Deep neurotherapeutic strategy for ischemic stroke via focused ultrasound-enhanced delivery of curcumin-loaded antioxidant nanoliposomes

Abstract

Ischemic stroke is the most common type of stroke, accounting for 80% of all stroke cases, and is increasingly affecting younger individuals due to changes in lifestyle, diet, and environment. Conventional treatments such as thrombolytic agents and mechanical thrombectomy can rapidly restore blood flow but are limited by narrow therapeutic time windows and complications like ischemia–reperfusion injury (IRI) and cerebral hemorrhage. These limitations highlight the need for strategies that provide both vascular and neuroprotective benefits. This study introduces a targeted nanotherapy approach using focused ultrasound (FUS) and Sonazoid™ microbubbles to temporarily open the blood–brain barrier (BBB) via cavitation, enabling drug delivery to hypoxic brain regions. Curcumin (CUR), a natural antioxidant with neuroprotective effects, was encapsulated in liposomes (CUR@LP) and functionalized with a stroke homing peptide (SHp) to form CUR@LP-SHp nanoparticles. After systemic administration and FUS-induced BBB modulation, CUR@LP-SHp accumulated in ischemic tissue. In a photothrombotic stroke model, this treatment significantly reduced infarct volume by 92% and decreased expression of glial fibrillary acidic protein (GFAP) by 80%. Histological and protein analyses confirmed reduced oxidative stress and enhanced preservation of vascular and neuronal function. This novel delivery platform improves CUR's bioavailability and brain targeting, offering a promising, safe, and effective therapeutic strategy for ischemic stroke.

---

1. Introduction

Ischemic stroke is one of the leading causes of death and long-term disability worldwide, with millions of individuals affected annually.¹ It results from an obstruction of cerebral blood flow, causing a cascade of pathophysiological events such as energy failure, excitotoxicity, oxidative stress, inflammation, and eventual neuronal cell death.² Despite decades of research, current treatment options remain limited and largely restricted to the acute phase.³ Thrombolytic therapy (e.g., recombinant tissue plasminogen activator) and mechanical thrombectomy are effective but constrained by narrow therapeutic windows and strict patient eligibility criteria.⁴ Consequently, there is a critical unmet need for therapies that can mitigate secondary injury mechanisms and extend the treatment window, while being minimally invasive and broadly applicable.⁵

One promising direction is to combine targeted drug delivery systems with non-invasive stimulation techniques to create synergistic therapeutic effects. In this context, near-infrared (NIR) irradiation has gained increasing attention for its ability to modulate biological processes by photobiomodulation (PBM). NIR light can penetrate deeply into biological tissues and has been shown to enhance mitochondrial function, promote angiogenesis, reduce oxidative stress, and suppress the release of inflammatory cytokines—processes that are highly relevant in the treatment of cerebral ischemia.⁶ Meanwhile, nanotechnology, particularly liposomal drug delivery systems, offers new opportunities to overcome the blood–brain barrier (BBB), improve pharmacokinetics, and enhance the bioavailability of therapeutic agents.⁷ Curcumin (CUR), a polyphenolic compound derived from Curcuma longa, exhibits strong antioxidant, anti-inflammatory, and neuroprotective properties, making it an attractive candidate for stroke therapy.^8–10 However, its clinical application is severely hindered by poor water solubility, rapid degradation in physiological environments, and low systemic bioavailability.¹¹ To address these limitations, this study synthesizes a nanoliposomal curcumin formulation (CUR@LP) using thin-film hydration and ultrasonic homogenization. The surface of the liposomes is further modified with DSPE-PEG2000-SHp, a stroke-homing peptide (SHp), to create CUR@LP-SHp—a targeted nanocarrier capable of actively homing to ischemic regions in the brain.^12–14

The novelty of this research lies in the integration of two advanced technologies: (1) targeted nanoliposomal drug delivery and (2) focused ultrasound (FUS)-induced blood–brain barrier modulation. This integrated platform enables efficient and localized delivery of CUR to ischemic brain regions.^15–17 This faceted approach enables precise, localized delivery and activation of CUR, maximizing therapeutic impact while minimizing off-target effects.¹⁸ This study presents a therapeutic strategy integrating targeted nanoliposomal delivery with focused ultrasound–mediated blood–brain barrier opening for ischemic stroke therapy.¹⁹ Comprehensive physicochemical analysis of CUR@LP-SHp reveals several advantages over free CUR and non-targeted formulations. Transmission electron microscopy and dynamic light scattering confirm that CUR is efficiently encapsulated within the bilayer membrane of the liposomes, addressing its solubility issues and shielding it from degradation in acidic environments.²⁰ The incorporation of DSPE-PEG2000-SHp not only imparts targeting capability but also contributes to colloidal stability, aided by the nanoliposomes’ surface negative charge that provides electrostatic repulsion to prevent aggregation.^21–23 These characteristics collectively enhance in vivo retention time and improve the pharmacokinetic profile of CUR.²⁴

Functionally, CUR within CUR@LP-SHp maintains its robust antioxidant activity, demonstrated by its ability to scavenge reactive oxygen species (ROS) and regulate oxidative stress at both the cellular and tissue levels.²⁵ These properties are critical in the post-ischemic environment, where oxidative damage exacerbates neuronal injury and delays recovery.²⁶ To test the therapeutic efficacy of this system, a photothrombotic ischemic stroke model was established in vivo using Rose Bengal dye and targeted laser exposure to induce focal cortical infarction.^27–29 To facilitate the entry of CUR@LP-SHp into the brain, FUS stimulation was applied in conjunction with intravenous microbubble injection. This strategy temporarily and safely opened the BBB, allowing the nanocomposite drug to cross into ischemic brain regions. CUR@LP-SHp was then administered via tail vein injection and allowed to circulate systemically.

The therapeutic strategy proposed in this study demonstrates multiple potential advantages. NIR irradiation acted as a non-invasive therapeutic modality, modulating local tissue responses through photobiomodulation while offering deep tissue penetration, with histological analyses showing well-preserved tissue architecture in surrounding regions. In addition, unlike systemic antioxidant administration, CUR@LP-SHp targets the ischemic brain region directly, reducing the required dosage and minimizing side effects. Immunofluorescence staining and histological analysis were performed to assess therapeutic outcomes, including neuronal survival, reduction in oxidative stress markers, and attenuation of neuroinflammation. Moreover, the use of FUS to open the BBB represents a cutting-edge, non-invasive strategy to overcome one of the greatest obstacles in central nervous system drug delivery.³⁰

2. Materials and methods

2.1 Chemicals and materials

The names, chemical formulas, and purity information of the drugs used in this study were obtained from their respective suppliers. All chemical reagents were of analytical grade with high purity and did not require further purification, as detailed below. For the formulation of the stroke-homing peptide-modified curcumin liposomes (CUR@LP-SHp), the following chemical compounds were used: DSPE-PEG2000 (1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000], C₁₃₃H₂₆₇N₂O₅₅P, purity >99%) was purchased from Avanti. DPPA (1,2-dihexadecanoyl-sn-glycero-3-phosphate, sodium salt, C₃₅H₆₈O₈PNa, purity >99%) was also obtained from Avanti. DPPC (1,2-dipalmitoyl-sn-glycero-3-phosphocholine, C₄₀H₈₀NO₈P, purity >99%) was supplied by MACKLIN. The stroke-homing peptide-modified DSPE-PEG2000-SHP (CLEVSRKNC, with the peptide sequence Asn–Gly–Arg, purity >99%) was purchased from Qiyuebio. Curcumin (C₂₁H₂₀O₆, purity >97%) was sourced from TCI. Pluronic F-127, a block copolymer of poly(propylene glycol) and poly(ethylene glycol), with general formula (C₅H₁₀O₂)_n and purity >99%, was obtained from Sigma-Aldrich. Glycerol (C₃H₈O₃, 99.5% purity) was provided by J. T. Baker, and ethanol (C₂H₅OH, 99% purity) was purchased from Fisher Chemical. For cell-based experiments, Dulbecco's Modified Eagle Medium (DMEM) was obtained from Thermo Fisher. Cobalt chloride (CoCl₂, purity >99%) was purchased from Sigma-Aldrich. DPPH (1,1-diphenyl-2-picrylhydrazyl, a free radical compound, C₁₈H₁₂N₅O₆, purity >95%) was obtained from ALFA. For animal experiments, Evans Blue dye (C₃₄H₂₄N₆Na₄O₁₄S₄, purity >99%) was purchased from Thermo Fisher. TTC (2,3,5-triphenyl-tetrazolium chloride solution, C₁₉H₁₅ClN₄) and Rose Bengal (4,5,6,7-tetrachloro-2′,4′,5′,7′-tetraiodofluorescein, C₂₀H₂Cl₄I₄Na₂O₅) were both sourced from Sigma-Aldrich. RIPA buffer, bovine serum albumin (BSA), and the Coomassie protein assay kit were all supplied by Thermo Fisher.

2.2 Free radical scavenging assay

Current methods for detecting antioxidant activity are based on spectrophotometric techniques that rely on hydrogen atom transfer (HAT) or single electron transfer (SET) mechanisms. DPPH is a commonly used lipophilic free radical reagent widely applied for evaluating antioxidant activity. The deep purple DPPH radical exhibits a maximum absorbance at 517 nm. When reduced by hydrogen atom donation from antioxidants, it becomes light yellow. The change in DPPH absorbance is used to assess the antioxidant activity of analytes. The chemical reactions are shown below:

DPPH˙ + ArOH → DPPH-H + ArO˙ (HAT mechanism) (1)

DPPH˙ + ArOH → DPPH + [ArOH]˙ (SET mechanism) (2)

(ArOH: phenolic antioxidant)

In this study, the antioxidant activities of three compounds—Resveratrol, Edaravone, and CUR—were evaluated using the DPPH assay. DPPH and the antioxidants were dissolved in ethanol and mixed at a concentration ratio of 100 μM to 30 μM. The mixture was stirred in the dark for 1 hour. UV-Vis spectroscopy was used to measure the maximum absorbance of DPPH at 517 nm. The remaining ROS and radical scavenging activity (RSA) were calculated using the following equations:

ROS% = (Abs\_sample/Abs\_control) × 100% (3)

RSA% = [(Abs\_control − Abs\_sample)/Abs\_control] × 100%. (4)

2.3 Synthesis of CUR-loaded liposomes and stability test

4.5 mg DPPC, 0.5 mg DSPE-PEG2000, 0.5 mg DSPE-PEG-SHp, and 0.25 mg DPPA were dissolved in 10 mL of ethanol. Then, 1.2 mL of 2 mM CUR (in ethanol) was added to the lipid solution to prepare 600 μM CUR@LP-SHp. After thorough mixing, the solution was transferred to a culture dish and left to evaporate the solvent overnight, forming a thin yellow lipid film. Then, 0.4 mL glycerol, 8 mg pluronic F-127, and 3.6 mL deionized water containing surfactant were added. The mixture was shaken at room temperature for 2 hours and homogenized using a sonicator (10 s pulse, 5 s pause) for 3 minutes. Finally, the solution was filtered through a 0.22 μm PTFE membrane to obtain bioavailable, stroke-targeted CUR liposomes (CUR@LP-SHp), as shown in Fig. S3. The synthesized nanoliposomes in this study are intended for use in systemic circulation. Therefore, their size stability was evaluated using dynamic light scattering (DLS) over 6 hours at 37 °C in deionized water to simulate physiological conditions. To assess the stability of CUR liposomes in phosphate-buffered saline at 37 °C, UV-Vis spectrophotometry was used to measure absorbance changes of CUR and CUR liposomes (CUR@LP) over 3 hours. The absorbance at 420 nm at each time point was compared to the initial value to calculate relative stability. To evaluate the stability of CUR@LP-SHp, the particle size distribution and zeta potential were measured in PBS and 50% fetal bovine serum (FBS) at 37 °C for 24 hours.

2.4 Encapsulation efficiency of CUR liposomes

UV-Vis spectrophotometry was used to measure the peak absorbance of CUR while subtracting background absorbance from liposomes to determine encapsulation efficiency (EE%). Based on the therapeutic effect result, 60 μM was chosen as the optimal concentration. Since CUR exhibits peak absorbance at 420 nm, the EE% was calculated using:

LC (%) = (EE% × expected CUR concentration) × 100%. (6)

2.5 Cell culture

Neuro-2a mouse neuroblastoma cells were used. They were cultured in high-glucose DMEM with 10% horse serum, 5% FBS, 2 mM L-glutamine, 1.5 mg mL⁻¹ sodium bicarbonate, and 1% penicillin/streptomycin in a humidified incubator at 37 °C with 5% CO₂. Subculturing was performed twice weekly.

2.6 Cell biocompatibility assay

Neuro-2a cells were seeded in 96-well plates at 1 × 10⁴ cells per mL and incubated for 24 hours. LP was diluted to different concentrations (3, 9, 27, 81, and 250 μg mL⁻¹) in a medium and co-cultured with cells for 24 hours. After removing the drug with PBS, 10% Alamar Blue was added and incubated for 1 hour to assess viability.

2.7 Hypoxia-induced cell model

CoCl₂ was used to induce hypoxia in Neuro-2a cells over a 5 day protocol. On Day 1, cells were seeded. On Day 2, 300 μM CoCl₂ was added. On Day 3, the CoCl₂-containing medium was replaced, and cells were treated with various concentrations (7.5–120 μM) of CUR@LP-SHp or Cur for 48 hours. On Day 5, viability was assessed with 10% Alamar Blue.

2.8 In vitro drug release study

The in vitro release of CUR was evaluated using a dialysis method. Briefly, 1 mL of the sample was loaded into a dialysis bag (molecular weight cutoff of 3500 Da) and sealed at both ends. The dialysis bag was then immersed in 50 mL of release medium consisting of PBS (pH 6.5 or 7.4) containing 0.1% (w/v) Tween-80 and 10% ethanol, and incubated in a horizontal shaker at 70 rpm and 37 ± 1 °C. At predetermined time points (0.5, 1, 2, 4, 8, 24, 48, and 72 h), 1 mL of the release medium was withdrawn and immediately replaced with an equal volume of fresh medium. The collected samples were stored at −80 °C until analysis. Absorbance was measured at 420 nm using 200 μL of the sample.

2.9 In vitro blood–brain barrier (BBB) penetration

An in vitro BBB-mimetic transwell system was established by seeding BMECs in the upper chamber with a 0.4 μm membrane and Neuro-2a cells in the lower chamber. After BMECs reached confluence, the two chambers were co-cultured for 24 h. DiD-labeled liposomal samples were then added to the upper chamber and incubated for 4 h. After incubation, Neuro-2a cells were washed, fixed, and counterstained with DAPI. DiD fluorescence was observed by confocal microscopy and quantified using ImageJ.

2.10 Reactive oxygen species (ROS) detection in hypoxic neuro-2a

Neuro-2a cells were seeded in 6-well plates and cultured for 24 h. Except for the normal group, cells were treated with 300 μM CoCl₂ for 24 h to induce hypoxia, followed by 48 h of treatment with LP, CUR@LP, or CUR@LP-SHp. After treatment, cells were washed and incubated with 20 μM DCFH-DA in the dark at 37 °C for 30 min. The cells were then washed, counterstained with DAPI, and mounted. Fluorescence images were acquired using a confocal microscope, and intracellular ROS levels were analyzed using ImageJ.

2.11 Immunohistochemistry for colocalization analysis

The hypoxic Neuro-2a cells were treated with DiD-labeled CUR@LP or CUR@LP-SHp, washed with PBS, and fixed with 4% paraformaldehyde. The cells were permeabilized with 0.1% Triton X-100, blocked with 1% BSA, and incubated with anti-fibrinogen primary antibody (1 : 200) overnight at 4 °C. After washing, the cells were incubated with Alexa Fluor 488-conjugated secondary antibody (1 : 1000) and DAPI. Fluorescence images were acquired by confocal microscopy, and colocalization was analyzed using ImageJ.

2.12 Focused ultrasound (FUS) to open the blood–brain barrier (BBB)

All animal procedures were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals of National Taiwan University Hospital and approved by the Institutional Animal Care and Use Committee (IACUC) of National Taiwan University Hospital (approval no. 20220360). C57BL/6 male mice were anesthetized, their scalps shaved, and given intraperitoneal anesthesia (5 μL per g body weight). The BBB was targeted at M1 (1.4 mm anterior, 1.1 mm left of bregma). The FUS parameters were selected according to previously reported BBB opening protocols with validated safety and efficacy.³¹ In brief, the mice were injected intravenously with 100 μL Sonazoid™ microbubbles (10⁷ MB mL⁻¹), followed by FUS (500 kHz, 30 mVpp, 5000 cycles, 1 s bursts for 1 minute). Then, 30 μL of 2% Evans Blue was injected. After 3 hours of circulation, mice were perfused and brains collected for sectioning (2 mm) and fluorescence microscopy to assess BBB opening.

2.13 Brain drug accumulation

Mice were grouped into FUS + CUR@LP, FUS + Cur, and saline controls. Following Sonazoid™ microbubble and FUS treatment, Cur or CUR@LP (25 mg kg⁻¹) was injected. After 3 hours, brains were harvested, homogenized with RIPA buffer, centrifuged, and fluorescence intensity was measured at 509 nm (excitation 430 nm). Autofluorescence from saline controls was subtracted. CUR concentration was calculated from standard curves, normalized by total protein (BSA standard), and expressed as:

2.14 Photothrombotic stroke model and therapeutic evaluation. Mice were anesthetized and injected with 10 μL g⁻¹ of 10 mg mL⁻¹ rose bengal (photosensitizer) and 10 μL g⁻¹ of carprofen (analgesic). After 15 minutes, the right scalp was cut, and M1 (1.4 mm anterior, 1.1 mm right) was irradiated for 15 minutes with a 561 nm laser at 25 mW, 5 cm distance. The scalp was sealed with tissue glue, and mice were left overnight. After inducing a stroke, mice were intravenously injected with 100 μL Sonazoid™ microbubbles and CUR@LP-SHp (25 mg kg⁻¹). FUS was applied for 1 minute on the contralateral M1 to transiently open the BBB and deliver the drug. Mice were left for 48 hours for recovery.

2.15 TTC staining of brain tissue

TTC staining was used to detect infarct regions. TTC is reduced by NADH in live mitochondria to red formazan (TTF), while ischemic regions remain white/pale due to reduced enzyme activity. A 2% TTC solution was prepared in saline, shielded from light, and pre-warmed to 37 °C. After sacrificing mice, brains were sectioned into 2 mm slices and incubated with TTC for 10 minutes at 50 rpm and 37 °C, then observed.

2.16 Western blotting

Brain tissues were lysed with RIPA buffer, centrifuged at 13 000 rpm, and the protein concentration was determined. Equal amounts (30 μg) were separated by SDS-PAGE, transferred to PVDF membranes, and blocked with 5% skim milk. Membranes were incubated with GFAP primary antibody (1 : 1000) and β-actin overnight at 4 °C, followed by HRP-conjugated secondary antibody (1 : 10 000). Protein bands were visualized with ECL Plus. Quantification was performed using ImageJ.

2.17 Tissue staining

For Hematoxylin and Eosin (H&E) staining, tissues were fixed in 4% paraformaldehyde overnight at 4 °C, then immersed in 30% sucrose PBS solution overnight. Sections were cut coronally and placed on gelatin-coated slides. After washing with graded ethanol and xylene, slides were stained with hematoxylin for 5 minutes, followed by 1% acid alcohol and Scott's blue. Eosin counterstaining was done for 2 minutes. Dehydrated slides were then observed under a light microscope. For Immuno-histochemistry (IHC) staining, brain slices (from +2 mm to −2 mm of bregma) were deparaffinized with 3% hydrogen peroxide in methanol, rehydrated, and treated with 0.01 M citrate buffer. Sections were blocked with goat serum and incubated overnight at 4 °C with primary antibodies: GFAP (1 : 2000) and NeuN (1 : 100). Secondary goat anti-rabbit IgG was used, and stained slices were mounted and analyzed by a light microscope.

3. Results and discussion

3.1 Characterization of CUR@LP

3.1.1. Pilot assay of antioxidant. Ischemic stroke occurs when a blood clot or other obstruction blocks blood flow to the brain, depriving brain tissue of oxygen and nutrients. This can lead to rapid cell death and long-term neurological damage if not treated promptly (Fig. S1). One of the major consequences of ischemic stroke is the overproduction of ROS, leading to oxidative stress, which exacerbates brain injury.³² Thus, antioxidants have shown promise in preclinical studies by reducing infarct size, improving neurological outcomes, and limiting oxidative damage after stroke. In this study, three kinds of antioxidants were investigated in the pilot assay.³³ Firstly, Edaravone (EDV) is a clinical drug with phenol-like properties that has been approved for the treatment of acute ischemic stroke.³⁴ Its unique pyrazolone structure undergoes keto–enol tautomerism, enabling potent free radical scavenging activity through both hydrogen atom transfer and electron transfer mechanisms. Secondly, Curcumin (CUR), a natural polyphenol derived from the rhizome of Curcuma longa, features a distinctive yellow pigment and contains a β-diketone moiety with two methoxy-phenol rings.³⁵

Resveratrol (RES), another natural polyphenol predominantly found in grapes, red wine, and various berries, is characterized by its trans-stilbene structure with three hydroxyl groups (Fig. 1b).³⁶ The DPPH assay is a widely used method to evaluate the antioxidant activity of compounds, particularly through their ability to scavenge free radicals (Fig. S2).³⁷ The absorption spectra change after a 1 hour interaction between DPPH and these antioxidant drugs are shown in Fig. 1c. A significant decrease in the DPPH absorption peak at 517 nm indicated a substantial reduction of free radicals. Using eqn (3) and (4), ROS% and RSA% were calculated, and CUR demonstrated superior antioxidant activity with ROS% of 36.6% (Fig. 1d) and RSA% of 63.4% (Fig. 1e). CUR's superior antioxidant activity is attributed to its structure, which features a heptadienone bridge with highly active carbon atoms that enable hydrogen proton transfer, while two methoxyphenol rings facilitate single electron transfer, providing a total of three antioxidant active sites. In comparison, EDV contains a pyrazole structure that undergoes keto–enol tautomerism to form a phenol-like structure, while RES features a single phenol structure. Both compounds have only one antioxidant active site, and their effectiveness is limited by DPPH's triphenyl ring steric hindrance. Based on these findings, CUR was selected as the neuroprotective agent for ischemic stroke treatment due to its superior antioxidant properties. This research utilized the thin-film hydration method and ultrasonic homogenization to synthesize nanoliposomal carriers, which were self-assembled to encapsulate the antioxidant drug CUR, producing CUR-loadedliposomes (CUR@LP). Following the concept in Fig. 1a, the first part explores the morphological structure identification, size, surface properties, and stability analysis of nanoliposomes, free radical scavenging activity analysis of antioxidant drugs, and drug encapsulation efficiency of curcumin-loaded nanoliposomes. To evaluate the changes in antioxidant activity of CUR encapsulated in nanoliposomes, the DPPH radical scavenging assay was used to measure the antioxidant activity of free CUR and nanoliposomal CUR, as shown in Fig. 1f and g. The ROS and RSA values for free CUR were 26.6% and 73.4%, respectively, while those for nanoliposomal CUR were 30.3% and 69.7%, respectively. These results indicate that CUR encapsulated in liposomes retains antioxidant activity nearly equivalent to that of free CUR. This is consistent with the theory that the high specific surface area of nanoliposomes enhances drug loading capacity, and the stability of the carrier also prevents CUR from degrading easily in solution. Therefore, CUR@LP improves the stability of CUR while preserving its antioxidant activity comparable to that of the free form.

Fig. 1 (a) A schematic illustration of the design concept for the synthesis of nanoliposome materials. (b) The molecular structures of EDV, RES, and CUR. (c) The absorption spectra of the reagent and (d) retain the ROS ratio of DPPH, CUR, EDV, and RES (n = 3). (e) RSA relative to DPPH, CUR, EDV, and RES (n = 3). (f) The retaining ROS ratio of DPPH, CUR, and CUR@LP (n = 3). (g) RSA values representing the radical scavenging activity relative to DPPH (n = 3). (The significance is marked as *p < 0.05, **p < 0.01, and ***p < 0.001 compared to the control group.)

3.1.2. Structural and morphological analysis of liposomes. Liposomes are nanoscale lipid-based vesicles typically ranging from 50 to 200 nm in size. They are valued for their biocompatibility, ability to enhance drug stability, and potential for targeted delivery, particularly in treatments for diseases such as cancer and neurological disorders, including ischemic stroke, making them versatile drug delivery systems (Fig. S3). The morphology of liposomes was analyzed using TEM, as shown in Fig. 2a and b. Since the liposomal structure lacks conductive metal components on its surface, a 2% phosphotungstic acid stain was used for negative staining to enhance electron density. This allowed for the monitoring of the liposomes under TEM at an acceleration voltage of 200 kV and a magnification of 30–80 kX. The images showed that the liposomes (LP) appeared as dark, uniformly dispersed spherical particles. Image J software analysis revealed an average particle diameter of 139 ± 35 nm, which meets the size requirements for biological nanocarriers. To understand the characteristics of LP and CUR@LP in solution, DLS was used to measure particle size and zeta potential. When LP was dispersed in PBS, the average size was measured at 130 ± 56 nm, while CUR@LP showed an average size of 130 ± 57 nm, as shown in Fig. 2c. This indicates that CUR encapsulation via self-assembly did not significantly affect particle size, as confirmed by TEM, and remained below 200 nm, thereby maintaining good bioavailability. The magnitude of the zeta potential depends on the net charge accumulated on the liposome surface. When LP was dispersed in a low ionic strength solution, the DPPC zwitterionic molecules’ phosphate groups (PO₄³⁻) were preferentially arranged on the outer membrane layer, while choline groups were arranged on the inner layer. This resulted in LP having a negative zeta potential of −20.5 ± 8.6 mV, as shown in Fig. 2d. After CUR encapsulation, CUR@LP showed a zeta potential of −43.4 ± 6.6 mV. This decrease in potential was attributed to the hydrophobic distribution of CUR within the liposomal bilayer. The absolute zeta potential values greater than 20 mV indicate good electrostatic repulsion between particles, preventing aggregation and maintaining good stability in solution. With stroke-homing peptide (SHp) modification, the zeta potential of CUR@LP-SHp showed a slight increase to −37.1 ± 6.6 mV compared to that of CUR@LP, as shown in Fig. S5a. Moreover, the average liposomal particle size increased to 146.2 ± 86.3 nm, which can be attributed to the surface-extended peptide moiety. These physicochemical changes indicate the successful surface modification of the liposomes with SHp, as shown in Fig. S5b.

Fig. 2 The TEM image of (a) LP and (b) CUR@LP. (c) DLS distribution and (d) zeta potential of LP and CUR@LP (n = 3). The absorbance of (e) CUR and (f) CUR@LP was monitored for 3 hours.

3.1.3. Stability assay. Stability analysis was performed based on hourly absorption measurements from Fig. S4b. CUR@LP maintained approximately 65% stability after 3 hours in PBS buffer, while free CUR's stability rapidly declined to about 20% within the same period. These results demonstrate that liposomal encapsulation effectively reduces CUR's interaction with the aqueous phase, thereby significantly enhancing its stability. Since this research delivers nanoliposomal drugs via systemic circulation in mice, stability was evaluated by simulating circulation at body temperature (37 °C) for 6 hours. DLS measurements showed that CUR@LP particle size remained stable at around 140 nm over 6 hours, with no significant changes, as shown in Fig. S4a. For the modified liposomes, the size distribution and zeta potential were measured in 50% FBS at physiological temperature for 24 hours to evaluate their stability under simulated physiological conditions. As shown in Fig. S5c, the hydrodynamic size of CUR@LP-SHp did not exhibit significant changes in FBS over 24 hours. In contrast, the zeta potential increased to −21.7 ± 1.6 mV after incubation in FBS, which may be attributed to the adsorption of serum proteins onto the liposomal surface; a slight increase was also observed within 24 hours, as shown in Fig. S5d. Despite the high colloidal stability of CUR@LP and CUR@LP-SHp under physiological and serum-containing conditions, an ideal drug delivery system should also enable efficient drug release at target sites. Therefore, the in vitro drug release behavior was further investigated under simulated physiological (pH 7.4) and ischemic (pH 6.5) conditions. As shown in Fig. S6, CUR exhibited a faster release profile at acidic pH than at neutral pH, indicating that acidic environments facilitate its release. Notably, approximately 46% of CUR was released at pH 6.5 after 72 hours, whereas only about 28% was released at pH 7.4. This accelerated release under ischemia-mimicking conditions suggests that acidic, pathological microenvironments may enhance curcumin availability at target sites, thereby enhancing therapeutic efficacy. Collectively, these results demonstrate favorable stability during circulation at 37 °C, without particle aggregation, and responsive release properties, supporting the potential suitability of CUR@LP and CUR@LP-SHp for in vivo drug delivery in brain disease treatment. The UV-visible spectroscopic analysis of CUR and CUR@LP was conducted in PBS at 37 °C over a 3 hour interaction period, as shown in Fig. 2e and f. The spectra reveal that CUR exhibits a characteristic absorption peak at 430 nm and a subtle shoulder peak at 365 nm. These peaks correspond to the conjugated ferulic acid structure and the single ferulic acid unit, respectively. The shoulder peak at 365 nm appears due to partial deprotonation of CUR when interacting with water in the PBS medium. In contrast, CUR@LP shows distinct spectral characteristics. The shoulder peak at 365 nm disappears due to CUR's encapsulation within the phospholipid bilayer of the liposomes, which reduces its interaction with the aqueous phase. This encapsulation maintains CUR's highly conjugated structure. The spectrum shows both blue and red shifts due to π–π* and n–π* transitions, resulting in an absorption peak at 425 nm and a shoulder peak at 450 nm.

3.1.4. Spectral analysis. CUR exhibits distinct absorption and fluorescence properties. The UV-visible absorption spectra were measured for CUR concentrations ranging from 5 to 80 μM, as shown in Fig. 3a. A prominent absorption peak was observed at 420 nm, with absorption values increasing in proportion to the concentration. The absorption-concentration calibration curve, plotted using the 420 nm absorption values (Fig. 3b), demonstrates a linear correlation between absorption and concentration. Fluorescence spectroscopy analysis of CUR in ethanol solution revealed a significant emission peak at 550 nm, as illustrated in Fig. 3c. However, the fluorescence intensity-concentration relationship Fig. 3d showed non-linear behavior, with self-quenching occurring at concentrations above 20 μM. Due to this non-linear fluorescence behavior, the drug encapsulation efficiency of CUR@LP was determined using the linear absorption-concentration relationship. When encapsulated in nanoliposomes, CUR displayed two absorption peaks: one at 420 nm and another at 450 nm. The redshift to 450 nm is attributed to n–π* transitions of CUR within the phospholipid bilayer, although this peak shows relatively lower intensity (Fig. 3e). Therefore, the 420 nm absorption value was used to calculate the drug encapsulation efficiency, which was determined to be 72%. Therefore, by substituting the absorbance value at 420 nm into eqn (5), the drug encapsulation efficiency was calculated to be 72%. By further substituting this encapsulation efficiency (EE%) into eqn (6), the drug loading concentration of 60 μM CUR@LP was determined to be 43 μM, as shown in Fig. 3f.

Fig. 3 The spectral characterization of CUR. (a) The absorbance spectra, (b) the calibration curve of absorbance, and (c) the fluorescence spectra of CUR. (d) The relationship between concentration and fluorescence. (e) The absorbance of LP, CUR@LP, and CUR. (f) Drug encapsulation efficiency and drug-loading concentration of 60 μM CUR@LP.

3.2 In vitro analysis

3.2.1. Cell viability. To determine the optimal concentration of nanocarrier for cells, normal Neuro-2a cells were cultured with various concentrations (3, 9, 27, 81, and 250 mg mL⁻¹) of LP for 24 hours (Fig. S7a). The results indicated that LP concentrations above 81 mg mL⁻¹ adversely affected cell growth, with 250 mg mL⁻¹ exhibiting significant cytotoxicity, resulting in a reduction of to 78%, as shown in Fig. S7b. Based on these findings, CUR@LP was synthesized using an 81 mg mL⁻¹ LP concentration for subsequent therapeutic analysis of CoCl₂-induced hypoxic cells.³⁸ To investigate the antioxidant activity of CUR and CUR@LP in treating hypoxic cells, Neuro-2a cells were exposed to 300 μM CoCl₂ for 24 hours to activate the hypoxia-inducible factor HIF-1α, simulating ischemic stroke conditions.³⁹ The cells were then treated with various concentrations of CUR, CUR@LP, and CUR@LP-SHp for two days. The cell viability results in Fig. S7c and S8 revealed that CoCl₂-induced hypoxic cells showed approximately 30% lower viability than normal cells. After two days of treatment, CUR@LP and CUR@LP-SHp demonstrated superior therapeutic efficacy compared to free Cur, with cell viability improving from 69% to 82% with 60 μM CUR@LP and CUR@LP-SHp treatment. However, the free CUR and CUR-loaded modified liposome at 120 μM show significant cytotoxicity. These results indicate that liposomal encapsulation enhances CUR's bioavailability. The improved therapeutic effect of CUR@LP and CUR@LP-SHp can be attributed to enhanced cellular uptake through endocytosis, CUR's antioxidant activity in eliminating intracellular ROS, and regulation of apoptotic factors and oxidative stress. This demonstrates CUR@LP and CUR@LP-SHp's effectiveness in restoring the viability of hypoxic cells. On the other hand, by exposure to 770 nm NIR, the cell viability was elevated to 83%. The improved therapeutic effect can be attributed to increased cell viability.

3.2.2. Targeted antioxidant delivery across the BBB. To explore the ability of the liposomal samples to cross the BBB, an in vitro BBB-mimetic transwell system was established. As shown in Fig. S9a, brain microvascular endothelial cells (BMECs) were seeded in the upper chamber and allowed to reach confluence to mimic the BBB. The upper chamber was then placed onto the lower chamber containing Neuro-2a cells, and the two chambers were co-cultured for 24 hours to facilitate barrier maturation. Subsequently, DiD-labeled liposomal samples were added to the upper chamber and incubated for 4 hours to allow transcellular transport. The liposomal samples’ penetration ability was then evaluated by detecting DiD fluorescence using confocal microscopy (Fig. S9b). Quantitative analysis revealed that the LP, CUR@LP, and CUR@LP-SHp groups exhibited significantly higher DiD fluorescence intensity than the control group, which showed only weak fluorescence (Fig. S9c), indicating effective transport across the BBB-mimetic BMEC layer and delivery of CUR into brain-like cellular environments. To further investigate the interaction between SHp-containing liposomes and hypoxic Neuro-2a cells, colocalization analysis was performed. As shown in Fig. S11, the CUR@LP exhibited limited overlap between DiD and fibrinogen signals, resulting in a low Pearson's correlation coefficient (r = 0.37). In contrast, CUR@LP-SHp showed markedly enhanced colocalization, as evidenced by the increased overlap in the merged images and a higher correlation coefficient (r = 0.60), suggesting that SHp modification promotes association with fibrinogen-enriched intracellular regions. Following the confirmation of BBB penetration and fibrinogen-associated targeting, the intracellular antioxidant effects of the liposomal formulations were further evaluated in CoCl₂-induced hypoxic Neuro-2a cells using DCF staining. As shown in Fig. S10a, CoCl₂-treated cells exhibited strong green fluorescence, indicating excessive intracellular ROS accumulation compared to the normal groups. In contrast, cells treated with CUR@LP and CUR@LP-SHp showed markedly reduced fluorescence signals. Quantitative analysis further demonstrated a significant decrease in ROS levels in the CUR@LP and CUR@LP-SHp groups compared with the non-treated hypoxic group (Fig. S11b), whereas only a moderate reduction was observed in the LP group. Taken together, these results suggest that SHp-modified liposomes facilitate efficient delivery into hypoxic neural cells within fibrinogen-enriched microenvironments. After cellular uptake, CUR may exert its neuroprotective effects by regulating oxidative stress, which is consistent with the observed decrease in intracellular ROS levels. In particular, CUR has been reported to modulate antioxidant-related signaling pathways, such as the Nrf2/ARE pathway, thereby promoting the expression of downstream antioxidant genes, such as HO-1, NQO1, and Gpx4, which contribute to cellular redox homeostasis.⁴⁰ This sequential enhancement in transport, targeting, cellular uptake, and functional activity highlights the integrated therapeutic potential of CUR@LP-SHp for ischemia-related brain disorders.

3.3 In vivo analysis

3.3.1. FUS-induced BBB opening. In this study, focused ultrasound was used to stimulate microbubbles in the brain, creating cavitation effects that temporarily open the BBB.⁴¹ This approach enables the delivery of CUR-loaded nanoliposomes through tail vein injection via systemic circulation (Fig. 4a).⁴² To verify successful BBB opening, Evans blue dye was administered through tail vein injection and allowed to circulate for 3 hours (Fig. 4b).

Fig. 4 Evans blue permeability analysis of BBB opening by FUS. The (a) scheme, (b) FUS setting, and (c) flow chart of the BBB opening test. (d) A microscope of Evans blue-stained tissue of the harvested brain. (e) A macroscope of Evans blue-stained tissue of brain sections. (f) Fluorescence imaging of Evans blue-stained tissue of brain sections.

Subsequently, the brain tissue was harvested and preserved in a 4 °C formalin solution for one day. Fluorescence microscopy and photography were used to document Evans blue leakage, as shown in Fig. 4c. As illustrated in Fig. 4d–f, the white arrows in the standard image indicate the FUS stimulation site in the left M1 region of the mouse brain (1.4 mm anterior to bregma, 1.1 mm left of the midline). Brain sections revealed visible blue staining from Evans blue leakage. Under fluorescence macroscope with a GFP filter, Evans blue was excited using a 500–600 nm laser source, exhibiting fluorescence emission at 680 nm. The fluorescence imaging showed a clear Evans blue signal near the left M1 region of the cerebrum, confirming the successful establishment of the focused ultrasound-mediated BBB opening model.

3.3.2. In vivo drug accumulation assay. Drug accumulation in mouse brain tissue was analyzed using two calibration curves by eqn (7): one for CUR fluorescence intensity versus concentration and another for BSA absorption versus concentration, as shown in Fig. S12a and S12b. To account for individual variations in drug absorption among animals, CUR content was normalized relative to total brain tissue protein content. This animal brain drug accumulation experiment was conducted in collaboration with Dr Yang Shu-Mei and research assistants from the Department of Rehabilitation, National Taiwan University Hospital. The study compared three drug delivery approaches: systemic delivery of free CUR, FUS-mediated delivery of free CUR (FUS + CUR), and FUS-mediated delivery of CUR-loaded liposomes (FUS + CUR@LP). The experimental protocol and scheme are illustrated in Fig. S12c and S12d. Results shown in Fig. S12e demonstrate that both free CUR and FUS + CUR groups showed significantly lower CUR accumulation compared to the FUS + CUR@LP group, with differences in brain accumulation reaching 10⁴ to 10⁹ orders of magnitude. These findings reveal that liposomal encapsulation effectively addresses the rapid metabolism and low bioavailability limitations of CUR. Moreover, when combined with FUS-induced opening of the BBB, this approach substantially enhances the efficiency of drug delivery to the brain.

3.3.3. In vivo photothrombotic ischemic stroke model. This research utilized a photothrombotic stroke model where rose bengal photosensitizer was administered via intraperitoneal injection in mice.⁴³ After 15 minutes of circulation, a 561 nm green laser was directed at the right M1 region of the mouse brain for 15 minutes.⁴⁴ The following day, brain tissue was harvested and subjected to TTC staining to evaluate the infarct area and location, as illustrated in Fig. 5a.⁴⁵ Brain sections with TTC staining were photographically documented, as shown in Fig. 5b. White arrows indicate the infarcted regions. A white circular region in the upper right hemisphere corresponding to the M1 infarct location demonstrates the most pronounced infarct area. This infarction resulted from rose bengal activation by 25 mW 561 nm laser irradiation, which generated singlet oxygen and ROS, leading to subsequent damage to blood vessels and surrounding tissues. The process involved interaction with platelets, fibrin, and coagulation factors, ultimately forming intravascular plaques leading to photothrombotic stroke. Due to individual variations in mouse cerebral vasculature, infarct areas showed varying degrees of prominence. In some cases, as shown in Fig. 5b, brain tissue damage was primarily evidenced by right hemisphere edema rather than distinct infarct regions.

Fig. 5 (a) The scheme, setting, and flow chart of the photothrombotic ischemic stroke model. (b) Image of TTC-staining infarct area of brain tissue. Evaluation of photothrombotic stroke combined with FUS stimulation. (c) Timeline diagram of photothrombotic stroke and FUS stimulation, (d) Evans blue leakage and TTC staining of cerebral infarction area, and (e) survival rate of mice one day after stroke and FUS stimulation.

To evaluate the safety and survival rate of bilateral brain stimulation in mice, FUS was applied to the contralesional M1 area one hour after photothrombotic stroke. Following FUS stimulation, Evans blue dye was injected through the tail vein to investigate brain permeability. The mice were sacrificed one day after the stroke, and TTC staining was performed to measure the infarct area. The dual staining method was used to observe Evans blue penetration from the left hemisphere to the right hemisphere, as shown in Fig. 5c. Brain sections were examined using both conventional photography and fluorescence microscopy, as shown in Fig. 5d. In conventional imaging, the stroke area in the right hemisphere showed no red coloration from TTC oxidation but instead appeared light blue due to Evans blue staining. This indicated that FUS stimulation of the contralesional M1 region induced localized opening of the BBB in the left hemisphere, whereas ischemic injury in the ipsilesional hemisphere increased baseline vascular permeability. The combined effects contributed to enhanced Evans blue penetration into the infarcted region. Fluorescence imaging revealed Evans blue signals at the FUS stimulation site, confirming successful BBB opening in the left hemisphere, accompanied by increased permeability in the damaged right hemisphere, resulting in enhanced Evans blue penetration. To assess the safety of this dual stimulation approach, mice were observed for a one-day recovery period to monitor their physiological stability. As shown in Fig. 5e, all three mice maintained stable vital signs under dual stimulation, demonstrating that the combination of stroke and FUS treatment was safe.

3.3.4. In vivo therapy of CUR@LP-SHp. To evaluate the therapeutic effects of CUR@LP and CUR@LP-SHp in photothrombotic stroke, both formulations were combined with FUS to treat stroke-induced mice, followed by a 48 hour recovery period (D + 2), as shown in Fig. 6a. We refer to this treatment as FUS-assisted liposomal drug delivery for photothrombotic ischemic stroke treatment, and the detailed flowchart is presented in Fig. S13. TTC staining was used to analyze changes in cerebral infarct volume ratios, as shown in Fig. 6b. In the photothrombotic stroke group (D + 1), significant infarct areas were observed in the right hemisphere of the brain sections. After CUR@LP-SHp treatment, the white infarct area was significantly smaller compared to both the stroke-only group and the CUR@LP treatment group, demonstrating the superior therapeutic efficacy of CUR@LP-SHp. Using ImageJ software to analyze the infarct volume ratios from TTC-stained brain sections, as shown in Fig. 6c, the average infarct volume ratios were 23.7%, 10.1%, and 1.8% for the stroke-only group, CUR@LP treatment group, and CUR@LP-SHp treatment group, respectively. Using the mathematical equation (eqn (8)) to calculate the quantification results further demonstrated the excellent therapeutic effect of CUR@LP-SHp, which reduced the infarct volume by approximately 92%. This reveals that modifying CUR liposomes with SHp enhances drug affinity for the ischemic stroke region, thereby achieving highly efficient therapeutic outcomes by significantly reducing infarct volume.

Fig. 6 (a) Timeline diagram of photothrombotic stroke treatment combining antioxidant nanoliposomal drugs with FUS, and (b) TTC staining showing changes in cerebral infarction in photothrombotic stroke mice treated with CUR@LP and CUR@LP-SHp. (c) Quantification graph of cerebral infarct volume using TTC staining (n = 3). (The significance is marked as *p < 0.05, **p < 0.01, and ***p < 0.001 compared to the control group).

3.3.5. Histochemical staining. To investigate glial cell expression in photothrombotic stroke mouse models treated with antioxidant nanoliposomal drugs, protein analysis was performed on samples isolated from the ischemic penumbra using GFAP as an astrocyte marker. As shown in Fig. 7b, no significant difference in GFAP expression was observed between the FUS-only group and the sham group, indicating that FUS alone did not induce astrocyte activation under non-ischemic conditions. In contrast, the photothrombotic stroke group exhibited significantly higher GFAP expression compared to the control group, indicating astrocyte proliferation in the ischemic stroke region. While this proliferation can protect surrounding healthy brain tissue from ischemic toxins, excessive activation can trigger inflammatory responses and impede axonal regeneration, hindering neuronal functional recovery. After treatment with the antioxidant nanoliposomal drug CUR@LP-SHp, GFAP expression was significantly reduced compared to the stroke group and was notably lower than in the CUR@LP treatment group. This demonstrates that the antioxidant and anti-inflammatory properties of CUR@LP-SHp effectively suppress astrocyte activation, achieving neuroprotective effects. To examine changes in GFAP protein density across treatment groups, protein quantification was performed using ImageJ. As shown in Fig. 7c, the relative GFAP protein density in the ischemic penumbra was four times higher than in normal mice. However, after two days of CUR@LP-SHp treatment, GFAP expression was significantly reduced by approximately half, effectively inhibiting astrocyte activation.

Fig. 7 (a) IHC staining of a brain section with GFAP. (b) Western blot analysis of GFAP expression. (c) Quantification graph of the signal of GFAP.

Following the protein analysis results showing GFAP-positive astrocyte overexpression in the ischemic stroke region, immunohistochemical staining was performed to further examine morphological changes in GFAP-positive astrocytes. Images were recorded at 40x magnification using optical microscopy, as shown in Fig. 7a. In the ischemic boundary zone, GFAP-positive astrocyte density increased substantially, with dense astrocytic fibers surrounding the ischemic boundary zone and forming an expanded glial scar wall (shown in the black frame in the figure). After treatment with CUR@LP (25 mg kg⁻¹), astrocytic fibers were still present in the ischemic stroke boundary region. However, treatment with CUR@LP-SHp (25 mg kg⁻¹) significantly reduced the number of GFAP-positive astrocytes, demonstrating their ability to repair neuronal damage in the ischemic stroke region, are consistent with the protein analysis results. To evaluate the neuroprotective efficacy of the formulations, NeuN was used as a neuronal marker to assess neuronal survival in the ischemic cortex (Fig. S15). In the photothrombotic stroke group, NeuN-positive neurons in the ischemic cortex were substantially reduced compared with the contralateral side, indicating significant neuronal loss following ischemic injury. In contrast, treatment with CUR@LP and CUR@LP-SHp resulted in an apparent preservation of NeuN-positive neurons in the ischemic region, suggesting improved neuronal survival. Using optical microscopy at 40× magnification, we examined H&E-stained images to investigate cellular morphological changes between the ischemic boundary zone and normal regions. As shown in Fig. 8a, compared to the normal region, neurons in the stroke boundary region of the stroke-only group showed significant pathological changes. Most neurons exhibited shrunken morphology, nuclear pyknosis, and vacuolization, as indicated by the white dotted area in the figure. This demonstrates that ischemic infarction caused extensive neuronal damage. After treatment with CUR@LP (25 mg kg⁻¹), while there was some reduction in neuronal vacuolization in the ischemic region, the cell nuclei remained smaller compared to those in the normal region. However, following treatment with CUR@LP-SHp (25 mg kg⁻¹), cellular vacuolization was markedly reduced, and nuclear size showed recovery approaching that of the normal group, as indicated by the white arrows in the figure. As shown in Fig. S16, LP and free CUR treatments failed to significantly reduce the infarct area, with distinct tissue vacuolization still visible within the demarcated ischemic regions. This reveals that the antioxidant nanoliposomal drug CUR@LP-SHp, through its antioxidant and anti-inflammatory properties, can improve ischemic brain neuronal damage and demonstrates potential neuroprotective effects without the nonspecific therapeutic effects from free CUR or the liposomal carrier alone.

Fig. 8 H&E tissue staining of mouse brain coronal sections showing ischemic and surrounding normal regions in (a) stroke-only group, CUR@LP, and SHp-CUR@LP (25 mg kg⁻¹) treatment groups, one hour after cerebral infarction. The NIR-LED PBM evaluation with control, 770 nm, and 810 nm groups. (b) Gross view of the harvested brain. (c) The infarction area of the brain slices (n = 3). (d) H&E staining results of brain slices. (The significance is marked as *p < 0.05 and ***p < 0.001 compared to the control group).

NIR PBM demonstrated therapeutic efficacy in treating photothrombotic stroke, as evaluated through multiple imaging and staining methods. The experiment compared three conditions: a control group (no treatment), and two treatment groups using 770 nm and 810 nm NIR PBM, respectively. As shown in Fig. 8b, the control group exhibited significant left frontal brain injury, while both NIR PBM treatment groups showed reduced damage areas, suggesting neuroprotective effects. TTC staining Fig. 8c revealed severe damage in the control group's left brain, characterized by light red coloration with black regions indicating infarction. The 770 nm-treated brains showed uniform red coloration, suggesting substantial recovery, while 810 nm-treated brains showed intermediate improvement with some remaining white areas. H&E staining (Fig. 8d) revealed detailed tissue changes. Control group samples showed a clear distinction between normal tissue (dense, dark pink) and stroke-affected areas (lighter, porous), indicating cellular necrosis. The 770 nm treatment group showed minimal boundary between healthy and damaged tissue, with consistently dense tissue structure similar to healthy controls. The 810 nm group showed moderate improvement in tissue density and organization, though less pronounced than the 770 nm group.

4. Conclusions

This study synthesized nanoliposomal CUR drugs using thin-film hydration and ultrasonic homogenization methods. The surface of CUR@LP was modified with DSPE-PEG2000-SHp to create ischemic stroke-targeted nanoliposomal CUR. Analysis of the nanocomposite materials’ morphology, size, antioxidant activity, and stability revealed that nanoliposomes could encapsulate CUR within their bilayer membranes, addressing its poor solubility. This encapsulation also protected CUR from degradation in acidic environments, preserving its antioxidant activity, extending its in vivo retention time, and increasing its bioavailability. The surface negative charge of the nanoliposomes provided electrostatic repulsion that enhanced CUR@LP dispersion, preventing aggregation and improving the biological size stability of the nanocarrier. CUR's excellent antioxidant active sites can eliminate reactive oxygen species and regulate oxidative stress at both cellular and tissue levels. These findings suggest that CUR@LP-SHp shows promising potential for treating ischemic stroke. In this work, a photothrombotic ischemic stroke model was established using Rose Bengal dye, combined with FUS stimulation to generate cavitation effects from microbubbles in the brain, temporarily opening the BBB. The stroke-targeted CUR@LP-SHpwas administered via tail vein injection, allowing the nanocomposite drug to enter the ischemic stroke region of the brain through the circulatory system. Subsequently, immunofluorescence tissue staining was used to evaluate the therapeutic efficacy of the nanocomposite drug CUR@LP-SHp.

Author contributions

T. Y. Su, G. T. Huang, and M. Y. Chang contributed to the study design and drafted the article. W. T. Huang participated in methodological operations. Y. L. Lin participated in data analysis and interpretation. C. C. Su, M. H. Chan, M. Y. Hsiao, and R. S. Liu contributed to the study conception and design, and were responsible for critical review of the manuscript and its final editing. All authors have read and approved the final version of the article.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

The data supporting the findings of this study are available within the article and its supplementary information (SI). The SI includes experimental schematics (Fig. S1–S3 and S13), material characterization (Fig. S4 and S5), drug release profiles (Fig. S6 and S12), in vitro cell studies (Fig. S7 and S8), confocal imaging analyses (Fig. S9–S11), Western blot data (Fig. S14), and histological evaluations (Fig. S15 and S16). The supplementary information is available online. See DOI: https://doi.org/10.1039/d5bm01596k.

Additional data related to this study are available from the corresponding author upon reasonable request.

Acknowledgements

This work was financially supported by the National Science and Technology Council (NSTC) in Taiwan through grants numbered NSTC 114-2113-M-002-001 to R. S. Liu and 114-2113-M-A49-031-MY3 to M. H. Chan.

References

1. D. Calvet, Rev. Med. Interne, 2016, 37, 19.
2. J. Wang, H. Zhang, D. L. Ni, W. P. Fan, J. X. Qu, Y. Y. Liu, Y. Y. Jin, Z. W. Cui, T. Y. Xu, Y. Wu, W. B. Bu and Z. W. Yao, Small, 2016, 12, 3591.
3. D. Barthels and H. Das, BBA, Mol. Basis Dis., 2020, 1866, 165260.
4. M. Gauberti, S. M. de Lizarrondo and D. Vivien, J. Thromb. Haemostasis, 2021, 19, 1618.
5. L. Catanese, J. Tarsia and M. Fisher, Circ. Res., 2017, 120, 541.
6. J. L. Kong, R. X. Chu and Y. Wang, Adv. Funct. Mater., 2022, 32, 2209405.
7. C. Li, S. H. Qin, Y. Wen, W. Zhao, Y. J. Huang and J. Liu, J. Controlled Release, 2022, 349, 902.
8. B. Salehi, Z. Stojanovic-Radic, J. Matejic, M. Sharifi-Rad, N. V. A. Kumar, N. Martins and J. Sharifi-Rad, Eur. J. Med. Chem., 2019, 163, 527.
9. Y. Peng, M. Y. Ao, B. H. Dong, Y. X. Jiang, L. Y. Yu, Z. M. Chen, C. J. Hu and R. C. Xu, Drug Des., Dev. Ther., 2021, 15, 4503.
10. L. Subedi and B. P. Gaire, ACS Chem. Neurosci., 2021, 12, 2562.
11. R. X. Chang, L. R. Chen, M. Qamar, Y. J. Wen, L. Z. Li, J. Y. Zhang, X. Li, E. Assadpour, T. Esatbeyoglu, M. S. Kharazmi, Y. Li and S. M. Jafari, Adv. Colloid Interface Sci., 2023, 318, 102933.
12. H. Y. Hong, J. S. Choi, Y. J. Kim, H. Y. Lee, W. Kwak, J. Yoo, J. T. Lee, T. H. Kwon, I. S. Kim, H. S. Han and B. H. Lee, J. Controlled Release, 2008, 131, 167.
13. Z. F. Dong, L. Tang, Y. Zhang, X. Y. Ma, Y. Yin, L. Kuang, Q. Fan, B. Y. Wang, X. Y. Hu, T. Y. Yin and Y. Z. Wang, Adv. Funct. Mater., 2024, 34, 2309167.
14. A. Grayston, Y. J. Zhang, M. Garcia-Gabilondo, M. Arrúe, A. Martin, P. Kopcansky, M. Timko, J. Kovac, O. Strbak, L. Castellote, S. Belloli, R. M. Moresco, M. Picchio, A. Roig and A. Rosell, J. Cereb. Blood Flow Metab., 2022, 42, 237.
15. A. Goiaei, U. Adhikari and M. L. Berkowitz, ACS Chem. Neurosci., 2015, 6, 1296.
16. J. Li, Y. Zhang, Z. C. Lou, M. X. Li, L. Cui, Z. R. Yang, L. J. Zhang, Y. Zhang, N. Gu and F. Yang, Small, 2022, 18, 2201123.
17. P. Machado, J. R. Eisenbrey, B. Cavanaugh and F. Forsberg, Ultrasound Med. Biol., 2014, 40, 2321.
18. G. C. Terstappen, A. H. Meyer, R. D. Bell and W. D. Zhang, Nat. Rev. Drug Discovery, 2021, 20, 362.
19. J. B. Xie, Z. Y. Shen, Y. Anraku, K. Kataoka and X. Y. Chen, Biomaterials, 2019, 224, 119491.
20. S. Khizar, N. Alrushaid, F. A. Khan, N. Zine, N. Jaffrezic-Renault, A. Errachid and A. Elaissari, Int. J. Pharm., 2023, 632, 122570.
21. D. Guimarães, A. Cavaco-Paulo and E. Nogueira, Int. J. Pharm., 2021, 601, 120571.
22. D. E. Large, R. G. Abdelmessih, E. A. Fink and D. T. Auguste, Adv. Drug Delivery Rev., 2021, 176, 113851.
23. T. M. Allen and P. R. Cullis, Adv. Drug Delivery Rev., 2013, 65, 36.
24. R. Liu, C. Luo, Z. Q. Pang, J. M. Zhang, S. B. Ruan, M. Y. Wu, L. Wang, T. Sun, N. Li, L. Han, J. J. Shi, Y. Y. Huang, W. S. Guo, S. J. Peng, W. H. Zhou and H. L. Gao, Chin. Chem. Lett., 2023, 34, 107518.
25. J. M. Ryu, H. J. Lee, Y. H. Jung, K. H. Lee, D. I. Kim, J. Y. Kim, S. H. Ko, G. E. Choi, I. I. Chai, E. J. Song, J. Y. Oh, S. J. Lee and H. J. Han, Int. J. Stem Cells, 2015, 8, 24.
26. B. Chen, J. J. Zhao, R. Zhang, L. L. Zhang, Q. Zhang, H. Yang and J. An, Nutr. Neurosci., 2022, 25, 1078.
27. A. B. Uzdensky, Transl. Stroke Res., 2018, 9, 437.
28. M. Parcheta, R. Swislocka, S. Orzechowska, M. Akimowicz, R. Choinska and W. Lewandowski, Materials, 2021, 14, 1984.
29. Y. T. Liu, P. F. Gong, G. B. Qi, H. Tang, R. S. Gui, C. C. Qi and S. Qin, Brain Sci., 2024, 14, 152.
30. F. Prada, M. Y. S. Kalani, K. Yagmurlu, P. Norat, M. Del Bene, F. DiMeco and N. F. Kassell, Neurotherapeutics, 2019, 16, 67.
31. H. A. Kamimura, J. Flament, J. Valette, A. Cafarelli, R. Aron Badin, P. Hantraye and B. Larrat, J. Cereb. Blood Flow Metab., 2019, 39, 1191–1203.
32. A. W. Shao, D. F. Lin, L. L. Wang, S. Tu, C. Lenahan and J. M. Zhang, Aging Dis., 2020, 11, 1537.
33. N. B. Sadeer, D. Montesano, S. Albrizio, G. Zengin and M. F. Mahomoodally, Antioxidants, 2020, 9, 709.
34. C. Minnelli, E. Laudadio, R. Galeazzi, D. Rusciano, T. Armeni, P. Stipa, M. Cantarini and G. Mobbili, Antioxidants, 2019, 8, 258.
35. Y. M. Niu, X. Y. Wang, S. H. Chai, Z. Y. Chen, X. Q. An and W. G. Shen, J. Agric. Food Chem., 2012, 60, 1865.
36. H. Song, S. Kang, Y. Yu, S. Y. Jung, K. Park, S. M. Kim, H. J. Kim, J. G. Kim and S. E. Kim, Int. J. Mol. Sci., 2023, 24, 3843.
37. K. Sirivibulkovit, S. Nouanthavong and Y. Sameenoi, Anal. Sci., 2018, 34, 795.
38. J. Muñoz-Sánchez and M. E. Chánez-Cárdenas, J. Appl. Toxicol., 2019, 39, 556.
39. V. K. Tripathi, S. A. Subramaniyan and I. Hwang, ACS Omega, 2019, 4, 20882.
40. C. Duan, H. Wang, D. Jiao, Y. Geng, Q. Wu, H. Yan and C. Li, Front. Pharmacol., 2022, 13, 889226.
41. Z. Izadifar, Z. Izadifar, D. Chapman and P. Babyn, J. Clin. Med., 2020, 9, 460.
42. Y. Meng, M. Goubran, J. S. Rabin, M. McSweeney, J. Ottoy, C. B. Pople, Y. X. Huang, A. Storace, M. Ozzoude, A. Bethune, B. Lam, W. Swardfager, C. Heyn, A. Abrahao, B. Davidson, C. Hamani, I. Aubert, H. Zetterberg, N. J. Ashton, T. K. Karikari, K. Blennow, S. E. Black, K. Hynynen and N. Lipsman, Brain, 2023, 146, 865.
43. F. Fluri, M. K. Schuhmann and C. Kleinschnitz, Drug Des., Dev. Ther., 2015, 9, 3445.
44. C. J. Sommer, Acta Neuropathol., 2017, 133, 245.
45. A. Benedek, K. Móricz, Z. Jurányi, G. Gigler, G. Lévay, L. G. Hársing, P. Mátyus, G. Szénási and M. Albert, Brain Res., 2006, 1116, 159.

---

Footnote

† Co-first authors: these authors contributed equally to this work.

---