3.1 Total phenolic, flavonoid, and anthocyanin content of ROE
Red onion is a source of phenolic compounds, which provides high antioxidant activity due to the high bioactive compound concentration that, for the most part, is present in the skin (outer part). However, most of these compounds are also found in the bulb (internal part), among which we can mention flavonoids, which are phenolic compounds responsible for conferring high antioxidant activity. In addition to flavonoids, red onions are rich in anthocyanins, a group of pigments belonging to the family of phenolic compounds responsible for eliminating free radicals, providing more excellent antioxidant activity (González-de-Peredo et al., 2021).
The ROE showed a total phenolic content of 1922.53 mg GAE/100g on a wet basis, which is higher than the data reported by Metrani et al. (2022), who evaluated two red onion cultivars and found values ranging from 140 to 157 mg GAE/100g of fresh sample and Salamatullah et al. (2021), who obtained values ranging from 64.35 to 815.93 mg GAE/100g on a dry basis in red onions subjected to different drying temperatures.
As for the total flavonoid content, the extract showed a content of 1897.62 mg QE/100g on a wet basis, a result also higher than Salamatullah et al. (2021), who obtained values ranging from 12 to 729.56 mg CA/100g on a dry basis when subjecting red onions to different drying temperatures. For the total anthocyanin content, the ROE showed a content of 11.36 mg C3GE/100g on a dry basis, corroborating other studies that also evaluated the ROE, including Oancea and Drághici (2013), who found values ranging from 0.12 to 7.93 mg/100g for different red onion cultivars. Metrani et al. (2020) also evaluated two red onion cultivars (‘Honeysuckle’ and ‘Sweet Italian’) and found total anthocyanin contents of 10.3 and 8.6 mg/100 g, respectively. Anthocyanins are responsible for exerting various biological effects; extracts rich in anthocyanins can prevent or reduce the risk of developing multiple pathologies, including obesity, cardiovascular disease, and the onset of cancer because they are antioxidant agents (Colina-Coca et al., 2017).
Differences in the total phenolic, flavonoid, and anthocyanin content may be related to the method used for extraction and identification, edaphoclimatic conditions, soil type, and management (Metrani et al., 2020; Salamatullah et al., 2021).
3.2 Volatile composition of ROE headspace by GC-MS (HS-SPME-GC-MS)
Volatile compounds contribute to the overall sensory attributes of food products and generally determine their acceptability and quality. For example, studies have been carried out in onions on the relationship between volatile compounds and sensory characteristics of bulbs, showing the importance of volatile composition for sensory attributes (Cozzolino et al., 2021; Lekshmi et al., 2014). Nine VOCs were detected and identified in the ROE sample (Supplementary Material Table S1). The heat map (Fig. 1) illustrates the quantification of identified volatile organic compounds belonging to the classes of ketones, hydrocarbons, aromatics, alcohol, and acids, among others, which are determined by the color of the rectangle, where, in the color scale, the highest concentration was represented by the color red.
The major compound was 3(2H)-furanone, 2-hexyl-5-methyl (~ 0.04 mg/mL), which belongs to the furanone class. This compound has already been reported in onion skin extract, such as by Constantin et al. (2021). This compound is considered a product of the rearrangement of sulfur; it has also been previously reported in onion ‘shallot’ and is attributed as a fatty acid breakdown product derived from the onion tear factor (Tocmo et al., 2014). The second volatile compound was squalene (~ 0.03 mg/mL), a natural 30-carbon triterpene found in the hydrolipidic profile of some plants, as reported in the literature for a volatile profile of onion (Lekshmi et al., 2014). Squalene is synthesized in all cell types because it is a critical intermediate in forming eukaryotic sterols and bacterial hopanoids.
For this reason, various plants have become valuable sources for squalene isolation, as it is widely present in nature but in substantial amounts. The third major compound, o-cymene (~ 0.02 mg/mL), is an aromatic hydrocarbon that has already been reported in essential oils extracted from different matrices, such as from plants of the genus Allium (Pande et al., 2015). The compounds 2-octanone and 2-decanone were also identified (~ 0.01 mg/mL). In black garlic (Allium sativum L.), the compound 2-octanone has already been identified by Najman et al. (2022), who reported that this volatile could be detected due to aging. The authors attributed that this substance can contribute to the roasted, burnt, caramelized, green, and fruity aroma. There are reports of the presence of eugenol (~ 0.01 mg/mL), in plants of the genus Allium, as in the essential oil of garlic (Allium sativum L.) (Dziri et al., 2014).
The eugenol identified in the extract (~ 0.01 mg/mL) is a phenylpropanoid, a phenolic aromatic compound, weakly acidic and slightly soluble in water and organic solvents. The terpinen-4-ol (~ 0.01 mg/mL) is a monomeric compound in many essential oils. Hexadecanoic acid (~ 0.01 mg/mL), which is a saturated fatty acid widely reported to be present in onions as well as onion extracts and onion skin extracts (Acharya et al., 2019; Constantin et al., 2021). Finally, the last volatile organic compound identified was 2,4,4-trimethyl-1,3-pentanediol 1-isobutyrate (~ 0.01 mg/mL), an alcoholic ester.
The characteristic odors and flavors and the biological activities of the onion are related to its composition of VOCs. Disulfides are the most abundant class of VOCs, followed by trisulfides (Cecchi et al., 2020; Cozzolino et al., 2021; Lekshmi et al., 2014). Nonetheless, these two classes were not observed in the ROE volatile analysis in this study, suggesting that the extract production process may affect the presence of these compounds. The studies found in the literature show a variable composition of VOCs referring to onions, mainly red. This is directly attributed to the parts of the onion, its origin, cultivar, species, and the edaphoclimatic conditions under which they are cultivated.
3.3 Apparent viscosity and electrical conductivity of polymer solutions
The polymeric solutions used in producing fibers with different concentrations of ROE were evaluated for apparent viscosity and electrical conductivity (Table 1) since these parameters influence the electrospinning behavior and the quality of the fibers produced (García-Moreno et al., 2018). A gradual increase in electrical conductivity can be observed when the amount of ROE added to zein polymeric solutions increases from 10 to 40% (v/w), associated with increased ionizable groups in the solutions. The same behavior was observed by Prietto et al. (2018), who added anthocyanins extracted from red cabbage to zein polymeric solutions to obtain a pH-sensitive indicator produced by electrospinning.
An inverse behavior was observed for apparent viscosity compared to electrical conductivity since as the concentration of ROE in polymeric solutions increased, there was a decrease in apparent viscosity (Table 1). Nevertheless, studies show the opposite behavior when incorporating a jambolan extract (Santos et al., 2022) and red cabbage extract (Prietto et al., 2018), rich in anthocyanins, in zein polymeric solutions. This difference may be related to the use of a liquid extract in the present, while the aforementioned authors used a dry extract. Thus, adding ROE increased the presence of liquids in the zein polymeric solution, decreasing viscosity as it was added and increasing its concentration. Polymer solutions with low viscosity tend to form fibers with beads along their length, as seen in our study (especially in fibers with 40% ROSE).
3.4 Loading capacity
The LC is a crucial parameter to quantify the concentration effect of the compounds loaded on the polymeric matrix. Some factors can influence this process, such as the polymer concentration and the bioactive compounds in the polymeric solution, which significantly increase or reduce loading.
The ROE concentration added to the zein electrospun fibers affected the LC (p < 0.05), according to the increase in ROE concentration, there was a reduction in LC from 91.5 to 77.3% (Table 1). We observed no significant difference (p < 0.05) between fibers with 10 and 20% ROE, with a drop in LC for fibers with 30 and 40% ROE, which also did not show a significant difference between each other.
Regarding fibers with 10% ROE, there was a loss of approximately 14.7 to 15.5% for fibers with 30 and 40% ROE in the LC, respectively. This behavior of decreasing LC as ROE was added may be related to the saturation of the active sites available in the polymer used (zein) due to the progressive increase of ROE in polymeric solutions or even a lower interaction of the material wall thickness and ROE or even losses during the electrospinning process.
In order to be considered an effective process, minimum values of 70% loading of the compounds are suggested so that the process becomes industrially viable (Cruz et al., 2022). The LC values reported in this study showed that the polymer used as wall material is highly effective for loading bioactive compounds. Furthermore, encapsulation aims to protect the compounds against external environmental factors, such as humidity, light, and oxygen, so that there is no loss of their bioactivity and functionality (Cruz et al., 2021). This can promote desirable characteristics in active packaging or a food matrix. In addition, encapsulation prevents the vital sensory attributes of onion from being transferred to the food product when undesirable.
Santos et al. (2022) evaluated the loading capacity of ultrafine fibers produced using zein as a wall material to encapsulate jambolan extract at concentrations of 20, 30, and 40% (w/w). The authors obtained values ranging from 58.0 to 66.8% of loading capacity. Cruz et al. (2023) reported that the loading capacity was 67 to 78% for ultrafine fibers based on sweet potato starch incorporated with 3, 6, and 9% of red onion skin extract. These differences found for the loading capacity may be related to the plant material analyzed or even to the method and parameters used to encapsulate the compounds.
3.5 Morphology and diameter of zein fibers
In order to evaluate the effects of incorporating different concentrations of ROE (0, 10, 20, 30, and 40%, v/v) on zein electrospun fibers production, the morphological characterization and size distribution of the materials obtained were carried out (Fig. 2). The micrographs show that the conditions used in the processing guaranteed to obtain random and continuous fibers. Furthermore, the incorporation of ROE, especially at the highest concentration (40%, v/v), produced fibers with the presence of beads, which did not occur for the fibers without ROE incorporation.
Nevertheless, the increased incorporation of ROE in the zein electrospun fibers ensured a reduction in diameter and greater homogeneity of the materials since the variation in diameters also decreased as the concentration of ROE ranged from 0% (512 ± 132 nm) to 40% (224 ± 50 nm). Thus, it was possible to obtain a reduction of 56.25% in the average diameter of the fibers by incorporating ROE at the highest concentration. According to Neo et al. (2013), polymeric solutions with higher viscosity generate greater molecular entanglement and, thus, increase the fibers’ diameter. The same trend was observed in our study, as the greater the viscosity of the polymeric solutions, the greater the diameter obtained.
Furthermore, as the electrical conductivity of the polymeric solutions increased (with the increase in ROE concentration), the fibers diameter decreased. This behavior is justified by a lower stretching of the solution jet, resulting in higher viscosity and lower electrical conductivity of the fibers-forming solutions, which, when combined, result in fibers with larger diameters.
3.6 Fourier transform infrared spectroscopy
The FTIR analysis was used to characterize the chemical nature of the electrospun fibers and their precursor materials (Fig. 3). A broad band centered at 3433 cm− 1 was observed for the unencapsulated ROE, characteristic of N-H stretching of proteins and O-H bond stretching of carbohydrates and water, whereas the band 2921 cm− 1 is characteristic of C-H stretching of aliphatic groups (Chen et al., 2020; Santos et al., 2022). The band around 1635 cm− 1 is attributed to the C-C stretch of the phenyl ring that, in Allium plants, is present in high levels as polyphenolic constituents (Lu et al., 2011). Additionally, at 1105 cm− 1, a “shoulder” peak is observed, which is characteristic of the carbohydrates in onions (Lu et al., 2011). Furthermore, in the range of 1000–950 cm− 1, the bands are attributed to the elongation of the C = C and C-OH bonds located in the aromatic rings of anthocyanins are observed (Santos et al., 2022).
Regarding the spectra obtained for the zein polymer and fibers 0% ROE, no alterations were observed (Fig. 3); both showed broad bands around 3500–3000 cm− 1, which are attributed to the elongation of the O-H and N-H bonds of the hydroxyl and amide groups present in zein. In the 3000 to 2850 cm− 1 range, bands are observed and refer to the stretching of C-H bonds and aliphatic CHX groups. In the range of 1648–1640 cm− 1, bands are observed, referring to the stretching of C = O of the major amide (Li et al., 2020). Finally, bands around 1535 and 1449 cm− 1 are observed, which refer to the elongation of C = O and C-N bonds, respectively (Neo et al., 2013), suggesting the electrospinning process did not affect the structure of the protein. However, it interrupted intramolecular bonds such as hydrogen bonds. Minor changes were observed in the FTIR spectra for fibers containing different concentrations of encapsulated ROE (10, 20, 30, and 40%) compared with the spectrum obtained for zein. In these spectra, a slight narrowing of the band centered around 3500 cm− 1 is observed and attributed to the elongation of the O-H and N-H bonds compared to pure zein. According to Santos et al. (2022), this narrowing may be related to intermolecular interactions (hydrogen bonds) between the polymer (zein) and anthocyanins present in ROE. In addition, the fibers containing different concentrations of ROE showed characteristic bands of the anthocyanin aromatic rings (bands around 1000–950 cm− 1), which are more evident at higher ROE concentrations in the fibers due to hydrogen bond formation between zein and ROE, thereby confirming the interaction between the polymer and ROE and its effective encapsulation.
3.7 Thermogravimetric analysis
Thermogravimetric analysis of zein electrospun fibers loaded with different concentrations of ROE and its constituents (pure ROE and pure zein) allowed us to investigate the mass loss of each sample as a function of temperature and the influence of the electrospinning process on these losses (Fig. 4). The ROE (unencapsulated) showed two phenomena of mass loss based on TGA and DTG curves (Figs. 4a and 4b). Within a greater mass degradation, the first event occurred in the approximate range of ~ 33 to 167 ºC, with a peak temperature at ~ 140 ºC. With the less mass loss, the second event occurred at a degradation range of roughly ~ 180 to 225°C and a peak temperature of ~ 204°C. These mass loss phenomena are probably related to the evaporation of water, volatile compounds, and the degradation of molecules sensitive to this temperature range.
During thermal analysis, the pure zein and electrospun fibers with 0 to 40% ROE showed similar behavior. These samples presented the first mass loss at peak temperature at ~ 50 ºC, which probably refers to the loss of water and volatile compounds in the samples. A second thermal event was observed around ~ 170°C and the third event, responsible for the greatest mass loss of pure zein and fibers, which occurred in the range of ~ 195 to ~ 400 ºC (Fig. 4a), as demonstrated by the highest rates in the DTG diagram (Fig. 4b). These events possibly refer to breaks in the polymer chains due to the cleavage of bonds and biopolymer degradation (Altan et al., 2018; Antunes et al., 2017; Neo et al., 2013). The electrospinning process could not significantly change the polymer (zein) behavior in the rate of mass loss as a function of temperature, given the similarity observed between the curves of pure zein and fibers 0% ROE (i.e., electrospun fibers containing only zein). Moreover, the process of encapsulation of ROE in zein fibers could improve the thermal stability of the extract due to an increase in degradation temperatures of encapsulated ROE into zein fibers (degradation range of ~ 195 to ~ 400 ºC) compared to unencapsulated ROE (degradation range of ~ 33 to ~ 167 ºC) (Fig. 4).
In addition, the similarities between the TGA and DTG curves of zein fibers incorporated with 10 to 40% ROE compared to pure zein fibers (fibers 0% REO) indicate efficient extract loading into the wall material, corroborating the findings for LC presented in Section 3.4. Similar behavior was observed by Altan et al. (2018) and Evangelho et al. (2019), who incorporated active compounds such as carvacrol and folic acid into zein ultrafine fibers, respectively.
3.8 Contact angle with water
The hydrophilicity of zein membranes with ROE was measured by contact angle with water (Fig. 5). At the initial moment of 2 s, after the drop fell, the zein fibers membranes showed lower values for contact angle as ROE was incorporated. After 60 s, the membranes continued to decrease in the contact angle, which was smaller as the concentration of ROE in the fibers increased (similar behavior to the data for 2 s of analysis) except for the fibers with 20 and 30% ROE, which showed no significant difference (p > 0.05). Possibly, the highest mean contact angle values for zein fiber membranes without the addition of ROE occurred due to the presence of hydrophobic amino acid residues in the protein, which can reduce its interaction with water and, consequently, make the membrane more hydrophobic or less hydrophilic (Prietto et al., 2018).
As for the fibers membranes with ROE, which showed a reduction in contact angles as ROE was added, it is likely due to the incorporation of anthocyanin molecules, which are highly hydrophilic and water-soluble compounds, increasing the interaction of the membranes with the drop of water and making them more hydrophilic (Santos et al., 2022). All the fibers in this study presented contact angles below 90°, indicating hydrophilic surfaces according to the classification reported by Ahmad et al. 2015.
The more excellent permeability of the fibers with the addition of ROE can be helpful in different applications, including in active packaging, allowing the compounds to interact with the food product in a controlled and specific way, providing greater preservation due to the antioxidant properties of ROE. Similar behavior was also observed by Santos et al. (2022) when evaluating the contact angle in zein fibers added with anthocyanins from jambolan (at concentrations of 20, 30, and 40%, w/w). The authors reported that the higher the anthocyanin concentrations, the greater the permeability of the fibers and, consequently, the lower the contact angle.
3.9 Antioxidant activities
The antioxidant activities of the unencapsulated ROE (Table 1) were more significant than its encapsulated form for all radicals evaluated (DPPH, NO, and OH). The lower antioxidant activity in electrospun fibers with ROE is possibly due to the non-complete release of compounds retained in the zein fibers. As for the DPPH radical, there was an increase in antioxidant activity with increasing ROE into zein fibers. This behavior helps to confirm ROE encapsulation and the effectiveness of the encapsulation technique.
Similar behavior was observed for the NO radical. However, no significant difference was observed for zein electrospun fibers added with 30 and 40% ROE. As for the OH radical, the fibers added with 20, 30, and 40% of ROE did not show significant differences. Pure zein electrospun fibers (i.e., 0% ROE) did not present antioxidant activity against the evaluated radicals.
Despite not being produced by the human body, the DPPH radical is widely used to determine antioxidant activity. It is one of the few stable radicals available on the market, capable of accepting an electron or hydrogen radical to become a stable molecule. The reaction for determining the antioxidant capacity of compounds against DPPH occurs when its radical form is reduced by an antioxidant (Radünz et al., 2021), in this case, ROE bioactive compounds. Nevertheless, reactive oxygen species (ROS) are constantly produced in the human body through oxidative metabolic processes, not showing toxic effects at low concentrations; NO and OH radicals stand out among the ROS. The OH is the most abundant and destructive species, acting on lipid peroxidation and promoting mutations at the DNA level by modifying the purine and pyrimidine bases. However, NO can act in lipid peroxidation, promoting the release of pro-apoptotic factors and cell death. In addition, they are capable of causing food product deterioration and are also related to several diseases in living organisms (Charrier & Anastasio, 2011; Radünz et al., 2021).
Although zein electrospun fibers added from ROE have lower antioxidant activities than unencapsulated ROE; it is essential to mention that NO and OH radicals are important molecules for maintaining the proper functioning of living organisms, and their complete inhibition would be harmful. Thus, lower rates of capturing these radicals are desirable and possibly sufficient to reduce the concentration of excess oxidant species and, consequently, control oxidative stress, protecting cells from damage that could be caused by these radicals (Radünz et al., 2021). In addition, some studies have demonstrated the antioxidant activity of onions and onion extracts, such as (Prietto et al., 2018; Colina-Coca et al., 2017; Zhao et al., 2021).
Natural extracts are sources of antioxidants, including bioactive compounds, whose activity is widely explored in the pharmaceutical, cosmetic, and food industries and can replace the synthetic antioxidants commonly used in the food and packaging industry. In addition, research reports that the antioxidant activity of volatile compounds is directly related to the total content of phenolic compounds in this type of sample (volatile oils and organic extracts). Thus, in this study, our findings showed that zein electrospun fibers added with different encapsulated ROE concentrations showed potential as natural antioxidants due to sufficient free radical scavenging capabilities, which can control oxidative processes in some foods, for example, with a short shelf life.