Selenium-doped carbon dots nanozymes hitchhiking tailored liposomes block neuronal pyroptosis through GPX4/ROS/NLRP3/GSDMD axis to attenuate ischemic stroke

Jiaxuan Hou; Li Yao; Yane Li; Enrui Xie; Jiawei Zhang; Hao Wu; Yuanyuan Zhu; Zhichao Deng; Chenxi Xu; Zongru Han; Lu Bai; Mingzhen Zhang; Shaoying Lu; Runqing Li; Hui Cai

doi:10.1186/s12951-026-04107-9

. 2026 Feb 7;24:226. doi: 10.1186/s12951-026-04107-9

Selenium-doped carbon dots nanozymes hitchhiking tailored liposomes block neuronal pyroptosis through GPX4/ROS/NLRP3/GSDMD axis to attenuate ischemic stroke

Jiaxuan Hou ¹, Li Yao ², Yane Li ¹, Enrui Xie ³, Jiawei Zhang ¹, Hao Wu ¹, Yuanyuan Zhu ⁴, Zhichao Deng ⁴, Chenxi Xu ⁴, Zongru Han ⁵, Lu Bai ⁶, Mingzhen Zhang ⁴, Shaoying Lu ^1,^✉, Runqing Li ^6,^7,^✉, Hui Cai ^1,^✉

PMCID: PMC12977847 PMID: 41654881

Abstract

Ischemia-reperfusion (I/R) injury is a critical contributor to adverse outcomes following stroke. During I/R injury, excessive production of reactive oxygen species (ROS) leads to various forms of neuronal cell death. Moreover, the blood-brain barrier (BBB) significantly hinders the delivery and efficacy of many neuroprotective agents. Given selenium’s crucial role in mitigating brain ischemia, we developed a selenium-based nanozyme encapsulated in glutathione (GSH)-conjugated liposomes to overcome these challenges. Specifically, we encapsulated selenium-doped carbon dot nanozymes (Se-CDs) within GSH-conjugated liposomes (Se-CD@LP-GSH) to enable targeted delivery and enhance therapeutic efficacy in ischemic stroke. This system demonstrates effective ROS scavenging capabilities both in vitro and in vivo, while also enhancing the biocompatibility of Se-CDs and their ability to cross the BBB. In the tMCAo model, Se-CD@LP-GSH reduces the neuronal death and infarct area following cerebral I/R injury, and promotes improvements in spatial learning ability and sensorimotor function. Mechanistically, Se-CD@LP-GSH promoted the upregulation of GPX4, an essential selenoprotein, thereby preserving mitochondrial function and suppressing ROS generation Consequently, the reduced ROS levels inhibit NLRP3/GSDMD-mediated neuronal pyroptosis during cerebral I/R injury. By improving the brain-targeting ability of Se-CDs via GSH-functionalized liposomal delivery, our work elucidates their neuroprotective efficacy and mechanistic basis, thus providing a translationally relevant strategy for ischemic stroke therapy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04107-9.

Keywords: Selenium-Doped carbon dots nanozymes, Glutathion peroxidase, GPX4/ROS/NLRP3/GSDMD axis, Ischemic stroke, Pyroptosis

Introduction

Stroke is the world’s second leading cause of mortality and a predominant cause of disability [1]. Ischemic stroke is a stroke subtype defined as infarction of the brain, spinal cord, or retina and accounts for 71% of all stroke cases globally [2, 3]. Contemporary clinical management prioritizes prompt reperfusion through intravenous thrombolysis or endovascular thrombectomy, interventions that can markedly enhance patient outcomes when implemented within a limited therapeutic window of 3 to 4.5 h following the onset of symptoms [4]. Nevertheless, this intervention, which is contingent upon timely administration, is applicable to only 2%–5% of stroke patients. Furthermore, among those who receive treatment, successful cerebral reperfusion is achieved in approximately 50% of cases [5, 6].

Once the optimal therapeutic window is missed, ischemia-reperfusion (I/R) injury ensues, exacerbating neural damage [2]. I/R injury is closely linked to mitochondrial dysfunction and oxidative stress, as a consequence of dysregulated reactive oxygen species (ROS) [7, 8]. This triggers a cascade of pathological processes, including oxidative stress, apoptosis, necrosis, ferroptosis, neuroinflammation, blood-brain barrier (BBB) disruption, and extracellular matrix (ECM) remodeling [9, 10]. Among the forms of regulated cell death, pyroptosis—a proinflammatory and lytic cell death mechanism—has emerged as a critical contributor to neuroinflammation following stroke [11]. Both neurons and glial cells are susceptible to pyroptotic death in the context of I/R injury, making it a promising therapeutic target [12, 13].

Given the complex pathological mechanism of ischemic stroke, various functional nanomaterials have gained attention for biomedical applications such as drug delivery, diagnostics, bioimaging, and therapeutic intervention [14]. In particular, nanozymes, a class of nanomaterials with intrinsic enzyme-like catalytic activities, have shown great promise in treating oxidative stress-related diseases [15]. Several nanozymes exhibit superoxide dismutase, catalase, and glutathione peroxidase (GPx)-like activities, enabling the breakdown of ROS into less reactive species like O₂ and H₂O₂ [16]. Selenium, an essential trace element, is incorporated into GPX4 — a selenoenzyme that restricts lipid-peroxidation–driven ferroptosis— and thereby supports cellular redox homeostasis after stroke [17]. By inhibiting ferroptosis and regulating iron metabolism, selenium likely contributes to preservation of the iron–furin–BDNF axis; these processes may also intersect with other regulated cell-death pathways such as pyroptosis and with consequent neuroinflammatory responses [18]. Recent studies have reported the therapeutic potential of selenium and selenium-based nanomaterials in I/R injury across various organs, including the brain [19–24]. These materials not only possess inherent nanozyme activity but can also upregulate the biosynthesis of selenium-containing antioxidant proteins, facilitating ROS scavenging and organ protection. Among these, selenium-doped carbon quantum dots (Se-CDs) have attracted significant attention due to their dual advantages: the intrinsic antioxidant properties and their role as precursors in the biosynthesis of the selenoprotein GPX4, coupled with the versatile surface chemistry and photoluminescent characteristics of carbon dots. This combination positions Se-CDs as a promising candidate for the treatment of ROS-related neurological disorders.

Nevertheless, the biocompatibility and brain-targeting capability of Se-CDs remain major challenges [15]. Liposomes have emerged as efficient nanocarriers for drug delivery due to their favorable safety profile and ability to encapsulate both hydrophilic and hydrophobic agents [25]. However, the BBB poses a formidable obstacle to brain-targeted drug delivery. Although BBB integrity can be transiently compromised during acute stroke due to oxidative damage, this disruption typically lasts only a few hours [26, 27]. To overcome this limitation, several strategies have aimed at enhancing brain-targeted delivery. Studies have confirmed that the sodium-dependent glutathione (GSH) transporter has a preferential expression in the central nervous system and the BBB [28]. Conjugation of GSH to liposomes markedly improved brain delivery of encapsulated drugs and nucleic acids while maintaining a favorable safety profile [29–32].

In this study, we developed selenium-doped carbon dots encapsulated within glutathione-conjugated liposomes (Se-CD@LP-GSH). Following intravenous administration, Se-CD@LP-GSH exhibited remarkable biocompatibility and preferentially accumulated in ischemic brain regions, thereby enhancing delivery efficiency. Functionally, Se-CD@LP-GSH effectively scavenged excessive ROS and reduced neuronal death after I/R injury. Mechanistically, this neuroprotective effect was associated with the upregulation of GPX4 expression in neurons, the preservation of mitochondrial function with consequent suppression of ROS production, and the attenuation of neuronal pyroptosis (Scheme 1).

Scheme 1 — Schematic representation of Se-CD@LP-GSH preparation and application in the treatment of cerebral I/R injury. A) Synthesis process of Se-CD@LP-GSH. B) Blood–brain barrier targeting and penetration capabilities of Se-CD@LP-GSH. C) ROS scavenging activity of Se-CD@LP-GSH. D) Functional recovery following treatment with Se-CD@LP-GSH. E) Potential mechanisms by which Se-CD@LP-GSH alleviates cerebral I/R injury. (Created with BioRender)

Materials and methods

Synthesis of Se-CDs

50 mg of L-selenocystine (Macklin, Shanghai, China) was prepared in 10 mL of deionised water. To facilitate dissolution, the pH was adjusted to approximately 9 using sodium hydroxide. The solution was kept stirring at 60 °C for 2 h. Next, the mixture was transferred to a hydrothermal reactor and maintained at 60 °C for 24 h to initiate the carbonization and doping process. The reaction mixture was then subjected to centrifugation at 12,000 rpm for 15 min to eliminate insoluble residues. The resulting supernatant was collected and subjected to dialysis against ultrapure water for 24 h using a dialysis bag with MWCO of 3500 Da to eliminate small molecular impurities. Finally, the purified solution was freeze-dried to obtain solid Se-CDs for subsequent use.

Synthesis of Se-CD@LP-GSH and Se-CD@LP-PEG

Liposomes encapsulating Se-CDs were prepared using the thin-film hydration method, followed by post-insertion of GSH-PEG or PEG micelles, as described previously [33]. Briefly, hydrogenated soy phosphatidylcholine (HSPC) (Ruixi, Xi’an, China), cholesterol (Ruixi, Xi’an, China) and DSPE-PEG2000 (Ruixi, Xi’an, China) were co-dissolved in chloroform at a ratio of 56.2: 38.5: 1. The lipid mixture was subjected to rotary evaporation under vacuum yielding a thin lipid film, followed by 12-hour high vacuum drying to remove residual organic solvent completely. The dried lipid film was rehydrated with PBS solution containing Se-CDs (5 mg), followed by incubation at 4 °C overnight to facilitate the formation of multilamellar vesicles. The resulting suspension was extruded sequentially via polycarbonate films with pore diameters of 400, 200, and 100 nm to obtain liposomes with uniform particle size distribution. For surface functionalization, DSPE-PEG2000-MAL and GSH were mixed at a molar ratio of 1:1.5 and allowed to react at room temperature for 2 h to achieve thiol-maleimide conjugation. The resulting GSH-PEG conjugates were incorporated into the liposomal membrane via ultrasonication. The final product, Se-CD@LP-GSH, was purified by dialysis against deionized water using a 100-kDa MWCO membrane to remove free Se-CDs and unreacted components. A control formulation, Se-CD@LP-PEG, was prepared using the same protocol, excluding GSH from the surface modification step. ICP-OES (Thermo iCAP 7200 ICP-OES, ThermoFisher, USA) was employed to determine the encapsulation efficiency and drug loading capacity of the liposomes.

Synthesis of FITC-GSH and DiO/DiL-Liposome

DiL/DiO (1 mg/mL) solution was added to the HSPC/cholesterol/DSPE-PEG2000 mixed solution. Dil/DiO-labeled liposomes were then synthesized using the thin-film hydration method under complete light-protected conditions, following the procedure described above.

The FITC solution (dissolved in DMSO) was added dropwise to the GSH solution (dissolved in 0.1 M NaHCO₃ buffer solution) at a molar ratio of 1:2 (FITC: GSH). The mixture was gently stirred under light-protected conditions at room temperature (pH 9.0) for 4–6 h. Upon completion of the reaction, 1 M Tris-HCl buffer (pH 7.4) was added to stop the reaction. The solution was left for 30 min to ensure complete inactivation of any unreacted FITC. The mixture was then poured dialyzed (MWCO 1 kDa) against 1 L of ultrapure water under light-protected conditions for 24 h, with water changes every 4 h to remove excess unreacted FITC. The purified FITC-GSH conjugate was subsequently lyophilized to obtain a fluorescent GSH powder, suitable for long-term storage. FITC-GSH/DiL-Liposomes were synthesized using the thin-film hydration method under complete light-protected conditions, following the procedure described above.

Synthesis of Cy5.5-Se-CDs

For fluorescent tagging, the carboxyl functionalities of Se-CDs were activated using a mixture of 100 µL EDC (10 mg/mL) and 200 µL NHS (10 mg/mL), followed by incubation at 25 °C for 30 min. Following activation, 1 mL of 5 mg/mL Se-CDs solution was added to the EDC/NHS mixture and allowed to react for 1 h at room temperature, protected from light. Subsequently, 2 mL of 0.25 mg/mL Cy5.5 solution (Ruixi, Xi’an, China) was added to the activated Se-CDs mixture. The reaction was stirred continuously for 24 h in the dark to facilitate covalent conjugation. Upon completion, the reaction mixture was dialyzed against ultrapure water for 24 h using a suitable dialysis membrane to remove unreacted dye and other small molecules. The purified Cy5.5-labeled Se-CDs (Cy5.5-Se-CDs) were then freeze-dried for further use. To prepare fluorescently labeled liposomes, Cy5.5-Se-CDs were encapsulated into liposomes using the same thin-film hydration and extrusion protocol as described above. Post-insertion of GSH-PEG or PEG micelles was subsequently performed to generate functionalized liposomes. The resulting formulations were denoted as Cy5.5-Se-CD@LP-GSH and Cy5.5-Se-CD@LP-PEG, respectively, corresponding to liposomes functionalized with GSH or PEG.

Cell culture

SH-SY5Y (ZQ0050) cell line was purchased from Zhong Qiao Xin Zhou Biotechnology Co., Ltd, China. HT-22 (CL-0697), bEnd.3 (CL-0598) and Raw 264.7 (CL-0190) cell lines were purchased from Pricella Biotechnology Co., Ltd, China. BV2 (STCC20009P-1) cell line was purchased from Wuhan Servicebio Technology Co., Ltd, China. Cell lines were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum and 100 U·mL⁻¹ penicillin–streptomycin, cultured at 37 °C in a humidified incubator containing 5% CO₂. Routine testing confirmed the absence of mycoplasma contamination throughout the study.

In vitro internalization and subcellular location of nanoparticles (NPs)

To investigate the cellular internalization and subcellular localization of NPs, Cy5.5-Se-CDs encapsulated in liposomes were employed. SH-SY5Y cells were incubated with either Cy5.5-Se-CD@LP-PEG or Cy5.5-Se-CD@LP-GSH at a concentration of 200 µg/mL for various time points (0, 2, 4, 6, and 8 h). Following incubation, cellular uptake of NPs was assessed qualitatively by fluorescence microscopy and quantitatively by flow cytometry (Becton Dickinson, USA). To visualize cellular morphology, cytoskeletal actin filaments were labeled using Actin-Tracker Green (C2201S, Beyotime, China), while nuclei were counterstained with DAPI (C1005, Beyotime, China).

To further evaluate subcellular localization and potential mitochondrial targeting, SH-SY5Y cells were incubated with the NPs for 12 h. Subsequently, mitochondria were stained using Mito-Tracker (C1996S, Beyotime, China), and the degree of colocalization between Cy5.5-labeled NPs and mitochondria was analyzed using fluorescence microscopy.

In vitro across the BBB

We established an in vitro BBB model employing bEnd.3 cells. Cells (1 × 10⁵/well) were plated onto the upper chambers of transwell inserts equipped with permeable polyester membranes (Corning, USA; pore size: 0.4 μm). The upper chamber was filled with phenol red-free medium. The medium was renewed at 2-day intervals, with cells were kept in culture for 14 days to allow the formation of a confluent and functionally mature monolayer mimicking the BBB. The quantity of the monolayer was evaluated by recording transendothelial electrical resistance (TEER) with a Millicell-ERS voltohmmeter (Millipore Millicell ERS3.0, USA). Only monolayers exhibiting TEER values ≥ 150 Ω·cm² were deemed acceptable for subsequent experiments, indicating a tight and selective endothelial barrier. To determine the permeability of NPs, liposomes loaded with Cy5.5-Se-CDs were placed in the upper chamber and maintained at 37 °C for 12 h. Post-incubation, the fluorescence intensity of Cy5.5 in the lower chamber medium was measured using a fluorescence spectrophotometer. A standard calibration curve was established by preparing serial dilutions of Cy5.5-Se-CDs in phenol red-free medium, enabling the quantification of nanoparticle transport across the BBB model.

To assess transcytosis and subsequent uptake by neuronal cells, HT-22 cells (5 × 10⁵/well) were plated in the lower chambers of the Transwell system. After a 24-hour pre-incubation to establish an adherent monolayer, the bEnd.3 monolayers in the upper chambers were co-cultured with the HT-22 cells. Cy5.5-Se-CD@DiO-LP-GSH was then introduced into the upper chambers and kept at 37 °C for 12 h. NPs internalization by HT-22 cells in the lower chamber was then observed using confocal laser scanning microscopy (Olympus FV3000, Japan), enabling visualization of NPs transcytosis and neuronal targeting.

Oxygen-glucose deprivation and reperfusion model (OGD/R)

SH-SY5Y or HT-22 cells were seeded in culture dishes and left overnight to adhere. Cells were pretreated with Se-CD@LP-GSH for 12 h. Following pretreatment, the medium was renewed with glucose- and serum-free medium, and cells were subjected to hypoxic conditions (95% N₂, 5% CO₂, 0.1% O₂) in a hypoxia chamber for 4 h. After hypoxia, the cells were further incubated with standard culture medium for 24 h prior to subsequent experiments.

In vitro cytoprotective study of Se-CD@LP-GSH

CCK8 assay: SH-SY5Y cells (1 × 10⁴/well) were plated in 96-well plates and left overnight to adhere. Cells were pretreated with Se-CD@LP-GSH for 12 h, then incubated in glucose- and serum-free medium under hypoxia (95% N₂, 5% CO₂, 0.1% O₂) for 4 h. Afterward, normoxic medium containing Se-CD@LP-GSH was added for 24 h. Cell viability was acquired by adding 100 µL CCK-8 per well, incubating 2 h at 37 °C, and reading absorbance at 450 nm.

Live/dead assay: SH-SY5Y cells (1 × 10⁵/well) were plated in 12-well plates. The establishment of the OGD/R model in SH-SY5Y cells and the pretreatment with Se-CD@LP-GSH were performed. After incubation, the medium was discarded, and each well was treated with PBS containing 2 µM Calcein-AM and 8 µM propidium iodide (PI) (E-CK-A354, Elabscience, China). Fluorescence microscopy was used to visualize stained cells.

Annexin V-FITC/ PI analysis: To assess apoptosis, SH-SY5Y and HT-22 cells were labeled with Annexin V-FITC and PI using the Annexin V-FITC/PI Apoptosis Detection Kit (AT101C, Multi Science, China), following the recommended protocol. The samples were then subjected to flow cytometric analysis.

JC-1 analysis: After harvesting SH-SY5Y cells, mitochondrial membrane potential was assessed using the JC-1 staining kit (c2006, Beyotime, China). Fluorescence microscopy and flow cytometry were employed to analyze the staining results.

In vitro ROS scavenging ability of Se-CD@LP-GSH

ROS levels in SH-SY5Y cells were evaluated using dichlorodihydrofluorescein diacetate (DCFH-DA). The OGD/R model in SH-SY5Y cells was established, and pretreatment with Se-CD@LP-GSH was conducted according to the previously outlined procedures. Subsequently, the culture medium was discarded, and each well was treated with medium containing 10 µM DCFH-DA and incubated at 37 °C for 30 min. The cells underwent three PBS washes before ROS levels were evaluated by fluorescence microscopy and flow cytometry.

Transfection

Small interfering RNA targeting GPX4 (si-GPX4) was purchased from GENEPharma Biotech Co., with the following sequence: sense, GCCAUCAAAUGGAACUUUA; antisense, UAAAGUUCCAUUUGAUGGCTT. HT-22 cells (7 × 10⁵/well) were plated in 6-well plates. After 1 day, the cells were transfected with si-GPX4 using the CALNPTM RNAi Transfection Kit (D-Nano Therapeutics, Beijing, China). The transfected cells were then utilized for subsequent experiments.

Construction of tMCAo model

Male C57BL/6J mice (8–10 weeks, 20–25 g) were supplied by the Experimental Animal Center of Xi’an Jiaotong University. Mice were housed under a 12-h light/dark cycle with unrestricted access to food and water. Randomization was applied for group assignment, and all evaluations were performed by investigators blinded to treatment conditions. All procedures complied with institutional guidelines and were approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University (No. XJTUAE2024-2118).

Following induction with 2% isoflurane in O₂, mice were positioned on a stereotaxic surgical platform, and anesthesia was maintained with 1–2% isoflurane. Under a surgical microscope, the carotid arteries were exposed by carefully isolating the common carotid artery (CCA), internal carotid artery (ICA), and external carotid artery (ECA). The peripheral end of the ECA was ligated with a silk suture as close to the distal extremity as possible. Another suture was loosely placed near the bifurcation between the ECA and CCA without ligation. Microvascular clamps were then used to occlude the CCA and the ICA proximal to the carotid bifurcation. A small incision was created between the two sutures on the ECA, and a suture was inserted into the lumen and advanced in the direction of the clamp on the CCA. The partially attached segment of the ECA was completely severed to allow full mobilization of the proximal ECA. The freed ECA was then gently pulled caudally to align it with the bifurcation of the ECA and ICA, facilitating insertion of the filament into the ICA. The suture on the proximal ECA was securely tied to fix the filament in place, and the vascular clamp on the ICA was gently removed. The filament was advanced further through the ICA until it reached the origin of the middle cerebral artery (MCA). The vascular clamp on the CCA was then removed, and the incision was closed. After 1 h of occlusion, the mouse was re-anesthetized, and the CCA was clamped again. The filament was withdrawn, and the proximal ECA stump was ligated. Postoperative monitoring was conducted during anesthetic recovery. Two hours after surgery, once the mice had resumed activity, neurological function was evaluated by gently lifting the mouse by the tail and observing whether it could turn to both sides. A unilateral turning behavior was considered an indication of successful stroke modeling. Mice with Longa brain injury scale neurological deficit scores of 2–3 were selected for subsequent experiments. Sham groups were subjected to identical surgical procedures, but without insertion of the filament. The overall mortality rate of stroke was approximate 25%.

TTC staining

Following euthanasia, mouse brains were immediately harvested and rapidly frozen at − 20 ℃ for 15–20 min. After a slight thawing at room temperature, the brain tissue was placed on an ice-cold surface and sectioned into slices with a sharp blade at a thickness of 2–3 mm, ensuring clean and even cross-sections. TTC (T8877, Sigma, USA) staining solution (2%) was pre-warmed in a 37 °C incubator for 30 min before use. The brain slices were placed in TTC solution (pre-heated) and kept at 37 °C for about 30 min in the dark. After staining, formalin fixative was slowly added to the container to fully immerse the tissue slices, thereby terminating the staining reaction. The brain slices were stored overnight at 4 °C in the dark.

Nissl staining and TUNEL/NeuN staining

Following euthanasia, 5 μm brain sections embedded in paraffin were subjected to deparaffinization in xylene (10 min), graded ethanol rehydration (100%, 90%, 70%, 2 min each), and a final rinse in distilled water. For Nissl staining, the sections were immersed in Nissl staining solution (C0117, Beyotime, China) at regular temperature for 10 min. For TUNEL/NeuN double staining, sections were incubated with the corresponding reaction mixture at ambient temperature for 1 h. After Nissl staining, tissue sections were cleared with xylene, mounted using neutral resin, and examined under a light microscope. For TUNEL staining, nuclei were stained with DAPI for 5 min, and fluorescence imaging was performed. Positively labeled cells were quantified in anatomically defined regions of each section and used for subsequent statistical analysis.

Targeting ability and tissue distribution of NPs in vivo

Following 2 h of reperfusion in the tMCAo model, Cy5.5-labeled NPs, including Cy5.5-Se-CDs, Cy5.5-Se-CD@LP-PEG, and Cy5.5-Se-CD@LP-GSH were administered intravenously into the mice. The fluorescence intensity in the brain was monitored at several time points using the In Vivo Imaging System (VISQUE, Viewworks, South Korea). At 12 h post-reperfusion, the mice were euthanized, and their brains and major organs were excised. These tissues were subsequently imaged using the IVIS system to assess NPs distribution. The brains were sliced into 2 mm and imaged with IVIS again.

To investigate the intracerebral distribution of NPs, brain frozen sections were prepared and Immunostaining for NeuN, a neuronal marker, was performed using secondary antibodies. Fluorescence imaging was conducted to visualize the localization of NPs within the brain tissue.

Immunofluorescent staining

Mouse brains were fixed in 4% paraformaldehyde (PFA) at 4 °C overnight, then sequentially immersed in increasing sucrose solutions (10–30%) for cryoprotection. They were subsequently embedded in OCT compound and sliced into 8 μm coronal sections. These sections were blocked using a solution of 1% goat serum and 0.1% Triton X-100 in PBS, followed by overnight incubation at 4 °C with primary antibodies targeting GSDMD, NLRP3, GPX4, and NeuN. After rinsing with PBS, fluorochrome-conjugated secondary antibodies were applied for 1 h, and nuclei were stained with DAPI. Fluorescent images were acquired using a fluorescence microscope. For cellular immunofluorescence, cells were fixed with 4% PFA and subjected to the identical staining procedure.

Immumohistochemical staining

24 h after tMCAo surgery, and the brains were extracted and fixed in 4% PFA overnight. Subsequently, the tissues were embedded in paraffin and sectioned. Tissue sections were subjected to deparaffinization using xylene, followed by rehydration in a descending series of ethanol dilutions. Antigen retrieval was conducted with a 10 mM sodium citrate buffer adjusted to pH 6.0. Following this, the tissue sections were incubated in normal goat serum for 30 min at 37 °C to minimize any non-specific binding. The sections were then treated overnight at 4 °C with primary antibodies: IL-18 antibody and IL-1β antibody. The sections were subsequently exposed to the appropriate biotinylated secondary antibodies at 37 °C for 30 min. After staining with 3,3′-diaminobenzidine (DAB) (D8001, Sigma-Aldrich, USA), positive staining was determined. Finally, the slides were observed and photographed using a DM600B automated microscope (Leica Microsystems, Heidelberg, Germany).

RNA extraction and quantitative real-time PCR analysis (qRT-PCR)

RNA was extracted from cultured cells and ischemic penumbra brain tissue using a spin column-based RNA Isolation Kit (R0027, Beyotime, Shanghai, China). cDNA was generated from extracted RNA by a Reverse Transcription Kit (RK20428, Abclonal, Wuhan, China) and amplified by qRT-PCR with SYBR Green Master Mix (RK21203, Abclonal, Wuhan, China). Gene expression quantification employed the 2^–ΔΔCT method normalized to β-actin. Primer sequences appear in Supplementary Table S1.

Western blotting assay

Protein extraction from cultured cells and brain tissue in the ischemic penumbra was carried out with RIPA buffer supplemented by protease and phosphatase inhibitors. (WB3100, NCM Biotech, China). The BCA method was employed to quantify protein levels. Samples containing equivalent protein quantities (30 µg) were separated using SDS-PAGE and subsequently electrotransferred onto PVDF membranes. These membranes underwent blocking in 5% skim milk for 1 h at ambient temperature, followed by overnight probing with primary antibodies at 4 °C. The membranes were subsequently incubated with appropriate species-matched secondary antibodies for 1 h at ambient temperature. Signals were visualized by ECL and analyzed with ImageJ. A comprehensive list of antibodies and dilution ratios is provided in Supplementary Table S2.

ELISA assay

IL-1β ELISA kits (SYP-MOO26, Upingbio Biology, China) were utilized to measure the concentrations of cytokine IL-1β. IL-18 ELISA kits (E-EL-MO730, Elabscience Biotechnology Co., Ltd, China) were employed to measure the concentrations of cytokine IL-18.

RNA sequencing

Brain tissue samples were collected 24 h post-operation from mice in the Sham, tMCAo, and tMCAo + Se-CD@LP-GSH groups (n = 3). Transcriptome sequencing was conducted by Beijing Biomarker Technologies Co., Ltd. Differentially expressed genes (DEGs) were determined using a fold change cutoff of > 1.5 and a P value < 0.05.

Statistical analysis

Data were analysed by GraphPad Prism 8.0. For comparisons between two groups, an unpaired two-tailed Student’s t-test was applied, whereas differences among three or more groups were evaluated using one-way analysis of variance. All data are expressed as mean ± SEM. Statistical significance was set at P < 0.05. (ns: not significant. *P < 0.05; **P < 0.01; ***P < 0.001)

Results

Preparation and characterization of Se-CDs

Se-CDs were synthesized via a hydrothermal method as illustrated in Fig. 1A, where L-selenocysteine was dissolved in deionized water under alkaline conditions and maintained at 60 °C for 24 h, resulting in the formation of Se-CDs. Transmission electron microscopy (TEM) was employed to examine the morphological characteristics of Se-CDs. The Se-CDs were found to be spherical, exhibiting a lattice structure, with diameters ranging from approximately 2 to 10 nm (Fig. 1B). UV/Vis absorption spectroscopy was used to study the optical properties of the Se-CDs. As shown in Fig. S1A, due to the presence of multiple electron transitions, two characteristic absorption peaks of aqueous Se-CDs were observed at approximately 285 nm and 340 nm, consistent with previously reported studies [34]. The fluorescence spectra of Se-CDs revealed the excitation and emission peaks at 364.6 nm and 452.6 nm, respectively (Fig. S1B). As shown in Fig. 1C, Se-CDs demonstrated excitation-dependent luminescence, which is similar to semiconductor quantum dots. The structural characteristics of the Se-CDs were further investigated using nuclear magnetic resonance (NMR) spectroscopy and X-ray photoelectron spectroscopy (XPS). As shown in Fig. 1D, the ¹H NMR spectrum exhibited signals in the range of 7–9 ppm, which are characteristic of protons attached to sp²-hybridized carbon atoms. This observation indicated the presence of aromatic structures within the Se-CDs. The XPS spectra showed that the Se-CDs were mainly composed of carbon, oxygen, nitrogen, and selenium (Fig. 1E), which showed a similar selenium content (4.9%) to the previous studies [34, 35]. In the deconvoluted N 1s spectrum (Fig. 1F), distinct peaks at 399.4 eV and 403.1 eV indicated the presence of pyridinic and pyrrolic nitrogen species, respectively. The deconvoluted XPS spectra of C 1s exhibited three major peaks (Fig. 1G). The most prominent peak at 284.8 eV should be assigned to the graphitic sp²-hybridized carbon structure of the Se-CDs. The peak at 286.6 eV should be associated with the presence of C–O, C–Se, and C–N bonds, while the peak observed around 288.2 eV corresponds to carbonyl (C = O) functional groups. Furthermore, the deconvoluted Se 3d spectrum, centered around 55 eV, confirmed the existence of C–Se–C structural units within the material [36].(Fig. 1H) Fourier transform infrared (FTIR) spectroscopy was utilized to further investigate the chemical composition and structural characteristics of the Se-CDs (Fig. S1C). A broad absorption band observed in the range of 3100–3600 cm⁻¹ should be ascribed to the stretching vibrations of –OH and –NH functional groups. The prominent absorption peak at 1602 cm⁻¹ was attributed to the stretching vibration of C = C bonds within a conjugated system, while the peaks observed near 1400 cm⁻¹ were ascribed to the C = C stretching vibrations characteristic of aromatic structures. Additionally, absorption bands in the range of 1200–900 cm⁻¹ were indicative of C–O, C–N, and C–Se stretching vibrations. The FTIR results were consistent with the XPS data and exhibited significant deviations from the spectral characteristics of L-selenocysteine, indicating a successful structural transformation during the synthesis of Se-CDs.

Following an ischemic stroke or subsequent reperfusion, the mitochondria generate an excessive amount of ROS, leading to oxidative stress-induced damage, particularly affecting neuronal cells [7]. Se-CDs have been reported to possess potent ROS-scavenging ability. In this study, we assessed the ROS-scavenging capacity of Se-CDs. The total antioxidant capacity of Se-CDs was assessed by measuring the absorbance of the ABTS•⁺ radical, which exhibited a characteristic blue-green coloration with a maximum absorption peak at 734 nm. As shown in Fig. 1I and Fig. S2A, the scavenging activity of Se-CDs exhibited a dose-dependent trend, reaching approximately 80% at a concentration of 20 µg/mL. This result highlighted the strong antioxidant capacity of Se-CDs. The NBT assay and TMB assay were used to assess the removal of superoxide anions (O₂^•−) and hydroxyl radicals (•OH) by Se-CDs, respectively. As shown in Fig. 1J-K and Fig. S2B, Se-CDs showed excellent O₂^•− and •OH scavenging abilities. Selenium is essential for natural glutathione peroxidase (GPx), which is known for its ability to scavenge H₂O₂. selenium NPs and metal selenides have been used to mimic GPx [37]. GPx-like activity of Se-CDs was measured by using the benzoic acid chromogenic method. GPx catalyzes the reaction between GSH and the benzoic acid-based chromogenic substrate, leading to the generation of a yellow-colored anionic product. The concentration of this anion is determined by measuring the absorbance at 422 nm. Based on the decrease in GSH levels, the GPx activity can be quantitatively calculated. As shown in Fig. 1L, the levels of GPx activity gradually increased with the concentration of Se-CDs. Meanwhile, the cellular GPx assay kit with NADPH was employed to assess the GPx activity of Se-CDs. It showed the same results as the benzoic acid chromogenic method (Fig. S2C). These results confirmed that Se-CDs have excellent anti-oxidant properties for scavenging with O₂^•−, •OH, and H₂O₂.

Preparation and characterization of Se-CD@LP-GSH

To increase the biocompatibility and brain-targeting ability, Se-CDs were encapsulated into liposomes. Figure 1M shows the synthetic procedure of liposomes and Se-CD@LP-GSH. Liposomes were synthesized utilizing the thin-film hydration technique as described in reference [36]. The preparation of Se-CD@LP-PEG adhered to the same protocol as that for Se-CD@LP-GSH, with the exception that GSH was excluded during the synthesis process. TEM was employed to observe the morphological characteristics of Se-CD@LP-GSH. As shown in Fig. 1N, Se-CD@LP-GSH displayed a monodisperse spherical morphology. The hydrodynamic diameters of Se-CD@LP-GSH and Se-CD@LP-PEG, measured by dynamic light scattering (DLS), were 115.9 ± 0.8 nm and 101.7 ± 0.5 nm, respectively, demonstrating good size uniformity and long-term stability after continuous measurements over 7 days (Fig. S3A, B). The zeta potentials of Se-CD@LP-GSH and Se-CD@LP-PEG were determined to be − 13.8 ± 2.0 mV and − 9.24 ± 1.04 mV, respectively, with only minor fluctuations observed over a 7-day period. This slight variation may be attributed to the gradual release of the encapsulated Se-CDs (Fig. S3C). To mimic the physiological circulation environment, we incubated Se-CD@LP-GSH in 10% FBS at 37 °C. As shown in Fig. S3D, the diameter of GSH gradually increased over time; however, the particle size remained below 200 nm on day 7. Meanwhile, we determined that the apparent blood circulation half-life of Se-CD@LP-GSH in healthy mice was 2.12 h (Fig. S3E). To confirm the presence of Se-CDs within liposomes, elemental mapping analysis was performed. As shown in Fig. 1O, Selenium was uniformly distributed throughout the liposomal structure, supporting successful encapsulation. The loading efficiency and loading content of Se-CD@LP-GSH were 46.4 ± 3.1% and 10.4 ± 0.7%, respectively. Furthermore, to visualize the conjugation of GSH to the liposomes, fluorescein isothiocyanate (FITC) and Dil were employed to pre-label GSH and liposomes, respectively. Fluorescence microscopy analysis demonstrated a pronounced co-localization of FITC and DiL signals within individual liposomes, confirming the successful functionalization of liposomes with GSH (Fig. 1P). Subsequently, the antioxidant capacity of Se-CD@LP-PEG and Se-CD@LP-GSH was systematically evaluated. As shown in Fig. 1Q, both formulations exhibited effective, dose-dependent suppression of ABTS•⁺ radicals, with no significant disparity observed between the two groups. To further investigate the ROS-scavenging properties, electron spin resonance (ESR) spectroscopy was performed using specific spin-trapping agents—DMPO for •OH and BMPO for O₂^•−. O₂^•− were produced in vitro using the xanthine/xanthine oxidase system, in which xanthine is oxidized to uric acid with concurrent generation of O₂^•−, and subsequently trapped by BMPO. As shown in Fig. 1R, all three NPs formulations exhibited varying degrees of O₂^•− scavenging activity. Hydroxyl radicals were generated via a classical Fenton reaction (Fe²⁺/H₂O₂) and subsequently captured by DMPO to form characteristic DMPO-OH adducts. As depicted in Fig. 1S, the Fenton system yielded distinct ESR signals indicative of •OH generation. Treatment with Se-CDs, Se-CD@LP-PEG, or Se-CD@LP-GSH resulted in a significant reduction in signal intensity, confirming their ability to scavenge hydroxyl radicals. Finally, to assess enzymatic antioxidant activity, a NADPH-based GPx assay kit was employed to measure the GPx-like catalytic activity of Se-CD@LP-PEG and Se-CD@LP-GSH. As illustrated in Fig. 1T, both of them exhibited dose-dependent GPx-mimicking activity. Collectively, these results confirmed that Se-CD@LP-PEG and Se-CD@LP-GSH possessed robust antioxidant properties, including the ability to scavenge O₂^•−, •OH, and H₂O₂. Moreover, the encapsulation of Se-CDs within liposomes did not impair their intrinsic enzymatic activity.

Evaluation of cellular uptake and BBB penetration of Se-CD@LP-GSH

Subsequently, we investigated the neuronal uptake and brain-targeting potential of these NPs. For this purpose, SH-SY5Y and HT-22 cell lines, which are widely recognized as in vitro neuronal models, were utilized in the study [38]. The sodium-dependent glutathione transporter has been shown to be selectively expressed in the central nervous system and the BBB, according to existing studies [28]. In this study, Se-CDs were labeled with the near-infrared fluorescent dye Cy5.5 and subsequently encapsulated within GSH- or PEG-functionalized liposomes. Flow cytometry and fluorescence imaging were used to compare the internalization of Cy5.5-Se-CD@LP-GSH and Cy5.5-Se-CD@LP-PEG by SH-SY5Y cells at different time points. As shown in Fig. 2A, the internalization of Se-CDs by SH-SY5Y cells exhibited a time-dependent pattern, which was significantly augmented upon conjugation with GSH-functionalized liposomes. This observation was further confirmed by fluorescence microscopy (Fig. S4). Mitochondria are key contributors to the pathophysiology of I/R injury [39]. To further determine the subcellular localization of NPs, SH-SY5Y cells were treated with Cy5.5-Se-CD@LP-GSH for 12 h, after which mitochondria were labeled using Mito-Tracker staining. Fluorescence microscopy demonstrated co-localization of Cy5.5-Se-CDs with mitochondria, which was further confirmed by plot profile analysis (Fig. 2B). The capacity to traverse the BBB is a fundamental requirement for the successful delivery of therapeutic agents targeting ischemic stroke. Although the BBB undergoes pathological disruption to some extent during I/R injury, studies have shown that the BBB remains in a fully open state for only a few hours [26, 27]. We studied the ability of NPs to cross the BBB monolayer in vitro. As shown in Fig. 2C, Se-CDs labeled with Cy5.5 were wrapped by PEG-liposomes or GSH-liposomes. Then Cy5.5-Se-CDs, Cy5.5-Se-CD@LP-PEG and Cy5.5-Se-CD@LP-GSH were incubated with bEnd.3 cells for 12 h. The medium in the lower chamber was collected, and the Cy5.5-Se-CDs were quantified by fluorescence spectroscopy. The NPs effectively penetrated the BBB. Approximately 20.78% of the Cy5.5-Se-CDs were carried across the bEnd.3 monolayer in the Cy5.5-Se-CD@LP-GSH group, which was significantly higher than those in the Cy5.5-Se-CDs group and the Cy5.5-Se-CD@LP-PEG group. To visualize the process of Se-CD@LP-GSH entry into HT-22 cells in the lower chamber, considering the species homology shared by HT-22 and bEnd.3 cells, we conducted a transwell assay for the co-culture of bEnd.3 and HT-22 cells to simulate the BBB model. Se-CD@LP-GSH exhibited no significant cytotoxicity toward either bEnd.3 or HT-22 cells, up to 400 µg/mL (Fig. S5). Cy5.5-Se-CDs were wrapped with GSH-liposomes labeled with DiO. As shown in Fig. 2D, the bEnd.3 cells were incubated with Cy5.5-Se-CD@DiO-LP-GSH for 12 h in the upper chamber, fluorescence imaging was used to examine the internalization of the penetrative Se-CD@LP-GSH in the lower chamber of HT-22 cells. Interestingly, the fluorescence signal of DiO-labeled GSH-liposomes was predominantly localized on the cell membrane, whereas the Cy5.5-Se-CDs exhibited a dispersed distribution within the cytoplasm (Fig. 2D). These findings indicated that Se-CD@LP-GSH enhanced the capacity of NPs to traverse the BBB, thereby facilitating increased neuronal uptake. Moreover, the observed co-localization of NPs with mitochondria in neuronal cells suggested a potential mechanism underlying their therapeutic efficacy.

Fig. 2 — Evaluation of cellular uptake and BBB penetration of Se-CD@LP-GSH. A) Flow cytometric analysis of SH-SY5Y cells treated with Se-CD@LP-PEG or Se-CD@LP-GSH for 0, 2, 4, 6, and 8 h. B) Fluorescence colocalization of Se-CD@LP-GSH (red) with mitochondria (green). C, D) Diagrammatic representation of the transwell assay used to establish a BBB coculture model. Quantification of the penetrative capacity of Cy5.5-labeled Se-CDs (C). Representative confocal fluorescence images showing the uptake of Se-CD@LP-GSH in HT-22 cells after 12 h of permeation through the BBB coculture system. Cy5.5-labeled Se-CDs are shown in red, and DiO-labeled liposomes are shown in green (D). **E-G**) IVIS images of Sham or tMCAo model mice treated with Cy5.5-Se-CDs, Cy5.5-Se-CD@LP-PEG, and Cy5.5-Se-CD@LP-GSH. Representative IVIS images of the mouse brain at various time points (E). Ex vivo IVIS images of major organs at 12 h after injection and statistical analysis of radiant efficiency in different groups. (F) Representative fluorescence images of brain slices collected following IVIS imaging and statistical analysis of radiant efficiency in various groups. H) Fluorescence image and intensity quantitative analysis of ischemic penumbra slices in different groups. (Data are expressed as mean ± SEM, n = 3. Groups: I, Sham + Se-CD@LP-GSH; II, tMCAo + Se-CDs; III, tMCAo + Se-CD@LP-PEG; IV, tMCAo + Se-CD@LP-GSH. *P<0.05, **P<0.01, ***P < 0.001)

To further assess the brain-targeting capability of NPs in vivo, a transient cerebral artery occlusion/reperfusion (tMCAo) model was established in the right hemisphere by inserting a suture to occlude the middle cerebral artery. Cy5.5-Se-CDs, Cy5.5-Se-CD@LP-PEG, and Cy5.5-Se-CD@LP-GSH were intravenously injected into Sham or tMCAo mice 2 h post-reperfusion, and fluorescence distribution was monitored using IVIS. Representative fluorescence images of the mouse brains at various time points are shown in Fig. 2E. Notably, the Cy5.5-Se-CD@LP-GSH-treated group displayed a markedly stronger fluorescence signal in the brain at 12 h. In the tMCAo group, the fluorescence signal progressively intensified over time, suggesting that the ischemic environment facilitated NPs accumulation, with the most significant effect observed in mice treated with Cy5.5-Se-CD@LP-GSH. Additionally, the biodistribution of NPs across major organs—including the heart, liver, spleen, lungs, kidneys and brain—was assessed by measuring Cy5.5 fluorescence in excised organs 12 h post-administration (Fig. 2F). Although high fluorescence signals were observed in primary metabolic organs such as the liver, kidneys, and spleen, the brain fluorescence intensity was significantly higher in the Cy5.5-Se-CD@LP-GSH-treated group compared to other groups. Fluorescence imaging of brain slices further corroborated these findings, demonstrating consistent patterns of NPs accumulation (Fig. 2G). To more accurately assess the ability of the NPs to traverse the BBB and be internalized by neurons within the ischemic hemisphere, immunofluorescence staining was performed using NeuN as a neuronal marker. As illustrated in Fig. 2H, NPs were internalized by neurons across all groups; However, the Se-CD@LP-GSH-treated tMCAo group exhibited the strongest fluorescence intensity and the highest degree of co-localization between NPs and neurons. Collectively, these results suggested that GSH-conjugated liposomes significantly enhanced the targeting efficiency of encapsulated Se-CDs to the ischemic region of the brain.

Se-CD@LP-GSH protected neurons from oxidative damage

To evaluate the neuroprotective effects of Se-CD@LP-GSH in vitro, we established a hypoxia-reoxygenation model using SH-SY5Y cells (Fig. 3A). To ensure the safety of the NPs, we first evaluated the cytotoxicity of Se-CDs and Se-CD@LP-GSH. We also evaluated the cytotoxicity of Se- Se-CD@LP-GSH on BV2 and RAW 264.7 cells. Following a 24-hour incubation of the NPs with SH-SY5Y, BV2 and RAW 264.7, cell viability was assessed using the CCK-8 assay. As seen in Fig. 3B, when the concentration of Se-CDs was below 50 µg/mL, the SH-SY5Y cell viability remained above 80%. As illustrated in Fig. 3C, Fig. S5C and Fig. S5D, even at a concentration of 400 µg/mL, Se-CD@LP-GSH did not exhibit significant cytotoxicity on SH-SY5Y, BV2 and RAW 264.7 cells.

Cerebral I/R injury is closely associated with oxidative stress, as the brain is particularly vulnerable to oxidative damage [40]. In vitro, OGD/R model is widely used to simulate ischemic stroke and investigate potential neuroprotective strategies. According to the results of the CCK-8 assay, OGD/R treatment reduced the viability of SH-SY5Y cells to 51.73% (Fig. 3D). However, pretreatment with varying concentrations of Se-CD@LP-GSH significantly improved cell viability, with the most pronounced effect observed at 200 µg/mL, indicating that Se-CD@LP-GSH effectively attenuated OGD/R-induced cytotoxicity (Fig. 3D). These findings were visually confirmed by the live/dead cell assay (Fig. 3E), which showed a marked increase in red fluorescence (dead cells) following OGD/R treatment compared to the control group. Treatment with Se-CD@LP-GSH dose-dependently decreased the proportion of dead cells. Apoptosis, one of the primary forms of cell death triggered by OGD/R, was further evaluated using Annexin V and PI co-staining, followed by flow cytometry analysis. As presented in Fig. 3F, OGD/R significantly increased the proportion of apoptotic cells to approximately 40%, predominantly in the early stage of apoptosis. In contrast, Se-CD@LP-GSH treatment reduced apoptosis in a dose-dependent manner, further demonstrating its cytoprotective effect. Mitochondria, the main source of intracellular ROS, play a pivotal role in oxidative stress-related cell death. To assess mitochondrial function, mitochondrial membrane potential (ΔΨm) was evaluated by JC-1 staining, a commonly used method for detecting mitochondrial depolarizatio. OGD/R caused a substantial loss of ΔΨm, indicative of mitochondrial dysfunction. Notably, Se-CD@LP-GSH pretreatment effectively mitigated this reduction in a concentration-dependent manner, suggesting that it helped preserve mitochondrial integrity (Fig. 3G, H). Since excessive ROS production is a hallmark of I/R injury, intracellular ROS levels were measured using the DCFH-DA fluorescent probe. Fluorescence microscopy revealed a significant increase in green fluorescence following OGD/R, indicative of elevated ROS levels (Fig. 3I). Pretreatment with Se-CD@LP-GSH led to a dose-dependent decrease in fluorescence intensity. These results were further corroborated by flow cytometry analysis, which quantitatively confirmed the ROS-scavenging effects of Se-CD@LP-GSH (Fig. 3J, S6). In summary, Se-CD@LP-GSH exerts neuroprotective effects in the OGD/R model by reducing oxidative stress, preserving mitochondrial membrane potential, inhibiting apoptosis, and enhancing overall cell survival. These findings underscore its potential as a therapeutic candidate for neurological disorders associated with I/R injury.

The neuroprotective effect of Se-CD@LP-GSH in the tMCAo model

Before intravenous injection, we first assessed the biosafety of Se-CD@LP-GSH. As shown in Fig. S7, the hemolysis assay indicated that Se-CD@LP-GSH did not induce significant hemolytic reactions. Subsequently, mice were intravenously injected with Se-CD@LP-GSH at a dose of 20 mg/kg, and blood samples along with major organs were collected on Day 0, 7, and 30 post-injection. H&E staining (Fig. S8A) revealed no obvious histopathological differences in the heart, liver, spleen, lungs, kidneys, or brain at different time points after injection. Furthermore, routine hematological and biochemical analyses were performed on the collected blood samples. As illustrated in Fig. S8B, both hematological parameters and biochemical indices showed minimal variation over time and remained within normal ranges. These findings demonstrated that Se-CD@LP-GSH exhibited favorable biosafety and was suitable for further evaluation in the treatment of cerebral I/R injury.

To further verify the protective effects of Se-CD@LP-GSH in vivo, the tMCAo mouse model was established to mimic cerebral I/R in the right brain. The mice were randomly divided into four groups: sham, tMCAo+Saline, tMCAo + Se-CD@LP-GSH (2.5 mg/kg), tMCAo + Se-CD@LP-GSH (5 mg/kg). The experimental groups received distinct pharmacological treatments via the caudal vein at 2 h post-surgery and again on the third postoperative day. On the fourth postoperative day, all mice were euthanized, and brain tissues were collected (Fig. 4A).

TTC staining was used to characterize the infarct volume of tMCAo mice in each group. As seen in Fig. 4B and C, tMCAo mice treated with Saline exhibited a significantly larger infarct area compared to the sham group. However, treatment with Se-CD@LP-GSH for three days resulted in a dose-dependent reduction in infarct volume. Furthermore, histopathological analysis of brain tissues was conducted. As shown in Fig. 4D and Fig. 4E, HE staining demonstrated that the tMCAo group exhibited the largest necrotic area in brain tissue, whereas Se-CD@LP-GSH significantly reduced the necrotic region. Nissl staining, an essential method for assessing neuronal damage in brain tissue, further confirmed these findings. As illustrated in Fig. 4F and G, compared with the sham-operated group, the number of Nissl bodies was markedly reduced following I/R injury, and Nissl bodies in the infarct area displayed irregular morphology. However, treatment with Se-CD@LP-GSH significantly increased the number of Nissl bodies. Oxidative stress induced by ROS is a key factor contributing to neuronal damage in the brain after I/R injury. Therefore, we used the Dihydroethidium (DHE) probe to assess ROS levels in mouse brain tissues. As shown in Fig. 4H and I, an increase in red fluorescence indicated a higher level of ROS. The tMCAo + Saline group exhibited markedly elevated ROS levels compared to the sham-operated group. However, treatment with Se-CD@LP-GSH significantly reduced ROS levels, and at a dose of 5 mg/kg. TUNEL/NeuN staining of brain tissue sections is widely used to assess early neuronal apoptosis in the brain. As shown in Fig. 4J and K, in the ipsilateral hippocampal region, the tMCAo+Saline group exhibited a significant increase in green fluorescence and a decrease in red fluorescence compared with the sham-operated group, indicating extensive neuronal apoptosis. However, treatment with different concentrations of Se-CD@LP-GSH markedly reduced neuronal apoptosis. Similar results were observed in the cerebral cortex, further confirming the neuroprotective effects of Se-CD@LP-GSH (Fig. 4L and M). These results demonstrated that Se-CD@LP-GSH effectively mitigated oxidative stress induced by I/R injury in stroke, thereby attenuating the subsequent neuronal damage and death.

Se-CD@LP-GSH attenuated long-term neurological deficits after tMCAo

The brain is the central organ of the nervous system in animals, controlling various physiological functions such as movement and sensation. Cerebral I/R injury often results in neurological damage. We assessed the effect of Se-CD@LP-GSH on the recovery of brain function following cerebral I/R injury using a series of neurobehavioral tests. Edaravone, a potent antioxidant agent, has been shown to improve outcomes following ischemic stroke by scavenging hydroxyl and superoxide radicals, reducing cerebral edema, and thereby preventing delayed neuronal death [40]. Edaravone has been recommended for the treatment of ischemic stroke by some clinical guidelines [41, 42]. In this study, it was used as a therapeutic positive control. According to the drug label and interspecies dose conversion based on body surface area, the initial dose administered to mice was 5 mg/kg. As shown in Fig. 5A, C57BL/6J male mice (8–10 weeks old) were randomly distributed into five groups. The tMCAo model was established on Day 1, and drug treatment was administered at 2 h, Day 3, and Day 5 post-surgery. Survival rate, body weight, modified neurological severity score (mNSS) score, adhesive test, and cylinder test were performed on Day 0, 3, 5, 7, 10, 14, 21, and 28. The water maze test was conducted between Days 22 and 28. We first observed the changes in the mortality rate of mice over 28 days following cerebral I/R injury. As shown in Fig. 5B and 100% of mice in the sham group survived for 28 days, while only 40% survived for 7 days after cerebral I/R injury, compared with 60% in the 2.5 mg/kg Se-CD@LP-GSH group, 80% in the 5 mg/kg Se-CD@LP-GSH group, and 80% in the Edaravone group. On Day 14, survival rates had decreased to 10%, 40%, 70%, and 50% in the tMCAo, 2.5 mg/kg Se-CD@LP-GSH, 5 mg/kg Se-CD@LP-GSH, and Edaravone groups, respectively. All mice in the model group treated with Saline had died by Day 23. Ultimately, the Se-CD@LP-GSH 5 mg/kg treatment group exhibited the highest survival rate of 70% by Day 28, significantly surpassing both the Edaravone treatment group and the 2.5 mg/kg Se-CD@LP-GSH group. Body weight is also a critical indicator of the recovery status of the mice. Within 3–5 days post-surgery, all groups showed significant body weight loss. However, all treatment groups exhibited varying degrees of body weight recovery, with the Se-CD@LP-GSH 5 mg/kg group demonstrating the least weight loss and the fastest recovery to baseline levels (Fig. 5C). In this study, we used a mNSS to assess the motor, sensory, balance, and reflex function deficits in mice following cerebral I/R injury. As illustrated in Fig. 5D, the cerebral I/R injury severely impaired the sensory-motor function of the mice. In the tMCAo group, the mNSS scores increased over time. In contrast, in the treatment groups, the mNSS scores peaked at Days 3–5 and then gradually decreased, with the Se-CD@LP-GSH 5 mg/kg group showing significantly better recovery compared to the 2.5 mg/kg group and the Edaravone group. The adhesive test, which requires mice to sense a small sticker on the affected forepaw and coordinate their movements to remove it, provides a more direct evaluation of the mice’s tactile and sensory-motor responses. As shown in Fig. 5E and F, mice took significantly longer to touch and remove the sticker compared to the sham group after cerebral I/R injury. The Se-CD@LP-GSH 5 mg/kg group demonstrated the shortest time to remove the sticker, further supporting the role of Se-CD@LP-GSH in promoting the recovery of long-term neurological deficits following cerebral I/R injury. In addition, the results of the cylinder test showed that mice in the Se-CD@LP-GSH 5 mg/Kg group exhibited the asymmetric rate more closely to that of the sham group, which confirmed that mice in the Se-CD@LP-GSH 5 mg/kg group had better recovery in physical coordination (Fig. 5G).

Fig. 5 — Se-CD@LP-GSH attenuated long-term neurological deficits induced after tMCAo. A) Schematic diagram of long-term neurological tests and experimental time axis (from Day 0 to Day 28). B) Survival rate changes over time. C) Body weight changes over time. D) mNSS scores of mice in each group. E-F) Statistical analysis of the adhesive removal test. G) Statistical analysis of the cylinder test. H-K) The results of the Morris water maze; Representative swimming trajectory plots used to assess learning and memory abilities in mice during the Morris water maze test (H); Statistical analysis of the latency to target (I); Statistical analysis of the distance to target (J); Statistical analysis of the time in target zone (K). (n = 4–10). (Data are expressed as mean ± SEM. I: tMCAo+Saline; II: tMCAo + Se-CD@LP-GSH (2.5 mg/kg); III: tMCAo + Se-CD@LP-GSH (5 mg/kg); IV: tMCAo+Edaravone (5 mg/kg). *P < 0.05, **P < 0.01, ***P < 0.001)

Finally, the water maze test was utilized to evaluate the effect of cerebral I/R injury on the long-term learning and memory abilities of the mice. As seen in Fig. S9, the tMCAo mice treated with Saline died at Day 23 after training, with the tracking trajectory showing that the affected limb exhibited weakness, and the mice rolled in the water. The representative movement trajectories in Fig. 5H and 5I show the variations in learning and memory abilities across the different groups. Although the tMCAo+Saline group mice died and were excluded from the analysis, the remaining data indicated that cerebral I/R injury led to an increased time and distance for the mice to locate the hidden platform. All treatment groups showed some degree of recovery, with the Se-CD@LP-GSH 5 mg/kg group finding the hidden platform in the shortest time and covering the shortest distance (Fig. 5I and J). Following platform removal, the spatial memory ability of the mice was assessed by measuring the time spent in the third quadrant. As shown in Fig. 5K, compared to the sham group, mice in the tMCAo group spent less time in the target quadrant, suggesting that cerebral I/R injury impaired spatial memory. After drug treatment, recovery of memory was observed, with the Se-CD@LP-GSH 5 mg/kg group spending the most time in the target area. The results of long-term neurobehavioral assessments strongly suggested that Se-CD@LP-GSH had substantial potential to attenuate cerebral I/R injury induced by tMCAo and to promote the recovery of neurological function.

Neuronal pyroptosis played a crucial role in the Se-CD@LP-GSH–mediated protection against cerebral I/R injury

To delineate molecular changes occurring after acute cerebral I/R injury and clarify the underlying mechanisms of Se-CD@LP-GSH in protecting the brain from I/R injury, we performed a transcriptomic study to explore possible genes that might play an important role. Principal component analysis (PCA) revealed distinct clusters of the three groups, which showed differentially expressed transcriptomes among the Sham group, the tMCAo group, and the tMCAo + Se-CD@LP-GSH group (Fig. 6A). DEGs were determined using P < 0.05 and a log2 fold change (FC) threshold, with genes showing log2(FC) > 1.5 classified as upregulated and those with log2(FC) < -1.5 as downregulated. Based on this criterion, a heatmap and a volcano plot were generated (Fig. 6B). The heatmap of transcriptomic data displayed significantly different mRNA profiles between the Sham group and the tMCAo group. In order to show the difference between the tMCAo group and the Se-CD@LP-GSH-treated group, the heatmap of the two groups was made (Fig. S10). Compared to the tMCAo group, transcriptomic analysis revealed 928 differentially expressed genes in response to Se-CD@LP-GSH treatment, among which 136 were significantly upregulated and 792 were downregulated (Fig. 6C).

Fig. 6 — Transcriptome Analysis for the Identification of Key Biological Pathways. A) PCA analysis of Sham, tMCAo, and tMCAo + Se-CD@LP-GSH groups. B) Heatmap of differentially expressed genes in Sham, tMCAo, and tMCAo + Se-CD@LP-GSH groups (fold change > 1.5 and P value < 0.05). C) Volcano plots for DEGs in Sham, tMCAo, and tMCAo + Se-CD@LP-GSH groups. D) KEGG pathway enrichment scatter plot of DEGs between Sham and tMCAo groups. E) KEGG pathway enrichment scatter plot of DEGs between tMCAo and tMCAo + Se-CD@LP-GSH groups. F) Heatmap of the differential expression of pyroptosis-related genes between tMCAo and tMCAo + Se-CD@LP-GSH groups. G) Representative fluorescence images of GSDMD protein expression. H. Western blot and statistical analysis of GSDMD and GSDMD-N protein expression. (Data are expressed as mean ± SEM, n = 3. Groups: I. Sham+Saline; II, Sham + Se-CD@LP-GSH; III, tMCAo+Saline; IV, tMCAo + Se-CD@LP-GSH. *P < 0.05, **P < 0.01, ***P < 0.001)

To further investigate the potential molecular biological mechanisms underlying cerebral I/R injury and the effects of Se-CD@LP-GSH, KEGG pathway analysis was conducted to identify enriched biological pathways associated with the DEGs. As shown in Fig. 6D, compared to the sham group, numerous pathways associated with cell death, oxidative stress, inflammation, and phagocytosis were significantly altered following cerebral I/R injury. After treatment with Se-CD@LP-GSH, the results further indicated that the DEGs were significantly involved in signaling pathways related to inflammation and cell death, and the alteration in the NOD-like receptor signaling pathway were most pronounced (Fig. 6E). Some important DEGs in NOD-like receptor signaling pathway were clearly shown in the heatmap (Fig. 6F). We found that these DEGs were elevated in the tMCAo group, such as Gasdermin D (GSDMD), Caspase 1, Caspase 4, and so on, while these were decreased after treatment with Se-CD@LP-GSH (Fig. 6C). GSDMD is one of the most important proteins of the gasdermin family [41]. The gasdermin family plays a crucial role as a key executor in regulating pyroptosis in various diseases [42]. The gasdermin family mediates the formation of non-selective ion channels through non-specific interactions with the cytoplasmic membrane. This process results in cellular swelling and the subsequent release of significant quantities of intracellular contents, thereby initiating inflammation and ultimately culminating in cell death [43]. The gasdermin family members (A/B/C/D/E) have been recognized as critical effectors of pyroptosis, among which GSDMD is most directly implicated in its initiation. The process by which GSDMD mediates pyroptotic cell death is regarded as a hallmark mechanism defining pyroptosis. To determine whether neurons undergo pyroptosis, we performed immunofluorescence double staining on mouse brain tissue. As shown in Fig. 6G, green fluorescence represents neuronal cells, while red fluorescence indicates GSDMD. We observed that many green-fluorescent-labeled neurons exhibited red fluorescence staining of GSDMD protein surrounding them in the tMCAo group. After treatment with Se-CD@LP-GSH, the red fluorescence staining of GSDMD protein was significantly reduced. Western blot analysis demonstrated a marked increase in GSDMD protein levels in the tMCAo group compared to the sham group (Fig. 6H). However, treatment with Se-CD@LP-GSH significantly attenuated the upregulation of GSDMD expression. GSDMD comprises two evolutionarily conserved domains: an N-terminal (NT) effector domain and a C-terminal (CT) autoinhibitory domain, bridged by a flexible loop region. GSDMD, in its unmodified form, has not been shown to exhibit cytotoxicity or execute cell lysis functions, as the CT domain oligomerizes with the NT domain, thereby inhibiting its activity. The N-terminal fragment of GSDMD (GSDMD-N) exerts its function by binding to lipids in the inner leaflet of the cell membrane to form pores [44]. Notably, extracellularly free GSDMD-N is incapable of initiating pyroptosis. Western blot analysis revealed an increased level of GSDMD-N protein in the tMCAo group compared to the sham group. However, treatment with Se-CD@LP-GSH led to a reduction in GSDMD-N expression (Fig. 6H). These results suggested that neuronal pyroptosis occurred following cerebral I/R injury, and that Se-CD@LP-GSH could alleviate I/R injury-induced pyroptosis.

Se-CD@LP-GSH modulated NLRP3, Caspase 1, and GPX4 expression in the tMCAo model

Pyroptotic cell death is primarily regulated via the canonical inflammatory pathway, which is mediated by Caspase 1 [42]. Caspase 1 mediates the cleavage of GSDMD, generating GSDMD-N, which exhibits pore-forming activity [45]. As shown in Fig. 6F, transcriptome sequencing results showed that Caspase 1 is upregulated in the tMCAo group and downregulated in the Se-CD@LP-GSH group. Guided by the transcriptome sequencing results, we determined the protein expression levels of Pro-caspase1 in the ischemic penumbra by western blot and observed significantly higher levels in the tMCAo group than in the sham group (Fig. 7B and C). However, after treatment with Se-CD@LP-GSH, the level of Caspase 1 protein was reduced. Caspase 1 typically exists in an inactive zymogen form and becomes activated only upon incorporation into assembled inflammasomes. Within the inflammasome complex, Pro-caspase1 undergoes autocatalytic cleavage, resulting in its conversion to the active form [46]. Meanwhile, western blot analysis demonstrated a notable upregulation of Cleaved caspase 1 protein levels in the tMCAo group compared to the sham group. However, treatment with Se-CD@LP-GSH effectively mitigated this increase, significantly reducing the expression of Cleaved caspase 1 (Fig. 7B and D). Studies have confirmed that Caspase 1 has a positive impact on the activation of IL-1β and IL-18 precursors, converting them into their mature forms to promote cell death [47, 48]. Immunohistochemistry results demonstrated a significant increase in IL-1β expression in the ischemic penumbra of tMCAo group than the sham group (Fig. 7E). However, the elevation was markedly reduced following treatment with Se-CD@LP-GSH. Similarly, qRT-PCR analysis revealed an upregulation of IL-1β in the tMCAo group, whereas the gene expression levels were partially decreased following Se-CD@LP-GSH treatment (Fig. 7F). The IL-1β ELISA kit was employed to quantitatively evaluate the protein level of IL-1β. As shown in Fig. 7G, IL-1β levels were increased in the tMCAo group and were partially reduced after Se-CD@LP-GSH treatment. The immunohistochemistry, qRT-PCR, and ELISA results for IL-18 showed a consistent trend with IL-1β (Fig. 7H-J). In addition to Pro-caspase 1, the NOD-like receptor (NLR) family plays a key role in inflammasome assembly by acting as a molecular sensor that recognizes exogenous and endogenous danger signals, such as potassium efflux, ROS production, and double-stranded DNA damage [49]. Among them, NLR family pyrin domain-containing 3 (NLRP3) serves as a crucial pattern recognition receptor in the cytoplasm, possessing both autoinhibitory and signal recognition capabilities [50]. As shown in Fig. 7B and K-M, the increased expression of NLRP3 was confirmed in the tMCAo group compared to the sham group, whereas this increase was markedly attenuated following treatment with Se-CD@LP-GSH. As shown in Fig. 7B and M, western blot indicated that Se-CD@LP-GSH did not directly alter NLRP3 expression. However, a significant release of ROS was observed during I/R injury in vivo and OGD/R in vitro in our experiment. Given that ROS could activate the NLRP3 inflammasome, modulating ROS levels may indirectly regulate NLRP3 expression [49, 51]. Our previous findings had demonstrated that Se-CD@LP-GSH effectively scavenged ROS both in vitro and in vivo. Consequently, these results suggested that Se-CD@LP-GSH might inhibit NLRP3 expression by reducing ROS levels, thereby mitigating neuronal pyroptosis. Furthermore, studies have found the crucial role of Selenium in regulating ferroptotic cell death, primarily through its co-translational incorporation into selenocysteine-containing proteins such as GPX4 [17]. As shown in Fig. 6F, KEGG pathway analysis also revealed alterations in ferroptosis-related signaling following treatment with Se-CD@LP-GSH, although it was not the predominant altered pathway. While the precise molecular mechanisms linking lipid peroxidation to ferroptosis remain unclear, GPX4, a selenoprotein, is recognized as a key regulator in this process. GPX4 catalyzes the reduction of lipid peroxides and mitigates the accumulation of lipid ROS, thereby protecting mitochondrial function and preventing cells from ferroptotic death. Based on the established contribution of GPX4 to ferroptosis regulation, we investigated whether Se-CD@LP-GSH modulates GPX4 expression, thereby protecting the mitochondrial function and then mitigating ROS-induced neuronal pyroptosis (Fig. 7A). To assess GPX4 expression in neurons, we performed immunofluorescence staining in the ischemic penumbra. As shown in Fig. 7N, red fluorescence-labeled GPX4 was significantly reduced in the tMCAo group compared to the other three groups. Interestingly, we observed a marked increase in GPX4 expression in the sham mice treated with Se-CD@LP-GSH compared to the untreated sham group. Notably, Se-CD@LP-GSH administration in tMCAo mice not only upregulated GPX4 expression but also resulted in co-localization of red fluorescence-labeled GPX4 with green fluorescence-labeled neurons. Western blot and qRT-PCR analyses further confirmed that GPX4 expression was downregulated in the tMCAo group but significantly increased in both sham and tMCAo mice following Se-CD@LP-GSH treatment (Fig. 7O-Q).

Fig. 7 — Se-CD@LP-GSH modulated the expression of NLRP3, Caspase 1, and the selenium-associated protein GPX4 in the tMCAo model. A) Schematic illustration of the potential molecular signaling pathways regulated by Se-CD@LP-GSH at the mechanistic level. B) Western blot of Pro-caspase 1 and Cleaved caspase 1 protein expression. C and D) Statistical analysis of Pro-caspase 1 (C) and Cleaved caspase 1 (D) protein expression. **E-G**) IHC staining (E), gene (F), and protein (G) expression level of IL-1β. **H-J**) IHC staining (H), gene (I), and protein (J) expression levels of IL-1β. K) Representative fluorescence images of NLRP3 protein expression. L) Gene expression of NLRP3. M) Protein quantification analysis of NLRP3 western blot results. N) Representative fluorescence images of GPX4 protein expression. **O-Q**) Western blot (O), gene expression (P) and protein quantification analysis (Q) of GPX4. (Data are expressed as mean ± SEM, n = 3. Groups: I: Sham+Saline; II: Sham + Se-CD@LP-GSH; III: tMCAo+Saline; IV: tMCAo + Se-CD@LP-GSH. *P < 0.05, **P < 0.01, ***P < 0.001)

Se-CD@LP-GSH upregulated GPX4 and inhibited NLRP3/GSDMD-mediated pyroptosis

To validate our hypothesis, HT-22 cells were subjected to OGD/R to mimic tMCAo-induced brain neurons in vitro, considering species homology. As shown in Fig. 8A, immunofluorescence results indicated that the expression of GSDMD was elevated in the OGD/R-induced HT-22 cell model compared to the control group. However, when Se-CD@LP-GSH was co-incubated with HT-22 cells before and after hypoxia, the expression of GSDMD was relatively reduced. The immunofluorescence results of NLRP3 also exhibited a similar trend (Fig. 8B). Western blot analysis revealed an increased level of GSDMD and NLRP3 protein in the OGD/R group compared to the control group. However, Se-CD@LP-GSH co-incubated with HT-22 cells led to a reduction in GSDMD and NLRP3 expression in a dose-dependent manner (Fig. 8C). Further, the mRNA expression of NLRP3, IL-1β, and IL-18 was examined. As shown in Fig. 8D, the mRNA expression of NLRP3, IL-1β, and IL-18 was all increased in the OGD/R group and decreased after co-incubation with Se-CD@LP-GSH in a dose-dependent manner. These results suggested that neuronal pyroptosis occurred in the OGD/R-induced HT-22 cell model, and that Se-CD@LP-GSH could alleviate OGD/R-induced pyroptosis.

Fig. 8 — Se-CD@LP-GSH upregulated GPX4 and inhibited NLRP3/GSDMD-mediated Pyroptosis in OGD/R-treated HT-22 cells. A and B) Representative fluorescence images of GSDMD and NLRP3 protein expression. C) Western blot of GSDMD and NLRP3 protein expression. D) Relative gene expression of NLRP3, IL-1β, and IL-18 in the OGD/R model of HT-22 treated with Se-CD@LP-GSH (n = 3). E) Representative fluorescence images of GPX4 protein expression. F) Western blot of GPX4 protein expression. G) Western blotting of NLRP3 and GPX4 expression in the OGD/R model of HT-22 cells treated with Se-CD@LP-GSH and si-GPX4. H) Annexin V/PI double staining was performed to evaluate OGD/R-induced apoptosis in HT-22 cells treated with Se-CD@LP-GSH and si-GPX4. (Data are expressed as mean ± SEM.)

The ability of selenium and Se-containing NPs to enhance cellular GPX4 expression has been demonstrated in various disease models [20, 24, 52]. Ishraq Alim et al. confirmed that Selenium could enhance GPX4 expression in neurons in a mouse stroke model [53]. In this study, immunofluorescence and western blot analyses were employed to evaluate the expression levels of GPX4. As illustrated in Fig. 8E, GPX4 expression was reduced in HT-22 cells subjected to OGD/R, whereas co-incubation with Se-CD@LP-GSH led to an upregulation of GPX4 in both control and OGD/R-treated cells. Western blot showed the same result, and Se-CD@LP-GSH co-incubated with HT-22 cells led to an increase in GPX4 expression in a dose-dependent manner (Fig. 8F).

In our previous experiment, we confirmed that Se-CD@LP-GSH reduced ROS production in an OGD/R-induced neuronal model. Based on these findings, we hypothesized that Se-CD@LP-GSH mitigated neuronal pyroptosis by enhancing GPX4 expression, thereby reducing OGD/R-induced ROS generation and subsequently downregulating NLRP3 expression. Next, we transfected HT-22 cells with si-GPX4 to block GPX4 expression and observed the effects of Se-CD@LP-GSH on NLRP3 expression. As shown in Fig. 8G, western blot analysis showed that the expression of NLRP3 did not significantly change in normal HT-22 cells transfected with si-GPX4. However, in the OGD/R-treated HT-22 cells, the expression of GPX4 had a significant decrease, and the expression of NLRP3 had a significant increase. After co-incubation with Se-CD@LP-GSH, GPX4 expression significantly increased and the expression of NLRP3 significantly decreased; However, this effect was reversed in OGD/R-induced HT-22 cells transfected with si-GPX4. Subsequently, apoptotic flow cytometry was used to assess whether si-GPX4 could reverse the protective effect of Se-CD@LP-GSH against OGD/R-induced apoptosis in HT-22 cells. As seen in Fig. 8H, Se-CD@LP-GSH treatment reduced the number of OGD-induced apoptotic cells from 18.91% to 6.22%, and si-GPX4 reversed the number up to 12.76%. These results suggested that Se-CD@LP-GSH alleviated NLRP3-mediated pyroptosis, at least in part, by enhancing GPX4 expression and scavenging ROS.

Discussion

Ischemic stroke is a devastating disease with high incidence and mortality. I/R injury leads to excessive production of ROS, making targeted ROS scavenging a promising therapeutic strategy. Owing to their favorable biocompatibility, high bioavailability and potent ROS–scavenging activity, nanozymes have been increasingly investigated for treating ischemic stroke [15, 54]. In addition, cerebral ischemia is accompanied by systemic selenium depletion [55]. When selenium availability is limited, nutritional selenium is preferentially distributed to the brain and to the testes in males [56, 57], highlighting the essential role of selenium in the brain. Based on these considerations, we synthesized and applied Se-CDs for the treatment of ischemic stroke. Se-CDs combine intrinsic nanozyme-like ROS–scavenging capability with the ability to replenish selenium, and our in vivo and in vitro experiments demonstrated their protective effects against cerebral I/R injury. To overcome the systemic distribution pattern of selenium and achieve pharmacological selenium levels in the brain, we encapsulated Se-CDs in targeted liposomes to enhance brain delivery. GSH–modified liposomes have been shown to facilitate brain targeting, although this strategy may also risk excessive accumulation. Nanoparticle clearance from the brain appears to depend more on particle composition than on individual physicochemical parameters such as size, zeta potential, surface functionalization, or rigidity. Clearance occurs primarily through perivascular glymphatic pathways that mediate exchange between cerebrospinal fluid and interstitial fluid [58, 59]. After being internalized by microglia, organic nanoparticles facilitate extracellular vesicles secretion and limit particle retention. In contrast, inorganic nanoparticles accumulate in microglia, trigger apoptosis, inflammation and lysosomal related transcriptional responses, enhance ERK1/2 signaling, and consequently suppress extracellular vesicles biogenesis [60]. Because our therapeutic platform uses liposome encapsulation, it may facilitate the clearance of Se-CDs from the brain. In this study, a 30-day safety evaluation revealed no apparent abnormalities in major organs or in circulating immune cell populations, supporting the biosafety of this formulation.

Previous studies have shown that, once internalized by cells, selenium nanoparticles are primarily metabolized into selenate, selenomethionine, selenite, Semethylselenocysteine, and selenocysteine, with selenocysteine likely serving as the predominant bioactive form in vivo [61]. Selenium nanoparticles can enhance the expression of multiple Selenium-related proteins—such as GPX1, GPX3, GPX4, SeLK and so on—which play essential roles in anti-inflammatory and antioxidant responses [61, 62]. Among these selenoproteins, the ability of selenium to enhance neuronal GPX4 expression after stroke has been well documented. Pharmacological selenium further augments GPX4 through coordinated activation of the transcription factors TFAP2C and Sp1, thereby promoting neuronal protection [53]. Consistent with this mechanism, Se-CD@LP-GSH treatment in our experiments increased GPX4 expression, reduced oxidative stress, and preserved mitochondrial homeostasis.

In the present study, we found that Se-CDs upregulated GPX4 and suppressed NLRP3/GSDMD-mediated pyroptosis in HT-22 cells subjected to OGD/R. GPX4 knockdown partially reversed the Se-CDs–mediated inhibition of NLRP3, which is consistent with a model in which reduced GPX4 synthesis precipitates mitochondrial dysfunction, increases ROS production, and thereby promotes NLRP3 inflammasome activation and neuronal pyroptosis following ischemic insult. GPX4 is also a central regulator of ferroptosis, but the relationship between ferroptosis and pyroptosis was not examined in depth in this work. Emerging evidence suggests that mitochondria are likely to play a pivotal role in the crosstalk between these two cell-death modalities [63]. Mitophagy clears damaged mitochondria and mtROS, thereby limiting both pyroptosis and ferroptosis, whereas ferritinophagy and mitochondrial iron accumulation elevate labile iron and ROS to drive lipid peroxidation and ferroptotic death. mtROS also amplifies pyroptosis via inflammasome- and caspase-dependent pathways [64], and although mitophagy can be protective, processes such as complex I inhibition or FUNDC1-mediated GPX4 recruitment may paradoxically increase mitophagy-dependent mtROS or deplete cytosolic GPX4, sensitizing cells to ferroptosis [65]. In our study, Se-CDs exhibited strong intrinsic ROS-scavenging activity and concurrently promoted GPX4 expression, providing a dual, synergistic mechanism to protect neurons from OGD/R-induced pyroptosis. Nonetheless, the precise mechanistic interplay between ferroptosis and pyroptosis — and the extent to which Se-CDs modulate this crosstalk — remains incompletely defined and warrants further investigation.

Although our findings indicate that Se-CDs enhance therapeutic efficacy in ischemic stroke, several challenges remain before such an approach can be translated into clinical practice. The synthesis of Se-CDs needs to be standardized, and batch-to-batch reproducibility must be rigorously established. In addition, the long-term safety of Se-CDs requires further verification. Finally, the pathophysiological mechanisms of cerebral ischemia are highly complex, and the effects of Se-CDs on other brain parenchymal and stromal cell types, as well as on the broader cerebral microenvironment, warrant further investigation.

Conclusion

In this study, we engineered Se-CD@LP-GSH nanoparticles characterized by superior biocompatibility, safety, and enhanced brain-targeting capabilities. In vitro analyses demonstrated that these nanoparticles effectively scavenged ROS and mitigated oxidative damage in neuronal cells. In vivo experiments revealed their efficacy in alleviating cerebral I/R injury and promoting neurological functional recovery. Mechanistically, our findings suggest that cerebral I/R injury triggers multiple forms of neuronal cell death, with pyroptosis identified as a significant contributor. Treatment with Se-CD@LP-GSH upregulated the expression of GPX4, a pivotal selenoprotein, thereby preserving mitochondrial function and suppressing ROS generation. This suppression of ROS subsequently downregulated the expression of NLRP3 and GSDMD, thereby mitigating pyroptosis associated with I/R injury. Collectively, these results indicate that Se-CD@LP-GSH nanoparticles possess potential as a therapy for cerebral I/R injury by attenuating oxidative stress and modulating regulated forms of neuronal cell death.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 2^{(725KB, pdf)}

Supplementary Material 1^{(2.9MB, docx)}

Acknowledgements

We thank Dr. Zijun Ren at the Instrument Analysis Center of Xi’an Jiaotong University for support during TEM experiments, Senior Engineer Lai Baochang from the Instrumentation and Equipment Center of the Translational Medicine Institute at Xi'an Jiaotong University for his technical support in this work. The authors express gratitude to Beijing Zhongkebaice Technology Service Co., Ltd. for conducting critical quantitative analyses that underpinned this study (www.zkbaice.cn). The authors appreciate the open access to the public databases that supported this work. (Biorender).

Author contributions

Jiaxuan Hou.: Investigation, Methodology, Experiments, Data curation, Formal analysis, Writing-original draft. Li Yao: Methodology, Data curation, Formal analysis, Writing- original draft. Yane Li: Conceptualization, Methodology, Formal analysis. Enrui Xie: Methodology, Formal analysis. Jiawei Zhang: Data curation, Writing review & editing. Hao Wu: Methodology, Formal analysis. Yuanyuan Zhu: Investigation, Data curation. Zhichao Deng: Data curation. Chenxi Xu: Data curation. Zongru Han: Revised draft & editing . Lu Bai: Conceptualization. Mingzhen Zhang: Supervision, Funding acquisition. Shaoying Lu: Conceptualization, Resources . Runqing Li: Conceptualization, Writing-review & editing. Hui Cai: Conceptualization, Writing-review & editing, Resources, Funding acquisition, Project administration.

Funding

This work was supported by the International Science and Technology Cooperation and Exchange Program of Shaanxi Province, China (No. 2015KW-052), Natural Science Foundation of Shannxi Province (2024JC-YBMS-781).

Data availability

Data will be available on request.

Declarations

Ethics approval and consent to participate

All animal experiments were conducted in accordance with institutional guidelines and were approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University, China (NO. XJTUAE2024-2118).

Consent for publication

All authors agreed to be published.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

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

Contributor Information

Shaoying Lu, Email: robertlu@mail.xjtu.edu.cn.

Runqing Li, Email: lirunqing718@xjtu.edu.cn.

Hui Cai, Email: caihui@xjtufh.edu.cn.

References

1.Feigin VL, Nguyen G, Cercy K, Johnson CO, Alam T, Parmar PG, et al. Global, Regional, and Country-Specific lifetime risks of Stroke, 1990 and 2016. N Engl J Med. 2018;379(25):2429–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Campbell BCV, De Silva DA, Macleod MR, Coutts SB, Schwamm LH, Davis SM, et al. Ischaemic stroke. Nat Reviews Disease Primers. 2019;5(1):70. [DOI] [PubMed] [Google Scholar]
3.Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, et al. An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American heart association/American stroke association. Stroke. 2013;44(7):2064–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Widimsky P, Snyder K, Sulzenko J, Hopkins LN, Stetkarova I. Acute ischaemic stroke: recent advances in reperfusion treatment. Eur Heart J. 2023;44(14):1205–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Adeoye O, Hornung R, Khatri P, Kleindorfer D. Recombinant tissue-type plasminogen activator use for ischemic stroke in the united states: a doubling of treatment rates over the course of 5 years. Stroke. 2011;42(7):1952–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Wu S, Wu B, Liu M, Chen Z, Wang W, Anderson CS, et al. Stroke in china: advances and challenges in epidemiology, prevention, and management. Lancet Neurol. 2019;18(4):394–405. [DOI] [PubMed] [Google Scholar]
7.Andrabi SS, Parvez S, Tabassum H. Ischemic stroke and mitochondria: mechanisms and targets. Protoplasma. 2020;257(2):335–43. [DOI] [PubMed] [Google Scholar]
8.Chamorro Á, Dirnagl U, Urra X, Planas AM. Neuroprotection in acute stroke: targeting excitotoxicity, oxidative and nitrosative stress, and inflammation. Lancet Neurol. 2016;15(8):869–81. [DOI] [PubMed] [Google Scholar]
9.Guo Z, Yu S, Chen X, Ye R, Zhu W, Liu X. NLRP3 is involved in ischemia/Reperfusion injury. CNS Neurol Disord Drug Target. 2016;15(6):699–712. [DOI] [PubMed] [Google Scholar]
10.Liu H, Wang L, Weng X, Chen H, Du Y, Diao C, et al. Inhibition of Brd4 alleviates renal ischemia/reperfusion injury-induced apoptosis and Endoplasmic reticulum stress by blocking FoxO4-mediated oxidative stress. Redox Biol. 2019;24:101195. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Long J, Sun Y, Liu S, Yang S, Chen C, Zhang Z, et al. Targeting pyroptosis as a preventive and therapeutic approach for stroke. Cell Death Discov. 2023;9(1):155. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Xu P, Tao C, Zhu Y, Wang G, Kong L, Li W et al. TAK1 mediates neuronal pyroptosis in early brain injury after subarachnoid hemorrhage. J Neuroinflamm. 2021;18(1):188. [DOI] [PMC free article] [PubMed]
13.Li J, Xu P, Hong Y, Xie Y, Peng M, Sun R et al. Lipocalin-2-mediated astrocyte pyroptosis promotes neuroinflammatory injury via NLRP3 inflammasome activation in cerebral ischemia/reperfusion injury. J Neuroinflamm. 2023;20(1):148. [DOI] [PMC free article] [PubMed]
14.Zhao P, Deng Y, Xiang G, Liu Y. Nanoparticle-Assisted sonosensitizers and their biomedical applications. Int J Nanomed. 2021;16:4615–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Zhang Y, Zhang L, Wang M, Li P. The applications of nanozymes in neurological diseases: from mechanism to design. Theranostics. 2023;13(8):2492–514. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Ren X, Chen D, Wang Y, Li H, Zhang Y, Chen H, et al. Nanozymes-recent development and biomedical applications. J Nanobiotechnol. 2022;20(1):92. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Ingold I, Berndt C, Schmitt S, Doll S, Poschmann G, Buday K, et al. Selenium utilization by GPX4 is required to prevent Hydroperoxide-Induced ferroptosis. Cell. 2018;172(3):409–22. e21. [DOI] [PubMed] [Google Scholar]
18.Daneshpour A, Leite MEN, Wagner KH, Sabico S, Al-Daghri NM, Aldisi D, et al. Selenium and brain aging: A comprehensive review with a focus on hippocampal neurogenesis. Ageing Res Rev. 2025;112:102898. [DOI] [PubMed] [Google Scholar]
19.Turovsky EA, Baryshev AS, Plotnikov EY. Selenium nanoparticles in protecting the brain from stroke: possible signaling and metabolic mechanisms. Nanomaterials. 2024;14(2):160. [DOI] [PMC free article] [PubMed]
20.Li J, Fu C, Feng B, Liu Q, Gu J, Khan MN, et al. Polyacrylic Acid-Coated Selenium‐Doped carbon Dots inhibit ferroptosis to alleviate Chemotherapy‐Associated acute kidney injury. Adv Sci. 2024;11:28. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Bai B, Qi S, Yang K, Yu X, Jian R, Zhang T, et al. Self-Assembly of Selenium-Doped carbon quantum Dots as antioxidants for hepatic Ischemia-Reperfusion injury management. Small. 2023;19(30):e2300217. [DOI] [PubMed] [Google Scholar]
22.Wang S, Chen Y, Han S, Liu Y, Gao J, Huang Y, et al. Selenium nanoparticles alleviate ischemia reperfusion injury-induced acute kidney injury by modulating GPx-1/NLRP3/Caspase-1 pathway. Theranostics. 2022;12(8):3882–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Tian R, Ma H, Ye W, Li Y, Wang S, Zhang Z et al. Se-Containing MOF coated Dual‐Fe‐Atom nanozymes with Multi‐Enzyme cascade activities protect against cerebral ischemic reperfusion injury. Adv Funct Mater. 2022;32(36):2204025.
24.Shi Y, Han L, Zhang X, Xie L, Pan P, Chen F. Selenium alleviates cerebral Ischemia/Reperfusion injury by regulating oxidative Stress, mitochondrial fusion and ferroptosis. Neurochem Res. 2022;47(10):2992–3002. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Large DE, Abdelmessih RG, Fink EA, Auguste DT. Liposome composition in drug delivery design, synthesis, characterization, and clinical application. Adv Drug Deliv Rev. 2021;176:113851. [DOI] [PubMed] [Google Scholar]
26.Sadeghian N, Shadman J, Moradi A, Ghasem Golmohammadi M, Panahpour H. Calcitriol protects the Blood-Brain barrier integrity against ischemic stroke and reduces vasogenic brain edema via antioxidant and antiapoptotic actions in rats. Brain Res Bull. 2019;150:281–9. [DOI] [PubMed] [Google Scholar]
27.Jiang X, Andjelkovic AV, Zhu L, Yang T, Bennett MVL, Chen J et al. Blood-brain barrier dysfunction and recovery after ischemic stroke. Prog Neurobiol. 2018;163–164:144–71. [DOI] [PMC free article] [PubMed]
28.Gaucher C, Boudier A, Bonetti J, Clarot I, Leroy P, Parent M. Glutathione: antioxidant properties dedicated to nanotechnologies. Antioxidants. 2018;7(5):62. [DOI] [PMC free article] [PubMed]
29.Rip J, Chen L, Hartman R, van den Heuvel A, Reijerkerk A, van Kregten J, et al. Glutathione pegylated liposomes: pharmacokinetics and delivery of cargo across the blood–brain barrier in rats. J Drug Target. 2014;22(5):460–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Rotman M, Welling MM, Bunschoten A, de Backer ME, Rip J, Nabuurs RJ, et al. Enhanced glutathione pegylated liposomal brain delivery of an anti-amyloid single domain antibody fragment in a mouse model for alzheimer’s disease. J Control Release. 2015;203:40–50. [DOI] [PubMed] [Google Scholar]
31.Maussang D, Rip J, van Kregten J, van den Heuvel A, van der Pol S, van der Boom B, et al. Glutathione conjugation dose-dependently increases brain-specific liposomal drug delivery in vitro and in vivo. Drug Discovery Today Technol. 2016;20:59–69. [DOI] [PubMed] [Google Scholar]
32.Birngruber T, Raml R, Gladdines W, Gatschelhofer C, Gander E, Ghosh A, et al. Enhanced doxorubicin delivery to the brain administered through glutathione pegylated liposomal doxorubicin (2B3-101) as compared with generic Caelyx,(^®)/Doxil(^®)--a cerebral open flow microperfusion pilot study. J Pharm Sci. 2014;103(7):1945–8. [DOI] [PubMed] [Google Scholar]
33.Elhissi AM, O’Neill MA, Roberts SA, Taylor KM. A calorimetric study of dimyristoylphosphatidylcholine phase transitions and steroid-liposome interactions for liposomes prepared by thin film and proliposome methods. Int J Pharm. 2006;320(1–2):124–30. [DOI] [PubMed] [Google Scholar]
34.Li F, Li T, Sun C, Xia J, Jiao Y, Xu H. Selenium-Doped carbon quantum Dots for Free‐Radical scavenging. Angew Chem Int Ed. 2017;56(33):9910–4. [DOI] [PubMed] [Google Scholar]
35.Rosenkrans ZT, Sun T, Jiang D, Chen W, Barnhart TE, Zhang Z, et al. Selenium-Doped carbon quantum Dots act as Broad-Spectrum antioxidants for acute kidney injury management. Adv Sci (Weinh). 2020;7(12):2000420. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Han P, Ma N, Ren H, Xu H, Li Z, Wang Z, et al. Oxidation-responsive micelles based on a selenium-containing polymeric superamphiphile. Langmuir. 2010;26(18):14414–8. [DOI] [PubMed] [Google Scholar]
37.Rotruck JT, Pope AL, Ganther HE, Swanson AB, Hafeman DG, Hoekstra WG. Selenium: biochemical role as a component of glutathione peroxidase. Science. 1973;179(4073):588–90. [DOI] [PubMed] [Google Scholar]
38.Li T, Zhao L, Li Y, Dang M, Lu J, Lu Z et al. PPM1K mediates metabolic disorder of branched-chain amino acid and regulates cerebral ischemia-reperfusion injury by activating ferroptosis in neurons. Cell Death Dis. 2023;14(9):634. [DOI] [PMC free article] [PubMed]
39.Bouchez C, Devin A. Mitochondrial biogenesis and mitochondrial reactive oxygen species (ROS): A complex relationship regulated by the cAMP/PKA signaling pathway. Cells. 2019;8(4):287. [DOI] [PMC free article] [PubMed]
40.Uttara B, Singh AV, Zamboni P, Mahajan RT. Oxidative stress and neurodegenerative diseases: a review of upstream and downstream antioxidant therapeutic options. Curr Neuropharmacol. 2009;7(1):65–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Ding J, Wang K, Liu W, She Y, Sun Q, Shi J, et al. Erratum: Pore-forming activity and structural autoinhibition of the gasdermin family. Nature. 2016;540(7631):150. [DOI] [PubMed] [Google Scholar]
42.Yu P, Zhang X, Liu N, Tang L, Peng C, Chen X. Pyroptosis: mechanisms and diseases. Signal Transduct Target Ther. 2021;6(1):128. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Xia S, Hollingsworth LRt, Wu H. Mechanism and regulation of Gasdermin-Mediated cell death. Cold Spring Harb Perspect Biol. 2020;12(3):a036400. [DOI] [PMC free article] [PubMed]
44.Liu X, Zhang Z, Ruan J, Pan Y, Magupalli VG, Wu H, et al. Inflammasome-activated gasdermin D causes pyroptosis by forming membrane pores. Nature. 2016;535(7610):153–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Man SM, Kanneganti TD. Regulation of inflammasome activation. Immunol Rev. 2015;265(1):6–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Sollberger G, Strittmatter GE, Garstkiewicz M, Sand J, Beer HD. Caspase-1: the inflammasome and beyond. Innate Immun. 2014;20(2):115–25. [DOI] [PubMed] [Google Scholar]
47.Bergsbaken T, Fink SL, Cookson BT. Pyroptosis: host cell death and inflammation. Nat Rev Microbiol. 2009;7(2):99–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Shi J, Zhao Y, Wang K, Shi X, Wang Y, Huang H, et al. Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death. Nature. 2015;526(7575):660–5. [DOI] [PubMed] [Google Scholar]
49.Lu A, Magupalli VG, Ruan J, Yin Q, Atianand MK, Vos MR, et al. Unified polymerization mechanism for the assembly of ASC-dependent inflammasomes. Cell. 2014;156(6):1193–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.He Y, Hara H, Núñez G. Mechanism and regulation of NLRP3 inflammasome activation. Trends Biochem Sci. 2016;41(12):1012–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Lin Q, Li S, Jiang N, Shao X, Zhang M, Jin H, et al. PINK1-parkin pathway of mitophagy protects against contrast-induced acute kidney injury via decreasing mitochondrial ROS and NLRP3 inflammasome activation. Redox Biol. 2019;26:101254. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Jia C, Xiang Z, Zhang P, Liu L, Zhu X, Yu R, et al. Selenium-SelK-GPX4 axis protects nucleus pulposus cells against mechanical overloading-induced ferroptosis and attenuates senescence of intervertebral disc. Cell Mol Life Sci. 2024;81(1):49. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Alim I, Caulfield JT, Chen Y, Swarup V, Geschwind DH, Ivanova E, et al. Selenium drives a transcriptional adaptive program to block ferroptosis and treat stroke. Cell. 2019;177(5):1262–e7925. [DOI] [PubMed] [Google Scholar]
54.He L, Huang G, Liu H, Sang C, Liu X, Chen T. Highly bioactive zeolitic imidazolate framework-8-capped nanotherapeutics for efficient reversal of reperfusion-induced injury in ischemic stroke. Sci Adv. 2020;6(12):eaay9751. [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Mirończuk A, Kapica-Topczewska K, Socha K, Soroczyńska J, Jamiołkowski J, Kułakowska A et al. Selenium, Copper, Zinc Concentrations and Cu/Zn, Cu/Se Molar Ratios in the Serum of Patients with Acute Ischemic Stroke in Northeastern Poland-A New Insight into Stroke Pathophysiology. Nutrients. 2021;13(7). [DOI] [PMC free article] [PubMed]
56.Brown DG, Burk RF. Selenium retention in tissues and sperm of rats fed a Torula yeast diet. J Nutr. 1973;103(1):102–8. [DOI] [PubMed] [Google Scholar]
57.Schomburg L, Schweizer U. Hierarchical regulation of Selenoprotein expression and sex-specific effects of selenium. Biochim Biophys Acta. 2009;1790(11):1453–62. [DOI] [PubMed] [Google Scholar]
58.Gu X, Song Q, Zhang Q, Huang M, Zheng M, Chen J, et al. Clearance of two organic nanoparticles from the brain via the paravascular pathway. J Control Release. 2020;322:31–41. [DOI] [PubMed] [Google Scholar]
59.Iliff JJ, Wang M, Liao Y, Plogg BA, Peng W, Gundersen GA, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med. 2012;4(147):147ra11. [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Gao J, Song Q, Gu X, Jiang G, Huang J, Tang Y et al. Intracerebral fate of organic and inorganic nanoparticles is dependent on microglial extracellular vesicle function. Nat Nanotechnol. 2023;19(3):376–386. [DOI] [PubMed]
61.Ouyang J, Deng B, Zou B, Li Y, Bu Q, Tian Y, et al. Oral hydrogel Microbeads-Mediated in situ synthesis of selenoproteins for regulating intestinal immunity and microbiota. J Am Chem Soc. 2023;145(22):12193–205. [DOI] [PubMed] [Google Scholar]
62.Zou B, Xiong Z, Yu Y, Shi S, Li X, Chen T. Rapid Selenoprotein activation by selenium nanoparticles to suppresses osteoclastogenesis and pathological bone loss. Adv Mater. 2024;36(27):e2401620. [DOI] [PubMed] [Google Scholar]
63.Zhao P, Yin S, Qiu Y, Sun C, Yu H. Ferroptosis and pyroptosis are connected through autophagy: a new perspective of overcoming drug resistance. Mol Cancer. 2025;24(1):23. [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Liu Y, Pan R, Ouyang Y, Gu W, Xiao T, Yang H, et al. Pyroptosis in health and disease: mechanisms, regulation and clinical perspective. Signal Transduct Target Ther. 2024;9(1):245. [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Basit F, van Oppen LM, Schöckel L, Bossenbroek HM, van Emst-de Vries SE, Hermeling JC, et al. Mitochondrial complex I Inhibition triggers a mitophagy-dependent ROS increase leading to necroptosis and ferroptosis in melanoma cells. Cell Death Dis. 2017;8(3):e2716. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 2^{(725KB, pdf)}

Supplementary Material 1^{(2.9MB, docx)}

Data Availability Statement

Data will be available on request.

[CR1] 1.Feigin VL, Nguyen G, Cercy K, Johnson CO, Alam T, Parmar PG, et al. Global, Regional, and Country-Specific lifetime risks of Stroke, 1990 and 2016. N Engl J Med. 2018;379(25):2429–37. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Campbell BCV, De Silva DA, Macleod MR, Coutts SB, Schwamm LH, Davis SM, et al. Ischaemic stroke. Nat Reviews Disease Primers. 2019;5(1):70. [DOI] [PubMed] [Google Scholar]

[CR3] 3.Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, et al. An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American heart association/American stroke association. Stroke. 2013;44(7):2064–89. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Widimsky P, Snyder K, Sulzenko J, Hopkins LN, Stetkarova I. Acute ischaemic stroke: recent advances in reperfusion treatment. Eur Heart J. 2023;44(14):1205–15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Adeoye O, Hornung R, Khatri P, Kleindorfer D. Recombinant tissue-type plasminogen activator use for ischemic stroke in the united states: a doubling of treatment rates over the course of 5 years. Stroke. 2011;42(7):1952–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR6] 6.Wu S, Wu B, Liu M, Chen Z, Wang W, Anderson CS, et al. Stroke in china: advances and challenges in epidemiology, prevention, and management. Lancet Neurol. 2019;18(4):394–405. [DOI] [PubMed] [Google Scholar]

[CR7] 7.Andrabi SS, Parvez S, Tabassum H. Ischemic stroke and mitochondria: mechanisms and targets. Protoplasma. 2020;257(2):335–43. [DOI] [PubMed] [Google Scholar]

[CR8] 8.Chamorro Á, Dirnagl U, Urra X, Planas AM. Neuroprotection in acute stroke: targeting excitotoxicity, oxidative and nitrosative stress, and inflammation. Lancet Neurol. 2016;15(8):869–81. [DOI] [PubMed] [Google Scholar]

[CR9] 9.Guo Z, Yu S, Chen X, Ye R, Zhu W, Liu X. NLRP3 is involved in ischemia/Reperfusion injury. CNS Neurol Disord Drug Target. 2016;15(6):699–712. [DOI] [PubMed] [Google Scholar]

[CR10] 10.Liu H, Wang L, Weng X, Chen H, Du Y, Diao C, et al. Inhibition of Brd4 alleviates renal ischemia/reperfusion injury-induced apoptosis and Endoplasmic reticulum stress by blocking FoxO4-mediated oxidative stress. Redox Biol. 2019;24:101195. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Long J, Sun Y, Liu S, Yang S, Chen C, Zhang Z, et al. Targeting pyroptosis as a preventive and therapeutic approach for stroke. Cell Death Discov. 2023;9(1):155. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Xu P, Tao C, Zhu Y, Wang G, Kong L, Li W et al. TAK1 mediates neuronal pyroptosis in early brain injury after subarachnoid hemorrhage. J Neuroinflamm. 2021;18(1):188. [DOI] [PMC free article] [PubMed]

[CR13] 13.Li J, Xu P, Hong Y, Xie Y, Peng M, Sun R et al. Lipocalin-2-mediated astrocyte pyroptosis promotes neuroinflammatory injury via NLRP3 inflammasome activation in cerebral ischemia/reperfusion injury. J Neuroinflamm. 2023;20(1):148. [DOI] [PMC free article] [PubMed]

[CR14] 14.Zhao P, Deng Y, Xiang G, Liu Y. Nanoparticle-Assisted sonosensitizers and their biomedical applications. Int J Nanomed. 2021;16:4615–30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Zhang Y, Zhang L, Wang M, Li P. The applications of nanozymes in neurological diseases: from mechanism to design. Theranostics. 2023;13(8):2492–514. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR16] 16.Ren X, Chen D, Wang Y, Li H, Zhang Y, Chen H, et al. Nanozymes-recent development and biomedical applications. J Nanobiotechnol. 2022;20(1):92. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Ingold I, Berndt C, Schmitt S, Doll S, Poschmann G, Buday K, et al. Selenium utilization by GPX4 is required to prevent Hydroperoxide-Induced ferroptosis. Cell. 2018;172(3):409–22. e21. [DOI] [PubMed] [Google Scholar]

[CR18] 18.Daneshpour A, Leite MEN, Wagner KH, Sabico S, Al-Daghri NM, Aldisi D, et al. Selenium and brain aging: A comprehensive review with a focus on hippocampal neurogenesis. Ageing Res Rev. 2025;112:102898. [DOI] [PubMed] [Google Scholar]

[CR19] 19.Turovsky EA, Baryshev AS, Plotnikov EY. Selenium nanoparticles in protecting the brain from stroke: possible signaling and metabolic mechanisms. Nanomaterials. 2024;14(2):160. [DOI] [PMC free article] [PubMed]

[CR20] 20.Li J, Fu C, Feng B, Liu Q, Gu J, Khan MN, et al. Polyacrylic Acid-Coated Selenium‐Doped carbon Dots inhibit ferroptosis to alleviate Chemotherapy‐Associated acute kidney injury. Adv Sci. 2024;11:28. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Bai B, Qi S, Yang K, Yu X, Jian R, Zhang T, et al. Self-Assembly of Selenium-Doped carbon quantum Dots as antioxidants for hepatic Ischemia-Reperfusion injury management. Small. 2023;19(30):e2300217. [DOI] [PubMed] [Google Scholar]

[CR22] 22.Wang S, Chen Y, Han S, Liu Y, Gao J, Huang Y, et al. Selenium nanoparticles alleviate ischemia reperfusion injury-induced acute kidney injury by modulating GPx-1/NLRP3/Caspase-1 pathway. Theranostics. 2022;12(8):3882–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Tian R, Ma H, Ye W, Li Y, Wang S, Zhang Z et al. Se-Containing MOF coated Dual‐Fe‐Atom nanozymes with Multi‐Enzyme cascade activities protect against cerebral ischemic reperfusion injury. Adv Funct Mater. 2022;32(36):2204025.

[CR24] 24.Shi Y, Han L, Zhang X, Xie L, Pan P, Chen F. Selenium alleviates cerebral Ischemia/Reperfusion injury by regulating oxidative Stress, mitochondrial fusion and ferroptosis. Neurochem Res. 2022;47(10):2992–3002. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Large DE, Abdelmessih RG, Fink EA, Auguste DT. Liposome composition in drug delivery design, synthesis, characterization, and clinical application. Adv Drug Deliv Rev. 2021;176:113851. [DOI] [PubMed] [Google Scholar]

[CR26] 26.Sadeghian N, Shadman J, Moradi A, Ghasem Golmohammadi M, Panahpour H. Calcitriol protects the Blood-Brain barrier integrity against ischemic stroke and reduces vasogenic brain edema via antioxidant and antiapoptotic actions in rats. Brain Res Bull. 2019;150:281–9. [DOI] [PubMed] [Google Scholar]

[CR27] 27.Jiang X, Andjelkovic AV, Zhu L, Yang T, Bennett MVL, Chen J et al. Blood-brain barrier dysfunction and recovery after ischemic stroke. Prog Neurobiol. 2018;163–164:144–71. [DOI] [PMC free article] [PubMed]

[CR28] 28.Gaucher C, Boudier A, Bonetti J, Clarot I, Leroy P, Parent M. Glutathione: antioxidant properties dedicated to nanotechnologies. Antioxidants. 2018;7(5):62. [DOI] [PMC free article] [PubMed]

[CR29] 29.Rip J, Chen L, Hartman R, van den Heuvel A, Reijerkerk A, van Kregten J, et al. Glutathione pegylated liposomes: pharmacokinetics and delivery of cargo across the blood–brain barrier in rats. J Drug Target. 2014;22(5):460–7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Rotman M, Welling MM, Bunschoten A, de Backer ME, Rip J, Nabuurs RJ, et al. Enhanced glutathione pegylated liposomal brain delivery of an anti-amyloid single domain antibody fragment in a mouse model for alzheimer’s disease. J Control Release. 2015;203:40–50. [DOI] [PubMed] [Google Scholar]

[CR31] 31.Maussang D, Rip J, van Kregten J, van den Heuvel A, van der Pol S, van der Boom B, et al. Glutathione conjugation dose-dependently increases brain-specific liposomal drug delivery in vitro and in vivo. Drug Discovery Today Technol. 2016;20:59–69. [DOI] [PubMed] [Google Scholar]

[CR32] 32.Birngruber T, Raml R, Gladdines W, Gatschelhofer C, Gander E, Ghosh A, et al. Enhanced doxorubicin delivery to the brain administered through glutathione pegylated liposomal doxorubicin (2B3-101) as compared with generic Caelyx,(^®)/Doxil(^®)--a cerebral open flow microperfusion pilot study. J Pharm Sci. 2014;103(7):1945–8. [DOI] [PubMed] [Google Scholar]

[CR33] 33.Elhissi AM, O’Neill MA, Roberts SA, Taylor KM. A calorimetric study of dimyristoylphosphatidylcholine phase transitions and steroid-liposome interactions for liposomes prepared by thin film and proliposome methods. Int J Pharm. 2006;320(1–2):124–30. [DOI] [PubMed] [Google Scholar]

[CR34] 34.Li F, Li T, Sun C, Xia J, Jiao Y, Xu H. Selenium-Doped carbon quantum Dots for Free‐Radical scavenging. Angew Chem Int Ed. 2017;56(33):9910–4. [DOI] [PubMed] [Google Scholar]

[CR35] 35.Rosenkrans ZT, Sun T, Jiang D, Chen W, Barnhart TE, Zhang Z, et al. Selenium-Doped carbon quantum Dots act as Broad-Spectrum antioxidants for acute kidney injury management. Adv Sci (Weinh). 2020;7(12):2000420. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Han P, Ma N, Ren H, Xu H, Li Z, Wang Z, et al. Oxidation-responsive micelles based on a selenium-containing polymeric superamphiphile. Langmuir. 2010;26(18):14414–8. [DOI] [PubMed] [Google Scholar]

[CR37] 37.Rotruck JT, Pope AL, Ganther HE, Swanson AB, Hafeman DG, Hoekstra WG. Selenium: biochemical role as a component of glutathione peroxidase. Science. 1973;179(4073):588–90. [DOI] [PubMed] [Google Scholar]

[CR38] 38.Li T, Zhao L, Li Y, Dang M, Lu J, Lu Z et al. PPM1K mediates metabolic disorder of branched-chain amino acid and regulates cerebral ischemia-reperfusion injury by activating ferroptosis in neurons. Cell Death Dis. 2023;14(9):634. [DOI] [PMC free article] [PubMed]

[CR39] 39.Bouchez C, Devin A. Mitochondrial biogenesis and mitochondrial reactive oxygen species (ROS): A complex relationship regulated by the cAMP/PKA signaling pathway. Cells. 2019;8(4):287. [DOI] [PMC free article] [PubMed]

[CR40] 40.Uttara B, Singh AV, Zamboni P, Mahajan RT. Oxidative stress and neurodegenerative diseases: a review of upstream and downstream antioxidant therapeutic options. Curr Neuropharmacol. 2009;7(1):65–74. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Ding J, Wang K, Liu W, She Y, Sun Q, Shi J, et al. Erratum: Pore-forming activity and structural autoinhibition of the gasdermin family. Nature. 2016;540(7631):150. [DOI] [PubMed] [Google Scholar]

[CR42] 42.Yu P, Zhang X, Liu N, Tang L, Peng C, Chen X. Pyroptosis: mechanisms and diseases. Signal Transduct Target Ther. 2021;6(1):128. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Xia S, Hollingsworth LRt, Wu H. Mechanism and regulation of Gasdermin-Mediated cell death. Cold Spring Harb Perspect Biol. 2020;12(3):a036400. [DOI] [PMC free article] [PubMed]

[CR44] 44.Liu X, Zhang Z, Ruan J, Pan Y, Magupalli VG, Wu H, et al. Inflammasome-activated gasdermin D causes pyroptosis by forming membrane pores. Nature. 2016;535(7610):153–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Man SM, Kanneganti TD. Regulation of inflammasome activation. Immunol Rev. 2015;265(1):6–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Sollberger G, Strittmatter GE, Garstkiewicz M, Sand J, Beer HD. Caspase-1: the inflammasome and beyond. Innate Immun. 2014;20(2):115–25. [DOI] [PubMed] [Google Scholar]

[CR47] 47.Bergsbaken T, Fink SL, Cookson BT. Pyroptosis: host cell death and inflammation. Nat Rev Microbiol. 2009;7(2):99–109. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Shi J, Zhao Y, Wang K, Shi X, Wang Y, Huang H, et al. Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death. Nature. 2015;526(7575):660–5. [DOI] [PubMed] [Google Scholar]

[CR49] 49.Lu A, Magupalli VG, Ruan J, Yin Q, Atianand MK, Vos MR, et al. Unified polymerization mechanism for the assembly of ASC-dependent inflammasomes. Cell. 2014;156(6):1193–206. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.He Y, Hara H, Núñez G. Mechanism and regulation of NLRP3 inflammasome activation. Trends Biochem Sci. 2016;41(12):1012–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Lin Q, Li S, Jiang N, Shao X, Zhang M, Jin H, et al. PINK1-parkin pathway of mitophagy protects against contrast-induced acute kidney injury via decreasing mitochondrial ROS and NLRP3 inflammasome activation. Redox Biol. 2019;26:101254. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Jia C, Xiang Z, Zhang P, Liu L, Zhu X, Yu R, et al. Selenium-SelK-GPX4 axis protects nucleus pulposus cells against mechanical overloading-induced ferroptosis and attenuates senescence of intervertebral disc. Cell Mol Life Sci. 2024;81(1):49. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Alim I, Caulfield JT, Chen Y, Swarup V, Geschwind DH, Ivanova E, et al. Selenium drives a transcriptional adaptive program to block ferroptosis and treat stroke. Cell. 2019;177(5):1262–e7925. [DOI] [PubMed] [Google Scholar]

[CR54] 54.He L, Huang G, Liu H, Sang C, Liu X, Chen T. Highly bioactive zeolitic imidazolate framework-8-capped nanotherapeutics for efficient reversal of reperfusion-induced injury in ischemic stroke. Sci Adv. 2020;6(12):eaay9751. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Mirończuk A, Kapica-Topczewska K, Socha K, Soroczyńska J, Jamiołkowski J, Kułakowska A et al. Selenium, Copper, Zinc Concentrations and Cu/Zn, Cu/Se Molar Ratios in the Serum of Patients with Acute Ischemic Stroke in Northeastern Poland-A New Insight into Stroke Pathophysiology. Nutrients. 2021;13(7). [DOI] [PMC free article] [PubMed]

[CR56] 56.Brown DG, Burk RF. Selenium retention in tissues and sperm of rats fed a Torula yeast diet. J Nutr. 1973;103(1):102–8. [DOI] [PubMed] [Google Scholar]

[CR57] 57.Schomburg L, Schweizer U. Hierarchical regulation of Selenoprotein expression and sex-specific effects of selenium. Biochim Biophys Acta. 2009;1790(11):1453–62. [DOI] [PubMed] [Google Scholar]

[CR58] 58.Gu X, Song Q, Zhang Q, Huang M, Zheng M, Chen J, et al. Clearance of two organic nanoparticles from the brain via the paravascular pathway. J Control Release. 2020;322:31–41. [DOI] [PubMed] [Google Scholar]

[CR59] 59.Iliff JJ, Wang M, Liao Y, Plogg BA, Peng W, Gundersen GA, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med. 2012;4(147):147ra11. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Gao J, Song Q, Gu X, Jiang G, Huang J, Tang Y et al. Intracerebral fate of organic and inorganic nanoparticles is dependent on microglial extracellular vesicle function. Nat Nanotechnol. 2023;19(3):376–386. [DOI] [PubMed]

[CR61] 61.Ouyang J, Deng B, Zou B, Li Y, Bu Q, Tian Y, et al. Oral hydrogel Microbeads-Mediated in situ synthesis of selenoproteins for regulating intestinal immunity and microbiota. J Am Chem Soc. 2023;145(22):12193–205. [DOI] [PubMed] [Google Scholar]

[CR62] 62.Zou B, Xiong Z, Yu Y, Shi S, Li X, Chen T. Rapid Selenoprotein activation by selenium nanoparticles to suppresses osteoclastogenesis and pathological bone loss. Adv Mater. 2024;36(27):e2401620. [DOI] [PubMed] [Google Scholar]

[CR63] 63.Zhao P, Yin S, Qiu Y, Sun C, Yu H. Ferroptosis and pyroptosis are connected through autophagy: a new perspective of overcoming drug resistance. Mol Cancer. 2025;24(1):23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR64] 64.Liu Y, Pan R, Ouyang Y, Gu W, Xiao T, Yang H, et al. Pyroptosis in health and disease: mechanisms, regulation and clinical perspective. Signal Transduct Target Ther. 2024;9(1):245. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Basit F, van Oppen LM, Schöckel L, Bossenbroek HM, van Emst-de Vries SE, Hermeling JC, et al. Mitochondrial complex I Inhibition triggers a mitophagy-dependent ROS increase leading to necroptosis and ferroptosis in melanoma cells. Cell Death Dis. 2017;8(3):e2716. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Selenium-doped carbon dots nanozymes hitchhiking tailored liposomes block neuronal pyroptosis through GPX4/ROS/NLRP3/GSDMD axis to attenuate ischemic stroke

Jiaxuan Hou

Li Yao

Yane Li

Enrui Xie

Jiawei Zhang

Hao Wu

Yuanyuan Zhu

Zhichao Deng

Chenxi Xu

Zongru Han

Lu Bai

Mingzhen Zhang

Shaoying Lu

Runqing Li

Hui Cai

Abstract

Supplementary Information

Introduction

Scheme 1.

Materials and methods

Synthesis of Se-CDs

Synthesis of Se-CD@LP-GSH and Se-CD@LP-PEG

Synthesis of FITC-GSH and DiO/DiL-Liposome

Synthesis of Cy5.5-Se-CDs

Cell culture

In vitro internalization and subcellular location of nanoparticles (NPs)

In vitro across the BBB

Oxygen-glucose deprivation and reperfusion model (OGD/R)

In vitro cytoprotective study of Se-CD@LP-GSH

In vitro ROS scavenging ability of Se-CD@LP-GSH

Transfection

Construction of tMCAo model

TTC staining

Nissl staining and TUNEL/NeuN staining

Targeting ability and tissue distribution of NPs in vivo

Immunofluorescent staining

Immumohistochemical staining

RNA extraction and quantitative real-time PCR analysis (qRT-PCR)

Western blotting assay

ELISA assay

RNA sequencing

Statistical analysis

Results

Preparation and characterization of Se-CDs

Fig. 1.

Preparation and characterization of Se-CD@LP-GSH

Evaluation of cellular uptake and BBB penetration of Se-CD@LP-GSH

Fig. 2.

Se-CD@LP-GSH protected neurons from oxidative damage

Fig. 3.

The neuroprotective effect of Se-CD@LP-GSH in the tMCAo model

Fig. 4.

Se-CD@LP-GSH attenuated long-term neurological deficits after tMCAo

Fig. 5.

Neuronal pyroptosis played a crucial role in the Se-CD@LP-GSH–mediated protection against cerebral I/R injury

Fig. 6.

Se-CD@LP-GSH modulated NLRP3, Caspase 1, and GPX4 expression in the tMCAo model

Fig. 7.

Se-CD@LP-GSH upregulated GPX4 and inhibited NLRP3/GSDMD-mediated pyroptosis

Fig. 8.

Discussion

Conclusion

Supplementary Information

Acknowledgements

Author contributions

Funding

Data availability

Declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Footnotes

Contributor Information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS