Biomimetic membrane-coating zinc sulfide nanoparticles for anti-inflammatory combined neuroprotective therapy for spinal cord injury

Abstract

Spinal cord regeneration remains challenging due to complex inflammatory microenvironments, imbalances in metal ions, and obstacles to neuronal regeneration following spinal cord injury (SCI). Herein, microglial cell membranes coated with zinc sulfide nanoparticles modified with albumin (ZnS@BSA@MM) were designed as an anti-inflammatory combined neuroprotective therapy for SCI. ZnS@BSA@MM NPs were constructed via albumin modification and membrane extrusion and exhibited ROS-scavenging abilities comparable to those of natural products and slow H₂S release under acidic conditions. In vitro and in vivo experiments demonstrated the outstanding therapeutic effects of the ZnS@BSA@MM. In detail, the released H₂S and Zn²⁺ not only inhibit microglial activation through the NF-κB signaling axis but also promote the axonal growth of neurons under pathological conditions. Notably, microglial cell membranes effectively deliver ZnS@BSA to the lesion area. Finally, ZnS@BSA@MM facilitated the axonal regeneration of neurons in SCI, suppressed inflammatory responses, and activated multiple pathways, including cytokine-cytokine receptor interactions, neuroactive ligand-receptor interactions, and cAMP signaling. Collectively, this work highlights the anti-inflammatory and neuroprotective effects of ZnS@BSA@MM NPs, featuring satisfactory H₂S release and Zn²⁺ supplementation under membrane-targeting conditions for SCI therapy.

Graphical abstract

Highlights

• •Microglial membrane-coated ZnS@BSA provides anti-inflammatory neuroprotection.
• •ZnS@BSA@MM scavenges ROS like natural products and releases H₂S slowly in acid.
• •In vitro and in vivo studies show significant therapeutic benefit in SCI.
• •ZnS@BSA@MM promotes axonal regeneration and suppresses neuroinflammation.
• •Mechanism supported by H₂S kinetics and Zn²⁺ supplementation in the lesion.

1. Introduction

The incidence of spinal cord injury (SCI) is increasing due to growth in construction, transportation, and sports, particularly in young males. Globally, approximately 30 million people live with SCI, significantly straining economic resources [1]. Primary SCI stems from direct impacts, leading to secondary damage processes, including ischemia, inflammation, and oxidative stress, which worsen injury [2]. Following traumatic SCI, phagocytic immune cells (microglia, neutrophils, and macrophages) are activated and recruited to the injury site. These cells exacerbate neuronal death by releasing reactive oxygen species (ROS), which not only promote inflammation but also intensify secondary damage [3]. Therefore, therapeutic strategies that focus on mitigating oxidative stress and the inflammatory response are crucial for improving outcomes after SCI.

Existing therapeutic approaches for SCI, including pharmacological agents, surgical procedures, and hyperbaric oxygen therapy, are aimed at facilitating axonal regeneration but have shown limited efficacy [4]. There is a critical need for the formulation of more effective protective strategies. Gasotransmitters, synthesized in vivo from specific substrates by rate-limiting enzymes, regulate oxidative stress, inflammation, and apoptosis [5]. These effects could be replicated by external donors for therapeutic application [[6], [7], [8]]. Therefore, the use of biological gases might provide a promising SCI treatment approach because of their multimodal properties.

Hydrogen sulfide (H₂S) mitigates neuritis symptoms and prevents secondary neuronal damage by reducing inflammatory cytokines, apoptosis, and oxidative stress [9,10]. H₂S reduces inflammation by diminishing NF-κB pathway activity and enhancing protein kinase B phosphorylation, which promotes a stable immune microenvironment in SCI [11]. Accordingly, there is increasing interest in identifying novel H₂S donors that could offer variable release rates, which is essential for avoiding the deleterious effects of rapid gas increases. Metal centers interact with likely HS⁻ to form H₂S [12]. H₂S exhibits binding affinity toward zinc centers, resulting in the inhibition of enzymes such as angiotensin-converting enzyme and carbonic anhydrase [13,14]. Zn²⁺, the second most abundant transition metal ion in organisms, plays numerous crucial roles in neurogenesis processes. Zn²⁺ deficiency reduces neurogenesis and decreases the expression of genes linked to neuron formation [15,16]. Zinc sulfide (ZnS) nanoparticles exhibit excellent biocompatibility, releasing H₂S to clear ROS and inhibit oxidative stress. Zn²⁺ released from ZnS promotes neuronal axon regeneration, exerting a role in nerve regeneration. These properties are particularly beneficial for SCI repair. However, due to the potential release of H₂S in the bloodstream might result in H₂S depletion, it is crucial to effectively deliver ZnS directly to the SCI site.

Biomimetic cell membrane-coated nanoparticles have gained attention for disease treatment because of their reduced immunogenicity, prolonged circulation time, and enhanced drug delivery efficiency [17,18]. Cell chemotaxis and associated chemokines confer targeted effects to the lesion site, facilitating the recruitment of microglia along with cell-derived biomimetic nanoparticles [19]. Metal sulfide nanoparticles, such as ZnS, often exhibit poor colloidal stability, easy aggregation, and cytotoxicity due to uncontrolled Zn²⁺ release under physiological conditions [20]. To address these issues, bovine serum albumin (BSA) was introduced as a stabilizing and biocompatible coating, yielding ZnS@BSA nanoparticles with improved dispersibility, reduced toxicity, and enhanced biocompatibility [21]. Herein, to supplement the concentration of Zn²⁺ post-SCI [22], microglial cell membranes coating zinc sulfide nanoparticles modified with albumin (ZnS@BSA@MM) were designed as anti-inflammatory combined neuroprotective therapy for SCI (Scheme 1). ZnS@BSA@MM NPs, developed via albumin modification and membrane extrusion, demonstrate ROS-scavenging abilities and slow H₂S release in acidic environments. In vitro and in vivo studies confirmed their therapeutic effects, as H₂S and Zn²⁺ inhibited microglial activation through the NF-κB pathway and promoted axonal growth. Microglial membranes enhanced targeted delivery to lesion sites, facilitating axonal regeneration, reducing inflammation, and activating pathways such as cytokine-cytokine receptor interaction and cAMP signaling. This study highlights ZnS@BSA@MM as a promising therapeutic material with significant anti-inflammatory and neuroprotective potential, offering a novel approach for SCI treatment.

Scheme 1.

Diagram of the delivery of ZnS@BSA@MM nanoparticles delivered to the SCI area via a microglia membrane coating to exert anti-inflammatory and neurotrophic effects. By using the targeting ability of microglial cell membranes, ZnS@BSA@MM nanoparticles are delivered to the SCI site. These nanoparticles release H₂S and Zn²⁺, which reduce immune cell inflammation and promote nerve cell growth, helping to restore movement and nerve function in SCI mice.

2. Results and discussion

2.1. Fabrication and characterization of ZnS@BSA NPs

Zn(CH₃COO)₂ and NaHS were co-loaded with the BSA protein to form ZnS@BSA nanomaterials via a self-assembly method (Fig. 1a). The diameter of ZnS@BSA was approximately 100 nm, indicating a uniform spherical structure with excellent crystallinity, as confirmed by transmission electron microscopy (TEM) (Fig. 1b). Furthermore, the successful synthesis of ZnS@BSA nanoparticles and the effective BSA coating were corroborated by EDS analysis, which identified the characteristic elements Zn and S (Fig. S1). ZnS@BSA showed excellent dispersion and stability in different solutions, such as ultrapure water, phosphate-buffered saline (PBS), and Dulbecco's modified Eagle medium (DMEM) (Fig. 1c). The zeta potential indicated the partial charge of ZnS@BSA in both the PBS and the DMEM solutions (Fig. 1d). The broader distribution in PBS was due to its high ionic strength, which compressed the electrical double layer and increased charge heterogeneity. In contrast, organic components in DMEM helped buffer surface charge, resulting in a more uniform negative potential [23]. Upon further membrane coating with microglial membranes, the zeta potential of ZnS@BSA became significantly more negative (Fig. S2a), owing to the negatively charged phospholipids and membrane proteins on the membrane surface. This shift confirmed successful membrane cloaking and contributed to enhanced colloidal stability. Consistently, DLS analysis revealed a stable hydrodynamic size of the ZnS@BSA@MM in ddH₂O, DMEM, and PBS without obvious aggregation (Fig. S2b). Although there was no statistically significant difference, the cell membrane exhibited characteristics of internal positive and external negative charges. This suggested that the charge properties of ZnS@BSA might enhance its adsorption onto cell membranes, facilitating membrane coating. Furthermore, atomic force microscopy (AFM) revealed no distinct adhesion force between ZnS@BSA and the microglial membrane (Fig. S3), indicating that the interaction during membrane cloaking was mainly nonspecific and likely driven by weak physical adsorption or electrostatic effects [24]. Moreover, X-ray diffraction (XRD) analysis revealed the crystal structure of ZnS@BSA (JCPDS #12–0688), revealing the main ZnS diffraction peaks (Fig. 1e).

Fig. 1.

Characterization of ZnS@BSA NPs. (a) Schematic diagram of the synthesis of ZnS@BSA nanoparticles. (b) TEM image of ZnS@BSA NPs. (c) DLS data of ZnS@BSA NPs in different solutions. (d) Zeta potential variation of ZnS@BSA in saline and culture medium. (e) XRD patterns and b-crystallite content quantification of ZnS@BSA NPs. (f) H₂S release performance of ZnS@BSA NPs at different concentrations or time using MB as the indicators. (g) The concentration changes of ABTS^•+ radicals after treatment with ZnS@BSA at different time points or concentrations. (h) Changes in the TMB concentration after treatment with ZnS@BSA at different time points or concentrations. (i) ESR spectra showing the time-dependent hydroxyl radical (·OH) scavenging activity of the ZnS@BSA nanoparticles at 0, 1, 5, and 10 min. (j)Comparison of ABTS^•+ radical scavenging activity among ZnS@BSA, curcumin, vitamin C, quercetin, and naringenin. (k) Illustration of the ROS scavenging process of ZnS@BSA.

The introduction of exogenous H₂S is promising for promoting SCI repair. As an endogenous gas transmitter, H₂S is involved in essential physiological processes and disease states, particularly in the development of central nervous system illnesses [25]. This highlights the potential of targeting H₂S-mediated signaling pathways for treating SCI. UV–vis–NIR spectra demonstrated that the methylene blue (MB) probe, which was used to detect hydrogen sulfide, exhibited increased H₂S gas generation ability with an increasing characteristic absorption peak in a concentration-dependent and time-dependent manner (Fig. 1f and Fig. S4). TEM imaging revealed the progressive degradation of ZnS@BSA nanoparticles in PBS (pH 6.5) over a 72 h period, characterized by increased surface erosion and eventual structural collapse (Fig. S5). These time-dependent morphological changes highlighted the acid-responsive nature of the nanosystem and supported its potential for sustained H₂S release under SCI-mimicking conditions. ZnS@BSA NPs required a stable particle size in the inflammatory microenvironment when targeted to the SCI site. The assessment of the change in the diameter of the ZnS@BSA NPs over time in different pH solutions confirmed their stability under pathological conditions. Moreover, ZnS@BSA remained stable in solutions of varying acidity, with a slight decrease in particle size noted in acidic environments, likely due to H₂S release (Fig. S6). Long-term incubation under acidic conditions led to a gradual decrease in nanoparticle size, indicating that ZnS@BSA possesses acid-responsive degradability (Fig. S7). To evaluate the release of H₂S from ZnS@BSA, lead acetate test paper was used, as it reacts with H₂S to form black lead sulfide (PbS) precipitates. ZnS@BSA was dispersed in PBS at different pH values to simulate the inflammatory microenvironment. The intensity of the solution color qualitatively indicated the concentration of generated H₂S, with a deeper color representing a higher concentration. Moreover, ZnS@BSA gradually degraded in the PBS solution at pH 5.5, and the color of the lead acetate test paper significantly darkened, indicating that the acidic microenvironment promoted H₂S release (Fig. S8).

2.2. Antioxidant performance of ZnS@BSA NPs

The overproduction of ROS is recognized as a key contributor to the formation of an inhibitory microenvironment post-SCI. ROS trigger oxidative stress and cytotoxic nerve excitation and lead to a severe inflammatory response [26]. H₂S, which has the ability to eliminate excess ROS, can effectively modulate the unfavorable microenvironment. The anti-inflammatory effects of ZnS@BSA NPs were extensively investigated. The free radical scavenging ability and antioxidant performance of ZnS@BSA were assessed through 2,2′-azinobis-(3-ethylbenzthiazoline-6-sulfonate) (ABTS) method. With increasing concentrations of ZnS@BSA, the color of the ABTS^•+ solution gradually lightened and eventually became clear, indicating the excellent ROS scavenging capability of ZnS@BSA. Similarly, with increasing treatment time, ZnS@BSA effectively cleared almost all the ROS (Fig. 1g and Fig. S9a). The ROS probe 3,3,5,5-tetramethylbenzidine (TMB) was also used to quantify the elimination of hydroxyl radicals (•OH) by the ZnS@BSA NPs. A gradual reduction in the characteristic absorption peak of TMB at approximately 652 nm was observed, suggesting that •OH was eliminated by the H₂S released from the ZnS@BSA NPs (Fig. 1h and Fig. S9b). ESR spectroscopy using DMPO further confirmed this, showing a marked reduction in •OH signals with ZnS@BSA treatment (Fig. 1i), demonstrating strong antioxidant activity. To clarify the ROS neutralizing capacity of ZnS@BSA, several comparisons were made using existing ROS scavengers as references [27]. Compared with natural ROS scavengers such as naringenin, curcumin, vitamin C, and quercetin, at equivalent molar concentrations, ZnS@BSA exhibited significantly stronger ROS scavenging capabilities, surpassing naringenin and curcumin (Fig. 1j). The mechanism by which ZnS@BSA alleviates oxidative stress may involve the release of H₂S, leading to the formation of sulfur-containing species such as -SH [28], which directly scavenge free radicals (Fig. 1k). The hydrophilic and eco-friendly ZnS@BSA showed potential as a substitute for natural products and was a promising ROS scavenger. In addition, under oxidative stress induced by 100 μM H₂O₂, ZnS@BSA significantly enhanced BV2 cell viability at 12 and 24 h (Fig. S11). JC-1 staining further revealed a slight inhibition of mitochondrial membrane depolarization, indicating limited early anti-apoptotic effects (Fig. S12). These findings suggested that ZnS@BSA NPs alleviated oxidative damage mainly by promoting cell survival.

2.3. Bioactivity of ZnS@BSA and its role in neuroinflammation

To evaluate the cytotoxicity of the ZnS@BSA NPs in vitro, cell viability was measured via the CCK-8 assay after exposure to various NP concentrations at different time points. At 10 μM, the ZnS@BSA NPs showed optimal internalization with 100 ± 5 % cell viability, while higher concentrations reduced viability (Fig. S13). Secondary injury offers a critical therapeutic window for the management of SCI, which is largely driven by the inflammatory cascade. Secondary injury creates a critical therapeutic window for SCI, driven by the inflammatory cascade. Microglial polarization regulates chronic inflammation, with the number of pro-inflammatory M1 microglia increasing in the acute phase to clear debris and trigger inflammation, while anti-inflammatory M2 microglia promote neural repair [29]. Additionally, to investigate the ability of ZnS@BSA NPs to modulate microglial polarization through ROS scavenging, the presence of M1 and M2 microglial cells was assessed via CD86 (M1 marker) and CD206 (M2 marker) antibody staining. Microglial cells treated with LPS exhibited a strong CD86⁺ signal, whereas those treatment with ZnS@BSA NPs presented a significant reduction in CD86⁺ signals (Fig. 2a). Similarly, the NaHS group presented reduced CD86 fluorescence, indicating that sodium hydrosulfide was effective. However, the ZnCl₂ group displayed no significant change in CD86 levels, suggesting that the reversal of M1 microglial polarization by ZnS@BSA is primarily due to the release of H₂S.

Fig. 2.

Regulatory effects of ZnS@BSA on immune cells and neuronal cells. (a) Immunofluorescence and quantification results of BV2 cells cultured in different groups under LPS conditions. M1 (CD86: red), M2 (CD206: green), and nuclei (DAPI: blue). Scale bar, 50 μm. (b) ZnS@BSA NPs regulate microglial M1 inhibition and M2 promotion through the NF-κB pathway. (c) Schematic illustration of ZnS@BSA modulating the microglial phenotype by inhibiting the LPS-activated NF-κB pathway. (d) FACS results showing intracellular ROS in BV2 cells stained with DCFH-DA after various treatments. (e) Schematic diagram of metal ion concentration detection in the CSF patients with SCI. (f) Changes in the concentrations of Zn²⁺ in the CSF of SCI patients compared with those in the CSF of non-SCI patients (lumbar puncture fluid from hydrocephalus patients). (g) The neurite length of PC12 cells cultured with different ions (Ca²⁺, Fe³⁺, Mg²⁺, Cu²⁺, Zn²⁺). Scale bar, 50 μm. (h) The neurite length of PC12 cells cultured with ZnS@BSA NPs in pathological condition. Scale bar, 50 μm. (i) Effect of microglial-conditioned medium treated with ZnS@BSA on the axon growth of PC12 cells, with a schematic on the left showing that ZnS@BSA inhibits the secretion of inflammatory mediators by microglia, thereby reducing their inhibitory effect on neuronal axon growth. The data are presented as the mean ± SD (n = 6) in (a), (b), as the mean ± SD (n = 12) in (f), and as mean ± SD (n = 5) in (d), (g), (h), (i). Statistical differences were determined by using the ANOVA with Bonferroni's multiple comparison test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ns: no significant; a.u. means arbitrary unit).

To investigate the anti-inflammatory effects of ZnS@BSA in microglia, Western blot analysis revealed that LPS stimulation activated and phosphorylated IkBα and p65 in the NF-κB pathway. However, treatment with ZnS@BSA markedly decreased the levels of activated IkBα, p65, and their phosphorylated forms, indicating strong suppression of the NF-κB pathway (Fig. 2b,Figure S14, Fig. S15). In the NaHS-treated group, there was no significant change in IkBα or p65 phosphorylation, likely due to the rapid release of H₂S into the culture medium, resulting in minimal intracellular uptake. In contrast, after being internalized intracellularly, ZnS@BSA released H₂S, thereby exerting its antioxidative stress effects. These findings suggested that ZnS@BSA NPs inhibited M1 polarization and promoted M2 microglial polarization through the NF-κB pathway (Fig. 2c).

Next, we utilized DCFH-DA probe to verify the antioxidative properties of the ZnS@BSA NPs in microglia. The ROS levels within BV2 cells were quantified on the basis of the fluorescence intensity of DCFH-DA. The H₂O₂ treatment group exhibited a significant DCFH-DA signal, indicating the induction of ROS by inflammatory mediators in microglia and a consequently strong inflammatory response. Conversely, both the NaHS and ZnS@BSA groups presented reduced ROS levels upon H₂O₂ stimulation (Fig. 2d). These findings suggested that H₂S released from ZnS@BSA NPs effectively inhibited ROS production in neuroimmune cells. Additionally, the impact of ZnS@BSA NPs on the gene expression of inflammatory cytokines in microglia was assessed. In both the ZnS@BSA and NaHS groups, the levels of proinflammatory cytokines such as TNF-α were lower, whereas the anti-inflammatory IL-4 was mildly elevated, compared to the LPS group (Fig. S16).

2.4. Functional role of ZnS@BSA in neuronal axon growth

Essential and trace elements are vital for growth, metabolism, and nervous system excitability [30,31]. Zn²⁺ plays a vital role in the spinal nervous system by regulating neuronal signal transmission, growth, repair, and synaptic plasticity. It helps maintain neuronal excitability, promote axonal regeneration, and reduce inflammation after injury. Zn²⁺ deficiency can impair spinal cord repair and regeneration, leading to neural dysfunction and affecting overall nervous system function. However, the changes in the Zn²⁺ concentration following SCI remain unclear. Therefore, in the early stage of this study, cerebrospinal fluid samples were collected from six patients (one with brain edema and five with traumatic SCI, all approved by the ethics committee). The differences in the concentrations of the metal ions (Zn²⁺, Fe³⁺, Mg²⁺, Ca²⁺, and Cu²⁺) present in the samples were measured via inductively coupled plasma (ICP) analysis to verify the necessity of supplementing exogenous Zn²⁺ for neuronal repair after SCI (Fig. 2e and Table S1). The results indicated that Cu²⁺ and Mg²⁺ were undetectable in both the sham and SCI CSF samples. However, the concentrations of Ca²⁺ and Fe³⁺ increased post-SCI, leading to pathological reactions such as calcium overload and oxidative stress, ultimately causing neuronal necrosis and apoptosis. Conversely, the concentration of Zn²⁺ was lower in the SCI group than in the sham group (Fig. 2f and Fig. S17). The decrease in the Zn²⁺ concentration after SCI may be related to altered expression of zinc transporters (ZnTs). ZnT1 and ZnT3, which are responsible for Zn²⁺ efflux in neurons, exhibit decreased expression during the acute phase of SCI, leading to Zn²⁺ homeostasis disruption and ultimately resulting in a reduction in the Zn²⁺ concentration [32]. The reduced Zn²⁺ concentration was related to changes in the excitability of neuronal cell membranes, while exogenous zinc supplementation potentially promoted the functional recovery of neurons after SCI [33]. To confirm the effects of various metal ions on neuronal axon growth, we cultured undifferentiated PC12 cells and treated them with physiological concentrations of Zn²⁺, Fe³⁺, Mg²⁺, Ca²⁺, and Cu²⁺. Fluorescence staining demonstrated that Zn²⁺ effectively induced neuronal axon differentiation, with axon lengths reaching approximately 40 μm (Fig. 2g). Therefore, the use of Zn²⁺ as a substrate to synthesize related nanomedicines holds promise for promoting neuronal regeneration and repair following SCI.

To investigate the influence of ZnS@BSA on axon length, PC12 cells with low differentiation were exposed to PBS, LPS, LPS + NaHS, LPS + Zn²⁺, or LPS + ZnS@BSA. LPS treatment under pathological conditions suppressed neurite outgrowth to about 5 μm. In contrast, the neurite lengths in the LPS + ZnCl₂ (approximately 20 μm) and LPS + ZnS@BSA (approximately 40 μm) groups remained significantly longer than those in the other groups (Fig. 2h). Although NaHS-derived H₂S reduced inflammation, it did not promote axonal extension. Collectively, these findings highlight the role of ZnS@BSA NPs in supporting neurite growth and creating a conducive environment for neural cell proliferation and axon development. To further validate these results under more physiologically relevant conditions, we conducted supplementary experiments using primary cortical neurons. ZnS@BSA treatment under both normal and LPS-induced inflammatory conditions significantly enhanced axonal outgrowth, as confirmed by F-actin immunofluorescence staining (Fig. S18). These results suggested that ZnS@BSA exerted robust neuroregenerative effects not only in model cell lines but also in primary neurons under both physiological and pathological conditions.

During the chronic phase of SCI, a gradual increase in microglial levels was observed after 2 weeks. The release of inflammatory mediators by M1 microglia disrupted axonal growth in neurons. Despite the induction of nerve growth factors promoting axonal growth, persistent low-level inflammation hindered the repair process of SCI. ZnS@BSA NPs successfully prevented the polarization of microglia into the M1 phenotype and attenuated the secretion of proinflammatory cytokines. Further investigations were warranted to determine whether ZnS@BSA NPs facilitate neuron axon growth by modulating M1/M2 microglial polarization. Therefore, microglia were cultured under various conditions with LPS treatment. After centrifugation, the cell supernatant was utilized to stimulate PC12 cells to observe axon length (Fig. 2i). Notably, the neurite length in the Zn²⁺ (around 47 μm) and ZnS@BSA (around 120 μm) groups was significantly longer compared to the other groups. Intriguingly, the culture medium derived from normal microglia promoted axonal growth (approximately 50 μm), suggesting the beneficial role of homeostatic microglia under physiological conditions in promoting axonal growth and neuronal proliferation. However, under inflammatory stimuli, microglial polarization was altered, leading to axonal growth inhibition (approximately 10 μm) due to factors in the culture medium. However, the addition of Zn²⁺ ameliorated this axonal inhibition effect. The optimal promotion of axonal growth was achieved only when Zn²⁺ was combined with anti-inflammatory agents. These findings collectively underscored the pivotal role of ZnS@BSA NPs in nerve formation through the modulation of microglial polarization in SCI.

2.5. Preparation and cellular internalization mechanism of ZnS@BSA@MM

Direct injection of nanoparticles may lead to immune system recognition, making it difficult to target the SCI site. Therefore, biological membranes have been used to disguise nanoparticles, optimizing disease theranostics by reducing side effects, extending circulation time, and improving targeting ability [17,34]. For example, microglial membrane (MM)-coated nanoparticles can evade immune attacks via self-markers [19]. The microglial cell membranes were prepared via ultrasound-assisted membrane disruption and centrifugation and then filtered through a series of filters with decreasing pore sizes, resulting in the repetitive destruction and self-assembly of ZnS@BSA@MM (Fig. 3a). TEM image showed an obvious core–shell structure of ZnS@BSA@MM, which is in good agreement with reported membrane-coated NPs (Fig. 3b) [35]. To verify the successful encapsulation of the nanomaterials by the cell membranes, ZnS@BSA@MM was separately stained with Zinquin ionophore (Zn²⁺ fluorescence probe) and WSP-1 (H₂S fluorescence probes). Flow cytometry analysis indicated that the cell membrane effectively encapsulated ZnS@BSA, demonstrating the ability to release H₂S and Zn²⁺ (Fig. 3c). To verify the successful encapsulation of ZnS@BSA by microglial membranes, Western blot analysis revealed significant expression of cell membrane proteins (CD9 and CD63) across the above groups. IBA1 was not significantly expressed in any group, indicating that the extracted cell membranes originated from resting microglia. However, β-actin levels were lower in both the microglial membrane and ZnS@BSA@MM groups, likely because β-actin was expressed mainly in the cytoplasm and rarely in the cell membrane (Fig. 3d). These data suggested the successful construction of cell membrane-encapsulated ZnS nanomaterials.

Fig. 3.

Preparation of ZnS@BSA@MM and study of its endocytosis mechanism study. (a) Schematic diagram of ZnS@BSA particles coated with microglial cell membranes constructed via the extrusion method. (b) TEM images of microglial cell membranes and ZnS@BSA@MM. Scale bar, 500 nm. (c) Detection of membrane-coated nanoparticle efficiency via flow cytometry. (d) WB detection of membrane-associated protein expression in the ZnS@BSA@MM. (e) Enrichment efficiency of different concentrations of membrane (Dil-labeled) coated with ZnS@BSA (Cy5.5-labeled) in microglial cells. (f–h) Immunofluorescence imaging analysis of the endocytic pathway of ZnS@BSA and ZnS@BSA@MM NPs. The scatter gram represents the colocalization analysis of Zn²⁺ or H₂S fluorescence from ZnS@BSA@MM with endosomes or the Golgi apparatus after endocytosis. The scatter plot shows an overall diagonal distribution, indicating colocalization between the red fluorescence (Rab5, Rab7, or Golgi apparatus) and ZnS@BSA@MM labeled green fluorescence. The early endosome marker Rab5 (f), the late endosome marker Rab7 (g), and the Golgi apparatus marker GM130 (h) were observed for 0.5 h and 6 h after the BV2 cells were exposed to the NPs.

To assess the intracellular distribution of ZnS@BSA@MM NPs in microglia, Dil-stained microglial cell membrane coatings and Cy5.5-labeled ZnS@BSA NPs were used. After 6 h of cellular uptake, immunofluorescence staining revealed co-localization of red fluorescence (ZnS@BSA NPs) and orange-red fluorescence (cell membrane) both on the microglial cell membrane and within the cytoplasm (Fig. 3e). Notably, increasing concentrations of ZnS@BSA@MM NPs did not induce organelle swelling, cell membrane rupture, or cytoplasmic and nuclear degradation. These findings demonstrate the excellent biocompatibility of ZnS@BSA@MM NPs with microglia, highlighting their potential in biomedical applications. During the process of the internalization of ZnS@BSA@MM NPs, we explored the localization of H₂S and Zn²⁺ within organelles via the use of cellular markers associated with endocytosis pathways, including Rab5, Rab7, and GM130. Microglia were exposed to 10 μM ZnS@BSA@MM for 30 min and 6 h. Initially, the NPs colocalized with Rab5 within 30 min, indicating entry into early endosomes with an internalization efficiency exceeding 70 % (Fig. 3f). Intriguingly, the colocalization of Rab7, a marker of late endosomes and autophagosomes, notably increased after 6 h, suggesting that Rab7 was transported to late endosomes following endosome maturation and subsequently localized to larger organelles (Fig. 3g). Furthermore, we observed colocalization with GM130, indicating the transfer of some NPs to the Golgi apparatus. Zn²⁺ also co-localizated with the Golgi apparatus, suggesting that ZnS@BSA@MM NPs might exit cells via the endoplasmic reticulum-Golgi pathway after entering late endosomes, potentially affecting other neuronal cells (Fig. 3h). These findings suggested a potential association with exocytosis due to colocalization with ZnS@BSA@MM NPs.

ZnS@BSA@MM has been internalized via the endosomal pathway, as shown by colocalization with early (Rab5) and late endosomes (Rab7), where acidic conditions induce partial degradation and release of H₂S and Zn²⁺. Importantly, significant colocalization with the Golgi apparatus indicated that a portion of the nanoparticles escape lysosomal degradation by trafficking to the Golgi, thus avoiding breakdown and enabling sustained intracellular release [36].

2.6. In vivo therapeutic effects of ZnS@BSA@MM in SCI model mice

To assess the in vivo therapeutic effects, ZnS@BSA@MM NPs were intravenously administered to mice via spinal cord transection. Sham-operated mice and those that received SCI, ZnS, microglial cell membrane (MM), or ZnS@BSA@MM NPs served as controls. Over a 14-week period, the mice in the SCI, ZnS@BSA, and MM groups exhibited paralysis, while those in the ZnS@BSA@MM group showed signs of recovery in motor function (Fig. 4a). Digital stereomicroscopy was employed to perform microsurgical procedures on C57BL/6J mice to establish a spinal cord transection injury model (Fig. 4b). The Basso Mouse Scale (BMS) was used to evaluate the locomotor behavior of the mice. The BMS decreased to 0 in all SCI model mice by post-injury day 0, confirming the successful establishment of the SCI model. By post-operation weeks 14, the BMS score (4.50) was notably greater in the ZnS@BSA@MM group than in the PBS (0.75), ZnS@BSA (1.75), and MM (2.20) groups (Fig. 4c and Video S1), which was consistent with previous observations. Furthermore, the body weights of the mice in the ZnS@BSA@MM group were greater than those in the other groups, suggesting that the malnutrition resulting from hindlimb paralysis induced by SCI was ameliorated (Fig. S19). The swimming assessment commenced 14 weeks post-SCI. Mice were placed in a transparent rectangular pool (50 × 30 × 10 cm) filled with 30 °C warm water, with a depth sufficient to prevent touching the pool's bottom. Their movements were recorded by a camera mounted on a rail along the pool's length. In the mice treated with systemic ZnS@BSA@MM, significant improvements in joint angle dispersion and motor function restoration of the ankle and knee joints were observed, particularly at the 14-week mark after SCI (Fig. 4d and Fig. S20).

Fig. 4.

Targeting efficiency and therapeutic application of ZnS@BSA@MM in SCI mouse models for motor function recovery. (a) Schedule of SCI mice treated with ZnS@BSA@MM NPs. IOCV: Injection via the caudal vein, t.i.w: three times a week. (b) Procedure for establishing the SCI mouse model. (c) Lower limb motor function over time was assessed via the BMS. (d) Swimming macroscopic images of the mice in each group. (e) Footprint test results of the mice in each group. (f) Muscle-electrophysiological testing of both hind limbs in mice. (g) Enrichment efficiency of fluorescent particles in various organs of SCI model mice across different groups. Abbreviations of organs: spinal cord (SC), heart (HR), liver (LVR), spleen (SPL), lung (LNG), and kidney (K). (h) Differences in the distribution of fluorescent particles in injured spinal cord tissue across different groups. (j) H&E staining of spinal cord tissues and bladder tissue from each group, blue arrow: muscular layer of urinary bladder. The data are presented as the mean ± SD (n = 9) in (c). Statistical differences were determined by using the ANOVA with Bonferroni's multiple comparison test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001).

Footprint analysis at 14 weeks post-injury revealed distinct patterns among the groups. In the sham group, the hind footprint consistently showed stable grounding and the longest forefoot stride length. Conversely, hind footprints in the PBS, ZnS@BSA, and MM groups presented a reduced print area and a significant absence of stride amplitude, indicating a notable decline in motor function. However, treatment with ZnS@BSA@MM resulted in a substantial improvement in the hind footprint area and an extension of the forefoot stride length, indicating a significant recovery in motor function (Fig. 4e). Additionally, electrophysiological research revealed that the administration of ZnS@BSA@MM NPs led to significantly shorter latency periods and significantly greater amplitudes of motor-evoked potentials (MEPs) at the 14-week post-SCI mark (Fig. 4f).

2.7. ZnS@BSA@MM NPs enriched at the SCI site

In the in vivo experiments, tail vein injection of ZnS@BSA resulted in minimal accumulation at the SCI site. However, ZnS@BSA coated with microglial cell membranes (ZnS@BSA@MM) significantly enhanced functional recovery. These findings suggested that microglial membrane encapsulation improved the enrichment of Zn²⁺ and H₂S at the injury site, increasing their therapeutic efficacy. To verify the biodistribution of ZnS@BSA@MM NPs labeled with Cy5.5, in vitro fluorescence was measured 24h following intravenous injection. Small animal fluorescence imaging revealed fluorescence signals in the SCI region in both the ZnS@BSA@MM and MM groups, indicating that the membrane coating could be delivered to the injury site (Fig. 4g). In contrast, ZnS@BSA primarily accumulated in the liver region, with partial recognition by splenic tissue and enrichment in the spleen area. Therefore, ZnS@BSA NPs were unable to act on the SCI area to promote neural function recovery, which was consistent with the BMS scoring results. Microglial cell membrane coating could reduce the uptake of NPs by innate immune phagocytes in the liver and spleen, allowing them to circulate through the bloodstream and act on the SCI area. The observations revealed substantial fluorescence nanoparticle enrichment in the SCI region in both the cell membrane and the membrane-encapsulated NP groups (Fig. 4h). To further verify the spatiotemporal targeting behavior, we performed in vivo fluorescence imaging at 0, 6, 12, 24, and 72 h post-injection of Cy5.5-labeled ZnS@BSA@MM (Fig. S21). The fluorescence signal peaked at 6–12 h and remained detectable for up to 72 h, indicating rapid accumulation and long-term retention at the injury site. These results confirm the efficient targeting capability and sustained presence of the nanoparticles in the injured spinal cord. To assess the restoration of Zn²⁺ levels, we quantified Zn²⁺ concentrations in spinal cord tissues at 7 days post-injury (Fig. S22). In SCI mice, Zn²⁺ levels dropped below the detection limit, indicating severe depletion. ZnS@BSA@MM treatment significantly increased the Zn²⁺ concentration, although it was not fully restored to normal levels, suggesting partial recovery of Zn²⁺ homeostasis. Additionally, Flow cytometry analysis revealed that 24 h after tail vein injection of ZnS@BSA@MM, the fluorescence intensities of Zn²⁺ and H₂S at the SCI site of the mice were significantly higher than those in the control group, confirming the effective targeting and therapeutic action of ZnS@BSA@MM in the SCI region (Fig. S23). In addition, HE staining of various organs indicated that intravenous injection of ZnS@BSA@MM did not cause organ toxicity or damage (Fig. S24). These findings underscored the enhanced targeted recognition capacity of ZnS@BSA@MM NPs due to the presence of the cell membrane, consequently amplifying the efficacy of ZnS within the SCI site.

2.8. ZnS@BSA@MM NPs affect in vivo neural regeneration

At the end of the 14-week period, all mice were sacrificed for histological analysis of the regenerated nerve tissue. Wound sites were still evident in the SCI and ZnS@BSA groups, whereas the MM group showed prominent cavities and scar tissue. In contrast, these structural defects were notably diminished in the ZnS@BSA@MM group (Fig. 4i). The H&E-stained sections also confirmed that the ZnS@BSA@MM NPs promoted SCI repair, as evidenced by the absence of significant defects or scarred areas compared with those in the other groups (Fig. 4j). The degree of detrusor muscle atrophy in mice after SCI is an important indicator for assessing bladder function. The thickness of the detrusor muscle and changes in muscle fibers serve as evaluation parameters for neural function recovery post-SCI. H&E staining revealed that in the control group, the detrusor muscle was disorganized and atrophied muscle fibers post-surgery, with a gradual worsening over time, along with extensive infiltration of connective tissue between muscle bundles. In contrast, there was a gradual restoration of detrusor muscle morphology in the ZnS@BSA@MM group at 14 weeks post-surgery, with no significant increase in the amount of connective tissue. The cross-sectional area of the detrusor muscle fibers gradually decreased over time in the control group, while the decrease in cross-sectional area was slower in the experimental group, and the percentage of connective tissue increased gradually over time (Fig. 4j).

Spinal cord regeneration encompasses various processes, including neuronal maturation, scar formation, and synaptic reorganization. Astrocytes serve as crucial components in CNS function, providing vital nutrients and safeguarding neurons. Under inflammatory conditions, astrocytes are activated and form glial scars in SCI, which act as chemical impediments to neuronal regeneration [37]. In a typical physiological setting, endogenous neural stem cells (eNSCs) remain quiescent, but they differentiate into astrocytes or neurons after SCI. Despite their potential to promote spinal cord regeneration, the presence of glial scars at the lesion site often obstructs the activity of NSCs [38]. Considering the promising potential of ZnS@BSA@MM NPs for axon regeneration in vitro, a dual-staining approach with β3-tubulin and GFAP was used to evaluate the distribution of neurons and astrocytes at the lesion sites (Fig. 5a). GFAP⁺ cells were primarily concentrated around the lesion border in the SCI and ZnS@BSA groups, with few neurons present. The MM group exhibited proper lesion healing but retained a significant cavity area. In contrast, the neuron count at the lesion site was markedly greater in the SCI group than in the SCI, ZnS@BSA, and MM groups. The neurons were evenly distributed, with moderate astrocytic presence throughout the spinal cord. These findings highlight the potential of ZnS@BSA@MM NPs to optimize the neuronal-astrocytic balance, promoting spinal cord regeneration.

Fig. 5.

ZnS@BSA@MM NPs promoted synaptic regeneration in vivo. (a) Immunofluorescence staining and quantification of β3-tubulin⁺/GFAP⁺ cells, NeuN⁺ cells, and SYP⁺ cells. β3-Tubulin is a neuronal marker indicating neuronal characteristics. GFAP is an astrocyte marker indicating glial cell characteristics. NeuN is a neuronal marker that specifically identifies mature neurons. SYP (Synaptophysin) is a synaptic marker that represents synaptic activity and the presence of neural networks. (b) Changes in neuron-related protein expression across different groups. MAP2: marker for mature neurons. ChAT: marker for motor neurons. Nestin: maker for neural stem cells. The data are presented as the mean ± SD (n = 5) in (a) and (b). Statistical differences were determined by using the ANOVA with Bonferroni's multiple comparison test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ns: no significant; a.u. means arbitrary unit).

Compared with the other experimental groups, the localized accumulation of ZnS@BSA@MM nanoparticles promoted the neural differentiation of eNSCs, this led to a higher number of NeuN⁺ cells being distributed within the lesion sites (Fig. 5a). Furthermore, synapses are critical for nerve repair and axonal extension. Zn²⁺ functions as an important signaling molecule at synapses, modulating protein synthesis in dendrites. Following treatment with the ZnS@BSA@MM nanoparticles, there was an evident increase in the synaptic distribution at the lesion sites (Fig. 5a). ChAT⁺ cells, identified as motor neurons in spinal cord tissue, were confirmed by immunofluorescence in the distal spinal cord of the ZnS@BSA@MM and MM groups. These findings indicated that the recovery of motor function in mice is closely related to the number of motor neurons, which was consistent with the trends in the BMS score (Fig. S25). Nestin⁺ cells, identified as neural stem cells in the SCI area, have the potential to differentiate into neuronal cells. The greater the distribution of neural stem cells after SCI is, the better the regenerative effects on nerve damage. Immunofluorescence analysis revealed a significant increase in the number of nestin-positive cells in the ZnS@BSA@MM group compared with those in the other groups. Additionally, the lesion area at the injury center was reduced compared with that in the other groups (Fig. S25).

Similarly, SCI tissues were collected for protein detection. Western blot results confirmed that the expression of the NeuN, Map2, and Chat proteins in the spinal cords of the mice in the ZnS@BSA@MM group was significantly greater than that in the control group. These findings indicated that the membrane-coated targeted zinc sulfide particles effectively promoted neuronal regeneration after SCI and induced the proliferation and differentiation of motor neurons and mature neurons (Fig. 5b). Moreover, via RNA transcriptome sequencing technology, the effects of ZnS@BSA@MM NPs on neurons and the inflammatory response in the early stage of SCI (14 days post-injury) were analyzed. The heatmap illustrated the expression profiles of the significantly differentially expressed genes (Fig. 6a). In the ZnS@BSA@MM group, inhibitory neurotransmitter-related genes (gabrb2, oprm1, glra1) were upregulated. Gabrb2 encodes the β2 subunit of the GABA A receptor, which reduces neuronal excitability. Oprm1 encodes the Mu-opioid receptor, which alleviates pain and reduces neuropathic pain post-SCI. Glra1 encodes the glycine receptor α1 subunit, facilitating chloride influx and reducing neuronal excitability. Transcriptome sequencing suggested that ZnS@BSA@MM alleviated neuronal hyperexcitability during the acute and subacute phases of SCI by increasing inhibitory neurotransmitter expression [39], protecting residual neurons and controlling pain and spasms. Additionally, the expression of inflammation-related genes (kng1, cxcl1) was downregulated. Kng1 encoded a high-molecular-weight kininogen that activated leukocytes and amplified inflammation. Cxcl1 attracted neutrophils to inflammation sites. The downregulation of these genes indicated that ZnS@BSA@MM reduces immune cell infiltration and activation, decreases inflammation and prevents secondary damage.

Fig. 6.

The mechanism by which ZnS@BSA@MM repairs SCI was determined via RNA sequencing analysis. (a) Heatmap of the differentially expressed genes. (b) Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the functional annotations of the spinal cord after treatment with ZnS@BSA@MM. (c) Gene Ontology analysis of the functional annotations of the differentially expressed genes. (d, e) ZnS@BSA@MM may inhibit the inflammatory response in the injured spinal cord through the NF-κB signaling pathway. (f) ZnS@BSA@MM inhibited the secretion of pro-inflammatory cytokines. (g) Schematic diagram of the in vivo treatment of SCI mice with ZnS@BSA@MM. The data are presented as the mean ± SD (n = 5) in (e) and (f). Statistical differences were determined by using the ANOVA with Bonferroni's multiple comparison test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ns: no significant; a.u. means arbitrary unit).

The KEGG results confirmed that ZnS@BSA@MM activated signaling pathways such as the ribosome, cytokine-cytokine receptor, neuroactive ligand-receptor, and cAMP pathways in the spinal cord (Fig. 6b). These findings suggested that the membrane-coated nanoparticles primarily regulate the activities of neural and immune cells through surface receptor binding and neurotransmitter release. Gene Ontology analysis revealed that the differentially expressed genes were related to immune cell proliferation and activation, neurotransmitter secretion, and ribosome formation, reflecting the multifunctional role of ZnS@BSA@MM (Fig. 6c). To validate the regulatory mechanism of inflammation during the acute phase of SCI, Western blot analysis revealed that ZnS@BSA@MM effectively inhibited the NF-κB pathway, as evidenced by the suppression of p65 and IκB protein phosphorylation (Fig. 6d–S26 and 6e). Additionally, the secretion of inflammatory factors such as IL-1β and TNF-α was significantly lower in the SCI group than in the other groups (Fig. 6f). Therefore, early inhibition of inflammation in SCI could reduce detrimental factors, creating a favorable environment for spinal cord regeneration. Overall, ZnS@BSA@MM NPs created multifunctional microenvironments that promoted neuronal homeostasis while simultaneously inhibiting neural inflammation. This dual action supported spinal cord regeneration with reduced scarring and enhanced functional recovery (Fig. 6g).

3. Conclusion

In summary, we developed a bioactive zinc-based nanoplatform (ZnS@BSA) for promoting neuronal axon regeneration and inhibiting neuroinflammatory responses. On the one hand, ZnS@BSA gradually released H₂S, which scavenged ROS, regulated microglial polarization, and inhibited the release of inflammatory cytokines. H₂S is known to inhibit NF-κB signaling by mediating S-sulfhydration of key cysteine residues in signaling proteins, thereby suppressing their phosphorylation and preventing p65 nuclear translocation [40,41]. Consistent with this mechanism, ZnS@BSA@MM treatment markedly reduced the phosphorylation levels of p65 and IκBα in our study, indicating that its anti-inflammatory effect is primarily mediated by H₂S-induced S-sulfhydration and subsequent NF-κB inhibition. On the other hand, ZnS@BSA degradation released Zn²⁺, compensating for Zn²⁺ deficiency after SCI and restoring the excitatory/inhibitory neuronal balance. Furthermore, combined treatment with ZnS@BSA and microglial cell membranes (ZnS@BSA@MM) achieved more satisfactory therapeutic outcomes for spinal cord injury (SCI), effectively regulating the pathological microenvironment of SCI through Zn²⁺ and H₂S infiltration. The dual strategy significantly improved the pathological changes at different stages of SCI. In the acute phase, it reduced the infiltration of activated immune cells such as macrophages and microglia, suppressing the cytokine storm.

In the subacute phase, the ratio of excitatory to inhibitory neurons was balanced, and in the chronic phase, Zn²⁺ was continuously supplemented to promote neuronal axon regeneration, restoring motor neuron function. These therapeutic effects were largely attributed to the microglia membrane coating, which provided inflammation-targeting and immune-evasive properties, enhancing nanoparticle accumulation and retention at the injury site. Combined with the pH-responsive sustained release of H₂S and Zn²⁺ from the ZnS@BSA core, the platform effectively facilitated ROS scavenging, microglial polarization, and neural repair. Our work emphasized the application of a bioactive ZnS@BSA@MM nanoplatform with “dual” efficiency in SCI treatment. This innovative research presents a promising injectable nanoparticle-based therapy for tissue repair engineering, highlighting its potential in the treatment of SCI.

3.1. Experimental section

Synthesis of ZnS@BSA NPs: 40.0 mg of BSA (BSA, B2064-10G, SIGMA) was dissolved in 6.0 mL of deionized water. Subsequently, 1.0 mL of Zn(CH₃COO)₂ solution (26 mg/mL) was added, and the mixture was stirred. Then, 3.0 mL of NaHS solution (50 mg/mL) and 1.0 mL of NaOH solution (0.1 M) were added to the mixture. Glutaraldehyde solution (20 μL) was added into the mixture to induce crosslinking. The reaction was allowed to proceed for 4 h at room temperature. After the reaction, the mixture was dialyzed (MW: 8000–14000) against deionized water for 4 h to obtain ZnS@BSA nanoparticles (ZnS@BSA NPs).

Characterization: The morphologies of the ZnS@BSA NPs were characterized by transmission electron microscopy (TEM, Tecnai F20). The crystal structure and surface chemical composition of ZnS@BSA were measured via X-ray diffraction (XRD, Panalytical Empyrean). The diameter and potential of the ZnS@BSA NPs were detected by a Zeta sizer Nano-ZS (Malvern Instruments, UK).

Hydrogen sulfide (H₂S) release from ZnS@BSA NPs: H₂S generation was qualitatively analyzed using MB probes. Initially, 200 μL of FeCl₂ (1 mg/mL) and 200 μL of H₂O₂ (10 mM) were added to 10 mL of deionized water. Then, 50 μL of MB (0.5 mM) was added to the solution. Next, ZnS@BSA NPs of various concentrations and different degradation times were added. The reduction in the absorbance of MB at approximately 664 nm indicated the release of H₂S by ZnS@BSA NPs. Additionally, H₂S release in cells was quantitatively analyzed via the WSP-1 probe (50 μM, 1 mL), and the fluorescence intensity was measured via a microplate reader (Ex = 465 nm, Em = 515 nm). Additionally, a piece of lead acetate ((CH₃COO)₂Pb) test paper was placed on top of the small Petri dishes. After reacting for 3 h at 37 °C, the color change of the test paper was observed.

Quantitative analysis of the elimination of •OH: The decomposition of •OH by ZnS@BSA NPs with H₂O₂ catalysis was detected via the TMB colorimetric method. First, 4 μL of TMB (80 mM) was added to 1 mL of ZnS@BSA solution containing 100 μM H₂O₂. The absorbance change at 654 nm was monitored to assess •OH elimination. Additionally, changes in the ABTS^•+ concentration were measured via the ABTS colorimetric method. The antioxidant capacity of ZnS@BSA NPs was evaluated by observing the absorbance change at 414 nm with a UV–vis spectrophotometer.

Microglia cell behavior induced by ZnS@BSA NPs: The BSA-related nanomaterials were filtered through a 0.2 μm filter to remove bacteria and then co-cultured with macrophage cells. BV2 microglial cells (2.0 × 10⁴ cells per well) were purchased from Wuhan Procell Life Science & Technology Co., Ltd., seeded on a 24-well plate with DMEM containing 10 % FBS, and treated with LPS (100 ng mL⁻¹) for a specified time. After stimulation, the samples were stained with CD86 (Abcam, ab239075, USA), CD206 (Abcam, ab125028, USA), and DAPI (Invitrogen, 62247, USA). Microglial polarization was observed via confocal laser scanning microscopy (CLSM, LSM 800 with Airyscan, ZEISS, Germany).

For Western blot analysis, microglial cells were cultured in a 6-well plate and treated as described earlier. Proteins were separated by SDS-PAGE and transferred to a PVDF membrane. The membranes were then incubated with primary antibodies (IκBα, phospho-IκBα, NF-κB p65, and phospho-NF-κB p65) and a secondary antibody (goat anti-rabbit IgG H + L HRP). The protein bands were detected via enhanced chemiluminescence reagents (Bi Yuntian, Wuxi, China) and quantified via ImageJ software. The levels of TNF-α, IL-1β, and IL-4 secreted by microglial cells were quantified via ELISA kits (mlbio, Shanghai, China).

Reactive oxygen species scavenging ability of ZnS@BSA in BV2 cells: BV2 microglial cells were incubated with 100 μM H₂O₂ for 12 h. After stimulation, the cells were stained with 20 μM 2′,7′-dichlorofluorescein diacetate (DCFH-DA) for 30 min to qualitatively analyze the ROS scavenging efficiency.

Detection of metal ions in cerebrospinal fluid: Cerebrospinal fluid (CSF) was collected from patients with varying degrees of spinal cord injury (n = 5) in the Department of Orthopedics at the Second Affiliated Hospital of Soochow University, and lumbar puncture drainage fluid from patients with cerebral edema was used as the control. Each patient had acute neurological damage, resulting in either complete or incomplete paralysis. There were three cases of lumbar injury and two cases of cervical injury, with one case of complete injury and four cases of incomplete injury. CSF from a 72-year-old female patient with hydrocephalus but without SCI, collected via lumbar puncture, was used as the sham group. The study received ethical approval. CSF samples from three SCI patients were obtained intraoperatively during dural incision or decompression procedures, where 3–5 mL of CSF was collected using an epidural catheter and centrifuged to remove cellular debris. The supernatant was then stored at −80 °C. For the non-SCI group, CSF samples were collected via lumbar cistern drainage from patients with cerebral edema. A catheter was placed at the L3-L4 interspace, and approximately 450 mL of CSF was drained daily; samples for this study were taken from the surplus fluid after routine clinical testing. The absolute concentrations of different ions (Zn²⁺, Fe³⁺, Mg²⁺, Ca²⁺, and Cu²⁺) were detected via inductively coupled plasma-optical emission spectroscopy (ICP-OES, Avio 200) to assess changes in ion concentrations in the CSF.

Proliferation and axon spreading of PC12 cells with ZnS@BSA nanoparticles: Low-differentiated PC12 cells were prepared. PC12 cells (2.0 × 10⁴mL⁻¹) were cultured in 24-well plates with DMEM containing 10 % FBS, treated with different concentrations of ZnS@BSA nanoparticles, and then stained with F-actin (FITC-phalloidin, Invitrogen, B1370, USA) and DAPI at a suitable concentration for further experiments. PC12 cells were treated with ZnCl₂, NaHS, and ZnS@BSA NPs, and then stained with FITC-phalloidin (Invitrogen, B1370, USA) and DAPI. The PC12 cells were incubated with LPS (100 ng mL⁻¹) for 24 h, The cells were subsequently stained with FITC-phalloidin to detect changes in neurite length under pathological conditions. The change in neurite length change in the PC12 cells was visualized via CLSM.

RNA extraction and RT-qPCR: RNA was extracted following the TRIzol reagent (Invitrogen) protocol with slight modifications. cDNA synthesis was carried out via the RT reagent kit with gDNA Eraser. Quantitative PCR (qPCR) was performed via the SYBR Green Master Mix Kit on a LightCycler 480 instrument. GAPDH served as an endogenous control.

Preparation of ZnS@BSA@MM: Using BV2 microglial cell lines as the source of cell membranes, BV2 cells were collected and washed twice with saline. The cells were then resuspended in a hypotonic solution (a mixture of 0.25 M sucrose, 1 mM EDTA, 20 mM HEPES-OH, and protease inhibitors). The cells were subjected to ultrasonic treatment in an ice bath via an ultrasonic processor (ultrasonication for 3 s, intervals of 7 s, 25 times). The mixture was subsequently centrifuged at 4000 rpm for 10 min to remove larger amounts of cell debris. The supernatant was collected and further centrifuged at 14800 rpm for 20 min. The pellet was collected, washed twice with saline, and centrifuged again at 14800 rpm for 20 min. The cell membrane pellet was resuspended. The cell membrane suspension was mixed with 400 mM ZnS@BSA and sonicated in an ice bath for 2 h. ZnS@BSA@MM was obtained by passing through 200 nm polycarbonate porous membranes 15 times and storing them in saline containing protease inhibitors. The effectiveness of the cell membrane-coated nanoparticles was evaluated via transmission electron microscopy and flow cytometry. Western blot (WB) analysis was employed to detect the protein expression of CD9, CD63, IBA-1, and β-actin in microglial cells, cell membranes, and ZnS@BSA@MM.

Fluorescence analysis of ZnS@BSA@MM: Cy5.5-labeled ZnS@BSA and Dil-labeled microglial cell membranes were used to prepare fluorescent ZnS@BSA@MM nanoparticles. These cells were co-incubated with microglial cells to observe their phagocytic efficiency at various concentrations of ZnS@BSA@MM. Although Cy5.5 and Dil-labeled NPs (red and orange-red) can be detected in cells, it is also necessary to evaluate the intracellular transport pathways and organelle distribution of the nanoparticles. The antibodies utilized included anti-EEA1 (ab2900; Abcam, Cambridge, MA, USA) for early endosomes, anti-Rab7 (ab126712; Abcam) for late endosomes, and anti-GM130 (ab169276; Abcam) for the Golgi apparatus. A goat anti-rabbit IgG H&L (Alexa Fluor 488, green) secondary antibody (ab150077; Abcam) was used to pinpoint the location of ZnS@BSA@MM NPs in cellular organelles. All the samples were examined via confocal microscopy.

Animal experiments of spinal cord injury: A mouse spinal cord transection model was employed to assess tissue regeneration [42]. The Ethics Committee of the Second Affiliated Hospital of Soochow University approved the animal experiments. Forty-five six-week-old female C57BL/6J mice (20 ± 1 g, provided by Soochow University Experimental Animal Center) were divided into five groups. The mice that received ZnS@BSA, MM, or ZnS@BSA@MM via tail vein injection were classified into the ZnS@BSA, MM, or ZnS@BSA@MM groups, respectively. Mice with transection defects that received normal saline were designated the SCI group, while those without SCI formed the sham group. The nanoparticle solution was sterilized by filtration prior to use.

The mice were anesthetized with 1 % pentobarbital sodium (125 μL/20 g) via intraperitoneal injection. A 1.5 cm incision was made at the T9-T10 level, exposing the spinous process and vertebral plate. A surgical blade was used to create a complete transverse injury at the T9 spinal cord level. The muscles and skin were sutured in layers, and each mouse received intramuscular antibiotics for 3 days post-surgery. Following SCI, either ZnS@BSA, MM, ZnS@BSA@MM nanoparticles (150 μL, 2 mg/mL), or normal saline (150 μL) were administered via tail vein injection. Injections were performed three times a week, continuing until week 14.

Function recovery and footprint analysis in SCI mice: Hindlimb recovery was evaluated via the Basso Mouse Scale (BMS) score, which ranges from 0 to 9 and measures changes in hind limb locomotion. Weekly BMS scores were recorded for the mice over a 14-week period. Additionally, the movement of the treated groups was videotaped, capturing hind limb movement and weight-bearing ability while walking on a horizontal plane. The observers, who were blinded to the SCI conditions of the mice, conducted and recorded the evaluations five times. All observers received prior training to ensure consistency in scoring. Before Footprint assessment, the mice were allowed to move freely in an open field for 5 min. For footprint analysis, the forelimbs were dipped in blue ink, and the hindlimbs were dipped in red ink. The mice were then guided to walk in a straight line on a sheet of white paper.

Louisville swimming scale (LSS) in SCI mice: At 14 weeks post-SCI, swimming performance was evaluated via the LSS. Each animal underwent a single 4-min swimming session per assessment [43]. Two impartial observers, blinded to the group assignments, scored the performance. Additionally, a third individual managed animal handling and recorded 1-min segments of each session with a digital video camera. All procedures were designed to minimize animal stress and discomfort: the water temperature was carefully controlled, the swimming duration was limited to 4 min per session, and the mice were gently dried afterward. The protocol strictly adhered to institutional animal care guidelines and was approved by the Animal Ethics Committee, ensuring humane treatment throughout the assessment.

Motor electrophysiology: At 14 weeks post-SCI, neural function was assessed via electrophysiological testing. The mice were deeply anesthetized with pentobarbital, and motor-evoked potentials (MEPs) were recorded via electrophysiology. A single 10-mA stimulus was administered to activate the motor cortex.

ZnS@BSA@MM NPs targeted to SCI for imaging: For fluorescence imaging, ZnS@BSA@MM-Cy5.5 NSs (150 μL, 2 mg/mL) or free Cy5.5 solution was administered intravenously. In this formulation, Cy5.5 was covalently conjugated to both the ZnS@BSA core and the microglial membrane (MM) coating, allowing accurate tracking of the entire nanoparticle system. After 48 h, the mice were euthanized, and the spinal cord, heart, liver, spleen, lungs, and kidneys were collected for imaging.

Histological analysis: At 14 weeks post-SCI, the mice were deeply anesthetized and perfused intracardially with normal saline followed by 4 % paraformaldehyde. A 1.5 cm segment of spinal cord tissue surrounding the injury site was harvested and sectioned longitudinally into 10 μm thick slices using a cryotome. These sections were then prepared for various experimental analyses. Hematoxylin and eosin (H&E) staining was used to assess the morphology and size of lesion cavities, while H&E and Masson's trichrome staining were applied to bladder tissue to evaluate neurogenic bladder recovery.

Immunofluorescence analyses: At 14 weeks post-operation, the mice were deeply euthanized and perfused intracardially with 50 mL of saline solution followed by 50 mL of 4 % paraformaldehyde. The spinal cord containing the lesion area was then dissected and dehydrated in a 30 % sucrose solution. Each group's spinal cord was photographed with a digital camera before being embedded in OCT-freezing medium. The samples were sectioned into 8 μm thick slices along the sagittal axis. Immunofluorescence staining was used to identify β3-tubulin (Abcam, ab78078, USA), GFAP (Abcam, ab7260, USA), NeuN (Abcam, ab177487, USA), ChAT (Affinity, DF6964, USA), Nestin (Affinity, DF7754, USA), and Synaptophysin (Abcam, ab32127, USA) at the lesion site.

Inflammation analyses in vivo: Spinal cord tissues from each group were collected to evaluate changes in the expression of NF-κB signaling axis-related proteins (IκBα, phospho-IκBα, NF-κB p65, and phospho-NF-κB p65) via Western blot analysis. The tissue supernatants were also collected to measure the secretion levels of the inflammatory cytokines IL-1β and TNF-α. To assess the impact of the ZnS@BSA@MM on spinal neuron regeneration and inflammation regulation, spinal cord tissues were subjected to RNA transcriptome sequencing. This analysis aimed to evaluate the changes in neuroregeneration and inflammation-related signaling genes in the ZnS@BSA@MM group.

Statistical methods: All experimental data were expressed as the mean ± standard deviation (SD). One-way analysis of variance (ANOVA) with Tukey's post hoc test was employed to analyze the results. Differences between groups were assessed via GraphPad Prism software, with p < 0.05 considered statistically significant.

CRediT authorship contribution statement

Qin Qin: Writing – original draft, Formal analysis, Data curation, Conceptualization. Bingrong Jin: Formal analysis, Data curation. Chaowen Bai: Formal analysis, Data curation. Zijie Zhou: Formal analysis, Data curation. Bingchen Shan: Formal analysis, Data curation. Chenhui Ding: Formal analysis, Data curation. Zhihui Han: Formal analysis, Data curation. Xi Wang: Formal analysis, Data curation. Hao Zhong: Formal analysis, Data curation. Kai Zhao: Formal analysis, Data curation. Hong Xie: Resources, Project administration. Xiang Gao: Writing – review & editing, Project administration. Liang Cheng: Writing – review & editing, Supervision, Project administration. Xiaozhong Zhou: Supervision, Resources.

Ethics approval and consent to participate

The study received ethical approval from the Ethics Committee of The Second Affiliated Hospital of Soochow University (Approval number: LK2024105).

Declaration of competing interest

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

Appendix A. Supplementary data

The following are the Supplementary data to this article.