Wound healing remains a problem because wounds caused by burns, chronic diseases, trauma, and post-surgery may be unavoidable for the proliferation of resistant microorganisms [1]. Thus, skin repair is indispensable, and the faster it occurs, the shorter the exposure time of the body to a series of pathogens [2]. Chronic wounds, which do not heal easily and generate a vicious cycle of inflammation and infection, impair the functional restoration of skin tissue [3].

Studies indicate that patients with chronic wounds have an affected quality of life compared to those who are not affected because of pain, difficulty in walking, dissatisfaction, anxiety, depression, social isolation, and constantly changing body image, causing a decrease in self-esteem and functional ability, and often preventing them from carrying out their routine activities [4–6].

Successful treatment of chronic wounds involves several factors, and complications, such as preexisting diseases, comorbidities, and advanced age, can make this process difficult. One such treatment is the use of topical antibiotics [7]. Mupirocin (MUP), a class of antibiotics used to treat wounds, inhibits the synthesis of bacterial proteins by the reversible and specific binding of isoleucyl tRNA synthetase [8]. It has excellent activity against gram-positive staphylococci, but its activity against gram-negative organisms is more variable [9].

Although MUP is used to treat various bacterial skin infections, it has some limitations such as a short half-life, high protein binding, and drug resistance. Therefore, new strategies for its administration have been adopted to improve patient safety, reduce drug resistance, and expand its delivery [10].

In recent years, polymer-based films have been developed as a means of sustained or controlled drug delivery, with adhesive properties, providing the drug to a specific location on the skin in a prolonged manner, and improving therapeutic effects [11]. Among polymers, hydrogels have the advantage of maintaining the environment hydrated and ideal for treatment, in addition to having low interfacial tension, high permeability to oxygen and water, absorption of excess exudates, easy trauma, and good mechanical properties [12, 13]. Particularly, carbohydrate-based hydrogels are good candidates for the construction of delivery systems because they are nontoxic, biocompatible, and biodegradable [14].

The novelty of our study was the production of biodegradable films based on sodium alginate and κ-carrageenan, loaded with MUP, which exhibited antibacterial activities against both gram-positive and gram-negative microorganisms. Films comprising different proportions of the polysaccharides were obtained through crosslinking with the Zn²⁺ ion and were applied as a matrix for the controlled delivery of MUP. Zn²⁺ was selected as the crosslinking agent because it is an essential micronutrient for human health with significant biological relevance and abundance in the epidermis. Topical zinc sulfate has also been used to treat wounds and improve healing [15]. The obtained films were characterized by infrared spectroscopy, Raman spectroscopy, micro-Raman imaging, and scanning electron microscopy (SEM). The degree of swelling, uniformity, release, permeation kinetics, and antimicrobial effects of the films against the four bacteria were studied.

Table 2 summarizes film uniformity data in relation to mass, thickness, and MUP concentration. It was observed that all films presented uniformity in the three parameters analyzed, presenting a low standard deviation. However, there was a difference in thickness and uniformity between the films, with F2 presenting a greater thickness and thus a greater mass. Figure 2 depicts the bar graph of the absorptive capacity and the results of the two-way ANOVA analysis. The data were evaluated for normality using the Kolmogorov-Smirnov test, and the homogeneity of variance was assessed using the Bartlett test. In all statistical tests, a significance level of p < 0.05 was employed. The statistical analyses were conducted using GraphPad Prism software, version 7.0 for Windows (GraphPad Software).

Figure 3 shows the Raman spectra of MUP and the carbohydrates used to obtain the films, and Table 3 summarizes the attempt to assign bands [21–25]. Figure 4 presents the Raman spectra of the films obtained with and without crosslinking to compare and identify the spectral changes. Thus, in the film spectra, it was possible to observe bands at 359, 893, and 1,093 cm^–1 and associate them with the presence of both carbohydrates. The bands at 736 and 853 cm^–1 are related to κ-carrageenan and those at 809 and 957 cm^–1 are related to sodium alginate.

Figure 5 presents the infrared spectra of MUP and the carbohydrates used to obtain the films, and Table 4 summarizes the assignment of the observed bands [25–33]. Figure 6 presents the infrared spectra of the films obtained with and without crosslinking to compare and identify the spectral changes. Bands associated with both carbohydrates were observed in the spectra of all samples.

Figure 7 presents the permeation data using the Franz cell and the polysulfone membrane in a medium of phosphate buffer and 5% methanol. First, MUP presents a good release for all films except for F3, as it was the only film with a permeation percentage lower than 60%. This may be related to the fact that the film presented lower degrees of swelling than the other films, which would lead to lower MUP release [1]. Furthermore, a controlled release process was observed in all films, as immediate permeation was not observed; this only occurred when MUP permeation was analyzed without the presence of polymeric film. To study the permeation and release kinetics, the data were fitted to zero-order, first-order, Higuchi, Peppas-Korsmeyer, and Weibull models. The equations for each model, parameters obtained, and statistical analyses for each film are summarized in Table 5.

Figure 8 shows the micro-Raman images of the κ-carrageenan/alginate films obtained by integrating the band at 864 cm^–1. This was chosen because it is a region where only κ-carrageenan shows a signal. In the Raman image of all the mixtures, 20 points were analyzed, and in all of them the presence of bands at 864 cm^–1 was observed, with more or less intensity, represented by blue to red coloration on the contour map.

Figure 9 shows SEM images of the films. The solvent evaporation method used in this study does not normally produce interesting architectural structures, such as electrospinning or phase separation methods [1, 34]. However, analysis using SEM can provide useful information on the miscibility of the polymers that form the film. Film F2 was the only film that presented a flat surface, as opposed to films F1 and F3, which presented an agglomerated surface after evaporation of the solvent. However, for all of them, phase separation was not observed.

Figure 10 depicts the bar graph representing the diameters of inhibition and the results of the one-way analysis of variance (adjusted p-value). The data were assessed for normality using the Kolmogorov-Smirnov test, and the homogeneity of variance was evaluated through the utilization of Bartlett’s test. The impact of the samples on bacterial growth was evaluated through one-way analysis of variance (ANOVA), followed by a Tukey test. In all statistical tests, a significance level of p < 0.05 was employed. The statistical analyses were conducted using GraphPad Prism software, version 7.0 for Windows (GraphPad Software).

The mechanical properties of a polymer are significantly influenced by the crosslinking density. The modulus or rigidity of polymers with a higher crosslinking density is typically enhanced [35]. In addition, polymer blends can change the properties of the system, such as the three-dimensional organization of the chains and, consequently, the thickness of the film [36]. In this context, the observed thicknesses of the polymeric films can be attributed to the distinct proportions of carbohydrates utilized in each film and the potential for varying degrees of crosslinking, given the disparate compositions of each film. Regarding the MUP concentration, it is observed that the films present values between 3.403 and 3.825%; however, these values are lower than the 8.0% added during the preparation of the polymeric film. This may be due to the non-total solubilization of MUP and its deposition in the Petri dish and/or the crosslinking and washing process that resulted in the MUP leaching process.

The swelling capacity of polymeric films is an important parameter when applied to dressings. Carbohydrates, such as carrageenan and sodium alginate, swell in physiological fluids [37, 38]. The water absorption capacity and rate of crosslinked and water swellable polymers can be explained in terms of the diffusion of water in the polymer network and the dynamics of hydrogen bond formation. Consequently, as the crosslinking density is responsible for forming the 3D structure of the materials, it will influence the water absorption process [39]. Thus, if the degree of crosslinking is insufficient, the interactions between the polymer chains will be readily disrupted, resulting in a fragile hydrogel that is susceptible to dissolution in water. Conversely, a high degree of crosslinking will result in a reduction in the hydrogel’s swelling capacity [40]. A statistical comparison of the samples reveals that films F1 and F2 exhibit comparable absorption capacity, exceeding that of sample F3. Additionally, mass loss is discernible in samples F1 and F2 after 120 and 180 minutes, respectively, whereas this phenomenon is not observed in film F3. This may be associated with the elevated proportion of κ-carrageenan in the film F3 and a greater degree of crosslinking.

The crosslinking process can be confirmed by comparing the Raman spectra of the films with and without crosslinking; thus, the appearance of a band at 1,650 cm^–1 is observed, which can be associated with the interaction of the alginate carboxylic group with Zn²⁺ ions. It was also observed that the band at 956 cm^–1 split into two in the spectra of the crosslinked films (F1 and F2) compared to the spectrum of the non-crosslinked films; for the F3 crosslinked film, this band was not present in the spectrum of the non-crosslinked film. As this band is associated only with sodium alginate and attributed to δ(C–O–H), it can indicate the interaction process between the Zn²⁺ ions and the alginate hydroxyl. This result corresponds with the literature, which describes the formation of a complex between this carbohydrate and Zn²⁺ ions [41]. In relation to κ-carrageenan, it is not possible to indicate the crosslinking process from the Raman spectra, because it would occur owing to the interaction of the OSO₃^– groups with Zn²⁺ and the band related to (S=O) appears superimposed on sodium alginate bands. However, the literature describes the formation and interaction of the carrageenan complex with Zn²⁺ [42, 43].

Regarding infrared, some changes indicated crosslinking through Zn²⁺ ions. The carbonyl stretches of sodium alginate (approximately 1,600 cm^–1) shifted to higher wavenumbers after crosslinking, which may be related to the formation of coordinated bonds between the carboxylate groups of alginate and Zn²⁺ ions. This is because coordination changes the length of the C=O bond, which produces a change in wavenumbers [43]. In relation to the carbonyl group of the F3 film, the band at 1,417 cm^–1 shifted to lower wavenumbers, indicating the possibility of an interaction between this chemical group and Zn²⁺. In relation to κ-carrageenan, one of the likely interactions with Zn²⁺ would be with the sulfonic group, which would alter the bands at 1,258, 848, and 704 cm^–1. In relation to the band at 1,258 cm^–1, a decrease in its relative intensity is observed after crosslinking. For the bands at 704 and 848 cm^–1, it is not possible to observe any change. However, in the literature, it is described that the interaction between ƛ-carrageenan and Zn²⁺ shifts the band at 1,258 cm^–1 to lower wave numbers; thus, it is possible to assume that the change observed in the present study also indicates this interaction [43].

The utilization of polymeric films for the regulated release of MUP has been the subject of numerous investigations. The permeation flow of films comprising high and low molecular weight chitosan was investigated using rabbit skin, and it was demonstrated that the sustained drug release effect could be maintained for up to 24 hours [44, 45]. The same period of controlled release was achieved using rabbit skin in hydrogels composed of acrylic acid and 2-acrylamido-2-methylpropane sulfonic acid. The in vitro release study of compounds containing chitosan, sodium alginate and Carbopol^® showed that MUP release was enhanced, with maximum release occurring at 12 h [46]. The release duration observed in this study was up to nine hours, a value that falls below that reported in comparable studies. However, it is important to note that the studies used rabbit skin as the membrane, which will inherently have a reduced release profile compared to a synthetic membrane [47].

In relation to the release kinetics models studied, it was observed that the Higuchi model did not fit well with the experimental data on the permeation of polymeric films, as it presented lower coefficients of determination. This can be explained by the fact that the polymeric films exhibit a high degree of swelling, which was not predicted by this model [48]. However, this model applies well to MUP permeation without polymeric films, because swelling does not occur. The polymeric and MUP films also fit the zero-order model because the coefficient of determination was close to 1.0. This indicated that the permeation and release processes were slow [49]. The Peppas-Korsmeyer model, all the samples presented a coefficient of determination close to 1.0, indicating a good fit with the experimental data. For polymeric films, all values of n were greater than 0.89, indicating that the mechanism of super-relaxation of polymeric chains is decisive for the release and permeation of MUP; however, for MUP permeation, the value of n was 0.7070, indicating non-Fickian transport [19]. The Weibull model, a good fit of the experimental data was also observed; therefore, it was possible to analyze parameter T₀, which presented values close to 10 min in all experiments. This may indicate that the polymeric films have MUP on their surface, which is immediately released; this is why the T₀ values of the films and isolated MUP are approximately equal [19].

Polymer blends often exhibit compositional and morphological heterogeneity that affects the overall properties of the material. Therefore, it is possible to better understand the influence of heterogeneity on material properties by obtaining a high-resolution spatial distribution of the components of a sample composed of different polymers. In this context, the spatial distribution of molecular species in polymer mixtures can be studied using micro-Raman image analysis [50]. The results demonstrate the presence of κ-carrageenan in all points of the sample, as well as homogeneity at a spatial resolution of approximately 1 µm. This may indicate the presence of similar properties in the different areas of the samples [51]. Similarly, the data obtained by SEM do not indicate the presence of phase separation in the samples.

Regarding microbiological tests, it can be observed that the positive control demonstrated a more pronounced inhibitory effect on all bacterial strains, with a statistically significant difference between the various films (p < 0.05). With respect to the polymeric films, the results for S. aureus, S. epidermidis, and P. aeruginosa were comparable, as evidenced by an adjusted p-value exceeding 0.05 in all comparisons. Regarding the inhibitory activities on E. coli, it can be observed that the F2 and F3 films show comparable inhibitory effects and more pronounced than those observed in the F1 sample. Similar findings have been previously documented in the scientific literature. Films composed of chitosan, sodium alginate, and Carbopol^® loaded with MUP demonstrated antimicrobial activity against a range of bacteria, including E. coli (ATCC 8739), Enterococcus hirae (ATCC 10541), S. aureus (ATCC 6538), P. aeruginosa (ATCC 27853), Bacillus cereus (ATCC 7064), and Klebsiella pneumoniae [1].