Valence-engineered catalysis-selectivity regulation of molybdenum oxide nanozyme for acute kidney injury therapy and post-cure assessment

Li, Liangyu; Liu, Xiaotong; Liu, Guanghe; Xu, Suying; Hu, Gaofei; Wang, Leyu

doi:10.1038/s41467-024-53047-1

Download PDF

Article
Open access
Published: 08 October 2024

Valence-engineered catalysis-selectivity regulation of molybdenum oxide nanozyme for acute kidney injury therapy and post-cure assessment

Nature Communications volume 15, Article number: 8720 (2024) Cite this article

11k Accesses
67 Citations
Metrics details

Subjects

Abstract

The optimization of the enzyme-like catalytic selectivity of nanozymes for specific reactive oxygen species (ROS)-related applications is significant, and meanwhile the real-time monitoring of ROS is really crucial for tracking the therapeutic process. Herein, we present a mild oxidation valence-engineering strategy to modulate the valence states of Mo in Pluronic F127-coated MoO_3-x nanozymes (denoted as MF-x, x: oxidation time) in a controlled manner aiming to improve their specificity of H₂O₂-associated catalytic reactions for specific therapy and monitoring of ROS-related diseases. Experimentally, MF-0 (Mo average valence 4.64) and MF-10 (Mo average valence 5.68) exhibit exclusively optimal catalase (CAT)- or peroxidase (POD)-like activity, respectively. Density functional theory (DFT) calculations verify the most favorable reaction path for both MF-0- and MF-10-catalyzed reaction processes based on free energy diagram and electronic structure analysis, disclosing the mechanism of the H₂O₂ activation pathway on the Mo-based nanozymes. Furthermore, MF-0 poses a strong potential in acute kidney injury (AKI) treatment, achieving excellent therapeutic outcomes in vitro and in vivo. Notably, the ROS-responsive photoacoustic imaging (PAI) signal of MF-0 during treatment guarantees real-time monitoring of the therapeutic effect and post-cure assessment in vivo, providing a highly desirable non-invasive diagnostic approach for ROS-related diseases.

2D vanadium carbide MXenzyme to alleviate ROS-mediated inflammatory and neurodegenerative diseases

Article Open access 13 April 2021

In vivo three-dimensional multispectral photoacoustic imaging of dual enzyme-driven cyclic cascade reaction for tumor catalytic therapy

Article Open access 11 March 2022

Screening and regulation of nanozyme activity via liquid metals coined electron rearrangement and phase engineering

Article Open access 26 March 2026

Introduction

Nanozymes can mimic the activity of natural enzymes, and have attracted considerable attention due to their advantages of low cost, high stability, mass production and durability^1,2. However, the main defect of nanozymes is poor reaction selectivity because of the lack of enzyme-like molecular recognition units with the spatial and componential cooperation observed in natural enzymes³. Oxidoreductases such as catalase (CAT)⁴, superoxide dismutase (SOD)⁵, peroxidase (POD)⁶, and oxidase (OXD)⁷ are in the spotlight for their ability of scavenging or generating reactive oxygen species (ROS) which are associated with the occurrence and development of various diseases^8,9,10. Therefore, it is of great significance to manipulate the selectivity of such nanozymes. To this end, plenty of efforts have been made on the intrinsic engineering of active centers in nanozymes, including atomic doping^11,12, coordination environment adjustment¹³ and crystal facet regulation^14,15,16. However, the modulation of selectivity of multiple enzymatic reaction pathways involving the same substrate has rarely been explored mechanistically and remains a challenge for use in specifically catalytic therapy of diseases. In particular, nanozymes with both CAT- and POD-like activities require H₂O₂ as substrate but produce O₂ and hydroxyl radicals (·OH), respectively, which will inevitably compete for H₂O₂ to a certain extent, affecting H₂O₂ utilization and performance in target applications¹⁴. Recently, we have proved that the valence engineering is a fascinating and powerful strategy in modulating the performance (activity) of nanozymes¹⁷. Specifically, the valence states of active sites have been adjusted by varying the calcination conditions for Co/TiO₂ single-atom nanozymes¹⁷, doping metal into spinel oxide ZnMn₂O₄¹⁸ and investigating the size effects on ruthenium nanoparticles (RuNPs)¹⁹. Correspondingly, with the increasing content of Co²⁺, Mn⁴⁺ or Ru⁴⁺, single or multiple enzyme-like activities remarkably improved, facilitating the treatment of ROS-related models such as tumor, inflammatory bowel disease and liver injury. To date, however, few guidelines exist for the design and regulation of reaction selectivity (pathway) of nanozymes with multi-activities through valence-engineering approach.

As an essential trace element for the survival of living organisms and a vital cofactor for molybdenum (Mo)-based natural enzymes, molybdenum is a low-cost transition metal with desirable biosafety²⁰. Due to the abundance of easily lost single electrons in the electronic configuration, Mo has a variety of oxidation states²¹, and the reversible transition between Mo (IV), Mo (V), and Mo (VI) imparts anti– and pro-oxidative enzyme-like activities to Mo-based nanomaterials²². Recent studies on Mo-based nanozymes have mainly focused on the regulation of enzyme-like catalytic activity and synergistic action of enzyme-mimicking active centers rather than the reaction selectivity. Taking molybdenum oxide as an example, single or multiple enzyme-mimicking activities could be improved by metal doping (Fe-MoO_v)²³, intercalation-activation (NH-MoO_3-x nanobelts)²⁴ and coordination-driven self-assembly (MoO_x-Cu-Cys)²⁵, of which the underlying mechanisms were ascribed to the generation of defect sites (Fe substitution and oxygen vacancy defect), Mo valence reduction and rich defects formed on MoO_3-x, and the electronic structure engineered through amino-acid-bridged electron transfer from single Cu atoms to the support, respectively. Nevertheless, the selectivity modulation of H₂O₂-associated enzymatic reactions such as CAT- and POD-like pathways with regard to Mo-based nanozymes still remains unexplored. In addition, valence charge transfer and transition between different valence states of Mo contributes to near-infrared local surface plasmon resonance (LSPR) absorption^26,27, hence, Mo-based nanomaterials can be applied as effective nanoagents for photothermal therapy (PTT) and photoacoustic imaging (PAI) as well^28,29.

Acute kidney injury (AKI) has been considered as one of the most common clinical complications with high morbidity and mortality, which is asymptomatic and has no characteristic clinical manifestations until extremely loss of renal function characterized by decreased renal excretion function and increased nitrogen metabolism accumulation^30,31. Therefore, effective treatment, precise diagnosis and post-cure assessment of AKI are urgently needed and of great clinical significance. Oxidative stress is the most dominant pathophysiological mechanism in the occurrence and development of AKI, which may lead to the abnormality of renal oxidative metabolism, thereby ROS produced in excess by renal-infiltrating triggers the damaging of kidneys and causes AKI. Thus ROS is definitely a key target for the prevention and treatment of AKI^32,33,34. Currently, the antioxidant N-acetyl cysteine (NAC) is mainly used in the clinical therapy of AKI³⁵, but it obviously suffers from high dose and low bioavailability. As such, nanozymes with promising ROS scavenging ability was proven to be excellent alternatives for the treatment of acute colitis³⁶, acute gout³⁷, acute kidney injury³⁸, androgenetic alopecia³⁹ and other ROS-related diseases^40,41.

In addition, current clinical diagnostic and post-cure evaluation for AKI mainly depends on the measurement of renal function indicators such as blood urea nitrogen (BUN) and serum creatinine (CRE) which are insensitive to early-stage kidney dysfunction. Although imaging techniques like magnetic resonance imaging (MRI), positron emission tomography (PET) and single-photon emission computed tomography (SPECT) have been employed to evaluate different stages of renal dysfunction, however, they suffer from low sensitivity, radiation risk or high cost. Non-invasive optical imaging in the near-infrared (NIR) region with deep penetration, high resolution and sensitivity, fast feedback and non-ionizing radiation enables the direct visualization and monitoring of deep tissues such as kidney in vivo^42,43,44,45. As such, it is highly desirable to develop renal activatable NIR optical sensors that specifically respond to the kidney related biomarkers (e.g. ROS-responsive). In especial, the theranostic probes for real-time imaging and amelioration of AKI are still lacking and challenging.

Herein, by taking redox-active molybdenum oxide (MoO_3-x) as a model matrix, we demonstrated how the valence state regulation of Mo affects the catalytic reaction selectivity, that is, how to tune the H₂O₂ activation pathway on the Mo-based nanozymes for entire steering the specificity between associated CAT-like and POD-like enzymatic reactions. The valence state of Mo in Pluronic F127-coated MoO_3-x NPs (hydrophilic) was well tuned by a mild oxidation approach which enabled the gradual increase of the average valence and re-distribution of the valence states between Mo⁶⁺, Mo⁵⁺, and Mo⁴⁺ in a controllable fashion. Simultaneously, the morphology, size, zeta potential, and phase composition of MF-x (x refers to the oxidation time implemented) remained almost unchanged. Experimental results manifested that MF-0 with a low average valence and MF-10 with a high average valence of Mo exhibited exclusively the optimal CAT- or POD-like activity, respectively (Fig. 1). Density functional theory (DFT) calculations explicitly disclosed that for MF-0, the free energy barrier of the rate-determining step (RDS) from H₂O₂ to O₂ was 1.21 eV, which was distinctly lower than that from H₂O₂ to ·OH (2.24 eV), suggesting the production of O₂ (CAT-like activity) was much more energetically advantageous. As for MF-10, the free energy barrier of the RDS from H₂O₂ to ·OH (0.89 eV) was significantly lower than that from H₂O₂ to O₂ (1.10 eV), indicating the vastly superior selectivity toward POD-like activity.

Fig. 1: Schematic illustration of the valence-engineered catalysis-selectivity regulation of molybdenum oxide nanozymes (MF-x) and application for acute kidney injury (AKI) therapy as well as the post-cure assessment through ROS-activatable photoacoustic imaging (PAI).

Interestingly, MF-0 posed a strong potential in AKI treatment, showing excellent therapeutic efficacy in vitro and in vivo, as well as desirable stability and biocompatibility. More intriguingly, MF-0 was able to induce the photoacoustic (PA) signal (driven by NIR light) variations with the consumption of ROS during treatment, guaranteeing the real-time and in situ monitoring of ROS in deep tissues. In AKI mice, the diminished PA signal (“off”) caused by the excess ROS, was turned on as ROS in the kidneys were scavenged in AKI-cured mice (treatment with MF-0) (Fig. 1). The ROS-responsive switchable PAI circumvented false-positive signals from nonspecific retention and successfully realized real-time monitoring of therapeutic process and post-cure assessment for AKI in vivo. This work provides a promising example on tuning the H₂O₂ activation pathway on Mo-based nanozymes for highly efficient catalytic therapy and post-cure assessment via non-invasive PAI of ROS-related diseases.

Results

Synthesis and characterization

Transmission electron microscope (TEM) image indicated that the hydrophobic MoO_3-x nanoparticles (NPs) were obtained with a uniform morphology and TEM observed size of ~8 nm (Fig. 2a). As shown in Fig. 2b, high-angle annular dark-field scanning transmission election microscope (HAADF-STEM) image further presented regular lattice fringes with lattice spacing of 0.247 and 0.351 nm, corresponding to the (–2 1 1) and (–1 1 1) crystal planes of tugarinovite MoO₂, respectively. Meanwhile, Mo and O were observed uniformly distributed in the particles according to the elemental mapping. In addition, XRD pattern of MoO_3-x NPs (Supplementary Fig. 1) matched that of MoO₂ (JCPDS 32-0671), further proving the successful preparation of molybdenum oxide NPs.

Fig. 2: Characterization of MoO3-x, MF-0, MF-2, MF-3.5, MF-5 and MF-10. — **Fig. 2: Characterization of MoO_3-x, MF-0, MF-2, MF-3.5, MF-5 and MF-10.**

Hydrophobic MoO_3-x was then transferred to hydrophilic NPs through coating of amphiphilic Pluronic F127, which were then oxidized for different times (0, 2, 3.5, 5 and 10 h) under the mild oxidation condition to form MF-x, namely, MF-0, MF-2, MF-3.5, MF-5 and MF-10, respectively, showing various enzyme-like activities. TEM images indicated that MF-x samples were well dispersed (Supplementary Fig. 2) with an average dynamic light scattering (DLS) size from ~21 to ~24 nm (Supplementary Fig. 3a), and zeta potential from –10.7 to –14.6 mV (Supplementary Fig. 3b). These slight variations in size and zeta potential would have negligible effects on the enzyme-like activity of nanozymes^46,47,48. Meanwhile, HAADF-STEM image and the elemental mapping of MF-0 (Supplementary Fig. 4) presented consistent microstructure with that of hydrophobic MoO_3-x NPs. FT-IR spectra (Supplementary Fig. 5) and thermogravimetric (TG) analysis (Supplementary Fig. 6) further confirmed the connectivity of organic components for the as-prepared nanozyme, that is, the presence of oleylamine and oleic acid as the surface ligands of hydrophobic MoO_3-x NPs, as well as the F127 polymer coating on hydrophilic MF-0. Furthermore, XPS fitting spectra for Mo 3d (3d_3/2 and 3d_5/2) revealed the mixed-valence states of Mo⁶⁺, Mo⁵⁺ and Mo⁴⁺ in MF-x, with significant differences in the content of each valence state among the samples (Fig. 2c). As illustrated in Fig. 2d, the percentage content of Mo⁴⁺ decreased from 56.3% for MF-0 to 6.5% for MF-10, and the percentage of Mo⁶⁺ increased from 20.5% to 74.8%. Accordingly, the average valence of Mo increased from 4.64 for MF-0 to 5.68 for MF-10. As expected, the elevation of high valence (Mo⁶⁺) content and the average valence of Mo were positively correlated with the oxidation time for valence tuning of Mo, thus performing the regulation of enzyme-like activity. UV-vis-NIR absorption spectra in Fig. 2e exhibited noticeable decrease of absorption from MF-0 to MF-10, with obviously different colors (MF-0 and MF-2: dark grey, MF-3.5: navy blue, MF-5 and MF-10: yellow), due to the variation of valence state of Mo.

Valence-engineered catalysis-selectivity regulation of MF-x and DFT calculations

The reversible transition among multiple valence states of Mo endowed Mo-based nanozymes with antioxidation and oxidation-related activities. With the aim to study the relationship between the catalysis selectivity and Mo valence states for MF-x, the two enzyme-like activities of CAT and POD that inevitably compete for H₂O₂ were investigated. Firstly, CAT-like activity of each sample was evaluated by testing oxygen production in the presence of H₂O₂. Compared with MF-0 and MF−2 that exhibited the most excellent capability for oxygen production, MF-3.5 showed weak capability, and MF-5/MF−10 indicated almost no CAT-like activity under both pH 7.4 (Fig. 3a) and 6.5 (Supplementary Fig. 7). That is, from MF-0 to MF-10, CAT-like activity gradually declined and presented a pH-independent manner. Then, 3,3,5,5-tetramethylbenzidine (TMB) oxidation assay was performed to compare POD-like activity of MF-x samples. As shown in Fig. 3b and Supplementary Fig. 8, the absorption intensity difference (at 650 nm) of TMB aqueous solution treated with MF-x at pH 6.5 gradually increased within 20 min, suggesting the opposite trend to that of CAT-like activity, meanwhile MF-10 exhibited the strongest POD-like activity. The characteristic 1:2:2:1 signal observed in electron spin resonance (ESR) spectra (Fig. 3c) manifested the production of hydroxyl radicals (·OH) catalyzed by MF-x. Notably, unlike at pH 6.5, the absorption intensity differences (at 650 nm) at pH 7.4 were almost unchanged (Supplementary Fig. 9), proving that the POD-like activity of MF-x acted in a pH-dependent manner. In short, MF-0 and MF-10 exhibited optimal CAT/POD-like activities, respectively.

**Fig. 3: Catalysis-selectivity regulation based on Mo valence and DFT calculations for the mechanism exploration.**

The POD-like activity enhanced with the increment of the average valence of Mo (from 4.64 of MF-0 to 5.68 of MF-10), while it was uncertain whether the summit has been reached (Fig. 3b), thus the oxidation time was prolonged to prepare MF-20 and MF-30 with higher proportion of Mo⁶⁺ for further elucidating the evolution of the POD-like activity. As expected, the percentage of Mo⁶⁺ in MF-20 and MF-30 increased to 79.4% and 89.5%, correspondingly the average valence of Mo attained to 5.73 and 5.89, respectively (Supplementary Fig. 10). UV-vis-NIR absorption spectra of MF-20 and MF-30 were analogous to that of MF-10 (Fig. 2e), likewise the DLS size and zeta potential (Supplementary Fig. 11). TMB oxidation assay in Supplementary Fig. 12 demonstrated that the POD-like activity of MF-20 was obviously higher than MF-30 but weaker as compared to MF-10 at pH 6.5 (Fig. 3b). Correspondingly, the production of ROS was not observed at pH 7.4, which is pH-dependent. Altogether, within a certain range, the POD-like activity of MF-x enhanced with the increase of the average Mo valance, and then declined.

To elucidate the difference of the reaction selectivity between MF-0 and MF-10, ESR spectra of the two samples was first conducted. As shown in Fig. 3d, MF-0 and MF-10 presented resonance signals at g = 2.003 with analogous intensity, indicating that there was little difference in the oxygen vacancy (O_V) concentration of the two samples. Thus, the contribution of surface O_V to the selectivity of H₂O₂-associated enzymatic reactions of MF-x samples was considered to be negligible. Furthermore, the rapid cycles of redox process were essential for the activity of nanozymes containing redox-active centers (Mo with different valence states herein)¹⁷, thus the valence state of Mo in MF-x was reasonably considered as the key factor to determine the catalysis selectivity. Additionally, density functional theory (DFT) calculations were carried out to elucidate the catalytic origin and underlying mechanism for the catalysis selectivity of MF-0 and MF-10. Firstly, it is noted that the optimization of the surface models (Supplementary Fig. 13 and Supplementary Fig. 14) was described in the section of experimental methods (calculation details). Next, as shown in Supplementary Fig. 15, the charge density difference for MF-0 and MF-10 intuitively displayed that Mo lost more electrons in MF-10 than in MF-0, indicating the higher valence state of Mo in MF-10, which was agree with the XPS results (Fig. 2d), and also verified the reliability of the computational modeling. To obtain the intrinsic rate-determining step (RDS) of the catalytic reaction pathways for the production of ·OH (POD-like activity) and O₂ (CAT-like activity) of both MF-0 and MF-10, the corresponding free energy diagrams were further investigated (Fig. 3e, Supplementary Fig. 16 and Supplementary Fig. 17). As for MF-0, the RDS of the ·OH formation pathway was *OH + *OH → *OH + ·OH (·OH desorption), and the RDS of the O₂ generation pathway was *O₂ → MF-0 + O₂ (O₂ desorption), with free energy barriers of 2.24 eV and 1.21 eV, respectively. The lower barrier for the RDS of the O₂ generation pathway indicated that MF-0 was more likely to catalyze H₂O₂ to generate O₂ (CAT-like activity) rather than ·OH (POD-like activity), which was thermodynamically favorable. Correspondingly for MF-10, the RDS of the ·OH formation pathway was *OH + *OH → *OH + ·OH (·OH desorption), and the RDS of the O₂ generation pathway was *O + *H₂O₂ → *OH + *OOH (proton transfer), with free energy barriers of 0.89 eV and 1.10 eV, respectively, suggesting the remarkable selectivity toward POD-like activity.

In addition, electronic structure analyses including the projected density of state (PDOS) analysis, charge density difference and the calculation of Bader charge were further performed to uncover the mechanism for the significant catalysis selectivity. As depicted in Fig. 3f, the position of the d-band center of the Mo orbital for MF-10 (–3.66 eV) was far away from the Fermi level compared with MF-0 (–1.77 eV), resulting in weaker interaction between the active sites of MF-10 and intermediate species, which further promoted the desorption of ·OH and accelerated the RDS for ·OH generation (Fig. 3e). Thus, MF-10 eventually exhibited far superior POD-like activity to MF-0. Moreover, as illustrated in Fig. 3g, Supplementary Fig. 18 and Supplementary Table 1, the charge transfer from MF-0 to 2*OH was 1.22 e and higher than that for MF-10 (0.77 e), indicating that MF-0 had stronger adsorption effect on the 2*OH. This also confirmed again the difficulty of ·OH desorption for MF-0 (with higher free energy barrier than MF-10) in the free energy diagrams (Fig. 3e), and thus MF-10 exhibited much better POD-like activity. Meanwhile, the charge transfer from MF-0 to *O₂ was 0.86 e, higher than MF-10 (0.25 e), suggesting the closer connection between MF-0 and *O₂, which was also in accordance with the endothermic and exothermic processes of the oxygen desorption step (*O₂ → MF-0/MF-10 + O₂) on MF-0 and MF-10, respectively (Fig. 3e).

Overall, the selectivity of H₂O₂-associated enzymatic reactions of MF-x could be tuned by a feasible mild oxidation valence-engineering strategy, as summarized in the schematic diagram (Fig. 3h). MF-0 (Mo average valence 4.64) and MF-10 (Mo average valence 5.68) exhibited exclusively efficient CAT- or POD-like activity, respectively.

ROS scavenging capacity of MF-0 in vitro

In view of the successful regulation of the catalysis selectivity of MF-x by the valence-engineering strategy, MF-0 with the optimal CAT-like performance was further investigated for antioxidation related biomedical applications. H₂O₂ and ·OH were selected as representative ROS for the evaluation of the in vitro antioxidant capacity of MF-0. As the concentration of H₂O₂ increased from 0 to 6 mM, the UV-vis-NIR absorption of MF-0 aqueous solution was gradually reduced and the color changed from dark to light (Fig. 4a), due to the oxidation of MF-0 during the consumption of H₂O₂. Additionally, by using the potassium titanium oxalate (PTO) probe, the inhibition of H₂O₂ was observed to be dependent on the MF-0 dosage, specifically, 87.3% of H₂O₂ was eliminated by MF-0 at a dose of 30 μg Mo/mL (Fig. 4b & Supplementary Fig. 19a). Next, methylene blue (MB) was utilized as an indicator to detect ·OH, thus evaluating the ·OH scavenging effect of MF-0. It was found that 81.2% of ·OH was removed by MF-0 at a dose of 10 μg Mo/mL, and an inhibition rate of 44.8% was still attained even at 1.0 μg Mo/mL (Fig. 4c & Supplementary Fig. 19b). Lastly, as shown in Fig. 4d & Supplementary Fig. 19c, up to 91.5% of ABTS· can be scavenged by MF-0 at a low dose of 10 μg Mo/mL, indicating the desirable clearance ability of MF-0 for oxidative free radicals like ABTS· as well. Although no scavenging ability of MF-0 for superoxide anion (·O₂^-) was found (Supplementary Fig. 20). Overall, MF-0 was proved to possess excellent broad-spectrum antioxidant properties. In addition, it is worth noting that a certain increase of Mo valence state (Supplementary Fig. 21) observed in MF-0 (meanwhile the TEM observed particle size was reduced to ~3 nm, Supplementary Fig. 22) during scavenging ROS will not cause POD-like activity to facilitate the accumulation of ·OH in neutral pH conditions like the renal microenvironment of AKI mice³⁸. Because the POD-like activity of MF-x acted in a pH-dependent manner, as mentioned above, even MF-10 with high Mo valence did not show any POD-like activity under pH 7.4 (Supplementary Fig. 9).

Prior to assessment of feasibility of MF-0 in cellular level, methyl thiazolyl tetrazolium (MTT) assay was first implemented to check the cytotoxicity of MF-0. Although the viability of MREpiC cells descended slightly with the increase of MF-0 concentration after incubation for 24 h, nearly 90% of the cells survived even at a high concentration of 400 μg/mL (Fig. 4e). Besides, the absorbance at 600 nm and morphology, as well as zeta potential and average DLS size of MF-0 changed little during 14 days of storage (Supplementary Fig. 23). These results showed that MF-0 owned desired stability and biocompatibility for further biomedical applications. Afterwards, as depicted in Fig. 4f, compared with the negative control group (no addition of H₂O₂ and MF-0), cell viability of the positive control group (only H₂O₂) significantly declined to 66.5% due to H₂O₂-induced excessive oxidative stress and thereby cell death. With the following addition of MF-0, the cell viability improved steadily and reached ~90% at the dose of 80 μg/mL (MF-0). Correspondingly, as shown in Fig. 4g, h, the fluorescence imaging of intracellular ROS by using DCFH-DA probe demonstrated that the group II (positive control) emerged the strongest green fluorescence of DCF (from oxidation of DCFH-DA by ROS), indicating the most severe oxidative stress occurred. While in MF-0 treated groups (III, IV, V), intracellular oxidative stress (ROS) was distinctly inhibited, and recovered to a comparative level with the negative control group (I) at a dose of 80 μg/mL. These results were in good agreement with those in above cell survival experiments (Fig. 4f), which confirmed that the introduction of MF-0 could scavenge intracellular ROS effectively and protect cells from oxidative stress-induced cell damages.

In vivo therapeutic assessments of MF-0 for AKI

Encouraged by the desirable in vitro ROS clearance capability of MF-0, its therapeutic effects on ROS-related acute kidney injury (AKI) in mice were further explored. First, MF-0 was labeled with the fluorescent dye IR780 (IR780@MF-0) to inspect its circulation, targetability, and biodistribution in AKI mice. The in vivo fluorescence imaging showed that IR780@MF-0 could obviously accumulate to the kidney within 10 min, and reached the maximum accumulation in 30 min and lasted for an appropriate period (with a slight attenuation in 60 min) (Supplementary Fig. 24). At the same time, the results of ICP-MS detection of Mo content (Supplementary Fig. 25) indicated that the retention of MF-0 in the kidneys of AKI mice attained the maximum (2.8% ID per gram of kidney) at 30 min of post-injection, and then decreased to around 2.1% at 120 min, which was consistent with the metabolic trend observed in in vivo fluorescence imaging. Furthermore, the high-resolution bio-TEM-EDX elemental mapping images of Mo for renal cortex sections of AKI mice treated with PBS and MF-0 were compared, which demonstrated the obvious presence of MF-0 in glomerular basement membrane (GBM) of AKI mice, although it has not been clearly observed in renal tubules at this stage (30 min of post-injection) (Supplementary Fig. 26). In the meantime, the content of Mo accumulated in the urine of AKI mice treated with MF-0 was significantly higher than that in the control group (AKI mice treated with PBS), and the amount of Mo in urine gradually increased with metabolic time (Supplementary Fig. 27), indicating that MF-0 could be further excreted into urine after passing through kidneys. Besides, the ICP-MS quantification data on the biodistribution of MF-0 in major organs (heart, liver, spleen, lung, kidney and intestine) of AKI mice at different time points of post-injection (0, 0.5, 2, 6, 12, 24, 72, 168 and 336 h) showed that MF-0 was mainly accumulated in the liver, spleen and lung of AKI mice, followed by the kidney, and reached its maximum accumulation in different organs at 2 h (liver, spleen and lung) of post-injection, and then was gradually metabolized out of the body within 14 days (Supplementary Fig. 28). These suggested that the nanoprobes were still retained at most in the reticuloendothelial system (RES) organs including liver, spleen and lung, a certain dose of MF-0 could be targeted to the kidneys of AKI mice via renal metabolic pathways, providing a favorable prerequisite for subsequent treatment design.

Timeline of mouse AKI modeling and treatment was presented in Fig. 5a, the AKI model was established via intramuscular injection of glycerin into both hind legs of dehydrated mice, and the mice subsequently showed symptoms such as oliguria, slow movement and hematuria. After 24 h all the mice were sacrificed, and kidney function test, H&E staining as well as dihydroethidium (DHE) staining were conducted for the therapeutic assessments. As shown in Fig. 5b, c, two crucial renal function indicators of AKI mice, namely, creatinine (CRE) and blood urea nitrogen (BUN) became significantly higher than those of healthy mice, indicating an apparent abnormality in renal function. In contrast, the AKI mice treated with NAC (a commonly used antioxidant) or MF-0 exhibited down-regulation of CRE and BUN, and the MF-0 treated groups recovered to a level comparable to that of healthy mice. Moreover, the survival rate of AKI mice treated with MF-0 was 100% during the treatment, which was obviously better than that of PBS treated group (Fig. 5d), and their body weight also maintained an increasing trend within 10 days (Supplementary Fig. 29). Afterwards, H&E staining of kidneys (Fig. 5e) provided intuitive evidence that in AKI mice, a large number of damaged renal tubules (marked with arrows) and casts (marked with asterisks) formed by precipitation of denatured proteins in the tubules, while in the NAC treatment group, the kidney injury was alleviated to a certain extent as evidenced by the reduction in the number of casts (a marker of more severe tubular damage), but there were still damaged renal tubules. More encouragingly, AKI mice treated with MF-0 recovered as normal, almost no such tissue damages were found, suggesting the superior therapeutic effect of MF-0 on AKI. In addition, DHE staining of renal tissues further evaluated the ROS level (showing red fluorescence) of each treatment group (Fig. 5f). As compared to the PBS and NAC treated groups, ROS level of MF-0 treated AKI mice was effectively inhibited to that of healthy mice, which revealed the desired antioxidant activity and targeting effect of MF-0 for AKI treatment. In other words, MF-0 could effectively eliminate excess ROS in the kidneys of AKI mice, thereby avoiding renal tubule damages and achieving prominent AKI therapeutic efficacy.

Fig. 5: Schematic diagram of AKI modeling and treatment process, and therapeutic assessments of MF-0 via intravenous injection for AKI. — **Fig. 5: Schematic diagram of AKI modeling and treatment process, and therapeutic assessments of MF-0 *via* intravenous injection for AKI.**

In vivo ROS-responsive PAI for post-cure assessment of AKI

In addition to the effective treatment of AKI, MF-0 was also endowed with ROS-responsive photoacoustic imaging (PAI) features for non-invasive and real-time post-cure assessment of AKI. It is well known that the PA signal of a material is highly dependent on its absorption, which was obviously reflected in the color change of MF-0 solution from dark grey to pale yellow after incubation with H₂O₂ (Fig. 6a). UV-vis-NIR absorption spectra presented a noticeable decreasing absorption of MF-0 solution along with the increment of H₂O₂ from 0 to 1 mM (Fig. 6b). Meanwhile, the absorption intensity ratio (A/A₀) of MF-0 (at 730 nm) incubated with 1 mM of H₂O₂ declined quickly in 5 min and reached the plateau in 10 min (Fig. 6c). Accordingly, PA intensity of MF-0 solution (at 730 nm) was observed negatively proportional to the concentration of H₂O₂ (Fig. 6d). Encouraged by the rapid and sensitive H₂O₂-responsive performance of MF-0, the in vivo PAI was further carried out to intuitively evaluate the therapeutic effect on AKI mice. As illustrated in Fig. 6e, PAI was performed at different post-injection time points for AKI mice and AKI-cured mice. Distinguished from healthy mice, ROS level in kidneys of AKI mice was much higher, causing a diminished PAI signal (“off” state). After treatment with MF-0, the excess ROS in kidneys were scavenged, meanwhile, the absorption of MF-0 was tuned by ROS consumption, thus switching on the PAI signal (turn “on”) in AKI-cured mice. As depicted in Fig. 6f, g, the evolution trend of PAI signal in kidneys among the three groups during the post-injection period of 120 min was analogous. PAI signal appeared in 10 min after injection of MF-0, then reached the maximum in 30 min, and attenuated to the level as the beginning of MF-0 injection in 120 min, indicating the rapid accumulation and metabolism of MF-0 in kidneys. As expected, the PAI intensity difference in AKI-cured mice at 30 min was ~2.1-fold stronger than that in AKI mice, and almost the same level as in healthy mice, successfully achieving ROS-responsive PAI for AKI therapeutic effect and post-cure assessments in vivo.

**Fig. 6: ROS-responsive photoacoustic imaging (PAI) for post-cure assessment of AKI mice.**

Absolutely, the biosafety assessment of MF-0 was also very worthy of attention. The hemolysis assay demonstrated that MF-0 did not cause hemolysis even at a concentration of 400 μg/mL (Supplementary Fig. 30). More importantly, the biological toxicity studies of the mice revealed that no obvious difference was found in liver function indicators and hematological parameters among the groups: healthy mice treated with PBS, AKI mice treated with PBS and AKI mice treated with MF-0. (Supplementary Fig. 31). In addition, it was evidenced by H&E staining (Supplementary Fig. 32) that MF-0 treated AKI mice had no significant damage such as necrosis, congestion, and hemorrhage in major organs (heart, liver, spleen, lung and intestine), further suggesting the favorable biosafety of MF-0.

Discussion

In summary, we developed a mild oxidation valence-engineering strategy to tune the selectivity of molybdenum oxide nanozyme. The results demonstrated that the CAT-like activity decreased and the POD-like activity enhanced along with the average valence of Mo rising from 4.64 to 5.68 in MF-x. MF-0 (Mo average valence 4.64) and MF-10 (Mo average valence 5.68) exhibited exclusively optimal CAT- or POD-like activity, respectively, indicating the high specificity toward different reaction pathways involving H₂O₂. DFT calculations also provided theoretical evidence for the point that the valence state of Mo played a leading role in the catalytic reaction selectivity, that is, the electronic structure of MF-x and the free energy barriers of the RDS from H₂O₂ to O₂ or ·OH on MF-x nanozymes were well tuned by the mild oxidation valence-engineering strategy. Both in vitro and in vivo experimental results demonstrated that MF-0 could effectively scavenge ROS such as H₂O₂ and ·OH, protecting cells from oxidative stress damage, thereby accomplishing highly efficient treatment of AKI. Meanwhile, MF-0 also presented reliable ROS-responsive photoacoustic imaging ability to achieve intuitively post-cure assessment of AKI in vivo, providing an advantageous deep tissue imaging tool for ROS-related diseases that lack effective non-invasive in vivo diagnostic and evaluation approaches. This work offers guidance for the rational design of redox nanozymes with high catalysis selectivity and H₂O₂ elimination, as well as new insights into the development of integrated nanoprobes for the treatment and post-cure imaging assessment of ROS-related diseases.

Methods

Chemicals and reagents

Oleylamine (OAm) was bought from Acros Organics (Belgium). Oleic acid (OA) and 11-Chloro-1,1’-di-n-propyl-3,3,3’,3’-tetramethyl-10,12-trimethyleneindatricarbocyanine iodide (IR780) was bought from Alfa Aesar (Shanghai). Ethanol, n-hexane, hydrogen peroxide (H₂O₂), dimethylsulfoxide (DMSO) and sodium citrate were supplied by Beijing Chemical Reagent Company Ltd. Sodium molybdate and 5,5’-dimethyl-1-pyrrolin-N-oxide (DMPO) were purchased from Macklin (Shanghai). Hydrochloric acid (HCl) was obtained from Tianjin Fuyu Fine Chemical Co., Ltd. Dichloromethane (CH₂Cl₂) and glycerin were obtained from Damao Chemical Reagent Factory (Tianjin). Pluronic F127 was from Sigma (Shanghai). Tris-hydrochloride (Tris-HCl) and Dulbecco’s modified Eagle medium (DMEM) were supplied by Beijing Kebio Biotechnology Co., Ltd. 3,3,5,5-tetramethylbenzidine (TMB), methyl thiazolyl tetrazolium (MTT) and 2’,7’-dichlorofluorescin diacetate (DCFH-DA) were bought from Beijing InnoChem Science & Technology Co., Ltd. N-acetyl cysteine (NAC) was purchased from Beyotime (Shanghai). Phosphate buffer solution (PBS) was from M&C Gene Technology (Beijing). Potassium titanium oxalate (PTO) was obtained from Shanghai Titan Scientific Co., Ltd. Methylene blue (MB) was acquired from Tianjin Guangfu Fine Chemical Research Institute. FeCl₂·4H₂O was acquired from Xiya Reagent (Shandong). Diammonium 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) was bought from Shanghai Yuanye Bio-Technology Co., Ltd. Potassium persulfate was supplied by Fuchen Chemical Reagent Co., Ltd. (Tianjin). Paraformaldehyde was obtained from Tianjin Chemical Reagent Research Institute. All of the abovementioned chemicals were of analytical grade and used as received without further purification. Ultrapure water used throughout all experiments was obtained from a Millipore Milli-Q purification system (U.S.).

Characterization

Transmission electron microscope (TEM) images were obtained using a JEM-1200EX (JEOL) transmission electron microscope at a voltage of 100 kV, and the data of size distribution was analyzed by Nano Measurer 1.2. High-angle annular dark-field scanning transmission election microscope (HAADF-STEM) images and energy-dispersive X-ray (EDX) elemental mapping were collected on a JEM-ARM200F (JEOL) spherical aberration-corrected transmission electron microscope at a voltage of 200 kV. The lattice spacing was measured by Digital Micrograph 3.7. Bio-TEM-EDX elemental mapping images were obtained using a HRTEM-JEM-F200 (JEOL) high-resolution field emission transmission electron microscope at a voltage of 100 kV. Fourier transform infrared (FT-IR) spectra were obtained on a Nexus 8700 Fourier transform infrared spectrophotometer (Nicolet). Thermogravimetric (TG) analysis was performed on a Mettler Toledo TGA/DSC3+ thermogravimeter. X-ray photoelectron spectrum (XPS) was carried out on a K-Alpha spectrometer (Thermo Fisher Scientific), and the data was analyzed by XPSPEAK 4.1. UV-vis-NIR absorption spectra were recorded on a UV-3600 spectrophotometer (Shimadzu). Oxygen production ability was measured using a F4-Standard Kit dissolved oxygen meter (Mettler Toledo). Electron spin resonance (ESR) spectra were acquired on an EMX-500 spectrometer (Bruker). Cell viability was detected using a Tecan Infinite F50 plate reader (Switzerland). Photographs of cell imaging and kidney tissue section staining were obtained using a model eclipse Ti2-U inverted fluorescence microscope (Nikon). The X-ray diffraction (XRD) patterns were acquired on a Bruker AXS D8-Advanced X-ray diffractometer with Cu Kα radiation (λ = 1.5418 Å), and the data was analyzed by MDI Jade 6. Dynamic light scattering (DLS) size distribution and zeta potential were carried out using a Zetasizer Nano-ZS90 zeta potential and size analyzer (Malvern). The content of Mo accumulated in the major organs (heart, liver, spleen, lung, kidney and intestine) and the urine of AKI mice was detected by an inductively coupled plasma-mass spectrometer (ICP-MS, Thermo Scientific iCAP 6000 series). Fluorescence imaging was performed on a IVIS Spectrum (PerkinElmer). Photoacoustic imaging (PAI) was performed on a multispectral optoacoustic tomography scanner (MSOT, iThera Medical 256-tf).

Synthesis of MoO_3-x nanoparticles (NPs)

MoO_3-x NPs were prepared through a hydrothermal method. Specifically, oleylamine (2 mL) and oleic acid (1 mL) were dissolved in a mixed solvent containing 6 mL of n-hexane and 4 mL of ethanol. Then, 2 mL of sodium molybdate aqueous solution (50 mg/mL) and 1.2 mL of hydrochloric acid (6 M) were added under vigorous stirring and keeping for 30 min. Subsequently, the mixture was transferred into a Teflon-lined autoclave (20 mL) and heated at 200 °C for 6 h. After cooling down to room temperature, the product was washed with n-hexane and ethanol and collected by centrifugation (8497 g, 10 min). Finally, the obtained MoO_3-x NPs were redispersed in dichloromethane for future use.

Preparation of hydrophilic MF-0, MF-2, MF-3.5, MF-5 and MF-10

MoO_3-x NPs (1 mg) dispersed in 200 μL of CH₂Cl₂ were added into 10 mL of ultrapure water containing amphiphilic Pluronic F127 (50 mg). The mixture was treated by ultrasonication (400 W) in ice bath for 8 min (3 s/3 s, on/off), causing the evaporation of CH₂Cl₂ and forming a uniform clear light grey solution. Then, the above solution was stirred vigorously at 37 °C in the open container for 0, 2, 3.5, 5 and 10 h respectively, followed by rotary evaporation to obtain the stock solution (1 mL) with various degrees of oxidation. The corresponding samples were named as MF-x, namely, MF-0, MF-2, MF-3.5, MF-5 and MF-10, respectively (mass concentration: 40 mg/mL; concentration of molybdenum (Mo): 400 μg/mL).

Catalase (CAT)-like activity of MF-x

CAT-like activity of the samples (MF-x) was indicated by the oxygen generation ability in the presence of H₂O₂. H₂O₂ solution (30%, 200 μL) was dispersed into 1.8 mL of Tris-HCl buffer (pH 6.5 or pH 7.4), and then the MF-x samples (200 μL) were added, respectively. The produced O₂ was measured at different time points.

Peroxidase (POD)-like activity of MF-x

TMB (3,3,5,5-tetramethylbenzidine) oxidation assay was carried out to evaluate the POD-like activity of samples, as the oxidized TMB (oxTMB) showing maximum absorption at 650 nm. 50 μL of TMB ethanolic solution (4 mg/mL) was added into 1.78 mL of Tris-HCl buffer (pH 6.5 or pH 7.4), and then treated with 50 μL of H₂O₂ solution (4 mM) and each of MF-x samples (120 μL), respectively. The absorption spectra of oxTMB solutions at different time points were recorded.

In addition, the type of produced ROS at pH 6.5 was further explored by ESR analysis. Briefly, 85 μL of Tris-HCl buffer (pH 6.5) was mixed with 10 μL of DMPO (a spin trapping agent, 5 M), 5 μL of H₂O₂ solution (4 mM) and 10 μL of each MF-x sample, respectively. After incubation for several minutes, ESR measurements were performed and the ROS was determined to be ·OH.

Hydrogen peroxide (H₂O₂) scavenging ability of MF-0

Since titanium ion can react with H₂O₂ to produce a stable orange-yellow complex with a maximum absorption at 381 nm, potassium titanium oxalate (PTO) probe was served to assess the H₂O₂ scavenging capacity of MF-0. MF-0 and H₂O₂ solution (20 mM) were mixed in ultrapure water and reacted for 10 min, then PTO (10 mM) was added, and the final concentrations of H₂O₂ and PTO were 0.5 mM and 1.0 mM. The absorbance of H₂O₂ solution with PTO (A_P) and a mixed solution of PTO, H₂O₂ and MF-0 (4, 10, 20 and 30 μg of Mo per mL) (A_Q) was measured at 381 nm. Due to a certain of absorption at 381 nm for MF-0, the absorbance was deducted to avoid interference. H₂O₂ inhibition rate (%) = [(A_P - A_Q)/A_P] × 100%.

Hydroxyl radical (·OH) scavenging ability of MF-0

Methylene blue (MB) can be bleached by ·OH, thus the absorbance of remaining MB was used to evaluate ·OH scavenging capacity of MF-0. First, 600 μL of FeCl₂·4H₂O (5 mM) mixed with 400 μL of H₂O₂ (5 mM) was prepared as ·OH working solution based on the Fenton reaction. After 5 min, MF-0 and MB (1 mM) was added successively and kept for another 50 min. The final concentrations of MB, FeCl₂·4H₂O and H₂O₂ were 15 μM, 1.5 mM and 1.0 mM, respectively. The absorbance of pure MB solution (A_M), MB solution with ·OH (A_N) and a mixed solution of MB, ·OH and MF-0 (1, 3, 5 and 10 μg of Mo per mL) (A_F) was measured at 664 nm. ·OH inhibition rate (%) = [(A_F - A_N)/(A_M - A_N)] × 100%.

ABTS radical (ABTS·) scavenging ability of MF-0

Diammonium 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) decolorization assay was carried out to assess ABTS radical scavenging capacity of MF-0. ABTS aqueous solution (7 mM) and potassium persulfate solution (2.45 mM) were mixed in a 1:1 volume ratio and stored in the dark for 12 h to generate ABTS radical. Subsequently, the absorbance of pure ABTS· solution (A_B) and ABTS· solution mixed with MF-0 (1, 3, 5 and 10 μg of Mo per mL) for 20 min (A_E) was measured at 730 nm. ABTS· inhibition rate (%) = [(A_B - A_E)/A_B] × 100%.

Superoxide anion (·O₂ ^-) scavenging ability investigation of MF-0

Under weakly basic conditions, pyrogallol can undergo autoxidation to produce ·O₂^- and colored intermediates whose absorption intensity is proportional to the amount of ·O₂^-. Therefore, pyrogallol was used to verify the ·O₂^- scavenging activity of MF-0. MF-0 or ascorbic acid (AA, an antioxidant as the positive control) was dispersed in Tris-HCl buffer (pH 8.0), then pyrogallol (final concentration 250 μM) was added and incubated for 20 min. The absorbance of pure pyrogallol solution (A_Z), a mixed solution of pyrogallol and MF-0 (A_X), as well as a mixed solution of pyrogallol and AA (A_Y) was measured at 319 nm. Due to a certain of absorption at 319 nm for MF-0, the absorbance was deducted to avoid interference. ·O₂^- inhibition rate (%) of MF-0 = [(A_Z - A_X)/A_Z] × 100%, ·O₂^- inhibition rate (%) of AA = [(A_Z - A_Y)/A_Z] × 100%.

Cytotoxicity test

Mouse kidney epithelial cells (MREpiC) were obtained from Hunan Fenghui Biotechnology Co., Ltd. MTT assay was implemented to assess the cytotoxicity of MF-0. MREpiC cells were seeded in a 96-well plate at 2 × 10⁴ cells per well and incubated for 24 h (37 °C, 5% CO₂). Subsequently, MF-0 dispersed in PBS (final concentration: 0, 50, 100, 200 and 400 μg/mL) were added into each well and cultured for 24 h. Then, 20 μL of MTT (5 mg/mL) was introduced and incubated for an additional 4 h. Finally, the absorbance of colored formazan produced from MTT and dissolved by DMSO was measured at 492 nm using a microplate reader.

Intracellular ROS scavenging by MF-0

MREpiC cells were seeded in a 96-well plate followed by culturing with MF-0 (final concentration: 0, 20, 50 and 80 μg/mL) in the presence of H₂O₂ (250 μM) for 24 h. The cell viability was measured by MTT assay to display the ROS scavenging efficiency.

In addition, as a ROS fluorescent imaging probe, DCFH-DA was used to intuitively explore the intracellular scavenging effect of MF-0. MREpiC cells were seed in a 6-well plate at 3 × 10⁵ cells per well and cultured for 24 h (37 °C, 5% CO₂). Afterwards, MF-0 dispersed in PBS (final concentration: 0, 20, 50 and 80 μg/mL) were added into each well followed by the addition of H₂O₂ (100 μM), and then cultured for 4 h. After washing by PBS, the cells were further incubated with DCFH-DA (10 μM) dispersed in DMEM for 30 min. Then, the cells were washed again and observed through a fluorescence microscope (λ_ex = 480 nm, λ_em = 525 nm). The quantification analysis of DCF fluorescence intensity was from ImageJ 1.52.

Acute kidney injury (AKI) model

All animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of the China-Japan Friendship Hospital (Beijing). Six weeks old of female BALB/c mice were provided for the animal experiments. The findings apply to both male and female animals. All mice were housed in a 12/12 h dark/light cycle with the standard conditions: Temperature, 20–25 °C; Relative humidity, 40–70%.

The mice were deprived of water but able to have food for 15 h, then 50% glycerin (8 mL/kg) was intramuscularly (i.m.) injected into both hind legs equally, followed by the provision of water and food as normal. Gradually, the mice appeared symptoms of AKI, such as reduced urine, slow movement and hematuria.

Biodistribution of MF-0 in AKI mice

For the sake of in vivo fluorescence (FL) imaging, MF-0 was labeled with a fluorescent dye IR780 to produce IR780@MF-0. Specifically, MoO_3-x NPs (1 mg) and IR780 (200 μg) were dispersed in 200 μL of CH₂Cl₂, and then added into 10 mL of ultrapure water containing Pluronic F127 (50 mg). The mixture was treated with ultrasonication for 8 min (3 s/3 s, on/off) to acquire a clear grey-green solution. AKI mice (back hair removed) were intravenously injected with 150 μL of IR780@MF-0 (600 μg of Mo per mL) to perform FL imaging at 10, 30, 60 and 120 min (λ_ex = 745 nm, λ_em = 820 nm).

Meanwhile, the amount of MF-0 accumulated in the various organs (heart, liver, spleen, lung, kidney and intestine) of AKI mice was evaluated. Two hours after the establishment of AKI model, 200 μL of MF-0 solution was intravenously injected into the mice. The mice were sacrificed and the organs were harvested at different post-injection time points (0, 0.5, 2, 6, 12, 24, 72, 168 and 336 h). After grinding, aqua regia soaking digestion, centrifugation (765 g, 6 min), filtration and dilution, the content of Mo was detected by ICP-MS.

In addition, the biodistribution of MF-0 nanoparticles in the kidney of AKI mice was further investigated by bio-TEM-EDX elemental mapping. After intravenous injection of MF-0 (200 μL) for 30 min, the mice were sacrificed and small slices of renal cortex were collected after the perfusion of the heart of mouse with 5 mL of normal saline and 5 mL of 4% paraformaldehyde solution consecutively. Then the renal cortex sections were fixed with 2.5% glutaraldehyde for 24 h, post-fixed with 1% osmium tetroxide and stained with lead citrate and uranium acetate for bio-TEM-EDX analysis.

In vivo AKI treatment by MF-0

Two hours after the establishment of AKI model (glycerol injection), therapeutic agents (200 μL) were intravenously (i.v.) injected into the mice of various treatment groups: (1) healthy mice treated with PBS; (2) healthy mice treated with MF-0 (800 μg of Mo per mL); (3) AKI mice treated with PBS; (4) AKI mice treated with the antioxidant NAC in PBS (800 μg/mL); (5) AKI mice treated with MF-0 (800 μg of Mo per mL) to assess the therapeutic effect of MF-0 in AKI mice. All the mice were sacrificed after 24 h of AKI model implementing. In order to evaluate the treatment effect of MF-0 for AKI, blood samples were centrifuged at 2000 rpm for 5 min to collect plasma for renal function test (creatinine and blood urea nitrogen). Meanwhile, a portion of kidneys were harvested and stored in 4% paraformaldehyde solution for hematoxylin and eosin (H&E) staining, and the remaining parts were frozen rapidly by liquid nitrogen and set at –80 °C for subsequent ROS staining experiments. The ROS staining images were processed by ImageJ 1.52.

In vivo photoacoustic imaging (PAI) of MF-0 for AKI post-cure assessment

The mice used for in vivo PAI were divided into three groups: (1) healthy mice; (2) AKI mice; and (3) AKI-cured mice. The back hair of the mice was removed to acquire clear imaging effects. 200 μL of MF-0 solution (800 μg of Mo per mL) were intravenously (i.v.) injected into the mice of different groups, and PA imaging were performed at 10, 30, 60 and 120 min after the injection of MF-0. For AKI mice, the MF-0 was injected 2 h after glycerol injection; and for AKI-cured mice, the MF-0 was given a second injection 24 h after glycerol injection. The corresponding data were obtained from viewMSOT 4.0.

Hemolysis assay

100 μL of blood was taken from the mice and dispersed in 5% sodium citrate solution, forming red blood cells (RBCs) suspension. Then 400 μL of MF-0 dispersed in PBS (50, 100, 200 and 400 μg/mL) were mixed with 100 μL of RBCs suspension. In the meantime, H₂O and PBS were served as positive and negative control groups, respectively. After incubation at 37 °C for 2 h, the supernatants from all groups were collected by centrifugation (543 g, 8 min). The obtained supernatants were transferred to a 96-well plate, and then the absorbance at 414 nm was measured. Hemolytic degree (%) = [(A_sample -A_negative)/(A_positive - A_negative)] × 100%.

Calculation details

All calculations were performed by using the projector augmented wave method in the framework of density functional theory (DFT), as implemented in the Vienna Ab initio Simulation Package (VASP 5.4.4)⁴⁹. The electron exchange-correlation interactions were parameterized by the generalized gradient approximation (GGA) with the Perdew-Burke-Ernzerhof (PBE)^50,51. Spin polarization effect was considered in this work. The Brillouin zone was sampled by the Γ-centered k-mesh with a resolution of 2π × 0.04 Å⁻¹ for geometry optimization and 2π × 0.02 Å⁻¹ for electronic structure calculation. The plane-wave energy cutoff was set to 500 eV, and the convergence tolerance for residual force and energy on each atom during structure relaxation were set to 0.02 eV/Å and 10^-5eV, respectively. To account for strongly correlated interactions in transition metal oxides, an additional Hubbard U term with U_eff values of 6.3 for Mo in MoO₃ and 3.0 for Mo in MoO₂ was applied^52,53,54,55.

A vacuum space of 15 Å was applied to avoid interactions between the neighboring configurations. For vdW correction, DFT-D3 method with Becke–Jonson damping was employed^56,57. The solvent effects were simulated using an implicit solvent model for the plane-wave DFT code VASP (VASPsol)⁵⁸. The solvent dielectric constant was set at 78.4 to simulate the solvent effects of aqueous environment⁵⁹.

Gibbs free energies were calculated from DFT total energies corrected by zero-point energy (ZPE), heat capacity (C_p), and entropy (TS), according to the expression^60,61,62:

$$G={E}_{{{{\rm{DFT}}}}}+{E}_{{{{\rm{ZPE}}}}}+\int {C}_{{{{\rm{p}}}}}{{{\rm{d}}}}T-{TS}$$

where E, ZPE, C_p, T, and S represent the DFT calculated energy, zero-point energy, heat capacity, temperature (298.15 K), and vibrational entropy, respectively. The thermodynamic correction was carried out by using the VASPKIT code. Structural and charge density visualization were conducted utilizing the VESTA software⁶³.

Surface models were built based on the different oxidation states of Mo compounds MoO₂ and MoO₃, respectively. Due to their lowest surface energies, the (100) surface of MoO₂ and the (010) surface of MoO₃ were selected for creating surface adsorption models with oxygen vacancies by removing one oxygen atom from the (100) surface of MoO₂ and the (010) surface of MoO₃, and the metal centers near the oxygen vacancies were used as the primary active site for subsequent calculations^64,65,66. Additionally, two different terminations of the MoO₃ (010) crystal surface were further calculated, and the configuration with the lowest surface energy was chosen (Supplementary Fig. 13). In our calculations, the atoms in the bottom were fixed in their bulk positions, and those in the top and second layers were allowed to relax (Supplementary Fig. 14). The optimized models for MF-0 and MF-10 in this work correspond to MoO₂ and MoO₃ mentioned above, respectively. The corresponding data were processed by VESTA 3.4.

Statistical analysis

The data were analyzed using Origin software (version 2022). All experiments were repeated at least 3 times and presented as means ± SD. Significance was calculated by one-sided Student’s t-test.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The data supporting the findings of the study are available within the paper and Supplementary Files (Supplementary Information and Supplementary Data 1), and from the corresponding author(s) upon request. Source data are provided with this paper.

References

Ji, S. et al. Matching the kinetics of natural enzymes with a single-atom iron nanozyme. Nat. Catal. 4, 407–417 (2021).
Article CAS Google Scholar
Huang, Y., Ren, J. & Qu, X. Nanozymes: classification, catalytic mechanisms, activity regulation, and applications. Chem. Rev. 119, 4357–4412 (2019).
Article CAS PubMed Google Scholar
Zhou, Q. et al. Cascaded nanozyme system with high reaction selectivity by substrate screening and channeling in a microfluidic device. Angew. Chem. Int. Ed. 61, e202112453 (2021).
Article Google Scholar
Chen, Y. et al. Atomic-level regulation of cobalt single-atom nanozymes: engineering high-efficiency catalase mimics. Angew. Chem. Int. Ed. 62, e202301879 (2023).
Article ADS CAS Google Scholar
Liu, C. et al. Intrinsic strain-mediated ultrathin ceria nanoantioxidant. J. Am. Chem. Soc. 145, 19086–19097 (2023).
Article CAS PubMed Google Scholar
Fan, H. et al. Surface ligand engineering ruthenium nanozyme superior to horseradish peroxidase for enhanced immunoassay. Adv. Mater. 36, 2300387 (2023).
Article Google Scholar
Chen, D. et al. Bioinspired porous three-coordinated single-atom Fe nanozyme with oxidase-like activity for tumor visual identification via glutathione. Nat. Commun. 14, 7127 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhang, L., Wang, H. & Qu, X. Biosystem-inspired engineering of nanozymes for biomedical applications. Adv. Mater. 36, 2211147 (2023).
Article Google Scholar
Zhu, Y. et al. Stimuli-responsive manganese single-atom nanozyme for tumor therapy via integrated cascade reactions. Angew. Chem. Int. Ed. 60, 9480–9488 (2021).
Article CAS Google Scholar
Jiang, W. et al. Chiral metal-organic frameworks incorporating nanozymes as neuroinflammation inhibitors for managing Parkinson’s disease. Nat. Commun. 14, 8137 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Hu, Y. et al. Nitrogen-doped carbon nanomaterials as highly active and specific peroxidase mimics. Chem. Mater. 30, 6431–6439 (2018).
Article ADS CAS Google Scholar
Zhang, S. et al. Modulation of the biocatalytic activity and selectivity of CeO₂ nanozymes via atomic doping engineering. Nanoscale 15, 4408–4419 (2023).
Article CAS PubMed Google Scholar
Wang, Y. et al. Coordination number regulation of molybdenum single-atom nanozyme peroxidase-like specificity. Chem. 7, 436–449 (2021).
Article CAS Google Scholar
Yuan, B. et al. Regulating the H₂O₂ activation pathway on a well-defined CeO₂ nanozyme allows the entire steering of its specificity between associated enzymatic reactions. ACS Nano 17, 17383–17393 (2023).
Article CAS PubMed Google Scholar
Ge, C. et al. Facet energy versus enzyme-like activities: the unexpected protection of palladium nanocrystals against oxidative damage. ACS Nano 10, 10436–10445 (2016).
Article CAS PubMed Google Scholar
Fang, G. et al. Differential Pd-nanocrystal facets demonstrate distinct antibacterial activity against Gram-positive and Gram-negative bacteria. Nat. Commun. 9, 129 (2018).
Article ADS PubMed PubMed Central Google Scholar
Wang, H. et al. Reducing valence states of Co active sites in a single-atom nanozyme for boosted tumor therapy. Adv. Funct. Mater. 32, 2200331 (2022).
Article CAS Google Scholar
Wang, Q. et al. A valence-engineered self-cascading antioxidant nanozyme for the therapy of inflammatory bowel disease. Angew. Chem. Int. Ed. 61, e202201101 (2022).
Article ADS CAS Google Scholar
Xia, F. et al. Ultrasmall ruthenium nanoparticles with boosted antioxidant activity upregulate regulatory T cells for highly efficient liver injury therapy. Small 18, 2201558 (2022).
Article CAS Google Scholar
Cao, M. et al. Molybdenum derived from nanomaterials incorporates into molybdenum enzymes and affects their activities in vivo. Nat. Nanotechnol. 16, 708–716 (2021).
Article ADS CAS PubMed Google Scholar
Chen, Y., Zhang, G., Ji, Q., Liu, H. & Qu, J. Triggering of low-valence molybdenum in multiphasic MoS₂ for effective reactive oxygen species output in catalytic Fenton-like reactions. ACS Appl. Mater. Interfaces 11, 26781–26788 (2019).
Article CAS PubMed Google Scholar
Zu, Y. et al. The age of bioinspired molybdenum-involved nanozymes: synthesis, catalytic mechanisms, and biomedical applications. View 2, 20200188 (2021).
Article CAS Google Scholar
Yu, B., Wang, W., Sun, W., Jiang, C. & Lu, L. Defect engineering enables synergistic action of enzyme-mimicking active centers for high-efficiency tumor therapy. J. Am. Chem. Soc. 143, 8855–8865 (2021).
Article CAS PubMed Google Scholar
Zhou, Z. et al. Intercalation-activated layered MoO₃ nanobelts as biodegradable nanozymes for tumor-specific photo-enhanced catalytic therapy. Angew. Chem. Int. Ed. 61, e202115939 (2022).
Article ADS CAS Google Scholar
Zhou, J. et al. Coordination-driven self-assembly strategy-activated Cu single-atom nanozymes for catalytic tumor-specific therapy. J. Am. Chem. Soc. 145, 4279–4293 (2023).
Article CAS Google Scholar
Bao, T. et al. One-pot synthesis of PEGylated plasmonic MoO_3-X hollow nanospheres for photoacoustic imaging guided chemo-photothermal combinational therapy of cancer. Biomaterials 76, 11–24 (2016).
Article CAS PubMed Google Scholar
Song, G. et al. Hydrophilic molybdenum oxide nanomaterials with controlled morphology and strong plasmonic absorption for photothermal ablation of cancer cells. ACS Appl. Mater. Interfaces 6, 3915–3922 (2014).
Article CAS PubMed Google Scholar
Wang, S. et al. A unique multifunctional nanoenzyme tailored for triggering tumor microenvironment activated NIR-II photoacoustic imaging and chemodynamic/photothermal combined therapy. Adv. Healthc. Mater. 11, 2102073 (2022).
Article CAS Google Scholar
Wang, Y. et al. Atomic-level nanorings (A-NRs) therapeutic agent for photoacoustic imaging and photothermal/photodynamic therapy of cancer. J. Am. Chem. Soc. 142, 1735–1739 (2020).
Article CAS PubMed Google Scholar
Kellum, J. A. & Prowle, J. R. Paradigms of acute kidney injury in the intensive care setting. Nat. Rev. Nephrol. 14, 217–230 (2018).
Article PubMed Google Scholar
Chawla, L. S., Eggers, P. W., Star, R. A. & Kimmel, P. L. Acute kidney injury and chronic kidney disease as interconnected syndromes. N. Engl. J. Med. 371, 58–66 (2014).
Article PubMed PubMed Central Google Scholar
Guo, C., Dong, G., Liang, X. & Dong, Z. Epigenetic regulation in AKI and kidney repair: mechanisms and therapeutic implications. Nat. Rev. Nephrol. 15, 220–239 (2019).
Article PubMed PubMed Central Google Scholar
Liu, T. et al. Ultrasmall copper-based nanoparticles for reactive oxygen species scavenging and alleviation of inflammation related diseases. Nat. Commun. 11, 2788 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Gu, J. et al. Cerium-luteolin nanocomplexes in managing inflammation-related diseases by antioxidant and immunoregulation. ACS Nano 18, 6229–6242 (2024).
Article CAS PubMed Google Scholar
Tepel, M. et al. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N. Engl. J. Med. 343, 180–184 (2000).
Article CAS PubMed Google Scholar
Liu, J. et al. Ruthenium-based metal-organic framework with reactive oxygen and nitrogen species scavenging activities for alleviating inflammation diseases. Nano Today 47, 101627 (2022).
Article CAS Google Scholar
Lin, A. et al. Self-cascade uricase/catalase mimics alleviate acute gout. Nano Lett. 22, 508–516 (2022).
Article ADS CAS PubMed Google Scholar
Weng, Q. et al. Catalytic activity tunable ceria nanoparticles prevent chemotherapy-induced acute kidney injury without interference with chemotherapeutics. Nat. Commun. 12, 1436 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Yuan, A. et al. Ceria nanozyme-integrated microneedles reshape the perifollicular microenvironment for androgenetic alopecia treatment. ACS Nano 15, 13759–13769 (2021).
Article CAS PubMed Google Scholar
Tian, R. et al. Se-containing MOF coated dual-Fe-atom nanozymes with multi-enzyme cascade activities protect against cerebral ischemic reperfusion injury. Adv. Funct. Mater. 32, 2204025 (2022).
Article CAS Google Scholar
Zhang, R. et al. Edge-site engineering of defective Fe–N₄ nanozymes with boosted catalase-like performance for retinal vasculopathies. Adv. Mater. 34, 2205324 (2022).
Article CAS Google Scholar
Cheng, P. et al. Fluoro-photoacoustic polymeric renal reporter for real-time dual imaging of acute kidney injury. Adv. Mater. 32, 1908530 (2020).
Article CAS Google Scholar
Chen, Y. et al. A promising NIR-II fluorescent sensor for peptide-mediated long-term monitoring of kidney dysfunction. Angew. Chem. Int. Ed. 60, 15809–15815 (2021).
Article ADS CAS Google Scholar
Xu, Y. et al. NIR-II photoacoustic-active DNA origami nanoantenna for early diagnosis and smart therapy of acute kidney injury. J. Am. Chem. Soc. 144, 23522–23533 (2022).
Article CAS PubMed Google Scholar
Ma, Y. et al. Rational design of a double-locked photoacoustic probe for precise in vivo imaging of cathepsin B in atherosclerotic plaques. J. Am. Chem. Soc. 145, 17881–17891 (2023).
Article CAS PubMed Google Scholar
Li, R. et al. Zwitterionic sulfhydryl sulfobetaine stabilized platinum nanoparticles for enhanced dopamine detection and antitumor ability. ACS Appl. Mater. Interfaces 14, 55201–55216 (2022).
Article CAS PubMed Google Scholar
Luo, W. et al. Self-catalyzed, self-limiting growth of glucose oxidase-mimicking gold nanoparticles. ACS Nano 4, 7451–7458 (2010).
Article CAS PubMed Google Scholar
Gao, L. et al. Intrinsic peroxidase-like activity of ferromagnetic nanoparticles. Nat. Nanotechnol. 2, 577–583 (2007).
Article ADS CAS PubMed Google Scholar
Kresse, G. & Furthmuller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).
Article ADS CAS Google Scholar
Ernzerhof, M. & Scuseria, G. E. Assessment of the Perdew–Burke–Ernzerhof exchange-correlation functional. J. Chem. Phys. 110, 5029–5036 (1999).
Article ADS CAS Google Scholar
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Article ADS CAS PubMed Google Scholar
Yu, M. et al. Interfacial stability, electronic property, and surface reactivity of α-MoO₃/γ-Al₂O₃ composites: DFT and DFT + U calculations. Comput. Mater. Sci. 153, 217–227 (2018).
Article CAS Google Scholar
Lei, Y. & Yan, M. DFT studies of selective oxidation of propene on the MoO₃(010) surface. Phys. Chem. Chem. Phys. 23, 2792–2804 (2021).
Article CAS PubMed Google Scholar
Gao, X. et al. High-performance Mo₂C/MWCNT electrocatalyst for MOR: comparison with MoO₂/MWCNT and MoO₃/MWCNT. Int. J. Hydrog. Energy 48, 32408–32419 (2023).
Article ADS CAS Google Scholar
Coquet, R. & Willock, D. J. The (010) surface of α-MoO₃, a DFT + U study. Phys. Chem. Chem. Phys. 7, 3819–3828 (2005).
Article CAS PubMed Google Scholar
Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).
Article ADS PubMed Google Scholar
Grimme, S., Ehrlich, S. & Goerigk, L. Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 32, 1456–1465 (2011).
Article CAS PubMed Google Scholar
Mathew, K., Sundararaman, R., Letchworth-Weaver, K., Arias, T. A. & Hennig, R. G. Implicit solvation model for density-functional study of nanocrystal surfaces and reaction pathways. J. Chem. Phys. 140, 084106 (2014).
Article ADS PubMed Google Scholar
Liu, J., Luo, F. & Li, J. Electrochemical potential-driven shift of frontier orbitals in M-N-C single-atom catalysts leading to inverted adsorption energies. J. Am. Chem. Soc. 145, 25264–25273 (2023).
Article CAS PubMed Google Scholar
Wang, J., Zheng, M., Zhao, X. & Fan, W. Structure-performance descriptors and the role of the axial oxygen atom on M-N₄-C single-atom catalysts for electrochemical CO₂ Reduction. ACS Catal. 12, 5441–5454 (2022).
Article CAS Google Scholar
Gao, S., Xiang, S., Shi, J., Zhang, W. & Zhao, L. Theoretical understanding of the electrochemical reaction barrier: a kinetic study of CO₂ reduction reaction on copper electrodes. Phys. Chem. Chem. Phys. 22, 9607–9615 (2020).
Article CAS PubMed Google Scholar
Gao, G., Zheng, F., Pan, F. & Wang, L. Theoretical investigation of 2D conductive microporous coordination polymers as Li-S battery cathode with ultrahigh energy density. Adv. Energy Mater. 8, 1801823 (2018).
Article Google Scholar
Momma, K. & Izumi, F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr. 44, 1272–1276 (2011).
Article ADS CAS Google Scholar
Shetty, M., Buesser, B., Román-Leshkov, Y. & Green, W. H. Computational investigation on hydrodeoxygenation (HDO) of acetone to propylene on α-MoO₃ (010) Surface. J. Phys. Chem. C. 121, 17848–17855 (2017).
Article CAS Google Scholar
Kashfi-Sadabad, R., Yazdani, S., Huan, T. D., Cai, Z. & Pettes, M. T. Role of oxygen vacancy defects in the electrocatalytic activity of substoichiometric molybdenum oxide. J. Phys. Chem. C. 122, 18212–18222 (2018).
Article CAS Google Scholar
Bose, A. L. & Agarwal, V. Oxygen healing and CO₂/H₂/anisole dissociation on reduced molybdenum oxide surfaces studied by density functional theory. ChemPhysChem 23, e202200510 (2022).
Article Google Scholar

Download references

Acknowledgements

This research was supported in part by the National Natural Science Foundation of China (22334002, L.W.), the National Natural Science Foundation of China (22076010, G.H.), Beijing Municipal Science and Technology Special Project (Z231100002723006, L.W.), and the Fundamental Research Funds for the Central Universities (JD2308 and XK2023-19, L.W.).

Author information

Authors and Affiliations

State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing, 100029, P.R. China
Liangyu Li, Xiaotong Liu, Guanghe Liu, Suying Xu, Gaofei Hu & Leyu Wang

Authors

Liangyu Li
View author publications
Search author on:PubMed Google Scholar
Xiaotong Liu
View author publications
Search author on:PubMed Google Scholar
Guanghe Liu
View author publications
Search author on:PubMed Google Scholar
Suying Xu
View author publications
Search author on:PubMed Google Scholar
Gaofei Hu
View author publications
Search author on:PubMed Google Scholar
Leyu Wang
View author publications
Search author on:PubMed Google Scholar

Contributions

L.W., G.H. and S.X. designed and supervised the research. L.L. conducted the experiments and X.L. and G.L. helped prepare the materials. L.W., G.H. and L.L. wrote and revised this manuscript. All the authors discussed the results and commented on the manuscript.

Corresponding authors

Correspondence to Gaofei Hu or Leyu Wang.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Jingxiang Zhao and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information (download PDF )

Peer Review File (download PDF )

Description of Additional Supplementary Files (download PDF )

Supplementary Data 1 (download ZIP )

Reporting Summary (download PDF )

Source data

Source Data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Li, L., Liu, X., Liu, G. et al. Valence-engineered catalysis-selectivity regulation of molybdenum oxide nanozyme for acute kidney injury therapy and post-cure assessment. Nat Commun 15, 8720 (2024). https://doi.org/10.1038/s41467-024-53047-1

Download citation

Received: 02 January 2024
Accepted: 30 September 2024
Published: 08 October 2024
Version of record: 08 October 2024
DOI: https://doi.org/10.1038/s41467-024-53047-1

This article is cited by

Ultrasmall platinum single-atom enzyme alleviates oxidative stress and macrophage polarization induced by acute kidney ischemia–reperfusion injury through inhibition of cell death storm
- Keng Ye
- Kongwen Lin
- Yanfang Xu
Journal of Nanobiotechnology (2025)
Tailoring the optoelectronic properties of MoOX nanoparticles: a novel microwave-assisted synthesis for near-infrared absorbing polyoxometallic clusters
- Daniel Truchan
- Adriana Hvizdošová Annušová
- Yoann Lalatonne
Discover Nano (2025)