Development of an Injectable, ECM-Derivative Embolic for the Treatment of Cerebral Saccular Aneurysms

Cerebral aneurysms are a source of neurological morbidity and mortality, most often as a result of rupture. The most common approach for treating aneurysms involves endovascular embolization using nonbiodegradable medical devices, such as platinum coils. However, the need for retreatment due to the recanalization of coil-treated aneurysms highlights the importance of exploring alternative solutions. In this study, we propose an injectable extracellular matrix-derived embolic formed in situ by Michael addition of gelatin-thiol (Gel-SH) and hyaluronic acid vinyl sulfone (HA-VS) that may be delivered with a therapeutic agent (here, RADA-SP) to fill and remodel aneurysmal tissue without leaving behind permanent foreign bodies. The injectable embolic material demonstrated rapid gelation under physiological conditions, forming a highly porous structure and allowing for cellular infiltration. The injectable embolic exhibited thrombogenic behavior in vitro that was comparable to that of alginate injectables. Furthermore, in vivo studies in a murine carotid aneurysm model demonstrated the successful embolization of a saccular aneurysm and extensive cellular infiltration both with and without RADA-SP at 3 weeks, with some evidence of increased vascular or fibrosis markers with RADA-SP incorporation. The results indicate that the developed embolic has inherent potential for acutely filling cerebrovascular aneurysms and encouraging the cellular infiltration that would be necessary for stable, chronic remodeling.


INTRODUCTION
Cerebral aneurysms are thought to develop at points of hemodynamic stress in intracranial arteries and are characterized by their bulging, saccular structure. 1,2Cerebral aneurysms affect approximately 2−3% of the adult population, with each aneurysm carrying an average annual rupture risk of 1% (although certain aneurysms can have a rupture risk as high as 50%) and a mortality rate of 50% when rupture occurs. 3ecently, the use of nondegradable platinum coils to embolize cerebral aneurysms has gained widespread acceptance for their treatment. 4Additionally, flow-diverting-device (FDD) stents with high metal coverage and low porosity have been introduced.While FDD has greater rates of aneurysm obliteration than intrasaccular therapy, careful patient selection is necessary for typical side-wall aneurysms, and extended antiplatelet therapy is required. 5espite its advantages, endovascular therapy using coiling is not without limitations and complications.These include wellknown issues such as coil migration, coil compaction, and recanalization. 6Recanalization has been observed in approximately 17% of patients who undergo bare coiling for aneurysms, necessitating retreatment in 10.3% of cases. 7,8he occurrence of recanalization is more common in the cases of large, giant, or wide-neck aneurysms.
As an alternative to metallic coiling, researchers have investigated injectable embolic agents to fill vascular defects including arteriovenous malformations and aneurysms. 9These include Onyx, n-butyl cyanoacrylate, and Embogel. 5Here, the target site is filled by administering a controlled amount of the embolic agent to occlude the aneurysm.However, Onyx and nbutyl cyanoacrylate have limitations primarily as a result of uncontrolled reflux into the parent artery, creating a higher risk of stroke. 10They are thus not used in the treatment of typical saccular aneurysms.EmboGel is an alginate-based embolic biomaterial that does not require organic solvents and achieves solidification by the introduction of calcium chloride.Although EmboGel is biocompatible, the resulting solidified gel with calcium chloride may be soft and fragile, leading to the potential for rapid degradation and washout of the parent artery.
Recently, extensive research has been conducted on gelatin for the development of injectables aimed at treating hemorrhage.Gelatin has been explored as a shear-thinning, 11−15 semisolid, 16,17 and cross-linkable hydrogel. 18,19hese studies have been motivated by the various characteristics of gelatin, such as its ability to trigger clot formation, low immunogenicity, and support for cell attachment and proliferation.Among these approaches, injectables composed of gelatin and nanoclay have shown promising results in hemorrhage control and embolization. 16,17However, it has been reported that silicate nanoplatelets can induce cytotoxicity through the generation of reactive oxygen species, damage to cell membrane interactions, and intracellular interaction mechanisms. 20n light of these concerns, the development of extracellular matrix (ECM)-derived cross-linkable hydrogels, such as those based on modified gelatin and hyaluronic acid, has garnered attention as injectable gels due to their cytocompatibility, biological functionality, and slow degradability relative to physically cross-linked gels. 18,19,21Recently, combined gelatin and hyaluronic acid hydrogels have been studied as injectable, cross-linking embolic agents. 19However, chemically crosslinkable and injectable ECM-mimicking gels of gelatin and hyaluronic acid have not been explored as a means to fill saccular aneurysms by embolization and to facilitate healing.
For achieving long-lasting exclusion and stabilization of the vessel wall without the placement of permanent foreign bodies, it would be ideal if an ECM-derived embolic was employed that initially facilitates thrombogenesis and subsequent cell attachment and retention, followed by gradual degradation and resorption.To meet those requirements, in this study, we developed a modified form of gelatin and hyaluronic acid to create an injectable and cross-linkable ECM-derived embolic agent.Our goal was to deliver the gel, with or without an additional therapeutic agent, to promote a healing response and facilitate tissue-in-growth to the aneurysm sac.As a model therapeutic agent, the self-assembling peptide (RADA) was conjugated with neuropeptide substance P (SP) (RADA-SP, Ac-RARADADARARADADAGGRPKPQQFFGLM-NH 2 ).RADA-SP was selected due to its reported ability to enhance the recruitment of endogenous cells into implanted scaffolds and expedite the process of wound healing. 22Also, the effectiveness of RADA-SP in recruiting vascular smooth muscle cells (vSMCs) is well-proven in published papers. 22,23The addition of RADA-SP to the injectable embolic aimed to foster increased in-growth of vSMCs to the target aneurysm, thereby potentially enhancing the elasticity of the newly formed tissue within the sac.The potential for these materials to fill and encourage cell ingrowth in cerebral saccular aneurysms was assessed in vitro and in vivo using a murine aneurysm model.

Synthesis and Characterization of Thiolated Gelatin (Gel-SH) and Vinyl Sulfonated Hyaluronic Acid (HA-VS).
Thiolated gelatin (Gel-SH) and vinyl sulfonated hyaluronic acid (HA-VS) were prepared using the methodology described previously. 21Briefly, Gel-SH was synthesized from gelatin and 3,3′dithiobis(propionic hydrazide) (DTPH) which is a reaction product of 3,3′-dithiopropionic acid dimethyl ester with hydrazine hydrate.A solution of 1.0 g of gelatin in 100 mL of distilled water was prepared, and 0.5 g of DTPH was added while stirring.The pH of the solution was adjusted to 4.75 by adding 1 M HCl, followed by the addition of 0.3 g of EDC.After 2 h, the reaction was stopped by adding 1 M NaOH solution to raise the pH to 7.0, then, 2.2 g of DTT was slowly added to the reactor, and the pH was increased to 8.5 using 1 M NaOH.The reaction proceeded for 24 h at room temperature.The pH was adjusted to 3.5 using 1 M HCl, and the acidified product was purified using dialysis membrane tubing (MWCO 12 000) in 0.03 M HCl.The resulting fine powder was obtained through lyophilization.
Hyaluronic acid (HA) and divinyl sulfone (DVS) were used to synthesize HA-VS.A solution containing 0.5 g of HA in 100 mL of 0.02 M NaOH was prepared.Then 1.3215 g of DVS was added to the reactor in an ice bath.The reaction was allowed to proceed for 30 min and then terminated by adjusting the pH to 5.0 using 1 M HCl solution.The product was purified using dialysis membrane tubing (MWCO 12,000) in distilled water at pH 5.3.The fine powder was obtained after lyophilization.The synthesized Gel-SH and HA-VS chemical structures were confirmed by 1 H-nuclear magnetic resonance ( 1 H NMR).
For the inverted vial gelation test, solutions of HA-VS and Gel-SH were prepared in 20 mM HEPES at concentrations of 1% (w/v) and 2% (w/v), respectively.A 0.5 mL HA-VS solution and a 0.5 mL Gel-SH solution were preheated at 37 °C and injected into a 3.7 mL glass vial using a 25 G needle.The vial containing the mixture was then incubated at 37 °C for 30 s, and the sol−gel phase transition was evaluated using the vial inverting method.The internal structure of the formed gel was examined by cross-sectional images of lyophilized scaffolds using a scanning electron microscope (SEM, JSM 6335F, JEOL, Tokyo, Japan).
A rheometer (AR2000EX, TA Instruments, New Castle, DE, USA) was used for evaluation of the rheological change at 37 °C during the gelation.A 0.5 mL portion of 1% (w/v) HA-VS solution and 0.5 mL of 2% (w/v) Gel-SH solution were injected into the center of the Peltier plate at the same time using separated nozzles.The shear storage modulus (G′) and loss modulus (G″) were measured at a strain of 5%.
The swelling ratio and degradation profiles of HA-VS/Gel-SH were evaluated in vitro.Solutions of 1% (w/v) HA-VS and 2% (w/v) Gel-SH were prepared in 20 mM HEPES, and 75 μL of each solution was injected into each well of a 96 well plate and then mixed and kept at 37 °C for 1 h.The HA-VS/Gel-SH gels were freeze-dried, washed with distilled (DI) water, and freeze-dried before use.For the swelling ratio, the mass of dried samples was measured and then immersed in Dulbecco's phosphate-buffered saline (DPBS) at 37 °C for 3 h.The mass of swollen samples was measured for the calculation of the swelling ratio (S) using the equation , where Ms and Md are the mass of swollen and dried samples, respectively (n = 5).
The degradation profile was evaluated by a change in mass for 4 weeks.The mass of samples (W 0 ) was measured and then immersed in DPBS for 4 weeks.At time points of 1, 2, 3, and 4 weeks, the samples were pulled out, washed with DPBS and DI water three times for each, and freeze-dried before checking the mass (W 1 ).The degradation of samples was evaluated by an equation change of mass (%) = ((W 1 − W 0 )/W 0 ) × 100 (n = 5).
2.3.HA-VS/Gel-SH Gelation with Iohexol.HA-VS was dissolved at 1% (w/v), and Gel-SH was dissolved at 2% (w/v) in 20 mM HEPES separately after being sterilized under UV for 10 min.Iohexol was then dissolved in the Gel-SH solution with various concentrations at 5%, 10%, or 15% (w/v, mass of iohexol/a total volume of HA-VS and Gel-SH mixture solution).75 μL of each solution HA-VS and Gel-SH+iohexol were injected into each 96 well at 37 °C and then kept in an incubator for 30 min followed by lyophilization.X-ray images of the resulting freeze-dried hydrogels were captured by using an OEC 9800 Plus X-ray machine.In addition, cross-sectional images of the freeze-dried hydrogels were obtained using SEM.Each sample was immersed in liquid nitrogen and sliced using a surgical-grade blade to expose the cross section.Before SEM imaging, the sliced samples were mounted on a sample holder and coated with a 5 nm layer of Pt.The average diameters of the pores were measured from the SEM images using a free photo editing software program Fiji (https://imagej.net/software/fiji/).
2.4.Blood-Contacting Test of HA-VS/Gel-SH with Iohexol.Fresh whole ovine blood was obtained for this study through jugular venipuncture, using a sodium citrate-containing tube.The collection and handling of the blood samples followed the guidelines set by the National Institute of Health (NIH) for the care and use of laboratory animals.All animal procedures conducted as part of this study were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Pittsburgh.
To investigate the gelation behavior of HA-VS/Gel-SH+iohexol in blood, solutions of 1% HA-VS and 2% Gel-SH+iohexol (10%) were prepared following the procedure described in section 2.3.Subsequently, 75 μL of each solution was injected into 50 μL of fresh ovine blood in individual wells of a 96-well plate, maintained at 37 °C.As a comparison, 0.2 M CaCl 2 (20v/v%, relative to the total blood volume) or 2% (w/v) alginic acid in PBS + 0.2 M CaCl 2 (20v/v %, relative to the total blood volume) were injected into 50 μL of fresh blood.After injection, the plate was placed in an incubator for 30 min.The samples were then washed with DPBS five times to remove any unattached blood cells, followed by fixation with 2.5% glutaraldehyde at room temperature for 2 h.The samples were subjected to lyophilization after being washed with distilled water.The internal structure of the freeze-dried hydrogels was examined by using cross-sectional SEM images.The number of entrapped or encapsulated red blood cells was counted from the SEM images.
To explore the interaction between blood cells and the scaffolds following gelation, lyophilized HA-VS/Gel-SH+iohexol (10% or 15% of the total mixture volume) and 2% alginic acid+0.2M CaCl 2 scaffolds were exposed to fresh ovine blood, and the results were observed by SEM imaging.The lyophilized hydrogels were prepared as described previously and immersed in 5 mL of fresh ovine blood.The blood-containing tubes with the samples added were maintained at 37 °C with gentle rocking for 3 h.Following blood contact, the samples were washed with DPBS five times to remove any unattached blood cells.Subsequently, the samples were fixed with 2.5% glutaraldehyde at room temperature for 2 h, followed by washing with distilled water.The samples were then lyophilized and sectioned by using a surgical blade in liquid nitrogen to examine their surface and cross-sectional morphologies through SEM imaging.The number of entrapped or encapsulated red blood cells was counted from the SEM images.
2.5.rSMC Penetration and Proliferation in HA-VS/Gel-SH with Iohexol.The penetration and proliferation of rat aortic smooth muscle cells (rSMCs) within the HA-VS/Gel-SH+iohexol hydrogel were assessed by the in vitro seeding of rSMCs followed by DAPI imaging.Lyophilized HA-VS/Gel-SH+iohexol (10% or 15% of the total mixture volume) and 2% alginic acid+0.2M CaCl 2 scaffolds were prepared following the aforementioned method.The samples were sterilized under UV light for 10 min and placed in separate wells of a 96-well plate.rSMCs were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (HI FBS) and 1% penicillin/streptomycin at 37 °C and 5% CO 2 .A volume of 150 μL of rSMCs at a concentration of 5 × 10 6 cells/mL was added to each well containing the sample.The culture medium was refreshed daily.After 1 and 7 days, the samples were transferred to fresh wells and washed with DPBS.Following fixation with 2.5% glutaraldehyde, the samples were embedded in an optimal cutting temperature (OCT) compound and then sectioned using a microtome-cryostat (CryoStar NX50, Thermo Scientific).The mounted slices were washed with PBS and stained with 4′,6diamidino-2-phenylindole (DAPI).The images were captured using a Zeiss fluorescence microscope.
2.6.In Vitro Cell Response to HA-VS/Gel-SH with Iohexol and RADA-SP.The gelation of HA-VS/Gel-SH with RADA-SP was tested in vitro, and their morphology was observed by SEM imaging.To prepare the gel incorporating RADA-SP, the following steps were taken: first, HA-VS was dissolved at a concentration of 1% (w/v) and Gel-SH was dissolved at a concentration of 2% (w/v) in 1 mL of 20 mM HEPES solution.Both solutions were sterilized under UV for 10 min.Then, iohexol was added to the Gel-SH solution at a concentration of 20% (w/v), and 2 mg of either RADA or RADA-SP was dissolved in 1 mL of the Gel-SH+iohexol solution.Next, 75 μL of each solution (HA-VS and Gel-SH+iohexol+RADA or RADA-SP) were separately injected into individual 96-well plates at a temperature of 37 °C.The plates were incubated for 30 min in an incubator and subsequently subjected to lyophilization.Finally, the cross-sectional morphologies of the resulting gels were observed by using SEM imaging.The average diameters of the pores were measured from the SEM images using a Fiji.
The cell responses to the RADA-SP-incorporated injectable embolic were evaluated and compared with those of RADA and RADA-SP free embolic and RADA-incorporated embolic in vitro.For the evaluation of the effect of RADA-SP, rat vascular endothelial cells (rECs, from aortic tissue) or rSMCs (150 μL of 1 × 10 7 cells/mL) were seeded on the lyophilized gels in each well of a 96-well plate and incubated for 7 d.The cell medium was changed daily.After incubation, samples were washed in PBS and fixed in 2% paraformaldehyde for 1 h at room temperature, followed by permeabilization and blocking in PBS containing 0.2% Triton X-100, 3% bovine serum albumin, and 10% goat serum for 2 h at room temperature.After blocking, samples were incubated with primary antibody (ab6994, rabbit polyclonal to Von Willebrand Factor, Abcam, Cambridge, UK, 1:300) or (ab5694, rabbit polyclonal to alpha-smooth muscle actin, Abcam, 1:100) overnight at 4 °C, then incubated with the solution containing secondary antibody (A21442, Alexa Fluor 594 chicken antirabbit IgG H&L, Invitrogen, Eugene, Oregon, USA, 1:500) or (A11034, Alexa Fluor 488 goat antirabbit IgG H&L, Invitrogen, 1:250) and DAPI for 2 h at room temperature.Primary and secondary antibodies were diluted in PBS containing 5% goat serum.Stained samples were kept in PBS until imaging.Fluorescent images were acquired using either a Zeiss fluorescence microscope (for VWF) or a Nikon A1 confocal laser scanning microscope (for αSMA).Image analysis was performed using either Zen lite (version 2.3) or Fiji.

In Vivo Studies Using a Mouse Aneurysm Model.
All animal experimentation conducted in this study adhered to a protocol approved by the IACUC at the University of Pittsburgh, in accordance with the guidelines set by the NIH.Given that cerebral aneurysm formation and rupture affect women at a 60% higher rate than men, 24,25 we used C57BL/6 female mice (Charles River) to establish murine carotid aneurysm.Briefly, mice were anesthetized with an isoflurane and oxygen gas mixture, and then microsurgical exposure of the right common carotid artery (RCCA) was performed under sterile conditions using an operating microscope.The RCCA was exposed, and then a latex cuff was placed around the vessel.The RCCA was bathed with porcine pancreatic elastase solution (10 U/1 mL of PBS) for 20 min, and the vessel was occluded distally.A saccular aneurysm formed over the next 3 weeks.For control purposes, sham-operated animals underwent microsurgical exposure to the RCCA, which was then bathed with phosphate-buffered saline rather than elastase solution and no distal occlusion of the artery.A Biomacromolecules second control group underwent the same procedure but received elastase but no follow-up treatment with an embolic.
Two different components HA-VS/Gel-SH with 10% (w/v) iohexol and HA-VS/Gel-SH with 10% iohexol and 2 mg of RADA-SP were tested in vivo using the murine model (n = 7 for each group).Here, a temporary aneurysm clip with a relatively gentle closing force of 69 gf (0.68N; MIZUHO Sugita Titanium temporary miniclip II) was placed at the neck of the aneurysm to prevent retrograde embolization.A 30G needle was then used to inject ∼20 μL of the embolic agent into the aneurysm.The gel mixture was allowed to cross-link for 5 min.The temporary clip was then removed, and the aneurysm was allowed to heal for 3 weeks, after which the carotid arteries that had been embolized or underwent a sham procedure were collected for proinflammatory cytokine assay, histologic evaluation, and immunofluorescence analysis.
2.8.Proinflammatory Cytokine Assay.For proinflammatory cytokine expression, to extract proteins from these arteries, a Cell Lysis Buffer from RayBiotech was utilized for their lysis.Subsequently, the isolated protein was applied onto standard mouse cytokine antibody array membranes (RayBiotech's Mouse Cytokine Array C1000), which enabled the analysis of 96 different cytokines.The membranes underwent steps involving incubating them with the protein sample, removing any unbound proteins through washing and treating them with a detection reagent.The emitted signals resulting from the bound cytokines were visualized by capturing film images.To assess the intensity of these signals, densitometry analysis was performed using ImageJ.Morpheus, software for matrix visualization and analysis (https://software.broadinstitute.org/morpheus)was employed to visualize the data.Additionally, data clustering was carried out by applying the hierarchical clustering algorithm within Morpheus, using the Spearman rank correlation similarity metric (n = 3 animals in each group).

Statistical Analysis.
The n-value refers to the number of replicates for each test.Data are presented as mean ± standard deviation (SD).One-way ANOVA along with Tukey's test was performed, and p < 0.05 was considered statistically significant.

Synthesis and Characterization of HA-VS and
Gel-SH and their Gelation with Iohexol.HA-VS and Gel-SH were successfully synthesized with the expectation of rapid gelation by the Michael Addition reaction when they were mixed at physiological temperature (37 °C) and pH (∼7.4) (Figure 1A).The chemical structure of synthesized HA-VS and Gel-SH were confirmed by 1 H NMR spectra, which showed new methylene sulfone (−CH 2 CH 2 SO 2 ) peaks at 6.25, 6.35, and 6.80 for HA-VS and new side chain methylene (−CH 2 CH 2 SH) peaks at 2.54 and 2.68 ppm for Gel-SH (Figure 1B). 21The conversion rate of the repeat unit of HA to HA-VS is 106.5% and that of Gel-SH is 48.1%; therefore, the substitution degree of VS and SH is 106.5% and 48.1%, respectively.The ratio between VS and SH was 1:0.45.
The results of the rheological change and reaction time of the mixture of 1% HA-VS and 2% Gel-SH showed a fast storage modulus increase for 5 min to 88 ± 1 Pa from 49 ± 1 Pa, and then reached 100 Pa within 10 min.The initial storage modulus may be influenced by the premixing of 1% HA-VS and 2% Gel-SH solutions on the Peltier plate before measurement.
In consideration of the future potential for angiographyguided transcatheter injection, the FDA-approved watersoluble contrast agent iohexol (Omnipaque) was dissolved in a 2% Gel-SH solution at concentrations of 5%, 10%, or 15% (w/v, mass of iohexol/a total volume of HA-VS and Gel-SH mixture solution).Following the lyophilization process, the resulting chemically cross-linked scaffolds containing iohexol were examined using X-ray and SEM imaging (Supplementary Figure 2).It was observed that all scaffolds with different concentrations of iohexol exhibited radiopacity, with higher concentrations of iohexol yielding greater radiopacity.The SEM images showed that all scaffolds, irrespective of iohexol concentration, displayed porous cross sections.The average pore size (diameter, μm) measured for the iohexol concentration of 5, 10, and 15% was 43 ± 14, 49 ± 12, and 47 ± 11, respectively.However, the scaffolds incorporating 10% and 15% iohexol were deemed better than scaffolds with 5% iohexol considering clear visualization under an X-ray machine.
3.2.In Vitro Blood-Contacting of HA-VS/Gel-SH with Iohexol.SEM images of the cross sections of gels injected into ovine blood are depicted in Figure 2. The HA-VS/Gel-SH gels revealed evident entrapment and encapsulation of blood cells (including red blood cells and platelets) on the porous walls of the gel.In contrast, SEM images of gels formed using 0.2 M CaCl 2 and 2% alginic acid + 0.2 M CaCl 2 displayed a less obvious blood cell entrapment in the cross section compared to HA-VS/Gel-SH gels.The number of encapsulated red blood cells was 0, 8 ± 3, 100, and 63 ± 10% for 20% 0.2 M CaCl 2 ,
From the rocking test with fresh ovine blood, it becomes evident that blood cells were deposited on the surface and penetrated the inner pores of the lyophilized scaffolds (which were formed by the chemically cross-linked HA-VS/Gel-SH +iohexol gels) (Figure 3).The lyophilized scaffold allowed the penetration of blood cells and supported the formation of blood clots within the scaffold.The number of encapsulated red blood cells was 100, 79 ± 6, 67 ± 8, and 72 ± 5% for surface and cross section of HA-VS/Gel-SH gel with 10% (w/ v) iohexol, and surface and cross section of HA-VS/Gel-SH gel with 15% (w/v) iohexol, respectively, when normalized to the surface of HA-VS/Gel-SH gel with 10% (w/v) iohexol as 100%.On the other hand, the scaffolds created using 2% alginic acid and 0.2 M CaCl 2 underwent the same procedure of being in contact with excess blood with gentle rocking for 3 h.However, these scaffolds were fragile and dispersed in the blood, leading to an inability to collect SEM images depicting the blood-contacted 2% alginic acid + 0.2 M CaCl 2 scaffolds.

In Vitro Cell
Response to the HA-VS/Gel-SH.The infiltration and proliferation of cells within the HA-VS/Gel-SH +iohexol gels were assessed for 7 days using rSMCs.The number of cells in the cross section of the gels was estimated by fluorescent images with DAPI staining and the intensity of the DAPI (Figure 3).It was observed that the HA-VS/Gel-SH gel containing 10% iohexol exhibited a significantly higher DAPI intensity on both day 1 and day 7 compared to the HA-VS/Gel-SH gel with 15% iohexol.Furthermore, the DAPI intensity on day 7 was significantly greater than on day 1 for both the 10% and 15% iohexol-containing HA-VS/Gel-SH gels.2% alginic acid + 20% 0.2 M CaCl 2 were tested as well, but data could not be collected because of degradation and the fragility of fragments.
The HA-VS/Gel-SH+iohexol gels containing RADA or RADA-SP exhibited porous cross-sectional images (Figure 4).The average pore size (diameter, μm) measured for the peptide-free, with RADA, or with RADA-SP was 43 ± 14, 35 ± 9, and 29 ± 7, respectively.Among the different types of prepared scaffolds (peptide-free, with RADA, or with RADA-SP) seeded with rECs or rSMCs, the gel incorporating RADA-SP demonstrated significantly higher expression of both von Willebrand factor (vWF) and alpha-smooth muscle actin (αSMA) compared to the peptide-free and with RADA gels.In the case of rEC seeding, the DAPI intensity in the cross section of the RADA-SP added gel was significantly lower, but the ratio of vWF/DAPI was significantly higher compared to other gel compositions (Supplementary Figure 5).When rSMCs were seeded, the gel incorporating RADA-SP displayed significantly higher DAPI intensity, αSMA expression, and αSMA/DAPI ratio compared to the other gel compositions (Figure 6)

In Vivo Injection of HA-VS/Gel-SH with Iohexol and RADA-SP Using Murine Carotid Aneurysm Model.
The murine carotid aneurysm model was used for 3 weeks of in vivo assessment of the injectable embolics (Figure 4A−C).From the proinflammatory cytokine array (Figure 4D), there were no significant differences between the two embolic groups in the expression of major cytokines IL-1β, IL-6, IL-17A, TNF-α, MMP-2, and MMP-3, which are mediators implicated in cerebral aneurysm formation and rupture.
Histologic evaluation of H&E, Trichrome, and PicroSirrus-Red showed successful embolization of the aneurysms by the injectable embolic (Figure 5).All of the ECM-mimicking injectable embolic with and without RADA-SP embolized the sac of the aneurysm and allowed cell infiltration.
The delivered RADA-SP was released within 3 weeks after the injection, and there was no remaining SP in the tissue.

DISCUSSION
Desirable properties for aneurysm embolization materials would include the promotion of clot formation to effectively exclude the aneurysm sac from the parental artery after embolization and subsequently support the growth of new tissue.Ultimately, the material should be remodeled by the infiltrating cells, allowing for permanent embolization without the presence of a foreign body.This would provide an optimal solution for treating intracranial aneurysms while minimizing potential complications and allowing for easier retreatment, if necessary.
HA and collagen are abundant ECM components broadly found in various tissues.Their cytocompatibility, support for cell adhesion, physical properties, and amenability to Biomacromolecules purification and modification have made derivatives of HA and collagen common building blocks for biomaterial design in the fields of tissue engineering and drug delivery systems.The use of HA and collagen derivatives (e.g., gelatin) in tissue engineering has shown promising results, as they can mimic critical aspects of the natural ECM and provide structural support.Moreover, their biodegradability ensures that these materials are gradually broken down in the body over time in native remodeling processes without leaving behind permanent foreign substances.Their versatility allows for various modifications and functionalization to tailor their properties for specific applications, such as controlled drug delivery systems.) murine common carotid artery is exposed and a latex cuff is placed underneath; 2) most distal portion of the artery is ligated using 8−0 nylon suture to create a stump; 3) the artery is incubated with elastase for 20 min; 4) at the end of the incubation period an aneurysm has formed, the latex cuff is removed and the neck incision is closed; 5) 3 weeks later the carotid aneurysm is re-exposed and a temporary clip is placed around the proximal part of the artery-aneurysm complex; 6) embolic is injected through a 30G needle; 7) waiting for 5 min; 8) the temporary clip is removed; and (C) intraoperative photographs of the injection.(D) Pro-inflammatory cytokine production (n = 3) after the injection of (a) HA-VS/Gel-SH with 10% iohexol and (b) HA-VS/Gel-SH with 10% iohexol and RADA-SP.
In this report, we applied a chemically cross-linked hydrogel by synthesizing vinyl sulfonated hyaluronic acid and thiolated gelatin (collagen hydrolyzate functionalized with thiol groups).These components form a hydrogel through the Michael Addition reaction when mixed under physiological conditions.The HA-VS and Gel-SH can be individually delivered to an intracranial saccular aneurysm through two separate channels using a microcatheter connected to a dual-barrel syringe.The delivered two solutions would mix in the sac of the aneurysm, forming a cross-linked ECM-derived matrix while initiating and propagating local thrombus formation. 14,28Following in situ gelation, blood can infiltrate the porous embolic material and adhere to its surfaces.This phenomenon was verified through in vitro blood contacting tests conducted using the cross-linked scaffolds.−31 During 3 h of blood contact, lyophilized hydrogels consisting of 2% alginic acid + 0.2 M CaCl 2 were observed to degrade in the blood.In contrast, the lyophilized HA-VS/Gel-SH + iohexol gels retained their shape while clot formation occurred.This different behavior between the 2% alginic acid + 0.2 M CaCl 2 hydrogels and the HA-VS/Gel-SH +iohexol hydrogels can be attributed to differences in the cross-linking mechanism (ionic complexation for alginic acid with calcium ions versus chemical cross-linking for HA-VS/ Gel-SH) and the extent of clot formation.
Based on the in vitro data regarding the penetration and proliferation of rSMCs, it was observed that HA-VS/Gel-SH with 10% iohexol had a higher number of cells within the cross section on both days 1 and 7, compared to HA-VS/Gel-SH with 15% iohexol.Iohexol exhibits only low systemic toxicity due to its low chemotoxicity and osmolality.However, a higher iohexol concentration might directly impact cell infiltration and proliferation within the hydrogel by changing the surface chemistry.HA-VS/Gel-SH with 10% iohexol was chosen for the next stage of the experiments.
In endovascular transcatheter embolization, both controlled localized thrombus formation and tissue ingrowth are key factors for successful aneurysm occlusion and healing.In addition to the thrombotic and cell support activity of gelatin, RADA-SP was loaded into the ECM-mimicking injectable embolic with the potential for stimulating improved healing.RADA-SP is a self-assembled peptide conjugated with SP.SP is a neuropeptide released from the peripheral terminals of sensory nerve fibers, and SP functions as both a neurotransmitter and a hormone.Studies have reported that RADA-SP has the ability to enhance the recruitment of endogenous cells into implanted scaffolds and expedite the process of wound healing. 22While the effects of RADA-SP on wound healing, including ischemia, bone tissue engineering, skin regeneration, and vascular regeneration, have been extensively studied and established, 22,23,32 its impact on aneurysm healing has not been investigated to our knowledge.
Self-assembling peptides, such as RADA, possess the ability to form stable β-sheet structures and undergo self-assembly into nanofibers (ranging from 5 to 10 nm) through van der Waals and ionic interactions.RADA-SP has previously exhibited a sustained effect in contrast to free SP by impeding the rapid release of water-soluble SP from the injection site and protecting it from enzymatic digestion and deactivation. 22,23,32n an example utilizing a mouse hind limb ischemia model, self-assembled RADA-SP maintained a higher concentration of SP at the target site compared to free SP for 28 days.Resulting in an accelerated wound healing process, attributed to enhanced recruitment of mesenchymal stem cells into the ischemic region. 22The sustained release of SP is thus believed to play a vital role in promoting the continuous recruitment of cells and creating a favorable microenvironment for healing.In this regard, it was hypothesized that delivering RADA-SP would aid in healing by facilitating cell recruitment and tissue in-growth into the target aneurysm sac.
For the ECM-mimicking injectable embolic incorporating RADA-SP, given the intricate nature of the physical (e.g., selfassembly of RADA-SP) and chemical (e.g., Michael Addition of HA-VS with Gel-SH) gelation mechanisms involved, it was considered whether HA-VS/Gel-SH with 10% iohexol, in combination with RADA or RADA-SP, displayed a porous interconnected morphology that facilitates cell penetration and metabolite transport.The results demonstrated that the prepared gels containing RADA or RADA-SP exhibited a porous structure.The inclusion of RADA-SP likely introduced variations in surface chemistry, pore size, and structure, which may account for patterns of different cell infiltration into the ECM-mimicking gel. 33Furthermore, the gel system incorporating RADA-SP demonstrated notably elevated expression of VWF and αSMA in comparison to other formulations confirmed in vitro.Consequently, the HA-VS/Gel-SH gel with RADA-SP exhibited the potential ability to recruit vSMCs and stimulate angiogenesis.This activity may hold potential for enhancing the healing process of an embolized aneurysm.IL-1β, TNF-α, and MMPs are mediators that play significant roles in the formation and rupture of cerebral aneurysms. 34−37 MMPs contribute to phenotypic modulation and SMC loss, macrophage activity, M1/M2 imbalance, leukocyte infiltration, vascular remodeling, and cell death. 38,39The expression levels of these mediators were evaluated using a cytokine expression array 3 weeks after implantation, and no significant differences were observed on both of the tested compositions of the injectable embolic.Based on the observation, it appears that integrating RADA-SP into the injectable embolic does not adversely impact cytokine expression associated with aneurysm development and rupture; however, a beneficial effect is also not apparent.
Histological analysis by H&E, trichrome, and picrosirius-red staining revealed that both injectable embolic with different components effectively filled the aneurysm sac and stimulated tissue ingrowth within 3 weeks.When RADA-SP was incorporated into the injectable embolic, potential capillary in-growth was observed.This angiogenesis effect was intended as intracranial aneurysm therapy but their effect on recanalization needs to be evaluated in longer-term animal studies.Immunofluorescent staining with biomarkers revealed that the RADA-SP incorporated injectable embolic showed significantly higher SMC infiltration reflected by αSMA.This result is along the lines of the in vitro αSMA expression data where the RADA-SP showed an effect on SMC proliferation.
Nevertheless, given the successful embolization involving macrophages and fibroblast infiltration with the ECM-derived injectable embolic lacking RADA-SP, it is plausible that the injectable embolic without RADA-SP could exhibit superior efficacy in saccular aneurysm embolization and subsequent healing.However, it is worth noting that the hydrogel we developed is capable of incorporating other potential prohealing molecules (e.g., TGF-β, MCP-1) for cerebral saccular aneurysm treatment.Also, the infiltration of SMCs may induce better elasticity of new tissue in the sac than fibroblast infiltration.Both more extended studies temporally and broader investigation of bioactive factor loading would be logical next steps for this effort.
Several clinical limitations exist with the described approach including: 1) the experimental method used for embolic delivery in vivo, 2) the potential for downstream embolization resulting in cerebral ischemia, and 3) the rapid formation of a low-oxygen environment resulting in local aneurysm wall ischemia and rupture.In this work, we used an open approach and a temporary clip to prevent accidental downstream embolization.This was done to minimize unnecessary animal suffering.However, in clinical practice, such embolization would have to be performed via endovascular balloon-assisted methods.The potential for downstream embolization due to leakage around the balloon and causing a stroke cannot be overstated.In fact, in the CAMEO trial, 10 a significant percentage of patients treated with Onyx had an intraprocedural complication resulting in permanent neurologic deficit.Another point of concern is that embolic use could result in local aneurysm wall ischemia and early rupture.Hoh et al. described the presence of endothelial cells and capillaries within both human and murine aneurysms. 40The function of these capillaries is unknown.However, rapid ischemia within the aneurysm wall could result in cellular apoptosis, a loss of structural integrity, and early rupture.

CONCLUSION
Thiolated gelatin (Gel-SH) and vinyl sulfonated hyaluronic acid (HA-VS) were successfully synthesized, and their composition was confirmed using 1 H NMR. By the rapid reaction of Gel-SH and HA-VS, upon mixing, ECM-derived, chemically cross-linked porous hydrogels were formed, incorporating the contrast agent iohexol.Furthermore, the bioactive, self-assembling peptide RADA-SP could be incorporated without compromising gel formation and structure.The injectable embolic system demonstrated promising potential for treating cerebral saccular aneurysms.In both in vitro and in vivo (murine) models, the embolic material exhibited rapid gelation, forming an in situ scaffold that enhanced thrombus formation and facilitated tissue ingrowth into the aneurysmal sac.Overall, the ECM-mimicking injectable embolic showed the potential for aneurysm treatment; although, further longerterm (over 3 months) in vivo studies may be necessary for a better understanding of their efficacy.

Figure 3 .
Figure 3.In vitro blood-contacting test where formed HA-VS/Gel-SH gel was lyophilized and contacted ovine blood for 3 h with gentle rocking (n = 4): (A) image of freeze-dried HA-VS/Gel-SH with 10% iohexol; (B) freeze-dried HA-VS/Gel-SH with 10% iohexol after blood contact; (C) HA-VS/Gel-SH with 15% iohexol after blood contact; (D) SEM images of surfaces and cross sections from C and D.

Figure 4 .
Figure 4.In vivo embolization studies using the mice aneurysm model: (A) diagram of schedule of the animal study; (B) aneurysm creation and embolization steps: 1) murine common carotid artery is exposed and a latex cuff is placed underneath; 2) most distal portion of the artery is ligated using 8−0 nylon suture to create a stump; 3) the artery is incubated with elastase for 20 min; 4) at the end of the incubation period an aneurysm has formed, the latex cuff is removed and the neck incision is closed; 5) 3 weeks later the carotid aneurysm is re-exposed and a temporary clip is placed around the proximal part of the artery-aneurysm complex; 6) embolic is injected through a 30G needle; 7) waiting for 5 min; 8) the temporary clip is removed; and (C) intraoperative photographs of the injection.(D) Pro-inflammatory cytokine production (n = 3) after the injection of (a) HA-VS/Gel-SH with 10% iohexol and (b) HA-VS/Gel-SH with 10% iohexol and RADA-SP.