1. Introduction
Hydroxytyrosol is one of the natural antioxidants present in olive-derived biomasses and olive oil [
1]. In particular, it is found in different by-products/wastes obtained during olive oil and pomace olive oil production such as olive leaves, olive pomace, exhausted olive pomace (or extracted olive pomace, defatted olive pomace) (EOP) and olive mill wastewater [
2]. Different preclinical studies have attributed anti-inflammatory, antiproliferative, proapoptotic, antimicrobial and neuroprotective properties to hydroxytyrosol [
3,
4,
5]. The healthy benefits of hydroxytyrosol have also been shown in clinical trials [
6]. It protects against lipid oxidation and prevents cardiovascular disease [
6] and in combination with vitamin E improves steatosis and hypertriglyceridemia in children [
7]. In fact, a health claim has been published by the European Food Safety Authority (EFSA) on the benefits of a minimum daily intake of 5 mg hydroxytyrosol and its derivatives through olive oil consumption [
6,
8].
Furthermore, hydroxytyrosol has a lipophilic and a hydrophilic feature that makes it soluble in water and fats [
1]. In this way, the EFSA has included hydroxytyrosol as a new safe antioxidant food ingredient to be added into fish and vegetable oils and spreadable fats [
9].
In olive-derived biomasses, other bioactive compounds can be found, e.g., the sugar alcohol mannitol and the triterpenes maslinic acid and oleanolic acid [
10,
11,
12]. The former compound has food preservative properties (increasing food shelf life by reducing sugar crystallization) and it is a low-calorie sweetener [
11,
13]. It is also currently used as a drug to treat acute stroke and as a protective and therapeutic agent during neurological or renal failure [
14]. Mannitol can be co-extracted with phenolic compounds using alcoholic-water solutions [
10]. In the case of the triterpenic acids, they also have important health and disease prevention properties [
15], such as anti-inflammatory, antihyperlipidemic, antitumoral and antimicrobial activities [
16,
17,
18,
19]. Triterpenic acids generally require pure or closely pure alcohols for solubilization, as shown some previous studies on olive leaves [
20] and olive pomace [
21].
Currently, novel extraction methods such as ultrasound-assisted extraction (UAE) are applied to obtain bioactive compounds from agri-food biomasses. Ultrasound is one of the key technologies to achieve the goal of sustainable “green” extraction and chemistry. UAE is a non-thermal technology that reduces the extraction time and the solvent-to-solid ratio compared, for example, with Soxhlet extraction [
22,
23,
24], as it is an environmentally and economically viable alternative to conventional extraction techniques [
24,
25]. UAE can produce cavitation, vibration and mixing of the media. These effects cause the cell wall of the materials to rupture and allow the extraction of the compounds of interest [
24,
26]. Industrial-scale systems are currently available, including bath systems from TierraTech and probe systems (batch and continuous) from Hielscher-Ultrasound Technology [
27]. Thus, the evaluation of the operational parameters affecting the extraction of the latter bioactive compounds at lab scale can be useful to move towards large scale and valorize EOP, the industrial final waste obtained in the olive sector in some countries.
In a previous work on EOP, water was used as an efficient and environmentally friendly solvent to recover phenolic compounds from EOP, but both high temperature and long extraction time were required, i.e., 85 °C and 90 min, respectively [
28]. This work means an advance in the study of this little explored waste for the recovery of bioactive compounds, including phenolic compounds and mannitol. To the best of our knowledge, this is the first time that ultrasound assisted water extraction (UAWE) of EOP has been optimized and the main operational parameters evaluated. In addition, a second ethanolic extraction step was used to recover triterpenic acids from this feedstock. This sequential extraction extends the opportunities to exploit EOP in a greener manner.
2. Materials and Methods
2.1. Chemicals, Reagents and Standards
Folin–Ciocalteu’s phenol reagent, 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonate) (ABTS™), 2,4,6,-tri(2pyridyl)-1,3,5,-triazine (TPTZ), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), orthophosphoric acid and the standard of gallic acid were provided by Sigma-Aldrich (St. Louis, MO, USA). Methanol and acetonitrile of HPLC grade was obtained from Honeywell (Morristown, NJ, USA) and PanReac AppliChem (Barcelona, Spain), respectively. The following standards were obtained from Extrasynthese (Genay, France): hydroxytyrosol, maslinic acid and oleanolic acid. Ultrapure water was obtained using a Milli-Q system (Millipore, Bedford, MA, USA).
2.2. Raw Material
Industrial EOP was obtained as pellets (average length, 14.5 mm; average diameter, 4.6 mm) from “Spuny SA” (Linares, Jaén, Spain). One part of the sample was milled to a particle size of around 1 mm in an Ultra Centrifugal Mill ZM 200 (Retsch, Haan, Germany) and the other part remained pelletized. Both samples were stored in a dry place until use.
2.3. Methodology
Figure 1 shows the experimental procedure followed in this work in order to extract phenolic compounds and mannitol from EOP by UAWE and to optimise the experimental conditions. After that, triterpenic acids were recovered in a second ethanolic extraction and the resulting fractions were characterized.
Firstly, the effect of milling was tested, while using a Box–Behnken design (BBD), UAWE operational parameters (extraction time, solid loading, and amplitude) were evaluated to obtain phenolic compounds and mannitol. Moreover, the extracts obtained under optimized conditions were further characterized by high-performance liquid chromatography (HPLC)-ion trap mass spectrometry (IT-MS) to profile minor hydroxytyrosol derivatives. The structural changes of the extracted solid obtained under UAWE optimized conditions were studied by scanning electron microscopy (SEM) and it was also chemically characterized. Then, that solid was subjected to a second extraction step using ethanol to obtain maslinic acid and oleanolic acid and the effect of the previous step was evaluated.
2.4. UAWE of EOP
UAWE was performed using an ultrasonic probe (Branson 550, Ultrasonics Corporation; Danbury, CT, USA) with a power of 550 W and a frequency of 50–60 KHz. A sonotrode of 30 mm of diameter was used. The sonotrode was immersed 1 cm deep into the samples (250 mL in vessels). The sonication was performed in continuous mode, i.e., the waves propagated continuously through the sample during the whole extraction time, without any interruption. The experiments were initiated at room temperature and the increment of temperature due to the sonication effects was measured at the end of each assay.
After extraction, the slurry was filtered under vacuum, obtaining two fractions: a liquid fraction (aqueous extract) and a solid fraction (extracted EOP solid). An aliquot of the aqueous extracts was filtered with a syringe filter (nylon, pore size 0.45 μm) (SinerLab Group, Madrid, Spain) and stored at −20 °C until analysis. The solid was oven-dried at 40 °C and then stored.
Firstly, some preliminary experiments were carried out with pelletized and milled EOP, using a solid loading of 8.5% (
w/
v) to test two amplitudes (20% and 80%) and extraction times (2 min and 13 min). All extraction assays were conducted in duplicate. Then, an experimental design was applied (
Section 2.5).
2.5. Box-Behnken Experimental Design for UAWE of EOP
The optimization of UAWE of bioactive compounds from the pelletized EOP was performed using a response surface methodology (RSM). As operational variables, ultrasound amplitude (20–80%), extraction time (2–18 min) and solid loading (2–15%, w/v) were studied. The BBD consisted of 17 experiments, including five central points (50% amplitude, 10 min and 8.5% of solid loading). These experiments were performed in random order.
The influence of each independent variable was determined according to the following equation:
where
y is the response variable,
xi and
xj are the operational variables in coded values ranging from −1 to 1,
β0,
βi,
βij and
βii are the regression coefficients calculated from the experimental results by the least-squares method. Analysis of variance (ANOVA) was used to establish the significance of the results. To determine the model goodness, the following parameters were evaluated: the coefficient of determination (R
2), adjusted determination coefficient (R
2 adj) and coefficient of variance (CV, percent). The significance of all terms in the polynomial equation was considered statistically different when
p < 0.05. After applying a multiple response optimization, the optimal conditions were reproduced (five replicates) to compare the experimental with the predicted data and assess the validity of the model. Additionally, these conditions were applied to obtain antioxidant compounds and mannitol from milled EOP for comparison.
After each extraction, the samples were filtered under vacuum, treated and stored as previously described.
2.6. Extraction Yield
It was determined gravimetrically. In brief, an aliquot of the extracts (2 mL) was poured into a 10 mL glass tube. Then, it was oven-dried at 105 °C until constant weight. All samples were measured in duplicate. The data were expressed as percent (g of extract/100 g of EOP, dry weight).
2.7. Analytical Determinations of the UAWE Extracts
2.7.1. Total Phenolic Content and Antioxidant Capacity
TPC was estimated using the Folin–Ciocalteu assay according to Gómez-Cruz et al. [
28]. The absorbance of each sample was measured at 760 nm in a Bio-Rad iMark
TM microplate absorbance reader (Hercules, CA, USA). Gallic acid was used as standard (concentration range 0–0.295 g/L) and the results expressed as grams of gallic acid equivalents (GAE)/L in the aqueous extract and milligrams of GAE/g of EOP (dry weight).
In addition, the antioxidant activity of the extracts obtained from EOP was determined by ferric reducing power assay (FRAP) and ABTS assay according to Gómez-Cruz et al. [
28] at 734 nm and 593 nm, respectively, using the aforementioned device. The first method is based on the reduction of Fe(TPTZ)
23+ to Fe(TPTZ)
22+ by the donation of electrons from the antioxidant compounds and the second method is based on the neutralization of the ABTS radical by the antioxidant compounds. For both assays, the results were expressed as mg Trolox equivalent (TE)/g of EOP (dry weight) using Trolox as standard (concentration range, 0–0.279 g/L for FRAP; 0–0.629 g/L for ABTS). TPC and antioxidant assays were carried out in triplicate.
2.7.2. Characterization of Phenolic Compounds and Mannitol by HPLC Analyses
For the determination of phenolic compounds, a Shimadzu Prominence HPLC equipment was applied according to Gómez-Cruz et al. [
28]. It was equipped with an SPD-M20A diode array detection (DAD). The analysis was performed using a BDS HYPERSIL C18 column (290 mm × 4.6 mm, 5 μm particle size) (Thermo Fisher Scientific Inc., Waltham, MA, USA). Retention time and UV absorption spectra allowed the identification of hydroxytyrosol by comparison with its commercial standard. This compound was quantified at 280 nm (
y = 19,113
x − 15,977; R
2, 1.0000) and the results were expressed in g/L in the aqueous extract and mg/g of EOP (dry weight).
In addition, HPLC-IT-MS and -MS
2 analyses were performed in an Agilent 1100 HPLC connected on-line to an Esquire 6000 IT (Bruker, Bremen, Germany) via an electrospray interface, according to Medfai et al. [
29]. A Kinetex core–shell C18 column (2.1 mm × 50 mm, 2.7 µm) (Phenomenex, Barcelona, Spain) was applied. MS and MS/MS spectra were recorded over the mass-to-charge (
m/
z) range of 100–1200 in the negative ionization mode. Auto MS/MS analyses were performed at 0.6 V. DataAnalysis (version 4.0) from Bruker was used to process the data.
For mannitol analysis, samples were previously conditioned with ion-exchange resins (Microionex MB200, Rohm Haas, Denmark) to remove impurities and then, filtered through 0.45 µm nylon membranes. It was then analyzed by HPLC equipped with refractive index detection (RID). Carbohydrate column (CARBOSep CHO-782 Pb, Transgenomic, Inc., Omaha, NE, USA) and ultrapure water as mobile phase were employed. The flow rate and the column temperature were set at 0.6 mL/min and 70 °C, respectively. The results were expressed as g/L in the aqueous extract and mg/g of EOP (dry weight) using the following curve (y = 8.82 × 105x + 6.63 × 104; R2, 0.9997).
2.8. Characterization of the Extracted EOP Solids after UAWE
2.8.1. Chemical Characterization
The methodology of the National Renewable Energy Laboratory (NREL) was used to chemically characterize the extracted EOP solids. First, the content in extractives was determined by Soxhlet extraction with water and ethanol according to NREL/TP-510-42619 [
30]. Then, cellulose, hemicellulose and lignin contents were also determined according to the NREL methodology [
31] and ash content according to NREL/TP-510-42622 [
32]. Moreover, the elemental composition (C, H, N and S) was analyzed using a TruSpec Micro device (Leco, St. Joseph, MI, USA).
2.8.2. Scanning Electron Microscopy
SEM (MERLIN from Carl Zeiss) (Oberkochen, Germany) was used to evaluate the morphological changes of the biomass before and after UAWE for both pelletized and milled EOP. Dry samples were fixed with double-sided adhesive tape mounted on SEM holders and metalized with gold. The samples were photographed at high vacuum, 5 kV and 100× and 1000× magnification, i.e., 500 μm and 50 μm scales, respectively.
2.9. Extraction of Triterpenic Acids from the Extracted EOP Solids after UAWE
The extraction of triterpenic acids was carried out according to Romero et al. [
33], with some modifications. The extractive agent was absolute ethanol and the extraction was performed at 10% (
w/
w) solid loading and room temperature (24 h, 150 rpm) in a rotary shaker (INFORS HT Ecotron, Surrey, UK). Each sample was centrifuged (MicroCen 16, Herolab, Germany) and filtered with a syringe filter (nylon, pore size 0.22 μm) (SinerLab Group, Madrid, Spain).
Extracted EOP solids (solid remaining of EOP) obtained after UAWE in the previous BBD experimental assays and under optimized conditions were subjected to a second extraction for recovering triterpenes acids. All extraction assays were conducted in duplicate and the extraction yield was determined as for the aqueous extracts obtained by UAWE. Triterpenic acids were quantified using external standards: maslinic acid (y = 8118.3x + 123,127; R2, 0.9968) and oleanolic acid (y = 10,717x + 48,413; R2, 0.9992). This determination was performed using the aforementioned HPLC-DAD conditions at 210 nm, according to Romero et al., (2017). The results were expressed as g/L in the ethanolic extracts and mg/g of EOP (dry weight).
2.10. Statistical Analysis
The statistical analysis of the experimental design was performed using Design-Expert® v8.0.7.1 software (Stat-Ease, Inc., Minneapolis, MN, USA), which also served to evaluate how the operational parameters affected the increment of temperature promoted by UAWE. In addition, the indirect effects of the UAWE operational variables on triterpene extraction were also studied with the latter software. An ANOVA analysis was also carried out using the post hoc test Fishers Least Significant Difference LSD using Statgraphics Centurion XVII (StatPoint Technologies, Inc., Warrenton, VA, USA). A correlation test and t-test were performed using Microsoft Office Excel 2007 (Redmond, WA, USA).
4. Conclusions
The present study revealed that UAWE can be applied to extract phenolic compounds, mostly hydroxytyrosol, and mannitol from the industrial pelletized EOP. After applying the response surface methodology, the optimal conditions that enabled to maximize simultaneously the yield, the extraction of phenolic compounds, including hydroxytyrosol, and mannitol were: 80% amplitude, 16 min and 11.5% solid loading (w/v). The TPC, hydroxytyrosol content and mannitol content were 40.04 mg GAE/g EOP, 6.42 mg/g EOP and 50.92 mg/g EOP, respectively. Besides hydroxytyrosol, other minor but interesting derivatives were characterized, including oleacein, verbascoside and oleuropein. Moreover, UAWE of EOP favored the subsequent recovery of maslinic and oleanolic acids by ethanolic extraction. Overall, UAWE is a promising green extraction procedure to valorize EOP and recover different type of bioactive compounds.