Chemical characterization and antioxidant properties of products and by‐products from Olea europaea L.

Abstract The products and by‐products of Olea europaea L.: olive fruits (primary agricultural product), oils (primary agro‐industrial product), pomaces (agro‐industrial processing by‐product), and leaves (agricultural practices by‐product), are promising sources of bioactive compounds. In the present study, qualitative and quantitative analyses of selected bioactive components in olive fruits, oils, and pomaces were performed. Total polyphenol content and antioxidant activity were analyzed in all samples (humid pomaces 2015: TPP, 26.0 ± 1.5–43.7 ± 3.0 g(GAEq)/kg DW; TEAC/ABTS, 189.5 ± 3.7–388.1 ± 12.0 mmol(Trx)kg DW). Radical (DPPH) quenching potential was analyzed via photometric and EPR methods, obtaining Vis/EPR signal ratio by 1.05 ± 0.45 and 1.66 ± 0.39 for fruits and pomaces, respectively. Through HPLC‐UV and HPLC‐MS/MS techniques, oleuropein and hydroxytyrosol, as well as selected hydroxycinnamic acids and flavonoids, were identified and quantified in olive fruits and pomaces. The main components were rutin, luteolin, and chlorogenic acid. Cytotoxic assay on fibroblast cells revealed toxic effects for selected extracts at highest tested concentrations (5%).

to the presence of minor antioxidant components, which are also responsible for its main organoleptic properties (such as the spice, bitter taste, distinguishing the freshly milled product), as well as its health-related properties, as preventing agents for cardiovascular diseases, atherosclerosis, and heart attacks, and antitumoural activities against colon and breast cancer (Batarseh & Kaddoumi, 2018;Bulotta et al., 2014;Rigacci & Stefani, 2016). Recently, olive polyphenols have been recognized as "health claims" by the EFSA, EU (Source: www.efsa.europa.eu). To this extent, it is very important to underline that antioxidant species, like carotenoids, tocopherols, and vitamins, can be found in many different foods (i.e., vegetables, cereals; Tamasi et al., 2015;Tamasi et al., 2019;Van Hung, 2016), whereas specific hydrophilic (poly)phenolic compounds (i.e., iridoids and secoiridoids, such as tyrosol, hydroxytyrosol, oleuropein, and ligstroside) are present, in great amount, only in EVOO, and related by-products of olive oil production.
The production of olive oil presents a challenge to agro-industrial waste management. Humid pomace, dry pomace, and mill wastewaters are produced in different quantities, based on specific milling technology. Mediterranean countries produce about 30 × 10 6 m 3 of mill waste. These by-products are pathogen-free and very rich in organic matter, in nutrients, and are also characterized by high levels of bioactive molecules (particularly, polyphenols), showing strong antimicrobial and phytotoxic activities and not easily biodegradable. For those bioactive properties, these by-products can be recovered and reused for the production of functional foods for human or animal consumption, as well as for diet supplements and cosmetics formulations (Gullón et al., 2018;Herrero et al., 2011;Kishikawa et al., 2015;Di Nunzio et al., 2018;Romero, Medina, Mateo, & Brenes, 2018;Sousa, Costa, Alexandre, & Prata, 2019;Vitali Čepo et al., 2018). For that reason, the use of the phenolic compounds extracted from olive by-products represents a great opportunity for the circular economy. Particular attention has been recently devoted to optimize nonconventional extraction procedures able to produce high-quality phytocomplexes by using nontoxic solvents. These protocols are usually assisted by ultrasound, microwave, or supercritical fluid extraction (SFE) using carbon dioxide as solvent (Chanioti & Tzia, 2018;Herrero, Pilar Sánchez-Camargo, Cifuentes, & Ibáñez, 2015;Xie et al., 2019). Several studies also indicate the possibility to increase the stability and the bioavailability of antioxidant and natural bioactive molecules using new carrier systems, like liposomes or polymeric micelles' formulations Leone et al., 2018Leone et al., , 2016Zhang, Huang, & Li, 2014).
Given this opportunity, the present study explored the chemical and nutraceutical characterization of products and by-products of O. europaea L.: olive fruits (primary agricultural products), EVVOs (primary agro-industrial products), pomaces (by-products from agroindustrial processing), collected at the harvestings in 2014-2015.
Selected polyphenols have been also identified and quantified through HPLC-UV and HPLC-MS/MS techniques, optimizing the analytical protocols on the basis of the chemical properties of the matrix and analytes.
The approach presented in this study focuses on the valorization of primary and secondary products from O. europaea L., highlighting the possibility to utilize the pomaces as source of bioactive molecules. This represents a challenge and a great opportunity from both environmental and economical points of view, building a model to increase the sustainability of agricultural and agro-industrial productions.

| Sample collection and storage
All samples were collected at harvesting/milling time in 2014 and 2015, from oil milling plants in southwest Tuscany (names are not reported for privacy reasons; Table S1). The samples of olive fruits, olive oil (extra-virgin), and pomace, coded as F1x-Y, EVOO1x-Y, and P1x-Y, were related to the same farm/oil mill. All the oil milling plants were based on two-phase technology (olive oil and humid pomace), except for samples 15-A and 15-C that were from three-phase systems (olive oil, dry pomace, and vegetation water).

| Olive fruits and pomaces
Aliquots of 0.250 g of each sample (analytically weighed) were defatted by 7 ml of n-hexane (twice), and the liquid phase was discarded.
The residual solid phase was then extracted (ultrasound assisted) by an hydroalcoholic mixture (EtOH/H 2 O, 80/20%, v/v; first extraction 5 ml, second and third extractions 2.5 ml; total volume extract, 10 ml). The extract was used as such, or dried under nitrogen flow (overnight) and then lyophilized. The dry extract was stored (darkness, −20 ± 1°C). Before HPLC analyses, the dry extract was reconstituted in 2 ml of solvent. In case of pomace samples, preliminary analyses comparing extracts with and without defatting process were carried out, obtaining results within 3% difference. Following this analysis, pomaces were usually extracted without previous defatting process.

| Trolox-Equivalent antioxidant capacity (TEAC) assays
Antioxidant activity was assayed by following the radical scavenger activity of free radicals ABTS •+ and DPPH • according to procedures previously reported (Brand-Williams, Cuvelier, & Berset, 1995;Re et al., 1999) with some modifications (Tamasi et al., 2019). The calibration curves were recorded by using standard solutions of Trolox, in the linear range, 0.20-20.00 µM (from a 0.55 mM mother solution in EtOH; R 2 > 0.990 were accepted for analyses; Figure S1b where Abs Trolox/Sample is the absorbance of the radical solution treated with standards or samples, and Abs Blank is the absorbance of the radical solution as such (ABTS •+ or DPPH • not treated solutions).
The results of TEAC were expressed as mmol Trolox equivalent per kg of dried sample (mmol(TrxEq)/kg DW).

| TEAC/ABTS assay
The ABTS •+ radical cation was prepared by incubation of a solution of ABTS (7 mM in EtOH) with a K 2 S 2 O 8 solution (140 mM in water) overnight (darkness, 4 ± 1°C) and dilution in EtOH before use. A known volume of this solution was treated with Trolox standard solutions or a known amount of extract (diluted, if necessary).
After 30 min of incubation in the dark, at 21 ± 2°C, the adsorption at 734 nm was recorded, against EtOH.

| TEAC/DPPH assays
A stock solution of DPPH • (0.10 mM in MeOH) was freshly prepared and used within 4 hr. A known volume of DPPH • solution was treated with Trolox standard solutions or a known amount of extract (diluted, if necessary). After 15 min of incubation in the dark, at 21 ± 2°C, the adsorption at 517 nm was recorded, against MeOH. In case of olive fruits and pomaces, the same experiment was carried out reading the DPPH • solution (blank and treated) via EPR spectroscopy. EPR spectra were acquired on continuouswave X-band (CW, 9GHz) using a Bruker E500 ELEXSYS Series spectrometer (Bruker, Italy), with the ER 4,122 SHQE cavity. EPR measurements were performed at 21 ± 1°C, 9.8 GHz microwave frequency, 0.1 mT modulation amplitude, and 4 mW microwave power. The sample was placed into a 3.0 mm ID × 4.0 mm OD, suprasil tube. In this case, stock solution of DPPH • was prepared (1.0 mM in MeOH) and the final concentration of radicals in each sample was 0.45 mM. The acquisition of EPR signal was carried out 15 min after the addition of the antioxidant (Trolox or extract; darkness, 21 ± 2°C; Figure S1d), and the antioxidant activity was calculated by the relative decrease in area (instead of absorbance).
The area of the EPR spectra was calculated by the double integral of the DPPH signal.

| HPLC-MS: identification and quantification of hydroxycinnamic acids and flavonoids
A reverse-phase column (Phenomenex Kinetex biphenyl, 10 × 2.1 mm, 5 μm particles, 100 Å pore, shell-core), with safeguard precolumn (Phenomenex Phenyl, 4 × 2.0 mm), thermostated at 35 ± 1°C, was used (Tamasi et al., 2019). The eluents were as  Table 1 and Figure S3. The analytical determination was carried out via external calibration, by using genistein as internal standard. Calibration showing correlation factors R 2 > 0.990 was accepted for analyses. The values for LOQ and LOD were also defined (Table 1). Details on cell culture and cell viability procedures were those reported in Bonechi et al. (2018). Briefly, the fibroblasts NIH3T3 were propagated in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal calf serum (FCS), 1% L-glutaminepenicillin-streptomycin, and 1% MEM nonessential amino acid, and maintained at 37°C, in a humidified atmosphere (5% CO 2 ). When at confluence, cells were washed with phosphate-buffered saline solution (0.1 M, PBS) and separated using a trypsin-EDTA solution and centrifuged (118 g, 5 min). The pellet was resuspended and diluted in medium solution and added by different concentrations (0.5, 1.0, and 5.0%, v/v) of tested solutions (standards or extracts). After 24 hr of incubation, cell viability was evaluated by neutral red uptake. The incubation medium was removed, and cells were washed with prewarmed PBS; then, the neutral red medium (NR, 0.33 g in 100 ml sterile water, then diluted 100 times) was added and the samples were further incubated at 37°C, 95% humidity, 5.0% CO 2 for 3 hr.

| Statistical data treatment
Three samples were collected for each type, pretreatments (extracts) were performed in triplicate for each sample, and triplicate analyses were performed for all measurements in each extract.
In two cases, they were higher than the maximum value for "extra-   Šarolić et al., 2014;Galvano et al., 2007). A linear correlation between TPP and TEAC/ABTS parameters (Figure 1a) was

| Olive fruit samples
TPP values for olive fruits ranged 12.0 ± 0.9-40.2 ± 1.2 g(GAEq)/ kg DW ( harvesting time, soil, and climatic conditions. As already mentioned, 2014 was an unusual year, with a Bactrocera oleae infestation in central Italy, and unusual weather conditions. The lower summer temperature and the higher summer humidity maximum ca. 22°C and 80%, respectively, ( Figure S4) strongly influenced the fruit ripening process, producing a marked decrease in final quantity. On the other hand, warmer winter temperatures allowed higher quantity of insects to survive and lay eggs inside the fruits (Rice, 2000). On the contrary, the 2015 summer was hot and dry, and followed by a colder and dryer winter, leading to an increase in production and fruits' quality.

TPP TEAC/ABTS TEAC/DPPH(Vis) TEAC/DPPH(EPR)
Olive fruits F14-A 19.2 ± 0.7 a 14.3 ± 0.5 a 55.5 ± 1.9 a 62.5 ± 2.3 a F14-B 12.0 ± 0.9 b 9.5 ± 0.8 b 16.0 ± 0.7 b 14.2 ± 0.9 b F14-C 13.7 ± 0.5 b 9.9 ± 0.6 b 14.7 ± 0.5 b 37.   Table 3). In addition to fruits' quality, the other important factor which strongly affects the content of polyphenols in olive pomace is olive oil production technology. The usage of hot water in three-phase mill systems brings about a lower antioxidant activity and polyphenol content, as revealed by two samples from the year 2015 that were very dry (P15-A and P15-C). This could be reasonably explained suggesting that, the added hot water, works as extragent, moving the polyphenols and other antioxidant species, to waste wasters. Other impact production process can be related to the possible seed removal, which also is a source of antioxidant compounds. Leaving seeds in production process could cause higher antioxidant activity for pomace than for fruits, as seeds were removed from fruits before their analysis.  Figure 1c). These latter samples were from geographical areas different from that of the other samples.

| HPLC-UV: identification and quantification of hydroxytyrosol and oleuropein in olive fruit and pomace samples
The hydroalcoholic (EtOH/H 2 O, 80/20%, v/v) extracts of olive fruits from 2015 harvest showed contents of hydroxytyrosol and oleuropein of the same order of magnitude, ranging 2.4 ± 0.2-6.8 ± 0.3, and 0.7 ± 0.1-9.8 ± 0.4 g/kg DW, respectively (Table 4 and Figure   S5). The minimum value for both analytes was found for sample Regarding the olive pomaces from 2015, oleuropein was not quantifiable (trace, <0.5 g/kg DW), whereas hydroxytyrosol ranged 5.3 ± 0.2-8.0 ± 0.3 g/kg DW, excluding the two dry pomace samples (P15-A and P15-C), for which the values were much lower (1.2 ± 0.1 and 0.4 ± 0.1 g/kg DW, respectively). The hydroxytyrosol contents correlate quite well with general antioxidant parameters (particularly with TPP; y = 5.1450x + 2.8544; R 2 = 0.807; p < .001, at 95% confidence interval, Figure 1d). Studies indicate that many factors, such as cultivar, geographic origin, pedo-climatic conditions, agronomical cultivation protocols (i.e., irrigation, fertilization, plant and soil treatments), ripening stage, and postharvest processing, strongly affect the phenolic profile of olive fruits, oil product, and pomace (Uylaşer & Yildiz, 2014). In particular, higher contents of oleuropein, related to the bitter taste to the drupes, have been mainly found in the skin of the fruit, and were reported as related to its ripening stage. Oleuropein undergoes enzymatic processes, by hydrolases and oxidases producing hydroxytyrosol and/or the quinone derivative (Scheme S1). These oxidative reactions also occur during the oil production process (malaxation stage), bringing about oleoside derivatives in olive pomace (Cardoso et al., 2005;Marsilio, 2001;Romero et al., 2004).
Quercetin and luteolin-7-O-rutinoside were one or two order (

| Principal component analysis
The data just above commented were confirmed, at a great extent,
On the contrary, fruits and EVOOs 2014 revealed toxic effect at 5% (v/v), with a major effect from EVOOs with respect to fruits.
For the 2015 harvest samples, none of the tested extracts were toxic at 0.5 and 1% (v/v) concentrations (Figure 3b and Figure S6), whereas 5% (v/v) treated cells showed a great decrease of viability. It was previously reported that quercetin and rutin modified mouse fibroblasts NIH3T3 viability at higher concentrations than those present in the extracts tested in this study (Araújo, de M.B. Costa, Pazini, Valadares, & de Oliveira, 2013;Bonechi et al., 2018).
However, this outcome may be reasonably explained as a result of the cumulative and synergic effects of several components and their metabolites.

| CON CLUS IONS
Qualitative and quantitative analyses of olive fruits, olive oils (primary product), and olive pomaces (by-product from technology) showed multiple factors that influenced the antioxidant properties and polyphenol components. These include genetic factors, fruit maturation stage, agronomical practices, geographical and pedoclimatic conditions, as well as production technologies (dry and humid pomaces). The results showed that pomace, in particular the humid by-product, is a promising source of bioactive and antioxidant compounds, without cytotoxic properties. Taking into account the human health benefits of antioxidant polyphenols and considering the importance of olive oil production in the Mediterranean basin, the possibility to utilize olive pomaces as source of nutraceuticals should be a priority. These materials, usually considered as waste products, could be used for the formulation of novel diet supplements and food fortifiers, as well as for applications in cosmetics. This approach allows the valorization of primary and secondary products from O. europaea L. and could be considered a model for other agriculture productions (e.g., viticulture, horticulture, cereal crops) to increase the sustainability of agricultural activities.

ACK N OWLED G M ENTS
Tuscan Regional Administration is acknowledged for funding the project NUTRIforOIL. The authors gratefully acknowledge factory partners that kindly gifted samples of EVOOs, olive fruits, and pomaces. Toscana Life Sciences Foundation is acknowledged for the use of HPLC-MS instrument.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.

E TH I C A L S TATEM ENT
This study does not involve neither human nor animal testing.