Amphotericin B, Sabouraud 2% Dextrose Agar (SDA) and silver nitrate (AgNO₃, 99.98%), were purchased from Merck. Chitosan (MW 50–190 kDa, deacetylation degree: ≥85%), Polyvinyl Alcohol (PVA, 99%), RPMI 1640 [supplemented with L-glutamine, 0.2% (w/v) of glucose and without sodium bicarbonate], Fetal Bovine Serum (FBS), L-cysteine (C₃H₇NO₂S, >99%), Cetyl trimethyl ammonium bromide (CTAB, 99%), Bovine Serum Albumin (BSA) and itraconazole (ITC, ≥98% Thin-Layer Chromatography) were purchased from Sigma-Aldrich, United States. Brain Heart Infusion (BHI) was acquired from Oxoid, Brazil. All chemicals used in this research were used without further purification.

Sporothrix brasiliensis ATCC 5110/MYA4823 and Sporothrix schenckii ATCC 1099-18/MYA4821 strains were obtained from the American Type Culture Collection (Manassas, VA 20108 United States) and stored on BHI plates at 4°C or cryopreserved at −20°C in BHI broth with 10% glycerol. For the inoculum of the in vitro tests, Sporothrix cells were cultured in BHI broth at 37°C to favor the development of their yeast phase. For the in vivo experiments, the inoculum was prepared as previously described (Artunduaga Bonilla et al., 2021b). Briefly, S. brasiliensis was cultivated on BHI plates for 7 days at 25°C prior to use. Cells (2 × 10⁶) were transferred to flasks containing 150 ml of BHI and then incubated at 37°C/150 rpm for 4 days. Yeasts were collected by centrifugation at 3.500 rpm for 10 min (4°C) and then washed three times in 0.1 M phosphate-buffered saline pH 7.4 (PBS). Yeast suspension (2 × 10⁸ cells/ml) were assessed for viability by trypan blue staining. Suspensions with cell viability greater than 90% were used for the tests (Della Terra et al., 2017; Artunduaga Bonilla et al., 2021b).

The silver chitosan nanocomposites tested in the present work were prepared as previously described by our group (Artunduaga Bonilla et al., 2021a). Briefly, a solution containing 4 ml of 0.01 M AgNO₃, 800 µL of 0.01 M cysteine and 1.2 ml of 0.01 M CTAB were mixed and stirred using a magnetic bar. Deionized water was added to a final volume of 20 ml and PVA added at a final concentration of 0.02% (w/v). Then, the suspension was sonicated at 37 KHz during 2 h. The appearance of a red-orange suspension indicates that AgNPs was formed (Artunduaga Bonilla et al., 2015; Almeida-Paes et al., 2017). Synthesis was performed at room temperature and in the presence of light. Once the AgNPs were characterized, additional steps were taken to incorporate them into chitosan. Initially, 10 ml of AgNPs (298 μg/ml) was added to 10 ml of chitosan (1%, w/v) in 1% acetic acid and the pH was adjusted to 12 with NaOH. The mixture was subjected to five cycles of 5 s pulses alternating with pauses of 5 s, using a sonicator probe (Sonopuls HD3200) at 150 W. The suspension was then stirred for 1 h at 300 rpm and room temperature (Artunduaga Bonilla et al., 2021a). Finally, the suspension pH was adjusted to 7.5 with NaOH. The nanocomposites were stored in amber flask at room temperature.

Physicochemical properties of the nanocomposites were determined according to our previous publication (Artunduaga Bonilla et al., 2021a). Succinctly, the AgNPs spectrum in the UV-visible region was observed using a spectrophotometer from wavelength of 300–600 nm. To determine the average hydrodynamic size and surface charge of the nanomaterial, Dynamic Light Scattering (DLS) and Zeta potential analysis were performed using a Brookhaven instrument (United States) with suspension at 100 μg/ml concentration. Quantification of silver using a ContrAA 700—High-Resolution Continuum Source Graphite Furnace Atomic Absorption Spectrometer (HR-CS GF AAS), Analytik jena AG, was performed. For all tests, the AgNPs@Chi concentration corresponds to the silver concentration. The size, morphology and crystalline nature of the AgNPs, were determined by transmission electron microscopy (TEM) and X-ray diffraction (XRD). For these analyses, a drop of AgNPs was dispensed on a Formvar coated copper grids (300 mesh) (polyvinyl resin, Ted Pella, Inc.) and dried in a vacuum desiccator. Images were acquired using a JEOL 1200 EX and FEI Tecnai G2 FEG transmission electron microscopes operating at 80 and 200 kV, respectively. For the AgNPs@Chi, the characterization was performed by Zeta potential analysis and TEM respectively, to compare the surface charge and observe the distribution of the AgNPs within the polymer.

In vitro silver release from chitosan matrix was carried out as previously demonstrated in our recent work (Artunduaga Bonilla et al., 2021a). Briefly, 50 μg/ml aliquots of AgNPs@Chi were added to a small dialysis bag and placed in a beaker containing 20 ml of 85 mM phosphate buffer, pH 7.4 and incubated at 37°C/100 rpm. Aliquots of the buffer (2 ml) were collected sporadically until day 52 and the concentration of silver diffused through the dialysis membrane was measured by atomic absorption spectroscopy.

Anti-Sporothrix tests were performed following the Clinical and Laboratory Standards Institute (CLSI) M27-A3 protocol with cell suspension adjustments. Yeasts were cultivated in BHI broth at 37°C with orbital agitation (150 rpm) for 3 days. Cells were then washed in PBS and adjusted to 5 × 10⁵ yeasts/ml in RPMI 1640. NCs were diluted in RPMI, and 100 µl volumes were added to 96-well plates in serial dilutions from 16 to 0.12 μg/ml. A volume of 100 µl of the yeast suspension was added to each well. Microplates were incubated at 37°C for 3 days. RPMI 1640 and itraconazole (ITC) were used as controls. The fungal growth was measured spectrophotometrically (λ: 530 nm) using a BioTeK ELx808 microplate reader (Netherlands). Minimal Inhibitory Concentration (MIC₅₀ and MIC₉₀) were considered as the lower concentrations of NCs able to reduce microbial growth in 50 and 90%, respectively, comparing to control conditions. Subsequently, 100 µl aliquots from each of the MIC₉₀ wells were suspended in 900 µl of RPMI and then incubated at 37°C for 48 h. Aliquots of 20 μl were then seeded in BHI plates and incubated for 48 h to check absence of fungal growth. In order to confirm the non-antifungal effect of the chemical reagents used to synthesize the silver nanocomposites, 100 µl aliquots of each working solution (CTAB, PVA, Chitosan) were incubated with Sporothrix yeasts following the broth microdilution protocol in 96-well plates, under the same conditions used to test NCs. All the experiments were carried out in triplicates.

The antifungal combinatory effect of NCs with ITC or amphotericin B (AMB) was investigated using the checkerboard test (Zhao et al., 2019). Tests using 96-well plates were performed according to methods of previous studies (Zhao et al., 2019; Artunduaga Bonilla et al., 2021a). Serial two-fold dilutions of each drug (AgNPs, AgNPs@Chi, ITC and AMB) were prepared in RPMI. The 96-well microplates were inoculated with 100 µl of 5 × 10⁵ cells/ml and incubated at 37°C for 3 days. The optical density (OD) was measured at 530 nm using a microplate reader (Biotek ELx808). The fractional inhibitory concentration index (FICi) was defined as (MIC combined/MIC drug A alone) + (MIC combined/MIC drug B alone). The drug interaction was considered synergistic when the FICi is ˂0.5, indifferent when the FICi is between 0.5 and 4, and antagonistic when the FICi is >4 (Zhao et al., 2019; Artunduaga Bonilla et al., 2021a).

Yeasts were treated with MIC₅₀ of the NCs (0.06 and 0.25 μg/ml of AgNPs and AgNPs@Chi, respectively) in RPMI 1640 at 37°C for 48 h. Then, cells were washed twice in PBS and fixed overnight with 2.5% glutaraldehyde, 4% formaldehyde and 10 mM calcium chloride in 0.1 M cacodylate buffer, pH 7.2 at 4°C. Post-fixation in 1% osmium tetroxide in cacodylate buffer containing 0.08% potassium ferrocyanide was performed for 2 h at room temperature. Subsequent, the fungal cells were dehydrated in acetone and embedded in spurr’s resin (Ted Pella Inc.). Ultrathin sections (70 nm) were acquired using a ultramicrotome (LEICA EM UC6). Sections placed onto copper grids (300 mesh) were stained with lead citrate and uranyl acetate and visualized in a transmission electron microscope (JEOL 1200 EX) operated at 80 Kv.

The hemolytic effect of silver nanocomposites was determined according to Shiny, Mukherjee, and Chandrasekaran 2014, with some modifications. Briefly, red cells present in defibrinated sheep blood (Ebafarma) were washed three times with PBS (1,000 g, 3 min). Then, 100 μl of 2 × 10⁸ cells/ml were added to 96-well plates with 100 μl of NCs in serial dilution from 16 to 0.25 μg/ml. The combinations were incubated at 37°C for 1 h. Relative hemoglobin concentration in supernatants after centrifugation at 1700 g for 5 min was monitored by measuring optical absorbance (Abs) at 540 nm using a microplate reader (Biotek ELx808). Red blood cells incubated with PBS were considered as a negative control, and those incubated with water were used as positive controls for hemolysis. Percent hemolysis was calculated using the formula: Hemolysis (%) = 100 * (Abs treated cells−Abs negative control)/(Abs positive control−Abs negative control). The experiment was carried out in triplicate and three independent tests were performed.

The scratch assay was performed according to You et al. (2017). Briefly, a total of 10⁶ cells of spontaneously immortalized human skin keratinocyte line (HaCaT), immortalized human fibroblasts (HFb) and NIH/3T3 murine fibroblasts were plated separately in six-well plates to obtain a confluent culture. The monolayer was then scraped straight with a 200 µl micropipette tip. Thereafter, cell debris was removed by washing the cells twice with PBS or culture medium. Silver nanocomposites suspended in RPMI 1640 and supplemented with 0.2% FBS (2 ml) were added carefully in the wells at concentration of 1 μg/ml of silver and immediately incubated at 37°C and 5% CO₂ for up to 72 h. To monitor the cell migration or scratch closure, the plates were placed under an inverted Olympus CKX41 microscope coupled to an AmScope MU500 camera and images were acquired every 24 h. The test was performed in three independent assays by duplicates.

The efficacy of the silver nanocomposites was investigated using a murine model for subcutaneous sporotrichosis (Della Terra et al., 2017). Groups of five female C57BL/6 mice weighing 16–18 g (7–8 weeks old) were obtained from the Núcleo de Criação de Animais de Laboratório of the Universidade Federal do Rio de Janeiro (UFRJ, RJ, Brazil). To develop the infections mice were inoculated by subcutaneous injection of 25 µl (5 × 10⁶ yeasts) into the left hind footpad, previously anesthetized with ketamine (100 mg/kg, ip) and xylazine (10 mg/kg, ip). Daily treatment with silver nanocomposites was started after lesion development (∼20 days). The application of AgNPs or AgNPs@Chi was by spray (three applications of ∼100 μl/at a time, from a suspension with 5 μg/ml of silver). The negative controls (ITC or PBS) were administered intravenously (i.v). Two independent tests with treatment and monitoring of injures were performed. The body weight, the paw diameter and the survival pattern were monitored in the second test. Digital Vernier caliper was used to measure the diameter of the paw.

The experimental data were analyzed by one-way ANOVA and multiple comparisons tests (Sidak’s, and Dunnett’s for in vitro tests and Tukey’s for the in vivo experiments data) using Graph Pad Prism seven software. Statistical significance was considered at p value ˂ 0.05.

Synthesis and characterization of the silver nanocomposites were carried out following previously published protocols (Artunduaga Bonilla et al., 2021a). The NCs used here displayed the same physical-chemical characteristics as observed previously by our group (Supplementary Figures S1, S2 for the physicochemical details).

Treatment with AgNPs inhibited ≥90% of S. brasiliensis and S. schenckii growth at concentrations of 0.12 and 0.25 μg/ml, respectively (Table 1). AgNPs functionalized with chitosan (AgNPs@Chi) displayed the same inhibitory activity at higher concentrations, ranging from 0.5 to 1 μg/ml. Thus, a 4-fold reduction in activity was observed. Remarkably, the effect of NCs was comparable and even more efficient than the activity exhibited by itraconazole (1 μg/ml for both species), a first-line drug to combat sporotrichosis. The reduced antifungal activity of AgNPs embedded in the chitosan occurred most probably due to the slower release of silver from the polymeric matrix. Similar concentrations of chemical reagents used as working solutions for the NCs synthesis did not impact fungal growth. The optical densities (OD) of the wells corresponding to the cells exposed to the chemical reagents ranged from 0.296 to 0.324, values similar to the fungal growth under control conditions (RPMI) (0.330 ± 0.08 OD).

To explore the possibility of using silver nanocomposites combined with ITC or AMB as an unconventional therapy against Sporothrix, the combinatory effect was studied by the checkerboard method. As seen in Table 2, none of the combinations showed synergism with either ITC or AMB against the S. brasiliensis and S. schenckii strains tested. We observed an indifferent or non-interactive effect (FICi ˃0.5–4) for all experiments.

Under control conditions, yeasts of S. brasiliensis showed a uniform cell wall and a continuous plasma membrane. Intracellularly, the yeasts displayed regular nucleus and mitochondria with a homogeneous electron-dense cytoplasm (Figures 1A,B). Treatment with AgNPs promoted deformities in yeast morphology (Figures 1C,D). The exposition of the fungus to AgNPs promoted plasma membrane discontinuity, rupture of organelles and a heterogeneous density at the cytoplasm (Figure 1D, white arrows). In addition, the cell wall displayed a decreased electron-density. The detachment of the cytoplasmic membrane was also observed with the disintegration of the surface of the cells treated with AgNPs@Chi (Figures 1E,F). The images are representative of the effects observed in 20–30 cells viewed at high magnification and for each treatment. The ultrastructural characteristics described for the cells treated with the experimental nanocomposites correspond to the alterations observed in more than 80% of the cells visualized by each group.

The hemolytic concentration of NCs capable of lysing 50% (HC₅₀) of the red blood cells was approximately 2 and 8 μg/ml for AgNPs and the functionalized compound, respectively (Figure 2). For concentrations below 1 μg/ml, a modest variability in toxicity was observed, ranging between 20 and 30% when using AgNPs, and 15–18% for the AgNPs@Chi. The cytotoxic effect was directly relative to the concentration of the NCs, with a reduced activity observed for the functionalized compound. Red blood cells in presence of PBS showed no signs of hemolysis.

Keratinocytes and fibroblast proliferation and/or migration was individually visualized by the scratch assay. The effect on cellular migration/tissue repair was investigated at 1 μg/ml of NCs, an efficient concentration to inhibit Sporothrix spp. growth. Figure 3 demonstrates that restoration of cell density for all cell lines was faster in the group stimulated with AgNPs@Chi. For HaCaT, cellular restoration in the presence of nanoparticles was considerably higher than control conditions (Figure 3A). For these cells the migration was quantified by counting their number in the scratch area, delimited by red lines, at each time point, confirming a better wound closure (p ˂ 0.05) after 48–72 h of exposure to the functionalized compound (Figure 3B). The cell density of human (Figure 3C) and murine (Figure 3D) fibroblasts, after 24 h of exposure to nanocomposites was considerably higher, making the cell number determination inaccessible. Qualitatively, a higher cell convergence with AgNPs@Chi was observed, reaching the total closure of the scratch in the first 48 h.

Considering the anti-Sporothrix efficacy of NCs in vitro along with a potential activity to stimulate keratinocytes and fibroblasts migration, we investigated their effect in a murine sporotrichosis model. The timeline followed for in vivo testing is shown in Figure 4. For the first test (Figure 4A), treatment was started on day 25 post-infection, once swelling and skin injure were evident in the paw of the mice. (Figure 5, day 1). AgNPs, AgNPs@Chi and ITC were administered separately. Treatment was performed daily for 35 days (Figure 5). Under these conditions the paw swelling of the control mice continued even at day 35 although the lesion has diminished. Treatment with AgNPs displayed paw recovery with swelling reduction at day 25. Finally, treatment with AgNPs@Chi generated the fastest recovery, reducing inflammation and skin lesion within 15 days of treatment (Figure 5). At the end of the experiment (day 35), animals treated with NCs showed complete recovery from the infection. Monitoring of the mice was performed for 3 weeks after the last day of treatment. Silver nanocomposites-treated mice had no development of new injures. Mice treated with ITC displayed inflammation reduction at endpoint test when the control group showed chronic lesions with mild variations in size and swelling. Remarkably, about 20% of the mice under control conditions died (data not shown).

Considering that the strain exhibited low virulence according to the minor lesions observed in mice, we used a recent pick from the same strain (S. brasiliensis 5110) with two passages in culture medium from frozen stock to ensure the protecting effect of the NCs. In this case, a severe infection was since the skin lesions were more serious (Figure 6A) and both swelling and injures were developed in less time, at 15 days’ post-infection (Figure 6, day 1). Treatment was started on day 15 after yeast’ inoculation, as indicated in the timeline scheme (Figure 4B). The NCs were administered in the same way as in the first test. The course of the infection was monitoring by the measure of the paw swelling and body weight.

The silver nanocomposites treatment reduced the severity of the infection resulting in a better prognosis since there was no loss of the paw, as observed at day 10 in the control group. Although all mice from the control group displayed a severe tissue loss, swelling measurements were continued at the base of the paw to monitor the infection progress, moving to the surrounding tissues. In this sense, from day 10 the swelling data correspond to the diameter of the paw’s base. Minimal variations were observed in the diameter of the base of the paw with no signs of fungal dissemination, suggesting that the infection was localized. Weight loss was lower in mice from the group treated with AgNPs@Chi comparing to control (p = 0.001) and ITC (p ˂ 0.05) groups (Figure 7A). In addition, AgNPs@Chi-treated mice showed paw swelling reduction, being significantly different from the ITC-treated mice (p ˂ 0.001) suggesting a decreased inflammatory response (Figure 7B). Until day 24 of treatment, the mice treated with NCs no showed significant difference in weight variation (Figure 7A). From day 14, a greater paw swelling of AgNPs-treated mice was observed when contrasted to those treated with ITC (p ˂ 0.05) and with the functionalized compound (p ˂ 0.001) (Figure 7B). AgNPs@Chi treated mice showed 70% survival, while 50% of those treated with AgNPs survived at the end of the experiment (Figure 6B).