Biocompatible Cyclodextrin–Cannabinoid Agar–Xanthan Gum Hydrogels for Controlled Delivery and Antimicrobial Soft-Tissue Biomedical Applications

Abstract

Agar/xanthan hydrogels incorporating hydroxypropyl-β-cyclodextrin (HP-β-CD) inclusion complexes of nonpsychoactive cannabinoids (cannabidiol, CBD; cannabinol, CBN) were developed as multifunctional biomaterials for controlled delivery and antimicrobial soft-tissue biomedical applications. Physicochemical characterization showed improved thermal stability, tunable hydrophilicity, and a porous network suitable for loading and release. In vitro release studies demonstrated sustained cannabinoid delivery over 72 h, following first-order and Fu–Kao kinetic models. The hydrogels exhibited selective antimicrobial activity against Staphylococcus aureus, while remaining inactive toward Escherichia coli. Cytocompatibility assays with human skin fibroblasts confirmed noncytotoxic behavior over 7 days for all hydrogel formulations tested. These results highlight that cyclodextrin–cannabinoid-loaded agar/xanthan hydrogels are promising biocompatible platforms for controlled delivery with antimicrobial activity, suitable for tissue engineering applications.

1. Introduction

Nowadays, there is a growing interest in the use of nonpsychoactive cannabinoids (NPCs) for cosmetic and biomedical purposes. Various innovative approaches have been initiated to enhance their application and address key challenges in practical use including prolonged release and stability. Numerous clinical studies have demonstrated the efficacy of NPCs, showing they are well tolerated by patients across inflammatory, pain, and dermatological indications without producing psychoactive effects. − Among nonpsychoactive cannabinoids, cannabidiol (CBD) and cannabinol (CBN) are structurally related but pharmacologically distinct, differing in oxidation state, lipophilicity, receptor engagement (CB1/CB2 and noncannabinoid targets), and reported profiles in terms of anti-inflammatory, neuroprotective, analgesic, and antibacterial effects. These divergences, together with their distinct pharmacokinetics and safety margins, mean that CBD and CBN are not interchangeable and may be suited for partially overlapping but different biomedical indications. − Moreover, rational combinations of NPCs in defined ratios may offer a route to tailor-made cannabinoid therapies with optimized efficacy and safety, highlighting the value of delivery systems that can accommodate more than one NPC in a controlled manner. In recent years, researchers have increasingly focused on utilizing NPCs to develop bioactive materials for biomedical applications in which the antimicrobial, antioxidant, anti-inflammatory, regenerative, and antithrombotic properties of NPCs are especially beneficial. , However, NPCs have a narrow therapeutic window, as they can be ineffective at low concentrations and potentially toxic at high concentrations. Additionally, current regulations regarding the use NPCs impose a very restrictive maximum recommended daily dosage, which is often not fully utilized due to improper methods of administration. Nevertheless, their utilization is limited by their physicochemical properties, rapid degradation, and restricted bioavailability.

To date, several methodologies have been investigated for the integration of NPCs into various delivery systems including inhalation through vaporization systems, transdermal patches, intranasal delivery via liquid suspensions, and oral or buccal sprays presented in soft capsules or oil droplets. − Among these delivery systems, those based on biocompatible polymers in the form of hydrogels are particularly interesting and promising for both therapeutic and commercial applications. , Moreover, to date, several technological challenges remain as open issues including but not limited to the spatial and temporal control of biomolecule release, managing interactions between the molecules and the matrix, bioactivity, biocompatibility, and controlling the degradability of the loading material. , To address these limitations, developing multifunctional hydrogels capable of adapting to complex physiological environments and effectively interacting with living systems remains a major technological challenge. In this context, various polymers have been successfully used as biocompatible carriers for developing hydrogels such as gelatin, cellulose, alginate, collagen, chitosan, agar, xanthan gum, as well as other synthetic biocompatible polymers, such as poly­(vinyl alcohol) (PVA), poly­(glyceryl methacrylate) (PGMA), polyacrylamide (PAM), polyethylene glycol (PEG), polyacrylic acid (PAA), polyvinylpyrrolidone (PVP), and poly­(ε-caprolactone) (PCL). , It is worth mentioning that among the available hydrogel-forming polymers, agar and xanthan gum, alone or in combination, are particularly attractive for biomedical and topical applications because they provide complementary biopharmaceutical advantages. , Indeed, agar forms thermally reversible gels with ideal structural properties and dimensional stability, which can act as mechanically stable scaffolds for localized drug delivery and wound coverage. For its part, xanthan gum exhibits high water-holding capacity, mucoadhesive behavior, and shear-thinning rheology, features that favor prolonged hydration of the tissue surface, intimate contact with the skin or mucosa, and easy spreading under mild shear. Thus, when combined, agar–xanthan gum networks would produce highly hydrated, mechanically stable matrices that can sustain drug release but simultaneously maintain a moist environment and have been widely reported as biocompatible for dermatological, mucosal, and regenerative applications. , To the best of our knowledge, there are still very few examples that combine preformed HPβ-CD inclusion complexes of multiple cannabinoids (beyond CBD alone) with simple, food-grade polysaccharide hydrogels and evaluate them as soft-tissue biomedical platforms. , In particular, no previous study has reported agar–xanthan gum hydrogels incorporating HPβ-CD complexes of cannabidiol (CBD) and cannabinol (CBN) and systematically examined their controlled delivery, antibacterial performance, and in vitro biocompatibility. , Also note that free CBD and CBN and their HPβ-CD inclusion complexes have been previously characterized in terms of solubility enhancement and biocompatibility in aqueous environments, so the remaining challenge is to understand how their incorporation into an agar–xanthan hydrogel matrix modulates release, antibacterial performance, and cytocompatibility in soft-tissue conditions. ,

Based on these considerations, in this work, hydroxypropyl-β-cyclodextrin (HPβ-CD) is exploited as a supramolecular carrier to solubilize highly hydrophobic cannabinoids and stably incorporate them into soft agar–xanthan gum hydrogels, enabling controlled delivery and antibacterial activity while preserving biocompatibility. Specifically, we develop and investigate new biocompatible hydrogels formulated from agar and xanthan gum and enhanced by HPβ-CD inclusion complexes (CICs) containing two nonpsychotropic cannabinoids (NPCs), cannabidiol (CBD), and cannabinol (CBN), thus going beyond the CBD-only focus of most previous studies. The physicochemical properties of these hydrogels are characterized together with their ability to efficiently and sustainedly release cannabinoids over extended periods, and their selective antibacterial activity against Staphylococcus aureus and Escherichia coli is evaluated. In addition, the in vitro biocompatibility of the materials is assessed using human skin fibroblast cultures to confirm their safety and support their potential for soft-tissue biomedical applications.

2. Materials and Methods

2.1. Materials

Agar (cas number 9002-18-0), xanthan gum from Xanthomonas campestris (cas number 11138-66-2), and glycerol (cas number 56-81-5) were acquired from Sigma-Aldrich (Madrid, Spain). Modified cyclodextrin 2-hydroxypropyl-β-cyclodextrin: C₄₂H₇₀–nO₃₅·(C₃H_7O)n, average M _W ∼ 1546 g/mol, purity ≥98% were obtained from AraChem (Eindhoven, The Netherlands). Cannabidiol and cannabinol certified reference materials (Supelco Cerilliant, C-045 & C-046 respectively) were obtained in methanol (1 mg/mL) from Merck (Darmstadt, Germany). Cannabidiol and cannabinol isolate crystal bulk employed for complexation was purchased from Ballinvestors groups S.L. (Murcia, Spain). HPLC grade water, methanol, and acetonitrile were purchased from Sigma-Aldrich (Madrid, Spain).

2.2. Methods

2.2.1. Fabrication of Hydrogels H-CD-NPCs

The nonpsychoactive cannabinoid–cyclodextrin (CD-NPCs) complexes utilized in this study were prepared according to a previously reported procedure. Briefly, hydroxypropyl-β-cyclodextrin (HP-β-CD) was dissolved in distilled water to a final concentration of 100 mM. Then, cannabidiol (CBD) and cannabinol (CBN) were added at its maximum inclusion capacity, and the mixture was sonicated in an ultrasonic bath until a clear solution was achieved. The resulting solution was further processed using a Buchi B-290 mini spray dryer equipped with a 1.5 mm nozzle cap and a 0.7 mm internal diameter nozzle (operating with 100% inlet air flow, 10% pump flow rate, 360 L/h compressed air, at an inlet temperature of 160 °C, and an outlet temperature of 70–80 °C). The dried powder was collected and stored at room temperature (25 °C) until further use. For this study, the inclusion complex (CIC) containing NPCs was integrated directly into subsequent hydrogel formulations.

Hydrogels were synthesized by combining xanthan gum, agar, and cyclodextrin complexes, either with or without nonpsychoactive cannabinoids (CD-NPCs). Specifically, four types of hydrogels were prepared as follows: (1) a nonloaded control hydrogel [H], (2) a hydrogel containing solely cyclodextrin [H-CD], (3) a hydrogel containing CD complexes with CBD [H-CD-CBD], and (4) a hydrogel containing CD complexes with CBN [H-CD-CBN]. Typically, to prepare 20 g of hydrogel, 0.4 g (2 wt %) of agar was dissolved in 10 g of deionized water by continuous stirring at 90 °C for 15 min. In parallel, 0.4 g (2 wt %) of xanthan gum was dissolved in a separate beaker with 9.2 g of deionized water at room temperature under constant stirring. The two solutions were then combined in a new vial and placed in a water bath at 90 °C for 15 min with moderate stirring. The homogenized mixture was subsequently poured into a 100 mm Petri dish to cool down until room temperature, after which it was cut into 8 mm diameter discs using a biopsy punch. For the CD- and CD-NPC-loaded hydrogel formulations (H-CD, H-CD-CBD, and H-CD-CBN), the same general procedure was followed, with the addition of 3 g (15 wt %) of either HP-β-CD, CD-CBD, or CD-CBN powders, which were gently homogenized into the xanthan gum solution prior to mixing with the agar solution. The preparation strategy for the cannabinoid-loaded hydrogels is summarized in Figure , which depicts the workflow formation of HP-β-CD/cannabinoid inclusion complexes followed by their incorporation into an agar–xanthan gum matrix through sequential dissolution, mixing, and thermal treatment steps to obtain standardized hydrogel discs.

1.

Schematic representation of cannabinoid encapsulation in hydroxypropyl-β-cyclodextrin (HP-β-CD) and subsequent preparation of agar–xanthan gum hydrogels containing HP-β-CD/cannabinoid inclusion complexes, including dissolution steps, casting of hydrogel and final post processing into 8 mm disc pieces.

For the cannabinoid-loaded hydrogels, the actual cannabinoid content per disc was estimated from the known drug loading of the HP-β-CD/NPC inclusion complexes. The detailed process yield (PY), encapsulation efficiency (EE), and drug loading (DL) of the HP-β-CD/CBD spray-dried powder (DL = 0.16% w/w CBD; 1.6 mg CBD/g complex) and for the HP-β-CD/CBN complex (DL = 0.10 ± 0.01% w/w CBN; 1.0 mg CBN/g complex) are reported in our previous literature studies. − In the present study, 20 g of hydrogel precursor were cast into a 100 mm Petri dish, yielding discs of 3 mm thickness, and 8 mm biopsy punches were used to obtain individual samples with an estimated mass of ∼0.35 g. Considering the 15 wt % content of HP-β-CD/cannabinoid inclusion complexes in the hydrogel formulation and the measured drug loading of the powders, this corresponds to a cannabinoid contents of ∼90 μg CBD per H-CD-CBD disc and ∼60 μg CBN per H-CD-CBN disc.

2.2.2. Scanning Electron Microscopy: Morphology and Microstructure

The characterization of the morphology and microstructure of the NPC-cyclodextrin complex hydrogels was evaluated using scanning electron microscopy (SEM). Before observation with SEM, the samples were freeze-dried. Further, samples were coated with a ∼5 nm thick platinum conductive layer (99.99% purity) using a sputter evaporator (Biorad-Polaron). Finally, the microstructure of the hydrogels was evaluated by using a Teneo FEI (Thermo Fisher Scientific, Waltham, MA, USA) operating at 2 kV accelerating voltage. Images of the samples were collected at magnification ranging from 100× to 3000×.

2.2.3. Contact Angle

Contact angle measurements were carried out by using the sessile drop method. For each hydrogel sample (H-CD, H-CD-CBD, and H-CD-CBN), three drops of distilled water were placed randomly on the surface. The contact angle of each drop was recorded, and the mean and standard deviation were calculated for each liquid and each sample.

2.2.4. FTIR Analysis

FTIR spectra were recorded using a PerkinElmer spectrometer (Norwalk, CT, USA) in the attenuated total reflection (ATR) mode. The instrument was equipped with a universal ATR accessory, featuring a diamond crystal and ZnSe focusing elements. Measurements were performed using a single reflection at a 45° incident angle. The analysis was conducted on samples at room temperature and ambient humidity conditions. Each spectrum was averaged from 32 scans, covering the range of 4000 to 650 cm^–1, with a spectral resolution of 4 cm^–1.

2.2.5. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) was performed to assess the thermal stability of the hydrogels, using a PerkinElmer Pyris Diamond TG-DTA Instruments in a nitrogen atmosphere to prevent oxidation. Approximatively 6 mg of sample were heated from 25 to 800 °C at a rate of 10 °C/min, with a nominal gas flow rate of 30 mL/min. Degradation peak temperatures (T _max) were determined from the first derivative thermogravimetric (DTG) curves.

2.2.6. Differential Scanning Calorimetry

The thermal properties of hydrogels were investigated using a Q2000 T zero differential scanning calorimeter (DSC) from TA Instruments (New Castle, DE, USA), featuring a liquid nitrogen accessory for rapid cooling. Approximately 5 mg of each sample were equilibrated at 25 °C before being heated to 300 °C at a rate of 10 °C/min. Measurements were conducted under a nitrogen flow of 30 mL/min, and the glass transition temperature (T _g) was determined using the first derivative method.

2.2.7. Wide-Angle X-ray Diffraction

Wide-angle X-ray diffraction (WAXD) patterns were obtained using a PANalytical PW3040/60 X’Pert PRO MPD diffractometer operating at 45 kV and 40 mA in a Bragg–Brentano geometry. The X-ray source was a high-power ceramic tube (PW3373/10 LFF) with a Cu anode. WAXD profiles were recorded with Ni-filtered Cu Kα radiation (λ = 0.15418 nm), employing a continuous scan rate of 0.04°/s and a step size of 0.0167°, over a 2θ range of 5° to 65°.

2.2.8. Swelling

For hydrogel swelling measurements, samples were weighed (W ₀), subsequently immersed in PBS and incubated at 37 °C for different time periods (0, 3, 7, 24, and 72 h). At each time period, samples were removed, excess liquid blotted away and reweighed to determine the weight after immersion (W _f). This procedure was done in triplicate (n = 3) for each time point to ensure accuracy and reliability. Swelling percentage (S %) was calculated using eq as follows

$$\mathrm{S}\%=((W_{\mathrm{f}}-W_0)/W_0)\times 100$$ 1

2.2.9. Protein Adhesion in Hydrogels

Protein adhesion assays were conducted using the Coomassie Brilliant Blue assay to evaluate the protein adsorption on material surfaces. Hydrogels were immersed in the fetal bovine serum (FBS) for 30 min for protein adhesion, then the FBS was removed, and the hydrogels were washed twice with PBS to eliminate nonadherent proteins. To fix the protein coating, hydrogels were immersed in a 4% paraformaldehyde (PFA) solution in PBS for 15 min, followed by 30 min in Coomassie staining solution. The hydrogels were washed twice with a destaining solution (methanol, acetic acid, and distilled water in a 40:10:50 ratio) for 30 min. The eluted solution was collected, and the absorbance was measured at 590 nm using a SpectraMax iD3 spectrophotometer. Control hydrogels without FBS treatment were also stained with Coomassie, and samples without hydrogels were treated similarly.

2.2.10. In Vitro Biodegradation and Remaining Mass (%)

In vitro hydrogel biodegradation tests were performed by monitoring sample mass over different time periods (1, 7, 14, 28, and 60 days). Hydrogels were weighed (initial dry weight, W ₀), immersed in PBS, and incubated at 37 °C. At each time point, samples were removed, dried in an oven at 37 °C for 24 h, and then reweighed (W _d) to obtain the dry weight after incubation. This procedure was performed in triplicate (n = 3) for each time point to ensure accuracy and reliability. Remaining mass (%) was calculated using eq

$$\mathrm{R}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}(\%)=(W_{\mathrm{d}}/W_0)\times 100$$ 2

where higher remaining mass values indicate lower biodegradation and greater hydrogel stability under the tested conditions. For completeness, weight loss (%) can be expressed as eq

$$\mathrm{W}\mathrm{L}\%=100-\mathrm{R}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}(\%)$$ 3

2.2.11. In Vitro Cumulative Release

Cumulative release experiments were performed using phosphate-buffered saline (PBS, pH ∼ 7.4) and water as the release medium. Hydrogel samples H-CD-CBD and H-CD-CBN were incubated in 10 mL PBS and water at 37 °C. At predetermined intervals, 2 mL aliquots were withdrawn for analysis and replaced with fresh PBS/water to maintain sink conditions. For the release quantification, a high-performance liquid chromatography (HPLC) protocol was employed using an Agilent 1200 system, equipped with an InfinityLab Poroshell 120 EC-C18 column (2.1 × 150 mm, 2.7 μm) maintained at 30 °C. Absorption spectra from confirmed maximum absorbance at 210 nm for CBD and 230 nm for CBN in agreement with literature studies. ,, Cumulative release was calculated using eq , where “Abs” is the measured absorbance intensity, CR represents the cumulative amount of released compound (mg), m and b are the slope and intercept obtained from the linear fit of the calibration curve prepared beforehand for each CNP complex. R _i indicates the amount of compound (mg/mL) removed (discarded) during each previous sampling interval and replaced with fresh PBS/water. Subscript i refers to the specific time interval analyzed. The prefactors 10 and 2 correspond to the total volume used in the release experiments and the volume removed at each time interval, respectively.

$$CR(\mathrm{m}\mathrm{g})=10((\mathrm{A}\mathrm{b}\mathrm{s}-b)/m)+2\sum _{i=1}^{n-1}Ri$$ 4

Calibration curves for cannabinoids were established over the concentration range of 0.005–0.2 mg/mL and yielded linear equations with R ² values exceeding 0.99. , To facilitate comparison between media and formulations, cumulative release was then expressed as a percentage of the initial drug loading (DL) per disc ( M ₀ ); thus, the cumulative release (%) at time t _n was calculated as

$$\mathrm{C}\mathrm{u}\mathrm{m}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e}(\%)=[\mathrm{C}\mathrm{R}({\mathit{t}}_{\{n\}})/{\mathit{M}}_{\{0\}}]\times 100$$ 5

where M ₀ ≈ 90 μg for CBD and M ₀ ≈ 60 μg for CBN, as determined from the known drug loading of the HP-β-CD/NPC inclusion complexes and the hydrogel composition. Higher release (%) values thus indicate a larger fraction of the loaded cannabinoid released from the hydrogels into the surrounding medium. −

The HPLC detection wavelengths were then set at 210 nm for CBD and 230 nm for CBN, subsequently sample volumes of 5 μL were injected with needle washing between runs. The elution was performed using a binary gradient: mobile phase A (0.5% acetic acid in water) and mobile phase B (acetonitrile). The protocol began at 67% B for 20 min, increased linearly to 95% B in 1 min, held at 95% B for 4 min, and returned to 67% B over the final 5 min, with a constant flow rate of 0.3 mL/min. Chromatographic data were acquired and processed using Agilent Chemstation software. This optimized protocol enabled the reliable detection and quantification of CBD and CBN within the 0.001–1 mg/mL range, demonstrating robust linearity (R ² > 0.99) across all calibration standards. It is worth mentioning that in all these release experiments, the species in solution is the hydrophilic cyclodextrin–cannabinoid inclusion complex (CD/CBD and CD/CBN), not the free hydrophobic cannabinoid (CBD or CBN). Indeed, at each sampling time, 2 mL of the release medium containing the CD/CBD or CD/CBN complexes are withdrawn and replaced with fresh PBS/water, thereby limiting accumulation or saturation of the released complexes in the medium. The cannabinoid content is quantified after dilution with 80% methanol (v/v) in a validated HPLC-DAD method, which fully dissociates the inclusion complexes. Therefore, the classical sink condition concept for poorly soluble free drugs does not strictly apply to this system.

2.2.12. Antimicrobial Activity

We employed two complementary approaches to assess antimicrobial activity and bacterial growth of the materials: (i) the disk diffusion assay, using hydrogel discs placed on inoculated plates to evaluate inhibition zones, and (ii) growth curve analysis, monitoring bacterial proliferation in the liquid culture in the presence of hydrogels using optical density measurements. The antimicrobial activity of the hydrogels was evaluated using standard reference strains, including Gram-positive S. aureus (ATCC 338662) and Gram-negative E. coli (ATCC 25922).

For the first method, Petri dishes containing Mueller–Hinton agar were utilized. Bacterial suspensions of E. coli and S. aureus were prepared and adjusted to a density of 0.25 McFarland standard. Then 100 μL of both bacteria suspensions were dispensed onto the agar plates with a sterile pipet tip and evenly spread over the agar surface. Hydrogel pieces were gently placed onto the Mueller–Hinton agar plates previously inoculated with the bacterial suspension, ensuring full contact between the hydrogel surface and the agar. Before placement, the samples were presoaked in a sterile Luria–Bertani (LB) broth to keep them hydrated and to facilitate drug diffusion throughout the hydrogel. Ciprofloxacin and penicillin discs were used as the control. The plates were incubated at 37 °C for 24 h, after which the results were documented with macroscopic photographs to measure inhibition zones and evaluate antimicrobial activity. It is worth mentioning that for the disk diffusion assays, each hydrogel formulation was tested with at least n = 3 technical replicates per condition on each experimental day. In addition, the experiments were repeated independently on three different weeks using fresh bacterial cultures and hydrogel batches (biological replicates), consistently yielding the same qualitative outcome. After incubation, the hydrogel discs were carefully lifted from the agar plates and the underlying area that had been in direct contact with the material was examined. In this region, heterogeneous clear/opaque patterns were observed immediately beneath the discs, indicating localized antimicrobial activity but making it difficult to define a single, well-delimited inhibition zone diameter by conventional disk diffusion criteria. Due to these constraints and the small halo sizes obtained under the tested bacterial concentrations, disk diffusion was used primarily as a qualitative confirmation of activity.

In parallel, bacterial growth was monitored by optical density measurements. The same bacterial suspensions (0.25 McF) employed for the Petri dish experiments were inoculated into tubes containing the various hydrogel samples, alongside a positive control (the LB broth with no added material) and negative controls (cultures containing ciprofloxacin or penicillin). The cultures were incubated with constant shaking at 37 °C for 24 h. Bacterial growth kinetics were monitored by recording optical density at 600 nm (OD600) at regular time intervals using a Plate Reader Spectrometer. Additionally, macroscopic photographs were taken at the start (t = 0 h) and end (t = 24 h) of incubation to visually assess turbidity as a qualitative indicator of growth.

For both the agar disk-diffusion and liquid culture assays, hydrogels were prepared by casting 20 g of precursors into 100 mm Petri dishes to obtain ∼3 mm thick sheets, from which 8 mm discs (≈0.35 g) were punched. Considering the 15 wt % content of HPβCD–cannabinoid inclusion complexes in the formulation and the measured loading of the spray-dried powders, these discs contained approximately 90 μg CBD (H-CD-CBD) or 60 μg CBN (H-CD-CBN). The same batches of hydrogels and disc preparations were used for both agar-based and liquid-culture antibacterial experiments, ensuring consistent composition and cannabinoid loading across all assays.

2.2.13. Cell Culture

In vitro cell culturing was performed using ATCC-CRL-2522 BJ, human skin fibroblast human (Homo sapiens) batch no: 70046146. Typically, 8 mm diameter and 3 mm thick hydrogel samples were transferred into a 48-well plate and were sterilized with UV light. Each hydrogel was conditioned for 1 h in complete culture media (DMEM supplemented with 10% (w/v) FBS, 1% (v/v) penicillin–streptomycin, and 2 mM GlutaMAX) at 37 °C. Fibroblast cells were then seeded onto the hydrogel at a concentration of 2.5 × 10⁴ cells in 48 well plates and incubated at 37 °C in a 5% CO₂ humidified atmosphere. The medium was refreshed every 3 days.

2.2.13.1. Cell Cytotoxicity

Mitochondrial activity of seeded fibroblast cells treated with the different hydrogel films was assessed through the enzymatic conversion of tetrazolium salt (MTT cell growth KIT, Sigma-Aldrich, Germany) after 1, 3, and 7 days of cell culture. Typically, fibroblast cells were seeded onto the hydrogel at a concentration of 2.5 × 10⁴ cells in 48 well plates and incubated at 37 °C in a 5% CO₂. After reaching each time period, culture media was removed from samples, and subsequently, 500 μL of MTT solution (5 mg/mL) was added to the cells. Then, well plates were incubated at 37 °C for 3 h, followed by the MTT remotion and rinsing, and 500 μL DMSO was added to dissolve the formazan crystals. Subsequently, 200 μL of supernatant from each sample was transferred into a well of a 96 well-plate and measured the absorbance at 570 nm with a Multimode Microplate Reader (Thermo Scientific). Cells without the presence of hydrogel were used as a cell control.

2.2.13.2. Cell Proliferation

Fibroblast cells were cultured for 1, 3, and 7 days in the presence of hydrogels at a concentration of 2.5 × 10⁴ cells in 48 well plates and incubated at 37 °C in a 5% CO₂. At each time point, cells were rinsed with PBS and incubated with 200 μL of 10% Alamar Blue solution (Yeasen, China) for 4 h. Subsequently, 100 μL were transferred to a dark 96-well-plate, and fluorescence was measured with a plate reader using an excitation/emission wavelength of 530/590 nm. Cells without the presence of hydrogel were used as a cell control.

2.2.13.3. Cell Loading Capacity

The cell loading capacity of the hydrogel was evaluated by seeding fibroblast at a density of 5 × 10⁵ cells in 24-well plates and incubated at 37 °C in a 5% CO₂ for 6 h. At the end of the experiment, the cells were harvested using 500 μL of 0.25% trypsin and counted with an automated cell counter (Invitrogen). The percentage of cell loading was calculated based on a control group of treated cells without the hydrogel and reported as the percentage of cell loading.

2.2.13.4. Histological Evaluation of Cell–Hydrogel Interactions (Hematoxylin and Eosin and Crystal Violet Staining)

After 1, 3, and 7 days in culture, the gels were washed with 1× PBS and fixed in 4% paraformaldehyde for 15 min at room temperature, followed by rinsing with Milli-Q water. For Hematoxylin andEosin (H&E), samples were stained with hematoxylin for 3 min, rinsed with water for 3 min, and stained with eosin for another 3 min. Then samples were rinsed in 70%, 96%, and 100% ethanol for 3 min each. For crystal violet (CV) staining, the same initial steps were followed, but after washing with Milli-Q water, samples were stained with 0.4% crystal violet for 3 min before rinsing with water again. Images were taken with a Leica ICC50HD camera mounted on a Leica DM500 microscope.

2.2.13.5. Statistics

All experiments were conducted in triplicate and results are expressed as mean ± standard error of the mean. Statistical analyses were accomplished by two-way ANOVA with multiple comparison tests after 1, 3, and 7 days of incubation. p-values < 0.05 were considered statistically significant. Statistical analysis was performed using the GraphPad Prism software (version 10.0; GraphPad Software Inc., USA).

3. Results and Discussion

3.1. Characterization of the Morphology and Structure of the Raw Materials and the Hydrogels

The SEM results are shown in Figures and for the raw materials and the corresponding hydrogels, respectively. Figure presents representative micrographs of CD powder, CBD powder, CBN powder, and their respective inclusion complexes (CD–CBD and CD–CBN), highlighting notable differences in particle morphology and surface characteristics among the pure substances and the cyclodextrin-based complexes. CD particles typically exhibit an egg-like morphology with relatively smooth surfaces, whereas CBD and CBN powders display the characteristic spike-like structures associated with cellulose-type domains. In the inclusion complexes (CD–CBD and CD–CBN), the dominant morphological features resemble those of CD, indicating that the CD host matrix governs the overall particle architecture, in good agreement with previous reports on cyclodextrin inclusion complexes and consistent with an efficient host–guest interaction that is expected to influence drug loading and release behavior.

2.

Representative SEM micrographs of: (a) CD powder; (b) CBD powder; (c) CBN powder; (d) CD–CBD inclusion complex powder; and (e) CD–CBN inclusion complex powder. Scale bars: (a–c) 400 μm and (d,e) 200 μm.

3.

Representative SEM images of hydrogel samples: (a) H, (b) H-CD, (c) H-CD-CBD, and (d) H-CD-CBN. The bar scale is 400 μm.

Figure shows representative SEM images and highlights the microstructural features of the different hydrogels. The control hydrogel (Figure a), formulated without a cyclodextrin or cyclodextrin-inclusion complex, exhibits a comparatively compact structure with densely packed walls and smaller, less interconnected surface cavities, indicative of a more homogeneous polymer network. − In contrast, the hydrogels incorporating CD, CD–CBD, or CD–CBN (Figure b–d) display a markedly rougher and more heterogeneous architecture with interconnected porosity, with micro- and macroporosities ranging from 5 mm to 400 μm, characterized by lamellar domains and irregular fissures that define an open, porous network. Although the overall morphology of the CD-containing hydrogels is broadly similar, a gradual increase in the apparent pore size and the extent of open voids can be observed as CD and then the CD–CBD and CD–CBN inclusion complexes are introduced. , This progressive enlargement of pores suggests that the presence of cyclodextrin and its inclusion complexes perturbs the packing of the polymer chains, promoting the formation of more expanded microdomains within the hydrogel matrix. Such microstructural changes are expected to facilitate greater fluid penetration and mass transport through the scaffold, which may in turn influence swelling behavior, drug diffusion kinetics, and cell–material interactions.

3.2. Contact Angle Measurement (Hydrophobicity)

Figure presents representative images of water droplets on hydrogels loaded with cyclodextrin–cannabinoid complexes (H, H-CD, H-CD-CBD, and H-CD-CBN). From these images, the average contact angle measurements (ACAMs) and their standard deviations (SDs) were determined using water as the probing liquid. The ACAM values indicate a similar wettability for H, H-CD, and H-CD-CBD, with average contact angles of 50.83 ± 2.49°, 50.80 ± 2.82°, and 46.24 ± 1.47°, respectively, whereas the H-CD-CBN hydrogel shows a slightly lower value of 41.36 ± 3.87°, consistent with a modest increase in surface hydrophilicity for this formulation. In all cases, the contact angles remain below 90°, confirming the hydrophilic character of the hydrogel surfaces and agrees with the former reported findings, a feature generally regarded as favorable for biomaterial applications due to improved aqueous compatibility and potential support for cell adhesion. ,

4.

Representative images of water droplets on hydrogels samples (a) H, (b) H-CD, (c) H-CD-CBD, and (d) H-CD-CBN showing the variation in the contact angle.

3.3. FTIR-ATR Spectroscopy

FTIR-ATR spectroscopy of the hydrogels was performed to investigate their structural characteristics and evaluate potential interactions between the different components (Figure a). A neat agar spectrum typically shows a broad absorption band at 3312 cm^–1 attributed to –OH group stretching, and a peak centered at 2894 cm^–1 ascribed to the C–H stretching of the ring methine hydrogen bonds. The band around 1644 cm^–1 corresponds to the stretching vibration of the conjugated peptide bond. Additionally, the peaks observed at 1367 cm^–1 and 1150 cm^–1 suggested the presence of ester sulfate groups. The intense band recorded at 1040 cm^–1 is common to all polysaccharides and is primarily attributed to the coupling of C–O and C–C stretching modes with C–O–H bending modes. In contrast, the strong absorbance at 930 cm^–1 is associated with the vibration of the C–O–C bridge in 3,6-anhydrogalactose. This band is highly characteristic of the agar family, as other polysaccharides found in red algae, such as cellulose and xylans, do not exhibit absorption at these specific wavenumbers in their infrared spectra. Finally, the band around 887 cm^–1 is related to the –CH stretching vibration of the β-d-galactose backbone. , For its part, xanthan spectrum’s typical features exhibited a broad absorption peak at 3274 cm^‑1 related to –OH stretching vibration and a band corresponding to the asymmetric stretching vibration of aliphatic –CH at 2911 cm^–1. The peaks around 1724 cm^–1 and 1607 cm^–1 corresponded to symmetric and asymmetric stretching vibrations of carbonyl groups found in esters, carboxylic acids, aldehydes, and ketones, respectively. Additionally, the absorption band at 1410 cm^–1 was associated with the symmetric stretching vibration of the carboxylate groups, whereas peaks at approximatively 1243 cm^–1 and 1150 cm^–1 were attributed to stretching vibration of –C–O and C–O–C of cyclic ethers of carboxylic acid salt, and the asymmetric stretching of C–O–C, respectively. At 1076 cm^–1, a band related to the C–O stretching vibrations in hydroxyl (–OH) groups appeared along with an intense band at 1016 cm^–1, which corresponds to C–O stretching in alcohol groups and is a characteristic peak for polysaccharides, highlighting the glycosidic linkages. ,, Both spectra are shown in Figure SI1a.

5.

Physico-chemical characterization of H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels: (a) FTIR-ATR spectra in the region between 1800 and 650 cm^–1; (b) TGA thermograms; (c) DSC thermograms; and (d) wide-angle X-ray diffraction patterns.

The spectrum of the agar/xanthan hydrogel (H) exhibited characteristic absorption bands of both polysaccharides. There were no significant changes compared with the neat polymers, indicating that no major derivatization occurred during gelation in the binary system. Since the bands of agar and xanthan gum overlap significantly in this binary system, a qualitative evaluation of potential changes in band positions and intensities indicating physical interactions between the two polysaccharides was conducted using spectrum subtraction (i.e., H hydrogel minus xanthan). The resulting subtraction spectrum is shown in Figure SI1b. Comparing this spectrum with that corresponding to agar revealed a significant reduction in the intensity of the agar absorption peak at 1040 cm^–1, along with a shift to a lower wavenumber, approximately 1015 cm^–1. The vibrational frequency of a functional group is mainly dependent on the force constant and the mass of the bonded atoms. The involvement of these groups in physical interactions can modify their electron distribution, thereby decreasing the force constant and shifting the frequencies to lower values. Additionally, a significant reduction in the intensity of the band at 887 cm^–1 was observed in the binary hydrogel. Most of the xanthan peaks exhibited no significant variation in the binary hydrogel, except for the peak associated with the CO asymmetric stretching. This peak showed a slight shift from 1607 cm^–1 to 1603 cm^–1, suggesting the involvement of these functional groups in hydrogen-bonding interactions with agar chains. When CD, CD-CBD, or CD-CBN complexes were added to the hydrogel formulation, the above-mentioned peak nearly disappeared. A similar variation was observed for the agar bands at 930 cm^–1 and 887 cm^–1. The intensity reduction of these adsorption bands is likely attributable to the enhanced physical interactions involving the corresponding groups, mediated by the presence of cyclodextrin.

3.4. Thermogravimetric Analysis

TGA assessed the thermal stability of hydrogel samples, and the results are shown in Figure b and listed Table . The thermal degradation of the investigated samples occurred through a multistep process. A first slight weight loss of comparable magnitude (12–15%) was observed for all tested hydrogels, in the temperature range between 55 and 85 °C, due to the evaporation of moisture from the agar, xanthan gum, and hydrogel matrices. The main thermal degradation for all samples occurred between 200 and 400 °C. For the binary hydrogel without cyclodextrin (H), the first thermal decomposition was observed at 298 °C, resulting in a weight loss of around 50% and attributed to the breakdown of the polymer backbone. A subsequent phase of degradation began at 423 °C, peaking at 501 °C, due to the presence of more thermally stable residues. The incorporation of CD, CD-CBD, and CD-CBN complexes into the hydrogel resulted in a shift to a higher degradation temperature (peak at 330 and 560 °C) confirming the establishment of physical interactions that enhance thermal stability, as highlighted by FTIR spectroscopy.

1. Thermal Parameters of Hydrogels Determined by TGA.

sample	T onset (°C)		T max (°C)
H	198	423	251	501
H-CD	240	482	331	571
H-CD-CBD	234	499	334	563
H-CD-CBN	235	487	332	564

3.5. Differential Scanning Calorimetry

Figure c displays DSC thermograms of all the hydrogels H, H-CD, H-CD-CBD, and H-CD-CBN. The melting peak of H hydrogel appeared composed by two peaks: the first at around 96 °C, belonging to the higher ordered structure and compact network of the xanthan, and a further one that could be ascribed to the agar at approximately 112 °C. When CD, CD-CBD, and CD-CBN were incorporated into the hydrogel a unique endothermal peak at lower temperatures (i.e., 92 °C for H-CD, 94 °C for H-CD-CBD, and 92 °C H-CD-CBN) was observed suggesting that the presence of both CD and solid inclusion complexes (CD-CBD and CD-CBN) reduce the melting temperature by weakening intermolecular forces between polysaccharide chains and disrupting the crystalline structure. At about 250 °C, the H hydrogel presented an exothermic peak at 246 °C due to some degradative phenomena. When CD, CD-CBD, and CD-CBN were incorporated into the hydrogel, the peak associated with degradation disappeared, and no other signs of degradation were observed. This confirms the formation of physical interactions that enhance thermal stability. These results were consistent with the findings from both TGA and FTIR analyses.

3.6. Wide-Angle X-Ray Diffraction

Diffraction patterns of H hydrogel and its complexes are reported in Figure d. The pattern of the H hydrogel presented two broad peaks at 2θ of 18.9° and 13.5° belonging to the xanthan gum and the agar, respectively. H-CD showed two broad halos of amorphous cyclodextrin, typically centered around 10.2° and 18.6° (2θ), reflecting a disordered arrangement of molecules. Indeed, these guest molecules, (Cannabidiol (CBD) and Cannabinol (CBN)) did affect the structure of H-CD hydrogel.

3.7. Swelling

The swelling behavior of the H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels in PBS revealed rapid water uptake during the first hours, followed by a slower approach to equilibrium up to 72 h (Figure a). All formulations reached swelling ratios close above 1500% to a maximum of 2600%, which indicates a highly hydrated network with extensive free volume and efficient solvent penetration throughout the hydrogel structure. The absence of statistically significant differences in both the kinetics and the maximum swelling capacity suggests that the introduction of cyclodextrin and the corresponding cannabinoid inclusion complexes does not substantially disrupt the polymer cross-linked architecture. In other words, CD and CD–drug entities behave more as embedded supramolecular domains than as additional permanent cross-linkers, preserving the global balance between elastic restoring forces and osmotic pressure that governs hydrogel expansion. The slight apparent variations between compositions can still provide useful information. The marginally lower swelling observed for some CD-containing networks may reflect local restrictions in chain mobility due to host–guest interactions or physical entanglements around the CD cavities, whereas the cannabinoid-loaded systems maintain comparable swelling despite the presence of hydrophobic moieties, indicating that the hydrophilic backbone dominates the overall solvent uptake. Similar behavior has been reported for glucose-based and superabsorbent hydrogels, where high equilibrium swelling (often above 1500–3000%) is achieved while maintaining structural integrity and where changes in the comonomer or filler content modulate the swelling profile only within a relatively narrow window. This agreement supports the view that the present hydrogels belong to the class of highly swollen yet mechanically coherent networks suitable for diffusion-controlled release and soft-tissue contact.

6.

(a) Swelling percentage of H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels over time (0, 3, 7, 24, and 72 h). The hydrogels swelled up to 2500% at 72 h, and neither swelling kinetics nor maximum swelling capacity showed significant differences between groups, despite slight apparent variations among compositions. (b) Protein adhesion quantified as OD at 590 nm using the Coomassie Brilliant Blue (CBB) assay for H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels. (c) In vitro biodegradation as remaining mass (%) of H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels over 0, 7, 14, 28, and 60 days. Overall, the hydrogels did not exhibit measurable biodegradation during the study period, and no significant differences were detected among groups. (d) Representative macroscopic images illustrating in vitro degradation, confirming the absence of marked differences in biodegradation kinetics among the hydrogels and the preservation of their structural integrity after 60 days. Unless otherwise stated, all functional assays were performed using PBS as the aqueous medium.

3.8. Protein Adhesion

Protein adsorption on the hydrogel surfaces, quantified by the Coomassie Brilliant Blue assay (Figure b), showed comparable optical density values for all tested formulations without significant differences, indicating no marked effect of CD or cannabinoid loading on the amount of protein bound under the chosen conditions. This result is consistent with the tendency of proteins to be adsorbed through hydrophilic interactions on the hydrogel surface. From a mechanistic perspective, the hydrated polymer chains and the bound water layer at the hydrogel interface form an energetically favorable environment that enhances protein denaturation and anchoring, leading to a bound protein layer. , The similarity in protein adhesion between the control H hydrogel and the CD-containing systems also suggests that the cyclodextrin cavities and their inclusion complexes are either partially shielded by the surrounding polymer or do not present sufficiently large hydrophobic patches to drastically promote serum protein anchoring at the macroscopic level. These features would be potentially advantageous for biomedical applications, because excessive protein fouling can trigger unwanted cell responses, inflammatory cascades, or rapid biofilm formation. Recently, studies on pectin-based beads and polysaccharide multilayer coatings have reported analogous trends, where hydrophilic carbohydrate–rich interfaces exhibit controlled swelling combined with relatively low levels of nonspecific protein adsorption and good in vitro cytocompatibility. On the basis of these results, the cyclodextrin inclusion complex with NPCs embedded into hydrogel matrices would promote a moderate protein surface adhesion, able to support biological fluids contact without strong fouling, while still allowing sufficient protein presence to mediate integrin-dependent cell attachment when required.

3.9. In Vitro Biodegradation and Remaining Mass (%) of Hydrogels

The in vitro biodegradation assay performed up to 60 days in PBS showed negligible mass loss for any of the hydrogel compositions, indicating that the networks behave as effectively nondegradable under these mild aqueous conditions (Figure c). The percentage of remaining mass remained close to 100% for the entire study, with overlapping values without significant differences for H, H-CD, H-CD-CBD, and H-CD-CBN. To support these findings from the qualitative point of view, the macroscopic images corroborated the quantitative data by showing preserved shape and structural integrity after prolonged incubation for up to 60 days (Figure d). This stability points to a dominant contribution of hydrolytically resistant bonds within the polymer backbone and cross-links and to the absence of labile groups that could undergo significant cleavage at physiological environments in the time frame considered. Such long-term persistence can be advantageous in scenarios where the material is intended to act as a long-lasting scaffold or reservoir, for example, for chronic drug delivery or as a support for repeated cell seeding, although it also implies that in vivo resorption will likely require enzymatic pathways or mechanical fragmentation rather than simple hydrolysis. As reported in the literature, PEG-based and click-cross-linked hydrogels with tunable but relatively slow degradation profiles have shown similar behavior, where the network remains intact for weeks to months in buffered saline while still allowing controlled erosion under more demanding conditions or in the presence of specific enzymes. Also, recent work on synthetic microporous hydrogels for 3D culture demonstrates that maintaining structural integrity over extended culture periods is crucial to preserving pore architecture and mechanical cues for embedded cells. Thus, the negligible degradation observed in the present hydrogel system suggests that these hydrogels are best categorized as long-term stable matrices, suitable for applications where structural persistence is required, and where degradation, if desired, must be engineered through additional labile segments or bioresponsive cross-linkers. In the context of soft-tissue biomedical applications, such long-term stability may be particularly advantageous for local controlled-release depots or protective soft-tissue coatings that must remain in place for weeks to months, where subsequent removal or replacement is clinically feasible and preferred over rapid in situ resorption.

3.10. In Vitro Cumulative Release

Figure shows the cumulative release profiles of cannabidiol (CBD) and cannabinol (CBN) from hydrogels containing hydroxypropyl-β-cyclodextrin inclusion complexes (H-CD-CBD and H-CD-CBN) over 72 h in water and phosphate-buffered saline (PBS). In these plots, the cumulative release is expressed as the percentage of the initial cannabinoid loading per disc (release (%)), calculated from the known drug content of the HP-β-CD/CBD and HP-β-CD/CBN powders and the hydrogel formulation (≈90 μg CBD or ≈60 μg CBN per disc, respectively). For its part, for CBD (Figure a), both media show a rapid release phase during the first hours, followed by a slower approach to a plateau. In PBS, the hydrogels reach the plateau (close to the complete release of the loaded CBD) within approximately 12–24 h, whereas in water the release is slightly slower, approaching similar percentages at later time points. In other words, the release approaches ∼100% of the initial 90 μg per disc within 12–24 h, whereas in water it reaches a comparable plateau somewhat later, at ∼90–95% of the initial loading. This medium-dependence is consistent with a predominantly diffusion-controlled process modulated by the decomplexation of the CD–CBD host–guest system and the different ionic strengths of PBS, which can favor CBD solubilization and slightly increase the effective concentration gradient between the hydrogel and the surrounding solution.

7.

Cumulative release (%) of cannabidiol (CBD) (a) and cannabinol (CBN) (b) from H-CD-CBD and H-CD-CBN hydrogels in water and PBS as release media, monitored over 72 h. Yellow and green circles represent release in water, red triangles, and blue squares represent release in PBS. Cumulative release (%) was calculated as [CR­( t _n )/ M ₀ ] × 100, where CR­( t _n ) is the cumulative mass of cannabinoid released at time t _n and M ₀ is the initial loading per disc (≈90 μg CBD or ≈60 μg CBN). The solid straight lines correspond to the fits obtained with first-order and Fu–Kao release kinetic models.

For CBN (Figure b), the cumulative release (%) also increases over time, but plateau values are lower than for CBD, reflecting the smaller initial loading per disc and the higher hydrophobicity of CBN. In both media, most of the loaded CBN is released within 48–72 h, with PBS again showing a somewhat faster approach to the plateau compared with water.

Therefore, in PBS, the CBN release reaches ∼85–90% of the initial 60 μg per disc by 48–72 h, while in water the plateau remains near ∼80–85%. The medium-dependent behavior of CBN likely reflects a balance between its solubilization in the aqueous phase, its association with the CD cavities, and its interactions with the polysaccharide network. Thus, the release profiles of CBD and CBN are compatible with diffusion through a highly swollen agar/xanthan network, modulated by host–guest equilibria between the cannabinoids and the HP-β-CD domains. The kinetic analysis summarized in Table , using first-order and Fu–Kao models, supports the predominance of diffusion-controlled release with minor contributions from hydrogel relaxation, in agreement with previous cyclodextrin-based hydrogel systems for hydrophobic drug delivery. These results further demonstrate through Figure that the fraction of the loaded CBD or CBN released from the hydrogels in each medium is directly linked to the validated initial loading per disc.

2. Kinetic Parameters of Cumulative CBD and CBN Release from H-CD Hydrogels in Water and PBS.

sample	medium	max (mg/mL)	half-life (h)	correlation coefficient
H-CD-CBD	water	0.107	2.52	0.997
H-CD-CBD	PBS	0.108	1.32	0.987
H-CD-CBN	water	0.044	2.02	0.995
H-CD-CBN	PBS	0.047	2.05	0.995

3.11. Antimicrobial Activity

Two complementary techniques were used to evaluate the antibacterial activity of the hydrogels: a disk diffusion assay (Kirby–Bauer method) on MH plates and a growth-curve assay in the liquid LB medium (Figure ). In the diffusion test, only the H-CD-CBD hydrogel produced a clear inhibition halo against S. aureus, whereas no marked zones of inhibition were observed for E. coli with any formulation (Figure a–d). This pronounced effect of H-CD-CBD on the Gram-positive strain contrasts with the minimal activity of the other hydrogels and is noteworthy considering the low CBD loading, with an approximate CD/CBD molar ratio of 1:80 (and ∼1:160 for the CD/CBN system), indicating that relatively small amounts of complexed CBD can still generate a strong local antibacterial response.

8.

Representative macroscopic images of the Petri dishes from the disk diffusion assay containing H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels, showing inhibition zones against: (a,c) E. coli and (b,d) S. aureus. Ciprofloxacin and penicillin were used as antibiotic controls in these experiments. (e–h) Bacterial cultures grown in the LB broth from left to right: control (no material), H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels, respectively. Panels e and f show the results for E. coli, whereas panels g and h correspond to S. aureus. The top row displays cultures at t = 0 h, and the bottom row shows the same cultures after 24 h of incubation at 37 °C. The arrow indicates the only culture that exhibited antimicrobial marked activity in the presence of H-CD-CBD. (i,j) Optical density (OD₆₀₀) measurements at 0 and 24 h for E. coli (i) and S. aureus (j). Error bars represent the standard deviation of three independent replicates.

The liquid-culture experiments provide both qualitative and quantitative confirmation of these trends. Photographs of the LB cultures after 24 h show that all conditions remain turbid for E. coli, whereas only the H-CD-CBD tubes appear completely clear for S. aureus, suggesting almost complete growth inhibition in this group (Figure e–h). This observation is further supported by the optical density measurements at 600 nm (OD600) for E. coli, all treatments reaching OD600 values in the 2–3 range, indicating that none of the hydrogels are able to significantly inhibit this Gram-negative strain under the tested conditions (Figure i,j). In contrast, for S. aureus, the H-CD-CBD hydrogel reduces OD600 to values close to the baseline, while the control, H, H-CD, and H-CD-CBN groups retain high OD600 values with somewhat variable growth, pointing to only slight inhibition effects in the absence of CBD. These results align with previous studies reporting that CBD and other cannabinoids display higher potency against Gram-positive bacteria than against Gram-negative species, largely due to the absence of an outer membrane and the presence of a thick peptidoglycan layer that is more susceptible to membrane-targeting and cell-wall-disruptive mechanisms. Reported MIC values for CBD are typically lower for S. aureus and related Gram-positive pathogens, whereas E. coli and other Gram-negative strains require substantially higher concentrations or show intrinsic tolerance, which agrees with the strong, selective response observed here for the H-CD-CBD hydrogel and the limited effect on E. coli under comparatively low CBD loadings. ,

3.12. In Vitro Biocompatibility

Figure a shows fibroblast cell viability for the different hydrogels (H, H-CD, H-CD-CBD, and H-CD-CBN) as measured by the MTT assay after 1, 3, and 7 days. No statistically significant differences either between time points or between hydrogel compositions, indicating that none of the formulations exerted detectable cytotoxic effects under the conditions tested. Numerically, the hydrogel without cyclodextrin (H) showed slightly lower viability values (approximately 79%, 86%, and 85% at days 1, 3, and 7, respectively), whereas the CD-containing formulations, particularly H-CD and H-CD-CBD, maintained viability close to or slightly above 100% relative to the H-CD reference, with H-CD-CBD reaching ∼99%, 101%, and 97% at the corresponding time points. Cells without materials were considered as the control.

9.

(a) Cell viability of fibroblasts assessed by the MTT assay after exposure to H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels for 1, 3, and 7 days. Data are expressed as percentage viability relative to the H-CD group. (b) Cell proliferation of fibroblasts cultured with the same hydrogel formulations and time points, measured by the Alamar Blue assay (RFU, relative fluorescence units). (c) Cell-loading capacity of H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels after 6 h of seeding, expressed as the percentage relative to the control (cells without treatment, no material). No significant differences were detected between groups in any of the assays.

For its part, cell proliferation, evaluated by the Alamar Blue assay (Figure b), exhibited a progressive increase in metabolic activity from day 1 to day 7 for all hydrogel groups. Fluorescence intensity units (RFUs) rose markedly between days 1 and 3 and continued to increase up to day 7, with no significant differences among H, H-CD, H-CD-CBD, and H-CD-CBN groups. This pattern indicates that fibroblasts remained metabolically active and continued to proliferate over time on all matrices and that the presence of cyclodextrin and the CD–cannabinoid complexes did not impair, and may even subtly support, cell growth. Thus, in Figure a–b, MTT and Alamar Blue show the same overall trend of high biocompatibility for all hydrogels, although minor differences in the absolute values are expected because MTT reflects mitochondrial reductase activity linked to formazan formation, whereas Alamar Blue monitors broader cellular metabolic activity through resazurin reduction. These small assay-dependent variations do not alter the conclusion that the CD- and CD–cannabinoid-containing hydrogels support viable, metabolically active fibroblasts over 7 days.

Furthermore, the cell-loading capacity assay (Figure c) confirmed the ability of these hydrogels to support initial cell adhesion. The percentage of adherent cells on H, H-CD, H-CD-CBD, and H-CD-CBN at 6 h was comparable to that of the control condition (cells on tissue-culture plastic), with no statistically significant differences among groups. Although the control showed slightly higher mean values, all hydrogel formulations maintained cell-loading capacities within the same range, suggesting that neither cyclodextrins nor their CBD or CBN inclusion complexes adversely affect early cell–material interactions. These findings agreed with previous reports showing that hydrophilic, highly swollen hydrogel networks based on polysaccharides or synthetic polymers generally exhibit good fibroblast viability and proliferation over several days, provided that no leachable cytotoxic components are present. Similar results have been described for CD-functionalized hydrogels, where the incorporation of β-CDs or HP-β-CDs did not reduce cell viability and in some cases even enhanced metabolic activity, likely due to improved surface hydration and protein adsorption profiles that favor cell adhesion. In the context of cannabinoid delivery, the absence of cytotoxicity for H-CD-CBD and H-CD-CBN at the tested loadings supports the view that complexation with CD effectively solubilizes CBD and CBN while keeping local concentrations in a biocompatible range for fibroblasts. ,,

3.13. Histological Evaluation of Cell–Hydrogel Interactions (Hematoxylin and Eosin and Crystal Violet Staining)

Bright-field microscopy mosaics of unstained cultures were first acquired to visualize the overall distribution of fibroblasts on the different hydrogels (Figure ). Subsequent Hematoxylin and Eosin and Crystal Violet staining (Figure ) provided substantially higher contrast, allowing a more detailed assessment of cell morphology, coverage, and organization on each composition. At day 1, cell distribution and cell numbers were comparable between the control (cells without hydrogels) and the hydrogel groups (H, H-CD, H-CD-CBD, and H-CD-CBN), with sparsely distributed fibroblasts already exhibiting an early spread morphology. By day 3, cells formed more continuous monolayers, with elongated spindle-shaped fibroblasts aligned along the surface, indicative of active adhesion and proliferation typically observed when fibroblasts adapt well to biomaterial substrates. At day 7, the culture surface appeared largely confluent in all conditions, and fibroblasts displayed an elongated, well-spread cytoskeleton with dense cell–cell contacts, consistent with robust growth and good compatibility with the different hydrogel compositions (Figure ).

10.

Bright-field micrographs of fibroblasts cultured on the control surface (no hydrogel), H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels after 1, 3, and 7 days. Images are shown at 4× magnification (scale bar = 100 μm) and 20× magnification (scale bar = 200 μm) to illustrate overall cell coverage and detailed cell morphology on each substrate.

11.

Representative micrographs of fibroblasts cultured on the control surface (no hydrogel), H, H-CD, H-CD-CBD, and H-CD-CBN hydrogels after 1, 3, and 7 days. For each time point, the upper row shows crystal violet staining, emphasizing overall cell density and surface coverage, and the lower row shows Hematoxylin and Eosin (H&E) staining, highlighting nuclear and cytoplasmic morphology. All conditions display elongated, well-spread fibroblasts and a progressive increase in cell coverage over time, indicating good cytocompatibility of the hydrogel matrices. Magnification 4×; scale bar = 100 μm.

In addition to the quantitative viability and proliferation assays, fibroblast morphology and surface coverage on the different substrates were assessed by crystal violet and hematoxylin and eosin (H&E) staining (Figure ). For each time point (days 1, 3 and 7), the upper panels correspond to crystal violet (CV) staining, which highlights overall cell density and distribution, whereas the lower panels show the corresponding H&E-stained fields, allowing a more detailed visualization of nuclei and cytoplasm. At day 1, both stains reveal sparsely distributed but already well-spread fibroblasts on the control surface and on all hydrogels (H, H-CD, H-CD-CBD, and H-CD-CBN), with elongated spindle-like cells and no evidence of rounding or detachment. By day 3, the cell density increases markedly across all conditions, and the stained images show the formation of more continuous fibroblast layers covering large areas of the hydrogel surfaces. At day 7, crystal violet and H&E staining demonstrate nearly confluent cell sheets with homogeneous, aligned fibroblasts on the control and on each hydrogel formulation, without detectable differences in cell shape or organization between CD-CBD/CBN-free and CD-CBD/CBN-containing hydrogel samples. The absence of necrotic areas or abnormal morphologies and the persistence of elongated, well-spread cells confirm that the hydrogels provide a cytocompatible interface that supports fibroblast adhesion, spreading, and long-term culture, in agreement with previous reports on fibroblast-compatible CD-based hydrogels.

In short, we stress that bright-field and fluorescence images (Figures and ) qualitatively corroborated the quantitative cytotoxicity, proliferation, and cell load capacity data, showing fibroblasts with an elongated, well-spread morphology and intimate attachment to the hydrogel surface, consistent with a healthy fibroblastic phenotype.

4. Conclusions

The agar–xanthan hydrogels developed in this work provide a suitable matrix into which hydroxypropyl-β-cyclodextrin (HPβ-CD) and its inclusion complexes with nonpsychoactive cannabinoids (CBD and CBN) can be efficiently incorporated. Structural and thermal analyses indicate that agar supplies a partially crystalline, stable framework, and the addition of xanthan gum contributes to the amorphous flexibility. For its part, it is demonstrated that the presence of HPβ-CD further enhances thermal stability without disrupting the polysaccharide network. Also, the use of molecular encapsulation strategies such as the use of cyclodextrin inclusion complex (CIC) in the form of CD/CBD and CD/CBN complexes does not markedly affect swelling, protein adsorption, or PBS-based degradation, suggesting that cannabinoids remain encapsulated while the hydrogel architecture is preserved. Indeed, both CBD and CBN are released in a sustained manner over 72 h in water and PBS, with diffusion-dominated kinetics consistent with medium-dependent host–guest equilibria. It is worth noting that hydrogels containing CD/CBD show selective antibacterial activity against S. aureus while remaining ineffective against E. coli, and all formulations are cytocompatible with fibroblasts, supporting adhesion and proliferation over 7 days. All these results demonstrate that combining agar/xanthan hydrogels with HPβ-CD–cannabinoid inclusion complexes yields multifunctional, biocompatible materials with controlled release and selective antibacterial activity, making them promising candidates for soft-tissue local controlled-release applications.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.6c05484.

• FTIR-ATR analysis of agar and xanthan and spectral subtraction data for hydrogel–xanthan system (PDF)

This research received financial support from Project BIOCONTROL, coordinated by OVIS INVESTIGACIONES, AIE, together with its partner organizations, within the framework of the “Financiación estructurada de proyectos de I + D por Agrupaciones de Interés Económico (A.I.E.)” and in collaboration with INVENTIUM. Additional funding was provided, in part, by the Era-Min Cofund 2023 initiative COOL&SMARTTIT, MCIU/AEI/10.13039/501100011033/UE.

The authors declare no competing financial interest.