Sodium Phytate-Incorporated Gelatin-Silicate Nanoplatelet Composites for Enhanced Cohesion and Hemostatic Function of Shear-Thinning Biomaterials

Abstract

Shear-thinning biomaterials (STBs) based on gelatin-silicate nanoplatelets (SNs) are emerging as an alternative to conventional coiling and clipping techniques in treatment of vascular anomalies. Improvements in cohesiveness of STB hydrogels pave the way toward their translational application in minimally invasive therapies such as endovascular embolization repair. In the present study, we employ sodium phytate (Phyt) additives to tune the electrostatic network of silicate nanoplatelets (SNs)-gelatin STBs, thereby promoting their mechanical integrity and facilitating injectability through standard catheters. We report that an optimized amount of Phyt can enhance storage modulus by ca. one order of magnitude and reduce injection force by ~58% without compromising biocompatibility and hydrogel wet stability. The Phyt additives were found to decrease the immune responses induced by SNs. In vitro embolization experiments suggested a significantly lower rate of failure in Phyt-incorporated STBs than in control groups. Furthermore, the addition of Phyt led to accelerated blood coagulation (reduced clotting time by ~45% compared to controls) due to the contribution of negatively charged phosphate groups, which could aid in prolonged durability of STB in coagulopathic patients. Therefore, the proposed approach can be an effective method for design of robust and injectable STBs for minimally invasive treatment of vascular malformations.

1. Introduction

Vascular anomalies account for a major portion of the surgical burden on the healthcare system: e.g., each aneurysm repair procedure can cost up to ~$100,000 in the United States;^[1] nevertheless, post-surgery complications such as rebleeding are still prevalent in patients treated with the conventional coiling and clipping surgeries.^[2] With the advent of injectable biomaterials, minimally invasive therapies are growing rapidly to replace traditional open surgeries in the treatment of vascular injuries and malformations.^[3] Several injectable polymers have been commercialized since the U.S. Food and Drug Administration (FDA) first approved endovascular aneurysm repair (EVAR) in 1999.^[3a] However, these products are generally associated with serious limitations: Onyx^™ is a popular catheter-injectable embolic agent based on of ethylene(vinyl alcohol) (EVOH) dissolved in dimethyl sulfoxide (DMSO), which induces cytotoxicity at the injection site.^[4] Histoacryl^®, another embolic agent based on adhesive n-butyl cyanoacrylates (NBCAs),^[5] suffers from catheter clogging (due to its uncontrollable curing), exothermic solidification,^[6] as well as stiffness mismatch with the surrounding soft tissues.^[7]

Injectable hydrogels have received enormous attention due to their versatility for integration with minimally invasive platforms, e.g., in transcatheter embolization.^[8] In addition to their favorable biocompatibility, various gelation mechanisms can be employed to trigger their solidification in situ in a mechanically tunable manner.^[9] Among these mechanisms, delivery of multi-component hydrogels via dual syringes (for chemical crosslinking systems) as well as body temperature-triggered physical gelation are limited by the time-sensitivity of the solidification process.^[10] An example of in situ crosslinking commercial embolic agents is EmboGel, which requires bulky coaxial catheters to deliver alginate and calcium chloride solutions for filling aneurysm sac.^[11] This issue also extends to photocrosslinkable hydrogels (such as UltraGel), as the catheter should be equipped with a light-guiding optical fiber in the cases where light cannot penetrate deep into the aneurysm site. Another crucial consideration for the development of injectable hydrogels is their water absorption capacity. Excessive swelling of crosslinked hydrogels poses the risk of vessel wall rupture; hence, water absorption should be minimized in embolic agents. Previous reports have associated EmboGel with insufficient mechanical strength, necessitating an additional endovascular stent to maintain hydrogel stability.^[11b] Therefore, an easy-to-use, water stable, and biocompatible injectable hydrogel with strong cohesion is in urgent demand to enable long-term durability and to reduce progressive complications such as hemorrhaging aneurysm ruptures.

Recent trends emphasize injectable shear-thinning biomaterials (STBs) as a simple and biocompatible solution for EVAR due to their self-association in the aneurysm sac without the need for external stimuli.^[12] Injectability of STBs through standard catheters is simply enabled by their flow under the application of shear stress due to the injection force.^[13] In our previous study, we engineered a gelatin-based injectable STB nanocomposite where the shear-thinning effects stemmed from the electrostatic interactions in colloidal solutions of highly charged silicate nanoplatelets (SNs).^[14] Hydrogels demonstrated stable dimensions in wet media (minimal swelling). Additionally, their cohesion was tuned through their composition to resist physiological pressures while maintaining their stability to avoid fragmentation and recanalization. In addition to their shear-thinning properties, SNs significantly promoted hemostatic properties of STBs,^[15] which synergistically contributed to hydrogel stability and thereby led to the full occlusion of porcine vasculatures in vivo.

Here, we aim to further enhance the mechanical integrity and hemostatic properties of SN-based STBs given their proven roles in hydrogel durability at the aneurysm site. For this purpose, negatively charged moieties were incorporated within the previously developed^[14] STB hydrogels. We hypothesize that these negative electrostatic interactions can strengthen intermolecular attractions for better cohesion and enable greater binding affinity toward blood components for rapid hemostasis. Various mechanical and physical characterizations are carried out to assess injectability and structural integrity of hydrogels. Hemostatic efficacy of the developed biomaterials is evaluated further via in vitro hemostatic studies. Finally, the capability of developed biomaterials for translational clinical applications is examined through biocompatibility and hemocompatibility evaluations in vitro.

2. Results and Discussion

2.1. Design of cohesive hemostatic shear-thinning hydrogels

To enhance mechanical cohesion and hemostatic effects in STB, a previously optimized formulation of gelatin-SN nanocomposites (Gel-SN) was supplemented by sodium phytate (Phyt) additives, resulting in Gel-SN-Phyt hydrogels (Figure 1a). Sodium phytate acts as a naturally derived chelating agent due to its rich content of negatively charged phosphate groups. These groups can form electrostatic interactions with the positively charged primary amine groups available on the gelatin backbone and SN components as shown in Figure 1a. In addition to increasing intermolecular associations (which can enhance overall cohesion), negative charges of STB can trigger the extrinsic coagulation cascade when they come in contact with blood components.^[15] Therefore, the injectable Gel-SN-Phyt STBs can be engineered for potential application in the minimally invasive treatment of aneurysms (Figure 1b). Note that Phyt is a biodegradable and biocompatible component obtained from rice bran, and is already used in biomedical applications due to its bioactivity^[16] and anticorrosion properties.^[17] The material compositions of the studied biomaterial formulations are represented in Table 1. The samples are labeled as Gel-SN-PhytX, where X=1, 2 represent the amount of Phyt in STBs, i.e., 4.17 and 8.33 mg/ml, respectively.

Figure 1. Development and characterization of shear-thinning biomaterials (STBs) based on gelatin (Gel) and silicate nanoplatelets (SNs) incorporated with sodium phytate (Phyt).

(a) Synthesis route of STB hydrogels. (b) Potential application of STB hydrogels for endovascular embolization. (c) Zeta potential characterization of constitutive components and STB hydrogels. (d) Scanning electron microscopy (SEM) images of pore distribution for different STB formulations. (e) Assessment of weight stability in wet media in terms of relative weight change over time. (f) Results of degradation study performed on different STB formulations.

Table 1.

Composition, labeling, and formulation of injectable shear-thinning biomaterial (STB) hydrogels.

Sample label	Total solid (mg/ml)	Silicate nanoplatelet (mg/ml)	Gelatin (mg/ml)	Sodium phytate (mg/ml)
Gel-SN	60	45	15	0.00
Gel-SN-Phyt1	64.17	45	15	4.17
Gel-SN-Phyt2	68.33	45	15	8.33

2.2. Characterization of shear-thinning biomaterials

The effect of Phyt additives on the hydrogels’ net charges was studied by measuring their zeta potential as shown in Figure 1c. The increased negative charges of SN-Phyt compared to SN samples suggests that the Phyt molecules were electrostatically associated with SN. Correspondingly, the negative charges introduced by SN components in Gel-SN nanocomposites was further increased with the addition of Phyt by ~×3 (from ~−5 mV to ~17 mV). Micro-scale structural characterization of hydrogels was performed via scanning electron microscopy (SEM) of the freeze-dried STBs (Figure 1d). In general, a denser distribution of gelatin (smaller pore sizes) was observed in Gel-SN-Phyt1 compared to the Gel-SN controls. This distribution pattern can be attributed to the effect of electrostatic interactions of Phyt additives within the gelatin macromolecules. Further addition of Phyt in Gel-SN-Phyt2 compared to Gel-SN-Phyt1 was associated with larger pore sizes indicating lower intermolecular attractions in hydrogels. This effect can be attributed to the disruption of SN’s self-assembly into the “house of cards” structure with the excess Phyt content, as they can occupy positively charged edges of SNs thereby reducing SN-SN attractions.

To evaluate weight-stability in wet conditions, mass changes of STBs in response to incubation in physiologically relevant media, i.e., phosphate-buffered saline (PBS) was characterized over time as shown in Figure 1e. As opposed to several other covalently crosslinked hydrogels,^[18] which suffer from tissue compression due to their swelling, the STBs exhibited minimal weight changes (and no visible volume change) over 48 h incubation at 37 °C. This effect can be explained by peripheral infusion of positive ions present in PBS within the STBs leading to association of negatively charged SN faces, which is in essence, in agreement with low solubility of SN in PBS.^[19] It was seen that the addition of Phyt did not alter weight stability of Gel-SN STBs significantly as the dry weight of hydrogels before and after soaking in PBS did not change significantly (freeze dried weight of 15.1±2.6 mg and 14.7±1.9 mg, respectively). A long-term degradation study was conducted by incubating STBs in an enzymatic solution as demonstrated in Figure 1f. Although the impact of 1 %w/v Phyt additives was not statistically significant, further addition of Phyt (i.e., 2 %w/v) compromised the hydrogel stability in wet media. The greater degradation rate of Gel-SN-Phyt2 can be also due to Phyt additives shielding the positively charged edges of SN fillers, thereby hindering their self-association into the “house-of-cards” structure as discussed above for Figure 1d. This process can facilitate the enzymatic cleavage of gelatin chains, which can eventually lead to their accelerated release. We further observed that gelation of SN solutions in the absence of gelatin was completely disrupted by adding an excess amount of Phyt (>~5 %w/v) to SN dispersions corroborating the gelation-disrupting effects of Phyt at high concentrations.

2.3. Rheological characterizations

To further understand the effects of Phyt-induced interactions on the mechanical properties of Gel-SN STBs, rheological tests were performed as illustrated in Figure 2a–c. The results of strain sweep tests are represented in Figure 2a. As can be seen, the addition of 1 %w/v Phyt led to a higher storage modulus (G’) compared to Gel-SN in the strain intervals before the yield point (elastic zone). This larger storage modulus is explained by the additional electrostatic junction zones with Phyt additives (i.e., interactions between negatively charged phosphate groups of Phyt and positively charged primary amine groups of gelatin as well as SN edges). Further addition of Phyt however, reversed this trend: a decrease in storage modulus observed in the static regime could be possibly due to Phyt molecules shielding the positive edges of SNs, thereby disrupting their association, which is in accord with the discussion in Figure 1d and f.

Figure 2. Rheological and mechanical characterization of gelatin (Gel)-silicate nanoplatelet (SN)-sodium phytate (Phyt) shear-thinning biomaterials (STBs).

(a) Strain sweep results demonstrating the effect of Phyt additives on the hydrogel storage modulus. (b) Viscosity-shear rate characteristics of STB samples demonstrating the shear-thinning behavior of hydrogels. (c) Time sweep test results showing hydrogels’ strain recovery under high strain (100%) and low strain (0.01%, highlighted in red) intervals. (d) Experimental setup for performing injectability tests. (e) Dimensions of catheter and needle outlets used for injectability tests. L, length. ID, internal diameter. (f) A typical force-time curve obtained from injection tests based on which the injection force was calculated. (g) The results of injection force measured at two different flow rates. (h) Effect of gauge outlet on injection force of STB hydrogels.

Further, a strain-softening was observed after the breaking point, corresponding to the decreasing trend of storage modulus with shear strain. Here, addition of Phyt in STBs was associated with an accelerated post-yield drop of G’ compared to that of Gel-SN. As discussed above, in the low strain regime (static elastic zone) where the molecules are the most interacting electrostatically, a higher storage modulus with addition of Phyt is explained by the larger density of electrostatic attractions. In the high strain regime after the breaking point however, re-association of electrostatic network is postponed due to Phyt slowing down or disrupting “house of cards” formation of SNs (because of Phyt blocking positive edges of SNs), which explains the accelerated softening of Phyt-incorporated STBs post-breaking point. This behavior renders higher shear stress sensitivity of STBs, which is in agreement with our previously published work^[20] and can facilitate their injection while enhancing hydrogel cohesion.

The decreasing trends of viscosity-shear rate characteristics in all STBs (Figure 2b) signifies their shear-thinning character due to the mechanically sensitive reconfiguration of the electrostatic network of SNs. It was observed that viscosity dropped at a higher rate in Phyt-incorporated STBs compared to Gel-SN samples (indicating a stronger shear-thinning effect), which agrees with the accelerated G’ drop in post-yield behavior seen in strain sweep tests (Figure 2a).

The rheological recoverability of shear-thinning effects is evaluated in time sweeps with alternating high (100%) and low strain (0.01%) intervals as shown in Figure 2c. All STBs could recover their mechanical integrity at low strain intervals after the network rupture at high strain intervals due to the restoration of broken ionic bonds. The Gel-SN-Phyt1 formed stronger electrostatic attractions and a more cohesive network compared to Gel-SN controls (due to larger G’ at low strain amplitudes); whereas their broken network at higher strain amplitudes was much softer than Gel-SN (due to lower G’), implying a more facile injectability. This effect was in agreement with the results of Figure 2a and supports the underlying mechanisms explained for higher storage modulus of Phyt-incorporated STBs in low strain amplitudes and their lower storage modulus in high strains (post-yield point).

Time sweeps were performed to examine long-term evolutions of rheological properties over 48 h (see Figure S1a). We found that the hydrogels continue to self-associate for ~24 h after which the mechanical properties remained steady with further aging. Furthermore, to investigate the effect of gelatin’s thermosensitivity on rheological characteristics after implantation in physiological conditions, the hydrogels were subjected to a heating profile from 25 °C to 37 °C while their G’ was monitored over time (Figure S1b). The values of G’ for all conditions remained stable over the course of testing, suggesting that gelatin’s potential phase transition in physiological conditions had a minimal impact on the overall structural integrity of STBs.

2.4. Injectability of shear-thinning biomaterials

Facile injectability of embolic agents is a critical requirement in their minimally invasive applications. Shear-thinning biomaterials are attractive candidates due to their softening upon application of shear-inducing injection forces, which facilitates injectability by physicians. Injectability of STBs was characterized by the experimental setup shown in Figure 2d. Needle gauges and catheters with various dimensions were used to study injectability of STBs as shown in Figure 2e. Figure 2f shows a typical injection force response of STBs while flowing through a 2.7F catheter. The force began with a sharp increase corresponding to the static friction between the plunger and syringe wall. Thereafter, the increase in force was continued up to the maximum force required to break the electrostatic bonds of the SN network. Then, the force reached a plateau, which corresponded to the dynamic flow of hydrogels through the catheters. A parametric study was conducted to understand the injection force for different flow rates, outlet types, and STB formulations (Figure 2g,h). In general, a larger injection force was required to flow the STBs at the higher flow rate, which agrees with Poiseuille’s law. Furthermore, the addition of 1 %w/v Phyt (Gel-SN-Phyt1) to Gel-SN nanocomposites significantly reduced the force required to enable their injection. This observation was in accord with the lower G’ of Gel-SN-Phyt compared to Gel-SN in rheological assessments of hydrogels (in the network-disrupted state) as discussed for Figure 2g. The difference between the Gel-SN-Phyt1 and Gel-SN-Phyt2 STBs was not found to be statistically significant. The injection forces of different needle and catheter types are presented in Figure 2h. According to Bernoulli’s law, the injection force is proportional to the length of needles (due to the head loss because of wall friction with the catheters) and inversely varies with the internal diameter of catheter/needles (ID). This effect can be seen in the results of injection force where an increase of ID from 0.34 mm to 0.84 mm, augmented the injection force by ca. ×2.

2.5. In vitro biocompatibility and hemocompatibility

The interactions between the STBs and living cells as well as blood components were evaluated to assess biocompatibility of hydrogels. For the former, an in vitro biocompatibility test was implemented by exposure of the human umbilical vein endothelial cells (HUVECs) to the hydrogel extracts. The live/dead assay performed on day 5 (see Figure 3a,b) demonstrated no significant difference in terms of cell viability with that of control and Gel-SN conditions. However, a minor number of dead cells appeared at high concentration of Phyt additives (i.e., in Gel-SN-Phyt2 groups). The potential cause of cell death in these hydrogels can be the excess Phyt content and their concentrated release leading to cytotoxic effects in STBs. Nevertheless, we note that cell viability in our optimized STB, i.e., Gel-SN-Phyt1 remained acceptable. The metabolic activity of cells was studied using PrestoBlue^™ assay as shown in Figure 3c to quantify cell proliferation after exposure to STBs. Although the SN groups exhibited a lower mean of metabolic activity, greater figures with no significant difference were measured across Gel-SN and Gel-SN-Phyt formulations.

Figure 3. Evaluation of hydrogel biocompatibility, hemocompatibility, and immunoactivity for shear-thinning biomaterials (STBs) based on gelatin (Gel)-silicate nanoplatelet (SN)-sodium phytate (Phyt).

(a) Fluorescent images of the live (green)/dead (red) assay performed on day 5. (b) Values of cell viability calculated based on the live/dead images for STB samples. (c) Results of PrestoBlue^™ assay quantifying metabolic activity of the cells exposed indirectly to the hydrogel samples. (d) Optical images of tubes containing hydrogels and centrifuged whole blood samples for hemolysis assay. (e) Hemolysis assay performed to evaluate hemocompatibility of hydrogels. DI, deionized. (f) Scatter plot of flow cytometry results to analyze immunoactivity of hydrogels. The expression levels of (g) CD80/CD86 and (g) CD206/CD163 positive cells after 1 day incubation.

Hemocompatibility of hydrogels was examined via hemolysis tests. The whole blood from human sources was exposed to the hydrogels for a certain period of time, and disruption of red blood cells (RBCs) was measured following a centrifugation process by monitoring the color changes of the supernatant (see Figure 3d,e). The results suggested that Phyt additives significantly promoted hemocompatibility of STBs according to the ~60% lower hemolysis ratio of Gel-SN-Phyt1 compared to Gel-SN. These results can be attributed to the Phyt additives shielding positive charges at the SN edges, which can eventually mitigate the cell membrane-disrupting effect of SNs induced by their positive charges.

2.6. Cellular immune function

The expression of co-stimulatory ligands by the cells were measured to evaluate the immunostimulatory activity of hydrogels (Figure 3f,g). The results showed (i) SN dose-dependent response of macrophages for both activation levels and (ii) enhanced anti-inflammatory polarization of macrophages in high-dose SN activation. The level of inflammatory and anti-inflammatory macrophages raised significantly in SN and Gel-SN compared to control groups by ~4-6×, whereas the activated and anti-inflammatory macrophages in Gel-SN-Phyt1 were found to be on the order of control samples (Figure S2). On day 2, the level of inflammatory and anti-inflammatory signal decreased and then the immune system activation and deactivation reached the normal level on day 4 (comparable to the control). Therefore, it can be inferred that the addition of Phyt could mitigate the immune system activation due to SN, thereby decreasing the risk of organ inflammation, damage, and autoimmune diseases in patients.

2.7. Hemostatic properties

Formation of blood clots around the STBs after injection can favor their stability in blood vessels in situ as it reduces the risk of biomaterial fragmentation (in addition to preventing local hemorrhage). Here, we study the hemostatic effects of Phyt additives in Gel-SN-based systems via an in vitro clotting time assay (as shown in Figure 4a,b). The SN content led to a decrease in time to hemostasis from ~28 min (blank controls) to ~11 min due to its strong electrostatic charges,^{[15, 21]} as seen in Figure 1a. This clotting time was further reduced by ~45% (i.e., to ~6 min clotting time) with the addition of Phyt (no significant difference was seen between 1 and 2 %w/v Phyt content) compared to Gel-SN samples. Formation of blood coagulation was also characterized through UV-vis spectroscopy for better quantification of hemostatic effects (Figure 4c). Again, the hemostatic effects of Phyt were confirmed by the lower absorption of Gel-SN-Phyt groups compared to Gel-SN controls. The promoted hemostatic effect with the addition of Phyt is explained by the introduced negative charges (phosphate groups in Phyt). These results are in accord with the previous studies, which demonstrated significant improvement in hemostatic function due to phosphate’s ability in triggering extrinsic coagulation cascade (through activation of factor V).^[22] Further, it has been reported previously that phosphate groups can delay clot lysis, thereby enhancing clot stability, which is essential to prevent clot migration to remote distances.^[23]

Figure 4. Hemostatic performance of shear-thinning biomaterials (STBs) developed by the addition of sodium phytate (Phyt) in gelatin (Gel) and silicate nanoplatelets (SNs) composites (i.e., Gel-SN-Phyt).

(a) Blood clotting assay showing time to blood coagulation for the whole blood samples exposed to different STB formulations. (b) Quantification of the relative decrease in blood clotting time compared to blank controls due to electrostatic charges in STB hydrogels. (c) The results of hemostatic assay based on UV-vis absorbance of supernatant blood after 6 min contact with the samples.

2.8. Embolization efficacy in vitro

Resistance of the injected STBs to shear stresses applied due to fluid flow was examined in an in vitro brain aneurysm model using the embolization setup demonstrated in Figure 5a. This resistance was monitored in terms of internal pressure inside the fluid circuit. The increasing trend of internal pressure over time is seen in Figure 5b. Here, the failure point is corresponded to an abrupt drop in pressure. The probability of occlusion (i.e., no failure below 220 mmHg), as well as failure (namely leakage, and ejection) was calculated for different flow rates resembling the flow in malformed vasculatures as shown in Figure 5c. For all tested flow rates, addition of Phyt at its optimal concentration (Gel-SN-Phyt1) significantly enhanced the occlusion efficacy of STBs: the success rate of ~16-33% in Gel-SN controls reached up to ~83% in Gel-SN-Phyt1, which we attribute it to the enhanced cohesion of STBs.

Figure 5. Evaluation of embolization capacity of shear-thinning biomaterials (STBs) developed based on sodium phytate (Phyt)-incorporated gelatin (Gel)-silicate nanoplatelets (SNs) (i.e., Gel-SN-Phyt).

(i) Schematic of the experimental setup used to characterize embolization capacity of the injected STB hydrogels. (j) Pressure-time responses of different hydrogel formulations at 30 ml/min flow rate. (k) Probability of successful occlusion and failure (i.e., ejection and occlusion) in response to 220 mmHg fluid pressure for flow rates of I: 30, II: 60, and III:100 ml/min. PBS, phosphate-buffered saline.

3. Conclusions and Prospects

Shear-thinning gelatin-based biomaterials such as gelatin-SN composites have shown promise in minimally invasive endovascular embolization as they offer excellent biocompatibility, superior wet-stability, simple use, and injectability using standard catheters. Hydrogel cohesion, which is aided by local blood coagulation, is key to prevent from fragmentation and recanalization thereby essential in complete occlusion of aneurysm sacs. While increasing SN concentration in Gel-SN composites renders STB hydrogels brittle, introducing negative charges in hydrogel formulations using Phyt could effectively enhance mechanical integrity and stress-sensitivity of shear-thinning behavior. Without making a negative impact on wet-stability and degradation behavior of STBs, a low concentration of Phyt (i.e., 1 %w/v) in previously optimized STB hydrogel is able to enhance the resistance against high fluid pressure in blood vessels, making STBs suitable for e.g., patients with high blood pressure. As far as coagulopathic patients are concerned, Phyt additives also offer promising hemostatic properties, accelerating blood coagulation by ~45%. Lastly, the optimized concentration of Phyt for effective improvement in cohesion and hemostatic functions (i.e., 1 %w/v) not only did not induce major toxicity in STB formulations, but also mitigated negative immunostimulatory effects of SNs. Therefore, it is inferred that electrostatic tuning of STB networks using negatively charged components such as Phyt can be a multifaceted approach in the design of biomaterials for minimally invasive therapies.

4. Experimental Section

Hydrogel preparation:

Stock solutions of gelatin (G1890, Sigma Aldrich) at 18 %w/v and 9 %w/v SNs (Laponite XLG, BYK) were prepared using deionized (DI) water according to previous protocols.^[15] To fabricate hydrogels, gelatin stock, silicate nanoplatelet stock, and DI water were mixed using a speed mixer (DAC 150.1 FV-K, FlackTek) at 3000 rpm for 15 min. For samples including Phyt (P8810, Sigma), the desired weight was first dissolved in the DI water before mixing the three components.

Dynamic light scattering:

For zeta potential analysis, hydrogel samples were dissolved in DI water before loading into a zeta potential analyzer (Zetasizer Nano ZS, Malvern Panalytical). The pH of solutions was in the range of ~5-6. The solutions were prepared at 2 mg/ml concentration except for the SNs where they were tested at the concentration used in hydrogel formulations.

Scanning electron microscopy:

To image materials with SEM, hydrogel samples were prepared by freezing at −80°C for 24 hours, freeze-drying for 48 hours, and coating with a thin layer of Au. Samples were sectioned for loading into an SEM plate. Scanning electron microscopy was performed with a high-resolution scanning microscope (Supra 40VP, Zeiss) with an accelerating voltage of 1 kV.

Swelling test:

The swelling ability of the materials was determined as the percentage of swelling according to the following equation: Swelling ratio = (W-W₀)/W₀ × 100% where W and W₀ are the wet and freeze-dried hydrogel weights, respectively. Previously freeze-dried material samples were incubated in PBS for specific time intervals. After soaking, swollen samples were removed from PBS and weighed immediately.

Degradation tests:

A volume of 1.5 mL collagenase (5 μg/mL) was used to incubate weighed materials samples (200 mg) in a 24-well plate at 37°C. At specified time points, samples were removed from the collagenase media, fully dried, and weighed. Material samples were fully dried by first freezing at −80°C for one day, then freeze-dried for two days. After weighing, samples were resubmerged in new collagenase media.

Rheological tests:

A rheometer (MCR 302, Anton Paar) with a 25 mm diameter sandblasted parallel plate geometry and 1 mm gap was used for mechanical testing: shear stress, viscosity, and storage modulus were all measured. All samples were equilibrated to room temperature and tests were run at 25°C. Frequency sweeps were performed at 1% strain, with sweeping frequencies between 0.001 and 100 Hz. Shear rate sweeps were performed at 10 points/decade, with shear rates between 0.001 and 100 s⁻¹. Oscillatory stress sweeps were carried out at 1 Hz between 0.01 and 100 Pa. Strain sweeps were conducted at 1 Hz between 0.01 and 1000%. Recovery tests were implemented at 1 Hz and 100% strain, which is much larger that the gels’ linear viscoelastic range. Then, 1% strain was applied for 5 min to track recovery of hydrogels.

Injectability test:

To test the injection force of materials, samples were loaded into 1 mL plastic syringes and injected through medical catheters (2.7F and 3.5F) and needles (18G and 23G) using standard Luer lock adapters. The outlets of the catheters were submerged in PBS at 37°C. The syringe plunger was depressed with an upper compression plate, with compression tested at two constant rates of 33.33 and 100 mm/min. The body of the syringe was fixed in place using a lower tensile grip. Injection force was measured mechanically by an Instron 5943 with a 1 kN load cell and recorded with Bluehill version 3 software.

In vitro embolization test:

To determine the gels’ ability to occlude blood flow, an in vitro bench test was performed. The system consisted of a pulsative pump pushing a solution into silicon tubes connected to an aneurysm phantom. The aneurysm phantom was filled with the gel using a 3.5F catheter. The pressure applied to the gel was recorded continuously using a manometer (PDA 100L, PCE) connected to a computer. As solution flowed through the system, pressure increased up to a maximum of 220 mmHg (a value that exceeds physiological blood systolic pressure); if the gel allowed liquid through or broke before reaching the maximum pressure, a sudden drop in pressure was recorded. Silicon tubes (3.175 mm inside diameter and 1.6 mm thickness (VWR)) were used to construct the system, as their mechanical properties are like those of arteries (Young’s modulus of 1.50 MPa). The influence of the gel composition and the effect of flow rate (30, 60, 100, and 200 ml/min) on embolization efficacy were studied (n=6).

In vitro biocompatibility:

For assessment of hydrogel biocompatibility, HUVECs were cultured in a Dulbecco’s modified eagle medium (DMEM). The HUVEC cells were seeded onto hydrogel-coated slides and cultured for 5 days. After an incubation period, slides were stained with a live/dead cytotoxicity assay using PrestoBlue^™ reagent. Slides were imaged using a fluorescent microscope (BZ-X710, Keyence). To quantify the live/dead results, cells from different areas were counted with ImageJ software (100× magnification). Cell viability (%) was defined as the ratio of living cells to the total number of cells.

Immunization studies:

To harvest bone marrow-derived macrophage cells (BMDMs), bone marrow was flushed first from femur and tibia extracted from C57BL/6 mice that were 5 to 6 weeks old. Isolated bone marrow cells were seeded at one × 10⁶ per dish in RPMI 1640 augmented with 55 μM β-mercaptoethanol, 10% FBS, 5 ng/ml of MCSF, and 100 U/ml penicillin. The culture media was exchanged (50% of volume) with fresh media on days three and five. After eight days, BMDMs were harvested and plated at one ×10⁶ cells per well in 12-well plates. Macrophages were Treated with SNs, Gel-SN, and various formulations of Gel-SN-Phyt. After 24, 36, 48, and 72 h, media were collected, and cells were stained for markers indicating the activation and polarization of macrophages. Cells were then analyzed using flow cytometry to determine the percentages of each population.

Hemolysis tests:

Blood samples were diluted 50× in 0.9 %w/v saline solution. Material samples were flattened into 15 mL Falcon tubes in a centrifuge (Sorvall Legend X1R, ThermoFisher Scientific), and an equal volume of diluted blood was added to each tube. For controls, equal volumes of diluted blood and either DI water (positive) or saline (negative) were added to an empty tube. All samples were incubated at 37°C for 2 hours in a shaker incubator (Labline Instruments) at 100 rpm. After incubation, samples were centrifuged at 2000 rpm for 15 min (Sorvall Legend X1R, ThermoFisher Scientific), and the supernatant from each sample was transferred into a 96-well plate. The absorbance of each well was read at 545 nm using an absorbance plate reader (Varioskan LUX, ThermoFisher Scientific). Percent hemolysis was calculated based on the following equation: %Hemolysis = (A_s-A_neg)/A_pos × 100% where A_s is the absorbance of supernatant, A_neg is the saline diluted blood absorbance, and A_pos is the DI water-diluted blood absorbance.

Clotting time assay:

Blood was mixed with 0.1 M CaCl₂ at a 1:10 ratio of CaCl₂ to blood. This mixture was vortexed for 10 s and pipetted into sequential wells on a 48-well plate. Wells were washed with 0.9% (w/v) saline solution at set time points to stop clot formation. The saline solution was aspirated from each well and the saline wash was repeated until all soluble components were removed. To test STB hemocompatibility, STB samples were loaded into syringes and injected into the base of the wells. The STB-loaded well plates were centrifuged (Sorvall Legend X1R, ThermoFisher Scientific) at 3000 rpm for 5 minutes to ensure the gels were level, and clotting analysis was performed as previously described. Clotting time trials were performed in triplicate, and the final clotting time was determined qualitatively when at least two of the three sample wells formed a uniform clot.

STB samples with similar qualitative results from the initial coagulation test were differentiated with a secondary coagulation test. Blood was prepared as in the initial coagulation test and pipetted into 1.5-mL Eppendorf tubes. For STB testing, gel samples were loaded into 1.5-mL Eppendorf tubes via syringe and centrifuged (MySpin 6, ThermoFisher Scientific) for 30 s to form a level surface before the addition of blood. A fixed amount of saline solution was added to each sample after set time points, and this wash was collected for absorbance measurements at 405 nm. The absorbance of each well was measured using an absorbance plate reader (Varioskan LUX, ThermoFisher Scientific).

Statistical analyses:

All the measurements were performed at least in triplicates unless noted otherwise. The values reported here represent the mean ± standard deviation. The significance of difference between the means was analyzed using a one-way analysis of variance (ANOVA) where p<0.05 was statistically significant.