Zinc Hexacyanoferrate Nanoenzymes Attenuate Neuroinflammation After Spinal Cord Injury by Modulating Mitochondrial Function and Microglial Polarization

Abstract

Background

Secondary pathology after acute spinal cord injury (SCI) is driven by pathological iron accumulation, reactive oxygen species (ROS) overproduction, and mitochondrial failure, which together sustain chronic neuroinflammation. Methylprednisolone (MP), the most widely used ROS-scavenging agent for acute SCI, provides limited clinical benefit and carries considerable side effects. To address this, we engineered zinc hexacyanoferrate nanoenzymes (ZnHPB) that chelate excess ferrous ions while displaying intrinsic SOD- and CAT-like activities, attenuating oxidative injury at its source rather than merely suppressing downstream inflammation. The defining feature of ZnHPB is a Fe(II)-driven self-reinforcing antioxidant loop in which captured ferrous species are converted into mixed-valence Prussian-blue phases, breaking the Fenton cascade that underpins iron- and ROS-driven secondary injury.

Methods

ZnHPB was prepared by polyvinylpyrrolidone-directed self-assembly and characterized for physicochemical and enzyme-mimetic properties. In vitro efficacy was evaluated in H₂O₂-challenged BV2 microglia in parallel with MP. C57BL/6 mice subjected to T8-T9 contusion SCI received ZnHPB (5 mg/kg, i.p.) under preventive or post-injury dosing regimens for 28 days, with a 60-day safety assessment. Outcome measures included cell viability, ROS levels, mitochondrial bioenergetics (ECAR/OCR), microglial phenotypic markers, plasma injury biomarkers, and histopathology of the lesion.

Results

In H₂O₂-challenged BV2 cells, ZnHPB improved viability more effectively than MP, with associated reductions in intracellular ROS, partial restoration of mitochondrial OXPHOS capacity, and alleviation of ferroptosis-related transcriptional signatures (GPX4, ACSL4, COX2). In SCI mice, ZnHPB preserved 86.71% (preventive) and 74.06% (post-injury) of white matter at the lesion epicenter and restored plasma CK, AST, and LDH toward physiological levels. Microglial phenotypic markers shifted from an M1-associated toward an M2-associated phenotype. No visceral toxicity was observed over 60 days.

Conclusion

By integrating ferrous-ion sequestration with ROS elimination and mitochondrial support, ZnHPB interrupts the self-perpetuating loop of oxidative stress and neuroinflammation that characterizes secondary SCI pathology. Under the conditions tested, ZnHPB compared favorably with MP across multiple preclinical readouts, supporting further mechanistic and translational evaluation.

Introduction

Antioxidant strategies have been increasingly explored in biomedical research,¹ with the potential to counteract a variety of pathological conditions including systemic toxicity,² inflammation,³ and metabolic dysfunction.⁴ The pathological accumulation of pro-inflammatory cytokines, driven by excessive production of reactive oxygen species (ROS) and/or deregulated metal-ion homeostasis, has emerged as a central theme in the pathogenesis of these diseases.⁵ Neuroinflammation in particular represents a destructive and life-threatening condition that commonly arises after acute trauma or during aging, and frequently leads to lasting neurological disability.^6,7 Trauma-induced neuroinflammation is a leading cause of permanent paraplegia and paralysis in patients with spinal cord injury (SCI) and remains a major unresolved challenge for clinicians.⁶ This inflammatory cascade in the spinal cord microenvironment not only worsens the immediate symptoms, but also delays recovery and may culminate in irreversible disability.⁸

The pathophysiology of SCI is widely believed to stem from the accumulation of ROS in the injured microenvironment, leading to a surge in superoxide anions, H₂O₂, and hydroxyl radicals.⁹ However, the limited clinical efficacy of methylprednisolone (MP), a direct ROS scavenger, has called the ROS-only hypothesis into question.⁸ Recent studies have revealed an additional dimension of post-SCI neuroinflammation: damaged spinal cord tissue exhibits marked intracellular accumulation of ferrous iron, accompanied by oxidative stress and mitochondrial dysfunction.¹⁰ This paradigm shift positions pathological iron accumulation as a key therapeutic target, complementing traditional approaches that address secondary oxidative stress alone. Combining iron chelation with ROS removal and mitochondrial support is therefore expected to improve neuroprotection and prognosis in SCI patients.

Although conventional antioxidants show strong radical-scavenging activity, their ability to chelate redox-active metals is limited compared with nanocatalysts based on hexacyanoferrate.¹¹ Multifunctional nanozymes with intrinsic antioxidant enzyme activity have been engineered and applied to various in vivo settings, including reperfusion injury, neurodegeneration, and aging-related pathology.¹¹ Through engineered electronic structures, manganese-containing nanoparticles,¹² vanadium-containing carbon nanozymes,¹³ and cerium oxide nanoparticles¹⁴ exhibit stable catalytic activity that mimics natural antioxidant enzymes such as superoxide dismutase (SOD) and catalase (CAT). Prussian blue (PB), an iron hexacyanoferrate(Fe₄[Fe(CN)₆]₃) framework, is approved by the US Food and Drug Administration as an antidote for radioactive cesium (Cs) or thallium (Tl) contamination and has well-documented biocompatibility in vivo. PB-like Prussian-blue analogue (PBA) nanoparticles, in which the iron centers are partially or fully replaced by transition-metal ions, have been developed as versatile nanotherapeutic platforms for drug delivery and antioxidant applications.^15,16 PBA nanoparticles exhibit iron-chelation properties owing to their very low solubility product constant (Ksp = 3.3×10⁻⁴¹), orders of magnitude lower than that of analogous ferrocyanides.¹⁷ This property endows PBA nanoparticles with a strong capacity to capture ferrous species. By engineering the kinetics of cation exchange, it also provides a versatile platform for biomedical intervention.¹⁷

Here, we introduce a zinc-based PBA nanomedicine (ZnHPB) assembled with polyvinylpyrrolidone, designed to sequester pathological ferrous ions while providing antioxidant neuroprotection in SCI (Figure 1). Our data show that ZnHPB binds pathological ferrous species and attenuates oxidative stress both in vitro and in vivo, with a comprehensive evaluation of biocompatibility, biosafety, and neuroprotective efficacy supporting its potential as an advanced therapeutic for secondary SCI injury.

Figure 1.

Scheme of neuroprotective strategy by ZnHPB.

Materials and Methods

Synthesis of ZnHPB

ZnCl₂·6H₂O (0.8 mmol) and polyvinylpyrrolidone (PVP; Mw = 10,000; 300 mg) were completely dissolved in 40 mL of aqueous HCl (pH 1.0) under vigorous stirring. A second aqueous solution (pH 1.0, 40 mL) containing potassium ferricyanide (K₃[Fe(CN)₆]; 0.8 mmol) was added dropwise at a rate of 80 mL/h, and the reaction was stirred at room temperature for 2 h. The mixture was dialyzed against deionized water for 6 h, then concentrated by rotary evaporation and freeze-dried to obtain the final ZnHPB product.

Quantitative Real-Time RT-PCR

Total RNA was extracted from collected samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), reverse-transcribed using a PrimeScript RT Reagent Kit (TaKaRa, Shiga, Japan), and quantified by qPCR on a QuantStudio 5 system (Applied Biosystems) with SYBR Green PCR Master Mix. Relative expression levels of target genes were normalized to the housekeeping gene beta-actin, and fold changes were calculated using the 2^(-DeltaDeltaCt) method. Primer sequences, amplicon sizes, and efficiency data are available from the corresponding author upon reasonable request.

Animals and Surgical Procedures

Male C57BL/6 mice (6–8 weeks old, 22–30 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. and housed under specific pathogen-free (SPF) conditions at the Animal Care Facility of Gannan Medical University, with free access to food and water and a 12 h/12 h light/dark cycle. After a 7-day acclimatization period, mice were anesthetized by inhalation of isoflurane (induction 3% in oxygen, maintenance 1.5–2% via nose cone) and subjected to moderate T8-T9 spinal cord contusion using a computer-controlled impactor (10 g rod, 1.5 mm displacement). Pre-emptive analgesia (buprenorphine 0.1 mg/kg, s.c.) was administered 30 min before surgery and then every 12 h for 72 h post-operatively in accordance with institutional analgesia and animal-care standards and local IACUC requirements. Post-operatively, mice received manual bladder expression twice daily until autonomic voiding recovered, and prophylactic antibiotic (gentamicin, 5 mg/kg i.p.) for 5 days. Animals were randomly assigned to treatment groups using a computer-generated randomization list, with the surgeon blinded to group allocation throughout the study. Group sizes (n per group) were: sham control n = 10, SCI + saline n = 14 (to account for post-injury mortality), SCI + ZnHPB preventive n = 12, SCI + ZnHPB post-injury n = 12, SCI + MP n = 12; total n = 60 for the 28-day study. ZnHPB was administered intraperitoneally at 5 mg/kg under two regimens: a preventive regimen (every 3 days, from 7 days before injury through day 28 post-injury) and a post-injury regimen (every 6 days, from day 0 through day 28 post-injury). MP was administered intraperitoneally at 30 mg/kg once daily for 3 days, starting immediately after injury; this reflects a commonly used preclinical regimen in rodent SCI models rather than the full clinical NASCIS protocol. Sham and SCI + saline controls received an equivalent volume of saline on the same schedule as the post-injury ZnHPB arm. Animals found in obvious distress or with > 20% body-weight loss were euthanized and excluded from analysis according to a pre-specified humane endpoint criterion. Terminal sacrifice was by deep isoflurane anesthesia followed by transcardial perfusion with ice-cold PBS and 4% paraformaldehyde. All animal procedures were approved by the Biomedical Research Ethics Committee of Gannan Medical University (Approval No. 2026378) and conducted in accordance with the ARRIVE 2.0 guidelines, the Guide for the Care and Use of Laboratory Animals (8th edition, National Research Council, 2011), and the Chinese national standard GB/T 35892–2018 (Laboratory Animal Welfare Guideline).

Behavioral Assessment

Locomotor recovery was assessed by two blinded investigators using the Basso Mouse Scale (BMS) at 0, 1, 3, 7, 14, 21, and 28 days post-injury.¹⁸ A score of 0 indicates no ankle movement and a score of 9 indicates normal function. Group means and inter-rater agreement (> 0.85) are reported; BMS scores were plotted over time for each group.

Histopathological Staining

Spinal cord tissues were snap-frozen, embedded in optimum cutting temperature (O.C.T.) compound (Sakura, Leiden, The Netherlands), and sectioned at −20 °C into 5-µm slices. Sections were permeabilized in 0.3% Triton X-100 in PBS for 30 min, blocked with 1% BSA for 45 min, and incubated with primary antibodies overnight at 4 °C. After washing, sections were stained with Alexa Fluor 647 and Alexa Fluor 488 secondary antibodies for 1 h at room temperature, counterstained with DAPI, and imaged on a confocal laser microscope (Zeiss LSM 880). Primary antibody details (target, host species, vendor, catalog number, RRID, working dilution, and antigen retrieval) are available from the corresponding author upon reasonable request.

Statistical Analysis

Data are presented as mean ± standard deviation (SD) of at least three independent experiments, with biological replicates indicated in each figure legend. Differences between groups were tested by one-way analysis of variance (ANOVA) followed by the Tukey-Kramer post-hoc test, using GraphPad Prism version 9.5.1 (GraphPad Software, San Diego, CA, USA). For BMS scores, a repeated-measures two-way ANOVA with Bonferroni correction was used. Differences were considered statistically significant at P < 0.05. Sample sizes were estimated a priori based on pilot data to detect a 30% difference with 80% power (alpha = 0.05) using G*Power 3.1.9.7 (Heinrich-Heine-Universität Düsseldorf).

Results

Synthesis and Characterization of ZnHPB

ZnHPB was synthesized via a single-step polyvinylpyrrolidone (PVP, Mw = 10 kDa)-directed self-assembly (Figure 2a). Mixing equimolar Zn(II) with potassium ferricyanide (K3[Fe(CN)6]) produced an immediate color transition from colorless to deep blue; subsequent dialysis and freeze-drying yielded the final nanoformulation. Rehydration in saline or PBS gave a uniform colloidal suspension with a mean particle size of 26.52 ± 5.22 nm by transmission electron microscopy (TEM) (Figure 2b; scale bars: 200 nm in the main panel, 30 nm in the inset) and a hydrodynamic diameter of 28.55 ± 7.23 nm by dynamic light scattering (DLS) (Figure 2c), consistent with good colloidal stability. Fourier-transform infrared (FT-IR) spectroscopy showed characteristic absorption bands attributable to the PVP polymer (Figure 2d), and the zeta potential of −15.3 ± 4.57 mV (Figure 2e) is consistent with the negatively charged Fe(CN)₆ surface that stabilizes the colloid under physiological conditions. Particle size remained unchanged over 7 days in PBS and plasma (Figure 2f), confirming robust colloidal stability. Together these data demonstrate that ZnHPB is a stable nanoplatform under physiological conditions capable of chelating ferrous ions and neutralizing oxidative stress.

Figure 2.

Preparation and characterization of ZnHPB. (a) Schematic illustration of the PVP-directed self-assembly of ZnHPB. (b) Transmission electron microscopy (TEM) image and (c) size distribution of ZnHPB (scale bars: 200 nm in the main panel, 30 nm in the inset). (d) FT-IR of ZnHPB. (e) Zeta potential of ZnHPB. (f) Size distribution of ZnHPB in PBS and plasma. The data are presented as mean ± standard deviation (SD) (n = 3).

Ferrous Ion Capture and the Multicatalytic Activities of ZnHPB

ZnHPB showed strong binding affinity for ferrous ions, with the initial PB-like framework modified by replacement of Zn(II) within the nanostructure (Figure 3a). To characterize this conversion, we titrated ZnHPB with Fe(II) in PBS (pH 7.4) containing H₂O₂ and recorded UV–vis spectra over increasing Fe(II) concentrations (Figure 3b). Addition of Fe(II) produced a characteristic broad absorption band spanning 500–1000 nm with a maximum near 700 nm, indicating intervalence charge transfer between Fe(II) and Fe(III) and confirming conversion to the canonical mixed-valence Prussian-blue structure (Figure 3b). Because this band requires the coexistence of Fe(II) and Fe(III), its intensity reports the extent of mixed-valence phase formation rather than loss of the framework. The absorbance at 720 nm was approximately 3.2-fold higher in H₂O₂ than in pH 7.4 buffer at the same wavelength (Figure 3b), indicating that H₂O₂ promotes the Fe(II)-driven conversion of ZnHPB to the Prussian-blue phase rather than destroying the framework. Cation exchange was monitored by dialyzing ZnHPB (10 mL, 20 µg/mL expressed as Zn) against excess Fe(II) and sampling the dialysate by inductively coupled plasma optical emission spectroscopy (ICP-OES). Iron content rapidly declined within 1 h while zinc content progressively approached an equimolar match, confirming that Fe(II) replaces Zn(II) in the framework (Figure 3c). These data establish that ZnHPB efficiently captures ferrous ions by stoichiometric cation exchange, releasing Zn(II) and yielding a PB-like end-product. Antioxidant activity was assessed next (Figure 3d). ZnHPB scavenged superoxide in an SOD activity assay, reaching approximately 60% scavenging efficiency after 10 min at both 1 and 5 µg/mL, indicating that the effect was already saturated at the lowest concentration tested (Figure 3e). Catalase-like activity increased with ZnHPB concentration and plateaued at higher concentrations (Figure 3f). Throughout, we use the terms SOD- and CAT-like to denote the observed scavenging activities, without implying full Michaelis-Menten enzymatic turnover. Electron spin resonance (ESR) showed that co-incubation of ZnHPB with H₂O₂ substantially decreased the H₂O₂ signal, with 5 µg/mL reducing the signal by roughly 50% (Figure 3g). Dissolved-oxygen measurements showed that net O2 generation was highest in the H₂O₂ + ZnHPB group, exceeding that in the H₂O₂ and ZnHPB groups alone, confirming that ZnHPB decomposes H₂O₂ to O₂ (Figure 3h). Cumulatively, these results indicate that ZnHPB combines stoichiometric ferrous capture with effective H₂O₂ (and to a lesser extent superoxide) scavenging, the principal activities responsible for its antioxidant effect.

Figure 3.

Ferrous capturing and antioxidation performance of ZnHPB. (a) Schematic illustration of the ferrous binding activity of ZnHPB as a Prussian blue analogue by reverting to the pristine Prussian blue nanoparticles. (b) UV–vis spectra of ZnHPB in PBS (pH 7.4) containing H₂O₂ at increasing Fe(II) concentrations. (c) Time-dependent elemental concentration profiles for Zn²⁺ and Fe²⁺ in the solution during the dialysis-membrane separation-regulated cation exchange for ZnHPB. (d) Schematic illustration of the proposed SOD- and CAT-like scavenging activity of ZnHPB. (e) SOD activity of ZnHPB. (f) CAT activity of ZnHPB. (g) ESR spectra of ZnHPB in eliminating H₂O₂. (h) Net O₂ generation in the H₂O₂, ZnHPB, and H₂O₂ + ZnHPB groups. The data are presented as mean ± standard deviation (SD) (n = 3).

ZnHPB Rescued Mitochondrial Dysfunction in vitro

The murine microglial BV2 cell line (ATCC Cat# YC-C035) was cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and used as an in vitro model to benchmark the protective effects of ZnHPB against MP (Figure 4a). To establish an oxidative-stress challenge, BV2 cells were exposed to H₂O₂ at 50–200 µM; Calcein-AM/PI staining showed dose-dependent cell death that plateaued above approximately 100 µM (Figure 4b), and MTT assays confirmed the loss of viability (Figure 4c). We therefore used 100 µM H₂O₂ as the standardized stress challenge for all downstream comparisons; this concentration was chosen based on established in vitro oxidative-stress models in the literature and corresponds to a moderate, sub-lethal oxidative challenge that mimics pathological ROS levels without overwhelming cellular antioxidant capacity. Co-incubation with either ZnHPB or MP significantly improved viability, with ZnHPB consistently providing greater protection across the concentration range tested (Figure 4d and e). These results support the capacity of ZnHPB to limit microglial death under oxidative stress and motivated further mechanistic evaluation against MP.

Figure 4.

In vitro cellular evaluations. (a) Schematic illustration of the in vitro cellular experiments. (b) Confocal microscopic images of Calcein-AM/PI-stained cells treated with varying doses of H₂O₂ (50–200 µM). (c) Viability of BV2 cells treated with varying doses of H₂O₂. Viability of BV2 cells treated with H₂O₂ supplemented with different concentrations of (d) MP or (e) ZnHPB. (f) Schematic illustration of the in vitro mitochondrial dysfunction assay. (g) Quantification of ROS levels after the different treatments. (h) Quantification of ECAR levels after the different treatments. (i) Quantification of OCR levels after the different treatments. Data are presented as mean ± standard deviation (SD) (n = 3 biological replicates). Statistical significance is indicated as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, and n.s. for not significant.

To profile the transcriptional response to H₂O₂, we performed RNA sequencing on BV2 microglia. H₂O₂ treatment significantly altered gene expression, with 3891 differentially expressed genes (DEGs, P < 0.05). Gene Ontology and volcano plot analysis revealed enrichment for membrane components and metal-ion binding, reflecting the major transcriptional shifts under oxidative stress (Supplementary Figure 1a and b). Pathway analysis highlighted apoptosis and iron-dependent processes, consistent with the multifaceted nature of H₂O₂-induced microglial injury (Supplementary Figure 1c and d). H₂O₂ upregulated COX2 and ACSL4 while downregulating GPX4, an expression pattern consistent with engagement of ferroptosis-related transcriptional programs. Co-incubation with ZnHPB (5 µg/mL) or MP (10 µg/mL) significantly attenuated these transcriptional changes, with ZnHPB showing a substantially larger effect on iron-pathway genes than MP (Supplementary Figure 1e–g).

To investigate metabolic consequences of oxidative stress, we profiled microglial bioenergetics using the Seahorse Extracellular Flux system (Figure 4f). ROS measurement and metabolic analysis were used to evaluate the protective effects of ZnHPB and MP in BV2 cells (Figure 4g–i). The DCFH-DA fluorescence probe showed that ZnHPB significantly lowered ROS levels in H₂O₂-treated microglia (Figure 4g). For mitochondrial function, we quantified the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) under oligomycin, FCCP, rotenone, and antimycin A challenge to derive oxidative phosphorylation (OXPHOS) parameters. ZnHPB-treated microglia displayed enhanced OXPHOS capacity and a partial shift from glycolysis back toward mitochondrial ATP production relative to H₂O₂-treated controls, which showed predominant glycolytic metabolism and suppressed OXPHOS (Figure 4h and i). These findings suggest that ZnHPB partially restores mitochondrial function in H₂O₂-stressed microglia, more effectively than MP under the conditions tested.

In vivo Neuroprotective Effect of Acute SCI Mice

To assess biosafety in vivo, ZnHPB (5 mg/kg) was administered intraperitoneally to C57BL/6 mice. During the 60-day observation period, body weight was stable and serum biochemistry remained within reference ranges (Supplementary Figures 2 and 3). These data support a favorable acute biosafety profile for ZnHPB at the dose and schedule used.

To evaluate in vivo efficacy, T8-T9 spinal cord contusion was performed in healthy C57BL/6 mice (Figure 5a). ZnHPB was administered in two regimens: preventive (5 mg/kg, every 3 days from 7 days before injury through day 28 post-injury) and post-injury (5 mg/kg, every 6 days from day 0 post-injury through day 28). Body weight and survival were monitored for 28 days (Figure 5b and c). Mice in SCI groups lost weight during the first post-operative week; four mice in the untreated SCI group died within 7 days. The untreated group showed an early mortality of approximately 29% (4 of 14 mice), predominantly attributable to post-injury bladder dysfunction and urinary retention that required twice-daily manual bladder expression; antibiotic prophylaxis (gentamicin, 5 mg/kg i.p. for 5 days) and fluid support were provided as standard post-operative care. This higher-than-expected mortality reflects the contusion severity chosen to maximize therapeutic differentiation and is acknowledged as a study limitation in the Discussion. ZnHPB-treated groups showed faster body-weight recovery and improved survival relative to untreated SCI controls. H&E staining of spinal cord sections revealed reduced lesion area in both preventive (86.71% white-matter preservation) and post-injury (74.06%) ZnHPB groups compared with the untreated SCI group (Figure 5d and e). Plasma creatine kinase (CK), aspartate aminotransferase (AST), and lactate dehydrogenase (LDH), markers of secondary tissue injury, returned toward control levels in ZnHPB-treated mice (Figure 5f–h). Survival monitoring over 28 days showed approximately 20% survival in the untreated SCI group (Figure 5c), consistent with the severity of the contusion model and post-injury complications despite supportive care.

Figure 5.

In vivo SCI recovery evaluation. (a) Dosing schedule of the in vivo SCI recovery evaluation. (b) Body weights and (c) survival rates of mice from different groups during the evaluation period. (d) H&E staining images of the spinal cord sections in different groups. (e) Quantification of the area of spared white matter after the different treatments. (f–h) Plasma CK (f), AST (g) and LDH (h) levels of each mouse in different groups at the end of the evaluation period. The data are presented as mean ± standard deviation (SD); group sizes were n = 10 (sham), 14 (SCI + saline), 12 (SCI + ZnHPB preventive), 12 (SCI + ZnHPB post-injury), and 12 (SCI + MP), as specified in the Animals and surgical procedures section. Statistical significance is indicated as *P < 0.05, **P < 0.01, and n.s. for not significant.

Effects on Microglial Polarization and Downstream Anti-Inflammatory Readouts

To assess potential anti-inflammatory properties, microglial polarization was evaluated after 14 days of ZnHPB treatment in SCI mice. Microglia were classified by marker expression: inducible nitric oxide synthase (iNOS), CD86, and IL-1β for the pro-inflammatory M1-associated phenotype, and CD206 for the anti-inflammatory M2-associated phenotype (Figure 6a). Immunofluorescence imaging showed that, relative to the SCI group, both preventive and post-injury ZnHPB regimens increased the number of F4/80-positive cells at the lesion while decreasing iNOS-positive M1-associated microglia (Figure 6b–e). qPCR analysis confirmed that pro-inflammatory cytokine transcripts (IL-1β and CD86) were elevated after SCI and were attenuated by ZnHPB, while the anti-inflammatory marker CD206 was increased (Figure 6f–h). Tissue and plasma levels of TNF, IL-6, IL-1β, GPX4, ACSL4, COX2, SOD, GSH, and MDA were quantified in parallel (Supplementary Figures 4–7). Collectively, these data indicate that ZnHPB shifts microglial phenotypic markers toward an M2-associated profile in vivo, accompanied by reduced pro-inflammatory cytokine load.

Figure 6.

In vivo anti-neuroinflammation evaluation. (a) Schematic illustration of the experimental design. (b) Immunofluorescence staining images of the spinal cord sections in different groups. Quantification of (c) F4/80-positive, (d) iNOS-positive, (e) percentage of iNOS⁺ F4/80⁺ double-positive cells after the different treatments. Quantification of (f) IL-1β, (g) CD86, (h) CD206 mRNA after the different treatments. The data are presented as mean ± standard deviation (SD) (n=6/group). Statistical significance is indicated as *P < 0.05, **P < 0.01, ***P < 0.001, and n.s. for not significant.

Chronic-Phase Spinal Cord Regeneration

Spinal cord regeneration involves scar remodeling and synaptic reconnection. To examine whether ZnHPB supports these later-phase events, we administered ZnHPB from day 28 post-injury (the chronic phase) onward. Immunofluorescence imaging showed that synapsin (Syn) immunoreactivity in peri-lesional regions was increased, indicating improved synaptic density (Supplementary Figure 8). Chondroitin sulfate proteoglycan (CSPG) immunoreactivity was reduced in ZnHPB-treated animals, consistent with reduced scar deposition at the lesion (Supplementary Figure 9). Together, these observations suggest that ZnHPB not only limits acute inflammation but also creates a permissive environment for spinal cord regeneration in the chronic phase of injury.

Discussion

Spinal cord injury is characterized by the accumulation of ROS in the lesion micro-environment, leading to a surge of superoxide anions, H₂O₂ and hydroxyl radicals. The limited clinical efficacy of methylprednisolone (MP), a direct ROS scavenger, has called the ROS-only hypothesis into question. Recent studies have revealed an additional dimension of post-SCI neuroinflammation: damaged spinal cord tissue exhibits a marked intracellular accumulation of ferrous species together with oxidative stress and mitochondrial dysfunction, identifying pathological iron accumulation as a key therapeutic target that complements traditional ROS-only approaches.

We therefore propose a combined treatment strategy that simultaneously removes pathological iron and improves ROS handling and mitochondrial respiration, with the goal of providing better neuroprotection and prognosis in SCI patients. Compared with conventional ferricyanide-based nanocatalysts, this approach more directly attacks the iron-ROS axis that drives secondary injury. Multifunctional nanozymes with antioxidant activity, including manganese-containing nanoparticles, vanadium-containing carbon nanozymes, and cerium oxide nanoparticles, have been engineered to mimic natural SOD/CAT activity. Prussian blue (PB) is an iron hexacyanoferrate (Fe4[Fe(CN)6]3) framework approved by the US FDA as an antidote for radioactive Cs/Tl contamination and has well-documented biocompatibility in vivo. By replacing the iron center with zinc, the PBA nanoparticles (ZnHPB) developed here gain a very low solubility product constant (Ksp = 3.3×10-41), endowing them with strong ferrous-capture capacity and rapid cation exchange with Fe(II). UV–vis spectroscopy showed enhanced visible-light absorption in the 500–1000 nm window after Fe(II) addition, confirming formation of the mixed-valence Prussian-blue phase; the higher absorbance at 720 nm in H₂O₂ than in pH 7.4 buffer indicates that H₂O₂ promotes this Fe(II)-driven conversion rather than decomposing the framework. ICP-OES monitoring of cation exchange confirmed that Fe(II) replaces Zn(II) in the framework, and yielding a PB-like phase, supporting the proposed self-reinforcing protective loop. From a translational standpoint, ZnHPB compares favorably with other strategies evaluated for SCI, including methylprednisolone and GM-1 ganglioside,¹⁹ NT3-chitosan,²⁰ and selenium-doped carbon quantum dots (Se-CQDs),²¹ because it combines injectability, low toxicity, and a simple one-step synthesis, although no direct head-to-head efficacy comparison with these agents has yet been performed.

The in vitro efficacy of ZnHPB was benchmarked directly against MP in H₂O₂-challenged BV2 microglia. PI staining and MTT assays established that H₂O₂ reliably induced microglial death, while co-incubation with ZnHPB significantly improved cell survival more effectively than MP across the tested concentration range. At the transcriptional level, H₂O₂ produced broad gene-expression changes, with significant enrichment for apoptosis- and iron-pathway genes; ZnHPB attenuated these signatures more strongly than MP, consistent with engagement of an iron-handling mechanism that MP does not share. ZnHPB also reduced intracellular ROS (DCFH-DA), partially restored OXPHOS, and reversed H₂O₂-induced mitochondrial dysfunction in microglia more effectively than MP under the conditions tested. These findings support ZnHPB as a multi-target neuroprotective candidate relative to MP, while we acknowledge that in vitro comparisons cannot fully predict in vivo performance.

For in vivo evaluation, intraperitoneal ZnHPB at 5 mg/kg did not produce visible organ pathology over the 60-day observation period. In the T8-T9 contusion model, ZnHPB treatment (preventive or post-injury) improved body-weight recovery, reduced lesion area, and normalized plasma CK, AST, and LDH, supporting overall neuroprotection. Immunofluorescence imaging showed that ZnHPB increased F4/80-positive microglia and decreased iNOS-positive M1-associated microglia, while qPCR analysis showed reduced pro-inflammatory cytokine transcripts (IL-1β, CD86) and increased CD206. Tissue and plasma levels of TNF, IL-6, IL-1β, GPX4, ACSL4, COX2, SOD, GSH, and MDA shifted in the expected anti-inflammatory direction, and ZnHPB compared favorably with MP across multiple readouts.

Several caveats should temper the conclusions drawn here. (i) The physicochemical characterization of ZnHPB is limited to a single batch; batch-to-batch reproducibility, long-term colloidal stability beyond 7 days, biodistribution, and the identity and safety of degradation products remain to be established in dedicated studies. (ii) The in vivo SCI evaluation was performed in a single moderate-contusion model; biodistribution, pharmacokinetics, and dose-response have not yet been systematically optimized. (iii) The MP comparison is at a single dose and a single regimen (30 mg/kg, i.p., daily for 3 days), which reflects a commonly used preclinical regimen rather than the full clinical NASCIS protocol, and should be interpreted accordingly. (iv) Functional recovery was not formally assessed: BMS scores were recorded but are not reported in the present manuscript, and no gait, footprint, or electrophysiological readouts were obtained. Whether the structural and biochemical improvements reported here translate into functional benefit therefore remains to be determined. (v) Mechanistic conclusions, including mitochondrial “restoration”, ferroptosis suppression, and M1-to-M2 repolarization, are based on indirect transcriptional and biochemical readouts and should be confirmed by additional protein-level and functional assays (eg, mitochondrial respiration in isolated mitochondria, lipid peroxidation assays, and Western blot for ferroptosis effectors) before being treated as definitive. In addition, whether the Prussian-blue phase formed after Fe(II) capture itself retains catalytic activity was not directly assessed in this study. (vi) Finally, our reading of microglial polarization using the classical M1/M2 framework simplifies a heterogeneous reality; single-cell transcriptomic studies²² have shown that microglia in CNS injury comprise multiple disease-associated and intermediate states that the M1/M2 dichotomy does not fully capture. The present M1/M2 results should therefore be interpreted as a convenient aggregate readout rather than a definitive binary switch.

Beyond these caveats, the present work fits within a broader precision nanomedicine trajectory. Quantum dots (QDs) have recently been highlighted as a promising class of nanoprobes for precision nanomedicine, owing to their tunable emission, photostability, and emerging compatibility with deep-tissue and image-guided applications.²³ In the context of CNS injury, where longitudinal, non-invasive monitoring of lesion evolution remains challenging, QD-based platforms may ultimately complement therapeutic nanozymes such as ZnHPB by enabling visualization of iron dynamics, ROS bursts, and microglial activation. Future hybrid systems that pair iron-handling nanozymes with QD-type reporters may therefore offer combined neuroprotection and image-guided evaluation of recovery, in line with the trajectory outlined for advanced precision nanomedicine.

Conclusion

We synthesized ZnHPB, a zinc hexacyanoferrate nanocatalyst prepared by PVP-directed self-assembly, which combines strong ferrous-capture capacity with intrinsic antioxidant activity. ZnHPB rescued H₂O₂-stressed microglia more effectively than MP in vitro, attenuated multiple readouts of secondary SCI injury in vivo, and was well tolerated at the dose tested. Together these data support ZnHPB as a candidate therapeutic for secondary injury after SCI and provide a rationale for further mechanism-focused and translational evaluation.

Funding Statement

This work was supported by the Science and Technology Research Project of Department of Education of Jiangxi Province [grant number GJJ2501316].

Data Sharing Statement

All raw data supporting the findings of this study are available from the corresponding author, Dr. Sen Lin, upon reasonable request.

Ethics Approval and Consent to Participate

All animal experiments were approved by the Biomedical Research Ethics Committee of Gannan Medical University (Approval No. 2026378). All animals were housed and all surgical and behavioral procedures were performed at the Animal Care Facility of Gannan Medical University. All procedures were performed in accordance with the ARRIVE 2.0 guidelines, the Guide for the Care and Use of Laboratory Animals (8th edition, National Research Council, USA, 2011), the AVMA Guidelines for the Euthanasia of Animals (2020 edition), and the Chinese national standard GB/T 35892-2018 (Laboratory Animal Welfare Guideline). No human participants or human tissue samples were involved in this study, and informed consent is therefore not applicable.

Author Contributions

R.L. and Y.L. conceived and designed the study; R.L. performed the experiments and analyzed the data; S.L. supervised the project and provided funding; R.L. wrote the original draft; S.L. and Y.L. critically revised the manuscript. All authors read and approved the final version of the manuscript. All authors contributed to data analysis, drafting or revising the article, have agreed on the journal to which the article will be submitted, gave final approval of the version to be published, and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.