Phenolic and Volatile Composition and Antioxidant Properties of the Leaf Extract of Brassica fruticulosa subsp. fruticulosa (Brassicaceae) Growing Wild in Sicily (Italy)

In continuation of research conducted on species of the spontaneous flora of Sicily (Italy) belonging to the Brassicaceae family, Brassica fruticulosa subsp. fruticulosa was selected. It is an edible species utilized in Sicilian traditional medicine. In this study, for the first time, the phenolic and the volatile compounds and the antioxidant properties of the hydroalcoholic extract obtained from the leaves of B. fruticulosa subsp. fruticulosa were characterized. Through HPLC-PDA/ESI-MS analysis, a total of 22 polyphenolic compounds (20 flavonoids and 2 phenolic acids) were identified, with 3-hydroxiferuloylsophoroside-7-O-glucoside (1.30 mg/g ± 0.01) and kaempferol-3-O-feruloylsophoroside-7-O-glucoside (1.28 mg/g ± 0.01) as the most abundant compounds. Through SPME-GC/MS several volatiles belonging to different chemical classes were characterized, with nitriles and aldehydes accounting for more than 54% of the whole volatile fraction. The extract of B. fruticulosa subsp. fruticulosa showed moderate activity in the DPPH assay (IC50 = 1.65 ± 0.08 mg/mL), weak reducing power (17.47 ± 0.65 ASE/mL), and good chelating properties (IC50 = 0.38 ± 0.02 mg/mL), reaching approximately 90% activity at the highest tested concentration. Lastly, the extract was non-toxic against Artemia salina, indicating its potential safety. According to the findings, it can be stated that B. fruticulosa subsp. fruticulosa represents a new valuable source of bioactive compounds.


Introduction
The Brassicaceae family (also called Cruciferae), order Brassicales, consisting of more than 300 genera and about 3500 species, includes a large number of vegetable crops recognized as rich sources of health-promoting phytochemicals [1,2]. ion (Fe 2+ )-chelating activity. Lastly, the toxicity of the extract was assessed by the brine shrimp (Artemia salina Leach) lethality bioassay. The Folin-Ciocâlteu assay is a recognized, widely used procedure for quantification of total phenolic compounds in plant extracts. It is a colorimetric method based on electron transfer reactions between the Folin-Ciocâlteu reagent and phenolics, giving rise to the formation of a blue chromophore with the maximum absorption at 765 nm. Generally, gallic acid is used as the reference standard compound, and the results are usually expressed as gallic acid equivalent [21]. In most cases, the antioxidant properties of plant extracts are explained by their total phenolic content with good correlation, confirming the value of this assay. Therefore, the determination of their total amount in the extract used for this study was performed.

Results and Discussion
The results of the Folin-Ciocâlteu assay showed that the total phenolic content of B. fruticulosa subsp. fruticulosa leaf extract was equal to 32.63 ± 1.11 mg gallic acid equivalent (GAE)/g extract. This content was close to that of the B. incana leaf extract previously investigated (37.20 ± 0.93 mg GAE/g extract) [5].

Identification of Phenolic Compounds by HPLC-PDA/ESI-MS
For the first time, the phenolic profile of the hydroalcoholic extract of the leaf of B. fruticulosa subsp. fruticulosa is reported. The HPLC-PDA chromatogram (λ = 330 nm) of the polyphenolic compounds occurring in the extract is shown in Figure 1. A total of 24 compounds were detected and, among them, according to retention times, as well as PDA, MS and MS/MS, and literature data, 22 were tentatively identified (Table 1) [5,[22][23][24][25][26][27][28]. Notably, most of them belonged to the flavonoid class, whereas only two were phenolic acids. Among flavonoids, 10 were kaempferol derivates, nine were quercetin derivates, and only two were isorhamnetin derivates. With regard to the two phenolic acids, they were sinapic and ferulic hydroxycinnamic acids in conjugation with a gentiobiose moiety.

Determination of Total Phenolic Content
The Folin-Ciocâlteu assay is a recognized, widely used procedure for quantification of total phenolic compounds in plant extracts. It is a colorimetric method based on electron transfer reactions between the Folin-Ciocâlteu reagent and phenolics, giving rise to the formation of a blue chromophore with the maximum absorption at 765 nm. Generally, gallic acid is used as the reference standard compound, and the results are usually expressed as gallic acid equivalent [21]. In most cases, the antioxidant properties of plant extracts are explained by their total phenolic content with good correlation, confirming the value of this assay. Therefore, the determination of their total amount in the extract used for this study was performed.
The results of the Folin-Ciocâlteu assay showed that the total phenolic content of B. fruticulosa subsp. fruticulosa leaf extract was equal to 32.63 ± 1.11 mg gallic acid equivalent (GAE)/g extract. This content was close to that of the B. incana leaf extract previously investigated (37.20 ± 0.93 mg GAE/g extract) [5].

Identification of Phenolic Compounds by HPLC-PDA/ESI-MS
For the first time, the phenolic profile of the hydroalcoholic extract of the leaf of B. fruticulosa subsp. fruticulosa is reported. The HPLC-PDA chromatogram (λ = 330 nm) of the polyphenolic compounds occurring in the extract is shown in Figure 1. A total of 24 compounds were detected and, among them, according to retention times, as well as PDA, MS and MS/MS, and literature data, 22 were tentatively identified (Table 1) [5,[22][23][24][25][26][27][28]. Notably, most of them belonged to the flavonoid class, whereas only two were phenolic acids. Among flavonoids, 10 were kaempferol derivates, nine were quercetin derivates, and only two were isorhamnetin derivates. With regard to the two phenolic acids, they were sinapic and ferulic hydroxycinnamic acids in conjugation with a gentiobiose moiety.   Table 1. As can be seen from Table 1, among the phenolic compounds identified, flavonols represented the most abundant constituents (11.1 mg/g extract), while phenolic acids were not quantified. Many of the compounds identified were previously reported to be constituents of Brassica juncea L. or B. incana [5,22].
In a previous study, we characterized the polyphenol compounds contained in the leaves of another Brassica wild species from Sicily, namely, B. incana, utilizing the same procedure of extraction reported here. By comparing the polyphenol profile of the leaf hydroalcoholic extract of B. fruticulosa subsp. fruticulosa with that of B. incana, a similar flavonoid pattern could be appreciated between the two species, with derivatives of the flavonols quercetin, kaempferol, and isorhamnetin, together with the hydroxycinnamic acids sinapic acid and ferulic acid. Nonetheless, some differences among the two species were highlighted; indeed, a greater number of flavonoid derivatives were detected in the B. fruticulosa subsp fruticulosa leaf extract, whereas the hydroxycinnamic acids identified in the B. incana extract were found to be more numerous and in conjugation with malic acid and glucose moieties, in addition to gentiobiose [5].
The identified compounds are well known secondary metabolites of plants [27] and, in particular, nitriles are common in Brassica species. Indeed, nitriles, as well as isothiocyanates, originate from the hydrolysis of glucosinolates, a group of compounds typical of Brassicaceae, Capparaceae, and Caricaceae families. The enzyme myrosinase, released upon tissue damage, hydrolyzes the β-D-S-glycosidic bonds of glucosinolates, releasing the sulfur-containing aglycone moieties that are unstable and undergo the Lossen rearrangement to form various breakdown products such as isothiocyanates, thiocyanates, nitriles, epithionitriles, and oxazolidine-2-thiones. The products of glucosinolate hydrolysis depend on various factors, such as the glucosinolate substrate, the reaction conditions, the presence of substances which can modify the action of the enzyme, and the plant pretreatments. It has been demonstrated that, if the hydrolytic reaction occurs under acidic conditions, low temperature, and low water levels, nitrile formation is favored. Nitriles are also favored by autolysis, rather than by the action of an exogenous source of the enzyme and by the presence of ferrous ion. Moreover, in fresh or freeze-thawed leaves, the glucosinolate hydrolysis produced mainly nitriles, whereas dry heating of the leaves decreased the proportion of nitrile formation and increased the proportion of isothiocyanate formation [29].
Although few data are present in the literature on the glucosinolate composition of B. fruticolosa [30], with none referring to leaves, the nitriles here identified are consistent with the structure of glucosinolates previously reported in Brassicaceae [31].
Among the minor constituents, safranal, β-cyclocitral, and β-ionone originate from the enzymatic breakdown of carotenoids. These terpenoids have been detected in the flowering top extract of B. incana [32] and in the hydroalcoholic extract of the aerial part of different Matthiola species, such as M. fruticolosa [28] and M. tricuspidata (our unpublished data). Furthermore, hexahydrofarnesyl acetone or phytone was detected in our previous studies on hydroalcoholic extract of B. incana and Matthiola spp. [28,32,33]; this ketone very common in plants arises from the oxidative degradation of (E)-phytol, a diterpene alcohol that occurs as a side-chain of chlorophyll a [34].
The results here reported are quite different from those described in our previous study on the volatiles of B. fruticolosa leaves [8]. This can be explained considering that we previously applied the SPME technique directly to the plant leaves, whereas, in this case, a hydroalcoholic extract of the leaves was considered.
Considering the volatile profile of the leaf hydroalcoholic extract of another Brassica species, namely, B. incana, only a few qualitative similarities emerged, whereas, from a quantitative point of view, in the volatile profile of B. incana leaf extract, isothiocyanates prevailed vs. nitriles [32].

Antioxidant Activity
Oxidative stress has been identified as the root cause of the development and progression of many diseases. In recent years, several epidemiological studies have indicated that a high intake of plant products is associated with a reduced risk of a number of chronic diseases, such as atherosclerosis and cancer, partly attributed to the compounds which possess antioxidant activity [35]. Bibliographic data show that many species belonging to the genus Brassica contain phenolic compounds, widely considered to be the most important specialized metabolites with antioxidant activity [24].
Antioxidant activity should be evaluated by the use of various methods in order to acquire a more complete antioxidant profile. In these assays, plant extracts are generally assessed for their function as reducing agents, hydrogen donors, singlet oxygen quenchers, or metal chelators [36]. When they react with free radicals by producing less reactive species or by interrupting the radical chain reaction, they are classified as primary (chain breaking) antioxidants; on the contrary, when they act by suppressing the formation of radicals and protecting against oxidative damage, they are defined as secondary (preventive) antioxidants [37]. Thus, three in vitro assays based on different approaches and mechanisms were used in order to determine the antioxidant capacity of B. fruticulosa subsp. fruticulosa extract. The primary antioxidant properties were examined using the DPPH assay, based on the hydrogen atom transfer (HAT) and electron transfer (ET) mechanisms, and the reducing power, an ET-based assay. The secondary antioxidant ability was determined by measuring the ferrous ion (Fe 2+ )-chelating activity.
The results of the DPPH test, utilized to establish the free-radical-scavenging properties of the extract, are shown in Figure 2A. Compared with the reference standard BHT, the hydroalcoholic leaf extract of B. fruticulosa subsp. fruticulosa exhibited moderate scavenging activity, dose-dependently, in the range of concentrations assayed (0.0625-2 mg/mL), reaching about 60% inhibition of the DPPH radical at the highest concentration tested. The IC 50 values confirmed the lower activity of the extract with respect to the standard BHT (1.65 ± 0.08 mg/mL and 0.07 ± 0.01 mg/mL, respectively). From the comparison of the scavenging activity of the extract with that highlighted for B. incana leaf extract (IC50 = 1.31 ± 0.05 mg/mL), previously investigated under the same experimental conditions, it is evident that B. fruticulosa subsp. fruticulosa extract had a slightly lower activity [5].
The reducing power reflects the ability to stop the radical chain reaction. In this assay, the presence of antioxidant compounds in the sample determines the reduction of Fe 3+ to the ferrous form (Fe 2+ ); this reduction is highlighted by spectrophotometric measurement (700 nm) of the change of yellow color of the test solution to various shades of green and From the comparison of the scavenging activity of the extract with that highlighted for B. incana leaf extract (IC 50 = 1.31 ± 0.05 mg/mL), previously investigated under the same experimental conditions, it is evident that B. fruticulosa subsp. fruticulosa extract had a slightly lower activity [5].
The reducing power reflects the ability to stop the radical chain reaction. In this assay, the presence of antioxidant compounds in the sample determines the reduction of Fe 3+ to the ferrous form (Fe 2+ ); this reduction is highlighted by spectrophotometric measurement (700 nm) of the change of yellow color of the test solution to various shades of green and blue, depending on the reducing power of the antioxidant sample [38]. Figure 2B shows the results of the reducing power of the hydroalcoholic leaf extract of B. fruticulosa subsp. fruticulosa; the extract exhibited mild, concentration-dependent, reducing power, as compared with the standard BHT. This was confirmed also by the ASE/mL values (17.47 ± 0.65 and 0.89 ± 0.06 ASE/mL, respectively). This result agrees with that previously reported for the extract of B. incana leaves [5].
The method of Fe 2+ -chelating activity utilized the reagent ferrozine, which can quantitatively form complexes with Fe 2+ ; in the presence of chelating agents, the complex formation is inhibited, with the result that the red color of the complex is decreased. Measurement of color reduction, therefore, allows the estimation of the chelating activity of the coexisting chelator [38].
The results of the in vitro antioxidant tests showed that B. fruticulosa subsp. fruticulosa extract acts as moderate primary antioxidant and possesses strong secondary antioxidant properties.
Taking into consideration that flavonoids are known to display metal-chelating effects [39], the good chelating activity of the extract may depend to some extent on the presence of flavonol derivatives, mostly of quercetin and kaempferol, detected by HPLC-PDA/ESI-MS analysis; however, the involvement of other polar constituents present in the phytocomplex cannot be excluded.

Artemia salina Leach Lethality Bioassay
The toxicity of the extract was assessed by the Artemia salina Leach bioassay. The brine shrimp lethality bioassay is extensively utilized as an alternative model for toxicity evaluation because it offers numerous advantages such as rapidity, cost-effectiveness, continuous availability of cysts (eggs), and ease of handling and maintenance under laboratory conditions [40]. It represents a simple technique for predicting the toxicity of plant extracts in order to consider their safety. The results of the bioassay showed the absence of toxicity against brine shrimp larvae for the extract of B. fruticulosa subsp. fruticulosa. Indeed, the median lethal concentration values were found to be above 1000 µg/mL, thus indicating their potential safety according to Clarkson's toxicity criterion [41]. These data are in agreement with those observed for the extracts of B. incana investigated in our previous work [5].

Plant Material and Extraction Procedure
The leaves of the Brassica fruticulosa subsp. fruticulosa were collected in the locality of Massa San Giorgio, on the Peloritani Mountains (Messina, Sicily, Italy), in October 2019. The taxonomic identification was confirmed by Prof. S. Ragusa, Department of Health Sciences, University Magna Graecia of Catanzaro (Catanzaro, Italy). A voucher specimen (1016/19) was deposited in the same Department.
After harvesting, the plant material was washed, blended, frozen, and then lyophilized. Subsequently, the leaves, finely ground, were subjected to a preventive maceration at 25 • C with 70% MeOH (1:10 w/v) for 1 h. The extraction was performed with 70% MeOH (1:10 w/v) in an ultrasonic bath at 50 • C for 15 min, repeated three times; then, the filtrates were combined and evaporated to dryness by a rotavapor. The yield of the leaf extract, referring to 100 g of lyophilized plant material, was 22.99%.

Determination of Total Phenolic Content
The total phenolic content of B. fruticulosa subsp. fruticulosa leaf extract was determined by the Folin-Ciocâlteu colorimetric method, using gallic acid as a standard phenolic compound [42]. An aliquot of 0.1 mL of each sample solution was mixed with 0.2 mL Folin-Ciocâlteu reagent, 2 mL of distilled water, and 1 mL of 15% Na 2 CO 3 . A linear calibration curve of gallic acid, in the range 125-500 µg/mL, was constructed. The absorbance was measured at 765 nm, after a 2 h incubation at room temperature, with a UV-1601 spectrophotometer (Shimadzu, Milan, Italy). The total phenolics were expressed as mg GAE/g of extract (dw) ± standard deviation (SD). The data were obtained from the average of three independent determinations.

Identification of Phenolic Compounds by HPLC-PDA/ESI-MS
The analyses were carried out using a Shimadzu HPLC system (Milan, Italy) equipped with a CBM-20A controller, LC-20AD pumps, a DGU-20A3 degasser, a SIL-20AC autosampler, an SPD-M20A photo diode array detector (PDA), and a triple-quadrupole mass analyzer (LCMS-8050, Shimadzu, Kyoto, Japan), equipped with an ESI interface, in positive and negative ionization mode. Data acquisition was performed by Shimadzu LabSolution software ver. 5.91.

PDA Conditions
The wavelength range was 200-400 nm, and the chromatograms were extracted at 280 nm. The time constant was 0.08 s, and the sample frequency was 40 Hz.

MS Conditions
The MS acquisition was performed using the ESI interface in negative ionization mode. Mass detection was performed in full scan mode in the spectral range 100-1400 m/z, with an interval of 0.5 s. Nitrogen (N 2 ) was used as a nebulizing gas at a flow rate of 3 L/min. The following settings were applied to the instrument: interface temperature, 300 • C; heat block, 400 • C; DL temperature, 250 • C; DL voltage, −34 V; probe voltage, 4.5 kV; Q-array voltage, 1.0 V; RF voltage, 90 V; detection gain, 1.0 kV.
The dried extract was solubilized in saturated sodium chloride solution to a final concentration of 10 mg/mL; then, 3 ± 0.1 mL of each extract solution was transferred to a 7 mL vial closed with a 'mininert' valve (Supelco, Bellefonte, PA, USA). For the volatile extraction, the sample was equilibrated for 15 min at 40 • C, and a DVB/CAR/PDMS fiber, 50/30 µm film thickness (Supelco, Bellefonte, PA, USA), was exposed for 15 min to the headspace of the sample maintained at 40 • C under continuous magnetic stirring. Finally, the SPME fiber was placed for 3 min into the injector port of the GC/MS, held at 260 • C, for the thermal desorption of the analytes onto the capillary GC column. The conditions were as follows: injection mode, splitless; oven temperature (1) 45 • C held for 5 min, then increased to 80 • C at a rate of 10 • C/min and to 240 • C at 2 • C/min, held at 240 • C for 5 min for polar column, (2) 45 • C increased to 160 • C at a rate of 3 • C/min and to 260 • C at 10 • C/min, held at 260 • C for 5 min for apolar column; carrier gas, helium at a constant flow of 1 mL/min; transfer line temperature, 250 • C; acquisition range, 40 to 360 m/z; scan speed, 1250. For the identification of the volatiles, mass spectral data, NIST' 14 (NIST/EPA/NIH Mass Spectra Library, version 2.0, Gaithersburg, MD, USA) and FFNSC 3.0 database, and linear retention indices (LRI) were used.

Free-Radical-Scavenging Activity
The free-radical-scavenging activity of the hydroalcoholic leaf extract of B. fruticulosa subsp. fruticulosa was evaluated using the DPPH (2,2-diphenyl-1-picrylhydrazyl) test [44]. DPPH is a stable radical in methanol with violet color because of delocalization of the spare electron throughout the molecule. When a proton is accepted in the reaction with the oxygen atom of a radical scavenger's OH group, the reduced DPPH-H (2,2-diphenyl-1-picrylhydrazine) is formed, which is yellow. The degree of discoloration indicates the amount of DPPH scavenged; a greater bleaching action indicates higher antioxidant activity, as reflected in a lower IC 50 value.
The extract was tested at different concentrations (0.0625-2 mg/mL) using butylated hydroxytoluene (BHT) as a reference compound. A volume of 0.5 mL of each sample solution was mixed with 3 mL of daily prepared methanol DPPH solution (0.1 M) and incubated for 20 min at room temperature in the dark. Then absorbance was measured at 517 nm using a model UV-1601 spectrophotometer (Shimadzu, Milan, Italy). The scavenging activity was measured as the decrease in absorbance of the samples versus DPPH standard solution.
Results were obtained from the average of three independent experiments, and they were expressed as the mean radical-scavenging activity percentage (%) ± SD and mean 50% inhibitory concentration (IC 50 ) ± SD.

Reducing Power Assay
The reducing power of the hydroalcoholic leaf extract of B. fruticulosa subsp. fruticulosa was determined according to the Fe 3+ -Fe 2+ transformation method [45]. The extract was tested in the range of 0.0625-2 mg/mL. A volume of 1 mL of each sample was mixed with 2.5 mL of phosphate buffer (0.2 M, pH 6.6) and 2.5 mL of 1% potassium ferricyanide (K 3 Fe(CN) 6 ). Following incubation at 50 • C for 20 min and rapid cooling, 2.5 mL of 10% trichloroacetic acid was added, and the mixture was centrifuged (3000 rpm, 10 min). Finally, 2.5 mL of the supernatant was mixed with 2.5 mL of distilled water and 0.5 mL of 0.1% ferric chloride (FeCl 3 ). After incubation for 10 min of at room temperature in the dark, the color change of the sample was estimated by measuring absorbance at 700 nm. The increased absorbance of the reaction mixture indicates an increase in reducing power. Ascorbic acid and butylated hydroxytoluene (BHT) were used as reference compounds. The results were obtained from the average of three independent experiments, and they were expressed as the mean absorbance values (700 nm) ± SD and ascorbic acid equivalent/mL of extract (ASE/mL) ± SD.

Ferrous Ion (Fe 2+ )-Chelating Activity Assay
The Fe 2+ -chelating activity of the hydroalcoholic leaf extract of B. fruticulosa subsp. fruticulosa was estimated by measuring the formation of the Fe 2+ -ferrozine complex [46]. The extract was tested in the range of 0.0625-2 mg/mL, and ethylenediaminetetraacetic acid (EDTA) was used as positive control. A volume of 1 mL of each sample was mixed with 0.5 mL of MeOH and 50 µL of 2 mM FeCl 2 . Then, 0.1 mL of 5 mM ferrozine was added to initiate the reaction; the mixture was shaken vigorously and incubated at room temperature in the dark for 10 min. The control contained FeCl 2 and ferrozine, which are complex formation molecules. The color change of the solutions was estimated by measuring absorbance spectrophotometrically at 562 nm. The results were obtained from the average of three independent experiments, and they were expressed as the mean inhibition of the ferrozine-Fe 2+ complex formation (%) ± SD and IC 50 ± SD.

Artemia Salina Leach Lethality Bioassay
The Artemia salina Leach (brine shrimp) lethality bioassay was employed to predict the toxicity of the leaf hydroalcoholic extract of B. fruticulosa subsp. fruticulosa [47]. Brine shrimp eggs were hatched in artificial seawater (33 g sea salt/L deionized water) by incubation under a 60 W lamp, providing direct light and warmth (24-26 • C). After hatching, 10 brine shrimp larvae were incubated at 25-28 • C in 5 mL of artificial seawater mixed with different amounts of the extract (10, 100, 500, and 1000 µg/mL). After 24 h, the numbers of surviving nauplii were counted using a magnifying glass. The experiments were conducted in triplicate for each concentration, and the median lethal concentration (LC 50 ) values were determined by Litchfield and Wilcoxon's method. The toxicity level of the extract was assessed according to the toxicity scale reported by Clarkson et al. [41]. Extractis considered non-toxic if the LC 50 is higher than 1000 µg/mL.

Conclusions
This work described the results of the phytochemical characterization and the antioxidant properties of the leaf hydroalcoholic extract of B. fruticulosa subsp. fruticulosa growing wild in Sicily (Italy), never investigated before. An in-depth overview of the qualitative-quantitative composition of the phenolic and volatile constituents of the leaves was attained. On the basis of the in vitro antioxidant assays performed, it can be stated that the B. fruticulosa subsp. fruticulosa leaf extract had much higher secondary than primary antioxidant properties. Furthermore, the extract was found to be non-toxic against brine shrimp larvae, indicative of its potential safety.
The obtained results provide a substantial contribution to the knowledge of B. fruticulosa subsp. fruticulosa so far little studied, indicating this wild edible species as a new valuable source of antioxidant compounds with potential health-promoting effects.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.