Can nanozymes make the leap to the clinic? Advances, hurdles, and prospects

Abstract

Nanozymes (enzyme-like catalytic nanomaterials) represent a new generation of artificial enzymes designed to overcome key limitations of their natural counterparts, including high production cost, low stability, and challenging storage. Their versatility spans biosensing, bioimaging, therapeutics, and environmental remediation, attracting widespread interest across disciplines. Recent advances have moved nanozymes beyond benchtop proof-of-concept, with multiple in vivo studies and early clinical investigations demonstrating their translational potential. However, progress toward the clinic remains constrained by limited mechanistic understanding under physiologically relevant conditions; concerns over safety, biodegradability, and long-term efficacy; and uncertainties in scalable manufacturing, market feasibility, and regulatory approval. This Review highlights recent breakthroughs in biomedical nanozymes and identifies key knowledge gaps that must be addressed to accelerate successful clinical translation and commercialization.

Introduction: promises and challenges of nanozymes

Nanozymes (enzyme-like catalytic nanomaterials) have attracted significant attention for their advantages over natural enzymes, including tunable catalytic activity, high stability, low cost, ease of use, and the unique physicochemical properties conferred by nanoscale design [1-3]. Their applications range from biosensing and bioimaging to therapeutics, engaging researchers across biology, chemistry, materials science, and nanotechnology. Increasingly, interdisciplinary teams are designing novel nanozymes with enhanced catalytic performance by mimicking and engineering the active sites of natural enzymes [4].

Despite rapid progress in this field, the clinical translation of nanozymes remains limited. Incomplete understanding of their catalytic mechanisms, biological interactions, and behavior in complex physiological environments continues to impede widespread adoption, regulatory approval, and clinical applications in biomedical settings. This Review highlights recent progress in biomedical nanozyme research (covering biofilm, cancer, and other disease treatments, as well as diagnostic and biomedical imaging), critically examines the hurdles to translation, and identifies key questions that must be addressed for clinical implementation (Figure 1, Key figure). By integrating knowledge of nanozyme composition, structure, and catalytic mechanisms with insights from recent preclinical and early clinical studies, we aim to inform strategies for advancing their use in disease prevention, diagnosis, and therapy.

Figure 1 Key figure. Overview of the current landscape of nanozyme research, highlighting major translational hurdles and potential approaches to facilitate their clinical application.

(A) Over 1,200 distinct nanozymes have been developed to mimic the catalytic activity of natural enzymes across diverse fields, from environmental monitoring to biomedical applications. Despite their promising potential in medicine, no nanozyme-based product has yet received regulatory approval for clinical use in humans from major agencies such as the Food and Drug Administration (FDA), European Medicines Agency (EMA), Pharmaceuticals and Medical Devices Agency of Japan (PMDA), and National Medical Products Administration (NMPA). The vast majority of nanozymes are still at the preclinical stage. Some nanozymes have been evaluated for biosensing using human samples, and, to the best of our knowledge, iron oxide and gold nanozymes are the only ones that have advanced to clinical studies. A carbon dot-based nanozyme is registered in the WHO International Clinical Trials Registry Platform (ICTRP; https://trialsearch.who.int/) for the treatment of chronic actinic dermatitis and solar dermatitis; however, no additional information is currently available. (B) Hurdles to clinical translation of nanozymes. (C) Potential solutions to facilitate clinical translation of nanozymes. PK, pharmacokinetics; BD, biodistribution; AI, artificial intelligence. Figure 1B,C created with BioRender.

Biomedical applications of nanozymes

The nanozyme field has rapidly progressed from in vitro demonstrations to robust in vivo evaluations. Preclinical studies in animal models have demonstrated that engineered nanozymes can function under physiological conditions, retain catalytic activity in complex biological environments, and achieve therapeutic outcomes in disease models ranging from bacterial infections to cancer. Encouragingly, an iron oxide nanozyme platform has entered early-stage clinical investigations for oral biofilm control [5,6] and a gold nanoparticle agent has been in trials as a treatment for amyotrophic lateral sclerosis, multiple sclerosis, and Parkinson’s disease [7-10]. Although these studies are limited in scale and scope, they represent an important shift toward real-world testing. While nanozyme therapies can also be classified by immunological mode of action, for example into pro-inflammatory (such as antibacterial and antitumor) and anti-inflammatory (such as cardiovascular, neurological, and bone-related) applications, in this Review we summarize recent studies by disease focus to facilitate discussion of clinical context and translational potential.

Biofilm therapy

Biofilms (see Glossary) are notoriously resistant to conventional antimicrobials and antibiotics, posing persistent challenges in oral and systemic health settings, representing a serious threat to global healthcare systems. Nanozymes have emerged as promising candidates for antimicrobial and antibiofilm therapies, offering catalytic activity that can be harnessed for targeted pathogen elimination and biofilm disruption.

One example is ferumoxytol (referred to as FerIONP), an FDA-approved iron oxide nanoparticle formulation for anemia treatment, which has shown therapeutic potential in oral health applications [5]. In a recent clinical study, FerIONP catalytically activates H₂O₂ to enable visual detection and selective inhibition of pathogenic bacterial species (such as Streptococcus mutans (S. mutans)) within acidogenic biofilms, while sparing commensal species (Figure 2A-D). Mechanistic studies linked this specificity to the preferential binding of FerIONP to S. mutans surface proteins, while high-resolution confocal imaging, spectroscopy, and genetics-based approaches confirmed its targeted antimicrobial binding and catalytic action under acidic pH. FerIONP treatment significantly reduces cariogenic biofilm formation (Figure 2C,D) and enamel demineralization in a human in situ study (Figure 2E), demonstrating its potential for dental caries (tooth decay) prevention.

Figure 2. Ferumoxytol nanozymes exhibit diagnostic and therapeutic effects against biofilms in humans.

(A) Schematic illustration of the selective catalytic, diagnostic, and therapeutic mechanism of ferumoxytol (FerIONP), highlighting its ability to detect biofilms on natural teeth. Under acidic pH, FerIONP generates ROS in the presence of H₂O₂, selectively damaging acid-producing pathogenic bacteria while sparing commensal species. FerIONP also enables diagnosis by oxidizing a chromogenic substrate (TMB) to generate a blue signal, allowing visual detection of biofilm accumulation. (B) FerIONP detects pathogenic biofilms. S. mutans biofilms are stained blue through FerIONP-mediated oxidation of TMB in the presence of H₂O₂, allowing differentiation from biofilms formed by Streptococcus oralis (S. oralis), a non-cariogenic commensal species. (C) Colony-forming units (CFU) of S. mutans from a subset of human subjects positive for S. mutans at baseline. This result shows that FerIONP/H₂O₂ treatment eliminates S. mutans viability in human dental plaque, whereas H₂O₂ alone has no significant effect, demonstrating the selective catalytic killing activity of FerIONP. (D) Representative confocal images of intact biofilms on enamel surfaces after treatment. Scale bar = 50 μm. (E) Surface microhardness (% SMH) change for the subset of enamel samples with S. mutans at baseline. This result shows that FerIONP/H₂O₂ significantly reduces enamel demineralization compared with vehicle control, indicating effective protection against acid-induced mineral loss. Reproduced from [5] under a Creative Commons license. (F) Schematic illustration showing the dual function of FerIONP in disrupting biofilms and promoting stem cell activity. FerIONP activates H₂O₂ to generate free radicals that disrupt and kill biofilm-forming bacteria (left), while simultaneously promoting stem-cell proliferation without inducing cytotoxicity (right), illustrating its dual antimicrobial and regenerative functions. (G) Schematic diagram of the clinical treatment regimen and sample collection. The workflow illustrates initial microbial sampling (S1), determination of the working length to ensure accurate root-canal instrumentation, delivery of the assigned treatment within the canal, and final sampling (S2) to assess the treatment’s antimicrobial effect. (H) Comparison of antimicrobial efficacy across different clinical protocols. The red dashed line indicates the relationship of the median of log reduction of FerIONP/H₂O₂ relative to the median log reduction of NaOCl. (I,J) qPCR results showing significant upregulation of osteogenic (THY1 and MSX1) (I) and chondrogenic (SOX9 and COL10A1) markers (J) in FerIONP-treated stem cells from the apical papilla (SCAPs). These data indicate that FerIONP treatment promotes osteogenic and chondrogenic differentiation in SCAPs. *P < 0.05. Adapted from [6] under a Creative Commons license.

FerIONP performance can be further enhanced through combination therapy. When mixed with low concentrations of stannous fluoride, FerIONP’s peroxidase-like activity increased significantly, exhibiting synergistic effects in bacterial killing, biofilm matrix degradation, and caries protection in vivo without adverse effects on host tissues or microbiome composition [11]. This combination approach may lead to effective prevention of tooth decay with reduced fluoride exposure.

FerIONP has also been evaluated clinically for treating chronic biofilm infections in patients with apical periodontitis [6], a widespread endodontic infection affecting approximately half of the global adult population. It exhibits robust antibiofilm activity, matching the antimicrobial efficacy of sodium hypochlorite (NaOCl; the clinical gold standard) (Figure 2H) while avoiding its caustic effects. FerIONP binds to a key endodontic pathogen Enterococcus faecalis and remains catalytically active, enabling localized bacterial killing without cytotoxic effect. Unexpectedly, it promotes the proliferation of stem cells from the apical papilla (SCAPs) and induces osteogenic/chondrogenic differentiation (Figure 2I,J) through activation of SCAP pluripotency and WNT/NOTCH signaling pathways, indicating dual antimicrobial and regenerative benefits. Despite these promising findings, longitudinal studies with larger cohorts are needed to evaluate its long-term therapeutic potential.

Nanozyme-based combination therapies are also being developed for other biofilm infections [12,13]. For instance, reduced graphene oxide (rGO)@FeS₂ nanozymes integrated with Lactobacillus in a hyaluronic acid (HA) hydrogel, termed rGO@FeS₂/Lactobacillus@HA (FeLab), produced lactic acid and H₂O₂ in situ, creating optimal acidic conditions for •OH generation and enabling effective clearance of Candida vaginitis in vivo (Figure 3A,B) [12]. Similarly, a nanozyme-bacteriophage system (phage@palladium (Pd)) targeted bacterial infections while catalyzing H₂O₂ into •OH under acidic environments [13], showing in vivo efficacy in both acute bacterial pneumonia and subcutaneous biofilm models (Figure 3C,D).

Figure 3: Nanozyme-based platforms exhibit in vivo antimicrobial, antibiofilm, and immunomodulatory therapies.

(A) Schematic representation of the Candida vaginitis microenvironment and FeLab-modulated microenvironmental regulation for Candida vaginitis therapy. The left panel illustrates key pathological features, including overgrowth of C. albicans, elevated vaginal pH, disruption of the vaginal microbiota, and mucosal damage. The right panel shows how FeLab modulates the microenvironment to restore homeostasis and facilitate effective treatment of Candida vaginitis. (B) In vivo cell viabilities of C. albicans in vaginal washes following the indicated treatments, showing that both FeLab and clotrimazole, a standard antifungal drug, significantly reduce fungal viability, demonstrating the strong antifungal activity of FeLab. *P < 0.05. Reproduced from [12] under a Creative Commons license. (C) Schematic illustration of the phage@Pd structure and its antimicrobial effect. (D) Bacterial content in skin tissue homogenates on day 2 post-treatment, showing the in vivo therapeutic effect of phage@Pd against biofilm-associated infection. Ciprofloxacin (CIP), a standard antibiotic. **P < 0.01 and ***P < 0.001. Reproduced, with permission, from [13]. (E-H) Nanozyme-based immunomodulatory treatment strategy for MRSA-infected diabetic wounds. (E) Schematic illustration of the programmed therapeutic process for MRSA-infected diabetic wounds using Janus liposozyme (TSeL). TSeL consists of a liposome-like selenoenzyme (SeL) with GPx activity that scavenges ROS to restore tissue redox and immune homeostasis. It encapsulates a photosensitizer (TDTM). Treatment begins with photo-induced ROS generation for MRSA eradication, followed by redox and immune homeostasis remodeling. (F) Representative images of MRSA-infected diabetic wounds following different treatments, showing that TSeL + L treatment results in markedly accelerated wound closure compared with the other groups over the 15-day post-surgery period. Scale bars = 2 mm. (G) Quantitative analysis of dihydroethidium (DHE) fluorescence intensity on day 15, showing reduced ROS levels in both TSeL- and TSeL + L-treated groups due to the antioxidative activity of the liposozyme. L, light. (H) Quantification of the regenerated epidermal length on day 15, showing that TSeL + L-treated wounds exhibit the greatest epidermal regeneration compared with the other groups. Reproduced, with permission, from [14].

Moreover, nanozymes can be combined with other therapeutic platforms. One example is the Janus liposozyme (TSeL) in combination with photodynamic therapy (PDT) designed for treatment of methicillin-resistant Staphylococcus aureus (MRSA)-infected diabetic wounds [14]. This approach utilizes PDT to generate reactive oxygen species (ROS) during the early stage of treatment, followed by nanozyme-mediated antioxidant activity to restore redox balance in the later stage, which promotes macrophage polarization and tissue regeneration, addressing both microbial burden and impaired healing (Figure 3E-H).

In addition, emerging approaches, including single-atom nanozymes and photo-responsive nanozymes, offer new routes to improve catalytic efficiency and substrate specificity [15-17]. For example, 2D metal-organic framework nanozymes containing single zinc sites (SZN-MOFs) achieved high peroxidase-like activity and potent antibiofilm effects at low H₂O₂ concentrations in both in vitro and in vivo studies [16].

Together, these studies illustrate the expanding versatility of nanozymes against biofilm-associated infections, highlighting opportunities for targeted, combination, and multifunctional designs. Nevertheless, important questions remain regarding their long-term efficacy, microbial targeting specificity, and successful translation for widespread clinical adoption.

Cancer treatment

Cancer remains one of the leading causes of death worldwide, and conventional treatments such as chemotherapy, radiotherapy, and surgery are often limited by severe side effects and drug resistance, necessitating safer and more effective approaches [18]. Nanozymes offer a promising alternative by generating ROS, depleting intracellular antioxidants, alleviating tumor hypoxia, and modulating the immune microenvironment. These mechanisms can overcome therapy resistance and sensitize tumors to radiotherapy [19,20], photothermal therapy [21,22], chemotherapy [23], photodynamic therapy [24], and immunotherapy [25,26], while minimizing damage to healthy tissues.

Recent work has extended beyond direct tumor cell killing to explore tumor-microbe interactions. Intratumoral bacteria can contribute to tumor progression, metastasis, and treatment resistance, while tumor vascular networks and the immunosuppressive microenvironment provide shelters for their localization [27]. Targeting these interactions represents a promising anticancer strategy. For example, a bovine serum albumin-copper single-atom nanozyme (BSA-Cu SAN) selectively eliminated Fusobacterium nucleatum (F. nucleatum) in colorectal tumors, disrupting pathogen-tumor symbiosis and enhancing cancer cell killing (Figure 4A-D) [28].

Figure 4. Nanozymes for antitumor therapy.

(A) Schematic illustration of the synthesis of bovine serum albumin-copper single-atom nanozyme (BSA-Cu SAN), developed to disrupt pathogen-tumor symbionts for antitumor therapy. (B) Photograph of tumors excised 21 days after intravenous (i.v.) injection of different treatments. (C,D) Fluorescence images of tumor tissue sections from the control (PBS) and treatment (4 mg/kg BSA-Cu SAN) groups (i.v.), with nucleic acids stained blue and F. nucleatum labeled in red. Compared to the control group, minimal red fluorescence is observed in the BSA-Cu SAN-treated group, indicating elimination of F. nucleatum within the tumor microenvironment. Scale bar = 50 μm. Reproduced from [28] under a Creative Commons license. (E) Schematic illustration of FeN₅ and FeN₄ single-atom nanozymes (SAzymes). (F) Confocal laser scanning microscopy images of 4T1 cells stained with 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) to assess intracellular ROS generation across different treatment groups. FeN₅-treated cells exhibit the strongest fluorescence signal, indicating the highest ROS production. Scale bar = 25 μm. (G) Tumor weights measured 15 days post-treatment following i.v. injection. MCS, iron (Fe)-free monodisperse carbon nanospheres. **P < 0.01 and ***P < 0.001. Reproduced, with permission, from [30]. (H) Electron paramagnetic resonance (EPR) spectra of the spin-trap molecule 5,5-dimethyl-1-pyrroline N-oxide (DMPO) under different treatment conditions using nitrogen/phosphorous co-doped graphene quantum dots (NPGQDs). The strong signal observed at pH 6.2 in the presence of H₂O₂ indicates that NPGQDs generate •OH specifically under acidic conditions, demonstrating their pH-responsive catalytic activity and tumor-relevant selectivity. (I) Tumor weights measured on day 15 post-treatment. (J) Quantification of ROS intensity in ex vivo tumors collected 24 h after injection of NPGQDs. **P < 0.01, ****P < 0.0001. Reproduced, with permission, from [31].

Some other designs mimic metabolic enzymes, such as FeMoO₄ nanocatalysts with dual Fe²⁺ and Mo⁴⁺ active sites that exhibit xanthine oxidoreductase-like activity. These nanocatalysts convert tumor-abundant xanthine into uric acid, reprogramming metabolism and promoting immune crosstalk [29]. Bioinspired single-atom nanozymes have also shown impressive gains in catalytic performance [23,30]. For example, a five-coordinated single-atom FeN₅ nanozyme (Figure 4E) produced higher ROS than MCS and FeN₄ groups (Figure 4F), achieving strong antitumor effects in vitro and in vivo (Figure 4G) without toxicity [30].

Multienzyme-like nanozymes enable catalytic cascade reactions or synergistic catalysis therapies that are potentially more effective than single enzyme-like nanozymes [32]. The MCMSFT nanozyme, for instance, leverages photothermal enhancement and combines catalytic functionalities capable of O₂ generation, •OH production, and glutathione (GSH) depletion. Upon near-infrared irradiation, the nanozyme’s photothermal effect enables tumor ablation and enhances its catalytic performance, thereby alleviating tumor hypoxia and inducing ferroptosis, apoptosis, and immunogenic cell death, eliminating primary tumors and suppressing metastases [33].

Meanwhile, interest in metal-free nanozymes is increasing to address potential toxicity concerns associated with metal-based nanozymes [31,34]. For example, nitrogen/phosphorus co-doped graphene quantum dots (NPGQDs) showed peroxidase-like activity under mildly acidic conditions, as evidenced by EPR detection of the •OH radical (Figure 4H) [31]. Intratumoral delivery of NPGQD resulted in tumor growth inhibition and localized ROS generation (Figure 4I,J).

Together, these studies illustrate how nanozymes can function as multifunctional platforms in oncology, capable of targeting cancer cells, remodeling the tumor microenvironment, and synergizing with other treatment modalities. The next challenge lies in translating these advances into clinically viable therapies with predictable performance and safety.

Other disease treatments

Nanozymes are being explored for a wide range of oxidative stress-related diseases [35], including depression [36-38], androgenetic alopecia [39-41], acute gout [42,43], Parkinson’s disease [44], Alzheimer’s disease [45], cardiovascular disease [46], inflammatory bowel disease [47-50], and urinary tract infection [51]. By modulating ROS levels, nanozymes can protect tissues from oxidative damage, restore normal function, and potentially complement or replace conventional therapies.

In neurological disorders, oxidative stress plays a critical role in disease progression. CeO₂@BSA nanozymes have demonstrated ROS-scavenging activity, the ability to penetrate the blood-brain barrier, rapid clearance, and minimal toxicity. In a mouse model of depression, they alleviated depressive-like behaviors and associated neuropathology, suggesting that targeting oxidative stress could open new avenues for nanozyme-based antidepressant strategies [36].

In dermatology, androgenetic alopecia (AGA), a common type of hair loss, is closely linked to oxidative stress-induced dysfunction of hair follicles. Recently, manganese thiophosphite (MnPS₃) was identified as a potent superoxide dismutase (SOD) mimic using machine learning approaches to accelerate nanozyme discovery. MnPS₃ was then incorporated into microneedle patches (MnMNP) for localized delivery to hair follicles. In preclinical AGA models, MnMNP promoted superior hair regrowth compared to minoxidil, even with less frequent application [39], underscoring both the therapeutic potential of MnPS₃ and the role of artificial intelligence (AI) in guiding nanozyme design.

In hematology, ferumoxytol has been shown to protect stressed hematopoietic stem cells (HSCs) by scavenging ROS. By modulating ROS, ferumoxytol preserves self-renewal and regenerative capacity of HSCs, facilitating recovery after irradiation, chemotherapy, infection, or aging [52]. This antioxidant function suggests potential clinical utility in supporting hematopoietic reconstitution during transplantation or other stress conditions.

Gouty arthritis (GA), a chronic inflammatory condition driven by urate crystal deposition and immune dysregulation, has also been treated with nanozymes. As an example, an M2-macrophage-derived, exosome-cloaked hybrid nanocarrier (D-N[EM2]) integrates enzyme-like activity and photothermal responsiveness to simultaneously remove urate crystals, reduce inflammation, and modulate the immune microenvironment [43], demonstrating feasibility to address metabolic dysregulation and immune modulation.

CNM-Au8, a gold nanocrystal formulation, acts as a nicotinamide adenine dinucleotide hydride (NADH)-dehydrogenase-mimicking nanozyme and has been clinically investigated for treating neurodegenerative diseases such as Parkinson’s disease, amyotrophic lateral sclerosis, and multiple sclerosis [7-10]. It regulates intracellular bioenergetic metabolism while reducing oxidative stress by catalyzing the oxidation of NADH to its oxidized counterpart (NAD⁺), thereby enhancing adenosine triphosphate (ATP) production. Owing to its biocompatibility, ability to cross the blood-brain barrier, tunable optical properties, and enzyme-mimicking activities, CNM-Au8 represents a promising therapeutic candidate. Despite encouraging results, the clinical translation of gold nanoparticle-based therapies also remains in its infancy, primarily limited by small patient cohorts and short follow-up durations.

Collectively, these advances illustrate the breadth of nanozyme applications beyond biofilm infection and cancer, highlighting their adaptability to diverse pathological contexts. Future work should focus on refining delivery strategies, enhancing target specificity, and conducting long-term safety and efficacy evaluations. In addition, comprehensive pharmacokinetic and biodistribution studies are required to evaluate chronic nanozyme accumulation and potential toxicities.

Diagnostic and biomedical imaging

Nanozymes have emerged as powerful tools for biomedical diagnostics, combining robust catalytic activity with high stability and tunability, while compatible with miniaturized platforms. In biosensing, they have been incorporated into colorimetric, fluorescent, electrochemical, chemiluminescent, and surface-enhanced Raman scattering assays for highly sensitive and selective detection. These platforms have been adapted for detecting small molecules (e.g., H₂O₂ [58,63], GSH [63], glucose [63], dopamine [64,72], uric acid [72]), proteins (e.g., insulin [73], cardiac troponin I [74], hepatitis B surface antigen [75]), nucleic acids [76], tumor- [60,77] and pathogen-specific (SARS-CoV-2 [62]) markers, and larger biological targets (e.g., bacteria [78,79], cells [80]), expanding diagnostic capabilities in oncology, infectious disease, metabolic disorders, and beyond.

One example is the Z/Ce@hemin nanozymes, constructed by anchoring hemin onto zeolitic imidazolate framework-8 (ZIF-8)-encapsulated ceria (CeO₂) nanoparticles, enabling sensitive H₂O₂ detection for monitoring oxidative stress in Parkinson’s disease [81]. Another is a bifunctional tannic acid-modified gold nanoflowers (TA@AuNFs), which integrates peroxidase- and glucose oxidase (GOx)-like activity for colorimetric glucose sensing (Figure 5A) with a self-assembled nanoarray configuration to enhance laser desorption/ionization mass spectrometry (LDI-MS) analysis (Figure 5B,C). This system enabled early kidney cancer diagnosis, subtype classification, and differentiation between benign and malignant tumors by analyzing metabolic fingerprints from urine samples [59].

Figure 5. Nanozyme-based platforms for disease diagnosis and image-guided therapy.

(A,B) Schematic illustration of the bifunctional smart nanoplatform TA@AuNFs for both colorimetric sensing and LDI-MS analysis. (A) TA@AuNFs with peroxidase- and GOx-mimetic properties for colorimetric glucose detection. (B) Self-assembled TA@AuNF nanoarrays (SA-TA@AuNFs) for LDI-MS analysis. In the presence of Fe³⁺, TA@AuNFs self-assemble into uniform 2D nanoarrays that enhance LDI-MS ionization efficiency and reproducibility for metabolic fingerprinting. (C) Representative mass spectra of urine samples from healthy controls (HC) and renal cell carcinoma (RCC) patients, showing distinct metabolic profiles between HC and RCC groups. Over 95% of the samples within each group exhibit high similarity scores above 0.8, indicating intra-group consistency. Reproduced, with permission, from [59]. (D) Schematic of CACN for precise cancer therapy. CACN disassembles in response to ATP and acidic pH in the tumor microenvironment, releasing Dox, enhancing ROS production, and enabling switchable MRI and fluorescence imaging for real-time monitoring of therapeutic activation. (E) (I) T₁-MRI and T₂-MRI images of 4T1 tumor-bearing mice after intravenous (i.v.) injection of Dox-CACN, showing a switchable MRI contrast effect. (II) Corresponding images after i.v. injection of Dox-NACN, indicating the absence of a switchable MRI effect. (III) Mean fluorescence intensity of tumors over time following i.v. injection of Dox-CACN, indicating time-dependent Dox release. (IV) Antitumor efficacy of Dox-CACN following i.v. injection, compared to the DPBS control group. Reproduced, with permission, from [82].

Integration with biomedical imaging further enhances nanozyme utility, enabling real-time monitoring of biodistribution, accumulation, targeting, and physiological effects. Imaging modalities such as fluorescence [67,82,83], photoacoustic [69,84], computed tomography (CT) [48], and magnetic resonance imaging (MRI) [67,82,83] not only reveal in vivo nanozyme behavior but also correlate catalytic activity with therapeutic outcomes. By facilitating disease monitoring and guidance, these approaches allow clinicians to adjust treatment in real time, thereby enhancing efficacy and minimizing side effects [85]. For instance, a catalytic nanoplatform (CACN), composed of DNA-functionalized iron oxide nanoparticles (Fe₃O₄ NPs) and doxorubicin (Dox), disassembles in response to tumor-specific ATP levels and acidity, triggering drug release, ROS generation, and simultaneous MRI and fluorescence signal changes (Figure 5D,E). The imaging feedback correlates with ROS generation and Dox release, enabling noninvasive, real-time monitoring of drug release, in situ catalysis, and therapeutic activation [82].

In another study, a ferritin-based oxygen self-supply system, tracked via fluorescence imaging and MRI, relieved tumor hypoxia and improved PDT efficacy, providing a comprehensive real-time evaluation of biodistribution, accumulation, and functional performance, which can guide therapeutic optimization [83]. Naha and others have reported that cerium oxide NPs, which have SOD and catalase-like activities, also inherently produce contrast for CT, as evidenced in models of inflammatory bowel disease [48,86,87].

Collectively, these advances position nanozymes as powerful tools across the diagnostic-therapeutic spectrum. Imaging-integrated systems exemplify the potential for “theranostic” platforms that not only guide and monitor treatment in real time but also enhance therapeutic precision. At the same time, nanozyme-based biosensors are paving the way for rapid, cost-effective, and point-of-care diagnostics. Moving forward, integrating biosensing with wearable devices, multiplexed detection systems, and biomarker discovery pipelines could further broaden clinical utility and real-time health monitoring as well as biomarker discovery.

Microrobotics as a convergence platform for nanozyme functions

Nanozyme robotics represent a rapidly emerging platform where catalytic function and robotic control are combined into a single, programmable system. By integrating therapeutic, diagnostic, and targeting capabilities [88,89], these microrobotic platforms have the potential to unify the diverse functionalities of nanozymes within one modular design.

Catalytic materials can be tailored to match the therapeutic goals, whether ROS generation for antimicrobial activity, metabolic modulation in tumors, or enzymatic cascade reactions for enhanced therapy. Diagnostic elements, such as imaging contrast agents or catalytic reporters, can be incorporated for real-time monitoring of biodistribution, catalysis, and treatment response. These functionalities can be integrated into robotic platforms during fabrication through techniques such as thin-film deposition, molding, microfluidics, and 3D printing [90,91]. Targeting is achieved through programmable navigation and environmental responsiveness, allowing site-specific delivery with minimal off-target effects. External stimuli such as magnetic fields, ultrasound, light, or chemical gradients can be used to remotely guide their movement and activation, enabling targeted navigation and precise intervention in anatomically complex or otherwise inaccessible sites [90,92]. In this way, nanozyme microrobots can tackle limitations of traditional bulk catalysis, such as low reaction specificity, limited mass transport, and inefficacy within complex or spatially constrained biological environments while adding a new dimension through mechanical element for physically degrading biological substances [93].

Recent advances illustrate the integrative potential of this convergence technology. For example, a single-atom Cu-doped BiOI microrobot, actuated optomagnetically, penetrated a viscous and confined sinus environment to deliver localized photothermal and ROS therapy against biofilms at infection sites (Figure 6A,B). The therapeutic efficacy is demonstrated through in vivo studies in a rabbit sinusitis model and ex vivo experiments in porcine sinuses, indicating translational potential [94]. However, the dependence on fiber-delivered illumination, while effective in anatomically accessible regions like the sinuses, may pose limitations for treating deeper or less accessible tissues without endoscopic intervention.

Figure 6. Nanozyme robotics.

(A) Schematic illustration of single-atom Cu-doped BiOI-based nanozyme robotic swarm through a magnetic-guided optical fiber for targeting biofilm infections and (B) their demonstration using ex vivo porcine sinus model for the biofilm infection treatment, highlighting the effective biofilm removal. Reproduced, with permission, from [94]. (C) Reconfigurable nanozyme-shelled microcapsule robots for precise navigation in complex, branched spaces and (D) their localized delivery of ROS at the apical site of a branched root canal model. The localized distribution of the blue color on one side of the root canal indicates the successful localized delivery of ROS at the target site. Reproduced from [95] under a Creative Commons license. (E) Schematic illustration of iron oxide NP-based nanozyme swarm for precision fungal targeting; (F) Targetability toward C. albicans through surface binding from the iron oxide NP-based nanozyme through a robotic dabbing motion under magnetic control. The panels show before and after treatment; arrows indicate the fungal biofilms covered by iron oxide nanozyme swarms and (G) their associated antifungal killing on murine mucosa using nanozyme-microrobots. **P < 0.01, ***P < 0.001, and n.d.= not detectable. Reproduced from [96] under a Creative Commons license. (H) Schematic illustration of nanozyme microrobots for pH-responsive Fe²⁺ release, ROS, and T₂/T₁ signal conversion. (I) In vivo imaging of representative T₂-weighted MR images of 4T1 tumor-bearing mice before and 1 day after i.v. injection of nanozymes with or without magnetic targeting. The yellow circles indicate the tumor. Scale bars = 5 mm. (J) The change in the relative tumor volume over time. ***P < 0.001. Reproduced, with permission, from [97].

Nanozyme microrobots can also be magnetically assembled into dynamic, self-reconfigurable structures that adapt to anatomical constraints and varying surface topographies to target intractable biofilms [95,98]. Magnetically reconfigurable Fe₃O₄/SiO₂ microcapsules can navigate intricate root canal geometries and varying anatomical confinements, while catalyzing rapid ROS generation at the infection site, enabling localized biofilm elimination (Figure 6C,D) [95]. Fungal biofilms have also been targeted using Fe₃O₄ NP microrobots (Figure 6E-G). These magnetically actuated NP assemblies bind strongly to C. albicans on oral mucosa and perform in situ catalysis, outperforming conventional oxidative or antifungal treatments [96].

In oncology, a representative example is the design of core-shell-structured iron carbide (Fe₅C₂@Fe₃O₄) nanozymes, which combine magnetic guidance, catalytic therapy, and MRI contrast, to enable spatially controlled, image-guided tumor treatment (Figure 6H-J). This nanozyme robotic design leverages intrinsic acidity and overproduction of H₂O₂ in the tumor environment to initiate a cascade reaction that depletes tumor antioxidants and amplifies ROS production, thereby enhancing catalytic cancer therapy in vivo through intratumoral •OH generation [97]. Additionally, the high magnetization of Fe₅C₂@Fe₃O₄ also provides T₁-weighted MRI contrast in acidic tumor microenvironments, allowing environment-responsive imaging and spatial control under MRI guidance.

By merging locomotion, environmental sensing, and catalytic activity, nanozyme microrobots are capable of navigating complex biological environments in a coordinated manner, performing diagnostics, and delivering therapy in situ with high precision [95,96,99]. To realize their potential, future work must optimize performance under physiological conditions, ensure safety, and develop closed-loop control with real-time feedback systems for autonomous operation. As these challenges are addressed, nanozyme microrobotics could serve as the ultimate theranostic platform, bridging the gap between multifunctional nanozyme design and precision medicine in real-world clinical settings.

Overall, these developments highlight the versatility of nanozymes across diagnostic and therapeutic modalities and provide a foundation for tackling the translational hurdles discussed next. The insights gained from preclinical and clinical studies, particularly regarding biodistribution, clearance, and performance in humans, are poised to inform the rational design of next-generation nanozymes for clinical use.

Challenges and opportunities in clinical translation

Over the past two decades, scientific advances have transformed nanozymes from an intriguing laboratory concept into multifunctional platforms with tangible biomedical impact, spanning antimicrobial therapy, cancer treatment, diagnostics, and bioimaging (Table 1). Yet, despite this breadth of innovation, the path to clinical translation remains far from straightforward. Many exist largely as proof-of-concept studies under highly controlled experimental conditions. Moving from bench to bedside will require addressing critical gaps in mechanistic understanding, performance predictability, safety, manufacturability, and regulatory approval. Below, we frame these issues as key questions whose resolution will be essential to unlock the full potential of nanozymes in real-world medicine.

Table 1. Representative examples of nanozymes in biomedical applications ^a.

Nanozyme	Enzyme-like activity	Application	Ref
Cu1.5Mn1.5O4	POD, OXD, and GPx	Wound healing	[53]
Au1Pd3 alloy	SOD and MPO	Tumor therapy	[54]
O-NZ	SOD and GPx	Brain injury	[55]
CPTH-AT	SOD and POD	Tumor immunotherapy	[56]
ACPCAH	SOD, CAT, GOx, POD, and NOS	Diabetic wound healing	[57]
PCNSs	POD	H2O2 detection	[58]
Pt/CeO2	Uricase and CAT	Gout therapy	[42]
Pt-MNs	SOD and POD	Androgenetic alopecia treatment	[41]
Ptzyme@D-ZIF	SOD and CAT	Parkinson’s disease	[44]
CuxO@EM-K	CAT, SOD, and GPx	Alzheimer’s disease	[45]
Pt@PCN222-Mn	SOD and CAT	Inflammatory bowel disease	[47]
FA-Fht	CAT	Cancer radiotherapy	[20]
TA@AuNFs	POD and GOx	Glucose detection	[59]
Pt1/PA	POD	Cancer biomarker detection	[60]
Au@Rh-ICG-CM	CAT	Bimodal imaging and PDT	[61]
Co-Fe@hemin	POD	SARS-CoV-2 antigen detection	[62]
D-CeO2	SOD and CAT	Inflammatory bowel disease	[48]
CuPd@H-C3N4	CAT, POD, and OXD	GSH, H2O2, and glucose detection	[63]
Ferumoxytol	CAT	HSC regeneration therapy	[52]
BSA-Cu NPs	POD	Dopamine detection	[64]
CeTA	SOD and CAT	Viral pneumonia treatment	[65]
GO-Pt30-AuPt5	POD	SARS-CoV-2 and H1N1 detection	[66]
Ag2S@Fe2C-DSPE-PEG-iRGD	POD	Breast cancer theranostics	[67]
Cu-Cl MOF	SOD	Corneal burn treatment	[68]
1-FCuSA	POD	In vivo imaging of •OH and tumor treatment	[69]
MOF@Se	SOD and GPx	Atherosclerosis therapy	[70]
Ce-MSCNV	SOD and CAT	Rheumatoid arthritis therapy	[71]

^aAbbreviations: O-NZ, oligomeric nanozyme; PCNSs, porous carbon nanosheets; Pt-MNs, dissolving microneedles loaded with platinum nanozymes; Ptzyme@D-ZIF, platinum nanozymes into D-chiral imidazolate zeolite frameworks; D-CeO₂, dextran-coated cerium oxide; HSC, haematopoietic stem cells; BSA-Cu NPs, bovine serum albumin templated copper nanoparticles; CeTA, cerium-based tannic acid; Cu-Cl MOF, chlorine-coordinated copper nodes metal organic framework; Ce-MSCNV, ceria nanoparticle-immobilized mesenchymal stem cell nanovesicle

Question 1: How should catalytic performance be balanced with clinical safety in nanozyme design?

High catalytic activity often comes with trade-offs in toxicity, biocompatibility, ease/cost of synthesis, and stability in vivo. The relationship between physicochemical properties and long-term biological effects remains incompletely understood, and achieving the right balance is a major hurdle for clinical translation. Catalytic efficiency depends on multiple factors, including composition, size, shape, surface chemistry, and the spatial arrangement of active sites. Precise control over these parameters can enhance substrate specificity, reaction rates, and operational stability under physiological conditions. However, many high-performance nanozymes incorporate metals, alloys, or other components that raise toxicity concerns about metal leaching, long-term accumulation, and off-target interactions.

Understanding structure-activity relationships, for example comparing single-atom systems (e.g., FeN₅) and traditional nanoparticle platforms (e.g., Fe₃O₄), is therefore critical for linking design features to biological outcomes. Strategies such as biodegradable coatings, biomimetic membranes, and targeted delivery can help mitigate toxicity (metal leaching, long-term accumulation, off-target effects). Bridging the gap between catalytic performance and safety will require deeper mechanistic insight into nanozyme-biomolecule interactions, along with standardized toxicology assessments that reflect realistic clinical use scenarios. Interdisciplinary efforts combining advanced materials engineering, in vivo imaging, and predictive computational modeling could accelerate the design of nanozymes that are both highly active and clinically safe.

Question 2: How can catalytic mechanisms be fully elucidated to enable rational nanozyme design under biologically relevant conditions?

Mechanistic insight is essential for predicting activity, tuning selectivity, and minimizing off-target effects, yet our current understanding is often based on indirect or solely in vitro evidence. Reaction pathways, intermediate states, and structure-activity relationships are still incompletely mapped, and many nanozyme designs are still driven by trial-and-error rather than predictive models that account for both material chemistry and host factors.

Addressing these gaps will require advanced characterization approaches, such as operando spectroscopy, cryo-electron microscopy, and molecular dynamics to capture reaction intermediates and dynamic processes in real time within complex biofluids. Equally important is linking in vitro catalytic parameters with in vivo performance to ensure that mechanistic findings translate into clinically meaningful outcomes. Artificial intelligence and machine learning can help integrate experimental and computational data, reveal hidden patterns, and guide mechanism-driven design. Emerging human-mimicking platforms, such as organoids and organs-on-chips [100,101], together with well-designed in vivo studies and in situ analytical techniques, will be crucial for validating mechanisms (even at the atomistic level) under physiological conditions and accelerating rational design for clinically relevant performance.

Question 3: What manufacturing challenges must be addressed to ensure reproducibility, scalability, and regulatory compliance for clinical-grade nanozymes?

Even highly effective nanozymes can fail in translation if they cannot be produced at scale with reproducible quality and surface homogeneity or meet regulatory expectations for safety and efficacy. Current Good Manufacturing Practice (cGMP)-compliant synthesis, batch-to-batch consistency, stability during storage, and regulatory classification remain key bottlenecks.

Tackling these hurdles requires a multi-pronged approach. Establishing standardized production protocols and implementing robust quality control systems are critical to ensure reproducibility and compliance. Early engagement with regulatory agencies can clarify product classification and data requirements, thereby streamlining approval pathways. Additional strategies include developing a workforce with expertise in cGMP manufacturing and regulatory processes, engaging with translational research teams, and collaborating with contract research organizations (CROs) for specialized support. Engagement with industry and relevant clinical communities via targeted outreach can further align development with real-world needs.

A common limitation in current research is the emphasis on increasingly complex multi-component nanozyme systems, which can be difficult to reproduce and scale. Translationally, nanozyme platforms of simple design offer significant advantages due to their well-defined composition, less complex synthesis, and more predictable in vivo behavior. Focusing on agents based on a single catalytic material, particularly those with prior clinical use, such as iron oxides or other clinically explored nanomaterials, enables systematic elucidation of structure-property-bioactivity relationships, facilitates cGMP-compliant scale-up, and reduces regulatory complexity compared with multi-component architectures. These streamlined platforms provide a clear foundation for evaluating long-term safety, biodistribution, and mechanistic consistency and likely represent the most practical entry point for moving nanozymes toward clinical translation. Once such single-material systems are well understood, they can also serve as modular backbones onto which additional functionalities (e.g., targeting ligands, imaging labels, or drug payloads) can be rationally incorporated. By contrast, highly complex multi-component systems, while scientifically attractive, introduce additional challenges for cGMP manufacturing scalability, quality control, and safety assessment, which may complicate their progression toward clinical translation.

Question 4: How can nanozymes be integrated into existing or multifunctional therapeutic and diagnostic platforms without compromising performance?

Standalone nanozymes may face adoption hurdles; however, integration into existing or emerging clinical workflows could accelerate translation. Co-design with clinical workflows, including sterilization, storage, delivery devices, or combination with existing therapeutics, and early regulatory planning are essential to ensure that multifunctionality enhances, rather than undermines, safety and efficacy.

Embedding nanozymes into established drug delivery systems, imaging agents, or implantable devices can expand their theranostic potential. Examples include combining nanozymes with immunotherapies to modulate the tumor microenvironment, pairing them with antimicrobials to enhance antibiofilm efficacy, incorporating them into regenerative scaffolds to promote tissue repair, or integrating them into biosensor arrays for real-time diagnostics. The central challenge lies in coordinating multiple functionalities without compromising their individual performance. Looking ahead, cross-platform integration could accelerate adoption while generating new multifunctional technologies that bridge therapy and diagnostics, with microrobotics emerging as a powerful convergence of therapy, diagnostics, and targeting.

Additional considerations.

Beyond materials design and manufacturing, experimental design and indication choice strongly influence translational potential. Greater emphasis on mechanistic studies in vivo, including appropriate randomization, blinding, and the use of clinically relevant control (such as current gold standard treatments), is needed to reduce bias and improve data quality, reproducibility, and comparability. Larger animal models that better mimic human physiology and disease can provide more reliable information on biodistribution and efficacy than rodent models alone.

While most nanozyme work to date has focused on cancer and bacterial infections, additional translational opportunities may lie in diseases with high unmet medical need and limited effective treatment options, such as certain neurodegenerative and fungal diseases. In these settings, the substantial disease burden and lack of alternatives may lower translational hurdles and make even modest therapeutic gains clinically meaningful. At the same time, many of these conditions are chronic, underscoring the importance of carefully evaluating long-term safety and dosing regimens. Focusing nanozyme development on such underexplored indications therefore represents an emerging and potentially advantageous direction for accelerating clinical translation.

In parallel, integrating systems biology and AI into nanozyme research can help uncover complex disease-associated biological networks and nanozyme interactions, enabling the development of more effective and specific nanozyme-based theranostics. Ultimately, increased investment in translational research is vital to bridge the gap between laboratory innovations and clinical applications, and to optimize the relationship among catalytic activity, safety, clinical feasibility, and therapeutic efficacy (Figure 7). Additionally, most reported nanozymes to date fall within the oxidoreductase family, highlighting an opportunity to engineer nanozymes with a broader range of catalytic activities. Nanozymes that mimic hydrolases (e.g., proteases, glycosidases, or phosphatases) could be used to remodel extracellular matrices, modulate signaling pathways, or degrade pathological aggregates, while transferase- or lyase-like nanozymes might enable targeted manipulation of metabolic or epigenetic pathways. Expanding rational design principles beyond redox catalysis could therefore open new avenues for nanozyme-based therapies and diagnostics that act on diverse biochemical processes.

Figure 7. Key features required for the clinical translation of nanozymes.

Concluding remarks and future perspectives

Nanozymes have emerged as powerful alternatives to natural enzymes, offering unique advantages such as tunable catalytic activity, high stability, and multifunctionality. Despite rapid advancements, their clinical translation remains limited, underscoring the need to bridge the gap between laboratory innovation and real-world therapeutic use. Addressing this challenge requires coordinated efforts across design, safety evaluation, scalable manufacturing, and regulatory alignment (see Outstanding questions).

Outstanding questions.

1. How can nanozyme design balance catalytic performance with long-term safety and biocompatibility in complex physiological environments?

2. What experimental and computational approaches are needed to fully elucidate nanozyme catalytic mechanisms under biologically relevant conditions?

3. How can manufacturing be standardized to ensure reproducibility, scalability, and regulatory compliance for clinical-grade nanozymes?

4. How can nanozymes be integrated into existing therapeutic workflows and diagnostic platforms without compromising performance?

5. What are the most promising clinical applications where nanozymes can provide clear advantages over existing technologies?

Translation will demand a deliberate shift from academic proof-of-concept studies to clinically oriented development. This includes building partnerships with clinicians, entrepreneurs and industry, securing translational funding, and developing scalable production pipelines under cGMP conditions while seeking regulatory approvals. Engaging with the biomedical community through clinically oriented collaborations, publications, and conferences will be essential to position nanozymes within relevant therapeutic contexts and advance their practical use in patient care.

Safety and biocompatibility remain central concerns. Many inorganic nanozymes are poorly degradable, posing risks of long-term accumulation and toxicity. Among current platforms, those based on iron oxides are generally considered to have one of the more favorable safety profiles, owing to their clinical use and FDA-approved formulations. Other inorganic nanozymes, such as gold or cerium oxide, are also relatively favorable starting points because their pharmacokinetics, organ distribution, and toxicity have already been extensively evaluated in preclinical studies. Safety uncertainties are more pronounced for nanozymes that incorporate high levels of less-studied or highly redox-active metals, or that possess complex, multifunctional, and poorly degradable architectures, which may carry increased risks of long-term accumulation, metal-ion release, immunotoxicity, off-target ROS generation, and chronic organ accumulation. We therefore suggest that early clinical translation focus on nanozymes based on materials with established or emerging safety records, while advancing strategies such as immunocompatible coatings, biodegradable formulations, localized or topical delivery routes, and the optimization of size and shape parameters to facilitate efficient clearance. Encapsulation in nanocarriers (e.g., hydrogels, liposomes) offers additional pharmacokinetics tunability while minimizing systemic exposure [102].

Structural and compositional complexity often complicates reproducibility and large-scale manufacturing. Priority should be given to clinically feasible nanozyme platforms with well-defined compositions and scalable synthesis routes, thereby creating a foundation for systematic, long-term efficacy and safety evaluations and detailed mechanistic studies. Once this foundation is established, more sophisticated multifunctional or stimuli-responsive systems can be rationally designed for advanced clinical applications.

Quantitatively assessing nanozyme catalytic activity in vivo remains a major challenge. Although nanozymes often exhibit strong ROS-regulating activity in vitro, translating and reliably measuring these effects in living systems is far more complex. Biological components in physiological environments can alter nanozyme stability, catalytic function, and targeting efficiency, making it difficult to distinguish intrinsic catalytic activity from host-mediated effects. For clinical effectiveness, nanozymes must reach specific tissues and subcellular compartments, selectively modulate distinct ROS or metabolite species, and maintain biocompatibility and biodegradability. Moreover, the context-dependent roles of ROS and other reactive intermediates across different diseases further complicate assay design and mechanistic interpretation. These hurdles highlight the need for advanced imaging, biosensing, and analytical tools capable of tracking in vivo catalytic events in real time by directly coupling signal changes to catalytic turnover. Complementary microphysiological models, together with computational modelling, will also be important for linking measured in vivo activity to nanozyme physicochemical properties and disease microenvironments.

Emerging directions highlight the potential of multifunctionality and convergence, in which diverse modalities are combined into a unified system to enhance performance. Integrating therapeutic and diagnostic functions, or combining nanozymes with probiotics, drugs, or biologics, could address complex pathologies more effectively than mono-functional designs. Microrobotic platforms represent a particularly exciting frontier, unifying targeting, catalysis, and imaging within programmable systems that enable real-time monitoring and adaptive therapy.

In summary, nanozymes represent a transformative class of catalytic nanomaterials with broad potential in biomedicine. Their successful translation will depend on rational design, rigorous safety evaluation, scalable synthesis, and regulatory alignment. Human-mimicking experimental models under physiologically relevant conditions and AI-driven approaches offer powerful tools to accelerate this transition by enabling predictive, data-driven design, mechanistic discovery, and enhanced clinical relevance. With continued interdisciplinary innovation and strategic problem-solving, nanozymes are well-positioned to move from promising laboratory tools to tangible clinical solutions.

Highlights.

• Nanozymes are emerging as versatile artificial enzymes with tunable catalytic, diagnostic, and therapeutic functions across diverse biomedical applications.

• Clinical translation is constrained by limited mechanistic understanding, lack of longitudinal evaluation, safety and biodegradability concerns, challenges in scalable manufacturing, and regulatory hurdles.

• Single-material nanozyme platforms are proposed as practical foundations for translation and as starting points for clinically guided expansion into more complex multifunctional systems.

• Integration with advanced microrobotic systems offers powerful convergence platforms for precision theranostics and minimally invasive interventions.

• This Review outlines strategies to overcome translational hurdles, offers perspectives on emerging directions, and frames key questions for future human efficacy and safety studies.

Glossary

Biofilm

a three-dimensional microbial community encased in self-produced extracellular polymeric substances (EPS) and adherent to an inert or living surface.

Catalase (CAT)

an enzyme that catalyzes the decomposition of hydrogen peroxide (H₂O₂) into molecular oxygen (O₂) and water (H₂O).

Ferroptosis

a type of regulated cell death driven by iron-dependent lipid peroxidation.

Glucose oxidase (GOx)

an enzyme that catalyzes the oxidation of glucose to gluconic acid and H₂O₂.

Glutathione peroxidase (GPx)

an enzyme that protects cells from oxidative damage by catalyzing the reduction of H₂O₂ and organic hydroperoxides to H₂O or their corresponding alcohols.

Multienzyme-like nanozymes

nanozymes that exhibit more than one enzyme-like activity.

Myeloperoxidase (MPO)

an enzyme that catalyzes the formation of hypochlorous acid (HOCl) from H₂O₂ and chloride ions (Cl⁻).

Nitric oxide synthase (NOS)

an enzyme that catalyzes the production of nitric oxide (NO).

Oxidase (OXD)

an enzyme that catalyzes the oxidation of substrates by O₂, resulting in the production of H₂O₂ or H₂O.

Peroxidase (POD)

an enzyme that catalyzes the oxidation of substrates by H₂O₂.

Reactive oxygen species (ROS)

oxygen-containing reactive molecules that include both free radicals, such as superoxide radical (O₂•⁻), hydroxyl radical (•OH), and nitric oxide radical (•NO), and non-radicals, such as H₂O₂, singlet oxygen (¹O₂), and HOCl.

Stannous fluoride

a widely used source of fluoride in oral health care products, such as toothpaste and mouth rinses.

Self-cascade reactions

a sequence of consecutive steps in which the product of one enzymatic reaction serves as the substrate for the next, creating a self-sustaining reaction cycle.

Single-atom nanozymes

nanozymes with atomically dispersed metal atoms, combining the benefits of both homogeneous and heterogeneous catalysis.

Superoxide dismutase (SOD)

an enzyme that catalyzes the dismutation of •O₂⁻ into H₂O₂ and O₂.

Uricase

an enzyme that catalyzes the oxidation of uric acid to allantoin.

Xanthine oxidoreductase

an enzyme that catalyzes the oxidation of hypoxanthine to xanthine and xanthine to uric acid.