Evaluation of photodynamic therapy on nanoparticles and films loaded-nanoparticles based on chitosan/alginate for curcumin delivery in oral biofilms

International Journal of Biological Macromolecules 240 (2023) 124489 Available online 17 April 2023 0141-8130/© 2023 Elsevier B.V. All rights reserved. Evaluation of photodynamic therapy on nanoparticles and films loaded-nanoparticles based on chitosan/alginate for curcumin delivery in oral biofilms Amanda Letícia Polli Silvestre a , Aline Martins dos Santos a , Analú Barros de Oliveira b , Túlio Morandin Ferrisse c , Fernanda Lourenç˜ ao Brighenti b , Andr´ eia Bagliotti Meneguin a , Marlus Chorilli a , * a UNESP, S˜ ao Paulo State University, School of Pharmaceutical Sciences, Department of Drugs and Pharmaceutics, Araraquara, SP 14800-903, Brazil b UNESP, S˜ ao Paulo State University, School of Dentistry, Department of Restorative Dentistry, 14801-385 Araraquara, SP, Brazil c UNESP, S˜ ao Paulo State University, School of Dentistry, Department of Dental Materials and Prosthodontics, 14801-385 Araraquara, SP, Brazil ARTICLE INFO Keywords: Polymeric nanoparticles Buccal films Curcumin Chitosan Sodium alginate Antimicrobial photodynamic therapy ABSTRACT Nanoparticles and nanoparticle-loaded films based on chitosan/sodium alginate with curcumin (CUR) are promising strategies to improve the efficacy of antimicrobial photodynamic therapy (aPDT) for the treatment of oral biofilms. This work aimed to develop and evaluate the nanoparticles based on chitosan and sodium alginate encapsulated with CUR dispersed in polymeric films associated with aPDT in oral biofilms. The NPs were obtained by polyelectrolytic complexation, and the films were prepared by solvent evaporation. The photodynamic effect was evaluated by counting Colony Forming Units (CFU/mL). Both systems showed adequate characterization parameters for CUR release. Nanoparticles controlled the release of CUR for a longer period than the nanoparticle-loaded films in simulated saliva media. Control and CUR-loaded nanoparticles showed a significant reduction of 3 log10 CFU/mL against S. mutans biofilms, compared to treatment without light. However, biofilms of S. mutans showed no photoinactivation effect using films loaded with nanoparticles even in the presence of light. These results demonstrate the potential of chitosan/sodium alginate nanoparticles associated with aPDT as carriers for the oral delivery of CUR, offering new possibilities to improve the treatment of dental caries and infections. This work will contribute to advances in the search for innovative delivery systems in dentistry. 1. Introduction Streptococcus mutans and Candida albicans are important oral pathogens and are frequently related to dental caries and oral candidiasis [1]. Both species are present in the mouth and coexist in harmony with the host, but any imbalance in the microbiota such as stress, use of antimicrobials, excess sugar consumption, and poor hygiene habits lead to the development of these diseases, causing pain, discomfort, and sensitivity during chewing [2,3]. Moreover, the interaction between S. mutans and C. albicans amplifies biofilms' virulence, making the treatment difficult [4]. Antimicrobial Photodynamic Therapy (aPDT) has been widely used in the medical field, including in dentistry, to reduce the damage caused to the tooth, decreasing the loss of dental tissue, lower the incidence of root canal treatments, and total tooth removals. This is due to the noninvasive antimicrobial action, unlikely acquired bacterial resistance and side effects, affordable value, and simple application [5]. This therapy consists of the use of a photosensitizing (PS) substance that, when irradiated with a specific wavelength light source, acts on the target cell, triggering photochemical reactions (Type I and Type II). In this reaction, radical species and singlet oxygen are produced, which cause the death of microorganisms without causing damage to the patient's tissues. [6]. The chemical mechanism of aPDT is established through the Jablonski diagram. This mechanism can be observed in a detailed review of the subject documented in the literature [7]. Curcumin (CUR) – a constituent extracted from the rhizome of the Curcuma longa plant - has several pharmacological properties, such as anticancer [8–10], anti-inflammatory [11–14], antioxidant [15–17], antimicrobial [18–20], and others. Some of these properties are enhanced with blue light activation association (450-492 nm) which * Corresponding author. E-mail address: [email protected] (M. Chorilli). Contents lists available at ScienceDirect International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac https://doi.org/10.1016/j.ijbiomac.2023.124489 Received 28 October 2022; Received in revised form 2 April 2023; Accepted 13 April 2023 International Journal of Biological Macromolecules 240 (2023) 124489 2 became safe in the range of the spectrum commonly used in dental offices, through the use of LED or light-type devices [21]. Despite being active with excellent biological properties. The CUR application is limited because of its hydrophobic nature, solubility in toxic organic solvents, and rapid degradation under various conditions (light, temperature, and alkaline conditions) [22]. In addition, it is coloration and strong pigmentation usually stain teeth and resins [23], making it necessary to develop new strategies to improve this clinical application. Based on that, overcoming the limitations of CUR by incorporating it into nanosystems (NS) associated with aPDT is a very interesting way to release it into the oral biofilm [24,25]. Among the NS, there is a growing interest in the study and development of polymeric nanoparticles (NP) mainly using biopolymers. Among the options of several biopolymers, chitosan (CS) and sodium alginate (SA) stand out due to their natural origin and properties such as biocompatibility, biodegradability, no toxicity, and low cost. Furthermore, both polymers provide mucoadhesion through different mechanisms: i) the interaction of CS with mucin sialic acid through electrostatic interactions, and ii) SA creates hydrogen bonds with mucin-like glycoproteins [26]. Because of this, both polymers have been widely reported for use in oral mucosa [27–33]. CS is a hydrophilic cationic linear polysaccharide extracted by chitin deacetylation and composed of 2-amino-2-deoxy-β-D-glucose (1–4) linked and 2-acetamido-2-deoxyunits monomeric compounds of β-Dglucopyranose [34]. SA is a hydrophilic anionic linear polysaccharide extracted from brown algae and consists of monomeric residues α -Lguluronate and β-D-mannuronate [35]. Due to the high capacity of electrostatic complexation between these polymers, NPs of CS/SA can be obtained through the CS amino groups with the carboxylic groups of the SA, being considered a simple, easy, and relatively fast technique, that can be performed at room temperature with moderate agitation, without the use of organic solvents or expensive equipment [36,37]. For this reason, the selection of polymers was based on the mucoadhesive properties of CS and SA to increase the biological interaction of the system with the biointerface, in addition to the antimicrobial activity of CS that can act synergistically with PDT and the drug [38,39]. Furthermore, NPs offer other advantages of nanometer size, high surface area, high drug load, drug protection against degradation, and release rate control [40,41]. When encapsulated in polymeric films, they form the basis of a new system for buccal drug delivery, with several examples of success achieved [29,42–45]. SA-based polymeric films are thin and flexible films, with a pleasant taste and texture, in addition to being mucoadhesive and easy to handle, which increases patient compliance [46,47]. Although there are many works in the literature using CS/SA NPs and SA films, this is the first work that explored a combination of films loaded with NPs containing CUR associated with aPDT for the treatment of oral biofilms. These strategies should contribute to the dental clinic since the high salivary flow in the oral cavity hinders the application of free CUR. In addition, this system will promote better moldability at the target site, due to the contact of the polymers with the mucosa, optimizing the application of aPDT [48,49]. Therefore, this work aimed to develop and characterize two systems, CUR-loaded CS/SA NPs (NP_CUR) and SA-based films containing CURloaded NPs (NPs-loaded film), aiming to evaluate the potential of photodynamic therapy in S. mutans and C. albicans biofilms. 2. Materials and methods 2.1. Materials All reagents used were analytical grade. Solutions and dilutions employed deionized water (18.2 MΩ cm) obtained from a Milli-Q system (Millipore). Low molecular weight chitosan (Mw ≈50,000–190,000 Da; deacetylation degree of 75–85 %) and sodium alginate (Mw ≈107 kDa) were obtained from Sigma Aldrich® (S˜ ao Paulo, Brazil). Curcumin was obtained from Inlab® (S˜ ao Paulo, Brazil). Streptococcus mutans (ATCC 25175) and Candida albicans (ATCC 18804) were provided by Fundaç˜ ao FioCruz. 2.2. Methods 2.2.1. Influence of pH on the zeta potential (ZP) of polymeric dispersions The analysis of ZP as a function of pH was performed by the multipurpose titrator equipment MPT-2 (Malvern Instruments, Germany) coupled with the Zetasizer Nano-ZS® (Malvern Instruments, Germany). Stock solutions of CS in acetic acid solution 0.1 M (1 mg mL −1 ) and of SA in purified water (1 mg mL −1 ) were tested. The pH was automatically adjusted by the equipment using ready-made solutions of NaOH (0.05 and 0.5 M) and HCl (0.5 M), in a pH range from 2.0 to 8.5 (in steps of 0.7 ±0.1). The assay was evaluated in triplicate [36]. 2.2.2. Preparation of NPs NPs were obtained by the polyelectrolytic complexation technique, according to the method proposed by [50] with minor modifications. The CS was dispersed overnight in acetic acid solution 0.1 M (0.3 mg mL −1 ) and SA in purified water (0.5 mg mL −1 ), followed by adjustment to pH 4.0 and 6.0 (values selected by testing the effect of pH on PZ, Section 2.2.1), respectively. Different proportions of CS to SA (0.05:1; 0.10:1; 0.30:1; 0.50:1; 1:1) (w/w) of each dispersion were tested. The CS dispersion was dropped slowly into the SA dispersion through a peristaltic pump-P1(GEHealthcare) at a flow rate of 20 mL h −1 under magnetic stirring (~100 rpm) for 30 min. NP_CUR was prepared by the previous CUR solubilization in ethanol (1.0 mg mL −1 ), which was then mixed with SA dispersion containing 0.5 % of the surfactant Tween 80®. 2.2.3. Characterization of NPs The NPs were analyzed using the Zetasizer Nano ZS® equipment (Malvern Instruments, United Kingdom) at 25 ◦C and a scattering angle of 173◦to evaluate parameters such as average hydrodynamic particle size (Z-average) and polydispersity index (PDI) by the dynamic light scattering technique (DLS) and the ZP by the electrophoretic light scattering technique. All samples were measured 24 h after obtaining the NPs. All analyzes were evaluated in triplicate and mean values and standard deviations were analyzed. Fourier Transform Infrared Spectra (FTIR) (Thermo Fisher Scientific, Waltham, MA, USA) analyzed the components' interaction in the nanoparticle formulation. The FTIR spectra were obtained in the spectral range from 4000 to 400 cm −1 , with a resolution of 4 cm −1 at room temperature. The physical stability of NP_control and NP_ CUR was determined by evaluating the ZP after storage at 4 ◦C for six months (26 weeks). The surface morphology of the NPs was examined by field emission gum scanning electron microscopy (FEG-SEM) on JEOL JSM-7500 F (Jeol Company, Japan). The analyzed samples were diluted at 1:30 (v/v) in a dispersion of Tween 20® (0.5 %) previously to the drying process for no agglomeration among the particles. Photomicrographs were taken at ×50,000 magnification [51]. For the encapsulation efficiency test (EE%), a known volume of NP_CUR was added to an Amicon centrifugal filter device with Ultracel-100 membrane (100 kDa, Millipore Corporation, Billerica, MA) and centrifuged (Sorvall TC) for 15 min at 6200 g [44]. The free CUR deposited on the bottom of the Amicon centrifugal filter was analyzed by HPLC - model 1220 Infinity LC (Agilent Technologies) at 425 nm, using previously developed and validated methodology [52] EE% was calculated according to Eq. (1). The results were expressed as a percentage and the analyzes were performed in triplicate. EE% =(The initial mass of CUR −Mass of CUR in NP suspension) /(The Initial mass of CUR) × 100 (1) 2.2.4. Preparation of buccal films The solvent casting technique obtained the buccal films, as previously reported by [46] with some modifications. To prepare the Film_control, SA (2 %, w/v) was dispersed in water containing 10 % of A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 3 glycerin (concerning the polymer mass) used as a plasticizer for the filmforming dispersions. The dispersions were kept under magnetic stirring for 60 min. Then, 17 mL were poured into Petri dishes (63.5 cm 2 area) and dried at 40 ◦C in an oven with forced air circulation for approximately 12 h. For the preparation of NPs-loaded film, the same protocol described above was followed, except for the substitution of 8 mL of SA dispersion for the CS/SA NPs dispersion, in the final ratio of 1:1.5 NPs: SA dispersion (v/v) which was stirred for a few minutes until complete homogenization. After drying, all films were carefully detached from the Petri dish and kept in a desiccator for further studies. 2.2.5. Characterization of films 2.2.5.1. Macroscopic observations and thickness analysis. The buccal films were macroscopically analyzed for the occurrence of bubbles and/ or fissures, transparency, and flexibility. The thickness of the samples was analyzed by a digital micrometer MDC-Lite (Mitutoyo®) measuring five different positions of each film. Analyzes were performed in six replicates and the mean thickness was calculated [53]. 2.2.5.2. Field emission gun scanning electron microscopy (FEG-SEM). The surface and cross-section of the films were examined by FEG-SEM on JEOL JSM-7500 F (Jeol Company, Japan), using an accelerating voltage of 10 kV. Surface photomicrographs were taken at 2000×magnification and the cross-sections of NPs-loaded film photomicrographs were taken at 10,000×and 30,000×magnification. 2.2.5.3. Mechanical properties. The mechanical properties of the samples were performed by the Universal Texture Analyzer TA-XT2 equipment (Stable Micro Systems) using a spherical probe (5 mm) and a metallic support with a 10 mm diameter orifice, where films were mounted. The probe was moved down at a speed of 1 mm s-1 until reached the film surface when the speed was kept constant (0.1 mm s-1) throughout the test. The trigger force was 0.005 kg, Force versus displacement curves were recorded until the film ruptured and the puncture strength (Ps), elongation at break (Eb), and energy at perforation (Ep) were determined according to [53]. 2.2.6. Liquid uptake ability An adapted Enslin apparatus was used to evaluate the liquid absorption profile of the film samples in a simulated saliva medium composed of NaCl, KCl, MgCl2, KH2PO4, K2HPO4, NaHCO3 and purified water adjusted for pH 7. The samples were cut into pieces of 1 cm 2 accurately weighed and placed in the sintered glass filter of the funnel. The liquid absorption profile was measured through the volume absorbed (in percentage) in time intervals of 1, 2, 5, 10, 30, 60, 90, and 120 min, about the initial mass of the sample. The analyzes were performed in triplicate and the profile was measured by [53]. 2.2.6.1. Water vapor permeability (WVP). WVP was evaluated by a gravimetric method. For this, samples were cut into circular pieces of the known thickness and placed on top of glass flasks (11 mm opening) containing 10 mL of water (relative humidity 100 %). The set was kept in a desiccator containing silica gel (relative humidity 0 %). Flasks were weighed at predetermined times (0, 24, 48, 72, 96, and 120 h) as the WVP is based on the transport of water vapor from the weight loss of the flasks. After the permeation analysis, WVP (g mm m −2 h −1 Pa −1 ) was determined according to [53]. 2.2.6.2. In vitro mucoadhesion. In vitro mucoadhesion was analyzed by the Universal Texture Analyzer TA-XT2 equipment (Stable Micro Systems). Mucin discs (mucin-type II, Sigma Aldrich®, USA), previously hydrated with artificial saliva solution at pH 7 at 37 ◦C for 1 min. The film samples were attached to a cylindrical probe (10 mm in diameter) with double-sided adhesive tape. To assess the maximum detachment force (N), which represents the mucoadhesive force, the probe was moved down at a speed of 10 mm min −1 toward the mucin disc until reach a predetermined compression force (0.5 N). The sample was introduced 1 mm deep from the disc surface and held for 60 s without applying force. The probe was removed at a speed of 20 mm min −1 and the maximum detachment force (N) was determined for each sample [46]. The analyzes were measured in triplicate. 2.3. In vitro release test The in vitro release study was performed on Franz's cell apparatus (Hanson Research Corporation, Chatsworth, CA), using an artificial polyethersulfone (PES) membrane with 0.45 μ m pores (Millipore®), diffusion area of 1.77 cm 2 , and volume of 7.0 mL in the receptor compartment. Each membrane was placed on top of the receptor compartment which was filled with artificial saliva solution (pH 7) containing 2 % Tween 80® to maintain sink conditions. After that, 1000 μ L of NP_CUR or a section of NPs-loaded film was applied in the donor compartment on the upper part of the membrane. The medium was kept in a thermostatic water bath at 37 ±0.5 ◦C, under constant agitation in a magnetic stirrer. Aliquots of 1 mL were automatically collected at predetermined time intervals (0, 5, 15, 30, 45, 60, 120, 240, 480, 720, 960, 1200, and 1440 min). The amount of CUR released was analyzed by HPLC at 425 nm, applying a previously validated methodology. Tests were carried out in triplicate. The release kinetics was performed by fitting the release curve with different mathematical models (First order, Higuchi, Korsmeyer-Peppas) to evaluate the drug release mechanism. 2.4. Microbiological tests 2.4.1. Tested groups and photodynamic therapy parameters The microbiological assays were performed for the following seven experimental groups: free CUR, control NPs (NP_control), CUR-loaded NPs (NP_CUR), control film (Film_control), CUR-loaded film (Film_CUR), and films containing CUR-loaded NPs (NPs-loaded film). The same CUR concentration was maintained for all samples and the tests were applied to S. mutans and C. albicans biofilms. All groups were treated with and without light. The parameters for performing the aPDT were pre-irradiation of 5 min, irradiation of the wells by a 460 nm blue LED light system (Biotable, MMO) with an energy dose of 15 J/cm 2 for 11 min and 53 s [54]. 2.4.2. Preparation of bacterial and fungal suspension S. mutans was reactivated and cultivated in BHI agar for 48 h and incubated in a microaerophilic atmosphere at 37 ◦C (±1 ◦C). Afterward, about 3 to 5 colonies were placed in a tube containing 10 mL of BactoTM BHI broth +1 % sucrose, vortexed until reaching turbidity of approximately 0.5 of the McFarland scale (~10 8 bacterial cells/mL). C. albicans was reactivated and cultivated in Sabouraud agar for 48 h. A preinoculum was prepared by adding 3 colonies to 10 mL of YNB broth (Yeast Nitrogen Base) supplemented with 100 mM glucose. After incubation for 16 h in aerobiosis at 37 ◦C, the inoculum was prepared using 500 μ L of the pre-inoculum and 9.5 mL of YNB broth with 100 mM glucose. This suspension was vortexed until it reached a turbidity of approximately 0.5 on the McFarland scale (~10 6 fungal cells/mL) [54]. 2.4.3. In vitro bacterial and fungal biofilm formation in 96-well plate bottoms Biofilms were formed at the bottom of 96-well plates [55]. Initially, 100 μ L of agar was created at the bottom of each well plate. To induce biofilm formation, a volume of 100 μ L of the bacterial and fungal suspensions was placed in each well. For S. mutans biofilms, after 24 h, the old culture media were aspirated and refreshed with 100 μ L of the culture medium (BHI broth) for each well, until completing 48 h of growth in a microaerophilic environment at 37 ◦C (±1 ◦C). For C. albicans biofilms, after 1 h and 30 min, the old culture media were aspirated and A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 4 refreshed with 100 uL of the culture medium (RPMI 1640). The media was refreshed again every 24 h until completing 48 h of incubation. After irradiation of the samples, biofilms were detached, and colonies were counted through serial dilutions from the resulting suspension. Next, 5 μ L of the resulting suspensions were pipetted from each well using the technique of drop-in Petri dishes (90x15mL) in the biofilms. The microbial colonies formed were counted after 48 h of incubation [55]. 2.5. Statistical analysis One-way analysis of variance (ANOVA) followed by multiple comparisons using Tukey's test was used to assess significant differences. Results are expressed as a mean ±standard deviation. The statistical analysis was performed using GraphPad Prism 9.0. 3. Results and discussion 3.1. Influence of pH on the ZP of polymeric dispersions The charge density of the polymer is related to its ionization, which can be controlled by the pH value of the dispersion medium [34,36,41]. In this regard, the ZP of CS and SA solutions was performed in the function of different pH values, represented in Fig. 1. CS exhibited high positive charge density, with values of ZP above +40.0 mV at a pH range between 2.0 and 6.0 (Fig. 1). After that, there was a reduction of charge density from pH 7.0, reaching ZP values around +32.2 mV, due to deprotonation of the amine groups, which is associated with the CS pKa value of about ~6,5 [36,41,56].In contrast, SA is a hydrophilic anionic polymer that presents two pKa values of 3.38 and 3.65 [57]. For values above the SA pKa, there was a significant increase of the negative charge density, mainly from pH 5.0, resulting in a ZP of −29.9 mV, which remained constant until pH 8.5. This behavior is in agreement with the literature data [37], confirming that, SA is chemically stable at pH values between 5.0 and 10.0. In this study, pH conditions of 4.0 and 6.0 were chosen for the preparation of NPs, based on high charge density obtained for both polymeric dispersions, with ZP values of +50.0 mV at pH 4.0 for CS, and −31.5 mV at pH 6.0 for SA. In these conditions, the raised charge density, along with the large ZP difference, favors the maximum electrostatic attractions between CS and SA optimizing the formation of NPs by the complexation process. On the other hand, when the charge density is low, for example at pH values above 7.0 for CS and below 3.0 for SA, electrostatic repulsion occurs between the functional groups of the polyelectrolytes, leading to the formation of a two-phase system, characterized by macroscopic phase separation. For an intermediate range of charge densities (pH values around 6.5 for CS and 3.5 for SA), a complexation close to equilibrium can occur, resulting in aggregation and/or phase separation but at the microscopic level (mesophase) [37,41]. Therefore, the control of the reaction medium pH can be defined experimentally by ZP values, as an important step for the formation and stabilization of polycation-polyanion self-assembled NPs. 3.2. Preparation and characterization of NPs The complexation process can be influenced not only by the physicochemical properties of the polymers (e.g., molecular weight, degree of deacetylation, charge density, and hydrophilicity) but also by secondary experimental conditions, such as polymeric proportion, polyanion: polycation ratio, pH of the reaction medium, ionic strength of the solution, order of addition of polymers, drug presence, among others [36,41,56]. NP_control and NP_CUR were successfully prepared (for all tested proportions), and opalescent dispersions without the formation of aggregates were observed, indicating NPs complexation (Fig. 2B). SEM photomicrographs showed the spherical shape and nanometric size, with a narrow size distribution without agglomeration for empty and drug-loaded NPs (Fig. 2C – NP_control). The incorporation of CUR did not change the shape of the particles, which remained spherical (Fig. 2C – NP_CUR); however, a relatively larger particle size was observed, probably because of the association of the drug with the polymer matrix, corroborating with DLS results. Overall, FEG-SEM photomicrographs indicated smaller particle size values than those obtained by the DLS technique for both samples. In Fig. 1. Zeta potential of sodium alginate (SA) and chitosan (CS) as a function of pH value. Results are expressed as mean ±SD; n =3. A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 5 DLS analysis, the size measurements are based on the equation of the diffusion coefficient of the particles dispersed in an aqueous medium, leading to hydration, and swelling of these particles. On the other hand, FEG-SEM analysis requires a drying process of the samples, which promotes the contraction of the polymeric network, resulting in smaller particles. This behavior was also observed with other CS-based systems [36,41,56]. Results of particle diameter, PDI, and ZP values of control NPs and drug-loaded NP are presented in Table 1. The control NPs showed particle sizes between 181.2 and 384.1 nm, with PDI values ranging from 0.47 to 0.22, demonstrating homogeneity of the size distribution (Table 1). PDI values between 0.05 and 0.5 indicate a narrow size distribution for polymeric NPs [58–60]. For the CUR-loaded NPs, the mean size ranged from 257.2 to 473.1 nm, indicating that the loading of CUR into the NPs promoted an increase in particle size. Probably, the association of the drug caused a change in the rearrangement of polymeric chains during the complexation process between CS and SA, expanding its structure and resulting in larger particles. Furthermore, it was possible to observe that the addition of CUR promoted an increase in PDI values in the samples without the drug. However, the size distribution remained homogeneous even after drug incorporation for samples with higher mass ratios of CS to SA (0.30:1; 0.50:1, and 1:1). It is important to note that the samples with a lower ratio of CS to SA (0.05:1 and 0.10:1), showed higher values of particle size and PDI for both NP_control and NP_CUR. A lower proportion of CS provides a lower number of molecules available for complexation with SA, resulting in weaker interactions between the two polyelectrolytes. Thus, fewer CS molecules can associate with SA molecules, leading to the formation of larger and incomplete particles. ZP values were negative for both samples, ranging from −44.1 to −26.2 mV for NP_control, while NP_CUR had a small reduction in ZP after drug association(−33.0 and −26.5 mV) (Table 1). PZ values equal to or close to −30 mV and +30 mV provide greater electrostatic repulsion between particles, which reduces the possibility of aggregation, ensuring system stability [61–63]. The physical stability data in Fig. 3 show that the ZP of the NP_control and NP_CUR did not change significantly during storage at 4 ◦C for up to six months (26 weeks) (p > 0.05). The proposed methodology originated NP with high EE% (Table 1), which ranged from 64.13 to 80.23 % for NP_CUR prepared with lower ratios of CS to SA (0.05:1; 0.10:1 and 0.30:1), while those prepared with higher ratios of CS to SA (0.50:1 and 1:1) had EE% between 32.91 and 33.17 %. These results showed that increasing the proportion of CS (up to the ratio limit of 0.30:1) promoted a greater association of CUR with the NPs. Probably these conditions resulted in stronger interactions between the protonated amino groups of CS and carboxylate groups of SA, which led to more favorable conformational arrangements during the complexation process, favoring the association of CUR with the NPs. In another study with CS/SA NPs, the authors observed the raise of EE% values with the ratio increased from 0.05:1 to 0.10:1 CS:SA (42.9–61.3 %) [65]. Another work reported that increasing the CS:SA ratio (0.05:1 to 0.15:1) improved drug entrapment efficiency (33.2–54.9 %), these results are similar to those found in this work [64]. In this work also was observed the higher proportions of CS to SA (0.50:1 and 1:1) led to a reduction of up to 2.4 times in the association efficiency. A greater amount of CS molecules probably displaces CUR out of the NPs during complexation, resulting in a lower drug entrapment. Based on these results, the sample at a 0.30:1 ratio of CS to SA, which had the highest EE% and smaller particle size and PDI, was chosen for further analyses of this study. Chemical interactions between CS, SA, and CUR in nanoparticles were investigated by FT-IR. FT-IR spectra of CS (a), SA (b), NP_control (c) CUR (d), and NP_CUR (e) are shown in Fig. 4. The FT-IR spectrum of CS (Fig. 4a) exhibited bands at 3424 cm −1 corresponding to the amine and hydroxyl groups, and at 2926 cm −1 assigned to the presence of −CH groups, at 1658 cm −1 attributed to (-CO) NH 2 of the primary amide, at 1585 cm −1 referring to the axial deformation of the NH 2 group, in addition to the bands between 1177 and 894 cm −1 associated with the β-1,4 glycosidic bond and its polysaccharide structure [66,67]. In the SA spectrum (Fig. 4b) a band was observed at 3443 cm −1 , which is indicative of the presence of −OH groups, and a typical band at 1608 cm −1 corresponding to the C – – O groups. The band at 1415 cm −1 corresponds to the presence of carboxyl groups, and at 1027 cm −1 is associated with C-O-C groups attributed to saccharide structure [68,69]. The CUR spectrum (Fig. 4d) showed the main characteristic peaks of the molecule at 3512 cm −1 referring to the phenolic group, at 1628 cm −1 which indicates the C – – C portion of the aromatic group, a band at 1509 cm −1 associated with the C – – O and C – – C groups and another band at 1427 cm −1 corresponding to the olephenic group (C – H). Furthermore, a significant intense band at 1280 cm −1 Fig. 2. Schematic illustration of the NPs preparation by the polyelectrolyte complexation (A), schematic illustration of supramolecular interactions (electrostatic, hydrophobic, Van der Waals, steric interactions, hydrogen, and iondipole bonds) between CS and SA and the visual characteristic of NP_control and NP_CUR at 0.30:1 ratio of CS to SA (B) and photomicrographs of NP_control and NP_CUR at 0.30:1 ratio of CS to SA (50,000×magnification) by FEG-SEM. Table 1 Particle size, polydispersity index (PDI), zeta potential (ZP), and drug entrapment efficiency (%EE) of control NPs (NP_control) and CUR-loaded NPs (NP_CUR) at different mass ratios of CS to SA. Results are expressed as mean ± SD; n =3. The mass ratio (CS:SA) Samples Particle size (nm) PDI ZP (mV) %EE 0.05:1 NP_control 278.6 ± 0.018 0.39 ± 0.020 –43.0 ± 0.029 – 0.05:1 NP_CUR 473.1 ± 3.660 0.69 ± 0.020 –28.8 ± 0.306 64.13 ± 0.04 0.10:1 NP_control 384.1 ± 0.035 0.47 ± 0.047 –44.1 ± 0.071 – 0.10:1 NP_CUR 468.9 ± 10.510 0.66 ± 0.043 –33.0 ± 0.493 71.80 ± 0.60 0.30:1 NP_control 181.2 ± 0.002 0.22 ± 0.044 –31.1 ± 0.030 – 0.30:1 NP_CUR 257.2 ± 5.851 0.35 ± 0.068 –26.5 ± 0.493 80.23 ± 0.16 0.50:1 NP_control 201.2 ± 0.006 0.23 ± 0.076 –26.2 ± 0.012 – 0.50:1 NP_CUR 307.8 ± 8.420 0.42 ± 0.062 –27.3 ± 0.666 33.17 ± 0.57 1:1 NP_control 184.1 ± 0.004 0.24 ± 0.026 –26.4 ± 0.006 – 1:1 NP_CUR 264.0 ± 8.954 0.37 ± 0.058 –29.6 ± 0.907 32.91 ± 0.92 A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 6 Fig. 3. Results of the stability test performed throughout measurements of PZ in the function of time after storage at 4 ◦C of the NP_control and NP_CUR samples for six months (26 weeks). Fig. 4. FT-IR spectra of CS (a), SA (b), NP_control (c) CUR (d), and NP_CUR (e). A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 7 attributed to the (C – O) phenolic group and at 1155 cm −1 attributed to the C-O-C group [70,71]. The NP_control (Fig. 4. c) and NP_CUR (Fig. 4. e) spectra showed elongation bands similar to those presented by SA and CS polymers at 3443 cm −1 and 2926 cm −1 , respectively [72,73]. It was observed that some characteristic bands of CS suffered small shifts from 1585 cm-1 to 1571 cm-1(NP_control), and 1565 cm −1 (NP_CUR) for the SA polymer. It was observed characteristic bands of 1415 cm −1 and 1027 cm −1 for both samples (NP_control and NP_CUR). In addition, for the NP_CUR sample, the characteristic bands of CUR at 1628 cm −1 and 1280 cm −1 were shifted to 1630 cm −1 and 1250 cm −1 [74,75]. The results indicate that new intermolecular bonds were established between the components for the formation of NPs. 4. Preparation and characterization of buccal films 4.1. Macroscopic and microscopic analysis The control films (Film_control) and films loaded with NPs (NPsloaded films) were successfully obtained by the solvent casting technique (Fig. 5). Of note, this technique employs mild reaction conditions, without using high temperatures, making it suitable for chemically unstable compounds, such as CUR. By using plasticizers such as glycerin, it is possible to modulate the mechanical properties of the films to improve flexibility, tenacity, traction, and strength, making them suitable for application in the oral cavity [76,77]. Macroscopic analyses showed that the control films were homogeneous, transparent, easy to handle, and without the presence of bubbles and cracks (Fig. 5A). The addition of NP_CUR did not change the macroscopic characteristics of the films, except for the color, resulting in yellow films slightly less translucent than the control, as shown in Fig. 5B. All films developed showed adequate thickness, ranging between 0.024 and 0.042 mm, which agrees with literature data, in which films with <0.05 mm of thickness improve patient adherence, as they mold more adequately to body surfaces and orifices and tend to be more mucoadhesive [78]. In another word, thinner films favor the feeling of comfort during oral application. Among the samples, the Film_control was significantly thicker than the NPs-loaded film (p <0.05). The presence of NP_CUR changed the physical properties of the SA film, leading to the construction of more compact structures and, consequently, thinner. FEG-SEM photomicrographs showed a smoother and continuous surface of the Film_control, as shown in Fig. 6A. In contrast, NPs-loaded film exhibited a rough surface in the form of scales, indicating a change in surface morphology promoted by the presence of NPs (Fig. 6B). This behavior can be attributed to the different rates of solvent evaporation that occur at aleatory points in the Petri dish, as NPs can be considered a barrier against evaporation of water. The cross-sections of NPs-loaded film samples showed a uniform distribution of NPs, characterized by their spherical shape, in addition to the dense and, pore-free structure of the film that was maintained (Fig. 6 C and D). 4.2. Mechanical properties According to Table 2, PS values of Film_control and NPs-loaded films were suitable, ranging from 19.30 to 27.05 N/m 2 , respectively, indicating the development of resistant films to the perforation. The association of NPs in the structure of the films did not significantly affect the strength (p >0.05), maintaining the formation of a compact and flexible polymer network capable of withstanding high forces before breaking. It is known that the addition of nanostructures to the polymeric network can result in different mechanical properties, such as the improvement of Ps and, less commonly, its decrease. The impact generated is directly related to the pattern of interaction with polymeric groups. For the latter case, the filling of the free spaces among the polymeric chains with nanostructures may reduce their molecular mobility, impairing the establishment of new intermolecular interactions (hydrogen bonds and Van der Walls interactions), which should lead to the building of weaker and less structured polymeric network [79]. It is important to emphasize that the variation of Ps values found in this work should not affect the performance of the films in clinical practice because, although the mechanical properties are dictated by the bulk properties, the interaction with the biological tissues is governed by the surface properties [80]. Furthermore, Rokaya and coworkers in an important review of polymeric films applied to dentistry state that there has been increasing use of hydrogel-based Fig. 5. Schematic illustration of the film's preparation by solvent casting method (A) and images of Film_control and NPs-loaded film (B). Fig. 6. FEG-SEM surface photomicrographs of control films (A) and films loaded with nanoparticles containing curcumin (B) (both at 2000×magnification). Cross-sections of films loaded with nanoparticles containing curcumin at 10,000×(C) and 30,000×(D) magnification. Table 2 Mechanical properties of Film_control and NPs-loaded films. Values of the puncture strenght (PS), elongation at break (Eb), and perforation energy (Ep) (mean ±SD). Samples PS(N/m 2 ) Eb (%) Ep (N/m 2 ) Film_control 27.05 ±3.75 6.47 ±0.83 20.09 ±7.45 NPs-loaded film 19.30 ±2.82 2.88 ±0.48 28.05 ±3.24 A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 8 films, considering their viscoelastic properties, permeability, and mechanical properties that mimic those of natural tissues [81]. Considering that Eb is usually measured before the breaking point of the material and expressed as the percentage difference between the initial length and the breaking moment [82], values of Eb for both films were similar (p >0.05), indicating that the addition of NPs did not influence the elongation at break, and adequate flexibility was kept. This behavior agrees with the values of energy spent in perforation (Ep). According to the literature, the Eb values found for sodium alginate control films are close to the values obtained in other studies [83,84]. Such differences may be related to several parameters applied during the film preparation, such as dry matter content, composition, handling, mixing time, drying parameters, final thickness, plasticizer presence, etc. Indeed, the mechanical properties of the films are affected by the chemical composition of the SA as it is known that an increased amount in the polymer chain of α -L-guluronate block units forms more rigid and significantly fewer flexible films than with β-D-mannuronate block units [85]. 4.3. Liquid uptake ability In general, the control and NPs-loaded films showed adequate liquid uptake ability in the simulated saliva medium, with values ranging between 491 and 454 %, respectively (Table 3). This behavior is related to the high hydrophilicity of SA, which allows greater interaction between water molecules and the hydrophilic groups present in its structure, favoring the mobility of the polymer chains and, consequently, the diffusion of the medium [86]. The NPs-loaded films showed less ability to take up liquid than the control film, indicating that the presence of CS/SA NPs favored the formation of a more packaged matrix, hindering the diffusion of liquids. This finding agrees with the thickness data for films containing nanoparticles. The liquid absorption capacity plays an important role during the drug release process as well as in the mucoadhesion process. In the case of mucoadhesion, as the film structure becomes fully hydrated, it starts to lose its structural integrity due to the mucin interaction with the SA chains, promoting disintegration and dissolution of them [82]. The swelling of SA is mainly related to the number of G blocks, so that, the liquid absorption are based on the interaction of SA ions with the simulated saliva medium. The presence of Na + , an osmotic pressure called the Donnan effect, promotes electrostatic repulsion among polymer chains and, consequently, causes swelling and diffusion of the polymeric network of the film. The values are dependent on the osmotic pressure difference, the mixing pressure (water molecules are attracted to the surface of the polymer chains), and the stiffness of the macromolecular network [85,87]. 4.4. Water vapor permeability (WVP) According to the WVP results (Table 3), it is possible to observe that the Film_control showed higher permeability than the NPs-loaded films. Highly polar macromolecules, such as SA polysaccharides, contain many hydrophilic groups in their structure, which leads to interaction with water molecules through hydrogen bonds, resulting in increased WVP [83]. The NPs-loaded films presented lower permeability to water molecules due to the association of NP_CUR in their structure, indicating an excellent barrier property. These observations agree with the thickness, SEM, and liquid uptake ability results, indicating the formation of a more compact network that makes it difficult for vapor permeation. It should also be noted that the addition of NPs in continuous films increases the water vapor diffusional path, which decreases the rates of WVP. In addition, the hydrophobic character of CUR can help to prevent the diffusion of water molecules in the polymer matrix [88]. The results found by the author's Costa et al. (2018) and Motelica et al. (2021) demonstrated that the use of hydrophobic drugs in SA films made it difficult the water uptake, and demonstrated similar WVP results to those found in this work [89,90]. It can also be seen in the work by Aristizabal-Gil and coauthors (2019) that the presence of ZnO nanoparticles also acts as a physical barrier and decreased water absorption values were founded [91]. 4.5. Mucoadhesion in vitro The in vitro mucoadhesion test evaluated the maximum force required to detach the film from the mucin disc, and the results are presented in Table 3. Mucoadhesion values were similar for both films (p >0.05), indicating that the association of NPs to SA films did not influence the mucoadhesion force, and good mucoadhesive properties were kept. The mucoadhesion process can be established by supramolecular interactions between mucin, present in the oral mucosa, and the drug delivery system [92]. The carboxyl and hydroxyl groups of SA are responsible for the formation of hydrogen bonds between the polymer and the mucus layer. This is consistent with liquid uptake ability results, since the high swelling capacity of the films promotes the flexibility of the chains, favoring the establishment of intimate contact with the mucosa through the interpenetration with the mucin chain [85,93]. The article by Wang et al. (2022) developed chitosan‑sodium alginate-ethyl cellulose film and obtained lower values (0.89 N to 2.02 N in different ratios) of mucoadhesion for oral disease films [94]. However, in the work of Farias and coauthors (2020), we can observe similar results of mucoadhesion of SA and gellan gum films (values of 3.33 N) for antibiofilm activity in oral diseases [46]. 4.6. In vitro release assay The in vitro release study was performed with NP_CUR and Npsloaded film. In vitro, CUR release profiles from samples tested in simulated saliva media are shown in Fig. 7. The NPs-loaded film showed the highest drug release rate, releasing 92 % of CUR within 24 h. On the other hand, NPs released about 50 % of the drug in simulated saliva media (pH 7) during the same time. Such behavior can be related to the association of CUR molecules at the interface and/or interspersed in the hydrophilic hydrogel, which can shorten the drug diffusion path through the polymer network, facilitating the release of CUR from NPs. According to Notario-P´ erez et al. (2020), the steps of mixing polymers and NPs during film preparation, the relatively long solvent casting, and the use of water as a solvent during film production, can be considered enhancers of drug diffusion [95]. The hydrophilic nature of SA and the high liquid absorption capacity of SA films may also be influencing drug release. For example, at pH 7.0, the carboxylic groups of the alginate ionize, weakens the hydrogen bond due to the electrostatic repulsion, and thus tend to increase the swelling and diffusion capacity of the drug. [85]. In contrast, the NPs demonstrated a significant decrease in drug release rate, which indicates that the polymer-drug electrostatic interactions were able to keep it within the nanostructure, promoting a sustained release of CUR in simulated saliva media. Das et al. (2015) have reported similar results with CS/SA NPs, in which only 51 % of Table 3 Water vapor permeability (WVP), liquid uptake ability, and mucoadhesive force of control and NPs-loaded films (mean ±SD). Samples WVP (×10 −6 g mm m −2 h −1 Pa −1 ) Liquid uptake ability (%) Mucoadhesive force (N) Film_control 6.27 ±1.26 491.00 ±1.91 3.36 ±0.26 NPs-loaded film 4.15 ±1.85 454.00 ±2.81 3.79 ±0.33 A.L.P. Silvestre et al. International Journal of Biological Macromolecules 240 (2023) 124489 9 CUR was released from NPs in 24 h in phosphate buffer (pH 7.4) [96]. First Order, Higuchi, and Korsmeyer-Peppas mathematical models were applied to the drug release profiles (Table 4), to better understand the mechanism that governs the process of CUR release from NPs and NPs-loaded films (Fig. 7). The CUR release data from NPs correlated better with the KorsmeyerPeppas model, with r 2 values of 0.9873, n value of 0.63, and k value of 6.15. According to Eq. (2), the Korsmeyer-Peppas model is based on the exponential correlation of drug release with time, in which Mt./M∞ represents the amount of drug released at a time/at infinite time t; n is the release exponent and k is a kinetic constant. Mt/M∞ =K×tn(2) For spheres, when the coefficient is between 0.43 <n <0.85, indicates that the drug release mechanism occurs through non-Fickian or anomalous diffusion. In this release mechanism, there is a superposition of two phenomena, where the release is controlled by swelling and diffusion, simultaneously [97]. The same behavior was observed for NPs-loaded films, which presented a better correlation with the Korsmeyer-Peppas model, with an r 2 value of 0.9129, a k value of 29.45, and an n value of 0.60. For films, the release coefficient values between 0.5 <n <1.0, characterizes the release mechanism by non-Fickian diffusion or anomalous transport. Thus, the release process occurs by polymer matrix swelling followed by drug molecules diffusion. These results showed that the diffusion rate of CUR through the polymer matrix of NPs and NPs-loaded films will be a limiting step in the release process. In addition, it is important to consider that the value of the kinetic release constant k was significantly lower for NPs (6.15) compared to NPs-loaded films (29.45), indicating a reduction in CUR diffusion rates, so that the drug was released in a sustained manner from NPs. 5. Microbiological tests The microbiological assays were performed for the following samples: free CUR, control NPs (NP_control), CUR-loaded NPs (NP_CUR), control film (Film_control), CUR-loaded film (Film_CUR), and films containing CUR-loaded NPs (NPs-loaded film). The same CUR concentration was maintained for all samples and the tests were applied to S. mutans and C. albicans. 5.1. Quantitative evaluation of viable bacterial cells from S. mutans e C. albicans biofilms For the evaluation of CFU/mL in S. mutans biofilms, the free CUR sample (in alcoholic absolute solution) showed the same log reduction both for treatment with and without light, with log 10 CFU/mL values of 3.45 and 3.73, respectively, as shown in Fig. 8A. As expected, this sample showed greater antimicrobial activity since the CUR is in solution and, therefore, it does not need to be released to provide activity, being in direct contact with the biofilm. For all films, no significant difference was observed between the treatment with and without light (p >0.05). On the other hand, for both Fig. 7. In vitro CUR release profiles from NPs and NPs-loaded films in simulated saliva media pH 7 (mean ±SD; n =6). Table 4 Release coefficients of mathematical models (First order, Higuchi, Weibull e Korsmeyer-Peppas) from release studies for NP_CUR and NPs-loaded films. Samples Mathematical models First Order Higuchi Korsmeyer-Peppas R 2 k n R 2 k n R 2 k n NP_CUR 0.9841 0.0290 – 0.9784 10.60 – 0.9873 6.15 0.6373 NPs-loaded films 0.8367 0.1626 – 0.8873 21.17 0.9128 29.45 0.6070 A.L.P. Silvestre et al.