Introduction

Pistacia lentiscus L., commonly known in Tunisia as Dharou or Lentisk, is a Mediterranean evergreen shrub belonging to the Anacardiaceae family producing bright red globose berries [1]. It is widely distributed in “extreme” ecosystems of the Mediterranean regions [2, 3].

Lentisk fruits are consumed raw or roasted. They give an edible oil that is traditionally utilized by Tunisians people in their daily diet as condiment and also in pastries and salads. It is used in phytotherapy for the treatment of many diseases such as rhumatism, respiratory allergies and scabies, and in the fabrication of antidiarrhea pills [4]. Lentisk oil is rich in saturated, monounsaturated and essential polyunsaturated (n-3) fatty acids including palmitic, oleic and linoleic acids (representing 23.96%, 48.37% and 23.31%, respectively) [5] as well as phytosterols, vitamins [6], and polyphenols [7]. The later are mainly represented by phenolic acids (2762.67 mg/kg oil), flavonols (377.71 mg/kg oil) and secoiridoids (366.71 mg/kg oil) that can be introduced into food and feed as antioxidants [2]. In addition, lentisk fruits are a source of phenolic acids (gallic acid, digallic acid, catechin), flavonoids (luteolin) [2, 8, 9], and anthocyanins that could be exploited as food colorants [10]. These metabolites are responsible for antioxidants and healthy properties of P. lentiscus fruits such as anti-inflammatory, anticancer [11], antidiabetic and hepatoprotective activities [2]. In fact, such metabolites act as free radical scavengers, reducing agent, hydrogen donors and as singlet oxygen quenchers [12]. Pistacia lentiscus fruits can be considered as a promising source of natural antioxidants that can be used in pharmaceutical and food industries to overcome the use of synthetic antioxidants such as butylated hydroxytoluene (BHT) and butylated hydroxyanisole (BHA) that have shown toxic and carcinogenic effects [13,14,15,16].

Various parameters such as solvent type and concentration, solid/liquid ratio, temperature, time, matrix structure and procedures are known to affect the extraction efficiency of antioxidants from plant matrix [17]. Generally, traditional extraction methods such as percolation, decoction, heat reflux extraction, soxhlet extraction and maceration are used for the recovery of phenolic compounds [18, 19]. However, these conventional methods require long time, involve large amounts of organic solvents, low efficient, and may cause degradations and losses of heat sensitive phenolic compounds [12,13,14,15,16,17,18,19,20,21]. The new emphasis on environmental protection, safety and economic concerns as well as green chemistry development are forcing industries to turn to greener extraction procedures [18,19,20,21,22,23]. Therefore, in recent year’s novel green extraction methods, including ultrasound-assisted extraction, microwave-assisted extraction, pressurized-assisted extraction and supercritical fluid, has gained considerable attention [19, 23,24,25]. These innovative green techniques permit time and energy consumption reduction, allow use of alternative Generally Recognized As Safe (GRAS) solvents and ensure a safe and high quality extract-product [20,21,22,23,24,25,26]. However, they require investment in high-cost instruments [17]. Bead milling is another green and eco-friendly process, which does not require high cost of investment and could be a promising alternative to ultrasound-assisted extraction [20,21,22,23,24,25,26,27,28]. This green extraction technique has been used in this study for the extraction of phenolic compounds from P. lentiscus fruits.

The choice of the adequate solvent constitutes also a crucial step for extraction process that should take into account several parameters such as polarity, selectivity, toxicity, environmental considerations of the solvent and the solubility of targeted compounds in the solvent [20,21,22,23,24,25,26,27,28,29]. Polar or medium polar solvents, such as water, ethanol, methanol, acetone or their various aqueous mixtures, were the most often used for the extraction of antioxidants from plant materials [30, 31]. Of these, acetone and methanol are highly toxic and are not recommended for the manufacture of food and pharmaceuticals as they can leave toxic residues [21,22,23,24,25,26,27,28,29,30,31,32].

Ethanol is the most widely used solvent given its numerous advantages such as availability, low price, purity, perfectly biodegradable. Also, it is a green solvent, non-toxic and safe for human consumption [33,34,35].

Several studies have shown that ethanol extraction efficiency can be improved by adding a certain amount of water [21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. In fact, the presence of water in the solvent system would enhance the mass transfer between the solid and the liquid by increasing the permeability of the vegetal matrix [32,33,34,35,36,37]. The optimal EtOH/H2O ratio will depend on the chemical nature of the compounds to be extracted but also on the conditions of extraction such as time, temperature and agitation speed and on the characteristics of the matrix [38]. This ratio is generally selected by default and scarcely studied. Furthermore, there is a lack of data about solvent effect on antioxidant content and activity of P. lentiscus fruits.

Recently, COSMO-RS predictive model is being used to predict the dissolving power and the selectivity of binary EtOH/H2O solvents for the extraction of antioxidants [20,21,22,23,24,25,26,27,28, 39].

In our current study, firstly a simulation approach using COSMO-RS software followed by an experimental approach based on various in vitro antioxidant assays (Folin–Ciocalteu reducing capacity, total flavonoid content, total monomeric anthocyanin content and radical scavenging capacity of extracts) were conducted. In further step, a chemical analysis by RP-HPLC was achieved in order to assess the impact of different EtOH/H2O concentrations on the extraction efficiency of phenolic compounds from P. lentiscus fruits and on their antioxidant activity (Fig. 1). So, the main goal of this work is to determine, using green strategies, the optimal EtOH/H2O ratio to obtain an extract rich in natural antioxidants and which could be safely incorporated in food formulations and supplements.

Fig. 1
figure 1

Scientific approach: COSMO-RS solubility predictions versus experimental solubility study

Full size image

Materials and methods

Reagents and standard

Folin–Ciocalteu reagent, anhydrous sodium carbonate (Na2CO3), gallic acid, 6-hydroxy-2,5,7,8-tetramethyl chroman-2-carboxylic acid (trolox) and catechin were purchased from Sigma-Aldrich ChemieGmbh (Steinheim, Germany). Sodium nitrite (NaNO2) solution, aluminium chloride hexahydrate solution (AlCl3, 6H2O) were purchased from Sigma-Aldrich (USA). For the extraction, only demineralized water and absolute ethanol (Deulep, France) were used. Water (HPLC), acetonitrile (HPLC), trifluoroacetic acid and all standards used were acquired from Sigma Aldrich (France).

Plant sampling

Pistacia lentiscus fruits were harvested from plants growing wild in Tabarka in North of Tunisia in November 2018 located at the following geographical coordinates obtained by GPS, (36° 5783,6′ N and 08° 5736,2′ E, altitude: 20 m). Plant identification was carried by Professor AbderrzekSmaoui (Biotechnology Center in Borj-Cedria Technopole, Tunisia). A voucher specimen was deposited at the herbarium of the Laboratory of Aromatic and Medicinal Plants, Biotechnology Center in Borj-Cedria Technopole under the “P.l.08003” number. Fruits were air-dried in the dark at room temperature and were ground using MF 10 basic microfine grinder drive, IKA® (Rotar speed 6500 rpm, MFsieves of 2 mm of diameter).

Extraction procedure: bead milling

Polyphenols were isolated from P. lentiscus fruits by bead milling system using ULTRA-TURRAX® Tube Drive (UTTD, Ika, Germany) operating in a 20 mL tube with 20 g of ceramic beads. 2.5 g of ground dried fruit was mixed with 25 mL of EtOH/H2O system at different proportions (100:0, 90:10, 80:20, 70:30, 60:40, 50:50, 40:60, 30:70, 20:80, 10:90, 0:100). The drive tube was operated at 4000 rpm for 30 min. The obtained extracts were then filtered through a Whatman filter paper (90 mm) using a Buchner funnel. The filtrate was evaporated at 40 °C using a rotary evaporator. Samples containing trace of water were then dried completely using a freeze dryer and stored at 4 °C until analysis.

In vitro antioxidant assays by colorimetric tests

Total phenolic content

Total phenolic content (TPC) of different extracts was assayed using the Folin–Ciocalteu reagent as described by Singleton and Rossi [40] with some modifications. Briefly, 125 µL of each diluted extract was added to 500 µL of distilled water and 125 µL of the Folin–Ciocalteu reagent. After 6 min, the final volume was brought up to 3 mL by adding 1250 µL of sodium carbonate (7%) followed by the addition of 1000 µL of distilled water. After incubation for 90 min in the dark and at room temperature (20 °C), the absorbance was measured at 760 nm using a UV–vis spectrophotometer (Biochrom, Libra S22, UK). The calibration curve was performed with gallic acid (concentrations ranging from 0 to 400 µg/mL) and total phenolic content was expressed as milligram of gallic acid equivalents (GAE) per gram of dry residue (mg GAE/g DR). Measurements were performed in triplicate.

Total flavonoid content

The total flavonoid content (TFC) was determined by a colorimetric assay developed by Dewanto et al. [41]. An aliquot of sample or standard solution of (+)-catechin (250 µL) was mixed with 75 µL of sodium nitrite (5%). The mixture was shaken vigorously for 6 min before adding, 150 µL of aluminum chloride (AlCl3, 6H2O, 10%) solution freshly prepared.

After five minutes, 500 µL of 1 M NaOH solution and 1525 µL of distillated water were added and the absorbance of the mixture was measured at 510 nm. Total flavonoids were expressed as mg of (+)-catechin equivalents per gram of dry residue (mg CE/g DR), through the calibration curve of (+)-catechin (concentrations ranging from 0 to 600 µg/mL). All samples were analyzed in triplicate.

Total anthocyanin content

The total anthocyanin content (TAC) was determined by the pH differential method [42]. An aliquot of each extract was diluted separately in 2 mL of potassium chloride (pH 1) and 2 mL of sodium acetate (pH 4.5) buffers. After 30 min of incubation, the absorbance was measured firstly at 520 and then at 700 nm using a UV–vis spectrophotometer (Biochrom, Libra S22, UK). Results were expressed in milligram cyanidin-3-glucoside equivalent per gram of dry residue (mg C3OG/g DR). All samples were analyzed in triplicate.

DPPH· radical scavenging activity

The 1.1-diphenyl-2-picrylhydrazyl (DPPH·) radical scavenging capacity of the extracts was estimated based on the method described by Hanato et al. [43]. 250 µL of (0.2 mM) DPPH-methanolic solution was added to 1 mL of each extracts at various concentrations. The obtained mixture was shaken vigorously and left standing in the dark for 30 min at room temperature, then the absorbance was measured spectrophotometrically at 517 nm. All samples were analyzed in triplicates and the final results were expressed as IC50 (µg/mL), which represent the concentration of sample required to scavenge 50% of DPPH· free radicals. The antiradical capacity (percent inhibition) was determined according to the following equation:

$$Percent\; inhibition \left( \% \right) = \frac{{\left( {{\text{A}}_{{{\text{control}}}} - {\text{A}}_{{{\text{sample}}}} } \right)}}{{{\text{A}}_{{{\text{control}}}} }}*100,$$

where Acontrol is the absorbance of the control at 30 min, and ASampleis the absorbance of the sample at 30 min.

Computational prediction: COSMO-RS

The COnductor-like Screening MOdel for Realistic Solvation (COSMO-RS), developed by Klamt [44], is a powerful computational method that combines a quantum chemical consideration (COSMO) with statistical thermodynamics (RS) to determine and predict the chemical potential of molecules in a liquid phase without any experimental data. This model was previously used in the field of extraction as a decision-making tool [28, 29, 45,46,47].

In this study, the model used is based on the prediction of the chemical potential of individual antioxidant in the various (EtOH/H2O) solvent mixtures. Calculation of the relative solubility of target polyphenols (gallic acid, cinnamic acid, benzoic acid, salicylic acid, caffeic acid, luteolin and catechin) in different EtOH/H2O mixtures was achieved by implementing this COSMO-RS model in COSMO therm software (C30 1401, CosmothermX14, COSMO logic GmbH &Co. KG). The relative solubility is calculated from the following equation:

$${\mathrm{log}}_{10}{(x}_{j})={\mathrm{log}}_{10}\left[\frac{\mathrm{exp}\left({\mu }_{j}^{pure}-{\mu }_{j}^{solvent}-\Delta {G}_{j,fusion}\right)}{RT}\right]$$
(1)

\({\upmu }_{\mathrm{j}}^{\mathrm{pure}}\): chemical potential of pure compound j (Joule/mol), \({\upmu }_{\mathrm{j}}^{\mathrm{solvent}}\): chemical potential of j at infinite dilution (Joule/mol), \(\mathrm{\Delta Gj},\mathrm{ fusion}\): free energy of fusion of j (Joule/mol), \({\mathrm{x}}_{\mathrm{j}}\)∶ solubility of j (g/g solvent).

Relative solubility is calculated in infinite dilution. The logarithm of the best solubility is set to 0 and all other solvents are ranked relatively to the best or reference solvent.

RP-HPLC analysis

High performance liquid chromatography (HPLC) was performed using an HPLC Waters chromatogram (Milford, MA) equipped with automatic injector (Waters e2695, USA) and a photodiode array detector (Waters 2998, USA). Separation of phenolic compounds was performed on a reverse phase RP-18 column (250 mm × 4.6 mm, 5 µm; Merck) thermostated at 25 °C. The mobile phase was composed of water and 0.025% trifluoroacetic acid (Eluent A) and acetonitrile (Eluent B). The flow rate was kept constant at 1 mL/min. The gradient program was as follows: 90 A/10 B (from 0 to 40 min), 50 A/50 B (from 40 to 41 min), 100% B (from 41 to 50 min) and 90 A/10 B (from 50 to 59 min). The injection volume was 20 µL. The compounds separated were monitored at 280 nm and identified by comparing retention times with known commercial pure standards.

The calibration curves were determined for compounds of interest at varying concentrations from 1 to 5 mg/mL. The analyses were performed in triplicata and the mean values were reported in mg/g extract.

Statistical analysis

All experiments were performed in triplicate and these values were then presented as mean values along with their standard derivations. The results are considered statistically significant at (P < 0.05). Principal component analysis (PCA) was conducted using XL-STAT software (2020).

Results and discussion

In order to evaluate the effect of EtOH concentration in EtOH/H2O mixture on the solubility of target molecules and to determine the best EtOH/H2O ratio for the extraction of antioxidants from P. lentiscus fruits, in silico study using COSMOthermX (version 17.02) and an experimental approach were employed.

In silico solubility study: COSMO-RS predictions

COSMO-RS simulations were performed to predict the solubility index of the major antioxidants of P. lentiscus fruits in pure EtOH, H2O and EtOH/H2O mixtures. These constituents, selected after exhaustive literature review, include four phenolic acids (gallic acid, caffeic acid, benzoic acid and salicylic acid), two flavonoids (catechin and luteolin) and three anthocyanins (delphenidin-3-O-glucside, cyanidin-3-O-glucoside and cyanidin-3-O-rutinoside). The relative solubility log10 (Xsolub) data are summarized in Table 1. It can be classified into three groups: low if log10 (Xsolub) is superior to − 4 (red colour), medium if log10 (Xsolub) is ranged between − 1 and − 4 (orange colour) and high if log10 (Xsolub) is ranged between 0 and − 1 (green colour). COSMO-RS predictions showed that only with EtOH 100% the log10 (Xsolub) of all solutes is equal to 0, thus making it the reference solvent. Among all tested mixture, EtOH/H2O (90/10) seems to be the best mixture for the extraction of all selected antioxidants presented in P. lentiscus fruits.

Table 1 COSMO-RS prediction results regarding the relative solubility of antioxidants of P. lentiscus fruit in the different solvent system
Full size table

Phenolic acids and flavonoids have a high solubility index in EtOH/H2O (80/20, 70/30 and 60/40) and present a moderate solubility in the rest of the solvent system. In particular, caffeic acid show high solubility also in EtOH/H2O (50/50, 40/60, 30/70, 20/80 and 10/90) and benzoic acid in EtOH/H2O (50/50, 40/60 and 30/70). Flavonoids (catechin and luteolin) exhibited low solubility index in 100% H2O.

Regarding anthocyanin, COSMO-RS predictions showed that hydro-alcoholic mixtures with high concentration in H2O (up to 50%) are not a good system to the solubilization of this class of solutes. In fact, all the three tested anthocyanin have a high solubility index only in EtOH (100%) and EtOH/H2O (90/10), except cyanidin-3-O-glucoside that present a high solubility also with EtOH/H2O (80/20). A similar trend was reported by Catena et al. [20]. Jacotet-Navarro et al. [39] who studied the effect of EtOH/H2O ratio on the extraction of antioxidants from rosemary reported that EtOH 100% was theoretically the best solvent for the solubilisation of these metabolites.

Experimental solubility study

Extraction yield

Bead milling extractions were performed to determine the effect of EtOH/H2O ratio on the extraction yields of phenolic compounds from P. lentiscus fruits under defined experimental conditions between 100/0 (EtOH/H2O) and 0/100 (EtOH/H2O).

The extraction yield ranged from 6.68 for EtOH (100%) to 20.14% for EtOH/H2O (30/70) (Table 2). EtOH (100%) exhibited the lowest extraction yield (6.68 ± 0.14%). In general, it can be noticed that the addition of H2O and the modification of EtOH concentration led to an improvement of the extraction yields to reach a maximum value of 20.14 ± 0.21% with EtOH/H2O system (30/70) which was probably due to the increased solubility of phenolic compounds, flavonoids and anthocyanins in the EtOH/H2O mixture. Similar trends were reported by Özbeket al. [21] who have studied the effect of EtOH concentration on the extraction of phenolic compounds from Pistacia vera L. hull and they have demonstrated that extraction yields increase with increasing EtOH concentration to reach a maximum at 50% and then decreased. In fact, according Cacace and Mazza [48], the modification of EtOH concentration in the solvent system affect the physical properties of the solvent such as density, dynamic viscosity, and dielectric constant. Furthermore, the solubility of solutes, in particular of phenolics, can be enhanced with the use of mixed solvent over a limited compositional range [48].

Table 2 Extraction yield, TPC, TFC and DPPH test of Pistacia lentiscus fruits extracts
Full size table

Total polyphenol contents

The total polyphenol contents (TPC) of P. lentiscus fruits in the different solvent systems are given in Table 2. As we can see from these data, EtOH/H2O concentration have statistically significant effect on TPC (P < 0.05). EtOH/H2O (60/40) and (70/30) exhibited the highest amounts of polyphenols (286.52 ± 1.51 and 292.04 ± 2.09 mg GAE/g DR, respectively) while the lowest amounts of polyphenols were obtained with pure solvents (EtOH 100% or H2O 100%) and EtOH/H2O (90/10). These findings were in accordance with previous reports suggesting that binary EtOH/H2O solvent systems were more effective for the recovery of phenolic compounds from plant matrix in comparison to mono-solvent system (EtOH or H2O) [21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37].

In addition, results showed that the TPC were strongly dependent on the concentration of H2O used in the different solvent system. In fact, TPC was improved increasing the H2O percentage from 10 to 30%, then, it did not significantly changes for 30 and 40% and decreased with further increase in H2O percentage. Similar trends were reported in previous works. Spigno et al. [49] reported that polyphenol content of EtOH extract from grape marc increased by increasing H2O in the mixture from 10 to 30%, kept constant for H2O contents between 30 and 60% and decreased for higher percentage. Yilamz and Toledo [50] demonstrated also that extraction of phenols from grape seeds using aqueous EtOH increased by adding H2O in the mixture from 0 to 30%, then kept constant from 30 to 50% and then decreased for higher percentage.

In general, the polarity of EtOH increased by adding H2O, which can improve the recovery of phenolic compounds by breaking the hydrogen bonds in polyphenols structure and ameliorate their solubility in organic solvents [51].

Total flavonoid contents

The solvent composition showed a significant (P < 0.05) effect on the TFC as reported in Table 2. Total flavonoids amount in P. lentiscus fruits ranged from 5.97 mg CE/g DR for EtOH/H2O (90/10) to 11.25 mg CE/g DR for EtOH/H2O (60/40). It can be noticed that P. lentiscus fruits are rich in polyphenols and not in flavonoids, as reported previously [2]. An increase in the amounts of flavonoids was observed by increasing the H2O percentage from 10 to 40%, then decreased for higher percentage (up to 40%). As can be seen in Table 2, the highest TFC of P. lentiscus fruit extracts was obtained with 60, 70 and 80% EtOH. Pure solvents (EtOH 100% and H2O 100%) gave similar flavonoids values (6.87 mg CE/g DR and 6.22 mg CE/g DR, respectively) and which are lower than those obtained with hydroalcoholic mixtures.

The same tendency was observed for TPC. These findings are in accordance with previous studies showing that binary EtOH/H2O solvent systems were more effective than pure solvents (EtOH or H2O) for the extraction of flavonoids from some herbs and spices [36]. Assefa et al. [52] highlighted also that the highest TFC of Yuzu peels was achieved using EtOH/H2O (60/40). However, other research found that the highest TFC of some popular tea sample was obtained with 50% EtOH [37]. These differences can be explained by the effect of plant matrix chemical composition on the dissolving power of the extraction solvent.

Among the tested solvent mixtures, EtOH/H2O (70/30) and EtOH/H2O (60/40) exhibited the highest amounts of flavonoids. This finding is in accordance with COSMO-RS prediction, which highlighted that flavonoids (catechin and luteolin) present a good solubility index in these solvent mixtures.

Total anthocyanin contents

The quantification of total anthocyanin in P. lentiscus fruits was achieved using the pH-differential method. Results showed that EtOH/H2O (80/20) extracts exhibited the highest anthocyanin contents (94 mg C3OG/100 g DR) (Table 2). In contrast, the lowest content (22.35 mg C3OG/100 g DR) was recorded in the aqueous extract (100% H2O). Results showed also that beyond 20% of H2O, the anthocyanin concentration decreases gradually with the increase in H2O concentration from 87 mg C3OG/100 g DR for EtOH/H2O (70/30) to 22.35 mg C3OG/100 g DR for EtOH 100%. These findings are in agreement with COSMO-RS predictions. Indeed, Cacace and Mazza [48], found that extraction of anthocyanins from blackcurrant is affected by the concentration of EtOH used and that 85% of EtOH is the best concentration for extraction of these metabolites.

Antioxidant activity

DPPH. assay was applied for the evaluation of antioxidant capacity of P. lentiscus fruits in the different solvent mixtures. The scavenging effect of the different extracts on the DPPH· radical was expressed as IC50 (Table 2). Results revealed that the IC50 values of P. lentiscus fruit extracts decreased with the increase in H2O concentration up to 30%. Further increase in H2O concentration (from 70 to 100%) lowered the radical scavenging activity of the extracts.

In general, all the extracts showed an appreciable ability to quench DPPH· radical with an IC50 values ranging from 2.39 to 8.94 µg/mL. EtOH/H2O (70/30) extract exhibited the highest antioxidant activity with an IC50 of about 2.39 µg/mL, and close to that of the positive control Trolox (2.56 µg/mL). The lowest DPPH· radical scavenging activity was obtained using pure solvents (EtOH 100% (8.94 µg/mL) and H2O 100% (8.35 µg/mL). These results sustain those of Özbek et al. [21] who found that Pistacia vera L. hull extracts obtained with aqueous organic solvents had better antiradical capacity than pure solvents.

The strongest antioxidant capacity of P. lentiscus extracts might be attributed to their richness in phenolic compounds in particular in polyphenols and anthocyanins [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53]. In fact, numerous studies showed that polyphenols contribute significantly to the total antioxidant activity of many medicinal plants due to their redox properties [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54].

Correlation between EtOH/H2O ratio, extraction yield, phenolic content and antioxidant activity

In order to assess the relationship between the content of phenolic compounds and the antioxidant activity of the different P. lentiscus fruits extracts, a principal component analysis was applied. This kind of analysis can be useful to establish mathematical correlations between the different factors and to highlight the relationships between the variables and the phenomena at the origin of these relations. PCA model indicated that the two first principal component explain 92.84% of the total variance with 66.3% and 26.50 for F1 and F2, respectively.

The PCA biplot (Fig. 2) showed that the different solvent mixtures could be classified into four different clusters based on yield, phenolic content and antioxidant activity. In fact, EtOH/H2O (90/10), (80/20), (70/30) and (60/40) were grouped together and formed the first cluster due to their richness on polyphenols, flavonoids and anthocyanins. The second cluster was formed by EtOH/H2O (50/50), (40/60) and (30/70). These solvent systems were characterized by high extraction yields but a low concentration on phenolic compounds. The third cluster was composed by EtOH/H2O (20/80), (10/90) and H2O (100%). In fact, extracts obtained with these solvents exhibited low phenolic concentration and low antioxidant capacity. EtOH 100% was closely separated from the other solvents mixtures and formed the fourth cluster. In addition, PCA biplot showed a positive correlation between phenolic contents, in particular TPC, TFC and TAC, and antioxidant activity (IC50). Indeed, the powerful antioxidant activity of Pistacia lentiscus fruits extracts can be attributed to their richness on phenolic compounds. These results are in good agreement with previous works highlighting the positive correlation between antioxidant potential and phenolic contents [2, 39, 55,56,57].

Fig. 2
figure 2

Biplot of P. lentiscus fruits extracts obtained from principal component analysis of data matrix of the phenolic compounds (TPC, TFC and TAC), yields and the antioxidant activity with first two principal component analysis

Full size image

RP-HPLC quantification

RP-HPLC profiles of P. lentiscus fruits extracts allowed the identification of seven phenolic compounds, including five phenolic acids (gallic acid, cinnamic acid, benzoic acid, salicylic acid and caffeic acid) and two flavonoids (luteolin and catechin) (Table 3). The results indicated that gallic acid was the major components of P. lentiscus fruits in all solvent mixtures. This later is a hallmark of Pistacia species [2]. Regarding the extraction capacity and selectivity of solvent mixtures, EtOH/H2O (70/30) showed the highest extracting ability for all metabolites. This could justify the significant antiradical activity of this extract. Indeed, Baratto et al. [58] have reported that the antioxidant activity of P. lentiscus could be attributed to the presence of gallic acid and its polygalloyl derivatives (5-O-galloyl, 3,5-di-O-galloyl).

Table 3 Phenolic composition of P. lentiscus fruits obtained by RP-HPLC
Full size table

In addition, previous studies indicated that the phenolic compounds isolated from P. lentiscus fruits are known to be potent antioxidants agents [2], which confers them strong biological properties encouraging their valorization in food and pharmaceutical industries. The sum of individual phenolic contents varied considerably and ranged from 2.57 mg/g DR for H2O (100%) to 9.61 mg/g DR for EtOH/H2O (70/30).

RP-HPLC results are in accordance with theoretical screening, showing that EtOH/H2O (70/30) was the best mixture for the solubilization of all antioxidant components from P. lentiscus fruits.

In silico solubility study versus experimental solubility determination

It is interesting to note that the COSMO-RS solubility predictions for P. lentiscus fruits antioxidants in the different solvent systems correlated well and were consistent with the experimental results.

Theoretically, the highest solubility values were obtained in EtOH/H2O systems at concentrations varying from (90/10) to (60/40) for phenolic acids and flavonoids. However, anthocyanin present high solubility values only in EtOH/H2O (90/10) and a moderate solubility in EtOH/H2O (80/20, 70/30, and 60/40) mixtures. The solubility of solutes decreases by increasing the H2O concentration up to 60%. The experimental results validated this prediction model. So, based on colorimetric tests and RP-HPLC quantification, EtOH/H2O (70/30) mixture was selected as the best solvent mixture to solubilize all studied Lentisk antioxidants. This mixture EtOH/H2O (70/30) was previously reported in literature [12], as solvent used for microwave assisted extraction of phenolic compounds from P. lentiscus leaves and fruits.

Conclusion

In this study, organic solvent–water mixture has been theoretically and experimentally proven to be more efficient in extracting antioxidant components than respective pure organic solvents. In fact, EtOH/H2O (70/30) exhibited the highest TPC, TFC, TAC and antioxidant capacity values and appeared as the best mixture for the extraction of antioxidants from P. lentiscus fruits. In addition, PCA analysis indicated that there was a good correlation between total phenolic content and the antioxidant capacity of P. lentiscus fruits extracts.