Flower and Leaf Extracts of Sambucus nigra L.: Application of Membrane Processes to Obtain Fractions with Antioxidant and Antityrosinase Properties

This study aimed at evaluating and comparing the chemical profile as well as the antityrosinase and antioxidant activities of ethanol (EtOH) and methanol (MeOH) extracts of Sambucus nigra L. (Adoxaceae) flowers and leaves in order to discover new candidates for food additives and cosmetic and pharmaceutical products. For this purpose, a novel lower-melting-point ethylene-chlorotrifluoroethylene (LMP ECTFE) nanofiltration (NF) membrane was employed in order to produce the concentrated fractions of S. nigra. Floral extracts were richer in phytochemicals in comparison to the leaf extracts. The High-performance liquid chromatography (HPLC) profile revealed rutin, quercetin, protocateuchic acid, 3,5-dicaffeoylquinic acid, and neochlorogenic acid as the most abundant compounds. Ferric reducing antioxidant power (FRAP), 2,2’-diphenil-1-picrylhydrazil (DPPH) radical scavenging, and 2,2’-azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) tests were used to investigate the antioxidant properties. NF retentate fractions of floral ethanol extracts exerted the highest tyrosinase inhibitory activity with an IC50 of 53.9 µg/mL and the highest ABTS radical scavenging activity (IC50 of 46.4 µg/mL). In conclusion, the present investigation revealed the potential benefits of NF application in S. nigra extracts processing, suggesting the use of retentate fractions as a promising source for antioxidant and tyrosinase inhibitory compounds which could pave the way for future applications.

Specialty Polymers (Bollate, Mi) is a lower-melting-point grade of standard Halar ® ECTFE, but is easily workable [22]. As reported in literature, ECTFE membranes were successfully applied in the pervaporation (PV )process for aqueous solutions of toluene (200-250 ppm), with an enrichment factor (β) of 894 and 36 g m -2 h -1 of toluene flux [23]. Falbo et al. reported the use of ECTFE PV membranes for binary azeotropic mixture of ethanol and cyclohexane (30.5% w/w and 69.5% w/w) obtaining flux and selectivity of 1.7 kg m 2 h and 15, respectively [24]. ECTFE flat membranes were also employed in a membrane condenser by Drioli et al. [25], with a water recovery between 35% and 55%. Moreover, ECTFE porous membranes fabricated using a ternary system showed excellent fouling resistance during the vacuum membrane distillation process (VMD), with a salt rejection (3.5 wt% NaCl) exceeding 99.99% [26]. Thanks to the properties of LMP ECTFE polymer, such as excellent chemical and mechanical resistance, these novel membranes result in a higher stability toward organic solvents and they are suitable for extraction processes based on the use of alcohols [22,27].

Nanofiltration Set-Up
LMP ECTFE NF membranes used in this work, named N2, were prepared and characterized as reported in detail by Ursino et al. [22]. Membrane properties are summarized in Table 1. Filtration tests were performed by using a high-pressure crossflow filtration cell (model HP4750) supplied by Sterlitech Corporation. The cell presented a volume of 300 mL and a diameter of 5.1 cm, with an effective membrane area of 20.4 cm 2 . The membrane was preventively conditioned by immersion for 24 h in the target pure solvent (EtOH or MeOH) before the experiments. After conditioning, the membrane was placed in the cell. The permeate was collected by applying an N 2 gas pressure (transmembrane pressure) at 12 bar and room temperature. Each test was conducted three times. The permeate flux (J) was calculated by the following equation: where V (L) is the volume of permeate, A (m 2 ) is the membrane area, and ∆t (h) is the operation time. The average and relative standard deviations were calculated. A schematic of the extraction and filtration procedure of S. nigra samples is illustrated in Figure 1.
The identification and quantification of the main phytochemicals that characterize S. nigra flower and leaf extracts were carried out from the retention times in comparison with authentic standards (astragalin, caffeic acid, chlorogenic acid, 3,5-dicaffeoylquinic acid, ferulic acid, isoquercetin, kaempferol, myricetin, neochlorogenic acid, p-coumaric acid, protocateuchic acid, quercetin, rosmarinic acid, and rutin). Data processing was carried out using Clarity Software (Chromatography Station for Windows). All analyses were performed in triplicate and results were expressed as mg/100 g of extract.

Ferric Reducing Antioxidant Power (FRAP) Test
This assay is based on the redox reaction that involves TPTZ (2,4,6-tripyridyl-s-triazine)-Fe 3+ complex. The FRAP reagent was freshly prepared by mixing 25 mL of acetate buffer 0.3 M (pH 3.6), 2.5 mL of TPTZ solution 10 mM in HCl 40 mM, and 2.5 mL of FeCl 3 20 mM. A concentration of 2.5 mg/mL of S. nigra extracts was prepared and 0.2 mL of this solution was mixed with 1.8 mL of FRAP reagent. The reduction of TPTZ-Fe 3+ complex to TPTZ-Fe 2+ complex was monitored at 595 nm. This change is directly related to the reducing power of electron-donating antioxidants present in the reaction mixture [28]. BHT was used as positive control.

DPPH Assay
The DPPH radical scavenging activity of S. nigra extracts was assessed by using DPPH assay as previously reported [28]. Reacting with an antioxidant, DPPH radicals were reduced and generated a change in color, read at 517 nm (Perkin Elmer Lambda 40 UV/VIS spectrophotometer, Milan, Italy).
The DPPH radical scavenging activity was calculated as a percentage of DPPH discoloration using the following equation: where A 0 is the absorbance of the DPPH solution and A is the absorbance of the sample. Ascorbic acid was used as positive control.

ABTS Assay
In this assay, the ABTS solution was prepared by the reaction of ABTS 7 mM and potassium persulphate 2.45 mM, and stored (for 12 h) at room temperature in the dark. The solution was diluted with ethanol to reach an absorbance of 0.70 at 734 nm. Then, 25 mL of different concentrations of S. nigra extracts were added to 2 mL of diluted ABTS solution and the absorbance was measured at 734 nm after 6 min. The ABTS radical scavenging ability of S. nigra extracts was calculated according to the following equation: where A is the absorbance of the control reaction and A is the absorbance in the presence of the sample [28]. Ascorbic acid was employed as positive control.

Relative Antioxidant Capacity Index (RACI)
RACI is a statistical index that provides a rank of antioxidant ability generated from different in vitro tests applied to the investigated samples [29]. Herein, data obtained from FRAP, ABTS, and DPPH tests were used to calculate a RACI value for S. nigra extracts.

Tyrosinase Inhibitory Activity Test
The mushroom tyrosinase inhibition assay was performed as previously described [30]. Briefly, 40 µL of mushroom tyrosinase solution (100 units/mL), 40 µL of L-tyrosine solution (0.1 mg/mL), 40 µL of S. nigra extract in hydroalcoholic solution, and 80 µL of phosphate-buffered saline (PBS) solution (25 mM, pH 6.8) were added to a 96-well microplate and incubated for 30 min at 37 • C. The amount of the produced dopachrome was measured at 492 nm before and after incubation. Kojic acid was used as positive control. The percentage of tyrosinase inhibition was calculated as follows: where A and A are the absorbance values of the blank solution after and before incubation, respectively; B and B are the absorbance values of the extract solution after and before incubation, respectively.

Statistical Analysis
The half-maximal inhibitory concentration (IC 50 ) was calculated by nonlinear regression with the use of GraphPad Prism version 6 for MS Windows (GraphPad Software, San Diego, CA, USA). The concentration-response curve was obtained by plotting the percentage inhibition vs concentration. The results, expressed as the mean values and standard deviations (SD), were analyzed by using the one-way ANOVA test and multicomparison Dunnett's test.

Evaluation of Membrane Productivity
LMP ECTFE N2 NF membrane was selected based on its properties and performance, as reported in Ursino et al. [22]. Indeed, LMP ECTFE N2 membrane presented a good mechanical performance and excellent chemical stability, with a low degree of swelling in different solvents, in particular, 4% and 6% for methanol and ethanol, respectively. The ethanol and methanol permeability at room temperature were of 3.0 and 3.6 L/m 2 h bar, respectively. These differences can be attributed to the molecular weight and the viscosity of both solvents [13,22]. The time evolution of permeate flux in the selected operating conditions for both leaf and flower extracts is reported in Figure 2.  (Figure 2b). Therefore, the membrane productivity with floral extracts was lower than that observed for leaf extracts. This trend can be attributed to the lower solubility of the leaf extracts; indeed, the total concentration of leaf and floral extracts was 0.006 mg/mL and 0.011 mg/mL, respectively. In addition, both floral extracts presented a higher quantity of particulate material resulting in higher turbidity in comparison with the leaf extracts.

Chemical Profile
S. nigra flowers and leaves were extracted by maceration using methanol and ethanol as solvents. The content of selected phenolic acids and flavonoids analyzed in crude extracts and retentate fractions of S. nigra flowers and leaves by HPLC-DAD is reported in Tables 2 and 3, respectively.  Overall, obtained data showed that target compounds exhibited higher content in methanol extracts compared to ethanol extracts. Many techniques, including maceration, subcritical water extraction, Soxhlet extraction, and ultrasound-assisted extraction, are used to recover antioxidants from plants. However, the extraction yield and the antioxidant effects depend not only on the employed extraction method but also on the solvent used for the extraction process. To recover polyphenols from plants, polar solvents are often used. Among these, ethanol has been known as a good solvent in addition to being safe for human consumption. Generally, methanol has been found to be more efficient in extraction of polyphenols with lower molecular weight [31,32]. Moreover, in agreement with previous studies, the flavonoid amounts are greater in flowers than in leaves [33][34][35][36].
Floral extracts were mainly characterized by the presence of rutin, protocateuchic acid, chlorogenic acid, neochlorogenic acid, kaempferol, and quercetin. As expected, with the exception of chlorogenic acid and neochlorogenic acid, retentate fractions of both methanol and ethanol extracts exhibited a higher content of target compounds in comparison with crude extracts. In addition, in the retentate fractions, caffeic acid and p-coumaric acid were not detected. These differences with crude extracts can be attributed to the permeation of such compounds through the NF membrane. According to the data in Table 2, concentration of floral extracts by NF increased the content of most bioactive compounds of both alcoholic extracts, including 3,5-dicaffeoylquinic acid, isoquercetin, kaempferol, quercetin, and rosmarinic acid by 120-180%.
Quercetin, neochlorogenic acid, isoquercetrin, chlorogenic acid, and kaempferol are the most abundant compounds in leaf extracts. However, the content of these constituents is lower than that found in the floral extracts. In addition, ferulic acid, protocateuchic acid, and rosmarinic acid were not identified. As for floral extracts, caffeic acid and p-coumaric acid were not detected in the retentate fractions of both methanol and ethanol extracts. Moreover, some target compounds, including chlorogenic acid, 3,5-dicaffeoylquinic acid, and isoquercetin, were more expressed in the crude extracts due to their permeation through the NF membrane (Table 3). On the other hand, the content of quercetin in both alcoholic extracts increased by 225%.

Antioxidant Activity
The antioxidant properties of S. nigra leaf and floral extracts and related NF retentate fractions were estimated by using two radical scavenging assays (ABTS and DPPH tests) and ferric reducing power (FRAP) test. Data are reported in Table 4. All samples showed a concentration-dependent activity. Data are given as media ± S.D. (n = 3). Differences within and between groups were evaluated by one-way analysis of variance test *** P < 0.0001 followed by a multicomparison Dunnett's test: ** P < 0.01 compared with the positive controls. a Samples tested at the concentration of 2.5 mg/mL.
As expected, both retentate fractions exhibited higher DPPH and ABTS radical scavenging activity than crude extracts. In particular, both leaf and flower retentate fractions of methanol extracts showed the highest activity with IC 50 values of 39.2 and 48.3 µg/mL, respectively, in DPPH assay. A similar trend was observed in the ABTS test where flower retentate fraction of methanol extract exhibited the highest potency with an IC 50 value of 46.4 µg/mL. Many chemical reactions including radicals involved iron for its ability to transfer single electrons, starting even with relatively nonreactive radicals [37]. Thus, the reduction of ferric ion is an important approach to investigate the antioxidant potential of samples. In this study, all investigated samples showed higher FRAP values than that reported for BHT used as positive control. In particular, the retentate fraction of leaf ethanol extract exhibited a FRAP value of 110.9 µM Fe(II)/g, which is 1.75 times higher than BHT. A comparable FRAP value was observed also in the retentate fraction of floral methanol extract. Pearson's correlation coefficient calculated between the sum of quantified compounds and the result of the FRAP assay evidenced an r value of 0.64. The integrated approach RACI was used to estimate the sample with the highest antioxidant potential. As reported in Table 4, both crude extract and retentate fraction of leaf methanol extracts showed promising RACI values of −0.52 and −0.51, respectively. The retentate of the floral methanol extract showed the highest RACI value (−0.01) for this set of samples.
Recently, Viapiana and Wesolowski [38] reported that infusion prepared from S. nigra flowers had higher mean DPPH and FRAP activities than the teas prepared from berries. The potential ABTS radical scavenging activity confirmed results obtained by Młynarczyk and Walkowiak-Tomczak [39] that demonstrated the methanol extract of fresh flowers was more potent than dried flower extract obtained in the same condition in scavenging ABTS radicals. Our data are in agreement also with those reported by Stoilova et al. [40] for commercial elder flower extract in which an IC 50 value of 0.152 µg/mL was found in the DPPH test. Moreover, S. nigra floral extracts exert a greater radical scavenging activity when compared to rutin since it caused 97.70% of DPPH inhibition at 10 µg/mL concentration, while 40 µg/mL of rutin are necessary to ensure DPPH radical inhibition of 77.47%.

Tyrosinase Inhibitory Activity
The search for natural and safe enzyme inhibitors to overcome skin problems is amongst the most investigated matters. These compounds are able to reduce the production of melanin. Melanin is important in protecting skin. However, the accumulation of anomalous quantities of melanin, induced by several factors including UV exposure and drugs, produces more pigmented patches that might become aesthetic problems. Tyrosinase is a key enzyme in the melanogenesis process. In fact, melanogenesis starts with the oxidation of L-tyrosine to dopaquinone by tyrosinase. Despite numerous studies, the actually used inhibitors of tyrosinase still show insufficient activity, low stability, or toxicity. Hence, the search for new tyrosinase inhibitors is very active [5]. Herein, S. nigra extracts were investigated for their potential inhibitory activity against mushroom tyrosinase. Data are reported in Table 5. Data are given as media ± S.D. (n = 3). Differences within and between groups were evaluated by one-way analysis of variance test *** P < 0.0001 followed by a multicomparison Dunnett's test: ** P < 0.01 compared with the positive control.
Generally, flowers are more active than leaves. The most interesting results are obtained by using methanol as a solvent of flower extraction with IC 50 values of 53.9 and 62.5 µg/mL for retentate fraction and crude extract, respectively. Both extracts are characterized by the presence of rutin (quercetin-3-O-rutinoside), quercetin, protocateuchic acid, 3,5-dicaffeoylquinic acid, and kaempferol as the main constituents. Tyrosinase was competitively inactivated by rutin with an IC 50 value of 6.8 mM and a Ki of 1.10 mM [41]. Rutin competed with L-DOPA in the active site pocket of the enzyme due to the property of copper chelating. A strong copper chelating effect of this flavonoid may be related to the presence of hydroxyl groups. Fan et al. [42] demonstrated that quercetin is able to inhibit both the mono-and di-phenolase activity of tyrosinase, and to inhibit the dopaquinone formation in a reversible competitive manner (IC 50 of 3.08 × 10 −5 mol/L). Studies of molecular docking demonstrated the ability of quercetin to chelate a copper with the 3',4'-dihydroxy groups in the active site of tyrosinase. The chelation may prevent the entrance of substrate with inhibition of the catalytic activity of the enzyme.

Conclusions
In this work, flower and leaf extracts of S. nigra were concentrated using a novel solvent-resistant LMP ECTFE NF membrane in order to concentrate bioactive phytochemicals. Extracts prepared from flowers contained more abundant phenolic compounds than those prepared from leaves and showed the most interesting bioactivities. In particular, the retentate fraction of flowers extracted by methanol exhibited the major tyrosinase inhibitory activity and a promising antioxidant potential. The results obtained in this work suggest that S. nigra, particularly its flowers, may be an important dietary source of natural antioxidant and tyrosinase inhibitory agents.