Controlled Delivery of Corticosteroids Using Tunable Tough Adhesives

Abstract

Hydrogel-based drug delivery systems typically aim to release drugs locally to tissue in an extended manner. Tissue adhesive alginate-polyacrylamide tough hydrogels were recently demonstrated to serve as an extended-release system for the corticosteroid triamcinolone acetonide. Here, we describe the stimuli-responsive controlled release of triamcinolone acetonide from the alginate-polyacrylamide tough hydrogel drug delivery systems (TADDS) and evolving new approaches to combine alginate-polyacrylamide tough hydrogel with drug-loaded nano and microparticles, generating composite TADDS. Stimulation with ultrasound pulses or temperature changes was demonstrated to control the release of triamcinolone acetonide from the TADDS. The incorporation of laponite nanoparticles or PLGA microparticles into the tough hydrogel was shown to further enhance the versatility to control and modulate the release of triamcinolone acetonide. A first technical exploration of a TADDS shelf-life concept was performed using lyophilization, where lyophilized TADDS were physically stable and the bioactive integrity of released triamcinolone acetonide was demonstrated. Given the tunability of properties, the TADDS are a suggested technology platform for controlled drug delivery.

Graphical Abstract

This study reports alginate-polyacrylamide tough hydrogel drug delivery systems (TADDS) for the stimuli-responsive release of triamcinolone acetonide. Stimulation with ultrasound pulses or temperature changes control the release of triamcinolone acetonide from the TADDS, and the incorporation of laponite nanoparticles or PLGA microparticles further extend release. Given its tunability, the TADDS are a technology platform for controlled drug delivery.

Introduction

Systemic administration of anti-inflammatory drugs presents challenges due to the large dose required, need for frequent administration, and potential side effects^[1][2]. Extended local delivery from hydrogel-based biomaterials is considered to limit systemic undesired side effects commonly associated with corticosteroids^[3] by increasing local drug bioavailability while decreasing the risk of drug-induced systemic effects^[4]. Conventional hydrogels, however, have limitations in drug loading and controlled release, as well as weak mechanical properties^[5][6]. Of particular note, greater drug loading typically results in greater burst release as drug is freely diffused through the hydrated polymer network, leading to unwanted systemic exposure^[6].

Hydrogels with high matrix toughness and adhesion (i.e., a tough adhesive hydrogel achieving stretches > 15 times their initial length, fracture energies > 1 kJ/m^2, and adhesion energies > 1000 J/m^2)^[7] are suggested to be advantageous for extended, local drug release due to their maintenance of toughness, stretchability, and adhesive strength during and after drug loading^[8]. The high adhesion energy of the TADDS allows for strong adherence to a target tissue during drug delivery on a variety of tissue surfaces^[9], as compared to other tissue adhesives with weaker adhesive strength^[10][11][12]. Unlike conventional hydrogels, its high matrix toughness and tensility also aid in providing a secure environment during release that may be advantageous in tissue environments frequently undergoing mechanical loading. Indeed, a tissue adhesive tough hydrogel drug delivery system allowed for very high drug loadings (up to 500mg/ml) and dissolution-driven extended release of corticosteroids in vitro and in vivo while offering an adherent surface that is advantageous for drug release to wet and dynamically moving tissues^[8]. A tissue adhesive tough hydrogel drug delivery system exhibiting further control over extended release and responsiveness to stimuli would offer novel therapeutic options for a broad spectrum of clinical indications.

To achieve control over the release kinetics of drugs, several strategies have been explored to accelerate or slow delivery through physical stimulation or structural modification. A stimuli-responsive drug delivery system may be useful in several clinical scenarios. It can titrate drug release to disease activity of inflammatory diseases, optimizing therapeutic efficacy and reducing the risks for drug-induced systemic toxicity^[13]. Transient bursts of drug release from ionically cross-linked alginate hydrogels were achieved through ultrasound^[14] or mechanical stimulation^[15]. Conversely, exposing temperature-responsive hydrogels to low temperatures actively slowed release^[16]. In addition, composite hydrogels embedded with laponite nanoparticles have been shown to promote sustained release due to their ionic and hydrogen bond interactions with positively charged drugs^[6]. A hydrogel system exhibiting these abilities is desired for its extended release and stimuli responsiveness to further tune release.

Corticosteroids possess powerful anti-inflammatory and immunomodulatory properties and are commonly used to treat a variety of conditions (e.g., autoimmune diseases, allergic reactions, chronic obstructive pulmonary disease). Ruptures and repetitive strain injuries to the joints, for example, are often accompanied by tissue inflammation treated with corticosteroid injections^[13]. Despite these beneficial clinical effects, their systemic use is also associated with serious risks, especially at high doses for extended periods, that include osteoporosis, adrenal insufficiency, and aseptic joint necrosis among many^[17][18].

Here, we explored the stimuli-responsive controlled release of the corticosteroid, triamcinolone acetonide, from a tough adhesive drug delivery system (TADDS) formed from an interpenetrating network of an ionically crosslinked polymer network (alginate) with a covalently crosslinked network (polyacrylamide)^[19][9][8]. While the system was capable of high drug loading and adhesion to underlying tissue surfaces, unlike conventional drug delivery systems, it’s responsiveness to different stimuli and capacity for further extended release through the incorporation of nanoparticle formulations had not been investigated^[8]. We examined ultrasound stimulation to control the release of triamcinolone acetonide from the TADDS through disruption of the calcium crosslinks in the hydrogel ionic network^[14] and assessed temperature and mechanical compression to control release of triamcinolone acetonide from the TADDS (Figure 1). Moreover, we incorporated laponite nanoparticles and PLGA microparticles into the tough hydrogel and assessed the modulation of drug release from the composite TADDS. Finally, we explored lyophilization as a technical shelf-life concept for the TADDS to better store the drug delivery system without significantly impacting its release kinetics or mechanical stability. In contrast to previous drug delivery systems^{[20][21][22][23][6][13][24]}, we demonstrate that the TADDS has high mechanical toughness, high drug loading capacity, and sustained or on demand drug release. Given the properties of the TADDS, this family of materials may have a broad set of drug delivery applications.

Figure 1 |. Physical stimuli and structural changes tested in developing the tissue adhesive tough adhesive drug delivery system (TADDS).

Schematic of the system, comprising the tough hydrogel dissipative matrix (alginate/PAAM), adhesive surface (chitosan), and drug (triamcinolone acetonide). Physical stimuli (ultrasound, temperature, mechanical compression) and particles (laponite, PLGA) were explored to control and modulate drug delivery.

Results

Dissolution-Driven Release of Triamcinolone Acetonide

Triamcinolone acetonide was loaded into the TADDS at magnitudes above its solubility limit in water. High frequency ultrasound (HFUS) imaging was used on hydrogels submerged in HBSS to visualize and quantify, and not intended to modify, the release of triamcinolone acetonide as associated with a decrease of echogenicity of the TADDS (Figure 2a). Triamcinolone acetonide gradually dissolved at the aqueous interface, initially at the surface of the TADDS and over time, in the core of the TADDS (Figure 2b). The dissolution-driven release of triamcinolone acetonide was extended for 2 weeks (Figure 2c), while no surface erosion and disintegration of the TADDS became apparent^[8] (Supplementary S1). The TADDS maintained peak adhesion energies (i.e., energy released when the TADDS comes into contact with a tissue surface) as compared to the tough adhesive hydrogel (TA) alone at high triamcinolone acetonide loadings (Figure 2d).

Figure 2 |. Release of triamcinolone acetonide (CORT) from the tough hydrogel drug delivery system (TADDS).

(a) Sagittal HFUS images were taken of the cylindrical TADDS, which had a larger diameter than height (b) over the course of 7 days of drug release (n=7 gels/timepoint) (100mg/ml loading). Blue dashed lines indicate the perimeter of the TADDS over time. White dashed lines indicate the perimeter of the drug encapsulated within the TADDS. TADDS were imaged while placed on a support surface (opaque material underneath TADDS). Scale bar = 1mm. (c) Cumulative release curve of CORT (100mg/ml loading). Mean values are shown, and error bars are ±s.d. (n=7 gels/timepoint), as analyzed by a one-way ANOVA. *P<0.05. (d) Peak adhesion energies (J/m²) of the drug loaded TADDS as compared to the unloaded tough adhesive hydrogel (TA) alone. Mean values are shown, and error bars are ±s.d. (n=4 gels/group), as analyzed by a Student’s t-test. *P<0.05.

Physical Stimuli Alters the Release Rate From the TADDS

Controlled release of triamcinolone acetonide from the TADDS in response to physical stimulation was explored through ultrasound sonication, temperature change, and mechanical compression. To avoid stimulating during any period of more rapid release, release from the TADDS was first tracked for three days in sink conditions as done previously. Then, ultrasound pulses stimulated the TADDS at 37°C (Figure 3a) which significantly increased the release of triamcinolone acetonide compared to the non-stimulated control TADDS (Figure 3b). With each successive round of stimulation, the release of triamcinolone acetonide was increased, and images comparing the two groups show a visible reduction in drug loading of the TADDS (Figure 3b).

Figure 3 |. The effects of physical stimuli on the release of triamcinolone acetonide (CORT) from the tough hydrogel drug delivery system (TADDS).

(a) Schematic of the experimental set-up for triggering release on-demand through ultrasound. (b) Cumulative release curve for ultrasound-stimulated drug delivery during a 5-hour study after 3 days of unstimulated release. Down arrows indicate stimulation for 3 minutes. Mean values are shown, and error bars are ±s.d. (n=3 gels/group), as analyzed by a two-way ANOVA (ultrasound stimulation (US) and time) with post hoc t-tests with Bonferroni corrections. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. (c) Schematic of the experimental set-up for stimulating release through temperature changes. (d) Cumulative release curve for the TADDS at physiological (37°C) or room (25°C) temperatures (n=3 gels/group). Mean values are shown, and error bars are +s.d. (n=3 samples/group), as analyzed by a two-way ANOVA (temperature and time) with post hoc t-tests with Bonferroni corrections. (e) Cumulative release curve for the TADDS placed at 37°C and at 4°C, with transient heating to 37°C on days 3, 6, and 9, as indicated by the down arrows. Mean values are shown, and error bars are ±s.d. (n=3 gels/group), as analyzed by a two-way ANOVA (temperature and time) with post hoc t-tests with Bonferroni corrections. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

The temperature of the release medium was also found to affect drug release from the TADDS (Figure 3c). At body temperature (37°C), release occurred significantly faster than at room temperature (Figure 3d). There was significantly greater burst release (i.e., rapid initial release of drug), almost 50% of the initial drug loading, at 37°C as compared to 25°C (Figure 3d). A sudden change to body temperature (4°C to 37°C) transiently accelerated release (Figure 3e).

To examine the ability of mechanical compression on the TADDS to stimulate release, the TADDS was cyclically loaded in compression (Supplementary S2). After three days of release in sink conditions, mechanical loading of the TADDS did not accelerate release (Supplementary S2). When examining the dynamic mechanical properties of the TADDS during cyclical loadings, the dynamic modulus, but not the tan(δ) (i.e., the tangent of the ratio of the storage modulus to the loss modulus), of the material decreased slightly with each successive mechanical stimulation (Supplementary S3).

Extending Release Through the Inclusion of Drug Sequestering Particles

Drug delivery from the TADDS in response to the inclusion of particles that sequester drug was next examined. The triamcinolone acetonide was first mixed with laponite nanoparticles to form composite hydrogels (Figure 4a). Addition of laponite (0 to 8% (wt/vol) of laponite), did not affect the tensile properties (i.e., maximum stretch and stress) of the TADDS (Supplementary S4), which can be beneficial in a dynamic tissue environment. Whereas existing adhesives do not successfully bear large deformation, the TADDS combines strong adhesion with high toughness and tensile strength, which can be critical when delivering drugs in dynamic tissue environments, such as tissue interfaces^[9][15]. In comparison to the TADDS, many hydrogels lose physical integrity during handling or functioning on mechanically bearing tissues, which may change the drug release property^[6]. When the relative laponite concentration was much greater than drug (40 mg/mL laponite and 1 mg/mL triamcinolone acetonide for a ratio of 40:1 laponite:drug), the composite TADDS demonstrated significantly slower burst release (Figure 4b). At higher drug loadings, however, fewer differences were observed between the composite and non-composite hydrogels (Supplementary S5).

Figure 4 |. The effects of laponite and PLGA particles on the release of triamcinolone acetonide (CORT) from the composite tough hydrogel drug delivery system (TADDS).

(a) Schematic of the molecular composition of the composite TADDS with laponite or PLGA particles. (b) Cumulative release curve for the laponite (40mg/ml) composite TADDS with 1mg/ml CORT, and control TADDS at same drug loading. Mean values are shown, and error bars are ±s.d. (n=3 gels/group), as analyzed by a two-way ANOVA (laponite (lap) and time) with post hoc t-tests with Bonferroni corrections. *P<0.05. (c) Cumulative release curve for the PLGA (40mg/ml) composite TADDS with ~10mg/ml CORT, as compared to TADDS with no PLGA at same drug loading. Mean values are shown, and error bars are ±s.d. (n=3 gels/group), as analyzed by a two-way ANOVA (PLGA and time) with post hoc t-tests with Bonferroni corrections. ****P<0.0001.

Triamcinolone acetonide particles encapsulated within PLGA microparticles were then immobilized in gel during the TADDS synthesis (~10mg/ml final triamcinolone acetonide loading in TADDS) (Figure 4a). The release duration of triamcinolone acetonide was extended in PLGA- triamcinolone acetonide groups as compared to triamcinolone acetonide alone (Figure 4c), and burst release was reduced in the PLGA composite hydrogels by nearly half (Figure 4c).

Shelf-life Concept– Physical Integrity of TADDS After Lyophilization

To determine whether the TADDS could be prepared for long-term storage, triamcinolone acetonide-loaded devices were lyophilized, and then subsequently rehydrated and analyzed (Figure 5a). Lyophilization of the TADDS introduced cavities into its structure due to crystal formation during the freeze-drying process (Supplementary S6) and increased the modulus of the hydrogel, as evidenced by a higher linear modulus (Figure 5b). Following lyophilization, alginate-PAAm hydrogels have been shown to be 3–4x as porous in the dry state and exhibit elevated swelling potential^[19]. Therefore, in the present experiments, although the gels were hydrated to the same baseline wet weight, the lyophilized TADDS would be expected to be stiffer due to increased hydrogen bonding of the polymer network at this hydration level. We speculate that the two conditions may be more comparable once fully swollen. However, little impact was found in terms of triamcinolone acetonide release (Figure 5c).

Figure 5 |. The effects of lyophilization on the release of (triamcinolone acetonide) CORT from and the mechanics of the tough hydrogel drug delivery system (TADDS).

(a) Schematic of the experimental set-up for lyophilizing the TADDS to extend storage life. (b) Linear modulus of TADDS before (−) and after (+) lyophilization. Mean values are shown, and error bars are ± s.d. (n=3 gels/group), as analyzed by a Student’s t-test. *P<0.05. (c) Cumulative release curve for the lyophilized TADDS, and control, unlyophilized TADDS. Mean values are shown, and error bars are ±s.d. (n=3 gels/group), as analyzed by a two-way ANOVA (lyophilization (lyo) and time) with post hoc t-tests with Bonferroni corrections. *P<0.05, **P<0.01.

Shelf-life Concept– Bioactive Integrity of Triamcinolone Acetonide

The bioactivity of triamcinolone acetonide was assessed, in terms of reducing cytokine and chemokine levels secreted by cultured murine macrophages (RAW264.7 cells) in response to exposure by the endotoxin lipopolysaccharide (LPS). Following stimulation with LPS, the secretion of 15 cytokines and chemokines were upregulated (Figure 6a). With exposure to triamcinolone acetonide, the secretion of these cytokines was reduced (Figure 6b, 6c). IL-6 levels decreased the most (i.e., over half reduction) by adding triamcinolone acetonide to the culture medium following LPS stimulation (Figure 6b). A few members of the monocyte chemotactic protein (MCP) and macrophage inflammatory protein (MIP), and tissue inhibitors of metalloproteinases (TIMP) families were upregulated under LPS activation, but this effect was diminished when greater concentrations of triamcinolone acetonide were added (Figure 6a, 6d, 6e). Among the chemokines, triamcinolone acetonide stimulation had the greatest effect on RANTES/CCL-5 (part of the MCP subfamily of CC chemokines) (Figure 6d). TIMP-1 and a few MIPs, including MIP-1α, MIP-1β, and MIP-2, also exhibited changes in expression levels (Figure 6a, 6e). Similar results were observed when triamcinolone acetonide was released directly from the TADDS (Supplementary S7).

Figure 6 |. Expression levels of select cytokines and chemokines in response to LPS and/or different dosages of triamcinolone acetonide (CORT).

(a) Heat map of all cytokines and chemokines examined in the presence of LPS and/or CORT. (b-c) Individual plots comparing pixel density levels of select cytokines with the highest readout (i.e., IL-6 and IL-1ra) in response to LPS stimulation and/or CORT treatment of 0.002 or 0.02mg/ml. (d-e) Individual plots comparing levels of select chemokines with the highest readout (i.e., RANTES and MIP-1α) in response to LPS stimulation and/or CORT treatment of 0.002 or 0.02mg/ml.

Discussion

This study investigated how the TADDS could achieve both sustained and on-demand drug delivery through external stimulation, while maintaining high drug loading and tough mechanical properties. Cell viability was maintained during release at a drug loading of 100mg/ml^[8], and the TADDS exhibited extended release of triamcinolone acetonide as tracked through HFUS imaging. This approach to monitor release provides a facile means to track drug release out of the depot in vivo. This is similar to other imaging methods, such as fluorescence imaging to track fluorophores mixed in hydrogel formulations^[25] or magnetic resonance imaging^[26] described to track T2 relaxation times of drug depots in vivo, but does not rely on fluorescently labeled drug and is much easier to use. However, tissue penetration of drug cannot be measured with HFUS imaging. Lyophilization of this system may also enable an extended shelf life while maintaining bioactivity of triamcinolone acetonide, a strategy applied in previous hydrogel drug delivery systems^[6][27][28]. As alginate-PAAm hydrogels have been shown to exhibit elevated swelling potential^[19], we speculate that the elevated modulus is due to the fact that the gels were hydrated to the same baseline wet weight.

To confirm the anti-inflammatory functions of the triamcinolone acetonide -delivering TADDS, the bioactivity of triamcinolone acetonide was assessed on cultured macrophages in response to endotoxin exposure. As hypothesized, the cytokine assays revealed that the released triamcinolone acetonide was found to modulate cytokine production of cultured macrophages. The secretion levels of various cytokines and chemokines under LPS stimulation, which is a potent inducer of inflammation, decreased with increasing triamcinolone acetonide concentrations. This TADDS may serve several clinical applications. Rather than administrating multiple doses of triamcinolone acetonide, the TADDS may achieve more targeted and extended drug delivery to an inflamed tissue of interest. As several therapeutics have been shown to attenuate inflammatory joint damage and systemic inflammatory symptoms, the TADDS may be useful for targeted drug delivery that is advantageous for localized inflammatory diseases, such as monoarthritis or atopic dermatitis.

Stimuli-responsive drug delivery can titrate release to accelerate delivery on demand during a disease flare-up and minimize unnecessary release during low disease activity, prolonging the residence time of the drug depot and thus the duration of therapeutic efficacy^[13]. Stimuli-responsive release may be preferable for optimal drug presentation in vivo to match desired durations of administration^[29], but previous biomaterial platforms that demonstrate extended release often do not explore this option^[6][16], suffer from low drug loading^[13], or deteriorate under stimulation^[5]. The application of external stimulation (i.e., ultrasound and temperature) to the TADDS was found to accelerate drug release relative to passive release. While ultrasound has been used to trigger drug release through polymer degradation^[30][31], the ionically crosslinked alginate network within the TADDS was likely instead temporarily disrupted to promote the release of drug^[14]. The sensitivity of the TADDS to varying frequency levels and durations of ultrasound stimulation can be further explored in future studies. In addition to ultrasound, the TADDS achieved accelerated release through elevated temperatures. As a predominantly dissolution-controlled system, higher temperatures are expected to increase the dissolution of triamcinolone acetonide. Conversely, a decrease in temperature resulted in slowing release from the TADDS. As drug dissolution occurs more quickly at higher temperatures, the TADDS can exploit this fundamental phenomenon because of its greatly increased drug loading capacity of hydrophobic drugs. Future studies will investigate the combined effects of ultrasound and temperature change stimulations on triamcinolone acetonide release. Future work will also develop and evaluate application of these various stimuli in clinically relevant ways. The tissue region of interest harboring the TADDS may be locally stimulated with ultrasound pulses using commercially available ultrasound devices. Temperature modulation may be possible using warming pads to increase the dissolution and diffusion of triamcinolone acetonide, whereas a decrease in temperature may be controlled through local cooling with ice. The responsiveness of the TADDS to controlled temperature stimuli can be especially beneficial for tissues, such as the superficial tendon, that may be amenable to heat transfer applied topically^[32]. In contrast to other methods required for stimuli-responsive depots, both stimuli are also externally applied, readily available, and commonly used in medicine^{[33][34][35][36]}.

Cyclic compressive loading has recently been reported to substantially improve functional recovery of skeletal muscle using a pressure cuff^[37]. Thus, compression was also investigated as a physical stimulus that can be applied to the target tissue around the TADDS to accelerate drug release but was not found to have an effect. Previous studies showed that under an external force, the ionic bonds of the short-chain polymers of alginate hydrogels disassociate, releasing more drug^[38]. We speculate that since triamcinolone acetonide has a zero formal charge and thus limited charge interactions with the hydrogel matrix, compressive stimuli did not modulate drug release from the TADDS.

To extend release, triamcinolone acetonide aggregated with or encapsulated within particles, such as laponite or PLGA, helped prevent burst release. It has been shown that non-ionic drugs, such as corticosteroids, can be retained in laponite largely through hydrogen bonding involving hydroxyl and carbonyl groups^[39]. With a high ratio of laponite:drug, the laponite composite TADDS provided greater control over release kinetics, as was consistent with previous studies examining drug release from laponite composite alginate hydrogels^[6]. As compared to free drug, triamcinolone acetonide encapsulated in PLGA microspheres must first diffuse through the polymer matrix as the PLGA hydrolytically degrades in the release conditions^[40]. Release from PLGA is controlled by three major mechanisms: 1) diffusion through the matrix, 2) water-mediated transport processes, and 3) polymer erosion and hydrolysis^[40]. When incorporated into PLGA, release was significantly extended in agreement with previous studies delivering triamcinolone acetonide encapsulated in microparticle-loaded hydrogels^[41]. Many drug delivery systems suffer from relatively short release durations (<5 d) and the trade-off between mechanical toughness and drug loading capacity ^[6]. For instance, while lyophilized PLGA microparticles were loaded with corticosteroid particles and embedded into a bulk biodegradable hydrogel to prolong release, such an approach came at the cost of a low loading efficiency of 1.6%^[24]. In contrast, the structurally modified TADDS maintained high mechanical properties alongside sustained releasing capacities in all three alterations with high loading efficiency. Compared to other anti-inflammatory hydrogel delivery systems^{[20][21][23][22][6][13][24]}, the TADDS achieved both high drug loading and extended release (Supplementary S8), while simultaneously offering control over release through stimuli-responsiveness and structural modifications.

Conclusions

This work presents a hydrogel-based adhesive for controlled and stimuli-responsive release to address needs of current hydrogel systems. The TADDS tolerates large deformations, simultaneously achieves high drug loading, and can release at desired time points under physical stimuli. The TADDS can also incorporate drug sequestering particles to extend drug delivery without sacrificing the tough hydrogel’s mechanical strength. Taken together, this study presents a biomaterial platform capable of on-demand and extended release of corticosteroids for potential applications in alleviating inflammation.

Methods

Tough Gel Drug Delivery System Synthesis

The drug loaded hydrogels were synthesized in a facile two-step protocol. (1) Drug loading: an alginate solution was prepared by dissolving a 10mL solution of 2.2% (wt/vol) sodium alginate (combining a high-molecular-weight (MW = 200kDa) and low molecular-weight (MW = 30kDa) alginate at 1:1 ratio; LF20/40 and LF20/40–5Mrad, (Pronova, Novamatrix Norway)) and 13.5% (wt/vol) acrylamide (Sigma A8887, St. Louis MO) in HBSS (Gibco), 36µL of 2% (wt/vol) N,N’- methylenebis (acrylamide) (Sigma M7279, St. Louis MO), and 8µL of N,N,N′,N′- tetramethylethylenediamine (TEMED) (Sigma, T7024). Triamcinolone acetonide, with a molecular weight of 434.5g/mol (Sandoz, Germany; Toronto Research Chemicals, Canada), was dispersed in the alginate-AAM solution under vigorous vortexing at 1, 10, or 100mg/ml. For experiments testing the effect of synthetic laponite (Laponite, BYK Additives & Instruments), 0, 20, 40, or 80mg/ml of laponite was also added to the solution and mixed to form composite tough hydrogels loaded with corticosteroid and laponite nanoparticles (n=3 gels/group). (2) Crosslinking: the solution of alginate, corticosteroid, and optionally laponite was syringe-mixed with 226µL of 6.6% ammonium persulfate (Sigma A9164, St. Louis MO) and 191µL of 0.75M calcium sulfate dihydrate (Sigma 31 221, St. Louis MO) and cast into molds (80 × 15 × 1.5mm³) sealed on both sides with glass and left to crosslink for 24h. After 24h, tough gel strips were removed from molds and stored in sealed plastic bags at 4°C. (3) For experiments testing the effect of lyophilization: the hydrogel (diameter = 3mm, thickness = 0.75mm) was frozen at −20°C, and then lyophilized with a freeze-dryer (SP Scientific Freezemobile) for 2–3 days. After lyophilization, the dry scaffold of the composite hydrogel was stored at 4°C before usage (n=3 gels).

PLGA Microparticles Preparation, Characterization Methods, and Drug Encapsulation

For preparation of triamcinolone loaded poly(D,L-lactide-co-glycolide) (PLGA) microparticles a solid in oil in water (S/O/W) emulsion solvent evaporation method was used. After dissolving the polymer in dichloromethane (15.9% (w/w)), micronized drug (average size of 3 µm) was added to the polymer solution and vortexed for 3 minutes (11000 rpm) to make sure that the drug crystals are homogeneously distributed into the polymer solution. Next, the organic phase was emulsified with the water phase PVA 0.166% (pH 7.4, room temperature, saturated with dichloromethane) through an inline homogenization (4000 rpm). The formed emulsion was transferred to a 3L vessel under mild stirring (500 rpm) over night for microparticles hardening and solvent evaporation. next day solidified microparticles were collected, washed 3 times with distilled water and vacuum dried.

Microparticle size and size distributions were measured using Helos apparatus (Helos/BR Cuvette 50, Sympatec GmbH, Germany) (Table S1). The geometry of the 50 mL stationary cuvette produces a long optical path for the laser beam (over 20 mm) in the sample liquid, thereby enabling the optimum concentration for reliable measurements. About 10 mg of microparticles were dispersed in 50 mL distilled water containing three drops of 10% (v/v) Tween 20 and vortexed for about one minute before measurement and the solution stirred during measurement with a PTFE-coated magnetic stirrer and the same water was used as blank.

The microparticles morphology was studied by scanning electron microscopy (Scanning Electron Microscope Gemini SEM/SUPRA 40, Manufacturer = Carl Zeiss Microscopy GmbH, Germany) (Supplementary S9). Microparticles were glued on a sample holder using conductive carbon paint and coated with palladium.

10 mg of microparticles were accurately weighed in and dissolved in 10 mL dimethyl sulfoxide. Of the obtained solution 2.0 µL were injected into the HPLC for analysis (Model: Waters Acquity UPLC with DAD and QDA SQ-MSD, column type: CORTECS^™ C18+, 2.1 mm x 100 mm, 2.7 μm, eluent A: water + 4.76% isopropanol + 0.05% formic acid + 3.75 mM ammonium acetate, eluent B: isopropanol + 0.05 % formic acid, gradient: from 1 to 60% B in 8.4 minutes, from 60 to 98% B in 1.0 minutes, flow: 0.4 mL/min). The drug was detected by DAD at a specific wavelength of 238 nm and recording the UV-spectrum from 210 – 450 nm. The mass spectrum was recorded from m/z 150 – 1200 in the positive and negative ionization mode, and the loading capacity (LC) was reported as the amount of loaded drug substance divided by the weight of the microparticles. The drug substance eluted at a retention-time of 3.47 minutes. Lambda max of the UV-spectrum was 241.8 nm.

For gamma irradiation, the oxygen head space of vials was replaced by nitrogen using a Telstar freeze dryer at 0.035 mbar and 20.0 °C. The vials were capped, crimped and gamma irradiated using 25kGy±5.0kGy. Samples were kept at 15°C during gamma irradiation.

Dissolution-driven Release Study

Each hydrogel (diameter = 3mm, thickness = 0.75mm) was soaked in sink conditions (i.e., the necessary volume to solubilize the entire drug load within the hydrogel) (26.5mL of the release medium HBSS) at 37°C on a shaker. The release medium was changed daily to maintain the release volume and maximize drug flux out of the hydrogels and collected; aliquots of 1mL were used for liquid chromatography-mass spectrometry (LC-MS) analysis. The release profiles of triamcinolone acetonide corticosteroid were characterized with an LC-MS system (Agilent 1290/6140, Gradient method; SIM mode) equipped with an HPLC column (Agilant Zorbax Rx-C18; internal diameter 2.1mm, length 150mm). The two mobile phases were 0.02% formic acid and methanol. The triamcinolone acetonide concentration was quantified through integration of the characteristic peak.

Ultrasound Stimulation

Hydrogel samples (diameter = 3mm, thickness = 0.75mm) (n=3 gels/group) were placed in sink conditions of buffer (Ca²⁺ supplemented HBSS, Gibco #14175095) for three days prior to ultrasound sonication and after any burst release of drug. During this time, the buffer was sampled daily for future LCMS analysis and replaced with new buffer immediately after sampling. After the third sampling, the hydrogels underwent ultrasound stimulation for 3 min using Vibra-Cell VCX120 Sonicator from Sonics (intensity: 20 kHz; 8.04 mm² applicator; pulse 40%). During stimulation, the temperature was controlled and monitored by keeping the hydrogels in a secondary container containing water in a 37°C heat bath. The hydrogels were kept from touching the sonicator probe. Hydrogels were placed in a rocking water bath set at 37°C for an hour before the next ultrasound stimulation. The buffer was sampled before each stimulation round and replaced afterwards with fresh Ca²⁺ supplemented HBSS buffer.

Temperature Stimulation

Tough hydrogel samples (diameter = 3mm, thickness = 0.75mm) (n=3 gels/group) were placed in sink conditions of buffer (HBSS) for three days prior to temperature stimulation and after any burst release of drug. Hydrogels were kept on a rocking platform at 4°C. During this time, the buffer was sampled daily for future LCMS analysis and replaced with new buffer that was kept at 4°C immediately after sampling. On dates 3, 6, and 9, the hydrogels were placed at 37°C for 24 hours before sampling. After sampling, the buffer was replaced with fresh room temperature stored buffer, and the samples were placed back in 4°C.

Mechanical Testing

Hydrogel samples (diameter = 3mm, thickness = 0.75mm) were placed in sink conditions of buffer (HBSS) at 37°C for the set number of days (i.e., 0, 3 or 11 days) prior to compression tests (n= 4 gels/group). During this time, the buffer was replaced with new buffer daily. Day 0 hydrogels were prepared immediately before testing and pre-swollen in HBSS for a few minutes before characterization. A universal testing machine with a 10N load cell (ElectroForce 3200, TA Instruments, New Castle, DE) was used for the compression tests of the hydrogels. The order for Bose testing was randomized, and hydrogels were hydrated, sealed, and stored at room temperature. Hydrogel measurements were recorded immediately prior to testing. Each sample was loaded onto testing platens and bathed in 100µL of mineral oil (Ward’s Science, St Catharines ON, Canada) to prevent drying and pre-loaded with a force of 1g. According to these protocols, first, a ramp of 0.5mm/s was applied to compress the hydrogel to 5% strain followed by a 5min hold period to allow the hydrogel to relax. To acquire the dynamic modulus, the tangent loss, the percentage of relaxation, and the stress-relaxation halftime, a dynamic loading of 20 cycles at 1Hz was applied after the stress-relaxation. Then, a new ramp of 0.5mm/s was applied to reach a compression level of 10, 15, and 20% strain with the same frequency sweep parameters. A frequency sweep test from 0.1–10Hz was also applied to analyze the effect of frequency on the dynamic modulus. Compression test data as previously mentioned were analyzed using custom MATLAB code.

For tensile testing, pure shear tests were conducted to measure matrix toughness. Rectangular hydrogel samples (20 × 5 × 1.5mm³) (n=5 gels/group) were gripped between moving upper grips and fixed lower grips of an Instron machine (Instron 3342, 10N load cell) and tested under tension at 100mm min⁻¹. From the stress–strain curves, the maximum stretch, maximum stress, and toughness were calculated using custom MATLAB code.

For peel adhesion testing, ultrapure chitosan (2%) (54 046, Heppe Medical Chitosan, Halle, Germany) and coupling reagents (sulfated NHS (Thermofisher, PG82071) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (Sigma E6383)) (12mg/ml) were quickly vortexed^[42]. This mixture was then applied to the surface of the tough gel (25μL cm⁻²), which was then compressed to the porcine skin surface. Adhesion energy was measured using 180° peeling tests under uniaxial tension^[9]. A section of the tough adhesive (15 × 1.5 × 80mm³) was adhered to a skin sample (Sierra Medical) with one end open, forming a bilayer with an edge crack. The back of the tough adhesive was also bonded to a rigid polyethylene terephthalate film with cyanoacrylate (Krazy Glue), in order to limit deformation to the crack tip. A mechanical testing system (Instron 3342, 50N load cell) was used to apply unidirectional tension, while recording the force and the extension. The loading rate was kept constant at 100mm/min.

Controlled Drug Release by Compressive Stimulation

Hydrogel samples (diameter = 3mm, thickness = 0.75mm) (n=4 gels/group) were placed in sink conditions of buffer (HBSS) at 37°C for three days prior to compressive stimulation and after any burst release of drug. During this time, the buffer was sampled daily for LCMS analysis and replaced with new buffer immediately after sampling. A universal testing machine with a 10N load cell (ElectroForce 3200, TA Instruments, New Castle DE) was used for compressive stimulation of the hydrogels. Each hydrogel was loaded onto the Bose platform and bathed in 100µL of mineral oil (Ward’s Science, St Catharines ON, Canada) to be stimulated at 20% strain and sampled at 100Hz frequency for 5min. After stimulation, the hydrogels and surrounding buffers were collected and incubated at 37°C for an hour before the next stimulation. The control group was bathed in 100µL of buffer for 5min without stimulation. For both the experimental and control groups, the buffer was sampled before each stimulation round and replaced afterwards with fresh HBSS buffer stored at 37°C.

Cell Culture

RAW 264.7 macrophages were obtained from the American Type Culture Collection (Manassas, VA) and maintained in DMEM (30–2002, ATCC; R&D Systems, Minneapolis MN) supplemented with 10% FBS (American Type Culture Collection; Manassas VA), 100U/mL penicillin, and 100ug/mL streptomycin at 37°C in a 5% CO₂ incubator.

Measurement of Cytokine Production

To investigate the effects of corticosteroid on cytokine levels from LPS-treated cells, RAW 264.7 cells (2000 cells/ml) seeded into T75 flasks were treated with 1) 100ng/ml LPS (Sigma #067M4117V), 2) 100ng/ml LPS and a TADDS loaded with 100mg/ml of corticosteroid following release after 24h, or 3) 100ng/ml LPS and 0.002 or 0.02mg/ml of corticosteroid for 24h at 37°C in a 5% CO₂ incubator. Cell-free supernatants were collected and stored at −20°C until assayed for cytokine levels. The pixel density levels of TNF-α, IL-6, IL-1β, and 21 additional cytokines in the supernatants of RAW 264.7 cell cultures were determined using the Proteome Profiler Mouse Cytokine Array Kit (ARY006, R&D Systems), according to the manufacturer’s instructions. This kit is a membrane-based sandwich immunoassay that determines the relative levels of selected cytokines by mixing samples with a cocktail of biotinylated detection antibodies and then incubating with the array membrane, which is spotted in duplicate, with capture antibodies to specific target proteins^[43]. The assay, by design, is semi-quantitative and is not performed in replicate.

HFUS Imaging

High frequency ultrasound imaging (Vevo3100, FUJIFILM VisualSonics, Toronto ON, Canada; 50MHz transducer) was used to image the corticosteroid particles embedded inside the hydrogels (n=7 gels). Axial images (30µm axial resolution) were acquired that captured the hydrogel (window size: 13.5mm x 13.5mm). The hydrogel regions without drug particles appear hypoechoic (black) in the images. Images were exported through the VevoLAB software and the cross-sectional area and echogenicity evaluated using custom MATLAB code.

CryoSEM Imaging

The frozen samples were prepared in a BAL-TEC (Leica) MED 020 equipped with a Freeze Fracture chamber and VCT 100. A knife fracture was performed when the samples reached −150C. The samples were then sublimed/etched at −100C for 10 minutes, then sputter coated with 10 nm of PT/Pd 80:20. The samples were transferred to the SEM using the VCT100 transfer shuttle into the Zeiss Ultra 55 equipped with the VCT 100 Cryo stage. The samples were imaged at −140 C at 2kv.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism (v8.4.3; GraphPad Software; San Diego CA). Drug release data was normalized to set the plateau value of the cumulative curve to 100% release. Comparison of cumulative drug release between hydrogel samples was performed using a two-way ANOVA (time and treatment) with post hoc t-tests with Bonferroni corrections. Comparison of mechanical properties between hydrogel samples was analyzed using a one-way ANOVA. Mean values and error bars ± sd for each data set are shown. Normality was checked using the D’Agostino & Pearson test. An adjusted P-value of < 0.05 was considered significant.