Novel Rapid In Situ Curable PEGDA/PEG Composite Hydrogel Drives Biological Closure of Aneurysm Neck

Abstract

Background: Endovascular treatments for cerebral aneurysms, including coil embolization, stent placement, and liquid embolization, are limited by recanalization, thrombosis-related risks, and insufficient biological regulation. Methods: A composite hydrogel based on poly­(ethylene glycol) diacrylate/poly­(ethylene glycol) (PEGDA/PEG) was synthesized by using a benzophenone/diphenylacetone (BP/DPA) photoinitiator system. Systematic in vitro characterizations were performed. For in vivo evaluation, carotid artery aneurysm models were established in white rabbits. Embolization outcomes were assessed using digital subtraction angiography (DSA) and color Doppler flow imaging (CDFI), while the underlying mechanisms were investigated through vascular tissue transcriptomic analysis and histological staining. Results: The 75 wt % PEGDA/PEG hydrogel achieved intravascular in situ curing within 60 s and exhibited a superhydrophilic surface with an appropriate elastic modulus. The hydrogel demonstrated excellent biocompatibility, promoted endothelial cell migration and vascular network formation, and enhanced CD31 expression. In vivo imaging confirmed complete aneurysm occlusion with the preservation of parent artery patency following embolization. Transcriptomic analysis identified claudin-1 (CLDN1) as a key regulatory gene, whose interaction with claudin-12 (CLDN12) and zonula occludens-1 (ZO-1) facilitated tight junction remodeling and endothelial migration at the aneurysm neck. Conclusions: This study demonstrates a bioactive embolization strategy integrating rapid in situ sealing, thrombus-independent embolization, and active biological repair, providing a conceptual framework for advancing aneurysm embolization materials from passive physical fillers to biologically driven therapeutic systems.

1. Introduction

Cerebral aneurysms, often referred to as “silent time bombs,” are characterized by abnormal dilation of cerebral vessel walls and carry a high risk of rupture. Aneurysmal rupture results in subarachnoid hemorrhage (SAH), one of the most catastrophic acute cerebrovascular events, with an acute-phase mortality rate of 30–40%. Among survivors, more than half suffer irreversible neurological impairments, including cognitive decline, motor dysfunction, or epileptic seizures. − Owing to its minimally invasive nature, endovascular embolization has become the first-line treatment for cerebral aneurysms. The fundamental principle of this approach involves delivering embolic agents to fill the aneurysm sac, thereby interrupting intra-aneurysmal blood flow while preserving the structural integrity and function of the parent artery. ,

Despite its widespread clinical adoption, endovascular embolization remains limited by material-related shortcomings. Conventional coils rely on mechanical filling with metallic wires, whose intrinsic hydrophobicity promotes platelet adhesion and fibrin deposition. In addition, the porous structure of coils fails to achieve complete flow isolation at the aneurysm neck, and gradual coil compaction under hemodynamic stress often leads to aneurysm recanalization. Consequently, approximately 10–30% of patients experience aneurysm recurrence due to coil compaction and persistent blood flow at the neck region. , Although flow-diverting devices (FDs) reduce recurrence by reshaping intra-aneurysmal hemodynamics, their high metal surface coverage increases the risk of perforator occlusion, necessitates long-term antiplatelet therapy, and elevates bleeding risk. Moreover, incomplete endothelialization and high device cost further restrict their clinical application. − Liquid embolic agents, such as Onyx, can fill complex aneurysm geometries via flow; however, their solidification process is difficult to precisely control, increasing the risk of nontarget embolization. In addition, these procedures require substantial operator experience, thereby increasing technical complexity and procedural risk. −

The hydrophobic metallic surfaces of coils and the grid-like, prothrombotic architecture of FDs underpin the conventional paradigm of “thrombosis-dependent embolization.” However, the thrombotic microenvironment generated by these materials inhibits endothelial cell migration, confining them to function primarily as passive physical barriers rather than biologically integrated repair platforms. As a result, durable biological sealing of the aneurysm neck remains difficult to achieve. This lack of biological regulation presents a persistent clinical dilemma in which clinicians must often balance insufficient embolization against elevated procedural risk. These limitations underscore the urgent need for next-generation embolic materials capable of dynamically interacting with the vascular microenvironment while enabling precise control over the solidification behavior.

An ideal intravascular embolic material should overcome dependence on thrombus formation and establish a dynamic balance among three essential requirements: (i) mechanical support sufficient to resist hemodynamic stress and prevent aneurysm rupture, (ii) interface biocompatibility that inhibits platelet adhesion and minimizes thrombosis, and (iii) active biological repair that promotes endothelialization and permanent aneurysm neck closure. , Poly­(ethylene glycol) diacrylate (PEGDA), owing to its unique chemical structure, offers a promising platform for addressing these challenges. The flexible −(OCH₂CH₂) _n – segments confer elasticity, allowing the material to accommodate vascular deformation in real time. − Although PEGDA has previously been explored as a physical filler for aneurysm embolization, existing systems exhibit prolonged curing times and lack the capacity to actively regulate tissue repair.

In this study, we developed a PEGDA/PEG composite hydrogel designed to integrate three key functional advantages. First, rapid photoinitiated curing within 60 s was achieved by optimizing a benzophenone (BP)/diphenylacetone (DPA) photoinitiator system, enabling precise in situ solidification under 365 nm ultraviolet irradiation, following intravascular delivery. Second, the superhydrophilic surface formed by high-density hydroxyl (−OH) groups within PEG effectively inhibits platelet adhesion, thereby creating a thrombus-free microenvironment conducive to vascular repair. Third, by modulating PEGDA concentration, the elastic modulus of the hydrogel can be tuned across the mechanical range of aneurysm walls (approximately 0.89 MPa) and normal arteries (approximately 0.4 MPa). Such mechanically mediated cues are hypothesized to influence endothelial behavior, including tight junction regulation, thereby supporting the biological closure of the aneurysm neck. Collectively, these features enable the PEGDA hydrogel to overcome the limitations of conventional embolic materials and function as a multifunctional platform integrating light-controlled curing, antithrombotic interface properties, and biological regulatory potential.

To validate the translational potential of this hydrogel, we systematically characterized its physicochemical and mechanical properties, including curing kinetics, elastic modulus, and surface wettability, and correlated these parameters with clinically relevant performance requirements. Furthermore, in vitro biological assays assessing biocompatibility, endothelial cell migration, and vascular network formation, together with in vivo animal model validation of embolization efficacy and aneurysm neck closure, were conducted to establish a clear “material property–clinical function” relationship. Through this integrated evaluation, all experiments and characterizations were designed to directly support the overarching goal of developing a high-performance embolic material for cerebral aneurysm treatment.

2. Materials and Methods

2.1. Materials

Poly­(ethylene glycol) diacrylate (PEGDA, average molecular weight 400 Da; Macklin, P902472, Shanghai, China) and poly­(ethylene glycol) (PEG, average molecular weight 200 Da; Macklin, P815607, Shanghai, China) were used as the primary hydrogel components. Benzophenone (BP; Macklin, CAS No. 119–61–9, Shanghai, China) and diphenylacetone (DPA; Shin-Etsu, CAS No. 451–40–1, Guangdong, China) were employed as the photoinitiator system.

An ultraviolet (UV) lamp (SUNmini; wavelength 365 nm; light intensity 6 mW/cm²; YUN SHENG, Shenzhen, China) was used for hydrogel photopolymerization. Human umbilical vein endothelial cells (HUVECs, catalog No. C5912) were purchased from Baidi Biotechnology Co., Ltd. (Zhejiang, China). A Live/Dead cell viability assay kit (KGaA9501, Jiangsu, China) was used for the cell viability assessment.

New Zealand White rabbits (male, 2.4–3.0 kg) were supplied by Zhuhai BesTest Biotechnology Co., Ltd. (Zhuhai, China) and used for in vivo aneurysm embolization experiments.

2.2. Material Preparation

PEGDA composite hydrogels were prepared using poly­(ethylene glycol) diacrylate (PEGDA, average molecular weight 400 Da) and poly­(ethylene glycol) (PEG, average molecular weight 200 Da) as the primary components. PEGDA was first dissolved in sterile deionized water (DI water; resistivity ≥ 18.2 MΩ·cm at 25 °C). Subsequently, PEG was added to the solution, and the mixture was mixed thoroughly until complete dissolution.

Under dark conditions, a photoinitiator system consisting of benzophenone (BP) and diphenylacetone (DPA) was added to the mixture at a total concentration of 0.5% (w/v), with a BP/DPA molar ratio of 1:1. The solution was continuously stirred until the photoinitiators were completely dissolved. During this process, exposure to ambient light was minimized to prevent premature photopolymerization. The resulting mixture was gently stirred to ensure homogeneity and then degassed by gentle shaking to minimize entrapped air bubbles, yielding a uniform prepolymer solution. The detailed compositions of each formulation are listed in Table .

1. Proportions of Each Raw Material.

sample	PEGDA (wt %)	PEG (wt %)	BP (wt %)	DPA (wt %)
A	60%	10%	0.5%	0.5%
B	65%	10%	0.5%	0.5%
C	70%	10%	0.5%	0.5%
D	75%	10%	0.5%	0.5%
E	80%	10%	0.5%	0.5%

Photopolymerization was performed using a UV lamp (SUNmini; wavelength 365 nm; light intensity 6 mW/cm²; YUN SHENG, Shenzhen, China). The prepolymer solution was poured into cylindrical molds (diameter: 11 mm; height: 3 mm) and irradiated under UV light for predetermined durations. All curing procedures were conducted in a dust-free environment to avoid contamination, and the liquid-to-solid transition was visually monitored to confirm complete gelation. After UV curing, the hydrogels were immediately transferred to sterile Petri dishes within a Class-100 laminar flow hood. Sterile DI water was added to fully immerse the hydrogels, which were then washed to remove residual unreacted monomers and photoinitiators. Specifically, the hydrogels were soaked in fresh DI water for 15 min with gentle agitation every 5 min, and this washing step was repeated three times (total washing time: 45 min). Following the final wash, excess surface water was removed using sterile absorbent paper (Scheme ).

1. Schematic Diagram Illustrating the Fabrication and Functional Mechanisms of the PEGDA Hydrogel for Cerebral Aneurysm Treatment .

^a Through optimization of the BP/DPA photoinitiator system, the PEGDA hydrogel undergoes in situ cross-linking within 60 s under 365 nm UV light irradiation, enabling precise filling of the aneurysm. Activation of the tight junction protein CLDN1 actively promotes biological closure of the aneurysm neck.

For in vitro experiments, including cell culture and mechanical testing, the washed hydrogels were used immediately. For in vivo embolization studies, the hydrogels were lyophilized under sterile conditions (0.120 mbar and −50 °C) and stored in sealed sterile containers until use.

2.3. Material Characterization

2.3.1. Scanning Electron Microscopy (SEM)

Scanning electron microscopy (SEM) was used to characterize the surface morphology of PEGDA hydrogels with different concentrations. Hydrogel samples (60–80 wt %) were frozen at −80 °C and subsequently lyophilized using a freeze-dryer (Labconco, −50 °C, 0.120 mbar) for 24 h. The dried samples were mounted on aluminum stubs and sputter-coated with a 7 nm layer of gold/palladium using a sputter coater (Leica, Wetzlar, Germany).

SEM observations were performed using a field-emission scanning electron microscope (GeminiSEM 300, ZEISS, Germany) at an accelerating voltage of 3.0 kV and a working distance of 7.1 mm. All SEM images presented in this study were acquired at a magnification of 5000× to ensure consistent comparison among sample groups. Scale bars (10 μm) were added directly to each SEM image in Figure .

1.

Curing behavior and microstructure of PEGDA hydrogels. (A) Schematic illustration of PEGDA hydrogel morphology under different UV irradiation times. (B) Representative images of PEGDA hydrogel before and after UV irradiation (365 nm, 60 s). (C) SEM images of PEGDA hydrogels with different concentrations. Using PEGDA as the primary matrix, supplemented with PEG, DPA, and BP, five PEGDA-based hydrogels with PEGDA concentrations of 60, 65, 70, 75, and 80 wt % were successfully prepared.

2.3.2. Mechanical Testing

The compressive mechanical properties of PEGDA hydrogels were evaluated by using a universal testing machine (CMT6103, China Meters Industrial Systems, China). Hydrogel samples were prepared in a cylindrical geometry with a diameter of 11 mm and a height of 3 mm. Each specimen was carefully placed between the compression platelets, ensuring proper alignment perpendicular to the loading direction.

Compression tests were performed at a constant cross-head speed of 1 mm/min at room temperature. The applied force and corresponding displacement were continuously recorded during compression, and force–displacement curves were automatically generated by the testing system. Measurements were terminated upon sample failure or when a predefined compression limit of the instrument was reached. Compressive stress–strain curves were obtained by normalizing force and displacement to the initial cross-sectional area and height of the samples, and the compressive strength and elastic modulus were calculated accordingly. All mechanical tests were conducted in triplicate (n = 3).

2.3.3. Fourier Transform Infrared Spectroscopy (FTIR)

Fourier transform infrared (FTIR) spectroscopy was used to characterize the chemical structure of PEGDA before and after photopolymerization. FTIR spectra of the PEGDA prepolymer solution and the cured PEGDA hydrogel were collected using an FTIR spectrometer (Nicolet iS20, Thermo Fisher Scientific, USA). Samples were prepared as thin films and placed directly in the sample chamber for analysis. Spectra were recorded over a wavenumber range of 4000–400 cm^–1 to cover the characteristic absorption bands of relevant functional groups.

2.3.4. Water Contact Angle Measurements

The surface wettability of the PEGDA hydrogels was evaluated by water contact angle measurements. Hydrogel samples were cut into flat slices with a thickness of approximately 3 mm and mounted on glass slides to ensure a smooth and level surface.

Contact angle measurements were performed using a contact angle goniometer equipped with a high-speed camera and a microsyringe (needle diameter: 0.5 mm). All measurements were conducted at ambient temperature (25 °C). A droplet of deionized water (2 μL) was gently deposited onto the hydrogel surface from a height of 2 mm. The interaction between the water droplet and the sample surface was recorded by a high-speed camera, and the contact angle image was captured 0.1 s after droplet stabilization on the hydrogel surface.

The contact angles were analyzed using instrument software (Krüss ADVANCE, Krüss GmbH, Germany). At least three independent measurements were performed for each sample, and the average value was reported as the final contact angle.

2.3.5. Swelling Ratio of PEGDA Hydrogel

The swelling behavior of the PEGDA hydrogels was evaluated under physiological conditions. After complete curing, hydrogel samples were gently blotted with absorbent paper to remove excess surface moisture and weighed to obtain the initial mass (m ₁). Each sample was then fully immersed in phosphate-buffered saline (PBS, pH 7.4) and incubated at 37 °C for 72 h to reach swelling equilibrium.

After incubation, the swollen hydrogels were removed from the PBS solution, and the excess surface liquid was carefully removed using absorbent paper. The swollen mass (m ₂) was immediately measured. Three independent samples were tested for each group (n = 3).

The swelling ratio was calculated according to the following equation

$$\mathrm{swelling}\mathrm{ratio}(\%)=\frac{m_2-m_1}{m_1}\times 100$$ 1

where m ₁ and m ₂ represent the initial mass of the hydrogel before swelling and the mass after equilibrium swelling, respectively.

2.3.6. Rheological Properties

Rheological characterization of the PEGDA hydrogels was performed by using a rotational rheometer (MCR 302, Anton Paar, Graz, Austria) operated in oscillatory time-sweep mode. All measurements were conducted at room temperature (25 °C) using a 25 mm diameter sand-blasted parallel-plate geometry with a fixed gap of 1 mm.

A 200 μL aliquot of the PEGDA prepolymer solution was carefully placed on the lower plate prior to curing. Time-dependent rheological measurements were carried out under ultraviolet (UV) irradiation (wavelength 360–365 nm, light intensity 6 mW/cm²), consistent with the curing conditions used for bulk hydrogel preparation. The samples were subjected to oscillatory shear at a constant strain of 1% and an angular frequency of 10 rad/s.

During the time-sweep measurement, the storage modulus (G′) and loss modulus (G″) were continuously recorded at 5 s intervals over a total duration of 120 s to monitor the sol–gel transition and evaluate the gelation behavior of the PEGDA hydrogels.

All rheological experiments were performed in triplicate for each formulation (n = 3).

2.4. In Vitro Experiments

2.4.1. Cell Culture of HUVECs

Human umbilical vein endothelial cells (HUVECs, catalog No. C5912) were obtained from Baidi Biotechnology Co., Ltd. (Zhejiang, China) and cultured at 37 °C in a humidified incubator with 5% CO₂.

Hydrogel extracts were prepared in accordance with the ISO 10993–5 and ISO 10993–12 guidelines. Briefly, fully cured PEGDA hydrogel samples from each formulation were immersed in Dulbecco’s modified Eagle’s medium (DMEM, high glucose; Gibco, Thermo Fisher Scientific, USA) at an extraction ratio of 0.2 g·mL^–1 and incubated at 37 °C for 24 h. The resulting extracts were sterilized by filtration through 0.22 μm membrane filters. Subsequently, fetal bovine serum (FBS, 10%) and penicillin–streptomycin (1%) were added to prepare the complete extraction media.

HUVECs in the experimental groups were cultured in the corresponding hydrogel extraction media, whereas cells in the control group were cultured in standard DMEM high-glucose medium supplemented with 10% FBS and 1% penicillin–streptomycin.

2.4.2. Live/Dead Staining

HUVECs were seeded into 6-well plates at a density of 1 × 10⁵ cells per well and allowed to adhere overnight under standard culture conditions. The experimental groups were subsequently cultured in the corresponding PEGDA hydrogel extraction media, while the control group was maintained in standard culture medium without hydrogel extracts. After 48 h of incubation, the cell viability was evaluated using a Live/Dead staining kit (KeyGEN BioTECH, KGaA9501, Jiangsu, China).

Cells were stained with Calcein AM to label viable cells (green fluorescence) and propidium iodide (PI) to label dead cells (red fluorescence) according to the manufacturer’s instructions. Fluorescence images were acquired by using a fluorescence microscope (Olympus, Japan). The proportions of live and dead cells were quantitatively analyzed using ImageJ software (National Institutes of Health, USA).

2.4.3. Cell Proliferation Assay

Cell proliferation was evaluated using both the Cell Counting Kit-8 (CCK-8) and 5-ethynyl-2′-deoxyuridine (EdU) incorporation assays.

2.4.3.1. CCK-8 Assay

HUVECs were seeded into 96-well plates at a density of 5 × 10³ cells per well with five replicate wells for each group. Cells were cultured in the corresponding PEGDA hydrogel extraction media or standard culture medium (control group). After incubation for 1, 3, and 5 days, CCK-8 reagent (Meilunbio, MA0218, Dalian, China) was added to each well at a volume ratio of 1:10 (CCK-8 solution to complete medium) and incubated at 37 °C for 1 h. The absorbance at 450 nm was measured using a microplate reader to evaluate cell proliferation.

2.4.3.2. EdU Incorporation Assay

Logarithmically growing HUVECs were seeded into 6-well plates at a density of 1 × 10⁵ cells per well and allowed to adhere for 12 h at 37 °C. The culture medium was then replaced with the corresponding hydrogel extraction media. After 48 h of incubation, cell proliferation was assessed using an EdU staining kit (CellorLab, CX002, Shanghai, China) according to the manufacturer’s instructions. Fluorescence images were acquired by using an inverted fluorescence microscope, and the proportion of EdU-positive cells was quantified using image analysis software. All experiments were performed in triplicate (n = 3).

2.4.4. Wound Healing Experiment

The migratory ability of HUVECs was evaluated by using a wound healing assay. Cells were seeded into 6-well plates at a density of 1 × 10⁵ cells per well and cultured until reaching complete confluence. Linear scratch wounds were generated using a sterile 200 μL pipette tip. The wells were gently washed once with phosphate-buffered saline (PBS) to remove the detached cells.

Immediately after scratching, images of the wound area were captured under an inverted microscope and recorded for 0 h. Subsequently, cells were cultured in hydrogel extraction media supplemented with 2% fetal bovine serum (FBS) for 24 h. After incubation, images of the same wound regions were captured again under identical imaging conditions.

The wound area at 0 h (S ₀) and 24 h (S ₂₄) was quantified using image analysis software, and the migration rate was calculated using the following equation

$$\mathrm{m}\mathrm{igrations}\mathrm{rate}(\%)=\frac{S_0-S_{24}}{S_0}\times 100$$ 2

where S ₀ represents the wound area at 0 h and S ₂₄ represents the wound area after 24 h of incubation.

2.4.5. Transwell Migration Experiments

Transwell migration assays were performed using 24-well Transwell chambers equipped with polycarbonate membranes with an 8 μm pore size. The lower chambers were filled with PEGDA hydrogel extraction media supplemented with 4% fetal bovine serum (FBS) to serve as a chemoattractant.

HUVECs were resuspended in a serum-free medium and seeded into the upper chambers at a density of 2 × 10⁴ cells per chamber. After incubation at 37 °C for 24 h, cells that had migrated through the membrane to the lower surface were fixed with 4% paraformaldehyde for 15 min, washed twice with phosphate-buffered saline (PBS), and gently wiped with cotton swabs to remove nonmigrated cells from the upper surface of the membrane. The migrated cells were subsequently stained with 1% crystal violet solution (Soleberg, G1062, Beijing, China).

Stained cells were observed and imaged by using an inverted optical microscope. For quantitative analysis, migrated cells were counted in five randomly selected fields per membrane by using ImageJ software, and the average number of migrated cells was calculated. All experiments were performed in triplicate (n = 3).

2.4.6. Hemolysis

Hemolysis assays were conducted to evaluate the blood compatibility of the PEGDA hydrogels. Sterile hydrogel extracts from each experimental group were prepared in normal saline by using the same extraction protocol described above.

Fresh blood samples were obtained from healthy adult volunteers following approval by the Ethics Committee of Guizhou Medical University (Approval No. 2025 Lun Shen No. 352). Written informed consent was obtained from all donors in accordance with the Declaration of Helsinki. Blood collection was performed by qualified medical personnel under aseptic conditions, and all samples were anonymized prior to use. Blood samples were collected into EDTA-containing anticoagulant tubes and centrifuged at 300g for 15 min to separate erythrocytes from the plasma. The erythrocyte fraction was collected, and the plasma layer was retained for subsequent platelet adhesion assays. The erythrocytes were diluted by adding 3 mL of an erythrocyte suspension to 11 mL of normal saline.

Subsequently, 400 μL of the diluted erythrocyte suspension was added to 1 mL of a sterile PEGDA hydrogel extract for each group. The mixtures were gently mixed and incubated at 37 °C for 3 h. Normal saline and 0.5% Triton X-100 were used as the negative and positive controls, respectively. After incubation, the samples were centrifuged at 1500g for 15 min. The supernatants were collected, photographed to document the extent of hemolysis, and quantitatively assessed by measuring the absorbance at 540 nm by using a microplate reader.

2.4.7. Platelet Adhesion Assay

Platelet adhesion assays were performed to evaluate the antithrombotic surface properties of the PEGDA hydrogels. Platelet suspensions obtained from the hemolysis experiments were used for subsequent adhesion tests. Three aliquots (1 mL each) of platelet suspension were collected, and the initial platelet concentration was determined using a fully automated hematology analyzer (Mindray BC-6900, Guangdong, China), which was recorded as the platelet count before contact (C _B).

Bare glass beads (diameter: 6 mm) were used as a positive control, while silicone-coated glass beads were used as a negative control. Cylindrical PEGDA hydrogel samples with a diameter of 11 mm and a height of 3 mm were used as test samples. The glass beads and hydrogel samples were designed to have approximately equivalent contact areas with the platelet suspension. Each sample was incubated with the platelet suspension at 37 °C for 3 h.

After incubation, the platelet concentration in the supernatant was measured again by using the hematology analyzer and recorded as the platelet count after contact (C _A). The platelet adhesion rate was calculated according to the following equation

$$\mathrm{platelet}\mathrm{adherence}\mathrm{rate}(\%)=\frac{C_{\mathrm{B}}-C_{\mathrm{A}}}{C_{\mathrm{B}}}\times 100$$ 3

where C _B represents the platelet count before contact and C _A represents the platelet count after contact.

2.4.8. Tube Forming Experiments

The angiogenic potential of PEGDA hydrogels was evaluated by using an in vitro tube formation assay. HUVECs were pretreated with the corresponding PEGDA hydrogel extraction media for 72 h prior to the experiment. Growth factor-reduced Matrigel (Beyotime, C0372, Shanghai, China) was diluted 1:1 with serum-free medium, and 150 μL of the diluted Matrigel was added to each well of a 24-well plate. The plate was incubated at 37 °C for 30 min to allow for complete gelation.

After Matrigel polymerization, HUVECs were seeded onto the Matrigel-coated wells at a density of 6 × 10⁴ cells per well and incubated at 37 °C for 4 h to allow tube-like structure formation. Following incubation, the cells were stained with Calcein AM (KeyGEN BioTECH, KGA9501, Jiangsu, China) according to the manufacturer’s instructions. Tubular networks were observed and imaged by using a fluorescence microscope.

Quantitative analysis of tube formation, including the number of nodes, meshes, and branches, was performed using image analysis software to assess the proangiogenic effect of PEGDA hydrogels. All experiments were conducted in triplicate (n = 3).

2.4.9. Immunofluorescence Experiments

Immunofluorescence staining was performed to evaluate the expression of endothelial markers in HUVECs. Cells were seeded into 12-well plates at a density of 3 × 10⁴ cells per well and allowed to adhere overnight. The culture medium was then replaced with the corresponding PEGDA hydrogel extraction medium, and the cells were incubated for an additional 48 h.

After treatment, cells were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100, and blocked with 3% bovine serum albumin (BSA). Cells were subsequently incubated overnight at 4 °C with primary antibodies against CD31 (1:100, M1511–8, Huaan Biologicals, Zhejiang, China) and VEGF (1:100, ET1605–11, Huaan Biologicals, Zhejiang, China). After being washed with phosphate-buffered saline, cells were incubated with the corresponding fluorescent secondary antibodies for 1 h at room temperature in the dark.

Cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Solarbio, Beijing, China) for 5 min. After mounting with an antifade mounting medium, fluorescence images were captured by using a fluorescence microscope (Zeiss, Germany). The fluorescence intensities of CD31 and VEGF were quantified by using ImageJ software.

2.4.10. Real-Time Quantitative Polymerase Chain Reaction (RT-qPCR)

Total RNA was extracted from HUVECs using TRIzol reagent according to the manufacturer’s instructions. RNA concentration and purity were determined prior to reverse transcription. Complementary DNA (cDNA) was synthesized using a reverse transcription kit (Tiangen, KR118, Beijing, China).

RT-qPCR was performed using a 10 μL reaction system containing a cDNA template, gene-specific primers, and SYBR Green PCR master mix (Takara, Japan). Amplification was carried out on a real-time PCR system (Bio-Rad CFX450, USA) to assess the mRNA expression levels of hypoxia-inducible factor-1α (HIF-1α), vascular endothelial growth factor (VEGF), and cluster of differentiation 31 (CD31). The relative gene expression levels were calculated using the 2^–ΔΔCt method, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) used as the internal reference gene for normalization (Table ).

2. Primer Sequences of Genes.

gene name	forward primer (5′ to 3′)	reverse primer (5′ to 3′)
GAPDH	CACCACCAACTGCTTAGC	TTCACCACCTTCTTGATGTC
HIF-1α	ACAAGTCACCACAGGACAG	AGGGAGAAAATCAAGTCG
VEGF	CTCCGTTTCTTCTTTGCCCAG	GCACAAGCCACAAGTGTTCC
CD31	ACCGTGACGGAATCCTTCTCT	GCTGGACTCCACTTTGCAC

2.4.11. Western Blot for Protein Expression Levels of CD31 and VEGF

Western blot analysis was performed to evaluate the protein expression levels of CD31 and VEGF in the HUVECs. Cells were seeded into six-well plates at a density of 1 × 10⁶ cells per well and allowed to adhere for 12 h. The culture medium was then replaced with 75% PEGDA hydrogel extraction medium, while the control group was maintained in a conventional culture medium without extract. After 72 h of incubation, cells were lysed using RIPA lysis buffer supplemented with 1% phenylmethylsulfonyl fluoride (PMSF) protease inhibitor to obtain total cellular proteins.

Protein concentrations were determined using a bicinchoninic acid (BCA) protein assay kit, and equal amounts of protein (30 μL per sample) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) using 12.5% gels (Epizyme, PG223, Shanghai, China). Proteins were then transferred onto poly­(vinylidene difluoride) (PVDF) membranes at a constant current of 200 mA for 150 min.

The membranes were blocked with 5% nonfat milk for 1 h at room temperature and subsequently incubated overnight at 4 °C with primary antibodies against HIF-1α, VEGF, and CD31 (all diluted 1:2500; Huaan Biologicals, Zhejiang, China). After washing three times with Tris-buffered saline containing Tween 20 (TBST) (5 min each), the membranes were incubated with the corresponding horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5000; Huaan Biologicals, Zhejiang, China) for 1 h at room temperature.

Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system and imaged with a chemiluminescence imaging instrument (Clinx, Shanghai, China). The grayscale intensities of the bands were quantified using ImageJ software to calculate the relative expression levels of HIF-1α, VEGF, and CD31.

2.5. In Vivo Experiments

2.5.1. Carotid Artery Aneurysm Modeling and Embolization

All animal experiments were conducted in accordance with the ethical guidelines approved by the Animal Ethics Committee of the Affiliated Hospital of Guizhou Medical University (Ethics No. 2403302). Male New Zealand White rabbits (2.4 to 3.0 kg) were acclimatized for 1 week prior to surgery. Animals were fasted for 12 h before the procedure and anesthetized using inhalational isoflurane anesthesia (RWD, R510, Guangdong, China).

After the induction of anesthesia, the rabbits were placed in the supine position. The surgical field was shaved and disinfected with an iodine solution. The common carotid artery and external jugular vein on one side were surgically exposed. A segment (∼2 cm) of the carotid artery was temporarily occluded at both ends by using microvascular clamps. A longitudinal arteriotomy (∼5 mm) was then created on the arterial wall.

A segment of the external jugular vein was harvested, ligated at one end, and trimmed to an appropriate size to form a venous pouch. Under microscopic visualization, the venous pouch was anastomosed to the lateral wall of the carotid artery using 10–0 nylon microsutures. After the vascular clamps were released to restore blood flow and expel residual air, the apex of the venous pouch was ligated, thereby completing the carotid artery aneurysm model.

Following successful aneurysm construction, endovascular embolization was performed via a transfemoral arterial approach under real-time X-ray fluoroscopic guidance using a digital subtraction angiography (DSA) system (Artis Q biplane, Siemens). A small oblique incision was made in the right femoral artery, and an 18G indwelling needle was inserted. A 1.7 F microcatheter (Rebar-18, Medtronic, USA) was advanced through the femoral artery and abdominal aorta into the common carotid artery. Iodixanol contrast agent (0.2–0.4 mL) was injected to confirm aneurysm formation.

Subsequently, the PEGDA hydrogel was slowly injected into the aneurysm sac through the microcatheter and cured in situ under 365 nm ultraviolet irradiation. Embolization outcomes were immediately evaluated by X-ray fluoroscopy. Postembolization hemodynamics at the aneurysm site were further assessed by using color Doppler flow imaging (CDFI) (Mindray, HH6–26002399, China).

2.5.2. Hematoxylin and Eosin (H&E), Masson’s Trichrome, and Immunofluorescence Staining of Pathological Sections

2.5.2.1. Hematoxylin and Eosin (H&E) staining

Rabbits were euthanized at predetermined time points, and embolized carotid artery aneurysm tissues were harvested at 7 and 28 days postoperatively. The specimens were fixed in phosphate-buffered formalin, dehydrated through a graded ethanol series, and embedded in paraffin. Paraffin-embedded tissues were sectioned into 4–5 μm slices and subjected to standard H&E staining.

Briefly, tissue sections were deparaffinized in xylene, rehydrated through decreasing concentrations of ethanol, and rinsed in distilled water. The sections were stained with hematoxylin for 3–5 min, differentiated, blued, and counterstained with eosin for 5 min. After dehydration and clearing, the sections were mounted with a neutral resin. Histological images were acquired by using a light microscope (Nikon Eclipse E100, DS-U3 camera, Nikon, Japan).

2.5.2.2. Masson’s Trichrome Staining

Paraffin sections were deparaffinized and rehydrated prior to Masson’s trichrome staining to evaluate collagen deposition at the aneurysm neck. Sections were sequentially stained with ferric hematoxylin, acid fuchsin, phosphomolybdic acid, and aniline blue according to the standard protocols. After differentiation, dehydration, and clearing, the sections were mounted with a neutral resin. Images were obtained using a light microscope (Nikon Eclipse E100, DS-U3 camera, Nikon, Japan).

2.5.2.3. Immunofluorescence (IF) Staining

Paraffin-embedded tissue sections collected 7 and 28 days postoperatively were used for immunofluorescence analysis. After deparaffinization and antigen retrieval, sections were blocked and incubated overnight at 4 °C with primary antibodies against CD31 (Proteintech, 11265–1-AP, Wuhan, China) and β-actin (Aifang Biologicals, AF7018, Hunan, China). Following washing with phosphate-buffered saline (PBS), the sections were incubated with the corresponding secondary antibodies.

Cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI), and fluorescence images were captured by using a confocal laser scanning microscope (SpinSR10, Olympus, Japan).

2.6. Transcriptome Sequencing

2.6.1. RNA Extraction and Sequencing

To further elucidate the molecular mechanisms underlying PEGDA hydrogel-mediated aneurysm neck remodeling, transcriptomic analysis was performed using a parallel Sprague–Dawley (SD) rat aneurysm model. Fresh carotid artery aneurysm tissues were harvested from SD rats in the control and PEGDA-treated groups (n = 5 per group) and immediately snap-frozen in liquid nitrogen for RNA preservation.

Total RNA was extracted using standard protocols. RNA concentration and purity were initially assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific), with acceptable purity defined by an OD260/280 ratio between 1.8 and 2.2. RNA integrity was preliminarily evaluated by agarose gel electrophoresis and further assessed using an Agilent 5300 Bioanalyzer. Samples with an RNA quality number (RQN) of >6.5 and a total RNA yield of at least 1 μg were considered suitable for downstream library construction.

Messenger RNA (mRNA) was enriched using oligo­(dT) magnetic beads and fragmented to an average length of approximately 300 bp. Double-stranded complementary DNA (cDNA) was subsequently synthesized, followed by end repair, A-tailing, adapter ligation, and PCR amplification. The resulting libraries were purified and subjected to high-throughput sequencing on a NovaSeq X Plus platform (Meiji Bio, Shanghai, China).

2.6.2. Transcriptome Data Analysis

After RNA sequencing, a total of 30,562 genes were quantified across all samples. Gene expression levels were normalized and expressed as transcripts per million (TPM), and these normalized data were used for all subsequent analyses aimed at exploring molecular alterations associated with PEGDA hydrogel-mediated aneurysm neck remodeling.

Differential gene expression analysis was performed between the PEGDA-treated and control groups. Prior to statistical testing, data preprocessing was conducted, including format validation and screening for missing values. Gene expression distributions were assessed using normality tests implemented in the SciPy software package. For genes exhibiting approximately normal distributions, independent-samples t tests were applied. Given the relatively small sample size, the nonparametric Mann–Whitney U test was additionally employed to assess distributional differences between groups. Genes with P < 0.05 in both statistical tests were considered significantly differentially expressed.

To enhance analytical robustness while maintaining biological interpretability, the intersection of the top 100 differentially expressed genes identified by both the t test and Mann–Whitney U test was selected as the candidate gene set for subsequent analyses.

Functional enrichment analyses, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome pathway analyses, were performed to characterize the biological processes, molecular functions, cellular components, and signaling pathways associated with the candidate genes. Subcellular localization analysis was further conducted to determine the predominant intracellular distribution of the encoded proteins.

Protein–protein interaction (PPI) networks were constructed by using the STRING database to explore potential functional associations among candidate gene products. Structural modeling and interaction prediction between candidate gene-encoded proteins and pathway-related proteins were performed using HelixFold3. In addition, molecular docking simulations were conducted using AutoDock Vina to evaluate putative interaction interfaces between key proteins within enriched signaling pathways. Protein energy frustration analysis was carried out using a Frustratometer, and the resulting structural features were visualized using PyMOL to facilitate interpretation of protein interaction dynamics.

2.7. Statistical Analysis

All experiments were performed with at least three independent biological replicates. Statistical significance between groups was assessed using Student’s t test for two-group comparisons or one-way analysis of variance (ANOVA) for multiple-group comparisons, followed by Tukey’s post hoc test for pairwise comparisons when applicable. Data analysis was conducted using SPSS version 22.0 (IBM, USA) and GraphPad Prism version 8.0 (GraphPad Software, USA).

Data are presented as mean ± standard deviation (SD). Statistical significance was defined as follows: ns, P > 0.05; P < 0.05; *P < 0.01; **P < 0.001; and ***P < 0.0001.

3. Results

3.1. Development and Characterization of PEGDA Hydrogels

The curing behavior of PEGDA hydrogels under ultraviolet (UV) irradiation was systematically evaluated at different exposure durations. After 15 s of UV irradiation, the PEGDA precursor remained in a liquid state. At 30 s, the material began to gel, exhibiting an intermediate liquid–solid transition state. When the irradiation time was extended to 45 s, the hydrogel became visibly hardened; however, residual liquid was still observed on the surface. Upon irradiation for 60 s, the hydrogel fully solidified into a stable structure with a diameter of 11 mm and a thickness of 3 mm. Prolonging the irradiation time to 90 s did not result in further dimensional changes. Based on these observations, 60 s was determined to be the optimal curing time for the PEGDA hydrogel system (Figure A).

The curing capability of the PEGDA hydrogel under the optimized irradiation conditions was further verified. Prior to UV exposure, the material was in a free-flowing liquid state, whereas after 60 s of irradiation at 365 nm, it rapidly transformed into a mechanically stable gel, maintaining its predefined shape (Figure B).

The microstructural morphology of PEGDA hydrogels with different concentrations was examined by scanning electron microscopy (SEM). At 60 wt %, the hydrogel surface exhibited pronounced roughness with irregular protrusions and uneven features. As the PEGDA concentration increased to 65–70 wt %, surface roughness was moderately reduced; however, fine textures and localized heterogeneity remained visible. In contrast, hydrogels with PEGDA concentrations of 75–80 wt % displayed a smooth and dense surface morphology, with surface irregularities and microcracks largely eliminated, resulting in a more uniform and compact microstructure (Figure C).

FTIR was employed to confirm the chemical composition of the hydrogels (Figure A). In the FTIR spectrum of PEGDA, the absorption band at approximately 2920 cm^–1 corresponds to the asymmetric and symmetric stretching vibrations of methylene (−CH₂−) groups, while the peak at ∼1720 cm^–1 is attributed to the stretching vibration of the ester carbonyl (CO) group. The bands around 1460 and 1380 cm^–1 are assigned to C–H bending vibrations, and the characteristic peak near 1096 cm^–1 corresponds to C–O–C stretching vibrations of the PEG backbone.

2.

Physicochemical and mechanical characterization of PEGDA hydrogels. (A) FTIR spectra of PEGDA-based hydrogels. (B–D) Compressive strength and elastic modulus of PEGDA hydrogels with different concentrations. (E) Swelling ratio of PEGDA hydrogels measured at 37.5 °C. (F–J) Water contact angle images and quantitative analysis. (H) Dynamic time-sweep rheological analysis. (n = 3; ns, P > 0.05; P < 0.05; *P < 0.01; **P < 0.001; ***P < 0.0001.).

In the PEG spectrum, the absorption band at ∼2864 cm^–1 is associated with C–H stretching vibrations of saturated hydrocarbon chains, whereas the peak near 1059 cm^–1 represents C–O stretching vibrations. The FTIR spectrum of the PEGDA/PEG/DPA/BP composite hydrogel exhibited characteristic peaks of each component, including the C–H stretching band at 2871.9 cm^–1 and the carbonyl-related band at 1725 cm^–1. Minor shifts in peak positions and intensities were observed compared with the spectra of the individual components, indicating successful incorporation and intermolecular interactions within the hydrogel network.

Mechanical testing demonstrated a pronounced dependence of hydrogel properties on the PEGDA concentration (Figure B–D). The compressive strength increased progressively with increasing PEGDA content, reaching 15.55 and 16.55 MPa for the 70 and 80 wt % groups, respectively, representing an approximately 74% increase compared with the 60 wt % group (8.92 MPa). Similarly, the elastic modulus exhibited a continuous upward trend across the concentration gradient, with values of 28.58 and 34.76 MPa for the 70 and 80 wt % formulations, respectively, falling within a stiffness range suitable for aneurysm cavity support.

The swelling behavior of the hydrogels remained relatively stable across all formulations, with swelling ratios ranging from 20.27% to 23.37% (Figure E). Notably, the 70 wt % hydrogel exhibited a swelling ratio of 21.53%, indicating limited water uptake and minimizing the potential risk of chronic vascular wall compression following implantation.

Surface wettability analysis revealed clear differences among the hydrogel formulations. Water contact angle measurements showed that the 75 wt % hydrogel exhibited a superhydrophilic surface, with a contact angle of 14.50°, which was significantly lower than those of the other concentration groups (Figure F–J). In contrast, the contact angle of the 80 wt % group increased sharply to 57.59°, suggesting that excessively high PEGDA concentrations may alter surface energy and compromise hydrophilicity.

Rheological analysis further confirmed the concentration-dependent gelation behavior of the PEGDA hydrogels. During dynamic time-sweep measurements, the storage modulus (G′) gradually exceeded the loss modulus (G″) as curing progressed, indicating a transition from a sol state to a stable cross-linked gel network (Figure H).

Collectively, these physicochemical and mechanical characteristics demonstrate that the PEGDA hydrogels satisfy the key requirements for rapid in situ embolization and mechanical compatibility with aneurysm walls, consistent with the design rationale proposed in the Introduction.

3.2. Biocompatibility Assessment

In accordance with ISO 10993 standards, extraction media were prepared for human umbilical vein endothelial cell (HUVEC) culture, and the biocompatibility of PEGDA hydrogels was systematically evaluated using cell viability, proliferation, hemolysis, and platelet adhesion assays.

Live/Dead staining demonstrated that PEGDA hydrogels at all tested concentrations exhibited excellent cytocompatibility (Figure A,D). Quantitative analysis revealed cell viability values of 99.79, 99.67, 99.59, 100, and 99.30% for the 60, 65, 70, 75, and 80 wt % groups, respectively, which were comparable to that of the blank control group (99.88%), with no statistically significant differences observed (P > 0.05). These results meet the requirements of ISO 10993–5 and confirm the favorable biosafety profile of the PEGDA hydrogels.

3.

Biocompatibility and hemocompatibility of PEGDA hydrogels. (A, D) Representative Live/Dead fluorescence images of HUVECs cultured with PEGDA hydrogel extracts and quantitative analysis of cell viability. (B, F) Hemolysis images and corresponding absorbance values at 540 nm for positive control (AP), negative control (AN), and PEGDA hydrogel groups. (E) CCK-8 assay results showing HUVEC proliferation at 1, 3, and 5 days. (C, J) Quantitative analysis of platelet adhesion rates on PEGDA hydrogels with different concentrations. (n = 3; ns, P > 0.05; P < 0.05; *P < 0.01; **P < 0.001; ***P < 0.0001).

Cell proliferation was further assessed using the CCK-8 assay. As shown in Figure E, no significant differences in HUVEC proliferation were detected among the PEGDA hydrogel groups and the blank control at 1, 3, or 5 days (P > 0.05), indicating that variations in PEGDA concentration did not adversely affect endothelial cell growth.

Hemocompatibility was evaluated by hemolysis testing. As presented in Figure B,F, the absorbance at 540 nm for all PEGDA hydrogel extracts showed no significant difference compared with the negative control group (AN, saline) (P > 0.05). In contrast, a significant increase in absorbance was observed in the positive control group (AP, Triton X-100) (P < 0.001). These findings indicate that PEGDA hydrogels do not induce erythrocyte rupture or hemolysis, demonstrating excellent blood compatibility.

Platelet adhesion assays further revealed the pronounced antithrombogenic properties of PEGDA hydrogels (Figure C,J). The positive control surface exhibited a high platelet adhesion rate of 79.64%, reflecting strong platelet aggregation on pro-adhesive substrates, whereas the negative control showed minimal adhesion (0.84%). In comparison, PEGDA hydrogels displayed markedly reduced platelet adhesion, with adhesion rates of 7.74 and 8.95% for the 60 and 65 wt % groups, respectively. With increasing PEGDA concentration, platelet adhesion further decreased to 3.75% at 70 wt %, 3.87% at 75 wt %, and 0.36% at 80 wt %, approaching the level of the negative control.

Collectively, the low platelet adhesion and absence of hemolytic activity observed for PEGDA/PEG hydrogels confirm the formation of a nonthrombogenic interface. This property effectively overcomes the thrombosis-dependent limitations of conventional embolic materials highlighted in the Introduction and provides a favorable biological microenvironment for subsequent endothelial repair and aneurysm neck remodeling.

EdU incorporation assays revealed distinct differences in endothelial cell proliferation among PEGDA hydrogel formulations with varying concentrations (Figure A,B). The 75 wt % group exhibited the highest proliferation rate (76.12%), which was significantly greater than that of the 60 wt % (61.36%), 70 wt % (68.78%), and 80 wt % (59.92%) groups (P < 0.05), indicating that this formulation most effectively supports endothelial cell proliferation.

4.

Endothelial cell proliferation and migration on PEGDA hydrogels. (A, B) Representative EdU staining images and quantitative analysis of proliferation rates. (C, D) Scratch wound images and corresponding quantitative analysis of wound closure at 0 and 24 h. (E, F) Transwell migration images and quantitative analysis of migrated cell numbers at 24 h. (n = 3; ns, P > 0.05; P < 0.05; *P < 0.01; **P < 0.001; ***P < 0.0001).

Endothelial cell migration was evaluated using scratch wound healing and Transwell migration assays. In the scratch wound healing assay (Figure C,D), all PEGDA hydrogel groups demonstrated significantly higher closure rates compared with the blank control group (60.80%). Notably, the 75 wt % group achieved the highest scratch closure rate of 85.93%, representing an approximate 41% increase relative to the control group, suggesting a pronounced enhancement in migratory capacity at this concentration.

Consistent with these findings, the Transwell migration assay showed that the number of cells migrating through the membrane was greatest in the 75 wt % group (406 cells), which was significantly higher than that observed in the control group (305.66 cells) and in the other PEGDA concentration groups (60 wt %: 323.33 cells; 65 wt %: 343 cells; 70 wt %: 375 cells; 80 wt %: 382 cells) (Figure E,F). This corresponds to an approximately 33% increase in migratory cell number relative to the control group, further confirming the superior pro-migratory effect of the 75 wt % PEGDA hydrogel.

Collectively, these results demonstrate that PEGDA hydrogels maintain excellent in vitro biocompatibility while actively promoting endothelial cell proliferation and migration. Among the tested formulations, the 75 wt % PEGDA hydrogel consistently exhibited the most favorable performance across multiple functional assays, supporting its potential role in facilitating endothelial organization and subsequent biological closure of the aneurysm neck.

3.3. In Vitro Assessment of Angiogenic Properties of PEGDA Hydrogels

The angiogenic potential of PEGDA hydrogels was evaluated by using an in vitro tube formation assay with HUVECs. Quantitative analysis revealed that the 75 wt % PEGDA hydrogel exerted distinct effects on vascular network organization compared with the control and other concentration groups (Figure A–D). Although the number of branches in the 75 wt % group was lower than that of the control group, the number of tube formation nodes reached 306, representing a 101% increase relative to the control group (152.33). , In addition, the number of meshes increased to 32.67, approximately 3.9-fold higher than that observed in the control group (8.33). These findings indicate that the 75 wt % PEGDA hydrogel promotes the formation of a denser and more interconnected vascular-like network, characterized by enhanced nodal connectivity and network complexity rather than by increased branch formation alone. In contrast, the number of nodes in the other PEGDA concentration groups (60–80 wt %) ranged from 192.33 to 267.67, remaining substantially lower than that observed in the 75 wt % group.

5.

In vitro angiogenic assessment of PEGDA hydrogels. (A) Representative tube formation images of HUVECs. (B–D) Quantitative analysis of tube formation parameters, including mesh number, node number, and branch number. (E, G, H) Immunofluorescence staining and quantitative analysis of VEGF and CD31 protein expression (n = 3). (F, I, J) Western blot analysis and corresponding quantification of VEGF and CD31 protein levels. (K–M) RT-qPCR analysis of HIF-1α, VEGF, and CD31 gene expression. (n = 3; ns, P > 0.05; P < 0.05; *P < 0.01; **P < 0.001; ***P < 0.0001).

Based on the integrated results of cytocompatibility, proliferation, migration, and tube formation assays, the 75 wt % PEGDA hydrogel was selected for subsequent mechanistic investigations.

Immunofluorescence staining further demonstrated that the 75 wt % PEGDA hydrogel selectively modulated the expression of endothelial-associated proteins (Figure E,G,H). Quantitative analysis showed a modest increase in VEGF fluorescence intensity in the 75 wt % group (26.95) compared with the control group (25.23), although this difference did not reach statistical significance (P > 0.05). In contrast, CD31 fluorescence intensity was significantly increased in the 75 wt % group, reaching 57.54, corresponding to a 53.1% increase relative to the control group (P < 0.05). These observations were further corroborated by Western blot analysis, which revealed a 23% upregulation of CD31 protein expression in the 75 wt % group compared with the control (P < 0.05), while VEGF protein levels remained unchanged (Figure F,I,J).

At the transcriptional level, RT-qPCR analysis demonstrated that PEGDA hydrogels exerted differential regulatory effects on angiogenesis-related genes. HIF-1α expression in the 75 wt % group was increased by approximately 1.25-fold relative to the control group (P < 0.05), suggesting mild activation of hypoxia-responsive signaling pathways. In contrast, no significant differences were detected in the mRNA expression levels of VEGF or CD31 among the groups, indicating that the PEGDA hydrogel promotes endothelial organization without inducing excessive angiogenic gene activation or impairing endothelialization.

3.4. Embolization Efficacy and Aneurysm Occlusion

To evaluate the embolization performance of the PEGDA hydrogel, in vitro arterial aneurysm models with wide necks, narrow necks, and irregular morphologies were fabricated by using three-dimensional printing technology. Embolization efficacy was first assessed in vitro using PEGDA hydrogel doped with 0.1% crystal violet for visualization (Figure S2), which demonstrated complete filling of the aneurysm cavities with minimal material leakage.

Subsequently, a total of 20 carotid artery aneurysm models were successfully established in New Zealand White rabbits using the venous pouch transplantation method. Among these, 10 aneurysms were treated by using balloon-assisted embolization (Video S1 and Figure A). Nontarget vessel embolization occurred in one case, indicating a low but non-negligible procedural risk. To further improve intravascular delivery precision and enhance translational feasibility, a customized fiber-optic catheter system was developed to enable controlled intravascular photopolymerization (Figure S3).

6.

In vivo embolization evaluation of carotid artery aneurysms in New Zealand White rabbits. (A) Schematic illustration of the aneurysm model and embolization procedure. (B) Representative digital subtraction angiography (DSA) images before and after PEGDA hydrogel embolization in posteroanterior (PA) and lateral (LAT) projections. (C) Color Doppler flow imaging (CDFI) showing blood flow patterns within the aneurysm region before and after embolization.

DSA was employed to evaluate embolization outcomes. In untreated aneurysm models, posteroanterior (PA) projection images clearly revealed aneurysm protrusions (AN) originating from the parent artery (Figure B1,B3). Following PEGDA hydrogel embolization, the aneurysm cavity was no longer visualized (red dashed circle), while the feeding artery (FA) remained patent (Figure B2,B4). These findings were further confirmed in lateral (LAT) projection images, in which the aneurysm was clearly visible prior to treatment (Figure B5,B7) but completely disappeared after embolization, with preservation of the parent artery and distal vessels (Figure B6,B8).

Partial contrast defects observed in the feeding artery on PA projections were absent on the corresponding LAT views, indicating that these findings resulted from projection overlap rather than true vascular stenosis or occlusion (Videos S2, S3, S4, and S5). This interpretation was further corroborated by Doppler imaging.

CDFI provided complementary hemodynamic evidence. In baseline control images (Figure C, Baseline), prominent blood flow signals were detected within the aneurysm cavity, which were characterized by turbulent flow patterns and flow acceleration at the aneurysm neck. Following embolization (Figure C, Treatment), blood flow signals within the aneurysm region were markedly attenuated, with the PEGDA hydrogel occupying the aneurysm cavity and effectively blocking intra-aneurysmal blood flow while preserving parent artery patency.

Collectively, these in vitro and in vivo imaging results demonstrate that the PEGDA/PEG hydrogel enables stable and effective aneurysm occlusion while maintaining parent vessel integrity, supporting its potential as a precise and controllable embolic material for endovascular aneurysm treatment.

At 7 and 28 days postembolization, New Zealand White rabbits were euthanized, and embolized aneurysms were harvested for macroscopic and histological evaluation. Gross examination revealed no significant adhesion between the embolized aneurysm models and the surrounding tissues, indicating favorable local tissue compatibility (Figure S4).

Hematoxylin and eosin (H&E) staining showed that in the control group (Baseline), the aneurysm cavity persisted at 7 days, accompanied by a dilated aneurysm neck. In contrast, in the PEGDA embolization group, the hydrogel was stably retained within the aneurysm cavity, and early signs of aneurysm neck repair were evident. By 28 days, the aneurysm cavity in the control group remained patent, and the aneurysm neck failed to achieve closure. Conversely, in the PEGDA embolization group, the hydrogel remained securely in place without extrusion, and the aneurysm neck was completely closed.

Masson’s trichrome staining further demonstrated progressive extracellular matrix remodeling at the aneurysm neck. At 7 days post-treatment, scattered collagen fiber deposition was observed in the embolization group, indicating an early reparative response. By 28 days, the fibrous layer at the aneurysm neck had markedly thickened, forming a continuous and dense collagen barrier. As indicated by the black triangular arrow, the aneurysm neck was entirely sealed, and the aneurysm cavity was structurally separated from the parent artery (Figure A). Importantly, histological examination of distal vascular segments in the 7-day embolization group confirmed preserved vascular patency on both H&E and Masson’s trichrome staining (Figure S5).

7.

Histological evaluation of carotid artery aneurysms in New Zealand White rabbits. (A) Representative H&E and Masson’s trichrome-stained sections of carotid artery aneurysms from control and PEGDA embolization groups at 7 and 28 days. (B) Immunofluorescence staining for CD31 in carotid artery aneurysm sections from control and PEGDA embolization groups at 7 and 28 days.

Immunofluorescence staining revealed consistent endothelialization patterns (Figure B). In the 7-day control group, CD31-positive signals appeared as scattered punctate staining within the aneurysm neck region with no continuous endothelial lining, indicating persistence of the aneurysm. In contrast, the PEGDA embolization group exhibited CD31-positive signals covering the aneurysm neck region, suggesting active endothelial cell migration. By 28 days, although CD31-positive signals increased in the control group, they remained discontinuous and failed to form a complete endothelial layer, indicating that spontaneous self-repair was insufficient to achieve aneurysm neck closure within this period. In contrast, the PEGDA embolization group displayed continuous and dense CD31-positive bands completely covering the aneurysm neck, confirming successful endothelial layer formation and definitive aneurysm neck closure.

Furthermore, combined DAPI staining (blue, nuclei) and β-actin staining (red, cytoskeletal protein) demonstrated an orderly nuclear arrangement and well-organized cytoskeletal architecture in the embolization group, further supporting complete endothelial remodeling and tissue integration.

To assess the reproducibility of embolization-induced aneurysm neck closure across species, a parallel aneurysm model was established in Sprague–Dawley rats. Histological analysis at 28 days postembolization similarly confirmed complete aneurysm neck closure in the rat model (Figure S6), supporting the robustness and cross-model applicability of the PEGDA hydrogel embolization strategy.

3.5. Exploring the Molecular Mechanisms of PEGDA Hydrogel-Mediated Aneurysm Repair Based on Transcriptomic Analysis

To investigate the molecular mechanisms underlying PEGDA hydrogel-mediated aneurysm neck remodeling, a parallel Sprague–Dawley (SD) rat aortic aneurysm model was established for transcriptomic analysis. Following sample collection, total RNA was extracted, subjected to quality control, and sequenced by using the NovaSeq X Plus platform. All samples met the quality control criteria, and no missing data were detected in either the treatment or control groups.

After data preprocessing, a total of 30,562 genes were quantified. Global distribution assessment indicated no significant deviation from normality at the data set level (P > 0.05), supporting the application of parametric statistical approaches. Differential gene expression between the PEGDA-treated and control groups was evaluated using both independent-sample t tests and Mann–Whitney U tests. The top 100 differentially expressed genes identified by each method were intersected, yielding seven overlapping candidate genes (Figure A).

8.

Identification and pathway analysis of CLDN1 based on transcriptomic profiling. (A) Venn diagram showing overlap of the top 100 differentially expressed genes identified by independent-samples t tests and Mann–Whitney U tests. (B) Box plots of CLDN1 expression levels in treatment and control groups. (C) Protein–protein interaction network of CLDN1. (D, E) GO enrichment analysis for biological processes and cellular components associated with CLDN1. (F) KEGG pathway enrichment analysis of CLDN1. (G) Reactome pathway enrichment analysis of CLDN1. (H) Protein domain and feature analysis of CLDN1. (I) Subcellular localization enrichment analysis of CLDN1. (J) KEGG tight junction pathway map highlighting CLDN1-related interactions.

Among these candidates, claudin-1 (CLDN1) exhibited the most significant differential expression between the groups. CLDN1 expression was markedly higher in the PEGDA-treated group compared with the control group (Figure B), with statistical significance confirmed by both the independent-samples t test (p = 0.000073) and the Mann–Whitney U test (p = 0.007397). Based on these results, CLDN1 was selected as the key candidate gene for subsequent pathway and interaction analyses.

Protein–protein interaction analysis using the STRING database revealed that CLDN1 occupies a central position within intercellular junction-related networks (Figure C). Functional enrichment analyses further demonstrated a strong association between CLDN1 and tight-junction-related biological processes. Gene Ontology (GO) enrichment analysis indicated significant involvement in cellular components and biological processes associated with intercellular junction organization (Figure D,E). Consistently, the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis identified the tight junction signaling pathway as the primary enriched pathway (Figure F), which was further supported by Reactome pathway analysis (Figure G). Additional analyses of protein domains, functional features, and subcellular localization consistently localized CLDN1 to cell–cell junction structures (Figure H,I).

Based on the KEGG tight junction pathway map (Figure J), CLDN1 was found to form a functional complex with claudin-12 (CLDN12). This complex interacts with zonula occludens-1 (ZO-1) via cingulin-mediated binding, thereby participating in tight junction assembly and regulation. Collectively, these transcriptomic findings identify CLDN1 as a key molecular node associated with PEGDA hydrogel-induced aneurysm neck remodeling.

Based on the FASTA sequence information on CLDN1, CLDN12, and ZO-1 obtained from the Protein Data Bank (PDB), HelixFold3 was employed to predict the three-dimensional structures and potential interaction interfaces among CLDN1, CLDN12, and ZO-1. The predicted protein–protein interaction models were visualized using PyMOL, including interaction interfaces (Figures A and S7) and surface electrostatic potential distributions (Figure B).

9.

Structural modeling and protein interaction analysis of the CLDN1–CLDN12–ZO-1 complex. (A) Schematic representation of predicted interaction sites within the CLDN1–CLDN12–ZO-1 protein complex. (B) Electrostatic surface potential maps of the interaction interfaces among CLDN1, CLDN12, and ZO-1. (C) Molecular docking analysis showing hydrogen bond interactions between CLDN1 and PEGDA molecular fragments. (D) Local energy frustration map of the CLDN1–CLDN12–ZO-1 complex; red lines indicate highly frustrated interactions, green lines indicate minimally frustrated interactions, and yellow lines denote hydrogen bond interactions. (E) Contact maps of the CLDN1–CLDN12–ZO-1 complex (left: configurational; right: mutational). (F) Density of contacts in sequence space for the CLDN1–CLDN12–ZO-1 complex (left: configurational; right: mutational). (G) Fraction of contacts across different frustration classes in the CLDN1–CLDN12–ZO-1 complex (left: configurational; right: mutational).

Structural modeling suggested that CLDN1, CLDN12, and ZO-1 form a multiprotein complex stabilized by hydrogen bond interactions across five distinct interface regions (Figure A). Analysis of surface electrostatic potentials revealed complementary charge distributions at the interaction surfaces, supporting the structural feasibility of complex formation (Figure B).

To explore potential interactions between the material and protein targets, AutoDock Vina was used to perform molecular docking analysis between CLDN1 and PEGDA molecular fragments. The docking results indicated the presence of hydrogen bond interactions between CLDN1 and PEGDA, suggesting a possible interface for the material–protein interaction at the molecular level (Figure C).

Furthermore, local energy frustration within the CLDN1–CLDN12–ZO-1 complex was evaluated using the Frustratometer. The analysis revealed regions of low and high local frustration at the protein–protein interaction interfaces (Figures D and S8). Notably, hydrogen-bond-rich regions at the CLDN1–CLDN12 and CLDN12–ZO-1 interfaces predominantly exhibited low local frustration, indicating energetically favorable and structurally stable interactions.

Additional configurational and mutational frustration analyses, including contact maps (Figure E), contact density distributions in sequence space (Figure F), and the fraction of contacts across different frustration classes (Figure G), further supported the energetic stability of the CLDN1–CLDN12–ZO-1 interaction network. Collectively, these in silico analyses provide structural and energetic insights into the potential role of CLDN1-centered tight junction complexes in PEGDA hydrogel-mediated aneurysm neck remodeling.

4. Discussion

Endovascular therapy has long been a primary treatment modality for aneurysms. However, embolization materials currently used in clinical practice still suffer from several limitations, including recanalization, thrombosis-related risks, and insufficient biological regulation. , The development of novel embolic materials is, therefore, of critical importance. Although poly­(ethylene glycol) diacrylate (PEGDA) has been widely applied in the field of biomaterials owing to its hydrophilic backbone and photo-cross-linkable properties, existing PEGDA-based embolic systems have not been systematically optimized to meet the core requirements of aneurysm embolizationnamely, rapid and controllable in situ curing, mechanical compatibility, and active biological repairthereby limiting their clinical applicability. −

Accordingly, this study was designed to address these unmet clinical demands through the development of a PEGDA/PEG composite hydrogel that integrates precise in situ curing, a nonthrombogenic interface, and the ability to actively promote biological closure of the aneurysm neck. The material performance was comprehensively evaluated across physicochemical characterization, in vitro biological assays, in vivo embolization studies, and multiscale mechanistic analyses, ensuring a coherent structure–function–mechanism framework.

Upon irradiation with 365 nm ultraviolet light, benzophenone (BP) transitions from the ground state (S₀) to the excited singlet state (S₁) and subsequently undergoes intersystem crossing to the triplet state (T₁), which exhibits a strong hydrogen abstraction capability. Diphenylacetone (DPA) acts as a hydrogen donor, with hydrogen atoms from its hydroxyl (−OH) or methylene (−CH₂) groups abstracted by BP (T₁), generating benzophenone radicals (BP^•) and DPA radicals (DPA^•). These radicals initiate addition to the acrylic ester double bonds (CC) of PEGDA, forming PEGDA macroradicals that trigger chain polymerization and cross-linking into a three-dimensional network. Because each PEGDA molecule contains acrylic ester groups at both termini, covalent C–C bonds are formed between adjacent PEGDA chains, constructing a “PEG segment–crosslink node” network. − Polyethylene glycol (PEG) serves as the flexible backbone of PEGDA. ,

Mechanical characterization demonstrated that the elastic modulus of PEGDA hydrogels across concentration gradients exceeded that of normal arterial tissue and aneurysm walls while remaining within a range compatible with vascular mechanics. In particular, the 75 wt % formulation exhibited an elastic modulus of 23.89 MPa, substantially higher than that of the aneurysm wall (0.89 MPa), enabling effective resistance to systolic blood pressure-induced wall expansion (120 mmHg ≈ 0.016 MPa) and thereby reducing rupture risk. Meanwhile, the low swelling ratio minimizes the likelihood of chronic vascular compression caused by excessive hydrogel expansion.

Surface wettability and microstructural analyses further revealed that the smooth, dense morphology of the 75 wt % hydrogel corresponds to its low platelet adhesion rate. Although the 80 wt % formulation showed a marginally lower platelet adhesion rate, its water contact angle increased sharply to 57.59°, suggesting that excessive PEGDA concentrations alter surface energy and compromise hydrophilicity. Collectively, these findings indicate that the 75 wt % formulation achieves an optimal balance between surface hydrophilicity and structural uniformity, forming a stable hydration layer through high-density hydroxyl groups while avoiding abnormal surface energy associated with excessive cross-linking. This balance directly aligns with the clinical design criteria outlined in the Introduction.

The superhydrophilic surface constructed by PEG segments effectively reduces fibrinogen adsorption through steric hindrance, and antiplatelet adhesion performance improves with increasing PEGDA concentration, approaching that of the negative control. This nonthrombogenic interface overcomes the thrombosis-dependent limitations of conventional embolic materials, such as coils and flow-diverting devices, and provides a platelet-activation-free microenvironment conducive to aneurysm neck tissue repair. − Such a thrombus-independent embolization strategy distinguishes this hydrogel from existing approaches and establishes the material basis for subsequent biological closure. ,,

Consistent with this material design rationale, in vitro Transwell migration, scratch, and tube formation assays demonstrated that the 75 wt % hydrogel significantly enhances endothelial cell migration and vascular network formationprocesses that are essential for aneurysm neck endothelialization and closure. − Unlike conventional embolic materials that rely on passive thrombus formation, PEGDA actively induces vascular remodeling. Upregulation of claudin-1 (CLDN1), a core tight junction protein, enhances the stability of the CLDN1–CLDN12–ZO-1 complex and enables dynamic tight junction remodeling that preserves endothelial barrier integrity while permitting controlled junctional plasticity to support cell migration. − This “plastic tight junction” mechanism provides a biological explanation for the active aneurysm neck closure observed in this study.

In vivo, digital subtraction angiography (DSA) and histological analyses jointly confirmed the efficacy of the PEGDA hydrogel. Following treatment, the aneurysm cavity was completely nonvisualized while the parent artery remained patent. Apparent defects observed in anteroposterior projections can be attributed to projection overlap rather than true vascular stenosis. Masson’s trichrome staining revealed progressive reduction of the aneurysm neck septum at 7 days and complete closure by 28 days, accompanied by substantial collagen deposition. Immunofluorescence analysis of CD31 further confirmed endothelialization at the aneurysm neck. These in vivo observations are in strong agreement with the in vitro findings, validating the hydrogel’s dual functionality in mechanical embolization and biological repair.

To further elucidate the molecular mechanisms underlying aneurysm neck remodeling, transcriptomic analysis was performed using a parallel Sprague–Dawley rat aneurysm model. While the rabbit model was specifically employed to validate embolization efficacy, imaging outcomes, and histopathological remodeling due to its vascular anatomy and interventional compatibility, Sprague–Dawley rats offer well-established advantages for high-throughput transcriptomic analysis, including greater RNA yield, improved tissue homogeneity, and more consistent RNA quality control. This complementary experimental design enables robust mechanistic interrogation at the molecular level, while preserving translational relevance in large-animal embolization models.

To systematically explore the biological pathways involved in hydrogel-mediated aneurysm neck remodeling, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome pathway analyses were performed. − Transcriptomic profiling identified CLDN1 as a key differentially expressed gene (t test, p = 0.000073) with significant enrichment of the KEGG tight junction pathway (hsa04530). Protein–protein interaction modeling and structural prediction using HelixFold3 and PyMOL, combined with local frustration analysis, supported a central role for CLDN1 in coordinating endothelial junction dynamics. − The CLDN1–CLDN12 complex was predicted to anchor to the actin cytoskeleton via ZO-1, enhancing intercellular adhesion while facilitating endothelial migration through β-catenin signaling. ,

Notably, stable VEGF and CD31 protein expression in the absence of corresponding transcriptional upregulation suggests that the PEGDA hydrogel preserves endothelial integrity primarily through regulation of protein localization and junctional organization rather than excessive angiogenic activation. This observation is consistent with the preserved parent artery patency observed in vivo. Structural modeling further demonstrated that the extracellular β-sheet domain of CLDN1 (PDB ID: 6 × 8H) forms a hydrogen-bonding network with CLDN12 and interacts with the PDZ domain of ZO-1, thereby ensuring tight junction stability under mechanical stress. ,

Taken together, these findings establish a clear mechanistic link among material stiffness, tight junction regulation, and endothelial behavior. This molecular-level “mechanical–biological signal” conversion represents a defining advantage of the PEGDA hydrogel over traditional embolic materials, enabling a paradigm shift from passive physical occlusion to active, biologically driven aneurysm neck closure.

Despite these encouraging results, several limitations of this study should be acknowledged. Although the long-term degradation behavior of PEGDA hydrogels under physiological conditions was systematically evaluated, the potential influence of prolonged ambient light exposure on the material stability was not independently assessed. Given that the hydrogel is designed for intravascular deployment and cured in situ under controlled ultraviolet irradiation, future studies will investigate the effects of extended ambient light exposure on physicochemical stability and mechanical integrity to comprehensively address long-term translational safety.

In addition, although the BP/DPA photoinitiator concentration (0.5 wt %) was selected based on preliminary feasibility experiments to ensure rapid and reproducible in situ curing within 60 s while maintaining excellent cytocompatibility and hemocompatibility, a systematic parametric optimization of photoinitiator concentration was not performed in the present study. This variable was intentionally fixed to preserve mechanistic clarity and avoid confounding effects while elucidating the PEGDA concentration-dependent structure–function–mechanism relationship. Future investigations will include comprehensive head-to-head evaluations across photoinitiator concentrations, focusing on curing kinetics, residual initiator content, and biological safety profiles.

In addition, direct head-to-head comparisons with clinically used embolic materials (e.g., detachable coils or liquid embolics, such as Onyx) were not included. This study was intentionally designed to elucidate the intrinsic structure–function–mechanism relationship of the PEGDA/PEG hydrogel and to validate its fundamentally different “biological closure” paradigm rather than conventional thrombosis-dependent physical occlusion. Future work will incorporate comparative evaluations with standard embolic materials to further assess embolization efficacy, procedural safety, and long-term outcomes under clinically relevant conditions.

5. Conclusions

To the best of our knowledge, this study demonstrates a previously unreported integration of PEGDA concentration-gradient design with CLDN1-mediated tight junction regulation, establishing a coherent translational framework linking material mechanical parameters, cellular behavior, and molecular mechanisms. The optimized 75 wt % PEGDA hydrogel upregulates CLDN1 expression, providing mechanical support to the aneurysm cavity while actively guiding endothelial cell migration and collagen deposition, thereby enabling effective biological closure of the aneurysm neck and overcoming the limitations of conventional embolic materials that lack biological regulatory capability.

This dual-functional design strategy, combining mechanical matching with biological induction, introduces a new paradigm for vascular embolization materials. By shifting the therapeutic concept from passive lesion filling toward active biological repair, this approach holds promise for reducing aneurysm recurrence, preserving parent artery integrity, and advancing the next generation of bioactive endovascular embolization therapies.

The data supporting the findings of this study are available within the article and its Supporting Information.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c09657.

• Aneurysm model (Video S1) (MP4)

• Anteroposterior view of DSA 1 (Video S2) (MP4)

• Anteroposterior view of DSA 2 (Video S3) (MP4)

• Lateral view of DSA 1 (Video S4) (MP4)

• Lateral view of DSA 2 (Video S5) (MP4)

• Anteroposterior view of the embolized aneurysm 1 (Video S6) (MP4)

• Anteroposterior view of the embolized aneurysm 2 (Video S7) (MP4)

• Lateral view of the embolized aneurysm 1 (Video S8) (MP4)

• Lateral view of the embolized aneurysm 2 (Video S9) (MP4)

• Additional experimental details, including in vitro degradation behavior of PEGDA/PEG hydrogels, extracorporeal embolization in 3D-printed aneurysm models, in vivo embolization procedures, and supplementary figures (Figures S1–S8) (PDF)

⊥.

X.W. and J.T. contributed equally to this work. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.