting antibacterial effects by increasing bacterial membrane permeability. Meanwhile, the system significantly reduces the pro-inflammatory factor TNF-α, elevates the anti-inflammatory factor TGF-β, promotes macrophage polarization toward the M2 phenotype, activates the Nrf2 pathway, and suppresses the NF-κB pathway, thereby modulating the immune microenvironment. In terms of osteogenesis, TM/BHT/CuTA enhances ALP activity and mineralized nodule formation in mouse pre-osteoblast MC3T3-E1 cells, upregulating the expression of COL1A1 and OCN, which confirms its pro-osteogenic potential.

Tang et al. recently developed a thermosensitive ionic liquid hydrogel system based on MoS₂ nanoflowers (ACIS@MLG).²⁰⁷ By incorporating galangin and L-cysteine, the system exhibits multi-enzyme activity and iron ion chelation capability. It not only effectively scavenges ROS but also regulates the AMPK/Nrf2/solute carrier family 7 member 11 (SLC7A11) signaling pathway, suppresses the expression of the ferroptosis-key protein acyl-CoA synthetase long-chain family member 4 (ACSL4), and upregulates GPX4 and ferritin heavy chain 1 (FTH1) levels, thereby alleviating lipid peroxidation damage. In animal models, this system significantly attenuated periodontal inflammation and promoted alveolar bone regeneration.

In summary, the research trajectory of antioxidant nanozymes in promoting bone regeneration in periodontitis has progressively evolved from an initial focus on basic antioxidant functions to current integrated designs emphasizing systematization and multifunctional synergy. The core trend of this evolution is reflected in the continuous material innovation, enabling nanozymes to achieve intelligent perception, precise regulation, and on-demand treatment of the complex pathological microenvironment. Looking ahead, further exploration in this field depends on elucidating the direct interaction mechanisms between nanozymes and key osteogenic signaling pathways, such as Wnt/β-catenin and BMP/Smad, and obtaining more clinically predictive empirical data combined with large animal models to accelerate clinical translation. Among them, the development of efficient active targeting strategies is of utmost importance. These approaches include leveraging ligand–receptor interactions, extracellular matrix (ECM)-binding peptides, biofilm-targeting motifs, or cell membrane camouflaging technologies to impart innate targeting capabilities to nanozymes,^208–211 thereby significantly enhancing their selective accumulation and therapeutic efficacy within the periodontal pocket, while minimizing off-target effects on healthy tissues. Meanwhile, attention to the synergistic effects of nanozymes with traditional growth factors, along with their emerging roles in regulating cellular energy metabolism and death modes, such as pyroptosis and ferroptosis, holds promise for deepening the understanding of the molecular network underlying periodontal tissue regeneration. This will lay a solid theoretical foundation for developing next-generation nanozyme therapeutic strategies that are more efficient and safer.

7. Conclusions and perspectives

This article systematically reviews the application mechanisms and research progress of antioxidant nanozymes in intervening in the repair of bone defects in periodontitis. Substantial evidence indicates that by mimicking the catalytic activities of natural antioxidant enzymes such as SOD, CAT, and GPx, nanozymes can efficiently eliminate excess ROS within the periodontal pocket, thereby disrupting the pathological vicious cycle formed by oxidative stress, local hypoxia, chronic inflammation, and microbial dysbiosis. Moreover, smart nanozymes responsive to pathological signals such as pH, ROS, and certain enzymes enable on-demand, localized regulation of activity, thereby significantly improving the precision of intervention in the complex periodontal microenvironment. In summary, the therapeutic strategy centered on intelligent antioxidant activity extends far beyond merely scavenging ROS. It establishes a stable redox balance that supports tissue regeneration and concurrently delivers multiple biological effects. These include direct antibacterial action, degradation of biofilms, modulation of immune polarization toward a pro-repair phenotype, protection of mitochondrial function, and activation of osteogenic signaling pathways. Consequently, it demonstrates potential superior to conventional approaches for the repair of periodontitis-related bone defects.

However, before antioxidant nanozymes can advance to clinical application, several key challenges must be addressed. Compared to natural enzymes, most nanozymes still exhibit inferior catalytic activity and substrate specificity, which may limit their effectiveness within complex biological environments. A central challenge lies in achieving “intelligent” regulation of ROS. Since physiological levels of ROS are essential for maintaining cellular signaling and normal metabolic functions, the role of antioxidant nanozymes should not be the complete elimination of ROS, but rather the precise restoration and maintenance of ROS concentrations within a homeostatic range. Excessive scavenging may disrupt normal physiological processes, while insufficient regulation fails to effectively mitigate oxidative stress. This demands that nanozymes possess the capability for dynamic sensing of and feedback to the local ROS microenvironment. Furthermore, although the multi-enzyme antioxidant activity of many nanozymes is a significant advantage, precisely controlling their spatiotemporal catalytic behavior in the complex in vivo environment to prevent off-target effects and potential unintended catalytic risks remains an unresolved problem.

Finally, antioxidant nanozymes face challenges common to the broader field of biomaterials translation. A systematic assessment of their biocompatibility and long-term in vivo safety is still required. There remains a lack of comprehensive research into their metabolic pathways within the complex oral environment, their potential for organ accumulation, and their long-term impact on the oral microbiome. While intelligently responsive designs are conceptually promising, their controllability, reliability, and reproducibility within the dynamic and heterogeneous physiological environment of the human body require rigorous validation. Simultaneously, achieving the large-scale, standardized, and Good Manufacturing Practice-compliant production of such complex nanosystems presents a practical engineering challenge that must be solved for industrialization. Moreover, as most current evidence is derived from small animal models, their efficacy in large animal models that are anatomically and physiologically closer to humans, as well as their true therapeutic effectiveness and long-term safety within the complex microenvironment of the human periodontium, need to be systematically confirmed through rigorous preclinical and clinical studies. Overcoming these translational barriers is critical for advancing antioxidant nanozymes from a promising laboratory technology to a novel clinical strategy for treating periodontitis.

Looking forward, several important directions are crucial for advancing this field. First, it is essential to deeply explore the synergistic effects of nanozymes with existing clinical treatment modalities. For example, combining them with periodontal foundational therapy and guided tissue regeneration techniques holds promise for developing more effective comprehensive treatment strategies. Of particular interest is the integration of nanozymes into smart biomaterials or scaffolds to synergistically regulate cell behavior and tissue regeneration, which represents an emerging strategy. For instance, the microfluidic bone-on-a-chip platform developed by Yang et al., which guides osteoblast behavior by mimicking the topological structure of the bone microenvironment,²¹² provides important insights for the design of such intelligent integrations. Second, the introduction of artificial intelligence and machine learning technologies will significantly aid the rational design of nanozymes. By constructing databases that correlate material “genes” with catalytic performance based on structure–activity relationships, the development of high-performance, customized nanozymes can be accelerated. At the mechanistic research level, future work should further elucidate the molecular pathways through which nanozymes regulate cellular energy metabolism and influence novel programmed cell death modes such as pyroptosis and ferroptosis, thereby enabling more precise intervention in immune responses and tissue regeneration processes. Ultimately, through interdisciplinary collaboration, establishing standardized safety and efficacy evaluation systems and systematically advancing preclinical and clinical studies are key to translating nanozymes from fundamental research findings into novel clinical strategies for the treatment of periodontitis.

Conflicts of interest

There are no conflicts to declare.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (82560194 to Kun Yang), High-Level Innovative Talents of Guizhou Province (gzwjrs2023-044 to Kun Yang), Zunyi City Science and Technology Program(Zunshikehe HZ (2023) 79 to Kun Yang), Zunyi Medical University "12345" Future Clinical Eminent Physician Program(20211019 to Kun Yang).

References

1. F. Cavalla and M. Hernández, in Periodontitis, ed. J. Santi-Rocca, Springer International Publishing, Cham, 2022, vol. 1373, pp. 195–208.
2. A. T. D. O. Marre, R. M. C. P. Domingues and L. A. Lobo, Braz. J. Microbiol., 2020, 51, 1483–1491.
3. S. Zhang, X. Zhang, J. Huang, N. Zhou, F. Zhang, J. Ren, W. Zhang, T. Wang, W. Xu, X. Luan, X. Huang, Z. Huang, J. Wu, J. Da, L. Liu, B. Zhang, Y. Li and H. Jin, Mater. Today Bio, 2025, 33, 102073.
4. X. Wang, C. Wang, M. Ren, Y. Shi, T. Dong, Z. Liu, J. Jiang and H. Wei, Stem Cell Res. Ther., 2025, 16, 470.
5. X. Zhang, F. Peng and D. Wang, J. Funct. Biomater., 2022, 13, 215.
6. B. Jiang, L. Fang, K. Wu, X. Yan and K. Fan, Theranostics, 2020, 10, 687–706.
7. R. Zhang, B. Jiang, K. Fan, L. Gao and X. Yan, Nat. Rev. Bioeng., 2024, 2, 849–868.
8. F. Cao, L. Zhang, Y. You, L. Zheng, J. Ren and X. Qu, Angew. Chem., Int. Ed., 2020, 59, 5108–5115.
9. Q. Zeng, H. Zhong, J. Liao, Q. Huo, B. Miao, L. Zeng, B. Zhang and G. Nie, Biomater. Sci., 2024, 12, 5150–5163.
10. Z. Ding, Q. Song, G. Wang, Z. Zhong, G. Zhong, H. Li, Y. Chen, X. Zhou, L. Liu and S. Yang, RSC Adv., 2024, 14, 17571–17582.
11. N. Singh, G. R. Sherin and G. Mugesh, Angew. Chem., Int. Ed., 2023, 62, e202301232.
12. Y.-J. Wei, J. Li, Z.-E. Hu, X. Xing, Z.-W. Zhou, Y. Yu, X.-Q. Yu, J. Zhang, Y.-H. Liu and N. Wang, J. Mater. Chem. B, 2023, 11, 6581–6594.
13. S. Dong, Y. Dong, T. Jia, S. Liu, J. Liu, D. Yang, F. He, S. Gai, P. Yang and J. Lin, Adv. Mater., 2020, 32, e2002439.
14. J. Wu, X. Wang, Q. Wang, Z. Lou, S. Li, Y. Zhu, L. Qin and H. Wei, Chem. Soc. Rev., 2019, 48, 1004–1076.
15. H. Fan, R. Zhang, K. Fan, L. Gao and X. Yan, ACS Nano, 2024, 18, 2533–2540.
16. I. Zare, D. Choi, J. Zhang, M. T. Yaraki, A. Ghaee, S. Z. Nasab, R. Taheri-Ledari, A. Maleki, A. Rahi, K. Fan and J. Lee, Nano Today, 2024, 56, 102276.
17. P. Jiang, L. Zhang, X. Liu, C. Ye, P. Zhu, T. Tan, D. Wang and Y. Wang, Nat. Commun., 2024, 15, 1010.
18. W. Gao, J. He, L. Chen, X. Meng, Y. Ma, L. Cheng, K. Tu, X. Gao, C. Liu, M. Zhang, K. Fan, D.-W. Pang and X. Yan, Nat. Commun., 2023, 14, 160.
19. A. V. Kozlov, S. Javadov and N. Sommer, Antioxidants, 2024, 13, 602.
20. L. Zhang, K. Wang, Y. Lei, Q. Li, E. C. Nice and C. Huang, Free Radicals Biol. Med., 2015, 89, 452–465.
21. M. J. Morgan and Z. Liu, Cell Res., 2011, 21, 103–115.
22. J. Luo, L. Li, W. Shi, K. Xu, Y. Shen and B. Dai, Front. Immunol., 2025, 16, 1611932.
23. I. Bernatova, Biomed Res. Int., 2014, 2014, 598271.
24. S. Huang, Y. Zhang, H. Shu, W. Liu, X. Zhou and X. Zhou, CNS Neurosci. Ther., 2024, 30, e14807.
25. W. Xu, X. Chen, Y. Wang, B. Fan, K. Guo, C. Yang, S. Yu, Y. Pang and S. Zhang, J. Cell. Physiol., 2020, 235, 3022–3032.
26. W. Sheng, B. Li, T. Sun, C. Zhu, Y. Li and W. Xu, Exp. Ther. Med., 2024, 27, 232.
27. X. Li, Y. Jiang, X. Liu, J. Fu, J. Du, Z. Luo, J. Xu, U. K. Bhawal, Y. Liu and L. Guo, Int. J. Oral Sci., 2023, 15, 51.
28. B. Zhang, C. Pan, C. Feng, C. Yan, Y. Yu, Z. Chen, C. Guo and X. Wang, Redox Rep., 2022, 27, 45–52.
29. S. Matsunaga, A. Kohda, S. Kamakura, J. Hayase, K. Miyano, A. Shiose and H. Sumimoto, Genes Cells, 2024, 29, 63–72.
30. W. Zhao, Y. Sun and B. Zhu, Sci. Rep., 2025, 15, 3901.
31. K. M. DeMars, A. O. McCrea, D. M. Siwarski, B. D. Sanz, C. Yang and E. Candelario-Jalil, Front. Neurosci., 2018, 12, 89.
32. A. Basic and G. Dahlén, Front. Oral Health, 2023, 4, 1210200.
33. T. Xia, Y. Zhu, K. Li, K. Hao, Y. Chai, H. Jiang, C. Lou, J. Yu, W. Yang, J. Wang, J. Deng and Z. Wang, J. Nanobiotechnol., 2024, 22, 103.
34. X. Jiao and G. Tian, Mol. Med. Rep., 2025, 32, 292.
35. R. Cheng, W. Liu, R. Zhang, Y. Feng, N. A. Bhowmick and T. Hu, Front. Cell. Infect. Microbiol., 2017, 7, 474.
36. G. Baima, M. Arce, M. Romandini and T. Van Dyke, J. Periodontal Res., 2025, 0, 1–34,  DOI:10.1111/jre.70040.
37. N. D'Aiuto, J. Hochmann, M. Millán, A. Di Paolo, R. Bologna-Molina, J. Sotelo Silveira and M. Arocena, Sci. Rep., 2022, 12, 21699.
38. K. Yang, S. Xu, H. Zhao, L. Liu, X. Lv, F. Hu, L. Wang and Q. Ji, Int. Immunopharmacol., 2021, 94, 107456.
39. F. Zurel, H. Bux, A. Diveli and Z. Rashid, Cureus, 2025, 17, e87436.
40. H.-J. Jang and J.-K. Yoon, Biomimetics, 2024, 9, 75–94.
41. M. Liu, H. Ding, H. Wang, M. Wang, X. Wu, L. Gan, L. Cheng and X. Li, J. Orthop. Surg., 2021, 16, 150.
42. C. Wang, M. Zhu, D. Yang, X. Hu, X. Wen and A. Liu, Int. J. Stem Cells, 2022, 15, 324–333.
43. J. Yang, J. Luo, X. Tian, Y. Zhao, Y. Li and X. Wu, Antioxidants, 2024, 13, 394–408.
44. J. Fang, M. Silva, R. Lin, W. Zhou, Y. Chen and W. Zheng, Oxid. Med. Cell. Longevity, 2021, 2021, 5574932.
45. B. Li, W. Li, Y. Liao, Z. Weng, Y. Chen, T. Ouchi, Y. Fan, Z. Zhao and L. Li, J. Adv. Res., 2025, 76, 387–403.
46. E. M. Michalak, M. L. Burr, A. J. Bannister and M. A. Dawson, Nat. Rev. Mol. Cell Biol., 2019, 20, 573–589.
47. E. L. Greer and Y. Shi, Nat. Rev. Genet., 2012, 13, 343–357.
48. A. E. Kane and D. A. Sinclair, Crit. Rev. Biochem. Mol. Biol., 2019, 54, 61–83.
49. R. Ren, A. Ocampo, G.-H. Liu and J. C. Izpisua Belmonte, Cell Metab., 2017, 26, 460–474.
50. X. Su, H. Zhang, F. Lei, R. Wang, T. Lin and L. Liao, J. Cell. Mol. Med., 2022, 26, 375–384.
51. Y. Chen, H. Wang, Q. Ni, T. Wang, C. Bao, Y. Geng, Y. Lu, Y. Cao, Y. Li, L. Li, Y. Xu and W. Sun, J. Dent. Res., 2023, 102, 767–776.
52. S. Thorbert-Mros, L. Larsson and T. Berglundh, J. Periodontal Res., 2015, 50, 535–543.
53. R. Toni, G. Di Conza, F. Barbaro, N. Zini, E. Consolini, D. Dallatana, M. Antoniel, E. Quarantini, M. Quarantini, S. Maioli, C. A. Bruni, L. Elviri, S. Panseri, S. Sprio, M. Sandri and A. Tampieri, Front. Immunol., 2020, 11, 1737.
54. N. Dutzan, J. E. Konkel, T. Greenwell-Wild and N. M. Moutsopoulos, Mucosal Immunol., 2016, 9, 1163–1172.
55. S. Pérez and S. Rius-Pérez, Antioxidants, 2022, 11, 1394.
56. J. Yuan, X. Liang, W. Zhou, J. Feng, Z. Wang, S. Shen, X. Guan, L. Zhao and F. Deng, Ann. Transl. Med., 2021, 9, 1578–1578.
57. M. Tanaka, H. Inoue, N. Takahashi and M. Uehara, Free Radicals Biol. Med., 2023, 205, 107–115.
58. A. Drakou, A. Kaspiris, E. Vasiliadis, D.-S. Evangelopoulos, D. Koulalis, A. Lenti, S. Chatziioannou and S. G. Pneumatikos, Ann. Case Rep., 2025, 10, 102174.
59. G. Cerqueni, A. Scalzone, C. Licini, P. Gentile and M. Mattioli-Belmonte, Mater. Sci. Eng., C, 2021, 130, 112433.
60. Z.-J. Ji, Y. Shi, X. Li, R. Hou, Y. Yang, Z.-Q. Liu, X.-C. Duan, Q. Liu, W.-D. Chen and D.-Y. Peng, Front. Pharmacol., 2022, 13, 910217.
61. Q. He, Z. Li, C. Meng, J. Wu, Y. Zhao and J. Zhao, Cells, 2019, 8, 897.
62. F. R. G. Rocha, A. E. Delitto, J. A. C. De Souza, L. A. González-Maldonado, S. M. Wallet and C. Rossa Junior, Sci. Rep., 2020, 10, 7823.
63. D. Averill-Bates, Biochim. Biophys. Acta, Mol. Cell Res., 2024, 1871, 119573.
64. G. M. Tannahill, A. M. Curtis, J. Adamik, E. M. Palsson-McDermott, A. F. McGettrick, G. Goel, C. Frezza, N. J. Bernard, B. Kelly, N. H. Foley, L. Zheng, A. Gardet, Z. Tong, S. S. Jany, S. C. Corr, M. Haneklaus, B. E. Caffrey, K. Pierce, S. Walmsley, F. C. Beasley, E. Cummins, V. Nizet, M. Whyte, C. T. Taylor, H. Lin, S. L. Masters, E. Gottlieb, V. P. Kelly, C. Clish, P. E. Auron, R. J. Xavier and L. A. J. O'Neill, Nature, 2013, 496, 238–242.
65. S. E. Corcoran and L. A. J. O'Neill, J. Clin. Invest., 2016, 126, 3699–3707.
66. B. D'Autréaux and M. B. Toledano, Nat. Rev. Mol. Cell Biol., 2007, 8, 813–824.
67. M. Mittal, M. R. Siddiqui, K. Tran, S. P. Reddy and A. B. Malik, Antioxid. Redox Signal., 2014, 20, 1126–1167.
68. K. Zhang, A. Mishra and C. Jagannath, Front. Immunol., 2024, 15, 1363938.
69. K. Zhang, M. L. Sowers, E. I. Cherryhomes, V. K. Singh, A. Mishra, B. I. Restrepo, A. Khan and C. Jagannath, Front. Immunol., 2023, 14, 1121495.
70. F. Sun, I. Abreu-Rodriguez, S. Ye, S. Gay, O. Distler, M. Neidhart and E. Karouzakis, PLoS One, 2019, 14, e0218551.
71. A. R. DiNardo, K. Rajapakshe, T. Nishiguchi, S. L. Grimm, G. Mtetwa, Q. Dlamini, J. Kahari, S. Mahapatra, A. Kay, G. Maphalala, E. M. Mace, G. Makedonas, J. D. Cirillo, M. G. Netea, R. Van Crevel, C. Coarfa and A. M. Mandalakas, J. Clin. Invest., 2020, 130, 3113–3123.
72. B. T. Weinert, C. Schölz, S. A. Wagner, V. Iesmantavicius, D. Su, J. A. Daniel and C. Choudhary, Cell Rep., 2013, 4, 842–851.
73. K. Zhang and C. Jagannath, Epigenet. Chromatin, 2025, 18, 16–36.
74. R. M. Touyz and A. M. Briones, Hypertens. Res., 2011, 34, 5–14.
75. I. Takac, K. Schröder and R. P. Brandes, Curr. Hypertens. Rep., 2012, 14, 70–78.
76. M. Ushio-Fukai and R. W. Alexander, Mol. Cell. Biochem., 2004, 264, 85–97.
77. S. C. Bir, G. K. Kolluru, K. Fang and C. G. Kevil, Semin. Cell Dev. Biol., 2012, 23, 745–757.
78. S. Chandra, M. Romero, A. Shatanawi, A. Alkilany, R. Caldwell and R. Caldwell, Br. J. Pharmacol., 2012, 165, 506–519.
79. C. G. Kevil, N. Okayama and J. S. Alexander, Am. J. Physiol.: Cell Physiol., 2001, 281, C1940–C1947.
80. S. Vepa, W. M. Scribner and V. Natarajan, Free Radicals Biol. Med., 1997, 22, 25–35.
81. A. Csiszar, K. E. Smith, A. Koller, G. Kaley, J. G. Edwards and Z. Ungvari, Circulation, 2005, 111, 2364–2372.
82. T. Shono, M. Ono, H. Izumi, S.-I. Jimi, K. Matsushima, T. Okamoto, K. Kohno and M. Kuwano, Mol. Cell. Biol., 1996, 16, 4231–4239.
83. Y.-W. Kim and T. V. Byzova, Blood, 2014, 123, 625–631.
84. U. Florczyk, A. Jazwa, M. Maleszewska, M. Mendel, K. Szade, M. Kozakowska, A. Grochot-Przeczek, M. Viscardi, S. Czauderna, K. Bukowska-Strakova, J. Kotlinowski, A. Jozkowicz, A. Loboda and J. Dulak, Antioxid. Redox Signal., 2014, 20, 1693–1708.
85. Z. Zhai, Y. Huang, Y. Zhang, L. Zhao and W. Li, Antioxidants, 2022, 11, 1564.
86. D. D. Zhang, Nat. Rev. Drug Discovery, 2025, 24, 421–444.
87. J. Peng, F. Li, M. Qiu, X. Xu, G. Liu and J. Ou, Discover Oncol., 2025, 16, 168–180.
88. C. Lee, M.-J. Kim, A. Kumar, H.-W. Lee, Y. Yang and Y. Kim, Signal Transduction Targeted Ther., 2025, 10, 170–210.
89. Y. Yang, X. Sun, Y. Yang and Y. Qie, Regen. Ther., 2024, 26, 1159–1167.
90. C. M. Warren, S. Ziyad, A. Briot, A. Der and M. L. Iruela-Arispe, Sci. Signaling, 2014, 7, ra1.
91. I. Edirisinghe, S.-R. Yang, H. Yao, S. Rajendrasozhan, S. Caito, D. Adenuga, C. Wong, A. Rahman, R. P. Phipps, Z.-G. Jin and I. Rahman, FASEB J., 2008, 22, 2297–2310.
92. H. J. Zhou, Z. Xu, Z. Wang, H. Zhang, Z. W. Zhuang, M. Simons and W. Min, Nat. Commun., 2018, 9, 3303.
93. G. M. Nardi, E. Ferrara, I. Converti, F. Cesarano, S. Scacco, R. Grassi, A. Gnoni, F. R. Grassi and B. Rapone, Int. J. Environ. Res. Public Health, 2020, 17, 2765.
94. R. Hoshino, N. Nakamura, T. Yamauchi, Y. Aoki, M. Miyabe, S. Sasajima, R. Ozaki, T. Sekiya, T. Sato, M. Tabuchi, K. Miyazawa and K. Naruse, J. Diabetes Invest., 2025, 16, 72–82.
95. A. Shaito, K. Aramouni, R. Assaf, A. Parenti, A. Orekhov, A. E. Yazbi, G. Pintus and A. H. Eid, Front. Biosci.-Landmark, 2022, 27, 105–119.
96. L. U. Pagan, M. J. Gomes, P. F. Martinez and M. P. Okoshi, Oxid. Med. Cell. Longevity, 2022, 2022, 9829505.
97. S. Ali, S. Mumtaz, H. A. Shakir, M. Khan, H. M. Tahir, S. Mumtaz, T. A. Mughal, A. Hassan, S. A. R. Kazmi, Sadia, M. Irfan and M. A. Khan, Mol. Genet. Genomic Med., 2021, 9, e1788.
98. P. Caprari, E. Profumo, S. Massimi, B. Buttari, R. Riganò, V. Regine, M. Gabbianelli, S. Rossi, R. Risoluti, S. Materazzi, G. Gullifa, L. Maffei and F. Sorrentino, Front. Mol. Biosci., 2023, 9, 1108896.
99. A. R. Cyr, L. V. Huckaby, S. S. Shiva and B. S. Zuckerbraun, Crit. Care Clin., 2020, 36, 307–321.
100. J. E. Meegan, J. A. Bastarache and L. B. Ware, Am. J. Physiol.: Lung Cell. Mol. Physiol., 2021, 321, L429–L439.
101. H. Sun, J. Xu, Y. Wang, S. Shen, X. Xu, L. Zhang and Q. Jiang, Bioact. Mater., 2023, 24, 477–496.
102. R. Zhang, L. Chen, Q. Liang, J. Xi, H. Zhao, Y. Jin, X. Gao, X. Yan, L. Gao and K. Fan, Nano Today, 2021, 41, 101317.
103. P. Malla, Y.-M. Wang and C.-H. Su, J. Nanobiotechnol., 2025, 23, 130–162.
104. F. Yan, D. Liu, B. Zhao, Y. Wang, Y. Wang, S. Yang and S. Li, Biomater. Adv., 2025, 169, 214193.
105. F. Manea, F. B. Houillon, L. Pasquato and P. Scrimin, Angew. Chem., Int. Ed., 2004, 43, 6165–6169.
106. L. Gao, J. Zhuang, L. Nie, J. Zhang, Y. Zhang, N. Gu, T. Wang, J. Feng, D. Yang, S. Perrett and X. Yan, Nat. Nanotechnol., 2007, 2, 577–583.
107. Y. Ai, Z. Hu, X. Liang, H. Sun, H. Xin and Q. Liang, Adv. Funct. Mater., 2022, 32, 2110432.
108. L. Gao, Y. Liu, D. Kim, Y. Li, G. Hwang, P. C. Naha, D. P. Cormode and H. Koo, Biomaterials, 2016, 101, 272–284.
109. S. H. Im, C. Y. Kim, Y. Jung, Y. Jang and S. H. Kim, Biomater. Sci., 2017, 5, 422–431.
110. P. C. Naha, Y. Liu, G. Hwang, Y. Huang, S. Gubara, V. Jonnakuti, A. Simon-Soro, D. Kim, L. Gao, H. Koo and D. P. Cormode, ACS Nano, 2019, 13, 4960–4971.
111. Y. Liu, P. C. Naha, G. Hwang, D. Kim, Y. Huang, A. Simon-Soro, H.-I. Jung, Z. Ren, Y. Li, S. Gubara, F. Alawi, D. Zero, A. T. Hara, D. P. Cormode and H. Koo, Nat. Commun., 2018, 9, 2920.
112. Z. Li, K. Mao, M. Yue, Y. Liu and K. Fan, Small, 2025, e07226.
113. N. T. M. Thao, H. D. K. Do, N. N. Nam, N. K. S. Tran, T. T. Dan and K. T. L. Trinh, Micromachines, 2023, 14, 1017.
114. S. Singh, Int. J. Biol. Macromol., 2024, 260, 129374.
115. R. Zhang, B. Jiang, K. Fan, L. Gao and X. Yan, Nat. Rev. Bioeng., 2024, 2, 849–868.
116. H. Li, Z. Zhang, J. Liu and H. Wang, Biomed. Eng. OnLine, 2025, 24, 41.
117. J. Wang, Y. Zhang, Q. Tang, Y. Zhang, Y. Yin and L. Chen, Antioxidants, 2024, 13, 789.
118. A. Lin, S. Liu and H. Wei, Particuology, 2023, 76, 32–45.
119. S. Zhou, H. Cai, X. He, Z. Tang and S. Lu, Coord. Chem. Rev., 2024, 500, 215536.
120. S. Wang, J. Liu, L. Zhou, H. Xu, D. Zhang, X. Zhang, Q. Wang and Q. Zhou, Regener. Biomater., 2024, 11, rbae082.
121. S. D. Copley, Curr. Opin. Struct. Biol., 2017, 47, 167–175.
122. O. Khersonsky and D. S. Tawfik, Annu. Rev. Biochem., 2010, 79, 471–505.
123. G. Tossetta, S. Fantone, F. Olivieri, R. Mazzucchelli, L. Togni, A. Santarelli, D. Marzioni and M. R. Rippo, Mol. Biol. Rep., 2025, 52, 771–786.
124. T. Pecchillo Cimmino, R. Ammendola, F. Cattaneo and G. Esposito, Int. J. Mol. Sci., 2023, 24, 2086.
125. H. Zhao, R. Zhang, X. Yan and K. Fan, J. Mater. Chem. B, 2021, 9, 6939–6957.
126. M. Mureseanu, I. Bleotu, C.-I. Spînu and N. Cioatera, Int. J. Mol. Sci., 2024, 25, 1094.
127. M. Liu, X. Sun, B. Chen, R. Dai, Z. Xi and H. Xu, Int. J. Mol. Sci., 2022, 23, 15893.
128. J. Fujii, T. Homma and T. Osaki, Antioxidants, 2022, 11, 501.
129. J. Yang, R. Zhang, H. Zhao, H. Qi, J. Li, J.-F. Li, X. Zhou, A. Wang, K. Fan, X. Yan and T. Zhang, Exploration, 2022, 2, 20210267.
130. M. Patel and B. J. Day, Trends Pharmacol. Sci., 1999, 20, 359–364.
131. R. Geng, R. Chang, Q. Zou, G. Shen, T. Jiao and X. Yan, Small, 2021, 17, 2008114.
132. W. Wu, Q. Wang, J. Chen, L. Huang, H. Zhang, K. Rong and S. Dong, Nanoscale, 2019, 11, 12603–12609.
133. L. Li, H. Li, L. Shi, L. Shi and T. Li, Langmuir, 2022, 38, 7272–7279.
134. S. Anwar, F. Alrumaihi, T. Sarwar, A. Y. Babiker, A. A. Khan, S. V. Prabhu and A. H. Rahmani, Biomolecules, 2024, 14, 697–728.
135. M. Alfonso-Prieto, X. Biarnés, P. Vidossich and C. Rovira, J. Am. Chem. Soc., 2009, 131, 11751–11761.
136. M. Galasso, S. Gambino, M. G. Romanelli, M. Donadelli and M. T. Scupoli, Free Radicals Biol. Med., 2021, 172, 264–272.
137. D. Xu, L. Wu, H. Yao and L. Zhao, Small, 2022, 18, 2203400.
138. L. Yang, S. Dong, S. Gai, D. Yang, H. Ding, L. Feng, G. Yang, Z. Rehman and P. Yang, Nano-Micro Lett., 2024, 16, 28–81.
139. Y. Chen, B. Jiang, H. Hao, H. Li, C. Qiu, X. Liang, Q. Qu, Z. Zhang, R. Gao, D. Duan, S. Ji, D. Wang and M. Liang, Angew. Chem., Int. Ed., 2023, 62, e202301879.
140. D. Wang, H. Wu, S. Z. F. Phua, G. Yang, W. Lim, L. Gu, C. Qian, H. Wang, Z. Guo, H. Chen and Y. Zhao, Nat. Commun., 2020, 11, 357.
141. A. I. Cota-Magaña, M. Vazquez-Moreno, A. Rocha-Aguado, S. Ángeles-Mejía, A. Valladares-Salgado, M. Díaz-Flores, N. E. López-Díazguerrero and M. Cruz, Antioxidants, 2024, 13, 457–470.
142. J. Pei, X. Pan, G. Wei and Y. Hua, Front. Pharmacol., 2023, 14, 1147414.
143. E. Lubos, J. Loscalzo and D. E. Handy, Antioxid. Redox Signal., 2011, 15, 1957–1997.
144. L. Flohe, W. A. Günzler and H. H. Schock, FEBS Lett., 1973, 32, 132–134.
145. B. Huchzermeyer, E. Menghani, P. Khardia and A. Shilu, Antioxidants, 2022, 11, 761.
146. M. R. Cerón, M. Izquierdo, N. Alegret, J. A. Valdez, A. Rodríguez-Fortea, M. M. Olmstead, A. L. Balch, J. M. Poblet and L. Echegoyen, Chem. Commun., 2016, 52, 64–67.
147. Y. Huang, E. Su, J. Ren and X. Qu, Nano Today, 2021, 38, 101205.
148. Y. Huang, C. Liu, F. Pu, Z. Liu, J. Ren and X. Qu, A GO–Se nanocomposite as an antioxidant nanozyme for cytoprotection, Chem. Commun., 2017, 53, 3082–3085.
149. S. Zhang, Y. Li, S. Sun, L. Liu, X. Mu, S. Liu, M. Jiao, X. Chen, K. Chen, H. Ma, T. Li, X. Liu, H. Wang, J. Zhang, J. Yang and X.-D. Zhang, Nat. Commun., 2022, 13, 4744.
150. Y. V. Nikitchenko, V. K. Klochkov, N. S. Kavok, N. A. Karpenko, S. L. Yefimova, V. P. Semynozhenko, I. V. Nikitchenko and A. I. Bozhkov, Biogerontology, 2025, 26, 14.
151. Y. V. Nikitchenko, V. K. Klochkov, N. S. Kavok, K. A. Averchenko, N. A. Karpenko, I. V. Nikitchenko, S. L. Yefimova and A. I. Bozhkov, Biol. Trace Elem. Res., 2021, 199, 4183–4192.
152. J. Wu, Y. Yu, Y. Cheng, C. Cheng, Y. Zhang, B. Jiang, X. Zhao, L. Miao and H. Wei, Angew. Chem., Int. Ed., 2021, 60, 1227–1234.
153. B. Zhu, J. Wu, T. Li, S. Liu, J. Guo, Y. Yu, X. Qiu, Y. Zhao, H. Peng, J. Zhang, L. Miao and H. Wei, Adv. Healthcare Mater., 2024, 13, 2302485.
154. S. Ghosh, P. Roy, S. Prasad and G. Mugesh, Faraday Discuss., 2022, 234, 284–303.
155. D. Li, D. Dai, G. Xiong, S. Lan and C. Zhang, Small, 2023, 19, 2205870.
156. J. Sheng, Y. Wu, H. Ding, K. Feng, Y. Shen, Y. Zhang and N. Gu, Adv. Mater., 2024, 36, 2211210.
157. Y. Huang, Z. Liu, C. Liu, E. Ju, Y. Zhang, J. Ren and X. Qu, Angew. Chem., Int. Ed., 2016, 55, 6646–6650.
158. Y. Ai, J. You, J. Gao, J. Wang, H. Sun, M. Ding and Q. Liang, Nano Res., 2021, 14, 2644–2653.
159. C. Chen, H. Wu, Q. Li, M. Liu, F. Yin, M. Wu, X. Wei, H. Wang, Z. Zha and F. Wang, Biomater. Sci., 2017, 5, 422–431.
160. Q. Liu, A. Zhang, R. Wang, Q. Zhang and D. Cui, Nano-Micro Lett., 2021, 13, 154.
161. T. Pirmohamed, J. M. Dowding, S. Singh, B. Wasserman, E. Heckert, A. S. Karakoti, J. E. S. King, S. Seal and W. T. Self, Chem. Commun., 2010, 46, 2736.
162. C. Korsvik, S. Patil, S. Seal and W. T. Self, Chem. Commun., 2007, 10, 1056–1058.
163. C. Xu and X. Qu, NPG Asia Mater., 2014, 6, e90.
164. F. Wei, C. J. Neal, T. S. Sakthivel, T. Kean, S. Seal and M. J. Coathup, Mater. Sci. Eng., C, 2021, 124, 112041.
165. J. Li, J. Wen, B. Li, W. Li, W. Qiao, J. Shen, W. Jin, X. Jiang, K. W. K. Yeung and P. K. Chu, Adv. Sci., 2018, 5, 1700678.
166. E. Alpaslan, H. Yazici, N. H. Golshan, K. S. Ziemer and T. J. Webster, ACS Biomater. Sci. Eng., 2015, 1, 1096–1103.
167. A. Asati, C. Kaittanis, S. Santra and J. M. Perez, Anal. Chem., 2011, 83, 2547–2553.
168. A. Asati, S. Santra, C. Kaittanis, S. Nath and J. M. Perez, Angew. Chem., Int. Ed., 2009, 48, 2308–2312.
169. M. Pešić, A. Podolski-Renić, S. Stojković, B. Matović, D. Zmejkoski, V. Kojić, G. Bogdanović, A. Pavićević, M. Mojović, A. Savić, I. Milenković, A. Kalauzi and K. Radotić, Chem.-Biol. Interact., 2015, 232, 85–93.
170. Y. Xie, S. Xiao, L. Huang, J. Guo, M. Bai, Y. Gao, H. Zhou, L. Qiu, C. Cheng and X. Han, ACS Nano, 2023, 17, 15097–15112.
171. J. Xi, R. Zhang, L. Wang, W. Xu, Q. Liang, J. Li, J. Jiang, Y. Yang, X. Yan, K. Fan and L. Gao, Adv. Funct. Mater., 2021, 31, 2007130.
172. Y. Shen, R. Xing, X. Xu, Y. Wang, R. Huang, R. Su, M. D. Dickey, W. Qi and J. Kong, Matter, 2025, 8, 102159.
173. S. Liu, Q. Bai, Y. Jiang, Y. Gao, Z. Chen, L. Shang, S. Zhang, L. Yu, D. Yang, N. Sui and Z. Zhu, Small, 2024, 20, 2308403.
174. X. Li, M. Qi, C. Li, B. Dong, J. Wang, M. D. Weir, S. Imazato, L. Du, C. D. Lynch, L. Xu, Y. Zhou, L. Wang and H. H. K. Xu, J. Mater. Chem. B, 2019, 7, 6955–6971.
175. X. Li, H. Liang, Y. Huang, Q. Hu, L. Liang, J. He, Q. Lu, G. Zhou, L. Chen, C. Li and X. Li, Colloids Surf., B, 2025, 252, 114685.
176. Y. Xu, Y. Luo, Z. Weng, H. Xu, W. Zhang, Q. Li, H. Liu, L. Liu, Y. Wang, X. Liu, L. Liao and X. Wang, ACS Nano, 2023, 17, 18732–18746.
177. F. Dong, Y. Liu, H. Dong, X. Sun, B. Xu, Q. Wu, X. Xu, Y. Jiang, H. Liu and Y. Wang, Nano Today, 2024, 56, 102280.
178. L. Zhang, Y. Gao, Z. Wang, Y. Qi, L. Li, T. Wang, D. Li and C. Wang, ACS Appl. Bio Mater., 2024, 7, 2851–2861.
179. Y. Li, F. Muhammad, X. Chen, D. Gu, W. Li, J. Tang, M. Cheng, J. Du, S. Qiao, Y. Deng, Q. Yu, H. Wei and L. Miao, Nano Today, 2025, 61, 102624.
180. L. Li, R.-X. Wu, J. Wang, R. Jin, X.-Y. Zhang, M. Xu, S.-Y. Li, R. Gao, B.-M. Tian, X.-T. He, F.-M. Chen and Y. Yin, Chem. Eng. J., 2025, 506, 160119.
181. Y. Xu, N. Huang, L. Pan, J. Du, L. Cao and X. Jiang, Adv. Healthcare Mater., 2025, e02690.
182. D. Manoil and G. N. Belibasakis, in Oral Biofilms in Health and Disease, ed. H. Koo, N. S. Jakubovics and B. P. Krom, Springer Nature Switzerland, Cham, 2025, vol. 14, pp. 311–323.
183. G. Hajishengallis, Nat. Rev. Immunol., 2015, 15, 30–44.
184. P. Wang, L. Wang, Y. Liu, Y. Mu, Y. Deng, A. Jia, Y. Zhan, J. Guo, Z. Li and F. Tong, ACS Appl. Mater. Interfaces, 2025, 17, 47969–47984.
185. S. Lin, T. Cui, Y. Jiang, J. Xie, D. Zhong, J. Jiang, D. Deng, M. Zhao, C. Xue, S. Gan, J. Qiu and X. Wang, J. Nanobiotechnol., 2025, 23, 189.
186. H. Chen, N. Yu, J. Wang, S. Zhang, L. Cao, M. Zhou, Z. Xu, S. Lin, S. Yin, X. Jiang and M. Zhu, Nano Today, 2024, 56, 102297.
187. C. Qu, T. Luo, R. Han, D. Mo, K. Shi, X. Li, Z. Bei, M. Pan, S. Tan, J. Liu and Z. Qian, Chem. Eng. J., 2025, 506, 160108.
188. W. Feng, X. Han, H. Hu, M. Chang, L. Ding, H. Xiang, Y. Chen and Y. Li, Nat. Commun., 2021, 12, 2203.
189. Y. Lin, Z. Zhu, Z. Xu, J. Chen, Z. Li, H. Huang, Y. Zhang, D. Chen, B. Liang and G. Wang, Adv. Healthcare Mater., 2025, 14, 2502180.
190. S. Li, Q. Ding, L. Zhang, F. Shi, C. Liu, T. Li, Y. Shi, M. Qi, L. Wang, B. Dong, S. Song, J. Sun, J. S. Kim and C. Li, J. Controlled Release, 2024, 370, 600–613.
191. W. Liu, E. Shi, H. Wu, Y. Liang, M. Chen, H. Zhang, R. Zhang, X. Li, Y. Wang and L. Zhang, Adv. Funct. Mater., 2024, 34, 2403386.
192. B. Bassani, M. Cucchiara, A. Butera, O. Kayali, A. Chiesa, M. T. Palano, F. Olmeo, M. Gallazzi, C. P. B. Dellavia, L. Mortara, L. Parisi and A. Bruno, Int. J. Mol. Sci., 2023, 24, 15370.
193. Y. Shan, J. Zhong, Q. Sun, W. Gao, C. Zhang, H. Chen, J. Zhou, Z. Ye, Q. Chen, Z. Mao and M. Wu, Bioact. Mater., 2025, 53, 161–177.
194. Y. Yu, S. Zhao, D. Gu, B. Zhu, H. Liu, W. Wu, J. Wu, H. Wei and L. Miao, Nanoscale, 2022, 14, 2628–2637.
195. J. Li, J. Wen, B. Li, W. Li, W. Qiao, J. Shen, W. Jin, X. Jiang, K. W. K. Yeung and P. K. Chu, Adv. Sci., 2018, 5, 1700678.
196. R. Wang, X. He, J. Bai, S. Su, R. Zhou, S. Gao, H. Liu and F. Zhou, ACS Appl. Mater. Interfaces, 2024, 16, 67373–67384.
197. T. Li, M. Shu, C. Zhu, Q. Liu, Y. Li, R. Wang, L. Chen, W. Shi, Z. Sun, Z. Hou, B. Fang and L. Xia, Adv. Sci., 2025, 12, 2413891.
198. S. Chen, S. Li and H. Wang, Oncol. Transl. Med., 2024, 10, 281–285.
199. N. Nagarsheth, M. S. Wicha and W. Zou, Nat. Rev. Immunol., 2017, 17, 559–572.
200. S. B. Willingham, J.-P. Volkmer, A. J. Gentles, D. Sahoo, P. Dalerba, S. S. Mitra, J. Wang, H. Contreras-Trujillo, R. Martin, J. D. Cohen, P. Lovelace, F. A. Scheeren, M. P. Chao, K. Weiskopf, C. Tang, A. K. Volkmer, T. J. Naik, T. A. Storm, A. R. Mosley, B. Edris, S. M. Schmid, C. K. Sun, M.-S. Chua, O. Murillo, P. Rajendran, A. C. Cha, R. K. Chin, D. Kim, M. Adorno, T. Raveh, D. Tseng, S. Jaiswal, P. Ø. Enger, G. K. Steinberg, G. Li, S. K. So, R. Majeti, G. R. Harsh, M. Van De Rijn, N. N. H. Teng, J. B. Sunwoo, A. A. Alizadeh, M. F. Clarke and I. L. Weissman, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 6662–6667.
201. J. Liu, J. He, Z. Zhang, L. Liu, Y. Cao, X. Zhang, X. Cai, X. Luo, X. Lei, N. Zhang, H. Wang, J. Chen, P. Liu, J. Tian, J. Liu, Y. Gao, H. Xu, C. Ma, S. Bai, Y. Zhang, Y. Jin, C. Zheng, B. Sui and F. Jin, Int. J. Oral Sci., 2025, 17, 56.
202. S. Ren, Y. Zhou, K. Zheng, X. Xu, J. Yang, X. Wang, L. Miao, H. Wei and Y. Xu, Bioact. Mater., 2022, 7, 242–253.
203. S. Ren, Y. Zhou, R. Fan, W. Peng, X. Xu, L. Li and Y. Xu, Chem. Eng. J., 2021, 423, 130207.
204. S. Zhang, H. Zhou, X. Liu, Y. Wu, S. Yang, J. Yin, M. H. Shubair and M. Zhang, ACS Appl. Nano Mater., 2025, 8, 9295–9311.
205. Y. Hu, X. Ma, H. Wan, P. Zhou, Y. Chen, H. Wang, X. Wang, J. Chen, R. Cai, Y. He and G. Tao, Small, 2025, e10869.
206. C. Zhu, K. Huang, T. Li, Y. Li, Y. Jin, R. Li, Z. Zhu, S. Yang, L. Xia and B. Fang, Biomaterials, 2025, 319, 123179.
207. M. Tang, R. Ma, Y. Ding, Y. Fei, Y. Wang, G. Wang, C. Zhang, Y. Chen, X. Dong, P. Chen, H. Li, H. Liu, C. Peng, J. Guo and S. Gui, J. Controlled Release, 2025, 384, 113928.
208. B. M. Conley, L. Yang, B. Bhujel, J. Luo, I. Han and K.-B. Lee, ACS Nano, 2023, 17, 3750–3764.
209. Q. Pan, W. Wu, D. Jing, W. Huang, Y. Cui, Z. Zhang, Z. Shao, H. Hu and W. Yang, ACS Nano, 2025, 19, 30786–30809.
210. Q. Liu, L. Jiang, K. Wang, J. Dai and X. Liu, Pharmaceutics, 2025, 17, 1301.
211. S. He, L. Ma, Q. Zheng, Z. Wang, W. Chen, Z. Yu, X. Yan and K. Fan, Bioact. Mater., 2024, 42, 284–298.
212. J. Yang, P. Duan, Q. Liu, H. Yu, F. Fang and X. Liu, Eur. Cells Mater., 2024, 47, 238–252.

---