**3. Discussion**

For future reproducibility of the experiment, the extract obtained by maceration followed by percolation was duly characterized in relation to dry matter, total phenolics content, ellagic acid, and pH. Total phenolics were determined only in samples that contained the pomegranate peel extract, and then the chemical formulation did not present any phenolic content in its composition. Polyphenols are effective hydrogen donors and are correlated to the number and position of hydroxyl groups and conjugations as well as the presence of donor electrons in the aromatic ring B, because of the ability of this aromatic ring to withstand the electron depletion located in the π electron system [22]. The antimicrobial activity of various polyphenols and plant extracts have been evaluated in pharmaceutical and food studies [23,24]. Some phenolic compounds present in sage, rosemary, thyme, hops, coriander, tea, cloves, and basil are known to exhibit antimicrobial effects against foodborne pathogens. Their mechanisms of action need to be further elucidated, and might be due to a plethora of phenolic compounds present in a very single plant extract. Furthermore, as the bioactive compounds in the extract presented antioxidant and anti-inflammatory activities, the antimicrobial potential of the pomegranate peel extract in the in vivo trials could show better results, and should be strongly stimulated in further studies. Regarding the multi conceptual nature of the term antioxidant and bringing it into the context of this study, some polyphenols present in low concentrations could prevent or reduce the extent of oxidative damage in mammalian cells. Taken together, these natural biomolecules could indirectly protect the cells and reduce the cytotoxicity of silver nanoparticles.

The correct selection of the plant and the standardization of the methods to obtain the extracts to be used as reducing or capping agen<sup>t</sup> in the nanosynthesis of metal particles should be preponderant when the green process is elected for the production of products in large scale. Additionally, a plethora of plants used in the phytosynthesis of metal nanoparticles [25–27] and the lack of information of the

extraction techniques used in the articles has hindered the comparison of the present results with those found in the literature. For instance, different values and methods of total phenolics quantification can be observed in the literature as described by Kalaycioglu et al. (2017) [28]. Similarly, other factors can interfere in the evaluation and comparisons of the extracts such as the chemical and genotypic composition of the plant, the variety and the soil type, the place of the plant origin, the harvest season, maturation method, aside from the solvent and the process used for the obtention of the pomegranate extract, among others [29].

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images showed the smallest particles obtained by conventional chemical synthesis, and DLS data confirmed these findings with mean sizes of 89 and 19 nm for green and chemical nanoparticles, respectively. The fission of colloidal particles of different sizes and shapes may be related to additives (salts, polymers), solvent properties (boiling temperature, affinity with created surfaces), the addition of nucleation, among others [30]. The reagents used in the chemical synthesis would produce particles with more predictable characteristics than the several substances and compounds present in the plant extract and used in the phytosynthesis route, which would interfere with the size and form of the nanoparticles and make phytosynthesis a challenge in controlling the reaction process and the morphological aspects of the particles. Moreover, the presence of different bioactive substances in the extract would reduce only a fraction of the silver ions present in the solution. The remaining silver ions would form other nuclei and further the growth of the previously formed silver nanoparticles [31]. This process is called Ostwald Ripening, where the largest particles consume the smaller ones and grow larger, where the dissolution of the smaller ones and deposition of ions on the surface of larger ones occur [32].

Almost 100% of ions reduction was observed for both synthesis routes. However, when the chemical silver nanoparticles were added to the formulation, a dissociation of ions from nearly 30% was observed when compared to chemical silver nanoparticles alone. This fact could be due to the presence of the components as carboxymethylcellulose and propylene glycol in the spray formulation which possibly favored the silver ion dissociation into the system [33]. The presence of oxygen or ligands for Ag+ in the formulations may increase the dissolution rate of AgNP and lead to increased dissolution through the formation of Ag+ complexes [34]. Ag+ in solution will interact with various ions and molecules that are present in aqueous media. Important ligands to be considered for Ag+ are sulfide and organic ligands such as the carboxylic acids group which are used as Ag coatings (e.g., citrate, lactate). Carboxyl ligands such as carboxymethylcellulose strongly bind Ag+, which may affect the dissolution of AgNP and the bioavailability of Ag+ [35].

Furthermore, the size of the Ag in the NP affects the extent and kinetics of the AgNP dissolution as the smallest nanoparticles dissolve faster and to a greater extent [36]. This would explain the difference in the dissolution of the nanoparticles in the formulations. Their dissolution is of high relevance for the possible toxic effects of AgNP as Ag+ appears in many cases to determine their toxicity [37]. This fact was not observed when green-synthesis was carried out. This could be related to several compounds present in the extract which would readily react with the released silver ions, or the encapsulation of the silver nanoparticles promoted by those phytocompounds may have avoided the silver ions dissociation from the silver nanoparticles and its release to the solution.

Zeta potential test demonstrated the stability of the silver nanoparticles, most notedly in the spray formulations. Electrical charges on the surface of the nanoparticles prevent agglomeration, and thus afford the stability of the nanoparticles [38,39]. Indeed, silver nanoparticles and spray formulations presented a mean of 70 mV, which indicates their high stability of silver nanoparticles [40].

Antimicrobial results are also promising for the silver nanoparticles as well as the pomegranate extract obtained. The formulations notably showed better results when compared with the input active only. This fact could be explained for the proper dispersion of the active inputs (silver nanoparticles and pomegranate peel extract) in the spray formulation. Additionally, a synergistic effect could have occurred between those active inputs and the methylparaben present in the formulation. In the literature, studies with an antimicrobial effect of pomegranate extract were conducted against *Staphylococcus aureus*, *Enterobacter aerogene*, *Salmonella typhi*, and *Klebsiella pneumonia* [41]. The MIC values obtained in this study for pomegranate extract were in accordance with Bakkiyaraj et al. (2013) [42] for both the microorganisms studied, and a difference was observed in *C. albicans*, but this fact may be explained by the difference between the *C. albicans* strains used in the studies.

Chemical-silver nanoparticles, in formulation or not, produced MIC values against *S. aureus* about 10-fold lower than those produced by Prema et al. (2017) [33] (60 μg/mL), who also produced silver nanoparticles stabilized with CMC. Indeed, the antimicrobial activity of chemical silver nanoparticles was also determined by Monteiro et al. (2011) [43] with MIC values for *C. albicans* (0.5 μg/mL) in accordance with this present study.

Noteworthy is the difference found in the present study in respect of cytotoxicity between the chemical and green routes to obtain silver nanoparticles. Studies have shown that silver nanoparticles produced with *Protium serratum* and *Nyctanthes arbortristis* extracts were biocompatible when tested in L929 fibroblasts [44,45]. It is believed that what makes the silver nanoparticle toxic to human cells is the type of reducing agen<sup>t</sup> used such as sodium citrate or sodium borohydride [46]. Even in conventional syntheses of silver nanoparticles, reagents are used that prevent the aggregation of these nanoparticles [47], which may further favor their cytotoxicity.

In the case of phytosynthesis of metal nanoparticles, plant extracts, aside from acting as reducing agents, would act to stabilize the particles against dissolution, hence reducing the toxicity of the silver nanoparticles solution. Furthermore, it is possible that some compounds in the extracts may have a synergistic effect with the silver nanoparticles [48], making them less toxic to human cells. Furthermore, extracts of *Punica granatum* have exhibited antioxidant [49] and anti-inflammatory [50] activity, and may have contributed to reducing the cytotoxicity of green- in comparison with chemical-silver nanoparticles.

In general, the stability assay (silver ions dosage, zeta potential, and antimicrobial activity) showed a high stabilizing capacity of the formulations. However, the spray formulations of green silver nanoparticles and pomegranate peel extract showed a significant reduction in the content of total phenolics in 14 and 28 days. The decrease in the content of total phenolics may have occurred due to the temperature variations inherent in the stability test, as occurred in the study of [51] where the temperature affected the total phenolics content in the roselle-mango juice blends. Moreover, in formulations containing green-silver nanoparticles, the components of the extract may have been degraded or associated with the nanoparticles, explaining the faster decrease of the total phenolics content when compared to the pomegranate extract formulation. Interestingly, ion dosage, zeta potential, and antimicrobial activity were not affected by different conditions of temperature, time, and humidity of the stability test.

Altogether, the reported results sugges<sup>t</sup> that the plant extract mediated syntheses of AgNP showed a pronounced lower cytotoxic effect in mouse fibroblast cells (L929) than the syntheses of AgNP by the chemical method. Of note is the implication that different sizes between the green- chemical-AgNP as well as the expected impurities sedimented on both obtained nanoparticles could have had on their toxicity. Although it is quite tricky to obtain AgNP with a well-defined form and size and prevent the particles aggregation [52], it is of importance to complement and support our findings, then strongly recommend an eco-friendly approach to producing green-AgNP and prototype wound-care sprays containing these particles.

#### **4. Materials and Methods**

#### *4.1. Plant Material and Preparation of Pomegranate Peel Extract*

Pomegranate samples were collected from a crop cultivated in Eixo (21◦0801 S, 51◦0606 W), Mirandópolis, São Paulo, Brazil, during May 2015. Pomegranate peels were separated and stove-dried at 50 ◦C, ground, and sieved to a granulometry lower than 2 mm. Peels were submitted to alcohol extraction using 70% ethanol by maceration, followed by the percolation process [53]. The extract was

characterized in relation to pH, dry matter, and total phenolics expressed as gallic acid. The chemical marker of pomegranate, ellagic acid, was also identified and quantified.

#### 4.1.1. Determination of Total Phenolics, pH, and Dry Matter

To determine the total phenolics, an analytical curve of gallic acid (Sigma-Aldrich Chemical Co., St. Louis, MO, USA) was carried out [54]. All extracts obtained and the standard solution of gallic acid were prepared in 50 mL volumetric flasks using water as the solvent. The samples were homogenized and, the flasks were brought to the ultrasonic bath for 30 min. A 0.5 mL aliquot was transferred to another 50 mL flask where 2.5 mL of Folin-Denis reagen<sup>t</sup> (Qhemis-High Purity, Hexis, São Paulo, Brazil) and 5.0 mL of 29% sodium carbonate (Cinética, São Paulo, Brazil) were added. The samples were protected from light and the readings were performed after 30 min in a UV-Vis spectrophotometer at 760 nm [53]. The pH was measured direct from a solution of 1% extract, using a pH kit (Merck KGaA, Darmstadt, Germany) and dry matter was calculated after drying on a sample stove at 105 ◦C and was expressed in percentage *<sup>w</sup>*/*<sup>w</sup>*. All data were analyzed in triplicate.

#### 4.1.2. Determination of the Ellagic Acid Content

A Shimadzu liquid chromatograph and a Shimpack ODS C18 (Shimadzu Corporation, Kyoto, Japan) reverse phase column (100 mm × 2.6 mm) were used to determine the ellagic acid content by high performance liquid chromatography (HPLC). Analytical conditions were optimized based on de Sousa et al. (2007) [55] with modifications. As the mobile phase, HPLC grade methanol and a 2% aqueous acetic acid solution with gradient elution (0–7 min, 20–72.5% v/v methanol, 7–7.5 min, 72.5–95% v/v methanol, 7.5–8.5 min 95% v/v methanol, 8.5–9 min 95–20% v/v methanol, 9–10 min 20% v/v methanol) were used. The flow rate was 1.0 mL/min, and the separation was achieved at 25 ◦C. The injection volume was 5 μL and the wavelength used was 254 nm. Peaks were determined by comparison with an authenticated ellagic acid standard. Briefly, the sample was transferred to a 20 mL volumetric flask which was diluted with HPLC grade methanol. Extraction was undertaken using a vortex for 5 min and ultrasonic bath for 1 h. For the extracts, samples were transferred to volumetric flasks of 10 mL, using methanol HPLC as the solvent. All samples were vortexed for 5 min and sonicated for 30 min. Samples were filtered through 0.45 μm filter. All samples were prepared in triplicate.

#### *4.2. Synthesis of Green-Silver Nanoparticles*

The protocols described by Gorup et al. (2011) [56] and Das et al. 2015 [57] with modifications were used to produce silver nanoparticles. Briefly, 3.5% of carboxymethylcellulose (CMC) (Labsynth, Diadema, Brazil), 20% of propylene glycol (PG) (Labsynth, Diadema, Brazil), 100 mM of silver nitrate (SN) (Merck KGaA, Darmstadt, Germany), pomegranate peel extract at 30 mg/mL, and water to make up 100% of the samples were used. Silver nanoparticles were not purified relative to the excess reagents. The reaction was carried out at 50 ◦C for 12 min, and it was selected based on previous results.

#### *4.3. Synthesis of Chemical-Silver Nanoparticles*

Chemical-silver nanoparticles were produced according to Gorup et al. [53]. AgNO3 (Merck KGaA, Darmstadt, Hesse, Germany) was dissolved in water, and brought to boiling at 90 ◦C. After 2 min of boiling, an aqueous solution of sodium citrate (Na3C6H5O7) (Merck KGaA, Darmstadt, Hesse, Germany) was added, and kept boiling for another 6 min until the solution reached a yellow amber color. The stoichiometric ratio was 1:3, respectively for AgNO3 and Na3C6H5O7. Silver nanoparticles were not purified relative to the excess reagents.

#### *4.4. Preparation of the Spray Formulations*

The reagents used were CMC (Labsynth, Diadema, Brazil), PG (Labsynth, Diadema, Brazil), and methylparaben (Labsynth, Diadema, Brazil) in a proportion of 0.1%, 7%, and 0.1%, respectively. The active inputs (green- or chemical-silver nanoparticles and pomegranate peel extract) concentrations were based on the minimum inhibitory concentration and cytotoxicity. Therefore, the final concentrations of active inputs in the spray formulations were: 337.5 μg/mL of green-silver nanoparticles, 5.55 μg/mL chemical-silver nanoparticles, and 94 μg/mL of crude peel extract dry matter.

#### *4.5. Characterization of the Silver Nanoparticles and the Spray Formulations*

#### 4.5.1. X-ray Diffraction (XRD), Dynamic Light Scattering (DLS), and Zeta Potential Analysis

A Shimadzu XRD diffractometer with a Cu Kα radiation operating at 30 kV and 30 mA and 2θ range from 35◦ to 85◦ with step scan of 0.02◦ and scan speed 0.2◦·min−<sup>1</sup> was used to perform XRD analysis. To collect silver nanoparticles patterns, the nanoparticles were deposited on the surface of a silicon substrate (Si) by dripping the aqueous colloidal dispersion on the substrate at room temperature until the solvent had evaporated.

DLS experiments were performed at room temperature and at a fixed angle of 173◦ on a Zetasizer Nano ZS (Malvern Instruments Ltd., Malvern, UK) equipped with 50 mW 533 nm laser and a digital auto correlator. The number-average values obtained were compared to the size distributions of the silver nanoparticles. For the zeta potential test a Zetasizer (Malvern instruments, Malvern, UK) with an MPT-2 titrator was used. Aliquots from each test suspension were obtained to conduct zeta potential, and mean values were obtained from three independent measurements.

## 4.5.2. TEM Analyzes

The nanocompounds morphology was characterized by TEM images in a Jeol JEM-100 CXII (JEOL USA Inc., Peabody, St. Louis, MO, USA) microscope equipped with Hamamatsu ORCA-HR digital camera.

#### *4.6. Silver Ions Dosage*

The dosages of free silver ions (Ag+) present in the compounds and spray formulations were performed to observe if the total amount of Ag added in the synthesis reaction was successfully reduced. A specific electrode 9616 BNWP (Thermo Scientific, Beverly, MA, USA) coupled to an ion analyzer (Orion 720 A+, Thermo Scientific, Beverly, MA, USA) was used. A 1000 μg Ag/mL standard was prepared by adding 1.57 g of AgNO3 to 1 L of deionized water. The combined electrode was calibrated with standards containing 6.25 to 100 μg Ag/mL to achieve equivalent silver concentrations in the compounds. A silver ionic strength adjuster solution (ISA, Cat. No. 940011) that provided a constant background ionic strength was used (1 mL of each sample/standard: 0.02 mL ISA).

#### *4.7. Stability Test of the Spray Formulations*

The spray formulations were submitted to a stability test with controlled conditions of temperature and time. This test was based on Anvisa protocols (Cosmetics stability guide ISBN 85-88233-15-0; Copyright© Anvisa, 2005) and the guide to stability studies (Ordinance No. 593 of 25 August 2000). Briefly, samples of each spray formulation were submitted to alternating cycles of temperature daily ranging from 40 to −5 ◦C for 28 days. The tests selected to evaluate the stability of the samples were ion dosage, total phenolics content, zeta potential, and minimal inhibitory concentration (MIC). All tests were done in the same conditions as described before, and were carried out at 0, 7, 14, and 28 days.

#### *4.8. Antimicrobial Activity of the Silver Nanoparticles and the Spray Formulations*

Minimal inhibitory concentration of the silver nanoparticles samples were determined following the instructions of the Clinical Laboratory Standards Institute with some modifications. The samples were first diluted in water and subsequently in culture medium specific for each microorganism, Mueller Hinton broth (BD Difco, Franklin Lakes, USA) for *Staphylococcus aureus* (ATCC 25923), and RPMI (Sigma-Aldrich, St. Louis, MO, USA) for *Candida albicans* SC 5314) [58]. The microorganisms were adjusted to 5 × 10<sup>5</sup> cells/mL for *S. aureus* and 5 × 10<sup>3</sup> cells/mL for *C. albicans*, and the plates were incubated for 24 h and 48 h in aerobiosis at 37 ◦C for *S. aureus* and *C. albicans*, respectively. After incubation, the plates were visually read. The assays were performed in triplicate.

## *4.9. Cytotoxicity Analysis*

For the evaluation of cytotoxicity, fibroblast cells of the L929 lineage were used. Cells were cultured in DMEM culture supplemented with 10% fetal bovine serum (FBS), penicillin G (100 U/mL) (Gibco®, Carlsbad, USA), streptomycin (100 μg/mL), amphotericin B (25 μg/mL) and incubated in a stove at 37 ◦C with 5% CO2. Cells were subcultured (5–7 days), using 0.9% saline to wash them and 0.25% trypsin to disintegrate them from the vial. After disruption, these cells were centrifuged at 1000 rpm for 10 min at 10 ◦C, resuspended in complete DMEM medium (supplemented with FBS), and cell counted in a Neubauer's chamber.

The sub-cultured third to eighth passage fibroblasts were inoculated into 96-well microplates at a density of 0.5 × 10<sup>5</sup> cells/well. They were then incubated at 37 ◦C with 5% CO2. After 24 h, 20 μL of different dilutions of each sample were added to the wells of the plate containing the cells in medium not supplemented with SBF (incomplete medium) and incubated. Twenty-four hours post-treatment, the medium was withdrawn, cells were washed with saline and 20 μL of resazurin (Sigma-Aldrich) 0.01% *w/v* in deionized H2O was added to each well containing 180 μL of DMEM medium supplemented with 10% Of SFB. The plates were then incubated for 4 h at 37 ◦C and fluorescence was measured at 540 and 590 nm for excitation and emission, respectively [59]. Cell viability was expressed as a percentage of viable cells when compared to the control group without treatment.

#### *4.10. Statistical Analysis*

GraphPad Prism software (GraphPad Software, Inc., La Jolla, CA, USA) was employed for the statistical analysis with a confidence level of 95%. Parametric statistical analyses were conducted with one-way ANOVA followed by Tukey's multiple comparison test for total phenols and zeta potential. For the ion test the statistical analyses was Dunn's multiple comparison test.
