GELATIN/CARBOXYMETHYLCHITOSAN- AND CANNABIDIOL-BASED MEMBRANES: PREPARATION AND PERSPECTIVES FOR ASSISTING IN THE PREVENTION OF BRONCHOASPIRATION

Abstract

Aspiration pneumonia is one of the leading causes of death among older adults with dementia, and to address this health problem, cannabidiol (CBD) stands out due to its neuroprotective properties. The aim of this study was to develop and characterize gelatin/carboxymethylchitosan membranes with cannabidiol as a potential sublingual delivery system for cannabidiol in neurological disorders associated with bronchoaspiration. The methodology involved the preparation of membranes with different gelatin/carboxymethylchitosan (CMQ) ratios (90:10, 80:20, and 70:30), using Tween 80 to ensure CBD dispersion. The membranes were characterized by optical microscopy (OM), Fourier transform infrared spectroscopy (FTIR), thermogravimetric analysis (TGA), solubility, swelling, and cytotoxicity. OM showed that increasing the CMQ fraction resulted in greater structural heterogeneity. pH values remained within the physiological range (7.1-7.4). FTIR suggests possible interactions among the components, indicating drug incorporation. Swelling and solubility assays demonstrated that the presence of CMQ and CBD reduced water uptake and mass loss in different simulated media. TGA indicated that CBD exerted a plasticizing effect. Cytotoxicity assays showed cell viability above 80%, confirming the biocompatibility of the system. These results reinforce the potential of the membranes as a sublingual delivery system for cannabidiol and as an auxiliary strategy for the prevention of bronchoaspiration.

Keywords:
biomaterials; biodegradable; biopolymers; dementia.

INTRODUCTION

Aspiration pneumonia is one of the leading causes of death among older adults with dementia, generating a high clinical and financial burden. Between 1999 and 2020, more than 335,000 older adults with Alzheimer’s disease died from this condition in the United States.¹ In a study conducted by Manabe et al.,² individuals with dementia presented approximately twice the risk of death from pneumonia compared with those without dementia. Treatment imposes a substantial burden on healthcare systems, with hospitalizations exceeding US$15,000 per case.³ This scenario is further aggravated by increasing life expectancy and contemporary factors that favor psychiatric disorders and dementias.⁴

The use of cannabidiol (CBD) derived from Cannabis sativa has emerged as a therapeutic approach for neurological diseases.⁵,⁶ CBD comprises a complex and variable mixture of approximately 500 compounds, including terpenoids, flavonoids, and around 100 cannabinoids.⁷ Unlike ∆9-tetrahydrocannabinol (THC), CBD does not produce psychoactive effects and is associated with analgesic, sedative, antiemetic, antispasmodic, and anti-inflammatory properties. Its molecular structure (C₂₁H₂₆O₂; molar mass 310.43 g mol^-1) is predominantly hydrophobic, which directly influences its solubility, interaction with polymeric matrices, and incorporation into drug delivery systems.⁸

To enhance these therapeutic effects and enable safe and controlled application in biological systems, the development of matrices and supports capable of efficiently incorporating and releasing CBD is required, directing attention to the field of biomaterials. For any material to be considered an effective biomaterial, biocompatibility is essential, ensuring interaction with living tissues without inducing toxicity or adverse immune responses.⁹

Carboxymethylchitosan (CMQ) stands out as a biomaterial because it is a chitosan-derived biopolymer with mucoadhesive properties, biocompatibility, biodegradability, and bioactivity, in addition to exhibiting solubility at physiological pH. This characteristic favors its application in the pharmaceutical industry as a controlled drug delivery system, contributing to increased bioavailability and reduced systemic adverse effects.¹⁰

Gelatin is suitable for the proposed system due to its filmforming capacity, biocompatibility, biodegradability, and intrinsic mucoadhesive properties. As a protein derived from denatured collagen, its structure contains amino, carboxyl, and hydroxyl groups capable of forming hydrogen bonds and promoting high affinity for water. These characteristics favor the formation of flexible and rapidly hydrating matrices, which is particularly advantageous for sublingual applications, where rapid swelling, intimate mucosal contact, and patient comfort are required. Furthermore, gelatin is widely reported to contribute to the structural integrity and elasticity of hybrid polymeric systems, aiding in the production of cohesive and homogeneous films. Thus, its combination with carboxymethylchitosan enables the development of a balanced system that integrates structural stability, controlled hydration, and adequate mucoadhesive potential for sublingual administration.¹¹

In addition to the properties of biomaterials used in delivery systems, the choice of the administration route is a determining factor for therapeutic efficacy, with sublingual administration representing a promising alternative to oral and parenteral routes. This route involves placing the drug under the tongue, where direct absorption occurs through the highly vascularized sublingual mucosa. This avoids first-pass hepatic metabolism, promoting a rapid onset of therapeutic effects.¹² Drugs absorbed via this route reach systemic circulation directly through venous drainage to the superior vena cava, with passive diffusion being the main absorption mechanism.¹³

Despite the growing interest in the development of biomaterials for controlled drug delivery systems, studies addressing proteinpolysaccharide hybrid systems incorporating cannabidiol for sublingual administration are still scarce, especially focusing on complications associated with dementia, such as bronchoaspiration. In this context, the present study proposes the development of gelatin/carboxymethylchitosan membranes containing cannabidiol, seeking to combine the film-forming and mucoadhesive properties of gelatin with the structural stability and biofunctional characteristics of carboxymethylchitosan. The membranes were obtained by the solvent evaporation (casting) method and characterized in terms of morphology, pH, infrared spectroscopy (FTIR), swelling behavior, solubility in different simulated media, thermal stability (TGA), and cytotoxicity, with the aim of evaluating their suitability as potential sublingual delivery systems.

EXPERIMENTAL

Materials

In this study, the following materials were used: carboxymethylchitosan (CMQ) produced at the Laboratory for Evaluation and Development of Biomaterials of the Northeast (CERTBIO); commercial type A gelatin obtained from porcine skin (Sigma-Aldrich^®); lysozyme; phosphate-buffered saline (PBS), pH 7.4 (Sigma-Aldrich^®); cannabidiol 20 mg mL^-1 (Prati-Donaduzzi); polyoxyethylene sorbitan monooleate (Tween 80^®, Merck); artificial saliva (Roval); citric acid (C₆H₈O₇), analytical grade (Vetec^®); and ultrapure water.

Methods

Preparation of the membranes Preparation of the carboxymethylchitosan solution

A 1% (m/v) carboxymethylchitosan (CMQ) solution was prepared using the CMQ polymer produced by the Northeast Biomaterials Evaluation and Development Laboratory (CERTBIO). Initially, 1.0 g of carboxymethylchitosan was accurately weighed on an analytical balance and gradually dispersed in approximately 80 mL of ultrapure water under continuous mechanical stirring at 200 rpm. The system was maintained at room temperature (~ 24 °C) for 60 min to ensure complete dissolution. After complete homogenization, the final volume was adjusted to 100 mL with ultrapure water in a volumetric flask, resulting in a final concentration of 1% (m/v).

Preparation of gelatin/carboxymethylchitosan solutions

Gelatin was dissolved in ultrapure water under mechanical stirring for 1 h at 40 °C to obtain a 1% (m/v) solution. After complete dissolution of gelatin at 40 °C, the carboxymethylchitosan solution was gradually added under continuous mechanical stirring using a propeller-type stirrer. The mixture was maintained under stirring at 300 rpm and 40 °C for 30 min to ensure complete homogenization and formation of a uniform polymeric solution. Gelatin/carboxymethylchitosan ratios of 90:10, 80:20, and 70:30 (v/v) were employed.

Preparation of gelatin/carboxymethylchitosan membranes containing cannabidiol

After obtaining the different gelatin/carboxymethylchitosan solutions, tween 80 was added at 2% (v/v). Subsequently, cannabidiol was incorporated (1 mL per plate), and the mixtures were maintained under magnetic stirring until complete homogenization. The solutions, with and without cannabidiol, were poured into acrylic Petri dishes (5.5 cm in diameter), with a standardized volume of 20 mL per dish. The dishes were then dried in an oven at 40 °C for 24 h to ensure complete solvent evaporation and membrane formation. The membranes were coded as presented in Table 1.

Table 1
Sample coding

Code	Sample
CMQ	carboxymethylchitosan membrane
G	gelatin membrane
G70CMQ30	gelatin (70%) and carboxymethylchitosan (30%) membranes
G80CMQ20	gelatin (80%) and carboxymethylchitosan (20%) membranes
G90CMQ10	gelatin (90%) and carboxymethylchitosan (10%) membranes
G70CMQ30CBD	gelatin (70%), carboxymethylchitosan (30%), and cannabidiol (1 mL) membranes
G80CMQ20CBD	gelatin (80%), carboxymethylchitosan (20%), and cannabidiol (1 mL) membranes
G90CMQ10CBD	gelatin (90%), carboxymethylchitosan (10%), and cannabidiol (1 mL) membranes

Characterization

To evaluate the developed membranes, different morphological, physicochemical, thermal, and biological characterizations were performed. The techniques employed included optical microscopy (OM), pH determination, Fourier transform infrared spectroscopy (FTIR), swelling degree, solubility, thermogravimetric analysis (TGA), and cytotoxicity assays.

Morphology

Optical microscopy was employed to evaluate the morphology and macroscopic aspects of the obtained samples. Analyses were performed using a Hirox optical microscope operating in reflection and transmission modes, equipped with 2D accessories and magnification ranging from 20× to 3500×, coupled to an image analysis workstation.

Surface morphology, pore size, and pore distribution were evaluated using a Shimadzu SuperScan SS500 scanning electron microscope (SEM). Samples were mounted on aluminum stubs with carbon tape and gold-sputter-coated with a thin gold layer (~ 10 nm) prior to analysis. Imaging was performed at 500× magnification under an acceleration voltage of 10-15 kV.

Physicochemical analyses

The determination of pH was carried out using a Micronal pH meter, model B 474, equipped with a glass electrode combined with an internal potassium chloride (KCl) solution saturated with silver chloride (AgCl). The equipment was previously calibrated with buffer solutions at pH 4.0 and 7.0 and the pH measurements were performed directly, without sample dilution, and in triplicate.

FTIR analyses were conducted using a PerkinElmer Spectrum 400 spectrometer. Spectra were recorded in the mid-infrared region from 4000 to 400 cm^-1, with 16 scans and a resolution of 4 cm^-1, using an attenuated total reflectance (ATR) accessory equipped with a diamond crystal. FTIR was employed to identify characteristic bands of functional groups present in the samples, investigate intermolecular interactions among the constituents, and assess chemical structure changes resulting from the incorporation of cannabidiol into the gelatin matrix.

The swelling degree of the membranes was determined using the equilibrium immersion method. Samples were weighed and immersed in ultrapure water for 24 h. Subsequently, they were removed, gently blotted with filter paper to eliminate excess surface water, and weighed using a digital balance. The swelling degree was calculated as:

(1) $$\mathrm{SD}(\mathrm{\%})=\frac{{\mathrm{W}}_{\mathrm{f}}-{\mathrm{W}}_{\mathrm{i}}}{{\mathrm{W}}_{\mathrm{i}}}\times 100$$

where W_f is the weight of the sample at time t, and W_i is the initial weight of the sample.

All samples were weighed before and after swelling to determine the extent of membrane swelling.

The solubility test was performed to evaluate film degradation over a period of 1 h, based on the ASTM F1635-11¹³ standard test method. The membranes were weighed on a digital balance before and after the test to determine mass loss. Samples were prepared in quintuplicate and divided into four groups: one immersed in PBS solution at a concentration of 1 mg mL^-1 as a control (pH ≈ 7), one in artificial saliva, one in lysozyme solution, and one at pH 4.0 using citric acid. The pH of the solutions was measured at the beginning and at the end of the assay to evaluate possible pH variations.

The membranes were incubated in a microbiological incubator at 37 °C and removed after 24 h, followed by washing with ultrapure water, drying for 24 h in an oven at 40 °C, and weighing on an analytical balance. Biodegradation results were obtained from the ratio between the final mass and the initial mass using:

(2) $$\mathrm{Mass}\mathrm{loss}(\mathrm{\%})=\frac{{\mathrm{W}}_0-{\mathrm{W}}_{\mathrm{t}}}{{\mathrm{W}}_0}\times 100$$

where W₀ is the initial mass and W_t is the mass at time t.

Thermal analysis

Thermogravimetric analysis (TGA) of the membranes was performed using a PerkinElmer Pyris 1 TGA instrument. Approximately 5.0 mg of each sample was weighed on a precision balance (± 0.1 mg). The samples were heated at a rate of 10 °C min^-1 under a nitrogen atmosphere with a flow rate of 50 mL min^-1, using aluminum crucibles. TG curves were recorded from 20 to 600 °C to evaluate the thermal stability and degradation profile of the membranes.

Biological analysis

The cytotoxicity of the membranes was evaluated by the indirect contact method, in accordance with ISO 10993-5.¹³ The samples were sterilized by ultraviolet (UV) light and immersed in 1 mL of cell culture medium supplemented with 10% (v/v) fetal bovine serum and 1% (v/v) antibiotic-antimycotic solution at 37 °C for 24 h to obtain the extracts. L929 cells at a density of 1 × 10⁵ cells well^-1 were seeded into 96-well tissue culture plates and allowed to adhere for 4 h, after which they were cultured with the extracts of each sample for 1 and 4 days in a humidified atmosphere containing 5% CO₂ at 37 °C. Plate readings and cell viability assessment were performed using a Victor X3 spectrophotometer (PerkinElmer, Massachusetts) by measuring absorbance at 492 nm.

RESULTS AND DISCUSSION

Morphology

The optical microscopy images (Figure 1) of the membrane composed solely of gelatin (G) reveal a relatively homogeneous surface, although marked by cavities and small irregularities distributed throughout the film. These defects are characteristic of the drying of gelatin colloidal solutions, which tend to form microbubbles or localized shrinkage during the film formation process. Despite these features, the surface exhibits continuity, with no major cracks or collapse regions, indicating the formation of a continuous polymeric film with a well-organized internal structure typical of protein-based biomaterials.¹⁴ This behavior serves as a reference for the analysis of the other formulations containing CMQ and CBD.

Figure 1
Optical microscopy images of gelatin/carboxymethylchitosan membranes containing different proportions of cannabidiol (CBD): G (control, without CBD), G80CMQ20CBD, G90CMQ10CBD and G70CMQ30CBD

Optical microscopy images revealed generally similar surface morphologies among the membranes. However, subtle differences in surface homogeneity and the presence of small microdomains were observed as the CMQ content increases. The G90CMQ10CBD membrane exhibited a relatively smoother and more continuous surface, while formulations with higher CMQ content (G80CMQ20CBD and G70CMQ30CBD) showed slightly increased heterogeneity and the presence of small structural irregularities. These variations suggest modest changes in the organization of the polymeric network associated with the composition of the membranes.¹⁵

SEM images (Figure 2) revealed that all membranes exhibited continuous surfaces without evident macroscopic defects, such as cracks or structural collapse. At the magnification employed, the samples appeared relatively homogeneous, with no pronounced discontinuities in the film structure. It is important to note that the magnification used in this analysis does not allow a detailed evaluation of finer surface features, such as porosity, roughness, or microstructural organization. Therefore, no definitive conclusions can be drawn regarding subtle morphological differences among the formulations based solely on the presented SEM images. Within this limitation, the results indicate that the incorporation of carboxymethylchitosan and cannabidiol did not visibly compromise the structural integrity of the polymeric matrix at the micrometric scale.¹⁶

Figure 2
SEM images of gelatin/carboxymethylchitosan membranes containing different proportions of cannabidiol (CBD): G (control, without CBD), G80CMQ20CBD, G90CMQ10CBD and G70CMQ30CBD

In the G90CMQ10CBD membrane, composed of 90% gelatin, 10% carboxymethylchitosan, and CBD, a similarly continuous and uniform surface was observed under the same magnification conditions, without evident macroscopic defects when compared to the G membrane. From a theoretical perspective, the combination of gelatin and carboxymethylchitosan is known to promote intermolecular interactions, particularly through hydrogen bonding between functional groups, contributing to the formation of integrated polymeric networks. Additionally, the incorporation of cannabidiol, due to its lipophilic nature, may influence the organization of the polymeric matrix. However, such effects cannot be directly inferred from the SEM images presented here.¹⁷,¹⁸

Overall, the SEM analysis provides a general assessment of surface continuity, while more detailed investigations using higher magnifications or complementary techniques would be required to evaluate microstructural features and their potential impact on biological interactions.

pH determination

Table 2 presents the pH values of the different formulations. The observed behavior reflects the combined influence of the individual components. Ultrapure water, with a pH of 6.2, establishes a slightly acidic baseline. CMQ, with a pH of 7.7, contributes to increasing the pH due to the ionization of carboxylate groups, which confer buffering capacity. Type A gelatin presents a pH close to neutral under the conditions evaluated in this study, although values between 3.8 and 5.5 have been reported by the manufacturer depending on the concentration and preparation conditions. It is important to clarify that the pH value attributed to CBD corresponds to its dispersion in the aqueous system in the presence of tween 80, rather than to pure CBD, given its hydrophobic nature.¹⁹

Table 2
pH values of the analyzed samples, including ultrapure water, drug, gelatin, carboxymethylchitosan (CMQ), and gelatin/CMQ membranes with and without cannabidiol

Sample	pH
Ultrapure water	6.21 ± 0.08
Drug (D)	6.58 ± 0.07
Gelatin	7.00 ± 0.01
CMQ	7.74 ± 0.03
G70CMQ30	7.16 ± 0.06
G80CMQ20	7.31 ± 0.08
G90CMQ10	7.43 ± 0.02
G70CMQ30CBD	7.35 ± 0.09
G80CMQ20CBD	7.27 ± 0.06
G90CMQ10CBD	7.23 ± 0.05

For the formulations without CBD (G70CMQ30, G80CMQ20, and G90CMQ10), a slight decrease in pH is observed with increasing gelatin content, reflecting the balance between the acidic contribution of gelatin and the alkaline character of CMQ. Upon incorporation of CBD, no uniform trend in pH variation was observed across all formulations. While some compositions exhibited a slight decrease in pH, others showed similar or slightly higher values compared to their respective controls. These variations suggest that the effect of CBD on the system depends on the specific composition and may be associated with local interactions within the polymeric matrix rather than with a direct and systematic influence on hydrogen ion concentration. Overall, all formulations remained within a near-neutral pH range, indicating suitability for physiological applications.

The FTIR spectra (Figure 3) show the characteristic bands of gelatin, CMQ, tween 80, and CBD, reflecting the chemical composition of the developed membranes. Gelatin exhibits typical amide bands, with amide I around 1650 cm^-1 and amide II near 1550 cm^-1, associated with C=O stretching and N-H bending vibrations, respectively. CMQ presents characteristic bands related to carboxylate groups, particularly in the region of 1600-1400 cm^-1 (Table 3).²⁰

Table 3
Characteristic FTIR bands of the analyzed samples

Wavenumber / cm-1	Functional group	Component
3300-3400	O-H / N-H stretching	gelatin / tween 80
2920-2850	C-H stretching (aliphatic)	CBD / tween 80 / polymers
1650	amide I (C=O stretching)	gelatin
1550	amide II (N-H bending and C-N stretching)	gelatin
1600-1580	COO- asymmetric stretching	CMQ
1420-1400	COO- symmetric stretching	CMQ
1500-1450	aromatic ring	CBD
1300-1100	C-O-C / C-O stretching	tween 80 / CMQ
1100-1000	C-O stretching	CMQ / tween 80
800-600	aromatic C-H bending	CBD

Figure 3
FTIR spectra of the isolated components, carboxymethylchitosan (CMQ), gelatin, Tween 80, and cannabidiol (CBD), and of the formulated membranes G90CMQ10CBD, G80CMQ20CBD, and G70CMQ30CBD

In the spectra of the hybrid membranes, these bands appear in overlapping regions, making it difficult to visually distinguish individual contributions or confirm significant spectral shifts. Therefore, while the coexistence of these bands indicates the presence of both polymers, direct evidence of strong intermolecular interactions based solely on FTIR is limited. However, previous studies²¹,²² have reported that interactions between gelatin and chitosan derivatives may manifest as band broadening, intensity variations, and partial overlapping rather than clear peak shifts, due to hydrogen bonding and electrostatic interactions between amino and carboxylate groups. Such spectral features may account for the overlapping patterns observed in the present study.

From a theoretical standpoint, interactions between gelatin and CMQ are expected due to hydrogen bonding and electrostatic interactions between amino and carboxylate groups, as widely reported in the literature. However, such interactions cannot be conclusively demonstrated from the present spectra.²³

Regarding CBD, its characteristic bands are not clearly distinguishable in the spectra of the membranes, likely due to its low concentration and overlap with polymer signals. To facilitate visualization, selected spectral regions were enlarged (Figure 4). Even so, the identification of CBD-specific bands remains limited.²⁴

Figure 4
Infrared spectra of CBD-containing samples and gelatin for comparison

The broad band observed in the region between 3200 and 3400 cm^-1 is commonly associated with O-H and N-H stretching vibrations. However, this region may also be influenced by residual water in the samples, as supported by TGA results. Therefore, caution must be exercised when attributing variations in this region exclusively to changes in hydrogen bonding.

Overall, the FTIR results confirm the presence of the components in the membranes but provide limited direct evidence of specific intermolecular interactions. Complementary techniques would be required to achieve a more detailed and conclusive structural characterization.²⁵-²⁸

Gelatin exhibited the highest degree of swelling (Figure 5), exceeding 1000%, reflecting its highly hydrophilic nature. In contrast, the hybrid membranes containing CMQ, such as G70CMQ30, G80CMQ20, and G90CMQ10, showed a reduction in this value.

Figure 5
Swelling degree of membranes formulated with the compositions G70CMQ30, G80CMQ20, and G90CMQ10 and their respective cannabidiol-containing versions (G70CMQ30CBD, G80CMQ20CBD, and G90CMQ10CBD), as well as the individual controls of gelatin (G) and CMQ

The incorporation of CBD into the respective formulations further reinforced this decreasing trend, as the less hydrophilic nature of the compound limits water uptake into the polymeric matrix. When compared among themselves, a clear gradation is observed: the higher the gelatin content, the greater the swelling, whereas higher CMQ content and the presence of CBD lead to reduced expansion.

The swelling behavior of the hybrid membranes is influenced by the interaction among gelatin, CMQ, and CBD. Gelatin, which constitutes the base material of the formulations, is inherently hydrophilic and highly expandable. This characteristic arises from the abundance of amino, carboxyl, and hydroxyl groups in its structure, which favor multiple hydrogen bonds with water molecules, thereby promoting intense fluid uptake. Such behavior has been widely reported in the literature,²⁹ highlighting the strong hydration capacity of gelatin in biological environments.

Solubility

The analysis of Table 4 reveals the solubilization behavior of the compounds in different solutions (PBS, lysozyme, artificial saliva, and citric acid) supplemented with cannabidiol over a maximum period of 1 h. The reported values represent the percentage mass loss of the samples over time, highlighting distinct mass loss profiles depending on the medium used.

Table 4
Solubilization of drug-loaded samples in different solutions: PBS, lysozyme, artificial saliva, and citric acid over a 20-min interval

Solution	Sample			Mass loss of compounds / %
		5 min	10 min	15 min	30 min	1 h
PBS	G G70CMQ30 G80CMQ20 G90CMQ10	21.70 15.93 18.65 20.70	40.53 38.40 38.99 39.71	63.10 59.15 60.09 61.25	89.67 86.63 87.65 88.31	- - - -
	G70CMQ30CBD	13.12	35.12	56.12	80.12	-
	G80CMQ20CBD	15.23	36.23	57.89	82.35	-
	G90CMQ10CBD	17.99	37.12	59.01	84.96	-
Lysozyme	G G70CMQ30 G80CMQ20 G90CMQ10	23.76 19.45 21.73 22.12	43.15 41.34 41.99 42.98	65.56 61.65 62.83 63.35	93.20 90.12 91.45 92.12	- - - -
	G70CMQ30CBD	18.43	38.12	58.23	86.23	-
	G80CMQ20CBD	20.64	39.41	59.35	88,52	-
	G90CMQ10CBD	21.24	40.21	60.21	89.78	-
Artificial saliva	G G70CMQ30 G80CMQ20 G90CMQ10	27.90 23.65 25.40 26.45	44.78 39.34 43.45 44.21	67.15 63.74 64.54 66.90	96.64 93.62 94.65 95.01	- - - -
	G70CMQ30CBD	22.86	38.01	62.10	92.32	-
	G80CMQ20CBD	24.87	42.87	63.15	93.78	-
	G90CMQ10CBD	25.99	43.71	65.56	94.81	-
Citric acid	G G70CMQ30 G80CMQ20 G90CMQ10	26.91 22.90 24.51 25.54	40.87 37.87 42.78 43.12	66.57 62.36 63.74 65.21	95.32 92.14 93.64 94.15	- - - -
	G70CMQ30CBD	21.63	36.64	61.06	91.24	-
	G80CMQ20CBD	23.78	41.17	62.31	92.41	-
	G90CMQ10CBD	25.01	41.99	64.53	93.89	-

Gelatin, due to its hydrophilic nature, tends to solubilize more readily, whereas CMQ increases structural cohesion and offers better resistance to swelling. This explains why the G sample presents higher values, reaching nearly 90% solubility within 30 min, while the hybrid formulations show a clear reduction in this effect. When CBD is incorporated, an additional decrease in solubility is observed, suggesting that the compound acts as a hydrophobic “filler,” reducing water penetration into the matrix.

In PBS, membranes containing CMQ (G70CMQ30, G80CMQ20, and G90CMQ10) already exhibit lower solubility from the earliest time points. The higher the CMQ concentration, the more controlled the dissolution process, reflecting the structural role of carboxylate groups, which create internal ionic interactions that restrict material disaggregation. Even after 30 min, the solubility of membranes with higher CMQ content remains reduced compared to pure gelatin. This initial resistance indicates that the polymeric network formed between gelatin and CMQ is more rigid and less vulnerable to rapid solvent penetration.

The presence of CBD reinforces this behavior. In PBS, CBDcontaining membranes show even lower solubility values, particularly G70CMQ30CBD, which exhibits the lowest dissolution rate throughout the entire assay. The inherent hydrophobicity of CBD reduces water-matrix interactions. In addition, CBD appears to act as a stabilizing element, decreasing the rate at which gelatin tends to lose its structural integrity. This effect is consistently observed across the three evaluated formulations.

When comparing the behavior in artificial saliva, all materials exhibit higher solubility than in PBS, reflecting the more complex composition of this medium, which contains enzymes and ions capable of destabilizing intermolecular interactions. Even so, the performance pattern is preserved: gelatin remains the most susceptible material, while CMQ- and CBD-containing formulations maintain lower solubility. Differences among samples become particularly pronounced within the first 10-15 min, a phase during which saliva promotes a more intense attack on the polymeric matrix.

In acidic medium, as observed in the other environments, gelatin and CBD-containing samples, particularly G70CMQ30CBD, exhibit higher solubility. This behavior is attributed to the possible protonation of amino groups present in gelatin and chitosan derivatives, a phenomenon that facilitates solvent interaction and matrix disaggregation. Such behavior is well described for natural polyamines and their derivatives. In addition, the acidic environment may accelerate partial hydrolysis of peptide or glycosidic bonds, contributing to faster mass loss in non-crosslinked materials.¹⁹

When comparing the samples, pure gelatin is consistently the most soluble, followed by membranes with intermediate CMQ proportions, whereas formulations combining higher CMQ content with CBD present the lowest dissolution rates. G70CMQ30CBD is the most stable in both media, suggesting that its denser matrix offers greater resistance, while G90CMQ10CBD exhibits slightly higher solubilization due to its higher gelatin fraction. Collectively, these results demonstrate that the balance among gelatin, CMQ, and CBD allows fine modulation of solubility, offering a range of formulations from rapidly biodegradable systems to more stable and controlled matrices, an advantageous feature in the design of biomaterials intended for oral or physiological environments.

The incorporation of CMQ into the gelatin matrix reduces swelling. Although CMQ is also hydrophilic, it exhibits a more rigid structural organization and a more stable polymeric network, which contributes to limiting membrane expansion.³⁰ This explains the intermediate values observed for G70CMQ30, G80CMQ20, and G90CMQ10, where increasing CMQ content progressively reduces swelling capacity.³¹ According to Aleksandr et al.,³² CMQ improves the mechanical and thermal stability of hybrid materials, reduces polymer chain mobility, and promotes more controlled water uptake, favoring the formation of structures that balance resistance and hydration.

The application of CMQ-containing membranes has been explored in wound dressings, hydrogels, and drug delivery platforms. According to Mehdi-Sefiani et al.³³ hybrid gelatin-CMQ systems exhibit adjustable swelling, biocompatibility, and superior performance in physiological environments.

The presence of CBD in the membranes acts as a hydrophobic modulator capable of reducing the affinity of the material for water, while reinforcing polymer matrix stability. This characteristic arises from the apolar nature of CBD, which hinders water penetration into the structure, creating an environment more resistant to excessive swelling.³⁴,³⁵ Hydrophobic modulation is widely used in applications requiring controlled interaction with fluids, such as transdermal drug delivery systems, where moisture stability and gradual release of the active compound are essential for therapeutic performance.¹⁸

Thermogravimetric analysis (TGA)

The TGA results of the membranes without CBD (Figure 6) show the typical thermal behavior of gelatin-based systems, with three main stages: initial water loss, polymer degradation, and final residue formation. As summarized in Table 5, the increase in CMQ content leads to higher water loss, reflecting its hydrophilic nature, and a decrease in the main degradation mass loss, indicating improved thermal stability. This behavior suggests that CMQ contributes to the formation of a more stable polymeric network, likely due to intermolecular interactions between the biopolymers. The residue content increases significantly in formulations with higher CMQ content, particularly in G70CMQ30, indicating enhanced char formation and improved resistance to thermal degradation.

Table 5
Thermogravimetric analysis (TGA) parameters of gelatin and gelatin/CMQ membranes, including initial weight loss, water loss temperature, main degradation, peak degradation temperature, and final residue

Sample	Initial weight loss / %	Sample water loss temperature / °C	Main degradation / %	Peak temperature / °C	Final residue / %
Gelatin	11.48	216.19	62.50	330.68	2.86
G80CMQ20	11.94	230.06	58.12	332.56	2.74
G90CMQ10	11.65	198.70	60.00	343.51	3.32
G70CMQ30	14.36	235.45	51.48	339.91	14.97

Figure 6
Thermogravimetric curve of the samples without CBD

The TGA results (Figure 6) for gelatin (G) show its typical thermal behavior, with three main stages of mass loss. The first stage, around 11.48%, corresponds to the elimination of physically adsorbed water. The second stage, accounting for approximately 62.50% mass loss, occurs between about 200-350 °C and is associated with degradation of the polypeptide chain, involving the cleavage of peptide bonds and decomposition of amino acids.¹⁸ The third stage, accounting for only 2.86%, represents the final carbonization of the organic residue.³⁶ These values confirm the low thermal stability of gelatin, a characteristic of natural biopolymers rich in thermolabile functional groups.

The incorporation of CBD (Figure 7 and Table 6) results in slight modifications in the thermal behavior of the membranes. In general, a reduction in thermal stability is observed, which may be attributed to the presence of thermally labile functional groups in CBD, including phenolic groups that can promote earlier degradation.

Table 6
Thermogravimetric analysis (TGA) parameters of gelatin/CMQ membranes containing CBD, including initial weight loss, water loss temperature, main degradation, peak degradation temperature, and final residue

Sample	Initial weight loss / %	Sample water loss temperature / °C	Main degradation / %	Peak temperatutre / °C	Final residue / %
G90CMQ10CBD	12.08	225.99	63.07	334.80	2.29
G80CMQ20CBD	11.74	199.67	56.47	334.63	3.87
G70CMQ30CBD	10.89	199.78	50.34	327.35	3.79

Figure 7
Thermogravimetric curve of the samples with CBD

Additionally, CBD may act as a plasticizing agent, increasing polymer chain mobility and contributing to a decrease in degradation temperature. However, this effect is modulated by CMQ content, as formulations with higher CMQ fractions exhibit greater resistance to thermal degradation even after CBD incorporation. Overall, the thermal stability of the system is governed by the balance between the stabilizing effect of CMQ and the destabilizing influence of CBD. This results in a tunable thermal behavior, depending on the composition of the membranes.

A more detailed comparison indicates that the effect of CBD on thermal behavior depends on the gelatin/CMQ ratio. In formulations with lower CMQ content, a more pronounced reduction in thermal stability is observed, consistent with the plasticizing effect of CBD and its thermolabile nature. In contrast, higher CMQ contents contribute to increased thermal resistance, partially compensating for the destabilizing influence of CBD due to the formation of a more stable polysaccharide-rich network. Additionally, slight increases in residual mass suggest contributions from both CMQ carbonization and aromatic byproducts generated during CBD degradation. Overall, these results indicate that the impact of CBD on thermal stability is composition-dependent and modulated by the relative proportion of CMQ in the system.³⁷,³⁸

Cytotoxicity

The cell viability results (Figure 8) show that all formulations, including gelatin, gelatin/CMQ membranes without CBD, and CBDloaded membranes, maintained viability values above 85%, indicating a favorable biocompatibility profile. According to ISO 10993-5,¹³ materials are considered non-cytotoxic when cell viability exceeds 70% relative to the control, a criterion met by all samples.

Figure 8
Cell viability of gelatin (G), gelatin/CMQ membranes without cannabidiol (G70CMQ30, G80CMQ20, and G90CMQ10), and cannabidiol-loaded membranes (G70CMQ30CBD, G80CMQ20CBD, and G90CMQ10CBD)

The gelatin control (G) exhibited the highest viability, consistent with its well-established biocompatibility. The incorporation of CMQ led to a slight decrease in cell viability compared to pure gelatin, which may be associated with its higher density of functional groups and increased surface reactivity. However, this reduction was not significant and remained within acceptable biological limits.

Importantly, the comparison between membranes with and without CBD shows no consistent or significant reduction in cell viability upon incorporation of the bioactive compound. The values observed for G70CMQ30, G80CMQ20, and G90CMQ10 are comparable to those of their respective CBD-containing counterparts, indicating that cannabidiol does not adversely affect cell viability under the tested conditions.

These findings suggest that the incorporation of CBD into the polymeric matrix does not induce cytotoxic effects and is compatible with the developed system. Similar results have been reported in the literature,³⁹-⁴² where CBD-containing biomaterials demonstrated good cytocompatibility in different cell models. Overall, the results confirm that both CMQ and CBD can be incorporated into gelatin-based membranes without compromising their biological performance, supporting their potential application in biomedical systems.

CONCLUSIONS

The present study successfully achieved its objective of developing and characterizing mucoadhesive membranes composed of gelatin and carboxymethylchitosan (CMQ) incorporated with cannabidiol (CBD), aiming at a safe therapeutic alternative for patients with dementia and dysphagia. The solvent evaporation (casting) method proved to be effective and reproducible for producing films with properties suitable for sublingual administration.

Physicochemical characterizations confirmed the formation of a stable and functional polymeric matrix. FTIR analysis confirmed the presence of the components and suggested their coexistence and possible interactions, although no definitive conclusions could be drawn based solely on the spectra.

From a thermal standpoint, the incorporation of CMQ influenced the thermal behavior of the membranes; however, no consistent trend of increased thermal stability with higher CMQ content could be conclusively established based on the present data. The addition of CBD resulted in slight modifications in thermal stability, likely related to its plasticizing effect and the presence of thermally labile functional groups.

Regarding interaction with biological fluids, the system exhibited adjustable and promising behavior. The introduction of CMQ and the hydrophobic nature of CBD reduced the swelling degree and controlled membrane solubility compared to pure gelatin. In an artificial saliva-simulating medium, the G70CMQ30CBD formulation showed the lowest erosion rate and greater dimensional stability, which are important characteristics for ensuring adequate sublingual residence time for drug absorption.

From a biological safety perspective, all formulations demonstrated high biocompatibility. Cytotoxicity assays using L929 cells revealed cell viability above 80%, meeting regulatory criteria for biomaterials and indicating that the combined presence of the polymeric matrix and cannabidiol does not induce cytotoxic effects.

Overall, these findings indicate that the developed membranes represent a promising platform for sublingual cannabidiol delivery and may be further explored in future studies targeting neurological conditions associated with bronchoaspiration.

ACKNOWLEDGMENTS

The authors thank CNPq-Brazil and CAPES-Brazil for financial support, and UFCG (PB, Brazil) and CERTBIO for support in this work.

DATA AVAILABILITY STATEMENT

All data used and analyzed in this study are fully presented within the body of the manuscript.

REFERENCES

• ¹ Jehangir, H.; Faizan, A.; Khan, D.; Shahzad, Z.; Nadir, M. A.; Neurodegener. Dis. Manage. 2026, 16, 43. [Crossref]
  » Crossref
• ² Manabe, T.; Fujikura, Y.; Mizukami, K.; Akatsu, H.; Kudo, K.; PloS One 2019, 14, e0213825. [Crossref]
  » Crossref
• ³ Wu, C. P.; Chen, Y. W.; Wang, M. J.; Pinelis, E.; Ann. Am. Thorac. Soc. 2017, 14, 874. [Crossref]
  » Crossref
• ⁴ Viegas, C.; Franco, G. R. R.; Rev. Virtual Quim. 2017, 9, 1. [Link] accessed in June 2026
  » Link
• ⁵ Campos, A. C.; Fogaça, M. V.; Sonego, A. B.; Guimarães, F. S.; Pharmacol. Res. 2016, 112, 119. [Crossref]
  » Crossref
• ⁶ Rosenberg, E. C.; Tsien, R. W.; Whalley, B. J.; Devinsky, O.; Neurotherapeutics 2015, 12, 747. [Crossref]
  » Crossref
• ⁷ ElSohly, M. A.; Radwan, M. M.; Gul, W.; Chandra, S.; Galal, A.; Phytocannabinoids: Unraveling the Complex Chemistry Pharmacology of Cannabis sativa; Kinghorn, A.; Falk, H.; Gibbons, S.; Kobayashi, J., eds.; Springer: German, 2017.
• ⁸ Pagano, C.; Navarra, G.; Coppola, L.; Avilia, G.; Bifulco, M.; Laezza, C.; Int. J. Mol. Sci. 2022, 23, 3344. [Crossref]
  » Crossref
• ⁹ Ghasemi-Mobarakeh, L.; Kolahreez, D.; Ramakrishna, S.; Williams, D.; Curr. Opin. Biomed. Eng. 2019, 10, 45. [Crossref]
  » Crossref
• ¹⁰ Galdino, T. P.; de Oliveira, L. C.; Lima, E. P. N.; Farias, R. I. R.; de Jesus, R. A.; de Queiroz, S. F. A. M.; Santos, A. C. Q.; Fook, M. V. L.; Research, Society and Development 2022, 11, e5011527634. [Crossref]
  » Crossref
• ¹¹ Dubashynskaya, N. V.; Petrova, V. A.; Skorik, Y. A.; Int. J. Mol. Sci. 2024, 25, 5359. [Crossref]
  » Crossref
• ¹² Fisher, D. M.; Chang, P.; Wada, D. R.; Dahan, A.; Palmer, P. P.; Anesthesiol. Clin. 2018, 128, 943. [Crossref]
  » Crossref
• ¹³ Lim, S. C. B.; Schug, S.; Krishnarajah, J.; Pain Medicine 2019, 20, 143 [Crossref]; ASTM F1635-11: Standard Test Method for in vitro Degradation Testing of Hydrolytically Degradable Polymer Resins and Fabricated Forms for Surgical Implants, Philadelphia, 2011; ISO 10993-5: Biological Evaluation of Medical Devices - Part 5: Tests for in vitro Cytotoxicity, ISO: Geneva, 2009.
  » Crossref
• ¹⁴ Patil, M. B.; Hiredesai, R. D.; Kulkarni, V. B.; Patil, P. S.; Chavan, S. S.; Mathad, S. N.; Journal of Environmental Engineering and Studies 2025, 10, 48. [Crossref]
  » Crossref
• ¹⁵ Franco, A.; Van Durme, B.; Van Vlierberghe, S.; Dupont-Gillain, C.; Tissue Eng., Part C 2024, 30, 307. [Crossref]
  » Crossref
• ¹⁶ Zakuwan, S. Z.; Ahmad, I.; Abu Tahrim, N.; Mohamed, F.; Polymers 2021, 13, 1176. [Crossref]
  » Crossref
• ¹⁷ Nisat T. N.; Farzana K. R.; Asaduz, Z.; Int. J. Mater. Sci. Appl. 2018, 7, 62. [Crossref]
  » Crossref
• ¹⁸ Lapmanee, S.; Bhubhanil, S.; Charoenphon, N.; Inchan, A.; Bunwatcharaphansakun, P.; Khongkow, M.; Namdee, K.; Gels 2024, 10, 843. [Crossref]
  » Crossref
• ¹⁹ Bhardwaj, N.; Ashraf, M. T.; Maitra, J.; Biopolymers 2025, 116, e23645. [Crossref]
  » Crossref
• ²⁰ Bhardwaj, N.; Maitra, J.; Polym. Eng. Sci. 2025, 65, 6519. [Crossref]
  » Crossref
• ²¹ Baroudi, I.; Simonnet-Jégat, C.; Roch-Marchal, C.; Leclerc-Laronze, N.; Livage, C.; Martineau, C.; Gervais, C.; Cadot, E.; Carn, F.; Fayolle, B.; Steunou, N.; Chem. Mater. 2015, 27, 1452. [Crossref]
  » Crossref
• ²² Wang, F.; Qiao, W.; Guo, W.; Li, Z.; Cai, X.; RSC Adv. 2022, 12, 5677. [Crossref]
  » Crossref
• ²³ Kalahal, P. B.; Sajjan, A. M.; Khan, T. M. Y.; Rajhi, A. A.; Achappa, S.; Banapurmath, N. R. M. A.; Duhduh, A. A.; Polymers 2022, 14, 5114. [Crossref]
  » Crossref
• ²⁴ Frost, C.; Wiedbrauk, S.; Graham, E.; Yepuri, N. R.; Nelson, A. R. J.; Boase, N. R. B.; Macromol. Chem. Phys. 2023, 224, 2300315. [Crossref]
  » Crossref
• ²⁵ Firdous, S. O.; Arif, K.; Wong, S. Y.; Li, X.; Arafat, M. T.; J. Appl. Polym. Sci. 2023, 140, e54572. [Crossref]
  » Crossref
• ²⁶ Zuev, Y. F.; Derkach, S. R.; Lunev, I. V.; Nikiforova, A. A.; Klimovitskaya, M. A.; Bogdanova, L. R.; Skvortsova, P. V.; Kurbanov, R. K.; Kazantseva, M. A.; Zueva, O. S.; Int. J. Mol. Sci. 2024, 25, 11738. [Crossref]
  » Crossref
• ²⁷ Kulshreshtha, A.; Modica, K. J.; Jayaraman, A.; Macromolecules 2019, 52, 2725. [Crossref]
  » Crossref
• ²⁸ Roberts, C.; Papantonakis, M.; McGill, R. A.; Kendziora, C.; Furstenberg, R.; Kim, Y.; Shabaev, A.; Lambrakos, S.; Proceedings of SPIE Defense + Commercial Sensing; Orlando, United States, 2021. [Crossref]
  » Crossref
• ²⁹ Huang, Y. C.; Lee, C. T.; Don, T. M.; J. Polym. Res. 2021, 28, 173. [Crossref]
  » Crossref
• ³⁰ Zhang, H. Q.; Zhou, J.; Wang, H.; Lan, D.; Peng, Y.; Li, Z.; Dai, F.; Polym. Adv. Technol. 2023, 34, 691. [Crossref]
  » Crossref
• ³¹ Razmjooee, K.; Ahmady, A. R.; Arabzadeh, N.; Ahmadi, S.; Saber-Samandari, S.; Toghraie, D.; J. Mater. Sci. Eng. B 2023, 295, 116597. [Crossref]
  » Crossref
• ³² Aleksandr, K.; Mikhail, L.; Aleksandr, P.; Gels 2024, 10, 786. [Crossref]
  » Crossref
• ³³ Mehdi-Sefiani, H.; Granados-Carrera, C. M.; Romero, A.; Chicardi, E.; Domínguez-Robles, J.; Perez-Puyana, V. M.; Gels 2024, 10, 419. [Crossref]
  » Crossref
• ³⁴ Kirk, R. D.; Akanji, T.; Li, H.; Shen, J.; Allababidi, S.; Seeram, N. P.; Bertin, M. J.; Ma, H.; Med. Cannabis Cannabinoids 2022, 5, 129. [Crossref]
  » Crossref
• ³⁵ Aoyama, H.; Ogawa, T.; Ozawa-Umeta, H.; Teshima, K.; Hashimoto, T.; Sudo, T.; Hashimoto, K.; Tsuda, T.; J. Nutr. Sci. Vitaminol. 2025, 71, 312. [Crossref]
  » Crossref
• ³⁶ Kuzema, P. O.; Stavinskaya, O. N.; Laguta, I. V.; Kazakova, O. A.; J. Therm. Anal. Calorim. 2015, 122, 1231. [Crossref]
  » Crossref
• ³⁷ Liu, R.; Qiao, C.; Liu, Q.; Yao, J.; Xu, J.; Soft Matter 2024, 20, 1603. [Crossref]
  » Crossref
• ³⁸ Jannah, C.; Hudiyanti, D.; Prasasty, V. D.; Siahaan, P.; AIP Conf. Proc. 2022, 2553, 020002. [Crossref]
  » Crossref
• ³⁹ Söpper, U.; Hoffmann, A.; Daniels, R.; Sci. Pharm. 2021, 89, 35. [Crossref]
  » Crossref
• ⁴⁰ Cîrloiu , G. S.; Segneanu, A. E.; Bejenaru, L. E.; Văruţ, M. C.; Bălăşoiu, R. M.; Călina, D.; Stoian, A. C.; Băluşescu, G.; Herea, D. D.; Ciocîlteu, M. V.; Biţă, A.; Mogoşanu, G. D.; Bejenaru, C.; Pharmaceutics 2025, 17, 1189. [Crossref]
  » Crossref
• ⁴¹ Sierakowska-Byczek, A.; Gałuszka, A.; Janus, Ł.; Radwan-Pragłowska, J.; Molecules 2024, 29, 5376. [Crossref]
  » Crossref
• ⁴² Blebea, N. M.; Schröder, V.; Mitea, G.; Iancu, V.; Apetroaei, M. M.; Apetroaei, M. R.; Negreş, S.; 2021 International Conference on e-Health and Bioengineering (EHB); Iasi, Romania, 2021. [Crossref]
  » Crossref

• Associate Editor handled this article:
  Livia Cristina R. M. da Frota

Publication Dates

• Publication in this collection
  27 July 2026
• Date of issue
  2026

History

• Received
  29 Dec 2025
• Accepted
  28 May 2026
• Published
  12 June 2026